From cd3c88ad7e9b5ae315e14784db256733ab43431b Mon Sep 17 00:00:00 2001 From: Stenzek Date: Sun, 24 Mar 2024 14:26:41 +1000 Subject: [PATCH] deps: Remove now-unused vendored libs --- dep/CMakeLists.txt | 8 - dep/libjpeg/CMakeLists.txt | 65 - dep/libjpeg/README | 374 - dep/libjpeg/change.log | 527 -- dep/libjpeg/include/jconfig.h | 171 - dep/libjpeg/include/jerror.h | 304 - dep/libjpeg/include/jmorecfg.h | 457 -- dep/libjpeg/include/jpegint.h | 445 -- dep/libjpeg/include/jpeglib.h | 1183 --- dep/libjpeg/libjpeg.vcxproj | 76 - dep/libjpeg/libjpeg.vcxproj.filters | 52 - dep/libjpeg/src/jaricom.c | 153 - dep/libjpeg/src/jcapimin.c | 288 - dep/libjpeg/src/jcapistd.c | 162 - dep/libjpeg/src/jcarith.c | 945 --- dep/libjpeg/src/jccoefct.c | 456 -- dep/libjpeg/src/jccolor.c | 598 -- dep/libjpeg/src/jcdctmgr.c | 466 -- dep/libjpeg/src/jchuff.c | 1656 ----- dep/libjpeg/src/jcinit.c | 249 - dep/libjpeg/src/jcmainct.c | 297 - dep/libjpeg/src/jcmarker.c | 717 -- dep/libjpeg/src/jcmaster.c | 675 -- dep/libjpeg/src/jcomapi.c | 244 - dep/libjpeg/src/jcparam.c | 591 -- dep/libjpeg/src/jcprepct.c | 358 - dep/libjpeg/src/jcsample.c | 545 -- dep/libjpeg/src/jctrans.c | 399 -- dep/libjpeg/src/jdapimin.c | 412 -- dep/libjpeg/src/jdapistd.c | 276 - dep/libjpeg/src/jdarith.c | 796 --- dep/libjpeg/src/jdatadst.c | 263 - dep/libjpeg/src/jdatasrc.c | 271 - dep/libjpeg/src/jdcoefct.c | 744 -- dep/libjpeg/src/jdcolor.c | 769 -- dep/libjpeg/src/jdct.h | 409 -- dep/libjpeg/src/jddctmgr.c | 384 - dep/libjpeg/src/jdhuff.c | 1559 ---- dep/libjpeg/src/jdinput.c | 657 -- dep/libjpeg/src/jdmainct.c | 511 -- dep/libjpeg/src/jdmarker.c | 1505 ---- dep/libjpeg/src/jdmaster.c | 532 -- dep/libjpeg/src/jdmerge.c | 437 -- dep/libjpeg/src/jdpostct.c | 290 - dep/libjpeg/src/jdsample.c | 341 - dep/libjpeg/src/jdtrans.c | 140 - dep/libjpeg/src/jerror.c | 253 - dep/libjpeg/src/jfdctflt.c | 176 - dep/libjpeg/src/jfdctfst.c | 232 - dep/libjpeg/src/jfdctint.c | 4415 ------------ dep/libjpeg/src/jidctflt.c | 238 - dep/libjpeg/src/jidctfst.c | 351 - dep/libjpeg/src/jidctint.c | 5240 -------------- dep/libjpeg/src/jinclude.h | 157 - dep/libjpeg/src/jmemmgr.c | 1115 --- dep/libjpeg/src/jmemnobs.c | 113 - dep/libjpeg/src/jmemsys.h | 198 - dep/libjpeg/src/jquant1.c | 851 --- dep/libjpeg/src/jquant2.c | 1311 ---- dep/libjpeg/src/jutils.c | 224 - dep/libjpeg/src/jversion.h | 14 - dep/libjpeg/src/transupp.c | 2433 ------- dep/libjpeg/src/transupp.h | 230 - dep/libpng/ANNOUNCE | 63 - dep/libpng/AUTHORS | 61 - dep/libpng/CHANGES | 6202 ---------------- dep/libpng/CMakeLists.txt | 39 - dep/libpng/LICENSE | 134 - dep/libpng/README | 184 - dep/libpng/TODO | 23 - dep/libpng/TRADEMARK | 8 - dep/libpng/include/png.h | 3250 --------- dep/libpng/include/pngconf.h | 623 -- dep/libpng/include/pnglibconf.h | 224 - dep/libpng/libpng.vcxproj | 61 - dep/libpng/libpng.vcxproj.filters | 52 - dep/libpng/src/arm/arm_init.c | 139 - dep/libpng/src/arm/filter_neon.S | 253 - dep/libpng/src/arm/filter_neon_intrinsics.c | 402 -- dep/libpng/src/arm/palette_neon_intrinsics.c | 151 - dep/libpng/src/intel/filter_sse2_intrinsics.c | 391 - dep/libpng/src/intel/intel_init.c | 52 - dep/libpng/src/png.c | 4559 ------------ dep/libpng/src/pngdebug.h | 153 - dep/libpng/src/pngerror.c | 957 --- dep/libpng/src/pngget.c | 1267 ---- dep/libpng/src/pnginfo.h | 267 - dep/libpng/src/pngmem.c | 284 - dep/libpng/src/pngpread.c | 1104 --- dep/libpng/src/pngpriv.h | 2221 ------ dep/libpng/src/pngread.c | 4228 ----------- dep/libpng/src/pngrio.c | 120 - dep/libpng/src/pngrtran.c | 5040 ------------- dep/libpng/src/pngrutil.c | 4680 ------------ dep/libpng/src/pngset.c | 1803 ----- dep/libpng/src/pngstruct.h | 479 -- dep/libpng/src/pngtrans.c | 868 --- dep/libpng/src/pngwio.c | 168 - dep/libpng/src/pngwrite.c | 2418 ------- dep/libpng/src/pngwtran.c | 575 -- dep/libpng/src/pngwutil.c | 2781 -------- dep/zlib/CMakeLists.txt | 33 - dep/zlib/ChangeLog | 1515 ---- dep/zlib/FAQ | 368 - dep/zlib/README | 115 - dep/zlib/include/zconf.h | 534 -- dep/zlib/include/zlib.h | 1912 ----- dep/zlib/src/adler32.c | 186 - dep/zlib/src/compress.c | 86 - dep/zlib/src/crc32.c | 442 -- dep/zlib/src/crc32.h | 441 -- dep/zlib/src/deflate.c | 2163 ------ dep/zlib/src/deflate.h | 349 - dep/zlib/src/gzclose.c | 25 - dep/zlib/src/gzguts.h | 220 - dep/zlib/src/gzlib.c | 637 -- dep/zlib/src/gzread.c | 654 -- dep/zlib/src/gzwrite.c | 665 -- dep/zlib/src/infback.c | 640 -- dep/zlib/src/inffast.c | 323 - dep/zlib/src/inffast.h | 11 - dep/zlib/src/inffixed.h | 94 - dep/zlib/src/inflate.c | 1561 ---- dep/zlib/src/inflate.h | 125 - dep/zlib/src/inftrees.c | 304 - dep/zlib/src/inftrees.h | 62 - dep/zlib/src/trees.c | 1203 ---- dep/zlib/src/trees.h | 128 - dep/zlib/src/uncompr.c | 93 - dep/zlib/src/zutil.c | 325 - dep/zlib/src/zutil.h | 271 - dep/zlib/zlib.vcxproj | 50 - dep/zlib/zlib.vcxproj.filters | 33 - dep/zstd/CMakeLists.txt | 38 - dep/zstd/LICENSE | 30 - dep/zstd/lib/common/bitstream.h | 478 -- dep/zstd/lib/common/compiler.h | 335 - dep/zstd/lib/common/cpu.h | 213 - dep/zstd/lib/common/debug.c | 24 - dep/zstd/lib/common/debug.h | 107 - dep/zstd/lib/common/entropy_common.c | 368 - dep/zstd/lib/common/error_private.c | 56 - dep/zstd/lib/common/error_private.h | 159 - dep/zstd/lib/common/fse.h | 717 -- dep/zstd/lib/common/fse_decompress.c | 403 -- dep/zstd/lib/common/huf.h | 364 - dep/zstd/lib/common/mem.h | 442 -- dep/zstd/lib/common/pool.c | 355 - dep/zstd/lib/common/pool.h | 84 - dep/zstd/lib/common/portability_macros.h | 137 - dep/zstd/lib/common/threading.c | 122 - dep/zstd/lib/common/threading.h | 155 - dep/zstd/lib/common/xxhash.c | 24 - dep/zstd/lib/common/xxhash.h | 5686 --------------- dep/zstd/lib/common/zstd_common.c | 83 - dep/zstd/lib/common/zstd_deps.h | 111 - dep/zstd/lib/common/zstd_internal.h | 493 -- dep/zstd/lib/common/zstd_trace.h | 163 - dep/zstd/lib/compress/clevels.h | 134 - dep/zstd/lib/compress/fse_compress.c | 741 -- dep/zstd/lib/compress/hist.c | 181 - dep/zstd/lib/compress/hist.h | 75 - dep/zstd/lib/compress/huf_compress.c | 1370 ---- dep/zstd/lib/compress/zstd_compress.c | 6327 ----------------- .../lib/compress/zstd_compress_internal.h | 1458 ---- .../lib/compress/zstd_compress_literals.c | 159 - .../lib/compress/zstd_compress_literals.h | 31 - .../lib/compress/zstd_compress_sequences.c | 442 -- .../lib/compress/zstd_compress_sequences.h | 54 - .../lib/compress/zstd_compress_superblock.c | 573 -- .../lib/compress/zstd_compress_superblock.h | 32 - dep/zstd/lib/compress/zstd_cwksp.h | 676 -- dep/zstd/lib/compress/zstd_double_fast.c | 696 -- dep/zstd/lib/compress/zstd_double_fast.h | 38 - dep/zstd/lib/compress/zstd_fast.c | 675 -- dep/zstd/lib/compress/zstd_fast.h | 37 - dep/zstd/lib/compress/zstd_lazy.c | 2104 ------ dep/zstd/lib/compress/zstd_lazy.h | 125 - dep/zstd/lib/compress/zstd_ldm.c | 724 -- dep/zstd/lib/compress/zstd_ldm.h | 117 - dep/zstd/lib/compress/zstd_ldm_geartab.h | 106 - dep/zstd/lib/compress/zstd_opt.c | 1446 ---- dep/zstd/lib/compress/zstd_opt.h | 56 - dep/zstd/lib/compress/zstdmt_compress.c | 1859 ----- dep/zstd/lib/compress/zstdmt_compress.h | 113 - dep/zstd/lib/decompress/huf_decompress.c | 1889 ----- .../lib/decompress/huf_decompress_amd64.S | 585 -- dep/zstd/lib/decompress/zstd_ddict.c | 244 - dep/zstd/lib/decompress/zstd_ddict.h | 44 - dep/zstd/lib/decompress/zstd_decompress.c | 2230 ------ .../lib/decompress/zstd_decompress_block.c | 2072 ------ .../lib/decompress/zstd_decompress_block.h | 68 - .../lib/decompress/zstd_decompress_internal.h | 236 - dep/zstd/lib/zdict.h | 452 -- dep/zstd/lib/zstd.h | 2575 ------- dep/zstd/lib/zstd_errors.h | 95 - dep/zstd/zstd.vcxproj | 76 - dep/zstd/zstd.vcxproj.filters | 189 - 198 files changed, 144411 deletions(-) delete mode 100644 dep/libjpeg/CMakeLists.txt delete mode 100644 dep/libjpeg/README delete mode 100644 dep/libjpeg/change.log delete mode 100644 dep/libjpeg/include/jconfig.h delete mode 100644 dep/libjpeg/include/jerror.h delete mode 100644 dep/libjpeg/include/jmorecfg.h delete mode 100644 dep/libjpeg/include/jpegint.h delete mode 100644 dep/libjpeg/include/jpeglib.h delete mode 100644 dep/libjpeg/libjpeg.vcxproj delete mode 100644 dep/libjpeg/libjpeg.vcxproj.filters delete mode 100644 dep/libjpeg/src/jaricom.c delete mode 100644 dep/libjpeg/src/jcapimin.c delete mode 100644 dep/libjpeg/src/jcapistd.c delete mode 100644 dep/libjpeg/src/jcarith.c delete mode 100644 dep/libjpeg/src/jccoefct.c delete mode 100644 dep/libjpeg/src/jccolor.c delete mode 100644 dep/libjpeg/src/jcdctmgr.c delete mode 100644 dep/libjpeg/src/jchuff.c delete mode 100644 dep/libjpeg/src/jcinit.c delete mode 100644 dep/libjpeg/src/jcmainct.c delete mode 100644 dep/libjpeg/src/jcmarker.c delete mode 100644 dep/libjpeg/src/jcmaster.c delete mode 100644 dep/libjpeg/src/jcomapi.c delete mode 100644 dep/libjpeg/src/jcparam.c delete mode 100644 dep/libjpeg/src/jcprepct.c delete mode 100644 dep/libjpeg/src/jcsample.c delete mode 100644 dep/libjpeg/src/jctrans.c delete mode 100644 dep/libjpeg/src/jdapimin.c delete mode 100644 dep/libjpeg/src/jdapistd.c delete mode 100644 dep/libjpeg/src/jdarith.c delete mode 100644 dep/libjpeg/src/jdatadst.c delete mode 100644 dep/libjpeg/src/jdatasrc.c delete mode 100644 dep/libjpeg/src/jdcoefct.c delete mode 100644 dep/libjpeg/src/jdcolor.c delete mode 100644 dep/libjpeg/src/jdct.h delete mode 100644 dep/libjpeg/src/jddctmgr.c delete mode 100644 dep/libjpeg/src/jdhuff.c delete mode 100644 dep/libjpeg/src/jdinput.c delete mode 100644 dep/libjpeg/src/jdmainct.c delete mode 100644 dep/libjpeg/src/jdmarker.c delete mode 100644 dep/libjpeg/src/jdmaster.c delete mode 100644 dep/libjpeg/src/jdmerge.c delete mode 100644 dep/libjpeg/src/jdpostct.c delete mode 100644 dep/libjpeg/src/jdsample.c delete mode 100644 dep/libjpeg/src/jdtrans.c delete mode 100644 dep/libjpeg/src/jerror.c delete mode 100644 dep/libjpeg/src/jfdctflt.c delete mode 100644 dep/libjpeg/src/jfdctfst.c delete mode 100644 dep/libjpeg/src/jfdctint.c delete mode 100644 dep/libjpeg/src/jidctflt.c delete mode 100644 dep/libjpeg/src/jidctfst.c delete mode 100644 dep/libjpeg/src/jidctint.c delete mode 100644 dep/libjpeg/src/jinclude.h delete mode 100644 dep/libjpeg/src/jmemmgr.c delete mode 100644 dep/libjpeg/src/jmemnobs.c delete mode 100644 dep/libjpeg/src/jmemsys.h delete mode 100644 dep/libjpeg/src/jquant1.c delete mode 100644 dep/libjpeg/src/jquant2.c delete mode 100644 dep/libjpeg/src/jutils.c delete mode 100644 dep/libjpeg/src/jversion.h delete mode 100644 dep/libjpeg/src/transupp.c delete mode 100644 dep/libjpeg/src/transupp.h delete mode 100644 dep/libpng/ANNOUNCE delete mode 100644 dep/libpng/AUTHORS delete mode 100644 dep/libpng/CHANGES delete mode 100644 dep/libpng/CMakeLists.txt delete mode 100644 dep/libpng/LICENSE delete mode 100644 dep/libpng/README delete mode 100644 dep/libpng/TODO delete mode 100644 dep/libpng/TRADEMARK delete mode 100644 dep/libpng/include/png.h delete mode 100644 dep/libpng/include/pngconf.h delete mode 100644 dep/libpng/include/pnglibconf.h delete mode 100644 dep/libpng/libpng.vcxproj delete mode 100644 dep/libpng/libpng.vcxproj.filters delete mode 100644 dep/libpng/src/arm/arm_init.c delete mode 100644 dep/libpng/src/arm/filter_neon.S delete mode 100644 dep/libpng/src/arm/filter_neon_intrinsics.c delete mode 100644 dep/libpng/src/arm/palette_neon_intrinsics.c delete mode 100644 dep/libpng/src/intel/filter_sse2_intrinsics.c delete mode 100644 dep/libpng/src/intel/intel_init.c delete mode 100644 dep/libpng/src/png.c delete mode 100644 dep/libpng/src/pngdebug.h delete mode 100644 dep/libpng/src/pngerror.c delete mode 100644 dep/libpng/src/pngget.c delete mode 100644 dep/libpng/src/pnginfo.h delete mode 100644 dep/libpng/src/pngmem.c delete mode 100644 dep/libpng/src/pngpread.c delete mode 100644 dep/libpng/src/pngpriv.h delete mode 100644 dep/libpng/src/pngread.c delete mode 100644 dep/libpng/src/pngrio.c delete mode 100644 dep/libpng/src/pngrtran.c delete mode 100644 dep/libpng/src/pngrutil.c delete mode 100644 dep/libpng/src/pngset.c delete mode 100644 dep/libpng/src/pngstruct.h delete mode 100644 dep/libpng/src/pngtrans.c delete mode 100644 dep/libpng/src/pngwio.c delete mode 100644 dep/libpng/src/pngwrite.c delete mode 100644 dep/libpng/src/pngwtran.c delete mode 100644 dep/libpng/src/pngwutil.c delete mode 100644 dep/zlib/CMakeLists.txt delete mode 100644 dep/zlib/ChangeLog delete mode 100644 dep/zlib/FAQ delete mode 100644 dep/zlib/README delete mode 100644 dep/zlib/include/zconf.h delete mode 100644 dep/zlib/include/zlib.h delete mode 100644 dep/zlib/src/adler32.c delete mode 100644 dep/zlib/src/compress.c delete mode 100644 dep/zlib/src/crc32.c delete mode 100644 dep/zlib/src/crc32.h delete mode 100644 dep/zlib/src/deflate.c delete mode 100644 dep/zlib/src/deflate.h delete mode 100644 dep/zlib/src/gzclose.c delete mode 100644 dep/zlib/src/gzguts.h delete mode 100644 dep/zlib/src/gzlib.c delete mode 100644 dep/zlib/src/gzread.c delete mode 100644 dep/zlib/src/gzwrite.c delete mode 100644 dep/zlib/src/infback.c delete mode 100644 dep/zlib/src/inffast.c delete mode 100644 dep/zlib/src/inffast.h delete mode 100644 dep/zlib/src/inffixed.h delete mode 100644 dep/zlib/src/inflate.c delete mode 100644 dep/zlib/src/inflate.h delete mode 100644 dep/zlib/src/inftrees.c delete mode 100644 dep/zlib/src/inftrees.h delete mode 100644 dep/zlib/src/trees.c delete mode 100644 dep/zlib/src/trees.h delete mode 100644 dep/zlib/src/uncompr.c delete mode 100644 dep/zlib/src/zutil.c delete mode 100644 dep/zlib/src/zutil.h delete mode 100644 dep/zlib/zlib.vcxproj delete mode 100644 dep/zlib/zlib.vcxproj.filters delete mode 100644 dep/zstd/CMakeLists.txt delete mode 100644 dep/zstd/LICENSE delete mode 100644 dep/zstd/lib/common/bitstream.h delete mode 100644 dep/zstd/lib/common/compiler.h delete mode 100644 dep/zstd/lib/common/cpu.h delete mode 100644 dep/zstd/lib/common/debug.c delete mode 100644 dep/zstd/lib/common/debug.h delete mode 100644 dep/zstd/lib/common/entropy_common.c delete mode 100644 dep/zstd/lib/common/error_private.c delete mode 100644 dep/zstd/lib/common/error_private.h delete mode 100644 dep/zstd/lib/common/fse.h delete mode 100644 dep/zstd/lib/common/fse_decompress.c delete mode 100644 dep/zstd/lib/common/huf.h delete mode 100644 dep/zstd/lib/common/mem.h delete mode 100644 dep/zstd/lib/common/pool.c delete mode 100644 dep/zstd/lib/common/pool.h delete mode 100644 dep/zstd/lib/common/portability_macros.h delete mode 100644 dep/zstd/lib/common/threading.c delete mode 100644 dep/zstd/lib/common/threading.h delete mode 100644 dep/zstd/lib/common/xxhash.c delete mode 100644 dep/zstd/lib/common/xxhash.h delete mode 100644 dep/zstd/lib/common/zstd_common.c delete mode 100644 dep/zstd/lib/common/zstd_deps.h delete mode 100644 dep/zstd/lib/common/zstd_internal.h delete mode 100644 dep/zstd/lib/common/zstd_trace.h delete mode 100644 dep/zstd/lib/compress/clevels.h delete mode 100644 dep/zstd/lib/compress/fse_compress.c delete mode 100644 dep/zstd/lib/compress/hist.c delete mode 100644 dep/zstd/lib/compress/hist.h delete mode 100644 dep/zstd/lib/compress/huf_compress.c delete mode 100644 dep/zstd/lib/compress/zstd_compress.c delete mode 100644 dep/zstd/lib/compress/zstd_compress_internal.h delete mode 100644 dep/zstd/lib/compress/zstd_compress_literals.c delete mode 100644 dep/zstd/lib/compress/zstd_compress_literals.h delete mode 100644 dep/zstd/lib/compress/zstd_compress_sequences.c delete mode 100644 dep/zstd/lib/compress/zstd_compress_sequences.h delete mode 100644 dep/zstd/lib/compress/zstd_compress_superblock.c delete mode 100644 dep/zstd/lib/compress/zstd_compress_superblock.h delete mode 100644 dep/zstd/lib/compress/zstd_cwksp.h delete mode 100644 dep/zstd/lib/compress/zstd_double_fast.c delete mode 100644 dep/zstd/lib/compress/zstd_double_fast.h delete mode 100644 dep/zstd/lib/compress/zstd_fast.c delete mode 100644 dep/zstd/lib/compress/zstd_fast.h delete mode 100644 dep/zstd/lib/compress/zstd_lazy.c delete mode 100644 dep/zstd/lib/compress/zstd_lazy.h delete mode 100644 dep/zstd/lib/compress/zstd_ldm.c delete mode 100644 dep/zstd/lib/compress/zstd_ldm.h delete mode 100644 dep/zstd/lib/compress/zstd_ldm_geartab.h delete mode 100644 dep/zstd/lib/compress/zstd_opt.c delete mode 100644 dep/zstd/lib/compress/zstd_opt.h delete mode 100644 dep/zstd/lib/compress/zstdmt_compress.c delete mode 100644 dep/zstd/lib/compress/zstdmt_compress.h delete mode 100644 dep/zstd/lib/decompress/huf_decompress.c delete mode 100644 dep/zstd/lib/decompress/huf_decompress_amd64.S delete mode 100644 dep/zstd/lib/decompress/zstd_ddict.c delete mode 100644 dep/zstd/lib/decompress/zstd_ddict.h delete mode 100644 dep/zstd/lib/decompress/zstd_decompress.c delete mode 100644 dep/zstd/lib/decompress/zstd_decompress_block.c delete mode 100644 dep/zstd/lib/decompress/zstd_decompress_block.h delete mode 100644 dep/zstd/lib/decompress/zstd_decompress_internal.h delete mode 100644 dep/zstd/lib/zdict.h delete mode 100644 dep/zstd/lib/zstd.h delete mode 100644 dep/zstd/lib/zstd_errors.h delete mode 100644 dep/zstd/zstd.vcxproj delete mode 100644 dep/zstd/zstd.vcxproj.filters diff --git a/dep/CMakeLists.txt b/dep/CMakeLists.txt index 438c8a10c..183c08d1a 100644 --- a/dep/CMakeLists.txt +++ b/dep/CMakeLists.txt @@ -30,14 +30,6 @@ disable_compiler_warnings_for_target(rcheevos) add_subdirectory(rapidyaml EXCLUDE_FROM_ALL) disable_compiler_warnings_for_target(rapidyaml) -# Build dependencies on Windows/Android. -if(WIN32 OR ANDROID) - add_subdirectory(zlib EXCLUDE_FROM_ALL) - add_subdirectory(zstd EXCLUDE_FROM_ALL) - add_subdirectory(libpng EXCLUDE_FROM_ALL) - add_subdirectory(libjpeg EXCLUDE_FROM_ALL) -endif() - if(ENABLE_CUBEB) add_subdirectory(cubeb EXCLUDE_FROM_ALL) disable_compiler_warnings_for_target(cubeb) diff --git a/dep/libjpeg/CMakeLists.txt b/dep/libjpeg/CMakeLists.txt deleted file mode 100644 index 3af958df4..000000000 --- a/dep/libjpeg/CMakeLists.txt +++ /dev/null @@ -1,65 +0,0 @@ -add_library(jpeg - include/jconfig.h - include/jerror.h - include/jmorecfg.h - include/jpegint.h - include/jpeglib.h - src/jaricom.c - src/jcapimin.c - src/jcapistd.c - src/jcarith.c - src/jccoefct.c - src/jccolor.c - src/jcdctmgr.c - src/jchuff.c - src/jcinit.c - src/jcmainct.c - src/jcmarker.c - src/jcmaster.c - src/jcomapi.c - src/jcparam.c - src/jcprepct.c - src/jcsample.c - src/jctrans.c - src/jdapimin.c - src/jdapistd.c - src/jdarith.c - src/jdatadst.c - src/jdatasrc.c - src/jdcoefct.c - src/jdcolor.c - src/jdct.h - src/jddctmgr.c - src/jdhuff.c - src/jdinput.c - src/jdmainct.c - src/jdmarker.c - src/jdmaster.c - src/jdmerge.c - src/jdpostct.c - src/jdsample.c - src/jdtrans.c - src/jerror.c - src/jfdctflt.c - src/jfdctfst.c - src/jfdctint.c - src/jidctflt.c - src/jidctfst.c - src/jidctint.c - src/jinclude.h - src/jmemmgr.c - src/jmemnobs.c - src/jmemsys.h - src/jquant1.c - src/jquant2.c - src/jutils.c - src/jversion.h - src/transupp.c - src/transupp.h -) - -target_include_directories(jpeg PUBLIC "${CMAKE_CURRENT_SOURCE_DIR}/include") -target_include_directories(jpeg PRIVATE "${CMAKE_CURRENT_SOURCE_DIR}/src") -disable_compiler_warnings_for_target(jpeg) - -add_library(JPEG::JPEG ALIAS jpeg) diff --git a/dep/libjpeg/README b/dep/libjpeg/README deleted file mode 100644 index 0f4ee1522..000000000 --- a/dep/libjpeg/README +++ /dev/null @@ -1,374 +0,0 @@ -The Independent JPEG Group's JPEG software -========================================== - -README for release 9f of 14-Jan-2024 -==================================== - -This distribution contains the ninth public release of the Independent JPEG -Group's free JPEG software. You are welcome to redistribute this software and -to use it for any purpose, subject to the conditions under LEGAL ISSUES, below. - -This software is the work of Tom Lane, Guido Vollbeding, Philip Gladstone, -Bill Allombert, Jim Boucher, Lee Crocker, Bob Friesenhahn, Ben Jackson, -John Korejwa, Julian Minguillon, Luis Ortiz, George Phillips, Davide Rossi, -Ge' Weijers, and other members of the Independent JPEG Group. - -IJG is not affiliated with the ISO/IEC JTC1/SC29/WG1 standards committee -(previously known as JPEG, together with ITU-T SG16). - - -DOCUMENTATION ROADMAP -===================== - -This file contains the following sections: - -OVERVIEW General description of JPEG and the IJG software. -LEGAL ISSUES Copyright, lack of warranty, terms of distribution. -REFERENCES Where to learn more about JPEG. -ARCHIVE LOCATIONS Where to find newer versions of this software. -ACKNOWLEDGMENTS Special thanks. -FILE FORMAT WARS Software *not* to get. -TO DO Plans for future IJG releases. - -Other documentation files in the distribution are: - -User documentation: - install.txt How to configure and install the IJG software. - usage.txt Usage instructions for cjpeg, djpeg, jpegtran, - rdjpgcom, and wrjpgcom. - *.1 Unix-style man pages for programs (same info as usage.txt). - wizard.txt Advanced usage instructions for JPEG wizards only. - cdaltui.txt Description of alternate user interface for cjpeg/djpeg. - change.log Version-to-version change highlights. -Programmer and internal documentation: - libjpeg.txt How to use the JPEG library in your own programs. - example.c Sample code for calling the JPEG library. - structure.txt Overview of the JPEG library's internal structure. - filelist.txt Road map of IJG files. - coderules.txt Coding style rules --- please read if you contribute code. - -Please read at least the files install.txt and usage.txt. Some information -can also be found in the JPEG FAQ (Frequently Asked Questions) article. See -ARCHIVE LOCATIONS below to find out where to obtain the FAQ article. - -If you want to understand how the JPEG code works, we suggest reading one or -more of the REFERENCES, then looking at the documentation files (in roughly -the order listed) before diving into the code. - - -OVERVIEW -======== - -This package contains C software to implement JPEG image encoding, decoding, -and transcoding. JPEG (pronounced "jay-peg") is a standardized compression -method for full-color and grayscale images. - -This software implements JPEG baseline, extended-sequential, and progressive -compression processes. Provision is made for supporting all variants of these -processes, although some uncommon parameter settings aren't implemented yet. -We have made no provision for supporting the hierarchical or lossless -processes defined in the standard. - -We provide a set of library routines for reading and writing JPEG image files, -plus two sample applications "cjpeg" and "djpeg", which use the library to -perform conversion between JPEG and some other popular image file formats. -The library is intended to be reused in other applications. - -In order to support file conversion and viewing software, we have included -considerable functionality beyond the bare JPEG coding/decoding capability; -for example, the color quantization modules are not strictly part of JPEG -decoding, but they are essential for output to colormapped file formats or -colormapped displays. These extra functions can be compiled out of the -library if not required for a particular application. - -We have also included "jpegtran", a utility for lossless transcoding between -different JPEG processes, and "rdjpgcom" and "wrjpgcom", two simple -applications for inserting and extracting textual comments in JFIF files. - -The emphasis in designing this software has been on achieving portability and -flexibility, while also making it fast enough to be useful. In particular, -the software is not intended to be read as a tutorial on JPEG. (See the -REFERENCES section for introductory material.) Rather, it is intended to -be reliable, portable, industrial-strength code. We do not claim to have -achieved that goal in every aspect of the software, but we strive for it. - -We welcome the use of this software as a component of commercial products. -No royalty is required, but we do ask for an acknowledgement in product -documentation, as described under LEGAL ISSUES. - - -LEGAL ISSUES -============ - -In plain English: - -1. We don't promise that this software works. (But if you find any bugs, - please let us know!) -2. You can use this software for whatever you want. You don't have to pay us. -3. You may not pretend that you wrote this software. If you use it in a - program, you must acknowledge somewhere in your documentation that - you've used the IJG code. - -In legalese: - -The authors make NO WARRANTY or representation, either express or implied, -with respect to this software, its quality, accuracy, merchantability, or -fitness for a particular purpose. This software is provided "AS IS", and you, -its user, assume the entire risk as to its quality and accuracy. - -This software is copyright (C) 1991-2024, Thomas G. Lane, Guido Vollbeding. -All Rights Reserved except as specified below. - -Permission is hereby granted to use, copy, modify, and distribute this -software (or portions thereof) for any purpose, without fee, subject to these -conditions: -(1) If any part of the source code for this software is distributed, then this -README file must be included, with this copyright and no-warranty notice -unaltered; and any additions, deletions, or changes to the original files -must be clearly indicated in accompanying documentation. -(2) If only executable code is distributed, then the accompanying -documentation must state that "this software is based in part on the work of -the Independent JPEG Group". -(3) Permission for use of this software is granted only if the user accepts -full responsibility for any undesirable consequences; the authors accept -NO LIABILITY for damages of any kind. - -These conditions apply to any software derived from or based on the IJG code, -not just to the unmodified library. If you use our work, you ought to -acknowledge us. - -Permission is NOT granted for the use of any IJG author's name or company name -in advertising or publicity relating to this software or products derived from -it. This software may be referred to only as "the Independent JPEG Group's -software". - -We specifically permit and encourage the use of this software as the basis of -commercial products, provided that all warranty or liability claims are -assumed by the product vendor. - - -The Unix configuration script "configure" was produced with GNU Autoconf. -It is copyright by the Free Software Foundation but is freely distributable. -The same holds for its supporting scripts (config.guess, config.sub, -ltmain.sh). Another support script, install-sh, is copyright by X Consortium -but is also freely distributable. - - -REFERENCES -========== - -We recommend reading one or more of these references before trying to -understand the innards of the JPEG software. - -The best short technical introduction to the JPEG compression algorithm is - Wallace, Gregory K. "The JPEG Still Picture Compression Standard", - Communications of the ACM, April 1991 (vol. 34 no. 4), pp. 30-44. -(Adjacent articles in that issue discuss MPEG motion picture compression, -applications of JPEG, and related topics.) If you don't have the CACM issue -handy, a PDF file containing a revised version of Wallace's article is -available at https://www.ijg.org/files/Wallace.JPEG.pdf. The file (actually -a preprint for an article that appeared in IEEE Trans. Consumer Electronics) -omits the sample images that appeared in CACM, but it includes corrections -and some added material. Note: the Wallace article is copyright ACM and IEEE, -and it may not be used for commercial purposes. - -A somewhat less technical, more leisurely introduction to JPEG can be found in -"The Data Compression Book" by Mark Nelson and Jean-loup Gailly, published by -M&T Books (New York), 2nd ed. 1996, ISBN 1-55851-434-1. This book provides -good explanations and example C code for a multitude of compression methods -including JPEG. It is an excellent source if you are comfortable reading C -code but don't know much about data compression in general. The book's JPEG -sample code is far from industrial-strength, but when you are ready to look -at a full implementation, you've got one here... - -The best currently available description of JPEG is the textbook "JPEG Still -Image Data Compression Standard" by William B. Pennebaker and Joan L. -Mitchell, published by Van Nostrand Reinhold, 1993, ISBN 0-442-01272-1. -Price US$59.95, 638 pp. The book includes the complete text of the ISO JPEG -standards (DIS 10918-1 and draft DIS 10918-2). -Although this is by far the most detailed and comprehensive exposition of -JPEG publicly available, we point out that it is still missing an explanation -of the most essential properties and algorithms of the underlying DCT -technology. -If you think that you know about DCT-based JPEG after reading this book, -then you are in delusion. The real fundamentals and corresponding potential -of DCT-based JPEG are not publicly known so far, and that is the reason for -all the mistaken developments taking place in the image coding domain. - -The original JPEG standard is divided into two parts, Part 1 being the actual -specification, while Part 2 covers compliance testing methods. Part 1 is -titled "Digital Compression and Coding of Continuous-tone Still Images, -Part 1: Requirements and guidelines" and has document numbers ISO/IEC IS -10918-1, ITU-T T.81. Part 2 is titled "Digital Compression and Coding of -Continuous-tone Still Images, Part 2: Compliance testing" and has document -numbers ISO/IEC IS 10918-2, ITU-T T.83. -IJG JPEG 8 introduced an implementation of the JPEG SmartScale extension -which is specified in two documents: A contributed document at ITU and ISO -with title "ITU-T JPEG-Plus Proposal for Extending ITU-T T.81 for Advanced -Image Coding", April 2006, Geneva, Switzerland. The latest version of this -document is Revision 3. And a contributed document ISO/IEC JTC1/SC29/WG1 N -5799 with title "Evolution of JPEG", June/July 2011, Berlin, Germany. -IJG JPEG 9 introduces a reversible color transform for improved lossless -compression which is described in a contributed document ISO/IEC JTC1/SC29/ -WG1 N 6080 with title "JPEG 9 Lossless Coding", June/July 2012, Paris, France. - -The JPEG standard does not specify all details of an interchangeable file -format. For the omitted details we follow the "JFIF" conventions, version 2. -JFIF version 1 has been adopted as Recommendation ITU-T T.871 (05/2011) : -Information technology - Digital compression and coding of continuous-tone -still images: JPEG File Interchange Format (JFIF). It is available as a -free download in PDF file format from https://www.itu.int/rec/T-REC-T.871. -A PDF file of the older JFIF document is available at -https://www.w3.org/Graphics/JPEG/jfif3.pdf. - -The TIFF 6.0 file format specification can be obtained by FTP from -ftp://ftp.sgi.com/graphics/tiff/TIFF6.ps.gz. The JPEG incorporation scheme -found in the TIFF 6.0 spec of 3-June-92 has a number of serious problems. -IJG does not recommend use of the TIFF 6.0 design (TIFF Compression tag 6). -Instead, we recommend the JPEG design proposed by TIFF Technical Note #2 -(Compression tag 7). Copies of this Note can be obtained from -https://www.ijg.org/files/. It is expected that the next revision -of the TIFF spec will replace the 6.0 JPEG design with the Note's design. -Although IJG's own code does not support TIFF/JPEG, the free libtiff library -uses our library to implement TIFF/JPEG per the Note. - - -ARCHIVE LOCATIONS -================= - -The "official" archive site for this software is www.ijg.org. -The most recent released version can always be found there in -directory "files". This particular version will be archived -in Windows-compatible "zip" archive format as -https://www.ijg.org/files/jpegsr9f.zip, and -in Unix-compatible "tar.gz" archive format as -https://www.ijg.org/files/jpegsrc.v9f.tar.gz. - -The JPEG FAQ (Frequently Asked Questions) article is a source of some -general information about JPEG. -It is available on the World Wide Web at http://www.faqs.org/faqs/jpeg-faq/ -and other news.answers archive sites, including the official news.answers -archive at rtfm.mit.edu: ftp://rtfm.mit.edu/pub/usenet/news.answers/jpeg-faq/. -If you don't have Web or FTP access, send e-mail to mail-server@rtfm.mit.edu -with body - send usenet/news.answers/jpeg-faq/part1 - send usenet/news.answers/jpeg-faq/part2 - - -ACKNOWLEDGMENTS -=============== - -Thank to Juergen Bruder for providing me with a copy of the common DCT -algorithm article, only to find out that I had come to the same result -in a more direct and comprehensible way with a more generative approach. - -Thank to Istvan Sebestyen and Joan L. Mitchell for inviting me to the -ITU JPEG (Study Group 16) meeting in Geneva, Switzerland. - -Thank to Thomas Wiegand and Gary Sullivan for inviting me to the -Joint Video Team (MPEG & ITU) meeting in Geneva, Switzerland. - -Thank to Thomas Richter and Daniel Lee for inviting me to the -ISO/IEC JTC1/SC29/WG1 (previously known as JPEG, together with ITU-T SG16) -meeting in Berlin, Germany. - -Thank to John Korejwa and Massimo Ballerini for inviting me to -fruitful consultations in Boston, MA and Milan, Italy. - -Thank to Hendrik Elstner, Roland Fassauer, Simone Zuck, Guenther -Maier-Gerber, Walter Stoeber, Fred Schmitz, and Norbert Braunagel -for corresponding business development. - -Thank to Nico Zschach and Dirk Stelling of the technical support team -at the Digital Images company in Halle for providing me with extra -equipment for configuration tests. - -Thank to Richard F. Lyon (then of Foveon Inc.) for fruitful -communication about JPEG configuration in Sigma Photo Pro software. - -Thank to Andrew Finkenstadt for hosting the ijg.org site. - -Thank to Thomas G. Lane for the original design and development -of this singular software package. - -Thank to Lars Goehler, Andreas Heinecke, Sebastian Fuss, -Yvonne Roebert, Andrej Werner, Ulf-Dietrich Braumann, -and Nina Ssymank for support and public relations. - - -FILE FORMAT WARS -================ - -The ISO/IEC JTC1/SC29/WG1 standards committee (previously known as JPEG, -together with ITU-T SG16) currently promotes different formats containing -the name "JPEG" which is misleading because these formats are incompatible -with original DCT-based JPEG and are based on faulty technologies. -IJG therefore does not and will not support such momentary mistakes -(see REFERENCES). -There exist also distributions under the name "OpenJPEG" promoting such -kind of formats which is misleading because they don't support original -JPEG images. -We have no sympathy for the promotion of inferior formats. Indeed, one of -the original reasons for developing this free software was to help force -convergence on common, interoperable format standards for JPEG files. -Don't use an incompatible file format! -(In any case, our decoder will remain capable of reading existing JPEG -image files indefinitely.) - -The ISO committee pretends to be "responsible for the popular JPEG" in their -public reports which is not true because they don't respond to actual -requirements for the maintenance of the original JPEG specification. -Furthermore, the ISO committee pretends to "ensure interoperability" with -their standards which is not true because their "standards" support only -application-specific and proprietary use cases and contain mathematically -incorrect code. - -There are currently different distributions in circulation containing the -name "libjpeg" which is misleading because they don't have the features and -are incompatible with formats supported by actual IJG libjpeg distributions. -One of those fakes is released by members of the ISO committee and just uses -the name of libjpeg for misdirection of people, similar to the abuse of the -name JPEG as described above, while having nothing in common with actual IJG -libjpeg distributions and containing mathematically incorrect code. -The other one claims to be a "derivative" or "fork" of the original libjpeg, -but violates the license conditions as described under LEGAL ISSUES above -and violates basic C programming properties. -We have no sympathy for the release of misleading, incorrect and illegal -distributions derived from obsolete code bases. -Don't use an obsolete code base! - -According to the UCC (Uniform Commercial Code) law, IJG has the lawful and -legal right to foreclose on certain standardization bodies and other -institutions or corporations that knowingly perform substantial and -systematic deceptive acts and practices, fraud, theft, and damaging of the -value of the people of this planet without their knowing, willing and -intentional consent. -The titles, ownership, and rights of these institutions and all their assets -are now duly secured and held in trust for the free people of this planet. -People of the planet, on every country, may have a financial interest in -the assets of these former principals, agents, and beneficiaries of the -foreclosed institutions and corporations. -IJG asserts what is: that each man, woman, and child has unalienable value -and rights granted and deposited in them by the Creator and not any one of -the people is subordinate to any artificial principality, corporate fiction -or the special interest of another without their appropriate knowing, -willing and intentional consent made by contract or accommodation agreement. -IJG expresses that which already was. -The people have already determined and demanded that public administration -entities, national governments, and their supporting judicial systems must -be fully transparent, accountable, and liable. -IJG has secured the value for all concerned free people of the planet. - -A partial list of foreclosed institutions and corporations ("Hall of Shame") -is currently prepared and will be published later. - - -TO DO -===== - -Version 9 is the second release of a new generation JPEG standard -to overcome the limitations of the original JPEG specification, -and is the first true source reference JPEG codec. -More features are being prepared for coming releases... - -Please send bug reports, offers of help, etc. to jpeg-info@ijg.org. diff --git a/dep/libjpeg/change.log b/dep/libjpeg/change.log deleted file mode 100644 index eae011312..000000000 --- a/dep/libjpeg/change.log +++ /dev/null @@ -1,527 +0,0 @@ -CHANGE LOG for Independent JPEG Group's JPEG software - - -Version 9f 14-Jan-2024 ------------------------ - -Add build system for C++Builder/RAD Studio. - -Add build system for Xcode (beside configure). - -Add ARM64EC (Emulation Compatible) platform support in the -Visual Studio build. - - -Version 9e 16-Jan-2022 ------------------------ - -Include alternate user interface files for cjpeg/djpeg. - -jcparam.c: change default chrominance DC quantization factor -for lossless support. Note: Requires rebuild of test images. - -rdgif.c, cderror.h: add sanity check for GIF image dimensions. -Thank to Casper Sun for cjpeg potential vulnerability report. - -Add ARM and ARM64 platform support in the Visual Studio build. - - -Version 9d 12-Jan-2020 ------------------------ - -Optimize the optimal Huffman code table generation to produce -slightly smaller files. Thank to John Korejwa for suggestion. -Note: Requires rebuild of testimgp.jpg. - -Decoding Huffman: Use default tables if tables are not defined. -Thank to Simone Azzalin for report (Motion JPEG), -and to Martin Strunz for hint. - -Add sanity check in optimal Huffman code table generation. -Thank to Adam Farley for suggestion. - -rdtarga.c: use read_byte(), with EOF check, instead of getc() -in read_*_pixel(). -Thank to Chijin Zhou for cjpeg potential vulnerability report. - -jmemnobs.c: respect the max_memory_to_use setting in -jpeg_mem_available() computation. Thank to Sheng Shu and -Dongdong She for djpeg potential vulnerability report. - -jdarith.c, jdhuff.c: avoid left shift of negative value -compiler warning in decode_mcu_AC_refine(). -Thank to Indu Bhagat for suggestion. - -Add x64 (64-bit) platform support, avoid compiler warnings. -Thank to Jonathan Potter, Feiyun Wang, and Sheng Shu for suggestion. - -Adjust libjpeg version specification for pkg-config file. -Thank to Chen Chen for suggestion. - -Restore GIF read and write support from libjpeg version 6a. -Thank to Wolfgang Werner (W.W.) Heinz for suggestion. - -Improve consistency in raw (downsampled) image data processing mode. -Thank to Zhongyuan Zhou for hint. - -Avoid out of bounds array read (AC derived table pointers) -in start pass in jdhuff.c. Thank to Peng Li for report. - -Improve code sanity (jdhuff.c). -Thank to Reza Mirzazade farkhani for reports. - -Add jpegtran -drop option; add options to the crop extension and wipe -to fill the extra area with content from the source image region, -instead of gray out. - - -Version 9c 14-Jan-2018 ------------------------ - -jpegtran: add an option to the -wipe switch to fill the region -with the average of adjacent blocks, instead of gray out. -Thank to Caitlyn Feddock and Maddie Ziegler for inspiration. - -Make range extension bits adjustable (in jpegint.h). -Thank to Robin Watts for suggestion. - -Provide macros for fflush() and ferror() in jinclude.h in order -to facilitate adaption by applications using an own FILE class. -Thank to Gerhard Huber for suggestion. - -Add libjpeg pkg-config file. Thank to Mark Lavi, Vincent Torri, -Patrick McMunn, and Huw Davies for suggestion. - -Add sanity checks in cjpeg image reader modules. -Thank to Bingchang, Liu for reports. - - -Version 9b 17-Jan-2016 ------------------------ - -Improvements and optimizations in DCT and color calculations. -Normalize range limit array composition and access pattern. -Thank to Sia Furler and Maddie Ziegler for inspiration. - -Use merged upsample with scaled DCT sizes larger than 8. -Thank to Taylor Hatala for inspiration. - -Check for excessive comment lengths in argument parsing in wrjpgcom.c. -Thank to Julian Cohen for hint. - -Add makefile.b32 for use with Borland C++ 32-bit (bcc32). -Thank to Joe Slater for contribution. - -Document 'f' specifier for jpegtran -crop specification. -Thank to Michele Martone for suggestion. - -Use defined value from header instead of hardwired number in rdswitch.c. -Thank to Robert Sprowson for hint. - - -Version 9a 19-Jan-2014 ------------------------ - -Add support for wide gamut color spaces (JFIF version 2). -Improve clarity and accuracy in color conversion modules. -Note: Requires rebuild of test images. - -Extend the bit depth support to all values from 8 to 12 -(BITS_IN_JSAMPLE configuration option in jmorecfg.h). -jpegtran now supports N bits sample data precision with all N from 8 to 12 -in a single instance. Thank to Roland Fassauer for inspiration. - -Try to resolve issues with new boolean type definition. -Thank also to v4hn for suggestion. - -Enable option to use default Huffman tables for lossless compression -(for hardware solution), and in this case improve lossless RGB compression -with reversible color transform. Thank to Benny Alexandar for hint. - -Extend the entropy decoding structure, so that extraneous bytes between -compressed scan data and following marker can be reported correctly. -Thank to Nigel Tao for hint. - -Add jpegtran -wipe option and extension for -crop. -Thank to Andrew Senior, David Clunie, and Josef Schmid for suggestion. - - -Version 9 13-Jan-2013 ----------------------- - -Add cjpeg -rgb1 option to create an RGB JPEG file, and insert -a simple reversible color transform into the processing which -significantly improves the compression. -The recommended command for lossless coding of RGB images is now -cjpeg -rgb1 -block 1 -arithmetic. -As said, this option improves the compression significantly, but -the files are not compatible with JPEG decoders prior to IJG v9 -due to the included color transform. -The used color transform and marker signaling is compatible with -other JPEG standards (e.g., JPEG-LS part 2). - -Remove the automatic de-ANSI-fication support (Automake 1.12). -Thank also to Nitin A Kamble for suggestion. - -Add remark for jpeg_mem_dest() in jdatadst.c. -Thank to Elie-Gregoire Khoury for the hint. - -Support files with invalid component identifiers (created -by Adobe PDF). Thank to Robin Watts for the suggestion. - -Adapt full buffer case in jcmainct.c for use with scaled DCT. -Thank to Sergii Biloshytskyi for the suggestion. - -Add type identifier for declaration of noreturn functions. -Thank to Brett L. Moore for the suggestion. - -Correct argument type in format string, avoid compiler warnings. -Thank to Vincent Torri for hint. - -Add missing #include directives in configuration checks, avoid -configuration errors. Thank to John Spencer for the hint. - - -Version 8d 15-Jan-2012 ------------------------ - -Add cjpeg -rgb option to create RGB JPEG files. -Using this switch suppresses the conversion from RGB -colorspace input to the default YCbCr JPEG colorspace. -This feature allows true lossless JPEG coding of RGB color images. -The recommended command for this purpose is currently -cjpeg -rgb -block 1 -arithmetic. -SmartScale capable decoder (introduced with IJG JPEG 8) required. -Thank to Michael Koch for the initial suggestion. - -Add option to disable the region adjustment in the transupp crop code. -Thank to Jeffrey Friedl for the suggestion. - -Thank to Richard Jones and Edd Dawson for various minor corrections. - -Thank to Akim Demaille for configure.ac cleanup. - - -Version 8c 16-Jan-2011 ------------------------ - -Add option to compression library and cjpeg (-block N) to use -different DCT block size. -All N from 1 to 16 are possible. Default is 8 (baseline format). -Larger values produce higher compression, -smaller values produce higher quality. -SmartScale capable decoder (introduced with IJG JPEG 8) required. - - -Version 8b 16-May-2010 ------------------------ - -Repair problem in new memory source manager with corrupt JPEG data. -Thank to Ted Campbell and Samuel Chun for the report. - -Repair problem in Makefile.am test target. -Thank to anonymous user for the report. - -Support MinGW installation with automatic configure. -Thank to Volker Grabsch for the suggestion. - - -Version 8a 28-Feb-2010 ------------------------ - -Writing tables-only datastreams via jpeg_write_tables works again. - -Support 32-bit BMPs (RGB image with Alpha channel) for read in cjpeg. -Thank to Brett Blackham for the suggestion. - -Improve accuracy in floating point IDCT calculation. -Thank to Robert Hooke for the hint. - - -Version 8 10-Jan-2010 ----------------------- - -jpegtran now supports the same -scale option as djpeg for "lossless" resize. -An implementation of the JPEG SmartScale extension is required for this -feature. A (draft) specification of the JPEG SmartScale extension is -available as a contributed document at ITU and ISO. Revision 2 or later -of the document is required (latest document version is Revision 3). -The SmartScale extension will enable more features beside lossless resize -in future implementations, as described in the document (new compression -options). - -Add sanity check in BMP reader module to avoid cjpeg crash for empty input -image (thank to Isaev Ildar of ISP RAS, Moscow, RU for reporting this error). - -Add data source and destination managers for read from and write to -memory buffers. New API functions jpeg_mem_src and jpeg_mem_dest. -Thank to Roberto Boni from Italy for the suggestion. - - -Version 7 27-Jun-2009 ----------------------- - -New scaled DCTs implemented. -djpeg now supports scalings N/8 with all N from 1 to 16. -cjpeg now supports scalings 8/N with all N from 1 to 16. -Scaled DCTs with size larger than 8 are now also used for resolving the -common 2x2 chroma subsampling case without additional spatial resampling. -Separate spatial resampling for those kind of files is now only necessary -for N>8 scaling cases. -Furthermore, separate scaled DCT functions are provided for direct resolving -of the common asymmetric subsampling cases (2x1 and 1x2) without additional -spatial resampling. - -cjpeg -quality option has been extended for support of separate quality -settings for luminance and chrominance (or in general, for every provided -quantization table slot). -New API function jpeg_default_qtables() and q_scale_factor array in library. - -Added -nosmooth option to cjpeg, complementary to djpeg. -New variable "do_fancy_downsampling" in library, complement to fancy -upsampling. Fancy upsampling now uses direct DCT scaling with sizes -larger than 8. The old method is not reversible and has been removed. - -Support arithmetic entropy encoding and decoding. -Added files jaricom.c, jcarith.c, jdarith.c. - -Straighten the file structure: -Removed files jidctred.c, jcphuff.c, jchuff.h, jdphuff.c, jdhuff.h. - -jpegtran has a new "lossless" cropping feature. - -Implement -perfect option in jpegtran, new API function -jtransform_perfect_transform() in transupp. (DP 204_perfect.dpatch) - -Better error messages for jpegtran fopen failure. -(DP 203_jpegtran_errmsg.dpatch) - -Fix byte order issue with 16bit PPM/PGM files in rdppm.c/wrppm.c: -according to Netpbm, the de facto standard implementation of the PNM formats, -the most significant byte is first. (DP 203_rdppm.dpatch) - -Add -raw option to rdjpgcom not to mangle the output. -(DP 205_rdjpgcom_raw.dpatch) - -Make rdjpgcom locale aware. (DP 201_rdjpgcom_locale.dpatch) - -Add extern "C" to jpeglib.h. -This avoids the need to put extern "C" { ... } around #include "jpeglib.h" -in your C++ application. Defining the symbol DONT_USE_EXTERN_C in the -configuration prevents this. (DP 202_jpeglib.h_c++.dpatch) - - -Version 6b 27-Mar-1998 ------------------------ - -jpegtran has new features for lossless image transformations (rotation -and flipping) as well as "lossless" reduction to grayscale. - -jpegtran now copies comments by default; it has a -copy switch to enable -copying all APPn blocks as well, or to suppress comments. (Formerly it -always suppressed comments and APPn blocks.) jpegtran now also preserves -JFIF version and resolution information. - -New decompressor library feature: COM and APPn markers found in the input -file can be saved in memory for later use by the application. (Before, -you had to code this up yourself with a custom marker processor.) - -There is an unused field "void * client_data" now in compress and decompress -parameter structs; this may be useful in some applications. - -JFIF version number information is now saved by the decoder and accepted by -the encoder. jpegtran uses this to copy the source file's version number, -to ensure "jpegtran -copy all" won't create bogus files that contain JFXX -extensions but claim to be version 1.01. Applications that generate their -own JFXX extension markers also (finally) have a supported way to cause the -encoder to emit JFIF version number 1.02. - -djpeg's trace mode reports JFIF 1.02 thumbnail images as such, rather -than as unknown APP0 markers. - -In -verbose mode, djpeg and rdjpgcom will try to print the contents of -APP12 markers as text. Some digital cameras store useful text information -in APP12 markers. - -Handling of truncated data streams is more robust: blocks beyond the one in -which the error occurs will be output as uniform gray, or left unchanged -if decoding a progressive JPEG. The appearance no longer depends on the -Huffman tables being used. - -Huffman tables are checked for validity much more carefully than before. - -To avoid the Unisys LZW patent, djpeg's GIF output capability has been -changed to produce "uncompressed GIFs", and cjpeg's GIF input capability -has been removed altogether. We're not happy about it either, but there -seems to be no good alternative. - -The configure script now supports building libjpeg as a shared library -on many flavors of Unix (all the ones that GNU libtool knows how to -build shared libraries for). Use "./configure --enable-shared" to -try this out. - -New jconfig file and makefiles for Microsoft Visual C++ and Developer Studio. -Also, a jconfig file and a build script for Metrowerks CodeWarrior -on Apple Macintosh. makefile.dj has been updated for DJGPP v2, and there -are miscellaneous other minor improvements in the makefiles. - -jmemmac.c now knows how to create temporary files following Mac System 7 -conventions. - -djpeg's -map switch is now able to read raw-format PPM files reliably. - -cjpeg -progressive -restart no longer generates any unnecessary DRI markers. - -Multiple calls to jpeg_simple_progression for a single JPEG object -no longer leak memory. - - -Version 6a 7-Feb-96 --------------------- - -Library initialization sequence modified to detect version mismatches -and struct field packing mismatches between library and calling application. -This change requires applications to be recompiled, but does not require -any application source code change. - -All routine declarations changed to the style "GLOBAL(type) name ...", -that is, GLOBAL, LOCAL, METHODDEF, EXTERN are now macros taking the -routine's return type as an argument. This makes it possible to add -Microsoft-style linkage keywords to all the routines by changing just -these macros. Note that any application code that was using these macros -will have to be changed. - -DCT coefficient quantization tables are now stored in normal array order -rather than zigzag order. Application code that calls jpeg_add_quant_table, -or otherwise manipulates quantization tables directly, will need to be -changed. If you need to make such code work with either older or newer -versions of the library, a test like "#if JPEG_LIB_VERSION >= 61" is -recommended. - -djpeg's trace capability now dumps DQT tables in natural order, not zigzag -order. This allows the trace output to be made into a "-qtables" file -more easily. - -New system-dependent memory manager module for use on Apple Macintosh. - -Fix bug in cjpeg's -smooth option: last one or two scanlines would be -duplicates of the prior line unless the image height mod 16 was 1 or 2. - -Repair minor problems in VMS, BCC, MC6 makefiles. - -New configure script based on latest GNU Autoconf. - -Correct the list of include files needed by MetroWerks C for ccommand(). - -Numerous small documentation updates. - - -Version 6 2-Aug-95 -------------------- - -Progressive JPEG support: library can read and write full progressive JPEG -files. A "buffered image" mode supports incremental decoding for on-the-fly -display of progressive images. Simply recompiling an existing IJG-v5-based -decoder with v6 should allow it to read progressive files, though of course -without any special progressive display. - -New "jpegtran" application performs lossless transcoding between different -JPEG formats; primarily, it can be used to convert baseline to progressive -JPEG and vice versa. In support of jpegtran, the library now allows lossless -reading and writing of JPEG files as DCT coefficient arrays. This ability -may be of use in other applications. - -Notes for programmers: -* We changed jpeg_start_decompress() to be able to suspend; this makes all -decoding modes available to suspending-input applications. However, -existing applications that use suspending input will need to be changed -to check the return value from jpeg_start_decompress(). You don't need to -do anything if you don't use a suspending data source. -* We changed the interface to the virtual array routines: access_virt_array -routines now take a count of the number of rows to access this time. The -last parameter to request_virt_array routines is now interpreted as the -maximum number of rows that may be accessed at once, but not necessarily -the height of every access. - - -Version 5b 15-Mar-95 ---------------------- - -Correct bugs with grayscale images having v_samp_factor > 1. - -jpeg_write_raw_data() now supports output suspension. - -Correct bugs in "configure" script for case of compiling in -a directory other than the one containing the source files. - -Repair bug in jquant1.c: sometimes didn't use as many colors as it could. - -Borland C makefile and jconfig file work under either MS-DOS or OS/2. - -Miscellaneous improvements to documentation. - - -Version 5a 7-Dec-94 --------------------- - -Changed color conversion roundoff behavior so that grayscale values are -represented exactly. (This causes test image files to change.) - -Make ordered dither use 16x16 instead of 4x4 pattern for a small quality -improvement. - -New configure script based on latest GNU Autoconf. -Fix configure script to handle CFLAGS correctly. -Rename *.auto files to *.cfg, so that configure script still works if -file names have been truncated for DOS. - -Fix bug in rdbmp.c: didn't allow for extra data between header and image. - -Modify rdppm.c/wrppm.c to handle 2-byte raw PPM/PGM formats for 12-bit data. - -Fix several bugs in rdrle.c. - -NEED_SHORT_EXTERNAL_NAMES option was broken. - -Revise jerror.h/jerror.c for more flexibility in message table. - -Repair oversight in jmemname.c NO_MKTEMP case: file could be there -but unreadable. - - -Version 5 24-Sep-94 --------------------- - -Version 5 represents a nearly complete redesign and rewrite of the IJG -software. Major user-visible changes include: - * Automatic configuration simplifies installation for most Unix systems. - * A range of speed vs. image quality tradeoffs are supported. - This includes resizing of an image during decompression: scaling down - by a factor of 1/2, 1/4, or 1/8 is handled very efficiently. - * New programs rdjpgcom and wrjpgcom allow insertion and extraction - of text comments in a JPEG file. - -The application programmer's interface to the library has changed completely. -Notable improvements include: - * We have eliminated the use of callback routines for handling the - uncompressed image data. The application now sees the library as a - set of routines that it calls to read or write image data on a - scanline-by-scanline basis. - * The application image data is represented in a conventional interleaved- - pixel format, rather than as a separate array for each color channel. - This can save a copying step in many programs. - * The handling of compressed data has been cleaned up: the application can - supply routines to source or sink the compressed data. It is possible to - suspend processing on source/sink buffer overrun, although this is not - supported in all operating modes. - * All static state has been eliminated from the library, so that multiple - instances of compression or decompression can be active concurrently. - * JPEG abbreviated datastream formats are supported, ie, quantization and - Huffman tables can be stored separately from the image data. - * And not only that, but the documentation of the library has improved - considerably! - - -The last widely used release before the version 5 rewrite was version 4A of -18-Feb-93. Change logs before that point have been discarded, since they -are not of much interest after the rewrite. diff --git a/dep/libjpeg/include/jconfig.h b/dep/libjpeg/include/jconfig.h deleted file mode 100644 index d1710ae7d..000000000 --- a/dep/libjpeg/include/jconfig.h +++ /dev/null @@ -1,171 +0,0 @@ -/* - * jconfig.txt - * - * Copyright (C) 1991-1994, Thomas G. Lane. - * Modified 2009-2013 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file documents the configuration options that are required to - * customize the JPEG software for a particular system. - * - * The actual configuration options for a particular installation are stored - * in jconfig.h. On many machines, jconfig.h can be generated automatically - * or copied from one of the "canned" jconfig files that we supply. But if - * you need to generate a jconfig.h file by hand, this file tells you how. - * - * DO NOT EDIT THIS FILE --- IT WON'T ACCOMPLISH ANYTHING. - * EDIT A COPY NAMED JCONFIG.H. - */ - - -/* - * These symbols indicate the properties of your machine or compiler. - * #define the symbol if yes, #undef it if no. - */ - -/* Does your compiler support function prototypes? - * (If not, you also need to use ansi2knr, see install.txt) - */ -#define HAVE_PROTOTYPES - -/* Does your compiler support the declaration "unsigned char" ? - * How about "unsigned short" ? - */ -#define HAVE_UNSIGNED_CHAR -#define HAVE_UNSIGNED_SHORT - -/* Define "void" as "char" if your compiler doesn't know about type void. - * NOTE: be sure to define void such that "void *" represents the most general - * pointer type, e.g., that returned by malloc(). - */ -/* #define void char */ - -/* Define "const" as empty if your compiler doesn't know the "const" keyword. - */ -/* #define const */ - -/* Define this if an ordinary "char" type is unsigned. - * If you're not sure, leaving it undefined will work at some cost in speed. - * If you defined HAVE_UNSIGNED_CHAR then the speed difference is minimal. - */ -#undef CHAR_IS_UNSIGNED - -/* Define this if your system has an ANSI-conforming file. - */ -#define HAVE_STDDEF_H - -/* Define this if your system has an ANSI-conforming file. - */ -#define HAVE_STDLIB_H - -/* Define this if your system does not have an ANSI/SysV , - * but does have a BSD-style . - */ -#undef NEED_BSD_STRINGS - -/* Define this if your system does not provide typedef size_t in any of the - * ANSI-standard places (stddef.h, stdlib.h, or stdio.h), but places it in - * instead. - */ -#undef NEED_SYS_TYPES_H - -/* For 80x86 machines, you need to define NEED_FAR_POINTERS, - * unless you are using a large-data memory model or 80386 flat-memory mode. - * On less brain-damaged CPUs this symbol must not be defined. - * (Defining this symbol causes large data structures to be referenced through - * "far" pointers and to be allocated with a special version of malloc.) - */ -#undef NEED_FAR_POINTERS - -/* Define this if your linker needs global names to be unique in less - * than the first 15 characters. - */ -#undef NEED_SHORT_EXTERNAL_NAMES - -/* Although a real ANSI C compiler can deal perfectly well with pointers to - * unspecified structures (see "incomplete types" in the spec), a few pre-ANSI - * and pseudo-ANSI compilers get confused. To keep one of these bozos happy, - * define INCOMPLETE_TYPES_BROKEN. This is not recommended unless you - * actually get "missing structure definition" warnings or errors while - * compiling the JPEG code. - */ -#undef INCOMPLETE_TYPES_BROKEN - -/* Define "boolean" as unsigned char, not enum, on Windows systems. - */ -#ifdef _WIN32 -#ifndef __RPCNDR_H__ /* don't conflict if rpcndr.h already read */ -typedef unsigned char boolean; -#endif -#ifndef FALSE /* in case these macros already exist */ -#define FALSE 0 /* values of boolean */ -#endif -#ifndef TRUE -#define TRUE 1 -#endif -#define HAVE_BOOLEAN /* prevent jmorecfg.h from redefining it */ -#endif - - -/* - * The following options affect code selection within the JPEG library, - * but they don't need to be visible to applications using the library. - * To minimize application namespace pollution, the symbols won't be - * defined unless JPEG_INTERNALS has been defined. - */ - -#ifdef JPEG_INTERNALS - -/* Define this if your compiler implements ">>" on signed values as a logical - * (unsigned) shift; leave it undefined if ">>" is a signed (arithmetic) shift, - * which is the normal and rational definition. - */ -#undef RIGHT_SHIFT_IS_UNSIGNED - - -#endif /* JPEG_INTERNALS */ - - -/* - * The remaining options do not affect the JPEG library proper, - * but only the sample applications cjpeg/djpeg (see cjpeg.c, djpeg.c). - * Other applications can ignore these. - */ - -#ifdef JPEG_CJPEG_DJPEG - -/* These defines indicate which image (non-JPEG) file formats are allowed. */ - -#define BMP_SUPPORTED /* BMP image file format */ -#define GIF_SUPPORTED /* GIF image file format */ -#define PPM_SUPPORTED /* PBMPLUS PPM/PGM image file format */ -#undef RLE_SUPPORTED /* Utah RLE image file format */ -#define TARGA_SUPPORTED /* Targa image file format */ - -/* Define this if you want to name both input and output files on the command - * line, rather than using stdout and optionally stdin. You MUST do this if - * your system can't cope with binary I/O to stdin/stdout. See comments at - * head of cjpeg.c or djpeg.c. - */ -#undef TWO_FILE_COMMANDLINE - -/* Define this if your system needs explicit cleanup of temporary files. - * This is crucial under MS-DOS, where the temporary "files" may be areas - * of extended memory; on most other systems it's not as important. - */ -#undef NEED_SIGNAL_CATCHER - -/* By default, we open image files with fopen(...,"rb") or fopen(...,"wb"). - * This is necessary on systems that distinguish text files from binary files, - * and is harmless on most systems that don't. If you have one of the rare - * systems that complains about the "b" spec, define this symbol. - */ -#undef DONT_USE_B_MODE - -/* Define this if you want percent-done progress reports from cjpeg/djpeg. - */ -#undef PROGRESS_REPORT - - -#endif /* JPEG_CJPEG_DJPEG */ diff --git a/dep/libjpeg/include/jerror.h b/dep/libjpeg/include/jerror.h deleted file mode 100644 index db608b9c0..000000000 --- a/dep/libjpeg/include/jerror.h +++ /dev/null @@ -1,304 +0,0 @@ -/* - * jerror.h - * - * Copyright (C) 1994-1997, Thomas G. Lane. - * Modified 1997-2018 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file defines the error and message codes for the JPEG library. - * Edit this file to add new codes, or to translate the message strings to - * some other language. - * A set of error-reporting macros are defined too. Some applications using - * the JPEG library may wish to include this file to get the error codes - * and/or the macros. - */ - -/* - * To define the enum list of message codes, include this file without - * defining macro JMESSAGE. To create a message string table, include it - * again with a suitable JMESSAGE definition (see jerror.c for an example). - */ -#ifndef JMESSAGE -#ifndef JERROR_H -/* First time through, define the enum list */ -#define JMAKE_ENUM_LIST -#else -/* Repeated inclusions of this file are no-ops unless JMESSAGE is defined */ -#define JMESSAGE(code,string) -#endif /* JERROR_H */ -#endif /* JMESSAGE */ - -#ifdef JMAKE_ENUM_LIST - -typedef enum { - -#define JMESSAGE(code,string) code , - -#endif /* JMAKE_ENUM_LIST */ - -JMESSAGE(JMSG_NOMESSAGE, "Bogus message code %d") /* Must be first entry! */ - -/* For maintenance convenience, list is alphabetical by message code name */ -JMESSAGE(JERR_BAD_ALIGN_TYPE, "ALIGN_TYPE is wrong, please fix") -JMESSAGE(JERR_BAD_ALLOC_CHUNK, "MAX_ALLOC_CHUNK is wrong, please fix") -JMESSAGE(JERR_BAD_BUFFER_MODE, "Bogus buffer control mode") -JMESSAGE(JERR_BAD_COMPONENT_ID, "Invalid component ID %d in SOS") -JMESSAGE(JERR_BAD_CROP_SPEC, "Invalid crop request") -JMESSAGE(JERR_BAD_DCT_COEF, "DCT coefficient out of range") -JMESSAGE(JERR_BAD_DCTSIZE, "DCT scaled block size %dx%d not supported") -JMESSAGE(JERR_BAD_DROP_SAMPLING, - "Component index %d: mismatching sampling ratio %d:%d, %d:%d, %c") -JMESSAGE(JERR_BAD_HUFF_TABLE, "Bogus Huffman table definition") -JMESSAGE(JERR_BAD_IN_COLORSPACE, "Bogus input colorspace") -JMESSAGE(JERR_BAD_J_COLORSPACE, "Bogus JPEG colorspace") -JMESSAGE(JERR_BAD_LENGTH, "Bogus marker length") -JMESSAGE(JERR_BAD_LIB_VERSION, - "Wrong JPEG library version: library is %d, caller expects %d") -JMESSAGE(JERR_BAD_MCU_SIZE, "Sampling factors too large for interleaved scan") -JMESSAGE(JERR_BAD_POOL_ID, "Invalid memory pool code %d") -JMESSAGE(JERR_BAD_PRECISION, "Unsupported JPEG data precision %d") -JMESSAGE(JERR_BAD_PROGRESSION, - "Invalid progressive parameters Ss=%d Se=%d Ah=%d Al=%d") -JMESSAGE(JERR_BAD_PROG_SCRIPT, - "Invalid progressive parameters at scan script entry %d") -JMESSAGE(JERR_BAD_SAMPLING, "Bogus sampling factors") -JMESSAGE(JERR_BAD_SCAN_SCRIPT, "Invalid scan script at entry %d") -JMESSAGE(JERR_BAD_STATE, "Improper call to JPEG library in state %d") -JMESSAGE(JERR_BAD_STRUCT_SIZE, - "JPEG parameter struct mismatch: library thinks size is %u, caller expects %u") -JMESSAGE(JERR_BAD_VIRTUAL_ACCESS, "Bogus virtual array access") -JMESSAGE(JERR_BUFFER_SIZE, "Buffer passed to JPEG library is too small") -JMESSAGE(JERR_CANT_SUSPEND, "Suspension not allowed here") -JMESSAGE(JERR_CCIR601_NOTIMPL, "CCIR601 sampling not implemented yet") -JMESSAGE(JERR_COMPONENT_COUNT, "Too many color components: %d, max %d") -JMESSAGE(JERR_CONVERSION_NOTIMPL, "Unsupported color conversion request") -JMESSAGE(JERR_DAC_INDEX, "Bogus DAC index %d") -JMESSAGE(JERR_DAC_VALUE, "Bogus DAC value 0x%x") -JMESSAGE(JERR_DHT_INDEX, "Bogus DHT index %d") -JMESSAGE(JERR_DQT_INDEX, "Bogus DQT index %d") -JMESSAGE(JERR_EMPTY_IMAGE, "Empty JPEG image (DNL not supported)") -JMESSAGE(JERR_EMS_READ, "Read from EMS failed") -JMESSAGE(JERR_EMS_WRITE, "Write to EMS failed") -JMESSAGE(JERR_EOI_EXPECTED, "Didn't expect more than one scan") -JMESSAGE(JERR_FILE_READ, "Input file read error") -JMESSAGE(JERR_FILE_WRITE, "Output file write error --- out of disk space?") -JMESSAGE(JERR_FRACT_SAMPLE_NOTIMPL, "Fractional sampling not implemented yet") -JMESSAGE(JERR_HUFF_CLEN_OUTOFBOUNDS, "Huffman code size table out of bounds") -JMESSAGE(JERR_HUFF_MISSING_CODE, "Missing Huffman code table entry") -JMESSAGE(JERR_IMAGE_TOO_BIG, "Maximum supported image dimension is %u pixels") -JMESSAGE(JERR_INPUT_EMPTY, "Empty input file") -JMESSAGE(JERR_INPUT_EOF, "Premature end of input file") -JMESSAGE(JERR_MISMATCHED_QUANT_TABLE, - "Cannot transcode due to multiple use of quantization table %d") -JMESSAGE(JERR_MISSING_DATA, "Scan script does not transmit all data") -JMESSAGE(JERR_MODE_CHANGE, "Invalid color quantization mode change") -JMESSAGE(JERR_NOTIMPL, "Not implemented yet") -JMESSAGE(JERR_NOT_COMPILED, "Requested feature was omitted at compile time") -JMESSAGE(JERR_NO_ARITH_TABLE, "Arithmetic table 0x%02x was not defined") -JMESSAGE(JERR_NO_BACKING_STORE, "Backing store not supported") -JMESSAGE(JERR_NO_HUFF_TABLE, "Huffman table 0x%02x was not defined") -JMESSAGE(JERR_NO_IMAGE, "JPEG datastream contains no image") -JMESSAGE(JERR_NO_QUANT_TABLE, "Quantization table 0x%02x was not defined") -JMESSAGE(JERR_NO_SOI, "Not a JPEG file: starts with 0x%02x 0x%02x") -JMESSAGE(JERR_OUT_OF_MEMORY, "Insufficient memory (case %d)") -JMESSAGE(JERR_QUANT_COMPONENTS, - "Cannot quantize more than %d color components") -JMESSAGE(JERR_QUANT_FEW_COLORS, "Cannot quantize to fewer than %d colors") -JMESSAGE(JERR_QUANT_MANY_COLORS, "Cannot quantize to more than %d colors") -JMESSAGE(JERR_SOF_BEFORE, "Invalid JPEG file structure: %s before SOF") -JMESSAGE(JERR_SOF_DUPLICATE, "Invalid JPEG file structure: two SOF markers") -JMESSAGE(JERR_SOF_NO_SOS, "Invalid JPEG file structure: missing SOS marker") -JMESSAGE(JERR_SOF_UNSUPPORTED, "Unsupported JPEG process: SOF type 0x%02x") -JMESSAGE(JERR_SOI_DUPLICATE, "Invalid JPEG file structure: two SOI markers") -JMESSAGE(JERR_TFILE_CREATE, "Failed to create temporary file %s") -JMESSAGE(JERR_TFILE_READ, "Read failed on temporary file") -JMESSAGE(JERR_TFILE_SEEK, "Seek failed on temporary file") -JMESSAGE(JERR_TFILE_WRITE, - "Write failed on temporary file --- out of disk space?") -JMESSAGE(JERR_TOO_LITTLE_DATA, "Application transferred too few scanlines") -JMESSAGE(JERR_UNKNOWN_MARKER, "Unsupported marker type 0x%02x") -JMESSAGE(JERR_VIRTUAL_BUG, "Virtual array controller messed up") -JMESSAGE(JERR_WIDTH_OVERFLOW, "Image too wide for this implementation") -JMESSAGE(JERR_XMS_READ, "Read from XMS failed") -JMESSAGE(JERR_XMS_WRITE, "Write to XMS failed") -JMESSAGE(JMSG_COPYRIGHT, JCOPYRIGHT) -JMESSAGE(JMSG_VERSION, JVERSION) -JMESSAGE(JTRC_16BIT_TABLES, - "Caution: quantization tables are too coarse for baseline JPEG") -JMESSAGE(JTRC_ADOBE, - "Adobe APP14 marker: version %d, flags 0x%04x 0x%04x, transform %d") -JMESSAGE(JTRC_APP0, "Unknown APP0 marker (not JFIF), length %u") -JMESSAGE(JTRC_APP14, "Unknown APP14 marker (not Adobe), length %u") -JMESSAGE(JTRC_DAC, "Define Arithmetic Table 0x%02x: 0x%02x") -JMESSAGE(JTRC_DHT, "Define Huffman Table 0x%02x") -JMESSAGE(JTRC_DQT, "Define Quantization Table %d precision %d") -JMESSAGE(JTRC_DRI, "Define Restart Interval %u") -JMESSAGE(JTRC_EMS_CLOSE, "Freed EMS handle %u") -JMESSAGE(JTRC_EMS_OPEN, "Obtained EMS handle %u") -JMESSAGE(JTRC_EOI, "End Of Image") -JMESSAGE(JTRC_HUFFBITS, " %3d %3d %3d %3d %3d %3d %3d %3d") -JMESSAGE(JTRC_JFIF, "JFIF APP0 marker: version %d.%02d, density %dx%d %d") -JMESSAGE(JTRC_JFIF_BADTHUMBNAILSIZE, - "Warning: thumbnail image size does not match data length %u") -JMESSAGE(JTRC_JFIF_EXTENSION, - "JFIF extension marker: type 0x%02x, length %u") -JMESSAGE(JTRC_JFIF_THUMBNAIL, " with %d x %d thumbnail image") -JMESSAGE(JTRC_MISC_MARKER, "Miscellaneous marker 0x%02x, length %u") -JMESSAGE(JTRC_PARMLESS_MARKER, "Unexpected marker 0x%02x") -JMESSAGE(JTRC_QUANTVALS, " %4u %4u %4u %4u %4u %4u %4u %4u") -JMESSAGE(JTRC_QUANT_3_NCOLORS, "Quantizing to %d = %d*%d*%d colors") -JMESSAGE(JTRC_QUANT_NCOLORS, "Quantizing to %d colors") -JMESSAGE(JTRC_QUANT_SELECTED, "Selected %d colors for quantization") -JMESSAGE(JTRC_RECOVERY_ACTION, "At marker 0x%02x, recovery action %d") -JMESSAGE(JTRC_RST, "RST%d") -JMESSAGE(JTRC_SMOOTH_NOTIMPL, - "Smoothing not supported with nonstandard sampling ratios") -JMESSAGE(JTRC_SOF, "Start Of Frame 0x%02x: width=%u, height=%u, components=%d") -JMESSAGE(JTRC_SOF_COMPONENT, " Component %d: %dhx%dv q=%d") -JMESSAGE(JTRC_SOI, "Start of Image") -JMESSAGE(JTRC_SOS, "Start Of Scan: %d components") -JMESSAGE(JTRC_SOS_COMPONENT, " Component %d: dc=%d ac=%d") -JMESSAGE(JTRC_SOS_PARAMS, " Ss=%d, Se=%d, Ah=%d, Al=%d") -JMESSAGE(JTRC_TFILE_CLOSE, "Closed temporary file %s") -JMESSAGE(JTRC_TFILE_OPEN, "Opened temporary file %s") -JMESSAGE(JTRC_THUMB_JPEG, - "JFIF extension marker: JPEG-compressed thumbnail image, length %u") -JMESSAGE(JTRC_THUMB_PALETTE, - "JFIF extension marker: palette thumbnail image, length %u") -JMESSAGE(JTRC_THUMB_RGB, - "JFIF extension marker: RGB thumbnail image, length %u") -JMESSAGE(JTRC_UNKNOWN_IDS, - "Unrecognized component IDs %d %d %d, assuming YCbCr") -JMESSAGE(JTRC_XMS_CLOSE, "Freed XMS handle %u") -JMESSAGE(JTRC_XMS_OPEN, "Obtained XMS handle %u") -JMESSAGE(JWRN_ADOBE_XFORM, "Unknown Adobe color transform code %d") -JMESSAGE(JWRN_ARITH_BAD_CODE, "Corrupt JPEG data: bad arithmetic code") -JMESSAGE(JWRN_BOGUS_PROGRESSION, - "Inconsistent progression sequence for component %d coefficient %d") -JMESSAGE(JWRN_EXTRANEOUS_DATA, - "Corrupt JPEG data: %u extraneous bytes before marker 0x%02x") -JMESSAGE(JWRN_HIT_MARKER, "Corrupt JPEG data: premature end of data segment") -JMESSAGE(JWRN_HUFF_BAD_CODE, "Corrupt JPEG data: bad Huffman code") -JMESSAGE(JWRN_JFIF_MAJOR, "Warning: unknown JFIF revision number %d.%02d") -JMESSAGE(JWRN_JPEG_EOF, "Premature end of JPEG file") -JMESSAGE(JWRN_MUST_RESYNC, - "Corrupt JPEG data: found marker 0x%02x instead of RST%d") -JMESSAGE(JWRN_NOT_SEQUENTIAL, "Invalid SOS parameters for sequential JPEG") -JMESSAGE(JWRN_TOO_MUCH_DATA, "Application transferred too many scanlines") - -#ifdef JMAKE_ENUM_LIST - - JMSG_LASTMSGCODE -} J_MESSAGE_CODE; - -#undef JMAKE_ENUM_LIST -#endif /* JMAKE_ENUM_LIST */ - -/* Zap JMESSAGE macro so that future re-inclusions do nothing by default */ -#undef JMESSAGE - - -#ifndef JERROR_H -#define JERROR_H - -/* Macros to simplify using the error and trace message stuff */ -/* The first parameter is either type of cinfo pointer */ - -/* Fatal errors (print message and exit) */ -#define ERREXIT(cinfo,code) \ - ((cinfo)->err->msg_code = (code), \ - (*(cinfo)->err->error_exit) ((j_common_ptr) (cinfo))) -#define ERREXIT1(cinfo,code,p1) \ - ((cinfo)->err->msg_code = (code), \ - (cinfo)->err->msg_parm.i[0] = (p1), \ - (*(cinfo)->err->error_exit) ((j_common_ptr) (cinfo))) -#define ERREXIT2(cinfo,code,p1,p2) \ - ((cinfo)->err->msg_code = (code), \ - (cinfo)->err->msg_parm.i[0] = (p1), \ - (cinfo)->err->msg_parm.i[1] = (p2), \ - (*(cinfo)->err->error_exit) ((j_common_ptr) (cinfo))) -#define ERREXIT3(cinfo,code,p1,p2,p3) \ - ((cinfo)->err->msg_code = (code), \ - (cinfo)->err->msg_parm.i[0] = (p1), \ - (cinfo)->err->msg_parm.i[1] = (p2), \ - (cinfo)->err->msg_parm.i[2] = (p3), \ - (*(cinfo)->err->error_exit) ((j_common_ptr) (cinfo))) -#define ERREXIT4(cinfo,code,p1,p2,p3,p4) \ - ((cinfo)->err->msg_code = (code), \ - (cinfo)->err->msg_parm.i[0] = (p1), \ - (cinfo)->err->msg_parm.i[1] = (p2), \ - (cinfo)->err->msg_parm.i[2] = (p3), \ - (cinfo)->err->msg_parm.i[3] = (p4), \ - (*(cinfo)->err->error_exit) ((j_common_ptr) (cinfo))) -#define ERREXIT6(cinfo,code,p1,p2,p3,p4,p5,p6) \ - ((cinfo)->err->msg_code = (code), \ - (cinfo)->err->msg_parm.i[0] = (p1), \ - (cinfo)->err->msg_parm.i[1] = (p2), \ - (cinfo)->err->msg_parm.i[2] = (p3), \ - (cinfo)->err->msg_parm.i[3] = (p4), \ - (cinfo)->err->msg_parm.i[4] = (p5), \ - (cinfo)->err->msg_parm.i[5] = (p6), \ - (*(cinfo)->err->error_exit) ((j_common_ptr) (cinfo))) -#define ERREXITS(cinfo,code,str) \ - ((cinfo)->err->msg_code = (code), \ - strncpy((cinfo)->err->msg_parm.s, (str), JMSG_STR_PARM_MAX), \ - (*(cinfo)->err->error_exit) ((j_common_ptr) (cinfo))) - -#define MAKESTMT(stuff) do { stuff } while (0) - -/* Nonfatal errors (we can keep going, but the data is probably corrupt) */ -#define WARNMS(cinfo,code) \ - ((cinfo)->err->msg_code = (code), \ - (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), -1)) -#define WARNMS1(cinfo,code,p1) \ - ((cinfo)->err->msg_code = (code), \ - (cinfo)->err->msg_parm.i[0] = (p1), \ - (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), -1)) -#define WARNMS2(cinfo,code,p1,p2) \ - ((cinfo)->err->msg_code = (code), \ - (cinfo)->err->msg_parm.i[0] = (p1), \ - (cinfo)->err->msg_parm.i[1] = (p2), \ - (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), -1)) - -/* Informational/debugging messages */ -#define TRACEMS(cinfo,lvl,code) \ - ((cinfo)->err->msg_code = (code), \ - (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl))) -#define TRACEMS1(cinfo,lvl,code,p1) \ - ((cinfo)->err->msg_code = (code), \ - (cinfo)->err->msg_parm.i[0] = (p1), \ - (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl))) -#define TRACEMS2(cinfo,lvl,code,p1,p2) \ - ((cinfo)->err->msg_code = (code), \ - (cinfo)->err->msg_parm.i[0] = (p1), \ - (cinfo)->err->msg_parm.i[1] = (p2), \ - (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl))) -#define TRACEMS3(cinfo,lvl,code,p1,p2,p3) \ - MAKESTMT(int * _mp = (cinfo)->err->msg_parm.i; \ - _mp[0] = (p1); _mp[1] = (p2); _mp[2] = (p3); \ - (cinfo)->err->msg_code = (code); \ - (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)); ) -#define TRACEMS4(cinfo,lvl,code,p1,p2,p3,p4) \ - MAKESTMT(int * _mp = (cinfo)->err->msg_parm.i; \ - _mp[0] = (p1); _mp[1] = (p2); _mp[2] = (p3); _mp[3] = (p4); \ - (cinfo)->err->msg_code = (code); \ - (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)); ) -#define TRACEMS5(cinfo,lvl,code,p1,p2,p3,p4,p5) \ - MAKESTMT(int * _mp = (cinfo)->err->msg_parm.i; \ - _mp[0] = (p1); _mp[1] = (p2); _mp[2] = (p3); _mp[3] = (p4); \ - _mp[4] = (p5); \ - (cinfo)->err->msg_code = (code); \ - (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)); ) -#define TRACEMS8(cinfo,lvl,code,p1,p2,p3,p4,p5,p6,p7,p8) \ - MAKESTMT(int * _mp = (cinfo)->err->msg_parm.i; \ - _mp[0] = (p1); _mp[1] = (p2); _mp[2] = (p3); _mp[3] = (p4); \ - _mp[4] = (p5); _mp[5] = (p6); _mp[6] = (p7); _mp[7] = (p8); \ - (cinfo)->err->msg_code = (code); \ - (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl)); ) -#define TRACEMSS(cinfo,lvl,code,str) \ - ((cinfo)->err->msg_code = (code), \ - strncpy((cinfo)->err->msg_parm.s, (str), JMSG_STR_PARM_MAX), \ - (*(cinfo)->err->emit_message) ((j_common_ptr) (cinfo), (lvl))) - -#endif /* JERROR_H */ diff --git a/dep/libjpeg/include/jmorecfg.h b/dep/libjpeg/include/jmorecfg.h deleted file mode 100644 index 4638d6af2..000000000 --- a/dep/libjpeg/include/jmorecfg.h +++ /dev/null @@ -1,457 +0,0 @@ -/* - * jmorecfg.h - * - * Copyright (C) 1991-1997, Thomas G. Lane. - * Modified 1997-2022 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains additional configuration options that customize the - * JPEG software for special applications or support machine-dependent - * optimizations. Most users will not need to touch this file. - */ - - -/* - * Define BITS_IN_JSAMPLE as either - * 8 for 8-bit sample values (the usual setting) - * 9 for 9-bit sample values - * 10 for 10-bit sample values - * 11 for 11-bit sample values - * 12 for 12-bit sample values - * Only 8, 9, 10, 11, and 12 bits sample data precision are supported for - * full-feature DCT processing. Further depths up to 16-bit may be added - * later for the lossless modes of operation. - * Run-time selection and conversion of data precision will be added later - * and are currently not supported, sorry. - * Exception: The transcoding part (jpegtran) supports all settings in a - * single instance, since it operates on the level of DCT coefficients and - * not sample values. The DCT coefficients are of the same type (16 bits) - * in all cases (see below). - */ - -#define BITS_IN_JSAMPLE 8 /* use 8, 9, 10, 11, or 12 */ - - -/* - * Maximum number of components (color channels) allowed in JPEG image. - * To meet the letter of the JPEG spec, set this to 255. However, darn - * few applications need more than 4 channels (maybe 5 for CMYK + alpha - * mask). We recommend 10 as a reasonable compromise; use 4 if you are - * really short on memory. (Each allowed component costs a hundred or so - * bytes of storage, whether actually used in an image or not.) - */ - -#define MAX_COMPONENTS 10 /* maximum number of image components */ - - -/* - * Basic data types. - * You may need to change these if you have a machine with unusual data - * type sizes; for example, "char" not 8 bits, "short" not 16 bits, - * or "long" not 32 bits. We don't care whether "int" is 16 or 32 bits, - * but it had better be at least 16. - */ - -/* Representation of a single sample (pixel element value). - * We frequently allocate large arrays of these, so it's important to keep - * them small. But if you have memory to burn and access to char or short - * arrays is very slow on your hardware, you might want to change these. - */ - -#if BITS_IN_JSAMPLE == 8 -/* JSAMPLE should be the smallest type that will hold the values 0..255. - * You can use a signed char by having GETJSAMPLE mask it with 0xFF. - */ - -#ifdef HAVE_UNSIGNED_CHAR - -typedef unsigned char JSAMPLE; -#define GETJSAMPLE(value) ((int) (value)) - -#else /* not HAVE_UNSIGNED_CHAR */ - -typedef char JSAMPLE; -#ifdef CHAR_IS_UNSIGNED -#define GETJSAMPLE(value) ((int) (value)) -#else -#define GETJSAMPLE(value) ((int) (value) & 0xFF) -#endif /* CHAR_IS_UNSIGNED */ - -#endif /* HAVE_UNSIGNED_CHAR */ - -#define MAXJSAMPLE 255 -#define CENTERJSAMPLE 128 - -#endif /* BITS_IN_JSAMPLE == 8 */ - - -#if BITS_IN_JSAMPLE == 9 -/* JSAMPLE should be the smallest type that will hold the values 0..511. - * On nearly all machines "short" will do nicely. - */ - -typedef short JSAMPLE; -#define GETJSAMPLE(value) ((int) (value)) - -#define MAXJSAMPLE 511 -#define CENTERJSAMPLE 256 - -#endif /* BITS_IN_JSAMPLE == 9 */ - - -#if BITS_IN_JSAMPLE == 10 -/* JSAMPLE should be the smallest type that will hold the values 0..1023. - * On nearly all machines "short" will do nicely. - */ - -typedef short JSAMPLE; -#define GETJSAMPLE(value) ((int) (value)) - -#define MAXJSAMPLE 1023 -#define CENTERJSAMPLE 512 - -#endif /* BITS_IN_JSAMPLE == 10 */ - - -#if BITS_IN_JSAMPLE == 11 -/* JSAMPLE should be the smallest type that will hold the values 0..2047. - * On nearly all machines "short" will do nicely. - */ - -typedef short JSAMPLE; -#define GETJSAMPLE(value) ((int) (value)) - -#define MAXJSAMPLE 2047 -#define CENTERJSAMPLE 1024 - -#endif /* BITS_IN_JSAMPLE == 11 */ - - -#if BITS_IN_JSAMPLE == 12 -/* JSAMPLE should be the smallest type that will hold the values 0..4095. - * On nearly all machines "short" will do nicely. - */ - -typedef short JSAMPLE; -#define GETJSAMPLE(value) ((int) (value)) - -#define MAXJSAMPLE 4095 -#define CENTERJSAMPLE 2048 - -#endif /* BITS_IN_JSAMPLE == 12 */ - - -/* Representation of a DCT frequency coefficient. - * This should be a signed value of at least 16 bits; "short" is usually OK. - * Again, we allocate large arrays of these, but you can change to int - * if you have memory to burn and "short" is really slow. - */ - -typedef short JCOEF; - - -/* Compressed datastreams are represented as arrays of JOCTET. - * These must be EXACTLY 8 bits wide, at least once they are written to - * external storage. Note that when using the stdio data source/destination - * managers, this is also the data type passed to fread/fwrite. - */ - -#ifdef HAVE_UNSIGNED_CHAR - -typedef unsigned char JOCTET; -#define GETJOCTET(value) (value) - -#else /* not HAVE_UNSIGNED_CHAR */ - -typedef char JOCTET; -#ifdef CHAR_IS_UNSIGNED -#define GETJOCTET(value) (value) -#else -#define GETJOCTET(value) ((value) & 0xFF) -#endif /* CHAR_IS_UNSIGNED */ - -#endif /* HAVE_UNSIGNED_CHAR */ - - -/* These typedefs are used for various table entries and so forth. - * They must be at least as wide as specified; but making them too big - * won't cost a huge amount of memory, so we don't provide special - * extraction code like we did for JSAMPLE. (In other words, these - * typedefs live at a different point on the speed/space tradeoff curve.) - */ - -/* UINT8 must hold at least the values 0..255. */ - -#ifdef HAVE_UNSIGNED_CHAR -typedef unsigned char UINT8; -#else /* not HAVE_UNSIGNED_CHAR */ -#ifdef CHAR_IS_UNSIGNED -typedef char UINT8; -#else /* not CHAR_IS_UNSIGNED */ -typedef short UINT8; -#endif /* CHAR_IS_UNSIGNED */ -#endif /* HAVE_UNSIGNED_CHAR */ - -/* UINT16 must hold at least the values 0..65535. */ - -#ifdef HAVE_UNSIGNED_SHORT -typedef unsigned short UINT16; -#else /* not HAVE_UNSIGNED_SHORT */ -typedef unsigned int UINT16; -#endif /* HAVE_UNSIGNED_SHORT */ - -/* INT16 must hold at least the values -32768..32767. */ - -#ifndef XMD_H /* X11/xmd.h correctly defines INT16 */ -typedef short INT16; -#endif - -/* INT32 must hold at least signed 32-bit values. */ - -#ifndef XMD_H /* X11/xmd.h correctly defines INT32 */ -#ifndef _BASETSD_H_ /* Microsoft defines it in basetsd.h */ -#ifndef _BASETSD_H /* MinGW is slightly different */ -#ifndef QGLOBAL_H /* Qt defines it in qglobal.h */ -typedef long INT32; -#endif -#endif -#endif -#endif - -/* Datatype used for image dimensions. The JPEG standard only supports - * images up to 64K*64K due to 16-bit fields in SOF markers. Therefore - * "unsigned int" is sufficient on all machines. However, if you need to - * handle larger images and you don't mind deviating from the spec, you - * can change this datatype. - */ - -typedef unsigned int JDIMENSION; - -#define JPEG_MAX_DIMENSION 65500L /* a tad under 64K to prevent overflows */ - - -/* These macros are used in all function definitions and extern declarations. - * You could modify them if you need to change function linkage conventions; - * in particular, you'll need to do that to make the library a Windows DLL. - * Another application is to make all functions global for use with debuggers - * or code profilers that require it. - */ - -/* a function called through method pointers: */ -#define METHODDEF(type) static type -/* a function used only in its module: */ -#define LOCAL(type) static type -/* a function referenced thru EXTERNs: */ -#define GLOBAL(type) type -/* a reference to a GLOBAL function: */ -#define EXTERN(type) extern type - - -/* This macro is used to declare a "method", that is, a function pointer. - * We want to supply prototype parameters if the compiler can cope. - * Note that the arglist parameter must be parenthesized! - * Again, you can customize this if you need special linkage keywords. - */ - -#ifdef HAVE_PROTOTYPES -#define JMETHOD(type,methodname,arglist) type (*methodname) arglist -#else -#define JMETHOD(type,methodname,arglist) type (*methodname) () -#endif - - -/* The noreturn type identifier is used to declare functions - * which cannot return. - * Compilers can thus create more optimized code and perform - * better checks for warnings and errors. - * Static analyzer tools can make improved inferences about - * execution paths and are prevented from giving false alerts. - * - * Unfortunately, the proposed specifications of corresponding - * extensions in the Dec 2011 ISO C standard revision (C11), - * GCC, MSVC, etc. are not viable. - * Thus we introduce a user defined type to declare noreturn - * functions at least for clarity. A proper compiler would - * have a suitable noreturn type to match in place of void. - */ - -#ifndef HAVE_NORETURN_T -typedef void noreturn_t; -#endif - - -/* Here is the pseudo-keyword for declaring pointers that must be "far" - * on 80x86 machines. Most of the specialized coding for 80x86 is handled - * by just saying "FAR *" where such a pointer is needed. In a few places - * explicit coding is needed; see uses of the NEED_FAR_POINTERS symbol. - */ - -#ifndef FAR -#ifdef NEED_FAR_POINTERS -#define FAR far -#else -#define FAR -#endif -#endif - - -/* - * On a few systems, type boolean and/or its values FALSE, TRUE may appear - * in standard header files. Or you may have conflicts with application- - * specific header files that you want to include together with these files. - * Defining HAVE_BOOLEAN before including jpeglib.h should make it work. - */ - -#ifndef HAVE_BOOLEAN -#if defined FALSE || defined TRUE || defined QGLOBAL_H -/* Qt3 defines FALSE and TRUE as "const" variables in qglobal.h */ -typedef int boolean; -#ifndef FALSE /* in case these macros already exist */ -#define FALSE 0 /* values of boolean */ -#endif -#ifndef TRUE -#define TRUE 1 -#endif -#else -typedef enum { FALSE = 0, TRUE = 1 } boolean; -#endif -#endif - - -/* - * The remaining options affect code selection within the JPEG library, - * but they don't need to be visible to most applications using the library. - * To minimize application namespace pollution, the symbols won't be - * defined unless JPEG_INTERNALS or JPEG_INTERNAL_OPTIONS has been defined. - */ - -#ifdef JPEG_INTERNALS -#define JPEG_INTERNAL_OPTIONS -#endif - -#ifdef JPEG_INTERNAL_OPTIONS - - -/* - * These defines indicate whether to include various optional functions. - * Undefining some of these symbols will produce a smaller but less capable - * library. Note that you can leave certain source files out of the - * compilation/linking process if you've #undef'd the corresponding symbols. - * (You may HAVE to do that if your compiler doesn't like null source files.) - */ - -/* Capability options common to encoder and decoder: */ - -#define DCT_ISLOW_SUPPORTED /* slow but accurate integer algorithm */ -#define DCT_IFAST_SUPPORTED /* faster, less accurate integer method */ -#define DCT_FLOAT_SUPPORTED /* floating-point: accurate, fast on fast HW */ - -/* Encoder capability options: */ - -#define C_ARITH_CODING_SUPPORTED /* Arithmetic coding back end? */ -#define C_MULTISCAN_FILES_SUPPORTED /* Multiple-scan JPEG files? */ -#define C_PROGRESSIVE_SUPPORTED /* Progressive JPEG? (Requires MULTISCAN) */ -#define DCT_SCALING_SUPPORTED /* Input rescaling via DCT? (Requires DCT_ISLOW) */ -#define ENTROPY_OPT_SUPPORTED /* Optimization of entropy coding parms? */ -/* Note: if you selected more than 8-bit data precision, it is dangerous to - * turn off ENTROPY_OPT_SUPPORTED. The standard Huffman tables are only - * good for 8-bit precision, so arithmetic coding is recommended for higher - * precision. The Huffman encoder normally uses entropy optimization to - * compute usable tables for higher precision. Otherwise, you'll have to - * supply different default Huffman tables. - * The exact same statements apply for progressive JPEG: the default tables - * don't work for progressive mode. (This may get fixed, however.) - */ -#define INPUT_SMOOTHING_SUPPORTED /* Input image smoothing option? */ - -/* Decoder capability options: */ - -#define D_ARITH_CODING_SUPPORTED /* Arithmetic coding back end? */ -#define D_MULTISCAN_FILES_SUPPORTED /* Multiple-scan JPEG files? */ -#define D_PROGRESSIVE_SUPPORTED /* Progressive JPEG? (Requires MULTISCAN) */ -#define IDCT_SCALING_SUPPORTED /* Output rescaling via IDCT? (Requires DCT_ISLOW) */ -#define SAVE_MARKERS_SUPPORTED /* jpeg_save_markers() needed? */ -#define BLOCK_SMOOTHING_SUPPORTED /* Block smoothing? (Progressive only) */ -#undef UPSAMPLE_SCALING_SUPPORTED /* Output rescaling at upsample stage? */ -#define UPSAMPLE_MERGING_SUPPORTED /* Fast path for sloppy upsampling? */ -#define QUANT_1PASS_SUPPORTED /* 1-pass color quantization? */ -#define QUANT_2PASS_SUPPORTED /* 2-pass color quantization? */ - -/* more capability options later, no doubt */ - - -/* - * Ordering of RGB data in scanlines passed to or from the application. - * If your application wants to deal with data in the order B,G,R, just - * #define JPEG_USE_RGB_CUSTOM in jconfig.h, or define your own custom - * order in jconfig.h and #define JPEG_HAVE_RGB_CUSTOM. - * You can also deal with formats such as R,G,B,X (one extra byte per pixel) - * by changing RGB_PIXELSIZE. - * Note that changing the offsets will also change - * the order in which colormap data is organized. - * RESTRICTIONS: - * 1. The sample applications cjpeg,djpeg do NOT support modified RGB formats. - * 2. The color quantizer modules will not behave desirably if RGB_PIXELSIZE - * is not 3 (they don't understand about dummy color components!). - * So you can't use color quantization if you change that value. - */ - -#ifndef JPEG_HAVE_RGB_CUSTOM -#ifdef JPEG_USE_RGB_CUSTOM -#define RGB_RED 2 /* Offset of Red in an RGB scanline element */ -#define RGB_GREEN 1 /* Offset of Green */ -#define RGB_BLUE 0 /* Offset of Blue */ -#else -#define RGB_RED 0 /* Offset of Red in an RGB scanline element */ -#define RGB_GREEN 1 /* Offset of Green */ -#define RGB_BLUE 2 /* Offset of Blue */ -#endif -#define RGB_PIXELSIZE 3 /* JSAMPLEs per RGB scanline element */ -#endif - - -/* Definitions for speed-related optimizations. */ - - -/* If your compiler supports inline functions, define INLINE - * as the inline keyword; otherwise define it as empty. - */ - -#ifndef INLINE -#ifdef __GNUC__ /* for instance, GNU C knows about inline */ -#define INLINE __inline__ -#endif -#ifndef INLINE -#define INLINE /* default is to define it as empty */ -#endif -#endif - - -/* On some machines (notably 68000 series) "int" is 32 bits, but multiplying - * two 16-bit shorts is faster than multiplying two ints. Define MULTIPLIER - * as short on such a machine. MULTIPLIER must be at least 16 bits wide. - */ - -#ifndef MULTIPLIER -#define MULTIPLIER int /* type for fastest integer multiply */ -#endif - - -/* FAST_FLOAT should be either float or double, whichever is done faster - * by your compiler. (Note that this type is only used in the floating point - * DCT routines, so it only matters if you've defined DCT_FLOAT_SUPPORTED.) - * Typically, float is faster in ANSI C compilers, while double is faster in - * pre-ANSI compilers (because they insist on converting to double anyway). - * The code below therefore chooses float if we have ANSI-style prototypes. - */ - -#ifndef FAST_FLOAT -#ifdef HAVE_PROTOTYPES -#define FAST_FLOAT float -#else -#define FAST_FLOAT double -#endif -#endif - -#endif /* JPEG_INTERNAL_OPTIONS */ diff --git a/dep/libjpeg/include/jpegint.h b/dep/libjpeg/include/jpegint.h deleted file mode 100644 index 3528bff5b..000000000 --- a/dep/libjpeg/include/jpegint.h +++ /dev/null @@ -1,445 +0,0 @@ -/* - * jpegint.h - * - * Copyright (C) 1991-1997, Thomas G. Lane. - * Modified 1997-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file provides common declarations for the various JPEG modules. - * These declarations are considered internal to the JPEG library; most - * applications using the library shouldn't need to include this file. - */ - - -/* Declarations for both compression & decompression */ - -typedef enum { /* Operating modes for buffer controllers */ - JBUF_PASS_THRU, /* Plain stripwise operation */ - /* Remaining modes require a full-image buffer to have been created */ - JBUF_SAVE_SOURCE, /* Run source subobject only, save output */ - JBUF_CRANK_DEST, /* Run dest subobject only, using saved data */ - JBUF_SAVE_AND_PASS /* Run both subobjects, save output */ -} J_BUF_MODE; - -/* Values of global_state field (jdapi.c has some dependencies on ordering!) */ -#define CSTATE_START 100 /* after create_compress */ -#define CSTATE_SCANNING 101 /* start_compress done, write_scanlines OK */ -#define CSTATE_RAW_OK 102 /* start_compress done, write_raw_data OK */ -#define CSTATE_WRCOEFS 103 /* jpeg_write_coefficients done */ -#define DSTATE_START 200 /* after create_decompress */ -#define DSTATE_INHEADER 201 /* reading header markers, no SOS yet */ -#define DSTATE_READY 202 /* found SOS, ready for start_decompress */ -#define DSTATE_PRELOAD 203 /* reading multiscan file in start_decompress*/ -#define DSTATE_PRESCAN 204 /* performing dummy pass for 2-pass quant */ -#define DSTATE_SCANNING 205 /* start_decompress done, read_scanlines OK */ -#define DSTATE_RAW_OK 206 /* start_decompress done, read_raw_data OK */ -#define DSTATE_BUFIMAGE 207 /* expecting jpeg_start_output */ -#define DSTATE_BUFPOST 208 /* looking for SOS/EOI in jpeg_finish_output */ -#define DSTATE_RDCOEFS 209 /* reading file in jpeg_read_coefficients */ -#define DSTATE_STOPPING 210 /* looking for EOI in jpeg_finish_decompress */ - - -/* Declarations for compression modules */ - -/* Master control module */ -struct jpeg_comp_master { - JMETHOD(void, prepare_for_pass, (j_compress_ptr cinfo)); - JMETHOD(void, pass_startup, (j_compress_ptr cinfo)); - JMETHOD(void, finish_pass, (j_compress_ptr cinfo)); - - /* State variables made visible to other modules */ - boolean call_pass_startup; /* True if pass_startup must be called */ - boolean is_last_pass; /* True during last pass */ -}; - -/* Main buffer control (downsampled-data buffer) */ -struct jpeg_c_main_controller { - JMETHOD(void, start_pass, (j_compress_ptr cinfo, J_BUF_MODE pass_mode)); - JMETHOD(void, process_data, (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JDIMENSION *in_row_ctr, - JDIMENSION in_rows_avail)); -}; - -/* Compression preprocessing (downsampling input buffer control) */ -struct jpeg_c_prep_controller { - JMETHOD(void, start_pass, (j_compress_ptr cinfo, J_BUF_MODE pass_mode)); - JMETHOD(void, pre_process_data, (j_compress_ptr cinfo, - JSAMPARRAY input_buf, - JDIMENSION *in_row_ctr, - JDIMENSION in_rows_avail, - JSAMPIMAGE output_buf, - JDIMENSION *out_row_group_ctr, - JDIMENSION out_row_groups_avail)); -}; - -/* Coefficient buffer control */ -struct jpeg_c_coef_controller { - JMETHOD(void, start_pass, (j_compress_ptr cinfo, J_BUF_MODE pass_mode)); - JMETHOD(boolean, compress_data, (j_compress_ptr cinfo, - JSAMPIMAGE input_buf)); -}; - -/* Colorspace conversion */ -struct jpeg_color_converter { - JMETHOD(void, start_pass, (j_compress_ptr cinfo)); - JMETHOD(void, color_convert, (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JSAMPIMAGE output_buf, - JDIMENSION output_row, int num_rows)); -}; - -/* Downsampling */ -struct jpeg_downsampler { - JMETHOD(void, start_pass, (j_compress_ptr cinfo)); - JMETHOD(void, downsample, (j_compress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION in_row_index, - JSAMPIMAGE output_buf, - JDIMENSION out_row_group_index)); - - boolean need_context_rows; /* TRUE if need rows above & below */ -}; - -/* Forward DCT (also controls coefficient quantization) */ -typedef JMETHOD(void, forward_DCT_ptr, - (j_compress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY sample_data, JBLOCKROW coef_blocks, - JDIMENSION start_col, JDIMENSION num_blocks)); - -struct jpeg_forward_dct { - JMETHOD(void, start_pass, (j_compress_ptr cinfo)); - /* It is useful to allow each component to have a separate FDCT method. */ - forward_DCT_ptr forward_DCT[MAX_COMPONENTS]; -}; - -/* Entropy encoding */ -struct jpeg_entropy_encoder { - JMETHOD(void, start_pass, (j_compress_ptr cinfo, boolean gather_statistics)); - JMETHOD(boolean, encode_mcu, (j_compress_ptr cinfo, JBLOCKARRAY MCU_data)); - JMETHOD(void, finish_pass, (j_compress_ptr cinfo)); -}; - -/* Marker writing */ -struct jpeg_marker_writer { - JMETHOD(void, write_file_header, (j_compress_ptr cinfo)); - JMETHOD(void, write_frame_header, (j_compress_ptr cinfo)); - JMETHOD(void, write_scan_header, (j_compress_ptr cinfo)); - JMETHOD(void, write_file_trailer, (j_compress_ptr cinfo)); - JMETHOD(void, write_tables_only, (j_compress_ptr cinfo)); - /* These routines are exported to allow insertion of extra markers */ - /* Probably only COM and APPn markers should be written this way */ - JMETHOD(void, write_marker_header, (j_compress_ptr cinfo, int marker, - unsigned int datalen)); - JMETHOD(void, write_marker_byte, (j_compress_ptr cinfo, int val)); -}; - - -/* Declarations for decompression modules */ - -/* Master control module */ -struct jpeg_decomp_master { - JMETHOD(void, prepare_for_output_pass, (j_decompress_ptr cinfo)); - JMETHOD(void, finish_output_pass, (j_decompress_ptr cinfo)); - - /* State variables made visible to other modules */ - boolean is_dummy_pass; /* True during 1st pass for 2-pass quant */ -}; - -/* Input control module */ -struct jpeg_input_controller { - JMETHOD(int, consume_input, (j_decompress_ptr cinfo)); - JMETHOD(void, reset_input_controller, (j_decompress_ptr cinfo)); - JMETHOD(void, start_input_pass, (j_decompress_ptr cinfo)); - JMETHOD(void, finish_input_pass, (j_decompress_ptr cinfo)); - - /* State variables made visible to other modules */ - boolean has_multiple_scans; /* True if file has multiple scans */ - boolean eoi_reached; /* True when EOI has been consumed */ -}; - -/* Main buffer control (downsampled-data buffer) */ -struct jpeg_d_main_controller { - JMETHOD(void, start_pass, (j_decompress_ptr cinfo, J_BUF_MODE pass_mode)); - JMETHOD(void, process_data, (j_decompress_ptr cinfo, - JSAMPARRAY output_buf, JDIMENSION *out_row_ctr, - JDIMENSION out_rows_avail)); -}; - -/* Coefficient buffer control */ -struct jpeg_d_coef_controller { - JMETHOD(void, start_input_pass, (j_decompress_ptr cinfo)); - JMETHOD(int, consume_data, (j_decompress_ptr cinfo)); - JMETHOD(void, start_output_pass, (j_decompress_ptr cinfo)); - JMETHOD(int, decompress_data, (j_decompress_ptr cinfo, - JSAMPIMAGE output_buf)); - /* Pointer to array of coefficient virtual arrays, or NULL if none */ - jvirt_barray_ptr *coef_arrays; -}; - -/* Decompression postprocessing (color quantization buffer control) */ -struct jpeg_d_post_controller { - JMETHOD(void, start_pass, (j_decompress_ptr cinfo, J_BUF_MODE pass_mode)); - JMETHOD(void, post_process_data, (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, - JDIMENSION *in_row_group_ctr, - JDIMENSION in_row_groups_avail, - JSAMPARRAY output_buf, - JDIMENSION *out_row_ctr, - JDIMENSION out_rows_avail)); -}; - -/* Marker reading & parsing */ -struct jpeg_marker_reader { - JMETHOD(void, reset_marker_reader, (j_decompress_ptr cinfo)); - /* Read markers until SOS or EOI. - * Returns same codes as are defined for jpeg_consume_input: - * JPEG_SUSPENDED, JPEG_REACHED_SOS, or JPEG_REACHED_EOI. - */ - JMETHOD(int, read_markers, (j_decompress_ptr cinfo)); - /* Read a restart marker --- exported for use by entropy decoder only */ - jpeg_marker_parser_method read_restart_marker; - - /* State of marker reader --- nominally internal, but applications - * supplying COM or APPn handlers might like to know the state. - */ - boolean saw_SOI; /* found SOI? */ - boolean saw_SOF; /* found SOF? */ - int next_restart_num; /* next restart number expected (0-7) */ - unsigned int discarded_bytes; /* # of bytes skipped looking for a marker */ -}; - -/* Entropy decoding */ -struct jpeg_entropy_decoder { - JMETHOD(void, start_pass, (j_decompress_ptr cinfo)); - JMETHOD(boolean, decode_mcu, (j_decompress_ptr cinfo, JBLOCKARRAY MCU_data)); - JMETHOD(void, finish_pass, (j_decompress_ptr cinfo)); -}; - -/* Inverse DCT (also performs dequantization) */ -typedef JMETHOD(void, inverse_DCT_method_ptr, - (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col)); - -struct jpeg_inverse_dct { - JMETHOD(void, start_pass, (j_decompress_ptr cinfo)); - /* It is useful to allow each component to have a separate IDCT method. */ - inverse_DCT_method_ptr inverse_DCT[MAX_COMPONENTS]; -}; - -/* Upsampling (note that upsampler must also call color converter) */ -struct jpeg_upsampler { - JMETHOD(void, start_pass, (j_decompress_ptr cinfo)); - JMETHOD(void, upsample, (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, - JDIMENSION *in_row_group_ctr, - JDIMENSION in_row_groups_avail, - JSAMPARRAY output_buf, - JDIMENSION *out_row_ctr, - JDIMENSION out_rows_avail)); - - boolean need_context_rows; /* TRUE if need rows above & below */ -}; - -/* Colorspace conversion */ -struct jpeg_color_deconverter { - JMETHOD(void, start_pass, (j_decompress_ptr cinfo)); - JMETHOD(void, color_convert, (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION input_row, - JSAMPARRAY output_buf, int num_rows)); -}; - -/* Color quantization or color precision reduction */ -struct jpeg_color_quantizer { - JMETHOD(void, start_pass, (j_decompress_ptr cinfo, boolean is_pre_scan)); - JMETHOD(void, color_quantize, (j_decompress_ptr cinfo, - JSAMPARRAY input_buf, JSAMPARRAY output_buf, - int num_rows)); - JMETHOD(void, finish_pass, (j_decompress_ptr cinfo)); - JMETHOD(void, new_color_map, (j_decompress_ptr cinfo)); -}; - - -/* Definition of range extension bits for decompression processes. - * See the comments with prepare_range_limit_table (in jdmaster.c) - * for more info. - * The recommended default value for normal applications is 2. - * Applications with special requirements may use a different value. - * For example, Ghostscript wants to use 3 for proper handling of - * wacky images with oversize coefficient values. - */ - -#define RANGE_BITS 2 -#define RANGE_CENTER (CENTERJSAMPLE << RANGE_BITS) - - -/* Miscellaneous useful macros */ - -#undef MAX -#define MAX(a,b) ((a) > (b) ? (a) : (b)) -#undef MIN -#define MIN(a,b) ((a) < (b) ? (a) : (b)) - - -/* We assume that right shift corresponds to signed division by 2 with - * rounding towards minus infinity. This is correct for typical "arithmetic - * shift" instructions that shift in copies of the sign bit. But some - * C compilers implement >> with an unsigned shift. For these machines you - * must define RIGHT_SHIFT_IS_UNSIGNED. - * RIGHT_SHIFT provides a proper signed right shift of an INT32 quantity. - * It is only applied with constant shift counts. SHIFT_TEMPS must be - * included in the variables of any routine using RIGHT_SHIFT. - */ - -#ifdef RIGHT_SHIFT_IS_UNSIGNED -#define SHIFT_TEMPS INT32 shift_temp; -#define RIGHT_SHIFT(x,shft) \ - ((shift_temp = (x)) < 0 ? \ - (shift_temp >> (shft)) | ((~((INT32) 0)) << (32-(shft))) : \ - (shift_temp >> (shft))) -#else -#define SHIFT_TEMPS -#define RIGHT_SHIFT(x,shft) ((x) >> (shft)) -#endif - -/* Descale and correctly round an INT32 value that's scaled by N bits. - * We assume RIGHT_SHIFT rounds towards minus infinity, so adding - * the fudge factor is correct for either sign of X. - */ - -#define DESCALE(x,n) RIGHT_SHIFT((x) + ((INT32) 1 << ((n)-1)), n) - - -/* Short forms of external names for systems with brain-damaged linkers. */ - -#ifdef NEED_SHORT_EXTERNAL_NAMES -#define jinit_compress_master jICompress -#define jinit_c_master_control jICMaster -#define jinit_c_main_controller jICMainC -#define jinit_c_prep_controller jICPrepC -#define jinit_c_coef_controller jICCoefC -#define jinit_color_converter jICColor -#define jinit_downsampler jIDownsampler -#define jinit_forward_dct jIFDCT -#define jinit_huff_encoder jIHEncoder -#define jinit_arith_encoder jIAEncoder -#define jinit_marker_writer jIMWriter -#define jinit_master_decompress jIDMaster -#define jinit_d_main_controller jIDMainC -#define jinit_d_coef_controller jIDCoefC -#define jinit_d_post_controller jIDPostC -#define jinit_input_controller jIInCtlr -#define jinit_marker_reader jIMReader -#define jinit_huff_decoder jIHDecoder -#define jinit_arith_decoder jIADecoder -#define jinit_inverse_dct jIIDCT -#define jinit_upsampler jIUpsampler -#define jinit_color_deconverter jIDColor -#define jinit_1pass_quantizer jI1Quant -#define jinit_2pass_quantizer jI2Quant -#define jinit_merged_upsampler jIMUpsampler -#define jinit_memory_mgr jIMemMgr -#define jdiv_round_up jDivRound -#define jround_up jRound -#define jzero_far jZeroFar -#define jcopy_sample_rows jCopySamples -#define jcopy_block_row jCopyBlocks -#define jpeg_zigzag_order jZIGTable -#define jpeg_natural_order jZAGTable -#define jpeg_natural_order7 jZAG7Table -#define jpeg_natural_order6 jZAG6Table -#define jpeg_natural_order5 jZAG5Table -#define jpeg_natural_order4 jZAG4Table -#define jpeg_natural_order3 jZAG3Table -#define jpeg_natural_order2 jZAG2Table -#define jpeg_aritab jAriTab -#endif /* NEED_SHORT_EXTERNAL_NAMES */ - - -/* On normal machines we can apply MEMCOPY() and MEMZERO() to sample arrays - * and coefficient-block arrays. This won't work on 80x86 because the arrays - * are FAR and we're assuming a small-pointer memory model. However, some - * DOS compilers provide far-pointer versions of memcpy() and memset() even - * in the small-model libraries. These will be used if USE_FMEM is defined. - * Otherwise, the routines in jutils.c do it the hard way. - */ - -#ifndef NEED_FAR_POINTERS /* normal case, same as regular macro */ -#define FMEMZERO(target,size) MEMZERO(target,size) -#else /* 80x86 case */ -#ifdef USE_FMEM -#define FMEMZERO(target,size) _fmemset((void FAR *)(target), 0, (size_t)(size)) -#else -EXTERN(void) jzero_far JPP((void FAR * target, size_t bytestozero)); -#define FMEMZERO(target,size) jzero_far(target, size) -#endif -#endif - - -/* Compression module initialization routines */ -EXTERN(void) jinit_compress_master JPP((j_compress_ptr cinfo)); -EXTERN(void) jinit_c_master_control JPP((j_compress_ptr cinfo, - boolean transcode_only)); -EXTERN(void) jinit_c_main_controller JPP((j_compress_ptr cinfo, - boolean need_full_buffer)); -EXTERN(void) jinit_c_prep_controller JPP((j_compress_ptr cinfo, - boolean need_full_buffer)); -EXTERN(void) jinit_c_coef_controller JPP((j_compress_ptr cinfo, - boolean need_full_buffer)); -EXTERN(void) jinit_color_converter JPP((j_compress_ptr cinfo)); -EXTERN(void) jinit_downsampler JPP((j_compress_ptr cinfo)); -EXTERN(void) jinit_forward_dct JPP((j_compress_ptr cinfo)); -EXTERN(void) jinit_huff_encoder JPP((j_compress_ptr cinfo)); -EXTERN(void) jinit_arith_encoder JPP((j_compress_ptr cinfo)); -EXTERN(void) jinit_marker_writer JPP((j_compress_ptr cinfo)); -/* Decompression module initialization routines */ -EXTERN(void) jinit_master_decompress JPP((j_decompress_ptr cinfo)); -EXTERN(void) jinit_d_main_controller JPP((j_decompress_ptr cinfo, - boolean need_full_buffer)); -EXTERN(void) jinit_d_coef_controller JPP((j_decompress_ptr cinfo, - boolean need_full_buffer)); -EXTERN(void) jinit_d_post_controller JPP((j_decompress_ptr cinfo, - boolean need_full_buffer)); -EXTERN(void) jinit_input_controller JPP((j_decompress_ptr cinfo)); -EXTERN(void) jinit_marker_reader JPP((j_decompress_ptr cinfo)); -EXTERN(void) jinit_huff_decoder JPP((j_decompress_ptr cinfo)); -EXTERN(void) jinit_arith_decoder JPP((j_decompress_ptr cinfo)); -EXTERN(void) jinit_inverse_dct JPP((j_decompress_ptr cinfo)); -EXTERN(void) jinit_upsampler JPP((j_decompress_ptr cinfo)); -EXTERN(void) jinit_color_deconverter JPP((j_decompress_ptr cinfo)); -EXTERN(void) jinit_1pass_quantizer JPP((j_decompress_ptr cinfo)); -EXTERN(void) jinit_2pass_quantizer JPP((j_decompress_ptr cinfo)); -EXTERN(void) jinit_merged_upsampler JPP((j_decompress_ptr cinfo)); -/* Memory manager initialization */ -EXTERN(void) jinit_memory_mgr JPP((j_common_ptr cinfo)); - -/* Utility routines in jutils.c */ -EXTERN(long) jdiv_round_up JPP((long a, long b)); -EXTERN(long) jround_up JPP((long a, long b)); -EXTERN(void) jcopy_sample_rows JPP((JSAMPARRAY input_array, - JSAMPARRAY output_array, - int num_rows, JDIMENSION num_cols)); -EXTERN(void) jcopy_block_row JPP((JBLOCKROW input_row, JBLOCKROW output_row, - JDIMENSION num_blocks)); -/* Constant tables in jutils.c */ -#if 0 /* This table is not actually needed in v6a */ -extern const int jpeg_zigzag_order[]; /* natural coef order to zigzag order */ -#endif -extern const int jpeg_natural_order[]; /* zigzag coef order to natural order */ -extern const int jpeg_natural_order7[]; /* zz to natural order for 7x7 block */ -extern const int jpeg_natural_order6[]; /* zz to natural order for 6x6 block */ -extern const int jpeg_natural_order5[]; /* zz to natural order for 5x5 block */ -extern const int jpeg_natural_order4[]; /* zz to natural order for 4x4 block */ -extern const int jpeg_natural_order3[]; /* zz to natural order for 3x3 block */ -extern const int jpeg_natural_order2[]; /* zz to natural order for 2x2 block */ - -/* Arithmetic coding probability estimation tables in jaricom.c */ -extern const INT32 jpeg_aritab[]; - -/* Suppress undefined-structure complaints if necessary. */ - -#ifdef INCOMPLETE_TYPES_BROKEN -#ifndef AM_MEMORY_MANAGER /* only jmemmgr.c defines these */ -struct jvirt_sarray_control { long dummy; }; -struct jvirt_barray_control { long dummy; }; -#endif -#endif /* INCOMPLETE_TYPES_BROKEN */ diff --git a/dep/libjpeg/include/jpeglib.h b/dep/libjpeg/include/jpeglib.h deleted file mode 100644 index e7e15ab2c..000000000 --- a/dep/libjpeg/include/jpeglib.h +++ /dev/null @@ -1,1183 +0,0 @@ -/* - * jpeglib.h - * - * Copyright (C) 1991-1998, Thomas G. Lane. - * Modified 2002-2022 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file defines the application interface for the JPEG library. - * Most applications using the library need only include this file, - * and perhaps jerror.h if they want to know the exact error codes. - */ - -#ifndef JPEGLIB_H -#define JPEGLIB_H - -/* - * First we include the configuration files that record how this - * installation of the JPEG library is set up. jconfig.h can be - * generated automatically for many systems. jmorecfg.h contains - * manual configuration options that most people need not worry about. - */ - -#ifndef JCONFIG_INCLUDED /* in case jinclude.h already did */ -#include "jconfig.h" /* widely used configuration options */ -#endif -#include "jmorecfg.h" /* seldom changed options */ - - -#ifdef __cplusplus -#ifndef DONT_USE_EXTERN_C -extern "C" { -#endif -#endif - -/* Version IDs for the JPEG library. - * Might be useful for tests like "#if JPEG_LIB_VERSION >= 90". - */ - -#define JPEG_LIB_VERSION 90 /* Compatibility version 9.0 */ -#define JPEG_LIB_VERSION_MAJOR 9 -#define JPEG_LIB_VERSION_MINOR 6 - - -/* Various constants determining the sizes of things. - * All of these are specified by the JPEG standard, - * so don't change them if you want to be compatible. - */ - -#define DCTSIZE 8 /* The basic DCT block is 8x8 coefficients */ -#define DCTSIZE2 64 /* DCTSIZE squared; # of elements in a block */ -#define NUM_QUANT_TBLS 4 /* Quantization tables are numbered 0..3 */ -#define NUM_HUFF_TBLS 4 /* Huffman tables are numbered 0..3 */ -#define NUM_ARITH_TBLS 16 /* Arith-coding tables are numbered 0..15 */ -#define MAX_COMPS_IN_SCAN 4 /* JPEG limit on # of components in one scan */ -#define MAX_SAMP_FACTOR 4 /* JPEG limit on sampling factors */ -/* Unfortunately, some bozo at Adobe saw no reason to be bound by the standard; - * the PostScript DCT filter can emit files with many more than 10 blocks/MCU. - * If you happen to run across such a file, you can up D_MAX_BLOCKS_IN_MCU - * to handle it. We even let you do this from the jconfig.h file. However, - * we strongly discourage changing C_MAX_BLOCKS_IN_MCU; just because Adobe - * sometimes emits noncompliant files doesn't mean you should too. - */ -#define C_MAX_BLOCKS_IN_MCU 10 /* compressor's limit on blocks per MCU */ -#ifndef D_MAX_BLOCKS_IN_MCU -#define D_MAX_BLOCKS_IN_MCU 10 /* decompressor's limit on blocks per MCU */ -#endif - - -/* Data structures for images (arrays of samples and of DCT coefficients). - * On 80x86 machines, the image arrays are too big for near pointers, - * but the pointer arrays can fit in near memory. - */ - -typedef JSAMPLE FAR *JSAMPROW; /* ptr to one image row of pixel samples. */ -typedef JSAMPROW *JSAMPARRAY; /* ptr to some rows (a 2-D sample array) */ -typedef JSAMPARRAY *JSAMPIMAGE; /* a 3-D sample array: top index is color */ - -typedef JCOEF JBLOCK[DCTSIZE2]; /* one block of coefficients */ -typedef JBLOCK FAR *JBLOCKROW; /* pointer to one row of coefficient blocks */ -typedef JBLOCKROW *JBLOCKARRAY; /* a 2-D array of coefficient blocks */ -typedef JBLOCKARRAY *JBLOCKIMAGE; /* a 3-D array of coefficient blocks */ - -typedef JCOEF FAR *JCOEFPTR; /* useful in a couple of places */ - - -/* Types for JPEG compression parameters and working tables. */ - - -/* DCT coefficient quantization tables. */ - -typedef struct { - /* This array gives the coefficient quantizers in natural array order - * (not the zigzag order in which they are stored in a JPEG DQT marker). - * CAUTION: IJG versions prior to v6a kept this array in zigzag order. - */ - UINT16 quantval[DCTSIZE2]; /* quantization step for each coefficient */ - /* This field is used only during compression. It's initialized FALSE when - * the table is created, and set TRUE when it's been output to the file. - * You could suppress output of a table by setting this to TRUE. - * (See jpeg_suppress_tables for an example.) - */ - boolean sent_table; /* TRUE when table has been output */ -} JQUANT_TBL; - - -/* Huffman coding tables. */ - -typedef struct { - /* These two fields directly represent the contents of a JPEG DHT marker */ - UINT8 bits[17]; /* bits[k] = # of symbols with codes of */ - /* length k bits; bits[0] is unused */ - UINT8 huffval[256]; /* The symbols, in order of incr code length */ - /* This field is used only during compression. It's initialized FALSE when - * the table is created, and set TRUE when it's been output to the file. - * You could suppress output of a table by setting this to TRUE. - * (See jpeg_suppress_tables for an example.) - */ - boolean sent_table; /* TRUE when table has been output */ -} JHUFF_TBL; - - -/* Basic info about one component (color channel). */ - -typedef struct { - /* These values are fixed over the whole image. */ - /* For compression, they must be supplied by parameter setup; */ - /* for decompression, they are read from the SOF marker. */ - int component_id; /* identifier for this component (0..255) */ - int component_index; /* its index in SOF or cinfo->comp_info[] */ - int h_samp_factor; /* horizontal sampling factor (1..4) */ - int v_samp_factor; /* vertical sampling factor (1..4) */ - int quant_tbl_no; /* quantization table selector (0..3) */ - /* These values may vary between scans. */ - /* For compression, they must be supplied by parameter setup; */ - /* for decompression, they are read from the SOS marker. */ - /* The decompressor output side may not use these variables. */ - int dc_tbl_no; /* DC entropy table selector (0..3) */ - int ac_tbl_no; /* AC entropy table selector (0..3) */ - - /* Remaining fields should be treated as private by applications. */ - - /* These values are computed during compression or decompression startup: */ - /* Component's size in DCT blocks. - * Any dummy blocks added to complete an MCU are not counted; therefore - * these values do not depend on whether a scan is interleaved or not. - */ - JDIMENSION width_in_blocks; - JDIMENSION height_in_blocks; - /* Size of a DCT block in samples, - * reflecting any scaling we choose to apply during the DCT step. - * Values from 1 to 16 are supported. - * Note that different components may receive different DCT scalings. - */ - int DCT_h_scaled_size; - int DCT_v_scaled_size; - /* The downsampled dimensions are the component's actual, unpadded number - * of samples at the main buffer (preprocessing/compression interface); - * DCT scaling is included, so - * downsampled_width = - * ceil(image_width * Hi/Hmax * DCT_h_scaled_size/block_size) - * and similarly for height. - */ - JDIMENSION downsampled_width; /* actual width in samples */ - JDIMENSION downsampled_height; /* actual height in samples */ - /* For decompression, in cases where some of the components will be - * ignored (eg grayscale output from YCbCr image), we can skip most - * computations for the unused components. - * For compression, some of the components will need further quantization - * scale by factor of 2 after DCT (eg BG_YCC output from normal RGB input). - * The field is first set TRUE for decompression, FALSE for compression - * in initial_setup, and then adapted in color conversion setup. - */ - boolean component_needed; - - /* These values are computed before starting a scan of the component. */ - /* The decompressor output side may not use these variables. */ - int MCU_width; /* number of blocks per MCU, horizontally */ - int MCU_height; /* number of blocks per MCU, vertically */ - int MCU_blocks; /* MCU_width * MCU_height */ - int MCU_sample_width; /* MCU width in samples: MCU_width * DCT_h_scaled_size */ - int last_col_width; /* # of non-dummy blocks across in last MCU */ - int last_row_height; /* # of non-dummy blocks down in last MCU */ - - /* Saved quantization table for component; NULL if none yet saved. - * See jdinput.c comments about the need for this information. - * This field is currently used only for decompression. - */ - JQUANT_TBL * quant_table; - - /* Private per-component storage for DCT or IDCT subsystem. */ - void * dct_table; -} jpeg_component_info; - - -/* The script for encoding a multiple-scan file is an array of these: */ - -typedef struct { - int comps_in_scan; /* number of components encoded in this scan */ - int component_index[MAX_COMPS_IN_SCAN]; /* their SOF/comp_info[] indexes */ - int Ss, Se; /* progressive JPEG spectral selection parms */ - int Ah, Al; /* progressive JPEG successive approx. parms */ -} jpeg_scan_info; - -/* The decompressor can save APPn and COM markers in a list of these: */ - -typedef struct jpeg_marker_struct FAR * jpeg_saved_marker_ptr; - -struct jpeg_marker_struct { - jpeg_saved_marker_ptr next; /* next in list, or NULL */ - UINT8 marker; /* marker code: JPEG_COM, or JPEG_APP0+n */ - unsigned int original_length; /* # bytes of data in the file */ - unsigned int data_length; /* # bytes of data saved at data[] */ - JOCTET FAR * data; /* the data contained in the marker */ - /* the marker length word is not counted in data_length or original_length */ -}; - -/* Known color spaces. */ - -typedef enum { - JCS_UNKNOWN, /* error/unspecified */ - JCS_GRAYSCALE, /* monochrome */ - JCS_RGB, /* red/green/blue, standard RGB (sRGB) */ - JCS_YCbCr, /* Y/Cb/Cr (also known as YUV), standard YCC */ - JCS_CMYK, /* C/M/Y/K */ - JCS_YCCK, /* Y/Cb/Cr/K */ - JCS_BG_RGB, /* big gamut red/green/blue, bg-sRGB */ - JCS_BG_YCC /* big gamut Y/Cb/Cr, bg-sYCC */ -} J_COLOR_SPACE; - -/* Supported color transforms. */ - -typedef enum { - JCT_NONE = 0, - JCT_SUBTRACT_GREEN = 1 -} J_COLOR_TRANSFORM; - -/* DCT/IDCT algorithm options. */ - -typedef enum { - JDCT_ISLOW, /* slow but accurate integer algorithm */ - JDCT_IFAST, /* faster, less accurate integer method */ - JDCT_FLOAT /* floating-point: accurate, fast on fast HW */ -} J_DCT_METHOD; - -#ifndef JDCT_DEFAULT /* may be overridden in jconfig.h */ -#define JDCT_DEFAULT JDCT_ISLOW -#endif -#ifndef JDCT_FASTEST /* may be overridden in jconfig.h */ -#define JDCT_FASTEST JDCT_IFAST -#endif - -/* Dithering options for decompression. */ - -typedef enum { - JDITHER_NONE, /* no dithering */ - JDITHER_ORDERED, /* simple ordered dither */ - JDITHER_FS /* Floyd-Steinberg error diffusion dither */ -} J_DITHER_MODE; - - -/* Common fields between JPEG compression and decompression master structs. */ - -#define jpeg_common_fields \ - struct jpeg_error_mgr * err; /* Error handler module */\ - struct jpeg_memory_mgr * mem; /* Memory manager module */\ - struct jpeg_progress_mgr * progress; /* Progress monitor, or NULL if none */\ - void * client_data; /* Available for use by application */\ - boolean is_decompressor; /* So common code can tell which is which */\ - int global_state /* For checking call sequence validity */ - -/* Routines that are to be used by both halves of the library are declared - * to receive a pointer to this structure. There are no actual instances of - * jpeg_common_struct, only of jpeg_compress_struct and jpeg_decompress_struct. - */ -struct jpeg_common_struct { - jpeg_common_fields; /* Fields common to both master struct types */ - /* Additional fields follow in an actual jpeg_compress_struct or - * jpeg_decompress_struct. All three structs must agree on these - * initial fields! (This would be a lot cleaner in C++.) - */ -}; - -typedef struct jpeg_common_struct * j_common_ptr; -typedef struct jpeg_compress_struct * j_compress_ptr; -typedef struct jpeg_decompress_struct * j_decompress_ptr; - - -/* Master record for a compression instance */ - -struct jpeg_compress_struct { - jpeg_common_fields; /* Fields shared with jpeg_decompress_struct */ - - /* Destination for compressed data */ - struct jpeg_destination_mgr * dest; - - /* Description of source image --- these fields must be filled in by - * outer application before starting compression. in_color_space must - * be correct before you can even call jpeg_set_defaults(). - */ - - JDIMENSION image_width; /* input image width */ - JDIMENSION image_height; /* input image height */ - int input_components; /* # of color components in input image */ - J_COLOR_SPACE in_color_space; /* colorspace of input image */ - - double input_gamma; /* image gamma of input image */ - - /* Compression parameters --- these fields must be set before calling - * jpeg_start_compress(). We recommend calling jpeg_set_defaults() to - * initialize everything to reasonable defaults, then changing anything - * the application specifically wants to change. That way you won't get - * burnt when new parameters are added. Also note that there are several - * helper routines to simplify changing parameters. - */ - - unsigned int scale_num, scale_denom; /* fraction by which to scale image */ - - JDIMENSION jpeg_width; /* scaled JPEG image width */ - JDIMENSION jpeg_height; /* scaled JPEG image height */ - /* Dimensions of actual JPEG image that will be written to file, - * derived from input dimensions by scaling factors above. - * These fields are computed by jpeg_start_compress(). - * You can also use jpeg_calc_jpeg_dimensions() to determine these values - * in advance of calling jpeg_start_compress(). - */ - - int data_precision; /* bits of precision in image data */ - - int num_components; /* # of color components in JPEG image */ - J_COLOR_SPACE jpeg_color_space; /* colorspace of JPEG image */ - - jpeg_component_info * comp_info; - /* comp_info[i] describes component that appears i'th in SOF */ - - JQUANT_TBL * quant_tbl_ptrs[NUM_QUANT_TBLS]; - int q_scale_factor[NUM_QUANT_TBLS]; - /* ptrs to coefficient quantization tables, or NULL if not defined, - * and corresponding scale factors (percentage, initialized 100). - */ - - JHUFF_TBL * dc_huff_tbl_ptrs[NUM_HUFF_TBLS]; - JHUFF_TBL * ac_huff_tbl_ptrs[NUM_HUFF_TBLS]; - /* ptrs to Huffman coding tables, or NULL if not defined */ - - UINT8 arith_dc_L[NUM_ARITH_TBLS]; /* L values for DC arith-coding tables */ - UINT8 arith_dc_U[NUM_ARITH_TBLS]; /* U values for DC arith-coding tables */ - UINT8 arith_ac_K[NUM_ARITH_TBLS]; /* Kx values for AC arith-coding tables */ - - int num_scans; /* # of entries in scan_info array */ - const jpeg_scan_info * scan_info; /* script for multi-scan file, or NULL */ - /* The default value of scan_info is NULL, which causes a single-scan - * sequential JPEG file to be emitted. To create a multi-scan file, - * set num_scans and scan_info to point to an array of scan definitions. - */ - - boolean raw_data_in; /* TRUE=caller supplies downsampled data */ - boolean arith_code; /* TRUE=arithmetic coding, FALSE=Huffman */ - boolean optimize_coding; /* TRUE=optimize entropy encoding parms */ - boolean CCIR601_sampling; /* TRUE=first samples are cosited */ - boolean do_fancy_downsampling; /* TRUE=apply fancy downsampling */ - int smoothing_factor; /* 1..100, or 0 for no input smoothing */ - J_DCT_METHOD dct_method; /* DCT algorithm selector */ - - /* The restart interval can be specified in absolute MCUs by setting - * restart_interval, or in MCU rows by setting restart_in_rows - * (in which case the correct restart_interval will be figured - * for each scan). - */ - unsigned int restart_interval; /* MCUs per restart, or 0 for no restart */ - int restart_in_rows; /* if > 0, MCU rows per restart interval */ - - /* Parameters controlling emission of special markers. */ - - boolean write_JFIF_header; /* should a JFIF marker be written? */ - UINT8 JFIF_major_version; /* What to write for the JFIF version number */ - UINT8 JFIF_minor_version; - /* These three values are not used by the JPEG code, merely copied */ - /* into the JFIF APP0 marker. density_unit can be 0 for unknown, */ - /* 1 for dots/inch, or 2 for dots/cm. Note that the pixel aspect */ - /* ratio is defined by X_density/Y_density even when density_unit=0. */ - UINT8 density_unit; /* JFIF code for pixel size units */ - UINT16 X_density; /* Horizontal pixel density */ - UINT16 Y_density; /* Vertical pixel density */ - boolean write_Adobe_marker; /* should an Adobe marker be written? */ - - J_COLOR_TRANSFORM color_transform; - /* Color transform identifier, writes LSE marker if nonzero */ - - /* State variable: index of next scanline to be written to - * jpeg_write_scanlines(). Application may use this to control its - * processing loop, e.g., "while (next_scanline < image_height)". - */ - - JDIMENSION next_scanline; /* 0 .. image_height-1 */ - - /* Remaining fields are known throughout compressor, but generally - * should not be touched by a surrounding application. - */ - - /* - * These fields are computed during compression startup - */ - boolean progressive_mode; /* TRUE if scan script uses progressive mode */ - int max_h_samp_factor; /* largest h_samp_factor */ - int max_v_samp_factor; /* largest v_samp_factor */ - - int min_DCT_h_scaled_size; /* smallest DCT_h_scaled_size of any component */ - int min_DCT_v_scaled_size; /* smallest DCT_v_scaled_size of any component */ - - JDIMENSION total_iMCU_rows; /* # of iMCU rows to be input to coef ctlr */ - /* The coefficient controller receives data in units of MCU rows as defined - * for fully interleaved scans (whether the JPEG file is interleaved or not). - * There are v_samp_factor * DCT_v_scaled_size sample rows of each component - * in an "iMCU" (interleaved MCU) row. - */ - - /* - * These fields are valid during any one scan. - * They describe the components and MCUs actually appearing in the scan. - */ - int comps_in_scan; /* # of JPEG components in this scan */ - jpeg_component_info * cur_comp_info[MAX_COMPS_IN_SCAN]; - /* *cur_comp_info[i] describes component that appears i'th in SOS */ - - JDIMENSION MCUs_per_row; /* # of MCUs across the image */ - JDIMENSION MCU_rows_in_scan; /* # of MCU rows in the image */ - - int blocks_in_MCU; /* # of DCT blocks per MCU */ - int MCU_membership[C_MAX_BLOCKS_IN_MCU]; - /* MCU_membership[i] is index in cur_comp_info of component owning */ - /* i'th block in an MCU */ - - int Ss, Se, Ah, Al; /* progressive JPEG parameters for scan */ - - int block_size; /* the basic DCT block size: 1..16 */ - const int * natural_order; /* natural-order position array */ - int lim_Se; /* min( Se, DCTSIZE2-1 ) */ - - /* - * Links to compression subobjects (methods and private variables of modules) - */ - struct jpeg_comp_master * master; - struct jpeg_c_main_controller * main; - struct jpeg_c_prep_controller * prep; - struct jpeg_c_coef_controller * coef; - struct jpeg_marker_writer * marker; - struct jpeg_color_converter * cconvert; - struct jpeg_downsampler * downsample; - struct jpeg_forward_dct * fdct; - struct jpeg_entropy_encoder * entropy; - jpeg_scan_info * script_space; /* workspace for jpeg_simple_progression */ - int script_space_size; -}; - - -/* Master record for a decompression instance */ - -struct jpeg_decompress_struct { - jpeg_common_fields; /* Fields shared with jpeg_compress_struct */ - - /* Source of compressed data */ - struct jpeg_source_mgr * src; - - /* Basic description of image --- filled in by jpeg_read_header(). */ - /* Application may inspect these values to decide how to process image. */ - - JDIMENSION image_width; /* nominal image width (from SOF marker) */ - JDIMENSION image_height; /* nominal image height */ - int num_components; /* # of color components in JPEG image */ - J_COLOR_SPACE jpeg_color_space; /* colorspace of JPEG image */ - - /* Decompression processing parameters --- these fields must be set before - * calling jpeg_start_decompress(). Note that jpeg_read_header() initializes - * them to default values. - */ - - J_COLOR_SPACE out_color_space; /* colorspace for output */ - - unsigned int scale_num, scale_denom; /* fraction by which to scale image */ - - double output_gamma; /* image gamma wanted in output */ - - boolean buffered_image; /* TRUE=multiple output passes */ - boolean raw_data_out; /* TRUE=downsampled data wanted */ - - J_DCT_METHOD dct_method; /* IDCT algorithm selector */ - boolean do_fancy_upsampling; /* TRUE=apply fancy upsampling */ - boolean do_block_smoothing; /* TRUE=apply interblock smoothing */ - - boolean quantize_colors; /* TRUE=colormapped output wanted */ - /* the following are ignored if not quantize_colors: */ - J_DITHER_MODE dither_mode; /* type of color dithering to use */ - boolean two_pass_quantize; /* TRUE=use two-pass color quantization */ - int desired_number_of_colors; /* max # colors to use in created colormap */ - /* these are significant only in buffered-image mode: */ - boolean enable_1pass_quant; /* enable future use of 1-pass quantizer */ - boolean enable_external_quant;/* enable future use of external colormap */ - boolean enable_2pass_quant; /* enable future use of 2-pass quantizer */ - - /* Description of actual output image that will be returned to application. - * These fields are computed by jpeg_start_decompress(). - * You can also use jpeg_calc_output_dimensions() to determine these values - * in advance of calling jpeg_start_decompress(). - */ - - JDIMENSION output_width; /* scaled image width */ - JDIMENSION output_height; /* scaled image height */ - int out_color_components; /* # of color components in out_color_space */ - int output_components; /* # of color components returned */ - /* output_components is 1 (a colormap index) when quantizing colors; - * otherwise it equals out_color_components. - */ - int rec_outbuf_height; /* min recommended height of scanline buffer */ - /* If the buffer passed to jpeg_read_scanlines() is less than this many rows - * high, space and time will be wasted due to unnecessary data copying. - * Usually rec_outbuf_height will be 1 or 2, at most 4. - */ - - /* When quantizing colors, the output colormap is described by these fields. - * The application can supply a colormap by setting colormap non-NULL before - * calling jpeg_start_decompress; otherwise a colormap is created during - * jpeg_start_decompress or jpeg_start_output. - * The map has out_color_components rows and actual_number_of_colors columns. - */ - int actual_number_of_colors; /* number of entries in use */ - JSAMPARRAY colormap; /* The color map as a 2-D pixel array */ - - /* State variables: these variables indicate the progress of decompression. - * The application may examine these but must not modify them. - */ - - /* Row index of next scanline to be read from jpeg_read_scanlines(). - * Application may use this to control its processing loop, e.g., - * "while (output_scanline < output_height)". - */ - JDIMENSION output_scanline; /* 0 .. output_height-1 */ - - /* Current input scan number and number of iMCU rows completed in scan. - * These indicate the progress of the decompressor input side. - */ - int input_scan_number; /* Number of SOS markers seen so far */ - JDIMENSION input_iMCU_row; /* Number of iMCU rows completed */ - - /* The "output scan number" is the notional scan being displayed by the - * output side. The decompressor will not allow output scan/row number - * to get ahead of input scan/row, but it can fall arbitrarily far behind. - */ - int output_scan_number; /* Nominal scan number being displayed */ - JDIMENSION output_iMCU_row; /* Number of iMCU rows read */ - - /* Current progression status. coef_bits[c][i] indicates the precision - * with which component c's DCT coefficient i (in zigzag order) is known. - * It is -1 when no data has yet been received, otherwise it is the point - * transform (shift) value for the most recent scan of the coefficient - * (thus, 0 at completion of the progression). - * This pointer is NULL when reading a non-progressive file. - */ - int (*coef_bits)[DCTSIZE2]; /* -1 or current Al value for each coef */ - - /* Internal JPEG parameters --- the application usually need not look at - * these fields. Note that the decompressor output side may not use - * any parameters that can change between scans. - */ - - /* Quantization and Huffman tables are carried forward across input - * datastreams when processing abbreviated JPEG datastreams. - */ - - JQUANT_TBL * quant_tbl_ptrs[NUM_QUANT_TBLS]; - /* ptrs to coefficient quantization tables, or NULL if not defined */ - - JHUFF_TBL * dc_huff_tbl_ptrs[NUM_HUFF_TBLS]; - JHUFF_TBL * ac_huff_tbl_ptrs[NUM_HUFF_TBLS]; - /* ptrs to Huffman coding tables, or NULL if not defined */ - - /* These parameters are never carried across datastreams, since they - * are given in SOF/SOS markers or defined to be reset by SOI. - */ - - int data_precision; /* bits of precision in image data */ - - jpeg_component_info * comp_info; - /* comp_info[i] describes component that appears i'th in SOF */ - - boolean is_baseline; /* TRUE if Baseline SOF0 encountered */ - boolean progressive_mode; /* TRUE if SOFn specifies progressive mode */ - boolean arith_code; /* TRUE=arithmetic coding, FALSE=Huffman */ - - UINT8 arith_dc_L[NUM_ARITH_TBLS]; /* L values for DC arith-coding tables */ - UINT8 arith_dc_U[NUM_ARITH_TBLS]; /* U values for DC arith-coding tables */ - UINT8 arith_ac_K[NUM_ARITH_TBLS]; /* Kx values for AC arith-coding tables */ - - unsigned int restart_interval; /* MCUs per restart interval, or 0 for no restart */ - - /* These fields record data obtained from optional markers recognized by - * the JPEG library. - */ - boolean saw_JFIF_marker; /* TRUE iff a JFIF APP0 marker was found */ - /* Data copied from JFIF marker; only valid if saw_JFIF_marker is TRUE: */ - UINT8 JFIF_major_version; /* JFIF version number */ - UINT8 JFIF_minor_version; - UINT8 density_unit; /* JFIF code for pixel size units */ - UINT16 X_density; /* Horizontal pixel density */ - UINT16 Y_density; /* Vertical pixel density */ - boolean saw_Adobe_marker; /* TRUE iff an Adobe APP14 marker was found */ - UINT8 Adobe_transform; /* Color transform code from Adobe marker */ - - J_COLOR_TRANSFORM color_transform; - /* Color transform identifier derived from LSE marker, otherwise zero */ - - boolean CCIR601_sampling; /* TRUE=first samples are cosited */ - - /* Aside from the specific data retained from APPn markers known to the - * library, the uninterpreted contents of any or all APPn and COM markers - * can be saved in a list for examination by the application. - */ - jpeg_saved_marker_ptr marker_list; /* Head of list of saved markers */ - - /* Remaining fields are known throughout decompressor, but generally - * should not be touched by a surrounding application. - */ - - /* - * These fields are computed during decompression startup - */ - int max_h_samp_factor; /* largest h_samp_factor */ - int max_v_samp_factor; /* largest v_samp_factor */ - - int min_DCT_h_scaled_size; /* smallest DCT_h_scaled_size of any component */ - int min_DCT_v_scaled_size; /* smallest DCT_v_scaled_size of any component */ - - JDIMENSION total_iMCU_rows; /* # of iMCU rows in image */ - /* The coefficient controller's input and output progress is measured in - * units of "iMCU" (interleaved MCU) rows. These are the same as MCU rows - * in fully interleaved JPEG scans, but are used whether the scan is - * interleaved or not. We define an iMCU row as v_samp_factor DCT block - * rows of each component. Therefore, the IDCT output contains - * v_samp_factor * DCT_v_scaled_size sample rows of a component per iMCU row. - */ - - JSAMPLE * sample_range_limit; /* table for fast range-limiting */ - - /* - * These fields are valid during any one scan. - * They describe the components and MCUs actually appearing in the scan. - * Note that the decompressor output side must not use these fields. - */ - int comps_in_scan; /* # of JPEG components in this scan */ - jpeg_component_info * cur_comp_info[MAX_COMPS_IN_SCAN]; - /* *cur_comp_info[i] describes component that appears i'th in SOS */ - - JDIMENSION MCUs_per_row; /* # of MCUs across the image */ - JDIMENSION MCU_rows_in_scan; /* # of MCU rows in the image */ - - int blocks_in_MCU; /* # of DCT blocks per MCU */ - int MCU_membership[D_MAX_BLOCKS_IN_MCU]; - /* MCU_membership[i] is index in cur_comp_info of component owning */ - /* i'th block in an MCU */ - - int Ss, Se, Ah, Al; /* progressive JPEG parameters for scan */ - - /* These fields are derived from Se of first SOS marker. - */ - int block_size; /* the basic DCT block size: 1..16 */ - const int * natural_order; /* natural-order position array for entropy decode */ - int lim_Se; /* min( Se, DCTSIZE2-1 ) for entropy decode */ - - /* This field is shared between entropy decoder and marker parser. - * It is either zero or the code of a JPEG marker that has been - * read from the data source, but has not yet been processed. - */ - int unread_marker; - - /* - * Links to decompression subobjects (methods, private variables of modules) - */ - struct jpeg_decomp_master * master; - struct jpeg_d_main_controller * main; - struct jpeg_d_coef_controller * coef; - struct jpeg_d_post_controller * post; - struct jpeg_input_controller * inputctl; - struct jpeg_marker_reader * marker; - struct jpeg_entropy_decoder * entropy; - struct jpeg_inverse_dct * idct; - struct jpeg_upsampler * upsample; - struct jpeg_color_deconverter * cconvert; - struct jpeg_color_quantizer * cquantize; -}; - - -/* "Object" declarations for JPEG modules that may be supplied or called - * directly by the surrounding application. - * As with all objects in the JPEG library, these structs only define the - * publicly visible methods and state variables of a module. Additional - * private fields may exist after the public ones. - */ - - -/* Error handler object */ - -struct jpeg_error_mgr { - /* Error exit handler: does not return to caller */ - JMETHOD(noreturn_t, error_exit, (j_common_ptr cinfo)); - /* Conditionally emit a trace or warning message */ - JMETHOD(void, emit_message, (j_common_ptr cinfo, int msg_level)); - /* Routine that actually outputs a trace or error message */ - JMETHOD(void, output_message, (j_common_ptr cinfo)); - /* Format a message string for the most recent JPEG error or message */ - JMETHOD(void, format_message, (j_common_ptr cinfo, char * buffer)); -#define JMSG_LENGTH_MAX 200 /* recommended size of format_message buffer */ - /* Reset error state variables at start of a new image */ - JMETHOD(void, reset_error_mgr, (j_common_ptr cinfo)); - - /* The message ID code and any parameters are saved here. - * A message can have one string parameter or up to 8 int parameters. - */ - int msg_code; -#define JMSG_STR_PARM_MAX 80 - union { - int i[8]; - char s[JMSG_STR_PARM_MAX]; - } msg_parm; - - /* Standard state variables for error facility */ - - int trace_level; /* max msg_level that will be displayed */ - - /* For recoverable corrupt-data errors, we emit a warning message, - * but keep going unless emit_message chooses to abort. emit_message - * should count warnings in num_warnings. The surrounding application - * can check for bad data by seeing if num_warnings is nonzero at the - * end of processing. - */ - long num_warnings; /* number of corrupt-data warnings */ - - /* These fields point to the table(s) of error message strings. - * An application can change the table pointer to switch to a different - * message list (typically, to change the language in which errors are - * reported). Some applications may wish to add additional error codes - * that will be handled by the JPEG library error mechanism; the second - * table pointer is used for this purpose. - * - * First table includes all errors generated by JPEG library itself. - * Error code 0 is reserved for a "no such error string" message. - */ - const char * const * jpeg_message_table; /* Library errors */ - int last_jpeg_message; /* Table contains strings 0..last_jpeg_message */ - /* Second table can be added by application (see cjpeg/djpeg for example). - * It contains strings numbered first_addon_message..last_addon_message. - */ - const char * const * addon_message_table; /* Non-library errors */ - int first_addon_message; /* code for first string in addon table */ - int last_addon_message; /* code for last string in addon table */ -}; - - -/* Progress monitor object */ - -struct jpeg_progress_mgr { - JMETHOD(void, progress_monitor, (j_common_ptr cinfo)); - - long pass_counter; /* work units completed in this pass */ - long pass_limit; /* total number of work units in this pass */ - int completed_passes; /* passes completed so far */ - int total_passes; /* total number of passes expected */ -}; - - -/* Data destination object for compression */ - -struct jpeg_destination_mgr { - JOCTET * next_output_byte; /* => next byte to write in buffer */ - size_t free_in_buffer; /* # of byte spaces remaining in buffer */ - - JMETHOD(void, init_destination, (j_compress_ptr cinfo)); - JMETHOD(boolean, empty_output_buffer, (j_compress_ptr cinfo)); - JMETHOD(void, term_destination, (j_compress_ptr cinfo)); -}; - - -/* Data source object for decompression */ - -struct jpeg_source_mgr { - const JOCTET * next_input_byte; /* => next byte to read from buffer */ - size_t bytes_in_buffer; /* # of bytes remaining in buffer */ - - JMETHOD(void, init_source, (j_decompress_ptr cinfo)); - JMETHOD(boolean, fill_input_buffer, (j_decompress_ptr cinfo)); - JMETHOD(void, skip_input_data, (j_decompress_ptr cinfo, long num_bytes)); - JMETHOD(boolean, resync_to_restart, (j_decompress_ptr cinfo, int desired)); - JMETHOD(void, term_source, (j_decompress_ptr cinfo)); -}; - - -/* Memory manager object. - * Allocates "small" objects (a few K total), "large" objects (tens of K), - * and "really big" objects (virtual arrays with backing store if needed). - * The memory manager does not allow individual objects to be freed; rather, - * each created object is assigned to a pool, and whole pools can be freed - * at once. This is faster and more convenient than remembering exactly what - * to free, especially where malloc()/free() are not too speedy. - * NB: alloc routines never return NULL. They exit to error_exit if not - * successful. - */ - -#define JPOOL_PERMANENT 0 /* lasts until master record is destroyed */ -#define JPOOL_IMAGE 1 /* lasts until done with image/datastream */ -#define JPOOL_NUMPOOLS 2 - -typedef struct jvirt_sarray_control * jvirt_sarray_ptr; -typedef struct jvirt_barray_control * jvirt_barray_ptr; - - -struct jpeg_memory_mgr { - /* Method pointers */ - JMETHOD(void *, alloc_small, (j_common_ptr cinfo, int pool_id, - size_t sizeofobject)); - JMETHOD(void FAR *, alloc_large, (j_common_ptr cinfo, int pool_id, - size_t sizeofobject)); - JMETHOD(JSAMPARRAY, alloc_sarray, (j_common_ptr cinfo, int pool_id, - JDIMENSION samplesperrow, - JDIMENSION numrows)); - JMETHOD(JBLOCKARRAY, alloc_barray, (j_common_ptr cinfo, int pool_id, - JDIMENSION blocksperrow, - JDIMENSION numrows)); - JMETHOD(jvirt_sarray_ptr, request_virt_sarray, (j_common_ptr cinfo, - int pool_id, - boolean pre_zero, - JDIMENSION samplesperrow, - JDIMENSION numrows, - JDIMENSION maxaccess)); - JMETHOD(jvirt_barray_ptr, request_virt_barray, (j_common_ptr cinfo, - int pool_id, - boolean pre_zero, - JDIMENSION blocksperrow, - JDIMENSION numrows, - JDIMENSION maxaccess)); - JMETHOD(void, realize_virt_arrays, (j_common_ptr cinfo)); - JMETHOD(JSAMPARRAY, access_virt_sarray, (j_common_ptr cinfo, - jvirt_sarray_ptr ptr, - JDIMENSION start_row, - JDIMENSION num_rows, - boolean writable)); - JMETHOD(JBLOCKARRAY, access_virt_barray, (j_common_ptr cinfo, - jvirt_barray_ptr ptr, - JDIMENSION start_row, - JDIMENSION num_rows, - boolean writable)); - JMETHOD(void, free_pool, (j_common_ptr cinfo, int pool_id)); - JMETHOD(void, self_destruct, (j_common_ptr cinfo)); - - /* Limit on memory allocation for this JPEG object. (Note that this is - * merely advisory, not a guaranteed maximum; it only affects the space - * used for virtual-array buffers.) May be changed by outer application - * after creating the JPEG object. - */ - long max_memory_to_use; - - /* Maximum allocation request accepted by alloc_large. */ - long max_alloc_chunk; -}; - - -/* Routine signature for application-supplied marker processing methods. - * Need not pass marker code since it is stored in cinfo->unread_marker. - */ -typedef JMETHOD(boolean, jpeg_marker_parser_method, (j_decompress_ptr cinfo)); - - -/* Declarations for routines called by application. - * The JPP macro hides prototype parameters from compilers that can't cope. - * Note JPP requires double parentheses. - */ - -#ifdef HAVE_PROTOTYPES -#define JPP(arglist) arglist -#else -#define JPP(arglist) () -#endif - - -/* Short forms of external names for systems with brain-damaged linkers. - * We shorten external names to be unique in the first six letters, which - * is good enough for all known systems. - * (If your compiler itself needs names to be unique in less than 15 - * characters, you are out of luck. Get a better compiler.) - */ - -#ifdef NEED_SHORT_EXTERNAL_NAMES -#define jpeg_std_error jStdError -#define jpeg_CreateCompress jCreaCompress -#define jpeg_CreateDecompress jCreaDecompress -#define jpeg_destroy_compress jDestCompress -#define jpeg_destroy_decompress jDestDecompress -#define jpeg_stdio_dest jStdDest -#define jpeg_stdio_src jStdSrc -#define jpeg_mem_dest jMemDest -#define jpeg_mem_src jMemSrc -#define jpeg_set_defaults jSetDefaults -#define jpeg_set_colorspace jSetColorspace -#define jpeg_default_colorspace jDefColorspace -#define jpeg_set_quality jSetQuality -#define jpeg_set_linear_quality jSetLQuality -#define jpeg_default_qtables jDefQTables -#define jpeg_add_quant_table jAddQuantTable -#define jpeg_quality_scaling jQualityScaling -#define jpeg_simple_progression jSimProgress -#define jpeg_suppress_tables jSuppressTables -#define jpeg_alloc_quant_table jAlcQTable -#define jpeg_alloc_huff_table jAlcHTable -#define jpeg_std_huff_table jStdHTable -#define jpeg_start_compress jStrtCompress -#define jpeg_write_scanlines jWrtScanlines -#define jpeg_finish_compress jFinCompress -#define jpeg_calc_jpeg_dimensions jCjpegDimensions -#define jpeg_write_raw_data jWrtRawData -#define jpeg_write_marker jWrtMarker -#define jpeg_write_m_header jWrtMHeader -#define jpeg_write_m_byte jWrtMByte -#define jpeg_write_tables jWrtTables -#define jpeg_read_header jReadHeader -#define jpeg_start_decompress jStrtDecompress -#define jpeg_read_scanlines jReadScanlines -#define jpeg_finish_decompress jFinDecompress -#define jpeg_read_raw_data jReadRawData -#define jpeg_has_multiple_scans jHasMultScn -#define jpeg_start_output jStrtOutput -#define jpeg_finish_output jFinOutput -#define jpeg_input_complete jInComplete -#define jpeg_new_colormap jNewCMap -#define jpeg_consume_input jConsumeInput -#define jpeg_core_output_dimensions jCoreDimensions -#define jpeg_calc_output_dimensions jCalcDimensions -#define jpeg_save_markers jSaveMarkers -#define jpeg_set_marker_processor jSetMarker -#define jpeg_read_coefficients jReadCoefs -#define jpeg_write_coefficients jWrtCoefs -#define jpeg_copy_critical_parameters jCopyCrit -#define jpeg_abort_compress jAbrtCompress -#define jpeg_abort_decompress jAbrtDecompress -#define jpeg_abort jAbort -#define jpeg_destroy jDestroy -#define jpeg_resync_to_restart jResyncRestart -#endif /* NEED_SHORT_EXTERNAL_NAMES */ - - -/* Default error-management setup */ -EXTERN(struct jpeg_error_mgr *) jpeg_std_error - JPP((struct jpeg_error_mgr * err)); - -/* Initialization of JPEG compression objects. - * jpeg_create_compress() and jpeg_create_decompress() are the exported - * names that applications should call. These expand to calls on - * jpeg_CreateCompress and jpeg_CreateDecompress with additional information - * passed for version mismatch checking. - * NB: you must set up the error-manager BEFORE calling jpeg_create_xxx. - */ -#define jpeg_create_compress(cinfo) \ - jpeg_CreateCompress((cinfo), JPEG_LIB_VERSION, \ - (size_t) sizeof(struct jpeg_compress_struct)) -#define jpeg_create_decompress(cinfo) \ - jpeg_CreateDecompress((cinfo), JPEG_LIB_VERSION, \ - (size_t) sizeof(struct jpeg_decompress_struct)) -EXTERN(void) jpeg_CreateCompress JPP((j_compress_ptr cinfo, - int version, size_t structsize)); -EXTERN(void) jpeg_CreateDecompress JPP((j_decompress_ptr cinfo, - int version, size_t structsize)); -/* Destruction of JPEG compression objects */ -EXTERN(void) jpeg_destroy_compress JPP((j_compress_ptr cinfo)); -EXTERN(void) jpeg_destroy_decompress JPP((j_decompress_ptr cinfo)); - -/* Standard data source and destination managers: stdio streams. */ -/* Caller is responsible for opening the file before and closing after. */ -EXTERN(void) jpeg_stdio_dest JPP((j_compress_ptr cinfo, FILE * outfile)); -EXTERN(void) jpeg_stdio_src JPP((j_decompress_ptr cinfo, FILE * infile)); - -/* Data source and destination managers: memory buffers. */ -EXTERN(void) jpeg_mem_dest JPP((j_compress_ptr cinfo, - unsigned char ** outbuffer, - size_t * outsize)); -EXTERN(void) jpeg_mem_src JPP((j_decompress_ptr cinfo, - const unsigned char * inbuffer, - size_t insize)); - -/* Default parameter setup for compression */ -EXTERN(void) jpeg_set_defaults JPP((j_compress_ptr cinfo)); -/* Compression parameter setup aids */ -EXTERN(void) jpeg_set_colorspace JPP((j_compress_ptr cinfo, - J_COLOR_SPACE colorspace)); -EXTERN(void) jpeg_default_colorspace JPP((j_compress_ptr cinfo)); -EXTERN(void) jpeg_set_quality JPP((j_compress_ptr cinfo, int quality, - boolean force_baseline)); -EXTERN(void) jpeg_set_linear_quality JPP((j_compress_ptr cinfo, - int scale_factor, - boolean force_baseline)); -EXTERN(void) jpeg_default_qtables JPP((j_compress_ptr cinfo, - boolean force_baseline)); -EXTERN(void) jpeg_add_quant_table JPP((j_compress_ptr cinfo, int which_tbl, - const unsigned int *basic_table, - int scale_factor, - boolean force_baseline)); -EXTERN(int) jpeg_quality_scaling JPP((int quality)); -EXTERN(void) jpeg_simple_progression JPP((j_compress_ptr cinfo)); -EXTERN(void) jpeg_suppress_tables JPP((j_compress_ptr cinfo, - boolean suppress)); -EXTERN(JQUANT_TBL *) jpeg_alloc_quant_table JPP((j_common_ptr cinfo)); -EXTERN(JHUFF_TBL *) jpeg_alloc_huff_table JPP((j_common_ptr cinfo)); -EXTERN(JHUFF_TBL *) jpeg_std_huff_table JPP((j_common_ptr cinfo, - boolean isDC, int tblno)); - -/* Main entry points for compression */ -EXTERN(void) jpeg_start_compress JPP((j_compress_ptr cinfo, - boolean write_all_tables)); -EXTERN(JDIMENSION) jpeg_write_scanlines JPP((j_compress_ptr cinfo, - JSAMPARRAY scanlines, - JDIMENSION num_lines)); -EXTERN(void) jpeg_finish_compress JPP((j_compress_ptr cinfo)); - -/* Precalculate JPEG dimensions for current compression parameters. */ -EXTERN(void) jpeg_calc_jpeg_dimensions JPP((j_compress_ptr cinfo)); - -/* Replaces jpeg_write_scanlines when writing raw downsampled data. */ -EXTERN(JDIMENSION) jpeg_write_raw_data JPP((j_compress_ptr cinfo, - JSAMPIMAGE data, - JDIMENSION num_lines)); - -/* Write a special marker. See libjpeg.txt concerning safe usage. */ -EXTERN(void) jpeg_write_marker - JPP((j_compress_ptr cinfo, int marker, - const JOCTET * dataptr, unsigned int datalen)); -/* Same, but piecemeal. */ -EXTERN(void) jpeg_write_m_header - JPP((j_compress_ptr cinfo, int marker, unsigned int datalen)); -EXTERN(void) jpeg_write_m_byte - JPP((j_compress_ptr cinfo, int val)); - -/* Alternate compression function: just write an abbreviated table file */ -EXTERN(void) jpeg_write_tables JPP((j_compress_ptr cinfo)); - -/* Decompression startup: read start of JPEG datastream to see what's there */ -EXTERN(int) jpeg_read_header JPP((j_decompress_ptr cinfo, - boolean require_image)); -/* Return value is one of: */ -#define JPEG_SUSPENDED 0 /* Suspended due to lack of input data */ -#define JPEG_HEADER_OK 1 /* Found valid image datastream */ -#define JPEG_HEADER_TABLES_ONLY 2 /* Found valid table-specs-only datastream */ -/* If you pass require_image = TRUE (normal case), you need not check for - * a TABLES_ONLY return code; an abbreviated file will cause an error exit. - * JPEG_SUSPENDED is only possible if you use a data source module that can - * give a suspension return (the stdio source module doesn't). - */ - -/* Main entry points for decompression */ -EXTERN(boolean) jpeg_start_decompress JPP((j_decompress_ptr cinfo)); -EXTERN(JDIMENSION) jpeg_read_scanlines JPP((j_decompress_ptr cinfo, - JSAMPARRAY scanlines, - JDIMENSION max_lines)); -EXTERN(boolean) jpeg_finish_decompress JPP((j_decompress_ptr cinfo)); - -/* Replaces jpeg_read_scanlines when reading raw downsampled data. */ -EXTERN(JDIMENSION) jpeg_read_raw_data JPP((j_decompress_ptr cinfo, - JSAMPIMAGE data, - JDIMENSION max_lines)); - -/* Additional entry points for buffered-image mode. */ -EXTERN(boolean) jpeg_has_multiple_scans JPP((j_decompress_ptr cinfo)); -EXTERN(boolean) jpeg_start_output JPP((j_decompress_ptr cinfo, - int scan_number)); -EXTERN(boolean) jpeg_finish_output JPP((j_decompress_ptr cinfo)); -EXTERN(boolean) jpeg_input_complete JPP((j_decompress_ptr cinfo)); -EXTERN(void) jpeg_new_colormap JPP((j_decompress_ptr cinfo)); -EXTERN(int) jpeg_consume_input JPP((j_decompress_ptr cinfo)); -/* Return value is one of: */ -/* #define JPEG_SUSPENDED 0 Suspended due to lack of input data */ -#define JPEG_REACHED_SOS 1 /* Reached start of new scan */ -#define JPEG_REACHED_EOI 2 /* Reached end of image */ -#define JPEG_ROW_COMPLETED 3 /* Completed one iMCU row */ -#define JPEG_SCAN_COMPLETED 4 /* Completed last iMCU row of a scan */ - -/* Precalculate output dimensions for current decompression parameters. */ -EXTERN(void) jpeg_core_output_dimensions JPP((j_decompress_ptr cinfo)); -EXTERN(void) jpeg_calc_output_dimensions JPP((j_decompress_ptr cinfo)); - -/* Control saving of COM and APPn markers into marker_list. */ -EXTERN(void) jpeg_save_markers - JPP((j_decompress_ptr cinfo, int marker_code, - unsigned int length_limit)); - -/* Install a special processing method for COM or APPn markers. */ -EXTERN(void) jpeg_set_marker_processor - JPP((j_decompress_ptr cinfo, int marker_code, - jpeg_marker_parser_method routine)); - -/* Read or write raw DCT coefficients --- useful for lossless transcoding. */ -EXTERN(jvirt_barray_ptr *) jpeg_read_coefficients JPP((j_decompress_ptr cinfo)); -EXTERN(void) jpeg_write_coefficients JPP((j_compress_ptr cinfo, - jvirt_barray_ptr * coef_arrays)); -EXTERN(void) jpeg_copy_critical_parameters JPP((j_decompress_ptr srcinfo, - j_compress_ptr dstinfo)); - -/* If you choose to abort compression or decompression before completing - * jpeg_finish_(de)compress, then you need to clean up to release memory, - * temporary files, etc. You can just call jpeg_destroy_(de)compress - * if you're done with the JPEG object, but if you want to clean it up and - * reuse it, call this: - */ -EXTERN(void) jpeg_abort_compress JPP((j_compress_ptr cinfo)); -EXTERN(void) jpeg_abort_decompress JPP((j_decompress_ptr cinfo)); - -/* Generic versions of jpeg_abort and jpeg_destroy that work on either - * flavor of JPEG object. These may be more convenient in some places. - */ -EXTERN(void) jpeg_abort JPP((j_common_ptr cinfo)); -EXTERN(void) jpeg_destroy JPP((j_common_ptr cinfo)); - -/* Default restart-marker-resync procedure for use by data source modules */ -EXTERN(boolean) jpeg_resync_to_restart JPP((j_decompress_ptr cinfo, - int desired)); - - -/* These marker codes are exported since applications and data source modules - * are likely to want to use them. - */ - -#define JPEG_RST0 0xD0 /* RST0 marker code */ -#define JPEG_EOI 0xD9 /* EOI marker code */ -#define JPEG_APP0 0xE0 /* APP0 marker code */ -#define JPEG_COM 0xFE /* COM marker code */ - - -/* If we have a brain-damaged compiler that emits warnings (or worse, errors) - * for structure definitions that are never filled in, keep it quiet by - * supplying dummy definitions for the various substructures. - */ - -#ifdef INCOMPLETE_TYPES_BROKEN -#ifndef JPEG_INTERNALS /* will be defined in jpegint.h */ -struct jvirt_sarray_control { long dummy; }; -struct jvirt_barray_control { long dummy; }; -struct jpeg_comp_master { long dummy; }; -struct jpeg_c_main_controller { long dummy; }; -struct jpeg_c_prep_controller { long dummy; }; -struct jpeg_c_coef_controller { long dummy; }; -struct jpeg_marker_writer { long dummy; }; -struct jpeg_color_converter { long dummy; }; -struct jpeg_downsampler { long dummy; }; -struct jpeg_forward_dct { long dummy; }; -struct jpeg_entropy_encoder { long dummy; }; -struct jpeg_decomp_master { long dummy; }; -struct jpeg_d_main_controller { long dummy; }; -struct jpeg_d_coef_controller { long dummy; }; -struct jpeg_d_post_controller { long dummy; }; -struct jpeg_input_controller { long dummy; }; -struct jpeg_marker_reader { long dummy; }; -struct jpeg_entropy_decoder { long dummy; }; -struct jpeg_inverse_dct { long dummy; }; -struct jpeg_upsampler { long dummy; }; -struct jpeg_color_deconverter { long dummy; }; -struct jpeg_color_quantizer { long dummy; }; -#endif /* JPEG_INTERNALS */ -#endif /* INCOMPLETE_TYPES_BROKEN */ - - -/* - * The JPEG library modules define JPEG_INTERNALS before including this file. - * The internal structure declarations are read only when that is true. - * Applications using the library should not include jpegint.h, but may wish - * to include jerror.h. - */ - -#ifdef JPEG_INTERNALS -#include "jpegint.h" /* fetch private declarations */ -#include "jerror.h" /* fetch error codes too */ -#endif - -#ifdef __cplusplus -#ifndef DONT_USE_EXTERN_C -} -#endif -#endif - -#endif /* JPEGLIB_H */ diff --git a/dep/libjpeg/libjpeg.vcxproj b/dep/libjpeg/libjpeg.vcxproj deleted file mode 100644 index 655271d41..000000000 --- a/dep/libjpeg/libjpeg.vcxproj +++ /dev/null @@ -1,76 +0,0 @@ - - - - - {EC3B6685-0B6E-4767-84AB-39B75EEAD2E2} - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - TurnOffAllWarnings - $(ProjectDir)include;$(ProjectDir)src;%(AdditionalIncludeDirectories) - - - - \ No newline at end of file diff --git a/dep/libjpeg/libjpeg.vcxproj.filters b/dep/libjpeg/libjpeg.vcxproj.filters deleted file mode 100644 index 9b5b8ef17..000000000 --- a/dep/libjpeg/libjpeg.vcxproj.filters +++ /dev/null @@ -1,52 +0,0 @@ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - intel - - - intel - - - arm - - - arm - - - arm - - - - - {9f24e95e-025d-4ed8-8c41-2fb1c7a36026} - - - {8316b9c1-8c00-4bc8-ace7-c9b864890f2d} - - - \ No newline at end of file diff --git a/dep/libjpeg/src/jaricom.c b/dep/libjpeg/src/jaricom.c deleted file mode 100644 index 690068861..000000000 --- a/dep/libjpeg/src/jaricom.c +++ /dev/null @@ -1,153 +0,0 @@ -/* - * jaricom.c - * - * Developed 1997-2011 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains probability estimation tables for common use in - * arithmetic entropy encoding and decoding routines. - * - * This data represents Table D.3 in the JPEG spec (D.2 in the draft), - * ISO/IEC IS 10918-1 and CCITT Recommendation ITU-T T.81, and Table 24 - * in the JBIG spec, ISO/IEC IS 11544 and CCITT Recommendation ITU-T T.82. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - -/* The following #define specifies the packing of the four components - * into the compact INT32 representation. - * Note that this formula must match the actual arithmetic encoder - * and decoder implementation. The implementation has to be changed - * if this formula is changed. - * The current organization is leaned on Markus Kuhn's JBIG - * implementation (jbig_tab.c). - */ - -#define V(i,a,b,c,d) (((INT32)a << 16) | ((INT32)c << 8) | ((INT32)d << 7) | b) - -const INT32 jpeg_aritab[113+1] = { -/* - * Index, Qe_Value, Next_Index_LPS, Next_Index_MPS, Switch_MPS - */ - V( 0, 0x5a1d, 1, 1, 1 ), - V( 1, 0x2586, 14, 2, 0 ), - V( 2, 0x1114, 16, 3, 0 ), - V( 3, 0x080b, 18, 4, 0 ), - V( 4, 0x03d8, 20, 5, 0 ), - V( 5, 0x01da, 23, 6, 0 ), - V( 6, 0x00e5, 25, 7, 0 ), - V( 7, 0x006f, 28, 8, 0 ), - V( 8, 0x0036, 30, 9, 0 ), - V( 9, 0x001a, 33, 10, 0 ), - V( 10, 0x000d, 35, 11, 0 ), - V( 11, 0x0006, 9, 12, 0 ), - V( 12, 0x0003, 10, 13, 0 ), - V( 13, 0x0001, 12, 13, 0 ), - V( 14, 0x5a7f, 15, 15, 1 ), - V( 15, 0x3f25, 36, 16, 0 ), - V( 16, 0x2cf2, 38, 17, 0 ), - V( 17, 0x207c, 39, 18, 0 ), - V( 18, 0x17b9, 40, 19, 0 ), - V( 19, 0x1182, 42, 20, 0 ), - V( 20, 0x0cef, 43, 21, 0 ), - V( 21, 0x09a1, 45, 22, 0 ), - V( 22, 0x072f, 46, 23, 0 ), - V( 23, 0x055c, 48, 24, 0 ), - V( 24, 0x0406, 49, 25, 0 ), - V( 25, 0x0303, 51, 26, 0 ), - V( 26, 0x0240, 52, 27, 0 ), - V( 27, 0x01b1, 54, 28, 0 ), - V( 28, 0x0144, 56, 29, 0 ), - V( 29, 0x00f5, 57, 30, 0 ), - V( 30, 0x00b7, 59, 31, 0 ), - V( 31, 0x008a, 60, 32, 0 ), - V( 32, 0x0068, 62, 33, 0 ), - V( 33, 0x004e, 63, 34, 0 ), - V( 34, 0x003b, 32, 35, 0 ), - V( 35, 0x002c, 33, 9, 0 ), - V( 36, 0x5ae1, 37, 37, 1 ), - V( 37, 0x484c, 64, 38, 0 ), - V( 38, 0x3a0d, 65, 39, 0 ), - V( 39, 0x2ef1, 67, 40, 0 ), - V( 40, 0x261f, 68, 41, 0 ), - V( 41, 0x1f33, 69, 42, 0 ), - V( 42, 0x19a8, 70, 43, 0 ), - V( 43, 0x1518, 72, 44, 0 ), - V( 44, 0x1177, 73, 45, 0 ), - V( 45, 0x0e74, 74, 46, 0 ), - V( 46, 0x0bfb, 75, 47, 0 ), - V( 47, 0x09f8, 77, 48, 0 ), - V( 48, 0x0861, 78, 49, 0 ), - V( 49, 0x0706, 79, 50, 0 ), - V( 50, 0x05cd, 48, 51, 0 ), - V( 51, 0x04de, 50, 52, 0 ), - V( 52, 0x040f, 50, 53, 0 ), - V( 53, 0x0363, 51, 54, 0 ), - V( 54, 0x02d4, 52, 55, 0 ), - V( 55, 0x025c, 53, 56, 0 ), - V( 56, 0x01f8, 54, 57, 0 ), - V( 57, 0x01a4, 55, 58, 0 ), - V( 58, 0x0160, 56, 59, 0 ), - V( 59, 0x0125, 57, 60, 0 ), - V( 60, 0x00f6, 58, 61, 0 ), - V( 61, 0x00cb, 59, 62, 0 ), - V( 62, 0x00ab, 61, 63, 0 ), - V( 63, 0x008f, 61, 32, 0 ), - V( 64, 0x5b12, 65, 65, 1 ), - V( 65, 0x4d04, 80, 66, 0 ), - V( 66, 0x412c, 81, 67, 0 ), - V( 67, 0x37d8, 82, 68, 0 ), - V( 68, 0x2fe8, 83, 69, 0 ), - V( 69, 0x293c, 84, 70, 0 ), - V( 70, 0x2379, 86, 71, 0 ), - V( 71, 0x1edf, 87, 72, 0 ), - V( 72, 0x1aa9, 87, 73, 0 ), - V( 73, 0x174e, 72, 74, 0 ), - V( 74, 0x1424, 72, 75, 0 ), - V( 75, 0x119c, 74, 76, 0 ), - V( 76, 0x0f6b, 74, 77, 0 ), - V( 77, 0x0d51, 75, 78, 0 ), - V( 78, 0x0bb6, 77, 79, 0 ), - V( 79, 0x0a40, 77, 48, 0 ), - V( 80, 0x5832, 80, 81, 1 ), - V( 81, 0x4d1c, 88, 82, 0 ), - V( 82, 0x438e, 89, 83, 0 ), - V( 83, 0x3bdd, 90, 84, 0 ), - V( 84, 0x34ee, 91, 85, 0 ), - V( 85, 0x2eae, 92, 86, 0 ), - V( 86, 0x299a, 93, 87, 0 ), - V( 87, 0x2516, 86, 71, 0 ), - V( 88, 0x5570, 88, 89, 1 ), - V( 89, 0x4ca9, 95, 90, 0 ), - V( 90, 0x44d9, 96, 91, 0 ), - V( 91, 0x3e22, 97, 92, 0 ), - V( 92, 0x3824, 99, 93, 0 ), - V( 93, 0x32b4, 99, 94, 0 ), - V( 94, 0x2e17, 93, 86, 0 ), - V( 95, 0x56a8, 95, 96, 1 ), - V( 96, 0x4f46, 101, 97, 0 ), - V( 97, 0x47e5, 102, 98, 0 ), - V( 98, 0x41cf, 103, 99, 0 ), - V( 99, 0x3c3d, 104, 100, 0 ), - V( 100, 0x375e, 99, 93, 0 ), - V( 101, 0x5231, 105, 102, 0 ), - V( 102, 0x4c0f, 106, 103, 0 ), - V( 103, 0x4639, 107, 104, 0 ), - V( 104, 0x415e, 103, 99, 0 ), - V( 105, 0x5627, 105, 106, 1 ), - V( 106, 0x50e7, 108, 107, 0 ), - V( 107, 0x4b85, 109, 103, 0 ), - V( 108, 0x5597, 110, 109, 0 ), - V( 109, 0x504f, 111, 107, 0 ), - V( 110, 0x5a10, 110, 111, 1 ), - V( 111, 0x5522, 112, 109, 0 ), - V( 112, 0x59eb, 112, 111, 1 ), -/* - * This last entry is used for fixed probability estimate of 0.5 - * as suggested in Section 10.3 Table 5 of ITU-T Rec. T.851. - */ - V( 113, 0x5a1d, 113, 113, 0 ) -}; diff --git a/dep/libjpeg/src/jcapimin.c b/dep/libjpeg/src/jcapimin.c deleted file mode 100644 index 639ce86f4..000000000 --- a/dep/libjpeg/src/jcapimin.c +++ /dev/null @@ -1,288 +0,0 @@ -/* - * jcapimin.c - * - * Copyright (C) 1994-1998, Thomas G. Lane. - * Modified 2003-2010 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains application interface code for the compression half - * of the JPEG library. These are the "minimum" API routines that may be - * needed in either the normal full-compression case or the transcoding-only - * case. - * - * Most of the routines intended to be called directly by an application - * are in this file or in jcapistd.c. But also see jcparam.c for - * parameter-setup helper routines, jcomapi.c for routines shared by - * compression and decompression, and jctrans.c for the transcoding case. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* - * Initialization of a JPEG compression object. - * The error manager must already be set up (in case memory manager fails). - */ - -GLOBAL(void) -jpeg_CreateCompress (j_compress_ptr cinfo, int version, size_t structsize) -{ - int i; - - /* Guard against version mismatches between library and caller. */ - cinfo->mem = NULL; /* so jpeg_destroy knows mem mgr not called */ - if (version != JPEG_LIB_VERSION) - ERREXIT2(cinfo, JERR_BAD_LIB_VERSION, JPEG_LIB_VERSION, version); - if (structsize != SIZEOF(struct jpeg_compress_struct)) - ERREXIT2(cinfo, JERR_BAD_STRUCT_SIZE, - (int) SIZEOF(struct jpeg_compress_struct), (int) structsize); - - /* For debugging purposes, we zero the whole master structure. - * But the application has already set the err pointer, and may have set - * client_data, so we have to save and restore those fields. - * Note: if application hasn't set client_data, tools like Purify may - * complain here. - */ - { - struct jpeg_error_mgr * err = cinfo->err; - void * client_data = cinfo->client_data; /* ignore Purify complaint here */ - MEMZERO(cinfo, SIZEOF(struct jpeg_compress_struct)); - cinfo->err = err; - cinfo->client_data = client_data; - } - cinfo->is_decompressor = FALSE; - - /* Initialize a memory manager instance for this object */ - jinit_memory_mgr((j_common_ptr) cinfo); - - /* Zero out pointers to permanent structures. */ - cinfo->progress = NULL; - cinfo->dest = NULL; - - cinfo->comp_info = NULL; - - for (i = 0; i < NUM_QUANT_TBLS; i++) { - cinfo->quant_tbl_ptrs[i] = NULL; - cinfo->q_scale_factor[i] = 100; - } - - for (i = 0; i < NUM_HUFF_TBLS; i++) { - cinfo->dc_huff_tbl_ptrs[i] = NULL; - cinfo->ac_huff_tbl_ptrs[i] = NULL; - } - - /* Must do it here for emit_dqt in case jpeg_write_tables is used */ - cinfo->block_size = DCTSIZE; - cinfo->natural_order = jpeg_natural_order; - cinfo->lim_Se = DCTSIZE2-1; - - cinfo->script_space = NULL; - - cinfo->input_gamma = 1.0; /* in case application forgets */ - - /* OK, I'm ready */ - cinfo->global_state = CSTATE_START; -} - - -/* - * Destruction of a JPEG compression object - */ - -GLOBAL(void) -jpeg_destroy_compress (j_compress_ptr cinfo) -{ - jpeg_destroy((j_common_ptr) cinfo); /* use common routine */ -} - - -/* - * Abort processing of a JPEG compression operation, - * but don't destroy the object itself. - */ - -GLOBAL(void) -jpeg_abort_compress (j_compress_ptr cinfo) -{ - jpeg_abort((j_common_ptr) cinfo); /* use common routine */ -} - - -/* - * Forcibly suppress or un-suppress all quantization and Huffman tables. - * Marks all currently defined tables as already written (if suppress) - * or not written (if !suppress). This will control whether they get emitted - * by a subsequent jpeg_start_compress call. - * - * This routine is exported for use by applications that want to produce - * abbreviated JPEG datastreams. It logically belongs in jcparam.c, but - * since it is called by jpeg_start_compress, we put it here --- otherwise - * jcparam.o would be linked whether the application used it or not. - */ - -GLOBAL(void) -jpeg_suppress_tables (j_compress_ptr cinfo, boolean suppress) -{ - int i; - JQUANT_TBL * qtbl; - JHUFF_TBL * htbl; - - for (i = 0; i < NUM_QUANT_TBLS; i++) { - if ((qtbl = cinfo->quant_tbl_ptrs[i]) != NULL) - qtbl->sent_table = suppress; - } - - for (i = 0; i < NUM_HUFF_TBLS; i++) { - if ((htbl = cinfo->dc_huff_tbl_ptrs[i]) != NULL) - htbl->sent_table = suppress; - if ((htbl = cinfo->ac_huff_tbl_ptrs[i]) != NULL) - htbl->sent_table = suppress; - } -} - - -/* - * Finish JPEG compression. - * - * If a multipass operating mode was selected, this may do a great deal of - * work including most of the actual output. - */ - -GLOBAL(void) -jpeg_finish_compress (j_compress_ptr cinfo) -{ - JDIMENSION iMCU_row; - - if (cinfo->global_state == CSTATE_SCANNING || - cinfo->global_state == CSTATE_RAW_OK) { - /* Terminate first pass */ - if (cinfo->next_scanline < cinfo->image_height) - ERREXIT(cinfo, JERR_TOO_LITTLE_DATA); - (*cinfo->master->finish_pass) (cinfo); - } else if (cinfo->global_state != CSTATE_WRCOEFS) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - /* Perform any remaining passes */ - while (! cinfo->master->is_last_pass) { - (*cinfo->master->prepare_for_pass) (cinfo); - for (iMCU_row = 0; iMCU_row < cinfo->total_iMCU_rows; iMCU_row++) { - if (cinfo->progress != NULL) { - cinfo->progress->pass_counter = (long) iMCU_row; - cinfo->progress->pass_limit = (long) cinfo->total_iMCU_rows; - (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo); - } - /* We bypass the main controller and invoke coef controller directly; - * all work is being done from the coefficient buffer. - */ - if (! (*cinfo->coef->compress_data) (cinfo, (JSAMPIMAGE) NULL)) - ERREXIT(cinfo, JERR_CANT_SUSPEND); - } - (*cinfo->master->finish_pass) (cinfo); - } - /* Write EOI, do final cleanup */ - (*cinfo->marker->write_file_trailer) (cinfo); - (*cinfo->dest->term_destination) (cinfo); - /* We can use jpeg_abort to release memory and reset global_state */ - jpeg_abort((j_common_ptr) cinfo); -} - - -/* - * Write a special marker. - * This is only recommended for writing COM or APPn markers. - * Must be called after jpeg_start_compress() and before - * first call to jpeg_write_scanlines() or jpeg_write_raw_data(). - */ - -GLOBAL(void) -jpeg_write_marker (j_compress_ptr cinfo, int marker, - const JOCTET *dataptr, unsigned int datalen) -{ - JMETHOD(void, write_marker_byte, (j_compress_ptr info, int val)); - - if (cinfo->next_scanline != 0 || - (cinfo->global_state != CSTATE_SCANNING && - cinfo->global_state != CSTATE_RAW_OK && - cinfo->global_state != CSTATE_WRCOEFS)) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - - (*cinfo->marker->write_marker_header) (cinfo, marker, datalen); - write_marker_byte = cinfo->marker->write_marker_byte; /* copy for speed */ - while (datalen--) { - (*write_marker_byte) (cinfo, *dataptr); - dataptr++; - } -} - -/* Same, but piecemeal. */ - -GLOBAL(void) -jpeg_write_m_header (j_compress_ptr cinfo, int marker, unsigned int datalen) -{ - if (cinfo->next_scanline != 0 || - (cinfo->global_state != CSTATE_SCANNING && - cinfo->global_state != CSTATE_RAW_OK && - cinfo->global_state != CSTATE_WRCOEFS)) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - - (*cinfo->marker->write_marker_header) (cinfo, marker, datalen); -} - -GLOBAL(void) -jpeg_write_m_byte (j_compress_ptr cinfo, int val) -{ - (*cinfo->marker->write_marker_byte) (cinfo, val); -} - - -/* - * Alternate compression function: just write an abbreviated table file. - * Before calling this, all parameters and a data destination must be set up. - * - * To produce a pair of files containing abbreviated tables and abbreviated - * image data, one would proceed as follows: - * - * initialize JPEG object - * set JPEG parameters - * set destination to table file - * jpeg_write_tables(cinfo); - * set destination to image file - * jpeg_start_compress(cinfo, FALSE); - * write data... - * jpeg_finish_compress(cinfo); - * - * jpeg_write_tables has the side effect of marking all tables written - * (same as jpeg_suppress_tables(..., TRUE)). Thus a subsequent start_compress - * will not re-emit the tables unless it is passed write_all_tables=TRUE. - */ - -GLOBAL(void) -jpeg_write_tables (j_compress_ptr cinfo) -{ - if (cinfo->global_state != CSTATE_START) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - - /* (Re)initialize error mgr and destination modules */ - (*cinfo->err->reset_error_mgr) ((j_common_ptr) cinfo); - (*cinfo->dest->init_destination) (cinfo); - /* Initialize the marker writer ... bit of a crock to do it here. */ - jinit_marker_writer(cinfo); - /* Write them tables! */ - (*cinfo->marker->write_tables_only) (cinfo); - /* And clean up. */ - (*cinfo->dest->term_destination) (cinfo); - /* - * In library releases up through v6a, we called jpeg_abort() here to free - * any working memory allocated by the destination manager and marker - * writer. Some applications had a problem with that: they allocated space - * of their own from the library memory manager, and didn't want it to go - * away during write_tables. So now we do nothing. This will cause a - * memory leak if an app calls write_tables repeatedly without doing a full - * compression cycle or otherwise resetting the JPEG object. However, that - * seems less bad than unexpectedly freeing memory in the normal case. - * An app that prefers the old behavior can call jpeg_abort for itself after - * each call to jpeg_write_tables(). - */ -} diff --git a/dep/libjpeg/src/jcapistd.c b/dep/libjpeg/src/jcapistd.c deleted file mode 100644 index 0917afa97..000000000 --- a/dep/libjpeg/src/jcapistd.c +++ /dev/null @@ -1,162 +0,0 @@ -/* - * jcapistd.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2013 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains application interface code for the compression half - * of the JPEG library. These are the "standard" API routines that are - * used in the normal full-compression case. They are not used by a - * transcoding-only application. Note that if an application links in - * jpeg_start_compress, it will end up linking in the entire compressor. - * We thus must separate this file from jcapimin.c to avoid linking the - * whole compression library into a transcoder. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* - * Compression initialization. - * Before calling this, all parameters and a data destination must be set up. - * - * We require a write_all_tables parameter as a failsafe check when writing - * multiple datastreams from the same compression object. Since prior runs - * will have left all the tables marked sent_table=TRUE, a subsequent run - * would emit an abbreviated stream (no tables) by default. This may be what - * is wanted, but for safety's sake it should not be the default behavior: - * programmers should have to make a deliberate choice to emit abbreviated - * images. Therefore the documentation and examples should encourage people - * to pass write_all_tables=TRUE; then it will take active thought to do the - * wrong thing. - */ - -GLOBAL(void) -jpeg_start_compress (j_compress_ptr cinfo, boolean write_all_tables) -{ - if (cinfo->global_state != CSTATE_START) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - - if (write_all_tables) - jpeg_suppress_tables(cinfo, FALSE); /* mark all tables to be written */ - - /* (Re)initialize error mgr and destination modules */ - (*cinfo->err->reset_error_mgr) ((j_common_ptr) cinfo); - (*cinfo->dest->init_destination) (cinfo); - /* Perform master selection of active modules */ - jinit_compress_master(cinfo); - /* Set up for the first pass */ - (*cinfo->master->prepare_for_pass) (cinfo); - /* Ready for application to drive first pass through jpeg_write_scanlines - * or jpeg_write_raw_data. - */ - cinfo->next_scanline = 0; - cinfo->global_state = (cinfo->raw_data_in ? CSTATE_RAW_OK : CSTATE_SCANNING); -} - - -/* - * Write some scanlines of data to the JPEG compressor. - * - * The return value will be the number of lines actually written. - * This should be less than the supplied num_lines only in case that - * the data destination module has requested suspension of the compressor, - * or if more than image_height scanlines are passed in. - * - * Note: we warn about excess calls to jpeg_write_scanlines() since - * this likely signals an application programmer error. However, - * excess scanlines passed in the last valid call are *silently* ignored, - * so that the application need not adjust num_lines for end-of-image - * when using a multiple-scanline buffer. - */ - -GLOBAL(JDIMENSION) -jpeg_write_scanlines (j_compress_ptr cinfo, JSAMPARRAY scanlines, - JDIMENSION num_lines) -{ - JDIMENSION row_ctr, rows_left; - - if (cinfo->global_state != CSTATE_SCANNING) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - if (cinfo->next_scanline >= cinfo->image_height) - WARNMS(cinfo, JWRN_TOO_MUCH_DATA); - - /* Call progress monitor hook if present */ - if (cinfo->progress != NULL) { - cinfo->progress->pass_counter = (long) cinfo->next_scanline; - cinfo->progress->pass_limit = (long) cinfo->image_height; - (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo); - } - - /* Give master control module another chance if this is first call to - * jpeg_write_scanlines. This lets output of the frame/scan headers be - * delayed so that application can write COM, etc, markers between - * jpeg_start_compress and jpeg_write_scanlines. - */ - if (cinfo->master->call_pass_startup) - (*cinfo->master->pass_startup) (cinfo); - - /* Ignore any extra scanlines at bottom of image. */ - rows_left = cinfo->image_height - cinfo->next_scanline; - if (num_lines > rows_left) - num_lines = rows_left; - - row_ctr = 0; - (*cinfo->main->process_data) (cinfo, scanlines, &row_ctr, num_lines); - cinfo->next_scanline += row_ctr; - return row_ctr; -} - - -/* - * Alternate entry point to write raw data. - * Processes exactly one iMCU row per call, unless suspended. - */ - -GLOBAL(JDIMENSION) -jpeg_write_raw_data (j_compress_ptr cinfo, JSAMPIMAGE data, - JDIMENSION num_lines) -{ - JDIMENSION lines_per_iMCU_row; - - if (cinfo->global_state != CSTATE_RAW_OK) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - if (cinfo->next_scanline >= cinfo->image_height) { - WARNMS(cinfo, JWRN_TOO_MUCH_DATA); - return 0; - } - - /* Call progress monitor hook if present */ - if (cinfo->progress != NULL) { - cinfo->progress->pass_counter = (long) cinfo->next_scanline; - cinfo->progress->pass_limit = (long) cinfo->image_height; - (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo); - } - - /* Give master control module another chance if this is first call to - * jpeg_write_raw_data. This lets output of the frame/scan headers be - * delayed so that application can write COM, etc, markers between - * jpeg_start_compress and jpeg_write_raw_data. - */ - if (cinfo->master->call_pass_startup) - (*cinfo->master->pass_startup) (cinfo); - - /* Verify that at least one iMCU row has been passed. */ - lines_per_iMCU_row = cinfo->max_v_samp_factor * cinfo->min_DCT_v_scaled_size; - if (num_lines < lines_per_iMCU_row) - ERREXIT(cinfo, JERR_BUFFER_SIZE); - - /* Directly compress the row. */ - if (! (*cinfo->coef->compress_data) (cinfo, data)) { - /* If compressor did not consume the whole row, suspend processing. */ - return 0; - } - - /* OK, we processed one iMCU row. */ - cinfo->next_scanline += lines_per_iMCU_row; - return lines_per_iMCU_row; -} diff --git a/dep/libjpeg/src/jcarith.c b/dep/libjpeg/src/jcarith.c deleted file mode 100644 index 1b45089a3..000000000 --- a/dep/libjpeg/src/jcarith.c +++ /dev/null @@ -1,945 +0,0 @@ -/* - * jcarith.c - * - * Developed 1997-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains portable arithmetic entropy encoding routines for JPEG - * (implementing the ISO/IEC IS 10918-1 and CCITT Recommendation ITU-T T.81). - * - * Both sequential and progressive modes are supported in this single module. - * - * Suspension is not currently supported in this module. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Expanded entropy encoder object for arithmetic encoding. */ - -typedef struct { - struct jpeg_entropy_encoder pub; /* public fields */ - - INT32 c; /* C register, base of coding interval, layout as in sec. D.1.3 */ - INT32 a; /* A register, normalized size of coding interval */ - INT32 sc; /* counter for stacked 0xFF values which might overflow */ - INT32 zc; /* counter for pending 0x00 output values which might * - * be discarded at the end ("Pacman" termination) */ - int ct; /* bit shift counter, determines when next byte will be written */ - int buffer; /* buffer for most recent output byte != 0xFF */ - - int last_dc_val[MAX_COMPS_IN_SCAN]; /* last DC coef for each component */ - int dc_context[MAX_COMPS_IN_SCAN]; /* context index for DC conditioning */ - - unsigned int restarts_to_go; /* MCUs left in this restart interval */ - int next_restart_num; /* next restart number to write (0-7) */ - - /* Pointers to statistics areas (these workspaces have image lifespan) */ - unsigned char * dc_stats[NUM_ARITH_TBLS]; - unsigned char * ac_stats[NUM_ARITH_TBLS]; - - /* Statistics bin for coding with fixed probability 0.5 */ - unsigned char fixed_bin[4]; -} arith_entropy_encoder; - -typedef arith_entropy_encoder * arith_entropy_ptr; - -/* The following two definitions specify the allocation chunk size - * for the statistics area. - * According to sections F.1.4.4.1.3 and F.1.4.4.2, we need at least - * 49 statistics bins for DC, and 245 statistics bins for AC coding. - * - * We use a compact representation with 1 byte per statistics bin, - * thus the numbers directly represent byte sizes. - * This 1 byte per statistics bin contains the meaning of the MPS - * (more probable symbol) in the highest bit (mask 0x80), and the - * index into the probability estimation state machine table - * in the lower bits (mask 0x7F). - */ - -#define DC_STAT_BINS 64 -#define AC_STAT_BINS 256 - -/* NOTE: Uncomment the following #define if you want to use the - * given formula for calculating the AC conditioning parameter Kx - * for spectral selection progressive coding in section G.1.3.2 - * of the spec (Kx = Kmin + SRL (8 + Se - Kmin) 4). - * Although the spec and P&M authors claim that this "has proven - * to give good results for 8 bit precision samples", I'm not - * convinced yet that this is really beneficial. - * Early tests gave only very marginal compression enhancements - * (a few - around 5 or so - bytes even for very large files), - * which would turn out rather negative if we'd suppress the - * DAC (Define Arithmetic Conditioning) marker segments for - * the default parameters in the future. - * Note that currently the marker writing module emits 12-byte - * DAC segments for a full-component scan in a color image. - * This is not worth worrying about IMHO. However, since the - * spec defines the default values to be used if the tables - * are omitted (unlike Huffman tables, which are required - * anyway), one might optimize this behaviour in the future, - * and then it would be disadvantageous to use custom tables if - * they don't provide sufficient gain to exceed the DAC size. - * - * On the other hand, I'd consider it as a reasonable result - * that the conditioning has no significant influence on the - * compression performance. This means that the basic - * statistical model is already rather stable. - * - * Thus, at the moment, we use the default conditioning values - * anyway, and do not use the custom formula. - * -#define CALCULATE_SPECTRAL_CONDITIONING - */ - -/* IRIGHT_SHIFT is like RIGHT_SHIFT, but works on int rather than INT32. - * We assume that int right shift is unsigned if INT32 right shift is, - * which should be safe. - */ - -#ifdef RIGHT_SHIFT_IS_UNSIGNED -#define ISHIFT_TEMPS int ishift_temp; -#define IRIGHT_SHIFT(x,shft) \ - ((ishift_temp = (x)) < 0 ? \ - (ishift_temp >> (shft)) | ((~0) << (16-(shft))) : \ - (ishift_temp >> (shft))) -#else -#define ISHIFT_TEMPS -#define IRIGHT_SHIFT(x,shft) ((x) >> (shft)) -#endif - - -LOCAL(void) -emit_byte (int val, j_compress_ptr cinfo) -/* Write next output byte; we do not support suspension in this module. */ -{ - struct jpeg_destination_mgr * dest = cinfo->dest; - - *dest->next_output_byte++ = (JOCTET) val; - if (--dest->free_in_buffer == 0) - if (! (*dest->empty_output_buffer) (cinfo)) - ERREXIT(cinfo, JERR_CANT_SUSPEND); -} - - -/* - * Finish up at the end of an arithmetic-compressed scan. - */ - -METHODDEF(void) -finish_pass (j_compress_ptr cinfo) -{ - arith_entropy_ptr e = (arith_entropy_ptr) cinfo->entropy; - INT32 temp; - - /* Section D.1.8: Termination of encoding */ - - /* Find the e->c in the coding interval with the largest - * number of trailing zero bits */ - if ((temp = (e->a - 1 + e->c) & 0xFFFF0000L) < e->c) - e->c = temp + 0x8000L; - else - e->c = temp; - /* Send remaining bytes to output */ - e->c <<= e->ct; - if (e->c & 0xF8000000L) { - /* One final overflow has to be handled */ - if (e->buffer >= 0) { - if (e->zc) - do emit_byte(0x00, cinfo); - while (--e->zc); - emit_byte(e->buffer + 1, cinfo); - if (e->buffer + 1 == 0xFF) - emit_byte(0x00, cinfo); - } - e->zc += e->sc; /* carry-over converts stacked 0xFF bytes to 0x00 */ - e->sc = 0; - } else { - if (e->buffer == 0) - ++e->zc; - else if (e->buffer >= 0) { - if (e->zc) - do emit_byte(0x00, cinfo); - while (--e->zc); - emit_byte(e->buffer, cinfo); - } - if (e->sc) { - if (e->zc) - do emit_byte(0x00, cinfo); - while (--e->zc); - do { - emit_byte(0xFF, cinfo); - emit_byte(0x00, cinfo); - } while (--e->sc); - } - } - /* Output final bytes only if they are not 0x00 */ - if (e->c & 0x7FFF800L) { - if (e->zc) /* output final pending zero bytes */ - do emit_byte(0x00, cinfo); - while (--e->zc); - emit_byte((int) ((e->c >> 19) & 0xFF), cinfo); - if (((e->c >> 19) & 0xFF) == 0xFF) - emit_byte(0x00, cinfo); - if (e->c & 0x7F800L) { - emit_byte((int) ((e->c >> 11) & 0xFF), cinfo); - if (((e->c >> 11) & 0xFF) == 0xFF) - emit_byte(0x00, cinfo); - } - } -} - - -/* - * The core arithmetic encoding routine (common in JPEG and JBIG). - * This needs to go as fast as possible. - * Machine-dependent optimization facilities - * are not utilized in this portable implementation. - * However, this code should be fairly efficient and - * may be a good base for further optimizations anyway. - * - * Parameter 'val' to be encoded may be 0 or 1 (binary decision). - * - * Note: I've added full "Pacman" termination support to the - * byte output routines, which is equivalent to the optional - * Discard_final_zeros procedure (Figure D.15) in the spec. - * Thus, we always produce the shortest possible output - * stream compliant to the spec (no trailing zero bytes, - * except for FF stuffing). - * - * I've also introduced a new scheme for accessing - * the probability estimation state machine table, - * derived from Markus Kuhn's JBIG implementation. - */ - -LOCAL(void) -arith_encode (j_compress_ptr cinfo, unsigned char *st, int val) -{ - register arith_entropy_ptr e = (arith_entropy_ptr) cinfo->entropy; - register unsigned char nl, nm; - register INT32 qe, temp; - register int sv; - - /* Fetch values from our compact representation of Table D.3(D.2): - * Qe values and probability estimation state machine - */ - sv = *st; - qe = jpeg_aritab[sv & 0x7F]; /* => Qe_Value */ - nl = qe & 0xFF; qe >>= 8; /* Next_Index_LPS + Switch_MPS */ - nm = qe & 0xFF; qe >>= 8; /* Next_Index_MPS */ - - /* Encode & estimation procedures per sections D.1.4 & D.1.5 */ - e->a -= qe; - if (val != (sv >> 7)) { - /* Encode the less probable symbol */ - if (e->a >= qe) { - /* If the interval size (qe) for the less probable symbol (LPS) - * is larger than the interval size for the MPS, then exchange - * the two symbols for coding efficiency, otherwise code the LPS - * as usual: */ - e->c += e->a; - e->a = qe; - } - *st = (sv & 0x80) ^ nl; /* Estimate_after_LPS */ - } else { - /* Encode the more probable symbol */ - if (e->a >= 0x8000L) - return; /* A >= 0x8000 -> ready, no renormalization required */ - if (e->a < qe) { - /* If the interval size (qe) for the less probable symbol (LPS) - * is larger than the interval size for the MPS, then exchange - * the two symbols for coding efficiency: */ - e->c += e->a; - e->a = qe; - } - *st = (sv & 0x80) ^ nm; /* Estimate_after_MPS */ - } - - /* Renormalization & data output per section D.1.6 */ - do { - e->a <<= 1; - e->c <<= 1; - if (--e->ct == 0) { - /* Another byte is ready for output */ - temp = e->c >> 19; - if (temp > 0xFF) { - /* Handle overflow over all stacked 0xFF bytes */ - if (e->buffer >= 0) { - if (e->zc) - do emit_byte(0x00, cinfo); - while (--e->zc); - emit_byte(e->buffer + 1, cinfo); - if (e->buffer + 1 == 0xFF) - emit_byte(0x00, cinfo); - } - e->zc += e->sc; /* carry-over converts stacked 0xFF bytes to 0x00 */ - e->sc = 0; - /* Note: The 3 spacer bits in the C register guarantee - * that the new buffer byte can't be 0xFF here - * (see page 160 in the P&M JPEG book). */ - /* New output byte, might overflow later */ - e->buffer = (int) (temp & 0xFF); - } else if (temp == 0xFF) { - ++e->sc; /* stack 0xFF byte (which might overflow later) */ - } else { - /* Output all stacked 0xFF bytes, they will not overflow any more */ - if (e->buffer == 0) - ++e->zc; - else if (e->buffer >= 0) { - if (e->zc) - do emit_byte(0x00, cinfo); - while (--e->zc); - emit_byte(e->buffer, cinfo); - } - if (e->sc) { - if (e->zc) - do emit_byte(0x00, cinfo); - while (--e->zc); - do { - emit_byte(0xFF, cinfo); - emit_byte(0x00, cinfo); - } while (--e->sc); - } - /* New output byte (can still overflow) */ - e->buffer = (int) (temp & 0xFF); - } - e->c &= 0x7FFFFL; - e->ct += 8; - } - } while (e->a < 0x8000L); -} - - -/* - * Emit a restart marker & resynchronize predictions. - */ - -LOCAL(void) -emit_restart (j_compress_ptr cinfo, int restart_num) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - int ci; - jpeg_component_info * compptr; - - finish_pass(cinfo); - - emit_byte(0xFF, cinfo); - emit_byte(JPEG_RST0 + restart_num, cinfo); - - /* Re-initialize statistics areas */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - /* DC needs no table for refinement scan */ - if (cinfo->Ss == 0 && cinfo->Ah == 0) { - MEMZERO(entropy->dc_stats[compptr->dc_tbl_no], DC_STAT_BINS); - /* Reset DC predictions to 0 */ - entropy->last_dc_val[ci] = 0; - entropy->dc_context[ci] = 0; - } - /* AC needs no table when not present */ - if (cinfo->Se) { - MEMZERO(entropy->ac_stats[compptr->ac_tbl_no], AC_STAT_BINS); - } - } - - /* Reset arithmetic encoding variables */ - entropy->c = 0; - entropy->a = 0x10000L; - entropy->sc = 0; - entropy->zc = 0; - entropy->ct = 11; - entropy->buffer = -1; /* empty */ -} - - -/* - * MCU encoding for DC initial scan (either spectral selection, - * or first pass of successive approximation). - */ - -METHODDEF(boolean) -encode_mcu_DC_first (j_compress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - unsigned char *st; - int blkn, ci, tbl; - int v, v2, m; - ISHIFT_TEMPS - - /* Emit restart marker if needed */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) { - emit_restart(cinfo, entropy->next_restart_num); - entropy->restarts_to_go = cinfo->restart_interval; - entropy->next_restart_num++; - entropy->next_restart_num &= 7; - } - entropy->restarts_to_go--; - } - - /* Encode the MCU data blocks */ - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - ci = cinfo->MCU_membership[blkn]; - tbl = cinfo->cur_comp_info[ci]->dc_tbl_no; - - /* Compute the DC value after the required point transform by Al. - * This is simply an arithmetic right shift. - */ - m = IRIGHT_SHIFT((int) (MCU_data[blkn][0][0]), cinfo->Al); - - /* Sections F.1.4.1 & F.1.4.4.1: Encoding of DC coefficients */ - - /* Table F.4: Point to statistics bin S0 for DC coefficient coding */ - st = entropy->dc_stats[tbl] + entropy->dc_context[ci]; - - /* Figure F.4: Encode_DC_DIFF */ - if ((v = m - entropy->last_dc_val[ci]) == 0) { - arith_encode(cinfo, st, 0); - entropy->dc_context[ci] = 0; /* zero diff category */ - } else { - entropy->last_dc_val[ci] = m; - arith_encode(cinfo, st, 1); - /* Figure F.6: Encoding nonzero value v */ - /* Figure F.7: Encoding the sign of v */ - if (v > 0) { - arith_encode(cinfo, st + 1, 0); /* Table F.4: SS = S0 + 1 */ - st += 2; /* Table F.4: SP = S0 + 2 */ - entropy->dc_context[ci] = 4; /* small positive diff category */ - } else { - v = -v; - arith_encode(cinfo, st + 1, 1); /* Table F.4: SS = S0 + 1 */ - st += 3; /* Table F.4: SN = S0 + 3 */ - entropy->dc_context[ci] = 8; /* small negative diff category */ - } - /* Figure F.8: Encoding the magnitude category of v */ - m = 0; - if (v -= 1) { - arith_encode(cinfo, st, 1); - m = 1; - v2 = v; - st = entropy->dc_stats[tbl] + 20; /* Table F.4: X1 = 20 */ - while (v2 >>= 1) { - arith_encode(cinfo, st, 1); - m <<= 1; - st += 1; - } - } - arith_encode(cinfo, st, 0); - /* Section F.1.4.4.1.2: Establish dc_context conditioning category */ - if (m < (int) ((1L << cinfo->arith_dc_L[tbl]) >> 1)) - entropy->dc_context[ci] = 0; /* zero diff category */ - else if (m > (int) ((1L << cinfo->arith_dc_U[tbl]) >> 1)) - entropy->dc_context[ci] += 8; /* large diff category */ - /* Figure F.9: Encoding the magnitude bit pattern of v */ - st += 14; - while (m >>= 1) - arith_encode(cinfo, st, (m & v) ? 1 : 0); - } - } - - return TRUE; -} - - -/* - * MCU encoding for AC initial scan (either spectral selection, - * or first pass of successive approximation). - */ - -METHODDEF(boolean) -encode_mcu_AC_first (j_compress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - const int * natural_order; - JBLOCKROW block; - unsigned char *st; - int tbl, k, ke; - int v, v2, m; - - /* Emit restart marker if needed */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) { - emit_restart(cinfo, entropy->next_restart_num); - entropy->restarts_to_go = cinfo->restart_interval; - entropy->next_restart_num++; - entropy->next_restart_num &= 7; - } - entropy->restarts_to_go--; - } - - natural_order = cinfo->natural_order; - - /* Encode the MCU data block */ - block = MCU_data[0]; - tbl = cinfo->cur_comp_info[0]->ac_tbl_no; - - /* Sections F.1.4.2 & F.1.4.4.2: Encoding of AC coefficients */ - - /* Establish EOB (end-of-block) index */ - ke = cinfo->Se; - do { - /* We must apply the point transform by Al. For AC coefficients this - * is an integer division with rounding towards 0. To do this portably - * in C, we shift after obtaining the absolute value. - */ - if ((v = (*block)[natural_order[ke]]) >= 0) { - if (v >>= cinfo->Al) break; - } else { - v = -v; - if (v >>= cinfo->Al) break; - } - } while (--ke); - - /* Figure F.5: Encode_AC_Coefficients */ - for (k = cinfo->Ss - 1; k < ke;) { - st = entropy->ac_stats[tbl] + 3 * k; - arith_encode(cinfo, st, 0); /* EOB decision */ - for (;;) { - if ((v = (*block)[natural_order[++k]]) >= 0) { - if (v >>= cinfo->Al) { - arith_encode(cinfo, st + 1, 1); - arith_encode(cinfo, entropy->fixed_bin, 0); - break; - } - } else { - v = -v; - if (v >>= cinfo->Al) { - arith_encode(cinfo, st + 1, 1); - arith_encode(cinfo, entropy->fixed_bin, 1); - break; - } - } - arith_encode(cinfo, st + 1, 0); - st += 3; - } - st += 2; - /* Figure F.8: Encoding the magnitude category of v */ - m = 0; - if (v -= 1) { - arith_encode(cinfo, st, 1); - m = 1; - v2 = v; - if (v2 >>= 1) { - arith_encode(cinfo, st, 1); - m <<= 1; - st = entropy->ac_stats[tbl] + - (k <= cinfo->arith_ac_K[tbl] ? 189 : 217); - while (v2 >>= 1) { - arith_encode(cinfo, st, 1); - m <<= 1; - st += 1; - } - } - } - arith_encode(cinfo, st, 0); - /* Figure F.9: Encoding the magnitude bit pattern of v */ - st += 14; - while (m >>= 1) - arith_encode(cinfo, st, (m & v) ? 1 : 0); - } - /* Encode EOB decision only if k < cinfo->Se */ - if (k < cinfo->Se) { - st = entropy->ac_stats[tbl] + 3 * k; - arith_encode(cinfo, st, 1); - } - - return TRUE; -} - - -/* - * MCU encoding for DC successive approximation refinement scan. - * Note: we assume such scans can be multi-component, - * although the spec is not very clear on the point. - */ - -METHODDEF(boolean) -encode_mcu_DC_refine (j_compress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - unsigned char *st; - int Al, blkn; - - /* Emit restart marker if needed */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) { - emit_restart(cinfo, entropy->next_restart_num); - entropy->restarts_to_go = cinfo->restart_interval; - entropy->next_restart_num++; - entropy->next_restart_num &= 7; - } - entropy->restarts_to_go--; - } - - st = entropy->fixed_bin; /* use fixed probability estimation */ - Al = cinfo->Al; - - /* Encode the MCU data blocks */ - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - /* We simply emit the Al'th bit of the DC coefficient value. */ - arith_encode(cinfo, st, (MCU_data[blkn][0][0] >> Al) & 1); - } - - return TRUE; -} - - -/* - * MCU encoding for AC successive approximation refinement scan. - */ - -METHODDEF(boolean) -encode_mcu_AC_refine (j_compress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - const int * natural_order; - JBLOCKROW block; - unsigned char *st; - int tbl, k, ke, kex; - int v; - - /* Emit restart marker if needed */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) { - emit_restart(cinfo, entropy->next_restart_num); - entropy->restarts_to_go = cinfo->restart_interval; - entropy->next_restart_num++; - entropy->next_restart_num &= 7; - } - entropy->restarts_to_go--; - } - - natural_order = cinfo->natural_order; - - /* Encode the MCU data block */ - block = MCU_data[0]; - tbl = cinfo->cur_comp_info[0]->ac_tbl_no; - - /* Section G.1.3.3: Encoding of AC coefficients */ - - /* Establish EOB (end-of-block) index */ - ke = cinfo->Se; - do { - /* We must apply the point transform by Al. For AC coefficients this - * is an integer division with rounding towards 0. To do this portably - * in C, we shift after obtaining the absolute value. - */ - if ((v = (*block)[natural_order[ke]]) >= 0) { - if (v >>= cinfo->Al) break; - } else { - v = -v; - if (v >>= cinfo->Al) break; - } - } while (--ke); - - /* Establish EOBx (previous stage end-of-block) index */ - for (kex = ke; kex > 0; kex--) - if ((v = (*block)[natural_order[kex]]) >= 0) { - if (v >>= cinfo->Ah) break; - } else { - v = -v; - if (v >>= cinfo->Ah) break; - } - - /* Figure G.10: Encode_AC_Coefficients_SA */ - for (k = cinfo->Ss - 1; k < ke;) { - st = entropy->ac_stats[tbl] + 3 * k; - if (k >= kex) - arith_encode(cinfo, st, 0); /* EOB decision */ - for (;;) { - if ((v = (*block)[natural_order[++k]]) >= 0) { - if (v >>= cinfo->Al) { - if (v >> 1) /* previously nonzero coef */ - arith_encode(cinfo, st + 2, (v & 1)); - else { /* newly nonzero coef */ - arith_encode(cinfo, st + 1, 1); - arith_encode(cinfo, entropy->fixed_bin, 0); - } - break; - } - } else { - v = -v; - if (v >>= cinfo->Al) { - if (v >> 1) /* previously nonzero coef */ - arith_encode(cinfo, st + 2, (v & 1)); - else { /* newly nonzero coef */ - arith_encode(cinfo, st + 1, 1); - arith_encode(cinfo, entropy->fixed_bin, 1); - } - break; - } - } - arith_encode(cinfo, st + 1, 0); - st += 3; - } - } - /* Encode EOB decision only if k < cinfo->Se */ - if (k < cinfo->Se) { - st = entropy->ac_stats[tbl] + 3 * k; - arith_encode(cinfo, st, 1); - } - - return TRUE; -} - - -/* - * Encode and output one MCU's worth of arithmetic-compressed coefficients. - */ - -METHODDEF(boolean) -encode_mcu (j_compress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - const int * natural_order; - JBLOCKROW block; - unsigned char *st; - int tbl, k, ke; - int v, v2, m; - int blkn, ci; - jpeg_component_info * compptr; - - /* Emit restart marker if needed */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) { - emit_restart(cinfo, entropy->next_restart_num); - entropy->restarts_to_go = cinfo->restart_interval; - entropy->next_restart_num++; - entropy->next_restart_num &= 7; - } - entropy->restarts_to_go--; - } - - natural_order = cinfo->natural_order; - - /* Encode the MCU data blocks */ - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - block = MCU_data[blkn]; - ci = cinfo->MCU_membership[blkn]; - compptr = cinfo->cur_comp_info[ci]; - - /* Sections F.1.4.1 & F.1.4.4.1: Encoding of DC coefficients */ - - tbl = compptr->dc_tbl_no; - - /* Table F.4: Point to statistics bin S0 for DC coefficient coding */ - st = entropy->dc_stats[tbl] + entropy->dc_context[ci]; - - /* Figure F.4: Encode_DC_DIFF */ - if ((v = (*block)[0] - entropy->last_dc_val[ci]) == 0) { - arith_encode(cinfo, st, 0); - entropy->dc_context[ci] = 0; /* zero diff category */ - } else { - entropy->last_dc_val[ci] = (*block)[0]; - arith_encode(cinfo, st, 1); - /* Figure F.6: Encoding nonzero value v */ - /* Figure F.7: Encoding the sign of v */ - if (v > 0) { - arith_encode(cinfo, st + 1, 0); /* Table F.4: SS = S0 + 1 */ - st += 2; /* Table F.4: SP = S0 + 2 */ - entropy->dc_context[ci] = 4; /* small positive diff category */ - } else { - v = -v; - arith_encode(cinfo, st + 1, 1); /* Table F.4: SS = S0 + 1 */ - st += 3; /* Table F.4: SN = S0 + 3 */ - entropy->dc_context[ci] = 8; /* small negative diff category */ - } - /* Figure F.8: Encoding the magnitude category of v */ - m = 0; - if (v -= 1) { - arith_encode(cinfo, st, 1); - m = 1; - v2 = v; - st = entropy->dc_stats[tbl] + 20; /* Table F.4: X1 = 20 */ - while (v2 >>= 1) { - arith_encode(cinfo, st, 1); - m <<= 1; - st += 1; - } - } - arith_encode(cinfo, st, 0); - /* Section F.1.4.4.1.2: Establish dc_context conditioning category */ - if (m < (int) ((1L << cinfo->arith_dc_L[tbl]) >> 1)) - entropy->dc_context[ci] = 0; /* zero diff category */ - else if (m > (int) ((1L << cinfo->arith_dc_U[tbl]) >> 1)) - entropy->dc_context[ci] += 8; /* large diff category */ - /* Figure F.9: Encoding the magnitude bit pattern of v */ - st += 14; - while (m >>= 1) - arith_encode(cinfo, st, (m & v) ? 1 : 0); - } - - /* Sections F.1.4.2 & F.1.4.4.2: Encoding of AC coefficients */ - - if ((ke = cinfo->lim_Se) == 0) continue; - tbl = compptr->ac_tbl_no; - - /* Establish EOB (end-of-block) index */ - do { - if ((*block)[natural_order[ke]]) break; - } while (--ke); - - /* Figure F.5: Encode_AC_Coefficients */ - for (k = 0; k < ke;) { - st = entropy->ac_stats[tbl] + 3 * k; - arith_encode(cinfo, st, 0); /* EOB decision */ - while ((v = (*block)[natural_order[++k]]) == 0) { - arith_encode(cinfo, st + 1, 0); - st += 3; - } - arith_encode(cinfo, st + 1, 1); - /* Figure F.6: Encoding nonzero value v */ - /* Figure F.7: Encoding the sign of v */ - if (v > 0) { - arith_encode(cinfo, entropy->fixed_bin, 0); - } else { - v = -v; - arith_encode(cinfo, entropy->fixed_bin, 1); - } - st += 2; - /* Figure F.8: Encoding the magnitude category of v */ - m = 0; - if (v -= 1) { - arith_encode(cinfo, st, 1); - m = 1; - v2 = v; - if (v2 >>= 1) { - arith_encode(cinfo, st, 1); - m <<= 1; - st = entropy->ac_stats[tbl] + - (k <= cinfo->arith_ac_K[tbl] ? 189 : 217); - while (v2 >>= 1) { - arith_encode(cinfo, st, 1); - m <<= 1; - st += 1; - } - } - } - arith_encode(cinfo, st, 0); - /* Figure F.9: Encoding the magnitude bit pattern of v */ - st += 14; - while (m >>= 1) - arith_encode(cinfo, st, (m & v) ? 1 : 0); - } - /* Encode EOB decision only if k < cinfo->lim_Se */ - if (k < cinfo->lim_Se) { - st = entropy->ac_stats[tbl] + 3 * k; - arith_encode(cinfo, st, 1); - } - } - - return TRUE; -} - - -/* - * Initialize for an arithmetic-compressed scan. - */ - -METHODDEF(void) -start_pass (j_compress_ptr cinfo, boolean gather_statistics) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - int ci, tbl; - jpeg_component_info * compptr; - - if (gather_statistics) - /* Make sure to avoid that in the master control logic! - * We are fully adaptive here and need no extra - * statistics gathering pass! - */ - ERREXIT(cinfo, JERR_NOT_COMPILED); - - /* We assume jcmaster.c already validated the progressive scan parameters. */ - - /* Select execution routines */ - if (cinfo->progressive_mode) { - if (cinfo->Ah == 0) { - if (cinfo->Ss == 0) - entropy->pub.encode_mcu = encode_mcu_DC_first; - else - entropy->pub.encode_mcu = encode_mcu_AC_first; - } else { - if (cinfo->Ss == 0) - entropy->pub.encode_mcu = encode_mcu_DC_refine; - else - entropy->pub.encode_mcu = encode_mcu_AC_refine; - } - } else - entropy->pub.encode_mcu = encode_mcu; - - /* Allocate & initialize requested statistics areas */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - /* DC needs no table for refinement scan */ - if (cinfo->Ss == 0 && cinfo->Ah == 0) { - tbl = compptr->dc_tbl_no; - if (tbl < 0 || tbl >= NUM_ARITH_TBLS) - ERREXIT1(cinfo, JERR_NO_ARITH_TABLE, tbl); - if (entropy->dc_stats[tbl] == NULL) - entropy->dc_stats[tbl] = (unsigned char *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, DC_STAT_BINS); - MEMZERO(entropy->dc_stats[tbl], DC_STAT_BINS); - /* Initialize DC predictions to 0 */ - entropy->last_dc_val[ci] = 0; - entropy->dc_context[ci] = 0; - } - /* AC needs no table when not present */ - if (cinfo->Se) { - tbl = compptr->ac_tbl_no; - if (tbl < 0 || tbl >= NUM_ARITH_TBLS) - ERREXIT1(cinfo, JERR_NO_ARITH_TABLE, tbl); - if (entropy->ac_stats[tbl] == NULL) - entropy->ac_stats[tbl] = (unsigned char *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, AC_STAT_BINS); - MEMZERO(entropy->ac_stats[tbl], AC_STAT_BINS); -#ifdef CALCULATE_SPECTRAL_CONDITIONING - if (cinfo->progressive_mode) - /* Section G.1.3.2: Set appropriate arithmetic conditioning value Kx */ - cinfo->arith_ac_K[tbl] = cinfo->Ss + ((8 + cinfo->Se - cinfo->Ss) >> 4); -#endif - } - } - - /* Initialize arithmetic encoding variables */ - entropy->c = 0; - entropy->a = 0x10000L; - entropy->sc = 0; - entropy->zc = 0; - entropy->ct = 11; - entropy->buffer = -1; /* empty */ - - /* Initialize restart stuff */ - entropy->restarts_to_go = cinfo->restart_interval; - entropy->next_restart_num = 0; -} - - -/* - * Module initialization routine for arithmetic entropy encoding. - */ - -GLOBAL(void) -jinit_arith_encoder (j_compress_ptr cinfo) -{ - arith_entropy_ptr entropy; - int i; - - entropy = (arith_entropy_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(arith_entropy_encoder)); - cinfo->entropy = &entropy->pub; - entropy->pub.start_pass = start_pass; - entropy->pub.finish_pass = finish_pass; - - /* Mark tables unallocated */ - for (i = 0; i < NUM_ARITH_TBLS; i++) { - entropy->dc_stats[i] = NULL; - entropy->ac_stats[i] = NULL; - } - - /* Initialize index for fixed probability estimation */ - entropy->fixed_bin[0] = 113; -} diff --git a/dep/libjpeg/src/jccoefct.c b/dep/libjpeg/src/jccoefct.c deleted file mode 100644 index 494aa2298..000000000 --- a/dep/libjpeg/src/jccoefct.c +++ /dev/null @@ -1,456 +0,0 @@ -/* - * jccoefct.c - * - * Copyright (C) 1994-1997, Thomas G. Lane. - * Modified 2003-2022 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains the coefficient buffer controller for compression. - * This controller is the top level of the JPEG compressor proper. - * The coefficient buffer lies between forward-DCT and entropy encoding steps. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* We use a full-image coefficient buffer when doing Huffman optimization, - * and also for writing multiple-scan JPEG files. In all cases, the DCT - * step is run during the first pass, and subsequent passes need only read - * the buffered coefficients. - */ -#ifdef ENTROPY_OPT_SUPPORTED -#define FULL_COEF_BUFFER_SUPPORTED -#else -#ifdef C_MULTISCAN_FILES_SUPPORTED -#define FULL_COEF_BUFFER_SUPPORTED -#endif -#endif - - -/* Private buffer controller object */ - -typedef struct { - struct jpeg_c_coef_controller pub; /* public fields */ - - JDIMENSION iMCU_row_num; /* iMCU row # within image */ - JDIMENSION MCU_ctr; /* counts MCUs processed in current row */ - int MCU_vert_offset; /* counts MCU rows within iMCU row */ - int MCU_rows_per_iMCU_row; /* number of such rows needed */ - - /* For single-pass compression, it's sufficient to buffer just one MCU - * (although this may prove a bit slow in practice). - * We append a workspace of C_MAX_BLOCKS_IN_MCU coefficient blocks, - * and reuse it for each MCU constructed and sent. - * In multi-pass modes, this array points to the current MCU's blocks - * within the virtual arrays. - */ - JBLOCKROW MCU_buffer[C_MAX_BLOCKS_IN_MCU]; - - /* In multi-pass modes, we need a virtual block array for each component. */ - jvirt_barray_ptr whole_image[MAX_COMPONENTS]; - - /* Workspace for single-pass compression (omitted otherwise). */ - JBLOCK blk_buffer[C_MAX_BLOCKS_IN_MCU]; -} my_coef_controller; - -typedef my_coef_controller * my_coef_ptr; - - -/* Forward declarations */ -METHODDEF(boolean) compress_data - JPP((j_compress_ptr cinfo, JSAMPIMAGE input_buf)); -#ifdef FULL_COEF_BUFFER_SUPPORTED -METHODDEF(boolean) compress_first_pass - JPP((j_compress_ptr cinfo, JSAMPIMAGE input_buf)); -METHODDEF(boolean) compress_output - JPP((j_compress_ptr cinfo, JSAMPIMAGE input_buf)); -#endif - - -LOCAL(void) -start_iMCU_row (j_compress_ptr cinfo) -/* Reset within-iMCU-row counters for a new row */ -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - - /* In an interleaved scan, an MCU row is the same as an iMCU row. - * In a noninterleaved scan, an iMCU row has v_samp_factor MCU rows. - * But at the bottom of the image, process only what's left. - */ - if (cinfo->comps_in_scan > 1) { - coef->MCU_rows_per_iMCU_row = 1; - } else { - if (coef->iMCU_row_num < (cinfo->total_iMCU_rows-1)) - coef->MCU_rows_per_iMCU_row = cinfo->cur_comp_info[0]->v_samp_factor; - else - coef->MCU_rows_per_iMCU_row = cinfo->cur_comp_info[0]->last_row_height; - } - - coef->MCU_ctr = 0; - coef->MCU_vert_offset = 0; -} - - -/* - * Initialize for a processing pass. - */ - -METHODDEF(void) -start_pass_coef (j_compress_ptr cinfo, J_BUF_MODE pass_mode) -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - - coef->iMCU_row_num = 0; - start_iMCU_row(cinfo); - - switch (pass_mode) { - case JBUF_PASS_THRU: - if (coef->whole_image[0] != NULL) - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - coef->pub.compress_data = compress_data; - break; -#ifdef FULL_COEF_BUFFER_SUPPORTED - case JBUF_SAVE_AND_PASS: - if (coef->whole_image[0] == NULL) - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - coef->pub.compress_data = compress_first_pass; - break; - case JBUF_CRANK_DEST: - if (coef->whole_image[0] == NULL) - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - coef->pub.compress_data = compress_output; - break; -#endif - default: - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - } -} - - -/* - * Process some data in the single-pass case. - * We process the equivalent of one fully interleaved MCU row ("iMCU" row) - * per call, ie, v_samp_factor block rows for each component in the image. - * Returns TRUE if the iMCU row is completed, FALSE if suspended. - * - * NB: input_buf contains a plane for each component in image, - * which we index according to the component's SOF position. - */ - -METHODDEF(boolean) -compress_data (j_compress_ptr cinfo, JSAMPIMAGE input_buf) -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - JDIMENSION MCU_col_num; /* index of current MCU within row */ - JDIMENSION last_MCU_col = cinfo->MCUs_per_row - 1; - JDIMENSION last_iMCU_row = cinfo->total_iMCU_rows - 1; - int ci, xindex, yindex, yoffset, blockcnt; - JBLOCKROW blkp; - JSAMPARRAY input_ptr; - JDIMENSION xpos; - jpeg_component_info *compptr; - forward_DCT_ptr forward_DCT; - - /* Loop to write as much as one whole iMCU row */ - for (yoffset = coef->MCU_vert_offset; yoffset < coef->MCU_rows_per_iMCU_row; - yoffset++) { - for (MCU_col_num = coef->MCU_ctr; MCU_col_num <= last_MCU_col; - MCU_col_num++) { - /* Determine where data comes from in input_buf and do the DCT thing. - * Each call on forward_DCT processes a horizontal row of DCT blocks as - * wide as an MCU. Dummy blocks at the right or bottom edge are filled in - * specially. The data in them does not matter for image reconstruction, - * so we fill them with values that will encode to the smallest amount of - * data, viz: all zeroes in the AC entries, DC entries equal to previous - * block's DC value. (Thanks to Thomas Kinsman for this idea.) - */ - blkp = coef->blk_buffer; /* pointer to current DCT block within MCU */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - forward_DCT = cinfo->fdct->forward_DCT[compptr->component_index]; - input_ptr = input_buf[compptr->component_index] + - yoffset * compptr->DCT_v_scaled_size; - /* ypos == (yoffset + yindex) * compptr->DCT_v_scaled_size */ - blockcnt = (MCU_col_num < last_MCU_col) ? compptr->MCU_width - : compptr->last_col_width; - xpos = MCU_col_num * compptr->MCU_sample_width; - for (yindex = 0; yindex < compptr->MCU_height; yindex++) { - if (coef->iMCU_row_num < last_iMCU_row || - yoffset + yindex < compptr->last_row_height) { - (*forward_DCT) (cinfo, compptr, input_ptr, blkp, - xpos, (JDIMENSION) blockcnt); - input_ptr += compptr->DCT_v_scaled_size; - blkp += blockcnt; - /* Dummy blocks at right edge */ - if ((xindex = compptr->MCU_width - blockcnt) == 0) - continue; - } else { - /* At bottom of image, need a whole row of dummy blocks */ - xindex = compptr->MCU_width; - } - /* Fill in any dummy blocks needed in this row */ - MEMZERO(blkp, xindex * SIZEOF(JBLOCK)); - do { - blkp[0][0] = blkp[-1][0]; - blkp++; - } while (--xindex); - } - } - /* Try to write the MCU. In event of a suspension failure, we will - * re-DCT the MCU on restart (a bit inefficient, could be fixed...) - */ - if (! (*cinfo->entropy->encode_mcu) (cinfo, coef->MCU_buffer)) { - /* Suspension forced; update state counters and exit */ - coef->MCU_vert_offset = yoffset; - coef->MCU_ctr = MCU_col_num; - return FALSE; - } - } - /* Completed an MCU row, but perhaps not an iMCU row */ - coef->MCU_ctr = 0; - } - /* Completed the iMCU row, advance counters for next one */ - coef->iMCU_row_num++; - start_iMCU_row(cinfo); - return TRUE; -} - - -#ifdef FULL_COEF_BUFFER_SUPPORTED - -/* - * Process some data in the first pass of a multi-pass case. - * We process the equivalent of one fully interleaved MCU row ("iMCU" row) - * per call, ie, v_samp_factor block rows for each component in the image. - * This amount of data is read from the source buffer, DCT'd and quantized, - * and saved into the virtual arrays. We also generate suitable dummy blocks - * as needed at the right and lower edges. (The dummy blocks are constructed - * in the virtual arrays, which have been padded appropriately.) This makes - * it possible for subsequent passes not to worry about real vs. dummy blocks. - * - * We must also emit the data to the entropy encoder. This is conveniently - * done by calling compress_output() after we've loaded the current strip - * of the virtual arrays. - * - * NB: input_buf contains a plane for each component in image. All - * components are DCT'd and loaded into the virtual arrays in this pass. - * However, it may be that only a subset of the components are emitted to - * the entropy encoder during this first pass; be careful about looking - * at the scan-dependent variables (MCU dimensions, etc). - */ - -METHODDEF(boolean) -compress_first_pass (j_compress_ptr cinfo, JSAMPIMAGE input_buf) -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - JDIMENSION last_iMCU_row = cinfo->total_iMCU_rows - 1; - JDIMENSION blocks_across, MCUs_across, MCUindex; - int bi, ci, h_samp_factor, block_row, block_rows, ndummy; - JCOEF lastDC; - jpeg_component_info *compptr; - JBLOCKARRAY buffer; - JBLOCKROW thisblockrow, lastblockrow; - JSAMPARRAY input_ptr; - forward_DCT_ptr forward_DCT; - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* Align the virtual buffer for this component. */ - buffer = (*cinfo->mem->access_virt_barray) - ((j_common_ptr) cinfo, coef->whole_image[ci], - coef->iMCU_row_num * compptr->v_samp_factor, - (JDIMENSION) compptr->v_samp_factor, TRUE); - /* Count non-dummy DCT block rows in this iMCU row. */ - if (coef->iMCU_row_num < last_iMCU_row) - block_rows = compptr->v_samp_factor; - else { - /* NB: can't use last_row_height here, since may not be set! */ - block_rows = (int) (compptr->height_in_blocks % compptr->v_samp_factor); - if (block_rows == 0) block_rows = compptr->v_samp_factor; - } - blocks_across = compptr->width_in_blocks; - h_samp_factor = compptr->h_samp_factor; - /* Count number of dummy blocks to be added at the right margin. */ - ndummy = (int) (blocks_across % h_samp_factor); - if (ndummy > 0) - ndummy = h_samp_factor - ndummy; - forward_DCT = cinfo->fdct->forward_DCT[ci]; - input_ptr = input_buf[ci]; - /* Perform DCT for all non-dummy blocks in this iMCU row. Each call - * on forward_DCT processes a complete horizontal row of DCT blocks. - */ - for (block_row = 0; block_row < block_rows; block_row++) { - thisblockrow = buffer[block_row]; - (*forward_DCT) (cinfo, compptr, input_ptr, thisblockrow, - (JDIMENSION) 0, blocks_across); - input_ptr += compptr->DCT_v_scaled_size; - if (ndummy > 0) { - /* Create dummy blocks at the right edge of the image. */ - thisblockrow += blocks_across; /* => first dummy block */ - FMEMZERO((void FAR *) thisblockrow, ndummy * SIZEOF(JBLOCK)); - lastDC = thisblockrow[-1][0]; - for (bi = 0; bi < ndummy; bi++) { - thisblockrow[bi][0] = lastDC; - } - } - } - /* If at end of image, create dummy block rows as needed. - * The tricky part here is that within each MCU, we want the DC values - * of the dummy blocks to match the last real block's DC value. - * This squeezes a few more bytes out of the resulting file... - */ - if (block_row < compptr->v_samp_factor) { - blocks_across += ndummy; /* include lower right corner */ - MCUs_across = blocks_across / h_samp_factor; - do { - thisblockrow = buffer[block_row]; - lastblockrow = buffer[block_row-1]; - FMEMZERO((void FAR *) thisblockrow, - (size_t) blocks_across * SIZEOF(JBLOCK)); - for (MCUindex = 0; MCUindex < MCUs_across; MCUindex++) { - lastDC = lastblockrow[h_samp_factor-1][0]; - for (bi = 0; bi < h_samp_factor; bi++) { - thisblockrow[bi][0] = lastDC; - } - thisblockrow += h_samp_factor; /* advance to next MCU in row */ - lastblockrow += h_samp_factor; - } - } while (++block_row < compptr->v_samp_factor); - } - } - /* NB: compress_output will increment iMCU_row_num if successful. - * A suspension return will result in redoing all the work above next time. - */ - - /* Emit data to the entropy encoder, sharing code with subsequent passes */ - return compress_output(cinfo, input_buf); -} - - -/* - * Process some data in subsequent passes of a multi-pass case. - * We process the equivalent of one fully interleaved MCU row ("iMCU" row) - * per call, ie, v_samp_factor block rows for each component in the scan. - * The data is obtained from the virtual arrays and fed to the entropy coder. - * Returns TRUE if the iMCU row is completed, FALSE if suspended. - * - * NB: input_buf is ignored; it is likely to be a NULL pointer. - */ - -METHODDEF(boolean) -compress_output (j_compress_ptr cinfo, JSAMPIMAGE input_buf) -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - JDIMENSION MCU_col_num; /* index of current MCU within row */ - int ci, xindex, yindex, yoffset; - JDIMENSION start_col; - JBLOCKARRAY blkp; - JBLOCKARRAY buffer[MAX_COMPS_IN_SCAN]; - JBLOCKROW buffer_ptr; - jpeg_component_info *compptr; - - /* Align the virtual buffers for the components used in this scan. - * NB: during first pass, this is safe only because the buffers will - * already be aligned properly, so jmemmgr.c won't need to do any I/O. - */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - buffer[ci] = (*cinfo->mem->access_virt_barray) - ((j_common_ptr) cinfo, coef->whole_image[compptr->component_index], - coef->iMCU_row_num * compptr->v_samp_factor, - (JDIMENSION) compptr->v_samp_factor, FALSE); - } - - /* Loop to process one whole iMCU row */ - for (yoffset = coef->MCU_vert_offset; yoffset < coef->MCU_rows_per_iMCU_row; - yoffset++) { - for (MCU_col_num = coef->MCU_ctr; MCU_col_num < cinfo->MCUs_per_row; - MCU_col_num++) { - /* Construct list of pointers to DCT blocks belonging to this MCU */ - blkp = coef->MCU_buffer; /* pointer to current DCT block within MCU */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - start_col = MCU_col_num * compptr->MCU_width; - for (yindex = 0; yindex < compptr->MCU_height; yindex++) { - buffer_ptr = buffer[ci][yoffset + yindex] + start_col; - xindex = compptr->MCU_width; - do { - *blkp++ = buffer_ptr++; - } while (--xindex); - } - } - /* Try to write the MCU. */ - if (! (*cinfo->entropy->encode_mcu) (cinfo, coef->MCU_buffer)) { - /* Suspension forced; update state counters and exit */ - coef->MCU_vert_offset = yoffset; - coef->MCU_ctr = MCU_col_num; - return FALSE; - } - } - /* Completed an MCU row, but perhaps not an iMCU row */ - coef->MCU_ctr = 0; - } - /* Completed the iMCU row, advance counters for next one */ - coef->iMCU_row_num++; - start_iMCU_row(cinfo); - return TRUE; -} - -#endif /* FULL_COEF_BUFFER_SUPPORTED */ - - -/* - * Initialize coefficient buffer controller. - */ - -GLOBAL(void) -jinit_c_coef_controller (j_compress_ptr cinfo, boolean need_full_buffer) -{ - my_coef_ptr coef; - - if (need_full_buffer) { -#ifdef FULL_COEF_BUFFER_SUPPORTED - /* Allocate a full-image virtual array for each component, */ - /* padded to a multiple of samp_factor DCT blocks in each direction. */ - int ci; - jpeg_component_info *compptr; - - coef = (my_coef_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - SIZEOF(my_coef_controller) - SIZEOF(coef->blk_buffer)); - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - coef->whole_image[ci] = (*cinfo->mem->request_virt_barray) - ((j_common_ptr) cinfo, JPOOL_IMAGE, FALSE, - (JDIMENSION) jround_up((long) compptr->width_in_blocks, - (long) compptr->h_samp_factor), - (JDIMENSION) jround_up((long) compptr->height_in_blocks, - (long) compptr->v_samp_factor), - (JDIMENSION) compptr->v_samp_factor); - } -#else - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); -#endif - } else { - /* We only need a single-MCU buffer. */ - JBLOCKARRAY blkp; - JBLOCKROW buffer_ptr; - int bi; - - coef = (my_coef_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_coef_controller)); - blkp = coef->MCU_buffer; - buffer_ptr = coef->blk_buffer; - bi = C_MAX_BLOCKS_IN_MCU; - do { - *blkp++ = buffer_ptr++; - } while (--bi); - coef->whole_image[0] = NULL; /* flag for no virtual arrays */ - } - - coef->pub.start_pass = start_pass_coef; - cinfo->coef = &coef->pub; -} diff --git a/dep/libjpeg/src/jccolor.c b/dep/libjpeg/src/jccolor.c deleted file mode 100644 index c028dd9db..000000000 --- a/dep/libjpeg/src/jccolor.c +++ /dev/null @@ -1,598 +0,0 @@ -/* - * jccolor.c - * - * Copyright (C) 1991-1996, Thomas G. Lane. - * Modified 2011-2023 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains input colorspace conversion routines. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Private subobject */ - -typedef struct { - struct jpeg_color_converter pub; /* public fields */ - - /* Private state for RGB->YCC conversion */ - INT32 * rgb_ycc_tab; /* => table for RGB to YCbCr conversion */ -} my_color_converter; - -typedef my_color_converter * my_cconvert_ptr; - - -/**************** RGB -> YCbCr conversion: most common case **************/ - -/* - * YCbCr is defined per Recommendation ITU-R BT.601-7 (03/2011), - * previously known as Recommendation CCIR 601-1, except that Cb and Cr - * are normalized to the range 0..MAXJSAMPLE rather than -0.5 .. 0.5. - * sRGB (standard RGB color space) is defined per IEC 61966-2-1:1999. - * sYCC (standard luma-chroma-chroma color space with extended gamut) - * is defined per IEC 61966-2-1:1999 Amendment A1:2003 Annex F. - * bg-sRGB and bg-sYCC (big gamut standard color spaces) - * are defined per IEC 61966-2-1:1999 Amendment A1:2003 Annex G. - * Note that the derived conversion coefficients given in some of these - * documents are imprecise. The general conversion equations are - * Y = Kr * R + (1 - Kr - Kb) * G + Kb * B - * Cb = (B - Y) / (1 - Kb) / K - * Cr = (R - Y) / (1 - Kr) / K - * With Kr = 0.299 and Kb = 0.114 (derived according to SMPTE RP 177-1993 - * from the 1953 FCC NTSC primaries and CIE Illuminant C), K = 2 for sYCC, - * the conversion equations to be implemented are therefore - * Y = 0.299 * R + 0.587 * G + 0.114 * B - * Cb = -0.168735892 * R - 0.331264108 * G + 0.5 * B + CENTERJSAMPLE - * Cr = 0.5 * R - 0.418687589 * G - 0.081312411 * B + CENTERJSAMPLE - * Note: older versions of the IJG code used a zero offset of MAXJSAMPLE/2, - * rather than CENTERJSAMPLE, for Cb and Cr. This gave equal positive and - * negative swings for Cb/Cr, but meant that grayscale values (Cb=Cr=0) - * were not represented exactly. Now we sacrifice exact representation of - * maximum red and maximum blue in order to get exact grayscales. - * - * To avoid floating-point arithmetic, we represent the fractional constants - * as integers scaled up by 2^16 (about 4 digits precision); we have to divide - * the products by 2^16, with appropriate rounding, to get the correct answer. - * - * For even more speed, we avoid doing any multiplications in the inner loop - * by precalculating the constants times R,G,B for all possible values. - * For 8-bit JSAMPLEs this is very reasonable (only 256 entries per table); - * for 9-bit to 12-bit samples it is still acceptable. It's not very - * reasonable for 16-bit samples, but if you want lossless storage - * you shouldn't be changing colorspace anyway. - * The CENTERJSAMPLE offsets and the rounding fudge-factor of 0.5 are included - * in the tables to save adding them separately in the inner loop. - */ - -#define SCALEBITS 16 /* speediest right-shift on some machines */ -#define CBCR_OFFSET ((INT32) CENTERJSAMPLE << SCALEBITS) -#define ONE_HALF ((INT32) 1 << (SCALEBITS-1)) -#define FIX(x) ((INT32) ((x) * (1L< Y section */ -#define G_Y_OFF (1*(MAXJSAMPLE+1)) /* offset to G => Y section */ -#define B_Y_OFF (2*(MAXJSAMPLE+1)) /* etc. */ -#define R_CB_OFF (3*(MAXJSAMPLE+1)) -#define G_CB_OFF (4*(MAXJSAMPLE+1)) -#define B_CB_OFF (5*(MAXJSAMPLE+1)) -#define R_CR_OFF B_CB_OFF /* B=>Cb, R=>Cr are the same */ -#define G_CR_OFF (6*(MAXJSAMPLE+1)) -#define B_CR_OFF (7*(MAXJSAMPLE+1)) -#define TABLE_SIZE (8*(MAXJSAMPLE+1)) - - -/* - * Initialize for RGB->YCC colorspace conversion. - */ - -METHODDEF(void) -rgb_ycc_start (j_compress_ptr cinfo) -{ - my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert; - INT32 * rgb_ycc_tab; - INT32 i; - - /* Allocate and fill in the conversion tables. */ - cconvert->rgb_ycc_tab = rgb_ycc_tab = (INT32 *) - (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE, - TABLE_SIZE * SIZEOF(INT32)); - - for (i = 0; i <= MAXJSAMPLE; i++) { - rgb_ycc_tab[i+R_Y_OFF] = FIX(0.299) * i; - rgb_ycc_tab[i+G_Y_OFF] = FIX(0.587) * i; - rgb_ycc_tab[i+B_Y_OFF] = FIX(0.114) * i + ONE_HALF; - rgb_ycc_tab[i+R_CB_OFF] = (- FIX(0.168735892)) * i; - rgb_ycc_tab[i+G_CB_OFF] = (- FIX(0.331264108)) * i; - /* We use a rounding fudge-factor of 0.5-epsilon for Cb and Cr. - * This ensures that the maximum output will round to MAXJSAMPLE - * not MAXJSAMPLE+1, and thus that we don't have to range-limit. - */ - rgb_ycc_tab[i+B_CB_OFF] = (i << (SCALEBITS-1)) + CBCR_OFFSET + ONE_HALF-1; -/* B=>Cb and R=>Cr tables are the same - rgb_ycc_tab[i+R_CR_OFF] = (i << (SCALEBITS-1)) + CBCR_OFFSET + ONE_HALF-1; -*/ - rgb_ycc_tab[i+G_CR_OFF] = (- FIX(0.418687589)) * i; - rgb_ycc_tab[i+B_CR_OFF] = (- FIX(0.081312411)) * i; - } -} - - -/* - * Convert some rows of samples to the JPEG colorspace. - * - * Note that we change from the application's interleaved-pixel format - * to our internal noninterleaved, one-plane-per-component format. The - * input buffer is therefore three times as wide as the output buffer. - * - * A starting row offset is provided only for the output buffer. The - * caller can easily adjust the passed input_buf value to accommodate - * any row offset required on that side. - */ - -METHODDEF(void) -rgb_ycc_convert (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JSAMPIMAGE output_buf, - JDIMENSION output_row, int num_rows) -{ - my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert; - register int r, g, b; - register INT32 * ctab = cconvert->rgb_ycc_tab; - register JSAMPROW inptr; - register JSAMPROW outptr0, outptr1, outptr2; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->image_width; - - while (--num_rows >= 0) { - inptr = *input_buf++; - outptr0 = output_buf[0][output_row]; - outptr1 = output_buf[1][output_row]; - outptr2 = output_buf[2][output_row]; - output_row++; - for (col = 0; col < num_cols; col++) { - r = GETJSAMPLE(inptr[RGB_RED]); - g = GETJSAMPLE(inptr[RGB_GREEN]); - b = GETJSAMPLE(inptr[RGB_BLUE]); - inptr += RGB_PIXELSIZE; - /* If the inputs are 0..MAXJSAMPLE, the outputs of these equations - * must be too; we do not need an explicit range-limiting operation. - * Hence the value being shifted is never negative, and we don't - * need the general RIGHT_SHIFT macro. - */ - /* Y */ - outptr0[col] = (JSAMPLE) - ((ctab[r+R_Y_OFF] + ctab[g+G_Y_OFF] + ctab[b+B_Y_OFF]) - >> SCALEBITS); - /* Cb */ - outptr1[col] = (JSAMPLE) - ((ctab[r+R_CB_OFF] + ctab[g+G_CB_OFF] + ctab[b+B_CB_OFF]) - >> SCALEBITS); - /* Cr */ - outptr2[col] = (JSAMPLE) - ((ctab[r+R_CR_OFF] + ctab[g+G_CR_OFF] + ctab[b+B_CR_OFF]) - >> SCALEBITS); - } - } -} - - -/**************** Cases other than RGB -> YCbCr **************/ - - -/* - * Convert some rows of samples to the JPEG colorspace. - * This version handles RGB->grayscale conversion, - * which is the same as the RGB->Y portion of RGB->YCbCr. - * We assume rgb_ycc_start has been called (we only use the Y tables). - */ - -METHODDEF(void) -rgb_gray_convert (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JSAMPIMAGE output_buf, - JDIMENSION output_row, int num_rows) -{ - my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert; - register INT32 y; - register INT32 * ctab = cconvert->rgb_ycc_tab; - register JSAMPROW inptr; - register JSAMPROW outptr; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->image_width; - - while (--num_rows >= 0) { - inptr = *input_buf++; - outptr = output_buf[0][output_row++]; - for (col = 0; col < num_cols; col++) { - y = ctab[R_Y_OFF + GETJSAMPLE(inptr[RGB_RED])]; - y += ctab[G_Y_OFF + GETJSAMPLE(inptr[RGB_GREEN])]; - y += ctab[B_Y_OFF + GETJSAMPLE(inptr[RGB_BLUE])]; - inptr += RGB_PIXELSIZE; - outptr[col] = (JSAMPLE) (y >> SCALEBITS); - } - } -} - - -/* - * Convert some rows of samples to the JPEG colorspace. - * This version handles Adobe-style CMYK->YCCK conversion, - * where we convert R=1-C, G=1-M, and B=1-Y to YCbCr using the - * same conversion as above, while passing K (black) unchanged. - * We assume rgb_ycc_start has been called. - */ - -METHODDEF(void) -cmyk_ycck_convert (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JSAMPIMAGE output_buf, - JDIMENSION output_row, int num_rows) -{ - my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert; - register int r, g, b; - register INT32 * ctab = cconvert->rgb_ycc_tab; - register JSAMPROW inptr; - register JSAMPROW outptr0, outptr1, outptr2, outptr3; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->image_width; - - while (--num_rows >= 0) { - inptr = *input_buf++; - outptr0 = output_buf[0][output_row]; - outptr1 = output_buf[1][output_row]; - outptr2 = output_buf[2][output_row]; - outptr3 = output_buf[3][output_row]; - output_row++; - for (col = 0; col < num_cols; col++) { - r = MAXJSAMPLE - GETJSAMPLE(inptr[0]); - g = MAXJSAMPLE - GETJSAMPLE(inptr[1]); - b = MAXJSAMPLE - GETJSAMPLE(inptr[2]); - /* K passes through as-is */ - outptr3[col] = inptr[3]; /* don't need GETJSAMPLE here */ - inptr += 4; - /* If the inputs are 0..MAXJSAMPLE, the outputs of these equations - * must be too; we do not need an explicit range-limiting operation. - * Hence the value being shifted is never negative, and we don't - * need the general RIGHT_SHIFT macro. - */ - /* Y */ - outptr0[col] = (JSAMPLE) - ((ctab[r+R_Y_OFF] + ctab[g+G_Y_OFF] + ctab[b+B_Y_OFF]) - >> SCALEBITS); - /* Cb */ - outptr1[col] = (JSAMPLE) - ((ctab[r+R_CB_OFF] + ctab[g+G_CB_OFF] + ctab[b+B_CB_OFF]) - >> SCALEBITS); - /* Cr */ - outptr2[col] = (JSAMPLE) - ((ctab[r+R_CR_OFF] + ctab[g+G_CR_OFF] + ctab[b+B_CR_OFF]) - >> SCALEBITS); - } - } -} - - -/* - * Convert some rows of samples to the JPEG colorspace. - * [R,G,B] to [R-G,G,B-G] conversion with modulo calculation - * (forward reversible color transform). - * This can be seen as an adaption of the general RGB->YCbCr - * conversion equation with Kr = Kb = 0, while replacing the - * normalization by modulo calculation. - */ - -METHODDEF(void) -rgb_rgb1_convert (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JSAMPIMAGE output_buf, - JDIMENSION output_row, int num_rows) -{ - register int r, g, b; - register JSAMPROW inptr; - register JSAMPROW outptr0, outptr1, outptr2; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->image_width; - - while (--num_rows >= 0) { - inptr = *input_buf++; - outptr0 = output_buf[0][output_row]; - outptr1 = output_buf[1][output_row]; - outptr2 = output_buf[2][output_row]; - output_row++; - for (col = 0; col < num_cols; col++) { - r = GETJSAMPLE(inptr[RGB_RED]); - g = GETJSAMPLE(inptr[RGB_GREEN]); - b = GETJSAMPLE(inptr[RGB_BLUE]); - inptr += RGB_PIXELSIZE; - /* Assume that MAXJSAMPLE+1 is a power of 2, so that the MOD - * (modulo) operator is equivalent to the bitmask operator AND. - */ - outptr0[col] = (JSAMPLE) ((r - g + CENTERJSAMPLE) & MAXJSAMPLE); - outptr1[col] = (JSAMPLE) g; - outptr2[col] = (JSAMPLE) ((b - g + CENTERJSAMPLE) & MAXJSAMPLE); - } - } -} - - -/* - * Convert some rows of samples to the JPEG colorspace. - * This version handles grayscale output with no conversion. - * The source can be either plain grayscale or YCC (since Y == gray). - */ - -METHODDEF(void) -grayscale_convert (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JSAMPIMAGE output_buf, - JDIMENSION output_row, int num_rows) -{ - register JSAMPROW inptr; - register JSAMPROW outptr; - register JDIMENSION count; - register int instride = cinfo->input_components; - JDIMENSION num_cols = cinfo->image_width; - - while (--num_rows >= 0) { - inptr = *input_buf++; - outptr = output_buf[0][output_row++]; - for (count = num_cols; count > 0; count--) { - *outptr++ = *inptr; /* don't need GETJSAMPLE() here */ - inptr += instride; - } - } -} - - -/* - * Convert some rows of samples to the JPEG colorspace. - * No colorspace conversion, but change from interleaved - * to separate-planes representation. - */ - -METHODDEF(void) -rgb_convert (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JSAMPIMAGE output_buf, - JDIMENSION output_row, int num_rows) -{ - register JSAMPROW inptr; - register JSAMPROW outptr0, outptr1, outptr2; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->image_width; - - while (--num_rows >= 0) { - inptr = *input_buf++; - outptr0 = output_buf[0][output_row]; - outptr1 = output_buf[1][output_row]; - outptr2 = output_buf[2][output_row]; - output_row++; - for (col = 0; col < num_cols; col++) { - /* We can dispense with GETJSAMPLE() here */ - outptr0[col] = inptr[RGB_RED]; - outptr1[col] = inptr[RGB_GREEN]; - outptr2[col] = inptr[RGB_BLUE]; - inptr += RGB_PIXELSIZE; - } - } -} - - -/* - * Convert some rows of samples to the JPEG colorspace. - * This version handles multi-component colorspaces without conversion. - * We assume input_components == num_components. - */ - -METHODDEF(void) -null_convert (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JSAMPIMAGE output_buf, - JDIMENSION output_row, int num_rows) -{ - register JSAMPROW inptr; - register JSAMPROW outptr; - register JDIMENSION count; - register int num_comps = cinfo->num_components; - JDIMENSION num_cols = cinfo->image_width; - int ci; - - while (--num_rows >= 0) { - /* It seems fastest to make a separate pass for each component. */ - for (ci = 0; ci < num_comps; ci++) { - inptr = input_buf[0] + ci; - outptr = output_buf[ci][output_row]; - for (count = num_cols; count > 0; count--) { - *outptr++ = *inptr; /* don't need GETJSAMPLE() here */ - inptr += num_comps; - } - } - input_buf++; - output_row++; - } -} - - -/* - * Empty method for start_pass. - */ - -METHODDEF(void) -null_method (j_compress_ptr cinfo) -{ - /* no work needed */ -} - - -/* - * Module initialization routine for input colorspace conversion. - */ - -GLOBAL(void) -jinit_color_converter (j_compress_ptr cinfo) -{ - my_cconvert_ptr cconvert; - - cconvert = (my_cconvert_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_color_converter)); - cinfo->cconvert = &cconvert->pub; - /* set start_pass to null method until we find out differently */ - cconvert->pub.start_pass = null_method; - - /* Make sure input_components agrees with in_color_space */ - switch (cinfo->in_color_space) { - case JCS_GRAYSCALE: - if (cinfo->input_components != 1) - ERREXIT(cinfo, JERR_BAD_IN_COLORSPACE); - break; - - case JCS_RGB: - case JCS_BG_RGB: -#if RGB_PIXELSIZE != 3 - if (cinfo->input_components != RGB_PIXELSIZE) - ERREXIT(cinfo, JERR_BAD_IN_COLORSPACE); - break; -#endif /* else share code with YCbCr */ - - case JCS_YCbCr: - case JCS_BG_YCC: - if (cinfo->input_components != 3) - ERREXIT(cinfo, JERR_BAD_IN_COLORSPACE); - break; - - case JCS_CMYK: - case JCS_YCCK: - if (cinfo->input_components != 4) - ERREXIT(cinfo, JERR_BAD_IN_COLORSPACE); - break; - - default: /* JCS_UNKNOWN can be anything */ - if (cinfo->input_components < 1) - ERREXIT(cinfo, JERR_BAD_IN_COLORSPACE); - } - - /* Support color transform only for RGB colorspaces */ - if (cinfo->color_transform && - cinfo->jpeg_color_space != JCS_RGB && - cinfo->jpeg_color_space != JCS_BG_RGB) - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - - /* Check num_components, set conversion method based on requested space */ - switch (cinfo->jpeg_color_space) { - case JCS_GRAYSCALE: - if (cinfo->num_components != 1) - ERREXIT(cinfo, JERR_BAD_J_COLORSPACE); - switch (cinfo->in_color_space) { - case JCS_GRAYSCALE: - case JCS_YCbCr: - case JCS_BG_YCC: - cconvert->pub.color_convert = grayscale_convert; - break; - case JCS_RGB: - cconvert->pub.start_pass = rgb_ycc_start; - cconvert->pub.color_convert = rgb_gray_convert; - break; - default: - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - } - break; - - case JCS_RGB: - case JCS_BG_RGB: - if (cinfo->num_components != 3) - ERREXIT(cinfo, JERR_BAD_J_COLORSPACE); - if (cinfo->in_color_space != cinfo->jpeg_color_space) - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - switch (cinfo->color_transform) { - case JCT_NONE: - cconvert->pub.color_convert = rgb_convert; - break; - case JCT_SUBTRACT_GREEN: - cconvert->pub.color_convert = rgb_rgb1_convert; - break; - default: - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - } - break; - - case JCS_YCbCr: - if (cinfo->num_components != 3) - ERREXIT(cinfo, JERR_BAD_J_COLORSPACE); - switch (cinfo->in_color_space) { - case JCS_RGB: - cconvert->pub.start_pass = rgb_ycc_start; - cconvert->pub.color_convert = rgb_ycc_convert; - break; - case JCS_YCbCr: - cconvert->pub.color_convert = null_convert; - break; - default: - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - } - break; - - case JCS_BG_YCC: - if (cinfo->num_components != 3) - ERREXIT(cinfo, JERR_BAD_J_COLORSPACE); - switch (cinfo->in_color_space) { - case JCS_RGB: - /* For conversion from normal RGB input to BG_YCC representation, - * the Cb/Cr values are first computed as usual, and then - * quantized further after DCT processing by a factor of - * 2 in reference to the nominal quantization factor. - */ - /* need quantization scale by factor of 2 after DCT */ - cinfo->comp_info[1].component_needed = TRUE; - cinfo->comp_info[2].component_needed = TRUE; - /* compute normal YCC first */ - cconvert->pub.start_pass = rgb_ycc_start; - cconvert->pub.color_convert = rgb_ycc_convert; - break; - case JCS_YCbCr: - /* need quantization scale by factor of 2 after DCT */ - cinfo->comp_info[1].component_needed = TRUE; - cinfo->comp_info[2].component_needed = TRUE; - /*FALLTHROUGH*/ - case JCS_BG_YCC: - /* Pass through for BG_YCC input */ - cconvert->pub.color_convert = null_convert; - break; - default: - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - } - break; - - case JCS_CMYK: - if (cinfo->num_components != 4) - ERREXIT(cinfo, JERR_BAD_J_COLORSPACE); - if (cinfo->in_color_space != JCS_CMYK) - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - cconvert->pub.color_convert = null_convert; - break; - - case JCS_YCCK: - if (cinfo->num_components != 4) - ERREXIT(cinfo, JERR_BAD_J_COLORSPACE); - switch (cinfo->in_color_space) { - case JCS_CMYK: - cconvert->pub.start_pass = rgb_ycc_start; - cconvert->pub.color_convert = cmyk_ycck_convert; - break; - case JCS_YCCK: - cconvert->pub.color_convert = null_convert; - break; - default: - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - } - break; - - default: /* allow null conversion of JCS_UNKNOWN */ - if (cinfo->jpeg_color_space != cinfo->in_color_space || - cinfo->num_components != cinfo->input_components) - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - cconvert->pub.color_convert = null_convert; - } -} diff --git a/dep/libjpeg/src/jcdctmgr.c b/dep/libjpeg/src/jcdctmgr.c deleted file mode 100644 index a48ccd814..000000000 --- a/dep/libjpeg/src/jcdctmgr.c +++ /dev/null @@ -1,466 +0,0 @@ -/* - * jcdctmgr.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2003-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains the forward-DCT management logic. - * This code selects a particular DCT implementation to be used, - * and it performs related housekeeping chores including coefficient - * quantization. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" -#include "jdct.h" /* Private declarations for DCT subsystem */ - - -/* Private subobject for this module */ - -typedef struct { - struct jpeg_forward_dct pub; /* public fields */ - - /* Pointer to the DCT routine actually in use */ - forward_DCT_method_ptr do_dct[MAX_COMPONENTS]; - -#ifdef DCT_FLOAT_SUPPORTED - /* Same as above for the floating-point case. */ - float_DCT_method_ptr do_float_dct[MAX_COMPONENTS]; -#endif -} my_fdct_controller; - -typedef my_fdct_controller * my_fdct_ptr; - - -/* The allocated post-DCT divisor tables -- big enough for any - * supported variant and not identical to the quant table entries, - * because of scaling (especially for an unnormalized DCT) -- - * are pointed to by dct_table in the per-component comp_info - * structures. Each table is given in normal array order. - */ - -typedef union { - DCTELEM int_array[DCTSIZE2]; -#ifdef DCT_FLOAT_SUPPORTED - FAST_FLOAT float_array[DCTSIZE2]; -#endif -} divisor_table; - - -/* The current scaled-DCT routines require ISLOW-style divisor tables, - * so be sure to compile that code if either ISLOW or SCALING is requested. - */ -#ifdef DCT_ISLOW_SUPPORTED -#define PROVIDE_ISLOW_TABLES -#else -#ifdef DCT_SCALING_SUPPORTED -#define PROVIDE_ISLOW_TABLES -#endif -#endif - - -/* - * Perform forward DCT on one or more blocks of a component. - * - * The input samples are taken from the sample_data[] array starting at - * position start_col, and moving to the right for any additional blocks. - * The quantized coefficients are returned in coef_blocks[]. - */ - -METHODDEF(void) -forward_DCT (j_compress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY sample_data, JBLOCKROW coef_blocks, - JDIMENSION start_col, JDIMENSION num_blocks) -/* This version is used for integer DCT implementations. */ -{ - /* This routine is heavily used, so it's worth coding it tightly. */ - my_fdct_ptr fdct = (my_fdct_ptr) cinfo->fdct; - forward_DCT_method_ptr do_dct = fdct->do_dct[compptr->component_index]; - DCTELEM * divisors = (DCTELEM *) compptr->dct_table; - DCTELEM workspace[DCTSIZE2]; /* work area for FDCT subroutine */ - JDIMENSION bi; - - for (bi = 0; bi < num_blocks; bi++, start_col += compptr->DCT_h_scaled_size) { - /* Perform the DCT */ - (*do_dct) (workspace, sample_data, start_col); - - /* Quantize/descale the coefficients, and store into coef_blocks[] */ - { register DCTELEM temp, qval; - register int i; - register JCOEFPTR output_ptr = coef_blocks[bi]; - - for (i = 0; i < DCTSIZE2; i++) { - qval = divisors[i]; - temp = workspace[i]; - /* Divide the coefficient value by qval, ensuring proper rounding. - * Since C does not specify the direction of rounding for negative - * quotients, we have to force the dividend positive for portability. - * - * In most files, at least half of the output values will be zero - * (at default quantization settings, more like three-quarters...) - * so we should ensure that this case is fast. On many machines, - * a comparison is enough cheaper than a divide to make a special test - * a win. Since both inputs will be nonnegative, we need only test - * for a < b to discover whether a/b is 0. - * If your machine's division is fast enough, define FAST_DIVIDE. - */ -#ifdef FAST_DIVIDE -#define DIVIDE_BY(a,b) a /= b -#else -#define DIVIDE_BY(a,b) if (a >= b) a /= b; else a = 0 -#endif - if (temp < 0) { - temp = -temp; - temp += qval>>1; /* for rounding */ - DIVIDE_BY(temp, qval); - temp = -temp; - } else { - temp += qval>>1; /* for rounding */ - DIVIDE_BY(temp, qval); - } - output_ptr[i] = (JCOEF) temp; - } - } - } -} - - -#ifdef DCT_FLOAT_SUPPORTED - -METHODDEF(void) -forward_DCT_float (j_compress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY sample_data, JBLOCKROW coef_blocks, - JDIMENSION start_col, JDIMENSION num_blocks) -/* This version is used for floating-point DCT implementations. */ -{ - /* This routine is heavily used, so it's worth coding it tightly. */ - my_fdct_ptr fdct = (my_fdct_ptr) cinfo->fdct; - float_DCT_method_ptr do_dct = fdct->do_float_dct[compptr->component_index]; - FAST_FLOAT * divisors = (FAST_FLOAT *) compptr->dct_table; - FAST_FLOAT workspace[DCTSIZE2]; /* work area for FDCT subroutine */ - JDIMENSION bi; - - for (bi = 0; bi < num_blocks; bi++, start_col += compptr->DCT_h_scaled_size) { - /* Perform the DCT */ - (*do_dct) (workspace, sample_data, start_col); - - /* Quantize/descale the coefficients, and store into coef_blocks[] */ - { register FAST_FLOAT temp; - register int i; - register JCOEFPTR output_ptr = coef_blocks[bi]; - - for (i = 0; i < DCTSIZE2; i++) { - /* Apply the quantization and scaling factor */ - temp = workspace[i] * divisors[i]; - /* Round to nearest integer. - * Since C does not specify the direction of rounding for negative - * quotients, we have to force the dividend positive for portability. - * The maximum coefficient size is +-16K (for 12-bit data), so this - * code should work for either 16-bit or 32-bit ints. - */ - output_ptr[i] = (JCOEF) ((int) (temp + (FAST_FLOAT) 16384.5) - 16384); - } - } - } -} - -#endif /* DCT_FLOAT_SUPPORTED */ - - -/* - * Initialize for a processing pass. - * Verify that all referenced Q-tables are present, and set up - * the divisor table for each one. - * In the current implementation, DCT of all components is done during - * the first pass, even if only some components will be output in the - * first scan. Hence all components should be examined here. - */ - -METHODDEF(void) -start_pass_fdctmgr (j_compress_ptr cinfo) -{ - my_fdct_ptr fdct = (my_fdct_ptr) cinfo->fdct; - int ci, qtblno, i; - jpeg_component_info *compptr; - int method = 0; - JQUANT_TBL * qtbl; - DCTELEM * dtbl; - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* Select the proper DCT routine for this component's scaling */ - switch ((compptr->DCT_h_scaled_size << 8) + compptr->DCT_v_scaled_size) { -#ifdef DCT_SCALING_SUPPORTED - case ((1 << 8) + 1): - fdct->do_dct[ci] = jpeg_fdct_1x1; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((2 << 8) + 2): - fdct->do_dct[ci] = jpeg_fdct_2x2; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((3 << 8) + 3): - fdct->do_dct[ci] = jpeg_fdct_3x3; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((4 << 8) + 4): - fdct->do_dct[ci] = jpeg_fdct_4x4; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((5 << 8) + 5): - fdct->do_dct[ci] = jpeg_fdct_5x5; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((6 << 8) + 6): - fdct->do_dct[ci] = jpeg_fdct_6x6; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((7 << 8) + 7): - fdct->do_dct[ci] = jpeg_fdct_7x7; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((9 << 8) + 9): - fdct->do_dct[ci] = jpeg_fdct_9x9; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((10 << 8) + 10): - fdct->do_dct[ci] = jpeg_fdct_10x10; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((11 << 8) + 11): - fdct->do_dct[ci] = jpeg_fdct_11x11; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((12 << 8) + 12): - fdct->do_dct[ci] = jpeg_fdct_12x12; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((13 << 8) + 13): - fdct->do_dct[ci] = jpeg_fdct_13x13; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((14 << 8) + 14): - fdct->do_dct[ci] = jpeg_fdct_14x14; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((15 << 8) + 15): - fdct->do_dct[ci] = jpeg_fdct_15x15; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((16 << 8) + 16): - fdct->do_dct[ci] = jpeg_fdct_16x16; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((16 << 8) + 8): - fdct->do_dct[ci] = jpeg_fdct_16x8; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((14 << 8) + 7): - fdct->do_dct[ci] = jpeg_fdct_14x7; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((12 << 8) + 6): - fdct->do_dct[ci] = jpeg_fdct_12x6; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((10 << 8) + 5): - fdct->do_dct[ci] = jpeg_fdct_10x5; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((8 << 8) + 4): - fdct->do_dct[ci] = jpeg_fdct_8x4; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((6 << 8) + 3): - fdct->do_dct[ci] = jpeg_fdct_6x3; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((4 << 8) + 2): - fdct->do_dct[ci] = jpeg_fdct_4x2; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((2 << 8) + 1): - fdct->do_dct[ci] = jpeg_fdct_2x1; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((8 << 8) + 16): - fdct->do_dct[ci] = jpeg_fdct_8x16; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((7 << 8) + 14): - fdct->do_dct[ci] = jpeg_fdct_7x14; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((6 << 8) + 12): - fdct->do_dct[ci] = jpeg_fdct_6x12; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((5 << 8) + 10): - fdct->do_dct[ci] = jpeg_fdct_5x10; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((4 << 8) + 8): - fdct->do_dct[ci] = jpeg_fdct_4x8; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((3 << 8) + 6): - fdct->do_dct[ci] = jpeg_fdct_3x6; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((2 << 8) + 4): - fdct->do_dct[ci] = jpeg_fdct_2x4; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; - case ((1 << 8) + 2): - fdct->do_dct[ci] = jpeg_fdct_1x2; - method = JDCT_ISLOW; /* jfdctint uses islow-style table */ - break; -#endif - case ((DCTSIZE << 8) + DCTSIZE): - switch (cinfo->dct_method) { -#ifdef DCT_ISLOW_SUPPORTED - case JDCT_ISLOW: - fdct->do_dct[ci] = jpeg_fdct_islow; - method = JDCT_ISLOW; - break; -#endif -#ifdef DCT_IFAST_SUPPORTED - case JDCT_IFAST: - fdct->do_dct[ci] = jpeg_fdct_ifast; - method = JDCT_IFAST; - break; -#endif -#ifdef DCT_FLOAT_SUPPORTED - case JDCT_FLOAT: - fdct->do_float_dct[ci] = jpeg_fdct_float; - method = JDCT_FLOAT; - break; -#endif - default: - ERREXIT(cinfo, JERR_NOT_COMPILED); - } - break; - default: - ERREXIT2(cinfo, JERR_BAD_DCTSIZE, - compptr->DCT_h_scaled_size, compptr->DCT_v_scaled_size); - } - qtblno = compptr->quant_tbl_no; - /* Make sure specified quantization table is present */ - if (qtblno < 0 || qtblno >= NUM_QUANT_TBLS || - cinfo->quant_tbl_ptrs[qtblno] == NULL) - ERREXIT1(cinfo, JERR_NO_QUANT_TABLE, qtblno); - qtbl = cinfo->quant_tbl_ptrs[qtblno]; - /* Create divisor table from quant table */ - switch (method) { -#ifdef PROVIDE_ISLOW_TABLES - case JDCT_ISLOW: - /* For LL&M IDCT method, divisors are equal to raw quantization - * coefficients multiplied by 8 (to counteract scaling). - */ - dtbl = (DCTELEM *) compptr->dct_table; - for (i = 0; i < DCTSIZE2; i++) { - dtbl[i] = - ((DCTELEM) qtbl->quantval[i]) << (compptr->component_needed ? 4 : 3); - } - fdct->pub.forward_DCT[ci] = forward_DCT; - break; -#endif -#ifdef DCT_IFAST_SUPPORTED - case JDCT_IFAST: - { - /* For AA&N IDCT method, divisors are equal to quantization - * coefficients scaled by scalefactor[row]*scalefactor[col], where - * scalefactor[0] = 1 - * scalefactor[k] = cos(k*PI/16) * sqrt(2) for k=1..7 - * We apply a further scale factor of 8. - */ -#define CONST_BITS 14 - static const INT16 aanscales[DCTSIZE2] = { - /* precomputed values scaled up by 14 bits */ - 16384, 22725, 21407, 19266, 16384, 12873, 8867, 4520, - 22725, 31521, 29692, 26722, 22725, 17855, 12299, 6270, - 21407, 29692, 27969, 25172, 21407, 16819, 11585, 5906, - 19266, 26722, 25172, 22654, 19266, 15137, 10426, 5315, - 16384, 22725, 21407, 19266, 16384, 12873, 8867, 4520, - 12873, 17855, 16819, 15137, 12873, 10114, 6967, 3552, - 8867, 12299, 11585, 10426, 8867, 6967, 4799, 2446, - 4520, 6270, 5906, 5315, 4520, 3552, 2446, 1247 - }; - SHIFT_TEMPS - - dtbl = (DCTELEM *) compptr->dct_table; - for (i = 0; i < DCTSIZE2; i++) { - dtbl[i] = (DCTELEM) - DESCALE(MULTIPLY16V16((INT32) qtbl->quantval[i], - (INT32) aanscales[i]), - compptr->component_needed ? CONST_BITS-4 : CONST_BITS-3); - } - } - fdct->pub.forward_DCT[ci] = forward_DCT; - break; -#endif -#ifdef DCT_FLOAT_SUPPORTED - case JDCT_FLOAT: - { - /* For float AA&N IDCT method, divisors are equal to quantization - * coefficients scaled by scalefactor[row]*scalefactor[col], where - * scalefactor[0] = 1 - * scalefactor[k] = cos(k*PI/16) * sqrt(2) for k=1..7 - * We apply a further scale factor of 8. - * What's actually stored is 1/divisor so that the inner loop can - * use a multiplication rather than a division. - */ - FAST_FLOAT * fdtbl = (FAST_FLOAT *) compptr->dct_table; - int row, col; - static const double aanscalefactor[DCTSIZE] = { - 1.0, 1.387039845, 1.306562965, 1.175875602, - 1.0, 0.785694958, 0.541196100, 0.275899379 - }; - - i = 0; - for (row = 0; row < DCTSIZE; row++) { - for (col = 0; col < DCTSIZE; col++) { - fdtbl[i] = (FAST_FLOAT) - (1.0 / ((double) qtbl->quantval[i] * - aanscalefactor[row] * aanscalefactor[col] * - (compptr->component_needed ? 16.0 : 8.0))); - i++; - } - } - } - fdct->pub.forward_DCT[ci] = forward_DCT_float; - break; -#endif - default: - ERREXIT(cinfo, JERR_NOT_COMPILED); - } - } -} - - -/* - * Initialize FDCT manager. - */ - -GLOBAL(void) -jinit_forward_dct (j_compress_ptr cinfo) -{ - my_fdct_ptr fdct; - int ci; - jpeg_component_info *compptr; - - fdct = (my_fdct_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_fdct_controller)); - cinfo->fdct = &fdct->pub; - fdct->pub.start_pass = start_pass_fdctmgr; - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* Allocate a divisor table for each component */ - compptr->dct_table = (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(divisor_table)); - } -} diff --git a/dep/libjpeg/src/jchuff.c b/dep/libjpeg/src/jchuff.c deleted file mode 100644 index 1f527b218..000000000 --- a/dep/libjpeg/src/jchuff.c +++ /dev/null @@ -1,1656 +0,0 @@ -/* - * jchuff.c - * - * Copyright (C) 1991-1997, Thomas G. Lane. - * Modified 2006-2023 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains Huffman entropy encoding routines. - * Both sequential and progressive modes are supported in this single module. - * - * Much of the complexity here has to do with supporting output suspension. - * If the data destination module demands suspension, we want to be able to - * back up to the start of the current MCU. To do this, we copy state - * variables into local working storage, and update them back to the - * permanent JPEG objects only upon successful completion of an MCU. - * - * We do not support output suspension for the progressive JPEG mode, since - * the library currently does not allow multiple-scan files to be written - * with output suspension. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* The legal range of a DCT coefficient is - * -1024 .. +1023 for 8-bit sample data precision; - * -16384 .. +16383 for 12-bit sample data precision. - * Hence the magnitude should always fit in sample data precision + 2 bits. - */ - -/* Derived data constructed for each Huffman table */ - -typedef struct { - unsigned int ehufco[256]; /* code for each symbol */ - char ehufsi[256]; /* length of code for each symbol */ - /* If no code has been allocated for a symbol S, ehufsi[S] contains 0 */ -} c_derived_tbl; - - -/* Expanded entropy encoder object for Huffman encoding. - * - * The savable_state subrecord contains fields that change within an MCU, - * but must not be updated permanently until we complete the MCU. - */ - -typedef struct { - INT32 put_buffer; /* current bit-accumulation buffer */ - int put_bits; /* # of bits now in it */ - int last_dc_val[MAX_COMPS_IN_SCAN]; /* last DC coef for each component */ -} savable_state; - -/* This macro is to work around compilers with missing or broken - * structure assignment. You'll need to fix this code if you have - * such a compiler and you change MAX_COMPS_IN_SCAN. - */ - -#ifndef NO_STRUCT_ASSIGN -#define ASSIGN_STATE(dest,src) ((dest) = (src)) -#else -#if MAX_COMPS_IN_SCAN == 4 -#define ASSIGN_STATE(dest,src) \ - ((dest).put_buffer = (src).put_buffer, \ - (dest).put_bits = (src).put_bits, \ - (dest).last_dc_val[0] = (src).last_dc_val[0], \ - (dest).last_dc_val[1] = (src).last_dc_val[1], \ - (dest).last_dc_val[2] = (src).last_dc_val[2], \ - (dest).last_dc_val[3] = (src).last_dc_val[3]) -#endif -#endif - - -typedef struct { - struct jpeg_entropy_encoder pub; /* public fields */ - - savable_state saved; /* Bit buffer & DC state at start of MCU */ - - /* These fields are NOT loaded into local working state. */ - unsigned int restarts_to_go; /* MCUs left in this restart interval */ - int next_restart_num; /* next restart number to write (0-7) */ - - /* Pointers to derived tables (these workspaces have image lifespan) */ - c_derived_tbl * dc_derived_tbls[NUM_HUFF_TBLS]; - c_derived_tbl * ac_derived_tbls[NUM_HUFF_TBLS]; - - /* Statistics tables for optimization */ - long * dc_count_ptrs[NUM_HUFF_TBLS]; - long * ac_count_ptrs[NUM_HUFF_TBLS]; - - /* Following fields used only in progressive mode */ - - /* Mode flag: TRUE for optimization, FALSE for actual data output */ - boolean gather_statistics; - - /* next_output_byte/free_in_buffer are local copies of cinfo->dest fields. - */ - JOCTET * next_output_byte; /* => next byte to write in buffer */ - size_t free_in_buffer; /* # of byte spaces remaining in buffer */ - j_compress_ptr cinfo; /* link to cinfo (needed for dump_buffer) */ - - /* Coding status for AC components */ - int ac_tbl_no; /* the table number of the single component */ - unsigned int EOBRUN; /* run length of EOBs */ - unsigned int BE; /* # of buffered correction bits before MCU */ - char * bit_buffer; /* buffer for correction bits (1 per char) */ - /* packing correction bits tightly would save some space but cost time... */ -} huff_entropy_encoder; - -typedef huff_entropy_encoder * huff_entropy_ptr; - -/* Working state while writing an MCU (sequential mode). - * This struct contains all the fields that are needed by subroutines. - */ - -typedef struct { - JOCTET * next_output_byte; /* => next byte to write in buffer */ - size_t free_in_buffer; /* # of byte spaces remaining in buffer */ - savable_state cur; /* Current bit buffer & DC state */ - j_compress_ptr cinfo; /* dump_buffer needs access to this */ -} working_state; - -/* MAX_CORR_BITS is the number of bits the AC refinement correction-bit - * buffer can hold. Larger sizes may slightly improve compression, but - * 1000 is already well into the realm of overkill. - * The minimum safe size is 64 bits. - */ - -#define MAX_CORR_BITS 1000 /* Max # of correction bits I can buffer */ - -/* IRIGHT_SHIFT is like RIGHT_SHIFT, but works on int rather than INT32. - * We assume that int right shift is unsigned if INT32 right shift is, - * which should be safe. - */ - -#ifdef RIGHT_SHIFT_IS_UNSIGNED -#define ISHIFT_TEMPS int ishift_temp; -#define IRIGHT_SHIFT(x,shft) \ - ((ishift_temp = (x)) < 0 ? \ - (ishift_temp >> (shft)) | ((~0) << (16-(shft))) : \ - (ishift_temp >> (shft))) -#else -#define ISHIFT_TEMPS -#define IRIGHT_SHIFT(x,shft) ((x) >> (shft)) -#endif - - -/* - * Compute the derived values for a Huffman table. - * This routine also performs some validation checks on the table. - */ - -LOCAL(void) -jpeg_make_c_derived_tbl (j_compress_ptr cinfo, boolean isDC, int tblno, - c_derived_tbl ** pdtbl) -{ - JHUFF_TBL *htbl; - c_derived_tbl *dtbl; - int p, i, l, lastp, si, maxsymbol; - char huffsize[257]; - unsigned int huffcode[257]; - unsigned int code; - - /* Note that huffsize[] and huffcode[] are filled in code-length order, - * paralleling the order of the symbols themselves in htbl->huffval[]. - */ - - /* Find the input Huffman table */ - if (tblno < 0 || tblno >= NUM_HUFF_TBLS) - ERREXIT1(cinfo, JERR_NO_HUFF_TABLE, tblno); - htbl = - isDC ? cinfo->dc_huff_tbl_ptrs[tblno] : cinfo->ac_huff_tbl_ptrs[tblno]; - if (htbl == NULL) - htbl = jpeg_std_huff_table((j_common_ptr) cinfo, isDC, tblno); - - /* Allocate a workspace if we haven't already done so. */ - if (*pdtbl == NULL) - *pdtbl = (c_derived_tbl *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(c_derived_tbl)); - dtbl = *pdtbl; - - /* Figure C.1: make table of Huffman code length for each symbol */ - - p = 0; - for (l = 1; l <= 16; l++) { - i = (int) htbl->bits[l]; - if (i < 0 || p + i > 256) /* protect against table overrun */ - ERREXIT(cinfo, JERR_BAD_HUFF_TABLE); - while (i--) - huffsize[p++] = (char) l; - } - huffsize[p] = 0; - lastp = p; - - /* Figure C.2: generate the codes themselves */ - /* We also validate that the counts represent a legal Huffman code tree. */ - - code = 0; - si = huffsize[0]; - p = 0; - while (huffsize[p]) { - while (((int) huffsize[p]) == si) { - huffcode[p++] = code; - code++; - } - /* code is now 1 more than the last code used for codelength si; but - * it must still fit in si bits, since no code is allowed to be all ones. - */ - if (((INT32) code) >= (((INT32) 1) << si)) - ERREXIT(cinfo, JERR_BAD_HUFF_TABLE); - code <<= 1; - si++; - } - - /* Figure C.3: generate encoding tables */ - /* These are code and size indexed by symbol value */ - - /* Set all codeless symbols to have code length 0; - * this lets us detect duplicate VAL entries here, and later - * allows emit_bits to detect any attempt to emit such symbols. - */ - MEMZERO(dtbl->ehufsi, SIZEOF(dtbl->ehufsi)); - - /* This is also a convenient place to check for out-of-range - * and duplicated VAL entries. We allow 0..255 for AC symbols - * but only 0..15 for DC. (We could constrain them further - * based on data depth and mode, but this seems enough.) - */ - maxsymbol = isDC ? 15 : 255; - - for (p = 0; p < lastp; p++) { - i = htbl->huffval[p]; - if (i < 0 || i > maxsymbol || dtbl->ehufsi[i]) - ERREXIT(cinfo, JERR_BAD_HUFF_TABLE); - dtbl->ehufco[i] = huffcode[p]; - dtbl->ehufsi[i] = huffsize[p]; - } -} - - -/* Outputting bytes to the file. - * NB: these must be called only when actually outputting, - * that is, entropy->gather_statistics == FALSE. - */ - -/* Emit a byte, taking 'action' if must suspend. */ -#define emit_byte_s(state,val,action) \ - { *(state)->next_output_byte++ = (JOCTET) (val); \ - if (--(state)->free_in_buffer == 0) \ - if (! dump_buffer_s(state)) \ - { action; } } - -/* Emit a byte */ -#define emit_byte_e(entropy,val) \ - { *(entropy)->next_output_byte++ = (JOCTET) (val); \ - if (--(entropy)->free_in_buffer == 0) \ - dump_buffer_e(entropy); } - - -LOCAL(boolean) -dump_buffer_s (working_state * state) -/* Empty the output buffer; return TRUE if successful, FALSE if must suspend */ -{ - struct jpeg_destination_mgr * dest = state->cinfo->dest; - - if (! (*dest->empty_output_buffer) (state->cinfo)) - return FALSE; - /* After a successful buffer dump, must reset buffer pointers */ - state->next_output_byte = dest->next_output_byte; - state->free_in_buffer = dest->free_in_buffer; - return TRUE; -} - - -LOCAL(void) -dump_buffer_e (huff_entropy_ptr entropy) -/* Empty the output buffer; we do not support suspension in this case. */ -{ - struct jpeg_destination_mgr * dest = entropy->cinfo->dest; - - if (! (*dest->empty_output_buffer) (entropy->cinfo)) - ERREXIT(entropy->cinfo, JERR_CANT_SUSPEND); - /* After a successful buffer dump, must reset buffer pointers */ - entropy->next_output_byte = dest->next_output_byte; - entropy->free_in_buffer = dest->free_in_buffer; -} - - -/* Outputting bits to the file */ - -/* Only the right 24 bits of put_buffer are used; the valid bits are - * left-justified in this part. At most 16 bits can be passed to emit_bits - * in one call, and we never retain more than 7 bits in put_buffer - * between calls, so 24 bits are sufficient. - */ - -INLINE -LOCAL(boolean) -emit_bits_s (working_state * state, unsigned int code, int size) -/* Emit some bits; return TRUE if successful, FALSE if must suspend */ -{ - /* This routine is heavily used, so it's worth coding tightly. */ - register INT32 put_buffer; - register int put_bits; - - /* if size is 0, caller used an invalid Huffman table entry */ - if (size == 0) - ERREXIT(state->cinfo, JERR_HUFF_MISSING_CODE); - - /* mask off any extra bits in code */ - put_buffer = ((INT32) code) & ((((INT32) 1) << size) - 1); - - /* new number of bits in buffer */ - put_bits = size + state->cur.put_bits; - - put_buffer <<= 24 - put_bits; /* align incoming bits */ - - /* and merge with old buffer contents */ - put_buffer |= state->cur.put_buffer; - - while (put_bits >= 8) { - int c = (int) ((put_buffer >> 16) & 0xFF); - - emit_byte_s(state, c, return FALSE); - if (c == 0xFF) { /* need to stuff a zero byte? */ - emit_byte_s(state, 0, return FALSE); - } - put_buffer <<= 8; - put_bits -= 8; - } - - state->cur.put_buffer = put_buffer; /* update state variables */ - state->cur.put_bits = put_bits; - - return TRUE; -} - - -INLINE -LOCAL(void) -emit_bits_e (huff_entropy_ptr entropy, unsigned int code, int size) -/* Emit some bits, unless we are in gather mode */ -{ - /* This routine is heavily used, so it's worth coding tightly. */ - register INT32 put_buffer; - register int put_bits; - - /* if size is 0, caller used an invalid Huffman table entry */ - if (size == 0) - ERREXIT(entropy->cinfo, JERR_HUFF_MISSING_CODE); - - if (entropy->gather_statistics) - return; /* do nothing if we're only getting stats */ - - /* mask off any extra bits in code */ - put_buffer = ((INT32) code) & ((((INT32) 1) << size) - 1); - - /* new number of bits in buffer */ - put_bits = size + entropy->saved.put_bits; - - put_buffer <<= 24 - put_bits; /* align incoming bits */ - - /* and merge with old buffer contents */ - put_buffer |= entropy->saved.put_buffer; - - while (put_bits >= 8) { - int c = (int) ((put_buffer >> 16) & 0xFF); - - emit_byte_e(entropy, c); - if (c == 0xFF) { /* need to stuff a zero byte? */ - emit_byte_e(entropy, 0); - } - put_buffer <<= 8; - put_bits -= 8; - } - - entropy->saved.put_buffer = put_buffer; /* update variables */ - entropy->saved.put_bits = put_bits; -} - - -LOCAL(boolean) -flush_bits_s (working_state * state) -{ - if (! emit_bits_s(state, 0x7F, 7)) /* fill any partial byte with ones */ - return FALSE; - state->cur.put_buffer = 0; /* and reset bit-buffer to empty */ - state->cur.put_bits = 0; - return TRUE; -} - - -LOCAL(void) -flush_bits_e (huff_entropy_ptr entropy) -{ - emit_bits_e(entropy, 0x7F, 7); /* fill any partial byte with ones */ - entropy->saved.put_buffer = 0; /* and reset bit-buffer to empty */ - entropy->saved.put_bits = 0; -} - - -/* - * Emit (or just count) a Huffman symbol. - */ - -INLINE -LOCAL(void) -emit_dc_symbol (huff_entropy_ptr entropy, int tbl_no, int symbol) -{ - if (entropy->gather_statistics) - entropy->dc_count_ptrs[tbl_no][symbol]++; - else { - c_derived_tbl * tbl = entropy->dc_derived_tbls[tbl_no]; - emit_bits_e(entropy, tbl->ehufco[symbol], tbl->ehufsi[symbol]); - } -} - - -INLINE -LOCAL(void) -emit_ac_symbol (huff_entropy_ptr entropy, int tbl_no, int symbol) -{ - if (entropy->gather_statistics) - entropy->ac_count_ptrs[tbl_no][symbol]++; - else { - c_derived_tbl * tbl = entropy->ac_derived_tbls[tbl_no]; - emit_bits_e(entropy, tbl->ehufco[symbol], tbl->ehufsi[symbol]); - } -} - - -/* - * Emit bits from a correction bit buffer. - */ - -LOCAL(void) -emit_buffered_bits (huff_entropy_ptr entropy, char * bufstart, - unsigned int nbits) -{ - if (entropy->gather_statistics) - return; /* no real work */ - - while (nbits > 0) { - emit_bits_e(entropy, (unsigned int) (*bufstart), 1); - bufstart++; - nbits--; - } -} - - -/* - * Emit any pending EOBRUN symbol. - */ - -LOCAL(void) -emit_eobrun (huff_entropy_ptr entropy) -{ - register int temp, nbits; - - if (entropy->EOBRUN > 0) { /* if there is any pending EOBRUN */ - temp = entropy->EOBRUN; - nbits = 0; - while ((temp >>= 1)) - nbits++; - /* safety check: shouldn't happen given limited correction-bit buffer */ - if (nbits > 14) - ERREXIT(entropy->cinfo, JERR_HUFF_MISSING_CODE); - - emit_ac_symbol(entropy, entropy->ac_tbl_no, nbits << 4); - if (nbits) - emit_bits_e(entropy, entropy->EOBRUN, nbits); - - entropy->EOBRUN = 0; - - /* Emit any buffered correction bits */ - emit_buffered_bits(entropy, entropy->bit_buffer, entropy->BE); - entropy->BE = 0; - } -} - - -/* - * Emit a restart marker & resynchronize predictions. - */ - -LOCAL(boolean) -emit_restart_s (working_state * state, int restart_num) -{ - int ci; - - if (! flush_bits_s(state)) - return FALSE; - - emit_byte_s(state, 0xFF, return FALSE); - emit_byte_s(state, JPEG_RST0 + restart_num, return FALSE); - - /* Re-initialize DC predictions to 0 */ - for (ci = 0; ci < state->cinfo->comps_in_scan; ci++) - state->cur.last_dc_val[ci] = 0; - - /* The restart counter is not updated until we successfully write the MCU. */ - - return TRUE; -} - - -LOCAL(void) -emit_restart_e (huff_entropy_ptr entropy, int restart_num) -{ - int ci; - - emit_eobrun(entropy); - - if (! entropy->gather_statistics) { - flush_bits_e(entropy); - emit_byte_e(entropy, 0xFF); - emit_byte_e(entropy, JPEG_RST0 + restart_num); - } - - if (entropy->cinfo->Ss == 0) { - /* Re-initialize DC predictions to 0 */ - for (ci = 0; ci < entropy->cinfo->comps_in_scan; ci++) - entropy->saved.last_dc_val[ci] = 0; - } else { - /* Re-initialize all AC-related fields to 0 */ - entropy->EOBRUN = 0; - entropy->BE = 0; - } -} - - -/* - * MCU encoding for DC initial scan (either spectral selection, - * or first pass of successive approximation). - */ - -METHODDEF(boolean) -encode_mcu_DC_first (j_compress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - register int temp, temp2; - register int nbits; - int max_coef_bits; - int blkn, ci, tbl; - ISHIFT_TEMPS - - entropy->next_output_byte = cinfo->dest->next_output_byte; - entropy->free_in_buffer = cinfo->dest->free_in_buffer; - - /* Emit restart marker if needed */ - if (cinfo->restart_interval) - if (entropy->restarts_to_go == 0) - emit_restart_e(entropy, entropy->next_restart_num); - - /* Since we're encoding a difference, the range limit is twice as much. */ - max_coef_bits = cinfo->data_precision + 3; - - /* Encode the MCU data blocks */ - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - ci = cinfo->MCU_membership[blkn]; - tbl = cinfo->cur_comp_info[ci]->dc_tbl_no; - - /* Compute the DC value after the required point transform by Al. - * This is simply an arithmetic right shift. - */ - temp = IRIGHT_SHIFT((int) (MCU_data[blkn][0][0]), cinfo->Al); - - /* DC differences are figured on the point-transformed values. */ - if ((temp2 = temp - entropy->saved.last_dc_val[ci]) == 0) { - /* Count/emit the Huffman-coded symbol for the number of bits */ - emit_dc_symbol(entropy, tbl, 0); - - continue; - } - - entropy->saved.last_dc_val[ci] = temp; - - /* Encode the DC coefficient difference per section G.1.2.1 */ - if ((temp = temp2) < 0) { - temp = -temp; /* temp is abs value of input */ - /* For a negative input, want temp2 = bitwise complement of abs(input) */ - /* This code assumes we are on a two's complement machine */ - temp2--; - } - - /* Find the number of bits needed for the magnitude of the coefficient */ - nbits = 0; - do nbits++; /* there must be at least one 1 bit */ - while ((temp >>= 1)); - /* Check for out-of-range coefficient values */ - if (nbits > max_coef_bits) - ERREXIT(cinfo, JERR_BAD_DCT_COEF); - - /* Count/emit the Huffman-coded symbol for the number of bits */ - emit_dc_symbol(entropy, tbl, nbits); - - /* Emit that number of bits of the value, if positive, */ - /* or the complement of its magnitude, if negative. */ - emit_bits_e(entropy, (unsigned int) temp2, nbits); - } - - cinfo->dest->next_output_byte = entropy->next_output_byte; - cinfo->dest->free_in_buffer = entropy->free_in_buffer; - - /* Update restart-interval state too */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) { - entropy->restarts_to_go = cinfo->restart_interval; - entropy->next_restart_num++; - entropy->next_restart_num &= 7; - } - entropy->restarts_to_go--; - } - - return TRUE; -} - - -/* - * MCU encoding for AC initial scan (either spectral selection, - * or first pass of successive approximation). - */ - -METHODDEF(boolean) -encode_mcu_AC_first (j_compress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - const int * natural_order; - JBLOCKROW block; - register int temp, temp2; - register int nbits; - register int r, k; - int Se, Al, max_coef_bits; - - entropy->next_output_byte = cinfo->dest->next_output_byte; - entropy->free_in_buffer = cinfo->dest->free_in_buffer; - - /* Emit restart marker if needed */ - if (cinfo->restart_interval) - if (entropy->restarts_to_go == 0) - emit_restart_e(entropy, entropy->next_restart_num); - - Se = cinfo->Se; - Al = cinfo->Al; - natural_order = cinfo->natural_order; - max_coef_bits = cinfo->data_precision + 2; - - /* Encode the MCU data block */ - block = MCU_data[0]; - - /* Encode the AC coefficients per section G.1.2.2, fig. G.3 */ - - r = 0; /* r = run length of zeros */ - - for (k = cinfo->Ss; k <= Se; k++) { - if ((temp = (*block)[natural_order[k]]) == 0) { - r++; - continue; - } - /* We must apply the point transform by Al. For AC coefficients this - * is an integer division with rounding towards 0. To do this portably - * in C, we shift after obtaining the absolute value; so the code is - * interwoven with finding the abs value (temp) and output bits (temp2). - */ - if (temp < 0) { - temp = -temp; /* temp is abs value of input */ - /* Apply the point transform, and watch out for case */ - /* that nonzero coef is zero after point transform. */ - if ((temp >>= Al) == 0) { - r++; - continue; - } - /* For a negative coef, want temp2 = bitwise complement of abs(coef) */ - temp2 = ~temp; - } else { - /* Apply the point transform, and watch out for case */ - /* that nonzero coef is zero after point transform. */ - if ((temp >>= Al) == 0) { - r++; - continue; - } - temp2 = temp; - } - - /* Emit any pending EOBRUN */ - if (entropy->EOBRUN > 0) - emit_eobrun(entropy); - /* if run length > 15, must emit special run-length-16 codes (0xF0) */ - while (r > 15) { - emit_ac_symbol(entropy, entropy->ac_tbl_no, 0xF0); - r -= 16; - } - - /* Find the number of bits needed for the magnitude of the coefficient */ - nbits = 0; - do nbits++; /* there must be at least one 1 bit */ - while ((temp >>= 1)); - /* Check for out-of-range coefficient values */ - if (nbits > max_coef_bits) - ERREXIT(cinfo, JERR_BAD_DCT_COEF); - - /* Count/emit Huffman symbol for run length / number of bits */ - emit_ac_symbol(entropy, entropy->ac_tbl_no, (r << 4) + nbits); - - /* Emit that number of bits of the value, if positive, */ - /* or the complement of its magnitude, if negative. */ - emit_bits_e(entropy, (unsigned int) temp2, nbits); - - r = 0; /* reset zero run length */ - } - - if (r > 0) { /* If there are trailing zeroes, */ - entropy->EOBRUN++; /* count an EOB */ - if (entropy->EOBRUN == 0x7FFF) - emit_eobrun(entropy); /* force it out to avoid overflow */ - } - - cinfo->dest->next_output_byte = entropy->next_output_byte; - cinfo->dest->free_in_buffer = entropy->free_in_buffer; - - /* Update restart-interval state too */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) { - entropy->restarts_to_go = cinfo->restart_interval; - entropy->next_restart_num++; - entropy->next_restart_num &= 7; - } - entropy->restarts_to_go--; - } - - return TRUE; -} - - -/* - * MCU encoding for DC successive approximation refinement scan. - * Note: we assume such scans can be multi-component, - * although the spec is not very clear on the point. - */ - -METHODDEF(boolean) -encode_mcu_DC_refine (j_compress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - int Al, blkn; - - entropy->next_output_byte = cinfo->dest->next_output_byte; - entropy->free_in_buffer = cinfo->dest->free_in_buffer; - - /* Emit restart marker if needed */ - if (cinfo->restart_interval) - if (entropy->restarts_to_go == 0) - emit_restart_e(entropy, entropy->next_restart_num); - - Al = cinfo->Al; - - /* Encode the MCU data blocks */ - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - /* We simply emit the Al'th bit of the DC coefficient value. */ - emit_bits_e(entropy, (unsigned int) (MCU_data[blkn][0][0] >> Al), 1); - } - - cinfo->dest->next_output_byte = entropy->next_output_byte; - cinfo->dest->free_in_buffer = entropy->free_in_buffer; - - /* Update restart-interval state too */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) { - entropy->restarts_to_go = cinfo->restart_interval; - entropy->next_restart_num++; - entropy->next_restart_num &= 7; - } - entropy->restarts_to_go--; - } - - return TRUE; -} - - -/* - * MCU encoding for AC successive approximation refinement scan. - */ - -METHODDEF(boolean) -encode_mcu_AC_refine (j_compress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - const int * natural_order; - JBLOCKROW block; - register int temp; - register int r, k; - int Se, Al; - int EOB; - char *BR_buffer; - unsigned int BR; - int absvalues[DCTSIZE2]; - - entropy->next_output_byte = cinfo->dest->next_output_byte; - entropy->free_in_buffer = cinfo->dest->free_in_buffer; - - /* Emit restart marker if needed */ - if (cinfo->restart_interval) - if (entropy->restarts_to_go == 0) - emit_restart_e(entropy, entropy->next_restart_num); - - Se = cinfo->Se; - Al = cinfo->Al; - natural_order = cinfo->natural_order; - - /* Encode the MCU data block */ - block = MCU_data[0]; - - /* It is convenient to make a pre-pass to determine the transformed - * coefficients' absolute values and the EOB position. - */ - EOB = 0; - for (k = cinfo->Ss; k <= Se; k++) { - temp = (*block)[natural_order[k]]; - /* We must apply the point transform by Al. For AC coefficients this - * is an integer division with rounding towards 0. To do this portably - * in C, we shift after obtaining the absolute value. - */ - if (temp < 0) - temp = -temp; /* temp is abs value of input */ - temp >>= Al; /* apply the point transform */ - absvalues[k] = temp; /* save abs value for main pass */ - if (temp == 1) - EOB = k; /* EOB = index of last newly-nonzero coef */ - } - - /* Encode the AC coefficients per section G.1.2.3, fig. G.7 */ - - r = 0; /* r = run length of zeros */ - BR = 0; /* BR = count of buffered bits added now */ - BR_buffer = entropy->bit_buffer + entropy->BE; /* Append bits to buffer */ - - for (k = cinfo->Ss; k <= Se; k++) { - if ((temp = absvalues[k]) == 0) { - r++; - continue; - } - - /* Emit any required ZRLs, but not if they can be folded into EOB */ - while (r > 15 && k <= EOB) { - /* emit any pending EOBRUN and the BE correction bits */ - emit_eobrun(entropy); - /* Emit ZRL */ - emit_ac_symbol(entropy, entropy->ac_tbl_no, 0xF0); - r -= 16; - /* Emit buffered correction bits that must be associated with ZRL */ - emit_buffered_bits(entropy, BR_buffer, BR); - BR_buffer = entropy->bit_buffer; /* BE bits are gone now */ - BR = 0; - } - - /* If the coef was previously nonzero, it only needs a correction bit. - * NOTE: a straight translation of the spec's figure G.7 would suggest - * that we also need to test r > 15. But if r > 15, we can only get here - * if k > EOB, which implies that this coefficient is not 1. - */ - if (temp > 1) { - /* The correction bit is the next bit of the absolute value. */ - BR_buffer[BR++] = (char) (temp & 1); - continue; - } - - /* Emit any pending EOBRUN and the BE correction bits */ - emit_eobrun(entropy); - - /* Count/emit Huffman symbol for run length / number of bits */ - emit_ac_symbol(entropy, entropy->ac_tbl_no, (r << 4) + 1); - - /* Emit output bit for newly-nonzero coef */ - temp = ((*block)[natural_order[k]] < 0) ? 0 : 1; - emit_bits_e(entropy, (unsigned int) temp, 1); - - /* Emit buffered correction bits that must be associated with this code */ - emit_buffered_bits(entropy, BR_buffer, BR); - BR_buffer = entropy->bit_buffer; /* BE bits are gone now */ - BR = 0; - r = 0; /* reset zero run length */ - } - - if (r > 0 || BR > 0) { /* If there are trailing zeroes, */ - entropy->EOBRUN++; /* count an EOB */ - entropy->BE += BR; /* concat my correction bits to older ones */ - /* We force out the EOB if we risk either: - * 1. overflow of the EOB counter; - * 2. overflow of the correction bit buffer during the next MCU. - */ - if (entropy->EOBRUN == 0x7FFF || entropy->BE > (MAX_CORR_BITS-DCTSIZE2+1)) - emit_eobrun(entropy); - } - - cinfo->dest->next_output_byte = entropy->next_output_byte; - cinfo->dest->free_in_buffer = entropy->free_in_buffer; - - /* Update restart-interval state too */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) { - entropy->restarts_to_go = cinfo->restart_interval; - entropy->next_restart_num++; - entropy->next_restart_num &= 7; - } - entropy->restarts_to_go--; - } - - return TRUE; -} - - -/* Encode a single block's worth of coefficients */ - -LOCAL(boolean) -encode_one_block (working_state * state, JCOEFPTR block, int last_dc_val, - c_derived_tbl *dctbl, c_derived_tbl *actbl) -{ - register int temp, temp2; - register int nbits; - register int r, k; - int Se = state->cinfo->lim_Se; - int max_coef_bits = state->cinfo->data_precision + 3; - const int * natural_order = state->cinfo->natural_order; - - /* Encode the DC coefficient difference per section F.1.2.1 */ - - if ((temp = block[0] - last_dc_val) == 0) { - /* Emit the Huffman-coded symbol for the number of bits */ - if (! emit_bits_s(state, dctbl->ehufco[0], dctbl->ehufsi[0])) - return FALSE; - } else { - if ((temp2 = temp) < 0) { - temp = -temp; /* temp is abs value of input */ - /* For a negative input, want temp2 = bitwise complement of abs(input) */ - /* This code assumes we are on a two's complement machine */ - temp2--; - } - - /* Find the number of bits needed for the magnitude of the coefficient */ - nbits = 0; - do nbits++; /* there must be at least one 1 bit */ - while ((temp >>= 1)); - /* Check for out-of-range coefficient values. - * Since we're encoding a difference, the range limit is twice as much. - */ - if (nbits > max_coef_bits) - ERREXIT(state->cinfo, JERR_BAD_DCT_COEF); - - /* Emit the Huffman-coded symbol for the number of bits */ - if (! emit_bits_s(state, dctbl->ehufco[nbits], dctbl->ehufsi[nbits])) - return FALSE; - - /* Emit that number of bits of the value, if positive, */ - /* or the complement of its magnitude, if negative. */ - if (! emit_bits_s(state, (unsigned int) temp2, nbits)) - return FALSE; - } - - /* Encode the AC coefficients per section F.1.2.2 */ - - r = 0; /* r = run length of zeros */ - - for (k = 1; k <= Se; k++) { - if ((temp = block[natural_order[k]]) == 0) { - r++; - continue; - } - - /* if run length > 15, must emit special run-length-16 codes (0xF0) */ - while (r > 15) { - if (! emit_bits_s(state, actbl->ehufco[0xF0], actbl->ehufsi[0xF0])) - return FALSE; - r -= 16; - } - - if ((temp2 = temp) < 0) { - temp = -temp; /* temp is abs value of input */ - /* For a negative coef, want temp2 = bitwise complement of abs(coef) */ - /* This code assumes we are on a two's complement machine */ - temp2--; - } - - /* Find the number of bits needed for the magnitude of the coefficient */ - nbits = 0; - do nbits++; /* there must be at least one 1 bit */ - while ((temp >>= 1)); - /* Check for out-of-range coefficient values. - * Use ">=" instead of ">" so can use the - * same one larger limit from DC check here. - */ - if (nbits >= max_coef_bits) - ERREXIT(state->cinfo, JERR_BAD_DCT_COEF); - - /* Emit Huffman symbol for run length / number of bits */ - temp = (r << 4) + nbits; - if (! emit_bits_s(state, actbl->ehufco[temp], actbl->ehufsi[temp])) - return FALSE; - - /* Emit that number of bits of the value, if positive, */ - /* or the complement of its magnitude, if negative. */ - if (! emit_bits_s(state, (unsigned int) temp2, nbits)) - return FALSE; - - r = 0; /* reset zero run length */ - } - - /* If the last coef(s) were zero, emit an end-of-block code */ - if (r > 0) - if (! emit_bits_s(state, actbl->ehufco[0], actbl->ehufsi[0])) - return FALSE; - - return TRUE; -} - - -/* - * Encode and output one MCU's worth of Huffman-compressed coefficients. - */ - -METHODDEF(boolean) -encode_mcu_huff (j_compress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - working_state state; - int blkn, ci; - jpeg_component_info * compptr; - - /* Load up working state */ - state.next_output_byte = cinfo->dest->next_output_byte; - state.free_in_buffer = cinfo->dest->free_in_buffer; - ASSIGN_STATE(state.cur, entropy->saved); - state.cinfo = cinfo; - - /* Emit restart marker if needed */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) - if (! emit_restart_s(&state, entropy->next_restart_num)) - return FALSE; - } - - /* Encode the MCU data blocks */ - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - ci = cinfo->MCU_membership[blkn]; - compptr = cinfo->cur_comp_info[ci]; - if (! encode_one_block(&state, - MCU_data[blkn][0], state.cur.last_dc_val[ci], - entropy->dc_derived_tbls[compptr->dc_tbl_no], - entropy->ac_derived_tbls[compptr->ac_tbl_no])) - return FALSE; - /* Update last_dc_val */ - state.cur.last_dc_val[ci] = MCU_data[blkn][0][0]; - } - - /* Completed MCU, so update state */ - cinfo->dest->next_output_byte = state.next_output_byte; - cinfo->dest->free_in_buffer = state.free_in_buffer; - ASSIGN_STATE(entropy->saved, state.cur); - - /* Update restart-interval state too */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) { - entropy->restarts_to_go = cinfo->restart_interval; - entropy->next_restart_num++; - entropy->next_restart_num &= 7; - } - entropy->restarts_to_go--; - } - - return TRUE; -} - - -/* - * Finish up at the end of a Huffman-compressed scan. - */ - -METHODDEF(void) -finish_pass_huff (j_compress_ptr cinfo) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - working_state state; - - if (cinfo->progressive_mode) { - entropy->next_output_byte = cinfo->dest->next_output_byte; - entropy->free_in_buffer = cinfo->dest->free_in_buffer; - - /* Flush out any buffered data */ - emit_eobrun(entropy); - flush_bits_e(entropy); - - cinfo->dest->next_output_byte = entropy->next_output_byte; - cinfo->dest->free_in_buffer = entropy->free_in_buffer; - } else { - /* Load up working state ... flush_bits needs it */ - state.next_output_byte = cinfo->dest->next_output_byte; - state.free_in_buffer = cinfo->dest->free_in_buffer; - ASSIGN_STATE(state.cur, entropy->saved); - state.cinfo = cinfo; - - /* Flush out the last data */ - if (! flush_bits_s(&state)) - ERREXIT(cinfo, JERR_CANT_SUSPEND); - - /* Update state */ - cinfo->dest->next_output_byte = state.next_output_byte; - cinfo->dest->free_in_buffer = state.free_in_buffer; - ASSIGN_STATE(entropy->saved, state.cur); - } -} - - -/* - * Huffman coding optimization. - * - * We first scan the supplied data and count the number of uses of each symbol - * that is to be Huffman-coded. (This process MUST agree with the code above.) - * Then we build a Huffman coding tree for the observed counts. - * Symbols which are not needed at all for the particular image are not - * assigned any code, which saves space in the DHT marker as well as in - * the compressed data. - */ - - -/* Process a single block's worth of coefficients */ - -LOCAL(void) -htest_one_block (j_compress_ptr cinfo, JCOEFPTR block, int last_dc_val, - long dc_counts[], long ac_counts[]) -{ - register int temp; - register int nbits; - register int r, k; - int Se = cinfo->lim_Se; - int max_coef_bits = cinfo->data_precision + 3; - const int * natural_order = cinfo->natural_order; - - /* Encode the DC coefficient difference per section F.1.2.1 */ - - if ((temp = block[0] - last_dc_val) == 0) { - /* Count the Huffman symbol for the number of bits */ - dc_counts[0]++; - } else { - if (temp < 0) - temp = -temp; /* temp is abs value of input */ - - /* Find the number of bits needed for the magnitude of the coefficient */ - nbits = 0; - do nbits++; /* there must be at least one 1 bit */ - while ((temp >>= 1)); - /* Check for out-of-range coefficient values. - * Since we're encoding a difference, the range limit is twice as much. - */ - if (nbits > max_coef_bits) - ERREXIT(cinfo, JERR_BAD_DCT_COEF); - - /* Count the Huffman symbol for the number of bits */ - dc_counts[nbits]++; - } - - /* Encode the AC coefficients per section F.1.2.2 */ - - r = 0; /* r = run length of zeros */ - - for (k = 1; k <= Se; k++) { - if ((temp = block[natural_order[k]]) == 0) { - r++; - continue; - } - - /* if run length > 15, must emit special run-length-16 codes (0xF0) */ - while (r > 15) { - ac_counts[0xF0]++; - r -= 16; - } - - if (temp < 0) - temp = -temp; /* temp is abs value of input */ - - /* Find the number of bits needed for the magnitude of the coefficient */ - nbits = 0; - do nbits++; /* there must be at least one 1 bit */ - while ((temp >>= 1)); - /* Check for out-of-range coefficient values. - * Use ">=" instead of ">" so can use the - * same one larger limit from DC check here. - */ - if (nbits >= max_coef_bits) - ERREXIT(cinfo, JERR_BAD_DCT_COEF); - - /* Count Huffman symbol for run length / number of bits */ - ac_counts[(r << 4) + nbits]++; - - r = 0; /* reset zero run length */ - } - - /* If the last coef(s) were zero, emit an end-of-block code */ - if (r > 0) - ac_counts[0]++; -} - - -/* - * Trial-encode one MCU's worth of Huffman-compressed coefficients. - * No data is actually output, so no suspension return is possible. - */ - -METHODDEF(boolean) -encode_mcu_gather (j_compress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - int blkn, ci; - jpeg_component_info * compptr; - - /* Take care of restart intervals if needed */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) { - /* Re-initialize DC predictions to 0 */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) - entropy->saved.last_dc_val[ci] = 0; - /* Update restart state */ - entropy->restarts_to_go = cinfo->restart_interval; - } - entropy->restarts_to_go--; - } - - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - ci = cinfo->MCU_membership[blkn]; - compptr = cinfo->cur_comp_info[ci]; - htest_one_block(cinfo, MCU_data[blkn][0], entropy->saved.last_dc_val[ci], - entropy->dc_count_ptrs[compptr->dc_tbl_no], - entropy->ac_count_ptrs[compptr->ac_tbl_no]); - entropy->saved.last_dc_val[ci] = MCU_data[blkn][0][0]; - } - - return TRUE; -} - - -/* - * Generate the best Huffman code table for the given counts, fill htbl. - * - * The JPEG standard requires that no symbol be assigned a codeword of all - * one bits (so that padding bits added at the end of a compressed segment - * can't look like a valid code). Because of the canonical ordering of - * codewords, this just means that there must be an unused slot in the - * longest codeword length category. Section K.2 of the JPEG spec suggests - * reserving such a slot by pretending that symbol 256 is a valid symbol - * with count 1. In theory that's not optimal; giving it count zero but - * including it in the symbol set anyway should give a better Huffman code. - * But the theoretically better code actually seems to come out worse in - * practice, because it produces more all-ones bytes (which incur stuffed - * zero bytes in the final file). In any case the difference is tiny. - * - * The JPEG standard requires Huffman codes to be no more than 16 bits long. - * If some symbols have a very small but nonzero probability, the Huffman tree - * must be adjusted to meet the code length restriction. We currently use - * the adjustment method suggested in JPEG section K.2. This method is *not* - * optimal; it may not choose the best possible limited-length code. But - * typically only very-low-frequency symbols will be given less-than-optimal - * lengths, so the code is almost optimal. Experimental comparisons against - * an optimal limited-length-code algorithm indicate that the difference is - * microscopic --- usually less than a hundredth of a percent of total size. - * So the extra complexity of an optimal algorithm doesn't seem worthwhile. - */ - -LOCAL(void) -jpeg_gen_optimal_table (j_compress_ptr cinfo, JHUFF_TBL * htbl, long freq[]) -{ -#define MAX_CLEN 32 /* assumed maximum initial code length */ - UINT8 bits[MAX_CLEN+1]; /* bits[k] = # of symbols with code length k */ - int codesize[257]; /* codesize[k] = code length of symbol k */ - int others[257]; /* next symbol in current branch of tree */ - int c1, c2, i, j; - UINT8 *p; - long v; - - freq[256] = 1; /* make sure 256 has a nonzero count */ - /* Including the pseudo-symbol 256 in the Huffman procedure guarantees - * that no real symbol is given code-value of all ones, because 256 - * will be placed last in the largest codeword category. - * In the symbol list build procedure this element serves as sentinel - * for the zero run loop. - */ - -#ifndef DONT_USE_FANCY_HUFF_OPT - - /* Build list of symbols sorted in order of descending frequency */ - /* This approach has several benefits (thank to John Korejwa for the idea): - * 1. - * If a codelength category is split during the length limiting procedure - * below, the feature that more frequent symbols are assigned shorter - * codewords remains valid for the adjusted code. - * 2. - * To reduce consecutive ones in a Huffman data stream (thus reducing the - * number of stuff bytes in JPEG) it is preferable to follow 0 branches - * (and avoid 1 branches) as much as possible. This is easily done by - * assigning symbols to leaves of the Huffman tree in order of decreasing - * frequency, with no secondary sort based on codelengths. - * 3. - * The symbol list can be built independently from the assignment of code - * lengths by the Huffman procedure below. - * Note: The symbol list build procedure must be performed first, because - * the Huffman procedure assigning the codelengths clobbers the frequency - * counts! - */ - - /* Here we use the others array as a linked list of nonzero frequencies - * to be sorted. Already sorted elements are removed from the list. - */ - - /* Building list */ - - /* This item does not correspond to a valid symbol frequency and is used - * as starting index. - */ - j = 256; - - for (i = 0;; i++) { - if (freq[i] == 0) /* skip zero frequencies */ - continue; - if (i > 255) - break; - others[j] = i; /* this symbol value */ - j = i; /* previous symbol value */ - } - others[j] = -1; /* mark end of list */ - - /* Sorting list */ - - p = htbl->huffval; - while ((c1 = others[256]) >= 0) { - v = freq[c1]; - i = c1; /* first symbol value */ - j = 256; /* pseudo symbol value for starting index */ - while ((c2 = others[c1]) >= 0) { - if (freq[c2] > v) { - v = freq[c2]; - i = c2; /* this symbol value */ - j = c1; /* previous symbol value */ - } - c1 = c2; - } - others[j] = others[i]; /* remove this symbol i from list */ - *p++ = (UINT8) i; - } - -#endif /* DONT_USE_FANCY_HUFF_OPT */ - - /* This algorithm is explained in section K.2 of the JPEG standard */ - - MEMZERO(bits, SIZEOF(bits)); - MEMZERO(codesize, SIZEOF(codesize)); - for (i = 0; i < 257; i++) - others[i] = -1; /* init links to empty */ - - /* Huffman's basic algorithm to assign optimal code lengths to symbols */ - - for (;;) { - /* Find the smallest nonzero frequency, set c1 = its symbol */ - /* In case of ties, take the larger symbol number */ - c1 = -1; - v = 1000000000L; - for (i = 0; i <= 256; i++) { - if (freq[i] && freq[i] <= v) { - v = freq[i]; - c1 = i; - } - } - - /* Find the next smallest nonzero frequency, set c2 = its symbol */ - /* In case of ties, take the larger symbol number */ - c2 = -1; - v = 1000000000L; - for (i = 0; i <= 256; i++) { - if (freq[i] && freq[i] <= v && i != c1) { - v = freq[i]; - c2 = i; - } - } - - /* Done if we've merged everything into one frequency */ - if (c2 < 0) - break; - - /* Else merge the two counts/trees */ - freq[c1] += freq[c2]; - freq[c2] = 0; - - /* Increment the codesize of everything in c1's tree branch */ - codesize[c1]++; - while (others[c1] >= 0) { - c1 = others[c1]; - codesize[c1]++; - } - - others[c1] = c2; /* chain c2 onto c1's tree branch */ - - /* Increment the codesize of everything in c2's tree branch */ - codesize[c2]++; - while (others[c2] >= 0) { - c2 = others[c2]; - codesize[c2]++; - } - } - - /* Now count the number of symbols of each code length */ - for (i = 0; i <= 256; i++) { - if (codesize[i]) { - /* The JPEG standard seems to think that this can't happen, */ - /* but I'm paranoid... */ - if (codesize[i] > MAX_CLEN) - ERREXIT(cinfo, JERR_HUFF_CLEN_OUTOFBOUNDS); - - bits[codesize[i]]++; - } - } - - /* JPEG doesn't allow symbols with code lengths over 16 bits, so if the pure - * Huffman procedure assigned any such lengths, we must adjust the coding. - * Here is what the JPEG spec says about how this next bit works: - * Since symbols are paired for the longest Huffman code, the symbols are - * removed from this length category two at a time. The prefix for the pair - * (which is one bit shorter) is allocated to one of the pair; then, - * skipping the BITS entry for that prefix length, a code word from the next - * shortest nonzero BITS entry is converted into a prefix for two code words - * one bit longer. - */ - - for (i = MAX_CLEN; i > 16; i--) { - while (bits[i] > 0) { - j = i - 2; /* find length of new prefix to be used */ - while (bits[j] == 0) { - if (j == 0) - ERREXIT(cinfo, JERR_HUFF_CLEN_OUTOFBOUNDS); - j--; - } - - bits[i] -= 2; /* remove two symbols */ - bits[i-1]++; /* one goes in this length */ - bits[j+1] += 2; /* two new symbols in this length */ - bits[j]--; /* symbol of this length is now a prefix */ - } - } - - /* Remove the count for the pseudo-symbol 256 from the largest codelength */ - while (bits[i] == 0) /* find largest codelength still in use */ - i--; - bits[i]--; - - /* Return final symbol counts (only for lengths 0..16) */ - MEMCOPY(htbl->bits, bits, SIZEOF(htbl->bits)); - -#ifdef DONT_USE_FANCY_HUFF_OPT - - /* Return a list of the symbols sorted by code length */ - /* Note: Due to the codelength changes made above, it can happen - * that more frequent symbols are assigned longer codewords. - */ - p = htbl->huffval; - for (i = 1; i <= MAX_CLEN; i++) { - for (j = 0; j <= 255; j++) { - if (codesize[j] == i) { - *p++ = (UINT8) j; - } - } - } - -#endif /* DONT_USE_FANCY_HUFF_OPT */ - - /* Set sent_table FALSE so updated table will be written to JPEG file. */ - htbl->sent_table = FALSE; -} - - -/* - * Finish up a statistics-gathering pass and create the new Huffman tables. - */ - -METHODDEF(void) -finish_pass_gather (j_compress_ptr cinfo) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - int ci, tbl; - jpeg_component_info * compptr; - JHUFF_TBL **htblptr; - boolean did_dc[NUM_HUFF_TBLS]; - boolean did_ac[NUM_HUFF_TBLS]; - - if (cinfo->progressive_mode) - /* Flush out buffered data (all we care about is counting the EOB symbol) */ - emit_eobrun(entropy); - - /* It's important not to apply jpeg_gen_optimal_table more than once - * per table, because it clobbers the input frequency counts! - */ - MEMZERO(did_dc, SIZEOF(did_dc)); - MEMZERO(did_ac, SIZEOF(did_ac)); - - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - /* DC needs no table for refinement scan */ - if (cinfo->Ss == 0 && cinfo->Ah == 0) { - tbl = compptr->dc_tbl_no; - if (! did_dc[tbl]) { - htblptr = & cinfo->dc_huff_tbl_ptrs[tbl]; - if (*htblptr == NULL) - *htblptr = jpeg_alloc_huff_table((j_common_ptr) cinfo); - jpeg_gen_optimal_table(cinfo, *htblptr, entropy->dc_count_ptrs[tbl]); - did_dc[tbl] = TRUE; - } - } - /* AC needs no table when not present */ - if (cinfo->Se) { - tbl = compptr->ac_tbl_no; - if (! did_ac[tbl]) { - htblptr = & cinfo->ac_huff_tbl_ptrs[tbl]; - if (*htblptr == NULL) - *htblptr = jpeg_alloc_huff_table((j_common_ptr) cinfo); - jpeg_gen_optimal_table(cinfo, *htblptr, entropy->ac_count_ptrs[tbl]); - did_ac[tbl] = TRUE; - } - } - } -} - - -/* - * Initialize for a Huffman-compressed scan. - * If gather_statistics is TRUE, we do not output anything during the scan, - * just count the Huffman symbols used and generate Huffman code tables. - */ - -METHODDEF(void) -start_pass_huff (j_compress_ptr cinfo, boolean gather_statistics) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - int ci, tbl; - jpeg_component_info * compptr; - - if (gather_statistics) - entropy->pub.finish_pass = finish_pass_gather; - else - entropy->pub.finish_pass = finish_pass_huff; - - if (cinfo->progressive_mode) { - entropy->cinfo = cinfo; - entropy->gather_statistics = gather_statistics; - - /* We assume jcmaster.c already validated the scan parameters. */ - - /* Select execution routine */ - if (cinfo->Ah == 0) { - if (cinfo->Ss == 0) - entropy->pub.encode_mcu = encode_mcu_DC_first; - else - entropy->pub.encode_mcu = encode_mcu_AC_first; - } else { - if (cinfo->Ss == 0) - entropy->pub.encode_mcu = encode_mcu_DC_refine; - else { - entropy->pub.encode_mcu = encode_mcu_AC_refine; - /* AC refinement needs a correction bit buffer */ - if (entropy->bit_buffer == NULL) - entropy->bit_buffer = (char *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, MAX_CORR_BITS * SIZEOF(char)); - } - } - - /* Initialize AC stuff */ - entropy->ac_tbl_no = cinfo->cur_comp_info[0]->ac_tbl_no; - entropy->EOBRUN = 0; - entropy->BE = 0; - } else { - if (gather_statistics) - entropy->pub.encode_mcu = encode_mcu_gather; - else - entropy->pub.encode_mcu = encode_mcu_huff; - } - - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - /* DC needs no table for refinement scan */ - if (cinfo->Ss == 0 && cinfo->Ah == 0) { - tbl = compptr->dc_tbl_no; - if (gather_statistics) { - /* Check for invalid table index */ - /* (make_c_derived_tbl does this in the other path) */ - if (tbl < 0 || tbl >= NUM_HUFF_TBLS) - ERREXIT1(cinfo, JERR_NO_HUFF_TABLE, tbl); - /* Allocate and zero the statistics tables */ - /* Note that jpeg_gen_optimal_table expects 257 entries in each table! */ - if (entropy->dc_count_ptrs[tbl] == NULL) - entropy->dc_count_ptrs[tbl] = (long *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, 257 * SIZEOF(long)); - MEMZERO(entropy->dc_count_ptrs[tbl], 257 * SIZEOF(long)); - } else { - /* Compute derived values for Huffman tables */ - /* We may do this more than once for a table, but it's not expensive */ - jpeg_make_c_derived_tbl(cinfo, TRUE, tbl, - & entropy->dc_derived_tbls[tbl]); - } - /* Initialize DC predictions to 0 */ - entropy->saved.last_dc_val[ci] = 0; - } - /* AC needs no table when not present */ - if (cinfo->Se) { - tbl = compptr->ac_tbl_no; - if (gather_statistics) { - if (tbl < 0 || tbl >= NUM_HUFF_TBLS) - ERREXIT1(cinfo, JERR_NO_HUFF_TABLE, tbl); - if (entropy->ac_count_ptrs[tbl] == NULL) - entropy->ac_count_ptrs[tbl] = (long *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, 257 * SIZEOF(long)); - MEMZERO(entropy->ac_count_ptrs[tbl], 257 * SIZEOF(long)); - } else { - jpeg_make_c_derived_tbl(cinfo, FALSE, tbl, - & entropy->ac_derived_tbls[tbl]); - } - } - } - - /* Initialize bit buffer to empty */ - entropy->saved.put_buffer = 0; - entropy->saved.put_bits = 0; - - /* Initialize restart stuff */ - entropy->restarts_to_go = cinfo->restart_interval; - entropy->next_restart_num = 0; -} - - -/* - * Module initialization routine for Huffman entropy encoding. - */ - -GLOBAL(void) -jinit_huff_encoder (j_compress_ptr cinfo) -{ - huff_entropy_ptr entropy; - int i; - - entropy = (huff_entropy_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(huff_entropy_encoder)); - cinfo->entropy = &entropy->pub; - entropy->pub.start_pass = start_pass_huff; - - /* Mark tables unallocated */ - for (i = 0; i < NUM_HUFF_TBLS; i++) { - entropy->dc_derived_tbls[i] = entropy->ac_derived_tbls[i] = NULL; - entropy->dc_count_ptrs[i] = entropy->ac_count_ptrs[i] = NULL; - } - - if (cinfo->progressive_mode) - entropy->bit_buffer = NULL; /* needed only in AC refinement scan */ -} diff --git a/dep/libjpeg/src/jcinit.c b/dep/libjpeg/src/jcinit.c deleted file mode 100644 index 2aea7ca2a..000000000 --- a/dep/libjpeg/src/jcinit.c +++ /dev/null @@ -1,249 +0,0 @@ -/* - * jcinit.c - * - * Copyright (C) 1991-1997, Thomas G. Lane. - * Modified 2003-2017 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains initialization logic for the JPEG compressor. - * This routine is in charge of selecting the modules to be executed and - * making an initialization call to each one. - * - * Logically, this code belongs in jcmaster.c. It's split out because - * linking this routine implies linking the entire compression library. - * For a transcoding-only application, we want to be able to use jcmaster.c - * without linking in the whole library. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* - * Compute JPEG image dimensions and related values. - * NOTE: this is exported for possible use by application. - * Hence it mustn't do anything that can't be done twice. - */ - -GLOBAL(void) -jpeg_calc_jpeg_dimensions (j_compress_ptr cinfo) -/* Do computations that are needed before master selection phase */ -{ - /* Sanity check on input image dimensions to prevent overflow in - * following calculations. - * We do check jpeg_width and jpeg_height in initial_setup in jcmaster.c, - * but image_width and image_height can come from arbitrary data, - * and we need some space for multiplication by block_size. - */ - if (((long) cinfo->image_width >> 24) || ((long) cinfo->image_height >> 24)) - ERREXIT1(cinfo, JERR_IMAGE_TOO_BIG, (unsigned int) JPEG_MAX_DIMENSION); - -#ifdef DCT_SCALING_SUPPORTED - - /* Compute actual JPEG image dimensions and DCT scaling choices. */ - if (cinfo->scale_num >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/1 scaling */ - cinfo->jpeg_width = cinfo->image_width * cinfo->block_size; - cinfo->jpeg_height = cinfo->image_height * cinfo->block_size; - cinfo->min_DCT_h_scaled_size = 1; - cinfo->min_DCT_v_scaled_size = 1; - } else if (cinfo->scale_num * 2 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/2 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 2L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 2L); - cinfo->min_DCT_h_scaled_size = 2; - cinfo->min_DCT_v_scaled_size = 2; - } else if (cinfo->scale_num * 3 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/3 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 3L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 3L); - cinfo->min_DCT_h_scaled_size = 3; - cinfo->min_DCT_v_scaled_size = 3; - } else if (cinfo->scale_num * 4 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/4 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 4L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 4L); - cinfo->min_DCT_h_scaled_size = 4; - cinfo->min_DCT_v_scaled_size = 4; - } else if (cinfo->scale_num * 5 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/5 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 5L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 5L); - cinfo->min_DCT_h_scaled_size = 5; - cinfo->min_DCT_v_scaled_size = 5; - } else if (cinfo->scale_num * 6 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/6 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 6L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 6L); - cinfo->min_DCT_h_scaled_size = 6; - cinfo->min_DCT_v_scaled_size = 6; - } else if (cinfo->scale_num * 7 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/7 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 7L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 7L); - cinfo->min_DCT_h_scaled_size = 7; - cinfo->min_DCT_v_scaled_size = 7; - } else if (cinfo->scale_num * 8 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/8 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 8L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 8L); - cinfo->min_DCT_h_scaled_size = 8; - cinfo->min_DCT_v_scaled_size = 8; - } else if (cinfo->scale_num * 9 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/9 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 9L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 9L); - cinfo->min_DCT_h_scaled_size = 9; - cinfo->min_DCT_v_scaled_size = 9; - } else if (cinfo->scale_num * 10 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/10 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 10L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 10L); - cinfo->min_DCT_h_scaled_size = 10; - cinfo->min_DCT_v_scaled_size = 10; - } else if (cinfo->scale_num * 11 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/11 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 11L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 11L); - cinfo->min_DCT_h_scaled_size = 11; - cinfo->min_DCT_v_scaled_size = 11; - } else if (cinfo->scale_num * 12 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/12 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 12L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 12L); - cinfo->min_DCT_h_scaled_size = 12; - cinfo->min_DCT_v_scaled_size = 12; - } else if (cinfo->scale_num * 13 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/13 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 13L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 13L); - cinfo->min_DCT_h_scaled_size = 13; - cinfo->min_DCT_v_scaled_size = 13; - } else if (cinfo->scale_num * 14 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/14 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 14L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 14L); - cinfo->min_DCT_h_scaled_size = 14; - cinfo->min_DCT_v_scaled_size = 14; - } else if (cinfo->scale_num * 15 >= cinfo->scale_denom * cinfo->block_size) { - /* Provide block_size/15 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 15L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 15L); - cinfo->min_DCT_h_scaled_size = 15; - cinfo->min_DCT_v_scaled_size = 15; - } else { - /* Provide block_size/16 scaling */ - cinfo->jpeg_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * cinfo->block_size, 16L); - cinfo->jpeg_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * cinfo->block_size, 16L); - cinfo->min_DCT_h_scaled_size = 16; - cinfo->min_DCT_v_scaled_size = 16; - } - -#else /* !DCT_SCALING_SUPPORTED */ - - /* Hardwire it to "no scaling" */ - cinfo->jpeg_width = cinfo->image_width; - cinfo->jpeg_height = cinfo->image_height; - cinfo->min_DCT_h_scaled_size = DCTSIZE; - cinfo->min_DCT_v_scaled_size = DCTSIZE; - -#endif /* DCT_SCALING_SUPPORTED */ -} - - -/* - * Master selection of compression modules. - * This is done once at the start of processing an image. We determine - * which modules will be used and give them appropriate initialization calls. - */ - -GLOBAL(void) -jinit_compress_master (j_compress_ptr cinfo) -{ - long samplesperrow; - JDIMENSION jd_samplesperrow; - - /* For now, precision must match compiled-in value... */ - if (cinfo->data_precision != BITS_IN_JSAMPLE) - ERREXIT1(cinfo, JERR_BAD_PRECISION, cinfo->data_precision); - - /* Sanity check on input image dimensions */ - if (cinfo->image_height <= 0 || cinfo->image_width <= 0 || - cinfo->input_components <= 0) - ERREXIT(cinfo, JERR_EMPTY_IMAGE); - - /* Width of an input scanline must be representable as JDIMENSION. */ - samplesperrow = (long) cinfo->image_width * (long) cinfo->input_components; - jd_samplesperrow = (JDIMENSION) samplesperrow; - if ((long) jd_samplesperrow != samplesperrow) - ERREXIT(cinfo, JERR_WIDTH_OVERFLOW); - - /* Compute JPEG image dimensions and related values. */ - jpeg_calc_jpeg_dimensions(cinfo); - - /* Initialize master control (includes parameter checking/processing) */ - jinit_c_master_control(cinfo, FALSE /* full compression */); - - /* Preprocessing */ - if (! cinfo->raw_data_in) { - jinit_color_converter(cinfo); - jinit_downsampler(cinfo); - jinit_c_prep_controller(cinfo, FALSE /* never need full buffer here */); - } - /* Forward DCT */ - jinit_forward_dct(cinfo); - /* Entropy encoding: either Huffman or arithmetic coding. */ - if (cinfo->arith_code) - jinit_arith_encoder(cinfo); - else { - jinit_huff_encoder(cinfo); - } - - /* Need a full-image coefficient buffer in any multi-pass mode. */ - jinit_c_coef_controller(cinfo, - (boolean) (cinfo->num_scans > 1 || cinfo->optimize_coding)); - jinit_c_main_controller(cinfo, FALSE /* never need full buffer here */); - - jinit_marker_writer(cinfo); - - /* We can now tell the memory manager to allocate virtual arrays. */ - (*cinfo->mem->realize_virt_arrays) ((j_common_ptr) cinfo); - - /* Write the datastream header (SOI) immediately. - * Frame and scan headers are postponed till later. - * This lets application insert special markers after the SOI. - */ - (*cinfo->marker->write_file_header) (cinfo); -} diff --git a/dep/libjpeg/src/jcmainct.c b/dep/libjpeg/src/jcmainct.c deleted file mode 100644 index 39b97902e..000000000 --- a/dep/libjpeg/src/jcmainct.c +++ /dev/null @@ -1,297 +0,0 @@ -/* - * jcmainct.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2003-2012 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains the main buffer controller for compression. - * The main buffer lies between the pre-processor and the JPEG - * compressor proper; it holds downsampled data in the JPEG colorspace. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Note: currently, there is no operating mode in which a full-image buffer - * is needed at this step. If there were, that mode could not be used with - * "raw data" input, since this module is bypassed in that case. However, - * we've left the code here for possible use in special applications. - */ -#undef FULL_MAIN_BUFFER_SUPPORTED - - -/* Private buffer controller object */ - -typedef struct { - struct jpeg_c_main_controller pub; /* public fields */ - - JDIMENSION cur_iMCU_row; /* number of current iMCU row */ - JDIMENSION rowgroup_ctr; /* counts row groups received in iMCU row */ - boolean suspended; /* remember if we suspended output */ - J_BUF_MODE pass_mode; /* current operating mode */ - - /* If using just a strip buffer, this points to the entire set of buffers - * (we allocate one for each component). In the full-image case, this - * points to the currently accessible strips of the virtual arrays. - */ - JSAMPARRAY buffer[MAX_COMPONENTS]; - -#ifdef FULL_MAIN_BUFFER_SUPPORTED - /* If using full-image storage, this array holds pointers to virtual-array - * control blocks for each component. Unused if not full-image storage. - */ - jvirt_sarray_ptr whole_image[MAX_COMPONENTS]; -#endif -} my_main_controller; - -typedef my_main_controller * my_main_ptr; - - -/* Forward declarations */ -METHODDEF(void) process_data_simple_main - JPP((j_compress_ptr cinfo, JSAMPARRAY input_buf, - JDIMENSION *in_row_ctr, JDIMENSION in_rows_avail)); -#ifdef FULL_MAIN_BUFFER_SUPPORTED -METHODDEF(void) process_data_buffer_main - JPP((j_compress_ptr cinfo, JSAMPARRAY input_buf, - JDIMENSION *in_row_ctr, JDIMENSION in_rows_avail)); -#endif - - -/* - * Initialize for a processing pass. - */ - -METHODDEF(void) -start_pass_main (j_compress_ptr cinfo, J_BUF_MODE pass_mode) -{ - my_main_ptr mainp = (my_main_ptr) cinfo->main; - - /* Do nothing in raw-data mode. */ - if (cinfo->raw_data_in) - return; - - mainp->cur_iMCU_row = 0; /* initialize counters */ - mainp->rowgroup_ctr = 0; - mainp->suspended = FALSE; - mainp->pass_mode = pass_mode; /* save mode for use by process_data */ - - switch (pass_mode) { - case JBUF_PASS_THRU: -#ifdef FULL_MAIN_BUFFER_SUPPORTED - if (mainp->whole_image[0] != NULL) - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); -#endif - mainp->pub.process_data = process_data_simple_main; - break; -#ifdef FULL_MAIN_BUFFER_SUPPORTED - case JBUF_SAVE_SOURCE: - case JBUF_CRANK_DEST: - case JBUF_SAVE_AND_PASS: - if (mainp->whole_image[0] == NULL) - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - mainp->pub.process_data = process_data_buffer_main; - break; -#endif - default: - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - break; - } -} - - -/* - * Process some data. - * This routine handles the simple pass-through mode, - * where we have only a strip buffer. - */ - -METHODDEF(void) -process_data_simple_main (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JDIMENSION *in_row_ctr, - JDIMENSION in_rows_avail) -{ - my_main_ptr mainp = (my_main_ptr) cinfo->main; - - while (mainp->cur_iMCU_row < cinfo->total_iMCU_rows) { - /* Read input data if we haven't filled the main buffer yet */ - if (mainp->rowgroup_ctr < (JDIMENSION) cinfo->min_DCT_v_scaled_size) - (*cinfo->prep->pre_process_data) (cinfo, - input_buf, in_row_ctr, in_rows_avail, - mainp->buffer, &mainp->rowgroup_ctr, - (JDIMENSION) cinfo->min_DCT_v_scaled_size); - - /* If we don't have a full iMCU row buffered, return to application for - * more data. Note that preprocessor will always pad to fill the iMCU row - * at the bottom of the image. - */ - if (mainp->rowgroup_ctr != (JDIMENSION) cinfo->min_DCT_v_scaled_size) - return; - - /* Send the completed row to the compressor */ - if (! (*cinfo->coef->compress_data) (cinfo, mainp->buffer)) { - /* If compressor did not consume the whole row, then we must need to - * suspend processing and return to the application. In this situation - * we pretend we didn't yet consume the last input row; otherwise, if - * it happened to be the last row of the image, the application would - * think we were done. - */ - if (! mainp->suspended) { - (*in_row_ctr)--; - mainp->suspended = TRUE; - } - return; - } - /* We did finish the row. Undo our little suspension hack if a previous - * call suspended; then mark the main buffer empty. - */ - if (mainp->suspended) { - (*in_row_ctr)++; - mainp->suspended = FALSE; - } - mainp->rowgroup_ctr = 0; - mainp->cur_iMCU_row++; - } -} - - -#ifdef FULL_MAIN_BUFFER_SUPPORTED - -/* - * Process some data. - * This routine handles all of the modes that use a full-size buffer. - */ - -METHODDEF(void) -process_data_buffer_main (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JDIMENSION *in_row_ctr, - JDIMENSION in_rows_avail) -{ - my_main_ptr mainp = (my_main_ptr) cinfo->main; - int ci; - jpeg_component_info *compptr; - boolean writing = (mainp->pass_mode != JBUF_CRANK_DEST); - - while (mainp->cur_iMCU_row < cinfo->total_iMCU_rows) { - /* Realign the virtual buffers if at the start of an iMCU row. */ - if (mainp->rowgroup_ctr == 0) { - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - mainp->buffer[ci] = (*cinfo->mem->access_virt_sarray) - ((j_common_ptr) cinfo, mainp->whole_image[ci], mainp->cur_iMCU_row * - ((JDIMENSION) (compptr->v_samp_factor * cinfo->min_DCT_v_scaled_size)), - (JDIMENSION) (compptr->v_samp_factor * cinfo->min_DCT_v_scaled_size), - writing); - } - /* In a read pass, pretend we just read some source data. */ - if (! writing) { - *in_row_ctr += (JDIMENSION) - (cinfo->max_v_samp_factor * cinfo->min_DCT_v_scaled_size); - mainp->rowgroup_ctr = (JDIMENSION) cinfo->min_DCT_v_scaled_size; - } - } - - /* If a write pass, read input data until the current iMCU row is full. */ - /* Note: preprocessor will pad if necessary to fill the last iMCU row. */ - if (writing) { - (*cinfo->prep->pre_process_data) (cinfo, - input_buf, in_row_ctr, in_rows_avail, - mainp->buffer, &mainp->rowgroup_ctr, - (JDIMENSION) cinfo->min_DCT_v_scaled_size); - /* Return to application if we need more data to fill the iMCU row. */ - if (mainp->rowgroup_ctr < (JDIMENSION) cinfo->min_DCT_v_scaled_size) - return; - } - - /* Emit data, unless this is a sink-only pass. */ - if (mainp->pass_mode != JBUF_SAVE_SOURCE) { - if (! (*cinfo->coef->compress_data) (cinfo, mainp->buffer)) { - /* If compressor did not consume the whole row, then we must need to - * suspend processing and return to the application. In this situation - * we pretend we didn't yet consume the last input row; otherwise, if - * it happened to be the last row of the image, the application would - * think we were done. - */ - if (! mainp->suspended) { - (*in_row_ctr)--; - mainp->suspended = TRUE; - } - return; - } - /* We did finish the row. Undo our little suspension hack if a previous - * call suspended; then mark the main buffer empty. - */ - if (mainp->suspended) { - (*in_row_ctr)++; - mainp->suspended = FALSE; - } - } - - /* If get here, we are done with this iMCU row. Mark buffer empty. */ - mainp->rowgroup_ctr = 0; - mainp->cur_iMCU_row++; - } -} - -#endif /* FULL_MAIN_BUFFER_SUPPORTED */ - - -/* - * Initialize main buffer controller. - */ - -GLOBAL(void) -jinit_c_main_controller (j_compress_ptr cinfo, boolean need_full_buffer) -{ - my_main_ptr mainp; - int ci; - jpeg_component_info *compptr; - - mainp = (my_main_ptr) - (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE, - SIZEOF(my_main_controller)); - cinfo->main = &mainp->pub; - mainp->pub.start_pass = start_pass_main; - - /* We don't need to create a buffer in raw-data mode. */ - if (cinfo->raw_data_in) - return; - - /* Create the buffer. It holds downsampled data, so each component - * may be of a different size. - */ - if (need_full_buffer) { -#ifdef FULL_MAIN_BUFFER_SUPPORTED - /* Allocate a full-image virtual array for each component */ - /* Note we pad the bottom to a multiple of the iMCU height */ - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - mainp->whole_image[ci] = (*cinfo->mem->request_virt_sarray) - ((j_common_ptr) cinfo, JPOOL_IMAGE, FALSE, - compptr->width_in_blocks * ((JDIMENSION) compptr->DCT_h_scaled_size), - ((JDIMENSION) jround_up((long) compptr->height_in_blocks, - (long) compptr->v_samp_factor)) * - ((JDIMENSION) cinfo->min_DCT_v_scaled_size), - (JDIMENSION) (compptr->v_samp_factor * compptr->DCT_v_scaled_size)); - } -#else - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); -#endif - } else { -#ifdef FULL_MAIN_BUFFER_SUPPORTED - mainp->whole_image[0] = NULL; /* flag for no virtual arrays */ -#endif - /* Allocate a strip buffer for each component */ - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - mainp->buffer[ci] = (*cinfo->mem->alloc_sarray) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - compptr->width_in_blocks * ((JDIMENSION) compptr->DCT_h_scaled_size), - (JDIMENSION) (compptr->v_samp_factor * compptr->DCT_v_scaled_size)); - } - } -} diff --git a/dep/libjpeg/src/jcmarker.c b/dep/libjpeg/src/jcmarker.c deleted file mode 100644 index 8874cd867..000000000 --- a/dep/libjpeg/src/jcmarker.c +++ /dev/null @@ -1,717 +0,0 @@ -/* - * jcmarker.c - * - * Copyright (C) 1991-1998, Thomas G. Lane. - * Modified 2003-2019 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains routines to write JPEG datastream markers. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -typedef enum { /* JPEG marker codes */ - M_SOF0 = 0xc0, - M_SOF1 = 0xc1, - M_SOF2 = 0xc2, - M_SOF3 = 0xc3, - - M_SOF5 = 0xc5, - M_SOF6 = 0xc6, - M_SOF7 = 0xc7, - - M_JPG = 0xc8, - M_SOF9 = 0xc9, - M_SOF10 = 0xca, - M_SOF11 = 0xcb, - - M_SOF13 = 0xcd, - M_SOF14 = 0xce, - M_SOF15 = 0xcf, - - M_DHT = 0xc4, - - M_DAC = 0xcc, - - M_RST0 = 0xd0, - M_RST1 = 0xd1, - M_RST2 = 0xd2, - M_RST3 = 0xd3, - M_RST4 = 0xd4, - M_RST5 = 0xd5, - M_RST6 = 0xd6, - M_RST7 = 0xd7, - - M_SOI = 0xd8, - M_EOI = 0xd9, - M_SOS = 0xda, - M_DQT = 0xdb, - M_DNL = 0xdc, - M_DRI = 0xdd, - M_DHP = 0xde, - M_EXP = 0xdf, - - M_APP0 = 0xe0, - M_APP1 = 0xe1, - M_APP2 = 0xe2, - M_APP3 = 0xe3, - M_APP4 = 0xe4, - M_APP5 = 0xe5, - M_APP6 = 0xe6, - M_APP7 = 0xe7, - M_APP8 = 0xe8, - M_APP9 = 0xe9, - M_APP10 = 0xea, - M_APP11 = 0xeb, - M_APP12 = 0xec, - M_APP13 = 0xed, - M_APP14 = 0xee, - M_APP15 = 0xef, - - M_JPG0 = 0xf0, - M_JPG8 = 0xf8, - M_JPG13 = 0xfd, - M_COM = 0xfe, - - M_TEM = 0x01, - - M_ERROR = 0x100 -} JPEG_MARKER; - - -/* Private state */ - -typedef struct { - struct jpeg_marker_writer pub; /* public fields */ - - unsigned int last_restart_interval; /* last DRI value emitted; 0 after SOI */ -} my_marker_writer; - -typedef my_marker_writer * my_marker_ptr; - - -/* - * Basic output routines. - * - * Note that we do not support suspension while writing a marker. - * Therefore, an application using suspension must ensure that there is - * enough buffer space for the initial markers (typ. 600-700 bytes) before - * calling jpeg_start_compress, and enough space to write the trailing EOI - * (a few bytes) before calling jpeg_finish_compress. Multipass compression - * modes are not supported at all with suspension, so those two are the only - * points where markers will be written. - */ - -LOCAL(void) -emit_byte (j_compress_ptr cinfo, int val) -/* Emit a byte */ -{ - struct jpeg_destination_mgr * dest = cinfo->dest; - - *(dest->next_output_byte)++ = (JOCTET) val; - if (--dest->free_in_buffer == 0) { - if (! (*dest->empty_output_buffer) (cinfo)) - ERREXIT(cinfo, JERR_CANT_SUSPEND); - } -} - - -LOCAL(void) -emit_marker (j_compress_ptr cinfo, JPEG_MARKER mark) -/* Emit a marker code */ -{ - emit_byte(cinfo, 0xFF); - emit_byte(cinfo, (int) mark); -} - - -LOCAL(void) -emit_2bytes (j_compress_ptr cinfo, int value) -/* Emit a 2-byte integer; these are always MSB first in JPEG files */ -{ - emit_byte(cinfo, (value >> 8) & 0xFF); - emit_byte(cinfo, value & 0xFF); -} - - -/* - * Routines to write specific marker types. - */ - -LOCAL(int) -emit_dqt (j_compress_ptr cinfo, int index) -/* Emit a DQT marker */ -/* Returns the precision used (0 = 8bits, 1 = 16bits) for baseline checking */ -{ - JQUANT_TBL * qtbl = cinfo->quant_tbl_ptrs[index]; - int prec; - int i; - - if (qtbl == NULL) - ERREXIT1(cinfo, JERR_NO_QUANT_TABLE, index); - - prec = 0; - for (i = 0; i <= cinfo->lim_Se; i++) { - if (qtbl->quantval[cinfo->natural_order[i]] > 255) - prec = 1; - } - - if (! qtbl->sent_table) { - emit_marker(cinfo, M_DQT); - - emit_2bytes(cinfo, - prec ? cinfo->lim_Se * 2 + 2 + 1 + 2 : cinfo->lim_Se + 1 + 1 + 2); - - emit_byte(cinfo, index + (prec<<4)); - - for (i = 0; i <= cinfo->lim_Se; i++) { - /* The table entries must be emitted in zigzag order. */ - unsigned int qval = qtbl->quantval[cinfo->natural_order[i]]; - if (prec) - emit_byte(cinfo, (int) (qval >> 8)); - emit_byte(cinfo, (int) (qval & 0xFF)); - } - - qtbl->sent_table = TRUE; - } - - return prec; -} - - -LOCAL(void) -emit_dht (j_compress_ptr cinfo, int index, boolean is_ac) -/* Emit a DHT marker */ -{ - JHUFF_TBL * htbl; - int length, i; - - if (is_ac) { - htbl = cinfo->ac_huff_tbl_ptrs[index]; - index += 0x10; /* output index has AC bit set */ - } else { - htbl = cinfo->dc_huff_tbl_ptrs[index]; - } - - if (htbl == NULL) - ERREXIT1(cinfo, JERR_NO_HUFF_TABLE, index); - - if (! htbl->sent_table) { - emit_marker(cinfo, M_DHT); - - length = 0; - for (i = 1; i <= 16; i++) - length += htbl->bits[i]; - - emit_2bytes(cinfo, length + 2 + 1 + 16); - emit_byte(cinfo, index); - - for (i = 1; i <= 16; i++) - emit_byte(cinfo, htbl->bits[i]); - - for (i = 0; i < length; i++) - emit_byte(cinfo, htbl->huffval[i]); - - htbl->sent_table = TRUE; - } -} - - -LOCAL(void) -emit_dac (j_compress_ptr cinfo) -/* Emit a DAC marker */ -/* Since the useful info is so small, we want to emit all the tables in */ -/* one DAC marker. Therefore this routine does its own scan of the table. */ -{ -#ifdef C_ARITH_CODING_SUPPORTED - char dc_in_use[NUM_ARITH_TBLS]; - char ac_in_use[NUM_ARITH_TBLS]; - int length, i; - jpeg_component_info *compptr; - - for (i = 0; i < NUM_ARITH_TBLS; i++) - dc_in_use[i] = ac_in_use[i] = 0; - - for (i = 0; i < cinfo->comps_in_scan; i++) { - compptr = cinfo->cur_comp_info[i]; - /* DC needs no table for refinement scan */ - if (cinfo->Ss == 0 && cinfo->Ah == 0) - dc_in_use[compptr->dc_tbl_no] = 1; - /* AC needs no table when not present */ - if (cinfo->Se) - ac_in_use[compptr->ac_tbl_no] = 1; - } - - length = 0; - for (i = 0; i < NUM_ARITH_TBLS; i++) - length += dc_in_use[i] + ac_in_use[i]; - - if (length) { - emit_marker(cinfo, M_DAC); - - emit_2bytes(cinfo, length*2 + 2); - - for (i = 0; i < NUM_ARITH_TBLS; i++) { - if (dc_in_use[i]) { - emit_byte(cinfo, i); - emit_byte(cinfo, cinfo->arith_dc_L[i] + (cinfo->arith_dc_U[i]<<4)); - } - if (ac_in_use[i]) { - emit_byte(cinfo, i + 0x10); - emit_byte(cinfo, cinfo->arith_ac_K[i]); - } - } - } -#endif /* C_ARITH_CODING_SUPPORTED */ -} - - -LOCAL(void) -emit_dri (j_compress_ptr cinfo) -/* Emit a DRI marker */ -{ - emit_marker(cinfo, M_DRI); - - emit_2bytes(cinfo, 4); /* fixed length */ - - emit_2bytes(cinfo, (int) cinfo->restart_interval); -} - - -LOCAL(void) -emit_lse_ict (j_compress_ptr cinfo) -/* Emit an LSE inverse color transform specification marker */ -{ - /* Support only 1 transform */ - if (cinfo->color_transform != JCT_SUBTRACT_GREEN || - cinfo->num_components < 3) - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - - emit_marker(cinfo, M_JPG8); - - emit_2bytes(cinfo, 24); /* fixed length */ - - emit_byte(cinfo, 0x0D); /* ID inverse transform specification */ - emit_2bytes(cinfo, MAXJSAMPLE); /* MAXTRANS */ - emit_byte(cinfo, 3); /* Nt=3 */ - emit_byte(cinfo, cinfo->comp_info[1].component_id); - emit_byte(cinfo, cinfo->comp_info[0].component_id); - emit_byte(cinfo, cinfo->comp_info[2].component_id); - emit_byte(cinfo, 0x80); /* F1: CENTER1=1, NORM1=0 */ - emit_2bytes(cinfo, 0); /* A(1,1)=0 */ - emit_2bytes(cinfo, 0); /* A(1,2)=0 */ - emit_byte(cinfo, 0); /* F2: CENTER2=0, NORM2=0 */ - emit_2bytes(cinfo, 1); /* A(2,1)=1 */ - emit_2bytes(cinfo, 0); /* A(2,2)=0 */ - emit_byte(cinfo, 0); /* F3: CENTER3=0, NORM3=0 */ - emit_2bytes(cinfo, 1); /* A(3,1)=1 */ - emit_2bytes(cinfo, 0); /* A(3,2)=0 */ -} - - -LOCAL(void) -emit_sof (j_compress_ptr cinfo, JPEG_MARKER code) -/* Emit a SOF marker */ -{ - int ci; - jpeg_component_info *compptr; - - emit_marker(cinfo, code); - - emit_2bytes(cinfo, 3 * cinfo->num_components + 2 + 5 + 1); /* length */ - - /* Make sure image isn't bigger than SOF field can handle */ - if ((long) cinfo->jpeg_height > 65535L || - (long) cinfo->jpeg_width > 65535L) - ERREXIT1(cinfo, JERR_IMAGE_TOO_BIG, (unsigned int) 65535); - - emit_byte(cinfo, cinfo->data_precision); - emit_2bytes(cinfo, (int) cinfo->jpeg_height); - emit_2bytes(cinfo, (int) cinfo->jpeg_width); - - emit_byte(cinfo, cinfo->num_components); - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - emit_byte(cinfo, compptr->component_id); - emit_byte(cinfo, (compptr->h_samp_factor << 4) + compptr->v_samp_factor); - emit_byte(cinfo, compptr->quant_tbl_no); - } -} - - -LOCAL(void) -emit_sos (j_compress_ptr cinfo) -/* Emit a SOS marker */ -{ - int i, td, ta; - jpeg_component_info *compptr; - - emit_marker(cinfo, M_SOS); - - emit_2bytes(cinfo, 2 * cinfo->comps_in_scan + 2 + 1 + 3); /* length */ - - emit_byte(cinfo, cinfo->comps_in_scan); - - for (i = 0; i < cinfo->comps_in_scan; i++) { - compptr = cinfo->cur_comp_info[i]; - emit_byte(cinfo, compptr->component_id); - - /* We emit 0 for unused field(s); this is recommended by the P&M text - * but does not seem to be specified in the standard. - */ - - /* DC needs no table for refinement scan */ - td = cinfo->Ss == 0 && cinfo->Ah == 0 ? compptr->dc_tbl_no : 0; - /* AC needs no table when not present */ - ta = cinfo->Se ? compptr->ac_tbl_no : 0; - - emit_byte(cinfo, (td << 4) + ta); - } - - emit_byte(cinfo, cinfo->Ss); - emit_byte(cinfo, cinfo->Se); - emit_byte(cinfo, (cinfo->Ah << 4) + cinfo->Al); -} - - -LOCAL(void) -emit_pseudo_sos (j_compress_ptr cinfo) -/* Emit a pseudo SOS marker */ -{ - emit_marker(cinfo, M_SOS); - - emit_2bytes(cinfo, 2 + 1 + 3); /* length */ - - emit_byte(cinfo, 0); /* Ns */ - - emit_byte(cinfo, 0); /* Ss */ - emit_byte(cinfo, cinfo->block_size * cinfo->block_size - 1); /* Se */ - emit_byte(cinfo, 0); /* Ah/Al */ -} - - -LOCAL(void) -emit_jfif_app0 (j_compress_ptr cinfo) -/* Emit a JFIF-compliant APP0 marker */ -{ - /* - * Length of APP0 block (2 bytes) - * Block ID (4 bytes - ASCII "JFIF") - * Zero byte (1 byte to terminate the ID string) - * Version Major, Minor (2 bytes - major first) - * Units (1 byte - 0x00 = none, 0x01 = inch, 0x02 = cm) - * Xdpu (2 bytes - dots per unit horizontal) - * Ydpu (2 bytes - dots per unit vertical) - * Thumbnail X size (1 byte) - * Thumbnail Y size (1 byte) - */ - - emit_marker(cinfo, M_APP0); - - emit_2bytes(cinfo, 2 + 4 + 1 + 2 + 1 + 2 + 2 + 1 + 1); /* length */ - - emit_byte(cinfo, 0x4A); /* Identifier: ASCII "JFIF" */ - emit_byte(cinfo, 0x46); - emit_byte(cinfo, 0x49); - emit_byte(cinfo, 0x46); - emit_byte(cinfo, 0); - emit_byte(cinfo, cinfo->JFIF_major_version); /* Version fields */ - emit_byte(cinfo, cinfo->JFIF_minor_version); - emit_byte(cinfo, cinfo->density_unit); /* Pixel size information */ - emit_2bytes(cinfo, (int) cinfo->X_density); - emit_2bytes(cinfo, (int) cinfo->Y_density); - emit_byte(cinfo, 0); /* No thumbnail image */ - emit_byte(cinfo, 0); -} - - -LOCAL(void) -emit_adobe_app14 (j_compress_ptr cinfo) -/* Emit an Adobe APP14 marker */ -{ - /* - * Length of APP14 block (2 bytes) - * Block ID (5 bytes - ASCII "Adobe") - * Version Number (2 bytes - currently 100) - * Flags0 (2 bytes - currently 0) - * Flags1 (2 bytes - currently 0) - * Color transform (1 byte) - * - * Although Adobe TN 5116 mentions Version = 101, all the Adobe files - * now in circulation seem to use Version = 100, so that's what we write. - * - * We write the color transform byte as 1 if the JPEG color space is - * YCbCr, 2 if it's YCCK, 0 otherwise. Adobe's definition has to do with - * whether the encoder performed a transformation, which is pretty useless. - */ - - emit_marker(cinfo, M_APP14); - - emit_2bytes(cinfo, 2 + 5 + 2 + 2 + 2 + 1); /* length */ - - emit_byte(cinfo, 0x41); /* Identifier: ASCII "Adobe" */ - emit_byte(cinfo, 0x64); - emit_byte(cinfo, 0x6F); - emit_byte(cinfo, 0x62); - emit_byte(cinfo, 0x65); - emit_2bytes(cinfo, 100); /* Version */ - emit_2bytes(cinfo, 0); /* Flags0 */ - emit_2bytes(cinfo, 0); /* Flags1 */ - switch (cinfo->jpeg_color_space) { - case JCS_YCbCr: - emit_byte(cinfo, 1); /* Color transform = 1 */ - break; - case JCS_YCCK: - emit_byte(cinfo, 2); /* Color transform = 2 */ - break; - default: - emit_byte(cinfo, 0); /* Color transform = 0 */ - } -} - - -/* - * These routines allow writing an arbitrary marker with parameters. - * The only intended use is to emit COM or APPn markers after calling - * write_file_header and before calling write_frame_header. - * Other uses are not guaranteed to produce desirable results. - * Counting the parameter bytes properly is the caller's responsibility. - */ - -METHODDEF(void) -write_marker_header (j_compress_ptr cinfo, int marker, unsigned int datalen) -/* Emit an arbitrary marker header */ -{ - if (datalen > (unsigned int) 65533) /* safety check */ - ERREXIT(cinfo, JERR_BAD_LENGTH); - - emit_marker(cinfo, (JPEG_MARKER) marker); - - emit_2bytes(cinfo, (int) (datalen + 2)); /* total length */ -} - -METHODDEF(void) -write_marker_byte (j_compress_ptr cinfo, int val) -/* Emit one byte of marker parameters following write_marker_header */ -{ - emit_byte(cinfo, val); -} - - -/* - * Write datastream header. - * This consists of an SOI and optional APPn markers. - * We recommend use of the JFIF marker, but not the Adobe marker, - * when using YCbCr or grayscale data. The JFIF marker is also used - * for other standard JPEG colorspaces. The Adobe marker is helpful - * to distinguish RGB, CMYK, and YCCK colorspaces. - * Note that an application can write additional header markers after - * jpeg_start_compress returns. - */ - -METHODDEF(void) -write_file_header (j_compress_ptr cinfo) -{ - my_marker_ptr marker = (my_marker_ptr) cinfo->marker; - - emit_marker(cinfo, M_SOI); /* first the SOI */ - - /* SOI is defined to reset restart interval to 0 */ - marker->last_restart_interval = 0; - - if (cinfo->write_JFIF_header) /* next an optional JFIF APP0 */ - emit_jfif_app0(cinfo); - if (cinfo->write_Adobe_marker) /* next an optional Adobe APP14 */ - emit_adobe_app14(cinfo); -} - - -/* - * Write frame header. - * This consists of DQT and SOFn markers, - * a conditional LSE marker and a conditional pseudo SOS marker. - * Note that we do not emit the SOF until we have emitted the DQT(s). - * This avoids compatibility problems with incorrect implementations that - * try to error-check the quant table numbers as soon as they see the SOF. - */ - -METHODDEF(void) -write_frame_header (j_compress_ptr cinfo) -{ - int ci, prec; - boolean is_baseline; - jpeg_component_info *compptr; - - /* Emit DQT for each quantization table. - * Note that emit_dqt() suppresses any duplicate tables. - */ - prec = 0; - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - prec += emit_dqt(cinfo, compptr->quant_tbl_no); - } - /* now prec is nonzero iff there are any 16-bit quant tables. */ - - /* Check for a non-baseline specification. - * Note we assume that Huffman table numbers won't be changed later. - */ - if (cinfo->arith_code || cinfo->progressive_mode || - cinfo->data_precision != 8 || cinfo->block_size != DCTSIZE) { - is_baseline = FALSE; - } else { - is_baseline = TRUE; - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - if (compptr->dc_tbl_no > 1 || compptr->ac_tbl_no > 1) - is_baseline = FALSE; - } - if (prec && is_baseline) { - is_baseline = FALSE; - /* If it's baseline except for quantizer size, warn the user */ - TRACEMS(cinfo, 0, JTRC_16BIT_TABLES); - } - } - - /* Emit the proper SOF marker */ - if (cinfo->arith_code) { - if (cinfo->progressive_mode) - emit_sof(cinfo, M_SOF10); /* SOF code for progressive arithmetic */ - else - emit_sof(cinfo, M_SOF9); /* SOF code for sequential arithmetic */ - } else { - if (cinfo->progressive_mode) - emit_sof(cinfo, M_SOF2); /* SOF code for progressive Huffman */ - else if (is_baseline) - emit_sof(cinfo, M_SOF0); /* SOF code for baseline implementation */ - else - emit_sof(cinfo, M_SOF1); /* SOF code for non-baseline Huffman file */ - } - - /* Check to emit LSE inverse color transform specification marker */ - if (cinfo->color_transform) - emit_lse_ict(cinfo); - - /* Check to emit pseudo SOS marker */ - if (cinfo->progressive_mode && cinfo->block_size != DCTSIZE) - emit_pseudo_sos(cinfo); -} - - -/* - * Write scan header. - * This consists of DHT or DAC markers, optional DRI, and SOS. - * Compressed data will be written following the SOS. - */ - -METHODDEF(void) -write_scan_header (j_compress_ptr cinfo) -{ - my_marker_ptr marker = (my_marker_ptr) cinfo->marker; - int i; - jpeg_component_info *compptr; - - if (cinfo->arith_code) { - /* Emit arith conditioning info. We may have some duplication - * if the file has multiple scans, but it's so small it's hardly - * worth worrying about. - */ - emit_dac(cinfo); - } else { - /* Emit Huffman tables. - * Note that emit_dht() suppresses any duplicate tables. - */ - for (i = 0; i < cinfo->comps_in_scan; i++) { - compptr = cinfo->cur_comp_info[i]; - /* DC needs no table for refinement scan */ - if (cinfo->Ss == 0 && cinfo->Ah == 0) - emit_dht(cinfo, compptr->dc_tbl_no, FALSE); - /* AC needs no table when not present */ - if (cinfo->Se) - emit_dht(cinfo, compptr->ac_tbl_no, TRUE); - } - } - - /* Emit DRI if required --- note that DRI value could change for each scan. - * We avoid wasting space with unnecessary DRIs, however. - */ - if (cinfo->restart_interval != marker->last_restart_interval) { - emit_dri(cinfo); - marker->last_restart_interval = cinfo->restart_interval; - } - - emit_sos(cinfo); -} - - -/* - * Write datastream trailer. - */ - -METHODDEF(void) -write_file_trailer (j_compress_ptr cinfo) -{ - emit_marker(cinfo, M_EOI); -} - - -/* - * Write an abbreviated table-specification datastream. - * This consists of SOI, DQT and DHT tables, and EOI. - * Any table that is defined and not marked sent_table = TRUE will be - * emitted. Note that all tables will be marked sent_table = TRUE at exit. - */ - -METHODDEF(void) -write_tables_only (j_compress_ptr cinfo) -{ - int i; - - emit_marker(cinfo, M_SOI); - - for (i = 0; i < NUM_QUANT_TBLS; i++) { - if (cinfo->quant_tbl_ptrs[i] != NULL) - (void) emit_dqt(cinfo, i); - } - - if (! cinfo->arith_code) { - for (i = 0; i < NUM_HUFF_TBLS; i++) { - if (cinfo->dc_huff_tbl_ptrs[i] != NULL) - emit_dht(cinfo, i, FALSE); - if (cinfo->ac_huff_tbl_ptrs[i] != NULL) - emit_dht(cinfo, i, TRUE); - } - } - - emit_marker(cinfo, M_EOI); -} - - -/* - * Initialize the marker writer module. - */ - -GLOBAL(void) -jinit_marker_writer (j_compress_ptr cinfo) -{ - my_marker_ptr marker; - - /* Create the subobject */ - marker = (my_marker_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_marker_writer)); - cinfo->marker = &marker->pub; - /* Initialize method pointers */ - marker->pub.write_file_header = write_file_header; - marker->pub.write_frame_header = write_frame_header; - marker->pub.write_scan_header = write_scan_header; - marker->pub.write_file_trailer = write_file_trailer; - marker->pub.write_tables_only = write_tables_only; - marker->pub.write_marker_header = write_marker_header; - marker->pub.write_marker_byte = write_marker_byte; - /* Initialize private state */ - marker->last_restart_interval = 0; -} diff --git a/dep/libjpeg/src/jcmaster.c b/dep/libjpeg/src/jcmaster.c deleted file mode 100644 index a70af0c02..000000000 --- a/dep/libjpeg/src/jcmaster.c +++ /dev/null @@ -1,675 +0,0 @@ -/* - * jcmaster.c - * - * Copyright (C) 1991-1997, Thomas G. Lane. - * Modified 2003-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains master control logic for the JPEG compressor. - * These routines are concerned with parameter validation, initial setup, - * and inter-pass control (determining the number of passes and the work - * to be done in each pass). - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Private state */ - -typedef enum { - main_pass, /* input data, also do first output step */ - huff_opt_pass, /* Huffman code optimization pass */ - output_pass /* data output pass */ -} c_pass_type; - -typedef struct { - struct jpeg_comp_master pub; /* public fields */ - - c_pass_type pass_type; /* the type of the current pass */ - - int pass_number; /* # of passes completed */ - int total_passes; /* total # of passes needed */ - - int scan_number; /* current index in scan_info[] */ -} my_comp_master; - -typedef my_comp_master * my_master_ptr; - - -/* - * Support routines that do various essential calculations. - */ - -LOCAL(void) -initial_setup (j_compress_ptr cinfo) -/* Do computations that are needed before master selection phase */ -{ - int ci, ssize; - jpeg_component_info *compptr; - - /* Sanity check on block_size */ - if (cinfo->block_size < 1 || cinfo->block_size > 16) - ERREXIT2(cinfo, JERR_BAD_DCTSIZE, cinfo->block_size, cinfo->block_size); - - /* Derive natural_order from block_size */ - switch (cinfo->block_size) { - case 2: cinfo->natural_order = jpeg_natural_order2; break; - case 3: cinfo->natural_order = jpeg_natural_order3; break; - case 4: cinfo->natural_order = jpeg_natural_order4; break; - case 5: cinfo->natural_order = jpeg_natural_order5; break; - case 6: cinfo->natural_order = jpeg_natural_order6; break; - case 7: cinfo->natural_order = jpeg_natural_order7; break; - default: cinfo->natural_order = jpeg_natural_order; - } - - /* Derive lim_Se from block_size */ - cinfo->lim_Se = cinfo->block_size < DCTSIZE ? - cinfo->block_size * cinfo->block_size - 1 : DCTSIZE2-1; - - /* Sanity check on image dimensions */ - if (cinfo->jpeg_height <= 0 || cinfo->jpeg_width <= 0 || - cinfo->num_components <= 0) - ERREXIT(cinfo, JERR_EMPTY_IMAGE); - - /* Make sure image isn't bigger than I can handle */ - if ((long) cinfo->jpeg_height > (long) JPEG_MAX_DIMENSION || - (long) cinfo->jpeg_width > (long) JPEG_MAX_DIMENSION) - ERREXIT1(cinfo, JERR_IMAGE_TOO_BIG, (unsigned int) JPEG_MAX_DIMENSION); - - /* Only 8 to 12 bits data precision are supported for DCT based JPEG */ - if (cinfo->data_precision < 8 || cinfo->data_precision > 12) - ERREXIT1(cinfo, JERR_BAD_PRECISION, cinfo->data_precision); - - /* Check that number of components won't exceed internal array sizes */ - if (cinfo->num_components > MAX_COMPONENTS) - ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->num_components, - MAX_COMPONENTS); - - /* Compute maximum sampling factors; check factor validity */ - cinfo->max_h_samp_factor = 1; - cinfo->max_v_samp_factor = 1; - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - if (compptr->h_samp_factor<=0 || compptr->h_samp_factor>MAX_SAMP_FACTOR || - compptr->v_samp_factor<=0 || compptr->v_samp_factor>MAX_SAMP_FACTOR) - ERREXIT(cinfo, JERR_BAD_SAMPLING); - cinfo->max_h_samp_factor = MAX(cinfo->max_h_samp_factor, - compptr->h_samp_factor); - cinfo->max_v_samp_factor = MAX(cinfo->max_v_samp_factor, - compptr->v_samp_factor); - } - - /* Compute dimensions of components */ - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* Fill in the correct component_index value; don't rely on application */ - compptr->component_index = ci; - /* In selecting the actual DCT scaling for each component, we try to - * scale down the chroma components via DCT scaling rather than downsampling. - * This saves time if the downsampler gets to use 1:1 scaling. - * Note this code adapts subsampling ratios which are powers of 2. - */ - ssize = 1; -#ifdef DCT_SCALING_SUPPORTED - if (! cinfo->raw_data_in) - while (cinfo->min_DCT_h_scaled_size * ssize <= - (cinfo->do_fancy_downsampling ? DCTSIZE : DCTSIZE / 2) && - (cinfo->max_h_samp_factor % (compptr->h_samp_factor * ssize * 2)) == - 0) { - ssize = ssize * 2; - } -#endif - compptr->DCT_h_scaled_size = cinfo->min_DCT_h_scaled_size * ssize; - ssize = 1; -#ifdef DCT_SCALING_SUPPORTED - if (! cinfo->raw_data_in) - while (cinfo->min_DCT_v_scaled_size * ssize <= - (cinfo->do_fancy_downsampling ? DCTSIZE : DCTSIZE / 2) && - (cinfo->max_v_samp_factor % (compptr->v_samp_factor * ssize * 2)) == - 0) { - ssize = ssize * 2; - } -#endif - compptr->DCT_v_scaled_size = cinfo->min_DCT_v_scaled_size * ssize; - - /* We don't support DCT ratios larger than 2. */ - if (compptr->DCT_h_scaled_size > compptr->DCT_v_scaled_size * 2) - compptr->DCT_h_scaled_size = compptr->DCT_v_scaled_size * 2; - else if (compptr->DCT_v_scaled_size > compptr->DCT_h_scaled_size * 2) - compptr->DCT_v_scaled_size = compptr->DCT_h_scaled_size * 2; - - /* Size in DCT blocks */ - compptr->width_in_blocks = (JDIMENSION) - jdiv_round_up((long) cinfo->jpeg_width * (long) compptr->h_samp_factor, - (long) (cinfo->max_h_samp_factor * cinfo->block_size)); - compptr->height_in_blocks = (JDIMENSION) - jdiv_round_up((long) cinfo->jpeg_height * (long) compptr->v_samp_factor, - (long) (cinfo->max_v_samp_factor * cinfo->block_size)); - /* Size in samples */ - compptr->downsampled_width = (JDIMENSION) - jdiv_round_up((long) cinfo->jpeg_width * - (long) (compptr->h_samp_factor * compptr->DCT_h_scaled_size), - (long) (cinfo->max_h_samp_factor * cinfo->block_size)); - compptr->downsampled_height = (JDIMENSION) - jdiv_round_up((long) cinfo->jpeg_height * - (long) (compptr->v_samp_factor * compptr->DCT_v_scaled_size), - (long) (cinfo->max_v_samp_factor * cinfo->block_size)); - /* Don't need quantization scale after DCT, - * until color conversion says otherwise. - */ - compptr->component_needed = FALSE; - } - - /* Compute number of fully interleaved MCU rows (number of times that - * main controller will call coefficient controller). - */ - cinfo->total_iMCU_rows = (JDIMENSION) - jdiv_round_up((long) cinfo->jpeg_height, - (long) (cinfo->max_v_samp_factor * cinfo->block_size)); -} - - -#ifdef C_MULTISCAN_FILES_SUPPORTED - -LOCAL(void) -validate_script (j_compress_ptr cinfo) -/* Verify that the scan script in cinfo->scan_info[] is valid; also - * determine whether it uses progressive JPEG, and set cinfo->progressive_mode. - */ -{ - const jpeg_scan_info * scanptr; - int scanno, ncomps, ci, coefi, thisi; - int Ss, Se, Ah, Al; - boolean component_sent[MAX_COMPONENTS]; -#ifdef C_PROGRESSIVE_SUPPORTED - int * last_bitpos_ptr; - int last_bitpos[MAX_COMPONENTS][DCTSIZE2]; - /* -1 until that coefficient has been seen; then last Al for it */ -#endif - - if (cinfo->num_scans <= 0) - ERREXIT1(cinfo, JERR_BAD_SCAN_SCRIPT, 0); - - /* For sequential JPEG, all scans must have Ss=0, Se=DCTSIZE2-1; - * for progressive JPEG, no scan can have this. - */ - scanptr = cinfo->scan_info; - if (scanptr->Ss != 0 || scanptr->Se != DCTSIZE2-1) { -#ifdef C_PROGRESSIVE_SUPPORTED - cinfo->progressive_mode = TRUE; - last_bitpos_ptr = & last_bitpos[0][0]; - for (ci = 0; ci < cinfo->num_components; ci++) - for (coefi = 0; coefi < DCTSIZE2; coefi++) - *last_bitpos_ptr++ = -1; -#else - ERREXIT(cinfo, JERR_NOT_COMPILED); -#endif - } else { - cinfo->progressive_mode = FALSE; - for (ci = 0; ci < cinfo->num_components; ci++) - component_sent[ci] = FALSE; - } - - for (scanno = 1; scanno <= cinfo->num_scans; scanptr++, scanno++) { - /* Validate component indexes */ - ncomps = scanptr->comps_in_scan; - if (ncomps <= 0 || ncomps > MAX_COMPS_IN_SCAN) - ERREXIT2(cinfo, JERR_COMPONENT_COUNT, ncomps, MAX_COMPS_IN_SCAN); - for (ci = 0; ci < ncomps; ci++) { - thisi = scanptr->component_index[ci]; - if (thisi < 0 || thisi >= cinfo->num_components) - ERREXIT1(cinfo, JERR_BAD_SCAN_SCRIPT, scanno); - /* Components must appear in SOF order within each scan */ - if (ci > 0 && thisi <= scanptr->component_index[ci-1]) - ERREXIT1(cinfo, JERR_BAD_SCAN_SCRIPT, scanno); - } - /* Validate progression parameters */ - Ss = scanptr->Ss; - Se = scanptr->Se; - Ah = scanptr->Ah; - Al = scanptr->Al; - if (cinfo->progressive_mode) { -#ifdef C_PROGRESSIVE_SUPPORTED - /* The JPEG spec simply gives the ranges 0..13 for Ah and Al, but that - * seems wrong: the upper bound ought to depend on data precision. - * Perhaps they really meant 0..N+1 for N-bit precision. - * Here we allow 0..10 for 8-bit data; Al larger than 10 results in - * out-of-range reconstructed DC values during the first DC scan, - * which might cause problems for some decoders. - */ - if (Ss < 0 || Ss >= DCTSIZE2 || Se < Ss || Se >= DCTSIZE2 || - Ah < 0 || Ah > (cinfo->data_precision > 8 ? 13 : 10) || - Al < 0 || Al > (cinfo->data_precision > 8 ? 13 : 10)) - ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno); - if (Ss == 0) { - if (Se != 0) /* DC and AC together not OK */ - ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno); - } else { - if (ncomps != 1) /* AC scans must be for only one component */ - ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno); - } - for (ci = 0; ci < ncomps; ci++) { - last_bitpos_ptr = & last_bitpos[scanptr->component_index[ci]][0]; - if (Ss != 0 && last_bitpos_ptr[0] < 0) /* AC without prior DC scan */ - ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno); - for (coefi = Ss; coefi <= Se; coefi++) { - if (last_bitpos_ptr[coefi] < 0) { - /* first scan of this coefficient */ - if (Ah != 0) - ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno); - } else { - /* not first scan */ - if (Ah != last_bitpos_ptr[coefi] || Al != Ah-1) - ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno); - } - last_bitpos_ptr[coefi] = Al; - } - } -#endif - } else { - /* For sequential JPEG, all progression parameters must be these: */ - if (Ss != 0 || Se != DCTSIZE2-1 || Ah != 0 || Al != 0) - ERREXIT1(cinfo, JERR_BAD_PROG_SCRIPT, scanno); - /* Make sure components are not sent twice */ - for (ci = 0; ci < ncomps; ci++) { - thisi = scanptr->component_index[ci]; - if (component_sent[thisi]) - ERREXIT1(cinfo, JERR_BAD_SCAN_SCRIPT, scanno); - component_sent[thisi] = TRUE; - } - } - } - - /* Now verify that everything got sent. */ - if (cinfo->progressive_mode) { -#ifdef C_PROGRESSIVE_SUPPORTED - /* For progressive mode, we only check that at least some DC data - * got sent for each component; the spec does not require that all bits - * of all coefficients be transmitted. Would it be wiser to enforce - * transmission of all coefficient bits?? - */ - for (ci = 0; ci < cinfo->num_components; ci++) { - if (last_bitpos[ci][0] < 0) - ERREXIT(cinfo, JERR_MISSING_DATA); - } -#endif - } else { - for (ci = 0; ci < cinfo->num_components; ci++) { - if (! component_sent[ci]) - ERREXIT(cinfo, JERR_MISSING_DATA); - } - } -} - - -LOCAL(void) -reduce_script (j_compress_ptr cinfo) -/* Adapt scan script for use with reduced block size; - * assume that script has been validated before. - */ -{ - jpeg_scan_info * scanptr; - int idxout, idxin; - - /* Circumvent const declaration for this function */ - scanptr = (jpeg_scan_info *) cinfo->scan_info; - idxout = 0; - - for (idxin = 0; idxin < cinfo->num_scans; idxin++) { - /* After skipping, idxout becomes smaller than idxin */ - if (idxin != idxout) - /* Copy rest of data; - * note we stay in given chunk of allocated memory. - */ - scanptr[idxout] = scanptr[idxin]; - if (scanptr[idxout].Ss > cinfo->lim_Se) - /* Entire scan out of range - skip this entry */ - continue; - if (scanptr[idxout].Se > cinfo->lim_Se) - /* Limit scan to end of block */ - scanptr[idxout].Se = cinfo->lim_Se; - idxout++; - } - - cinfo->num_scans = idxout; -} - -#endif /* C_MULTISCAN_FILES_SUPPORTED */ - - -LOCAL(void) -select_scan_parameters (j_compress_ptr cinfo) -/* Set up the scan parameters for the current scan */ -{ - int ci; - -#ifdef C_MULTISCAN_FILES_SUPPORTED - if (cinfo->scan_info != NULL) { - /* Prepare for current scan --- the script is already validated */ - my_master_ptr master = (my_master_ptr) cinfo->master; - const jpeg_scan_info * scanptr = cinfo->scan_info + master->scan_number; - - cinfo->comps_in_scan = scanptr->comps_in_scan; - for (ci = 0; ci < scanptr->comps_in_scan; ci++) { - cinfo->cur_comp_info[ci] = - &cinfo->comp_info[scanptr->component_index[ci]]; - } - if (cinfo->progressive_mode) { - cinfo->Ss = scanptr->Ss; - cinfo->Se = scanptr->Se; - cinfo->Ah = scanptr->Ah; - cinfo->Al = scanptr->Al; - return; - } - } - else -#endif - { - /* Prepare for single sequential-JPEG scan containing all components */ - if (cinfo->num_components > MAX_COMPS_IN_SCAN) - ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->num_components, - MAX_COMPS_IN_SCAN); - cinfo->comps_in_scan = cinfo->num_components; - for (ci = 0; ci < cinfo->num_components; ci++) { - cinfo->cur_comp_info[ci] = &cinfo->comp_info[ci]; - } - } - cinfo->Ss = 0; - cinfo->Se = cinfo->block_size * cinfo->block_size - 1; - cinfo->Ah = 0; - cinfo->Al = 0; -} - - -LOCAL(void) -per_scan_setup (j_compress_ptr cinfo) -/* Do computations that are needed before processing a JPEG scan */ -/* cinfo->comps_in_scan and cinfo->cur_comp_info[] are already set */ -{ - int ci, mcublks, tmp; - jpeg_component_info *compptr; - - if (cinfo->comps_in_scan == 1) { - - /* Noninterleaved (single-component) scan */ - compptr = cinfo->cur_comp_info[0]; - - /* Overall image size in MCUs */ - cinfo->MCUs_per_row = compptr->width_in_blocks; - cinfo->MCU_rows_in_scan = compptr->height_in_blocks; - - /* For noninterleaved scan, always one block per MCU */ - compptr->MCU_width = 1; - compptr->MCU_height = 1; - compptr->MCU_blocks = 1; - compptr->MCU_sample_width = compptr->DCT_h_scaled_size; - compptr->last_col_width = 1; - /* For noninterleaved scans, it is convenient to define last_row_height - * as the number of block rows present in the last iMCU row. - */ - tmp = (int) (compptr->height_in_blocks % compptr->v_samp_factor); - if (tmp == 0) tmp = compptr->v_samp_factor; - compptr->last_row_height = tmp; - - /* Prepare array describing MCU composition */ - cinfo->blocks_in_MCU = 1; - cinfo->MCU_membership[0] = 0; - - } else { - - /* Interleaved (multi-component) scan */ - if (cinfo->comps_in_scan <= 0 || cinfo->comps_in_scan > MAX_COMPS_IN_SCAN) - ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->comps_in_scan, - MAX_COMPS_IN_SCAN); - - /* Overall image size in MCUs */ - cinfo->MCUs_per_row = (JDIMENSION) - jdiv_round_up((long) cinfo->jpeg_width, - (long) (cinfo->max_h_samp_factor * cinfo->block_size)); - cinfo->MCU_rows_in_scan = cinfo->total_iMCU_rows; - - cinfo->blocks_in_MCU = 0; - - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - /* Sampling factors give # of blocks of component in each MCU */ - compptr->MCU_width = compptr->h_samp_factor; - compptr->MCU_height = compptr->v_samp_factor; - compptr->MCU_blocks = compptr->MCU_width * compptr->MCU_height; - compptr->MCU_sample_width = compptr->MCU_width * compptr->DCT_h_scaled_size; - /* Figure number of non-dummy blocks in last MCU column & row */ - tmp = (int) (compptr->width_in_blocks % compptr->MCU_width); - if (tmp == 0) tmp = compptr->MCU_width; - compptr->last_col_width = tmp; - tmp = (int) (compptr->height_in_blocks % compptr->MCU_height); - if (tmp == 0) tmp = compptr->MCU_height; - compptr->last_row_height = tmp; - /* Prepare array describing MCU composition */ - mcublks = compptr->MCU_blocks; - if (cinfo->blocks_in_MCU + mcublks > C_MAX_BLOCKS_IN_MCU) - ERREXIT(cinfo, JERR_BAD_MCU_SIZE); - while (mcublks-- > 0) { - cinfo->MCU_membership[cinfo->blocks_in_MCU++] = ci; - } - } - - } - - /* Convert restart specified in rows to actual MCU count. */ - /* Note that count must fit in 16 bits, so we provide limiting. */ - if (cinfo->restart_in_rows > 0) { - long nominal = (long) cinfo->restart_in_rows * (long) cinfo->MCUs_per_row; - cinfo->restart_interval = (unsigned int) MIN(nominal, 65535L); - } -} - - -/* - * Per-pass setup. - * This is called at the beginning of each pass. We determine which modules - * will be active during this pass and give them appropriate start_pass calls. - * We also set is_last_pass to indicate whether any more passes will be - * required. - */ - -METHODDEF(void) -prepare_for_pass (j_compress_ptr cinfo) -{ - my_master_ptr master = (my_master_ptr) cinfo->master; - - switch (master->pass_type) { - case main_pass: - /* Initial pass: will collect input data, and do either Huffman - * optimization or data output for the first scan. - */ - select_scan_parameters(cinfo); - per_scan_setup(cinfo); - if (! cinfo->raw_data_in) { - (*cinfo->cconvert->start_pass) (cinfo); - (*cinfo->downsample->start_pass) (cinfo); - (*cinfo->prep->start_pass) (cinfo, JBUF_PASS_THRU); - } - (*cinfo->fdct->start_pass) (cinfo); - (*cinfo->entropy->start_pass) (cinfo, cinfo->optimize_coding); - (*cinfo->coef->start_pass) (cinfo, - (master->total_passes > 1 ? - JBUF_SAVE_AND_PASS : JBUF_PASS_THRU)); - (*cinfo->main->start_pass) (cinfo, JBUF_PASS_THRU); - if (cinfo->optimize_coding) { - /* No immediate data output; postpone writing frame/scan headers */ - master->pub.call_pass_startup = FALSE; - } else { - /* Will write frame/scan headers at first jpeg_write_scanlines call */ - master->pub.call_pass_startup = TRUE; - } - break; -#ifdef ENTROPY_OPT_SUPPORTED - case huff_opt_pass: - /* Do Huffman optimization for a scan after the first one. */ - select_scan_parameters(cinfo); - per_scan_setup(cinfo); - if (cinfo->Ss != 0 || cinfo->Ah == 0) { - (*cinfo->entropy->start_pass) (cinfo, TRUE); - (*cinfo->coef->start_pass) (cinfo, JBUF_CRANK_DEST); - master->pub.call_pass_startup = FALSE; - break; - } - /* Special case: Huffman DC refinement scans need no Huffman table - * and therefore we can skip the optimization pass for them. - */ - master->pass_type = output_pass; - master->pass_number++; - /*FALLTHROUGH*/ -#endif - case output_pass: - /* Do a data-output pass. */ - /* We need not repeat per-scan setup if prior optimization pass did it. */ - if (! cinfo->optimize_coding) { - select_scan_parameters(cinfo); - per_scan_setup(cinfo); - } - (*cinfo->entropy->start_pass) (cinfo, FALSE); - (*cinfo->coef->start_pass) (cinfo, JBUF_CRANK_DEST); - /* We emit frame/scan headers now */ - if (master->scan_number == 0) - (*cinfo->marker->write_frame_header) (cinfo); - (*cinfo->marker->write_scan_header) (cinfo); - master->pub.call_pass_startup = FALSE; - break; - default: - ERREXIT(cinfo, JERR_NOT_COMPILED); - } - - master->pub.is_last_pass = (master->pass_number == master->total_passes-1); - - /* Set up progress monitor's pass info if present */ - if (cinfo->progress != NULL) { - cinfo->progress->completed_passes = master->pass_number; - cinfo->progress->total_passes = master->total_passes; - } -} - - -/* - * Special start-of-pass hook. - * This is called by jpeg_write_scanlines if call_pass_startup is TRUE. - * In single-pass processing, we need this hook because we don't want to - * write frame/scan headers during jpeg_start_compress; we want to let the - * application write COM markers etc. between jpeg_start_compress and the - * jpeg_write_scanlines loop. - * In multi-pass processing, this routine is not used. - */ - -METHODDEF(void) -pass_startup (j_compress_ptr cinfo) -{ - cinfo->master->call_pass_startup = FALSE; /* reset flag so call only once */ - - (*cinfo->marker->write_frame_header) (cinfo); - (*cinfo->marker->write_scan_header) (cinfo); -} - - -/* - * Finish up at end of pass. - */ - -METHODDEF(void) -finish_pass_master (j_compress_ptr cinfo) -{ - my_master_ptr master = (my_master_ptr) cinfo->master; - - /* The entropy coder always needs an end-of-pass call, - * either to analyze statistics or to flush its output buffer. - */ - (*cinfo->entropy->finish_pass) (cinfo); - - /* Update state for next pass */ - switch (master->pass_type) { - case main_pass: - /* next pass is either output of scan 0 (after optimization) - * or output of scan 1 (if no optimization). - */ - master->pass_type = output_pass; - if (! cinfo->optimize_coding) - master->scan_number++; - break; - case huff_opt_pass: - /* next pass is always output of current scan */ - master->pass_type = output_pass; - break; - case output_pass: - /* next pass is either optimization or output of next scan */ - if (cinfo->optimize_coding) - master->pass_type = huff_opt_pass; - master->scan_number++; - break; - } - - master->pass_number++; -} - - -/* - * Initialize master compression control. - */ - -GLOBAL(void) -jinit_c_master_control (j_compress_ptr cinfo, boolean transcode_only) -{ - my_master_ptr master; - - master = (my_master_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_comp_master)); - cinfo->master = &master->pub; - master->pub.prepare_for_pass = prepare_for_pass; - master->pub.pass_startup = pass_startup; - master->pub.finish_pass = finish_pass_master; - master->pub.is_last_pass = FALSE; - - /* Validate parameters, determine derived values */ - initial_setup(cinfo); - - if (cinfo->scan_info != NULL) { -#ifdef C_MULTISCAN_FILES_SUPPORTED - validate_script(cinfo); - if (cinfo->block_size < DCTSIZE) - reduce_script(cinfo); -#else - ERREXIT(cinfo, JERR_NOT_COMPILED); -#endif - } else { - cinfo->progressive_mode = FALSE; - cinfo->num_scans = 1; - } - - if (cinfo->optimize_coding) - cinfo->arith_code = FALSE; /* disable arithmetic coding */ - else if (! cinfo->arith_code && - (cinfo->progressive_mode || - (cinfo->block_size > 1 && cinfo->block_size < DCTSIZE))) - /* TEMPORARY HACK ??? */ - /* assume default tables no good for progressive or reduced AC mode */ - cinfo->optimize_coding = TRUE; /* force Huffman optimization */ - - /* Initialize my private state */ - if (transcode_only) { - /* no main pass in transcoding */ - if (cinfo->optimize_coding) - master->pass_type = huff_opt_pass; - else - master->pass_type = output_pass; - } else { - /* for normal compression, first pass is always this type: */ - master->pass_type = main_pass; - } - master->scan_number = 0; - master->pass_number = 0; - if (cinfo->optimize_coding) - master->total_passes = cinfo->num_scans * 2; - else - master->total_passes = cinfo->num_scans; -} diff --git a/dep/libjpeg/src/jcomapi.c b/dep/libjpeg/src/jcomapi.c deleted file mode 100644 index 678c5d12d..000000000 --- a/dep/libjpeg/src/jcomapi.c +++ /dev/null @@ -1,244 +0,0 @@ -/* - * jcomapi.c - * - * Copyright (C) 1994-1997, Thomas G. Lane. - * Modified 2019 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains application interface routines that are used for both - * compression and decompression. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* - * Abort processing of a JPEG compression or decompression operation, - * but don't destroy the object itself. - * - * For this, we merely clean up all the nonpermanent memory pools. - * Note that temp files (virtual arrays) are not allowed to belong to - * the permanent pool, so we will be able to close all temp files here. - * Closing a data source or destination, if necessary, is the application's - * responsibility. - */ - -GLOBAL(void) -jpeg_abort (j_common_ptr cinfo) -{ - int pool; - - /* Do nothing if called on a not-initialized or destroyed JPEG object. */ - if (cinfo->mem == NULL) - return; - - /* Releasing pools in reverse order might help avoid fragmentation - * with some (brain-damaged) malloc libraries. - */ - for (pool = JPOOL_NUMPOOLS-1; pool > JPOOL_PERMANENT; pool--) { - (*cinfo->mem->free_pool) (cinfo, pool); - } - - /* Reset overall state for possible reuse of object */ - if (cinfo->is_decompressor) { - cinfo->global_state = DSTATE_START; - /* Try to keep application from accessing now-deleted marker list. - * A bit kludgy to do it here, but this is the most central place. - */ - ((j_decompress_ptr) cinfo)->marker_list = NULL; - } else { - cinfo->global_state = CSTATE_START; - } -} - - -/* - * Destruction of a JPEG object. - * - * Everything gets deallocated except the master jpeg_compress_struct itself - * and the error manager struct. Both of these are supplied by the application - * and must be freed, if necessary, by the application. (Often they are on - * the stack and so don't need to be freed anyway.) - * Closing a data source or destination, if necessary, is the application's - * responsibility. - */ - -GLOBAL(void) -jpeg_destroy (j_common_ptr cinfo) -{ - /* We need only tell the memory manager to release everything. */ - /* NB: mem pointer is NULL if memory mgr failed to initialize. */ - if (cinfo->mem != NULL) - (*cinfo->mem->self_destruct) (cinfo); - cinfo->mem = NULL; /* be safe if jpeg_destroy is called twice */ - cinfo->global_state = 0; /* mark it destroyed */ -} - - -/* - * Convenience routines for allocating quantization and Huffman tables. - * (Would jutils.c be a more reasonable place to put these?) - */ - -GLOBAL(JQUANT_TBL *) -jpeg_alloc_quant_table (j_common_ptr cinfo) -{ - JQUANT_TBL *tbl; - - tbl = (JQUANT_TBL *) - (*cinfo->mem->alloc_small) (cinfo, JPOOL_PERMANENT, SIZEOF(JQUANT_TBL)); - tbl->sent_table = FALSE; /* make sure this is false in any new table */ - return tbl; -} - - -GLOBAL(JHUFF_TBL *) -jpeg_alloc_huff_table (j_common_ptr cinfo) -{ - JHUFF_TBL *tbl; - - tbl = (JHUFF_TBL *) - (*cinfo->mem->alloc_small) (cinfo, JPOOL_PERMANENT, SIZEOF(JHUFF_TBL)); - tbl->sent_table = FALSE; /* make sure this is false in any new table */ - return tbl; -} - - -/* - * Set up the standard Huffman tables (cf. JPEG standard section K.3). - * IMPORTANT: these are only valid for 8-bit data precision! - * (Would jutils.c be a more reasonable place to put this?) - */ - -GLOBAL(JHUFF_TBL *) -jpeg_std_huff_table (j_common_ptr cinfo, boolean isDC, int tblno) -{ - JHUFF_TBL **htblptr, *htbl; - const UINT8 *bits, *val; - int nsymbols, len; - - static const UINT8 bits_dc_luminance[17] = - { /* 0-base */ 0, 0, 1, 5, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0, 0, 0 }; - static const UINT8 val_dc_luminance[] = - { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 }; - - static const UINT8 bits_dc_chrominance[17] = - { /* 0-base */ 0, 0, 3, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 0, 0, 0, 0 }; - static const UINT8 val_dc_chrominance[] = - { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 }; - - static const UINT8 bits_ac_luminance[17] = - { /* 0-base */ 0, 0, 2, 1, 3, 3, 2, 4, 3, 5, 5, 4, 4, 0, 0, 1, 0x7d }; - static const UINT8 val_ac_luminance[] = - { 0x01, 0x02, 0x03, 0x00, 0x04, 0x11, 0x05, 0x12, - 0x21, 0x31, 0x41, 0x06, 0x13, 0x51, 0x61, 0x07, - 0x22, 0x71, 0x14, 0x32, 0x81, 0x91, 0xa1, 0x08, - 0x23, 0x42, 0xb1, 0xc1, 0x15, 0x52, 0xd1, 0xf0, - 0x24, 0x33, 0x62, 0x72, 0x82, 0x09, 0x0a, 0x16, - 0x17, 0x18, 0x19, 0x1a, 0x25, 0x26, 0x27, 0x28, - 0x29, 0x2a, 0x34, 0x35, 0x36, 0x37, 0x38, 0x39, - 0x3a, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, 0x49, - 0x4a, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, 0x59, - 0x5a, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, 0x69, - 0x6a, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, 0x79, - 0x7a, 0x83, 0x84, 0x85, 0x86, 0x87, 0x88, 0x89, - 0x8a, 0x92, 0x93, 0x94, 0x95, 0x96, 0x97, 0x98, - 0x99, 0x9a, 0xa2, 0xa3, 0xa4, 0xa5, 0xa6, 0xa7, - 0xa8, 0xa9, 0xaa, 0xb2, 0xb3, 0xb4, 0xb5, 0xb6, - 0xb7, 0xb8, 0xb9, 0xba, 0xc2, 0xc3, 0xc4, 0xc5, - 0xc6, 0xc7, 0xc8, 0xc9, 0xca, 0xd2, 0xd3, 0xd4, - 0xd5, 0xd6, 0xd7, 0xd8, 0xd9, 0xda, 0xe1, 0xe2, - 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, 0xe8, 0xe9, 0xea, - 0xf1, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8, - 0xf9, 0xfa }; - - static const UINT8 bits_ac_chrominance[17] = - { /* 0-base */ 0, 0, 2, 1, 2, 4, 4, 3, 4, 7, 5, 4, 4, 0, 1, 2, 0x77 }; - static const UINT8 val_ac_chrominance[] = - { 0x00, 0x01, 0x02, 0x03, 0x11, 0x04, 0x05, 0x21, - 0x31, 0x06, 0x12, 0x41, 0x51, 0x07, 0x61, 0x71, - 0x13, 0x22, 0x32, 0x81, 0x08, 0x14, 0x42, 0x91, - 0xa1, 0xb1, 0xc1, 0x09, 0x23, 0x33, 0x52, 0xf0, - 0x15, 0x62, 0x72, 0xd1, 0x0a, 0x16, 0x24, 0x34, - 0xe1, 0x25, 0xf1, 0x17, 0x18, 0x19, 0x1a, 0x26, - 0x27, 0x28, 0x29, 0x2a, 0x35, 0x36, 0x37, 0x38, - 0x39, 0x3a, 0x43, 0x44, 0x45, 0x46, 0x47, 0x48, - 0x49, 0x4a, 0x53, 0x54, 0x55, 0x56, 0x57, 0x58, - 0x59, 0x5a, 0x63, 0x64, 0x65, 0x66, 0x67, 0x68, - 0x69, 0x6a, 0x73, 0x74, 0x75, 0x76, 0x77, 0x78, - 0x79, 0x7a, 0x82, 0x83, 0x84, 0x85, 0x86, 0x87, - 0x88, 0x89, 0x8a, 0x92, 0x93, 0x94, 0x95, 0x96, - 0x97, 0x98, 0x99, 0x9a, 0xa2, 0xa3, 0xa4, 0xa5, - 0xa6, 0xa7, 0xa8, 0xa9, 0xaa, 0xb2, 0xb3, 0xb4, - 0xb5, 0xb6, 0xb7, 0xb8, 0xb9, 0xba, 0xc2, 0xc3, - 0xc4, 0xc5, 0xc6, 0xc7, 0xc8, 0xc9, 0xca, 0xd2, - 0xd3, 0xd4, 0xd5, 0xd6, 0xd7, 0xd8, 0xd9, 0xda, - 0xe2, 0xe3, 0xe4, 0xe5, 0xe6, 0xe7, 0xe8, 0xe9, - 0xea, 0xf2, 0xf3, 0xf4, 0xf5, 0xf6, 0xf7, 0xf8, - 0xf9, 0xfa }; - - if (cinfo->is_decompressor) { - if (isDC) - htblptr = ((j_decompress_ptr) cinfo)->dc_huff_tbl_ptrs; - else - htblptr = ((j_decompress_ptr) cinfo)->ac_huff_tbl_ptrs; - } else { - if (isDC) - htblptr = ((j_compress_ptr) cinfo)->dc_huff_tbl_ptrs; - else - htblptr = ((j_compress_ptr) cinfo)->ac_huff_tbl_ptrs; - } - - switch (tblno) { - case 0: - if (isDC) { - bits = bits_dc_luminance; - val = val_dc_luminance; - } else { - bits = bits_ac_luminance; - val = val_ac_luminance; - } - break; - case 1: - if (isDC) { - bits = bits_dc_chrominance; - val = val_dc_chrominance; - } else { - bits = bits_ac_chrominance; - val = val_ac_chrominance; - } - break; - default: - ERREXIT1(cinfo, JERR_NO_HUFF_TABLE, tblno); - return NULL; /* avoid compiler warnings for uninitialized variables */ - } - - if (htblptr[tblno] == NULL) - htblptr[tblno] = jpeg_alloc_huff_table(cinfo); - - htbl = htblptr[tblno]; - - /* Copy the number-of-symbols-of-each-code-length counts */ - MEMCOPY(htbl->bits, bits, SIZEOF(htbl->bits)); - - /* Validate the counts. We do this here mainly so we can copy the right - * number of symbols from the val[] array, without risking marching off - * the end of memory. jxhuff.c will do a more thorough test later. - */ - nsymbols = 0; - for (len = 1; len <= 16; len++) - nsymbols += bits[len]; - if (nsymbols > 256) - ERREXIT(cinfo, JERR_BAD_HUFF_TABLE); - - if (nsymbols > 0) - MEMCOPY(htbl->huffval, val, nsymbols * SIZEOF(UINT8)); - - /* Initialize sent_table FALSE so table will be written to JPEG file. */ - htbl->sent_table = FALSE; - - return htbl; -} diff --git a/dep/libjpeg/src/jcparam.c b/dep/libjpeg/src/jcparam.c deleted file mode 100644 index 261ae86ca..000000000 --- a/dep/libjpeg/src/jcparam.c +++ /dev/null @@ -1,591 +0,0 @@ -/* - * jcparam.c - * - * Copyright (C) 1991-1998, Thomas G. Lane. - * Modified 2003-2022 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains optional default-setting code for the JPEG compressor. - * Applications do not have to use this file, but those that don't use it - * must know a lot more about the innards of the JPEG code. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* - * Quantization table setup routines - */ - -GLOBAL(void) -jpeg_add_quant_table (j_compress_ptr cinfo, int which_tbl, - const unsigned int *basic_table, - int scale_factor, boolean force_baseline) -/* Define a quantization table equal to the basic_table times - * a scale factor (given as a percentage). - * If force_baseline is TRUE, the computed quantization table entries - * are limited to 1..255 for JPEG baseline compatibility. - */ -{ - JQUANT_TBL ** qtblptr; - int i; - long temp; - - /* Safety check to ensure start_compress not called yet. */ - if (cinfo->global_state != CSTATE_START) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - - if (which_tbl < 0 || which_tbl >= NUM_QUANT_TBLS) - ERREXIT1(cinfo, JERR_DQT_INDEX, which_tbl); - - qtblptr = & cinfo->quant_tbl_ptrs[which_tbl]; - - if (*qtblptr == NULL) - *qtblptr = jpeg_alloc_quant_table((j_common_ptr) cinfo); - - for (i = 0; i < DCTSIZE2; i++) { - temp = ((long) basic_table[i] * scale_factor + 50L) / 100L; - /* limit the values to the valid range */ - if (temp <= 0L) temp = 1L; - if (temp > 32767L) temp = 32767L; /* max quantizer needed for 12 bits */ - if (force_baseline && temp > 255L) - temp = 255L; /* limit to baseline range if requested */ - (*qtblptr)->quantval[i] = (UINT16) temp; - } - - /* Initialize sent_table FALSE so table will be written to JPEG file. */ - (*qtblptr)->sent_table = FALSE; -} - - -/* These are the sample quantization tables given in JPEG spec section K.1. - * NOTE: chrominance DC value is changed from 17 to 16 for lossless support. - * The spec says that the values given produce "good" quality, - * and when divided by 2, "very good" quality. - */ -static const unsigned int std_luminance_quant_tbl[DCTSIZE2] = { - 16, 11, 10, 16, 24, 40, 51, 61, - 12, 12, 14, 19, 26, 58, 60, 55, - 14, 13, 16, 24, 40, 57, 69, 56, - 14, 17, 22, 29, 51, 87, 80, 62, - 18, 22, 37, 56, 68, 109, 103, 77, - 24, 35, 55, 64, 81, 104, 113, 92, - 49, 64, 78, 87, 103, 121, 120, 101, - 72, 92, 95, 98, 112, 100, 103, 99 -}; -static const unsigned int std_chrominance_quant_tbl[DCTSIZE2] = { - 16, 18, 24, 47, 99, 99, 99, 99, - 18, 21, 26, 66, 99, 99, 99, 99, - 24, 26, 56, 99, 99, 99, 99, 99, - 47, 66, 99, 99, 99, 99, 99, 99, - 99, 99, 99, 99, 99, 99, 99, 99, - 99, 99, 99, 99, 99, 99, 99, 99, - 99, 99, 99, 99, 99, 99, 99, 99, - 99, 99, 99, 99, 99, 99, 99, 99 -}; - - -GLOBAL(void) -jpeg_default_qtables (j_compress_ptr cinfo, boolean force_baseline) -/* Set or change the 'quality' (quantization) setting, using default tables - * and straight percentage-scaling quality scales. - * This entry point allows different scalings for luminance and chrominance. - */ -{ - /* Set up two quantization tables using the specified scaling */ - jpeg_add_quant_table(cinfo, 0, std_luminance_quant_tbl, - cinfo->q_scale_factor[0], force_baseline); - jpeg_add_quant_table(cinfo, 1, std_chrominance_quant_tbl, - cinfo->q_scale_factor[1], force_baseline); -} - - -GLOBAL(void) -jpeg_set_linear_quality (j_compress_ptr cinfo, int scale_factor, - boolean force_baseline) -/* Set or change the 'quality' (quantization) setting, using default tables - * and a straight percentage-scaling quality scale. In most cases it's better - * to use jpeg_set_quality (below); this entry point is provided for - * applications that insist on a linear percentage scaling. - */ -{ - /* Set up two quantization tables using the specified scaling */ - jpeg_add_quant_table(cinfo, 0, std_luminance_quant_tbl, - scale_factor, force_baseline); - jpeg_add_quant_table(cinfo, 1, std_chrominance_quant_tbl, - scale_factor, force_baseline); -} - - -GLOBAL(int) -jpeg_quality_scaling (int quality) -/* Convert a user-specified quality rating to a percentage scaling factor - * for an underlying quantization table, using our recommended scaling curve. - * The input 'quality' factor should be 0 (terrible) to 100 (very good). - */ -{ - /* Safety limit on quality factor. Convert 0 to 1 to avoid zero divide. */ - if (quality <= 0) quality = 1; - if (quality > 100) quality = 100; - - /* The basic table is used as-is (scaling 100) for a quality of 50. - * Qualities 50..100 are converted to scaling percentage 200 - 2*Q; - * note that at Q=100 the scaling is 0, which will cause jpeg_add_quant_table - * to make all the table entries 1 (hence, minimum quantization loss). - * Qualities 1..50 are converted to scaling percentage 5000/Q. - */ - if (quality < 50) - quality = 5000 / quality; - else - quality = 200 - quality*2; - - return quality; -} - - -GLOBAL(void) -jpeg_set_quality (j_compress_ptr cinfo, int quality, boolean force_baseline) -/* Set or change the 'quality' (quantization) setting, using default tables. - * This is the standard quality-adjusting entry point for typical user - * interfaces; only those who want detailed control over quantization tables - * would use the preceding routines directly. - */ -{ - /* Convert user 0-100 rating to percentage scaling */ - quality = jpeg_quality_scaling(quality); - - /* Set up standard quality tables */ - jpeg_set_linear_quality(cinfo, quality, force_baseline); -} - - -/* - * Reset standard Huffman tables - */ - -LOCAL(void) -std_huff_tables (j_compress_ptr cinfo) -{ - if (cinfo->dc_huff_tbl_ptrs[0] != NULL) - (void) jpeg_std_huff_table((j_common_ptr) cinfo, TRUE, 0); - - if (cinfo->ac_huff_tbl_ptrs[0] != NULL) - (void) jpeg_std_huff_table((j_common_ptr) cinfo, FALSE, 0); - - if (cinfo->dc_huff_tbl_ptrs[1] != NULL) - (void) jpeg_std_huff_table((j_common_ptr) cinfo, TRUE, 1); - - if (cinfo->ac_huff_tbl_ptrs[1] != NULL) - (void) jpeg_std_huff_table((j_common_ptr) cinfo, FALSE, 1); -} - - -/* - * Default parameter setup for compression. - * - * Applications that don't choose to use this routine must do their - * own setup of all these parameters. Alternately, you can call this - * to establish defaults and then alter parameters selectively. This - * is the recommended approach since, if we add any new parameters, - * your code will still work (they'll be set to reasonable defaults). - */ - -GLOBAL(void) -jpeg_set_defaults (j_compress_ptr cinfo) -{ - int i; - - /* Safety check to ensure start_compress not called yet. */ - if (cinfo->global_state != CSTATE_START) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - - /* Allocate comp_info array large enough for maximum component count. - * Array is made permanent in case application wants to compress - * multiple images at same param settings. - */ - if (cinfo->comp_info == NULL) - cinfo->comp_info = (jpeg_component_info *) - (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_PERMANENT, - MAX_COMPONENTS * SIZEOF(jpeg_component_info)); - - /* Initialize everything not dependent on the color space */ - - cinfo->scale_num = 1; /* 1:1 scaling */ - cinfo->scale_denom = 1; - cinfo->data_precision = BITS_IN_JSAMPLE; - /* Set up two quantization tables using default quality of 75 */ - jpeg_set_quality(cinfo, 75, TRUE); - /* Reset standard Huffman tables */ - std_huff_tables(cinfo); - - /* Initialize default arithmetic coding conditioning */ - for (i = 0; i < NUM_ARITH_TBLS; i++) { - cinfo->arith_dc_L[i] = 0; - cinfo->arith_dc_U[i] = 1; - cinfo->arith_ac_K[i] = 5; - } - - /* Default is no multiple-scan output */ - cinfo->scan_info = NULL; - cinfo->num_scans = 0; - - /* Expect normal source image, not raw downsampled data */ - cinfo->raw_data_in = FALSE; - - /* The standard Huffman tables are only valid for 8-bit data precision. - * If the precision is higher, use arithmetic coding. - * (Alternatively, using Huffman coding would be possible with forcing - * optimization on so that usable tables will be computed, or by - * supplying default tables that are valid for the desired precision.) - * Otherwise, use Huffman coding by default. - */ - cinfo->arith_code = cinfo->data_precision > 8 ? TRUE : FALSE; - - /* By default, don't do extra passes to optimize entropy coding */ - cinfo->optimize_coding = FALSE; - - /* By default, use the simpler non-cosited sampling alignment */ - cinfo->CCIR601_sampling = FALSE; - - /* By default, apply fancy downsampling */ - cinfo->do_fancy_downsampling = TRUE; - - /* No input smoothing */ - cinfo->smoothing_factor = 0; - - /* DCT algorithm preference */ - cinfo->dct_method = JDCT_DEFAULT; - - /* No restart markers */ - cinfo->restart_interval = 0; - cinfo->restart_in_rows = 0; - - /* Fill in default JFIF marker parameters. Note that whether the marker - * will actually be written is determined by jpeg_set_colorspace. - * - * By default, the library emits JFIF version code 1.01. - * An application that wants to emit JFIF 1.02 extension markers should set - * JFIF_minor_version to 2. We could probably get away with just defaulting - * to 1.02, but there may still be some decoders in use that will complain - * about that; saying 1.01 should minimize compatibility problems. - * - * For wide gamut colorspaces (BG_RGB and BG_YCC), the major version will be - * overridden by jpeg_set_colorspace and set to 2. - */ - cinfo->JFIF_major_version = 1; /* Default JFIF version = 1.01 */ - cinfo->JFIF_minor_version = 1; - cinfo->density_unit = 0; /* Pixel size is unknown by default */ - cinfo->X_density = 1; /* Pixel aspect ratio is square by default */ - cinfo->Y_density = 1; - - /* No color transform */ - cinfo->color_transform = JCT_NONE; - - /* Choose JPEG colorspace based on input space, set defaults accordingly */ - - jpeg_default_colorspace(cinfo); -} - - -/* - * Select an appropriate JPEG colorspace for in_color_space. - */ - -GLOBAL(void) -jpeg_default_colorspace (j_compress_ptr cinfo) -{ - switch (cinfo->in_color_space) { - case JCS_UNKNOWN: - jpeg_set_colorspace(cinfo, JCS_UNKNOWN); - break; - case JCS_GRAYSCALE: - jpeg_set_colorspace(cinfo, JCS_GRAYSCALE); - break; - case JCS_RGB: - jpeg_set_colorspace(cinfo, JCS_YCbCr); - break; - case JCS_YCbCr: - jpeg_set_colorspace(cinfo, JCS_YCbCr); - break; - case JCS_CMYK: - jpeg_set_colorspace(cinfo, JCS_CMYK); /* By default, no translation */ - break; - case JCS_YCCK: - jpeg_set_colorspace(cinfo, JCS_YCCK); - break; - case JCS_BG_RGB: - /* No translation for now -- conversion to BG_YCC not yet supportet */ - jpeg_set_colorspace(cinfo, JCS_BG_RGB); - break; - case JCS_BG_YCC: - jpeg_set_colorspace(cinfo, JCS_BG_YCC); - break; - default: - ERREXIT(cinfo, JERR_BAD_IN_COLORSPACE); - } -} - - -/* - * Set the JPEG colorspace, and choose colorspace-dependent default values. - */ - -GLOBAL(void) -jpeg_set_colorspace (j_compress_ptr cinfo, J_COLOR_SPACE colorspace) -{ - jpeg_component_info * compptr; - int ci; - -#define SET_COMP(index,id,hsamp,vsamp,quant,dctbl,actbl) \ - (compptr = &cinfo->comp_info[index], \ - compptr->component_id = (id), \ - compptr->h_samp_factor = (hsamp), \ - compptr->v_samp_factor = (vsamp), \ - compptr->quant_tbl_no = (quant), \ - compptr->dc_tbl_no = (dctbl), \ - compptr->ac_tbl_no = (actbl) ) - - /* Safety check to ensure start_compress not called yet. */ - if (cinfo->global_state != CSTATE_START) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - - /* For all colorspaces, we use Q and Huff tables 0 for luminance components, - * tables 1 for chrominance components. - */ - - cinfo->jpeg_color_space = colorspace; - - cinfo->write_JFIF_header = FALSE; /* No marker for non-JFIF colorspaces */ - cinfo->write_Adobe_marker = FALSE; /* write no Adobe marker by default */ - - switch (colorspace) { - case JCS_UNKNOWN: - cinfo->num_components = cinfo->input_components; - if (cinfo->num_components < 1 || cinfo->num_components > MAX_COMPONENTS) - ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->num_components, - MAX_COMPONENTS); - for (ci = 0; ci < cinfo->num_components; ci++) { - SET_COMP(ci, ci, 1,1, 0, 0,0); - } - break; - case JCS_GRAYSCALE: - cinfo->write_JFIF_header = TRUE; /* Write a JFIF marker */ - cinfo->num_components = 1; - /* JFIF specifies component ID 1 */ - SET_COMP(0, 0x01, 1,1, 0, 0,0); - break; - case JCS_RGB: - cinfo->write_Adobe_marker = TRUE; /* write Adobe marker to flag RGB */ - cinfo->num_components = 3; - SET_COMP(0, 0x52 /* 'R' */, 1,1, - cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0, - cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0, - cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0); - SET_COMP(1, 0x47 /* 'G' */, 1,1, 0, 0,0); - SET_COMP(2, 0x42 /* 'B' */, 1,1, - cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0, - cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0, - cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0); - break; - case JCS_YCbCr: - cinfo->write_JFIF_header = TRUE; /* Write a JFIF marker */ - cinfo->num_components = 3; - /* JFIF specifies component IDs 1,2,3 */ - /* We default to 2x2 subsamples of chrominance */ - SET_COMP(0, 0x01, 2,2, 0, 0,0); - SET_COMP(1, 0x02, 1,1, 1, 1,1); - SET_COMP(2, 0x03, 1,1, 1, 1,1); - break; - case JCS_CMYK: - cinfo->write_Adobe_marker = TRUE; /* write Adobe marker to flag CMYK */ - cinfo->num_components = 4; - SET_COMP(0, 0x43 /* 'C' */, 1,1, 0, 0,0); - SET_COMP(1, 0x4D /* 'M' */, 1,1, 0, 0,0); - SET_COMP(2, 0x59 /* 'Y' */, 1,1, 0, 0,0); - SET_COMP(3, 0x4B /* 'K' */, 1,1, 0, 0,0); - break; - case JCS_YCCK: - cinfo->write_Adobe_marker = TRUE; /* write Adobe marker to flag YCCK */ - cinfo->num_components = 4; - SET_COMP(0, 0x01, 2,2, 0, 0,0); - SET_COMP(1, 0x02, 1,1, 1, 1,1); - SET_COMP(2, 0x03, 1,1, 1, 1,1); - SET_COMP(3, 0x04, 2,2, 0, 0,0); - break; - case JCS_BG_RGB: - cinfo->write_JFIF_header = TRUE; /* Write a JFIF marker */ - cinfo->JFIF_major_version = 2; /* Set JFIF major version = 2 */ - cinfo->num_components = 3; - /* Add offset 0x20 to the normal R/G/B component IDs */ - SET_COMP(0, 0x72 /* 'r' */, 1,1, - cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0, - cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0, - cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0); - SET_COMP(1, 0x67 /* 'g' */, 1,1, 0, 0,0); - SET_COMP(2, 0x62 /* 'b' */, 1,1, - cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0, - cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0, - cinfo->color_transform == JCT_SUBTRACT_GREEN ? 1 : 0); - break; - case JCS_BG_YCC: - cinfo->write_JFIF_header = TRUE; /* Write a JFIF marker */ - cinfo->JFIF_major_version = 2; /* Set JFIF major version = 2 */ - cinfo->num_components = 3; - /* Add offset 0x20 to the normal Cb/Cr component IDs */ - /* We default to 2x2 subsamples of chrominance */ - SET_COMP(0, 0x01, 2,2, 0, 0,0); - SET_COMP(1, 0x22, 1,1, 1, 1,1); - SET_COMP(2, 0x23, 1,1, 1, 1,1); - break; - default: - ERREXIT(cinfo, JERR_BAD_J_COLORSPACE); - } -} - - -#ifdef C_PROGRESSIVE_SUPPORTED - -LOCAL(jpeg_scan_info *) -fill_a_scan (jpeg_scan_info * scanptr, int ci, - int Ss, int Se, int Ah, int Al) -/* Support routine: generate one scan for specified component */ -{ - scanptr->comps_in_scan = 1; - scanptr->component_index[0] = ci; - scanptr->Ss = Ss; - scanptr->Se = Se; - scanptr->Ah = Ah; - scanptr->Al = Al; - scanptr++; - return scanptr; -} - -LOCAL(jpeg_scan_info *) -fill_scans (jpeg_scan_info * scanptr, int ncomps, - int Ss, int Se, int Ah, int Al) -/* Support routine: generate one scan for each component */ -{ - int ci; - - for (ci = 0; ci < ncomps; ci++) { - scanptr->comps_in_scan = 1; - scanptr->component_index[0] = ci; - scanptr->Ss = Ss; - scanptr->Se = Se; - scanptr->Ah = Ah; - scanptr->Al = Al; - scanptr++; - } - return scanptr; -} - -LOCAL(jpeg_scan_info *) -fill_dc_scans (jpeg_scan_info * scanptr, int ncomps, int Ah, int Al) -/* Support routine: generate interleaved DC scan if possible, else N scans */ -{ - int ci; - - if (ncomps <= MAX_COMPS_IN_SCAN) { - /* Single interleaved DC scan */ - scanptr->comps_in_scan = ncomps; - for (ci = 0; ci < ncomps; ci++) - scanptr->component_index[ci] = ci; - scanptr->Ss = scanptr->Se = 0; - scanptr->Ah = Ah; - scanptr->Al = Al; - scanptr++; - } else { - /* Noninterleaved DC scan for each component */ - scanptr = fill_scans(scanptr, ncomps, 0, 0, Ah, Al); - } - return scanptr; -} - - -/* - * Create a recommended progressive-JPEG script. - * cinfo->num_components and cinfo->jpeg_color_space must be correct. - */ - -GLOBAL(void) -jpeg_simple_progression (j_compress_ptr cinfo) -{ - int ncomps = cinfo->num_components; - int nscans; - jpeg_scan_info * scanptr; - - /* Safety check to ensure start_compress not called yet. */ - if (cinfo->global_state != CSTATE_START) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - - /* Figure space needed for script. Calculation must match code below! */ - if (ncomps == 3 && - (cinfo->jpeg_color_space == JCS_YCbCr || - cinfo->jpeg_color_space == JCS_BG_YCC)) { - /* Custom script for YCC color images. */ - nscans = 10; - } else { - /* All-purpose script for other color spaces. */ - if (ncomps > MAX_COMPS_IN_SCAN) - nscans = 6 * ncomps; /* 2 DC + 4 AC scans per component */ - else - nscans = 2 + 4 * ncomps; /* 2 DC scans; 4 AC scans per component */ - } - - /* Allocate space for script. - * We need to put it in the permanent pool in case the application performs - * multiple compressions without changing the settings. To avoid a memory - * leak if jpeg_simple_progression is called repeatedly for the same JPEG - * object, we try to re-use previously allocated space, and we allocate - * enough space to handle YCC even if initially asked for grayscale. - */ - if (cinfo->script_space == NULL || cinfo->script_space_size < nscans) { - cinfo->script_space_size = MAX(nscans, 10); - cinfo->script_space = (jpeg_scan_info *) - (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_PERMANENT, - cinfo->script_space_size * SIZEOF(jpeg_scan_info)); - } - scanptr = cinfo->script_space; - cinfo->scan_info = scanptr; - cinfo->num_scans = nscans; - - if (ncomps == 3 && - (cinfo->jpeg_color_space == JCS_YCbCr || - cinfo->jpeg_color_space == JCS_BG_YCC)) { - /* Custom script for YCC color images. */ - /* Initial DC scan */ - scanptr = fill_dc_scans(scanptr, ncomps, 0, 1); - /* Initial AC scan: get some luma data out in a hurry */ - scanptr = fill_a_scan(scanptr, 0, 1, 5, 0, 2); - /* Chroma data is too small to be worth expending many scans on */ - scanptr = fill_a_scan(scanptr, 2, 1, 63, 0, 1); - scanptr = fill_a_scan(scanptr, 1, 1, 63, 0, 1); - /* Complete spectral selection for luma AC */ - scanptr = fill_a_scan(scanptr, 0, 6, 63, 0, 2); - /* Refine next bit of luma AC */ - scanptr = fill_a_scan(scanptr, 0, 1, 63, 2, 1); - /* Finish DC successive approximation */ - scanptr = fill_dc_scans(scanptr, ncomps, 1, 0); - /* Finish AC successive approximation */ - scanptr = fill_a_scan(scanptr, 2, 1, 63, 1, 0); - scanptr = fill_a_scan(scanptr, 1, 1, 63, 1, 0); - /* Luma bottom bit comes last since it's usually largest scan */ - scanptr = fill_a_scan(scanptr, 0, 1, 63, 1, 0); - } else { - /* All-purpose script for other color spaces. */ - /* Successive approximation first pass */ - scanptr = fill_dc_scans(scanptr, ncomps, 0, 1); - scanptr = fill_scans(scanptr, ncomps, 1, 5, 0, 2); - scanptr = fill_scans(scanptr, ncomps, 6, 63, 0, 2); - /* Successive approximation second pass */ - scanptr = fill_scans(scanptr, ncomps, 1, 63, 2, 1); - /* Successive approximation final pass */ - scanptr = fill_dc_scans(scanptr, ncomps, 1, 0); - scanptr = fill_scans(scanptr, ncomps, 1, 63, 1, 0); - } -} - -#endif /* C_PROGRESSIVE_SUPPORTED */ diff --git a/dep/libjpeg/src/jcprepct.c b/dep/libjpeg/src/jcprepct.c deleted file mode 100644 index 586964bd4..000000000 --- a/dep/libjpeg/src/jcprepct.c +++ /dev/null @@ -1,358 +0,0 @@ -/* - * jcprepct.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2003-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains the compression preprocessing controller. - * This controller manages the color conversion, downsampling, - * and edge expansion steps. - * - * Most of the complexity here is associated with buffering input rows - * as required by the downsampler. See the comments at the head of - * jcsample.c for the downsampler's needs. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* At present, jcsample.c can request context rows only for smoothing. - * In the future, we might also need context rows for CCIR601 sampling - * or other more-complex downsampling procedures. The code to support - * context rows should be compiled only if needed. - */ -#ifdef INPUT_SMOOTHING_SUPPORTED -#define CONTEXT_ROWS_SUPPORTED -#endif - - -/* - * For the simple (no-context-row) case, we just need to buffer one - * row group's worth of pixels for the downsampling step. At the bottom of - * the image, we pad to a full row group by replicating the last pixel row. - * The downsampler's last output row is then replicated if needed to pad - * out to a full iMCU row. - * - * When providing context rows, we must buffer three row groups' worth of - * pixels. Three row groups are physically allocated, but the row pointer - * arrays are made five row groups high, with the extra pointers above and - * below "wrapping around" to point to the last and first real row groups. - * This allows the downsampler to access the proper context rows. - * At the top and bottom of the image, we create dummy context rows by - * copying the first or last real pixel row. This copying could be avoided - * by pointer hacking as is done in jdmainct.c, but it doesn't seem worth the - * trouble on the compression side. - */ - - -/* Private buffer controller object */ - -typedef struct { - struct jpeg_c_prep_controller pub; /* public fields */ - - /* Downsampling input buffer. This buffer holds color-converted data - * until we have enough to do a downsample step. - */ - JSAMPARRAY color_buf[MAX_COMPONENTS]; - - JDIMENSION rows_to_go; /* counts rows remaining in source image */ - int next_buf_row; /* index of next row to store in color_buf */ - -#ifdef CONTEXT_ROWS_SUPPORTED /* only needed for context case */ - int this_row_group; /* starting row index of group to process */ - int next_buf_stop; /* downsample when we reach this index */ -#endif -} my_prep_controller; - -typedef my_prep_controller * my_prep_ptr; - - -/* - * Initialize for a processing pass. - */ - -METHODDEF(void) -start_pass_prep (j_compress_ptr cinfo, J_BUF_MODE pass_mode) -{ - my_prep_ptr prep = (my_prep_ptr) cinfo->prep; - - if (pass_mode != JBUF_PASS_THRU) - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - - /* Initialize total-height counter for detecting bottom of image */ - prep->rows_to_go = cinfo->image_height; - /* Mark the conversion buffer empty */ - prep->next_buf_row = 0; -#ifdef CONTEXT_ROWS_SUPPORTED - /* Preset additional state variables for context mode. - * These aren't used in non-context mode, so we needn't test which mode. - */ - prep->this_row_group = 0; - /* Set next_buf_stop to stop after two row groups have been read in. */ - prep->next_buf_stop = 2 * cinfo->max_v_samp_factor; -#endif -} - - -/* - * Expand an image vertically from height input_rows to height output_rows, - * by duplicating the bottom row. - */ - -LOCAL(void) -expand_bottom_edge (JSAMPARRAY image_data, JDIMENSION num_cols, - int input_rows, int output_rows) -{ - register int row; - - for (row = input_rows; row < output_rows; row++) { - jcopy_sample_rows(image_data + input_rows - 1, - image_data + row, - 1, num_cols); - } -} - - -/* - * Process some data in the simple no-context case. - * - * Preprocessor output data is counted in "row groups". A row group - * is defined to be v_samp_factor sample rows of each component. - * Downsampling will produce this much data from each max_v_samp_factor - * input rows. - */ - -METHODDEF(void) -pre_process_data (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JDIMENSION *in_row_ctr, - JDIMENSION in_rows_avail, - JSAMPIMAGE output_buf, JDIMENSION *out_row_group_ctr, - JDIMENSION out_row_groups_avail) -{ - my_prep_ptr prep = (my_prep_ptr) cinfo->prep; - int numrows, ci; - JDIMENSION inrows; - jpeg_component_info * compptr; - - while (*in_row_ctr < in_rows_avail && - *out_row_group_ctr < out_row_groups_avail) { - /* Do color conversion to fill the conversion buffer. */ - inrows = in_rows_avail - *in_row_ctr; - numrows = cinfo->max_v_samp_factor - prep->next_buf_row; - numrows = (int) MIN((JDIMENSION) numrows, inrows); - (*cinfo->cconvert->color_convert) (cinfo, input_buf + *in_row_ctr, - prep->color_buf, - (JDIMENSION) prep->next_buf_row, - numrows); - *in_row_ctr += numrows; - prep->next_buf_row += numrows; - prep->rows_to_go -= numrows; - /* If at bottom of image, pad to fill the conversion buffer. */ - if (prep->rows_to_go == 0 && - prep->next_buf_row < cinfo->max_v_samp_factor) { - for (ci = 0; ci < cinfo->num_components; ci++) { - expand_bottom_edge(prep->color_buf[ci], cinfo->image_width, - prep->next_buf_row, cinfo->max_v_samp_factor); - } - prep->next_buf_row = cinfo->max_v_samp_factor; - } - /* If we've filled the conversion buffer, empty it. */ - if (prep->next_buf_row == cinfo->max_v_samp_factor) { - (*cinfo->downsample->downsample) (cinfo, - prep->color_buf, (JDIMENSION) 0, - output_buf, *out_row_group_ctr); - prep->next_buf_row = 0; - (*out_row_group_ctr)++; - } - /* If at bottom of image, pad the output to a full iMCU height. - * Note we assume the caller is providing a one-iMCU-height output buffer! - */ - if (prep->rows_to_go == 0 && - *out_row_group_ctr < out_row_groups_avail) { - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - numrows = (compptr->v_samp_factor * compptr->DCT_v_scaled_size) / - cinfo->min_DCT_v_scaled_size; - expand_bottom_edge(output_buf[ci], - compptr->width_in_blocks * compptr->DCT_h_scaled_size, - (int) (*out_row_group_ctr * numrows), - (int) (out_row_groups_avail * numrows)); - } - *out_row_group_ctr = out_row_groups_avail; - break; /* can exit outer loop without test */ - } - } -} - - -#ifdef CONTEXT_ROWS_SUPPORTED - -/* - * Process some data in the context case. - */ - -METHODDEF(void) -pre_process_context (j_compress_ptr cinfo, - JSAMPARRAY input_buf, JDIMENSION *in_row_ctr, - JDIMENSION in_rows_avail, - JSAMPIMAGE output_buf, JDIMENSION *out_row_group_ctr, - JDIMENSION out_row_groups_avail) -{ - my_prep_ptr prep = (my_prep_ptr) cinfo->prep; - int numrows, ci; - int buf_height = cinfo->max_v_samp_factor * 3; - JDIMENSION inrows; - - while (*out_row_group_ctr < out_row_groups_avail) { - if (*in_row_ctr < in_rows_avail) { - /* Do color conversion to fill the conversion buffer. */ - inrows = in_rows_avail - *in_row_ctr; - numrows = prep->next_buf_stop - prep->next_buf_row; - numrows = (int) MIN((JDIMENSION) numrows, inrows); - (*cinfo->cconvert->color_convert) (cinfo, input_buf + *in_row_ctr, - prep->color_buf, - (JDIMENSION) prep->next_buf_row, - numrows); - /* Pad at top of image, if first time through */ - if (prep->rows_to_go == cinfo->image_height) { - for (ci = 0; ci < cinfo->num_components; ci++) { - int row; - for (row = 1; row <= cinfo->max_v_samp_factor; row++) { - jcopy_sample_rows(prep->color_buf[ci], - prep->color_buf[ci] - row, - 1, cinfo->image_width); - } - } - } - *in_row_ctr += numrows; - prep->next_buf_row += numrows; - prep->rows_to_go -= numrows; - } else { - /* Return for more data, unless we are at the bottom of the image. */ - if (prep->rows_to_go != 0) - break; - /* When at bottom of image, pad to fill the conversion buffer. */ - if (prep->next_buf_row < prep->next_buf_stop) { - for (ci = 0; ci < cinfo->num_components; ci++) { - expand_bottom_edge(prep->color_buf[ci], cinfo->image_width, - prep->next_buf_row, prep->next_buf_stop); - } - prep->next_buf_row = prep->next_buf_stop; - } - } - /* If we've gotten enough data, downsample a row group. */ - if (prep->next_buf_row == prep->next_buf_stop) { - (*cinfo->downsample->downsample) (cinfo, - prep->color_buf, - (JDIMENSION) prep->this_row_group, - output_buf, *out_row_group_ctr); - (*out_row_group_ctr)++; - /* Advance pointers with wraparound as necessary. */ - prep->this_row_group += cinfo->max_v_samp_factor; - if (prep->this_row_group >= buf_height) - prep->this_row_group = 0; - if (prep->next_buf_row >= buf_height) - prep->next_buf_row = 0; - prep->next_buf_stop = prep->next_buf_row + cinfo->max_v_samp_factor; - } - } -} - - -/* - * Create the wrapped-around downsampling input buffer needed for context mode. - */ - -LOCAL(void) -create_context_buffer (j_compress_ptr cinfo) -{ - my_prep_ptr prep = (my_prep_ptr) cinfo->prep; - int rgroup_height = cinfo->max_v_samp_factor; - int ci, i; - jpeg_component_info * compptr; - JSAMPARRAY true_buffer, fake_buffer; - - /* Grab enough space for fake row pointers for all the components; - * we need five row groups' worth of pointers for each component. - */ - fake_buffer = (JSAMPARRAY) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - (cinfo->num_components * 5 * rgroup_height) * SIZEOF(JSAMPROW)); - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* Allocate the actual buffer space (3 row groups) for this component. - * We make the buffer wide enough to allow the downsampler to edge-expand - * horizontally within the buffer, if it so chooses. - */ - true_buffer = (*cinfo->mem->alloc_sarray) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - (JDIMENSION) (((long) compptr->width_in_blocks * - cinfo->min_DCT_h_scaled_size * - cinfo->max_h_samp_factor) / compptr->h_samp_factor), - (JDIMENSION) (3 * rgroup_height)); - /* Copy true buffer row pointers into the middle of the fake row array */ - MEMCOPY(fake_buffer + rgroup_height, true_buffer, - 3 * rgroup_height * SIZEOF(JSAMPROW)); - /* Fill in the above and below wraparound pointers */ - for (i = 0; i < rgroup_height; i++) { - fake_buffer[i] = true_buffer[2 * rgroup_height + i]; - fake_buffer[4 * rgroup_height + i] = true_buffer[i]; - } - prep->color_buf[ci] = fake_buffer + rgroup_height; - fake_buffer += 5 * rgroup_height; /* point to space for next component */ - } -} - -#endif /* CONTEXT_ROWS_SUPPORTED */ - - -/* - * Initialize preprocessing controller. - */ - -GLOBAL(void) -jinit_c_prep_controller (j_compress_ptr cinfo, boolean need_full_buffer) -{ - my_prep_ptr prep; - int ci; - jpeg_component_info * compptr; - - if (need_full_buffer) /* safety check */ - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - - prep = (my_prep_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_prep_controller)); - cinfo->prep = &prep->pub; - prep->pub.start_pass = start_pass_prep; - - /* Allocate the color conversion buffer. - * We make the buffer wide enough to allow the downsampler to edge-expand - * horizontally within the buffer, if it so chooses. - */ - if (cinfo->downsample->need_context_rows) { - /* Set up to provide context rows */ -#ifdef CONTEXT_ROWS_SUPPORTED - prep->pub.pre_process_data = pre_process_context; - create_context_buffer(cinfo); -#else - ERREXIT(cinfo, JERR_NOT_COMPILED); -#endif - } else { - /* No context, just make it tall enough for one row group */ - prep->pub.pre_process_data = pre_process_data; - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - prep->color_buf[ci] = (*cinfo->mem->alloc_sarray) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - (JDIMENSION) (((long) compptr->width_in_blocks * - cinfo->min_DCT_h_scaled_size * - cinfo->max_h_samp_factor) / compptr->h_samp_factor), - (JDIMENSION) cinfo->max_v_samp_factor); - } - } -} diff --git a/dep/libjpeg/src/jcsample.c b/dep/libjpeg/src/jcsample.c deleted file mode 100644 index 2372c4173..000000000 --- a/dep/libjpeg/src/jcsample.c +++ /dev/null @@ -1,545 +0,0 @@ -/* - * jcsample.c - * - * Copyright (C) 1991-1996, Thomas G. Lane. - * Modified 2003-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains downsampling routines. - * - * Downsampling input data is counted in "row groups". A row group - * is defined to be max_v_samp_factor pixel rows of each component, - * from which the downsampler produces v_samp_factor sample rows. - * A single row group is processed in each call to the downsampler module. - * - * The downsampler is responsible for edge-expansion of its output data - * to fill an integral number of DCT blocks horizontally. The source buffer - * may be modified if it is helpful for this purpose (the source buffer is - * allocated wide enough to correspond to the desired output width). - * The caller (the prep controller) is responsible for vertical padding. - * - * The downsampler may request "context rows" by setting need_context_rows - * during startup. In this case, the input arrays will contain at least - * one row group's worth of pixels above and below the passed-in data; - * the caller will create dummy rows at image top and bottom by replicating - * the first or last real pixel row. - * - * An excellent reference for image resampling is - * Digital Image Warping, George Wolberg, 1990. - * Pub. by IEEE Computer Society Press, Los Alamitos, CA. ISBN 0-8186-8944-7. - * - * The downsampling algorithm used here is a simple average of the source - * pixels covered by the output pixel. The hi-falutin sampling literature - * refers to this as a "box filter". In general the characteristics of a box - * filter are not very good, but for the specific cases we normally use (1:1 - * and 2:1 ratios) the box is equivalent to a "triangle filter" which is not - * nearly so bad. If you intend to use other sampling ratios, you'd be well - * advised to improve this code. - * - * A simple input-smoothing capability is provided. This is mainly intended - * for cleaning up color-dithered GIF input files (if you find it inadequate, - * we suggest using an external filtering program such as pnmconvol). When - * enabled, each input pixel P is replaced by a weighted sum of itself and its - * eight neighbors. P's weight is 1-8*SF and each neighbor's weight is SF, - * where SF = (smoothing_factor / 1024). - * Currently, smoothing is only supported for 2h2v sampling factors. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Pointer to routine to downsample a single component */ -typedef JMETHOD(void, downsample1_ptr, - (j_compress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY input_data, JSAMPARRAY output_data)); - -/* Private subobject */ - -typedef struct { - struct jpeg_downsampler pub; /* public fields */ - - /* Downsampling method pointers, one per component */ - downsample1_ptr methods[MAX_COMPONENTS]; - - /* Height of an output row group for each component. */ - int rowgroup_height[MAX_COMPONENTS]; - - /* These arrays save pixel expansion factors so that int_downsample need not - * recompute them each time. They are unused for other downsampling methods. - */ - UINT8 h_expand[MAX_COMPONENTS]; - UINT8 v_expand[MAX_COMPONENTS]; -} my_downsampler; - -typedef my_downsampler * my_downsample_ptr; - - -/* - * Initialize for a downsampling pass. - */ - -METHODDEF(void) -start_pass_downsample (j_compress_ptr cinfo) -{ - /* no work for now */ -} - - -/* - * Expand a component horizontally from width input_cols to width output_cols, - * by duplicating the rightmost samples. - */ - -LOCAL(void) -expand_right_edge (JSAMPARRAY image_data, int num_rows, - JDIMENSION input_cols, JDIMENSION output_cols) -{ - register JSAMPROW ptr; - register JSAMPLE pixval; - register int count; - int row; - int numcols = (int) (output_cols - input_cols); - - if (numcols > 0) { - for (row = 0; row < num_rows; row++) { - ptr = image_data[row] + input_cols; - pixval = ptr[-1]; /* don't need GETJSAMPLE() here */ - for (count = numcols; count > 0; count--) - *ptr++ = pixval; - } - } -} - - -/* - * Do downsampling for a whole row group (all components). - * - * In this version we simply downsample each component independently. - */ - -METHODDEF(void) -sep_downsample (j_compress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION in_row_index, - JSAMPIMAGE output_buf, JDIMENSION out_row_group_index) -{ - my_downsample_ptr downsample = (my_downsample_ptr) cinfo->downsample; - int ci; - jpeg_component_info * compptr; - JSAMPARRAY in_ptr, out_ptr; - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - in_ptr = input_buf[ci] + in_row_index; - out_ptr = output_buf[ci] + - (out_row_group_index * downsample->rowgroup_height[ci]); - (*downsample->methods[ci]) (cinfo, compptr, in_ptr, out_ptr); - } -} - - -/* - * Downsample pixel values of a single component. - * One row group is processed per call. - * This version handles arbitrary integral sampling ratios, without smoothing. - * Note that this version is not actually used for customary sampling ratios. - */ - -METHODDEF(void) -int_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY input_data, JSAMPARRAY output_data) -{ - my_downsample_ptr downsample = (my_downsample_ptr) cinfo->downsample; - int inrow, outrow, h_expand, v_expand, numpix, numpix2, h, v; - JDIMENSION outcol, outcol_h; /* outcol_h == outcol*h_expand */ - JDIMENSION output_cols = compptr->width_in_blocks * compptr->DCT_h_scaled_size; - JSAMPROW inptr, outptr; - INT32 outvalue; - - h_expand = downsample->h_expand[compptr->component_index]; - v_expand = downsample->v_expand[compptr->component_index]; - numpix = h_expand * v_expand; - numpix2 = numpix/2; - - /* Expand input data enough to let all the output samples be generated - * by the standard loop. Special-casing padded output would be more - * efficient. - */ - expand_right_edge(input_data, cinfo->max_v_samp_factor, - cinfo->image_width, output_cols * h_expand); - - inrow = outrow = 0; - while (inrow < cinfo->max_v_samp_factor) { - outptr = output_data[outrow]; - for (outcol = 0, outcol_h = 0; outcol < output_cols; - outcol++, outcol_h += h_expand) { - outvalue = 0; - for (v = 0; v < v_expand; v++) { - inptr = input_data[inrow+v] + outcol_h; - for (h = 0; h < h_expand; h++) { - outvalue += (INT32) GETJSAMPLE(*inptr++); - } - } - *outptr++ = (JSAMPLE) ((outvalue + numpix2) / numpix); - } - inrow += v_expand; - outrow++; - } -} - - -/* - * Downsample pixel values of a single component. - * This version handles the special case of a full-size component, - * without smoothing. - */ - -METHODDEF(void) -fullsize_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY input_data, JSAMPARRAY output_data) -{ - /* Copy the data */ - jcopy_sample_rows(input_data, output_data, - cinfo->max_v_samp_factor, cinfo->image_width); - /* Edge-expand */ - expand_right_edge(output_data, cinfo->max_v_samp_factor, cinfo->image_width, - compptr->width_in_blocks * compptr->DCT_h_scaled_size); -} - - -/* - * Downsample pixel values of a single component. - * This version handles the common case of 2:1 horizontal and 1:1 vertical, - * without smoothing. - * - * A note about the "bias" calculations: when rounding fractional values to - * integer, we do not want to always round 0.5 up to the next integer. - * If we did that, we'd introduce a noticeable bias towards larger values. - * Instead, this code is arranged so that 0.5 will be rounded up or down at - * alternate pixel locations (a simple ordered dither pattern). - */ - -METHODDEF(void) -h2v1_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY input_data, JSAMPARRAY output_data) -{ - int inrow; - JDIMENSION outcol; - JDIMENSION output_cols = compptr->width_in_blocks * compptr->DCT_h_scaled_size; - register JSAMPROW inptr, outptr; - register int bias; - - /* Expand input data enough to let all the output samples be generated - * by the standard loop. Special-casing padded output would be more - * efficient. - */ - expand_right_edge(input_data, cinfo->max_v_samp_factor, - cinfo->image_width, output_cols * 2); - - for (inrow = 0; inrow < cinfo->max_v_samp_factor; inrow++) { - outptr = output_data[inrow]; - inptr = input_data[inrow]; - bias = 0; /* bias = 0,1,0,1,... for successive samples */ - for (outcol = 0; outcol < output_cols; outcol++) { - *outptr++ = (JSAMPLE) ((GETJSAMPLE(*inptr) + GETJSAMPLE(inptr[1]) - + bias) >> 1); - bias ^= 1; /* 0=>1, 1=>0 */ - inptr += 2; - } - } -} - - -/* - * Downsample pixel values of a single component. - * This version handles the standard case of 2:1 horizontal and 2:1 vertical, - * without smoothing. - */ - -METHODDEF(void) -h2v2_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY input_data, JSAMPARRAY output_data) -{ - int inrow, outrow; - JDIMENSION outcol; - JDIMENSION output_cols = compptr->width_in_blocks * compptr->DCT_h_scaled_size; - register JSAMPROW inptr0, inptr1, outptr; - register int bias; - - /* Expand input data enough to let all the output samples be generated - * by the standard loop. Special-casing padded output would be more - * efficient. - */ - expand_right_edge(input_data, cinfo->max_v_samp_factor, - cinfo->image_width, output_cols * 2); - - inrow = outrow = 0; - while (inrow < cinfo->max_v_samp_factor) { - outptr = output_data[outrow]; - inptr0 = input_data[inrow]; - inptr1 = input_data[inrow+1]; - bias = 1; /* bias = 1,2,1,2,... for successive samples */ - for (outcol = 0; outcol < output_cols; outcol++) { - *outptr++ = (JSAMPLE) ((GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[1]) + - GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]) - + bias) >> 2); - bias ^= 3; /* 1=>2, 2=>1 */ - inptr0 += 2; inptr1 += 2; - } - inrow += 2; - outrow++; - } -} - - -#ifdef INPUT_SMOOTHING_SUPPORTED - -/* - * Downsample pixel values of a single component. - * This version handles the standard case of 2:1 horizontal and 2:1 vertical, - * with smoothing. One row of context is required. - */ - -METHODDEF(void) -h2v2_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY input_data, JSAMPARRAY output_data) -{ - int inrow, outrow; - JDIMENSION colctr; - JDIMENSION output_cols = compptr->width_in_blocks * compptr->DCT_h_scaled_size; - register JSAMPROW inptr0, inptr1, above_ptr, below_ptr, outptr; - INT32 membersum, neighsum, memberscale, neighscale; - - /* Expand input data enough to let all the output samples be generated - * by the standard loop. Special-casing padded output would be more - * efficient. - */ - expand_right_edge(input_data - 1, cinfo->max_v_samp_factor + 2, - cinfo->image_width, output_cols * 2); - - /* We don't bother to form the individual "smoothed" input pixel values; - * we can directly compute the output which is the average of the four - * smoothed values. Each of the four member pixels contributes a fraction - * (1-8*SF) to its own smoothed image and a fraction SF to each of the three - * other smoothed pixels, therefore a total fraction (1-5*SF)/4 to the final - * output. The four corner-adjacent neighbor pixels contribute a fraction - * SF to just one smoothed pixel, or SF/4 to the final output; while the - * eight edge-adjacent neighbors contribute SF to each of two smoothed - * pixels, or SF/2 overall. In order to use integer arithmetic, these - * factors are scaled by 2^16 = 65536. - * Also recall that SF = smoothing_factor / 1024. - */ - - memberscale = 16384 - cinfo->smoothing_factor * 80; /* scaled (1-5*SF)/4 */ - neighscale = cinfo->smoothing_factor * 16; /* scaled SF/4 */ - - inrow = outrow = 0; - while (inrow < cinfo->max_v_samp_factor) { - outptr = output_data[outrow]; - inptr0 = input_data[inrow]; - inptr1 = input_data[inrow+1]; - above_ptr = input_data[inrow-1]; - below_ptr = input_data[inrow+2]; - - /* Special case for first column: pretend column -1 is same as column 0 */ - membersum = GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[1]) + - GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]); - neighsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(above_ptr[1]) + - GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[1]) + - GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[2]) + - GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[2]); - neighsum += neighsum; - neighsum += GETJSAMPLE(*above_ptr) + GETJSAMPLE(above_ptr[2]) + - GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[2]); - membersum = membersum * memberscale + neighsum * neighscale; - *outptr++ = (JSAMPLE) ((membersum + 32768) >> 16); - inptr0 += 2; inptr1 += 2; above_ptr += 2; below_ptr += 2; - - for (colctr = output_cols - 2; colctr > 0; colctr--) { - /* sum of pixels directly mapped to this output element */ - membersum = GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[1]) + - GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]); - /* sum of edge-neighbor pixels */ - neighsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(above_ptr[1]) + - GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[1]) + - GETJSAMPLE(inptr0[-1]) + GETJSAMPLE(inptr0[2]) + - GETJSAMPLE(inptr1[-1]) + GETJSAMPLE(inptr1[2]); - /* The edge-neighbors count twice as much as corner-neighbors */ - neighsum += neighsum; - /* Add in the corner-neighbors */ - neighsum += GETJSAMPLE(above_ptr[-1]) + GETJSAMPLE(above_ptr[2]) + - GETJSAMPLE(below_ptr[-1]) + GETJSAMPLE(below_ptr[2]); - /* form final output scaled up by 2^16 */ - membersum = membersum * memberscale + neighsum * neighscale; - /* round, descale and output it */ - *outptr++ = (JSAMPLE) ((membersum + 32768) >> 16); - inptr0 += 2; inptr1 += 2; above_ptr += 2; below_ptr += 2; - } - - /* Special case for last column */ - membersum = GETJSAMPLE(*inptr0) + GETJSAMPLE(inptr0[1]) + - GETJSAMPLE(*inptr1) + GETJSAMPLE(inptr1[1]); - neighsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(above_ptr[1]) + - GETJSAMPLE(*below_ptr) + GETJSAMPLE(below_ptr[1]) + - GETJSAMPLE(inptr0[-1]) + GETJSAMPLE(inptr0[1]) + - GETJSAMPLE(inptr1[-1]) + GETJSAMPLE(inptr1[1]); - neighsum += neighsum; - neighsum += GETJSAMPLE(above_ptr[-1]) + GETJSAMPLE(above_ptr[1]) + - GETJSAMPLE(below_ptr[-1]) + GETJSAMPLE(below_ptr[1]); - membersum = membersum * memberscale + neighsum * neighscale; - *outptr = (JSAMPLE) ((membersum + 32768) >> 16); - - inrow += 2; - outrow++; - } -} - - -/* - * Downsample pixel values of a single component. - * This version handles the special case of a full-size component, - * with smoothing. One row of context is required. - */ - -METHODDEF(void) -fullsize_smooth_downsample (j_compress_ptr cinfo, jpeg_component_info *compptr, - JSAMPARRAY input_data, JSAMPARRAY output_data) -{ - int inrow; - JDIMENSION colctr; - JDIMENSION output_cols = compptr->width_in_blocks * compptr->DCT_h_scaled_size; - register JSAMPROW inptr, above_ptr, below_ptr, outptr; - INT32 membersum, neighsum, memberscale, neighscale; - int colsum, lastcolsum, nextcolsum; - - /* Expand input data enough to let all the output samples be generated - * by the standard loop. Special-casing padded output would be more - * efficient. - */ - expand_right_edge(input_data - 1, cinfo->max_v_samp_factor + 2, - cinfo->image_width, output_cols); - - /* Each of the eight neighbor pixels contributes a fraction SF to the - * smoothed pixel, while the main pixel contributes (1-8*SF). In order - * to use integer arithmetic, these factors are multiplied by 2^16 = 65536. - * Also recall that SF = smoothing_factor / 1024. - */ - - memberscale = 65536L - cinfo->smoothing_factor * 512L; /* scaled 1-8*SF */ - neighscale = cinfo->smoothing_factor * 64; /* scaled SF */ - - for (inrow = 0; inrow < cinfo->max_v_samp_factor; inrow++) { - outptr = output_data[inrow]; - inptr = input_data[inrow]; - above_ptr = input_data[inrow-1]; - below_ptr = input_data[inrow+1]; - - /* Special case for first column */ - colsum = GETJSAMPLE(*above_ptr++) + GETJSAMPLE(*below_ptr++) + - GETJSAMPLE(*inptr); - membersum = GETJSAMPLE(*inptr++); - nextcolsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(*below_ptr) + - GETJSAMPLE(*inptr); - neighsum = colsum + (colsum - membersum) + nextcolsum; - membersum = membersum * memberscale + neighsum * neighscale; - *outptr++ = (JSAMPLE) ((membersum + 32768) >> 16); - lastcolsum = colsum; colsum = nextcolsum; - - for (colctr = output_cols - 2; colctr > 0; colctr--) { - membersum = GETJSAMPLE(*inptr++); - above_ptr++; below_ptr++; - nextcolsum = GETJSAMPLE(*above_ptr) + GETJSAMPLE(*below_ptr) + - GETJSAMPLE(*inptr); - neighsum = lastcolsum + (colsum - membersum) + nextcolsum; - membersum = membersum * memberscale + neighsum * neighscale; - *outptr++ = (JSAMPLE) ((membersum + 32768) >> 16); - lastcolsum = colsum; colsum = nextcolsum; - } - - /* Special case for last column */ - membersum = GETJSAMPLE(*inptr); - neighsum = lastcolsum + (colsum - membersum) + colsum; - membersum = membersum * memberscale + neighsum * neighscale; - *outptr = (JSAMPLE) ((membersum + 32768) >> 16); - - } -} - -#endif /* INPUT_SMOOTHING_SUPPORTED */ - - -/* - * Module initialization routine for downsampling. - * Note that we must select a routine for each component. - */ - -GLOBAL(void) -jinit_downsampler (j_compress_ptr cinfo) -{ - my_downsample_ptr downsample; - int ci; - jpeg_component_info * compptr; - boolean smoothok = TRUE; - int h_in_group, v_in_group, h_out_group, v_out_group; - - downsample = (my_downsample_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_downsampler)); - cinfo->downsample = &downsample->pub; - downsample->pub.start_pass = start_pass_downsample; - downsample->pub.downsample = sep_downsample; - downsample->pub.need_context_rows = FALSE; - - if (cinfo->CCIR601_sampling) - ERREXIT(cinfo, JERR_CCIR601_NOTIMPL); - - /* Verify we can handle the sampling factors, and set up method pointers */ - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* Compute size of an "output group" for DCT scaling. This many samples - * are to be converted from max_h_samp_factor * max_v_samp_factor pixels. - */ - h_out_group = (compptr->h_samp_factor * compptr->DCT_h_scaled_size) / - cinfo->min_DCT_h_scaled_size; - v_out_group = (compptr->v_samp_factor * compptr->DCT_v_scaled_size) / - cinfo->min_DCT_v_scaled_size; - h_in_group = cinfo->max_h_samp_factor; - v_in_group = cinfo->max_v_samp_factor; - downsample->rowgroup_height[ci] = v_out_group; /* save for use later */ - if (h_in_group == h_out_group && v_in_group == v_out_group) { -#ifdef INPUT_SMOOTHING_SUPPORTED - if (cinfo->smoothing_factor) { - downsample->methods[ci] = fullsize_smooth_downsample; - downsample->pub.need_context_rows = TRUE; - } else -#endif - downsample->methods[ci] = fullsize_downsample; - } else if (h_in_group == h_out_group * 2 && - v_in_group == v_out_group) { - smoothok = FALSE; - downsample->methods[ci] = h2v1_downsample; - } else if (h_in_group == h_out_group * 2 && - v_in_group == v_out_group * 2) { -#ifdef INPUT_SMOOTHING_SUPPORTED - if (cinfo->smoothing_factor) { - downsample->methods[ci] = h2v2_smooth_downsample; - downsample->pub.need_context_rows = TRUE; - } else -#endif - downsample->methods[ci] = h2v2_downsample; - } else if ((h_in_group % h_out_group) == 0 && - (v_in_group % v_out_group) == 0) { - smoothok = FALSE; - downsample->methods[ci] = int_downsample; - downsample->h_expand[ci] = (UINT8) (h_in_group / h_out_group); - downsample->v_expand[ci] = (UINT8) (v_in_group / v_out_group); - } else - ERREXIT(cinfo, JERR_FRACT_SAMPLE_NOTIMPL); - } - -#ifdef INPUT_SMOOTHING_SUPPORTED - if (cinfo->smoothing_factor && !smoothok) - TRACEMS(cinfo, 0, JTRC_SMOOTH_NOTIMPL); -#endif -} diff --git a/dep/libjpeg/src/jctrans.c b/dep/libjpeg/src/jctrans.c deleted file mode 100644 index 261dd2996..000000000 --- a/dep/libjpeg/src/jctrans.c +++ /dev/null @@ -1,399 +0,0 @@ -/* - * jctrans.c - * - * Copyright (C) 1995-1998, Thomas G. Lane. - * Modified 2000-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains library routines for transcoding compression, - * that is, writing raw DCT coefficient arrays to an output JPEG file. - * The routines in jcapimin.c will also be needed by a transcoder. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Forward declarations */ -LOCAL(void) transencode_master_selection - JPP((j_compress_ptr cinfo, jvirt_barray_ptr * coef_arrays)); -LOCAL(void) transencode_coef_controller - JPP((j_compress_ptr cinfo, jvirt_barray_ptr * coef_arrays)); - - -/* - * Compression initialization for writing raw-coefficient data. - * Before calling this, all parameters and a data destination must be set up. - * Call jpeg_finish_compress() to actually write the data. - * - * The number of passed virtual arrays must match cinfo->num_components. - * Note that the virtual arrays need not be filled or even realized at - * the time write_coefficients is called; indeed, if the virtual arrays - * were requested from this compression object's memory manager, they - * typically will be realized during this routine and filled afterwards. - */ - -GLOBAL(void) -jpeg_write_coefficients (j_compress_ptr cinfo, jvirt_barray_ptr * coef_arrays) -{ - if (cinfo->global_state != CSTATE_START) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - /* Mark all tables to be written */ - jpeg_suppress_tables(cinfo, FALSE); - /* (Re)initialize error mgr and destination modules */ - (*cinfo->err->reset_error_mgr) ((j_common_ptr) cinfo); - (*cinfo->dest->init_destination) (cinfo); - /* Perform master selection of active modules */ - transencode_master_selection(cinfo, coef_arrays); - /* Wait for jpeg_finish_compress() call */ - cinfo->next_scanline = 0; /* so jpeg_write_marker works */ - cinfo->global_state = CSTATE_WRCOEFS; -} - - -/* - * Initialize the compression object with default parameters, - * then copy from the source object all parameters needed for lossless - * transcoding. Parameters that can be varied without loss (such as - * scan script and Huffman optimization) are left in their default states. - */ - -GLOBAL(void) -jpeg_copy_critical_parameters (j_decompress_ptr srcinfo, - j_compress_ptr dstinfo) -{ - JQUANT_TBL ** qtblptr; - jpeg_component_info *incomp, *outcomp; - JQUANT_TBL *c_quant, *slot_quant; - int tblno, ci, coefi; - - /* Safety check to ensure start_compress not called yet. */ - if (dstinfo->global_state != CSTATE_START) - ERREXIT1(dstinfo, JERR_BAD_STATE, dstinfo->global_state); - /* Copy fundamental image dimensions */ - dstinfo->image_width = srcinfo->image_width; - dstinfo->image_height = srcinfo->image_height; - dstinfo->input_components = srcinfo->num_components; - dstinfo->in_color_space = srcinfo->jpeg_color_space; - dstinfo->jpeg_width = srcinfo->output_width; - dstinfo->jpeg_height = srcinfo->output_height; - dstinfo->min_DCT_h_scaled_size = srcinfo->min_DCT_h_scaled_size; - dstinfo->min_DCT_v_scaled_size = srcinfo->min_DCT_v_scaled_size; - /* Initialize all parameters to default values */ - jpeg_set_defaults(dstinfo); - /* jpeg_set_defaults may choose wrong colorspace, eg YCbCr if input is RGB. - * Fix it to get the right header markers for the image colorspace. - * Note: Entropy table assignment in jpeg_set_colorspace - * depends on color_transform. - * Adaption is also required for setting the appropriate - * entropy coding mode dependent on image data precision. - */ - dstinfo->color_transform = srcinfo->color_transform; - jpeg_set_colorspace(dstinfo, srcinfo->jpeg_color_space); - dstinfo->data_precision = srcinfo->data_precision; - dstinfo->arith_code = srcinfo->data_precision > 8 ? TRUE : FALSE; - dstinfo->CCIR601_sampling = srcinfo->CCIR601_sampling; - /* Copy the source's quantization tables. */ - for (tblno = 0; tblno < NUM_QUANT_TBLS; tblno++) { - if (srcinfo->quant_tbl_ptrs[tblno] != NULL) { - qtblptr = & dstinfo->quant_tbl_ptrs[tblno]; - if (*qtblptr == NULL) - *qtblptr = jpeg_alloc_quant_table((j_common_ptr) dstinfo); - MEMCOPY((*qtblptr)->quantval, - srcinfo->quant_tbl_ptrs[tblno]->quantval, - SIZEOF((*qtblptr)->quantval)); - (*qtblptr)->sent_table = FALSE; - } - } - /* Copy the source's per-component info. - * Note we assume jpeg_set_defaults has allocated the dest comp_info array. - */ - dstinfo->num_components = srcinfo->num_components; - if (dstinfo->num_components < 1 || dstinfo->num_components > MAX_COMPONENTS) - ERREXIT2(dstinfo, JERR_COMPONENT_COUNT, dstinfo->num_components, - MAX_COMPONENTS); - for (ci = 0, incomp = srcinfo->comp_info, outcomp = dstinfo->comp_info; - ci < dstinfo->num_components; ci++, incomp++, outcomp++) { - outcomp->component_id = incomp->component_id; - outcomp->h_samp_factor = incomp->h_samp_factor; - outcomp->v_samp_factor = incomp->v_samp_factor; - outcomp->quant_tbl_no = incomp->quant_tbl_no; - /* Make sure saved quantization table for component matches the qtable - * slot. If not, the input file re-used this qtable slot. - * IJG encoder currently cannot duplicate this. - */ - tblno = outcomp->quant_tbl_no; - if (tblno < 0 || tblno >= NUM_QUANT_TBLS || - srcinfo->quant_tbl_ptrs[tblno] == NULL) - ERREXIT1(dstinfo, JERR_NO_QUANT_TABLE, tblno); - slot_quant = srcinfo->quant_tbl_ptrs[tblno]; - c_quant = incomp->quant_table; - if (c_quant != NULL) { - for (coefi = 0; coefi < DCTSIZE2; coefi++) { - if (c_quant->quantval[coefi] != slot_quant->quantval[coefi]) - ERREXIT1(dstinfo, JERR_MISMATCHED_QUANT_TABLE, tblno); - } - } - /* Note: we do not copy the source's entropy table assignments; - * instead we rely on jpeg_set_colorspace to have made a suitable choice. - */ - } - /* Also copy JFIF version and resolution information, if available. - * Strictly speaking this isn't "critical" info, but it's nearly - * always appropriate to copy it if available. In particular, - * if the application chooses to copy JFIF 1.02 extension markers from - * the source file, we need to copy the version to make sure we don't - * emit a file that has 1.02 extensions but a claimed version of 1.01. - */ - if (srcinfo->saw_JFIF_marker) { - if (srcinfo->JFIF_major_version == 1 || - srcinfo->JFIF_major_version == 2) { - dstinfo->JFIF_major_version = srcinfo->JFIF_major_version; - dstinfo->JFIF_minor_version = srcinfo->JFIF_minor_version; - } - dstinfo->density_unit = srcinfo->density_unit; - dstinfo->X_density = srcinfo->X_density; - dstinfo->Y_density = srcinfo->Y_density; - } -} - - -LOCAL(void) -jpeg_calc_trans_dimensions (j_compress_ptr cinfo) -/* Do computations that are needed before master selection phase */ -{ - if (cinfo->min_DCT_h_scaled_size != cinfo->min_DCT_v_scaled_size) - ERREXIT2(cinfo, JERR_BAD_DCTSIZE, - cinfo->min_DCT_h_scaled_size, cinfo->min_DCT_v_scaled_size); - - cinfo->block_size = cinfo->min_DCT_h_scaled_size; -} - - -/* - * Master selection of compression modules for transcoding. - * This substitutes for jcinit.c's initialization of the full compressor. - */ - -LOCAL(void) -transencode_master_selection (j_compress_ptr cinfo, - jvirt_barray_ptr * coef_arrays) -{ - /* Do computations that are needed before master selection phase */ - jpeg_calc_trans_dimensions(cinfo); - - /* Initialize master control (includes parameter checking/processing) */ - jinit_c_master_control(cinfo, TRUE /* transcode only */); - - /* Entropy encoding: either Huffman or arithmetic coding. */ - if (cinfo->arith_code) - jinit_arith_encoder(cinfo); - else { - jinit_huff_encoder(cinfo); - } - - /* We need a special coefficient buffer controller. */ - transencode_coef_controller(cinfo, coef_arrays); - - jinit_marker_writer(cinfo); - - /* We can now tell the memory manager to allocate virtual arrays. */ - (*cinfo->mem->realize_virt_arrays) ((j_common_ptr) cinfo); - - /* Write the datastream header (SOI, JFIF) immediately. - * Frame and scan headers are postponed till later. - * This lets application insert special markers after the SOI. - */ - (*cinfo->marker->write_file_header) (cinfo); -} - - -/* - * The rest of this file is a special implementation of the coefficient - * buffer controller. This is similar to jccoefct.c, but it handles only - * output from presupplied virtual arrays. Furthermore, we generate any - * dummy padding blocks on-the-fly rather than expecting them to be present - * in the arrays. - */ - -/* Private buffer controller object */ - -typedef struct { - struct jpeg_c_coef_controller pub; /* public fields */ - - JDIMENSION iMCU_row_num; /* iMCU row # within image */ - JDIMENSION MCU_ctr; /* counts MCUs processed in current row */ - int MCU_vert_offset; /* counts MCU rows within iMCU row */ - int MCU_rows_per_iMCU_row; /* number of such rows needed */ - - /* Virtual block array for each component. */ - jvirt_barray_ptr * whole_image; - - /* Workspace for constructing dummy blocks at right/bottom edges. */ - JBLOCK dummy_buffer[C_MAX_BLOCKS_IN_MCU]; -} my_coef_controller; - -typedef my_coef_controller * my_coef_ptr; - - -LOCAL(void) -start_iMCU_row (j_compress_ptr cinfo) -/* Reset within-iMCU-row counters for a new row */ -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - - /* In an interleaved scan, an MCU row is the same as an iMCU row. - * In a noninterleaved scan, an iMCU row has v_samp_factor MCU rows. - * But at the bottom of the image, process only what's left. - */ - if (cinfo->comps_in_scan > 1) { - coef->MCU_rows_per_iMCU_row = 1; - } else { - if (coef->iMCU_row_num < (cinfo->total_iMCU_rows-1)) - coef->MCU_rows_per_iMCU_row = cinfo->cur_comp_info[0]->v_samp_factor; - else - coef->MCU_rows_per_iMCU_row = cinfo->cur_comp_info[0]->last_row_height; - } - - coef->MCU_ctr = 0; - coef->MCU_vert_offset = 0; -} - - -/* - * Initialize for a processing pass. - */ - -METHODDEF(void) -start_pass_coef (j_compress_ptr cinfo, J_BUF_MODE pass_mode) -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - - if (pass_mode != JBUF_CRANK_DEST) - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - - coef->iMCU_row_num = 0; - start_iMCU_row(cinfo); -} - - -/* - * Process some data. - * We process the equivalent of one fully interleaved MCU row ("iMCU" row) - * per call, ie, v_samp_factor block rows for each component in the scan. - * The data is obtained from the virtual arrays and fed to the entropy coder. - * Returns TRUE if the iMCU row is completed, FALSE if suspended. - * - * NB: input_buf is ignored; it is likely to be a NULL pointer. - */ - -METHODDEF(boolean) -compress_output (j_compress_ptr cinfo, JSAMPIMAGE input_buf) -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - JDIMENSION MCU_col_num; /* index of current MCU within row */ - JDIMENSION last_MCU_col = cinfo->MCUs_per_row - 1; - JDIMENSION last_iMCU_row = cinfo->total_iMCU_rows - 1; - int blkn, ci, xindex, yindex, yoffset, blockcnt; - JDIMENSION start_col; - JBLOCKARRAY buffer[MAX_COMPS_IN_SCAN]; - JBLOCKROW MCU_buffer[C_MAX_BLOCKS_IN_MCU]; - JBLOCKROW buffer_ptr; - jpeg_component_info *compptr; - - /* Align the virtual buffers for the components used in this scan. */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - buffer[ci] = (*cinfo->mem->access_virt_barray) - ((j_common_ptr) cinfo, coef->whole_image[compptr->component_index], - coef->iMCU_row_num * compptr->v_samp_factor, - (JDIMENSION) compptr->v_samp_factor, FALSE); - } - - /* Loop to process one whole iMCU row */ - for (yoffset = coef->MCU_vert_offset; yoffset < coef->MCU_rows_per_iMCU_row; - yoffset++) { - for (MCU_col_num = coef->MCU_ctr; MCU_col_num <= last_MCU_col; - MCU_col_num++) { - /* Construct list of pointers to DCT blocks belonging to this MCU */ - blkn = 0; /* index of current DCT block within MCU */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - blockcnt = (MCU_col_num < last_MCU_col) ? compptr->MCU_width - : compptr->last_col_width; - start_col = MCU_col_num * compptr->MCU_width; - for (yindex = 0; yindex < compptr->MCU_height; yindex++) { - if (coef->iMCU_row_num < last_iMCU_row || - yoffset + yindex < compptr->last_row_height) { - /* Fill in pointers to real blocks in this row */ - buffer_ptr = buffer[ci][yoffset + yindex] + start_col; - xindex = blockcnt; - do { - MCU_buffer[blkn++] = buffer_ptr++; - } while (--xindex); - /* Dummy blocks at right edge */ - if ((xindex = compptr->MCU_width - blockcnt) == 0) - continue; - } else { - /* At bottom of image, need a whole row of dummy blocks */ - xindex = compptr->MCU_width; - } - /* Fill in any dummy blocks needed in this row. - * Dummy blocks are filled in the same way as in jccoefct.c: - * all zeroes in the AC entries, DC entries equal to previous - * block's DC value. The init routine has already zeroed the - * AC entries, so we need only set the DC entries correctly. - */ - buffer_ptr = coef->dummy_buffer + blkn; - do { - buffer_ptr[0][0] = MCU_buffer[blkn-1][0][0]; - MCU_buffer[blkn++] = buffer_ptr++; - } while (--xindex); - } - } - /* Try to write the MCU. */ - if (! (*cinfo->entropy->encode_mcu) (cinfo, MCU_buffer)) { - /* Suspension forced; update state counters and exit */ - coef->MCU_vert_offset = yoffset; - coef->MCU_ctr = MCU_col_num; - return FALSE; - } - } - /* Completed an MCU row, but perhaps not an iMCU row */ - coef->MCU_ctr = 0; - } - /* Completed the iMCU row, advance counters for next one */ - coef->iMCU_row_num++; - start_iMCU_row(cinfo); - return TRUE; -} - - -/* - * Initialize coefficient buffer controller. - * - * Each passed coefficient array must be the right size for that - * coefficient: width_in_blocks wide and height_in_blocks high, - * with unitheight at least v_samp_factor. - */ - -LOCAL(void) -transencode_coef_controller (j_compress_ptr cinfo, - jvirt_barray_ptr * coef_arrays) -{ - my_coef_ptr coef; - - coef = (my_coef_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_coef_controller)); - cinfo->coef = &coef->pub; - coef->pub.start_pass = start_pass_coef; - coef->pub.compress_data = compress_output; - - /* Save pointer to virtual arrays */ - coef->whole_image = coef_arrays; - - /* Pre-zero space for dummy DCT blocks */ - MEMZERO(coef->dummy_buffer, SIZEOF(coef->dummy_buffer)); -} diff --git a/dep/libjpeg/src/jdapimin.c b/dep/libjpeg/src/jdapimin.c deleted file mode 100644 index 785e52722..000000000 --- a/dep/libjpeg/src/jdapimin.c +++ /dev/null @@ -1,412 +0,0 @@ -/* - * jdapimin.c - * - * Copyright (C) 1994-1998, Thomas G. Lane. - * Modified 2009-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains application interface code for the decompression half - * of the JPEG library. These are the "minimum" API routines that may be - * needed in either the normal full-decompression case or the - * transcoding-only case. - * - * Most of the routines intended to be called directly by an application - * are in this file or in jdapistd.c. But also see jcomapi.c for routines - * shared by compression and decompression, and jdtrans.c for the transcoding - * case. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* - * Initialization of a JPEG decompression object. - * The error manager must already be set up (in case memory manager fails). - */ - -GLOBAL(void) -jpeg_CreateDecompress (j_decompress_ptr cinfo, int version, size_t structsize) -{ - int i; - - /* Guard against version mismatches between library and caller. */ - cinfo->mem = NULL; /* so jpeg_destroy knows mem mgr not called */ - if (version != JPEG_LIB_VERSION) - ERREXIT2(cinfo, JERR_BAD_LIB_VERSION, JPEG_LIB_VERSION, version); - if (structsize != SIZEOF(struct jpeg_decompress_struct)) - ERREXIT2(cinfo, JERR_BAD_STRUCT_SIZE, - (int) SIZEOF(struct jpeg_decompress_struct), (int) structsize); - - /* For debugging purposes, we zero the whole master structure. - * But the application has already set the err pointer, and may have set - * client_data, so we have to save and restore those fields. - * Note: if application hasn't set client_data, tools like Purify may - * complain here. - */ - { - struct jpeg_error_mgr * err = cinfo->err; - void * client_data = cinfo->client_data; /* ignore Purify complaint here */ - MEMZERO(cinfo, SIZEOF(struct jpeg_decompress_struct)); - cinfo->err = err; - cinfo->client_data = client_data; - } - cinfo->is_decompressor = TRUE; - - /* Initialize a memory manager instance for this object */ - jinit_memory_mgr((j_common_ptr) cinfo); - - /* Zero out pointers to permanent structures. */ - cinfo->progress = NULL; - cinfo->src = NULL; - - for (i = 0; i < NUM_QUANT_TBLS; i++) - cinfo->quant_tbl_ptrs[i] = NULL; - - for (i = 0; i < NUM_HUFF_TBLS; i++) { - cinfo->dc_huff_tbl_ptrs[i] = NULL; - cinfo->ac_huff_tbl_ptrs[i] = NULL; - } - - /* Initialize marker processor so application can override methods - * for COM, APPn markers before calling jpeg_read_header. - */ - cinfo->marker_list = NULL; - jinit_marker_reader(cinfo); - - /* And initialize the overall input controller. */ - jinit_input_controller(cinfo); - - /* OK, I'm ready */ - cinfo->global_state = DSTATE_START; -} - - -/* - * Destruction of a JPEG decompression object - */ - -GLOBAL(void) -jpeg_destroy_decompress (j_decompress_ptr cinfo) -{ - jpeg_destroy((j_common_ptr) cinfo); /* use common routine */ -} - - -/* - * Abort processing of a JPEG decompression operation, - * but don't destroy the object itself. - */ - -GLOBAL(void) -jpeg_abort_decompress (j_decompress_ptr cinfo) -{ - jpeg_abort((j_common_ptr) cinfo); /* use common routine */ -} - - -/* - * Set default decompression parameters. - */ - -LOCAL(void) -default_decompress_parms (j_decompress_ptr cinfo) -{ - int cid0, cid1, cid2, cid3; - - /* Guess the input colorspace, and set output colorspace accordingly. */ - /* Note application may override our guesses. */ - switch (cinfo->num_components) { - case 1: - cinfo->jpeg_color_space = JCS_GRAYSCALE; - cinfo->out_color_space = JCS_GRAYSCALE; - break; - - case 3: - cid0 = cinfo->comp_info[0].component_id; - cid1 = cinfo->comp_info[1].component_id; - cid2 = cinfo->comp_info[2].component_id; - - /* For robust detection of standard colorspaces - * regardless of the presence of special markers, - * check component IDs from SOF marker first. - */ - if (cid0 == 0x01 && cid1 == 0x02 && cid2 == 0x03) - cinfo->jpeg_color_space = JCS_YCbCr; - else if (cid0 == 0x01 && cid1 == 0x22 && cid2 == 0x23) - cinfo->jpeg_color_space = JCS_BG_YCC; - else if (cid0 == 0x52 && cid1 == 0x47 && cid2 == 0x42) - cinfo->jpeg_color_space = JCS_RGB; /* ASCII 'R', 'G', 'B' */ - else if (cid0 == 0x72 && cid1 == 0x67 && cid2 == 0x62) - cinfo->jpeg_color_space = JCS_BG_RGB; /* ASCII 'r', 'g', 'b' */ - else if (cinfo->saw_JFIF_marker) - cinfo->jpeg_color_space = JCS_YCbCr; /* assume it's YCbCr */ - else if (cinfo->saw_Adobe_marker) { - switch (cinfo->Adobe_transform) { - case 0: - cinfo->jpeg_color_space = JCS_RGB; - break; - case 1: - cinfo->jpeg_color_space = JCS_YCbCr; - break; - default: - WARNMS1(cinfo, JWRN_ADOBE_XFORM, cinfo->Adobe_transform); - cinfo->jpeg_color_space = JCS_YCbCr; /* assume it's YCbCr */ - } - } else { - TRACEMS3(cinfo, 1, JTRC_UNKNOWN_IDS, cid0, cid1, cid2); - cinfo->jpeg_color_space = JCS_YCbCr; /* assume it's YCbCr */ - } - /* Always guess RGB is proper output colorspace. */ - cinfo->out_color_space = JCS_RGB; - break; - - case 4: - cid0 = cinfo->comp_info[0].component_id; - cid1 = cinfo->comp_info[1].component_id; - cid2 = cinfo->comp_info[2].component_id; - cid3 = cinfo->comp_info[3].component_id; - - /* For robust detection of standard colorspaces - * regardless of the presence of special markers, - * check component IDs from SOF marker first. - */ - if (cid0 == 0x01 && cid1 == 0x02 && cid2 == 0x03 && cid3 == 0x04) - cinfo->jpeg_color_space = JCS_YCCK; - else if (cid0 == 0x43 && cid1 == 0x4D && cid2 == 0x59 && cid3 == 0x4B) - cinfo->jpeg_color_space = JCS_CMYK; /* ASCII 'C', 'M', 'Y', 'K' */ - else if (cinfo->saw_Adobe_marker) { - switch (cinfo->Adobe_transform) { - case 0: - cinfo->jpeg_color_space = JCS_CMYK; - break; - case 2: - cinfo->jpeg_color_space = JCS_YCCK; - break; - default: - WARNMS1(cinfo, JWRN_ADOBE_XFORM, cinfo->Adobe_transform); - cinfo->jpeg_color_space = JCS_YCCK; /* assume it's YCCK */ - } - } else { - /* Unknown IDs and no special markers, assume straight CMYK. */ - cinfo->jpeg_color_space = JCS_CMYK; - } - cinfo->out_color_space = JCS_CMYK; - break; - - default: - cinfo->jpeg_color_space = JCS_UNKNOWN; - cinfo->out_color_space = JCS_UNKNOWN; - } - - /* Set defaults for other decompression parameters. */ - cinfo->scale_num = cinfo->block_size; /* 1:1 scaling */ - cinfo->scale_denom = cinfo->block_size; - cinfo->output_gamma = 1.0; - cinfo->buffered_image = FALSE; - cinfo->raw_data_out = FALSE; - cinfo->dct_method = JDCT_DEFAULT; - cinfo->do_fancy_upsampling = TRUE; - cinfo->do_block_smoothing = TRUE; - cinfo->quantize_colors = FALSE; - /* We set these in case application only sets quantize_colors. */ - cinfo->dither_mode = JDITHER_FS; -#ifdef QUANT_2PASS_SUPPORTED - cinfo->two_pass_quantize = TRUE; -#else - cinfo->two_pass_quantize = FALSE; -#endif - cinfo->desired_number_of_colors = 256; - cinfo->colormap = NULL; - /* Initialize for no mode change in buffered-image mode. */ - cinfo->enable_1pass_quant = FALSE; - cinfo->enable_external_quant = FALSE; - cinfo->enable_2pass_quant = FALSE; -} - - -/* - * Decompression startup: read start of JPEG datastream to see what's there. - * Need only initialize JPEG object and supply a data source before calling. - * - * This routine will read as far as the first SOS marker (ie, actual start of - * compressed data), and will save all tables and parameters in the JPEG - * object. It will also initialize the decompression parameters to default - * values, and finally return JPEG_HEADER_OK. On return, the application may - * adjust the decompression parameters and then call jpeg_start_decompress. - * (Or, if the application only wanted to determine the image parameters, - * the data need not be decompressed. In that case, call jpeg_abort or - * jpeg_destroy to release any temporary space.) - * If an abbreviated (tables only) datastream is presented, the routine will - * return JPEG_HEADER_TABLES_ONLY upon reaching EOI. The application may then - * re-use the JPEG object to read the abbreviated image datastream(s). - * It is unnecessary (but OK) to call jpeg_abort in this case. - * The JPEG_SUSPENDED return code only occurs if the data source module - * requests suspension of the decompressor. In this case the application - * should load more source data and then re-call jpeg_read_header to resume - * processing. - * If a non-suspending data source is used and require_image is TRUE, then the - * return code need not be inspected since only JPEG_HEADER_OK is possible. - * - * This routine is now just a front end to jpeg_consume_input, with some - * extra error checking. - */ - -GLOBAL(int) -jpeg_read_header (j_decompress_ptr cinfo, boolean require_image) -{ - int retcode; - - if (cinfo->global_state != DSTATE_START && - cinfo->global_state != DSTATE_INHEADER) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - - retcode = jpeg_consume_input(cinfo); - - switch (retcode) { - case JPEG_REACHED_SOS: - retcode = JPEG_HEADER_OK; - break; - case JPEG_REACHED_EOI: - if (require_image) /* Complain if application wanted an image */ - ERREXIT(cinfo, JERR_NO_IMAGE); - /* Reset to start state; it would be safer to require the application to - * call jpeg_abort, but we can't change it now for compatibility reasons. - * A side effect is to free any temporary memory (there shouldn't be any). - */ - jpeg_abort((j_common_ptr) cinfo); /* sets state = DSTATE_START */ - retcode = JPEG_HEADER_TABLES_ONLY; - break; - case JPEG_SUSPENDED: - /* no work */ - break; - } - - return retcode; -} - - -/* - * Consume data in advance of what the decompressor requires. - * This can be called at any time once the decompressor object has - * been created and a data source has been set up. - * - * This routine is essentially a state machine that handles a couple - * of critical state-transition actions, namely initial setup and - * transition from header scanning to ready-for-start_decompress. - * All the actual input is done via the input controller's consume_input - * method. - */ - -GLOBAL(int) -jpeg_consume_input (j_decompress_ptr cinfo) -{ - int retcode = JPEG_SUSPENDED; - - /* NB: every possible DSTATE value should be listed in this switch */ - switch (cinfo->global_state) { - case DSTATE_START: - /* Start-of-datastream actions: reset appropriate modules */ - (*cinfo->inputctl->reset_input_controller) (cinfo); - /* Initialize application's data source module */ - (*cinfo->src->init_source) (cinfo); - cinfo->global_state = DSTATE_INHEADER; - /*FALLTHROUGH*/ - case DSTATE_INHEADER: - retcode = (*cinfo->inputctl->consume_input) (cinfo); - if (retcode == JPEG_REACHED_SOS) { /* Found SOS, prepare to decompress */ - /* Set up default parameters based on header data */ - default_decompress_parms(cinfo); - /* Set global state: ready for start_decompress */ - cinfo->global_state = DSTATE_READY; - } - break; - case DSTATE_READY: - /* Can't advance past first SOS until start_decompress is called */ - retcode = JPEG_REACHED_SOS; - break; - case DSTATE_PRELOAD: - case DSTATE_PRESCAN: - case DSTATE_SCANNING: - case DSTATE_RAW_OK: - case DSTATE_BUFIMAGE: - case DSTATE_BUFPOST: - case DSTATE_STOPPING: - retcode = (*cinfo->inputctl->consume_input) (cinfo); - break; - default: - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - } - return retcode; -} - - -/* - * Have we finished reading the input file? - */ - -GLOBAL(boolean) -jpeg_input_complete (j_decompress_ptr cinfo) -{ - /* Check for valid jpeg object */ - if (cinfo->global_state < DSTATE_START || - cinfo->global_state > DSTATE_STOPPING) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - return cinfo->inputctl->eoi_reached; -} - - -/* - * Is there more than one scan? - */ - -GLOBAL(boolean) -jpeg_has_multiple_scans (j_decompress_ptr cinfo) -{ - /* Only valid after jpeg_read_header completes */ - if (cinfo->global_state < DSTATE_READY || - cinfo->global_state > DSTATE_STOPPING) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - return cinfo->inputctl->has_multiple_scans; -} - - -/* - * Finish JPEG decompression. - * - * This will normally just verify the file trailer and release temp storage. - * - * Returns FALSE if suspended. The return value need be inspected only if - * a suspending data source is used. - */ - -GLOBAL(boolean) -jpeg_finish_decompress (j_decompress_ptr cinfo) -{ - if ((cinfo->global_state == DSTATE_SCANNING || - cinfo->global_state == DSTATE_RAW_OK) && ! cinfo->buffered_image) { - /* Terminate final pass of non-buffered mode */ - if (cinfo->output_scanline < cinfo->output_height) - ERREXIT(cinfo, JERR_TOO_LITTLE_DATA); - (*cinfo->master->finish_output_pass) (cinfo); - cinfo->global_state = DSTATE_STOPPING; - } else if (cinfo->global_state == DSTATE_BUFIMAGE) { - /* Finishing after a buffered-image operation */ - cinfo->global_state = DSTATE_STOPPING; - } else if (cinfo->global_state != DSTATE_STOPPING) { - /* STOPPING = repeat call after a suspension, anything else is error */ - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - } - /* Read until EOI */ - while (! cinfo->inputctl->eoi_reached) { - if ((*cinfo->inputctl->consume_input) (cinfo) == JPEG_SUSPENDED) - return FALSE; /* Suspend, come back later */ - } - /* Do final cleanup */ - (*cinfo->src->term_source) (cinfo); - /* We can use jpeg_abort to release memory and reset global_state */ - jpeg_abort((j_common_ptr) cinfo); - return TRUE; -} diff --git a/dep/libjpeg/src/jdapistd.c b/dep/libjpeg/src/jdapistd.c deleted file mode 100644 index 7f3a78b25..000000000 --- a/dep/libjpeg/src/jdapistd.c +++ /dev/null @@ -1,276 +0,0 @@ -/* - * jdapistd.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2002-2013 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains application interface code for the decompression half - * of the JPEG library. These are the "standard" API routines that are - * used in the normal full-decompression case. They are not used by a - * transcoding-only application. Note that if an application links in - * jpeg_start_decompress, it will end up linking in the entire decompressor. - * We thus must separate this file from jdapimin.c to avoid linking the - * whole decompression library into a transcoder. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Forward declarations */ -LOCAL(boolean) output_pass_setup JPP((j_decompress_ptr cinfo)); - - -/* - * Decompression initialization. - * jpeg_read_header must be completed before calling this. - * - * If a multipass operating mode was selected, this will do all but the - * last pass, and thus may take a great deal of time. - * - * Returns FALSE if suspended. The return value need be inspected only if - * a suspending data source is used. - */ - -GLOBAL(boolean) -jpeg_start_decompress (j_decompress_ptr cinfo) -{ - if (cinfo->global_state == DSTATE_READY) { - /* First call: initialize master control, select active modules */ - jinit_master_decompress(cinfo); - if (cinfo->buffered_image) { - /* No more work here; expecting jpeg_start_output next */ - cinfo->global_state = DSTATE_BUFIMAGE; - return TRUE; - } - cinfo->global_state = DSTATE_PRELOAD; - } - if (cinfo->global_state == DSTATE_PRELOAD) { - /* If file has multiple scans, absorb them all into the coef buffer */ - if (cinfo->inputctl->has_multiple_scans) { -#ifdef D_MULTISCAN_FILES_SUPPORTED - for (;;) { - int retcode; - /* Call progress monitor hook if present */ - if (cinfo->progress != NULL) - (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo); - /* Absorb some more input */ - retcode = (*cinfo->inputctl->consume_input) (cinfo); - if (retcode == JPEG_SUSPENDED) - return FALSE; - if (retcode == JPEG_REACHED_EOI) - break; - /* Advance progress counter if appropriate */ - if (cinfo->progress != NULL && - (retcode == JPEG_ROW_COMPLETED || retcode == JPEG_REACHED_SOS)) { - if (++cinfo->progress->pass_counter >= cinfo->progress->pass_limit) { - /* jdmaster underestimated number of scans; ratchet up one scan */ - cinfo->progress->pass_limit += (long) cinfo->total_iMCU_rows; - } - } - } -#else - ERREXIT(cinfo, JERR_NOT_COMPILED); -#endif /* D_MULTISCAN_FILES_SUPPORTED */ - } - cinfo->output_scan_number = cinfo->input_scan_number; - } else if (cinfo->global_state != DSTATE_PRESCAN) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - /* Perform any dummy output passes, and set up for the final pass */ - return output_pass_setup(cinfo); -} - - -/* - * Set up for an output pass, and perform any dummy pass(es) needed. - * Common subroutine for jpeg_start_decompress and jpeg_start_output. - * Entry: global_state = DSTATE_PRESCAN only if previously suspended. - * Exit: If done, returns TRUE and sets global_state for proper output mode. - * If suspended, returns FALSE and sets global_state = DSTATE_PRESCAN. - */ - -LOCAL(boolean) -output_pass_setup (j_decompress_ptr cinfo) -{ - if (cinfo->global_state != DSTATE_PRESCAN) { - /* First call: do pass setup */ - (*cinfo->master->prepare_for_output_pass) (cinfo); - cinfo->output_scanline = 0; - cinfo->global_state = DSTATE_PRESCAN; - } - /* Loop over any required dummy passes */ - while (cinfo->master->is_dummy_pass) { -#ifdef QUANT_2PASS_SUPPORTED - /* Crank through the dummy pass */ - while (cinfo->output_scanline < cinfo->output_height) { - JDIMENSION last_scanline; - /* Call progress monitor hook if present */ - if (cinfo->progress != NULL) { - cinfo->progress->pass_counter = (long) cinfo->output_scanline; - cinfo->progress->pass_limit = (long) cinfo->output_height; - (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo); - } - /* Process some data */ - last_scanline = cinfo->output_scanline; - (*cinfo->main->process_data) (cinfo, (JSAMPARRAY) NULL, - &cinfo->output_scanline, (JDIMENSION) 0); - if (cinfo->output_scanline == last_scanline) - return FALSE; /* No progress made, must suspend */ - } - /* Finish up dummy pass, and set up for another one */ - (*cinfo->master->finish_output_pass) (cinfo); - (*cinfo->master->prepare_for_output_pass) (cinfo); - cinfo->output_scanline = 0; -#else - ERREXIT(cinfo, JERR_NOT_COMPILED); -#endif /* QUANT_2PASS_SUPPORTED */ - } - /* Ready for application to drive output pass through - * jpeg_read_scanlines or jpeg_read_raw_data. - */ - cinfo->global_state = cinfo->raw_data_out ? DSTATE_RAW_OK : DSTATE_SCANNING; - return TRUE; -} - - -/* - * Read some scanlines of data from the JPEG decompressor. - * - * The return value will be the number of lines actually read. - * This may be less than the number requested in several cases, - * including bottom of image, data source suspension, and operating - * modes that emit multiple scanlines at a time. - * - * Note: we warn about excess calls to jpeg_read_scanlines() since - * this likely signals an application programmer error. However, - * an oversize buffer (max_lines > scanlines remaining) is not an error. - */ - -GLOBAL(JDIMENSION) -jpeg_read_scanlines (j_decompress_ptr cinfo, JSAMPARRAY scanlines, - JDIMENSION max_lines) -{ - JDIMENSION row_ctr; - - if (cinfo->global_state != DSTATE_SCANNING) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - if (cinfo->output_scanline >= cinfo->output_height) { - WARNMS(cinfo, JWRN_TOO_MUCH_DATA); - return 0; - } - - /* Call progress monitor hook if present */ - if (cinfo->progress != NULL) { - cinfo->progress->pass_counter = (long) cinfo->output_scanline; - cinfo->progress->pass_limit = (long) cinfo->output_height; - (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo); - } - - /* Process some data */ - row_ctr = 0; - (*cinfo->main->process_data) (cinfo, scanlines, &row_ctr, max_lines); - cinfo->output_scanline += row_ctr; - return row_ctr; -} - - -/* - * Alternate entry point to read raw data. - * Processes exactly one iMCU row per call, unless suspended. - */ - -GLOBAL(JDIMENSION) -jpeg_read_raw_data (j_decompress_ptr cinfo, JSAMPIMAGE data, - JDIMENSION max_lines) -{ - JDIMENSION lines_per_iMCU_row; - - if (cinfo->global_state != DSTATE_RAW_OK) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - if (cinfo->output_scanline >= cinfo->output_height) { - WARNMS(cinfo, JWRN_TOO_MUCH_DATA); - return 0; - } - - /* Call progress monitor hook if present */ - if (cinfo->progress != NULL) { - cinfo->progress->pass_counter = (long) cinfo->output_scanline; - cinfo->progress->pass_limit = (long) cinfo->output_height; - (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo); - } - - /* Verify that at least one iMCU row can be returned. */ - lines_per_iMCU_row = cinfo->max_v_samp_factor * cinfo->min_DCT_v_scaled_size; - if (max_lines < lines_per_iMCU_row) - ERREXIT(cinfo, JERR_BUFFER_SIZE); - - /* Decompress directly into user's buffer. */ - if (! (*cinfo->coef->decompress_data) (cinfo, data)) - return 0; /* suspension forced, can do nothing more */ - - /* OK, we processed one iMCU row. */ - cinfo->output_scanline += lines_per_iMCU_row; - return lines_per_iMCU_row; -} - - -/* Additional entry points for buffered-image mode. */ - -#ifdef D_MULTISCAN_FILES_SUPPORTED - -/* - * Initialize for an output pass in buffered-image mode. - */ - -GLOBAL(boolean) -jpeg_start_output (j_decompress_ptr cinfo, int scan_number) -{ - if (cinfo->global_state != DSTATE_BUFIMAGE && - cinfo->global_state != DSTATE_PRESCAN) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - /* Limit scan number to valid range */ - if (scan_number <= 0) - scan_number = 1; - if (cinfo->inputctl->eoi_reached && - scan_number > cinfo->input_scan_number) - scan_number = cinfo->input_scan_number; - cinfo->output_scan_number = scan_number; - /* Perform any dummy output passes, and set up for the real pass */ - return output_pass_setup(cinfo); -} - - -/* - * Finish up after an output pass in buffered-image mode. - * - * Returns FALSE if suspended. The return value need be inspected only if - * a suspending data source is used. - */ - -GLOBAL(boolean) -jpeg_finish_output (j_decompress_ptr cinfo) -{ - if ((cinfo->global_state == DSTATE_SCANNING || - cinfo->global_state == DSTATE_RAW_OK) && cinfo->buffered_image) { - /* Terminate this pass. */ - /* We do not require the whole pass to have been completed. */ - (*cinfo->master->finish_output_pass) (cinfo); - cinfo->global_state = DSTATE_BUFPOST; - } else if (cinfo->global_state != DSTATE_BUFPOST) { - /* BUFPOST = repeat call after a suspension, anything else is error */ - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - } - /* Read markers looking for SOS or EOI */ - while (cinfo->input_scan_number <= cinfo->output_scan_number && - ! cinfo->inputctl->eoi_reached) { - if ((*cinfo->inputctl->consume_input) (cinfo) == JPEG_SUSPENDED) - return FALSE; /* Suspend, come back later */ - } - cinfo->global_state = DSTATE_BUFIMAGE; - return TRUE; -} - -#endif /* D_MULTISCAN_FILES_SUPPORTED */ diff --git a/dep/libjpeg/src/jdarith.c b/dep/libjpeg/src/jdarith.c deleted file mode 100644 index 2c9abe23f..000000000 --- a/dep/libjpeg/src/jdarith.c +++ /dev/null @@ -1,796 +0,0 @@ -/* - * jdarith.c - * - * Developed 1997-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains portable arithmetic entropy decoding routines for JPEG - * (implementing the ISO/IEC IS 10918-1 and CCITT Recommendation ITU-T T.81). - * - * Both sequential and progressive modes are supported in this single module. - * - * Suspension is not currently supported in this module. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Expanded entropy decoder object for arithmetic decoding. */ - -typedef struct { - struct jpeg_entropy_decoder pub; /* public fields */ - - INT32 c; /* C register, base of coding interval + input bit buffer */ - INT32 a; /* A register, normalized size of coding interval */ - int ct; /* bit shift counter, # of bits left in bit buffer part of C */ - /* init: ct = -16 */ - /* run: ct = 0..7 */ - /* error: ct = -1 */ - int last_dc_val[MAX_COMPS_IN_SCAN]; /* last DC coef for each component */ - int dc_context[MAX_COMPS_IN_SCAN]; /* context index for DC conditioning */ - - unsigned int restarts_to_go; /* MCUs left in this restart interval */ - - /* Pointers to statistics areas (these workspaces have image lifespan) */ - unsigned char * dc_stats[NUM_ARITH_TBLS]; - unsigned char * ac_stats[NUM_ARITH_TBLS]; - - /* Statistics bin for coding with fixed probability 0.5 */ - unsigned char fixed_bin[4]; -} arith_entropy_decoder; - -typedef arith_entropy_decoder * arith_entropy_ptr; - -/* The following two definitions specify the allocation chunk size - * for the statistics area. - * According to sections F.1.4.4.1.3 and F.1.4.4.2, we need at least - * 49 statistics bins for DC, and 245 statistics bins for AC coding. - * - * We use a compact representation with 1 byte per statistics bin, - * thus the numbers directly represent byte sizes. - * This 1 byte per statistics bin contains the meaning of the MPS - * (more probable symbol) in the highest bit (mask 0x80), and the - * index into the probability estimation state machine table - * in the lower bits (mask 0x7F). - */ - -#define DC_STAT_BINS 64 -#define AC_STAT_BINS 256 - - -LOCAL(int) -get_byte (j_decompress_ptr cinfo) -/* Read next input byte; we do not support suspension in this module. */ -{ - struct jpeg_source_mgr * src = cinfo->src; - - if (src->bytes_in_buffer == 0) - if (! (*src->fill_input_buffer) (cinfo)) - ERREXIT(cinfo, JERR_CANT_SUSPEND); - src->bytes_in_buffer--; - return GETJOCTET(*src->next_input_byte++); -} - - -/* - * The core arithmetic decoding routine (common in JPEG and JBIG). - * This needs to go as fast as possible. - * Machine-dependent optimization facilities - * are not utilized in this portable implementation. - * However, this code should be fairly efficient and - * may be a good base for further optimizations anyway. - * - * Return value is 0 or 1 (binary decision). - * - * Note: I've changed the handling of the code base & bit - * buffer register C compared to other implementations - * based on the standards layout & procedures. - * While it also contains both the actual base of the - * coding interval (16 bits) and the next-bits buffer, - * the cut-point between these two parts is floating - * (instead of fixed) with the bit shift counter CT. - * Thus, we also need only one (variable instead of - * fixed size) shift for the LPS/MPS decision, and - * we can do away with any renormalization update - * of C (except for new data insertion, of course). - * - * I've also introduced a new scheme for accessing - * the probability estimation state machine table, - * derived from Markus Kuhn's JBIG implementation. - */ - -LOCAL(int) -arith_decode (j_decompress_ptr cinfo, unsigned char *st) -{ - register arith_entropy_ptr e = (arith_entropy_ptr) cinfo->entropy; - register unsigned char nl, nm; - register INT32 qe, temp; - register int sv, data; - - /* Renormalization & data input per section D.2.6 */ - while (e->a < 0x8000L) { - if (--e->ct < 0) { - /* Need to fetch next data byte */ - if (cinfo->unread_marker) - data = 0; /* stuff zero data */ - else { - data = get_byte(cinfo); /* read next input byte */ - if (data == 0xFF) { /* zero stuff or marker code */ - do data = get_byte(cinfo); - while (data == 0xFF); /* swallow extra 0xFF bytes */ - if (data == 0) - data = 0xFF; /* discard stuffed zero byte */ - else { - /* Note: Different from the Huffman decoder, hitting - * a marker while processing the compressed data - * segment is legal in arithmetic coding. - * The convention is to supply zero data - * then until decoding is complete. - */ - cinfo->unread_marker = data; - data = 0; - } - } - } - e->c = (e->c << 8) | data; /* insert data into C register */ - if ((e->ct += 8) < 0) /* update bit shift counter */ - /* Need more initial bytes */ - if (++e->ct == 0) - /* Got 2 initial bytes -> re-init A and exit loop */ - e->a = 0x8000L; /* => e->a = 0x10000L after loop exit */ - } - e->a <<= 1; - } - - /* Fetch values from our compact representation of Table D.3(D.2): - * Qe values and probability estimation state machine - */ - sv = *st; - qe = jpeg_aritab[sv & 0x7F]; /* => Qe_Value */ - nl = qe & 0xFF; qe >>= 8; /* Next_Index_LPS + Switch_MPS */ - nm = qe & 0xFF; qe >>= 8; /* Next_Index_MPS */ - - /* Decode & estimation procedures per sections D.2.4 & D.2.5 */ - temp = e->a - qe; - e->a = temp; - temp <<= e->ct; - if (e->c >= temp) { - e->c -= temp; - /* Conditional LPS (less probable symbol) exchange */ - if (e->a < qe) { - e->a = qe; - *st = (sv & 0x80) ^ nm; /* Estimate_after_MPS */ - } else { - e->a = qe; - *st = (sv & 0x80) ^ nl; /* Estimate_after_LPS */ - sv ^= 0x80; /* Exchange LPS/MPS */ - } - } else if (e->a < 0x8000L) { - /* Conditional MPS (more probable symbol) exchange */ - if (e->a < qe) { - *st = (sv & 0x80) ^ nl; /* Estimate_after_LPS */ - sv ^= 0x80; /* Exchange LPS/MPS */ - } else { - *st = (sv & 0x80) ^ nm; /* Estimate_after_MPS */ - } - } - - return sv >> 7; -} - - -/* - * Check for a restart marker & resynchronize decoder. - */ - -LOCAL(void) -process_restart (j_decompress_ptr cinfo) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - int ci; - jpeg_component_info * compptr; - - /* Advance past the RSTn marker */ - if (! (*cinfo->marker->read_restart_marker) (cinfo)) - ERREXIT(cinfo, JERR_CANT_SUSPEND); - - /* Re-initialize statistics areas */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - if (! cinfo->progressive_mode || (cinfo->Ss == 0 && cinfo->Ah == 0)) { - MEMZERO(entropy->dc_stats[compptr->dc_tbl_no], DC_STAT_BINS); - /* Reset DC predictions to 0 */ - entropy->last_dc_val[ci] = 0; - entropy->dc_context[ci] = 0; - } - if ((! cinfo->progressive_mode && cinfo->lim_Se) || - (cinfo->progressive_mode && cinfo->Ss)) { - MEMZERO(entropy->ac_stats[compptr->ac_tbl_no], AC_STAT_BINS); - } - } - - /* Reset arithmetic decoding variables */ - entropy->c = 0; - entropy->a = 0; - entropy->ct = -16; /* force reading 2 initial bytes to fill C */ - - /* Reset restart counter */ - entropy->restarts_to_go = cinfo->restart_interval; -} - - -/* - * Arithmetic MCU decoding. - * Each of these routines decodes and returns one MCU's worth of - * arithmetic-compressed coefficients. - * The coefficients are reordered from zigzag order into natural array order, - * but are not dequantized. - * - * The i'th block of the MCU is stored into the block pointed to by - * MCU_data[i]. WE ASSUME THIS AREA IS INITIALLY ZEROED BY THE CALLER. - */ - -/* - * MCU decoding for DC initial scan (either spectral selection, - * or first pass of successive approximation). - */ - -METHODDEF(boolean) -decode_mcu_DC_first (j_decompress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - JBLOCKROW block; - unsigned char *st; - int blkn, ci, tbl, sign; - int v, m; - - /* Process restart marker if needed */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) - process_restart(cinfo); - entropy->restarts_to_go--; - } - - if (entropy->ct == -1) return TRUE; /* if error do nothing */ - - /* Outer loop handles each block in the MCU */ - - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - block = MCU_data[blkn]; - ci = cinfo->MCU_membership[blkn]; - tbl = cinfo->cur_comp_info[ci]->dc_tbl_no; - - /* Sections F.2.4.1 & F.1.4.4.1: Decoding of DC coefficients */ - - /* Table F.4: Point to statistics bin S0 for DC coefficient coding */ - st = entropy->dc_stats[tbl] + entropy->dc_context[ci]; - - /* Figure F.19: Decode_DC_DIFF */ - if (arith_decode(cinfo, st) == 0) - entropy->dc_context[ci] = 0; - else { - /* Figure F.21: Decoding nonzero value v */ - /* Figure F.22: Decoding the sign of v */ - sign = arith_decode(cinfo, st + 1); - st += 2; st += sign; - /* Figure F.23: Decoding the magnitude category of v */ - if ((m = arith_decode(cinfo, st)) != 0) { - st = entropy->dc_stats[tbl] + 20; /* Table F.4: X1 = 20 */ - while (arith_decode(cinfo, st)) { - if ((m <<= 1) == (int) 0x8000U) { - WARNMS(cinfo, JWRN_ARITH_BAD_CODE); - entropy->ct = -1; /* magnitude overflow */ - return TRUE; - } - st += 1; - } - } - /* Section F.1.4.4.1.2: Establish dc_context conditioning category */ - if (m < (int) ((1L << cinfo->arith_dc_L[tbl]) >> 1)) - entropy->dc_context[ci] = 0; /* zero diff category */ - else if (m > (int) ((1L << cinfo->arith_dc_U[tbl]) >> 1)) - entropy->dc_context[ci] = 12 + (sign * 4); /* large diff category */ - else - entropy->dc_context[ci] = 4 + (sign * 4); /* small diff category */ - v = m; - /* Figure F.24: Decoding the magnitude bit pattern of v */ - st += 14; - while (m >>= 1) - if (arith_decode(cinfo, st)) v |= m; - v += 1; if (sign) v = -v; - entropy->last_dc_val[ci] += v; - } - - /* Scale and output the DC coefficient (assumes jpeg_natural_order[0]=0) */ - (*block)[0] = (JCOEF) (entropy->last_dc_val[ci] << cinfo->Al); - } - - return TRUE; -} - - -/* - * MCU decoding for AC initial scan (either spectral selection, - * or first pass of successive approximation). - */ - -METHODDEF(boolean) -decode_mcu_AC_first (j_decompress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - JBLOCKROW block; - unsigned char *st; - int tbl, sign, k; - int v, m; - const int * natural_order; - - /* Process restart marker if needed */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) - process_restart(cinfo); - entropy->restarts_to_go--; - } - - if (entropy->ct == -1) return TRUE; /* if error do nothing */ - - natural_order = cinfo->natural_order; - - /* There is always only one block per MCU */ - block = MCU_data[0]; - tbl = cinfo->cur_comp_info[0]->ac_tbl_no; - - /* Sections F.2.4.2 & F.1.4.4.2: Decoding of AC coefficients */ - - /* Figure F.20: Decode_AC_coefficients */ - k = cinfo->Ss - 1; - do { - st = entropy->ac_stats[tbl] + 3 * k; - if (arith_decode(cinfo, st)) break; /* EOB flag */ - for (;;) { - k++; - if (arith_decode(cinfo, st + 1)) break; - st += 3; - if (k >= cinfo->Se) { - WARNMS(cinfo, JWRN_ARITH_BAD_CODE); - entropy->ct = -1; /* spectral overflow */ - return TRUE; - } - } - /* Figure F.21: Decoding nonzero value v */ - /* Figure F.22: Decoding the sign of v */ - sign = arith_decode(cinfo, entropy->fixed_bin); - st += 2; - /* Figure F.23: Decoding the magnitude category of v */ - if ((m = arith_decode(cinfo, st)) != 0) { - if (arith_decode(cinfo, st)) { - m <<= 1; - st = entropy->ac_stats[tbl] + - (k <= cinfo->arith_ac_K[tbl] ? 189 : 217); - while (arith_decode(cinfo, st)) { - if ((m <<= 1) == (int) 0x8000U) { - WARNMS(cinfo, JWRN_ARITH_BAD_CODE); - entropy->ct = -1; /* magnitude overflow */ - return TRUE; - } - st += 1; - } - } - } - v = m; - /* Figure F.24: Decoding the magnitude bit pattern of v */ - st += 14; - while (m >>= 1) - if (arith_decode(cinfo, st)) v |= m; - v += 1; if (sign) v = -v; - /* Scale and output coefficient in natural (dezigzagged) order */ - (*block)[natural_order[k]] = (JCOEF) (v << cinfo->Al); - } while (k < cinfo->Se); - - return TRUE; -} - - -/* - * MCU decoding for DC successive approximation refinement scan. - * Note: we assume such scans can be multi-component, - * although the spec is not very clear on the point. - */ - -METHODDEF(boolean) -decode_mcu_DC_refine (j_decompress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - unsigned char *st; - JCOEF p1; - int blkn; - - /* Process restart marker if needed */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) - process_restart(cinfo); - entropy->restarts_to_go--; - } - - st = entropy->fixed_bin; /* use fixed probability estimation */ - p1 = 1 << cinfo->Al; /* 1 in the bit position being coded */ - - /* Outer loop handles each block in the MCU */ - - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - /* Encoded data is simply the next bit of the two's-complement DC value */ - if (arith_decode(cinfo, st)) - MCU_data[blkn][0][0] |= p1; - } - - return TRUE; -} - - -/* - * MCU decoding for AC successive approximation refinement scan. - */ - -METHODDEF(boolean) -decode_mcu_AC_refine (j_decompress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - JBLOCKROW block; - JCOEFPTR thiscoef; - unsigned char *st; - int tbl, k, kex; - JCOEF p1, m1; - const int * natural_order; - - /* Process restart marker if needed */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) - process_restart(cinfo); - entropy->restarts_to_go--; - } - - if (entropy->ct == -1) return TRUE; /* if error do nothing */ - - natural_order = cinfo->natural_order; - - /* There is always only one block per MCU */ - block = MCU_data[0]; - tbl = cinfo->cur_comp_info[0]->ac_tbl_no; - - p1 = 1 << cinfo->Al; /* 1 in the bit position being coded */ - m1 = -p1; /* -1 in the bit position being coded */ - - /* Establish EOBx (previous stage end-of-block) index */ - kex = cinfo->Se; - do { - if ((*block)[natural_order[kex]]) break; - } while (--kex); - - k = cinfo->Ss - 1; - do { - st = entropy->ac_stats[tbl] + 3 * k; - if (k >= kex) - if (arith_decode(cinfo, st)) break; /* EOB flag */ - for (;;) { - thiscoef = *block + natural_order[++k]; - if (*thiscoef) { /* previously nonzero coef */ - if (arith_decode(cinfo, st + 2)) { - if (*thiscoef < 0) - *thiscoef += m1; - else - *thiscoef += p1; - } - break; - } - if (arith_decode(cinfo, st + 1)) { /* newly nonzero coef */ - if (arith_decode(cinfo, entropy->fixed_bin)) - *thiscoef = m1; - else - *thiscoef = p1; - break; - } - st += 3; - if (k >= cinfo->Se) { - WARNMS(cinfo, JWRN_ARITH_BAD_CODE); - entropy->ct = -1; /* spectral overflow */ - return TRUE; - } - } - } while (k < cinfo->Se); - - return TRUE; -} - - -/* - * Decode one MCU's worth of arithmetic-compressed coefficients. - */ - -METHODDEF(boolean) -decode_mcu (j_decompress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - jpeg_component_info * compptr; - JBLOCKROW block; - unsigned char *st; - int blkn, ci, tbl, sign, k; - int v, m; - const int * natural_order; - - /* Process restart marker if needed */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) - process_restart(cinfo); - entropy->restarts_to_go--; - } - - if (entropy->ct == -1) return TRUE; /* if error do nothing */ - - natural_order = cinfo->natural_order; - - /* Outer loop handles each block in the MCU */ - - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - block = MCU_data[blkn]; - ci = cinfo->MCU_membership[blkn]; - compptr = cinfo->cur_comp_info[ci]; - - /* Sections F.2.4.1 & F.1.4.4.1: Decoding of DC coefficients */ - - tbl = compptr->dc_tbl_no; - - /* Table F.4: Point to statistics bin S0 for DC coefficient coding */ - st = entropy->dc_stats[tbl] + entropy->dc_context[ci]; - - /* Figure F.19: Decode_DC_DIFF */ - if (arith_decode(cinfo, st) == 0) - entropy->dc_context[ci] = 0; - else { - /* Figure F.21: Decoding nonzero value v */ - /* Figure F.22: Decoding the sign of v */ - sign = arith_decode(cinfo, st + 1); - st += 2; st += sign; - /* Figure F.23: Decoding the magnitude category of v */ - if ((m = arith_decode(cinfo, st)) != 0) { - st = entropy->dc_stats[tbl] + 20; /* Table F.4: X1 = 20 */ - while (arith_decode(cinfo, st)) { - if ((m <<= 1) == (int) 0x8000U) { - WARNMS(cinfo, JWRN_ARITH_BAD_CODE); - entropy->ct = -1; /* magnitude overflow */ - return TRUE; - } - st += 1; - } - } - /* Section F.1.4.4.1.2: Establish dc_context conditioning category */ - if (m < (int) ((1L << cinfo->arith_dc_L[tbl]) >> 1)) - entropy->dc_context[ci] = 0; /* zero diff category */ - else if (m > (int) ((1L << cinfo->arith_dc_U[tbl]) >> 1)) - entropy->dc_context[ci] = 12 + (sign * 4); /* large diff category */ - else - entropy->dc_context[ci] = 4 + (sign * 4); /* small diff category */ - v = m; - /* Figure F.24: Decoding the magnitude bit pattern of v */ - st += 14; - while (m >>= 1) - if (arith_decode(cinfo, st)) v |= m; - v += 1; if (sign) v = -v; - entropy->last_dc_val[ci] += v; - } - - (*block)[0] = (JCOEF) entropy->last_dc_val[ci]; - - /* Sections F.2.4.2 & F.1.4.4.2: Decoding of AC coefficients */ - - if (cinfo->lim_Se == 0) continue; - tbl = compptr->ac_tbl_no; - k = 0; - - /* Figure F.20: Decode_AC_coefficients */ - do { - st = entropy->ac_stats[tbl] + 3 * k; - if (arith_decode(cinfo, st)) break; /* EOB flag */ - for (;;) { - k++; - if (arith_decode(cinfo, st + 1)) break; - st += 3; - if (k >= cinfo->lim_Se) { - WARNMS(cinfo, JWRN_ARITH_BAD_CODE); - entropy->ct = -1; /* spectral overflow */ - return TRUE; - } - } - /* Figure F.21: Decoding nonzero value v */ - /* Figure F.22: Decoding the sign of v */ - sign = arith_decode(cinfo, entropy->fixed_bin); - st += 2; - /* Figure F.23: Decoding the magnitude category of v */ - if ((m = arith_decode(cinfo, st)) != 0) { - if (arith_decode(cinfo, st)) { - m <<= 1; - st = entropy->ac_stats[tbl] + - (k <= cinfo->arith_ac_K[tbl] ? 189 : 217); - while (arith_decode(cinfo, st)) { - if ((m <<= 1) == (int) 0x8000U) { - WARNMS(cinfo, JWRN_ARITH_BAD_CODE); - entropy->ct = -1; /* magnitude overflow */ - return TRUE; - } - st += 1; - } - } - } - v = m; - /* Figure F.24: Decoding the magnitude bit pattern of v */ - st += 14; - while (m >>= 1) - if (arith_decode(cinfo, st)) v |= m; - v += 1; if (sign) v = -v; - (*block)[natural_order[k]] = (JCOEF) v; - } while (k < cinfo->lim_Se); - } - - return TRUE; -} - - -/* - * Initialize for an arithmetic-compressed scan. - */ - -METHODDEF(void) -start_pass (j_decompress_ptr cinfo) -{ - arith_entropy_ptr entropy = (arith_entropy_ptr) cinfo->entropy; - int ci, tbl; - jpeg_component_info * compptr; - - if (cinfo->progressive_mode) { - /* Validate progressive scan parameters */ - if (cinfo->Ss == 0) { - if (cinfo->Se != 0) - goto bad; - } else { - /* need not check Ss/Se < 0 since they came from unsigned bytes */ - if (cinfo->Se < cinfo->Ss || cinfo->Se > cinfo->lim_Se) - goto bad; - /* AC scans may have only one component */ - if (cinfo->comps_in_scan != 1) - goto bad; - } - if (cinfo->Ah != 0) { - /* Successive approximation refinement scan: must have Al = Ah-1. */ - if (cinfo->Ah-1 != cinfo->Al) - goto bad; - } - if (cinfo->Al > 13) { /* need not check for < 0 */ - bad: - ERREXIT4(cinfo, JERR_BAD_PROGRESSION, - cinfo->Ss, cinfo->Se, cinfo->Ah, cinfo->Al); - } - /* Update progression status, and verify that scan order is legal. - * Note that inter-scan inconsistencies are treated as warnings - * not fatal errors ... not clear if this is right way to behave. - */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - int coefi, cindex = cinfo->cur_comp_info[ci]->component_index; - int *coef_bit_ptr = & cinfo->coef_bits[cindex][0]; - if (cinfo->Ss && coef_bit_ptr[0] < 0) /* AC without prior DC scan */ - WARNMS2(cinfo, JWRN_BOGUS_PROGRESSION, cindex, 0); - for (coefi = cinfo->Ss; coefi <= cinfo->Se; coefi++) { - int expected = (coef_bit_ptr[coefi] < 0) ? 0 : coef_bit_ptr[coefi]; - if (cinfo->Ah != expected) - WARNMS2(cinfo, JWRN_BOGUS_PROGRESSION, cindex, coefi); - coef_bit_ptr[coefi] = cinfo->Al; - } - } - /* Select MCU decoding routine */ - if (cinfo->Ah == 0) { - if (cinfo->Ss == 0) - entropy->pub.decode_mcu = decode_mcu_DC_first; - else - entropy->pub.decode_mcu = decode_mcu_AC_first; - } else { - if (cinfo->Ss == 0) - entropy->pub.decode_mcu = decode_mcu_DC_refine; - else - entropy->pub.decode_mcu = decode_mcu_AC_refine; - } - } else { - /* Check that the scan parameters Ss, Se, Ah/Al are OK for sequential JPEG. - * This ought to be an error condition, but we make it a warning. - */ - if (cinfo->Ss != 0 || cinfo->Ah != 0 || cinfo->Al != 0 || - (cinfo->Se < DCTSIZE2 && cinfo->Se != cinfo->lim_Se)) - WARNMS(cinfo, JWRN_NOT_SEQUENTIAL); - /* Select MCU decoding routine */ - entropy->pub.decode_mcu = decode_mcu; - } - - /* Allocate & initialize requested statistics areas */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - if (! cinfo->progressive_mode || (cinfo->Ss == 0 && cinfo->Ah == 0)) { - tbl = compptr->dc_tbl_no; - if (tbl < 0 || tbl >= NUM_ARITH_TBLS) - ERREXIT1(cinfo, JERR_NO_ARITH_TABLE, tbl); - if (entropy->dc_stats[tbl] == NULL) - entropy->dc_stats[tbl] = (unsigned char *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, DC_STAT_BINS); - MEMZERO(entropy->dc_stats[tbl], DC_STAT_BINS); - /* Initialize DC predictions to 0 */ - entropy->last_dc_val[ci] = 0; - entropy->dc_context[ci] = 0; - } - if ((! cinfo->progressive_mode && cinfo->lim_Se) || - (cinfo->progressive_mode && cinfo->Ss)) { - tbl = compptr->ac_tbl_no; - if (tbl < 0 || tbl >= NUM_ARITH_TBLS) - ERREXIT1(cinfo, JERR_NO_ARITH_TABLE, tbl); - if (entropy->ac_stats[tbl] == NULL) - entropy->ac_stats[tbl] = (unsigned char *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, AC_STAT_BINS); - MEMZERO(entropy->ac_stats[tbl], AC_STAT_BINS); - } - } - - /* Initialize arithmetic decoding variables */ - entropy->c = 0; - entropy->a = 0; - entropy->ct = -16; /* force reading 2 initial bytes to fill C */ - - /* Initialize restart counter */ - entropy->restarts_to_go = cinfo->restart_interval; -} - - -/* - * Finish up at the end of an arithmetic-compressed scan. - */ - -METHODDEF(void) -finish_pass (j_decompress_ptr cinfo) -{ - /* no work necessary here */ -} - - -/* - * Module initialization routine for arithmetic entropy decoding. - */ - -GLOBAL(void) -jinit_arith_decoder (j_decompress_ptr cinfo) -{ - arith_entropy_ptr entropy; - int i; - - entropy = (arith_entropy_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(arith_entropy_decoder)); - cinfo->entropy = &entropy->pub; - entropy->pub.start_pass = start_pass; - entropy->pub.finish_pass = finish_pass; - - /* Mark tables unallocated */ - for (i = 0; i < NUM_ARITH_TBLS; i++) { - entropy->dc_stats[i] = NULL; - entropy->ac_stats[i] = NULL; - } - - /* Initialize index for fixed probability estimation */ - entropy->fixed_bin[0] = 113; - - if (cinfo->progressive_mode) { - /* Create progression status table */ - int *coef_bit_ptr, ci; - cinfo->coef_bits = (int (*)[DCTSIZE2]) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - cinfo->num_components * DCTSIZE2 * SIZEOF(int)); - coef_bit_ptr = & cinfo->coef_bits[0][0]; - for (ci = 0; ci < cinfo->num_components; ci++) - for (i = 0; i < DCTSIZE2; i++) - *coef_bit_ptr++ = -1; - } -} diff --git a/dep/libjpeg/src/jdatadst.c b/dep/libjpeg/src/jdatadst.c deleted file mode 100644 index b3b4798ea..000000000 --- a/dep/libjpeg/src/jdatadst.c +++ /dev/null @@ -1,263 +0,0 @@ -/* - * jdatadst.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2009-2022 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains compression data destination routines for the case of - * emitting JPEG data to memory or to a file (or any stdio stream). - * While these routines are sufficient for most applications, - * some will want to use a different destination manager. - * IMPORTANT: we assume that fwrite() will correctly transcribe an array of - * JOCTETs into 8-bit-wide elements on external storage. If char is wider - * than 8 bits on your machine, you may need to do some tweaking. - */ - -/* this is not a core library module, so it doesn't define JPEG_INTERNALS */ -#include "jinclude.h" -#include "jpeglib.h" -#include "jerror.h" - -#ifndef HAVE_STDLIB_H /* should declare malloc(),free() */ -extern void * malloc JPP((size_t size)); -extern void free JPP((void *ptr)); -#endif - - -/* Expanded data destination object for stdio output */ - -#define OUTPUT_BUF_SIZE 4096 /* choose an efficiently fwrite'able size */ - -typedef struct { - struct jpeg_destination_mgr pub; /* public fields */ - - FILE * outfile; /* target stream */ - JOCTET buffer[OUTPUT_BUF_SIZE]; /* output buffer */ -} my_destination_mgr; - -typedef my_destination_mgr * my_dest_ptr; - - -/* Expanded data destination object for memory output */ - -typedef struct { - struct jpeg_destination_mgr pub; /* public fields */ - - unsigned char ** outbuffer; /* target buffer */ - size_t * outsize; - unsigned char * newbuffer; /* newly allocated buffer */ - JOCTET * buffer; /* start of buffer */ - size_t bufsize; -} my_mem_destination_mgr; - -typedef my_mem_destination_mgr * my_mem_dest_ptr; - - -/* - * Initialize destination --- called by jpeg_start_compress - * before any data is actually written. - */ - -METHODDEF(void) -init_destination (j_compress_ptr cinfo) -{ - my_dest_ptr dest = (my_dest_ptr) cinfo->dest; - - dest->pub.next_output_byte = dest->buffer; - dest->pub.free_in_buffer = OUTPUT_BUF_SIZE; -} - -METHODDEF(void) -init_mem_destination (j_compress_ptr cinfo) -{ - /* no work necessary here */ -} - - -/* - * Empty the output buffer --- called whenever buffer fills up. - * - * In typical applications, this should write the entire output buffer - * (ignoring the current state of next_output_byte & free_in_buffer), - * reset the pointer & count to the start of the buffer, and return TRUE - * indicating that the buffer has been dumped. - * - * In applications that need to be able to suspend compression due to output - * overrun, a FALSE return indicates that the buffer cannot be emptied now. - * In this situation, the compressor will return to its caller (possibly with - * an indication that it has not accepted all the supplied scanlines). The - * application should resume compression after it has made more room in the - * output buffer. Note that there are substantial restrictions on the use of - * suspension --- see the documentation. - * - * When suspending, the compressor will back up to a convenient restart point - * (typically the start of the current MCU). next_output_byte & free_in_buffer - * indicate where the restart point will be if the current call returns FALSE. - * Data beyond this point will be regenerated after resumption, so do not - * write it out when emptying the buffer externally. - */ - -METHODDEF(boolean) -empty_output_buffer (j_compress_ptr cinfo) -{ - my_dest_ptr dest = (my_dest_ptr) cinfo->dest; - - if (JFWRITE(dest->outfile, dest->buffer, OUTPUT_BUF_SIZE) != - (size_t) OUTPUT_BUF_SIZE) - ERREXIT(cinfo, JERR_FILE_WRITE); - - dest->pub.next_output_byte = dest->buffer; - dest->pub.free_in_buffer = OUTPUT_BUF_SIZE; - - return TRUE; -} - -METHODDEF(boolean) -empty_mem_output_buffer (j_compress_ptr cinfo) -{ - size_t nextsize; - JOCTET * nextbuffer; - my_mem_dest_ptr dest = (my_mem_dest_ptr) cinfo->dest; - - /* Try to allocate new buffer with double size */ - nextsize = dest->bufsize * 2; - nextbuffer = (JOCTET *) malloc(nextsize); - - if (nextbuffer == NULL) - ERREXIT1(cinfo, JERR_OUT_OF_MEMORY, 11); - - MEMCOPY(nextbuffer, dest->buffer, dest->bufsize); - - if (dest->newbuffer != NULL) - free(dest->newbuffer); - - dest->newbuffer = nextbuffer; - - dest->pub.next_output_byte = nextbuffer + dest->bufsize; - dest->pub.free_in_buffer = dest->bufsize; - - dest->buffer = nextbuffer; - dest->bufsize = nextsize; - - return TRUE; -} - - -/* - * Terminate destination --- called by jpeg_finish_compress - * after all data has been written. Usually needs to flush buffer. - * - * NB: *not* called by jpeg_abort or jpeg_destroy; surrounding - * application must deal with any cleanup that should happen even - * for error exit. - */ - -METHODDEF(void) -term_destination (j_compress_ptr cinfo) -{ - my_dest_ptr dest = (my_dest_ptr) cinfo->dest; - size_t datacount = OUTPUT_BUF_SIZE - dest->pub.free_in_buffer; - - /* Write any data remaining in the buffer */ - if (datacount > 0) { - if (JFWRITE(dest->outfile, dest->buffer, datacount) != datacount) - ERREXIT(cinfo, JERR_FILE_WRITE); - } - JFFLUSH(dest->outfile); - /* Make sure we wrote the output file OK */ - if (JFERROR(dest->outfile)) - ERREXIT(cinfo, JERR_FILE_WRITE); -} - -METHODDEF(void) -term_mem_destination (j_compress_ptr cinfo) -{ - my_mem_dest_ptr dest = (my_mem_dest_ptr) cinfo->dest; - - *dest->outbuffer = dest->buffer; - *dest->outsize = dest->bufsize - dest->pub.free_in_buffer; -} - - -/* - * Prepare for output to a stdio stream. - * The caller must have already opened the stream, - * and is responsible for closing it after finishing compression. - */ - -GLOBAL(void) -jpeg_stdio_dest (j_compress_ptr cinfo, FILE * outfile) -{ - my_dest_ptr dest; - - /* The destination object is made permanent so that multiple JPEG images - * can be written to the same file without re-executing jpeg_stdio_dest. - * This makes it dangerous to use this manager and a different destination - * manager serially with the same JPEG object, because their private object - * sizes may be different. Caveat programmer. - */ - if (cinfo->dest == NULL) { /* first time for this JPEG object? */ - cinfo->dest = (struct jpeg_destination_mgr *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_PERMANENT, SIZEOF(my_destination_mgr)); - } - - dest = (my_dest_ptr) cinfo->dest; - dest->pub.init_destination = init_destination; - dest->pub.empty_output_buffer = empty_output_buffer; - dest->pub.term_destination = term_destination; - dest->outfile = outfile; -} - - -/* - * Prepare for output to a memory buffer. - * The caller may supply an own initial buffer with appropriate size. - * Otherwise, or when the actual data output exceeds the given size, - * the library adapts the buffer size as necessary. - * The standard library functions malloc/free are used for allocating - * larger memory, so the buffer is available to the application after - * finishing compression, and then the application is responsible for - * freeing the requested memory. - * Note: An initial buffer supplied by the caller is expected to be - * managed by the application. The library does not free such buffer - * when allocating a larger buffer. - */ - -GLOBAL(void) -jpeg_mem_dest (j_compress_ptr cinfo, - unsigned char ** outbuffer, size_t * outsize) -{ - my_mem_dest_ptr dest; - - if (outbuffer == NULL || outsize == NULL) /* sanity check */ - ERREXIT(cinfo, JERR_BUFFER_SIZE); - - /* The destination object is made permanent so that multiple JPEG images - * can be written to the same buffer without re-executing jpeg_mem_dest. - */ - if (cinfo->dest == NULL) { /* first time for this JPEG object? */ - cinfo->dest = (struct jpeg_destination_mgr *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_PERMANENT, SIZEOF(my_mem_destination_mgr)); - } - - dest = (my_mem_dest_ptr) cinfo->dest; - dest->pub.init_destination = init_mem_destination; - dest->pub.empty_output_buffer = empty_mem_output_buffer; - dest->pub.term_destination = term_mem_destination; - dest->outbuffer = outbuffer; - dest->outsize = outsize; - dest->newbuffer = NULL; - - if (*outbuffer == NULL || *outsize == 0) { - /* Allocate initial buffer */ - dest->newbuffer = *outbuffer = (unsigned char *) malloc(OUTPUT_BUF_SIZE); - if (dest->newbuffer == NULL) - ERREXIT1(cinfo, JERR_OUT_OF_MEMORY, 10); - *outsize = OUTPUT_BUF_SIZE; - } - - dest->pub.next_output_byte = dest->buffer = *outbuffer; - dest->pub.free_in_buffer = dest->bufsize = *outsize; -} diff --git a/dep/libjpeg/src/jdatasrc.c b/dep/libjpeg/src/jdatasrc.c deleted file mode 100644 index fd7a1a594..000000000 --- a/dep/libjpeg/src/jdatasrc.c +++ /dev/null @@ -1,271 +0,0 @@ -/* - * jdatasrc.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2009-2022 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains decompression data source routines for the case of - * reading JPEG data from memory or from a file (or any stdio stream). - * While these routines are sufficient for most applications, - * some will want to use a different source manager. - * IMPORTANT: we assume that fread() will correctly transcribe an array of - * JOCTETs from 8-bit-wide elements on external storage. If char is wider - * than 8 bits on your machine, you may need to do some tweaking. - */ - -/* this is not a core library module, so it doesn't define JPEG_INTERNALS */ -#include "jinclude.h" -#include "jpeglib.h" -#include "jerror.h" - - -/* Expanded data source object for stdio input */ - -#define INPUT_BUF_SIZE 4096 /* choose an efficiently fread'able size */ - -typedef struct { - struct jpeg_source_mgr pub; /* public fields */ - - FILE * infile; /* source stream */ - JOCTET buffer[INPUT_BUF_SIZE]; /* input buffer */ - boolean start_of_file; /* have we gotten any data yet? */ -} my_source_mgr; - -typedef my_source_mgr * my_src_ptr; - - -/* - * Initialize source --- called by jpeg_read_header - * before any data is actually read. - */ - -METHODDEF(void) -init_source (j_decompress_ptr cinfo) -{ - my_src_ptr src = (my_src_ptr) cinfo->src; - - /* We reset the empty-input-file flag for each image, - * but we don't clear the input buffer. - * This is correct behavior for reading a series of images from one source. - */ - src->start_of_file = TRUE; -} - -METHODDEF(void) -init_mem_source (j_decompress_ptr cinfo) -{ - /* no work necessary here */ -} - - -/* - * Fill the input buffer --- called whenever buffer is emptied. - * - * In typical applications, this should read fresh data into the buffer - * (ignoring the current state of next_input_byte & bytes_in_buffer), - * reset the pointer & count to the start of the buffer, and return TRUE - * indicating that the buffer has been reloaded. It is not necessary to - * fill the buffer entirely, only to obtain at least one more byte. - * - * There is no such thing as an EOF return. If the end of the file has been - * reached, the routine has a choice of ERREXIT() or inserting fake data into - * the buffer. In most cases, generating a warning message and inserting a - * fake EOI marker is the best course of action --- this will allow the - * decompressor to output however much of the image is there. However, - * the resulting error message is misleading if the real problem is an empty - * input file, so we handle that case specially. - * - * In applications that need to be able to suspend compression due to input - * not being available yet, a FALSE return indicates that no more data can be - * obtained right now, but more may be forthcoming later. In this situation, - * the decompressor will return to its caller (with an indication of the - * number of scanlines it has read, if any). The application should resume - * decompression after it has loaded more data into the input buffer. Note - * that there are substantial restrictions on the use of suspension --- see - * the documentation. - * - * When suspending, the decompressor will back up to a convenient restart point - * (typically the start of the current MCU). next_input_byte & bytes_in_buffer - * indicate where the restart point will be if the current call returns FALSE. - * Data beyond this point must be rescanned after resumption, so move it to - * the front of the buffer rather than discarding it. - */ - -METHODDEF(boolean) -fill_input_buffer (j_decompress_ptr cinfo) -{ - my_src_ptr src = (my_src_ptr) cinfo->src; - size_t nbytes; - - nbytes = JFREAD(src->infile, src->buffer, INPUT_BUF_SIZE); - - if (nbytes <= 0) { - if (src->start_of_file) /* Treat empty input file as fatal error */ - ERREXIT(cinfo, JERR_INPUT_EMPTY); - WARNMS(cinfo, JWRN_JPEG_EOF); - /* Insert a fake EOI marker */ - src->buffer[0] = (JOCTET) 0xFF; - src->buffer[1] = (JOCTET) JPEG_EOI; - nbytes = 2; - } - - src->pub.next_input_byte = src->buffer; - src->pub.bytes_in_buffer = nbytes; - src->start_of_file = FALSE; - - return TRUE; -} - -METHODDEF(boolean) -fill_mem_input_buffer (j_decompress_ptr cinfo) -{ - static const JOCTET mybuffer[4] = { - (JOCTET) 0xFF, (JOCTET) JPEG_EOI, 0, 0 - }; - - /* The whole JPEG data is expected to reside in the supplied memory - * buffer, so any request for more data beyond the given buffer size - * is treated as an error. - */ - WARNMS(cinfo, JWRN_JPEG_EOF); - - /* Insert a fake EOI marker */ - - cinfo->src->next_input_byte = mybuffer; - cinfo->src->bytes_in_buffer = 2; - - return TRUE; -} - - -/* - * Skip data --- used to skip over a potentially large amount of - * uninteresting data (such as an APPn marker). - * - * Writers of suspendable-input applications must note that skip_input_data - * is not granted the right to give a suspension return. If the skip extends - * beyond the data currently in the buffer, the buffer can be marked empty so - * that the next read will cause a fill_input_buffer call that can suspend. - * Arranging for additional bytes to be discarded before reloading the input - * buffer is the application writer's problem. - */ - -METHODDEF(void) -skip_input_data (j_decompress_ptr cinfo, long num_bytes) -{ - struct jpeg_source_mgr * src = cinfo->src; - size_t nbytes; - - /* Just a dumb implementation for now. Could use fseek() except - * it doesn't work on pipes. Not clear that being smart is worth - * any trouble anyway --- large skips are infrequent. - */ - if (num_bytes > 0) { - nbytes = (size_t) num_bytes; - while (nbytes > src->bytes_in_buffer) { - nbytes -= src->bytes_in_buffer; - (void) (*src->fill_input_buffer) (cinfo); - /* note we assume that fill_input_buffer will never return FALSE, - * so suspension need not be handled. - */ - } - src->next_input_byte += nbytes; - src->bytes_in_buffer -= nbytes; - } -} - - -/* - * An additional method that can be provided by data source modules is the - * resync_to_restart method for error recovery in the presence of RST markers. - * For the moment, this source module just uses the default resync method - * provided by the JPEG library. That method assumes that no backtracking - * is possible. - */ - - -/* - * Terminate source --- called by jpeg_finish_decompress - * after all data has been read. Often a no-op. - * - * NB: *not* called by jpeg_abort or jpeg_destroy; surrounding - * application must deal with any cleanup that should happen even - * for error exit. - */ - -METHODDEF(void) -term_source (j_decompress_ptr cinfo) -{ - /* no work necessary here */ -} - - -/* - * Prepare for input from a stdio stream. - * The caller must have already opened the stream, - * and is responsible for closing it after finishing decompression. - */ - -GLOBAL(void) -jpeg_stdio_src (j_decompress_ptr cinfo, FILE * infile) -{ - my_src_ptr src; - - /* The source object including the input buffer is made permanent so that - * a series of JPEG images can be read from the same file by calling - * jpeg_stdio_src only before the first one. (If we discarded the buffer - * at the end of one image, we'd likely lose the start of the next one.) - * This makes it unsafe to use this manager and a different source - * manager serially with the same JPEG object. Caveat programmer. - */ - if (cinfo->src == NULL) { /* first time for this JPEG object? */ - cinfo->src = (struct jpeg_source_mgr *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_PERMANENT, SIZEOF(my_source_mgr)); - } - - src = (my_src_ptr) cinfo->src; - src->pub.init_source = init_source; - src->pub.fill_input_buffer = fill_input_buffer; - src->pub.skip_input_data = skip_input_data; - src->pub.resync_to_restart = jpeg_resync_to_restart; /* use default method */ - src->pub.term_source = term_source; - src->infile = infile; - src->pub.bytes_in_buffer = 0; /* forces fill_input_buffer on first read */ - src->pub.next_input_byte = NULL; /* until buffer loaded */ -} - - -/* - * Prepare for input from a supplied memory buffer. - * The buffer must contain the whole JPEG data. - */ - -GLOBAL(void) -jpeg_mem_src (j_decompress_ptr cinfo, - const unsigned char * inbuffer, size_t insize) -{ - struct jpeg_source_mgr * src; - - if (inbuffer == NULL || insize == 0) /* Treat empty input as fatal error */ - ERREXIT(cinfo, JERR_INPUT_EMPTY); - - /* The source object is made permanent so that a series of JPEG images - * can be read from the same buffer by calling jpeg_mem_src only before - * the first one. - */ - if (cinfo->src == NULL) { /* first time for this JPEG object? */ - cinfo->src = (struct jpeg_source_mgr *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_PERMANENT, SIZEOF(struct jpeg_source_mgr)); - } - - src = cinfo->src; - src->init_source = init_mem_source; - src->fill_input_buffer = fill_mem_input_buffer; - src->skip_input_data = skip_input_data; - src->resync_to_restart = jpeg_resync_to_restart; /* use default method */ - src->term_source = term_source; - src->bytes_in_buffer = insize; - src->next_input_byte = (const JOCTET *) inbuffer; -} diff --git a/dep/libjpeg/src/jdcoefct.c b/dep/libjpeg/src/jdcoefct.c deleted file mode 100644 index 79ba42014..000000000 --- a/dep/libjpeg/src/jdcoefct.c +++ /dev/null @@ -1,744 +0,0 @@ -/* - * jdcoefct.c - * - * Copyright (C) 1994-1997, Thomas G. Lane. - * Modified 2002-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains the coefficient buffer controller for decompression. - * This controller is the top level of the JPEG decompressor proper. - * The coefficient buffer lies between entropy decoding and inverse-DCT steps. - * - * In buffered-image mode, this controller is the interface between - * input-oriented processing and output-oriented processing. - * Also, the input side (only) is used when reading a file for transcoding. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Block smoothing is only applicable for progressive JPEG, so: */ -#ifndef D_PROGRESSIVE_SUPPORTED -#undef BLOCK_SMOOTHING_SUPPORTED -#endif - - -/* Private buffer controller object */ - -typedef struct { - struct jpeg_d_coef_controller pub; /* public fields */ - - /* These variables keep track of the current location of the input side. */ - /* cinfo->input_iMCU_row is also used for this. */ - JDIMENSION MCU_ctr; /* counts MCUs processed in current row */ - int MCU_vert_offset; /* counts MCU rows within iMCU row */ - int MCU_rows_per_iMCU_row; /* number of such rows needed */ - - /* The output side's location is represented by cinfo->output_iMCU_row. */ - - /* In single-pass modes, it's sufficient to buffer just one MCU. - * We append a workspace of D_MAX_BLOCKS_IN_MCU coefficient blocks, - * and let the entropy decoder write into that workspace each time. - * In multi-pass modes, this array points to the current MCU's blocks - * within the virtual arrays; it is used only by the input side. - */ - JBLOCKROW MCU_buffer[D_MAX_BLOCKS_IN_MCU]; - -#ifdef D_MULTISCAN_FILES_SUPPORTED - /* In multi-pass modes, we need a virtual block array for each component. */ - jvirt_barray_ptr whole_image[MAX_COMPONENTS]; -#endif - -#ifdef BLOCK_SMOOTHING_SUPPORTED - /* When doing block smoothing, we latch coefficient Al values here */ - int * coef_bits_latch; -#define SAVED_COEFS 6 /* we save coef_bits[0..5] */ -#endif - - /* Workspace for single-pass modes (omitted otherwise). */ - JBLOCK blk_buffer[D_MAX_BLOCKS_IN_MCU]; -} my_coef_controller; - -typedef my_coef_controller * my_coef_ptr; - - -/* Forward declarations */ -METHODDEF(int) decompress_onepass - JPP((j_decompress_ptr cinfo, JSAMPIMAGE output_buf)); -#ifdef D_MULTISCAN_FILES_SUPPORTED -METHODDEF(int) decompress_data - JPP((j_decompress_ptr cinfo, JSAMPIMAGE output_buf)); -#endif -#ifdef BLOCK_SMOOTHING_SUPPORTED -LOCAL(boolean) smoothing_ok JPP((j_decompress_ptr cinfo)); -METHODDEF(int) decompress_smooth_data - JPP((j_decompress_ptr cinfo, JSAMPIMAGE output_buf)); -#endif - - -LOCAL(void) -start_iMCU_row (j_decompress_ptr cinfo) -/* Reset within-iMCU-row counters for a new row (input side) */ -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - - /* In an interleaved scan, an MCU row is the same as an iMCU row. - * In a noninterleaved scan, an iMCU row has v_samp_factor MCU rows. - * But at the bottom of the image, process only what's left. - */ - if (cinfo->comps_in_scan > 1) { - coef->MCU_rows_per_iMCU_row = 1; - } else { - if (cinfo->input_iMCU_row < (cinfo->total_iMCU_rows-1)) - coef->MCU_rows_per_iMCU_row = cinfo->cur_comp_info[0]->v_samp_factor; - else - coef->MCU_rows_per_iMCU_row = cinfo->cur_comp_info[0]->last_row_height; - } - - coef->MCU_ctr = 0; - coef->MCU_vert_offset = 0; -} - - -/* - * Initialize for an input processing pass. - */ - -METHODDEF(void) -start_input_pass (j_decompress_ptr cinfo) -{ - cinfo->input_iMCU_row = 0; - start_iMCU_row(cinfo); -} - - -/* - * Initialize for an output processing pass. - */ - -METHODDEF(void) -start_output_pass (j_decompress_ptr cinfo) -{ -#ifdef BLOCK_SMOOTHING_SUPPORTED - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - - /* If multipass, check to see whether to use block smoothing on this pass */ - if (coef->pub.coef_arrays != NULL) { - if (cinfo->do_block_smoothing && smoothing_ok(cinfo)) - coef->pub.decompress_data = decompress_smooth_data; - else - coef->pub.decompress_data = decompress_data; - } -#endif - cinfo->output_iMCU_row = 0; -} - - -/* - * Decompress and return some data in the single-pass case. - * Always attempts to emit one fully interleaved MCU row ("iMCU" row). - * Input and output must run in lockstep since we have only a one-MCU buffer. - * Return value is JPEG_ROW_COMPLETED, JPEG_SCAN_COMPLETED, or JPEG_SUSPENDED. - * - * NB: output_buf contains a plane for each component in image, - * which we index according to the component's SOF position. - */ - -METHODDEF(int) -decompress_onepass (j_decompress_ptr cinfo, JSAMPIMAGE output_buf) -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - JDIMENSION MCU_col_num; /* index of current MCU within row */ - JDIMENSION last_MCU_col = cinfo->MCUs_per_row - 1; - JDIMENSION last_iMCU_row = cinfo->total_iMCU_rows - 1; - int ci, xindex, yindex, yoffset, useful_width; - JBLOCKROW blkp; - JSAMPARRAY output_ptr; - JDIMENSION start_col, output_col; - jpeg_component_info *compptr; - inverse_DCT_method_ptr inverse_DCT; - - /* Loop to process as much as one whole iMCU row */ - for (yoffset = coef->MCU_vert_offset; yoffset < coef->MCU_rows_per_iMCU_row; - yoffset++) { - for (MCU_col_num = coef->MCU_ctr; MCU_col_num <= last_MCU_col; - MCU_col_num++) { - blkp = coef->blk_buffer; /* pointer to current DCT block within MCU */ - /* Try to fetch an MCU. Entropy decoder expects buffer to be zeroed. */ - if (cinfo->lim_Se) /* can bypass in DC only case */ - MEMZERO(blkp, cinfo->blocks_in_MCU * SIZEOF(JBLOCK)); - if (! (*cinfo->entropy->decode_mcu) (cinfo, coef->MCU_buffer)) { - /* Suspension forced; update state counters and exit */ - coef->MCU_vert_offset = yoffset; - coef->MCU_ctr = MCU_col_num; - return JPEG_SUSPENDED; - } - /* Determine where data should go in output_buf and do the IDCT thing. - * We skip dummy blocks at the right and bottom edges (but blkp gets - * incremented past them!). - */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - /* Don't bother to IDCT an uninteresting component. */ - if (! compptr->component_needed) { - blkp += compptr->MCU_blocks; - continue; - } - inverse_DCT = cinfo->idct->inverse_DCT[compptr->component_index]; - output_ptr = output_buf[compptr->component_index] + - yoffset * compptr->DCT_v_scaled_size; - useful_width = (MCU_col_num < last_MCU_col) ? compptr->MCU_width - : compptr->last_col_width; - start_col = MCU_col_num * compptr->MCU_sample_width; - for (yindex = 0; yindex < compptr->MCU_height; yindex++) { - if (cinfo->input_iMCU_row < last_iMCU_row || - yoffset + yindex < compptr->last_row_height) { - output_col = start_col; - for (xindex = 0; xindex < useful_width; xindex++) { - (*inverse_DCT) (cinfo, compptr, (JCOEFPTR) (blkp + xindex), - output_ptr, output_col); - output_col += compptr->DCT_h_scaled_size; - } - output_ptr += compptr->DCT_v_scaled_size; - } - blkp += compptr->MCU_width; - } - } - } - /* Completed an MCU row, but perhaps not an iMCU row */ - coef->MCU_ctr = 0; - } - /* Completed the iMCU row, advance counters for next one */ - cinfo->output_iMCU_row++; - if (++(cinfo->input_iMCU_row) <= last_iMCU_row) { - start_iMCU_row(cinfo); - return JPEG_ROW_COMPLETED; - } - /* Completed the scan */ - (*cinfo->inputctl->finish_input_pass) (cinfo); - return JPEG_SCAN_COMPLETED; -} - - -/* - * Dummy consume-input routine for single-pass operation. - */ - -METHODDEF(int) -dummy_consume_data (j_decompress_ptr cinfo) -{ - return JPEG_SUSPENDED; /* Always indicate nothing was done */ -} - - -#ifdef D_MULTISCAN_FILES_SUPPORTED - -/* - * Consume input data and store it in the full-image coefficient buffer. - * We read as much as one fully interleaved MCU row ("iMCU" row) per call, - * ie, v_samp_factor block rows for each component in the scan. - * Return value is JPEG_ROW_COMPLETED, JPEG_SCAN_COMPLETED, or JPEG_SUSPENDED. - */ - -METHODDEF(int) -consume_data (j_decompress_ptr cinfo) -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - JDIMENSION MCU_col_num; /* index of current MCU within row */ - int ci, xindex, yindex, yoffset; - JDIMENSION start_col; - JBLOCKARRAY blkp; - JBLOCKARRAY buffer[MAX_COMPS_IN_SCAN]; - JBLOCKROW buffer_ptr; - jpeg_component_info *compptr; - - /* Align the virtual buffers for the components used in this scan. */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - buffer[ci] = (*cinfo->mem->access_virt_barray) - ((j_common_ptr) cinfo, coef->whole_image[compptr->component_index], - cinfo->input_iMCU_row * compptr->v_samp_factor, - (JDIMENSION) compptr->v_samp_factor, TRUE); - /* Note: entropy decoder expects buffer to be zeroed, - * but this is handled automatically by the memory manager - * because we requested a pre-zeroed array. - */ - } - - /* Loop to process one whole iMCU row */ - for (yoffset = coef->MCU_vert_offset; yoffset < coef->MCU_rows_per_iMCU_row; - yoffset++) { - for (MCU_col_num = coef->MCU_ctr; MCU_col_num < cinfo->MCUs_per_row; - MCU_col_num++) { - /* Construct list of pointers to DCT blocks belonging to this MCU */ - blkp = coef->MCU_buffer; /* pointer to current DCT block within MCU */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - start_col = MCU_col_num * compptr->MCU_width; - for (yindex = 0; yindex < compptr->MCU_height; yindex++) { - buffer_ptr = buffer[ci][yoffset + yindex] + start_col; - xindex = compptr->MCU_width; - do { - *blkp++ = buffer_ptr++; - } while (--xindex); - } - } - /* Try to fetch the MCU. */ - if (! (*cinfo->entropy->decode_mcu) (cinfo, coef->MCU_buffer)) { - /* Suspension forced; update state counters and exit */ - coef->MCU_vert_offset = yoffset; - coef->MCU_ctr = MCU_col_num; - return JPEG_SUSPENDED; - } - } - /* Completed an MCU row, but perhaps not an iMCU row */ - coef->MCU_ctr = 0; - } - /* Completed the iMCU row, advance counters for next one */ - if (++(cinfo->input_iMCU_row) < cinfo->total_iMCU_rows) { - start_iMCU_row(cinfo); - return JPEG_ROW_COMPLETED; - } - /* Completed the scan */ - (*cinfo->inputctl->finish_input_pass) (cinfo); - return JPEG_SCAN_COMPLETED; -} - - -/* - * Decompress and return some data in the multi-pass case. - * Always attempts to emit one fully interleaved MCU row ("iMCU" row). - * Return value is JPEG_ROW_COMPLETED, JPEG_SCAN_COMPLETED, or JPEG_SUSPENDED. - * - * NB: output_buf contains a plane for each component in image. - */ - -METHODDEF(int) -decompress_data (j_decompress_ptr cinfo, JSAMPIMAGE output_buf) -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - JDIMENSION last_iMCU_row = cinfo->total_iMCU_rows - 1; - JDIMENSION block_num; - int ci, block_row, block_rows; - JBLOCKARRAY buffer; - JBLOCKROW buffer_ptr; - JSAMPARRAY output_ptr; - JDIMENSION output_col; - jpeg_component_info *compptr; - inverse_DCT_method_ptr inverse_DCT; - - /* Force some input to be done if we are getting ahead of the input. */ - while (cinfo->input_scan_number < cinfo->output_scan_number || - (cinfo->input_scan_number == cinfo->output_scan_number && - cinfo->input_iMCU_row <= cinfo->output_iMCU_row)) { - if ((*cinfo->inputctl->consume_input)(cinfo) == JPEG_SUSPENDED) - return JPEG_SUSPENDED; - } - - /* OK, output from the virtual arrays. */ - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* Don't bother to IDCT an uninteresting component. */ - if (! compptr->component_needed) - continue; - /* Align the virtual buffer for this component. */ - buffer = (*cinfo->mem->access_virt_barray) - ((j_common_ptr) cinfo, coef->whole_image[ci], - cinfo->output_iMCU_row * compptr->v_samp_factor, - (JDIMENSION) compptr->v_samp_factor, FALSE); - /* Count non-dummy DCT block rows in this iMCU row. */ - if (cinfo->output_iMCU_row < last_iMCU_row) - block_rows = compptr->v_samp_factor; - else { - /* NB: can't use last_row_height here; it is input-side-dependent! */ - block_rows = (int) (compptr->height_in_blocks % compptr->v_samp_factor); - if (block_rows == 0) block_rows = compptr->v_samp_factor; - } - inverse_DCT = cinfo->idct->inverse_DCT[ci]; - output_ptr = output_buf[ci]; - /* Loop over all DCT blocks to be processed. */ - for (block_row = 0; block_row < block_rows; block_row++) { - buffer_ptr = buffer[block_row]; - output_col = 0; - for (block_num = 0; block_num < compptr->width_in_blocks; block_num++) { - (*inverse_DCT) (cinfo, compptr, (JCOEFPTR) buffer_ptr, - output_ptr, output_col); - buffer_ptr++; - output_col += compptr->DCT_h_scaled_size; - } - output_ptr += compptr->DCT_v_scaled_size; - } - } - - if (++(cinfo->output_iMCU_row) <= last_iMCU_row) - return JPEG_ROW_COMPLETED; - return JPEG_SCAN_COMPLETED; -} - -#endif /* D_MULTISCAN_FILES_SUPPORTED */ - - -#ifdef BLOCK_SMOOTHING_SUPPORTED - -/* - * This code applies interblock smoothing as described by section K.8 - * of the JPEG standard: the first 5 AC coefficients are estimated from - * the DC values of a DCT block and its 8 neighboring blocks. - * We apply smoothing only for progressive JPEG decoding, and only if - * the coefficients it can estimate are not yet known to full precision. - */ - -/* Natural-order array positions of the first 5 zigzag-order coefficients */ -#define Q01_POS 1 -#define Q10_POS 8 -#define Q20_POS 16 -#define Q11_POS 9 -#define Q02_POS 2 - -/* - * Determine whether block smoothing is applicable and safe. - * We also latch the current states of the coef_bits[] entries for the - * AC coefficients; otherwise, if the input side of the decompressor - * advances into a new scan, we might think the coefficients are known - * more accurately than they really are. - */ - -LOCAL(boolean) -smoothing_ok (j_decompress_ptr cinfo) -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - boolean smoothing_useful = FALSE; - int ci, coefi; - jpeg_component_info *compptr; - JQUANT_TBL * qtable; - int * coef_bits; - int * coef_bits_latch; - - if (! cinfo->progressive_mode || cinfo->coef_bits == NULL) - return FALSE; - - /* Allocate latch area if not already done */ - if (coef->coef_bits_latch == NULL) - coef->coef_bits_latch = (int *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - cinfo->num_components * (SAVED_COEFS * SIZEOF(int))); - coef_bits_latch = coef->coef_bits_latch; - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* All components' quantization values must already be latched. */ - if ((qtable = compptr->quant_table) == NULL) - return FALSE; - /* Verify DC & first 5 AC quantizers are nonzero to avoid zero-divide. */ - if (qtable->quantval[0] == 0 || - qtable->quantval[Q01_POS] == 0 || - qtable->quantval[Q10_POS] == 0 || - qtable->quantval[Q20_POS] == 0 || - qtable->quantval[Q11_POS] == 0 || - qtable->quantval[Q02_POS] == 0) - return FALSE; - /* DC values must be at least partly known for all components. */ - coef_bits = cinfo->coef_bits[ci]; - if (coef_bits[0] < 0) - return FALSE; - /* Block smoothing is helpful if some AC coefficients remain inaccurate. */ - for (coefi = 1; coefi <= 5; coefi++) { - coef_bits_latch[coefi] = coef_bits[coefi]; - if (coef_bits[coefi] != 0) - smoothing_useful = TRUE; - } - coef_bits_latch += SAVED_COEFS; - } - - return smoothing_useful; -} - - -/* - * Variant of decompress_data for use when doing block smoothing. - */ - -METHODDEF(int) -decompress_smooth_data (j_decompress_ptr cinfo, JSAMPIMAGE output_buf) -{ - my_coef_ptr coef = (my_coef_ptr) cinfo->coef; - JDIMENSION last_iMCU_row = cinfo->total_iMCU_rows - 1; - JDIMENSION block_num, last_block_column; - int ci, block_row, block_rows, access_rows; - JBLOCKARRAY buffer; - JBLOCKROW buffer_ptr, prev_block_row, next_block_row; - JSAMPARRAY output_ptr; - JDIMENSION output_col; - jpeg_component_info *compptr; - inverse_DCT_method_ptr inverse_DCT; - boolean first_row, last_row; - JBLOCK workspace; - int *coef_bits; - JQUANT_TBL *quanttbl; - INT32 Q00,Q01,Q02,Q10,Q11,Q20, num; - int DC1,DC2,DC3,DC4,DC5,DC6,DC7,DC8,DC9; - int Al, pred; - - /* Force some input to be done if we are getting ahead of the input. */ - while (cinfo->input_scan_number <= cinfo->output_scan_number && - ! cinfo->inputctl->eoi_reached) { - if (cinfo->input_scan_number == cinfo->output_scan_number) { - /* If input is working on current scan, we ordinarily want it to - * have completed the current row. But if input scan is DC, - * we want it to keep one row ahead so that next block row's DC - * values are up to date. - */ - JDIMENSION delta = (cinfo->Ss == 0) ? 1 : 0; - if (cinfo->input_iMCU_row > cinfo->output_iMCU_row+delta) - break; - } - if ((*cinfo->inputctl->consume_input)(cinfo) == JPEG_SUSPENDED) - return JPEG_SUSPENDED; - } - - /* OK, output from the virtual arrays. */ - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* Don't bother to IDCT an uninteresting component. */ - if (! compptr->component_needed) - continue; - /* Count non-dummy DCT block rows in this iMCU row. */ - if (cinfo->output_iMCU_row < last_iMCU_row) { - block_rows = compptr->v_samp_factor; - access_rows = block_rows * 2; /* this and next iMCU row */ - last_row = FALSE; - } else { - /* NB: can't use last_row_height here; it is input-side-dependent! */ - block_rows = (int) (compptr->height_in_blocks % compptr->v_samp_factor); - if (block_rows == 0) block_rows = compptr->v_samp_factor; - access_rows = block_rows; /* this iMCU row only */ - last_row = TRUE; - } - /* Align the virtual buffer for this component. */ - if (cinfo->output_iMCU_row > 0) { - access_rows += compptr->v_samp_factor; /* prior iMCU row too */ - buffer = (*cinfo->mem->access_virt_barray) - ((j_common_ptr) cinfo, coef->whole_image[ci], - (cinfo->output_iMCU_row - 1) * compptr->v_samp_factor, - (JDIMENSION) access_rows, FALSE); - buffer += compptr->v_samp_factor; /* point to current iMCU row */ - first_row = FALSE; - } else { - buffer = (*cinfo->mem->access_virt_barray) - ((j_common_ptr) cinfo, coef->whole_image[ci], - (JDIMENSION) 0, (JDIMENSION) access_rows, FALSE); - first_row = TRUE; - } - /* Fetch component-dependent info */ - coef_bits = coef->coef_bits_latch + (ci * SAVED_COEFS); - quanttbl = compptr->quant_table; - Q00 = quanttbl->quantval[0]; - Q01 = quanttbl->quantval[Q01_POS]; - Q10 = quanttbl->quantval[Q10_POS]; - Q20 = quanttbl->quantval[Q20_POS]; - Q11 = quanttbl->quantval[Q11_POS]; - Q02 = quanttbl->quantval[Q02_POS]; - inverse_DCT = cinfo->idct->inverse_DCT[ci]; - output_ptr = output_buf[ci]; - /* Loop over all DCT blocks to be processed. */ - for (block_row = 0; block_row < block_rows; block_row++) { - buffer_ptr = buffer[block_row]; - if (first_row && block_row == 0) - prev_block_row = buffer_ptr; - else - prev_block_row = buffer[block_row-1]; - if (last_row && block_row == block_rows-1) - next_block_row = buffer_ptr; - else - next_block_row = buffer[block_row+1]; - /* We fetch the surrounding DC values using a sliding-register approach. - * Initialize all nine here so as to do the right thing on narrow pics. - */ - DC1 = DC2 = DC3 = (int) prev_block_row[0][0]; - DC4 = DC5 = DC6 = (int) buffer_ptr[0][0]; - DC7 = DC8 = DC9 = (int) next_block_row[0][0]; - output_col = 0; - last_block_column = compptr->width_in_blocks - 1; - for (block_num = 0; block_num <= last_block_column; block_num++) { - /* Fetch current DCT block into workspace so we can modify it. */ - jcopy_block_row(buffer_ptr, (JBLOCKROW) workspace, (JDIMENSION) 1); - /* Update DC values */ - if (block_num < last_block_column) { - DC3 = (int) prev_block_row[1][0]; - DC6 = (int) buffer_ptr[1][0]; - DC9 = (int) next_block_row[1][0]; - } - /* Compute coefficient estimates per K.8. - * An estimate is applied only if coefficient is still zero, - * and is not known to be fully accurate. - */ - /* AC01 */ - if ((Al=coef_bits[1]) != 0 && workspace[1] == 0) { - num = 36 * Q00 * (DC4 - DC6); - if (num >= 0) { - pred = (int) (((Q01<<7) + num) / (Q01<<8)); - if (Al > 0 && pred >= (1< 0 && pred >= (1<= 0) { - pred = (int) (((Q10<<7) + num) / (Q10<<8)); - if (Al > 0 && pred >= (1< 0 && pred >= (1<= 0) { - pred = (int) (((Q20<<7) + num) / (Q20<<8)); - if (Al > 0 && pred >= (1< 0 && pred >= (1<= 0) { - pred = (int) (((Q11<<7) + num) / (Q11<<8)); - if (Al > 0 && pred >= (1< 0 && pred >= (1<= 0) { - pred = (int) (((Q02<<7) + num) / (Q02<<8)); - if (Al > 0 && pred >= (1< 0 && pred >= (1<DCT_h_scaled_size; - } - output_ptr += compptr->DCT_v_scaled_size; - } - } - - if (++(cinfo->output_iMCU_row) <= last_iMCU_row) - return JPEG_ROW_COMPLETED; - return JPEG_SCAN_COMPLETED; -} - -#endif /* BLOCK_SMOOTHING_SUPPORTED */ - - -/* - * Initialize coefficient buffer controller. - */ - -GLOBAL(void) -jinit_d_coef_controller (j_decompress_ptr cinfo, boolean need_full_buffer) -{ - my_coef_ptr coef; - - if (need_full_buffer) { -#ifdef D_MULTISCAN_FILES_SUPPORTED - /* Allocate a full-image virtual array for each component, */ - /* padded to a multiple of samp_factor DCT blocks in each direction. */ - /* Note we ask for a pre-zeroed array. */ - int ci, access_rows; - jpeg_component_info *compptr; - - coef = (my_coef_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - SIZEOF(my_coef_controller) - SIZEOF(coef->blk_buffer)); - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - access_rows = compptr->v_samp_factor; -#ifdef BLOCK_SMOOTHING_SUPPORTED - /* If block smoothing could be used, need a bigger window */ - if (cinfo->progressive_mode) - access_rows *= 3; -#endif - coef->whole_image[ci] = (*cinfo->mem->request_virt_barray) - ((j_common_ptr) cinfo, JPOOL_IMAGE, TRUE, - (JDIMENSION) jround_up((long) compptr->width_in_blocks, - (long) compptr->h_samp_factor), - (JDIMENSION) jround_up((long) compptr->height_in_blocks, - (long) compptr->v_samp_factor), - (JDIMENSION) access_rows); - } - coef->pub.consume_data = consume_data; - coef->pub.decompress_data = decompress_data; - coef->pub.coef_arrays = coef->whole_image; /* link to virtual arrays */ -#else - ERREXIT(cinfo, JERR_NOT_COMPILED); -#endif - } else { - /* We only need a single-MCU buffer. */ - JBLOCKARRAY blkp; - JBLOCKROW buffer_ptr; - int bi; - - coef = (my_coef_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_coef_controller)); - buffer_ptr = coef->blk_buffer; - if (cinfo->lim_Se == 0) /* DC only case: want to bypass later */ - MEMZERO(buffer_ptr, SIZEOF(coef->blk_buffer)); - blkp = coef->MCU_buffer; - bi = D_MAX_BLOCKS_IN_MCU; - do { - *blkp++ = buffer_ptr++; - } while (--bi); - coef->pub.consume_data = dummy_consume_data; - coef->pub.decompress_data = decompress_onepass; - coef->pub.coef_arrays = NULL; /* flag for no virtual arrays */ - } - - coef->pub.start_input_pass = start_input_pass; - coef->pub.start_output_pass = start_output_pass; -#ifdef BLOCK_SMOOTHING_SUPPORTED - coef->coef_bits_latch = NULL; -#endif - cinfo->coef = &coef->pub; -} diff --git a/dep/libjpeg/src/jdcolor.c b/dep/libjpeg/src/jdcolor.c deleted file mode 100644 index 6b40fb534..000000000 --- a/dep/libjpeg/src/jdcolor.c +++ /dev/null @@ -1,769 +0,0 @@ -/* - * jdcolor.c - * - * Copyright (C) 1991-1997, Thomas G. Lane. - * Modified 2011-2023 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains output colorspace conversion routines. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -#if RANGE_BITS < 2 - /* Deliberate syntax err */ - Sorry, this code requires 2 or more range extension bits. -#endif - - -/* Private subobject */ - -typedef struct { - struct jpeg_color_deconverter pub; /* public fields */ - - /* Private state for YCbCr->RGB and BG_YCC->RGB conversion */ - int * Cr_r_tab; /* => table for Cr to R conversion */ - int * Cb_b_tab; /* => table for Cb to B conversion */ - INT32 * Cr_g_tab; /* => table for Cr to G conversion */ - INT32 * Cb_g_tab; /* => table for Cb to G conversion */ - - /* Private state for RGB->Y conversion */ - INT32 * R_y_tab; /* => table for R to Y conversion */ - INT32 * G_y_tab; /* => table for G to Y conversion */ - INT32 * B_y_tab; /* => table for B to Y conversion */ -} my_color_deconverter; - -typedef my_color_deconverter * my_cconvert_ptr; - - -/*************** YCbCr -> RGB conversion: most common case **************/ -/*************** BG_YCC -> RGB conversion: less common case **************/ -/*************** RGB -> Y conversion: less common case **************/ - -/* - * YCbCr is defined per Recommendation ITU-R BT.601-7 (03/2011), - * previously known as Recommendation CCIR 601-1, except that Cb and Cr - * are normalized to the range 0..MAXJSAMPLE rather than -0.5 .. 0.5. - * sRGB (standard RGB color space) is defined per IEC 61966-2-1:1999. - * sYCC (standard luma-chroma-chroma color space with extended gamut) - * is defined per IEC 61966-2-1:1999 Amendment A1:2003 Annex F. - * bg-sRGB and bg-sYCC (big gamut standard color spaces) - * are defined per IEC 61966-2-1:1999 Amendment A1:2003 Annex G. - * Note that the derived conversion coefficients given in some of these - * documents are imprecise. The general conversion equations are - * - * R = Y + K * (1 - Kr) * Cr - * G = Y - K * (Kb * (1 - Kb) * Cb + Kr * (1 - Kr) * Cr) / (1 - Kr - Kb) - * B = Y + K * (1 - Kb) * Cb - * - * Y = Kr * R + (1 - Kr - Kb) * G + Kb * B - * - * With Kr = 0.299 and Kb = 0.114 (derived according to SMPTE RP 177-1993 - * from the 1953 FCC NTSC primaries and CIE Illuminant C), K = 2 for sYCC, - * the conversion equations to be implemented are therefore - * - * R = Y + 1.402 * Cr - * G = Y - 0.344136286 * Cb - 0.714136286 * Cr - * B = Y + 1.772 * Cb - * - * Y = 0.299 * R + 0.587 * G + 0.114 * B - * - * where Cb and Cr represent the incoming values less CENTERJSAMPLE. - * For bg-sYCC, with K = 4, the equations are - * - * R = Y + 2.804 * Cr - * G = Y - 0.688272572 * Cb - 1.428272572 * Cr - * B = Y + 3.544 * Cb - * - * To avoid floating-point arithmetic, we represent the fractional constants - * as integers scaled up by 2^16 (about 4 digits precision); we have to divide - * the products by 2^16, with appropriate rounding, to get the correct answer. - * Notice that Y, being an integral input, does not contribute any fraction - * so it need not participate in the rounding. - * - * For even more speed, we avoid doing any multiplications in the inner loop - * by precalculating the constants times Cb and Cr for all possible values. - * For 8-bit JSAMPLEs this is very reasonable (only 256 entries per table); - * for 9-bit to 12-bit samples it is still acceptable. It's not very - * reasonable for 16-bit samples, but if you want lossless storage - * you shouldn't be changing colorspace anyway. - * The Cr=>R and Cb=>B values can be rounded to integers in advance; - * the values for the G calculation are left scaled up, - * since we must add them together before rounding. - */ - -#define SCALEBITS 16 /* speediest right-shift on some machines */ -#define ONE_HALF ((INT32) 1 << (SCALEBITS-1)) -#define FIX(x) ((INT32) ((x) * (1L<RGB and BG_YCC->RGB colorspace conversion. - */ - -LOCAL(void) -build_ycc_rgb_table (j_decompress_ptr cinfo) -/* Normal case, sYCC */ -{ - my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert; - int i; - INT32 x; - SHIFT_TEMPS - - cconvert->Cr_r_tab = (int *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(int)); - cconvert->Cb_b_tab = (int *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(int)); - cconvert->Cr_g_tab = (INT32 *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(INT32)); - cconvert->Cb_g_tab = (INT32 *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(INT32)); - - for (i = 0, x = -CENTERJSAMPLE; i <= MAXJSAMPLE; i++, x++) { - /* i is the actual input pixel value, in the range 0..MAXJSAMPLE */ - /* The Cb or Cr value we are thinking of is x = i - CENTERJSAMPLE */ - /* Cr=>R value is nearest int to 1.402 * x */ - cconvert->Cr_r_tab[i] = (int) DESCALE(FIX(1.402) * x, SCALEBITS); - /* Cb=>B value is nearest int to 1.772 * x */ - cconvert->Cb_b_tab[i] = (int) DESCALE(FIX(1.772) * x, SCALEBITS); - /* Cr=>G value is scaled-up -0.714136286 * x */ - cconvert->Cr_g_tab[i] = (- FIX(0.714136286)) * x; - /* Cb=>G value is scaled-up -0.344136286 * x */ - /* We also add in ONE_HALF so that need not do it in inner loop */ - cconvert->Cb_g_tab[i] = (- FIX(0.344136286)) * x + ONE_HALF; - } -} - - -LOCAL(void) -build_bg_ycc_rgb_table (j_decompress_ptr cinfo) -/* Wide gamut case, bg-sYCC */ -{ - my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert; - int i; - INT32 x; - SHIFT_TEMPS - - cconvert->Cr_r_tab = (int *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(int)); - cconvert->Cb_b_tab = (int *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(int)); - cconvert->Cr_g_tab = (INT32 *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(INT32)); - cconvert->Cb_g_tab = (INT32 *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(INT32)); - - for (i = 0, x = -CENTERJSAMPLE; i <= MAXJSAMPLE; i++, x++) { - /* i is the actual input pixel value, in the range 0..MAXJSAMPLE */ - /* The Cb or Cr value we are thinking of is x = i - CENTERJSAMPLE */ - /* Cr=>R value is nearest int to 2.804 * x */ - cconvert->Cr_r_tab[i] = (int) DESCALE(FIX(2.804) * x, SCALEBITS); - /* Cb=>B value is nearest int to 3.544 * x */ - cconvert->Cb_b_tab[i] = (int) DESCALE(FIX(3.544) * x, SCALEBITS); - /* Cr=>G value is scaled-up -1.428272572 * x */ - cconvert->Cr_g_tab[i] = (- FIX(1.428272572)) * x; - /* Cb=>G value is scaled-up -0.688272572 * x */ - /* We also add in ONE_HALF so that need not do it in inner loop */ - cconvert->Cb_g_tab[i] = (- FIX(0.688272572)) * x + ONE_HALF; - } -} - - -/* - * Convert some rows of samples to the output colorspace. - * - * Note that we change from noninterleaved, one-plane-per-component format - * to interleaved-pixel format. The output buffer is therefore three times - * as wide as the input buffer. - * - * A starting row offset is provided only for the input buffer. The caller - * can easily adjust the passed output_buf value to accommodate any row - * offset required on that side. - */ - -METHODDEF(void) -ycc_rgb_convert (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION input_row, - JSAMPARRAY output_buf, int num_rows) -{ - my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert; - register int y, cb, cr; - register JSAMPROW outptr; - register JSAMPROW inptr0, inptr1, inptr2; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->output_width; - /* copy these pointers into registers if possible */ - register JSAMPLE * range_limit = cinfo->sample_range_limit; - register int * Crrtab = cconvert->Cr_r_tab; - register int * Cbbtab = cconvert->Cb_b_tab; - register INT32 * Crgtab = cconvert->Cr_g_tab; - register INT32 * Cbgtab = cconvert->Cb_g_tab; - SHIFT_TEMPS - - while (--num_rows >= 0) { - inptr0 = input_buf[0][input_row]; - inptr1 = input_buf[1][input_row]; - inptr2 = input_buf[2][input_row]; - input_row++; - outptr = *output_buf++; - for (col = 0; col < num_cols; col++) { - y = GETJSAMPLE(inptr0[col]); - cb = GETJSAMPLE(inptr1[col]); - cr = GETJSAMPLE(inptr2[col]); - /* Range-limiting is essential due to noise introduced by DCT losses, - * for extended gamut (sYCC) and wide gamut (bg-sYCC) encodings. - */ - outptr[RGB_RED] = range_limit[y + Crrtab[cr]]; - outptr[RGB_GREEN] = range_limit[y + - ((int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], - SCALEBITS))]; - outptr[RGB_BLUE] = range_limit[y + Cbbtab[cb]]; - outptr += RGB_PIXELSIZE; - } - } -} - - -/**************** Cases other than YCC -> RGB ****************/ - - -/* - * Initialize for RGB->grayscale colorspace conversion. - */ - -LOCAL(void) -build_rgb_y_table (j_decompress_ptr cinfo) -{ - my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert; - INT32 i; - - cconvert->R_y_tab = (INT32 *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(INT32)); - cconvert->G_y_tab = (INT32 *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(INT32)); - cconvert->B_y_tab = (INT32 *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(INT32)); - - for (i = 0; i <= MAXJSAMPLE; i++) { - cconvert->R_y_tab[i] = FIX(0.299) * i; - cconvert->G_y_tab[i] = FIX(0.587) * i; - cconvert->B_y_tab[i] = FIX(0.114) * i + ONE_HALF; - } -} - - -/* - * Convert RGB to grayscale. - */ - -METHODDEF(void) -rgb_gray_convert (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION input_row, - JSAMPARRAY output_buf, int num_rows) -{ - my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert; - register INT32 y; - register INT32 * Rytab = cconvert->R_y_tab; - register INT32 * Gytab = cconvert->G_y_tab; - register INT32 * Bytab = cconvert->B_y_tab; - register JSAMPROW outptr; - register JSAMPROW inptr0, inptr1, inptr2; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->output_width; - - while (--num_rows >= 0) { - inptr0 = input_buf[0][input_row]; - inptr1 = input_buf[1][input_row]; - inptr2 = input_buf[2][input_row]; - input_row++; - outptr = *output_buf++; - for (col = 0; col < num_cols; col++) { - y = Rytab[GETJSAMPLE(inptr0[col])]; - y += Gytab[GETJSAMPLE(inptr1[col])]; - y += Bytab[GETJSAMPLE(inptr2[col])]; - outptr[col] = (JSAMPLE) (y >> SCALEBITS); - } - } -} - - -/* - * Convert some rows of samples to the output colorspace. - * [R-G,G,B-G] to [R,G,B] conversion with modulo calculation - * (inverse color transform). - * This can be seen as an adaption of the general YCbCr->RGB - * conversion equation with Kr = Kb = 0, while replacing the - * normalization by modulo calculation. - */ - -METHODDEF(void) -rgb1_rgb_convert (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION input_row, - JSAMPARRAY output_buf, int num_rows) -{ - register int r, g, b; - register JSAMPROW outptr; - register JSAMPROW inptr0, inptr1, inptr2; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->output_width; - - while (--num_rows >= 0) { - inptr0 = input_buf[0][input_row]; - inptr1 = input_buf[1][input_row]; - inptr2 = input_buf[2][input_row]; - input_row++; - outptr = *output_buf++; - for (col = 0; col < num_cols; col++) { - r = GETJSAMPLE(inptr0[col]); - g = GETJSAMPLE(inptr1[col]); - b = GETJSAMPLE(inptr2[col]); - /* Assume that MAXJSAMPLE+1 is a power of 2, so that the MOD - * (modulo) operator is equivalent to the bitmask operator AND. - */ - outptr[RGB_RED] = (JSAMPLE) ((r + g - CENTERJSAMPLE) & MAXJSAMPLE); - outptr[RGB_GREEN] = (JSAMPLE) g; - outptr[RGB_BLUE] = (JSAMPLE) ((b + g - CENTERJSAMPLE) & MAXJSAMPLE); - outptr += RGB_PIXELSIZE; - } - } -} - - -/* - * [R-G,G,B-G] to grayscale conversion with modulo calculation - * (inverse color transform). - */ - -METHODDEF(void) -rgb1_gray_convert (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION input_row, - JSAMPARRAY output_buf, int num_rows) -{ - my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert; - register int r, g, b; - register INT32 y; - register INT32 * Rytab = cconvert->R_y_tab; - register INT32 * Gytab = cconvert->G_y_tab; - register INT32 * Bytab = cconvert->B_y_tab; - register JSAMPROW outptr; - register JSAMPROW inptr0, inptr1, inptr2; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->output_width; - - while (--num_rows >= 0) { - inptr0 = input_buf[0][input_row]; - inptr1 = input_buf[1][input_row]; - inptr2 = input_buf[2][input_row]; - input_row++; - outptr = *output_buf++; - for (col = 0; col < num_cols; col++) { - r = GETJSAMPLE(inptr0[col]); - g = GETJSAMPLE(inptr1[col]); - b = GETJSAMPLE(inptr2[col]); - /* Assume that MAXJSAMPLE+1 is a power of 2, so that the MOD - * (modulo) operator is equivalent to the bitmask operator AND. - */ - y = Rytab[(r + g - CENTERJSAMPLE) & MAXJSAMPLE]; - y += Gytab[g]; - y += Bytab[(b + g - CENTERJSAMPLE) & MAXJSAMPLE]; - outptr[col] = (JSAMPLE) (y >> SCALEBITS); - } - } -} - - -/* - * Convert some rows of samples to the output colorspace. - * No colorspace change, but conversion from separate-planes - * to interleaved representation. - */ - -METHODDEF(void) -rgb_convert (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION input_row, - JSAMPARRAY output_buf, int num_rows) -{ - register JSAMPROW outptr; - register JSAMPROW inptr0, inptr1, inptr2; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->output_width; - - while (--num_rows >= 0) { - inptr0 = input_buf[0][input_row]; - inptr1 = input_buf[1][input_row]; - inptr2 = input_buf[2][input_row]; - input_row++; - outptr = *output_buf++; - for (col = 0; col < num_cols; col++) { - /* We can dispense with GETJSAMPLE() here */ - outptr[RGB_RED] = inptr0[col]; - outptr[RGB_GREEN] = inptr1[col]; - outptr[RGB_BLUE] = inptr2[col]; - outptr += RGB_PIXELSIZE; - } - } -} - - -/* - * Color conversion for no colorspace change: just copy the data, - * converting from separate-planes to interleaved representation. - * Note: Omit uninteresting components in output buffer. - */ - -METHODDEF(void) -null_convert (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION input_row, - JSAMPARRAY output_buf, int num_rows) -{ - register JSAMPROW outptr; - register JSAMPROW inptr; - register JDIMENSION count; - register int out_comps = cinfo->out_color_components; - JDIMENSION num_cols = cinfo->output_width; - JSAMPROW startptr; - int ci; - jpeg_component_info *compptr; - - while (--num_rows >= 0) { - /* It seems fastest to make a separate pass for each component. */ - startptr = *output_buf++; - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - if (! compptr->component_needed) - continue; /* skip uninteresting component */ - inptr = input_buf[ci][input_row]; - outptr = startptr++; - for (count = num_cols; count > 0; count--) { - *outptr = *inptr++; /* don't need GETJSAMPLE() here */ - outptr += out_comps; - } - } - input_row++; - } -} - - -/* - * Color conversion for grayscale: just copy the data. - * This also works for YCC -> grayscale conversion, in which - * we just copy the Y (luminance) component and ignore chrominance. - */ - -METHODDEF(void) -grayscale_convert (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION input_row, - JSAMPARRAY output_buf, int num_rows) -{ - jcopy_sample_rows(input_buf[0] + input_row, output_buf, - num_rows, cinfo->output_width); -} - - -/* - * Convert grayscale to RGB: just duplicate the graylevel three times. - * This is provided to support applications that don't want to cope - * with grayscale as a separate case. - */ - -METHODDEF(void) -gray_rgb_convert (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION input_row, - JSAMPARRAY output_buf, int num_rows) -{ - register JSAMPROW outptr; - register JSAMPROW inptr; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->output_width; - - while (--num_rows >= 0) { - inptr = input_buf[0][input_row++]; - outptr = *output_buf++; - for (col = 0; col < num_cols; col++) { - /* We can dispense with GETJSAMPLE() here */ - outptr[RGB_RED] = outptr[RGB_GREEN] = outptr[RGB_BLUE] = inptr[col]; - outptr += RGB_PIXELSIZE; - } - } -} - - -/* - * Convert some rows of samples to the output colorspace. - * This version handles Adobe-style YCCK->CMYK conversion, - * where we convert YCbCr to R=1-C, G=1-M, and B=1-Y using the - * same conversion as above, while passing K (black) unchanged. - * We assume build_ycc_rgb_table has been called. - */ - -METHODDEF(void) -ycck_cmyk_convert (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION input_row, - JSAMPARRAY output_buf, int num_rows) -{ - my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert; - register int y, cb, cr; - register JSAMPROW outptr; - register JSAMPROW inptr0, inptr1, inptr2, inptr3; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->output_width; - /* copy these pointers into registers if possible */ - register JSAMPLE * range_limit = cinfo->sample_range_limit; - register int * Crrtab = cconvert->Cr_r_tab; - register int * Cbbtab = cconvert->Cb_b_tab; - register INT32 * Crgtab = cconvert->Cr_g_tab; - register INT32 * Cbgtab = cconvert->Cb_g_tab; - SHIFT_TEMPS - - while (--num_rows >= 0) { - inptr0 = input_buf[0][input_row]; - inptr1 = input_buf[1][input_row]; - inptr2 = input_buf[2][input_row]; - inptr3 = input_buf[3][input_row]; - input_row++; - outptr = *output_buf++; - for (col = 0; col < num_cols; col++) { - y = GETJSAMPLE(inptr0[col]); - cb = GETJSAMPLE(inptr1[col]); - cr = GETJSAMPLE(inptr2[col]); - /* Range-limiting is essential due to noise introduced by DCT losses, - * and for extended gamut encodings (sYCC). - */ - outptr[0] = range_limit[MAXJSAMPLE - (y + Crrtab[cr])]; /* red */ - outptr[1] = range_limit[MAXJSAMPLE - (y + /* green */ - ((int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], - SCALEBITS)))]; - outptr[2] = range_limit[MAXJSAMPLE - (y + Cbbtab[cb])]; /* blue */ - /* K passes through unchanged */ - outptr[3] = inptr3[col]; /* don't need GETJSAMPLE here */ - outptr += 4; - } - } -} - - -/* - * Convert CMYK to YK part of YCCK for colorless output. - * We assume build_rgb_y_table has been called. - */ - -METHODDEF(void) -cmyk_yk_convert (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION input_row, - JSAMPARRAY output_buf, int num_rows) -{ - my_cconvert_ptr cconvert = (my_cconvert_ptr) cinfo->cconvert; - register INT32 y; - register INT32 * Rytab = cconvert->R_y_tab; - register INT32 * Gytab = cconvert->G_y_tab; - register INT32 * Bytab = cconvert->B_y_tab; - register JSAMPROW outptr; - register JSAMPROW inptr0, inptr1, inptr2, inptr3; - register JDIMENSION col; - JDIMENSION num_cols = cinfo->output_width; - - while (--num_rows >= 0) { - inptr0 = input_buf[0][input_row]; - inptr1 = input_buf[1][input_row]; - inptr2 = input_buf[2][input_row]; - inptr3 = input_buf[3][input_row]; - input_row++; - outptr = *output_buf++; - for (col = 0; col < num_cols; col++) { - y = Rytab[MAXJSAMPLE - GETJSAMPLE(inptr0[col])]; - y += Gytab[MAXJSAMPLE - GETJSAMPLE(inptr1[col])]; - y += Bytab[MAXJSAMPLE - GETJSAMPLE(inptr2[col])]; - outptr[0] = (JSAMPLE) (y >> SCALEBITS); - /* K passes through unchanged */ - outptr[1] = inptr3[col]; /* don't need GETJSAMPLE here */ - outptr += 2; - } - } -} - - -/* - * Empty method for start_pass. - */ - -METHODDEF(void) -start_pass_dcolor (j_decompress_ptr cinfo) -{ - /* no work needed */ -} - - -/* - * Module initialization routine for output colorspace conversion. - */ - -GLOBAL(void) -jinit_color_deconverter (j_decompress_ptr cinfo) -{ - my_cconvert_ptr cconvert; - int ci, i; - - cconvert = (my_cconvert_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_color_deconverter)); - cinfo->cconvert = &cconvert->pub; - cconvert->pub.start_pass = start_pass_dcolor; - - /* Make sure num_components agrees with jpeg_color_space */ - switch (cinfo->jpeg_color_space) { - case JCS_GRAYSCALE: - if (cinfo->num_components != 1) - ERREXIT(cinfo, JERR_BAD_J_COLORSPACE); - break; - - case JCS_RGB: - case JCS_YCbCr: - case JCS_BG_RGB: - case JCS_BG_YCC: - if (cinfo->num_components != 3) - ERREXIT(cinfo, JERR_BAD_J_COLORSPACE); - break; - - case JCS_CMYK: - case JCS_YCCK: - if (cinfo->num_components != 4) - ERREXIT(cinfo, JERR_BAD_J_COLORSPACE); - break; - - default: /* JCS_UNKNOWN can be anything */ - if (cinfo->num_components < 1) - ERREXIT(cinfo, JERR_BAD_J_COLORSPACE); - } - - /* Support color transform only for RGB colorspaces */ - if (cinfo->color_transform && - cinfo->jpeg_color_space != JCS_RGB && - cinfo->jpeg_color_space != JCS_BG_RGB) - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - - /* Set out_color_components and conversion method based on requested space. - * Also adjust the component_needed flags for any unused components, - * so that earlier pipeline stages can avoid useless computation. - */ - - switch (cinfo->out_color_space) { - case JCS_GRAYSCALE: - cinfo->out_color_components = 1; - switch (cinfo->jpeg_color_space) { - case JCS_GRAYSCALE: - case JCS_YCbCr: - case JCS_BG_YCC: - cconvert->pub.color_convert = grayscale_convert; - /* For color->grayscale conversion, only the Y (0) component is needed */ - for (ci = 1; ci < cinfo->num_components; ci++) - cinfo->comp_info[ci].component_needed = FALSE; - break; - case JCS_RGB: - switch (cinfo->color_transform) { - case JCT_NONE: - cconvert->pub.color_convert = rgb_gray_convert; - break; - case JCT_SUBTRACT_GREEN: - cconvert->pub.color_convert = rgb1_gray_convert; - break; - default: - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - } - build_rgb_y_table(cinfo); - break; - default: - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - } - break; - - case JCS_RGB: - cinfo->out_color_components = RGB_PIXELSIZE; - switch (cinfo->jpeg_color_space) { - case JCS_GRAYSCALE: - cconvert->pub.color_convert = gray_rgb_convert; - break; - case JCS_YCbCr: - cconvert->pub.color_convert = ycc_rgb_convert; - build_ycc_rgb_table(cinfo); - break; - case JCS_BG_YCC: - cconvert->pub.color_convert = ycc_rgb_convert; - build_bg_ycc_rgb_table(cinfo); - break; - case JCS_RGB: - switch (cinfo->color_transform) { - case JCT_NONE: - cconvert->pub.color_convert = rgb_convert; - break; - case JCT_SUBTRACT_GREEN: - cconvert->pub.color_convert = rgb1_rgb_convert; - break; - default: - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - } - break; - default: - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - } - break; - - case JCS_BG_RGB: - if (cinfo->jpeg_color_space != JCS_BG_RGB) - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - cinfo->out_color_components = RGB_PIXELSIZE; - switch (cinfo->color_transform) { - case JCT_NONE: - cconvert->pub.color_convert = rgb_convert; - break; - case JCT_SUBTRACT_GREEN: - cconvert->pub.color_convert = rgb1_rgb_convert; - break; - default: - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - } - break; - - case JCS_CMYK: - if (cinfo->jpeg_color_space != JCS_YCCK) - goto def_label; - cinfo->out_color_components = 4; - cconvert->pub.color_convert = ycck_cmyk_convert; - build_ycc_rgb_table(cinfo); - break; - - case JCS_YCCK: - if (cinfo->jpeg_color_space != JCS_CMYK || - /* Support only YK part of YCCK for colorless output */ - ! cinfo->comp_info[0].component_needed || - cinfo->comp_info[1].component_needed || - cinfo->comp_info[2].component_needed || - ! cinfo->comp_info[3].component_needed) - goto def_label; - cinfo->out_color_components = 2; - /* Need all components on input side */ - cinfo->comp_info[1].component_needed = TRUE; - cinfo->comp_info[2].component_needed = TRUE; - cconvert->pub.color_convert = cmyk_yk_convert; - build_rgb_y_table(cinfo); - break; - - default: def_label: /* permit null conversion to same output space */ - if (cinfo->out_color_space != cinfo->jpeg_color_space) - /* unsupported non-null conversion */ - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - i = 0; - for (ci = 0; ci < cinfo->num_components; ci++) - if (cinfo->comp_info[ci].component_needed) - i++; /* count output color components */ - cinfo->out_color_components = i; - cconvert->pub.color_convert = null_convert; - } - - if (cinfo->quantize_colors) - cinfo->output_components = 1; /* single colormapped output component */ - else - cinfo->output_components = cinfo->out_color_components; -} diff --git a/dep/libjpeg/src/jdct.h b/dep/libjpeg/src/jdct.h deleted file mode 100644 index 0f251590c..000000000 --- a/dep/libjpeg/src/jdct.h +++ /dev/null @@ -1,409 +0,0 @@ -/* - * jdct.h - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2002-2023 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This include file contains common declarations for the forward and - * inverse DCT modules. These declarations are private to the DCT managers - * (jcdctmgr.c, jddctmgr.c) and the individual DCT algorithms. - * The individual DCT algorithms are kept in separate files to ease - * machine-dependent tuning (e.g., assembly coding). - */ - - -/* - * A forward DCT routine is given a pointer to an input sample array and - * a pointer to a work area of type DCTELEM[]; the DCT is to be performed - * in-place in that buffer. Type DCTELEM is int for 8-bit samples, INT32 - * for 12-bit samples. (NOTE: Floating-point DCT implementations use an - * array of type FAST_FLOAT, instead.) - * The input data is to be fetched from the sample array starting at a - * specified column. (Any row offset needed will be applied to the array - * pointer before it is passed to the FDCT code.) - * Note that the number of samples fetched by the FDCT routine is - * DCT_h_scaled_size * DCT_v_scaled_size. - * The DCT outputs are returned scaled up by a factor of 8; they therefore - * have a range of +-8K for 8-bit data, +-128K for 12-bit data. This - * convention improves accuracy in integer implementations and saves some - * work in floating-point ones. - * Quantization of the output coefficients is done by jcdctmgr.c. - */ - -#if BITS_IN_JSAMPLE == 8 -typedef int DCTELEM; /* 16 or 32 bits is fine */ -#else -typedef INT32 DCTELEM; /* must have 32 bits */ -#endif - -typedef JMETHOD(void, forward_DCT_method_ptr, (DCTELEM * data, - JSAMPARRAY sample_data, - JDIMENSION start_col)); -typedef JMETHOD(void, float_DCT_method_ptr, (FAST_FLOAT * data, - JSAMPARRAY sample_data, - JDIMENSION start_col)); - - -/* - * An inverse DCT routine is given a pointer to the input JBLOCK and a pointer - * to an output sample array. The routine must dequantize the input data as - * well as perform the IDCT; for dequantization, it uses the multiplier table - * pointed to by compptr->dct_table. The output data is to be placed into the - * sample array starting at a specified column. (Any row offset needed will - * be applied to the array pointer before it is passed to the IDCT code.) - * Note that the number of samples emitted by the IDCT routine is - * DCT_h_scaled_size * DCT_v_scaled_size. - */ - -/* typedef inverse_DCT_method_ptr is declared in jpegint.h */ - -/* - * Each IDCT routine has its own ideas about the best dct_table element type. - */ - -typedef MULTIPLIER ISLOW_MULT_TYPE; /* short or int, whichever is faster */ -#if BITS_IN_JSAMPLE == 8 -typedef MULTIPLIER IFAST_MULT_TYPE; /* 16 bits is OK, use short if faster */ -#define IFAST_SCALE_BITS 2 /* fractional bits in scale factors */ -#else -typedef INT32 IFAST_MULT_TYPE; /* need 32 bits for scaled quantizers */ -#define IFAST_SCALE_BITS 13 /* fractional bits in scale factors */ -#endif -typedef FAST_FLOAT FLOAT_MULT_TYPE; /* preferred floating type */ - - -/* - * Each IDCT routine is responsible for range-limiting its results and - * converting them to unsigned form (0..MAXJSAMPLE). The raw outputs could - * be quite far out of range if the input data is corrupt, so a bulletproof - * range-limiting step is required. We use a mask-and-table-lookup method - * to do the combined operations quickly, assuming that RANGE_CENTER - * (defined in jpegint.h) is a power of 2. See the comments with - * prepare_range_limit_table (in jdmaster.c) for more info. - */ - -#define RANGE_MASK (RANGE_CENTER * 2 - 1) -#define RANGE_SUBSET (RANGE_CENTER - CENTERJSAMPLE) - -#define IDCT_range_limit(cinfo) ((cinfo)->sample_range_limit - RANGE_SUBSET) - - -/* Short forms of external names for systems with brain-damaged linkers. */ - -#ifdef NEED_SHORT_EXTERNAL_NAMES -#define jpeg_fdct_islow jFDislow -#define jpeg_fdct_ifast jFDifast -#define jpeg_fdct_float jFDfloat -#define jpeg_fdct_7x7 jFD7x7 -#define jpeg_fdct_6x6 jFD6x6 -#define jpeg_fdct_5x5 jFD5x5 -#define jpeg_fdct_4x4 jFD4x4 -#define jpeg_fdct_3x3 jFD3x3 -#define jpeg_fdct_2x2 jFD2x2 -#define jpeg_fdct_1x1 jFD1x1 -#define jpeg_fdct_9x9 jFD9x9 -#define jpeg_fdct_10x10 jFD10x10 -#define jpeg_fdct_11x11 jFD11x11 -#define jpeg_fdct_12x12 jFD12x12 -#define jpeg_fdct_13x13 jFD13x13 -#define jpeg_fdct_14x14 jFD14x14 -#define jpeg_fdct_15x15 jFD15x15 -#define jpeg_fdct_16x16 jFD16x16 -#define jpeg_fdct_16x8 jFD16x8 -#define jpeg_fdct_14x7 jFD14x7 -#define jpeg_fdct_12x6 jFD12x6 -#define jpeg_fdct_10x5 jFD10x5 -#define jpeg_fdct_8x4 jFD8x4 -#define jpeg_fdct_6x3 jFD6x3 -#define jpeg_fdct_4x2 jFD4x2 -#define jpeg_fdct_2x1 jFD2x1 -#define jpeg_fdct_8x16 jFD8x16 -#define jpeg_fdct_7x14 jFD7x14 -#define jpeg_fdct_6x12 jFD6x12 -#define jpeg_fdct_5x10 jFD5x10 -#define jpeg_fdct_4x8 jFD4x8 -#define jpeg_fdct_3x6 jFD3x6 -#define jpeg_fdct_2x4 jFD2x4 -#define jpeg_fdct_1x2 jFD1x2 -#define jpeg_idct_islow jRDislow -#define jpeg_idct_ifast jRDifast -#define jpeg_idct_float jRDfloat -#define jpeg_idct_7x7 jRD7x7 -#define jpeg_idct_6x6 jRD6x6 -#define jpeg_idct_5x5 jRD5x5 -#define jpeg_idct_4x4 jRD4x4 -#define jpeg_idct_3x3 jRD3x3 -#define jpeg_idct_2x2 jRD2x2 -#define jpeg_idct_1x1 jRD1x1 -#define jpeg_idct_9x9 jRD9x9 -#define jpeg_idct_10x10 jRD10x10 -#define jpeg_idct_11x11 jRD11x11 -#define jpeg_idct_12x12 jRD12x12 -#define jpeg_idct_13x13 jRD13x13 -#define jpeg_idct_14x14 jRD14x14 -#define jpeg_idct_15x15 jRD15x15 -#define jpeg_idct_16x16 jRD16x16 -#define jpeg_idct_16x8 jRD16x8 -#define jpeg_idct_14x7 jRD14x7 -#define jpeg_idct_12x6 jRD12x6 -#define jpeg_idct_10x5 jRD10x5 -#define jpeg_idct_8x4 jRD8x4 -#define jpeg_idct_6x3 jRD6x3 -#define jpeg_idct_4x2 jRD4x2 -#define jpeg_idct_2x1 jRD2x1 -#define jpeg_idct_8x16 jRD8x16 -#define jpeg_idct_7x14 jRD7x14 -#define jpeg_idct_6x12 jRD6x12 -#define jpeg_idct_5x10 jRD5x10 -#define jpeg_idct_4x8 jRD4x8 -#define jpeg_idct_3x6 jRD3x6 -#define jpeg_idct_2x4 jRD2x4 -#define jpeg_idct_1x2 jRD1x2 -#endif /* NEED_SHORT_EXTERNAL_NAMES */ - -/* Extern declarations for the forward and inverse DCT routines. */ - -EXTERN(void) jpeg_fdct_islow - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_ifast - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_float - JPP((FAST_FLOAT * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_7x7 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_6x6 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_5x5 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_4x4 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_3x3 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_2x2 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_1x1 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_9x9 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_10x10 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_11x11 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_12x12 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_13x13 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_14x14 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_15x15 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_16x16 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_16x8 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_14x7 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_12x6 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_10x5 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_8x4 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_6x3 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_4x2 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_2x1 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_8x16 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_7x14 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_6x12 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_5x10 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_4x8 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_3x6 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_2x4 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); -EXTERN(void) jpeg_fdct_1x2 - JPP((DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col)); - -EXTERN(void) jpeg_idct_islow - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_ifast - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_float - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_7x7 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_6x6 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_5x5 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_4x4 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_3x3 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_2x2 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_1x1 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_9x9 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_10x10 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_11x11 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_12x12 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_13x13 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_14x14 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_15x15 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_16x16 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_16x8 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_14x7 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_12x6 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_10x5 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_8x4 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_6x3 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_4x2 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_2x1 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_8x16 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_7x14 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_6x12 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_5x10 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_4x8 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_3x6 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_2x4 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); -EXTERN(void) jpeg_idct_1x2 - JPP((j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, JSAMPARRAY output_buf, JDIMENSION output_col)); - - -/* - * Macros for handling fixed-point arithmetic; these are used by many - * but not all of the DCT/IDCT modules. - * - * All values are expected to be of type INT32. - * Fractional constants are scaled left by CONST_BITS bits. - * CONST_BITS is defined within each module using these macros, - * and may differ from one module to the next. - */ - -#define ONE ((INT32) 1) -#define CONST_SCALE (ONE << CONST_BITS) - -/* Convert a positive real constant to an integer scaled by CONST_SCALE. - * Caution: some C compilers fail to reduce "FIX(constant)" at compile time, - * thus causing a lot of useless floating-point operations at run time. - */ - -#define FIX(x) ((INT32) ((x) * CONST_SCALE + 0.5)) - -/* Multiply an INT32 variable by an INT32 constant to yield an INT32 result. - * This macro is used only when the two inputs will actually be no more than - * 16 bits wide, so that a 16x16->32 bit multiply can be used instead of a - * full 32x32 multiply. This provides a useful speedup on many machines. - * Unfortunately there is no way to specify a 16x16->32 multiply portably - * in C, but some C compilers will do the right thing if you provide the - * correct combination of casts. - */ - -#ifdef SHORTxSHORT_32 /* may work if 'int' is 32 bits */ -#define MULTIPLY16C16(var,const) (((INT16) (var)) * ((INT16) (const))) -#endif -#ifdef SHORTxLCONST_32 /* known to work with Microsoft C 6.0 */ -#define MULTIPLY16C16(var,const) (((INT16) (var)) * ((INT32) (const))) -#endif - -#ifndef MULTIPLY16C16 /* default definition */ -#define MULTIPLY16C16(var,const) ((var) * (const)) -#endif - -/* Same except both inputs are variables. */ - -#ifdef SHORTxSHORT_32 /* may work if 'int' is 32 bits */ -#define MULTIPLY16V16(var1,var2) (((INT16) (var1)) * ((INT16) (var2))) -#endif - -#ifndef MULTIPLY16V16 /* default definition */ -#define MULTIPLY16V16(var1,var2) ((var1) * (var2)) -#endif - -/* Like RIGHT_SHIFT, but applies to a DCTELEM. - * We assume that int right shift is unsigned if INT32 right shift is. - */ - -#ifdef RIGHT_SHIFT_IS_UNSIGNED -#define ISHIFT_TEMPS DCTELEM ishift_temp; -#if BITS_IN_JSAMPLE == 8 -#define DCTELEMBITS 16 /* DCTELEM may be 16 or 32 bits */ -#else -#define DCTELEMBITS 32 /* DCTELEM must be 32 bits */ -#endif -#define IRIGHT_SHIFT(x,shft) \ - ((ishift_temp = (x)) < 0 ? \ - (ishift_temp >> (shft)) | ((~((DCTELEM) 0)) << (DCTELEMBITS-(shft))) : \ - (ishift_temp >> (shft))) -#else -#define ISHIFT_TEMPS -#define IRIGHT_SHIFT(x,shft) ((x) >> (shft)) -#endif diff --git a/dep/libjpeg/src/jddctmgr.c b/dep/libjpeg/src/jddctmgr.c deleted file mode 100644 index 9ecfbb510..000000000 --- a/dep/libjpeg/src/jddctmgr.c +++ /dev/null @@ -1,384 +0,0 @@ -/* - * jddctmgr.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2002-2013 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains the inverse-DCT management logic. - * This code selects a particular IDCT implementation to be used, - * and it performs related housekeeping chores. No code in this file - * is executed per IDCT step, only during output pass setup. - * - * Note that the IDCT routines are responsible for performing coefficient - * dequantization as well as the IDCT proper. This module sets up the - * dequantization multiplier table needed by the IDCT routine. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" -#include "jdct.h" /* Private declarations for DCT subsystem */ - - -/* - * The decompressor input side (jdinput.c) saves away the appropriate - * quantization table for each component at the start of the first scan - * involving that component. (This is necessary in order to correctly - * decode files that reuse Q-table slots.) - * When we are ready to make an output pass, the saved Q-table is converted - * to a multiplier table that will actually be used by the IDCT routine. - * The multiplier table contents are IDCT-method-dependent. To support - * application changes in IDCT method between scans, we can remake the - * multiplier tables if necessary. - * In buffered-image mode, the first output pass may occur before any data - * has been seen for some components, and thus before their Q-tables have - * been saved away. To handle this case, multiplier tables are preset - * to zeroes; the result of the IDCT will be a neutral gray level. - */ - - -/* Private subobject for this module */ - -typedef struct { - struct jpeg_inverse_dct pub; /* public fields */ - - /* This array contains the IDCT method code that each multiplier table - * is currently set up for, or -1 if it's not yet set up. - * The actual multiplier tables are pointed to by dct_table in the - * per-component comp_info structures. - */ - int cur_method[MAX_COMPONENTS]; -} my_idct_controller; - -typedef my_idct_controller * my_idct_ptr; - - -/* Allocated multiplier tables: big enough for any supported variant */ - -typedef union { - ISLOW_MULT_TYPE islow_array[DCTSIZE2]; -#ifdef DCT_IFAST_SUPPORTED - IFAST_MULT_TYPE ifast_array[DCTSIZE2]; -#endif -#ifdef DCT_FLOAT_SUPPORTED - FLOAT_MULT_TYPE float_array[DCTSIZE2]; -#endif -} multiplier_table; - - -/* The current scaled-IDCT routines require ISLOW-style multiplier tables, - * so be sure to compile that code if either ISLOW or SCALING is requested. - */ -#ifdef DCT_ISLOW_SUPPORTED -#define PROVIDE_ISLOW_TABLES -#else -#ifdef IDCT_SCALING_SUPPORTED -#define PROVIDE_ISLOW_TABLES -#endif -#endif - - -/* - * Prepare for an output pass. - * Here we select the proper IDCT routine for each component and build - * a matching multiplier table. - */ - -METHODDEF(void) -start_pass (j_decompress_ptr cinfo) -{ - my_idct_ptr idct = (my_idct_ptr) cinfo->idct; - int ci, i; - jpeg_component_info *compptr; - int method = 0; - inverse_DCT_method_ptr method_ptr = NULL; - JQUANT_TBL * qtbl; - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* Select the proper IDCT routine for this component's scaling */ - switch ((compptr->DCT_h_scaled_size << 8) + compptr->DCT_v_scaled_size) { -#ifdef IDCT_SCALING_SUPPORTED - case ((1 << 8) + 1): - method_ptr = jpeg_idct_1x1; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((2 << 8) + 2): - method_ptr = jpeg_idct_2x2; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((3 << 8) + 3): - method_ptr = jpeg_idct_3x3; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((4 << 8) + 4): - method_ptr = jpeg_idct_4x4; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((5 << 8) + 5): - method_ptr = jpeg_idct_5x5; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((6 << 8) + 6): - method_ptr = jpeg_idct_6x6; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((7 << 8) + 7): - method_ptr = jpeg_idct_7x7; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((9 << 8) + 9): - method_ptr = jpeg_idct_9x9; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((10 << 8) + 10): - method_ptr = jpeg_idct_10x10; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((11 << 8) + 11): - method_ptr = jpeg_idct_11x11; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((12 << 8) + 12): - method_ptr = jpeg_idct_12x12; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((13 << 8) + 13): - method_ptr = jpeg_idct_13x13; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((14 << 8) + 14): - method_ptr = jpeg_idct_14x14; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((15 << 8) + 15): - method_ptr = jpeg_idct_15x15; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((16 << 8) + 16): - method_ptr = jpeg_idct_16x16; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((16 << 8) + 8): - method_ptr = jpeg_idct_16x8; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((14 << 8) + 7): - method_ptr = jpeg_idct_14x7; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((12 << 8) + 6): - method_ptr = jpeg_idct_12x6; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((10 << 8) + 5): - method_ptr = jpeg_idct_10x5; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((8 << 8) + 4): - method_ptr = jpeg_idct_8x4; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((6 << 8) + 3): - method_ptr = jpeg_idct_6x3; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((4 << 8) + 2): - method_ptr = jpeg_idct_4x2; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((2 << 8) + 1): - method_ptr = jpeg_idct_2x1; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((8 << 8) + 16): - method_ptr = jpeg_idct_8x16; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((7 << 8) + 14): - method_ptr = jpeg_idct_7x14; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((6 << 8) + 12): - method_ptr = jpeg_idct_6x12; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((5 << 8) + 10): - method_ptr = jpeg_idct_5x10; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((4 << 8) + 8): - method_ptr = jpeg_idct_4x8; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((3 << 8) + 6): - method_ptr = jpeg_idct_3x6; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((2 << 8) + 4): - method_ptr = jpeg_idct_2x4; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; - case ((1 << 8) + 2): - method_ptr = jpeg_idct_1x2; - method = JDCT_ISLOW; /* jidctint uses islow-style table */ - break; -#endif - case ((DCTSIZE << 8) + DCTSIZE): - switch (cinfo->dct_method) { -#ifdef DCT_ISLOW_SUPPORTED - case JDCT_ISLOW: - method_ptr = jpeg_idct_islow; - method = JDCT_ISLOW; - break; -#endif -#ifdef DCT_IFAST_SUPPORTED - case JDCT_IFAST: - method_ptr = jpeg_idct_ifast; - method = JDCT_IFAST; - break; -#endif -#ifdef DCT_FLOAT_SUPPORTED - case JDCT_FLOAT: - method_ptr = jpeg_idct_float; - method = JDCT_FLOAT; - break; -#endif - default: - ERREXIT(cinfo, JERR_NOT_COMPILED); - break; - } - break; - default: - ERREXIT2(cinfo, JERR_BAD_DCTSIZE, - compptr->DCT_h_scaled_size, compptr->DCT_v_scaled_size); - break; - } - idct->pub.inverse_DCT[ci] = method_ptr; - /* Create multiplier table from quant table. - * However, we can skip this if the component is uninteresting - * or if we already built the table. Also, if no quant table - * has yet been saved for the component, we leave the - * multiplier table all-zero; we'll be reading zeroes from the - * coefficient controller's buffer anyway. - */ - if (! compptr->component_needed || idct->cur_method[ci] == method) - continue; - qtbl = compptr->quant_table; - if (qtbl == NULL) /* happens if no data yet for component */ - continue; - idct->cur_method[ci] = method; - switch (method) { -#ifdef PROVIDE_ISLOW_TABLES - case JDCT_ISLOW: - { - /* For LL&M IDCT method, multipliers are equal to raw quantization - * coefficients, but are stored as ints to ensure access efficiency. - */ - ISLOW_MULT_TYPE * ismtbl = (ISLOW_MULT_TYPE *) compptr->dct_table; - for (i = 0; i < DCTSIZE2; i++) { - ismtbl[i] = (ISLOW_MULT_TYPE) qtbl->quantval[i]; - } - } - break; -#endif -#ifdef DCT_IFAST_SUPPORTED - case JDCT_IFAST: - { - /* For AA&N IDCT method, multipliers are equal to quantization - * coefficients scaled by scalefactor[row]*scalefactor[col], where - * scalefactor[0] = 1 - * scalefactor[k] = cos(k*PI/16) * sqrt(2) for k=1..7 - * For integer operation, the multiplier table is to be scaled by - * IFAST_SCALE_BITS. - */ - IFAST_MULT_TYPE * ifmtbl = (IFAST_MULT_TYPE *) compptr->dct_table; -#define CONST_BITS 14 - static const INT16 aanscales[DCTSIZE2] = { - /* precomputed values scaled up by 14 bits */ - 16384, 22725, 21407, 19266, 16384, 12873, 8867, 4520, - 22725, 31521, 29692, 26722, 22725, 17855, 12299, 6270, - 21407, 29692, 27969, 25172, 21407, 16819, 11585, 5906, - 19266, 26722, 25172, 22654, 19266, 15137, 10426, 5315, - 16384, 22725, 21407, 19266, 16384, 12873, 8867, 4520, - 12873, 17855, 16819, 15137, 12873, 10114, 6967, 3552, - 8867, 12299, 11585, 10426, 8867, 6967, 4799, 2446, - 4520, 6270, 5906, 5315, 4520, 3552, 2446, 1247 - }; - SHIFT_TEMPS - - for (i = 0; i < DCTSIZE2; i++) { - ifmtbl[i] = (IFAST_MULT_TYPE) - DESCALE(MULTIPLY16V16((INT32) qtbl->quantval[i], - (INT32) aanscales[i]), - CONST_BITS-IFAST_SCALE_BITS); - } - } - break; -#endif -#ifdef DCT_FLOAT_SUPPORTED - case JDCT_FLOAT: - { - /* For float AA&N IDCT method, multipliers are equal to quantization - * coefficients scaled by scalefactor[row]*scalefactor[col], where - * scalefactor[0] = 1 - * scalefactor[k] = cos(k*PI/16) * sqrt(2) for k=1..7 - * We apply a further scale factor of 1/8. - */ - FLOAT_MULT_TYPE * fmtbl = (FLOAT_MULT_TYPE *) compptr->dct_table; - int row, col; - static const double aanscalefactor[DCTSIZE] = { - 1.0, 1.387039845, 1.306562965, 1.175875602, - 1.0, 0.785694958, 0.541196100, 0.275899379 - }; - - i = 0; - for (row = 0; row < DCTSIZE; row++) { - for (col = 0; col < DCTSIZE; col++) { - fmtbl[i] = (FLOAT_MULT_TYPE) - ((double) qtbl->quantval[i] * - aanscalefactor[row] * aanscalefactor[col] * 0.125); - i++; - } - } - } - break; -#endif - default: - ERREXIT(cinfo, JERR_NOT_COMPILED); - break; - } - } -} - - -/* - * Initialize IDCT manager. - */ - -GLOBAL(void) -jinit_inverse_dct (j_decompress_ptr cinfo) -{ - my_idct_ptr idct; - int ci; - jpeg_component_info *compptr; - - idct = (my_idct_ptr) - (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE, - SIZEOF(my_idct_controller)); - cinfo->idct = &idct->pub; - idct->pub.start_pass = start_pass; - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* Allocate and pre-zero a multiplier table for each component */ - compptr->dct_table = - (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE, - SIZEOF(multiplier_table)); - MEMZERO(compptr->dct_table, SIZEOF(multiplier_table)); - /* Mark multiplier table not yet set up for any method */ - idct->cur_method[ci] = -1; - } -} diff --git a/dep/libjpeg/src/jdhuff.c b/dep/libjpeg/src/jdhuff.c deleted file mode 100644 index f175f0c32..000000000 --- a/dep/libjpeg/src/jdhuff.c +++ /dev/null @@ -1,1559 +0,0 @@ -/* - * jdhuff.c - * - * Copyright (C) 1991-1997, Thomas G. Lane. - * Modified 2006-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains Huffman entropy decoding routines. - * Both sequential and progressive modes are supported in this single module. - * - * Much of the complexity here has to do with supporting input suspension. - * If the data source module demands suspension, we want to be able to back - * up to the start of the current MCU. To do this, we copy state variables - * into local working storage, and update them back to the permanent - * storage only upon successful completion of an MCU. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Derived data constructed for each Huffman table */ - -#define HUFF_LOOKAHEAD 8 /* # of bits of lookahead */ - -typedef struct { - /* Basic tables: (element [0] of each array is unused) */ - INT32 maxcode[18]; /* largest code of length k (-1 if none) */ - /* (maxcode[17] is a sentinel to ensure jpeg_huff_decode terminates) */ - INT32 valoffset[17]; /* huffval[] offset for codes of length k */ - /* valoffset[k] = huffval[] index of 1st symbol of code length k, less - * the smallest code of length k; so given a code of length k, the - * corresponding symbol is huffval[code + valoffset[k]] - */ - - /* Link to public Huffman table (needed only in jpeg_huff_decode) */ - JHUFF_TBL *pub; - - /* Lookahead tables: indexed by the next HUFF_LOOKAHEAD bits of - * the input data stream. If the next Huffman code is no more - * than HUFF_LOOKAHEAD bits long, we can obtain its length and - * the corresponding symbol directly from these tables. - */ - int look_nbits[1< 32 bits on your machine, and shifting/masking longs is - * reasonably fast, making bit_buf_type be long and setting BIT_BUF_SIZE - * appropriately should be a win. Unfortunately we can't define the size - * with something like #define BIT_BUF_SIZE (sizeof(bit_buf_type)*8) - * because not all machines measure sizeof in 8-bit bytes. - */ - -typedef struct { /* Bitreading state saved across MCUs */ - bit_buf_type get_buffer; /* current bit-extraction buffer */ - int bits_left; /* # of unused bits in it */ -} bitread_perm_state; - -typedef struct { /* Bitreading working state within an MCU */ - /* Current data source location */ - /* We need a copy, rather than munging the original, in case of suspension */ - const JOCTET * next_input_byte; /* => next byte to read from source */ - size_t bytes_in_buffer; /* # of bytes remaining in source buffer */ - /* Bit input buffer --- note these values are kept in register variables, - * not in this struct, inside the inner loops. - */ - bit_buf_type get_buffer; /* current bit-extraction buffer */ - int bits_left; /* # of unused bits in it */ - /* Pointer needed by jpeg_fill_bit_buffer. */ - j_decompress_ptr cinfo; /* back link to decompress master record */ -} bitread_working_state; - -/* Macros to declare and load/save bitread local variables. */ -#define BITREAD_STATE_VARS \ - register bit_buf_type get_buffer; \ - register int bits_left; \ - bitread_working_state br_state - -#define BITREAD_LOAD_STATE(cinfop,permstate) \ - br_state.cinfo = cinfop; \ - br_state.next_input_byte = cinfop->src->next_input_byte; \ - br_state.bytes_in_buffer = cinfop->src->bytes_in_buffer; \ - get_buffer = permstate.get_buffer; \ - bits_left = permstate.bits_left; - -#define BITREAD_SAVE_STATE(cinfop,permstate) \ - cinfop->src->next_input_byte = br_state.next_input_byte; \ - cinfop->src->bytes_in_buffer = br_state.bytes_in_buffer; \ - permstate.get_buffer = get_buffer; \ - permstate.bits_left = bits_left - -/* - * These macros provide the in-line portion of bit fetching. - * Use CHECK_BIT_BUFFER to ensure there are N bits in get_buffer - * before using GET_BITS, PEEK_BITS, or DROP_BITS. - * The variables get_buffer and bits_left are assumed to be locals, - * but the state struct might not be (jpeg_huff_decode needs this). - * CHECK_BIT_BUFFER(state,n,action); - * Ensure there are N bits in get_buffer; if suspend, take action. - * val = GET_BITS(n); - * Fetch next N bits. - * val = PEEK_BITS(n); - * Fetch next N bits without removing them from the buffer. - * DROP_BITS(n); - * Discard next N bits. - * The value N should be a simple variable, not an expression, because it - * is evaluated multiple times. - */ - -#define CHECK_BIT_BUFFER(state,nbits,action) \ - { if (bits_left < (nbits)) { \ - if (! jpeg_fill_bit_buffer(&(state),get_buffer,bits_left,nbits)) \ - { action; } \ - get_buffer = (state).get_buffer; bits_left = (state).bits_left; } } - -#define GET_BITS(nbits) \ - (((int) (get_buffer >> (bits_left -= (nbits)))) & BIT_MASK(nbits)) - -#define PEEK_BITS(nbits) \ - (((int) (get_buffer >> (bits_left - (nbits)))) & BIT_MASK(nbits)) - -#define DROP_BITS(nbits) \ - (bits_left -= (nbits)) - - -/* - * Code for extracting next Huffman-coded symbol from input bit stream. - * Again, this is time-critical and we make the main paths be macros. - * - * We use a lookahead table to process codes of up to HUFF_LOOKAHEAD bits - * without looping. Usually, more than 95% of the Huffman codes will be 8 - * or fewer bits long. The few overlength codes are handled with a loop, - * which need not be inline code. - * - * Notes about the HUFF_DECODE macro: - * 1. Near the end of the data segment, we may fail to get enough bits - * for a lookahead. In that case, we do it the hard way. - * 2. If the lookahead table contains no entry, the next code must be - * more than HUFF_LOOKAHEAD bits long. - * 3. jpeg_huff_decode returns -1 if forced to suspend. - */ - -#define HUFF_DECODE(result,state,htbl,failaction,slowlabel) \ -{ register int nb, look; \ - if (bits_left < HUFF_LOOKAHEAD) { \ - if (! jpeg_fill_bit_buffer(&state,get_buffer,bits_left, 0)) {failaction;} \ - get_buffer = state.get_buffer; bits_left = state.bits_left; \ - if (bits_left < HUFF_LOOKAHEAD) { \ - nb = 1; goto slowlabel; \ - } \ - } \ - look = PEEK_BITS(HUFF_LOOKAHEAD); \ - if ((nb = htbl->look_nbits[look]) != 0) { \ - DROP_BITS(nb); \ - result = htbl->look_sym[look]; \ - } else { \ - nb = HUFF_LOOKAHEAD+1; \ -slowlabel: \ - if ((result=jpeg_huff_decode(&state,get_buffer,bits_left,htbl,nb)) < 0) \ - { failaction; } \ - get_buffer = state.get_buffer; bits_left = state.bits_left; \ - } \ -} - - -/* - * Expanded entropy decoder object for Huffman decoding. - * - * The savable_state subrecord contains fields that change within an MCU, - * but must not be updated permanently until we complete the MCU. - */ - -typedef struct { - unsigned int EOBRUN; /* remaining EOBs in EOBRUN */ - int last_dc_val[MAX_COMPS_IN_SCAN]; /* last DC coef for each component */ -} savable_state; - -/* This macro is to work around compilers with missing or broken - * structure assignment. You'll need to fix this code if you have - * such a compiler and you change MAX_COMPS_IN_SCAN. - */ - -#ifndef NO_STRUCT_ASSIGN -#define ASSIGN_STATE(dest,src) ((dest) = (src)) -#else -#if MAX_COMPS_IN_SCAN == 4 -#define ASSIGN_STATE(dest,src) \ - ((dest).EOBRUN = (src).EOBRUN, \ - (dest).last_dc_val[0] = (src).last_dc_val[0], \ - (dest).last_dc_val[1] = (src).last_dc_val[1], \ - (dest).last_dc_val[2] = (src).last_dc_val[2], \ - (dest).last_dc_val[3] = (src).last_dc_val[3]) -#endif -#endif - - -typedef struct { - struct jpeg_entropy_decoder pub; /* public fields */ - - /* These fields are loaded into local variables at start of each MCU. - * In case of suspension, we exit WITHOUT updating them. - */ - bitread_perm_state bitstate; /* Bit buffer at start of MCU */ - savable_state saved; /* Other state at start of MCU */ - - /* These fields are NOT loaded into local working state. */ - boolean insufficient_data; /* set TRUE after emitting warning */ - unsigned int restarts_to_go; /* MCUs left in this restart interval */ - - /* Following two fields used only in progressive mode */ - - /* Pointers to derived tables (these workspaces have image lifespan) */ - d_derived_tbl * derived_tbls[NUM_HUFF_TBLS]; - - d_derived_tbl * ac_derived_tbl; /* active table during an AC scan */ - - /* Following fields used only in sequential mode */ - - /* Pointers to derived tables (these workspaces have image lifespan) */ - d_derived_tbl * dc_derived_tbls[NUM_HUFF_TBLS]; - d_derived_tbl * ac_derived_tbls[NUM_HUFF_TBLS]; - - /* Precalculated info set up by start_pass for use in decode_mcu: */ - - /* Pointers to derived tables to be used for each block within an MCU */ - d_derived_tbl * dc_cur_tbls[D_MAX_BLOCKS_IN_MCU]; - d_derived_tbl * ac_cur_tbls[D_MAX_BLOCKS_IN_MCU]; - /* Whether we care about the DC and AC coefficient values for each block */ - int coef_limit[D_MAX_BLOCKS_IN_MCU]; -} huff_entropy_decoder; - -typedef huff_entropy_decoder * huff_entropy_ptr; - - -static const int jpeg_zigzag_order[8][8] = { - { 0, 1, 5, 6, 14, 15, 27, 28 }, - { 2, 4, 7, 13, 16, 26, 29, 42 }, - { 3, 8, 12, 17, 25, 30, 41, 43 }, - { 9, 11, 18, 24, 31, 40, 44, 53 }, - { 10, 19, 23, 32, 39, 45, 52, 54 }, - { 20, 22, 33, 38, 46, 51, 55, 60 }, - { 21, 34, 37, 47, 50, 56, 59, 61 }, - { 35, 36, 48, 49, 57, 58, 62, 63 } -}; - -static const int jpeg_zigzag_order7[7][7] = { - { 0, 1, 5, 6, 14, 15, 27 }, - { 2, 4, 7, 13, 16, 26, 28 }, - { 3, 8, 12, 17, 25, 29, 38 }, - { 9, 11, 18, 24, 30, 37, 39 }, - { 10, 19, 23, 31, 36, 40, 45 }, - { 20, 22, 32, 35, 41, 44, 46 }, - { 21, 33, 34, 42, 43, 47, 48 } -}; - -static const int jpeg_zigzag_order6[6][6] = { - { 0, 1, 5, 6, 14, 15 }, - { 2, 4, 7, 13, 16, 25 }, - { 3, 8, 12, 17, 24, 26 }, - { 9, 11, 18, 23, 27, 32 }, - { 10, 19, 22, 28, 31, 33 }, - { 20, 21, 29, 30, 34, 35 } -}; - -static const int jpeg_zigzag_order5[5][5] = { - { 0, 1, 5, 6, 14 }, - { 2, 4, 7, 13, 15 }, - { 3, 8, 12, 16, 21 }, - { 9, 11, 17, 20, 22 }, - { 10, 18, 19, 23, 24 } -}; - -static const int jpeg_zigzag_order4[4][4] = { - { 0, 1, 5, 6 }, - { 2, 4, 7, 12 }, - { 3, 8, 11, 13 }, - { 9, 10, 14, 15 } -}; - -static const int jpeg_zigzag_order3[3][3] = { - { 0, 1, 5 }, - { 2, 4, 6 }, - { 3, 7, 8 } -}; - -static const int jpeg_zigzag_order2[2][2] = { - { 0, 1 }, - { 2, 3 } -}; - - -/* - * Compute the derived values for a Huffman table. - * This routine also performs some validation checks on the table. - */ - -LOCAL(void) -jpeg_make_d_derived_tbl (j_decompress_ptr cinfo, boolean isDC, int tblno, - d_derived_tbl ** pdtbl) -{ - JHUFF_TBL *htbl; - d_derived_tbl *dtbl; - int p, i, l, si, numsymbols; - int lookbits, ctr; - char huffsize[257]; - unsigned int huffcode[257]; - unsigned int code; - - /* Note that huffsize[] and huffcode[] are filled in code-length order, - * paralleling the order of the symbols themselves in htbl->huffval[]. - */ - - /* Find the input Huffman table */ - if (tblno < 0 || tblno >= NUM_HUFF_TBLS) - ERREXIT1(cinfo, JERR_NO_HUFF_TABLE, tblno); - htbl = - isDC ? cinfo->dc_huff_tbl_ptrs[tblno] : cinfo->ac_huff_tbl_ptrs[tblno]; - if (htbl == NULL) - htbl = jpeg_std_huff_table((j_common_ptr) cinfo, isDC, tblno); - - /* Allocate a workspace if we haven't already done so. */ - if (*pdtbl == NULL) - *pdtbl = (d_derived_tbl *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(d_derived_tbl)); - dtbl = *pdtbl; - dtbl->pub = htbl; /* fill in back link */ - - /* Figure C.1: make table of Huffman code length for each symbol */ - - p = 0; - for (l = 1; l <= 16; l++) { - i = (int) htbl->bits[l]; - if (i < 0 || p + i > 256) /* protect against table overrun */ - ERREXIT(cinfo, JERR_BAD_HUFF_TABLE); - while (i--) - huffsize[p++] = (char) l; - } - huffsize[p] = 0; - numsymbols = p; - - /* Figure C.2: generate the codes themselves */ - /* We also validate that the counts represent a legal Huffman code tree. */ - - code = 0; - si = huffsize[0]; - p = 0; - while (huffsize[p]) { - while (((int) huffsize[p]) == si) { - huffcode[p++] = code; - code++; - } - /* code is now 1 more than the last code used for codelength si; but - * it must still fit in si bits, since no code is allowed to be all ones. - */ - if (((INT32) code) >= (((INT32) 1) << si)) - ERREXIT(cinfo, JERR_BAD_HUFF_TABLE); - code <<= 1; - si++; - } - - /* Figure F.15: generate decoding tables for bit-sequential decoding */ - - p = 0; - for (l = 1; l <= 16; l++) { - if (htbl->bits[l]) { - /* valoffset[l] = huffval[] index of 1st symbol of code length l, - * minus the minimum code of length l - */ - dtbl->valoffset[l] = (INT32) p - (INT32) huffcode[p]; - p += htbl->bits[l]; - dtbl->maxcode[l] = huffcode[p-1]; /* maximum code of length l */ - } else { - dtbl->maxcode[l] = -1; /* -1 if no codes of this length */ - } - } - dtbl->maxcode[17] = 0xFFFFFL; /* ensures jpeg_huff_decode terminates */ - - /* Compute lookahead tables to speed up decoding. - * First we set all the table entries to 0, indicating "too long"; - * then we iterate through the Huffman codes that are short enough and - * fill in all the entries that correspond to bit sequences starting - * with that code. - */ - - MEMZERO(dtbl->look_nbits, SIZEOF(dtbl->look_nbits)); - - p = 0; - for (l = 1; l <= HUFF_LOOKAHEAD; l++) { - for (i = 1; i <= (int) htbl->bits[l]; i++, p++) { - /* l = current code's length, p = its index in huffcode[] & huffval[]. */ - /* Generate left-justified code followed by all possible bit sequences */ - lookbits = huffcode[p] << (HUFF_LOOKAHEAD-l); - for (ctr = 1 << (HUFF_LOOKAHEAD-l); ctr > 0; ctr--) { - dtbl->look_nbits[lookbits] = l; - dtbl->look_sym[lookbits] = htbl->huffval[p]; - lookbits++; - } - } - } - - /* Validate symbols as being reasonable. - * For AC tables, we make no check, but accept all byte values 0..255. - * For DC tables, we require the symbols to be in range 0..15. - * (Tighter bounds could be applied depending on the data depth and mode, - * but this is sufficient to ensure safe decoding.) - */ - if (isDC) { - for (i = 0; i < numsymbols; i++) { - int sym = htbl->huffval[i]; - if (sym < 0 || sym > 15) - ERREXIT(cinfo, JERR_BAD_HUFF_TABLE); - } - } -} - - -/* - * Out-of-line code for bit fetching. - * Note: current values of get_buffer and bits_left are passed as parameters, - * but are returned in the corresponding fields of the state struct. - * - * On most machines MIN_GET_BITS should be 25 to allow the full 32-bit width - * of get_buffer to be used. (On machines with wider words, an even larger - * buffer could be used.) However, on some machines 32-bit shifts are - * quite slow and take time proportional to the number of places shifted. - * (This is true with most PC compilers, for instance.) In this case it may - * be a win to set MIN_GET_BITS to the minimum value of 15. This reduces the - * average shift distance at the cost of more calls to jpeg_fill_bit_buffer. - */ - -#ifdef SLOW_SHIFT_32 -#define MIN_GET_BITS 15 /* minimum allowable value */ -#else -#define MIN_GET_BITS (BIT_BUF_SIZE-7) -#endif - - -LOCAL(boolean) -jpeg_fill_bit_buffer (bitread_working_state * state, - register bit_buf_type get_buffer, register int bits_left, - int nbits) -/* Load up the bit buffer to a depth of at least nbits */ -{ - /* Copy heavily used state fields into locals (hopefully registers) */ - register const JOCTET * next_input_byte = state->next_input_byte; - register size_t bytes_in_buffer = state->bytes_in_buffer; - j_decompress_ptr cinfo = state->cinfo; - - /* Attempt to load at least MIN_GET_BITS bits into get_buffer. */ - /* (It is assumed that no request will be for more than that many bits.) */ - /* We fail to do so only if we hit a marker or are forced to suspend. */ - - if (cinfo->unread_marker == 0) { /* cannot advance past a marker */ - while (bits_left < MIN_GET_BITS) { - register int c; - - /* Attempt to read a byte */ - if (bytes_in_buffer == 0) { - if (! (*cinfo->src->fill_input_buffer) (cinfo)) - return FALSE; - next_input_byte = cinfo->src->next_input_byte; - bytes_in_buffer = cinfo->src->bytes_in_buffer; - } - bytes_in_buffer--; - c = GETJOCTET(*next_input_byte++); - - /* If it's 0xFF, check and discard stuffed zero byte */ - if (c == 0xFF) { - /* Loop here to discard any padding FF's on terminating marker, - * so that we can save a valid unread_marker value. NOTE: we will - * accept multiple FF's followed by a 0 as meaning a single FF data - * byte. This data pattern is not valid according to the standard. - */ - do { - if (bytes_in_buffer == 0) { - if (! (*cinfo->src->fill_input_buffer) (cinfo)) - return FALSE; - next_input_byte = cinfo->src->next_input_byte; - bytes_in_buffer = cinfo->src->bytes_in_buffer; - } - bytes_in_buffer--; - c = GETJOCTET(*next_input_byte++); - } while (c == 0xFF); - - if (c == 0) { - /* Found FF/00, which represents an FF data byte */ - c = 0xFF; - } else { - /* Oops, it's actually a marker indicating end of compressed data. - * Save the marker code for later use. - * Fine point: it might appear that we should save the marker into - * bitread working state, not straight into permanent state. But - * once we have hit a marker, we cannot need to suspend within the - * current MCU, because we will read no more bytes from the data - * source. So it is OK to update permanent state right away. - */ - cinfo->unread_marker = c; - /* See if we need to insert some fake zero bits. */ - goto no_more_bytes; - } - } - - /* OK, load c into get_buffer */ - get_buffer = (get_buffer << 8) | c; - bits_left += 8; - } /* end while */ - } else { - no_more_bytes: - /* We get here if we've read the marker that terminates the compressed - * data segment. There should be enough bits in the buffer register - * to satisfy the request; if so, no problem. - */ - if (nbits > bits_left) { - /* Uh-oh. Report corrupted data to user and stuff zeroes into - * the data stream, so that we can produce some kind of image. - * We use a nonvolatile flag to ensure that only one warning message - * appears per data segment. - */ - if (! ((huff_entropy_ptr) cinfo->entropy)->insufficient_data) { - WARNMS(cinfo, JWRN_HIT_MARKER); - ((huff_entropy_ptr) cinfo->entropy)->insufficient_data = TRUE; - } - /* Fill the buffer with zero bits */ - get_buffer <<= MIN_GET_BITS - bits_left; - bits_left = MIN_GET_BITS; - } - } - - /* Unload the local registers */ - state->next_input_byte = next_input_byte; - state->bytes_in_buffer = bytes_in_buffer; - state->get_buffer = get_buffer; - state->bits_left = bits_left; - - return TRUE; -} - - -/* - * Figure F.12: extend sign bit. - * On some machines, a shift and sub will be faster than a table lookup. - */ - -#ifdef AVOID_TABLES - -#define BIT_MASK(nbits) ((1<<(nbits))-1) -#define HUFF_EXTEND(x,s) ((x) < (1<<((s)-1)) ? (x) - ((1<<(s))-1) : (x)) - -#else - -#define BIT_MASK(nbits) bmask[nbits] -#define HUFF_EXTEND(x,s) ((x) <= bmask[(s) - 1] ? (x) - bmask[s] : (x)) - -static const int bmask[16] = /* bmask[n] is mask for n rightmost bits */ - { 0, 0x0001, 0x0003, 0x0007, 0x000F, 0x001F, 0x003F, 0x007F, 0x00FF, - 0x01FF, 0x03FF, 0x07FF, 0x0FFF, 0x1FFF, 0x3FFF, 0x7FFF }; - -#endif /* AVOID_TABLES */ - - -/* - * Out-of-line code for Huffman code decoding. - */ - -LOCAL(int) -jpeg_huff_decode (bitread_working_state * state, - register bit_buf_type get_buffer, register int bits_left, - d_derived_tbl * htbl, int min_bits) -{ - register int l = min_bits; - register INT32 code; - - /* HUFF_DECODE has determined that the code is at least min_bits */ - /* bits long, so fetch that many bits in one swoop. */ - - CHECK_BIT_BUFFER(*state, l, return -1); - code = GET_BITS(l); - - /* Collect the rest of the Huffman code one bit at a time. */ - /* This is per Figure F.16 in the JPEG spec. */ - - while (code > htbl->maxcode[l]) { - code <<= 1; - CHECK_BIT_BUFFER(*state, 1, return -1); - code |= GET_BITS(1); - l++; - } - - /* Unload the local registers */ - state->get_buffer = get_buffer; - state->bits_left = bits_left; - - /* With garbage input we may reach the sentinel value l = 17. */ - - if (l > 16) { - WARNMS(state->cinfo, JWRN_HUFF_BAD_CODE); - return 0; /* fake a zero as the safest result */ - } - - return htbl->pub->huffval[ (int) (code + htbl->valoffset[l]) ]; -} - - -/* - * Finish up at the end of a Huffman-compressed scan. - */ - -METHODDEF(void) -finish_pass_huff (j_decompress_ptr cinfo) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - - /* Throw away any unused bits remaining in bit buffer; */ - /* include any full bytes in next_marker's count of discarded bytes */ - cinfo->marker->discarded_bytes += entropy->bitstate.bits_left / 8; - entropy->bitstate.bits_left = 0; -} - - -/* - * Check for a restart marker & resynchronize decoder. - * Returns FALSE if must suspend. - */ - -LOCAL(boolean) -process_restart (j_decompress_ptr cinfo) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - int ci; - - finish_pass_huff(cinfo); - - /* Advance past the RSTn marker */ - if (! (*cinfo->marker->read_restart_marker) (cinfo)) - return FALSE; - - /* Re-initialize DC predictions to 0 */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) - entropy->saved.last_dc_val[ci] = 0; - /* Re-init EOB run count, too */ - entropy->saved.EOBRUN = 0; - - /* Reset restart counter */ - entropy->restarts_to_go = cinfo->restart_interval; - - /* Reset out-of-data flag, unless read_restart_marker left us smack up - * against a marker. In that case we will end up treating the next data - * segment as empty, and we can avoid producing bogus output pixels by - * leaving the flag set. - */ - if (cinfo->unread_marker == 0) - entropy->insufficient_data = FALSE; - - return TRUE; -} - - -/* - * Huffman MCU decoding. - * Each of these routines decodes and returns one MCU's worth of - * Huffman-compressed coefficients. - * The coefficients are reordered from zigzag order into natural array order, - * but are not dequantized. - * - * The i'th block of the MCU is stored into the block pointed to by - * MCU_data[i]. WE ASSUME THIS AREA IS INITIALLY ZEROED BY THE CALLER. - * (Wholesale zeroing is usually a little faster than retail...) - * - * We return FALSE if data source requested suspension. In that case no - * changes have been made to permanent state. (Exception: some output - * coefficients may already have been assigned. This is harmless for - * spectral selection, since we'll just re-assign them on the next call. - * Successive approximation AC refinement has to be more careful, however.) - */ - -/* - * MCU decoding for DC initial scan (either spectral selection, - * or first pass of successive approximation). - */ - -METHODDEF(boolean) -decode_mcu_DC_first (j_decompress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - int Al = cinfo->Al; - register int s, r; - int blkn, ci; - JBLOCKROW block; - BITREAD_STATE_VARS; - savable_state state; - d_derived_tbl * tbl; - jpeg_component_info * compptr; - - /* Process restart marker if needed; may have to suspend */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) - if (! process_restart(cinfo)) - return FALSE; - } - - /* If we've run out of data, just leave the MCU set to zeroes. - * This way, we return uniform gray for the remainder of the segment. - */ - if (! entropy->insufficient_data) { - - /* Load up working state */ - BITREAD_LOAD_STATE(cinfo, entropy->bitstate); - ASSIGN_STATE(state, entropy->saved); - - /* Outer loop handles each block in the MCU */ - - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - block = MCU_data[blkn]; - ci = cinfo->MCU_membership[blkn]; - compptr = cinfo->cur_comp_info[ci]; - tbl = entropy->derived_tbls[compptr->dc_tbl_no]; - - /* Decode a single block's worth of coefficients */ - - /* Section F.2.2.1: decode the DC coefficient difference */ - HUFF_DECODE(s, br_state, tbl, return FALSE, label1); - if (s) { - CHECK_BIT_BUFFER(br_state, s, return FALSE); - r = GET_BITS(s); - s = HUFF_EXTEND(r, s); - } - - /* Convert DC difference to actual value, update last_dc_val */ - s += state.last_dc_val[ci]; - state.last_dc_val[ci] = s; - /* Scale and output the coefficient (assumes jpeg_natural_order[0]=0) */ - (*block)[0] = (JCOEF) (s << Al); - } - - /* Completed MCU, so update state */ - BITREAD_SAVE_STATE(cinfo, entropy->bitstate); - ASSIGN_STATE(entropy->saved, state); - } - - /* Account for restart interval if using restarts */ - if (cinfo->restart_interval) - entropy->restarts_to_go--; - - return TRUE; -} - - -/* - * MCU decoding for AC initial scan (either spectral selection, - * or first pass of successive approximation). - */ - -METHODDEF(boolean) -decode_mcu_AC_first (j_decompress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - register int s, k, r; - unsigned int EOBRUN; - int Se, Al; - const int * natural_order; - JBLOCKROW block; - BITREAD_STATE_VARS; - d_derived_tbl * tbl; - - /* Process restart marker if needed; may have to suspend */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) - if (! process_restart(cinfo)) - return FALSE; - } - - /* If we've run out of data, just leave the MCU set to zeroes. - * This way, we return uniform gray for the remainder of the segment. - */ - if (! entropy->insufficient_data) { - - /* Load up working state. - * We can avoid loading/saving bitread state if in an EOB run. - */ - EOBRUN = entropy->saved.EOBRUN; /* only part of saved state we need */ - - /* There is always only one block per MCU */ - - if (EOBRUN) /* if it's a band of zeroes... */ - EOBRUN--; /* ...process it now (we do nothing) */ - else { - BITREAD_LOAD_STATE(cinfo, entropy->bitstate); - Se = cinfo->Se; - Al = cinfo->Al; - natural_order = cinfo->natural_order; - block = MCU_data[0]; - tbl = entropy->ac_derived_tbl; - - for (k = cinfo->Ss; k <= Se; k++) { - HUFF_DECODE(s, br_state, tbl, return FALSE, label2); - r = s >> 4; - s &= 15; - if (s) { - k += r; - CHECK_BIT_BUFFER(br_state, s, return FALSE); - r = GET_BITS(s); - s = HUFF_EXTEND(r, s); - /* Scale and output coefficient in natural (dezigzagged) order */ - (*block)[natural_order[k]] = (JCOEF) (s << Al); - } else { - if (r != 15) { /* EOBr, run length is 2^r + appended bits */ - if (r) { /* EOBr, r > 0 */ - EOBRUN = 1 << r; - CHECK_BIT_BUFFER(br_state, r, return FALSE); - r = GET_BITS(r); - EOBRUN += r; - EOBRUN--; /* this band is processed at this moment */ - } - break; /* force end-of-band */ - } - k += 15; /* ZRL: skip 15 zeroes in band */ - } - } - - BITREAD_SAVE_STATE(cinfo, entropy->bitstate); - } - - /* Completed MCU, so update state */ - entropy->saved.EOBRUN = EOBRUN; /* only part of saved state we need */ - } - - /* Account for restart interval if using restarts */ - if (cinfo->restart_interval) - entropy->restarts_to_go--; - - return TRUE; -} - - -/* - * MCU decoding for DC successive approximation refinement scan. - * Note: we assume such scans can be multi-component, - * although the spec is not very clear on the point. - */ - -METHODDEF(boolean) -decode_mcu_DC_refine (j_decompress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - JCOEF p1; - int blkn; - BITREAD_STATE_VARS; - - /* Process restart marker if needed; may have to suspend */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) - if (! process_restart(cinfo)) - return FALSE; - } - - /* Not worth the cycles to check insufficient_data here, - * since we will not change the data anyway if we read zeroes. - */ - - /* Load up working state */ - BITREAD_LOAD_STATE(cinfo, entropy->bitstate); - - p1 = 1 << cinfo->Al; /* 1 in the bit position being coded */ - - /* Outer loop handles each block in the MCU */ - - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - /* Encoded data is simply the next bit of the two's-complement DC value */ - CHECK_BIT_BUFFER(br_state, 1, return FALSE); - if (GET_BITS(1)) - MCU_data[blkn][0][0] |= p1; - /* Note: since we use |=, repeating the assignment later is safe */ - } - - /* Completed MCU, so update state */ - BITREAD_SAVE_STATE(cinfo, entropy->bitstate); - - /* Account for restart interval if using restarts */ - if (cinfo->restart_interval) - entropy->restarts_to_go--; - - return TRUE; -} - - -/* - * MCU decoding for AC successive approximation refinement scan. - */ - -METHODDEF(boolean) -decode_mcu_AC_refine (j_decompress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - register int s, k, r; - unsigned int EOBRUN; - int Se; - JCOEF p1, m1; - const int * natural_order; - JBLOCKROW block; - JCOEFPTR thiscoef; - BITREAD_STATE_VARS; - d_derived_tbl * tbl; - int num_newnz; - int newnz_pos[DCTSIZE2]; - - /* Process restart marker if needed; may have to suspend */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) - if (! process_restart(cinfo)) - return FALSE; - } - - /* If we've run out of data, don't modify the MCU. - */ - if (! entropy->insufficient_data) { - - Se = cinfo->Se; - p1 = 1 << cinfo->Al; /* 1 in the bit position being coded */ - m1 = -p1; /* -1 in the bit position being coded */ - natural_order = cinfo->natural_order; - - /* Load up working state */ - BITREAD_LOAD_STATE(cinfo, entropy->bitstate); - EOBRUN = entropy->saved.EOBRUN; /* only part of saved state we need */ - - /* There is always only one block per MCU */ - block = MCU_data[0]; - tbl = entropy->ac_derived_tbl; - - /* If we are forced to suspend, we must undo the assignments to any newly - * nonzero coefficients in the block, because otherwise we'd get confused - * next time about which coefficients were already nonzero. - * But we need not undo addition of bits to already-nonzero coefficients; - * instead, we can test the current bit to see if we already did it. - */ - num_newnz = 0; - - /* initialize coefficient loop counter to start of band */ - k = cinfo->Ss; - - if (EOBRUN == 0) { - do { - HUFF_DECODE(s, br_state, tbl, goto undoit, label3); - r = s >> 4; - s &= 15; - if (s) { - if (s != 1) /* size of new coef should always be 1 */ - WARNMS(cinfo, JWRN_HUFF_BAD_CODE); - CHECK_BIT_BUFFER(br_state, 1, goto undoit); - if (GET_BITS(1)) - s = p1; /* newly nonzero coef is positive */ - else - s = m1; /* newly nonzero coef is negative */ - } else { - if (r != 15) { - EOBRUN = 1 << r; /* EOBr, run length is 2^r + appended bits */ - if (r) { - CHECK_BIT_BUFFER(br_state, r, goto undoit); - r = GET_BITS(r); - EOBRUN += r; - } - break; /* rest of block is handled by EOB logic */ - } - /* note s = 0 for processing ZRL */ - } - /* Advance over already-nonzero coefs and r still-zero coefs, - * appending correction bits to the nonzeroes. A correction bit is 1 - * if the absolute value of the coefficient must be increased. - */ - do { - thiscoef = *block + natural_order[k]; - if (*thiscoef) { - CHECK_BIT_BUFFER(br_state, 1, goto undoit); - if (GET_BITS(1)) { - if ((*thiscoef & p1) == 0) { /* do nothing if already set it */ - if (*thiscoef >= 0) - *thiscoef += p1; - else - *thiscoef += m1; - } - } - } else { - if (--r < 0) - break; /* reached target zero coefficient */ - } - k++; - } while (k <= Se); - if (s) { - int pos = natural_order[k]; - /* Output newly nonzero coefficient */ - (*block)[pos] = (JCOEF) s; - /* Remember its position in case we have to suspend */ - newnz_pos[num_newnz++] = pos; - } - k++; - } while (k <= Se); - } - - if (EOBRUN) { - /* Scan any remaining coefficient positions after the end-of-band - * (the last newly nonzero coefficient, if any). Append a correction - * bit to each already-nonzero coefficient. A correction bit is 1 - * if the absolute value of the coefficient must be increased. - */ - do { - thiscoef = *block + natural_order[k]; - if (*thiscoef) { - CHECK_BIT_BUFFER(br_state, 1, goto undoit); - if (GET_BITS(1)) { - if ((*thiscoef & p1) == 0) { /* do nothing if already changed it */ - if (*thiscoef >= 0) - *thiscoef += p1; - else - *thiscoef += m1; - } - } - } - k++; - } while (k <= Se); - /* Count one block completed in EOB run */ - EOBRUN--; - } - - /* Completed MCU, so update state */ - BITREAD_SAVE_STATE(cinfo, entropy->bitstate); - entropy->saved.EOBRUN = EOBRUN; /* only part of saved state we need */ - } - - /* Account for restart interval if using restarts */ - if (cinfo->restart_interval) - entropy->restarts_to_go--; - - return TRUE; - -undoit: - /* Re-zero any output coefficients that we made newly nonzero */ - while (num_newnz) - (*block)[newnz_pos[--num_newnz]] = 0; - - return FALSE; -} - - -/* - * Decode one MCU's worth of Huffman-compressed coefficients, - * partial blocks. - */ - -METHODDEF(boolean) -decode_mcu_sub (j_decompress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - const int * natural_order; - int Se, blkn; - BITREAD_STATE_VARS; - savable_state state; - - /* Process restart marker if needed; may have to suspend */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) - if (! process_restart(cinfo)) - return FALSE; - } - - /* If we've run out of data, just leave the MCU set to zeroes. - * This way, we return uniform gray for the remainder of the segment. - */ - if (! entropy->insufficient_data) { - - natural_order = cinfo->natural_order; - Se = cinfo->lim_Se; - - /* Load up working state */ - BITREAD_LOAD_STATE(cinfo, entropy->bitstate); - ASSIGN_STATE(state, entropy->saved); - - /* Outer loop handles each block in the MCU */ - - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - JBLOCKROW block = MCU_data[blkn]; - d_derived_tbl * htbl; - register int s, k, r; - int coef_limit, ci; - - /* Decode a single block's worth of coefficients */ - - /* Section F.2.2.1: decode the DC coefficient difference */ - htbl = entropy->dc_cur_tbls[blkn]; - HUFF_DECODE(s, br_state, htbl, return FALSE, label1); - - htbl = entropy->ac_cur_tbls[blkn]; - k = 1; - coef_limit = entropy->coef_limit[blkn]; - if (coef_limit) { - /* Convert DC difference to actual value, update last_dc_val */ - if (s) { - CHECK_BIT_BUFFER(br_state, s, return FALSE); - r = GET_BITS(s); - s = HUFF_EXTEND(r, s); - } - ci = cinfo->MCU_membership[blkn]; - s += state.last_dc_val[ci]; - state.last_dc_val[ci] = s; - /* Output the DC coefficient */ - (*block)[0] = (JCOEF) s; - - /* Section F.2.2.2: decode the AC coefficients */ - /* Since zeroes are skipped, output area must be cleared beforehand */ - for (; k < coef_limit; k++) { - HUFF_DECODE(s, br_state, htbl, return FALSE, label2); - - r = s >> 4; - s &= 15; - - if (s) { - k += r; - CHECK_BIT_BUFFER(br_state, s, return FALSE); - r = GET_BITS(s); - s = HUFF_EXTEND(r, s); - /* Output coefficient in natural (dezigzagged) order. - * Note: the extra entries in natural_order[] will save us - * if k > Se, which could happen if the data is corrupted. - */ - (*block)[natural_order[k]] = (JCOEF) s; - } else { - if (r != 15) - goto EndOfBlock; - k += 15; - } - } - } else { - if (s) { - CHECK_BIT_BUFFER(br_state, s, return FALSE); - DROP_BITS(s); - } - } - - /* Section F.2.2.2: decode the AC coefficients */ - /* In this path we just discard the values */ - for (; k <= Se; k++) { - HUFF_DECODE(s, br_state, htbl, return FALSE, label3); - - r = s >> 4; - s &= 15; - - if (s) { - k += r; - CHECK_BIT_BUFFER(br_state, s, return FALSE); - DROP_BITS(s); - } else { - if (r != 15) - break; - k += 15; - } - } - - EndOfBlock: ; - } - - /* Completed MCU, so update state */ - BITREAD_SAVE_STATE(cinfo, entropy->bitstate); - ASSIGN_STATE(entropy->saved, state); - } - - /* Account for restart interval if using restarts */ - if (cinfo->restart_interval) - entropy->restarts_to_go--; - - return TRUE; -} - - -/* - * Decode one MCU's worth of Huffman-compressed coefficients, - * full-size blocks. - */ - -METHODDEF(boolean) -decode_mcu (j_decompress_ptr cinfo, JBLOCKARRAY MCU_data) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - int blkn; - BITREAD_STATE_VARS; - savable_state state; - - /* Process restart marker if needed; may have to suspend */ - if (cinfo->restart_interval) { - if (entropy->restarts_to_go == 0) - if (! process_restart(cinfo)) - return FALSE; - } - - /* If we've run out of data, just leave the MCU set to zeroes. - * This way, we return uniform gray for the remainder of the segment. - */ - if (! entropy->insufficient_data) { - - /* Load up working state */ - BITREAD_LOAD_STATE(cinfo, entropy->bitstate); - ASSIGN_STATE(state, entropy->saved); - - /* Outer loop handles each block in the MCU */ - - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - JBLOCKROW block = MCU_data[blkn]; - d_derived_tbl * htbl; - register int s, k, r; - int coef_limit, ci; - - /* Decode a single block's worth of coefficients */ - - /* Section F.2.2.1: decode the DC coefficient difference */ - htbl = entropy->dc_cur_tbls[blkn]; - HUFF_DECODE(s, br_state, htbl, return FALSE, label1); - - htbl = entropy->ac_cur_tbls[blkn]; - k = 1; - coef_limit = entropy->coef_limit[blkn]; - if (coef_limit) { - /* Convert DC difference to actual value, update last_dc_val */ - if (s) { - CHECK_BIT_BUFFER(br_state, s, return FALSE); - r = GET_BITS(s); - s = HUFF_EXTEND(r, s); - } - ci = cinfo->MCU_membership[blkn]; - s += state.last_dc_val[ci]; - state.last_dc_val[ci] = s; - /* Output the DC coefficient */ - (*block)[0] = (JCOEF) s; - - /* Section F.2.2.2: decode the AC coefficients */ - /* Since zeroes are skipped, output area must be cleared beforehand */ - for (; k < coef_limit; k++) { - HUFF_DECODE(s, br_state, htbl, return FALSE, label2); - - r = s >> 4; - s &= 15; - - if (s) { - k += r; - CHECK_BIT_BUFFER(br_state, s, return FALSE); - r = GET_BITS(s); - s = HUFF_EXTEND(r, s); - /* Output coefficient in natural (dezigzagged) order. - * Note: the extra entries in jpeg_natural_order[] will save us - * if k >= DCTSIZE2, which could happen if the data is corrupted. - */ - (*block)[jpeg_natural_order[k]] = (JCOEF) s; - } else { - if (r != 15) - goto EndOfBlock; - k += 15; - } - } - } else { - if (s) { - CHECK_BIT_BUFFER(br_state, s, return FALSE); - DROP_BITS(s); - } - } - - /* Section F.2.2.2: decode the AC coefficients */ - /* In this path we just discard the values */ - for (; k < DCTSIZE2; k++) { - HUFF_DECODE(s, br_state, htbl, return FALSE, label3); - - r = s >> 4; - s &= 15; - - if (s) { - k += r; - CHECK_BIT_BUFFER(br_state, s, return FALSE); - DROP_BITS(s); - } else { - if (r != 15) - break; - k += 15; - } - } - - EndOfBlock: ; - } - - /* Completed MCU, so update state */ - BITREAD_SAVE_STATE(cinfo, entropy->bitstate); - ASSIGN_STATE(entropy->saved, state); - } - - /* Account for restart interval if using restarts */ - if (cinfo->restart_interval) - entropy->restarts_to_go--; - - return TRUE; -} - - -/* - * Initialize for a Huffman-compressed scan. - */ - -METHODDEF(void) -start_pass_huff_decoder (j_decompress_ptr cinfo) -{ - huff_entropy_ptr entropy = (huff_entropy_ptr) cinfo->entropy; - int ci, blkn, tbl, i; - jpeg_component_info * compptr; - - if (cinfo->progressive_mode) { - /* Validate progressive scan parameters */ - if (cinfo->Ss == 0) { - if (cinfo->Se != 0) - goto bad; - } else { - /* need not check Ss/Se < 0 since they came from unsigned bytes */ - if (cinfo->Se < cinfo->Ss || cinfo->Se > cinfo->lim_Se) - goto bad; - /* AC scans may have only one component */ - if (cinfo->comps_in_scan != 1) - goto bad; - } - if (cinfo->Ah != 0) { - /* Successive approximation refinement scan: must have Al = Ah-1. */ - if (cinfo->Ah-1 != cinfo->Al) - goto bad; - } - if (cinfo->Al > 13) { /* need not check for < 0 */ - /* Arguably the maximum Al value should be less than 13 for 8-bit - * precision, but the spec doesn't say so, and we try to be liberal - * about what we accept. Note: large Al values could result in - * out-of-range DC coefficients during early scans, leading to bizarre - * displays due to overflows in the IDCT math. But we won't crash. - */ - bad: - ERREXIT4(cinfo, JERR_BAD_PROGRESSION, - cinfo->Ss, cinfo->Se, cinfo->Ah, cinfo->Al); - } - /* Update progression status, and verify that scan order is legal. - * Note that inter-scan inconsistencies are treated as warnings - * not fatal errors ... not clear if this is right way to behave. - */ - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - int coefi, cindex = cinfo->cur_comp_info[ci]->component_index; - int *coef_bit_ptr = & cinfo->coef_bits[cindex][0]; - if (cinfo->Ss && coef_bit_ptr[0] < 0) /* AC without prior DC scan */ - WARNMS2(cinfo, JWRN_BOGUS_PROGRESSION, cindex, 0); - for (coefi = cinfo->Ss; coefi <= cinfo->Se; coefi++) { - int expected = (coef_bit_ptr[coefi] < 0) ? 0 : coef_bit_ptr[coefi]; - if (cinfo->Ah != expected) - WARNMS2(cinfo, JWRN_BOGUS_PROGRESSION, cindex, coefi); - coef_bit_ptr[coefi] = cinfo->Al; - } - } - - /* Select MCU decoding routine */ - if (cinfo->Ah == 0) { - if (cinfo->Ss == 0) - entropy->pub.decode_mcu = decode_mcu_DC_first; - else - entropy->pub.decode_mcu = decode_mcu_AC_first; - } else { - if (cinfo->Ss == 0) - entropy->pub.decode_mcu = decode_mcu_DC_refine; - else - entropy->pub.decode_mcu = decode_mcu_AC_refine; - } - - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - /* Make sure requested tables are present, and compute derived tables. - * We may build same derived table more than once, but it's not expensive. - */ - if (cinfo->Ss == 0) { - if (cinfo->Ah == 0) { /* DC refinement needs no table */ - tbl = compptr->dc_tbl_no; - jpeg_make_d_derived_tbl(cinfo, TRUE, tbl, - & entropy->derived_tbls[tbl]); - } - } else { - tbl = compptr->ac_tbl_no; - jpeg_make_d_derived_tbl(cinfo, FALSE, tbl, - & entropy->derived_tbls[tbl]); - /* remember the single active table */ - entropy->ac_derived_tbl = entropy->derived_tbls[tbl]; - } - /* Initialize DC predictions to 0 */ - entropy->saved.last_dc_val[ci] = 0; - } - - /* Initialize private state variables */ - entropy->saved.EOBRUN = 0; - } else { - /* Check that the scan parameters Ss, Se, Ah/Al are OK for sequential JPEG. - * This ought to be an error condition, but we make it a warning because - * there are some baseline files out there with all zeroes in these bytes. - */ - if (cinfo->Ss != 0 || cinfo->Ah != 0 || cinfo->Al != 0 || - ((cinfo->is_baseline || cinfo->Se < DCTSIZE2) && - cinfo->Se != cinfo->lim_Se)) - WARNMS(cinfo, JWRN_NOT_SEQUENTIAL); - - /* Select MCU decoding routine */ - /* We retain the hard-coded case for full-size blocks. - * This is not necessary, but it appears that this version is slightly - * more performant in the given implementation. - * With an improved implementation we would prefer a single optimized - * function. - */ - if (cinfo->lim_Se != DCTSIZE2-1) - entropy->pub.decode_mcu = decode_mcu_sub; - else - entropy->pub.decode_mcu = decode_mcu; - - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - /* Compute derived values for Huffman tables */ - /* We may do this more than once for a table, but it's not expensive */ - tbl = compptr->dc_tbl_no; - jpeg_make_d_derived_tbl(cinfo, TRUE, tbl, - & entropy->dc_derived_tbls[tbl]); - if (cinfo->lim_Se) { /* AC needs no table when not present */ - tbl = compptr->ac_tbl_no; - jpeg_make_d_derived_tbl(cinfo, FALSE, tbl, - & entropy->ac_derived_tbls[tbl]); - } - /* Initialize DC predictions to 0 */ - entropy->saved.last_dc_val[ci] = 0; - } - - /* Precalculate decoding info for each block in an MCU of this scan */ - for (blkn = 0; blkn < cinfo->blocks_in_MCU; blkn++) { - ci = cinfo->MCU_membership[blkn]; - compptr = cinfo->cur_comp_info[ci]; - /* Precalculate which table to use for each block */ - entropy->dc_cur_tbls[blkn] = entropy->dc_derived_tbls[compptr->dc_tbl_no]; - entropy->ac_cur_tbls[blkn] = /* AC needs no table when not present */ - cinfo->lim_Se ? entropy->ac_derived_tbls[compptr->ac_tbl_no] : NULL; - /* Decide whether we really care about the coefficient values */ - if (compptr->component_needed) { - ci = compptr->DCT_v_scaled_size; - i = compptr->DCT_h_scaled_size; - switch (cinfo->lim_Se) { - case (1*1-1): - entropy->coef_limit[blkn] = 1; - break; - case (2*2-1): - if (ci <= 0 || ci > 2) ci = 2; - if (i <= 0 || i > 2) i = 2; - entropy->coef_limit[blkn] = 1 + jpeg_zigzag_order2[ci - 1][i - 1]; - break; - case (3*3-1): - if (ci <= 0 || ci > 3) ci = 3; - if (i <= 0 || i > 3) i = 3; - entropy->coef_limit[blkn] = 1 + jpeg_zigzag_order3[ci - 1][i - 1]; - break; - case (4*4-1): - if (ci <= 0 || ci > 4) ci = 4; - if (i <= 0 || i > 4) i = 4; - entropy->coef_limit[blkn] = 1 + jpeg_zigzag_order4[ci - 1][i - 1]; - break; - case (5*5-1): - if (ci <= 0 || ci > 5) ci = 5; - if (i <= 0 || i > 5) i = 5; - entropy->coef_limit[blkn] = 1 + jpeg_zigzag_order5[ci - 1][i - 1]; - break; - case (6*6-1): - if (ci <= 0 || ci > 6) ci = 6; - if (i <= 0 || i > 6) i = 6; - entropy->coef_limit[blkn] = 1 + jpeg_zigzag_order6[ci - 1][i - 1]; - break; - case (7*7-1): - if (ci <= 0 || ci > 7) ci = 7; - if (i <= 0 || i > 7) i = 7; - entropy->coef_limit[blkn] = 1 + jpeg_zigzag_order7[ci - 1][i - 1]; - break; - default: - if (ci <= 0 || ci > 8) ci = 8; - if (i <= 0 || i > 8) i = 8; - entropy->coef_limit[blkn] = 1 + jpeg_zigzag_order[ci - 1][i - 1]; - } - } else { - entropy->coef_limit[blkn] = 0; - } - } - } - - /* Initialize bitread state variables */ - entropy->bitstate.bits_left = 0; - entropy->bitstate.get_buffer = 0; /* unnecessary, but keeps Purify quiet */ - entropy->insufficient_data = FALSE; - - /* Initialize restart counter */ - entropy->restarts_to_go = cinfo->restart_interval; -} - - -/* - * Module initialization routine for Huffman entropy decoding. - */ - -GLOBAL(void) -jinit_huff_decoder (j_decompress_ptr cinfo) -{ - huff_entropy_ptr entropy; - int i; - - entropy = (huff_entropy_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(huff_entropy_decoder)); - cinfo->entropy = &entropy->pub; - entropy->pub.start_pass = start_pass_huff_decoder; - entropy->pub.finish_pass = finish_pass_huff; - - if (cinfo->progressive_mode) { - /* Create progression status table */ - int *coef_bit_ptr, ci; - cinfo->coef_bits = (int (*)[DCTSIZE2]) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - cinfo->num_components * DCTSIZE2 * SIZEOF(int)); - coef_bit_ptr = & cinfo->coef_bits[0][0]; - for (ci = 0; ci < cinfo->num_components; ci++) - for (i = 0; i < DCTSIZE2; i++) - *coef_bit_ptr++ = -1; - - /* Mark derived tables unallocated */ - for (i = 0; i < NUM_HUFF_TBLS; i++) { - entropy->derived_tbls[i] = NULL; - } - } else { - /* Mark derived tables unallocated */ - for (i = 0; i < NUM_HUFF_TBLS; i++) { - entropy->dc_derived_tbls[i] = entropy->ac_derived_tbls[i] = NULL; - } - } -} diff --git a/dep/libjpeg/src/jdinput.c b/dep/libjpeg/src/jdinput.c deleted file mode 100644 index 29fbef90b..000000000 --- a/dep/libjpeg/src/jdinput.c +++ /dev/null @@ -1,657 +0,0 @@ -/* - * jdinput.c - * - * Copyright (C) 1991-1997, Thomas G. Lane. - * Modified 2002-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains input control logic for the JPEG decompressor. - * These routines are concerned with controlling the decompressor's input - * processing (marker reading and coefficient decoding). The actual input - * reading is done in jdmarker.c, jdhuff.c, and jdarith.c. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Private state */ - -typedef struct { - struct jpeg_input_controller pub; /* public fields */ - - int inheaders; /* Nonzero until first SOS is reached */ -} my_input_controller; - -typedef my_input_controller * my_inputctl_ptr; - - -/* Forward declarations */ -METHODDEF(int) consume_markers JPP((j_decompress_ptr cinfo)); - - -/* - * Routines to calculate various quantities related to the size of the image. - */ - - -/* - * Compute output image dimensions and related values. - * NOTE: this is exported for possible use by application. - * Hence it mustn't do anything that can't be done twice. - */ - -GLOBAL(void) -jpeg_core_output_dimensions (j_decompress_ptr cinfo) -/* Do computations that are needed before master selection phase. - * This function is used for transcoding and full decompression. - */ -{ -#ifdef IDCT_SCALING_SUPPORTED - int ci; - jpeg_component_info *compptr; - - /* Compute actual output image dimensions and DCT scaling choices. */ - if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom) { - /* Provide 1/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 1; - cinfo->min_DCT_v_scaled_size = 1; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 2) { - /* Provide 2/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 2L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 2L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 2; - cinfo->min_DCT_v_scaled_size = 2; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 3) { - /* Provide 3/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 3L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 3L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 3; - cinfo->min_DCT_v_scaled_size = 3; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 4) { - /* Provide 4/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 4L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 4L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 4; - cinfo->min_DCT_v_scaled_size = 4; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 5) { - /* Provide 5/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 5L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 5L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 5; - cinfo->min_DCT_v_scaled_size = 5; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 6) { - /* Provide 6/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 6L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 6L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 6; - cinfo->min_DCT_v_scaled_size = 6; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 7) { - /* Provide 7/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 7L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 7L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 7; - cinfo->min_DCT_v_scaled_size = 7; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 8) { - /* Provide 8/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 8L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 8L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 8; - cinfo->min_DCT_v_scaled_size = 8; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 9) { - /* Provide 9/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 9L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 9L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 9; - cinfo->min_DCT_v_scaled_size = 9; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 10) { - /* Provide 10/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 10L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 10L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 10; - cinfo->min_DCT_v_scaled_size = 10; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 11) { - /* Provide 11/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 11L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 11L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 11; - cinfo->min_DCT_v_scaled_size = 11; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 12) { - /* Provide 12/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 12L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 12L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 12; - cinfo->min_DCT_v_scaled_size = 12; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 13) { - /* Provide 13/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 13L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 13L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 13; - cinfo->min_DCT_v_scaled_size = 13; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 14) { - /* Provide 14/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 14L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 14L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 14; - cinfo->min_DCT_v_scaled_size = 14; - } else if (cinfo->scale_num * cinfo->block_size <= cinfo->scale_denom * 15) { - /* Provide 15/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 15L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 15L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 15; - cinfo->min_DCT_v_scaled_size = 15; - } else { - /* Provide 16/block_size scaling */ - cinfo->output_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * 16L, (long) cinfo->block_size); - cinfo->output_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * 16L, (long) cinfo->block_size); - cinfo->min_DCT_h_scaled_size = 16; - cinfo->min_DCT_v_scaled_size = 16; - } - - /* Recompute dimensions of components */ - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - compptr->DCT_h_scaled_size = cinfo->min_DCT_h_scaled_size; - compptr->DCT_v_scaled_size = cinfo->min_DCT_v_scaled_size; - } - -#else /* !IDCT_SCALING_SUPPORTED */ - - /* Hardwire it to "no scaling" */ - cinfo->output_width = cinfo->image_width; - cinfo->output_height = cinfo->image_height; - /* initial_setup has already initialized DCT_scaled_size, - * and has computed unscaled downsampled_width and downsampled_height. - */ - -#endif /* IDCT_SCALING_SUPPORTED */ -} - - -LOCAL(void) -initial_setup (j_decompress_ptr cinfo) -/* Called once, when first SOS marker is reached */ -{ - int ci; - jpeg_component_info *compptr; - - /* Make sure image isn't bigger than I can handle */ - if ((long) cinfo->image_height > (long) JPEG_MAX_DIMENSION || - (long) cinfo->image_width > (long) JPEG_MAX_DIMENSION) - ERREXIT1(cinfo, JERR_IMAGE_TOO_BIG, (unsigned int) JPEG_MAX_DIMENSION); - - /* Only 8 to 12 bits data precision are supported for DCT based JPEG */ - if (cinfo->data_precision < 8 || cinfo->data_precision > 12) - ERREXIT1(cinfo, JERR_BAD_PRECISION, cinfo->data_precision); - - /* Check that number of components won't exceed internal array sizes */ - if (cinfo->num_components > MAX_COMPONENTS) - ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->num_components, - MAX_COMPONENTS); - - /* Compute maximum sampling factors; check factor validity */ - cinfo->max_h_samp_factor = 1; - cinfo->max_v_samp_factor = 1; - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - if (compptr->h_samp_factor<=0 || compptr->h_samp_factor>MAX_SAMP_FACTOR || - compptr->v_samp_factor<=0 || compptr->v_samp_factor>MAX_SAMP_FACTOR) - ERREXIT(cinfo, JERR_BAD_SAMPLING); - cinfo->max_h_samp_factor = MAX(cinfo->max_h_samp_factor, - compptr->h_samp_factor); - cinfo->max_v_samp_factor = MAX(cinfo->max_v_samp_factor, - compptr->v_samp_factor); - } - - /* Derive block_size, natural_order, and lim_Se */ - if (cinfo->is_baseline || (cinfo->progressive_mode && - cinfo->comps_in_scan)) { /* no pseudo SOS marker */ - cinfo->block_size = DCTSIZE; - cinfo->natural_order = jpeg_natural_order; - cinfo->lim_Se = DCTSIZE2-1; - } else - switch (cinfo->Se) { - case (1*1-1): - cinfo->block_size = 1; - cinfo->natural_order = jpeg_natural_order; /* not needed */ - cinfo->lim_Se = cinfo->Se; - break; - case (2*2-1): - cinfo->block_size = 2; - cinfo->natural_order = jpeg_natural_order2; - cinfo->lim_Se = cinfo->Se; - break; - case (3*3-1): - cinfo->block_size = 3; - cinfo->natural_order = jpeg_natural_order3; - cinfo->lim_Se = cinfo->Se; - break; - case (4*4-1): - cinfo->block_size = 4; - cinfo->natural_order = jpeg_natural_order4; - cinfo->lim_Se = cinfo->Se; - break; - case (5*5-1): - cinfo->block_size = 5; - cinfo->natural_order = jpeg_natural_order5; - cinfo->lim_Se = cinfo->Se; - break; - case (6*6-1): - cinfo->block_size = 6; - cinfo->natural_order = jpeg_natural_order6; - cinfo->lim_Se = cinfo->Se; - break; - case (7*7-1): - cinfo->block_size = 7; - cinfo->natural_order = jpeg_natural_order7; - cinfo->lim_Se = cinfo->Se; - break; - case (8*8-1): - cinfo->block_size = 8; - cinfo->natural_order = jpeg_natural_order; - cinfo->lim_Se = DCTSIZE2-1; - break; - case (9*9-1): - cinfo->block_size = 9; - cinfo->natural_order = jpeg_natural_order; - cinfo->lim_Se = DCTSIZE2-1; - break; - case (10*10-1): - cinfo->block_size = 10; - cinfo->natural_order = jpeg_natural_order; - cinfo->lim_Se = DCTSIZE2-1; - break; - case (11*11-1): - cinfo->block_size = 11; - cinfo->natural_order = jpeg_natural_order; - cinfo->lim_Se = DCTSIZE2-1; - break; - case (12*12-1): - cinfo->block_size = 12; - cinfo->natural_order = jpeg_natural_order; - cinfo->lim_Se = DCTSIZE2-1; - break; - case (13*13-1): - cinfo->block_size = 13; - cinfo->natural_order = jpeg_natural_order; - cinfo->lim_Se = DCTSIZE2-1; - break; - case (14*14-1): - cinfo->block_size = 14; - cinfo->natural_order = jpeg_natural_order; - cinfo->lim_Se = DCTSIZE2-1; - break; - case (15*15-1): - cinfo->block_size = 15; - cinfo->natural_order = jpeg_natural_order; - cinfo->lim_Se = DCTSIZE2-1; - break; - case (16*16-1): - cinfo->block_size = 16; - cinfo->natural_order = jpeg_natural_order; - cinfo->lim_Se = DCTSIZE2-1; - break; - default: - ERREXIT4(cinfo, JERR_BAD_PROGRESSION, - cinfo->Ss, cinfo->Se, cinfo->Ah, cinfo->Al); - } - - /* We initialize DCT_scaled_size and min_DCT_scaled_size to block_size. - * In the full decompressor, - * this will be overridden by jpeg_calc_output_dimensions in jdmaster.c; - * but in the transcoder, - * jpeg_calc_output_dimensions is not used, so we must do it here. - */ - cinfo->min_DCT_h_scaled_size = cinfo->block_size; - cinfo->min_DCT_v_scaled_size = cinfo->block_size; - - /* Compute dimensions of components */ - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - compptr->DCT_h_scaled_size = cinfo->block_size; - compptr->DCT_v_scaled_size = cinfo->block_size; - /* Size in DCT blocks */ - compptr->width_in_blocks = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * (long) compptr->h_samp_factor, - (long) (cinfo->max_h_samp_factor * cinfo->block_size)); - compptr->height_in_blocks = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * (long) compptr->v_samp_factor, - (long) (cinfo->max_v_samp_factor * cinfo->block_size)); - /* downsampled_width and downsampled_height will also be overridden by - * jdmaster.c if we are doing full decompression. The transcoder library - * doesn't use these values, but the calling application might. - */ - /* Size in samples */ - compptr->downsampled_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * (long) compptr->h_samp_factor, - (long) cinfo->max_h_samp_factor); - compptr->downsampled_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * (long) compptr->v_samp_factor, - (long) cinfo->max_v_samp_factor); - /* Mark component needed, until color conversion says otherwise */ - compptr->component_needed = TRUE; - /* Mark no quantization table yet saved for component */ - compptr->quant_table = NULL; - } - - /* Compute number of fully interleaved MCU rows. */ - cinfo->total_iMCU_rows = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height, - (long) (cinfo->max_v_samp_factor * cinfo->block_size)); - - /* Decide whether file contains multiple scans */ - if (cinfo->comps_in_scan < cinfo->num_components || cinfo->progressive_mode) - cinfo->inputctl->has_multiple_scans = TRUE; - else - cinfo->inputctl->has_multiple_scans = FALSE; -} - - -LOCAL(void) -per_scan_setup (j_decompress_ptr cinfo) -/* Do computations that are needed before processing a JPEG scan */ -/* cinfo->comps_in_scan and cinfo->cur_comp_info[] were set from SOS marker */ -{ - int ci, mcublks, tmp; - jpeg_component_info *compptr; - - if (cinfo->comps_in_scan == 1) { - - /* Noninterleaved (single-component) scan */ - compptr = cinfo->cur_comp_info[0]; - - /* Overall image size in MCUs */ - cinfo->MCUs_per_row = compptr->width_in_blocks; - cinfo->MCU_rows_in_scan = compptr->height_in_blocks; - - /* For noninterleaved scan, always one block per MCU */ - compptr->MCU_width = 1; - compptr->MCU_height = 1; - compptr->MCU_blocks = 1; - compptr->MCU_sample_width = compptr->DCT_h_scaled_size; - compptr->last_col_width = 1; - /* For noninterleaved scans, it is convenient to define last_row_height - * as the number of block rows present in the last iMCU row. - */ - tmp = (int) (compptr->height_in_blocks % compptr->v_samp_factor); - if (tmp == 0) tmp = compptr->v_samp_factor; - compptr->last_row_height = tmp; - - /* Prepare array describing MCU composition */ - cinfo->blocks_in_MCU = 1; - cinfo->MCU_membership[0] = 0; - - } else { - - /* Interleaved (multi-component) scan */ - if (cinfo->comps_in_scan <= 0 || cinfo->comps_in_scan > MAX_COMPS_IN_SCAN) - ERREXIT2(cinfo, JERR_COMPONENT_COUNT, cinfo->comps_in_scan, - MAX_COMPS_IN_SCAN); - - /* Overall image size in MCUs */ - cinfo->MCUs_per_row = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width, - (long) (cinfo->max_h_samp_factor * cinfo->block_size)); - cinfo->MCU_rows_in_scan = cinfo->total_iMCU_rows; - - cinfo->blocks_in_MCU = 0; - - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - /* Sampling factors give # of blocks of component in each MCU */ - compptr->MCU_width = compptr->h_samp_factor; - compptr->MCU_height = compptr->v_samp_factor; - compptr->MCU_blocks = compptr->MCU_width * compptr->MCU_height; - compptr->MCU_sample_width = compptr->MCU_width * compptr->DCT_h_scaled_size; - /* Figure number of non-dummy blocks in last MCU column & row */ - tmp = (int) (compptr->width_in_blocks % compptr->MCU_width); - if (tmp == 0) tmp = compptr->MCU_width; - compptr->last_col_width = tmp; - tmp = (int) (compptr->height_in_blocks % compptr->MCU_height); - if (tmp == 0) tmp = compptr->MCU_height; - compptr->last_row_height = tmp; - /* Prepare array describing MCU composition */ - mcublks = compptr->MCU_blocks; - if (cinfo->blocks_in_MCU + mcublks > D_MAX_BLOCKS_IN_MCU) - ERREXIT(cinfo, JERR_BAD_MCU_SIZE); - while (mcublks-- > 0) { - cinfo->MCU_membership[cinfo->blocks_in_MCU++] = ci; - } - } - - } -} - - -/* - * Save away a copy of the Q-table referenced by each component present - * in the current scan, unless already saved during a prior scan. - * - * In a multiple-scan JPEG file, the encoder could assign different components - * the same Q-table slot number, but change table definitions between scans - * so that each component uses a different Q-table. (The IJG encoder is not - * currently capable of doing this, but other encoders might.) Since we want - * to be able to dequantize all the components at the end of the file, this - * means that we have to save away the table actually used for each component. - * We do this by copying the table at the start of the first scan containing - * the component. - * The JPEG spec prohibits the encoder from changing the contents of a Q-table - * slot between scans of a component using that slot. If the encoder does so - * anyway, this decoder will simply use the Q-table values that were current - * at the start of the first scan for the component. - * - * The decompressor output side looks only at the saved quant tables, - * not at the current Q-table slots. - */ - -LOCAL(void) -latch_quant_tables (j_decompress_ptr cinfo) -{ - int ci, qtblno; - jpeg_component_info *compptr; - JQUANT_TBL * qtbl; - - for (ci = 0; ci < cinfo->comps_in_scan; ci++) { - compptr = cinfo->cur_comp_info[ci]; - /* No work if we already saved Q-table for this component */ - if (compptr->quant_table != NULL) - continue; - /* Make sure specified quantization table is present */ - qtblno = compptr->quant_tbl_no; - if (qtblno < 0 || qtblno >= NUM_QUANT_TBLS || - cinfo->quant_tbl_ptrs[qtblno] == NULL) - ERREXIT1(cinfo, JERR_NO_QUANT_TABLE, qtblno); - /* OK, save away the quantization table */ - qtbl = (JQUANT_TBL *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(JQUANT_TBL)); - MEMCOPY(qtbl, cinfo->quant_tbl_ptrs[qtblno], SIZEOF(JQUANT_TBL)); - compptr->quant_table = qtbl; - } -} - - -/* - * Initialize the input modules to read a scan of compressed data. - * The first call to this is done by jdmaster.c after initializing - * the entire decompressor (during jpeg_start_decompress). - * Subsequent calls come from consume_markers, below. - */ - -METHODDEF(void) -start_input_pass (j_decompress_ptr cinfo) -{ - per_scan_setup(cinfo); - latch_quant_tables(cinfo); - (*cinfo->entropy->start_pass) (cinfo); - (*cinfo->coef->start_input_pass) (cinfo); - cinfo->inputctl->consume_input = cinfo->coef->consume_data; -} - - -/* - * Finish up after inputting a compressed-data scan. - * This is called by the coefficient controller after it's read all - * the expected data of the scan. - */ - -METHODDEF(void) -finish_input_pass (j_decompress_ptr cinfo) -{ - (*cinfo->entropy->finish_pass) (cinfo); - cinfo->inputctl->consume_input = consume_markers; -} - - -/* - * Read JPEG markers before, between, or after compressed-data scans. - * Change state as necessary when a new scan is reached. - * Return value is JPEG_SUSPENDED, JPEG_REACHED_SOS, or JPEG_REACHED_EOI. - * - * The consume_input method pointer points either here or to the - * coefficient controller's consume_data routine, depending on whether - * we are reading a compressed data segment or inter-segment markers. - * - * Note: This function should NOT return a pseudo SOS marker (with zero - * component number) to the caller. A pseudo marker received by - * read_markers is processed and then skipped for other markers. - */ - -METHODDEF(int) -consume_markers (j_decompress_ptr cinfo) -{ - my_inputctl_ptr inputctl = (my_inputctl_ptr) cinfo->inputctl; - int val; - - if (inputctl->pub.eoi_reached) /* After hitting EOI, read no further */ - return JPEG_REACHED_EOI; - - for (;;) { /* Loop to pass pseudo SOS marker */ - val = (*cinfo->marker->read_markers) (cinfo); - - switch (val) { - case JPEG_REACHED_SOS: /* Found SOS */ - if (inputctl->inheaders) { /* 1st SOS */ - if (inputctl->inheaders == 1) - initial_setup(cinfo); - if (cinfo->comps_in_scan == 0) { /* pseudo SOS marker */ - inputctl->inheaders = 2; - break; - } - inputctl->inheaders = 0; - /* Note: start_input_pass must be called by jdmaster.c - * before any more input can be consumed. jdapimin.c is - * responsible for enforcing this sequencing. - */ - } else { /* 2nd or later SOS marker */ - if (! inputctl->pub.has_multiple_scans) - ERREXIT(cinfo, JERR_EOI_EXPECTED); /* Oops, I wasn't expecting this! */ - if (cinfo->comps_in_scan == 0) /* unexpected pseudo SOS marker */ - break; - start_input_pass(cinfo); - } - return val; - case JPEG_REACHED_EOI: /* Found EOI */ - inputctl->pub.eoi_reached = TRUE; - if (inputctl->inheaders) { /* Tables-only datastream, apparently */ - if (cinfo->marker->saw_SOF) - ERREXIT(cinfo, JERR_SOF_NO_SOS); - } else { - /* Prevent infinite loop in coef ctlr's decompress_data routine - * if user set output_scan_number larger than number of scans. - */ - if (cinfo->output_scan_number > cinfo->input_scan_number) - cinfo->output_scan_number = cinfo->input_scan_number; - } - return val; - case JPEG_SUSPENDED: - return val; - default: - return val; - } - } -} - - -/* - * Reset state to begin a fresh datastream. - */ - -METHODDEF(void) -reset_input_controller (j_decompress_ptr cinfo) -{ - my_inputctl_ptr inputctl = (my_inputctl_ptr) cinfo->inputctl; - - inputctl->pub.consume_input = consume_markers; - inputctl->pub.has_multiple_scans = FALSE; /* "unknown" would be better */ - inputctl->pub.eoi_reached = FALSE; - inputctl->inheaders = 1; - /* Reset other modules */ - (*cinfo->err->reset_error_mgr) ((j_common_ptr) cinfo); - (*cinfo->marker->reset_marker_reader) (cinfo); - /* Reset progression state -- would be cleaner if entropy decoder did this */ - cinfo->coef_bits = NULL; -} - - -/* - * Initialize the input controller module. - * This is called only once, when the decompression object is created. - */ - -GLOBAL(void) -jinit_input_controller (j_decompress_ptr cinfo) -{ - my_inputctl_ptr inputctl; - - /* Create subobject in permanent pool */ - inputctl = (my_inputctl_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_PERMANENT, SIZEOF(my_input_controller)); - cinfo->inputctl = &inputctl->pub; - /* Initialize method pointers */ - inputctl->pub.consume_input = consume_markers; - inputctl->pub.reset_input_controller = reset_input_controller; - inputctl->pub.start_input_pass = start_input_pass; - inputctl->pub.finish_input_pass = finish_input_pass; - /* Initialize state: can't use reset_input_controller since we don't - * want to try to reset other modules yet. - */ - inputctl->pub.has_multiple_scans = FALSE; /* "unknown" would be better */ - inputctl->pub.eoi_reached = FALSE; - inputctl->inheaders = 1; -} diff --git a/dep/libjpeg/src/jdmainct.c b/dep/libjpeg/src/jdmainct.c deleted file mode 100644 index 1cd66d853..000000000 --- a/dep/libjpeg/src/jdmainct.c +++ /dev/null @@ -1,511 +0,0 @@ -/* - * jdmainct.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2002-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains the main buffer controller for decompression. - * The main buffer lies between the JPEG decompressor proper and the - * post-processor; it holds downsampled data in the JPEG colorspace. - * - * Note that this code is bypassed in raw-data mode, since the application - * supplies the equivalent of the main buffer in that case. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* - * In the current system design, the main buffer need never be a full-image - * buffer; any full-height buffers will be found inside the coefficient or - * postprocessing controllers. Nonetheless, the main controller is not - * trivial. Its responsibility is to provide context rows for upsampling/ - * rescaling, and doing this in an efficient fashion is a bit tricky. - * - * Postprocessor input data is counted in "row groups". A row group is - * defined to be (v_samp_factor * DCT_v_scaled_size / min_DCT_v_scaled_size) - * sample rows of each component. (We require DCT_scaled_size values to be - * chosen such that these numbers are integers. In practice DCT_scaled_size - * values will likely be powers of two, so we actually have the stronger - * condition that DCT_scaled_size / min_DCT_scaled_size is an integer.) - * Upsampling will typically produce max_v_samp_factor pixel rows from each - * row group (times any additional scale factor that the upsampler is - * applying). - * - * The coefficient controller will deliver data to us one iMCU row at a time; - * each iMCU row contains v_samp_factor * DCT_v_scaled_size sample rows, or - * exactly min_DCT_v_scaled_size row groups. (This amount of data corresponds - * to one row of MCUs when the image is fully interleaved.) Note that the - * number of sample rows varies across components, but the number of row - * groups does not. Some garbage sample rows may be included in the last iMCU - * row at the bottom of the image. - * - * Depending on the vertical scaling algorithm used, the upsampler may need - * access to the sample row(s) above and below its current input row group. - * The upsampler is required to set need_context_rows TRUE at global selection - * time if so. When need_context_rows is FALSE, this controller can simply - * obtain one iMCU row at a time from the coefficient controller and dole it - * out as row groups to the postprocessor. - * - * When need_context_rows is TRUE, this controller guarantees that the buffer - * passed to postprocessing contains at least one row group's worth of samples - * above and below the row group(s) being processed. Note that the context - * rows "above" the first passed row group appear at negative row offsets in - * the passed buffer. At the top and bottom of the image, the required - * context rows are manufactured by duplicating the first or last real sample - * row; this avoids having special cases in the upsampling inner loops. - * - * The amount of context is fixed at one row group just because that's a - * convenient number for this controller to work with. The existing - * upsamplers really only need one sample row of context. An upsampler - * supporting arbitrary output rescaling might wish for more than one row - * group of context when shrinking the image; tough, we don't handle that. - * (This is justified by the assumption that downsizing will be handled mostly - * by adjusting the DCT_scaled_size values, so that the actual scale factor at - * the upsample step needn't be much less than one.) - * - * To provide the desired context, we have to retain the last two row groups - * of one iMCU row while reading in the next iMCU row. (The last row group - * can't be processed until we have another row group for its below-context, - * and so we have to save the next-to-last group too for its above-context.) - * We could do this most simply by copying data around in our buffer, but - * that'd be very slow. We can avoid copying any data by creating a rather - * strange pointer structure. Here's how it works. We allocate a workspace - * consisting of M+2 row groups (where M = min_DCT_v_scaled_size is the number - * of row groups per iMCU row). We create two sets of redundant pointers to - * the workspace. Labeling the physical row groups 0 to M+1, the synthesized - * pointer lists look like this: - * M+1 M-1 - * master pointer --> 0 master pointer --> 0 - * 1 1 - * ... ... - * M-3 M-3 - * M-2 M - * M-1 M+1 - * M M-2 - * M+1 M-1 - * 0 0 - * We read alternate iMCU rows using each master pointer; thus the last two - * row groups of the previous iMCU row remain un-overwritten in the workspace. - * The pointer lists are set up so that the required context rows appear to - * be adjacent to the proper places when we pass the pointer lists to the - * upsampler. - * - * The above pictures describe the normal state of the pointer lists. - * At top and bottom of the image, we diddle the pointer lists to duplicate - * the first or last sample row as necessary (this is cheaper than copying - * sample rows around). - * - * This scheme breaks down if M < 2, ie, min_DCT_v_scaled_size is 1. In that - * situation each iMCU row provides only one row group so the buffering logic - * must be different (eg, we must read two iMCU rows before we can emit the - * first row group). For now, we simply do not support providing context - * rows when min_DCT_v_scaled_size is 1. That combination seems unlikely to - * be worth providing --- if someone wants a 1/8th-size preview, they probably - * want it quick and dirty, so a context-free upsampler is sufficient. - */ - - -/* Private buffer controller object */ - -typedef struct { - struct jpeg_d_main_controller pub; /* public fields */ - - /* Pointer to allocated workspace (M or M+2 row groups). */ - JSAMPARRAY buffer[MAX_COMPONENTS]; - - JDIMENSION rowgroup_ctr; /* counts row groups output to postprocessor */ - JDIMENSION rowgroups_avail; /* row groups available to postprocessor */ - - /* Remaining fields are only used in the context case. */ - - boolean buffer_full; /* Have we gotten an iMCU row from decoder? */ - - /* These are the master pointers to the funny-order pointer lists. */ - JSAMPIMAGE xbuffer[2]; /* pointers to weird pointer lists */ - - int whichptr; /* indicates which pointer set is now in use */ - int context_state; /* process_data state machine status */ - JDIMENSION iMCU_row_ctr; /* counts iMCU rows to detect image top/bot */ -} my_main_controller; - -typedef my_main_controller * my_main_ptr; - -/* context_state values: */ -#define CTX_PREPARE_FOR_IMCU 0 /* need to prepare for MCU row */ -#define CTX_PROCESS_IMCU 1 /* feeding iMCU to postprocessor */ -#define CTX_POSTPONED_ROW 2 /* feeding postponed row group */ - - -/* Forward declarations */ -METHODDEF(void) process_data_simple_main - JPP((j_decompress_ptr cinfo, JSAMPARRAY output_buf, - JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail)); -METHODDEF(void) process_data_context_main - JPP((j_decompress_ptr cinfo, JSAMPARRAY output_buf, - JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail)); -#ifdef QUANT_2PASS_SUPPORTED -METHODDEF(void) process_data_crank_post - JPP((j_decompress_ptr cinfo, JSAMPARRAY output_buf, - JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail)); -#endif - - -LOCAL(void) -alloc_funny_pointers (j_decompress_ptr cinfo) -/* Allocate space for the funny pointer lists. - * This is done only once, not once per pass. - */ -{ - my_main_ptr mainp = (my_main_ptr) cinfo->main; - int ci, rgroup; - int M = cinfo->min_DCT_v_scaled_size; - jpeg_component_info *compptr; - JSAMPARRAY xbuf; - - /* Get top-level space for component array pointers. - * We alloc both arrays with one call to save a few cycles. - */ - mainp->xbuffer[0] = (JSAMPIMAGE) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - cinfo->num_components * 2 * SIZEOF(JSAMPARRAY)); - mainp->xbuffer[1] = mainp->xbuffer[0] + cinfo->num_components; - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - if (! compptr->component_needed) - continue; /* skip uninteresting component */ - rgroup = (compptr->v_samp_factor * compptr->DCT_v_scaled_size) / - cinfo->min_DCT_v_scaled_size; /* height of a row group of component */ - /* Get space for pointer lists --- M+4 row groups in each list. - * We alloc both pointer lists with one call to save a few cycles. - */ - xbuf = (JSAMPARRAY) (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, - JPOOL_IMAGE, 2 * (rgroup * (M + 4)) * SIZEOF(JSAMPROW)); - xbuf += rgroup; /* want one row group at negative offsets */ - mainp->xbuffer[0][ci] = xbuf; - xbuf += rgroup * (M + 4); - mainp->xbuffer[1][ci] = xbuf; - } -} - - -LOCAL(void) -make_funny_pointers (j_decompress_ptr cinfo) -/* Create the funny pointer lists discussed in the comments above. - * The actual workspace is already allocated (in mainp->buffer), - * and the space for the pointer lists is allocated too. - * This routine just fills in the curiously ordered lists. - * This will be repeated at the beginning of each pass. - */ -{ - my_main_ptr mainp = (my_main_ptr) cinfo->main; - int ci, i, rgroup; - int M = cinfo->min_DCT_v_scaled_size; - jpeg_component_info *compptr; - JSAMPARRAY buf, xbuf0, xbuf1; - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - if (! compptr->component_needed) - continue; /* skip uninteresting component */ - rgroup = (compptr->v_samp_factor * compptr->DCT_v_scaled_size) / - cinfo->min_DCT_v_scaled_size; /* height of a row group of component */ - xbuf0 = mainp->xbuffer[0][ci]; - xbuf1 = mainp->xbuffer[1][ci]; - /* First copy the workspace pointers as-is */ - buf = mainp->buffer[ci]; - for (i = 0; i < rgroup * (M + 2); i++) { - xbuf0[i] = xbuf1[i] = buf[i]; - } - /* In the second list, put the last four row groups in swapped order */ - for (i = 0; i < rgroup * 2; i++) { - xbuf1[rgroup*(M-2) + i] = buf[rgroup*M + i]; - xbuf1[rgroup*M + i] = buf[rgroup*(M-2) + i]; - } - /* The wraparound pointers at top and bottom will be filled later - * (see set_wraparound_pointers, below). Initially we want the "above" - * pointers to duplicate the first actual data line. This only needs - * to happen in xbuffer[0]. - */ - for (i = 0; i < rgroup; i++) { - xbuf0[i - rgroup] = xbuf0[0]; - } - } -} - - -LOCAL(void) -set_wraparound_pointers (j_decompress_ptr cinfo) -/* Set up the "wraparound" pointers at top and bottom of the pointer lists. - * This changes the pointer list state from top-of-image to the normal state. - */ -{ - my_main_ptr mainp = (my_main_ptr) cinfo->main; - int ci, i, rgroup; - int M = cinfo->min_DCT_v_scaled_size; - jpeg_component_info *compptr; - JSAMPARRAY xbuf0, xbuf1; - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - if (! compptr->component_needed) - continue; /* skip uninteresting component */ - rgroup = (compptr->v_samp_factor * compptr->DCT_v_scaled_size) / - cinfo->min_DCT_v_scaled_size; /* height of a row group of component */ - xbuf0 = mainp->xbuffer[0][ci]; - xbuf1 = mainp->xbuffer[1][ci]; - for (i = 0; i < rgroup; i++) { - xbuf0[i - rgroup] = xbuf0[rgroup*(M+1) + i]; - xbuf1[i - rgroup] = xbuf1[rgroup*(M+1) + i]; - xbuf0[rgroup*(M+2) + i] = xbuf0[i]; - xbuf1[rgroup*(M+2) + i] = xbuf1[i]; - } - } -} - - -LOCAL(void) -set_bottom_pointers (j_decompress_ptr cinfo) -/* Change the pointer lists to duplicate the last sample row at the bottom - * of the image. whichptr indicates which xbuffer holds the final iMCU row. - * Also sets rowgroups_avail to indicate number of nondummy row groups in row. - */ -{ - my_main_ptr mainp = (my_main_ptr) cinfo->main; - int ci, i, rgroup, iMCUheight, rows_left; - jpeg_component_info *compptr; - JSAMPARRAY xbuf; - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - if (! compptr->component_needed) - continue; /* skip uninteresting component */ - /* Count sample rows in one iMCU row and in one row group */ - iMCUheight = compptr->v_samp_factor * compptr->DCT_v_scaled_size; - rgroup = iMCUheight / cinfo->min_DCT_v_scaled_size; - /* Count nondummy sample rows remaining for this component */ - rows_left = (int) (compptr->downsampled_height % (JDIMENSION) iMCUheight); - if (rows_left == 0) rows_left = iMCUheight; - /* Count nondummy row groups. Should get same answer for each component, - * so we need only do it once. - */ - if (ci == 0) { - mainp->rowgroups_avail = (JDIMENSION) ((rows_left-1) / rgroup + 1); - } - /* Duplicate the last real sample row rgroup*2 times; this pads out the - * last partial rowgroup and ensures at least one full rowgroup of context. - */ - xbuf = mainp->xbuffer[mainp->whichptr][ci]; - for (i = 0; i < rgroup * 2; i++) { - xbuf[rows_left + i] = xbuf[rows_left-1]; - } - } -} - - -/* - * Initialize for a processing pass. - */ - -METHODDEF(void) -start_pass_main (j_decompress_ptr cinfo, J_BUF_MODE pass_mode) -{ - my_main_ptr mainp = (my_main_ptr) cinfo->main; - - switch (pass_mode) { - case JBUF_PASS_THRU: - if (cinfo->upsample->need_context_rows) { - mainp->pub.process_data = process_data_context_main; - make_funny_pointers(cinfo); /* Create the xbuffer[] lists */ - mainp->whichptr = 0; /* Read first iMCU row into xbuffer[0] */ - mainp->context_state = CTX_PREPARE_FOR_IMCU; - mainp->iMCU_row_ctr = 0; - mainp->buffer_full = FALSE; /* Mark buffer empty */ - } else { - /* Simple case with no context needed */ - mainp->pub.process_data = process_data_simple_main; - mainp->rowgroup_ctr = mainp->rowgroups_avail; /* Mark buffer empty */ - } - break; -#ifdef QUANT_2PASS_SUPPORTED - case JBUF_CRANK_DEST: - /* For last pass of 2-pass quantization, just crank the postprocessor */ - mainp->pub.process_data = process_data_crank_post; - break; -#endif - default: - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - } -} - - -/* - * Process some data. - * This handles the simple case where no context is required. - */ - -METHODDEF(void) -process_data_simple_main (j_decompress_ptr cinfo, JSAMPARRAY output_buf, - JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail) -{ - my_main_ptr mainp = (my_main_ptr) cinfo->main; - - /* Read input data if we haven't filled the main buffer yet */ - if (mainp->rowgroup_ctr >= mainp->rowgroups_avail) { - if (! (*cinfo->coef->decompress_data) (cinfo, mainp->buffer)) - return; /* suspension forced, can do nothing more */ - mainp->rowgroup_ctr = 0; /* OK, we have an iMCU row to work with */ - } - - /* Note: at the bottom of the image, we may pass extra garbage row groups - * to the postprocessor. The postprocessor has to check for bottom - * of image anyway (at row resolution), so no point in us doing it too. - */ - - /* Feed the postprocessor */ - (*cinfo->post->post_process_data) (cinfo, mainp->buffer, - &mainp->rowgroup_ctr, mainp->rowgroups_avail, - output_buf, out_row_ctr, out_rows_avail); -} - - -/* - * Process some data. - * This handles the case where context rows must be provided. - */ - -METHODDEF(void) -process_data_context_main (j_decompress_ptr cinfo, JSAMPARRAY output_buf, - JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail) -{ - my_main_ptr mainp = (my_main_ptr) cinfo->main; - - /* Read input data if we haven't filled the main buffer yet */ - if (! mainp->buffer_full) { - if (! (*cinfo->coef->decompress_data) (cinfo, - mainp->xbuffer[mainp->whichptr])) - return; /* suspension forced, can do nothing more */ - mainp->buffer_full = TRUE; /* OK, we have an iMCU row to work with */ - mainp->iMCU_row_ctr++; /* count rows received */ - } - - /* Postprocessor typically will not swallow all the input data it is handed - * in one call (due to filling the output buffer first). Must be prepared - * to exit and restart. This switch lets us keep track of how far we got. - * Note that each case falls through to the next on successful completion. - */ - switch (mainp->context_state) { - case CTX_POSTPONED_ROW: - /* Call postprocessor using previously set pointers for postponed row */ - (*cinfo->post->post_process_data) (cinfo, mainp->xbuffer[mainp->whichptr], - &mainp->rowgroup_ctr, mainp->rowgroups_avail, - output_buf, out_row_ctr, out_rows_avail); - if (mainp->rowgroup_ctr < mainp->rowgroups_avail) - return; /* Need to suspend */ - mainp->context_state = CTX_PREPARE_FOR_IMCU; - if (*out_row_ctr >= out_rows_avail) - return; /* Postprocessor exactly filled output buf */ - /*FALLTHROUGH*/ - case CTX_PREPARE_FOR_IMCU: - /* Prepare to process first M-1 row groups of this iMCU row */ - mainp->rowgroup_ctr = 0; - mainp->rowgroups_avail = (JDIMENSION) (cinfo->min_DCT_v_scaled_size - 1); - /* Check for bottom of image: if so, tweak pointers to "duplicate" - * the last sample row, and adjust rowgroups_avail to ignore padding rows. - */ - if (mainp->iMCU_row_ctr == cinfo->total_iMCU_rows) - set_bottom_pointers(cinfo); - mainp->context_state = CTX_PROCESS_IMCU; - /*FALLTHROUGH*/ - case CTX_PROCESS_IMCU: - /* Call postprocessor using previously set pointers */ - (*cinfo->post->post_process_data) (cinfo, mainp->xbuffer[mainp->whichptr], - &mainp->rowgroup_ctr, mainp->rowgroups_avail, - output_buf, out_row_ctr, out_rows_avail); - if (mainp->rowgroup_ctr < mainp->rowgroups_avail) - return; /* Need to suspend */ - /* After the first iMCU, change wraparound pointers to normal state */ - if (mainp->iMCU_row_ctr == 1) - set_wraparound_pointers(cinfo); - /* Prepare to load new iMCU row using other xbuffer list */ - mainp->whichptr ^= 1; /* 0=>1 or 1=>0 */ - mainp->buffer_full = FALSE; - /* Still need to process last row group of this iMCU row, */ - /* which is saved at index M+1 of the other xbuffer */ - mainp->rowgroup_ctr = (JDIMENSION) (cinfo->min_DCT_v_scaled_size + 1); - mainp->rowgroups_avail = (JDIMENSION) (cinfo->min_DCT_v_scaled_size + 2); - mainp->context_state = CTX_POSTPONED_ROW; - } -} - - -/* - * Process some data. - * Final pass of two-pass quantization: just call the postprocessor. - * Source data will be the postprocessor controller's internal buffer. - */ - -#ifdef QUANT_2PASS_SUPPORTED - -METHODDEF(void) -process_data_crank_post (j_decompress_ptr cinfo, JSAMPARRAY output_buf, - JDIMENSION *out_row_ctr, JDIMENSION out_rows_avail) -{ - (*cinfo->post->post_process_data) (cinfo, (JSAMPIMAGE) NULL, - (JDIMENSION *) NULL, (JDIMENSION) 0, - output_buf, out_row_ctr, out_rows_avail); -} - -#endif /* QUANT_2PASS_SUPPORTED */ - - -/* - * Initialize main buffer controller. - */ - -GLOBAL(void) -jinit_d_main_controller (j_decompress_ptr cinfo, boolean need_full_buffer) -{ - my_main_ptr mainp; - int ci, rgroup, ngroups; - jpeg_component_info *compptr; - - mainp = (my_main_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_main_controller)); - cinfo->main = &mainp->pub; - mainp->pub.start_pass = start_pass_main; - - if (need_full_buffer) /* shouldn't happen */ - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - - /* Allocate the workspace. - * ngroups is the number of row groups we need. - */ - if (cinfo->upsample->need_context_rows) { - if (cinfo->min_DCT_v_scaled_size < 2) /* unsupported, see comments above */ - ERREXIT(cinfo, JERR_NOTIMPL); - alloc_funny_pointers(cinfo); /* Alloc space for xbuffer[] lists */ - ngroups = cinfo->min_DCT_v_scaled_size + 2; - } else { - /* There are always min_DCT_v_scaled_size row groups in an iMCU row. */ - ngroups = cinfo->min_DCT_v_scaled_size; - mainp->rowgroups_avail = (JDIMENSION) ngroups; - } - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - if (! compptr->component_needed) - continue; /* skip uninteresting component */ - rgroup = (compptr->v_samp_factor * compptr->DCT_v_scaled_size) / - cinfo->min_DCT_v_scaled_size; /* height of a row group of component */ - mainp->buffer[ci] = (*cinfo->mem->alloc_sarray) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - compptr->width_in_blocks * ((JDIMENSION) compptr->DCT_h_scaled_size), - (JDIMENSION) (rgroup * ngroups)); - } -} diff --git a/dep/libjpeg/src/jdmarker.c b/dep/libjpeg/src/jdmarker.c deleted file mode 100644 index c10fde605..000000000 --- a/dep/libjpeg/src/jdmarker.c +++ /dev/null @@ -1,1505 +0,0 @@ -/* - * jdmarker.c - * - * Copyright (C) 1991-1998, Thomas G. Lane. - * Modified 2009-2019 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains routines to decode JPEG datastream markers. - * Most of the complexity arises from our desire to support input - * suspension: if not all of the data for a marker is available, - * we must exit back to the application. On resumption, we reprocess - * the marker. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -typedef enum { /* JPEG marker codes */ - M_SOF0 = 0xc0, - M_SOF1 = 0xc1, - M_SOF2 = 0xc2, - M_SOF3 = 0xc3, - - M_SOF5 = 0xc5, - M_SOF6 = 0xc6, - M_SOF7 = 0xc7, - - M_JPG = 0xc8, - M_SOF9 = 0xc9, - M_SOF10 = 0xca, - M_SOF11 = 0xcb, - - M_SOF13 = 0xcd, - M_SOF14 = 0xce, - M_SOF15 = 0xcf, - - M_DHT = 0xc4, - - M_DAC = 0xcc, - - M_RST0 = 0xd0, - M_RST1 = 0xd1, - M_RST2 = 0xd2, - M_RST3 = 0xd3, - M_RST4 = 0xd4, - M_RST5 = 0xd5, - M_RST6 = 0xd6, - M_RST7 = 0xd7, - - M_SOI = 0xd8, - M_EOI = 0xd9, - M_SOS = 0xda, - M_DQT = 0xdb, - M_DNL = 0xdc, - M_DRI = 0xdd, - M_DHP = 0xde, - M_EXP = 0xdf, - - M_APP0 = 0xe0, - M_APP1 = 0xe1, - M_APP2 = 0xe2, - M_APP3 = 0xe3, - M_APP4 = 0xe4, - M_APP5 = 0xe5, - M_APP6 = 0xe6, - M_APP7 = 0xe7, - M_APP8 = 0xe8, - M_APP9 = 0xe9, - M_APP10 = 0xea, - M_APP11 = 0xeb, - M_APP12 = 0xec, - M_APP13 = 0xed, - M_APP14 = 0xee, - M_APP15 = 0xef, - - M_JPG0 = 0xf0, - M_JPG8 = 0xf8, - M_JPG13 = 0xfd, - M_COM = 0xfe, - - M_TEM = 0x01, - - M_ERROR = 0x100 -} JPEG_MARKER; - - -/* Private state */ - -typedef struct { - struct jpeg_marker_reader pub; /* public fields */ - - /* Application-overridable marker processing methods */ - jpeg_marker_parser_method process_COM; - jpeg_marker_parser_method process_APPn[16]; - - /* Limit on marker data length to save for each marker type */ - unsigned int length_limit_COM; - unsigned int length_limit_APPn[16]; - - /* Status of COM/APPn marker saving */ - jpeg_saved_marker_ptr cur_marker; /* NULL if not processing a marker */ - unsigned int bytes_read; /* data bytes read so far in marker */ - /* Note: cur_marker is not linked into marker_list until it's all read. */ -} my_marker_reader; - -typedef my_marker_reader * my_marker_ptr; - - -/* - * Macros for fetching data from the data source module. - * - * At all times, cinfo->src->next_input_byte and ->bytes_in_buffer reflect - * the current restart point; we update them only when we have reached a - * suitable place to restart if a suspension occurs. - */ - -/* Declare and initialize local copies of input pointer/count */ -#define INPUT_VARS(cinfo) \ - struct jpeg_source_mgr * datasrc = (cinfo)->src; \ - const JOCTET * next_input_byte = datasrc->next_input_byte; \ - size_t bytes_in_buffer = datasrc->bytes_in_buffer - -/* Unload the local copies --- do this only at a restart boundary */ -#define INPUT_SYNC(cinfo) \ - ( datasrc->next_input_byte = next_input_byte, \ - datasrc->bytes_in_buffer = bytes_in_buffer ) - -/* Reload the local copies --- used only in MAKE_BYTE_AVAIL */ -#define INPUT_RELOAD(cinfo) \ - ( next_input_byte = datasrc->next_input_byte, \ - bytes_in_buffer = datasrc->bytes_in_buffer ) - -/* Internal macro for INPUT_BYTE and INPUT_2BYTES: make a byte available. - * Note we do *not* do INPUT_SYNC before calling fill_input_buffer, - * but we must reload the local copies after a successful fill. - */ -#define MAKE_BYTE_AVAIL(cinfo,action) \ - if (bytes_in_buffer == 0) { \ - if (! (*datasrc->fill_input_buffer) (cinfo)) \ - { action; } \ - INPUT_RELOAD(cinfo); \ - } - -/* Read a byte into variable V. - * If must suspend, take the specified action (typically "return FALSE"). - */ -#define INPUT_BYTE(cinfo,V,action) \ - MAKESTMT( MAKE_BYTE_AVAIL(cinfo,action); \ - bytes_in_buffer--; \ - V = GETJOCTET(*next_input_byte++); ) - -/* As above, but read two bytes interpreted as an unsigned 16-bit integer. - * V should be declared unsigned int or perhaps INT32. - */ -#define INPUT_2BYTES(cinfo,V,action) \ - MAKESTMT( MAKE_BYTE_AVAIL(cinfo,action); \ - bytes_in_buffer--; \ - V = ((unsigned int) GETJOCTET(*next_input_byte++)) << 8; \ - MAKE_BYTE_AVAIL(cinfo,action); \ - bytes_in_buffer--; \ - V += GETJOCTET(*next_input_byte++); ) - - -/* - * Routines to process JPEG markers. - * - * Entry condition: JPEG marker itself has been read and its code saved - * in cinfo->unread_marker; input restart point is just after the marker. - * - * Exit: if return TRUE, have read and processed any parameters, and have - * updated the restart point to point after the parameters. - * If return FALSE, was forced to suspend before reaching end of - * marker parameters; restart point has not been moved. Same routine - * will be called again after application supplies more input data. - * - * This approach to suspension assumes that all of a marker's parameters - * can fit into a single input bufferload. This should hold for "normal" - * markers. Some COM/APPn markers might have large parameter segments - * that might not fit. If we are simply dropping such a marker, we use - * skip_input_data to get past it, and thereby put the problem on the - * source manager's shoulders. If we are saving the marker's contents - * into memory, we use a slightly different convention: when forced to - * suspend, the marker processor updates the restart point to the end of - * what it's consumed (ie, the end of the buffer) before returning FALSE. - * On resumption, cinfo->unread_marker still contains the marker code, - * but the data source will point to the next chunk of marker data. - * The marker processor must retain internal state to deal with this. - * - * Note that we don't bother to avoid duplicate trace messages if a - * suspension occurs within marker parameters. Other side effects - * require more care. - */ - - -LOCAL(boolean) -get_soi (j_decompress_ptr cinfo) -/* Process an SOI marker */ -{ - int i; - - TRACEMS(cinfo, 1, JTRC_SOI); - - if (cinfo->marker->saw_SOI) - ERREXIT(cinfo, JERR_SOI_DUPLICATE); - - /* Reset all parameters that are defined to be reset by SOI */ - - for (i = 0; i < NUM_ARITH_TBLS; i++) { - cinfo->arith_dc_L[i] = 0; - cinfo->arith_dc_U[i] = 1; - cinfo->arith_ac_K[i] = 5; - } - cinfo->restart_interval = 0; - - /* Set initial assumptions for colorspace etc */ - - cinfo->jpeg_color_space = JCS_UNKNOWN; - cinfo->color_transform = JCT_NONE; - cinfo->CCIR601_sampling = FALSE; /* Assume non-CCIR sampling??? */ - - cinfo->saw_JFIF_marker = FALSE; - cinfo->JFIF_major_version = 1; /* set default JFIF APP0 values */ - cinfo->JFIF_minor_version = 1; - cinfo->density_unit = 0; - cinfo->X_density = 1; - cinfo->Y_density = 1; - cinfo->saw_Adobe_marker = FALSE; - cinfo->Adobe_transform = 0; - - cinfo->marker->saw_SOI = TRUE; - - return TRUE; -} - - -LOCAL(boolean) -get_sof (j_decompress_ptr cinfo, boolean is_baseline, boolean is_prog, - boolean is_arith) -/* Process a SOFn marker */ -{ - INT32 length; - int c, ci, i; - jpeg_component_info * compptr; - INPUT_VARS(cinfo); - - cinfo->is_baseline = is_baseline; - cinfo->progressive_mode = is_prog; - cinfo->arith_code = is_arith; - - INPUT_2BYTES(cinfo, length, return FALSE); - - INPUT_BYTE(cinfo, cinfo->data_precision, return FALSE); - INPUT_2BYTES(cinfo, cinfo->image_height, return FALSE); - INPUT_2BYTES(cinfo, cinfo->image_width, return FALSE); - INPUT_BYTE(cinfo, cinfo->num_components, return FALSE); - - length -= 8; - - TRACEMS4(cinfo, 1, JTRC_SOF, cinfo->unread_marker, - (int) cinfo->image_width, (int) cinfo->image_height, - cinfo->num_components); - - if (cinfo->marker->saw_SOF) - ERREXIT(cinfo, JERR_SOF_DUPLICATE); - - /* We don't support files in which the image height is initially specified */ - /* as 0 and is later redefined by DNL. As long as we have to check that, */ - /* might as well have a general sanity check. */ - if (cinfo->image_height <= 0 || cinfo->image_width <= 0 || - cinfo->num_components <= 0) - ERREXIT(cinfo, JERR_EMPTY_IMAGE); - - if (length != (cinfo->num_components * 3)) - ERREXIT(cinfo, JERR_BAD_LENGTH); - - if (cinfo->comp_info == NULL) /* do only once, even if suspend */ - cinfo->comp_info = (jpeg_component_info *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - cinfo->num_components * SIZEOF(jpeg_component_info)); - - for (ci = 0; ci < cinfo->num_components; ci++) { - INPUT_BYTE(cinfo, c, return FALSE); - /* Check to see whether component id has already been seen */ - /* (in violation of the spec, but unfortunately seen in some */ - /* files). If so, create "fake" component id equal to the */ - /* max id seen so far + 1. */ - for (i = 0, compptr = cinfo->comp_info; i < ci; i++, compptr++) { - if (c == compptr->component_id) { - compptr = cinfo->comp_info; - c = compptr->component_id; - compptr++; - for (i = 1; i < ci; i++, compptr++) { - if (compptr->component_id > c) c = compptr->component_id; - } - c++; - break; - } - } - compptr->component_id = c; - compptr->component_index = ci; - INPUT_BYTE(cinfo, c, return FALSE); - compptr->h_samp_factor = (c >> 4) & 15; - compptr->v_samp_factor = (c ) & 15; - INPUT_BYTE(cinfo, compptr->quant_tbl_no, return FALSE); - - TRACEMS4(cinfo, 1, JTRC_SOF_COMPONENT, - compptr->component_id, compptr->h_samp_factor, - compptr->v_samp_factor, compptr->quant_tbl_no); - } - - cinfo->marker->saw_SOF = TRUE; - - INPUT_SYNC(cinfo); - return TRUE; -} - - -LOCAL(boolean) -get_sos (j_decompress_ptr cinfo) -/* Process a SOS marker */ -{ - INT32 length; - int c, ci, i, n; - jpeg_component_info * compptr; - INPUT_VARS(cinfo); - - if (! cinfo->marker->saw_SOF) - ERREXITS(cinfo, JERR_SOF_BEFORE, "SOS"); - - INPUT_2BYTES(cinfo, length, return FALSE); - - INPUT_BYTE(cinfo, n, return FALSE); /* Number of components */ - - TRACEMS1(cinfo, 1, JTRC_SOS, n); - - if (length != (n * 2 + 6) || n > MAX_COMPS_IN_SCAN || - (n == 0 && !cinfo->progressive_mode)) - /* pseudo SOS marker only allowed in progressive mode */ - ERREXIT(cinfo, JERR_BAD_LENGTH); - - cinfo->comps_in_scan = n; - - /* Collect the component-spec parameters */ - - for (i = 0; i < n; i++) { - INPUT_BYTE(cinfo, c, return FALSE); - - /* Detect the case where component id's are not unique, and, if so, */ - /* create a fake component id using the same logic as in get_sof. */ - /* Note: This also ensures that all of the SOF components are */ - /* referenced in the single scan case, which prevents access to */ - /* uninitialized memory in later decoding stages. */ - for (ci = 0; ci < i; ci++) { - if (c == cinfo->cur_comp_info[ci]->component_id) { - c = cinfo->cur_comp_info[0]->component_id; - for (ci = 1; ci < i; ci++) { - compptr = cinfo->cur_comp_info[ci]; - if (compptr->component_id > c) c = compptr->component_id; - } - c++; - break; - } - } - - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - if (c == compptr->component_id) - goto id_found; - } - - ERREXIT1(cinfo, JERR_BAD_COMPONENT_ID, c); - - id_found: - - cinfo->cur_comp_info[i] = compptr; - INPUT_BYTE(cinfo, c, return FALSE); - compptr->dc_tbl_no = (c >> 4) & 15; - compptr->ac_tbl_no = (c ) & 15; - - TRACEMS3(cinfo, 1, JTRC_SOS_COMPONENT, compptr->component_id, - compptr->dc_tbl_no, compptr->ac_tbl_no); - } - - /* Collect the additional scan parameters Ss, Se, Ah/Al. */ - INPUT_BYTE(cinfo, c, return FALSE); - cinfo->Ss = c; - INPUT_BYTE(cinfo, c, return FALSE); - cinfo->Se = c; - INPUT_BYTE(cinfo, c, return FALSE); - cinfo->Ah = (c >> 4) & 15; - cinfo->Al = (c ) & 15; - - TRACEMS4(cinfo, 1, JTRC_SOS_PARAMS, cinfo->Ss, cinfo->Se, - cinfo->Ah, cinfo->Al); - - /* Prepare to scan data & restart markers */ - cinfo->marker->next_restart_num = 0; - - /* Count another (non-pseudo) SOS marker */ - if (n) cinfo->input_scan_number++; - - INPUT_SYNC(cinfo); - return TRUE; -} - - -#ifdef D_ARITH_CODING_SUPPORTED - -LOCAL(boolean) -get_dac (j_decompress_ptr cinfo) -/* Process a DAC marker */ -{ - INT32 length; - int index, val; - INPUT_VARS(cinfo); - - INPUT_2BYTES(cinfo, length, return FALSE); - length -= 2; - - while (length > 0) { - INPUT_BYTE(cinfo, index, return FALSE); - INPUT_BYTE(cinfo, val, return FALSE); - - length -= 2; - - TRACEMS2(cinfo, 1, JTRC_DAC, index, val); - - if (index < 0 || index >= (2*NUM_ARITH_TBLS)) - ERREXIT1(cinfo, JERR_DAC_INDEX, index); - - if (index >= NUM_ARITH_TBLS) { /* define AC table */ - cinfo->arith_ac_K[index-NUM_ARITH_TBLS] = (UINT8) val; - } else { /* define DC table */ - cinfo->arith_dc_L[index] = (UINT8) (val & 0x0F); - cinfo->arith_dc_U[index] = (UINT8) (val >> 4); - if (cinfo->arith_dc_L[index] > cinfo->arith_dc_U[index]) - ERREXIT1(cinfo, JERR_DAC_VALUE, val); - } - } - - if (length != 0) - ERREXIT(cinfo, JERR_BAD_LENGTH); - - INPUT_SYNC(cinfo); - return TRUE; -} - -#else /* ! D_ARITH_CODING_SUPPORTED */ - -#define get_dac(cinfo) skip_variable(cinfo) - -#endif /* D_ARITH_CODING_SUPPORTED */ - - -LOCAL(boolean) -get_dht (j_decompress_ptr cinfo) -/* Process a DHT marker */ -{ - INT32 length; - UINT8 bits[17]; - UINT8 huffval[256]; - int i, index, count; - JHUFF_TBL **htblptr; - INPUT_VARS(cinfo); - - INPUT_2BYTES(cinfo, length, return FALSE); - length -= 2; - - while (length > 16) { - INPUT_BYTE(cinfo, index, return FALSE); - - TRACEMS1(cinfo, 1, JTRC_DHT, index); - - bits[0] = 0; - count = 0; - for (i = 1; i <= 16; i++) { - INPUT_BYTE(cinfo, bits[i], return FALSE); - count += bits[i]; - } - - length -= 1 + 16; - - TRACEMS8(cinfo, 2, JTRC_HUFFBITS, - bits[1], bits[2], bits[3], bits[4], - bits[5], bits[6], bits[7], bits[8]); - TRACEMS8(cinfo, 2, JTRC_HUFFBITS, - bits[9], bits[10], bits[11], bits[12], - bits[13], bits[14], bits[15], bits[16]); - - /* Here we just do minimal validation of the counts to avoid walking - * off the end of our table space. jdhuff.c will check more carefully. - */ - if (count > 256 || ((INT32) count) > length) - ERREXIT(cinfo, JERR_BAD_HUFF_TABLE); - - for (i = 0; i < count; i++) - INPUT_BYTE(cinfo, huffval[i], return FALSE); - - length -= count; - - if (index & 0x10) { /* AC table definition */ - index -= 0x10; - htblptr = &cinfo->ac_huff_tbl_ptrs[index]; - } else { /* DC table definition */ - htblptr = &cinfo->dc_huff_tbl_ptrs[index]; - } - - if (index < 0 || index >= NUM_HUFF_TBLS) - ERREXIT1(cinfo, JERR_DHT_INDEX, index); - - if (*htblptr == NULL) - *htblptr = jpeg_alloc_huff_table((j_common_ptr) cinfo); - - MEMCOPY((*htblptr)->bits, bits, SIZEOF((*htblptr)->bits)); - if (count > 0) - MEMCOPY((*htblptr)->huffval, huffval, count * SIZEOF(UINT8)); - } - - if (length != 0) - ERREXIT(cinfo, JERR_BAD_LENGTH); - - INPUT_SYNC(cinfo); - return TRUE; -} - - -LOCAL(boolean) -get_dqt (j_decompress_ptr cinfo) -/* Process a DQT marker */ -{ - INT32 length, count, i; - int n, prec; - unsigned int tmp; - JQUANT_TBL *quant_ptr; - const int *natural_order; - INPUT_VARS(cinfo); - - INPUT_2BYTES(cinfo, length, return FALSE); - length -= 2; - - while (length > 0) { - length--; - INPUT_BYTE(cinfo, n, return FALSE); - prec = n >> 4; - n &= 0x0F; - - TRACEMS2(cinfo, 1, JTRC_DQT, n, prec); - - if (n >= NUM_QUANT_TBLS) - ERREXIT1(cinfo, JERR_DQT_INDEX, n); - - if (cinfo->quant_tbl_ptrs[n] == NULL) - cinfo->quant_tbl_ptrs[n] = jpeg_alloc_quant_table((j_common_ptr) cinfo); - quant_ptr = cinfo->quant_tbl_ptrs[n]; - - if (prec) { - if (length < DCTSIZE2 * 2) { - /* Initialize full table for safety. */ - for (i = 0; i < DCTSIZE2; i++) { - quant_ptr->quantval[i] = 1; - } - count = length >> 1; - } else - count = DCTSIZE2; - } else { - if (length < DCTSIZE2) { - /* Initialize full table for safety. */ - for (i = 0; i < DCTSIZE2; i++) { - quant_ptr->quantval[i] = 1; - } - count = length; - } else - count = DCTSIZE2; - } - - switch ((int) count) { - case (2*2): natural_order = jpeg_natural_order2; break; - case (3*3): natural_order = jpeg_natural_order3; break; - case (4*4): natural_order = jpeg_natural_order4; break; - case (5*5): natural_order = jpeg_natural_order5; break; - case (6*6): natural_order = jpeg_natural_order6; break; - case (7*7): natural_order = jpeg_natural_order7; break; - default: natural_order = jpeg_natural_order; - } - - for (i = 0; i < count; i++) { - if (prec) - INPUT_2BYTES(cinfo, tmp, return FALSE); - else - INPUT_BYTE(cinfo, tmp, return FALSE); - /* We convert the zigzag-order table to natural array order. */ - quant_ptr->quantval[natural_order[i]] = (UINT16) tmp; - } - - if (cinfo->err->trace_level >= 2) { - for (i = 0; i < DCTSIZE2; i += 8) { - TRACEMS8(cinfo, 2, JTRC_QUANTVALS, - quant_ptr->quantval[i], quant_ptr->quantval[i+1], - quant_ptr->quantval[i+2], quant_ptr->quantval[i+3], - quant_ptr->quantval[i+4], quant_ptr->quantval[i+5], - quant_ptr->quantval[i+6], quant_ptr->quantval[i+7]); - } - } - - length -= count; - if (prec) length -= count; - } - - if (length != 0) - ERREXIT(cinfo, JERR_BAD_LENGTH); - - INPUT_SYNC(cinfo); - return TRUE; -} - - -LOCAL(boolean) -get_dri (j_decompress_ptr cinfo) -/* Process a DRI marker */ -{ - INT32 length; - unsigned int tmp; - INPUT_VARS(cinfo); - - INPUT_2BYTES(cinfo, length, return FALSE); - - if (length != 4) - ERREXIT(cinfo, JERR_BAD_LENGTH); - - INPUT_2BYTES(cinfo, tmp, return FALSE); - - TRACEMS1(cinfo, 1, JTRC_DRI, tmp); - - cinfo->restart_interval = tmp; - - INPUT_SYNC(cinfo); - return TRUE; -} - - -LOCAL(boolean) -get_lse (j_decompress_ptr cinfo) -/* Process an LSE marker */ -{ - INT32 length; - unsigned int tmp; - int cid; - INPUT_VARS(cinfo); - - if (! cinfo->marker->saw_SOF) - ERREXITS(cinfo, JERR_SOF_BEFORE, "LSE"); - - if (cinfo->num_components < 3) goto bad; - - INPUT_2BYTES(cinfo, length, return FALSE); - - if (length != 24) - ERREXIT(cinfo, JERR_BAD_LENGTH); - - INPUT_BYTE(cinfo, tmp, return FALSE); - if (tmp != 0x0D) /* ID inverse transform specification */ - ERREXIT1(cinfo, JERR_UNKNOWN_MARKER, cinfo->unread_marker); - INPUT_2BYTES(cinfo, tmp, return FALSE); - if (tmp != MAXJSAMPLE) goto bad; /* MAXTRANS */ - INPUT_BYTE(cinfo, tmp, return FALSE); - if (tmp != 3) goto bad; /* Nt=3 */ - INPUT_BYTE(cinfo, cid, return FALSE); - if (cid != cinfo->comp_info[1].component_id) goto bad; - INPUT_BYTE(cinfo, cid, return FALSE); - if (cid != cinfo->comp_info[0].component_id) goto bad; - INPUT_BYTE(cinfo, cid, return FALSE); - if (cid != cinfo->comp_info[2].component_id) goto bad; - INPUT_BYTE(cinfo, tmp, return FALSE); - if (tmp != 0x80) goto bad; /* F1: CENTER1=1, NORM1=0 */ - INPUT_2BYTES(cinfo, tmp, return FALSE); - if (tmp != 0) goto bad; /* A(1,1)=0 */ - INPUT_2BYTES(cinfo, tmp, return FALSE); - if (tmp != 0) goto bad; /* A(1,2)=0 */ - INPUT_BYTE(cinfo, tmp, return FALSE); - if (tmp != 0) goto bad; /* F2: CENTER2=0, NORM2=0 */ - INPUT_2BYTES(cinfo, tmp, return FALSE); - if (tmp != 1) goto bad; /* A(2,1)=1 */ - INPUT_2BYTES(cinfo, tmp, return FALSE); - if (tmp != 0) goto bad; /* A(2,2)=0 */ - INPUT_BYTE(cinfo, tmp, return FALSE); - if (tmp != 0) goto bad; /* F3: CENTER3=0, NORM3=0 */ - INPUT_2BYTES(cinfo, tmp, return FALSE); - if (tmp != 1) goto bad; /* A(3,1)=1 */ - INPUT_2BYTES(cinfo, tmp, return FALSE); - if (tmp != 0) { /* A(3,2)=0 */ - bad: - ERREXIT(cinfo, JERR_CONVERSION_NOTIMPL); - } - - /* OK, valid transform that we can handle. */ - cinfo->color_transform = JCT_SUBTRACT_GREEN; - - INPUT_SYNC(cinfo); - return TRUE; -} - - -/* - * Routines for processing APPn and COM markers. - * These are either saved in memory or discarded, per application request. - * APP0 and APP14 are specially checked to see if they are - * JFIF and Adobe markers, respectively. - */ - -#define APP0_DATA_LEN 14 /* Length of interesting data in APP0 */ -#define APP14_DATA_LEN 12 /* Length of interesting data in APP14 */ -#define APPN_DATA_LEN 14 /* Must be the largest of the above!! */ - - -LOCAL(void) -examine_app0 (j_decompress_ptr cinfo, JOCTET FAR * data, - unsigned int datalen, INT32 remaining) -/* Examine first few bytes from an APP0. - * Take appropriate action if it is a JFIF marker. - * datalen is # of bytes at data[], remaining is length of rest of marker data. - */ -{ - INT32 totallen = (INT32) datalen + remaining; - - if (datalen >= APP0_DATA_LEN && - GETJOCTET(data[0]) == 0x4A && - GETJOCTET(data[1]) == 0x46 && - GETJOCTET(data[2]) == 0x49 && - GETJOCTET(data[3]) == 0x46 && - GETJOCTET(data[4]) == 0) { - /* Found JFIF APP0 marker: save info */ - cinfo->saw_JFIF_marker = TRUE; - cinfo->JFIF_major_version = GETJOCTET(data[5]); - cinfo->JFIF_minor_version = GETJOCTET(data[6]); - cinfo->density_unit = GETJOCTET(data[7]); - cinfo->X_density = (GETJOCTET(data[8]) << 8) + GETJOCTET(data[9]); - cinfo->Y_density = (GETJOCTET(data[10]) << 8) + GETJOCTET(data[11]); - /* Check version. - * Major version must be 1 or 2, anything else signals an incompatible - * change. - * (We used to treat this as an error, but now it's a nonfatal warning, - * because some bozo at Hijaak couldn't read the spec.) - * Minor version should be 0..2, but process anyway if newer. - */ - if (cinfo->JFIF_major_version != 1 && cinfo->JFIF_major_version != 2) - WARNMS2(cinfo, JWRN_JFIF_MAJOR, - cinfo->JFIF_major_version, cinfo->JFIF_minor_version); - /* Generate trace messages */ - TRACEMS5(cinfo, 1, JTRC_JFIF, - cinfo->JFIF_major_version, cinfo->JFIF_minor_version, - cinfo->X_density, cinfo->Y_density, cinfo->density_unit); - /* Validate thumbnail dimensions and issue appropriate messages */ - if (GETJOCTET(data[12]) | GETJOCTET(data[13])) - TRACEMS2(cinfo, 1, JTRC_JFIF_THUMBNAIL, - GETJOCTET(data[12]), GETJOCTET(data[13])); - totallen -= APP0_DATA_LEN; - if (totallen != - ((INT32)GETJOCTET(data[12]) * (INT32)GETJOCTET(data[13]) * (INT32) 3)) - TRACEMS1(cinfo, 1, JTRC_JFIF_BADTHUMBNAILSIZE, (int) totallen); - } else if (datalen >= 6 && - GETJOCTET(data[0]) == 0x4A && - GETJOCTET(data[1]) == 0x46 && - GETJOCTET(data[2]) == 0x58 && - GETJOCTET(data[3]) == 0x58 && - GETJOCTET(data[4]) == 0) { - /* Found JFIF "JFXX" extension APP0 marker */ - /* The library doesn't actually do anything with these, - * but we try to produce a helpful trace message. - */ - switch (GETJOCTET(data[5])) { - case 0x10: - TRACEMS1(cinfo, 1, JTRC_THUMB_JPEG, (int) totallen); - break; - case 0x11: - TRACEMS1(cinfo, 1, JTRC_THUMB_PALETTE, (int) totallen); - break; - case 0x13: - TRACEMS1(cinfo, 1, JTRC_THUMB_RGB, (int) totallen); - break; - default: - TRACEMS2(cinfo, 1, JTRC_JFIF_EXTENSION, - GETJOCTET(data[5]), (int) totallen); - } - } else { - /* Start of APP0 does not match "JFIF" or "JFXX", or too short */ - TRACEMS1(cinfo, 1, JTRC_APP0, (int) totallen); - } -} - - -LOCAL(void) -examine_app14 (j_decompress_ptr cinfo, JOCTET FAR * data, - unsigned int datalen, INT32 remaining) -/* Examine first few bytes from an APP14. - * Take appropriate action if it is an Adobe marker. - * datalen is # of bytes at data[], remaining is length of rest of marker data. - */ -{ - unsigned int version, flags0, flags1, transform; - - if (datalen >= APP14_DATA_LEN && - GETJOCTET(data[0]) == 0x41 && - GETJOCTET(data[1]) == 0x64 && - GETJOCTET(data[2]) == 0x6F && - GETJOCTET(data[3]) == 0x62 && - GETJOCTET(data[4]) == 0x65) { - /* Found Adobe APP14 marker */ - version = (GETJOCTET(data[5]) << 8) + GETJOCTET(data[6]); - flags0 = (GETJOCTET(data[7]) << 8) + GETJOCTET(data[8]); - flags1 = (GETJOCTET(data[9]) << 8) + GETJOCTET(data[10]); - transform = GETJOCTET(data[11]); - TRACEMS4(cinfo, 1, JTRC_ADOBE, version, flags0, flags1, transform); - cinfo->saw_Adobe_marker = TRUE; - cinfo->Adobe_transform = (UINT8) transform; - } else { - /* Start of APP14 does not match "Adobe", or too short */ - TRACEMS1(cinfo, 1, JTRC_APP14, (int) (datalen + remaining)); - } -} - - -METHODDEF(boolean) -get_interesting_appn (j_decompress_ptr cinfo) -/* Process an APP0 or APP14 marker without saving it */ -{ - INT32 length; - JOCTET b[APPN_DATA_LEN]; - unsigned int i, numtoread; - INPUT_VARS(cinfo); - - INPUT_2BYTES(cinfo, length, return FALSE); - length -= 2; - - /* get the interesting part of the marker data */ - if (length >= APPN_DATA_LEN) - numtoread = APPN_DATA_LEN; - else if (length > 0) - numtoread = (unsigned int) length; - else - numtoread = 0; - for (i = 0; i < numtoread; i++) - INPUT_BYTE(cinfo, b[i], return FALSE); - length -= numtoread; - - /* process it */ - switch (cinfo->unread_marker) { - case M_APP0: - examine_app0(cinfo, (JOCTET FAR *) b, numtoread, length); - break; - case M_APP14: - examine_app14(cinfo, (JOCTET FAR *) b, numtoread, length); - break; - default: - /* can't get here unless jpeg_save_markers chooses wrong processor */ - ERREXIT1(cinfo, JERR_UNKNOWN_MARKER, cinfo->unread_marker); - } - - /* skip any remaining data -- could be lots */ - INPUT_SYNC(cinfo); - if (length > 0) - (*cinfo->src->skip_input_data) (cinfo, (long) length); - - return TRUE; -} - - -#ifdef SAVE_MARKERS_SUPPORTED - -METHODDEF(boolean) -save_marker (j_decompress_ptr cinfo) -/* Save an APPn or COM marker into the marker list */ -{ - my_marker_ptr marker = (my_marker_ptr) cinfo->marker; - jpeg_saved_marker_ptr cur_marker = marker->cur_marker; - unsigned int bytes_read, data_length; - JOCTET FAR * data; - INT32 length = 0; - INPUT_VARS(cinfo); - - if (cur_marker == NULL) { - /* begin reading a marker */ - INPUT_2BYTES(cinfo, length, return FALSE); - length -= 2; - if (length >= 0) { /* watch out for bogus length word */ - /* figure out how much we want to save */ - unsigned int limit; - if (cinfo->unread_marker == (int) M_COM) - limit = marker->length_limit_COM; - else - limit = marker->length_limit_APPn[cinfo->unread_marker - (int) M_APP0]; - if ((unsigned int) length < limit) - limit = (unsigned int) length; - /* allocate and initialize the marker item */ - cur_marker = (jpeg_saved_marker_ptr) - (*cinfo->mem->alloc_large) ((j_common_ptr) cinfo, JPOOL_IMAGE, - SIZEOF(struct jpeg_marker_struct) + limit); - cur_marker->next = NULL; - cur_marker->marker = (UINT8) cinfo->unread_marker; - cur_marker->original_length = (unsigned int) length; - cur_marker->data_length = limit; - /* data area is just beyond the jpeg_marker_struct */ - data = cur_marker->data = (JOCTET FAR *) (cur_marker + 1); - marker->cur_marker = cur_marker; - marker->bytes_read = 0; - bytes_read = 0; - data_length = limit; - } else { - /* deal with bogus length word */ - bytes_read = data_length = 0; - data = NULL; - } - } else { - /* resume reading a marker */ - bytes_read = marker->bytes_read; - data_length = cur_marker->data_length; - data = cur_marker->data + bytes_read; - } - - while (bytes_read < data_length) { - INPUT_SYNC(cinfo); /* move the restart point to here */ - marker->bytes_read = bytes_read; - /* If there's not at least one byte in buffer, suspend */ - MAKE_BYTE_AVAIL(cinfo, return FALSE); - /* Copy bytes with reasonable rapidity */ - while (bytes_read < data_length && bytes_in_buffer > 0) { - *data++ = *next_input_byte++; - bytes_in_buffer--; - bytes_read++; - } - } - - /* Done reading what we want to read */ - if (cur_marker != NULL) { /* will be NULL if bogus length word */ - /* Add new marker to end of list */ - if (cinfo->marker_list == NULL) { - cinfo->marker_list = cur_marker; - } else { - jpeg_saved_marker_ptr prev = cinfo->marker_list; - while (prev->next != NULL) - prev = prev->next; - prev->next = cur_marker; - } - /* Reset pointer & calc remaining data length */ - data = cur_marker->data; - length = cur_marker->original_length - data_length; - } - /* Reset to initial state for next marker */ - marker->cur_marker = NULL; - - /* Process the marker if interesting; else just make a generic trace msg */ - switch (cinfo->unread_marker) { - case M_APP0: - examine_app0(cinfo, data, data_length, length); - break; - case M_APP14: - examine_app14(cinfo, data, data_length, length); - break; - default: - TRACEMS2(cinfo, 1, JTRC_MISC_MARKER, cinfo->unread_marker, - (int) (data_length + length)); - } - - /* skip any remaining data -- could be lots */ - INPUT_SYNC(cinfo); /* do before skip_input_data */ - if (length > 0) - (*cinfo->src->skip_input_data) (cinfo, (long) length); - - return TRUE; -} - -#endif /* SAVE_MARKERS_SUPPORTED */ - - -METHODDEF(boolean) -skip_variable (j_decompress_ptr cinfo) -/* Skip over an unknown or uninteresting variable-length marker */ -{ - INT32 length; - INPUT_VARS(cinfo); - - INPUT_2BYTES(cinfo, length, return FALSE); - length -= 2; - - TRACEMS2(cinfo, 1, JTRC_MISC_MARKER, cinfo->unread_marker, (int) length); - - INPUT_SYNC(cinfo); /* do before skip_input_data */ - if (length > 0) - (*cinfo->src->skip_input_data) (cinfo, (long) length); - - return TRUE; -} - - -/* - * Find the next JPEG marker, save it in cinfo->unread_marker. - * Returns FALSE if had to suspend before reaching a marker; - * in that case cinfo->unread_marker is unchanged. - * - * Note that the result might not be a valid marker code, - * but it will never be 0 or FF. - */ - -LOCAL(boolean) -next_marker (j_decompress_ptr cinfo) -{ - int c; - INPUT_VARS(cinfo); - - for (;;) { - INPUT_BYTE(cinfo, c, return FALSE); - /* Skip any non-FF bytes. - * This may look a bit inefficient, but it will not occur in a valid file. - * We sync after each discarded byte so that a suspending data source - * can discard the byte from its buffer. - */ - while (c != 0xFF) { - cinfo->marker->discarded_bytes++; - INPUT_SYNC(cinfo); - INPUT_BYTE(cinfo, c, return FALSE); - } - /* This loop swallows any duplicate FF bytes. Extra FFs are legal as - * pad bytes, so don't count them in discarded_bytes. We assume there - * will not be so many consecutive FF bytes as to overflow a suspending - * data source's input buffer. - */ - do { - INPUT_BYTE(cinfo, c, return FALSE); - } while (c == 0xFF); - if (c != 0) - break; /* found a valid marker, exit loop */ - /* Reach here if we found a stuffed-zero data sequence (FF/00). - * Discard it and loop back to try again. - */ - cinfo->marker->discarded_bytes += 2; - INPUT_SYNC(cinfo); - } - - if (cinfo->marker->discarded_bytes != 0) { - WARNMS2(cinfo, JWRN_EXTRANEOUS_DATA, cinfo->marker->discarded_bytes, c); - cinfo->marker->discarded_bytes = 0; - } - - cinfo->unread_marker = c; - - INPUT_SYNC(cinfo); - return TRUE; -} - - -LOCAL(boolean) -first_marker (j_decompress_ptr cinfo) -/* Like next_marker, but used to obtain the initial SOI marker. */ -/* For this marker, we do not allow preceding garbage or fill; otherwise, - * we might well scan an entire input file before realizing it ain't JPEG. - * If an application wants to process non-JFIF files, it must seek to the - * SOI before calling the JPEG library. - */ -{ - int c, c2; - INPUT_VARS(cinfo); - - INPUT_BYTE(cinfo, c, return FALSE); - INPUT_BYTE(cinfo, c2, return FALSE); - if (c != 0xFF || c2 != (int) M_SOI) - ERREXIT2(cinfo, JERR_NO_SOI, c, c2); - - cinfo->unread_marker = c2; - - INPUT_SYNC(cinfo); - return TRUE; -} - - -/* - * Read markers until SOS or EOI. - * - * Returns same codes as are defined for jpeg_consume_input: - * JPEG_SUSPENDED, JPEG_REACHED_SOS, or JPEG_REACHED_EOI. - * - * Note: This function may return a pseudo SOS marker (with zero - * component number) for treat by input controller's consume_input. - * consume_input itself should filter out (skip) the pseudo marker - * after processing for the caller. - */ - -METHODDEF(int) -read_markers (j_decompress_ptr cinfo) -{ - /* Outer loop repeats once for each marker. */ - for (;;) { - /* Collect the marker proper, unless we already did. */ - /* NB: first_marker() enforces the requirement that SOI appear first. */ - if (cinfo->unread_marker == 0) { - if (! cinfo->marker->saw_SOI) { - if (! first_marker(cinfo)) - return JPEG_SUSPENDED; - } else { - if (! next_marker(cinfo)) - return JPEG_SUSPENDED; - } - } - /* At this point cinfo->unread_marker contains the marker code and the - * input point is just past the marker proper, but before any parameters. - * A suspension will cause us to return with this state still true. - */ - switch (cinfo->unread_marker) { - case M_SOI: - if (! get_soi(cinfo)) - return JPEG_SUSPENDED; - break; - - case M_SOF0: /* Baseline */ - if (! get_sof(cinfo, TRUE, FALSE, FALSE)) - return JPEG_SUSPENDED; - break; - - case M_SOF1: /* Extended sequential, Huffman */ - if (! get_sof(cinfo, FALSE, FALSE, FALSE)) - return JPEG_SUSPENDED; - break; - - case M_SOF2: /* Progressive, Huffman */ - if (! get_sof(cinfo, FALSE, TRUE, FALSE)) - return JPEG_SUSPENDED; - break; - - case M_SOF9: /* Extended sequential, arithmetic */ - if (! get_sof(cinfo, FALSE, FALSE, TRUE)) - return JPEG_SUSPENDED; - break; - - case M_SOF10: /* Progressive, arithmetic */ - if (! get_sof(cinfo, FALSE, TRUE, TRUE)) - return JPEG_SUSPENDED; - break; - - /* Currently unsupported SOFn types */ - case M_SOF3: /* Lossless, Huffman */ - case M_SOF5: /* Differential sequential, Huffman */ - case M_SOF6: /* Differential progressive, Huffman */ - case M_SOF7: /* Differential lossless, Huffman */ - case M_JPG: /* Reserved for JPEG extensions */ - case M_SOF11: /* Lossless, arithmetic */ - case M_SOF13: /* Differential sequential, arithmetic */ - case M_SOF14: /* Differential progressive, arithmetic */ - case M_SOF15: /* Differential lossless, arithmetic */ - ERREXIT1(cinfo, JERR_SOF_UNSUPPORTED, cinfo->unread_marker); - break; - - case M_SOS: - if (! get_sos(cinfo)) - return JPEG_SUSPENDED; - cinfo->unread_marker = 0; /* processed the marker */ - return JPEG_REACHED_SOS; - - case M_EOI: - TRACEMS(cinfo, 1, JTRC_EOI); - cinfo->unread_marker = 0; /* processed the marker */ - return JPEG_REACHED_EOI; - - case M_DAC: - if (! get_dac(cinfo)) - return JPEG_SUSPENDED; - break; - - case M_DHT: - if (! get_dht(cinfo)) - return JPEG_SUSPENDED; - break; - - case M_DQT: - if (! get_dqt(cinfo)) - return JPEG_SUSPENDED; - break; - - case M_DRI: - if (! get_dri(cinfo)) - return JPEG_SUSPENDED; - break; - - case M_JPG8: - if (! get_lse(cinfo)) - return JPEG_SUSPENDED; - break; - - case M_APP0: - case M_APP1: - case M_APP2: - case M_APP3: - case M_APP4: - case M_APP5: - case M_APP6: - case M_APP7: - case M_APP8: - case M_APP9: - case M_APP10: - case M_APP11: - case M_APP12: - case M_APP13: - case M_APP14: - case M_APP15: - if (! (*((my_marker_ptr) cinfo->marker)->process_APPn[ - cinfo->unread_marker - (int) M_APP0]) (cinfo)) - return JPEG_SUSPENDED; - break; - - case M_COM: - if (! (*((my_marker_ptr) cinfo->marker)->process_COM) (cinfo)) - return JPEG_SUSPENDED; - break; - - case M_RST0: /* these are all parameterless */ - case M_RST1: - case M_RST2: - case M_RST3: - case M_RST4: - case M_RST5: - case M_RST6: - case M_RST7: - case M_TEM: - TRACEMS1(cinfo, 1, JTRC_PARMLESS_MARKER, cinfo->unread_marker); - break; - - case M_DNL: /* Ignore DNL ... perhaps the wrong thing */ - if (! skip_variable(cinfo)) - return JPEG_SUSPENDED; - break; - - default: /* must be DHP, EXP, JPGn, or RESn */ - /* For now, we treat the reserved markers as fatal errors since they are - * likely to be used to signal incompatible JPEG Part 3 extensions. - * Once the JPEG 3 version-number marker is well defined, this code - * ought to change! - */ - ERREXIT1(cinfo, JERR_UNKNOWN_MARKER, cinfo->unread_marker); - } - /* Successfully processed marker, so reset state variable */ - cinfo->unread_marker = 0; - } /* end loop */ -} - - -/* - * Read a restart marker, which is expected to appear next in the datastream; - * if the marker is not there, take appropriate recovery action. - * Returns FALSE if suspension is required. - * - * This is called by the entropy decoder after it has read an appropriate - * number of MCUs. cinfo->unread_marker may be nonzero if the entropy decoder - * has already read a marker from the data source. Under normal conditions - * cinfo->unread_marker will be reset to 0 before returning; if not reset, - * it holds a marker which the decoder will be unable to read past. - */ - -METHODDEF(boolean) -read_restart_marker (j_decompress_ptr cinfo) -{ - /* Obtain a marker unless we already did. */ - /* Note that next_marker will complain if it skips any data. */ - if (cinfo->unread_marker == 0) { - if (! next_marker(cinfo)) - return FALSE; - } - - if (cinfo->unread_marker == - ((int) M_RST0 + cinfo->marker->next_restart_num)) { - /* Normal case --- swallow the marker and let entropy decoder continue */ - TRACEMS1(cinfo, 3, JTRC_RST, cinfo->marker->next_restart_num); - cinfo->unread_marker = 0; - } else { - /* Uh-oh, the restart markers have been messed up. */ - /* Let the data source manager determine how to resync. */ - if (! (*cinfo->src->resync_to_restart) (cinfo, - cinfo->marker->next_restart_num)) - return FALSE; - } - - /* Update next-restart state */ - cinfo->marker->next_restart_num = (cinfo->marker->next_restart_num + 1) & 7; - - return TRUE; -} - - -/* - * This is the default resync_to_restart method for data source managers - * to use if they don't have any better approach. Some data source managers - * may be able to back up, or may have additional knowledge about the data - * which permits a more intelligent recovery strategy; such managers would - * presumably supply their own resync method. - * - * read_restart_marker calls resync_to_restart if it finds a marker other than - * the restart marker it was expecting. (This code is *not* used unless - * a nonzero restart interval has been declared.) cinfo->unread_marker is - * the marker code actually found (might be anything, except 0 or FF). - * The desired restart marker number (0..7) is passed as a parameter. - * This routine is supposed to apply whatever error recovery strategy seems - * appropriate in order to position the input stream to the next data segment. - * Note that cinfo->unread_marker is treated as a marker appearing before - * the current data-source input point; usually it should be reset to zero - * before returning. - * Returns FALSE if suspension is required. - * - * This implementation is substantially constrained by wanting to treat the - * input as a data stream; this means we can't back up. Therefore, we have - * only the following actions to work with: - * 1. Simply discard the marker and let the entropy decoder resume at next - * byte of file. - * 2. Read forward until we find another marker, discarding intervening - * data. (In theory we could look ahead within the current bufferload, - * without having to discard data if we don't find the desired marker. - * This idea is not implemented here, in part because it makes behavior - * dependent on buffer size and chance buffer-boundary positions.) - * 3. Leave the marker unread (by failing to zero cinfo->unread_marker). - * This will cause the entropy decoder to process an empty data segment, - * inserting dummy zeroes, and then we will reprocess the marker. - * - * #2 is appropriate if we think the desired marker lies ahead, while #3 is - * appropriate if the found marker is a future restart marker (indicating - * that we have missed the desired restart marker, probably because it got - * corrupted). - * We apply #2 or #3 if the found marker is a restart marker no more than - * two counts behind or ahead of the expected one. We also apply #2 if the - * found marker is not a legal JPEG marker code (it's certainly bogus data). - * If the found marker is a restart marker more than 2 counts away, we do #1 - * (too much risk that the marker is erroneous; with luck we will be able to - * resync at some future point). - * For any valid non-restart JPEG marker, we apply #3. This keeps us from - * overrunning the end of a scan. An implementation limited to single-scan - * files might find it better to apply #2 for markers other than EOI, since - * any other marker would have to be bogus data in that case. - */ - -GLOBAL(boolean) -jpeg_resync_to_restart (j_decompress_ptr cinfo, int desired) -{ - int marker = cinfo->unread_marker; - int action = 1; - - /* Always put up a warning. */ - WARNMS2(cinfo, JWRN_MUST_RESYNC, marker, desired); - - /* Outer loop handles repeated decision after scanning forward. */ - for (;;) { - if (marker < (int) M_SOF0) - action = 2; /* invalid marker */ - else if (marker < (int) M_RST0 || marker > (int) M_RST7) - action = 3; /* valid non-restart marker */ - else { - if (marker == ((int) M_RST0 + ((desired+1) & 7)) || - marker == ((int) M_RST0 + ((desired+2) & 7))) - action = 3; /* one of the next two expected restarts */ - else if (marker == ((int) M_RST0 + ((desired-1) & 7)) || - marker == ((int) M_RST0 + ((desired-2) & 7))) - action = 2; /* a prior restart, so advance */ - else - action = 1; /* desired restart or too far away */ - } - TRACEMS2(cinfo, 4, JTRC_RECOVERY_ACTION, marker, action); - switch (action) { - case 1: - /* Discard marker and let entropy decoder resume processing. */ - cinfo->unread_marker = 0; - return TRUE; - case 2: - /* Scan to the next marker, and repeat the decision loop. */ - if (! next_marker(cinfo)) - return FALSE; - marker = cinfo->unread_marker; - break; - case 3: - /* Return without advancing past this marker. */ - /* Entropy decoder will be forced to process an empty segment. */ - return TRUE; - } - } /* end loop */ -} - - -/* - * Reset marker processing state to begin a fresh datastream. - */ - -METHODDEF(void) -reset_marker_reader (j_decompress_ptr cinfo) -{ - my_marker_ptr marker = (my_marker_ptr) cinfo->marker; - - cinfo->comp_info = NULL; /* until allocated by get_sof */ - cinfo->input_scan_number = 0; /* no SOS seen yet */ - cinfo->unread_marker = 0; /* no pending marker */ - marker->pub.saw_SOI = FALSE; /* set internal state too */ - marker->pub.saw_SOF = FALSE; - marker->pub.discarded_bytes = 0; - marker->cur_marker = NULL; -} - - -/* - * Initialize the marker reader module. - * This is called only once, when the decompression object is created. - */ - -GLOBAL(void) -jinit_marker_reader (j_decompress_ptr cinfo) -{ - my_marker_ptr marker; - int i; - - /* Create subobject in permanent pool */ - marker = (my_marker_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_PERMANENT, SIZEOF(my_marker_reader)); - cinfo->marker = &marker->pub; - /* Initialize public method pointers */ - marker->pub.reset_marker_reader = reset_marker_reader; - marker->pub.read_markers = read_markers; - marker->pub.read_restart_marker = read_restart_marker; - /* Initialize COM/APPn processing. - * By default, we examine and then discard APP0 and APP14, - * but simply discard COM and all other APPn. - */ - marker->process_COM = skip_variable; - marker->length_limit_COM = 0; - for (i = 0; i < 16; i++) { - marker->process_APPn[i] = skip_variable; - marker->length_limit_APPn[i] = 0; - } - marker->process_APPn[0] = get_interesting_appn; - marker->process_APPn[14] = get_interesting_appn; - /* Reset marker processing state */ - reset_marker_reader(cinfo); -} - - -/* - * Control saving of COM and APPn markers into marker_list. - */ - -#ifdef SAVE_MARKERS_SUPPORTED - -GLOBAL(void) -jpeg_save_markers (j_decompress_ptr cinfo, int marker_code, - unsigned int length_limit) -{ - my_marker_ptr marker = (my_marker_ptr) cinfo->marker; - long maxlength; - jpeg_marker_parser_method processor; - - /* Length limit mustn't be larger than what we can allocate - * (should only be a concern in a 16-bit environment). - */ - maxlength = cinfo->mem->max_alloc_chunk - SIZEOF(struct jpeg_marker_struct); - if (((long) length_limit) > maxlength) - length_limit = (unsigned int) maxlength; - - /* Choose processor routine to use. - * APP0/APP14 have special requirements. - */ - if (length_limit) { - processor = save_marker; - /* If saving APP0/APP14, save at least enough for our internal use. */ - if (marker_code == (int) M_APP0 && length_limit < APP0_DATA_LEN) - length_limit = APP0_DATA_LEN; - else if (marker_code == (int) M_APP14 && length_limit < APP14_DATA_LEN) - length_limit = APP14_DATA_LEN; - } else { - processor = skip_variable; - /* If discarding APP0/APP14, use our regular on-the-fly processor. */ - if (marker_code == (int) M_APP0 || marker_code == (int) M_APP14) - processor = get_interesting_appn; - } - - if (marker_code == (int) M_COM) { - marker->process_COM = processor; - marker->length_limit_COM = length_limit; - } else if (marker_code >= (int) M_APP0 && marker_code <= (int) M_APP15) { - marker->process_APPn[marker_code - (int) M_APP0] = processor; - marker->length_limit_APPn[marker_code - (int) M_APP0] = length_limit; - } else - ERREXIT1(cinfo, JERR_UNKNOWN_MARKER, marker_code); -} - -#endif /* SAVE_MARKERS_SUPPORTED */ - - -/* - * Install a special processing method for COM or APPn markers. - */ - -GLOBAL(void) -jpeg_set_marker_processor (j_decompress_ptr cinfo, int marker_code, - jpeg_marker_parser_method routine) -{ - my_marker_ptr marker = (my_marker_ptr) cinfo->marker; - - if (marker_code == (int) M_COM) - marker->process_COM = routine; - else if (marker_code >= (int) M_APP0 && marker_code <= (int) M_APP15) - marker->process_APPn[marker_code - (int) M_APP0] = routine; - else - ERREXIT1(cinfo, JERR_UNKNOWN_MARKER, marker_code); -} diff --git a/dep/libjpeg/src/jdmaster.c b/dep/libjpeg/src/jdmaster.c deleted file mode 100644 index 3070b7bb4..000000000 --- a/dep/libjpeg/src/jdmaster.c +++ /dev/null @@ -1,532 +0,0 @@ -/* - * jdmaster.c - * - * Copyright (C) 1991-1997, Thomas G. Lane. - * Modified 2002-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains master control logic for the JPEG decompressor. - * These routines are concerned with selecting the modules to be executed - * and with determining the number of passes and the work to be done in each - * pass. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Private state */ - -typedef struct { - struct jpeg_decomp_master pub; /* public fields */ - - int pass_number; /* # of passes completed */ - - boolean using_merged_upsample; /* TRUE if using merged upsample/cconvert */ - - /* Saved references to initialized quantizer modules, - * in case we need to switch modes. - */ - struct jpeg_color_quantizer * quantizer_1pass; - struct jpeg_color_quantizer * quantizer_2pass; -} my_decomp_master; - -typedef my_decomp_master * my_master_ptr; - - -/* - * Determine whether merged upsample/color conversion should be used. - * CRUCIAL: this must match the actual capabilities of jdmerge.c! - */ - -LOCAL(boolean) -use_merged_upsample (j_decompress_ptr cinfo) -{ -#ifdef UPSAMPLE_MERGING_SUPPORTED - /* Merging is the equivalent of plain box-filter upsampling. */ - /* The following condition is only needed if fancy shall select - * a different upsampling method. In our current implementation - * fancy only affects the DCT scaling, thus we can use fancy - * upsampling and merged upsample simultaneously, in particular - * with scaled DCT sizes larger than the default DCTSIZE. - */ -#if 0 - if (cinfo->do_fancy_upsampling) - return FALSE; -#endif - if (cinfo->CCIR601_sampling) - return FALSE; - /* jdmerge.c only supports YCC=>RGB color conversion */ - if ((cinfo->jpeg_color_space != JCS_YCbCr && - cinfo->jpeg_color_space != JCS_BG_YCC) || - cinfo->num_components != 3 || - cinfo->out_color_space != JCS_RGB || - cinfo->out_color_components != RGB_PIXELSIZE || - cinfo->color_transform) - return FALSE; - /* and it only handles 2h1v or 2h2v sampling ratios */ - if (cinfo->comp_info[0].h_samp_factor != 2 || - cinfo->comp_info[1].h_samp_factor != 1 || - cinfo->comp_info[2].h_samp_factor != 1 || - cinfo->comp_info[0].v_samp_factor > 2 || - cinfo->comp_info[1].v_samp_factor != 1 || - cinfo->comp_info[2].v_samp_factor != 1) - return FALSE; - /* furthermore, it doesn't work if we've scaled the IDCTs differently */ - if (cinfo->comp_info[0].DCT_h_scaled_size != cinfo->min_DCT_h_scaled_size || - cinfo->comp_info[1].DCT_h_scaled_size != cinfo->min_DCT_h_scaled_size || - cinfo->comp_info[2].DCT_h_scaled_size != cinfo->min_DCT_h_scaled_size || - cinfo->comp_info[0].DCT_v_scaled_size != cinfo->min_DCT_v_scaled_size || - cinfo->comp_info[1].DCT_v_scaled_size != cinfo->min_DCT_v_scaled_size || - cinfo->comp_info[2].DCT_v_scaled_size != cinfo->min_DCT_v_scaled_size) - return FALSE; - /* ??? also need to test for upsample-time rescaling, when & if supported */ - return TRUE; /* by golly, it'll work... */ -#else - return FALSE; -#endif -} - - -/* - * Compute output image dimensions and related values. - * NOTE: this is exported for possible use by application. - * Hence it mustn't do anything that can't be done twice. - * Also note that it may be called before the master module is initialized! - */ - -GLOBAL(void) -jpeg_calc_output_dimensions (j_decompress_ptr cinfo) -/* Do computations that are needed before master selection phase. - * This function is used for full decompression. - */ -{ - int ci, i; - jpeg_component_info *compptr; - - /* Prevent application from calling me at wrong times */ - if (cinfo->global_state != DSTATE_READY) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - - /* Compute core output image dimensions and DCT scaling choices. */ - jpeg_core_output_dimensions(cinfo); - -#ifdef IDCT_SCALING_SUPPORTED - - /* In selecting the actual DCT scaling for each component, we try to - * scale up the chroma components via IDCT scaling rather than upsampling. - * This saves time if the upsampler gets to use 1:1 scaling. - * Note this code adapts subsampling ratios which are powers of 2. - */ - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - int ssize = 1; - if (! cinfo->raw_data_out) - while (cinfo->min_DCT_h_scaled_size * ssize <= - (cinfo->do_fancy_upsampling ? DCTSIZE : DCTSIZE / 2) && - (cinfo->max_h_samp_factor % (compptr->h_samp_factor * ssize * 2)) == - 0) { - ssize = ssize * 2; - } - compptr->DCT_h_scaled_size = cinfo->min_DCT_h_scaled_size * ssize; - ssize = 1; - if (! cinfo->raw_data_out) - while (cinfo->min_DCT_v_scaled_size * ssize <= - (cinfo->do_fancy_upsampling ? DCTSIZE : DCTSIZE / 2) && - (cinfo->max_v_samp_factor % (compptr->v_samp_factor * ssize * 2)) == - 0) { - ssize = ssize * 2; - } - compptr->DCT_v_scaled_size = cinfo->min_DCT_v_scaled_size * ssize; - - /* We don't support IDCT ratios larger than 2. */ - if (compptr->DCT_h_scaled_size > compptr->DCT_v_scaled_size * 2) - compptr->DCT_h_scaled_size = compptr->DCT_v_scaled_size * 2; - else if (compptr->DCT_v_scaled_size > compptr->DCT_h_scaled_size * 2) - compptr->DCT_v_scaled_size = compptr->DCT_h_scaled_size * 2; - - /* Recompute downsampled dimensions of components; - * application needs to know these if using raw downsampled data. - */ - /* Size in samples, after IDCT scaling */ - compptr->downsampled_width = (JDIMENSION) - jdiv_round_up((long) cinfo->image_width * - (long) (compptr->h_samp_factor * compptr->DCT_h_scaled_size), - (long) (cinfo->max_h_samp_factor * cinfo->block_size)); - compptr->downsampled_height = (JDIMENSION) - jdiv_round_up((long) cinfo->image_height * - (long) (compptr->v_samp_factor * compptr->DCT_v_scaled_size), - (long) (cinfo->max_v_samp_factor * cinfo->block_size)); - } - -#endif /* IDCT_SCALING_SUPPORTED */ - - /* Report number of components in selected colorspace. */ - /* This should correspond to the actual code in the color conversion module. */ - switch (cinfo->out_color_space) { - case JCS_GRAYSCALE: - cinfo->out_color_components = 1; - break; - case JCS_RGB: - case JCS_BG_RGB: - cinfo->out_color_components = RGB_PIXELSIZE; - break; - default: /* YCCK <=> CMYK conversion or same colorspace as in file */ - i = 0; - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) - if (compptr->component_needed) - i++; /* count output color components */ - cinfo->out_color_components = i; - } - cinfo->output_components = (cinfo->quantize_colors ? 1 : - cinfo->out_color_components); - - /* See if upsampler will want to emit more than one row at a time */ - if (use_merged_upsample(cinfo)) - cinfo->rec_outbuf_height = cinfo->max_v_samp_factor; - else - cinfo->rec_outbuf_height = 1; -} - - -/* - * Several decompression processes need to range-limit values to the range - * 0..MAXJSAMPLE; the input value may fall somewhat outside this range - * due to noise introduced by quantization, roundoff error, etc. These - * processes are inner loops and need to be as fast as possible. On most - * machines, particularly CPUs with pipelines or instruction prefetch, - * a (subscript-check-less) C table lookup - * x = sample_range_limit[x]; - * is faster than explicit tests - * if (x < 0) x = 0; - * else if (x > MAXJSAMPLE) x = MAXJSAMPLE; - * These processes all use a common table prepared by the routine below. - * - * For most steps we can mathematically guarantee that the initial value - * of x is within 2*(MAXJSAMPLE+1) of the legal range, so a table running - * from -2*(MAXJSAMPLE+1) to 3*MAXJSAMPLE+2 is sufficient. But for the - * initial limiting step (just after the IDCT), a wildly out-of-range value - * is possible if the input data is corrupt. To avoid any chance of indexing - * off the end of memory and getting a bad-pointer trap, we perform the - * post-IDCT limiting thus: - * x = (sample_range_limit - SUBSET)[(x + CENTER) & MASK]; - * where MASK is 2 bits wider than legal sample data, ie 10 bits for 8-bit - * samples. Under normal circumstances this is more than enough range and - * a correct output will be generated; with bogus input data the mask will - * cause wraparound, and we will safely generate a bogus-but-in-range output. - * For the post-IDCT step, we want to convert the data from signed to unsigned - * representation by adding CENTERJSAMPLE at the same time that we limit it. - * This is accomplished with SUBSET = CENTER - CENTERJSAMPLE. - * - * Note that the table is allocated in near data space on PCs; it's small - * enough and used often enough to justify this. - */ - -LOCAL(void) -prepare_range_limit_table (j_decompress_ptr cinfo) -/* Allocate and fill in the sample_range_limit table */ -{ - JSAMPLE * table; - int i; - - table = (JSAMPLE *) (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, - JPOOL_IMAGE, (RANGE_CENTER * 2 + MAXJSAMPLE + 1) * SIZEOF(JSAMPLE)); - /* First segment of range limit table: limit[x] = 0 for x < 0 */ - MEMZERO(table, RANGE_CENTER * SIZEOF(JSAMPLE)); - table += RANGE_CENTER; /* allow negative subscripts of table */ - cinfo->sample_range_limit = table; - /* Main part of range limit table: limit[x] = x */ - for (i = 0; i <= MAXJSAMPLE; i++) - table[i] = (JSAMPLE) i; - /* End of range limit table: limit[x] = MAXJSAMPLE for x > MAXJSAMPLE */ - for (; i <= MAXJSAMPLE + RANGE_CENTER; i++) - table[i] = MAXJSAMPLE; -} - - -/* - * Master selection of decompression modules. - * This is done once at jpeg_start_decompress time. We determine - * which modules will be used and give them appropriate initialization calls. - * We also initialize the decompressor input side to begin consuming data. - * - * Since jpeg_read_header has finished, we know what is in the SOF - * and (first) SOS markers. We also have all the application parameter - * settings. - */ - -LOCAL(void) -master_selection (j_decompress_ptr cinfo) -{ - my_master_ptr master = (my_master_ptr) cinfo->master; - boolean use_c_buffer; - long samplesperrow; - JDIMENSION jd_samplesperrow; - - /* For now, precision must match compiled-in value... */ - if (cinfo->data_precision != BITS_IN_JSAMPLE) - ERREXIT1(cinfo, JERR_BAD_PRECISION, cinfo->data_precision); - - /* Initialize dimensions and other stuff */ - jpeg_calc_output_dimensions(cinfo); - prepare_range_limit_table(cinfo); - - /* Sanity check on image dimensions */ - if (cinfo->output_height <= 0 || cinfo->output_width <= 0 || - cinfo->out_color_components <= 0) - ERREXIT(cinfo, JERR_EMPTY_IMAGE); - - /* Width of an output scanline must be representable as JDIMENSION. */ - samplesperrow = (long) cinfo->output_width * (long) cinfo->out_color_components; - jd_samplesperrow = (JDIMENSION) samplesperrow; - if ((long) jd_samplesperrow != samplesperrow) - ERREXIT(cinfo, JERR_WIDTH_OVERFLOW); - - /* Initialize my private state */ - master->pass_number = 0; - master->using_merged_upsample = use_merged_upsample(cinfo); - - /* Color quantizer selection */ - master->quantizer_1pass = NULL; - master->quantizer_2pass = NULL; - /* No mode changes if not using buffered-image mode. */ - if (! cinfo->quantize_colors || ! cinfo->buffered_image) { - cinfo->enable_1pass_quant = FALSE; - cinfo->enable_external_quant = FALSE; - cinfo->enable_2pass_quant = FALSE; - } - if (cinfo->quantize_colors) { - if (cinfo->raw_data_out) - ERREXIT(cinfo, JERR_NOTIMPL); - /* 2-pass quantizer only works in 3-component color space. */ - if (cinfo->out_color_components != 3) { - cinfo->enable_1pass_quant = TRUE; - cinfo->enable_external_quant = FALSE; - cinfo->enable_2pass_quant = FALSE; - cinfo->colormap = NULL; - } else if (cinfo->colormap != NULL) { - cinfo->enable_external_quant = TRUE; - } else if (cinfo->two_pass_quantize) { - cinfo->enable_2pass_quant = TRUE; - } else { - cinfo->enable_1pass_quant = TRUE; - } - - if (cinfo->enable_1pass_quant) { -#ifdef QUANT_1PASS_SUPPORTED - jinit_1pass_quantizer(cinfo); - master->quantizer_1pass = cinfo->cquantize; -#else - ERREXIT(cinfo, JERR_NOT_COMPILED); -#endif - } - - /* We use the 2-pass code to map to external colormaps. */ - if (cinfo->enable_2pass_quant || cinfo->enable_external_quant) { -#ifdef QUANT_2PASS_SUPPORTED - jinit_2pass_quantizer(cinfo); - master->quantizer_2pass = cinfo->cquantize; -#else - ERREXIT(cinfo, JERR_NOT_COMPILED); -#endif - } - /* If both quantizers are initialized, the 2-pass one is left active; - * this is necessary for starting with quantization to an external map. - */ - } - - /* Post-processing: in particular, color conversion first */ - if (! cinfo->raw_data_out) { - if (master->using_merged_upsample) { -#ifdef UPSAMPLE_MERGING_SUPPORTED - jinit_merged_upsampler(cinfo); /* does color conversion too */ -#else - ERREXIT(cinfo, JERR_NOT_COMPILED); -#endif - } else { - jinit_color_deconverter(cinfo); - jinit_upsampler(cinfo); - } - jinit_d_post_controller(cinfo, cinfo->enable_2pass_quant); - } - /* Inverse DCT */ - jinit_inverse_dct(cinfo); - /* Entropy decoding: either Huffman or arithmetic coding. */ - if (cinfo->arith_code) - jinit_arith_decoder(cinfo); - else { - jinit_huff_decoder(cinfo); - } - - /* Initialize principal buffer controllers. */ - use_c_buffer = cinfo->inputctl->has_multiple_scans || cinfo->buffered_image; - jinit_d_coef_controller(cinfo, use_c_buffer); - - if (! cinfo->raw_data_out) - jinit_d_main_controller(cinfo, FALSE /* never need full buffer here */); - - /* We can now tell the memory manager to allocate virtual arrays. */ - (*cinfo->mem->realize_virt_arrays) ((j_common_ptr) cinfo); - - /* Initialize input side of decompressor to consume first scan. */ - (*cinfo->inputctl->start_input_pass) (cinfo); - -#ifdef D_MULTISCAN_FILES_SUPPORTED - /* If jpeg_start_decompress will read the whole file, initialize - * progress monitoring appropriately. The input step is counted - * as one pass. - */ - if (cinfo->progress != NULL && ! cinfo->buffered_image && - cinfo->inputctl->has_multiple_scans) { - int nscans; - /* Estimate number of scans to set pass_limit. */ - if (cinfo->progressive_mode) { - /* Arbitrarily estimate 2 interleaved DC scans + 3 AC scans/component. */ - nscans = 2 + 3 * cinfo->num_components; - } else { - /* For a nonprogressive multiscan file, estimate 1 scan per component. */ - nscans = cinfo->num_components; - } - cinfo->progress->pass_counter = 0L; - cinfo->progress->pass_limit = (long) cinfo->total_iMCU_rows * nscans; - cinfo->progress->completed_passes = 0; - cinfo->progress->total_passes = (cinfo->enable_2pass_quant ? 3 : 2); - /* Count the input pass as done */ - master->pass_number++; - } -#endif /* D_MULTISCAN_FILES_SUPPORTED */ -} - - -/* - * Per-pass setup. - * This is called at the beginning of each output pass. We determine which - * modules will be active during this pass and give them appropriate - * start_pass calls. We also set is_dummy_pass to indicate whether this - * is a "real" output pass or a dummy pass for color quantization. - * (In the latter case, jdapistd.c will crank the pass to completion.) - */ - -METHODDEF(void) -prepare_for_output_pass (j_decompress_ptr cinfo) -{ - my_master_ptr master = (my_master_ptr) cinfo->master; - - if (master->pub.is_dummy_pass) { -#ifdef QUANT_2PASS_SUPPORTED - /* Final pass of 2-pass quantization */ - master->pub.is_dummy_pass = FALSE; - (*cinfo->cquantize->start_pass) (cinfo, FALSE); - (*cinfo->post->start_pass) (cinfo, JBUF_CRANK_DEST); - (*cinfo->main->start_pass) (cinfo, JBUF_CRANK_DEST); -#else - ERREXIT(cinfo, JERR_NOT_COMPILED); -#endif /* QUANT_2PASS_SUPPORTED */ - } else { - if (cinfo->quantize_colors && cinfo->colormap == NULL) { - /* Select new quantization method */ - if (cinfo->two_pass_quantize && cinfo->enable_2pass_quant) { - cinfo->cquantize = master->quantizer_2pass; - master->pub.is_dummy_pass = TRUE; - } else if (cinfo->enable_1pass_quant) { - cinfo->cquantize = master->quantizer_1pass; - } else { - ERREXIT(cinfo, JERR_MODE_CHANGE); - } - } - (*cinfo->idct->start_pass) (cinfo); - (*cinfo->coef->start_output_pass) (cinfo); - if (! cinfo->raw_data_out) { - if (! master->using_merged_upsample) - (*cinfo->cconvert->start_pass) (cinfo); - (*cinfo->upsample->start_pass) (cinfo); - if (cinfo->quantize_colors) - (*cinfo->cquantize->start_pass) (cinfo, master->pub.is_dummy_pass); - (*cinfo->post->start_pass) (cinfo, - (master->pub.is_dummy_pass ? JBUF_SAVE_AND_PASS : JBUF_PASS_THRU)); - (*cinfo->main->start_pass) (cinfo, JBUF_PASS_THRU); - } - } - - /* Set up progress monitor's pass info if present */ - if (cinfo->progress != NULL) { - cinfo->progress->completed_passes = master->pass_number; - cinfo->progress->total_passes = master->pass_number + - (master->pub.is_dummy_pass ? 2 : 1); - /* In buffered-image mode, we assume one more output pass if EOI not - * yet reached, but no more passes if EOI has been reached. - */ - if (cinfo->buffered_image && ! cinfo->inputctl->eoi_reached) { - cinfo->progress->total_passes += (cinfo->enable_2pass_quant ? 2 : 1); - } - } -} - - -/* - * Finish up at end of an output pass. - */ - -METHODDEF(void) -finish_output_pass (j_decompress_ptr cinfo) -{ - my_master_ptr master = (my_master_ptr) cinfo->master; - - if (cinfo->quantize_colors) - (*cinfo->cquantize->finish_pass) (cinfo); - master->pass_number++; -} - - -#ifdef D_MULTISCAN_FILES_SUPPORTED - -/* - * Switch to a new external colormap between output passes. - */ - -GLOBAL(void) -jpeg_new_colormap (j_decompress_ptr cinfo) -{ - my_master_ptr master = (my_master_ptr) cinfo->master; - - /* Prevent application from calling me at wrong times */ - if (cinfo->global_state != DSTATE_BUFIMAGE) - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - - if (cinfo->quantize_colors && cinfo->enable_external_quant && - cinfo->colormap != NULL) { - /* Select 2-pass quantizer for external colormap use */ - cinfo->cquantize = master->quantizer_2pass; - /* Notify quantizer of colormap change */ - (*cinfo->cquantize->new_color_map) (cinfo); - master->pub.is_dummy_pass = FALSE; /* just in case */ - } else - ERREXIT(cinfo, JERR_MODE_CHANGE); -} - -#endif /* D_MULTISCAN_FILES_SUPPORTED */ - - -/* - * Initialize master decompression control and select active modules. - * This is performed at the start of jpeg_start_decompress. - */ - -GLOBAL(void) -jinit_master_decompress (j_decompress_ptr cinfo) -{ - my_master_ptr master; - - master = (my_master_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_decomp_master)); - cinfo->master = &master->pub; - master->pub.prepare_for_output_pass = prepare_for_output_pass; - master->pub.finish_output_pass = finish_output_pass; - - master->pub.is_dummy_pass = FALSE; - - master_selection(cinfo); -} diff --git a/dep/libjpeg/src/jdmerge.c b/dep/libjpeg/src/jdmerge.c deleted file mode 100644 index 0d16821be..000000000 --- a/dep/libjpeg/src/jdmerge.c +++ /dev/null @@ -1,437 +0,0 @@ -/* - * jdmerge.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2013-2022 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains code for merged upsampling/color conversion. - * - * This file combines functions from jdsample.c and jdcolor.c; - * read those files first to understand what's going on. - * - * When the chroma components are to be upsampled by simple replication - * (ie, box filtering), we can save some work in color conversion by - * calculating all the output pixels corresponding to a pair of chroma - * samples at one time. In the conversion equations - * R = Y + K1 * Cr - * G = Y + K2 * Cb + K3 * Cr - * B = Y + K4 * Cb - * only the Y term varies among the group of pixels corresponding to a pair - * of chroma samples, so the rest of the terms can be calculated just once. - * At typical sampling ratios, this eliminates half or three-quarters - * of the multiplications needed for color conversion. - * - * This file currently provides implementations for the following cases: - * YCC => RGB color conversion only (YCbCr or BG_YCC). - * Sampling ratios of 2h1v or 2h2v. - * No scaling needed at upsample time. - * Corner-aligned (non-CCIR601) sampling alignment. - * Other special cases could be added, but in most applications these - * are the only common cases. (For uncommon cases we fall back on - * the more general code in jdsample.c and jdcolor.c.) - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - -#ifdef UPSAMPLE_MERGING_SUPPORTED - - -#if RANGE_BITS < 2 - /* Deliberate syntax err */ - Sorry, this code requires 2 or more range extension bits. -#endif - - -/* Private subobject */ - -typedef struct { - struct jpeg_upsampler pub; /* public fields */ - - /* Pointer to routine to do actual upsampling/conversion of one row group */ - JMETHOD(void, upmethod, (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr, - JSAMPARRAY output_buf)); - - /* Private state for YCC->RGB conversion */ - int * Cr_r_tab; /* => table for Cr to R conversion */ - int * Cb_b_tab; /* => table for Cb to B conversion */ - INT32 * Cr_g_tab; /* => table for Cr to G conversion */ - INT32 * Cb_g_tab; /* => table for Cb to G conversion */ - - /* For 2:1 vertical sampling, we produce two output rows at a time. - * We need a "spare" row buffer to hold the second output row if the - * application provides just a one-row buffer; we also use the spare - * to discard the dummy last row if the image height is odd. - */ - JSAMPROW spare_row; - boolean spare_full; /* T if spare buffer is occupied */ - - JDIMENSION out_row_width; /* samples per output row */ - JDIMENSION rows_to_go; /* counts rows remaining in image */ -} my_upsampler; - -typedef my_upsampler * my_upsample_ptr; - -#define SCALEBITS 16 /* speediest right-shift on some machines */ -#define ONE_HALF ((INT32) 1 << (SCALEBITS-1)) -#define FIX(x) ((INT32) ((x) * (1L<RGB and BG_YCC->RGB colorspace conversion. - * This is taken directly from jdcolor.c; see that file for more info. - */ - -LOCAL(void) -build_ycc_rgb_table (j_decompress_ptr cinfo) -/* Normal case, sYCC */ -{ - my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample; - int i; - INT32 x; - SHIFT_TEMPS - - upsample->Cr_r_tab = (int *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(int)); - upsample->Cb_b_tab = (int *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(int)); - upsample->Cr_g_tab = (INT32 *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(INT32)); - upsample->Cb_g_tab = (INT32 *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(INT32)); - - for (i = 0, x = -CENTERJSAMPLE; i <= MAXJSAMPLE; i++, x++) { - /* i is the actual input pixel value, in the range 0..MAXJSAMPLE */ - /* The Cb or Cr value we are thinking of is x = i - CENTERJSAMPLE */ - /* Cr=>R value is nearest int to 1.402 * x */ - upsample->Cr_r_tab[i] = (int) DESCALE(FIX(1.402) * x, SCALEBITS); - /* Cb=>B value is nearest int to 1.772 * x */ - upsample->Cb_b_tab[i] = (int) DESCALE(FIX(1.772) * x, SCALEBITS); - /* Cr=>G value is scaled-up -0.714136286 * x */ - upsample->Cr_g_tab[i] = (- FIX(0.714136286)) * x; - /* Cb=>G value is scaled-up -0.344136286 * x */ - /* We also add in ONE_HALF so that need not do it in inner loop */ - upsample->Cb_g_tab[i] = (- FIX(0.344136286)) * x + ONE_HALF; - } -} - - -LOCAL(void) -build_bg_ycc_rgb_table (j_decompress_ptr cinfo) -/* Wide gamut case, bg-sYCC */ -{ - my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample; - int i; - INT32 x; - SHIFT_TEMPS - - upsample->Cr_r_tab = (int *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(int)); - upsample->Cb_b_tab = (int *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(int)); - upsample->Cr_g_tab = (INT32 *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(INT32)); - upsample->Cb_g_tab = (INT32 *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE+1) * SIZEOF(INT32)); - - for (i = 0, x = -CENTERJSAMPLE; i <= MAXJSAMPLE; i++, x++) { - /* i is the actual input pixel value, in the range 0..MAXJSAMPLE */ - /* The Cb or Cr value we are thinking of is x = i - CENTERJSAMPLE */ - /* Cr=>R value is nearest int to 2.804 * x */ - upsample->Cr_r_tab[i] = (int) DESCALE(FIX(2.804) * x, SCALEBITS); - /* Cb=>B value is nearest int to 3.544 * x */ - upsample->Cb_b_tab[i] = (int) DESCALE(FIX(3.544) * x, SCALEBITS); - /* Cr=>G value is scaled-up -1.428272572 * x */ - upsample->Cr_g_tab[i] = (- FIX(1.428272572)) * x; - /* Cb=>G value is scaled-up -0.688272572 * x */ - /* We also add in ONE_HALF so that need not do it in inner loop */ - upsample->Cb_g_tab[i] = (- FIX(0.688272572)) * x + ONE_HALF; - } -} - - -/* - * Initialize for an upsampling pass. - */ - -METHODDEF(void) -start_pass_merged_upsample (j_decompress_ptr cinfo) -{ - my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample; - - /* Mark the spare buffer empty */ - upsample->spare_full = FALSE; - /* Initialize total-height counter for detecting bottom of image */ - upsample->rows_to_go = cinfo->output_height; -} - - -/* - * Control routine to do upsampling (and color conversion). - * - * The control routine just handles the row buffering considerations. - */ - -METHODDEF(void) -merged_2v_upsample (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr, - JDIMENSION in_row_groups_avail, - JSAMPARRAY output_buf, JDIMENSION *out_row_ctr, - JDIMENSION out_rows_avail) -/* 2:1 vertical sampling case: may need a spare row. */ -{ - my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample; - JSAMPROW work_ptrs[2]; - JDIMENSION num_rows; /* number of rows returned to caller */ - - if (upsample->spare_full) { - /* If we have a spare row saved from a previous cycle, just return it. */ - jcopy_sample_rows(& upsample->spare_row, output_buf + *out_row_ctr, - 1, upsample->out_row_width); - num_rows = 1; - upsample->spare_full = FALSE; - } else { - /* Figure number of rows to return to caller. */ - num_rows = 2; - /* Not more than the distance to the end of the image. */ - if (num_rows > upsample->rows_to_go) - num_rows = upsample->rows_to_go; - /* And not more than what the client can accept: */ - out_rows_avail -= *out_row_ctr; - if (num_rows > out_rows_avail) - num_rows = out_rows_avail; - /* Create output pointer array for upsampler. */ - work_ptrs[0] = output_buf[*out_row_ctr]; - if (num_rows > 1) { - work_ptrs[1] = output_buf[*out_row_ctr + 1]; - } else { - work_ptrs[1] = upsample->spare_row; - upsample->spare_full = TRUE; - } - /* Now do the upsampling. */ - (*upsample->upmethod) (cinfo, input_buf, *in_row_group_ctr, work_ptrs); - } - - /* Adjust counts */ - *out_row_ctr += num_rows; - upsample->rows_to_go -= num_rows; - /* When the buffer is emptied, declare this input row group consumed */ - if (! upsample->spare_full) - (*in_row_group_ctr)++; -} - - -METHODDEF(void) -merged_1v_upsample (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr, - JDIMENSION in_row_groups_avail, - JSAMPARRAY output_buf, JDIMENSION *out_row_ctr, - JDIMENSION out_rows_avail) -/* 1:1 vertical sampling case: much easier, never need a spare row. */ -{ - my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample; - - /* Just do the upsampling. */ - (*upsample->upmethod) (cinfo, input_buf, *in_row_group_ctr, - output_buf + *out_row_ctr); - /* Adjust counts */ - (*out_row_ctr)++; - (*in_row_group_ctr)++; -} - - -/* - * These are the routines invoked by the control routines to do - * the actual upsampling/conversion. One row group is processed per call. - * - * Note: since we may be writing directly into application-supplied buffers, - * we have to be honest about the output width; we can't assume the buffer - * has been rounded up to an even width. - */ - - -/* - * Upsample and color convert for the case of 2:1 horizontal and 1:1 vertical. - */ - -METHODDEF(void) -h2v1_merged_upsample (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr, - JSAMPARRAY output_buf) -{ - my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample; - register int y, cred, cgreen, cblue; - int cb, cr; - register JSAMPROW outptr; - JSAMPROW inptr0, inptr1, inptr2; - JDIMENSION col; - /* copy these pointers into registers if possible */ - register JSAMPLE * range_limit = cinfo->sample_range_limit; - int * Crrtab = upsample->Cr_r_tab; - int * Cbbtab = upsample->Cb_b_tab; - INT32 * Crgtab = upsample->Cr_g_tab; - INT32 * Cbgtab = upsample->Cb_g_tab; - SHIFT_TEMPS - - inptr0 = input_buf[0][in_row_group_ctr]; - inptr1 = input_buf[1][in_row_group_ctr]; - inptr2 = input_buf[2][in_row_group_ctr]; - outptr = output_buf[0]; - /* Loop for each pair of output pixels */ - for (col = cinfo->output_width >> 1; col > 0; col--) { - /* Do the chroma part of the calculation */ - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); - cgreen = (int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); - cblue = Cbbtab[cb]; - cred = Crrtab[cr]; - /* Fetch 2 Y values and emit 2 pixels */ - y = GETJSAMPLE(*inptr0++); - outptr[RGB_RED] = range_limit[y + cred]; - outptr[RGB_GREEN] = range_limit[y + cgreen]; - outptr[RGB_BLUE] = range_limit[y + cblue]; - outptr += RGB_PIXELSIZE; - y = GETJSAMPLE(*inptr0++); - outptr[RGB_RED] = range_limit[y + cred]; - outptr[RGB_GREEN] = range_limit[y + cgreen]; - outptr[RGB_BLUE] = range_limit[y + cblue]; - outptr += RGB_PIXELSIZE; - } - /* If image width is odd, do the last output column separately */ - if (cinfo->output_width & 1) { - y = GETJSAMPLE(*inptr0); - cb = GETJSAMPLE(*inptr1); - cr = GETJSAMPLE(*inptr2); - outptr[RGB_RED] = range_limit[y + Crrtab[cr]]; - outptr[RGB_GREEN] = range_limit[y + - ((int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], - SCALEBITS))]; - outptr[RGB_BLUE] = range_limit[y + Cbbtab[cb]]; - } -} - - -/* - * Upsample and color convert for the case of 2:1 horizontal and 2:1 vertical. - */ - -METHODDEF(void) -h2v2_merged_upsample (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION in_row_group_ctr, - JSAMPARRAY output_buf) -{ - my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample; - register int y, cred, cgreen, cblue; - int cb, cr; - register JSAMPROW outptr0, outptr1; - JSAMPROW inptr00, inptr01, inptr1, inptr2; - JDIMENSION col; - /* copy these pointers into registers if possible */ - register JSAMPLE * range_limit = cinfo->sample_range_limit; - int * Crrtab = upsample->Cr_r_tab; - int * Cbbtab = upsample->Cb_b_tab; - INT32 * Crgtab = upsample->Cr_g_tab; - INT32 * Cbgtab = upsample->Cb_g_tab; - SHIFT_TEMPS - - inptr00 = input_buf[0][in_row_group_ctr*2]; - inptr01 = input_buf[0][in_row_group_ctr*2 + 1]; - inptr1 = input_buf[1][in_row_group_ctr]; - inptr2 = input_buf[2][in_row_group_ctr]; - outptr0 = output_buf[0]; - outptr1 = output_buf[1]; - /* Loop for each group of output pixels */ - for (col = cinfo->output_width >> 1; col > 0; col--) { - /* Do the chroma part of the calculation */ - cb = GETJSAMPLE(*inptr1++); - cr = GETJSAMPLE(*inptr2++); - cgreen = (int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); - cblue = Cbbtab[cb]; - cred = Crrtab[cr]; - /* Fetch 4 Y values and emit 4 pixels */ - y = GETJSAMPLE(*inptr00++); - outptr0[RGB_RED] = range_limit[y + cred]; - outptr0[RGB_GREEN] = range_limit[y + cgreen]; - outptr0[RGB_BLUE] = range_limit[y + cblue]; - outptr0 += RGB_PIXELSIZE; - y = GETJSAMPLE(*inptr00++); - outptr0[RGB_RED] = range_limit[y + cred]; - outptr0[RGB_GREEN] = range_limit[y + cgreen]; - outptr0[RGB_BLUE] = range_limit[y + cblue]; - outptr0 += RGB_PIXELSIZE; - y = GETJSAMPLE(*inptr01++); - outptr1[RGB_RED] = range_limit[y + cred]; - outptr1[RGB_GREEN] = range_limit[y + cgreen]; - outptr1[RGB_BLUE] = range_limit[y + cblue]; - outptr1 += RGB_PIXELSIZE; - y = GETJSAMPLE(*inptr01++); - outptr1[RGB_RED] = range_limit[y + cred]; - outptr1[RGB_GREEN] = range_limit[y + cgreen]; - outptr1[RGB_BLUE] = range_limit[y + cblue]; - outptr1 += RGB_PIXELSIZE; - } - /* If image width is odd, do the last output column separately */ - if (cinfo->output_width & 1) { - cb = GETJSAMPLE(*inptr1); - cr = GETJSAMPLE(*inptr2); - cgreen = (int) RIGHT_SHIFT(Cbgtab[cb] + Crgtab[cr], SCALEBITS); - cblue = Cbbtab[cb]; - cred = Crrtab[cr]; - y = GETJSAMPLE(*inptr00); - outptr0[RGB_RED] = range_limit[y + cred]; - outptr0[RGB_GREEN] = range_limit[y + cgreen]; - outptr0[RGB_BLUE] = range_limit[y + cblue]; - y = GETJSAMPLE(*inptr01); - outptr1[RGB_RED] = range_limit[y + cred]; - outptr1[RGB_GREEN] = range_limit[y + cgreen]; - outptr1[RGB_BLUE] = range_limit[y + cblue]; - } -} - - -/* - * Module initialization routine for merged upsampling/color conversion. - * - * NB: this is called under the conditions determined by use_merged_upsample() - * in jdmaster.c. That routine MUST correspond to the actual capabilities - * of this module; no safety checks are made here. - */ - -GLOBAL(void) -jinit_merged_upsampler (j_decompress_ptr cinfo) -{ - my_upsample_ptr upsample; - - upsample = (my_upsample_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_upsampler)); - cinfo->upsample = &upsample->pub; - upsample->pub.start_pass = start_pass_merged_upsample; - upsample->pub.need_context_rows = FALSE; - - upsample->out_row_width = cinfo->output_width * cinfo->out_color_components; - - if (cinfo->max_v_samp_factor == 2) { - upsample->pub.upsample = merged_2v_upsample; - upsample->upmethod = h2v2_merged_upsample; - /* Allocate a spare row buffer */ - upsample->spare_row = (JSAMPROW) (*cinfo->mem->alloc_large) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - (size_t) upsample->out_row_width * SIZEOF(JSAMPLE)); - } else { - upsample->pub.upsample = merged_1v_upsample; - upsample->upmethod = h2v1_merged_upsample; - /* No spare row needed */ - upsample->spare_row = NULL; - } - - if (cinfo->jpeg_color_space == JCS_BG_YCC) - build_bg_ycc_rgb_table(cinfo); - else - build_ycc_rgb_table(cinfo); -} - -#endif /* UPSAMPLE_MERGING_SUPPORTED */ diff --git a/dep/libjpeg/src/jdpostct.c b/dep/libjpeg/src/jdpostct.c deleted file mode 100644 index 571563d72..000000000 --- a/dep/libjpeg/src/jdpostct.c +++ /dev/null @@ -1,290 +0,0 @@ -/* - * jdpostct.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains the decompression postprocessing controller. - * This controller manages the upsampling, color conversion, and color - * quantization/reduction steps; specifically, it controls the buffering - * between upsample/color conversion and color quantization/reduction. - * - * If no color quantization/reduction is required, then this module has no - * work to do, and it just hands off to the upsample/color conversion code. - * An integrated upsample/convert/quantize process would replace this module - * entirely. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Private buffer controller object */ - -typedef struct { - struct jpeg_d_post_controller pub; /* public fields */ - - /* Color quantization source buffer: this holds output data from - * the upsample/color conversion step to be passed to the quantizer. - * For two-pass color quantization, we need a full-image buffer; - * for one-pass operation, a strip buffer is sufficient. - */ - jvirt_sarray_ptr whole_image; /* virtual array, or NULL if one-pass */ - JSAMPARRAY buffer; /* strip buffer, or current strip of virtual */ - JDIMENSION strip_height; /* buffer size in rows */ - /* for two-pass mode only: */ - JDIMENSION starting_row; /* row # of first row in current strip */ - JDIMENSION next_row; /* index of next row to fill/empty in strip */ -} my_post_controller; - -typedef my_post_controller * my_post_ptr; - - -/* Forward declarations */ -METHODDEF(void) post_process_1pass - JPP((j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr, - JDIMENSION in_row_groups_avail, - JSAMPARRAY output_buf, JDIMENSION *out_row_ctr, - JDIMENSION out_rows_avail)); -#ifdef QUANT_2PASS_SUPPORTED -METHODDEF(void) post_process_prepass - JPP((j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr, - JDIMENSION in_row_groups_avail, - JSAMPARRAY output_buf, JDIMENSION *out_row_ctr, - JDIMENSION out_rows_avail)); -METHODDEF(void) post_process_2pass - JPP((j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr, - JDIMENSION in_row_groups_avail, - JSAMPARRAY output_buf, JDIMENSION *out_row_ctr, - JDIMENSION out_rows_avail)); -#endif - - -/* - * Initialize for a processing pass. - */ - -METHODDEF(void) -start_pass_dpost (j_decompress_ptr cinfo, J_BUF_MODE pass_mode) -{ - my_post_ptr post = (my_post_ptr) cinfo->post; - - switch (pass_mode) { - case JBUF_PASS_THRU: - if (cinfo->quantize_colors) { - /* Single-pass processing with color quantization. */ - post->pub.post_process_data = post_process_1pass; - /* We could be doing buffered-image output before starting a 2-pass - * color quantization; in that case, jinit_d_post_controller did not - * allocate a strip buffer. Use the virtual-array buffer as workspace. - */ - if (post->buffer == NULL) { - post->buffer = (*cinfo->mem->access_virt_sarray) - ((j_common_ptr) cinfo, post->whole_image, - (JDIMENSION) 0, post->strip_height, TRUE); - } - } else { - /* For single-pass processing without color quantization, - * I have no work to do; just call the upsampler directly. - */ - post->pub.post_process_data = cinfo->upsample->upsample; - } - break; -#ifdef QUANT_2PASS_SUPPORTED - case JBUF_SAVE_AND_PASS: - /* First pass of 2-pass quantization */ - if (post->whole_image == NULL) - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - post->pub.post_process_data = post_process_prepass; - break; - case JBUF_CRANK_DEST: - /* Second pass of 2-pass quantization */ - if (post->whole_image == NULL) - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - post->pub.post_process_data = post_process_2pass; - break; -#endif /* QUANT_2PASS_SUPPORTED */ - default: - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); - break; - } - post->starting_row = post->next_row = 0; -} - - -/* - * Process some data in the one-pass (strip buffer) case. - * This is used for color precision reduction as well as one-pass quantization. - */ - -METHODDEF(void) -post_process_1pass (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr, - JDIMENSION in_row_groups_avail, - JSAMPARRAY output_buf, JDIMENSION *out_row_ctr, - JDIMENSION out_rows_avail) -{ - my_post_ptr post = (my_post_ptr) cinfo->post; - JDIMENSION num_rows, max_rows; - - /* Fill the buffer, but not more than what we can dump out in one go. */ - /* Note we rely on the upsampler to detect bottom of image. */ - max_rows = out_rows_avail - *out_row_ctr; - if (max_rows > post->strip_height) - max_rows = post->strip_height; - num_rows = 0; - (*cinfo->upsample->upsample) (cinfo, - input_buf, in_row_group_ctr, in_row_groups_avail, - post->buffer, &num_rows, max_rows); - /* Quantize and emit data. */ - (*cinfo->cquantize->color_quantize) (cinfo, - post->buffer, output_buf + *out_row_ctr, (int) num_rows); - *out_row_ctr += num_rows; -} - - -#ifdef QUANT_2PASS_SUPPORTED - -/* - * Process some data in the first pass of 2-pass quantization. - */ - -METHODDEF(void) -post_process_prepass (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr, - JDIMENSION in_row_groups_avail, - JSAMPARRAY output_buf, JDIMENSION *out_row_ctr, - JDIMENSION out_rows_avail) -{ - my_post_ptr post = (my_post_ptr) cinfo->post; - JDIMENSION old_next_row, num_rows; - - /* Reposition virtual buffer if at start of strip. */ - if (post->next_row == 0) { - post->buffer = (*cinfo->mem->access_virt_sarray) - ((j_common_ptr) cinfo, post->whole_image, - post->starting_row, post->strip_height, TRUE); - } - - /* Upsample some data (up to a strip height's worth). */ - old_next_row = post->next_row; - (*cinfo->upsample->upsample) (cinfo, - input_buf, in_row_group_ctr, in_row_groups_avail, - post->buffer, &post->next_row, post->strip_height); - - /* Allow quantizer to scan new data. No data is emitted, */ - /* but we advance out_row_ctr so outer loop can tell when we're done. */ - if (post->next_row > old_next_row) { - num_rows = post->next_row - old_next_row; - (*cinfo->cquantize->color_quantize) (cinfo, post->buffer + old_next_row, - (JSAMPARRAY) NULL, (int) num_rows); - *out_row_ctr += num_rows; - } - - /* Advance if we filled the strip. */ - if (post->next_row >= post->strip_height) { - post->starting_row += post->strip_height; - post->next_row = 0; - } -} - - -/* - * Process some data in the second pass of 2-pass quantization. - */ - -METHODDEF(void) -post_process_2pass (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr, - JDIMENSION in_row_groups_avail, - JSAMPARRAY output_buf, JDIMENSION *out_row_ctr, - JDIMENSION out_rows_avail) -{ - my_post_ptr post = (my_post_ptr) cinfo->post; - JDIMENSION num_rows, max_rows; - - /* Reposition virtual buffer if at start of strip. */ - if (post->next_row == 0) { - post->buffer = (*cinfo->mem->access_virt_sarray) - ((j_common_ptr) cinfo, post->whole_image, - post->starting_row, post->strip_height, FALSE); - } - - /* Determine number of rows to emit. */ - num_rows = post->strip_height - post->next_row; /* available in strip */ - max_rows = out_rows_avail - *out_row_ctr; /* available in output area */ - if (num_rows > max_rows) - num_rows = max_rows; - /* We have to check bottom of image here, can't depend on upsampler. */ - max_rows = cinfo->output_height - post->starting_row; - if (num_rows > max_rows) - num_rows = max_rows; - - /* Quantize and emit data. */ - (*cinfo->cquantize->color_quantize) (cinfo, - post->buffer + post->next_row, output_buf + *out_row_ctr, - (int) num_rows); - *out_row_ctr += num_rows; - - /* Advance if we filled the strip. */ - post->next_row += num_rows; - if (post->next_row >= post->strip_height) { - post->starting_row += post->strip_height; - post->next_row = 0; - } -} - -#endif /* QUANT_2PASS_SUPPORTED */ - - -/* - * Initialize postprocessing controller. - */ - -GLOBAL(void) -jinit_d_post_controller (j_decompress_ptr cinfo, boolean need_full_buffer) -{ - my_post_ptr post; - - post = (my_post_ptr) - (*cinfo->mem->alloc_small) ((j_common_ptr) cinfo, JPOOL_IMAGE, - SIZEOF(my_post_controller)); - cinfo->post = (struct jpeg_d_post_controller *) post; - post->pub.start_pass = start_pass_dpost; - post->whole_image = NULL; /* flag for no virtual arrays */ - post->buffer = NULL; /* flag for no strip buffer */ - - /* Create the quantization buffer, if needed */ - if (cinfo->quantize_colors) { - /* The buffer strip height is max_v_samp_factor, which is typically - * an efficient number of rows for upsampling to return. - * (In the presence of output rescaling, we might want to be smarter?) - */ - post->strip_height = (JDIMENSION) cinfo->max_v_samp_factor; - if (need_full_buffer) { - /* Two-pass color quantization: need full-image storage. */ - /* We round up the number of rows to a multiple of the strip height. */ -#ifdef QUANT_2PASS_SUPPORTED - post->whole_image = (*cinfo->mem->request_virt_sarray) - ((j_common_ptr) cinfo, JPOOL_IMAGE, FALSE, - cinfo->output_width * cinfo->out_color_components, - (JDIMENSION) jround_up((long) cinfo->output_height, - (long) post->strip_height), - post->strip_height); -#else - ERREXIT(cinfo, JERR_BAD_BUFFER_MODE); -#endif /* QUANT_2PASS_SUPPORTED */ - } else { - /* One-pass color quantization: just make a strip buffer. */ - post->buffer = (*cinfo->mem->alloc_sarray) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - cinfo->output_width * cinfo->out_color_components, - post->strip_height); - } - } -} diff --git a/dep/libjpeg/src/jdsample.c b/dep/libjpeg/src/jdsample.c deleted file mode 100644 index 15afeafe3..000000000 --- a/dep/libjpeg/src/jdsample.c +++ /dev/null @@ -1,341 +0,0 @@ -/* - * jdsample.c - * - * Copyright (C) 1991-1996, Thomas G. Lane. - * Modified 2002-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains upsampling routines. - * - * Upsampling input data is counted in "row groups". A row group - * is defined to be (v_samp_factor * DCT_v_scaled_size / min_DCT_v_scaled_size) - * sample rows of each component. Upsampling will normally produce - * max_v_samp_factor pixel rows from each row group (but this could vary - * if the upsampler is applying a scale factor of its own). - * - * An excellent reference for image resampling is - * Digital Image Warping, George Wolberg, 1990. - * Pub. by IEEE Computer Society Press, Los Alamitos, CA. ISBN 0-8186-8944-7. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Pointer to routine to upsample a single component */ -typedef JMETHOD(void, upsample1_ptr, - (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY input_data, JSAMPIMAGE output_data_ptr)); - -/* Private subobject */ - -typedef struct { - struct jpeg_upsampler pub; /* public fields */ - - /* Color conversion buffer. When using separate upsampling and color - * conversion steps, this buffer holds one upsampled row group until it - * has been color converted and output. - * Note: we do not allocate any storage for component(s) which are full-size, - * ie do not need rescaling. The corresponding entry of color_buf[] is - * simply set to point to the input data array, thereby avoiding copying. - */ - JSAMPARRAY color_buf[MAX_COMPONENTS]; - - /* Per-component upsampling method pointers */ - upsample1_ptr methods[MAX_COMPONENTS]; - - int next_row_out; /* counts rows emitted from color_buf */ - JDIMENSION rows_to_go; /* counts rows remaining in image */ - - /* Height of an input row group for each component. */ - int rowgroup_height[MAX_COMPONENTS]; - - /* These arrays save pixel expansion factors so that int_expand need not - * recompute them each time. They are unused for other upsampling methods. - */ - UINT8 h_expand[MAX_COMPONENTS]; - UINT8 v_expand[MAX_COMPONENTS]; -} my_upsampler; - -typedef my_upsampler * my_upsample_ptr; - - -/* - * Initialize for an upsampling pass. - */ - -METHODDEF(void) -start_pass_upsample (j_decompress_ptr cinfo) -{ - my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample; - - /* Mark the conversion buffer empty */ - upsample->next_row_out = cinfo->max_v_samp_factor; - /* Initialize total-height counter for detecting bottom of image */ - upsample->rows_to_go = cinfo->output_height; -} - - -/* - * Control routine to do upsampling (and color conversion). - * - * In this version we upsample each component independently. - * We upsample one row group into the conversion buffer, then apply - * color conversion a row at a time. - */ - -METHODDEF(void) -sep_upsample (j_decompress_ptr cinfo, - JSAMPIMAGE input_buf, JDIMENSION *in_row_group_ctr, - JDIMENSION in_row_groups_avail, - JSAMPARRAY output_buf, JDIMENSION *out_row_ctr, - JDIMENSION out_rows_avail) -{ - my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample; - int ci; - jpeg_component_info * compptr; - JDIMENSION num_rows; - - /* Fill the conversion buffer, if it's empty */ - if (upsample->next_row_out >= cinfo->max_v_samp_factor) { - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* Don't bother to upsample an uninteresting component. */ - if (! compptr->component_needed) - continue; - /* Invoke per-component upsample method. Notice we pass a POINTER - * to color_buf[ci], so that fullsize_upsample can change it. - */ - (*upsample->methods[ci]) (cinfo, compptr, - input_buf[ci] + (*in_row_group_ctr * upsample->rowgroup_height[ci]), - upsample->color_buf + ci); - } - upsample->next_row_out = 0; - } - - /* Color-convert and emit rows */ - - /* How many we have in the buffer: */ - num_rows = (JDIMENSION) (cinfo->max_v_samp_factor - upsample->next_row_out); - /* Not more than the distance to the end of the image. Need this test - * in case the image height is not a multiple of max_v_samp_factor: - */ - if (num_rows > upsample->rows_to_go) - num_rows = upsample->rows_to_go; - /* And not more than what the client can accept: */ - out_rows_avail -= *out_row_ctr; - if (num_rows > out_rows_avail) - num_rows = out_rows_avail; - - (*cinfo->cconvert->color_convert) (cinfo, upsample->color_buf, - (JDIMENSION) upsample->next_row_out, - output_buf + *out_row_ctr, - (int) num_rows); - - /* Adjust counts */ - *out_row_ctr += num_rows; - upsample->rows_to_go -= num_rows; - upsample->next_row_out += num_rows; - /* When the buffer is emptied, declare this input row group consumed */ - if (upsample->next_row_out >= cinfo->max_v_samp_factor) - (*in_row_group_ctr)++; -} - - -/* - * These are the routines invoked by sep_upsample to upsample pixel values - * of a single component. One row group is processed per call. - */ - - -/* - * For full-size components, we just make color_buf[ci] point at the - * input buffer, and thus avoid copying any data. Note that this is - * safe only because sep_upsample doesn't declare the input row group - * "consumed" until we are done color converting and emitting it. - */ - -METHODDEF(void) -fullsize_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY input_data, JSAMPIMAGE output_data_ptr) -{ - *output_data_ptr = input_data; -} - - -/* - * This version handles any integral sampling ratios. - * This is not used for typical JPEG files, so it need not be fast. - * Nor, for that matter, is it particularly accurate: the algorithm is - * simple replication of the input pixel onto the corresponding output - * pixels. The hi-falutin sampling literature refers to this as a - * "box filter". A box filter tends to introduce visible artifacts, - * so if you are actually going to use 3:1 or 4:1 sampling ratios - * you would be well advised to improve this code. - */ - -METHODDEF(void) -int_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY input_data, JSAMPIMAGE output_data_ptr) -{ - my_upsample_ptr upsample = (my_upsample_ptr) cinfo->upsample; - JSAMPARRAY output_data, output_end; - register JSAMPROW inptr, outptr; - register JSAMPLE invalue; - register int h; - JSAMPROW outend; - int h_expand, v_expand; - - h_expand = upsample->h_expand[compptr->component_index]; - v_expand = upsample->v_expand[compptr->component_index]; - - output_data = *output_data_ptr; - output_end = output_data + cinfo->max_v_samp_factor; - for (; output_data < output_end; output_data += v_expand) { - /* Generate one output row with proper horizontal expansion */ - inptr = *input_data++; - outptr = *output_data; - outend = outptr + cinfo->output_width; - while (outptr < outend) { - invalue = *inptr++; /* don't need GETJSAMPLE() here */ - for (h = h_expand; h > 0; h--) { - *outptr++ = invalue; - } - } - /* Generate any additional output rows by duplicating the first one */ - if (v_expand > 1) { - jcopy_sample_rows(output_data, output_data + 1, - v_expand - 1, cinfo->output_width); - } - } -} - - -/* - * Fast processing for the common case of 2:1 horizontal and 1:1 vertical. - * It's still a box filter. - */ - -METHODDEF(void) -h2v1_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY input_data, JSAMPIMAGE output_data_ptr) -{ - JSAMPARRAY output_data = *output_data_ptr; - register JSAMPROW inptr, outptr; - register JSAMPLE invalue; - JSAMPROW outend; - int outrow; - - for (outrow = 0; outrow < cinfo->max_v_samp_factor; outrow++) { - inptr = input_data[outrow]; - outptr = output_data[outrow]; - outend = outptr + cinfo->output_width; - while (outptr < outend) { - invalue = *inptr++; /* don't need GETJSAMPLE() here */ - *outptr++ = invalue; - *outptr++ = invalue; - } - } -} - - -/* - * Fast processing for the common case of 2:1 horizontal and 2:1 vertical. - * It's still a box filter. - */ - -METHODDEF(void) -h2v2_upsample (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JSAMPARRAY input_data, JSAMPIMAGE output_data_ptr) -{ - JSAMPARRAY output_data, output_end; - register JSAMPROW inptr, outptr; - register JSAMPLE invalue; - JSAMPROW outend; - - output_data = *output_data_ptr; - output_end = output_data + cinfo->max_v_samp_factor; - for (; output_data < output_end; output_data += 2) { - inptr = *input_data++; - outptr = *output_data; - outend = outptr + cinfo->output_width; - while (outptr < outend) { - invalue = *inptr++; /* don't need GETJSAMPLE() here */ - *outptr++ = invalue; - *outptr++ = invalue; - } - jcopy_sample_rows(output_data, output_data + 1, - 1, cinfo->output_width); - } -} - - -/* - * Module initialization routine for upsampling. - */ - -GLOBAL(void) -jinit_upsampler (j_decompress_ptr cinfo) -{ - my_upsample_ptr upsample; - int ci; - jpeg_component_info * compptr; - int h_in_group, v_in_group, h_out_group, v_out_group; - - upsample = (my_upsample_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_upsampler)); - cinfo->upsample = &upsample->pub; - upsample->pub.start_pass = start_pass_upsample; - upsample->pub.upsample = sep_upsample; - upsample->pub.need_context_rows = FALSE; /* until we find out differently */ - - if (cinfo->CCIR601_sampling) /* this isn't supported */ - ERREXIT(cinfo, JERR_CCIR601_NOTIMPL); - - /* Verify we can handle the sampling factors, select per-component methods, - * and create storage as needed. - */ - for (ci = 0, compptr = cinfo->comp_info; ci < cinfo->num_components; - ci++, compptr++) { - /* Don't bother to upsample an uninteresting component. */ - if (! compptr->component_needed) - continue; - /* Compute size of an "input group" after IDCT scaling. This many samples - * are to be converted to max_h_samp_factor * max_v_samp_factor pixels. - */ - h_in_group = (compptr->h_samp_factor * compptr->DCT_h_scaled_size) / - cinfo->min_DCT_h_scaled_size; - v_in_group = (compptr->v_samp_factor * compptr->DCT_v_scaled_size) / - cinfo->min_DCT_v_scaled_size; - h_out_group = cinfo->max_h_samp_factor; - v_out_group = cinfo->max_v_samp_factor; - upsample->rowgroup_height[ci] = v_in_group; /* save for use later */ - if (h_in_group == h_out_group && v_in_group == v_out_group) { - /* Fullsize components can be processed without any work. */ - upsample->methods[ci] = fullsize_upsample; - continue; /* don't need to allocate buffer */ - } - if (h_in_group * 2 == h_out_group && v_in_group == v_out_group) { - /* Special case for 2h1v upsampling */ - upsample->methods[ci] = h2v1_upsample; - } else if (h_in_group * 2 == h_out_group && - v_in_group * 2 == v_out_group) { - /* Special case for 2h2v upsampling */ - upsample->methods[ci] = h2v2_upsample; - } else if ((h_out_group % h_in_group) == 0 && - (v_out_group % v_in_group) == 0) { - /* Generic integral-factors upsampling method */ - upsample->methods[ci] = int_upsample; - upsample->h_expand[ci] = (UINT8) (h_out_group / h_in_group); - upsample->v_expand[ci] = (UINT8) (v_out_group / v_in_group); - } else - ERREXIT(cinfo, JERR_FRACT_SAMPLE_NOTIMPL); - upsample->color_buf[ci] = (*cinfo->mem->alloc_sarray) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - (JDIMENSION) jround_up((long) cinfo->output_width, - (long) cinfo->max_h_samp_factor), - (JDIMENSION) cinfo->max_v_samp_factor); - } -} diff --git a/dep/libjpeg/src/jdtrans.c b/dep/libjpeg/src/jdtrans.c deleted file mode 100644 index 22dd47fb5..000000000 --- a/dep/libjpeg/src/jdtrans.c +++ /dev/null @@ -1,140 +0,0 @@ -/* - * jdtrans.c - * - * Copyright (C) 1995-1997, Thomas G. Lane. - * Modified 2000-2009 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains library routines for transcoding decompression, - * that is, reading raw DCT coefficient arrays from an input JPEG file. - * The routines in jdapimin.c will also be needed by a transcoder. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* Forward declarations */ -LOCAL(void) transdecode_master_selection JPP((j_decompress_ptr cinfo)); - - -/* - * Read the coefficient arrays from a JPEG file. - * jpeg_read_header must be completed before calling this. - * - * The entire image is read into a set of virtual coefficient-block arrays, - * one per component. The return value is a pointer to the array of - * virtual-array descriptors. These can be manipulated directly via the - * JPEG memory manager, or handed off to jpeg_write_coefficients(). - * To release the memory occupied by the virtual arrays, call - * jpeg_finish_decompress() when done with the data. - * - * An alternative usage is to simply obtain access to the coefficient arrays - * during a buffered-image-mode decompression operation. This is allowed - * after any jpeg_finish_output() call. The arrays can be accessed until - * jpeg_finish_decompress() is called. (Note that any call to the library - * may reposition the arrays, so don't rely on access_virt_barray() results - * to stay valid across library calls.) - * - * Returns NULL if suspended. This case need be checked only if - * a suspending data source is used. - */ - -GLOBAL(jvirt_barray_ptr *) -jpeg_read_coefficients (j_decompress_ptr cinfo) -{ - if (cinfo->global_state == DSTATE_READY) { - /* First call: initialize active modules */ - transdecode_master_selection(cinfo); - cinfo->global_state = DSTATE_RDCOEFS; - } - if (cinfo->global_state == DSTATE_RDCOEFS) { - /* Absorb whole file into the coef buffer */ - for (;;) { - int retcode; - /* Call progress monitor hook if present */ - if (cinfo->progress != NULL) - (*cinfo->progress->progress_monitor) ((j_common_ptr) cinfo); - /* Absorb some more input */ - retcode = (*cinfo->inputctl->consume_input) (cinfo); - if (retcode == JPEG_SUSPENDED) - return NULL; - if (retcode == JPEG_REACHED_EOI) - break; - /* Advance progress counter if appropriate */ - if (cinfo->progress != NULL && - (retcode == JPEG_ROW_COMPLETED || retcode == JPEG_REACHED_SOS)) { - if (++cinfo->progress->pass_counter >= cinfo->progress->pass_limit) { - /* startup underestimated number of scans; ratchet up one scan */ - cinfo->progress->pass_limit += (long) cinfo->total_iMCU_rows; - } - } - } - /* Set state so that jpeg_finish_decompress does the right thing */ - cinfo->global_state = DSTATE_STOPPING; - } - /* At this point we should be in state DSTATE_STOPPING if being used - * standalone, or in state DSTATE_BUFIMAGE if being invoked to get access - * to the coefficients during a full buffered-image-mode decompression. - */ - if ((cinfo->global_state == DSTATE_STOPPING || - cinfo->global_state == DSTATE_BUFIMAGE) && cinfo->buffered_image) { - return cinfo->coef->coef_arrays; - } - /* Oops, improper usage */ - ERREXIT1(cinfo, JERR_BAD_STATE, cinfo->global_state); - return NULL; /* keep compiler happy */ -} - - -/* - * Master selection of decompression modules for transcoding. - * This substitutes for jdmaster.c's initialization of the full decompressor. - */ - -LOCAL(void) -transdecode_master_selection (j_decompress_ptr cinfo) -{ - /* This is effectively a buffered-image operation. */ - cinfo->buffered_image = TRUE; - - /* Compute output image dimensions and related values. */ - jpeg_core_output_dimensions(cinfo); - - /* Entropy decoding: either Huffman or arithmetic coding. */ - if (cinfo->arith_code) - jinit_arith_decoder(cinfo); - else { - jinit_huff_decoder(cinfo); - } - - /* Always get a full-image coefficient buffer. */ - jinit_d_coef_controller(cinfo, TRUE); - - /* We can now tell the memory manager to allocate virtual arrays. */ - (*cinfo->mem->realize_virt_arrays) ((j_common_ptr) cinfo); - - /* Initialize input side of decompressor to consume first scan. */ - (*cinfo->inputctl->start_input_pass) (cinfo); - - /* Initialize progress monitoring. */ - if (cinfo->progress != NULL) { - int nscans; - /* Estimate number of scans to set pass_limit. */ - if (cinfo->progressive_mode) { - /* Arbitrarily estimate 2 interleaved DC scans + 3 AC scans/component. */ - nscans = 2 + 3 * cinfo->num_components; - } else if (cinfo->inputctl->has_multiple_scans) { - /* For a nonprogressive multiscan file, estimate 1 scan per component. */ - nscans = cinfo->num_components; - } else { - nscans = 1; - } - cinfo->progress->pass_counter = 0L; - cinfo->progress->pass_limit = (long) cinfo->total_iMCU_rows * nscans; - cinfo->progress->completed_passes = 0; - cinfo->progress->total_passes = 1; - } -} diff --git a/dep/libjpeg/src/jerror.c b/dep/libjpeg/src/jerror.c deleted file mode 100644 index 7163af699..000000000 --- a/dep/libjpeg/src/jerror.c +++ /dev/null @@ -1,253 +0,0 @@ -/* - * jerror.c - * - * Copyright (C) 1991-1998, Thomas G. Lane. - * Modified 2012-2015 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains simple error-reporting and trace-message routines. - * These are suitable for Unix-like systems and others where writing to - * stderr is the right thing to do. Many applications will want to replace - * some or all of these routines. - * - * If you define USE_WINDOWS_MESSAGEBOX in jconfig.h or in the makefile, - * you get a Windows-specific hack to display error messages in a dialog box. - * It ain't much, but it beats dropping error messages into the bit bucket, - * which is what happens to output to stderr under most Windows C compilers. - * - * These routines are used by both the compression and decompression code. - */ - -#ifdef USE_WINDOWS_MESSAGEBOX -#include -#endif - -/* this is not a core library module, so it doesn't define JPEG_INTERNALS */ -#include "jinclude.h" -#include "jpeglib.h" -#include "jversion.h" -#include "jerror.h" - -#ifndef EXIT_FAILURE /* define exit() codes if not provided */ -#define EXIT_FAILURE 1 -#endif - - -/* - * Create the message string table. - * We do this from the master message list in jerror.h by re-reading - * jerror.h with a suitable definition for macro JMESSAGE. - * The message table is made an external symbol just in case any applications - * want to refer to it directly. - */ - -#ifdef NEED_SHORT_EXTERNAL_NAMES -#define jpeg_std_message_table jMsgTable -#endif - -#define JMESSAGE(code,string) string , - -const char * const jpeg_std_message_table[] = { -#include "jerror.h" - NULL -}; - - -/* - * Error exit handler: must not return to caller. - * - * Applications may override this if they want to get control back after - * an error. Typically one would longjmp somewhere instead of exiting. - * The setjmp buffer can be made a private field within an expanded error - * handler object. Note that the info needed to generate an error message - * is stored in the error object, so you can generate the message now or - * later, at your convenience. - * You should make sure that the JPEG object is cleaned up (with jpeg_abort - * or jpeg_destroy) at some point. - */ - -METHODDEF(noreturn_t) -error_exit (j_common_ptr cinfo) -{ - /* Always display the message */ - (*cinfo->err->output_message) (cinfo); - - /* Let the memory manager delete any temp files before we die */ - jpeg_destroy(cinfo); - - exit(EXIT_FAILURE); -} - - -/* - * Actual output of an error or trace message. - * Applications may override this method to send JPEG messages somewhere - * other than stderr. - * - * On Windows, printing to stderr is generally completely useless, - * so we provide optional code to produce an error-dialog popup. - * Most Windows applications will still prefer to override this routine, - * but if they don't, it'll do something at least marginally useful. - * - * NOTE: to use the library in an environment that doesn't support the - * C stdio library, you may have to delete the call to fprintf() entirely, - * not just not use this routine. - */ - -METHODDEF(void) -output_message (j_common_ptr cinfo) -{ - char buffer[JMSG_LENGTH_MAX]; - - /* Create the message */ - (*cinfo->err->format_message) (cinfo, buffer); - -#ifdef USE_WINDOWS_MESSAGEBOX - /* Display it in a message dialog box */ - MessageBox(GetActiveWindow(), buffer, "JPEG Library Error", - MB_OK | MB_ICONERROR); -#else - /* Send it to stderr, adding a newline */ - fprintf(stderr, "%s\n", buffer); -#endif -} - - -/* - * Decide whether to emit a trace or warning message. - * msg_level is one of: - * -1: recoverable corrupt-data warning, may want to abort. - * 0: important advisory messages (always display to user). - * 1: first level of tracing detail. - * 2,3,...: successively more detailed tracing messages. - * An application might override this method if it wanted to abort on warnings - * or change the policy about which messages to display. - */ - -METHODDEF(void) -emit_message (j_common_ptr cinfo, int msg_level) -{ - struct jpeg_error_mgr * err = cinfo->err; - - if (msg_level < 0) { - /* It's a warning message. Since corrupt files may generate many warnings, - * the policy implemented here is to show only the first warning, - * unless trace_level >= 3. - */ - if (err->num_warnings == 0 || err->trace_level >= 3) - (*err->output_message) (cinfo); - /* Always count warnings in num_warnings. */ - err->num_warnings++; - } else { - /* It's a trace message. Show it if trace_level >= msg_level. */ - if (err->trace_level >= msg_level) - (*err->output_message) (cinfo); - } -} - - -/* - * Format a message string for the most recent JPEG error or message. - * The message is stored into buffer, which should be at least JMSG_LENGTH_MAX - * characters. Note that no '\n' character is added to the string. - * Few applications should need to override this method. - */ - -METHODDEF(void) -format_message (j_common_ptr cinfo, char * buffer) -{ - struct jpeg_error_mgr * err = cinfo->err; - int msg_code = err->msg_code; - const char * msgtext = NULL; - const char * msgptr; - char ch; - boolean isstring; - - /* Look up message string in proper table */ - if (msg_code > 0 && msg_code <= err->last_jpeg_message) { - msgtext = err->jpeg_message_table[msg_code]; - } else if (err->addon_message_table != NULL && - msg_code >= err->first_addon_message && - msg_code <= err->last_addon_message) { - msgtext = err->addon_message_table[msg_code - err->first_addon_message]; - } - - /* Defend against bogus message number */ - if (msgtext == NULL) { - err->msg_parm.i[0] = msg_code; - msgtext = err->jpeg_message_table[0]; - } - - /* Check for string parameter, as indicated by %s in the message text */ - isstring = FALSE; - msgptr = msgtext; - while ((ch = *msgptr++) != '\0') { - if (ch == '%') { - if (*msgptr == 's') isstring = TRUE; - break; - } - } - - /* Format the message into the passed buffer */ - if (isstring) - sprintf(buffer, msgtext, err->msg_parm.s); - else - sprintf(buffer, msgtext, - err->msg_parm.i[0], err->msg_parm.i[1], - err->msg_parm.i[2], err->msg_parm.i[3], - err->msg_parm.i[4], err->msg_parm.i[5], - err->msg_parm.i[6], err->msg_parm.i[7]); -} - - -/* - * Reset error state variables at start of a new image. - * This is called during compression startup to reset trace/error - * processing to default state, without losing any application-specific - * method pointers. An application might possibly want to override - * this method if it has additional error processing state. - */ - -METHODDEF(void) -reset_error_mgr (j_common_ptr cinfo) -{ - cinfo->err->num_warnings = 0; - /* trace_level is not reset since it is an application-supplied parameter */ - cinfo->err->msg_code = 0; /* may be useful as a flag for "no error" */ -} - - -/* - * Fill in the standard error-handling methods in a jpeg_error_mgr object. - * Typical call is: - * struct jpeg_compress_struct cinfo; - * struct jpeg_error_mgr err; - * - * cinfo.err = jpeg_std_error(&err); - * after which the application may override some of the methods. - */ - -GLOBAL(struct jpeg_error_mgr *) -jpeg_std_error (struct jpeg_error_mgr * err) -{ - err->error_exit = error_exit; - err->emit_message = emit_message; - err->output_message = output_message; - err->format_message = format_message; - err->reset_error_mgr = reset_error_mgr; - - err->trace_level = 0; /* default = no tracing */ - err->num_warnings = 0; /* no warnings emitted yet */ - err->msg_code = 0; /* may be useful as a flag for "no error" */ - - /* Initialize message table pointers */ - err->jpeg_message_table = jpeg_std_message_table; - err->last_jpeg_message = (int) JMSG_LASTMSGCODE - 1; - - err->addon_message_table = NULL; - err->first_addon_message = 0; /* for safety */ - err->last_addon_message = 0; - - return err; -} diff --git a/dep/libjpeg/src/jfdctflt.c b/dep/libjpeg/src/jfdctflt.c deleted file mode 100644 index 013f29e0c..000000000 --- a/dep/libjpeg/src/jfdctflt.c +++ /dev/null @@ -1,176 +0,0 @@ -/* - * jfdctflt.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2003-2017 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains a floating-point implementation of the - * forward DCT (Discrete Cosine Transform). - * - * This implementation should be more accurate than either of the integer - * DCT implementations. However, it may not give the same results on all - * machines because of differences in roundoff behavior. Speed will depend - * on the hardware's floating point capacity. - * - * A 2-D DCT can be done by 1-D DCT on each row followed by 1-D DCT - * on each column. Direct algorithms are also available, but they are - * much more complex and seem not to be any faster when reduced to code. - * - * This implementation is based on Arai, Agui, and Nakajima's algorithm for - * scaled DCT. Their original paper (Trans. IEICE E-71(11):1095) is in - * Japanese, but the algorithm is described in the Pennebaker & Mitchell - * JPEG textbook (see REFERENCES section in file README). The following code - * is based directly on figure 4-8 in P&M. - * While an 8-point DCT cannot be done in less than 11 multiplies, it is - * possible to arrange the computation so that many of the multiplies are - * simple scalings of the final outputs. These multiplies can then be - * folded into the multiplications or divisions by the JPEG quantization - * table entries. The AA&N method leaves only 5 multiplies and 29 adds - * to be done in the DCT itself. - * The primary disadvantage of this method is that with a fixed-point - * implementation, accuracy is lost due to imprecise representation of the - * scaled quantization values. However, that problem does not arise if - * we use floating point arithmetic. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" -#include "jdct.h" /* Private declarations for DCT subsystem */ - -#ifdef DCT_FLOAT_SUPPORTED - - -/* - * This module is specialized to the case DCTSIZE = 8. - */ - -#if DCTSIZE != 8 - Sorry, this code only copes with 8x8 DCT blocks. /* deliberate syntax err */ -#endif - - -/* - * Perform the forward DCT on one block of samples. - * - * cK represents cos(K*pi/16). - */ - -GLOBAL(void) -jpeg_fdct_float (FAST_FLOAT * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - FAST_FLOAT tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7; - FAST_FLOAT tmp10, tmp11, tmp12, tmp13; - FAST_FLOAT z1, z2, z3, z4, z5, z11, z13; - FAST_FLOAT *dataptr; - JSAMPROW elemptr; - int ctr; - - /* Pass 1: process rows. */ - - dataptr = data; - for (ctr = 0; ctr < DCTSIZE; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Load data into workspace */ - tmp0 = (FAST_FLOAT) (GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[7])); - tmp7 = (FAST_FLOAT) (GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[7])); - tmp1 = (FAST_FLOAT) (GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[6])); - tmp6 = (FAST_FLOAT) (GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[6])); - tmp2 = (FAST_FLOAT) (GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[5])); - tmp5 = (FAST_FLOAT) (GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[5])); - tmp3 = (FAST_FLOAT) (GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[4])); - tmp4 = (FAST_FLOAT) (GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[4])); - - /* Even part */ - - tmp10 = tmp0 + tmp3; /* phase 2 */ - tmp13 = tmp0 - tmp3; - tmp11 = tmp1 + tmp2; - tmp12 = tmp1 - tmp2; - - /* Apply unsigned->signed conversion. */ - dataptr[0] = tmp10 + tmp11 - 8 * CENTERJSAMPLE; /* phase 3 */ - dataptr[4] = tmp10 - tmp11; - - z1 = (tmp12 + tmp13) * ((FAST_FLOAT) 0.707106781); /* c4 */ - dataptr[2] = tmp13 + z1; /* phase 5 */ - dataptr[6] = tmp13 - z1; - - /* Odd part */ - - tmp10 = tmp4 + tmp5; /* phase 2 */ - tmp11 = tmp5 + tmp6; - tmp12 = tmp6 + tmp7; - - /* The rotator is modified from fig 4-8 to avoid extra negations. */ - z5 = (tmp10 - tmp12) * ((FAST_FLOAT) 0.382683433); /* c6 */ - z2 = ((FAST_FLOAT) 0.541196100) * tmp10 + z5; /* c2-c6 */ - z4 = ((FAST_FLOAT) 1.306562965) * tmp12 + z5; /* c2+c6 */ - z3 = tmp11 * ((FAST_FLOAT) 0.707106781); /* c4 */ - - z11 = tmp7 + z3; /* phase 5 */ - z13 = tmp7 - z3; - - dataptr[5] = z13 + z2; /* phase 6 */ - dataptr[3] = z13 - z2; - dataptr[1] = z11 + z4; - dataptr[7] = z11 - z4; - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. */ - - dataptr = data; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*7]; - tmp7 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*7]; - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*6]; - tmp6 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*6]; - tmp2 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*5]; - tmp5 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*5]; - tmp3 = dataptr[DCTSIZE*3] + dataptr[DCTSIZE*4]; - tmp4 = dataptr[DCTSIZE*3] - dataptr[DCTSIZE*4]; - - /* Even part */ - - tmp10 = tmp0 + tmp3; /* phase 2 */ - tmp13 = tmp0 - tmp3; - tmp11 = tmp1 + tmp2; - tmp12 = tmp1 - tmp2; - - dataptr[DCTSIZE*0] = tmp10 + tmp11; /* phase 3 */ - dataptr[DCTSIZE*4] = tmp10 - tmp11; - - z1 = (tmp12 + tmp13) * ((FAST_FLOAT) 0.707106781); /* c4 */ - dataptr[DCTSIZE*2] = tmp13 + z1; /* phase 5 */ - dataptr[DCTSIZE*6] = tmp13 - z1; - - /* Odd part */ - - tmp10 = tmp4 + tmp5; /* phase 2 */ - tmp11 = tmp5 + tmp6; - tmp12 = tmp6 + tmp7; - - /* The rotator is modified from fig 4-8 to avoid extra negations. */ - z5 = (tmp10 - tmp12) * ((FAST_FLOAT) 0.382683433); /* c6 */ - z2 = ((FAST_FLOAT) 0.541196100) * tmp10 + z5; /* c2-c6 */ - z4 = ((FAST_FLOAT) 1.306562965) * tmp12 + z5; /* c2+c6 */ - z3 = tmp11 * ((FAST_FLOAT) 0.707106781); /* c4 */ - - z11 = tmp7 + z3; /* phase 5 */ - z13 = tmp7 - z3; - - dataptr[DCTSIZE*5] = z13 + z2; /* phase 6 */ - dataptr[DCTSIZE*3] = z13 - z2; - dataptr[DCTSIZE*1] = z11 + z4; - dataptr[DCTSIZE*7] = z11 - z4; - - dataptr++; /* advance pointer to next column */ - } -} - -#endif /* DCT_FLOAT_SUPPORTED */ diff --git a/dep/libjpeg/src/jfdctfst.c b/dep/libjpeg/src/jfdctfst.c deleted file mode 100644 index 5e4e01722..000000000 --- a/dep/libjpeg/src/jfdctfst.c +++ /dev/null @@ -1,232 +0,0 @@ -/* - * jfdctfst.c - * - * Copyright (C) 1994-1996, Thomas G. Lane. - * Modified 2003-2017 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains a fast, not so accurate integer implementation of the - * forward DCT (Discrete Cosine Transform). - * - * A 2-D DCT can be done by 1-D DCT on each row followed by 1-D DCT - * on each column. Direct algorithms are also available, but they are - * much more complex and seem not to be any faster when reduced to code. - * - * This implementation is based on Arai, Agui, and Nakajima's algorithm for - * scaled DCT. Their original paper (Trans. IEICE E-71(11):1095) is in - * Japanese, but the algorithm is described in the Pennebaker & Mitchell - * JPEG textbook (see REFERENCES section in file README). The following code - * is based directly on figure 4-8 in P&M. - * While an 8-point DCT cannot be done in less than 11 multiplies, it is - * possible to arrange the computation so that many of the multiplies are - * simple scalings of the final outputs. These multiplies can then be - * folded into the multiplications or divisions by the JPEG quantization - * table entries. The AA&N method leaves only 5 multiplies and 29 adds - * to be done in the DCT itself. - * The primary disadvantage of this method is that with fixed-point math, - * accuracy is lost due to imprecise representation of the scaled - * quantization values. The smaller the quantization table entry, the less - * precise the scaled value, so this implementation does worse with high- - * quality-setting files than with low-quality ones. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" -#include "jdct.h" /* Private declarations for DCT subsystem */ - -#ifdef DCT_IFAST_SUPPORTED - - -/* - * This module is specialized to the case DCTSIZE = 8. - */ - -#if DCTSIZE != 8 - Sorry, this code only copes with 8x8 DCT blocks. /* deliberate syntax err */ -#endif - - -/* Scaling decisions are generally the same as in the LL&M algorithm; - * see jfdctint.c for more details. However, we choose to descale - * (right shift) multiplication products as soon as they are formed, - * rather than carrying additional fractional bits into subsequent additions. - * This compromises accuracy slightly, but it lets us save a few shifts. - * More importantly, 16-bit arithmetic is then adequate (for 8-bit samples) - * everywhere except in the multiplications proper; this saves a good deal - * of work on 16-bit-int machines. - * - * Again to save a few shifts, the intermediate results between pass 1 and - * pass 2 are not upscaled, but are represented only to integral precision. - * - * A final compromise is to represent the multiplicative constants to only - * 8 fractional bits, rather than 13. This saves some shifting work on some - * machines, and may also reduce the cost of multiplication (since there - * are fewer one-bits in the constants). - */ - -#define CONST_BITS 8 - - -/* Some C compilers fail to reduce "FIX(constant)" at compile time, thus - * causing a lot of useless floating-point operations at run time. - * To get around this we use the following pre-calculated constants. - * If you change CONST_BITS you may want to add appropriate values. - * (With a reasonable C compiler, you can just rely on the FIX() macro...) - */ - -#if CONST_BITS == 8 -#define FIX_0_382683433 ((INT32) 98) /* FIX(0.382683433) */ -#define FIX_0_541196100 ((INT32) 139) /* FIX(0.541196100) */ -#define FIX_0_707106781 ((INT32) 181) /* FIX(0.707106781) */ -#define FIX_1_306562965 ((INT32) 334) /* FIX(1.306562965) */ -#else -#define FIX_0_382683433 FIX(0.382683433) -#define FIX_0_541196100 FIX(0.541196100) -#define FIX_0_707106781 FIX(0.707106781) -#define FIX_1_306562965 FIX(1.306562965) -#endif - - -/* We can gain a little more speed, with a further compromise in accuracy, - * by omitting the addition in a descaling shift. This yields an incorrectly - * rounded result half the time... - */ - -#ifndef USE_ACCURATE_ROUNDING -#undef DESCALE -#define DESCALE(x,n) RIGHT_SHIFT(x, n) -#endif - - -/* Multiply a DCTELEM variable by an INT32 constant, and immediately - * descale to yield a DCTELEM result. - */ - -#define MULTIPLY(var,const) ((DCTELEM) DESCALE((var) * (const), CONST_BITS)) - - -/* - * Perform the forward DCT on one block of samples. - * - * cK represents cos(K*pi/16). - */ - -GLOBAL(void) -jpeg_fdct_ifast (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - DCTELEM tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7; - DCTELEM tmp10, tmp11, tmp12, tmp13; - DCTELEM z1, z2, z3, z4, z5, z11, z13; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pass 1: process rows. */ - - dataptr = data; - for (ctr = 0; ctr < DCTSIZE; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Load data into workspace */ - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[7]); - tmp7 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[7]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[6]); - tmp6 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[6]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[5]); - tmp5 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[5]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[4]); - tmp4 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[4]); - - /* Even part */ - - tmp10 = tmp0 + tmp3; /* phase 2 */ - tmp13 = tmp0 - tmp3; - tmp11 = tmp1 + tmp2; - tmp12 = tmp1 - tmp2; - - /* Apply unsigned->signed conversion. */ - dataptr[0] = tmp10 + tmp11 - 8 * CENTERJSAMPLE; /* phase 3 */ - dataptr[4] = tmp10 - tmp11; - - z1 = MULTIPLY(tmp12 + tmp13, FIX_0_707106781); /* c4 */ - dataptr[2] = tmp13 + z1; /* phase 5 */ - dataptr[6] = tmp13 - z1; - - /* Odd part */ - - tmp10 = tmp4 + tmp5; /* phase 2 */ - tmp11 = tmp5 + tmp6; - tmp12 = tmp6 + tmp7; - - /* The rotator is modified from fig 4-8 to avoid extra negations. */ - z5 = MULTIPLY(tmp10 - tmp12, FIX_0_382683433); /* c6 */ - z2 = MULTIPLY(tmp10, FIX_0_541196100) + z5; /* c2-c6 */ - z4 = MULTIPLY(tmp12, FIX_1_306562965) + z5; /* c2+c6 */ - z3 = MULTIPLY(tmp11, FIX_0_707106781); /* c4 */ - - z11 = tmp7 + z3; /* phase 5 */ - z13 = tmp7 - z3; - - dataptr[5] = z13 + z2; /* phase 6 */ - dataptr[3] = z13 - z2; - dataptr[1] = z11 + z4; - dataptr[7] = z11 - z4; - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. */ - - dataptr = data; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*7]; - tmp7 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*7]; - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*6]; - tmp6 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*6]; - tmp2 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*5]; - tmp5 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*5]; - tmp3 = dataptr[DCTSIZE*3] + dataptr[DCTSIZE*4]; - tmp4 = dataptr[DCTSIZE*3] - dataptr[DCTSIZE*4]; - - /* Even part */ - - tmp10 = tmp0 + tmp3; /* phase 2 */ - tmp13 = tmp0 - tmp3; - tmp11 = tmp1 + tmp2; - tmp12 = tmp1 - tmp2; - - dataptr[DCTSIZE*0] = tmp10 + tmp11; /* phase 3 */ - dataptr[DCTSIZE*4] = tmp10 - tmp11; - - z1 = MULTIPLY(tmp12 + tmp13, FIX_0_707106781); /* c4 */ - dataptr[DCTSIZE*2] = tmp13 + z1; /* phase 5 */ - dataptr[DCTSIZE*6] = tmp13 - z1; - - /* Odd part */ - - tmp10 = tmp4 + tmp5; /* phase 2 */ - tmp11 = tmp5 + tmp6; - tmp12 = tmp6 + tmp7; - - /* The rotator is modified from fig 4-8 to avoid extra negations. */ - z5 = MULTIPLY(tmp10 - tmp12, FIX_0_382683433); /* c6 */ - z2 = MULTIPLY(tmp10, FIX_0_541196100) + z5; /* c2-c6 */ - z4 = MULTIPLY(tmp12, FIX_1_306562965) + z5; /* c2+c6 */ - z3 = MULTIPLY(tmp11, FIX_0_707106781); /* c4 */ - - z11 = tmp7 + z3; /* phase 5 */ - z13 = tmp7 - z3; - - dataptr[DCTSIZE*5] = z13 + z2; /* phase 6 */ - dataptr[DCTSIZE*3] = z13 - z2; - dataptr[DCTSIZE*1] = z11 + z4; - dataptr[DCTSIZE*7] = z11 - z4; - - dataptr++; /* advance pointer to next column */ - } -} - -#endif /* DCT_IFAST_SUPPORTED */ diff --git a/dep/libjpeg/src/jfdctint.c b/dep/libjpeg/src/jfdctint.c deleted file mode 100644 index 05df4750b..000000000 --- a/dep/libjpeg/src/jfdctint.c +++ /dev/null @@ -1,4415 +0,0 @@ -/* - * jfdctint.c - * - * Copyright (C) 1991-1996, Thomas G. Lane. - * Modification developed 2003-2018 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains a slow-but-accurate integer implementation of the - * forward DCT (Discrete Cosine Transform). - * - * A 2-D DCT can be done by 1-D DCT on each row followed by 1-D DCT - * on each column. Direct algorithms are also available, but they are - * much more complex and seem not to be any faster when reduced to code. - * - * This implementation is based on an algorithm described in - * C. Loeffler, A. Ligtenberg and G. Moschytz, "Practical Fast 1-D DCT - * Algorithms with 11 Multiplications", Proc. Int'l. Conf. on Acoustics, - * Speech, and Signal Processing 1989 (ICASSP '89), pp. 988-991. - * The primary algorithm described there uses 11 multiplies and 29 adds. - * We use their alternate method with 12 multiplies and 32 adds. - * The advantage of this method is that no data path contains more than one - * multiplication; this allows a very simple and accurate implementation in - * scaled fixed-point arithmetic, with a minimal number of shifts. - * - * We also provide FDCT routines with various input sample block sizes for - * direct resolution reduction or enlargement and for direct resolving the - * common 2x1 and 1x2 subsampling cases without additional resampling: NxN - * (N=1...16), 2NxN, and Nx2N (N=1...8) pixels for one 8x8 output DCT block. - * - * For N<8 we fill the remaining block coefficients with zero. - * For N>8 we apply a partial N-point FDCT on the input samples, computing - * just the lower 8 frequency coefficients and discarding the rest. - * - * We must scale the output coefficients of the N-point FDCT appropriately - * to the standard 8-point FDCT level by 8/N per 1-D pass. This scaling - * is folded into the constant multipliers (pass 2) and/or final/initial - * shifting. - * - * CAUTION: We rely on the FIX() macro except for the N=1,2,4,8 cases - * since there would be too many additional constants to pre-calculate. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" -#include "jdct.h" /* Private declarations for DCT subsystem */ - -#ifdef DCT_ISLOW_SUPPORTED - - -/* - * This module is specialized to the case DCTSIZE = 8. - */ - -#if DCTSIZE != 8 - Sorry, this code only copes with 8x8 DCT blocks. /* deliberate syntax err */ -#endif - - -/* - * The poop on this scaling stuff is as follows: - * - * Each 1-D DCT step produces outputs which are a factor of sqrt(N) - * larger than the true DCT outputs. The final outputs are therefore - * a factor of N larger than desired; since N=8 this can be cured by - * a simple right shift at the end of the algorithm. The advantage of - * this arrangement is that we save two multiplications per 1-D DCT, - * because the y0 and y4 outputs need not be divided by sqrt(N). - * In the IJG code, this factor of 8 is removed by the quantization step - * (in jcdctmgr.c), NOT in this module. - * - * We have to do addition and subtraction of the integer inputs, which - * is no problem, and multiplication by fractional constants, which is - * a problem to do in integer arithmetic. We multiply all the constants - * by CONST_SCALE and convert them to integer constants (thus retaining - * CONST_BITS bits of precision in the constants). After doing a - * multiplication we have to divide the product by CONST_SCALE, with proper - * rounding, to produce the correct output. This division can be done - * cheaply as a right shift of CONST_BITS bits. We postpone shifting - * as long as possible so that partial sums can be added together with - * full fractional precision. - * - * The outputs of the first pass are scaled up by PASS1_BITS bits so that - * they are represented to better-than-integral precision. These outputs - * require BITS_IN_JSAMPLE + PASS1_BITS + 3 bits; this fits in a 16-bit word - * with the recommended scaling. (For 12-bit sample data, the intermediate - * array is INT32 anyway.) - * - * To avoid overflow of the 32-bit intermediate results in pass 2, we must - * have BITS_IN_JSAMPLE + CONST_BITS + PASS1_BITS <= 26. Error analysis - * shows that the values given below are the most effective. - */ - -#if BITS_IN_JSAMPLE == 8 -#define CONST_BITS 13 -#define PASS1_BITS 2 -#else -#define CONST_BITS 13 -#define PASS1_BITS 1 /* lose a little precision to avoid overflow */ -#endif - -/* Some C compilers fail to reduce "FIX(constant)" at compile time, thus - * causing a lot of useless floating-point operations at run time. - * To get around this we use the following pre-calculated constants. - * If you change CONST_BITS you may want to add appropriate values. - * (With a reasonable C compiler, you can just rely on the FIX() macro...) - */ - -#if CONST_BITS == 13 -#define FIX_0_298631336 ((INT32) 2446) /* FIX(0.298631336) */ -#define FIX_0_390180644 ((INT32) 3196) /* FIX(0.390180644) */ -#define FIX_0_541196100 ((INT32) 4433) /* FIX(0.541196100) */ -#define FIX_0_765366865 ((INT32) 6270) /* FIX(0.765366865) */ -#define FIX_0_899976223 ((INT32) 7373) /* FIX(0.899976223) */ -#define FIX_1_175875602 ((INT32) 9633) /* FIX(1.175875602) */ -#define FIX_1_501321110 ((INT32) 12299) /* FIX(1.501321110) */ -#define FIX_1_847759065 ((INT32) 15137) /* FIX(1.847759065) */ -#define FIX_1_961570560 ((INT32) 16069) /* FIX(1.961570560) */ -#define FIX_2_053119869 ((INT32) 16819) /* FIX(2.053119869) */ -#define FIX_2_562915447 ((INT32) 20995) /* FIX(2.562915447) */ -#define FIX_3_072711026 ((INT32) 25172) /* FIX(3.072711026) */ -#else -#define FIX_0_298631336 FIX(0.298631336) -#define FIX_0_390180644 FIX(0.390180644) -#define FIX_0_541196100 FIX(0.541196100) -#define FIX_0_765366865 FIX(0.765366865) -#define FIX_0_899976223 FIX(0.899976223) -#define FIX_1_175875602 FIX(1.175875602) -#define FIX_1_501321110 FIX(1.501321110) -#define FIX_1_847759065 FIX(1.847759065) -#define FIX_1_961570560 FIX(1.961570560) -#define FIX_2_053119869 FIX(2.053119869) -#define FIX_2_562915447 FIX(2.562915447) -#define FIX_3_072711026 FIX(3.072711026) -#endif - - -/* Multiply an INT32 variable by an INT32 constant to yield an INT32 result. - * For 8-bit samples with the recommended scaling, all the variable - * and constant values involved are no more than 16 bits wide, so a - * 16x16->32 bit multiply can be used instead of a full 32x32 multiply. - * For 12-bit samples, a full 32-bit multiplication will be needed. - */ - -#if BITS_IN_JSAMPLE == 8 -#define MULTIPLY(var,const) MULTIPLY16C16(var,const) -#else -#define MULTIPLY(var,const) ((var) * (const)) -#endif - - -/* - * Perform the forward DCT on one block of samples. - */ - -GLOBAL(void) -jpeg_fdct_islow (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3; - INT32 tmp10, tmp11, tmp12, tmp13; - INT32 z1; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * cK represents sqrt(2) * cos(K*pi/16). - */ - - dataptr = data; - for (ctr = 0; ctr < DCTSIZE; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part per LL&M figure 1 --- note that published figure is faulty; - * rotator "c1" should be "c6". - */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[7]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[6]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[5]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[4]); - - tmp10 = tmp0 + tmp3; - tmp12 = tmp0 - tmp3; - tmp11 = tmp1 + tmp2; - tmp13 = tmp1 - tmp2; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[7]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[6]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[5]); - tmp3 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[4]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) ((tmp10 + tmp11 - 8 * CENTERJSAMPLE) << PASS1_BITS); - dataptr[4] = (DCTELEM) ((tmp10 - tmp11) << PASS1_BITS); - - z1 = MULTIPLY(tmp12 + tmp13, FIX_0_541196100); /* c6 */ - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS-PASS1_BITS-1); - - dataptr[2] = (DCTELEM) - RIGHT_SHIFT(z1 + MULTIPLY(tmp12, FIX_0_765366865), /* c2-c6 */ - CONST_BITS-PASS1_BITS); - dataptr[6] = (DCTELEM) - RIGHT_SHIFT(z1 - MULTIPLY(tmp13, FIX_1_847759065), /* c2+c6 */ - CONST_BITS-PASS1_BITS); - - /* Odd part per figure 8 --- note paper omits factor of sqrt(2). - * i0..i3 in the paper are tmp0..tmp3 here. - */ - - tmp12 = tmp0 + tmp2; - tmp13 = tmp1 + tmp3; - - z1 = MULTIPLY(tmp12 + tmp13, FIX_1_175875602); /* c3 */ - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS-PASS1_BITS-1); - - tmp12 = MULTIPLY(tmp12, - FIX_0_390180644); /* -c3+c5 */ - tmp13 = MULTIPLY(tmp13, - FIX_1_961570560); /* -c3-c5 */ - tmp12 += z1; - tmp13 += z1; - - z1 = MULTIPLY(tmp0 + tmp3, - FIX_0_899976223); /* -c3+c7 */ - tmp0 = MULTIPLY(tmp0, FIX_1_501321110); /* c1+c3-c5-c7 */ - tmp3 = MULTIPLY(tmp3, FIX_0_298631336); /* -c1+c3+c5-c7 */ - tmp0 += z1 + tmp12; - tmp3 += z1 + tmp13; - - z1 = MULTIPLY(tmp1 + tmp2, - FIX_2_562915447); /* -c1-c3 */ - tmp1 = MULTIPLY(tmp1, FIX_3_072711026); /* c1+c3+c5-c7 */ - tmp2 = MULTIPLY(tmp2, FIX_2_053119869); /* c1+c3-c5+c7 */ - tmp1 += z1 + tmp13; - tmp2 += z1 + tmp12; - - dataptr[1] = (DCTELEM) RIGHT_SHIFT(tmp0, CONST_BITS-PASS1_BITS); - dataptr[3] = (DCTELEM) RIGHT_SHIFT(tmp1, CONST_BITS-PASS1_BITS); - dataptr[5] = (DCTELEM) RIGHT_SHIFT(tmp2, CONST_BITS-PASS1_BITS); - dataptr[7] = (DCTELEM) RIGHT_SHIFT(tmp3, CONST_BITS-PASS1_BITS); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * cK represents sqrt(2) * cos(K*pi/16). - */ - - dataptr = data; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part per LL&M figure 1 --- note that published figure is faulty; - * rotator "c1" should be "c6". - */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*7]; - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*6]; - tmp2 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*5]; - tmp3 = dataptr[DCTSIZE*3] + dataptr[DCTSIZE*4]; - - /* Add fudge factor here for final descale. */ - tmp10 = tmp0 + tmp3 + (ONE << (PASS1_BITS-1)); - tmp12 = tmp0 - tmp3; - tmp11 = tmp1 + tmp2; - tmp13 = tmp1 - tmp2; - - tmp0 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*7]; - tmp1 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*6]; - tmp2 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*5]; - tmp3 = dataptr[DCTSIZE*3] - dataptr[DCTSIZE*4]; - - dataptr[DCTSIZE*0] = (DCTELEM) RIGHT_SHIFT(tmp10 + tmp11, PASS1_BITS); - dataptr[DCTSIZE*4] = (DCTELEM) RIGHT_SHIFT(tmp10 - tmp11, PASS1_BITS); - - z1 = MULTIPLY(tmp12 + tmp13, FIX_0_541196100); /* c6 */ - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS+PASS1_BITS-1); - - dataptr[DCTSIZE*2] = (DCTELEM) - RIGHT_SHIFT(z1 + MULTIPLY(tmp12, FIX_0_765366865), /* c2-c6 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*6] = (DCTELEM) - RIGHT_SHIFT(z1 - MULTIPLY(tmp13, FIX_1_847759065), /* c2+c6 */ - CONST_BITS+PASS1_BITS); - - /* Odd part per figure 8 --- note paper omits factor of sqrt(2). - * i0..i3 in the paper are tmp0..tmp3 here. - */ - - tmp12 = tmp0 + tmp2; - tmp13 = tmp1 + tmp3; - - z1 = MULTIPLY(tmp12 + tmp13, FIX_1_175875602); /* c3 */ - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS+PASS1_BITS-1); - - tmp12 = MULTIPLY(tmp12, - FIX_0_390180644); /* -c3+c5 */ - tmp13 = MULTIPLY(tmp13, - FIX_1_961570560); /* -c3-c5 */ - tmp12 += z1; - tmp13 += z1; - - z1 = MULTIPLY(tmp0 + tmp3, - FIX_0_899976223); /* -c3+c7 */ - tmp0 = MULTIPLY(tmp0, FIX_1_501321110); /* c1+c3-c5-c7 */ - tmp3 = MULTIPLY(tmp3, FIX_0_298631336); /* -c1+c3+c5-c7 */ - tmp0 += z1 + tmp12; - tmp3 += z1 + tmp13; - - z1 = MULTIPLY(tmp1 + tmp2, - FIX_2_562915447); /* -c1-c3 */ - tmp1 = MULTIPLY(tmp1, FIX_3_072711026); /* c1+c3+c5-c7 */ - tmp2 = MULTIPLY(tmp2, FIX_2_053119869); /* c1+c3-c5+c7 */ - tmp1 += z1 + tmp13; - tmp2 += z1 + tmp12; - - dataptr[DCTSIZE*1] = (DCTELEM) RIGHT_SHIFT(tmp0, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*3] = (DCTELEM) RIGHT_SHIFT(tmp1, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*5] = (DCTELEM) RIGHT_SHIFT(tmp2, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*7] = (DCTELEM) RIGHT_SHIFT(tmp3, CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - } -} - -#ifdef DCT_SCALING_SUPPORTED - - -/* - * Perform the forward DCT on a 7x7 sample block. - */ - -GLOBAL(void) -jpeg_fdct_7x7 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3; - INT32 tmp10, tmp11, tmp12; - INT32 z1, z2, z3; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * cK represents sqrt(2) * cos(K*pi/14). - */ - - dataptr = data; - for (ctr = 0; ctr < 7; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[6]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[5]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[4]); - tmp3 = GETJSAMPLE(elemptr[3]); - - tmp10 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[6]); - tmp11 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[5]); - tmp12 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[4]); - - z1 = tmp0 + tmp2; - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((z1 + tmp1 + tmp3 - 7 * CENTERJSAMPLE) << PASS1_BITS); - tmp3 += tmp3; - z1 -= tmp3; - z1 -= tmp3; - z1 = MULTIPLY(z1, FIX(0.353553391)); /* (c2+c6-c4)/2 */ - z2 = MULTIPLY(tmp0 - tmp2, FIX(0.920609002)); /* (c2+c4-c6)/2 */ - z3 = MULTIPLY(tmp1 - tmp2, FIX(0.314692123)); /* c6 */ - dataptr[2] = (DCTELEM) DESCALE(z1 + z2 + z3, CONST_BITS-PASS1_BITS); - z1 -= z2; - z2 = MULTIPLY(tmp0 - tmp1, FIX(0.881747734)); /* c4 */ - dataptr[4] = (DCTELEM) - DESCALE(z2 + z3 - MULTIPLY(tmp1 - tmp3, FIX(0.707106781)), /* c2+c6-c4 */ - CONST_BITS-PASS1_BITS); - dataptr[6] = (DCTELEM) DESCALE(z1 + z2, CONST_BITS-PASS1_BITS); - - /* Odd part */ - - tmp1 = MULTIPLY(tmp10 + tmp11, FIX(0.935414347)); /* (c3+c1-c5)/2 */ - tmp2 = MULTIPLY(tmp10 - tmp11, FIX(0.170262339)); /* (c3+c5-c1)/2 */ - tmp0 = tmp1 - tmp2; - tmp1 += tmp2; - tmp2 = MULTIPLY(tmp11 + tmp12, - FIX(1.378756276)); /* -c1 */ - tmp1 += tmp2; - tmp3 = MULTIPLY(tmp10 + tmp12, FIX(0.613604268)); /* c5 */ - tmp0 += tmp3; - tmp2 += tmp3 + MULTIPLY(tmp12, FIX(1.870828693)); /* c3+c1-c5 */ - - dataptr[1] = (DCTELEM) DESCALE(tmp0, CONST_BITS-PASS1_BITS); - dataptr[3] = (DCTELEM) DESCALE(tmp1, CONST_BITS-PASS1_BITS); - dataptr[5] = (DCTELEM) DESCALE(tmp2, CONST_BITS-PASS1_BITS); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/7)**2 = 64/49, which we fold - * into the constant multipliers: - * cK now represents sqrt(2) * cos(K*pi/14) * 64/49. - */ - - dataptr = data; - for (ctr = 0; ctr < 7; ctr++) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*6]; - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*5]; - tmp2 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*4]; - tmp3 = dataptr[DCTSIZE*3]; - - tmp10 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*6]; - tmp11 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*5]; - tmp12 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*4]; - - z1 = tmp0 + tmp2; - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(z1 + tmp1 + tmp3, FIX(1.306122449)), /* 64/49 */ - CONST_BITS+PASS1_BITS); - tmp3 += tmp3; - z1 -= tmp3; - z1 -= tmp3; - z1 = MULTIPLY(z1, FIX(0.461784020)); /* (c2+c6-c4)/2 */ - z2 = MULTIPLY(tmp0 - tmp2, FIX(1.202428084)); /* (c2+c4-c6)/2 */ - z3 = MULTIPLY(tmp1 - tmp2, FIX(0.411026446)); /* c6 */ - dataptr[DCTSIZE*2] = (DCTELEM) DESCALE(z1 + z2 + z3, CONST_BITS+PASS1_BITS); - z1 -= z2; - z2 = MULTIPLY(tmp0 - tmp1, FIX(1.151670509)); /* c4 */ - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(z2 + z3 - MULTIPLY(tmp1 - tmp3, FIX(0.923568041)), /* c2+c6-c4 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*6] = (DCTELEM) DESCALE(z1 + z2, CONST_BITS+PASS1_BITS); - - /* Odd part */ - - tmp1 = MULTIPLY(tmp10 + tmp11, FIX(1.221765677)); /* (c3+c1-c5)/2 */ - tmp2 = MULTIPLY(tmp10 - tmp11, FIX(0.222383464)); /* (c3+c5-c1)/2 */ - tmp0 = tmp1 - tmp2; - tmp1 += tmp2; - tmp2 = MULTIPLY(tmp11 + tmp12, - FIX(1.800824523)); /* -c1 */ - tmp1 += tmp2; - tmp3 = MULTIPLY(tmp10 + tmp12, FIX(0.801442310)); /* c5 */ - tmp0 += tmp3; - tmp2 += tmp3 + MULTIPLY(tmp12, FIX(2.443531355)); /* c3+c1-c5 */ - - dataptr[DCTSIZE*1] = (DCTELEM) DESCALE(tmp0, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*3] = (DCTELEM) DESCALE(tmp1, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*5] = (DCTELEM) DESCALE(tmp2, CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 6x6 sample block. - */ - -GLOBAL(void) -jpeg_fdct_6x6 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2; - INT32 tmp10, tmp11, tmp12; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * cK represents sqrt(2) * cos(K*pi/12). - */ - - dataptr = data; - for (ctr = 0; ctr < 6; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[5]); - tmp11 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[4]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[3]); - - tmp10 = tmp0 + tmp2; - tmp12 = tmp0 - tmp2; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[5]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[4]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[3]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp10 + tmp11 - 6 * CENTERJSAMPLE) << PASS1_BITS); - dataptr[2] = (DCTELEM) - DESCALE(MULTIPLY(tmp12, FIX(1.224744871)), /* c2 */ - CONST_BITS-PASS1_BITS); - dataptr[4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp11 - tmp11, FIX(0.707106781)), /* c4 */ - CONST_BITS-PASS1_BITS); - - /* Odd part */ - - tmp10 = DESCALE(MULTIPLY(tmp0 + tmp2, FIX(0.366025404)), /* c5 */ - CONST_BITS-PASS1_BITS); - - dataptr[1] = (DCTELEM) (tmp10 + ((tmp0 + tmp1) << PASS1_BITS)); - dataptr[3] = (DCTELEM) ((tmp0 - tmp1 - tmp2) << PASS1_BITS); - dataptr[5] = (DCTELEM) (tmp10 + ((tmp2 - tmp1) << PASS1_BITS)); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/6)**2 = 16/9, which we fold - * into the constant multipliers: - * cK now represents sqrt(2) * cos(K*pi/12) * 16/9. - */ - - dataptr = data; - for (ctr = 0; ctr < 6; ctr++) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*5]; - tmp11 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*4]; - tmp2 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*3]; - - tmp10 = tmp0 + tmp2; - tmp12 = tmp0 - tmp2; - - tmp0 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*5]; - tmp1 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*4]; - tmp2 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*3]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 + tmp11, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(MULTIPLY(tmp12, FIX(2.177324216)), /* c2 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp11 - tmp11, FIX(1.257078722)), /* c4 */ - CONST_BITS+PASS1_BITS); - - /* Odd part */ - - tmp10 = MULTIPLY(tmp0 + tmp2, FIX(0.650711829)); /* c5 */ - - dataptr[DCTSIZE*1] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp0 + tmp1, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*3] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 - tmp1 - tmp2, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*5] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp2 - tmp1, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 5x5 sample block. - */ - -GLOBAL(void) -jpeg_fdct_5x5 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2; - INT32 tmp10, tmp11; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * We scale the results further by 2 as part of output adaption - * scaling for different DCT size. - * cK represents sqrt(2) * cos(K*pi/10). - */ - - dataptr = data; - for (ctr = 0; ctr < 5; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[4]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[3]); - tmp2 = GETJSAMPLE(elemptr[2]); - - tmp10 = tmp0 + tmp1; - tmp11 = tmp0 - tmp1; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[4]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[3]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp10 + tmp2 - 5 * CENTERJSAMPLE) << (PASS1_BITS+1)); - tmp11 = MULTIPLY(tmp11, FIX(0.790569415)); /* (c2+c4)/2 */ - tmp10 -= tmp2 << 2; - tmp10 = MULTIPLY(tmp10, FIX(0.353553391)); /* (c2-c4)/2 */ - dataptr[2] = (DCTELEM) DESCALE(tmp11 + tmp10, CONST_BITS-PASS1_BITS-1); - dataptr[4] = (DCTELEM) DESCALE(tmp11 - tmp10, CONST_BITS-PASS1_BITS-1); - - /* Odd part */ - - tmp10 = MULTIPLY(tmp0 + tmp1, FIX(0.831253876)); /* c3 */ - - dataptr[1] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp0, FIX(0.513743148)), /* c1-c3 */ - CONST_BITS-PASS1_BITS-1); - dataptr[3] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp1, FIX(2.176250899)), /* c1+c3 */ - CONST_BITS-PASS1_BITS-1); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/5)**2 = 64/25, which we partially - * fold into the constant multipliers (other part was done in pass 1): - * cK now represents sqrt(2) * cos(K*pi/10) * 32/25. - */ - - dataptr = data; - for (ctr = 0; ctr < 5; ctr++) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*4]; - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*3]; - tmp2 = dataptr[DCTSIZE*2]; - - tmp10 = tmp0 + tmp1; - tmp11 = tmp0 - tmp1; - - tmp0 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*4]; - tmp1 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*3]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 + tmp2, FIX(1.28)), /* 32/25 */ - CONST_BITS+PASS1_BITS); - tmp11 = MULTIPLY(tmp11, FIX(1.011928851)); /* (c2+c4)/2 */ - tmp10 -= tmp2 << 2; - tmp10 = MULTIPLY(tmp10, FIX(0.452548340)); /* (c2-c4)/2 */ - dataptr[DCTSIZE*2] = (DCTELEM) DESCALE(tmp11 + tmp10, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*4] = (DCTELEM) DESCALE(tmp11 - tmp10, CONST_BITS+PASS1_BITS); - - /* Odd part */ - - tmp10 = MULTIPLY(tmp0 + tmp1, FIX(1.064004961)); /* c3 */ - - dataptr[DCTSIZE*1] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp0, FIX(0.657591230)), /* c1-c3 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*3] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp1, FIX(2.785601151)), /* c1+c3 */ - CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 4x4 sample block. - */ - -GLOBAL(void) -jpeg_fdct_4x4 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1; - INT32 tmp10, tmp11; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * We must also scale the output by (8/4)**2 = 2**2, which we add here. - * cK represents sqrt(2) * cos(K*pi/16) [refers to 8-point FDCT]. - */ - - dataptr = data; - for (ctr = 0; ctr < 4; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[3]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[2]); - - tmp10 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[3]); - tmp11 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[2]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp0 + tmp1 - 4 * CENTERJSAMPLE) << (PASS1_BITS+2)); - dataptr[2] = (DCTELEM) ((tmp0 - tmp1) << (PASS1_BITS+2)); - - /* Odd part */ - - tmp0 = MULTIPLY(tmp10 + tmp11, FIX_0_541196100); /* c6 */ - /* Add fudge factor here for final descale. */ - tmp0 += ONE << (CONST_BITS-PASS1_BITS-3); - - dataptr[1] = (DCTELEM) - RIGHT_SHIFT(tmp0 + MULTIPLY(tmp10, FIX_0_765366865), /* c2-c6 */ - CONST_BITS-PASS1_BITS-2); - dataptr[3] = (DCTELEM) - RIGHT_SHIFT(tmp0 - MULTIPLY(tmp11, FIX_1_847759065), /* c2+c6 */ - CONST_BITS-PASS1_BITS-2); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * cK represents sqrt(2) * cos(K*pi/16) [refers to 8-point FDCT]. - */ - - dataptr = data; - for (ctr = 0; ctr < 4; ctr++) { - /* Even part */ - - /* Add fudge factor here for final descale. */ - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*3] + (ONE << (PASS1_BITS-1)); - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*2]; - - tmp10 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*3]; - tmp11 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*2]; - - dataptr[DCTSIZE*0] = (DCTELEM) RIGHT_SHIFT(tmp0 + tmp1, PASS1_BITS); - dataptr[DCTSIZE*2] = (DCTELEM) RIGHT_SHIFT(tmp0 - tmp1, PASS1_BITS); - - /* Odd part */ - - tmp0 = MULTIPLY(tmp10 + tmp11, FIX_0_541196100); /* c6 */ - /* Add fudge factor here for final descale. */ - tmp0 += ONE << (CONST_BITS+PASS1_BITS-1); - - dataptr[DCTSIZE*1] = (DCTELEM) - RIGHT_SHIFT(tmp0 + MULTIPLY(tmp10, FIX_0_765366865), /* c2-c6 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*3] = (DCTELEM) - RIGHT_SHIFT(tmp0 - MULTIPLY(tmp11, FIX_1_847759065), /* c2+c6 */ - CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 3x3 sample block. - */ - -GLOBAL(void) -jpeg_fdct_3x3 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * We scale the results further by 2**2 as part of output adaption - * scaling for different DCT size. - * cK represents sqrt(2) * cos(K*pi/6). - */ - - dataptr = data; - for (ctr = 0; ctr < 3; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[2]); - tmp1 = GETJSAMPLE(elemptr[1]); - - tmp2 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[2]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp0 + tmp1 - 3 * CENTERJSAMPLE) << (PASS1_BITS+2)); - dataptr[2] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 - tmp1 - tmp1, FIX(0.707106781)), /* c2 */ - CONST_BITS-PASS1_BITS-2); - - /* Odd part */ - - dataptr[1] = (DCTELEM) - DESCALE(MULTIPLY(tmp2, FIX(1.224744871)), /* c1 */ - CONST_BITS-PASS1_BITS-2); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/3)**2 = 64/9, which we partially - * fold into the constant multipliers (other part was done in pass 1): - * cK now represents sqrt(2) * cos(K*pi/6) * 16/9. - */ - - dataptr = data; - for (ctr = 0; ctr < 3; ctr++) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*2]; - tmp1 = dataptr[DCTSIZE*1]; - - tmp2 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*2]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 + tmp1, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 - tmp1 - tmp1, FIX(1.257078722)), /* c2 */ - CONST_BITS+PASS1_BITS); - - /* Odd part */ - - dataptr[DCTSIZE*1] = (DCTELEM) - DESCALE(MULTIPLY(tmp2, FIX(2.177324216)), /* c1 */ - CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 2x2 sample block. - */ - -GLOBAL(void) -jpeg_fdct_2x2 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - DCTELEM tmp0, tmp1, tmp2, tmp3; - JSAMPROW elemptr; - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT. - */ - - /* Row 0 */ - elemptr = sample_data[0] + start_col; - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[1]); - tmp1 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[1]); - - /* Row 1 */ - elemptr = sample_data[1] + start_col; - - tmp2 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[1]); - tmp3 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[1]); - - /* Pass 2: process columns. - * We leave the results scaled up by an overall factor of 8. - * We must also scale the output by (8/2)**2 = 2**4. - */ - - /* Column 0 */ - /* Apply unsigned->signed conversion. */ - data[DCTSIZE*0] = (tmp0 + tmp2 - 4 * CENTERJSAMPLE) << 4; - data[DCTSIZE*1] = (tmp0 - tmp2) << 4; - - /* Column 1 */ - data[DCTSIZE*0+1] = (tmp1 + tmp3) << 4; - data[DCTSIZE*1+1] = (tmp1 - tmp3) << 4; -} - - -/* - * Perform the forward DCT on a 1x1 sample block. - */ - -GLOBAL(void) -jpeg_fdct_1x1 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - DCTELEM dcval; - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - dcval = GETJSAMPLE(sample_data[0][start_col]); - - /* We leave the result scaled up by an overall factor of 8. */ - /* We must also scale the output by (8/1)**2 = 2**6. */ - /* Apply unsigned->signed conversion. */ - data[0] = (dcval - CENTERJSAMPLE) << 6; -} - - -/* - * Perform the forward DCT on a 9x9 sample block. - */ - -GLOBAL(void) -jpeg_fdct_9x9 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4; - INT32 tmp10, tmp11, tmp12, tmp13; - INT32 z1, z2; - DCTELEM workspace[8]; - DCTELEM *dataptr; - DCTELEM *wsptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * we scale the results further by 2 as part of output adaption - * scaling for different DCT size. - * cK represents sqrt(2) * cos(K*pi/18). - */ - - dataptr = data; - ctr = 0; - for (;;) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[8]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[7]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[6]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[5]); - tmp4 = GETJSAMPLE(elemptr[4]); - - tmp10 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[8]); - tmp11 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[7]); - tmp12 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[6]); - tmp13 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[5]); - - z1 = tmp0 + tmp2 + tmp3; - z2 = tmp1 + tmp4; - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) ((z1 + z2 - 9 * CENTERJSAMPLE) << 1); - dataptr[6] = (DCTELEM) - DESCALE(MULTIPLY(z1 - z2 - z2, FIX(0.707106781)), /* c6 */ - CONST_BITS-1); - z1 = MULTIPLY(tmp0 - tmp2, FIX(1.328926049)); /* c2 */ - z2 = MULTIPLY(tmp1 - tmp4 - tmp4, FIX(0.707106781)); /* c6 */ - dataptr[2] = (DCTELEM) - DESCALE(MULTIPLY(tmp2 - tmp3, FIX(1.083350441)) /* c4 */ - + z1 + z2, CONST_BITS-1); - dataptr[4] = (DCTELEM) - DESCALE(MULTIPLY(tmp3 - tmp0, FIX(0.245575608)) /* c8 */ - + z1 - z2, CONST_BITS-1); - - /* Odd part */ - - dataptr[3] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp12 - tmp13, FIX(1.224744871)), /* c3 */ - CONST_BITS-1); - - tmp11 = MULTIPLY(tmp11, FIX(1.224744871)); /* c3 */ - tmp0 = MULTIPLY(tmp10 + tmp12, FIX(0.909038955)); /* c5 */ - tmp1 = MULTIPLY(tmp10 + tmp13, FIX(0.483689525)); /* c7 */ - - dataptr[1] = (DCTELEM) DESCALE(tmp11 + tmp0 + tmp1, CONST_BITS-1); - - tmp2 = MULTIPLY(tmp12 - tmp13, FIX(1.392728481)); /* c1 */ - - dataptr[5] = (DCTELEM) DESCALE(tmp0 - tmp11 - tmp2, CONST_BITS-1); - dataptr[7] = (DCTELEM) DESCALE(tmp1 - tmp11 + tmp2, CONST_BITS-1); - - ctr++; - - if (ctr != DCTSIZE) { - if (ctr == 9) - break; /* Done. */ - dataptr += DCTSIZE; /* advance pointer to next row */ - } else - dataptr = workspace; /* switch pointer to extended workspace */ - } - - /* Pass 2: process columns. - * We leave the results scaled up by an overall factor of 8. - * We must also scale the output by (8/9)**2 = 64/81, which we partially - * fold into the constant multipliers and final/initial shifting: - * cK now represents sqrt(2) * cos(K*pi/18) * 128/81. - */ - - dataptr = data; - wsptr = workspace; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + wsptr[DCTSIZE*0]; - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*7]; - tmp2 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*6]; - tmp3 = dataptr[DCTSIZE*3] + dataptr[DCTSIZE*5]; - tmp4 = dataptr[DCTSIZE*4]; - - tmp10 = dataptr[DCTSIZE*0] - wsptr[DCTSIZE*0]; - tmp11 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*7]; - tmp12 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*6]; - tmp13 = dataptr[DCTSIZE*3] - dataptr[DCTSIZE*5]; - - z1 = tmp0 + tmp2 + tmp3; - z2 = tmp1 + tmp4; - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(z1 + z2, FIX(1.580246914)), /* 128/81 */ - CONST_BITS+2); - dataptr[DCTSIZE*6] = (DCTELEM) - DESCALE(MULTIPLY(z1 - z2 - z2, FIX(1.117403309)), /* c6 */ - CONST_BITS+2); - z1 = MULTIPLY(tmp0 - tmp2, FIX(2.100031287)); /* c2 */ - z2 = MULTIPLY(tmp1 - tmp4 - tmp4, FIX(1.117403309)); /* c6 */ - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(MULTIPLY(tmp2 - tmp3, FIX(1.711961190)) /* c4 */ - + z1 + z2, CONST_BITS+2); - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(MULTIPLY(tmp3 - tmp0, FIX(0.388070096)) /* c8 */ - + z1 - z2, CONST_BITS+2); - - /* Odd part */ - - dataptr[DCTSIZE*3] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp12 - tmp13, FIX(1.935399303)), /* c3 */ - CONST_BITS+2); - - tmp11 = MULTIPLY(tmp11, FIX(1.935399303)); /* c3 */ - tmp0 = MULTIPLY(tmp10 + tmp12, FIX(1.436506004)); /* c5 */ - tmp1 = MULTIPLY(tmp10 + tmp13, FIX(0.764348879)); /* c7 */ - - dataptr[DCTSIZE*1] = (DCTELEM) - DESCALE(tmp11 + tmp0 + tmp1, CONST_BITS+2); - - tmp2 = MULTIPLY(tmp12 - tmp13, FIX(2.200854883)); /* c1 */ - - dataptr[DCTSIZE*5] = (DCTELEM) - DESCALE(tmp0 - tmp11 - tmp2, CONST_BITS+2); - dataptr[DCTSIZE*7] = (DCTELEM) - DESCALE(tmp1 - tmp11 + tmp2, CONST_BITS+2); - - dataptr++; /* advance pointer to next column */ - wsptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 10x10 sample block. - */ - -GLOBAL(void) -jpeg_fdct_10x10 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14; - DCTELEM workspace[8*2]; - DCTELEM *dataptr; - DCTELEM *wsptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * we scale the results further by 2 as part of output adaption - * scaling for different DCT size. - * cK represents sqrt(2) * cos(K*pi/20). - */ - - dataptr = data; - ctr = 0; - for (;;) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[9]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[8]); - tmp12 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[7]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[6]); - tmp4 = GETJSAMPLE(elemptr[4]) + GETJSAMPLE(elemptr[5]); - - tmp10 = tmp0 + tmp4; - tmp13 = tmp0 - tmp4; - tmp11 = tmp1 + tmp3; - tmp14 = tmp1 - tmp3; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[9]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[8]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[7]); - tmp3 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[6]); - tmp4 = GETJSAMPLE(elemptr[4]) - GETJSAMPLE(elemptr[5]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp10 + tmp11 + tmp12 - 10 * CENTERJSAMPLE) << 1); - tmp12 += tmp12; - dataptr[4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp12, FIX(1.144122806)) - /* c4 */ - MULTIPLY(tmp11 - tmp12, FIX(0.437016024)), /* c8 */ - CONST_BITS-1); - tmp10 = MULTIPLY(tmp13 + tmp14, FIX(0.831253876)); /* c6 */ - dataptr[2] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp13, FIX(0.513743148)), /* c2-c6 */ - CONST_BITS-1); - dataptr[6] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp14, FIX(2.176250899)), /* c2+c6 */ - CONST_BITS-1); - - /* Odd part */ - - tmp10 = tmp0 + tmp4; - tmp11 = tmp1 - tmp3; - dataptr[5] = (DCTELEM) ((tmp10 - tmp11 - tmp2) << 1); - tmp2 <<= CONST_BITS; - dataptr[1] = (DCTELEM) - DESCALE(MULTIPLY(tmp0, FIX(1.396802247)) + /* c1 */ - MULTIPLY(tmp1, FIX(1.260073511)) + tmp2 + /* c3 */ - MULTIPLY(tmp3, FIX(0.642039522)) + /* c7 */ - MULTIPLY(tmp4, FIX(0.221231742)), /* c9 */ - CONST_BITS-1); - tmp12 = MULTIPLY(tmp0 - tmp4, FIX(0.951056516)) - /* (c3+c7)/2 */ - MULTIPLY(tmp1 + tmp3, FIX(0.587785252)); /* (c1-c9)/2 */ - tmp13 = MULTIPLY(tmp10 + tmp11, FIX(0.309016994)) + /* (c3-c7)/2 */ - (tmp11 << (CONST_BITS - 1)) - tmp2; - dataptr[3] = (DCTELEM) DESCALE(tmp12 + tmp13, CONST_BITS-1); - dataptr[7] = (DCTELEM) DESCALE(tmp12 - tmp13, CONST_BITS-1); - - ctr++; - - if (ctr != DCTSIZE) { - if (ctr == 10) - break; /* Done. */ - dataptr += DCTSIZE; /* advance pointer to next row */ - } else - dataptr = workspace; /* switch pointer to extended workspace */ - } - - /* Pass 2: process columns. - * We leave the results scaled up by an overall factor of 8. - * We must also scale the output by (8/10)**2 = 16/25, which we partially - * fold into the constant multipliers and final/initial shifting: - * cK now represents sqrt(2) * cos(K*pi/20) * 32/25. - */ - - dataptr = data; - wsptr = workspace; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + wsptr[DCTSIZE*1]; - tmp1 = dataptr[DCTSIZE*1] + wsptr[DCTSIZE*0]; - tmp12 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*7]; - tmp3 = dataptr[DCTSIZE*3] + dataptr[DCTSIZE*6]; - tmp4 = dataptr[DCTSIZE*4] + dataptr[DCTSIZE*5]; - - tmp10 = tmp0 + tmp4; - tmp13 = tmp0 - tmp4; - tmp11 = tmp1 + tmp3; - tmp14 = tmp1 - tmp3; - - tmp0 = dataptr[DCTSIZE*0] - wsptr[DCTSIZE*1]; - tmp1 = dataptr[DCTSIZE*1] - wsptr[DCTSIZE*0]; - tmp2 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*7]; - tmp3 = dataptr[DCTSIZE*3] - dataptr[DCTSIZE*6]; - tmp4 = dataptr[DCTSIZE*4] - dataptr[DCTSIZE*5]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 + tmp11 + tmp12, FIX(1.28)), /* 32/25 */ - CONST_BITS+2); - tmp12 += tmp12; - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp12, FIX(1.464477191)) - /* c4 */ - MULTIPLY(tmp11 - tmp12, FIX(0.559380511)), /* c8 */ - CONST_BITS+2); - tmp10 = MULTIPLY(tmp13 + tmp14, FIX(1.064004961)); /* c6 */ - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp13, FIX(0.657591230)), /* c2-c6 */ - CONST_BITS+2); - dataptr[DCTSIZE*6] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp14, FIX(2.785601151)), /* c2+c6 */ - CONST_BITS+2); - - /* Odd part */ - - tmp10 = tmp0 + tmp4; - tmp11 = tmp1 - tmp3; - dataptr[DCTSIZE*5] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp11 - tmp2, FIX(1.28)), /* 32/25 */ - CONST_BITS+2); - tmp2 = MULTIPLY(tmp2, FIX(1.28)); /* 32/25 */ - dataptr[DCTSIZE*1] = (DCTELEM) - DESCALE(MULTIPLY(tmp0, FIX(1.787906876)) + /* c1 */ - MULTIPLY(tmp1, FIX(1.612894094)) + tmp2 + /* c3 */ - MULTIPLY(tmp3, FIX(0.821810588)) + /* c7 */ - MULTIPLY(tmp4, FIX(0.283176630)), /* c9 */ - CONST_BITS+2); - tmp12 = MULTIPLY(tmp0 - tmp4, FIX(1.217352341)) - /* (c3+c7)/2 */ - MULTIPLY(tmp1 + tmp3, FIX(0.752365123)); /* (c1-c9)/2 */ - tmp13 = MULTIPLY(tmp10 + tmp11, FIX(0.395541753)) + /* (c3-c7)/2 */ - MULTIPLY(tmp11, FIX(0.64)) - tmp2; /* 16/25 */ - dataptr[DCTSIZE*3] = (DCTELEM) DESCALE(tmp12 + tmp13, CONST_BITS+2); - dataptr[DCTSIZE*7] = (DCTELEM) DESCALE(tmp12 - tmp13, CONST_BITS+2); - - dataptr++; /* advance pointer to next column */ - wsptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on an 11x11 sample block. - */ - -GLOBAL(void) -jpeg_fdct_11x11 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14; - INT32 z1, z2, z3; - DCTELEM workspace[8*3]; - DCTELEM *dataptr; - DCTELEM *wsptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * we scale the results further by 2 as part of output adaption - * scaling for different DCT size. - * cK represents sqrt(2) * cos(K*pi/22). - */ - - dataptr = data; - ctr = 0; - for (;;) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[10]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[9]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[8]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[7]); - tmp4 = GETJSAMPLE(elemptr[4]) + GETJSAMPLE(elemptr[6]); - tmp5 = GETJSAMPLE(elemptr[5]); - - tmp10 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[10]); - tmp11 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[9]); - tmp12 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[8]); - tmp13 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[7]); - tmp14 = GETJSAMPLE(elemptr[4]) - GETJSAMPLE(elemptr[6]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp0 + tmp1 + tmp2 + tmp3 + tmp4 + tmp5 - 11 * CENTERJSAMPLE) << 1); - tmp5 += tmp5; - tmp0 -= tmp5; - tmp1 -= tmp5; - tmp2 -= tmp5; - tmp3 -= tmp5; - tmp4 -= tmp5; - z1 = MULTIPLY(tmp0 + tmp3, FIX(1.356927976)) + /* c2 */ - MULTIPLY(tmp2 + tmp4, FIX(0.201263574)); /* c10 */ - z2 = MULTIPLY(tmp1 - tmp3, FIX(0.926112931)); /* c6 */ - z3 = MULTIPLY(tmp0 - tmp1, FIX(1.189712156)); /* c4 */ - dataptr[2] = (DCTELEM) - DESCALE(z1 + z2 - MULTIPLY(tmp3, FIX(1.018300590)) /* c2+c8-c6 */ - - MULTIPLY(tmp4, FIX(1.390975730)), /* c4+c10 */ - CONST_BITS-1); - dataptr[4] = (DCTELEM) - DESCALE(z2 + z3 + MULTIPLY(tmp1, FIX(0.062335650)) /* c4-c6-c10 */ - - MULTIPLY(tmp2, FIX(1.356927976)) /* c2 */ - + MULTIPLY(tmp4, FIX(0.587485545)), /* c8 */ - CONST_BITS-1); - dataptr[6] = (DCTELEM) - DESCALE(z1 + z3 - MULTIPLY(tmp0, FIX(1.620527200)) /* c2+c4-c6 */ - - MULTIPLY(tmp2, FIX(0.788749120)), /* c8+c10 */ - CONST_BITS-1); - - /* Odd part */ - - tmp1 = MULTIPLY(tmp10 + tmp11, FIX(1.286413905)); /* c3 */ - tmp2 = MULTIPLY(tmp10 + tmp12, FIX(1.068791298)); /* c5 */ - tmp3 = MULTIPLY(tmp10 + tmp13, FIX(0.764581576)); /* c7 */ - tmp0 = tmp1 + tmp2 + tmp3 - MULTIPLY(tmp10, FIX(1.719967871)) /* c7+c5+c3-c1 */ - + MULTIPLY(tmp14, FIX(0.398430003)); /* c9 */ - tmp4 = MULTIPLY(tmp11 + tmp12, - FIX(0.764581576)); /* -c7 */ - tmp5 = MULTIPLY(tmp11 + tmp13, - FIX(1.399818907)); /* -c1 */ - tmp1 += tmp4 + tmp5 + MULTIPLY(tmp11, FIX(1.276416582)) /* c9+c7+c1-c3 */ - - MULTIPLY(tmp14, FIX(1.068791298)); /* c5 */ - tmp10 = MULTIPLY(tmp12 + tmp13, FIX(0.398430003)); /* c9 */ - tmp2 += tmp4 + tmp10 - MULTIPLY(tmp12, FIX(1.989053629)) /* c9+c5+c3-c7 */ - + MULTIPLY(tmp14, FIX(1.399818907)); /* c1 */ - tmp3 += tmp5 + tmp10 + MULTIPLY(tmp13, FIX(1.305598626)) /* c1+c5-c9-c7 */ - - MULTIPLY(tmp14, FIX(1.286413905)); /* c3 */ - - dataptr[1] = (DCTELEM) DESCALE(tmp0, CONST_BITS-1); - dataptr[3] = (DCTELEM) DESCALE(tmp1, CONST_BITS-1); - dataptr[5] = (DCTELEM) DESCALE(tmp2, CONST_BITS-1); - dataptr[7] = (DCTELEM) DESCALE(tmp3, CONST_BITS-1); - - ctr++; - - if (ctr != DCTSIZE) { - if (ctr == 11) - break; /* Done. */ - dataptr += DCTSIZE; /* advance pointer to next row */ - } else - dataptr = workspace; /* switch pointer to extended workspace */ - } - - /* Pass 2: process columns. - * We leave the results scaled up by an overall factor of 8. - * We must also scale the output by (8/11)**2 = 64/121, which we partially - * fold into the constant multipliers and final/initial shifting: - * cK now represents sqrt(2) * cos(K*pi/22) * 128/121. - */ - - dataptr = data; - wsptr = workspace; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + wsptr[DCTSIZE*2]; - tmp1 = dataptr[DCTSIZE*1] + wsptr[DCTSIZE*1]; - tmp2 = dataptr[DCTSIZE*2] + wsptr[DCTSIZE*0]; - tmp3 = dataptr[DCTSIZE*3] + dataptr[DCTSIZE*7]; - tmp4 = dataptr[DCTSIZE*4] + dataptr[DCTSIZE*6]; - tmp5 = dataptr[DCTSIZE*5]; - - tmp10 = dataptr[DCTSIZE*0] - wsptr[DCTSIZE*2]; - tmp11 = dataptr[DCTSIZE*1] - wsptr[DCTSIZE*1]; - tmp12 = dataptr[DCTSIZE*2] - wsptr[DCTSIZE*0]; - tmp13 = dataptr[DCTSIZE*3] - dataptr[DCTSIZE*7]; - tmp14 = dataptr[DCTSIZE*4] - dataptr[DCTSIZE*6]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 + tmp1 + tmp2 + tmp3 + tmp4 + tmp5, - FIX(1.057851240)), /* 128/121 */ - CONST_BITS+2); - tmp5 += tmp5; - tmp0 -= tmp5; - tmp1 -= tmp5; - tmp2 -= tmp5; - tmp3 -= tmp5; - tmp4 -= tmp5; - z1 = MULTIPLY(tmp0 + tmp3, FIX(1.435427942)) + /* c2 */ - MULTIPLY(tmp2 + tmp4, FIX(0.212906922)); /* c10 */ - z2 = MULTIPLY(tmp1 - tmp3, FIX(0.979689713)); /* c6 */ - z3 = MULTIPLY(tmp0 - tmp1, FIX(1.258538479)); /* c4 */ - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(z1 + z2 - MULTIPLY(tmp3, FIX(1.077210542)) /* c2+c8-c6 */ - - MULTIPLY(tmp4, FIX(1.471445400)), /* c4+c10 */ - CONST_BITS+2); - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(z2 + z3 + MULTIPLY(tmp1, FIX(0.065941844)) /* c4-c6-c10 */ - - MULTIPLY(tmp2, FIX(1.435427942)) /* c2 */ - + MULTIPLY(tmp4, FIX(0.621472312)), /* c8 */ - CONST_BITS+2); - dataptr[DCTSIZE*6] = (DCTELEM) - DESCALE(z1 + z3 - MULTIPLY(tmp0, FIX(1.714276708)) /* c2+c4-c6 */ - - MULTIPLY(tmp2, FIX(0.834379234)), /* c8+c10 */ - CONST_BITS+2); - - /* Odd part */ - - tmp1 = MULTIPLY(tmp10 + tmp11, FIX(1.360834544)); /* c3 */ - tmp2 = MULTIPLY(tmp10 + tmp12, FIX(1.130622199)); /* c5 */ - tmp3 = MULTIPLY(tmp10 + tmp13, FIX(0.808813568)); /* c7 */ - tmp0 = tmp1 + tmp2 + tmp3 - MULTIPLY(tmp10, FIX(1.819470145)) /* c7+c5+c3-c1 */ - + MULTIPLY(tmp14, FIX(0.421479672)); /* c9 */ - tmp4 = MULTIPLY(tmp11 + tmp12, - FIX(0.808813568)); /* -c7 */ - tmp5 = MULTIPLY(tmp11 + tmp13, - FIX(1.480800167)); /* -c1 */ - tmp1 += tmp4 + tmp5 + MULTIPLY(tmp11, FIX(1.350258864)) /* c9+c7+c1-c3 */ - - MULTIPLY(tmp14, FIX(1.130622199)); /* c5 */ - tmp10 = MULTIPLY(tmp12 + tmp13, FIX(0.421479672)); /* c9 */ - tmp2 += tmp4 + tmp10 - MULTIPLY(tmp12, FIX(2.104122847)) /* c9+c5+c3-c7 */ - + MULTIPLY(tmp14, FIX(1.480800167)); /* c1 */ - tmp3 += tmp5 + tmp10 + MULTIPLY(tmp13, FIX(1.381129125)) /* c1+c5-c9-c7 */ - - MULTIPLY(tmp14, FIX(1.360834544)); /* c3 */ - - dataptr[DCTSIZE*1] = (DCTELEM) DESCALE(tmp0, CONST_BITS+2); - dataptr[DCTSIZE*3] = (DCTELEM) DESCALE(tmp1, CONST_BITS+2); - dataptr[DCTSIZE*5] = (DCTELEM) DESCALE(tmp2, CONST_BITS+2); - dataptr[DCTSIZE*7] = (DCTELEM) DESCALE(tmp3, CONST_BITS+2); - - dataptr++; /* advance pointer to next column */ - wsptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 12x12 sample block. - */ - -GLOBAL(void) -jpeg_fdct_12x12 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15; - DCTELEM workspace[8*4]; - DCTELEM *dataptr; - DCTELEM *wsptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT. - * cK represents sqrt(2) * cos(K*pi/24). - */ - - dataptr = data; - ctr = 0; - for (;;) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[11]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[10]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[9]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[8]); - tmp4 = GETJSAMPLE(elemptr[4]) + GETJSAMPLE(elemptr[7]); - tmp5 = GETJSAMPLE(elemptr[5]) + GETJSAMPLE(elemptr[6]); - - tmp10 = tmp0 + tmp5; - tmp13 = tmp0 - tmp5; - tmp11 = tmp1 + tmp4; - tmp14 = tmp1 - tmp4; - tmp12 = tmp2 + tmp3; - tmp15 = tmp2 - tmp3; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[11]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[10]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[9]); - tmp3 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[8]); - tmp4 = GETJSAMPLE(elemptr[4]) - GETJSAMPLE(elemptr[7]); - tmp5 = GETJSAMPLE(elemptr[5]) - GETJSAMPLE(elemptr[6]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) (tmp10 + tmp11 + tmp12 - 12 * CENTERJSAMPLE); - dataptr[6] = (DCTELEM) (tmp13 - tmp14 - tmp15); - dataptr[4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp12, FIX(1.224744871)), /* c4 */ - CONST_BITS); - dataptr[2] = (DCTELEM) - DESCALE(tmp14 - tmp15 + MULTIPLY(tmp13 + tmp15, FIX(1.366025404)), /* c2 */ - CONST_BITS); - - /* Odd part */ - - tmp10 = MULTIPLY(tmp1 + tmp4, FIX_0_541196100); /* c9 */ - tmp14 = tmp10 + MULTIPLY(tmp1, FIX_0_765366865); /* c3-c9 */ - tmp15 = tmp10 - MULTIPLY(tmp4, FIX_1_847759065); /* c3+c9 */ - tmp12 = MULTIPLY(tmp0 + tmp2, FIX(1.121971054)); /* c5 */ - tmp13 = MULTIPLY(tmp0 + tmp3, FIX(0.860918669)); /* c7 */ - tmp10 = tmp12 + tmp13 + tmp14 - MULTIPLY(tmp0, FIX(0.580774953)) /* c5+c7-c1 */ - + MULTIPLY(tmp5, FIX(0.184591911)); /* c11 */ - tmp11 = MULTIPLY(tmp2 + tmp3, - FIX(0.184591911)); /* -c11 */ - tmp12 += tmp11 - tmp15 - MULTIPLY(tmp2, FIX(2.339493912)) /* c1+c5-c11 */ - + MULTIPLY(tmp5, FIX(0.860918669)); /* c7 */ - tmp13 += tmp11 - tmp14 + MULTIPLY(tmp3, FIX(0.725788011)) /* c1+c11-c7 */ - - MULTIPLY(tmp5, FIX(1.121971054)); /* c5 */ - tmp11 = tmp15 + MULTIPLY(tmp0 - tmp3, FIX(1.306562965)) /* c3 */ - - MULTIPLY(tmp2 + tmp5, FIX_0_541196100); /* c9 */ - - dataptr[1] = (DCTELEM) DESCALE(tmp10, CONST_BITS); - dataptr[3] = (DCTELEM) DESCALE(tmp11, CONST_BITS); - dataptr[5] = (DCTELEM) DESCALE(tmp12, CONST_BITS); - dataptr[7] = (DCTELEM) DESCALE(tmp13, CONST_BITS); - - ctr++; - - if (ctr != DCTSIZE) { - if (ctr == 12) - break; /* Done. */ - dataptr += DCTSIZE; /* advance pointer to next row */ - } else - dataptr = workspace; /* switch pointer to extended workspace */ - } - - /* Pass 2: process columns. - * We leave the results scaled up by an overall factor of 8. - * We must also scale the output by (8/12)**2 = 4/9, which we partially - * fold into the constant multipliers and final shifting: - * cK now represents sqrt(2) * cos(K*pi/24) * 8/9. - */ - - dataptr = data; - wsptr = workspace; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + wsptr[DCTSIZE*3]; - tmp1 = dataptr[DCTSIZE*1] + wsptr[DCTSIZE*2]; - tmp2 = dataptr[DCTSIZE*2] + wsptr[DCTSIZE*1]; - tmp3 = dataptr[DCTSIZE*3] + wsptr[DCTSIZE*0]; - tmp4 = dataptr[DCTSIZE*4] + dataptr[DCTSIZE*7]; - tmp5 = dataptr[DCTSIZE*5] + dataptr[DCTSIZE*6]; - - tmp10 = tmp0 + tmp5; - tmp13 = tmp0 - tmp5; - tmp11 = tmp1 + tmp4; - tmp14 = tmp1 - tmp4; - tmp12 = tmp2 + tmp3; - tmp15 = tmp2 - tmp3; - - tmp0 = dataptr[DCTSIZE*0] - wsptr[DCTSIZE*3]; - tmp1 = dataptr[DCTSIZE*1] - wsptr[DCTSIZE*2]; - tmp2 = dataptr[DCTSIZE*2] - wsptr[DCTSIZE*1]; - tmp3 = dataptr[DCTSIZE*3] - wsptr[DCTSIZE*0]; - tmp4 = dataptr[DCTSIZE*4] - dataptr[DCTSIZE*7]; - tmp5 = dataptr[DCTSIZE*5] - dataptr[DCTSIZE*6]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 + tmp11 + tmp12, FIX(0.888888889)), /* 8/9 */ - CONST_BITS+1); - dataptr[DCTSIZE*6] = (DCTELEM) - DESCALE(MULTIPLY(tmp13 - tmp14 - tmp15, FIX(0.888888889)), /* 8/9 */ - CONST_BITS+1); - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp12, FIX(1.088662108)), /* c4 */ - CONST_BITS+1); - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(MULTIPLY(tmp14 - tmp15, FIX(0.888888889)) + /* 8/9 */ - MULTIPLY(tmp13 + tmp15, FIX(1.214244803)), /* c2 */ - CONST_BITS+1); - - /* Odd part */ - - tmp10 = MULTIPLY(tmp1 + tmp4, FIX(0.481063200)); /* c9 */ - tmp14 = tmp10 + MULTIPLY(tmp1, FIX(0.680326102)); /* c3-c9 */ - tmp15 = tmp10 - MULTIPLY(tmp4, FIX(1.642452502)); /* c3+c9 */ - tmp12 = MULTIPLY(tmp0 + tmp2, FIX(0.997307603)); /* c5 */ - tmp13 = MULTIPLY(tmp0 + tmp3, FIX(0.765261039)); /* c7 */ - tmp10 = tmp12 + tmp13 + tmp14 - MULTIPLY(tmp0, FIX(0.516244403)) /* c5+c7-c1 */ - + MULTIPLY(tmp5, FIX(0.164081699)); /* c11 */ - tmp11 = MULTIPLY(tmp2 + tmp3, - FIX(0.164081699)); /* -c11 */ - tmp12 += tmp11 - tmp15 - MULTIPLY(tmp2, FIX(2.079550144)) /* c1+c5-c11 */ - + MULTIPLY(tmp5, FIX(0.765261039)); /* c7 */ - tmp13 += tmp11 - tmp14 + MULTIPLY(tmp3, FIX(0.645144899)) /* c1+c11-c7 */ - - MULTIPLY(tmp5, FIX(0.997307603)); /* c5 */ - tmp11 = tmp15 + MULTIPLY(tmp0 - tmp3, FIX(1.161389302)) /* c3 */ - - MULTIPLY(tmp2 + tmp5, FIX(0.481063200)); /* c9 */ - - dataptr[DCTSIZE*1] = (DCTELEM) DESCALE(tmp10, CONST_BITS+1); - dataptr[DCTSIZE*3] = (DCTELEM) DESCALE(tmp11, CONST_BITS+1); - dataptr[DCTSIZE*5] = (DCTELEM) DESCALE(tmp12, CONST_BITS+1); - dataptr[DCTSIZE*7] = (DCTELEM) DESCALE(tmp13, CONST_BITS+1); - - dataptr++; /* advance pointer to next column */ - wsptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 13x13 sample block. - */ - -GLOBAL(void) -jpeg_fdct_13x13 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15; - INT32 z1, z2; - DCTELEM workspace[8*5]; - DCTELEM *dataptr; - DCTELEM *wsptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT. - * cK represents sqrt(2) * cos(K*pi/26). - */ - - dataptr = data; - ctr = 0; - for (;;) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[12]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[11]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[10]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[9]); - tmp4 = GETJSAMPLE(elemptr[4]) + GETJSAMPLE(elemptr[8]); - tmp5 = GETJSAMPLE(elemptr[5]) + GETJSAMPLE(elemptr[7]); - tmp6 = GETJSAMPLE(elemptr[6]); - - tmp10 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[12]); - tmp11 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[11]); - tmp12 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[10]); - tmp13 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[9]); - tmp14 = GETJSAMPLE(elemptr[4]) - GETJSAMPLE(elemptr[8]); - tmp15 = GETJSAMPLE(elemptr[5]) - GETJSAMPLE(elemptr[7]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - (tmp0 + tmp1 + tmp2 + tmp3 + tmp4 + tmp5 + tmp6 - 13 * CENTERJSAMPLE); - tmp6 += tmp6; - tmp0 -= tmp6; - tmp1 -= tmp6; - tmp2 -= tmp6; - tmp3 -= tmp6; - tmp4 -= tmp6; - tmp5 -= tmp6; - dataptr[2] = (DCTELEM) - DESCALE(MULTIPLY(tmp0, FIX(1.373119086)) + /* c2 */ - MULTIPLY(tmp1, FIX(1.058554052)) + /* c6 */ - MULTIPLY(tmp2, FIX(0.501487041)) - /* c10 */ - MULTIPLY(tmp3, FIX(0.170464608)) - /* c12 */ - MULTIPLY(tmp4, FIX(0.803364869)) - /* c8 */ - MULTIPLY(tmp5, FIX(1.252223920)), /* c4 */ - CONST_BITS); - z1 = MULTIPLY(tmp0 - tmp2, FIX(1.155388986)) - /* (c4+c6)/2 */ - MULTIPLY(tmp3 - tmp4, FIX(0.435816023)) - /* (c2-c10)/2 */ - MULTIPLY(tmp1 - tmp5, FIX(0.316450131)); /* (c8-c12)/2 */ - z2 = MULTIPLY(tmp0 + tmp2, FIX(0.096834934)) - /* (c4-c6)/2 */ - MULTIPLY(tmp3 + tmp4, FIX(0.937303064)) + /* (c2+c10)/2 */ - MULTIPLY(tmp1 + tmp5, FIX(0.486914739)); /* (c8+c12)/2 */ - - dataptr[4] = (DCTELEM) DESCALE(z1 + z2, CONST_BITS); - dataptr[6] = (DCTELEM) DESCALE(z1 - z2, CONST_BITS); - - /* Odd part */ - - tmp1 = MULTIPLY(tmp10 + tmp11, FIX(1.322312651)); /* c3 */ - tmp2 = MULTIPLY(tmp10 + tmp12, FIX(1.163874945)); /* c5 */ - tmp3 = MULTIPLY(tmp10 + tmp13, FIX(0.937797057)) + /* c7 */ - MULTIPLY(tmp14 + tmp15, FIX(0.338443458)); /* c11 */ - tmp0 = tmp1 + tmp2 + tmp3 - - MULTIPLY(tmp10, FIX(2.020082300)) + /* c3+c5+c7-c1 */ - MULTIPLY(tmp14, FIX(0.318774355)); /* c9-c11 */ - tmp4 = MULTIPLY(tmp14 - tmp15, FIX(0.937797057)) - /* c7 */ - MULTIPLY(tmp11 + tmp12, FIX(0.338443458)); /* c11 */ - tmp5 = MULTIPLY(tmp11 + tmp13, - FIX(1.163874945)); /* -c5 */ - tmp1 += tmp4 + tmp5 + - MULTIPLY(tmp11, FIX(0.837223564)) - /* c5+c9+c11-c3 */ - MULTIPLY(tmp14, FIX(2.341699410)); /* c1+c7 */ - tmp6 = MULTIPLY(tmp12 + tmp13, - FIX(0.657217813)); /* -c9 */ - tmp2 += tmp4 + tmp6 - - MULTIPLY(tmp12, FIX(1.572116027)) + /* c1+c5-c9-c11 */ - MULTIPLY(tmp15, FIX(2.260109708)); /* c3+c7 */ - tmp3 += tmp5 + tmp6 + - MULTIPLY(tmp13, FIX(2.205608352)) - /* c3+c5+c9-c7 */ - MULTIPLY(tmp15, FIX(1.742345811)); /* c1+c11 */ - - dataptr[1] = (DCTELEM) DESCALE(tmp0, CONST_BITS); - dataptr[3] = (DCTELEM) DESCALE(tmp1, CONST_BITS); - dataptr[5] = (DCTELEM) DESCALE(tmp2, CONST_BITS); - dataptr[7] = (DCTELEM) DESCALE(tmp3, CONST_BITS); - - ctr++; - - if (ctr != DCTSIZE) { - if (ctr == 13) - break; /* Done. */ - dataptr += DCTSIZE; /* advance pointer to next row */ - } else - dataptr = workspace; /* switch pointer to extended workspace */ - } - - /* Pass 2: process columns. - * We leave the results scaled up by an overall factor of 8. - * We must also scale the output by (8/13)**2 = 64/169, which we partially - * fold into the constant multipliers and final shifting: - * cK now represents sqrt(2) * cos(K*pi/26) * 128/169. - */ - - dataptr = data; - wsptr = workspace; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + wsptr[DCTSIZE*4]; - tmp1 = dataptr[DCTSIZE*1] + wsptr[DCTSIZE*3]; - tmp2 = dataptr[DCTSIZE*2] + wsptr[DCTSIZE*2]; - tmp3 = dataptr[DCTSIZE*3] + wsptr[DCTSIZE*1]; - tmp4 = dataptr[DCTSIZE*4] + wsptr[DCTSIZE*0]; - tmp5 = dataptr[DCTSIZE*5] + dataptr[DCTSIZE*7]; - tmp6 = dataptr[DCTSIZE*6]; - - tmp10 = dataptr[DCTSIZE*0] - wsptr[DCTSIZE*4]; - tmp11 = dataptr[DCTSIZE*1] - wsptr[DCTSIZE*3]; - tmp12 = dataptr[DCTSIZE*2] - wsptr[DCTSIZE*2]; - tmp13 = dataptr[DCTSIZE*3] - wsptr[DCTSIZE*1]; - tmp14 = dataptr[DCTSIZE*4] - wsptr[DCTSIZE*0]; - tmp15 = dataptr[DCTSIZE*5] - dataptr[DCTSIZE*7]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 + tmp1 + tmp2 + tmp3 + tmp4 + tmp5 + tmp6, - FIX(0.757396450)), /* 128/169 */ - CONST_BITS+1); - tmp6 += tmp6; - tmp0 -= tmp6; - tmp1 -= tmp6; - tmp2 -= tmp6; - tmp3 -= tmp6; - tmp4 -= tmp6; - tmp5 -= tmp6; - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(MULTIPLY(tmp0, FIX(1.039995521)) + /* c2 */ - MULTIPLY(tmp1, FIX(0.801745081)) + /* c6 */ - MULTIPLY(tmp2, FIX(0.379824504)) - /* c10 */ - MULTIPLY(tmp3, FIX(0.129109289)) - /* c12 */ - MULTIPLY(tmp4, FIX(0.608465700)) - /* c8 */ - MULTIPLY(tmp5, FIX(0.948429952)), /* c4 */ - CONST_BITS+1); - z1 = MULTIPLY(tmp0 - tmp2, FIX(0.875087516)) - /* (c4+c6)/2 */ - MULTIPLY(tmp3 - tmp4, FIX(0.330085509)) - /* (c2-c10)/2 */ - MULTIPLY(tmp1 - tmp5, FIX(0.239678205)); /* (c8-c12)/2 */ - z2 = MULTIPLY(tmp0 + tmp2, FIX(0.073342435)) - /* (c4-c6)/2 */ - MULTIPLY(tmp3 + tmp4, FIX(0.709910013)) + /* (c2+c10)/2 */ - MULTIPLY(tmp1 + tmp5, FIX(0.368787494)); /* (c8+c12)/2 */ - - dataptr[DCTSIZE*4] = (DCTELEM) DESCALE(z1 + z2, CONST_BITS+1); - dataptr[DCTSIZE*6] = (DCTELEM) DESCALE(z1 - z2, CONST_BITS+1); - - /* Odd part */ - - tmp1 = MULTIPLY(tmp10 + tmp11, FIX(1.001514908)); /* c3 */ - tmp2 = MULTIPLY(tmp10 + tmp12, FIX(0.881514751)); /* c5 */ - tmp3 = MULTIPLY(tmp10 + tmp13, FIX(0.710284161)) + /* c7 */ - MULTIPLY(tmp14 + tmp15, FIX(0.256335874)); /* c11 */ - tmp0 = tmp1 + tmp2 + tmp3 - - MULTIPLY(tmp10, FIX(1.530003162)) + /* c3+c5+c7-c1 */ - MULTIPLY(tmp14, FIX(0.241438564)); /* c9-c11 */ - tmp4 = MULTIPLY(tmp14 - tmp15, FIX(0.710284161)) - /* c7 */ - MULTIPLY(tmp11 + tmp12, FIX(0.256335874)); /* c11 */ - tmp5 = MULTIPLY(tmp11 + tmp13, - FIX(0.881514751)); /* -c5 */ - tmp1 += tmp4 + tmp5 + - MULTIPLY(tmp11, FIX(0.634110155)) - /* c5+c9+c11-c3 */ - MULTIPLY(tmp14, FIX(1.773594819)); /* c1+c7 */ - tmp6 = MULTIPLY(tmp12 + tmp13, - FIX(0.497774438)); /* -c9 */ - tmp2 += tmp4 + tmp6 - - MULTIPLY(tmp12, FIX(1.190715098)) + /* c1+c5-c9-c11 */ - MULTIPLY(tmp15, FIX(1.711799069)); /* c3+c7 */ - tmp3 += tmp5 + tmp6 + - MULTIPLY(tmp13, FIX(1.670519935)) - /* c3+c5+c9-c7 */ - MULTIPLY(tmp15, FIX(1.319646532)); /* c1+c11 */ - - dataptr[DCTSIZE*1] = (DCTELEM) DESCALE(tmp0, CONST_BITS+1); - dataptr[DCTSIZE*3] = (DCTELEM) DESCALE(tmp1, CONST_BITS+1); - dataptr[DCTSIZE*5] = (DCTELEM) DESCALE(tmp2, CONST_BITS+1); - dataptr[DCTSIZE*7] = (DCTELEM) DESCALE(tmp3, CONST_BITS+1); - - dataptr++; /* advance pointer to next column */ - wsptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 14x14 sample block. - */ - -GLOBAL(void) -jpeg_fdct_14x14 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15, tmp16; - DCTELEM workspace[8*6]; - DCTELEM *dataptr; - DCTELEM *wsptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT. - * cK represents sqrt(2) * cos(K*pi/28). - */ - - dataptr = data; - ctr = 0; - for (;;) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[13]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[12]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[11]); - tmp13 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[10]); - tmp4 = GETJSAMPLE(elemptr[4]) + GETJSAMPLE(elemptr[9]); - tmp5 = GETJSAMPLE(elemptr[5]) + GETJSAMPLE(elemptr[8]); - tmp6 = GETJSAMPLE(elemptr[6]) + GETJSAMPLE(elemptr[7]); - - tmp10 = tmp0 + tmp6; - tmp14 = tmp0 - tmp6; - tmp11 = tmp1 + tmp5; - tmp15 = tmp1 - tmp5; - tmp12 = tmp2 + tmp4; - tmp16 = tmp2 - tmp4; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[13]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[12]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[11]); - tmp3 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[10]); - tmp4 = GETJSAMPLE(elemptr[4]) - GETJSAMPLE(elemptr[9]); - tmp5 = GETJSAMPLE(elemptr[5]) - GETJSAMPLE(elemptr[8]); - tmp6 = GETJSAMPLE(elemptr[6]) - GETJSAMPLE(elemptr[7]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - (tmp10 + tmp11 + tmp12 + tmp13 - 14 * CENTERJSAMPLE); - tmp13 += tmp13; - dataptr[4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp13, FIX(1.274162392)) + /* c4 */ - MULTIPLY(tmp11 - tmp13, FIX(0.314692123)) - /* c12 */ - MULTIPLY(tmp12 - tmp13, FIX(0.881747734)), /* c8 */ - CONST_BITS); - - tmp10 = MULTIPLY(tmp14 + tmp15, FIX(1.105676686)); /* c6 */ - - dataptr[2] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp14, FIX(0.273079590)) /* c2-c6 */ - + MULTIPLY(tmp16, FIX(0.613604268)), /* c10 */ - CONST_BITS); - dataptr[6] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp15, FIX(1.719280954)) /* c6+c10 */ - - MULTIPLY(tmp16, FIX(1.378756276)), /* c2 */ - CONST_BITS); - - /* Odd part */ - - tmp10 = tmp1 + tmp2; - tmp11 = tmp5 - tmp4; - dataptr[7] = (DCTELEM) (tmp0 - tmp10 + tmp3 - tmp11 - tmp6); - tmp3 <<= CONST_BITS; - tmp10 = MULTIPLY(tmp10, - FIX(0.158341681)); /* -c13 */ - tmp11 = MULTIPLY(tmp11, FIX(1.405321284)); /* c1 */ - tmp10 += tmp11 - tmp3; - tmp11 = MULTIPLY(tmp0 + tmp2, FIX(1.197448846)) + /* c5 */ - MULTIPLY(tmp4 + tmp6, FIX(0.752406978)); /* c9 */ - dataptr[5] = (DCTELEM) - DESCALE(tmp10 + tmp11 - MULTIPLY(tmp2, FIX(2.373959773)) /* c3+c5-c13 */ - + MULTIPLY(tmp4, FIX(1.119999435)), /* c1+c11-c9 */ - CONST_BITS); - tmp12 = MULTIPLY(tmp0 + tmp1, FIX(1.334852607)) + /* c3 */ - MULTIPLY(tmp5 - tmp6, FIX(0.467085129)); /* c11 */ - dataptr[3] = (DCTELEM) - DESCALE(tmp10 + tmp12 - MULTIPLY(tmp1, FIX(0.424103948)) /* c3-c9-c13 */ - - MULTIPLY(tmp5, FIX(3.069855259)), /* c1+c5+c11 */ - CONST_BITS); - dataptr[1] = (DCTELEM) - DESCALE(tmp11 + tmp12 + tmp3 + tmp6 - - MULTIPLY(tmp0 + tmp6, FIX(1.126980169)), /* c3+c5-c1 */ - CONST_BITS); - - ctr++; - - if (ctr != DCTSIZE) { - if (ctr == 14) - break; /* Done. */ - dataptr += DCTSIZE; /* advance pointer to next row */ - } else - dataptr = workspace; /* switch pointer to extended workspace */ - } - - /* Pass 2: process columns. - * We leave the results scaled up by an overall factor of 8. - * We must also scale the output by (8/14)**2 = 16/49, which we partially - * fold into the constant multipliers and final shifting: - * cK now represents sqrt(2) * cos(K*pi/28) * 32/49. - */ - - dataptr = data; - wsptr = workspace; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + wsptr[DCTSIZE*5]; - tmp1 = dataptr[DCTSIZE*1] + wsptr[DCTSIZE*4]; - tmp2 = dataptr[DCTSIZE*2] + wsptr[DCTSIZE*3]; - tmp13 = dataptr[DCTSIZE*3] + wsptr[DCTSIZE*2]; - tmp4 = dataptr[DCTSIZE*4] + wsptr[DCTSIZE*1]; - tmp5 = dataptr[DCTSIZE*5] + wsptr[DCTSIZE*0]; - tmp6 = dataptr[DCTSIZE*6] + dataptr[DCTSIZE*7]; - - tmp10 = tmp0 + tmp6; - tmp14 = tmp0 - tmp6; - tmp11 = tmp1 + tmp5; - tmp15 = tmp1 - tmp5; - tmp12 = tmp2 + tmp4; - tmp16 = tmp2 - tmp4; - - tmp0 = dataptr[DCTSIZE*0] - wsptr[DCTSIZE*5]; - tmp1 = dataptr[DCTSIZE*1] - wsptr[DCTSIZE*4]; - tmp2 = dataptr[DCTSIZE*2] - wsptr[DCTSIZE*3]; - tmp3 = dataptr[DCTSIZE*3] - wsptr[DCTSIZE*2]; - tmp4 = dataptr[DCTSIZE*4] - wsptr[DCTSIZE*1]; - tmp5 = dataptr[DCTSIZE*5] - wsptr[DCTSIZE*0]; - tmp6 = dataptr[DCTSIZE*6] - dataptr[DCTSIZE*7]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 + tmp11 + tmp12 + tmp13, - FIX(0.653061224)), /* 32/49 */ - CONST_BITS+1); - tmp13 += tmp13; - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp13, FIX(0.832106052)) + /* c4 */ - MULTIPLY(tmp11 - tmp13, FIX(0.205513223)) - /* c12 */ - MULTIPLY(tmp12 - tmp13, FIX(0.575835255)), /* c8 */ - CONST_BITS+1); - - tmp10 = MULTIPLY(tmp14 + tmp15, FIX(0.722074570)); /* c6 */ - - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp14, FIX(0.178337691)) /* c2-c6 */ - + MULTIPLY(tmp16, FIX(0.400721155)), /* c10 */ - CONST_BITS+1); - dataptr[DCTSIZE*6] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp15, FIX(1.122795725)) /* c6+c10 */ - - MULTIPLY(tmp16, FIX(0.900412262)), /* c2 */ - CONST_BITS+1); - - /* Odd part */ - - tmp10 = tmp1 + tmp2; - tmp11 = tmp5 - tmp4; - dataptr[DCTSIZE*7] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 - tmp10 + tmp3 - tmp11 - tmp6, - FIX(0.653061224)), /* 32/49 */ - CONST_BITS+1); - tmp3 = MULTIPLY(tmp3 , FIX(0.653061224)); /* 32/49 */ - tmp10 = MULTIPLY(tmp10, - FIX(0.103406812)); /* -c13 */ - tmp11 = MULTIPLY(tmp11, FIX(0.917760839)); /* c1 */ - tmp10 += tmp11 - tmp3; - tmp11 = MULTIPLY(tmp0 + tmp2, FIX(0.782007410)) + /* c5 */ - MULTIPLY(tmp4 + tmp6, FIX(0.491367823)); /* c9 */ - dataptr[DCTSIZE*5] = (DCTELEM) - DESCALE(tmp10 + tmp11 - MULTIPLY(tmp2, FIX(1.550341076)) /* c3+c5-c13 */ - + MULTIPLY(tmp4, FIX(0.731428202)), /* c1+c11-c9 */ - CONST_BITS+1); - tmp12 = MULTIPLY(tmp0 + tmp1, FIX(0.871740478)) + /* c3 */ - MULTIPLY(tmp5 - tmp6, FIX(0.305035186)); /* c11 */ - dataptr[DCTSIZE*3] = (DCTELEM) - DESCALE(tmp10 + tmp12 - MULTIPLY(tmp1, FIX(0.276965844)) /* c3-c9-c13 */ - - MULTIPLY(tmp5, FIX(2.004803435)), /* c1+c5+c11 */ - CONST_BITS+1); - dataptr[DCTSIZE*1] = (DCTELEM) - DESCALE(tmp11 + tmp12 + tmp3 - - MULTIPLY(tmp0, FIX(0.735987049)) /* c3+c5-c1 */ - - MULTIPLY(tmp6, FIX(0.082925825)), /* c9-c11-c13 */ - CONST_BITS+1); - - dataptr++; /* advance pointer to next column */ - wsptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 15x15 sample block. - */ - -GLOBAL(void) -jpeg_fdct_15x15 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15, tmp16; - INT32 z1, z2, z3; - DCTELEM workspace[8*7]; - DCTELEM *dataptr; - DCTELEM *wsptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT. - * cK represents sqrt(2) * cos(K*pi/30). - */ - - dataptr = data; - ctr = 0; - for (;;) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[14]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[13]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[12]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[11]); - tmp4 = GETJSAMPLE(elemptr[4]) + GETJSAMPLE(elemptr[10]); - tmp5 = GETJSAMPLE(elemptr[5]) + GETJSAMPLE(elemptr[9]); - tmp6 = GETJSAMPLE(elemptr[6]) + GETJSAMPLE(elemptr[8]); - tmp7 = GETJSAMPLE(elemptr[7]); - - tmp10 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[14]); - tmp11 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[13]); - tmp12 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[12]); - tmp13 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[11]); - tmp14 = GETJSAMPLE(elemptr[4]) - GETJSAMPLE(elemptr[10]); - tmp15 = GETJSAMPLE(elemptr[5]) - GETJSAMPLE(elemptr[9]); - tmp16 = GETJSAMPLE(elemptr[6]) - GETJSAMPLE(elemptr[8]); - - z1 = tmp0 + tmp4 + tmp5; - z2 = tmp1 + tmp3 + tmp6; - z3 = tmp2 + tmp7; - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) (z1 + z2 + z3 - 15 * CENTERJSAMPLE); - z3 += z3; - dataptr[6] = (DCTELEM) - DESCALE(MULTIPLY(z1 - z3, FIX(1.144122806)) - /* c6 */ - MULTIPLY(z2 - z3, FIX(0.437016024)), /* c12 */ - CONST_BITS); - tmp2 += ((tmp1 + tmp4) >> 1) - tmp7 - tmp7; - z1 = MULTIPLY(tmp3 - tmp2, FIX(1.531135173)) - /* c2+c14 */ - MULTIPLY(tmp6 - tmp2, FIX(2.238241955)); /* c4+c8 */ - z2 = MULTIPLY(tmp5 - tmp2, FIX(0.798468008)) - /* c8-c14 */ - MULTIPLY(tmp0 - tmp2, FIX(0.091361227)); /* c2-c4 */ - z3 = MULTIPLY(tmp0 - tmp3, FIX(1.383309603)) + /* c2 */ - MULTIPLY(tmp6 - tmp5, FIX(0.946293579)) + /* c8 */ - MULTIPLY(tmp1 - tmp4, FIX(0.790569415)); /* (c6+c12)/2 */ - - dataptr[2] = (DCTELEM) DESCALE(z1 + z3, CONST_BITS); - dataptr[4] = (DCTELEM) DESCALE(z2 + z3, CONST_BITS); - - /* Odd part */ - - tmp2 = MULTIPLY(tmp10 - tmp12 - tmp13 + tmp15 + tmp16, - FIX(1.224744871)); /* c5 */ - tmp1 = MULTIPLY(tmp10 - tmp14 - tmp15, FIX(1.344997024)) + /* c3 */ - MULTIPLY(tmp11 - tmp13 - tmp16, FIX(0.831253876)); /* c9 */ - tmp12 = MULTIPLY(tmp12, FIX(1.224744871)); /* c5 */ - tmp4 = MULTIPLY(tmp10 - tmp16, FIX(1.406466353)) + /* c1 */ - MULTIPLY(tmp11 + tmp14, FIX(1.344997024)) + /* c3 */ - MULTIPLY(tmp13 + tmp15, FIX(0.575212477)); /* c11 */ - tmp0 = MULTIPLY(tmp13, FIX(0.475753014)) - /* c7-c11 */ - MULTIPLY(tmp14, FIX(0.513743148)) + /* c3-c9 */ - MULTIPLY(tmp16, FIX(1.700497885)) + tmp4 + tmp12; /* c1+c13 */ - tmp3 = MULTIPLY(tmp10, - FIX(0.355500862)) - /* -(c1-c7) */ - MULTIPLY(tmp11, FIX(2.176250899)) - /* c3+c9 */ - MULTIPLY(tmp15, FIX(0.869244010)) + tmp4 - tmp12; /* c11+c13 */ - - dataptr[1] = (DCTELEM) DESCALE(tmp0, CONST_BITS); - dataptr[3] = (DCTELEM) DESCALE(tmp1, CONST_BITS); - dataptr[5] = (DCTELEM) DESCALE(tmp2, CONST_BITS); - dataptr[7] = (DCTELEM) DESCALE(tmp3, CONST_BITS); - - ctr++; - - if (ctr != DCTSIZE) { - if (ctr == 15) - break; /* Done. */ - dataptr += DCTSIZE; /* advance pointer to next row */ - } else - dataptr = workspace; /* switch pointer to extended workspace */ - } - - /* Pass 2: process columns. - * We leave the results scaled up by an overall factor of 8. - * We must also scale the output by (8/15)**2 = 64/225, which we partially - * fold into the constant multipliers and final shifting: - * cK now represents sqrt(2) * cos(K*pi/30) * 256/225. - */ - - dataptr = data; - wsptr = workspace; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + wsptr[DCTSIZE*6]; - tmp1 = dataptr[DCTSIZE*1] + wsptr[DCTSIZE*5]; - tmp2 = dataptr[DCTSIZE*2] + wsptr[DCTSIZE*4]; - tmp3 = dataptr[DCTSIZE*3] + wsptr[DCTSIZE*3]; - tmp4 = dataptr[DCTSIZE*4] + wsptr[DCTSIZE*2]; - tmp5 = dataptr[DCTSIZE*5] + wsptr[DCTSIZE*1]; - tmp6 = dataptr[DCTSIZE*6] + wsptr[DCTSIZE*0]; - tmp7 = dataptr[DCTSIZE*7]; - - tmp10 = dataptr[DCTSIZE*0] - wsptr[DCTSIZE*6]; - tmp11 = dataptr[DCTSIZE*1] - wsptr[DCTSIZE*5]; - tmp12 = dataptr[DCTSIZE*2] - wsptr[DCTSIZE*4]; - tmp13 = dataptr[DCTSIZE*3] - wsptr[DCTSIZE*3]; - tmp14 = dataptr[DCTSIZE*4] - wsptr[DCTSIZE*2]; - tmp15 = dataptr[DCTSIZE*5] - wsptr[DCTSIZE*1]; - tmp16 = dataptr[DCTSIZE*6] - wsptr[DCTSIZE*0]; - - z1 = tmp0 + tmp4 + tmp5; - z2 = tmp1 + tmp3 + tmp6; - z3 = tmp2 + tmp7; - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(z1 + z2 + z3, FIX(1.137777778)), /* 256/225 */ - CONST_BITS+2); - z3 += z3; - dataptr[DCTSIZE*6] = (DCTELEM) - DESCALE(MULTIPLY(z1 - z3, FIX(1.301757503)) - /* c6 */ - MULTIPLY(z2 - z3, FIX(0.497227121)), /* c12 */ - CONST_BITS+2); - tmp2 += ((tmp1 + tmp4) >> 1) - tmp7 - tmp7; - z1 = MULTIPLY(tmp3 - tmp2, FIX(1.742091575)) - /* c2+c14 */ - MULTIPLY(tmp6 - tmp2, FIX(2.546621957)); /* c4+c8 */ - z2 = MULTIPLY(tmp5 - tmp2, FIX(0.908479156)) - /* c8-c14 */ - MULTIPLY(tmp0 - tmp2, FIX(0.103948774)); /* c2-c4 */ - z3 = MULTIPLY(tmp0 - tmp3, FIX(1.573898926)) + /* c2 */ - MULTIPLY(tmp6 - tmp5, FIX(1.076671805)) + /* c8 */ - MULTIPLY(tmp1 - tmp4, FIX(0.899492312)); /* (c6+c12)/2 */ - - dataptr[DCTSIZE*2] = (DCTELEM) DESCALE(z1 + z3, CONST_BITS+2); - dataptr[DCTSIZE*4] = (DCTELEM) DESCALE(z2 + z3, CONST_BITS+2); - - /* Odd part */ - - tmp2 = MULTIPLY(tmp10 - tmp12 - tmp13 + tmp15 + tmp16, - FIX(1.393487498)); /* c5 */ - tmp1 = MULTIPLY(tmp10 - tmp14 - tmp15, FIX(1.530307725)) + /* c3 */ - MULTIPLY(tmp11 - tmp13 - tmp16, FIX(0.945782187)); /* c9 */ - tmp12 = MULTIPLY(tmp12, FIX(1.393487498)); /* c5 */ - tmp4 = MULTIPLY(tmp10 - tmp16, FIX(1.600246161)) + /* c1 */ - MULTIPLY(tmp11 + tmp14, FIX(1.530307725)) + /* c3 */ - MULTIPLY(tmp13 + tmp15, FIX(0.654463974)); /* c11 */ - tmp0 = MULTIPLY(tmp13, FIX(0.541301207)) - /* c7-c11 */ - MULTIPLY(tmp14, FIX(0.584525538)) + /* c3-c9 */ - MULTIPLY(tmp16, FIX(1.934788705)) + tmp4 + tmp12; /* c1+c13 */ - tmp3 = MULTIPLY(tmp10, - FIX(0.404480980)) - /* -(c1-c7) */ - MULTIPLY(tmp11, FIX(2.476089912)) - /* c3+c9 */ - MULTIPLY(tmp15, FIX(0.989006518)) + tmp4 - tmp12; /* c11+c13 */ - - dataptr[DCTSIZE*1] = (DCTELEM) DESCALE(tmp0, CONST_BITS+2); - dataptr[DCTSIZE*3] = (DCTELEM) DESCALE(tmp1, CONST_BITS+2); - dataptr[DCTSIZE*5] = (DCTELEM) DESCALE(tmp2, CONST_BITS+2); - dataptr[DCTSIZE*7] = (DCTELEM) DESCALE(tmp3, CONST_BITS+2); - - dataptr++; /* advance pointer to next column */ - wsptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 16x16 sample block. - */ - -GLOBAL(void) -jpeg_fdct_16x16 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15, tmp16, tmp17; - DCTELEM workspace[DCTSIZE2]; - DCTELEM *dataptr; - DCTELEM *wsptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * cK represents sqrt(2) * cos(K*pi/32). - */ - - dataptr = data; - ctr = 0; - for (;;) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[15]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[14]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[13]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[12]); - tmp4 = GETJSAMPLE(elemptr[4]) + GETJSAMPLE(elemptr[11]); - tmp5 = GETJSAMPLE(elemptr[5]) + GETJSAMPLE(elemptr[10]); - tmp6 = GETJSAMPLE(elemptr[6]) + GETJSAMPLE(elemptr[9]); - tmp7 = GETJSAMPLE(elemptr[7]) + GETJSAMPLE(elemptr[8]); - - tmp10 = tmp0 + tmp7; - tmp14 = tmp0 - tmp7; - tmp11 = tmp1 + tmp6; - tmp15 = tmp1 - tmp6; - tmp12 = tmp2 + tmp5; - tmp16 = tmp2 - tmp5; - tmp13 = tmp3 + tmp4; - tmp17 = tmp3 - tmp4; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[15]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[14]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[13]); - tmp3 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[12]); - tmp4 = GETJSAMPLE(elemptr[4]) - GETJSAMPLE(elemptr[11]); - tmp5 = GETJSAMPLE(elemptr[5]) - GETJSAMPLE(elemptr[10]); - tmp6 = GETJSAMPLE(elemptr[6]) - GETJSAMPLE(elemptr[9]); - tmp7 = GETJSAMPLE(elemptr[7]) - GETJSAMPLE(elemptr[8]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp10 + tmp11 + tmp12 + tmp13 - 16 * CENTERJSAMPLE) << PASS1_BITS); - dataptr[4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp13, FIX(1.306562965)) + /* c4[16] = c2[8] */ - MULTIPLY(tmp11 - tmp12, FIX_0_541196100), /* c12[16] = c6[8] */ - CONST_BITS-PASS1_BITS); - - tmp10 = MULTIPLY(tmp17 - tmp15, FIX(0.275899379)) + /* c14[16] = c7[8] */ - MULTIPLY(tmp14 - tmp16, FIX(1.387039845)); /* c2[16] = c1[8] */ - - dataptr[2] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp15, FIX(1.451774982)) /* c6+c14 */ - + MULTIPLY(tmp16, FIX(2.172734804)), /* c2+c10 */ - CONST_BITS-PASS1_BITS); - dataptr[6] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp14, FIX(0.211164243)) /* c2-c6 */ - - MULTIPLY(tmp17, FIX(1.061594338)), /* c10+c14 */ - CONST_BITS-PASS1_BITS); - - /* Odd part */ - - tmp11 = MULTIPLY(tmp0 + tmp1, FIX(1.353318001)) + /* c3 */ - MULTIPLY(tmp6 - tmp7, FIX(0.410524528)); /* c13 */ - tmp12 = MULTIPLY(tmp0 + tmp2, FIX(1.247225013)) + /* c5 */ - MULTIPLY(tmp5 + tmp7, FIX(0.666655658)); /* c11 */ - tmp13 = MULTIPLY(tmp0 + tmp3, FIX(1.093201867)) + /* c7 */ - MULTIPLY(tmp4 - tmp7, FIX(0.897167586)); /* c9 */ - tmp14 = MULTIPLY(tmp1 + tmp2, FIX(0.138617169)) + /* c15 */ - MULTIPLY(tmp6 - tmp5, FIX(1.407403738)); /* c1 */ - tmp15 = MULTIPLY(tmp1 + tmp3, - FIX(0.666655658)) + /* -c11 */ - MULTIPLY(tmp4 + tmp6, - FIX(1.247225013)); /* -c5 */ - tmp16 = MULTIPLY(tmp2 + tmp3, - FIX(1.353318001)) + /* -c3 */ - MULTIPLY(tmp5 - tmp4, FIX(0.410524528)); /* c13 */ - tmp10 = tmp11 + tmp12 + tmp13 - - MULTIPLY(tmp0, FIX(2.286341144)) + /* c7+c5+c3-c1 */ - MULTIPLY(tmp7, FIX(0.779653625)); /* c15+c13-c11+c9 */ - tmp11 += tmp14 + tmp15 + MULTIPLY(tmp1, FIX(0.071888074)) /* c9-c3-c15+c11 */ - - MULTIPLY(tmp6, FIX(1.663905119)); /* c7+c13+c1-c5 */ - tmp12 += tmp14 + tmp16 - MULTIPLY(tmp2, FIX(1.125726048)) /* c7+c5+c15-c3 */ - + MULTIPLY(tmp5, FIX(1.227391138)); /* c9-c11+c1-c13 */ - tmp13 += tmp15 + tmp16 + MULTIPLY(tmp3, FIX(1.065388962)) /* c15+c3+c11-c7 */ - + MULTIPLY(tmp4, FIX(2.167985692)); /* c1+c13+c5-c9 */ - - dataptr[1] = (DCTELEM) DESCALE(tmp10, CONST_BITS-PASS1_BITS); - dataptr[3] = (DCTELEM) DESCALE(tmp11, CONST_BITS-PASS1_BITS); - dataptr[5] = (DCTELEM) DESCALE(tmp12, CONST_BITS-PASS1_BITS); - dataptr[7] = (DCTELEM) DESCALE(tmp13, CONST_BITS-PASS1_BITS); - - ctr++; - - if (ctr != DCTSIZE) { - if (ctr == DCTSIZE * 2) - break; /* Done. */ - dataptr += DCTSIZE; /* advance pointer to next row */ - } else - dataptr = workspace; /* switch pointer to extended workspace */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/16)**2 = 1/2**2. - * cK represents sqrt(2) * cos(K*pi/32). - */ - - dataptr = data; - wsptr = workspace; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + wsptr[DCTSIZE*7]; - tmp1 = dataptr[DCTSIZE*1] + wsptr[DCTSIZE*6]; - tmp2 = dataptr[DCTSIZE*2] + wsptr[DCTSIZE*5]; - tmp3 = dataptr[DCTSIZE*3] + wsptr[DCTSIZE*4]; - tmp4 = dataptr[DCTSIZE*4] + wsptr[DCTSIZE*3]; - tmp5 = dataptr[DCTSIZE*5] + wsptr[DCTSIZE*2]; - tmp6 = dataptr[DCTSIZE*6] + wsptr[DCTSIZE*1]; - tmp7 = dataptr[DCTSIZE*7] + wsptr[DCTSIZE*0]; - - tmp10 = tmp0 + tmp7; - tmp14 = tmp0 - tmp7; - tmp11 = tmp1 + tmp6; - tmp15 = tmp1 - tmp6; - tmp12 = tmp2 + tmp5; - tmp16 = tmp2 - tmp5; - tmp13 = tmp3 + tmp4; - tmp17 = tmp3 - tmp4; - - tmp0 = dataptr[DCTSIZE*0] - wsptr[DCTSIZE*7]; - tmp1 = dataptr[DCTSIZE*1] - wsptr[DCTSIZE*6]; - tmp2 = dataptr[DCTSIZE*2] - wsptr[DCTSIZE*5]; - tmp3 = dataptr[DCTSIZE*3] - wsptr[DCTSIZE*4]; - tmp4 = dataptr[DCTSIZE*4] - wsptr[DCTSIZE*3]; - tmp5 = dataptr[DCTSIZE*5] - wsptr[DCTSIZE*2]; - tmp6 = dataptr[DCTSIZE*6] - wsptr[DCTSIZE*1]; - tmp7 = dataptr[DCTSIZE*7] - wsptr[DCTSIZE*0]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(tmp10 + tmp11 + tmp12 + tmp13, PASS1_BITS+2); - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp13, FIX(1.306562965)) + /* c4[16] = c2[8] */ - MULTIPLY(tmp11 - tmp12, FIX_0_541196100), /* c12[16] = c6[8] */ - CONST_BITS+PASS1_BITS+2); - - tmp10 = MULTIPLY(tmp17 - tmp15, FIX(0.275899379)) + /* c14[16] = c7[8] */ - MULTIPLY(tmp14 - tmp16, FIX(1.387039845)); /* c2[16] = c1[8] */ - - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp15, FIX(1.451774982)) /* c6+c14 */ - + MULTIPLY(tmp16, FIX(2.172734804)), /* c2+10 */ - CONST_BITS+PASS1_BITS+2); - dataptr[DCTSIZE*6] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp14, FIX(0.211164243)) /* c2-c6 */ - - MULTIPLY(tmp17, FIX(1.061594338)), /* c10+c14 */ - CONST_BITS+PASS1_BITS+2); - - /* Odd part */ - - tmp11 = MULTIPLY(tmp0 + tmp1, FIX(1.353318001)) + /* c3 */ - MULTIPLY(tmp6 - tmp7, FIX(0.410524528)); /* c13 */ - tmp12 = MULTIPLY(tmp0 + tmp2, FIX(1.247225013)) + /* c5 */ - MULTIPLY(tmp5 + tmp7, FIX(0.666655658)); /* c11 */ - tmp13 = MULTIPLY(tmp0 + tmp3, FIX(1.093201867)) + /* c7 */ - MULTIPLY(tmp4 - tmp7, FIX(0.897167586)); /* c9 */ - tmp14 = MULTIPLY(tmp1 + tmp2, FIX(0.138617169)) + /* c15 */ - MULTIPLY(tmp6 - tmp5, FIX(1.407403738)); /* c1 */ - tmp15 = MULTIPLY(tmp1 + tmp3, - FIX(0.666655658)) + /* -c11 */ - MULTIPLY(tmp4 + tmp6, - FIX(1.247225013)); /* -c5 */ - tmp16 = MULTIPLY(tmp2 + tmp3, - FIX(1.353318001)) + /* -c3 */ - MULTIPLY(tmp5 - tmp4, FIX(0.410524528)); /* c13 */ - tmp10 = tmp11 + tmp12 + tmp13 - - MULTIPLY(tmp0, FIX(2.286341144)) + /* c7+c5+c3-c1 */ - MULTIPLY(tmp7, FIX(0.779653625)); /* c15+c13-c11+c9 */ - tmp11 += tmp14 + tmp15 + MULTIPLY(tmp1, FIX(0.071888074)) /* c9-c3-c15+c11 */ - - MULTIPLY(tmp6, FIX(1.663905119)); /* c7+c13+c1-c5 */ - tmp12 += tmp14 + tmp16 - MULTIPLY(tmp2, FIX(1.125726048)) /* c7+c5+c15-c3 */ - + MULTIPLY(tmp5, FIX(1.227391138)); /* c9-c11+c1-c13 */ - tmp13 += tmp15 + tmp16 + MULTIPLY(tmp3, FIX(1.065388962)) /* c15+c3+c11-c7 */ - + MULTIPLY(tmp4, FIX(2.167985692)); /* c1+c13+c5-c9 */ - - dataptr[DCTSIZE*1] = (DCTELEM) DESCALE(tmp10, CONST_BITS+PASS1_BITS+2); - dataptr[DCTSIZE*3] = (DCTELEM) DESCALE(tmp11, CONST_BITS+PASS1_BITS+2); - dataptr[DCTSIZE*5] = (DCTELEM) DESCALE(tmp12, CONST_BITS+PASS1_BITS+2); - dataptr[DCTSIZE*7] = (DCTELEM) DESCALE(tmp13, CONST_BITS+PASS1_BITS+2); - - dataptr++; /* advance pointer to next column */ - wsptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 16x8 sample block. - * - * 16-point FDCT in pass 1 (rows), 8-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_16x8 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15, tmp16, tmp17; - INT32 z1; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * 16-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/32). - */ - - dataptr = data; - ctr = 0; - for (ctr = 0; ctr < DCTSIZE; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[15]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[14]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[13]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[12]); - tmp4 = GETJSAMPLE(elemptr[4]) + GETJSAMPLE(elemptr[11]); - tmp5 = GETJSAMPLE(elemptr[5]) + GETJSAMPLE(elemptr[10]); - tmp6 = GETJSAMPLE(elemptr[6]) + GETJSAMPLE(elemptr[9]); - tmp7 = GETJSAMPLE(elemptr[7]) + GETJSAMPLE(elemptr[8]); - - tmp10 = tmp0 + tmp7; - tmp14 = tmp0 - tmp7; - tmp11 = tmp1 + tmp6; - tmp15 = tmp1 - tmp6; - tmp12 = tmp2 + tmp5; - tmp16 = tmp2 - tmp5; - tmp13 = tmp3 + tmp4; - tmp17 = tmp3 - tmp4; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[15]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[14]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[13]); - tmp3 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[12]); - tmp4 = GETJSAMPLE(elemptr[4]) - GETJSAMPLE(elemptr[11]); - tmp5 = GETJSAMPLE(elemptr[5]) - GETJSAMPLE(elemptr[10]); - tmp6 = GETJSAMPLE(elemptr[6]) - GETJSAMPLE(elemptr[9]); - tmp7 = GETJSAMPLE(elemptr[7]) - GETJSAMPLE(elemptr[8]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp10 + tmp11 + tmp12 + tmp13 - 16 * CENTERJSAMPLE) << PASS1_BITS); - dataptr[4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp13, FIX(1.306562965)) + /* c4[16] = c2[8] */ - MULTIPLY(tmp11 - tmp12, FIX_0_541196100), /* c12[16] = c6[8] */ - CONST_BITS-PASS1_BITS); - - tmp10 = MULTIPLY(tmp17 - tmp15, FIX(0.275899379)) + /* c14[16] = c7[8] */ - MULTIPLY(tmp14 - tmp16, FIX(1.387039845)); /* c2[16] = c1[8] */ - - dataptr[2] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp15, FIX(1.451774982)) /* c6+c14 */ - + MULTIPLY(tmp16, FIX(2.172734804)), /* c2+c10 */ - CONST_BITS-PASS1_BITS); - dataptr[6] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp14, FIX(0.211164243)) /* c2-c6 */ - - MULTIPLY(tmp17, FIX(1.061594338)), /* c10+c14 */ - CONST_BITS-PASS1_BITS); - - /* Odd part */ - - tmp11 = MULTIPLY(tmp0 + tmp1, FIX(1.353318001)) + /* c3 */ - MULTIPLY(tmp6 - tmp7, FIX(0.410524528)); /* c13 */ - tmp12 = MULTIPLY(tmp0 + tmp2, FIX(1.247225013)) + /* c5 */ - MULTIPLY(tmp5 + tmp7, FIX(0.666655658)); /* c11 */ - tmp13 = MULTIPLY(tmp0 + tmp3, FIX(1.093201867)) + /* c7 */ - MULTIPLY(tmp4 - tmp7, FIX(0.897167586)); /* c9 */ - tmp14 = MULTIPLY(tmp1 + tmp2, FIX(0.138617169)) + /* c15 */ - MULTIPLY(tmp6 - tmp5, FIX(1.407403738)); /* c1 */ - tmp15 = MULTIPLY(tmp1 + tmp3, - FIX(0.666655658)) + /* -c11 */ - MULTIPLY(tmp4 + tmp6, - FIX(1.247225013)); /* -c5 */ - tmp16 = MULTIPLY(tmp2 + tmp3, - FIX(1.353318001)) + /* -c3 */ - MULTIPLY(tmp5 - tmp4, FIX(0.410524528)); /* c13 */ - tmp10 = tmp11 + tmp12 + tmp13 - - MULTIPLY(tmp0, FIX(2.286341144)) + /* c7+c5+c3-c1 */ - MULTIPLY(tmp7, FIX(0.779653625)); /* c15+c13-c11+c9 */ - tmp11 += tmp14 + tmp15 + MULTIPLY(tmp1, FIX(0.071888074)) /* c9-c3-c15+c11 */ - - MULTIPLY(tmp6, FIX(1.663905119)); /* c7+c13+c1-c5 */ - tmp12 += tmp14 + tmp16 - MULTIPLY(tmp2, FIX(1.125726048)) /* c7+c5+c15-c3 */ - + MULTIPLY(tmp5, FIX(1.227391138)); /* c9-c11+c1-c13 */ - tmp13 += tmp15 + tmp16 + MULTIPLY(tmp3, FIX(1.065388962)) /* c15+c3+c11-c7 */ - + MULTIPLY(tmp4, FIX(2.167985692)); /* c1+c13+c5-c9 */ - - dataptr[1] = (DCTELEM) DESCALE(tmp10, CONST_BITS-PASS1_BITS); - dataptr[3] = (DCTELEM) DESCALE(tmp11, CONST_BITS-PASS1_BITS); - dataptr[5] = (DCTELEM) DESCALE(tmp12, CONST_BITS-PASS1_BITS); - dataptr[7] = (DCTELEM) DESCALE(tmp13, CONST_BITS-PASS1_BITS); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by 8/16 = 1/2. - * 8-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/16). - */ - - dataptr = data; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part per LL&M figure 1 --- note that published figure is faulty; - * rotator "c1" should be "c6". - */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*7]; - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*6]; - tmp2 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*5]; - tmp3 = dataptr[DCTSIZE*3] + dataptr[DCTSIZE*4]; - - tmp10 = tmp0 + tmp3; - tmp12 = tmp0 - tmp3; - tmp11 = tmp1 + tmp2; - tmp13 = tmp1 - tmp2; - - tmp0 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*7]; - tmp1 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*6]; - tmp2 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*5]; - tmp3 = dataptr[DCTSIZE*3] - dataptr[DCTSIZE*4]; - - dataptr[DCTSIZE*0] = (DCTELEM) DESCALE(tmp10 + tmp11, PASS1_BITS+1); - dataptr[DCTSIZE*4] = (DCTELEM) DESCALE(tmp10 - tmp11, PASS1_BITS+1); - - z1 = MULTIPLY(tmp12 + tmp13, FIX_0_541196100); /* c6 */ - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(z1 + MULTIPLY(tmp12, FIX_0_765366865), /* c2-c6 */ - CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*6] = (DCTELEM) - DESCALE(z1 - MULTIPLY(tmp13, FIX_1_847759065), /* c2+c6 */ - CONST_BITS+PASS1_BITS+1); - - /* Odd part per figure 8 --- note paper omits factor of sqrt(2). - * i0..i3 in the paper are tmp0..tmp3 here. - */ - - tmp12 = tmp0 + tmp2; - tmp13 = tmp1 + tmp3; - - z1 = MULTIPLY(tmp12 + tmp13, FIX_1_175875602); /* c3 */ - tmp12 = MULTIPLY(tmp12, - FIX_0_390180644); /* -c3+c5 */ - tmp13 = MULTIPLY(tmp13, - FIX_1_961570560); /* -c3-c5 */ - tmp12 += z1; - tmp13 += z1; - - z1 = MULTIPLY(tmp0 + tmp3, - FIX_0_899976223); /* -c3+c7 */ - tmp0 = MULTIPLY(tmp0, FIX_1_501321110); /* c1+c3-c5-c7 */ - tmp3 = MULTIPLY(tmp3, FIX_0_298631336); /* -c1+c3+c5-c7 */ - tmp0 += z1 + tmp12; - tmp3 += z1 + tmp13; - - z1 = MULTIPLY(tmp1 + tmp2, - FIX_2_562915447); /* -c1-c3 */ - tmp1 = MULTIPLY(tmp1, FIX_3_072711026); /* c1+c3+c5-c7 */ - tmp2 = MULTIPLY(tmp2, FIX_2_053119869); /* c1+c3-c5+c7 */ - tmp1 += z1 + tmp13; - tmp2 += z1 + tmp12; - - dataptr[DCTSIZE*1] = (DCTELEM) DESCALE(tmp0, CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*3] = (DCTELEM) DESCALE(tmp1, CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*5] = (DCTELEM) DESCALE(tmp2, CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*7] = (DCTELEM) DESCALE(tmp3, CONST_BITS+PASS1_BITS+1); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 14x7 sample block. - * - * 14-point FDCT in pass 1 (rows), 7-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_14x7 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15, tmp16; - INT32 z1, z2, z3; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Zero bottom row of output coefficient block. */ - MEMZERO(&data[DCTSIZE*7], SIZEOF(DCTELEM) * DCTSIZE); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * 14-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/28). - */ - - dataptr = data; - for (ctr = 0; ctr < 7; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[13]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[12]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[11]); - tmp13 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[10]); - tmp4 = GETJSAMPLE(elemptr[4]) + GETJSAMPLE(elemptr[9]); - tmp5 = GETJSAMPLE(elemptr[5]) + GETJSAMPLE(elemptr[8]); - tmp6 = GETJSAMPLE(elemptr[6]) + GETJSAMPLE(elemptr[7]); - - tmp10 = tmp0 + tmp6; - tmp14 = tmp0 - tmp6; - tmp11 = tmp1 + tmp5; - tmp15 = tmp1 - tmp5; - tmp12 = tmp2 + tmp4; - tmp16 = tmp2 - tmp4; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[13]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[12]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[11]); - tmp3 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[10]); - tmp4 = GETJSAMPLE(elemptr[4]) - GETJSAMPLE(elemptr[9]); - tmp5 = GETJSAMPLE(elemptr[5]) - GETJSAMPLE(elemptr[8]); - tmp6 = GETJSAMPLE(elemptr[6]) - GETJSAMPLE(elemptr[7]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp10 + tmp11 + tmp12 + tmp13 - 14 * CENTERJSAMPLE) << PASS1_BITS); - tmp13 += tmp13; - dataptr[4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp13, FIX(1.274162392)) + /* c4 */ - MULTIPLY(tmp11 - tmp13, FIX(0.314692123)) - /* c12 */ - MULTIPLY(tmp12 - tmp13, FIX(0.881747734)), /* c8 */ - CONST_BITS-PASS1_BITS); - - tmp10 = MULTIPLY(tmp14 + tmp15, FIX(1.105676686)); /* c6 */ - - dataptr[2] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp14, FIX(0.273079590)) /* c2-c6 */ - + MULTIPLY(tmp16, FIX(0.613604268)), /* c10 */ - CONST_BITS-PASS1_BITS); - dataptr[6] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp15, FIX(1.719280954)) /* c6+c10 */ - - MULTIPLY(tmp16, FIX(1.378756276)), /* c2 */ - CONST_BITS-PASS1_BITS); - - /* Odd part */ - - tmp10 = tmp1 + tmp2; - tmp11 = tmp5 - tmp4; - dataptr[7] = (DCTELEM) ((tmp0 - tmp10 + tmp3 - tmp11 - tmp6) << PASS1_BITS); - tmp3 <<= CONST_BITS; - tmp10 = MULTIPLY(tmp10, - FIX(0.158341681)); /* -c13 */ - tmp11 = MULTIPLY(tmp11, FIX(1.405321284)); /* c1 */ - tmp10 += tmp11 - tmp3; - tmp11 = MULTIPLY(tmp0 + tmp2, FIX(1.197448846)) + /* c5 */ - MULTIPLY(tmp4 + tmp6, FIX(0.752406978)); /* c9 */ - dataptr[5] = (DCTELEM) - DESCALE(tmp10 + tmp11 - MULTIPLY(tmp2, FIX(2.373959773)) /* c3+c5-c13 */ - + MULTIPLY(tmp4, FIX(1.119999435)), /* c1+c11-c9 */ - CONST_BITS-PASS1_BITS); - tmp12 = MULTIPLY(tmp0 + tmp1, FIX(1.334852607)) + /* c3 */ - MULTIPLY(tmp5 - tmp6, FIX(0.467085129)); /* c11 */ - dataptr[3] = (DCTELEM) - DESCALE(tmp10 + tmp12 - MULTIPLY(tmp1, FIX(0.424103948)) /* c3-c9-c13 */ - - MULTIPLY(tmp5, FIX(3.069855259)), /* c1+c5+c11 */ - CONST_BITS-PASS1_BITS); - dataptr[1] = (DCTELEM) - DESCALE(tmp11 + tmp12 + tmp3 + tmp6 - - MULTIPLY(tmp0 + tmp6, FIX(1.126980169)), /* c3+c5-c1 */ - CONST_BITS-PASS1_BITS); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/14)*(8/7) = 32/49, which we - * partially fold into the constant multipliers and final shifting: - * 7-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/14) * 64/49. - */ - - dataptr = data; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*6]; - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*5]; - tmp2 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*4]; - tmp3 = dataptr[DCTSIZE*3]; - - tmp10 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*6]; - tmp11 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*5]; - tmp12 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*4]; - - z1 = tmp0 + tmp2; - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(z1 + tmp1 + tmp3, FIX(1.306122449)), /* 64/49 */ - CONST_BITS+PASS1_BITS+1); - tmp3 += tmp3; - z1 -= tmp3; - z1 -= tmp3; - z1 = MULTIPLY(z1, FIX(0.461784020)); /* (c2+c6-c4)/2 */ - z2 = MULTIPLY(tmp0 - tmp2, FIX(1.202428084)); /* (c2+c4-c6)/2 */ - z3 = MULTIPLY(tmp1 - tmp2, FIX(0.411026446)); /* c6 */ - dataptr[DCTSIZE*2] = (DCTELEM) DESCALE(z1 + z2 + z3, CONST_BITS+PASS1_BITS+1); - z1 -= z2; - z2 = MULTIPLY(tmp0 - tmp1, FIX(1.151670509)); /* c4 */ - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(z2 + z3 - MULTIPLY(tmp1 - tmp3, FIX(0.923568041)), /* c2+c6-c4 */ - CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*6] = (DCTELEM) DESCALE(z1 + z2, CONST_BITS+PASS1_BITS+1); - - /* Odd part */ - - tmp1 = MULTIPLY(tmp10 + tmp11, FIX(1.221765677)); /* (c3+c1-c5)/2 */ - tmp2 = MULTIPLY(tmp10 - tmp11, FIX(0.222383464)); /* (c3+c5-c1)/2 */ - tmp0 = tmp1 - tmp2; - tmp1 += tmp2; - tmp2 = MULTIPLY(tmp11 + tmp12, - FIX(1.800824523)); /* -c1 */ - tmp1 += tmp2; - tmp3 = MULTIPLY(tmp10 + tmp12, FIX(0.801442310)); /* c5 */ - tmp0 += tmp3; - tmp2 += tmp3 + MULTIPLY(tmp12, FIX(2.443531355)); /* c3+c1-c5 */ - - dataptr[DCTSIZE*1] = (DCTELEM) DESCALE(tmp0, CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*3] = (DCTELEM) DESCALE(tmp1, CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*5] = (DCTELEM) DESCALE(tmp2, CONST_BITS+PASS1_BITS+1); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 12x6 sample block. - * - * 12-point FDCT in pass 1 (rows), 6-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_12x6 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Zero 2 bottom rows of output coefficient block. */ - MEMZERO(&data[DCTSIZE*6], SIZEOF(DCTELEM) * DCTSIZE * 2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * 12-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/24). - */ - - dataptr = data; - for (ctr = 0; ctr < 6; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[11]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[10]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[9]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[8]); - tmp4 = GETJSAMPLE(elemptr[4]) + GETJSAMPLE(elemptr[7]); - tmp5 = GETJSAMPLE(elemptr[5]) + GETJSAMPLE(elemptr[6]); - - tmp10 = tmp0 + tmp5; - tmp13 = tmp0 - tmp5; - tmp11 = tmp1 + tmp4; - tmp14 = tmp1 - tmp4; - tmp12 = tmp2 + tmp3; - tmp15 = tmp2 - tmp3; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[11]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[10]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[9]); - tmp3 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[8]); - tmp4 = GETJSAMPLE(elemptr[4]) - GETJSAMPLE(elemptr[7]); - tmp5 = GETJSAMPLE(elemptr[5]) - GETJSAMPLE(elemptr[6]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp10 + tmp11 + tmp12 - 12 * CENTERJSAMPLE) << PASS1_BITS); - dataptr[6] = (DCTELEM) ((tmp13 - tmp14 - tmp15) << PASS1_BITS); - dataptr[4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp12, FIX(1.224744871)), /* c4 */ - CONST_BITS-PASS1_BITS); - dataptr[2] = (DCTELEM) - DESCALE(tmp14 - tmp15 + MULTIPLY(tmp13 + tmp15, FIX(1.366025404)), /* c2 */ - CONST_BITS-PASS1_BITS); - - /* Odd part */ - - tmp10 = MULTIPLY(tmp1 + tmp4, FIX_0_541196100); /* c9 */ - tmp14 = tmp10 + MULTIPLY(tmp1, FIX_0_765366865); /* c3-c9 */ - tmp15 = tmp10 - MULTIPLY(tmp4, FIX_1_847759065); /* c3+c9 */ - tmp12 = MULTIPLY(tmp0 + tmp2, FIX(1.121971054)); /* c5 */ - tmp13 = MULTIPLY(tmp0 + tmp3, FIX(0.860918669)); /* c7 */ - tmp10 = tmp12 + tmp13 + tmp14 - MULTIPLY(tmp0, FIX(0.580774953)) /* c5+c7-c1 */ - + MULTIPLY(tmp5, FIX(0.184591911)); /* c11 */ - tmp11 = MULTIPLY(tmp2 + tmp3, - FIX(0.184591911)); /* -c11 */ - tmp12 += tmp11 - tmp15 - MULTIPLY(tmp2, FIX(2.339493912)) /* c1+c5-c11 */ - + MULTIPLY(tmp5, FIX(0.860918669)); /* c7 */ - tmp13 += tmp11 - tmp14 + MULTIPLY(tmp3, FIX(0.725788011)) /* c1+c11-c7 */ - - MULTIPLY(tmp5, FIX(1.121971054)); /* c5 */ - tmp11 = tmp15 + MULTIPLY(tmp0 - tmp3, FIX(1.306562965)) /* c3 */ - - MULTIPLY(tmp2 + tmp5, FIX_0_541196100); /* c9 */ - - dataptr[1] = (DCTELEM) DESCALE(tmp10, CONST_BITS-PASS1_BITS); - dataptr[3] = (DCTELEM) DESCALE(tmp11, CONST_BITS-PASS1_BITS); - dataptr[5] = (DCTELEM) DESCALE(tmp12, CONST_BITS-PASS1_BITS); - dataptr[7] = (DCTELEM) DESCALE(tmp13, CONST_BITS-PASS1_BITS); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/12)*(8/6) = 8/9, which we - * partially fold into the constant multipliers and final shifting: - * 6-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/12) * 16/9. - */ - - dataptr = data; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*5]; - tmp11 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*4]; - tmp2 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*3]; - - tmp10 = tmp0 + tmp2; - tmp12 = tmp0 - tmp2; - - tmp0 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*5]; - tmp1 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*4]; - tmp2 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*3]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 + tmp11, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(MULTIPLY(tmp12, FIX(2.177324216)), /* c2 */ - CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp11 - tmp11, FIX(1.257078722)), /* c4 */ - CONST_BITS+PASS1_BITS+1); - - /* Odd part */ - - tmp10 = MULTIPLY(tmp0 + tmp2, FIX(0.650711829)); /* c5 */ - - dataptr[DCTSIZE*1] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp0 + tmp1, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*3] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 - tmp1 - tmp2, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*5] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp2 - tmp1, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS+1); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 10x5 sample block. - * - * 10-point FDCT in pass 1 (rows), 5-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_10x5 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Zero 3 bottom rows of output coefficient block. */ - MEMZERO(&data[DCTSIZE*5], SIZEOF(DCTELEM) * DCTSIZE * 3); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * 10-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/20). - */ - - dataptr = data; - for (ctr = 0; ctr < 5; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[9]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[8]); - tmp12 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[7]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[6]); - tmp4 = GETJSAMPLE(elemptr[4]) + GETJSAMPLE(elemptr[5]); - - tmp10 = tmp0 + tmp4; - tmp13 = tmp0 - tmp4; - tmp11 = tmp1 + tmp3; - tmp14 = tmp1 - tmp3; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[9]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[8]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[7]); - tmp3 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[6]); - tmp4 = GETJSAMPLE(elemptr[4]) - GETJSAMPLE(elemptr[5]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp10 + tmp11 + tmp12 - 10 * CENTERJSAMPLE) << PASS1_BITS); - tmp12 += tmp12; - dataptr[4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp12, FIX(1.144122806)) - /* c4 */ - MULTIPLY(tmp11 - tmp12, FIX(0.437016024)), /* c8 */ - CONST_BITS-PASS1_BITS); - tmp10 = MULTIPLY(tmp13 + tmp14, FIX(0.831253876)); /* c6 */ - dataptr[2] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp13, FIX(0.513743148)), /* c2-c6 */ - CONST_BITS-PASS1_BITS); - dataptr[6] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp14, FIX(2.176250899)), /* c2+c6 */ - CONST_BITS-PASS1_BITS); - - /* Odd part */ - - tmp10 = tmp0 + tmp4; - tmp11 = tmp1 - tmp3; - dataptr[5] = (DCTELEM) ((tmp10 - tmp11 - tmp2) << PASS1_BITS); - tmp2 <<= CONST_BITS; - dataptr[1] = (DCTELEM) - DESCALE(MULTIPLY(tmp0, FIX(1.396802247)) + /* c1 */ - MULTIPLY(tmp1, FIX(1.260073511)) + tmp2 + /* c3 */ - MULTIPLY(tmp3, FIX(0.642039522)) + /* c7 */ - MULTIPLY(tmp4, FIX(0.221231742)), /* c9 */ - CONST_BITS-PASS1_BITS); - tmp12 = MULTIPLY(tmp0 - tmp4, FIX(0.951056516)) - /* (c3+c7)/2 */ - MULTIPLY(tmp1 + tmp3, FIX(0.587785252)); /* (c1-c9)/2 */ - tmp13 = MULTIPLY(tmp10 + tmp11, FIX(0.309016994)) + /* (c3-c7)/2 */ - (tmp11 << (CONST_BITS - 1)) - tmp2; - dataptr[3] = (DCTELEM) DESCALE(tmp12 + tmp13, CONST_BITS-PASS1_BITS); - dataptr[7] = (DCTELEM) DESCALE(tmp12 - tmp13, CONST_BITS-PASS1_BITS); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/10)*(8/5) = 32/25, which we - * fold into the constant multipliers: - * 5-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/10) * 32/25. - */ - - dataptr = data; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*4]; - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*3]; - tmp2 = dataptr[DCTSIZE*2]; - - tmp10 = tmp0 + tmp1; - tmp11 = tmp0 - tmp1; - - tmp0 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*4]; - tmp1 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*3]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 + tmp2, FIX(1.28)), /* 32/25 */ - CONST_BITS+PASS1_BITS); - tmp11 = MULTIPLY(tmp11, FIX(1.011928851)); /* (c2+c4)/2 */ - tmp10 -= tmp2 << 2; - tmp10 = MULTIPLY(tmp10, FIX(0.452548340)); /* (c2-c4)/2 */ - dataptr[DCTSIZE*2] = (DCTELEM) DESCALE(tmp11 + tmp10, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*4] = (DCTELEM) DESCALE(tmp11 - tmp10, CONST_BITS+PASS1_BITS); - - /* Odd part */ - - tmp10 = MULTIPLY(tmp0 + tmp1, FIX(1.064004961)); /* c3 */ - - dataptr[DCTSIZE*1] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp0, FIX(0.657591230)), /* c1-c3 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*3] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp1, FIX(2.785601151)), /* c1+c3 */ - CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on an 8x4 sample block. - * - * 8-point FDCT in pass 1 (rows), 4-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_8x4 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3; - INT32 tmp10, tmp11, tmp12, tmp13; - INT32 z1; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Zero 4 bottom rows of output coefficient block. */ - MEMZERO(&data[DCTSIZE*4], SIZEOF(DCTELEM) * DCTSIZE * 4); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * We must also scale the output by 8/4 = 2, which we add here. - * 8-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/16). - */ - - dataptr = data; - for (ctr = 0; ctr < 4; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part per LL&M figure 1 --- note that published figure is faulty; - * rotator "c1" should be "c6". - */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[7]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[6]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[5]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[4]); - - tmp10 = tmp0 + tmp3; - tmp12 = tmp0 - tmp3; - tmp11 = tmp1 + tmp2; - tmp13 = tmp1 - tmp2; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[7]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[6]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[5]); - tmp3 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[4]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp10 + tmp11 - 8 * CENTERJSAMPLE) << (PASS1_BITS+1)); - dataptr[4] = (DCTELEM) ((tmp10 - tmp11) << (PASS1_BITS+1)); - - z1 = MULTIPLY(tmp12 + tmp13, FIX_0_541196100); /* c6 */ - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS-PASS1_BITS-2); - - dataptr[2] = (DCTELEM) - RIGHT_SHIFT(z1 + MULTIPLY(tmp12, FIX_0_765366865), /* c2-c6 */ - CONST_BITS-PASS1_BITS-1); - dataptr[6] = (DCTELEM) - RIGHT_SHIFT(z1 - MULTIPLY(tmp13, FIX_1_847759065), /* c2+c6 */ - CONST_BITS-PASS1_BITS-1); - - /* Odd part per figure 8 --- note paper omits factor of sqrt(2). - * i0..i3 in the paper are tmp0..tmp3 here. - */ - - tmp12 = tmp0 + tmp2; - tmp13 = tmp1 + tmp3; - - z1 = MULTIPLY(tmp12 + tmp13, FIX_1_175875602); /* c3 */ - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS-PASS1_BITS-2); - - tmp12 = MULTIPLY(tmp12, - FIX_0_390180644); /* -c3+c5 */ - tmp13 = MULTIPLY(tmp13, - FIX_1_961570560); /* -c3-c5 */ - tmp12 += z1; - tmp13 += z1; - - z1 = MULTIPLY(tmp0 + tmp3, - FIX_0_899976223); /* -c3+c7 */ - tmp0 = MULTIPLY(tmp0, FIX_1_501321110); /* c1+c3-c5-c7 */ - tmp3 = MULTIPLY(tmp3, FIX_0_298631336); /* -c1+c3+c5-c7 */ - tmp0 += z1 + tmp12; - tmp3 += z1 + tmp13; - - z1 = MULTIPLY(tmp1 + tmp2, - FIX_2_562915447); /* -c1-c3 */ - tmp1 = MULTIPLY(tmp1, FIX_3_072711026); /* c1+c3+c5-c7 */ - tmp2 = MULTIPLY(tmp2, FIX_2_053119869); /* c1+c3-c5+c7 */ - tmp1 += z1 + tmp13; - tmp2 += z1 + tmp12; - - dataptr[1] = (DCTELEM) RIGHT_SHIFT(tmp0, CONST_BITS-PASS1_BITS-1); - dataptr[3] = (DCTELEM) RIGHT_SHIFT(tmp1, CONST_BITS-PASS1_BITS-1); - dataptr[5] = (DCTELEM) RIGHT_SHIFT(tmp2, CONST_BITS-PASS1_BITS-1); - dataptr[7] = (DCTELEM) RIGHT_SHIFT(tmp3, CONST_BITS-PASS1_BITS-1); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * 4-point FDCT kernel, - * cK represents sqrt(2) * cos(K*pi/16) [refers to 8-point FDCT]. - */ - - dataptr = data; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - /* Add fudge factor here for final descale. */ - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*3] + (ONE << (PASS1_BITS-1)); - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*2]; - - tmp10 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*3]; - tmp11 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*2]; - - dataptr[DCTSIZE*0] = (DCTELEM) RIGHT_SHIFT(tmp0 + tmp1, PASS1_BITS); - dataptr[DCTSIZE*2] = (DCTELEM) RIGHT_SHIFT(tmp0 - tmp1, PASS1_BITS); - - /* Odd part */ - - tmp0 = MULTIPLY(tmp10 + tmp11, FIX_0_541196100); /* c6 */ - /* Add fudge factor here for final descale. */ - tmp0 += ONE << (CONST_BITS+PASS1_BITS-1); - - dataptr[DCTSIZE*1] = (DCTELEM) - RIGHT_SHIFT(tmp0 + MULTIPLY(tmp10, FIX_0_765366865), /* c2-c6 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*3] = (DCTELEM) - RIGHT_SHIFT(tmp0 - MULTIPLY(tmp11, FIX_1_847759065), /* c2+c6 */ - CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 6x3 sample block. - * - * 6-point FDCT in pass 1 (rows), 3-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_6x3 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2; - INT32 tmp10, tmp11, tmp12; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * We scale the results further by 2 as part of output adaption - * scaling for different DCT size. - * 6-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/12). - */ - - dataptr = data; - for (ctr = 0; ctr < 3; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[5]); - tmp11 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[4]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[3]); - - tmp10 = tmp0 + tmp2; - tmp12 = tmp0 - tmp2; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[5]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[4]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[3]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp10 + tmp11 - 6 * CENTERJSAMPLE) << (PASS1_BITS+1)); - dataptr[2] = (DCTELEM) - DESCALE(MULTIPLY(tmp12, FIX(1.224744871)), /* c2 */ - CONST_BITS-PASS1_BITS-1); - dataptr[4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp11 - tmp11, FIX(0.707106781)), /* c4 */ - CONST_BITS-PASS1_BITS-1); - - /* Odd part */ - - tmp10 = DESCALE(MULTIPLY(tmp0 + tmp2, FIX(0.366025404)), /* c5 */ - CONST_BITS-PASS1_BITS-1); - - dataptr[1] = (DCTELEM) (tmp10 + ((tmp0 + tmp1) << (PASS1_BITS+1))); - dataptr[3] = (DCTELEM) ((tmp0 - tmp1 - tmp2) << (PASS1_BITS+1)); - dataptr[5] = (DCTELEM) (tmp10 + ((tmp2 - tmp1) << (PASS1_BITS+1))); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/6)*(8/3) = 32/9, which we partially - * fold into the constant multipliers (other part was done in pass 1): - * 3-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/6) * 16/9. - */ - - dataptr = data; - for (ctr = 0; ctr < 6; ctr++) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*2]; - tmp1 = dataptr[DCTSIZE*1]; - - tmp2 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*2]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 + tmp1, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 - tmp1 - tmp1, FIX(1.257078722)), /* c2 */ - CONST_BITS+PASS1_BITS); - - /* Odd part */ - - dataptr[DCTSIZE*1] = (DCTELEM) - DESCALE(MULTIPLY(tmp2, FIX(2.177324216)), /* c1 */ - CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 4x2 sample block. - * - * 4-point FDCT in pass 1 (rows), 2-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_4x2 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - DCTELEM tmp0, tmp2, tmp10, tmp12, tmp4, tmp5; - INT32 tmp1, tmp3, tmp11, tmp13; - INT32 z1, z2, z3; - JSAMPROW elemptr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT. - * 4-point FDCT kernel, - * cK represents sqrt(2) * cos(K*pi/16) [refers to 8-point FDCT]. - */ - - /* Row 0 */ - elemptr = sample_data[0] + start_col; - - /* Even part */ - - tmp4 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[3]); - tmp5 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[2]); - - tmp0 = tmp4 + tmp5; - tmp2 = tmp4 - tmp5; - - /* Odd part */ - - z2 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[3]); - z3 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[2]); - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS-3-1); - tmp1 = z1 + MULTIPLY(z2, FIX_0_765366865); /* c2-c6 */ - tmp3 = z1 - MULTIPLY(z3, FIX_1_847759065); /* c2+c6 */ - - /* Row 1 */ - elemptr = sample_data[1] + start_col; - - /* Even part */ - - tmp4 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[3]); - tmp5 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[2]); - - tmp10 = tmp4 + tmp5; - tmp12 = tmp4 - tmp5; - - /* Odd part */ - - z2 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[3]); - z3 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[2]); - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - tmp11 = z1 + MULTIPLY(z2, FIX_0_765366865); /* c2-c6 */ - tmp13 = z1 - MULTIPLY(z3, FIX_1_847759065); /* c2+c6 */ - - /* Pass 2: process columns. - * We leave the results scaled up by an overall factor of 8. - * We must also scale the output by (8/4)*(8/2) = 2**3. - */ - - /* Column 0 */ - /* Apply unsigned->signed conversion. */ - data[DCTSIZE*0] = (tmp0 + tmp10 - 8 * CENTERJSAMPLE) << 3; - data[DCTSIZE*1] = (tmp0 - tmp10) << 3; - - /* Column 1 */ - data[DCTSIZE*0+1] = (DCTELEM) RIGHT_SHIFT(tmp1 + tmp11, CONST_BITS-3); - data[DCTSIZE*1+1] = (DCTELEM) RIGHT_SHIFT(tmp1 - tmp11, CONST_BITS-3); - - /* Column 2 */ - data[DCTSIZE*0+2] = (tmp2 + tmp12) << 3; - data[DCTSIZE*1+2] = (tmp2 - tmp12) << 3; - - /* Column 3 */ - data[DCTSIZE*0+3] = (DCTELEM) RIGHT_SHIFT(tmp3 + tmp13, CONST_BITS-3); - data[DCTSIZE*1+3] = (DCTELEM) RIGHT_SHIFT(tmp3 - tmp13, CONST_BITS-3); -} - - -/* - * Perform the forward DCT on a 2x1 sample block. - * - * 2-point FDCT in pass 1 (rows), 1-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_2x1 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - DCTELEM tmp0, tmp1; - JSAMPROW elemptr; - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - elemptr = sample_data[0] + start_col; - - tmp0 = GETJSAMPLE(elemptr[0]); - tmp1 = GETJSAMPLE(elemptr[1]); - - /* We leave the results scaled up by an overall factor of 8. - * We must also scale the output by (8/2)*(8/1) = 2**5. - */ - - /* Even part */ - - /* Apply unsigned->signed conversion. */ - data[0] = (tmp0 + tmp1 - 2 * CENTERJSAMPLE) << 5; - - /* Odd part */ - - data[1] = (tmp0 - tmp1) << 5; -} - - -/* - * Perform the forward DCT on an 8x16 sample block. - * - * 8-point FDCT in pass 1 (rows), 16-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_8x16 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15, tmp16, tmp17; - INT32 z1; - DCTELEM workspace[DCTSIZE2]; - DCTELEM *dataptr; - DCTELEM *wsptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * 8-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/16). - */ - - dataptr = data; - ctr = 0; - for (;;) { - elemptr = sample_data[ctr] + start_col; - - /* Even part per LL&M figure 1 --- note that published figure is faulty; - * rotator "c1" should be "c6". - */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[7]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[6]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[5]); - tmp3 = GETJSAMPLE(elemptr[3]) + GETJSAMPLE(elemptr[4]); - - tmp10 = tmp0 + tmp3; - tmp12 = tmp0 - tmp3; - tmp11 = tmp1 + tmp2; - tmp13 = tmp1 - tmp2; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[7]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[6]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[5]); - tmp3 = GETJSAMPLE(elemptr[3]) - GETJSAMPLE(elemptr[4]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) ((tmp10 + tmp11 - 8 * CENTERJSAMPLE) << PASS1_BITS); - dataptr[4] = (DCTELEM) ((tmp10 - tmp11) << PASS1_BITS); - - z1 = MULTIPLY(tmp12 + tmp13, FIX_0_541196100); /* c6 */ - dataptr[2] = (DCTELEM) - DESCALE(z1 + MULTIPLY(tmp12, FIX_0_765366865), /* c2-c6 */ - CONST_BITS-PASS1_BITS); - dataptr[6] = (DCTELEM) - DESCALE(z1 - MULTIPLY(tmp13, FIX_1_847759065), /* c2+c6 */ - CONST_BITS-PASS1_BITS); - - /* Odd part per figure 8 --- note paper omits factor of sqrt(2). - * i0..i3 in the paper are tmp0..tmp3 here. - */ - - tmp12 = tmp0 + tmp2; - tmp13 = tmp1 + tmp3; - - z1 = MULTIPLY(tmp12 + tmp13, FIX_1_175875602); /* c3 */ - tmp12 = MULTIPLY(tmp12, - FIX_0_390180644); /* -c3+c5 */ - tmp13 = MULTIPLY(tmp13, - FIX_1_961570560); /* -c3-c5 */ - tmp12 += z1; - tmp13 += z1; - - z1 = MULTIPLY(tmp0 + tmp3, - FIX_0_899976223); /* -c3+c7 */ - tmp0 = MULTIPLY(tmp0, FIX_1_501321110); /* c1+c3-c5-c7 */ - tmp3 = MULTIPLY(tmp3, FIX_0_298631336); /* -c1+c3+c5-c7 */ - tmp0 += z1 + tmp12; - tmp3 += z1 + tmp13; - - z1 = MULTIPLY(tmp1 + tmp2, - FIX_2_562915447); /* -c1-c3 */ - tmp1 = MULTIPLY(tmp1, FIX_3_072711026); /* c1+c3+c5-c7 */ - tmp2 = MULTIPLY(tmp2, FIX_2_053119869); /* c1+c3-c5+c7 */ - tmp1 += z1 + tmp13; - tmp2 += z1 + tmp12; - - dataptr[1] = (DCTELEM) DESCALE(tmp0, CONST_BITS-PASS1_BITS); - dataptr[3] = (DCTELEM) DESCALE(tmp1, CONST_BITS-PASS1_BITS); - dataptr[5] = (DCTELEM) DESCALE(tmp2, CONST_BITS-PASS1_BITS); - dataptr[7] = (DCTELEM) DESCALE(tmp3, CONST_BITS-PASS1_BITS); - - ctr++; - - if (ctr != DCTSIZE) { - if (ctr == DCTSIZE * 2) - break; /* Done. */ - dataptr += DCTSIZE; /* advance pointer to next row */ - } else - dataptr = workspace; /* switch pointer to extended workspace */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by 8/16 = 1/2. - * 16-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/32). - */ - - dataptr = data; - wsptr = workspace; - for (ctr = DCTSIZE-1; ctr >= 0; ctr--) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + wsptr[DCTSIZE*7]; - tmp1 = dataptr[DCTSIZE*1] + wsptr[DCTSIZE*6]; - tmp2 = dataptr[DCTSIZE*2] + wsptr[DCTSIZE*5]; - tmp3 = dataptr[DCTSIZE*3] + wsptr[DCTSIZE*4]; - tmp4 = dataptr[DCTSIZE*4] + wsptr[DCTSIZE*3]; - tmp5 = dataptr[DCTSIZE*5] + wsptr[DCTSIZE*2]; - tmp6 = dataptr[DCTSIZE*6] + wsptr[DCTSIZE*1]; - tmp7 = dataptr[DCTSIZE*7] + wsptr[DCTSIZE*0]; - - tmp10 = tmp0 + tmp7; - tmp14 = tmp0 - tmp7; - tmp11 = tmp1 + tmp6; - tmp15 = tmp1 - tmp6; - tmp12 = tmp2 + tmp5; - tmp16 = tmp2 - tmp5; - tmp13 = tmp3 + tmp4; - tmp17 = tmp3 - tmp4; - - tmp0 = dataptr[DCTSIZE*0] - wsptr[DCTSIZE*7]; - tmp1 = dataptr[DCTSIZE*1] - wsptr[DCTSIZE*6]; - tmp2 = dataptr[DCTSIZE*2] - wsptr[DCTSIZE*5]; - tmp3 = dataptr[DCTSIZE*3] - wsptr[DCTSIZE*4]; - tmp4 = dataptr[DCTSIZE*4] - wsptr[DCTSIZE*3]; - tmp5 = dataptr[DCTSIZE*5] - wsptr[DCTSIZE*2]; - tmp6 = dataptr[DCTSIZE*6] - wsptr[DCTSIZE*1]; - tmp7 = dataptr[DCTSIZE*7] - wsptr[DCTSIZE*0]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(tmp10 + tmp11 + tmp12 + tmp13, PASS1_BITS+1); - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp13, FIX(1.306562965)) + /* c4[16] = c2[8] */ - MULTIPLY(tmp11 - tmp12, FIX_0_541196100), /* c12[16] = c6[8] */ - CONST_BITS+PASS1_BITS+1); - - tmp10 = MULTIPLY(tmp17 - tmp15, FIX(0.275899379)) + /* c14[16] = c7[8] */ - MULTIPLY(tmp14 - tmp16, FIX(1.387039845)); /* c2[16] = c1[8] */ - - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp15, FIX(1.451774982)) /* c6+c14 */ - + MULTIPLY(tmp16, FIX(2.172734804)), /* c2+c10 */ - CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*6] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp14, FIX(0.211164243)) /* c2-c6 */ - - MULTIPLY(tmp17, FIX(1.061594338)), /* c10+c14 */ - CONST_BITS+PASS1_BITS+1); - - /* Odd part */ - - tmp11 = MULTIPLY(tmp0 + tmp1, FIX(1.353318001)) + /* c3 */ - MULTIPLY(tmp6 - tmp7, FIX(0.410524528)); /* c13 */ - tmp12 = MULTIPLY(tmp0 + tmp2, FIX(1.247225013)) + /* c5 */ - MULTIPLY(tmp5 + tmp7, FIX(0.666655658)); /* c11 */ - tmp13 = MULTIPLY(tmp0 + tmp3, FIX(1.093201867)) + /* c7 */ - MULTIPLY(tmp4 - tmp7, FIX(0.897167586)); /* c9 */ - tmp14 = MULTIPLY(tmp1 + tmp2, FIX(0.138617169)) + /* c15 */ - MULTIPLY(tmp6 - tmp5, FIX(1.407403738)); /* c1 */ - tmp15 = MULTIPLY(tmp1 + tmp3, - FIX(0.666655658)) + /* -c11 */ - MULTIPLY(tmp4 + tmp6, - FIX(1.247225013)); /* -c5 */ - tmp16 = MULTIPLY(tmp2 + tmp3, - FIX(1.353318001)) + /* -c3 */ - MULTIPLY(tmp5 - tmp4, FIX(0.410524528)); /* c13 */ - tmp10 = tmp11 + tmp12 + tmp13 - - MULTIPLY(tmp0, FIX(2.286341144)) + /* c7+c5+c3-c1 */ - MULTIPLY(tmp7, FIX(0.779653625)); /* c15+c13-c11+c9 */ - tmp11 += tmp14 + tmp15 + MULTIPLY(tmp1, FIX(0.071888074)) /* c9-c3-c15+c11 */ - - MULTIPLY(tmp6, FIX(1.663905119)); /* c7+c13+c1-c5 */ - tmp12 += tmp14 + tmp16 - MULTIPLY(tmp2, FIX(1.125726048)) /* c7+c5+c15-c3 */ - + MULTIPLY(tmp5, FIX(1.227391138)); /* c9-c11+c1-c13 */ - tmp13 += tmp15 + tmp16 + MULTIPLY(tmp3, FIX(1.065388962)) /* c15+c3+c11-c7 */ - + MULTIPLY(tmp4, FIX(2.167985692)); /* c1+c13+c5-c9 */ - - dataptr[DCTSIZE*1] = (DCTELEM) DESCALE(tmp10, CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*3] = (DCTELEM) DESCALE(tmp11, CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*5] = (DCTELEM) DESCALE(tmp12, CONST_BITS+PASS1_BITS+1); - dataptr[DCTSIZE*7] = (DCTELEM) DESCALE(tmp13, CONST_BITS+PASS1_BITS+1); - - dataptr++; /* advance pointer to next column */ - wsptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 7x14 sample block. - * - * 7-point FDCT in pass 1 (rows), 14-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_7x14 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15, tmp16; - INT32 z1, z2, z3; - DCTELEM workspace[8*6]; - DCTELEM *dataptr; - DCTELEM *wsptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * 7-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/14). - */ - - dataptr = data; - ctr = 0; - for (;;) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[6]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[5]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[4]); - tmp3 = GETJSAMPLE(elemptr[3]); - - tmp10 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[6]); - tmp11 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[5]); - tmp12 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[4]); - - z1 = tmp0 + tmp2; - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((z1 + tmp1 + tmp3 - 7 * CENTERJSAMPLE) << PASS1_BITS); - tmp3 += tmp3; - z1 -= tmp3; - z1 -= tmp3; - z1 = MULTIPLY(z1, FIX(0.353553391)); /* (c2+c6-c4)/2 */ - z2 = MULTIPLY(tmp0 - tmp2, FIX(0.920609002)); /* (c2+c4-c6)/2 */ - z3 = MULTIPLY(tmp1 - tmp2, FIX(0.314692123)); /* c6 */ - dataptr[2] = (DCTELEM) DESCALE(z1 + z2 + z3, CONST_BITS-PASS1_BITS); - z1 -= z2; - z2 = MULTIPLY(tmp0 - tmp1, FIX(0.881747734)); /* c4 */ - dataptr[4] = (DCTELEM) - DESCALE(z2 + z3 - MULTIPLY(tmp1 - tmp3, FIX(0.707106781)), /* c2+c6-c4 */ - CONST_BITS-PASS1_BITS); - dataptr[6] = (DCTELEM) DESCALE(z1 + z2, CONST_BITS-PASS1_BITS); - - /* Odd part */ - - tmp1 = MULTIPLY(tmp10 + tmp11, FIX(0.935414347)); /* (c3+c1-c5)/2 */ - tmp2 = MULTIPLY(tmp10 - tmp11, FIX(0.170262339)); /* (c3+c5-c1)/2 */ - tmp0 = tmp1 - tmp2; - tmp1 += tmp2; - tmp2 = MULTIPLY(tmp11 + tmp12, - FIX(1.378756276)); /* -c1 */ - tmp1 += tmp2; - tmp3 = MULTIPLY(tmp10 + tmp12, FIX(0.613604268)); /* c5 */ - tmp0 += tmp3; - tmp2 += tmp3 + MULTIPLY(tmp12, FIX(1.870828693)); /* c3+c1-c5 */ - - dataptr[1] = (DCTELEM) DESCALE(tmp0, CONST_BITS-PASS1_BITS); - dataptr[3] = (DCTELEM) DESCALE(tmp1, CONST_BITS-PASS1_BITS); - dataptr[5] = (DCTELEM) DESCALE(tmp2, CONST_BITS-PASS1_BITS); - - ctr++; - - if (ctr != DCTSIZE) { - if (ctr == 14) - break; /* Done. */ - dataptr += DCTSIZE; /* advance pointer to next row */ - } else - dataptr = workspace; /* switch pointer to extended workspace */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/7)*(8/14) = 32/49, which we - * fold into the constant multipliers: - * 14-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/28) * 32/49. - */ - - dataptr = data; - wsptr = workspace; - for (ctr = 0; ctr < 7; ctr++) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + wsptr[DCTSIZE*5]; - tmp1 = dataptr[DCTSIZE*1] + wsptr[DCTSIZE*4]; - tmp2 = dataptr[DCTSIZE*2] + wsptr[DCTSIZE*3]; - tmp13 = dataptr[DCTSIZE*3] + wsptr[DCTSIZE*2]; - tmp4 = dataptr[DCTSIZE*4] + wsptr[DCTSIZE*1]; - tmp5 = dataptr[DCTSIZE*5] + wsptr[DCTSIZE*0]; - tmp6 = dataptr[DCTSIZE*6] + dataptr[DCTSIZE*7]; - - tmp10 = tmp0 + tmp6; - tmp14 = tmp0 - tmp6; - tmp11 = tmp1 + tmp5; - tmp15 = tmp1 - tmp5; - tmp12 = tmp2 + tmp4; - tmp16 = tmp2 - tmp4; - - tmp0 = dataptr[DCTSIZE*0] - wsptr[DCTSIZE*5]; - tmp1 = dataptr[DCTSIZE*1] - wsptr[DCTSIZE*4]; - tmp2 = dataptr[DCTSIZE*2] - wsptr[DCTSIZE*3]; - tmp3 = dataptr[DCTSIZE*3] - wsptr[DCTSIZE*2]; - tmp4 = dataptr[DCTSIZE*4] - wsptr[DCTSIZE*1]; - tmp5 = dataptr[DCTSIZE*5] - wsptr[DCTSIZE*0]; - tmp6 = dataptr[DCTSIZE*6] - dataptr[DCTSIZE*7]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 + tmp11 + tmp12 + tmp13, - FIX(0.653061224)), /* 32/49 */ - CONST_BITS+PASS1_BITS); - tmp13 += tmp13; - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp13, FIX(0.832106052)) + /* c4 */ - MULTIPLY(tmp11 - tmp13, FIX(0.205513223)) - /* c12 */ - MULTIPLY(tmp12 - tmp13, FIX(0.575835255)), /* c8 */ - CONST_BITS+PASS1_BITS); - - tmp10 = MULTIPLY(tmp14 + tmp15, FIX(0.722074570)); /* c6 */ - - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp14, FIX(0.178337691)) /* c2-c6 */ - + MULTIPLY(tmp16, FIX(0.400721155)), /* c10 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*6] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp15, FIX(1.122795725)) /* c6+c10 */ - - MULTIPLY(tmp16, FIX(0.900412262)), /* c2 */ - CONST_BITS+PASS1_BITS); - - /* Odd part */ - - tmp10 = tmp1 + tmp2; - tmp11 = tmp5 - tmp4; - dataptr[DCTSIZE*7] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 - tmp10 + tmp3 - tmp11 - tmp6, - FIX(0.653061224)), /* 32/49 */ - CONST_BITS+PASS1_BITS); - tmp3 = MULTIPLY(tmp3 , FIX(0.653061224)); /* 32/49 */ - tmp10 = MULTIPLY(tmp10, - FIX(0.103406812)); /* -c13 */ - tmp11 = MULTIPLY(tmp11, FIX(0.917760839)); /* c1 */ - tmp10 += tmp11 - tmp3; - tmp11 = MULTIPLY(tmp0 + tmp2, FIX(0.782007410)) + /* c5 */ - MULTIPLY(tmp4 + tmp6, FIX(0.491367823)); /* c9 */ - dataptr[DCTSIZE*5] = (DCTELEM) - DESCALE(tmp10 + tmp11 - MULTIPLY(tmp2, FIX(1.550341076)) /* c3+c5-c13 */ - + MULTIPLY(tmp4, FIX(0.731428202)), /* c1+c11-c9 */ - CONST_BITS+PASS1_BITS); - tmp12 = MULTIPLY(tmp0 + tmp1, FIX(0.871740478)) + /* c3 */ - MULTIPLY(tmp5 - tmp6, FIX(0.305035186)); /* c11 */ - dataptr[DCTSIZE*3] = (DCTELEM) - DESCALE(tmp10 + tmp12 - MULTIPLY(tmp1, FIX(0.276965844)) /* c3-c9-c13 */ - - MULTIPLY(tmp5, FIX(2.004803435)), /* c1+c5+c11 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*1] = (DCTELEM) - DESCALE(tmp11 + tmp12 + tmp3 - - MULTIPLY(tmp0, FIX(0.735987049)) /* c3+c5-c1 */ - - MULTIPLY(tmp6, FIX(0.082925825)), /* c9-c11-c13 */ - CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - wsptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 6x12 sample block. - * - * 6-point FDCT in pass 1 (rows), 12-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_6x12 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4, tmp5; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15; - DCTELEM workspace[8*4]; - DCTELEM *dataptr; - DCTELEM *wsptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * 6-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/12). - */ - - dataptr = data; - ctr = 0; - for (;;) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[5]); - tmp11 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[4]); - tmp2 = GETJSAMPLE(elemptr[2]) + GETJSAMPLE(elemptr[3]); - - tmp10 = tmp0 + tmp2; - tmp12 = tmp0 - tmp2; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[5]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[4]); - tmp2 = GETJSAMPLE(elemptr[2]) - GETJSAMPLE(elemptr[3]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp10 + tmp11 - 6 * CENTERJSAMPLE) << PASS1_BITS); - dataptr[2] = (DCTELEM) - DESCALE(MULTIPLY(tmp12, FIX(1.224744871)), /* c2 */ - CONST_BITS-PASS1_BITS); - dataptr[4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp11 - tmp11, FIX(0.707106781)), /* c4 */ - CONST_BITS-PASS1_BITS); - - /* Odd part */ - - tmp10 = DESCALE(MULTIPLY(tmp0 + tmp2, FIX(0.366025404)), /* c5 */ - CONST_BITS-PASS1_BITS); - - dataptr[1] = (DCTELEM) (tmp10 + ((tmp0 + tmp1) << PASS1_BITS)); - dataptr[3] = (DCTELEM) ((tmp0 - tmp1 - tmp2) << PASS1_BITS); - dataptr[5] = (DCTELEM) (tmp10 + ((tmp2 - tmp1) << PASS1_BITS)); - - ctr++; - - if (ctr != DCTSIZE) { - if (ctr == 12) - break; /* Done. */ - dataptr += DCTSIZE; /* advance pointer to next row */ - } else - dataptr = workspace; /* switch pointer to extended workspace */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/6)*(8/12) = 8/9, which we - * fold into the constant multipliers: - * 12-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/24) * 8/9. - */ - - dataptr = data; - wsptr = workspace; - for (ctr = 0; ctr < 6; ctr++) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + wsptr[DCTSIZE*3]; - tmp1 = dataptr[DCTSIZE*1] + wsptr[DCTSIZE*2]; - tmp2 = dataptr[DCTSIZE*2] + wsptr[DCTSIZE*1]; - tmp3 = dataptr[DCTSIZE*3] + wsptr[DCTSIZE*0]; - tmp4 = dataptr[DCTSIZE*4] + dataptr[DCTSIZE*7]; - tmp5 = dataptr[DCTSIZE*5] + dataptr[DCTSIZE*6]; - - tmp10 = tmp0 + tmp5; - tmp13 = tmp0 - tmp5; - tmp11 = tmp1 + tmp4; - tmp14 = tmp1 - tmp4; - tmp12 = tmp2 + tmp3; - tmp15 = tmp2 - tmp3; - - tmp0 = dataptr[DCTSIZE*0] - wsptr[DCTSIZE*3]; - tmp1 = dataptr[DCTSIZE*1] - wsptr[DCTSIZE*2]; - tmp2 = dataptr[DCTSIZE*2] - wsptr[DCTSIZE*1]; - tmp3 = dataptr[DCTSIZE*3] - wsptr[DCTSIZE*0]; - tmp4 = dataptr[DCTSIZE*4] - dataptr[DCTSIZE*7]; - tmp5 = dataptr[DCTSIZE*5] - dataptr[DCTSIZE*6]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 + tmp11 + tmp12, FIX(0.888888889)), /* 8/9 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*6] = (DCTELEM) - DESCALE(MULTIPLY(tmp13 - tmp14 - tmp15, FIX(0.888888889)), /* 8/9 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp12, FIX(1.088662108)), /* c4 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(MULTIPLY(tmp14 - tmp15, FIX(0.888888889)) + /* 8/9 */ - MULTIPLY(tmp13 + tmp15, FIX(1.214244803)), /* c2 */ - CONST_BITS+PASS1_BITS); - - /* Odd part */ - - tmp10 = MULTIPLY(tmp1 + tmp4, FIX(0.481063200)); /* c9 */ - tmp14 = tmp10 + MULTIPLY(tmp1, FIX(0.680326102)); /* c3-c9 */ - tmp15 = tmp10 - MULTIPLY(tmp4, FIX(1.642452502)); /* c3+c9 */ - tmp12 = MULTIPLY(tmp0 + tmp2, FIX(0.997307603)); /* c5 */ - tmp13 = MULTIPLY(tmp0 + tmp3, FIX(0.765261039)); /* c7 */ - tmp10 = tmp12 + tmp13 + tmp14 - MULTIPLY(tmp0, FIX(0.516244403)) /* c5+c7-c1 */ - + MULTIPLY(tmp5, FIX(0.164081699)); /* c11 */ - tmp11 = MULTIPLY(tmp2 + tmp3, - FIX(0.164081699)); /* -c11 */ - tmp12 += tmp11 - tmp15 - MULTIPLY(tmp2, FIX(2.079550144)) /* c1+c5-c11 */ - + MULTIPLY(tmp5, FIX(0.765261039)); /* c7 */ - tmp13 += tmp11 - tmp14 + MULTIPLY(tmp3, FIX(0.645144899)) /* c1+c11-c7 */ - - MULTIPLY(tmp5, FIX(0.997307603)); /* c5 */ - tmp11 = tmp15 + MULTIPLY(tmp0 - tmp3, FIX(1.161389302)) /* c3 */ - - MULTIPLY(tmp2 + tmp5, FIX(0.481063200)); /* c9 */ - - dataptr[DCTSIZE*1] = (DCTELEM) DESCALE(tmp10, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*3] = (DCTELEM) DESCALE(tmp11, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*5] = (DCTELEM) DESCALE(tmp12, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*7] = (DCTELEM) DESCALE(tmp13, CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - wsptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 5x10 sample block. - * - * 5-point FDCT in pass 1 (rows), 10-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_5x10 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp4; - INT32 tmp10, tmp11, tmp12, tmp13, tmp14; - DCTELEM workspace[8*2]; - DCTELEM *dataptr; - DCTELEM *wsptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * 5-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/10). - */ - - dataptr = data; - ctr = 0; - for (;;) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[4]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[3]); - tmp2 = GETJSAMPLE(elemptr[2]); - - tmp10 = tmp0 + tmp1; - tmp11 = tmp0 - tmp1; - - tmp0 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[4]); - tmp1 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[3]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp10 + tmp2 - 5 * CENTERJSAMPLE) << PASS1_BITS); - tmp11 = MULTIPLY(tmp11, FIX(0.790569415)); /* (c2+c4)/2 */ - tmp10 -= tmp2 << 2; - tmp10 = MULTIPLY(tmp10, FIX(0.353553391)); /* (c2-c4)/2 */ - dataptr[2] = (DCTELEM) DESCALE(tmp11 + tmp10, CONST_BITS-PASS1_BITS); - dataptr[4] = (DCTELEM) DESCALE(tmp11 - tmp10, CONST_BITS-PASS1_BITS); - - /* Odd part */ - - tmp10 = MULTIPLY(tmp0 + tmp1, FIX(0.831253876)); /* c3 */ - - dataptr[1] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp0, FIX(0.513743148)), /* c1-c3 */ - CONST_BITS-PASS1_BITS); - dataptr[3] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp1, FIX(2.176250899)), /* c1+c3 */ - CONST_BITS-PASS1_BITS); - - ctr++; - - if (ctr != DCTSIZE) { - if (ctr == 10) - break; /* Done. */ - dataptr += DCTSIZE; /* advance pointer to next row */ - } else - dataptr = workspace; /* switch pointer to extended workspace */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/5)*(8/10) = 32/25, which we - * fold into the constant multipliers: - * 10-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/20) * 32/25. - */ - - dataptr = data; - wsptr = workspace; - for (ctr = 0; ctr < 5; ctr++) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + wsptr[DCTSIZE*1]; - tmp1 = dataptr[DCTSIZE*1] + wsptr[DCTSIZE*0]; - tmp12 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*7]; - tmp3 = dataptr[DCTSIZE*3] + dataptr[DCTSIZE*6]; - tmp4 = dataptr[DCTSIZE*4] + dataptr[DCTSIZE*5]; - - tmp10 = tmp0 + tmp4; - tmp13 = tmp0 - tmp4; - tmp11 = tmp1 + tmp3; - tmp14 = tmp1 - tmp3; - - tmp0 = dataptr[DCTSIZE*0] - wsptr[DCTSIZE*1]; - tmp1 = dataptr[DCTSIZE*1] - wsptr[DCTSIZE*0]; - tmp2 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*7]; - tmp3 = dataptr[DCTSIZE*3] - dataptr[DCTSIZE*6]; - tmp4 = dataptr[DCTSIZE*4] - dataptr[DCTSIZE*5]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 + tmp11 + tmp12, FIX(1.28)), /* 32/25 */ - CONST_BITS+PASS1_BITS); - tmp12 += tmp12; - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp12, FIX(1.464477191)) - /* c4 */ - MULTIPLY(tmp11 - tmp12, FIX(0.559380511)), /* c8 */ - CONST_BITS+PASS1_BITS); - tmp10 = MULTIPLY(tmp13 + tmp14, FIX(1.064004961)); /* c6 */ - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp13, FIX(0.657591230)), /* c2-c6 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*6] = (DCTELEM) - DESCALE(tmp10 - MULTIPLY(tmp14, FIX(2.785601151)), /* c2+c6 */ - CONST_BITS+PASS1_BITS); - - /* Odd part */ - - tmp10 = tmp0 + tmp4; - tmp11 = tmp1 - tmp3; - dataptr[DCTSIZE*5] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp11 - tmp2, FIX(1.28)), /* 32/25 */ - CONST_BITS+PASS1_BITS); - tmp2 = MULTIPLY(tmp2, FIX(1.28)); /* 32/25 */ - dataptr[DCTSIZE*1] = (DCTELEM) - DESCALE(MULTIPLY(tmp0, FIX(1.787906876)) + /* c1 */ - MULTIPLY(tmp1, FIX(1.612894094)) + tmp2 + /* c3 */ - MULTIPLY(tmp3, FIX(0.821810588)) + /* c7 */ - MULTIPLY(tmp4, FIX(0.283176630)), /* c9 */ - CONST_BITS+PASS1_BITS); - tmp12 = MULTIPLY(tmp0 - tmp4, FIX(1.217352341)) - /* (c3+c7)/2 */ - MULTIPLY(tmp1 + tmp3, FIX(0.752365123)); /* (c1-c9)/2 */ - tmp13 = MULTIPLY(tmp10 + tmp11, FIX(0.395541753)) + /* (c3-c7)/2 */ - MULTIPLY(tmp11, FIX(0.64)) - tmp2; /* 16/25 */ - dataptr[DCTSIZE*3] = (DCTELEM) DESCALE(tmp12 + tmp13, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*7] = (DCTELEM) DESCALE(tmp12 - tmp13, CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - wsptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 4x8 sample block. - * - * 4-point FDCT in pass 1 (rows), 8-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_4x8 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3; - INT32 tmp10, tmp11, tmp12, tmp13; - INT32 z1; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * We must also scale the output by 8/4 = 2, which we add here. - * 4-point FDCT kernel, - * cK represents sqrt(2) * cos(K*pi/16) [refers to 8-point FDCT]. - */ - - dataptr = data; - for (ctr = 0; ctr < DCTSIZE; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[3]); - tmp1 = GETJSAMPLE(elemptr[1]) + GETJSAMPLE(elemptr[2]); - - tmp10 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[3]); - tmp11 = GETJSAMPLE(elemptr[1]) - GETJSAMPLE(elemptr[2]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp0 + tmp1 - 4 * CENTERJSAMPLE) << (PASS1_BITS+1)); - dataptr[2] = (DCTELEM) ((tmp0 - tmp1) << (PASS1_BITS+1)); - - /* Odd part */ - - tmp0 = MULTIPLY(tmp10 + tmp11, FIX_0_541196100); /* c6 */ - /* Add fudge factor here for final descale. */ - tmp0 += ONE << (CONST_BITS-PASS1_BITS-2); - - dataptr[1] = (DCTELEM) - RIGHT_SHIFT(tmp0 + MULTIPLY(tmp10, FIX_0_765366865), /* c2-c6 */ - CONST_BITS-PASS1_BITS-1); - dataptr[3] = (DCTELEM) - RIGHT_SHIFT(tmp0 - MULTIPLY(tmp11, FIX_1_847759065), /* c2+c6 */ - CONST_BITS-PASS1_BITS-1); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * 8-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/16). - */ - - dataptr = data; - for (ctr = 0; ctr < 4; ctr++) { - /* Even part per LL&M figure 1 --- note that published figure is faulty; - * rotator "c1" should be "c6". - */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*7]; - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*6]; - tmp2 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*5]; - tmp3 = dataptr[DCTSIZE*3] + dataptr[DCTSIZE*4]; - - /* Add fudge factor here for final descale. */ - tmp10 = tmp0 + tmp3 + (ONE << (PASS1_BITS-1)); - tmp12 = tmp0 - tmp3; - tmp11 = tmp1 + tmp2; - tmp13 = tmp1 - tmp2; - - tmp0 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*7]; - tmp1 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*6]; - tmp2 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*5]; - tmp3 = dataptr[DCTSIZE*3] - dataptr[DCTSIZE*4]; - - dataptr[DCTSIZE*0] = (DCTELEM) RIGHT_SHIFT(tmp10 + tmp11, PASS1_BITS); - dataptr[DCTSIZE*4] = (DCTELEM) RIGHT_SHIFT(tmp10 - tmp11, PASS1_BITS); - - z1 = MULTIPLY(tmp12 + tmp13, FIX_0_541196100); /* c6 */ - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS+PASS1_BITS-1); - - dataptr[DCTSIZE*2] = (DCTELEM) - RIGHT_SHIFT(z1 + MULTIPLY(tmp12, FIX_0_765366865), /* c2-c6 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*6] = (DCTELEM) - RIGHT_SHIFT(z1 - MULTIPLY(tmp13, FIX_1_847759065), /* c2+c6 */ - CONST_BITS+PASS1_BITS); - - /* Odd part per figure 8 --- note paper omits factor of sqrt(2). - * i0..i3 in the paper are tmp0..tmp3 here. - */ - - tmp12 = tmp0 + tmp2; - tmp13 = tmp1 + tmp3; - - z1 = MULTIPLY(tmp12 + tmp13, FIX_1_175875602); /* c3 */ - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS+PASS1_BITS-1); - - tmp12 = MULTIPLY(tmp12, - FIX_0_390180644); /* -c3+c5 */ - tmp13 = MULTIPLY(tmp13, - FIX_1_961570560); /* -c3-c5 */ - tmp12 += z1; - tmp13 += z1; - - z1 = MULTIPLY(tmp0 + tmp3, - FIX_0_899976223); /* -c3+c7 */ - tmp0 = MULTIPLY(tmp0, FIX_1_501321110); /* c1+c3-c5-c7 */ - tmp3 = MULTIPLY(tmp3, FIX_0_298631336); /* -c1+c3+c5-c7 */ - tmp0 += z1 + tmp12; - tmp3 += z1 + tmp13; - - z1 = MULTIPLY(tmp1 + tmp2, - FIX_2_562915447); /* -c1-c3 */ - tmp1 = MULTIPLY(tmp1, FIX_3_072711026); /* c1+c3+c5-c7 */ - tmp2 = MULTIPLY(tmp2, FIX_2_053119869); /* c1+c3-c5+c7 */ - tmp1 += z1 + tmp13; - tmp2 += z1 + tmp12; - - dataptr[DCTSIZE*1] = (DCTELEM) RIGHT_SHIFT(tmp0, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*3] = (DCTELEM) RIGHT_SHIFT(tmp1, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*5] = (DCTELEM) RIGHT_SHIFT(tmp2, CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*7] = (DCTELEM) RIGHT_SHIFT(tmp3, CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 3x6 sample block. - * - * 3-point FDCT in pass 1 (rows), 6-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_3x6 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1, tmp2; - INT32 tmp10, tmp11, tmp12; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * We scale the results further by 2 as part of output adaption - * scaling for different DCT size. - * 3-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/6). - */ - - dataptr = data; - for (ctr = 0; ctr < 6; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]) + GETJSAMPLE(elemptr[2]); - tmp1 = GETJSAMPLE(elemptr[1]); - - tmp2 = GETJSAMPLE(elemptr[0]) - GETJSAMPLE(elemptr[2]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) - ((tmp0 + tmp1 - 3 * CENTERJSAMPLE) << (PASS1_BITS+1)); - dataptr[2] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 - tmp1 - tmp1, FIX(0.707106781)), /* c2 */ - CONST_BITS-PASS1_BITS-1); - - /* Odd part */ - - dataptr[1] = (DCTELEM) - DESCALE(MULTIPLY(tmp2, FIX(1.224744871)), /* c1 */ - CONST_BITS-PASS1_BITS-1); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We remove the PASS1_BITS scaling, but leave the results scaled up - * by an overall factor of 8. - * We must also scale the output by (8/6)*(8/3) = 32/9, which we partially - * fold into the constant multipliers (other part was done in pass 1): - * 6-point FDCT kernel, cK represents sqrt(2) * cos(K*pi/12) * 16/9. - */ - - dataptr = data; - for (ctr = 0; ctr < 3; ctr++) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*5]; - tmp11 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*4]; - tmp2 = dataptr[DCTSIZE*2] + dataptr[DCTSIZE*3]; - - tmp10 = tmp0 + tmp2; - tmp12 = tmp0 - tmp2; - - tmp0 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*5]; - tmp1 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*4]; - tmp2 = dataptr[DCTSIZE*2] - dataptr[DCTSIZE*3]; - - dataptr[DCTSIZE*0] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 + tmp11, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*2] = (DCTELEM) - DESCALE(MULTIPLY(tmp12, FIX(2.177324216)), /* c2 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*4] = (DCTELEM) - DESCALE(MULTIPLY(tmp10 - tmp11 - tmp11, FIX(1.257078722)), /* c4 */ - CONST_BITS+PASS1_BITS); - - /* Odd part */ - - tmp10 = MULTIPLY(tmp0 + tmp2, FIX(0.650711829)); /* c5 */ - - dataptr[DCTSIZE*1] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp0 + tmp1, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*3] = (DCTELEM) - DESCALE(MULTIPLY(tmp0 - tmp1 - tmp2, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS); - dataptr[DCTSIZE*5] = (DCTELEM) - DESCALE(tmp10 + MULTIPLY(tmp2 - tmp1, FIX(1.777777778)), /* 16/9 */ - CONST_BITS+PASS1_BITS); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 2x4 sample block. - * - * 2-point FDCT in pass 1 (rows), 4-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_2x4 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - INT32 tmp0, tmp1; - INT32 tmp10, tmp11; - DCTELEM *dataptr; - JSAMPROW elemptr; - int ctr; - SHIFT_TEMPS - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: process rows. - * Note results are scaled up by sqrt(8) compared to a true DCT. - */ - - dataptr = data; - for (ctr = 0; ctr < 4; ctr++) { - elemptr = sample_data[ctr] + start_col; - - /* Even part */ - - tmp0 = GETJSAMPLE(elemptr[0]); - tmp1 = GETJSAMPLE(elemptr[1]); - - /* Apply unsigned->signed conversion. */ - dataptr[0] = (DCTELEM) (tmp0 + tmp1 - 2 * CENTERJSAMPLE); - - /* Odd part */ - - dataptr[1] = (DCTELEM) (tmp0 - tmp1); - - dataptr += DCTSIZE; /* advance pointer to next row */ - } - - /* Pass 2: process columns. - * We leave the results scaled up by an overall factor of 8. - * We must also scale the output by (8/2)*(8/4) = 2**3. - * 4-point FDCT kernel, - * cK represents sqrt(2) * cos(K*pi/16) [refers to 8-point FDCT]. - */ - - dataptr = data; - for (ctr = 0; ctr < 2; ctr++) { - /* Even part */ - - tmp0 = dataptr[DCTSIZE*0] + dataptr[DCTSIZE*3]; - tmp1 = dataptr[DCTSIZE*1] + dataptr[DCTSIZE*2]; - - tmp10 = dataptr[DCTSIZE*0] - dataptr[DCTSIZE*3]; - tmp11 = dataptr[DCTSIZE*1] - dataptr[DCTSIZE*2]; - - dataptr[DCTSIZE*0] = (DCTELEM) ((tmp0 + tmp1) << 3); - dataptr[DCTSIZE*2] = (DCTELEM) ((tmp0 - tmp1) << 3); - - /* Odd part */ - - tmp0 = MULTIPLY(tmp10 + tmp11, FIX_0_541196100); /* c6 */ - /* Add fudge factor here for final descale. */ - tmp0 += ONE << (CONST_BITS-3-1); - - dataptr[DCTSIZE*1] = (DCTELEM) - RIGHT_SHIFT(tmp0 + MULTIPLY(tmp10, FIX_0_765366865), /* c2-c6 */ - CONST_BITS-3); - dataptr[DCTSIZE*3] = (DCTELEM) - RIGHT_SHIFT(tmp0 - MULTIPLY(tmp11, FIX_1_847759065), /* c2+c6 */ - CONST_BITS-3); - - dataptr++; /* advance pointer to next column */ - } -} - - -/* - * Perform the forward DCT on a 1x2 sample block. - * - * 1-point FDCT in pass 1 (rows), 2-point in pass 2 (columns). - */ - -GLOBAL(void) -jpeg_fdct_1x2 (DCTELEM * data, JSAMPARRAY sample_data, JDIMENSION start_col) -{ - DCTELEM tmp0, tmp1; - - /* Pre-zero output coefficient block. */ - MEMZERO(data, SIZEOF(DCTELEM) * DCTSIZE2); - - /* Pass 1: empty. */ - - /* Pass 2: process columns. - * We leave the results scaled up by an overall factor of 8. - * We must also scale the output by (8/1)*(8/2) = 2**5. - */ - - /* Even part */ - - tmp0 = GETJSAMPLE(sample_data[0][start_col]); - tmp1 = GETJSAMPLE(sample_data[1][start_col]); - - /* Apply unsigned->signed conversion. */ - data[DCTSIZE*0] = (tmp0 + tmp1 - 2 * CENTERJSAMPLE) << 5; - - /* Odd part */ - - data[DCTSIZE*1] = (tmp0 - tmp1) << 5; -} - -#endif /* DCT_SCALING_SUPPORTED */ -#endif /* DCT_ISLOW_SUPPORTED */ diff --git a/dep/libjpeg/src/jidctflt.c b/dep/libjpeg/src/jidctflt.c deleted file mode 100644 index e33a2b5e4..000000000 --- a/dep/libjpeg/src/jidctflt.c +++ /dev/null @@ -1,238 +0,0 @@ -/* - * jidctflt.c - * - * Copyright (C) 1994-1998, Thomas G. Lane. - * Modified 2010-2017 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains a floating-point implementation of the - * inverse DCT (Discrete Cosine Transform). In the IJG code, this routine - * must also perform dequantization of the input coefficients. - * - * This implementation should be more accurate than either of the integer - * IDCT implementations. However, it may not give the same results on all - * machines because of differences in roundoff behavior. Speed will depend - * on the hardware's floating point capacity. - * - * A 2-D IDCT can be done by 1-D IDCT on each column followed by 1-D IDCT - * on each row (or vice versa, but it's more convenient to emit a row at - * a time). Direct algorithms are also available, but they are much more - * complex and seem not to be any faster when reduced to code. - * - * This implementation is based on Arai, Agui, and Nakajima's algorithm for - * scaled DCT. Their original paper (Trans. IEICE E-71(11):1095) is in - * Japanese, but the algorithm is described in the Pennebaker & Mitchell - * JPEG textbook (see REFERENCES section in file README). The following code - * is based directly on figure 4-8 in P&M. - * While an 8-point DCT cannot be done in less than 11 multiplies, it is - * possible to arrange the computation so that many of the multiplies are - * simple scalings of the final outputs. These multiplies can then be - * folded into the multiplications or divisions by the JPEG quantization - * table entries. The AA&N method leaves only 5 multiplies and 29 adds - * to be done in the DCT itself. - * The primary disadvantage of this method is that with a fixed-point - * implementation, accuracy is lost due to imprecise representation of the - * scaled quantization values. However, that problem does not arise if - * we use floating point arithmetic. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" -#include "jdct.h" /* Private declarations for DCT subsystem */ - -#ifdef DCT_FLOAT_SUPPORTED - - -/* - * This module is specialized to the case DCTSIZE = 8. - */ - -#if DCTSIZE != 8 - Sorry, this code only copes with 8x8 DCT blocks. /* deliberate syntax err */ -#endif - - -/* Dequantize a coefficient by multiplying it by the multiplier-table - * entry; produce a float result. - */ - -#define DEQUANTIZE(coef,quantval) (((FAST_FLOAT) (coef)) * (quantval)) - - -/* - * Perform dequantization and inverse DCT on one block of coefficients. - * - * cK represents cos(K*pi/16). - */ - -GLOBAL(void) -jpeg_idct_float (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - FAST_FLOAT tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7; - FAST_FLOAT tmp10, tmp11, tmp12, tmp13; - FAST_FLOAT z5, z10, z11, z12, z13; - JCOEFPTR inptr; - FLOAT_MULT_TYPE * quantptr; - FAST_FLOAT * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - FAST_FLOAT workspace[DCTSIZE2]; /* buffers data between passes */ - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (FLOAT_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = DCTSIZE; ctr > 0; ctr--) { - /* Due to quantization, we will usually find that many of the input - * coefficients are zero, especially the AC terms. We can exploit this - * by short-circuiting the IDCT calculation for any column in which all - * the AC terms are zero. In that case each output is equal to the - * DC coefficient (with scale factor as needed). - * With typical images and quantization tables, half or more of the - * column DCT calculations can be simplified this way. - */ - - if (inptr[DCTSIZE*1] == 0 && inptr[DCTSIZE*2] == 0 && - inptr[DCTSIZE*3] == 0 && inptr[DCTSIZE*4] == 0 && - inptr[DCTSIZE*5] == 0 && inptr[DCTSIZE*6] == 0 && - inptr[DCTSIZE*7] == 0) { - /* AC terms all zero */ - FAST_FLOAT dcval = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - - wsptr[DCTSIZE*0] = dcval; - wsptr[DCTSIZE*1] = dcval; - wsptr[DCTSIZE*2] = dcval; - wsptr[DCTSIZE*3] = dcval; - wsptr[DCTSIZE*4] = dcval; - wsptr[DCTSIZE*5] = dcval; - wsptr[DCTSIZE*6] = dcval; - wsptr[DCTSIZE*7] = dcval; - - inptr++; /* advance pointers to next column */ - quantptr++; - wsptr++; - continue; - } - - /* Even part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - tmp2 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - tmp3 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - tmp10 = tmp0 + tmp2; /* phase 3 */ - tmp11 = tmp0 - tmp2; - - tmp13 = tmp1 + tmp3; /* phases 5-3 */ - tmp12 = (tmp1 - tmp3) * ((FAST_FLOAT) 1.414213562) - tmp13; /* 2*c4 */ - - tmp0 = tmp10 + tmp13; /* phase 2 */ - tmp3 = tmp10 - tmp13; - tmp1 = tmp11 + tmp12; - tmp2 = tmp11 - tmp12; - - /* Odd part */ - - tmp4 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - tmp5 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - tmp6 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - tmp7 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - - z13 = tmp6 + tmp5; /* phase 6 */ - z10 = tmp6 - tmp5; - z11 = tmp4 + tmp7; - z12 = tmp4 - tmp7; - - tmp7 = z11 + z13; /* phase 5 */ - tmp11 = (z11 - z13) * ((FAST_FLOAT) 1.414213562); /* 2*c4 */ - - z5 = (z10 + z12) * ((FAST_FLOAT) 1.847759065); /* 2*c2 */ - tmp10 = z5 - z12 * ((FAST_FLOAT) 1.082392200); /* 2*(c2-c6) */ - tmp12 = z5 - z10 * ((FAST_FLOAT) 2.613125930); /* 2*(c2+c6) */ - - tmp6 = tmp12 - tmp7; /* phase 2 */ - tmp5 = tmp11 - tmp6; - tmp4 = tmp10 - tmp5; - - wsptr[DCTSIZE*0] = tmp0 + tmp7; - wsptr[DCTSIZE*7] = tmp0 - tmp7; - wsptr[DCTSIZE*1] = tmp1 + tmp6; - wsptr[DCTSIZE*6] = tmp1 - tmp6; - wsptr[DCTSIZE*2] = tmp2 + tmp5; - wsptr[DCTSIZE*5] = tmp2 - tmp5; - wsptr[DCTSIZE*3] = tmp3 + tmp4; - wsptr[DCTSIZE*4] = tmp3 - tmp4; - - inptr++; /* advance pointers to next column */ - quantptr++; - wsptr++; - } - - /* Pass 2: process rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < DCTSIZE; ctr++) { - outptr = output_buf[ctr] + output_col; - /* Rows of zeroes can be exploited in the same way as we did with columns. - * However, the column calculation has created many nonzero AC terms, so - * the simplification applies less often (typically 5% to 10% of the time). - * And testing floats for zero is relatively expensive, so we don't bother. - */ - - /* Even part */ - - /* Prepare range-limit and float->int conversion */ - z5 = wsptr[0] + (((FAST_FLOAT) RANGE_CENTER) + ((FAST_FLOAT) 0.5)); - tmp10 = z5 + wsptr[4]; - tmp11 = z5 - wsptr[4]; - - tmp13 = wsptr[2] + wsptr[6]; - tmp12 = (wsptr[2] - wsptr[6]) * - ((FAST_FLOAT) 1.414213562) - tmp13; /* 2*c4 */ - - tmp0 = tmp10 + tmp13; - tmp3 = tmp10 - tmp13; - tmp1 = tmp11 + tmp12; - tmp2 = tmp11 - tmp12; - - /* Odd part */ - - z13 = wsptr[5] + wsptr[3]; - z10 = wsptr[5] - wsptr[3]; - z11 = wsptr[1] + wsptr[7]; - z12 = wsptr[1] - wsptr[7]; - - tmp7 = z11 + z13; /* phase 5 */ - tmp11 = (z11 - z13) * ((FAST_FLOAT) 1.414213562); /* 2*c4 */ - - z5 = (z10 + z12) * ((FAST_FLOAT) 1.847759065); /* 2*c2 */ - tmp10 = z5 - z12 * ((FAST_FLOAT) 1.082392200); /* 2*(c2-c6) */ - tmp12 = z5 - z10 * ((FAST_FLOAT) 2.613125930); /* 2*(c2+c6) */ - - tmp6 = tmp12 - tmp7; /* phase 2 */ - tmp5 = tmp11 - tmp6; - tmp4 = tmp10 - tmp5; - - /* Final output stage: float->int conversion and range-limit */ - - outptr[0] = range_limit[(int) (tmp0 + tmp7) & RANGE_MASK]; - outptr[7] = range_limit[(int) (tmp0 - tmp7) & RANGE_MASK]; - outptr[1] = range_limit[(int) (tmp1 + tmp6) & RANGE_MASK]; - outptr[6] = range_limit[(int) (tmp1 - tmp6) & RANGE_MASK]; - outptr[2] = range_limit[(int) (tmp2 + tmp5) & RANGE_MASK]; - outptr[5] = range_limit[(int) (tmp2 - tmp5) & RANGE_MASK]; - outptr[3] = range_limit[(int) (tmp3 + tmp4) & RANGE_MASK]; - outptr[4] = range_limit[(int) (tmp3 - tmp4) & RANGE_MASK]; - - wsptr += DCTSIZE; /* advance pointer to next row */ - } -} - -#endif /* DCT_FLOAT_SUPPORTED */ diff --git a/dep/libjpeg/src/jidctfst.c b/dep/libjpeg/src/jidctfst.c deleted file mode 100644 index 1ac3e39cb..000000000 --- a/dep/libjpeg/src/jidctfst.c +++ /dev/null @@ -1,351 +0,0 @@ -/* - * jidctfst.c - * - * Copyright (C) 1994-1998, Thomas G. Lane. - * Modified 2015-2017 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains a fast, not so accurate integer implementation of the - * inverse DCT (Discrete Cosine Transform). In the IJG code, this routine - * must also perform dequantization of the input coefficients. - * - * A 2-D IDCT can be done by 1-D IDCT on each column followed by 1-D IDCT - * on each row (or vice versa, but it's more convenient to emit a row at - * a time). Direct algorithms are also available, but they are much more - * complex and seem not to be any faster when reduced to code. - * - * This implementation is based on Arai, Agui, and Nakajima's algorithm for - * scaled DCT. Their original paper (Trans. IEICE E-71(11):1095) is in - * Japanese, but the algorithm is described in the Pennebaker & Mitchell - * JPEG textbook (see REFERENCES section in file README). The following code - * is based directly on figure 4-8 in P&M. - * While an 8-point DCT cannot be done in less than 11 multiplies, it is - * possible to arrange the computation so that many of the multiplies are - * simple scalings of the final outputs. These multiplies can then be - * folded into the multiplications or divisions by the JPEG quantization - * table entries. The AA&N method leaves only 5 multiplies and 29 adds - * to be done in the DCT itself. - * The primary disadvantage of this method is that with fixed-point math, - * accuracy is lost due to imprecise representation of the scaled - * quantization values. The smaller the quantization table entry, the less - * precise the scaled value, so this implementation does worse with high- - * quality-setting files than with low-quality ones. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" -#include "jdct.h" /* Private declarations for DCT subsystem */ - -#ifdef DCT_IFAST_SUPPORTED - - -/* - * This module is specialized to the case DCTSIZE = 8. - */ - -#if DCTSIZE != 8 - Sorry, this code only copes with 8x8 DCT blocks. /* deliberate syntax err */ -#endif - - -/* Scaling decisions are generally the same as in the LL&M algorithm; - * see jidctint.c for more details. However, we choose to descale - * (right shift) multiplication products as soon as they are formed, - * rather than carrying additional fractional bits into subsequent additions. - * This compromises accuracy slightly, but it lets us save a few shifts. - * More importantly, 16-bit arithmetic is then adequate (for 8-bit samples) - * everywhere except in the multiplications proper; this saves a good deal - * of work on 16-bit-int machines. - * - * The dequantized coefficients are not integers because the AA&N scaling - * factors have been incorporated. We represent them scaled up by PASS1_BITS, - * so that the first and second IDCT rounds have the same input scaling. - * For 8-bit JSAMPLEs, we choose IFAST_SCALE_BITS = PASS1_BITS so as to - * avoid a descaling shift; this compromises accuracy rather drastically - * for small quantization table entries, but it saves a lot of shifts. - * For 12-bit JSAMPLEs, there's no hope of using 16x16 multiplies anyway, - * so we use a much larger scaling factor to preserve accuracy. - * - * A final compromise is to represent the multiplicative constants to only - * 8 fractional bits, rather than 13. This saves some shifting work on some - * machines, and may also reduce the cost of multiplication (since there - * are fewer one-bits in the constants). - */ - -#if BITS_IN_JSAMPLE == 8 -#define CONST_BITS 8 -#define PASS1_BITS 2 -#else -#define CONST_BITS 8 -#define PASS1_BITS 1 /* lose a little precision to avoid overflow */ -#endif - -/* Some C compilers fail to reduce "FIX(constant)" at compile time, thus - * causing a lot of useless floating-point operations at run time. - * To get around this we use the following pre-calculated constants. - * If you change CONST_BITS you may want to add appropriate values. - * (With a reasonable C compiler, you can just rely on the FIX() macro...) - */ - -#if CONST_BITS == 8 -#define FIX_1_082392200 ((INT32) 277) /* FIX(1.082392200) */ -#define FIX_1_414213562 ((INT32) 362) /* FIX(1.414213562) */ -#define FIX_1_847759065 ((INT32) 473) /* FIX(1.847759065) */ -#define FIX_2_613125930 ((INT32) 669) /* FIX(2.613125930) */ -#else -#define FIX_1_082392200 FIX(1.082392200) -#define FIX_1_414213562 FIX(1.414213562) -#define FIX_1_847759065 FIX(1.847759065) -#define FIX_2_613125930 FIX(2.613125930) -#endif - - -/* We can gain a little more speed, with a further compromise in accuracy, - * by omitting the addition in a descaling shift. This yields an incorrectly - * rounded result half the time... - */ - -#ifndef USE_ACCURATE_ROUNDING -#undef DESCALE -#define DESCALE(x,n) RIGHT_SHIFT(x, n) -#endif - - -/* Multiply a DCTELEM variable by an INT32 constant, and immediately - * descale to yield a DCTELEM result. - */ - -#define MULTIPLY(var,const) ((DCTELEM) DESCALE((var) * (const), CONST_BITS)) - - -/* Dequantize a coefficient by multiplying it by the multiplier-table - * entry; produce a DCTELEM result. For 8-bit data a 16x16->16 - * multiplication will do. For 12-bit data, the multiplier table is - * declared INT32, so a 32-bit multiply will be used. - */ - -#if BITS_IN_JSAMPLE == 8 -#define DEQUANTIZE(coef,quantval) (((IFAST_MULT_TYPE) (coef)) * (quantval)) -#else -#define DEQUANTIZE(coef,quantval) \ - DESCALE((coef)*(quantval), IFAST_SCALE_BITS-PASS1_BITS) -#endif - - -/* - * Perform dequantization and inverse DCT on one block of coefficients. - * - * cK represents cos(K*pi/16). - */ - -GLOBAL(void) -jpeg_idct_ifast (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - DCTELEM tmp0, tmp1, tmp2, tmp3, tmp4, tmp5, tmp6, tmp7; - DCTELEM tmp10, tmp11, tmp12, tmp13; - DCTELEM z5, z10, z11, z12, z13; - JCOEFPTR inptr; - IFAST_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[DCTSIZE2]; /* buffers data between passes */ - SHIFT_TEMPS /* for DESCALE */ - ISHIFT_TEMPS /* for IRIGHT_SHIFT */ - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (IFAST_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = DCTSIZE; ctr > 0; ctr--) { - /* Due to quantization, we will usually find that many of the input - * coefficients are zero, especially the AC terms. We can exploit this - * by short-circuiting the IDCT calculation for any column in which all - * the AC terms are zero. In that case each output is equal to the - * DC coefficient (with scale factor as needed). - * With typical images and quantization tables, half or more of the - * column DCT calculations can be simplified this way. - */ - - if (inptr[DCTSIZE*1] == 0 && inptr[DCTSIZE*2] == 0 && - inptr[DCTSIZE*3] == 0 && inptr[DCTSIZE*4] == 0 && - inptr[DCTSIZE*5] == 0 && inptr[DCTSIZE*6] == 0 && - inptr[DCTSIZE*7] == 0) { - /* AC terms all zero */ - int dcval = (int) DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - - wsptr[DCTSIZE*0] = dcval; - wsptr[DCTSIZE*1] = dcval; - wsptr[DCTSIZE*2] = dcval; - wsptr[DCTSIZE*3] = dcval; - wsptr[DCTSIZE*4] = dcval; - wsptr[DCTSIZE*5] = dcval; - wsptr[DCTSIZE*6] = dcval; - wsptr[DCTSIZE*7] = dcval; - - inptr++; /* advance pointers to next column */ - quantptr++; - wsptr++; - continue; - } - - /* Even part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - tmp2 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - tmp3 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - tmp10 = tmp0 + tmp2; /* phase 3 */ - tmp11 = tmp0 - tmp2; - - tmp13 = tmp1 + tmp3; /* phases 5-3 */ - tmp12 = MULTIPLY(tmp1 - tmp3, FIX_1_414213562) - tmp13; /* 2*c4 */ - - tmp0 = tmp10 + tmp13; /* phase 2 */ - tmp3 = tmp10 - tmp13; - tmp1 = tmp11 + tmp12; - tmp2 = tmp11 - tmp12; - - /* Odd part */ - - tmp4 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - tmp5 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - tmp6 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - tmp7 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - - z13 = tmp6 + tmp5; /* phase 6 */ - z10 = tmp6 - tmp5; - z11 = tmp4 + tmp7; - z12 = tmp4 - tmp7; - - tmp7 = z11 + z13; /* phase 5 */ - tmp11 = MULTIPLY(z11 - z13, FIX_1_414213562); /* 2*c4 */ - - z5 = MULTIPLY(z10 + z12, FIX_1_847759065); /* 2*c2 */ - tmp10 = z5 - MULTIPLY(z12, FIX_1_082392200); /* 2*(c2-c6) */ - tmp12 = z5 - MULTIPLY(z10, FIX_2_613125930); /* 2*(c2+c6) */ - - tmp6 = tmp12 - tmp7; /* phase 2 */ - tmp5 = tmp11 - tmp6; - tmp4 = tmp10 - tmp5; - - wsptr[DCTSIZE*0] = (int) (tmp0 + tmp7); - wsptr[DCTSIZE*7] = (int) (tmp0 - tmp7); - wsptr[DCTSIZE*1] = (int) (tmp1 + tmp6); - wsptr[DCTSIZE*6] = (int) (tmp1 - tmp6); - wsptr[DCTSIZE*2] = (int) (tmp2 + tmp5); - wsptr[DCTSIZE*5] = (int) (tmp2 - tmp5); - wsptr[DCTSIZE*3] = (int) (tmp3 + tmp4); - wsptr[DCTSIZE*4] = (int) (tmp3 - tmp4); - - inptr++; /* advance pointers to next column */ - quantptr++; - wsptr++; - } - - /* Pass 2: process rows from work array, store into output array. - * Note that we must descale the results by a factor of 8 == 2**3, - * and also undo the PASS1_BITS scaling. - */ - - wsptr = workspace; - for (ctr = 0; ctr < DCTSIZE; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Add range center and fudge factor for final descale and range-limit. */ - z5 = (DCTELEM) wsptr[0] + - ((((DCTELEM) RANGE_CENTER) << (PASS1_BITS+3)) + - (1 << (PASS1_BITS+2))); - - /* Rows of zeroes can be exploited in the same way as we did with columns. - * However, the column calculation has created many nonzero AC terms, so - * the simplification applies less often (typically 5% to 10% of the time). - * On machines with very fast multiplication, it's possible that the - * test takes more time than it's worth. In that case this section - * may be commented out. - */ - -#ifndef NO_ZERO_ROW_TEST - if (wsptr[1] == 0 && wsptr[2] == 0 && wsptr[3] == 0 && wsptr[4] == 0 && - wsptr[5] == 0 && wsptr[6] == 0 && wsptr[7] == 0) { - /* AC terms all zero */ - JSAMPLE dcval = range_limit[(int) IRIGHT_SHIFT(z5, PASS1_BITS+3) - & RANGE_MASK]; - - outptr[0] = dcval; - outptr[1] = dcval; - outptr[2] = dcval; - outptr[3] = dcval; - outptr[4] = dcval; - outptr[5] = dcval; - outptr[6] = dcval; - outptr[7] = dcval; - - wsptr += DCTSIZE; /* advance pointer to next row */ - continue; - } -#endif - - /* Even part */ - - tmp10 = z5 + (DCTELEM) wsptr[4]; - tmp11 = z5 - (DCTELEM) wsptr[4]; - - tmp13 = (DCTELEM) wsptr[2] + (DCTELEM) wsptr[6]; - tmp12 = MULTIPLY((DCTELEM) wsptr[2] - (DCTELEM) wsptr[6], - FIX_1_414213562) - tmp13; /* 2*c4 */ - - tmp0 = tmp10 + tmp13; - tmp3 = tmp10 - tmp13; - tmp1 = tmp11 + tmp12; - tmp2 = tmp11 - tmp12; - - /* Odd part */ - - z13 = (DCTELEM) wsptr[5] + (DCTELEM) wsptr[3]; - z10 = (DCTELEM) wsptr[5] - (DCTELEM) wsptr[3]; - z11 = (DCTELEM) wsptr[1] + (DCTELEM) wsptr[7]; - z12 = (DCTELEM) wsptr[1] - (DCTELEM) wsptr[7]; - - tmp7 = z11 + z13; /* phase 5 */ - tmp11 = MULTIPLY(z11 - z13, FIX_1_414213562); /* 2*c4 */ - - z5 = MULTIPLY(z10 + z12, FIX_1_847759065); /* 2*c2 */ - tmp10 = z5 - MULTIPLY(z12, FIX_1_082392200); /* 2*(c2-c6) */ - tmp12 = z5 - MULTIPLY(z10, FIX_2_613125930); /* 2*(c2+c6) */ - - tmp6 = tmp12 - tmp7; /* phase 2 */ - tmp5 = tmp11 - tmp6; - tmp4 = tmp10 - tmp5; - - /* Final output stage: scale down by a factor of 8 and range-limit */ - - outptr[0] = range_limit[(int) IRIGHT_SHIFT(tmp0 + tmp7, PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) IRIGHT_SHIFT(tmp0 - tmp7, PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) IRIGHT_SHIFT(tmp1 + tmp6, PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) IRIGHT_SHIFT(tmp1 - tmp6, PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) IRIGHT_SHIFT(tmp2 + tmp5, PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) IRIGHT_SHIFT(tmp2 - tmp5, PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) IRIGHT_SHIFT(tmp3 + tmp4, PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) IRIGHT_SHIFT(tmp3 - tmp4, PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += DCTSIZE; /* advance pointer to next row */ - } -} - -#endif /* DCT_IFAST_SUPPORTED */ diff --git a/dep/libjpeg/src/jidctint.c b/dep/libjpeg/src/jidctint.c deleted file mode 100644 index e30ec8c8a..000000000 --- a/dep/libjpeg/src/jidctint.c +++ /dev/null @@ -1,5240 +0,0 @@ -/* - * jidctint.c - * - * Copyright (C) 1991-1998, Thomas G. Lane. - * Modification developed 2002-2018 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains a slow-but-accurate integer implementation of the - * inverse DCT (Discrete Cosine Transform). In the IJG code, this routine - * must also perform dequantization of the input coefficients. - * - * A 2-D IDCT can be done by 1-D IDCT on each column followed by 1-D IDCT - * on each row (or vice versa, but it's more convenient to emit a row at - * a time). Direct algorithms are also available, but they are much more - * complex and seem not to be any faster when reduced to code. - * - * This implementation is based on an algorithm described in - * C. Loeffler, A. Ligtenberg and G. Moschytz, "Practical Fast 1-D DCT - * Algorithms with 11 Multiplications", Proc. Int'l. Conf. on Acoustics, - * Speech, and Signal Processing 1989 (ICASSP '89), pp. 988-991. - * The primary algorithm described there uses 11 multiplies and 29 adds. - * We use their alternate method with 12 multiplies and 32 adds. - * The advantage of this method is that no data path contains more than one - * multiplication; this allows a very simple and accurate implementation in - * scaled fixed-point arithmetic, with a minimal number of shifts. - * - * We also provide IDCT routines with various output sample block sizes for - * direct resolution reduction or enlargement and for direct resolving the - * common 2x1 and 1x2 subsampling cases without additional resampling: NxN - * (N=1...16), 2NxN, and Nx2N (N=1...8) pixels for one 8x8 input DCT block. - * - * For N<8 we simply take the corresponding low-frequency coefficients of - * the 8x8 input DCT block and apply an NxN point IDCT on the sub-block - * to yield the downscaled outputs. - * This can be seen as direct low-pass downsampling from the DCT domain - * point of view rather than the usual spatial domain point of view, - * yielding significant computational savings and results at least - * as good as common bilinear (averaging) spatial downsampling. - * - * For N>8 we apply a partial NxN IDCT on the 8 input coefficients as - * lower frequencies and higher frequencies assumed to be zero. - * It turns out that the computational effort is similar to the 8x8 IDCT - * regarding the output size. - * Furthermore, the scaling and descaling is the same for all IDCT sizes. - * - * CAUTION: We rely on the FIX() macro except for the N=1,2,4,8 cases - * since there would be too many additional constants to pre-calculate. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" -#include "jdct.h" /* Private declarations for DCT subsystem */ - -#ifdef DCT_ISLOW_SUPPORTED - - -/* - * This module is specialized to the case DCTSIZE = 8. - */ - -#if DCTSIZE != 8 - Sorry, this code only copes with 8x8 DCT blocks. /* deliberate syntax err */ -#endif - - -/* - * The poop on this scaling stuff is as follows: - * - * Each 1-D IDCT step produces outputs which are a factor of sqrt(N) - * larger than the true IDCT outputs. The final outputs are therefore - * a factor of N larger than desired; since N=8 this can be cured by - * a simple right shift at the end of the algorithm. The advantage of - * this arrangement is that we save two multiplications per 1-D IDCT, - * because the y0 and y4 inputs need not be divided by sqrt(N). - * - * We have to do addition and subtraction of the integer inputs, which - * is no problem, and multiplication by fractional constants, which is - * a problem to do in integer arithmetic. We multiply all the constants - * by CONST_SCALE and convert them to integer constants (thus retaining - * CONST_BITS bits of precision in the constants). After doing a - * multiplication we have to divide the product by CONST_SCALE, with proper - * rounding, to produce the correct output. This division can be done - * cheaply as a right shift of CONST_BITS bits. We postpone shifting - * as long as possible so that partial sums can be added together with - * full fractional precision. - * - * The outputs of the first pass are scaled up by PASS1_BITS bits so that - * they are represented to better-than-integral precision. These outputs - * require BITS_IN_JSAMPLE + PASS1_BITS + 3 bits; this fits in a 16-bit word - * with the recommended scaling. (To scale up 12-bit sample data further, an - * intermediate INT32 array would be needed.) - * - * To avoid overflow of the 32-bit intermediate results in pass 2, we must - * have BITS_IN_JSAMPLE + CONST_BITS + PASS1_BITS <= 26. Error analysis - * shows that the values given below are the most effective. - */ - -#if BITS_IN_JSAMPLE == 8 -#define CONST_BITS 13 -#define PASS1_BITS 2 -#else -#define CONST_BITS 13 -#define PASS1_BITS 1 /* lose a little precision to avoid overflow */ -#endif - -/* Some C compilers fail to reduce "FIX(constant)" at compile time, thus - * causing a lot of useless floating-point operations at run time. - * To get around this we use the following pre-calculated constants. - * If you change CONST_BITS you may want to add appropriate values. - * (With a reasonable C compiler, you can just rely on the FIX() macro...) - */ - -#if CONST_BITS == 13 -#define FIX_0_298631336 ((INT32) 2446) /* FIX(0.298631336) */ -#define FIX_0_390180644 ((INT32) 3196) /* FIX(0.390180644) */ -#define FIX_0_541196100 ((INT32) 4433) /* FIX(0.541196100) */ -#define FIX_0_765366865 ((INT32) 6270) /* FIX(0.765366865) */ -#define FIX_0_899976223 ((INT32) 7373) /* FIX(0.899976223) */ -#define FIX_1_175875602 ((INT32) 9633) /* FIX(1.175875602) */ -#define FIX_1_501321110 ((INT32) 12299) /* FIX(1.501321110) */ -#define FIX_1_847759065 ((INT32) 15137) /* FIX(1.847759065) */ -#define FIX_1_961570560 ((INT32) 16069) /* FIX(1.961570560) */ -#define FIX_2_053119869 ((INT32) 16819) /* FIX(2.053119869) */ -#define FIX_2_562915447 ((INT32) 20995) /* FIX(2.562915447) */ -#define FIX_3_072711026 ((INT32) 25172) /* FIX(3.072711026) */ -#else -#define FIX_0_298631336 FIX(0.298631336) -#define FIX_0_390180644 FIX(0.390180644) -#define FIX_0_541196100 FIX(0.541196100) -#define FIX_0_765366865 FIX(0.765366865) -#define FIX_0_899976223 FIX(0.899976223) -#define FIX_1_175875602 FIX(1.175875602) -#define FIX_1_501321110 FIX(1.501321110) -#define FIX_1_847759065 FIX(1.847759065) -#define FIX_1_961570560 FIX(1.961570560) -#define FIX_2_053119869 FIX(2.053119869) -#define FIX_2_562915447 FIX(2.562915447) -#define FIX_3_072711026 FIX(3.072711026) -#endif - - -/* Multiply an INT32 variable by an INT32 constant to yield an INT32 result. - * For 8-bit samples with the recommended scaling, all the variable - * and constant values involved are no more than 16 bits wide, so a - * 16x16->32 bit multiply can be used instead of a full 32x32 multiply. - * For 12-bit samples, a full 32-bit multiplication will be needed. - */ - -#if BITS_IN_JSAMPLE == 8 -#define MULTIPLY(var,const) MULTIPLY16C16(var,const) -#else -#define MULTIPLY(var,const) ((var) * (const)) -#endif - - -/* Dequantize a coefficient by multiplying it by the multiplier-table - * entry; produce an int result. In this module, both inputs and result - * are 16 bits or less, so either int or short multiply will work. - */ - -#define DEQUANTIZE(coef,quantval) (((ISLOW_MULT_TYPE) (coef)) * (quantval)) - - -/* - * Perform dequantization and inverse DCT on one block of coefficients. - * - * Optimized algorithm with 12 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/16). - */ - -GLOBAL(void) -jpeg_idct_islow (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3; - INT32 tmp10, tmp11, tmp12, tmp13; - INT32 z1, z2, z3; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[DCTSIZE2]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * Note results are scaled up by sqrt(8) compared to a true IDCT; - * furthermore, we scale the results by 2**PASS1_BITS. - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = DCTSIZE; ctr > 0; ctr--) { - /* Due to quantization, we will usually find that many of the input - * coefficients are zero, especially the AC terms. We can exploit this - * by short-circuiting the IDCT calculation for any column in which all - * the AC terms are zero. In that case each output is equal to the - * DC coefficient (with scale factor as needed). - * With typical images and quantization tables, half or more of the - * column DCT calculations can be simplified this way. - */ - - if (inptr[DCTSIZE*1] == 0 && inptr[DCTSIZE*2] == 0 && - inptr[DCTSIZE*3] == 0 && inptr[DCTSIZE*4] == 0 && - inptr[DCTSIZE*5] == 0 && inptr[DCTSIZE*6] == 0 && - inptr[DCTSIZE*7] == 0) { - /* AC terms all zero */ - int dcval = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]) << PASS1_BITS; - - wsptr[DCTSIZE*0] = dcval; - wsptr[DCTSIZE*1] = dcval; - wsptr[DCTSIZE*2] = dcval; - wsptr[DCTSIZE*3] = dcval; - wsptr[DCTSIZE*4] = dcval; - wsptr[DCTSIZE*5] = dcval; - wsptr[DCTSIZE*6] = dcval; - wsptr[DCTSIZE*7] = dcval; - - inptr++; /* advance pointers to next column */ - quantptr++; - wsptr++; - continue; - } - - /* Even part: reverse the even part of the forward DCT. - * The rotator is c(-6). - */ - - z2 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - z3 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z2 <<= CONST_BITS; - z3 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - z2 += ONE << (CONST_BITS-PASS1_BITS-1); - - tmp0 = z2 + z3; - tmp1 = z2 - z3; - - z2 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z3 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - tmp2 = z1 + MULTIPLY(z2, FIX_0_765366865); /* c2-c6 */ - tmp3 = z1 - MULTIPLY(z3, FIX_1_847759065); /* c2+c6 */ - - tmp10 = tmp0 + tmp2; - tmp13 = tmp0 - tmp2; - tmp11 = tmp1 + tmp3; - tmp12 = tmp1 - tmp3; - - /* Odd part per figure 8; the matrix is unitary and hence its - * transpose is its inverse. i0..i3 are y7,y5,y3,y1 respectively. - */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - tmp1 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - tmp2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - tmp3 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - - z2 = tmp0 + tmp2; - z3 = tmp1 + tmp3; - - z1 = MULTIPLY(z2 + z3, FIX_1_175875602); /* c3 */ - z2 = MULTIPLY(z2, - FIX_1_961570560); /* -c3-c5 */ - z3 = MULTIPLY(z3, - FIX_0_390180644); /* -c3+c5 */ - z2 += z1; - z3 += z1; - - z1 = MULTIPLY(tmp0 + tmp3, - FIX_0_899976223); /* -c3+c7 */ - tmp0 = MULTIPLY(tmp0, FIX_0_298631336); /* -c1+c3+c5-c7 */ - tmp3 = MULTIPLY(tmp3, FIX_1_501321110); /* c1+c3-c5-c7 */ - tmp0 += z1 + z2; - tmp3 += z1 + z3; - - z1 = MULTIPLY(tmp1 + tmp2, - FIX_2_562915447); /* -c1-c3 */ - tmp1 = MULTIPLY(tmp1, FIX_2_053119869); /* c1+c3-c5+c7 */ - tmp2 = MULTIPLY(tmp2, FIX_3_072711026); /* c1+c3+c5-c7 */ - tmp1 += z1 + z3; - tmp2 += z1 + z2; - - /* Final output stage: inputs are tmp10..tmp13, tmp0..tmp3 */ - - wsptr[DCTSIZE*0] = (int) RIGHT_SHIFT(tmp10 + tmp3, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*7] = (int) RIGHT_SHIFT(tmp10 - tmp3, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*1] = (int) RIGHT_SHIFT(tmp11 + tmp2, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*6] = (int) RIGHT_SHIFT(tmp11 - tmp2, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*2] = (int) RIGHT_SHIFT(tmp12 + tmp1, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*5] = (int) RIGHT_SHIFT(tmp12 - tmp1, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*3] = (int) RIGHT_SHIFT(tmp13 + tmp0, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*4] = (int) RIGHT_SHIFT(tmp13 - tmp0, CONST_BITS-PASS1_BITS); - - inptr++; /* advance pointers to next column */ - quantptr++; - wsptr++; - } - - /* Pass 2: process rows from work array, store into output array. - * Note that we must descale the results by a factor of 8 == 2**3, - * and also undo the PASS1_BITS scaling. - */ - - wsptr = workspace; - for (ctr = 0; ctr < DCTSIZE; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Add range center and fudge factor for final descale and range-limit. */ - z2 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - - /* Rows of zeroes can be exploited in the same way as we did with columns. - * However, the column calculation has created many nonzero AC terms, so - * the simplification applies less often (typically 5% to 10% of the time). - * On machines with very fast multiplication, it's possible that the - * test takes more time than it's worth. In that case this section - * may be commented out. - */ - -#ifndef NO_ZERO_ROW_TEST - if (wsptr[1] == 0 && wsptr[2] == 0 && wsptr[3] == 0 && wsptr[4] == 0 && - wsptr[5] == 0 && wsptr[6] == 0 && wsptr[7] == 0) { - /* AC terms all zero */ - JSAMPLE dcval = range_limit[(int) RIGHT_SHIFT(z2, PASS1_BITS+3) - & RANGE_MASK]; - - outptr[0] = dcval; - outptr[1] = dcval; - outptr[2] = dcval; - outptr[3] = dcval; - outptr[4] = dcval; - outptr[5] = dcval; - outptr[6] = dcval; - outptr[7] = dcval; - - wsptr += DCTSIZE; /* advance pointer to next row */ - continue; - } -#endif - - /* Even part: reverse the even part of the forward DCT. - * The rotator is c(-6). - */ - - z3 = (INT32) wsptr[4]; - - tmp0 = (z2 + z3) << CONST_BITS; - tmp1 = (z2 - z3) << CONST_BITS; - - z2 = (INT32) wsptr[2]; - z3 = (INT32) wsptr[6]; - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - tmp2 = z1 + MULTIPLY(z2, FIX_0_765366865); /* c2-c6 */ - tmp3 = z1 - MULTIPLY(z3, FIX_1_847759065); /* c2+c6 */ - - tmp10 = tmp0 + tmp2; - tmp13 = tmp0 - tmp2; - tmp11 = tmp1 + tmp3; - tmp12 = tmp1 - tmp3; - - /* Odd part per figure 8; the matrix is unitary and hence its - * transpose is its inverse. i0..i3 are y7,y5,y3,y1 respectively. - */ - - tmp0 = (INT32) wsptr[7]; - tmp1 = (INT32) wsptr[5]; - tmp2 = (INT32) wsptr[3]; - tmp3 = (INT32) wsptr[1]; - - z2 = tmp0 + tmp2; - z3 = tmp1 + tmp3; - - z1 = MULTIPLY(z2 + z3, FIX_1_175875602); /* c3 */ - z2 = MULTIPLY(z2, - FIX_1_961570560); /* -c3-c5 */ - z3 = MULTIPLY(z3, - FIX_0_390180644); /* -c3+c5 */ - z2 += z1; - z3 += z1; - - z1 = MULTIPLY(tmp0 + tmp3, - FIX_0_899976223); /* -c3+c7 */ - tmp0 = MULTIPLY(tmp0, FIX_0_298631336); /* -c1+c3+c5-c7 */ - tmp3 = MULTIPLY(tmp3, FIX_1_501321110); /* c1+c3-c5-c7 */ - tmp0 += z1 + z2; - tmp3 += z1 + z3; - - z1 = MULTIPLY(tmp1 + tmp2, - FIX_2_562915447); /* -c1-c3 */ - tmp1 = MULTIPLY(tmp1, FIX_2_053119869); /* c1+c3-c5+c7 */ - tmp2 = MULTIPLY(tmp2, FIX_3_072711026); /* c1+c3+c5-c7 */ - tmp1 += z1 + z3; - tmp2 += z1 + z2; - - /* Final output stage: inputs are tmp10..tmp13, tmp0..tmp3 */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp3, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp3, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp11 + tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp11 - tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp12 + tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp12 - tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp13 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp13 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += DCTSIZE; /* advance pointer to next row */ - } -} - -#ifdef IDCT_SCALING_SUPPORTED - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a reduced-size 7x7 output block. - * - * Optimized algorithm with 12 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/14). - */ - -GLOBAL(void) -jpeg_idct_7x7 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp1, tmp2, tmp10, tmp11, tmp12, tmp13; - INT32 z1, z2, z3; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[7*7]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 7; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp13 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp13 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp13 += ONE << (CONST_BITS-PASS1_BITS-1); - - z1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z2 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z3 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - tmp10 = MULTIPLY(z2 - z3, FIX(0.881747734)); /* c4 */ - tmp12 = MULTIPLY(z1 - z2, FIX(0.314692123)); /* c6 */ - tmp11 = tmp10 + tmp12 + tmp13 - MULTIPLY(z2, FIX(1.841218003)); /* c2+c4-c6 */ - tmp0 = z1 + z3; - z2 -= tmp0; - tmp0 = MULTIPLY(tmp0, FIX(1.274162392)) + tmp13; /* c2 */ - tmp10 += tmp0 - MULTIPLY(z3, FIX(0.077722536)); /* c2-c4-c6 */ - tmp12 += tmp0 - MULTIPLY(z1, FIX(2.470602249)); /* c2+c4+c6 */ - tmp13 += MULTIPLY(z2, FIX(1.414213562)); /* c0 */ - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - - tmp1 = MULTIPLY(z1 + z2, FIX(0.935414347)); /* (c3+c1-c5)/2 */ - tmp2 = MULTIPLY(z1 - z2, FIX(0.170262339)); /* (c3+c5-c1)/2 */ - tmp0 = tmp1 - tmp2; - tmp1 += tmp2; - tmp2 = MULTIPLY(z2 + z3, - FIX(1.378756276)); /* -c1 */ - tmp1 += tmp2; - z2 = MULTIPLY(z1 + z3, FIX(0.613604268)); /* c5 */ - tmp0 += z2; - tmp2 += z2 + MULTIPLY(z3, FIX(1.870828693)); /* c3+c1-c5 */ - - /* Final output stage */ - - wsptr[7*0] = (int) RIGHT_SHIFT(tmp10 + tmp0, CONST_BITS-PASS1_BITS); - wsptr[7*6] = (int) RIGHT_SHIFT(tmp10 - tmp0, CONST_BITS-PASS1_BITS); - wsptr[7*1] = (int) RIGHT_SHIFT(tmp11 + tmp1, CONST_BITS-PASS1_BITS); - wsptr[7*5] = (int) RIGHT_SHIFT(tmp11 - tmp1, CONST_BITS-PASS1_BITS); - wsptr[7*2] = (int) RIGHT_SHIFT(tmp12 + tmp2, CONST_BITS-PASS1_BITS); - wsptr[7*4] = (int) RIGHT_SHIFT(tmp12 - tmp2, CONST_BITS-PASS1_BITS); - wsptr[7*3] = (int) RIGHT_SHIFT(tmp13, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 7 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 7; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp13 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp13 <<= CONST_BITS; - - z1 = (INT32) wsptr[2]; - z2 = (INT32) wsptr[4]; - z3 = (INT32) wsptr[6]; - - tmp10 = MULTIPLY(z2 - z3, FIX(0.881747734)); /* c4 */ - tmp12 = MULTIPLY(z1 - z2, FIX(0.314692123)); /* c6 */ - tmp11 = tmp10 + tmp12 + tmp13 - MULTIPLY(z2, FIX(1.841218003)); /* c2+c4-c6 */ - tmp0 = z1 + z3; - z2 -= tmp0; - tmp0 = MULTIPLY(tmp0, FIX(1.274162392)) + tmp13; /* c2 */ - tmp10 += tmp0 - MULTIPLY(z3, FIX(0.077722536)); /* c2-c4-c6 */ - tmp12 += tmp0 - MULTIPLY(z1, FIX(2.470602249)); /* c2+c4+c6 */ - tmp13 += MULTIPLY(z2, FIX(1.414213562)); /* c0 */ - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - - tmp1 = MULTIPLY(z1 + z2, FIX(0.935414347)); /* (c3+c1-c5)/2 */ - tmp2 = MULTIPLY(z1 - z2, FIX(0.170262339)); /* (c3+c5-c1)/2 */ - tmp0 = tmp1 - tmp2; - tmp1 += tmp2; - tmp2 = MULTIPLY(z2 + z3, - FIX(1.378756276)); /* -c1 */ - tmp1 += tmp2; - z2 = MULTIPLY(z1 + z3, FIX(0.613604268)); /* c5 */ - tmp0 += z2; - tmp2 += z2 + MULTIPLY(z3, FIX(1.870828693)); /* c3+c1-c5 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp11 + tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp11 - tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp12 + tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp12 - tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 7; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a reduced-size 6x6 output block. - * - * Optimized algorithm with 3 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/12). - */ - -GLOBAL(void) -jpeg_idct_6x6 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp1, tmp2, tmp10, tmp11, tmp12; - INT32 z1, z2, z3; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[6*6]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 6; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp0 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp0 += ONE << (CONST_BITS-PASS1_BITS-1); - tmp2 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - tmp10 = MULTIPLY(tmp2, FIX(0.707106781)); /* c4 */ - tmp1 = tmp0 + tmp10; - tmp11 = RIGHT_SHIFT(tmp0 - tmp10 - tmp10, CONST_BITS-PASS1_BITS); - tmp10 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - tmp0 = MULTIPLY(tmp10, FIX(1.224744871)); /* c2 */ - tmp10 = tmp1 + tmp0; - tmp12 = tmp1 - tmp0; - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - tmp1 = MULTIPLY(z1 + z3, FIX(0.366025404)); /* c5 */ - tmp0 = tmp1 + ((z1 + z2) << CONST_BITS); - tmp2 = tmp1 + ((z3 - z2) << CONST_BITS); - tmp1 = (z1 - z2 - z3) << PASS1_BITS; - - /* Final output stage */ - - wsptr[6*0] = (int) RIGHT_SHIFT(tmp10 + tmp0, CONST_BITS-PASS1_BITS); - wsptr[6*5] = (int) RIGHT_SHIFT(tmp10 - tmp0, CONST_BITS-PASS1_BITS); - wsptr[6*1] = (int) (tmp11 + tmp1); - wsptr[6*4] = (int) (tmp11 - tmp1); - wsptr[6*2] = (int) RIGHT_SHIFT(tmp12 + tmp2, CONST_BITS-PASS1_BITS); - wsptr[6*3] = (int) RIGHT_SHIFT(tmp12 - tmp2, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 6 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 6; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp0 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp0 <<= CONST_BITS; - tmp2 = (INT32) wsptr[4]; - tmp10 = MULTIPLY(tmp2, FIX(0.707106781)); /* c4 */ - tmp1 = tmp0 + tmp10; - tmp11 = tmp0 - tmp10 - tmp10; - tmp10 = (INT32) wsptr[2]; - tmp0 = MULTIPLY(tmp10, FIX(1.224744871)); /* c2 */ - tmp10 = tmp1 + tmp0; - tmp12 = tmp1 - tmp0; - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - tmp1 = MULTIPLY(z1 + z3, FIX(0.366025404)); /* c5 */ - tmp0 = tmp1 + ((z1 + z2) << CONST_BITS); - tmp2 = tmp1 + ((z3 - z2) << CONST_BITS); - tmp1 = (z1 - z2 - z3) << CONST_BITS; - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp11 + tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp11 - tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp12 + tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp12 - tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 6; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a reduced-size 5x5 output block. - * - * Optimized algorithm with 5 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/10). - */ - -GLOBAL(void) -jpeg_idct_5x5 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp1, tmp10, tmp11, tmp12; - INT32 z1, z2, z3; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[5*5]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 5; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp12 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp12 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp12 += ONE << (CONST_BITS-PASS1_BITS-1); - tmp0 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - tmp1 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z1 = MULTIPLY(tmp0 + tmp1, FIX(0.790569415)); /* (c2+c4)/2 */ - z2 = MULTIPLY(tmp0 - tmp1, FIX(0.353553391)); /* (c2-c4)/2 */ - z3 = tmp12 + z2; - tmp10 = z3 + z1; - tmp11 = z3 - z1; - tmp12 -= z2 << 2; - - /* Odd part */ - - z2 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z3 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - - z1 = MULTIPLY(z2 + z3, FIX(0.831253876)); /* c3 */ - tmp0 = z1 + MULTIPLY(z2, FIX(0.513743148)); /* c1-c3 */ - tmp1 = z1 - MULTIPLY(z3, FIX(2.176250899)); /* c1+c3 */ - - /* Final output stage */ - - wsptr[5*0] = (int) RIGHT_SHIFT(tmp10 + tmp0, CONST_BITS-PASS1_BITS); - wsptr[5*4] = (int) RIGHT_SHIFT(tmp10 - tmp0, CONST_BITS-PASS1_BITS); - wsptr[5*1] = (int) RIGHT_SHIFT(tmp11 + tmp1, CONST_BITS-PASS1_BITS); - wsptr[5*3] = (int) RIGHT_SHIFT(tmp11 - tmp1, CONST_BITS-PASS1_BITS); - wsptr[5*2] = (int) RIGHT_SHIFT(tmp12, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 5 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 5; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp12 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp12 <<= CONST_BITS; - tmp0 = (INT32) wsptr[2]; - tmp1 = (INT32) wsptr[4]; - z1 = MULTIPLY(tmp0 + tmp1, FIX(0.790569415)); /* (c2+c4)/2 */ - z2 = MULTIPLY(tmp0 - tmp1, FIX(0.353553391)); /* (c2-c4)/2 */ - z3 = tmp12 + z2; - tmp10 = z3 + z1; - tmp11 = z3 - z1; - tmp12 -= z2 << 2; - - /* Odd part */ - - z2 = (INT32) wsptr[1]; - z3 = (INT32) wsptr[3]; - - z1 = MULTIPLY(z2 + z3, FIX(0.831253876)); /* c3 */ - tmp0 = z1 + MULTIPLY(z2, FIX(0.513743148)); /* c1-c3 */ - tmp1 = z1 - MULTIPLY(z3, FIX(2.176250899)); /* c1+c3 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp11 + tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp11 - tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 5; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a reduced-size 4x4 output block. - * - * Optimized algorithm with 3 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/16) [refers to 8-point IDCT]. - */ - -GLOBAL(void) -jpeg_idct_4x4 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp2, tmp10, tmp12; - INT32 z1, z2, z3; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[4*4]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 4; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp2 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - - tmp10 = (tmp0 + tmp2) << PASS1_BITS; - tmp12 = (tmp0 - tmp2) << PASS1_BITS; - - /* Odd part */ - /* Same rotation as in the even part of the 8x8 LL&M IDCT */ - - z2 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z3 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS-PASS1_BITS-1); - tmp0 = RIGHT_SHIFT(z1 + MULTIPLY(z2, FIX_0_765366865), /* c2-c6 */ - CONST_BITS-PASS1_BITS); - tmp2 = RIGHT_SHIFT(z1 - MULTIPLY(z3, FIX_1_847759065), /* c2+c6 */ - CONST_BITS-PASS1_BITS); - - /* Final output stage */ - - wsptr[4*0] = (int) (tmp10 + tmp0); - wsptr[4*3] = (int) (tmp10 - tmp0); - wsptr[4*1] = (int) (tmp12 + tmp2); - wsptr[4*2] = (int) (tmp12 - tmp2); - } - - /* Pass 2: process 4 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 4; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp0 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp2 = (INT32) wsptr[2]; - - tmp10 = (tmp0 + tmp2) << CONST_BITS; - tmp12 = (tmp0 - tmp2) << CONST_BITS; - - /* Odd part */ - /* Same rotation as in the even part of the 8x8 LL&M IDCT */ - - z2 = (INT32) wsptr[1]; - z3 = (INT32) wsptr[3]; - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - tmp0 = z1 + MULTIPLY(z2, FIX_0_765366865); /* c2-c6 */ - tmp2 = z1 - MULTIPLY(z3, FIX_1_847759065); /* c2+c6 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp12 + tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp12 - tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 4; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a reduced-size 3x3 output block. - * - * Optimized algorithm with 2 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/6). - */ - -GLOBAL(void) -jpeg_idct_3x3 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp2, tmp10, tmp12; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[3*3]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 3; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp0 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp0 += ONE << (CONST_BITS-PASS1_BITS-1); - tmp2 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - tmp12 = MULTIPLY(tmp2, FIX(0.707106781)); /* c2 */ - tmp10 = tmp0 + tmp12; - tmp2 = tmp0 - tmp12 - tmp12; - - /* Odd part */ - - tmp12 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - tmp0 = MULTIPLY(tmp12, FIX(1.224744871)); /* c1 */ - - /* Final output stage */ - - wsptr[3*0] = (int) RIGHT_SHIFT(tmp10 + tmp0, CONST_BITS-PASS1_BITS); - wsptr[3*2] = (int) RIGHT_SHIFT(tmp10 - tmp0, CONST_BITS-PASS1_BITS); - wsptr[3*1] = (int) RIGHT_SHIFT(tmp2, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 3 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 3; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp0 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp0 <<= CONST_BITS; - tmp2 = (INT32) wsptr[2]; - tmp12 = MULTIPLY(tmp2, FIX(0.707106781)); /* c2 */ - tmp10 = tmp0 + tmp12; - tmp2 = tmp0 - tmp12 - tmp12; - - /* Odd part */ - - tmp12 = (INT32) wsptr[1]; - tmp0 = MULTIPLY(tmp12, FIX(1.224744871)); /* c1 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 3; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a reduced-size 2x2 output block. - * - * Multiplication-less algorithm. - */ - -GLOBAL(void) -jpeg_idct_2x2 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - DCTELEM tmp0, tmp1, tmp2, tmp3, tmp4, tmp5; - ISLOW_MULT_TYPE * quantptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - ISHIFT_TEMPS - - /* Pass 1: process columns from input. */ - - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - - /* Column 0 */ - tmp4 = DEQUANTIZE(coef_block[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp5 = DEQUANTIZE(coef_block[DCTSIZE*1], quantptr[DCTSIZE*1]); - /* Add range center and fudge factor for final descale and range-limit. */ - tmp4 += (((DCTELEM) RANGE_CENTER) << 3) + (1 << 2); - - tmp0 = tmp4 + tmp5; - tmp2 = tmp4 - tmp5; - - /* Column 1 */ - tmp4 = DEQUANTIZE(coef_block[DCTSIZE*0+1], quantptr[DCTSIZE*0+1]); - tmp5 = DEQUANTIZE(coef_block[DCTSIZE*1+1], quantptr[DCTSIZE*1+1]); - - tmp1 = tmp4 + tmp5; - tmp3 = tmp4 - tmp5; - - /* Pass 2: process 2 rows, store into output array. */ - - /* Row 0 */ - outptr = output_buf[0] + output_col; - - outptr[0] = range_limit[(int) IRIGHT_SHIFT(tmp0 + tmp1, 3) & RANGE_MASK]; - outptr[1] = range_limit[(int) IRIGHT_SHIFT(tmp0 - tmp1, 3) & RANGE_MASK]; - - /* Row 1 */ - outptr = output_buf[1] + output_col; - - outptr[0] = range_limit[(int) IRIGHT_SHIFT(tmp2 + tmp3, 3) & RANGE_MASK]; - outptr[1] = range_limit[(int) IRIGHT_SHIFT(tmp2 - tmp3, 3) & RANGE_MASK]; -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a reduced-size 1x1 output block. - * - * We hardly need an inverse DCT routine for this: just take the - * average pixel value, which is one-eighth of the DC coefficient. - */ - -GLOBAL(void) -jpeg_idct_1x1 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - DCTELEM dcval; - ISLOW_MULT_TYPE * quantptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - ISHIFT_TEMPS - - /* 1x1 is trivial: just take the DC coefficient divided by 8. */ - - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - - dcval = DEQUANTIZE(coef_block[0], quantptr[0]); - /* Add range center and fudge factor for descale and range-limit. */ - dcval += (((DCTELEM) RANGE_CENTER) << 3) + (1 << 2); - - output_buf[0][output_col] = - range_limit[(int) IRIGHT_SHIFT(dcval, 3) & RANGE_MASK]; -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 9x9 output block. - * - * Optimized algorithm with 10 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/18). - */ - -GLOBAL(void) -jpeg_idct_9x9 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp10, tmp11, tmp12, tmp13, tmp14; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*9]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp0 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp0 += ONE << (CONST_BITS-PASS1_BITS-1); - - z1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z2 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z3 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - tmp3 = MULTIPLY(z3, FIX(0.707106781)); /* c6 */ - tmp1 = tmp0 + tmp3; - tmp2 = tmp0 - tmp3 - tmp3; - - tmp0 = MULTIPLY(z1 - z2, FIX(0.707106781)); /* c6 */ - tmp11 = tmp2 + tmp0; - tmp14 = tmp2 - tmp0 - tmp0; - - tmp0 = MULTIPLY(z1 + z2, FIX(1.328926049)); /* c2 */ - tmp2 = MULTIPLY(z1, FIX(1.083350441)); /* c4 */ - tmp3 = MULTIPLY(z2, FIX(0.245575608)); /* c8 */ - - tmp10 = tmp1 + tmp0 - tmp3; - tmp12 = tmp1 - tmp0 + tmp2; - tmp13 = tmp1 - tmp2 + tmp3; - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - z4 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - - z2 = MULTIPLY(z2, - FIX(1.224744871)); /* -c3 */ - - tmp2 = MULTIPLY(z1 + z3, FIX(0.909038955)); /* c5 */ - tmp3 = MULTIPLY(z1 + z4, FIX(0.483689525)); /* c7 */ - tmp0 = tmp2 + tmp3 - z2; - tmp1 = MULTIPLY(z3 - z4, FIX(1.392728481)); /* c1 */ - tmp2 += z2 - tmp1; - tmp3 += z2 + tmp1; - tmp1 = MULTIPLY(z1 - z3 - z4, FIX(1.224744871)); /* c3 */ - - /* Final output stage */ - - wsptr[8*0] = (int) RIGHT_SHIFT(tmp10 + tmp0, CONST_BITS-PASS1_BITS); - wsptr[8*8] = (int) RIGHT_SHIFT(tmp10 - tmp0, CONST_BITS-PASS1_BITS); - wsptr[8*1] = (int) RIGHT_SHIFT(tmp11 + tmp1, CONST_BITS-PASS1_BITS); - wsptr[8*7] = (int) RIGHT_SHIFT(tmp11 - tmp1, CONST_BITS-PASS1_BITS); - wsptr[8*2] = (int) RIGHT_SHIFT(tmp12 + tmp2, CONST_BITS-PASS1_BITS); - wsptr[8*6] = (int) RIGHT_SHIFT(tmp12 - tmp2, CONST_BITS-PASS1_BITS); - wsptr[8*3] = (int) RIGHT_SHIFT(tmp13 + tmp3, CONST_BITS-PASS1_BITS); - wsptr[8*5] = (int) RIGHT_SHIFT(tmp13 - tmp3, CONST_BITS-PASS1_BITS); - wsptr[8*4] = (int) RIGHT_SHIFT(tmp14, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 9 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 9; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp0 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp0 <<= CONST_BITS; - - z1 = (INT32) wsptr[2]; - z2 = (INT32) wsptr[4]; - z3 = (INT32) wsptr[6]; - - tmp3 = MULTIPLY(z3, FIX(0.707106781)); /* c6 */ - tmp1 = tmp0 + tmp3; - tmp2 = tmp0 - tmp3 - tmp3; - - tmp0 = MULTIPLY(z1 - z2, FIX(0.707106781)); /* c6 */ - tmp11 = tmp2 + tmp0; - tmp14 = tmp2 - tmp0 - tmp0; - - tmp0 = MULTIPLY(z1 + z2, FIX(1.328926049)); /* c2 */ - tmp2 = MULTIPLY(z1, FIX(1.083350441)); /* c4 */ - tmp3 = MULTIPLY(z2, FIX(0.245575608)); /* c8 */ - - tmp10 = tmp1 + tmp0 - tmp3; - tmp12 = tmp1 - tmp0 + tmp2; - tmp13 = tmp1 - tmp2 + tmp3; - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - z4 = (INT32) wsptr[7]; - - z2 = MULTIPLY(z2, - FIX(1.224744871)); /* -c3 */ - - tmp2 = MULTIPLY(z1 + z3, FIX(0.909038955)); /* c5 */ - tmp3 = MULTIPLY(z1 + z4, FIX(0.483689525)); /* c7 */ - tmp0 = tmp2 + tmp3 - z2; - tmp1 = MULTIPLY(z3 - z4, FIX(1.392728481)); /* c1 */ - tmp2 += z2 - tmp1; - tmp3 += z2 + tmp1; - tmp1 = MULTIPLY(z1 - z3 - z4, FIX(1.224744871)); /* c3 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[8] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp11 + tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp11 - tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp12 + tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp12 - tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp13 + tmp3, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp13 - tmp3, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 8; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 10x10 output block. - * - * Optimized algorithm with 12 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/20). - */ - -GLOBAL(void) -jpeg_idct_10x10 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp10, tmp11, tmp12, tmp13, tmp14; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24; - INT32 z1, z2, z3, z4, z5; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*10]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - z3 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - z3 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - z3 += ONE << (CONST_BITS-PASS1_BITS-1); - z4 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z1 = MULTIPLY(z4, FIX(1.144122806)); /* c4 */ - z2 = MULTIPLY(z4, FIX(0.437016024)); /* c8 */ - tmp10 = z3 + z1; - tmp11 = z3 - z2; - - tmp22 = RIGHT_SHIFT(z3 - ((z1 - z2) << 1), /* c0 = (c4-c8)*2 */ - CONST_BITS-PASS1_BITS); - - z2 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z3 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - z1 = MULTIPLY(z2 + z3, FIX(0.831253876)); /* c6 */ - tmp12 = z1 + MULTIPLY(z2, FIX(0.513743148)); /* c2-c6 */ - tmp13 = z1 - MULTIPLY(z3, FIX(2.176250899)); /* c2+c6 */ - - tmp20 = tmp10 + tmp12; - tmp24 = tmp10 - tmp12; - tmp21 = tmp11 + tmp13; - tmp23 = tmp11 - tmp13; - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - z4 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - - tmp11 = z2 + z4; - tmp13 = z2 - z4; - - tmp12 = MULTIPLY(tmp13, FIX(0.309016994)); /* (c3-c7)/2 */ - z5 = z3 << CONST_BITS; - - z2 = MULTIPLY(tmp11, FIX(0.951056516)); /* (c3+c7)/2 */ - z4 = z5 + tmp12; - - tmp10 = MULTIPLY(z1, FIX(1.396802247)) + z2 + z4; /* c1 */ - tmp14 = MULTIPLY(z1, FIX(0.221231742)) - z2 + z4; /* c9 */ - - z2 = MULTIPLY(tmp11, FIX(0.587785252)); /* (c1-c9)/2 */ - z4 = z5 - tmp12 - (tmp13 << (CONST_BITS - 1)); - - tmp12 = (z1 - tmp13 - z3) << PASS1_BITS; - - tmp11 = MULTIPLY(z1, FIX(1.260073511)) - z2 - z4; /* c3 */ - tmp13 = MULTIPLY(z1, FIX(0.642039522)) - z2 + z4; /* c7 */ - - /* Final output stage */ - - wsptr[8*0] = (int) RIGHT_SHIFT(tmp20 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*9] = (int) RIGHT_SHIFT(tmp20 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*1] = (int) RIGHT_SHIFT(tmp21 + tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*8] = (int) RIGHT_SHIFT(tmp21 - tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*2] = (int) (tmp22 + tmp12); - wsptr[8*7] = (int) (tmp22 - tmp12); - wsptr[8*3] = (int) RIGHT_SHIFT(tmp23 + tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*6] = (int) RIGHT_SHIFT(tmp23 - tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*4] = (int) RIGHT_SHIFT(tmp24 + tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*5] = (int) RIGHT_SHIFT(tmp24 - tmp14, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 10 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 10; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - z3 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - z3 <<= CONST_BITS; - z4 = (INT32) wsptr[4]; - z1 = MULTIPLY(z4, FIX(1.144122806)); /* c4 */ - z2 = MULTIPLY(z4, FIX(0.437016024)); /* c8 */ - tmp10 = z3 + z1; - tmp11 = z3 - z2; - - tmp22 = z3 - ((z1 - z2) << 1); /* c0 = (c4-c8)*2 */ - - z2 = (INT32) wsptr[2]; - z3 = (INT32) wsptr[6]; - - z1 = MULTIPLY(z2 + z3, FIX(0.831253876)); /* c6 */ - tmp12 = z1 + MULTIPLY(z2, FIX(0.513743148)); /* c2-c6 */ - tmp13 = z1 - MULTIPLY(z3, FIX(2.176250899)); /* c2+c6 */ - - tmp20 = tmp10 + tmp12; - tmp24 = tmp10 - tmp12; - tmp21 = tmp11 + tmp13; - tmp23 = tmp11 - tmp13; - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - z3 <<= CONST_BITS; - z4 = (INT32) wsptr[7]; - - tmp11 = z2 + z4; - tmp13 = z2 - z4; - - tmp12 = MULTIPLY(tmp13, FIX(0.309016994)); /* (c3-c7)/2 */ - - z2 = MULTIPLY(tmp11, FIX(0.951056516)); /* (c3+c7)/2 */ - z4 = z3 + tmp12; - - tmp10 = MULTIPLY(z1, FIX(1.396802247)) + z2 + z4; /* c1 */ - tmp14 = MULTIPLY(z1, FIX(0.221231742)) - z2 + z4; /* c9 */ - - z2 = MULTIPLY(tmp11, FIX(0.587785252)); /* (c1-c9)/2 */ - z4 = z3 - tmp12 - (tmp13 << (CONST_BITS - 1)); - - tmp12 = ((z1 - tmp13) << CONST_BITS) - z3; - - tmp11 = MULTIPLY(z1, FIX(1.260073511)) - z2 - z4; /* c3 */ - tmp13 = MULTIPLY(z1, FIX(0.642039522)) - z2 + z4; /* c7 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[9] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[8] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp23 + tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp23 - tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp24 + tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp24 - tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 8; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing an 11x11 output block. - * - * Optimized algorithm with 24 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/22). - */ - -GLOBAL(void) -jpeg_idct_11x11 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp10, tmp11, tmp12, tmp13, tmp14; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24, tmp25; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*11]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp10 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp10 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp10 += ONE << (CONST_BITS-PASS1_BITS-1); - - z1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z2 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z3 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - tmp20 = MULTIPLY(z2 - z3, FIX(2.546640132)); /* c2+c4 */ - tmp23 = MULTIPLY(z2 - z1, FIX(0.430815045)); /* c2-c6 */ - z4 = z1 + z3; - tmp24 = MULTIPLY(z4, - FIX(1.155664402)); /* -(c2-c10) */ - z4 -= z2; - tmp25 = tmp10 + MULTIPLY(z4, FIX(1.356927976)); /* c2 */ - tmp21 = tmp20 + tmp23 + tmp25 - - MULTIPLY(z2, FIX(1.821790775)); /* c2+c4+c10-c6 */ - tmp20 += tmp25 + MULTIPLY(z3, FIX(2.115825087)); /* c4+c6 */ - tmp23 += tmp25 - MULTIPLY(z1, FIX(1.513598477)); /* c6+c8 */ - tmp24 += tmp25; - tmp22 = tmp24 - MULTIPLY(z3, FIX(0.788749120)); /* c8+c10 */ - tmp24 += MULTIPLY(z2, FIX(1.944413522)) - /* c2+c8 */ - MULTIPLY(z1, FIX(1.390975730)); /* c4+c10 */ - tmp25 = tmp10 - MULTIPLY(z4, FIX(1.414213562)); /* c0 */ - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - z4 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - - tmp11 = z1 + z2; - tmp14 = MULTIPLY(tmp11 + z3 + z4, FIX(0.398430003)); /* c9 */ - tmp11 = MULTIPLY(tmp11, FIX(0.887983902)); /* c3-c9 */ - tmp12 = MULTIPLY(z1 + z3, FIX(0.670361295)); /* c5-c9 */ - tmp13 = tmp14 + MULTIPLY(z1 + z4, FIX(0.366151574)); /* c7-c9 */ - tmp10 = tmp11 + tmp12 + tmp13 - - MULTIPLY(z1, FIX(0.923107866)); /* c7+c5+c3-c1-2*c9 */ - z1 = tmp14 - MULTIPLY(z2 + z3, FIX(1.163011579)); /* c7+c9 */ - tmp11 += z1 + MULTIPLY(z2, FIX(2.073276588)); /* c1+c7+3*c9-c3 */ - tmp12 += z1 - MULTIPLY(z3, FIX(1.192193623)); /* c3+c5-c7-c9 */ - z1 = MULTIPLY(z2 + z4, - FIX(1.798248910)); /* -(c1+c9) */ - tmp11 += z1; - tmp13 += z1 + MULTIPLY(z4, FIX(2.102458632)); /* c1+c5+c9-c7 */ - tmp14 += MULTIPLY(z2, - FIX(1.467221301)) + /* -(c5+c9) */ - MULTIPLY(z3, FIX(1.001388905)) - /* c1-c9 */ - MULTIPLY(z4, FIX(1.684843907)); /* c3+c9 */ - - /* Final output stage */ - - wsptr[8*0] = (int) RIGHT_SHIFT(tmp20 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*10] = (int) RIGHT_SHIFT(tmp20 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*1] = (int) RIGHT_SHIFT(tmp21 + tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*9] = (int) RIGHT_SHIFT(tmp21 - tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*2] = (int) RIGHT_SHIFT(tmp22 + tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*8] = (int) RIGHT_SHIFT(tmp22 - tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*3] = (int) RIGHT_SHIFT(tmp23 + tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*7] = (int) RIGHT_SHIFT(tmp23 - tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*4] = (int) RIGHT_SHIFT(tmp24 + tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*6] = (int) RIGHT_SHIFT(tmp24 - tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*5] = (int) RIGHT_SHIFT(tmp25, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 11 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 11; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp10 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp10 <<= CONST_BITS; - - z1 = (INT32) wsptr[2]; - z2 = (INT32) wsptr[4]; - z3 = (INT32) wsptr[6]; - - tmp20 = MULTIPLY(z2 - z3, FIX(2.546640132)); /* c2+c4 */ - tmp23 = MULTIPLY(z2 - z1, FIX(0.430815045)); /* c2-c6 */ - z4 = z1 + z3; - tmp24 = MULTIPLY(z4, - FIX(1.155664402)); /* -(c2-c10) */ - z4 -= z2; - tmp25 = tmp10 + MULTIPLY(z4, FIX(1.356927976)); /* c2 */ - tmp21 = tmp20 + tmp23 + tmp25 - - MULTIPLY(z2, FIX(1.821790775)); /* c2+c4+c10-c6 */ - tmp20 += tmp25 + MULTIPLY(z3, FIX(2.115825087)); /* c4+c6 */ - tmp23 += tmp25 - MULTIPLY(z1, FIX(1.513598477)); /* c6+c8 */ - tmp24 += tmp25; - tmp22 = tmp24 - MULTIPLY(z3, FIX(0.788749120)); /* c8+c10 */ - tmp24 += MULTIPLY(z2, FIX(1.944413522)) - /* c2+c8 */ - MULTIPLY(z1, FIX(1.390975730)); /* c4+c10 */ - tmp25 = tmp10 - MULTIPLY(z4, FIX(1.414213562)); /* c0 */ - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - z4 = (INT32) wsptr[7]; - - tmp11 = z1 + z2; - tmp14 = MULTIPLY(tmp11 + z3 + z4, FIX(0.398430003)); /* c9 */ - tmp11 = MULTIPLY(tmp11, FIX(0.887983902)); /* c3-c9 */ - tmp12 = MULTIPLY(z1 + z3, FIX(0.670361295)); /* c5-c9 */ - tmp13 = tmp14 + MULTIPLY(z1 + z4, FIX(0.366151574)); /* c7-c9 */ - tmp10 = tmp11 + tmp12 + tmp13 - - MULTIPLY(z1, FIX(0.923107866)); /* c7+c5+c3-c1-2*c9 */ - z1 = tmp14 - MULTIPLY(z2 + z3, FIX(1.163011579)); /* c7+c9 */ - tmp11 += z1 + MULTIPLY(z2, FIX(2.073276588)); /* c1+c7+3*c9-c3 */ - tmp12 += z1 - MULTIPLY(z3, FIX(1.192193623)); /* c3+c5-c7-c9 */ - z1 = MULTIPLY(z2 + z4, - FIX(1.798248910)); /* -(c1+c9) */ - tmp11 += z1; - tmp13 += z1 + MULTIPLY(z4, FIX(2.102458632)); /* c1+c5+c9-c7 */ - tmp14 += MULTIPLY(z2, - FIX(1.467221301)) + /* -(c5+c9) */ - MULTIPLY(z3, FIX(1.001388905)) - /* c1-c9 */ - MULTIPLY(z4, FIX(1.684843907)); /* c3+c9 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[10] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[9] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[8] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp23 + tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp23 - tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp24 + tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp24 - tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp25, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 8; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 12x12 output block. - * - * Optimized algorithm with 15 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/24). - */ - -GLOBAL(void) -jpeg_idct_12x12 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24, tmp25; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*12]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - z3 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - z3 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - z3 += ONE << (CONST_BITS-PASS1_BITS-1); - - z4 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z4 = MULTIPLY(z4, FIX(1.224744871)); /* c4 */ - - tmp10 = z3 + z4; - tmp11 = z3 - z4; - - z1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z4 = MULTIPLY(z1, FIX(1.366025404)); /* c2 */ - z1 <<= CONST_BITS; - z2 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - z2 <<= CONST_BITS; - - tmp12 = z1 - z2; - - tmp21 = z3 + tmp12; - tmp24 = z3 - tmp12; - - tmp12 = z4 + z2; - - tmp20 = tmp10 + tmp12; - tmp25 = tmp10 - tmp12; - - tmp12 = z4 - z1 - z2; - - tmp22 = tmp11 + tmp12; - tmp23 = tmp11 - tmp12; - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - z4 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - - tmp11 = MULTIPLY(z2, FIX(1.306562965)); /* c3 */ - tmp14 = MULTIPLY(z2, - FIX_0_541196100); /* -c9 */ - - tmp10 = z1 + z3; - tmp15 = MULTIPLY(tmp10 + z4, FIX(0.860918669)); /* c7 */ - tmp12 = tmp15 + MULTIPLY(tmp10, FIX(0.261052384)); /* c5-c7 */ - tmp10 = tmp12 + tmp11 + MULTIPLY(z1, FIX(0.280143716)); /* c1-c5 */ - tmp13 = MULTIPLY(z3 + z4, - FIX(1.045510580)); /* -(c7+c11) */ - tmp12 += tmp13 + tmp14 - MULTIPLY(z3, FIX(1.478575242)); /* c1+c5-c7-c11 */ - tmp13 += tmp15 - tmp11 + MULTIPLY(z4, FIX(1.586706681)); /* c1+c11 */ - tmp15 += tmp14 - MULTIPLY(z1, FIX(0.676326758)) - /* c7-c11 */ - MULTIPLY(z4, FIX(1.982889723)); /* c5+c7 */ - - z1 -= z4; - z2 -= z3; - z3 = MULTIPLY(z1 + z2, FIX_0_541196100); /* c9 */ - tmp11 = z3 + MULTIPLY(z1, FIX_0_765366865); /* c3-c9 */ - tmp14 = z3 - MULTIPLY(z2, FIX_1_847759065); /* c3+c9 */ - - /* Final output stage */ - - wsptr[8*0] = (int) RIGHT_SHIFT(tmp20 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*11] = (int) RIGHT_SHIFT(tmp20 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*1] = (int) RIGHT_SHIFT(tmp21 + tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*10] = (int) RIGHT_SHIFT(tmp21 - tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*2] = (int) RIGHT_SHIFT(tmp22 + tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*9] = (int) RIGHT_SHIFT(tmp22 - tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*3] = (int) RIGHT_SHIFT(tmp23 + tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*8] = (int) RIGHT_SHIFT(tmp23 - tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*4] = (int) RIGHT_SHIFT(tmp24 + tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*7] = (int) RIGHT_SHIFT(tmp24 - tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*5] = (int) RIGHT_SHIFT(tmp25 + tmp15, CONST_BITS-PASS1_BITS); - wsptr[8*6] = (int) RIGHT_SHIFT(tmp25 - tmp15, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 12 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 12; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - z3 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - z3 <<= CONST_BITS; - - z4 = (INT32) wsptr[4]; - z4 = MULTIPLY(z4, FIX(1.224744871)); /* c4 */ - - tmp10 = z3 + z4; - tmp11 = z3 - z4; - - z1 = (INT32) wsptr[2]; - z4 = MULTIPLY(z1, FIX(1.366025404)); /* c2 */ - z1 <<= CONST_BITS; - z2 = (INT32) wsptr[6]; - z2 <<= CONST_BITS; - - tmp12 = z1 - z2; - - tmp21 = z3 + tmp12; - tmp24 = z3 - tmp12; - - tmp12 = z4 + z2; - - tmp20 = tmp10 + tmp12; - tmp25 = tmp10 - tmp12; - - tmp12 = z4 - z1 - z2; - - tmp22 = tmp11 + tmp12; - tmp23 = tmp11 - tmp12; - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - z4 = (INT32) wsptr[7]; - - tmp11 = MULTIPLY(z2, FIX(1.306562965)); /* c3 */ - tmp14 = MULTIPLY(z2, - FIX_0_541196100); /* -c9 */ - - tmp10 = z1 + z3; - tmp15 = MULTIPLY(tmp10 + z4, FIX(0.860918669)); /* c7 */ - tmp12 = tmp15 + MULTIPLY(tmp10, FIX(0.261052384)); /* c5-c7 */ - tmp10 = tmp12 + tmp11 + MULTIPLY(z1, FIX(0.280143716)); /* c1-c5 */ - tmp13 = MULTIPLY(z3 + z4, - FIX(1.045510580)); /* -(c7+c11) */ - tmp12 += tmp13 + tmp14 - MULTIPLY(z3, FIX(1.478575242)); /* c1+c5-c7-c11 */ - tmp13 += tmp15 - tmp11 + MULTIPLY(z4, FIX(1.586706681)); /* c1+c11 */ - tmp15 += tmp14 - MULTIPLY(z1, FIX(0.676326758)) - /* c7-c11 */ - MULTIPLY(z4, FIX(1.982889723)); /* c5+c7 */ - - z1 -= z4; - z2 -= z3; - z3 = MULTIPLY(z1 + z2, FIX_0_541196100); /* c9 */ - tmp11 = z3 + MULTIPLY(z1, FIX_0_765366865); /* c3-c9 */ - tmp14 = z3 - MULTIPLY(z2, FIX_1_847759065); /* c3+c9 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[11] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[10] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[9] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp23 + tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[8] = range_limit[(int) RIGHT_SHIFT(tmp23 - tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp24 + tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp24 - tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp25 + tmp15, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp25 - tmp15, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 8; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 13x13 output block. - * - * Optimized algorithm with 29 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/26). - */ - -GLOBAL(void) -jpeg_idct_13x13 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24, tmp25, tmp26; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*13]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - z1 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS-PASS1_BITS-1); - - z2 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z3 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z4 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - tmp10 = z3 + z4; - tmp11 = z3 - z4; - - tmp12 = MULTIPLY(tmp10, FIX(1.155388986)); /* (c4+c6)/2 */ - tmp13 = MULTIPLY(tmp11, FIX(0.096834934)) + z1; /* (c4-c6)/2 */ - - tmp20 = MULTIPLY(z2, FIX(1.373119086)) + tmp12 + tmp13; /* c2 */ - tmp22 = MULTIPLY(z2, FIX(0.501487041)) - tmp12 + tmp13; /* c10 */ - - tmp12 = MULTIPLY(tmp10, FIX(0.316450131)); /* (c8-c12)/2 */ - tmp13 = MULTIPLY(tmp11, FIX(0.486914739)) + z1; /* (c8+c12)/2 */ - - tmp21 = MULTIPLY(z2, FIX(1.058554052)) - tmp12 + tmp13; /* c6 */ - tmp25 = MULTIPLY(z2, - FIX(1.252223920)) + tmp12 + tmp13; /* c4 */ - - tmp12 = MULTIPLY(tmp10, FIX(0.435816023)); /* (c2-c10)/2 */ - tmp13 = MULTIPLY(tmp11, FIX(0.937303064)) - z1; /* (c2+c10)/2 */ - - tmp23 = MULTIPLY(z2, - FIX(0.170464608)) - tmp12 - tmp13; /* c12 */ - tmp24 = MULTIPLY(z2, - FIX(0.803364869)) + tmp12 - tmp13; /* c8 */ - - tmp26 = MULTIPLY(tmp11 - z2, FIX(1.414213562)) + z1; /* c0 */ - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - z4 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - - tmp11 = MULTIPLY(z1 + z2, FIX(1.322312651)); /* c3 */ - tmp12 = MULTIPLY(z1 + z3, FIX(1.163874945)); /* c5 */ - tmp15 = z1 + z4; - tmp13 = MULTIPLY(tmp15, FIX(0.937797057)); /* c7 */ - tmp10 = tmp11 + tmp12 + tmp13 - - MULTIPLY(z1, FIX(2.020082300)); /* c7+c5+c3-c1 */ - tmp14 = MULTIPLY(z2 + z3, - FIX(0.338443458)); /* -c11 */ - tmp11 += tmp14 + MULTIPLY(z2, FIX(0.837223564)); /* c5+c9+c11-c3 */ - tmp12 += tmp14 - MULTIPLY(z3, FIX(1.572116027)); /* c1+c5-c9-c11 */ - tmp14 = MULTIPLY(z2 + z4, - FIX(1.163874945)); /* -c5 */ - tmp11 += tmp14; - tmp13 += tmp14 + MULTIPLY(z4, FIX(2.205608352)); /* c3+c5+c9-c7 */ - tmp14 = MULTIPLY(z3 + z4, - FIX(0.657217813)); /* -c9 */ - tmp12 += tmp14; - tmp13 += tmp14; - tmp15 = MULTIPLY(tmp15, FIX(0.338443458)); /* c11 */ - tmp14 = tmp15 + MULTIPLY(z1, FIX(0.318774355)) - /* c9-c11 */ - MULTIPLY(z2, FIX(0.466105296)); /* c1-c7 */ - z1 = MULTIPLY(z3 - z2, FIX(0.937797057)); /* c7 */ - tmp14 += z1; - tmp15 += z1 + MULTIPLY(z3, FIX(0.384515595)) - /* c3-c7 */ - MULTIPLY(z4, FIX(1.742345811)); /* c1+c11 */ - - /* Final output stage */ - - wsptr[8*0] = (int) RIGHT_SHIFT(tmp20 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*12] = (int) RIGHT_SHIFT(tmp20 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*1] = (int) RIGHT_SHIFT(tmp21 + tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*11] = (int) RIGHT_SHIFT(tmp21 - tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*2] = (int) RIGHT_SHIFT(tmp22 + tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*10] = (int) RIGHT_SHIFT(tmp22 - tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*3] = (int) RIGHT_SHIFT(tmp23 + tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*9] = (int) RIGHT_SHIFT(tmp23 - tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*4] = (int) RIGHT_SHIFT(tmp24 + tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*8] = (int) RIGHT_SHIFT(tmp24 - tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*5] = (int) RIGHT_SHIFT(tmp25 + tmp15, CONST_BITS-PASS1_BITS); - wsptr[8*7] = (int) RIGHT_SHIFT(tmp25 - tmp15, CONST_BITS-PASS1_BITS); - wsptr[8*6] = (int) RIGHT_SHIFT(tmp26, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 13 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 13; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - z1 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - z1 <<= CONST_BITS; - - z2 = (INT32) wsptr[2]; - z3 = (INT32) wsptr[4]; - z4 = (INT32) wsptr[6]; - - tmp10 = z3 + z4; - tmp11 = z3 - z4; - - tmp12 = MULTIPLY(tmp10, FIX(1.155388986)); /* (c4+c6)/2 */ - tmp13 = MULTIPLY(tmp11, FIX(0.096834934)) + z1; /* (c4-c6)/2 */ - - tmp20 = MULTIPLY(z2, FIX(1.373119086)) + tmp12 + tmp13; /* c2 */ - tmp22 = MULTIPLY(z2, FIX(0.501487041)) - tmp12 + tmp13; /* c10 */ - - tmp12 = MULTIPLY(tmp10, FIX(0.316450131)); /* (c8-c12)/2 */ - tmp13 = MULTIPLY(tmp11, FIX(0.486914739)) + z1; /* (c8+c12)/2 */ - - tmp21 = MULTIPLY(z2, FIX(1.058554052)) - tmp12 + tmp13; /* c6 */ - tmp25 = MULTIPLY(z2, - FIX(1.252223920)) + tmp12 + tmp13; /* c4 */ - - tmp12 = MULTIPLY(tmp10, FIX(0.435816023)); /* (c2-c10)/2 */ - tmp13 = MULTIPLY(tmp11, FIX(0.937303064)) - z1; /* (c2+c10)/2 */ - - tmp23 = MULTIPLY(z2, - FIX(0.170464608)) - tmp12 - tmp13; /* c12 */ - tmp24 = MULTIPLY(z2, - FIX(0.803364869)) + tmp12 - tmp13; /* c8 */ - - tmp26 = MULTIPLY(tmp11 - z2, FIX(1.414213562)) + z1; /* c0 */ - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - z4 = (INT32) wsptr[7]; - - tmp11 = MULTIPLY(z1 + z2, FIX(1.322312651)); /* c3 */ - tmp12 = MULTIPLY(z1 + z3, FIX(1.163874945)); /* c5 */ - tmp15 = z1 + z4; - tmp13 = MULTIPLY(tmp15, FIX(0.937797057)); /* c7 */ - tmp10 = tmp11 + tmp12 + tmp13 - - MULTIPLY(z1, FIX(2.020082300)); /* c7+c5+c3-c1 */ - tmp14 = MULTIPLY(z2 + z3, - FIX(0.338443458)); /* -c11 */ - tmp11 += tmp14 + MULTIPLY(z2, FIX(0.837223564)); /* c5+c9+c11-c3 */ - tmp12 += tmp14 - MULTIPLY(z3, FIX(1.572116027)); /* c1+c5-c9-c11 */ - tmp14 = MULTIPLY(z2 + z4, - FIX(1.163874945)); /* -c5 */ - tmp11 += tmp14; - tmp13 += tmp14 + MULTIPLY(z4, FIX(2.205608352)); /* c3+c5+c9-c7 */ - tmp14 = MULTIPLY(z3 + z4, - FIX(0.657217813)); /* -c9 */ - tmp12 += tmp14; - tmp13 += tmp14; - tmp15 = MULTIPLY(tmp15, FIX(0.338443458)); /* c11 */ - tmp14 = tmp15 + MULTIPLY(z1, FIX(0.318774355)) - /* c9-c11 */ - MULTIPLY(z2, FIX(0.466105296)); /* c1-c7 */ - z1 = MULTIPLY(z3 - z2, FIX(0.937797057)); /* c7 */ - tmp14 += z1; - tmp15 += z1 + MULTIPLY(z3, FIX(0.384515595)) - /* c3-c7 */ - MULTIPLY(z4, FIX(1.742345811)); /* c1+c11 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[12] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[11] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[10] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp23 + tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[9] = range_limit[(int) RIGHT_SHIFT(tmp23 - tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp24 + tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[8] = range_limit[(int) RIGHT_SHIFT(tmp24 - tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp25 + tmp15, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp25 - tmp15, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp26, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 8; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 14x14 output block. - * - * Optimized algorithm with 20 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/28). - */ - -GLOBAL(void) -jpeg_idct_14x14 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15, tmp16; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24, tmp25, tmp26; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*14]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - z1 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS-PASS1_BITS-1); - z4 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z2 = MULTIPLY(z4, FIX(1.274162392)); /* c4 */ - z3 = MULTIPLY(z4, FIX(0.314692123)); /* c12 */ - z4 = MULTIPLY(z4, FIX(0.881747734)); /* c8 */ - - tmp10 = z1 + z2; - tmp11 = z1 + z3; - tmp12 = z1 - z4; - - tmp23 = RIGHT_SHIFT(z1 - ((z2 + z3 - z4) << 1), /* c0 = (c4+c12-c8)*2 */ - CONST_BITS-PASS1_BITS); - - z1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z2 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - z3 = MULTIPLY(z1 + z2, FIX(1.105676686)); /* c6 */ - - tmp13 = z3 + MULTIPLY(z1, FIX(0.273079590)); /* c2-c6 */ - tmp14 = z3 - MULTIPLY(z2, FIX(1.719280954)); /* c6+c10 */ - tmp15 = MULTIPLY(z1, FIX(0.613604268)) - /* c10 */ - MULTIPLY(z2, FIX(1.378756276)); /* c2 */ - - tmp20 = tmp10 + tmp13; - tmp26 = tmp10 - tmp13; - tmp21 = tmp11 + tmp14; - tmp25 = tmp11 - tmp14; - tmp22 = tmp12 + tmp15; - tmp24 = tmp12 - tmp15; - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - z4 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - tmp13 = z4 << CONST_BITS; - - tmp14 = z1 + z3; - tmp11 = MULTIPLY(z1 + z2, FIX(1.334852607)); /* c3 */ - tmp12 = MULTIPLY(tmp14, FIX(1.197448846)); /* c5 */ - tmp10 = tmp11 + tmp12 + tmp13 - MULTIPLY(z1, FIX(1.126980169)); /* c3+c5-c1 */ - tmp14 = MULTIPLY(tmp14, FIX(0.752406978)); /* c9 */ - tmp16 = tmp14 - MULTIPLY(z1, FIX(1.061150426)); /* c9+c11-c13 */ - z1 -= z2; - tmp15 = MULTIPLY(z1, FIX(0.467085129)) - tmp13; /* c11 */ - tmp16 += tmp15; - z1 += z4; - z4 = MULTIPLY(z2 + z3, - FIX(0.158341681)) - tmp13; /* -c13 */ - tmp11 += z4 - MULTIPLY(z2, FIX(0.424103948)); /* c3-c9-c13 */ - tmp12 += z4 - MULTIPLY(z3, FIX(2.373959773)); /* c3+c5-c13 */ - z4 = MULTIPLY(z3 - z2, FIX(1.405321284)); /* c1 */ - tmp14 += z4 + tmp13 - MULTIPLY(z3, FIX(1.6906431334)); /* c1+c9-c11 */ - tmp15 += z4 + MULTIPLY(z2, FIX(0.674957567)); /* c1+c11-c5 */ - - tmp13 = (z1 - z3) << PASS1_BITS; - - /* Final output stage */ - - wsptr[8*0] = (int) RIGHT_SHIFT(tmp20 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*13] = (int) RIGHT_SHIFT(tmp20 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*1] = (int) RIGHT_SHIFT(tmp21 + tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*12] = (int) RIGHT_SHIFT(tmp21 - tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*2] = (int) RIGHT_SHIFT(tmp22 + tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*11] = (int) RIGHT_SHIFT(tmp22 - tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*3] = (int) (tmp23 + tmp13); - wsptr[8*10] = (int) (tmp23 - tmp13); - wsptr[8*4] = (int) RIGHT_SHIFT(tmp24 + tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*9] = (int) RIGHT_SHIFT(tmp24 - tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*5] = (int) RIGHT_SHIFT(tmp25 + tmp15, CONST_BITS-PASS1_BITS); - wsptr[8*8] = (int) RIGHT_SHIFT(tmp25 - tmp15, CONST_BITS-PASS1_BITS); - wsptr[8*6] = (int) RIGHT_SHIFT(tmp26 + tmp16, CONST_BITS-PASS1_BITS); - wsptr[8*7] = (int) RIGHT_SHIFT(tmp26 - tmp16, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 14 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 14; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - z1 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - z1 <<= CONST_BITS; - z4 = (INT32) wsptr[4]; - z2 = MULTIPLY(z4, FIX(1.274162392)); /* c4 */ - z3 = MULTIPLY(z4, FIX(0.314692123)); /* c12 */ - z4 = MULTIPLY(z4, FIX(0.881747734)); /* c8 */ - - tmp10 = z1 + z2; - tmp11 = z1 + z3; - tmp12 = z1 - z4; - - tmp23 = z1 - ((z2 + z3 - z4) << 1); /* c0 = (c4+c12-c8)*2 */ - - z1 = (INT32) wsptr[2]; - z2 = (INT32) wsptr[6]; - - z3 = MULTIPLY(z1 + z2, FIX(1.105676686)); /* c6 */ - - tmp13 = z3 + MULTIPLY(z1, FIX(0.273079590)); /* c2-c6 */ - tmp14 = z3 - MULTIPLY(z2, FIX(1.719280954)); /* c6+c10 */ - tmp15 = MULTIPLY(z1, FIX(0.613604268)) - /* c10 */ - MULTIPLY(z2, FIX(1.378756276)); /* c2 */ - - tmp20 = tmp10 + tmp13; - tmp26 = tmp10 - tmp13; - tmp21 = tmp11 + tmp14; - tmp25 = tmp11 - tmp14; - tmp22 = tmp12 + tmp15; - tmp24 = tmp12 - tmp15; - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - z4 = (INT32) wsptr[7]; - z4 <<= CONST_BITS; - - tmp14 = z1 + z3; - tmp11 = MULTIPLY(z1 + z2, FIX(1.334852607)); /* c3 */ - tmp12 = MULTIPLY(tmp14, FIX(1.197448846)); /* c5 */ - tmp10 = tmp11 + tmp12 + z4 - MULTIPLY(z1, FIX(1.126980169)); /* c3+c5-c1 */ - tmp14 = MULTIPLY(tmp14, FIX(0.752406978)); /* c9 */ - tmp16 = tmp14 - MULTIPLY(z1, FIX(1.061150426)); /* c9+c11-c13 */ - z1 -= z2; - tmp15 = MULTIPLY(z1, FIX(0.467085129)) - z4; /* c11 */ - tmp16 += tmp15; - tmp13 = MULTIPLY(z2 + z3, - FIX(0.158341681)) - z4; /* -c13 */ - tmp11 += tmp13 - MULTIPLY(z2, FIX(0.424103948)); /* c3-c9-c13 */ - tmp12 += tmp13 - MULTIPLY(z3, FIX(2.373959773)); /* c3+c5-c13 */ - tmp13 = MULTIPLY(z3 - z2, FIX(1.405321284)); /* c1 */ - tmp14 += tmp13 + z4 - MULTIPLY(z3, FIX(1.6906431334)); /* c1+c9-c11 */ - tmp15 += tmp13 + MULTIPLY(z2, FIX(0.674957567)); /* c1+c11-c5 */ - - tmp13 = ((z1 - z3) << CONST_BITS) + z4; - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[13] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[12] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[11] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp23 + tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[10] = range_limit[(int) RIGHT_SHIFT(tmp23 - tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp24 + tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[9] = range_limit[(int) RIGHT_SHIFT(tmp24 - tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp25 + tmp15, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[8] = range_limit[(int) RIGHT_SHIFT(tmp25 - tmp15, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp26 + tmp16, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp26 - tmp16, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 8; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 15x15 output block. - * - * Optimized algorithm with 22 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/30). - */ - -GLOBAL(void) -jpeg_idct_15x15 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15, tmp16; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24, tmp25, tmp26, tmp27; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*15]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - z1 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS-PASS1_BITS-1); - - z2 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z3 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z4 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - tmp10 = MULTIPLY(z4, FIX(0.437016024)); /* c12 */ - tmp11 = MULTIPLY(z4, FIX(1.144122806)); /* c6 */ - - tmp12 = z1 - tmp10; - tmp13 = z1 + tmp11; - z1 -= (tmp11 - tmp10) << 1; /* c0 = (c6-c12)*2 */ - - z4 = z2 - z3; - z3 += z2; - tmp10 = MULTIPLY(z3, FIX(1.337628990)); /* (c2+c4)/2 */ - tmp11 = MULTIPLY(z4, FIX(0.045680613)); /* (c2-c4)/2 */ - z2 = MULTIPLY(z2, FIX(1.439773946)); /* c4+c14 */ - - tmp20 = tmp13 + tmp10 + tmp11; - tmp23 = tmp12 - tmp10 + tmp11 + z2; - - tmp10 = MULTIPLY(z3, FIX(0.547059574)); /* (c8+c14)/2 */ - tmp11 = MULTIPLY(z4, FIX(0.399234004)); /* (c8-c14)/2 */ - - tmp25 = tmp13 - tmp10 - tmp11; - tmp26 = tmp12 + tmp10 - tmp11 - z2; - - tmp10 = MULTIPLY(z3, FIX(0.790569415)); /* (c6+c12)/2 */ - tmp11 = MULTIPLY(z4, FIX(0.353553391)); /* (c6-c12)/2 */ - - tmp21 = tmp12 + tmp10 + tmp11; - tmp24 = tmp13 - tmp10 + tmp11; - tmp11 += tmp11; - tmp22 = z1 + tmp11; /* c10 = c6-c12 */ - tmp27 = z1 - tmp11 - tmp11; /* c0 = (c6-c12)*2 */ - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z4 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - z3 = MULTIPLY(z4, FIX(1.224744871)); /* c5 */ - z4 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - - tmp13 = z2 - z4; - tmp15 = MULTIPLY(z1 + tmp13, FIX(0.831253876)); /* c9 */ - tmp11 = tmp15 + MULTIPLY(z1, FIX(0.513743148)); /* c3-c9 */ - tmp14 = tmp15 - MULTIPLY(tmp13, FIX(2.176250899)); /* c3+c9 */ - - tmp13 = MULTIPLY(z2, - FIX(0.831253876)); /* -c9 */ - tmp15 = MULTIPLY(z2, - FIX(1.344997024)); /* -c3 */ - z2 = z1 - z4; - tmp12 = z3 + MULTIPLY(z2, FIX(1.406466353)); /* c1 */ - - tmp10 = tmp12 + MULTIPLY(z4, FIX(2.457431844)) - tmp15; /* c1+c7 */ - tmp16 = tmp12 - MULTIPLY(z1, FIX(1.112434820)) + tmp13; /* c1-c13 */ - tmp12 = MULTIPLY(z2, FIX(1.224744871)) - z3; /* c5 */ - z2 = MULTIPLY(z1 + z4, FIX(0.575212477)); /* c11 */ - tmp13 += z2 + MULTIPLY(z1, FIX(0.475753014)) - z3; /* c7-c11 */ - tmp15 += z2 - MULTIPLY(z4, FIX(0.869244010)) + z3; /* c11+c13 */ - - /* Final output stage */ - - wsptr[8*0] = (int) RIGHT_SHIFT(tmp20 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*14] = (int) RIGHT_SHIFT(tmp20 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*1] = (int) RIGHT_SHIFT(tmp21 + tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*13] = (int) RIGHT_SHIFT(tmp21 - tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*2] = (int) RIGHT_SHIFT(tmp22 + tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*12] = (int) RIGHT_SHIFT(tmp22 - tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*3] = (int) RIGHT_SHIFT(tmp23 + tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*11] = (int) RIGHT_SHIFT(tmp23 - tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*4] = (int) RIGHT_SHIFT(tmp24 + tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*10] = (int) RIGHT_SHIFT(tmp24 - tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*5] = (int) RIGHT_SHIFT(tmp25 + tmp15, CONST_BITS-PASS1_BITS); - wsptr[8*9] = (int) RIGHT_SHIFT(tmp25 - tmp15, CONST_BITS-PASS1_BITS); - wsptr[8*6] = (int) RIGHT_SHIFT(tmp26 + tmp16, CONST_BITS-PASS1_BITS); - wsptr[8*8] = (int) RIGHT_SHIFT(tmp26 - tmp16, CONST_BITS-PASS1_BITS); - wsptr[8*7] = (int) RIGHT_SHIFT(tmp27, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 15 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 15; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - z1 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - z1 <<= CONST_BITS; - - z2 = (INT32) wsptr[2]; - z3 = (INT32) wsptr[4]; - z4 = (INT32) wsptr[6]; - - tmp10 = MULTIPLY(z4, FIX(0.437016024)); /* c12 */ - tmp11 = MULTIPLY(z4, FIX(1.144122806)); /* c6 */ - - tmp12 = z1 - tmp10; - tmp13 = z1 + tmp11; - z1 -= (tmp11 - tmp10) << 1; /* c0 = (c6-c12)*2 */ - - z4 = z2 - z3; - z3 += z2; - tmp10 = MULTIPLY(z3, FIX(1.337628990)); /* (c2+c4)/2 */ - tmp11 = MULTIPLY(z4, FIX(0.045680613)); /* (c2-c4)/2 */ - z2 = MULTIPLY(z2, FIX(1.439773946)); /* c4+c14 */ - - tmp20 = tmp13 + tmp10 + tmp11; - tmp23 = tmp12 - tmp10 + tmp11 + z2; - - tmp10 = MULTIPLY(z3, FIX(0.547059574)); /* (c8+c14)/2 */ - tmp11 = MULTIPLY(z4, FIX(0.399234004)); /* (c8-c14)/2 */ - - tmp25 = tmp13 - tmp10 - tmp11; - tmp26 = tmp12 + tmp10 - tmp11 - z2; - - tmp10 = MULTIPLY(z3, FIX(0.790569415)); /* (c6+c12)/2 */ - tmp11 = MULTIPLY(z4, FIX(0.353553391)); /* (c6-c12)/2 */ - - tmp21 = tmp12 + tmp10 + tmp11; - tmp24 = tmp13 - tmp10 + tmp11; - tmp11 += tmp11; - tmp22 = z1 + tmp11; /* c10 = c6-c12 */ - tmp27 = z1 - tmp11 - tmp11; /* c0 = (c6-c12)*2 */ - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z4 = (INT32) wsptr[5]; - z3 = MULTIPLY(z4, FIX(1.224744871)); /* c5 */ - z4 = (INT32) wsptr[7]; - - tmp13 = z2 - z4; - tmp15 = MULTIPLY(z1 + tmp13, FIX(0.831253876)); /* c9 */ - tmp11 = tmp15 + MULTIPLY(z1, FIX(0.513743148)); /* c3-c9 */ - tmp14 = tmp15 - MULTIPLY(tmp13, FIX(2.176250899)); /* c3+c9 */ - - tmp13 = MULTIPLY(z2, - FIX(0.831253876)); /* -c9 */ - tmp15 = MULTIPLY(z2, - FIX(1.344997024)); /* -c3 */ - z2 = z1 - z4; - tmp12 = z3 + MULTIPLY(z2, FIX(1.406466353)); /* c1 */ - - tmp10 = tmp12 + MULTIPLY(z4, FIX(2.457431844)) - tmp15; /* c1+c7 */ - tmp16 = tmp12 - MULTIPLY(z1, FIX(1.112434820)) + tmp13; /* c1-c13 */ - tmp12 = MULTIPLY(z2, FIX(1.224744871)) - z3; /* c5 */ - z2 = MULTIPLY(z1 + z4, FIX(0.575212477)); /* c11 */ - tmp13 += z2 + MULTIPLY(z1, FIX(0.475753014)) - z3; /* c7-c11 */ - tmp15 += z2 - MULTIPLY(z4, FIX(0.869244010)) + z3; /* c11+c13 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[14] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[13] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[12] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp23 + tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[11] = range_limit[(int) RIGHT_SHIFT(tmp23 - tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp24 + tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[10] = range_limit[(int) RIGHT_SHIFT(tmp24 - tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp25 + tmp15, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[9] = range_limit[(int) RIGHT_SHIFT(tmp25 - tmp15, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp26 + tmp16, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[8] = range_limit[(int) RIGHT_SHIFT(tmp26 - tmp16, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp27, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 8; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 16x16 output block. - * - * Optimized algorithm with 28 multiplications in the 1-D kernel. - * cK represents sqrt(2) * cos(K*pi/32). - */ - -GLOBAL(void) -jpeg_idct_16x16 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp10, tmp11, tmp12, tmp13; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24, tmp25, tmp26, tmp27; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*16]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp0 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp0 += ONE << (CONST_BITS-PASS1_BITS-1); - - z1 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - tmp1 = MULTIPLY(z1, FIX(1.306562965)); /* c4[16] = c2[8] */ - tmp2 = MULTIPLY(z1, FIX_0_541196100); /* c12[16] = c6[8] */ - - tmp10 = tmp0 + tmp1; - tmp11 = tmp0 - tmp1; - tmp12 = tmp0 + tmp2; - tmp13 = tmp0 - tmp2; - - z1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z2 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - z3 = z1 - z2; - z4 = MULTIPLY(z3, FIX(0.275899379)); /* c14[16] = c7[8] */ - z3 = MULTIPLY(z3, FIX(1.387039845)); /* c2[16] = c1[8] */ - - tmp0 = z3 + MULTIPLY(z2, FIX_2_562915447); /* (c6+c2)[16] = (c3+c1)[8] */ - tmp1 = z4 + MULTIPLY(z1, FIX_0_899976223); /* (c6-c14)[16] = (c3-c7)[8] */ - tmp2 = z3 - MULTIPLY(z1, FIX(0.601344887)); /* (c2-c10)[16] = (c1-c5)[8] */ - tmp3 = z4 - MULTIPLY(z2, FIX(0.509795579)); /* (c10-c14)[16] = (c5-c7)[8] */ - - tmp20 = tmp10 + tmp0; - tmp27 = tmp10 - tmp0; - tmp21 = tmp12 + tmp1; - tmp26 = tmp12 - tmp1; - tmp22 = tmp13 + tmp2; - tmp25 = tmp13 - tmp2; - tmp23 = tmp11 + tmp3; - tmp24 = tmp11 - tmp3; - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - z4 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - - tmp11 = z1 + z3; - - tmp1 = MULTIPLY(z1 + z2, FIX(1.353318001)); /* c3 */ - tmp2 = MULTIPLY(tmp11, FIX(1.247225013)); /* c5 */ - tmp3 = MULTIPLY(z1 + z4, FIX(1.093201867)); /* c7 */ - tmp10 = MULTIPLY(z1 - z4, FIX(0.897167586)); /* c9 */ - tmp11 = MULTIPLY(tmp11, FIX(0.666655658)); /* c11 */ - tmp12 = MULTIPLY(z1 - z2, FIX(0.410524528)); /* c13 */ - tmp0 = tmp1 + tmp2 + tmp3 - - MULTIPLY(z1, FIX(2.286341144)); /* c7+c5+c3-c1 */ - tmp13 = tmp10 + tmp11 + tmp12 - - MULTIPLY(z1, FIX(1.835730603)); /* c9+c11+c13-c15 */ - z1 = MULTIPLY(z2 + z3, FIX(0.138617169)); /* c15 */ - tmp1 += z1 + MULTIPLY(z2, FIX(0.071888074)); /* c9+c11-c3-c15 */ - tmp2 += z1 - MULTIPLY(z3, FIX(1.125726048)); /* c5+c7+c15-c3 */ - z1 = MULTIPLY(z3 - z2, FIX(1.407403738)); /* c1 */ - tmp11 += z1 - MULTIPLY(z3, FIX(0.766367282)); /* c1+c11-c9-c13 */ - tmp12 += z1 + MULTIPLY(z2, FIX(1.971951411)); /* c1+c5+c13-c7 */ - z2 += z4; - z1 = MULTIPLY(z2, - FIX(0.666655658)); /* -c11 */ - tmp1 += z1; - tmp3 += z1 + MULTIPLY(z4, FIX(1.065388962)); /* c3+c11+c15-c7 */ - z2 = MULTIPLY(z2, - FIX(1.247225013)); /* -c5 */ - tmp10 += z2 + MULTIPLY(z4, FIX(3.141271809)); /* c1+c5+c9-c13 */ - tmp12 += z2; - z2 = MULTIPLY(z3 + z4, - FIX(1.353318001)); /* -c3 */ - tmp2 += z2; - tmp3 += z2; - z2 = MULTIPLY(z4 - z3, FIX(0.410524528)); /* c13 */ - tmp10 += z2; - tmp11 += z2; - - /* Final output stage */ - - wsptr[8*0] = (int) RIGHT_SHIFT(tmp20 + tmp0, CONST_BITS-PASS1_BITS); - wsptr[8*15] = (int) RIGHT_SHIFT(tmp20 - tmp0, CONST_BITS-PASS1_BITS); - wsptr[8*1] = (int) RIGHT_SHIFT(tmp21 + tmp1, CONST_BITS-PASS1_BITS); - wsptr[8*14] = (int) RIGHT_SHIFT(tmp21 - tmp1, CONST_BITS-PASS1_BITS); - wsptr[8*2] = (int) RIGHT_SHIFT(tmp22 + tmp2, CONST_BITS-PASS1_BITS); - wsptr[8*13] = (int) RIGHT_SHIFT(tmp22 - tmp2, CONST_BITS-PASS1_BITS); - wsptr[8*3] = (int) RIGHT_SHIFT(tmp23 + tmp3, CONST_BITS-PASS1_BITS); - wsptr[8*12] = (int) RIGHT_SHIFT(tmp23 - tmp3, CONST_BITS-PASS1_BITS); - wsptr[8*4] = (int) RIGHT_SHIFT(tmp24 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*11] = (int) RIGHT_SHIFT(tmp24 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*5] = (int) RIGHT_SHIFT(tmp25 + tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*10] = (int) RIGHT_SHIFT(tmp25 - tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*6] = (int) RIGHT_SHIFT(tmp26 + tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*9] = (int) RIGHT_SHIFT(tmp26 - tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*7] = (int) RIGHT_SHIFT(tmp27 + tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*8] = (int) RIGHT_SHIFT(tmp27 - tmp13, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 16 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 16; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp0 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp0 <<= CONST_BITS; - - z1 = (INT32) wsptr[4]; - tmp1 = MULTIPLY(z1, FIX(1.306562965)); /* c4[16] = c2[8] */ - tmp2 = MULTIPLY(z1, FIX_0_541196100); /* c12[16] = c6[8] */ - - tmp10 = tmp0 + tmp1; - tmp11 = tmp0 - tmp1; - tmp12 = tmp0 + tmp2; - tmp13 = tmp0 - tmp2; - - z1 = (INT32) wsptr[2]; - z2 = (INT32) wsptr[6]; - z3 = z1 - z2; - z4 = MULTIPLY(z3, FIX(0.275899379)); /* c14[16] = c7[8] */ - z3 = MULTIPLY(z3, FIX(1.387039845)); /* c2[16] = c1[8] */ - - tmp0 = z3 + MULTIPLY(z2, FIX_2_562915447); /* (c6+c2)[16] = (c3+c1)[8] */ - tmp1 = z4 + MULTIPLY(z1, FIX_0_899976223); /* (c6-c14)[16] = (c3-c7)[8] */ - tmp2 = z3 - MULTIPLY(z1, FIX(0.601344887)); /* (c2-c10)[16] = (c1-c5)[8] */ - tmp3 = z4 - MULTIPLY(z2, FIX(0.509795579)); /* (c10-c14)[16] = (c5-c7)[8] */ - - tmp20 = tmp10 + tmp0; - tmp27 = tmp10 - tmp0; - tmp21 = tmp12 + tmp1; - tmp26 = tmp12 - tmp1; - tmp22 = tmp13 + tmp2; - tmp25 = tmp13 - tmp2; - tmp23 = tmp11 + tmp3; - tmp24 = tmp11 - tmp3; - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - z4 = (INT32) wsptr[7]; - - tmp11 = z1 + z3; - - tmp1 = MULTIPLY(z1 + z2, FIX(1.353318001)); /* c3 */ - tmp2 = MULTIPLY(tmp11, FIX(1.247225013)); /* c5 */ - tmp3 = MULTIPLY(z1 + z4, FIX(1.093201867)); /* c7 */ - tmp10 = MULTIPLY(z1 - z4, FIX(0.897167586)); /* c9 */ - tmp11 = MULTIPLY(tmp11, FIX(0.666655658)); /* c11 */ - tmp12 = MULTIPLY(z1 - z2, FIX(0.410524528)); /* c13 */ - tmp0 = tmp1 + tmp2 + tmp3 - - MULTIPLY(z1, FIX(2.286341144)); /* c7+c5+c3-c1 */ - tmp13 = tmp10 + tmp11 + tmp12 - - MULTIPLY(z1, FIX(1.835730603)); /* c9+c11+c13-c15 */ - z1 = MULTIPLY(z2 + z3, FIX(0.138617169)); /* c15 */ - tmp1 += z1 + MULTIPLY(z2, FIX(0.071888074)); /* c9+c11-c3-c15 */ - tmp2 += z1 - MULTIPLY(z3, FIX(1.125726048)); /* c5+c7+c15-c3 */ - z1 = MULTIPLY(z3 - z2, FIX(1.407403738)); /* c1 */ - tmp11 += z1 - MULTIPLY(z3, FIX(0.766367282)); /* c1+c11-c9-c13 */ - tmp12 += z1 + MULTIPLY(z2, FIX(1.971951411)); /* c1+c5+c13-c7 */ - z2 += z4; - z1 = MULTIPLY(z2, - FIX(0.666655658)); /* -c11 */ - tmp1 += z1; - tmp3 += z1 + MULTIPLY(z4, FIX(1.065388962)); /* c3+c11+c15-c7 */ - z2 = MULTIPLY(z2, - FIX(1.247225013)); /* -c5 */ - tmp10 += z2 + MULTIPLY(z4, FIX(3.141271809)); /* c1+c5+c9-c13 */ - tmp12 += z2; - z2 = MULTIPLY(z3 + z4, - FIX(1.353318001)); /* -c3 */ - tmp2 += z2; - tmp3 += z2; - z2 = MULTIPLY(z4 - z3, FIX(0.410524528)); /* c13 */ - tmp10 += z2; - tmp11 += z2; - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[15] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[14] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[13] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp23 + tmp3, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[12] = range_limit[(int) RIGHT_SHIFT(tmp23 - tmp3, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp24 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[11] = range_limit[(int) RIGHT_SHIFT(tmp24 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp25 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[10] = range_limit[(int) RIGHT_SHIFT(tmp25 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp26 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[9] = range_limit[(int) RIGHT_SHIFT(tmp26 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp27 + tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[8] = range_limit[(int) RIGHT_SHIFT(tmp27 - tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 8; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 16x8 output block. - * - * 8-point IDCT in pass 1 (columns), 16-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_16x8 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp10, tmp11, tmp12, tmp13; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24, tmp25, tmp26, tmp27; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*8]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * Note results are scaled up by sqrt(8) compared to a true IDCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * 8-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/16). - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = DCTSIZE; ctr > 0; ctr--) { - /* Due to quantization, we will usually find that many of the input - * coefficients are zero, especially the AC terms. We can exploit this - * by short-circuiting the IDCT calculation for any column in which all - * the AC terms are zero. In that case each output is equal to the - * DC coefficient (with scale factor as needed). - * With typical images and quantization tables, half or more of the - * column DCT calculations can be simplified this way. - */ - - if (inptr[DCTSIZE*1] == 0 && inptr[DCTSIZE*2] == 0 && - inptr[DCTSIZE*3] == 0 && inptr[DCTSIZE*4] == 0 && - inptr[DCTSIZE*5] == 0 && inptr[DCTSIZE*6] == 0 && - inptr[DCTSIZE*7] == 0) { - /* AC terms all zero */ - int dcval = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]) << PASS1_BITS; - - wsptr[DCTSIZE*0] = dcval; - wsptr[DCTSIZE*1] = dcval; - wsptr[DCTSIZE*2] = dcval; - wsptr[DCTSIZE*3] = dcval; - wsptr[DCTSIZE*4] = dcval; - wsptr[DCTSIZE*5] = dcval; - wsptr[DCTSIZE*6] = dcval; - wsptr[DCTSIZE*7] = dcval; - - inptr++; /* advance pointers to next column */ - quantptr++; - wsptr++; - continue; - } - - /* Even part: reverse the even part of the forward DCT. - * The rotator is c(-6). - */ - - z2 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - z3 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z2 <<= CONST_BITS; - z3 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - z2 += ONE << (CONST_BITS-PASS1_BITS-1); - - tmp0 = z2 + z3; - tmp1 = z2 - z3; - - z2 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z3 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - tmp2 = z1 + MULTIPLY(z2, FIX_0_765366865); /* c2-c6 */ - tmp3 = z1 - MULTIPLY(z3, FIX_1_847759065); /* c2+c6 */ - - tmp10 = tmp0 + tmp2; - tmp13 = tmp0 - tmp2; - tmp11 = tmp1 + tmp3; - tmp12 = tmp1 - tmp3; - - /* Odd part per figure 8; the matrix is unitary and hence its - * transpose is its inverse. i0..i3 are y7,y5,y3,y1 respectively. - */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - tmp1 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - tmp2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - tmp3 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - - z2 = tmp0 + tmp2; - z3 = tmp1 + tmp3; - - z1 = MULTIPLY(z2 + z3, FIX_1_175875602); /* c3 */ - z2 = MULTIPLY(z2, - FIX_1_961570560); /* -c3-c5 */ - z3 = MULTIPLY(z3, - FIX_0_390180644); /* -c3+c5 */ - z2 += z1; - z3 += z1; - - z1 = MULTIPLY(tmp0 + tmp3, - FIX_0_899976223); /* -c3+c7 */ - tmp0 = MULTIPLY(tmp0, FIX_0_298631336); /* -c1+c3+c5-c7 */ - tmp3 = MULTIPLY(tmp3, FIX_1_501321110); /* c1+c3-c5-c7 */ - tmp0 += z1 + z2; - tmp3 += z1 + z3; - - z1 = MULTIPLY(tmp1 + tmp2, - FIX_2_562915447); /* -c1-c3 */ - tmp1 = MULTIPLY(tmp1, FIX_2_053119869); /* c1+c3-c5+c7 */ - tmp2 = MULTIPLY(tmp2, FIX_3_072711026); /* c1+c3+c5-c7 */ - tmp1 += z1 + z3; - tmp2 += z1 + z2; - - /* Final output stage: inputs are tmp10..tmp13, tmp0..tmp3 */ - - wsptr[DCTSIZE*0] = (int) RIGHT_SHIFT(tmp10 + tmp3, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*7] = (int) RIGHT_SHIFT(tmp10 - tmp3, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*1] = (int) RIGHT_SHIFT(tmp11 + tmp2, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*6] = (int) RIGHT_SHIFT(tmp11 - tmp2, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*2] = (int) RIGHT_SHIFT(tmp12 + tmp1, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*5] = (int) RIGHT_SHIFT(tmp12 - tmp1, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*3] = (int) RIGHT_SHIFT(tmp13 + tmp0, CONST_BITS-PASS1_BITS); - wsptr[DCTSIZE*4] = (int) RIGHT_SHIFT(tmp13 - tmp0, CONST_BITS-PASS1_BITS); - - inptr++; /* advance pointers to next column */ - quantptr++; - wsptr++; - } - - /* Pass 2: process 8 rows from work array, store into output array. - * 16-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/32). - */ - - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp0 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp0 <<= CONST_BITS; - - z1 = (INT32) wsptr[4]; - tmp1 = MULTIPLY(z1, FIX(1.306562965)); /* c4[16] = c2[8] */ - tmp2 = MULTIPLY(z1, FIX_0_541196100); /* c12[16] = c6[8] */ - - tmp10 = tmp0 + tmp1; - tmp11 = tmp0 - tmp1; - tmp12 = tmp0 + tmp2; - tmp13 = tmp0 - tmp2; - - z1 = (INT32) wsptr[2]; - z2 = (INT32) wsptr[6]; - z3 = z1 - z2; - z4 = MULTIPLY(z3, FIX(0.275899379)); /* c14[16] = c7[8] */ - z3 = MULTIPLY(z3, FIX(1.387039845)); /* c2[16] = c1[8] */ - - tmp0 = z3 + MULTIPLY(z2, FIX_2_562915447); /* (c6+c2)[16] = (c3+c1)[8] */ - tmp1 = z4 + MULTIPLY(z1, FIX_0_899976223); /* (c6-c14)[16] = (c3-c7)[8] */ - tmp2 = z3 - MULTIPLY(z1, FIX(0.601344887)); /* (c2-c10)[16] = (c1-c5)[8] */ - tmp3 = z4 - MULTIPLY(z2, FIX(0.509795579)); /* (c10-c14)[16] = (c5-c7)[8] */ - - tmp20 = tmp10 + tmp0; - tmp27 = tmp10 - tmp0; - tmp21 = tmp12 + tmp1; - tmp26 = tmp12 - tmp1; - tmp22 = tmp13 + tmp2; - tmp25 = tmp13 - tmp2; - tmp23 = tmp11 + tmp3; - tmp24 = tmp11 - tmp3; - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - z4 = (INT32) wsptr[7]; - - tmp11 = z1 + z3; - - tmp1 = MULTIPLY(z1 + z2, FIX(1.353318001)); /* c3 */ - tmp2 = MULTIPLY(tmp11, FIX(1.247225013)); /* c5 */ - tmp3 = MULTIPLY(z1 + z4, FIX(1.093201867)); /* c7 */ - tmp10 = MULTIPLY(z1 - z4, FIX(0.897167586)); /* c9 */ - tmp11 = MULTIPLY(tmp11, FIX(0.666655658)); /* c11 */ - tmp12 = MULTIPLY(z1 - z2, FIX(0.410524528)); /* c13 */ - tmp0 = tmp1 + tmp2 + tmp3 - - MULTIPLY(z1, FIX(2.286341144)); /* c7+c5+c3-c1 */ - tmp13 = tmp10 + tmp11 + tmp12 - - MULTIPLY(z1, FIX(1.835730603)); /* c9+c11+c13-c15 */ - z1 = MULTIPLY(z2 + z3, FIX(0.138617169)); /* c15 */ - tmp1 += z1 + MULTIPLY(z2, FIX(0.071888074)); /* c9+c11-c3-c15 */ - tmp2 += z1 - MULTIPLY(z3, FIX(1.125726048)); /* c5+c7+c15-c3 */ - z1 = MULTIPLY(z3 - z2, FIX(1.407403738)); /* c1 */ - tmp11 += z1 - MULTIPLY(z3, FIX(0.766367282)); /* c1+c11-c9-c13 */ - tmp12 += z1 + MULTIPLY(z2, FIX(1.971951411)); /* c1+c5+c13-c7 */ - z2 += z4; - z1 = MULTIPLY(z2, - FIX(0.666655658)); /* -c11 */ - tmp1 += z1; - tmp3 += z1 + MULTIPLY(z4, FIX(1.065388962)); /* c3+c11+c15-c7 */ - z2 = MULTIPLY(z2, - FIX(1.247225013)); /* -c5 */ - tmp10 += z2 + MULTIPLY(z4, FIX(3.141271809)); /* c1+c5+c9-c13 */ - tmp12 += z2; - z2 = MULTIPLY(z3 + z4, - FIX(1.353318001)); /* -c3 */ - tmp2 += z2; - tmp3 += z2; - z2 = MULTIPLY(z4 - z3, FIX(0.410524528)); /* c13 */ - tmp10 += z2; - tmp11 += z2; - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[15] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[14] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[13] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp23 + tmp3, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[12] = range_limit[(int) RIGHT_SHIFT(tmp23 - tmp3, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp24 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[11] = range_limit[(int) RIGHT_SHIFT(tmp24 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp25 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[10] = range_limit[(int) RIGHT_SHIFT(tmp25 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp26 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[9] = range_limit[(int) RIGHT_SHIFT(tmp26 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp27 + tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[8] = range_limit[(int) RIGHT_SHIFT(tmp27 - tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 8; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 14x7 output block. - * - * 7-point IDCT in pass 1 (columns), 14-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_14x7 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15, tmp16; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24, tmp25, tmp26; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*7]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * 7-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/14). - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp23 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp23 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp23 += ONE << (CONST_BITS-PASS1_BITS-1); - - z1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z2 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z3 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - tmp20 = MULTIPLY(z2 - z3, FIX(0.881747734)); /* c4 */ - tmp22 = MULTIPLY(z1 - z2, FIX(0.314692123)); /* c6 */ - tmp21 = tmp20 + tmp22 + tmp23 - MULTIPLY(z2, FIX(1.841218003)); /* c2+c4-c6 */ - tmp10 = z1 + z3; - z2 -= tmp10; - tmp10 = MULTIPLY(tmp10, FIX(1.274162392)) + tmp23; /* c2 */ - tmp20 += tmp10 - MULTIPLY(z3, FIX(0.077722536)); /* c2-c4-c6 */ - tmp22 += tmp10 - MULTIPLY(z1, FIX(2.470602249)); /* c2+c4+c6 */ - tmp23 += MULTIPLY(z2, FIX(1.414213562)); /* c0 */ - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - - tmp11 = MULTIPLY(z1 + z2, FIX(0.935414347)); /* (c3+c1-c5)/2 */ - tmp12 = MULTIPLY(z1 - z2, FIX(0.170262339)); /* (c3+c5-c1)/2 */ - tmp10 = tmp11 - tmp12; - tmp11 += tmp12; - tmp12 = MULTIPLY(z2 + z3, - FIX(1.378756276)); /* -c1 */ - tmp11 += tmp12; - z2 = MULTIPLY(z1 + z3, FIX(0.613604268)); /* c5 */ - tmp10 += z2; - tmp12 += z2 + MULTIPLY(z3, FIX(1.870828693)); /* c3+c1-c5 */ - - /* Final output stage */ - - wsptr[8*0] = (int) RIGHT_SHIFT(tmp20 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*6] = (int) RIGHT_SHIFT(tmp20 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*1] = (int) RIGHT_SHIFT(tmp21 + tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*5] = (int) RIGHT_SHIFT(tmp21 - tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*2] = (int) RIGHT_SHIFT(tmp22 + tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*4] = (int) RIGHT_SHIFT(tmp22 - tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*3] = (int) RIGHT_SHIFT(tmp23, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 7 rows from work array, store into output array. - * 14-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/28). - */ - - wsptr = workspace; - for (ctr = 0; ctr < 7; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - z1 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - z1 <<= CONST_BITS; - z4 = (INT32) wsptr[4]; - z2 = MULTIPLY(z4, FIX(1.274162392)); /* c4 */ - z3 = MULTIPLY(z4, FIX(0.314692123)); /* c12 */ - z4 = MULTIPLY(z4, FIX(0.881747734)); /* c8 */ - - tmp10 = z1 + z2; - tmp11 = z1 + z3; - tmp12 = z1 - z4; - - tmp23 = z1 - ((z2 + z3 - z4) << 1); /* c0 = (c4+c12-c8)*2 */ - - z1 = (INT32) wsptr[2]; - z2 = (INT32) wsptr[6]; - - z3 = MULTIPLY(z1 + z2, FIX(1.105676686)); /* c6 */ - - tmp13 = z3 + MULTIPLY(z1, FIX(0.273079590)); /* c2-c6 */ - tmp14 = z3 - MULTIPLY(z2, FIX(1.719280954)); /* c6+c10 */ - tmp15 = MULTIPLY(z1, FIX(0.613604268)) - /* c10 */ - MULTIPLY(z2, FIX(1.378756276)); /* c2 */ - - tmp20 = tmp10 + tmp13; - tmp26 = tmp10 - tmp13; - tmp21 = tmp11 + tmp14; - tmp25 = tmp11 - tmp14; - tmp22 = tmp12 + tmp15; - tmp24 = tmp12 - tmp15; - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - z4 = (INT32) wsptr[7]; - z4 <<= CONST_BITS; - - tmp14 = z1 + z3; - tmp11 = MULTIPLY(z1 + z2, FIX(1.334852607)); /* c3 */ - tmp12 = MULTIPLY(tmp14, FIX(1.197448846)); /* c5 */ - tmp10 = tmp11 + tmp12 + z4 - MULTIPLY(z1, FIX(1.126980169)); /* c3+c5-c1 */ - tmp14 = MULTIPLY(tmp14, FIX(0.752406978)); /* c9 */ - tmp16 = tmp14 - MULTIPLY(z1, FIX(1.061150426)); /* c9+c11-c13 */ - z1 -= z2; - tmp15 = MULTIPLY(z1, FIX(0.467085129)) - z4; /* c11 */ - tmp16 += tmp15; - tmp13 = MULTIPLY(z2 + z3, - FIX(0.158341681)) - z4; /* -c13 */ - tmp11 += tmp13 - MULTIPLY(z2, FIX(0.424103948)); /* c3-c9-c13 */ - tmp12 += tmp13 - MULTIPLY(z3, FIX(2.373959773)); /* c3+c5-c13 */ - tmp13 = MULTIPLY(z3 - z2, FIX(1.405321284)); /* c1 */ - tmp14 += tmp13 + z4 - MULTIPLY(z3, FIX(1.6906431334)); /* c1+c9-c11 */ - tmp15 += tmp13 + MULTIPLY(z2, FIX(0.674957567)); /* c1+c11-c5 */ - - tmp13 = ((z1 - z3) << CONST_BITS) + z4; - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[13] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[12] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[11] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp23 + tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[10] = range_limit[(int) RIGHT_SHIFT(tmp23 - tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp24 + tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[9] = range_limit[(int) RIGHT_SHIFT(tmp24 - tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp25 + tmp15, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[8] = range_limit[(int) RIGHT_SHIFT(tmp25 - tmp15, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp26 + tmp16, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp26 - tmp16, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 8; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 12x6 output block. - * - * 6-point IDCT in pass 1 (columns), 12-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_12x6 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24, tmp25; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*6]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * 6-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/12). - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp10 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp10 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp10 += ONE << (CONST_BITS-PASS1_BITS-1); - tmp12 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - tmp20 = MULTIPLY(tmp12, FIX(0.707106781)); /* c4 */ - tmp11 = tmp10 + tmp20; - tmp21 = RIGHT_SHIFT(tmp10 - tmp20 - tmp20, CONST_BITS-PASS1_BITS); - tmp20 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - tmp10 = MULTIPLY(tmp20, FIX(1.224744871)); /* c2 */ - tmp20 = tmp11 + tmp10; - tmp22 = tmp11 - tmp10; - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - tmp11 = MULTIPLY(z1 + z3, FIX(0.366025404)); /* c5 */ - tmp10 = tmp11 + ((z1 + z2) << CONST_BITS); - tmp12 = tmp11 + ((z3 - z2) << CONST_BITS); - tmp11 = (z1 - z2 - z3) << PASS1_BITS; - - /* Final output stage */ - - wsptr[8*0] = (int) RIGHT_SHIFT(tmp20 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*5] = (int) RIGHT_SHIFT(tmp20 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*1] = (int) (tmp21 + tmp11); - wsptr[8*4] = (int) (tmp21 - tmp11); - wsptr[8*2] = (int) RIGHT_SHIFT(tmp22 + tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*3] = (int) RIGHT_SHIFT(tmp22 - tmp12, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 6 rows from work array, store into output array. - * 12-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/24). - */ - - wsptr = workspace; - for (ctr = 0; ctr < 6; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - z3 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - z3 <<= CONST_BITS; - - z4 = (INT32) wsptr[4]; - z4 = MULTIPLY(z4, FIX(1.224744871)); /* c4 */ - - tmp10 = z3 + z4; - tmp11 = z3 - z4; - - z1 = (INT32) wsptr[2]; - z4 = MULTIPLY(z1, FIX(1.366025404)); /* c2 */ - z1 <<= CONST_BITS; - z2 = (INT32) wsptr[6]; - z2 <<= CONST_BITS; - - tmp12 = z1 - z2; - - tmp21 = z3 + tmp12; - tmp24 = z3 - tmp12; - - tmp12 = z4 + z2; - - tmp20 = tmp10 + tmp12; - tmp25 = tmp10 - tmp12; - - tmp12 = z4 - z1 - z2; - - tmp22 = tmp11 + tmp12; - tmp23 = tmp11 - tmp12; - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - z4 = (INT32) wsptr[7]; - - tmp11 = MULTIPLY(z2, FIX(1.306562965)); /* c3 */ - tmp14 = MULTIPLY(z2, - FIX_0_541196100); /* -c9 */ - - tmp10 = z1 + z3; - tmp15 = MULTIPLY(tmp10 + z4, FIX(0.860918669)); /* c7 */ - tmp12 = tmp15 + MULTIPLY(tmp10, FIX(0.261052384)); /* c5-c7 */ - tmp10 = tmp12 + tmp11 + MULTIPLY(z1, FIX(0.280143716)); /* c1-c5 */ - tmp13 = MULTIPLY(z3 + z4, - FIX(1.045510580)); /* -(c7+c11) */ - tmp12 += tmp13 + tmp14 - MULTIPLY(z3, FIX(1.478575242)); /* c1+c5-c7-c11 */ - tmp13 += tmp15 - tmp11 + MULTIPLY(z4, FIX(1.586706681)); /* c1+c11 */ - tmp15 += tmp14 - MULTIPLY(z1, FIX(0.676326758)) - /* c7-c11 */ - MULTIPLY(z4, FIX(1.982889723)); /* c5+c7 */ - - z1 -= z4; - z2 -= z3; - z3 = MULTIPLY(z1 + z2, FIX_0_541196100); /* c9 */ - tmp11 = z3 + MULTIPLY(z1, FIX_0_765366865); /* c3-c9 */ - tmp14 = z3 - MULTIPLY(z2, FIX_1_847759065); /* c3+c9 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[11] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[10] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[9] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp23 + tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[8] = range_limit[(int) RIGHT_SHIFT(tmp23 - tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp24 + tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp24 - tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp25 + tmp15, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp25 - tmp15, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 8; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 10x5 output block. - * - * 5-point IDCT in pass 1 (columns), 10-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_10x5 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp10, tmp11, tmp12, tmp13, tmp14; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*5]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * 5-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/10). - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp12 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp12 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp12 += ONE << (CONST_BITS-PASS1_BITS-1); - tmp13 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - tmp14 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z1 = MULTIPLY(tmp13 + tmp14, FIX(0.790569415)); /* (c2+c4)/2 */ - z2 = MULTIPLY(tmp13 - tmp14, FIX(0.353553391)); /* (c2-c4)/2 */ - z3 = tmp12 + z2; - tmp10 = z3 + z1; - tmp11 = z3 - z1; - tmp12 -= z2 << 2; - - /* Odd part */ - - z2 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z3 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - - z1 = MULTIPLY(z2 + z3, FIX(0.831253876)); /* c3 */ - tmp13 = z1 + MULTIPLY(z2, FIX(0.513743148)); /* c1-c3 */ - tmp14 = z1 - MULTIPLY(z3, FIX(2.176250899)); /* c1+c3 */ - - /* Final output stage */ - - wsptr[8*0] = (int) RIGHT_SHIFT(tmp10 + tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*4] = (int) RIGHT_SHIFT(tmp10 - tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*1] = (int) RIGHT_SHIFT(tmp11 + tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*3] = (int) RIGHT_SHIFT(tmp11 - tmp14, CONST_BITS-PASS1_BITS); - wsptr[8*2] = (int) RIGHT_SHIFT(tmp12, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 5 rows from work array, store into output array. - * 10-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/20). - */ - - wsptr = workspace; - for (ctr = 0; ctr < 5; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - z3 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - z3 <<= CONST_BITS; - z4 = (INT32) wsptr[4]; - z1 = MULTIPLY(z4, FIX(1.144122806)); /* c4 */ - z2 = MULTIPLY(z4, FIX(0.437016024)); /* c8 */ - tmp10 = z3 + z1; - tmp11 = z3 - z2; - - tmp22 = z3 - ((z1 - z2) << 1); /* c0 = (c4-c8)*2 */ - - z2 = (INT32) wsptr[2]; - z3 = (INT32) wsptr[6]; - - z1 = MULTIPLY(z2 + z3, FIX(0.831253876)); /* c6 */ - tmp12 = z1 + MULTIPLY(z2, FIX(0.513743148)); /* c2-c6 */ - tmp13 = z1 - MULTIPLY(z3, FIX(2.176250899)); /* c2+c6 */ - - tmp20 = tmp10 + tmp12; - tmp24 = tmp10 - tmp12; - tmp21 = tmp11 + tmp13; - tmp23 = tmp11 - tmp13; - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - z3 <<= CONST_BITS; - z4 = (INT32) wsptr[7]; - - tmp11 = z2 + z4; - tmp13 = z2 - z4; - - tmp12 = MULTIPLY(tmp13, FIX(0.309016994)); /* (c3-c7)/2 */ - - z2 = MULTIPLY(tmp11, FIX(0.951056516)); /* (c3+c7)/2 */ - z4 = z3 + tmp12; - - tmp10 = MULTIPLY(z1, FIX(1.396802247)) + z2 + z4; /* c1 */ - tmp14 = MULTIPLY(z1, FIX(0.221231742)) - z2 + z4; /* c9 */ - - z2 = MULTIPLY(tmp11, FIX(0.587785252)); /* (c1-c9)/2 */ - z4 = z3 - tmp12 - (tmp13 << (CONST_BITS - 1)); - - tmp12 = ((z1 - tmp13) << CONST_BITS) - z3; - - tmp11 = MULTIPLY(z1, FIX(1.260073511)) - z2 - z4; /* c3 */ - tmp13 = MULTIPLY(z1, FIX(0.642039522)) - z2 + z4; /* c7 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[9] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[8] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp23 + tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp23 - tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp24 + tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp24 - tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 8; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing an 8x4 output block. - * - * 4-point IDCT in pass 1 (columns), 8-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_8x4 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3; - INT32 tmp10, tmp11, tmp12, tmp13; - INT32 z1, z2, z3; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*4]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * 4-point IDCT kernel, - * cK represents sqrt(2) * cos(K*pi/16) [refers to 8-point IDCT]. - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp2 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - - tmp10 = (tmp0 + tmp2) << PASS1_BITS; - tmp12 = (tmp0 - tmp2) << PASS1_BITS; - - /* Odd part */ - /* Same rotation as in the even part of the 8x8 LL&M IDCT */ - - z2 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z3 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS-PASS1_BITS-1); - tmp0 = RIGHT_SHIFT(z1 + MULTIPLY(z2, FIX_0_765366865), /* c2-c6 */ - CONST_BITS-PASS1_BITS); - tmp2 = RIGHT_SHIFT(z1 - MULTIPLY(z3, FIX_1_847759065), /* c2+c6 */ - CONST_BITS-PASS1_BITS); - - /* Final output stage */ - - wsptr[8*0] = (int) (tmp10 + tmp0); - wsptr[8*3] = (int) (tmp10 - tmp0); - wsptr[8*1] = (int) (tmp12 + tmp2); - wsptr[8*2] = (int) (tmp12 - tmp2); - } - - /* Pass 2: process rows from work array, store into output array. - * Note that we must descale the results by a factor of 8 == 2**3, - * and also undo the PASS1_BITS scaling. - * 8-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/16). - */ - - wsptr = workspace; - for (ctr = 0; ctr < 4; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part: reverse the even part of the forward DCT. - * The rotator is c(-6). - */ - - /* Add range center and fudge factor for final descale and range-limit. */ - z2 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - z3 = (INT32) wsptr[4]; - - tmp0 = (z2 + z3) << CONST_BITS; - tmp1 = (z2 - z3) << CONST_BITS; - - z2 = (INT32) wsptr[2]; - z3 = (INT32) wsptr[6]; - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - tmp2 = z1 + MULTIPLY(z2, FIX_0_765366865); /* c2-c6 */ - tmp3 = z1 - MULTIPLY(z3, FIX_1_847759065); /* c2+c6 */ - - tmp10 = tmp0 + tmp2; - tmp13 = tmp0 - tmp2; - tmp11 = tmp1 + tmp3; - tmp12 = tmp1 - tmp3; - - /* Odd part per figure 8; the matrix is unitary and hence its - * transpose is its inverse. i0..i3 are y7,y5,y3,y1 respectively. - */ - - tmp0 = (INT32) wsptr[7]; - tmp1 = (INT32) wsptr[5]; - tmp2 = (INT32) wsptr[3]; - tmp3 = (INT32) wsptr[1]; - - z2 = tmp0 + tmp2; - z3 = tmp1 + tmp3; - - z1 = MULTIPLY(z2 + z3, FIX_1_175875602); /* c3 */ - z2 = MULTIPLY(z2, - FIX_1_961570560); /* -c3-c5 */ - z3 = MULTIPLY(z3, - FIX_0_390180644); /* -c3+c5 */ - z2 += z1; - z3 += z1; - - z1 = MULTIPLY(tmp0 + tmp3, - FIX_0_899976223); /* -c3+c7 */ - tmp0 = MULTIPLY(tmp0, FIX_0_298631336); /* -c1+c3+c5-c7 */ - tmp3 = MULTIPLY(tmp3, FIX_1_501321110); /* c1+c3-c5-c7 */ - tmp0 += z1 + z2; - tmp3 += z1 + z3; - - z1 = MULTIPLY(tmp1 + tmp2, - FIX_2_562915447); /* -c1-c3 */ - tmp1 = MULTIPLY(tmp1, FIX_2_053119869); /* c1+c3-c5+c7 */ - tmp2 = MULTIPLY(tmp2, FIX_3_072711026); /* c1+c3+c5-c7 */ - tmp1 += z1 + z3; - tmp2 += z1 + z2; - - /* Final output stage: inputs are tmp10..tmp13, tmp0..tmp3 */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp3, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp3, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp11 + tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp11 - tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp12 + tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp12 - tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp13 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp13 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += DCTSIZE; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 6x3 output block. - * - * 3-point IDCT in pass 1 (columns), 6-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_6x3 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp1, tmp2, tmp10, tmp11, tmp12; - INT32 z1, z2, z3; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[6*3]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * 3-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/6). - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 6; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp0 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp0 += ONE << (CONST_BITS-PASS1_BITS-1); - tmp2 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - tmp12 = MULTIPLY(tmp2, FIX(0.707106781)); /* c2 */ - tmp10 = tmp0 + tmp12; - tmp2 = tmp0 - tmp12 - tmp12; - - /* Odd part */ - - tmp12 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - tmp0 = MULTIPLY(tmp12, FIX(1.224744871)); /* c1 */ - - /* Final output stage */ - - wsptr[6*0] = (int) RIGHT_SHIFT(tmp10 + tmp0, CONST_BITS-PASS1_BITS); - wsptr[6*2] = (int) RIGHT_SHIFT(tmp10 - tmp0, CONST_BITS-PASS1_BITS); - wsptr[6*1] = (int) RIGHT_SHIFT(tmp2, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 3 rows from work array, store into output array. - * 6-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/12). - */ - - wsptr = workspace; - for (ctr = 0; ctr < 3; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp0 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp0 <<= CONST_BITS; - tmp2 = (INT32) wsptr[4]; - tmp10 = MULTIPLY(tmp2, FIX(0.707106781)); /* c4 */ - tmp1 = tmp0 + tmp10; - tmp11 = tmp0 - tmp10 - tmp10; - tmp10 = (INT32) wsptr[2]; - tmp0 = MULTIPLY(tmp10, FIX(1.224744871)); /* c2 */ - tmp10 = tmp1 + tmp0; - tmp12 = tmp1 - tmp0; - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - tmp1 = MULTIPLY(z1 + z3, FIX(0.366025404)); /* c5 */ - tmp0 = tmp1 + ((z1 + z2) << CONST_BITS); - tmp2 = tmp1 + ((z3 - z2) << CONST_BITS); - tmp1 = (z1 - z2 - z3) << CONST_BITS; - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp11 + tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp11 - tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp12 + tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp12 - tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 6; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 4x2 output block. - * - * 2-point IDCT in pass 1 (columns), 4-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_4x2 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp2, tmp10, tmp12; - INT32 z1, z2, z3; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - INT32 * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - INT32 workspace[4*2]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 4; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp10 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - - /* Odd part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - - /* Final output stage */ - - wsptr[4*0] = tmp10 + tmp0; - wsptr[4*1] = tmp10 - tmp0; - } - - /* Pass 2: process 2 rows from work array, store into output array. - * 4-point IDCT kernel, - * cK represents sqrt(2) * cos(K*pi/16) [refers to 8-point IDCT]. - */ - - wsptr = workspace; - for (ctr = 0; ctr < 2; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp0 = wsptr[0] + ((((INT32) RANGE_CENTER) << 3) + (ONE << 2)); - tmp2 = wsptr[2]; - - tmp10 = (tmp0 + tmp2) << CONST_BITS; - tmp12 = (tmp0 - tmp2) << CONST_BITS; - - /* Odd part */ - /* Same rotation as in the even part of the 8x8 LL&M IDCT */ - - z2 = wsptr[1]; - z3 = wsptr[3]; - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - tmp0 = z1 + MULTIPLY(z2, FIX_0_765366865); /* c2-c6 */ - tmp2 = z1 - MULTIPLY(z3, FIX_1_847759065); /* c2+c6 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp0, - CONST_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp0, - CONST_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp12 + tmp2, - CONST_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp12 - tmp2, - CONST_BITS+3) - & RANGE_MASK]; - - wsptr += 4; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 2x1 output block. - * - * 1-point IDCT in pass 1 (columns), 2-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_2x1 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - DCTELEM tmp0, tmp1; - ISLOW_MULT_TYPE * quantptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - ISHIFT_TEMPS - - /* Pass 1: empty. */ - - /* Pass 2: process 1 row from input, store into output array. */ - - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - outptr = output_buf[0] + output_col; - - /* Even part */ - - tmp0 = DEQUANTIZE(coef_block[0], quantptr[0]); - /* Add range center and fudge factor for final descale and range-limit. */ - tmp0 += (((DCTELEM) RANGE_CENTER) << 3) + (1 << 2); - - /* Odd part */ - - tmp1 = DEQUANTIZE(coef_block[1], quantptr[1]); - - /* Final output stage */ - - outptr[0] = range_limit[(int) IRIGHT_SHIFT(tmp0 + tmp1, 3) & RANGE_MASK]; - outptr[1] = range_limit[(int) IRIGHT_SHIFT(tmp0 - tmp1, 3) & RANGE_MASK]; -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing an 8x16 output block. - * - * 16-point IDCT in pass 1 (columns), 8-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_8x16 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3, tmp10, tmp11, tmp12, tmp13; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24, tmp25, tmp26, tmp27; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[8*16]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * 16-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/32). - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp0 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp0 += ONE << (CONST_BITS-PASS1_BITS-1); - - z1 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - tmp1 = MULTIPLY(z1, FIX(1.306562965)); /* c4[16] = c2[8] */ - tmp2 = MULTIPLY(z1, FIX_0_541196100); /* c12[16] = c6[8] */ - - tmp10 = tmp0 + tmp1; - tmp11 = tmp0 - tmp1; - tmp12 = tmp0 + tmp2; - tmp13 = tmp0 - tmp2; - - z1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z2 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - z3 = z1 - z2; - z4 = MULTIPLY(z3, FIX(0.275899379)); /* c14[16] = c7[8] */ - z3 = MULTIPLY(z3, FIX(1.387039845)); /* c2[16] = c1[8] */ - - tmp0 = z3 + MULTIPLY(z2, FIX_2_562915447); /* (c6+c2)[16] = (c3+c1)[8] */ - tmp1 = z4 + MULTIPLY(z1, FIX_0_899976223); /* (c6-c14)[16] = (c3-c7)[8] */ - tmp2 = z3 - MULTIPLY(z1, FIX(0.601344887)); /* (c2-c10)[16] = (c1-c5)[8] */ - tmp3 = z4 - MULTIPLY(z2, FIX(0.509795579)); /* (c10-c14)[16] = (c5-c7)[8] */ - - tmp20 = tmp10 + tmp0; - tmp27 = tmp10 - tmp0; - tmp21 = tmp12 + tmp1; - tmp26 = tmp12 - tmp1; - tmp22 = tmp13 + tmp2; - tmp25 = tmp13 - tmp2; - tmp23 = tmp11 + tmp3; - tmp24 = tmp11 - tmp3; - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - z4 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - - tmp11 = z1 + z3; - - tmp1 = MULTIPLY(z1 + z2, FIX(1.353318001)); /* c3 */ - tmp2 = MULTIPLY(tmp11, FIX(1.247225013)); /* c5 */ - tmp3 = MULTIPLY(z1 + z4, FIX(1.093201867)); /* c7 */ - tmp10 = MULTIPLY(z1 - z4, FIX(0.897167586)); /* c9 */ - tmp11 = MULTIPLY(tmp11, FIX(0.666655658)); /* c11 */ - tmp12 = MULTIPLY(z1 - z2, FIX(0.410524528)); /* c13 */ - tmp0 = tmp1 + tmp2 + tmp3 - - MULTIPLY(z1, FIX(2.286341144)); /* c7+c5+c3-c1 */ - tmp13 = tmp10 + tmp11 + tmp12 - - MULTIPLY(z1, FIX(1.835730603)); /* c9+c11+c13-c15 */ - z1 = MULTIPLY(z2 + z3, FIX(0.138617169)); /* c15 */ - tmp1 += z1 + MULTIPLY(z2, FIX(0.071888074)); /* c9+c11-c3-c15 */ - tmp2 += z1 - MULTIPLY(z3, FIX(1.125726048)); /* c5+c7+c15-c3 */ - z1 = MULTIPLY(z3 - z2, FIX(1.407403738)); /* c1 */ - tmp11 += z1 - MULTIPLY(z3, FIX(0.766367282)); /* c1+c11-c9-c13 */ - tmp12 += z1 + MULTIPLY(z2, FIX(1.971951411)); /* c1+c5+c13-c7 */ - z2 += z4; - z1 = MULTIPLY(z2, - FIX(0.666655658)); /* -c11 */ - tmp1 += z1; - tmp3 += z1 + MULTIPLY(z4, FIX(1.065388962)); /* c3+c11+c15-c7 */ - z2 = MULTIPLY(z2, - FIX(1.247225013)); /* -c5 */ - tmp10 += z2 + MULTIPLY(z4, FIX(3.141271809)); /* c1+c5+c9-c13 */ - tmp12 += z2; - z2 = MULTIPLY(z3 + z4, - FIX(1.353318001)); /* -c3 */ - tmp2 += z2; - tmp3 += z2; - z2 = MULTIPLY(z4 - z3, FIX(0.410524528)); /* c13 */ - tmp10 += z2; - tmp11 += z2; - - /* Final output stage */ - - wsptr[8*0] = (int) RIGHT_SHIFT(tmp20 + tmp0, CONST_BITS-PASS1_BITS); - wsptr[8*15] = (int) RIGHT_SHIFT(tmp20 - tmp0, CONST_BITS-PASS1_BITS); - wsptr[8*1] = (int) RIGHT_SHIFT(tmp21 + tmp1, CONST_BITS-PASS1_BITS); - wsptr[8*14] = (int) RIGHT_SHIFT(tmp21 - tmp1, CONST_BITS-PASS1_BITS); - wsptr[8*2] = (int) RIGHT_SHIFT(tmp22 + tmp2, CONST_BITS-PASS1_BITS); - wsptr[8*13] = (int) RIGHT_SHIFT(tmp22 - tmp2, CONST_BITS-PASS1_BITS); - wsptr[8*3] = (int) RIGHT_SHIFT(tmp23 + tmp3, CONST_BITS-PASS1_BITS); - wsptr[8*12] = (int) RIGHT_SHIFT(tmp23 - tmp3, CONST_BITS-PASS1_BITS); - wsptr[8*4] = (int) RIGHT_SHIFT(tmp24 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*11] = (int) RIGHT_SHIFT(tmp24 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[8*5] = (int) RIGHT_SHIFT(tmp25 + tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*10] = (int) RIGHT_SHIFT(tmp25 - tmp11, CONST_BITS-PASS1_BITS); - wsptr[8*6] = (int) RIGHT_SHIFT(tmp26 + tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*9] = (int) RIGHT_SHIFT(tmp26 - tmp12, CONST_BITS-PASS1_BITS); - wsptr[8*7] = (int) RIGHT_SHIFT(tmp27 + tmp13, CONST_BITS-PASS1_BITS); - wsptr[8*8] = (int) RIGHT_SHIFT(tmp27 - tmp13, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process rows from work array, store into output array. - * Note that we must descale the results by a factor of 8 == 2**3, - * and also undo the PASS1_BITS scaling. - * 8-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/16). - */ - - wsptr = workspace; - for (ctr = 0; ctr < 16; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part: reverse the even part of the forward DCT. - * The rotator is c(-6). - */ - - /* Add range center and fudge factor for final descale and range-limit. */ - z2 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - z3 = (INT32) wsptr[4]; - - tmp0 = (z2 + z3) << CONST_BITS; - tmp1 = (z2 - z3) << CONST_BITS; - - z2 = (INT32) wsptr[2]; - z3 = (INT32) wsptr[6]; - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - tmp2 = z1 + MULTIPLY(z2, FIX_0_765366865); /* c2-c6 */ - tmp3 = z1 - MULTIPLY(z3, FIX_1_847759065); /* c2+c6 */ - - tmp10 = tmp0 + tmp2; - tmp13 = tmp0 - tmp2; - tmp11 = tmp1 + tmp3; - tmp12 = tmp1 - tmp3; - - /* Odd part per figure 8; the matrix is unitary and hence its - * transpose is its inverse. i0..i3 are y7,y5,y3,y1 respectively. - */ - - tmp0 = (INT32) wsptr[7]; - tmp1 = (INT32) wsptr[5]; - tmp2 = (INT32) wsptr[3]; - tmp3 = (INT32) wsptr[1]; - - z2 = tmp0 + tmp2; - z3 = tmp1 + tmp3; - - z1 = MULTIPLY(z2 + z3, FIX_1_175875602); /* c3 */ - z2 = MULTIPLY(z2, - FIX_1_961570560); /* -c3-c5 */ - z3 = MULTIPLY(z3, - FIX_0_390180644); /* -c3+c5 */ - z2 += z1; - z3 += z1; - - z1 = MULTIPLY(tmp0 + tmp3, - FIX_0_899976223); /* -c3+c7 */ - tmp0 = MULTIPLY(tmp0, FIX_0_298631336); /* -c1+c3+c5-c7 */ - tmp3 = MULTIPLY(tmp3, FIX_1_501321110); /* c1+c3-c5-c7 */ - tmp0 += z1 + z2; - tmp3 += z1 + z3; - - z1 = MULTIPLY(tmp1 + tmp2, - FIX_2_562915447); /* -c1-c3 */ - tmp1 = MULTIPLY(tmp1, FIX_2_053119869); /* c1+c3-c5+c7 */ - tmp2 = MULTIPLY(tmp2, FIX_3_072711026); /* c1+c3+c5-c7 */ - tmp1 += z1 + z3; - tmp2 += z1 + z2; - - /* Final output stage: inputs are tmp10..tmp13, tmp0..tmp3 */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp3, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[7] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp3, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp11 + tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp11 - tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp12 + tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp12 - tmp1, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp13 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp13 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += DCTSIZE; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 7x14 output block. - * - * 14-point IDCT in pass 1 (columns), 7-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_7x14 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15, tmp16; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24, tmp25, tmp26; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[7*14]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * 14-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/28). - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 7; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - z1 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - z1 += ONE << (CONST_BITS-PASS1_BITS-1); - z4 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z2 = MULTIPLY(z4, FIX(1.274162392)); /* c4 */ - z3 = MULTIPLY(z4, FIX(0.314692123)); /* c12 */ - z4 = MULTIPLY(z4, FIX(0.881747734)); /* c8 */ - - tmp10 = z1 + z2; - tmp11 = z1 + z3; - tmp12 = z1 - z4; - - tmp23 = RIGHT_SHIFT(z1 - ((z2 + z3 - z4) << 1), /* c0 = (c4+c12-c8)*2 */ - CONST_BITS-PASS1_BITS); - - z1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z2 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - z3 = MULTIPLY(z1 + z2, FIX(1.105676686)); /* c6 */ - - tmp13 = z3 + MULTIPLY(z1, FIX(0.273079590)); /* c2-c6 */ - tmp14 = z3 - MULTIPLY(z2, FIX(1.719280954)); /* c6+c10 */ - tmp15 = MULTIPLY(z1, FIX(0.613604268)) - /* c10 */ - MULTIPLY(z2, FIX(1.378756276)); /* c2 */ - - tmp20 = tmp10 + tmp13; - tmp26 = tmp10 - tmp13; - tmp21 = tmp11 + tmp14; - tmp25 = tmp11 - tmp14; - tmp22 = tmp12 + tmp15; - tmp24 = tmp12 - tmp15; - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - z4 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - tmp13 = z4 << CONST_BITS; - - tmp14 = z1 + z3; - tmp11 = MULTIPLY(z1 + z2, FIX(1.334852607)); /* c3 */ - tmp12 = MULTIPLY(tmp14, FIX(1.197448846)); /* c5 */ - tmp10 = tmp11 + tmp12 + tmp13 - MULTIPLY(z1, FIX(1.126980169)); /* c3+c5-c1 */ - tmp14 = MULTIPLY(tmp14, FIX(0.752406978)); /* c9 */ - tmp16 = tmp14 - MULTIPLY(z1, FIX(1.061150426)); /* c9+c11-c13 */ - z1 -= z2; - tmp15 = MULTIPLY(z1, FIX(0.467085129)) - tmp13; /* c11 */ - tmp16 += tmp15; - z1 += z4; - z4 = MULTIPLY(z2 + z3, - FIX(0.158341681)) - tmp13; /* -c13 */ - tmp11 += z4 - MULTIPLY(z2, FIX(0.424103948)); /* c3-c9-c13 */ - tmp12 += z4 - MULTIPLY(z3, FIX(2.373959773)); /* c3+c5-c13 */ - z4 = MULTIPLY(z3 - z2, FIX(1.405321284)); /* c1 */ - tmp14 += z4 + tmp13 - MULTIPLY(z3, FIX(1.6906431334)); /* c1+c9-c11 */ - tmp15 += z4 + MULTIPLY(z2, FIX(0.674957567)); /* c1+c11-c5 */ - - tmp13 = (z1 - z3) << PASS1_BITS; - - /* Final output stage */ - - wsptr[7*0] = (int) RIGHT_SHIFT(tmp20 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[7*13] = (int) RIGHT_SHIFT(tmp20 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[7*1] = (int) RIGHT_SHIFT(tmp21 + tmp11, CONST_BITS-PASS1_BITS); - wsptr[7*12] = (int) RIGHT_SHIFT(tmp21 - tmp11, CONST_BITS-PASS1_BITS); - wsptr[7*2] = (int) RIGHT_SHIFT(tmp22 + tmp12, CONST_BITS-PASS1_BITS); - wsptr[7*11] = (int) RIGHT_SHIFT(tmp22 - tmp12, CONST_BITS-PASS1_BITS); - wsptr[7*3] = (int) (tmp23 + tmp13); - wsptr[7*10] = (int) (tmp23 - tmp13); - wsptr[7*4] = (int) RIGHT_SHIFT(tmp24 + tmp14, CONST_BITS-PASS1_BITS); - wsptr[7*9] = (int) RIGHT_SHIFT(tmp24 - tmp14, CONST_BITS-PASS1_BITS); - wsptr[7*5] = (int) RIGHT_SHIFT(tmp25 + tmp15, CONST_BITS-PASS1_BITS); - wsptr[7*8] = (int) RIGHT_SHIFT(tmp25 - tmp15, CONST_BITS-PASS1_BITS); - wsptr[7*6] = (int) RIGHT_SHIFT(tmp26 + tmp16, CONST_BITS-PASS1_BITS); - wsptr[7*7] = (int) RIGHT_SHIFT(tmp26 - tmp16, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 14 rows from work array, store into output array. - * 7-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/14). - */ - - wsptr = workspace; - for (ctr = 0; ctr < 14; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp23 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp23 <<= CONST_BITS; - - z1 = (INT32) wsptr[2]; - z2 = (INT32) wsptr[4]; - z3 = (INT32) wsptr[6]; - - tmp20 = MULTIPLY(z2 - z3, FIX(0.881747734)); /* c4 */ - tmp22 = MULTIPLY(z1 - z2, FIX(0.314692123)); /* c6 */ - tmp21 = tmp20 + tmp22 + tmp23 - MULTIPLY(z2, FIX(1.841218003)); /* c2+c4-c6 */ - tmp10 = z1 + z3; - z2 -= tmp10; - tmp10 = MULTIPLY(tmp10, FIX(1.274162392)) + tmp23; /* c2 */ - tmp20 += tmp10 - MULTIPLY(z3, FIX(0.077722536)); /* c2-c4-c6 */ - tmp22 += tmp10 - MULTIPLY(z1, FIX(2.470602249)); /* c2+c4+c6 */ - tmp23 += MULTIPLY(z2, FIX(1.414213562)); /* c0 */ - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - - tmp11 = MULTIPLY(z1 + z2, FIX(0.935414347)); /* (c3+c1-c5)/2 */ - tmp12 = MULTIPLY(z1 - z2, FIX(0.170262339)); /* (c3+c5-c1)/2 */ - tmp10 = tmp11 - tmp12; - tmp11 += tmp12; - tmp12 = MULTIPLY(z2 + z3, - FIX(1.378756276)); /* -c1 */ - tmp11 += tmp12; - z2 = MULTIPLY(z1 + z3, FIX(0.613604268)); /* c5 */ - tmp10 += z2; - tmp12 += z2 + MULTIPLY(z3, FIX(1.870828693)); /* c3+c1-c5 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[6] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp23, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 7; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 6x12 output block. - * - * 12-point IDCT in pass 1 (columns), 6-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_6x12 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp10, tmp11, tmp12, tmp13, tmp14, tmp15; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24, tmp25; - INT32 z1, z2, z3, z4; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[6*12]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * 12-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/24). - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 6; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - z3 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - z3 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - z3 += ONE << (CONST_BITS-PASS1_BITS-1); - - z4 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z4 = MULTIPLY(z4, FIX(1.224744871)); /* c4 */ - - tmp10 = z3 + z4; - tmp11 = z3 - z4; - - z1 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z4 = MULTIPLY(z1, FIX(1.366025404)); /* c2 */ - z1 <<= CONST_BITS; - z2 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - z2 <<= CONST_BITS; - - tmp12 = z1 - z2; - - tmp21 = z3 + tmp12; - tmp24 = z3 - tmp12; - - tmp12 = z4 + z2; - - tmp20 = tmp10 + tmp12; - tmp25 = tmp10 - tmp12; - - tmp12 = z4 - z1 - z2; - - tmp22 = tmp11 + tmp12; - tmp23 = tmp11 - tmp12; - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - z4 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - - tmp11 = MULTIPLY(z2, FIX(1.306562965)); /* c3 */ - tmp14 = MULTIPLY(z2, - FIX_0_541196100); /* -c9 */ - - tmp10 = z1 + z3; - tmp15 = MULTIPLY(tmp10 + z4, FIX(0.860918669)); /* c7 */ - tmp12 = tmp15 + MULTIPLY(tmp10, FIX(0.261052384)); /* c5-c7 */ - tmp10 = tmp12 + tmp11 + MULTIPLY(z1, FIX(0.280143716)); /* c1-c5 */ - tmp13 = MULTIPLY(z3 + z4, - FIX(1.045510580)); /* -(c7+c11) */ - tmp12 += tmp13 + tmp14 - MULTIPLY(z3, FIX(1.478575242)); /* c1+c5-c7-c11 */ - tmp13 += tmp15 - tmp11 + MULTIPLY(z4, FIX(1.586706681)); /* c1+c11 */ - tmp15 += tmp14 - MULTIPLY(z1, FIX(0.676326758)) - /* c7-c11 */ - MULTIPLY(z4, FIX(1.982889723)); /* c5+c7 */ - - z1 -= z4; - z2 -= z3; - z3 = MULTIPLY(z1 + z2, FIX_0_541196100); /* c9 */ - tmp11 = z3 + MULTIPLY(z1, FIX_0_765366865); /* c3-c9 */ - tmp14 = z3 - MULTIPLY(z2, FIX_1_847759065); /* c3+c9 */ - - /* Final output stage */ - - wsptr[6*0] = (int) RIGHT_SHIFT(tmp20 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[6*11] = (int) RIGHT_SHIFT(tmp20 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[6*1] = (int) RIGHT_SHIFT(tmp21 + tmp11, CONST_BITS-PASS1_BITS); - wsptr[6*10] = (int) RIGHT_SHIFT(tmp21 - tmp11, CONST_BITS-PASS1_BITS); - wsptr[6*2] = (int) RIGHT_SHIFT(tmp22 + tmp12, CONST_BITS-PASS1_BITS); - wsptr[6*9] = (int) RIGHT_SHIFT(tmp22 - tmp12, CONST_BITS-PASS1_BITS); - wsptr[6*3] = (int) RIGHT_SHIFT(tmp23 + tmp13, CONST_BITS-PASS1_BITS); - wsptr[6*8] = (int) RIGHT_SHIFT(tmp23 - tmp13, CONST_BITS-PASS1_BITS); - wsptr[6*4] = (int) RIGHT_SHIFT(tmp24 + tmp14, CONST_BITS-PASS1_BITS); - wsptr[6*7] = (int) RIGHT_SHIFT(tmp24 - tmp14, CONST_BITS-PASS1_BITS); - wsptr[6*5] = (int) RIGHT_SHIFT(tmp25 + tmp15, CONST_BITS-PASS1_BITS); - wsptr[6*6] = (int) RIGHT_SHIFT(tmp25 - tmp15, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 12 rows from work array, store into output array. - * 6-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/12). - */ - - wsptr = workspace; - for (ctr = 0; ctr < 12; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp10 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp10 <<= CONST_BITS; - tmp12 = (INT32) wsptr[4]; - tmp20 = MULTIPLY(tmp12, FIX(0.707106781)); /* c4 */ - tmp11 = tmp10 + tmp20; - tmp21 = tmp10 - tmp20 - tmp20; - tmp20 = (INT32) wsptr[2]; - tmp10 = MULTIPLY(tmp20, FIX(1.224744871)); /* c2 */ - tmp20 = tmp11 + tmp10; - tmp22 = tmp11 - tmp10; - - /* Odd part */ - - z1 = (INT32) wsptr[1]; - z2 = (INT32) wsptr[3]; - z3 = (INT32) wsptr[5]; - tmp11 = MULTIPLY(z1 + z3, FIX(0.366025404)); /* c5 */ - tmp10 = tmp11 + ((z1 + z2) << CONST_BITS); - tmp12 = tmp11 + ((z3 - z2) << CONST_BITS); - tmp11 = (z1 - z2 - z3) << CONST_BITS; - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp20 + tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[5] = range_limit[(int) RIGHT_SHIFT(tmp20 - tmp10, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp21 + tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp21 - tmp11, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp22 + tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp22 - tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 6; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 5x10 output block. - * - * 10-point IDCT in pass 1 (columns), 5-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_5x10 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp10, tmp11, tmp12, tmp13, tmp14; - INT32 tmp20, tmp21, tmp22, tmp23, tmp24; - INT32 z1, z2, z3, z4, z5; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[5*10]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * 10-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/20). - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 5; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - z3 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - z3 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - z3 += ONE << (CONST_BITS-PASS1_BITS-1); - z4 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z1 = MULTIPLY(z4, FIX(1.144122806)); /* c4 */ - z2 = MULTIPLY(z4, FIX(0.437016024)); /* c8 */ - tmp10 = z3 + z1; - tmp11 = z3 - z2; - - tmp22 = RIGHT_SHIFT(z3 - ((z1 - z2) << 1), /* c0 = (c4-c8)*2 */ - CONST_BITS-PASS1_BITS); - - z2 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z3 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - z1 = MULTIPLY(z2 + z3, FIX(0.831253876)); /* c6 */ - tmp12 = z1 + MULTIPLY(z2, FIX(0.513743148)); /* c2-c6 */ - tmp13 = z1 - MULTIPLY(z3, FIX(2.176250899)); /* c2+c6 */ - - tmp20 = tmp10 + tmp12; - tmp24 = tmp10 - tmp12; - tmp21 = tmp11 + tmp13; - tmp23 = tmp11 - tmp13; - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - z4 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - - tmp11 = z2 + z4; - tmp13 = z2 - z4; - - tmp12 = MULTIPLY(tmp13, FIX(0.309016994)); /* (c3-c7)/2 */ - z5 = z3 << CONST_BITS; - - z2 = MULTIPLY(tmp11, FIX(0.951056516)); /* (c3+c7)/2 */ - z4 = z5 + tmp12; - - tmp10 = MULTIPLY(z1, FIX(1.396802247)) + z2 + z4; /* c1 */ - tmp14 = MULTIPLY(z1, FIX(0.221231742)) - z2 + z4; /* c9 */ - - z2 = MULTIPLY(tmp11, FIX(0.587785252)); /* (c1-c9)/2 */ - z4 = z5 - tmp12 - (tmp13 << (CONST_BITS - 1)); - - tmp12 = (z1 - tmp13 - z3) << PASS1_BITS; - - tmp11 = MULTIPLY(z1, FIX(1.260073511)) - z2 - z4; /* c3 */ - tmp13 = MULTIPLY(z1, FIX(0.642039522)) - z2 + z4; /* c7 */ - - /* Final output stage */ - - wsptr[5*0] = (int) RIGHT_SHIFT(tmp20 + tmp10, CONST_BITS-PASS1_BITS); - wsptr[5*9] = (int) RIGHT_SHIFT(tmp20 - tmp10, CONST_BITS-PASS1_BITS); - wsptr[5*1] = (int) RIGHT_SHIFT(tmp21 + tmp11, CONST_BITS-PASS1_BITS); - wsptr[5*8] = (int) RIGHT_SHIFT(tmp21 - tmp11, CONST_BITS-PASS1_BITS); - wsptr[5*2] = (int) (tmp22 + tmp12); - wsptr[5*7] = (int) (tmp22 - tmp12); - wsptr[5*3] = (int) RIGHT_SHIFT(tmp23 + tmp13, CONST_BITS-PASS1_BITS); - wsptr[5*6] = (int) RIGHT_SHIFT(tmp23 - tmp13, CONST_BITS-PASS1_BITS); - wsptr[5*4] = (int) RIGHT_SHIFT(tmp24 + tmp14, CONST_BITS-PASS1_BITS); - wsptr[5*5] = (int) RIGHT_SHIFT(tmp24 - tmp14, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 10 rows from work array, store into output array. - * 5-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/10). - */ - - wsptr = workspace; - for (ctr = 0; ctr < 10; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp12 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp12 <<= CONST_BITS; - tmp13 = (INT32) wsptr[2]; - tmp14 = (INT32) wsptr[4]; - z1 = MULTIPLY(tmp13 + tmp14, FIX(0.790569415)); /* (c2+c4)/2 */ - z2 = MULTIPLY(tmp13 - tmp14, FIX(0.353553391)); /* (c2-c4)/2 */ - z3 = tmp12 + z2; - tmp10 = z3 + z1; - tmp11 = z3 - z1; - tmp12 -= z2 << 2; - - /* Odd part */ - - z2 = (INT32) wsptr[1]; - z3 = (INT32) wsptr[3]; - - z1 = MULTIPLY(z2 + z3, FIX(0.831253876)); /* c3 */ - tmp13 = z1 + MULTIPLY(z2, FIX(0.513743148)); /* c1-c3 */ - tmp14 = z1 - MULTIPLY(z3, FIX(2.176250899)); /* c1+c3 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[4] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp13, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp11 + tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp11 - tmp14, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp12, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 5; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 4x8 output block. - * - * 8-point IDCT in pass 1 (columns), 4-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_4x8 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp1, tmp2, tmp3; - INT32 tmp10, tmp11, tmp12, tmp13; - INT32 z1, z2, z3; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[4*8]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * Note results are scaled up by sqrt(8) compared to a true IDCT; - * furthermore, we scale the results by 2**PASS1_BITS. - * 8-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/16). - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 4; ctr > 0; ctr--) { - /* Due to quantization, we will usually find that many of the input - * coefficients are zero, especially the AC terms. We can exploit this - * by short-circuiting the IDCT calculation for any column in which all - * the AC terms are zero. In that case each output is equal to the - * DC coefficient (with scale factor as needed). - * With typical images and quantization tables, half or more of the - * column DCT calculations can be simplified this way. - */ - - if (inptr[DCTSIZE*1] == 0 && inptr[DCTSIZE*2] == 0 && - inptr[DCTSIZE*3] == 0 && inptr[DCTSIZE*4] == 0 && - inptr[DCTSIZE*5] == 0 && inptr[DCTSIZE*6] == 0 && - inptr[DCTSIZE*7] == 0) { - /* AC terms all zero */ - int dcval = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]) << PASS1_BITS; - - wsptr[4*0] = dcval; - wsptr[4*1] = dcval; - wsptr[4*2] = dcval; - wsptr[4*3] = dcval; - wsptr[4*4] = dcval; - wsptr[4*5] = dcval; - wsptr[4*6] = dcval; - wsptr[4*7] = dcval; - - inptr++; /* advance pointers to next column */ - quantptr++; - wsptr++; - continue; - } - - /* Even part: reverse the even part of the forward DCT. - * The rotator is c(-6). - */ - - z2 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - z3 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - z2 <<= CONST_BITS; - z3 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - z2 += ONE << (CONST_BITS-PASS1_BITS-1); - - tmp0 = z2 + z3; - tmp1 = z2 - z3; - - z2 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - z3 = DEQUANTIZE(inptr[DCTSIZE*6], quantptr[DCTSIZE*6]); - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - tmp2 = z1 + MULTIPLY(z2, FIX_0_765366865); /* c2-c6 */ - tmp3 = z1 - MULTIPLY(z3, FIX_1_847759065); /* c2+c6 */ - - tmp10 = tmp0 + tmp2; - tmp13 = tmp0 - tmp2; - tmp11 = tmp1 + tmp3; - tmp12 = tmp1 - tmp3; - - /* Odd part per figure 8; the matrix is unitary and hence its - * transpose is its inverse. i0..i3 are y7,y5,y3,y1 respectively. - */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*7], quantptr[DCTSIZE*7]); - tmp1 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - tmp2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - tmp3 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - - z2 = tmp0 + tmp2; - z3 = tmp1 + tmp3; - - z1 = MULTIPLY(z2 + z3, FIX_1_175875602); /* c3 */ - z2 = MULTIPLY(z2, - FIX_1_961570560); /* -c3-c5 */ - z3 = MULTIPLY(z3, - FIX_0_390180644); /* -c3+c5 */ - z2 += z1; - z3 += z1; - - z1 = MULTIPLY(tmp0 + tmp3, - FIX_0_899976223); /* -c3+c7 */ - tmp0 = MULTIPLY(tmp0, FIX_0_298631336); /* -c1+c3+c5-c7 */ - tmp3 = MULTIPLY(tmp3, FIX_1_501321110); /* c1+c3-c5-c7 */ - tmp0 += z1 + z2; - tmp3 += z1 + z3; - - z1 = MULTIPLY(tmp1 + tmp2, - FIX_2_562915447); /* -c1-c3 */ - tmp1 = MULTIPLY(tmp1, FIX_2_053119869); /* c1+c3-c5+c7 */ - tmp2 = MULTIPLY(tmp2, FIX_3_072711026); /* c1+c3+c5-c7 */ - tmp1 += z1 + z3; - tmp2 += z1 + z2; - - /* Final output stage: inputs are tmp10..tmp13, tmp0..tmp3 */ - - wsptr[4*0] = (int) RIGHT_SHIFT(tmp10 + tmp3, CONST_BITS-PASS1_BITS); - wsptr[4*7] = (int) RIGHT_SHIFT(tmp10 - tmp3, CONST_BITS-PASS1_BITS); - wsptr[4*1] = (int) RIGHT_SHIFT(tmp11 + tmp2, CONST_BITS-PASS1_BITS); - wsptr[4*6] = (int) RIGHT_SHIFT(tmp11 - tmp2, CONST_BITS-PASS1_BITS); - wsptr[4*2] = (int) RIGHT_SHIFT(tmp12 + tmp1, CONST_BITS-PASS1_BITS); - wsptr[4*5] = (int) RIGHT_SHIFT(tmp12 - tmp1, CONST_BITS-PASS1_BITS); - wsptr[4*3] = (int) RIGHT_SHIFT(tmp13 + tmp0, CONST_BITS-PASS1_BITS); - wsptr[4*4] = (int) RIGHT_SHIFT(tmp13 - tmp0, CONST_BITS-PASS1_BITS); - - inptr++; /* advance pointers to next column */ - quantptr++; - wsptr++; - } - - /* Pass 2: process 8 rows from work array, store into output array. - * 4-point IDCT kernel, - * cK represents sqrt(2) * cos(K*pi/16) [refers to 8-point IDCT]. - */ - - wsptr = workspace; - for (ctr = 0; ctr < 8; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp0 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp2 = (INT32) wsptr[2]; - - tmp10 = (tmp0 + tmp2) << CONST_BITS; - tmp12 = (tmp0 - tmp2) << CONST_BITS; - - /* Odd part */ - /* Same rotation as in the even part of the 8x8 LL&M IDCT */ - - z2 = (INT32) wsptr[1]; - z3 = (INT32) wsptr[3]; - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - tmp0 = z1 + MULTIPLY(z2, FIX_0_765366865); /* c2-c6 */ - tmp2 = z1 - MULTIPLY(z3, FIX_1_847759065); /* c2+c6 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[3] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp12 + tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp12 - tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 4; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 3x6 output block. - * - * 6-point IDCT in pass 1 (columns), 3-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_3x6 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp1, tmp2, tmp10, tmp11, tmp12; - INT32 z1, z2, z3; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - int * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - int workspace[3*6]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * 6-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/12). - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 3; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp0 <<= CONST_BITS; - /* Add fudge factor here for final descale. */ - tmp0 += ONE << (CONST_BITS-PASS1_BITS-1); - tmp2 = DEQUANTIZE(inptr[DCTSIZE*4], quantptr[DCTSIZE*4]); - tmp10 = MULTIPLY(tmp2, FIX(0.707106781)); /* c4 */ - tmp1 = tmp0 + tmp10; - tmp11 = RIGHT_SHIFT(tmp0 - tmp10 - tmp10, CONST_BITS-PASS1_BITS); - tmp10 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - tmp0 = MULTIPLY(tmp10, FIX(1.224744871)); /* c2 */ - tmp10 = tmp1 + tmp0; - tmp12 = tmp1 - tmp0; - - /* Odd part */ - - z1 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z2 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - z3 = DEQUANTIZE(inptr[DCTSIZE*5], quantptr[DCTSIZE*5]); - tmp1 = MULTIPLY(z1 + z3, FIX(0.366025404)); /* c5 */ - tmp0 = tmp1 + ((z1 + z2) << CONST_BITS); - tmp2 = tmp1 + ((z3 - z2) << CONST_BITS); - tmp1 = (z1 - z2 - z3) << PASS1_BITS; - - /* Final output stage */ - - wsptr[3*0] = (int) RIGHT_SHIFT(tmp10 + tmp0, CONST_BITS-PASS1_BITS); - wsptr[3*5] = (int) RIGHT_SHIFT(tmp10 - tmp0, CONST_BITS-PASS1_BITS); - wsptr[3*1] = (int) (tmp11 + tmp1); - wsptr[3*4] = (int) (tmp11 - tmp1); - wsptr[3*2] = (int) RIGHT_SHIFT(tmp12 + tmp2, CONST_BITS-PASS1_BITS); - wsptr[3*3] = (int) RIGHT_SHIFT(tmp12 - tmp2, CONST_BITS-PASS1_BITS); - } - - /* Pass 2: process 6 rows from work array, store into output array. - * 3-point IDCT kernel, cK represents sqrt(2) * cos(K*pi/6). - */ - - wsptr = workspace; - for (ctr = 0; ctr < 6; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp0 = (INT32) wsptr[0] + - ((((INT32) RANGE_CENTER) << (PASS1_BITS+3)) + - (ONE << (PASS1_BITS+2))); - tmp0 <<= CONST_BITS; - tmp2 = (INT32) wsptr[2]; - tmp12 = MULTIPLY(tmp2, FIX(0.707106781)); /* c2 */ - tmp10 = tmp0 + tmp12; - tmp2 = tmp0 - tmp12 - tmp12; - - /* Odd part */ - - tmp12 = (INT32) wsptr[1]; - tmp0 = MULTIPLY(tmp12, FIX(1.224744871)); /* c1 */ - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[2] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp0, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp2, - CONST_BITS+PASS1_BITS+3) - & RANGE_MASK]; - - wsptr += 3; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 2x4 output block. - * - * 4-point IDCT in pass 1 (columns), 2-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_2x4 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - INT32 tmp0, tmp2, tmp10, tmp12; - INT32 z1, z2, z3; - JCOEFPTR inptr; - ISLOW_MULT_TYPE * quantptr; - INT32 * wsptr; - JSAMPROW outptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - int ctr; - INT32 workspace[2*4]; /* buffers data between passes */ - SHIFT_TEMPS - - /* Pass 1: process columns from input, store into work array. - * 4-point IDCT kernel, - * cK represents sqrt(2) * cos(K*pi/16) [refers to 8-point IDCT]. - */ - - inptr = coef_block; - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - wsptr = workspace; - for (ctr = 0; ctr < 2; ctr++, inptr++, quantptr++, wsptr++) { - /* Even part */ - - tmp0 = DEQUANTIZE(inptr[DCTSIZE*0], quantptr[DCTSIZE*0]); - tmp2 = DEQUANTIZE(inptr[DCTSIZE*2], quantptr[DCTSIZE*2]); - - tmp10 = (tmp0 + tmp2) << CONST_BITS; - tmp12 = (tmp0 - tmp2) << CONST_BITS; - - /* Odd part */ - /* Same rotation as in the even part of the 8x8 LL&M IDCT */ - - z2 = DEQUANTIZE(inptr[DCTSIZE*1], quantptr[DCTSIZE*1]); - z3 = DEQUANTIZE(inptr[DCTSIZE*3], quantptr[DCTSIZE*3]); - - z1 = MULTIPLY(z2 + z3, FIX_0_541196100); /* c6 */ - tmp0 = z1 + MULTIPLY(z2, FIX_0_765366865); /* c2-c6 */ - tmp2 = z1 - MULTIPLY(z3, FIX_1_847759065); /* c2+c6 */ - - /* Final output stage */ - - wsptr[2*0] = tmp10 + tmp0; - wsptr[2*3] = tmp10 - tmp0; - wsptr[2*1] = tmp12 + tmp2; - wsptr[2*2] = tmp12 - tmp2; - } - - /* Pass 2: process 4 rows from work array, store into output array. */ - - wsptr = workspace; - for (ctr = 0; ctr < 4; ctr++) { - outptr = output_buf[ctr] + output_col; - - /* Even part */ - - /* Add range center and fudge factor for final descale and range-limit. */ - tmp10 = wsptr[0] + - ((((INT32) RANGE_CENTER) << (CONST_BITS+3)) + - (ONE << (CONST_BITS+2))); - - /* Odd part */ - - tmp0 = wsptr[1]; - - /* Final output stage */ - - outptr[0] = range_limit[(int) RIGHT_SHIFT(tmp10 + tmp0, CONST_BITS+3) - & RANGE_MASK]; - outptr[1] = range_limit[(int) RIGHT_SHIFT(tmp10 - tmp0, CONST_BITS+3) - & RANGE_MASK]; - - wsptr += 2; /* advance pointer to next row */ - } -} - - -/* - * Perform dequantization and inverse DCT on one block of coefficients, - * producing a 1x2 output block. - * - * 2-point IDCT in pass 1 (columns), 1-point in pass 2 (rows). - */ - -GLOBAL(void) -jpeg_idct_1x2 (j_decompress_ptr cinfo, jpeg_component_info * compptr, - JCOEFPTR coef_block, - JSAMPARRAY output_buf, JDIMENSION output_col) -{ - DCTELEM tmp0, tmp1; - ISLOW_MULT_TYPE * quantptr; - JSAMPLE *range_limit = IDCT_range_limit(cinfo); - ISHIFT_TEMPS - - /* Process 1 column from input, store into output array. */ - - quantptr = (ISLOW_MULT_TYPE *) compptr->dct_table; - - /* Even part */ - - tmp0 = DEQUANTIZE(coef_block[DCTSIZE*0], quantptr[DCTSIZE*0]); - /* Add range center and fudge factor for final descale and range-limit. */ - tmp0 += (((DCTELEM) RANGE_CENTER) << 3) + (1 << 2); - - /* Odd part */ - - tmp1 = DEQUANTIZE(coef_block[DCTSIZE*1], quantptr[DCTSIZE*1]); - - /* Final output stage */ - - output_buf[0][output_col] = - range_limit[(int) IRIGHT_SHIFT(tmp0 + tmp1, 3) & RANGE_MASK]; - output_buf[1][output_col] = - range_limit[(int) IRIGHT_SHIFT(tmp0 - tmp1, 3) & RANGE_MASK]; -} - -#endif /* IDCT_SCALING_SUPPORTED */ -#endif /* DCT_ISLOW_SUPPORTED */ diff --git a/dep/libjpeg/src/jinclude.h b/dep/libjpeg/src/jinclude.h deleted file mode 100644 index 12ea8cd2f..000000000 --- a/dep/libjpeg/src/jinclude.h +++ /dev/null @@ -1,157 +0,0 @@ -/* - * jinclude.h - * - * Copyright (C) 1991-1994, Thomas G. Lane. - * Modified 2017-2022 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file exists to provide a single place to fix any problems with - * including the wrong system include files. (Common problems are taken - * care of by the standard jconfig symbols, but on really weird systems - * you may have to edit this file.) - * - * NOTE: this file is NOT intended to be included by applications using - * the JPEG library. Most applications need only include jpeglib.h. - */ - - -/* Include auto-config file to find out which system include files we need. */ - -#include "jconfig.h" /* auto configuration options */ -#define JCONFIG_INCLUDED /* so that jpeglib.h doesn't do it again */ - -/* - * We need the NULL macro and size_t typedef. - * On an ANSI-conforming system it is sufficient to include . - * Otherwise, we get them from or ; we may have to - * pull in as well. - * Note that the core JPEG library does not require ; - * only the default error handler and data source/destination modules do. - * But we must pull it in because of the references to FILE in jpeglib.h. - * You can remove those references if you want to compile without . - */ - -#ifdef HAVE_STDDEF_H -#include -#endif - -#ifdef HAVE_STDLIB_H -#include -#endif - -#ifdef NEED_SYS_TYPES_H -#include -#endif - -#include - -/* - * We need memory copying and zeroing functions, plus strncpy(). - * ANSI and System V implementations declare these in . - * BSD doesn't have the mem() functions, but it does have bcopy()/bzero(). - * Some systems may declare memset and memcpy in . - * - * NOTE: we assume the size parameters to these functions are of type size_t. - * Change the casts in these macros if not! - */ - -#ifdef NEED_BSD_STRINGS - -#include -#define MEMZERO(target,size) bzero((void *)(target), (size_t)(size)) -#define MEMCOPY(dest,src,size) bcopy((const void *)(src), (void *)(dest), (size_t)(size)) - -#else /* not BSD, assume ANSI/SysV string lib */ - -#include -#define MEMZERO(target,size) memset((void *)(target), 0, (size_t)(size)) -#define MEMCOPY(dest,src,size) memcpy((void *)(dest), (const void *)(src), (size_t)(size)) - -#endif - -/* - * In ANSI C, and indeed any rational implementation, size_t is also the - * type returned by sizeof(). However, it seems there are some irrational - * implementations out there, in which sizeof() returns an int even though - * size_t is defined as long or unsigned long. To ensure consistent results - * we always use this SIZEOF() macro in place of using sizeof() directly. - */ - -#define SIZEOF(object) ((size_t) sizeof(object)) - -/* - * The modules that use fread() and fwrite() always invoke them through - * these macros. On some systems you may need to twiddle the argument casts. - * CAUTION: argument order is different from underlying functions! - * - * Furthermore, macros are provided for fflush() and ferror() in order - * to facilitate adaption by applications using an own FILE class. - * - * You can define your own custom file I/O functions in jconfig.h and - * #define JPEG_HAVE_FILE_IO_CUSTOM there to prevent redefinition here. - * - * You can #define JPEG_USE_FILE_IO_CUSTOM in jconfig.h to use custom file - * I/O functions implemented in Delphi VCL (Visual Component Library) - * in Vcl.Imaging.jpeg.pas for the TJPEGImage component utilizing - * the Delphi RTL (Run-Time Library) TMemoryStream component: - * - * procedure jpeg_stdio_src(var cinfo: jpeg_decompress_struct; - * input_file: TStream); external; - * - * procedure jpeg_stdio_dest(var cinfo: jpeg_compress_struct; - * output_file: TStream); external; - * - * function jfread(var buf; recsize, reccount: Integer; S: TStream): Integer; - * begin - * Result := S.Read(buf, recsize * reccount); - * end; - * - * function jfwrite(const buf; recsize, reccount: Integer; S: TStream): Integer; - * begin - * Result := S.Write(buf, recsize * reccount); - * end; - * - * function jfflush(S: TStream): Integer; - * begin - * Result := 0; - * end; - * - * function jferror(S: TStream): Integer; - * begin - * Result := 0; - * end; - * - * TMemoryStream of Delphi RTL has the distinctive feature to provide dynamic - * memory buffer management with a file/stream-based interface, particularly for - * the write (output) operation, which is easier to apply compared with direct - * implementations as given in jdatadst.c for memory destination. Those direct - * implementations of dynamic memory write tend to be more difficult to use, - * so providing an option like TMemoryStream may be a useful alternative. - * - * The CFile/CMemFile classes of the Microsoft Foundation Class (MFC) Library - * may be used in a similar fashion. - */ - -#ifndef JPEG_HAVE_FILE_IO_CUSTOM -#ifdef JPEG_USE_FILE_IO_CUSTOM -extern size_t jfread(void * __ptr, size_t __size, size_t __n, FILE * __stream); -extern size_t jfwrite(const void * __ptr, size_t __size, size_t __n, FILE * __stream); -extern int jfflush(FILE * __stream); -extern int jferror(FILE * __fp); - -#define JFREAD(file,buf,sizeofbuf) \ - ((size_t) jfread((void *) (buf), (size_t) 1, (size_t) (sizeofbuf), (file))) -#define JFWRITE(file,buf,sizeofbuf) \ - ((size_t) jfwrite((const void *) (buf), (size_t) 1, (size_t) (sizeofbuf), (file))) -#define JFFLUSH(file) jfflush(file) -#define JFERROR(file) jferror(file) -#else -#define JFREAD(file,buf,sizeofbuf) \ - ((size_t) fread((void *) (buf), (size_t) 1, (size_t) (sizeofbuf), (file))) -#define JFWRITE(file,buf,sizeofbuf) \ - ((size_t) fwrite((const void *) (buf), (size_t) 1, (size_t) (sizeofbuf), (file))) -#define JFFLUSH(file) fflush(file) -#define JFERROR(file) ferror(file) -#endif -#endif diff --git a/dep/libjpeg/src/jmemmgr.c b/dep/libjpeg/src/jmemmgr.c deleted file mode 100644 index 40377de47..000000000 --- a/dep/libjpeg/src/jmemmgr.c +++ /dev/null @@ -1,1115 +0,0 @@ -/* - * jmemmgr.c - * - * Copyright (C) 1991-1997, Thomas G. Lane. - * Modified 2011-2019 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains the JPEG system-independent memory management - * routines. This code is usable across a wide variety of machines; most - * of the system dependencies have been isolated in a separate file. - * The major functions provided here are: - * * pool-based allocation and freeing of memory; - * * policy decisions about how to divide available memory among the - * virtual arrays; - * * control logic for swapping virtual arrays between main memory and - * backing storage. - * The separate system-dependent file provides the actual backing-storage - * access code, and it contains the policy decision about how much total - * main memory to use. - * This file is system-dependent in the sense that some of its functions - * are unnecessary in some systems. For example, if there is enough virtual - * memory so that backing storage will never be used, much of the virtual - * array control logic could be removed. (Of course, if you have that much - * memory then you shouldn't care about a little bit of unused code...) - */ - -#define JPEG_INTERNALS -#define AM_MEMORY_MANAGER /* we define jvirt_Xarray_control structs */ -#include "jinclude.h" -#include "jpeglib.h" -#include "jmemsys.h" /* import the system-dependent declarations */ - -#ifndef NO_GETENV -#ifndef HAVE_STDLIB_H /* should declare getenv() */ -extern char * getenv JPP((const char * name)); -#endif -#endif - - -/* - * Some important notes: - * The allocation routines provided here must never return NULL. - * They should exit to error_exit if unsuccessful. - * - * It's not a good idea to try to merge the sarray and barray routines, - * even though they are textually almost the same, because samples are - * usually stored as bytes while coefficients are shorts or ints. Thus, - * in machines where byte pointers have a different representation from - * word pointers, the resulting machine code could not be the same. - */ - - -/* - * Many machines require storage alignment: longs must start on 4-byte - * boundaries, doubles on 8-byte boundaries, etc. On such machines, malloc() - * always returns pointers that are multiples of the worst-case alignment - * requirement, and we had better do so too. - * There isn't any really portable way to determine the worst-case alignment - * requirement. This module assumes that the alignment requirement is - * multiples of sizeof(ALIGN_TYPE). - * By default, we define ALIGN_TYPE as double. This is necessary on some - * workstations (where doubles really do need 8-byte alignment) and will work - * fine on nearly everything. If your machine has lesser alignment needs, - * you can save a few bytes by making ALIGN_TYPE smaller. - * The only place I know of where this will NOT work is certain Macintosh - * 680x0 compilers that define double as a 10-byte IEEE extended float. - * Doing 10-byte alignment is counterproductive because longwords won't be - * aligned well. Put "#define ALIGN_TYPE long" in jconfig.h if you have - * such a compiler. - */ - -#ifndef ALIGN_TYPE /* so can override from jconfig.h */ -#define ALIGN_TYPE double -#endif - - -/* - * We allocate objects from "pools", where each pool is gotten with a single - * request to jpeg_get_small() or jpeg_get_large(). There is no per-object - * overhead within a pool, except for alignment padding. Each pool has a - * header with a link to the next pool of the same class. - * Small and large pool headers are identical except that the latter's - * link pointer must be FAR on 80x86 machines. - * Notice that the "real" header fields are union'ed with a dummy ALIGN_TYPE - * field. This forces the compiler to make SIZEOF(small_pool_hdr) a multiple - * of the alignment requirement of ALIGN_TYPE. - */ - -typedef union small_pool_struct * small_pool_ptr; - -typedef union small_pool_struct { - struct { - small_pool_ptr next; /* next in list of pools */ - size_t bytes_used; /* how many bytes already used within pool */ - size_t bytes_left; /* bytes still available in this pool */ - } hdr; - ALIGN_TYPE dummy; /* included in union to ensure alignment */ -} small_pool_hdr; - -typedef union large_pool_struct FAR * large_pool_ptr; - -typedef union large_pool_struct { - struct { - large_pool_ptr next; /* next in list of pools */ - size_t bytes_used; /* how many bytes already used within pool */ - size_t bytes_left; /* bytes still available in this pool */ - } hdr; - ALIGN_TYPE dummy; /* included in union to ensure alignment */ -} large_pool_hdr; - - -/* - * Here is the full definition of a memory manager object. - */ - -typedef struct { - struct jpeg_memory_mgr pub; /* public fields */ - - /* Each pool identifier (lifetime class) names a linked list of pools. */ - small_pool_ptr small_list[JPOOL_NUMPOOLS]; - large_pool_ptr large_list[JPOOL_NUMPOOLS]; - - /* Since we only have one lifetime class of virtual arrays, only one - * linked list is necessary (for each datatype). Note that the virtual - * array control blocks being linked together are actually stored somewhere - * in the small-pool list. - */ - jvirt_sarray_ptr virt_sarray_list; - jvirt_barray_ptr virt_barray_list; - - /* This counts total space obtained from jpeg_get_small/large */ - size_t total_space_allocated; - - /* alloc_sarray and alloc_barray set this value for use by virtual - * array routines. - */ - JDIMENSION last_rowsperchunk; /* from most recent alloc_sarray/barray */ -} my_memory_mgr; - -typedef my_memory_mgr * my_mem_ptr; - - -/* - * The control blocks for virtual arrays. - * Note that these blocks are allocated in the "small" pool area. - * System-dependent info for the associated backing store (if any) is hidden - * inside the backing_store_info struct. - */ - -struct jvirt_sarray_control { - JSAMPARRAY mem_buffer; /* => the in-memory buffer */ - JDIMENSION rows_in_array; /* total virtual array height */ - JDIMENSION samplesperrow; /* width of array (and of memory buffer) */ - JDIMENSION maxaccess; /* max rows accessed by access_virt_sarray */ - JDIMENSION rows_in_mem; /* height of memory buffer */ - JDIMENSION rowsperchunk; /* allocation chunk size in mem_buffer */ - JDIMENSION cur_start_row; /* first logical row # in the buffer */ - JDIMENSION first_undef_row; /* row # of first uninitialized row */ - boolean pre_zero; /* pre-zero mode requested? */ - boolean dirty; /* do current buffer contents need written? */ - boolean b_s_open; /* is backing-store data valid? */ - jvirt_sarray_ptr next; /* link to next virtual sarray control block */ - backing_store_info b_s_info; /* System-dependent control info */ -}; - -struct jvirt_barray_control { - JBLOCKARRAY mem_buffer; /* => the in-memory buffer */ - JDIMENSION rows_in_array; /* total virtual array height */ - JDIMENSION blocksperrow; /* width of array (and of memory buffer) */ - JDIMENSION maxaccess; /* max rows accessed by access_virt_barray */ - JDIMENSION rows_in_mem; /* height of memory buffer */ - JDIMENSION rowsperchunk; /* allocation chunk size in mem_buffer */ - JDIMENSION cur_start_row; /* first logical row # in the buffer */ - JDIMENSION first_undef_row; /* row # of first uninitialized row */ - boolean pre_zero; /* pre-zero mode requested? */ - boolean dirty; /* do current buffer contents need written? */ - boolean b_s_open; /* is backing-store data valid? */ - jvirt_barray_ptr next; /* link to next virtual barray control block */ - backing_store_info b_s_info; /* System-dependent control info */ -}; - - -#ifdef MEM_STATS /* optional extra stuff for statistics */ - -LOCAL(void) -print_mem_stats (j_common_ptr cinfo, int pool_id) -{ - my_mem_ptr mem = (my_mem_ptr) cinfo->mem; - small_pool_ptr shdr_ptr; - large_pool_ptr lhdr_ptr; - - /* Since this is only a debugging stub, we can cheat a little by using - * fprintf directly rather than going through the trace message code. - * This is helpful because message parm array can't handle longs. - */ - fprintf(stderr, "Freeing pool %d, total space = %ld\n", - pool_id, (long) mem->total_space_allocated); - - for (lhdr_ptr = mem->large_list[pool_id]; lhdr_ptr != NULL; - lhdr_ptr = lhdr_ptr->hdr.next) { - fprintf(stderr, " Large chunk used %ld\n", - (long) lhdr_ptr->hdr.bytes_used); - } - - for (shdr_ptr = mem->small_list[pool_id]; shdr_ptr != NULL; - shdr_ptr = shdr_ptr->hdr.next) { - fprintf(stderr, " Small chunk used %ld free %ld\n", - (long) shdr_ptr->hdr.bytes_used, - (long) shdr_ptr->hdr.bytes_left); - } -} - -#endif /* MEM_STATS */ - - -LOCAL(noreturn_t) -out_of_memory (j_common_ptr cinfo, int which) -/* Report an out-of-memory error and stop execution */ -/* If we compiled MEM_STATS support, report alloc requests before dying */ -{ -#ifdef MEM_STATS - cinfo->err->trace_level = 2; /* force self_destruct to report stats */ -#endif - ERREXIT1(cinfo, JERR_OUT_OF_MEMORY, which); -} - - -/* - * Allocation of "small" objects. - * - * For these, we use pooled storage. When a new pool must be created, - * we try to get enough space for the current request plus a "slop" factor, - * where the slop will be the amount of leftover space in the new pool. - * The speed vs. space tradeoff is largely determined by the slop values. - * A different slop value is provided for each pool class (lifetime), - * and we also distinguish the first pool of a class from later ones. - * NOTE: the values given work fairly well on both 16- and 32-bit-int - * machines, but may be too small if longs are 64 bits or more. - */ - -static const size_t first_pool_slop[JPOOL_NUMPOOLS] = -{ - 1600, /* first PERMANENT pool */ - 16000 /* first IMAGE pool */ -}; - -static const size_t extra_pool_slop[JPOOL_NUMPOOLS] = -{ - 0, /* additional PERMANENT pools */ - 5000 /* additional IMAGE pools */ -}; - -#define MIN_SLOP 50 /* greater than 0 to avoid futile looping */ - - -METHODDEF(void *) -alloc_small (j_common_ptr cinfo, int pool_id, size_t sizeofobject) -/* Allocate a "small" object */ -{ - my_mem_ptr mem = (my_mem_ptr) cinfo->mem; - small_pool_ptr hdr_ptr, prev_hdr_ptr; - size_t odd_bytes, min_request, slop; - char * data_ptr; - - /* Check for unsatisfiable request (do now to ensure no overflow below) */ - if (sizeofobject > (size_t) MAX_ALLOC_CHUNK - SIZEOF(small_pool_hdr)) - out_of_memory(cinfo, 1); /* request exceeds malloc's ability */ - - /* Round up the requested size to a multiple of SIZEOF(ALIGN_TYPE) */ - odd_bytes = sizeofobject % SIZEOF(ALIGN_TYPE); - if (odd_bytes > 0) - sizeofobject += SIZEOF(ALIGN_TYPE) - odd_bytes; - - /* See if space is available in any existing pool */ - if (pool_id < 0 || pool_id >= JPOOL_NUMPOOLS) - ERREXIT1(cinfo, JERR_BAD_POOL_ID, pool_id); /* safety check */ - prev_hdr_ptr = NULL; - hdr_ptr = mem->small_list[pool_id]; - while (hdr_ptr != NULL) { - if (hdr_ptr->hdr.bytes_left >= sizeofobject) - break; /* found pool with enough space */ - prev_hdr_ptr = hdr_ptr; - hdr_ptr = hdr_ptr->hdr.next; - } - - /* Time to make a new pool? */ - if (hdr_ptr == NULL) { - /* min_request is what we need now, slop is what will be leftover */ - min_request = sizeofobject + SIZEOF(small_pool_hdr); - if (prev_hdr_ptr == NULL) /* first pool in class? */ - slop = first_pool_slop[pool_id]; - else - slop = extra_pool_slop[pool_id]; - /* Don't ask for more than MAX_ALLOC_CHUNK */ - if (slop > (size_t) MAX_ALLOC_CHUNK - min_request) - slop = (size_t) MAX_ALLOC_CHUNK - min_request; - /* Try to get space, if fail reduce slop and try again */ - for (;;) { - hdr_ptr = (small_pool_ptr) jpeg_get_small(cinfo, min_request + slop); - if (hdr_ptr != NULL) - break; - slop /= 2; - if (slop < MIN_SLOP) /* give up when it gets real small */ - out_of_memory(cinfo, 2); /* jpeg_get_small failed */ - } - mem->total_space_allocated += min_request + slop; - /* Success, initialize the new pool header and add to end of list */ - hdr_ptr->hdr.next = NULL; - hdr_ptr->hdr.bytes_used = 0; - hdr_ptr->hdr.bytes_left = sizeofobject + slop; - if (prev_hdr_ptr == NULL) /* first pool in class? */ - mem->small_list[pool_id] = hdr_ptr; - else - prev_hdr_ptr->hdr.next = hdr_ptr; - } - - /* OK, allocate the object from the current pool */ - data_ptr = (char *) (hdr_ptr + 1); /* point to first data byte in pool */ - data_ptr += hdr_ptr->hdr.bytes_used; /* point to place for object */ - hdr_ptr->hdr.bytes_used += sizeofobject; - hdr_ptr->hdr.bytes_left -= sizeofobject; - - return (void *) data_ptr; -} - - -/* - * Allocation of "large" objects. - * - * The external semantics of these are the same as "small" objects, - * except that FAR pointers are used on 80x86. However the pool - * management heuristics are quite different. We assume that each - * request is large enough that it may as well be passed directly to - * jpeg_get_large; the pool management just links everything together - * so that we can free it all on demand. - * Note: the major use of "large" objects is in JSAMPARRAY and JBLOCKARRAY - * structures. The routines that create these structures (see below) - * deliberately bunch rows together to ensure a large request size. - */ - -METHODDEF(void FAR *) -alloc_large (j_common_ptr cinfo, int pool_id, size_t sizeofobject) -/* Allocate a "large" object */ -{ - my_mem_ptr mem = (my_mem_ptr) cinfo->mem; - large_pool_ptr hdr_ptr; - size_t odd_bytes; - - /* Check for unsatisfiable request (do now to ensure no overflow below) */ - if (sizeofobject > (size_t) MAX_ALLOC_CHUNK - SIZEOF(large_pool_hdr)) - out_of_memory(cinfo, 3); /* request exceeds malloc's ability */ - - /* Round up the requested size to a multiple of SIZEOF(ALIGN_TYPE) */ - odd_bytes = sizeofobject % SIZEOF(ALIGN_TYPE); - if (odd_bytes > 0) - sizeofobject += SIZEOF(ALIGN_TYPE) - odd_bytes; - - /* Always make a new pool */ - if (pool_id < 0 || pool_id >= JPOOL_NUMPOOLS) - ERREXIT1(cinfo, JERR_BAD_POOL_ID, pool_id); /* safety check */ - - hdr_ptr = (large_pool_ptr) jpeg_get_large(cinfo, sizeofobject + - SIZEOF(large_pool_hdr)); - if (hdr_ptr == NULL) - out_of_memory(cinfo, 4); /* jpeg_get_large failed */ - mem->total_space_allocated += sizeofobject + SIZEOF(large_pool_hdr); - - /* Success, initialize the new pool header and add to list */ - hdr_ptr->hdr.next = mem->large_list[pool_id]; - /* We maintain space counts in each pool header for statistical purposes, - * even though they are not needed for allocation. - */ - hdr_ptr->hdr.bytes_used = sizeofobject; - hdr_ptr->hdr.bytes_left = 0; - mem->large_list[pool_id] = hdr_ptr; - - return (void FAR *) (hdr_ptr + 1); /* point to first data byte in pool */ -} - - -/* - * Creation of 2-D sample arrays. - * The pointers are in near heap, the samples themselves in FAR heap. - * - * To minimize allocation overhead and to allow I/O of large contiguous - * blocks, we allocate the sample rows in groups of as many rows as possible - * without exceeding MAX_ALLOC_CHUNK total bytes per allocation request. - * NB: the virtual array control routines, later in this file, know about - * this chunking of rows. The rowsperchunk value is left in the mem manager - * object so that it can be saved away if this sarray is the workspace for - * a virtual array. - */ - -METHODDEF(JSAMPARRAY) -alloc_sarray (j_common_ptr cinfo, int pool_id, - JDIMENSION samplesperrow, JDIMENSION numrows) -/* Allocate a 2-D sample array */ -{ - my_mem_ptr mem = (my_mem_ptr) cinfo->mem; - JSAMPARRAY result; - JSAMPROW workspace; - JDIMENSION rowsperchunk, currow, i; - long ltemp; - - /* Calculate max # of rows allowed in one allocation chunk */ - ltemp = (MAX_ALLOC_CHUNK - SIZEOF(large_pool_hdr)) / - ((long) samplesperrow * SIZEOF(JSAMPLE)); - if (ltemp <= 0) - ERREXIT(cinfo, JERR_WIDTH_OVERFLOW); - if (ltemp < (long) numrows) - rowsperchunk = (JDIMENSION) ltemp; - else - rowsperchunk = numrows; - mem->last_rowsperchunk = rowsperchunk; - - /* Get space for row pointers (small object) */ - result = (JSAMPARRAY) alloc_small(cinfo, pool_id, - (size_t) numrows * SIZEOF(JSAMPROW)); - - /* Get the rows themselves (large objects) */ - currow = 0; - while (currow < numrows) { - rowsperchunk = MIN(rowsperchunk, numrows - currow); - workspace = (JSAMPROW) alloc_large(cinfo, pool_id, - (size_t) rowsperchunk * (size_t) samplesperrow * SIZEOF(JSAMPLE)); - for (i = rowsperchunk; i > 0; i--) { - result[currow++] = workspace; - workspace += samplesperrow; - } - } - - return result; -} - - -/* - * Creation of 2-D coefficient-block arrays. - * This is essentially the same as the code for sample arrays, above. - */ - -METHODDEF(JBLOCKARRAY) -alloc_barray (j_common_ptr cinfo, int pool_id, - JDIMENSION blocksperrow, JDIMENSION numrows) -/* Allocate a 2-D coefficient-block array */ -{ - my_mem_ptr mem = (my_mem_ptr) cinfo->mem; - JBLOCKARRAY result; - JBLOCKROW workspace; - JDIMENSION rowsperchunk, currow, i; - long ltemp; - - /* Calculate max # of rows allowed in one allocation chunk */ - ltemp = (MAX_ALLOC_CHUNK - SIZEOF(large_pool_hdr)) / - ((long) blocksperrow * SIZEOF(JBLOCK)); - if (ltemp <= 0) - ERREXIT(cinfo, JERR_WIDTH_OVERFLOW); - if (ltemp < (long) numrows) - rowsperchunk = (JDIMENSION) ltemp; - else - rowsperchunk = numrows; - mem->last_rowsperchunk = rowsperchunk; - - /* Get space for row pointers (small object) */ - result = (JBLOCKARRAY) alloc_small(cinfo, pool_id, - (size_t) numrows * SIZEOF(JBLOCKROW)); - - /* Get the rows themselves (large objects) */ - currow = 0; - while (currow < numrows) { - rowsperchunk = MIN(rowsperchunk, numrows - currow); - workspace = (JBLOCKROW) alloc_large(cinfo, pool_id, - (size_t) rowsperchunk * (size_t) blocksperrow * SIZEOF(JBLOCK)); - for (i = rowsperchunk; i > 0; i--) { - result[currow++] = workspace; - workspace += blocksperrow; - } - } - - return result; -} - - -/* - * About virtual array management: - * - * The above "normal" array routines are only used to allocate strip buffers - * (as wide as the image, but just a few rows high). Full-image-sized buffers - * are handled as "virtual" arrays. The array is still accessed a strip at a - * time, but the memory manager must save the whole array for repeated - * accesses. The intended implementation is that there is a strip buffer in - * memory (as high as is possible given the desired memory limit), plus a - * backing file that holds the rest of the array. - * - * The request_virt_array routines are told the total size of the image and - * the maximum number of rows that will be accessed at once. The in-memory - * buffer must be at least as large as the maxaccess value. - * - * The request routines create control blocks but not the in-memory buffers. - * That is postponed until realize_virt_arrays is called. At that time the - * total amount of space needed is known (approximately, anyway), so free - * memory can be divided up fairly. - * - * The access_virt_array routines are responsible for making a specific strip - * area accessible (after reading or writing the backing file, if necessary). - * Note that the access routines are told whether the caller intends to modify - * the accessed strip; during a read-only pass this saves having to rewrite - * data to disk. The access routines are also responsible for pre-zeroing - * any newly accessed rows, if pre-zeroing was requested. - * - * In current usage, the access requests are usually for nonoverlapping - * strips; that is, successive access start_row numbers differ by exactly - * num_rows = maxaccess. This means we can get good performance with simple - * buffer dump/reload logic, by making the in-memory buffer be a multiple - * of the access height; then there will never be accesses across bufferload - * boundaries. The code will still work with overlapping access requests, - * but it doesn't handle bufferload overlaps very efficiently. - */ - - -METHODDEF(jvirt_sarray_ptr) -request_virt_sarray (j_common_ptr cinfo, int pool_id, boolean pre_zero, - JDIMENSION samplesperrow, JDIMENSION numrows, - JDIMENSION maxaccess) -/* Request a virtual 2-D sample array */ -{ - my_mem_ptr mem = (my_mem_ptr) cinfo->mem; - jvirt_sarray_ptr result; - - /* Only IMAGE-lifetime virtual arrays are currently supported */ - if (pool_id != JPOOL_IMAGE) - ERREXIT1(cinfo, JERR_BAD_POOL_ID, pool_id); /* safety check */ - - /* get control block */ - result = (jvirt_sarray_ptr) alloc_small(cinfo, pool_id, - SIZEOF(struct jvirt_sarray_control)); - - result->mem_buffer = NULL; /* marks array not yet realized */ - result->rows_in_array = numrows; - result->samplesperrow = samplesperrow; - result->maxaccess = maxaccess; - result->pre_zero = pre_zero; - result->b_s_open = FALSE; /* no associated backing-store object */ - result->next = mem->virt_sarray_list; /* add to list of virtual arrays */ - mem->virt_sarray_list = result; - - return result; -} - - -METHODDEF(jvirt_barray_ptr) -request_virt_barray (j_common_ptr cinfo, int pool_id, boolean pre_zero, - JDIMENSION blocksperrow, JDIMENSION numrows, - JDIMENSION maxaccess) -/* Request a virtual 2-D coefficient-block array */ -{ - my_mem_ptr mem = (my_mem_ptr) cinfo->mem; - jvirt_barray_ptr result; - - /* Only IMAGE-lifetime virtual arrays are currently supported */ - if (pool_id != JPOOL_IMAGE) - ERREXIT1(cinfo, JERR_BAD_POOL_ID, pool_id); /* safety check */ - - /* get control block */ - result = (jvirt_barray_ptr) alloc_small(cinfo, pool_id, - SIZEOF(struct jvirt_barray_control)); - - result->mem_buffer = NULL; /* marks array not yet realized */ - result->rows_in_array = numrows; - result->blocksperrow = blocksperrow; - result->maxaccess = maxaccess; - result->pre_zero = pre_zero; - result->b_s_open = FALSE; /* no associated backing-store object */ - result->next = mem->virt_barray_list; /* add to list of virtual arrays */ - mem->virt_barray_list = result; - - return result; -} - - -METHODDEF(void) -realize_virt_arrays (j_common_ptr cinfo) -/* Allocate the in-memory buffers for any unrealized virtual arrays */ -{ - my_mem_ptr mem = (my_mem_ptr) cinfo->mem; - long bytesperrow, space_per_minheight, maximum_space; - long avail_mem, minheights, max_minheights; - jvirt_sarray_ptr sptr; - jvirt_barray_ptr bptr; - - /* Compute the minimum space needed (maxaccess rows in each buffer) - * and the maximum space needed (full image height in each buffer). - * These may be of use to the system-dependent jpeg_mem_available routine. - */ - space_per_minheight = 0; - maximum_space = 0; - for (sptr = mem->virt_sarray_list; sptr != NULL; sptr = sptr->next) { - if (sptr->mem_buffer == NULL) { /* if not realized yet */ - bytesperrow = (long) sptr->samplesperrow * SIZEOF(JSAMPLE); - space_per_minheight += (long) sptr->maxaccess * bytesperrow; - maximum_space += (long) sptr->rows_in_array * bytesperrow; - } - } - for (bptr = mem->virt_barray_list; bptr != NULL; bptr = bptr->next) { - if (bptr->mem_buffer == NULL) { /* if not realized yet */ - bytesperrow = (long) bptr->blocksperrow * SIZEOF(JBLOCK); - space_per_minheight += (long) bptr->maxaccess * bytesperrow; - maximum_space += (long) bptr->rows_in_array * bytesperrow; - } - } - - if (space_per_minheight <= 0) - return; /* no unrealized arrays, no work */ - - /* Determine amount of memory to actually use; this is system-dependent. */ - avail_mem = jpeg_mem_available(cinfo, space_per_minheight, maximum_space, - (long) mem->total_space_allocated); - - /* If the maximum space needed is available, make all the buffers full - * height; otherwise parcel it out with the same number of minheights - * in each buffer. - */ - if (avail_mem >= maximum_space) - max_minheights = 1000000000L; - else { - max_minheights = avail_mem / space_per_minheight; - /* If there doesn't seem to be enough space, try to get the minimum - * anyway. This allows a "stub" implementation of jpeg_mem_available(). - */ - if (max_minheights <= 0) - max_minheights = 1; - } - - /* Allocate the in-memory buffers and initialize backing store as needed. */ - - for (sptr = mem->virt_sarray_list; sptr != NULL; sptr = sptr->next) { - if (sptr->mem_buffer == NULL) { /* if not realized yet */ - minheights = ((long) sptr->rows_in_array - 1L) / sptr->maxaccess + 1L; - if (minheights <= max_minheights) { - /* This buffer fits in memory */ - sptr->rows_in_mem = sptr->rows_in_array; - } else { - /* It doesn't fit in memory, create backing store. */ - sptr->rows_in_mem = (JDIMENSION) (max_minheights * sptr->maxaccess); - jpeg_open_backing_store(cinfo, & sptr->b_s_info, - (long) sptr->rows_in_array * - (long) sptr->samplesperrow * - (long) SIZEOF(JSAMPLE)); - sptr->b_s_open = TRUE; - } - sptr->mem_buffer = alloc_sarray(cinfo, JPOOL_IMAGE, - sptr->samplesperrow, sptr->rows_in_mem); - sptr->rowsperchunk = mem->last_rowsperchunk; - sptr->cur_start_row = 0; - sptr->first_undef_row = 0; - sptr->dirty = FALSE; - } - } - - for (bptr = mem->virt_barray_list; bptr != NULL; bptr = bptr->next) { - if (bptr->mem_buffer == NULL) { /* if not realized yet */ - minheights = ((long) bptr->rows_in_array - 1L) / bptr->maxaccess + 1L; - if (minheights <= max_minheights) { - /* This buffer fits in memory */ - bptr->rows_in_mem = bptr->rows_in_array; - } else { - /* It doesn't fit in memory, create backing store. */ - bptr->rows_in_mem = (JDIMENSION) (max_minheights * bptr->maxaccess); - jpeg_open_backing_store(cinfo, & bptr->b_s_info, - (long) bptr->rows_in_array * - (long) bptr->blocksperrow * - (long) SIZEOF(JBLOCK)); - bptr->b_s_open = TRUE; - } - bptr->mem_buffer = alloc_barray(cinfo, JPOOL_IMAGE, - bptr->blocksperrow, bptr->rows_in_mem); - bptr->rowsperchunk = mem->last_rowsperchunk; - bptr->cur_start_row = 0; - bptr->first_undef_row = 0; - bptr->dirty = FALSE; - } - } -} - - -LOCAL(void) -do_sarray_io (j_common_ptr cinfo, jvirt_sarray_ptr ptr, boolean writing) -/* Do backing store read or write of a virtual sample array */ -{ - long bytesperrow, file_offset, byte_count, rows, thisrow, i; - - bytesperrow = (long) ptr->samplesperrow * SIZEOF(JSAMPLE); - file_offset = (long) ptr->cur_start_row * bytesperrow; - /* Loop to read or write each allocation chunk in mem_buffer */ - for (i = 0; i < (long) ptr->rows_in_mem; i += ptr->rowsperchunk) { - /* One chunk, but check for short chunk at end of buffer */ - rows = MIN((long) ptr->rowsperchunk, (long) ptr->rows_in_mem - i); - /* Transfer no more than is currently defined */ - thisrow = (long) ptr->cur_start_row + i; - rows = MIN(rows, (long) ptr->first_undef_row - thisrow); - /* Transfer no more than fits in file */ - rows = MIN(rows, (long) ptr->rows_in_array - thisrow); - if (rows <= 0) /* this chunk might be past end of file! */ - break; - byte_count = rows * bytesperrow; - if (writing) - (*ptr->b_s_info.write_backing_store) (cinfo, & ptr->b_s_info, - (void FAR *) ptr->mem_buffer[i], - file_offset, byte_count); - else - (*ptr->b_s_info.read_backing_store) (cinfo, & ptr->b_s_info, - (void FAR *) ptr->mem_buffer[i], - file_offset, byte_count); - file_offset += byte_count; - } -} - - -LOCAL(void) -do_barray_io (j_common_ptr cinfo, jvirt_barray_ptr ptr, boolean writing) -/* Do backing store read or write of a virtual coefficient-block array */ -{ - long bytesperrow, file_offset, byte_count, rows, thisrow, i; - - bytesperrow = (long) ptr->blocksperrow * SIZEOF(JBLOCK); - file_offset = (long) ptr->cur_start_row * bytesperrow; - /* Loop to read or write each allocation chunk in mem_buffer */ - for (i = 0; i < (long) ptr->rows_in_mem; i += ptr->rowsperchunk) { - /* One chunk, but check for short chunk at end of buffer */ - rows = MIN((long) ptr->rowsperchunk, (long) ptr->rows_in_mem - i); - /* Transfer no more than is currently defined */ - thisrow = (long) ptr->cur_start_row + i; - rows = MIN(rows, (long) ptr->first_undef_row - thisrow); - /* Transfer no more than fits in file */ - rows = MIN(rows, (long) ptr->rows_in_array - thisrow); - if (rows <= 0) /* this chunk might be past end of file! */ - break; - byte_count = rows * bytesperrow; - if (writing) - (*ptr->b_s_info.write_backing_store) (cinfo, & ptr->b_s_info, - (void FAR *) ptr->mem_buffer[i], - file_offset, byte_count); - else - (*ptr->b_s_info.read_backing_store) (cinfo, & ptr->b_s_info, - (void FAR *) ptr->mem_buffer[i], - file_offset, byte_count); - file_offset += byte_count; - } -} - - -METHODDEF(JSAMPARRAY) -access_virt_sarray (j_common_ptr cinfo, jvirt_sarray_ptr ptr, - JDIMENSION start_row, JDIMENSION num_rows, - boolean writable) -/* Access the part of a virtual sample array starting at start_row */ -/* and extending for num_rows rows. writable is true if */ -/* caller intends to modify the accessed area. */ -{ - JDIMENSION end_row = start_row + num_rows; - JDIMENSION undef_row; - - /* debugging check */ - if (end_row > ptr->rows_in_array || num_rows > ptr->maxaccess || - ptr->mem_buffer == NULL) - ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS); - - /* Make the desired part of the virtual array accessible */ - if (start_row < ptr->cur_start_row || - end_row > ptr->cur_start_row + ptr->rows_in_mem) { - if (! ptr->b_s_open) - ERREXIT(cinfo, JERR_VIRTUAL_BUG); - /* Flush old buffer contents if necessary */ - if (ptr->dirty) { - do_sarray_io(cinfo, ptr, TRUE); - ptr->dirty = FALSE; - } - /* Decide what part of virtual array to access. - * Algorithm: if target address > current window, assume forward scan, - * load starting at target address. If target address < current window, - * assume backward scan, load so that target area is top of window. - * Note that when switching from forward write to forward read, will have - * start_row = 0, so the limiting case applies and we load from 0 anyway. - */ - if (start_row > ptr->cur_start_row) { - ptr->cur_start_row = start_row; - } else { - /* use long arithmetic here to avoid overflow & unsigned problems */ - long ltemp; - - ltemp = (long) end_row - (long) ptr->rows_in_mem; - if (ltemp < 0) - ltemp = 0; /* don't fall off front end of file */ - ptr->cur_start_row = (JDIMENSION) ltemp; - } - /* Read in the selected part of the array. - * During the initial write pass, we will do no actual read - * because the selected part is all undefined. - */ - do_sarray_io(cinfo, ptr, FALSE); - } - /* Ensure the accessed part of the array is defined; prezero if needed. - * To improve locality of access, we only prezero the part of the array - * that the caller is about to access, not the entire in-memory array. - */ - if (ptr->first_undef_row < end_row) { - if (ptr->first_undef_row < start_row) { - if (writable) /* writer skipped over a section of array */ - ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS); - undef_row = start_row; /* but reader is allowed to read ahead */ - } else { - undef_row = ptr->first_undef_row; - } - if (writable) - ptr->first_undef_row = end_row; - if (ptr->pre_zero) { - size_t bytesperrow = (size_t) ptr->samplesperrow * SIZEOF(JSAMPLE); - undef_row -= ptr->cur_start_row; /* make indexes relative to buffer */ - end_row -= ptr->cur_start_row; - while (undef_row < end_row) { - FMEMZERO((void FAR *) ptr->mem_buffer[undef_row], bytesperrow); - undef_row++; - } - } else { - if (! writable) /* reader looking at undefined data */ - ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS); - } - } - /* Flag the buffer dirty if caller will write in it */ - if (writable) - ptr->dirty = TRUE; - /* Return address of proper part of the buffer */ - return ptr->mem_buffer + (start_row - ptr->cur_start_row); -} - - -METHODDEF(JBLOCKARRAY) -access_virt_barray (j_common_ptr cinfo, jvirt_barray_ptr ptr, - JDIMENSION start_row, JDIMENSION num_rows, - boolean writable) -/* Access the part of a virtual block array starting at start_row */ -/* and extending for num_rows rows. writable is true if */ -/* caller intends to modify the accessed area. */ -{ - JDIMENSION end_row = start_row + num_rows; - JDIMENSION undef_row; - - /* debugging check */ - if (end_row > ptr->rows_in_array || num_rows > ptr->maxaccess || - ptr->mem_buffer == NULL) - ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS); - - /* Make the desired part of the virtual array accessible */ - if (start_row < ptr->cur_start_row || - end_row > ptr->cur_start_row + ptr->rows_in_mem) { - if (! ptr->b_s_open) - ERREXIT(cinfo, JERR_VIRTUAL_BUG); - /* Flush old buffer contents if necessary */ - if (ptr->dirty) { - do_barray_io(cinfo, ptr, TRUE); - ptr->dirty = FALSE; - } - /* Decide what part of virtual array to access. - * Algorithm: if target address > current window, assume forward scan, - * load starting at target address. If target address < current window, - * assume backward scan, load so that target area is top of window. - * Note that when switching from forward write to forward read, will have - * start_row = 0, so the limiting case applies and we load from 0 anyway. - */ - if (start_row > ptr->cur_start_row) { - ptr->cur_start_row = start_row; - } else { - /* use long arithmetic here to avoid overflow & unsigned problems */ - long ltemp; - - ltemp = (long) end_row - (long) ptr->rows_in_mem; - if (ltemp < 0) - ltemp = 0; /* don't fall off front end of file */ - ptr->cur_start_row = (JDIMENSION) ltemp; - } - /* Read in the selected part of the array. - * During the initial write pass, we will do no actual read - * because the selected part is all undefined. - */ - do_barray_io(cinfo, ptr, FALSE); - } - /* Ensure the accessed part of the array is defined; prezero if needed. - * To improve locality of access, we only prezero the part of the array - * that the caller is about to access, not the entire in-memory array. - */ - if (ptr->first_undef_row < end_row) { - if (ptr->first_undef_row < start_row) { - if (writable) /* writer skipped over a section of array */ - ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS); - undef_row = start_row; /* but reader is allowed to read ahead */ - } else { - undef_row = ptr->first_undef_row; - } - if (writable) - ptr->first_undef_row = end_row; - if (ptr->pre_zero) { - size_t bytesperrow = (size_t) ptr->blocksperrow * SIZEOF(JBLOCK); - undef_row -= ptr->cur_start_row; /* make indexes relative to buffer */ - end_row -= ptr->cur_start_row; - while (undef_row < end_row) { - FMEMZERO((void FAR *) ptr->mem_buffer[undef_row], bytesperrow); - undef_row++; - } - } else { - if (! writable) /* reader looking at undefined data */ - ERREXIT(cinfo, JERR_BAD_VIRTUAL_ACCESS); - } - } - /* Flag the buffer dirty if caller will write in it */ - if (writable) - ptr->dirty = TRUE; - /* Return address of proper part of the buffer */ - return ptr->mem_buffer + (start_row - ptr->cur_start_row); -} - - -/* - * Release all objects belonging to a specified pool. - */ - -METHODDEF(void) -free_pool (j_common_ptr cinfo, int pool_id) -{ - my_mem_ptr mem = (my_mem_ptr) cinfo->mem; - small_pool_ptr shdr_ptr; - large_pool_ptr lhdr_ptr; - size_t space_freed; - - if (pool_id < 0 || pool_id >= JPOOL_NUMPOOLS) - ERREXIT1(cinfo, JERR_BAD_POOL_ID, pool_id); /* safety check */ - -#ifdef MEM_STATS - if (cinfo->err->trace_level > 1) - print_mem_stats(cinfo, pool_id); /* print pool's memory usage statistics */ -#endif - - /* If freeing IMAGE pool, close any virtual arrays first */ - if (pool_id == JPOOL_IMAGE) { - jvirt_sarray_ptr sptr; - jvirt_barray_ptr bptr; - - for (sptr = mem->virt_sarray_list; sptr != NULL; sptr = sptr->next) { - if (sptr->b_s_open) { /* there may be no backing store */ - sptr->b_s_open = FALSE; /* prevent recursive close if error */ - (*sptr->b_s_info.close_backing_store) (cinfo, & sptr->b_s_info); - } - } - mem->virt_sarray_list = NULL; - for (bptr = mem->virt_barray_list; bptr != NULL; bptr = bptr->next) { - if (bptr->b_s_open) { /* there may be no backing store */ - bptr->b_s_open = FALSE; /* prevent recursive close if error */ - (*bptr->b_s_info.close_backing_store) (cinfo, & bptr->b_s_info); - } - } - mem->virt_barray_list = NULL; - } - - /* Release large objects */ - lhdr_ptr = mem->large_list[pool_id]; - mem->large_list[pool_id] = NULL; - - while (lhdr_ptr != NULL) { - large_pool_ptr next_lhdr_ptr = lhdr_ptr->hdr.next; - space_freed = lhdr_ptr->hdr.bytes_used + - lhdr_ptr->hdr.bytes_left + - SIZEOF(large_pool_hdr); - jpeg_free_large(cinfo, (void FAR *) lhdr_ptr, space_freed); - mem->total_space_allocated -= space_freed; - lhdr_ptr = next_lhdr_ptr; - } - - /* Release small objects */ - shdr_ptr = mem->small_list[pool_id]; - mem->small_list[pool_id] = NULL; - - while (shdr_ptr != NULL) { - small_pool_ptr next_shdr_ptr = shdr_ptr->hdr.next; - space_freed = shdr_ptr->hdr.bytes_used + - shdr_ptr->hdr.bytes_left + - SIZEOF(small_pool_hdr); - jpeg_free_small(cinfo, (void *) shdr_ptr, space_freed); - mem->total_space_allocated -= space_freed; - shdr_ptr = next_shdr_ptr; - } -} - - -/* - * Close up shop entirely. - * Note that this cannot be called unless cinfo->mem is non-NULL. - */ - -METHODDEF(void) -self_destruct (j_common_ptr cinfo) -{ - int pool; - - /* Close all backing store, release all memory. - * Releasing pools in reverse order might help avoid fragmentation - * with some (brain-damaged) malloc libraries. - */ - for (pool = JPOOL_NUMPOOLS-1; pool >= JPOOL_PERMANENT; pool--) { - free_pool(cinfo, pool); - } - - /* Release the memory manager control block too. */ - jpeg_free_small(cinfo, (void *) cinfo->mem, SIZEOF(my_memory_mgr)); - cinfo->mem = NULL; /* ensures I will be called only once */ - - jpeg_mem_term(cinfo); /* system-dependent cleanup */ -} - - -/* - * Memory manager initialization. - * When this is called, only the error manager pointer is valid in cinfo! - */ - -GLOBAL(void) -jinit_memory_mgr (j_common_ptr cinfo) -{ - my_mem_ptr mem; - long max_to_use; - int pool; - size_t test_mac; - - cinfo->mem = NULL; /* for safety if init fails */ - - /* Check for configuration errors. - * SIZEOF(ALIGN_TYPE) should be a power of 2; otherwise, it probably - * doesn't reflect any real hardware alignment requirement. - * The test is a little tricky: for X>0, X and X-1 have no one-bits - * in common if and only if X is a power of 2, ie has only one one-bit. - * Some compilers may give an "unreachable code" warning here; ignore it. - */ - if ((SIZEOF(ALIGN_TYPE) & (SIZEOF(ALIGN_TYPE)-1)) != 0) - ERREXIT(cinfo, JERR_BAD_ALIGN_TYPE); - /* MAX_ALLOC_CHUNK must be representable as type size_t, and must be - * a multiple of SIZEOF(ALIGN_TYPE). - * Again, an "unreachable code" warning may be ignored here. - * But a "constant too large" warning means you need to fix MAX_ALLOC_CHUNK. - */ - test_mac = (size_t) MAX_ALLOC_CHUNK; - if ((long) test_mac != MAX_ALLOC_CHUNK || - (MAX_ALLOC_CHUNK % SIZEOF(ALIGN_TYPE)) != 0) - ERREXIT(cinfo, JERR_BAD_ALLOC_CHUNK); - - max_to_use = jpeg_mem_init(cinfo); /* system-dependent initialization */ - - /* Attempt to allocate memory manager's control block */ - mem = (my_mem_ptr) jpeg_get_small(cinfo, SIZEOF(my_memory_mgr)); - - if (mem == NULL) { - jpeg_mem_term(cinfo); /* system-dependent cleanup */ - ERREXIT1(cinfo, JERR_OUT_OF_MEMORY, 0); - } - - /* OK, fill in the method pointers */ - mem->pub.alloc_small = alloc_small; - mem->pub.alloc_large = alloc_large; - mem->pub.alloc_sarray = alloc_sarray; - mem->pub.alloc_barray = alloc_barray; - mem->pub.request_virt_sarray = request_virt_sarray; - mem->pub.request_virt_barray = request_virt_barray; - mem->pub.realize_virt_arrays = realize_virt_arrays; - mem->pub.access_virt_sarray = access_virt_sarray; - mem->pub.access_virt_barray = access_virt_barray; - mem->pub.free_pool = free_pool; - mem->pub.self_destruct = self_destruct; - - /* Make MAX_ALLOC_CHUNK accessible to other modules */ - mem->pub.max_alloc_chunk = MAX_ALLOC_CHUNK; - - /* Initialize working state */ - mem->pub.max_memory_to_use = max_to_use; - - for (pool = JPOOL_NUMPOOLS-1; pool >= JPOOL_PERMANENT; pool--) { - mem->small_list[pool] = NULL; - mem->large_list[pool] = NULL; - } - mem->virt_sarray_list = NULL; - mem->virt_barray_list = NULL; - - mem->total_space_allocated = SIZEOF(my_memory_mgr); - - /* Declare ourselves open for business */ - cinfo->mem = &mem->pub; - - /* Check for an environment variable JPEGMEM; if found, override the - * default max_memory setting from jpeg_mem_init. Note that the - * surrounding application may again override this value. - * If your system doesn't support getenv(), define NO_GETENV to disable - * this feature. - */ -#ifndef NO_GETENV - { char * memenv; - - if ((memenv = getenv("JPEGMEM")) != NULL) { - char ch = 'x'; - - if (sscanf(memenv, "%ld%c", &max_to_use, &ch) > 0) { - if (ch == 'm' || ch == 'M') - max_to_use *= 1000L; - mem->pub.max_memory_to_use = max_to_use * 1000L; - } - } - } -#endif - -} diff --git a/dep/libjpeg/src/jmemnobs.c b/dep/libjpeg/src/jmemnobs.c deleted file mode 100644 index a364cd822..000000000 --- a/dep/libjpeg/src/jmemnobs.c +++ /dev/null @@ -1,113 +0,0 @@ -/* - * jmemnobs.c - * - * Copyright (C) 1992-1996, Thomas G. Lane. - * Modified 2019 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file provides a really simple implementation of the system- - * dependent portion of the JPEG memory manager. This implementation - * assumes that no backing-store files are needed: all required space - * can be obtained from malloc(). - * This is very portable in the sense that it'll compile on almost anything, - * but you'd better have lots of main memory (or virtual memory) if you want - * to process big images. - * Note that the max_memory_to_use option is respected by this implementation. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" -#include "jmemsys.h" /* import the system-dependent declarations */ - -#ifndef HAVE_STDLIB_H /* should declare malloc(),free() */ -extern void * malloc JPP((size_t size)); -extern void free JPP((void *ptr)); -#endif - - -/* - * Memory allocation and freeing are controlled by the regular library - * routines malloc() and free(). - */ - -GLOBAL(void *) -jpeg_get_small (j_common_ptr cinfo, size_t sizeofobject) -{ - return (void *) malloc(sizeofobject); -} - -GLOBAL(void) -jpeg_free_small (j_common_ptr cinfo, void * object, size_t sizeofobject) -{ - free(object); -} - - -/* - * "Large" objects are treated the same as "small" ones. - * NB: although we include FAR keywords in the routine declarations, - * this file won't actually work in 80x86 small/medium model; at least, - * you probably won't be able to process useful-size images in only 64KB. - */ - -GLOBAL(void FAR *) -jpeg_get_large (j_common_ptr cinfo, size_t sizeofobject) -{ - return (void FAR *) malloc(sizeofobject); -} - -GLOBAL(void) -jpeg_free_large (j_common_ptr cinfo, void FAR * object, size_t sizeofobject) -{ - free(object); -} - - -/* - * This routine computes the total memory space available for allocation. - */ - -GLOBAL(long) -jpeg_mem_available (j_common_ptr cinfo, long min_bytes_needed, - long max_bytes_needed, long already_allocated) -{ - if (cinfo->mem->max_memory_to_use) - return cinfo->mem->max_memory_to_use - already_allocated; - - /* Here we say, "we got all you want bud!" */ - return max_bytes_needed; -} - - -/* - * Backing store (temporary file) management. - * Since jpeg_mem_available always promised the moon, - * this should never be called and we can just error out. - */ - -GLOBAL(void) -jpeg_open_backing_store (j_common_ptr cinfo, backing_store_ptr info, - long total_bytes_needed) -{ - ERREXIT(cinfo, JERR_NO_BACKING_STORE); -} - - -/* - * These routines take care of any system-dependent initialization and - * cleanup required. Here, there isn't any. - */ - -GLOBAL(long) -jpeg_mem_init (j_common_ptr cinfo) -{ - return 0; /* just set max_memory_to_use to 0 */ -} - -GLOBAL(void) -jpeg_mem_term (j_common_ptr cinfo) -{ - /* no work */ -} diff --git a/dep/libjpeg/src/jmemsys.h b/dep/libjpeg/src/jmemsys.h deleted file mode 100644 index 6c3c6d348..000000000 --- a/dep/libjpeg/src/jmemsys.h +++ /dev/null @@ -1,198 +0,0 @@ -/* - * jmemsys.h - * - * Copyright (C) 1992-1997, Thomas G. Lane. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This include file defines the interface between the system-independent - * and system-dependent portions of the JPEG memory manager. No other - * modules need include it. (The system-independent portion is jmemmgr.c; - * there are several different versions of the system-dependent portion.) - * - * This file works as-is for the system-dependent memory managers supplied - * in the IJG distribution. You may need to modify it if you write a - * custom memory manager. If system-dependent changes are needed in - * this file, the best method is to #ifdef them based on a configuration - * symbol supplied in jconfig.h, as we have done with USE_MSDOS_MEMMGR - * and USE_MAC_MEMMGR. - */ - - -/* Short forms of external names for systems with brain-damaged linkers. */ - -#ifdef NEED_SHORT_EXTERNAL_NAMES -#define jpeg_get_small jGetSmall -#define jpeg_free_small jFreeSmall -#define jpeg_get_large jGetLarge -#define jpeg_free_large jFreeLarge -#define jpeg_mem_available jMemAvail -#define jpeg_open_backing_store jOpenBackStore -#define jpeg_mem_init jMemInit -#define jpeg_mem_term jMemTerm -#endif /* NEED_SHORT_EXTERNAL_NAMES */ - - -/* - * These two functions are used to allocate and release small chunks of - * memory. (Typically the total amount requested through jpeg_get_small is - * no more than 20K or so; this will be requested in chunks of a few K each.) - * Behavior should be the same as for the standard library functions malloc - * and free; in particular, jpeg_get_small must return NULL on failure. - * On most systems, these ARE malloc and free. jpeg_free_small is passed the - * size of the object being freed, just in case it's needed. - * On an 80x86 machine using small-data memory model, these manage near heap. - */ - -EXTERN(void *) jpeg_get_small JPP((j_common_ptr cinfo, size_t sizeofobject)); -EXTERN(void) jpeg_free_small JPP((j_common_ptr cinfo, void * object, - size_t sizeofobject)); - -/* - * These two functions are used to allocate and release large chunks of - * memory (up to the total free space designated by jpeg_mem_available). - * The interface is the same as above, except that on an 80x86 machine, - * far pointers are used. On most other machines these are identical to - * the jpeg_get/free_small routines; but we keep them separate anyway, - * in case a different allocation strategy is desirable for large chunks. - */ - -EXTERN(void FAR *) jpeg_get_large JPP((j_common_ptr cinfo, - size_t sizeofobject)); -EXTERN(void) jpeg_free_large JPP((j_common_ptr cinfo, void FAR * object, - size_t sizeofobject)); - -/* - * The macro MAX_ALLOC_CHUNK designates the maximum number of bytes that may - * be requested in a single call to jpeg_get_large (and jpeg_get_small for that - * matter, but that case should never come into play). This macro is needed - * to model the 64Kb-segment-size limit of far addressing on 80x86 machines. - * On those machines, we expect that jconfig.h will provide a proper value. - * On machines with 32-bit flat address spaces, any large constant may be used. - * - * NB: jmemmgr.c expects that MAX_ALLOC_CHUNK will be representable as type - * size_t and will be a multiple of sizeof(align_type). - */ - -#ifndef MAX_ALLOC_CHUNK /* may be overridden in jconfig.h */ -#define MAX_ALLOC_CHUNK 1000000000L -#endif - -/* - * This routine computes the total space still available for allocation by - * jpeg_get_large. If more space than this is needed, backing store will be - * used. NOTE: any memory already allocated must not be counted. - * - * There is a minimum space requirement, corresponding to the minimum - * feasible buffer sizes; jmemmgr.c will request that much space even if - * jpeg_mem_available returns zero. The maximum space needed, enough to hold - * all working storage in memory, is also passed in case it is useful. - * Finally, the total space already allocated is passed. If no better - * method is available, cinfo->mem->max_memory_to_use - already_allocated - * is often a suitable calculation. - * - * It is OK for jpeg_mem_available to underestimate the space available - * (that'll just lead to more backing-store access than is really necessary). - * However, an overestimate will lead to failure. Hence it's wise to subtract - * a slop factor from the true available space. 5% should be enough. - * - * On machines with lots of virtual memory, any large constant may be returned. - * Conversely, zero may be returned to always use the minimum amount of memory. - */ - -EXTERN(long) jpeg_mem_available JPP((j_common_ptr cinfo, - long min_bytes_needed, - long max_bytes_needed, - long already_allocated)); - - -/* - * This structure holds whatever state is needed to access a single - * backing-store object. The read/write/close method pointers are called - * by jmemmgr.c to manipulate the backing-store object; all other fields - * are private to the system-dependent backing store routines. - */ - -#define TEMP_NAME_LENGTH 64 /* max length of a temporary file's name */ - - -#ifdef USE_MSDOS_MEMMGR /* DOS-specific junk */ - -typedef unsigned short XMSH; /* type of extended-memory handles */ -typedef unsigned short EMSH; /* type of expanded-memory handles */ - -typedef union { - short file_handle; /* DOS file handle if it's a temp file */ - XMSH xms_handle; /* handle if it's a chunk of XMS */ - EMSH ems_handle; /* handle if it's a chunk of EMS */ -} handle_union; - -#endif /* USE_MSDOS_MEMMGR */ - -#ifdef USE_MAC_MEMMGR /* Mac-specific junk */ -#include -#endif /* USE_MAC_MEMMGR */ - - -typedef struct backing_store_struct * backing_store_ptr; - -typedef struct backing_store_struct { - /* Methods for reading/writing/closing this backing-store object */ - JMETHOD(void, read_backing_store, (j_common_ptr cinfo, - backing_store_ptr info, - void FAR * buffer_address, - long file_offset, long byte_count)); - JMETHOD(void, write_backing_store, (j_common_ptr cinfo, - backing_store_ptr info, - void FAR * buffer_address, - long file_offset, long byte_count)); - JMETHOD(void, close_backing_store, (j_common_ptr cinfo, - backing_store_ptr info)); - - /* Private fields for system-dependent backing-store management */ -#ifdef USE_MSDOS_MEMMGR - /* For the MS-DOS manager (jmemdos.c), we need: */ - handle_union handle; /* reference to backing-store storage object */ - char temp_name[TEMP_NAME_LENGTH]; /* name if it's a file */ -#else -#ifdef USE_MAC_MEMMGR - /* For the Mac manager (jmemmac.c), we need: */ - short temp_file; /* file reference number to temp file */ - FSSpec tempSpec; /* the FSSpec for the temp file */ - char temp_name[TEMP_NAME_LENGTH]; /* name if it's a file */ -#else - /* For a typical implementation with temp files, we need: */ - FILE * temp_file; /* stdio reference to temp file */ - char temp_name[TEMP_NAME_LENGTH]; /* name of temp file */ -#endif -#endif -} backing_store_info; - - -/* - * Initial opening of a backing-store object. This must fill in the - * read/write/close pointers in the object. The read/write routines - * may take an error exit if the specified maximum file size is exceeded. - * (If jpeg_mem_available always returns a large value, this routine can - * just take an error exit.) - */ - -EXTERN(void) jpeg_open_backing_store JPP((j_common_ptr cinfo, - backing_store_ptr info, - long total_bytes_needed)); - - -/* - * These routines take care of any system-dependent initialization and - * cleanup required. jpeg_mem_init will be called before anything is - * allocated (and, therefore, nothing in cinfo is of use except the error - * manager pointer). It should return a suitable default value for - * max_memory_to_use; this may subsequently be overridden by the surrounding - * application. (Note that max_memory_to_use is only important if - * jpeg_mem_available chooses to consult it ... no one else will.) - * jpeg_mem_term may assume that all requested memory has been freed and that - * all opened backing-store objects have been closed. - */ - -EXTERN(long) jpeg_mem_init JPP((j_common_ptr cinfo)); -EXTERN(void) jpeg_mem_term JPP((j_common_ptr cinfo)); diff --git a/dep/libjpeg/src/jquant1.c b/dep/libjpeg/src/jquant1.c deleted file mode 100644 index 60b1843e7..000000000 --- a/dep/libjpeg/src/jquant1.c +++ /dev/null @@ -1,851 +0,0 @@ -/* - * jquant1.c - * - * Copyright (C) 1991-1996, Thomas G. Lane. - * Modified 2011-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains 1-pass color quantization (color mapping) routines. - * These routines provide mapping to a fixed color map using equally spaced - * color values. Optional Floyd-Steinberg or ordered dithering is available. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - -#ifdef QUANT_1PASS_SUPPORTED - - -/* - * The main purpose of 1-pass quantization is to provide a fast, if not very - * high quality, colormapped output capability. A 2-pass quantizer usually - * gives better visual quality; however, for quantized grayscale output this - * quantizer is perfectly adequate. Dithering is highly recommended with this - * quantizer, though you can turn it off if you really want to. - * - * In 1-pass quantization the colormap must be chosen in advance of seeing the - * image. We use a map consisting of all combinations of Ncolors[i] color - * values for the i'th component. The Ncolors[] values are chosen so that - * their product, the total number of colors, is no more than that requested. - * (In most cases, the product will be somewhat less.) - * - * Since the colormap is orthogonal, the representative value for each color - * component can be determined without considering the other components; - * then these indexes can be combined into a colormap index by a standard - * N-dimensional-array-subscript calculation. Most of the arithmetic involved - * can be precalculated and stored in the lookup table colorindex[]. - * colorindex[i][j] maps pixel value j in component i to the nearest - * representative value (grid plane) for that component; this index is - * multiplied by the array stride for component i, so that the - * index of the colormap entry closest to a given pixel value is just - * sum( colorindex[component-number][pixel-component-value] ) - * Aside from being fast, this scheme allows for variable spacing between - * representative values with no additional lookup cost. - * - * If gamma correction has been applied in color conversion, it might be wise - * to adjust the color grid spacing so that the representative colors are - * equidistant in linear space. At this writing, gamma correction is not - * implemented by jdcolor, so nothing is done here. - */ - - -/* Declarations for ordered dithering. - * - * We use a standard 16x16 ordered dither array. The basic concept of ordered - * dithering is described in many references, for instance Dale Schumacher's - * chapter II.2 of Graphics Gems II (James Arvo, ed. Academic Press, 1991). - * In place of Schumacher's comparisons against a "threshold" value, we add a - * "dither" value to the input pixel and then round the result to the nearest - * output value. The dither value is equivalent to (0.5 - threshold) times - * the distance between output values. For ordered dithering, we assume that - * the output colors are equally spaced; if not, results will probably be - * worse, since the dither may be too much or too little at a given point. - * - * The normal calculation would be to form pixel value + dither, range-limit - * this to 0..MAXJSAMPLE, and then index into the colorindex table as usual. - * We can skip the separate range-limiting step by extending the colorindex - * table in both directions. - */ - -#define ODITHER_SIZE 16 /* dimension of dither matrix */ -/* NB: if ODITHER_SIZE is not a power of 2, ODITHER_MASK uses will break */ -#define ODITHER_CELLS (ODITHER_SIZE*ODITHER_SIZE) /* # cells in matrix */ -#define ODITHER_MASK (ODITHER_SIZE-1) /* mask for wrapping around counters */ - -typedef int ODITHER_MATRIX[ODITHER_SIZE][ODITHER_SIZE]; -typedef int (*ODITHER_MATRIX_PTR)[ODITHER_SIZE]; - -static const UINT8 base_dither_matrix[ODITHER_SIZE][ODITHER_SIZE] = { - /* Bayer's order-4 dither array. Generated by the code given in - * Stephen Hawley's article "Ordered Dithering" in Graphics Gems I. - * The values in this array must range from 0 to ODITHER_CELLS-1. - */ - { 0,192, 48,240, 12,204, 60,252, 3,195, 51,243, 15,207, 63,255 }, - { 128, 64,176,112,140, 76,188,124,131, 67,179,115,143, 79,191,127 }, - { 32,224, 16,208, 44,236, 28,220, 35,227, 19,211, 47,239, 31,223 }, - { 160, 96,144, 80,172,108,156, 92,163, 99,147, 83,175,111,159, 95 }, - { 8,200, 56,248, 4,196, 52,244, 11,203, 59,251, 7,199, 55,247 }, - { 136, 72,184,120,132, 68,180,116,139, 75,187,123,135, 71,183,119 }, - { 40,232, 24,216, 36,228, 20,212, 43,235, 27,219, 39,231, 23,215 }, - { 168,104,152, 88,164,100,148, 84,171,107,155, 91,167,103,151, 87 }, - { 2,194, 50,242, 14,206, 62,254, 1,193, 49,241, 13,205, 61,253 }, - { 130, 66,178,114,142, 78,190,126,129, 65,177,113,141, 77,189,125 }, - { 34,226, 18,210, 46,238, 30,222, 33,225, 17,209, 45,237, 29,221 }, - { 162, 98,146, 82,174,110,158, 94,161, 97,145, 81,173,109,157, 93 }, - { 10,202, 58,250, 6,198, 54,246, 9,201, 57,249, 5,197, 53,245 }, - { 138, 74,186,122,134, 70,182,118,137, 73,185,121,133, 69,181,117 }, - { 42,234, 26,218, 38,230, 22,214, 41,233, 25,217, 37,229, 21,213 }, - { 170,106,154, 90,166,102,150, 86,169,105,153, 89,165,101,149, 85 } -}; - - -/* Declarations for Floyd-Steinberg dithering. - * - * Errors are accumulated into the array fserrors[], at a resolution of - * 1/16th of a pixel count. The error at a given pixel is propagated - * to its not-yet-processed neighbors using the standard F-S fractions, - * ... (here) 7/16 - * 3/16 5/16 1/16 - * We work left-to-right on even rows, right-to-left on odd rows. - * - * We can get away with a single array (holding one row's worth of errors) - * by using it to store the current row's errors at pixel columns not yet - * processed, but the next row's errors at columns already processed. We - * need only a few extra variables to hold the errors immediately around the - * current column. (If we are lucky, those variables are in registers, but - * even if not, they're probably cheaper to access than array elements are.) - * - * The fserrors[] array is indexed [component#][position]. - * We provide (#columns + 2) entries per component; the extra entry at each - * end saves us from special-casing the first and last pixels. - * - * Note: on a wide image, we might not have enough room in a PC's near data - * segment to hold the error array; so it is allocated with alloc_large. - */ - -#if BITS_IN_JSAMPLE == 8 -typedef INT16 FSERROR; /* 16 bits should be enough */ -typedef int LOCFSERROR; /* use 'int' for calculation temps */ -#else -typedef INT32 FSERROR; /* may need more than 16 bits */ -typedef INT32 LOCFSERROR; /* be sure calculation temps are big enough */ -#endif - -typedef FSERROR FAR *FSERRPTR; /* pointer to error array (in FAR storage!) */ - - -/* Private subobject */ - -#define MAX_Q_COMPS 4 /* max components I can handle */ - -typedef struct { - struct jpeg_color_quantizer pub; /* public fields */ - - /* Initially allocated colormap is saved here */ - JSAMPARRAY sv_colormap; /* The color map as a 2-D pixel array */ - int sv_actual; /* number of entries in use */ - - JSAMPARRAY colorindex; /* Precomputed mapping for speed */ - /* colorindex[i][j] = index of color closest to pixel value j in component i, - * premultiplied as described above. Since colormap indexes must fit into - * JSAMPLEs, the entries of this array will too. - */ - boolean is_padded; /* is the colorindex padded for odither? */ - - int Ncolors[MAX_Q_COMPS]; /* # of values alloced to each component */ - - /* Variables for ordered dithering */ - int row_index; /* cur row's vertical index in dither matrix */ - ODITHER_MATRIX_PTR odither[MAX_Q_COMPS]; /* one dither array per component */ - - /* Variables for Floyd-Steinberg dithering */ - FSERRPTR fserrors[MAX_Q_COMPS]; /* accumulated errors */ - boolean on_odd_row; /* flag to remember which row we are on */ -} my_cquantizer; - -typedef my_cquantizer * my_cquantize_ptr; - - -/* - * Policy-making subroutines for create_colormap and create_colorindex. - * These routines determine the colormap to be used. The rest of the module - * only assumes that the colormap is orthogonal. - * - * * select_ncolors decides how to divvy up the available colors - * among the components. - * * output_value defines the set of representative values for a component. - * * largest_input_value defines the mapping from input values to - * representative values for a component. - * Note that the latter two routines may impose different policies for - * different components, though this is not currently done. - */ - - -LOCAL(int) -select_ncolors (j_decompress_ptr cinfo, int Ncolors[]) -/* Determine allocation of desired colors to components, */ -/* and fill in Ncolors[] array to indicate choice. */ -/* Return value is total number of colors (product of Ncolors[] values). */ -{ - int nc = cinfo->out_color_components; /* number of color components */ - int max_colors = cinfo->desired_number_of_colors; - int total_colors, iroot, i, j; - boolean changed; - long temp; - static const int RGB_order[3] = { RGB_GREEN, RGB_RED, RGB_BLUE }; - - /* We can allocate at least the nc'th root of max_colors per component. */ - /* Compute floor(nc'th root of max_colors). */ - iroot = 1; - do { - iroot++; - temp = iroot; /* set temp = iroot ** nc */ - for (i = 1; i < nc; i++) - temp *= iroot; - } while (temp <= (long) max_colors); /* repeat till iroot exceeds root */ - iroot--; /* now iroot = floor(root) */ - - /* Must have at least 2 color values per component */ - if (iroot < 2) - ERREXIT1(cinfo, JERR_QUANT_FEW_COLORS, (int) temp); - - /* Initialize to iroot color values for each component */ - total_colors = 1; - for (i = 0; i < nc; i++) { - Ncolors[i] = iroot; - total_colors *= iroot; - } - /* We may be able to increment the count for one or more components without - * exceeding max_colors, though we know not all can be incremented. - * Sometimes, the first component can be incremented more than once! - * (Example: for 16 colors, we start at 2*2*2, go to 3*2*2, then 4*2*2.) - * In RGB colorspace, try to increment G first, then R, then B. - */ - do { - changed = FALSE; - for (i = 0; i < nc; i++) { - j = (cinfo->out_color_space == JCS_RGB ? RGB_order[i] : i); - /* calculate new total_colors if Ncolors[j] is incremented */ - temp = total_colors / Ncolors[j]; - temp *= Ncolors[j]+1; /* done in long arith to avoid oflo */ - if (temp > (long) max_colors) - break; /* won't fit, done with this pass */ - Ncolors[j]++; /* OK, apply the increment */ - total_colors = (int) temp; - changed = TRUE; - } - } while (changed); - - return total_colors; -} - - -LOCAL(int) -output_value (j_decompress_ptr cinfo, int ci, int j, int maxj) -/* Return j'th output value, where j will range from 0 to maxj */ -/* The output values must fall in 0..MAXJSAMPLE in increasing order */ -{ - /* We always provide values 0 and MAXJSAMPLE for each component; - * any additional values are equally spaced between these limits. - * (Forcing the upper and lower values to the limits ensures that - * dithering can't produce a color outside the selected gamut.) - */ - return (int) (((INT32) j * MAXJSAMPLE + maxj/2) / maxj); -} - - -LOCAL(int) -largest_input_value (j_decompress_ptr cinfo, int ci, int j, int maxj) -/* Return largest input value that should map to j'th output value */ -/* Must have largest(j=0) >= 0, and largest(j=maxj) >= MAXJSAMPLE */ -{ - /* Breakpoints are halfway between values returned by output_value */ - return (int) (((INT32) (2*j + 1) * MAXJSAMPLE + maxj) / (2*maxj)); -} - - -/* - * Create the colormap. - */ - -LOCAL(void) -create_colormap (j_decompress_ptr cinfo) -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - JSAMPARRAY colormap; /* Created colormap */ - int total_colors; /* Number of distinct output colors */ - int i,j,k, nci, blksize, blkdist, ptr, val; - - /* Select number of colors for each component */ - total_colors = select_ncolors(cinfo, cquantize->Ncolors); - - /* Report selected color counts */ - if (cinfo->out_color_components == 3) - TRACEMS4(cinfo, 1, JTRC_QUANT_3_NCOLORS, - total_colors, cquantize->Ncolors[0], - cquantize->Ncolors[1], cquantize->Ncolors[2]); - else - TRACEMS1(cinfo, 1, JTRC_QUANT_NCOLORS, total_colors); - - /* Allocate and fill in the colormap. */ - /* The colors are ordered in the map in standard row-major order, */ - /* i.e. rightmost (highest-indexed) color changes most rapidly. */ - - colormap = (*cinfo->mem->alloc_sarray) ((j_common_ptr) cinfo, JPOOL_IMAGE, - (JDIMENSION) total_colors, (JDIMENSION) cinfo->out_color_components); - - /* blksize is number of adjacent repeated entries for a component */ - /* blkdist is distance between groups of identical entries for a component */ - blkdist = total_colors; - - for (i = 0; i < cinfo->out_color_components; i++) { - /* fill in colormap entries for i'th color component */ - nci = cquantize->Ncolors[i]; /* # of distinct values for this color */ - blksize = blkdist / nci; - for (j = 0; j < nci; j++) { - /* Compute j'th output value (out of nci) for component */ - val = output_value(cinfo, i, j, nci-1); - /* Fill in all colormap entries that have this value of this component */ - for (ptr = j * blksize; ptr < total_colors; ptr += blkdist) { - /* fill in blksize entries beginning at ptr */ - for (k = 0; k < blksize; k++) - colormap[i][ptr+k] = (JSAMPLE) val; - } - } - blkdist = blksize; /* blksize of this color is blkdist of next */ - } - - /* Save the colormap in private storage, - * where it will survive color quantization mode changes. - */ - cquantize->sv_colormap = colormap; - cquantize->sv_actual = total_colors; -} - - -/* - * Create the color index table. - */ - -LOCAL(void) -create_colorindex (j_decompress_ptr cinfo) -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - JSAMPROW indexptr; - int i,j,k, nci, blksize, val, pad; - - /* For ordered dither, we pad the color index tables by MAXJSAMPLE in - * each direction (input index values can be -MAXJSAMPLE .. 2*MAXJSAMPLE). - * This is not necessary in the other dithering modes. However, we - * flag whether it was done in case user changes dithering mode. - */ - if (cinfo->dither_mode == JDITHER_ORDERED) { - pad = MAXJSAMPLE*2; - cquantize->is_padded = TRUE; - } else { - pad = 0; - cquantize->is_padded = FALSE; - } - - cquantize->colorindex = (*cinfo->mem->alloc_sarray) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - (JDIMENSION) (MAXJSAMPLE+1 + pad), - (JDIMENSION) cinfo->out_color_components); - - /* blksize is number of adjacent repeated entries for a component */ - blksize = cquantize->sv_actual; - - for (i = 0; i < cinfo->out_color_components; i++) { - /* fill in colorindex entries for i'th color component */ - nci = cquantize->Ncolors[i]; /* # of distinct values for this color */ - blksize = blksize / nci; - - /* adjust colorindex pointers to provide padding at negative indexes. */ - if (pad) - cquantize->colorindex[i] += MAXJSAMPLE; - - /* in loop, val = index of current output value, */ - /* and k = largest j that maps to current val */ - indexptr = cquantize->colorindex[i]; - val = 0; - k = largest_input_value(cinfo, i, 0, nci-1); - for (j = 0; j <= MAXJSAMPLE; j++) { - while (j > k) /* advance val if past boundary */ - k = largest_input_value(cinfo, i, ++val, nci-1); - /* premultiply so that no multiplication needed in main processing */ - indexptr[j] = (JSAMPLE) (val * blksize); - } - /* Pad at both ends if necessary */ - if (pad) - for (j = 1; j <= MAXJSAMPLE; j++) { - indexptr[-j] = indexptr[0]; - indexptr[MAXJSAMPLE+j] = indexptr[MAXJSAMPLE]; - } - } -} - - -/* - * Create an ordered-dither array for a component having ncolors - * distinct output values. - */ - -LOCAL(ODITHER_MATRIX_PTR) -make_odither_array (j_decompress_ptr cinfo, int ncolors) -{ - ODITHER_MATRIX_PTR odither; - int j,k; - INT32 num,den; - - odither = (ODITHER_MATRIX_PTR) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(ODITHER_MATRIX)); - /* The inter-value distance for this color is MAXJSAMPLE/(ncolors-1). - * Hence the dither value for the matrix cell with fill order f - * (f=0..N-1) should be (N-1-2*f)/(2*N) * MAXJSAMPLE/(ncolors-1). - * On 16-bit-int machine, be careful to avoid overflow. - */ - den = 2 * ODITHER_CELLS * ((INT32) (ncolors - 1)); - for (j = 0; j < ODITHER_SIZE; j++) { - for (k = 0; k < ODITHER_SIZE; k++) { - num = ((INT32) (ODITHER_CELLS-1 - 2*((int)base_dither_matrix[j][k]))) - * MAXJSAMPLE; - /* Ensure round towards zero despite C's lack of consistency - * about rounding negative values in integer division... - */ - odither[j][k] = (int) (num<0 ? -((-num)/den) : num/den); - } - } - return odither; -} - - -/* - * Create the ordered-dither tables. - * Components having the same number of representative colors may - * share a dither table. - */ - -LOCAL(void) -create_odither_tables (j_decompress_ptr cinfo) -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - ODITHER_MATRIX_PTR odither; - int i, j, nci; - - for (i = 0; i < cinfo->out_color_components; i++) { - nci = cquantize->Ncolors[i]; /* # of distinct values for this color */ - odither = NULL; /* search for matching prior component */ - for (j = 0; j < i; j++) { - if (nci == cquantize->Ncolors[j]) { - odither = cquantize->odither[j]; - break; - } - } - if (odither == NULL) /* need a new table? */ - odither = make_odither_array(cinfo, nci); - cquantize->odither[i] = odither; - } -} - - -/* - * Map some rows of pixels to the output colormapped representation. - */ - -METHODDEF(void) -color_quantize (j_decompress_ptr cinfo, JSAMPARRAY input_buf, - JSAMPARRAY output_buf, int num_rows) -/* General case, no dithering */ -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - JSAMPARRAY colorindex = cquantize->colorindex; - register int pixcode, ci; - register JSAMPROW ptrin, ptrout; - int row; - JDIMENSION col; - JDIMENSION width = cinfo->output_width; - register int nc = cinfo->out_color_components; - - for (row = 0; row < num_rows; row++) { - ptrin = input_buf[row]; - ptrout = output_buf[row]; - for (col = width; col > 0; col--) { - pixcode = 0; - for (ci = 0; ci < nc; ci++) { - pixcode += GETJSAMPLE(colorindex[ci][GETJSAMPLE(*ptrin++)]); - } - *ptrout++ = (JSAMPLE) pixcode; - } - } -} - - -METHODDEF(void) -color_quantize3 (j_decompress_ptr cinfo, JSAMPARRAY input_buf, - JSAMPARRAY output_buf, int num_rows) -/* Fast path for out_color_components==3, no dithering */ -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - register int pixcode; - register JSAMPROW ptrin, ptrout; - JSAMPROW colorindex0 = cquantize->colorindex[0]; - JSAMPROW colorindex1 = cquantize->colorindex[1]; - JSAMPROW colorindex2 = cquantize->colorindex[2]; - int row; - JDIMENSION col; - JDIMENSION width = cinfo->output_width; - - for (row = 0; row < num_rows; row++) { - ptrin = input_buf[row]; - ptrout = output_buf[row]; - for (col = width; col > 0; col--) { - pixcode = GETJSAMPLE(colorindex0[GETJSAMPLE(*ptrin++)]); - pixcode += GETJSAMPLE(colorindex1[GETJSAMPLE(*ptrin++)]); - pixcode += GETJSAMPLE(colorindex2[GETJSAMPLE(*ptrin++)]); - *ptrout++ = (JSAMPLE) pixcode; - } - } -} - - -METHODDEF(void) -quantize_ord_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf, - JSAMPARRAY output_buf, int num_rows) -/* General case, with ordered dithering */ -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - register JSAMPROW input_ptr; - register JSAMPROW output_ptr; - JSAMPROW colorindex_ci; - int * dither; /* points to active row of dither matrix */ - int row_index, col_index; /* current indexes into dither matrix */ - int nc = cinfo->out_color_components; - int ci; - int row; - JDIMENSION col; - JDIMENSION width = cinfo->output_width; - - for (row = 0; row < num_rows; row++) { - /* Initialize output values to 0 so can process components separately */ - FMEMZERO((void FAR *) output_buf[row], (size_t) width * SIZEOF(JSAMPLE)); - row_index = cquantize->row_index; - for (ci = 0; ci < nc; ci++) { - input_ptr = input_buf[row] + ci; - output_ptr = output_buf[row]; - colorindex_ci = cquantize->colorindex[ci]; - dither = cquantize->odither[ci][row_index]; - col_index = 0; - - for (col = width; col > 0; col--) { - /* Form pixel value + dither, range-limit to 0..MAXJSAMPLE, - * select output value, accumulate into output code for this pixel. - * Range-limiting need not be done explicitly, as we have extended - * the colorindex table to produce the right answers for out-of-range - * inputs. The maximum dither is +- MAXJSAMPLE; this sets the - * required amount of padding. - */ - *output_ptr += colorindex_ci[GETJSAMPLE(*input_ptr)+dither[col_index]]; - input_ptr += nc; - output_ptr++; - col_index = (col_index + 1) & ODITHER_MASK; - } - } - /* Advance row index for next row */ - row_index = (row_index + 1) & ODITHER_MASK; - cquantize->row_index = row_index; - } -} - - -METHODDEF(void) -quantize3_ord_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf, - JSAMPARRAY output_buf, int num_rows) -/* Fast path for out_color_components==3, with ordered dithering */ -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - register int pixcode; - register JSAMPROW input_ptr; - register JSAMPROW output_ptr; - JSAMPROW colorindex0 = cquantize->colorindex[0]; - JSAMPROW colorindex1 = cquantize->colorindex[1]; - JSAMPROW colorindex2 = cquantize->colorindex[2]; - int * dither0; /* points to active row of dither matrix */ - int * dither1; - int * dither2; - int row_index, col_index; /* current indexes into dither matrix */ - int row; - JDIMENSION col; - JDIMENSION width = cinfo->output_width; - - for (row = 0; row < num_rows; row++) { - row_index = cquantize->row_index; - input_ptr = input_buf[row]; - output_ptr = output_buf[row]; - dither0 = cquantize->odither[0][row_index]; - dither1 = cquantize->odither[1][row_index]; - dither2 = cquantize->odither[2][row_index]; - col_index = 0; - - for (col = width; col > 0; col--) { - pixcode = GETJSAMPLE(colorindex0[GETJSAMPLE(*input_ptr++) + - dither0[col_index]]); - pixcode += GETJSAMPLE(colorindex1[GETJSAMPLE(*input_ptr++) + - dither1[col_index]]); - pixcode += GETJSAMPLE(colorindex2[GETJSAMPLE(*input_ptr++) + - dither2[col_index]]); - *output_ptr++ = (JSAMPLE) pixcode; - col_index = (col_index + 1) & ODITHER_MASK; - } - row_index = (row_index + 1) & ODITHER_MASK; - cquantize->row_index = row_index; - } -} - - -METHODDEF(void) -quantize_fs_dither (j_decompress_ptr cinfo, JSAMPARRAY input_buf, - JSAMPARRAY output_buf, int num_rows) -/* General case, with Floyd-Steinberg dithering */ -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - register LOCFSERROR cur; /* current error or pixel value */ - LOCFSERROR belowerr; /* error for pixel below cur */ - LOCFSERROR bpreverr; /* error for below/prev col */ - LOCFSERROR bnexterr; /* error for below/next col */ - LOCFSERROR delta; - register FSERRPTR errorptr; /* => fserrors[] at column before current */ - register JSAMPROW input_ptr; - register JSAMPROW output_ptr; - JSAMPROW colorindex_ci; - JSAMPROW colormap_ci; - int pixcode; - int nc = cinfo->out_color_components; - int dir; /* 1 for left-to-right, -1 for right-to-left */ - int dirnc; /* dir * nc */ - int ci; - int row; - JDIMENSION col; - JDIMENSION width = cinfo->output_width; - JSAMPLE *range_limit = cinfo->sample_range_limit; - SHIFT_TEMPS - - for (row = 0; row < num_rows; row++) { - /* Initialize output values to 0 so can process components separately */ - FMEMZERO((void FAR *) output_buf[row], (size_t) width * SIZEOF(JSAMPLE)); - for (ci = 0; ci < nc; ci++) { - input_ptr = input_buf[row] + ci; - output_ptr = output_buf[row]; - if (cquantize->on_odd_row) { - /* work right to left in this row */ - input_ptr += (width-1) * nc; /* so point to rightmost pixel */ - output_ptr += width-1; - dir = -1; - dirnc = -nc; - errorptr = cquantize->fserrors[ci] + (width+1); /* => entry after last column */ - } else { - /* work left to right in this row */ - dir = 1; - dirnc = nc; - errorptr = cquantize->fserrors[ci]; /* => entry before first column */ - } - colorindex_ci = cquantize->colorindex[ci]; - colormap_ci = cquantize->sv_colormap[ci]; - /* Preset error values: no error propagated to first pixel from left */ - cur = 0; - /* and no error propagated to row below yet */ - belowerr = bpreverr = 0; - - for (col = width; col > 0; col--) { - /* cur holds the error propagated from the previous pixel on the - * current line. Add the error propagated from the previous line - * to form the complete error correction term for this pixel, and - * round the error term (which is expressed * 16) to an integer. - * RIGHT_SHIFT rounds towards minus infinity, so adding 8 is correct - * for either sign of the error value. - * Note: errorptr points to *previous* column's array entry. - */ - cur = RIGHT_SHIFT(cur + errorptr[dir] + 8, 4); - /* Form pixel value + error, and range-limit to 0..MAXJSAMPLE. - * The maximum error is +- MAXJSAMPLE; this sets the required size - * of the range_limit array. - */ - cur += GETJSAMPLE(*input_ptr); - cur = GETJSAMPLE(range_limit[cur]); - /* Select output value, accumulate into output code for this pixel */ - pixcode = GETJSAMPLE(colorindex_ci[cur]); - *output_ptr += (JSAMPLE) pixcode; - /* Compute actual representation error at this pixel */ - /* Note: we can do this even though we don't have the final */ - /* pixel code, because the colormap is orthogonal. */ - cur -= GETJSAMPLE(colormap_ci[pixcode]); - /* Compute error fractions to be propagated to adjacent pixels. - * Add these into the running sums, and simultaneously shift the - * next-line error sums left by 1 column. - */ - bnexterr = cur; - delta = cur * 2; - cur += delta; /* form error * 3 */ - errorptr[0] = (FSERROR) (bpreverr + cur); - cur += delta; /* form error * 5 */ - bpreverr = belowerr + cur; - belowerr = bnexterr; - cur += delta; /* form error * 7 */ - /* At this point cur contains the 7/16 error value to be propagated - * to the next pixel on the current line, and all the errors for the - * next line have been shifted over. We are therefore ready to move on. - */ - input_ptr += dirnc; /* advance input ptr to next column */ - output_ptr += dir; /* advance output ptr to next column */ - errorptr += dir; /* advance errorptr to current column */ - } - /* Post-loop cleanup: we must unload the final error value into the - * final fserrors[] entry. Note we need not unload belowerr because - * it is for the dummy column before or after the actual array. - */ - errorptr[0] = (FSERROR) bpreverr; /* unload prev err into array */ - } - cquantize->on_odd_row = (cquantize->on_odd_row ? FALSE : TRUE); - } -} - - -/* - * Allocate workspace for Floyd-Steinberg errors. - */ - -LOCAL(void) -alloc_fs_workspace (j_decompress_ptr cinfo) -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - size_t arraysize; - int i; - - arraysize = ((size_t) cinfo->output_width + (size_t) 2) * SIZEOF(FSERROR); - for (i = 0; i < cinfo->out_color_components; i++) { - cquantize->fserrors[i] = (FSERRPTR) (*cinfo->mem->alloc_large) - ((j_common_ptr) cinfo, JPOOL_IMAGE, arraysize); - } -} - - -/* - * Initialize for one-pass color quantization. - */ - -METHODDEF(void) -start_pass_1_quant (j_decompress_ptr cinfo, boolean is_pre_scan) -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - size_t arraysize; - int i; - - /* Install my colormap. */ - cinfo->colormap = cquantize->sv_colormap; - cinfo->actual_number_of_colors = cquantize->sv_actual; - - /* Initialize for desired dithering mode. */ - switch (cinfo->dither_mode) { - case JDITHER_NONE: - if (cinfo->out_color_components == 3) - cquantize->pub.color_quantize = color_quantize3; - else - cquantize->pub.color_quantize = color_quantize; - break; - case JDITHER_ORDERED: - if (cinfo->out_color_components == 3) - cquantize->pub.color_quantize = quantize3_ord_dither; - else - cquantize->pub.color_quantize = quantize_ord_dither; - cquantize->row_index = 0; /* initialize state for ordered dither */ - /* If user changed to ordered dither from another mode, - * we must recreate the color index table with padding. - * This will cost extra space, but probably isn't very likely. - */ - if (! cquantize->is_padded) - create_colorindex(cinfo); - /* Create ordered-dither tables if we didn't already. */ - if (cquantize->odither[0] == NULL) - create_odither_tables(cinfo); - break; - case JDITHER_FS: - cquantize->pub.color_quantize = quantize_fs_dither; - cquantize->on_odd_row = FALSE; /* initialize state for F-S dither */ - /* Allocate Floyd-Steinberg workspace if didn't already. */ - if (cquantize->fserrors[0] == NULL) - alloc_fs_workspace(cinfo); - /* Initialize the propagated errors to zero. */ - arraysize = ((size_t) cinfo->output_width + (size_t) 2) * SIZEOF(FSERROR); - for (i = 0; i < cinfo->out_color_components; i++) - FMEMZERO((void FAR *) cquantize->fserrors[i], arraysize); - break; - default: - ERREXIT(cinfo, JERR_NOT_COMPILED); - } -} - - -/* - * Finish up at the end of the pass. - */ - -METHODDEF(void) -finish_pass_1_quant (j_decompress_ptr cinfo) -{ - /* no work in 1-pass case */ -} - - -/* - * Switch to a new external colormap between output passes. - * Shouldn't get to this module! - */ - -METHODDEF(void) -new_color_map_1_quant (j_decompress_ptr cinfo) -{ - ERREXIT(cinfo, JERR_MODE_CHANGE); -} - - -/* - * Module initialization routine for 1-pass color quantization. - */ - -GLOBAL(void) -jinit_1pass_quantizer (j_decompress_ptr cinfo) -{ - my_cquantize_ptr cquantize; - - cquantize = (my_cquantize_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_cquantizer)); - cinfo->cquantize = &cquantize->pub; - cquantize->pub.start_pass = start_pass_1_quant; - cquantize->pub.finish_pass = finish_pass_1_quant; - cquantize->pub.new_color_map = new_color_map_1_quant; - cquantize->fserrors[0] = NULL; /* Flag FS workspace not allocated */ - cquantize->odither[0] = NULL; /* Also flag odither arrays not allocated */ - - /* Make sure my internal arrays won't overflow */ - if (cinfo->out_color_components > MAX_Q_COMPS) - ERREXIT1(cinfo, JERR_QUANT_COMPONENTS, MAX_Q_COMPS); - /* Make sure colormap indexes can be represented by JSAMPLEs */ - if (cinfo->desired_number_of_colors > (MAXJSAMPLE+1)) - ERREXIT1(cinfo, JERR_QUANT_MANY_COLORS, MAXJSAMPLE+1); - - /* Create the colormap and color index table. */ - create_colormap(cinfo); - create_colorindex(cinfo); - - /* Allocate Floyd-Steinberg workspace now if requested. - * We do this now since it is FAR storage and may affect the memory - * manager's space calculations. If the user changes to FS dither - * mode in a later pass, we will allocate the space then, and will - * possibly overrun the max_memory_to_use setting. - */ - if (cinfo->dither_mode == JDITHER_FS) - alloc_fs_workspace(cinfo); -} - -#endif /* QUANT_1PASS_SUPPORTED */ diff --git a/dep/libjpeg/src/jquant2.c b/dep/libjpeg/src/jquant2.c deleted file mode 100644 index 662b9bcef..000000000 --- a/dep/libjpeg/src/jquant2.c +++ /dev/null @@ -1,1311 +0,0 @@ -/* - * jquant2.c - * - * Copyright (C) 1991-1996, Thomas G. Lane. - * Modified 2011-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains 2-pass color quantization (color mapping) routines. - * These routines provide selection of a custom color map for an image, - * followed by mapping of the image to that color map, with optional - * Floyd-Steinberg dithering. - * It is also possible to use just the second pass to map to an arbitrary - * externally-given color map. - * - * Note: ordered dithering is not supported, since there isn't any fast - * way to compute intercolor distances; it's unclear that ordered dither's - * fundamental assumptions even hold with an irregularly spaced color map. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - -#ifdef QUANT_2PASS_SUPPORTED - - -/* - * This module implements the well-known Heckbert paradigm for color - * quantization. Most of the ideas used here can be traced back to - * Heckbert's seminal paper - * Heckbert, Paul. "Color Image Quantization for Frame Buffer Display", - * Proc. SIGGRAPH '82, Computer Graphics v.16 #3 (July 1982), pp 297-304. - * - * In the first pass over the image, we accumulate a histogram showing the - * usage count of each possible color. To keep the histogram to a reasonable - * size, we reduce the precision of the input; typical practice is to retain - * 5 or 6 bits per color, so that 8 or 4 different input values are counted - * in the same histogram cell. - * - * Next, the color-selection step begins with a box representing the whole - * color space, and repeatedly splits the "largest" remaining box until we - * have as many boxes as desired colors. Then the mean color in each - * remaining box becomes one of the possible output colors. - * - * The second pass over the image maps each input pixel to the closest output - * color (optionally after applying a Floyd-Steinberg dithering correction). - * This mapping is logically trivial, but making it go fast enough requires - * considerable care. - * - * Heckbert-style quantizers vary a good deal in their policies for choosing - * the "largest" box and deciding where to cut it. The particular policies - * used here have proved out well in experimental comparisons, but better ones - * may yet be found. - * - * In earlier versions of the IJG code, this module quantized in YCbCr color - * space, processing the raw upsampled data without a color conversion step. - * This allowed the color conversion math to be done only once per colormap - * entry, not once per pixel. However, that optimization precluded other - * useful optimizations (such as merging color conversion with upsampling) - * and it also interfered with desired capabilities such as quantizing to an - * externally-supplied colormap. We have therefore abandoned that approach. - * The present code works in the post-conversion color space, typically RGB. - * - * To improve the visual quality of the results, we actually work in scaled - * RGB space, giving G distances more weight than R, and R in turn more than - * B. To do everything in integer math, we must use integer scale factors. - * The 2/3/1 scale factors used here correspond loosely to the relative - * weights of the colors in the NTSC grayscale equation. - * If you want to use this code to quantize a non-RGB color space, you'll - * probably need to change these scale factors. - */ - -#define R_SCALE 2 /* scale R distances by this much */ -#define G_SCALE 3 /* scale G distances by this much */ -#define B_SCALE 1 /* and B by this much */ - -/* Relabel R/G/B as components 0/1/2, respecting the RGB ordering defined - * in jmorecfg.h. As the code stands, it will do the right thing for R,G,B - * and B,G,R orders. If you define some other weird order in jmorecfg.h, - * you'll get compile errors until you extend this logic. In that case - * you'll probably want to tweak the histogram sizes too. - */ - -#if RGB_RED == 0 -#define C0_SCALE R_SCALE -#endif -#if RGB_BLUE == 0 -#define C0_SCALE B_SCALE -#endif -#if RGB_GREEN == 1 -#define C1_SCALE G_SCALE -#endif -#if RGB_RED == 2 -#define C2_SCALE R_SCALE -#endif -#if RGB_BLUE == 2 -#define C2_SCALE B_SCALE -#endif - - -/* - * First we have the histogram data structure and routines for creating it. - * - * The number of bits of precision can be adjusted by changing these symbols. - * We recommend keeping 6 bits for G and 5 each for R and B. - * If you have plenty of memory and cycles, 6 bits all around gives marginally - * better results; if you are short of memory, 5 bits all around will save - * some space but degrade the results. - * To maintain a fully accurate histogram, we'd need to allocate a "long" - * (preferably unsigned long) for each cell. In practice this is overkill; - * we can get by with 16 bits per cell. Few of the cell counts will overflow, - * and clamping those that do overflow to the maximum value will give close- - * enough results. This reduces the recommended histogram size from 256Kb - * to 128Kb, which is a useful savings on PC-class machines. - * (In the second pass the histogram space is re-used for pixel mapping data; - * in that capacity, each cell must be able to store zero to the number of - * desired colors. 16 bits/cell is plenty for that too.) - * Since the JPEG code is intended to run in small memory model on 80x86 - * machines, we can't just allocate the histogram in one chunk. Instead - * of a true 3-D array, we use a row of pointers to 2-D arrays. Each - * pointer corresponds to a C0 value (typically 2^5 = 32 pointers) and - * each 2-D array has 2^6*2^5 = 2048 or 2^6*2^6 = 4096 entries. Note that - * on 80x86 machines, the pointer row is in near memory but the actual - * arrays are in far memory (same arrangement as we use for image arrays). - */ - -#define MAXNUMCOLORS (MAXJSAMPLE+1) /* maximum size of colormap */ - -/* These will do the right thing for either R,G,B or B,G,R color order, - * but you may not like the results for other color orders. - */ -#define HIST_C0_BITS 5 /* bits of precision in R/B histogram */ -#define HIST_C1_BITS 6 /* bits of precision in G histogram */ -#define HIST_C2_BITS 5 /* bits of precision in B/R histogram */ - -/* Number of elements along histogram axes. */ -#define HIST_C0_ELEMS (1<cquantize; - register JSAMPROW ptr; - register histptr histp; - register hist3d histogram = cquantize->histogram; - int row; - JDIMENSION col; - JDIMENSION width = cinfo->output_width; - - for (row = 0; row < num_rows; row++) { - ptr = input_buf[row]; - for (col = width; col > 0; col--) { - /* get pixel value and index into the histogram */ - histp = & histogram[GETJSAMPLE(ptr[0]) >> C0_SHIFT] - [GETJSAMPLE(ptr[1]) >> C1_SHIFT] - [GETJSAMPLE(ptr[2]) >> C2_SHIFT]; - /* increment, check for overflow and undo increment if so. */ - if (++(*histp) <= 0) - (*histp)--; - ptr += 3; - } - } -} - - -/* - * Next we have the really interesting routines: selection of a colormap - * given the completed histogram. - * These routines work with a list of "boxes", each representing a rectangular - * subset of the input color space (to histogram precision). - */ - -typedef struct { - /* The bounds of the box (inclusive); expressed as histogram indexes */ - int c0min, c0max; - int c1min, c1max; - int c2min, c2max; - /* The volume (actually 2-norm) of the box */ - INT32 volume; - /* The number of nonzero histogram cells within this box */ - long colorcount; -} box; - -typedef box * boxptr; - - -LOCAL(boxptr) -find_biggest_color_pop (boxptr boxlist, int numboxes) -/* Find the splittable box with the largest color population */ -/* Returns NULL if no splittable boxes remain */ -{ - register boxptr boxp; - register int i; - register long maxc = 0; - boxptr which = NULL; - - for (i = 0, boxp = boxlist; i < numboxes; i++, boxp++) { - if (boxp->colorcount > maxc && boxp->volume > 0) { - which = boxp; - maxc = boxp->colorcount; - } - } - return which; -} - - -LOCAL(boxptr) -find_biggest_volume (boxptr boxlist, int numboxes) -/* Find the splittable box with the largest (scaled) volume */ -/* Returns NULL if no splittable boxes remain */ -{ - register boxptr boxp; - register int i; - register INT32 maxv = 0; - boxptr which = NULL; - - for (i = 0, boxp = boxlist; i < numboxes; i++, boxp++) { - if (boxp->volume > maxv) { - which = boxp; - maxv = boxp->volume; - } - } - return which; -} - - -LOCAL(void) -update_box (j_decompress_ptr cinfo, boxptr boxp) -/* Shrink the min/max bounds of a box to enclose only nonzero elements, */ -/* and recompute its volume and population */ -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - hist3d histogram = cquantize->histogram; - histptr histp; - int c0,c1,c2; - int c0min,c0max,c1min,c1max,c2min,c2max; - INT32 dist0,dist1,dist2; - long ccount; - - c0min = boxp->c0min; c0max = boxp->c0max; - c1min = boxp->c1min; c1max = boxp->c1max; - c2min = boxp->c2min; c2max = boxp->c2max; - - if (c0max > c0min) - for (c0 = c0min; c0 <= c0max; c0++) - for (c1 = c1min; c1 <= c1max; c1++) { - histp = & histogram[c0][c1][c2min]; - for (c2 = c2min; c2 <= c2max; c2++) - if (*histp++ != 0) { - boxp->c0min = c0min = c0; - goto have_c0min; - } - } - have_c0min: - if (c0max > c0min) - for (c0 = c0max; c0 >= c0min; c0--) - for (c1 = c1min; c1 <= c1max; c1++) { - histp = & histogram[c0][c1][c2min]; - for (c2 = c2min; c2 <= c2max; c2++) - if (*histp++ != 0) { - boxp->c0max = c0max = c0; - goto have_c0max; - } - } - have_c0max: - if (c1max > c1min) - for (c1 = c1min; c1 <= c1max; c1++) - for (c0 = c0min; c0 <= c0max; c0++) { - histp = & histogram[c0][c1][c2min]; - for (c2 = c2min; c2 <= c2max; c2++) - if (*histp++ != 0) { - boxp->c1min = c1min = c1; - goto have_c1min; - } - } - have_c1min: - if (c1max > c1min) - for (c1 = c1max; c1 >= c1min; c1--) - for (c0 = c0min; c0 <= c0max; c0++) { - histp = & histogram[c0][c1][c2min]; - for (c2 = c2min; c2 <= c2max; c2++) - if (*histp++ != 0) { - boxp->c1max = c1max = c1; - goto have_c1max; - } - } - have_c1max: - if (c2max > c2min) - for (c2 = c2min; c2 <= c2max; c2++) - for (c0 = c0min; c0 <= c0max; c0++) { - histp = & histogram[c0][c1min][c2]; - for (c1 = c1min; c1 <= c1max; c1++, histp += HIST_C2_ELEMS) - if (*histp != 0) { - boxp->c2min = c2min = c2; - goto have_c2min; - } - } - have_c2min: - if (c2max > c2min) - for (c2 = c2max; c2 >= c2min; c2--) - for (c0 = c0min; c0 <= c0max; c0++) { - histp = & histogram[c0][c1min][c2]; - for (c1 = c1min; c1 <= c1max; c1++, histp += HIST_C2_ELEMS) - if (*histp != 0) { - boxp->c2max = c2max = c2; - goto have_c2max; - } - } - have_c2max: - - /* Update box volume. - * We use 2-norm rather than real volume here; this biases the method - * against making long narrow boxes, and it has the side benefit that - * a box is splittable iff norm > 0. - * Since the differences are expressed in histogram-cell units, - * we have to shift back to JSAMPLE units to get consistent distances; - * after which, we scale according to the selected distance scale factors. - */ - dist0 = ((c0max - c0min) << C0_SHIFT) * C0_SCALE; - dist1 = ((c1max - c1min) << C1_SHIFT) * C1_SCALE; - dist2 = ((c2max - c2min) << C2_SHIFT) * C2_SCALE; - boxp->volume = dist0*dist0 + dist1*dist1 + dist2*dist2; - - /* Now scan remaining volume of box and compute population */ - ccount = 0; - for (c0 = c0min; c0 <= c0max; c0++) - for (c1 = c1min; c1 <= c1max; c1++) { - histp = & histogram[c0][c1][c2min]; - for (c2 = c2min; c2 <= c2max; c2++, histp++) - if (*histp != 0) { - ccount++; - } - } - boxp->colorcount = ccount; -} - - -LOCAL(int) -median_cut (j_decompress_ptr cinfo, boxptr boxlist, int numboxes, - int desired_colors) -/* Repeatedly select and split the largest box until we have enough boxes */ -{ - int n,lb; - int c0,c1,c2,cmax; - register boxptr b1,b2; - - while (numboxes < desired_colors) { - /* Select box to split. - * Current algorithm: by population for first half, then by volume. - */ - if (numboxes*2 <= desired_colors) { - b1 = find_biggest_color_pop(boxlist, numboxes); - } else { - b1 = find_biggest_volume(boxlist, numboxes); - } - if (b1 == NULL) /* no splittable boxes left! */ - break; - b2 = &boxlist[numboxes]; /* where new box will go */ - /* Copy the color bounds to the new box. */ - b2->c0max = b1->c0max; b2->c1max = b1->c1max; b2->c2max = b1->c2max; - b2->c0min = b1->c0min; b2->c1min = b1->c1min; b2->c2min = b1->c2min; - /* Choose which axis to split the box on. - * Current algorithm: longest scaled axis. - * See notes in update_box about scaling distances. - */ - c0 = ((b1->c0max - b1->c0min) << C0_SHIFT) * C0_SCALE; - c1 = ((b1->c1max - b1->c1min) << C1_SHIFT) * C1_SCALE; - c2 = ((b1->c2max - b1->c2min) << C2_SHIFT) * C2_SCALE; - /* We want to break any ties in favor of green, then red, blue last. - * This code does the right thing for R,G,B or B,G,R color orders only. - */ -#if RGB_RED == 0 - cmax = c1; n = 1; - if (c0 > cmax) { cmax = c0; n = 0; } - if (c2 > cmax) { n = 2; } -#else - cmax = c1; n = 1; - if (c2 > cmax) { cmax = c2; n = 2; } - if (c0 > cmax) { n = 0; } -#endif - /* Choose split point along selected axis, and update box bounds. - * Current algorithm: split at halfway point. - * (Since the box has been shrunk to minimum volume, - * any split will produce two nonempty subboxes.) - * Note that lb value is max for lower box, so must be < old max. - */ - switch (n) { - case 0: - lb = (b1->c0max + b1->c0min) / 2; - b1->c0max = lb; - b2->c0min = lb+1; - break; - case 1: - lb = (b1->c1max + b1->c1min) / 2; - b1->c1max = lb; - b2->c1min = lb+1; - break; - case 2: - lb = (b1->c2max + b1->c2min) / 2; - b1->c2max = lb; - b2->c2min = lb+1; - break; - } - /* Update stats for boxes */ - update_box(cinfo, b1); - update_box(cinfo, b2); - numboxes++; - } - return numboxes; -} - - -LOCAL(void) -compute_color (j_decompress_ptr cinfo, boxptr boxp, int icolor) -/* Compute representative color for a box, put it in colormap[icolor] */ -{ - /* Current algorithm: mean weighted by pixels (not colors) */ - /* Note it is important to get the rounding correct! */ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - hist3d histogram = cquantize->histogram; - histptr histp; - int c0,c1,c2; - int c0min,c0max,c1min,c1max,c2min,c2max; - long count; - long total = 0; - long c0total = 0; - long c1total = 0; - long c2total = 0; - - c0min = boxp->c0min; c0max = boxp->c0max; - c1min = boxp->c1min; c1max = boxp->c1max; - c2min = boxp->c2min; c2max = boxp->c2max; - - for (c0 = c0min; c0 <= c0max; c0++) - for (c1 = c1min; c1 <= c1max; c1++) { - histp = & histogram[c0][c1][c2min]; - for (c2 = c2min; c2 <= c2max; c2++) { - if ((count = *histp++) != 0) { - total += count; - c0total += ((c0 << C0_SHIFT) + ((1<>1)) * count; - c1total += ((c1 << C1_SHIFT) + ((1<>1)) * count; - c2total += ((c2 << C2_SHIFT) + ((1<>1)) * count; - } - } - } - - cinfo->colormap[0][icolor] = (JSAMPLE) ((c0total + (total>>1)) / total); - cinfo->colormap[1][icolor] = (JSAMPLE) ((c1total + (total>>1)) / total); - cinfo->colormap[2][icolor] = (JSAMPLE) ((c2total + (total>>1)) / total); -} - - -LOCAL(void) -select_colors (j_decompress_ptr cinfo, int desired_colors) -/* Master routine for color selection */ -{ - boxptr boxlist; - int numboxes; - int i; - - /* Allocate workspace for box list */ - boxlist = (boxptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, desired_colors * SIZEOF(box)); - /* Initialize one box containing whole space */ - numboxes = 1; - boxlist[0].c0min = 0; - boxlist[0].c0max = MAXJSAMPLE >> C0_SHIFT; - boxlist[0].c1min = 0; - boxlist[0].c1max = MAXJSAMPLE >> C1_SHIFT; - boxlist[0].c2min = 0; - boxlist[0].c2max = MAXJSAMPLE >> C2_SHIFT; - /* Shrink it to actually-used volume and set its statistics */ - update_box(cinfo, & boxlist[0]); - /* Perform median-cut to produce final box list */ - numboxes = median_cut(cinfo, boxlist, numboxes, desired_colors); - /* Compute the representative color for each box, fill colormap */ - for (i = 0; i < numboxes; i++) - compute_color(cinfo, & boxlist[i], i); - cinfo->actual_number_of_colors = numboxes; - TRACEMS1(cinfo, 1, JTRC_QUANT_SELECTED, numboxes); -} - - -/* - * These routines are concerned with the time-critical task of mapping input - * colors to the nearest color in the selected colormap. - * - * We re-use the histogram space as an "inverse color map", essentially a - * cache for the results of nearest-color searches. All colors within a - * histogram cell will be mapped to the same colormap entry, namely the one - * closest to the cell's center. This may not be quite the closest entry to - * the actual input color, but it's almost as good. A zero in the cache - * indicates we haven't found the nearest color for that cell yet; the array - * is cleared to zeroes before starting the mapping pass. When we find the - * nearest color for a cell, its colormap index plus one is recorded in the - * cache for future use. The pass2 scanning routines call fill_inverse_cmap - * when they need to use an unfilled entry in the cache. - * - * Our method of efficiently finding nearest colors is based on the "locally - * sorted search" idea described by Heckbert and on the incremental distance - * calculation described by Spencer W. Thomas in chapter III.1 of Graphics - * Gems II (James Arvo, ed. Academic Press, 1991). Thomas points out that - * the distances from a given colormap entry to each cell of the histogram can - * be computed quickly using an incremental method: the differences between - * distances to adjacent cells themselves differ by a constant. This allows a - * fairly fast implementation of the "brute force" approach of computing the - * distance from every colormap entry to every histogram cell. Unfortunately, - * it needs a work array to hold the best-distance-so-far for each histogram - * cell (because the inner loop has to be over cells, not colormap entries). - * The work array elements have to be INT32s, so the work array would need - * 256Kb at our recommended precision. This is not feasible in DOS machines. - * - * To get around these problems, we apply Thomas' method to compute the - * nearest colors for only the cells within a small subbox of the histogram. - * The work array need be only as big as the subbox, so the memory usage - * problem is solved. Furthermore, we need not fill subboxes that are never - * referenced in pass2; many images use only part of the color gamut, so a - * fair amount of work is saved. An additional advantage of this - * approach is that we can apply Heckbert's locality criterion to quickly - * eliminate colormap entries that are far away from the subbox; typically - * three-fourths of the colormap entries are rejected by Heckbert's criterion, - * and we need not compute their distances to individual cells in the subbox. - * The speed of this approach is heavily influenced by the subbox size: too - * small means too much overhead, too big loses because Heckbert's criterion - * can't eliminate as many colormap entries. Empirically the best subbox - * size seems to be about 1/512th of the histogram (1/8th in each direction). - * - * Thomas' article also describes a refined method which is asymptotically - * faster than the brute-force method, but it is also far more complex and - * cannot efficiently be applied to small subboxes. It is therefore not - * useful for programs intended to be portable to DOS machines. On machines - * with plenty of memory, filling the whole histogram in one shot with Thomas' - * refined method might be faster than the present code --- but then again, - * it might not be any faster, and it's certainly more complicated. - */ - - -/* log2(histogram cells in update box) for each axis; this can be adjusted */ -#define BOX_C0_LOG (HIST_C0_BITS-3) -#define BOX_C1_LOG (HIST_C1_BITS-3) -#define BOX_C2_LOG (HIST_C2_BITS-3) - -#define BOX_C0_ELEMS (1<actual_number_of_colors; - int maxc0, maxc1, maxc2; - int centerc0, centerc1, centerc2; - int i, x, ncolors; - INT32 minmaxdist, min_dist, max_dist, tdist; - INT32 mindist[MAXNUMCOLORS]; /* min distance to colormap entry i */ - - /* Compute true coordinates of update box's upper corner and center. - * Actually we compute the coordinates of the center of the upper-corner - * histogram cell, which are the upper bounds of the volume we care about. - * Note that since ">>" rounds down, the "center" values may be closer to - * min than to max; hence comparisons to them must be "<=", not "<". - */ - maxc0 = minc0 + ((1 << BOX_C0_SHIFT) - (1 << C0_SHIFT)); - centerc0 = (minc0 + maxc0) >> 1; - maxc1 = minc1 + ((1 << BOX_C1_SHIFT) - (1 << C1_SHIFT)); - centerc1 = (minc1 + maxc1) >> 1; - maxc2 = minc2 + ((1 << BOX_C2_SHIFT) - (1 << C2_SHIFT)); - centerc2 = (minc2 + maxc2) >> 1; - - /* For each color in colormap, find: - * 1. its minimum squared-distance to any point in the update box - * (zero if color is within update box); - * 2. its maximum squared-distance to any point in the update box. - * Both of these can be found by considering only the corners of the box. - * We save the minimum distance for each color in mindist[]; - * only the smallest maximum distance is of interest. - */ - minmaxdist = 0x7FFFFFFFL; - - for (i = 0; i < numcolors; i++) { - /* We compute the squared-c0-distance term, then add in the other two. */ - x = GETJSAMPLE(cinfo->colormap[0][i]); - if (x < minc0) { - tdist = (x - minc0) * C0_SCALE; - min_dist = tdist*tdist; - tdist = (x - maxc0) * C0_SCALE; - max_dist = tdist*tdist; - } else if (x > maxc0) { - tdist = (x - maxc0) * C0_SCALE; - min_dist = tdist*tdist; - tdist = (x - minc0) * C0_SCALE; - max_dist = tdist*tdist; - } else { - /* within cell range so no contribution to min_dist */ - min_dist = 0; - if (x <= centerc0) { - tdist = (x - maxc0) * C0_SCALE; - max_dist = tdist*tdist; - } else { - tdist = (x - minc0) * C0_SCALE; - max_dist = tdist*tdist; - } - } - - x = GETJSAMPLE(cinfo->colormap[1][i]); - if (x < minc1) { - tdist = (x - minc1) * C1_SCALE; - min_dist += tdist*tdist; - tdist = (x - maxc1) * C1_SCALE; - max_dist += tdist*tdist; - } else if (x > maxc1) { - tdist = (x - maxc1) * C1_SCALE; - min_dist += tdist*tdist; - tdist = (x - minc1) * C1_SCALE; - max_dist += tdist*tdist; - } else { - /* within cell range so no contribution to min_dist */ - if (x <= centerc1) { - tdist = (x - maxc1) * C1_SCALE; - max_dist += tdist*tdist; - } else { - tdist = (x - minc1) * C1_SCALE; - max_dist += tdist*tdist; - } - } - - x = GETJSAMPLE(cinfo->colormap[2][i]); - if (x < minc2) { - tdist = (x - minc2) * C2_SCALE; - min_dist += tdist*tdist; - tdist = (x - maxc2) * C2_SCALE; - max_dist += tdist*tdist; - } else if (x > maxc2) { - tdist = (x - maxc2) * C2_SCALE; - min_dist += tdist*tdist; - tdist = (x - minc2) * C2_SCALE; - max_dist += tdist*tdist; - } else { - /* within cell range so no contribution to min_dist */ - if (x <= centerc2) { - tdist = (x - maxc2) * C2_SCALE; - max_dist += tdist*tdist; - } else { - tdist = (x - minc2) * C2_SCALE; - max_dist += tdist*tdist; - } - } - - mindist[i] = min_dist; /* save away the results */ - if (max_dist < minmaxdist) - minmaxdist = max_dist; - } - - /* Now we know that no cell in the update box is more than minmaxdist - * away from some colormap entry. Therefore, only colors that are - * within minmaxdist of some part of the box need be considered. - */ - ncolors = 0; - for (i = 0; i < numcolors; i++) { - if (mindist[i] <= minmaxdist) - colorlist[ncolors++] = (JSAMPLE) i; - } - return ncolors; -} - - -LOCAL(void) -find_best_colors (j_decompress_ptr cinfo, int minc0, int minc1, int minc2, - int numcolors, JSAMPLE colorlist[], JSAMPLE bestcolor[]) -/* Find the closest colormap entry for each cell in the update box, - * given the list of candidate colors prepared by find_nearby_colors. - * Return the indexes of the closest entries in the bestcolor[] array. - * This routine uses Thomas' incremental distance calculation method to - * find the distance from a colormap entry to successive cells in the box. - */ -{ - int ic0, ic1, ic2; - int i, icolor; - register INT32 * bptr; /* pointer into bestdist[] array */ - JSAMPLE * cptr; /* pointer into bestcolor[] array */ - INT32 dist0, dist1; /* initial distance values */ - register INT32 dist2; /* current distance in inner loop */ - INT32 xx0, xx1; /* distance increments */ - register INT32 xx2; - INT32 inc0, inc1, inc2; /* initial values for increments */ - /* This array holds the distance to the nearest-so-far color for each cell */ - INT32 bestdist[BOX_C0_ELEMS * BOX_C1_ELEMS * BOX_C2_ELEMS]; - - /* Initialize best-distance for each cell of the update box */ - bptr = bestdist; - for (i = BOX_C0_ELEMS*BOX_C1_ELEMS*BOX_C2_ELEMS-1; i >= 0; i--) - *bptr++ = 0x7FFFFFFFL; - - /* For each color selected by find_nearby_colors, - * compute its distance to the center of each cell in the box. - * If that's less than best-so-far, update best distance and color number. - */ - - /* Nominal steps between cell centers ("x" in Thomas article) */ -#define STEP_C0 ((1 << C0_SHIFT) * C0_SCALE) -#define STEP_C1 ((1 << C1_SHIFT) * C1_SCALE) -#define STEP_C2 ((1 << C2_SHIFT) * C2_SCALE) - - for (i = 0; i < numcolors; i++) { - icolor = GETJSAMPLE(colorlist[i]); - /* Compute (square of) distance from minc0/c1/c2 to this color */ - inc0 = (minc0 - GETJSAMPLE(cinfo->colormap[0][icolor])) * C0_SCALE; - dist0 = inc0*inc0; - inc1 = (minc1 - GETJSAMPLE(cinfo->colormap[1][icolor])) * C1_SCALE; - dist0 += inc1*inc1; - inc2 = (minc2 - GETJSAMPLE(cinfo->colormap[2][icolor])) * C2_SCALE; - dist0 += inc2*inc2; - /* Form the initial difference increments */ - inc0 = inc0 * (2 * STEP_C0) + STEP_C0 * STEP_C0; - inc1 = inc1 * (2 * STEP_C1) + STEP_C1 * STEP_C1; - inc2 = inc2 * (2 * STEP_C2) + STEP_C2 * STEP_C2; - /* Now loop over all cells in box, updating distance per Thomas method */ - bptr = bestdist; - cptr = bestcolor; - xx0 = inc0; - for (ic0 = BOX_C0_ELEMS-1; ic0 >= 0; ic0--) { - dist1 = dist0; - xx1 = inc1; - for (ic1 = BOX_C1_ELEMS-1; ic1 >= 0; ic1--) { - dist2 = dist1; - xx2 = inc2; - for (ic2 = BOX_C2_ELEMS-1; ic2 >= 0; ic2--) { - if (dist2 < *bptr) { - *bptr = dist2; - *cptr = (JSAMPLE) icolor; - } - dist2 += xx2; - xx2 += 2 * STEP_C2 * STEP_C2; - bptr++; - cptr++; - } - dist1 += xx1; - xx1 += 2 * STEP_C1 * STEP_C1; - } - dist0 += xx0; - xx0 += 2 * STEP_C0 * STEP_C0; - } - } -} - - -LOCAL(void) -fill_inverse_cmap (j_decompress_ptr cinfo, int c0, int c1, int c2) -/* Fill the inverse-colormap entries in the update box that contains */ -/* histogram cell c0/c1/c2. (Only that one cell MUST be filled, but */ -/* we can fill as many others as we wish.) */ -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - hist3d histogram = cquantize->histogram; - int minc0, minc1, minc2; /* lower left corner of update box */ - int ic0, ic1, ic2; - register JSAMPLE * cptr; /* pointer into bestcolor[] array */ - register histptr cachep; /* pointer into main cache array */ - /* This array lists the candidate colormap indexes. */ - JSAMPLE colorlist[MAXNUMCOLORS]; - int numcolors; /* number of candidate colors */ - /* This array holds the actually closest colormap index for each cell. */ - JSAMPLE bestcolor[BOX_C0_ELEMS * BOX_C1_ELEMS * BOX_C2_ELEMS]; - - /* Convert cell coordinates to update box ID */ - c0 >>= BOX_C0_LOG; - c1 >>= BOX_C1_LOG; - c2 >>= BOX_C2_LOG; - - /* Compute true coordinates of update box's origin corner. - * Actually we compute the coordinates of the center of the corner - * histogram cell, which are the lower bounds of the volume we care about. - */ - minc0 = (c0 << BOX_C0_SHIFT) + ((1 << C0_SHIFT) >> 1); - minc1 = (c1 << BOX_C1_SHIFT) + ((1 << C1_SHIFT) >> 1); - minc2 = (c2 << BOX_C2_SHIFT) + ((1 << C2_SHIFT) >> 1); - - /* Determine which colormap entries are close enough to be candidates - * for the nearest entry to some cell in the update box. - */ - numcolors = find_nearby_colors(cinfo, minc0, minc1, minc2, colorlist); - - /* Determine the actually nearest colors. */ - find_best_colors(cinfo, minc0, minc1, minc2, numcolors, colorlist, - bestcolor); - - /* Save the best color numbers (plus 1) in the main cache array */ - c0 <<= BOX_C0_LOG; /* convert ID back to base cell indexes */ - c1 <<= BOX_C1_LOG; - c2 <<= BOX_C2_LOG; - cptr = bestcolor; - for (ic0 = 0; ic0 < BOX_C0_ELEMS; ic0++) { - for (ic1 = 0; ic1 < BOX_C1_ELEMS; ic1++) { - cachep = & histogram[c0+ic0][c1+ic1][c2]; - for (ic2 = 0; ic2 < BOX_C2_ELEMS; ic2++) { - *cachep++ = (histcell) (GETJSAMPLE(*cptr++) + 1); - } - } - } -} - - -/* - * Map some rows of pixels to the output colormapped representation. - */ - -METHODDEF(void) -pass2_no_dither (j_decompress_ptr cinfo, - JSAMPARRAY input_buf, JSAMPARRAY output_buf, int num_rows) -/* This version performs no dithering */ -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - hist3d histogram = cquantize->histogram; - register JSAMPROW inptr, outptr; - register histptr cachep; - register int c0, c1, c2; - int row; - JDIMENSION col; - JDIMENSION width = cinfo->output_width; - - for (row = 0; row < num_rows; row++) { - inptr = input_buf[row]; - outptr = output_buf[row]; - for (col = width; col > 0; col--) { - /* get pixel value and index into the cache */ - c0 = GETJSAMPLE(*inptr++) >> C0_SHIFT; - c1 = GETJSAMPLE(*inptr++) >> C1_SHIFT; - c2 = GETJSAMPLE(*inptr++) >> C2_SHIFT; - cachep = & histogram[c0][c1][c2]; - /* If we have not seen this color before, find nearest colormap entry */ - /* and update the cache */ - if (*cachep == 0) - fill_inverse_cmap(cinfo, c0,c1,c2); - /* Now emit the colormap index for this cell */ - *outptr++ = (JSAMPLE) (*cachep - 1); - } - } -} - - -METHODDEF(void) -pass2_fs_dither (j_decompress_ptr cinfo, - JSAMPARRAY input_buf, JSAMPARRAY output_buf, int num_rows) -/* This version performs Floyd-Steinberg dithering */ -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - hist3d histogram = cquantize->histogram; - register LOCFSERROR cur0, cur1, cur2; /* current error or pixel value */ - LOCFSERROR belowerr0, belowerr1, belowerr2; /* error for pixel below cur */ - LOCFSERROR bpreverr0, bpreverr1, bpreverr2; /* error for below/prev col */ - register FSERRPTR errorptr; /* => fserrors[] at column before current */ - JSAMPROW inptr; /* => current input pixel */ - JSAMPROW outptr; /* => current output pixel */ - histptr cachep; - int dir; /* +1 or -1 depending on direction */ - int dir3; /* 3*dir, for advancing inptr & errorptr */ - int row; - JDIMENSION col; - JDIMENSION width = cinfo->output_width; - JSAMPLE *range_limit = cinfo->sample_range_limit; - int *error_limit = cquantize->error_limiter; - JSAMPROW colormap0 = cinfo->colormap[0]; - JSAMPROW colormap1 = cinfo->colormap[1]; - JSAMPROW colormap2 = cinfo->colormap[2]; - SHIFT_TEMPS - - for (row = 0; row < num_rows; row++) { - inptr = input_buf[row]; - outptr = output_buf[row]; - if (cquantize->on_odd_row) { - /* work right to left in this row */ - inptr += (width-1) * 3; /* so point to rightmost pixel */ - outptr += width-1; - dir = -1; - dir3 = -3; - errorptr = cquantize->fserrors + (width+1)*3; /* => entry after last column */ - cquantize->on_odd_row = FALSE; /* flip for next time */ - } else { - /* work left to right in this row */ - dir = 1; - dir3 = 3; - errorptr = cquantize->fserrors; /* => entry before first real column */ - cquantize->on_odd_row = TRUE; /* flip for next time */ - } - /* Preset error values: no error propagated to first pixel from left */ - cur0 = cur1 = cur2 = 0; - /* and no error propagated to row below yet */ - belowerr0 = belowerr1 = belowerr2 = 0; - bpreverr0 = bpreverr1 = bpreverr2 = 0; - - for (col = width; col > 0; col--) { - /* curN holds the error propagated from the previous pixel on the - * current line. Add the error propagated from the previous line - * to form the complete error correction term for this pixel, and - * round the error term (which is expressed * 16) to an integer. - * RIGHT_SHIFT rounds towards minus infinity, so adding 8 is correct - * for either sign of the error value. - * Note: errorptr points to *previous* column's array entry. - */ - cur0 = RIGHT_SHIFT(cur0 + errorptr[dir3+0] + 8, 4); - cur1 = RIGHT_SHIFT(cur1 + errorptr[dir3+1] + 8, 4); - cur2 = RIGHT_SHIFT(cur2 + errorptr[dir3+2] + 8, 4); - /* Limit the error using transfer function set by init_error_limit. - * See comments with init_error_limit for rationale. - */ - cur0 = error_limit[cur0]; - cur1 = error_limit[cur1]; - cur2 = error_limit[cur2]; - /* Form pixel value + error, and range-limit to 0..MAXJSAMPLE. - * The maximum error is +- MAXJSAMPLE (or less with error limiting); - * this sets the required size of the range_limit array. - */ - cur0 += GETJSAMPLE(inptr[0]); - cur1 += GETJSAMPLE(inptr[1]); - cur2 += GETJSAMPLE(inptr[2]); - cur0 = GETJSAMPLE(range_limit[cur0]); - cur1 = GETJSAMPLE(range_limit[cur1]); - cur2 = GETJSAMPLE(range_limit[cur2]); - /* Index into the cache with adjusted pixel value */ - cachep = & histogram[cur0>>C0_SHIFT][cur1>>C1_SHIFT][cur2>>C2_SHIFT]; - /* If we have not seen this color before, find nearest colormap */ - /* entry and update the cache */ - if (*cachep == 0) - fill_inverse_cmap(cinfo, cur0>>C0_SHIFT,cur1>>C1_SHIFT,cur2>>C2_SHIFT); - /* Now emit the colormap index for this cell */ - { register int pixcode = *cachep - 1; - *outptr = (JSAMPLE) pixcode; - /* Compute representation error for this pixel */ - cur0 -= GETJSAMPLE(colormap0[pixcode]); - cur1 -= GETJSAMPLE(colormap1[pixcode]); - cur2 -= GETJSAMPLE(colormap2[pixcode]); - } - /* Compute error fractions to be propagated to adjacent pixels. - * Add these into the running sums, and simultaneously shift the - * next-line error sums left by 1 column. - */ - { register LOCFSERROR bnexterr, delta; - - bnexterr = cur0; /* Process component 0 */ - delta = cur0 * 2; - cur0 += delta; /* form error * 3 */ - errorptr[0] = (FSERROR) (bpreverr0 + cur0); - cur0 += delta; /* form error * 5 */ - bpreverr0 = belowerr0 + cur0; - belowerr0 = bnexterr; - cur0 += delta; /* form error * 7 */ - bnexterr = cur1; /* Process component 1 */ - delta = cur1 * 2; - cur1 += delta; /* form error * 3 */ - errorptr[1] = (FSERROR) (bpreverr1 + cur1); - cur1 += delta; /* form error * 5 */ - bpreverr1 = belowerr1 + cur1; - belowerr1 = bnexterr; - cur1 += delta; /* form error * 7 */ - bnexterr = cur2; /* Process component 2 */ - delta = cur2 * 2; - cur2 += delta; /* form error * 3 */ - errorptr[2] = (FSERROR) (bpreverr2 + cur2); - cur2 += delta; /* form error * 5 */ - bpreverr2 = belowerr2 + cur2; - belowerr2 = bnexterr; - cur2 += delta; /* form error * 7 */ - } - /* At this point curN contains the 7/16 error value to be propagated - * to the next pixel on the current line, and all the errors for the - * next line have been shifted over. We are therefore ready to move on. - */ - inptr += dir3; /* Advance pixel pointers to next column */ - outptr += dir; - errorptr += dir3; /* advance errorptr to current column */ - } - /* Post-loop cleanup: we must unload the final error values into the - * final fserrors[] entry. Note we need not unload belowerrN because - * it is for the dummy column before or after the actual array. - */ - errorptr[0] = (FSERROR) bpreverr0; /* unload prev errs into array */ - errorptr[1] = (FSERROR) bpreverr1; - errorptr[2] = (FSERROR) bpreverr2; - } -} - - -/* - * Initialize the error-limiting transfer function (lookup table). - * The raw F-S error computation can potentially compute error values of up to - * +- MAXJSAMPLE. But we want the maximum correction applied to a pixel to be - * much less, otherwise obviously wrong pixels will be created. (Typical - * effects include weird fringes at color-area boundaries, isolated bright - * pixels in a dark area, etc.) The standard advice for avoiding this problem - * is to ensure that the "corners" of the color cube are allocated as output - * colors; then repeated errors in the same direction cannot cause cascading - * error buildup. However, that only prevents the error from getting - * completely out of hand; Aaron Giles reports that error limiting improves - * the results even with corner colors allocated. - * A simple clamping of the error values to about +- MAXJSAMPLE/8 works pretty - * well, but the smoother transfer function used below is even better. Thanks - * to Aaron Giles for this idea. - */ - -LOCAL(void) -init_error_limit (j_decompress_ptr cinfo) -/* Allocate and fill in the error_limiter table */ -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - int * table; - int in, out; - - table = (int *) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, (MAXJSAMPLE*2+1) * SIZEOF(int)); - table += MAXJSAMPLE; /* so can index -MAXJSAMPLE .. +MAXJSAMPLE */ - cquantize->error_limiter = table; - -#define STEPSIZE ((MAXJSAMPLE+1)/16) - /* Map errors 1:1 up to +- MAXJSAMPLE/16 */ - out = 0; - for (in = 0; in < STEPSIZE; in++, out++) { - table[in] = out; table[-in] = -out; - } - /* Map errors 1:2 up to +- 3*MAXJSAMPLE/16 */ - for (; in < STEPSIZE*3; in++, out += (in&1) ? 0 : 1) { - table[in] = out; table[-in] = -out; - } - /* Clamp the rest to final out value (which is (MAXJSAMPLE+1)/8) */ - for (; in <= MAXJSAMPLE; in++) { - table[in] = out; table[-in] = -out; - } -#undef STEPSIZE -} - - -/* - * Finish up at the end of each pass. - */ - -METHODDEF(void) -finish_pass1 (j_decompress_ptr cinfo) -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - - /* Select the representative colors and fill in cinfo->colormap */ - cinfo->colormap = cquantize->sv_colormap; - select_colors(cinfo, cquantize->desired); - /* Force next pass to zero the color index table */ - cquantize->needs_zeroed = TRUE; -} - - -METHODDEF(void) -finish_pass2 (j_decompress_ptr cinfo) -{ - /* no work */ -} - - -/* - * Initialize for each processing pass. - */ - -METHODDEF(void) -start_pass_2_quant (j_decompress_ptr cinfo, boolean is_pre_scan) -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - hist3d histogram = cquantize->histogram; - int i; - - /* Only F-S dithering or no dithering is supported. */ - /* If user asks for ordered dither, give him F-S. */ - if (cinfo->dither_mode != JDITHER_NONE) - cinfo->dither_mode = JDITHER_FS; - - if (is_pre_scan) { - /* Set up method pointers */ - cquantize->pub.color_quantize = prescan_quantize; - cquantize->pub.finish_pass = finish_pass1; - cquantize->needs_zeroed = TRUE; /* Always zero histogram */ - } else { - /* Set up method pointers */ - if (cinfo->dither_mode == JDITHER_FS) - cquantize->pub.color_quantize = pass2_fs_dither; - else - cquantize->pub.color_quantize = pass2_no_dither; - cquantize->pub.finish_pass = finish_pass2; - - /* Make sure color count is acceptable */ - i = cinfo->actual_number_of_colors; - if (i < 1) - ERREXIT1(cinfo, JERR_QUANT_FEW_COLORS, 1); - if (i > MAXNUMCOLORS) - ERREXIT1(cinfo, JERR_QUANT_MANY_COLORS, MAXNUMCOLORS); - - if (cinfo->dither_mode == JDITHER_FS) { - size_t arraysize = ((size_t) cinfo->output_width + (size_t) 2) - * (3 * SIZEOF(FSERROR)); - /* Allocate Floyd-Steinberg workspace if we didn't already. */ - if (cquantize->fserrors == NULL) - cquantize->fserrors = (FSERRPTR) (*cinfo->mem->alloc_large) - ((j_common_ptr) cinfo, JPOOL_IMAGE, arraysize); - /* Initialize the propagated errors to zero. */ - FMEMZERO((void FAR *) cquantize->fserrors, arraysize); - /* Make the error-limit table if we didn't already. */ - if (cquantize->error_limiter == NULL) - init_error_limit(cinfo); - cquantize->on_odd_row = FALSE; - } - - } - /* Zero the histogram or inverse color map, if necessary */ - if (cquantize->needs_zeroed) { - for (i = 0; i < HIST_C0_ELEMS; i++) { - FMEMZERO((void FAR *) histogram[i], - HIST_C1_ELEMS*HIST_C2_ELEMS * SIZEOF(histcell)); - } - cquantize->needs_zeroed = FALSE; - } -} - - -/* - * Switch to a new external colormap between output passes. - */ - -METHODDEF(void) -new_color_map_2_quant (j_decompress_ptr cinfo) -{ - my_cquantize_ptr cquantize = (my_cquantize_ptr) cinfo->cquantize; - - /* Reset the inverse color map */ - cquantize->needs_zeroed = TRUE; -} - - -/* - * Module initialization routine for 2-pass color quantization. - */ - -GLOBAL(void) -jinit_2pass_quantizer (j_decompress_ptr cinfo) -{ - my_cquantize_ptr cquantize; - int i; - - cquantize = (my_cquantize_ptr) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, SIZEOF(my_cquantizer)); - cinfo->cquantize = &cquantize->pub; - cquantize->pub.start_pass = start_pass_2_quant; - cquantize->pub.new_color_map = new_color_map_2_quant; - cquantize->fserrors = NULL; /* flag optional arrays not allocated */ - cquantize->error_limiter = NULL; - - /* Make sure jdmaster didn't give me a case I can't handle */ - if (cinfo->out_color_components != 3) - ERREXIT(cinfo, JERR_NOTIMPL); - - /* Allocate the histogram/inverse colormap storage */ - cquantize->histogram = (hist3d) (*cinfo->mem->alloc_small) - ((j_common_ptr) cinfo, JPOOL_IMAGE, HIST_C0_ELEMS * SIZEOF(hist2d)); - for (i = 0; i < HIST_C0_ELEMS; i++) { - cquantize->histogram[i] = (hist2d) (*cinfo->mem->alloc_large) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - HIST_C1_ELEMS*HIST_C2_ELEMS * SIZEOF(histcell)); - } - cquantize->needs_zeroed = TRUE; /* histogram is garbage now */ - - /* Allocate storage for the completed colormap, if required. - * We do this now since it is FAR storage and may affect - * the memory manager's space calculations. - */ - if (cinfo->enable_2pass_quant) { - /* Make sure color count is acceptable */ - int desired = cinfo->desired_number_of_colors; - /* Lower bound on # of colors ... somewhat arbitrary as long as > 0 */ - if (desired < 8) - ERREXIT1(cinfo, JERR_QUANT_FEW_COLORS, 8); - /* Make sure colormap indexes can be represented by JSAMPLEs */ - if (desired > MAXNUMCOLORS) - ERREXIT1(cinfo, JERR_QUANT_MANY_COLORS, MAXNUMCOLORS); - cquantize->sv_colormap = (*cinfo->mem->alloc_sarray) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - (JDIMENSION) desired, (JDIMENSION) 3); - cquantize->desired = desired; - } else - cquantize->sv_colormap = NULL; - - /* Only F-S dithering or no dithering is supported. */ - /* If user asks for ordered dither, give him F-S. */ - if (cinfo->dither_mode != JDITHER_NONE) - cinfo->dither_mode = JDITHER_FS; - - /* Allocate Floyd-Steinberg workspace if necessary. - * This isn't really needed until pass 2, but again it is FAR storage. - * Although we will cope with a later change in dither_mode, - * we do not promise to honor max_memory_to_use if dither_mode changes. - */ - if (cinfo->dither_mode == JDITHER_FS) { - cquantize->fserrors = (FSERRPTR) (*cinfo->mem->alloc_large) - ((j_common_ptr) cinfo, JPOOL_IMAGE, - ((size_t) cinfo->output_width + (size_t) 2) * (3 * SIZEOF(FSERROR))); - /* Might as well create the error-limiting table too. */ - init_error_limit(cinfo); - } -} - -#endif /* QUANT_2PASS_SUPPORTED */ diff --git a/dep/libjpeg/src/jutils.c b/dep/libjpeg/src/jutils.c deleted file mode 100644 index 31e16dfb5..000000000 --- a/dep/libjpeg/src/jutils.c +++ /dev/null @@ -1,224 +0,0 @@ -/* - * jutils.c - * - * Copyright (C) 1991-1996, Thomas G. Lane. - * Modified 2009-2020 by Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains tables and miscellaneous utility routines needed - * for both compression and decompression. - * Note we prefix all global names with "j" to minimize conflicts with - * a surrounding application. - */ - -#define JPEG_INTERNALS -#include "jinclude.h" -#include "jpeglib.h" - - -/* - * jpeg_zigzag_order[i] is the zigzag-order position of the i'th element - * of a DCT block read in natural order (left to right, top to bottom). - */ - -#if 0 /* This table is not actually needed in v6a */ - -const int jpeg_zigzag_order[DCTSIZE2] = { - 0, 1, 5, 6, 14, 15, 27, 28, - 2, 4, 7, 13, 16, 26, 29, 42, - 3, 8, 12, 17, 25, 30, 41, 43, - 9, 11, 18, 24, 31, 40, 44, 53, - 10, 19, 23, 32, 39, 45, 52, 54, - 20, 22, 33, 38, 46, 51, 55, 60, - 21, 34, 37, 47, 50, 56, 59, 61, - 35, 36, 48, 49, 57, 58, 62, 63 -}; - -#endif - -/* - * jpeg_natural_order[i] is the natural-order position of the i'th element - * of zigzag order. - * - * When reading corrupted data, the Huffman decoders could attempt - * to reference an entry beyond the end of this array (if the decoded - * zero run length reaches past the end of the block). To prevent - * wild stores without adding an inner-loop test, we put some extra - * "63"s after the real entries. This will cause the extra coefficient - * to be stored in location 63 of the block, not somewhere random. - * The worst case would be a run-length of 15, which means we need 16 - * fake entries. - */ - -const int jpeg_natural_order[DCTSIZE2+16] = { - 0, 1, 8, 16, 9, 2, 3, 10, - 17, 24, 32, 25, 18, 11, 4, 5, - 12, 19, 26, 33, 40, 48, 41, 34, - 27, 20, 13, 6, 7, 14, 21, 28, - 35, 42, 49, 56, 57, 50, 43, 36, - 29, 22, 15, 23, 30, 37, 44, 51, - 58, 59, 52, 45, 38, 31, 39, 46, - 53, 60, 61, 54, 47, 55, 62, 63, - 63, 63, 63, 63, 63, 63, 63, 63, /* extra entries for safety in decoder */ - 63, 63, 63, 63, 63, 63, 63, 63 -}; - -const int jpeg_natural_order7[7*7+16] = { - 0, 1, 8, 16, 9, 2, 3, 10, - 17, 24, 32, 25, 18, 11, 4, 5, - 12, 19, 26, 33, 40, 48, 41, 34, - 27, 20, 13, 6, 14, 21, 28, 35, - 42, 49, 50, 43, 36, 29, 22, 30, - 37, 44, 51, 52, 45, 38, 46, 53, - 54, - 63, 63, 63, 63, 63, 63, 63, 63, /* extra entries for safety in decoder */ - 63, 63, 63, 63, 63, 63, 63, 63 -}; - -const int jpeg_natural_order6[6*6+16] = { - 0, 1, 8, 16, 9, 2, 3, 10, - 17, 24, 32, 25, 18, 11, 4, 5, - 12, 19, 26, 33, 40, 41, 34, 27, - 20, 13, 21, 28, 35, 42, 43, 36, - 29, 37, 44, 45, - 63, 63, 63, 63, 63, 63, 63, 63, /* extra entries for safety in decoder */ - 63, 63, 63, 63, 63, 63, 63, 63 -}; - -const int jpeg_natural_order5[5*5+16] = { - 0, 1, 8, 16, 9, 2, 3, 10, - 17, 24, 32, 25, 18, 11, 4, 12, - 19, 26, 33, 34, 27, 20, 28, 35, - 36, - 63, 63, 63, 63, 63, 63, 63, 63, /* extra entries for safety in decoder */ - 63, 63, 63, 63, 63, 63, 63, 63 -}; - -const int jpeg_natural_order4[4*4+16] = { - 0, 1, 8, 16, 9, 2, 3, 10, - 17, 24, 25, 18, 11, 19, 26, 27, - 63, 63, 63, 63, 63, 63, 63, 63, /* extra entries for safety in decoder */ - 63, 63, 63, 63, 63, 63, 63, 63 -}; - -const int jpeg_natural_order3[3*3+16] = { - 0, 1, 8, 16, 9, 2, 10, 17, - 18, - 63, 63, 63, 63, 63, 63, 63, 63, /* extra entries for safety in decoder */ - 63, 63, 63, 63, 63, 63, 63, 63 -}; - -const int jpeg_natural_order2[2*2+16] = { - 0, 1, 8, 9, - 63, 63, 63, 63, 63, 63, 63, 63, /* extra entries for safety in decoder */ - 63, 63, 63, 63, 63, 63, 63, 63 -}; - - -/* - * Arithmetic utilities - */ - -GLOBAL(long) -jdiv_round_up (long a, long b) -/* Compute a/b rounded up to next integer, ie, ceil(a/b) */ -/* Assumes a >= 0, b > 0 */ -{ - return (a + b - 1L) / b; -} - - -GLOBAL(long) -jround_up (long a, long b) -/* Compute a rounded up to next multiple of b, ie, ceil(a/b)*b */ -/* Assumes a >= 0, b > 0 */ -{ - a += b - 1L; - return a - (a % b); -} - - -/* On normal machines we can apply MEMCOPY() and MEMZERO() to sample arrays - * and coefficient-block arrays. This won't work on 80x86 because the arrays - * are FAR and we're assuming a small-pointer memory model. However, some - * DOS compilers provide far-pointer versions of memcpy() and memset() even - * in the small-model libraries. These will be used if USE_FMEM is defined. - * Otherwise, the routines below do it the hard way. (The performance cost - * is not all that great, because these routines aren't very heavily used.) - */ - -#ifndef NEED_FAR_POINTERS /* normal case, same as regular macro */ -#define FMEMCOPY(dest,src,size) MEMCOPY(dest,src,size) -#else /* 80x86 case, define if we can */ -#ifdef USE_FMEM -#define FMEMCOPY(dest,src,size) _fmemcpy((void FAR *)(dest), (const void FAR *)(src), (size_t)(size)) -#else -/* This function is for use by the FMEMZERO macro defined in jpegint.h. - * Do not call this function directly, use the FMEMZERO macro instead. - */ -GLOBAL(void) -jzero_far (void FAR * target, size_t bytestozero) -/* Zero out a chunk of FAR memory. */ -/* This might be sample-array data, block-array data, or alloc_large data. */ -{ - register char FAR * ptr = (char FAR *) target; - register size_t count; - - for (count = bytestozero; count > 0; count--) { - *ptr++ = 0; - } -} -#endif -#endif - - -GLOBAL(void) -jcopy_sample_rows (JSAMPARRAY input_array, - JSAMPARRAY output_array, - int num_rows, JDIMENSION num_cols) -/* Copy some rows of samples from one place to another. - * num_rows rows are copied from *input_array++ to *output_array++; - * these areas may overlap for duplication. - * The source and destination arrays must be at least as wide as num_cols. - */ -{ - register JSAMPROW inptr, outptr; -#ifdef FMEMCOPY - register size_t count = (size_t) num_cols * SIZEOF(JSAMPLE); -#else - register JDIMENSION count; -#endif - register int row; - - for (row = num_rows; row > 0; row--) { - inptr = *input_array++; - outptr = *output_array++; -#ifdef FMEMCOPY - FMEMCOPY(outptr, inptr, count); -#else - for (count = num_cols; count > 0; count--) - *outptr++ = *inptr++; /* needn't bother with GETJSAMPLE() here */ -#endif - } -} - - -GLOBAL(void) -jcopy_block_row (JBLOCKROW input_row, JBLOCKROW output_row, - JDIMENSION num_blocks) -/* Copy a row of coefficient blocks from one place to another. */ -{ -#ifdef FMEMCOPY - FMEMCOPY(output_row, input_row, (size_t) num_blocks * (DCTSIZE2 * SIZEOF(JCOEF))); -#else - register JCOEFPTR inptr, outptr; - register long count; - - inptr = (JCOEFPTR) input_row; - outptr = (JCOEFPTR) output_row; - for (count = (long) num_blocks * DCTSIZE2; count > 0; count--) { - *outptr++ = *inptr++; - } -#endif -} diff --git a/dep/libjpeg/src/jversion.h b/dep/libjpeg/src/jversion.h deleted file mode 100644 index df53ef5e5..000000000 --- a/dep/libjpeg/src/jversion.h +++ /dev/null @@ -1,14 +0,0 @@ -/* - * jversion.h - * - * Copyright (C) 1991-2024, Thomas G. Lane, Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains software version identification. - */ - - -#define JVERSION "9f 14-Jan-2024" - -#define JCOPYRIGHT "Copyright (C) 2024, Thomas G. Lane, Guido Vollbeding" diff --git a/dep/libjpeg/src/transupp.c b/dep/libjpeg/src/transupp.c deleted file mode 100644 index 6518936a0..000000000 --- a/dep/libjpeg/src/transupp.c +++ /dev/null @@ -1,2433 +0,0 @@ -/* - * transupp.c - * - * Copyright (C) 1997-2023, Thomas G. Lane, Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains image transformation routines and other utility code - * used by the jpegtran sample application. These are NOT part of the core - * JPEG library. But we keep these routines separate from jpegtran.c to - * ease the task of maintaining jpegtran-like programs that have other user - * interfaces. - */ - -/* Although this file really shouldn't have access to the library internals, - * it's helpful to let it call jround_up() and jcopy_block_row(). - * Also, the (switchable) virtual memory adaptation code for - * the drop feature has dependencies on library internals. - */ -#define JPEG_INTERNALS - -#include "jinclude.h" -#include "jpeglib.h" -#include "transupp.h" /* My own external interface */ -#include /* to declare isdigit() */ - - -#if TRANSFORMS_SUPPORTED - -/* - * Lossless image transformation routines. These routines work on DCT - * coefficient arrays and thus do not require any lossy decompression - * or recompression of the image. - * Thanks to Guido Vollbeding for the initial design and code of this feature, - * and to Ben Jackson for introducing the cropping feature. - * - * Horizontal flipping is done in-place, using a single top-to-bottom - * pass through the virtual source array. It will thus be much the - * fastest option for images larger than main memory. - * - * The other routines require a set of destination virtual arrays, so they - * need twice as much memory as jpegtran normally does. The destination - * arrays are always written in normal scan order (top to bottom) because - * the virtual array manager expects this. The source arrays will be scanned - * in the corresponding order, which means multiple passes through the source - * arrays for most of the transforms. That could result in much thrashing - * if the image is larger than main memory. - * - * If cropping or trimming is involved, the destination arrays may be smaller - * than the source arrays. Note it is not possible to do horizontal flip - * in-place when a nonzero Y crop offset is specified, since we'd have to move - * data from one block row to another but the virtual array manager doesn't - * guarantee we can touch more than one row at a time. So in that case, - * we have to use a separate destination array. - * - * Some notes about the operating environment of the individual transform - * routines: - * 1. Both the source and destination virtual arrays are allocated from the - * source JPEG object, and therefore should be manipulated by calling the - * source's memory manager. - * 2. The destination's component count should be used. It may be smaller - * than the source's when forcing to grayscale. - * 3. Likewise the destination's sampling factors should be used. When - * forcing to grayscale the destination's sampling factors will be all 1, - * and we may as well take that as the effective iMCU size. - * 4. When "trim" is in effect, the destination's dimensions will be the - * trimmed values but the source's will be untrimmed. - * 5. When "crop" is in effect, the destination's dimensions will be the - * cropped values but the source's will be uncropped. Each transform - * routine is responsible for picking up source data starting at the - * correct X and Y offset for the crop region. (The X and Y offsets - * passed to the transform routines are measured in iMCU blocks of the - * destination.) - * 6. All the routines assume that the source and destination buffers are - * padded out to a full iMCU boundary. This is true, although for the - * source buffer it is an undocumented property of jdcoefct.c. - */ - - -/* Drop code may be used with or without virtual memory adaptation code. - * This code has some dependencies on internal library behavior, so you - * may choose to disable it. For example, it doesn't make a difference - * if you only use jmemnobs anyway. - */ -#ifndef DROP_REQUEST_FROM_SRC -#define DROP_REQUEST_FROM_SRC 1 /* 0 disables adaptation */ -#endif - - -#if DROP_REQUEST_FROM_SRC -/* Force jpeg_read_coefficients to request - * the virtual coefficient arrays from - * the source decompression object. - */ -METHODDEF(jvirt_barray_ptr) -drop_request_virt_barray (j_common_ptr cinfo, int pool_id, boolean pre_zero, - JDIMENSION blocksperrow, JDIMENSION numrows, - JDIMENSION maxaccess) -{ - j_common_ptr srcinfo = (j_common_ptr) cinfo->client_data; - - return (*srcinfo->mem->request_virt_barray) - (srcinfo, pool_id, pre_zero, - blocksperrow, numrows, maxaccess); -} - - -/* Force jpeg_read_coefficients to return - * after requesting and before accessing - * the virtual coefficient arrays. - */ -METHODDEF(int) -drop_consume_input (j_decompress_ptr cinfo) -{ - return JPEG_SUSPENDED; -} - - -METHODDEF(void) -drop_start_input_pass (j_decompress_ptr cinfo) -{ - cinfo->inputctl->consume_input = drop_consume_input; -} - - -LOCAL(void) -drop_request_from_src (j_decompress_ptr dropinfo, j_decompress_ptr srcinfo) -{ - void *save_client_data; - JMETHOD(jvirt_barray_ptr, save_request_virt_barray, - (j_common_ptr cinfo, int pool_id, boolean pre_zero, - JDIMENSION blocksperrow, JDIMENSION numrows, JDIMENSION maxaccess)); - JMETHOD(void, save_start_input_pass, (j_decompress_ptr cinfo)); - - /* Set custom method pointers, save original pointers */ - save_client_data = dropinfo->client_data; - dropinfo->client_data = (void *) srcinfo; - save_request_virt_barray = dropinfo->mem->request_virt_barray; - dropinfo->mem->request_virt_barray = drop_request_virt_barray; - save_start_input_pass = dropinfo->inputctl->start_input_pass; - dropinfo->inputctl->start_input_pass = drop_start_input_pass; - - /* Execute only initialization part. - * Requested coefficient arrays will be realized later by the srcinfo object. - * Next call to the same function will then do the actual data reading. - * NB: since we request the coefficient arrays from another object, - * the inherent realization call is effectively a no-op. - */ - (void) jpeg_read_coefficients(dropinfo); - - /* Reset method pointers */ - dropinfo->client_data = save_client_data; - dropinfo->mem->request_virt_barray = save_request_virt_barray; - dropinfo->inputctl->start_input_pass = save_start_input_pass; - /* Do input initialization for first scan now, - * which also resets the consume_input method. - */ - (*save_start_input_pass)(dropinfo); -} -#endif /* DROP_REQUEST_FROM_SRC */ - - -LOCAL(void) -dequant_comp (j_decompress_ptr cinfo, jpeg_component_info *compptr, - jvirt_barray_ptr coef_array, JQUANT_TBL *qtblptr1) -{ - JDIMENSION blk_x, blk_y; - int offset_y, k; - JQUANT_TBL *qtblptr; - JBLOCKARRAY buffer; - JBLOCKROW block; - JCOEFPTR ptr; - - qtblptr = compptr->quant_table; - for (blk_y = 0; blk_y < compptr->height_in_blocks; - blk_y += compptr->v_samp_factor) { - buffer = (*cinfo->mem->access_virt_barray) - ((j_common_ptr) cinfo, coef_array, blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - block = buffer[offset_y]; - for (blk_x = 0; blk_x < compptr->width_in_blocks; blk_x++) { - ptr = block[blk_x]; - for (k = 0; k < DCTSIZE2; k++) - if (qtblptr->quantval[k] != qtblptr1->quantval[k]) - ptr[k] *= qtblptr->quantval[k] / qtblptr1->quantval[k]; - } - } - } -} - - -LOCAL(void) -requant_comp (j_decompress_ptr cinfo, jpeg_component_info *compptr, - jvirt_barray_ptr coef_array, JQUANT_TBL *qtblptr1) -{ - JDIMENSION blk_x, blk_y; - int offset_y, k, temp, qval; - JQUANT_TBL *qtblptr; - JBLOCKARRAY buffer; - JBLOCKROW block; - JCOEFPTR ptr; - - qtblptr = compptr->quant_table; - for (blk_y = 0; blk_y < compptr->height_in_blocks; - blk_y += compptr->v_samp_factor) { - buffer = (*cinfo->mem->access_virt_barray) - ((j_common_ptr) cinfo, coef_array, blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - block = buffer[offset_y]; - for (blk_x = 0; blk_x < compptr->width_in_blocks; blk_x++) { - ptr = block[blk_x]; - for (k = 0; k < DCTSIZE2; k++) { - qval = qtblptr1->quantval[k]; - if (qval == 0) continue; - temp = qtblptr->quantval[k]; - if (temp == qval) continue; - temp *= ptr[k]; - /* The following quantization code is a copy from jcdctmgr.c */ -#ifdef FAST_DIVIDE -#define DIVIDE_BY(a,b) a /= b -#else -#define DIVIDE_BY(a,b) if (a >= b) a /= b; else a = 0 -#endif - if (temp < 0) { - temp = -temp; - temp += qval>>1; /* for rounding */ - DIVIDE_BY(temp, qval); - temp = -temp; - } else { - temp += qval>>1; /* for rounding */ - DIVIDE_BY(temp, qval); - } - ptr[k] = (JCOEF) temp; - } - } - } - } -} - - -/* Calculate largest common denominator with Euclid's algorithm. - */ -LOCAL(JCOEF) -largest_common_denominator(JCOEF a, JCOEF b) -{ - JCOEF c; - - while (b) { - c = a % b; - a = b; - b = c; - } - - return a; -} - - -LOCAL(void) -adjust_quant(j_decompress_ptr srcinfo, jvirt_barray_ptr *src_coef_arrays, - j_decompress_ptr dropinfo, jvirt_barray_ptr *drop_coef_arrays, - boolean trim, j_compress_ptr dstinfo) -{ - jpeg_component_info *compptr1, *compptr2; - JQUANT_TBL *qtblptr1, *qtblptr2, *qtblptr3; - int ci, k; - - for (ci = 0; ci < dstinfo->num_components && - ci < dropinfo->num_components; ci++) { - compptr1 = srcinfo->comp_info + ci; - compptr2 = dropinfo->comp_info + ci; - qtblptr1 = compptr1->quant_table; - qtblptr2 = compptr2->quant_table; - for (k = 0; k < DCTSIZE2; k++) { - if (qtblptr1->quantval[k] != qtblptr2->quantval[k]) { - if (trim) - requant_comp(dropinfo, compptr2, drop_coef_arrays[ci], qtblptr1); - else { - qtblptr3 = dstinfo->quant_tbl_ptrs[compptr1->quant_tbl_no]; - for (k = 0; k < DCTSIZE2; k++) - if (qtblptr1->quantval[k] != qtblptr2->quantval[k]) - qtblptr3->quantval[k] = largest_common_denominator - (qtblptr1->quantval[k], qtblptr2->quantval[k]); - dequant_comp(srcinfo, compptr1, src_coef_arrays[ci], qtblptr3); - dequant_comp(dropinfo, compptr2, drop_coef_arrays[ci], qtblptr3); - } - break; - } - } - } -} - - -LOCAL(void) -do_drop (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - j_decompress_ptr dropinfo, jvirt_barray_ptr *drop_coef_arrays, - JDIMENSION drop_width, JDIMENSION drop_height) -/* Drop. If the dropinfo component number is smaller than the destination's, - * we fill in the remaining components with zero. This provides the feature - * of dropping grayscale into (arbitrarily sampled) color images. - */ -{ - JDIMENSION comp_width, comp_height; - JDIMENSION blk_y, x_drop_blocks, y_drop_blocks; - int ci, offset_y; - JBLOCKARRAY src_buffer, dst_buffer; - jpeg_component_info *compptr; - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - comp_width = drop_width * compptr->h_samp_factor; - comp_height = drop_height * compptr->v_samp_factor; - x_drop_blocks = x_crop_offset * compptr->h_samp_factor; - y_drop_blocks = y_crop_offset * compptr->v_samp_factor; - for (blk_y = 0; blk_y < comp_height; blk_y += compptr->v_samp_factor) { - dst_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], blk_y + y_drop_blocks, - (JDIMENSION) compptr->v_samp_factor, TRUE); - if (ci < dropinfo->num_components) { -#if DROP_REQUEST_FROM_SRC - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, drop_coef_arrays[ci], blk_y, -#else - src_buffer = (*dropinfo->mem->access_virt_barray) - ((j_common_ptr) dropinfo, drop_coef_arrays[ci], blk_y, -#endif - (JDIMENSION) compptr->v_samp_factor, FALSE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - jcopy_block_row(src_buffer[offset_y], - dst_buffer[offset_y] + x_drop_blocks, - comp_width); - } - } else { - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - FMEMZERO(dst_buffer[offset_y] + x_drop_blocks, - comp_width * SIZEOF(JBLOCK)); - } - } - } - } -} - - -LOCAL(void) -do_crop (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - jvirt_barray_ptr *dst_coef_arrays) -/* Crop. This is only used when no rotate/flip is requested with the crop. */ -{ - JDIMENSION dst_blk_y, x_crop_blocks, y_crop_blocks; - int ci, offset_y; - JBLOCKARRAY src_buffer, dst_buffer; - jpeg_component_info *compptr; - - /* We simply have to copy the right amount of data (the destination's - * image size) starting at the given X and Y offsets in the source. - */ - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - x_crop_blocks = x_crop_offset * compptr->h_samp_factor; - y_crop_blocks = y_crop_offset * compptr->v_samp_factor; - for (dst_blk_y = 0; dst_blk_y < compptr->height_in_blocks; - dst_blk_y += compptr->v_samp_factor) { - dst_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, dst_coef_arrays[ci], dst_blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_y + y_crop_blocks, - (JDIMENSION) compptr->v_samp_factor, FALSE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - jcopy_block_row(src_buffer[offset_y] + x_crop_blocks, - dst_buffer[offset_y], - compptr->width_in_blocks); - } - } - } -} - - -LOCAL(void) -do_crop_ext_zero (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - jvirt_barray_ptr *dst_coef_arrays) -/* Crop. This is only used when no rotate/flip is requested with the crop. - * Extension: If the destination size is larger than the source, we fill in - * the extra area with zero (neutral gray). Note we also have to zero partial - * iMCUs at the right and bottom edge of the source image area in this case. - */ -{ - JDIMENSION MCU_cols, MCU_rows, comp_width, comp_height; - JDIMENSION dst_blk_y, x_crop_blocks, y_crop_blocks; - int ci, offset_y; - JBLOCKARRAY src_buffer, dst_buffer; - jpeg_component_info *compptr; - - MCU_cols = srcinfo->output_width / - (dstinfo->max_h_samp_factor * dstinfo->min_DCT_h_scaled_size); - MCU_rows = srcinfo->output_height / - (dstinfo->max_v_samp_factor * dstinfo->min_DCT_v_scaled_size); - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - comp_width = MCU_cols * compptr->h_samp_factor; - comp_height = MCU_rows * compptr->v_samp_factor; - x_crop_blocks = x_crop_offset * compptr->h_samp_factor; - y_crop_blocks = y_crop_offset * compptr->v_samp_factor; - for (dst_blk_y = 0; dst_blk_y < compptr->height_in_blocks; - dst_blk_y += compptr->v_samp_factor) { - dst_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, dst_coef_arrays[ci], dst_blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - if (dstinfo->jpeg_height > srcinfo->output_height) { - if (dst_blk_y < y_crop_blocks || - dst_blk_y >= y_crop_blocks + comp_height) { - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - FMEMZERO(dst_buffer[offset_y], - compptr->width_in_blocks * SIZEOF(JBLOCK)); - } - continue; - } - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_y - y_crop_blocks, - (JDIMENSION) compptr->v_samp_factor, FALSE); - } else { - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_y + y_crop_blocks, - (JDIMENSION) compptr->v_samp_factor, FALSE); - } - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - if (dstinfo->jpeg_width > srcinfo->output_width) { - if (x_crop_blocks > 0) { - FMEMZERO(dst_buffer[offset_y], - x_crop_blocks * SIZEOF(JBLOCK)); - } - jcopy_block_row(src_buffer[offset_y], - dst_buffer[offset_y] + x_crop_blocks, - comp_width); - if (compptr->width_in_blocks > x_crop_blocks + comp_width) { - FMEMZERO(dst_buffer[offset_y] + - x_crop_blocks + comp_width, - (compptr->width_in_blocks - - x_crop_blocks - comp_width) * SIZEOF(JBLOCK)); - } - } else { - jcopy_block_row(src_buffer[offset_y] + x_crop_blocks, - dst_buffer[offset_y], - compptr->width_in_blocks); - } - } - } - } -} - - -LOCAL(void) -do_crop_ext_flat (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - jvirt_barray_ptr *dst_coef_arrays) -/* Crop. This is only used when no rotate/flip is requested with the crop. - * Extension: The destination width is larger than the source and we fill in - * the extra area with the DC of the adjacent block. Note we also have to - * fill partial iMCUs at the right and bottom edge of the source image area - * in this case. - */ -{ - JDIMENSION MCU_cols, MCU_rows, comp_width, comp_height; - JDIMENSION dst_blk_x, dst_blk_y, x_crop_blocks, y_crop_blocks; - int ci, offset_y; - JCOEF dc; - JBLOCKARRAY src_buffer, dst_buffer; - jpeg_component_info *compptr; - - MCU_cols = srcinfo->output_width / - (dstinfo->max_h_samp_factor * dstinfo->min_DCT_h_scaled_size); - MCU_rows = srcinfo->output_height / - (dstinfo->max_v_samp_factor * dstinfo->min_DCT_v_scaled_size); - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - comp_width = MCU_cols * compptr->h_samp_factor; - comp_height = MCU_rows * compptr->v_samp_factor; - x_crop_blocks = x_crop_offset * compptr->h_samp_factor; - y_crop_blocks = y_crop_offset * compptr->v_samp_factor; - for (dst_blk_y = 0; dst_blk_y < compptr->height_in_blocks; - dst_blk_y += compptr->v_samp_factor) { - dst_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, dst_coef_arrays[ci], dst_blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - if (dstinfo->jpeg_height > srcinfo->output_height) { - if (dst_blk_y < y_crop_blocks || - dst_blk_y >= y_crop_blocks + comp_height) { - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - FMEMZERO(dst_buffer[offset_y], - compptr->width_in_blocks * SIZEOF(JBLOCK)); - } - continue; - } - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_y - y_crop_blocks, - (JDIMENSION) compptr->v_samp_factor, FALSE); - } else { - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_y + y_crop_blocks, - (JDIMENSION) compptr->v_samp_factor, FALSE); - } - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - if (x_crop_blocks > 0) { - FMEMZERO(dst_buffer[offset_y], - x_crop_blocks * SIZEOF(JBLOCK)); - dc = src_buffer[offset_y][0][0]; - for (dst_blk_x = 0; dst_blk_x < x_crop_blocks; dst_blk_x++) { - dst_buffer[offset_y][dst_blk_x][0] = dc; - } - } - jcopy_block_row(src_buffer[offset_y], - dst_buffer[offset_y] + x_crop_blocks, - comp_width); - if (compptr->width_in_blocks > x_crop_blocks + comp_width) { - FMEMZERO(dst_buffer[offset_y] + - x_crop_blocks + comp_width, - (compptr->width_in_blocks - - x_crop_blocks - comp_width) * SIZEOF(JBLOCK)); - dc = src_buffer[offset_y][comp_width - 1][0]; - for (dst_blk_x = x_crop_blocks + comp_width; - dst_blk_x < compptr->width_in_blocks; dst_blk_x++) { - dst_buffer[offset_y][dst_blk_x][0] = dc; - } - } - } - } - } -} - - -LOCAL(void) -do_crop_ext_reflect (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - jvirt_barray_ptr *dst_coef_arrays) -/* Crop. This is only used when no rotate/flip is requested with the crop. - * Extension: The destination width is larger than the source and we fill in - * the extra area with repeated reflections of the source region. Note we - * also have to fill partial iMCUs at the right and bottom edge of the source - * image area in this case. - */ -{ - JDIMENSION MCU_cols, MCU_rows, comp_width, comp_height, src_blk_x; - JDIMENSION dst_blk_x, dst_blk_y, x_crop_blocks, y_crop_blocks; - int ci, k, offset_y; - JBLOCKARRAY src_buffer, dst_buffer; - JBLOCKROW src_row_ptr, dst_row_ptr; - JCOEFPTR src_ptr, dst_ptr; - jpeg_component_info *compptr; - - MCU_cols = srcinfo->output_width / - (dstinfo->max_h_samp_factor * dstinfo->min_DCT_h_scaled_size); - MCU_rows = srcinfo->output_height / - (dstinfo->max_v_samp_factor * dstinfo->min_DCT_v_scaled_size); - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - comp_width = MCU_cols * compptr->h_samp_factor; - comp_height = MCU_rows * compptr->v_samp_factor; - x_crop_blocks = x_crop_offset * compptr->h_samp_factor; - y_crop_blocks = y_crop_offset * compptr->v_samp_factor; - for (dst_blk_y = 0; dst_blk_y < compptr->height_in_blocks; - dst_blk_y += compptr->v_samp_factor) { - dst_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, dst_coef_arrays[ci], dst_blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - if (dstinfo->jpeg_height > srcinfo->output_height) { - if (dst_blk_y < y_crop_blocks || - dst_blk_y >= y_crop_blocks + comp_height) { - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - FMEMZERO(dst_buffer[offset_y], - compptr->width_in_blocks * SIZEOF(JBLOCK)); - } - continue; - } - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_y - y_crop_blocks, - (JDIMENSION) compptr->v_samp_factor, FALSE); - } else { - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_y + y_crop_blocks, - (JDIMENSION) compptr->v_samp_factor, FALSE); - } - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - /* Copy source region */ - jcopy_block_row(src_buffer[offset_y], - dst_buffer[offset_y] + x_crop_blocks, - comp_width); - if (x_crop_blocks > 0) { - /* Reflect to left */ - dst_row_ptr = dst_buffer[offset_y] + x_crop_blocks; - for (dst_blk_x = x_crop_blocks; dst_blk_x > 0;) { - src_row_ptr = dst_row_ptr; /* (re)set axis of reflection */ - for (src_blk_x = comp_width; src_blk_x > 0 && dst_blk_x > 0; - src_blk_x--, dst_blk_x--) { - dst_ptr = *--dst_row_ptr; /* destination goes left */ - src_ptr = *src_row_ptr++; /* source goes right */ - /* this unrolled loop doesn't need to know which row it's on... */ - for (k = 0; k < DCTSIZE2; k += 2) { - *dst_ptr++ = *src_ptr++; /* copy even column */ - *dst_ptr++ = - *src_ptr++; /* copy odd column with sign change */ - } - } - } - } - if (compptr->width_in_blocks > x_crop_blocks + comp_width) { - /* Reflect to right */ - dst_row_ptr = dst_buffer[offset_y] + x_crop_blocks + comp_width; - for (dst_blk_x = compptr->width_in_blocks - x_crop_blocks - comp_width; - dst_blk_x > 0;) { - src_row_ptr = dst_row_ptr; /* (re)set axis of reflection */ - for (src_blk_x = comp_width; src_blk_x > 0 && dst_blk_x > 0; - src_blk_x--, dst_blk_x--) { - dst_ptr = *dst_row_ptr++; /* destination goes right */ - src_ptr = *--src_row_ptr; /* source goes left */ - /* this unrolled loop doesn't need to know which row it's on... */ - for (k = 0; k < DCTSIZE2; k += 2) { - *dst_ptr++ = *src_ptr++; /* copy even column */ - *dst_ptr++ = - *src_ptr++; /* copy odd column with sign change */ - } - } - } - } - } - } - } -} - - -LOCAL(void) -do_wipe (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - JDIMENSION drop_width, JDIMENSION drop_height) -/* Wipe - drop content of specified area, fill with zero (neutral gray) */ -{ - JDIMENSION x_wipe_blocks, wipe_width; - JDIMENSION y_wipe_blocks, wipe_bottom; - int ci, offset_y; - JBLOCKARRAY buffer; - jpeg_component_info *compptr; - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - x_wipe_blocks = x_crop_offset * compptr->h_samp_factor; - wipe_width = drop_width * compptr->h_samp_factor; - y_wipe_blocks = y_crop_offset * compptr->v_samp_factor; - wipe_bottom = drop_height * compptr->v_samp_factor + y_wipe_blocks; - for (; y_wipe_blocks < wipe_bottom; - y_wipe_blocks += compptr->v_samp_factor) { - buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], y_wipe_blocks, - (JDIMENSION) compptr->v_samp_factor, TRUE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - FMEMZERO(buffer[offset_y] + x_wipe_blocks, - wipe_width * SIZEOF(JBLOCK)); - } - } - } -} - - -LOCAL(void) -do_flatten (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - JDIMENSION drop_width, JDIMENSION drop_height) -/* Flatten - drop content of specified area, similar to wipe, - * but fill with average of adjacent blocks, instead of zero. - */ -{ - JDIMENSION x_wipe_blocks, wipe_width, wipe_right; - JDIMENSION y_wipe_blocks, wipe_bottom, blk_x; - int ci, offset_y, dc_left_value, dc_right_value, average; - JBLOCKARRAY buffer; - jpeg_component_info *compptr; - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - x_wipe_blocks = x_crop_offset * compptr->h_samp_factor; - wipe_width = drop_width * compptr->h_samp_factor; - wipe_right = wipe_width + x_wipe_blocks; - y_wipe_blocks = y_crop_offset * compptr->v_samp_factor; - wipe_bottom = drop_height * compptr->v_samp_factor + y_wipe_blocks; - for (; y_wipe_blocks < wipe_bottom; - y_wipe_blocks += compptr->v_samp_factor) { - buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], y_wipe_blocks, - (JDIMENSION) compptr->v_samp_factor, TRUE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - FMEMZERO(buffer[offset_y] + x_wipe_blocks, - wipe_width * SIZEOF(JBLOCK)); - if (x_wipe_blocks > 0) { - dc_left_value = buffer[offset_y][x_wipe_blocks - 1][0]; - if (wipe_right < compptr->width_in_blocks) { - dc_right_value = buffer[offset_y][wipe_right][0]; - average = (dc_left_value + dc_right_value) >> 1; - } else { - average = dc_left_value; - } - } else if (wipe_right < compptr->width_in_blocks) { - average = buffer[offset_y][wipe_right][0]; - } else continue; - for (blk_x = x_wipe_blocks; blk_x < wipe_right; blk_x++) { - buffer[offset_y][blk_x][0] = (JCOEF) average; - } - } - } - } -} - - -LOCAL(void) -do_reflect (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - JDIMENSION drop_width, JDIMENSION drop_height) -/* Reflect - drop content of specified area, similar to wipe, but - * fill with repeated reflections of the outside area, instead of zero. - * NB: y_crop_offset is assumed to be zero. - */ -{ - JDIMENSION x_wipe_blocks, wipe_width; - JDIMENSION y_wipe_blocks, wipe_bottom; - JDIMENSION src_blk_x, dst_blk_x; - int ci, k, offset_y; - JBLOCKARRAY buffer; - JBLOCKROW src_row_ptr, dst_row_ptr; - JCOEFPTR src_ptr, dst_ptr; - jpeg_component_info *compptr; - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - x_wipe_blocks = x_crop_offset * compptr->h_samp_factor; - wipe_width = drop_width * compptr->h_samp_factor; - wipe_bottom = drop_height * compptr->v_samp_factor; - for (y_wipe_blocks = 0; y_wipe_blocks < wipe_bottom; - y_wipe_blocks += compptr->v_samp_factor) { - buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], y_wipe_blocks, - (JDIMENSION) compptr->v_samp_factor, TRUE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - if (x_wipe_blocks > 0) { - /* Reflect from left */ - dst_row_ptr = buffer[offset_y] + x_wipe_blocks; - for (dst_blk_x = wipe_width; dst_blk_x > 0;) { - src_row_ptr = dst_row_ptr; /* (re)set axis of reflection */ - for (src_blk_x = x_wipe_blocks; - src_blk_x > 0 && dst_blk_x > 0; src_blk_x--, dst_blk_x--) { - dst_ptr = *dst_row_ptr++; /* destination goes right */ - src_ptr = *--src_row_ptr; /* source goes left */ - /* this unrolled loop doesn't need to know which row it's on... */ - for (k = 0; k < DCTSIZE2; k += 2) { - *dst_ptr++ = *src_ptr++; /* copy even column */ - *dst_ptr++ = - *src_ptr++; /* copy odd column with sign change */ - } - } - } - } else if (compptr->width_in_blocks > x_wipe_blocks + wipe_width) { - /* Reflect from right */ - dst_row_ptr = buffer[offset_y] + x_wipe_blocks + wipe_width; - for (dst_blk_x = wipe_width; dst_blk_x > 0;) { - src_row_ptr = dst_row_ptr; /* (re)set axis of reflection */ - src_blk_x = compptr->width_in_blocks - x_wipe_blocks - wipe_width; - for (; src_blk_x > 0 && dst_blk_x > 0; src_blk_x--, dst_blk_x--) { - dst_ptr = *--dst_row_ptr; /* destination goes left */ - src_ptr = *src_row_ptr++; /* source goes right */ - /* this unrolled loop doesn't need to know which row it's on... */ - for (k = 0; k < DCTSIZE2; k += 2) { - *dst_ptr++ = *src_ptr++; /* copy even column */ - *dst_ptr++ = - *src_ptr++; /* copy odd column with sign change */ - } - } - } - } else { - FMEMZERO(buffer[offset_y] + x_wipe_blocks, - wipe_width * SIZEOF(JBLOCK)); - } - } - } - } -} - - -LOCAL(void) -do_flip_h_no_crop (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, - jvirt_barray_ptr *src_coef_arrays) -/* Horizontal flip; done in-place, so no separate dest array is required. - * NB: this only works when y_crop_offset is zero. - */ -{ - JDIMENSION MCU_cols, comp_width, blk_x, blk_y, x_crop_blocks; - int ci, k, offset_y; - JBLOCKARRAY buffer; - JCOEFPTR ptr1, ptr2; - JCOEF temp1, temp2; - jpeg_component_info *compptr; - - /* Horizontal mirroring of DCT blocks is accomplished by swapping - * pairs of blocks in-place. Within a DCT block, we perform horizontal - * mirroring by changing the signs of odd-numbered columns. - * Partial iMCUs at the right edge are left untouched. - */ - MCU_cols = srcinfo->output_width / - (dstinfo->max_h_samp_factor * dstinfo->min_DCT_h_scaled_size); - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - comp_width = MCU_cols * compptr->h_samp_factor; - x_crop_blocks = x_crop_offset * compptr->h_samp_factor; - for (blk_y = 0; blk_y < compptr->height_in_blocks; - blk_y += compptr->v_samp_factor) { - buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - /* Do the mirroring */ - for (blk_x = 0; blk_x * 2 < comp_width; blk_x++) { - ptr1 = buffer[offset_y][blk_x]; - ptr2 = buffer[offset_y][comp_width - blk_x - 1]; - /* this unrolled loop doesn't need to know which row it's on... */ - for (k = 0; k < DCTSIZE2; k += 2) { - temp1 = *ptr1; /* swap even column */ - temp2 = *ptr2; - *ptr1++ = temp2; - *ptr2++ = temp1; - temp1 = *ptr1; /* swap odd column with sign change */ - temp2 = *ptr2; - *ptr1++ = -temp2; - *ptr2++ = -temp1; - } - } - if (x_crop_blocks > 0) { - /* Now left-justify the portion of the data to be kept. - * We can't use a single jcopy_block_row() call because that routine - * depends on memcpy(), whose behavior is unspecified for overlapping - * source and destination areas. Sigh. - */ - for (blk_x = 0; blk_x < compptr->width_in_blocks; blk_x++) { - jcopy_block_row(buffer[offset_y] + blk_x + x_crop_blocks, - buffer[offset_y] + blk_x, - (JDIMENSION) 1); - } - } - } - } - } -} - - -LOCAL(void) -do_flip_h (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - jvirt_barray_ptr *dst_coef_arrays) -/* Horizontal flip in general cropping case */ -{ - JDIMENSION MCU_cols, comp_width, dst_blk_x, dst_blk_y; - JDIMENSION x_crop_blocks, y_crop_blocks; - int ci, k, offset_y; - JBLOCKARRAY src_buffer, dst_buffer; - JBLOCKROW src_row_ptr, dst_row_ptr; - JCOEFPTR src_ptr, dst_ptr; - jpeg_component_info *compptr; - - /* Here we must output into a separate array because we can't touch - * different rows of a single virtual array simultaneously. Otherwise, - * this is essentially the same as the routine above. - */ - MCU_cols = srcinfo->output_width / - (dstinfo->max_h_samp_factor * dstinfo->min_DCT_h_scaled_size); - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - comp_width = MCU_cols * compptr->h_samp_factor; - x_crop_blocks = x_crop_offset * compptr->h_samp_factor; - y_crop_blocks = y_crop_offset * compptr->v_samp_factor; - for (dst_blk_y = 0; dst_blk_y < compptr->height_in_blocks; - dst_blk_y += compptr->v_samp_factor) { - dst_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, dst_coef_arrays[ci], dst_blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_y + y_crop_blocks, - (JDIMENSION) compptr->v_samp_factor, FALSE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - dst_row_ptr = dst_buffer[offset_y]; - src_row_ptr = src_buffer[offset_y]; - for (dst_blk_x = 0; dst_blk_x < compptr->width_in_blocks; dst_blk_x++) { - if (x_crop_blocks + dst_blk_x < comp_width) { - /* Do the mirrorable blocks */ - dst_ptr = dst_row_ptr[dst_blk_x]; - src_ptr = src_row_ptr[comp_width - x_crop_blocks - dst_blk_x - 1]; - /* this unrolled loop doesn't need to know which row it's on... */ - for (k = 0; k < DCTSIZE2; k += 2) { - *dst_ptr++ = *src_ptr++; /* copy even column */ - *dst_ptr++ = - *src_ptr++; /* copy odd column with sign change */ - } - } else { - /* Copy last partial block(s) verbatim */ - jcopy_block_row(src_row_ptr + dst_blk_x + x_crop_blocks, - dst_row_ptr + dst_blk_x, - (JDIMENSION) 1); - } - } - } - } - } -} - - -LOCAL(void) -do_flip_v (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - jvirt_barray_ptr *dst_coef_arrays) -/* Vertical flip */ -{ - JDIMENSION MCU_rows, comp_height, dst_blk_x, dst_blk_y; - JDIMENSION x_crop_blocks, y_crop_blocks; - int ci, i, j, offset_y; - JBLOCKARRAY src_buffer, dst_buffer; - JBLOCKROW src_row_ptr, dst_row_ptr; - JCOEFPTR src_ptr, dst_ptr; - jpeg_component_info *compptr; - - /* We output into a separate array because we can't touch different - * rows of the source virtual array simultaneously. Otherwise, this - * is a pretty straightforward analog of horizontal flip. - * Within a DCT block, vertical mirroring is done by changing the signs - * of odd-numbered rows. - * Partial iMCUs at the bottom edge are copied verbatim. - */ - MCU_rows = srcinfo->output_height / - (dstinfo->max_v_samp_factor * dstinfo->min_DCT_v_scaled_size); - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - comp_height = MCU_rows * compptr->v_samp_factor; - x_crop_blocks = x_crop_offset * compptr->h_samp_factor; - y_crop_blocks = y_crop_offset * compptr->v_samp_factor; - for (dst_blk_y = 0; dst_blk_y < compptr->height_in_blocks; - dst_blk_y += compptr->v_samp_factor) { - dst_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, dst_coef_arrays[ci], dst_blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - if (y_crop_blocks + dst_blk_y < comp_height) { - /* Row is within the mirrorable area. */ - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - comp_height - y_crop_blocks - dst_blk_y - - (JDIMENSION) compptr->v_samp_factor, - (JDIMENSION) compptr->v_samp_factor, FALSE); - } else { - /* Bottom-edge blocks will be copied verbatim. */ - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_y + y_crop_blocks, - (JDIMENSION) compptr->v_samp_factor, FALSE); - } - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - if (y_crop_blocks + dst_blk_y < comp_height) { - /* Row is within the mirrorable area. */ - dst_row_ptr = dst_buffer[offset_y]; - src_row_ptr = src_buffer[compptr->v_samp_factor - offset_y - 1]; - src_row_ptr += x_crop_blocks; - for (dst_blk_x = 0; dst_blk_x < compptr->width_in_blocks; - dst_blk_x++) { - dst_ptr = dst_row_ptr[dst_blk_x]; - src_ptr = src_row_ptr[dst_blk_x]; - for (i = 0; i < DCTSIZE; i += 2) { - /* copy even row */ - for (j = 0; j < DCTSIZE; j++) - *dst_ptr++ = *src_ptr++; - /* copy odd row with sign change */ - for (j = 0; j < DCTSIZE; j++) - *dst_ptr++ = - *src_ptr++; - } - } - } else { - /* Just copy row verbatim. */ - jcopy_block_row(src_buffer[offset_y] + x_crop_blocks, - dst_buffer[offset_y], - compptr->width_in_blocks); - } - } - } - } -} - - -LOCAL(void) -do_transpose (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - jvirt_barray_ptr *dst_coef_arrays) -/* Transpose source into destination */ -{ - JDIMENSION dst_blk_x, dst_blk_y, x_crop_blocks, y_crop_blocks; - int ci, i, j, offset_x, offset_y; - JBLOCKARRAY src_buffer, dst_buffer; - JCOEFPTR src_ptr, dst_ptr; - jpeg_component_info *compptr; - - /* Transposing pixels within a block just requires transposing the - * DCT coefficients. - * Partial iMCUs at the edges require no special treatment; we simply - * process all the available DCT blocks for every component. - */ - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - x_crop_blocks = x_crop_offset * compptr->h_samp_factor; - y_crop_blocks = y_crop_offset * compptr->v_samp_factor; - for (dst_blk_y = 0; dst_blk_y < compptr->height_in_blocks; - dst_blk_y += compptr->v_samp_factor) { - dst_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, dst_coef_arrays[ci], dst_blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - for (dst_blk_x = 0; dst_blk_x < compptr->width_in_blocks; - dst_blk_x += compptr->h_samp_factor) { - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_x + x_crop_blocks, - (JDIMENSION) compptr->h_samp_factor, FALSE); - for (offset_x = 0; offset_x < compptr->h_samp_factor; offset_x++) { - dst_ptr = dst_buffer[offset_y][dst_blk_x + offset_x]; - src_ptr = src_buffer[offset_x][dst_blk_y + offset_y + y_crop_blocks]; - for (i = 0; i < DCTSIZE; i++) - for (j = 0; j < DCTSIZE; j++) - dst_ptr[j*DCTSIZE+i] = src_ptr[i*DCTSIZE+j]; - } - } - } - } - } -} - - -LOCAL(void) -do_rot_90 (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - jvirt_barray_ptr *dst_coef_arrays) -/* 90 degree rotation is equivalent to - * 1. Transposing the image; - * 2. Horizontal mirroring. - * These two steps are merged into a single processing routine. - */ -{ - JDIMENSION MCU_cols, comp_width, dst_blk_x, dst_blk_y; - JDIMENSION x_crop_blocks, y_crop_blocks; - int ci, i, j, offset_x, offset_y; - JBLOCKARRAY src_buffer, dst_buffer; - JCOEFPTR src_ptr, dst_ptr; - jpeg_component_info *compptr; - - /* Because of the horizontal mirror step, we can't process partial iMCUs - * at the (output) right edge properly. They just get transposed and - * not mirrored. - */ - MCU_cols = srcinfo->output_height / - (dstinfo->max_h_samp_factor * dstinfo->min_DCT_h_scaled_size); - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - comp_width = MCU_cols * compptr->h_samp_factor; - x_crop_blocks = x_crop_offset * compptr->h_samp_factor; - y_crop_blocks = y_crop_offset * compptr->v_samp_factor; - for (dst_blk_y = 0; dst_blk_y < compptr->height_in_blocks; - dst_blk_y += compptr->v_samp_factor) { - dst_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, dst_coef_arrays[ci], dst_blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - for (dst_blk_x = 0; dst_blk_x < compptr->width_in_blocks; - dst_blk_x += compptr->h_samp_factor) { - if (x_crop_blocks + dst_blk_x < comp_width) { - /* Block is within the mirrorable area. */ - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - comp_width - x_crop_blocks - dst_blk_x - - (JDIMENSION) compptr->h_samp_factor, - (JDIMENSION) compptr->h_samp_factor, FALSE); - } else { - /* Edge blocks are transposed but not mirrored. */ - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_x + x_crop_blocks, - (JDIMENSION) compptr->h_samp_factor, FALSE); - } - for (offset_x = 0; offset_x < compptr->h_samp_factor; offset_x++) { - dst_ptr = dst_buffer[offset_y][dst_blk_x + offset_x]; - if (x_crop_blocks + dst_blk_x < comp_width) { - /* Block is within the mirrorable area. */ - src_ptr = src_buffer[compptr->h_samp_factor - offset_x - 1] - [dst_blk_y + offset_y + y_crop_blocks]; - for (i = 0; i < DCTSIZE; i++) { - for (j = 0; j < DCTSIZE; j++) - dst_ptr[j*DCTSIZE+i] = src_ptr[i*DCTSIZE+j]; - i++; - for (j = 0; j < DCTSIZE; j++) - dst_ptr[j*DCTSIZE+i] = -src_ptr[i*DCTSIZE+j]; - } - } else { - /* Edge blocks are transposed but not mirrored. */ - src_ptr = src_buffer[offset_x] - [dst_blk_y + offset_y + y_crop_blocks]; - for (i = 0; i < DCTSIZE; i++) - for (j = 0; j < DCTSIZE; j++) - dst_ptr[j*DCTSIZE+i] = src_ptr[i*DCTSIZE+j]; - } - } - } - } - } - } -} - - -LOCAL(void) -do_rot_270 (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - jvirt_barray_ptr *dst_coef_arrays) -/* 270 degree rotation is equivalent to - * 1. Horizontal mirroring; - * 2. Transposing the image. - * These two steps are merged into a single processing routine. - */ -{ - JDIMENSION MCU_rows, comp_height, dst_blk_x, dst_blk_y; - JDIMENSION x_crop_blocks, y_crop_blocks; - int ci, i, j, offset_x, offset_y; - JBLOCKARRAY src_buffer, dst_buffer; - JCOEFPTR src_ptr, dst_ptr; - jpeg_component_info *compptr; - - /* Because of the horizontal mirror step, we can't process partial iMCUs - * at the (output) bottom edge properly. They just get transposed and - * not mirrored. - */ - MCU_rows = srcinfo->output_width / - (dstinfo->max_v_samp_factor * dstinfo->min_DCT_v_scaled_size); - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - comp_height = MCU_rows * compptr->v_samp_factor; - x_crop_blocks = x_crop_offset * compptr->h_samp_factor; - y_crop_blocks = y_crop_offset * compptr->v_samp_factor; - for (dst_blk_y = 0; dst_blk_y < compptr->height_in_blocks; - dst_blk_y += compptr->v_samp_factor) { - dst_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, dst_coef_arrays[ci], dst_blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - for (dst_blk_x = 0; dst_blk_x < compptr->width_in_blocks; - dst_blk_x += compptr->h_samp_factor) { - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_x + x_crop_blocks, - (JDIMENSION) compptr->h_samp_factor, FALSE); - for (offset_x = 0; offset_x < compptr->h_samp_factor; offset_x++) { - dst_ptr = dst_buffer[offset_y][dst_blk_x + offset_x]; - if (y_crop_blocks + dst_blk_y < comp_height) { - /* Block is within the mirrorable area. */ - src_ptr = src_buffer[offset_x] - [comp_height - y_crop_blocks - dst_blk_y - offset_y - 1]; - for (i = 0; i < DCTSIZE; i++) { - for (j = 0; j < DCTSIZE; j++) { - dst_ptr[j*DCTSIZE+i] = src_ptr[i*DCTSIZE+j]; - j++; - dst_ptr[j*DCTSIZE+i] = -src_ptr[i*DCTSIZE+j]; - } - } - } else { - /* Edge blocks are transposed but not mirrored. */ - src_ptr = src_buffer[offset_x] - [dst_blk_y + offset_y + y_crop_blocks]; - for (i = 0; i < DCTSIZE; i++) - for (j = 0; j < DCTSIZE; j++) - dst_ptr[j*DCTSIZE+i] = src_ptr[i*DCTSIZE+j]; - } - } - } - } - } - } -} - - -LOCAL(void) -do_rot_180 (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - jvirt_barray_ptr *dst_coef_arrays) -/* 180 degree rotation is equivalent to - * 1. Vertical mirroring; - * 2. Horizontal mirroring. - * These two steps are merged into a single processing routine. - */ -{ - JDIMENSION MCU_cols, MCU_rows, comp_width, comp_height, dst_blk_x, dst_blk_y; - JDIMENSION x_crop_blocks, y_crop_blocks; - int ci, i, j, offset_y; - JBLOCKARRAY src_buffer, dst_buffer; - JBLOCKROW src_row_ptr, dst_row_ptr; - JCOEFPTR src_ptr, dst_ptr; - jpeg_component_info *compptr; - - MCU_cols = srcinfo->output_width / - (dstinfo->max_h_samp_factor * dstinfo->min_DCT_h_scaled_size); - MCU_rows = srcinfo->output_height / - (dstinfo->max_v_samp_factor * dstinfo->min_DCT_v_scaled_size); - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - comp_width = MCU_cols * compptr->h_samp_factor; - comp_height = MCU_rows * compptr->v_samp_factor; - x_crop_blocks = x_crop_offset * compptr->h_samp_factor; - y_crop_blocks = y_crop_offset * compptr->v_samp_factor; - for (dst_blk_y = 0; dst_blk_y < compptr->height_in_blocks; - dst_blk_y += compptr->v_samp_factor) { - dst_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, dst_coef_arrays[ci], dst_blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - if (y_crop_blocks + dst_blk_y < comp_height) { - /* Row is within the vertically mirrorable area. */ - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - comp_height - y_crop_blocks - dst_blk_y - - (JDIMENSION) compptr->v_samp_factor, - (JDIMENSION) compptr->v_samp_factor, FALSE); - } else { - /* Bottom-edge rows are only mirrored horizontally. */ - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_y + y_crop_blocks, - (JDIMENSION) compptr->v_samp_factor, FALSE); - } - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - dst_row_ptr = dst_buffer[offset_y]; - if (y_crop_blocks + dst_blk_y < comp_height) { - /* Row is within the mirrorable area. */ - src_row_ptr = src_buffer[compptr->v_samp_factor - offset_y - 1]; - for (dst_blk_x = 0; dst_blk_x < compptr->width_in_blocks; dst_blk_x++) { - dst_ptr = dst_row_ptr[dst_blk_x]; - if (x_crop_blocks + dst_blk_x < comp_width) { - /* Process the blocks that can be mirrored both ways. */ - src_ptr = src_row_ptr[comp_width - x_crop_blocks - dst_blk_x - 1]; - for (i = 0; i < DCTSIZE; i += 2) { - /* For even row, negate every odd column. */ - for (j = 0; j < DCTSIZE; j += 2) { - *dst_ptr++ = *src_ptr++; - *dst_ptr++ = - *src_ptr++; - } - /* For odd row, negate every even column. */ - for (j = 0; j < DCTSIZE; j += 2) { - *dst_ptr++ = - *src_ptr++; - *dst_ptr++ = *src_ptr++; - } - } - } else { - /* Any remaining right-edge blocks are only mirrored vertically. */ - src_ptr = src_row_ptr[x_crop_blocks + dst_blk_x]; - for (i = 0; i < DCTSIZE; i += 2) { - for (j = 0; j < DCTSIZE; j++) - *dst_ptr++ = *src_ptr++; - for (j = 0; j < DCTSIZE; j++) - *dst_ptr++ = - *src_ptr++; - } - } - } - } else { - /* Remaining rows are just mirrored horizontally. */ - src_row_ptr = src_buffer[offset_y]; - for (dst_blk_x = 0; dst_blk_x < compptr->width_in_blocks; dst_blk_x++) { - if (x_crop_blocks + dst_blk_x < comp_width) { - /* Process the blocks that can be mirrored. */ - dst_ptr = dst_row_ptr[dst_blk_x]; - src_ptr = src_row_ptr[comp_width - x_crop_blocks - dst_blk_x - 1]; - for (i = 0; i < DCTSIZE2; i += 2) { - *dst_ptr++ = *src_ptr++; - *dst_ptr++ = - *src_ptr++; - } - } else { - /* Any remaining right-edge blocks are only copied. */ - jcopy_block_row(src_row_ptr + dst_blk_x + x_crop_blocks, - dst_row_ptr + dst_blk_x, - (JDIMENSION) 1); - } - } - } - } - } - } -} - - -LOCAL(void) -do_transverse (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JDIMENSION x_crop_offset, JDIMENSION y_crop_offset, - jvirt_barray_ptr *src_coef_arrays, - jvirt_barray_ptr *dst_coef_arrays) -/* Transverse transpose is equivalent to - * 1. 180 degree rotation; - * 2. Transposition; - * or - * 1. Horizontal mirroring; - * 2. Transposition; - * 3. Horizontal mirroring. - * These steps are merged into a single processing routine. - */ -{ - JDIMENSION MCU_cols, MCU_rows, comp_width, comp_height, dst_blk_x, dst_blk_y; - JDIMENSION x_crop_blocks, y_crop_blocks; - int ci, i, j, offset_x, offset_y; - JBLOCKARRAY src_buffer, dst_buffer; - JCOEFPTR src_ptr, dst_ptr; - jpeg_component_info *compptr; - - MCU_cols = srcinfo->output_height / - (dstinfo->max_h_samp_factor * dstinfo->min_DCT_h_scaled_size); - MCU_rows = srcinfo->output_width / - (dstinfo->max_v_samp_factor * dstinfo->min_DCT_v_scaled_size); - - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - comp_width = MCU_cols * compptr->h_samp_factor; - comp_height = MCU_rows * compptr->v_samp_factor; - x_crop_blocks = x_crop_offset * compptr->h_samp_factor; - y_crop_blocks = y_crop_offset * compptr->v_samp_factor; - for (dst_blk_y = 0; dst_blk_y < compptr->height_in_blocks; - dst_blk_y += compptr->v_samp_factor) { - dst_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, dst_coef_arrays[ci], dst_blk_y, - (JDIMENSION) compptr->v_samp_factor, TRUE); - for (offset_y = 0; offset_y < compptr->v_samp_factor; offset_y++) { - for (dst_blk_x = 0; dst_blk_x < compptr->width_in_blocks; - dst_blk_x += compptr->h_samp_factor) { - if (x_crop_blocks + dst_blk_x < comp_width) { - /* Block is within the mirrorable area. */ - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - comp_width - x_crop_blocks - dst_blk_x - - (JDIMENSION) compptr->h_samp_factor, - (JDIMENSION) compptr->h_samp_factor, FALSE); - } else { - src_buffer = (*srcinfo->mem->access_virt_barray) - ((j_common_ptr) srcinfo, src_coef_arrays[ci], - dst_blk_x + x_crop_blocks, - (JDIMENSION) compptr->h_samp_factor, FALSE); - } - for (offset_x = 0; offset_x < compptr->h_samp_factor; offset_x++) { - dst_ptr = dst_buffer[offset_y][dst_blk_x + offset_x]; - if (y_crop_blocks + dst_blk_y < comp_height) { - if (x_crop_blocks + dst_blk_x < comp_width) { - /* Block is within the mirrorable area. */ - src_ptr = src_buffer[compptr->h_samp_factor - offset_x - 1] - [comp_height - y_crop_blocks - dst_blk_y - offset_y - 1]; - for (i = 0; i < DCTSIZE; i++) { - for (j = 0; j < DCTSIZE; j++) { - dst_ptr[j*DCTSIZE+i] = src_ptr[i*DCTSIZE+j]; - j++; - dst_ptr[j*DCTSIZE+i] = -src_ptr[i*DCTSIZE+j]; - } - i++; - for (j = 0; j < DCTSIZE; j++) { - dst_ptr[j*DCTSIZE+i] = -src_ptr[i*DCTSIZE+j]; - j++; - dst_ptr[j*DCTSIZE+i] = src_ptr[i*DCTSIZE+j]; - } - } - } else { - /* Right-edge blocks are mirrored in y only */ - src_ptr = src_buffer[offset_x] - [comp_height - y_crop_blocks - dst_blk_y - offset_y - 1]; - for (i = 0; i < DCTSIZE; i++) { - for (j = 0; j < DCTSIZE; j++) { - dst_ptr[j*DCTSIZE+i] = src_ptr[i*DCTSIZE+j]; - j++; - dst_ptr[j*DCTSIZE+i] = -src_ptr[i*DCTSIZE+j]; - } - } - } - } else { - if (x_crop_blocks + dst_blk_x < comp_width) { - /* Bottom-edge blocks are mirrored in x only */ - src_ptr = src_buffer[compptr->h_samp_factor - offset_x - 1] - [dst_blk_y + offset_y + y_crop_blocks]; - for (i = 0; i < DCTSIZE; i++) { - for (j = 0; j < DCTSIZE; j++) - dst_ptr[j*DCTSIZE+i] = src_ptr[i*DCTSIZE+j]; - i++; - for (j = 0; j < DCTSIZE; j++) - dst_ptr[j*DCTSIZE+i] = -src_ptr[i*DCTSIZE+j]; - } - } else { - /* At lower right corner, just transpose, no mirroring */ - src_ptr = src_buffer[offset_x] - [dst_blk_y + offset_y + y_crop_blocks]; - for (i = 0; i < DCTSIZE; i++) - for (j = 0; j < DCTSIZE; j++) - dst_ptr[j*DCTSIZE+i] = src_ptr[i*DCTSIZE+j]; - } - } - } - } - } - } - } -} - - -/* Parse an unsigned integer: subroutine for jtransform_parse_crop_spec. - * Returns TRUE if valid integer found, FALSE if not. - * *strptr is advanced over the digit string, and *result is set to its value. - */ - -LOCAL(boolean) -jt_read_integer (const char ** strptr, JDIMENSION * result) -{ - const char * ptr = *strptr; - JDIMENSION val = 0; - - for (; isdigit(*ptr); ptr++) { - val = val * 10 + (JDIMENSION) (*ptr - '0'); - } - *result = val; - if (ptr == *strptr) - return FALSE; /* oops, no digits */ - *strptr = ptr; - return TRUE; -} - - -/* Parse a crop specification (written in X11 geometry style). - * The routine returns TRUE if the spec string is valid, FALSE if not. - * - * The crop spec string should have the format - * [{fr}]x[{fr}]{+-}{+-} - * where width, height, xoffset, and yoffset are unsigned integers. - * Each of the elements can be omitted to indicate a default value. - * (A weakness of this style is that it is not possible to omit xoffset - * while specifying yoffset, since they look alike.) - * - * This code is loosely based on XParseGeometry from the X11 distribution. - */ - -GLOBAL(boolean) -jtransform_parse_crop_spec (jpeg_transform_info *info, const char *spec) -{ - info->crop = FALSE; - info->crop_width_set = JCROP_UNSET; - info->crop_height_set = JCROP_UNSET; - info->crop_xoffset_set = JCROP_UNSET; - info->crop_yoffset_set = JCROP_UNSET; - - if (isdigit(*spec)) { - /* fetch width */ - if (! jt_read_integer(&spec, &info->crop_width)) - return FALSE; - if (*spec == 'f' || *spec == 'F') { - spec++; - info->crop_width_set = JCROP_FORCE; - } else if (*spec == 'r' || *spec == 'R') { - spec++; - info->crop_width_set = JCROP_REFLECT; - } else - info->crop_width_set = JCROP_POS; - } - if (*spec == 'x' || *spec == 'X') { - /* fetch height */ - spec++; - if (! jt_read_integer(&spec, &info->crop_height)) - return FALSE; - if (*spec == 'f' || *spec == 'F') { - spec++; - info->crop_height_set = JCROP_FORCE; - } else if (*spec == 'r' || *spec == 'R') { - spec++; - info->crop_height_set = JCROP_REFLECT; - } else - info->crop_height_set = JCROP_POS; - } - if (*spec == '+' || *spec == '-') { - /* fetch xoffset */ - info->crop_xoffset_set = (*spec == '-') ? JCROP_NEG : JCROP_POS; - spec++; - if (! jt_read_integer(&spec, &info->crop_xoffset)) - return FALSE; - } - if (*spec == '+' || *spec == '-') { - /* fetch yoffset */ - info->crop_yoffset_set = (*spec == '-') ? JCROP_NEG : JCROP_POS; - spec++; - if (! jt_read_integer(&spec, &info->crop_yoffset)) - return FALSE; - } - /* We had better have gotten to the end of the string. */ - if (*spec != '\0') - return FALSE; - info->crop = TRUE; - return TRUE; -} - - -/* Trim off any partial iMCUs on the indicated destination edge */ - -LOCAL(void) -trim_right_edge (jpeg_transform_info *info, JDIMENSION full_width) -{ - JDIMENSION MCU_cols; - - MCU_cols = info->output_width / info->iMCU_sample_width; - if (MCU_cols > 0 && info->x_crop_offset + MCU_cols == - full_width / info->iMCU_sample_width) - info->output_width = MCU_cols * info->iMCU_sample_width; -} - -LOCAL(void) -trim_bottom_edge (jpeg_transform_info *info, JDIMENSION full_height) -{ - JDIMENSION MCU_rows; - - MCU_rows = info->output_height / info->iMCU_sample_height; - if (MCU_rows > 0 && info->y_crop_offset + MCU_rows == - full_height / info->iMCU_sample_height) - info->output_height = MCU_rows * info->iMCU_sample_height; -} - - -/* Request any required workspace. - * - * This routine figures out the size that the output image will be - * (which implies that all the transform parameters must be set before - * it is called). - * - * We allocate the workspace virtual arrays from the source decompression - * object, so that all the arrays (both the original data and the workspace) - * will be taken into account while making memory management decisions. - * Hence, this routine must be called after jpeg_read_header (which reads - * the image dimensions) and before jpeg_read_coefficients (which realizes - * the source's virtual arrays). - * - * This function returns FALSE right away if -perfect is given - * and transformation is not perfect. Otherwise returns TRUE. - */ - -GLOBAL(boolean) -jtransform_request_workspace (j_decompress_ptr srcinfo, - jpeg_transform_info *info) -{ - jvirt_barray_ptr *coef_arrays; - boolean need_workspace, transpose_it; - jpeg_component_info *compptr; - JDIMENSION xoffset, yoffset, dtemp; - JDIMENSION width_in_iMCUs, height_in_iMCUs; - JDIMENSION width_in_blocks, height_in_blocks; - int itemp, ci, h_samp_factor, v_samp_factor; - - /* Determine number of components in output image */ - if (info->force_grayscale && - (srcinfo->jpeg_color_space == JCS_YCbCr || - srcinfo->jpeg_color_space == JCS_BG_YCC) && - srcinfo->num_components == 3) - /* We'll only process the first component */ - info->num_components = 1; - else - /* Process all the components */ - info->num_components = srcinfo->num_components; - - /* Compute output image dimensions and related values. */ - jpeg_core_output_dimensions(srcinfo); - - /* Return right away if -perfect is given and transformation is not perfect. - */ - if (info->perfect) { - if (info->num_components == 1) { - if (!jtransform_perfect_transform(srcinfo->output_width, - srcinfo->output_height, - srcinfo->min_DCT_h_scaled_size, - srcinfo->min_DCT_v_scaled_size, - info->transform)) - return FALSE; - } else { - if (!jtransform_perfect_transform(srcinfo->output_width, - srcinfo->output_height, - srcinfo->max_h_samp_factor * srcinfo->min_DCT_h_scaled_size, - srcinfo->max_v_samp_factor * srcinfo->min_DCT_v_scaled_size, - info->transform)) - return FALSE; - } - } - - /* If there is only one output component, force the iMCU size to be 1; - * else use the source iMCU size. (This allows us to do the right thing - * when reducing color to grayscale, and also provides a handy way of - * cleaning up "funny" grayscale images whose sampling factors are not 1x1.) - */ - switch (info->transform) { - case JXFORM_TRANSPOSE: - case JXFORM_TRANSVERSE: - case JXFORM_ROT_90: - case JXFORM_ROT_270: - info->output_width = srcinfo->output_height; - info->output_height = srcinfo->output_width; - if (info->num_components == 1) { - info->iMCU_sample_width = srcinfo->min_DCT_v_scaled_size; - info->iMCU_sample_height = srcinfo->min_DCT_h_scaled_size; - } else { - info->iMCU_sample_width = - srcinfo->max_v_samp_factor * srcinfo->min_DCT_v_scaled_size; - info->iMCU_sample_height = - srcinfo->max_h_samp_factor * srcinfo->min_DCT_h_scaled_size; - } - break; - default: - info->output_width = srcinfo->output_width; - info->output_height = srcinfo->output_height; - if (info->num_components == 1) { - info->iMCU_sample_width = srcinfo->min_DCT_h_scaled_size; - info->iMCU_sample_height = srcinfo->min_DCT_v_scaled_size; - } else { - info->iMCU_sample_width = - srcinfo->max_h_samp_factor * srcinfo->min_DCT_h_scaled_size; - info->iMCU_sample_height = - srcinfo->max_v_samp_factor * srcinfo->min_DCT_v_scaled_size; - } - } - - /* If cropping has been requested, compute the crop area's position and - * dimensions, ensuring that its upper left corner falls at an iMCU boundary. - */ - if (info->crop) { - /* Insert default values for unset crop parameters */ - if (info->crop_xoffset_set == JCROP_UNSET) - info->crop_xoffset = 0; /* default to +0 */ - if (info->crop_yoffset_set == JCROP_UNSET) - info->crop_yoffset = 0; /* default to +0 */ - if (info->crop_width_set == JCROP_UNSET) { - if (info->crop_xoffset >= info->output_width) - ERREXIT(srcinfo, JERR_BAD_CROP_SPEC); - info->crop_width = info->output_width - info->crop_xoffset; - } else { - /* Check for crop extension */ - if (info->crop_width > info->output_width) { - /* Crop extension does not work when transforming! */ - if (info->transform != JXFORM_NONE || - info->crop_xoffset >= info->crop_width || - info->crop_xoffset > info->crop_width - info->output_width) - ERREXIT(srcinfo, JERR_BAD_CROP_SPEC); - } else { - if (info->crop_xoffset >= info->output_width || - info->crop_width <= 0 || - info->crop_xoffset > info->output_width - info->crop_width) - ERREXIT(srcinfo, JERR_BAD_CROP_SPEC); - } - } - if (info->crop_height_set == JCROP_UNSET) { - if (info->crop_yoffset >= info->output_height) - ERREXIT(srcinfo, JERR_BAD_CROP_SPEC); - info->crop_height = info->output_height - info->crop_yoffset; - } else { - /* Check for crop extension */ - if (info->crop_height > info->output_height) { - /* Crop extension does not work when transforming! */ - if (info->transform != JXFORM_NONE || - info->crop_yoffset >= info->crop_height || - info->crop_yoffset > info->crop_height - info->output_height) - ERREXIT(srcinfo, JERR_BAD_CROP_SPEC); - } else { - if (info->crop_yoffset >= info->output_height || - info->crop_height <= 0 || - info->crop_yoffset > info->output_height - info->crop_height) - ERREXIT(srcinfo, JERR_BAD_CROP_SPEC); - } - } - /* Convert negative crop offsets into regular offsets */ - if (info->crop_xoffset_set != JCROP_NEG) - xoffset = info->crop_xoffset; - else if (info->crop_width > info->output_width) /* crop extension */ - xoffset = info->crop_width - info->output_width - info->crop_xoffset; - else - xoffset = info->output_width - info->crop_width - info->crop_xoffset; - if (info->crop_yoffset_set != JCROP_NEG) - yoffset = info->crop_yoffset; - else if (info->crop_height > info->output_height) /* crop extension */ - yoffset = info->crop_height - info->output_height - info->crop_yoffset; - else - yoffset = info->output_height - info->crop_height - info->crop_yoffset; - /* Now adjust so that upper left corner falls at an iMCU boundary */ - switch (info->transform) { - case JXFORM_DROP: - /* Ensure the effective drop region will not exceed the requested */ - itemp = info->iMCU_sample_width; - dtemp = itemp - 1 - ((xoffset + itemp - 1) % itemp); - xoffset += dtemp; - if (info->crop_width <= dtemp) - info->drop_width = 0; - else if (xoffset + info->crop_width - dtemp == info->output_width) - /* Matching right edge: include partial iMCU */ - info->drop_width = (info->crop_width - dtemp + itemp - 1) / itemp; - else - info->drop_width = (info->crop_width - dtemp) / itemp; - itemp = info->iMCU_sample_height; - dtemp = itemp - 1 - ((yoffset + itemp - 1) % itemp); - yoffset += dtemp; - if (info->crop_height <= dtemp) - info->drop_height = 0; - else if (yoffset + info->crop_height - dtemp == info->output_height) - /* Matching bottom edge: include partial iMCU */ - info->drop_height = (info->crop_height - dtemp + itemp - 1) / itemp; - else - info->drop_height = (info->crop_height - dtemp) / itemp; - /* Check if sampling factors match for dropping */ - if (info->drop_width != 0 && info->drop_height != 0) - for (ci = 0; ci < info->num_components && - ci < info->drop_ptr->num_components; ci++) { - if (info->drop_ptr->comp_info[ci].h_samp_factor * - srcinfo->max_h_samp_factor != - srcinfo->comp_info[ci].h_samp_factor * - info->drop_ptr->max_h_samp_factor) - ERREXIT6(srcinfo, JERR_BAD_DROP_SAMPLING, ci, - info->drop_ptr->comp_info[ci].h_samp_factor, - info->drop_ptr->max_h_samp_factor, - srcinfo->comp_info[ci].h_samp_factor, - srcinfo->max_h_samp_factor, 'h'); - if (info->drop_ptr->comp_info[ci].v_samp_factor * - srcinfo->max_v_samp_factor != - srcinfo->comp_info[ci].v_samp_factor * - info->drop_ptr->max_v_samp_factor) - ERREXIT6(srcinfo, JERR_BAD_DROP_SAMPLING, ci, - info->drop_ptr->comp_info[ci].v_samp_factor, - info->drop_ptr->max_v_samp_factor, - srcinfo->comp_info[ci].v_samp_factor, - srcinfo->max_v_samp_factor, 'v'); - } - break; - case JXFORM_WIPE: - /* Ensure the effective wipe region will cover the requested */ - info->drop_width = (JDIMENSION) jdiv_round_up - ((long) (info->crop_width + (xoffset % info->iMCU_sample_width)), - (long) info->iMCU_sample_width); - info->drop_height = (JDIMENSION) jdiv_round_up - ((long) (info->crop_height + (yoffset % info->iMCU_sample_height)), - (long) info->iMCU_sample_height); - break; - default: - /* Ensure the effective crop region will cover the requested */ - if (info->crop_width_set == JCROP_FORCE || - info->crop_width > info->output_width) - info->output_width = info->crop_width; - else - info->output_width = - info->crop_width + (xoffset % info->iMCU_sample_width); - if (info->crop_height_set == JCROP_FORCE || - info->crop_height > info->output_height) - info->output_height = info->crop_height; - else - info->output_height = - info->crop_height + (yoffset % info->iMCU_sample_height); - } - /* Save x/y offsets measured in iMCUs */ - info->x_crop_offset = xoffset / info->iMCU_sample_width; - info->y_crop_offset = yoffset / info->iMCU_sample_height; - } else { - info->x_crop_offset = 0; - info->y_crop_offset = 0; - } - - /* Figure out whether we need workspace arrays, - * and if so whether they are transposed relative to the source. - */ - need_workspace = FALSE; - transpose_it = FALSE; - switch (info->transform) { - case JXFORM_NONE: - if (info->x_crop_offset != 0 || info->y_crop_offset != 0 || - info->output_width > srcinfo->output_width || - info->output_height > srcinfo->output_height) - need_workspace = TRUE; - /* No workspace needed if neither cropping nor transforming */ - break; - case JXFORM_FLIP_H: - if (info->trim) - trim_right_edge(info, srcinfo->output_width); - if (info->y_crop_offset != 0) - need_workspace = TRUE; - /* do_flip_h_no_crop doesn't need a workspace array */ - break; - case JXFORM_FLIP_V: - if (info->trim) - trim_bottom_edge(info, srcinfo->output_height); - /* Need workspace arrays having same dimensions as source image. */ - need_workspace = TRUE; - break; - case JXFORM_TRANSPOSE: - /* transpose does NOT have to trim anything */ - /* Need workspace arrays having transposed dimensions. */ - need_workspace = TRUE; - transpose_it = TRUE; - break; - case JXFORM_TRANSVERSE: - if (info->trim) { - trim_right_edge(info, srcinfo->output_height); - trim_bottom_edge(info, srcinfo->output_width); - } - /* Need workspace arrays having transposed dimensions. */ - need_workspace = TRUE; - transpose_it = TRUE; - break; - case JXFORM_ROT_90: - if (info->trim) - trim_right_edge(info, srcinfo->output_height); - /* Need workspace arrays having transposed dimensions. */ - need_workspace = TRUE; - transpose_it = TRUE; - break; - case JXFORM_ROT_180: - if (info->trim) { - trim_right_edge(info, srcinfo->output_width); - trim_bottom_edge(info, srcinfo->output_height); - } - /* Need workspace arrays having same dimensions as source image. */ - need_workspace = TRUE; - break; - case JXFORM_ROT_270: - if (info->trim) - trim_bottom_edge(info, srcinfo->output_width); - /* Need workspace arrays having transposed dimensions. */ - need_workspace = TRUE; - transpose_it = TRUE; - break; - case JXFORM_WIPE: - break; - case JXFORM_DROP: -#if DROP_REQUEST_FROM_SRC - drop_request_from_src(info->drop_ptr, srcinfo); -#endif - break; - } - - /* Allocate workspace if needed. - * Note that we allocate arrays padded out to the next iMCU boundary, - * so that transform routines need not worry about missing edge blocks. - */ - if (need_workspace) { - coef_arrays = (jvirt_barray_ptr *) (*srcinfo->mem->alloc_small) - ((j_common_ptr) srcinfo, JPOOL_IMAGE, - SIZEOF(jvirt_barray_ptr) * info->num_components); - width_in_iMCUs = (JDIMENSION) jdiv_round_up - ((long) info->output_width, (long) info->iMCU_sample_width); - height_in_iMCUs = (JDIMENSION) jdiv_round_up - ((long) info->output_height, (long) info->iMCU_sample_height); - for (ci = 0; ci < info->num_components; ci++) { - compptr = srcinfo->comp_info + ci; - if (info->num_components == 1) { - /* we're going to force samp factors to 1x1 in this case */ - h_samp_factor = v_samp_factor = 1; - } else if (transpose_it) { - h_samp_factor = compptr->v_samp_factor; - v_samp_factor = compptr->h_samp_factor; - } else { - h_samp_factor = compptr->h_samp_factor; - v_samp_factor = compptr->v_samp_factor; - } - width_in_blocks = width_in_iMCUs * h_samp_factor; - height_in_blocks = height_in_iMCUs * v_samp_factor; - coef_arrays[ci] = (*srcinfo->mem->request_virt_barray) - ((j_common_ptr) srcinfo, JPOOL_IMAGE, FALSE, - width_in_blocks, height_in_blocks, (JDIMENSION) v_samp_factor); - } - info->workspace_coef_arrays = coef_arrays; - } else - info->workspace_coef_arrays = NULL; - - return TRUE; -} - - -/* Transpose destination image parameters */ - -LOCAL(void) -transpose_critical_parameters (j_compress_ptr dstinfo) -{ - int tblno, i, j, ci, itemp; - jpeg_component_info *compptr; - JQUANT_TBL *qtblptr; - JDIMENSION jtemp; - UINT16 qtemp; - - /* Transpose image dimensions */ - jtemp = dstinfo->image_width; - dstinfo->image_width = dstinfo->image_height; - dstinfo->image_height = jtemp; - itemp = dstinfo->min_DCT_h_scaled_size; - dstinfo->min_DCT_h_scaled_size = dstinfo->min_DCT_v_scaled_size; - dstinfo->min_DCT_v_scaled_size = itemp; - - /* Transpose sampling factors */ - for (ci = 0; ci < dstinfo->num_components; ci++) { - compptr = dstinfo->comp_info + ci; - itemp = compptr->h_samp_factor; - compptr->h_samp_factor = compptr->v_samp_factor; - compptr->v_samp_factor = itemp; - } - - /* Transpose quantization tables */ - for (tblno = 0; tblno < NUM_QUANT_TBLS; tblno++) { - qtblptr = dstinfo->quant_tbl_ptrs[tblno]; - if (qtblptr != NULL) { - for (i = 0; i < DCTSIZE; i++) { - for (j = 0; j < i; j++) { - qtemp = qtblptr->quantval[i*DCTSIZE+j]; - qtblptr->quantval[i*DCTSIZE+j] = qtblptr->quantval[j*DCTSIZE+i]; - qtblptr->quantval[j*DCTSIZE+i] = qtemp; - } - } - } - } -} - - -/* Adjust Exif image parameters. - * - * We try to adjust the Tags ExifImageWidth and ExifImageHeight if possible. - */ - -LOCAL(void) -adjust_exif_parameters (JOCTET FAR * data, unsigned int length, - JDIMENSION new_width, JDIMENSION new_height) -{ - boolean is_motorola; /* Flag for byte order */ - unsigned int number_of_tags, tagnum; - unsigned int firstoffset, offset; - JDIMENSION new_value; - - if (length < 12) return; /* Length of an IFD entry */ - - /* Discover byte order */ - if (GETJOCTET(data[0]) == 0x49 && GETJOCTET(data[1]) == 0x49) - is_motorola = FALSE; - else if (GETJOCTET(data[0]) == 0x4D && GETJOCTET(data[1]) == 0x4D) - is_motorola = TRUE; - else - return; - - /* Check Tag Mark */ - if (is_motorola) { - if (GETJOCTET(data[2]) != 0) return; - if (GETJOCTET(data[3]) != 0x2A) return; - } else { - if (GETJOCTET(data[3]) != 0) return; - if (GETJOCTET(data[2]) != 0x2A) return; - } - - /* Get first IFD offset (offset to IFD0) */ - if (is_motorola) { - if (GETJOCTET(data[4]) != 0) return; - if (GETJOCTET(data[5]) != 0) return; - firstoffset = GETJOCTET(data[6]); - firstoffset <<= 8; - firstoffset += GETJOCTET(data[7]); - } else { - if (GETJOCTET(data[7]) != 0) return; - if (GETJOCTET(data[6]) != 0) return; - firstoffset = GETJOCTET(data[5]); - firstoffset <<= 8; - firstoffset += GETJOCTET(data[4]); - } - if (firstoffset > length - 2) return; /* check end of data segment */ - - /* Get the number of directory entries contained in this IFD */ - if (is_motorola) { - number_of_tags = GETJOCTET(data[firstoffset]); - number_of_tags <<= 8; - number_of_tags += GETJOCTET(data[firstoffset+1]); - } else { - number_of_tags = GETJOCTET(data[firstoffset+1]); - number_of_tags <<= 8; - number_of_tags += GETJOCTET(data[firstoffset]); - } - if (number_of_tags == 0) return; - firstoffset += 2; - - /* Search for ExifSubIFD offset Tag in IFD0 */ - for (;;) { - if (firstoffset > length - 12) return; /* check end of data segment */ - /* Get Tag number */ - if (is_motorola) { - tagnum = GETJOCTET(data[firstoffset]); - tagnum <<= 8; - tagnum += GETJOCTET(data[firstoffset+1]); - } else { - tagnum = GETJOCTET(data[firstoffset+1]); - tagnum <<= 8; - tagnum += GETJOCTET(data[firstoffset]); - } - if (tagnum == 0x8769) break; /* found ExifSubIFD offset Tag */ - if (--number_of_tags == 0) return; - firstoffset += 12; - } - - /* Get the ExifSubIFD offset */ - if (is_motorola) { - if (GETJOCTET(data[firstoffset+8]) != 0) return; - if (GETJOCTET(data[firstoffset+9]) != 0) return; - offset = GETJOCTET(data[firstoffset+10]); - offset <<= 8; - offset += GETJOCTET(data[firstoffset+11]); - } else { - if (GETJOCTET(data[firstoffset+11]) != 0) return; - if (GETJOCTET(data[firstoffset+10]) != 0) return; - offset = GETJOCTET(data[firstoffset+9]); - offset <<= 8; - offset += GETJOCTET(data[firstoffset+8]); - } - if (offset > length - 2) return; /* check end of data segment */ - - /* Get the number of directory entries contained in this SubIFD */ - if (is_motorola) { - number_of_tags = GETJOCTET(data[offset]); - number_of_tags <<= 8; - number_of_tags += GETJOCTET(data[offset+1]); - } else { - number_of_tags = GETJOCTET(data[offset+1]); - number_of_tags <<= 8; - number_of_tags += GETJOCTET(data[offset]); - } - if (number_of_tags < 2) return; - offset += 2; - - /* Search for ExifImageWidth and ExifImageHeight Tags in this SubIFD */ - do { - if (offset > length - 12) return; /* check end of data segment */ - /* Get Tag number */ - if (is_motorola) { - tagnum = GETJOCTET(data[offset]); - tagnum <<= 8; - tagnum += GETJOCTET(data[offset+1]); - } else { - tagnum = GETJOCTET(data[offset+1]); - tagnum <<= 8; - tagnum += GETJOCTET(data[offset]); - } - if (tagnum == 0xA002 || tagnum == 0xA003) { - if (tagnum == 0xA002) - new_value = new_width; /* ExifImageWidth Tag */ - else - new_value = new_height; /* ExifImageHeight Tag */ - if (is_motorola) { - data[offset+2] = 0; /* Format = unsigned long (4 octets) */ - data[offset+3] = 4; - data[offset+4] = 0; /* Number Of Components = 1 */ - data[offset+5] = 0; - data[offset+6] = 0; - data[offset+7] = 1; - data[offset+8] = 0; - data[offset+9] = 0; - data[offset+10] = (JOCTET)((new_value >> 8) & 0xFF); - data[offset+11] = (JOCTET)(new_value & 0xFF); - } else { - data[offset+2] = 4; /* Format = unsigned long (4 octets) */ - data[offset+3] = 0; - data[offset+4] = 1; /* Number Of Components = 1 */ - data[offset+5] = 0; - data[offset+6] = 0; - data[offset+7] = 0; - data[offset+8] = (JOCTET)(new_value & 0xFF); - data[offset+9] = (JOCTET)((new_value >> 8) & 0xFF); - data[offset+10] = 0; - data[offset+11] = 0; - } - } - offset += 12; - } while (--number_of_tags); -} - - -/* Adjust output image parameters as needed. - * - * This must be called after jpeg_copy_critical_parameters() - * and before jpeg_write_coefficients(). - * - * The return value is the set of virtual coefficient arrays to be written - * (either the ones allocated by jtransform_request_workspace, or the - * original source data arrays). The caller will need to pass this value - * to jpeg_write_coefficients(). - */ - -GLOBAL(jvirt_barray_ptr *) -jtransform_adjust_parameters (j_decompress_ptr srcinfo, - j_compress_ptr dstinfo, - jvirt_barray_ptr *src_coef_arrays, - jpeg_transform_info *info) -{ - /* If force-to-grayscale is requested, adjust destination parameters */ - if (info->force_grayscale) { - /* First, ensure we have YCC or grayscale data, and that the source's - * Y channel is full resolution. (No reasonable person would make Y - * be less than full resolution, so actually coping with that case - * isn't worth extra code space. But we check it to avoid crashing.) - */ - if ((((dstinfo->jpeg_color_space == JCS_YCbCr || - dstinfo->jpeg_color_space == JCS_BG_YCC) && - dstinfo->num_components == 3) || - (dstinfo->jpeg_color_space == JCS_GRAYSCALE && - dstinfo->num_components == 1)) && - srcinfo->comp_info[0].h_samp_factor == srcinfo->max_h_samp_factor && - srcinfo->comp_info[0].v_samp_factor == srcinfo->max_v_samp_factor) { - /* We use jpeg_set_colorspace to make sure subsidiary settings get fixed - * properly. Among other things, it sets the target h_samp_factor & - * v_samp_factor to 1, which typically won't match the source. - * We have to preserve the source's quantization table number, however. - */ - int sv_quant_tbl_no = dstinfo->comp_info[0].quant_tbl_no; - jpeg_set_colorspace(dstinfo, JCS_GRAYSCALE); - dstinfo->comp_info[0].quant_tbl_no = sv_quant_tbl_no; - } else { - /* Sorry, can't do it */ - ERREXIT(dstinfo, JERR_CONVERSION_NOTIMPL); - } - } else if (info->num_components == 1) { - /* For a single-component source, we force the destination sampling factors - * to 1x1, with or without force_grayscale. This is useful because some - * decoders choke on grayscale images with other sampling factors. - */ - dstinfo->comp_info[0].h_samp_factor = 1; - dstinfo->comp_info[0].v_samp_factor = 1; - } - - /* Correct the destination's image dimensions as necessary - * for rotate/flip, resize, and crop operations. - */ - dstinfo->jpeg_width = info->output_width; - dstinfo->jpeg_height = info->output_height; - - /* Transpose destination image parameters, adjust quantization */ - switch (info->transform) { - case JXFORM_TRANSPOSE: - case JXFORM_TRANSVERSE: - case JXFORM_ROT_90: - case JXFORM_ROT_270: - transpose_critical_parameters(dstinfo); - break; - case JXFORM_DROP: - if (info->drop_width != 0 && info->drop_height != 0) - adjust_quant(srcinfo, src_coef_arrays, - info->drop_ptr, info->drop_coef_arrays, - info->trim, dstinfo); - break; - default: - break; - } - - /* Adjust Exif properties */ - if (srcinfo->marker_list != NULL && - srcinfo->marker_list->marker == JPEG_APP0+1 && - srcinfo->marker_list->data_length >= 6 && - GETJOCTET(srcinfo->marker_list->data[0]) == 0x45 && - GETJOCTET(srcinfo->marker_list->data[1]) == 0x78 && - GETJOCTET(srcinfo->marker_list->data[2]) == 0x69 && - GETJOCTET(srcinfo->marker_list->data[3]) == 0x66 && - GETJOCTET(srcinfo->marker_list->data[4]) == 0 && - GETJOCTET(srcinfo->marker_list->data[5]) == 0) { - /* Suppress output of JFIF marker */ - dstinfo->write_JFIF_header = FALSE; - /* Adjust Exif image parameters */ - if (dstinfo->jpeg_width != srcinfo->image_width || - dstinfo->jpeg_height != srcinfo->image_height) - /* Align data segment to start of TIFF structure for parsing */ - adjust_exif_parameters(srcinfo->marker_list->data + 6, - srcinfo->marker_list->data_length - 6, - dstinfo->jpeg_width, dstinfo->jpeg_height); - } - - /* Return the appropriate output data set */ - if (info->workspace_coef_arrays != NULL) - return info->workspace_coef_arrays; - return src_coef_arrays; -} - - -/* Execute the actual transformation, if any. - * - * This must be called *after* jpeg_write_coefficients, because it depends - * on jpeg_write_coefficients to have computed subsidiary values such as - * the per-component width and height fields in the destination object. - * - * Note that some transformations will modify the source data arrays! - */ - -GLOBAL(void) -jtransform_execute_transform (j_decompress_ptr srcinfo, - j_compress_ptr dstinfo, - jvirt_barray_ptr *src_coef_arrays, - jpeg_transform_info *info) -{ - jvirt_barray_ptr *dst_coef_arrays = info->workspace_coef_arrays; - - /* Note: conditions tested here should match those in switch statement - * in jtransform_request_workspace() - */ - switch (info->transform) { - case JXFORM_NONE: - if (info->output_width > srcinfo->output_width || - info->output_height > srcinfo->output_height) { - if (info->output_width > srcinfo->output_width && - info->crop_width_set == JCROP_REFLECT) - do_crop_ext_reflect(srcinfo, dstinfo, - info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, dst_coef_arrays); - else if (info->output_width > srcinfo->output_width && - info->crop_width_set == JCROP_FORCE) - do_crop_ext_flat(srcinfo, dstinfo, - info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, dst_coef_arrays); - else - do_crop_ext_zero(srcinfo, dstinfo, - info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, dst_coef_arrays); - } else if (info->x_crop_offset != 0 || info->y_crop_offset != 0) - do_crop(srcinfo, dstinfo, info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, dst_coef_arrays); - break; - case JXFORM_FLIP_H: - if (info->y_crop_offset != 0) - do_flip_h(srcinfo, dstinfo, info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, dst_coef_arrays); - else - do_flip_h_no_crop(srcinfo, dstinfo, info->x_crop_offset, - src_coef_arrays); - break; - case JXFORM_FLIP_V: - do_flip_v(srcinfo, dstinfo, info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, dst_coef_arrays); - break; - case JXFORM_TRANSPOSE: - do_transpose(srcinfo, dstinfo, info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, dst_coef_arrays); - break; - case JXFORM_TRANSVERSE: - do_transverse(srcinfo, dstinfo, info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, dst_coef_arrays); - break; - case JXFORM_ROT_90: - do_rot_90(srcinfo, dstinfo, info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, dst_coef_arrays); - break; - case JXFORM_ROT_180: - do_rot_180(srcinfo, dstinfo, info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, dst_coef_arrays); - break; - case JXFORM_ROT_270: - do_rot_270(srcinfo, dstinfo, info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, dst_coef_arrays); - break; - case JXFORM_WIPE: - if (info->crop_width_set == JCROP_REFLECT && - info->y_crop_offset == 0 && info->drop_height == - (JDIMENSION) jdiv_round_up - ((long) info->output_height, (long) info->iMCU_sample_height) && - (info->x_crop_offset == 0 || - info->x_crop_offset + info->drop_width == - (JDIMENSION) jdiv_round_up - ((long) info->output_width, (long) info->iMCU_sample_width))) - do_reflect(srcinfo, dstinfo, info->x_crop_offset, - src_coef_arrays, info->drop_width, info->drop_height); - else if (info->crop_width_set == JCROP_FORCE) - do_flatten(srcinfo, dstinfo, info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, info->drop_width, info->drop_height); - else - do_wipe(srcinfo, dstinfo, info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, info->drop_width, info->drop_height); - break; - case JXFORM_DROP: - if (info->drop_width != 0 && info->drop_height != 0) - do_drop(srcinfo, dstinfo, info->x_crop_offset, info->y_crop_offset, - src_coef_arrays, info->drop_ptr, info->drop_coef_arrays, - info->drop_width, info->drop_height); - break; - } -} - -/* jtransform_perfect_transform - * - * Determine whether lossless transformation is perfectly - * possible for a specified image and transformation. - * - * Inputs: - * image_width, image_height: source image dimensions. - * MCU_width, MCU_height: pixel dimensions of MCU. - * transform: transformation identifier. - * Parameter sources from initialized jpeg_struct - * (after reading source header): - * image_width = cinfo.image_width - * image_height = cinfo.image_height - * MCU_width = cinfo.max_h_samp_factor * cinfo.block_size - * MCU_height = cinfo.max_v_samp_factor * cinfo.block_size - * Result: - * TRUE = perfect transformation possible - * FALSE = perfect transformation not possible - * (may use custom action then) - */ - -GLOBAL(boolean) -jtransform_perfect_transform(JDIMENSION image_width, JDIMENSION image_height, - int MCU_width, int MCU_height, - JXFORM_CODE transform) -{ - boolean result = TRUE; /* initialize TRUE */ - - switch (transform) { - case JXFORM_FLIP_H: - case JXFORM_ROT_270: - if (image_width % (JDIMENSION) MCU_width) - result = FALSE; - break; - case JXFORM_FLIP_V: - case JXFORM_ROT_90: - if (image_height % (JDIMENSION) MCU_height) - result = FALSE; - break; - case JXFORM_TRANSVERSE: - case JXFORM_ROT_180: - if (image_width % (JDIMENSION) MCU_width) - result = FALSE; - if (image_height % (JDIMENSION) MCU_height) - result = FALSE; - break; - default: - break; - } - - return result; -} - -#endif /* TRANSFORMS_SUPPORTED */ - - -/* Setup decompression object to save desired markers in memory. - * This must be called before jpeg_read_header() to have the desired effect. - */ - -GLOBAL(void) -jcopy_markers_setup (j_decompress_ptr srcinfo, JCOPY_OPTION option) -{ -#ifdef SAVE_MARKERS_SUPPORTED - int m; - - /* Save comments except under NONE option */ - if (option != JCOPYOPT_NONE) { - jpeg_save_markers(srcinfo, JPEG_COM, 0xFFFF); - } - /* Save all types of APPn markers iff ALL option */ - if (option == JCOPYOPT_ALL) { - for (m = 0; m < 16; m++) - jpeg_save_markers(srcinfo, JPEG_APP0 + m, 0xFFFF); - } -#endif /* SAVE_MARKERS_SUPPORTED */ -} - -/* Copy markers saved in the given source object to the destination object. - * This should be called just after jpeg_start_compress() or - * jpeg_write_coefficients(). - * Note that those routines will have written the SOI, and also the - * JFIF APP0 or Adobe APP14 markers if selected. - */ - -GLOBAL(void) -jcopy_markers_execute (j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JCOPY_OPTION option) -{ - jpeg_saved_marker_ptr marker; - - /* In the current implementation, we don't actually need to examine the - * option flag here; we just copy everything that got saved. - * But to avoid confusion, we do not output JFIF and Adobe APP14 markers - * if the encoder library already wrote one. - */ - for (marker = srcinfo->marker_list; marker != NULL; marker = marker->next) { - if (dstinfo->write_JFIF_header && - marker->marker == JPEG_APP0 && - marker->data_length >= 5 && - GETJOCTET(marker->data[0]) == 0x4A && - GETJOCTET(marker->data[1]) == 0x46 && - GETJOCTET(marker->data[2]) == 0x49 && - GETJOCTET(marker->data[3]) == 0x46 && - GETJOCTET(marker->data[4]) == 0) - continue; /* reject duplicate JFIF */ - if (dstinfo->write_Adobe_marker && - marker->marker == JPEG_APP0+14 && - marker->data_length >= 5 && - GETJOCTET(marker->data[0]) == 0x41 && - GETJOCTET(marker->data[1]) == 0x64 && - GETJOCTET(marker->data[2]) == 0x6F && - GETJOCTET(marker->data[3]) == 0x62 && - GETJOCTET(marker->data[4]) == 0x65) - continue; /* reject duplicate Adobe */ -#ifdef NEED_FAR_POINTERS - /* We could use jpeg_write_marker if the data weren't FAR... */ - { - unsigned int i; - jpeg_write_m_header(dstinfo, marker->marker, marker->data_length); - for (i = 0; i < marker->data_length; i++) - jpeg_write_m_byte(dstinfo, marker->data[i]); - } -#else - jpeg_write_marker(dstinfo, marker->marker, - marker->data, marker->data_length); -#endif - } -} diff --git a/dep/libjpeg/src/transupp.h b/dep/libjpeg/src/transupp.h deleted file mode 100644 index a8ba16ad7..000000000 --- a/dep/libjpeg/src/transupp.h +++ /dev/null @@ -1,230 +0,0 @@ -/* - * transupp.h - * - * Copyright (C) 1997-2019, Thomas G. Lane, Guido Vollbeding. - * This file is part of the Independent JPEG Group's software. - * For conditions of distribution and use, see the accompanying README file. - * - * This file contains declarations for image transformation routines and - * other utility code used by the jpegtran sample application. These are - * NOT part of the core JPEG library. But we keep these routines separate - * from jpegtran.c to ease the task of maintaining jpegtran-like programs - * that have other user interfaces. - * - * NOTE: all the routines declared here have very specific requirements - * about when they are to be executed during the reading and writing of the - * source and destination files. See the comments in transupp.c, or see - * jpegtran.c for an example of correct usage. - */ - -/* If you happen not to want the image transform support, disable it here */ -#ifndef TRANSFORMS_SUPPORTED -#define TRANSFORMS_SUPPORTED 1 /* 0 disables transform code */ -#endif - -/* - * Although rotating and flipping data expressed as DCT coefficients is not - * hard, there is an asymmetry in the JPEG format specification for images - * whose dimensions aren't multiples of the iMCU size. The right and bottom - * image edges are padded out to the next iMCU boundary with junk data; but - * no padding is possible at the top and left edges. If we were to flip - * the whole image including the pad data, then pad garbage would become - * visible at the top and/or left, and real pixels would disappear into the - * pad margins --- perhaps permanently, since encoders & decoders may not - * bother to preserve DCT blocks that appear to be completely outside the - * nominal image area. So, we have to exclude any partial iMCUs from the - * basic transformation. - * - * Transpose is the only transformation that can handle partial iMCUs at the - * right and bottom edges completely cleanly. flip_h can flip partial iMCUs - * at the bottom, but leaves any partial iMCUs at the right edge untouched. - * Similarly flip_v leaves any partial iMCUs at the bottom edge untouched. - * The other transforms are defined as combinations of these basic transforms - * and process edge blocks in a way that preserves the equivalence. - * - * The "trim" option causes untransformable partial iMCUs to be dropped; - * this is not strictly lossless, but it usually gives the best-looking - * result for odd-size images. Note that when this option is active, - * the expected mathematical equivalences between the transforms may not hold. - * (For example, -rot 270 -trim trims only the bottom edge, but -rot 90 -trim - * followed by -rot 180 -trim trims both edges.) - * - * We also offer a lossless-crop option, which discards data outside a given - * image region but losslessly preserves what is inside. Like the rotate and - * flip transforms, lossless crop is restricted by the current JPEG format: the - * upper left corner of the selected region must fall on an iMCU boundary. If - * this does not hold for the given crop parameters, we silently move the upper - * left corner up and/or left to make it so, simultaneously increasing the - * region dimensions to keep the lower right crop corner unchanged. (Thus, the - * output image covers at least the requested region, but may cover more.) - * The adjustment of the region dimensions may be optionally disabled. - * - * A complementary lossless-wipe option is provided to discard (gray out) data - * inside a given image region while losslessly preserving what is outside. - * Another option is lossless-drop, which replaces data at a given image - * position by another image. Both source images must have the same - * subsampling values. It is best if they also have the same quantization, - * otherwise quantization adaption occurs. The trim option can be used with - * the drop option to requantize the drop file to the source file. - * - * We also provide a lossless-resize option, which is kind of a lossless-crop - * operation in the DCT coefficient block domain - it discards higher-order - * coefficients and losslessly preserves lower-order coefficients of a - * sub-block. - * - * Rotate/flip transform, resize, and crop can be requested together in a - * single invocation. The crop is applied last --- that is, the crop region - * is specified in terms of the destination image after transform/resize. - * - * We also offer a "force to grayscale" option, which simply discards the - * chrominance channels of a YCbCr image. This is lossless in the sense that - * the luminance channel is preserved exactly. It's not the same kind of - * thing as the rotate/flip transformations, but it's convenient to handle it - * as part of this package, mainly because the transformation routines have to - * be aware of the option to know how many components to work on. - */ - - -/* Short forms of external names for systems with brain-damaged linkers. */ - -#ifdef NEED_SHORT_EXTERNAL_NAMES -#define jtransform_parse_crop_spec jTrParCrop -#define jtransform_request_workspace jTrRequest -#define jtransform_adjust_parameters jTrAdjust -#define jtransform_execute_transform jTrExec -#define jtransform_perfect_transform jTrPerfect -#define jcopy_markers_setup jCMrkSetup -#define jcopy_markers_execute jCMrkExec -#endif /* NEED_SHORT_EXTERNAL_NAMES */ - - -/* - * Codes for supported types of image transformations. - */ - -typedef enum { - JXFORM_NONE, /* no transformation */ - JXFORM_FLIP_H, /* horizontal flip */ - JXFORM_FLIP_V, /* vertical flip */ - JXFORM_TRANSPOSE, /* transpose across UL-to-LR axis */ - JXFORM_TRANSVERSE, /* transpose across UR-to-LL axis */ - JXFORM_ROT_90, /* 90-degree clockwise rotation */ - JXFORM_ROT_180, /* 180-degree rotation */ - JXFORM_ROT_270, /* 270-degree clockwise (or 90 ccw) */ - JXFORM_WIPE, /* wipe */ - JXFORM_DROP /* drop */ -} JXFORM_CODE; - -/* - * Codes for crop parameters, which can individually be unspecified, - * positive or negative for xoffset or yoffset, - * positive or force or reflect for width or height. - */ - -typedef enum { - JCROP_UNSET, - JCROP_POS, - JCROP_NEG, - JCROP_FORCE, - JCROP_REFLECT -} JCROP_CODE; - -/* - * Transform parameters struct. - * NB: application must not change any elements of this struct after - * calling jtransform_request_workspace. - */ - -typedef struct { - /* Options: set by caller */ - JXFORM_CODE transform; /* image transform operator */ - boolean perfect; /* if TRUE, fail if partial MCUs are requested */ - boolean trim; /* if TRUE, trim partial MCUs as needed */ - boolean force_grayscale; /* if TRUE, convert color image to grayscale */ - boolean crop; /* if TRUE, crop or wipe source image, or drop */ - - /* Crop parameters: application need not set these unless crop is TRUE. - * These can be filled in by jtransform_parse_crop_spec(). - */ - JDIMENSION crop_width; /* Width of selected region */ - JCROP_CODE crop_width_set; /* (force disables adjustment) */ - JDIMENSION crop_height; /* Height of selected region */ - JCROP_CODE crop_height_set; /* (force disables adjustment) */ - JDIMENSION crop_xoffset; /* X offset of selected region */ - JCROP_CODE crop_xoffset_set; /* (negative measures from right edge) */ - JDIMENSION crop_yoffset; /* Y offset of selected region */ - JCROP_CODE crop_yoffset_set; /* (negative measures from bottom edge) */ - - /* Drop parameters: set by caller for drop request */ - j_decompress_ptr drop_ptr; - jvirt_barray_ptr * drop_coef_arrays; - - /* Internal workspace: caller should not touch these */ - int num_components; /* # of components in workspace */ - jvirt_barray_ptr * workspace_coef_arrays; /* workspace for transformations */ - JDIMENSION output_width; /* cropped destination dimensions */ - JDIMENSION output_height; - JDIMENSION x_crop_offset; /* destination crop offsets measured in iMCUs */ - JDIMENSION y_crop_offset; - JDIMENSION drop_width; /* drop/wipe dimensions measured in iMCUs */ - JDIMENSION drop_height; - int iMCU_sample_width; /* destination iMCU size */ - int iMCU_sample_height; -} jpeg_transform_info; - - -#if TRANSFORMS_SUPPORTED - -/* Parse a crop specification (written in X11 geometry style) */ -EXTERN(boolean) jtransform_parse_crop_spec - JPP((jpeg_transform_info *info, const char *spec)); -/* Request any required workspace */ -EXTERN(boolean) jtransform_request_workspace - JPP((j_decompress_ptr srcinfo, jpeg_transform_info *info)); -/* Adjust output image parameters */ -EXTERN(jvirt_barray_ptr *) jtransform_adjust_parameters - JPP((j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - jvirt_barray_ptr *src_coef_arrays, - jpeg_transform_info *info)); -/* Execute the actual transformation, if any */ -EXTERN(void) jtransform_execute_transform - JPP((j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - jvirt_barray_ptr *src_coef_arrays, - jpeg_transform_info *info)); -/* Determine whether lossless transformation is perfectly - * possible for a specified image and transformation. - */ -EXTERN(boolean) jtransform_perfect_transform - JPP((JDIMENSION image_width, JDIMENSION image_height, - int MCU_width, int MCU_height, - JXFORM_CODE transform)); - -/* jtransform_execute_transform used to be called - * jtransform_execute_transformation, but some compilers complain about - * routine names that long. This macro is here to avoid breaking any - * old source code that uses the original name... - */ -#define jtransform_execute_transformation jtransform_execute_transform - -#endif /* TRANSFORMS_SUPPORTED */ - - -/* - * Support for copying optional markers from source to destination file. - */ - -typedef enum { - JCOPYOPT_NONE, /* copy no optional markers */ - JCOPYOPT_COMMENTS, /* copy only comment (COM) markers */ - JCOPYOPT_ALL /* copy all optional markers */ -} JCOPY_OPTION; - -#define JCOPYOPT_DEFAULT JCOPYOPT_COMMENTS /* recommended default */ - -/* Setup decompression object to save desired markers in memory */ -EXTERN(void) jcopy_markers_setup - JPP((j_decompress_ptr srcinfo, JCOPY_OPTION option)); -/* Copy markers saved in the given source object to the destination object */ -EXTERN(void) jcopy_markers_execute - JPP((j_decompress_ptr srcinfo, j_compress_ptr dstinfo, - JCOPY_OPTION option)); diff --git a/dep/libpng/ANNOUNCE b/dep/libpng/ANNOUNCE deleted file mode 100644 index bc147adb7..000000000 --- a/dep/libpng/ANNOUNCE +++ /dev/null @@ -1,63 +0,0 @@ -libpng 1.6.43 - February 23, 2024 -================================= - -This is a public release of libpng, intended for use in production code. - - -Files available for download ----------------------------- - -Source files with LF line endings (for Unix/Linux): - - * libpng-1.6.43.tar.xz (LZMA-compressed, recommended) - * libpng-1.6.43.tar.gz (deflate-compressed) - -Source files with CRLF line endings (for Windows): - - * lpng1643.7z (LZMA-compressed, recommended) - * lpng1643.zip (deflate-compressed) - -Other information: - - * README.md - * LICENSE.md - * AUTHORS.md - * TRADEMARK.md - - -Changes from version 1.6.42 to version 1.6.43 ---------------------------------------------- - - * Fixed the row width check in png_check_IHDR(). - This corrected a bug that was specific to the 16-bit platforms, - and removed a spurious compiler warning from the 64-bit builds. - (Reported by Jacek Caban; fixed by John Bowler) - * Added eXIf chunk support to the push-mode reader in pngpread.c. - (Contributed by Chris Blume) - * Added contrib/pngexif for the benefit of the users who would like - to inspect the content of eXIf chunks. - * Added contrib/conftest/basic.dfa, a basic build-time configuration. - (Contributed by John Bowler) - * Fixed a preprocessor condition in pngread.c that broke build-time - configurations like contrib/conftest/pngcp.dfa. - (Contributed by John Bowler) - * Added CMake build support for LoongArch LSX. - (Contributed by GuXiWei) - * Fixed a CMake build error that occurred under a peculiar state of the - dependency tree. This was a regression introduced in libpng-1.6.41. - (Contributed by Dan Rosser) - * Marked the installed libpng headers as system headers in CMake. - (Contributed by Benjamin Buch) - * Updated the build support for RISCOS. - (Contributed by Cameron Cawley) - * Updated the makefiles to allow cross-platform builds to initialize - conventional make variables like AR and ARFLAGS. - * Added various improvements to the CI scripts in areas like version - consistency verification and text linting. - * Added version consistency verification to pngtest.c also. - - -Send comments/corrections/commendations to png-mng-implement at lists.sf.net. -Subscription is required; visit -https://lists.sourceforge.net/lists/listinfo/png-mng-implement -to subscribe. diff --git a/dep/libpng/AUTHORS b/dep/libpng/AUTHORS deleted file mode 100644 index 544341694..000000000 --- a/dep/libpng/AUTHORS +++ /dev/null @@ -1,61 +0,0 @@ -PNG REFERENCE LIBRARY AUTHORS -============================= - -This is the list of PNG Reference Library ("libpng") Contributing -Authors, for copyright and licensing purposes. - - * Adam Richter - * Andreas Dilger - * Chris Blume - * Cosmin Truta - * Dave Martindale - * Eric S. Raymond - * Gilles Vollant - * Glenn Randers-Pehrson - * Greg Roelofs - * Guy Eric Schalnat - * James Yu - * John Bowler - * Kevin Bracey - * Magnus Holmgren - * Mandar Sahastrabuddhe - * Mans Rullgard - * Matt Sarett - * Mike Klein - * Pascal Massimino - * Paul Schmidt - * Philippe Antoine - * Qiang Zhou - * Sam Bushell - * Samuel Williams - * Simon-Pierre Cadieux - * Tim Wegner - * Tom Lane - * Tom Tanner - * Vadim Barkov - * Willem van Schaik - * Zhijie Liang - * Apple Inc. - - Zixu Wang (王子旭) - * Arm Holdings - - Richard Townsend - * Google Inc. - - Dan Field - - Leon Scroggins III - - Matt Sarett - - Mike Klein - - Sami Boukortt - - Wan-Teh Chang - * Loongson Technology Corporation Ltd. - - GuXiWei (顾希伟) - - JinBo (金波) - - ZhangLixia (张利霞) - -The build projects, the build scripts, the test scripts, and other -files in the "projects", "scripts" and "tests" directories, have -other copyright owners, but are released under the libpng license. - -Some files in the "ci" and "contrib" directories, as well as some -of the tools-generated files that are distributed with libpng, have -other copyright owners, and are released under other open source -licenses. diff --git a/dep/libpng/CHANGES b/dep/libpng/CHANGES deleted file mode 100644 index 441b57ecf..000000000 --- a/dep/libpng/CHANGES +++ /dev/null @@ -1,6202 +0,0 @@ -CHANGES - changes for libpng - -version 0.1 [March 29, 1995] - initial work-in-progress release - -version 0.2 [April 1, 1995] - added reader into png.h - fixed small problems in stub file - -version 0.3 [April 8, 1995] - added pull reader - split up pngwrite.c to several files - added pnglib.txt - added example.c - cleaned up writer, adding a few new transformations - fixed some bugs in writer - interfaced with zlib 0.5 - added K&R support - added check for 64 KB blocks for 16 bit machines - -version 0.4 [April 26, 1995] - cleaned up code and commented code - simplified time handling into png_time - created png_color_16 and png_color_8 to handle color needs - cleaned up color type defines - fixed various bugs - made various names more consistent - interfaced with zlib 0.71 - cleaned up zTXt reader and writer (using zlib's Reset functions) - split transformations into pngrtran.c and pngwtran.c - -version 0.5 [April 30, 1995] - interfaced with zlib 0.8 - fixed many reading and writing bugs - saved using 3 spaces instead of tabs - -version 0.6 [May 1, 1995] - first beta release - added png_large_malloc() and png_large_free() - added png_size_t - cleaned up some compiler warnings - added png_start_read_image() - -version 0.7 [June 24, 1995] - cleaned up lots of bugs - finished dithering and other stuff - added test program - changed name from pnglib to libpng - -version 0.71 [June 26, 1995] - changed pngtest.png for zlib 0.93 - fixed error in libpng.txt and example.c - -version 0.8 [August 20, 1995] - cleaned up some bugs - added png_set_filler() - split up pngstub.c into pngmem.c, pngio.c, and pngerror.c - added #define's to remove unwanted code - moved png_info_init() to png.c - added old_size into png_realloc() - added functions to manually set filtering and compression info - changed compression parameters based on image type - optimized filter selection code - added version info - changed external functions passing floats to doubles (k&r problems?) - put all the configurable stuff in pngconf.h - enabled png_set_shift to work with paletted images on read - added png_read_update_info() - updates info structure with transformations - -Version 0.81 [August, 1995] - incorporated Tim Wegner's medium model code (thanks, Tim) - -Version 0.82 [September, 1995] - [unspecified changes] - -Version 0.85 [December, 1995] - added more medium model code (almost everything's a far) - added i/o, error, and memory callback functions - fixed some bugs (16-bit, 4-bit interlaced, etc.) - added first run progressive reader (barely tested) - -Version 0.86 [January, 1996] - fixed bugs - improved documentation - -Version 0.87 [January, 1996] - fixed medium model bugs - fixed other bugs introduced in 0.85 and 0.86 - added some minor documentation - -Version 0.88 [January, 1996] - fixed progressive bugs - replaced tabs with spaces - cleaned up documentation - added callbacks for read/write and warning/error functions - -Version 0.89 [June 5, 1996] - Added new initialization API to make libpng work better with shared libs - we now have png_create_read_struct(), png_create_write_struct(), - png_create_info_struct(), png_destroy_read_struct(), and - png_destroy_write_struct() instead of the separate calls to - malloc and png_read_init(), png_info_init(), and png_write_init() - Changed warning/error callback functions to fix bug - this means you - should use the new initialization API if you were using the old - png_set_message_fn() calls, and that the old API no longer exists - so that people are aware that they need to change their code - Changed filter selection API to allow selection of multiple filters - since it didn't work in previous versions of libpng anyways - Optimized filter selection code - Fixed png_set_background() to allow using an arbitrary RGB color for - paletted images - Fixed gamma and background correction for paletted images, so - png_correct_palette is not needed unless you are correcting an - external palette (you will need to #define PNG_CORRECT_PALETTE_SUPPORTED - in pngconf.h) - if nobody uses this, it may disappear in the future. - Fixed bug with Borland 64K memory allocation (Alexander Lehmann) - Fixed bug in interlace handling (Smarasderagd, I think) - Added more error checking for writing and image to reduce invalid files - Separated read and write functions so that they won't both be linked - into a binary when only reading or writing functionality is used - New pngtest image also has interlacing and zTXt - Updated documentation to reflect new API - -Version 0.89c [June 17, 1996] - Bug fixes. - -Version 0.90 [January, 1997] - Made CRC errors/warnings on critical and ancillary chunks configurable - libpng will use the zlib CRC routines by (compile-time) default - Changed DOS small/medium model memory support - needs zlib 1.04 (Tim Wegner) - Added external C++ wrapper statements to png.h (Gilles Dauphin) - Allow PNG file to be read when some or all of file signature has already - been read from the beginning of the stream. ****This affects the size - of info_struct and invalidates all programs that use a shared libpng**** - Fixed png_filler() declarations - Fixed? background color conversions - Fixed order of error function pointers to match documentation - Current chunk name is now available in png_struct to reduce the number - of nearly identical error messages (will simplify multi-lingual - support when available) - Try to get ready for unknown-chunk callback functions: - - previously read critical chunks are flagged, so the chunk handling - routines can determine if the chunk is in the right place - - all chunk handling routines have the same prototypes, so we will - be able to handle all chunks via a callback mechanism - Try to fix Linux "setjmp" buffer size problems - Removed png_large_malloc, png_large_free, and png_realloc functions. - -Version 0.95 [March, 1997] - Fixed bug in pngwutil.c allocating "up_row" twice and "avg_row" never - Fixed bug in PNG file signature compares when start != 0 - Changed parameter type of png_set_filler(...filler...) from png_byte - to png_uint_32 - Added test for MACOS to ensure that both math.h and fp.h are not #included - Added macros for libpng to be compiled as a Windows DLL (Andreas Kupries) - Added "packswap" transformation, which changes the endianness of - packed-pixel bytes (Kevin Bracey) - Added "strip_alpha" transformation, which removes the alpha channel of - input images without using it (not necessarily a good idea) - Added "swap_alpha" transformation, which puts the alpha channel in front - of the color bytes instead of after - Removed all implicit variable tests which assume NULL == 0 (I think) - Changed several variables to "png_size_t" to show 16/32-bit limitations - Added new pCAL chunk read/write support - Added experimental filter selection weighting (Greg Roelofs) - Removed old png_set_rgbx() and png_set_xrgb() functions that have been - obsolete for about 2 years now (use png_set_filler() instead) - Added macros to read 16- and 32-bit ints directly from buffer, to be - used only on those systems that support it (namely PowerPC and 680x0) - With some testing, this may become the default for MACOS/PPC systems. - Only calculate CRC on data if we are going to use it - Added macros for zTXt compression type PNG_zTXt_COMPRESSION_??? - Added macros for simple libpng debugging output selectable at compile time - Removed PNG_READ_END_MODE in progressive reader (Smarasderagd) - More description of info_struct in libpng.txt and png.h - More instructions in example.c - More chunk types tested in pngtest.c - Renamed pngrcb.c to pngset.c, and all png_read_ functions to be - png_set_. We now have corresponding png_get_ - functions in pngget.c to get information in info_ptr. This isolates - the application from the internal organization of png_info_struct - (good for shared library implementations). - -Version 0.96 [May, 1997] - Fixed serious bug with < 8bpp images introduced in 0.95 - Fixed 256-color transparency bug (Greg Roelofs) - Fixed up documentation (Greg Roelofs, Laszlo Nyul) - Fixed "error" in pngconf.h for Linux setjmp() behavior - Fixed DOS medium model support (Tim Wegner) - Fixed png_check_keyword() for case with error in static string text - Added read of CRC after IEND chunk for embedded PNGs (Laszlo Nyul) - Added typecasts to quiet compiler errors - Added more debugging info - -Version 0.97 [January, 1998] - Removed PNG_USE_OWN_CRC capability - Relocated png_set_crc_action from pngrutil.c to pngrtran.c - Fixed typecasts of "new_key", etc. (Andreas Dilger) - Added RFC 1152 [sic] date support - Fixed bug in gamma handling of 4-bit grayscale - Added 2-bit grayscale gamma handling (Glenn R-P) - Added more typecasts. 65536L becomes (png_uint_32)65536L, etc. (Glenn R-P) - Minor corrections in libpng.txt - Added simple sRGB support (Glenn R-P) - Easier conditional compiling, e.g., - define PNG_READ/WRITE_NOT_FULLY_SUPPORTED; - all configurable options can be selected from command line instead - of having to edit pngconf.h (Glenn R-P) - Fixed memory leak in pngwrite.c (free info_ptr->text) (Glenn R-P) - Added more conditions for png_do_background, to avoid changing - black pixels to background when a background is supplied and - no pixels are transparent - Repaired PNG_NO_STDIO behavior - Tested NODIV support and made it default behavior (Greg Roelofs) - Added "-m" option and PNGTEST_DEBUG_MEMORY to pngtest (John Bowler) - Regularized version numbering scheme and bumped shared-library major - version number to 2 to avoid problems with libpng 0.89 apps - (Greg Roelofs) - -Version 0.98 [January, 1998] - Cleaned up some typos in libpng.txt and in code documentation - Fixed memory leaks in pCAL chunk processing (Glenn R-P and John Bowler) - Cosmetic change "display_gamma" to "screen_gamma" in pngrtran.c - Changed recommendation about file_gamma for PC images to .51 from .45, - in example.c and libpng.txt, added comments to distinguish between - screen_gamma, viewing_gamma, and display_gamma. - Changed all references to RFC1152 to read RFC1123 and changed the - PNG_TIME_RFC1152_SUPPORTED macro to PNG_TIME_RFC1123_SUPPORTED - Added png_invert_alpha capability (Glenn R-P -- suggestion by Jon Vincent) - Changed srgb_intent from png_byte to int to avoid compiler bugs - -Version 0.99 [January 30, 1998] - Free info_ptr->text instead of end_info_ptr->text in pngread.c (John Bowler) - Fixed a longstanding "packswap" bug in pngtrans.c - Fixed some inconsistencies in pngconf.h that prevented compiling with - PNG_READ_GAMMA_SUPPORTED and PNG_READ_hIST_SUPPORTED undefined - Fixed some typos and made other minor rearrangement of libpng.txt (Andreas) - Changed recommendation about file_gamma for PC images to .50 from .51 in - example.c and libpng.txt, and changed file_gamma for sRGB images to .45 - Added a number of functions to access information from the png structure - png_get_image_height(), etc. (Glenn R-P, suggestion by Brad Pettit) - Added TARGET_MACOS similar to zlib-1.0.8 - Define PNG_ALWAYS_EXTERN when __MWERKS__ && WIN32 are defined - Added type casting to all png_malloc() function calls - -Version 0.99a [January 31, 1998] - Added type casts and parentheses to all returns that return a value.(Tim W.) - -Version 0.99b [February 4, 1998] - Added type cast png_uint_32 on malloc function calls where needed. - Changed type of num_hist from png_uint_32 to int (same as num_palette). - Added checks for rowbytes overflow, in case png_size_t is less than 32 bits. - Renamed makefile.elf to makefile.lnx. - -Version 0.99c [February 7, 1998] - More type casting. Removed erroneous overflow test in pngmem.c. - Added png_buffered_memcpy() and png_buffered_memset(), apply them to rowbytes. - Added UNIX manual pages libpng.3 (incorporating libpng.txt) and png.5. - -Version 0.99d [February 11, 1998] - Renamed "far_to_near()" "png_far_to_near()" - Revised libpng.3 - Version 99c "buffered" operations didn't work as intended. Replaced them - with png_memcpy_check() and png_memset_check(). - Added many "if (png_ptr == NULL) return" to quell compiler warnings about - unused png_ptr, mostly in pngget.c and pngset.c. - Check for overlength tRNS chunk present when indexed-color PLTE is read. - Cleaned up spelling errors in libpng.3/libpng.txt - Corrected a problem with png_get_tRNS() which returned undefined trans array - -Version 0.99e [February 28, 1998] - Corrected png_get_tRNS() again. - Add parentheses for easier reading of pngget.c, fixed "||" should be "&&". - Touched up example.c to make more of it compileable, although the entire - file still can't be compiled (Willem van Schaik) - Fixed a bug in png_do_shift() (Bryan Tsai) - Added a space in png.h prototype for png_write_chunk_start() - Replaced pngtest.png with one created with zlib 1.1.1 - Changed pngtest to report PASS even when file size is different (Jean-loup G.) - Corrected some logic errors in png_do_invert_alpha() (Chris Patterson) - -Version 0.99f [March 5, 1998] - Corrected a bug in pngpread() introduced in version 99c (Kevin Bracey) - Moved makefiles into a "scripts" directory, and added INSTALL instruction file - Added makefile.os2 and pngos2.def (A. Zabolotny) and makefile.s2x (W. Sebok) - Added pointers to "note on libpng versions" in makefile.lnx and README - Added row callback feature when reading and writing nonprogressive rows - and added a test of this feature in pngtest.c - Added user transform callbacks, with test of the feature in pngtest.c - -Version 0.99g [March 6, 1998, morning] - Minor changes to pngtest.c to suppress compiler warnings. - Removed "beta" language from documentation. - -Version 0.99h [March 6, 1998, evening] - Minor changes to previous minor changes to pngtest.c - Changed PNG_READ_NOT_FULLY_SUPPORTED to PNG_READ_TRANSFORMS_NOT_SUPPORTED - and added PNG_PROGRESSIVE_READ_NOT_SUPPORTED macro - Added user transform capability - -Version 1.00 [March 7, 1998] - Changed several typedefs in pngrutil.c - Added makefile.wat (Pawel Mrochen), updated makefile.tc3 (Willem van Schaik) - Replaced "while(1)" with "for(;;)" - Added PNGARG() to prototypes in pngtest.c and removed some prototypes - Updated some of the makefiles (Tom Lane) - Changed some typedefs (s_start, etc.) in pngrutil.c - Fixed dimensions of "short_months" array in pngwrite.c - Replaced ansi2knr.c with the one from jpeg-v6 - -Version 1.0.0 [March 8, 1998] - Changed name from 1.00 to 1.0.0 (Adam Costello) - Added smakefile.ppc (with SCOPTIONS.ppc) for Amiga PPC (Andreas Kleinert) - -Version 1.0.0a [March 9, 1998] - Fixed three bugs in pngrtran.c to make gamma+background handling consistent - (Greg Roelofs) - Changed format of the PNG_LIBPNG_VER integer to xyyzz instead of xyz - for major, minor, and bugfix releases. This is 10001. (Adam Costello, - Tom Lane) - Make months range from 1-12 in png_convert_to_rfc1123 - -Version 1.0.0b [March 13, 1998] - Quieted compiler complaints about two empty "for" loops in pngrutil.c - Minor changes to makefile.s2x - Removed #ifdef/#endif around a png_free() in pngread.c - -Version 1.0.1 [March 14, 1998] - Changed makefile.s2x to reduce security risk of using a relative pathname - Fixed some typos in the documentation (Greg). - Fixed a problem with value of "channels" returned by png_read_update_info() - -Version 1.0.1a [April 21, 1998] - Optimized Paeth calculations by replacing abs() function calls with intrinsics - plus other loop optimizations. Improves avg decoding speed by about 20%. - Commented out i386istic "align" compiler flags in makefile.lnx. - Reduced the default warning level in some makefiles, to make them consistent. - Removed references to IJG and JPEG in the ansi2knr.c copyright statement. - Fixed a bug in png_do_strip_filler with XXRRGGBB => RRGGBB transformation. - Added grayscale and 16-bit capability to png_do_read_filler(). - Fixed a bug in pngset.c, introduced in version 0.99c, that sets rowbytes - too large when writing an image with bit_depth < 8 (Bob Dellaca). - Corrected some bugs in the experimental weighted filtering heuristics. - Moved a misplaced pngrutil code block that truncates tRNS if it has more - than num_palette entries -- test was done before num_palette was defined. - Fixed a png_convert_to_rfc1123() bug that converts day 31 to 0 (Steve Eddins). - Changed compiler flags in makefile.wat for better optimization - (Pawel Mrochen). - -Version 1.0.1b [May 2, 1998] - Relocated png_do_gray_to_rgb() within png_do_read_transformations() (Greg). - Relocated the png_composite macros from pngrtran.c to png.h (Greg). - Added makefile.sco (contributed by Mike Hopkirk). - Fixed two bugs (missing definitions of "istop") introduced in libpng-1.0.1a. - Fixed a bug in pngrtran.c that would set channels=5 under some circumstances. - More work on the Paeth-filtering, achieving imperceptible speedup - (A Kleinert). - More work on loop optimization which may help when compiled with C++ - compilers. - Added warnings when people try to use transforms they've defined out. - Collapsed 4 "i" and "c" loops into single "i" loops in pngrtran and pngwtran. - Revised paragraph about png_set_expand() in libpng.txt and libpng.3 (Greg) - -Version 1.0.1c [May 11, 1998] - Fixed a bug in pngrtran.c (introduced in libpng-1.0.1a) where the masks for - filler bytes should have been 0xff instead of 0xf. - Added max_pixel_depth=32 in pngrutil.c when using FILLER with palette images. - Moved PNG_WRITE_WEIGHTED_FILTER_SUPPORTED and PNG_WRITE_FLUSH_SUPPORTED - out of the PNG_WRITE_TRANSFORMS_NOT_SUPPORTED block of pngconf.h - Added "PNG_NO_WRITE_TRANSFORMS" etc., as alternatives for *_NOT_SUPPORTED, - for consistency, in pngconf.h - Added individual "ifndef PNG_NO_[CAPABILITY]" in pngconf.h to make it easier - to remove unwanted capabilities via the compile line - Made some corrections to grammar (which, it's) in documentation (Greg). - Corrected example.c, use of row_pointers in png_write_image(). - -Version 1.0.1d [May 24, 1998] - Corrected several statements that used side effects illegally in pngrutil.c - and pngtrans.c, that were introduced in version 1.0.1b - Revised png_read_rows() to avoid repeated if-testing for NULL (A Kleinert) - More corrections to example.c, use of row_pointers in png_write_image() - and png_read_rows(). - Added pngdll.mak and pngdef.pas to scripts directory, contributed by - Bob Dellaca, to make a png32bd.dll with Borland C++ 4.5 - Fixed error in example.c with png_set_text: num_text is 3, not 2 (Guido V.) - Changed several loops from count-down to count-up, for consistency. - -Version 1.0.1e [June 6, 1998] - Revised libpng.txt and libpng.3 description of png_set_read|write_fn(), and - added warnings when people try to set png_read_fn and png_write_fn in - the same structure. - Added a test such that png_do_gamma will be done when num_trans==0 - for truecolor images that have defined a background. This corrects an - error that was introduced in libpng-0.90 that can cause gamma processing - to be skipped. - Added tests in png.h to include "trans" and "trans_values" in structures - when PNG_READ_BACKGROUND_SUPPORTED or PNG_READ_EXPAND_SUPPORTED is defined. - Add png_free(png_ptr->time_buffer) in png_destroy_read_struct() - Moved png_convert_to_rfc_1123() from pngwrite.c to png.c - Added capability for user-provided malloc_fn() and free_fn() functions, - and revised pngtest.c to demonstrate their use, replacing the - PNGTEST_DEBUG_MEM feature. - Added makefile.w32, for Microsoft C++ 4.0 and later (Tim Wegner). - -Version 1.0.2 [June 14, 1998] - Fixed two bugs in makefile.bor . - -Version 1.0.2a [December 30, 1998] - Replaced and extended code that was removed from png_set_filler() in 1.0.1a. - Fixed a bug in png_do_filler() that made it fail to write filler bytes in - the left-most pixel of each row (Kevin Bracey). - Changed "static pngcharp tIME_string" to "static char tIME_string[30]" - in pngtest.c (Duncan Simpson). - Fixed a bug in pngtest.c that caused pngtest to try to write a tIME chunk - even when no tIME chunk was present in the source file. - Fixed a problem in pngrutil.c: gray_to_rgb didn't always work with 16-bit. - Fixed a problem in png_read_push_finish_row(), which would not skip some - passes that it should skip, for images that are less than 3 pixels high. - Interchanged the order of calls to png_do_swap() and png_do_shift() - in pngwtran.c (John Cromer). - Added #ifdef PNG_DEBUG/#endif surrounding use of PNG_DEBUG in png.h . - Changed "bad adaptive filter type" from error to warning in pngrutil.c . - Fixed a documentation error about default filtering with 8-bit indexed-color. - Separated the PNG_NO_STDIO macro into PNG_NO_STDIO and PNG_NO_CONSOLE_IO - (L. Peter Deutsch). - Added png_set_rgb_to_gray() and png_get_rgb_to_gray_status() functions. - Added png_get_copyright() and png_get_header_version() functions. - Revised comments on png_set_progressive_read_fn() in libpng.txt and example.c - Added information about debugging in libpng.txt and libpng.3 . - Changed "ln -sf" to "ln -s -f" in makefile.s2x, makefile.lnx, and - makefile.sco. - Removed lines after Dynamic Dependencies" in makefile.aco . - Revised makefile.dec to make a shared library (Jeremie Petit). - Removed trailing blanks from all files. - -Version 1.0.2a [January 6, 1999] - Removed misplaced #endif and #ifdef PNG_NO_EXTERN near the end of png.h - Added "if" tests to silence complaints about unused png_ptr in png.h and png.c - Changed "check_if_png" function in example.c to return true (nonzero) if PNG. - Changed libpng.txt to demonstrate png_sig_cmp() instead of png_check_sig() - which is obsolete. - -Version 1.0.3 [January 14, 1999] - Added makefile.hux, for Hewlett Packard HPUX 10.20 and 11.00 (Jim Rice) - Added a statement of Y2K compliance in png.h, libpng.3, and Y2KINFO. - -Version 1.0.3a [August 12, 1999] - Added check for PNG_READ_INTERLACE_SUPPORTED in pngread.c; issue a warning - if an attempt is made to read an interlaced image when it's not supported. - Added check if png_ptr->trans is defined before freeing it in pngread.c - Modified the Y2K statement to include versions back to version 0.71 - Fixed a bug in the check for valid IHDR bit_depth/color_types in pngrutil.c - Modified makefile.wat (added -zp8 flag, ".symbolic", changed some comments) - Replaced leading blanks with tab characters in makefile.hux - Changed "dworkin.wustl.edu" to "ccrc.wustl.edu" in various documents. - Changed (float)red and (float)green to (double)red, (double)green - in png_set_rgb_to_gray() to avoid "promotion" problems in AIX. - Fixed a bug in pngconf.h that omitted when PNG_DEBUG==0 (K Bracey). - Reformatted libpng.3 and libpngpf.3 with proper fonts (script by J. vanZandt). - Updated documentation to refer to the PNG-1.2 specification. - Removed ansi2knr.c and left pointers to the latest source for ansi2knr.c - in makefile.knr, INSTALL, and README (L. Peter Deutsch) - Fixed bugs in calculation of the length of rowbytes when adding alpha - channels to 16-bit images, in pngrtran.c (Chris Nokleberg) - Added function png_set_user_transform_info() to store user_transform_ptr, - user_depth, and user_channels into the png_struct, and a function - png_get_user_transform_ptr() to retrieve the pointer (Chris Nokleberg) - Added function png_set_empty_plte_permitted() to make libpng useable - in MNG applications. - Corrected the typedef for png_free_ptr in png.h (Jesse Jones). - Correct gamma with srgb is 45455 instead of 45000 in pngrutil.c, to be - consistent with PNG-1.2, and allow variance of 500 before complaining. - Added assembler code contributed by Intel in file pngvcrd.c and modified - makefile.w32 to use it (Nirav Chhatrapati, INTEL Corporation, - Gilles Vollant) - Changed "ln -s -f" to "ln -f -s" in the makefiles to make Solaris happy. - Added some aliases for png_set_expand() in pngrtran.c, namely - png_set_expand_PLTE(), png_set_expand_depth(), and png_set_expand_tRNS() - (Greg Roelofs, in "PNG: The Definitive Guide"). - Added makefile.beo for BEOS on X86, contributed by Sander Stok. - -Version 1.0.3b [August 26, 1999] - Replaced 2147483647L several places with PNG_MAX_UINT macro, defined in png.h - Changed leading blanks to tabs in all makefiles. - Define PNG_USE_PNGVCRD in makefile.w32, to get MMX assembler code. - Made alternate versions of png_set_expand() in pngrtran.c, namely - png_set_gray_1_2_4_to_8, png_set_palette_to_rgb, and png_set_tRNS_to_alpha - (Greg Roelofs, in "PNG: The Definitive Guide"). Deleted the 1.0.3a aliases. - Relocated start of 'extern "C"' block in png.h so it doesn't include pngconf.h - Revised calculation of num_blocks in pngmem.c to avoid a potentially - negative shift distance, whose results are undefined in the C language. - Added a check in pngset.c to prevent writing multiple tIME chunks. - Added a check in pngwrite.c to detect invalid small window_bits sizes. - -Version 1.0.3d [September 4, 1999] - Fixed type casting of igamma in pngrutil.c - Added new png_expand functions to scripts/pngdef.pas and pngos2.def - Added a demo read_user_transform_fn that examines the row filters in pngtest.c - -Version 1.0.4 [September 24, 1999, not distributed publicly] - Define PNG_ALWAYS_EXTERN in pngconf.h if __STDC__ is defined - Delete #define PNG_INTERNAL and include "png.h" from pngasmrd.h - Made several minor corrections to pngtest.c - Renamed the makefiles with longer but more user friendly extensions. - Copied the PNG copyright and license to a separate LICENSE file. - Revised documentation, png.h, and example.c to remove reference to - "viewing_gamma" which no longer appears in the PNG specification. - Revised pngvcrd.c to use MMX code for interlacing only on the final pass. - Updated pngvcrd.c to use the faster C filter algorithms from libpng-1.0.1a - Split makefile.win32vc into two versions, makefile.vcawin32 (uses MMX - assembler code) and makefile.vcwin32 (doesn't). - Added a CPU timing report to pngtest.c (enabled by defining PNGTEST_TIMING) - Added a copy of pngnow.png to the distribution. - -Version 1.0.4a [September 25, 1999] - Increase max_pixel_depth in pngrutil.c if a user transform needs it. - Changed several division operations to right-shifts in pngvcrd.c - -Version 1.0.4b [September 30, 1999] - Added parentheses in line 3732 of pngvcrd.c - Added a comment in makefile.linux warning about buggy -O3 in pgcc 2.95.1 - -Version 1.0.4c [October 1, 1999] - Added a "png_check_version" function in png.c and pngtest.c that will generate - a helpful compiler error if an old png.h is found in the search path. - Changed type of png_user_transform_depth|channels from int to png_byte. - Added "Libpng is OSI Certified Open Source Software" statement to png.h - -Version 1.0.4d [October 6, 1999] - Changed 0.45 to 0.45455 in png_set_sRGB() - Removed unused PLTE entries from pngnow.png - Re-enabled some parts of pngvcrd.c (png_combine_row) that work properly. - -Version 1.0.4e [October 10, 1999] - Fixed sign error in pngvcrd.c (Greg Roelofs) - Replaced some instances of memcpy with simple assignments in pngvcrd (GR-P) - -Version 1.0.4f [October 15, 1999] - Surrounded example.c code with #if 0 .. #endif to prevent people from - inadvertently trying to compile it. - Changed png_get_header_version() from a function to a macro in png.h - Added type casting mostly in pngrtran.c and pngwtran.c - Removed some pointless "ptr = NULL" in pngmem.c - Added a "contrib" directory containing the source code from Greg's book. - -Version 1.0.5 [October 15, 1999] - Minor editing of the INSTALL and README files. - -Version 1.0.5a [October 23, 1999] - Added contrib/pngsuite and contrib/pngminus (Willem van Schaik) - Fixed a typo in the png_set_sRGB() function call in example.c (Jan Nijtmans) - Further optimization and bugfix of pngvcrd.c - Revised pngset.c so that it does not allocate or free memory in the user's - text_ptr structure. Instead, it makes its own copy. - Created separate write_end_info_struct in pngtest.c for a more severe test. - Added code in pngwrite.c to free info_ptr->text[i].key to stop a memory leak. - -Version 1.0.5b [November 23, 1999] - Moved PNG_FLAG_HAVE_CHUNK_HEADER, PNG_FLAG_BACKGROUND_IS_GRAY and - PNG_FLAG_WROTE_tIME from flags to mode. - Added png_write_info_before_PLTE() function. - Fixed some typecasting in contrib/gregbook/*.c - Updated scripts/makevms.com and added makevms.com to contrib/gregbook - and contrib/pngminus (Martin Zinser) - -Version 1.0.5c [November 26, 1999] - Moved png_get_header_version from png.h to png.c, to accommodate ansi2knr. - Removed all global arrays (according to PNG_NO_GLOBAL_ARRAYS macro), to - accommodate making DLL's: Moved usr_png_ver from global variable to function - png_get_header_ver() in png.c. Moved png_sig to png_sig_bytes in png.c and - eliminated use of png_sig in pngwutil.c. Moved the various png_CHNK arrays - into pngtypes.h. Eliminated use of global png_pass arrays. Declared the - png_CHNK and png_pass arrays to be "const". Made the global arrays - available to applications (although none are used in libpng itself) when - PNG_NO_GLOBAL_ARRAYS is not defined or when PNG_GLOBAL_ARRAYS is defined. - Removed some extraneous "-I" from contrib/pngminus/makefile.std - Changed the PNG_sRGB_INTENT macros in png.h to be consistent with PNG-1.2. - Change PNG_SRGB_INTENT to PNG_sRGB_INTENT in libpng.txt and libpng.3 - -Version 1.0.5d [November 29, 1999] - Add type cast (png_const_charp) two places in png.c - Eliminated pngtypes.h; use macros instead to declare PNG_CHNK arrays. - Renamed "PNG_GLOBAL_ARRAYS" to "PNG_USE_GLOBAL_ARRAYS" and made available - to applications a macro "PNG_USE_LOCAL_ARRAYS". - comment out (with #ifdef) all the new declarations when - PNG_USE_GLOBAL_ARRAYS is defined. - Added PNG_EXPORT_VAR macro to accommodate making DLL's. - -Version 1.0.5e [November 30, 1999] - Added iCCP, iTXt, and sPLT support; added "lang" member to the png_text - structure; refactored the inflate/deflate support to make adding new chunks - with trailing compressed parts easier in the future, and added new functions - png_free_iCCP, png_free_pCAL, png_free_sPLT, png_free_text, png_get_iCCP, - png_get_spalettes, png_set_iCCP, png_set_spalettes (Eric S. Raymond). - NOTE: Applications that write text chunks MUST define png_text->lang - before calling png_set_text(). It must be set to NULL if you want to - write tEXt or zTXt chunks. If you want your application to be able to - run with older versions of libpng, use - - #ifdef PNG_iTXt_SUPPORTED - png_text[i].lang = NULL; - #endif - - Changed png_get_oFFs() and png_set_oFFs() to use signed rather than unsigned - offsets (Eric S. Raymond). - Combined PNG_READ_cHNK_SUPPORTED and PNG_WRITE_cHNK_SUPPORTED macros into - PNG_cHNK_SUPPORTED and combined the three types of PNG_text_SUPPORTED - macros, leaving the separate macros also available. - Removed comments on #endifs at the end of many short, non-nested #if-blocks. - -Version 1.0.5f [December 6, 1999] - Changed makefile.solaris to issue a warning about potential problems when - the ucb "ld" is in the path ahead of the ccs "ld". - Removed "- [date]" from the "synopsis" line in libpng.3 and libpngpf.3. - Added sCAL chunk support (Eric S. Raymond). - -Version 1.0.5g [December 7, 1999] - Fixed "png_free_spallettes" typo in png.h - Added code to handle new chunks in pngpread.c - Moved PNG_CHNK string macro definitions outside of PNG_NO_EXTERN block - Added "translated_key" to png_text structure and png_write_iTXt(). - Added code in pngwrite.c to work around a newly discovered zlib bug. - -Version 1.0.5h [December 10, 1999] - NOTE: regarding the note for version 1.0.5e, the following must also - be included in your code: - png_text[i].translated_key = NULL; - Unknown chunk handling is now supported. - Option to eliminate all floating point support was added. Some new - fixed-point functions such as png_set_gAMA_fixed() were added. - Expanded tabs and removed trailing blanks in source files. - -Version 1.0.5i [December 13, 1999] - Added some type casts to silence compiler warnings. - Renamed "png_free_spalette" to "png_free_spalettes" for consistency. - Removed leading blanks from a #define in pngvcrd.c - Added some parameters to the new png_set_keep_unknown_chunks() function. - Added a test for up->location != 0 in the first instance of writing - unknown chunks in pngwrite.c - Changed "num" to "i" in png_free_spalettes() and png_free_unknowns() to - prevent recursion. - Added png_free_hIST() function. - Various patches to fix bugs in the sCAL and integer cHRM processing, - and to add some convenience macros for use with sCAL. - -Version 1.0.5j [December 21, 1999] - Changed "unit" parameter of png_write_sCAL from png_byte to int, to work - around buggy compilers. - Added new type "png_fixed_point" for integers that hold float*100000 values - Restored backward compatibility of tEXt/zTXt chunk processing: - Restored the first four members of png_text to the same order as v.1.0.5d. - Added members "lang_key" and "itxt_length" to png_text struct. Set - text_length=0 when "text" contains iTXt data. Use the "compression" - member to distinguish among tEXt/zTXt/iTXt types. Added - PNG_ITXT_COMPRESSION_NONE (1) and PNG_ITXT_COMPRESSION_zTXt(2) macros. - The "Note" above, about backward incompatibility of libpng-1.0.5e, no - longer applies. - Fixed png_read|write_iTXt() to read|write parameters in the right order, - and to write the iTXt chunk after IDAT if it appears in the end_ptr. - Added pnggccrd.c, version of pngvcrd.c Intel assembler for gcc (Greg Roelofs) - Reversed the order of trying to write floating-point and fixed-point gAMA. - -Version 1.0.5k [December 27, 1999] - Added many parentheses, e.g., "if (a && b & c)" becomes "if (a && (b & c))" - Added png_handle_as_unknown() function (Glenn) - Added png_free_chunk_list() function and chunk_list and num_chunk_list members - of png_ptr. - Eliminated erroneous warnings about multiple sPLT chunks and sPLT-after-PLTE. - Fixed a libpng-1.0.5h bug in pngrutil.c that was issuing erroneous warnings - about ignoring incorrect gAMA with sRGB (gAMA was in fact not ignored) - Added png_free_tRNS(); png_set_tRNS() now malloc's its own trans array (ESR). - Define png_get_int_32 when oFFs chunk is supported as well as when pCAL is. - Changed type of proflen from png_int_32 to png_uint_32 in png_get_iCCP(). - -Version 1.0.5l [January 1, 2000] - Added functions png_set_read_user_chunk_fn() and png_get_user_chunk_ptr() - for setting a callback function to handle unknown chunks and for - retrieving the associated user pointer (Glenn). - -Version 1.0.5m [January 7, 2000] - Added high-level functions png_read_png(), png_write_png(), png_free_pixels(). - -Version 1.0.5n [January 9, 2000] - Added png_free_PLTE() function, and modified png_set_PLTE() to malloc its - own memory for info_ptr->palette. This makes it safe for the calling - application to free its copy of the palette any time after it calls - png_set_PLTE(). - -Version 1.0.5o [January 20, 2000] - Cosmetic changes only (removed some trailing blanks and TABs) - -Version 1.0.5p [January 31, 2000] - Renamed pngdll.mak to makefile.bd32 - Cosmetic changes in pngtest.c - -Version 1.0.5q [February 5, 2000] - Relocated the makefile.solaris warning about PATH problems. - Fixed pngvcrd.c bug by pushing/popping registers in mmxsupport (Bruce Oberg) - Revised makefile.gcmmx - Added PNG_SETJMP_SUPPORTED, PNG_SETJMP_NOT_SUPPORTED, and PNG_ABORT() macros - -Version 1.0.5r [February 7, 2000] - Removed superfluous prototype for png_get_itxt from png.h - Fixed a bug in pngrtran.c that improperly expanded the background color. - Return *num_text=0 from png_get_text() when appropriate, and fix documentation - of png_get_text() in libpng.txt/libpng.3. - -Version 1.0.5s [February 18, 2000] - Added "png_jmp_env()" macro to pngconf.h, to help people migrate to the - new error handler that's planned for the next libpng release, and changed - example.c, pngtest.c, and contrib programs to use this macro. - Revised some of the DLL-export macros in pngconf.h (Greg Roelofs) - Fixed a bug in png_read_png() that caused it to fail to expand some images - that it should have expanded. - Fixed some mistakes in the unused and undocumented INCH_CONVERSIONS functions - in pngget.c - Changed the allocation of palette, history, and trans arrays back to - the version 1.0.5 method (linking instead of copying) which restores - backward compatibility with version 1.0.5. Added some remarks about - that in example.c. Added "free_me" member to info_ptr and png_ptr - and added png_free_data() function. - Updated makefile.linux and makefile.gccmmx to make directories conditionally. - Made cosmetic changes to pngasmrd.h - Added png_set_rows() and png_get_rows(), for use with png_read|write_png(). - Modified png_read_png() to allocate info_ptr->row_pointers only if it - hasn't already been allocated. - -Version 1.0.5t [March 4, 2000] - Changed png_jmp_env() migration aiding macro to png_jmpbuf(). - Fixed "interlace" typo (should be "interlaced") in contrib/gregbook/read2-x.c - Fixed bug with use of PNG_BEFORE_IHDR bit in png_ptr->mode, introduced when - PNG_FLAG_HAVE_CHUNK_HEADER was moved into png_ptr->mode in version 1.0.5b - Files in contrib/gregbook were revised to use png_jmpbuf() and to select - a 24-bit visual if one is available, and to allow abbreviated options. - Files in contrib/pngminus were revised to use the png_jmpbuf() macro. - Removed spaces in makefile.linux and makefile.gcmmx, introduced in 1.0.5s - -Version 1.0.5u [March 5, 2000] - Simplified the code that detects old png.h in png.c and pngtest.c - Renamed png_spalette (_p, _pp) to png_sPLT_t (_tp, _tpp) - Increased precision of rgb_to_gray calculations from 8 to 15 bits and - added png_set_rgb_to_gray_fixed() function. - Added makefile.bc32 (32-bit Borland C++, C mode) - -Version 1.0.5v [March 11, 2000] - Added some parentheses to the png_jmpbuf macro definition. - Updated references to the zlib home page, which has moved to freesoftware.com. - Corrected bugs in documentation regarding png_read_row() and png_write_row(). - Updated documentation of png_rgb_to_gray calculations in libpng.3/libpng.txt. - Renamed makefile.borland,turboc3 back to makefile.bor,tc3 as in version 1.0.3, - revised borland makefiles; added makefile.ibmvac3 and makefile.gcc (Cosmin) - -Version 1.0.6 [March 20, 2000] - Minor revisions of makefile.bor, libpng.txt, and gregbook/rpng2-win.c - Added makefile.sggcc (SGI IRIX with gcc) - -Version 1.0.6d [April 7, 2000] - Changed sprintf() to strcpy() in png_write_sCAL_s() to work without STDIO - Added data_length parameter to png_decompress_chunk() function - Revised documentation to remove reference to abandoned png_free_chnk functions - Fixed an error in png_rgb_to_gray_fixed() - Revised example.c, usage of png_destroy_write_struct(). - Renamed makefile.ibmvac3 to makefile.ibmc, added libpng.icc IBM project file - Added a check for info_ptr->free_me&PNG_FREE_TEXT when freeing text in png.c - Simplify png_sig_bytes() function to remove use of non-ISO-C strdup(). - -Version 1.0.6e [April 9, 2000] - Added png_data_freer() function. - In the code that checks for over-length tRNS chunks, added check of - info_ptr->num_trans as well as png_ptr->num_trans (Matthias Benckmann) - Minor revisions of libpng.txt/libpng.3. - Check for existing data and free it if the free_me flag is set, in png_set_*() - and png_handle_*(). - Only define PNG_WEIGHTED_FILTERS_SUPPORTED when PNG_FLOATING_POINT_SUPPORTED - is defined. - Changed several instances of PNG_NO_CONSOLE_ID to PNG_NO_STDIO in pngrutil.c - and mentioned the purposes of the two macros in libpng.txt/libpng.3. - -Version 1.0.6f [April 14, 2000] - Revised png_set_iCCP() and png_set_rows() to avoid prematurely freeing data. - Add checks in png_set_text() for NULL members of the input text structure. - Revised libpng.txt/libpng.3. - Removed superfluous prototype for png_set_iTXt from png.h - Removed "else" from pngread.c, after png_error(), and changed "0" to "length". - Changed several png_errors about malformed ancillary chunks to png_warnings. - -Version 1.0.6g [April 24, 2000] - Added png_pass-* arrays to pnggccrd.c when PNG_USE_LOCAL_ARRAYS is defined. - Relocated paragraph about png_set_background() in libpng.3/libpng.txt - and other revisions (Matthias Benckmann) - Relocated info_ptr->free_me, png_ptr->free_me, and other info_ptr and - png_ptr members to restore binary compatibility with libpng-1.0.5 - (breaks compatibility with libpng-1.0.6). - -Version 1.0.6h [April 24, 2000] - Changed shared library so-number pattern from 2.x.y.z to xy.z (this builds - libpng.so.10 & libpng.so.10.6h instead of libpng.so.2 & libpng.so.2.1.0.6h) - This is a temporary change for test purposes. - -Version 1.0.6i [May 2, 2000] - Rearranged some members at the end of png_info and png_struct, to put - unknown_chunks_num and free_me within the original size of the png_structs - and free_me, png_read_user_fn, and png_free_fn within the original png_info, - because some old applications allocate the structs directly instead of - using png_create_*(). - Added documentation of user memory functions in libpng.txt/libpng.3 - Modified png_read_png so that it will use user_allocated row_pointers - if present, unless free_me directs that it be freed, and added description - of the use of png_set_rows() and png_get_rows() in libpng.txt/libpng.3. - Added PNG_LEGACY_SUPPORTED macro, and #ifdef out all new (since version - 1.00) members of png_struct and png_info, to regain binary compatibility - when you define this macro. Capabilities lost in this event - are user transforms (new in version 1.0.0),the user transform pointer - (new in version 1.0.2), rgb_to_gray (new in 1.0.5), iCCP, sCAL, sPLT, - the high-level interface, and unknown chunks support (all new in 1.0.6). - This was necessary because of old applications that allocate the structs - directly as authors were instructed to do in libpng-0.88 and earlier, - instead of using png_create_*(). - Added modes PNG_CREATED_READ_STRUCT and PNG_CREATED_WRITE_STRUCT which - can be used to detect codes that directly allocate the structs, and - code to check these modes in png_read_init() and png_write_init() and - generate a libpng error if the modes aren't set and PNG_LEGACY_SUPPORTED - was not defined. - Added makefile.intel and updated makefile.watcom (Pawel Mrochen) - -Version 1.0.6j [May 3, 2000] - Overloaded png_read_init() and png_write_init() with macros that convert - calls to png_read_init_2() or png_write_init_2() that check the version - and structure sizes. - -Version 1.0.7beta11 [May 7, 2000] - Removed the new PNG_CREATED_READ_STRUCT and PNG_CREATED_WRITE_STRUCT modes - which are no longer used. - Eliminated the three new members of png_text when PNG_LEGACY_SUPPORTED is - defined or when neither PNG_READ_iTXt_SUPPORTED nor PNG_WRITE_iTXt_SUPPORTED - is defined. - Made PNG_NO_READ|WRITE_iTXt the default setting, to avoid memory - overrun when old applications fill the info_ptr->text structure directly. - Added PNGAPI macro, and added it to the definitions of all exported functions. - Relocated version macro definitions ahead of the includes of zlib.h and - pngconf.h in png.h. - -Version 1.0.7beta12 [May 12, 2000] - Revised pngset.c to avoid a problem with expanding the png_debug macro. - Deleted some extraneous defines from pngconf.h - Made PNG_NO_CONSOLE_IO the default condition when PNG_BUILD_DLL is defined. - Use MSC _RPTn debugging instead of fprintf if _MSC_VER is defined. - Added png_access_version_number() function. - Check for mask&PNG_FREE_CHNK (for TEXT, SCAL, PCAL) in png_free_data(). - Expanded libpng.3/libpng.txt information about png_data_freer(). - -Version 1.0.7beta14 [May 17, 2000] (beta13 was not published) - Changed pnggccrd.c and pngvcrd.c to handle bad adaptive filter types as - warnings instead of errors, as pngrutil.c does. - Set the PNG_INFO_IDAT valid flag in png_set_rows() so png_write_png() - will actually write IDATs. - Made the default PNG_USE_LOCAL_ARRAYS depend on PNG_DLL instead of WIN32. - Make png_free_data() ignore its final parameter except when freeing data - that can have multiple instances (text, sPLT, unknowns). - Fixed a new bug in png_set_rows(). - Removed info_ptr->valid tests from png_free_data(), as in version 1.0.5. - Added png_set_invalid() function. - Fixed incorrect illustrations of png_destroy_write_struct() in example.c. - -Version 1.0.7beta15 [May 30, 2000] - Revised the deliberately erroneous Linux setjmp code in pngconf.h to produce - fewer error messages. - Rearranged checks for Z_OK to check the most likely path first in pngpread.c - and pngwutil.c. - Added checks in pngtest.c for png_create_*() returning NULL, and mentioned - in libpng.txt/libpng.3 the need for applications to check this. - Changed names of png_default_*() functions in pngtest to pngtest_*(). - Changed return type of png_get_x|y_offset_*() from png_uint_32 to png_int_32. - Fixed some bugs in the unused PNG_INCH_CONVERSIONS functions in pngget.c - Set each pointer to NULL after freeing it in png_free_data(). - Worked around a problem in pngconf.h; AIX's strings.h defines an "index" - macro that conflicts with libpng's png_color_16.index. (Dimitri - Papadapoulos) - Added "msvc" directory with MSVC++ project files (Simon-Pierre Cadieux). - -Version 1.0.7beta16 [June 4, 2000] - Revised the workaround of AIX string.h "index" bug. - Added a check for overlength PLTE chunk in pngrutil.c. - Added PNG_NO_POINTER_INDEXING macro to use array-indexing instead of pointer - indexing in pngrutil.c and pngwutil.c to accommodate a buggy compiler. - Added a warning in png_decompress_chunk() when it runs out of data, e.g. - when it tries to read an erroneous PhotoShop iCCP chunk. - Added PNG_USE_DLL macro. - Revised the copyright/disclaimer/license notice. - Added contrib/msvctest directory - -Version 1.0.7rc1 [June 9, 2000] - Corrected the definition of PNG_TRANSFORM_INVERT_ALPHA (0x0400 not 0x0200) - Added contrib/visupng directory (Willem van Schaik) - -Version 1.0.7beta18 [June 23, 2000] - Revised PNGAPI definition, and pngvcrd.c to work with __GCC__ - and do not redefine PNGAPI if it is passed in via a compiler directive. - Revised visupng/PngFile.c to remove returns from within the Try block. - Removed leading underscores from "_PNG_H" and "_PNG_SAVE_BSD_SOURCE" macros. - Updated contrib/visupng/cexcept.h to version 1.0.0. - Fixed bugs in pngwrite.c and pngwutil.c that prevented writing iCCP chunks. - -Version 1.0.7rc2 [June 28, 2000] - Updated license to include disclaimers required by UCITA. - Fixed "DJBPP" typo in pnggccrd.c introduced in beta18. - -Version 1.0.7 [July 1, 2000] - Revised the definition of "trans_values" in libpng.3/libpng.txt - -Version 1.0.8beta1 [July 8, 2000] - Added png_free(png_ptr, key) two places in pngpread.c to stop memory leaks. - Changed PNG_NO_STDIO to PNG_NO_CONSOLE_IO, several places in pngrutil.c and - pngwutil.c. - Changed PNG_EXPORT_VAR to use PNG_IMPEXP, in pngconf.h. - Removed unused "#include " from png.c - Added WindowsCE support. - Revised pnggccrd.c to work with gcc-2.95.2 and in the Cygwin environment. - -Version 1.0.8beta2 [July 10, 2000] - Added project files to the wince directory and made further revisions - of pngtest.c, pngrio.c, and pngwio.c in support of WindowsCE. - -Version 1.0.8beta3 [July 11, 2000] - Only set the PNG_FLAG_FREE_TRNS or PNG_FREE_TRNS flag in png_handle_tRNS() - for indexed-color input files to avoid potential double-freeing trans array - under some unusual conditions; problem was introduced in version 1.0.6f. - Further revisions to pngtest.c and files in the wince subdirectory. - -Version 1.0.8beta4 [July 14, 2000] - Added the files pngbar.png and pngbar.jpg to the distribution. - Added makefile.cygwin, and cygwin support in pngconf.h - Added PNG_NO_ZALLOC_ZERO macro (makes png_zalloc skip zeroing memory) - -Version 1.0.8rc1 [July 16, 2000] - Revised png_debug() macros and statements to eliminate compiler warnings. - -Version 1.0.8 [July 24, 2000] - Added png_flush() in pngwrite.c, after png_write_IEND(). - Updated makefile.hpux to build a shared library. - -Version 1.0.9beta1 [November 10, 2000] - Fixed typo in scripts/makefile.hpux - Updated makevms.com in scripts and contrib/* and contrib/* (Martin Zinser) - Fixed sequence-point bug in contrib/pngminus/png2pnm (Martin Zinser) - Changed "cdrom.com" in documentation to "libpng.org" - Revised pnggccrd.c to get it all working, and updated makefile.gcmmx (Greg). - Changed type of "params" from voidp to png_voidp in png_read|write_png(). - Make sure PNGAPI and PNG_IMPEXP are defined in pngconf.h. - Revised the 3 instances of WRITEFILE in pngtest.c. - Relocated "msvc" and "wince" project subdirectories into "dll" subdirectory. - Updated png.rc in dll/msvc project - Revised makefile.dec to define and use LIBPATH and INCPATH - Increased size of global png_libpng_ver[] array from 12 to 18 chars. - Made global png_libpng_ver[], png_sig[] and png_pass_*[] arrays const. - Removed duplicate png_crc_finish() from png_handle_bKGD() function. - Added a warning when application calls png_read_update_info() multiple times. - Revised makefile.cygwin - Fixed bugs in iCCP support in pngrutil.c and pngwutil.c. - Replaced png_set_empty_plte_permitted() with png_permit_mng_features(). - -Version 1.0.9beta2 [November 19, 2000] - Renamed the "dll" subdirectory "projects". - Added borland project files to "projects" subdirectory. - Set VS_FF_PRERELEASE and VS_FF_PATCHED flags in msvc/png.rc when appropriate. - Add error message in png_set_compression_buffer_size() when malloc fails. - -Version 1.0.9beta3 [November 23, 2000] - Revised PNG_LIBPNG_BUILD_TYPE macro in png.h, used in the msvc project. - Removed the png_flush() in pngwrite.c that crashes some applications - that don't set png_output_flush_fn. - Added makefile.macosx and makefile.aix to scripts directory. - -Version 1.0.9beta4 [December 1, 2000] - Change png_chunk_warning to png_warning in png_check_keyword(). - Increased the first part of msg buffer from 16 to 18 in png_chunk_error(). - -Version 1.0.9beta5 [December 15, 2000] - Added support for filter method 64 (for PNG datastreams embedded in MNG). - -Version 1.0.9beta6 [December 18, 2000] - Revised png_set_filter() to accept filter method 64 when appropriate. - Added new PNG_HAVE_PNG_SIGNATURE bit to png_ptr->mode and use it to - help prevent applications from using MNG features in PNG datastreams. - Added png_permit_mng_features() function. - Revised libpng.3/libpng.txt. Changed "filter type" to "filter method". - -Version 1.0.9rc1 [December 23, 2000] - Revised test for PNG_HAVE_PNG_SIGNATURE in pngrutil.c - Fixed error handling of unknown compression type in png_decompress_chunk(). - In pngconf.h, define __cdecl when _MSC_VER is defined. - -Version 1.0.9beta7 [December 28, 2000] - Changed PNG_TEXT_COMPRESSION_zTXt to PNG_COMPRESSION_TYPE_BASE several places. - Revised memory management in png_set_hIST and png_handle_hIST in a backward - compatible manner. PLTE and tRNS were revised similarly. - Revised the iCCP chunk reader to ignore trailing garbage. - -Version 1.0.9beta8 [January 12, 2001] - Moved pngasmrd.h into pngconf.h. - Improved handling of out-of-spec garbage iCCP chunks generated by PhotoShop. - -Version 1.0.9beta9 [January 15, 2001] - Added png_set_invalid, png_permit_mng_features, and png_mmx_supported to - wince and msvc project module definition files. - Minor revision of makefile.cygwin. - Fixed bug with progressive reading of narrow interlaced images in pngpread.c - -Version 1.0.9beta10 [January 16, 2001] - Do not typedef png_FILE_p in pngconf.h when PNG_NO_STDIO is defined. - Fixed "png_mmx_supported" typo in project definition files. - -Version 1.0.9beta11 [January 19, 2001] - Updated makefile.sgi to make shared library. - Removed png_mmx_support() function and disabled PNG_MNG_FEATURES_SUPPORTED - by default, for the benefit of DLL forward compatibility. These will - be re-enabled in version 1.2.0. - -Version 1.0.9rc2 [January 22, 2001] - Revised cygwin support. - -Version 1.0.9 [January 31, 2001] - Added check of cygwin's ALL_STATIC in pngconf.h - Added "-nommx" parameter to contrib/gregbook/rpng2-win and rpng2-x demos. - -Version 1.0.10beta1 [March 14, 2001] - Revised makefile.dec, makefile.sgi, and makefile.sggcc; added makefile.hpgcc. - Reformatted libpng.3 to eliminate bad line breaks. - Added checks for _mmx_supported in the read_filter_row function of pnggccrd.c - Added prototype for png_mmx_support() near the top of pnggccrd.c - Moved some error checking from png_handle_IHDR to png_set_IHDR. - Added PNG_NO_READ_SUPPORTED and PNG_NO_WRITE_SUPPORTED macros. - Revised png_mmx_support() function in pnggccrd.c - Restored version 1.0.8 PNG_WRITE_EMPTY_PLTE_SUPPORTED behavior in pngwutil.c - Fixed memory leak in contrib/visupng/PngFile.c - Fixed bugs in png_combine_row() in pnggccrd.c and pngvcrd.c (C version) - Added warnings when retrieving or setting gamma=0. - Increased the first part of msg buffer from 16 to 18 in png_chunk_warning(). - -Version 1.0.10rc1 [March 23, 2001] - Changed all instances of memcpy, strcpy, and strlen to png_memcpy, png_strcpy, - and png_strlen. - Revised png_mmx_supported() function in pnggccrd.c to return proper value. - Fixed bug in progressive reading (pngpread.c) with small images (height < 8). - -Version 1.0.10 [March 30, 2001] - Deleted extraneous space (introduced in 1.0.9) from line 42 of makefile.cygwin - Added beos project files (Chris Herborth) - -Version 1.0.11beta1 [April 3, 2001] - Added type casts on several png_malloc() calls (Dimitri Papadapoulos). - Removed a no-longer needed AIX work-around from pngconf.h - Changed several "//" single-line comments to C-style in pnggccrd.c - -Version 1.0.11beta2 [April 11, 2001] - Removed PNGAPI from several functions whose prototypes did not have PNGAPI. - Updated scripts/pngos2.def - -Version 1.0.11beta3 [April 14, 2001] - Added checking the results of many instances of png_malloc() for NULL - -Version 1.0.11beta4 [April 20, 2001] - Undid the changes from version 1.0.11beta3. Added a check for NULL return - from user's malloc_fn(). - Removed some useless type casts of the NULL pointer. - Added makefile.netbsd - -Version 1.0.11 [April 27, 2001] - Revised makefile.netbsd - -Version 1.0.12beta1 [May 14, 2001] - Test for Windows platform in pngconf.h when including malloc.h (Emmanuel Blot) - Updated makefile.cygwin and handling of Cygwin's ALL_STATIC in pngconf.h - Added some never-to-be-executed code in pnggccrd.c to quiet compiler warnings. - Eliminated the png_error about apps using png_read|write_init(). Instead, - libpng will reallocate the png_struct and info_struct if they are too small. - This retains future binary compatibility for old applications written for - libpng-0.88 and earlier. - -Version 1.2.0beta1 [May 6, 2001] - Bumped DLLNUM to 2. - Re-enabled PNG_MNG_FEATURES_SUPPORTED and enabled PNG_ASSEMBLER_CODE_SUPPORTED - by default. - Added runtime selection of MMX features. - Added png_set_strip_error_numbers function and related macros. - -Version 1.2.0beta2 [May 7, 2001] - Finished merging 1.2.0beta1 with version 1.0.11 - Added a check for attempts to read or write PLTE in grayscale PNG datastreams. - -Version 1.2.0beta3 [May 17, 2001] - Enabled user memory function by default. - Modified png_create_struct so it passes user mem_ptr to user memory allocator. - Increased png_mng_features flag from png_byte to png_uint_32. - Bumped shared-library (so-number) and dll-number to 3. - -Version 1.2.0beta4 [June 23, 2001] - Check for missing profile length field in iCCP chunk and free chunk_data - in case of truncated iCCP chunk. - Bumped shared-library number to 3 in makefile.sgi and makefile.sggcc - Bumped dll-number from 2 to 3 in makefile.cygwin - Revised contrib/gregbook/rpng*-x.c to avoid a memory leak and to exit cleanly - if user attempts to run it on an 8-bit display. - Updated contrib/gregbook - Use png_malloc instead of png_zalloc to allocate palette in pngset.c - Updated makefile.ibmc - Added some typecasts to eliminate gcc 3.0 warnings. Changed prototypes - of png_write_oFFS width and height from png_uint_32 to png_int_32. - Updated example.c - Revised prototypes for png_debug_malloc and png_debug_free in pngtest.c - -Version 1.2.0beta5 [August 8, 2001] - Revised contrib/gregbook - Revised makefile.gcmmx - Revised pnggccrd.c to conditionally compile some thread-unsafe code only - when PNG_THREAD_UNSAFE_OK is defined. - Added tests to prevent pngwutil.c from writing a bKGD or tRNS chunk with - value exceeding 2^bit_depth-1 - Revised makefile.sgi and makefile.sggcc - Replaced calls to fprintf(stderr,...) with png_warning() in pnggccrd.c - Removed restriction that do_invert_mono only operate on 1-bit opaque files - -Version 1.2.0 [September 1, 2001] - Changed a png_warning() to png_debug() in pnggccrd.c - Fixed contrib/gregbook/rpng-x.c, rpng2-x.c to avoid crash with XFreeGC(). - -Version 1.2.1beta1 [October 19, 2001] - Revised makefile.std in contrib/pngminus - Include background_1 in png_struct regardless of gamma support. - Revised makefile.netbsd and makefile.macosx, added makefile.darwin. - Revised example.c to provide more details about using row_callback(). - -Version 1.2.1beta2 [October 25, 2001] - Added type cast to each NULL appearing in a function call, except for - WINCE functions. - Added makefile.so9. - -Version 1.2.1beta3 [October 27, 2001] - Removed type casts from all NULLs. - Simplified png_create_struct_2(). - -Version 1.2.1beta4 [November 7, 2001] - Revised png_create_info_struct() and png_creat_struct_2(). - Added error message if png_write_info() was omitted. - Type cast NULLs appearing in function calls when _NO_PROTO or - PNG_TYPECAST_NULL is defined. - -Version 1.2.1rc1 [November 24, 2001] - Type cast NULLs appearing in function calls except when PNG_NO_TYPECAST_NULL - is defined. - Changed typecast of "size" argument to png_size_t in pngmem.c calls to - the user malloc_fn, to agree with the prototype in png.h - Added a pop/push operation to pnggccrd.c, to preserve Eflag (Maxim Sobolev) - Updated makefile.sgi to recognize LIBPATH and INCPATH. - Updated various makefiles so "make clean" does not remove previous major - version of the shared library. - -Version 1.2.1rc2 [December 4, 2001] - Always allocate 256-entry internal palette, hist, and trans arrays, to - avoid out-of-bounds memory reference caused by invalid PNG datastreams. - Added a check for prefix_length > data_length in iCCP chunk handler. - -Version 1.2.1 [December 7, 2001] - None. - -Version 1.2.2beta1 [February 22, 2002] - Fixed a bug with reading the length of iCCP profiles (Larry Reeves). - Revised makefile.linux, makefile.gcmmx, and makefile.sgi to generate - libpng.a, libpng12.so (not libpng.so.3), and libpng12/png.h - Revised makefile.darwin to remove "-undefined suppress" option. - Added checks for gamma and chromaticity values over 21474.83, which exceed - the limit for PNG unsigned 32-bit integers when encoded. - Revised calls to png_create_read_struct() and png_create_write_struct() - for simpler debugging. - Revised png_zalloc() so zlib handles errors (uses PNG_FLAG_MALLOC_NULL_MEM_OK) - -Version 1.2.2beta2 [February 23, 2002] - Check chunk_length and idat_size for invalid (over PNG_MAX_UINT) lengths. - Check for invalid image dimensions in png_get_IHDR. - Added missing "fi;" in the install target of the SGI makefiles. - Added install-static to all makefiles that make shared libraries. - Always do gamma compensation when image is partially transparent. - -Version 1.2.2beta3 [March 7, 2002] - Compute background.gray and background_1.gray even when color_type is RGB - in case image gets reduced to gray later. - Modified shared-library makefiles to install pkgconfig/libpngNN.pc. - Export (with PNGAPI) png_zalloc, png_zfree, and png_handle_as_unknown - Removed unused png_write_destroy_info prototype from png.h - Eliminated incorrect use of width_mmx from pnggccrd.c in pixel_bytes == 8 case - Added install-shared target to all makefiles that make shared libraries. - Stopped a double free of palette, hist, and trans when not using free_me. - Added makefile.32sunu for Sun Ultra 32 and makefile.64sunu for Sun Ultra 64. - -Version 1.2.2beta4 [March 8, 2002] - Compute background.gray and background_1.gray even when color_type is RGB - in case image gets reduced to gray later (Jason Summers). - Relocated a misplaced /bin/rm in the "install-shared" makefile targets - Added PNG_1_0_X macro which can be used to build a 1.0.x-compatible library. - -Version 1.2.2beta5 [March 26, 2002] - Added missing PNGAPI to several function definitions. - Check for invalid bit_depth or color_type in png_get_IHDR(), and - check for missing PLTE or IHDR in png_push_read_chunk() (Matthias Clasen). - Revised iTXt support to accept NULL for lang and lang_key. - Compute gamma for color components of background even when color_type is gray. - Changed "()" to "{}" in scripts/libpng.pc.in. - Revised makefiles to put png.h and pngconf.h only in $prefix/include/libpngNN - Revised makefiles to make symlink to libpng.so.NN in addition to libpngNN.so - -Version 1.2.2beta6 [March 31, 2002] - -Version 1.0.13beta1 [March 31, 2002] - Prevent png_zalloc() from trying to memset memory that it failed to acquire. - Add typecasts of PNG_MAX_UINT in pngset_cHRM_fixed() (Matt Holgate). - Ensure that the right function (user or default) is used to free the - png_struct after an error in png_create_read_struct_2(). - -Version 1.2.2rc1 [April 7, 2002] - -Version 1.0.13rc1 [April 7, 2002] - Save the ebx register in pnggccrd.c (Sami Farin) - Add "mem_ptr = png_ptr->mem_ptr" in png_destroy_write_struct() (Paul Gardner). - Updated makefiles to put headers in include/libpng and remove old include/*.h. - -Version 1.2.2 [April 15, 2002] - -Version 1.0.13 [April 15, 2002] - Revised description of png_set_filter() in libpng.3/libpng.txt. - Revised makefile.netbsd and added makefile.neNNbsd and makefile.freebsd - -Version 1.0.13patch01 [April 17, 2002] - -Version 1.2.2patch01 [April 17, 2002] - Changed ${PNGMAJ}.${PNGVER} bug to ${PNGVER} in makefile.sgi and - makefile.sggcc - Fixed VER -> PNGVER typo in makefile.macosx and added install-static to - install - Added install: target to makefile.32sunu and makefile.64sunu - -Version 1.0.13patch03 [April 18, 2002] - -Version 1.2.2patch03 [April 18, 2002] - Revised 15 makefiles to link libpng.a to libpngNN.a and the include libpng - subdirectory to libpngNN subdirectory without the full pathname. - Moved generation of libpng.pc from "install" to "all" in 15 makefiles. - -Version 1.2.3rc1 [April 28, 2002] - Added install-man target to 15 makefiles (Dimitri Papadopolous-Orfanos). - Added $(DESTDIR) feature to 24 makefiles (Tim Mooney) - Fixed bug with $prefix, should be $(prefix) in makefile.hpux. - Updated cygwin-specific portion of pngconf.h and revised makefile.cygwin - Added a link from libpngNN.pc to libpng.pc in 15 makefiles. - Added links from include/libpngNN/*.h to include/*.h in 24 makefiles. - Revised makefile.darwin to make relative links without full pathname. - Added setjmp() at the end of png_create_*_struct_2() in case user forgets - to put one in their application. - Restored png_zalloc() and png_zfree() prototypes to version 1.2.1 and - removed them from module definition files. - -Version 1.2.3rc2 [May 1, 2002] - Fixed bug in reporting number of channels in pngget.c and pngset.c, - that was introduced in version 1.2.2beta5. - Exported png_zalloc(), png_zfree(), png_default_read(), png_default_write(), - png_default_flush(), and png_push_fill_buffer() and included them in - module definition files. - Added "libpng.pc" dependency to the "install-shared" target in 15 makefiles. - -Version 1.2.3rc3 [May 1, 2002] - Revised prototype for png_default_flush() - Remove old libpng.pc and libpngNN.pc before installing new ones. - -Version 1.2.3rc4 [May 2, 2002] - Typos in *.def files (png_default_read|write -> png_default_read|write_data) - In makefiles, changed rm libpng.NN.pc to rm libpngNN.pc - Added libpng-config and libpngNN-config and modified makefiles to install - them. - Changed $(MANPATH) to $(DESTDIR)$(MANPATH) in makefiles - Added "Win32 DLL VB" configuration to projects/msvc/libpng.dsp - -Version 1.2.3rc5 [May 11, 2002] - Changed "error" and "message" in prototypes to "error_message" and - "warning_message" to avoid namespace conflict. - Revised 15 makefiles to build libpng-config from libpng-config-*.in - Once more restored png_zalloc and png_zfree to regular nonexported form. - Restored png_default_read|write_data, png_default_flush, png_read_fill_buffer - to nonexported form, but with PNGAPI, and removed them from module def - files. - -Version 1.2.3rc6 [May 14, 2002] - Removed "PNGAPI" from png_zalloc() and png_zfree() in png.c - Changed "Gz" to "Gd" in projects/msvc/libpng.dsp and zlib.dsp. - Removed leftover libpng-config "sed" script from four makefiles. - Revised libpng-config creating script in 16 makefiles. - -Version 1.2.3 [May 22, 2002] - Revised libpng-config target in makefile.cygwin. - Removed description of png_set_mem_fn() from documentation. - Revised makefile.freebsd. - Minor cosmetic changes to 15 makefiles, e.g., $(DI) = $(DESTDIR)/$(INCDIR). - Revised projects/msvc/README.txt - Changed -lpng to -lpngNN in LDFLAGS in several makefiles. - -Version 1.2.4beta1 [May 24, 2002] - Added libpng.pc and libpng-config to "all:" target in 16 makefiles. - Fixed bug in 16 makefiles: $(DESTDIR)/$(LIBPATH) to $(DESTDIR)$(LIBPATH) - Added missing "\" before closing double quote in makefile.gcmmx. - Plugged various memory leaks; added png_malloc_warn() and png_set_text_2() - functions. - -Version 1.2.4beta2 [June 25, 2002] - Plugged memory leak of png_ptr->current_text (Matt Holgate). - Check for buffer overflow before reading CRC in pngpread.c (Warwick Allison) - Added -soname to the loader flags in makefile.dec, makefile.sgi, and - makefile.sggcc. - Added "test-installed" target to makefile.linux, makefile.gcmmx, - makefile.sgi, and makefile.sggcc. - -Version 1.2.4beta3 [June 28, 2002] - Plugged memory leak of row_buf in pngtest.c when there is a png_error(). - Detect buffer overflow in pngpread.c when IDAT is corrupted with extra data. - Added "test-installed" target to makefile.32sunu, makefile.64sunu, - makefile.beos, makefile.darwin, makefile.dec, makefile.macosx, - makefile.solaris, makefile.hpux, makefile.hpgcc, and makefile.so9. - -Version 1.2.4rc1 and 1.0.14rc1 [July 2, 2002] - Added "test-installed" target to makefile.cygwin and makefile.sco. - Revised pnggccrd.c to be able to back out version 1.0.x via PNG_1_0_X macro. - -Version 1.2.4 and 1.0.14 [July 8, 2002] - Changed png_warning() to png_error() when width is too large to process. - -Version 1.2.4patch01 [July 20, 2002] - Revised makefile.cygwin to use DLL number 12 instead of 13. - -Version 1.2.5beta1 [August 6, 2002] - Added code to contrib/gregbook/readpng2.c to ignore unused chunks. - Replaced toucan.png in contrib/gregbook (it has been corrupt since 1.0.11) - Removed some stray *.o files from contrib/gregbook. - Changed png_error() to png_warning() about "Too much data" in pngpread.c - and about "Extra compressed data" in pngrutil.c. - Prevent png_ptr->pass from exceeding 7 in png_push_finish_row(). - Updated makefile.hpgcc - Updated png.c and pnggccrd.c handling of return from png_mmx_support() - -Version 1.2.5beta2 [August 15, 2002] - Only issue png_warning() about "Too much data" in pngpread.c when avail_in - is nonzero. - Updated makefiles to install a separate libpng.so.3 with its own rpath. - -Version 1.2.5rc1 and 1.0.15rc1 [August 24, 2002] - Revised makefiles to not remove previous minor versions of shared libraries. - -Version 1.2.5rc2 and 1.0.15rc2 [September 16, 2002] - Revised 13 makefiles to remove "-lz" and "-L$(ZLIBLIB)", etc., from shared - library loader directive. - Added missing "$OBJSDLL" line to makefile.gcmmx. - Added missing "; fi" to makefile.32sunu. - -Version 1.2.5rc3 and 1.0.15rc3 [September 18, 2002] - Revised libpng-config script. - -Version 1.2.5 and 1.0.15 [October 3, 2002] - Revised makefile.macosx, makefile.darwin, makefile.hpgcc, and makefile.hpux, - and makefile.aix. - Relocated two misplaced PNGAPI lines in pngtest.c - -Version 1.2.6beta1 [October 22, 2002] - Commented out warning about uninitialized mmx_support in pnggccrd.c. - Changed "IBMCPP__" flag to "__IBMCPP__" in pngconf.h. - Relocated two more misplaced PNGAPI lines in pngtest.c - Fixed memory overrun bug in png_do_read_filler() with 16-bit datastreams, - introduced in version 1.0.2. - Revised makefile.macosx, makefile.dec, makefile.aix, and makefile.32sunu. - -Version 1.2.6beta2 [November 1, 2002] - Added libpng-config "--ldopts" output. - Added "AR=ar" and "ARFLAGS=rc" and changed "ar rc" to "$(AR) $(ARFLAGS)" - in makefiles. - -Version 1.2.6beta3 [July 18, 2004] - Reverted makefile changes from version 1.2.6beta2 and some of the changes - from version 1.2.6beta1; these will be postponed until version 1.2.7. - Version 1.2.6 is going to be a simple bugfix release. - Changed the one instance of "ln -sf" to "ln -f -s" in each Sun makefile. - Fixed potential overrun in pngerror.c by using strncpy instead of memcpy. - Added "#!/bin/sh" at the top of configure, for recognition of the - 'x' flag under Cygwin (Cosmin). - Optimized vacuous tests that silence compiler warnings, in png.c (Cosmin). - Added support for PNG_USER_CONFIG, in pngconf.h (Cosmin). - Fixed the special memory handler for Borland C under DOS, in pngmem.c - (Cosmin). - Removed some spurious assignments in pngrutil.c (Cosmin). - Replaced 65536 with 65536L, and 0xffff with 0xffffL, to silence warnings - on 16-bit platforms (Cosmin). - Enclosed shift op expressions in parentheses, to silence warnings (Cosmin). - Used proper type png_fixed_point, to avoid problems on 16-bit platforms, - in png_handle_sRGB() (Cosmin). - Added compression_type to png_struct, and optimized the window size - inside the deflate stream (Cosmin). - Fixed definition of isnonalpha(), in pngerror.c and pngrutil.c (Cosmin). - Fixed handling of unknown chunks that come after IDAT (Cosmin). - Allowed png_error() and png_warning() to work even if png_ptr == NULL - (Cosmin). - Replaced row_info->rowbytes with row_bytes in png_write_find_filter() - (Cosmin). - Fixed definition of PNG_LIBPNG_VER_DLLNUM (Simon-Pierre). - Used PNG_LIBPNG_VER and PNG_LIBPNG_VER_STRING instead of the hardcoded - values in png.c (Simon-Pierre, Cosmin). - Initialized png_libpng_ver[] with PNG_LIBPNG_VER_STRING (Simon-Pierre). - Replaced PNG_LIBPNG_VER_MAJOR with PNG_LIBPNG_VER_DLLNUM in png.rc - (Simon-Pierre). - Moved the definition of PNG_HEADER_VERSION_STRING near the definitions - of the other PNG_LIBPNG_VER_... symbols in png.h (Cosmin). - Relocated #ifndef PNGAPI guards in pngconf.h (Simon-Pierre, Cosmin). - Updated scripts/makefile.vc(a)win32 (Cosmin). - Updated the MSVC project (Simon-Pierre, Cosmin). - Updated the Borland C++ Builder project (Cosmin). - Avoided access to asm_flags in pngvcrd.c, if PNG_1_0_X is defined (Cosmin). - Commented out warning about uninitialized mmx_support in pngvcrd.c (Cosmin). - Removed scripts/makefile.bd32 and scripts/pngdef.pas (Cosmin). - Added extra guard around inclusion of Turbo C memory headers, in pngconf.h - (Cosmin). - Renamed projects/msvc/ to projects/visualc6/, and projects/borland/ to - projects/cbuilder5/ (Cosmin). - Moved projects/visualc6/png32ms.def to scripts/pngw32.def, - and projects/visualc6/png.rc to scripts/pngw32.rc (Cosmin). - Added projects/visualc6/pngtest.dsp; removed contrib/msvctest/ (Cosmin). - Changed line endings to DOS style in cbuilder5 and visualc6 files, even - in the tar.* distributions (Cosmin). - Updated contrib/visupng/VisualPng.dsp (Cosmin). - Updated contrib/visupng/cexcept.h to version 2.0.0 (Cosmin). - Added a separate distribution with "configure" and supporting files (Junichi). - -Version 1.2.6beta4 [July 28, 2004] - Added user ability to change png_size_t via a PNG_SIZE_T macro. - Added png_sizeof() and png_convert_size() functions. - Added PNG_SIZE_MAX (maximum value of a png_size_t variable. - Added check in png_malloc_default() for (size_t)size != (png_uint_32)size - which would indicate an overflow. - Changed sPLT failure action from png_error to png_warning and abandon chunk. - Changed sCAL and iCCP failures from png_error to png_warning and abandon. - Added png_get_uint_31(png_ptr, buf) function. - Added PNG_UINT_32_MAX macro. - Renamed PNG_MAX_UINT to PNG_UINT_31_MAX. - Made png_zalloc() issue a png_warning and return NULL on potential - overflow. - Turn on PNG_NO_ZALLOC_ZERO by default in version 1.2.x - Revised "clobber list" in pnggccrd.c so it will compile under gcc-3.4. - Revised Borland portion of png_malloc() to return NULL or issue - png_error() according to setting of PNG_FLAG_MALLOC_NULL_MEM_OK. - Added PNG_NO_SEQUENTIAL_READ_SUPPORTED macro to conditionally remove - sequential read support. - Added some "#if PNG_WRITE_SUPPORTED" blocks. - Added #ifdef to remove some redundancy in png_malloc_default(). - Use png_malloc instead of png_zalloc to allocate the palette. - -Version 1.0.16rc1 and 1.2.6rc1 [August 4, 2004] - Fixed buffer overflow vulnerability (CVE-2004-0597) in png_handle_tRNS(). - Fixed NULL dereference vulnerability (CVE-2004-0598) in png_handle_iCCP(). - Fixed integer overflow vulnerability (CVE-2004-0599) in png_read_png(). - Fixed some harmless bugs in png_handle_sBIT, etc, that would cause - duplicate chunk types to go undetected. - Fixed some timestamps in the -config version - Rearranged order of processing of color types in png_handle_tRNS(). - Added ROWBYTES macro to calculate rowbytes without integer overflow. - Updated makefile.darwin and removed makefile.macosx from scripts directory. - Imposed default one million column, one-million row limits on the image - dimensions, and added png_set_user_limits() function to override them. - Revised use of PNG_SET_USER_LIMITS_SUPPORTED macro. - Fixed wrong cast of returns from png_get_user_width|height_max(). - Changed some "keep the compiler happy" from empty statements to returns, - Revised libpng.txt to remove 1.2.x stuff from the 1.0.x distribution - -Version 1.0.16rc2 and 1.2.6rc2 [August 7, 2004] - Revised makefile.darwin and makefile.solaris. Removed makefile.macosx. - Revised pngtest's png_debug_malloc() to use png_malloc() instead of - png_malloc_default() which is not supposed to be exported. - Fixed off-by-one error in one of the conversions to PNG_ROWBYTES() in - pngpread.c. Bug was introduced in 1.2.6rc1. - Fixed bug in RGB to RGBX transformation introduced in 1.2.6rc1. - Fixed old bug in RGB to Gray transformation. - Fixed problem with 64-bit compilers by casting arguments to abs() - to png_int_32. - Changed "ln -sf" to "ln -f -s" in three makefiles (solaris, sco, so9). - Changed "HANDLE_CHUNK_*" to "PNG_HANDLE_CHUNK_*" (Cosmin) - Added "-@/bin/rm -f $(DL)/$(LIBNAME).so.$(PNGMAJ)" to 15 *NIX makefiles. - Added code to update the row_info->colortype in png_do_read_filler() (MSB). - -Version 1.0.16rc3 and 1.2.6rc3 [August 9, 2004] - Eliminated use of "abs()" in testing cHRM and gAMA values, to avoid - trouble with some 64-bit compilers. Created PNG_OUT_OF_RANGE() macro. - Revised documentation of png_set_keep_unknown_chunks(). - Check handle_as_unknown status in pngpread.c, as in pngread.c previously. - Moved "PNG_HANDLE_CHUNK_*" macros out of PNG_INTERNAL section of png.h - Added "rim" definitions for CONST4 and CONST6 in pnggccrd.c - -Version 1.0.16rc4 and 1.2.6rc4 [August 10, 2004] - Fixed mistake in pngtest.c introduced in 1.2.6rc2 (declaration of - "pinfo" was out of place). - -Version 1.0.16rc5 and 1.2.6rc5 [August 10, 2004] - Moved "PNG_HANDLE_CHUNK_*" macros out of PNG_ASSEMBLER_CODE_SUPPORTED - section of png.h where they were inadvertently placed in version rc3. - -Version 1.2.6 and 1.0.16 [August 15, 2004] - Revised pngtest so memory allocation testing is only done when PNG_DEBUG==1. - -Version 1.2.7beta1 [August 26, 2004] - Removed unused pngasmrd.h file. - Removed references to uu.net for archived files. Added references to - PNG Spec (second edition) and the PNG ISO/IEC Standard. - Added "test-dd" target in 15 makefiles, to run pngtest in DESTDIR. - Fixed bug with "optimized window size" in the IDAT datastream, that - causes libpng to write PNG files with incorrect zlib header bytes. - -Version 1.2.7beta2 [August 28, 2004] - Fixed bug with sCAL chunk and big-endian machines (David Munro). - Undid new code added in 1.2.6rc2 to update the color_type in - png_set_filler(). - Added png_set_add_alpha() that updates color type. - -Version 1.0.17rc1 and 1.2.7rc1 [September 4, 2004] - Revised png_set_strip_filler() to not remove alpha if color_type has alpha. - -Version 1.2.7 and 1.0.17 [September 12, 2004] - Added makefile.hp64 - Changed projects/msvc/png32ms.def to scripts/png32ms.def in makefile.cygwin - -Version 1.2.8beta1 [November 1, 2004] - Fixed bug in png_text_compress() that would fail to complete a large block. - Fixed bug, introduced in libpng-1.2.7, that overruns a buffer during - strip alpha operation in png_do_strip_filler(). - Added PNG_1_2_X definition in pngconf.h - Use #ifdef to comment out png_info_init in png.c and png_read_init in - pngread.c (as of 1.3.0) - -Version 1.2.8beta2 [November 2, 2004] - Reduce color_type to a nonalpha type after strip alpha operation in - png_do_strip_filler(). - -Version 1.2.8beta3 [November 3, 2004] - Revised definitions of PNG_MAX_UINT_32, PNG_MAX_SIZE, and PNG_MAXSUM - -Version 1.2.8beta4 [November 12, 2004] - Fixed (again) definition of PNG_LIBPNG_VER_DLLNUM in png.h (Cosmin). - Added PNG_LIBPNG_BUILD_PRIVATE in png.h (Cosmin). - Set png_ptr->zstream.data_type to Z_BINARY, to avoid unnecessary detection - of data type in deflate (Cosmin). - Deprecated but continue to support SPECIALBUILD and PRIVATEBUILD in favor of - PNG_LIBPNG_BUILD_SPECIAL_STRING and PNG_LIBPNG_BUILD_PRIVATE_STRING. - -Version 1.2.8beta5 [November 20, 2004] - Use png_ptr->flags instead of png_ptr->transformations to pass - PNG_STRIP_ALPHA info to png_do_strip_filler(), to preserve ABI - compatibility. - Revised handling of SPECIALBUILD, PRIVATEBUILD, - PNG_LIBPNG_BUILD_SPECIAL_STRING and PNG_LIBPNG_BUILD_PRIVATE_STRING. - -Version 1.2.8rc1 [November 24, 2004] - Moved handling of BUILD macros from pngconf.h to png.h - Added definition of PNG_LIBPNG_BASE_TYPE in png.h, inadvertently - omitted from beta5. - Revised scripts/pngw32.rc - Despammed mailing addresses by masking "@" with "at". - Inadvertently installed a supposedly faster test version of pngrutil.c - -Version 1.2.8rc2 [November 26, 2004] - Added two missing "\" in png.h - Change tests in pngread.c and pngpread.c to - if (png_ptr->transformations || (png_ptr->flags&PNG_FLAG_STRIP_ALPHA)) - png_do_read_transformations(png_ptr); - -Version 1.2.8rc3 [November 28, 2004] - Reverted pngrutil.c to version libpng-1.2.8beta5. - Added scripts/makefile.elf with supporting code in pngconf.h for symbol - versioning (John Bowler). - -Version 1.2.8rc4 [November 29, 2004] - Added projects/visualc7 (Simon-pierre). - -Version 1.2.8rc5 [November 29, 2004] - Fixed new typo in scripts/pngw32.rc - -Version 1.2.8 [December 3, 2004] - Removed projects/visualc7, added projects/visualc71. - -Version 1.2.9beta1 [February 21, 2006] - Initialized some structure members in pngwutil.c to avoid gcc-4.0.0 complaints - Revised man page and libpng.txt to make it clear that one should not call - png_read_end or png_write_end after png_read_png or png_write_png. - Updated references to png-mng-implement mailing list. - Fixed an incorrect typecast in pngrutil.c - Added PNG_NO_READ_SUPPORTED conditional for making a write-only library. - Added PNG_NO_WRITE_INTERLACING_SUPPORTED conditional. - Optimized alpha-inversion loops in pngwtran.c - Moved test for nonzero gamma outside of png_build_gamma_table() in pngrtran.c - Make sure num_trans is <= 256 before copying data in png_set_tRNS(). - Make sure num_palette is <= 256 before copying data in png_set_PLTE(). - Interchanged order of write_swap_alpha and write_invert_alpha transforms. - Added parentheses in the definition of PNG_LIBPNG_BUILD_TYPE (Cosmin). - Optimized zlib window flag (CINFO) in contrib/pngsuite/*.png (Cosmin). - Updated scripts/makefile.bc32 for Borland C++ 5.6 (Cosmin). - Exported png_get_uint_32, png_save_uint_32, png_get_uint_16, png_save_uint_16, - png_get_int_32, png_save_int_32, png_get_uint_31 (Cosmin). - Added type cast (png_byte) in png_write_sCAL() (Cosmin). - Fixed scripts/makefile.cygwin (Christian Biesinger, Cosmin). - Default iTXt support was inadvertently enabled. - -Version 1.2.9beta2 [February 21, 2006] - Check for png_rgb_to_gray and png_gray_to_rgb read transformations before - checking for png_read_dither in pngrtran.c - Revised checking of chromaticity limits to accommodate extended RGB - colorspace (John Denker). - Changed line endings in some of the project files to CRLF, even in the - "Unix" tar distributions (Cosmin). - Made png_get_int_32 and png_save_int_32 always available (Cosmin). - Updated scripts/pngos2.def, scripts/pngw32.def and projects/wince/png32ce.def - with the newly exported functions. - Eliminated distributions without the "configure" script. - Updated INSTALL instructions. - -Version 1.2.9beta3 [February 24, 2006] - Fixed CRCRLF line endings in contrib/visupng/VisualPng.dsp - Made libpng.pc respect EXEC_PREFIX (D. P. Kreil, J. Bowler) - Removed reference to pngasmrd.h from Makefile.am - Renamed CHANGES to ChangeLog. - Renamed LICENSE to COPYING. - Renamed ANNOUNCE to NEWS. - Created AUTHORS file. - -Version 1.2.9beta4 [March 3, 2006] - Changed definition of PKGCONFIG from $prefix/lib to $libdir in configure.ac - Reverted to filenames LICENSE and ANNOUNCE; removed AUTHORS and COPYING. - Removed newline from the end of some error and warning messages. - Removed test for sqrt() from configure.ac and configure. - Made swap tables in pngtrans.c PNG_CONST (Carlo Bramix). - Disabled default iTXt support that was inadvertently enabled in - libpng-1.2.9beta1. - Added "OS2" to list of systems that don't need underscores, in pnggccrd.c - Removed libpng version and date from *.c files. - -Version 1.2.9beta5 [March 4, 2006] - Removed trailing blanks from source files. - Put version and date of latest change in each source file, and changed - copyright year accordingly. - More cleanup of configure.ac, Makefile.am, and associated scripts. - Restored scripts/makefile.elf which was inadvertently deleted. - -Version 1.2.9beta6 [March 6, 2006] - Fixed typo (RELEASE) in configuration files. - -Version 1.2.9beta7 [March 7, 2006] - Removed libpng.vers and libpng.sym from libpng12_la_SOURCES in Makefile.am - Fixed inconsistent #ifdef's around png_sig_bytes() and png_set_sCAL_s() - in png.h. - Updated makefile.elf as suggested by debian. - Made cosmetic changes to some makefiles, adding LN_SF and other macros. - Made some makefiles accept "exec_prefix". - -Version 1.2.9beta8 [March 9, 2006] - Fixed some "#if defined (..." which should be "#if defined(..." - Bug introduced in libpng-1.2.8. - Fixed inconsistency in definition of png_default_read_data() - Restored blank that was lost from makefile.sggcc "clean" target in beta7. - Revised calculation of "current" and "major" for irix in ltmain.sh - Changed "mkdir" to "MKDIR_P" in some makefiles. - Separated PNG_EXPAND and PNG_EXPAND_tRNS. - Added png_set_expand_gray_1_2_4_to_8() and deprecated - png_set_gray_1_2_4_to_8() which also expands tRNS to alpha. - -Version 1.2.9beta9 [March 10, 2006] - Include "config.h" in pngconf.h when available. - Added some checks for NULL png_ptr or NULL info_ptr (timeless) - -Version 1.2.9beta10 [March 20, 2006] - Removed extra CR from contrib/visualpng/VisualPng.dsw (Cosmin) - Made pnggccrd.c PIC-compliant (Christian Aichinger). - Added makefile.mingw (Wolfgang Glas). - Revised pngconf.h MMX checking. - -Version 1.2.9beta11 [March 22, 2006] - Fixed out-of-order declaration in pngwrite.c that was introduced in beta9 - Simplified some makefiles by using LIBSO, LIBSOMAJ, and LIBSOVER macros. - -Version 1.2.9rc1 [March 31, 2006] - Defined PNG_USER_PRIVATEBUILD when including "pngusr.h" (Cosmin). - Removed nonsensical assertion check from pngtest.c (Cosmin). - -Version 1.2.9 [April 14, 2006] - Revised makefile.beos and added "none" selector in ltmain.sh - -Version 1.2.10beta1 [April 15, 2006] - Renamed "config.h" to "png_conf.h" and revised Makefile.am to add - -DPNG_BUILDING_LIBPNG to compile directive, and modified pngconf.h - to include png_conf.h only when PNG_BUILDING_LIBPNG is defined. - -Version 1.2.10beta2 [April 15, 2006] - Manually updated Makefile.in and configure. Changed png_conf.h.in - back to config.h. - -Version 1.2.10beta3 [April 15, 2006] - Change png_conf.h back to config.h in pngconf.h. - -Version 1.2.10beta4 [April 16, 2006] - Change PNG_BUILDING_LIBPNG to PNG_CONFIGURE_LIBPNG in config/Makefile*. - -Version 1.2.10beta5 [April 16, 2006] - Added a configure check for compiling assembler code in pnggccrd.c - -Version 1.2.10beta6 [April 17, 2006] - Revised the configure check for pnggccrd.c - Moved -DPNG_CONFIGURE_LIBPNG into @LIBPNG_DEFINES@ - Added @LIBPNG_DEFINES@ to arguments when building libpng.sym - -Version 1.2.10beta7 [April 18, 2006] - Change "exec_prefix=$prefix" to "exec_prefix=$(prefix)" in makefiles. - -Version 1.2.10rc1 [April 19, 2006] - Ensure pngconf.h doesn't define both PNG_USE_PNGGCCRD and PNG_USE_PNGVCRD - Fixed "LN_FS" typo in makefile.sco and makefile.solaris. - -Version 1.2.10rc2 [April 20, 2006] - Added a backslash between -DPNG_CONFIGURE_LIBPNG and -DPNG_NO_ASSEMBLER_CODE - in configure.ac and configure - Made the configure warning about versioned symbols less arrogant. - -Version 1.2.10rc3 [April 21, 2006] - Added a note in libpng.txt that png_set_sig_bytes(8) can be used when - writing an embedded PNG without the 8-byte signature. - Revised makefiles and configure to avoid making links to libpng.so.* - -Version 1.2.10 [April 23, 2006] - Reverted configure to "rc2" state. - -Version 1.2.11beta1 [May 31, 2006] - scripts/libpng.pc.in contained "configure" style version info and would - not work with makefiles. - The shared-library makefiles were linking to libpng.so.0 instead of - libpng.so.3 compatibility as the library. - -Version 1.2.11beta2 [June 2, 2006] - Increased sprintf buffer from 50 to 52 chars in pngrutil.c to avoid - buffer overflow. - Fixed bug in example.c (png_set_palette_rgb -> png_set_palette_to_rgb) - -Version 1.2.11beta3 [June 5, 2006] - Prepended "#! /bin/sh" to ltmail.sh and contrib/pngminus/*.sh (Cosmin). - Removed the accidental leftover Makefile.in~ (Cosmin). - Avoided potential buffer overflow and optimized buffer in - png_write_sCAL(), png_write_sCAL_s() (Cosmin). - Removed the include directories and libraries from CFLAGS and LDFLAGS - in scripts/makefile.gcc (Nelson A. de Oliveira, Cosmin). - -Version 1.2.11beta4 [June 6, 2006] - Allow zero-length IDAT chunks after the entire zlib datastream, but not - after another intervening chunk type. - -Version 1.0.19rc1, 1.2.11rc1 [June 13, 2006] - Deleted extraneous square brackets from [config.h] in configure.ac - -Version 1.0.19rc2, 1.2.11rc2 [June 14, 2006] - Added prototypes for PNG_INCH_CONVERSIONS functions to png.h - Revised INSTALL and autogen.sh - Fixed typo in several makefiles (-W1 should be -Wl) - Added typedef for png_int_32 and png_uint_32 on 64-bit systems. - -Version 1.0.19rc3, 1.2.11rc3 [June 15, 2006] - Removed the new typedefs for 64-bit systems (delay until version 1.4.0) - Added one zero element to png_gamma_shift[] array in pngrtran.c to avoid - reading out of bounds. - -Version 1.0.19rc4, 1.2.11rc4 [June 15, 2006] - Really removed the new typedefs for 64-bit systems. - -Version 1.0.19rc5, 1.2.11rc5 [June 22, 2006] - Removed png_sig_bytes entry from scripts/pngw32.def - -Version 1.0.19, 1.2.11 [June 26, 2006] - None. - -Version 1.0.20, 1.2.12 [June 27, 2006] - Really increased sprintf buffer from 50 to 52 chars in pngrutil.c to avoid - buffer overflow. - -Version 1.2.13beta1 [October 2, 2006] - Removed AC_FUNC_MALLOC from configure.ac - Work around Intel-Mac compiler bug by setting PNG_NO_MMX_CODE in pngconf.h - Change "logical" to "bitwise" throughout documentation. - Detect and fix attempt to write wrong iCCP profile length (CVE-2006-7244) - -Version 1.0.21, 1.2.13 [November 14, 2006] - Fix potential buffer overflow in sPLT chunk handler. - Fix Makefile.am to not try to link to noexistent files. - Check all exported functions for NULL png_ptr. - -Version 1.2.14beta1 [November 17, 2006] - Relocated three misplaced tests for NULL png_ptr. - Built Makefile.in with automake-1.9.6 instead of 1.9.2. - Build configure with autoconf-2.60 instead of 2.59 - -Version 1.2.14beta2 [November 17, 2006] - Added some typecasts in png_zalloc(). - -Version 1.2.14rc1 [November 20, 2006] - Changed "strtod" to "png_strtod" in pngrutil.c - -Version 1.0.22, 1.2.14 [November 27, 2006] - Added missing "$(srcdir)" in Makefile.am and Makefile.in - -Version 1.2.15beta1 [December 3, 2006] - Generated configure with autoconf-2.61 instead of 2.60 - Revised configure.ac to update libpng.pc and libpng-config. - -Version 1.2.15beta2 [December 3, 2006] - Always export MMX asm functions, just stubs if not building pnggccrd.c - -Version 1.2.15beta3 [December 4, 2006] - Add "png_bytep" typecast to profile while calculating length in pngwutil.c - -Version 1.2.15beta4 [December 7, 2006] - Added scripts/CMakeLists.txt - Changed PNG_NO_ASSEMBLER_CODE to PNG_NO_MMX_CODE in scripts, like 1.4.0beta - -Version 1.2.15beta5 [December 7, 2006] - Changed some instances of PNG_ASSEMBLER_* to PNG_MMX_* in pnggccrd.c - Revised scripts/CMakeLists.txt - -Version 1.2.15beta6 [December 13, 2006] - Revised scripts/CMakeLists.txt and configure.ac - -Version 1.2.15rc1 [December 18, 2006] - Revised scripts/CMakeLists.txt - -Version 1.2.15rc2 [December 21, 2006] - Added conditional #undef jmpbuf in pngtest.c to undo #define in AIX headers. - Added scripts/makefile.nommx - -Version 1.2.15rc3 [December 25, 2006] - Fixed shared library numbering error that was introduced in 1.2.15beta6. - -Version 1.2.15rc4 [December 27, 2006] - Fixed handling of rgb_to_gray when png_ptr->color.gray isn't set. - -Version 1.2.15rc5 [December 31, 2006] - Revised handling of rgb_to_gray. - -Version 1.2.15 [January 5, 2007] - Added some (unsigned long) typecasts in pngtest.c to avoid printing errors. - -Version 1.2.16beta1 [January 6, 2007] - Fix bugs in makefile.nommx - -Version 1.2.16beta2 [January 16, 2007] - Revised scripts/CMakeLists.txt - -Version 1.2.16 [January 31, 2007] - No changes. - -Version 1.2.17beta1 [March 6, 2007] - Revised scripts/CMakeLists.txt to install both shared and static libraries. - Deleted a redundant line from pngset.c. - -Version 1.2.17beta2 [April 26, 2007] - Relocated misplaced test for png_ptr == NULL in pngpread.c - Change "==" to "&" for testing PNG_RGB_TO_GRAY_ERR & PNG_RGB_TO_GRAY_WARN - flags. - Changed remaining instances of PNG_ASSEMBLER_* to PNG_MMX_* - Added pngerror() when write_IHDR fails in deflateInit2(). - Added "const" to some array declarations. - Mention examples of libpng usage in the libpng*.txt and libpng.3 documents. - -Version 1.2.17rc1 [May 4, 2007] - No changes. - -Version 1.2.17rc2 [May 8, 2007] - Moved several PNG_HAVE_* macros out of PNG_INTERNAL because applications - calling set_unknown_chunk_location() need them. - Changed transformation flag from PNG_EXPAND_tRNS to PNG_EXPAND in - png_set_expand_gray_1_2_4_to_8(). - Added png_ptr->unknown_chunk to hold working unknown chunk data, so it - can be free'ed in case of error. Revised unknown chunk handling in - pngrutil.c and pngpread.c to use this structure. - -Version 1.2.17rc3 [May 8, 2007] - Revised symbol-handling in configure script. - -Version 1.2.17rc4 [May 10, 2007] - Revised unknown chunk handling to avoid storing unknown critical chunks. - -Version 1.0.25 [May 15, 2007] -Version 1.2.17 [May 15, 2007] - Added "png_ptr->num_trans=0" before error return in png_handle_tRNS, - to eliminate a vulnerability (CVE-2007-2445, CERT VU#684664) - -Version 1.0.26 [May 15, 2007] -Version 1.2.18 [May 15, 2007] - Reverted the libpng-1.2.17rc3 change to symbol-handling in configure script - -Version 1.2.19beta1 [May 18, 2007] - Changed "const static" to "static PNG_CONST" everywhere, mostly undoing - change of libpng-1.2.17beta2. Changed other "const" to "PNG_CONST" - Changed some handling of unused parameters, to avoid compiler warnings. - "if (unused == NULL) return;" becomes "unused = unused". - -Version 1.2.19beta2 [May 18, 2007] - Only use the valid bits of tRNS value in png_do_expand() (Brian Cartier) - -Version 1.2.19beta3 [May 19, 2007] - Add some "png_byte" typecasts in png_check_keyword() and write new_key - instead of key in zTXt chunk (Kevin Ryde). - -Version 1.2.19beta4 [May 21, 2007] - Add png_snprintf() function and use it in place of sprint() for improved - defense against buffer overflows. - -Version 1.2.19beta5 [May 21, 2007] - Fixed png_handle_tRNS() to only use the valid bits of tRNS value. - Changed handling of more unused parameters, to avoid compiler warnings. - Removed some PNG_CONST in pngwutil.c to avoid compiler warnings. - -Version 1.2.19beta6 [May 22, 2007] - Added some #ifdef PNG_MMX_CODE_SUPPORTED where needed in pngvcrd.c - Added a special "_MSC_VER" case that defines png_snprintf to _snprintf - -Version 1.2.19beta7 [May 22, 2007] - Squelched png_squelch_warnings() in pnggccrd.c and added - an #ifdef PNG_MMX_CODE_SUPPORTED block around the declarations that caused - the warnings that png_squelch_warnings was squelching. - -Version 1.2.19beta8 [May 22, 2007] - Removed __MMX__ from test in pngconf.h. - -Version 1.2.19beta9 [May 23, 2007] - Made png_squelch_warnings() available via PNG_SQUELCH_WARNINGS macro. - Revised png_squelch_warnings() so it might work. - Updated makefile.sgcc and makefile.solaris; added makefile.solaris-x86. - -Version 1.2.19beta10 [May 24, 2007] - Resquelched png_squelch_warnings(), use "__attribute__((used))" instead. - -Version 1.4.0beta1 [April 20, 2006] - Enabled iTXt support (changes png_struct, thus requires so-number change). - Cleaned up PNG_ASSEMBLER_CODE_SUPPORTED vs PNG_MMX_CODE_SUPPORTED - Eliminated PNG_1_0_X and PNG_1_2_X macros. - Removed deprecated functions png_read_init, png_write_init, png_info_init, - png_permit_empty_plte, png_set_gray_1_2_4_to_8, png_check_sig, and - removed the deprecated macro PNG_MAX_UINT. - Moved "PNG_INTERNAL" parts of png.h and pngconf.h into pngintrn.h - Removed many WIN32_WCE #ifdefs (Cosmin). - Reduced dependency on C-runtime library when on Windows (Simon-Pierre) - Replaced sprintf() with png_sprintf() (Simon-Pierre) - -Version 1.4.0beta2 [April 20, 2006] - Revised makefiles and configure to avoid making links to libpng.so.* - Moved some leftover MMX-related defines from pngconf.h to pngintrn.h - Updated scripts/pngos2.def, pngw32.def, and projects/wince/png32ce.def - -Version 1.4.0beta3 [May 10, 2006] - Updated scripts/pngw32.def to comment out MMX functions. - Added PNG_NO_GET_INT_32 and PNG_NO_SAVE_INT_32 macros. - Scripts/libpng.pc.in contained "configure" style version info and would - not work with makefiles. - Revised pngconf.h and added pngconf.h.in, so makefiles and configure can - pass defines to libpng and applications. - -Version 1.4.0beta4 [May 11, 2006] - Revised configure.ac, Makefile.am, and many of the makefiles to write - their defines in pngconf.h. - -Version 1.4.0beta5 [May 15, 2006] - Added a missing semicolon in Makefile.am and Makefile.in - Deleted extraneous square brackets from configure.ac - -Version 1.4.0beta6 [June 2, 2006] - Increased sprintf buffer from 50 to 52 chars in pngrutil.c to avoid - buffer overflow. - Changed sonum from 0 to 1. - Removed unused prototype for png_check_sig() from png.h - -Version 1.4.0beta7 [June 16, 2006] - Exported png_write_sig (Cosmin). - Optimized buffer in png_handle_cHRM() (Cosmin). - Set pHYs = 2835 x 2835 pixels per meter, and added - sCAL = 0.352778e-3 x 0.352778e-3 meters, in pngtest.png (Cosmin). - Added png_set_benign_errors(), png_benign_error(), png_chunk_benign_error(). - Added typedef for png_int_32 and png_uint_32 on 64-bit systems. - Added "(unsigned long)" typecast on png_uint_32 variables in printf lists. - -Version 1.4.0beta8 [June 22, 2006] - Added demonstration of user chunk support in pngtest.c, to support the - public sTER chunk and a private vpAg chunk. - -Version 1.4.0beta9 [July 3, 2006] - Removed ordinals from scripts/pngw32.def and removed png_info_int and - png_set_gray_1_2_4_to_8 entries. - Inline call of png_get_uint_32() in png_get_uint_31(). - Use png_get_uint_31() to get vpAg width and height in pngtest.c - Removed WINCE and Netware projects. - Removed standalone Y2KINFO file. - -Version 1.4.0beta10 [July 12, 2006] - Eliminated automatic copy of pngconf.h to pngconf.h.in from configure and - some makefiles, because it was not working reliably. Instead, distribute - pngconf.h.in along with pngconf.h and cause configure and some of the - makefiles to update pngconf.h from pngconf.h.in. - Added pngconf.h to DEPENDENCIES in Makefile.am - -Version 1.4.0beta11 [August 19, 2006] - Removed AC_FUNC_MALLOC from configure.ac. - Added a warning when writing iCCP profile with mismatched profile length. - Patched pnggccrd.c to assemble on x86_64 platforms. - Moved chunk header reading into a separate function png_read_chunk_header() - in pngrutil.c. The chunk header (len+sig) is now serialized in a single - operation (Cosmin). - Implemented support for I/O states. Added png_ptr member io_state, and - functions png_get_io_chunk_name() and png_get_io_state() in pngget.c - (Cosmin). - Added png_get_io_chunk_name and png_get_io_state to scripts/*.def (Cosmin). - Renamed scripts/pngw32.* to scripts/pngwin.* (Cosmin). - Removed the include directories and libraries from CFLAGS and LDFLAGS - in scripts/makefile.gcc (Cosmin). - Used png_save_uint_32() to set vpAg width and height in pngtest.c (Cosmin). - Cast to proper type when getting/setting vpAg units in pngtest.c (Cosmin). - Added pngintrn.h to the Visual C++ projects (Cosmin). - Removed scripts/list (Cosmin). - Updated copyright year in scripts/pngwin.def (Cosmin). - Removed PNG_TYPECAST_NULL and used standard NULL consistently (Cosmin). - Disallowed the user to redefine png_size_t, and enforced a consistent use - of png_size_t across libpng (Cosmin). - Changed the type of png_ptr->rowbytes, PNG_ROWBYTES() and friends - to png_size_t (Cosmin). - Removed png_convert_size() and replaced png_sizeof with sizeof (Cosmin). - Removed some unnecessary type casts (Cosmin). - Changed prototype of png_get_compression_buffer_size() and - png_set_compression_buffer_size() to work with png_size_t instead of - png_uint_32 (Cosmin). - Removed png_memcpy_check() and png_memset_check() (Cosmin). - Fixed a typo (png_byte --> png_bytep) in libpng.3 and libpng.txt (Cosmin). - Clarified that png_zalloc() does not clear the allocated memory, - and png_zalloc() and png_zfree() cannot be PNGAPI (Cosmin). - Renamed png_mem_size_t to png_alloc_size_t, fixed its definition in - pngconf.h, and used it in all memory allocation functions (Cosmin). - Renamed pngintrn.h to pngpriv.h, added a comment at the top of the file - mentioning that the symbols declared in that file are private, and - updated the scripts and the Visual C++ projects accordingly (Cosmin). - Removed circular references between pngconf.h and pngconf.h.in in - scripts/makefile.vc*win32 (Cosmin). - Removing trailing '.' from the warning and error messages (Cosmin). - Added pngdefs.h that is built by makefile or configure, instead of - pngconf.h.in (Glenn). - Detect and fix attempt to write wrong iCCP profile length. - -Version 1.4.0beta12 [October 19, 2006] - Changed "logical" to "bitwise" in the documentation. - Work around Intel-Mac compiler bug by setting PNG_NO_MMX_CODE in pngconf.h - Add a typecast to stifle compiler warning in pngrutil.c - -Version 1.4.0beta13 [November 10, 2006] - Fix potential buffer overflow in sPLT chunk handler. - Fix Makefile.am to not try to link to noexistent files. - -Version 1.4.0beta14 [November 15, 2006] - Check all exported functions for NULL png_ptr. - -Version 1.4.0beta15 [November 17, 2006] - Relocated two misplaced tests for NULL png_ptr. - Built Makefile.in with automake-1.9.6 instead of 1.9.2. - Build configure with autoconf-2.60 instead of 2.59 - Add "install: all" in Makefile.am so "configure; make install" will work. - -Version 1.4.0beta16 [November 17, 2006] - Added a typecast in png_zalloc(). - -Version 1.4.0beta17 [December 4, 2006] - Changed "new_key[79] = '\0';" to "(*new_key)[79] = '\0';" in pngwutil.c - Add "png_bytep" typecast to profile while calculating length in pngwutil.c - -Version 1.4.0beta18 [December 7, 2006] - Added scripts/CMakeLists.txt - -Version 1.4.0beta19 [May 16, 2007] - Revised scripts/CMakeLists.txt - Rebuilt configure and Makefile.in with newer tools. - Added conditional #undef jmpbuf in pngtest.c to undo #define in AIX headers. - Added scripts/makefile.nommx - -Version 1.4.0beta20 [July 9, 2008] - Moved several PNG_HAVE_* macros from pngpriv.h to png.h because applications - calling set_unknown_chunk_location() need them. - Moved several macro definitions from pngpriv.h to pngconf.h - Merge with changes to the 1.2.X branch, as of 1.2.30beta04. - Deleted all use of the MMX assembler code and Intel-licensed optimizations. - Revised makefile.mingw - -Version 1.4.0beta21 [July 21, 2008] - Moved local array "chunkdata" from pngrutil.c to the png_struct, so - it will be freed by png_read_destroy() in case of a read error (Kurt - Christensen). - -Version 1.4.0beta22 [July 21, 2008] - Change "purpose" and "buffer" to png_ptr->chunkdata to avoid memory leaking. - -Version 1.4.0beta23 [July 22, 2008] - Change "chunkdata = NULL" to "png_ptr->chunkdata = NULL" several places in - png_decompress_chunk(). - -Version 1.4.0beta24 [July 25, 2008] - Change all remaining "chunkdata" to "png_ptr->chunkdata" in - png_decompress_chunk(), and remove "chunkdata" from parameter list. - Put a call to png_check_chunk_name() in png_read_chunk_header(). - Revised png_check_chunk_name() to reject a name with a lowercase 3rd byte. - Removed two calls to png_check_chunk_name() occurring later in the process. - Define PNG_NO_ERROR_NUMBERS by default in pngconf.h - -Version 1.4.0beta25 [July 30, 2008] - Added a call to png_check_chunk_name() in pngpread.c - Reverted png_check_chunk_name() to accept a name with a lowercase 3rd byte. - Added png_push_have_buffer() function to pngpread.c - Eliminated PNG_BIG_ENDIAN_SUPPORTED and associated png_get_* macros. - Made inline expansion of png_get_*() optional with PNG_USE_READ_MACROS. - Eliminated all PNG_USELESS_TESTS and PNG_CORRECT_PALETTE_SUPPORTED code. - Synced contrib directory and configure files with libpng-1.2.30beta06. - Eliminated no-longer-used pngdefs.h (but it's still built in the makefiles) - Relocated a misplaced "#endif /* PNG_NO_WRITE_FILTER */" in pngwutil.c - -Version 1.4.0beta26 [August 4, 2008] - Removed png_push_have_buffer() function in pngpread.c. It increased the - compiled library size slightly. - Changed "-Wall" to "-W -Wall" in the CFLAGS in all makefiles (Cosmin Truta) - Declared png_ptr "volatile" in pngread.c and pngwrite.c to avoid warnings. - Updated contrib/visupng/cexcept.h to version 2.0.1 - Added PNG_LITERAL_CHARACTER macros for #, [, and ]. - -Version 1.4.0beta27 [August 5, 2008] - Revised usage of PNG_LITERAL_SHARP in pngerror.c. - Moved newline character from individual png_debug messages into the - png_debug macros. - Allow user to #define their own png_debug, png_debug1, and png_debug2. - -Version 1.4.0beta28 [August 5, 2008] - Revised usage of PNG_LITERAL_SHARP in pngerror.c. - Added PNG_STRING_NEWLINE macro - -Version 1.4.0beta29 [August 9, 2008] - Revised usage of PNG_STRING_NEWLINE to work on non-ISO compilers. - Added PNG_STRING_COPYRIGHT macro. - Added non-ISO versions of png_debug macros. - -Version 1.4.0beta30 [August 14, 2008] - Added premultiplied alpha feature (Volker Wiendl). - -Version 1.4.0beta31 [August 18, 2008] - Moved png_set_premultiply_alpha from pngtrans.c to pngrtran.c - Removed extra crc check at the end of png_handle_cHRM(). Bug introduced - in libpng-1.4.0beta20. - -Version 1.4.0beta32 [August 19, 2008] - Added PNG_WRITE_FLUSH_SUPPORTED block around new png_flush() call. - Revised PNG_NO_STDIO version of png_write_flush() - -Version 1.4.0beta33 [August 20, 2008] - Added png_get|set_chunk_cache_max() to limit the total number of sPLT, - text, and unknown chunks that can be stored. - -Version 1.4.0beta34 [September 6, 2008] - Shortened tIME_string to 29 bytes in pngtest.c - Fixed off-by-one error introduced in png_push_read_zTXt() function in - libpng-1.2.30beta04/pngpread.c (Harald van Dijk) - -Version 1.4.0beta35 [October 6, 2008] - Changed "trans_values" to "trans_color". - Changed so-number from 0 to 14. Some OS do not like 0. - Revised makefile.darwin to fix shared library numbering. - Change png_set_gray_1_2_4_to_8() to png_set_expand_gray_1_2_4_to_8() - in example.c (debian bug report) - -Version 1.4.0beta36 [October 25, 2008] - Sync with tEXt vulnerability fix in libpng-1.2.33rc02. - -Version 1.4.0beta37 [November 13, 2008] - Added png_check_cHRM in png.c and moved checking from pngget.c, pngrutil.c, - and pngwrite.c - -Version 1.4.0beta38 [November 22, 2008] - Added check for zero-area RGB cHRM triangle in png_check_cHRM() and - png_check_cHRM_fixed(). - -Version 1.4.0beta39 [November 23, 2008] - Revised png_warning() to write its message on standard output by default - when warning_fn is NULL. - -Version 1.4.0beta40 [November 24, 2008] - Eliminated png_check_cHRM(). Instead, always use png_check_cHRM_fixed(). - In png_check_cHRM_fixed(), ensure white_y is > 0, and removed redundant - check for all-zero coordinates that is detected by the triangle check. - -Version 1.4.0beta41 [November 26, 2008] - Fixed string vs pointer-to-string error in png_check_keyword(). - Rearranged test expressions in png_check_cHRM_fixed() to avoid internal - overflows. - Added PNG_NO_CHECK_cHRM conditional. - -Version 1.4.0beta42, 43 [December 1, 2008] - Merge png_debug with version 1.2.34beta04. - -Version 1.4.0beta44 [December 6, 2008] - Removed redundant check for key==NULL before calling png_check_keyword() - to ensure that new_key gets initialized and removed extra warning - (Merge with version 1.2.34beta05 -- Arvan Pritchard). - -Version 1.4.0beta45 [December 9, 2008] - In png_write_png(), respect the placement of the filler bytes in an earlier - call to png_set_filler() (Jim Barry). - -Version 1.4.0beta46 [December 10, 2008] - Undid previous change and added PNG_TRANSFORM_STRIP_FILLER_BEFORE and - PNG_TRANSFORM_STRIP_FILLER_AFTER conditionals and deprecated - PNG_TRANSFORM_STRIP_FILLER (Jim Barry). - -Version 1.4.0beta47 [December 15, 2008] - Support for dithering was disabled by default, because it has never - been well tested and doesn't work very well. The code has not - been removed, however, and can be enabled by building libpng with - PNG_READ_DITHER_SUPPORTED defined. - -Version 1.4.0beta48 [February 14, 2009] - Added new exported function png_calloc(). - Combined several instances of png_malloc(); png_memset() into png_calloc(). - Removed prototype for png_freeptr() that was added in libpng-1.4.0beta24 - but was never defined. - -Version 1.4.0beta49 [February 28, 2009] - Added png_fileno() macro to pngconf.h, used in pngwio.c - Corrected order of #ifdef's in png_debug definition in png.h - Fixed bug introduced in libpng-1.4.0beta48 with the memset arguments - for pcal_params. - Fixed order of #ifdef directives in the png_debug defines in png.h - (bug introduced in libpng-1.2.34/1.4.0beta29). - Revised comments in png_set_read_fn() and png_set_write_fn(). - -Version 1.4.0beta50 [March 18, 2009] - Use png_calloc() instead of png_malloc() to allocate big_row_buf when - reading an interlaced file, to avoid a possible UMR. - Undid revision of PNG_NO_STDIO version of png_write_flush(). Users - having trouble with fflush() can build with PNG_NO_WRITE_FLUSH defined - or supply their own flush_fn() replacement. - Revised libpng*.txt and png.h documentation about use of png_write_flush() - and png_set_write_fn(). - Removed fflush() from pngtest.c. - Added "#define PNG_NO_WRITE_FLUSH" to contrib/pngminim/encoder/pngusr.h - -Version 1.4.0beta51 [March 21, 2009] - Removed new png_fileno() macro from pngconf.h . - -Version 1.4.0beta52 [March 27, 2009] - Relocated png_do_chop() ahead of building gamma tables in pngrtran.c - This avoids building 16-bit gamma tables unnecessarily. - Removed fflush() from pngtest.c. - Added "#define PNG_NO_WRITE_FLUSH" to contrib/pngminim/encoder/pngusr.h - Added a section on differences between 1.0.x and 1.2.x to libpng.3/libpng.txt - -Version 1.4.0beta53 [April 1, 2009] - Removed some remaining MMX macros from pngpriv.h - Fixed potential memory leak of "new_name" in png_write_iCCP() (Ralph Giles) - -Version 1.4.0beta54 [April 13, 2009] - Added "ifndef PNG_SKIP_SETJMP_CHECK" block in pngconf.h to allow - application code writers to bypass the check for multiple inclusion - of setjmp.h when they know that it is safe to ignore the situation. - Eliminated internal use of setjmp() in pngread.c and pngwrite.c - Reordered ancillary chunks in pngtest.png to be the same as what - pngtest now produces, and made some cosmetic changes to pngtest output. - Eliminated deprecated png_read_init_3() and png_write_init_3() functions. - -Version 1.4.0beta55 [April 15, 2009] - Simplified error handling in pngread.c and pngwrite.c by putting - the new png_read_cleanup() and png_write_cleanup() functions inline. - -Version 1.4.0beta56 [April 25, 2009] - Renamed "user_chunk_data" to "my_user_chunk_data" in pngtest.c to suppress - "shadowed declaration" warning from gcc-4.3.3. - Renamed "gamma" to "png_gamma" in pngset.c to avoid "shadowed declaration" - warning about a global "gamma" variable in math.h on some platforms. - -Version 1.4.0beta57 [May 2, 2009] - Removed prototype for png_freeptr() that was added in libpng-1.4.0beta24 - but was never defined (again). - Rebuilt configure scripts with autoconf-2.63 instead of 2.62 - Removed pngprefs.h and MMX from makefiles - -Version 1.4.0beta58 [May 14, 2009] - Changed pngw32.def to pngwin.def in makefile.mingw (typo was introduced - in beta57). - Clarified usage of sig_bit versus sig_bit_p in example.c (Vincent Torri) - -Version 1.4.0beta59 [May 15, 2009] - Reformatted sources in libpng style (3-space indentation, comment format) - Fixed typo in libpng docs (PNG_FILTER_AVE should be PNG_FILTER_AVG) - Added sections about the git repository and our coding style to the - documentation - Relocated misplaced #endif in pngwrite.c, sCAL chunk handler. - -Version 1.4.0beta60 [May 19, 2009] - Conditionally compile png_read_finish_row() which is not used by - progressive readers. - Added contrib/pngminim/preader to demonstrate building minimal progressive - decoder, based on contrib/gregbook with embedded libpng and zlib. - -Version 1.4.0beta61 [May 20, 2009] - In contrib/pngminim/*, renamed "makefile.std" to "makefile", since there - is only one makefile in those directories, and revised the README files - accordingly. - More reformatting of comments, mostly to capitalize sentences. - -Version 1.4.0beta62 [June 2, 2009] - Added "#define PNG_NO_WRITE_SWAP" to contrib/pngminim/encoder/pngusr.h - and "define PNG_NO_READ_SWAP" to decoder/pngusr.h and preader/pngusr.h - Reformatted several remaining "else statement" into two lines. - Added a section to the libpng documentation about using png_get_io_ptr() - in configure scripts to detect the presence of libpng. - -Version 1.4.0beta63 [June 15, 2009] - Revised libpng*.txt and libpng.3 to mention calling png_set_IHDR() - multiple times and to specify the sample order in the tRNS chunk, - because the ISO PNG specification has a typo in the tRNS table. - Changed several PNG_UNKNOWN_CHUNK_SUPPORTED to - PNG_HANDLE_AS_UNKNOWN_SUPPORTED, to make the png_set_keep mechanism - available for ignoring known chunks even when not saving unknown chunks. - Adopted preference for consistent use of "#ifdef" and "#ifndef" versus - "#if defined()" and "if !defined()" where possible. - -Version 1.4.0beta64 [June 24, 2009] - Eliminated PNG_LEGACY_SUPPORTED code. - Moved the various unknown chunk macro definitions outside of the - PNG_READ|WRITE_ANCILLARY_CHUNK_SUPPORTED blocks. - -Version 1.4.0beta65 [June 26, 2009] - Added a reference to the libpng license in each file. - -Version 1.4.0beta66 [June 27, 2009] - Refer to the libpng license instead of the libpng license in each file. - -Version 1.4.0beta67 [July 6, 2009] - Relocated INVERT_ALPHA within png_read_png() and png_write_png(). - Added high-level API transform PNG_TRANSFORM_GRAY_TO_RGB. - Added an "xcode" project to the projects directory (Alam Arias). - -Version 1.4.0beta68 [July 19, 2009] - Avoid some tests in filter selection in pngwutil.c - -Version 1.4.0beta69 [July 25, 2009] - Simplified the new filter-selection test. This runs faster in the - common "PNG_ALL_FILTERS" and PNG_FILTER_NONE cases. - Removed extraneous declaration from the new call to png_read_gray_to_rgb() - (bug introduced in libpng-1.4.0beta67). - Fixed up xcode project (Alam Arias) - Added a prototype for png_64bit_product() in png.c - -Version 1.4.0beta70 [July 27, 2009] - Avoid a possible NULL dereference in debug build, in png_set_text_2(). - (bug introduced in libpng-0.95, discovered by Evan Rouault) - -Version 1.4.0beta71 [July 29, 2009] - Rebuilt configure scripts with autoconf-2.64. - -Version 1.4.0beta72 [August 1, 2009] - Replaced *.tar.lzma with *.tar.xz in distribution. Get the xz codec - from . - -Version 1.4.0beta73 [August 1, 2009] - Reject attempt to write iCCP chunk with negative embedded profile length - (JD Chen) (CVE-2009-5063). - -Version 1.4.0beta74 [August 8, 2009] - Changed png_ptr and info_ptr member "trans" to "trans_alpha". - -Version 1.4.0beta75 [August 21, 2009] - Removed an extra png_debug() recently added to png_write_find_filter(). - Fixed incorrect #ifdef in pngset.c regarding unknown chunk support. - -Version 1.4.0beta76 [August 22, 2009] - Moved an incorrectly located test in png_read_row() in pngread.c - -Version 1.4.0beta77 [August 27, 2009] - Removed lpXYZ.tar.bz2 (with CRLF), KNOWNBUG, libpng-x.y.z-KNOWNBUG.txt, - and the "noconfig" files from the distribution. - Moved CMakeLists.txt from scripts into the main libpng directory. - Various bugfixes and improvements to CMakeLists.txt (Philip Lowman) - -Version 1.4.0beta78 [August 31, 2009] - Converted all PNG_NO_* tests to PNG_*_SUPPORTED everywhere except pngconf.h - Eliminated PNG_NO_FREE_ME and PNG_FREE_ME_SUPPORTED macros. - Use png_malloc plus a loop instead of png_calloc() to initialize - row_pointers in png_read_png(). - -Version 1.4.0beta79 [September 1, 2009] - Eliminated PNG_GLOBAL_ARRAYS and PNG_LOCAL_ARRAYS; always use local arrays. - Eliminated PNG_CALLOC_SUPPORTED macro and always provide png_calloc(). - -Version 1.4.0beta80 [September 17, 2009] - Removed scripts/libpng.icc - Changed typecast of filler from png_byte to png_uint_16 in png_set_filler(). - (Dennis Gustafsson) - Fixed typo introduced in beta78 in pngtest.c ("#if def " should be "#ifdef ") - -Version 1.4.0beta81 [September 23, 2009] - Eliminated unused PNG_FLAG_FREE_* defines from pngpriv.h - Expanded TAB characters in pngrtran.c - Removed PNG_CONST from all "PNG_CONST PNG_CHNK" declarations to avoid - compiler complaints about doubly declaring things "const". - Changed all "#if [!]defined(X)" to "if[n]def X" where possible. - Eliminated unused png_ptr->row_buf_size - -Version 1.4.0beta82 [September 25, 2009] - Moved redundant IHDR checking into new png_check_IHDR() in png.c - and report all errors found in the IHDR data. - Eliminated useless call to png_check_cHRM() from pngset.c - -Version 1.4.0beta83 [September 25, 2009] - Revised png_check_IHDR() to eliminate bogus complaint about filter_type. - -Version 1.4.0beta84 [September 30, 2009] - Fixed some inconsistent indentation in pngconf.h - Revised png_check_IHDR() to add a test for width variable less than 32-bit. - -Version 1.4.0beta85 [October 1, 2009] - Revised png_check_IHDR() again, to check info_ptr members instead of - the contents of the returned parameters. - -Version 1.4.0beta86 [October 9, 2009] - Updated the "xcode" project (Alam Arias). - Eliminated a shadowed declaration of "pp" in png_handle_sPLT(). - -Version 1.4.0rc01 [October 19, 2009] - Trivial cosmetic changes. - -Version 1.4.0beta87 [October 30, 2009] - Moved version 1.4.0 back into beta. - -Version 1.4.0beta88 [October 30, 2009] - Revised libpng*.txt section about differences between 1.2.x and 1.4.0 - because most of the new features have now been ported back to 1.2.41 - -Version 1.4.0beta89 [November 1, 2009] - More bugfixes and improvements to CMakeLists.txt (Philip Lowman) - Removed a harmless extra png_set_invert_alpha() from pngwrite.c - Apply png_user_chunk_cache_max within png_decompress_chunk(). - Merged libpng-1.2.41.txt with libpng-1.4.0.txt where appropriate. - -Version 1.4.0beta90 [November 2, 2009] - Removed all remaining WIN32_WCE #ifdefs except those involving the - time.h "tm" structure - -Version 1.4.0beta91 [November 3, 2009] - Updated scripts/pngw32.def and projects/wince/png32ce.def - Copied projects/wince/png32ce.def to the scripts directory. - Added scripts/makefile.wce - Patched ltmain.sh for wince support. - Added PNG_CONVERT_tIME_SUPPORTED macro. - -Version 1.4.0beta92 [November 4, 2009] - Make inclusion of time.h in pngconf.h depend on PNG_CONVERT_tIME_SUPPORTED - Make #define PNG_CONVERT_tIME_SUPPORTED depend on PNG_WRITE_tIME_SUPPORTED - Revised libpng*.txt to describe differences from 1.2.40 to 1.4.0 (instead - of differences from 1.2.41 to 1.4.0) - -Version 1.4.0beta93 [November 7, 2009] - Added PNG_DEPSTRUCT, PNG_DEPRECATED, PNG_USE_RESULT, PNG_NORETURN, and - PNG_ALLOCATED macros to detect deprecated direct access to the - png_struct or info_struct members and other deprecated usage in - applications (John Bowler). - Updated scripts/makefile* to add "-DPNG_CONFIGURE_LIBPNG" to CFLAGS, - to prevent warnings about direct access to png structs by libpng - functions while building libpng. They need to be tested, especially - those using compilers other than gcc. - Updated projects/visualc6 and visualc71 with "/d PNG_CONFIGURE_LIBPNG". - They should work but still need to be updated to remove - references to pnggccrd.c or pngvcrd.c and ASM building. - Added README.txt to the beos, cbuilder5, netware, and xcode projects warning - that they need to be updated, to remove references to pnggccrd.c and - pngvcrd.c and to depend on pngpriv.h - Removed three direct references to read_info_ptr members in pngtest.c - that were detected by the new PNG_DEPSTRUCT macro. - Moved the png_debug macro definitions and the png_read_destroy(), - png_write_destroy() and png_far_to_near() prototypes from png.h - to pngpriv.h (John Bowler) - Moved the synopsis lines for png_read_destroy(), png_write_destroy() - png_debug(), png_debug1(), and png_debug2() from libpng.3 to libpngpf.3. - -Version 1.4.0beta94 [November 9, 2009] - Removed the obsolete, unused pnggccrd.c and pngvcrd.c files. - Updated CMakeLists.txt to add "-DPNG_CONFIGURE_LIBPNG" to the definitions. - Removed dependency of pngtest.o on pngpriv.h in the makefiles. - Only #define PNG_DEPSTRUCT, etc. in pngconf.h if not already defined. - -Version 1.4.0beta95 [November 10, 2009] - Changed png_check_sig() to !png_sig_cmp() in contrib programs. - Added -DPNG_CONFIGURE_LIBPNG to contrib/pngminm/*/makefile - Changed png_check_sig() to !png_sig_cmp() in contrib programs. - Corrected the png_get_IHDR() call in contrib/gregbook/readpng2.c - Changed pngminim/*/gather.sh to stop trying to remove pnggccrd.c and pngvcrd.c - Added dependency on pngpriv.h in contrib/pngminim/*/makefile - -Version 1.4.0beta96 [November 12, 2009] - Renamed scripts/makefile.wce to scripts/makefile.cegcc - Revised Makefile.am to use libpng.sys while building libpng.so - so that only PNG_EXPORT functions are exported. - Removed the deprecated png_check_sig() function/macro. - Removed recently removed function names from scripts/*.def - Revised pngtest.png to put chunks in the same order written by pngtest - (evidently the same change made in libpng-1.0beta54 was lost). - Added PNG_PRIVATE macro definition in pngconf.h for possible future use. - -Version 1.4.0beta97 [November 13, 2009] - Restored pngtest.png to the libpng-1.4.0beta7 version. - Removed projects/beos and netware.txt; no one seems to be supporting them. - Revised Makefile.in - -Version 1.4.0beta98 [November 13, 2009] - Added the "xcode" project to zip distributions, - Fixed a typo in scripts/pngwin.def introduced in beta97. - -Version 1.4.0beta99 [November 14, 2009] - Moved libpng-config.in and libpng.pc-configure.in out of the scripts - directory, to libpng-config.in and libpng-pc.in, respectively, and - modified Makefile.am and configure.ac accordingly. Now "configure" - needs nothing from the "scripts" directory. - Avoid redefining PNG_CONST in pngconf.h - -Version 1.4.0beta100 [November 14, 2009] - Removed ASM builds from projects/visualc6 and projects/visualc71 - Removed scripts/makefile.nommx and makefile.vcawin32 - Revised CMakeLists.txt to account for new location of libpng-config.in - and libpng-pc.in - Updated INSTALL to reflect removal and relocation of files. - -Version 1.4.0beta101 [November 14, 2009] - Restored the binary files (*.jpg, *.png, some project files) that were - accidentally deleted from the zip and 7z distributions when the xcode - project was added. - -Version 1.4.0beta102 [November 18, 2009] - Added libpng-config.in and libpng-pc.in to the zip and 7z distributions. - Fixed a typo in projects/visualc6/pngtest.dsp, introduced in beta100. - Moved descriptions of makefiles and other scripts out of INSTALL into - scripts/README.txt - Updated the copyright year in scripts/pngwin.rc from 2006 to 2009. - -Version 1.4.0beta103 [November 21, 2009] - Removed obsolete comments about ASM from projects/visualc71/README_zlib.txt - Align row_buf on 16-byte boundary in memory. - Restored the PNG_WRITE_FLUSH_AFTER_IEND_SUPPORTED guard around the call - to png_flush() after png_write_IEND(). See 1.4.0beta32, 1.4.0beta50 - changes above and 1.2.30, 1.2.30rc01 and rc03 in 1.2.41 CHANGES. Someone - needs this feature. - Make the 'png_jmpbuf' macro expand to a call that records the correct - longjmp function as well as returning a pointer to the setjmp - jmp_buf buffer, and marked direct access to jmpbuf 'deprecated'. - (John Bowler) - -Version 1.4.0beta104 [November 22, 2009] - Removed png_longjmp_ptr from scripts/*.def and libpng.3 - Rebuilt configure scripts with autoconf-2.65 - -Version 1.4.0beta105 [November 25, 2009] - Use fast integer PNG_DIVIDE_BY_255() or PNG_DIVIDE_BY_65535() - to accomplish alpha premultiplication when - PNG_READ_COMPOSITE_NODIV_SUPPORTED is defined. - Changed "/255" to "/255.0" in background calculations to make it clear - that the 255 is used as a double. - -Version 1.4.0beta106 [November 27, 2009] - Removed premultiplied alpha feature. - -Version 1.4.0beta107 [December 4, 2009] - Updated README - Added "#define PNG_NO_PEDANTIC_WARNINGS" in the libpng source files. - Removed "-DPNG_CONFIGURE_LIBPNG" from the makefiles and projects. - Revised scripts/makefile.netbsd, makefile.openbsd, and makefile.sco - to put png.h and pngconf.h in $prefix/include, like the other scripts, - instead of in $prefix/include/libpng. Also revised makefile.sco - to put them in $prefix/include/libpng15 instead of in - $prefix/include/libpng/libpng15. - -Version 1.4.0beta108 [December 11, 2009] - Removed leftover "-DPNG_CONFIGURE_LIBPNG" from contrib/pngminim/*/makefile - Relocated png_do_chop() to its original position in pngrtran.c; the - change in version 1.2.41beta08 caused transparency to be handled wrong - in some 16-bit datastreams (Yusaku Sugai). - -Version 1.4.0beta109 [December 13, 2009] - Added "bit_depth" parameter to the private png_build_gamma_table() function. - Pass bit_depth=8 to png_build_gamma_table() when bit_depth is 16 but the - PNG_16_TO_8 transform has been set, to avoid unnecessary build of 16-bit - tables. - -Version 1.4.0rc02 [December 20, 2009] - Declared png_cleanup_needed "volatile" in pngread.c and pngwrite.c - -Version 1.4.0rc03 [December 22, 2009] - Renamed libpng-pc.in back to libpng.pc.in and revised CMakeLists.txt - (revising the change in 1.4.0beta99) - -Version 1.4.0rc04 [December 25, 2009] - Swapped PNG_UNKNOWN_CHUNKS_SUPPORTED and PNG_HANDLE_AS_UNKNOWN_SUPPORTED - in pngset.c to be consistent with other changes in version 1.2.38. - -Version 1.4.0rc05 [December 25, 2009] - Changed "libpng-pc.in" to "libpng.pc.in" in configure.ac, configure, and - Makefile.in to be consistent with changes in libpng-1.4.0rc03 - -Version 1.4.0rc06 [December 29, 2009] - Reverted the gamma_table changes from libpng-1.4.0beta109. - Fixed some indentation errors. - -Version 1.4.0rc07 [January 1, 2010] - Revised libpng*.txt and libpng.3 about 1.2.x->1.4.x differences. - Use png_calloc() instead of png_malloc(); png_memset() in pngrutil.c - Update copyright year to 2010. - -Version 1.4.0rc08 [January 2, 2010] - Avoid deprecated references to png_ptr-io_ptr and png_ptr->error_ptr - in pngtest.c - -Version 1.4.0 [January 3, 2010] - No changes. - -Version 1.4.1beta01 [January 8, 2010] - Updated CMakeLists.txt for consistent indentation and to avoid an - unclosed if-statement warning (Philip Lowman). - Revised Makefile.am and Makefile.in to remove references to Y2KINFO, - KNOWNBUG, and libpng.la (Robert Schwebel). - Revised the makefiles to install the same files and symbolic - links as configure, except for libpng.la and libpng14.la. - Make png_set|get_compression_buffer_size() available even when - PNG_WRITE_SUPPORTED is not enabled. - Revised Makefile.am and Makefile.in to simplify their maintenance. - Revised scripts/makefile.linux to install a link to libpng14.so.14.1 - -Version 1.4.1beta02 [January 9, 2010] - Revised the rest of the makefiles to install a link to libpng14.so.14.1 - -Version 1.4.1beta03 [January 10, 2010] - Removed png_set_premultiply_alpha() from scripts/*.def - -Version 1.4.1rc01 [January 16, 2010] - No changes. - -Version 1.4.1beta04 [January 23, 2010] - Revised png_decompress_chunk() to improve speed and memory usage when - decoding large chunks. - Added png_set|get_chunk_malloc_max() functions. - -Version 1.4.1beta05 [January 26, 2010] - Relocated "int k" declaration in pngtest.c to minimize its scope. - -Version 1.4.1beta06 [January 28, 2010] - Revised png_decompress_chunk() to use a two-pass method suggested by - John Bowler. - -Version 1.4.1beta07 [February 6, 2010] - Folded some long lines in the source files. - Added definable PNG_USER_CHUNK_CACHE_MAX, PNG_USER_CHUNK_MALLOC_MAX, - and a PNG_USER_LIMITS_SUPPORTED flag. - Eliminated use of png_ptr->irowbytes and reused the slot in png_ptr as - png_ptr->png_user_chunk_malloc_max. - Revised png_push_save_buffer() to do fewer but larger png_malloc() calls. - -Version 1.4.1beta08 [February 6, 2010] - Minor cleanup and updating of dates and copyright year. - -Version 1.5.0beta01 [February 7, 2010] - Moved declaration of png_struct into private pngstruct.h and png_info - into pnginfo.h - -Version 1.4.1beta09 and 1.5.0beta02 [February 7, 2010] - Reverted to original png_push_save_buffer() code. - -Version 1.4.1beta10 and 1.5.0beta03 [February 8, 2010] - Return allocated "old_buffer" in png_push_save_buffer() before - calling png_error(), to avoid a potential memory leak. - Updated configure script to use SO number 15. - -Version 1.5.0beta04 [February 9, 2010] - Removed malformed "incomplete struct declaration" of png_info from png.h - -Version 1.5.0beta05 [February 12, 2010] - Removed PNG_DEPSTRUCT markup in pngstruct.h and pnginfo.h, and undid the - linewrapping that it entailed. - Revised comments in pngstruct.h and pnginfo.h and added pointers to - the libpng license. - Changed PNG_INTERNAL to PNG_EXPOSE_INTERNAL_STRUCTURES - Removed the cbuilder5 project, which has not been updated to 1.4.0. - -Version 1.4.1beta12 and 1.5.0beta06 [February 14, 2010] - Fixed type declaration of png_get_chunk_malloc_max() in pngget.c (Daisuke - Nishikawa) - -Version 1.5.0beta07 [omitted] - -Version 1.5.0beta08 [February 19, 2010] - Changed #ifdef PNG_NO_STDIO_SUPPORTED to #ifdef PNG_NO_CONSOLE_IO_SUPPORTED - wherever png_snprintf() is used to construct error and warning messages. - Noted in scripts/makefile.mingw that it expects to be run under MSYS. - Removed obsolete unused MMX-querying support from contrib/gregbook - Added exported png_longjmp() function. - Removed the AIX redefinition of jmpbuf in png.h - Added -D_ALLSOURCE in configure.ac, makefile.aix, and CMakeLists.txt - when building on AIX. - -Version 1.5.0beta09 [February 19, 2010] - Removed -D_ALLSOURCE from configure.ac, makefile.aix, and CMakeLists.txt. - Changed the name of png_ptr->jmpbuf to png_ptr->png_jmpbuf in pngstruct.h - -Version 1.5.0beta10 [February 25, 2010] - Removed unused gzio.c from contrib/pngminim gather and makefile scripts - Removed replacement error handlers from contrib/gregbook. Because of - the new png_longjmp() function they are no longer needed. - -Version 1.5.0beta11 [March 6, 2010] - Removed checking for already-included setjmp.h from pngconf.h - Fixed inconsistent indentations and made numerous cosmetic changes. - Revised the "SEE ALSO" style of libpng.3, libpngpf.3, and png.5 - -Version 1.5.0beta12 [March 9, 2010] - Moved "#include png.h" inside pngpriv.h and removed "#include png.h" from - the source files, along with "#define PNG_EXPOSE_INTERNAL_STRUCTURES" - and "#define PNG_NO_PEDANTIC_WARNINGS" (John Bowler). - Created new pngdebug.h and moved debug definitions there. - -Version 1.5.0beta13 [March 10, 2010] - Protect pngstruct.h, pnginfo.h, and pngdebug.h from being included twice. - Revise the "#ifdef" blocks in png_inflate() so it will compile when neither - PNG_USER_CHUNK_MALLOC_MAX nor PNG_SET_CHUNK_MALLOC_LIMIT_SUPPORTED - is defined. - Removed unused png_measure_compressed_chunk() from pngpriv.h and libpngpf.3 - Moved the 'config.h' support from pngconf.h to pngpriv.h - Removed PNGAPI from the png_longjmp_ptr typedef. - Eliminated dependence of pngtest.c on the private pngdebug.h file. - Make all png_debug macros into *unterminated* statements or - expressions (i.e. a trailing ';' must always be added) and correct - the format statements in various png_debug messages. - -Version 1.5.0beta14 [March 14, 2010] - Removed direct access to png_ptr->io_ptr from the Windows code in pngtest.c - Revised Makefile.am to account for recent additions and replacements. - Corrected CE and OS/2 DEF files (scripts/png*def) for symbols removed and - added ordinal numbers to the Windows DEF file and corrected the duplicated - ordinal numbers on CE symbols that are commented out. - Added back in export symbols that can be present in the Windows build but - are disabled by default. - PNG_EXPORT changed to include an 'ordinal' field for DEF file generation. - PNG_CALLBACK added to make callback definitions uniform. PNGAPI split - into PNGCAPI (base C form), PNGAPI (exports) and PNGCBAPI (callbacks), - and appropriate changes made to all files. Cygwin builds re-hinged to - allow procedure call standard changes and to remove the need for the DEF - file (fixes build on Cygwin). - Enabled 'attribute' warnings that are relevant to library APIs and callbacks. - Changed rules for generation of the various symbol files and added a new - rule for a DEF file (which is also added to the distribution). - Updated the symbol file generation to stop it adding spurious spaces - to EOL (coming from preprocessor macro expansion). Added a facility - to join tokens in the output and rewrite *.dfn to use this. - Eliminated scripts/*.def in favor of libpng.def; updated projects/visualc71 - and removed scripts/makefile.cygwin. - Made PNG_BUILD_DLL safe: it can be set whenever a DLL is being built. - Removed the include of sys/types.h - apparently unnecessary now on the - platforms on which it happened (all but Mac OS and RISC OS). - Moved the Mac OS test into pngpriv.h (the only place it is used.) - -Version 1.5.0beta15 [March 17, 2010] - Added symbols.chk target to Makefile.am to validate the symbols in png.h - against the new DEF file scripts/symbols.def. - Changed the default DEF file back to pngwin.def. - Removed makefile.mingw. - Eliminated PNG_NO_EXTERN and PNG_ALL_EXTERN - -Version 1.5.0beta16 [April 1, 2010] - Make png_text_struct independent of PNG_iTXt_SUPPORTED, so that - fields are initialized in all configurations. The READ/WRITE - macros (PNG_(READ|WRITE)_iTXt_SUPPORTED) still function as - before to disable code to actually read or write iTXt chunks - and iTXt_SUPPORTED can be used to detect presence of either - read or write support (but it is probably better to check for - the one actually required - read or write.) - Combined multiple png_warning() calls for a single error. - Restored the macro definition of png_check_sig(). - -Version 1.5.0beta17 [April 17, 2010] - Added some "(long)" typecasts to printf calls in png_handle_cHRM(). - Documented the fact that png_set_dither() was disabled since libpng-1.4.0. - Reenabled png_set_dither() but renamed it to png_set_quantize() to reflect - more accurately what it actually does. At the same time, renamed - the PNG_DITHER_[RED,GREEN_BLUE]_BITS macros to - PNG_QUANTIZE_[RED,GREEN,BLUE]_BITS. - Added some "(long)" typecasts to printf calls in png_handle_cHRM(). - Freeze build-time only configuration in the build. - In all prior versions of libpng most configuration options - controlled by compiler #defines had to be repeated by the - application code that used libpng. This patch changes this - so that compilation options that can only be changed at build - time are frozen in the build. Options that are compiler - dependent (and those that are system dependent) are evaluated - each time - pngconf.h holds these. Options that can be changed - per-file in the application are in png.h. Frozen options are - in the new installed header file pnglibconf.h (John Bowler) - Removed the xcode project because it has not been updated to work - with libpng-1.5.0. - Removed the ability to include optional pngusr.h - -Version 1.5.0beta18 [April 17, 2010] - Restored the ability to include optional pngusr.h - Moved replacements for png_error() and png_warning() from the - contrib/pngminim project to pngerror.c, for use when warnings or - errors are disabled via PNG_NO_WARN or PNG_NO_ERROR_TEXT, to avoid - storing unneeded error/warning text. - Updated contrib/pngminim project to work with the new pnglibconf.h - Added some PNG_NO_* defines to contrib/pngminim/*/pngusr.h to save space. - -Version 1.5.0beta19 [April 24, 2010] - Added PNG_{READ,WRITE}_INT_FUNCTIONS_SUPPORTED. This allows the functions - to read and write ints to be disabled independently of PNG_USE_READ_MACROS, - which allows libpng to be built with the functions even though the default - is to use the macros - this allows applications to choose at app build - time whether or not to use macros (previously impossible because the - functions weren't in the default build.) - Changed Windows calling convention back to __cdecl for API functions. - For Windows/x86 platforms only: - __stdcall is no longer needed for Visual Basic, so libpng-1.5.0 uses - __cdecl throughout (both API functions and callbacks) on Windows/x86 - platforms. - Replaced visualc6 and visualc71 projects with new vstudio project - Relaxed the overly-restrictive permissions of some files. - -Version 1.5.0beta20 [April 24, 2010] - Relaxed more overly-restrictive permissions of some files. - -Version 1.5.0beta21 [April 27, 2010] - Removed some unwanted binary bytes and changed CRLF to NEWLINE in the new - vstudio project files, and some trivial editing of some files in the - scripts directory. - Set PNG_NO_READ_BGR, PNG_NO_IO_STATE, and PNG_NO_TIME_RFC1123 in - contrib/pngminim/decoder/pngusr.h to make a smaller decoder application. - -Version 1.5.0beta22 [April 28, 2010] - Fixed dependencies of GET_INT_32 - it does not require READ_INT_FUNCTIONS - because it has a macro equivalent. - Improved the options.awk script; added an "everything off" option. - Revised contrib/pngminim to use the "everything off" option in pngusr.dfa. - -Version 1.5.0beta23 [April 29, 2010] - Corrected PNG_REMOVED macro to take five arguments. - The macro was documented with two arguments (name,ordinal), however - the symbol checking .dfn files assumed five arguments. The five - argument form seems more useful so it is changed to that. - Corrected PNG_UNKNOWN_CHUNKS_SUPPORTED to PNG_HANDLE_AS_UNKNOWN_SUPPORTED - in gregbook/readpng2.c - Corrected protection of png_get_user_transform_ptr. The API declaration in - png.h is removed if both READ and WRITE USER_TRANSFORM are turned off - but was left defined in pngtrans.c - Added logunsupported=1 to cause pnglibconf.h to document disabled options. - This makes the installed pnglibconf.h more readable but causes no - other change. The intention is that users of libpng will find it - easier to understand if an API they need is missing. - Include png_reset_zstream() in png.c only when PNG_READ_SUPPORTED is defined. - Removed dummy_inflate.c from contrib/pngminim/encoder - Removed contrib/pngminim/*/gather.sh; gathering is now done in the makefile. - -Version 1.5.0beta24 [May 7, 2010] - Use bitwise "&" instead of arithmetic mod in pngrutil.c calculation of the - offset of the png_ptr->rowbuf pointer into png_ptr->big_row_buf. - Added more blank lines for readability. - -Version 1.5.0beta25 [June 18, 2010] - In pngpread.c: png_push_have_row() add check for new_row > height - Removed the now-redundant check for out-of-bounds new_row from example.c - -Version 1.5.0beta26 [June 18, 2010] - In pngpread.c: png_push_process_row() add check for too many rows. - -Version 1.5.0beta27 [June 18, 2010] - Removed the check added in beta25 as it is now redundant. - -Version 1.5.0beta28 [June 20, 2010] - Rewrote png_process_IDAT_data to consistently treat extra data as warnings - and handle end conditions more cleanly. - Removed the new (beta26) check in png_push_process_row(). - -Version 1.5.0beta29 [June 21, 2010] - Revised scripts/options.awk to work on Sunos (but still doesn't work) - Added comment to options.awk and contrib/pngminim/*/makefile to try nawk. - -Version 1.5.0beta30 [June 22, 2010] - Stop memory leak when reading a malformed sCAL chunk. - -Version 1.5.0beta31 [June 26, 2010] - Revised pngpread.c patch of beta28 to avoid an endless loop. - Removed some trailing blanks. - -Version 1.5.0beta32 [June 26, 2010] - Removed leftover scripts/options.patch and scripts/options.rej - -Version 1.5.0beta33 [July 6, 3010] - Made FIXED and FLOATING options consistent in the APIs they enable and - disable. Corrected scripts/options.awk to handle both command line - options and options specified in the .dfa files. - Changed char *msg to PNG_CONST char *msg in pngrutil.c - Make png_set_sRGB_gAMA_and_cHRM set values using either the fixed or - floating point APIs, but not both. - Reversed patch to remove error handler when the jmp_buf is stored in the - main program structure, not the png_struct. - The error handler is needed because the default handler in libpng will - always use the jmp_buf in the library control structure; this is never - set. The gregbook code is a useful example because, even though it - uses setjmp/longjmp, it shows how error handling can be implemented - using control mechanisms not directly supported by libpng. The - technique will work correctly with mechanisms such as Microsoft - Structure Exceptions or C++ exceptions (compiler willing - note that gcc - does not by default support interworking of C and C++ error handling.) - Reverted changes to call png_longjmp in contrib/gregbook where it is not - appropriate. If mainprog->jmpbuf is used by setjmp, then png_longjmp - cannot be used. - Changed "extern PNG_EXPORT" to "PNG_EXPORT" in png.h (Jan Nijtmans) - Changed "extern" to "PNG_EXTERN" in pngpriv.h (except for the 'extern "C" {') - -Version 1.5.0beta34 [July 12, 2010] - Put #ifndef PNG_EXTERN, #endif around the define PNG_EXTERN in pngpriv.h - -Version 1.5.0beta35 [July 24, 2010] - Removed some newly-added TAB characters. - Added -DNO_PNG_SNPRINTF to CFLAGS in scripts/makefile.dj2 - Moved the definition of png_snprintf() outside of the enclosing - #ifdef blocks in pngconf.h - -Version 1.5.0beta36 [July 29, 2010] - Patches by John Bowler: - Fixed point APIs are now supported throughout (no missing APIs). - Internal fixed point arithmetic support exists for all internal floating - point operations. - sCAL validates the floating point strings it is passed. - Safe, albeit rudimentary, Watcom support is provided by PNG_API_RULE==2 - Two new APIs exist to get the number of passes without turning on the - PNG_INTERLACE transform and to get the number of rows in the current - pass. - A new test program, pngvalid.c, validates the gamma code. - Errors in the 16-bit gamma correction (overflows) have been corrected. - cHRM chunk testing is done consistently (previously the floating point - API bypassed it, because the test really didn't work on FP, now the test - is performed on the actual values to be stored in the PNG file so it - works in the FP case too.) - Most floating point APIs now simply call the fixed point APIs after - converting the values to the fixed point form used in the PNG file. - The standard headers no longer include zlib.h, which is currently only - required for pngstruct.h and can therefore be internal. - Revised png_get_int_32 to undo the PNG two's complement representation of - negative numbers. - -Version 1.5.0beta37 [July 30, 2010] - Added a typecast in png_get_int_32() in png.h and pngrutil.h to avoid - a compiler warning. - Replaced oFFs 0,0 with oFFs -10,20 in pngtest.png - -Version 1.5.0beta38 [July 31, 2010] - Implemented remaining "_fixed" functions. - Corrected a number of recently introduced warnings mostly resulting from - safe but uncast assignments to shorter integers. Also added a zlib - VStudio release library project because the latest zlib Official Windows - build does not include such a thing. - Revised png_get_int_16() to be similar to png_get_int_32(). - Restored projects/visualc71. - -Version 1.5.0beta39 [August 2, 2010] - VisualC/GCC warning fixes, VisualC build fixes - The changes include support for function attributes in VC in addition to - those already present in GCC - necessary because without these some - warnings are unavoidable. Fixes include signed/unsigned fixes in - pngvalid and checks with gcc -Wall -Wextra -Wunused. - VC requires function attributes on function definitions as well as - declarations, PNG_FUNCTION has been added to enable this and the - relevant function definitions changed. - -Version 1.5.0beta40 [August 6, 2010] - Correct use of _WINDOWS_ in pngconf.h - Removed png_mem_ #defines; they are no longer used. - Added the sRGB chunk to pngtest.png - -Version 1.5.0beta41 [August 11, 2010] - Added the cHRM chunk to pngtest.png - Don't try to use version-script with cygwin/mingw. - Revised contrib/gregbook to work under cygwin/mingw. - -Version 1.5.0beta42 [August 18, 2010] - Add .dll.a to the list of extensions to be symlinked by Makefile.am (Yaakov) - Made all API functions that have const arguments and constant string - literal pointers declare them (John Bowler). - -Version 1.5.0beta43 [August 20, 2010] - Removed spurious tabs, shorten long lines (no source change) - Also added scripts/chkfmt to validate the format of all the files that can - reasonably be validated (it is suggested to run "make distclean" before - checking, because some machine generated files have long lines.) - Reformatted the CHANGES file to be more consistent throughout. - Made changes to address various issues identified by GCC, mostly - signed/unsigned and shortening problems on assignment but also a few - difficult to optimize (for GCC) loops. - Fixed non-GCC fixed point builds. In png.c a declaration was misplaced - in an earlier update. Fixed to declare the auto variables at the head. - Use cexcept.h in pngvalid.c. - -Version 1.5.0beta44 [August 24, 2010] - Updated CMakeLists.txt to use CMAKE_INSTALL_LIBDIR variable; useful for - installing libpng in /usr/lib64 (Funda Wang). - Revised CMakeLists.txt to put the man pages in share/man/man* not man/man* - Revised CMakeLists.txt to make symlinks instead of copies when installing. - Changed PNG_LIB_NAME from pngNN to libpngNN in CMakeLists.txt (Philip Lowman) - Implemented memory checks within pngvalid - Reformatted/rearranged pngvalid.c to assist use of progressive reader. - Check interlaced images in pngvalid - Clarified pngusr.h comments in pnglibconf.dfa - Simplified the pngvalid error-handling code now that cexcept.h is in place. - Implemented progressive reader in pngvalid.c for standard tests - Implemented progressive read in pngvalid.c gamma tests - Turn on progressive reader in pngvalid.c by default and tidy code. - -Version 1.5.0beta45 [August 26, 2010] - Added an explicit make step to projects/vstudio for pnglibconf.h - Also corrected zlib.vcxproj into which Visual Studio had introduced - what it calls an "authoring error". The change to make pnglibconf.h - simply copies the file; in the future it may actually generate the - file from scripts/pnglibconf.dfa as the other build systems do. - Changed pngvalid to work when floating point APIs are disabled - Renamed the prebuilt scripts/pnglibconf.h to scripts/pnglibconf.h.prebuilt - Supply default values for PNG_USER_PRIVATEBUILD and PNG_USER_DLLFNAME_POSTFIX - in pngpriv.h in case the user neglected to define them in their pngusr.h - -Version 1.5.0beta46 [August 28, 2010] - Added new private header files to libpng_sources in CMakeLists.txt - Added PNG_READ_16BIT, PNG_WRITE_16BIT, and PNG_16BIT options. - Added reference to scripts/pnglibconf.h.prebuilt in the visualc71 project. - -Version 1.5.0beta47 [September 11, 2010] - Fixed a number of problems with 64-bit compilation reported by Visual - Studio 2010 (John Bowler). - -Version 1.5.0beta48 [October 4, 2010] - Updated CMakeLists.txt (Philip Lowman). - Revised autogen.sh to recognize and use $AUTOCONF, $AUTOMAKE, $AUTOHEADER, - $AUTOPOINT, $ACLOCAL and $LIBTOOLIZE - Fixed problem with symbols creation in Makefile.am which was assuming that - all versions of ccp write to standard output by default (Martin Banky). The - bug was introduced in libpng-1.2.9beta5. - Removed unused mkinstalldirs. - -Version 1.5.0beta49 [October 8, 2010] - Undid Makefile.am revision of 1.5.0beta48. - -Version 1.5.0beta50 [October 14, 2010] - Revised Makefile.in to account for mkinstalldirs being removed. - Added some "(unsigned long)" typecasts in printf statements in pngvalid.c. - Suppressed a compiler warning in png_handle_sPLT(). - Check for out-of-range text compression mode in png_set_text(). - -Version 1.5.0beta51 [October 15, 2010] - Changed embedded dates to "(PENDING RELEASE) in beta releases (and future - rc releases) to minimize the difference between releases. - -Version 1.5.0beta52 [October 16, 2010] - Restored some of the embedded dates (in png.h, png.c, documentation, etc.) - -Version 1.5.0beta53 [October 18, 2010] - Updated INSTALL to mention using "make maintainer-clean" and to remove - obsolete statement about a custom ltmain.sh - Disabled "color-tests" by default in Makefile.am so it will work with - automake versions earlier than 1.11.1 - Use document name "libpng-manual.txt" instead of "libpng-.txt" - to simplify version differences. - Removed obsolete remarks about setjmp handling from INSTALL. - Revised and renamed the typedef in png.h and png.c that was designed - to catch library and header mismatch. - -Version 1.5.0beta54 [November 10, 2010] - Require 48 bytes, not 64 bytes, for big_row_buf in overflow checks. - Used a consistent structure for the pngget.c functions. - -Version 1.5.0beta55 [November 21, 2010] - Revised png_get_uint_32, png_get_int_32, png_get_uint_16 (Cosmin) - Moved reading of file signature into png_read_sig (Cosmin) - Fixed atomicity of chunk header serialization (Cosmin) - Added test for io_state in pngtest.c (Cosmin) - Added "#!/bin/sh" at the top of contrib/pngminim/*/gather.sh scripts. - Changes to remove gcc warnings (John Bowler) - Certain optional gcc warning flags resulted in warnings in libpng code. - With these changes only -Wconversion and -Wcast-qual cannot be turned on. - Changes are trivial rearrangements of code. -Wconversion is not possible - for pngrutil.c (because of the widespread use of += et al on variables - smaller than (int) or (unsigned int)) and -Wcast-qual is not possible - with pngwio.c and pngwutil.c because the 'write' callback and zlib - compression both fail to declare their input buffers with 'const'. - -Version 1.5.0beta56 [December 7, 2010] - Added the private PNG_UNUSED() macro definition in pngpriv.h. - Added some commentary about PNG_EXPORT in png.h and pngconf.h - Revised PNG_EXPORT() macro and added PNG_EXPORTA() macro, with the - objective of simplifying and improving the cosmetic appearance of png.h. - Fixed some incorrect "=" macro names in pnglibconf.dfa - Included documentation of changes in 1.5.0 from 1.4.x in libpng-manual.txt - -Version 1.5.0beta57 [December 9, 2010] - Documented the pngvalid gamma error summary with additional comments and - print statements. - Improved missing symbol handling in checksym.awk; symbols missing in both - the old and new files can now be optionally ignored, treated as errors - or warnings. - Removed references to pngvcrd.c and pnggccrd.c from the vstudio project. - Updated "libpng14" to "libpng15" in the visualc71 project. - Enabled the strip16 tests in pngvalid.` - Don't display test results (except PASS/FAIL) when running "make test". - Instead put them in pngtest-log.txt - Added "--with-zprefix=" to configure.ac - Updated the prebuilt configuration files to autoconf version 2.68 - -Version 1.5.0beta58 [December 19, 2010] - Fixed interlace image handling and add test cases (John Bowler) - Fixed the clean rule in Makefile.am to remove pngtest-log.txt - Made minor changes to work around warnings in gcc 3.4 - -Version 1.5.0rc01 [December 27, 2010] - No changes. - -Version 1.5.0rc02 [December 27, 2010] - Eliminated references to the scripts/*.def files in project/visualc71. - -Version 1.5.0rc03 [December 28, 2010] - Eliminated scripts/*.def and revised Makefile.am accordingly - -Version 1.5.0rc04 [December 29, 2010] - Fixed bug in background transformation handling in pngrtran.c (it was - looking for the flag in png_ptr->transformations instead of in - png_ptr->flags) (David Raymond). - -Version 1.5.0rc05 [December 31, 2010] - Fixed typo in a comment in CMakeLists.txt (libpng14 => libpng15) (Cosmin) - -Version 1.5.0rc06 [January 4, 2011] - Changed the new configure option "zprefix=string" to "zlib-prefix=string" - -Version 1.5.0rc07 [January 4, 2011] - Updated copyright year. - -Version 1.5.0 [January 6, 2011] - No changes. - -version 1.5.1beta01 [January 8, 2011] - Added description of png_set_crc_action() to the manual. - Added a note in the manual that the type of the iCCP profile was changed - from png_charpp to png_bytepp in png_get_iCCP(). This change happened - in version 1.5.0beta36 but is not noted in the CHANGES. Similarly, - it was changed from png_charpp to png_const_bytepp in png_set_iCCP(). - Ensure that png_rgb_to_gray ignores palette mapped images, if libpng - internally happens to call it with one, and fixed a failure to handle - palette mapped images correctly. This fixes CVE-2690. - -Version 1.5.1beta02 [January 14, 2011] - Fixed a bug in handling of interlaced images (bero at arklinux.org). - Updated CMakeLists.txt (Clifford Yapp) - -Version 1.5.1beta03 [January 14, 2011] - Fixed typecasting of some png_debug() statements (Cosmin) - -Version 1.5.1beta04 [January 16, 2011] - Updated documentation of png_set|get_tRNS() (Thomas Klausner). - Mentioned in the documentation that applications must #include "zlib.h" - if they need access to anything in zlib.h, and that a number of - macros such as png_memset() are no longer accessible by applications. - Corrected pngvalid gamma test "sample" function to access all of the color - samples of each pixel, instead of sampling the red channel three times. - Prefixed variable names index, div, exp, gamma with "png_" to avoid "shadow" - warnings, and (mistakenly) changed png_exp() to exp(). - -Version 1.5.1beta05 [January 16, 2011] - Changed variable names png_index, png_div, png_exp, and png_gamma to - char_index, divisor, exp_b10, and gamma_val, respectively, and - changed exp() back to png_exp(). - -Version 1.5.1beta06 [January 20, 2011] - Prevent png_push_crc_skip() from hanging while reading an unknown chunk - or an over-large compressed zTXt chunk with the progressive reader. - Eliminated more GCC "shadow" warnings. - Revised png_fixed() in png.c to avoid compiler warning about reaching the - end without returning anything. - -Version 1.5.1beta07 [January 22, 2011] - In the manual, describe the png_get_IHDR() arguments in the correct order. - Added const_png_structp and const_png_infop types, and used them in - prototypes for most png_get_*() functions. - -Version 1.5.1beta08 [January 23, 2011] - Added png_get_io_chunk_type() and deprecated png_get_io_chunk_name() - Added synopses for the IO_STATE functions and other missing synopses - to the manual. Removed the synopses from libpngpf.3 because they - were out of date and no longer useful. Better information can be - obtained by reading the prototypes and comments in pngpriv.h - Attempted to fix cpp on Solaris with S. Studio 12 cc, fix build - Added a make macro DFNCPP that is a CPP that will accept the tokens in - a .dfn file and adds configure stuff to test for such a CPP. ./configure - should fail if one is not available. - Corrected const_png_ in png.h to png_const_ to avoid polluting the namespace. - Added png_get_current_row_number and png_get_current_pass_number for the - benefit of the user transform callback. - Added png_process_data_pause and png_process_data_skip for the benefit of - progressive readers that need to stop data processing or want to optimize - skipping of unread data (e.g., if the reader marks a chunk to be skipped.) - -Version 1.5.1beta09 [January 24, 2011] - Enhanced pngvalid, corrected an error in gray_to_rgb, corrected doc error. - pngvalid contains tests of transforms, which tests are currently disabled - because they are incompletely tested. gray_to_rgb was failing to expand - the bit depth for smaller bit depth images; this seems to be a long - standing error and resulted, apparently, in invalid output - (CVE-2011-0408, CERT VU#643140). The documentation did not accurately - describe what libpng really does when converting RGB to gray. - -Version 1.5.1beta10 [January 27, 2010] - Fixed incorrect examples of callback prototypes in the manual, that were - introduced in libpng-1.0.0. - In addition the order of the png_get_uint macros with respect to the - relevant function definitions has been reversed. This helps the - preprocessing of the symbol files be more robust. Furthermore, the - symbol file preprocessing now uses -DPNG_NO_USE_READ_MACROS even when - the library may actually be built with PNG_USE_READ_MACROS; this stops - the read macros interfering with the symbol file format. - Made the manual, synopses, and function prototypes use the function - argument names file_gamma, int_file_gamma, and srgb_intent consistently. - -Version 1.5.1beta11 [January 28, 2011] - Changed PNG_UNUSED from "param=param;" to "{if(param){}}". - Corrected local variable type in new API png_process_data_skip() - The type was self-evidently incorrect but only causes problems on 64-bit - architectures. - Added transform tests to pngvalid and simplified the arguments. - -Version 1.5.1rc01 [January 29, 2011] - No changes. - -Version 1.5.1rc02 [January 31, 2011] - Added a request in the manual that applications do not use "png_" or - "PNG_" to begin any of their own symbols. - Changed PNG_UNUSED to "(void)param;" and updated the commentary in pngpriv.h - -Version 1.5.1 [February 3, 2011] - No changes. - -Version 1.5.2beta01 [February 13, 2011] - More -Wshadow fixes for older gcc compilers. Older gcc versions apparently - check formal parameters names in function declarations (as well as - definitions) to see if they match a name in the global namespace. - Revised PNG_EXPORTA macro to not use an empty parameter, to accommodate the - old VisualC++ preprocessor. - Turned on interlace handling in png_read_png(). - Fixed gcc pedantic warnings. - Handle longjmp in Cygwin. - Fixed png_get_current_row_number() in the interlaced case. - Cleaned up ALPHA flags and transformations. - Implemented expansion to 16 bits. - -Version 1.5.2beta02 [February 19, 2011] - Fixed mistake in the descriptions of user read_transform and write_transform - function prototypes in the manual. The row_info struct is png_row_infop. - Reverted png_get_current_row_number() to previous (1.5.2beta01) behavior. - Corrected png_get_current_row_number documentation - Fixed the read/write row callback documentation. - This documents the current behavior, where the callback is called after - every row with information pertaining to the next row. - -Version 1.5.2beta03 [March 3, 2011] - Fixed scripts/makefile.vcwin32 - Updated contrib/pngsuite/README to add the word "modify". - Define PNG_ALLOCATED to blank when _MSC_VER<1300. - -Version 1.5.2rc01 [March 19, 2011] - Define remaining attributes to blank when MSC_VER<1300. - ifdef out mask arrays in pngread.c when interlacing is not supported. - -Version 1.5.2rc02 [March 22, 2011] - Added a hint to try CPP=/bin/cpp if "cpp -E" fails in scripts/pnglibconf.mak - and in contrib/pngminim/*/makefile, eg., on SunOS 5.10, and removed "strip" - from the makefiles. - Fixed a bug (present since libpng-1.0.7) that makes png_handle_sPLT() fail - to compile when PNG_NO_POINTER_INDEXING is defined (Chubanov Kirill) - -Version 1.5.2rc03 [March 24, 2011] - Don't include standard header files in png.h while building the symbol table, - to avoid cpp failure on SunOS (introduced PNG_BUILDING_SYMBOL_TABLE macro). - -Version 1.5.2 [March 31, 2011] - No changes. - -Version 1.5.3beta01 [April 1, 2011] - Re-initialize the zlib compressor before compressing non-IDAT chunks. - Added API functions (png_set_text_compression_level() and four others) to - set parameters for zlib compression of non-IDAT chunks. - -Version 1.5.3beta02 [April 3, 2011] - Updated scripts/symbols.def with new API functions. - Only compile the new zlib re-initializing code when text or iCCP is - supported, using PNG_WRITE_COMPRESSED_TEXT_SUPPORTED macro. - Improved the optimization of the zlib CMF byte (see libpng-1.2.6beta03). - Optimize the zlib CMF byte in non-IDAT compressed chunks - -Version 1.5.3beta03 [April 16, 2011] - Fixed gcc -ansi -pedantic compile. A strict ANSI system does not have - snprintf, and the "__STRICT_ANSI__" detects that condition more reliably - than __STDC__ (John Bowler). - Removed the PNG_PTR_NORETURN attribute because it too dangerous. It tells - the compiler that a user supplied callback (the error handler) does not - return, yet there is no guarantee in practice that the application code - will correctly implement the error handler because the compiler only - issues a warning if there is a mistake (John Bowler). - Removed the no-longer-used PNG_DEPSTRUCT macro. - Updated the zlib version to 1.2.5 in the VStudio project. - Fixed 64-bit builds where png_uint_32 is smaller than png_size_t in - pngwutil.c (John Bowler). - Fixed bug with stripping the filler or alpha channel when writing, that - was introduced in libpng-1.5.2beta01 (bug report by Andrew Church). - -Version 1.5.3beta04 [April 27, 2011] - Updated pngtest.png with the new zlib CMF optimization. - Cleaned up conditional compilation code and of background/gamma handling - Internal changes only except a new option to avoid compiling the - png_build_grayscale_palette API (which is not used at all internally.) - The main change is to move the transform tests (READ_TRANSFORMS, - WRITE_TRANSFORMS) up one level to the caller of the APIs. This avoids - calls to spurious functions if all transforms are disabled and slightly - simplifies those functions. Pngvalid modified to handle this. - A minor change is to stop the strip_16 and expand_16 interfaces from - disabling each other; this allows the future alpha premultiplication - code to use 16-bit intermediate values while still producing 8-bit output. - png_do_background and png_do_gamma have been simplified to take a single - pointer to the png_struct rather than pointers to every item required - from the png_struct. This makes no practical difference to the internal - code. - A serious bug in the pngvalid internal routine 'standard_display_init' has - been fixed - this failed to initialize the red channel and accidentally - initialized the alpha channel twice. - Changed png_struct jmp_buf member name from png_jmpbuf to tmp_jmpbuf to - avoid a possible clash with the png_jmpbuf macro on some platforms. - -Version 1.5.3beta05 [May 6, 2011] - Added the "_POSIX_SOURCE" feature test macro to ensure libpng sees the - correct API. _POSIX_SOURCE is defined in pngpriv.h, pngtest.c and - pngvalid.c to ensure that POSIX conformant systems disable non-POSIX APIs. - Removed png_snprintf and added formatted warning messages. This change adds - internal APIs to allow png_warning messages to have parameters without - requiring the host OS to implement snprintf. As a side effect the - dependency of the tIME-supporting RFC1132 code on stdio is removed and - PNG_NO_WARNINGS does actually work now. - Pass "" instead of '\0' to png_default_error() in png_err(). This mistake - was introduced in libpng-1.2.20beta01. This fixes CVE-2011-2691. - Added PNG_WRITE_OPTIMIZE_CMF_SUPPORTED macro to make the zlib "CMF" byte - optimization configurable. - IDAT compression failed if preceded by a compressed text chunk (bug - introduced in libpng-1.5.3beta01-02). This was because the attempt to - reset the zlib stream in png_write_IDAT happened after the first IDAT - chunk had been deflated - much too late. In this change internal - functions were added to claim/release the z_stream and, hopefully, make - the code more robust. Also deflateEnd checking is added - previously - libpng would ignore an error at the end of the stream. - -Version 1.5.3beta06 [May 8, 2011] - Removed the -D_ALL_SOURCE from definitions for AIX in CMakeLists.txt - Implemented premultiplied alpha support: png_set_alpha_mode API - -Version 1.5.3beta07 [May 11, 2011] - Added expand_16 support to the high level interface. - Added named value and 'flag' gamma support to png_set_gamma. Made a minor - change from the previous (unreleased) ABI/API to hide the exact value used - for Macs - it's not a good idea to embed this in the ABI! - Moved macro definitions for PNG_HAVE_IHDR, PNG_HAVE_PLTE, and PNG_AFTER_IDAT - from pngpriv.h to png.h because they must be visible to applications - that call png_set_unknown_chunks(). - Check for up->location !PNG_AFTER_IDAT when writing unknown chunks - before IDAT. - -Version 1.5.3beta08 [May 16, 2011] - Improved "pngvalid --speed" to exclude more of pngvalid from the time. - Documented png_set_alpha_mode(), other changes in libpng.3/libpng-manual.txt - The cHRM chunk now sets the defaults for png_set_rgb_to_gray() (when negative - parameters are supplied by the caller), while in the absence of cHRM - sRGB/Rec 709 values are still used. This introduced a divide-by-zero - bug in png_handle_cHRM(). - The bKGD chunk no longer overwrites the background value set by - png_set_background(), allowing the latter to be used before the file - header is read. It never performed any useful function to override - the default anyway. - Added memory overwrite and palette image checks to pngvalid.c - Previously palette image code was poorly checked. Since the transformation - code has a special palette path in most cases this was a severe weakness. - Minor cleanup and some extra checking in pngrutil.c and pngrtran.c. When - expanding an indexed image, always expand to RGBA if transparency is - present. - -Version 1.5.3beta09 [May 17, 2011] - Reversed earlier 1.5.3 change of transformation order; move png_expand_16 - back where it was. The change doesn't work because it requires 16-bit - gamma tables when the code only generates 8-bit ones. This fails - silently; the libpng code just doesn't do any gamma correction. Moving - the tests back leaves the old, inaccurate, 8-bit gamma calculations, but - these are clearly better than none! - -Version 1.5.3beta10 [May 20, 2011] - - png_set_background() and png_expand_16() did not work together correctly. - This problem is present in 1.5.2; if png_set_background is called with - need_expand false and the matching 16 bit color libpng erroneously just - treats it as an 8-bit color because of where png_do_expand_16 is in the - transform list. This simple fix reduces the supplied colour to 8-bits, - so it gets smashed, but this is better than the current behavior. - Added tests for expand16, more fixes for palette image tests to pngvalid. - Corrects the code for palette image tests and disables attempts to - validate palette colors. - -Version 1.5.3rc01 [June 3, 2011] - No changes. - -Version 1.5.3rc02 [June 8, 2011] - Fixed uninitialized memory read in png_format_buffer() (Bug report by - Frank Busse, CVE-2011-2501, related to CVE-2004-0421). - -Version 1.5.3beta11 [June 11, 2011] - Fixed png_handle_sCAL which is broken in 1.5. This fixes CVE 2011-2692. - Added sCAL to pngtest.png - Revised documentation about png_set_user_limits() to say that it also affects - png writing. - Revised handling of png_set_user_limits() so that it can increase the - limit beyond the PNG_USER_WIDTH|HEIGHT_MAX; previously it could only - reduce it. - Make the 16-to-8 scaling accurate. Dividing by 256 with no rounding is - wrong (high by one) 25% of the time. Dividing by 257 with rounding is - wrong in 128 out of 65536 cases. Getting the right answer all the time - without division is easy. - Added "_SUPPORTED" to the PNG_WRITE_CUSTOMIZE_ZTXT_COMPRESSION macro. - Added projects/owatcom, an IDE project for OpenWatcom to replace - scripts/makefile.watcom. This project works with OpenWatcom 1.9. The - IDE autogenerates appropriate makefiles (libpng.mk) for batch processing. - The project is configurable, unlike the Visual Studio project, so long - as the developer has an awk. - Changed png_set_gAMA to limit the gamma value range so that the inverse - of the stored value cannot overflow the fixed point representation, - and changed other things OpenWatcom warns about. - Revised pngvalid.c to test PNG_ALPHA_MODE_SUPPORTED correctly. This allows - pngvalid to build when ALPHA_MODE is not supported, which is required if - it is to build on libpng 1.4. - Removed string/memory macros that are no longer used and are not - necessarily fully supportable, particularly png_strncpy and png_snprintf. - Added log option to pngvalid.c and attempted to improve gamma messages. - -Version 1.5.3 [omitted] - People found the presence of a beta release following an rc release - to be confusing; therefore we bump the version to libpng-1.5.4beta01 - and there will be no libpng-1.5.3 release. - -Version 1.5.4beta01 [June 14, 2011] - Made it possible to undefine PNG_READ_16_TO_8_ACCURATE_SCALE_SUPPORTED - to get the same (inaccurate) output as libpng-1.5.2 and earlier. - Moved definitions of PNG_HAVE_IHDR, PNG_AFTER_IDAT, and PNG_HAVE_PLTE - outside of an unknown-chunk block in png.h because they are also - needed for other uses. - -Version 1.5.4beta02 [June 14, 2011] - Fixed and clarified LEGACY 16-to-8 scaling code. - Added png_set_chop_16() API, to match inaccurate results from previous - libpng versions. - Removed the ACCURATE and LEGACY options (they are no longer useable) - Use the old scaling method for background if png_set_chop_16() was - called. - Made png_set_chop_16() API removeable by disabling PNG_CHOP_16_TO_8_SUPPORTED - -Version 1.5.4beta03 [June 15, 2011] - Fixed a problem in png_do_expand_palette() exposed by optimization in - 1.5.3beta06 - Also removed a spurious and confusing "trans" member ("trans") from png_info. - The palette expand optimization prevented expansion to an intermediate RGBA - form if tRNS was present but alpha was marked to be stripped; this exposed - a check for tRNS in png_do_expand_palette() which is inconsistent with the - code elsewhere in libpng. - Correction to the expand_16 code; removed extra instance of - png_set_scale_16_to_8 from pngpriv.h - -Version 1.5.4beta04 [June 16, 2011] - Added a missing "#ifdef PNG_READ_BACKGROUND_SUPPORTED/#endif" in pngrtran.c - Added PNG_TRANSFORM_CHOP_16 to the high-level read transforms. - Made PNG_READ_16_TO_8_ACCURATE_SCALE configurable again. If this is - not enabled, png_set_strip_16() and png_do_scale_16_to_8() aren't built. - Revised contrib/visupng, gregbook, and pngminim to demonstrate chop_16_to_8 - -Version 1.5.4beta05 [June 16, 2011] - Renamed png_set_strip_16() to png_set_scale_16() and renamed - png_set_chop_16() to png_set_strip(16) in an attempt to minimize the - behavior changes between libpng14 and libpng15. - -Version 1.5.4beta06 [June 18, 2011] - Fixed new bug that was causing both strip_16 and scale_16 to be applied. - -Version 1.5.4beta07 [June 19, 2011] - Fixed pngvalid, simplified macros, added checking for 0 in sCAL. - The ACCURATE scale macro is no longer defined in 1.5 - call the - png_scale_16_to_8 API. Made sure that PNG_READ_16_TO_8 is still defined - if the png_strip_16_to_8 API is present. png_check_fp_number now - maintains some state so that positive, negative and zero values are - identified. sCAL uses these to be strictly spec conformant. - -Version 1.5.4beta08 [June 23, 2011] - Fixed pngvalid if ACCURATE_SCALE is defined. - Updated scripts/pnglibconf.h.prebuilt. - -Version 1.5.4rc01 [June 30, 2011] - Define PNG_ALLOCATED to "restrict" only if MSC_VER >= 1400. - -Version 1.5.4 [July 7, 2011] - No changes. - -Version 1.5.5beta01 [July 13, 2011] - Fixed some typos and made other minor changes in the manual. - Updated contrib/pngminus/makefile.std (Samuli Souminen) - -Version 1.5.5beta02 [July 14, 2011] - Revised Makefile.am and Makefile.in to look in the right directory for - pnglibconf.h.prebuilt - -Version 1.5.5beta03 [July 27, 2011] - Enabled compilation with g++ compiler. This compiler does not recognize - the file extension, so it always compiles with C++ rules. Made minor - changes to pngrutil.c to cast results where C++ expects it but C does not. - Minor editing of libpng.3 and libpng-manual.txt. - -Version 1.5.5beta04 [July 29, 2011] - Revised CMakeLists.txt (Clifford Yapp) - Updated commentary about the png_rgb_to_gray() default coefficients - in the manual and in pngrtran.c - -Version 1.5.5beta05 [August 17, 2011] - Prevent unexpected API exports from non-libpng DLLs on Windows. The "_DLL" - is removed from the test of whether a DLL is being built (this erroneously - caused the libpng APIs to be marked as DLL exports in static builds under - Microsoft Visual Studio). Almost all of the libpng building configuration - is moved from pngconf.h to pngpriv.h, but PNG_DLL_EXPORT remains in - pngconf.h, though, so that it is colocated with the import definition (it - is no longer used anywhere in the installed headers). The VStudio project - definitions have been cleaned up: "_USRDLL" has been removed from the - static library builds (this was incorrect), and PNG_USE_DLL has been added - to pngvalid to test the functionality (pngtest does not supply it, - deliberately). The spurious "_EXPORTS" has been removed from the - libpng build (all these errors were a result of copy/paste between project - configurations.) - Added new types and internal functions for CIE RGB end point handling to - pngpriv.h (functions yet to be implemented). - -Version 1.5.5beta06 [August 26, 2011] - Ensure the CMAKE_LIBRARY_OUTPUT_DIRECTORY is set in CMakeLists.txt - (Clifford Yap) - Fixes to rgb_to_gray and cHRM XYZ APIs (John Bowler): - The rgb_to_gray code had errors when combined with gamma correction. - Some pixels were treated as true grey when they weren't and such pixels - and true grey ones were not gamma corrected (the original value of the - red component was used instead). APIs to get and set cHRM using color - space end points have been added and the rgb_to_gray code that defaults - based on cHRM, and the divide-by-zero bug in png_handle_cHRM (CERT - VU#477046, CVE-2011-3328, introduced in 1.5.4) have been corrected. - A considerable number of tests has been added to pngvalid for the - rgb_to_gray transform. - Arithmetic errors in rgb_to_gray whereby the calculated gray value was - truncated to the bit depth rather than rounded have been fixed except in - the 8-bit non-gamma-corrected case (where consistency seems more important - than correctness.) The code still has considerable inaccuracies in the - 8-bit case because 8-bit linear arithmetic is used. - -Version 1.5.5beta07 [September 7, 2011] - Added "$(ARCH)" option to makefile.darwin - Added SunOS support to configure.ac and Makefile.am - Changed png_chunk_benign_error() to png_warning() in png.c, in - png_XYZ_from_xy_checked(). - -Version 1.5.5beta08 [September 10, 2011] - Fixed 64-bit compilation errors (gcc). The errors fixed relate - to conditions where types that are 32 bits in the GCC 32-bit - world (uLong and png_size_t) become 64 bits in the 64-bit - world. This produces potential truncation errors which the - compiler correctly flags. - Relocated new HAVE_SOLARIS_LD definition in configure.ac - Constant changes for 64-bit compatibility (removal of L suffixes). The - 16-bit cases still use "L" as we don't have a 16-bit test system. - -Version 1.5.5rc01 [September 15, 2011] - Removed "L" suffixes in pngpriv.h - -Version 1.5.5 [September 22, 2011] - No changes. - -Version 1.5.6beta01 [September 22, 2011] - Fixed some 64-bit type conversion warnings in pngrtran.c - Moved row_info from png_struct to a local variable. - The various interlace mask arrays have been made into arrays of - bytes and made PNG_CONST and static (previously some arrays were - marked PNG_CONST and some weren't). - Additional checks have been added to the transform code to validate the - pixel depths after the transforms on both read and write. - Removed some redundant code from pngwrite.c, in png_destroy_write_struct(). - Changed chunk reading/writing code to use png_uint_32 instead of png_byte[4]. - This removes the need to allocate temporary strings for chunk names on - the stack in the read/write code. Unknown chunk handling still uses the - string form because this is exposed in the API. - -Version 1.5.6beta02 [September 26, 2011] - Added a note in the manual the png_read_update_info() must be called only - once with a particular info_ptr. - Fixed a typo in the definition of the new PNG_STRING_FROM_CHUNK(s,c) macro. - -Version 1.5.6beta03 [September 28, 2011] - Revised test-pngtest.sh to report FAIL when pngtest fails. - Added "--strict" option to pngtest, to report FAIL when the failure is - only because the resulting valid files are different. - Revised CMakeLists.txt to work with mingw and removed some material from - CMakeLists.txt that is no longer useful in libpng-1.5. - -Version 1.5.6beta04 [October 5, 2011] - Fixed typo in Makefile.in and Makefile.am ("-M Wl" should be "-M -Wl")." - -Version 1.5.6beta05 [October 12, 2011] - Speed up png_combine_row() for interlaced images. This reduces the generality - of the code, allowing it to be optimized for Adam7 interlace. The masks - passed to png_combine_row() are now generated internally, avoiding - some code duplication and localizing the interlace handling somewhat. - Align png_struct::row_buf - previously it was always unaligned, caused by - a bug in the code that attempted to align it; the code needs to subtract - one from the pointer to take account of the filter byte prepended to - each row. - Optimized png_combine_row() when rows are aligned. This gains a small - percentage for 16-bit and 32-bit pixels in the typical case where the - output row buffers are appropriately aligned. The optimization was not - previously possible because the png_struct buffer was always misaligned. - Fixed bug in png_write_chunk_header() debug print, introduced in 1.5.6beta01. - -Version 1.5.6beta06 [October 17, 2011] - Removed two redundant tests for uninitialized row. - Fixed a relatively harmless memory overwrite in compressed text writing - with a 1 byte zlib buffer. - Add ability to call png_read_update_info multiple times to pngvalid.c. - Fixes for multiple calls to png_read_update_info. These fixes attend to - most of the errors revealed in pngvalid, however doing the gamma work - twice results in inaccuracies that can't be easily fixed. There is now - a warning in the code if this is going to happen. - Turned on multiple png_read_update_info in pngvalid transform tests. - Prevent libpng from overwriting unused bits at the end of the image when - it is not byte aligned, while reading. Prior to libpng-1.5.6 libpng would - overwrite the partial byte at the end of each row if the row width was not - an exact multiple of 8 bits and the image is not interlaced. - -Version 1.5.6beta07 [October 21, 2011] - Made png_ptr->prev_row an aligned pointer into png_ptr->big_prev_row - (Mans Rullgard). - -Version 1.5.6rc01 [October 26, 2011] - Changed misleading "Missing PLTE before cHRM" warning to "Out of place cHRM" - -Version 1.5.6rc02 [October 27, 2011] - Added LSR() macro to defend against buggy compilers that evaluate non-taken - code branches and complain about out-of-range shifts. - -Version 1.5.6rc03 [October 28, 2011] - Renamed the LSR() macro to PNG_LSR() and added PNG_LSL() macro. - Fixed compiler warnings with Intel and MSYS compilers. The logical shift - fix for Microsoft Visual C is required by other compilers, so this - enables that fix for all compilers when using compile-time constants. - Under MSYS 'byte' is a name declared in a system header file, so we - changed the name of a local variable to avoid the warnings that result. - Added #define PNG_ALIGN_TYPE PNG_ALIGN_NONE to contrib/pngminim/*/pngusr.h - -Version 1.5.6 [November 3, 2011] - No changes. - -Version 1.5.7beta01 [November 4, 2011] - Added support for ARM processor, when decoding all PNG up-filtered rows - and any other-filtered rows with 3 or 4 bytes per pixel (Mans Rullgard). - Fixed bug in pngvalid on early allocation failure; fixed type cast in - pngmem.c; pngvalid would attempt to call png_error() if the allocation - of a png_struct or png_info failed. This would probably have led to a - crash. The pngmem.c implementation of png_malloc() included a cast - to png_size_t which would fail on large allocations on 16-bit systems. - Fix for the preprocessor of the Intel C compiler. The preprocessor - splits adjacent @ signs with a space; this changes the concatenation - token from @-@-@ to PNG_JOIN; that should work with all compiler - preprocessors. - Paeth filter speed improvements from work by Siarhei Siamashka. This - changes the 'Paeth' reconstruction function to improve the GCC code - generation on x86. The changes are only part of the suggested ones; - just the changes that definitely improve speed and remain simple. - The changes also slightly increase the clarity of the code. - -Version 1.5.7beta02 [November 11, 2011] - Check compression_type parameter in png_get_iCCP and remove spurious - casts. The compression_type parameter is always assigned to, so must - be non-NULL. The cast of the profile length potentially truncated the - value unnecessarily on a 16-bit int system, so the cast of the (byte) - compression type to (int) is specified by ANSI-C anyway. - Fixed FP division by zero in pngvalid.c; the 'test_pixel' code left - the sBIT fields in the test pixel as 0, which resulted in a floating - point division by zero which was irrelevant but causes systems where - FP exceptions cause a crash. Added code to pngvalid to turn on FP - exceptions if the appropriate glibc support is there to ensure this is - tested in the future. - Updated scripts/pnglibconf.mak and scripts/makefile.std to handle the - new PNG_JOIN macro. - Added versioning to pnglibconf.h comments. - Simplified read/write API initial version; basic read/write tested on - a variety of images, limited documentation (in the header file.) - Installed more accurate linear to sRGB conversion tables. The slightly - modified tables reduce the number of 16-bit values that - convert to an off-by-one 8-bit value. The "makesRGB.c" code that was used - to generate the tables is now in a contrib/sRGBtables sub-directory. - -Version 1.5.7beta03 [November 17, 2011] - Removed PNG_CONST from the sRGB table declarations in pngpriv.h and png.c - Added run-time detection of NEON support. - Added contrib/libtests; includes simplified API test and timing test and - a color conversion utility for rapid checking of failed 'pngstest' results. - Multiple transform bug fixes plus a work-round for double gamma correction. - libpng does not support more than one transform that requires linear data - at once - if this is tried typically the results is double gamma - correction. Since the simplified APIs can need rgb to gray combined with - a compose operation it is necessary to do one of these outside the main - libpng transform code. This check-in also contains fixes to various bugs - in the simplified APIs themselves and to some bugs in compose and rgb to - gray (on palette) itself. - Fixes for C++ compilation using g++ When libpng source is compiled - using g++. The compiler imposes C++ rules on the C source; thus it - is desirable to make the source work with either C or C++ rules - without throwing away useful error information. This change adds - png_voidcast to allow C semantic (void*) cases or the corresponding - C++ static_cast operation, as appropriate. - Added --noexecstack to assembler file compilation. GCC does not set - this on assembler compilation, even though it does on C compilation. - This creates security issues if assembler code is enabled; the - work-around is to set it by default in the flags for $(CCAS) - Work around compilers that don't support declaration of const data. Some - compilers fault 'extern const' data declarations (because the data is - not initialized); this turns on const-ness only for compilers where - this is known to work. - -Version 1.5.7beta04 [November 17, 2011] - Since the gcc driver does not recognize the --noexecstack flag, we must - use the -Wa prefix to have it passed through to the assembler. - Also removed a duplicate setting of this flag. - Added files that were omitted from the libpng-1.5.7beta03 zip distribution. - -Version 1.5.7beta05 [November 25, 2011] - Removed "zTXt" from warning in generic chunk decompression function. - Validate time settings passed to png_set_tIME() and png_convert_to_rfc1123() - (Frank Busse). Note: This prevented CVE-2015-7981 from affecting - libpng-1.5.7 and later. - Added MINGW support to CMakeLists.txt - Reject invalid compression flag or method when reading the iTXt chunk. - Backed out 'simplified' API changes. The API seems too complex and there - is a lack of consensus or enthusiasm for the proposals. The API also - reveals significant bugs inside libpng (double gamma correction and the - known bug of being unable to retrieve a corrected palette). It seems - better to wait until the bugs, at least, are corrected. - Moved pngvalid.c into contrib/libtests - Rebuilt Makefile.in, configure, etc., with autoconf-2.68 - -Version 1.5.7rc01 [December 1, 2011] - Replaced an "#if" with "#ifdef" in pngrtran.c - Revised #if PNG_DO_BC block in png.c (use #ifdef and add #else) - -Version 1.5.7rc02 [December 5, 2011] - Revised project files and contrib/pngvalid/pngvalid.c to account for - the relocation of pngvalid into contrib/libtests. - Revised pngconf.h to use " __declspec(restrict)" only when MSC_VER >= 1400, - as in libpng-1.5.4. - Put CRLF line endings in the owatcom project files. - -Version 1.5.7rc03 [December 7, 2011] - Updated CMakeLists.txt to account for the relocation of pngvalid.c - -Version 1.5.7 [December 15, 2011] - Minor fixes to pngvalid.c for gcc 4.6.2 compatibility to remove warnings - reported by earlier versions. - Fixed minor memset/sizeof errors in pngvalid.c. - -Version 1.6.0beta01 [December 15, 2011] - Removed machine-generated configure files from the GIT repository (they will - continue to appear in the tarball distributions and in the libpng15 and - earlier GIT branches). - Restored the new 'simplified' API, which was started in libpng-1.5.7beta02 - but later deleted from libpng-1.5.7beta05. - Added example programs for the new 'simplified' API. - Added ANSI-C (C90) headers and require them, and take advantage of the - change. Also fixed some of the projects/* and contrib/* files that needed - updates for libpng16 and the move of pngvalid.c. - With this change the required ANSI-C header files are assumed to exist: the - implementation must provide float.h, limits.h, stdarg.h and stddef.h and - libpng relies on limits.h and stddef.h existing and behaving as defined - (the other two required headers aren't used). Non-ANSI systems that don't - have stddef.h or limits.h will have to provide an appropriate fake - containing the relevant types and #defines. - Dropped support for 16-bit platforms. The use of FAR/far has been eliminated - and the definition of png_alloc_size_t is now controlled by a flag so - that 'small size_t' systems can select it if necessary. Libpng 1.6 may - not currently work on such systems -- it seems likely that it will - ask 'malloc' for more than 65535 bytes with any image that has a - sufficiently large row size (rather than simply failing to read such - images). - New tools directory containing tools used to generate libpng code. - Fixed race conditions in parallel make builds. With higher degrees of - parallelism during 'make' the use of the same temporary file names such - as 'dfn*' can result in a race where a temporary file from one arm of the - build is deleted or overwritten in another arm. This changes the - temporary files for suffix rules to always use $* and ensures that the - non-suffix rules use unique file names. - -Version 1.6.0beta02 [December 21, 2011] - Correct configure builds where build and source directories are separate. - The include path of 'config.h' was erroneously made relative in pngvalid.c - in libpng 1.5.7. - -Version 1.6.0beta03 [December 22, 2011] - Start-up code size improvements, error handler flexibility. These changes - alter how the tricky allocation of the initial png_struct and png_info - structures are handled. png_info is now handled in pretty much the same - way as everything else, except that the allocations handle NULL return - silently. png_struct is changed in a similar way on allocation and on - deallocation a 'safety' error handler is put in place (which should never - be required). The error handler itself is changed to permit mismatches - in the application and libpng error buffer size; however, this means a - silent change to the API to return the jmp_buf if the size doesn't match - the size from the libpng compilation; libpng now allocates the memory and - this may fail. Overall these changes result in slight code size - reductions; however, this is a reduction in code that is always executed - so is particularly valuable. Overall on a 64-bit system the libpng DLL - decreases in code size by 1733 bytes. pngerror.o increases in size by - about 465 bytes because of the new functionality. - Added png_convert_to_rfc1123_buffer() and deprecated png_convert_to_rfc1123() - to avoid including a spurious buffer in the png_struct. - -Version 1.6.0beta04 [December 30, 2011] - Regenerated configure scripts with automake-1.11.2 - Eliminated png_info_destroy(). It is now used only in png.c and only calls - one other internal function and memset(). - Enabled png_get_sCAL_fixed() if floating point APIs are enabled. Previously - it was disabled whenever internal fixed point arithmetic was selected, - which meant it didn't exist even on systems where FP was available but not - preferred. - Added pngvalid.c compile time checks for const APIs. - Implemented 'restrict' for png_info and png_struct. Because of the way - libpng works both png_info and png_struct are always accessed via a - single pointer. This means adding C99 'restrict' to the pointer gives - the compiler some opportunity to optimize the code. This change allows - that. - Moved AC_MSG_CHECKING([if libraries can be versioned]) later to the proper - location in configure.ac (Gilles Espinasse). - Changed png_memcpy to C assignment where appropriate. Changed all those - uses of png_memcpy that were doing a simple assignment to assignments - (all those cases where the thing being copied is a non-array C L-value). - Added some error checking to png_set_*() routines. - Removed the reference to the non-exported function png_memcpy() from - example.c. - Fixed the Visual C 64-bit build - it requires jmp_buf to be aligned, but - it had become misaligned. - Revised contrib/pngminus/pnm2png.c to avoid warnings when png_uint_32 - and unsigned long are of different sizes. - -Version 1.6.0beta05 [January 15, 2012] - Updated manual with description of the simplified API (copied from png.h) - Fix bug in pngerror.c: some long warnings were being improperly truncated - (CVE-2011-3464, bug introduced in libpng-1.5.3beta05). - -Version 1.6.0beta06 [January 24, 2012] - Added palette support to the simplified APIs. This commit - changes some of the macro definitions in png.h, app code - may need corresponding changes. - Increased the formatted warning buffer to 192 bytes. - Added color-map support to simplified API. This is an initial version for - review; the documentation has not yet been updated. - Fixed Min/GW uninstall to remove libpng.dll.a - -Version 1.6.0beta07 [January 28, 2012] - Eliminated Intel icc/icl compiler warnings. The Intel (GCC derived) - compiler issues slightly different warnings from those issued by the - current versions of GCC. This eliminates those warnings by - adding/removing casts and small code rewrites. - Updated configure.ac from autoupdate: added --enable-werror option. - Also some layout regularization and removal of introduced tab characters - (replaced with 3-character indentation). Obsolete macros identified by - autoupdate have been removed; the replacements are all in 2.59 so - the pre-req hasn't been changed. --enable-werror checks for support - for -Werror (or the given argument) in the compiler. This mimics the - gcc configure option by allowing -Werror to be turned on safely; without - the option the tests written in configure itself fail compilation because - they cause compiler warnings. - Rewrote autogen.sh to run autoreconf instead of running tools one-by-one. - Conditionalize the install rules for MINGW and CYGWIN in CMakeLists.txt and - set CMAKE_LIBRARY_OUTPUT_DIRECTORY to "lib" on all platforms (C. Yapp). - Freeze libtool files in the 'scripts' directory. This version of autogen.sh - attempts to dissuade people from running it when it is not, or should not, - be necessary. In fact, autogen.sh does not work when run in a libpng - directory extracted from a tar distribution anymore. You must run it in - a GIT clone instead. - Added two images to contrib/pngsuite (1-bit and 2-bit transparent grayscale), - and renamed three whose names were inconsistent with those in - pngsuite/README.txt. - -Version 1.6.0beta08 [February 1, 2012] - Fixed Image::colormap misalignment in pngstest.c - Check libtool/libtoolize version number (2.4.2) in configure.ac - Divide test-pngstest.sh into separate pngstest runs for basic and - transparent images. - Moved automake options to AM_INIT_AUTOMAKE in configure.ac - Added color-tests, silent-rules (Not yet implemented in Makefile.am) and - version checking to configure.ac - Improved pngstest speed by not doing redundant tests and add const to - the background parameter of png_image_finish_read. The --background - option is now done automagically only when required, so that command-line - option no longer exists. - Cleaned up pngpriv.h to consistently declare all functions and data. - Also eliminated PNG_CONST_DATA, which is apparently not needed but we - can't be sure until it is gone. - Added symbol prefixing that allows all the libpng external symbols - to be prefixed (suggested by Reuben Hawkins). - Updated "ftbb*.png" list in the owatcom and vstudio projects. - Fixed 'prefix' builds on clean systems. The generation of pngprefix.h - should not require itself. - Updated INSTALL to explain that autogen.sh must be run in a GIT clone, - not in a libpng directory extracted from a tar distribution. - -Version 1.6.0beta09 [February 1, 2012] - Reverted the prebuilt configure files to libpng-1.6.0beta05 condition. - -Version 1.6.0beta10 [February 3, 2012] - Added Z_SOLO for zlib-1.2.6+ and correct pngstest tests - Updated list of test images in CMakeLists.txt - Updated the prebuilt configure files to current condition. - Revised INSTALL information about autogen.sh; it works in tar distributions. - -Version 1.6.0beta11 [February 16, 2012] - Fix character count in pngstest command in projects/owatcom/pngstest.tgt - Revised test-pngstest.sh to report PASS/FAIL for each image. - Updated documentation about the simplified API. - Corrected estimate of error in libpng png_set_rgb_to_gray API. The API is - extremely inaccurate for sRGB conversions because it uses an 8-bit - intermediate linear value and it does not use the sRGB transform, so it - suffers from the known instability in gamma transforms for values close - to 0 (see Poynton). The net result is that the calculation has a maximum - error of 14.99/255; 0.5/255^(1/2.2). pngstest now uses 15 for the - permitted 8-bit error. This may still not be enough because of arithmetic - error. - Removed some unused arrays (with #ifdef) from png_read_push_finish_row(). - Fixed a memory overwrite bug in simplified read of RGB PNG with - non-linear gamma Also bugs in the error checking in pngread.c and changed - quite a lot of the checks in pngstest.c to be correct; either correctly - written or not over-optimistic. The pngstest changes are insufficient to - allow all possible RGB transforms to be passed; pngstest cmppixel needs - to be rewritten to make it clearer which errors it allows and then changed - to permit known inaccuracies. - Removed tests for no-longer-used *_EMPTY_PLTE_SUPPORTED from pngstruct.h - Fixed fixed/float API export conditionals. 1) If FIXED_POINT or - FLOATING_POINT options were switched off, png.h ended up with lone ';' - characters. This is not valid ANSI-C outside a function. The ';' - characters have been moved inside the definition of PNG_FP_EXPORT and - PNG_FIXED_EXPORT. 2) If either option was switched off, the declaration - of the corresponding functions were completely omitted, even though some - of them are still used internally. The result is still valid, but - produces warnings from gcc with some warning options (including -Wall). The - fix is to cause png.h to declare the functions with PNG_INTERNAL_FUNCTION - when png.h is included from pngpriv.h. - Check for invalid palette index while reading paletted PNG. When one is - found, issue a warning and increase png_ptr->num_palette accordingly. - Apps are responsible for checking to see if that happened. - -Version 1.6.0beta12 [February 18, 2012] - Do not increase num_palette on invalid_index. - Relocated check for invalid palette index to pngrtran.c, after unpacking - the sub-8-bit pixels. - Fixed CVE-2011-3026 buffer overrun bug. This bug was introduced when - iCCP chunk support was added at libpng-1.0.6. Deal more correctly with the - test on iCCP chunk length. Also removed spurious casts that may hide - problems on 16-bit systems. - -Version 1.6.0beta13 [February 24, 2012] - Eliminated redundant png_push_read_tEXt|zTXt|iTXt|unknown code from - pngpread.c and use the sequential png_handle_tEXt, etc., in pngrutil.c; - now that png_ptr->buffer is inaccessible to applications, the special - handling is no longer useful. - Added PNG_SAFE_LIMITS feature to pnglibconf.dfa, pngpriv.h, and new - pngusr.dfa to reset the user limits to safe ones if PNG_SAFE_LIMITS is - defined. To enable, use "CPPFLAGS=-DPNG_SAFE_LIMITS_SUPPORTED=1" on the - configure command or put #define PNG_SAFE_LIMITS_SUPPORTED in - pnglibconf.h.prebuilt and pnglibconf.h. - -Version 1.6.0beta14 [February 27, 2012] - Added information about the new limits in the manual. - Updated Makefile.in - -Version 1.6.0beta15 [March 2, 2012] - Removed unused "current_text" members of png_struct and the png_free() - of png_ptr->current_text from pngread.c - Rewrote pngstest.c for substantial speed improvement. - Fixed transparent pixel and 16-bit rgb tests in pngstest and removed a - spurious check in pngwrite.c - Added PNG_IMAGE_FLAG_FAST for the benefit of applications that store - intermediate files, or intermediate in-memory data, while processing - image data with the simplified API. The option makes the files larger - but faster to write and read. pngstest now uses this by default; this - can be disabled with the --slow option. - Improved pngstest fine tuning of error numbers, new test file generator. - The generator generates images that test the full range of sample values, - allow the error numbers in pngstest to be tuned and checked. makepng - also allows generation of images with extra chunks, although this is - still work-in-progress. - Added check for invalid palette index while reading. - Fixed some bugs in ICC profile writing. The code should now accept - all potentially valid ICC profiles and reject obviously invalid ones. - It now uses png_error() to do so rather than casually writing a PNG - without the necessary color data. - Removed whitespace from the end of lines in all source files and scripts. - -Version 1.6.0beta16 [March 6, 2012] - Relocated palette-index checking function from pngrutil.c to pngtrans.c - Added palette-index checking while writing. - Changed png_inflate() and calling routines to avoid overflow problems. - This is an intermediate check-in that solves the immediate problems and - introduces one performance improvement (avoiding a copy via png_ptr->zbuf.) - Further changes will be made to make ICC profile handling more secure. - Fixed build warnings (MSVC, GCC, GCC v3). Cygwin GCC with default options - declares 'index' as a global, causing a warning if it is used as a local - variable. GCC 64-bit warns about assigning a (size_t) (unsigned 64-bit) - to an (int) (signed 32-bit). MSVC, however, warns about using the - unary '-' operator on an unsigned value (even though it is well defined - by ANSI-C to be ~x+1). The padding calculation was changed to use a - different method. Removed the tests on png_ptr->pass. - Added contrib/libtests/tarith.c to test internal arithmetic functions from - png.c. This is a libpng maintainer program used to validate changes to the - internal arithmetic functions. - Made read 'inflate' handling like write 'deflate' handling. The read - code now claims and releases png_ptr->zstream, like the write code. - The bug whereby the progressive reader failed to release the zstream - is now fixed, all initialization is delayed, and the code checks for - changed parameters on deflate rather than always calling - deflatedEnd/deflateInit. - Validate the zTXt strings in pngvalid. - Added code to validate the windowBits value passed to deflateInit2(). - If the call to deflateInit2() is wrong a png_warning will be issued - (in fact this is harmless, but the PNG data produced may be sub-optimal). - -Version 1.6.0beta17 [March 10, 2012] - Fixed PNG_LIBPNG_BUILD_BASE_TYPE definition. - Reject all iCCP chunks after the first, even if the first one is invalid. - Deflate/inflate was reworked to move common zlib calls into single - functions [rw]util.c. A new shared keyword check routine was also added - and the 'zbuf' is no longer allocated on progressive read. It is now - possible to call png_inflate() incrementally. A warning is no longer - issued if the language tag or translated keyword in the iTXt chunk - has zero length. - If benign errors are disabled use maximum window on ancillary inflate. - This works round a bug introduced in 1.5.4 where compressed ancillary - chunks could end up with a too-small windowBits value in the deflate - header. - -Version 1.6.0beta18 [March 16, 2012] - Issue a png_benign_error() instead of png_warning() about bad palette index. - In pngtest, treat benign errors as errors if "-strict" is present. - Fixed an off-by-one error in the palette index checking function. - Fixed a compiler warning under Cygwin (Windows-7, 32-bit system) - Revised example.c to put text strings in a temporary character array - instead of directly assigning string constants to png_textp members. - This avoids compiler warnings when -Wwrite-strings is enabled. - Added output flushing to aid debugging under Visual Studio. Unfortunately - this is necessary because the VS2010 output window otherwise simply loses - the error messages on error (they weren't flushed to the window before - the process exited, apparently!) - Added configuration support for benign errors and changed the read - default. Also changed some warnings in the iCCP and sRGB handling - from to benign errors. Configuration now makes read benign - errors warnings and write benign errors to errors by default (thus - changing the behavior on read). The simplified API always forces - read benign errors to warnings (regardless of the system default, unless - this is disabled in which case the simplified API can't be built.) - -Version 1.6.0beta19 [March 18, 2012] - Work around for duplicate row start calls; added warning messages. - This turns on PNG_FLAG_DETECT_UNINITIALIZED to detect app code that - fails to call one of the 'start' routines (not enabled in libpng-1.5 - because it is technically an API change, since it did normally work - before.) It also makes duplicate calls to png_read_start_row (an - internal function called at the start of the image read) benign, as - they were before changes to use png_inflate_claim. Somehow webkit is - causing this to happen; this is probably a mis-feature in the zlib - changes so this commit is only a work-round. - Removed erroneous setting of DETECT_UNINITIALIZED and added more - checks. The code now does a png_error if an attempt is made to do the - row initialization twice; this is an application error and it has - serious consequences because the transform data in png_struct is - changed by each call. - Added application error reporting and added chunk names to read - benign errors; also added --strict to pngstest - not enabled - yet because a warning is produced. - Avoid the double gamma correction warning in the simplified API. - This allows the --strict option to pass in the pngstest checks - -Version 1.6.0beta20 [March 29, 2012] - Changed chunk handler warnings into benign errors, incrementally load iCCP - Added checksum-icc.c to contrib/tools - Prevent PNG_EXPAND+PNG_SHIFT doing the shift twice. - Recognize known sRGB ICC profiles while reading; prefer writing the - iCCP profile over writing the sRGB chunk, controlled by the - PNG_sRGB_PROFILE_CHECKS option. - Revised png_set_text_2() to avoid potential memory corruption (fixes - CVE-2011-3048, also known as CVE-2012-3425). - -Version 1.6.0beta21 [April 27, 2012] - Revised scripts/makefile.darwin: use system zlib; remove quotes around - architecture list; add missing ppc architecture; add architecture options - to shared library link; don't try to create a shared lib based on missing - RELEASE variable. - Enable png_set_check_for_invalid_index() for both read and write. - Removed #ifdef PNG_HANDLE_AS_UNKNOWN_SUPPORTED in pngpriv.h around - declaration of png_handle_unknown(). - Added -lssp_nonshared in a comment in scripts/makefile.freebsd - and changed deprecated NOOBJ and NOPROFILE to NO_OBJ and NO_PROFILE. - -Version 1.6.0beta22 [May 23, 2012] - Removed need for -Wno-cast-align with clang. clang correctly warns on - alignment increasing pointer casts when -Wcast-align is passed. This - fixes the cases that clang warns about either by eliminating the - casts from png_bytep to png_uint_16p (pngread.c), or, for pngrutil.c - where the cast is previously verified or pngstest.c where it is OK, by - introducing new png_aligncast macros to do the cast in a way that clang - accepts. - -Version 1.6.0beta23 [June 6, 2012] - Revised CMakeLists.txt to not attempt to make a symlink under mingw. - Made fixes for new optimization warnings from gcc 4.7.0. The compiler - performs an optimization which is safe; however it then warns about it. - Changing the type of 'palette_number' in pngvalid.c removes the warning. - Do not depend upon a GCC feature macro being available for use in generating - the linker mapfile symbol prefix. - Improved performance of new do_check_palette_indexes() function (only - update the value when it actually increases, move test for whether - the check is wanted out of the function. - -Version 1.6.0beta24 [June 7, 2012] - Don't check palette indexes if num_palette is 0 (as it can be in MNG files). - -Version 1.6.0beta25 [June 16, 2012] - Revised png_set_keep_unknown_chunks() so num_chunks < 0 means ignore all - unknown chunks and all known chunks except for IHDR, PLTE, tRNS, IDAT, - and IEND. Previously it only meant ignore all unknown chunks, the - same as num_chunks == 0. Revised png_image_skip_unused_chunks() to - provide a list of chunks to be processed instead of a list of chunks to - ignore. Revised contrib/gregbook/readpng2.c accordingly. - -Version 1.6.0beta26 [July 10, 2012] - Removed scripts/makefile.cegcc from the *.zip and *.7z distributions; it - depends on configure, which is not included in those archives. - Moved scripts/chkfmt to contrib/tools. - Changed "a+w" to "u+w" in Makefile.in to fix CVE-2012-3386. - -Version 1.6.0beta27 [August 11, 2012] - Do not compile PNG_DEPRECATED, PNG_ALLOC and PNG_PRIVATE when __GNUC__ < 3. - Do not use __restrict when GNUC is <= 3.1 - Removed references to png_zalloc() and png_zfree() from the manual. - Fixed configurations where floating point is completely disabled. Because - of the changes to support symbol prefixing PNG_INTERNAL_FUNCTION declares - floating point APIs during libpng builds even if they are completely - disabled. This requires the png floating point types (png_double*) to be - declared even though the functions are never actually defined. This - change provides a dummy definition so that the declarations work, yet any - implementation will fail to compile because of an incomplete type. - Re-eliminated the use of strcpy() in pngtest.c. An unnecessary use of - strcpy() was accidentally re-introduced in libpng16; this change replaces - it with strncpy(). - Eliminated use of png_sizeof(); use sizeof() instead. - Use a consistent style for (sizeof type) and (sizeof (array)) - Cleanup of png_set_filler(). This function does very different things on - read and write. In libpng 1.6 the two cases can be distinguished and - considerable code cleanup, and extra error checking, is possible. This - makes calls on the write side that have no effect be ignored with a - png_app_error(), which can be disabled in the app using - png_set_benign_errors(), and removes the spurious use of usr_channels - on the read side. - Insist on autotools 1.12.1 for git builds because there are security issues - with 1.12 and insisting on anything less would allow 1.12 to be used. - Removed info_ptr->signature[8] from WRITE-only builds. - Add some conditions for compiling png_fixed(). This is a small function - but it requires "-lm" on some platforms. - Cause pngtest --strict to fail on any warning from libpng (not just errors) - and cause it not to fail at the comparison step if libpng lacks support - for writing chunks that it reads from the input (currently only implemented - for compressed text chunks). - Make all three "make check" test programs work without READ or WRITE support. - Now "make check" will succeed even if libpng is compiled with -DPNG_NO_READ - or -DPNG_NO_WRITE. The tests performed are reduced, but the basic reading - and writing of a PNG file is always tested by one or more of the tests. - Consistently use strlen(), memset(), memcpy(), and memcmp() instead of the - png_strlen(), png_memset(), png_memcpy(), and png_memcmp() macros. - Removed the png_sizeof(), png_strlen(), png_memset(), png_memcpy(), and - png_memcmp() macros. - Work around gcc 3.x and Microsoft Visual Studio 2010 complaints. Both object - to the split initialization of num_chunks. - -Version 1.6.0beta28 [August 29, 2012] - Unknown handling fixes and clean up. This adds more correct option - control of the unknown handling, corrects the pre-existing bug where - the per-chunk 'keep' setting is ignored and makes it possible to skip - IDAT chunks in the sequential reader (broken in earlier 1.6 versions). - There is a new test program, test-unknown.c, which is a work in progress - (not currently part of the test suite). Comments in the header files now - explain how the unknown handling works. - Allow fine grain control of unknown chunk APIs. This change allows - png_set_keep_unknown_chunks() to be turned off if not required and causes - both read and write to behave appropriately (on read this is only possible - if the user callback is used to handle unknown chunks). The change - also removes the support for storing unknown chunks in the info_struct - if the only unknown handling enabled is via the callback, allowing libpng - to be configured with callback reading and none of the unnecessary code. - Corrected fix for unknown handling in pngtest. This reinstates the - libpng handling of unknown chunks other than vpAg and sTER (including - unsafe-to-copy chunks which were dropped before) and eliminates the - repositioning of vpAg and sTER in pngtest.png by changing pngtest.png - (so the chunks are where libpng would put them). - Added "tunknown" test and corrected a logic error in png_handle_unknown() - when SAVE support is absent. Moved the shell test scripts for - contrib/libtests from the libpng top directory to contrib/libtests. - png_handle_unknown() must always read or skip the chunk, if - SAVE_UNKNOWN_CHUNKS is turned off *and* the application does not set - a user callback an unknown chunk will not be read, leading to a read - error, which was revealed by the "tunknown" test. - Cleaned up and corrected ICC profile handling. - contrib/libtests/makepng: corrected 'rgb' and 'gray' cases. profile_error - messages could be truncated; made a correct buffer size calculation and - adjusted pngerror.c appropriately. png_icc_check_* checking improved; - changed the functions to receive the correct color type of the PNG on read - or write and check that it matches the color space of the profile (despite - what the comments said before, there is danger in assuming the app will - cope correctly with an RGB profile on a grayscale image and, since it - violates the PNG spec, allowing it is certain to produce inconsistent - app behavior and might even cause app crashes.) Check that profiles - contain the tags needed to process the PNG (tags all required by the ICC - spec). Removed unused PNG_STATIC from pngpriv.h. - -Version 1.6.0beta29 [September 4, 2012] - Fixed the simplified API example programs to add the *colormap parameter - to several of he API and improved the error message if the version field - is not set. - Added contrib/examples/* to the *.zip and *.7z distributions. - Updated simplified API synopses and description of the png_image structure - in the manual. - Made makepng and pngtest produce identical PNGs, add "--relaxed" option - to pngtest. The "--relaxed" option turns off the benign errors that are - enabled by default in pre-RC builds. makepng can now write ICC profiles - where the length has not been extended to a multiple of 4, and pngtest - now intercepts all libpng errors, allowing the previously-introduced - "--strict test" on no warnings to actually work. - Improved ICC profile handling including cHRM chunk generation and fixed - Cygwin+MSVC build errors. The ICC profile handling now includes more - checking. Several errors that caused rejection of the profile are now - handled with a warning in such a way that the invalid profiles will be - read by default in release (but not pre-RC) builds but will not be - written by default. The easy part of handling the cHRM chunk is written, - where the ICC profile contains the required data. The more difficult - part plus guessing a gAMA value requires code to pass selected RGB values - through the profile. - -Version 1.6.0beta30 [October 24, 2012] - Changed ICC profile matrix/vector types to not depend on array type rules. - By the ANSI-C standard the new types should be identical to the previous - versions, and all known versions of gcc tested with the previous versions - except for GCC-4.2.1 work with this version. The change makes the ANSI-C - rule that const applied to an array of elements applies instead to the - elements in the array moot by explicitly applying const to the base - elements of the png_icc_matrix and png_icc_vector types. The accidental - (harmless) 'const' previously applied to the parameters of two of the - functions have also been removed. - Added a work around for GCC 4.2 optimization bug. - Marked the broken (bad white point) original HP sRGB profiles correctly and - correct comments. - Added -DZ_SOLO to contrib/pngminim/*/makefile to work with zlib-1.2.7 - Use /MDd for vstudio debug builds. Also added pngunkown to the vstudio - builds, fixed build errors and corrected a minor exit code error in - pngvalid if the 'touch' file name is invalid. - Add updated WARNING file to projects/vstudio from libpng 1.5/vstudio - Fixed build when using #define PNG_NO_READ_GAMMA in png_do_compose() in - pngrtran.c (Domani Hannes). - -Version 1.6.0beta31 [November 1, 2012] - Undid the erroneous change to vstudio/pngvalid build in libpng-1.6.0beta30. - Made pngvalid so that it will build outside the libpng source tree. - Made builds -DPNG_NO_READ_GAMMA compile (the unit tests still fail). - Made PNG_NO_READ_GAMMA switch off interfaces that depend on READ_GAMMA. - Prior to 1.6.0 switching off READ_GAMMA did unpredictable things to the - interfaces that use it (specifically, png_do_background in 1.4 would - simply display composite for grayscale images but do composition - with the incorrect arithmetic for color ones). In 1.6 the semantic - of -DPNG_NO_READ_GAMMA is changed to simply disable any interface that - depends on it; this obliges people who set it to consider whether they - really want it off if they happen to use any of the interfaces in - question (typically most users who disable it won't). - Fixed GUIDs in projects/vstudio. Some were duplicated or missing, - resulting in VS2010 having to update the files. - Removed non-working ICC profile support code that was mostly added to - libpng-1.6.0beta29 and beta30. There was too much code for too little - gain; implementing full ICC color correction may be desirable but is left - up to applications. - -Version 1.6.0beta32 [November 25, 2012] - Fixed an intermittent SEGV in pngstest due to an uninitialized array element. - Added the ability for contrib/libtests/makepng.c to make a PNG with just one - color. This is useful for debugging pngstest color inaccuracy reports. - Fixed error checking in the simplified write API (Olaf van der Spek) - Made png_user_version_check() ok to use with libpng version 1.10.x and later. - -Version 1.6.0beta33 [December 15, 2012] - Fixed typo in png.c (PNG_SET_CHUNK_MALLOC_MAX should be PNG_CHUNK_MALLOC_MAX) - that causes the MALLOC_MAX limit not to work (John Bowler) - Change png_warning() to png_app_error() in pngwrite.c and comment the - fall-through condition. - Change png_warning() to png_app_warning() in png_write_tRNS(). - Rearranged the ARM-NEON optimizations: Isolated the machine specific code - to the hardware subdirectory and added comments to pngrutil.c so that - implementors of other optimizations know what to do. - Fixed cases of unquoted DESTDIR in Makefile.am - Rebuilt Makefile.in, etc., with autoconf-2.69 and automake-1.12.5. - -Version 1.6.0beta34 [December 19, 2012] - Cleaned up whitespace in the synopsis portion of the manpage "libpng.3" - Disassembled the version number in scripts/options.awk (necessary for - building on SunOs). - -Version 1.6.0beta35 [December 23, 2012] - Made default Zlib compression settings be configurable. This adds #defines to - pnglibconf.h to control the defaults. - Fixed Windows build issues, enabled ARM compilation. Various warnings issued - by earlier versions of GCC fixed for Cygwin and Min/GW (which both use old - GCCs.) ARM support is enabled by default in zlib.props (unsupported by - Microsoft) and ARM compilation is made possible by deleting the check for - x86. The test programs cannot be run because they are not signed. - -Version 1.6.0beta36 [January 2, 2013] - Discontinued distributing libpng-1.x.x.tar.bz2. - Discontinued distributing libpng-1.7.0-1.6.0-diff.txt and similar. - Rebuilt configure with autoconf-2.69 (inadvertently not done in beta33) - Fixed 'make distcheck' on SUN OS - libpng.so was not being removed - -Version 1.6.0beta37 [January 10, 2013] - Fixed conceivable but difficult to repro overflow. Also added two test - programs to generate and test a PNG which should have the problem. - -Version 1.6.0beta39 [January 19, 2013] - Again corrected attempt at overflow detection in png_set_unknown_chunks() - (CVE-2013-7353). Added overflow detection in png_set_sPLT() and - png_set_text_2() (CVE-2013-7354). - -Version 1.6.0beta40 [January 20, 2013] - Use consistent handling of overflows in text, sPLT and unknown png_set_* APIs - -Version 1.6.0rc01 [January 26, 2013] - No changes. - -Version 1.6.0rc02 [February 4, 2013] - Added png_get_palette_max() function. - -Version 1.6.0rc03 [February 5, 2013] - Fixed the png_get_palette_max API. - -Version 1.6.0rc04 [February 7, 2013] - Turn serial tests back on (recently turned off by autotools upgrade). - -Version 1.6.0rc05 [February 8, 2013] - Update manual about png_get_palette_max(). - -Version 1.6.0rc06 [February 9, 2013] - Fixed missing dependency in --prefix builds The intermediate - internal 'prefix.h' file can only be generated correctly after - pnglibconf.h, however the dependency was not in Makefile.am. The - symptoms are unpredictable depending on the order make chooses to - build pngprefix.h and pnglibconf.h, often the error goes unnoticed - because there is a system pnglibconf.h to use instead. - -Version 1.6.0rc07 [February 10, 2013] - Enclosed the new png_get_palette_max in #ifdef PNG_GET_PALETTE_MAX_SUPPORTED - block, and revised pnglibconf.h and pnglibconf.h.prebuilt accordingly. - -Version 1.6.0rc08 [February 10, 2013] - Fix typo in png.h #ifdef - -Version 1.6.0 [February 14, 2013] - No changes. - -Version 1.6.1beta01 [February 16, 2013] - Made symbol prefixing work with the ARM neon optimizations. Also allow - pngpriv.h to be included for preprocessor definitions only, so it can - be used in non-C/C++ files. Back ported from libpng 1.7. - Made sRGB check numbers consistent. - Ported libpng 1.5 options.awk/dfn file handling to 1.6, fixed one bug. - Removed cc -E workround, corrected png_get_palette_max API Tested on - SUN OS cc 5.9, which demonstrates the tokenization problem previously - avoided by using /lib/cpp. Since all .dfn output is now protected in - double quotes unless it is to be macro substituted the fix should - work everywhere. - Enabled parallel tests - back ported from libpng-1.7. - scripts/pnglibconf.dfa formatting improvements back ported from libpng17. - Fixed a race condition in the creation of the build 'scripts' directory - while building with a parallel make. - Use approved/supported Android method to check for NEON, use Linux/POSIX - 1003.1 API to check /proc/self/auxv avoiding buffer allocation and other - library calls (ported from libpng15). - -Version 1.6.1beta02 [February 19, 2013] - Use parentheses more consistently in "#if defined(MACRO)" tests. - Folded long lines. - Reenabled code to allow zero length PLTE chunks for MNG. - -Version 1.6.1beta03 [February 22, 2013] - Fixed ALIGNED_MEMORY support. - Added a new configure option: - --enable-arm-neon=always will stop the run-time checks. New checks - within arm/arm_init.c will cause the code not to be compiled unless - __ARM_NEON__ is set. This should make it fail safe (if someone asks - for it on then the build will fail if it can't be done.) - Updated the INSTALL document. - -Version 1.6.1beta04 [February 27, 2013] - Revised INSTALL to recommend using CPPFLAGS instead of INCLUDES. - Revised scripts/makefile.freebsd to respect ZLIBLIB and ZLIBINC. - Revised scripts/dfn.awk to work with the buggy MSYS awk that has trouble - with CRLF line endings. - -Version 1.6.1beta05 [March 1, 2013] - Avoid a possible memory leak in contrib/gregbook/readpng.c - -Version 1.6.1beta06 [March 4, 2013] - Better documentation of unknown handling API interactions. - Corrected Android builds and corrected libpng.vers with symbol - prefixing. It also makes those tests compile and link on Android. - Added an API png_set_option() to set optimization options externally, - providing an alternative and general solution for the non-portable - run-time tests used by the ARM Neon code, using the PNG_ARM_NEON option. - The order of settings vs options in pnglibconf.h is reversed to allow - settings to depend on options and options can now set (or override) the - defaults for settings. - -Version 1.6.1beta07 [March 7, 2013] - Corrected simplified API default gamma for color-mapped output, added - a flag to change default. In 1.6.0 when the simplified API was used - to produce color-mapped output from an input image with no gamma - information the gamma assumed for the input could be different from - that assumed for non-color-mapped output. In particular 16-bit depth - input files were assumed to be sRGB encoded, whereas in the 'direct' - case they were assumed to have linear data. This was an error. The - fix makes the simplified API treat all input files the same way and - adds a new flag to the png_image::flags member to allow the - application/user to specify that 16-bit files contain sRGB data - rather than the default linear. - Fixed bugs in the pngpixel and makepng test programs. - -Version 1.6.1beta08 [March 7, 2013] - Fixed CMakelists.txt to allow building a single variant of the library - (Claudio Bley): - Introduced a PNG_LIB_TARGETS variable that lists all activated library - targets. It is an error if this variable ends up empty, ie. you have - to build at least one library variant. - Made the *_COPY targets only depend on library targets actually being build. - Use PNG_LIB_TARGETS to unify a code path. - Changed the CREATE_SYMLINK macro to expect the full path to a file as the - first argument. When symlinking the filename component of that path is - determined and used as the link target. - Use copy_if_different in the CREATE_SYMLINK macro. - -Version 1.6.1beta09 [March 13, 2013] - Eliminated two warnings from the Intel C compiler. The warnings are - technically valid, although a reasonable treatment of division would - show it to be incorrect. - -Version 1.6.1rc01 [March 21, 2013] - No changes. - -Version 1.6.1 [March 28, 2013] - No changes. - -Version 1.6.2beta01 [April 14, 2013] - Updated documentation of 1.5.x to 1.6.x changes in iCCP chunk handling. - Fixed incorrect warning of excess deflate data. End condition - the - warning would be produced if the end of the deflate stream wasn't read - in the last row. The warning is harmless. - Corrected the test on user transform changes on read. It was in the - png_set of the transform function, but that doesn't matter unless the - transform function changes the rowbuf size, and that is only valid if - transform_info is called. - Corrected a misplaced closing bracket in contrib/libtests/pngvalid.c - (Flavio Medeiros). - Corrected length written to uncompressed iTXt chunks (Samuli Suominen). - Bug was introduced in libpng-1.6.0. - -Version 1.6.2rc01 [April 18, 2013] - Added contrib/tools/fixitxt.c, to repair the erroneous iTXt chunk length - written by libpng-1.6.0 and 1.6.1. - Disallow storing sRGB information when the sRGB is not supported. - -Version 1.6.2rc02 [April 18, 2013] - Merge pngtest.c with libpng-1.7.0 - -Version 1.6.2rc03 [April 22, 2013] - Trivial spelling cleanup. - -Version 1.6.2rc04 and 1.6.2rc05 [omitted] - -Version 1.6.2rc06 [April 24, 2013] - Reverted to version 1.6.2rc03. Recent changes to arm/neon support - have been ported to libpng-1.7.0beta09 and will reappear in version - 1.6.3beta01. - -Version 1.6.2 [April 25, 2013] - No changes. - -Version 1.6.3beta01 [April 25, 2013] - Revised stack marking in arm/filter_neon.S and configure.ac. - Ensure that NEON filter stuff is completely disabled when switched 'off'. - Previously the ARM NEON specific files were still built if the option - was switched 'off' as opposed to being explicitly disabled. - -Version 1.6.3beta02 [April 26, 2013] - Test for 'arm*' not just 'arm' in the host_cpu configure variable. - Rebuilt the configure scripts. - -Version 1.6.3beta03 [April 30, 2013] - Expanded manual paragraph about writing private chunks, particularly - the need to call png_set_keep_unknown_chunks() when writing them. - Avoid dereferencing NULL pointer possibly returned from - png_create_write_struct() (Andrew Church). - -Version 1.6.3beta05 [May 9, 2013] - Calculate our own zlib windowBits when decoding rather than trusting the - CMF bytes in the PNG datastream. - Added an option to force maximum window size for inflating, which was - the behavior of libpng15 and earlier, via a new PNG_MAXIMUM_INFLATE_WINDOW - option for png_set_options(). - Added png-fix-itxt and png-fix-too-far-back to the built programs and - removed warnings from the source code and timepng that are revealed as - a result. - Detect wrong libpng versions linked to png-fix-too-far-back, which currently - only works with libpng versions that can be made to reliably fail when - the deflate data contains an out-of-window reference. This means only - 1.6 and later. - Fixed gnu issues: g++ needs a static_cast, gcc 4.4.7 has a broken warning - message which it is easier to work round than ignore. - Updated contrib/pngminus/pnm2png.c (Paul Stewart): - Check for EOF - Ignore "#" delimited comments in input file to pnm2png.c. - Fixed whitespace handling - Added a call to png_set_packing() - Initialize dimension values so if sscanf fails at least we have known - invalid values. - Attempt to detect configuration issues with png-fix-too-far-back, which - requires both the correct libpng and the correct zlib to function - correctly. - Check ZLIB_VERNUM for mismatches, enclose #error in quotes - Added information in the documentation about problems with and fixes for - the bad CRC and bad iTXt chunk situations. - -Version 1.6.3beta06 [May 12, 2013] - Allow contrib/pngminus/pnm2png.c to compile without WRITE_INVERT and - WRITE_PACK supported (writes error message that it can't read P1 or - P4 PBM files). - Improved png-fix-too-far-back usage message, added --suffix option. - Revised contrib/pngminim/*/makefile to generate pnglibconf.h with the - right zlib header files. - Separated CPPFLAGS and CFLAGS in contrib/pngminim/*/makefile - -Version 1.6.3beta07 [June 8, 2013] - Removed a redundant test in png_set_IHDR(). - Added set(CMAKE_CONFIGURATION_TYPES ...) to CMakeLists.txt (Andrew Hundt) - Deleted set(CMAKE_BUILD_TYPE) block from CMakeLists.txt - Enclose the prototypes for the simplified write API in - #ifdef PNG_STDIO_SUPPORTED/#endif - Make ARM NEON support work at compile time (not just configure time). - This moves the test on __ARM_NEON__ into pngconf.h to avoid issues when - using a compiler that compiles for multiple architectures at one time. - Removed PNG_FILTER_OPTIMIZATIONS and PNG_ARM_NEON_SUPPORTED from - pnglibconf.h, allowing more of the decisions to be made internally - (pngpriv.h) during the compile. Without this, symbol prefixing is broken - under certain circumstances on ARM platforms. Now only the API parts of - the optimizations ('check' vs 'api') are exposed in the public header files - except that the new setting PNG_ARM_NEON_OPT documents how libpng makes the - decision about whether or not to use the optimizations. - Protect symbol prefixing against CC/CPPFLAGS/CFLAGS usage. - Previous iOS/Xcode fixes for the ARM NEON optimizations moved the test - on __ARM_NEON__ from configure time to compile time. This breaks symbol - prefixing because the definition of the special png_init_filter_functions - call was hidden at configure time if the relevant compiler arguments are - passed in CFLAGS as opposed to CC. This change attempts to avoid all - the confusion that would result by declaring the init function even when - it is not used, so that it will always get prefixed. - -Version 1.6.3beta08 [June 18, 2013] - Revised libpng.3 so that "doclifter" can process it. - -Version 1.6.3beta09 [June 27, 2013] - Revised example.c to illustrate use of PNG_DEFAULT_sRGB and PNG_GAMMA_MAC_18 - as parameters for png_set_gamma(). These have been available since - libpng-1.5.4. - Renamed contrib/tools/png-fix-too-far-back.c to pngfix.c and revised it - to check all compressed chunks known to libpng. - -Version 1.6.3beta10 [July 5, 2013] - Updated documentation to show default behavior of benign errors correctly. - Only compile ARM code when PNG_READ_SUPPORTED is defined. - Fixed undefined behavior in contrib/tools/pngfix.c and added new strip - option. pngfix relied on undefined behavior and even a simple change from - gcc to g++ caused it to fail. The new strip option 'unsafe' has been - implemented and is the default if --max is given. Option names have - been clarified, with --strip=transform now stripping the bKGD chunk, - which was stripped previously with --strip=unused. - Added all documented chunk types to pngpriv.h - Unified pngfix.c source with libpng17. - -Version 1.6.3rc01 [July 11, 2013] - No changes. - -Version 1.6.3 [July 18, 2013] - Revised manual about changes in iTXt chunk handling made in libpng-1.6.0. - Added "/* SAFE */" comments in pngrutil.c and pngrtran.c where warnings - may be erroneously issued by code-checking applications. - -Version 1.6.4beta01 [August 21, 2013] - Added information about png_set_options() to the manual. - Delay calling png_init_filter_functions() until a row with nonzero filter - is found. - -Version 1.6.4beta02 [August 30, 2013] - Fixed inconsistent conditional compilation of png_chunk_unknown_handling() - prototype, definition, and usage. Made it depend on - PNG_HANDLE_AS_UNKNOWN_SUPPORTED everywhere. - -Version 1.6.4rc01 [September 5, 2013] - No changes. - -Version 1.6.4 [September 12, 2013] - No changes. - -Version 1.6.5 [September 14, 2013] - Removed two stray lines of code from arm/arm_init.c. - -Version 1.6.6 [September 16, 2013] - Removed two stray lines of code from arm/arm_init.c, again. - -Version 1.6.7beta01 [September 30, 2013] - Revised unknown chunk code to correct several bugs in the NO_SAVE_/NO_WRITE - combination - Allow HANDLE_AS_UNKNOWN to work when other options are configured off. Also - fixed the pngminim makefiles to work when $(MAKEFLAGS) contains stuff - which terminates the make options (as by default in recent versions of - Gentoo). - Avoid up-cast warnings in pngvalid.c. On ARM the alignment requirements of - png_modifier are greater than that of png_store and as a consequence - compilation of pngvalid.c results in a warning about increased alignment - requirements because of the bare cast to (png_modifier*). The code is safe, - because the pointer is known to point to a stack allocated png_modifier, - but this change avoids the warning. - Fixed default behavior of ARM_NEON_API. If the ARM NEON API option was - compiled without the CHECK option it defaulted to on, not off. - Check user callback behavior in pngunknown.c. Previous versions compiled - if SAVE_UNKNOWN was not available but did nothing since the callback - was never implemented. - Merged pngunknown.c with 1.7 version and back ported 1.7 improvements/fixes - -Version 1.6.7beta02 [October 12, 2013] - Made changes for compatibility with automake 1.14: - 1) Added the 'compile' program to the list of programs that must be cleaned - in autogen.sh - 2) Added 'subdir-objects' which causes .c files in sub-directories to be - compiled such that the corresponding .o files are also in the - sub-directory. This is because automake 1.14 warns that the - current behavior of compiling to the top level directory may be removed - in the future. - 3) Updated dependencies on pnglibconf.h to match the new .o locations and - added all the files in contrib/libtests and contrib/tools that depend - on pnglibconf.h - 4) Added 'BUILD_SOURCES = pnglibconf.h'; this is the automake recommended - way of handling the dependencies of sources that are machine generated; - unfortunately it only works if the user does 'make all' or 'make check', - so the dependencies (3) are still required. - Cleaned up (char*) casts of zlib messages. The latest version of the Intel C - compiler complains about casting a string literal as (char*), so copied the - treatment of z_const from the library code into pngfix.c - Simplified error message code in pngunknown. The simplification has the - useful side effect of avoiding a bogus warning generated by the latest - version of the Intel C compiler (it objects to - condition ? string-literal : string-literal). - Make autogen.sh work with automake 1.13 as well as 1.14. Do this by always - removing the 1.14 'compile' script but never checking for it. - -Version 1.6.7beta03 [October 19, 2013] - Added ARMv8 support (James Yu ). Added file - arm/filter_neon_intrinsics.c; enable with -mfpu=neon. - Revised pngvalid to generate size images with as many filters as it can - manage, limited by the number of rows. - Cleaned up ARM NEON compilation handling. The tests are now in pngpriv.h - and detect the broken GCC compilers. - -Version 1.6.7beta04 [October 26, 2013] - Allow clang derived from older GCC versions to use ARM intrinsics. This - causes all clang builds that use -mfpu=neon to use the intrinsics code, - not the assembler code. This has only been tested on iOS 7. It may be - necessary to exclude some earlier clang versions but this seems unlikely. - Changed NEON implementation selection mechanism. This allows assembler - or intrinsics to be turned on at compile time during the build by defining - PNG_ARM_NEON_IMPLEMENTATION to the correct value (2 or 1). This macro - is undefined by default and the build type is selected in pngpriv.h. - -Version 1.6.7rc01 [November 2, 2013] - No changes. - -Version 1.6.7rc02 [November 7, 2013] - Fixed #include in filter_neon_intrinsics.c and ctype macros. The ctype char - checking macros take an unsigned char argument, not a signed char. - -Version 1.6.7 [November 14, 2013] - No changes. - -Version 1.6.8beta01 [November 24, 2013] - Moved prototype for png_handle_unknown() in pngpriv.h outside of - the #ifdef PNG_SET_UNKNOWN_CHUNKS_SUPPORTED/#endif block. - Added "-Wall" to CFLAGS in contrib/pngminim/*/makefile - Conditionally compile some unused functions reported by -Wall in - pngminim. - Fixed 'minimal' builds. Various obviously useful minimal configurations - don't build because of missing contrib/libtests test programs and - overly complex dependencies in scripts/pnglibconf.dfa. This change - adds contrib/conftest/*.dfa files that can be used in automatic build - scripts to ensure that these configurations continue to build. - Enabled WRITE_INVERT and WRITE_PACK in contrib/pngminim/encoder. - Fixed pngvalid 'fail' function declaration on the Intel C Compiler. - This reverts to the previous 'static' implementation and works round - the 'unused static function' warning by using PNG_UNUSED(). - -Version 1.6.8beta02 [November 30, 2013] - Removed or marked PNG_UNUSED some harmless "dead assignments" reported - by clang scan-build. - Changed tabs to 3 spaces in png_debug macros and changed '"%s"m' - to '"%s" m' to improve portability among compilers. - Changed png_free_default() to free() in pngtest.c - -Version 1.6.8rc01 [December 12, 2013] - Tidied up pngfix inits and fixed pngtest no-write builds. - -Version 1.6.8rc02 [December 14, 2013] - Handle zero-length PLTE chunk or NULL palette with png_error() - instead of png_chunk_report(), which by default issues a warning - rather than an error, leading to later reading from a NULL pointer - (png_ptr->palette) in png_do_expand_palette(). This is CVE-2013-6954 - and VU#650142. Libpng-1.6.1 through 1.6.7 are vulnerable. - Libpng-1.6.0 and earlier do not have this bug. - -Version 1.6.8 [December 19, 2013] - No changes. - -Version 1.6.9beta01 [December 26, 2013] - Bookkeeping: Moved functions around (no changes). Moved transform - function definitions before the place where they are called so that - they can be made static. Move the intrapixel functions and the - grayscale palette builder out of the png?tran.c files. The latter - isn't a transform function and is no longer used internally, and the - former MNG specific functions are better placed in pngread/pngwrite.c - Made transform implementation functions static. This makes the internal - functions called by png_do_{read|write}_transformations static. On an - x86-64 DLL build (Gentoo Linux) this reduces the size of the text - segment of the DLL by 1208 bytes, about 0.6%. It also simplifies - maintenance by removing the declarations from pngpriv.h and allowing - easier changes to the internal interfaces. - Rebuilt configure scripts with automake-1.14.1 and autoconf-2.69 - in the tar distributions. - -Version 1.6.9beta02 [January 1, 2014] - Added checks for libpng 1.5 to pngvalid.c. This supports the use of - this version of pngvalid in libpng 1.5 - Merged with pngvalid.c from libpng-1.7 changes to create a single - pngvalid.c - Removed #error macro from contrib/tools/pngfix.c (Thomas Klausner). - Merged pngrio.c, pngtrans.c, pngwio.c, and pngerror.c with libpng-1.7.0 - Merged libpng-1.7.0 changes to make no-interlace configurations work - with test programs. - Revised pngvalid.c to support libpng 1.5, which does not support the - PNG_MAXIMUM_INFLATE_WINDOW option, so #define it out when appropriate in - pngvalid.c - Allow unversioned links created on install to be disabled in configure. - In configure builds 'make install' changes/adds links like png.h - and libpng.a to point to the newly installed, versioned, files (e.g. - libpng17/png.h and libpng17.a). Three new configure options and some - rearrangement of Makefile.am allow creation of these links to be disabled. - -Version 1.6.9beta03 [January 10, 2014] - Removed potentially misleading warning from png_check_IHDR(). - -Version 1.6.9beta04 [January 20, 2014] - Updated scripts/makefile.* to use CPPFLAGS (Cosmin). - Added clang attribute support (Cosmin). - -Version 1.6.9rc01 [January 28, 2014] - No changes. - -Version 1.6.9rc02 [January 30, 2014] - Quiet an uninitialized memory warning from VC2013 in png_get_png(). - -Version 1.6.9 [February 6, 2014] - -Version 1.6.10beta01 [February 9, 2014] - Backported changes from libpng-1.7.0beta30 and beta31: - Fixed a large number of instances where PNGCBAPI was omitted from - function definitions. - Added pngimage test program for png_read_png() and png_write_png() - with two new test scripts. - Removed dependence on !PNG_READ_EXPAND_SUPPORTED for calling - png_set_packing() in png_read_png(). - Fixed combination of ~alpha with shift. On read invert alpha, processing - occurred after shift processing, which causes the final values to be - outside the range that should be produced by the shift. Reversing the - order on read makes the two transforms work together correctly and mirrors - the order used on write. - Do not read invalid sBIT chunks. Previously libpng only checked sBIT - values on write, so a malicious PNG writer could therefore cause - the read code to return an invalid sBIT chunk, which might lead to - application errors or crashes. Such chunks are now skipped (with - chunk_benign_error). - Make png_read_png() and png_write_png() prototypes in png.h depend - upon PNG_READ_SUPPORTED and PNG_WRITE_SUPPORTED. - Support builds with unsupported PNG_TRANSFORM_* values. All of the - PNG_TRANSFORM_* values are always defined in png.h and, because they - are used for both read and write in some cases, it is not reliable - to #if out ones that are totally unsupported. This change adds error - detection in png_read_image() and png_write_image() to do a - png_app_error() if the app requests something that cannot be done - and it adds corresponding code to pngimage.c to handle such options - by not attempting to test them. - -Version 1.6.10beta02 [February 23, 2014] - Moved redefines of png_error(), png_warning(), png_chunk_error(), - and png_chunk_warning() from pngpriv.h to png.h to make them visible - to libpng-calling applications. - Moved OS dependent code from arm/arm_init.c, to allow the included - implementation of the ARM NEON discovery function to be set at - build-time and provide sample implementations from the current code in the - contrib/arm-neon subdirectory. The __linux__ code has also been changed to - compile and link on Android by using /proc/cpuinfo, and the old linux code - is in contrib/arm-neon/linux-auxv.c. The new code avoids POSIX and Linux - dependencies apart from opening /proc/cpuinfo and is C90 compliant. - Check for info_ptr == NULL early in png_read_end() so we don't need to - run all the png_handle_*() and depend on them to return if info_ptr == NULL. - This improves the performance of png_read_end(png_ptr, NULL) and makes - it more robust against future programming errors. - Check for __has_extension before using it in pngconf.h, to - support older Clang versions (Jeremy Sequoia). - Treat CRC error handling with png_set_crc_action(), instead of with - png_set_benign_errors(), which has been the case since libpng-1.6.0beta18. - Use a user warning handler in contrib/gregbook/readpng2.c instead of default, - so warnings will be put on stderr even if libpng has CONSOLE_IO disabled. - Added png_ptr->process_mode = PNG_READ_IDAT_MODE in png_push_read_chunk - after recognizing the IDAT chunk, which avoids an infinite loop while - reading a datastream whose first IDAT chunk is of zero-length. - This fixes CERT VU#684412 and CVE-2014-0333. - Don't recognize known sRGB profiles as sRGB if they have been hacked, - but don't reject them and don't issue a copyright violation warning. - -Version 1.6.10beta03 [February 25, 2014] - Moved some documentation from png.h to libpng.3 and libpng-manual.txt - Minor editing of contrib/arm-neon/README and contrib/examples/*.c - -Version 1.6.10rc01 [February 27, 2014] - Fixed typos in the manual and in scripts/pnglibconf.dfa (CFLAGS -> CPPFLAGS - and PNG_USR_CONFIG -> PNG_USER_CONFIG). - -Version 1.6.10rc02 [February 28, 2014] - Removed unreachable return statement after png_chunk_error() - in pngrutil.c - -Version 1.6.10rc03 [March 4, 2014] - Un-deprecated png_data_freer(). - -Version 1.6.10 [March 6, 2014] - No changes. - -Version 1.6.11beta01 [March 17, 2014] - Use "if (value != 0)" instead of "if (value)" consistently. - Changed ZlibSrcDir from 1.2.5 to 1.2.8 in projects/vstudio. - Moved configuration information from the manual to the INSTALL file. - -Version 1.6.11beta02 [April 6, 2014] - Removed #if/#else/#endif from inside two pow() calls in pngvalid.c because - they were handled improperly by Portland Group's PGI-14.1 - PGI-14.3 - when using its "__builtin_pow()" function. - Silence 'unused parameter' build warnings (Cosmin Truta). - $(CP) is now used alongside $(RM_F). Also, use 'copy' instead of 'cp' - where applicable, and applied other minor makefile changes (Cosmin). - Don't warn about invalid dimensions exceeding user limits (Cosmin). - Allow an easy replacement of the default pre-built configuration - header with a custom header, via the make PNGLIBCONF_H_PREBUILT - macro (Cosmin). - -Version 1.6.11beta03 [April 6, 2014] - Fixed a typo in pngrutil.c, introduced in libpng-1.5.6, that interferes - with "blocky" expansion of sub-8-bit interlaced PNG files (Eric Huss). - Optionally use __builtin_bswap16() in png_do_swap(). - -Version 1.6.11beta04 [April 19, 2014] - Made progressive reading of interlaced images consistent with the - behavior of the sequential reader and consistent with the manual, by - moving some code out of the PNG_READ_INTERLACING_SUPPORTED blocks. The - row_callback now receives the proper pass number and unexpanded rows, when - png_combine_row() isn't built or used, and png_set_interlace_handling() - is not called. - Allow PNG_sRGB_PROFILE_CHECKING = (-1) to mean no sRGB profile checking. - -Version 1.6.11beta05 [April 26, 2014] - Do not reject ICC V2 profiles that lack padding (Kai-Uwe Behrmann). - Relocated closing bracket of the sRGB profile test loop to avoid getting - "Not recognizing known sRGB profile that has been edited" warning for - ICC V2 profiles that lack the MD5 signature in the profile header. - -Version 1.6.11beta06 [May 19, 2014] - Added PNG_SKIP_sRGB_CHECK_PROFILE choice for png_set_option(). - -Version 1.6.11rc01 [May 27, 2014] - No changes. - -Version 1.6.11rc02 [June 3, 2014] - Test ZLIB_VERNUM instead of PNG_ZLIB_VERNUM in contrib/tools/pngfix.c - -Version 1.6.11 [June 5, 2014] - No changes. - -Version 1.6.12rc01 [June 6, 2014] - Relocated new code from 1.6.11beta06 in png.c to a point after the - declarations (Max Stepin). - -Version 1.6.12rc02 [June 7, 2014] - Changed file permissions of contrib/tools/intgamma.sh, - test-driver, and compile from 0644 to 0755 (Cosmin). - -Version 1.6.12rc03 [June 8, 2014] - Ensure "__has_attribute()" macro exists before trying to use it with - old clang compilers (MacPorts Ticket #43939). - -Version 1.6.12 [June 12, 2014] - No changes. - -Version 1.6.13beta01 [July 4, 2014] - Quieted -Wsign-compare and -Wclobber compiler warnings in - contrib/pngminus/*.c - Added "(void) png_ptr;" where needed in contrib/gregbook to quiet - compiler complaints about unused pointers. - Split a long output string in contrib/gregbook/rpng2-x.c. - Added "PNG_SET_OPTION" requirement for sRGB chunk support to pnglibconf.dfa, - Needed for write-only support (John Bowler). - Changed "if defined(__ARM_NEON__)" to - "if (defined(__ARM_NEON__) || defined(__ARM_NEON))" (James Wu). - Fixed clang no-warning builds: png_digit was defined but never used. - -Version 1.6.13beta02 [July 21, 2014] - Fixed an incorrect separator ("/" should be "\") in scripts/makefile.vcwin32 - (bug report from Wolfgang S. Kechel). Bug was introduced in libpng-1.6.11. - Also fixed makefile.bc32, makefile.bor, makefile.msc, makefile.intel, and - makefile.tc3 similarly. - -Version 1.6.13beta03 [August 3, 2014] - Removed scripts/makefile.elf. It has not worked since libpng-1.5.0beta14 - due to elimination of the PNG_FUNCTION_EXPORT and PNG_DATA_EXPORT - definitions from pngconf.h. - Ensure that CMakeLists.txt makes the target "lib" directory before making - symbolic link into it (SourceForge bug report #226 by Rolf Timmermans). - -Version 1.6.13beta04 [August 8, 2014] - Added opinion that the ECCN (Export Control Classification Number) for - libpng is EAR99 to the README file. - Eliminated use of "$<" in makefile explicit rules, when copying - $PNGLIBCONF_H_PREBUILT. This does not work on some versions of make; - bug introduced in libpng version 1.6.11. - -Version 1.6.13rc01 [August 14, 2014] - Made "ccopts" agree with "CFLAGS" in scripts/makefile.hp* and makefile.*sunu - -Version 1.6.13 [August 21, 2014] - No changes. - -Version 1.6.14beta01 [September 14, 2014] - Guard usage of png_ptr->options with #ifdef PNG_SET_OPTION_SUPPORTED. - Do not build contrib/tools/pngfix.c when PNG_SETJMP_NOT_SUPPORTED, - to allow "make" to complete without setjmp support (bug report by - Claudio Fontana) - Add "#include " to contrib/tools/pngfix.c (John Bowler) - -Version 1.6.14beta02 [September 18, 2014] - Use nanosleep() instead of usleep() in contrib/gregbook/rpng2-x.c - because usleep() is deprecated. - Define usleep() in contrib/gregbook/rpng2-x.c if not already defined - in unistd.h and nanosleep() is not available; fixes error introduced - in libpng-1.6.13. - Disable floating point exception handling in pngvalid.c when - PNG_FLOATING_ARITHMETIC is not supported (bug report by "zootus - at users.sourceforge.net"). - -Version 1.6.14beta03 [September 19, 2014] - Define FE_DIVBYZERO, FE_INVALID, and FE_OVERFLOW in pngvalid.c if not - already defined. Revert floating point exception handling in pngvalid.c - to version 1.6.14beta01 behavior. - -Version 1.6.14beta04 [September 27, 2014] - Fixed incorrect handling of the iTXt compression flag in pngrutil.c - (bug report by Shunsaku Hirata). Bug was introduced in libpng-1.6.0. - -Version 1.6.14beta05 [October 1, 2014] - Added "option READ_iCCP enables READ_COMPRESSED_TEXT" to pnglibconf.dfa - -Version 1.6.14beta06 [October 5, 2014] - Removed unused "text_len" parameter from private function png_write_zTXt(). - Conditionally compile some code in png_deflate_claim(), when - PNG_WARNINGS_SUPPORTED and PNG_ERROR_TEXT_SUPPORTED are disabled. - Replaced repeated code in pngpread.c with PNG_PUSH_SAVE_BUFFER_IF_FULL. - Added "chunk iTXt enables TEXT" and "chunk zTXt enables TEXT" - to pnglibconf.dfa. - Removed "option READ_COMPRESSED_TEXT enables READ_TEXT" from pnglibconf.dfa, - to make it possible to configure a libpng that supports iCCP but not TEXT. - -Version 1.6.14beta07 [October 7, 2014] - Removed "option WRITE_COMPRESSED_TEXT enables WRITE_TEXT" from pnglibconf.dfa - Only mark text chunks as written after successfully writing them. - -Version 1.6.14rc01 [October 15, 2014] - Fixed some typos in comments. - -Version 1.6.14rc02 [October 17, 2014] - Changed png_convert_to_rfc_1123() to png_convert_to_rfc_1123_buffer() - in the manual, to reflect the change made in libpng-1.6.0. - Updated README file to explain that direct access to the png_struct - and info_struct members has not been permitted since libpng-1.5.0. - -Version 1.6.14 [October 23, 2014] - No changes. - -Version 1.6.15beta01 [October 29, 2014] - Changed "if (!x)" to "if (x == 0)" and "if (x)" to "if (x != 0)" - Simplified png_free_data(). - Added missing "ptr = NULL" after some instances of png_free(). - -Version 1.6.15beta02 [November 1, 2014] - Changed remaining "if (!x)" to "if (x == 0)" and "if (x)" to "if (x != 0)" - -Version 1.6.15beta03 [November 3, 2014] - Added PNG_USE_ARM_NEON configuration flag (Marcin Juszkiewicz). - -Version 1.6.15beta04 [November 4, 2014] - Removed new PNG_USE_ARM_NEON configuration flag and made a one-line - revision to configure.ac to support ARM on aarch64 instead (John Bowler). - -Version 1.6.15beta05 [November 5, 2014] - Use png_get_libpng_ver(NULL) instead of PNG_LIBPNG_VER_STRING in - example.c, pngtest.c, and applications in the contrib directory. - Fixed an out-of-range read in png_user_version_check() (Bug report from - Qixue Xiao, CVE-2015-8540). - Simplified and future-proofed png_user_version_check(). - Fixed GCC unsigned int->float warnings. Various versions of GCC - seem to generate warnings when an unsigned value is implicitly - converted to double. This is probably a GCC bug but this change - avoids the issue by explicitly converting to (int) where safe. - Free all allocated memory in pngimage. The file buffer cache was left - allocated at the end of the program, harmless but it causes memory - leak reports from clang. - Fixed array size calculations to avoid warnings. At various points - in the code the number of elements in an array is calculated using - sizeof. This generates a compile time constant of type (size_t) which - is then typically assigned to an (unsigned int) or (int). Some versions - of GCC on 64-bit systems warn about the apparent narrowing, even though - the same compiler does apparently generate the correct, in-range, - numeric constant. This adds appropriate, safe, casts to make the - warnings go away. - -Version 1.6.15beta06 [November 6, 2014] - Reverted use png_get_libpng_ver(NULL) instead of PNG_LIBPNG_VER_STRING - in the manual, example.c, pngtest.c, and applications in the contrib - directory. It was incorrect advice. - -Version 1.6.15beta07 [November 7, 2014] - Removed #ifdef PNG_16BIT_SUPPORTED/#endif around png_product2(); it is - needed by png_reciprocal2(). - Added #ifdef PNG_16BIT_SUPPORTED/#endif around png_log16bit() and - png_do_swap(). - Changed all "#endif /* PNG_FEATURE_SUPPORTED */" to "#endif /* FEATURE */" - -Version 1.6.15beta08 [November 8, 2014] - More housecleaning in *.h - -Version 1.6.15rc01 [November 13, 2014] - -Version 1.6.15rc02 [November 14, 2014] - The macros passed in the command line to Borland make were ignored if - similarly-named macros were already defined in makefiles. This behavior - is different from POSIX make and other make programs. Surround the - macro definitions with ifndef guards (Cosmin). - -Version 1.6.15rc03 [November 16, 2014] - Added "-D_CRT_SECURE_NO_WARNINGS" to CFLAGS in scripts/makefile.vcwin32. - Removed the obsolete $ARCH variable from scripts/makefile.darwin. - -Version 1.6.15 [November 20, 2014] - No changes. - -Version 1.6.16beta01 [December 14, 2014] - Added ".align 2" to arm/filter_neon.S to support old GAS assemblers that - don't do alignment correctly. - Revised Makefile.am and scripts/symbols.dfn to work with MinGW/MSYS - (Bob Friesenhahn). - -Version 1.6.16beta02 [December 15, 2014] - Revised Makefile.am and scripts/*.dfn again to work with MinGW/MSYS; - renamed scripts/*.dfn to scripts/*.c (John Bowler). - -Version 1.6.16beta03 [December 21, 2014] - Quiet a "comparison always true" warning in pngstest.c (John Bowler). - -Version 1.6.16rc01 [December 21, 2014] - Restored a test on width that was removed from png.c at libpng-1.6.9 - (Bug report by Alex Eubanks, CVE-2015-0973). - -Version 1.6.16rc02 [December 21, 2014] - Undid the update to pngrutil.c in 1.6.16rc01. - -Version 1.6.16rc03 [December 21, 2014] - Fixed an overflow in png_combine_row() with very wide interlaced images - (Bug report and fix by John Bowler, CVE-2014-9495). - -Version 1.6.16 [December 22, 2014] - No changes. - -Version 1.6.17beta01 [January 29, 2015] - Removed duplicate PNG_SAFE_LIMITS_SUPPORTED handling from pngconf.h - Corrected the width limit calculation in png_check_IHDR(). - Removed user limits from pngfix. Also pass NULL pointers to - png_read_row to skip the unnecessary row de-interlace stuff. - Added testing of png_set_packing() to pngvalid.c - Regenerated configure scripts in the *.tar distributions with libtool-2.4.4 - Implement previously untested cases of libpng transforms in pngvalid.c - Fixed byte order in png_do_read_filler() with 16-bit input. Previously - the high and low bytes of the filler, from png_set_filler() or from - png_set_add_alpha(), were read in the wrong order. - Made the check for out-of-range values in png_set_tRNS() detect - values that are exactly 2^bit_depth, and work on 16-bit platforms. - Merged some parts of libpng-1.6.17beta01 and libpng-1.7.0beta47. - Added #ifndef __COVERITY__ where needed in png.c, pngrutil.c and - pngset.c to avoid warnings about dead code. - Added "& 0xff" to many instances of expressions that are typecast - to (png_byte), to avoid Coverity warnings. - -Version 1.6.17beta02 [February 7, 2015] - Work around one more Coverity-scan dead-code warning. - Do not build png_product2() when it is unused. - -Version 1.6.17beta03 [February 17, 2015] - Display user limits in the output from pngtest. - Eliminated the PNG_SAFE_LIMITS macro and restored the 1-million-column - and 1-million-row default limits in pnglibconf.dfa, that can be reset - by the user at build time or run time. This provides a more robust - defense against DOS and as-yet undiscovered overflows. - -Version 1.6.17beta04 [February 21, 2015] - Added PNG_WRITE_CUSTOMIZE_COMPRESSION_SUPPORTED macro, on by default. - Allow user to call png_get_IHDR() with NULL arguments (Reuben Hawkins). - Rebuilt configure scripts with automake-1.15 and libtool-2.4.6 - -Version 1.6.17beta05 [February 25, 2015] - Restored compiling of png_reciprocal2 with PNG_NO_16BIT. - -Version 1.6.17beta06 [February 27, 2015] - Moved png_set_filter() prototype into a PNG_WRITE_SUPPORTED block - of png.h. - Avoid runtime checks when converting integer to png_byte with - Visual Studio (Sergey Kosarevsky) - -Version 1.6.17rc01 [March 4, 2015] - No changes. - -Version 1.6.17rc02 [March 9, 2015] - Removed some comments that the configure script did not handle - properly from scripts/pnglibconf.dfa and pnglibconf.h.prebuilt. - Free the unknown_chunks structure even when it contains no data. - -Version 1.6.17rc03 [March 12, 2015] - Updated CMakeLists.txt to add OSX framework, change YES/NO to ON/OFF - for consistency, and remove some useless tests (Alexey Petruchik). - -Version 1.6.17rc04 [March 16, 2015] - Remove pnglibconf.h, pnglibconf.c, and pnglibconf.out instead of - pnglibconf.* in "make clean" (Cosmin). - Fix bug in calculation of maxbits, in png_write_sBIT, introduced - in libpng-1.6.17beta01 (John Bowler). - -Version 1.6.17rc05 [March 21, 2015] - Define PNG_FILTER_* and PNG_FILTER_VALUE_* in png.h even when WRITE - is not supported (John Bowler). This fixes an error introduced in - libpng-1.6.17beta06. - Reverted "& 0xff" additions of version 1.6.17beta01. Libpng passes - the Coverity scan without them. - -Version 1.6.17rc06 [March 23, 2015] - Remove pnglibconf.dfn and pnglibconf.pre with "make clean". - Reformatted some "&0xff" instances to "& 0xff". - Fixed simplified 8-bit-linear to sRGB alpha. The calculated alpha - value was wrong. It's not clear if this affected the final stored - value; in the obvious code path the upper and lower 8-bits of the - alpha value were identical and the alpha was truncated to 8-bits - rather than dividing by 257 (John Bowler). - -Version 1.6.17 [March 26, 2015] - No changes. - -Version 1.6.18beta01 [April 1, 2015] - Removed PNG_SET_CHUNK_[CACHE|MALLOC]_LIMIT_SUPPORTED macros. They - have been combined with PNG_SET_USER_LIMITS_SUPPORTED (resolves - bug report by Andrew Church). - Fixed rgb_to_gray checks and added tRNS checks to pngvalid.c. This - fixes some arithmetic errors that caused some tests to fail on - some 32-bit platforms (Bug reports by Peter Breitenlohner [i686] - and Petr Gajdos [i586]). - -Version 1.6.18beta02 [April 26, 2015] - Suppressed some warnings from the Borland C++ 5.5.1/5.82 compiler - (Bug report by Viktor Szakats). - -Version 1.6.18beta03 [May 6, 2015] - Replaced "unexpected" with an integer (0xabadca11) in pngset.c - where a long was expected, to avoid a compiler warning when PNG_DEBUG > 1. - Added contrib/examples/simpleover.c, to demonstrate how to handle - alpha compositing of multiple images, using the "simplified API" - and an example PNG generation tool, contrib/examples/genpng.c - (John Bowler). - -Version 1.6.18beta04 [May 20, 2015] - PNG_RELEASE_BUILD replaces tests where the code depended on the build base - type and can be defined on the command line, allowing testing in beta - builds (John Bowler). - Avoid Coverity issue 80858 (REVERSE NULL) in pngtest.c PNG_DEBUG builds. - Avoid a harmless potential integer overflow in png_XYZ_from_xy() (Bug - report from Christopher Ferris). - -Version 1.6.18beta05 [May 31, 2015] - Backport filter selection code from libpng-1.7.0beta51, to combine - sub_row, up_row, avg_row, and paeth_row into try_row and tst_row. - Changed png_voidcast(), etc., to voidcast(), etc., in contrib/tools/pngfix.c - to avoid confusion with the libpng private macros. - Fixed old cut&paste bug in the weighted filter selection code in - pngwutil.c, introduced in libpng-0.95, March 1997. - -Version 1.6.18beta06 [June 1, 2015] - Removed WRITE_WEIGHTED_FILTERED code, to save a few kbytes of the - compiled library size. It never worked properly and as far as we can - tell, no one uses it. The png_set_filter_heuristics() and - png_set_filter_heuristics_fixed() APIs are retained but deprecated - and do nothing. - -Version 1.6.18beta07 [June 6, 2015] - Removed non-working progressive reader 'skip' function. This - function has apparently never been used. It was implemented - to support back-door modification of png_struct in libpng-1.4.x - but (because it does nothing and cannot do anything) was apparently - never tested (John Bowler). - Fixed cexcept.h in which GCC 5 now reports that one of the auto - variables in the Try macro needs to be volatile to prevent value - being lost over the setjmp (John Bowler). - Fixed NO_WRITE_FILTER and -Wconversion build breaks (John Bowler). - Fix g++ build breaks (John Bowler). - Quieted some Coverity issues in pngfix.c, png-fix-itxt.c, pngvalid.c, - pngstest.c, and pngimage.c. Most seem harmless, but png-fix-itxt - would only work with iTXt chunks with length 255 or less. - Added #ifdef's to contrib/examples programs so people don't try - to compile them without the minimum required support enabled - (suggested by Flavio Medeiros). - -Version 1.6.18beta08 [June 30, 2015] - Eliminated the final two Coverity defects (insecure temporary file - handling in contrib/libtests/pngstest.c; possible overflow of - unsigned char in contrib/tools/png-fix-itxt.c). To use the "secure" - file handling, define PNG_USE_MKSTEMP, otherwise "tmpfile()" will - be used. - Removed some unused WEIGHTED_FILTER macros from png.h and pngstruct.h - -Version 1.6.18beta09 [July 5, 2015] - Removed some useless typecasts from contrib/tools/png-fix-itxt.c - Fixed a new signed-unsigned comparison in pngrtran.c (Max Stepin). - Replaced arbitrary use of 'extern' with #define PNG_LINKAGE_*. To - preserve API compatibility, the new defines all default to "extern" - (requested by Jan Nijtmans). - -Version 1.6.18rc01 [July 9, 2015] - Belatedly added Mans Rullgard and James Yu to the list of Contributing - Authors. - -Version 1.6.18rc02 [July 12, 2015] - Restored unused FILTER_HEURISTIC macros removed at libpng-1.6.18beta08 - to png.h to avoid compatibility warnings. - -Version 1.6.18rc03 [July 15, 2015] - Minor changes to the man page - -Version 1.6.18 [July 23, 2015] - No changes. - -Version 1.6.19beta01 [July 30, 2015] - Updated obsolete information about the simplified API macros in the - manual pages (Bug report by Arc Riley). - Avoid potentially dereferencing NULL info_ptr in png_info_init_3(). - Rearranged png.h to put the major sections in the same order as - in libpng17. - Eliminated unused PNG_COST_SHIFT, PNG_WEIGHT_SHIFT, PNG_COST_FACTOR, and - PNG_WEIGHT_FACTOR macros. - Suppressed some warnings from the Borland C++ 5.5.1/5.82 compiler - (Bug report by Viktor Szakats). Several warnings remain and are - unavoidable, where we test for overflow. - Fixed potential leak of png_pixels in contrib/pngminus/pnm2png.c - Fixed uninitialized variable in contrib/gregbook/rpng2-x.c - -Version 1.6.19beta02 [August 19, 2015] - Moved config.h.in~ from the "libpng_autotools_files" list to the - "libpng_autotools_extra" list in autogen.sh because it was causing a - false positive for missing files (bug report by Robert C. Seacord). - Removed unreachable "break" statements in png.c, pngread.c, and pngrtran.c - to suppress clang warnings (Bug report by Viktor Szakats). - Fixed some bad links in the man page. - Changed "n bit" to "n-bit" in comments. - Added signed/unsigned 16-bit safety net. This removes the dubious - 0x8000 flag definitions on 16-bit systems. They aren't supported - yet the defs *probably* work, however it seems much safer to do this - and be advised if anyone, contrary to advice, is building libpng 1.6 - on a 16-bit system. It also adds back various switch default clauses - for GCC; GCC errors out if they are not present (with an appropriately - high level of warnings). - Safely convert num_bytes to a png_byte in png_set_sig_bytes() (Robert - Seacord). - Fixed the recently reported 1's complement security issue by replacing - the value that is illegal in the PNG spec, in both signed and unsigned - values, with 0. Illegal unsigned values (anything greater than or equal - to 0x80000000) can still pass through, but since these are not illegal - in ANSI-C (unlike 0x80000000 in the signed case) the checking that - occurs later can catch them (John Bowler). - -Version 1.6.19beta03 [September 26, 2015] - Fixed png_save_int_32 when int is not 2's complement (John Bowler). - Updated libpng16 with all the recent test changes from libpng17, - including changes to pngvalid.c to ensure that the original, - distributed, version of contrib/visupng/cexcept.h can be used - (John Bowler). - pngvalid contains the correction to the use of SAVE/STORE_ - UNKNOWN_CHUNKS; a bug revealed by changes in libpng 1.7. More - tests contain the --strict option to detect warnings and the - pngvalid-standard test has been corrected so that it does not - turn on progressive-read. There is a separate test which does - that. (John Bowler) - Also made some signed/unsigned fixes. - Make pngstest error limits version specific. Splitting the machine - generated error structs out to a file allows the values to be updated - without changing pngstest.c itself. Since libpng 1.6 and 1.7 have - slightly different error limits this simplifies maintenance. The - makepngs.sh script has also been updated to more accurately reflect - current problems in libpng 1.7 (John Bowler). - Incorporated new test PNG files into make check. tests/pngstest-* - are changed so that the new test files are divided into 8 groups by - gamma and alpha channel. These tests have considerably better code - and pixel-value coverage than contrib/pngsuite; however,coverage is - still incomplete (John Bowler). - Removed the '--strict' in 1.6 because of the double-gamma-correction - warning, updated pngstest-errors.h for the errors detected with the - new contrib/testspngs PNG test files (John Bowler). - -Version 1.6.19beta04 [October 15, 2015] - Worked around rgb-to-gray issues in libpng 1.6. The previous - attempts to ignore the errors in the code aren't quite enough to - deal with the 'channel selection' encoding added to libpng 1.7; abort. - pngvalid.c is changed to drop this encoding in prior versions. - Fixed 'pow' macros in pngvalid.c. It is legal for 'pow' to be a - macro, therefore the argument list cannot contain preprocessing - directives. Make sure pow is a function where this happens. This is - a minimal safe fix, the issue only arises in non-performance-critical - code (bug report by Curtis Leach, fix by John Bowler). - Added sPLT support to pngtest.c - -Version 1.6.19rc01 [October 23, 2015] - No changes. - -Version 1.6.19rc02 [October 31, 2015] - Prevent setting or writing over-length PLTE chunk (Cosmin Truta). - Silently truncate over-length PLTE chunk while reading. - Libpng incorrectly calculated the output rowbytes when the application - decreased either the number of channels or the bit depth (or both) in - a user transform. This was safe; libpng overallocated buffer space - (potentially by quite a lot; up to 4 times the amount required) but, - from 1.5.4 on, resulted in a png_error (John Bowler). - -Version 1.6.19rc03 [November 3, 2015] - Fixed some inconsequential cut-and-paste typos in png_set_cHRM_XYZ_fixed(). - Clarified COPYRIGHT information to state explicitly that versions - are derived from previous versions. - Removed much of the long list of previous versions from png.h and - libpng.3. - -Version 1.6.19rc04 [November 5, 2015] - Fixed new bug with CRC error after reading an over-length palette - (bug report by Cosmin Truta) (CVE-2015-8126). - -Version 1.6.19 [November 12, 2015] - Cleaned up coding style in png_handle_PLTE(). - -Version 1.6.20beta01 [November 20, 2015] - Avoid potential pointer overflow/underflow in png_handle_sPLT() and - png_handle_pCAL() (Bug report by John Regehr). - -Version 1.6.20beta02 [November 23, 2015] - Fixed incorrect implementation of png_set_PLTE() that uses png_ptr - not info_ptr, that left png_set_PLTE() open to the CVE-2015-8126 - vulnerability. Fixes CVE-2015-8472. - -Version 1.6.20beta03 [November 24, 2015] - Backported tests from libpng-1.7.0beta69. - -Version 1.6.20rc01 [November 26, 2015] - Fixed an error in handling of bad zlib CMINFO field in pngfix, found by - American Fuzzy Lop, reported by Brian Carpenter. inflate() doesn't - immediately fault a bad CMINFO field; instead a 'too far back' error - happens later (at least some times). pngfix failed to limit CMINFO to - the allowed values but then assumed that window_bits was in range, - triggering an assert. The bug is mostly harmless; the PNG file cannot - be fixed. - -Version 1.6.20rc02 [November 29, 2015] - In libpng 1.6 zlib initialization was changed to use the window size - in the zlib stream, not a fixed value. This causes some invalid images, - where CINFO is too large, to display 'correctly' if the rest of the - data is valid. This provides a workaround for zlib versions where the - error arises (ones that support the API change to use the window size - in the stream). - -Version 1.6.20 [December 3, 2015] - No changes. - -Version 1.6.21beta01 [December 11, 2015] - Fixed syntax "$(command)" in tests/pngstest that some shells other than - bash could not parse (Bug report by Nelson Beebe). Use `command` instead. - -Version 1.6.21beta02 [December 14, 2015] - Moved png_check_keyword() from pngwutil.c to pngset.c - Removed LE/BE dependencies in pngvalid, to 'fix' the current problem - in the BigEndian tests by not testing it, making the BE code the same - as the LE version. - Fixes to pngvalid for various reduced build configurations (eliminate unused - statics) and a fix for the case in rgb_to_gray when the digitize option - reduces graylo to 0, producing a large error. - -Version 1.6.21beta03 [December 18, 2015] - Widened the 'limit' check on the internally calculated error limits in - the 'DIGITIZE' case (the code used prior to 1.7 for rgb_to_gray error - checks) and changed the check to only operate in non-release builds - (base build type not RC or RELEASE.) - Fixed undefined behavior in pngvalid.c, undefined because - (png_byte) << shift is undefined if it changes the signed bit - (because png_byte is promoted to int). The libpng exported functions - png_get_uint_32 and png_get_uint_16 handle this. (Bug reported by - David Drysdale as a result of reports from UBSAN in clang 3.8). - This changes pngvalid to use BE random numbers; this used to produce - errors but these should not be fixed as a result of the previous changes. - -Version 1.6.21rc01 [January 4, 2016] - In projects/vstudio, combined readme.txt and WARNING into README.txt - -Version 1.6.21rc02 [January 7, 2016] - Relocated assert() in contrib/tools/pngfix.c, bug found by American - Fuzzy Lop, reported by Brian Carpenter. - Marked 'limit' UNUSED in transform_range_check(). This only affects - release builds. - -Version 1.6.21 [January 15, 2016] - Worked around a false-positive Coverity issue in pngvalid.c. - -Version 1.6.22beta01 [January 23, 2016] - Changed PNG_USE_MKSTEMP to __COVERITY__ to select alternate - "tmpfile()" implementation in contrib/libtests/pngstest.c - Fixed NO_STDIO build of pngunknown.c to skip calling png_init_io() - if there is no stdio.h support. - Added a png_image_write_to_memory() API and a number of assist macros - to allow an application that uses the simplified API write to bypass - stdio and write directly to memory. - Added some warnings (png.h) and some check code to detect *possible* - overflow in the ROW_STRIDE and simplified image SIZE macros. This - disallows image width/height/format that *might* overflow. This is - a quiet API change that limits in-memory image size (uncompressed) to - less than 4GByte and image row size (stride) to less than 2GByte. - Revised workaround for false-positive Coverity issue in pngvalid.c. - -Version 1.6.22beta02 [February 8, 2016] - Only use exit(77) in configure builds. - Corrected error in PNG_IMAGE_PNG_SIZE_MAX. This new macro underreported - the palette size because it failed to take into account that the memory - palette has to be expanded to full RGB when it is written to PNG. - Updated CMakeLists.txt, added supporting scripts/gen*.cmake.in - and test.cmake.in (Roger Leigh). - Relaxed limit checks on gamma values in pngrtran.c. As suggested in - the comments gamma values outside the range currently permitted - by png_set_alpha_mode are useful for HDR data encoding. These values - are already permitted by png_set_gamma so it is reasonable caution to - extend the png_set_alpha_mode range as HDR imaging systems are starting - to emerge. - -Version 1.6.22beta03 [March 9, 2016] - Added a common-law trademark notice and export control information - to the LICENSE file, png.h, and the man page. - Restored "& 0xff" in png_save_uint_16() and png_save_uint_32() that - were accidentally removed from libpng-1.6.17. - Changed PNG_INFO_cHNK and PNG_FREE_cHNK from 0xnnnn to 0xnnnnU in png.h - (Robert C. Seacord). - Removed dubious "#if INT_MAX" test from png.h that was added to - libpng-1.6.19beta02 (John Bowler). - Add ${INCLUDES} in scripts/genout.cmake.in (Bug report by Nixon Kwok). - Updated LICENSE to say files in the contrib directory are not - necessarily under the libpng license, and that some makefiles have - other copyright owners. - Added INTEL-SSE2 support (Mike Klein and Matt Sarett, Google, Inc.). - Made contrib/libtests/timepng more robust. The code no longer gives - up/fails on invalid PNG data, it just skips it (with error messages). - The code no longer fails on PNG files with data beyond IEND. Options - exist to use png_read_png (reading the whole image, not by row) and, in - that case, to apply any of the supported transforms. This makes for - more realistic testing; the decoded data actually gets used in a - meaningful fashion (John Bowler). - Fixed some misleading indentation (Krishnaraj Bhat). - -Version 1.6.22beta04 [April 5, 2016] - Force GCC compilation to C89 if needed (Dagobert Michelsen). - SSE filter speed improvements for bpp=3: - memcpy-free implementations of load3() / store3(). - call load3() only when needed at the end of a scanline. - -Version 1.6.22beta05 [April 27, 2016] - Added PNG_FAST_FILTERS macro (defined as - PNG_FILTER_NONE|PNG_FILTER_SUB|PNG_FILTER_UP). - Various fixes for contrib/libtests/timepng.c - Moved INTEL-SSE code from pngpriv.h into contrib/intel/intel_sse.patch. - Fixed typo (missing underscore) in #define PNG_READ_16_TO_8_SUPPORTED - (Bug report by Y.Ohashik). - -Version 1.6.22beta06 [May 5, 2016] - Rebased contrib/intel_sse.patch. - Quieted two Coverity issues in contrib/libtests/timepng.c. - Fixed issues with scripts/genout.cmake.in (David Capello, Nixon Kwok): - Added support to use multiple directories in ZLIBINCDIR variable, - Fixed CMAKE_C_FLAGS with multiple values when genout is compiled on MSVC, - Fixed pnglibconf.c compilation on OS X including the sysroot path. - -Version 1.6.22rc01 [May 14, 2016] - No changes. - -Version 1.6.22rc02 [May 16, 2016] - Removed contrib/timepng from default build; it does not build on platforms - that don't supply clock_gettime(). - -Version 1.6.22rc03 [May 17, 2016] - Restored contrib/timepng to default build but check for the presence - of clock_gettime() in configure.ac and Makefile.am. - -Version 1.6.22 [May 26, 2016] - No changes. - -Version 1.6.23beta01 [May 29, 2016] - Stop a potential memory leak in png_set_tRNS() (Bug report by Ted Ying). - Fixed the progressive reader to handle empty first IDAT chunk properly - (patch by Timothy Nikkel). This bug was introduced in libpng-1.6.0 and - only affected the libpng16 branch. - Added tests in pngvalid.c to check zero-length IDAT chunks in various - positions. Fixed the sequential reader to handle these more robustly - (John Bowler). - -Version 1.6.23rc01 [June 2, 2016] - Corrected progressive read input buffer in pngvalid.c. The previous version - the code invariably passed just one byte at a time to libpng. The intent - was to pass a random number of bytes in the range 0..511. - Moved sse2 prototype from pngpriv.h to contrib/intel/intel_sse.patch. - Added missing ")" in pngerror.c (Matt Sarrett). - -Version 1.6.23rc02 [June 4, 2016] - Fixed undefined behavior in png_push_save_buffer(). Do not call - memcpy() with a null source, even if count is zero (Leon Scroggins III). - -Version 1.6.23 [June 9, 2016] - Fixed bad link to RFC2083 in png.5 (Nikola Forro). - -Version 1.6.24beta01 [June 11, 2016] - Avoid potential overflow of the PNG_IMAGE_SIZE macro. This macro - is not used within libpng, but is used in some of the examples. - -Version 1.6.24beta02 [June 23, 2016] - Correct filter heuristic overflow handling. This was broken when the - write filter code was moved out-of-line; if there is a single filter and - the heuristic sum overflows the calculation of the filtered line is not - completed. In versions prior to 1.6 the code was duplicated in-line - and the check not performed, so the filter operation completed; however, - in the multi-filter case where the sum is performed the 'none' filter would - be selected if all the sums overflowed, even if it wasn't in the filter - list. The fix to the first problem is simply to provide PNG_SIZE_MAX as - the current lmins sum value; this means the sum can never exceed it and - overflows silently. A reasonable compiler that does choose to inline - the code will simply eliminate the sum check. - The fix to the second problem is to use high precision arithmetic (this is - implemented in 1.7), however a simple safe fix here is to chose the lowest - numbered filter in the list from png_set_filter (this only works if the - first problem is also fixed) (John Bowler). - Use a more efficient absolute value calculation on SSE2 (Matthieu Darbois). - Fixed the case where PNG_IMAGE_BUFFER_SIZE can overflow in the application - as a result of the application using an increased 'row_stride'; previously - png_image_finish_read only checked for overflow on the base calculation of - components. (I.e. it checked for overflow of a 32-bit number on the total - number of pixel components in the output format, not the possibly padded row - length and not the number of bytes, which for linear formats is twice the - number of components.) - MSVC does not like '-(unsigned)', so replaced it with 0U-(unsigned) - MSVC does not like (uInt) = -(unsigned) (i.e. as an initializer), unless - the conversion is explicitly invoked by a cast. - Put the SKIP definition in the correct place. It needs to come after the - png.h include (see all the other .c files in contrib/libtests) because it - depends on PNG_LIBPNG_VER. - Removed the three compile warning options from the individual project - files into the zlib.props globals. It increases the warning level from 4 - to All and adds a list of the warnings that need to be turned off. This is - semi-documentary; the intent is to tell libpng users which warnings have - been examined and judged non-fixable at present. The warning about - structure padding is fixable, but it would be a significant change (moving - structure members around). - -Version 1.6.24beta03 [July 4, 2016] - Optimized absolute value calculation in filter selection, similar to - code in the PAETH decoder in pngrutil.c. Build with PNG_USE_ABS to - use this. - Added pngcp to the build together with a pngcp.dfa configuration test. - Added high resolution timing to pngcp. - Added "Common linking failures" section to INSTALL. - Relocated misplaced #endif in png.c sRGB profile checking. - Fixed two Coverity issues in pngcp.c. - -Version 1.6.24beta04 [July 8, 2016] - Avoid filter-selection heuristic sum calculations in cases where only one - filter is a candidate for selection. This trades off code size (added - private png_setup_*_row_only() functions) for speed. - -Version 1.6.24beta05 [July 13, 2016] - Fixed some indentation to comply with our coding style. - Added contrib/tools/reindent. - -Version 1.6.24beta06 [July 18, 2016] - Fixed more indentation to comply with our coding style. - Eliminated unnecessary tests of boolean png_isaligned() vs 0. - -Version 1.6.24rc01 [July 25, 2016] - No changes. - -Version 1.6.24rc02 [August 1, 2016] - Conditionally compile SSE2 headers in contrib/intel/intel_sse.patch - Conditionally compile png_decompress_chunk(). - -Version 1.6.24rc03 [August 2, 2016] - Conditionally compile ARM_NEON headers in pngpriv.h - Updated contrib/intel/intel_sse.patch - -Version 1.6.24[August 4, 2016] - No changes. - -Version 1.6.25beta01 [August 12, 2016] - Reject oversized iCCP profile immediately. - Cleaned up PNG_DEBUG compile of pngtest.c. - Conditionally compile png_inflate(). - -Version 1.6.25beta02 [August 18, 2016] - Don't install pngcp; it conflicts with pngcp in the pngtools package. - Minor editing of INSTALL, (whitespace, added copyright line) - -Version 1.6.25rc01 [August 24, 2016] - No changes. - -Version 1.6.25rc02 [August 29, 2016] - Added MIPS support (Mandar Sahastrabuddhe ). - Only the UP filter is currently implemented. - -Version 1.6.25rc03 [August 29, 2016] - Rebased contrib/intel/intel_sse.patch after the MIPS implementation. - -Version 1.6.25rc04 [August 30, 2016] - Added MIPS support for SUB, AVG, and PAETH filters (Mandar Sahastrabuddhe). - -Version 1.6.25rc05 [August 30, 2016] - Rebased contrib/intel/intel_sse.patch after the MIPS implementation update.. - -Version 1.6.25 [September 1, 2016] - No changes. - -Version 1.6.26beta01 [September 26, 2016] - Fixed handling zero length IDAT in pngfix (bug report by Agostino Sarubbo, - bugfix by John Bowler). - Do not issue a png_error() on read in png_set_pCAL() because png_handle_pCAL - has allocated memory that libpng needs to free. - Conditionally compile png_set_benign_errors() in pngread.c and pngtest.c - Issue a png_benign_error instead of a png_error on ADLER32 mismatch - while decoding compressed data chunks. - Changed PNG_ZLIB_VERNUM to ZLIB_VERNUM in pngpriv.h, pngstruct.h, and - pngrutil.c. - If CRC handling of critical chunks has been set to PNG_CRC_QUIET_USE, - ignore the ADLER32 checksum in the IDAT chunk as well as the chunk CRCs. - Issue png_benign_error() on ADLER32 checksum mismatch instead of png_error(). - Add tests/badcrc.png and tests/badadler.png to tests/pngtest. - Merged pngtest.c with libpng-1.7.0beta84/pngtest.c - -Version 1.6.26beta02 [October 1, 2016] - Updated the documentation about CRC and ADLER32 handling. - Quieted 117 warnings from clang-3.8 in pngtrans.c, pngread.c, - pngwrite.c, pngunknown.c, and pngvalid.c. - Quieted 58 (out of 144) -Wconversion compiler warnings by changing - flag definitions in pngpriv.h from 0xnnnn to 0xnnnnU and trivial changes - in png.c, pngread.c, and pngwutil.c. - -Version 1.6.26beta03 [October 2, 2016] - Removed contrib/libtests/*.orig and *.rej that slipped into the tarballs. - Quieted the 86 remaining -Wconversion compiler warnings by - revising the png_isaligned() macro and trivial changes in png.c, - pngerror.c, pngget.c, pngmem.c, pngset.c, pngrtran.c, pngrutil.c, - pngwtran.c, pngwrite.c, and pngwutil.c. - -Version 1.6.26beta04 [October 3, 2016] - Quieted (bogus?) clang warnings about "absolute value has no effect" - when PNG_USE_ABS is defined. - Fixed offsets in contrib/intel/intel_sse.patch - -Version 1.6.26beta05 [October 6, 2016] - Changed integer constant 4294967294 to unsigned 4294967294U in pngconf.h - to avoid a signed/unsigned compare in the preprocessor. - -Version 1.6.26beta06 [October 7, 2016] - Use zlib-1.2.8.1 inflateValidate() instead of inflateReset2() to - optionally avoid ADLER32 evaluation. - -Version 1.6.26rc01 [October 12, 2016] - No changes. - -Version 1.6.26 [October 20, 2016] - Cosmetic change, "ptr != 0" to "ptr != NULL" in png.c and pngrutil.c - Despammed email addresses (replaced "@" with " at "). - -Version 1.6.27beta01 [November 2, 2016] - Restrict the new ADLER32-skipping to IDAT chunks. It broke iCCP chunk - handling: an erroneous iCCP chunk would throw a png_error and reject the - entire PNG image instead of rejecting just the iCCP chunk with a warning, - if built with zlib-1.2.8.1. - -Version 1.6.27rc01 [December 27, 2016] - Control ADLER32 checking with new PNG_IGNORE_ADLER32 option. Fixes - an endless loop when handling erroneous ADLER32 checksums; bug - introduced in libpng-1.6.26. - Removed the use of a macro containing the pre-processor 'defined' - operator. It is unclear whether this is valid; a macro that - "generates" 'defined' is not permitted, but the use of the word - "generates" within the C90 standard seems to imply more than simple - substitution of an expression itself containing a well-formed defined - operation. - Added ARM support to CMakeLists.txt (Andreas Franek). - -Version 1.6.27 [December 29, 2016] - Fixed a potential null pointer dereference in png_set_text_2() (bug report - and patch by Patrick Keshishian, CVE-2016-10087). - -Version 1.6.28rc01 [January 3, 2017] - Fixed arm/aarch64 detection in CMakeLists.txt (Gianfranco Costamagna). - Added option to Cmake build allowing a custom location of zlib to be - specified in a scenario where libpng is being built as a subproject - alongside zlib by another project (Sam Serrels). - Changed png_ptr->options from a png_byte to png_uint_32, to accommodate - up to 16 options. - -Version 1.6.28rc02 [January 4, 2017] - Added "include(GNUInstallDirs)" to CMakeLists.txt (Gianfranco Costamagna). - Moved SSE2 optimization code into the main libpng source directory. - Configure libpng with "configure --enable-intel-sse" or compile - libpng with "-DPNG_INTEL_SSE" in CPPFLAGS to enable it. - -Version 1.6.28rc03 [January 4, 2017] - Backed out the SSE optimization and last CMakeLists.txt to allow time for QA. - -Version 1.6.28 [January 5, 2017] - No changes. - -Version 1.6.29beta01 [January 12, 2017] - Readded "include(GNUInstallDirs)" to CMakeLists.txt (Gianfranco Costamagna). - Moved SSE2 optimization code into the main libpng source directory. - Configure libpng with "configure --enable-intel-sse" or compile - libpng with "-DPNG_INTEL_SSE" in CPPFLAGS to enable it. - Simplified conditional compilation in pngvalid.c, for AIX (Michael Felt). - -Version 1.6.29beta02 [February 22, 2017] - Avoid conditional directives that break statements in pngrutil.c (Romero - Malaquias) - The contrib/examples/pngtopng.c recovery code was in the wrong "if" - branches; the comments were correct. - Added code for PowerPC VSX optimisation (Vadim Barkov). - -Version 1.6.29beta03 [March 1, 2017] - Avoid potential overflow of shift operations in png_do_expand() (Aaron Boxer). - Change test ZLIB_VERNUM >= 0x1281 to ZLIB_VERNUM >= 0x1290 in pngrutil.c - because Solaris 11 distributes zlib-1.2.8.f that is older than 1.2.8.1, - as suggested in zlib FAQ, item 24. - Suppress clang warnings about implicit sign changes in png.c - -Version 1.6.29 [March 16, 2017] - No changes. - -Version 1.6.30beta01 [April 1, 2017] - Added missing "$(CPPFLAGS)" to the compile line for c.pic.o in - makefile.linux and makefile.solaris-x86 (Cosmin). - Revised documentation of png_get_error_ptr() in the libpng manual. - Silence clang -Wcomma and const drop warnings (Viktor Szakats). - Update Sourceforge URLs in documentation (https instead of http). - -Version 1.6.30beta02 [April 22, 2017] - Document need to check for integer overflow when allocating a pixel - buffer for multiple rows in contrib/gregbook, contrib/pngminus, - example.c, and in the manual (suggested by Jaeseung Choi). This - is similar to the bug reported against pngquant in CVE-2016-5735. - Removed reference to the obsolete PNG_SAFE_LIMITS macro in the documentation. - -Version 1.6.30beta03 [May 22, 2017] - Check for integer overflow in contrib/visupng and contrib/tools/genpng. - Do not double evaluate CMAKE_SYSTEM_PROCESSOR in CMakeLists.txt. - Test CMAKE_HOST_WIN32 instead of WIN32 in CMakeLists.txt. - Fix some URL in documentation. - -Version 1.6.30beta04 [June 7, 2017] - Avoid writing an empty IDAT when the last IDAT exactly fills the - compression buffer (bug report by Brian Baird). This bug was - introduced in libpng-1.6.0. - -Version 1.6.30rc01 [June 14, 2017] - No changes. - -Version 1.6.30rc02 [June 25, 2017] - Update copyright year in pnglibconf.h, make ltmain.sh executable. - Add a reference to the libpng.download site in README. - -Version 1.6.30 [June 28, 2017] - No changes. - -Version 1.6.31beta01 [July 5, 2017] - Guard the definition of _POSIX_SOURCE in pngpriv.h (AIX already defines it; - bug report by Michael Felt). - Revised pngpriv.h to work around failure to compile arm/filter_neon.S - ("typedef" directive is unrecognized by the assembler). The problem - was introduced in libpng-1.6.30beta01. - Added "Requires: zlib" to libpng.pc.in (Pieter Neerincx). - Added special case for FreeBSD in arm/filter_neon.S (Maya Rashish). - -Version 1.6.31beta02 [July 8, 2017] - Added instructions for disabling hardware optimizations in INSTALL. - Added "--enable-hardware-optimizations" configuration flag to enable - or disable all hardware optimizations with one flag. - -Version 1.6.31beta03 [July 9, 2017] - Updated CMakeLists.txt to add INTEL_SSE and MIPS_MSA platforms. - Changed "int" to "png_size_t" in intel/filter_sse2.c to prevent - possible integer overflow (Bug report by John Bowler). - Quieted "declaration after statement" warnings in intel/filter_sse2.c. - Added scripts/makefile-linux-opt, which has hardware optimizations enabled. - -Version 1.6.31beta04 [July 11, 2017] - Removed one of the GCC-7.1.0 'strict-overflow' warnings that result when - integers appear on both sides of a compare. Worked around the others by - forcing the strict-overflow setting in the relevant functions to a level - where they are not reported (John Bowler). - Changed "FALL THROUGH" comments to "FALLTHROUGH" because GCC doesn't like - the space. - Worked around some C-style casts from (void*) because g++ 5.4.0 objects - to them. - Increased the buffer size for 'sprint' to pass the gcc 7.1.0 'sprint - overflow' check that is on by default with -Wall -Wextra. - -Version 1.6.31beta05 [July 13, 2017] - Added eXIf chunk support. - -Version 1.6.31beta06 [July 17, 2017] - Added a minimal eXIf chunk (with Orientation and FocalLengthIn35mmFilm - tags) to pngtest.png. - -Version 1.6.31beta07 [July 18, 2017] - Revised the eXIf chunk in pngtest.png to fix "Bad IFD1 Directory" warning. - -Version 1.6.31rc01 [July 19, 2017] - No changes. - -Version 1.6.31rc02 [July 25, 2017] - Fixed typo in example.c (png_free_image should be png_image_free) (Bug - report by John Smith) - -Version 1.6.31 [July 27, 2017] - No changes. - -Version 1.6.32beta01 [July 31, 2017] - Avoid possible NULL dereference in png_handle_eXIf when benign_errors - are allowed. Avoid leaking the input buffer "eXIf_buf". - Eliminated png_ptr->num_exif member from pngstruct.h and added num_exif - to arguments for png_get_eXIf() and png_set_eXIf(). - Added calls to png_handle_eXIf(() in pngread.c and png_write_eXIf() in - pngwrite.c, and made various other fixes to png_write_eXIf(). - Changed name of png_get_eXIF and png_set_eXIf() to png_get_eXIf_1() and - png_set_eXIf_1(), respectively, to avoid breaking API compatibility - with libpng-1.6.31. - -Version 1.6.32beta02 [August 1, 2017] - Updated contrib/libtests/pngunknown.c with eXIf chunk. - -Version 1.6.32beta03 [August 2, 2017] - Initialized btoa[] in pngstest.c - Stop memory leak when returning from png_handle_eXIf() with an error - (Bug report from the OSS-fuzz project). - -Version 1.6.32beta04 [August 2, 2017] - Replaced local eXIf_buf with info_ptr->eXIf_buf in png_handle_eXIf(). - Update libpng.3 and libpng-manual.txt about eXIf functions. - -Version 1.6.32beta05 [August 2, 2017] - Restored png_get_eXIf() and png_set_eXIf() to maintain API compatibility. - -Version 1.6.32beta06 [August 2, 2017] - Removed png_get_eXIf_1() and png_set_eXIf_1(). - -Version 1.6.32beta07 [August 3, 2017] - Check length of all chunks except IDAT against user limit to fix an - OSS-fuzz issue (Fixes CVE-2017-12652). - -Version 1.6.32beta08 [August 3, 2017] - Check length of IDAT against maximum possible IDAT size, accounting - for height, rowbytes, interlacing and zlib/deflate overhead. - Restored png_get_eXIf_1() and png_set_eXIf_1(), because strlen(eXIf_buf) - does not work (the eXIf chunk data can contain zeroes). - -Version 1.6.32beta09 [August 3, 2017] - Require cmake-2.8.8 in CMakeLists.txt. Revised symlink creation, - no longer using deprecated cmake LOCATION feature (Clifford Yapp). - Fixed five-byte error in the calculation of IDAT maximum possible size. - -Version 1.6.32beta10 [August 5, 2017] - Moved chunk-length check into a png_check_chunk_length() private - function (Suggested by Max Stepin). - Moved bad pngs from tests to contrib/libtests/crashers - Moved testing of bad pngs into a separate tests/pngtest-badpngs script - Added the --xfail (expected FAIL) option to pngtest.c. It writes XFAIL - in the output but PASS for the libpng test. - Require cmake-3.0.2 in CMakeLists.txt (Clifford Yapp). - Fix "const" declaration info_ptr argument to png_get_eXIf_1() and the - num_exif argument to png_get_eXIf_1() (Github Issue 171). - -Version 1.6.32beta11 [August 7, 2017] - Added "eXIf" to "chunks_to_ignore[]" in png_set_keep_unknown_chunks(). - Added huge_IDAT.png and empty_ancillary_chunks.png to testpngs/crashers. - Make pngtest --strict, --relax, --xfail options imply -m (multiple). - Removed unused chunk_name parameter from png_check_chunk_length(). - Relocated setting free_me for eXIf data, to stop an OSS-fuzz leak. - Initialize profile_header[] in png_handle_iCCP() to fix OSS-fuzz issue. - Initialize png_ptr->row_buf[0] to 255 in png_read_row() to fix OSS-fuzz UMR. - Attempt to fix a UMR in png_set_text_2() to fix OSS-fuzz issue. - Increase minimum zlib stream from 9 to 14 in png_handle_iCCP(), to account - for the minimum 'deflate' stream, and relocate the test to a point - after the keyword has been read. - Check that the eXIf chunk has at least 2 bytes and begins with "II" or "MM". - -Version 1.6.32rc01 [August 18, 2017] - Added a set of "huge_xxxx_chunk.png" files to contrib/testpngs/crashers, - one for each known chunk type, with length = 2GB-1. - Check for 0 return from png_get_rowbytes() and added some (size_t) typecasts - in contrib/pngminus/*.c to stop some Coverity issues (162705, 162706, - and 162707). - Renamed chunks in contrib/testpngs/crashers to avoid having files whose - names differ only in case; this causes problems with some platforms - (github issue #172). - -Version 1.6.32rc02 [August 22, 2017] - Added contrib/oss-fuzz directory which contains files used by the oss-fuzz - project (https://github.com/google/oss-fuzz/tree/master/projects/libpng). - -Version 1.6.32 [August 24, 2017] - No changes. - -Version 1.6.33beta01 [August 28, 2017] - Added PNGMINUS_UNUSED macro to contrib/pngminus/p*.c and added missing - parenthesis in contrib/pngminus/pnm2png.c (bug report by Christian Hesse). - Fixed off-by-one error in png_do_check_palette_indexes() (Bug report - by Mick P., Source Forge Issue #269). - -Version 1.6.33beta02 [September 3, 2017] - Initialize png_handler.row_ptr in contrib/oss-fuzz/libpng_read_fuzzer.cc - to fix shortlived oss-fuzz issue 3234. - Compute a larger limit on IDAT because some applications write a deflate - buffer for each row (Bug report by Andrew Church). - Use current date (DATE) instead of release-date (RDATE) in last - changed date of contrib/oss-fuzz files. - Enabled ARM support in CMakeLists.txt (Bernd Kuhls). - -Version 1.6.33beta03 [September 14, 2017] - Fixed incorrect typecast of some arguments to png_malloc() and - png_calloc() that were png_uint_32 instead of png_alloc_size_t - (Bug report by "irwir" in Github libpng issue #175). - Use pnglibconf.h.prebuilt when building for ANDROID with cmake (Github - issue 162, by rcdailey). - -Version 1.6.33rc01 [September 20, 2017] - Initialize memory allocated by png_inflate to zero, using memset, to - stop an oss-fuzz "use of uninitialized value" detection in png_set_text_2() - due to truncated iTXt or zTXt chunk. - Initialize memory allocated by png_read_buffer to zero, using memset, to - stop an oss-fuzz "use of uninitialized value" detection in - png_icc_check_tag_table() due to truncated iCCP chunk. - Removed a redundant test (suggested by "irwir" in Github issue #180). - -Version 1.6.33rc02 [September 23, 2017] - Added an interlaced version of each file in contrib/pngsuite. - Relocate new memset() call in pngrutil.c. - Removed more redundant tests (suggested by "irwir" in Github issue #180). - Add support for loading images with associated alpha in the Simplified - API (Samuel Williams). - -Version 1.6.33 [September 28, 2017] - Revert contrib/oss-fuzz/libpng_read_fuzzer.cc to libpng-1.6.32 state. - Initialize png_handler.row_ptr in contrib/oss-fuzz/libpng_read_fuzzer.cc - Add end_info structure and png_read_end() to the libpng fuzzer. - -Version 1.6.34 [September 29, 2017] - Removed contrib/pngsuite/i*.png; some of them caused test failures. - -Version 1.6.35beta01 [March 6, 2018] - Restored 21 of the contrib/pngsuite/i*.png, which do not cause test - failures. Placed the remainder in contrib/pngsuite/interlaced/i*.png. - Added calls to png_set_*() transforms commonly used by browsers to - the fuzzer. - Removed some unnecessary brackets in pngrtran.c - Fixed miscellaneous typos (Patch by github user "luzpaz"). - Change "ASM C" to "C ASM" in CMakeLists.txt - Fixed incorrect handling of bKGD chunk in sub-8-bit files (Cosmin) - Added hardware optimization directories to zip and 7z distributions. - Fixed incorrect bitmask for options. - Fixed many spelling typos. - -Version 1.6.35beta02 [March 28, 2018] - Make png_get_iCCP consistent with man page (allow compression-type argument - to be NULL, bug report by Lenard Szolnoki). - -Version 1.6.35 [July 15, 2018] - Replaced the remaining uses of png_size_t with size_t (Cosmin) - Fixed the calculation of row_factor in png_check_chunk_length - (reported by Thuan Pham in SourceForge issue #278) - Added missing parentheses to a macro definition - (suggested by "irwir" in GitHub issue #216) - -Version 1.6.36 [December 1, 2018] - Optimized png_do_expand_palette for ARM processors. - Improved performance by around 10-22% on a recent ARM Chromebook. - (Contributed by Richard Townsend, ARM Holdings) - Fixed manipulation of machine-specific optimization options. - (Contributed by Vicki Pfau) - Used memcpy instead of manual pointer arithmetic on Intel SSE2. - (Contributed by Samuel Williams) - Fixed build errors with MSVC on ARM64. - (Contributed by Zhijie Liang) - Fixed detection of libm in CMakeLists. - (Contributed by Cameron Cawley) - Fixed incorrect creation of pkg-config file in CMakeLists. - (Contributed by Kyle Bentley) - Fixed the CMake build on Windows MSYS by avoiding symlinks. - Fixed a build warning on OpenBSD. - (Contributed by Theo Buehler) - Fixed various typos in comments. - (Contributed by "luz.paz") - Raised the minimum required CMake version from 3.0.2 to 3.1. - Removed yet more of the vestigial support for pre-ANSI C compilers. - Removed ancient makefiles for ancient systems that have been broken - across all previous libpng-1.6.x versions. - Removed the Y2K compliance statement and the export control - information. - Applied various code style and documentation fixes. - -Version 1.6.37 [April 14, 2019] - Fixed a use-after-free vulnerability (CVE-2019-7317) in png_image_free. - Fixed a memory leak in the ARM NEON implementation of png_do_expand_palette. - Fixed a memory leak in pngtest.c. - Fixed two vulnerabilities (CVE-2018-14048, CVE-2018-14550) in - contrib/pngminus; refactor. - Changed the license of contrib/pngminus to MIT; refresh makefile and docs. - (Contributed by Willem van Schaik) - Fixed a typo in the libpng license v2. - (Contributed by Miguel Ojeda) - Added makefiles for AddressSanitizer-enabled builds. - Cleaned up various makefiles. - -Version 1.6.38 [September 14, 2022] - Added configurations and scripts for continuous integration. - Fixed various errors in the handling of tRNS, hIST and eXIf. - Implemented many stability improvements across all platforms. - Updated the internal documentation. - -Version 1.6.39 [November 20, 2022] - Changed the error handler of oversized chunks (i.e. larger than - PNG_USER_CHUNK_MALLOC_MAX) from png_chunk_error to png_benign_error. - Fixed a buffer overflow error in contrib/tools/pngfix. - Fixed a memory leak (CVE-2019-6129) in contrib/tools/pngcp. - Disabled the ARM Neon optimizations by default in the CMake file, - following the default behavior of the configure script. - Allowed configure.ac to work with the trunk version of autoconf. - Removed the support for "install" targets from the legacy makefiles; - removed the obsolete makefile.cegcc. - Cleaned up the code and updated the internal documentation. - -Version 1.6.40 [June 21, 2023] - Fixed the eXIf chunk multiplicity checks. - Fixed a memory leak in pCAL processing. - Corrected the validity report about tRNS inside png_get_valid(). - Fixed various build issues on *BSD, Mac and Windows. - Updated the configurations and the scripts for continuous integration. - Cleaned up the code, the build scripts, and the documentation. - -Version 1.6.41 [January 24, 2024] - Added SIMD-optimized code for the LoongArch LSX hardware. - (Contributed by GuXiWei, JinBo and ZhangLixia) - Fixed the run-time discovery of MIPS MSA hardware. - (Contributed by Sui Jingfeng) - Fixed an off-by-one error in the function png_do_check_palette_indexes(), - which failed to recognize errors that might have existed in the first - column of a broken palette-encoded image. This was a benign regression - accidentally introduced in libpng-1.6.33. No pixel was harmed. - (Contributed by Adam Richter; reviewed by John Bowler) - Fixed, improved and modernized the contrib/pngminus programs, i.e., - png2pnm.c and pnm2png.c - Removed old and peculiar portability hacks that were meant to silence - warnings issued by gcc version 7.1 alone. - (Contributed by John Bowler) - Fixed and modernized the CMake file, and raised the minimum required - CMake version from 3.1 to 3.6. - (Contributed by Clinton Ingram, Timothy Lyanguzov, Tyler Kropp, et al.) - Allowed the configure script to disable the building of auxiliary tools - and tests, thus catching up with the CMake file. - (Contributed by Carlo Bramini) - Fixed a build issue on Mac. - (Contributed by Zixu Wang) - Moved the Autoconf macro files to scripts/autoconf. - Moved the CMake files (except for the main CMakeLists.txt) to - scripts/cmake and moved the list of their contributing authors to - scripts/cmake/AUTHORS.md - Updated the CI configurations and scripts. - Relicensed the CI scripts to the MIT License. - Improved the test coverage. - (Contributed by John Bowler) - -Version 1.6.42 [January 29, 2024] - Fixed the implementation of the macro function png_check_sig(). - This was an API regression, introduced in libpng-1.6.41. - (Reported by Matthieu Darbois) - Fixed and updated the libpng manual. - -Version 1.6.43 [February 23, 2024] - Fixed the row width check in png_check_IHDR(). - This corrected a bug that was specific to the 16-bit platforms, - and removed a spurious compiler warning from the 64-bit builds. - (Reported by Jacek Caban; fixed by John Bowler) - Added eXIf chunk support to the push-mode reader in pngpread.c. - (Contributed by Chris Blume) - Added contrib/pngexif for the benefit of the users who would like - to inspect the content of eXIf chunks. - Added contrib/conftest/basic.dfa, a basic build-time configuration. - (Contributed by John Bowler) - Fixed a preprocessor condition in pngread.c that broke build-time - configurations like contrib/conftest/pngcp.dfa. - (Contributed by John Bowler) - Added CMake build support for LoongArch LSX. - (Contributed by GuXiWei) - Fixed a CMake build error that occurred under a peculiar state of the - dependency tree. This was a regression introduced in libpng-1.6.41. - (Contributed by Dan Rosser) - Marked the installed libpng headers as system headers in CMake. - (Contributed by Benjamin Buch) - Updated the build support for RISCOS. - (Contributed by Cameron Cawley) - Updated the makefiles to allow cross-platform builds to initialize - conventional make variables like AR and ARFLAGS. - Added various improvements to the CI scripts in areas like version - consistency verification and text linting. - Added version consistency verification to pngtest.c also. - -Send comments/corrections/commendations to png-mng-implement at lists.sf.net. -Subscription is required; visit -https://lists.sourceforge.net/lists/listinfo/png-mng-implement -to subscribe. diff --git a/dep/libpng/CMakeLists.txt b/dep/libpng/CMakeLists.txt deleted file mode 100644 index 82b01bfb8..000000000 --- a/dep/libpng/CMakeLists.txt +++ /dev/null @@ -1,39 +0,0 @@ -add_library(png - src/png.c - src/pngerror.c - src/pngget.c - src/pngmem.c - src/pngpread.c - src/pngread.c - src/pngrio.c - src/pngrtran.c - src/pngrutil.c - src/pngset.c - src/pngtrans.c - src/pngwio.c - src/pngwrite.c - src/pngwtran.c - src/pngwutil.c -) - -if(CPU_ARCH_X64) - target_sources(png PRIVATE - src/intel/filter_sse2_intrinsics.c - src/intel/intel_init.c - ) - target_compile_definitions(png PRIVATE "PNG_INTEL_SSE") -elseif(CPU_ARCH_ARM32 OR CPU_ARCH_ARM64) - target_sources(png PRIVATE - src/arm/arm_init.c - src/arm/filter_neon.S - src/arm/filter_neon_intrinsics.c - src/arm/palette_neon_intrinsics.c - ) -endif() - -target_link_libraries(png PRIVATE ZLIB::ZLIB) -target_include_directories(png PUBLIC "${CMAKE_CURRENT_SOURCE_DIR}/include") -target_include_directories(png PRIVATE "${CMAKE_CURRENT_SOURCE_DIR}/src") -disable_compiler_warnings_for_target(png) - -add_library(PNG::PNG ALIAS png) diff --git a/dep/libpng/LICENSE b/dep/libpng/LICENSE deleted file mode 100644 index 25f298f0f..000000000 --- a/dep/libpng/LICENSE +++ /dev/null @@ -1,134 +0,0 @@ -COPYRIGHT NOTICE, DISCLAIMER, and LICENSE -========================================= - -PNG Reference Library License version 2 ---------------------------------------- - - * Copyright (c) 1995-2024 The PNG Reference Library Authors. - * Copyright (c) 2018-2024 Cosmin Truta. - * Copyright (c) 2000-2002, 2004, 2006-2018 Glenn Randers-Pehrson. - * Copyright (c) 1996-1997 Andreas Dilger. - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - -The software is supplied "as is", without warranty of any kind, -express or implied, including, without limitation, the warranties -of merchantability, fitness for a particular purpose, title, and -non-infringement. In no event shall the Copyright owners, or -anyone distributing the software, be liable for any damages or -other liability, whether in contract, tort or otherwise, arising -from, out of, or in connection with the software, or the use or -other dealings in the software, even if advised of the possibility -of such damage. - -Permission is hereby granted to use, copy, modify, and distribute -this software, or portions hereof, for any purpose, without fee, -subject to the following restrictions: - - 1. The origin of this software must not be misrepresented; you - must not claim that you wrote the original software. If you - use this software in a product, an acknowledgment in the product - documentation would be appreciated, but is not required. - - 2. Altered source versions must be plainly marked as such, and must - not be misrepresented as being the original software. - - 3. This Copyright notice may not be removed or altered from any - source or altered source distribution. - - -PNG Reference Library License version 1 (for libpng 0.5 through 1.6.35) ------------------------------------------------------------------------ - -libpng versions 1.0.7, July 1, 2000, through 1.6.35, July 15, 2018 are -Copyright (c) 2000-2002, 2004, 2006-2018 Glenn Randers-Pehrson, are -derived from libpng-1.0.6, and are distributed according to the same -disclaimer and license as libpng-1.0.6 with the following individuals -added to the list of Contributing Authors: - - Simon-Pierre Cadieux - Eric S. Raymond - Mans Rullgard - Cosmin Truta - Gilles Vollant - James Yu - Mandar Sahastrabuddhe - Google Inc. - Vadim Barkov - -and with the following additions to the disclaimer: - - There is no warranty against interference with your enjoyment of - the library or against infringement. There is no warranty that our - efforts or the library will fulfill any of your particular purposes - or needs. This library is provided with all faults, and the entire - risk of satisfactory quality, performance, accuracy, and effort is - with the user. - -Some files in the "contrib" directory and some configure-generated -files that are distributed with libpng have other copyright owners, and -are released under other open source licenses. - -libpng versions 0.97, January 1998, through 1.0.6, March 20, 2000, are -Copyright (c) 1998-2000 Glenn Randers-Pehrson, are derived from -libpng-0.96, and are distributed according to the same disclaimer and -license as libpng-0.96, with the following individuals added to the -list of Contributing Authors: - - Tom Lane - Glenn Randers-Pehrson - Willem van Schaik - -libpng versions 0.89, June 1996, through 0.96, May 1997, are -Copyright (c) 1996-1997 Andreas Dilger, are derived from libpng-0.88, -and are distributed according to the same disclaimer and license as -libpng-0.88, with the following individuals added to the list of -Contributing Authors: - - John Bowler - Kevin Bracey - Sam Bushell - Magnus Holmgren - Greg Roelofs - Tom Tanner - -Some files in the "scripts" directory have other copyright owners, -but are released under this license. - -libpng versions 0.5, May 1995, through 0.88, January 1996, are -Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - -For the purposes of this copyright and license, "Contributing Authors" -is defined as the following set of individuals: - - Andreas Dilger - Dave Martindale - Guy Eric Schalnat - Paul Schmidt - Tim Wegner - -The PNG Reference Library is supplied "AS IS". The Contributing -Authors and Group 42, Inc. disclaim all warranties, expressed or -implied, including, without limitation, the warranties of -merchantability and of fitness for any purpose. The Contributing -Authors and Group 42, Inc. assume no liability for direct, indirect, -incidental, special, exemplary, or consequential damages, which may -result from the use of the PNG Reference Library, even if advised of -the possibility of such damage. - -Permission is hereby granted to use, copy, modify, and distribute this -source code, or portions hereof, for any purpose, without fee, subject -to the following restrictions: - - 1. The origin of this source code must not be misrepresented. - - 2. Altered versions must be plainly marked as such and must not - be misrepresented as being the original source. - - 3. This Copyright notice may not be removed or altered from any - source or altered source distribution. - -The Contributing Authors and Group 42, Inc. specifically permit, -without fee, and encourage the use of this source code as a component -to supporting the PNG file format in commercial products. If you use -this source code in a product, acknowledgment is not required but would -be appreciated. diff --git a/dep/libpng/README b/dep/libpng/README deleted file mode 100644 index a6ca3ae9f..000000000 --- a/dep/libpng/README +++ /dev/null @@ -1,184 +0,0 @@ -README for libpng version 1.6.43 -================================ - -See the note about version numbers near the top of `png.h`. -See `INSTALL` for instructions on how to install libpng. - -Libpng comes in several distribution formats. Get `libpng-*.tar.gz` -or `libpng-*.tar.xz` if you want UNIX-style line endings in the text -files, or `lpng*.7z` or `lpng*.zip` if you want DOS-style line endings. - -For a detailed description on using libpng, read `libpng-manual.txt`. -For examples of libpng in a program, see `example.c` and `pngtest.c`. -For usage information and restrictions (what little they are) on libpng, -see `png.h`. For a description on using zlib (the compression library -used by libpng) and zlib's restrictions, see `zlib.h`. - -You should use zlib 1.0.4 or later to run this, but it _may_ work with -versions as old as zlib 0.95. Even so, there are bugs in older zlib -versions which can cause the output of invalid compression streams for -some images. - -You should also note that zlib is a compression library that is useful -for more things than just PNG files. You can use zlib as a drop-in -replacement for `fread()` and `fwrite()`, if you are so inclined. - -zlib should be available at the same place that libpng is, or at -https://zlib.net . - -You may also want a copy of the PNG specification. It is available -as an RFC, a W3C Recommendation, and an ISO/IEC Standard. You can find -these at http://www.libpng.org/pub/png/pngdocs.html . - -This code is currently being archived at https://libpng.sourceforge.io -in the download area, and at http://libpng.download/src . - -This release, based in a large way on Glenn's, Guy's and Andreas' -earlier work, was created and will be supported by myself and the PNG -development group. - -Send comments, corrections and commendations to `png-mng-implement` -at `lists.sourceforge.net`. (Subscription is required; visit -https://lists.sourceforge.net/lists/listinfo/png-mng-implement -to subscribe.) - -Send general questions about the PNG specification to `png-mng-misc` -at `lists.sourceforge.net`. (Subscription is required; visit -https://lists.sourceforge.net/lists/listinfo/png-mng-misc -to subscribe.) - -Historical notes ----------------- - -The libpng library has been in extensive use and testing since mid-1995. -Version 0.89, published a year later, was the first official release. -By late 1997, it had finally gotten to the stage where there hadn't -been significant changes to the API in some time, and people have a bad -feeling about libraries with versions below 1.0. Version 1.0.0 was -released in March 1998. - -Note that some of the changes to the `png_info` structure render this -version of the library binary incompatible with libpng-0.89 or -earlier versions if you are using a shared library. The type of the -`filler` parameter for `png_set_filler()` has changed from `png_byte` -to `png_uint_32`, which will affect shared-library applications that -use this function. - -To avoid problems with changes to the internals of the `info_struct`, -new APIs have been made available in 0.95 to avoid direct application -access to `info_ptr`. These functions are the `png_set_` and -`png_get_` functions. These functions should be used when -accessing/storing the `info_struct` data, rather than manipulating it -directly, to avoid such problems in the future. - -It is important to note that the APIs did not make current programs -that access the info struct directly incompatible with the new -library, through libpng-1.2.x. In libpng-1.4.x, which was meant to -be a transitional release, members of the `png_struct` and the -`info_struct` can still be accessed, but the compiler will issue a -warning about deprecated usage. Since libpng-1.5.0, direct access -to these structs is not allowed, and the definitions of the structs -reside in private `pngstruct.h` and `pnginfo.h` header files that are -not accessible to applications. It is strongly suggested that new -programs use the new APIs (as shown in `example.c` and `pngtest.c`), -and older programs be converted to the new format, to facilitate -upgrades in the future. - -The additions since 0.89 include the ability to read from a PNG stream -which has had some (or all) of the signature bytes read by the calling -application. This also allows the reading of embedded PNG streams that -do not have the PNG file signature. As well, it is now possible to set -the library action on the detection of chunk CRC errors. It is possible -to set different actions based on whether the CRC error occurred in a -critical or an ancillary chunk. - -The additions since 0.90 include the ability to compile libpng as a -Windows DLL, and new APIs for accessing data in the `info_struct`. -Experimental functions included the ability to set weighting and cost -factors for row filter selection, direct reads of integers from buffers -on big-endian processors that support misaligned data access, faster -methods of doing alpha composition, and more accurate 16-to-8 bit color -conversion. Some of these experimental functions, such as the weighted -filter heuristics, have since been removed. - -Files included in this distribution ------------------------------------ - - ANNOUNCE => Announcement of this version, with recent changes - AUTHORS => List of contributing authors - CHANGES => Description of changes between libpng versions - INSTALL => Instructions to install libpng - LICENSE => License to use and redistribute libpng - README => This file - TODO => Things not implemented in the current library - TRADEMARK => Trademark information - example.c => Example code for using libpng functions - libpng.3 => Manual page for libpng (includes libpng-manual.txt) - libpng-manual.txt => Description of libpng and its functions - libpngpf.3 => Manual page for libpng's private functions (deprecated) - png.5 => Manual page for the PNG format - png.c => Basic interface functions common to library - png.h => Library function and interface declarations (public) - pngpriv.h => Library function and interface declarations (private) - pngconf.h => System specific library configuration (public) - pngstruct.h => png_struct declaration (private) - pnginfo.h => png_info struct declaration (private) - pngdebug.h => debugging macros (private) - pngerror.c => Error/warning message I/O functions - pngget.c => Functions for retrieving info from struct - pngmem.c => Memory handling functions - pngbar.png => PNG logo, 88x31 - pngnow.png => PNG logo, 98x31 - pngpread.c => Progressive reading functions - pngread.c => Read data/helper high-level functions - pngrio.c => Lowest-level data read I/O functions - pngrtran.c => Read data transformation functions - pngrutil.c => Read data utility functions - pngset.c => Functions for storing data into the info_struct - pngtest.c => Library test program - pngtest.png => Library test sample image - pngtrans.c => Common data transformation functions - pngwio.c => Lowest-level write I/O functions - pngwrite.c => High-level write functions - pngwtran.c => Write data transformations - pngwutil.c => Write utility functions - arm/ => Optimized code for ARM Neon - intel/ => Optimized code for INTEL SSE2 - loongarch/ => Optimized code for LoongArch LSX - mips/ => Optimized code for MIPS MSA and MIPS MMI - powerpc/ => Optimized code for PowerPC VSX - ci/ => Scripts for continuous integration - contrib/ => External contributions - arm-neon/ => Optimized code for the ARM-NEON platform - mips-msa/ => Optimized code for the MIPS-MSA platform - powerpc-vsx/ => Optimized code for the POWERPC-VSX platform - examples/ => Examples of libpng usage - gregbook/ => Source code for PNG reading and writing, from - "PNG: The Definitive Guide" by Greg Roelofs, - O'Reilly, 1999 - libtests/ => Test programs - oss-fuzz/ => Files used by the OSS-Fuzz project for fuzz-testing - libpng - pngexif/ => Program to inspect the EXIF information in PNG files - pngminim/ => Minimal decoder, encoder, and progressive decoder - programs demonstrating the use of pngusr.dfa - pngminus/ => Simple pnm2png and png2pnm programs - pngsuite/ => Test images - testpngs/ => Test images - tools/ => Various tools - visupng/ => VisualPng, a Windows viewer for PNG images - projects/ => Project files and workspaces for various IDEs - owatcom/ => OpenWatcom project - visualc71/ => Microsoft Visual C++ 7.1 workspace - vstudio/ => Microsoft Visual Studio workspace - scripts/ => Scripts and makefiles for building libpng - (see scripts/README.txt for the complete list) - tests/ => Test scripts - -Good luck, and happy coding! - - * Cosmin Truta (current maintainer, since 2018) - * Glenn Randers-Pehrson (former maintainer, 1998-2018) - * Andreas Eric Dilger (former maintainer, 1996-1997) - * Guy Eric Schalnat (original author and former maintainer, 1995-1996) - (formerly of Group 42, Inc.) diff --git a/dep/libpng/TODO b/dep/libpng/TODO deleted file mode 100644 index 562dab069..000000000 --- a/dep/libpng/TODO +++ /dev/null @@ -1,23 +0,0 @@ -TODO - list of things to do for libpng: - -* Fix all defects (duh!) -* Better C++ wrapper / full C++ implementation (?) -* Fix the problems with C++ and 'extern "C"'. -* cHRM transformation. -* Palette creation. -* "grayscale->palette" transformation and "palette->grayscale" detection. -* Improved dithering. -* Multi-lingual error and warning message support. -* Complete sRGB transformation. (Currently it simply uses gamma=0.45455.) -* Man pages for function calls. -* Better documentation. -* Better filter selection - (e.g., counting huffman bits/precompression; filter inertia; filter costs). -* Histogram creation. -* Text conversion between different code pages (e.g., Latin-1 -> Mac). -* Avoid building gamma tables whenever possible. -* Greater precision in changing to linear gamma for compositing against - background, and in doing rgb-to-gray transformations. -* Investigate pre-incremented loop counters and other loop constructions. -* Interpolated method of handling interlacing. -* More validations for libpng transformations. diff --git a/dep/libpng/TRADEMARK b/dep/libpng/TRADEMARK deleted file mode 100644 index ac667187d..000000000 --- a/dep/libpng/TRADEMARK +++ /dev/null @@ -1,8 +0,0 @@ -TRADEMARK -========= - -The name "libpng" has not been registered by the Copyright owners -as a trademark in any jurisdiction. However, because libpng has -been distributed and maintained world-wide, continually since 1995, -the Copyright owners claim "common-law trademark protection" in any -jurisdiction where common-law trademark is recognized. diff --git a/dep/libpng/include/png.h b/dep/libpng/include/png.h deleted file mode 100644 index 83d390312..000000000 --- a/dep/libpng/include/png.h +++ /dev/null @@ -1,3250 +0,0 @@ - -/* png.h - header file for PNG reference library - * - * libpng version 1.6.43 - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. (See LICENSE, below.) - * - * Authors and maintainers: - * libpng versions 0.71, May 1995, through 0.88, January 1996: Guy Schalnat - * libpng versions 0.89, June 1996, through 0.96, May 1997: Andreas Dilger - * libpng versions 0.97, January 1998, through 1.6.35, July 2018: - * Glenn Randers-Pehrson - * libpng versions 1.6.36, December 2018, through 1.6.43, February 2024: - * Cosmin Truta - * See also "Contributing Authors", below. - */ - -/* - * COPYRIGHT NOTICE, DISCLAIMER, and LICENSE - * ========================================= - * - * PNG Reference Library License version 2 - * --------------------------------------- - * - * * Copyright (c) 1995-2024 The PNG Reference Library Authors. - * * Copyright (c) 2018-2024 Cosmin Truta. - * * Copyright (c) 2000-2002, 2004, 2006-2018 Glenn Randers-Pehrson. - * * Copyright (c) 1996-1997 Andreas Dilger. - * * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * The software is supplied "as is", without warranty of any kind, - * express or implied, including, without limitation, the warranties - * of merchantability, fitness for a particular purpose, title, and - * non-infringement. In no event shall the Copyright owners, or - * anyone distributing the software, be liable for any damages or - * other liability, whether in contract, tort or otherwise, arising - * from, out of, or in connection with the software, or the use or - * other dealings in the software, even if advised of the possibility - * of such damage. - * - * Permission is hereby granted to use, copy, modify, and distribute - * this software, or portions hereof, for any purpose, without fee, - * subject to the following restrictions: - * - * 1. The origin of this software must not be misrepresented; you - * must not claim that you wrote the original software. If you - * use this software in a product, an acknowledgment in the product - * documentation would be appreciated, but is not required. - * - * 2. Altered source versions must be plainly marked as such, and must - * not be misrepresented as being the original software. - * - * 3. This Copyright notice may not be removed or altered from any - * source or altered source distribution. - * - * - * PNG Reference Library License version 1 (for libpng 0.5 through 1.6.35) - * ----------------------------------------------------------------------- - * - * libpng versions 1.0.7, July 1, 2000, through 1.6.35, July 15, 2018 are - * Copyright (c) 2000-2002, 2004, 2006-2018 Glenn Randers-Pehrson, are - * derived from libpng-1.0.6, and are distributed according to the same - * disclaimer and license as libpng-1.0.6 with the following individuals - * added to the list of Contributing Authors: - * - * Simon-Pierre Cadieux - * Eric S. Raymond - * Mans Rullgard - * Cosmin Truta - * Gilles Vollant - * James Yu - * Mandar Sahastrabuddhe - * Google Inc. - * Vadim Barkov - * - * and with the following additions to the disclaimer: - * - * There is no warranty against interference with your enjoyment of - * the library or against infringement. There is no warranty that our - * efforts or the library will fulfill any of your particular purposes - * or needs. This library is provided with all faults, and the entire - * risk of satisfactory quality, performance, accuracy, and effort is - * with the user. - * - * Some files in the "contrib" directory and some configure-generated - * files that are distributed with libpng have other copyright owners, and - * are released under other open source licenses. - * - * libpng versions 0.97, January 1998, through 1.0.6, March 20, 2000, are - * Copyright (c) 1998-2000 Glenn Randers-Pehrson, are derived from - * libpng-0.96, and are distributed according to the same disclaimer and - * license as libpng-0.96, with the following individuals added to the - * list of Contributing Authors: - * - * Tom Lane - * Glenn Randers-Pehrson - * Willem van Schaik - * - * libpng versions 0.89, June 1996, through 0.96, May 1997, are - * Copyright (c) 1996-1997 Andreas Dilger, are derived from libpng-0.88, - * and are distributed according to the same disclaimer and license as - * libpng-0.88, with the following individuals added to the list of - * Contributing Authors: - * - * John Bowler - * Kevin Bracey - * Sam Bushell - * Magnus Holmgren - * Greg Roelofs - * Tom Tanner - * - * Some files in the "scripts" directory have other copyright owners, - * but are released under this license. - * - * libpng versions 0.5, May 1995, through 0.88, January 1996, are - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * For the purposes of this copyright and license, "Contributing Authors" - * is defined as the following set of individuals: - * - * Andreas Dilger - * Dave Martindale - * Guy Eric Schalnat - * Paul Schmidt - * Tim Wegner - * - * The PNG Reference Library is supplied "AS IS". The Contributing - * Authors and Group 42, Inc. disclaim all warranties, expressed or - * implied, including, without limitation, the warranties of - * merchantability and of fitness for any purpose. The Contributing - * Authors and Group 42, Inc. assume no liability for direct, indirect, - * incidental, special, exemplary, or consequential damages, which may - * result from the use of the PNG Reference Library, even if advised of - * the possibility of such damage. - * - * Permission is hereby granted to use, copy, modify, and distribute this - * source code, or portions hereof, for any purpose, without fee, subject - * to the following restrictions: - * - * 1. The origin of this source code must not be misrepresented. - * - * 2. Altered versions must be plainly marked as such and must not - * be misrepresented as being the original source. - * - * 3. This Copyright notice may not be removed or altered from any - * source or altered source distribution. - * - * The Contributing Authors and Group 42, Inc. specifically permit, - * without fee, and encourage the use of this source code as a component - * to supporting the PNG file format in commercial products. If you use - * this source code in a product, acknowledgment is not required but would - * be appreciated. - * - * END OF COPYRIGHT NOTICE, DISCLAIMER, and LICENSE. - * - * TRADEMARK - * ========= - * - * The name "libpng" has not been registered by the Copyright owners - * as a trademark in any jurisdiction. However, because libpng has - * been distributed and maintained world-wide, continually since 1995, - * the Copyright owners claim "common-law trademark protection" in any - * jurisdiction where common-law trademark is recognized. - */ - -/* - * A "png_get_copyright" function is available, for convenient use in "about" - * boxes and the like: - * - * printf("%s", png_get_copyright(NULL)); - * - * Also, the PNG logo (in PNG format, of course) is supplied in the - * files "pngbar.png" and "pngbar.jpg (88x31) and "pngnow.png" (98x31). - */ - -/* - * The contributing authors would like to thank all those who helped - * with testing, bug fixes, and patience. This wouldn't have been - * possible without all of you. - * - * Thanks to Frank J. T. Wojcik for helping with the documentation. - */ - -/* Note about libpng version numbers: - * - * Due to various miscommunications, unforeseen code incompatibilities - * and occasional factors outside the authors' control, version numbering - * on the library has not always been consistent and straightforward. - * The following table summarizes matters since version 0.89c, which was - * the first widely used release: - * - * source png.h png.h shared-lib - * version string int version - * ------- ------ ----- ---------- - * 0.89c "1.0 beta 3" 0.89 89 1.0.89 - * 0.90 "1.0 beta 4" 0.90 90 0.90 [should have been 2.0.90] - * 0.95 "1.0 beta 5" 0.95 95 0.95 [should have been 2.0.95] - * 0.96 "1.0 beta 6" 0.96 96 0.96 [should have been 2.0.96] - * 0.97b "1.00.97 beta 7" 1.00.97 97 1.0.1 [should have been 2.0.97] - * 0.97c 0.97 97 2.0.97 - * 0.98 0.98 98 2.0.98 - * 0.99 0.99 98 2.0.99 - * 0.99a-m 0.99 99 2.0.99 - * 1.00 1.00 100 2.1.0 [100 should be 10000] - * 1.0.0 (from here on, the 100 2.1.0 [100 should be 10000] - * 1.0.1 png.h string is 10001 2.1.0 - * 1.0.1a-e identical to the 10002 from here on, the shared library - * 1.0.2 source version) 10002 is 2.V where V is the source code - * 1.0.2a-b 10003 version, except as noted. - * 1.0.3 10003 - * 1.0.3a-d 10004 - * 1.0.4 10004 - * 1.0.4a-f 10005 - * 1.0.5 (+ 2 patches) 10005 - * 1.0.5a-d 10006 - * 1.0.5e-r 10100 (not source compatible) - * 1.0.5s-v 10006 (not binary compatible) - * 1.0.6 (+ 3 patches) 10006 (still binary incompatible) - * 1.0.6d-f 10007 (still binary incompatible) - * 1.0.6g 10007 - * 1.0.6h 10007 10.6h (testing xy.z so-numbering) - * 1.0.6i 10007 10.6i - * 1.0.6j 10007 2.1.0.6j (incompatible with 1.0.0) - * 1.0.7beta11-14 DLLNUM 10007 2.1.0.7beta11-14 (binary compatible) - * 1.0.7beta15-18 1 10007 2.1.0.7beta15-18 (binary compatible) - * 1.0.7rc1-2 1 10007 2.1.0.7rc1-2 (binary compatible) - * 1.0.7 1 10007 (still compatible) - * ... - * 1.0.69 10 10069 10.so.0.69[.0] - * ... - * 1.2.59 13 10259 12.so.0.59[.0] - * ... - * 1.4.20 14 10420 14.so.0.20[.0] - * ... - * 1.5.30 15 10530 15.so.15.30[.0] - * ... - * 1.6.43 16 10643 16.so.16.43[.0] - * - * Henceforth the source version will match the shared-library major and - * minor numbers; the shared-library major version number will be used for - * changes in backward compatibility, as it is intended. - * The PNG_LIBPNG_VER macro, which is not used within libpng but is - * available for applications, is an unsigned integer of the form XYYZZ - * corresponding to the source version X.Y.Z (leading zeros in Y and Z). - * Beta versions were given the previous public release number plus a - * letter, until version 1.0.6j; from then on they were given the upcoming - * public release number plus "betaNN" or "rcNN". - * - * Binary incompatibility exists only when applications make direct access - * to the info_ptr or png_ptr members through png.h, and the compiled - * application is loaded with a different version of the library. - * - * See libpng.txt or libpng.3 for more information. The PNG specification - * is available as a W3C Recommendation and as an ISO/IEC Standard; see - * - */ - -#ifndef PNG_H -#define PNG_H - -/* This is not the place to learn how to use libpng. The file libpng-manual.txt - * describes how to use libpng, and the file example.c summarizes it - * with some code on which to build. This file is useful for looking - * at the actual function definitions and structure components. If that - * file has been stripped from your copy of libpng, you can find it at - * - * - * If you just need to read a PNG file and don't want to read the documentation - * skip to the end of this file and read the section entitled 'simplified API'. - */ - -/* Version information for png.h - this should match the version in png.c */ -#define PNG_LIBPNG_VER_STRING "1.6.43" -#define PNG_HEADER_VERSION_STRING " libpng version " PNG_LIBPNG_VER_STRING "\n" - -/* The versions of shared library builds should stay in sync, going forward */ -#define PNG_LIBPNG_VER_SHAREDLIB 16 -#define PNG_LIBPNG_VER_SONUM PNG_LIBPNG_VER_SHAREDLIB /* [Deprecated] */ -#define PNG_LIBPNG_VER_DLLNUM PNG_LIBPNG_VER_SHAREDLIB /* [Deprecated] */ - -/* These should match the first 3 components of PNG_LIBPNG_VER_STRING: */ -#define PNG_LIBPNG_VER_MAJOR 1 -#define PNG_LIBPNG_VER_MINOR 6 -#define PNG_LIBPNG_VER_RELEASE 43 - -/* This should be zero for a public release, or non-zero for a - * development version. - */ -#define PNG_LIBPNG_VER_BUILD 0 - -/* Release Status */ -#define PNG_LIBPNG_BUILD_ALPHA 1 -#define PNG_LIBPNG_BUILD_BETA 2 -#define PNG_LIBPNG_BUILD_RC 3 -#define PNG_LIBPNG_BUILD_STABLE 4 -#define PNG_LIBPNG_BUILD_RELEASE_STATUS_MASK 7 - -/* Release-Specific Flags */ -#define PNG_LIBPNG_BUILD_PATCH 8 /* Can be OR'ed with - PNG_LIBPNG_BUILD_STABLE only */ -#define PNG_LIBPNG_BUILD_PRIVATE 16 /* Cannot be OR'ed with - PNG_LIBPNG_BUILD_SPECIAL */ -#define PNG_LIBPNG_BUILD_SPECIAL 32 /* Cannot be OR'ed with - PNG_LIBPNG_BUILD_PRIVATE */ - -#define PNG_LIBPNG_BUILD_BASE_TYPE PNG_LIBPNG_BUILD_STABLE - -/* Careful here. At one time, Guy wanted to use 082, but that - * would be octal. We must not include leading zeros. - * Versions 0.7 through 1.0.0 were in the range 0 to 100 here - * (only version 1.0.0 was mis-numbered 100 instead of 10000). - * From version 1.0.1 it is: - * XXYYZZ, where XX=major, YY=minor, ZZ=release - */ -#define PNG_LIBPNG_VER 10643 /* 1.6.43 */ - -/* Library configuration: these options cannot be changed after - * the library has been built. - */ -#ifndef PNGLCONF_H -/* If pnglibconf.h is missing, you can - * copy scripts/pnglibconf.h.prebuilt to pnglibconf.h - */ -# include "pnglibconf.h" -#endif - -#ifndef PNG_VERSION_INFO_ONLY -/* Machine specific configuration. */ -# include "pngconf.h" -#endif - -/* - * Added at libpng-1.2.8 - * - * Ref MSDN: Private as priority over Special - * VS_FF_PRIVATEBUILD File *was not* built using standard release - * procedures. If this value is given, the StringFileInfo block must - * contain a PrivateBuild string. - * - * VS_FF_SPECIALBUILD File *was* built by the original company using - * standard release procedures but is a variation of the standard - * file of the same version number. If this value is given, the - * StringFileInfo block must contain a SpecialBuild string. - */ - -#ifdef PNG_USER_PRIVATEBUILD /* From pnglibconf.h */ -# define PNG_LIBPNG_BUILD_TYPE \ - (PNG_LIBPNG_BUILD_BASE_TYPE | PNG_LIBPNG_BUILD_PRIVATE) -#else -# ifdef PNG_LIBPNG_SPECIALBUILD -# define PNG_LIBPNG_BUILD_TYPE \ - (PNG_LIBPNG_BUILD_BASE_TYPE | PNG_LIBPNG_BUILD_SPECIAL) -# else -# define PNG_LIBPNG_BUILD_TYPE (PNG_LIBPNG_BUILD_BASE_TYPE) -# endif -#endif - -#ifndef PNG_VERSION_INFO_ONLY - -/* Inhibit C++ name-mangling for libpng functions but not for system calls. */ -#ifdef __cplusplus -extern "C" { -#endif /* __cplusplus */ - -/* Version information for C files, stored in png.c. This had better match - * the version above. - */ -#define png_libpng_ver png_get_header_ver(NULL) - -/* This file is arranged in several sections: - * - * 1. [omitted] - * 2. Any configuration options that can be specified by for the application - * code when it is built. (Build time configuration is in pnglibconf.h) - * 3. Type definitions (base types are defined in pngconf.h), structure - * definitions. - * 4. Exported library functions. - * 5. Simplified API. - * 6. Implementation options. - * - * The library source code has additional files (principally pngpriv.h) that - * allow configuration of the library. - */ - -/* Section 1: [omitted] */ - -/* Section 2: run time configuration - * See pnglibconf.h for build time configuration - * - * Run time configuration allows the application to choose between - * implementations of certain arithmetic APIs. The default is set - * at build time and recorded in pnglibconf.h, but it is safe to - * override these (and only these) settings. Note that this won't - * change what the library does, only application code, and the - * settings can (and probably should) be made on a per-file basis - * by setting the #defines before including png.h - * - * Use macros to read integers from PNG data or use the exported - * functions? - * PNG_USE_READ_MACROS: use the macros (see below) Note that - * the macros evaluate their argument multiple times. - * PNG_NO_USE_READ_MACROS: call the relevant library function. - * - * Use the alternative algorithm for compositing alpha samples that - * does not use division? - * PNG_READ_COMPOSITE_NODIV_SUPPORTED: use the 'no division' - * algorithm. - * PNG_NO_READ_COMPOSITE_NODIV: use the 'division' algorithm. - * - * How to handle benign errors if PNG_ALLOW_BENIGN_ERRORS is - * false? - * PNG_ALLOW_BENIGN_ERRORS: map calls to the benign error - * APIs to png_warning. - * Otherwise the calls are mapped to png_error. - */ - -/* Section 3: type definitions, including structures and compile time - * constants. - * See pngconf.h for base types that vary by machine/system - */ - -/* This triggers a compiler error in png.c, if png.c and png.h - * do not agree upon the version number. - */ -typedef char* png_libpng_version_1_6_43; - -/* Basic control structions. Read libpng-manual.txt or libpng.3 for more info. - * - * png_struct is the cache of information used while reading or writing a single - * PNG file. One of these is always required, although the simplified API - * (below) hides the creation and destruction of it. - */ -typedef struct png_struct_def png_struct; -typedef const png_struct * png_const_structp; -typedef png_struct * png_structp; -typedef png_struct * * png_structpp; - -/* png_info contains information read from or to be written to a PNG file. One - * or more of these must exist while reading or creating a PNG file. The - * information is not used by libpng during read but is used to control what - * gets written when a PNG file is created. "png_get_" function calls read - * information during read and "png_set_" functions calls write information - * when creating a PNG. - * been moved into a separate header file that is not accessible to - * applications. Read libpng-manual.txt or libpng.3 for more info. - */ -typedef struct png_info_def png_info; -typedef png_info * png_infop; -typedef const png_info * png_const_infop; -typedef png_info * * png_infopp; - -/* Types with names ending 'p' are pointer types. The corresponding types with - * names ending 'rp' are identical pointer types except that the pointer is - * marked 'restrict', which means that it is the only pointer to the object - * passed to the function. Applications should not use the 'restrict' types; - * it is always valid to pass 'p' to a pointer with a function argument of the - * corresponding 'rp' type. Different compilers have different rules with - * regard to type matching in the presence of 'restrict'. For backward - * compatibility libpng callbacks never have 'restrict' in their parameters and, - * consequentially, writing portable application code is extremely difficult if - * an attempt is made to use 'restrict'. - */ -typedef png_struct * PNG_RESTRICT png_structrp; -typedef const png_struct * PNG_RESTRICT png_const_structrp; -typedef png_info * PNG_RESTRICT png_inforp; -typedef const png_info * PNG_RESTRICT png_const_inforp; - -/* Three color definitions. The order of the red, green, and blue, (and the - * exact size) is not important, although the size of the fields need to - * be png_byte or png_uint_16 (as defined below). - */ -typedef struct png_color_struct -{ - png_byte red; - png_byte green; - png_byte blue; -} png_color; -typedef png_color * png_colorp; -typedef const png_color * png_const_colorp; -typedef png_color * * png_colorpp; - -typedef struct png_color_16_struct -{ - png_byte index; /* used for palette files */ - png_uint_16 red; /* for use in red green blue files */ - png_uint_16 green; - png_uint_16 blue; - png_uint_16 gray; /* for use in grayscale files */ -} png_color_16; -typedef png_color_16 * png_color_16p; -typedef const png_color_16 * png_const_color_16p; -typedef png_color_16 * * png_color_16pp; - -typedef struct png_color_8_struct -{ - png_byte red; /* for use in red green blue files */ - png_byte green; - png_byte blue; - png_byte gray; /* for use in grayscale files */ - png_byte alpha; /* for alpha channel files */ -} png_color_8; -typedef png_color_8 * png_color_8p; -typedef const png_color_8 * png_const_color_8p; -typedef png_color_8 * * png_color_8pp; - -/* - * The following two structures are used for the in-core representation - * of sPLT chunks. - */ -typedef struct png_sPLT_entry_struct -{ - png_uint_16 red; - png_uint_16 green; - png_uint_16 blue; - png_uint_16 alpha; - png_uint_16 frequency; -} png_sPLT_entry; -typedef png_sPLT_entry * png_sPLT_entryp; -typedef const png_sPLT_entry * png_const_sPLT_entryp; -typedef png_sPLT_entry * * png_sPLT_entrypp; - -/* When the depth of the sPLT palette is 8 bits, the color and alpha samples - * occupy the LSB of their respective members, and the MSB of each member - * is zero-filled. The frequency member always occupies the full 16 bits. - */ - -typedef struct png_sPLT_struct -{ - png_charp name; /* palette name */ - png_byte depth; /* depth of palette samples */ - png_sPLT_entryp entries; /* palette entries */ - png_int_32 nentries; /* number of palette entries */ -} png_sPLT_t; -typedef png_sPLT_t * png_sPLT_tp; -typedef const png_sPLT_t * png_const_sPLT_tp; -typedef png_sPLT_t * * png_sPLT_tpp; - -#ifdef PNG_TEXT_SUPPORTED -/* png_text holds the contents of a text/ztxt/itxt chunk in a PNG file, - * and whether that contents is compressed or not. The "key" field - * points to a regular zero-terminated C string. The "text" fields can be a - * regular C string, an empty string, or a NULL pointer. - * However, the structure returned by png_get_text() will always contain - * the "text" field as a regular zero-terminated C string (possibly - * empty), never a NULL pointer, so it can be safely used in printf() and - * other string-handling functions. Note that the "itxt_length", "lang", and - * "lang_key" members of the structure only exist when the library is built - * with iTXt chunk support. Prior to libpng-1.4.0 the library was built by - * default without iTXt support. Also note that when iTXt *is* supported, - * the "lang" and "lang_key" fields contain NULL pointers when the - * "compression" field contains * PNG_TEXT_COMPRESSION_NONE or - * PNG_TEXT_COMPRESSION_zTXt. Note that the "compression value" is not the - * same as what appears in the PNG tEXt/zTXt/iTXt chunk's "compression flag" - * which is always 0 or 1, or its "compression method" which is always 0. - */ -typedef struct png_text_struct -{ - int compression; /* compression value: - -1: tEXt, none - 0: zTXt, deflate - 1: iTXt, none - 2: iTXt, deflate */ - png_charp key; /* keyword, 1-79 character description of "text" */ - png_charp text; /* comment, may be an empty string (ie "") - or a NULL pointer */ - size_t text_length; /* length of the text string */ - size_t itxt_length; /* length of the itxt string */ - png_charp lang; /* language code, 0-79 characters - or a NULL pointer */ - png_charp lang_key; /* keyword translated UTF-8 string, 0 or more - chars or a NULL pointer */ -} png_text; -typedef png_text * png_textp; -typedef const png_text * png_const_textp; -typedef png_text * * png_textpp; -#endif - -/* Supported compression types for text in PNG files (tEXt, and zTXt). - * The values of the PNG_TEXT_COMPRESSION_ defines should NOT be changed. */ -#define PNG_TEXT_COMPRESSION_NONE_WR -3 -#define PNG_TEXT_COMPRESSION_zTXt_WR -2 -#define PNG_TEXT_COMPRESSION_NONE -1 -#define PNG_TEXT_COMPRESSION_zTXt 0 -#define PNG_ITXT_COMPRESSION_NONE 1 -#define PNG_ITXT_COMPRESSION_zTXt 2 -#define PNG_TEXT_COMPRESSION_LAST 3 /* Not a valid value */ - -/* png_time is a way to hold the time in an machine independent way. - * Two conversions are provided, both from time_t and struct tm. There - * is no portable way to convert to either of these structures, as far - * as I know. If you know of a portable way, send it to me. As a side - * note - PNG has always been Year 2000 compliant! - */ -typedef struct png_time_struct -{ - png_uint_16 year; /* full year, as in, 1995 */ - png_byte month; /* month of year, 1 - 12 */ - png_byte day; /* day of month, 1 - 31 */ - png_byte hour; /* hour of day, 0 - 23 */ - png_byte minute; /* minute of hour, 0 - 59 */ - png_byte second; /* second of minute, 0 - 60 (for leap seconds) */ -} png_time; -typedef png_time * png_timep; -typedef const png_time * png_const_timep; -typedef png_time * * png_timepp; - -#if defined(PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED) ||\ - defined(PNG_USER_CHUNKS_SUPPORTED) -/* png_unknown_chunk is a structure to hold queued chunks for which there is - * no specific support. The idea is that we can use this to queue - * up private chunks for output even though the library doesn't actually - * know about their semantics. - * - * The data in the structure is set by libpng on read and used on write. - */ -typedef struct png_unknown_chunk_t -{ - png_byte name[5]; /* Textual chunk name with '\0' terminator */ - png_byte *data; /* Data, should not be modified on read! */ - size_t size; - - /* On write 'location' must be set using the flag values listed below. - * Notice that on read it is set by libpng however the values stored have - * more bits set than are listed below. Always treat the value as a - * bitmask. On write set only one bit - setting multiple bits may cause the - * chunk to be written in multiple places. - */ - png_byte location; /* mode of operation at read time */ -} -png_unknown_chunk; - -typedef png_unknown_chunk * png_unknown_chunkp; -typedef const png_unknown_chunk * png_const_unknown_chunkp; -typedef png_unknown_chunk * * png_unknown_chunkpp; -#endif - -/* Flag values for the unknown chunk location byte. */ -#define PNG_HAVE_IHDR 0x01 -#define PNG_HAVE_PLTE 0x02 -#define PNG_AFTER_IDAT 0x08 - -/* Maximum positive integer used in PNG is (2^31)-1 */ -#define PNG_UINT_31_MAX ((png_uint_32)0x7fffffffL) -#define PNG_UINT_32_MAX ((png_uint_32)(-1)) -#define PNG_SIZE_MAX ((size_t)(-1)) - -/* These are constants for fixed point values encoded in the - * PNG specification manner (x100000) - */ -#define PNG_FP_1 100000 -#define PNG_FP_HALF 50000 -#define PNG_FP_MAX ((png_fixed_point)0x7fffffffL) -#define PNG_FP_MIN (-PNG_FP_MAX) - -/* These describe the color_type field in png_info. */ -/* color type masks */ -#define PNG_COLOR_MASK_PALETTE 1 -#define PNG_COLOR_MASK_COLOR 2 -#define PNG_COLOR_MASK_ALPHA 4 - -/* color types. Note that not all combinations are legal */ -#define PNG_COLOR_TYPE_GRAY 0 -#define PNG_COLOR_TYPE_PALETTE (PNG_COLOR_MASK_COLOR | PNG_COLOR_MASK_PALETTE) -#define PNG_COLOR_TYPE_RGB (PNG_COLOR_MASK_COLOR) -#define PNG_COLOR_TYPE_RGB_ALPHA (PNG_COLOR_MASK_COLOR | PNG_COLOR_MASK_ALPHA) -#define PNG_COLOR_TYPE_GRAY_ALPHA (PNG_COLOR_MASK_ALPHA) -/* aliases */ -#define PNG_COLOR_TYPE_RGBA PNG_COLOR_TYPE_RGB_ALPHA -#define PNG_COLOR_TYPE_GA PNG_COLOR_TYPE_GRAY_ALPHA - -/* This is for compression type. PNG 1.0-1.2 only define the single type. */ -#define PNG_COMPRESSION_TYPE_BASE 0 /* Deflate method 8, 32K window */ -#define PNG_COMPRESSION_TYPE_DEFAULT PNG_COMPRESSION_TYPE_BASE - -/* This is for filter type. PNG 1.0-1.2 only define the single type. */ -#define PNG_FILTER_TYPE_BASE 0 /* Single row per-byte filtering */ -#define PNG_INTRAPIXEL_DIFFERENCING 64 /* Used only in MNG datastreams */ -#define PNG_FILTER_TYPE_DEFAULT PNG_FILTER_TYPE_BASE - -/* These are for the interlacing type. These values should NOT be changed. */ -#define PNG_INTERLACE_NONE 0 /* Non-interlaced image */ -#define PNG_INTERLACE_ADAM7 1 /* Adam7 interlacing */ -#define PNG_INTERLACE_LAST 2 /* Not a valid value */ - -/* These are for the oFFs chunk. These values should NOT be changed. */ -#define PNG_OFFSET_PIXEL 0 /* Offset in pixels */ -#define PNG_OFFSET_MICROMETER 1 /* Offset in micrometers (1/10^6 meter) */ -#define PNG_OFFSET_LAST 2 /* Not a valid value */ - -/* These are for the pCAL chunk. These values should NOT be changed. */ -#define PNG_EQUATION_LINEAR 0 /* Linear transformation */ -#define PNG_EQUATION_BASE_E 1 /* Exponential base e transform */ -#define PNG_EQUATION_ARBITRARY 2 /* Arbitrary base exponential transform */ -#define PNG_EQUATION_HYPERBOLIC 3 /* Hyperbolic sine transformation */ -#define PNG_EQUATION_LAST 4 /* Not a valid value */ - -/* These are for the sCAL chunk. These values should NOT be changed. */ -#define PNG_SCALE_UNKNOWN 0 /* unknown unit (image scale) */ -#define PNG_SCALE_METER 1 /* meters per pixel */ -#define PNG_SCALE_RADIAN 2 /* radians per pixel */ -#define PNG_SCALE_LAST 3 /* Not a valid value */ - -/* These are for the pHYs chunk. These values should NOT be changed. */ -#define PNG_RESOLUTION_UNKNOWN 0 /* pixels/unknown unit (aspect ratio) */ -#define PNG_RESOLUTION_METER 1 /* pixels/meter */ -#define PNG_RESOLUTION_LAST 2 /* Not a valid value */ - -/* These are for the sRGB chunk. These values should NOT be changed. */ -#define PNG_sRGB_INTENT_PERCEPTUAL 0 -#define PNG_sRGB_INTENT_RELATIVE 1 -#define PNG_sRGB_INTENT_SATURATION 2 -#define PNG_sRGB_INTENT_ABSOLUTE 3 -#define PNG_sRGB_INTENT_LAST 4 /* Not a valid value */ - -/* This is for text chunks */ -#define PNG_KEYWORD_MAX_LENGTH 79 - -/* Maximum number of entries in PLTE/sPLT/tRNS arrays */ -#define PNG_MAX_PALETTE_LENGTH 256 - -/* These determine if an ancillary chunk's data has been successfully read - * from the PNG header, or if the application has filled in the corresponding - * data in the info_struct to be written into the output file. The values - * of the PNG_INFO_ defines should NOT be changed. - */ -#define PNG_INFO_gAMA 0x0001U -#define PNG_INFO_sBIT 0x0002U -#define PNG_INFO_cHRM 0x0004U -#define PNG_INFO_PLTE 0x0008U -#define PNG_INFO_tRNS 0x0010U -#define PNG_INFO_bKGD 0x0020U -#define PNG_INFO_hIST 0x0040U -#define PNG_INFO_pHYs 0x0080U -#define PNG_INFO_oFFs 0x0100U -#define PNG_INFO_tIME 0x0200U -#define PNG_INFO_pCAL 0x0400U -#define PNG_INFO_sRGB 0x0800U /* GR-P, 0.96a */ -#define PNG_INFO_iCCP 0x1000U /* ESR, 1.0.6 */ -#define PNG_INFO_sPLT 0x2000U /* ESR, 1.0.6 */ -#define PNG_INFO_sCAL 0x4000U /* ESR, 1.0.6 */ -#define PNG_INFO_IDAT 0x8000U /* ESR, 1.0.6 */ -#define PNG_INFO_eXIf 0x10000U /* GR-P, 1.6.31 */ - -/* This is used for the transformation routines, as some of them - * change these values for the row. It also should enable using - * the routines for other purposes. - */ -typedef struct png_row_info_struct -{ - png_uint_32 width; /* width of row */ - size_t rowbytes; /* number of bytes in row */ - png_byte color_type; /* color type of row */ - png_byte bit_depth; /* bit depth of row */ - png_byte channels; /* number of channels (1, 2, 3, or 4) */ - png_byte pixel_depth; /* bits per pixel (depth * channels) */ -} png_row_info; - -typedef png_row_info * png_row_infop; -typedef png_row_info * * png_row_infopp; - -/* These are the function types for the I/O functions and for the functions - * that allow the user to override the default I/O functions with his or her - * own. The png_error_ptr type should match that of user-supplied warning - * and error functions, while the png_rw_ptr type should match that of the - * user read/write data functions. Note that the 'write' function must not - * modify the buffer it is passed. The 'read' function, on the other hand, is - * expected to return the read data in the buffer. - */ -typedef PNG_CALLBACK(void, *png_error_ptr, (png_structp, png_const_charp)); -typedef PNG_CALLBACK(void, *png_rw_ptr, (png_structp, png_bytep, size_t)); -typedef PNG_CALLBACK(void, *png_flush_ptr, (png_structp)); -typedef PNG_CALLBACK(void, *png_read_status_ptr, (png_structp, png_uint_32, - int)); -typedef PNG_CALLBACK(void, *png_write_status_ptr, (png_structp, png_uint_32, - int)); - -#ifdef PNG_PROGRESSIVE_READ_SUPPORTED -typedef PNG_CALLBACK(void, *png_progressive_info_ptr, (png_structp, png_infop)); -typedef PNG_CALLBACK(void, *png_progressive_end_ptr, (png_structp, png_infop)); - -/* The following callback receives png_uint_32 row_number, int pass for the - * png_bytep data of the row. When transforming an interlaced image the - * row number is the row number within the sub-image of the interlace pass, so - * the value will increase to the height of the sub-image (not the full image) - * then reset to 0 for the next pass. - * - * Use PNG_ROW_FROM_PASS_ROW(row, pass) and PNG_COL_FROM_PASS_COL(col, pass) to - * find the output pixel (x,y) given an interlaced sub-image pixel - * (row,col,pass). (See below for these macros.) - */ -typedef PNG_CALLBACK(void, *png_progressive_row_ptr, (png_structp, png_bytep, - png_uint_32, int)); -#endif - -#if defined(PNG_READ_USER_TRANSFORM_SUPPORTED) || \ - defined(PNG_WRITE_USER_TRANSFORM_SUPPORTED) -typedef PNG_CALLBACK(void, *png_user_transform_ptr, (png_structp, png_row_infop, - png_bytep)); -#endif - -#ifdef PNG_USER_CHUNKS_SUPPORTED -typedef PNG_CALLBACK(int, *png_user_chunk_ptr, (png_structp, - png_unknown_chunkp)); -#endif -#ifdef PNG_UNKNOWN_CHUNKS_SUPPORTED -/* not used anywhere */ -/* typedef PNG_CALLBACK(void, *png_unknown_chunk_ptr, (png_structp)); */ -#endif - -#ifdef PNG_SETJMP_SUPPORTED -/* This must match the function definition in , and the application - * must include this before png.h to obtain the definition of jmp_buf. The - * function is required to be PNG_NORETURN, but this is not checked. If the - * function does return the application will crash via an abort() or similar - * system level call. - * - * If you get a warning here while building the library you may need to make - * changes to ensure that pnglibconf.h records the calling convention used by - * your compiler. This may be very difficult - try using a different compiler - * to build the library! - */ -PNG_FUNCTION(void, (PNGCAPI *png_longjmp_ptr), PNGARG((jmp_buf, int)), typedef); -#endif - -/* Transform masks for the high-level interface */ -#define PNG_TRANSFORM_IDENTITY 0x0000 /* read and write */ -#define PNG_TRANSFORM_STRIP_16 0x0001 /* read only */ -#define PNG_TRANSFORM_STRIP_ALPHA 0x0002 /* read only */ -#define PNG_TRANSFORM_PACKING 0x0004 /* read and write */ -#define PNG_TRANSFORM_PACKSWAP 0x0008 /* read and write */ -#define PNG_TRANSFORM_EXPAND 0x0010 /* read only */ -#define PNG_TRANSFORM_INVERT_MONO 0x0020 /* read and write */ -#define PNG_TRANSFORM_SHIFT 0x0040 /* read and write */ -#define PNG_TRANSFORM_BGR 0x0080 /* read and write */ -#define PNG_TRANSFORM_SWAP_ALPHA 0x0100 /* read and write */ -#define PNG_TRANSFORM_SWAP_ENDIAN 0x0200 /* read and write */ -#define PNG_TRANSFORM_INVERT_ALPHA 0x0400 /* read and write */ -#define PNG_TRANSFORM_STRIP_FILLER 0x0800 /* write only */ -/* Added to libpng-1.2.34 */ -#define PNG_TRANSFORM_STRIP_FILLER_BEFORE PNG_TRANSFORM_STRIP_FILLER -#define PNG_TRANSFORM_STRIP_FILLER_AFTER 0x1000 /* write only */ -/* Added to libpng-1.4.0 */ -#define PNG_TRANSFORM_GRAY_TO_RGB 0x2000 /* read only */ -/* Added to libpng-1.5.4 */ -#define PNG_TRANSFORM_EXPAND_16 0x4000 /* read only */ -#if ~0U > 0xffffU /* or else this might break on a 16-bit machine */ -#define PNG_TRANSFORM_SCALE_16 0x8000 /* read only */ -#endif - -/* Flags for MNG supported features */ -#define PNG_FLAG_MNG_EMPTY_PLTE 0x01 -#define PNG_FLAG_MNG_FILTER_64 0x04 -#define PNG_ALL_MNG_FEATURES 0x05 - -/* NOTE: prior to 1.5 these functions had no 'API' style declaration, - * this allowed the zlib default functions to be used on Windows - * platforms. In 1.5 the zlib default malloc (which just calls malloc and - * ignores the first argument) should be completely compatible with the - * following. - */ -typedef PNG_CALLBACK(png_voidp, *png_malloc_ptr, (png_structp, - png_alloc_size_t)); -typedef PNG_CALLBACK(void, *png_free_ptr, (png_structp, png_voidp)); - -/* Section 4: exported functions - * Here are the function definitions most commonly used. This is not - * the place to find out how to use libpng. See libpng-manual.txt for the - * full explanation, see example.c for the summary. This just provides - * a simple one line description of the use of each function. - * - * The PNG_EXPORT() and PNG_EXPORTA() macros used below are defined in - * pngconf.h and in the *.dfn files in the scripts directory. - * - * PNG_EXPORT(ordinal, type, name, (args)); - * - * ordinal: ordinal that is used while building - * *.def files. The ordinal value is only - * relevant when preprocessing png.h with - * the *.dfn files for building symbol table - * entries, and are removed by pngconf.h. - * type: return type of the function - * name: function name - * args: function arguments, with types - * - * When we wish to append attributes to a function prototype we use - * the PNG_EXPORTA() macro instead. - * - * PNG_EXPORTA(ordinal, type, name, (args), attributes); - * - * ordinal, type, name, and args: same as in PNG_EXPORT(). - * attributes: function attributes - */ - -/* Returns the version number of the library */ -PNG_EXPORT(1, png_uint_32, png_access_version_number, (void)); - -/* Tell lib we have already handled the first magic bytes. - * Handling more than 8 bytes from the beginning of the file is an error. - */ -PNG_EXPORT(2, void, png_set_sig_bytes, (png_structrp png_ptr, int num_bytes)); - -/* Check sig[start] through sig[start + num_to_check - 1] to see if it's a - * PNG file. Returns zero if the supplied bytes match the 8-byte PNG - * signature, and non-zero otherwise. Having num_to_check == 0 or - * start > 7 will always fail (i.e. return non-zero). - */ -PNG_EXPORT(3, int, png_sig_cmp, (png_const_bytep sig, size_t start, - size_t num_to_check)); - -/* Simple signature checking function. This is the same as calling - * png_check_sig(sig, n) := (png_sig_cmp(sig, 0, n) == 0). - */ -#define png_check_sig(sig, n) (png_sig_cmp((sig), 0, (n)) == 0) /* DEPRECATED */ - -/* Allocate and initialize png_ptr struct for reading, and any other memory. */ -PNG_EXPORTA(4, png_structp, png_create_read_struct, - (png_const_charp user_png_ver, png_voidp error_ptr, - png_error_ptr error_fn, png_error_ptr warn_fn), - PNG_ALLOCATED); - -/* Allocate and initialize png_ptr struct for writing, and any other memory */ -PNG_EXPORTA(5, png_structp, png_create_write_struct, - (png_const_charp user_png_ver, png_voidp error_ptr, png_error_ptr error_fn, - png_error_ptr warn_fn), - PNG_ALLOCATED); - -PNG_EXPORT(6, size_t, png_get_compression_buffer_size, - (png_const_structrp png_ptr)); - -PNG_EXPORT(7, void, png_set_compression_buffer_size, (png_structrp png_ptr, - size_t size)); - -/* Moved from pngconf.h in 1.4.0 and modified to ensure setjmp/longjmp - * match up. - */ -#ifdef PNG_SETJMP_SUPPORTED -/* This function returns the jmp_buf built in to *png_ptr. It must be - * supplied with an appropriate 'longjmp' function to use on that jmp_buf - * unless the default error function is overridden in which case NULL is - * acceptable. The size of the jmp_buf is checked against the actual size - * allocated by the library - the call will return NULL on a mismatch - * indicating an ABI mismatch. - */ -PNG_EXPORT(8, jmp_buf*, png_set_longjmp_fn, (png_structrp png_ptr, - png_longjmp_ptr longjmp_fn, size_t jmp_buf_size)); -# define png_jmpbuf(png_ptr) \ - (*png_set_longjmp_fn((png_ptr), longjmp, (sizeof (jmp_buf)))) -#else -# define png_jmpbuf(png_ptr) \ - (LIBPNG_WAS_COMPILED_WITH__PNG_NO_SETJMP) -#endif -/* This function should be used by libpng applications in place of - * longjmp(png_ptr->jmpbuf, val). If longjmp_fn() has been set, it - * will use it; otherwise it will call PNG_ABORT(). This function was - * added in libpng-1.5.0. - */ -PNG_EXPORTA(9, void, png_longjmp, (png_const_structrp png_ptr, int val), - PNG_NORETURN); - -#ifdef PNG_READ_SUPPORTED -/* Reset the compression stream */ -PNG_EXPORTA(10, int, png_reset_zstream, (png_structrp png_ptr), PNG_DEPRECATED); -#endif - -/* New functions added in libpng-1.0.2 (not enabled by default until 1.2.0) */ -#ifdef PNG_USER_MEM_SUPPORTED -PNG_EXPORTA(11, png_structp, png_create_read_struct_2, - (png_const_charp user_png_ver, png_voidp error_ptr, png_error_ptr error_fn, - png_error_ptr warn_fn, - png_voidp mem_ptr, png_malloc_ptr malloc_fn, png_free_ptr free_fn), - PNG_ALLOCATED); -PNG_EXPORTA(12, png_structp, png_create_write_struct_2, - (png_const_charp user_png_ver, png_voidp error_ptr, png_error_ptr error_fn, - png_error_ptr warn_fn, - png_voidp mem_ptr, png_malloc_ptr malloc_fn, png_free_ptr free_fn), - PNG_ALLOCATED); -#endif - -/* Write the PNG file signature. */ -PNG_EXPORT(13, void, png_write_sig, (png_structrp png_ptr)); - -/* Write a PNG chunk - size, type, (optional) data, CRC. */ -PNG_EXPORT(14, void, png_write_chunk, (png_structrp png_ptr, png_const_bytep - chunk_name, png_const_bytep data, size_t length)); - -/* Write the start of a PNG chunk - length and chunk name. */ -PNG_EXPORT(15, void, png_write_chunk_start, (png_structrp png_ptr, - png_const_bytep chunk_name, png_uint_32 length)); - -/* Write the data of a PNG chunk started with png_write_chunk_start(). */ -PNG_EXPORT(16, void, png_write_chunk_data, (png_structrp png_ptr, - png_const_bytep data, size_t length)); - -/* Finish a chunk started with png_write_chunk_start() (includes CRC). */ -PNG_EXPORT(17, void, png_write_chunk_end, (png_structrp png_ptr)); - -/* Allocate and initialize the info structure */ -PNG_EXPORTA(18, png_infop, png_create_info_struct, (png_const_structrp png_ptr), - PNG_ALLOCATED); - -/* DEPRECATED: this function allowed init structures to be created using the - * default allocation method (typically malloc). Use is deprecated in 1.6.0 and - * the API will be removed in the future. - */ -PNG_EXPORTA(19, void, png_info_init_3, (png_infopp info_ptr, - size_t png_info_struct_size), PNG_DEPRECATED); - -/* Writes all the PNG information before the image. */ -PNG_EXPORT(20, void, png_write_info_before_PLTE, - (png_structrp png_ptr, png_const_inforp info_ptr)); -PNG_EXPORT(21, void, png_write_info, - (png_structrp png_ptr, png_const_inforp info_ptr)); - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -/* Read the information before the actual image data. */ -PNG_EXPORT(22, void, png_read_info, - (png_structrp png_ptr, png_inforp info_ptr)); -#endif - -#ifdef PNG_TIME_RFC1123_SUPPORTED - /* Convert to a US string format: there is no localization support in this - * routine. The original implementation used a 29 character buffer in - * png_struct, this will be removed in future versions. - */ -#if PNG_LIBPNG_VER < 10700 -/* To do: remove this from libpng17 (and from libpng17/png.c and pngstruct.h) */ -PNG_EXPORTA(23, png_const_charp, png_convert_to_rfc1123, (png_structrp png_ptr, - png_const_timep ptime),PNG_DEPRECATED); -#endif -PNG_EXPORT(241, int, png_convert_to_rfc1123_buffer, (char out[29], - png_const_timep ptime)); -#endif - -#ifdef PNG_CONVERT_tIME_SUPPORTED -/* Convert from a struct tm to png_time */ -PNG_EXPORT(24, void, png_convert_from_struct_tm, (png_timep ptime, - const struct tm * ttime)); - -/* Convert from time_t to png_time. Uses gmtime() */ -PNG_EXPORT(25, void, png_convert_from_time_t, (png_timep ptime, time_t ttime)); -#endif /* CONVERT_tIME */ - -#ifdef PNG_READ_EXPAND_SUPPORTED -/* Expand data to 24-bit RGB, or 8-bit grayscale, with alpha if available. */ -PNG_EXPORT(26, void, png_set_expand, (png_structrp png_ptr)); -PNG_EXPORT(27, void, png_set_expand_gray_1_2_4_to_8, (png_structrp png_ptr)); -PNG_EXPORT(28, void, png_set_palette_to_rgb, (png_structrp png_ptr)); -PNG_EXPORT(29, void, png_set_tRNS_to_alpha, (png_structrp png_ptr)); -#endif - -#ifdef PNG_READ_EXPAND_16_SUPPORTED -/* Expand to 16-bit channels, forces conversion of palette to RGB and expansion - * of a tRNS chunk if present. - */ -PNG_EXPORT(221, void, png_set_expand_16, (png_structrp png_ptr)); -#endif - -#if defined(PNG_READ_BGR_SUPPORTED) || defined(PNG_WRITE_BGR_SUPPORTED) -/* Use blue, green, red order for pixels. */ -PNG_EXPORT(30, void, png_set_bgr, (png_structrp png_ptr)); -#endif - -#ifdef PNG_READ_GRAY_TO_RGB_SUPPORTED -/* Expand the grayscale to 24-bit RGB if necessary. */ -PNG_EXPORT(31, void, png_set_gray_to_rgb, (png_structrp png_ptr)); -#endif - -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED -/* Reduce RGB to grayscale. */ -#define PNG_ERROR_ACTION_NONE 1 -#define PNG_ERROR_ACTION_WARN 2 -#define PNG_ERROR_ACTION_ERROR 3 -#define PNG_RGB_TO_GRAY_DEFAULT (-1)/*for red/green coefficients*/ - -PNG_FP_EXPORT(32, void, png_set_rgb_to_gray, (png_structrp png_ptr, - int error_action, double red, double green)) -PNG_FIXED_EXPORT(33, void, png_set_rgb_to_gray_fixed, (png_structrp png_ptr, - int error_action, png_fixed_point red, png_fixed_point green)) - -PNG_EXPORT(34, png_byte, png_get_rgb_to_gray_status, (png_const_structrp - png_ptr)); -#endif - -#ifdef PNG_BUILD_GRAYSCALE_PALETTE_SUPPORTED -PNG_EXPORT(35, void, png_build_grayscale_palette, (int bit_depth, - png_colorp palette)); -#endif - -#ifdef PNG_READ_ALPHA_MODE_SUPPORTED -/* How the alpha channel is interpreted - this affects how the color channels - * of a PNG file are returned to the calling application when an alpha channel, - * or a tRNS chunk in a palette file, is present. - * - * This has no effect on the way pixels are written into a PNG output - * datastream. The color samples in a PNG datastream are never premultiplied - * with the alpha samples. - * - * The default is to return data according to the PNG specification: the alpha - * channel is a linear measure of the contribution of the pixel to the - * corresponding composited pixel, and the color channels are unassociated - * (not premultiplied). The gamma encoded color channels must be scaled - * according to the contribution and to do this it is necessary to undo - * the encoding, scale the color values, perform the composition and re-encode - * the values. This is the 'PNG' mode. - * - * The alternative is to 'associate' the alpha with the color information by - * storing color channel values that have been scaled by the alpha. - * image. These are the 'STANDARD', 'ASSOCIATED' or 'PREMULTIPLIED' modes - * (the latter being the two common names for associated alpha color channels). - * - * For the 'OPTIMIZED' mode, a pixel is treated as opaque only if the alpha - * value is equal to the maximum value. - * - * The final choice is to gamma encode the alpha channel as well. This is - * broken because, in practice, no implementation that uses this choice - * correctly undoes the encoding before handling alpha composition. Use this - * choice only if other serious errors in the software or hardware you use - * mandate it; the typical serious error is for dark halos to appear around - * opaque areas of the composited PNG image because of arithmetic overflow. - * - * The API function png_set_alpha_mode specifies which of these choices to use - * with an enumerated 'mode' value and the gamma of the required output: - */ -#define PNG_ALPHA_PNG 0 /* according to the PNG standard */ -#define PNG_ALPHA_STANDARD 1 /* according to Porter/Duff */ -#define PNG_ALPHA_ASSOCIATED 1 /* as above; this is the normal practice */ -#define PNG_ALPHA_PREMULTIPLIED 1 /* as above */ -#define PNG_ALPHA_OPTIMIZED 2 /* 'PNG' for opaque pixels, else 'STANDARD' */ -#define PNG_ALPHA_BROKEN 3 /* the alpha channel is gamma encoded */ - -PNG_FP_EXPORT(227, void, png_set_alpha_mode, (png_structrp png_ptr, int mode, - double output_gamma)) -PNG_FIXED_EXPORT(228, void, png_set_alpha_mode_fixed, (png_structrp png_ptr, - int mode, png_fixed_point output_gamma)) -#endif - -#if defined(PNG_GAMMA_SUPPORTED) || defined(PNG_READ_ALPHA_MODE_SUPPORTED) -/* The output_gamma value is a screen gamma in libpng terminology: it expresses - * how to decode the output values, not how they are encoded. - */ -#define PNG_DEFAULT_sRGB -1 /* sRGB gamma and color space */ -#define PNG_GAMMA_MAC_18 -2 /* Old Mac '1.8' gamma and color space */ -#define PNG_GAMMA_sRGB 220000 /* Television standards--matches sRGB gamma */ -#define PNG_GAMMA_LINEAR PNG_FP_1 /* Linear */ -#endif - -/* The following are examples of calls to png_set_alpha_mode to achieve the - * required overall gamma correction and, where necessary, alpha - * premultiplication. - * - * png_set_alpha_mode(pp, PNG_ALPHA_PNG, PNG_DEFAULT_sRGB); - * This is the default libpng handling of the alpha channel - it is not - * pre-multiplied into the color components. In addition the call states - * that the output is for a sRGB system and causes all PNG files without gAMA - * chunks to be assumed to be encoded using sRGB. - * - * png_set_alpha_mode(pp, PNG_ALPHA_PNG, PNG_GAMMA_MAC); - * In this case the output is assumed to be something like an sRGB conformant - * display preceded by a power-law lookup table of power 1.45. This is how - * early Mac systems behaved. - * - * png_set_alpha_mode(pp, PNG_ALPHA_STANDARD, PNG_GAMMA_LINEAR); - * This is the classic Jim Blinn approach and will work in academic - * environments where everything is done by the book. It has the shortcoming - * of assuming that input PNG data with no gamma information is linear - this - * is unlikely to be correct unless the PNG files where generated locally. - * Most of the time the output precision will be so low as to show - * significant banding in dark areas of the image. - * - * png_set_expand_16(pp); - * png_set_alpha_mode(pp, PNG_ALPHA_STANDARD, PNG_DEFAULT_sRGB); - * This is a somewhat more realistic Jim Blinn inspired approach. PNG files - * are assumed to have the sRGB encoding if not marked with a gamma value and - * the output is always 16 bits per component. This permits accurate scaling - * and processing of the data. If you know that your input PNG files were - * generated locally you might need to replace PNG_DEFAULT_sRGB with the - * correct value for your system. - * - * png_set_alpha_mode(pp, PNG_ALPHA_OPTIMIZED, PNG_DEFAULT_sRGB); - * If you just need to composite the PNG image onto an existing background - * and if you control the code that does this you can use the optimization - * setting. In this case you just copy completely opaque pixels to the - * output. For pixels that are not completely transparent (you just skip - * those) you do the composition math using png_composite or png_composite_16 - * below then encode the resultant 8-bit or 16-bit values to match the output - * encoding. - * - * Other cases - * If neither the PNG nor the standard linear encoding work for you because - * of the software or hardware you use then you have a big problem. The PNG - * case will probably result in halos around the image. The linear encoding - * will probably result in a washed out, too bright, image (it's actually too - * contrasty.) Try the ALPHA_OPTIMIZED mode above - this will probably - * substantially reduce the halos. Alternatively try: - * - * png_set_alpha_mode(pp, PNG_ALPHA_BROKEN, PNG_DEFAULT_sRGB); - * This option will also reduce the halos, but there will be slight dark - * halos round the opaque parts of the image where the background is light. - * In the OPTIMIZED mode the halos will be light halos where the background - * is dark. Take your pick - the halos are unavoidable unless you can get - * your hardware/software fixed! (The OPTIMIZED approach is slightly - * faster.) - * - * When the default gamma of PNG files doesn't match the output gamma. - * If you have PNG files with no gamma information png_set_alpha_mode allows - * you to provide a default gamma, but it also sets the output gamma to the - * matching value. If you know your PNG files have a gamma that doesn't - * match the output you can take advantage of the fact that - * png_set_alpha_mode always sets the output gamma but only sets the PNG - * default if it is not already set: - * - * png_set_alpha_mode(pp, PNG_ALPHA_PNG, PNG_DEFAULT_sRGB); - * png_set_alpha_mode(pp, PNG_ALPHA_PNG, PNG_GAMMA_MAC); - * The first call sets both the default and the output gamma values, the - * second call overrides the output gamma without changing the default. This - * is easier than achieving the same effect with png_set_gamma. You must use - * PNG_ALPHA_PNG for the first call - internal checking in png_set_alpha will - * fire if more than one call to png_set_alpha_mode and png_set_background is - * made in the same read operation, however multiple calls with PNG_ALPHA_PNG - * are ignored. - */ - -#ifdef PNG_READ_STRIP_ALPHA_SUPPORTED -PNG_EXPORT(36, void, png_set_strip_alpha, (png_structrp png_ptr)); -#endif - -#if defined(PNG_READ_SWAP_ALPHA_SUPPORTED) || \ - defined(PNG_WRITE_SWAP_ALPHA_SUPPORTED) -PNG_EXPORT(37, void, png_set_swap_alpha, (png_structrp png_ptr)); -#endif - -#if defined(PNG_READ_INVERT_ALPHA_SUPPORTED) || \ - defined(PNG_WRITE_INVERT_ALPHA_SUPPORTED) -PNG_EXPORT(38, void, png_set_invert_alpha, (png_structrp png_ptr)); -#endif - -#if defined(PNG_READ_FILLER_SUPPORTED) || defined(PNG_WRITE_FILLER_SUPPORTED) -/* Add a filler byte to 8-bit or 16-bit Gray or 24-bit or 48-bit RGB images. */ -PNG_EXPORT(39, void, png_set_filler, (png_structrp png_ptr, png_uint_32 filler, - int flags)); -/* The values of the PNG_FILLER_ defines should NOT be changed */ -# define PNG_FILLER_BEFORE 0 -# define PNG_FILLER_AFTER 1 -/* Add an alpha byte to 8-bit or 16-bit Gray or 24-bit or 48-bit RGB images. */ -PNG_EXPORT(40, void, png_set_add_alpha, (png_structrp png_ptr, - png_uint_32 filler, int flags)); -#endif /* READ_FILLER || WRITE_FILLER */ - -#if defined(PNG_READ_SWAP_SUPPORTED) || defined(PNG_WRITE_SWAP_SUPPORTED) -/* Swap bytes in 16-bit depth files. */ -PNG_EXPORT(41, void, png_set_swap, (png_structrp png_ptr)); -#endif - -#if defined(PNG_READ_PACK_SUPPORTED) || defined(PNG_WRITE_PACK_SUPPORTED) -/* Use 1 byte per pixel in 1, 2, or 4-bit depth files. */ -PNG_EXPORT(42, void, png_set_packing, (png_structrp png_ptr)); -#endif - -#if defined(PNG_READ_PACKSWAP_SUPPORTED) || \ - defined(PNG_WRITE_PACKSWAP_SUPPORTED) -/* Swap packing order of pixels in bytes. */ -PNG_EXPORT(43, void, png_set_packswap, (png_structrp png_ptr)); -#endif - -#if defined(PNG_READ_SHIFT_SUPPORTED) || defined(PNG_WRITE_SHIFT_SUPPORTED) -/* Converts files to legal bit depths. */ -PNG_EXPORT(44, void, png_set_shift, (png_structrp png_ptr, png_const_color_8p - true_bits)); -#endif - -#if defined(PNG_READ_INTERLACING_SUPPORTED) || \ - defined(PNG_WRITE_INTERLACING_SUPPORTED) -/* Have the code handle the interlacing. Returns the number of passes. - * MUST be called before png_read_update_info or png_start_read_image, - * otherwise it will not have the desired effect. Note that it is still - * necessary to call png_read_row or png_read_rows png_get_image_height - * times for each pass. -*/ -PNG_EXPORT(45, int, png_set_interlace_handling, (png_structrp png_ptr)); -#endif - -#if defined(PNG_READ_INVERT_SUPPORTED) || defined(PNG_WRITE_INVERT_SUPPORTED) -/* Invert monochrome files */ -PNG_EXPORT(46, void, png_set_invert_mono, (png_structrp png_ptr)); -#endif - -#ifdef PNG_READ_BACKGROUND_SUPPORTED -/* Handle alpha and tRNS by replacing with a background color. Prior to - * libpng-1.5.4 this API must not be called before the PNG file header has been - * read. Doing so will result in unexpected behavior and possible warnings or - * errors if the PNG file contains a bKGD chunk. - */ -PNG_FP_EXPORT(47, void, png_set_background, (png_structrp png_ptr, - png_const_color_16p background_color, int background_gamma_code, - int need_expand, double background_gamma)) -PNG_FIXED_EXPORT(215, void, png_set_background_fixed, (png_structrp png_ptr, - png_const_color_16p background_color, int background_gamma_code, - int need_expand, png_fixed_point background_gamma)) -#endif -#ifdef PNG_READ_BACKGROUND_SUPPORTED -# define PNG_BACKGROUND_GAMMA_UNKNOWN 0 -# define PNG_BACKGROUND_GAMMA_SCREEN 1 -# define PNG_BACKGROUND_GAMMA_FILE 2 -# define PNG_BACKGROUND_GAMMA_UNIQUE 3 -#endif - -#ifdef PNG_READ_SCALE_16_TO_8_SUPPORTED -/* Scale a 16-bit depth file down to 8-bit, accurately. */ -PNG_EXPORT(229, void, png_set_scale_16, (png_structrp png_ptr)); -#endif - -#ifdef PNG_READ_STRIP_16_TO_8_SUPPORTED -#define PNG_READ_16_TO_8_SUPPORTED /* Name prior to 1.5.4 */ -/* Strip the second byte of information from a 16-bit depth file. */ -PNG_EXPORT(48, void, png_set_strip_16, (png_structrp png_ptr)); -#endif - -#ifdef PNG_READ_QUANTIZE_SUPPORTED -/* Turn on quantizing, and reduce the palette to the number of colors - * available. - */ -PNG_EXPORT(49, void, png_set_quantize, (png_structrp png_ptr, - png_colorp palette, int num_palette, int maximum_colors, - png_const_uint_16p histogram, int full_quantize)); -#endif - -#ifdef PNG_READ_GAMMA_SUPPORTED -/* The threshold on gamma processing is configurable but hard-wired into the - * library. The following is the floating point variant. - */ -#define PNG_GAMMA_THRESHOLD (PNG_GAMMA_THRESHOLD_FIXED*.00001) - -/* Handle gamma correction. Screen_gamma=(display_exponent). - * NOTE: this API simply sets the screen and file gamma values. It will - * therefore override the value for gamma in a PNG file if it is called after - * the file header has been read - use with care - call before reading the PNG - * file for best results! - * - * These routines accept the same gamma values as png_set_alpha_mode (described - * above). The PNG_GAMMA_ defines and PNG_DEFAULT_sRGB can be passed to either - * API (floating point or fixed.) Notice, however, that the 'file_gamma' value - * is the inverse of a 'screen gamma' value. - */ -PNG_FP_EXPORT(50, void, png_set_gamma, (png_structrp png_ptr, - double screen_gamma, double override_file_gamma)) -PNG_FIXED_EXPORT(208, void, png_set_gamma_fixed, (png_structrp png_ptr, - png_fixed_point screen_gamma, png_fixed_point override_file_gamma)) -#endif - -#ifdef PNG_WRITE_FLUSH_SUPPORTED -/* Set how many lines between output flushes - 0 for no flushing */ -PNG_EXPORT(51, void, png_set_flush, (png_structrp png_ptr, int nrows)); -/* Flush the current PNG output buffer */ -PNG_EXPORT(52, void, png_write_flush, (png_structrp png_ptr)); -#endif - -/* Optional update palette with requested transformations */ -PNG_EXPORT(53, void, png_start_read_image, (png_structrp png_ptr)); - -/* Optional call to update the users info structure */ -PNG_EXPORT(54, void, png_read_update_info, (png_structrp png_ptr, - png_inforp info_ptr)); - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -/* Read one or more rows of image data. */ -PNG_EXPORT(55, void, png_read_rows, (png_structrp png_ptr, png_bytepp row, - png_bytepp display_row, png_uint_32 num_rows)); -#endif - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -/* Read a row of data. */ -PNG_EXPORT(56, void, png_read_row, (png_structrp png_ptr, png_bytep row, - png_bytep display_row)); -#endif - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -/* Read the whole image into memory at once. */ -PNG_EXPORT(57, void, png_read_image, (png_structrp png_ptr, png_bytepp image)); -#endif - -/* Write a row of image data */ -PNG_EXPORT(58, void, png_write_row, (png_structrp png_ptr, - png_const_bytep row)); - -/* Write a few rows of image data: (*row) is not written; however, the type - * is declared as writeable to maintain compatibility with previous versions - * of libpng and to allow the 'display_row' array from read_rows to be passed - * unchanged to write_rows. - */ -PNG_EXPORT(59, void, png_write_rows, (png_structrp png_ptr, png_bytepp row, - png_uint_32 num_rows)); - -/* Write the image data */ -PNG_EXPORT(60, void, png_write_image, (png_structrp png_ptr, png_bytepp image)); - -/* Write the end of the PNG file. */ -PNG_EXPORT(61, void, png_write_end, (png_structrp png_ptr, - png_inforp info_ptr)); - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -/* Read the end of the PNG file. */ -PNG_EXPORT(62, void, png_read_end, (png_structrp png_ptr, png_inforp info_ptr)); -#endif - -/* Free any memory associated with the png_info_struct */ -PNG_EXPORT(63, void, png_destroy_info_struct, (png_const_structrp png_ptr, - png_infopp info_ptr_ptr)); - -/* Free any memory associated with the png_struct and the png_info_structs */ -PNG_EXPORT(64, void, png_destroy_read_struct, (png_structpp png_ptr_ptr, - png_infopp info_ptr_ptr, png_infopp end_info_ptr_ptr)); - -/* Free any memory associated with the png_struct and the png_info_structs */ -PNG_EXPORT(65, void, png_destroy_write_struct, (png_structpp png_ptr_ptr, - png_infopp info_ptr_ptr)); - -/* Set the libpng method of handling chunk CRC errors */ -PNG_EXPORT(66, void, png_set_crc_action, (png_structrp png_ptr, int crit_action, - int ancil_action)); - -/* Values for png_set_crc_action() say how to handle CRC errors in - * ancillary and critical chunks, and whether to use the data contained - * therein. Note that it is impossible to "discard" data in a critical - * chunk. For versions prior to 0.90, the action was always error/quit, - * whereas in version 0.90 and later, the action for CRC errors in ancillary - * chunks is warn/discard. These values should NOT be changed. - * - * value action:critical action:ancillary - */ -#define PNG_CRC_DEFAULT 0 /* error/quit warn/discard data */ -#define PNG_CRC_ERROR_QUIT 1 /* error/quit error/quit */ -#define PNG_CRC_WARN_DISCARD 2 /* (INVALID) warn/discard data */ -#define PNG_CRC_WARN_USE 3 /* warn/use data warn/use data */ -#define PNG_CRC_QUIET_USE 4 /* quiet/use data quiet/use data */ -#define PNG_CRC_NO_CHANGE 5 /* use current value use current value */ - -#ifdef PNG_WRITE_SUPPORTED -/* These functions give the user control over the scan-line filtering in - * libpng and the compression methods used by zlib. These functions are - * mainly useful for testing, as the defaults should work with most users. - * Those users who are tight on memory or want faster performance at the - * expense of compression can modify them. See the compression library - * header file (zlib.h) for an explanation of the compression functions. - */ - -/* Set the filtering method(s) used by libpng. Currently, the only valid - * value for "method" is 0. - */ -PNG_EXPORT(67, void, png_set_filter, (png_structrp png_ptr, int method, - int filters)); -#endif /* WRITE */ - -/* Flags for png_set_filter() to say which filters to use. The flags - * are chosen so that they don't conflict with real filter types - * below, in case they are supplied instead of the #defined constants. - * These values should NOT be changed. - */ -#define PNG_NO_FILTERS 0x00 -#define PNG_FILTER_NONE 0x08 -#define PNG_FILTER_SUB 0x10 -#define PNG_FILTER_UP 0x20 -#define PNG_FILTER_AVG 0x40 -#define PNG_FILTER_PAETH 0x80 -#define PNG_FAST_FILTERS (PNG_FILTER_NONE | PNG_FILTER_SUB | PNG_FILTER_UP) -#define PNG_ALL_FILTERS (PNG_FAST_FILTERS | PNG_FILTER_AVG | PNG_FILTER_PAETH) - -/* Filter values (not flags) - used in pngwrite.c, pngwutil.c for now. - * These defines should NOT be changed. - */ -#define PNG_FILTER_VALUE_NONE 0 -#define PNG_FILTER_VALUE_SUB 1 -#define PNG_FILTER_VALUE_UP 2 -#define PNG_FILTER_VALUE_AVG 3 -#define PNG_FILTER_VALUE_PAETH 4 -#define PNG_FILTER_VALUE_LAST 5 - -#ifdef PNG_WRITE_SUPPORTED -#ifdef PNG_WRITE_WEIGHTED_FILTER_SUPPORTED /* DEPRECATED */ -PNG_FP_EXPORT(68, void, png_set_filter_heuristics, (png_structrp png_ptr, - int heuristic_method, int num_weights, png_const_doublep filter_weights, - png_const_doublep filter_costs)) -PNG_FIXED_EXPORT(209, void, png_set_filter_heuristics_fixed, - (png_structrp png_ptr, int heuristic_method, int num_weights, - png_const_fixed_point_p filter_weights, - png_const_fixed_point_p filter_costs)) -#endif /* WRITE_WEIGHTED_FILTER */ - -/* The following are no longer used and will be removed from libpng-1.7: */ -#define PNG_FILTER_HEURISTIC_DEFAULT 0 /* Currently "UNWEIGHTED" */ -#define PNG_FILTER_HEURISTIC_UNWEIGHTED 1 /* Used by libpng < 0.95 */ -#define PNG_FILTER_HEURISTIC_WEIGHTED 2 /* Experimental feature */ -#define PNG_FILTER_HEURISTIC_LAST 3 /* Not a valid value */ - -/* Set the library compression level. Currently, valid values range from - * 0 - 9, corresponding directly to the zlib compression levels 0 - 9 - * (0 - no compression, 9 - "maximal" compression). Note that tests have - * shown that zlib compression levels 3-6 usually perform as well as level 9 - * for PNG images, and do considerably fewer calculations. In the future, - * these values may not correspond directly to the zlib compression levels. - */ -#ifdef PNG_WRITE_CUSTOMIZE_COMPRESSION_SUPPORTED -PNG_EXPORT(69, void, png_set_compression_level, (png_structrp png_ptr, - int level)); - -PNG_EXPORT(70, void, png_set_compression_mem_level, (png_structrp png_ptr, - int mem_level)); - -PNG_EXPORT(71, void, png_set_compression_strategy, (png_structrp png_ptr, - int strategy)); - -/* If PNG_WRITE_OPTIMIZE_CMF_SUPPORTED is defined, libpng will use a - * smaller value of window_bits if it can do so safely. - */ -PNG_EXPORT(72, void, png_set_compression_window_bits, (png_structrp png_ptr, - int window_bits)); - -PNG_EXPORT(73, void, png_set_compression_method, (png_structrp png_ptr, - int method)); -#endif /* WRITE_CUSTOMIZE_COMPRESSION */ - -#ifdef PNG_WRITE_CUSTOMIZE_ZTXT_COMPRESSION_SUPPORTED -/* Also set zlib parameters for compressing non-IDAT chunks */ -PNG_EXPORT(222, void, png_set_text_compression_level, (png_structrp png_ptr, - int level)); - -PNG_EXPORT(223, void, png_set_text_compression_mem_level, (png_structrp png_ptr, - int mem_level)); - -PNG_EXPORT(224, void, png_set_text_compression_strategy, (png_structrp png_ptr, - int strategy)); - -/* If PNG_WRITE_OPTIMIZE_CMF_SUPPORTED is defined, libpng will use a - * smaller value of window_bits if it can do so safely. - */ -PNG_EXPORT(225, void, png_set_text_compression_window_bits, - (png_structrp png_ptr, int window_bits)); - -PNG_EXPORT(226, void, png_set_text_compression_method, (png_structrp png_ptr, - int method)); -#endif /* WRITE_CUSTOMIZE_ZTXT_COMPRESSION */ -#endif /* WRITE */ - -/* These next functions are called for input/output, memory, and error - * handling. They are in the file pngrio.c, pngwio.c, and pngerror.c, - * and call standard C I/O routines such as fread(), fwrite(), and - * fprintf(). These functions can be made to use other I/O routines - * at run time for those applications that need to handle I/O in a - * different manner by calling png_set_???_fn(). See libpng-manual.txt for - * more information. - */ - -#ifdef PNG_STDIO_SUPPORTED -/* Initialize the input/output for the PNG file to the default functions. */ -PNG_EXPORT(74, void, png_init_io, (png_structrp png_ptr, png_FILE_p fp)); -#endif - -/* Replace the (error and abort), and warning functions with user - * supplied functions. If no messages are to be printed you must still - * write and use replacement functions. The replacement error_fn should - * still do a longjmp to the last setjmp location if you are using this - * method of error handling. If error_fn or warning_fn is NULL, the - * default function will be used. - */ - -PNG_EXPORT(75, void, png_set_error_fn, (png_structrp png_ptr, - png_voidp error_ptr, png_error_ptr error_fn, png_error_ptr warning_fn)); - -/* Return the user pointer associated with the error functions */ -PNG_EXPORT(76, png_voidp, png_get_error_ptr, (png_const_structrp png_ptr)); - -/* Replace the default data output functions with a user supplied one(s). - * If buffered output is not used, then output_flush_fn can be set to NULL. - * If PNG_WRITE_FLUSH_SUPPORTED is not defined at libpng compile time - * output_flush_fn will be ignored (and thus can be NULL). - * It is probably a mistake to use NULL for output_flush_fn if - * write_data_fn is not also NULL unless you have built libpng with - * PNG_WRITE_FLUSH_SUPPORTED undefined, because in this case libpng's - * default flush function, which uses the standard *FILE structure, will - * be used. - */ -PNG_EXPORT(77, void, png_set_write_fn, (png_structrp png_ptr, png_voidp io_ptr, - png_rw_ptr write_data_fn, png_flush_ptr output_flush_fn)); - -/* Replace the default data input function with a user supplied one. */ -PNG_EXPORT(78, void, png_set_read_fn, (png_structrp png_ptr, png_voidp io_ptr, - png_rw_ptr read_data_fn)); - -/* Return the user pointer associated with the I/O functions */ -PNG_EXPORT(79, png_voidp, png_get_io_ptr, (png_const_structrp png_ptr)); - -PNG_EXPORT(80, void, png_set_read_status_fn, (png_structrp png_ptr, - png_read_status_ptr read_row_fn)); - -PNG_EXPORT(81, void, png_set_write_status_fn, (png_structrp png_ptr, - png_write_status_ptr write_row_fn)); - -#ifdef PNG_USER_MEM_SUPPORTED -/* Replace the default memory allocation functions with user supplied one(s). */ -PNG_EXPORT(82, void, png_set_mem_fn, (png_structrp png_ptr, png_voidp mem_ptr, - png_malloc_ptr malloc_fn, png_free_ptr free_fn)); -/* Return the user pointer associated with the memory functions */ -PNG_EXPORT(83, png_voidp, png_get_mem_ptr, (png_const_structrp png_ptr)); -#endif - -#ifdef PNG_READ_USER_TRANSFORM_SUPPORTED -PNG_EXPORT(84, void, png_set_read_user_transform_fn, (png_structrp png_ptr, - png_user_transform_ptr read_user_transform_fn)); -#endif - -#ifdef PNG_WRITE_USER_TRANSFORM_SUPPORTED -PNG_EXPORT(85, void, png_set_write_user_transform_fn, (png_structrp png_ptr, - png_user_transform_ptr write_user_transform_fn)); -#endif - -#ifdef PNG_USER_TRANSFORM_PTR_SUPPORTED -PNG_EXPORT(86, void, png_set_user_transform_info, (png_structrp png_ptr, - png_voidp user_transform_ptr, int user_transform_depth, - int user_transform_channels)); -/* Return the user pointer associated with the user transform functions */ -PNG_EXPORT(87, png_voidp, png_get_user_transform_ptr, - (png_const_structrp png_ptr)); -#endif - -#ifdef PNG_USER_TRANSFORM_INFO_SUPPORTED -/* Return information about the row currently being processed. Note that these - * APIs do not fail but will return unexpected results if called outside a user - * transform callback. Also note that when transforming an interlaced image the - * row number is the row number within the sub-image of the interlace pass, so - * the value will increase to the height of the sub-image (not the full image) - * then reset to 0 for the next pass. - * - * Use PNG_ROW_FROM_PASS_ROW(row, pass) and PNG_COL_FROM_PASS_COL(col, pass) to - * find the output pixel (x,y) given an interlaced sub-image pixel - * (row,col,pass). (See below for these macros.) - */ -PNG_EXPORT(217, png_uint_32, png_get_current_row_number, (png_const_structrp)); -PNG_EXPORT(218, png_byte, png_get_current_pass_number, (png_const_structrp)); -#endif - -#ifdef PNG_READ_USER_CHUNKS_SUPPORTED -/* This callback is called only for *unknown* chunks. If - * PNG_HANDLE_AS_UNKNOWN_SUPPORTED is set then it is possible to set known - * chunks to be treated as unknown, however in this case the callback must do - * any processing required by the chunk (e.g. by calling the appropriate - * png_set_ APIs.) - * - * There is no write support - on write, by default, all the chunks in the - * 'unknown' list are written in the specified position. - * - * The integer return from the callback function is interpreted thus: - * - * negative: An error occurred; png_chunk_error will be called. - * zero: The chunk was not handled, the chunk will be saved. A critical - * chunk will cause an error at this point unless it is to be saved. - * positive: The chunk was handled, libpng will ignore/discard it. - * - * See "INTERACTION WITH USER CHUNK CALLBACKS" below for important notes about - * how this behavior will change in libpng 1.7 - */ -PNG_EXPORT(88, void, png_set_read_user_chunk_fn, (png_structrp png_ptr, - png_voidp user_chunk_ptr, png_user_chunk_ptr read_user_chunk_fn)); -#endif - -#ifdef PNG_USER_CHUNKS_SUPPORTED -PNG_EXPORT(89, png_voidp, png_get_user_chunk_ptr, (png_const_structrp png_ptr)); -#endif - -#ifdef PNG_PROGRESSIVE_READ_SUPPORTED -/* Sets the function callbacks for the push reader, and a pointer to a - * user-defined structure available to the callback functions. - */ -PNG_EXPORT(90, void, png_set_progressive_read_fn, (png_structrp png_ptr, - png_voidp progressive_ptr, png_progressive_info_ptr info_fn, - png_progressive_row_ptr row_fn, png_progressive_end_ptr end_fn)); - -/* Returns the user pointer associated with the push read functions */ -PNG_EXPORT(91, png_voidp, png_get_progressive_ptr, - (png_const_structrp png_ptr)); - -/* Function to be called when data becomes available */ -PNG_EXPORT(92, void, png_process_data, (png_structrp png_ptr, - png_inforp info_ptr, png_bytep buffer, size_t buffer_size)); - -/* A function which may be called *only* within png_process_data to stop the - * processing of any more data. The function returns the number of bytes - * remaining, excluding any that libpng has cached internally. A subsequent - * call to png_process_data must supply these bytes again. If the argument - * 'save' is set to true the routine will first save all the pending data and - * will always return 0. - */ -PNG_EXPORT(219, size_t, png_process_data_pause, (png_structrp, int save)); - -/* A function which may be called *only* outside (after) a call to - * png_process_data. It returns the number of bytes of data to skip in the - * input. Normally it will return 0, but if it returns a non-zero value the - * application must skip than number of bytes of input data and pass the - * following data to the next call to png_process_data. - */ -PNG_EXPORT(220, png_uint_32, png_process_data_skip, (png_structrp)); - -/* Function that combines rows. 'new_row' is a flag that should come from - * the callback and be non-NULL if anything needs to be done; the library - * stores its own version of the new data internally and ignores the passed - * in value. - */ -PNG_EXPORT(93, void, png_progressive_combine_row, (png_const_structrp png_ptr, - png_bytep old_row, png_const_bytep new_row)); -#endif /* PROGRESSIVE_READ */ - -PNG_EXPORTA(94, png_voidp, png_malloc, (png_const_structrp png_ptr, - png_alloc_size_t size), PNG_ALLOCATED); -/* Added at libpng version 1.4.0 */ -PNG_EXPORTA(95, png_voidp, png_calloc, (png_const_structrp png_ptr, - png_alloc_size_t size), PNG_ALLOCATED); - -/* Added at libpng version 1.2.4 */ -PNG_EXPORTA(96, png_voidp, png_malloc_warn, (png_const_structrp png_ptr, - png_alloc_size_t size), PNG_ALLOCATED); - -/* Frees a pointer allocated by png_malloc() */ -PNG_EXPORT(97, void, png_free, (png_const_structrp png_ptr, png_voidp ptr)); - -/* Free data that was allocated internally */ -PNG_EXPORT(98, void, png_free_data, (png_const_structrp png_ptr, - png_inforp info_ptr, png_uint_32 free_me, int num)); - -/* Reassign the responsibility for freeing existing data, whether allocated - * by libpng or by the application; this works on the png_info structure passed - * in, without changing the state for other png_info structures. - */ -PNG_EXPORT(99, void, png_data_freer, (png_const_structrp png_ptr, - png_inforp info_ptr, int freer, png_uint_32 mask)); - -/* Assignments for png_data_freer */ -#define PNG_DESTROY_WILL_FREE_DATA 1 -#define PNG_SET_WILL_FREE_DATA 1 -#define PNG_USER_WILL_FREE_DATA 2 -/* Flags for png_ptr->free_me and info_ptr->free_me */ -#define PNG_FREE_HIST 0x0008U -#define PNG_FREE_ICCP 0x0010U -#define PNG_FREE_SPLT 0x0020U -#define PNG_FREE_ROWS 0x0040U -#define PNG_FREE_PCAL 0x0080U -#define PNG_FREE_SCAL 0x0100U -#ifdef PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED -# define PNG_FREE_UNKN 0x0200U -#endif -/* PNG_FREE_LIST 0x0400U removed in 1.6.0 because it is ignored */ -#define PNG_FREE_PLTE 0x1000U -#define PNG_FREE_TRNS 0x2000U -#define PNG_FREE_TEXT 0x4000U -#define PNG_FREE_EXIF 0x8000U /* Added at libpng-1.6.31 */ -#define PNG_FREE_ALL 0xffffU -#define PNG_FREE_MUL 0x4220U /* PNG_FREE_SPLT|PNG_FREE_TEXT|PNG_FREE_UNKN */ - -#ifdef PNG_USER_MEM_SUPPORTED -PNG_EXPORTA(100, png_voidp, png_malloc_default, (png_const_structrp png_ptr, - png_alloc_size_t size), PNG_ALLOCATED PNG_DEPRECATED); -PNG_EXPORTA(101, void, png_free_default, (png_const_structrp png_ptr, - png_voidp ptr), PNG_DEPRECATED); -#endif - -#ifdef PNG_ERROR_TEXT_SUPPORTED -/* Fatal error in PNG image of libpng - can't continue */ -PNG_EXPORTA(102, void, png_error, (png_const_structrp png_ptr, - png_const_charp error_message), PNG_NORETURN); - -/* The same, but the chunk name is prepended to the error string. */ -PNG_EXPORTA(103, void, png_chunk_error, (png_const_structrp png_ptr, - png_const_charp error_message), PNG_NORETURN); - -#else -/* Fatal error in PNG image of libpng - can't continue */ -PNG_EXPORTA(104, void, png_err, (png_const_structrp png_ptr), PNG_NORETURN); -# define png_error(s1,s2) png_err(s1) -# define png_chunk_error(s1,s2) png_err(s1) -#endif - -#ifdef PNG_WARNINGS_SUPPORTED -/* Non-fatal error in libpng. Can continue, but may have a problem. */ -PNG_EXPORT(105, void, png_warning, (png_const_structrp png_ptr, - png_const_charp warning_message)); - -/* Non-fatal error in libpng, chunk name is prepended to message. */ -PNG_EXPORT(106, void, png_chunk_warning, (png_const_structrp png_ptr, - png_const_charp warning_message)); -#else -# define png_warning(s1,s2) ((void)(s1)) -# define png_chunk_warning(s1,s2) ((void)(s1)) -#endif - -#ifdef PNG_BENIGN_ERRORS_SUPPORTED -/* Benign error in libpng. Can continue, but may have a problem. - * User can choose whether to handle as a fatal error or as a warning. */ -PNG_EXPORT(107, void, png_benign_error, (png_const_structrp png_ptr, - png_const_charp warning_message)); - -#ifdef PNG_READ_SUPPORTED -/* Same, chunk name is prepended to message (only during read) */ -PNG_EXPORT(108, void, png_chunk_benign_error, (png_const_structrp png_ptr, - png_const_charp warning_message)); -#endif - -PNG_EXPORT(109, void, png_set_benign_errors, - (png_structrp png_ptr, int allowed)); -#else -# ifdef PNG_ALLOW_BENIGN_ERRORS -# define png_benign_error png_warning -# define png_chunk_benign_error png_chunk_warning -# else -# define png_benign_error png_error -# define png_chunk_benign_error png_chunk_error -# endif -#endif - -/* The png_set_ functions are for storing values in the png_info_struct. - * Similarly, the png_get_ calls are used to read values from the - * png_info_struct, either storing the parameters in the passed variables, or - * setting pointers into the png_info_struct where the data is stored. The - * png_get_ functions return a non-zero value if the data was available - * in info_ptr, or return zero and do not change any of the parameters if the - * data was not available. - * - * These functions should be used instead of directly accessing png_info - * to avoid problems with future changes in the size and internal layout of - * png_info_struct. - */ -/* Returns "flag" if chunk data is valid in info_ptr. */ -PNG_EXPORT(110, png_uint_32, png_get_valid, (png_const_structrp png_ptr, - png_const_inforp info_ptr, png_uint_32 flag)); - -/* Returns number of bytes needed to hold a transformed row. */ -PNG_EXPORT(111, size_t, png_get_rowbytes, (png_const_structrp png_ptr, - png_const_inforp info_ptr)); - -#ifdef PNG_INFO_IMAGE_SUPPORTED -/* Returns row_pointers, which is an array of pointers to scanlines that was - * returned from png_read_png(). - */ -PNG_EXPORT(112, png_bytepp, png_get_rows, (png_const_structrp png_ptr, - png_const_inforp info_ptr)); - -/* Set row_pointers, which is an array of pointers to scanlines for use - * by png_write_png(). - */ -PNG_EXPORT(113, void, png_set_rows, (png_const_structrp png_ptr, - png_inforp info_ptr, png_bytepp row_pointers)); -#endif - -/* Returns number of color channels in image. */ -PNG_EXPORT(114, png_byte, png_get_channels, (png_const_structrp png_ptr, - png_const_inforp info_ptr)); - -#ifdef PNG_EASY_ACCESS_SUPPORTED -/* Returns image width in pixels. */ -PNG_EXPORT(115, png_uint_32, png_get_image_width, (png_const_structrp png_ptr, - png_const_inforp info_ptr)); - -/* Returns image height in pixels. */ -PNG_EXPORT(116, png_uint_32, png_get_image_height, (png_const_structrp png_ptr, - png_const_inforp info_ptr)); - -/* Returns image bit_depth. */ -PNG_EXPORT(117, png_byte, png_get_bit_depth, (png_const_structrp png_ptr, - png_const_inforp info_ptr)); - -/* Returns image color_type. */ -PNG_EXPORT(118, png_byte, png_get_color_type, (png_const_structrp png_ptr, - png_const_inforp info_ptr)); - -/* Returns image filter_type. */ -PNG_EXPORT(119, png_byte, png_get_filter_type, (png_const_structrp png_ptr, - png_const_inforp info_ptr)); - -/* Returns image interlace_type. */ -PNG_EXPORT(120, png_byte, png_get_interlace_type, (png_const_structrp png_ptr, - png_const_inforp info_ptr)); - -/* Returns image compression_type. */ -PNG_EXPORT(121, png_byte, png_get_compression_type, (png_const_structrp png_ptr, - png_const_inforp info_ptr)); - -/* Returns image resolution in pixels per meter, from pHYs chunk data. */ -PNG_EXPORT(122, png_uint_32, png_get_pixels_per_meter, - (png_const_structrp png_ptr, png_const_inforp info_ptr)); -PNG_EXPORT(123, png_uint_32, png_get_x_pixels_per_meter, - (png_const_structrp png_ptr, png_const_inforp info_ptr)); -PNG_EXPORT(124, png_uint_32, png_get_y_pixels_per_meter, - (png_const_structrp png_ptr, png_const_inforp info_ptr)); - -/* Returns pixel aspect ratio, computed from pHYs chunk data. */ -PNG_FP_EXPORT(125, float, png_get_pixel_aspect_ratio, - (png_const_structrp png_ptr, png_const_inforp info_ptr)) -PNG_FIXED_EXPORT(210, png_fixed_point, png_get_pixel_aspect_ratio_fixed, - (png_const_structrp png_ptr, png_const_inforp info_ptr)) - -/* Returns image x, y offset in pixels or microns, from oFFs chunk data. */ -PNG_EXPORT(126, png_int_32, png_get_x_offset_pixels, - (png_const_structrp png_ptr, png_const_inforp info_ptr)); -PNG_EXPORT(127, png_int_32, png_get_y_offset_pixels, - (png_const_structrp png_ptr, png_const_inforp info_ptr)); -PNG_EXPORT(128, png_int_32, png_get_x_offset_microns, - (png_const_structrp png_ptr, png_const_inforp info_ptr)); -PNG_EXPORT(129, png_int_32, png_get_y_offset_microns, - (png_const_structrp png_ptr, png_const_inforp info_ptr)); - -#endif /* EASY_ACCESS */ - -#ifdef PNG_READ_SUPPORTED -/* Returns pointer to signature string read from PNG header */ -PNG_EXPORT(130, png_const_bytep, png_get_signature, (png_const_structrp png_ptr, - png_const_inforp info_ptr)); -#endif - -#ifdef PNG_bKGD_SUPPORTED -PNG_EXPORT(131, png_uint_32, png_get_bKGD, (png_const_structrp png_ptr, - png_inforp info_ptr, png_color_16p *background)); -#endif - -#ifdef PNG_bKGD_SUPPORTED -PNG_EXPORT(132, void, png_set_bKGD, (png_const_structrp png_ptr, - png_inforp info_ptr, png_const_color_16p background)); -#endif - -#ifdef PNG_cHRM_SUPPORTED -PNG_FP_EXPORT(133, png_uint_32, png_get_cHRM, (png_const_structrp png_ptr, - png_const_inforp info_ptr, double *white_x, double *white_y, double *red_x, - double *red_y, double *green_x, double *green_y, double *blue_x, - double *blue_y)) -PNG_FP_EXPORT(230, png_uint_32, png_get_cHRM_XYZ, (png_const_structrp png_ptr, - png_const_inforp info_ptr, double *red_X, double *red_Y, double *red_Z, - double *green_X, double *green_Y, double *green_Z, double *blue_X, - double *blue_Y, double *blue_Z)) -PNG_FIXED_EXPORT(134, png_uint_32, png_get_cHRM_fixed, - (png_const_structrp png_ptr, png_const_inforp info_ptr, - png_fixed_point *int_white_x, png_fixed_point *int_white_y, - png_fixed_point *int_red_x, png_fixed_point *int_red_y, - png_fixed_point *int_green_x, png_fixed_point *int_green_y, - png_fixed_point *int_blue_x, png_fixed_point *int_blue_y)) -PNG_FIXED_EXPORT(231, png_uint_32, png_get_cHRM_XYZ_fixed, - (png_const_structrp png_ptr, png_const_inforp info_ptr, - png_fixed_point *int_red_X, png_fixed_point *int_red_Y, - png_fixed_point *int_red_Z, png_fixed_point *int_green_X, - png_fixed_point *int_green_Y, png_fixed_point *int_green_Z, - png_fixed_point *int_blue_X, png_fixed_point *int_blue_Y, - png_fixed_point *int_blue_Z)) -#endif - -#ifdef PNG_cHRM_SUPPORTED -PNG_FP_EXPORT(135, void, png_set_cHRM, (png_const_structrp png_ptr, - png_inforp info_ptr, - double white_x, double white_y, double red_x, double red_y, double green_x, - double green_y, double blue_x, double blue_y)) -PNG_FP_EXPORT(232, void, png_set_cHRM_XYZ, (png_const_structrp png_ptr, - png_inforp info_ptr, double red_X, double red_Y, double red_Z, - double green_X, double green_Y, double green_Z, double blue_X, - double blue_Y, double blue_Z)) -PNG_FIXED_EXPORT(136, void, png_set_cHRM_fixed, (png_const_structrp png_ptr, - png_inforp info_ptr, png_fixed_point int_white_x, - png_fixed_point int_white_y, png_fixed_point int_red_x, - png_fixed_point int_red_y, png_fixed_point int_green_x, - png_fixed_point int_green_y, png_fixed_point int_blue_x, - png_fixed_point int_blue_y)) -PNG_FIXED_EXPORT(233, void, png_set_cHRM_XYZ_fixed, (png_const_structrp png_ptr, - png_inforp info_ptr, png_fixed_point int_red_X, png_fixed_point int_red_Y, - png_fixed_point int_red_Z, png_fixed_point int_green_X, - png_fixed_point int_green_Y, png_fixed_point int_green_Z, - png_fixed_point int_blue_X, png_fixed_point int_blue_Y, - png_fixed_point int_blue_Z)) -#endif - -#ifdef PNG_eXIf_SUPPORTED -PNG_EXPORT(246, png_uint_32, png_get_eXIf, (png_const_structrp png_ptr, - png_inforp info_ptr, png_bytep *exif)); -PNG_EXPORT(247, void, png_set_eXIf, (png_const_structrp png_ptr, - png_inforp info_ptr, png_bytep exif)); - -PNG_EXPORT(248, png_uint_32, png_get_eXIf_1, (png_const_structrp png_ptr, - png_const_inforp info_ptr, png_uint_32 *num_exif, png_bytep *exif)); -PNG_EXPORT(249, void, png_set_eXIf_1, (png_const_structrp png_ptr, - png_inforp info_ptr, png_uint_32 num_exif, png_bytep exif)); -#endif - -#ifdef PNG_gAMA_SUPPORTED -PNG_FP_EXPORT(137, png_uint_32, png_get_gAMA, (png_const_structrp png_ptr, - png_const_inforp info_ptr, double *file_gamma)) -PNG_FIXED_EXPORT(138, png_uint_32, png_get_gAMA_fixed, - (png_const_structrp png_ptr, png_const_inforp info_ptr, - png_fixed_point *int_file_gamma)) -#endif - -#ifdef PNG_gAMA_SUPPORTED -PNG_FP_EXPORT(139, void, png_set_gAMA, (png_const_structrp png_ptr, - png_inforp info_ptr, double file_gamma)) -PNG_FIXED_EXPORT(140, void, png_set_gAMA_fixed, (png_const_structrp png_ptr, - png_inforp info_ptr, png_fixed_point int_file_gamma)) -#endif - -#ifdef PNG_hIST_SUPPORTED -PNG_EXPORT(141, png_uint_32, png_get_hIST, (png_const_structrp png_ptr, - png_inforp info_ptr, png_uint_16p *hist)); -PNG_EXPORT(142, void, png_set_hIST, (png_const_structrp png_ptr, - png_inforp info_ptr, png_const_uint_16p hist)); -#endif - -PNG_EXPORT(143, png_uint_32, png_get_IHDR, (png_const_structrp png_ptr, - png_const_inforp info_ptr, png_uint_32 *width, png_uint_32 *height, - int *bit_depth, int *color_type, int *interlace_method, - int *compression_method, int *filter_method)); - -PNG_EXPORT(144, void, png_set_IHDR, (png_const_structrp png_ptr, - png_inforp info_ptr, png_uint_32 width, png_uint_32 height, int bit_depth, - int color_type, int interlace_method, int compression_method, - int filter_method)); - -#ifdef PNG_oFFs_SUPPORTED -PNG_EXPORT(145, png_uint_32, png_get_oFFs, (png_const_structrp png_ptr, - png_const_inforp info_ptr, png_int_32 *offset_x, png_int_32 *offset_y, - int *unit_type)); -#endif - -#ifdef PNG_oFFs_SUPPORTED -PNG_EXPORT(146, void, png_set_oFFs, (png_const_structrp png_ptr, - png_inforp info_ptr, png_int_32 offset_x, png_int_32 offset_y, - int unit_type)); -#endif - -#ifdef PNG_pCAL_SUPPORTED -PNG_EXPORT(147, png_uint_32, png_get_pCAL, (png_const_structrp png_ptr, - png_inforp info_ptr, png_charp *purpose, png_int_32 *X0, - png_int_32 *X1, int *type, int *nparams, png_charp *units, - png_charpp *params)); -#endif - -#ifdef PNG_pCAL_SUPPORTED -PNG_EXPORT(148, void, png_set_pCAL, (png_const_structrp png_ptr, - png_inforp info_ptr, png_const_charp purpose, png_int_32 X0, png_int_32 X1, - int type, int nparams, png_const_charp units, png_charpp params)); -#endif - -#ifdef PNG_pHYs_SUPPORTED -PNG_EXPORT(149, png_uint_32, png_get_pHYs, (png_const_structrp png_ptr, - png_const_inforp info_ptr, png_uint_32 *res_x, png_uint_32 *res_y, - int *unit_type)); -#endif - -#ifdef PNG_pHYs_SUPPORTED -PNG_EXPORT(150, void, png_set_pHYs, (png_const_structrp png_ptr, - png_inforp info_ptr, png_uint_32 res_x, png_uint_32 res_y, int unit_type)); -#endif - -PNG_EXPORT(151, png_uint_32, png_get_PLTE, (png_const_structrp png_ptr, - png_inforp info_ptr, png_colorp *palette, int *num_palette)); - -PNG_EXPORT(152, void, png_set_PLTE, (png_structrp png_ptr, - png_inforp info_ptr, png_const_colorp palette, int num_palette)); - -#ifdef PNG_sBIT_SUPPORTED -PNG_EXPORT(153, png_uint_32, png_get_sBIT, (png_const_structrp png_ptr, - png_inforp info_ptr, png_color_8p *sig_bit)); -#endif - -#ifdef PNG_sBIT_SUPPORTED -PNG_EXPORT(154, void, png_set_sBIT, (png_const_structrp png_ptr, - png_inforp info_ptr, png_const_color_8p sig_bit)); -#endif - -#ifdef PNG_sRGB_SUPPORTED -PNG_EXPORT(155, png_uint_32, png_get_sRGB, (png_const_structrp png_ptr, - png_const_inforp info_ptr, int *file_srgb_intent)); -#endif - -#ifdef PNG_sRGB_SUPPORTED -PNG_EXPORT(156, void, png_set_sRGB, (png_const_structrp png_ptr, - png_inforp info_ptr, int srgb_intent)); -PNG_EXPORT(157, void, png_set_sRGB_gAMA_and_cHRM, (png_const_structrp png_ptr, - png_inforp info_ptr, int srgb_intent)); -#endif - -#ifdef PNG_iCCP_SUPPORTED -PNG_EXPORT(158, png_uint_32, png_get_iCCP, (png_const_structrp png_ptr, - png_inforp info_ptr, png_charpp name, int *compression_type, - png_bytepp profile, png_uint_32 *proflen)); -#endif - -#ifdef PNG_iCCP_SUPPORTED -PNG_EXPORT(159, void, png_set_iCCP, (png_const_structrp png_ptr, - png_inforp info_ptr, png_const_charp name, int compression_type, - png_const_bytep profile, png_uint_32 proflen)); -#endif - -#ifdef PNG_sPLT_SUPPORTED -PNG_EXPORT(160, int, png_get_sPLT, (png_const_structrp png_ptr, - png_inforp info_ptr, png_sPLT_tpp entries)); -#endif - -#ifdef PNG_sPLT_SUPPORTED -PNG_EXPORT(161, void, png_set_sPLT, (png_const_structrp png_ptr, - png_inforp info_ptr, png_const_sPLT_tp entries, int nentries)); -#endif - -#ifdef PNG_TEXT_SUPPORTED -/* png_get_text also returns the number of text chunks in *num_text */ -PNG_EXPORT(162, int, png_get_text, (png_const_structrp png_ptr, - png_inforp info_ptr, png_textp *text_ptr, int *num_text)); -#endif - -/* Note while png_set_text() will accept a structure whose text, - * language, and translated keywords are NULL pointers, the structure - * returned by png_get_text will always contain regular - * zero-terminated C strings. They might be empty strings but - * they will never be NULL pointers. - */ - -#ifdef PNG_TEXT_SUPPORTED -PNG_EXPORT(163, void, png_set_text, (png_const_structrp png_ptr, - png_inforp info_ptr, png_const_textp text_ptr, int num_text)); -#endif - -#ifdef PNG_tIME_SUPPORTED -PNG_EXPORT(164, png_uint_32, png_get_tIME, (png_const_structrp png_ptr, - png_inforp info_ptr, png_timep *mod_time)); -#endif - -#ifdef PNG_tIME_SUPPORTED -PNG_EXPORT(165, void, png_set_tIME, (png_const_structrp png_ptr, - png_inforp info_ptr, png_const_timep mod_time)); -#endif - -#ifdef PNG_tRNS_SUPPORTED -PNG_EXPORT(166, png_uint_32, png_get_tRNS, (png_const_structrp png_ptr, - png_inforp info_ptr, png_bytep *trans_alpha, int *num_trans, - png_color_16p *trans_color)); -#endif - -#ifdef PNG_tRNS_SUPPORTED -PNG_EXPORT(167, void, png_set_tRNS, (png_structrp png_ptr, - png_inforp info_ptr, png_const_bytep trans_alpha, int num_trans, - png_const_color_16p trans_color)); -#endif - -#ifdef PNG_sCAL_SUPPORTED -PNG_FP_EXPORT(168, png_uint_32, png_get_sCAL, (png_const_structrp png_ptr, - png_const_inforp info_ptr, int *unit, double *width, double *height)) -#if defined(PNG_FLOATING_ARITHMETIC_SUPPORTED) || \ - defined(PNG_FLOATING_POINT_SUPPORTED) -/* NOTE: this API is currently implemented using floating point arithmetic, - * consequently it can only be used on systems with floating point support. - * In any case the range of values supported by png_fixed_point is small and it - * is highly recommended that png_get_sCAL_s be used instead. - */ -PNG_FIXED_EXPORT(214, png_uint_32, png_get_sCAL_fixed, - (png_const_structrp png_ptr, png_const_inforp info_ptr, int *unit, - png_fixed_point *width, png_fixed_point *height)) -#endif -PNG_EXPORT(169, png_uint_32, png_get_sCAL_s, - (png_const_structrp png_ptr, png_const_inforp info_ptr, int *unit, - png_charpp swidth, png_charpp sheight)); - -PNG_FP_EXPORT(170, void, png_set_sCAL, (png_const_structrp png_ptr, - png_inforp info_ptr, int unit, double width, double height)) -PNG_FIXED_EXPORT(213, void, png_set_sCAL_fixed, (png_const_structrp png_ptr, - png_inforp info_ptr, int unit, png_fixed_point width, - png_fixed_point height)) -PNG_EXPORT(171, void, png_set_sCAL_s, (png_const_structrp png_ptr, - png_inforp info_ptr, int unit, - png_const_charp swidth, png_const_charp sheight)); -#endif /* sCAL */ - -#ifdef PNG_SET_UNKNOWN_CHUNKS_SUPPORTED -/* Provide the default handling for all unknown chunks or, optionally, for - * specific unknown chunks. - * - * NOTE: prior to 1.6.0 the handling specified for particular chunks on read was - * ignored and the default was used, the per-chunk setting only had an effect on - * write. If you wish to have chunk-specific handling on read in code that must - * work on earlier versions you must use a user chunk callback to specify the - * desired handling (keep or discard.) - * - * The 'keep' parameter is a PNG_HANDLE_CHUNK_ value as listed below. The - * parameter is interpreted as follows: - * - * READ: - * PNG_HANDLE_CHUNK_AS_DEFAULT: - * Known chunks: do normal libpng processing, do not keep the chunk (but - * see the comments below about PNG_HANDLE_AS_UNKNOWN_SUPPORTED) - * Unknown chunks: for a specific chunk use the global default, when used - * as the default discard the chunk data. - * PNG_HANDLE_CHUNK_NEVER: - * Discard the chunk data. - * PNG_HANDLE_CHUNK_IF_SAFE: - * Keep the chunk data if the chunk is not critical else raise a chunk - * error. - * PNG_HANDLE_CHUNK_ALWAYS: - * Keep the chunk data. - * - * If the chunk data is saved it can be retrieved using png_get_unknown_chunks, - * below. Notice that specifying "AS_DEFAULT" as a global default is equivalent - * to specifying "NEVER", however when "AS_DEFAULT" is used for specific chunks - * it simply resets the behavior to the libpng default. - * - * INTERACTION WITH USER CHUNK CALLBACKS: - * The per-chunk handling is always used when there is a png_user_chunk_ptr - * callback and the callback returns 0; the chunk is then always stored *unless* - * it is critical and the per-chunk setting is other than ALWAYS. Notice that - * the global default is *not* used in this case. (In effect the per-chunk - * value is incremented to at least IF_SAFE.) - * - * IMPORTANT NOTE: this behavior will change in libpng 1.7 - the global and - * per-chunk defaults will be honored. If you want to preserve the current - * behavior when your callback returns 0 you must set PNG_HANDLE_CHUNK_IF_SAFE - * as the default - if you don't do this libpng 1.6 will issue a warning. - * - * If you want unhandled unknown chunks to be discarded in libpng 1.6 and - * earlier simply return '1' (handled). - * - * PNG_HANDLE_AS_UNKNOWN_SUPPORTED: - * If this is *not* set known chunks will always be handled by libpng and - * will never be stored in the unknown chunk list. Known chunks listed to - * png_set_keep_unknown_chunks will have no effect. If it is set then known - * chunks listed with a keep other than AS_DEFAULT will *never* be processed - * by libpng, in addition critical chunks must either be processed by the - * callback or saved. - * - * The IHDR and IEND chunks must not be listed. Because this turns off the - * default handling for chunks that would otherwise be recognized the - * behavior of libpng transformations may well become incorrect! - * - * WRITE: - * When writing chunks the options only apply to the chunks specified by - * png_set_unknown_chunks (below), libpng will *always* write known chunks - * required by png_set_ calls and will always write the core critical chunks - * (as required for PLTE). - * - * Each chunk in the png_set_unknown_chunks list is looked up in the - * png_set_keep_unknown_chunks list to find the keep setting, this is then - * interpreted as follows: - * - * PNG_HANDLE_CHUNK_AS_DEFAULT: - * Write safe-to-copy chunks and write other chunks if the global - * default is set to _ALWAYS, otherwise don't write this chunk. - * PNG_HANDLE_CHUNK_NEVER: - * Do not write the chunk. - * PNG_HANDLE_CHUNK_IF_SAFE: - * Write the chunk if it is safe-to-copy, otherwise do not write it. - * PNG_HANDLE_CHUNK_ALWAYS: - * Write the chunk. - * - * Note that the default behavior is effectively the opposite of the read case - - * in read unknown chunks are not stored by default, in write they are written - * by default. Also the behavior of PNG_HANDLE_CHUNK_IF_SAFE is very different - * - on write the safe-to-copy bit is checked, on read the critical bit is - * checked and on read if the chunk is critical an error will be raised. - * - * num_chunks: - * =========== - * If num_chunks is positive, then the "keep" parameter specifies the manner - * for handling only those chunks appearing in the chunk_list array, - * otherwise the chunk list array is ignored. - * - * If num_chunks is 0 the "keep" parameter specifies the default behavior for - * unknown chunks, as described above. - * - * If num_chunks is negative, then the "keep" parameter specifies the manner - * for handling all unknown chunks plus all chunks recognized by libpng - * except for the IHDR, PLTE, tRNS, IDAT, and IEND chunks (which continue to - * be processed by libpng. - */ -#ifdef PNG_HANDLE_AS_UNKNOWN_SUPPORTED -PNG_EXPORT(172, void, png_set_keep_unknown_chunks, (png_structrp png_ptr, - int keep, png_const_bytep chunk_list, int num_chunks)); -#endif /* HANDLE_AS_UNKNOWN */ - -/* The "keep" PNG_HANDLE_CHUNK_ parameter for the specified chunk is returned; - * the result is therefore true (non-zero) if special handling is required, - * false for the default handling. - */ -PNG_EXPORT(173, int, png_handle_as_unknown, (png_const_structrp png_ptr, - png_const_bytep chunk_name)); -#endif /* SET_UNKNOWN_CHUNKS */ - -#ifdef PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED -PNG_EXPORT(174, void, png_set_unknown_chunks, (png_const_structrp png_ptr, - png_inforp info_ptr, png_const_unknown_chunkp unknowns, - int num_unknowns)); - /* NOTE: prior to 1.6.0 this routine set the 'location' field of the added - * unknowns to the location currently stored in the png_struct. This is - * invariably the wrong value on write. To fix this call the following API - * for each chunk in the list with the correct location. If you know your - * code won't be compiled on earlier versions you can rely on - * png_set_unknown_chunks(write-ptr, png_get_unknown_chunks(read-ptr)) doing - * the correct thing. - */ - -PNG_EXPORT(175, void, png_set_unknown_chunk_location, - (png_const_structrp png_ptr, png_inforp info_ptr, int chunk, int location)); - -PNG_EXPORT(176, int, png_get_unknown_chunks, (png_const_structrp png_ptr, - png_inforp info_ptr, png_unknown_chunkpp entries)); -#endif - -/* Png_free_data() will turn off the "valid" flag for anything it frees. - * If you need to turn it off for a chunk that your application has freed, - * you can use png_set_invalid(png_ptr, info_ptr, PNG_INFO_CHNK); - */ -PNG_EXPORT(177, void, png_set_invalid, (png_const_structrp png_ptr, - png_inforp info_ptr, int mask)); - -#ifdef PNG_INFO_IMAGE_SUPPORTED -/* The "params" pointer is currently not used and is for future expansion. */ -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -PNG_EXPORT(178, void, png_read_png, (png_structrp png_ptr, png_inforp info_ptr, - int transforms, png_voidp params)); -#endif -#ifdef PNG_WRITE_SUPPORTED -PNG_EXPORT(179, void, png_write_png, (png_structrp png_ptr, png_inforp info_ptr, - int transforms, png_voidp params)); -#endif -#endif - -PNG_EXPORT(180, png_const_charp, png_get_copyright, - (png_const_structrp png_ptr)); -PNG_EXPORT(181, png_const_charp, png_get_header_ver, - (png_const_structrp png_ptr)); -PNG_EXPORT(182, png_const_charp, png_get_header_version, - (png_const_structrp png_ptr)); -PNG_EXPORT(183, png_const_charp, png_get_libpng_ver, - (png_const_structrp png_ptr)); - -#ifdef PNG_MNG_FEATURES_SUPPORTED -PNG_EXPORT(184, png_uint_32, png_permit_mng_features, (png_structrp png_ptr, - png_uint_32 mng_features_permitted)); -#endif - -/* For use in png_set_keep_unknown, added to version 1.2.6 */ -#define PNG_HANDLE_CHUNK_AS_DEFAULT 0 -#define PNG_HANDLE_CHUNK_NEVER 1 -#define PNG_HANDLE_CHUNK_IF_SAFE 2 -#define PNG_HANDLE_CHUNK_ALWAYS 3 -#define PNG_HANDLE_CHUNK_LAST 4 - -/* Strip the prepended error numbers ("#nnn ") from error and warning - * messages before passing them to the error or warning handler. - */ -#ifdef PNG_ERROR_NUMBERS_SUPPORTED -PNG_EXPORT(185, void, png_set_strip_error_numbers, (png_structrp png_ptr, - png_uint_32 strip_mode)); -#endif - -/* Added in libpng-1.2.6 */ -#ifdef PNG_SET_USER_LIMITS_SUPPORTED -PNG_EXPORT(186, void, png_set_user_limits, (png_structrp png_ptr, - png_uint_32 user_width_max, png_uint_32 user_height_max)); -PNG_EXPORT(187, png_uint_32, png_get_user_width_max, - (png_const_structrp png_ptr)); -PNG_EXPORT(188, png_uint_32, png_get_user_height_max, - (png_const_structrp png_ptr)); -/* Added in libpng-1.4.0 */ -PNG_EXPORT(189, void, png_set_chunk_cache_max, (png_structrp png_ptr, - png_uint_32 user_chunk_cache_max)); -PNG_EXPORT(190, png_uint_32, png_get_chunk_cache_max, - (png_const_structrp png_ptr)); -/* Added in libpng-1.4.1 */ -PNG_EXPORT(191, void, png_set_chunk_malloc_max, (png_structrp png_ptr, - png_alloc_size_t user_chunk_cache_max)); -PNG_EXPORT(192, png_alloc_size_t, png_get_chunk_malloc_max, - (png_const_structrp png_ptr)); -#endif - -#if defined(PNG_INCH_CONVERSIONS_SUPPORTED) -PNG_EXPORT(193, png_uint_32, png_get_pixels_per_inch, - (png_const_structrp png_ptr, png_const_inforp info_ptr)); - -PNG_EXPORT(194, png_uint_32, png_get_x_pixels_per_inch, - (png_const_structrp png_ptr, png_const_inforp info_ptr)); - -PNG_EXPORT(195, png_uint_32, png_get_y_pixels_per_inch, - (png_const_structrp png_ptr, png_const_inforp info_ptr)); - -PNG_FP_EXPORT(196, float, png_get_x_offset_inches, - (png_const_structrp png_ptr, png_const_inforp info_ptr)) -#ifdef PNG_FIXED_POINT_SUPPORTED /* otherwise not implemented. */ -PNG_FIXED_EXPORT(211, png_fixed_point, png_get_x_offset_inches_fixed, - (png_const_structrp png_ptr, png_const_inforp info_ptr)) -#endif - -PNG_FP_EXPORT(197, float, png_get_y_offset_inches, (png_const_structrp png_ptr, - png_const_inforp info_ptr)) -#ifdef PNG_FIXED_POINT_SUPPORTED /* otherwise not implemented. */ -PNG_FIXED_EXPORT(212, png_fixed_point, png_get_y_offset_inches_fixed, - (png_const_structrp png_ptr, png_const_inforp info_ptr)) -#endif - -# ifdef PNG_pHYs_SUPPORTED -PNG_EXPORT(198, png_uint_32, png_get_pHYs_dpi, (png_const_structrp png_ptr, - png_const_inforp info_ptr, png_uint_32 *res_x, png_uint_32 *res_y, - int *unit_type)); -# endif /* pHYs */ -#endif /* INCH_CONVERSIONS */ - -/* Added in libpng-1.4.0 */ -#ifdef PNG_IO_STATE_SUPPORTED -PNG_EXPORT(199, png_uint_32, png_get_io_state, (png_const_structrp png_ptr)); - -/* Removed from libpng 1.6; use png_get_io_chunk_type. */ -PNG_REMOVED(200, png_const_bytep, png_get_io_chunk_name, (png_structrp png_ptr), - PNG_DEPRECATED) - -PNG_EXPORT(216, png_uint_32, png_get_io_chunk_type, - (png_const_structrp png_ptr)); - -/* The flags returned by png_get_io_state() are the following: */ -# define PNG_IO_NONE 0x0000 /* no I/O at this moment */ -# define PNG_IO_READING 0x0001 /* currently reading */ -# define PNG_IO_WRITING 0x0002 /* currently writing */ -# define PNG_IO_SIGNATURE 0x0010 /* currently at the file signature */ -# define PNG_IO_CHUNK_HDR 0x0020 /* currently at the chunk header */ -# define PNG_IO_CHUNK_DATA 0x0040 /* currently at the chunk data */ -# define PNG_IO_CHUNK_CRC 0x0080 /* currently at the chunk crc */ -# define PNG_IO_MASK_OP 0x000f /* current operation: reading/writing */ -# define PNG_IO_MASK_LOC 0x00f0 /* current location: sig/hdr/data/crc */ -#endif /* IO_STATE */ - -/* Interlace support. The following macros are always defined so that if - * libpng interlace handling is turned off the macros may be used to handle - * interlaced images within the application. - */ -#define PNG_INTERLACE_ADAM7_PASSES 7 - -/* Two macros to return the first row and first column of the original, - * full, image which appears in a given pass. 'pass' is in the range 0 - * to 6 and the result is in the range 0 to 7. - */ -#define PNG_PASS_START_ROW(pass) (((1&~(pass))<<(3-((pass)>>1)))&7) -#define PNG_PASS_START_COL(pass) (((1& (pass))<<(3-(((pass)+1)>>1)))&7) - -/* A macro to return the offset between pixels in the output row for a pair of - * pixels in the input - effectively the inverse of the 'COL_SHIFT' macro that - * follows. Note that ROW_OFFSET is the offset from one row to the next whereas - * COL_OFFSET is from one column to the next, within a row. - */ -#define PNG_PASS_ROW_OFFSET(pass) ((pass)>2?(8>>(((pass)-1)>>1)):8) -#define PNG_PASS_COL_OFFSET(pass) (1<<((7-(pass))>>1)) - -/* Two macros to help evaluate the number of rows or columns in each - * pass. This is expressed as a shift - effectively log2 of the number or - * rows or columns in each 8x8 tile of the original image. - */ -#define PNG_PASS_ROW_SHIFT(pass) ((pass)>2?(8-(pass))>>1:3) -#define PNG_PASS_COL_SHIFT(pass) ((pass)>1?(7-(pass))>>1:3) - -/* Hence two macros to determine the number of rows or columns in a given - * pass of an image given its height or width. In fact these macros may - * return non-zero even though the sub-image is empty, because the other - * dimension may be empty for a small image. - */ -#define PNG_PASS_ROWS(height, pass) (((height)+(((1<>PNG_PASS_ROW_SHIFT(pass)) -#define PNG_PASS_COLS(width, pass) (((width)+(((1<>PNG_PASS_COL_SHIFT(pass)) - -/* For the reader row callbacks (both progressive and sequential) it is - * necessary to find the row in the output image given a row in an interlaced - * image, so two more macros: - */ -#define PNG_ROW_FROM_PASS_ROW(y_in, pass) \ - (((y_in)<>(((7-(off))-(pass))<<2)) & 0xF) | \ - ((0x01145AF0>>(((7-(off))-(pass))<<2)) & 0xF0)) - -#define PNG_ROW_IN_INTERLACE_PASS(y, pass) \ - ((PNG_PASS_MASK(pass,0) >> ((y)&7)) & 1) -#define PNG_COL_IN_INTERLACE_PASS(x, pass) \ - ((PNG_PASS_MASK(pass,1) >> ((x)&7)) & 1) - -#ifdef PNG_READ_COMPOSITE_NODIV_SUPPORTED -/* With these routines we avoid an integer divide, which will be slower on - * most machines. However, it does take more operations than the corresponding - * divide method, so it may be slower on a few RISC systems. There are two - * shifts (by 8 or 16 bits) and an addition, versus a single integer divide. - * - * Note that the rounding factors are NOT supposed to be the same! 128 and - * 32768 are correct for the NODIV code; 127 and 32767 are correct for the - * standard method. - * - * [Optimized code by Greg Roelofs and Mark Adler...blame us for bugs. :-) ] - */ - - /* fg and bg should be in `gamma 1.0' space; alpha is the opacity */ - -# define png_composite(composite, fg, alpha, bg) \ - { \ - png_uint_16 temp = (png_uint_16)((png_uint_16)(fg) \ - * (png_uint_16)(alpha) \ - + (png_uint_16)(bg)*(png_uint_16)(255 \ - - (png_uint_16)(alpha)) + 128); \ - (composite) = (png_byte)(((temp + (temp >> 8)) >> 8) & 0xff); \ - } - -# define png_composite_16(composite, fg, alpha, bg) \ - { \ - png_uint_32 temp = (png_uint_32)((png_uint_32)(fg) \ - * (png_uint_32)(alpha) \ - + (png_uint_32)(bg)*(65535 \ - - (png_uint_32)(alpha)) + 32768); \ - (composite) = (png_uint_16)(0xffff & ((temp + (temp >> 16)) >> 16)); \ - } - -#else /* Standard method using integer division */ - -# define png_composite(composite, fg, alpha, bg) \ - (composite) = \ - (png_byte)(0xff & (((png_uint_16)(fg) * (png_uint_16)(alpha) + \ - (png_uint_16)(bg) * (png_uint_16)(255 - (png_uint_16)(alpha)) + \ - 127) / 255)) - -# define png_composite_16(composite, fg, alpha, bg) \ - (composite) = \ - (png_uint_16)(0xffff & (((png_uint_32)(fg) * (png_uint_32)(alpha) + \ - (png_uint_32)(bg)*(png_uint_32)(65535 - (png_uint_32)(alpha)) + \ - 32767) / 65535)) -#endif /* READ_COMPOSITE_NODIV */ - -#ifdef PNG_READ_INT_FUNCTIONS_SUPPORTED -PNG_EXPORT(201, png_uint_32, png_get_uint_32, (png_const_bytep buf)); -PNG_EXPORT(202, png_uint_16, png_get_uint_16, (png_const_bytep buf)); -PNG_EXPORT(203, png_int_32, png_get_int_32, (png_const_bytep buf)); -#endif - -PNG_EXPORT(204, png_uint_32, png_get_uint_31, (png_const_structrp png_ptr, - png_const_bytep buf)); -/* No png_get_int_16 -- may be added if there's a real need for it. */ - -/* Place a 32-bit number into a buffer in PNG byte order (big-endian). */ -#ifdef PNG_WRITE_INT_FUNCTIONS_SUPPORTED -PNG_EXPORT(205, void, png_save_uint_32, (png_bytep buf, png_uint_32 i)); -#endif -#ifdef PNG_SAVE_INT_32_SUPPORTED -PNG_EXPORT(206, void, png_save_int_32, (png_bytep buf, png_int_32 i)); -#endif - -/* Place a 16-bit number into a buffer in PNG byte order. - * The parameter is declared unsigned int, not png_uint_16, - * just to avoid potential problems on pre-ANSI C compilers. - */ -#ifdef PNG_WRITE_INT_FUNCTIONS_SUPPORTED -PNG_EXPORT(207, void, png_save_uint_16, (png_bytep buf, unsigned int i)); -/* No png_save_int_16 -- may be added if there's a real need for it. */ -#endif - -#ifdef PNG_USE_READ_MACROS -/* Inline macros to do direct reads of bytes from the input buffer. - * The png_get_int_32() routine assumes we are using two's complement - * format for negative values, which is almost certainly true. - */ -# define PNG_get_uint_32(buf) \ - (((png_uint_32)(*(buf)) << 24) + \ - ((png_uint_32)(*((buf) + 1)) << 16) + \ - ((png_uint_32)(*((buf) + 2)) << 8) + \ - ((png_uint_32)(*((buf) + 3)))) - - /* From libpng-1.4.0 until 1.4.4, the png_get_uint_16 macro (but not the - * function) incorrectly returned a value of type png_uint_32. - */ -# define PNG_get_uint_16(buf) \ - ((png_uint_16) \ - (((unsigned int)(*(buf)) << 8) + \ - ((unsigned int)(*((buf) + 1))))) - -# define PNG_get_int_32(buf) \ - ((png_int_32)((*(buf) & 0x80) \ - ? -((png_int_32)(((png_get_uint_32(buf)^0xffffffffU)+1U)&0x7fffffffU)) \ - : (png_int_32)png_get_uint_32(buf))) - -/* If PNG_PREFIX is defined the same thing as below happens in pnglibconf.h, - * but defining a macro name prefixed with PNG_PREFIX. - */ -# ifndef PNG_PREFIX -# define png_get_uint_32(buf) PNG_get_uint_32(buf) -# define png_get_uint_16(buf) PNG_get_uint_16(buf) -# define png_get_int_32(buf) PNG_get_int_32(buf) -# endif -#else -# ifdef PNG_PREFIX - /* No macros; revert to the (redefined) function */ -# define PNG_get_uint_32 (png_get_uint_32) -# define PNG_get_uint_16 (png_get_uint_16) -# define PNG_get_int_32 (png_get_int_32) -# endif -#endif - -#ifdef PNG_CHECK_FOR_INVALID_INDEX_SUPPORTED -PNG_EXPORT(242, void, png_set_check_for_invalid_index, - (png_structrp png_ptr, int allowed)); -# ifdef PNG_GET_PALETTE_MAX_SUPPORTED -PNG_EXPORT(243, int, png_get_palette_max, (png_const_structp png_ptr, - png_const_infop info_ptr)); -# endif -#endif /* CHECK_FOR_INVALID_INDEX */ - -/******************************************************************************* - * Section 5: SIMPLIFIED API - ******************************************************************************* - * - * Please read the documentation in libpng-manual.txt (TODO: write said - * documentation) if you don't understand what follows. - * - * The simplified API hides the details of both libpng and the PNG file format - * itself. It allows PNG files to be read into a very limited number of - * in-memory bitmap formats or to be written from the same formats. If these - * formats do not accommodate your needs then you can, and should, use the more - * sophisticated APIs above - these support a wide variety of in-memory formats - * and a wide variety of sophisticated transformations to those formats as well - * as a wide variety of APIs to manipulate ancillary information. - * - * To read a PNG file using the simplified API: - * - * 1) Declare a 'png_image' structure (see below) on the stack, set the - * version field to PNG_IMAGE_VERSION and the 'opaque' pointer to NULL - * (this is REQUIRED, your program may crash if you don't do it.) - * 2) Call the appropriate png_image_begin_read... function. - * 3) Set the png_image 'format' member to the required sample format. - * 4) Allocate a buffer for the image and, if required, the color-map. - * 5) Call png_image_finish_read to read the image and, if required, the - * color-map into your buffers. - * - * There are no restrictions on the format of the PNG input itself; all valid - * color types, bit depths, and interlace methods are acceptable, and the - * input image is transformed as necessary to the requested in-memory format - * during the png_image_finish_read() step. The only caveat is that if you - * request a color-mapped image from a PNG that is full-color or makes - * complex use of an alpha channel the transformation is extremely lossy and the - * result may look terrible. - * - * To write a PNG file using the simplified API: - * - * 1) Declare a 'png_image' structure on the stack and memset() it to all zero. - * 2) Initialize the members of the structure that describe the image, setting - * the 'format' member to the format of the image samples. - * 3) Call the appropriate png_image_write... function with a pointer to the - * image and, if necessary, the color-map to write the PNG data. - * - * png_image is a structure that describes the in-memory format of an image - * when it is being read or defines the in-memory format of an image that you - * need to write: - */ -#if defined(PNG_SIMPLIFIED_READ_SUPPORTED) || \ - defined(PNG_SIMPLIFIED_WRITE_SUPPORTED) - -#define PNG_IMAGE_VERSION 1 - -typedef struct png_control *png_controlp; -typedef struct -{ - png_controlp opaque; /* Initialize to NULL, free with png_image_free */ - png_uint_32 version; /* Set to PNG_IMAGE_VERSION */ - png_uint_32 width; /* Image width in pixels (columns) */ - png_uint_32 height; /* Image height in pixels (rows) */ - png_uint_32 format; /* Image format as defined below */ - png_uint_32 flags; /* A bit mask containing informational flags */ - png_uint_32 colormap_entries; - /* Number of entries in the color-map */ - - /* In the event of an error or warning the following field will be set to a - * non-zero value and the 'message' field will contain a '\0' terminated - * string with the libpng error or warning message. If both warnings and - * an error were encountered, only the error is recorded. If there - * are multiple warnings, only the first one is recorded. - * - * The upper 30 bits of this value are reserved, the low two bits contain - * a value as follows: - */ -# define PNG_IMAGE_WARNING 1 -# define PNG_IMAGE_ERROR 2 - /* - * The result is a two-bit code such that a value more than 1 indicates - * a failure in the API just called: - * - * 0 - no warning or error - * 1 - warning - * 2 - error - * 3 - error preceded by warning - */ -# define PNG_IMAGE_FAILED(png_cntrl) ((((png_cntrl).warning_or_error)&0x03)>1) - - png_uint_32 warning_or_error; - - char message[64]; -} png_image, *png_imagep; - -/* The samples of the image have one to four channels whose components have - * original values in the range 0 to 1.0: - * - * 1: A single gray or luminance channel (G). - * 2: A gray/luminance channel and an alpha channel (GA). - * 3: Three red, green, blue color channels (RGB). - * 4: Three color channels and an alpha channel (RGBA). - * - * The components are encoded in one of two ways: - * - * a) As a small integer, value 0..255, contained in a single byte. For the - * alpha channel the original value is simply value/255. For the color or - * luminance channels the value is encoded according to the sRGB specification - * and matches the 8-bit format expected by typical display devices. - * - * The color/gray channels are not scaled (pre-multiplied) by the alpha - * channel and are suitable for passing to color management software. - * - * b) As a value in the range 0..65535, contained in a 2-byte integer. All - * channels can be converted to the original value by dividing by 65535; all - * channels are linear. Color channels use the RGB encoding (RGB end-points) of - * the sRGB specification. This encoding is identified by the - * PNG_FORMAT_FLAG_LINEAR flag below. - * - * When the simplified API needs to convert between sRGB and linear colorspaces, - * the actual sRGB transfer curve defined in the sRGB specification (see the - * article at ) is used, not the gamma=1/2.2 - * approximation used elsewhere in libpng. - * - * When an alpha channel is present it is expected to denote pixel coverage - * of the color or luminance channels and is returned as an associated alpha - * channel: the color/gray channels are scaled (pre-multiplied) by the alpha - * value. - * - * The samples are either contained directly in the image data, between 1 and 8 - * bytes per pixel according to the encoding, or are held in a color-map indexed - * by bytes in the image data. In the case of a color-map the color-map entries - * are individual samples, encoded as above, and the image data has one byte per - * pixel to select the relevant sample from the color-map. - */ - -/* PNG_FORMAT_* - * - * #defines to be used in png_image::format. Each #define identifies a - * particular layout of sample data and, if present, alpha values. There are - * separate defines for each of the two component encodings. - * - * A format is built up using single bit flag values. All combinations are - * valid. Formats can be built up from the flag values or you can use one of - * the predefined values below. When testing formats always use the FORMAT_FLAG - * macros to test for individual features - future versions of the library may - * add new flags. - * - * When reading or writing color-mapped images the format should be set to the - * format of the entries in the color-map then png_image_{read,write}_colormap - * called to read or write the color-map and set the format correctly for the - * image data. Do not set the PNG_FORMAT_FLAG_COLORMAP bit directly! - * - * NOTE: libpng can be built with particular features disabled. If you see - * compiler errors because the definition of one of the following flags has been - * compiled out it is because libpng does not have the required support. It is - * possible, however, for the libpng configuration to enable the format on just - * read or just write; in that case you may see an error at run time. You can - * guard against this by checking for the definition of the appropriate - * "_SUPPORTED" macro, one of: - * - * PNG_SIMPLIFIED_{READ,WRITE}_{BGR,AFIRST}_SUPPORTED - */ -#define PNG_FORMAT_FLAG_ALPHA 0x01U /* format with an alpha channel */ -#define PNG_FORMAT_FLAG_COLOR 0x02U /* color format: otherwise grayscale */ -#define PNG_FORMAT_FLAG_LINEAR 0x04U /* 2-byte channels else 1-byte */ -#define PNG_FORMAT_FLAG_COLORMAP 0x08U /* image data is color-mapped */ - -#ifdef PNG_FORMAT_BGR_SUPPORTED -# define PNG_FORMAT_FLAG_BGR 0x10U /* BGR colors, else order is RGB */ -#endif - -#ifdef PNG_FORMAT_AFIRST_SUPPORTED -# define PNG_FORMAT_FLAG_AFIRST 0x20U /* alpha channel comes first */ -#endif - -#define PNG_FORMAT_FLAG_ASSOCIATED_ALPHA 0x40U /* alpha channel is associated */ - -/* Commonly used formats have predefined macros. - * - * First the single byte (sRGB) formats: - */ -#define PNG_FORMAT_GRAY 0 -#define PNG_FORMAT_GA PNG_FORMAT_FLAG_ALPHA -#define PNG_FORMAT_AG (PNG_FORMAT_GA|PNG_FORMAT_FLAG_AFIRST) -#define PNG_FORMAT_RGB PNG_FORMAT_FLAG_COLOR -#define PNG_FORMAT_BGR (PNG_FORMAT_FLAG_COLOR|PNG_FORMAT_FLAG_BGR) -#define PNG_FORMAT_RGBA (PNG_FORMAT_RGB|PNG_FORMAT_FLAG_ALPHA) -#define PNG_FORMAT_ARGB (PNG_FORMAT_RGBA|PNG_FORMAT_FLAG_AFIRST) -#define PNG_FORMAT_BGRA (PNG_FORMAT_BGR|PNG_FORMAT_FLAG_ALPHA) -#define PNG_FORMAT_ABGR (PNG_FORMAT_BGRA|PNG_FORMAT_FLAG_AFIRST) - -/* Then the linear 2-byte formats. When naming these "Y" is used to - * indicate a luminance (gray) channel. - */ -#define PNG_FORMAT_LINEAR_Y PNG_FORMAT_FLAG_LINEAR -#define PNG_FORMAT_LINEAR_Y_ALPHA (PNG_FORMAT_FLAG_LINEAR|PNG_FORMAT_FLAG_ALPHA) -#define PNG_FORMAT_LINEAR_RGB (PNG_FORMAT_FLAG_LINEAR|PNG_FORMAT_FLAG_COLOR) -#define PNG_FORMAT_LINEAR_RGB_ALPHA \ - (PNG_FORMAT_FLAG_LINEAR|PNG_FORMAT_FLAG_COLOR|PNG_FORMAT_FLAG_ALPHA) - -/* With color-mapped formats the image data is one byte for each pixel, the byte - * is an index into the color-map which is formatted as above. To obtain a - * color-mapped format it is sufficient just to add the PNG_FOMAT_FLAG_COLORMAP - * to one of the above definitions, or you can use one of the definitions below. - */ -#define PNG_FORMAT_RGB_COLORMAP (PNG_FORMAT_RGB|PNG_FORMAT_FLAG_COLORMAP) -#define PNG_FORMAT_BGR_COLORMAP (PNG_FORMAT_BGR|PNG_FORMAT_FLAG_COLORMAP) -#define PNG_FORMAT_RGBA_COLORMAP (PNG_FORMAT_RGBA|PNG_FORMAT_FLAG_COLORMAP) -#define PNG_FORMAT_ARGB_COLORMAP (PNG_FORMAT_ARGB|PNG_FORMAT_FLAG_COLORMAP) -#define PNG_FORMAT_BGRA_COLORMAP (PNG_FORMAT_BGRA|PNG_FORMAT_FLAG_COLORMAP) -#define PNG_FORMAT_ABGR_COLORMAP (PNG_FORMAT_ABGR|PNG_FORMAT_FLAG_COLORMAP) - -/* PNG_IMAGE macros - * - * These are convenience macros to derive information from a png_image - * structure. The PNG_IMAGE_SAMPLE_ macros return values appropriate to the - * actual image sample values - either the entries in the color-map or the - * pixels in the image. The PNG_IMAGE_PIXEL_ macros return corresponding values - * for the pixels and will always return 1 for color-mapped formats. The - * remaining macros return information about the rows in the image and the - * complete image. - * - * NOTE: All the macros that take a png_image::format parameter are compile time - * constants if the format parameter is, itself, a constant. Therefore these - * macros can be used in array declarations and case labels where required. - * Similarly the macros are also pre-processor constants (sizeof is not used) so - * they can be used in #if tests. - * - * First the information about the samples. - */ -#define PNG_IMAGE_SAMPLE_CHANNELS(fmt)\ - (((fmt)&(PNG_FORMAT_FLAG_COLOR|PNG_FORMAT_FLAG_ALPHA))+1) - /* Return the total number of channels in a given format: 1..4 */ - -#define PNG_IMAGE_SAMPLE_COMPONENT_SIZE(fmt)\ - ((((fmt) & PNG_FORMAT_FLAG_LINEAR) >> 2)+1) - /* Return the size in bytes of a single component of a pixel or color-map - * entry (as appropriate) in the image: 1 or 2. - */ - -#define PNG_IMAGE_SAMPLE_SIZE(fmt)\ - (PNG_IMAGE_SAMPLE_CHANNELS(fmt) * PNG_IMAGE_SAMPLE_COMPONENT_SIZE(fmt)) - /* This is the size of the sample data for one sample. If the image is - * color-mapped it is the size of one color-map entry (and image pixels are - * one byte in size), otherwise it is the size of one image pixel. - */ - -#define PNG_IMAGE_MAXIMUM_COLORMAP_COMPONENTS(fmt)\ - (PNG_IMAGE_SAMPLE_CHANNELS(fmt) * 256) - /* The maximum size of the color-map required by the format expressed in a - * count of components. This can be used to compile-time allocate a - * color-map: - * - * png_uint_16 colormap[PNG_IMAGE_MAXIMUM_COLORMAP_COMPONENTS(linear_fmt)]; - * - * png_byte colormap[PNG_IMAGE_MAXIMUM_COLORMAP_COMPONENTS(sRGB_fmt)]; - * - * Alternatively use the PNG_IMAGE_COLORMAP_SIZE macro below to use the - * information from one of the png_image_begin_read_ APIs and dynamically - * allocate the required memory. - */ - -/* Corresponding information about the pixels */ -#define PNG_IMAGE_PIXEL_(test,fmt)\ - (((fmt)&PNG_FORMAT_FLAG_COLORMAP)?1:test(fmt)) - -#define PNG_IMAGE_PIXEL_CHANNELS(fmt)\ - PNG_IMAGE_PIXEL_(PNG_IMAGE_SAMPLE_CHANNELS,fmt) - /* The number of separate channels (components) in a pixel; 1 for a - * color-mapped image. - */ - -#define PNG_IMAGE_PIXEL_COMPONENT_SIZE(fmt)\ - PNG_IMAGE_PIXEL_(PNG_IMAGE_SAMPLE_COMPONENT_SIZE,fmt) - /* The size, in bytes, of each component in a pixel; 1 for a color-mapped - * image. - */ - -#define PNG_IMAGE_PIXEL_SIZE(fmt) PNG_IMAGE_PIXEL_(PNG_IMAGE_SAMPLE_SIZE,fmt) - /* The size, in bytes, of a complete pixel; 1 for a color-mapped image. */ - -/* Information about the whole row, or whole image */ -#define PNG_IMAGE_ROW_STRIDE(image)\ - (PNG_IMAGE_PIXEL_CHANNELS((image).format) * (image).width) - /* Return the total number of components in a single row of the image; this - * is the minimum 'row stride', the minimum count of components between each - * row. For a color-mapped image this is the minimum number of bytes in a - * row. - * - * WARNING: this macro overflows for some images with more than one component - * and very large image widths. libpng will refuse to process an image where - * this macro would overflow. - */ - -#define PNG_IMAGE_BUFFER_SIZE(image, row_stride)\ - (PNG_IMAGE_PIXEL_COMPONENT_SIZE((image).format)*(image).height*(row_stride)) - /* Return the size, in bytes, of an image buffer given a png_image and a row - * stride - the number of components to leave space for in each row. - * - * WARNING: this macro overflows a 32-bit integer for some large PNG images, - * libpng will refuse to process an image where such an overflow would occur. - */ - -#define PNG_IMAGE_SIZE(image)\ - PNG_IMAGE_BUFFER_SIZE(image, PNG_IMAGE_ROW_STRIDE(image)) - /* Return the size, in bytes, of the image in memory given just a png_image; - * the row stride is the minimum stride required for the image. - */ - -#define PNG_IMAGE_COLORMAP_SIZE(image)\ - (PNG_IMAGE_SAMPLE_SIZE((image).format) * (image).colormap_entries) - /* Return the size, in bytes, of the color-map of this image. If the image - * format is not a color-map format this will return a size sufficient for - * 256 entries in the given format; check PNG_FORMAT_FLAG_COLORMAP if - * you don't want to allocate a color-map in this case. - */ - -/* PNG_IMAGE_FLAG_* - * - * Flags containing additional information about the image are held in the - * 'flags' field of png_image. - */ -#define PNG_IMAGE_FLAG_COLORSPACE_NOT_sRGB 0x01 - /* This indicates that the RGB values of the in-memory bitmap do not - * correspond to the red, green and blue end-points defined by sRGB. - */ - -#define PNG_IMAGE_FLAG_FAST 0x02 - /* On write emphasise speed over compression; the resultant PNG file will be - * larger but will be produced significantly faster, particular for large - * images. Do not use this option for images which will be distributed, only - * used it when producing intermediate files that will be read back in - * repeatedly. For a typical 24-bit image the option will double the read - * speed at the cost of increasing the image size by 25%, however for many - * more compressible images the PNG file can be 10 times larger with only a - * slight speed gain. - */ - -#define PNG_IMAGE_FLAG_16BIT_sRGB 0x04 - /* On read if the image is a 16-bit per component image and there is no gAMA - * or sRGB chunk assume that the components are sRGB encoded. Notice that - * images output by the simplified API always have gamma information; setting - * this flag only affects the interpretation of 16-bit images from an - * external source. It is recommended that the application expose this flag - * to the user; the user can normally easily recognize the difference between - * linear and sRGB encoding. This flag has no effect on write - the data - * passed to the write APIs must have the correct encoding (as defined - * above.) - * - * If the flag is not set (the default) input 16-bit per component data is - * assumed to be linear. - * - * NOTE: the flag can only be set after the png_image_begin_read_ call, - * because that call initializes the 'flags' field. - */ - -#ifdef PNG_SIMPLIFIED_READ_SUPPORTED -/* READ APIs - * --------- - * - * The png_image passed to the read APIs must have been initialized by setting - * the png_controlp field 'opaque' to NULL (or, safer, memset the whole thing.) - */ -#ifdef PNG_STDIO_SUPPORTED -PNG_EXPORT(234, int, png_image_begin_read_from_file, (png_imagep image, - const char *file_name)); - /* The named file is opened for read and the image header is filled in - * from the PNG header in the file. - */ - -PNG_EXPORT(235, int, png_image_begin_read_from_stdio, (png_imagep image, - FILE* file)); - /* The PNG header is read from the stdio FILE object. */ -#endif /* STDIO */ - -PNG_EXPORT(236, int, png_image_begin_read_from_memory, (png_imagep image, - png_const_voidp memory, size_t size)); - /* The PNG header is read from the given memory buffer. */ - -PNG_EXPORT(237, int, png_image_finish_read, (png_imagep image, - png_const_colorp background, void *buffer, png_int_32 row_stride, - void *colormap)); - /* Finish reading the image into the supplied buffer and clean up the - * png_image structure. - * - * row_stride is the step, in byte or 2-byte units as appropriate, - * between adjacent rows. A positive stride indicates that the top-most row - * is first in the buffer - the normal top-down arrangement. A negative - * stride indicates that the bottom-most row is first in the buffer. - * - * background need only be supplied if an alpha channel must be removed from - * a png_byte format and the removal is to be done by compositing on a solid - * color; otherwise it may be NULL and any composition will be done directly - * onto the buffer. The value is an sRGB color to use for the background, - * for grayscale output the green channel is used. - * - * background must be supplied when an alpha channel must be removed from a - * single byte color-mapped output format, in other words if: - * - * 1) The original format from png_image_begin_read_from_* had - * PNG_FORMAT_FLAG_ALPHA set. - * 2) The format set by the application does not. - * 3) The format set by the application has PNG_FORMAT_FLAG_COLORMAP set and - * PNG_FORMAT_FLAG_LINEAR *not* set. - * - * For linear output removing the alpha channel is always done by compositing - * on black and background is ignored. - * - * colormap must be supplied when PNG_FORMAT_FLAG_COLORMAP is set. It must - * be at least the size (in bytes) returned by PNG_IMAGE_COLORMAP_SIZE. - * image->colormap_entries will be updated to the actual number of entries - * written to the colormap; this may be less than the original value. - */ - -PNG_EXPORT(238, void, png_image_free, (png_imagep image)); - /* Free any data allocated by libpng in image->opaque, setting the pointer to - * NULL. May be called at any time after the structure is initialized. - */ -#endif /* SIMPLIFIED_READ */ - -#ifdef PNG_SIMPLIFIED_WRITE_SUPPORTED -/* WRITE APIS - * ---------- - * For write you must initialize a png_image structure to describe the image to - * be written. To do this use memset to set the whole structure to 0 then - * initialize fields describing your image. - * - * version: must be set to PNG_IMAGE_VERSION - * opaque: must be initialized to NULL - * width: image width in pixels - * height: image height in rows - * format: the format of the data (image and color-map) you wish to write - * flags: set to 0 unless one of the defined flags applies; set - * PNG_IMAGE_FLAG_COLORSPACE_NOT_sRGB for color format images where the RGB - * values do not correspond to the colors in sRGB. - * colormap_entries: set to the number of entries in the color-map (0 to 256) - */ -#ifdef PNG_SIMPLIFIED_WRITE_STDIO_SUPPORTED -PNG_EXPORT(239, int, png_image_write_to_file, (png_imagep image, - const char *file, int convert_to_8bit, const void *buffer, - png_int_32 row_stride, const void *colormap)); - /* Write the image to the named file. */ - -PNG_EXPORT(240, int, png_image_write_to_stdio, (png_imagep image, FILE *file, - int convert_to_8_bit, const void *buffer, png_int_32 row_stride, - const void *colormap)); - /* Write the image to the given (FILE*). */ -#endif /* SIMPLIFIED_WRITE_STDIO */ - -/* With all write APIs if image is in one of the linear formats with 16-bit - * data then setting convert_to_8_bit will cause the output to be an 8-bit PNG - * gamma encoded according to the sRGB specification, otherwise a 16-bit linear - * encoded PNG file is written. - * - * With color-mapped data formats the colormap parameter point to a color-map - * with at least image->colormap_entries encoded in the specified format. If - * the format is linear the written PNG color-map will be converted to sRGB - * regardless of the convert_to_8_bit flag. - * - * With all APIs row_stride is handled as in the read APIs - it is the spacing - * from one row to the next in component sized units (1 or 2 bytes) and if - * negative indicates a bottom-up row layout in the buffer. If row_stride is - * zero, libpng will calculate it for you from the image width and number of - * channels. - * - * Note that the write API does not support interlacing, sub-8-bit pixels or - * most ancillary chunks. If you need to write text chunks (e.g. for copyright - * notices) you need to use one of the other APIs. - */ - -PNG_EXPORT(245, int, png_image_write_to_memory, (png_imagep image, void *memory, - png_alloc_size_t * PNG_RESTRICT memory_bytes, int convert_to_8_bit, - const void *buffer, png_int_32 row_stride, const void *colormap)); - /* Write the image to the given memory buffer. The function both writes the - * whole PNG data stream to *memory and updates *memory_bytes with the count - * of bytes written. - * - * 'memory' may be NULL. In this case *memory_bytes is not read however on - * success the number of bytes which would have been written will still be - * stored in *memory_bytes. On failure *memory_bytes will contain 0. - * - * If 'memory' is not NULL it must point to memory[*memory_bytes] of - * writeable memory. - * - * If the function returns success memory[*memory_bytes] (if 'memory' is not - * NULL) contains the written PNG data. *memory_bytes will always be less - * than or equal to the original value. - * - * If the function returns false and *memory_bytes was not changed an error - * occurred during write. If *memory_bytes was changed, or is not 0 if - * 'memory' was NULL, the write would have succeeded but for the memory - * buffer being too small. *memory_bytes contains the required number of - * bytes and will be bigger that the original value. - */ - -#define png_image_write_get_memory_size(image, size, convert_to_8_bit, buffer,\ - row_stride, colormap)\ - png_image_write_to_memory(&(image), 0, &(size), convert_to_8_bit, buffer,\ - row_stride, colormap) - /* Return the amount of memory in 'size' required to compress this image. - * The png_image structure 'image' must be filled in as in the above - * function and must not be changed before the actual write call, the buffer - * and all other parameters must also be identical to that in the final - * write call. The 'size' variable need not be initialized. - * - * NOTE: the macro returns true/false, if false is returned 'size' will be - * set to zero and the write failed and probably will fail if tried again. - */ - -/* You can pre-allocate the buffer by making sure it is of sufficient size - * regardless of the amount of compression achieved. The buffer size will - * always be bigger than the original image and it will never be filled. The - * following macros are provided to assist in allocating the buffer. - */ -#define PNG_IMAGE_DATA_SIZE(image) (PNG_IMAGE_SIZE(image)+(image).height) - /* The number of uncompressed bytes in the PNG byte encoding of the image; - * uncompressing the PNG IDAT data will give this number of bytes. - * - * NOTE: while PNG_IMAGE_SIZE cannot overflow for an image in memory this - * macro can because of the extra bytes used in the PNG byte encoding. You - * need to avoid this macro if your image size approaches 2^30 in width or - * height. The same goes for the remainder of these macros; they all produce - * bigger numbers than the actual in-memory image size. - */ -#ifndef PNG_ZLIB_MAX_SIZE -# define PNG_ZLIB_MAX_SIZE(b) ((b)+(((b)+7U)>>3)+(((b)+63U)>>6)+11U) - /* An upper bound on the number of compressed bytes given 'b' uncompressed - * bytes. This is based on deflateBounds() in zlib; different - * implementations of zlib compression may conceivably produce more data so - * if your zlib implementation is not zlib itself redefine this macro - * appropriately. - */ -#endif - -#define PNG_IMAGE_COMPRESSED_SIZE_MAX(image)\ - PNG_ZLIB_MAX_SIZE((png_alloc_size_t)PNG_IMAGE_DATA_SIZE(image)) - /* An upper bound on the size of the data in the PNG IDAT chunks. */ - -#define PNG_IMAGE_PNG_SIZE_MAX_(image, image_size)\ - ((8U/*sig*/+25U/*IHDR*/+16U/*gAMA*/+44U/*cHRM*/+12U/*IEND*/+\ - (((image).format&PNG_FORMAT_FLAG_COLORMAP)?/*colormap: PLTE, tRNS*/\ - 12U+3U*(image).colormap_entries/*PLTE data*/+\ - (((image).format&PNG_FORMAT_FLAG_ALPHA)?\ - 12U/*tRNS*/+(image).colormap_entries:0U):0U)+\ - 12U)+(12U*((image_size)/PNG_ZBUF_SIZE))/*IDAT*/+(image_size)) - /* A helper for the following macro; if your compiler cannot handle the - * following macro use this one with the result of - * PNG_IMAGE_COMPRESSED_SIZE_MAX(image) as the second argument (most - * compilers should handle this just fine.) - */ - -#define PNG_IMAGE_PNG_SIZE_MAX(image)\ - PNG_IMAGE_PNG_SIZE_MAX_(image, PNG_IMAGE_COMPRESSED_SIZE_MAX(image)) - /* An upper bound on the total length of the PNG data stream for 'image'. - * The result is of type png_alloc_size_t, on 32-bit systems this may - * overflow even though PNG_IMAGE_DATA_SIZE does not overflow; the write will - * run out of buffer space but return a corrected size which should work. - */ -#endif /* SIMPLIFIED_WRITE */ -/******************************************************************************* - * END OF SIMPLIFIED API - ******************************************************************************/ -#endif /* SIMPLIFIED_{READ|WRITE} */ - -/******************************************************************************* - * Section 6: IMPLEMENTATION OPTIONS - ******************************************************************************* - * - * Support for arbitrary implementation-specific optimizations. The API allows - * particular options to be turned on or off. 'Option' is the number of the - * option and 'onoff' is 0 (off) or non-0 (on). The value returned is given - * by the PNG_OPTION_ defines below. - * - * HARDWARE: normally hardware capabilities, such as the Intel SSE instructions, - * are detected at run time, however sometimes it may be impossible - * to do this in user mode, in which case it is necessary to discover - * the capabilities in an OS specific way. Such capabilities are - * listed here when libpng has support for them and must be turned - * ON by the application if present. - * - * SOFTWARE: sometimes software optimizations actually result in performance - * decrease on some architectures or systems, or with some sets of - * PNG images. 'Software' options allow such optimizations to be - * selected at run time. - */ -#ifdef PNG_SET_OPTION_SUPPORTED -#ifdef PNG_ARM_NEON_API_SUPPORTED -# define PNG_ARM_NEON 0 /* HARDWARE: ARM Neon SIMD instructions supported */ -#endif -#define PNG_MAXIMUM_INFLATE_WINDOW 2 /* SOFTWARE: force maximum window */ -#define PNG_SKIP_sRGB_CHECK_PROFILE 4 /* SOFTWARE: Check ICC profile for sRGB */ -#ifdef PNG_MIPS_MSA_API_SUPPORTED -# define PNG_MIPS_MSA 6 /* HARDWARE: MIPS Msa SIMD instructions supported */ -#endif -#ifdef PNG_DISABLE_ADLER32_CHECK_SUPPORTED -# define PNG_IGNORE_ADLER32 8 /* SOFTWARE: disable Adler32 check on IDAT */ -#endif -#ifdef PNG_POWERPC_VSX_API_SUPPORTED -# define PNG_POWERPC_VSX 10 /* HARDWARE: PowerPC VSX SIMD instructions - * supported */ -#endif -#ifdef PNG_MIPS_MMI_API_SUPPORTED -# define PNG_MIPS_MMI 12 /* HARDWARE: MIPS MMI SIMD instructions supported */ -#endif - -#define PNG_OPTION_NEXT 14 /* Next option - numbers must be even */ - -/* Return values: NOTE: there are four values and 'off' is *not* zero */ -#define PNG_OPTION_UNSET 0 /* Unset - defaults to off */ -#define PNG_OPTION_INVALID 1 /* Option number out of range */ -#define PNG_OPTION_OFF 2 -#define PNG_OPTION_ON 3 - -PNG_EXPORT(244, int, png_set_option, (png_structrp png_ptr, int option, - int onoff)); -#endif /* SET_OPTION */ - -/******************************************************************************* - * END OF HARDWARE AND SOFTWARE OPTIONS - ******************************************************************************/ - -/* Maintainer: Put new public prototypes here ^, in libpng.3, in project - * defs, and in scripts/symbols.def. - */ - -/* The last ordinal number (this is the *last* one already used; the next - * one to use is one more than this.) - */ -#ifdef PNG_EXPORT_LAST_ORDINAL - PNG_EXPORT_LAST_ORDINAL(249); -#endif - -#ifdef __cplusplus -} -#endif - -#endif /* PNG_VERSION_INFO_ONLY */ -/* Do not put anything past this line */ -#endif /* PNG_H */ diff --git a/dep/libpng/include/pngconf.h b/dep/libpng/include/pngconf.h deleted file mode 100644 index 000d7b1a8..000000000 --- a/dep/libpng/include/pngconf.h +++ /dev/null @@ -1,623 +0,0 @@ - -/* pngconf.h - machine-configurable file for libpng - * - * libpng version 1.6.43 - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2016,2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - * - * Any machine specific code is near the front of this file, so if you - * are configuring libpng for a machine, you may want to read the section - * starting here down to where it starts to typedef png_color, png_text, - * and png_info. - */ - -#ifndef PNGCONF_H -#define PNGCONF_H - -#ifndef PNG_BUILDING_SYMBOL_TABLE /* else includes may cause problems */ - -/* From libpng 1.6.0 libpng requires an ANSI X3.159-1989 ("ISOC90") compliant C - * compiler for correct compilation. The following header files are required by - * the standard. If your compiler doesn't provide these header files, or they - * do not match the standard, you will need to provide/improve them. - */ -#include -#include - -/* Library header files. These header files are all defined by ISOC90; libpng - * expects conformant implementations, however, an ISOC90 conformant system need - * not provide these header files if the functionality cannot be implemented. - * In this case it will be necessary to disable the relevant parts of libpng in - * the build of pnglibconf.h. - * - * Prior to 1.6.0 string.h was included here; the API changes in 1.6.0 to not - * include this unnecessary header file. - */ - -#ifdef PNG_STDIO_SUPPORTED - /* Required for the definition of FILE: */ -# include -#endif - -#ifdef PNG_SETJMP_SUPPORTED - /* Required for the definition of jmp_buf and the declaration of longjmp: */ -# include -#endif - -#ifdef PNG_CONVERT_tIME_SUPPORTED - /* Required for struct tm: */ -# include -#endif - -#endif /* PNG_BUILDING_SYMBOL_TABLE */ - -/* Prior to 1.6.0, it was possible to turn off 'const' in declarations, - * using PNG_NO_CONST. This is no longer supported. - */ -#define PNG_CONST const /* backward compatibility only */ - -/* This controls optimization of the reading of 16-bit and 32-bit - * values from PNG files. It can be set on a per-app-file basis: it - * just changes whether a macro is used when the function is called. - * The library builder sets the default; if read functions are not - * built into the library the macro implementation is forced on. - */ -#ifndef PNG_READ_INT_FUNCTIONS_SUPPORTED -# define PNG_USE_READ_MACROS -#endif -#if !defined(PNG_NO_USE_READ_MACROS) && !defined(PNG_USE_READ_MACROS) -# if PNG_DEFAULT_READ_MACROS -# define PNG_USE_READ_MACROS -# endif -#endif - -/* COMPILER SPECIFIC OPTIONS. - * - * These options are provided so that a variety of difficult compilers - * can be used. Some are fixed at build time (e.g. PNG_API_RULE - * below) but still have compiler specific implementations, others - * may be changed on a per-file basis when compiling against libpng. - */ - -/* The PNGARG macro was used in versions of libpng prior to 1.6.0 to protect - * against legacy (pre ISOC90) compilers that did not understand function - * prototypes. It is not required for modern C compilers. - */ -#ifndef PNGARG -# define PNGARG(arglist) arglist -#endif - -/* Function calling conventions. - * ============================= - * Normally it is not necessary to specify to the compiler how to call - * a function - it just does it - however on x86 systems derived from - * Microsoft and Borland C compilers ('IBM PC', 'DOS', 'Windows' systems - * and some others) there are multiple ways to call a function and the - * default can be changed on the compiler command line. For this reason - * libpng specifies the calling convention of every exported function and - * every function called via a user supplied function pointer. This is - * done in this file by defining the following macros: - * - * PNGAPI Calling convention for exported functions. - * PNGCBAPI Calling convention for user provided (callback) functions. - * PNGCAPI Calling convention used by the ANSI-C library (required - * for longjmp callbacks and sometimes used internally to - * specify the calling convention for zlib). - * - * These macros should never be overridden. If it is necessary to - * change calling convention in a private build this can be done - * by setting PNG_API_RULE (which defaults to 0) to one of the values - * below to select the correct 'API' variants. - * - * PNG_API_RULE=0 Use PNGCAPI - the 'C' calling convention - throughout. - * This is correct in every known environment. - * PNG_API_RULE=1 Use the operating system convention for PNGAPI and - * the 'C' calling convention (from PNGCAPI) for - * callbacks (PNGCBAPI). This is no longer required - * in any known environment - if it has to be used - * please post an explanation of the problem to the - * libpng mailing list. - * - * These cases only differ if the operating system does not use the C - * calling convention, at present this just means the above cases - * (x86 DOS/Windows systems) and, even then, this does not apply to - * Cygwin running on those systems. - * - * Note that the value must be defined in pnglibconf.h so that what - * the application uses to call the library matches the conventions - * set when building the library. - */ - -/* Symbol export - * ============= - * When building a shared library it is almost always necessary to tell - * the compiler which symbols to export. The png.h macro 'PNG_EXPORT' - * is used to mark the symbols. On some systems these symbols can be - * extracted at link time and need no special processing by the compiler, - * on other systems the symbols are flagged by the compiler and just - * the declaration requires a special tag applied (unfortunately) in a - * compiler dependent way. Some systems can do either. - * - * A small number of older systems also require a symbol from a DLL to - * be flagged to the program that calls it. This is a problem because - * we do not know in the header file included by application code that - * the symbol will come from a shared library, as opposed to a statically - * linked one. For this reason the application must tell us by setting - * the magic flag PNG_USE_DLL to turn on the special processing before - * it includes png.h. - * - * Four additional macros are used to make this happen: - * - * PNG_IMPEXP The magic (if any) to cause a symbol to be exported from - * the build or imported if PNG_USE_DLL is set - compiler - * and system specific. - * - * PNG_EXPORT_TYPE(type) A macro that pre or appends PNG_IMPEXP to - * 'type', compiler specific. - * - * PNG_DLL_EXPORT Set to the magic to use during a libpng build to - * make a symbol exported from the DLL. Not used in the - * public header files; see pngpriv.h for how it is used - * in the libpng build. - * - * PNG_DLL_IMPORT Set to the magic to force the libpng symbols to come - * from a DLL - used to define PNG_IMPEXP when - * PNG_USE_DLL is set. - */ - -/* System specific discovery. - * ========================== - * This code is used at build time to find PNG_IMPEXP, the API settings - * and PNG_EXPORT_TYPE(), it may also set a macro to indicate the DLL - * import processing is possible. On Windows systems it also sets - * compiler-specific macros to the values required to change the calling - * conventions of the various functions. - */ -#if defined(_WIN32) || defined(__WIN32__) || defined(__NT__) || \ - defined(__CYGWIN__) - /* Windows system (DOS doesn't support DLLs). Includes builds under Cygwin or - * MinGW on any architecture currently supported by Windows. Also includes - * Watcom builds but these need special treatment because they are not - * compatible with GCC or Visual C because of different calling conventions. - */ -# if PNG_API_RULE == 2 - /* If this line results in an error, either because __watcall is not - * understood or because of a redefine just below you cannot use *this* - * build of the library with the compiler you are using. *This* build was - * build using Watcom and applications must also be built using Watcom! - */ -# define PNGCAPI __watcall -# endif - -# if defined(__GNUC__) || (defined(_MSC_VER) && (_MSC_VER >= 800)) -# define PNGCAPI __cdecl -# if PNG_API_RULE == 1 - /* If this line results in an error __stdcall is not understood and - * PNG_API_RULE should not have been set to '1'. - */ -# define PNGAPI __stdcall -# endif -# else - /* An older compiler, or one not detected (erroneously) above, - * if necessary override on the command line to get the correct - * variants for the compiler. - */ -# ifndef PNGCAPI -# define PNGCAPI _cdecl -# endif -# if PNG_API_RULE == 1 && !defined(PNGAPI) -# define PNGAPI _stdcall -# endif -# endif /* compiler/api */ - - /* NOTE: PNGCBAPI always defaults to PNGCAPI. */ - -# if defined(PNGAPI) && !defined(PNG_USER_PRIVATEBUILD) -# error "PNG_USER_PRIVATEBUILD must be defined if PNGAPI is changed" -# endif - -# if (defined(_MSC_VER) && _MSC_VER < 800) ||\ - (defined(__BORLANDC__) && __BORLANDC__ < 0x500) - /* older Borland and MSC - * compilers used '__export' and required this to be after - * the type. - */ -# ifndef PNG_EXPORT_TYPE -# define PNG_EXPORT_TYPE(type) type PNG_IMPEXP -# endif -# define PNG_DLL_EXPORT __export -# else /* newer compiler */ -# define PNG_DLL_EXPORT __declspec(dllexport) -# ifndef PNG_DLL_IMPORT -# define PNG_DLL_IMPORT __declspec(dllimport) -# endif -# endif /* compiler */ - -#else /* !Windows */ -# if (defined(__IBMC__) || defined(__IBMCPP__)) && defined(__OS2__) -# define PNGAPI _System -# else /* !Windows/x86 && !OS/2 */ - /* Use the defaults, or define PNG*API on the command line (but - * this will have to be done for every compile!) - */ -# endif /* other system, !OS/2 */ -#endif /* !Windows/x86 */ - -/* Now do all the defaulting . */ -#ifndef PNGCAPI -# define PNGCAPI -#endif -#ifndef PNGCBAPI -# define PNGCBAPI PNGCAPI -#endif -#ifndef PNGAPI -# define PNGAPI PNGCAPI -#endif - -/* PNG_IMPEXP may be set on the compilation system command line or (if not set) - * then in an internal header file when building the library, otherwise (when - * using the library) it is set here. - */ -#ifndef PNG_IMPEXP -# if defined(PNG_USE_DLL) && defined(PNG_DLL_IMPORT) - /* This forces use of a DLL, disallowing static linking */ -# define PNG_IMPEXP PNG_DLL_IMPORT -# endif - -# ifndef PNG_IMPEXP -# define PNG_IMPEXP -# endif -#endif - -/* In 1.5.2 the definition of PNG_FUNCTION has been changed to always treat - * 'attributes' as a storage class - the attributes go at the start of the - * function definition, and attributes are always appended regardless of the - * compiler. This considerably simplifies these macros but may cause problems - * if any compilers both need function attributes and fail to handle them as - * a storage class (this is unlikely.) - */ -#ifndef PNG_FUNCTION -# define PNG_FUNCTION(type, name, args, attributes) attributes type name args -#endif - -#ifndef PNG_EXPORT_TYPE -# define PNG_EXPORT_TYPE(type) PNG_IMPEXP type -#endif - - /* The ordinal value is only relevant when preprocessing png.h for symbol - * table entries, so we discard it here. See the .dfn files in the - * scripts directory. - */ - -#ifndef PNG_EXPORTA -# define PNG_EXPORTA(ordinal, type, name, args, attributes) \ - PNG_FUNCTION(PNG_EXPORT_TYPE(type), (PNGAPI name), PNGARG(args), \ - PNG_LINKAGE_API attributes) -#endif - -/* ANSI-C (C90) does not permit a macro to be invoked with an empty argument, - * so make something non-empty to satisfy the requirement: - */ -#define PNG_EMPTY /*empty list*/ - -#define PNG_EXPORT(ordinal, type, name, args) \ - PNG_EXPORTA(ordinal, type, name, args, PNG_EMPTY) - -/* Use PNG_REMOVED to comment out a removed interface. */ -#ifndef PNG_REMOVED -# define PNG_REMOVED(ordinal, type, name, args, attributes) -#endif - -#ifndef PNG_CALLBACK -# define PNG_CALLBACK(type, name, args) type (PNGCBAPI name) PNGARG(args) -#endif - -/* Support for compiler specific function attributes. These are used - * so that where compiler support is available incorrect use of API - * functions in png.h will generate compiler warnings. - * - * Added at libpng-1.2.41. - */ - -#ifndef PNG_NO_PEDANTIC_WARNINGS -# ifndef PNG_PEDANTIC_WARNINGS_SUPPORTED -# define PNG_PEDANTIC_WARNINGS_SUPPORTED -# endif -#endif - -#ifdef PNG_PEDANTIC_WARNINGS_SUPPORTED - /* Support for compiler specific function attributes. These are used - * so that where compiler support is available, incorrect use of API - * functions in png.h will generate compiler warnings. Added at libpng - * version 1.2.41. Disabling these removes the warnings but may also produce - * less efficient code. - */ -# if defined(__clang__) && defined(__has_attribute) - /* Clang defines both __clang__ and __GNUC__. Check __clang__ first. */ -# if !defined(PNG_USE_RESULT) && __has_attribute(__warn_unused_result__) -# define PNG_USE_RESULT __attribute__((__warn_unused_result__)) -# endif -# if !defined(PNG_NORETURN) && __has_attribute(__noreturn__) -# define PNG_NORETURN __attribute__((__noreturn__)) -# endif -# if !defined(PNG_ALLOCATED) && __has_attribute(__malloc__) -# define PNG_ALLOCATED __attribute__((__malloc__)) -# endif -# if !defined(PNG_DEPRECATED) && __has_attribute(__deprecated__) -# define PNG_DEPRECATED __attribute__((__deprecated__)) -# endif -# if !defined(PNG_PRIVATE) -# ifdef __has_extension -# if __has_extension(attribute_unavailable_with_message) -# define PNG_PRIVATE __attribute__((__unavailable__(\ - "This function is not exported by libpng."))) -# endif -# endif -# endif -# ifndef PNG_RESTRICT -# define PNG_RESTRICT __restrict -# endif - -# elif defined(__GNUC__) -# ifndef PNG_USE_RESULT -# define PNG_USE_RESULT __attribute__((__warn_unused_result__)) -# endif -# ifndef PNG_NORETURN -# define PNG_NORETURN __attribute__((__noreturn__)) -# endif -# if __GNUC__ >= 3 -# ifndef PNG_ALLOCATED -# define PNG_ALLOCATED __attribute__((__malloc__)) -# endif -# ifndef PNG_DEPRECATED -# define PNG_DEPRECATED __attribute__((__deprecated__)) -# endif -# ifndef PNG_PRIVATE -# if 0 /* Doesn't work so we use deprecated instead*/ -# define PNG_PRIVATE \ - __attribute__((warning("This function is not exported by libpng."))) -# else -# define PNG_PRIVATE \ - __attribute__((__deprecated__)) -# endif -# endif -# if ((__GNUC__ > 3) || !defined(__GNUC_MINOR__) || (__GNUC_MINOR__ >= 1)) -# ifndef PNG_RESTRICT -# define PNG_RESTRICT __restrict -# endif -# endif /* __GNUC__.__GNUC_MINOR__ > 3.0 */ -# endif /* __GNUC__ >= 3 */ - -# elif defined(_MSC_VER) && (_MSC_VER >= 1300) -# ifndef PNG_USE_RESULT -# define PNG_USE_RESULT /* not supported */ -# endif -# ifndef PNG_NORETURN -# define PNG_NORETURN __declspec(noreturn) -# endif -# ifndef PNG_ALLOCATED -# if (_MSC_VER >= 1400) -# define PNG_ALLOCATED __declspec(restrict) -# endif -# endif -# ifndef PNG_DEPRECATED -# define PNG_DEPRECATED __declspec(deprecated) -# endif -# ifndef PNG_PRIVATE -# define PNG_PRIVATE __declspec(deprecated) -# endif -# ifndef PNG_RESTRICT -# if (_MSC_VER >= 1400) -# define PNG_RESTRICT __restrict -# endif -# endif - -# elif defined(__WATCOMC__) -# ifndef PNG_RESTRICT -# define PNG_RESTRICT __restrict -# endif -# endif -#endif /* PNG_PEDANTIC_WARNINGS */ - -#ifndef PNG_DEPRECATED -# define PNG_DEPRECATED /* Use of this function is deprecated */ -#endif -#ifndef PNG_USE_RESULT -# define PNG_USE_RESULT /* The result of this function must be checked */ -#endif -#ifndef PNG_NORETURN -# define PNG_NORETURN /* This function does not return */ -#endif -#ifndef PNG_ALLOCATED -# define PNG_ALLOCATED /* The result of the function is new memory */ -#endif -#ifndef PNG_PRIVATE -# define PNG_PRIVATE /* This is a private libpng function */ -#endif -#ifndef PNG_RESTRICT -# define PNG_RESTRICT /* The C99 "restrict" feature */ -#endif - -#ifndef PNG_FP_EXPORT /* A floating point API. */ -# ifdef PNG_FLOATING_POINT_SUPPORTED -# define PNG_FP_EXPORT(ordinal, type, name, args)\ - PNG_EXPORT(ordinal, type, name, args); -# else /* No floating point APIs */ -# define PNG_FP_EXPORT(ordinal, type, name, args) -# endif -#endif -#ifndef PNG_FIXED_EXPORT /* A fixed point API. */ -# ifdef PNG_FIXED_POINT_SUPPORTED -# define PNG_FIXED_EXPORT(ordinal, type, name, args)\ - PNG_EXPORT(ordinal, type, name, args); -# else /* No fixed point APIs */ -# define PNG_FIXED_EXPORT(ordinal, type, name, args) -# endif -#endif - -#ifndef PNG_BUILDING_SYMBOL_TABLE -/* Some typedefs to get us started. These should be safe on most of the common - * platforms. - * - * png_uint_32 and png_int_32 may, currently, be larger than required to hold a - * 32-bit value however this is not normally advisable. - * - * png_uint_16 and png_int_16 should always be two bytes in size - this is - * verified at library build time. - * - * png_byte must always be one byte in size. - * - * The checks below use constants from limits.h, as defined by the ISOC90 - * standard. - */ -#if CHAR_BIT == 8 && UCHAR_MAX == 255 - typedef unsigned char png_byte; -#else -# error "libpng requires 8-bit bytes" -#endif - -#if INT_MIN == -32768 && INT_MAX == 32767 - typedef int png_int_16; -#elif SHRT_MIN == -32768 && SHRT_MAX == 32767 - typedef short png_int_16; -#else -# error "libpng requires a signed 16-bit type" -#endif - -#if UINT_MAX == 65535 - typedef unsigned int png_uint_16; -#elif USHRT_MAX == 65535 - typedef unsigned short png_uint_16; -#else -# error "libpng requires an unsigned 16-bit type" -#endif - -#if INT_MIN < -2147483646 && INT_MAX > 2147483646 - typedef int png_int_32; -#elif LONG_MIN < -2147483646 && LONG_MAX > 2147483646 - typedef long int png_int_32; -#else -# error "libpng requires a signed 32-bit (or more) type" -#endif - -#if UINT_MAX > 4294967294U - typedef unsigned int png_uint_32; -#elif ULONG_MAX > 4294967294U - typedef unsigned long int png_uint_32; -#else -# error "libpng requires an unsigned 32-bit (or more) type" -#endif - -/* Prior to 1.6.0, it was possible to disable the use of size_t and ptrdiff_t. - * From 1.6.0 onwards, an ISO C90 compiler, as well as a standard-compliant - * behavior of sizeof and ptrdiff_t are required. - * The legacy typedefs are provided here for backwards compatibility. - */ -typedef size_t png_size_t; -typedef ptrdiff_t png_ptrdiff_t; - -/* libpng needs to know the maximum value of 'size_t' and this controls the - * definition of png_alloc_size_t, below. This maximum value of size_t limits - * but does not control the maximum allocations the library makes - there is - * direct application control of this through png_set_user_limits(). - */ -#ifndef PNG_SMALL_SIZE_T - /* Compiler specific tests for systems where size_t is known to be less than - * 32 bits (some of these systems may no longer work because of the lack of - * 'far' support; see above.) - */ -# if (defined(__TURBOC__) && !defined(__FLAT__)) ||\ - (defined(_MSC_VER) && defined(MAXSEG_64K)) -# define PNG_SMALL_SIZE_T -# endif -#endif - -/* png_alloc_size_t is guaranteed to be no smaller than size_t, and no smaller - * than png_uint_32. Casts from size_t or png_uint_32 to png_alloc_size_t are - * not necessary; in fact, it is recommended not to use them at all, so that - * the compiler can complain when something turns out to be problematic. - * - * Casts in the other direction (from png_alloc_size_t to size_t or - * png_uint_32) should be explicitly applied; however, we do not expect to - * encounter practical situations that require such conversions. - * - * PNG_SMALL_SIZE_T must be defined if the maximum value of size_t is less than - * 4294967295 - i.e. less than the maximum value of png_uint_32. - */ -#ifdef PNG_SMALL_SIZE_T - typedef png_uint_32 png_alloc_size_t; -#else - typedef size_t png_alloc_size_t; -#endif - -/* Prior to 1.6.0 libpng offered limited support for Microsoft C compiler - * implementations of Intel CPU specific support of user-mode segmented address - * spaces, where 16-bit pointers address more than 65536 bytes of memory using - * separate 'segment' registers. The implementation requires two different - * types of pointer (only one of which includes the segment value.) - * - * If required this support is available in version 1.2 of libpng and may be - * available in versions through 1.5, although the correctness of the code has - * not been verified recently. - */ - -/* Typedef for floating-point numbers that are converted to fixed-point with a - * multiple of 100,000, e.g., gamma - */ -typedef png_int_32 png_fixed_point; - -/* Add typedefs for pointers */ -typedef void * png_voidp; -typedef const void * png_const_voidp; -typedef png_byte * png_bytep; -typedef const png_byte * png_const_bytep; -typedef png_uint_32 * png_uint_32p; -typedef const png_uint_32 * png_const_uint_32p; -typedef png_int_32 * png_int_32p; -typedef const png_int_32 * png_const_int_32p; -typedef png_uint_16 * png_uint_16p; -typedef const png_uint_16 * png_const_uint_16p; -typedef png_int_16 * png_int_16p; -typedef const png_int_16 * png_const_int_16p; -typedef char * png_charp; -typedef const char * png_const_charp; -typedef png_fixed_point * png_fixed_point_p; -typedef const png_fixed_point * png_const_fixed_point_p; -typedef size_t * png_size_tp; -typedef const size_t * png_const_size_tp; - -#ifdef PNG_STDIO_SUPPORTED -typedef FILE * png_FILE_p; -#endif - -#ifdef PNG_FLOATING_POINT_SUPPORTED -typedef double * png_doublep; -typedef const double * png_const_doublep; -#endif - -/* Pointers to pointers; i.e. arrays */ -typedef png_byte * * png_bytepp; -typedef png_uint_32 * * png_uint_32pp; -typedef png_int_32 * * png_int_32pp; -typedef png_uint_16 * * png_uint_16pp; -typedef png_int_16 * * png_int_16pp; -typedef const char * * png_const_charpp; -typedef char * * png_charpp; -typedef png_fixed_point * * png_fixed_point_pp; -#ifdef PNG_FLOATING_POINT_SUPPORTED -typedef double * * png_doublepp; -#endif - -/* Pointers to pointers to pointers; i.e., pointer to array */ -typedef char * * * png_charppp; - -#endif /* PNG_BUILDING_SYMBOL_TABLE */ - -#endif /* PNGCONF_H */ diff --git a/dep/libpng/include/pnglibconf.h b/dep/libpng/include/pnglibconf.h deleted file mode 100644 index 83f09fbe7..000000000 --- a/dep/libpng/include/pnglibconf.h +++ /dev/null @@ -1,224 +0,0 @@ -/* pnglibconf.h - library build configuration */ - -/* libpng version 1.6.43 */ - -/* Copyright (c) 2018-2024 Cosmin Truta */ -/* Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson */ - -/* This code is released under the libpng license. */ -/* For conditions of distribution and use, see the disclaimer */ -/* and license in png.h */ - -/* pnglibconf.h */ -/* Machine generated file: DO NOT EDIT */ -/* Derived from: scripts/pnglibconf.dfa */ -#ifndef PNGLCONF_H -#define PNGLCONF_H -/* options */ -#define PNG_16BIT_SUPPORTED -#define PNG_ALIGNED_MEMORY_SUPPORTED -/*#undef PNG_ARM_NEON_API_SUPPORTED*/ -/*#undef PNG_ARM_NEON_CHECK_SUPPORTED*/ -#define PNG_BENIGN_ERRORS_SUPPORTED -#define PNG_BENIGN_READ_ERRORS_SUPPORTED -/*#undef PNG_BENIGN_WRITE_ERRORS_SUPPORTED*/ -#define PNG_BUILD_GRAYSCALE_PALETTE_SUPPORTED -#define PNG_CHECK_FOR_INVALID_INDEX_SUPPORTED -#define PNG_COLORSPACE_SUPPORTED -#define PNG_CONSOLE_IO_SUPPORTED -#define PNG_CONVERT_tIME_SUPPORTED -/*#undef PNG_DISABLE_ADLER32_CHECK_SUPPORTED*/ -#define PNG_EASY_ACCESS_SUPPORTED -/*#undef PNG_ERROR_NUMBERS_SUPPORTED*/ -#define PNG_ERROR_TEXT_SUPPORTED -#define PNG_FIXED_POINT_SUPPORTED -#define PNG_FLOATING_ARITHMETIC_SUPPORTED -#define PNG_FLOATING_POINT_SUPPORTED -#define PNG_FORMAT_AFIRST_SUPPORTED -#define PNG_FORMAT_BGR_SUPPORTED -#define PNG_GAMMA_SUPPORTED -#define PNG_GET_PALETTE_MAX_SUPPORTED -#define PNG_HANDLE_AS_UNKNOWN_SUPPORTED -#define PNG_INCH_CONVERSIONS_SUPPORTED -#define PNG_INFO_IMAGE_SUPPORTED -#define PNG_IO_STATE_SUPPORTED -/*#undef PNG_MIPS_MMI_API_SUPPORTED*/ -/*#undef PNG_MIPS_MMI_CHECK_SUPPORTED*/ -/*#undef PNG_MIPS_MSA_API_SUPPORTED*/ -/*#undef PNG_MIPS_MSA_CHECK_SUPPORTED*/ -#define PNG_MNG_FEATURES_SUPPORTED -#define PNG_POINTER_INDEXING_SUPPORTED -/*#undef PNG_POWERPC_VSX_API_SUPPORTED*/ -/*#undef PNG_POWERPC_VSX_CHECK_SUPPORTED*/ -#define PNG_PROGRESSIVE_READ_SUPPORTED -#define PNG_READ_16BIT_SUPPORTED -#define PNG_READ_ALPHA_MODE_SUPPORTED -#define PNG_READ_ANCILLARY_CHUNKS_SUPPORTED -#define PNG_READ_BACKGROUND_SUPPORTED -#define PNG_READ_BGR_SUPPORTED -#define PNG_READ_CHECK_FOR_INVALID_INDEX_SUPPORTED -#define PNG_READ_COMPOSITE_NODIV_SUPPORTED -#define PNG_READ_COMPRESSED_TEXT_SUPPORTED -#define PNG_READ_EXPAND_16_SUPPORTED -#define PNG_READ_EXPAND_SUPPORTED -#define PNG_READ_FILLER_SUPPORTED -#define PNG_READ_GAMMA_SUPPORTED -#define PNG_READ_GET_PALETTE_MAX_SUPPORTED -#define PNG_READ_GRAY_TO_RGB_SUPPORTED -#define PNG_READ_INTERLACING_SUPPORTED -#define PNG_READ_INT_FUNCTIONS_SUPPORTED -#define PNG_READ_INVERT_ALPHA_SUPPORTED -#define PNG_READ_INVERT_SUPPORTED -#define PNG_READ_OPT_PLTE_SUPPORTED -#define PNG_READ_PACKSWAP_SUPPORTED -#define PNG_READ_PACK_SUPPORTED -#define PNG_READ_QUANTIZE_SUPPORTED -#define PNG_READ_RGB_TO_GRAY_SUPPORTED -#define PNG_READ_SCALE_16_TO_8_SUPPORTED -#define PNG_READ_SHIFT_SUPPORTED -#define PNG_READ_STRIP_16_TO_8_SUPPORTED -#define PNG_READ_STRIP_ALPHA_SUPPORTED -#define PNG_READ_SUPPORTED -#define PNG_READ_SWAP_ALPHA_SUPPORTED -#define PNG_READ_SWAP_SUPPORTED -#define PNG_READ_TEXT_SUPPORTED -#define PNG_READ_TRANSFORMS_SUPPORTED -#define PNG_READ_UNKNOWN_CHUNKS_SUPPORTED -#define PNG_READ_USER_CHUNKS_SUPPORTED -#define PNG_READ_USER_TRANSFORM_SUPPORTED -#define PNG_READ_bKGD_SUPPORTED -#define PNG_READ_cHRM_SUPPORTED -#define PNG_READ_eXIf_SUPPORTED -#define PNG_READ_gAMA_SUPPORTED -#define PNG_READ_hIST_SUPPORTED -#define PNG_READ_iCCP_SUPPORTED -#define PNG_READ_iTXt_SUPPORTED -#define PNG_READ_oFFs_SUPPORTED -#define PNG_READ_pCAL_SUPPORTED -#define PNG_READ_pHYs_SUPPORTED -#define PNG_READ_sBIT_SUPPORTED -#define PNG_READ_sCAL_SUPPORTED -#define PNG_READ_sPLT_SUPPORTED -#define PNG_READ_sRGB_SUPPORTED -#define PNG_READ_tEXt_SUPPORTED -#define PNG_READ_tIME_SUPPORTED -#define PNG_READ_tRNS_SUPPORTED -#define PNG_READ_zTXt_SUPPORTED -#define PNG_SAVE_INT_32_SUPPORTED -#define PNG_SAVE_UNKNOWN_CHUNKS_SUPPORTED -#define PNG_SEQUENTIAL_READ_SUPPORTED -#define PNG_SETJMP_SUPPORTED -#define PNG_SET_OPTION_SUPPORTED -#define PNG_SET_UNKNOWN_CHUNKS_SUPPORTED -#define PNG_SET_USER_LIMITS_SUPPORTED -#define PNG_SIMPLIFIED_READ_AFIRST_SUPPORTED -#define PNG_SIMPLIFIED_READ_BGR_SUPPORTED -#define PNG_SIMPLIFIED_READ_SUPPORTED -#define PNG_SIMPLIFIED_WRITE_AFIRST_SUPPORTED -#define PNG_SIMPLIFIED_WRITE_BGR_SUPPORTED -#define PNG_SIMPLIFIED_WRITE_STDIO_SUPPORTED -#define PNG_SIMPLIFIED_WRITE_SUPPORTED -#define PNG_STDIO_SUPPORTED -#define PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED -#define PNG_TEXT_SUPPORTED -#define PNG_TIME_RFC1123_SUPPORTED -#define PNG_UNKNOWN_CHUNKS_SUPPORTED -#define PNG_USER_CHUNKS_SUPPORTED -#define PNG_USER_LIMITS_SUPPORTED -#define PNG_USER_MEM_SUPPORTED -#define PNG_USER_TRANSFORM_INFO_SUPPORTED -#define PNG_USER_TRANSFORM_PTR_SUPPORTED -#define PNG_WARNINGS_SUPPORTED -#define PNG_WRITE_16BIT_SUPPORTED -#define PNG_WRITE_ANCILLARY_CHUNKS_SUPPORTED -#define PNG_WRITE_BGR_SUPPORTED -#define PNG_WRITE_CHECK_FOR_INVALID_INDEX_SUPPORTED -#define PNG_WRITE_COMPRESSED_TEXT_SUPPORTED -#define PNG_WRITE_CUSTOMIZE_COMPRESSION_SUPPORTED -#define PNG_WRITE_CUSTOMIZE_ZTXT_COMPRESSION_SUPPORTED -#define PNG_WRITE_FILLER_SUPPORTED -#define PNG_WRITE_FILTER_SUPPORTED -#define PNG_WRITE_FLUSH_SUPPORTED -#define PNG_WRITE_GET_PALETTE_MAX_SUPPORTED -#define PNG_WRITE_INTERLACING_SUPPORTED -#define PNG_WRITE_INT_FUNCTIONS_SUPPORTED -#define PNG_WRITE_INVERT_ALPHA_SUPPORTED -#define PNG_WRITE_INVERT_SUPPORTED -#define PNG_WRITE_OPTIMIZE_CMF_SUPPORTED -#define PNG_WRITE_PACKSWAP_SUPPORTED -#define PNG_WRITE_PACK_SUPPORTED -#define PNG_WRITE_SHIFT_SUPPORTED -#define PNG_WRITE_SUPPORTED -#define PNG_WRITE_SWAP_ALPHA_SUPPORTED -#define PNG_WRITE_SWAP_SUPPORTED -#define PNG_WRITE_TEXT_SUPPORTED -#define PNG_WRITE_TRANSFORMS_SUPPORTED -#define PNG_WRITE_UNKNOWN_CHUNKS_SUPPORTED -#define PNG_WRITE_USER_TRANSFORM_SUPPORTED -#define PNG_WRITE_WEIGHTED_FILTER_SUPPORTED -#define PNG_WRITE_bKGD_SUPPORTED -#define PNG_WRITE_cHRM_SUPPORTED -#define PNG_WRITE_eXIf_SUPPORTED -#define PNG_WRITE_gAMA_SUPPORTED -#define PNG_WRITE_hIST_SUPPORTED -#define PNG_WRITE_iCCP_SUPPORTED -#define PNG_WRITE_iTXt_SUPPORTED -#define PNG_WRITE_oFFs_SUPPORTED -#define PNG_WRITE_pCAL_SUPPORTED -#define PNG_WRITE_pHYs_SUPPORTED -#define PNG_WRITE_sBIT_SUPPORTED -#define PNG_WRITE_sCAL_SUPPORTED -#define PNG_WRITE_sPLT_SUPPORTED -#define PNG_WRITE_sRGB_SUPPORTED -#define PNG_WRITE_tEXt_SUPPORTED -#define PNG_WRITE_tIME_SUPPORTED -#define PNG_WRITE_tRNS_SUPPORTED -#define PNG_WRITE_zTXt_SUPPORTED -#define PNG_bKGD_SUPPORTED -#define PNG_cHRM_SUPPORTED -#define PNG_eXIf_SUPPORTED -#define PNG_gAMA_SUPPORTED -#define PNG_hIST_SUPPORTED -#define PNG_iCCP_SUPPORTED -#define PNG_iTXt_SUPPORTED -#define PNG_oFFs_SUPPORTED -#define PNG_pCAL_SUPPORTED -#define PNG_pHYs_SUPPORTED -#define PNG_sBIT_SUPPORTED -#define PNG_sCAL_SUPPORTED -#define PNG_sPLT_SUPPORTED -#define PNG_sRGB_SUPPORTED -#define PNG_tEXt_SUPPORTED -#define PNG_tIME_SUPPORTED -#define PNG_tRNS_SUPPORTED -#define PNG_zTXt_SUPPORTED -/* end of options */ -/* settings */ -#define PNG_API_RULE 0 -#define PNG_DEFAULT_READ_MACROS 1 -#define PNG_GAMMA_THRESHOLD_FIXED 5000 -#define PNG_IDAT_READ_SIZE PNG_ZBUF_SIZE -#define PNG_INFLATE_BUF_SIZE 1024 -#define PNG_LINKAGE_API extern -#define PNG_LINKAGE_CALLBACK extern -#define PNG_LINKAGE_DATA extern -#define PNG_LINKAGE_FUNCTION extern -#define PNG_MAX_GAMMA_8 11 -#define PNG_QUANTIZE_BLUE_BITS 5 -#define PNG_QUANTIZE_GREEN_BITS 5 -#define PNG_QUANTIZE_RED_BITS 5 -#define PNG_TEXT_Z_DEFAULT_COMPRESSION (-1) -#define PNG_TEXT_Z_DEFAULT_STRATEGY 0 -#define PNG_USER_CHUNK_CACHE_MAX 1000 -#define PNG_USER_CHUNK_MALLOC_MAX 8000000 -#define PNG_USER_HEIGHT_MAX 1000000 -#define PNG_USER_WIDTH_MAX 1000000 -#define PNG_ZBUF_SIZE 8192 -#define PNG_ZLIB_VERNUM 0 /* unknown */ -#define PNG_Z_DEFAULT_COMPRESSION (-1) -#define PNG_Z_DEFAULT_NOFILTER_STRATEGY 0 -#define PNG_Z_DEFAULT_STRATEGY 1 -#define PNG_sCAL_PRECISION 5 -#define PNG_sRGB_PROFILE_CHECKS 2 -/* end of settings */ -#endif /* PNGLCONF_H */ diff --git a/dep/libpng/libpng.vcxproj b/dep/libpng/libpng.vcxproj deleted file mode 100644 index 219b94786..000000000 --- a/dep/libpng/libpng.vcxproj +++ /dev/null @@ -1,61 +0,0 @@ - - - - - {9FD2ABCD-2DCD-4302-BE5C-DF0BA8431FA5} - - - - - - - - - - - - - true - - - true - - - true - - - true - - - true - - - - - - - - - - - - - - - - - - - - {7ff9fdb9-d504-47db-a16a-b08071999620} - - - - - - TurnOffAllWarnings - $(ProjectDir)include;$(ProjectDir)src;$(ProjectDir)..\zlib\include;%(AdditionalIncludeDirectories) - - - - \ No newline at end of file diff --git a/dep/libpng/libpng.vcxproj.filters b/dep/libpng/libpng.vcxproj.filters deleted file mode 100644 index 9b5b8ef17..000000000 --- a/dep/libpng/libpng.vcxproj.filters +++ /dev/null @@ -1,52 +0,0 @@ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - intel - - - intel - - - arm - - - arm - - - arm - - - - - {9f24e95e-025d-4ed8-8c41-2fb1c7a36026} - - - {8316b9c1-8c00-4bc8-ace7-c9b864890f2d} - - - \ No newline at end of file diff --git a/dep/libpng/src/arm/arm_init.c b/dep/libpng/src/arm/arm_init.c deleted file mode 100644 index 84d05556f..000000000 --- a/dep/libpng/src/arm/arm_init.c +++ /dev/null @@ -1,139 +0,0 @@ - -/* arm_init.c - NEON optimised filter functions - * - * Copyright (c) 2018-2022 Cosmin Truta - * Copyright (c) 2014,2016 Glenn Randers-Pehrson - * Written by Mans Rullgard, 2011. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -/* This module requires POSIX 1003.1 functions. */ -#define _POSIX_SOURCE 1 - -#include "../pngpriv.h" - -#ifdef PNG_READ_SUPPORTED - -#if PNG_ARM_NEON_OPT > 0 -#ifdef PNG_ARM_NEON_CHECK_SUPPORTED /* Do run-time checks */ -/* WARNING: it is strongly recommended that you do not build libpng with - * run-time checks for CPU features if at all possible. In the case of the ARM - * NEON instructions there is no processor-specific way of detecting the - * presence of the required support, therefore run-time detection is extremely - * OS specific. - * - * You may set the macro PNG_ARM_NEON_FILE to the file name of file containing - * a fragment of C source code which defines the png_have_neon function. There - * are a number of implementations in contrib/arm-neon, but the only one that - * has partial support is contrib/arm-neon/linux.c - a generic Linux - * implementation which reads /proc/cpufino. - */ -#include /* for sig_atomic_t */ - -#ifndef PNG_ARM_NEON_FILE -# if defined(__aarch64__) || defined(_M_ARM64) - /* ARM Neon is expected to be unconditionally available on ARM64. */ -# error "PNG_ARM_NEON_CHECK_SUPPORTED must not be defined on ARM64" -# elif defined(__ARM_NEON__) || defined(__ARM_NEON) - /* ARM Neon is expected to be available on the target CPU architecture. */ -# error "PNG_ARM_NEON_CHECK_SUPPORTED must not be defined on this CPU arch" -# elif defined(__linux__) -# define PNG_ARM_NEON_FILE "contrib/arm-neon/linux.c" -# else -# error "No support for run-time ARM Neon checking; use compile-time options" -# endif -#endif - -static int png_have_neon(png_structp png_ptr); -#ifdef PNG_ARM_NEON_FILE -# include PNG_ARM_NEON_FILE -#endif -#endif /* PNG_ARM_NEON_CHECK_SUPPORTED */ - -#ifndef PNG_ALIGNED_MEMORY_SUPPORTED -# error "ALIGNED_MEMORY is required; set: -DPNG_ALIGNED_MEMORY_SUPPORTED" -#endif - -void -png_init_filter_functions_neon(png_structp pp, unsigned int bpp) -{ - /* The switch statement is compiled in for ARM_NEON_API, the call to - * png_have_neon is compiled in for ARM_NEON_CHECK. If both are defined - * the check is only performed if the API has not set the NEON option on - * or off explicitly. In this case the check controls what happens. - * - * If the CHECK is not compiled in and the option is UNSET the behavior prior - * to 1.6.7 was to use the NEON code - this was a bug caused by having the - * wrong order of the 'ON' and 'default' cases. UNSET now defaults to OFF, - * as documented in png.h - */ - png_debug(1, "in png_init_filter_functions_neon"); -#ifdef PNG_ARM_NEON_API_SUPPORTED - switch ((pp->options >> PNG_ARM_NEON) & 3) - { - case PNG_OPTION_UNSET: - /* Allow the run-time check to execute if it has been enabled - - * thus both API and CHECK can be turned on. If it isn't supported - * this case will fall through to the 'default' below, which just - * returns. - */ -#endif /* PNG_ARM_NEON_API_SUPPORTED */ -#ifdef PNG_ARM_NEON_CHECK_SUPPORTED - { - static volatile sig_atomic_t no_neon = -1; /* not checked */ - - if (no_neon < 0) - no_neon = !png_have_neon(pp); - - if (no_neon) - return; - } -#ifdef PNG_ARM_NEON_API_SUPPORTED - break; -#endif -#endif /* PNG_ARM_NEON_CHECK_SUPPORTED */ - -#ifdef PNG_ARM_NEON_API_SUPPORTED - default: /* OFF or INVALID */ - return; - - case PNG_OPTION_ON: - /* Option turned on */ - break; - } -#endif - - /* IMPORTANT: any new external functions used here must be declared using - * PNG_INTERNAL_FUNCTION in ../pngpriv.h. This is required so that the - * 'prefix' option to configure works: - * - * ./configure --with-libpng-prefix=foobar_ - * - * Verify you have got this right by running the above command, doing a build - * and examining pngprefix.h; it must contain a #define for every external - * function you add. (Notice that this happens automatically for the - * initialization function.) - */ - pp->read_filter[PNG_FILTER_VALUE_UP-1] = png_read_filter_row_up_neon; - - if (bpp == 3) - { - pp->read_filter[PNG_FILTER_VALUE_SUB-1] = png_read_filter_row_sub3_neon; - pp->read_filter[PNG_FILTER_VALUE_AVG-1] = png_read_filter_row_avg3_neon; - pp->read_filter[PNG_FILTER_VALUE_PAETH-1] = - png_read_filter_row_paeth3_neon; - } - - else if (bpp == 4) - { - pp->read_filter[PNG_FILTER_VALUE_SUB-1] = png_read_filter_row_sub4_neon; - pp->read_filter[PNG_FILTER_VALUE_AVG-1] = png_read_filter_row_avg4_neon; - pp->read_filter[PNG_FILTER_VALUE_PAETH-1] = - png_read_filter_row_paeth4_neon; - } -} -#endif /* PNG_ARM_NEON_OPT > 0 */ -#endif /* READ */ diff --git a/dep/libpng/src/arm/filter_neon.S b/dep/libpng/src/arm/filter_neon.S deleted file mode 100644 index 2308aad13..000000000 --- a/dep/libpng/src/arm/filter_neon.S +++ /dev/null @@ -1,253 +0,0 @@ - -/* filter_neon.S - NEON optimised filter functions - * - * Copyright (c) 2018 Cosmin Truta - * Copyright (c) 2014,2017 Glenn Randers-Pehrson - * Written by Mans Rullgard, 2011. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -/* This is required to get the symbol renames, which are #defines, and the - * definitions (or not) of PNG_ARM_NEON_OPT and PNG_ARM_NEON_IMPLEMENTATION. - */ -#define PNG_VERSION_INFO_ONLY -#include "../pngpriv.h" - -#if (defined(__linux__) || defined(__FreeBSD__)) && defined(__ELF__) -.section .note.GNU-stack,"",%progbits /* mark stack as non-executable */ -#endif - -#ifdef PNG_READ_SUPPORTED - -/* Assembler NEON support - only works for 32-bit ARM (i.e. it does not work for - * ARM64). The code in arm/filter_neon_intrinsics.c supports ARM64, however it - * only works if -mfpu=neon is specified on the GCC command line. See pngpriv.h - * for the logic which sets PNG_USE_ARM_NEON_ASM: - */ -#if PNG_ARM_NEON_IMPLEMENTATION == 2 /* hand-coded assembler */ - -#if PNG_ARM_NEON_OPT > 0 - -#ifdef __ELF__ -# define ELF -#else -# define ELF @ -#endif - - .arch armv7-a - .fpu neon - -.macro func name, export=0 - .macro endfunc -ELF .size \name, . - \name - .endfunc - .purgem endfunc - .endm - .text - - /* Explicitly specifying alignment here because some versions of - * GAS don't align code correctly. This is harmless in correctly - * written versions of GAS. - */ - .align 2 - - .if \export - .global \name - .endif -ELF .type \name, STT_FUNC - .func \name -\name: -.endm - -func png_read_filter_row_sub4_neon, export=1 - ldr r3, [r0, #4] @ rowbytes - vmov.i8 d3, #0 -1: - vld4.32 {d4[],d5[],d6[],d7[]}, [r1,:128] - vadd.u8 d0, d3, d4 - vadd.u8 d1, d0, d5 - vadd.u8 d2, d1, d6 - vadd.u8 d3, d2, d7 - vst4.32 {d0[0],d1[0],d2[0],d3[0]},[r1,:128]! - subs r3, r3, #16 - bgt 1b - - bx lr -endfunc - -func png_read_filter_row_sub3_neon, export=1 - ldr r3, [r0, #4] @ rowbytes - vmov.i8 d3, #0 - mov r0, r1 - mov r2, #3 - mov r12, #12 - vld1.8 {q11}, [r0], r12 -1: - vext.8 d5, d22, d23, #3 - vadd.u8 d0, d3, d22 - vext.8 d6, d22, d23, #6 - vadd.u8 d1, d0, d5 - vext.8 d7, d23, d23, #1 - vld1.8 {q11}, [r0], r12 - vst1.32 {d0[0]}, [r1,:32], r2 - vadd.u8 d2, d1, d6 - vst1.32 {d1[0]}, [r1], r2 - vadd.u8 d3, d2, d7 - vst1.32 {d2[0]}, [r1], r2 - vst1.32 {d3[0]}, [r1], r2 - subs r3, r3, #12 - bgt 1b - - bx lr -endfunc - -func png_read_filter_row_up_neon, export=1 - ldr r3, [r0, #4] @ rowbytes -1: - vld1.8 {q0}, [r1,:128] - vld1.8 {q1}, [r2,:128]! - vadd.u8 q0, q0, q1 - vst1.8 {q0}, [r1,:128]! - subs r3, r3, #16 - bgt 1b - - bx lr -endfunc - -func png_read_filter_row_avg4_neon, export=1 - ldr r12, [r0, #4] @ rowbytes - vmov.i8 d3, #0 -1: - vld4.32 {d4[],d5[],d6[],d7[]}, [r1,:128] - vld4.32 {d16[],d17[],d18[],d19[]},[r2,:128]! - vhadd.u8 d0, d3, d16 - vadd.u8 d0, d0, d4 - vhadd.u8 d1, d0, d17 - vadd.u8 d1, d1, d5 - vhadd.u8 d2, d1, d18 - vadd.u8 d2, d2, d6 - vhadd.u8 d3, d2, d19 - vadd.u8 d3, d3, d7 - vst4.32 {d0[0],d1[0],d2[0],d3[0]},[r1,:128]! - subs r12, r12, #16 - bgt 1b - - bx lr -endfunc - -func png_read_filter_row_avg3_neon, export=1 - push {r4,lr} - ldr r12, [r0, #4] @ rowbytes - vmov.i8 d3, #0 - mov r0, r1 - mov r4, #3 - mov lr, #12 - vld1.8 {q11}, [r0], lr -1: - vld1.8 {q10}, [r2], lr - vext.8 d5, d22, d23, #3 - vhadd.u8 d0, d3, d20 - vext.8 d17, d20, d21, #3 - vadd.u8 d0, d0, d22 - vext.8 d6, d22, d23, #6 - vhadd.u8 d1, d0, d17 - vext.8 d18, d20, d21, #6 - vadd.u8 d1, d1, d5 - vext.8 d7, d23, d23, #1 - vld1.8 {q11}, [r0], lr - vst1.32 {d0[0]}, [r1,:32], r4 - vhadd.u8 d2, d1, d18 - vst1.32 {d1[0]}, [r1], r4 - vext.8 d19, d21, d21, #1 - vadd.u8 d2, d2, d6 - vhadd.u8 d3, d2, d19 - vst1.32 {d2[0]}, [r1], r4 - vadd.u8 d3, d3, d7 - vst1.32 {d3[0]}, [r1], r4 - subs r12, r12, #12 - bgt 1b - - pop {r4,pc} -endfunc - -.macro paeth rx, ra, rb, rc - vaddl.u8 q12, \ra, \rb @ a + b - vaddl.u8 q15, \rc, \rc @ 2*c - vabdl.u8 q13, \rb, \rc @ pa - vabdl.u8 q14, \ra, \rc @ pb - vabd.u16 q15, q12, q15 @ pc - vcle.u16 q12, q13, q14 @ pa <= pb - vcle.u16 q13, q13, q15 @ pa <= pc - vcle.u16 q14, q14, q15 @ pb <= pc - vand q12, q12, q13 @ pa <= pb && pa <= pc - vmovn.u16 d28, q14 - vmovn.u16 \rx, q12 - vbsl d28, \rb, \rc - vbsl \rx, \ra, d28 -.endm - -func png_read_filter_row_paeth4_neon, export=1 - ldr r12, [r0, #4] @ rowbytes - vmov.i8 d3, #0 - vmov.i8 d20, #0 -1: - vld4.32 {d4[],d5[],d6[],d7[]}, [r1,:128] - vld4.32 {d16[],d17[],d18[],d19[]},[r2,:128]! - paeth d0, d3, d16, d20 - vadd.u8 d0, d0, d4 - paeth d1, d0, d17, d16 - vadd.u8 d1, d1, d5 - paeth d2, d1, d18, d17 - vadd.u8 d2, d2, d6 - paeth d3, d2, d19, d18 - vmov d20, d19 - vadd.u8 d3, d3, d7 - vst4.32 {d0[0],d1[0],d2[0],d3[0]},[r1,:128]! - subs r12, r12, #16 - bgt 1b - - bx lr -endfunc - -func png_read_filter_row_paeth3_neon, export=1 - push {r4,lr} - ldr r12, [r0, #4] @ rowbytes - vmov.i8 d3, #0 - vmov.i8 d4, #0 - mov r0, r1 - mov r4, #3 - mov lr, #12 - vld1.8 {q11}, [r0], lr -1: - vld1.8 {q10}, [r2], lr - paeth d0, d3, d20, d4 - vext.8 d5, d22, d23, #3 - vadd.u8 d0, d0, d22 - vext.8 d17, d20, d21, #3 - paeth d1, d0, d17, d20 - vst1.32 {d0[0]}, [r1,:32], r4 - vext.8 d6, d22, d23, #6 - vadd.u8 d1, d1, d5 - vext.8 d18, d20, d21, #6 - paeth d2, d1, d18, d17 - vext.8 d7, d23, d23, #1 - vld1.8 {q11}, [r0], lr - vst1.32 {d1[0]}, [r1], r4 - vadd.u8 d2, d2, d6 - vext.8 d19, d21, d21, #1 - paeth d3, d2, d19, d18 - vst1.32 {d2[0]}, [r1], r4 - vmov d4, d19 - vadd.u8 d3, d3, d7 - vst1.32 {d3[0]}, [r1], r4 - subs r12, r12, #12 - bgt 1b - - pop {r4,pc} -endfunc -#endif /* PNG_ARM_NEON_OPT > 0 */ -#endif /* PNG_ARM_NEON_IMPLEMENTATION == 2 (assembler) */ -#endif /* READ */ diff --git a/dep/libpng/src/arm/filter_neon_intrinsics.c b/dep/libpng/src/arm/filter_neon_intrinsics.c deleted file mode 100644 index 4466d48b2..000000000 --- a/dep/libpng/src/arm/filter_neon_intrinsics.c +++ /dev/null @@ -1,402 +0,0 @@ - -/* filter_neon_intrinsics.c - NEON optimised filter functions - * - * Copyright (c) 2018 Cosmin Truta - * Copyright (c) 2014,2016 Glenn Randers-Pehrson - * Written by James Yu , October 2013. - * Based on filter_neon.S, written by Mans Rullgard, 2011. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -#include "../pngpriv.h" - -#ifdef PNG_READ_SUPPORTED - -/* This code requires -mfpu=neon on the command line: */ -#if PNG_ARM_NEON_IMPLEMENTATION == 1 /* intrinsics code from pngpriv.h */ - -#if defined(_MSC_VER) && !defined(__clang__) && defined(_M_ARM64) -# include -#else -# include -#endif - -/* libpng row pointers are not necessarily aligned to any particular boundary, - * however this code will only work with appropriate alignment. arm/arm_init.c - * checks for this (and will not compile unless it is done). This code uses - * variants of png_aligncast to avoid compiler warnings. - */ -#define png_ptr(type,pointer) png_aligncast(type *,pointer) -#define png_ptrc(type,pointer) png_aligncastconst(const type *,pointer) - -/* The following relies on a variable 'temp_pointer' being declared with type - * 'type'. This is written this way just to hide the GCC strict aliasing - * warning; note that the code is safe because there never is an alias between - * the input and output pointers. - * - * When compiling with MSVC ARM64, the png_ldr macro can't be passed directly - * to vst4_lane_u32, because of an internal compiler error inside MSVC. - * To avoid this compiler bug, we use a temporary variable (vdest_val) to store - * the result of png_ldr. - */ -#define png_ldr(type,pointer)\ - (temp_pointer = png_ptr(type,pointer), *temp_pointer) - -#if PNG_ARM_NEON_OPT > 0 - -void -png_read_filter_row_up_neon(png_row_infop row_info, png_bytep row, - png_const_bytep prev_row) -{ - png_bytep rp = row; - png_bytep rp_stop = row + row_info->rowbytes; - png_const_bytep pp = prev_row; - - png_debug(1, "in png_read_filter_row_up_neon"); - - for (; rp < rp_stop; rp += 16, pp += 16) - { - uint8x16_t qrp, qpp; - - qrp = vld1q_u8(rp); - qpp = vld1q_u8(pp); - qrp = vaddq_u8(qrp, qpp); - vst1q_u8(rp, qrp); - } -} - -void -png_read_filter_row_sub3_neon(png_row_infop row_info, png_bytep row, - png_const_bytep prev_row) -{ - png_bytep rp = row; - png_bytep rp_stop = row + row_info->rowbytes; - - uint8x16_t vtmp = vld1q_u8(rp); - uint8x8x2_t *vrpt = png_ptr(uint8x8x2_t, &vtmp); - uint8x8x2_t vrp = *vrpt; - - uint8x8x4_t vdest; - vdest.val[3] = vdup_n_u8(0); - - png_debug(1, "in png_read_filter_row_sub3_neon"); - - for (; rp < rp_stop;) - { - uint8x8_t vtmp1, vtmp2; - uint32x2_t *temp_pointer; - - vtmp1 = vext_u8(vrp.val[0], vrp.val[1], 3); - vdest.val[0] = vadd_u8(vdest.val[3], vrp.val[0]); - vtmp2 = vext_u8(vrp.val[0], vrp.val[1], 6); - vdest.val[1] = vadd_u8(vdest.val[0], vtmp1); - - vtmp1 = vext_u8(vrp.val[1], vrp.val[1], 1); - vdest.val[2] = vadd_u8(vdest.val[1], vtmp2); - vdest.val[3] = vadd_u8(vdest.val[2], vtmp1); - - vtmp = vld1q_u8(rp + 12); - vrpt = png_ptr(uint8x8x2_t, &vtmp); - vrp = *vrpt; - - vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[0]), 0); - rp += 3; - vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[1]), 0); - rp += 3; - vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[2]), 0); - rp += 3; - vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[3]), 0); - rp += 3; - } - - PNG_UNUSED(prev_row) -} - -void -png_read_filter_row_sub4_neon(png_row_infop row_info, png_bytep row, - png_const_bytep prev_row) -{ - png_bytep rp = row; - png_bytep rp_stop = row + row_info->rowbytes; - - uint8x8x4_t vdest; - vdest.val[3] = vdup_n_u8(0); - - png_debug(1, "in png_read_filter_row_sub4_neon"); - - for (; rp < rp_stop; rp += 16) - { - uint32x2x4_t vtmp = vld4_u32(png_ptr(uint32_t,rp)); - uint8x8x4_t *vrpt = png_ptr(uint8x8x4_t,&vtmp); - uint8x8x4_t vrp = *vrpt; - uint32x2x4_t *temp_pointer; - uint32x2x4_t vdest_val; - - vdest.val[0] = vadd_u8(vdest.val[3], vrp.val[0]); - vdest.val[1] = vadd_u8(vdest.val[0], vrp.val[1]); - vdest.val[2] = vadd_u8(vdest.val[1], vrp.val[2]); - vdest.val[3] = vadd_u8(vdest.val[2], vrp.val[3]); - - vdest_val = png_ldr(uint32x2x4_t, &vdest); - vst4_lane_u32(png_ptr(uint32_t,rp), vdest_val, 0); - } - - PNG_UNUSED(prev_row) -} - -void -png_read_filter_row_avg3_neon(png_row_infop row_info, png_bytep row, - png_const_bytep prev_row) -{ - png_bytep rp = row; - png_const_bytep pp = prev_row; - png_bytep rp_stop = row + row_info->rowbytes; - - uint8x16_t vtmp; - uint8x8x2_t *vrpt; - uint8x8x2_t vrp; - uint8x8x4_t vdest; - vdest.val[3] = vdup_n_u8(0); - - vtmp = vld1q_u8(rp); - vrpt = png_ptr(uint8x8x2_t,&vtmp); - vrp = *vrpt; - - png_debug(1, "in png_read_filter_row_avg3_neon"); - - for (; rp < rp_stop; pp += 12) - { - uint8x8_t vtmp1, vtmp2, vtmp3; - - uint8x8x2_t *vppt; - uint8x8x2_t vpp; - - uint32x2_t *temp_pointer; - - vtmp = vld1q_u8(pp); - vppt = png_ptr(uint8x8x2_t,&vtmp); - vpp = *vppt; - - vtmp1 = vext_u8(vrp.val[0], vrp.val[1], 3); - vdest.val[0] = vhadd_u8(vdest.val[3], vpp.val[0]); - vdest.val[0] = vadd_u8(vdest.val[0], vrp.val[0]); - - vtmp2 = vext_u8(vpp.val[0], vpp.val[1], 3); - vtmp3 = vext_u8(vrp.val[0], vrp.val[1], 6); - vdest.val[1] = vhadd_u8(vdest.val[0], vtmp2); - vdest.val[1] = vadd_u8(vdest.val[1], vtmp1); - - vtmp2 = vext_u8(vpp.val[0], vpp.val[1], 6); - vtmp1 = vext_u8(vrp.val[1], vrp.val[1], 1); - - vtmp = vld1q_u8(rp + 12); - vrpt = png_ptr(uint8x8x2_t,&vtmp); - vrp = *vrpt; - - vdest.val[2] = vhadd_u8(vdest.val[1], vtmp2); - vdest.val[2] = vadd_u8(vdest.val[2], vtmp3); - - vtmp2 = vext_u8(vpp.val[1], vpp.val[1], 1); - - vdest.val[3] = vhadd_u8(vdest.val[2], vtmp2); - vdest.val[3] = vadd_u8(vdest.val[3], vtmp1); - - vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[0]), 0); - rp += 3; - vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[1]), 0); - rp += 3; - vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[2]), 0); - rp += 3; - vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[3]), 0); - rp += 3; - } -} - -void -png_read_filter_row_avg4_neon(png_row_infop row_info, png_bytep row, - png_const_bytep prev_row) -{ - png_bytep rp = row; - png_bytep rp_stop = row + row_info->rowbytes; - png_const_bytep pp = prev_row; - - uint8x8x4_t vdest; - vdest.val[3] = vdup_n_u8(0); - - png_debug(1, "in png_read_filter_row_avg4_neon"); - - for (; rp < rp_stop; rp += 16, pp += 16) - { - uint32x2x4_t vtmp; - uint8x8x4_t *vrpt, *vppt; - uint8x8x4_t vrp, vpp; - uint32x2x4_t *temp_pointer; - uint32x2x4_t vdest_val; - - vtmp = vld4_u32(png_ptr(uint32_t,rp)); - vrpt = png_ptr(uint8x8x4_t,&vtmp); - vrp = *vrpt; - vtmp = vld4_u32(png_ptrc(uint32_t,pp)); - vppt = png_ptr(uint8x8x4_t,&vtmp); - vpp = *vppt; - - vdest.val[0] = vhadd_u8(vdest.val[3], vpp.val[0]); - vdest.val[0] = vadd_u8(vdest.val[0], vrp.val[0]); - vdest.val[1] = vhadd_u8(vdest.val[0], vpp.val[1]); - vdest.val[1] = vadd_u8(vdest.val[1], vrp.val[1]); - vdest.val[2] = vhadd_u8(vdest.val[1], vpp.val[2]); - vdest.val[2] = vadd_u8(vdest.val[2], vrp.val[2]); - vdest.val[3] = vhadd_u8(vdest.val[2], vpp.val[3]); - vdest.val[3] = vadd_u8(vdest.val[3], vrp.val[3]); - - vdest_val = png_ldr(uint32x2x4_t, &vdest); - vst4_lane_u32(png_ptr(uint32_t,rp), vdest_val, 0); - } -} - -static uint8x8_t -paeth(uint8x8_t a, uint8x8_t b, uint8x8_t c) -{ - uint8x8_t d, e; - uint16x8_t p1, pa, pb, pc; - - p1 = vaddl_u8(a, b); /* a + b */ - pc = vaddl_u8(c, c); /* c * 2 */ - pa = vabdl_u8(b, c); /* pa */ - pb = vabdl_u8(a, c); /* pb */ - pc = vabdq_u16(p1, pc); /* pc */ - - p1 = vcleq_u16(pa, pb); /* pa <= pb */ - pa = vcleq_u16(pa, pc); /* pa <= pc */ - pb = vcleq_u16(pb, pc); /* pb <= pc */ - - p1 = vandq_u16(p1, pa); /* pa <= pb && pa <= pc */ - - d = vmovn_u16(pb); - e = vmovn_u16(p1); - - d = vbsl_u8(d, b, c); - e = vbsl_u8(e, a, d); - - return e; -} - -void -png_read_filter_row_paeth3_neon(png_row_infop row_info, png_bytep row, - png_const_bytep prev_row) -{ - png_bytep rp = row; - png_const_bytep pp = prev_row; - png_bytep rp_stop = row + row_info->rowbytes; - - uint8x16_t vtmp; - uint8x8x2_t *vrpt; - uint8x8x2_t vrp; - uint8x8_t vlast = vdup_n_u8(0); - uint8x8x4_t vdest; - vdest.val[3] = vdup_n_u8(0); - - vtmp = vld1q_u8(rp); - vrpt = png_ptr(uint8x8x2_t,&vtmp); - vrp = *vrpt; - - png_debug(1, "in png_read_filter_row_paeth3_neon"); - - for (; rp < rp_stop; pp += 12) - { - uint8x8x2_t *vppt; - uint8x8x2_t vpp; - uint8x8_t vtmp1, vtmp2, vtmp3; - uint32x2_t *temp_pointer; - - vtmp = vld1q_u8(pp); - vppt = png_ptr(uint8x8x2_t,&vtmp); - vpp = *vppt; - - vdest.val[0] = paeth(vdest.val[3], vpp.val[0], vlast); - vdest.val[0] = vadd_u8(vdest.val[0], vrp.val[0]); - - vtmp1 = vext_u8(vrp.val[0], vrp.val[1], 3); - vtmp2 = vext_u8(vpp.val[0], vpp.val[1], 3); - vdest.val[1] = paeth(vdest.val[0], vtmp2, vpp.val[0]); - vdest.val[1] = vadd_u8(vdest.val[1], vtmp1); - - vtmp1 = vext_u8(vrp.val[0], vrp.val[1], 6); - vtmp3 = vext_u8(vpp.val[0], vpp.val[1], 6); - vdest.val[2] = paeth(vdest.val[1], vtmp3, vtmp2); - vdest.val[2] = vadd_u8(vdest.val[2], vtmp1); - - vtmp1 = vext_u8(vrp.val[1], vrp.val[1], 1); - vtmp2 = vext_u8(vpp.val[1], vpp.val[1], 1); - - vtmp = vld1q_u8(rp + 12); - vrpt = png_ptr(uint8x8x2_t,&vtmp); - vrp = *vrpt; - - vdest.val[3] = paeth(vdest.val[2], vtmp2, vtmp3); - vdest.val[3] = vadd_u8(vdest.val[3], vtmp1); - - vlast = vtmp2; - - vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[0]), 0); - rp += 3; - vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[1]), 0); - rp += 3; - vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[2]), 0); - rp += 3; - vst1_lane_u32(png_ptr(uint32_t,rp), png_ldr(uint32x2_t,&vdest.val[3]), 0); - rp += 3; - } -} - -void -png_read_filter_row_paeth4_neon(png_row_infop row_info, png_bytep row, - png_const_bytep prev_row) -{ - png_bytep rp = row; - png_bytep rp_stop = row + row_info->rowbytes; - png_const_bytep pp = prev_row; - - uint8x8_t vlast = vdup_n_u8(0); - uint8x8x4_t vdest; - vdest.val[3] = vdup_n_u8(0); - - png_debug(1, "in png_read_filter_row_paeth4_neon"); - - for (; rp < rp_stop; rp += 16, pp += 16) - { - uint32x2x4_t vtmp; - uint8x8x4_t *vrpt, *vppt; - uint8x8x4_t vrp, vpp; - uint32x2x4_t *temp_pointer; - uint32x2x4_t vdest_val; - - vtmp = vld4_u32(png_ptr(uint32_t,rp)); - vrpt = png_ptr(uint8x8x4_t,&vtmp); - vrp = *vrpt; - vtmp = vld4_u32(png_ptrc(uint32_t,pp)); - vppt = png_ptr(uint8x8x4_t,&vtmp); - vpp = *vppt; - - vdest.val[0] = paeth(vdest.val[3], vpp.val[0], vlast); - vdest.val[0] = vadd_u8(vdest.val[0], vrp.val[0]); - vdest.val[1] = paeth(vdest.val[0], vpp.val[1], vpp.val[0]); - vdest.val[1] = vadd_u8(vdest.val[1], vrp.val[1]); - vdest.val[2] = paeth(vdest.val[1], vpp.val[2], vpp.val[1]); - vdest.val[2] = vadd_u8(vdest.val[2], vrp.val[2]); - vdest.val[3] = paeth(vdest.val[2], vpp.val[3], vpp.val[2]); - vdest.val[3] = vadd_u8(vdest.val[3], vrp.val[3]); - - vlast = vpp.val[3]; - - vdest_val = png_ldr(uint32x2x4_t, &vdest); - vst4_lane_u32(png_ptr(uint32_t,rp), vdest_val, 0); - } -} - -#endif /* PNG_ARM_NEON_OPT > 0 */ -#endif /* PNG_ARM_NEON_IMPLEMENTATION == 1 (intrinsics) */ -#endif /* READ */ diff --git a/dep/libpng/src/arm/palette_neon_intrinsics.c b/dep/libpng/src/arm/palette_neon_intrinsics.c deleted file mode 100644 index 92c7d6f9f..000000000 --- a/dep/libpng/src/arm/palette_neon_intrinsics.c +++ /dev/null @@ -1,151 +0,0 @@ - -/* palette_neon_intrinsics.c - NEON optimised palette expansion functions - * - * Copyright (c) 2018-2019 Cosmin Truta - * Copyright (c) 2017-2018 Arm Holdings. All rights reserved. - * Written by Richard Townsend , February 2017. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -#include "../pngpriv.h" - -#if PNG_ARM_NEON_IMPLEMENTATION == 1 - -#if defined(_MSC_VER) && !defined(__clang__) && defined(_M_ARM64) -# include -#else -# include -#endif - -/* Build an RGBA8 palette from the separate RGB and alpha palettes. */ -void -png_riffle_palette_neon(png_structrp png_ptr) -{ - png_const_colorp palette = png_ptr->palette; - png_bytep riffled_palette = png_ptr->riffled_palette; - png_const_bytep trans_alpha = png_ptr->trans_alpha; - int num_trans = png_ptr->num_trans; - int i; - - /* Initially black, opaque. */ - uint8x16x4_t w = {{ - vdupq_n_u8(0x00), - vdupq_n_u8(0x00), - vdupq_n_u8(0x00), - vdupq_n_u8(0xff), - }}; - - png_debug(1, "in png_riffle_palette_neon"); - - /* First, riffle the RGB colours into an RGBA8 palette. - * The alpha component is set to opaque for now. - */ - for (i = 0; i < 256; i += 16) - { - uint8x16x3_t v = vld3q_u8((png_const_bytep)(palette + i)); - w.val[0] = v.val[0]; - w.val[1] = v.val[1]; - w.val[2] = v.val[2]; - vst4q_u8(riffled_palette + (i << 2), w); - } - - /* Fix up the missing transparency values. */ - for (i = 0; i < num_trans; i++) - riffled_palette[(i << 2) + 3] = trans_alpha[i]; -} - -/* Expands a palettized row into RGBA8. */ -int -png_do_expand_palette_rgba8_neon(png_structrp png_ptr, png_row_infop row_info, - png_const_bytep row, png_bytepp ssp, png_bytepp ddp) -{ - png_uint_32 row_width = row_info->width; - const png_uint_32 *riffled_palette = - (const png_uint_32 *)png_ptr->riffled_palette; - const png_uint_32 pixels_per_chunk = 4; - png_uint_32 i; - - png_debug(1, "in png_do_expand_palette_rgba8_neon"); - - PNG_UNUSED(row) - if (row_width < pixels_per_chunk) - return 0; - - /* This function originally gets the last byte of the output row. - * The NEON part writes forward from a given position, so we have - * to seek this back by 4 pixels x 4 bytes. - */ - *ddp = *ddp - ((pixels_per_chunk * sizeof(png_uint_32)) - 1); - - for (i = 0; i < row_width; i += pixels_per_chunk) - { - uint32x4_t cur; - png_bytep sp = *ssp - i, dp = *ddp - (i << 2); - cur = vld1q_dup_u32 (riffled_palette + *(sp - 3)); - cur = vld1q_lane_u32(riffled_palette + *(sp - 2), cur, 1); - cur = vld1q_lane_u32(riffled_palette + *(sp - 1), cur, 2); - cur = vld1q_lane_u32(riffled_palette + *(sp - 0), cur, 3); - vst1q_u32((void *)dp, cur); - } - if (i != row_width) - { - /* Remove the amount that wasn't processed. */ - i -= pixels_per_chunk; - } - - /* Decrement output pointers. */ - *ssp = *ssp - i; - *ddp = *ddp - (i << 2); - return i; -} - -/* Expands a palettized row into RGB8. */ -int -png_do_expand_palette_rgb8_neon(png_structrp png_ptr, png_row_infop row_info, - png_const_bytep row, png_bytepp ssp, png_bytepp ddp) -{ - png_uint_32 row_width = row_info->width; - png_const_bytep palette = (png_const_bytep)png_ptr->palette; - const png_uint_32 pixels_per_chunk = 8; - png_uint_32 i; - - png_debug(1, "in png_do_expand_palette_rgb8_neon"); - - PNG_UNUSED(row) - if (row_width <= pixels_per_chunk) - return 0; - - /* Seeking this back by 8 pixels x 3 bytes. */ - *ddp = *ddp - ((pixels_per_chunk * sizeof(png_color)) - 1); - - for (i = 0; i < row_width; i += pixels_per_chunk) - { - uint8x8x3_t cur; - png_bytep sp = *ssp - i, dp = *ddp - ((i << 1) + i); - cur = vld3_dup_u8(palette + sizeof(png_color) * (*(sp - 7))); - cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 6)), cur, 1); - cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 5)), cur, 2); - cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 4)), cur, 3); - cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 3)), cur, 4); - cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 2)), cur, 5); - cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 1)), cur, 6); - cur = vld3_lane_u8(palette + sizeof(png_color) * (*(sp - 0)), cur, 7); - vst3_u8((void *)dp, cur); - } - - if (i != row_width) - { - /* Remove the amount that wasn't processed. */ - i -= pixels_per_chunk; - } - - /* Decrement output pointers. */ - *ssp = *ssp - i; - *ddp = *ddp - ((i << 1) + i); - return i; -} - -#endif /* PNG_ARM_NEON_IMPLEMENTATION */ diff --git a/dep/libpng/src/intel/filter_sse2_intrinsics.c b/dep/libpng/src/intel/filter_sse2_intrinsics.c deleted file mode 100644 index d3c0fe9e2..000000000 --- a/dep/libpng/src/intel/filter_sse2_intrinsics.c +++ /dev/null @@ -1,391 +0,0 @@ - -/* filter_sse2_intrinsics.c - SSE2 optimized filter functions - * - * Copyright (c) 2018 Cosmin Truta - * Copyright (c) 2016-2017 Glenn Randers-Pehrson - * Written by Mike Klein and Matt Sarett - * Derived from arm/filter_neon_intrinsics.c - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -#include "../pngpriv.h" - -#ifdef PNG_READ_SUPPORTED - -#if PNG_INTEL_SSE_IMPLEMENTATION > 0 - -#include - -/* Functions in this file look at most 3 pixels (a,b,c) to predict the 4th (d). - * They're positioned like this: - * prev: c b - * row: a d - * The Sub filter predicts d=a, Avg d=(a+b)/2, and Paeth predicts d to be - * whichever of a, b, or c is closest to p=a+b-c. - */ - -static __m128i load4(const void* p) { - int tmp; - memcpy(&tmp, p, sizeof(tmp)); - return _mm_cvtsi32_si128(tmp); -} - -static void store4(void* p, __m128i v) { - int tmp = _mm_cvtsi128_si32(v); - memcpy(p, &tmp, sizeof(int)); -} - -static __m128i load3(const void* p) { - png_uint_32 tmp = 0; - memcpy(&tmp, p, 3); - return _mm_cvtsi32_si128(tmp); -} - -static void store3(void* p, __m128i v) { - int tmp = _mm_cvtsi128_si32(v); - memcpy(p, &tmp, 3); -} - -void png_read_filter_row_sub3_sse2(png_row_infop row_info, png_bytep row, - png_const_bytep prev) -{ - /* The Sub filter predicts each pixel as the previous pixel, a. - * There is no pixel to the left of the first pixel. It's encoded directly. - * That works with our main loop if we just say that left pixel was zero. - */ - size_t rb; - - __m128i a, d = _mm_setzero_si128(); - - png_debug(1, "in png_read_filter_row_sub3_sse2"); - - rb = row_info->rowbytes; - while (rb >= 4) { - a = d; d = load4(row); - d = _mm_add_epi8(d, a); - store3(row, d); - - row += 3; - rb -= 3; - } - if (rb > 0) { - a = d; d = load3(row); - d = _mm_add_epi8(d, a); - store3(row, d); - - row += 3; - rb -= 3; - } - PNG_UNUSED(prev) -} - -void png_read_filter_row_sub4_sse2(png_row_infop row_info, png_bytep row, - png_const_bytep prev) -{ - /* The Sub filter predicts each pixel as the previous pixel, a. - * There is no pixel to the left of the first pixel. It's encoded directly. - * That works with our main loop if we just say that left pixel was zero. - */ - size_t rb; - - __m128i a, d = _mm_setzero_si128(); - - png_debug(1, "in png_read_filter_row_sub4_sse2"); - - rb = row_info->rowbytes+4; - while (rb > 4) { - a = d; d = load4(row); - d = _mm_add_epi8(d, a); - store4(row, d); - - row += 4; - rb -= 4; - } - PNG_UNUSED(prev) -} - -void png_read_filter_row_avg3_sse2(png_row_infop row_info, png_bytep row, - png_const_bytep prev) -{ - /* The Avg filter predicts each pixel as the (truncated) average of a and b. - * There's no pixel to the left of the first pixel. Luckily, it's - * predicted to be half of the pixel above it. So again, this works - * perfectly with our loop if we make sure a starts at zero. - */ - - size_t rb; - - const __m128i zero = _mm_setzero_si128(); - - __m128i b; - __m128i a, d = zero; - - png_debug(1, "in png_read_filter_row_avg3_sse2"); - rb = row_info->rowbytes; - while (rb >= 4) { - __m128i avg; - b = load4(prev); - a = d; d = load4(row ); - - /* PNG requires a truncating average, so we can't just use _mm_avg_epu8 */ - avg = _mm_avg_epu8(a,b); - /* ...but we can fix it up by subtracting off 1 if it rounded up. */ - avg = _mm_sub_epi8(avg, _mm_and_si128(_mm_xor_si128(a,b), - _mm_set1_epi8(1))); - d = _mm_add_epi8(d, avg); - store3(row, d); - - prev += 3; - row += 3; - rb -= 3; - } - if (rb > 0) { - __m128i avg; - b = load3(prev); - a = d; d = load3(row ); - - /* PNG requires a truncating average, so we can't just use _mm_avg_epu8 */ - avg = _mm_avg_epu8(a,b); - /* ...but we can fix it up by subtracting off 1 if it rounded up. */ - avg = _mm_sub_epi8(avg, _mm_and_si128(_mm_xor_si128(a,b), - _mm_set1_epi8(1))); - - d = _mm_add_epi8(d, avg); - store3(row, d); - - prev += 3; - row += 3; - rb -= 3; - } -} - -void png_read_filter_row_avg4_sse2(png_row_infop row_info, png_bytep row, - png_const_bytep prev) -{ - /* The Avg filter predicts each pixel as the (truncated) average of a and b. - * There's no pixel to the left of the first pixel. Luckily, it's - * predicted to be half of the pixel above it. So again, this works - * perfectly with our loop if we make sure a starts at zero. - */ - size_t rb; - const __m128i zero = _mm_setzero_si128(); - __m128i b; - __m128i a, d = zero; - - png_debug(1, "in png_read_filter_row_avg4_sse2"); - - rb = row_info->rowbytes+4; - while (rb > 4) { - __m128i avg; - b = load4(prev); - a = d; d = load4(row ); - - /* PNG requires a truncating average, so we can't just use _mm_avg_epu8 */ - avg = _mm_avg_epu8(a,b); - /* ...but we can fix it up by subtracting off 1 if it rounded up. */ - avg = _mm_sub_epi8(avg, _mm_and_si128(_mm_xor_si128(a,b), - _mm_set1_epi8(1))); - - d = _mm_add_epi8(d, avg); - store4(row, d); - - prev += 4; - row += 4; - rb -= 4; - } -} - -/* Returns |x| for 16-bit lanes. */ -static __m128i abs_i16(__m128i x) { -#if PNG_INTEL_SSE_IMPLEMENTATION >= 2 - return _mm_abs_epi16(x); -#else - /* Read this all as, return x<0 ? -x : x. - * To negate two's complement, you flip all the bits then add 1. - */ - __m128i is_negative = _mm_cmplt_epi16(x, _mm_setzero_si128()); - - /* Flip negative lanes. */ - x = _mm_xor_si128(x, is_negative); - - /* +1 to negative lanes, else +0. */ - x = _mm_sub_epi16(x, is_negative); - return x; -#endif -} - -/* Bytewise c ? t : e. */ -static __m128i if_then_else(__m128i c, __m128i t, __m128i e) { -#if PNG_INTEL_SSE_IMPLEMENTATION >= 3 - return _mm_blendv_epi8(e,t,c); -#else - return _mm_or_si128(_mm_and_si128(c, t), _mm_andnot_si128(c, e)); -#endif -} - -void png_read_filter_row_paeth3_sse2(png_row_infop row_info, png_bytep row, - png_const_bytep prev) -{ - /* Paeth tries to predict pixel d using the pixel to the left of it, a, - * and two pixels from the previous row, b and c: - * prev: c b - * row: a d - * The Paeth function predicts d to be whichever of a, b, or c is nearest to - * p=a+b-c. - * - * The first pixel has no left context, and so uses an Up filter, p = b. - * This works naturally with our main loop's p = a+b-c if we force a and c - * to zero. - * Here we zero b and d, which become c and a respectively at the start of - * the loop. - */ - size_t rb; - const __m128i zero = _mm_setzero_si128(); - __m128i c, b = zero, - a, d = zero; - - png_debug(1, "in png_read_filter_row_paeth3_sse2"); - - rb = row_info->rowbytes; - while (rb >= 4) { - /* It's easiest to do this math (particularly, deal with pc) with 16-bit - * intermediates. - */ - __m128i pa,pb,pc,smallest,nearest; - c = b; b = _mm_unpacklo_epi8(load4(prev), zero); - a = d; d = _mm_unpacklo_epi8(load4(row ), zero); - - /* (p-a) == (a+b-c - a) == (b-c) */ - - pa = _mm_sub_epi16(b,c); - - /* (p-b) == (a+b-c - b) == (a-c) */ - pb = _mm_sub_epi16(a,c); - - /* (p-c) == (a+b-c - c) == (a+b-c-c) == (b-c)+(a-c) */ - pc = _mm_add_epi16(pa,pb); - - pa = abs_i16(pa); /* |p-a| */ - pb = abs_i16(pb); /* |p-b| */ - pc = abs_i16(pc); /* |p-c| */ - - smallest = _mm_min_epi16(pc, _mm_min_epi16(pa, pb)); - - /* Paeth breaks ties favoring a over b over c. */ - nearest = if_then_else(_mm_cmpeq_epi16(smallest, pa), a, - if_then_else(_mm_cmpeq_epi16(smallest, pb), b, - c)); - - /* Note `_epi8`: we need addition to wrap modulo 255. */ - d = _mm_add_epi8(d, nearest); - store3(row, _mm_packus_epi16(d,d)); - - prev += 3; - row += 3; - rb -= 3; - } - if (rb > 0) { - /* It's easiest to do this math (particularly, deal with pc) with 16-bit - * intermediates. - */ - __m128i pa,pb,pc,smallest,nearest; - c = b; b = _mm_unpacklo_epi8(load3(prev), zero); - a = d; d = _mm_unpacklo_epi8(load3(row ), zero); - - /* (p-a) == (a+b-c - a) == (b-c) */ - pa = _mm_sub_epi16(b,c); - - /* (p-b) == (a+b-c - b) == (a-c) */ - pb = _mm_sub_epi16(a,c); - - /* (p-c) == (a+b-c - c) == (a+b-c-c) == (b-c)+(a-c) */ - pc = _mm_add_epi16(pa,pb); - - pa = abs_i16(pa); /* |p-a| */ - pb = abs_i16(pb); /* |p-b| */ - pc = abs_i16(pc); /* |p-c| */ - - smallest = _mm_min_epi16(pc, _mm_min_epi16(pa, pb)); - - /* Paeth breaks ties favoring a over b over c. */ - nearest = if_then_else(_mm_cmpeq_epi16(smallest, pa), a, - if_then_else(_mm_cmpeq_epi16(smallest, pb), b, - c)); - - /* Note `_epi8`: we need addition to wrap modulo 255. */ - d = _mm_add_epi8(d, nearest); - store3(row, _mm_packus_epi16(d,d)); - - prev += 3; - row += 3; - rb -= 3; - } -} - -void png_read_filter_row_paeth4_sse2(png_row_infop row_info, png_bytep row, - png_const_bytep prev) -{ - /* Paeth tries to predict pixel d using the pixel to the left of it, a, - * and two pixels from the previous row, b and c: - * prev: c b - * row: a d - * The Paeth function predicts d to be whichever of a, b, or c is nearest to - * p=a+b-c. - * - * The first pixel has no left context, and so uses an Up filter, p = b. - * This works naturally with our main loop's p = a+b-c if we force a and c - * to zero. - * Here we zero b and d, which become c and a respectively at the start of - * the loop. - */ - size_t rb; - const __m128i zero = _mm_setzero_si128(); - __m128i pa,pb,pc,smallest,nearest; - __m128i c, b = zero, - a, d = zero; - - png_debug(1, "in png_read_filter_row_paeth4_sse2"); - - rb = row_info->rowbytes+4; - while (rb > 4) { - /* It's easiest to do this math (particularly, deal with pc) with 16-bit - * intermediates. - */ - c = b; b = _mm_unpacklo_epi8(load4(prev), zero); - a = d; d = _mm_unpacklo_epi8(load4(row ), zero); - - /* (p-a) == (a+b-c - a) == (b-c) */ - pa = _mm_sub_epi16(b,c); - - /* (p-b) == (a+b-c - b) == (a-c) */ - pb = _mm_sub_epi16(a,c); - - /* (p-c) == (a+b-c - c) == (a+b-c-c) == (b-c)+(a-c) */ - pc = _mm_add_epi16(pa,pb); - - pa = abs_i16(pa); /* |p-a| */ - pb = abs_i16(pb); /* |p-b| */ - pc = abs_i16(pc); /* |p-c| */ - - smallest = _mm_min_epi16(pc, _mm_min_epi16(pa, pb)); - - /* Paeth breaks ties favoring a over b over c. */ - nearest = if_then_else(_mm_cmpeq_epi16(smallest, pa), a, - if_then_else(_mm_cmpeq_epi16(smallest, pb), b, - c)); - - /* Note `_epi8`: we need addition to wrap modulo 255. */ - d = _mm_add_epi8(d, nearest); - store4(row, _mm_packus_epi16(d,d)); - - prev += 4; - row += 4; - rb -= 4; - } -} - -#endif /* PNG_INTEL_SSE_IMPLEMENTATION > 0 */ -#endif /* READ */ diff --git a/dep/libpng/src/intel/intel_init.c b/dep/libpng/src/intel/intel_init.c deleted file mode 100644 index 2f8168b7c..000000000 --- a/dep/libpng/src/intel/intel_init.c +++ /dev/null @@ -1,52 +0,0 @@ - -/* intel_init.c - SSE2 optimized filter functions - * - * Copyright (c) 2018 Cosmin Truta - * Copyright (c) 2016-2017 Glenn Randers-Pehrson - * Written by Mike Klein and Matt Sarett, Google, Inc. - * Derived from arm/arm_init.c - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -#include "../pngpriv.h" - -#ifdef PNG_READ_SUPPORTED -#if PNG_INTEL_SSE_IMPLEMENTATION > 0 - -void -png_init_filter_functions_sse2(png_structp pp, unsigned int bpp) -{ - /* The techniques used to implement each of these filters in SSE operate on - * one pixel at a time. - * So they generally speed up 3bpp images about 3x, 4bpp images about 4x. - * They can scale up to 6 and 8 bpp images and down to 2 bpp images, - * but they'd not likely have any benefit for 1bpp images. - * Most of these can be implemented using only MMX and 64-bit registers, - * but they end up a bit slower than using the equally-ubiquitous SSE2. - */ - png_debug(1, "in png_init_filter_functions_sse2"); - if (bpp == 3) - { - pp->read_filter[PNG_FILTER_VALUE_SUB-1] = png_read_filter_row_sub3_sse2; - pp->read_filter[PNG_FILTER_VALUE_AVG-1] = png_read_filter_row_avg3_sse2; - pp->read_filter[PNG_FILTER_VALUE_PAETH-1] = - png_read_filter_row_paeth3_sse2; - } - else if (bpp == 4) - { - pp->read_filter[PNG_FILTER_VALUE_SUB-1] = png_read_filter_row_sub4_sse2; - pp->read_filter[PNG_FILTER_VALUE_AVG-1] = png_read_filter_row_avg4_sse2; - pp->read_filter[PNG_FILTER_VALUE_PAETH-1] = - png_read_filter_row_paeth4_sse2; - } - - /* No need optimize PNG_FILTER_VALUE_UP. The compiler should - * autovectorize. - */ -} - -#endif /* PNG_INTEL_SSE_IMPLEMENTATION > 0 */ -#endif /* PNG_READ_SUPPORTED */ diff --git a/dep/libpng/src/png.c b/dep/libpng/src/png.c deleted file mode 100644 index 9ed315700..000000000 --- a/dep/libpng/src/png.c +++ /dev/null @@ -1,4559 +0,0 @@ - -/* png.c - location for general purpose libpng functions - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -#include "pngpriv.h" - -/* Generate a compiler error if there is an old png.h in the search path. */ -typedef png_libpng_version_1_6_43 Your_png_h_is_not_version_1_6_43; - -/* Tells libpng that we have already handled the first "num_bytes" bytes - * of the PNG file signature. If the PNG data is embedded into another - * stream we can set num_bytes = 8 so that libpng will not attempt to read - * or write any of the magic bytes before it starts on the IHDR. - */ - -#ifdef PNG_READ_SUPPORTED -void PNGAPI -png_set_sig_bytes(png_structrp png_ptr, int num_bytes) -{ - unsigned int nb = (unsigned int)num_bytes; - - png_debug(1, "in png_set_sig_bytes"); - - if (png_ptr == NULL) - return; - - if (num_bytes < 0) - nb = 0; - - if (nb > 8) - png_error(png_ptr, "Too many bytes for PNG signature"); - - png_ptr->sig_bytes = (png_byte)nb; -} - -/* Checks whether the supplied bytes match the PNG signature. We allow - * checking less than the full 8-byte signature so that those apps that - * already read the first few bytes of a file to determine the file type - * can simply check the remaining bytes for extra assurance. Returns - * an integer less than, equal to, or greater than zero if sig is found, - * respectively, to be less than, to match, or be greater than the correct - * PNG signature (this is the same behavior as strcmp, memcmp, etc). - */ -int PNGAPI -png_sig_cmp(png_const_bytep sig, size_t start, size_t num_to_check) -{ - static const png_byte png_signature[8] = {137, 80, 78, 71, 13, 10, 26, 10}; - - if (num_to_check > 8) - num_to_check = 8; - - else if (num_to_check < 1) - return -1; - - if (start > 7) - return -1; - - if (start + num_to_check > 8) - num_to_check = 8 - start; - - return memcmp(&sig[start], &png_signature[start], num_to_check); -} - -#endif /* READ */ - -#if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) -/* Function to allocate memory for zlib */ -PNG_FUNCTION(voidpf /* PRIVATE */, -png_zalloc,(voidpf png_ptr, uInt items, uInt size),PNG_ALLOCATED) -{ - png_alloc_size_t num_bytes = size; - - if (png_ptr == NULL) - return NULL; - - if (items >= (~(png_alloc_size_t)0)/size) - { - png_warning (png_voidcast(png_structrp, png_ptr), - "Potential overflow in png_zalloc()"); - return NULL; - } - - num_bytes *= items; - return png_malloc_warn(png_voidcast(png_structrp, png_ptr), num_bytes); -} - -/* Function to free memory for zlib */ -void /* PRIVATE */ -png_zfree(voidpf png_ptr, voidpf ptr) -{ - png_free(png_voidcast(png_const_structrp,png_ptr), ptr); -} - -/* Reset the CRC variable to 32 bits of 1's. Care must be taken - * in case CRC is > 32 bits to leave the top bits 0. - */ -void /* PRIVATE */ -png_reset_crc(png_structrp png_ptr) -{ - /* The cast is safe because the crc is a 32-bit value. */ - png_ptr->crc = (png_uint_32)crc32(0, Z_NULL, 0); -} - -/* Calculate the CRC over a section of data. We can only pass as - * much data to this routine as the largest single buffer size. We - * also check that this data will actually be used before going to the - * trouble of calculating it. - */ -void /* PRIVATE */ -png_calculate_crc(png_structrp png_ptr, png_const_bytep ptr, size_t length) -{ - int need_crc = 1; - - if (PNG_CHUNK_ANCILLARY(png_ptr->chunk_name) != 0) - { - if ((png_ptr->flags & PNG_FLAG_CRC_ANCILLARY_MASK) == - (PNG_FLAG_CRC_ANCILLARY_USE | PNG_FLAG_CRC_ANCILLARY_NOWARN)) - need_crc = 0; - } - - else /* critical */ - { - if ((png_ptr->flags & PNG_FLAG_CRC_CRITICAL_IGNORE) != 0) - need_crc = 0; - } - - /* 'uLong' is defined in zlib.h as unsigned long; this means that on some - * systems it is a 64-bit value. crc32, however, returns 32 bits so the - * following cast is safe. 'uInt' may be no more than 16 bits, so it is - * necessary to perform a loop here. - */ - if (need_crc != 0 && length > 0) - { - uLong crc = png_ptr->crc; /* Should never issue a warning */ - - do - { - uInt safe_length = (uInt)length; -#ifndef __COVERITY__ - if (safe_length == 0) - safe_length = (uInt)-1; /* evil, but safe */ -#endif - - crc = crc32(crc, ptr, safe_length); - - /* The following should never issue compiler warnings; if they do the - * target system has characteristics that will probably violate other - * assumptions within the libpng code. - */ - ptr += safe_length; - length -= safe_length; - } - while (length > 0); - - /* And the following is always safe because the crc is only 32 bits. */ - png_ptr->crc = (png_uint_32)crc; - } -} - -/* Check a user supplied version number, called from both read and write - * functions that create a png_struct. - */ -int -png_user_version_check(png_structrp png_ptr, png_const_charp user_png_ver) -{ - /* Libpng versions 1.0.0 and later are binary compatible if the version - * string matches through the second '.'; we must recompile any - * applications that use any older library version. - */ - - if (user_png_ver != NULL) - { - int i = -1; - int found_dots = 0; - - do - { - i++; - if (user_png_ver[i] != PNG_LIBPNG_VER_STRING[i]) - png_ptr->flags |= PNG_FLAG_LIBRARY_MISMATCH; - if (user_png_ver[i] == '.') - found_dots++; - } while (found_dots < 2 && user_png_ver[i] != 0 && - PNG_LIBPNG_VER_STRING[i] != 0); - } - - else - png_ptr->flags |= PNG_FLAG_LIBRARY_MISMATCH; - - if ((png_ptr->flags & PNG_FLAG_LIBRARY_MISMATCH) != 0) - { -#ifdef PNG_WARNINGS_SUPPORTED - size_t pos = 0; - char m[128]; - - pos = png_safecat(m, (sizeof m), pos, - "Application built with libpng-"); - pos = png_safecat(m, (sizeof m), pos, user_png_ver); - pos = png_safecat(m, (sizeof m), pos, " but running with "); - pos = png_safecat(m, (sizeof m), pos, PNG_LIBPNG_VER_STRING); - PNG_UNUSED(pos) - - png_warning(png_ptr, m); -#endif - -#ifdef PNG_ERROR_NUMBERS_SUPPORTED - png_ptr->flags = 0; -#endif - - return 0; - } - - /* Success return. */ - return 1; -} - -/* Generic function to create a png_struct for either read or write - this - * contains the common initialization. - */ -PNG_FUNCTION(png_structp /* PRIVATE */, -png_create_png_struct,(png_const_charp user_png_ver, png_voidp error_ptr, - png_error_ptr error_fn, png_error_ptr warn_fn, png_voidp mem_ptr, - png_malloc_ptr malloc_fn, png_free_ptr free_fn),PNG_ALLOCATED) -{ - png_struct create_struct; -# ifdef PNG_SETJMP_SUPPORTED - jmp_buf create_jmp_buf; -# endif - - /* This temporary stack-allocated structure is used to provide a place to - * build enough context to allow the user provided memory allocator (if any) - * to be called. - */ - memset(&create_struct, 0, (sizeof create_struct)); - - /* Added at libpng-1.2.6 */ -# ifdef PNG_USER_LIMITS_SUPPORTED - create_struct.user_width_max = PNG_USER_WIDTH_MAX; - create_struct.user_height_max = PNG_USER_HEIGHT_MAX; - -# ifdef PNG_USER_CHUNK_CACHE_MAX - /* Added at libpng-1.2.43 and 1.4.0 */ - create_struct.user_chunk_cache_max = PNG_USER_CHUNK_CACHE_MAX; -# endif - -# ifdef PNG_USER_CHUNK_MALLOC_MAX - /* Added at libpng-1.2.43 and 1.4.1, required only for read but exists - * in png_struct regardless. - */ - create_struct.user_chunk_malloc_max = PNG_USER_CHUNK_MALLOC_MAX; -# endif -# endif - - /* The following two API calls simply set fields in png_struct, so it is safe - * to do them now even though error handling is not yet set up. - */ -# ifdef PNG_USER_MEM_SUPPORTED - png_set_mem_fn(&create_struct, mem_ptr, malloc_fn, free_fn); -# else - PNG_UNUSED(mem_ptr) - PNG_UNUSED(malloc_fn) - PNG_UNUSED(free_fn) -# endif - - /* (*error_fn) can return control to the caller after the error_ptr is set, - * this will result in a memory leak unless the error_fn does something - * extremely sophisticated. The design lacks merit but is implicit in the - * API. - */ - png_set_error_fn(&create_struct, error_ptr, error_fn, warn_fn); - -# ifdef PNG_SETJMP_SUPPORTED - if (!setjmp(create_jmp_buf)) -# endif - { -# ifdef PNG_SETJMP_SUPPORTED - /* Temporarily fake out the longjmp information until we have - * successfully completed this function. This only works if we have - * setjmp() support compiled in, but it is safe - this stuff should - * never happen. - */ - create_struct.jmp_buf_ptr = &create_jmp_buf; - create_struct.jmp_buf_size = 0; /*stack allocation*/ - create_struct.longjmp_fn = longjmp; -# endif - /* Call the general version checker (shared with read and write code): - */ - if (png_user_version_check(&create_struct, user_png_ver) != 0) - { - png_structrp png_ptr = png_voidcast(png_structrp, - png_malloc_warn(&create_struct, (sizeof *png_ptr))); - - if (png_ptr != NULL) - { - /* png_ptr->zstream holds a back-pointer to the png_struct, so - * this can only be done now: - */ - create_struct.zstream.zalloc = png_zalloc; - create_struct.zstream.zfree = png_zfree; - create_struct.zstream.opaque = png_ptr; - -# ifdef PNG_SETJMP_SUPPORTED - /* Eliminate the local error handling: */ - create_struct.jmp_buf_ptr = NULL; - create_struct.jmp_buf_size = 0; - create_struct.longjmp_fn = 0; -# endif - - *png_ptr = create_struct; - - /* This is the successful return point */ - return png_ptr; - } - } - } - - /* A longjmp because of a bug in the application storage allocator or a - * simple failure to allocate the png_struct. - */ - return NULL; -} - -/* Allocate the memory for an info_struct for the application. */ -PNG_FUNCTION(png_infop,PNGAPI -png_create_info_struct,(png_const_structrp png_ptr),PNG_ALLOCATED) -{ - png_inforp info_ptr; - - png_debug(1, "in png_create_info_struct"); - - if (png_ptr == NULL) - return NULL; - - /* Use the internal API that does not (or at least should not) error out, so - * that this call always returns ok. The application typically sets up the - * error handling *after* creating the info_struct because this is the way it - * has always been done in 'example.c'. - */ - info_ptr = png_voidcast(png_inforp, png_malloc_base(png_ptr, - (sizeof *info_ptr))); - - if (info_ptr != NULL) - memset(info_ptr, 0, (sizeof *info_ptr)); - - return info_ptr; -} - -/* This function frees the memory associated with a single info struct. - * Normally, one would use either png_destroy_read_struct() or - * png_destroy_write_struct() to free an info struct, but this may be - * useful for some applications. From libpng 1.6.0 this function is also used - * internally to implement the png_info release part of the 'struct' destroy - * APIs. This ensures that all possible approaches free the same data (all of - * it). - */ -void PNGAPI -png_destroy_info_struct(png_const_structrp png_ptr, png_infopp info_ptr_ptr) -{ - png_inforp info_ptr = NULL; - - png_debug(1, "in png_destroy_info_struct"); - - if (png_ptr == NULL) - return; - - if (info_ptr_ptr != NULL) - info_ptr = *info_ptr_ptr; - - if (info_ptr != NULL) - { - /* Do this first in case of an error below; if the app implements its own - * memory management this can lead to png_free calling png_error, which - * will abort this routine and return control to the app error handler. - * An infinite loop may result if it then tries to free the same info - * ptr. - */ - *info_ptr_ptr = NULL; - - png_free_data(png_ptr, info_ptr, PNG_FREE_ALL, -1); - memset(info_ptr, 0, (sizeof *info_ptr)); - png_free(png_ptr, info_ptr); - } -} - -/* Initialize the info structure. This is now an internal function (0.89) - * and applications using it are urged to use png_create_info_struct() - * instead. Use deprecated in 1.6.0, internal use removed (used internally it - * is just a memset). - * - * NOTE: it is almost inconceivable that this API is used because it bypasses - * the user-memory mechanism and the user error handling/warning mechanisms in - * those cases where it does anything other than a memset. - */ -PNG_FUNCTION(void,PNGAPI -png_info_init_3,(png_infopp ptr_ptr, size_t png_info_struct_size), - PNG_DEPRECATED) -{ - png_inforp info_ptr = *ptr_ptr; - - png_debug(1, "in png_info_init_3"); - - if (info_ptr == NULL) - return; - - if ((sizeof (png_info)) > png_info_struct_size) - { - *ptr_ptr = NULL; - /* The following line is why this API should not be used: */ - free(info_ptr); - info_ptr = png_voidcast(png_inforp, png_malloc_base(NULL, - (sizeof *info_ptr))); - if (info_ptr == NULL) - return; - *ptr_ptr = info_ptr; - } - - /* Set everything to 0 */ - memset(info_ptr, 0, (sizeof *info_ptr)); -} - -void PNGAPI -png_data_freer(png_const_structrp png_ptr, png_inforp info_ptr, - int freer, png_uint_32 mask) -{ - png_debug(1, "in png_data_freer"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - if (freer == PNG_DESTROY_WILL_FREE_DATA) - info_ptr->free_me |= mask; - - else if (freer == PNG_USER_WILL_FREE_DATA) - info_ptr->free_me &= ~mask; - - else - png_error(png_ptr, "Unknown freer parameter in png_data_freer"); -} - -void PNGAPI -png_free_data(png_const_structrp png_ptr, png_inforp info_ptr, png_uint_32 mask, - int num) -{ - png_debug(1, "in png_free_data"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - -#ifdef PNG_TEXT_SUPPORTED - /* Free text item num or (if num == -1) all text items */ - if (info_ptr->text != NULL && - ((mask & PNG_FREE_TEXT) & info_ptr->free_me) != 0) - { - if (num != -1) - { - png_free(png_ptr, info_ptr->text[num].key); - info_ptr->text[num].key = NULL; - } - - else - { - int i; - - for (i = 0; i < info_ptr->num_text; i++) - png_free(png_ptr, info_ptr->text[i].key); - - png_free(png_ptr, info_ptr->text); - info_ptr->text = NULL; - info_ptr->num_text = 0; - info_ptr->max_text = 0; - } - } -#endif - -#ifdef PNG_tRNS_SUPPORTED - /* Free any tRNS entry */ - if (((mask & PNG_FREE_TRNS) & info_ptr->free_me) != 0) - { - info_ptr->valid &= ~PNG_INFO_tRNS; - png_free(png_ptr, info_ptr->trans_alpha); - info_ptr->trans_alpha = NULL; - info_ptr->num_trans = 0; - } -#endif - -#ifdef PNG_sCAL_SUPPORTED - /* Free any sCAL entry */ - if (((mask & PNG_FREE_SCAL) & info_ptr->free_me) != 0) - { - png_free(png_ptr, info_ptr->scal_s_width); - png_free(png_ptr, info_ptr->scal_s_height); - info_ptr->scal_s_width = NULL; - info_ptr->scal_s_height = NULL; - info_ptr->valid &= ~PNG_INFO_sCAL; - } -#endif - -#ifdef PNG_pCAL_SUPPORTED - /* Free any pCAL entry */ - if (((mask & PNG_FREE_PCAL) & info_ptr->free_me) != 0) - { - png_free(png_ptr, info_ptr->pcal_purpose); - png_free(png_ptr, info_ptr->pcal_units); - info_ptr->pcal_purpose = NULL; - info_ptr->pcal_units = NULL; - - if (info_ptr->pcal_params != NULL) - { - int i; - - for (i = 0; i < info_ptr->pcal_nparams; i++) - png_free(png_ptr, info_ptr->pcal_params[i]); - - png_free(png_ptr, info_ptr->pcal_params); - info_ptr->pcal_params = NULL; - } - info_ptr->valid &= ~PNG_INFO_pCAL; - } -#endif - -#ifdef PNG_iCCP_SUPPORTED - /* Free any profile entry */ - if (((mask & PNG_FREE_ICCP) & info_ptr->free_me) != 0) - { - png_free(png_ptr, info_ptr->iccp_name); - png_free(png_ptr, info_ptr->iccp_profile); - info_ptr->iccp_name = NULL; - info_ptr->iccp_profile = NULL; - info_ptr->valid &= ~PNG_INFO_iCCP; - } -#endif - -#ifdef PNG_sPLT_SUPPORTED - /* Free a given sPLT entry, or (if num == -1) all sPLT entries */ - if (info_ptr->splt_palettes != NULL && - ((mask & PNG_FREE_SPLT) & info_ptr->free_me) != 0) - { - if (num != -1) - { - png_free(png_ptr, info_ptr->splt_palettes[num].name); - png_free(png_ptr, info_ptr->splt_palettes[num].entries); - info_ptr->splt_palettes[num].name = NULL; - info_ptr->splt_palettes[num].entries = NULL; - } - - else - { - int i; - - for (i = 0; i < info_ptr->splt_palettes_num; i++) - { - png_free(png_ptr, info_ptr->splt_palettes[i].name); - png_free(png_ptr, info_ptr->splt_palettes[i].entries); - } - - png_free(png_ptr, info_ptr->splt_palettes); - info_ptr->splt_palettes = NULL; - info_ptr->splt_palettes_num = 0; - info_ptr->valid &= ~PNG_INFO_sPLT; - } - } -#endif - -#ifdef PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED - if (info_ptr->unknown_chunks != NULL && - ((mask & PNG_FREE_UNKN) & info_ptr->free_me) != 0) - { - if (num != -1) - { - png_free(png_ptr, info_ptr->unknown_chunks[num].data); - info_ptr->unknown_chunks[num].data = NULL; - } - - else - { - int i; - - for (i = 0; i < info_ptr->unknown_chunks_num; i++) - png_free(png_ptr, info_ptr->unknown_chunks[i].data); - - png_free(png_ptr, info_ptr->unknown_chunks); - info_ptr->unknown_chunks = NULL; - info_ptr->unknown_chunks_num = 0; - } - } -#endif - -#ifdef PNG_eXIf_SUPPORTED - /* Free any eXIf entry */ - if (((mask & PNG_FREE_EXIF) & info_ptr->free_me) != 0) - { -# ifdef PNG_READ_eXIf_SUPPORTED - if (info_ptr->eXIf_buf) - { - png_free(png_ptr, info_ptr->eXIf_buf); - info_ptr->eXIf_buf = NULL; - } -# endif - if (info_ptr->exif) - { - png_free(png_ptr, info_ptr->exif); - info_ptr->exif = NULL; - } - info_ptr->valid &= ~PNG_INFO_eXIf; - } -#endif - -#ifdef PNG_hIST_SUPPORTED - /* Free any hIST entry */ - if (((mask & PNG_FREE_HIST) & info_ptr->free_me) != 0) - { - png_free(png_ptr, info_ptr->hist); - info_ptr->hist = NULL; - info_ptr->valid &= ~PNG_INFO_hIST; - } -#endif - - /* Free any PLTE entry that was internally allocated */ - if (((mask & PNG_FREE_PLTE) & info_ptr->free_me) != 0) - { - png_free(png_ptr, info_ptr->palette); - info_ptr->palette = NULL; - info_ptr->valid &= ~PNG_INFO_PLTE; - info_ptr->num_palette = 0; - } - -#ifdef PNG_INFO_IMAGE_SUPPORTED - /* Free any image bits attached to the info structure */ - if (((mask & PNG_FREE_ROWS) & info_ptr->free_me) != 0) - { - if (info_ptr->row_pointers != NULL) - { - png_uint_32 row; - for (row = 0; row < info_ptr->height; row++) - png_free(png_ptr, info_ptr->row_pointers[row]); - - png_free(png_ptr, info_ptr->row_pointers); - info_ptr->row_pointers = NULL; - } - info_ptr->valid &= ~PNG_INFO_IDAT; - } -#endif - - if (num != -1) - mask &= ~PNG_FREE_MUL; - - info_ptr->free_me &= ~mask; -} -#endif /* READ || WRITE */ - -/* This function returns a pointer to the io_ptr associated with the user - * functions. The application should free any memory associated with this - * pointer before png_write_destroy() or png_read_destroy() are called. - */ -png_voidp PNGAPI -png_get_io_ptr(png_const_structrp png_ptr) -{ - if (png_ptr == NULL) - return NULL; - - return png_ptr->io_ptr; -} - -#if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) -# ifdef PNG_STDIO_SUPPORTED -/* Initialize the default input/output functions for the PNG file. If you - * use your own read or write routines, you can call either png_set_read_fn() - * or png_set_write_fn() instead of png_init_io(). If you have defined - * PNG_NO_STDIO or otherwise disabled PNG_STDIO_SUPPORTED, you must use a - * function of your own because "FILE *" isn't necessarily available. - */ -void PNGAPI -png_init_io(png_structrp png_ptr, png_FILE_p fp) -{ - png_debug(1, "in png_init_io"); - - if (png_ptr == NULL) - return; - - png_ptr->io_ptr = (png_voidp)fp; -} -# endif - -# ifdef PNG_SAVE_INT_32_SUPPORTED -/* PNG signed integers are saved in 32-bit 2's complement format. ANSI C-90 - * defines a cast of a signed integer to an unsigned integer either to preserve - * the value, if it is positive, or to calculate: - * - * (UNSIGNED_MAX+1) + integer - * - * Where UNSIGNED_MAX is the appropriate maximum unsigned value, so when the - * negative integral value is added the result will be an unsigned value - * corresponding to the 2's complement representation. - */ -void PNGAPI -png_save_int_32(png_bytep buf, png_int_32 i) -{ - png_save_uint_32(buf, (png_uint_32)i); -} -# endif - -# ifdef PNG_TIME_RFC1123_SUPPORTED -/* Convert the supplied time into an RFC 1123 string suitable for use in - * a "Creation Time" or other text-based time string. - */ -int PNGAPI -png_convert_to_rfc1123_buffer(char out[29], png_const_timep ptime) -{ - static const char short_months[12][4] = - {"Jan", "Feb", "Mar", "Apr", "May", "Jun", - "Jul", "Aug", "Sep", "Oct", "Nov", "Dec"}; - - if (out == NULL) - return 0; - - if (ptime->year > 9999 /* RFC1123 limitation */ || - ptime->month == 0 || ptime->month > 12 || - ptime->day == 0 || ptime->day > 31 || - ptime->hour > 23 || ptime->minute > 59 || - ptime->second > 60) - return 0; - - { - size_t pos = 0; - char number_buf[5] = {0, 0, 0, 0, 0}; /* enough for a four-digit year */ - -# define APPEND_STRING(string) pos = png_safecat(out, 29, pos, (string)) -# define APPEND_NUMBER(format, value)\ - APPEND_STRING(PNG_FORMAT_NUMBER(number_buf, format, (value))) -# define APPEND(ch) if (pos < 28) out[pos++] = (ch) - - APPEND_NUMBER(PNG_NUMBER_FORMAT_u, (unsigned)ptime->day); - APPEND(' '); - APPEND_STRING(short_months[(ptime->month - 1)]); - APPEND(' '); - APPEND_NUMBER(PNG_NUMBER_FORMAT_u, ptime->year); - APPEND(' '); - APPEND_NUMBER(PNG_NUMBER_FORMAT_02u, (unsigned)ptime->hour); - APPEND(':'); - APPEND_NUMBER(PNG_NUMBER_FORMAT_02u, (unsigned)ptime->minute); - APPEND(':'); - APPEND_NUMBER(PNG_NUMBER_FORMAT_02u, (unsigned)ptime->second); - APPEND_STRING(" +0000"); /* This reliably terminates the buffer */ - PNG_UNUSED (pos) - -# undef APPEND -# undef APPEND_NUMBER -# undef APPEND_STRING - } - - return 1; -} - -# if PNG_LIBPNG_VER < 10700 -/* To do: remove the following from libpng-1.7 */ -/* Original API that uses a private buffer in png_struct. - * Deprecated because it causes png_struct to carry a spurious temporary - * buffer (png_struct::time_buffer), better to have the caller pass this in. - */ -png_const_charp PNGAPI -png_convert_to_rfc1123(png_structrp png_ptr, png_const_timep ptime) -{ - if (png_ptr != NULL) - { - /* The only failure above if png_ptr != NULL is from an invalid ptime */ - if (png_convert_to_rfc1123_buffer(png_ptr->time_buffer, ptime) == 0) - png_warning(png_ptr, "Ignoring invalid time value"); - - else - return png_ptr->time_buffer; - } - - return NULL; -} -# endif /* LIBPNG_VER < 10700 */ -# endif /* TIME_RFC1123 */ - -#endif /* READ || WRITE */ - -png_const_charp PNGAPI -png_get_copyright(png_const_structrp png_ptr) -{ - PNG_UNUSED(png_ptr) /* Silence compiler warning about unused png_ptr */ -#ifdef PNG_STRING_COPYRIGHT - return PNG_STRING_COPYRIGHT -#else - return PNG_STRING_NEWLINE \ - "libpng version 1.6.43" PNG_STRING_NEWLINE \ - "Copyright (c) 2018-2024 Cosmin Truta" PNG_STRING_NEWLINE \ - "Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson" \ - PNG_STRING_NEWLINE \ - "Copyright (c) 1996-1997 Andreas Dilger" PNG_STRING_NEWLINE \ - "Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc." \ - PNG_STRING_NEWLINE; -#endif -} - -/* The following return the library version as a short string in the - * format 1.0.0 through 99.99.99zz. To get the version of *.h files - * used with your application, print out PNG_LIBPNG_VER_STRING, which - * is defined in png.h. - * Note: now there is no difference between png_get_libpng_ver() and - * png_get_header_ver(). Due to the version_nn_nn_nn typedef guard, - * it is guaranteed that png.c uses the correct version of png.h. - */ -png_const_charp PNGAPI -png_get_libpng_ver(png_const_structrp png_ptr) -{ - /* Version of *.c files used when building libpng */ - return png_get_header_ver(png_ptr); -} - -png_const_charp PNGAPI -png_get_header_ver(png_const_structrp png_ptr) -{ - /* Version of *.h files used when building libpng */ - PNG_UNUSED(png_ptr) /* Silence compiler warning about unused png_ptr */ - return PNG_LIBPNG_VER_STRING; -} - -png_const_charp PNGAPI -png_get_header_version(png_const_structrp png_ptr) -{ - /* Returns longer string containing both version and date */ - PNG_UNUSED(png_ptr) /* Silence compiler warning about unused png_ptr */ -#ifdef __STDC__ - return PNG_HEADER_VERSION_STRING -# ifndef PNG_READ_SUPPORTED - " (NO READ SUPPORT)" -# endif - PNG_STRING_NEWLINE; -#else - return PNG_HEADER_VERSION_STRING; -#endif -} - -#ifdef PNG_BUILD_GRAYSCALE_PALETTE_SUPPORTED -/* NOTE: this routine is not used internally! */ -/* Build a grayscale palette. Palette is assumed to be 1 << bit_depth - * large of png_color. This lets grayscale images be treated as - * paletted. Most useful for gamma correction and simplification - * of code. This API is not used internally. - */ -void PNGAPI -png_build_grayscale_palette(int bit_depth, png_colorp palette) -{ - int num_palette; - int color_inc; - int i; - int v; - - png_debug(1, "in png_do_build_grayscale_palette"); - - if (palette == NULL) - return; - - switch (bit_depth) - { - case 1: - num_palette = 2; - color_inc = 0xff; - break; - - case 2: - num_palette = 4; - color_inc = 0x55; - break; - - case 4: - num_palette = 16; - color_inc = 0x11; - break; - - case 8: - num_palette = 256; - color_inc = 1; - break; - - default: - num_palette = 0; - color_inc = 0; - break; - } - - for (i = 0, v = 0; i < num_palette; i++, v += color_inc) - { - palette[i].red = (png_byte)(v & 0xff); - palette[i].green = (png_byte)(v & 0xff); - palette[i].blue = (png_byte)(v & 0xff); - } -} -#endif - -#ifdef PNG_SET_UNKNOWN_CHUNKS_SUPPORTED -int PNGAPI -png_handle_as_unknown(png_const_structrp png_ptr, png_const_bytep chunk_name) -{ - /* Check chunk_name and return "keep" value if it's on the list, else 0 */ - png_const_bytep p, p_end; - - if (png_ptr == NULL || chunk_name == NULL || png_ptr->num_chunk_list == 0) - return PNG_HANDLE_CHUNK_AS_DEFAULT; - - p_end = png_ptr->chunk_list; - p = p_end + png_ptr->num_chunk_list*5; /* beyond end */ - - /* The code is the fifth byte after each four byte string. Historically this - * code was always searched from the end of the list, this is no longer - * necessary because the 'set' routine handles duplicate entries correctly. - */ - do /* num_chunk_list > 0, so at least one */ - { - p -= 5; - - if (memcmp(chunk_name, p, 4) == 0) - return p[4]; - } - while (p > p_end); - - /* This means that known chunks should be processed and unknown chunks should - * be handled according to the value of png_ptr->unknown_default; this can be - * confusing because, as a result, there are two levels of defaulting for - * unknown chunks. - */ - return PNG_HANDLE_CHUNK_AS_DEFAULT; -} - -#if defined(PNG_READ_UNKNOWN_CHUNKS_SUPPORTED) ||\ - defined(PNG_HANDLE_AS_UNKNOWN_SUPPORTED) -int /* PRIVATE */ -png_chunk_unknown_handling(png_const_structrp png_ptr, png_uint_32 chunk_name) -{ - png_byte chunk_string[5]; - - PNG_CSTRING_FROM_CHUNK(chunk_string, chunk_name); - return png_handle_as_unknown(png_ptr, chunk_string); -} -#endif /* READ_UNKNOWN_CHUNKS || HANDLE_AS_UNKNOWN */ -#endif /* SET_UNKNOWN_CHUNKS */ - -#ifdef PNG_READ_SUPPORTED -/* This function, added to libpng-1.0.6g, is untested. */ -int PNGAPI -png_reset_zstream(png_structrp png_ptr) -{ - if (png_ptr == NULL) - return Z_STREAM_ERROR; - - /* WARNING: this resets the window bits to the maximum! */ - return inflateReset(&png_ptr->zstream); -} -#endif /* READ */ - -/* This function was added to libpng-1.0.7 */ -png_uint_32 PNGAPI -png_access_version_number(void) -{ - /* Version of *.c files used when building libpng */ - return (png_uint_32)PNG_LIBPNG_VER; -} - -#if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) -/* Ensure that png_ptr->zstream.msg holds some appropriate error message string. - * If it doesn't 'ret' is used to set it to something appropriate, even in cases - * like Z_OK or Z_STREAM_END where the error code is apparently a success code. - */ -void /* PRIVATE */ -png_zstream_error(png_structrp png_ptr, int ret) -{ - /* Translate 'ret' into an appropriate error string, priority is given to the - * one in zstream if set. This always returns a string, even in cases like - * Z_OK or Z_STREAM_END where the error code is a success code. - */ - if (png_ptr->zstream.msg == NULL) switch (ret) - { - default: - case Z_OK: - png_ptr->zstream.msg = PNGZ_MSG_CAST("unexpected zlib return code"); - break; - - case Z_STREAM_END: - /* Normal exit */ - png_ptr->zstream.msg = PNGZ_MSG_CAST("unexpected end of LZ stream"); - break; - - case Z_NEED_DICT: - /* This means the deflate stream did not have a dictionary; this - * indicates a bogus PNG. - */ - png_ptr->zstream.msg = PNGZ_MSG_CAST("missing LZ dictionary"); - break; - - case Z_ERRNO: - /* gz APIs only: should not happen */ - png_ptr->zstream.msg = PNGZ_MSG_CAST("zlib IO error"); - break; - - case Z_STREAM_ERROR: - /* internal libpng error */ - png_ptr->zstream.msg = PNGZ_MSG_CAST("bad parameters to zlib"); - break; - - case Z_DATA_ERROR: - png_ptr->zstream.msg = PNGZ_MSG_CAST("damaged LZ stream"); - break; - - case Z_MEM_ERROR: - png_ptr->zstream.msg = PNGZ_MSG_CAST("insufficient memory"); - break; - - case Z_BUF_ERROR: - /* End of input or output; not a problem if the caller is doing - * incremental read or write. - */ - png_ptr->zstream.msg = PNGZ_MSG_CAST("truncated"); - break; - - case Z_VERSION_ERROR: - png_ptr->zstream.msg = PNGZ_MSG_CAST("unsupported zlib version"); - break; - - case PNG_UNEXPECTED_ZLIB_RETURN: - /* Compile errors here mean that zlib now uses the value co-opted in - * pngpriv.h for PNG_UNEXPECTED_ZLIB_RETURN; update the switch above - * and change pngpriv.h. Note that this message is "... return", - * whereas the default/Z_OK one is "... return code". - */ - png_ptr->zstream.msg = PNGZ_MSG_CAST("unexpected zlib return"); - break; - } -} - -/* png_convert_size: a PNGAPI but no longer in png.h, so deleted - * at libpng 1.5.5! - */ - -/* Added at libpng version 1.2.34 and 1.4.0 (moved from pngset.c) */ -#ifdef PNG_GAMMA_SUPPORTED /* always set if COLORSPACE */ -static int -png_colorspace_check_gamma(png_const_structrp png_ptr, - png_colorspacerp colorspace, png_fixed_point gAMA, int from) - /* This is called to check a new gamma value against an existing one. The - * routine returns false if the new gamma value should not be written. - * - * 'from' says where the new gamma value comes from: - * - * 0: the new gamma value is the libpng estimate for an ICC profile - * 1: the new gamma value comes from a gAMA chunk - * 2: the new gamma value comes from an sRGB chunk - */ -{ - png_fixed_point gtest; - - if ((colorspace->flags & PNG_COLORSPACE_HAVE_GAMMA) != 0 && - (png_muldiv(>est, colorspace->gamma, PNG_FP_1, gAMA) == 0 || - png_gamma_significant(gtest) != 0)) - { - /* Either this is an sRGB image, in which case the calculated gamma - * approximation should match, or this is an image with a profile and the - * value libpng calculates for the gamma of the profile does not match the - * value recorded in the file. The former, sRGB, case is an error, the - * latter is just a warning. - */ - if ((colorspace->flags & PNG_COLORSPACE_FROM_sRGB) != 0 || from == 2) - { - png_chunk_report(png_ptr, "gamma value does not match sRGB", - PNG_CHUNK_ERROR); - /* Do not overwrite an sRGB value */ - return from == 2; - } - - else /* sRGB tag not involved */ - { - png_chunk_report(png_ptr, "gamma value does not match libpng estimate", - PNG_CHUNK_WARNING); - return from == 1; - } - } - - return 1; -} - -void /* PRIVATE */ -png_colorspace_set_gamma(png_const_structrp png_ptr, - png_colorspacerp colorspace, png_fixed_point gAMA) -{ - /* Changed in libpng-1.5.4 to limit the values to ensure overflow can't - * occur. Since the fixed point representation is asymmetrical it is - * possible for 1/gamma to overflow the limit of 21474 and this means the - * gamma value must be at least 5/100000 and hence at most 20000.0. For - * safety the limits here are a little narrower. The values are 0.00016 to - * 6250.0, which are truly ridiculous gamma values (and will produce - * displays that are all black or all white.) - * - * In 1.6.0 this test replaces the ones in pngrutil.c, in the gAMA chunk - * handling code, which only required the value to be >0. - */ - png_const_charp errmsg; - - if (gAMA < 16 || gAMA > 625000000) - errmsg = "gamma value out of range"; - -# ifdef PNG_READ_gAMA_SUPPORTED - /* Allow the application to set the gamma value more than once */ - else if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0 && - (colorspace->flags & PNG_COLORSPACE_FROM_gAMA) != 0) - errmsg = "duplicate"; -# endif - - /* Do nothing if the colorspace is already invalid */ - else if ((colorspace->flags & PNG_COLORSPACE_INVALID) != 0) - return; - - else - { - if (png_colorspace_check_gamma(png_ptr, colorspace, gAMA, - 1/*from gAMA*/) != 0) - { - /* Store this gamma value. */ - colorspace->gamma = gAMA; - colorspace->flags |= - (PNG_COLORSPACE_HAVE_GAMMA | PNG_COLORSPACE_FROM_gAMA); - } - - /* At present if the check_gamma test fails the gamma of the colorspace is - * not updated however the colorspace is not invalidated. This - * corresponds to the case where the existing gamma comes from an sRGB - * chunk or profile. An error message has already been output. - */ - return; - } - - /* Error exit - errmsg has been set. */ - colorspace->flags |= PNG_COLORSPACE_INVALID; - png_chunk_report(png_ptr, errmsg, PNG_CHUNK_WRITE_ERROR); -} - -void /* PRIVATE */ -png_colorspace_sync_info(png_const_structrp png_ptr, png_inforp info_ptr) -{ - if ((info_ptr->colorspace.flags & PNG_COLORSPACE_INVALID) != 0) - { - /* Everything is invalid */ - info_ptr->valid &= ~(PNG_INFO_gAMA|PNG_INFO_cHRM|PNG_INFO_sRGB| - PNG_INFO_iCCP); - -# ifdef PNG_COLORSPACE_SUPPORTED - /* Clean up the iCCP profile now if it won't be used. */ - png_free_data(png_ptr, info_ptr, PNG_FREE_ICCP, -1/*not used*/); -# else - PNG_UNUSED(png_ptr) -# endif - } - - else - { -# ifdef PNG_COLORSPACE_SUPPORTED - /* Leave the INFO_iCCP flag set if the pngset.c code has already set - * it; this allows a PNG to contain a profile which matches sRGB and - * yet still have that profile retrievable by the application. - */ - if ((info_ptr->colorspace.flags & PNG_COLORSPACE_MATCHES_sRGB) != 0) - info_ptr->valid |= PNG_INFO_sRGB; - - else - info_ptr->valid &= ~PNG_INFO_sRGB; - - if ((info_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_ENDPOINTS) != 0) - info_ptr->valid |= PNG_INFO_cHRM; - - else - info_ptr->valid &= ~PNG_INFO_cHRM; -# endif - - if ((info_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_GAMMA) != 0) - info_ptr->valid |= PNG_INFO_gAMA; - - else - info_ptr->valid &= ~PNG_INFO_gAMA; - } -} - -#ifdef PNG_READ_SUPPORTED -void /* PRIVATE */ -png_colorspace_sync(png_const_structrp png_ptr, png_inforp info_ptr) -{ - if (info_ptr == NULL) /* reduce code size; check here not in the caller */ - return; - - info_ptr->colorspace = png_ptr->colorspace; - png_colorspace_sync_info(png_ptr, info_ptr); -} -#endif -#endif /* GAMMA */ - -#ifdef PNG_COLORSPACE_SUPPORTED -/* Added at libpng-1.5.5 to support read and write of true CIEXYZ values for - * cHRM, as opposed to using chromaticities. These internal APIs return - * non-zero on a parameter error. The X, Y and Z values are required to be - * positive and less than 1.0. - */ -static int -png_xy_from_XYZ(png_xy *xy, const png_XYZ *XYZ) -{ - png_int_32 d, dwhite, whiteX, whiteY; - - d = XYZ->red_X + XYZ->red_Y + XYZ->red_Z; - if (png_muldiv(&xy->redx, XYZ->red_X, PNG_FP_1, d) == 0) - return 1; - if (png_muldiv(&xy->redy, XYZ->red_Y, PNG_FP_1, d) == 0) - return 1; - dwhite = d; - whiteX = XYZ->red_X; - whiteY = XYZ->red_Y; - - d = XYZ->green_X + XYZ->green_Y + XYZ->green_Z; - if (png_muldiv(&xy->greenx, XYZ->green_X, PNG_FP_1, d) == 0) - return 1; - if (png_muldiv(&xy->greeny, XYZ->green_Y, PNG_FP_1, d) == 0) - return 1; - dwhite += d; - whiteX += XYZ->green_X; - whiteY += XYZ->green_Y; - - d = XYZ->blue_X + XYZ->blue_Y + XYZ->blue_Z; - if (png_muldiv(&xy->bluex, XYZ->blue_X, PNG_FP_1, d) == 0) - return 1; - if (png_muldiv(&xy->bluey, XYZ->blue_Y, PNG_FP_1, d) == 0) - return 1; - dwhite += d; - whiteX += XYZ->blue_X; - whiteY += XYZ->blue_Y; - - /* The reference white is simply the sum of the end-point (X,Y,Z) vectors, - * thus: - */ - if (png_muldiv(&xy->whitex, whiteX, PNG_FP_1, dwhite) == 0) - return 1; - if (png_muldiv(&xy->whitey, whiteY, PNG_FP_1, dwhite) == 0) - return 1; - - return 0; -} - -static int -png_XYZ_from_xy(png_XYZ *XYZ, const png_xy *xy) -{ - png_fixed_point red_inverse, green_inverse, blue_scale; - png_fixed_point left, right, denominator; - - /* Check xy and, implicitly, z. Note that wide gamut color spaces typically - * have end points with 0 tristimulus values (these are impossible end - * points, but they are used to cover the possible colors). We check - * xy->whitey against 5, not 0, to avoid a possible integer overflow. - */ - if (xy->redx < 0 || xy->redx > PNG_FP_1) return 1; - if (xy->redy < 0 || xy->redy > PNG_FP_1-xy->redx) return 1; - if (xy->greenx < 0 || xy->greenx > PNG_FP_1) return 1; - if (xy->greeny < 0 || xy->greeny > PNG_FP_1-xy->greenx) return 1; - if (xy->bluex < 0 || xy->bluex > PNG_FP_1) return 1; - if (xy->bluey < 0 || xy->bluey > PNG_FP_1-xy->bluex) return 1; - if (xy->whitex < 0 || xy->whitex > PNG_FP_1) return 1; - if (xy->whitey < 5 || xy->whitey > PNG_FP_1-xy->whitex) return 1; - - /* The reverse calculation is more difficult because the original tristimulus - * value had 9 independent values (red,green,blue)x(X,Y,Z) however only 8 - * derived values were recorded in the cHRM chunk; - * (red,green,blue,white)x(x,y). This loses one degree of freedom and - * therefore an arbitrary ninth value has to be introduced to undo the - * original transformations. - * - * Think of the original end-points as points in (X,Y,Z) space. The - * chromaticity values (c) have the property: - * - * C - * c = --------- - * X + Y + Z - * - * For each c (x,y,z) from the corresponding original C (X,Y,Z). Thus the - * three chromaticity values (x,y,z) for each end-point obey the - * relationship: - * - * x + y + z = 1 - * - * This describes the plane in (X,Y,Z) space that intersects each axis at the - * value 1.0; call this the chromaticity plane. Thus the chromaticity - * calculation has scaled each end-point so that it is on the x+y+z=1 plane - * and chromaticity is the intersection of the vector from the origin to the - * (X,Y,Z) value with the chromaticity plane. - * - * To fully invert the chromaticity calculation we would need the three - * end-point scale factors, (red-scale, green-scale, blue-scale), but these - * were not recorded. Instead we calculated the reference white (X,Y,Z) and - * recorded the chromaticity of this. The reference white (X,Y,Z) would have - * given all three of the scale factors since: - * - * color-C = color-c * color-scale - * white-C = red-C + green-C + blue-C - * = red-c*red-scale + green-c*green-scale + blue-c*blue-scale - * - * But cHRM records only white-x and white-y, so we have lost the white scale - * factor: - * - * white-C = white-c*white-scale - * - * To handle this the inverse transformation makes an arbitrary assumption - * about white-scale: - * - * Assume: white-Y = 1.0 - * Hence: white-scale = 1/white-y - * Or: red-Y + green-Y + blue-Y = 1.0 - * - * Notice the last statement of the assumption gives an equation in three of - * the nine values we want to calculate. 8 more equations come from the - * above routine as summarised at the top above (the chromaticity - * calculation): - * - * Given: color-x = color-X / (color-X + color-Y + color-Z) - * Hence: (color-x - 1)*color-X + color.x*color-Y + color.x*color-Z = 0 - * - * This is 9 simultaneous equations in the 9 variables "color-C" and can be - * solved by Cramer's rule. Cramer's rule requires calculating 10 9x9 matrix - * determinants, however this is not as bad as it seems because only 28 of - * the total of 90 terms in the various matrices are non-zero. Nevertheless - * Cramer's rule is notoriously numerically unstable because the determinant - * calculation involves the difference of large, but similar, numbers. It is - * difficult to be sure that the calculation is stable for real world values - * and it is certain that it becomes unstable where the end points are close - * together. - * - * So this code uses the perhaps slightly less optimal but more - * understandable and totally obvious approach of calculating color-scale. - * - * This algorithm depends on the precision in white-scale and that is - * (1/white-y), so we can immediately see that as white-y approaches 0 the - * accuracy inherent in the cHRM chunk drops off substantially. - * - * libpng arithmetic: a simple inversion of the above equations - * ------------------------------------------------------------ - * - * white_scale = 1/white-y - * white-X = white-x * white-scale - * white-Y = 1.0 - * white-Z = (1 - white-x - white-y) * white_scale - * - * white-C = red-C + green-C + blue-C - * = red-c*red-scale + green-c*green-scale + blue-c*blue-scale - * - * This gives us three equations in (red-scale,green-scale,blue-scale) where - * all the coefficients are now known: - * - * red-x*red-scale + green-x*green-scale + blue-x*blue-scale - * = white-x/white-y - * red-y*red-scale + green-y*green-scale + blue-y*blue-scale = 1 - * red-z*red-scale + green-z*green-scale + blue-z*blue-scale - * = (1 - white-x - white-y)/white-y - * - * In the last equation color-z is (1 - color-x - color-y) so we can add all - * three equations together to get an alternative third: - * - * red-scale + green-scale + blue-scale = 1/white-y = white-scale - * - * So now we have a Cramer's rule solution where the determinants are just - * 3x3 - far more tractible. Unfortunately 3x3 determinants still involve - * multiplication of three coefficients so we can't guarantee to avoid - * overflow in the libpng fixed point representation. Using Cramer's rule in - * floating point is probably a good choice here, but it's not an option for - * fixed point. Instead proceed to simplify the first two equations by - * eliminating what is likely to be the largest value, blue-scale: - * - * blue-scale = white-scale - red-scale - green-scale - * - * Hence: - * - * (red-x - blue-x)*red-scale + (green-x - blue-x)*green-scale = - * (white-x - blue-x)*white-scale - * - * (red-y - blue-y)*red-scale + (green-y - blue-y)*green-scale = - * 1 - blue-y*white-scale - * - * And now we can trivially solve for (red-scale,green-scale): - * - * green-scale = - * (white-x - blue-x)*white-scale - (red-x - blue-x)*red-scale - * ----------------------------------------------------------- - * green-x - blue-x - * - * red-scale = - * 1 - blue-y*white-scale - (green-y - blue-y) * green-scale - * --------------------------------------------------------- - * red-y - blue-y - * - * Hence: - * - * red-scale = - * ( (green-x - blue-x) * (white-y - blue-y) - - * (green-y - blue-y) * (white-x - blue-x) ) / white-y - * ------------------------------------------------------------------------- - * (green-x - blue-x)*(red-y - blue-y)-(green-y - blue-y)*(red-x - blue-x) - * - * green-scale = - * ( (red-y - blue-y) * (white-x - blue-x) - - * (red-x - blue-x) * (white-y - blue-y) ) / white-y - * ------------------------------------------------------------------------- - * (green-x - blue-x)*(red-y - blue-y)-(green-y - blue-y)*(red-x - blue-x) - * - * Accuracy: - * The input values have 5 decimal digits of accuracy. The values are all in - * the range 0 < value < 1, so simple products are in the same range but may - * need up to 10 decimal digits to preserve the original precision and avoid - * underflow. Because we are using a 32-bit signed representation we cannot - * match this; the best is a little over 9 decimal digits, less than 10. - * - * The approach used here is to preserve the maximum precision within the - * signed representation. Because the red-scale calculation above uses the - * difference between two products of values that must be in the range -1..+1 - * it is sufficient to divide the product by 7; ceil(100,000/32767*2). The - * factor is irrelevant in the calculation because it is applied to both - * numerator and denominator. - * - * Note that the values of the differences of the products of the - * chromaticities in the above equations tend to be small, for example for - * the sRGB chromaticities they are: - * - * red numerator: -0.04751 - * green numerator: -0.08788 - * denominator: -0.2241 (without white-y multiplication) - * - * The resultant Y coefficients from the chromaticities of some widely used - * color space definitions are (to 15 decimal places): - * - * sRGB - * 0.212639005871510 0.715168678767756 0.072192315360734 - * Kodak ProPhoto - * 0.288071128229293 0.711843217810102 0.000085653960605 - * Adobe RGB - * 0.297344975250536 0.627363566255466 0.075291458493998 - * Adobe Wide Gamut RGB - * 0.258728243040113 0.724682314948566 0.016589442011321 - */ - /* By the argument, above overflow should be impossible here. The return - * value of 2 indicates an internal error to the caller. - */ - if (png_muldiv(&left, xy->greenx-xy->bluex, xy->redy - xy->bluey, 7) == 0) - return 2; - if (png_muldiv(&right, xy->greeny-xy->bluey, xy->redx - xy->bluex, 7) == 0) - return 2; - denominator = left - right; - - /* Now find the red numerator. */ - if (png_muldiv(&left, xy->greenx-xy->bluex, xy->whitey-xy->bluey, 7) == 0) - return 2; - if (png_muldiv(&right, xy->greeny-xy->bluey, xy->whitex-xy->bluex, 7) == 0) - return 2; - - /* Overflow is possible here and it indicates an extreme set of PNG cHRM - * chunk values. This calculation actually returns the reciprocal of the - * scale value because this allows us to delay the multiplication of white-y - * into the denominator, which tends to produce a small number. - */ - if (png_muldiv(&red_inverse, xy->whitey, denominator, left-right) == 0 || - red_inverse <= xy->whitey /* r+g+b scales = white scale */) - return 1; - - /* Similarly for green_inverse: */ - if (png_muldiv(&left, xy->redy-xy->bluey, xy->whitex-xy->bluex, 7) == 0) - return 2; - if (png_muldiv(&right, xy->redx-xy->bluex, xy->whitey-xy->bluey, 7) == 0) - return 2; - if (png_muldiv(&green_inverse, xy->whitey, denominator, left-right) == 0 || - green_inverse <= xy->whitey) - return 1; - - /* And the blue scale, the checks above guarantee this can't overflow but it - * can still produce 0 for extreme cHRM values. - */ - blue_scale = png_reciprocal(xy->whitey) - png_reciprocal(red_inverse) - - png_reciprocal(green_inverse); - if (blue_scale <= 0) - return 1; - - - /* And fill in the png_XYZ: */ - if (png_muldiv(&XYZ->red_X, xy->redx, PNG_FP_1, red_inverse) == 0) - return 1; - if (png_muldiv(&XYZ->red_Y, xy->redy, PNG_FP_1, red_inverse) == 0) - return 1; - if (png_muldiv(&XYZ->red_Z, PNG_FP_1 - xy->redx - xy->redy, PNG_FP_1, - red_inverse) == 0) - return 1; - - if (png_muldiv(&XYZ->green_X, xy->greenx, PNG_FP_1, green_inverse) == 0) - return 1; - if (png_muldiv(&XYZ->green_Y, xy->greeny, PNG_FP_1, green_inverse) == 0) - return 1; - if (png_muldiv(&XYZ->green_Z, PNG_FP_1 - xy->greenx - xy->greeny, PNG_FP_1, - green_inverse) == 0) - return 1; - - if (png_muldiv(&XYZ->blue_X, xy->bluex, blue_scale, PNG_FP_1) == 0) - return 1; - if (png_muldiv(&XYZ->blue_Y, xy->bluey, blue_scale, PNG_FP_1) == 0) - return 1; - if (png_muldiv(&XYZ->blue_Z, PNG_FP_1 - xy->bluex - xy->bluey, blue_scale, - PNG_FP_1) == 0) - return 1; - - return 0; /*success*/ -} - -static int -png_XYZ_normalize(png_XYZ *XYZ) -{ - png_int_32 Y; - - if (XYZ->red_Y < 0 || XYZ->green_Y < 0 || XYZ->blue_Y < 0 || - XYZ->red_X < 0 || XYZ->green_X < 0 || XYZ->blue_X < 0 || - XYZ->red_Z < 0 || XYZ->green_Z < 0 || XYZ->blue_Z < 0) - return 1; - - /* Normalize by scaling so the sum of the end-point Y values is PNG_FP_1. - * IMPLEMENTATION NOTE: ANSI requires signed overflow not to occur, therefore - * relying on addition of two positive values producing a negative one is not - * safe. - */ - Y = XYZ->red_Y; - if (0x7fffffff - Y < XYZ->green_X) - return 1; - Y += XYZ->green_Y; - if (0x7fffffff - Y < XYZ->blue_X) - return 1; - Y += XYZ->blue_Y; - - if (Y != PNG_FP_1) - { - if (png_muldiv(&XYZ->red_X, XYZ->red_X, PNG_FP_1, Y) == 0) - return 1; - if (png_muldiv(&XYZ->red_Y, XYZ->red_Y, PNG_FP_1, Y) == 0) - return 1; - if (png_muldiv(&XYZ->red_Z, XYZ->red_Z, PNG_FP_1, Y) == 0) - return 1; - - if (png_muldiv(&XYZ->green_X, XYZ->green_X, PNG_FP_1, Y) == 0) - return 1; - if (png_muldiv(&XYZ->green_Y, XYZ->green_Y, PNG_FP_1, Y) == 0) - return 1; - if (png_muldiv(&XYZ->green_Z, XYZ->green_Z, PNG_FP_1, Y) == 0) - return 1; - - if (png_muldiv(&XYZ->blue_X, XYZ->blue_X, PNG_FP_1, Y) == 0) - return 1; - if (png_muldiv(&XYZ->blue_Y, XYZ->blue_Y, PNG_FP_1, Y) == 0) - return 1; - if (png_muldiv(&XYZ->blue_Z, XYZ->blue_Z, PNG_FP_1, Y) == 0) - return 1; - } - - return 0; -} - -static int -png_colorspace_endpoints_match(const png_xy *xy1, const png_xy *xy2, int delta) -{ - /* Allow an error of +/-0.01 (absolute value) on each chromaticity */ - if (PNG_OUT_OF_RANGE(xy1->whitex, xy2->whitex,delta) || - PNG_OUT_OF_RANGE(xy1->whitey, xy2->whitey,delta) || - PNG_OUT_OF_RANGE(xy1->redx, xy2->redx, delta) || - PNG_OUT_OF_RANGE(xy1->redy, xy2->redy, delta) || - PNG_OUT_OF_RANGE(xy1->greenx, xy2->greenx,delta) || - PNG_OUT_OF_RANGE(xy1->greeny, xy2->greeny,delta) || - PNG_OUT_OF_RANGE(xy1->bluex, xy2->bluex, delta) || - PNG_OUT_OF_RANGE(xy1->bluey, xy2->bluey, delta)) - return 0; - return 1; -} - -/* Added in libpng-1.6.0, a different check for the validity of a set of cHRM - * chunk chromaticities. Earlier checks used to simply look for the overflow - * condition (where the determinant of the matrix to solve for XYZ ends up zero - * because the chromaticity values are not all distinct.) Despite this it is - * theoretically possible to produce chromaticities that are apparently valid - * but that rapidly degrade to invalid, potentially crashing, sets because of - * arithmetic inaccuracies when calculations are performed on them. The new - * check is to round-trip xy -> XYZ -> xy and then check that the result is - * within a small percentage of the original. - */ -static int -png_colorspace_check_xy(png_XYZ *XYZ, const png_xy *xy) -{ - int result; - png_xy xy_test; - - /* As a side-effect this routine also returns the XYZ endpoints. */ - result = png_XYZ_from_xy(XYZ, xy); - if (result != 0) - return result; - - result = png_xy_from_XYZ(&xy_test, XYZ); - if (result != 0) - return result; - - if (png_colorspace_endpoints_match(xy, &xy_test, - 5/*actually, the math is pretty accurate*/) != 0) - return 0; - - /* Too much slip */ - return 1; -} - -/* This is the check going the other way. The XYZ is modified to normalize it - * (another side-effect) and the xy chromaticities are returned. - */ -static int -png_colorspace_check_XYZ(png_xy *xy, png_XYZ *XYZ) -{ - int result; - png_XYZ XYZtemp; - - result = png_XYZ_normalize(XYZ); - if (result != 0) - return result; - - result = png_xy_from_XYZ(xy, XYZ); - if (result != 0) - return result; - - XYZtemp = *XYZ; - return png_colorspace_check_xy(&XYZtemp, xy); -} - -/* Used to check for an endpoint match against sRGB */ -static const png_xy sRGB_xy = /* From ITU-R BT.709-3 */ -{ - /* color x y */ - /* red */ 64000, 33000, - /* green */ 30000, 60000, - /* blue */ 15000, 6000, - /* white */ 31270, 32900 -}; - -static int -png_colorspace_set_xy_and_XYZ(png_const_structrp png_ptr, - png_colorspacerp colorspace, const png_xy *xy, const png_XYZ *XYZ, - int preferred) -{ - if ((colorspace->flags & PNG_COLORSPACE_INVALID) != 0) - return 0; - - /* The consistency check is performed on the chromaticities; this factors out - * variations because of the normalization (or not) of the end point Y - * values. - */ - if (preferred < 2 && - (colorspace->flags & PNG_COLORSPACE_HAVE_ENDPOINTS) != 0) - { - /* The end points must be reasonably close to any we already have. The - * following allows an error of up to +/-.001 - */ - if (png_colorspace_endpoints_match(xy, &colorspace->end_points_xy, - 100) == 0) - { - colorspace->flags |= PNG_COLORSPACE_INVALID; - png_benign_error(png_ptr, "inconsistent chromaticities"); - return 0; /* failed */ - } - - /* Only overwrite with preferred values */ - if (preferred == 0) - return 1; /* ok, but no change */ - } - - colorspace->end_points_xy = *xy; - colorspace->end_points_XYZ = *XYZ; - colorspace->flags |= PNG_COLORSPACE_HAVE_ENDPOINTS; - - /* The end points are normally quoted to two decimal digits, so allow +/-0.01 - * on this test. - */ - if (png_colorspace_endpoints_match(xy, &sRGB_xy, 1000) != 0) - colorspace->flags |= PNG_COLORSPACE_ENDPOINTS_MATCH_sRGB; - - else - colorspace->flags &= PNG_COLORSPACE_CANCEL( - PNG_COLORSPACE_ENDPOINTS_MATCH_sRGB); - - return 2; /* ok and changed */ -} - -int /* PRIVATE */ -png_colorspace_set_chromaticities(png_const_structrp png_ptr, - png_colorspacerp colorspace, const png_xy *xy, int preferred) -{ - /* We must check the end points to ensure they are reasonable - in the past - * color management systems have crashed as a result of getting bogus - * colorant values, while this isn't the fault of libpng it is the - * responsibility of libpng because PNG carries the bomb and libpng is in a - * position to protect against it. - */ - png_XYZ XYZ; - - switch (png_colorspace_check_xy(&XYZ, xy)) - { - case 0: /* success */ - return png_colorspace_set_xy_and_XYZ(png_ptr, colorspace, xy, &XYZ, - preferred); - - case 1: - /* We can't invert the chromaticities so we can't produce value XYZ - * values. Likely as not a color management system will fail too. - */ - colorspace->flags |= PNG_COLORSPACE_INVALID; - png_benign_error(png_ptr, "invalid chromaticities"); - break; - - default: - /* libpng is broken; this should be a warning but if it happens we - * want error reports so for the moment it is an error. - */ - colorspace->flags |= PNG_COLORSPACE_INVALID; - png_error(png_ptr, "internal error checking chromaticities"); - } - - return 0; /* failed */ -} - -int /* PRIVATE */ -png_colorspace_set_endpoints(png_const_structrp png_ptr, - png_colorspacerp colorspace, const png_XYZ *XYZ_in, int preferred) -{ - png_XYZ XYZ = *XYZ_in; - png_xy xy; - - switch (png_colorspace_check_XYZ(&xy, &XYZ)) - { - case 0: - return png_colorspace_set_xy_and_XYZ(png_ptr, colorspace, &xy, &XYZ, - preferred); - - case 1: - /* End points are invalid. */ - colorspace->flags |= PNG_COLORSPACE_INVALID; - png_benign_error(png_ptr, "invalid end points"); - break; - - default: - colorspace->flags |= PNG_COLORSPACE_INVALID; - png_error(png_ptr, "internal error checking chromaticities"); - } - - return 0; /* failed */ -} - -#if defined(PNG_sRGB_SUPPORTED) || defined(PNG_iCCP_SUPPORTED) -/* Error message generation */ -static char -png_icc_tag_char(png_uint_32 byte) -{ - byte &= 0xff; - if (byte >= 32 && byte <= 126) - return (char)byte; - else - return '?'; -} - -static void -png_icc_tag_name(char *name, png_uint_32 tag) -{ - name[0] = '\''; - name[1] = png_icc_tag_char(tag >> 24); - name[2] = png_icc_tag_char(tag >> 16); - name[3] = png_icc_tag_char(tag >> 8); - name[4] = png_icc_tag_char(tag ); - name[5] = '\''; -} - -static int -is_ICC_signature_char(png_alloc_size_t it) -{ - return it == 32 || (it >= 48 && it <= 57) || (it >= 65 && it <= 90) || - (it >= 97 && it <= 122); -} - -static int -is_ICC_signature(png_alloc_size_t it) -{ - return is_ICC_signature_char(it >> 24) /* checks all the top bits */ && - is_ICC_signature_char((it >> 16) & 0xff) && - is_ICC_signature_char((it >> 8) & 0xff) && - is_ICC_signature_char(it & 0xff); -} - -static int -png_icc_profile_error(png_const_structrp png_ptr, png_colorspacerp colorspace, - png_const_charp name, png_alloc_size_t value, png_const_charp reason) -{ - size_t pos; - char message[196]; /* see below for calculation */ - - if (colorspace != NULL) - colorspace->flags |= PNG_COLORSPACE_INVALID; - - pos = png_safecat(message, (sizeof message), 0, "profile '"); /* 9 chars */ - pos = png_safecat(message, pos+79, pos, name); /* Truncate to 79 chars */ - pos = png_safecat(message, (sizeof message), pos, "': "); /* +2 = 90 */ - if (is_ICC_signature(value) != 0) - { - /* So 'value' is at most 4 bytes and the following cast is safe */ - png_icc_tag_name(message+pos, (png_uint_32)value); - pos += 6; /* total +8; less than the else clause */ - message[pos++] = ':'; - message[pos++] = ' '; - } -# ifdef PNG_WARNINGS_SUPPORTED - else - { - char number[PNG_NUMBER_BUFFER_SIZE]; /* +24 = 114 */ - - pos = png_safecat(message, (sizeof message), pos, - png_format_number(number, number+(sizeof number), - PNG_NUMBER_FORMAT_x, value)); - pos = png_safecat(message, (sizeof message), pos, "h: "); /* +2 = 116 */ - } -# endif - /* The 'reason' is an arbitrary message, allow +79 maximum 195 */ - pos = png_safecat(message, (sizeof message), pos, reason); - PNG_UNUSED(pos) - - /* This is recoverable, but make it unconditionally an app_error on write to - * avoid writing invalid ICC profiles into PNG files (i.e., we handle them - * on read, with a warning, but on write unless the app turns off - * application errors the PNG won't be written.) - */ - png_chunk_report(png_ptr, message, - (colorspace != NULL) ? PNG_CHUNK_ERROR : PNG_CHUNK_WRITE_ERROR); - - return 0; -} -#endif /* sRGB || iCCP */ - -#ifdef PNG_sRGB_SUPPORTED -int /* PRIVATE */ -png_colorspace_set_sRGB(png_const_structrp png_ptr, png_colorspacerp colorspace, - int intent) -{ - /* sRGB sets known gamma, end points and (from the chunk) intent. */ - /* IMPORTANT: these are not necessarily the values found in an ICC profile - * because ICC profiles store values adapted to a D50 environment; it is - * expected that the ICC profile mediaWhitePointTag will be D50; see the - * checks and code elsewhere to understand this better. - * - * These XYZ values, which are accurate to 5dp, produce rgb to gray - * coefficients of (6968,23435,2366), which are reduced (because they add up - * to 32769 not 32768) to (6968,23434,2366). These are the values that - * libpng has traditionally used (and are the best values given the 15bit - * algorithm used by the rgb to gray code.) - */ - static const png_XYZ sRGB_XYZ = /* D65 XYZ (*not* the D50 adapted values!) */ - { - /* color X Y Z */ - /* red */ 41239, 21264, 1933, - /* green */ 35758, 71517, 11919, - /* blue */ 18048, 7219, 95053 - }; - - /* Do nothing if the colorspace is already invalidated. */ - if ((colorspace->flags & PNG_COLORSPACE_INVALID) != 0) - return 0; - - /* Check the intent, then check for existing settings. It is valid for the - * PNG file to have cHRM or gAMA chunks along with sRGB, but the values must - * be consistent with the correct values. If, however, this function is - * called below because an iCCP chunk matches sRGB then it is quite - * conceivable that an older app recorded incorrect gAMA and cHRM because of - * an incorrect calculation based on the values in the profile - this does - * *not* invalidate the profile (though it still produces an error, which can - * be ignored.) - */ - if (intent < 0 || intent >= PNG_sRGB_INTENT_LAST) - return png_icc_profile_error(png_ptr, colorspace, "sRGB", - (png_alloc_size_t)intent, "invalid sRGB rendering intent"); - - if ((colorspace->flags & PNG_COLORSPACE_HAVE_INTENT) != 0 && - colorspace->rendering_intent != intent) - return png_icc_profile_error(png_ptr, colorspace, "sRGB", - (png_alloc_size_t)intent, "inconsistent rendering intents"); - - if ((colorspace->flags & PNG_COLORSPACE_FROM_sRGB) != 0) - { - png_benign_error(png_ptr, "duplicate sRGB information ignored"); - return 0; - } - - /* If the standard sRGB cHRM chunk does not match the one from the PNG file - * warn but overwrite the value with the correct one. - */ - if ((colorspace->flags & PNG_COLORSPACE_HAVE_ENDPOINTS) != 0 && - !png_colorspace_endpoints_match(&sRGB_xy, &colorspace->end_points_xy, - 100)) - png_chunk_report(png_ptr, "cHRM chunk does not match sRGB", - PNG_CHUNK_ERROR); - - /* This check is just done for the error reporting - the routine always - * returns true when the 'from' argument corresponds to sRGB (2). - */ - (void)png_colorspace_check_gamma(png_ptr, colorspace, PNG_GAMMA_sRGB_INVERSE, - 2/*from sRGB*/); - - /* intent: bugs in GCC force 'int' to be used as the parameter type. */ - colorspace->rendering_intent = (png_uint_16)intent; - colorspace->flags |= PNG_COLORSPACE_HAVE_INTENT; - - /* endpoints */ - colorspace->end_points_xy = sRGB_xy; - colorspace->end_points_XYZ = sRGB_XYZ; - colorspace->flags |= - (PNG_COLORSPACE_HAVE_ENDPOINTS|PNG_COLORSPACE_ENDPOINTS_MATCH_sRGB); - - /* gamma */ - colorspace->gamma = PNG_GAMMA_sRGB_INVERSE; - colorspace->flags |= PNG_COLORSPACE_HAVE_GAMMA; - - /* Finally record that we have an sRGB profile */ - colorspace->flags |= - (PNG_COLORSPACE_MATCHES_sRGB|PNG_COLORSPACE_FROM_sRGB); - - return 1; /* set */ -} -#endif /* sRGB */ - -#ifdef PNG_iCCP_SUPPORTED -/* Encoded value of D50 as an ICC XYZNumber. From the ICC 2010 spec the value - * is XYZ(0.9642,1.0,0.8249), which scales to: - * - * (63189.8112, 65536, 54060.6464) - */ -static const png_byte D50_nCIEXYZ[12] = - { 0x00, 0x00, 0xf6, 0xd6, 0x00, 0x01, 0x00, 0x00, 0x00, 0x00, 0xd3, 0x2d }; - -static int /* bool */ -icc_check_length(png_const_structrp png_ptr, png_colorspacerp colorspace, - png_const_charp name, png_uint_32 profile_length) -{ - if (profile_length < 132) - return png_icc_profile_error(png_ptr, colorspace, name, profile_length, - "too short"); - return 1; -} - -#ifdef PNG_READ_iCCP_SUPPORTED -int /* PRIVATE */ -png_icc_check_length(png_const_structrp png_ptr, png_colorspacerp colorspace, - png_const_charp name, png_uint_32 profile_length) -{ - if (!icc_check_length(png_ptr, colorspace, name, profile_length)) - return 0; - - /* This needs to be here because the 'normal' check is in - * png_decompress_chunk, yet this happens after the attempt to - * png_malloc_base the required data. We only need this on read; on write - * the caller supplies the profile buffer so libpng doesn't allocate it. See - * the call to icc_check_length below (the write case). - */ -# ifdef PNG_SET_USER_LIMITS_SUPPORTED - else if (png_ptr->user_chunk_malloc_max > 0 && - png_ptr->user_chunk_malloc_max < profile_length) - return png_icc_profile_error(png_ptr, colorspace, name, profile_length, - "exceeds application limits"); -# elif PNG_USER_CHUNK_MALLOC_MAX > 0 - else if (PNG_USER_CHUNK_MALLOC_MAX < profile_length) - return png_icc_profile_error(png_ptr, colorspace, name, profile_length, - "exceeds libpng limits"); -# else /* !SET_USER_LIMITS */ - /* This will get compiled out on all 32-bit and better systems. */ - else if (PNG_SIZE_MAX < profile_length) - return png_icc_profile_error(png_ptr, colorspace, name, profile_length, - "exceeds system limits"); -# endif /* !SET_USER_LIMITS */ - - return 1; -} -#endif /* READ_iCCP */ - -int /* PRIVATE */ -png_icc_check_header(png_const_structrp png_ptr, png_colorspacerp colorspace, - png_const_charp name, png_uint_32 profile_length, - png_const_bytep profile/* first 132 bytes only */, int color_type) -{ - png_uint_32 temp; - - /* Length check; this cannot be ignored in this code because profile_length - * is used later to check the tag table, so even if the profile seems over - * long profile_length from the caller must be correct. The caller can fix - * this up on read or write by just passing in the profile header length. - */ - temp = png_get_uint_32(profile); - if (temp != profile_length) - return png_icc_profile_error(png_ptr, colorspace, name, temp, - "length does not match profile"); - - temp = (png_uint_32) (*(profile+8)); - if (temp > 3 && (profile_length & 3)) - return png_icc_profile_error(png_ptr, colorspace, name, profile_length, - "invalid length"); - - temp = png_get_uint_32(profile+128); /* tag count: 12 bytes/tag */ - if (temp > 357913930 || /* (2^32-4-132)/12: maximum possible tag count */ - profile_length < 132+12*temp) /* truncated tag table */ - return png_icc_profile_error(png_ptr, colorspace, name, temp, - "tag count too large"); - - /* The 'intent' must be valid or we can't store it, ICC limits the intent to - * 16 bits. - */ - temp = png_get_uint_32(profile+64); - if (temp >= 0xffff) /* The ICC limit */ - return png_icc_profile_error(png_ptr, colorspace, name, temp, - "invalid rendering intent"); - - /* This is just a warning because the profile may be valid in future - * versions. - */ - if (temp >= PNG_sRGB_INTENT_LAST) - (void)png_icc_profile_error(png_ptr, NULL, name, temp, - "intent outside defined range"); - - /* At this point the tag table can't be checked because it hasn't necessarily - * been loaded; however, various header fields can be checked. These checks - * are for values permitted by the PNG spec in an ICC profile; the PNG spec - * restricts the profiles that can be passed in an iCCP chunk (they must be - * appropriate to processing PNG data!) - */ - - /* Data checks (could be skipped). These checks must be independent of the - * version number; however, the version number doesn't accommodate changes in - * the header fields (just the known tags and the interpretation of the - * data.) - */ - temp = png_get_uint_32(profile+36); /* signature 'ascp' */ - if (temp != 0x61637370) - return png_icc_profile_error(png_ptr, colorspace, name, temp, - "invalid signature"); - - /* Currently the PCS illuminant/adopted white point (the computational - * white point) are required to be D50, - * however the profile contains a record of the illuminant so perhaps ICC - * expects to be able to change this in the future (despite the rationale in - * the introduction for using a fixed PCS adopted white.) Consequently the - * following is just a warning. - */ - if (memcmp(profile+68, D50_nCIEXYZ, 12) != 0) - (void)png_icc_profile_error(png_ptr, NULL, name, 0/*no tag value*/, - "PCS illuminant is not D50"); - - /* The PNG spec requires this: - * "If the iCCP chunk is present, the image samples conform to the colour - * space represented by the embedded ICC profile as defined by the - * International Color Consortium [ICC]. The colour space of the ICC profile - * shall be an RGB colour space for colour images (PNG colour types 2, 3, and - * 6), or a greyscale colour space for greyscale images (PNG colour types 0 - * and 4)." - * - * This checking code ensures the embedded profile (on either read or write) - * conforms to the specification requirements. Notice that an ICC 'gray' - * color-space profile contains the information to transform the monochrome - * data to XYZ or L*a*b (according to which PCS the profile uses) and this - * should be used in preference to the standard libpng K channel replication - * into R, G and B channels. - * - * Previously it was suggested that an RGB profile on grayscale data could be - * handled. However it it is clear that using an RGB profile in this context - * must be an error - there is no specification of what it means. Thus it is - * almost certainly more correct to ignore the profile. - */ - temp = png_get_uint_32(profile+16); /* data colour space field */ - switch (temp) - { - case 0x52474220: /* 'RGB ' */ - if ((color_type & PNG_COLOR_MASK_COLOR) == 0) - return png_icc_profile_error(png_ptr, colorspace, name, temp, - "RGB color space not permitted on grayscale PNG"); - break; - - case 0x47524159: /* 'GRAY' */ - if ((color_type & PNG_COLOR_MASK_COLOR) != 0) - return png_icc_profile_error(png_ptr, colorspace, name, temp, - "Gray color space not permitted on RGB PNG"); - break; - - default: - return png_icc_profile_error(png_ptr, colorspace, name, temp, - "invalid ICC profile color space"); - } - - /* It is up to the application to check that the profile class matches the - * application requirements; the spec provides no guidance, but it's pretty - * weird if the profile is not scanner ('scnr'), monitor ('mntr'), printer - * ('prtr') or 'spac' (for generic color spaces). Issue a warning in these - * cases. Issue an error for device link or abstract profiles - these don't - * contain the records necessary to transform the color-space to anything - * other than the target device (and not even that for an abstract profile). - * Profiles of these classes may not be embedded in images. - */ - temp = png_get_uint_32(profile+12); /* profile/device class */ - switch (temp) - { - case 0x73636e72: /* 'scnr' */ - case 0x6d6e7472: /* 'mntr' */ - case 0x70727472: /* 'prtr' */ - case 0x73706163: /* 'spac' */ - /* All supported */ - break; - - case 0x61627374: /* 'abst' */ - /* May not be embedded in an image */ - return png_icc_profile_error(png_ptr, colorspace, name, temp, - "invalid embedded Abstract ICC profile"); - - case 0x6c696e6b: /* 'link' */ - /* DeviceLink profiles cannot be interpreted in a non-device specific - * fashion, if an app uses the AToB0Tag in the profile the results are - * undefined unless the result is sent to the intended device, - * therefore a DeviceLink profile should not be found embedded in a - * PNG. - */ - return png_icc_profile_error(png_ptr, colorspace, name, temp, - "unexpected DeviceLink ICC profile class"); - - case 0x6e6d636c: /* 'nmcl' */ - /* A NamedColor profile is also device specific, however it doesn't - * contain an AToB0 tag that is open to misinterpretation. Almost - * certainly it will fail the tests below. - */ - (void)png_icc_profile_error(png_ptr, NULL, name, temp, - "unexpected NamedColor ICC profile class"); - break; - - default: - /* To allow for future enhancements to the profile accept unrecognized - * profile classes with a warning, these then hit the test below on the - * tag content to ensure they are backward compatible with one of the - * understood profiles. - */ - (void)png_icc_profile_error(png_ptr, NULL, name, temp, - "unrecognized ICC profile class"); - break; - } - - /* For any profile other than a device link one the PCS must be encoded - * either in XYZ or Lab. - */ - temp = png_get_uint_32(profile+20); - switch (temp) - { - case 0x58595a20: /* 'XYZ ' */ - case 0x4c616220: /* 'Lab ' */ - break; - - default: - return png_icc_profile_error(png_ptr, colorspace, name, temp, - "unexpected ICC PCS encoding"); - } - - return 1; -} - -int /* PRIVATE */ -png_icc_check_tag_table(png_const_structrp png_ptr, png_colorspacerp colorspace, - png_const_charp name, png_uint_32 profile_length, - png_const_bytep profile /* header plus whole tag table */) -{ - png_uint_32 tag_count = png_get_uint_32(profile+128); - png_uint_32 itag; - png_const_bytep tag = profile+132; /* The first tag */ - - /* First scan all the tags in the table and add bits to the icc_info value - * (temporarily in 'tags'). - */ - for (itag=0; itag < tag_count; ++itag, tag += 12) - { - png_uint_32 tag_id = png_get_uint_32(tag+0); - png_uint_32 tag_start = png_get_uint_32(tag+4); /* must be aligned */ - png_uint_32 tag_length = png_get_uint_32(tag+8);/* not padded */ - - /* The ICC specification does not exclude zero length tags, therefore the - * start might actually be anywhere if there is no data, but this would be - * a clear abuse of the intent of the standard so the start is checked for - * being in range. All defined tag types have an 8 byte header - a 4 byte - * type signature then 0. - */ - - /* This is a hard error; potentially it can cause read outside the - * profile. - */ - if (tag_start > profile_length || tag_length > profile_length - tag_start) - return png_icc_profile_error(png_ptr, colorspace, name, tag_id, - "ICC profile tag outside profile"); - - if ((tag_start & 3) != 0) - { - /* CNHP730S.icc shipped with Microsoft Windows 64 violates this; it is - * only a warning here because libpng does not care about the - * alignment. - */ - (void)png_icc_profile_error(png_ptr, NULL, name, tag_id, - "ICC profile tag start not a multiple of 4"); - } - } - - return 1; /* success, maybe with warnings */ -} - -#ifdef PNG_sRGB_SUPPORTED -#if PNG_sRGB_PROFILE_CHECKS >= 0 -/* Information about the known ICC sRGB profiles */ -static const struct -{ - png_uint_32 adler, crc, length; - png_uint_32 md5[4]; - png_byte have_md5; - png_byte is_broken; - png_uint_16 intent; - -# define PNG_MD5(a,b,c,d) { a, b, c, d }, (a!=0)||(b!=0)||(c!=0)||(d!=0) -# define PNG_ICC_CHECKSUM(adler, crc, md5, intent, broke, date, length, fname)\ - { adler, crc, length, md5, broke, intent }, - -} png_sRGB_checks[] = -{ - /* This data comes from contrib/tools/checksum-icc run on downloads of - * all four ICC sRGB profiles from www.color.org. - */ - /* adler32, crc32, MD5[4], intent, date, length, file-name */ - PNG_ICC_CHECKSUM(0x0a3fd9f6, 0x3b8772b9, - PNG_MD5(0x29f83dde, 0xaff255ae, 0x7842fae4, 0xca83390d), 0, 0, - "2009/03/27 21:36:31", 3048, "sRGB_IEC61966-2-1_black_scaled.icc") - - /* ICC sRGB v2 perceptual no black-compensation: */ - PNG_ICC_CHECKSUM(0x4909e5e1, 0x427ebb21, - PNG_MD5(0xc95bd637, 0xe95d8a3b, 0x0df38f99, 0xc1320389), 1, 0, - "2009/03/27 21:37:45", 3052, "sRGB_IEC61966-2-1_no_black_scaling.icc") - - PNG_ICC_CHECKSUM(0xfd2144a1, 0x306fd8ae, - PNG_MD5(0xfc663378, 0x37e2886b, 0xfd72e983, 0x8228f1b8), 0, 0, - "2009/08/10 17:28:01", 60988, "sRGB_v4_ICC_preference_displayclass.icc") - - /* ICC sRGB v4 perceptual */ - PNG_ICC_CHECKSUM(0x209c35d2, 0xbbef7812, - PNG_MD5(0x34562abf, 0x994ccd06, 0x6d2c5721, 0xd0d68c5d), 0, 0, - "2007/07/25 00:05:37", 60960, "sRGB_v4_ICC_preference.icc") - - /* The following profiles have no known MD5 checksum. If there is a match - * on the (empty) MD5 the other fields are used to attempt a match and - * a warning is produced. The first two of these profiles have a 'cprt' tag - * which suggests that they were also made by Hewlett Packard. - */ - PNG_ICC_CHECKSUM(0xa054d762, 0x5d5129ce, - PNG_MD5(0x00000000, 0x00000000, 0x00000000, 0x00000000), 1, 0, - "2004/07/21 18:57:42", 3024, "sRGB_IEC61966-2-1_noBPC.icc") - - /* This is a 'mntr' (display) profile with a mediaWhitePointTag that does not - * match the D50 PCS illuminant in the header (it is in fact the D65 values, - * so the white point is recorded as the un-adapted value.) The profiles - * below only differ in one byte - the intent - and are basically the same as - * the previous profile except for the mediaWhitePointTag error and a missing - * chromaticAdaptationTag. - */ - PNG_ICC_CHECKSUM(0xf784f3fb, 0x182ea552, - PNG_MD5(0x00000000, 0x00000000, 0x00000000, 0x00000000), 0, 1/*broken*/, - "1998/02/09 06:49:00", 3144, "HP-Microsoft sRGB v2 perceptual") - - PNG_ICC_CHECKSUM(0x0398f3fc, 0xf29e526d, - PNG_MD5(0x00000000, 0x00000000, 0x00000000, 0x00000000), 1, 1/*broken*/, - "1998/02/09 06:49:00", 3144, "HP-Microsoft sRGB v2 media-relative") -}; - -static int -png_compare_ICC_profile_with_sRGB(png_const_structrp png_ptr, - png_const_bytep profile, uLong adler) -{ - /* The quick check is to verify just the MD5 signature and trust the - * rest of the data. Because the profile has already been verified for - * correctness this is safe. png_colorspace_set_sRGB will check the 'intent' - * field too, so if the profile has been edited with an intent not defined - * by sRGB (but maybe defined by a later ICC specification) the read of - * the profile will fail at that point. - */ - - png_uint_32 length = 0; - png_uint_32 intent = 0x10000; /* invalid */ -#if PNG_sRGB_PROFILE_CHECKS > 1 - uLong crc = 0; /* the value for 0 length data */ -#endif - unsigned int i; - -#ifdef PNG_SET_OPTION_SUPPORTED - /* First see if PNG_SKIP_sRGB_CHECK_PROFILE has been set to "on" */ - if (((png_ptr->options >> PNG_SKIP_sRGB_CHECK_PROFILE) & 3) == - PNG_OPTION_ON) - return 0; -#endif - - for (i=0; i < (sizeof png_sRGB_checks) / (sizeof png_sRGB_checks[0]); ++i) - { - if (png_get_uint_32(profile+84) == png_sRGB_checks[i].md5[0] && - png_get_uint_32(profile+88) == png_sRGB_checks[i].md5[1] && - png_get_uint_32(profile+92) == png_sRGB_checks[i].md5[2] && - png_get_uint_32(profile+96) == png_sRGB_checks[i].md5[3]) - { - /* This may be one of the old HP profiles without an MD5, in that - * case we can only use the length and Adler32 (note that these - * are not used by default if there is an MD5!) - */ -# if PNG_sRGB_PROFILE_CHECKS == 0 - if (png_sRGB_checks[i].have_md5 != 0) - return 1+png_sRGB_checks[i].is_broken; -# endif - - /* Profile is unsigned or more checks have been configured in. */ - if (length == 0) - { - length = png_get_uint_32(profile); - intent = png_get_uint_32(profile+64); - } - - /* Length *and* intent must match */ - if (length == (png_uint_32) png_sRGB_checks[i].length && - intent == (png_uint_32) png_sRGB_checks[i].intent) - { - /* Now calculate the adler32 if not done already. */ - if (adler == 0) - { - adler = adler32(0, NULL, 0); - adler = adler32(adler, profile, length); - } - - if (adler == png_sRGB_checks[i].adler) - { - /* These basic checks suggest that the data has not been - * modified, but if the check level is more than 1 perform - * our own crc32 checksum on the data. - */ -# if PNG_sRGB_PROFILE_CHECKS > 1 - if (crc == 0) - { - crc = crc32(0, NULL, 0); - crc = crc32(crc, profile, length); - } - - /* So this check must pass for the 'return' below to happen. - */ - if (crc == png_sRGB_checks[i].crc) -# endif - { - if (png_sRGB_checks[i].is_broken != 0) - { - /* These profiles are known to have bad data that may cause - * problems if they are used, therefore attempt to - * discourage their use, skip the 'have_md5' warning below, - * which is made irrelevant by this error. - */ - png_chunk_report(png_ptr, "known incorrect sRGB profile", - PNG_CHUNK_ERROR); - } - - /* Warn that this being done; this isn't even an error since - * the profile is perfectly valid, but it would be nice if - * people used the up-to-date ones. - */ - else if (png_sRGB_checks[i].have_md5 == 0) - { - png_chunk_report(png_ptr, - "out-of-date sRGB profile with no signature", - PNG_CHUNK_WARNING); - } - - return 1+png_sRGB_checks[i].is_broken; - } - } - -# if PNG_sRGB_PROFILE_CHECKS > 0 - /* The signature matched, but the profile had been changed in some - * way. This probably indicates a data error or uninformed hacking. - * Fall through to "no match". - */ - png_chunk_report(png_ptr, - "Not recognizing known sRGB profile that has been edited", - PNG_CHUNK_WARNING); - break; -# endif - } - } - } - - return 0; /* no match */ -} - -void /* PRIVATE */ -png_icc_set_sRGB(png_const_structrp png_ptr, - png_colorspacerp colorspace, png_const_bytep profile, uLong adler) -{ - /* Is this profile one of the known ICC sRGB profiles? If it is, just set - * the sRGB information. - */ - if (png_compare_ICC_profile_with_sRGB(png_ptr, profile, adler) != 0) - (void)png_colorspace_set_sRGB(png_ptr, colorspace, - (int)/*already checked*/png_get_uint_32(profile+64)); -} -#endif /* PNG_sRGB_PROFILE_CHECKS >= 0 */ -#endif /* sRGB */ - -int /* PRIVATE */ -png_colorspace_set_ICC(png_const_structrp png_ptr, png_colorspacerp colorspace, - png_const_charp name, png_uint_32 profile_length, png_const_bytep profile, - int color_type) -{ - if ((colorspace->flags & PNG_COLORSPACE_INVALID) != 0) - return 0; - - if (icc_check_length(png_ptr, colorspace, name, profile_length) != 0 && - png_icc_check_header(png_ptr, colorspace, name, profile_length, profile, - color_type) != 0 && - png_icc_check_tag_table(png_ptr, colorspace, name, profile_length, - profile) != 0) - { -# if defined(PNG_sRGB_SUPPORTED) && PNG_sRGB_PROFILE_CHECKS >= 0 - /* If no sRGB support, don't try storing sRGB information */ - png_icc_set_sRGB(png_ptr, colorspace, profile, 0); -# endif - return 1; - } - - /* Failure case */ - return 0; -} -#endif /* iCCP */ - -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED -void /* PRIVATE */ -png_colorspace_set_rgb_coefficients(png_structrp png_ptr) -{ - /* Set the rgb_to_gray coefficients from the colorspace. */ - if (png_ptr->rgb_to_gray_coefficients_set == 0 && - (png_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_ENDPOINTS) != 0) - { - /* png_set_background has not been called, get the coefficients from the Y - * values of the colorspace colorants. - */ - png_fixed_point r = png_ptr->colorspace.end_points_XYZ.red_Y; - png_fixed_point g = png_ptr->colorspace.end_points_XYZ.green_Y; - png_fixed_point b = png_ptr->colorspace.end_points_XYZ.blue_Y; - png_fixed_point total = r+g+b; - - if (total > 0 && - r >= 0 && png_muldiv(&r, r, 32768, total) && r >= 0 && r <= 32768 && - g >= 0 && png_muldiv(&g, g, 32768, total) && g >= 0 && g <= 32768 && - b >= 0 && png_muldiv(&b, b, 32768, total) && b >= 0 && b <= 32768 && - r+g+b <= 32769) - { - /* We allow 0 coefficients here. r+g+b may be 32769 if two or - * all of the coefficients were rounded up. Handle this by - * reducing the *largest* coefficient by 1; this matches the - * approach used for the default coefficients in pngrtran.c - */ - int add = 0; - - if (r+g+b > 32768) - add = -1; - else if (r+g+b < 32768) - add = 1; - - if (add != 0) - { - if (g >= r && g >= b) - g += add; - else if (r >= g && r >= b) - r += add; - else - b += add; - } - - /* Check for an internal error. */ - if (r+g+b != 32768) - png_error(png_ptr, - "internal error handling cHRM coefficients"); - - else - { - png_ptr->rgb_to_gray_red_coeff = (png_uint_16)r; - png_ptr->rgb_to_gray_green_coeff = (png_uint_16)g; - } - } - - /* This is a png_error at present even though it could be ignored - - * it should never happen, but it is important that if it does, the - * bug is fixed. - */ - else - png_error(png_ptr, "internal error handling cHRM->XYZ"); - } -} -#endif /* READ_RGB_TO_GRAY */ - -#endif /* COLORSPACE */ - -void /* PRIVATE */ -png_check_IHDR(png_const_structrp png_ptr, - png_uint_32 width, png_uint_32 height, int bit_depth, - int color_type, int interlace_type, int compression_type, - int filter_type) -{ - int error = 0; - - /* Check for width and height valid values */ - if (width == 0) - { - png_warning(png_ptr, "Image width is zero in IHDR"); - error = 1; - } - - if (width > PNG_UINT_31_MAX) - { - png_warning(png_ptr, "Invalid image width in IHDR"); - error = 1; - } - - /* The bit mask on the first line below must be at least as big as a - * png_uint_32. "~7U" is not adequate on 16-bit systems because it will - * be an unsigned 16-bit value. Casting to (png_alloc_size_t) makes the - * type of the result at least as bit (in bits) as the RHS of the > operator - * which also avoids a common warning on 64-bit systems that the comparison - * of (png_uint_32) against the constant value on the RHS will always be - * false. - */ - if (((width + 7) & ~(png_alloc_size_t)7) > - (((PNG_SIZE_MAX - - 48 /* big_row_buf hack */ - - 1) /* filter byte */ - / 8) /* 8-byte RGBA pixels */ - - 1)) /* extra max_pixel_depth pad */ - { - /* The size of the row must be within the limits of this architecture. - * Because the read code can perform arbitrary transformations the - * maximum size is checked here. Because the code in png_read_start_row - * adds extra space "for safety's sake" in several places a conservative - * limit is used here. - * - * NOTE: it would be far better to check the size that is actually used, - * but the effect in the real world is minor and the changes are more - * extensive, therefore much more dangerous and much more difficult to - * write in a way that avoids compiler warnings. - */ - png_warning(png_ptr, "Image width is too large for this architecture"); - error = 1; - } - -#ifdef PNG_SET_USER_LIMITS_SUPPORTED - if (width > png_ptr->user_width_max) -#else - if (width > PNG_USER_WIDTH_MAX) -#endif - { - png_warning(png_ptr, "Image width exceeds user limit in IHDR"); - error = 1; - } - - if (height == 0) - { - png_warning(png_ptr, "Image height is zero in IHDR"); - error = 1; - } - - if (height > PNG_UINT_31_MAX) - { - png_warning(png_ptr, "Invalid image height in IHDR"); - error = 1; - } - -#ifdef PNG_SET_USER_LIMITS_SUPPORTED - if (height > png_ptr->user_height_max) -#else - if (height > PNG_USER_HEIGHT_MAX) -#endif - { - png_warning(png_ptr, "Image height exceeds user limit in IHDR"); - error = 1; - } - - /* Check other values */ - if (bit_depth != 1 && bit_depth != 2 && bit_depth != 4 && - bit_depth != 8 && bit_depth != 16) - { - png_warning(png_ptr, "Invalid bit depth in IHDR"); - error = 1; - } - - if (color_type < 0 || color_type == 1 || - color_type == 5 || color_type > 6) - { - png_warning(png_ptr, "Invalid color type in IHDR"); - error = 1; - } - - if (((color_type == PNG_COLOR_TYPE_PALETTE) && bit_depth > 8) || - ((color_type == PNG_COLOR_TYPE_RGB || - color_type == PNG_COLOR_TYPE_GRAY_ALPHA || - color_type == PNG_COLOR_TYPE_RGB_ALPHA) && bit_depth < 8)) - { - png_warning(png_ptr, "Invalid color type/bit depth combination in IHDR"); - error = 1; - } - - if (interlace_type >= PNG_INTERLACE_LAST) - { - png_warning(png_ptr, "Unknown interlace method in IHDR"); - error = 1; - } - - if (compression_type != PNG_COMPRESSION_TYPE_BASE) - { - png_warning(png_ptr, "Unknown compression method in IHDR"); - error = 1; - } - -#ifdef PNG_MNG_FEATURES_SUPPORTED - /* Accept filter_method 64 (intrapixel differencing) only if - * 1. Libpng was compiled with PNG_MNG_FEATURES_SUPPORTED and - * 2. Libpng did not read a PNG signature (this filter_method is only - * used in PNG datastreams that are embedded in MNG datastreams) and - * 3. The application called png_permit_mng_features with a mask that - * included PNG_FLAG_MNG_FILTER_64 and - * 4. The filter_method is 64 and - * 5. The color_type is RGB or RGBA - */ - if ((png_ptr->mode & PNG_HAVE_PNG_SIGNATURE) != 0 && - png_ptr->mng_features_permitted != 0) - png_warning(png_ptr, "MNG features are not allowed in a PNG datastream"); - - if (filter_type != PNG_FILTER_TYPE_BASE) - { - if (!((png_ptr->mng_features_permitted & PNG_FLAG_MNG_FILTER_64) != 0 && - (filter_type == PNG_INTRAPIXEL_DIFFERENCING) && - ((png_ptr->mode & PNG_HAVE_PNG_SIGNATURE) == 0) && - (color_type == PNG_COLOR_TYPE_RGB || - color_type == PNG_COLOR_TYPE_RGB_ALPHA))) - { - png_warning(png_ptr, "Unknown filter method in IHDR"); - error = 1; - } - - if ((png_ptr->mode & PNG_HAVE_PNG_SIGNATURE) != 0) - { - png_warning(png_ptr, "Invalid filter method in IHDR"); - error = 1; - } - } - -#else - if (filter_type != PNG_FILTER_TYPE_BASE) - { - png_warning(png_ptr, "Unknown filter method in IHDR"); - error = 1; - } -#endif - - if (error == 1) - png_error(png_ptr, "Invalid IHDR data"); -} - -#if defined(PNG_sCAL_SUPPORTED) || defined(PNG_pCAL_SUPPORTED) -/* ASCII to fp functions */ -/* Check an ASCII formatted floating point value, see the more detailed - * comments in pngpriv.h - */ -/* The following is used internally to preserve the sticky flags */ -#define png_fp_add(state, flags) ((state) |= (flags)) -#define png_fp_set(state, value) ((state) = (value) | ((state) & PNG_FP_STICKY)) - -int /* PRIVATE */ -png_check_fp_number(png_const_charp string, size_t size, int *statep, - size_t *whereami) -{ - int state = *statep; - size_t i = *whereami; - - while (i < size) - { - int type; - /* First find the type of the next character */ - switch (string[i]) - { - case 43: type = PNG_FP_SAW_SIGN; break; - case 45: type = PNG_FP_SAW_SIGN + PNG_FP_NEGATIVE; break; - case 46: type = PNG_FP_SAW_DOT; break; - case 48: type = PNG_FP_SAW_DIGIT; break; - case 49: case 50: case 51: case 52: - case 53: case 54: case 55: case 56: - case 57: type = PNG_FP_SAW_DIGIT + PNG_FP_NONZERO; break; - case 69: - case 101: type = PNG_FP_SAW_E; break; - default: goto PNG_FP_End; - } - - /* Now deal with this type according to the current - * state, the type is arranged to not overlap the - * bits of the PNG_FP_STATE. - */ - switch ((state & PNG_FP_STATE) + (type & PNG_FP_SAW_ANY)) - { - case PNG_FP_INTEGER + PNG_FP_SAW_SIGN: - if ((state & PNG_FP_SAW_ANY) != 0) - goto PNG_FP_End; /* not a part of the number */ - - png_fp_add(state, type); - break; - - case PNG_FP_INTEGER + PNG_FP_SAW_DOT: - /* Ok as trailer, ok as lead of fraction. */ - if ((state & PNG_FP_SAW_DOT) != 0) /* two dots */ - goto PNG_FP_End; - - else if ((state & PNG_FP_SAW_DIGIT) != 0) /* trailing dot? */ - png_fp_add(state, type); - - else - png_fp_set(state, PNG_FP_FRACTION | type); - - break; - - case PNG_FP_INTEGER + PNG_FP_SAW_DIGIT: - if ((state & PNG_FP_SAW_DOT) != 0) /* delayed fraction */ - png_fp_set(state, PNG_FP_FRACTION | PNG_FP_SAW_DOT); - - png_fp_add(state, type | PNG_FP_WAS_VALID); - - break; - - case PNG_FP_INTEGER + PNG_FP_SAW_E: - if ((state & PNG_FP_SAW_DIGIT) == 0) - goto PNG_FP_End; - - png_fp_set(state, PNG_FP_EXPONENT); - - break; - - /* case PNG_FP_FRACTION + PNG_FP_SAW_SIGN: - goto PNG_FP_End; ** no sign in fraction */ - - /* case PNG_FP_FRACTION + PNG_FP_SAW_DOT: - goto PNG_FP_End; ** Because SAW_DOT is always set */ - - case PNG_FP_FRACTION + PNG_FP_SAW_DIGIT: - png_fp_add(state, type | PNG_FP_WAS_VALID); - break; - - case PNG_FP_FRACTION + PNG_FP_SAW_E: - /* This is correct because the trailing '.' on an - * integer is handled above - so we can only get here - * with the sequence ".E" (with no preceding digits). - */ - if ((state & PNG_FP_SAW_DIGIT) == 0) - goto PNG_FP_End; - - png_fp_set(state, PNG_FP_EXPONENT); - - break; - - case PNG_FP_EXPONENT + PNG_FP_SAW_SIGN: - if ((state & PNG_FP_SAW_ANY) != 0) - goto PNG_FP_End; /* not a part of the number */ - - png_fp_add(state, PNG_FP_SAW_SIGN); - - break; - - /* case PNG_FP_EXPONENT + PNG_FP_SAW_DOT: - goto PNG_FP_End; */ - - case PNG_FP_EXPONENT + PNG_FP_SAW_DIGIT: - png_fp_add(state, PNG_FP_SAW_DIGIT | PNG_FP_WAS_VALID); - - break; - - /* case PNG_FP_EXPONEXT + PNG_FP_SAW_E: - goto PNG_FP_End; */ - - default: goto PNG_FP_End; /* I.e. break 2 */ - } - - /* The character seems ok, continue. */ - ++i; - } - -PNG_FP_End: - /* Here at the end, update the state and return the correct - * return code. - */ - *statep = state; - *whereami = i; - - return (state & PNG_FP_SAW_DIGIT) != 0; -} - - -/* The same but for a complete string. */ -int -png_check_fp_string(png_const_charp string, size_t size) -{ - int state=0; - size_t char_index=0; - - if (png_check_fp_number(string, size, &state, &char_index) != 0 && - (char_index == size || string[char_index] == 0)) - return state /* must be non-zero - see above */; - - return 0; /* i.e. fail */ -} -#endif /* pCAL || sCAL */ - -#ifdef PNG_sCAL_SUPPORTED -# ifdef PNG_FLOATING_POINT_SUPPORTED -/* Utility used below - a simple accurate power of ten from an integral - * exponent. - */ -static double -png_pow10(int power) -{ - int recip = 0; - double d = 1; - - /* Handle negative exponent with a reciprocal at the end because - * 10 is exact whereas .1 is inexact in base 2 - */ - if (power < 0) - { - if (power < DBL_MIN_10_EXP) return 0; - recip = 1; power = -power; - } - - if (power > 0) - { - /* Decompose power bitwise. */ - double mult = 10; - do - { - if (power & 1) d *= mult; - mult *= mult; - power >>= 1; - } - while (power > 0); - - if (recip != 0) d = 1/d; - } - /* else power is 0 and d is 1 */ - - return d; -} - -/* Function to format a floating point value in ASCII with a given - * precision. - */ -void /* PRIVATE */ -png_ascii_from_fp(png_const_structrp png_ptr, png_charp ascii, size_t size, - double fp, unsigned int precision) -{ - /* We use standard functions from math.h, but not printf because - * that would require stdio. The caller must supply a buffer of - * sufficient size or we will png_error. The tests on size and - * the space in ascii[] consumed are indicated below. - */ - if (precision < 1) - precision = DBL_DIG; - - /* Enforce the limit of the implementation precision too. */ - if (precision > DBL_DIG+1) - precision = DBL_DIG+1; - - /* Basic sanity checks */ - if (size >= precision+5) /* See the requirements below. */ - { - if (fp < 0) - { - fp = -fp; - *ascii++ = 45; /* '-' PLUS 1 TOTAL 1 */ - --size; - } - - if (fp >= DBL_MIN && fp <= DBL_MAX) - { - int exp_b10; /* A base 10 exponent */ - double base; /* 10^exp_b10 */ - - /* First extract a base 10 exponent of the number, - * the calculation below rounds down when converting - * from base 2 to base 10 (multiply by log10(2) - - * 0.3010, but 77/256 is 0.3008, so exp_b10 needs to - * be increased. Note that the arithmetic shift - * performs a floor() unlike C arithmetic - using a - * C multiply would break the following for negative - * exponents. - */ - (void)frexp(fp, &exp_b10); /* exponent to base 2 */ - - exp_b10 = (exp_b10 * 77) >> 8; /* <= exponent to base 10 */ - - /* Avoid underflow here. */ - base = png_pow10(exp_b10); /* May underflow */ - - while (base < DBL_MIN || base < fp) - { - /* And this may overflow. */ - double test = png_pow10(exp_b10+1); - - if (test <= DBL_MAX) - { - ++exp_b10; base = test; - } - - else - break; - } - - /* Normalize fp and correct exp_b10, after this fp is in the - * range [.1,1) and exp_b10 is both the exponent and the digit - * *before* which the decimal point should be inserted - * (starting with 0 for the first digit). Note that this - * works even if 10^exp_b10 is out of range because of the - * test on DBL_MAX above. - */ - fp /= base; - while (fp >= 1) - { - fp /= 10; ++exp_b10; - } - - /* Because of the code above fp may, at this point, be - * less than .1, this is ok because the code below can - * handle the leading zeros this generates, so no attempt - * is made to correct that here. - */ - - { - unsigned int czero, clead, cdigits; - char exponent[10]; - - /* Allow up to two leading zeros - this will not lengthen - * the number compared to using E-n. - */ - if (exp_b10 < 0 && exp_b10 > -3) /* PLUS 3 TOTAL 4 */ - { - czero = 0U-exp_b10; /* PLUS 2 digits: TOTAL 3 */ - exp_b10 = 0; /* Dot added below before first output. */ - } - else - czero = 0; /* No zeros to add */ - - /* Generate the digit list, stripping trailing zeros and - * inserting a '.' before a digit if the exponent is 0. - */ - clead = czero; /* Count of leading zeros */ - cdigits = 0; /* Count of digits in list. */ - - do - { - double d; - - fp *= 10; - /* Use modf here, not floor and subtract, so that - * the separation is done in one step. At the end - * of the loop don't break the number into parts so - * that the final digit is rounded. - */ - if (cdigits+czero+1 < precision+clead) - fp = modf(fp, &d); - - else - { - d = floor(fp + .5); - - if (d > 9) - { - /* Rounding up to 10, handle that here. */ - if (czero > 0) - { - --czero; d = 1; - if (cdigits == 0) --clead; - } - else - { - while (cdigits > 0 && d > 9) - { - int ch = *--ascii; - - if (exp_b10 != (-1)) - ++exp_b10; - - else if (ch == 46) - { - ch = *--ascii; ++size; - /* Advance exp_b10 to '1', so that the - * decimal point happens after the - * previous digit. - */ - exp_b10 = 1; - } - - --cdigits; - d = ch - 47; /* I.e. 1+(ch-48) */ - } - - /* Did we reach the beginning? If so adjust the - * exponent but take into account the leading - * decimal point. - */ - if (d > 9) /* cdigits == 0 */ - { - if (exp_b10 == (-1)) - { - /* Leading decimal point (plus zeros?), if - * we lose the decimal point here it must - * be reentered below. - */ - int ch = *--ascii; - - if (ch == 46) - { - ++size; exp_b10 = 1; - } - - /* Else lost a leading zero, so 'exp_b10' is - * still ok at (-1) - */ - } - else - ++exp_b10; - - /* In all cases we output a '1' */ - d = 1; - } - } - } - fp = 0; /* Guarantees termination below. */ - } - - if (d == 0) - { - ++czero; - if (cdigits == 0) ++clead; - } - else - { - /* Included embedded zeros in the digit count. */ - cdigits += czero - clead; - clead = 0; - - while (czero > 0) - { - /* exp_b10 == (-1) means we just output the decimal - * place - after the DP don't adjust 'exp_b10' any - * more! - */ - if (exp_b10 != (-1)) - { - if (exp_b10 == 0) - { - *ascii++ = 46; --size; - } - /* PLUS 1: TOTAL 4 */ - --exp_b10; - } - *ascii++ = 48; --czero; - } - - if (exp_b10 != (-1)) - { - if (exp_b10 == 0) - { - *ascii++ = 46; --size; /* counted above */ - } - - --exp_b10; - } - *ascii++ = (char)(48 + (int)d); ++cdigits; - } - } - while (cdigits+czero < precision+clead && fp > DBL_MIN); - - /* The total output count (max) is now 4+precision */ - - /* Check for an exponent, if we don't need one we are - * done and just need to terminate the string. At this - * point, exp_b10==(-1) is effectively a flag: it got - * to '-1' because of the decrement, after outputting - * the decimal point above. (The exponent required is - * *not* -1.) - */ - if (exp_b10 >= (-1) && exp_b10 <= 2) - { - /* The following only happens if we didn't output the - * leading zeros above for negative exponent, so this - * doesn't add to the digit requirement. Note that the - * two zeros here can only be output if the two leading - * zeros were *not* output, so this doesn't increase - * the output count. - */ - while (exp_b10-- > 0) *ascii++ = 48; - - *ascii = 0; - - /* Total buffer requirement (including the '\0') is - * 5+precision - see check at the start. - */ - return; - } - - /* Here if an exponent is required, adjust size for - * the digits we output but did not count. The total - * digit output here so far is at most 1+precision - no - * decimal point and no leading or trailing zeros have - * been output. - */ - size -= cdigits; - - *ascii++ = 69; --size; /* 'E': PLUS 1 TOTAL 2+precision */ - - /* The following use of an unsigned temporary avoids ambiguities in - * the signed arithmetic on exp_b10 and permits GCC at least to do - * better optimization. - */ - { - unsigned int uexp_b10; - - if (exp_b10 < 0) - { - *ascii++ = 45; --size; /* '-': PLUS 1 TOTAL 3+precision */ - uexp_b10 = 0U-exp_b10; - } - - else - uexp_b10 = 0U+exp_b10; - - cdigits = 0; - - while (uexp_b10 > 0) - { - exponent[cdigits++] = (char)(48 + uexp_b10 % 10); - uexp_b10 /= 10; - } - } - - /* Need another size check here for the exponent digits, so - * this need not be considered above. - */ - if (size > cdigits) - { - while (cdigits > 0) *ascii++ = exponent[--cdigits]; - - *ascii = 0; - - return; - } - } - } - else if (!(fp >= DBL_MIN)) - { - *ascii++ = 48; /* '0' */ - *ascii = 0; - return; - } - else - { - *ascii++ = 105; /* 'i' */ - *ascii++ = 110; /* 'n' */ - *ascii++ = 102; /* 'f' */ - *ascii = 0; - return; - } - } - - /* Here on buffer too small. */ - png_error(png_ptr, "ASCII conversion buffer too small"); -} -# endif /* FLOATING_POINT */ - -# ifdef PNG_FIXED_POINT_SUPPORTED -/* Function to format a fixed point value in ASCII. - */ -void /* PRIVATE */ -png_ascii_from_fixed(png_const_structrp png_ptr, png_charp ascii, - size_t size, png_fixed_point fp) -{ - /* Require space for 10 decimal digits, a decimal point, a minus sign and a - * trailing \0, 13 characters: - */ - if (size > 12) - { - png_uint_32 num; - - /* Avoid overflow here on the minimum integer. */ - if (fp < 0) - { - *ascii++ = 45; num = (png_uint_32)(-fp); - } - else - num = (png_uint_32)fp; - - if (num <= 0x80000000) /* else overflowed */ - { - unsigned int ndigits = 0, first = 16 /* flag value */; - char digits[10] = {0}; - - while (num) - { - /* Split the low digit off num: */ - unsigned int tmp = num/10; - num -= tmp*10; - digits[ndigits++] = (char)(48 + num); - /* Record the first non-zero digit, note that this is a number - * starting at 1, it's not actually the array index. - */ - if (first == 16 && num > 0) - first = ndigits; - num = tmp; - } - - if (ndigits > 0) - { - while (ndigits > 5) *ascii++ = digits[--ndigits]; - /* The remaining digits are fractional digits, ndigits is '5' or - * smaller at this point. It is certainly not zero. Check for a - * non-zero fractional digit: - */ - if (first <= 5) - { - unsigned int i; - *ascii++ = 46; /* decimal point */ - /* ndigits may be <5 for small numbers, output leading zeros - * then ndigits digits to first: - */ - i = 5; - while (ndigits < i) - { - *ascii++ = 48; --i; - } - while (ndigits >= first) *ascii++ = digits[--ndigits]; - /* Don't output the trailing zeros! */ - } - } - else - *ascii++ = 48; - - /* And null terminate the string: */ - *ascii = 0; - return; - } - } - - /* Here on buffer too small. */ - png_error(png_ptr, "ASCII conversion buffer too small"); -} -# endif /* FIXED_POINT */ -#endif /* SCAL */ - -#if defined(PNG_FLOATING_POINT_SUPPORTED) && \ - !defined(PNG_FIXED_POINT_MACRO_SUPPORTED) && \ - (defined(PNG_gAMA_SUPPORTED) || defined(PNG_cHRM_SUPPORTED) || \ - defined(PNG_sCAL_SUPPORTED) || defined(PNG_READ_BACKGROUND_SUPPORTED) || \ - defined(PNG_READ_RGB_TO_GRAY_SUPPORTED)) || \ - (defined(PNG_sCAL_SUPPORTED) && \ - defined(PNG_FLOATING_ARITHMETIC_SUPPORTED)) -png_fixed_point -png_fixed(png_const_structrp png_ptr, double fp, png_const_charp text) -{ - double r = floor(100000 * fp + .5); - - if (r > 2147483647. || r < -2147483648.) - png_fixed_error(png_ptr, text); - -# ifndef PNG_ERROR_TEXT_SUPPORTED - PNG_UNUSED(text) -# endif - - return (png_fixed_point)r; -} -#endif - -#if defined(PNG_GAMMA_SUPPORTED) || defined(PNG_COLORSPACE_SUPPORTED) ||\ - defined(PNG_INCH_CONVERSIONS_SUPPORTED) || defined(PNG_READ_pHYs_SUPPORTED) -/* muldiv functions */ -/* This API takes signed arguments and rounds the result to the nearest - * integer (or, for a fixed point number - the standard argument - to - * the nearest .00001). Overflow and divide by zero are signalled in - * the result, a boolean - true on success, false on overflow. - */ -int -png_muldiv(png_fixed_point_p res, png_fixed_point a, png_int_32 times, - png_int_32 divisor) -{ - /* Return a * times / divisor, rounded. */ - if (divisor != 0) - { - if (a == 0 || times == 0) - { - *res = 0; - return 1; - } - else - { -#ifdef PNG_FLOATING_ARITHMETIC_SUPPORTED - double r = a; - r *= times; - r /= divisor; - r = floor(r+.5); - - /* A png_fixed_point is a 32-bit integer. */ - if (r <= 2147483647. && r >= -2147483648.) - { - *res = (png_fixed_point)r; - return 1; - } -#else - int negative = 0; - png_uint_32 A, T, D; - png_uint_32 s16, s32, s00; - - if (a < 0) - negative = 1, A = -a; - else - A = a; - - if (times < 0) - negative = !negative, T = -times; - else - T = times; - - if (divisor < 0) - negative = !negative, D = -divisor; - else - D = divisor; - - /* Following can't overflow because the arguments only - * have 31 bits each, however the result may be 32 bits. - */ - s16 = (A >> 16) * (T & 0xffff) + - (A & 0xffff) * (T >> 16); - /* Can't overflow because the a*times bit is only 30 - * bits at most. - */ - s32 = (A >> 16) * (T >> 16) + (s16 >> 16); - s00 = (A & 0xffff) * (T & 0xffff); - - s16 = (s16 & 0xffff) << 16; - s00 += s16; - - if (s00 < s16) - ++s32; /* carry */ - - if (s32 < D) /* else overflow */ - { - /* s32.s00 is now the 64-bit product, do a standard - * division, we know that s32 < D, so the maximum - * required shift is 31. - */ - int bitshift = 32; - png_fixed_point result = 0; /* NOTE: signed */ - - while (--bitshift >= 0) - { - png_uint_32 d32, d00; - - if (bitshift > 0) - d32 = D >> (32-bitshift), d00 = D << bitshift; - - else - d32 = 0, d00 = D; - - if (s32 > d32) - { - if (s00 < d00) --s32; /* carry */ - s32 -= d32, s00 -= d00, result += 1<= d00) - s32 = 0, s00 -= d00, result += 1<= (D >> 1)) - ++result; - - if (negative != 0) - result = -result; - - /* Check for overflow. */ - if ((negative != 0 && result <= 0) || - (negative == 0 && result >= 0)) - { - *res = result; - return 1; - } - } -#endif - } - } - - return 0; -} -#endif /* READ_GAMMA || INCH_CONVERSIONS */ - -#if defined(PNG_READ_GAMMA_SUPPORTED) || defined(PNG_INCH_CONVERSIONS_SUPPORTED) -/* The following is for when the caller doesn't much care about the - * result. - */ -png_fixed_point -png_muldiv_warn(png_const_structrp png_ptr, png_fixed_point a, png_int_32 times, - png_int_32 divisor) -{ - png_fixed_point result; - - if (png_muldiv(&result, a, times, divisor) != 0) - return result; - - png_warning(png_ptr, "fixed point overflow ignored"); - return 0; -} -#endif - -#ifdef PNG_GAMMA_SUPPORTED /* more fixed point functions for gamma */ -/* Calculate a reciprocal, return 0 on div-by-zero or overflow. */ -png_fixed_point -png_reciprocal(png_fixed_point a) -{ -#ifdef PNG_FLOATING_ARITHMETIC_SUPPORTED - double r = floor(1E10/a+.5); - - if (r <= 2147483647. && r >= -2147483648.) - return (png_fixed_point)r; -#else - png_fixed_point res; - - if (png_muldiv(&res, 100000, 100000, a) != 0) - return res; -#endif - - return 0; /* error/overflow */ -} - -/* This is the shared test on whether a gamma value is 'significant' - whether - * it is worth doing gamma correction. - */ -int /* PRIVATE */ -png_gamma_significant(png_fixed_point gamma_val) -{ - return gamma_val < PNG_FP_1 - PNG_GAMMA_THRESHOLD_FIXED || - gamma_val > PNG_FP_1 + PNG_GAMMA_THRESHOLD_FIXED; -} -#endif - -#ifdef PNG_READ_GAMMA_SUPPORTED -#ifdef PNG_16BIT_SUPPORTED -/* A local convenience routine. */ -static png_fixed_point -png_product2(png_fixed_point a, png_fixed_point b) -{ - /* The required result is 1/a * 1/b; the following preserves accuracy. */ -#ifdef PNG_FLOATING_ARITHMETIC_SUPPORTED - double r = a * 1E-5; - r *= b; - r = floor(r+.5); - - if (r <= 2147483647. && r >= -2147483648.) - return (png_fixed_point)r; -#else - png_fixed_point res; - - if (png_muldiv(&res, a, b, 100000) != 0) - return res; -#endif - - return 0; /* overflow */ -} -#endif /* 16BIT */ - -/* The inverse of the above. */ -png_fixed_point -png_reciprocal2(png_fixed_point a, png_fixed_point b) -{ - /* The required result is 1/a * 1/b; the following preserves accuracy. */ -#ifdef PNG_FLOATING_ARITHMETIC_SUPPORTED - if (a != 0 && b != 0) - { - double r = 1E15/a; - r /= b; - r = floor(r+.5); - - if (r <= 2147483647. && r >= -2147483648.) - return (png_fixed_point)r; - } -#else - /* This may overflow because the range of png_fixed_point isn't symmetric, - * but this API is only used for the product of file and screen gamma so it - * doesn't matter that the smallest number it can produce is 1/21474, not - * 1/100000 - */ - png_fixed_point res = png_product2(a, b); - - if (res != 0) - return png_reciprocal(res); -#endif - - return 0; /* overflow */ -} -#endif /* READ_GAMMA */ - -#ifdef PNG_READ_GAMMA_SUPPORTED /* gamma table code */ -#ifndef PNG_FLOATING_ARITHMETIC_SUPPORTED -/* Fixed point gamma. - * - * The code to calculate the tables used below can be found in the shell script - * contrib/tools/intgamma.sh - * - * To calculate gamma this code implements fast log() and exp() calls using only - * fixed point arithmetic. This code has sufficient precision for either 8-bit - * or 16-bit sample values. - * - * The tables used here were calculated using simple 'bc' programs, but C double - * precision floating point arithmetic would work fine. - * - * 8-bit log table - * This is a table of -log(value/255)/log(2) for 'value' in the range 128 to - * 255, so it's the base 2 logarithm of a normalized 8-bit floating point - * mantissa. The numbers are 32-bit fractions. - */ -static const png_uint_32 -png_8bit_l2[128] = -{ - 4270715492U, 4222494797U, 4174646467U, 4127164793U, 4080044201U, 4033279239U, - 3986864580U, 3940795015U, 3895065449U, 3849670902U, 3804606499U, 3759867474U, - 3715449162U, 3671346997U, 3627556511U, 3584073329U, 3540893168U, 3498011834U, - 3455425220U, 3413129301U, 3371120137U, 3329393864U, 3287946700U, 3246774933U, - 3205874930U, 3165243125U, 3124876025U, 3084770202U, 3044922296U, 3005329011U, - 2965987113U, 2926893432U, 2888044853U, 2849438323U, 2811070844U, 2772939474U, - 2735041326U, 2697373562U, 2659933400U, 2622718104U, 2585724991U, 2548951424U, - 2512394810U, 2476052606U, 2439922311U, 2404001468U, 2368287663U, 2332778523U, - 2297471715U, 2262364947U, 2227455964U, 2192742551U, 2158222529U, 2123893754U, - 2089754119U, 2055801552U, 2022034013U, 1988449497U, 1955046031U, 1921821672U, - 1888774511U, 1855902668U, 1823204291U, 1790677560U, 1758320682U, 1726131893U, - 1694109454U, 1662251657U, 1630556815U, 1599023271U, 1567649391U, 1536433567U, - 1505374214U, 1474469770U, 1443718700U, 1413119487U, 1382670639U, 1352370686U, - 1322218179U, 1292211689U, 1262349810U, 1232631153U, 1203054352U, 1173618059U, - 1144320946U, 1115161701U, 1086139034U, 1057251672U, 1028498358U, 999877854U, - 971388940U, 943030410U, 914801076U, 886699767U, 858725327U, 830876614U, - 803152505U, 775551890U, 748073672U, 720716771U, 693480120U, 666362667U, - 639363374U, 612481215U, 585715177U, 559064263U, 532527486U, 506103872U, - 479792461U, 453592303U, 427502463U, 401522014U, 375650043U, 349885648U, - 324227938U, 298676034U, 273229066U, 247886176U, 222646516U, 197509248U, - 172473545U, 147538590U, 122703574U, 97967701U, 73330182U, 48790236U, - 24347096U, 0U - -#if 0 - /* The following are the values for 16-bit tables - these work fine for the - * 8-bit conversions but produce very slightly larger errors in the 16-bit - * log (about 1.2 as opposed to 0.7 absolute error in the final value). To - * use these all the shifts below must be adjusted appropriately. - */ - 65166, 64430, 63700, 62976, 62257, 61543, 60835, 60132, 59434, 58741, 58054, - 57371, 56693, 56020, 55352, 54689, 54030, 53375, 52726, 52080, 51439, 50803, - 50170, 49542, 48918, 48298, 47682, 47070, 46462, 45858, 45257, 44661, 44068, - 43479, 42894, 42312, 41733, 41159, 40587, 40020, 39455, 38894, 38336, 37782, - 37230, 36682, 36137, 35595, 35057, 34521, 33988, 33459, 32932, 32408, 31887, - 31369, 30854, 30341, 29832, 29325, 28820, 28319, 27820, 27324, 26830, 26339, - 25850, 25364, 24880, 24399, 23920, 23444, 22970, 22499, 22029, 21562, 21098, - 20636, 20175, 19718, 19262, 18808, 18357, 17908, 17461, 17016, 16573, 16132, - 15694, 15257, 14822, 14390, 13959, 13530, 13103, 12678, 12255, 11834, 11415, - 10997, 10582, 10168, 9756, 9346, 8937, 8531, 8126, 7723, 7321, 6921, 6523, - 6127, 5732, 5339, 4947, 4557, 4169, 3782, 3397, 3014, 2632, 2251, 1872, 1495, - 1119, 744, 372 -#endif -}; - -static png_int_32 -png_log8bit(unsigned int x) -{ - unsigned int lg2 = 0; - /* Each time 'x' is multiplied by 2, 1 must be subtracted off the final log, - * because the log is actually negate that means adding 1. The final - * returned value thus has the range 0 (for 255 input) to 7.994 (for 1 - * input), return -1 for the overflow (log 0) case, - so the result is - * always at most 19 bits. - */ - if ((x &= 0xff) == 0) - return -1; - - if ((x & 0xf0) == 0) - lg2 = 4, x <<= 4; - - if ((x & 0xc0) == 0) - lg2 += 2, x <<= 2; - - if ((x & 0x80) == 0) - lg2 += 1, x <<= 1; - - /* result is at most 19 bits, so this cast is safe: */ - return (png_int_32)((lg2 << 16) + ((png_8bit_l2[x-128]+32768)>>16)); -} - -/* The above gives exact (to 16 binary places) log2 values for 8-bit images, - * for 16-bit images we use the most significant 8 bits of the 16-bit value to - * get an approximation then multiply the approximation by a correction factor - * determined by the remaining up to 8 bits. This requires an additional step - * in the 16-bit case. - * - * We want log2(value/65535), we have log2(v'/255), where: - * - * value = v' * 256 + v'' - * = v' * f - * - * So f is value/v', which is equal to (256+v''/v') since v' is in the range 128 - * to 255 and v'' is in the range 0 to 255 f will be in the range 256 to less - * than 258. The final factor also needs to correct for the fact that our 8-bit - * value is scaled by 255, whereas the 16-bit values must be scaled by 65535. - * - * This gives a final formula using a calculated value 'x' which is value/v' and - * scaling by 65536 to match the above table: - * - * log2(x/257) * 65536 - * - * Since these numbers are so close to '1' we can use simple linear - * interpolation between the two end values 256/257 (result -368.61) and 258/257 - * (result 367.179). The values used below are scaled by a further 64 to give - * 16-bit precision in the interpolation: - * - * Start (256): -23591 - * Zero (257): 0 - * End (258): 23499 - */ -#ifdef PNG_16BIT_SUPPORTED -static png_int_32 -png_log16bit(png_uint_32 x) -{ - unsigned int lg2 = 0; - - /* As above, but now the input has 16 bits. */ - if ((x &= 0xffff) == 0) - return -1; - - if ((x & 0xff00) == 0) - lg2 = 8, x <<= 8; - - if ((x & 0xf000) == 0) - lg2 += 4, x <<= 4; - - if ((x & 0xc000) == 0) - lg2 += 2, x <<= 2; - - if ((x & 0x8000) == 0) - lg2 += 1, x <<= 1; - - /* Calculate the base logarithm from the top 8 bits as a 28-bit fractional - * value. - */ - lg2 <<= 28; - lg2 += (png_8bit_l2[(x>>8)-128]+8) >> 4; - - /* Now we need to interpolate the factor, this requires a division by the top - * 8 bits. Do this with maximum precision. - */ - x = ((x << 16) + (x >> 9)) / (x >> 8); - - /* Since we divided by the top 8 bits of 'x' there will be a '1' at 1<<24, - * the value at 1<<16 (ignoring this) will be 0 or 1; this gives us exactly - * 16 bits to interpolate to get the low bits of the result. Round the - * answer. Note that the end point values are scaled by 64 to retain overall - * precision and that 'lg2' is current scaled by an extra 12 bits, so adjust - * the overall scaling by 6-12. Round at every step. - */ - x -= 1U << 24; - - if (x <= 65536U) /* <= '257' */ - lg2 += ((23591U * (65536U-x)) + (1U << (16+6-12-1))) >> (16+6-12); - - else - lg2 -= ((23499U * (x-65536U)) + (1U << (16+6-12-1))) >> (16+6-12); - - /* Safe, because the result can't have more than 20 bits: */ - return (png_int_32)((lg2 + 2048) >> 12); -} -#endif /* 16BIT */ - -/* The 'exp()' case must invert the above, taking a 20-bit fixed point - * logarithmic value and returning a 16 or 8-bit number as appropriate. In - * each case only the low 16 bits are relevant - the fraction - since the - * integer bits (the top 4) simply determine a shift. - * - * The worst case is the 16-bit distinction between 65535 and 65534. This - * requires perhaps spurious accuracy in the decoding of the logarithm to - * distinguish log2(65535/65534.5) - 10^-5 or 17 bits. There is little chance - * of getting this accuracy in practice. - * - * To deal with this the following exp() function works out the exponent of the - * fractional part of the logarithm by using an accurate 32-bit value from the - * top four fractional bits then multiplying in the remaining bits. - */ -static const png_uint_32 -png_32bit_exp[16] = -{ - /* NOTE: the first entry is deliberately set to the maximum 32-bit value. */ - 4294967295U, 4112874773U, 3938502376U, 3771522796U, 3611622603U, 3458501653U, - 3311872529U, 3171459999U, 3037000500U, 2908241642U, 2784941738U, 2666869345U, - 2553802834U, 2445529972U, 2341847524U, 2242560872U -}; - -/* Adjustment table; provided to explain the numbers in the code below. */ -#if 0 -for (i=11;i>=0;--i){ print i, " ", (1 - e(-(2^i)/65536*l(2))) * 2^(32-i), "\n"} - 11 44937.64284865548751208448 - 10 45180.98734845585101160448 - 9 45303.31936980687359311872 - 8 45364.65110595323018870784 - 7 45395.35850361789624614912 - 6 45410.72259715102037508096 - 5 45418.40724413220722311168 - 4 45422.25021786898173001728 - 3 45424.17186732298419044352 - 2 45425.13273269940811464704 - 1 45425.61317555035558641664 - 0 45425.85339951654943850496 -#endif - -static png_uint_32 -png_exp(png_fixed_point x) -{ - if (x > 0 && x <= 0xfffff) /* Else overflow or zero (underflow) */ - { - /* Obtain a 4-bit approximation */ - png_uint_32 e = png_32bit_exp[(x >> 12) & 0x0f]; - - /* Incorporate the low 12 bits - these decrease the returned value by - * multiplying by a number less than 1 if the bit is set. The multiplier - * is determined by the above table and the shift. Notice that the values - * converge on 45426 and this is used to allow linear interpolation of the - * low bits. - */ - if (x & 0x800) - e -= (((e >> 16) * 44938U) + 16U) >> 5; - - if (x & 0x400) - e -= (((e >> 16) * 45181U) + 32U) >> 6; - - if (x & 0x200) - e -= (((e >> 16) * 45303U) + 64U) >> 7; - - if (x & 0x100) - e -= (((e >> 16) * 45365U) + 128U) >> 8; - - if (x & 0x080) - e -= (((e >> 16) * 45395U) + 256U) >> 9; - - if (x & 0x040) - e -= (((e >> 16) * 45410U) + 512U) >> 10; - - /* And handle the low 6 bits in a single block. */ - e -= (((e >> 16) * 355U * (x & 0x3fU)) + 256U) >> 9; - - /* Handle the upper bits of x. */ - e >>= x >> 16; - return e; - } - - /* Check for overflow */ - if (x <= 0) - return png_32bit_exp[0]; - - /* Else underflow */ - return 0; -} - -static png_byte -png_exp8bit(png_fixed_point lg2) -{ - /* Get a 32-bit value: */ - png_uint_32 x = png_exp(lg2); - - /* Convert the 32-bit value to 0..255 by multiplying by 256-1. Note that the - * second, rounding, step can't overflow because of the first, subtraction, - * step. - */ - x -= x >> 8; - return (png_byte)(((x + 0x7fffffU) >> 24) & 0xff); -} - -#ifdef PNG_16BIT_SUPPORTED -static png_uint_16 -png_exp16bit(png_fixed_point lg2) -{ - /* Get a 32-bit value: */ - png_uint_32 x = png_exp(lg2); - - /* Convert the 32-bit value to 0..65535 by multiplying by 65536-1: */ - x -= x >> 16; - return (png_uint_16)((x + 32767U) >> 16); -} -#endif /* 16BIT */ -#endif /* FLOATING_ARITHMETIC */ - -png_byte -png_gamma_8bit_correct(unsigned int value, png_fixed_point gamma_val) -{ - if (value > 0 && value < 255) - { -# ifdef PNG_FLOATING_ARITHMETIC_SUPPORTED - /* 'value' is unsigned, ANSI-C90 requires the compiler to correctly - * convert this to a floating point value. This includes values that - * would overflow if 'value' were to be converted to 'int'. - * - * Apparently GCC, however, does an intermediate conversion to (int) - * on some (ARM) but not all (x86) platforms, possibly because of - * hardware FP limitations. (E.g. if the hardware conversion always - * assumes the integer register contains a signed value.) This results - * in ANSI-C undefined behavior for large values. - * - * Other implementations on the same machine might actually be ANSI-C90 - * conformant and therefore compile spurious extra code for the large - * values. - * - * We can be reasonably sure that an unsigned to float conversion - * won't be faster than an int to float one. Therefore this code - * assumes responsibility for the undefined behavior, which it knows - * can't happen because of the check above. - * - * Note the argument to this routine is an (unsigned int) because, on - * 16-bit platforms, it is assigned a value which might be out of - * range for an (int); that would result in undefined behavior in the - * caller if the *argument* ('value') were to be declared (int). - */ - double r = floor(255*pow((int)/*SAFE*/value/255.,gamma_val*.00001)+.5); - return (png_byte)r; -# else - png_int_32 lg2 = png_log8bit(value); - png_fixed_point res; - - if (png_muldiv(&res, gamma_val, lg2, PNG_FP_1) != 0) - return png_exp8bit(res); - - /* Overflow. */ - value = 0; -# endif - } - - return (png_byte)(value & 0xff); -} - -#ifdef PNG_16BIT_SUPPORTED -png_uint_16 -png_gamma_16bit_correct(unsigned int value, png_fixed_point gamma_val) -{ - if (value > 0 && value < 65535) - { -# ifdef PNG_FLOATING_ARITHMETIC_SUPPORTED - /* The same (unsigned int)->(double) constraints apply here as above, - * however in this case the (unsigned int) to (int) conversion can - * overflow on an ANSI-C90 compliant system so the cast needs to ensure - * that this is not possible. - */ - double r = floor(65535*pow((png_int_32)value/65535., - gamma_val*.00001)+.5); - return (png_uint_16)r; -# else - png_int_32 lg2 = png_log16bit(value); - png_fixed_point res; - - if (png_muldiv(&res, gamma_val, lg2, PNG_FP_1) != 0) - return png_exp16bit(res); - - /* Overflow. */ - value = 0; -# endif - } - - return (png_uint_16)value; -} -#endif /* 16BIT */ - -/* This does the right thing based on the bit_depth field of the - * png_struct, interpreting values as 8-bit or 16-bit. While the result - * is nominally a 16-bit value if bit depth is 8 then the result is - * 8-bit (as are the arguments.) - */ -png_uint_16 /* PRIVATE */ -png_gamma_correct(png_structrp png_ptr, unsigned int value, - png_fixed_point gamma_val) -{ - if (png_ptr->bit_depth == 8) - return png_gamma_8bit_correct(value, gamma_val); - -#ifdef PNG_16BIT_SUPPORTED - else - return png_gamma_16bit_correct(value, gamma_val); -#else - /* should not reach this */ - return 0; -#endif /* 16BIT */ -} - -#ifdef PNG_16BIT_SUPPORTED -/* Internal function to build a single 16-bit table - the table consists of - * 'num' 256 entry subtables, where 'num' is determined by 'shift' - the amount - * to shift the input values right (or 16-number_of_signifiant_bits). - * - * The caller is responsible for ensuring that the table gets cleaned up on - * png_error (i.e. if one of the mallocs below fails) - i.e. the *table argument - * should be somewhere that will be cleaned. - */ -static void -png_build_16bit_table(png_structrp png_ptr, png_uint_16pp *ptable, - unsigned int shift, png_fixed_point gamma_val) -{ - /* Various values derived from 'shift': */ - unsigned int num = 1U << (8U - shift); -#ifdef PNG_FLOATING_ARITHMETIC_SUPPORTED - /* CSE the division and work round wacky GCC warnings (see the comments - * in png_gamma_8bit_correct for where these come from.) - */ - double fmax = 1.0 / (((png_int_32)1 << (16U - shift)) - 1); -#endif - unsigned int max = (1U << (16U - shift)) - 1U; - unsigned int max_by_2 = 1U << (15U - shift); - unsigned int i; - - png_uint_16pp table = *ptable = - (png_uint_16pp)png_calloc(png_ptr, num * (sizeof (png_uint_16p))); - - for (i = 0; i < num; i++) - { - png_uint_16p sub_table = table[i] = - (png_uint_16p)png_malloc(png_ptr, 256 * (sizeof (png_uint_16))); - - /* The 'threshold' test is repeated here because it can arise for one of - * the 16-bit tables even if the others don't hit it. - */ - if (png_gamma_significant(gamma_val) != 0) - { - /* The old code would overflow at the end and this would cause the - * 'pow' function to return a result >1, resulting in an - * arithmetic error. This code follows the spec exactly; ig is - * the recovered input sample, it always has 8-16 bits. - * - * We want input * 65535/max, rounded, the arithmetic fits in 32 - * bits (unsigned) so long as max <= 32767. - */ - unsigned int j; - for (j = 0; j < 256; j++) - { - png_uint_32 ig = (j << (8-shift)) + i; -# ifdef PNG_FLOATING_ARITHMETIC_SUPPORTED - /* Inline the 'max' scaling operation: */ - /* See png_gamma_8bit_correct for why the cast to (int) is - * required here. - */ - double d = floor(65535.*pow(ig*fmax, gamma_val*.00001)+.5); - sub_table[j] = (png_uint_16)d; -# else - if (shift != 0) - ig = (ig * 65535U + max_by_2)/max; - - sub_table[j] = png_gamma_16bit_correct(ig, gamma_val); -# endif - } - } - else - { - /* We must still build a table, but do it the fast way. */ - unsigned int j; - - for (j = 0; j < 256; j++) - { - png_uint_32 ig = (j << (8-shift)) + i; - - if (shift != 0) - ig = (ig * 65535U + max_by_2)/max; - - sub_table[j] = (png_uint_16)ig; - } - } - } -} - -/* NOTE: this function expects the *inverse* of the overall gamma transformation - * required. - */ -static void -png_build_16to8_table(png_structrp png_ptr, png_uint_16pp *ptable, - unsigned int shift, png_fixed_point gamma_val) -{ - unsigned int num = 1U << (8U - shift); - unsigned int max = (1U << (16U - shift))-1U; - unsigned int i; - png_uint_32 last; - - png_uint_16pp table = *ptable = - (png_uint_16pp)png_calloc(png_ptr, num * (sizeof (png_uint_16p))); - - /* 'num' is the number of tables and also the number of low bits of low - * bits of the input 16-bit value used to select a table. Each table is - * itself indexed by the high 8 bits of the value. - */ - for (i = 0; i < num; i++) - table[i] = (png_uint_16p)png_malloc(png_ptr, - 256 * (sizeof (png_uint_16))); - - /* 'gamma_val' is set to the reciprocal of the value calculated above, so - * pow(out,g) is an *input* value. 'last' is the last input value set. - * - * In the loop 'i' is used to find output values. Since the output is - * 8-bit there are only 256 possible values. The tables are set up to - * select the closest possible output value for each input by finding - * the input value at the boundary between each pair of output values - * and filling the table up to that boundary with the lower output - * value. - * - * The boundary values are 0.5,1.5..253.5,254.5. Since these are 9-bit - * values the code below uses a 16-bit value in i; the values start at - * 128.5 (for 0.5) and step by 257, for a total of 254 values (the last - * entries are filled with 255). Start i at 128 and fill all 'last' - * table entries <= 'max' - */ - last = 0; - for (i = 0; i < 255; ++i) /* 8-bit output value */ - { - /* Find the corresponding maximum input value */ - png_uint_16 out = (png_uint_16)(i * 257U); /* 16-bit output value */ - - /* Find the boundary value in 16 bits: */ - png_uint_32 bound = png_gamma_16bit_correct(out+128U, gamma_val); - - /* Adjust (round) to (16-shift) bits: */ - bound = (bound * max + 32768U)/65535U + 1U; - - while (last < bound) - { - table[last & (0xffU >> shift)][last >> (8U - shift)] = out; - last++; - } - } - - /* And fill in the final entries. */ - while (last < (num << 8)) - { - table[last & (0xff >> shift)][last >> (8U - shift)] = 65535U; - last++; - } -} -#endif /* 16BIT */ - -/* Build a single 8-bit table: same as the 16-bit case but much simpler (and - * typically much faster). Note that libpng currently does no sBIT processing - * (apparently contrary to the spec) so a 256-entry table is always generated. - */ -static void -png_build_8bit_table(png_structrp png_ptr, png_bytepp ptable, - png_fixed_point gamma_val) -{ - unsigned int i; - png_bytep table = *ptable = (png_bytep)png_malloc(png_ptr, 256); - - if (png_gamma_significant(gamma_val) != 0) - for (i=0; i<256; i++) - table[i] = png_gamma_8bit_correct(i, gamma_val); - - else - for (i=0; i<256; ++i) - table[i] = (png_byte)(i & 0xff); -} - -/* Used from png_read_destroy and below to release the memory used by the gamma - * tables. - */ -void /* PRIVATE */ -png_destroy_gamma_table(png_structrp png_ptr) -{ - png_free(png_ptr, png_ptr->gamma_table); - png_ptr->gamma_table = NULL; - -#ifdef PNG_16BIT_SUPPORTED - if (png_ptr->gamma_16_table != NULL) - { - int i; - int istop = (1 << (8 - png_ptr->gamma_shift)); - for (i = 0; i < istop; i++) - { - png_free(png_ptr, png_ptr->gamma_16_table[i]); - } - png_free(png_ptr, png_ptr->gamma_16_table); - png_ptr->gamma_16_table = NULL; - } -#endif /* 16BIT */ - -#if defined(PNG_READ_BACKGROUND_SUPPORTED) || \ - defined(PNG_READ_ALPHA_MODE_SUPPORTED) || \ - defined(PNG_READ_RGB_TO_GRAY_SUPPORTED) - png_free(png_ptr, png_ptr->gamma_from_1); - png_ptr->gamma_from_1 = NULL; - png_free(png_ptr, png_ptr->gamma_to_1); - png_ptr->gamma_to_1 = NULL; - -#ifdef PNG_16BIT_SUPPORTED - if (png_ptr->gamma_16_from_1 != NULL) - { - int i; - int istop = (1 << (8 - png_ptr->gamma_shift)); - for (i = 0; i < istop; i++) - { - png_free(png_ptr, png_ptr->gamma_16_from_1[i]); - } - png_free(png_ptr, png_ptr->gamma_16_from_1); - png_ptr->gamma_16_from_1 = NULL; - } - if (png_ptr->gamma_16_to_1 != NULL) - { - int i; - int istop = (1 << (8 - png_ptr->gamma_shift)); - for (i = 0; i < istop; i++) - { - png_free(png_ptr, png_ptr->gamma_16_to_1[i]); - } - png_free(png_ptr, png_ptr->gamma_16_to_1); - png_ptr->gamma_16_to_1 = NULL; - } -#endif /* 16BIT */ -#endif /* READ_BACKGROUND || READ_ALPHA_MODE || RGB_TO_GRAY */ -} - -/* We build the 8- or 16-bit gamma tables here. Note that for 16-bit - * tables, we don't make a full table if we are reducing to 8-bit in - * the future. Note also how the gamma_16 tables are segmented so that - * we don't need to allocate > 64K chunks for a full 16-bit table. - */ -void /* PRIVATE */ -png_build_gamma_table(png_structrp png_ptr, int bit_depth) -{ - png_debug(1, "in png_build_gamma_table"); - - /* Remove any existing table; this copes with multiple calls to - * png_read_update_info. The warning is because building the gamma tables - * multiple times is a performance hit - it's harmless but the ability to - * call png_read_update_info() multiple times is new in 1.5.6 so it seems - * sensible to warn if the app introduces such a hit. - */ - if (png_ptr->gamma_table != NULL || png_ptr->gamma_16_table != NULL) - { - png_warning(png_ptr, "gamma table being rebuilt"); - png_destroy_gamma_table(png_ptr); - } - - if (bit_depth <= 8) - { - png_build_8bit_table(png_ptr, &png_ptr->gamma_table, - png_ptr->screen_gamma > 0 ? - png_reciprocal2(png_ptr->colorspace.gamma, - png_ptr->screen_gamma) : PNG_FP_1); - -#if defined(PNG_READ_BACKGROUND_SUPPORTED) || \ - defined(PNG_READ_ALPHA_MODE_SUPPORTED) || \ - defined(PNG_READ_RGB_TO_GRAY_SUPPORTED) - if ((png_ptr->transformations & (PNG_COMPOSE | PNG_RGB_TO_GRAY)) != 0) - { - png_build_8bit_table(png_ptr, &png_ptr->gamma_to_1, - png_reciprocal(png_ptr->colorspace.gamma)); - - png_build_8bit_table(png_ptr, &png_ptr->gamma_from_1, - png_ptr->screen_gamma > 0 ? - png_reciprocal(png_ptr->screen_gamma) : - png_ptr->colorspace.gamma/* Probably doing rgb_to_gray */); - } -#endif /* READ_BACKGROUND || READ_ALPHA_MODE || RGB_TO_GRAY */ - } -#ifdef PNG_16BIT_SUPPORTED - else - { - png_byte shift, sig_bit; - - if ((png_ptr->color_type & PNG_COLOR_MASK_COLOR) != 0) - { - sig_bit = png_ptr->sig_bit.red; - - if (png_ptr->sig_bit.green > sig_bit) - sig_bit = png_ptr->sig_bit.green; - - if (png_ptr->sig_bit.blue > sig_bit) - sig_bit = png_ptr->sig_bit.blue; - } - else - sig_bit = png_ptr->sig_bit.gray; - - /* 16-bit gamma code uses this equation: - * - * ov = table[(iv & 0xff) >> gamma_shift][iv >> 8] - * - * Where 'iv' is the input color value and 'ov' is the output value - - * pow(iv, gamma). - * - * Thus the gamma table consists of up to 256 256-entry tables. The table - * is selected by the (8-gamma_shift) most significant of the low 8 bits - * of the color value then indexed by the upper 8 bits: - * - * table[low bits][high 8 bits] - * - * So the table 'n' corresponds to all those 'iv' of: - * - * ..<(n+1 << gamma_shift)-1> - * - */ - if (sig_bit > 0 && sig_bit < 16U) - /* shift == insignificant bits */ - shift = (png_byte)((16U - sig_bit) & 0xff); - - else - shift = 0; /* keep all 16 bits */ - - if ((png_ptr->transformations & (PNG_16_TO_8 | PNG_SCALE_16_TO_8)) != 0) - { - /* PNG_MAX_GAMMA_8 is the number of bits to keep - effectively - * the significant bits in the *input* when the output will - * eventually be 8 bits. By default it is 11. - */ - if (shift < (16U - PNG_MAX_GAMMA_8)) - shift = (16U - PNG_MAX_GAMMA_8); - } - - if (shift > 8U) - shift = 8U; /* Guarantees at least one table! */ - - png_ptr->gamma_shift = shift; - - /* NOTE: prior to 1.5.4 this test used to include PNG_BACKGROUND (now - * PNG_COMPOSE). This effectively smashed the background calculation for - * 16-bit output because the 8-bit table assumes the result will be - * reduced to 8 bits. - */ - if ((png_ptr->transformations & (PNG_16_TO_8 | PNG_SCALE_16_TO_8)) != 0) - png_build_16to8_table(png_ptr, &png_ptr->gamma_16_table, shift, - png_ptr->screen_gamma > 0 ? png_product2(png_ptr->colorspace.gamma, - png_ptr->screen_gamma) : PNG_FP_1); - - else - png_build_16bit_table(png_ptr, &png_ptr->gamma_16_table, shift, - png_ptr->screen_gamma > 0 ? png_reciprocal2(png_ptr->colorspace.gamma, - png_ptr->screen_gamma) : PNG_FP_1); - -#if defined(PNG_READ_BACKGROUND_SUPPORTED) || \ - defined(PNG_READ_ALPHA_MODE_SUPPORTED) || \ - defined(PNG_READ_RGB_TO_GRAY_SUPPORTED) - if ((png_ptr->transformations & (PNG_COMPOSE | PNG_RGB_TO_GRAY)) != 0) - { - png_build_16bit_table(png_ptr, &png_ptr->gamma_16_to_1, shift, - png_reciprocal(png_ptr->colorspace.gamma)); - - /* Notice that the '16 from 1' table should be full precision, however - * the lookup on this table still uses gamma_shift, so it can't be. - * TODO: fix this. - */ - png_build_16bit_table(png_ptr, &png_ptr->gamma_16_from_1, shift, - png_ptr->screen_gamma > 0 ? png_reciprocal(png_ptr->screen_gamma) : - png_ptr->colorspace.gamma/* Probably doing rgb_to_gray */); - } -#endif /* READ_BACKGROUND || READ_ALPHA_MODE || RGB_TO_GRAY */ - } -#endif /* 16BIT */ -} -#endif /* READ_GAMMA */ - -/* HARDWARE OR SOFTWARE OPTION SUPPORT */ -#ifdef PNG_SET_OPTION_SUPPORTED -int PNGAPI -png_set_option(png_structrp png_ptr, int option, int onoff) -{ - if (png_ptr != NULL && option >= 0 && option < PNG_OPTION_NEXT && - (option & 1) == 0) - { - png_uint_32 mask = 3U << option; - png_uint_32 setting = (2U + (onoff != 0)) << option; - png_uint_32 current = png_ptr->options; - - png_ptr->options = (png_uint_32)((current & ~mask) | setting); - - return (int)(current & mask) >> option; - } - - return PNG_OPTION_INVALID; -} -#endif - -/* sRGB support */ -#if defined(PNG_SIMPLIFIED_READ_SUPPORTED) ||\ - defined(PNG_SIMPLIFIED_WRITE_SUPPORTED) -/* sRGB conversion tables; these are machine generated with the code in - * contrib/tools/makesRGB.c. The actual sRGB transfer curve defined in the - * specification (see the article at https://en.wikipedia.org/wiki/SRGB) - * is used, not the gamma=1/2.2 approximation use elsewhere in libpng. - * The sRGB to linear table is exact (to the nearest 16-bit linear fraction). - * The inverse (linear to sRGB) table has accuracies as follows: - * - * For all possible (255*65535+1) input values: - * - * error: -0.515566 - 0.625971, 79441 (0.475369%) of readings inexact - * - * For the input values corresponding to the 65536 16-bit values: - * - * error: -0.513727 - 0.607759, 308 (0.469978%) of readings inexact - * - * In all cases the inexact readings are only off by one. - */ - -#ifdef PNG_SIMPLIFIED_READ_SUPPORTED -/* The convert-to-sRGB table is only currently required for read. */ -const png_uint_16 png_sRGB_table[256] = -{ - 0,20,40,60,80,99,119,139, - 159,179,199,219,241,264,288,313, - 340,367,396,427,458,491,526,562, - 599,637,677,718,761,805,851,898, - 947,997,1048,1101,1156,1212,1270,1330, - 1391,1453,1517,1583,1651,1720,1790,1863, - 1937,2013,2090,2170,2250,2333,2418,2504, - 2592,2681,2773,2866,2961,3058,3157,3258, - 3360,3464,3570,3678,3788,3900,4014,4129, - 4247,4366,4488,4611,4736,4864,4993,5124, - 5257,5392,5530,5669,5810,5953,6099,6246, - 6395,6547,6700,6856,7014,7174,7335,7500, - 7666,7834,8004,8177,8352,8528,8708,8889, - 9072,9258,9445,9635,9828,10022,10219,10417, - 10619,10822,11028,11235,11446,11658,11873,12090, - 12309,12530,12754,12980,13209,13440,13673,13909, - 14146,14387,14629,14874,15122,15371,15623,15878, - 16135,16394,16656,16920,17187,17456,17727,18001, - 18277,18556,18837,19121,19407,19696,19987,20281, - 20577,20876,21177,21481,21787,22096,22407,22721, - 23038,23357,23678,24002,24329,24658,24990,25325, - 25662,26001,26344,26688,27036,27386,27739,28094, - 28452,28813,29176,29542,29911,30282,30656,31033, - 31412,31794,32179,32567,32957,33350,33745,34143, - 34544,34948,35355,35764,36176,36591,37008,37429, - 37852,38278,38706,39138,39572,40009,40449,40891, - 41337,41785,42236,42690,43147,43606,44069,44534, - 45002,45473,45947,46423,46903,47385,47871,48359, - 48850,49344,49841,50341,50844,51349,51858,52369, - 52884,53401,53921,54445,54971,55500,56032,56567, - 57105,57646,58190,58737,59287,59840,60396,60955, - 61517,62082,62650,63221,63795,64372,64952,65535 -}; -#endif /* SIMPLIFIED_READ */ - -/* The base/delta tables are required for both read and write (but currently - * only the simplified versions.) - */ -const png_uint_16 png_sRGB_base[512] = -{ - 128,1782,3383,4644,5675,6564,7357,8074, - 8732,9346,9921,10463,10977,11466,11935,12384, - 12816,13233,13634,14024,14402,14769,15125,15473, - 15812,16142,16466,16781,17090,17393,17690,17981, - 18266,18546,18822,19093,19359,19621,19879,20133, - 20383,20630,20873,21113,21349,21583,21813,22041, - 22265,22487,22707,22923,23138,23350,23559,23767, - 23972,24175,24376,24575,24772,24967,25160,25352, - 25542,25730,25916,26101,26284,26465,26645,26823, - 27000,27176,27350,27523,27695,27865,28034,28201, - 28368,28533,28697,28860,29021,29182,29341,29500, - 29657,29813,29969,30123,30276,30429,30580,30730, - 30880,31028,31176,31323,31469,31614,31758,31902, - 32045,32186,32327,32468,32607,32746,32884,33021, - 33158,33294,33429,33564,33697,33831,33963,34095, - 34226,34357,34486,34616,34744,34873,35000,35127, - 35253,35379,35504,35629,35753,35876,35999,36122, - 36244,36365,36486,36606,36726,36845,36964,37083, - 37201,37318,37435,37551,37668,37783,37898,38013, - 38127,38241,38354,38467,38580,38692,38803,38915, - 39026,39136,39246,39356,39465,39574,39682,39790, - 39898,40005,40112,40219,40325,40431,40537,40642, - 40747,40851,40955,41059,41163,41266,41369,41471, - 41573,41675,41777,41878,41979,42079,42179,42279, - 42379,42478,42577,42676,42775,42873,42971,43068, - 43165,43262,43359,43456,43552,43648,43743,43839, - 43934,44028,44123,44217,44311,44405,44499,44592, - 44685,44778,44870,44962,45054,45146,45238,45329, - 45420,45511,45601,45692,45782,45872,45961,46051, - 46140,46229,46318,46406,46494,46583,46670,46758, - 46846,46933,47020,47107,47193,47280,47366,47452, - 47538,47623,47709,47794,47879,47964,48048,48133, - 48217,48301,48385,48468,48552,48635,48718,48801, - 48884,48966,49048,49131,49213,49294,49376,49458, - 49539,49620,49701,49782,49862,49943,50023,50103, - 50183,50263,50342,50422,50501,50580,50659,50738, - 50816,50895,50973,51051,51129,51207,51285,51362, - 51439,51517,51594,51671,51747,51824,51900,51977, - 52053,52129,52205,52280,52356,52432,52507,52582, - 52657,52732,52807,52881,52956,53030,53104,53178, - 53252,53326,53400,53473,53546,53620,53693,53766, - 53839,53911,53984,54056,54129,54201,54273,54345, - 54417,54489,54560,54632,54703,54774,54845,54916, - 54987,55058,55129,55199,55269,55340,55410,55480, - 55550,55620,55689,55759,55828,55898,55967,56036, - 56105,56174,56243,56311,56380,56448,56517,56585, - 56653,56721,56789,56857,56924,56992,57059,57127, - 57194,57261,57328,57395,57462,57529,57595,57662, - 57728,57795,57861,57927,57993,58059,58125,58191, - 58256,58322,58387,58453,58518,58583,58648,58713, - 58778,58843,58908,58972,59037,59101,59165,59230, - 59294,59358,59422,59486,59549,59613,59677,59740, - 59804,59867,59930,59993,60056,60119,60182,60245, - 60308,60370,60433,60495,60558,60620,60682,60744, - 60806,60868,60930,60992,61054,61115,61177,61238, - 61300,61361,61422,61483,61544,61605,61666,61727, - 61788,61848,61909,61969,62030,62090,62150,62211, - 62271,62331,62391,62450,62510,62570,62630,62689, - 62749,62808,62867,62927,62986,63045,63104,63163, - 63222,63281,63340,63398,63457,63515,63574,63632, - 63691,63749,63807,63865,63923,63981,64039,64097, - 64155,64212,64270,64328,64385,64443,64500,64557, - 64614,64672,64729,64786,64843,64900,64956,65013, - 65070,65126,65183,65239,65296,65352,65409,65465 -}; - -const png_byte png_sRGB_delta[512] = -{ - 207,201,158,129,113,100,90,82,77,72,68,64,61,59,56,54, - 52,50,49,47,46,45,43,42,41,40,39,39,38,37,36,36, - 35,34,34,33,33,32,32,31,31,30,30,30,29,29,28,28, - 28,27,27,27,27,26,26,26,25,25,25,25,24,24,24,24, - 23,23,23,23,23,22,22,22,22,22,22,21,21,21,21,21, - 21,20,20,20,20,20,20,20,20,19,19,19,19,19,19,19, - 19,18,18,18,18,18,18,18,18,18,18,17,17,17,17,17, - 17,17,17,17,17,17,16,16,16,16,16,16,16,16,16,16, - 16,16,16,16,15,15,15,15,15,15,15,15,15,15,15,15, - 15,15,15,15,14,14,14,14,14,14,14,14,14,14,14,14, - 14,14,14,14,14,14,14,13,13,13,13,13,13,13,13,13, - 13,13,13,13,13,13,13,13,13,13,13,13,13,13,12,12, - 12,12,12,12,12,12,12,12,12,12,12,12,12,12,12,12, - 12,12,12,12,12,12,12,12,12,12,12,12,11,11,11,11, - 11,11,11,11,11,11,11,11,11,11,11,11,11,11,11,11, - 11,11,11,11,11,11,11,11,11,11,11,11,11,11,11,11, - 11,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10, - 10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10, - 10,10,10,10,10,10,10,10,10,10,10,10,10,10,10,10, - 10,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,9,9,9,9,9,9,9,9,9,9,9,9,9,9,9, - 9,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8, - 8,8,8,8,8,8,8,8,8,7,7,7,7,7,7,7, - 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, - 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7, - 7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7 -}; -#endif /* SIMPLIFIED READ/WRITE sRGB support */ - -/* SIMPLIFIED READ/WRITE SUPPORT */ -#if defined(PNG_SIMPLIFIED_READ_SUPPORTED) ||\ - defined(PNG_SIMPLIFIED_WRITE_SUPPORTED) -static int -png_image_free_function(png_voidp argument) -{ - png_imagep image = png_voidcast(png_imagep, argument); - png_controlp cp = image->opaque; - png_control c; - - /* Double check that we have a png_ptr - it should be impossible to get here - * without one. - */ - if (cp->png_ptr == NULL) - return 0; - - /* First free any data held in the control structure. */ -# ifdef PNG_STDIO_SUPPORTED - if (cp->owned_file != 0) - { - FILE *fp = png_voidcast(FILE*, cp->png_ptr->io_ptr); - cp->owned_file = 0; - - /* Ignore errors here. */ - if (fp != NULL) - { - cp->png_ptr->io_ptr = NULL; - (void)fclose(fp); - } - } -# endif - - /* Copy the control structure so that the original, allocated, version can be - * safely freed. Notice that a png_error here stops the remainder of the - * cleanup, but this is probably fine because that would indicate bad memory - * problems anyway. - */ - c = *cp; - image->opaque = &c; - png_free(c.png_ptr, cp); - - /* Then the structures, calling the correct API. */ - if (c.for_write != 0) - { -# ifdef PNG_SIMPLIFIED_WRITE_SUPPORTED - png_destroy_write_struct(&c.png_ptr, &c.info_ptr); -# else - png_error(c.png_ptr, "simplified write not supported"); -# endif - } - else - { -# ifdef PNG_SIMPLIFIED_READ_SUPPORTED - png_destroy_read_struct(&c.png_ptr, &c.info_ptr, NULL); -# else - png_error(c.png_ptr, "simplified read not supported"); -# endif - } - - /* Success. */ - return 1; -} - -void PNGAPI -png_image_free(png_imagep image) -{ - /* Safely call the real function, but only if doing so is safe at this point - * (if not inside an error handling context). Otherwise assume - * png_safe_execute will call this API after the return. - */ - if (image != NULL && image->opaque != NULL && - image->opaque->error_buf == NULL) - { - png_image_free_function(image); - image->opaque = NULL; - } -} - -int /* PRIVATE */ -png_image_error(png_imagep image, png_const_charp error_message) -{ - /* Utility to log an error. */ - png_safecat(image->message, (sizeof image->message), 0, error_message); - image->warning_or_error |= PNG_IMAGE_ERROR; - png_image_free(image); - return 0; -} - -#endif /* SIMPLIFIED READ/WRITE */ -#endif /* READ || WRITE */ diff --git a/dep/libpng/src/pngdebug.h b/dep/libpng/src/pngdebug.h deleted file mode 100644 index 00d5a4569..000000000 --- a/dep/libpng/src/pngdebug.h +++ /dev/null @@ -1,153 +0,0 @@ - -/* pngdebug.h - Debugging macros for libpng, also used in pngtest.c - * - * Copyright (c) 2018 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2013 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -/* Define PNG_DEBUG at compile time for debugging information. Higher - * numbers for PNG_DEBUG mean more debugging information. This has - * only been added since version 0.95 so it is not implemented throughout - * libpng yet, but more support will be added as needed. - * - * png_debug[1-2]?(level, message ,arg{0-2}) - * Expands to a statement (either a simple expression or a compound - * do..while(0) statement) that outputs a message with parameter - * substitution if PNG_DEBUG is defined to 2 or more. If PNG_DEBUG - * is undefined, 0 or 1 every png_debug expands to a simple expression - * (actually ((void)0)). - * - * level: level of detail of message, starting at 0. A level 'n' - * message is preceded by 'n' 3-space indentations (not implemented - * on Microsoft compilers unless PNG_DEBUG_FILE is also - * defined, to allow debug DLL compilation with no standard IO). - * message: a printf(3) style text string. A trailing '\n' is added - * to the message. - * arg: 0 to 2 arguments for printf(3) style substitution in message. - */ -#ifndef PNGDEBUG_H -#define PNGDEBUG_H -/* These settings control the formatting of messages in png.c and pngerror.c */ -/* Moved to pngdebug.h at 1.5.0 */ -# ifndef PNG_LITERAL_SHARP -# define PNG_LITERAL_SHARP 0x23 -# endif -# ifndef PNG_LITERAL_LEFT_SQUARE_BRACKET -# define PNG_LITERAL_LEFT_SQUARE_BRACKET 0x5b -# endif -# ifndef PNG_LITERAL_RIGHT_SQUARE_BRACKET -# define PNG_LITERAL_RIGHT_SQUARE_BRACKET 0x5d -# endif -# ifndef PNG_STRING_NEWLINE -# define PNG_STRING_NEWLINE "\n" -# endif - -#ifdef PNG_DEBUG -# if (PNG_DEBUG > 0) -# if !defined(PNG_DEBUG_FILE) && defined(_MSC_VER) -# include -# if (PNG_DEBUG > 1) -# ifndef _DEBUG -# define _DEBUG -# endif -# ifndef png_debug -# define png_debug(l,m) _RPT0(_CRT_WARN,m PNG_STRING_NEWLINE) -# endif -# ifndef png_debug1 -# define png_debug1(l,m,p1) _RPT1(_CRT_WARN,m PNG_STRING_NEWLINE,p1) -# endif -# ifndef png_debug2 -# define png_debug2(l,m,p1,p2) \ - _RPT2(_CRT_WARN,m PNG_STRING_NEWLINE,p1,p2) -# endif -# endif -# else /* PNG_DEBUG_FILE || !_MSC_VER */ -# ifndef PNG_STDIO_SUPPORTED -# include /* not included yet */ -# endif -# ifndef PNG_DEBUG_FILE -# define PNG_DEBUG_FILE stderr -# endif /* PNG_DEBUG_FILE */ - -# if (PNG_DEBUG > 1) -# ifdef __STDC__ -# ifndef png_debug -# define png_debug(l,m) \ - do { \ - int num_tabs=l; \ - fprintf(PNG_DEBUG_FILE,"%s" m PNG_STRING_NEWLINE,(num_tabs==1 ? " " : \ - (num_tabs==2 ? " " : (num_tabs>2 ? " " : "")))); \ - } while (0) -# endif -# ifndef png_debug1 -# define png_debug1(l,m,p1) \ - do { \ - int num_tabs=l; \ - fprintf(PNG_DEBUG_FILE,"%s" m PNG_STRING_NEWLINE,(num_tabs==1 ? " " : \ - (num_tabs==2 ? " " : (num_tabs>2 ? " " : ""))),p1); \ - } while (0) -# endif -# ifndef png_debug2 -# define png_debug2(l,m,p1,p2) \ - do { \ - int num_tabs=l; \ - fprintf(PNG_DEBUG_FILE,"%s" m PNG_STRING_NEWLINE,(num_tabs==1 ? " " : \ - (num_tabs==2 ? " " : (num_tabs>2 ? " " : ""))),p1,p2);\ - } while (0) -# endif -# else /* __STDC __ */ -# ifndef png_debug -# define png_debug(l,m) \ - do { \ - int num_tabs=l; \ - char format[256]; \ - snprintf(format,256,"%s%s%s",(num_tabs==1 ? "\t" : \ - (num_tabs==2 ? "\t\t":(num_tabs>2 ? "\t\t\t":""))), \ - m,PNG_STRING_NEWLINE); \ - fprintf(PNG_DEBUG_FILE,format); \ - } while (0) -# endif -# ifndef png_debug1 -# define png_debug1(l,m,p1) \ - do { \ - int num_tabs=l; \ - char format[256]; \ - snprintf(format,256,"%s%s%s",(num_tabs==1 ? "\t" : \ - (num_tabs==2 ? "\t\t":(num_tabs>2 ? "\t\t\t":""))), \ - m,PNG_STRING_NEWLINE); \ - fprintf(PNG_DEBUG_FILE,format,p1); \ - } while (0) -# endif -# ifndef png_debug2 -# define png_debug2(l,m,p1,p2) \ - do { \ - int num_tabs=l; \ - char format[256]; \ - snprintf(format,256,"%s%s%s",(num_tabs==1 ? "\t" : \ - (num_tabs==2 ? "\t\t":(num_tabs>2 ? "\t\t\t":""))), \ - m,PNG_STRING_NEWLINE); \ - fprintf(PNG_DEBUG_FILE,format,p1,p2); \ - } while (0) -# endif -# endif /* __STDC __ */ -# endif /* (PNG_DEBUG > 1) */ - -# endif /* _MSC_VER */ -# endif /* (PNG_DEBUG > 0) */ -#endif /* PNG_DEBUG */ -#ifndef png_debug -# define png_debug(l, m) ((void)0) -#endif -#ifndef png_debug1 -# define png_debug1(l, m, p1) ((void)0) -#endif -#ifndef png_debug2 -# define png_debug2(l, m, p1, p2) ((void)0) -#endif -#endif /* PNGDEBUG_H */ diff --git a/dep/libpng/src/pngerror.c b/dep/libpng/src/pngerror.c deleted file mode 100644 index 29ebda794..000000000 --- a/dep/libpng/src/pngerror.c +++ /dev/null @@ -1,957 +0,0 @@ - -/* pngerror.c - stub functions for i/o and memory allocation - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2017 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - * - * This file provides a location for all error handling. Users who - * need special error handling are expected to write replacement functions - * and use png_set_error_fn() to use those functions. See the instructions - * at each function. - */ - -#include "pngpriv.h" - -#if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) - -static PNG_FUNCTION(void, png_default_error,PNGARG((png_const_structrp png_ptr, - png_const_charp error_message)),PNG_NORETURN); - -#ifdef PNG_WARNINGS_SUPPORTED -static void /* PRIVATE */ -png_default_warning PNGARG((png_const_structrp png_ptr, - png_const_charp warning_message)); -#endif /* WARNINGS */ - -/* This function is called whenever there is a fatal error. This function - * should not be changed. If there is a need to handle errors differently, - * you should supply a replacement error function and use png_set_error_fn() - * to replace the error function at run-time. - */ -#ifdef PNG_ERROR_TEXT_SUPPORTED -PNG_FUNCTION(void,PNGAPI -png_error,(png_const_structrp png_ptr, png_const_charp error_message), - PNG_NORETURN) -{ -#ifdef PNG_ERROR_NUMBERS_SUPPORTED - char msg[16]; - if (png_ptr != NULL) - { - if ((png_ptr->flags & - (PNG_FLAG_STRIP_ERROR_NUMBERS|PNG_FLAG_STRIP_ERROR_TEXT)) != 0) - { - if (*error_message == PNG_LITERAL_SHARP) - { - /* Strip "#nnnn " from beginning of error message. */ - int offset; - for (offset = 1; offset<15; offset++) - if (error_message[offset] == ' ') - break; - - if ((png_ptr->flags & PNG_FLAG_STRIP_ERROR_TEXT) != 0) - { - int i; - for (i = 0; i < offset - 1; i++) - msg[i] = error_message[i + 1]; - msg[i - 1] = '\0'; - error_message = msg; - } - - else - error_message += offset; - } - - else - { - if ((png_ptr->flags & PNG_FLAG_STRIP_ERROR_TEXT) != 0) - { - msg[0] = '0'; - msg[1] = '\0'; - error_message = msg; - } - } - } - } -#endif - if (png_ptr != NULL && png_ptr->error_fn != NULL) - (*(png_ptr->error_fn))(png_constcast(png_structrp,png_ptr), - error_message); - - /* If the custom handler doesn't exist, or if it returns, - use the default handler, which will not return. */ - png_default_error(png_ptr, error_message); -} -#else -PNG_FUNCTION(void,PNGAPI -png_err,(png_const_structrp png_ptr),PNG_NORETURN) -{ - /* Prior to 1.5.2 the error_fn received a NULL pointer, expressed - * erroneously as '\0', instead of the empty string "". This was - * apparently an error, introduced in libpng-1.2.20, and png_default_error - * will crash in this case. - */ - if (png_ptr != NULL && png_ptr->error_fn != NULL) - (*(png_ptr->error_fn))(png_constcast(png_structrp,png_ptr), ""); - - /* If the custom handler doesn't exist, or if it returns, - use the default handler, which will not return. */ - png_default_error(png_ptr, ""); -} -#endif /* ERROR_TEXT */ - -/* Utility to safely appends strings to a buffer. This never errors out so - * error checking is not required in the caller. - */ -size_t -png_safecat(png_charp buffer, size_t bufsize, size_t pos, - png_const_charp string) -{ - if (buffer != NULL && pos < bufsize) - { - if (string != NULL) - while (*string != '\0' && pos < bufsize-1) - buffer[pos++] = *string++; - - buffer[pos] = '\0'; - } - - return pos; -} - -#if defined(PNG_WARNINGS_SUPPORTED) || defined(PNG_TIME_RFC1123_SUPPORTED) -/* Utility to dump an unsigned value into a buffer, given a start pointer and - * and end pointer (which should point just *beyond* the end of the buffer!) - * Returns the pointer to the start of the formatted string. - */ -png_charp -png_format_number(png_const_charp start, png_charp end, int format, - png_alloc_size_t number) -{ - int count = 0; /* number of digits output */ - int mincount = 1; /* minimum number required */ - int output = 0; /* digit output (for the fixed point format) */ - - *--end = '\0'; - - /* This is written so that the loop always runs at least once, even with - * number zero. - */ - while (end > start && (number != 0 || count < mincount)) - { - - static const char digits[] = "0123456789ABCDEF"; - - switch (format) - { - case PNG_NUMBER_FORMAT_fixed: - /* Needs five digits (the fraction) */ - mincount = 5; - if (output != 0 || number % 10 != 0) - { - *--end = digits[number % 10]; - output = 1; - } - number /= 10; - break; - - case PNG_NUMBER_FORMAT_02u: - /* Expects at least 2 digits. */ - mincount = 2; - /* FALLTHROUGH */ - - case PNG_NUMBER_FORMAT_u: - *--end = digits[number % 10]; - number /= 10; - break; - - case PNG_NUMBER_FORMAT_02x: - /* This format expects at least two digits */ - mincount = 2; - /* FALLTHROUGH */ - - case PNG_NUMBER_FORMAT_x: - *--end = digits[number & 0xf]; - number >>= 4; - break; - - default: /* an error */ - number = 0; - break; - } - - /* Keep track of the number of digits added */ - ++count; - - /* Float a fixed number here: */ - if ((format == PNG_NUMBER_FORMAT_fixed) && (count == 5) && (end > start)) - { - /* End of the fraction, but maybe nothing was output? In that case - * drop the decimal point. If the number is a true zero handle that - * here. - */ - if (output != 0) - *--end = '.'; - else if (number == 0) /* and !output */ - *--end = '0'; - } - } - - return end; -} -#endif - -#ifdef PNG_WARNINGS_SUPPORTED -/* This function is called whenever there is a non-fatal error. This function - * should not be changed. If there is a need to handle warnings differently, - * you should supply a replacement warning function and use - * png_set_error_fn() to replace the warning function at run-time. - */ -void PNGAPI -png_warning(png_const_structrp png_ptr, png_const_charp warning_message) -{ - int offset = 0; - if (png_ptr != NULL) - { -#ifdef PNG_ERROR_NUMBERS_SUPPORTED - if ((png_ptr->flags & - (PNG_FLAG_STRIP_ERROR_NUMBERS|PNG_FLAG_STRIP_ERROR_TEXT)) != 0) -#endif - { - if (*warning_message == PNG_LITERAL_SHARP) - { - for (offset = 1; offset < 15; offset++) - if (warning_message[offset] == ' ') - break; - } - } - } - if (png_ptr != NULL && png_ptr->warning_fn != NULL) - (*(png_ptr->warning_fn))(png_constcast(png_structrp,png_ptr), - warning_message + offset); - else - png_default_warning(png_ptr, warning_message + offset); -} - -/* These functions support 'formatted' warning messages with up to - * PNG_WARNING_PARAMETER_COUNT parameters. In the format string the parameter - * is introduced by @, where 'number' starts at 1. This follows the - * standard established by X/Open for internationalizable error messages. - */ -void -png_warning_parameter(png_warning_parameters p, int number, - png_const_charp string) -{ - if (number > 0 && number <= PNG_WARNING_PARAMETER_COUNT) - (void)png_safecat(p[number-1], (sizeof p[number-1]), 0, string); -} - -void -png_warning_parameter_unsigned(png_warning_parameters p, int number, int format, - png_alloc_size_t value) -{ - char buffer[PNG_NUMBER_BUFFER_SIZE] = {0}; - png_warning_parameter(p, number, PNG_FORMAT_NUMBER(buffer, format, value)); -} - -void -png_warning_parameter_signed(png_warning_parameters p, int number, int format, - png_int_32 value) -{ - png_alloc_size_t u; - png_charp str; - char buffer[PNG_NUMBER_BUFFER_SIZE] = {0}; - - /* Avoid overflow by doing the negate in a png_alloc_size_t: */ - u = (png_alloc_size_t)value; - if (value < 0) - u = ~u + 1; - - str = PNG_FORMAT_NUMBER(buffer, format, u); - - if (value < 0 && str > buffer) - *--str = '-'; - - png_warning_parameter(p, number, str); -} - -void -png_formatted_warning(png_const_structrp png_ptr, png_warning_parameters p, - png_const_charp message) -{ - /* The internal buffer is just 192 bytes - enough for all our messages, - * overflow doesn't happen because this code checks! If someone figures - * out how to send us a message longer than 192 bytes, all that will - * happen is that the message will be truncated appropriately. - */ - size_t i = 0; /* Index in the msg[] buffer: */ - char msg[192]; - - /* Each iteration through the following loop writes at most one character - * to msg[i++] then returns here to validate that there is still space for - * the trailing '\0'. It may (in the case of a parameter) read more than - * one character from message[]; it must check for '\0' and continue to the - * test if it finds the end of string. - */ - while (i<(sizeof msg)-1 && *message != '\0') - { - /* '@' at end of string is now just printed (previously it was skipped); - * it is an error in the calling code to terminate the string with @. - */ - if (p != NULL && *message == '@' && message[1] != '\0') - { - int parameter_char = *++message; /* Consume the '@' */ - static const char valid_parameters[] = "123456789"; - int parameter = 0; - - /* Search for the parameter digit, the index in the string is the - * parameter to use. - */ - while (valid_parameters[parameter] != parameter_char && - valid_parameters[parameter] != '\0') - ++parameter; - - /* If the parameter digit is out of range it will just get printed. */ - if (parameter < PNG_WARNING_PARAMETER_COUNT) - { - /* Append this parameter */ - png_const_charp parm = p[parameter]; - png_const_charp pend = p[parameter] + (sizeof p[parameter]); - - /* No need to copy the trailing '\0' here, but there is no guarantee - * that parm[] has been initialized, so there is no guarantee of a - * trailing '\0': - */ - while (i<(sizeof msg)-1 && *parm != '\0' && parm < pend) - msg[i++] = *parm++; - - /* Consume the parameter digit too: */ - ++message; - continue; - } - - /* else not a parameter and there is a character after the @ sign; just - * copy that. This is known not to be '\0' because of the test above. - */ - } - - /* At this point *message can't be '\0', even in the bad parameter case - * above where there is a lone '@' at the end of the message string. - */ - msg[i++] = *message++; - } - - /* i is always less than (sizeof msg), so: */ - msg[i] = '\0'; - - /* And this is the formatted message. It may be larger than - * PNG_MAX_ERROR_TEXT, but that is only used for 'chunk' errors and these - * are not (currently) formatted. - */ - png_warning(png_ptr, msg); -} -#endif /* WARNINGS */ - -#ifdef PNG_BENIGN_ERRORS_SUPPORTED -void PNGAPI -png_benign_error(png_const_structrp png_ptr, png_const_charp error_message) -{ - if ((png_ptr->flags & PNG_FLAG_BENIGN_ERRORS_WARN) != 0) - { -# ifdef PNG_READ_SUPPORTED - if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0 && - png_ptr->chunk_name != 0) - png_chunk_warning(png_ptr, error_message); - else -# endif - png_warning(png_ptr, error_message); - } - - else - { -# ifdef PNG_READ_SUPPORTED - if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0 && - png_ptr->chunk_name != 0) - png_chunk_error(png_ptr, error_message); - else -# endif - png_error(png_ptr, error_message); - } - -# ifndef PNG_ERROR_TEXT_SUPPORTED - PNG_UNUSED(error_message) -# endif -} - -void /* PRIVATE */ -png_app_warning(png_const_structrp png_ptr, png_const_charp error_message) -{ - if ((png_ptr->flags & PNG_FLAG_APP_WARNINGS_WARN) != 0) - png_warning(png_ptr, error_message); - else - png_error(png_ptr, error_message); - -# ifndef PNG_ERROR_TEXT_SUPPORTED - PNG_UNUSED(error_message) -# endif -} - -void /* PRIVATE */ -png_app_error(png_const_structrp png_ptr, png_const_charp error_message) -{ - if ((png_ptr->flags & PNG_FLAG_APP_ERRORS_WARN) != 0) - png_warning(png_ptr, error_message); - else - png_error(png_ptr, error_message); - -# ifndef PNG_ERROR_TEXT_SUPPORTED - PNG_UNUSED(error_message) -# endif -} -#endif /* BENIGN_ERRORS */ - -#define PNG_MAX_ERROR_TEXT 196 /* Currently limited by profile_error in png.c */ -#if defined(PNG_WARNINGS_SUPPORTED) || \ - (defined(PNG_READ_SUPPORTED) && defined(PNG_ERROR_TEXT_SUPPORTED)) -/* These utilities are used internally to build an error message that relates - * to the current chunk. The chunk name comes from png_ptr->chunk_name, - * which is used to prefix the message. The message is limited in length - * to 63 bytes. The name characters are output as hex digits wrapped in [] - * if the character is invalid. - */ -#define isnonalpha(c) ((c) < 65 || (c) > 122 || ((c) > 90 && (c) < 97)) -static const char png_digit[16] = { - '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', - 'A', 'B', 'C', 'D', 'E', 'F' -}; - -static void /* PRIVATE */ -png_format_buffer(png_const_structrp png_ptr, png_charp buffer, png_const_charp - error_message) -{ - png_uint_32 chunk_name = png_ptr->chunk_name; - int iout = 0, ishift = 24; - - while (ishift >= 0) - { - int c = (int)(chunk_name >> ishift) & 0xff; - - ishift -= 8; - if (isnonalpha(c) != 0) - { - buffer[iout++] = PNG_LITERAL_LEFT_SQUARE_BRACKET; - buffer[iout++] = png_digit[(c & 0xf0) >> 4]; - buffer[iout++] = png_digit[c & 0x0f]; - buffer[iout++] = PNG_LITERAL_RIGHT_SQUARE_BRACKET; - } - - else - { - buffer[iout++] = (char)c; - } - } - - if (error_message == NULL) - buffer[iout] = '\0'; - - else - { - int iin = 0; - - buffer[iout++] = ':'; - buffer[iout++] = ' '; - - while (iin < PNG_MAX_ERROR_TEXT-1 && error_message[iin] != '\0') - buffer[iout++] = error_message[iin++]; - - /* iin < PNG_MAX_ERROR_TEXT, so the following is safe: */ - buffer[iout] = '\0'; - } -} -#endif /* WARNINGS || ERROR_TEXT */ - -#if defined(PNG_READ_SUPPORTED) && defined(PNG_ERROR_TEXT_SUPPORTED) -PNG_FUNCTION(void,PNGAPI -png_chunk_error,(png_const_structrp png_ptr, png_const_charp error_message), - PNG_NORETURN) -{ - char msg[18+PNG_MAX_ERROR_TEXT]; - if (png_ptr == NULL) - png_error(png_ptr, error_message); - - else - { - png_format_buffer(png_ptr, msg, error_message); - png_error(png_ptr, msg); - } -} -#endif /* READ && ERROR_TEXT */ - -#ifdef PNG_WARNINGS_SUPPORTED -void PNGAPI -png_chunk_warning(png_const_structrp png_ptr, png_const_charp warning_message) -{ - char msg[18+PNG_MAX_ERROR_TEXT]; - if (png_ptr == NULL) - png_warning(png_ptr, warning_message); - - else - { - png_format_buffer(png_ptr, msg, warning_message); - png_warning(png_ptr, msg); - } -} -#endif /* WARNINGS */ - -#ifdef PNG_READ_SUPPORTED -#ifdef PNG_BENIGN_ERRORS_SUPPORTED -void PNGAPI -png_chunk_benign_error(png_const_structrp png_ptr, png_const_charp - error_message) -{ - if ((png_ptr->flags & PNG_FLAG_BENIGN_ERRORS_WARN) != 0) - png_chunk_warning(png_ptr, error_message); - - else - png_chunk_error(png_ptr, error_message); - -# ifndef PNG_ERROR_TEXT_SUPPORTED - PNG_UNUSED(error_message) -# endif -} -#endif -#endif /* READ */ - -void /* PRIVATE */ -png_chunk_report(png_const_structrp png_ptr, png_const_charp message, int error) -{ -# ifndef PNG_WARNINGS_SUPPORTED - PNG_UNUSED(message) -# endif - - /* This is always supported, but for just read or just write it - * unconditionally does the right thing. - */ -# if defined(PNG_READ_SUPPORTED) && defined(PNG_WRITE_SUPPORTED) - if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0) -# endif - -# ifdef PNG_READ_SUPPORTED - { - if (error < PNG_CHUNK_ERROR) - png_chunk_warning(png_ptr, message); - - else - png_chunk_benign_error(png_ptr, message); - } -# endif - -# if defined(PNG_READ_SUPPORTED) && defined(PNG_WRITE_SUPPORTED) - else if ((png_ptr->mode & PNG_IS_READ_STRUCT) == 0) -# endif - -# ifdef PNG_WRITE_SUPPORTED - { - if (error < PNG_CHUNK_WRITE_ERROR) - png_app_warning(png_ptr, message); - - else - png_app_error(png_ptr, message); - } -# endif -} - -#ifdef PNG_ERROR_TEXT_SUPPORTED -#ifdef PNG_FLOATING_POINT_SUPPORTED -PNG_FUNCTION(void, -png_fixed_error,(png_const_structrp png_ptr, png_const_charp name),PNG_NORETURN) -{ -# define fixed_message "fixed point overflow in " -# define fixed_message_ln ((sizeof fixed_message)-1) - unsigned int iin; - char msg[fixed_message_ln+PNG_MAX_ERROR_TEXT]; - memcpy(msg, fixed_message, fixed_message_ln); - iin = 0; - if (name != NULL) - while (iin < (PNG_MAX_ERROR_TEXT-1) && name[iin] != 0) - { - msg[fixed_message_ln + iin] = name[iin]; - ++iin; - } - msg[fixed_message_ln + iin] = 0; - png_error(png_ptr, msg); -} -#endif -#endif - -#ifdef PNG_SETJMP_SUPPORTED -/* This API only exists if ANSI-C style error handling is used, - * otherwise it is necessary for png_default_error to be overridden. - */ -jmp_buf* PNGAPI -png_set_longjmp_fn(png_structrp png_ptr, png_longjmp_ptr longjmp_fn, - size_t jmp_buf_size) -{ - /* From libpng 1.6.0 the app gets one chance to set a 'jmpbuf_size' value - * and it must not change after that. Libpng doesn't care how big the - * buffer is, just that it doesn't change. - * - * If the buffer size is no *larger* than the size of jmp_buf when libpng is - * compiled a built in jmp_buf is returned; this preserves the pre-1.6.0 - * semantics that this call will not fail. If the size is larger, however, - * the buffer is allocated and this may fail, causing the function to return - * NULL. - */ - if (png_ptr == NULL) - return NULL; - - if (png_ptr->jmp_buf_ptr == NULL) - { - png_ptr->jmp_buf_size = 0; /* not allocated */ - - if (jmp_buf_size <= (sizeof png_ptr->jmp_buf_local)) - png_ptr->jmp_buf_ptr = &png_ptr->jmp_buf_local; - - else - { - png_ptr->jmp_buf_ptr = png_voidcast(jmp_buf *, - png_malloc_warn(png_ptr, jmp_buf_size)); - - if (png_ptr->jmp_buf_ptr == NULL) - return NULL; /* new NULL return on OOM */ - - png_ptr->jmp_buf_size = jmp_buf_size; - } - } - - else /* Already allocated: check the size */ - { - size_t size = png_ptr->jmp_buf_size; - - if (size == 0) - { - size = (sizeof png_ptr->jmp_buf_local); - if (png_ptr->jmp_buf_ptr != &png_ptr->jmp_buf_local) - { - /* This is an internal error in libpng: somehow we have been left - * with a stack allocated jmp_buf when the application regained - * control. It's always possible to fix this up, but for the moment - * this is a png_error because that makes it easy to detect. - */ - png_error(png_ptr, "Libpng jmp_buf still allocated"); - /* png_ptr->jmp_buf_ptr = &png_ptr->jmp_buf_local; */ - } - } - - if (size != jmp_buf_size) - { - png_warning(png_ptr, "Application jmp_buf size changed"); - return NULL; /* caller will probably crash: no choice here */ - } - } - - /* Finally fill in the function, now we have a satisfactory buffer. It is - * valid to change the function on every call. - */ - png_ptr->longjmp_fn = longjmp_fn; - return png_ptr->jmp_buf_ptr; -} - -void /* PRIVATE */ -png_free_jmpbuf(png_structrp png_ptr) -{ - if (png_ptr != NULL) - { - jmp_buf *jb = png_ptr->jmp_buf_ptr; - - /* A size of 0 is used to indicate a local, stack, allocation of the - * pointer; used here and in png.c - */ - if (jb != NULL && png_ptr->jmp_buf_size > 0) - { - - /* This stuff is so that a failure to free the error control structure - * does not leave libpng in a state with no valid error handling: the - * free always succeeds, if there is an error it gets ignored. - */ - if (jb != &png_ptr->jmp_buf_local) - { - /* Make an internal, libpng, jmp_buf to return here */ - jmp_buf free_jmp_buf; - - if (!setjmp(free_jmp_buf)) - { - png_ptr->jmp_buf_ptr = &free_jmp_buf; /* come back here */ - png_ptr->jmp_buf_size = 0; /* stack allocation */ - png_ptr->longjmp_fn = longjmp; - png_free(png_ptr, jb); /* Return to setjmp on error */ - } - } - } - - /* *Always* cancel everything out: */ - png_ptr->jmp_buf_size = 0; - png_ptr->jmp_buf_ptr = NULL; - png_ptr->longjmp_fn = 0; - } -} -#endif - -/* This is the default error handling function. Note that replacements for - * this function MUST NOT RETURN, or the program will likely crash. This - * function is used by default, or if the program supplies NULL for the - * error function pointer in png_set_error_fn(). - */ -static PNG_FUNCTION(void /* PRIVATE */, -png_default_error,(png_const_structrp png_ptr, png_const_charp error_message), - PNG_NORETURN) -{ -#ifdef PNG_CONSOLE_IO_SUPPORTED -#ifdef PNG_ERROR_NUMBERS_SUPPORTED - /* Check on NULL only added in 1.5.4 */ - if (error_message != NULL && *error_message == PNG_LITERAL_SHARP) - { - /* Strip "#nnnn " from beginning of error message. */ - int offset; - char error_number[16]; - for (offset = 0; offset<15; offset++) - { - error_number[offset] = error_message[offset + 1]; - if (error_message[offset] == ' ') - break; - } - - if ((offset > 1) && (offset < 15)) - { - error_number[offset - 1] = '\0'; - fprintf(stderr, "libpng error no. %s: %s", - error_number, error_message + offset + 1); - fprintf(stderr, PNG_STRING_NEWLINE); - } - - else - { - fprintf(stderr, "libpng error: %s, offset=%d", - error_message, offset); - fprintf(stderr, PNG_STRING_NEWLINE); - } - } - else -#endif - { - fprintf(stderr, "libpng error: %s", error_message ? error_message : - "undefined"); - fprintf(stderr, PNG_STRING_NEWLINE); - } -#else - PNG_UNUSED(error_message) /* Make compiler happy */ -#endif - png_longjmp(png_ptr, 1); -} - -PNG_FUNCTION(void,PNGAPI -png_longjmp,(png_const_structrp png_ptr, int val),PNG_NORETURN) -{ -#ifdef PNG_SETJMP_SUPPORTED - if (png_ptr != NULL && png_ptr->longjmp_fn != NULL && - png_ptr->jmp_buf_ptr != NULL) - png_ptr->longjmp_fn(*png_ptr->jmp_buf_ptr, val); -#else - PNG_UNUSED(png_ptr) - PNG_UNUSED(val) -#endif - - /* If control reaches this point, png_longjmp() must not return. The only - * choice is to terminate the whole process (or maybe the thread); to do - * this the ANSI-C abort() function is used unless a different method is - * implemented by overriding the default configuration setting for - * PNG_ABORT(). - */ - PNG_ABORT(); -} - -#ifdef PNG_WARNINGS_SUPPORTED -/* This function is called when there is a warning, but the library thinks - * it can continue anyway. Replacement functions don't have to do anything - * here if you don't want them to. In the default configuration, png_ptr is - * not used, but it is passed in case it may be useful. - */ -static void /* PRIVATE */ -png_default_warning(png_const_structrp png_ptr, png_const_charp warning_message) -{ -#ifdef PNG_CONSOLE_IO_SUPPORTED -# ifdef PNG_ERROR_NUMBERS_SUPPORTED - if (*warning_message == PNG_LITERAL_SHARP) - { - int offset; - char warning_number[16]; - for (offset = 0; offset < 15; offset++) - { - warning_number[offset] = warning_message[offset + 1]; - if (warning_message[offset] == ' ') - break; - } - - if ((offset > 1) && (offset < 15)) - { - warning_number[offset + 1] = '\0'; - fprintf(stderr, "libpng warning no. %s: %s", - warning_number, warning_message + offset); - fprintf(stderr, PNG_STRING_NEWLINE); - } - - else - { - fprintf(stderr, "libpng warning: %s", - warning_message); - fprintf(stderr, PNG_STRING_NEWLINE); - } - } - else -# endif - - { - fprintf(stderr, "libpng warning: %s", warning_message); - fprintf(stderr, PNG_STRING_NEWLINE); - } -#else - PNG_UNUSED(warning_message) /* Make compiler happy */ -#endif - PNG_UNUSED(png_ptr) /* Make compiler happy */ -} -#endif /* WARNINGS */ - -/* This function is called when the application wants to use another method - * of handling errors and warnings. Note that the error function MUST NOT - * return to the calling routine or serious problems will occur. The return - * method used in the default routine calls longjmp(png_ptr->jmp_buf_ptr, 1) - */ -void PNGAPI -png_set_error_fn(png_structrp png_ptr, png_voidp error_ptr, - png_error_ptr error_fn, png_error_ptr warning_fn) -{ - if (png_ptr == NULL) - return; - - png_ptr->error_ptr = error_ptr; - png_ptr->error_fn = error_fn; -#ifdef PNG_WARNINGS_SUPPORTED - png_ptr->warning_fn = warning_fn; -#else - PNG_UNUSED(warning_fn) -#endif -} - - -/* This function returns a pointer to the error_ptr associated with the user - * functions. The application should free any memory associated with this - * pointer before png_write_destroy and png_read_destroy are called. - */ -png_voidp PNGAPI -png_get_error_ptr(png_const_structrp png_ptr) -{ - if (png_ptr == NULL) - return NULL; - - return (png_voidp)png_ptr->error_ptr; -} - - -#ifdef PNG_ERROR_NUMBERS_SUPPORTED -void PNGAPI -png_set_strip_error_numbers(png_structrp png_ptr, png_uint_32 strip_mode) -{ - if (png_ptr != NULL) - { - png_ptr->flags &= - ((~(PNG_FLAG_STRIP_ERROR_NUMBERS | - PNG_FLAG_STRIP_ERROR_TEXT))&strip_mode); - } -} -#endif - -#if defined(PNG_SIMPLIFIED_READ_SUPPORTED) ||\ - defined(PNG_SIMPLIFIED_WRITE_SUPPORTED) - /* Currently the above both depend on SETJMP_SUPPORTED, however it would be - * possible to implement without setjmp support just so long as there is some - * way to handle the error return here: - */ -PNG_FUNCTION(void /* PRIVATE */, (PNGCBAPI -png_safe_error),(png_structp png_nonconst_ptr, png_const_charp error_message), - PNG_NORETURN) -{ - png_const_structrp png_ptr = png_nonconst_ptr; - png_imagep image = png_voidcast(png_imagep, png_ptr->error_ptr); - - /* An error is always logged here, overwriting anything (typically a warning) - * that is already there: - */ - if (image != NULL) - { - png_safecat(image->message, (sizeof image->message), 0, error_message); - image->warning_or_error |= PNG_IMAGE_ERROR; - - /* Retrieve the jmp_buf from within the png_control, making this work for - * C++ compilation too is pretty tricky: C++ wants a pointer to the first - * element of a jmp_buf, but C doesn't tell us the type of that. - */ - if (image->opaque != NULL && image->opaque->error_buf != NULL) - longjmp(png_control_jmp_buf(image->opaque), 1); - - /* Missing longjmp buffer, the following is to help debugging: */ - { - size_t pos = png_safecat(image->message, (sizeof image->message), 0, - "bad longjmp: "); - png_safecat(image->message, (sizeof image->message), pos, - error_message); - } - } - - /* Here on an internal programming error. */ - abort(); -} - -#ifdef PNG_WARNINGS_SUPPORTED -void /* PRIVATE */ PNGCBAPI -png_safe_warning(png_structp png_nonconst_ptr, png_const_charp warning_message) -{ - png_const_structrp png_ptr = png_nonconst_ptr; - png_imagep image = png_voidcast(png_imagep, png_ptr->error_ptr); - - /* A warning is only logged if there is no prior warning or error. */ - if (image->warning_or_error == 0) - { - png_safecat(image->message, (sizeof image->message), 0, warning_message); - image->warning_or_error |= PNG_IMAGE_WARNING; - } -} -#endif - -int /* PRIVATE */ -png_safe_execute(png_imagep image, int (*function)(png_voidp), png_voidp arg) -{ - png_voidp saved_error_buf = image->opaque->error_buf; - jmp_buf safe_jmpbuf; - int result; - - /* Safely execute function(arg), with png_error returning back here. */ - if (setjmp(safe_jmpbuf) == 0) - { - image->opaque->error_buf = safe_jmpbuf; - result = function(arg); - image->opaque->error_buf = saved_error_buf; - return result; - } - - /* On png_error, return via longjmp, pop the jmpbuf, and free the image. */ - image->opaque->error_buf = saved_error_buf; - png_image_free(image); - return 0; -} -#endif /* SIMPLIFIED READ || SIMPLIFIED_WRITE */ -#endif /* READ || WRITE */ diff --git a/dep/libpng/src/pngget.c b/dep/libpng/src/pngget.c deleted file mode 100644 index 1084b268f..000000000 --- a/dep/libpng/src/pngget.c +++ /dev/null @@ -1,1267 +0,0 @@ - -/* pngget.c - retrieval of values from info struct - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - * - */ - -#include "pngpriv.h" - -#if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) - -png_uint_32 PNGAPI -png_get_valid(png_const_structrp png_ptr, png_const_inforp info_ptr, - png_uint_32 flag) -{ - if (png_ptr != NULL && info_ptr != NULL) - { -#ifdef PNG_READ_tRNS_SUPPORTED - /* png_handle_PLTE() may have canceled a valid tRNS chunk but left the - * 'valid' flag for the detection of duplicate chunks. Do not report a - * valid tRNS chunk in this case. - */ - if (flag == PNG_INFO_tRNS && png_ptr->num_trans == 0) - return 0; -#endif - - return info_ptr->valid & flag; - } - - return 0; -} - -size_t PNGAPI -png_get_rowbytes(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - if (png_ptr != NULL && info_ptr != NULL) - return info_ptr->rowbytes; - - return 0; -} - -#ifdef PNG_INFO_IMAGE_SUPPORTED -png_bytepp PNGAPI -png_get_rows(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - if (png_ptr != NULL && info_ptr != NULL) - return info_ptr->row_pointers; - - return 0; -} -#endif - -#ifdef PNG_EASY_ACCESS_SUPPORTED -/* Easy access to info, added in libpng-0.99 */ -png_uint_32 PNGAPI -png_get_image_width(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - if (png_ptr != NULL && info_ptr != NULL) - return info_ptr->width; - - return 0; -} - -png_uint_32 PNGAPI -png_get_image_height(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - if (png_ptr != NULL && info_ptr != NULL) - return info_ptr->height; - - return 0; -} - -png_byte PNGAPI -png_get_bit_depth(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - if (png_ptr != NULL && info_ptr != NULL) - return info_ptr->bit_depth; - - return 0; -} - -png_byte PNGAPI -png_get_color_type(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - if (png_ptr != NULL && info_ptr != NULL) - return info_ptr->color_type; - - return 0; -} - -png_byte PNGAPI -png_get_filter_type(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - if (png_ptr != NULL && info_ptr != NULL) - return info_ptr->filter_type; - - return 0; -} - -png_byte PNGAPI -png_get_interlace_type(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - if (png_ptr != NULL && info_ptr != NULL) - return info_ptr->interlace_type; - - return 0; -} - -png_byte PNGAPI -png_get_compression_type(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - if (png_ptr != NULL && info_ptr != NULL) - return info_ptr->compression_type; - - return 0; -} - -png_uint_32 PNGAPI -png_get_x_pixels_per_meter(png_const_structrp png_ptr, png_const_inforp - info_ptr) -{ -#ifdef PNG_pHYs_SUPPORTED - png_debug(1, "in png_get_x_pixels_per_meter"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_pHYs) != 0) - { - if (info_ptr->phys_unit_type == PNG_RESOLUTION_METER) - return info_ptr->x_pixels_per_unit; - } -#else - PNG_UNUSED(png_ptr) - PNG_UNUSED(info_ptr) -#endif - - return 0; -} - -png_uint_32 PNGAPI -png_get_y_pixels_per_meter(png_const_structrp png_ptr, png_const_inforp - info_ptr) -{ -#ifdef PNG_pHYs_SUPPORTED - png_debug(1, "in png_get_y_pixels_per_meter"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_pHYs) != 0) - { - if (info_ptr->phys_unit_type == PNG_RESOLUTION_METER) - return info_ptr->y_pixels_per_unit; - } -#else - PNG_UNUSED(png_ptr) - PNG_UNUSED(info_ptr) -#endif - - return 0; -} - -png_uint_32 PNGAPI -png_get_pixels_per_meter(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ -#ifdef PNG_pHYs_SUPPORTED - png_debug(1, "in png_get_pixels_per_meter"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_pHYs) != 0) - { - if (info_ptr->phys_unit_type == PNG_RESOLUTION_METER && - info_ptr->x_pixels_per_unit == info_ptr->y_pixels_per_unit) - return info_ptr->x_pixels_per_unit; - } -#else - PNG_UNUSED(png_ptr) - PNG_UNUSED(info_ptr) -#endif - - return 0; -} - -#ifdef PNG_FLOATING_POINT_SUPPORTED -float PNGAPI -png_get_pixel_aspect_ratio(png_const_structrp png_ptr, png_const_inforp - info_ptr) -{ -#ifdef PNG_READ_pHYs_SUPPORTED - png_debug(1, "in png_get_pixel_aspect_ratio"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_pHYs) != 0) - { - if (info_ptr->x_pixels_per_unit != 0) - return (float)info_ptr->y_pixels_per_unit - / (float)info_ptr->x_pixels_per_unit; - } -#else - PNG_UNUSED(png_ptr) - PNG_UNUSED(info_ptr) -#endif - - return (float)0.0; -} -#endif - -#ifdef PNG_FIXED_POINT_SUPPORTED -png_fixed_point PNGAPI -png_get_pixel_aspect_ratio_fixed(png_const_structrp png_ptr, - png_const_inforp info_ptr) -{ -#ifdef PNG_READ_pHYs_SUPPORTED - png_debug(1, "in png_get_pixel_aspect_ratio_fixed"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_pHYs) != 0 && - info_ptr->x_pixels_per_unit > 0 && info_ptr->y_pixels_per_unit > 0 && - info_ptr->x_pixels_per_unit <= PNG_UINT_31_MAX && - info_ptr->y_pixels_per_unit <= PNG_UINT_31_MAX) - { - png_fixed_point res; - - /* The following casts work because a PNG 4 byte integer only has a valid - * range of 0..2^31-1; otherwise the cast might overflow. - */ - if (png_muldiv(&res, (png_int_32)info_ptr->y_pixels_per_unit, PNG_FP_1, - (png_int_32)info_ptr->x_pixels_per_unit) != 0) - return res; - } -#else - PNG_UNUSED(png_ptr) - PNG_UNUSED(info_ptr) -#endif - - return 0; -} -#endif - -png_int_32 PNGAPI -png_get_x_offset_microns(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ -#ifdef PNG_oFFs_SUPPORTED - png_debug(1, "in png_get_x_offset_microns"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_oFFs) != 0) - { - if (info_ptr->offset_unit_type == PNG_OFFSET_MICROMETER) - return info_ptr->x_offset; - } -#else - PNG_UNUSED(png_ptr) - PNG_UNUSED(info_ptr) -#endif - - return 0; -} - -png_int_32 PNGAPI -png_get_y_offset_microns(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ -#ifdef PNG_oFFs_SUPPORTED - png_debug(1, "in png_get_y_offset_microns"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_oFFs) != 0) - { - if (info_ptr->offset_unit_type == PNG_OFFSET_MICROMETER) - return info_ptr->y_offset; - } -#else - PNG_UNUSED(png_ptr) - PNG_UNUSED(info_ptr) -#endif - - return 0; -} - -png_int_32 PNGAPI -png_get_x_offset_pixels(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ -#ifdef PNG_oFFs_SUPPORTED - png_debug(1, "in png_get_x_offset_pixels"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_oFFs) != 0) - { - if (info_ptr->offset_unit_type == PNG_OFFSET_PIXEL) - return info_ptr->x_offset; - } -#else - PNG_UNUSED(png_ptr) - PNG_UNUSED(info_ptr) -#endif - - return 0; -} - -png_int_32 PNGAPI -png_get_y_offset_pixels(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ -#ifdef PNG_oFFs_SUPPORTED - png_debug(1, "in png_get_y_offset_pixels"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_oFFs) != 0) - { - if (info_ptr->offset_unit_type == PNG_OFFSET_PIXEL) - return info_ptr->y_offset; - } -#else - PNG_UNUSED(png_ptr) - PNG_UNUSED(info_ptr) -#endif - - return 0; -} - -#ifdef PNG_INCH_CONVERSIONS_SUPPORTED -static png_uint_32 -ppi_from_ppm(png_uint_32 ppm) -{ -#if 0 - /* The conversion is *(2.54/100), in binary (32 digits): - * .00000110100000001001110101001001 - */ - png_uint_32 t1001, t1101; - ppm >>= 1; /* .1 */ - t1001 = ppm + (ppm >> 3); /* .1001 */ - t1101 = t1001 + (ppm >> 1); /* .1101 */ - ppm >>= 20; /* .000000000000000000001 */ - t1101 += t1101 >> 15; /* .1101000000000001101 */ - t1001 >>= 11; /* .000000000001001 */ - t1001 += t1001 >> 12; /* .000000000001001000000001001 */ - ppm += t1001; /* .000000000001001000001001001 */ - ppm += t1101; /* .110100000001001110101001001 */ - return (ppm + 16) >> 5;/* .00000110100000001001110101001001 */ -#else - /* The argument is a PNG unsigned integer, so it is not permitted - * to be bigger than 2^31. - */ - png_fixed_point result; - if (ppm <= PNG_UINT_31_MAX && png_muldiv(&result, (png_int_32)ppm, 127, - 5000) != 0) - return (png_uint_32)result; - - /* Overflow. */ - return 0; -#endif -} - -png_uint_32 PNGAPI -png_get_pixels_per_inch(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - return ppi_from_ppm(png_get_pixels_per_meter(png_ptr, info_ptr)); -} - -png_uint_32 PNGAPI -png_get_x_pixels_per_inch(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - return ppi_from_ppm(png_get_x_pixels_per_meter(png_ptr, info_ptr)); -} - -png_uint_32 PNGAPI -png_get_y_pixels_per_inch(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - return ppi_from_ppm(png_get_y_pixels_per_meter(png_ptr, info_ptr)); -} - -#ifdef PNG_FIXED_POINT_SUPPORTED -static png_fixed_point -png_fixed_inches_from_microns(png_const_structrp png_ptr, png_int_32 microns) -{ - /* Convert from meters * 1,000,000 to inches * 100,000, meters to - * inches is simply *(100/2.54), so we want *(10/2.54) == 500/127. - * Notice that this can overflow - a warning is output and 0 is - * returned. - */ - return png_muldiv_warn(png_ptr, microns, 500, 127); -} - -png_fixed_point PNGAPI -png_get_x_offset_inches_fixed(png_const_structrp png_ptr, - png_const_inforp info_ptr) -{ - return png_fixed_inches_from_microns(png_ptr, - png_get_x_offset_microns(png_ptr, info_ptr)); -} -#endif - -#ifdef PNG_FIXED_POINT_SUPPORTED -png_fixed_point PNGAPI -png_get_y_offset_inches_fixed(png_const_structrp png_ptr, - png_const_inforp info_ptr) -{ - return png_fixed_inches_from_microns(png_ptr, - png_get_y_offset_microns(png_ptr, info_ptr)); -} -#endif - -#ifdef PNG_FLOATING_POINT_SUPPORTED -float PNGAPI -png_get_x_offset_inches(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - /* To avoid the overflow do the conversion directly in floating - * point. - */ - return (float)(png_get_x_offset_microns(png_ptr, info_ptr) * .00003937); -} -#endif - -#ifdef PNG_FLOATING_POINT_SUPPORTED -float PNGAPI -png_get_y_offset_inches(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - /* To avoid the overflow do the conversion directly in floating - * point. - */ - return (float)(png_get_y_offset_microns(png_ptr, info_ptr) * .00003937); -} -#endif - -#ifdef PNG_pHYs_SUPPORTED -png_uint_32 PNGAPI -png_get_pHYs_dpi(png_const_structrp png_ptr, png_const_inforp info_ptr, - png_uint_32 *res_x, png_uint_32 *res_y, int *unit_type) -{ - png_uint_32 retval = 0; - - png_debug1(1, "in %s retrieval function", "pHYs"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_pHYs) != 0) - { - if (res_x != NULL) - { - *res_x = info_ptr->x_pixels_per_unit; - retval |= PNG_INFO_pHYs; - } - - if (res_y != NULL) - { - *res_y = info_ptr->y_pixels_per_unit; - retval |= PNG_INFO_pHYs; - } - - if (unit_type != NULL) - { - *unit_type = (int)info_ptr->phys_unit_type; - retval |= PNG_INFO_pHYs; - - if (*unit_type == 1) - { - if (res_x != NULL) *res_x = (png_uint_32)(*res_x * .0254 + .50); - if (res_y != NULL) *res_y = (png_uint_32)(*res_y * .0254 + .50); - } - } - } - - return retval; -} -#endif /* pHYs */ -#endif /* INCH_CONVERSIONS */ - -/* png_get_channels really belongs in here, too, but it's been around longer */ - -#endif /* EASY_ACCESS */ - - -png_byte PNGAPI -png_get_channels(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - if (png_ptr != NULL && info_ptr != NULL) - return info_ptr->channels; - - return 0; -} - -#ifdef PNG_READ_SUPPORTED -png_const_bytep PNGAPI -png_get_signature(png_const_structrp png_ptr, png_const_inforp info_ptr) -{ - if (png_ptr != NULL && info_ptr != NULL) - return info_ptr->signature; - - return NULL; -} -#endif - -#ifdef PNG_bKGD_SUPPORTED -png_uint_32 PNGAPI -png_get_bKGD(png_const_structrp png_ptr, png_inforp info_ptr, - png_color_16p *background) -{ - png_debug1(1, "in %s retrieval function", "bKGD"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_bKGD) != 0 && - background != NULL) - { - *background = &(info_ptr->background); - return PNG_INFO_bKGD; - } - - return 0; -} -#endif - -#ifdef PNG_cHRM_SUPPORTED -/* The XYZ APIs were added in 1.5.5 to take advantage of the code added at the - * same time to correct the rgb grayscale coefficient defaults obtained from the - * cHRM chunk in 1.5.4 - */ -# ifdef PNG_FLOATING_POINT_SUPPORTED -png_uint_32 PNGAPI -png_get_cHRM(png_const_structrp png_ptr, png_const_inforp info_ptr, - double *white_x, double *white_y, double *red_x, double *red_y, - double *green_x, double *green_y, double *blue_x, double *blue_y) -{ - png_debug1(1, "in %s retrieval function", "cHRM"); - - /* Quiet API change: this code used to only return the end points if a cHRM - * chunk was present, but the end points can also come from iCCP or sRGB - * chunks, so in 1.6.0 the png_get_ APIs return the end points regardless and - * the png_set_ APIs merely check that set end points are mutually - * consistent. - */ - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_ENDPOINTS) != 0) - { - if (white_x != NULL) - *white_x = png_float(png_ptr, - info_ptr->colorspace.end_points_xy.whitex, "cHRM white X"); - if (white_y != NULL) - *white_y = png_float(png_ptr, - info_ptr->colorspace.end_points_xy.whitey, "cHRM white Y"); - if (red_x != NULL) - *red_x = png_float(png_ptr, info_ptr->colorspace.end_points_xy.redx, - "cHRM red X"); - if (red_y != NULL) - *red_y = png_float(png_ptr, info_ptr->colorspace.end_points_xy.redy, - "cHRM red Y"); - if (green_x != NULL) - *green_x = png_float(png_ptr, - info_ptr->colorspace.end_points_xy.greenx, "cHRM green X"); - if (green_y != NULL) - *green_y = png_float(png_ptr, - info_ptr->colorspace.end_points_xy.greeny, "cHRM green Y"); - if (blue_x != NULL) - *blue_x = png_float(png_ptr, info_ptr->colorspace.end_points_xy.bluex, - "cHRM blue X"); - if (blue_y != NULL) - *blue_y = png_float(png_ptr, info_ptr->colorspace.end_points_xy.bluey, - "cHRM blue Y"); - return PNG_INFO_cHRM; - } - - return 0; -} - -png_uint_32 PNGAPI -png_get_cHRM_XYZ(png_const_structrp png_ptr, png_const_inforp info_ptr, - double *red_X, double *red_Y, double *red_Z, double *green_X, - double *green_Y, double *green_Z, double *blue_X, double *blue_Y, - double *blue_Z) -{ - png_debug1(1, "in %s retrieval function", "cHRM_XYZ(float)"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_ENDPOINTS) != 0) - { - if (red_X != NULL) - *red_X = png_float(png_ptr, info_ptr->colorspace.end_points_XYZ.red_X, - "cHRM red X"); - if (red_Y != NULL) - *red_Y = png_float(png_ptr, info_ptr->colorspace.end_points_XYZ.red_Y, - "cHRM red Y"); - if (red_Z != NULL) - *red_Z = png_float(png_ptr, info_ptr->colorspace.end_points_XYZ.red_Z, - "cHRM red Z"); - if (green_X != NULL) - *green_X = png_float(png_ptr, - info_ptr->colorspace.end_points_XYZ.green_X, "cHRM green X"); - if (green_Y != NULL) - *green_Y = png_float(png_ptr, - info_ptr->colorspace.end_points_XYZ.green_Y, "cHRM green Y"); - if (green_Z != NULL) - *green_Z = png_float(png_ptr, - info_ptr->colorspace.end_points_XYZ.green_Z, "cHRM green Z"); - if (blue_X != NULL) - *blue_X = png_float(png_ptr, - info_ptr->colorspace.end_points_XYZ.blue_X, "cHRM blue X"); - if (blue_Y != NULL) - *blue_Y = png_float(png_ptr, - info_ptr->colorspace.end_points_XYZ.blue_Y, "cHRM blue Y"); - if (blue_Z != NULL) - *blue_Z = png_float(png_ptr, - info_ptr->colorspace.end_points_XYZ.blue_Z, "cHRM blue Z"); - return PNG_INFO_cHRM; - } - - return 0; -} -# endif - -# ifdef PNG_FIXED_POINT_SUPPORTED -png_uint_32 PNGAPI -png_get_cHRM_XYZ_fixed(png_const_structrp png_ptr, png_const_inforp info_ptr, - png_fixed_point *int_red_X, png_fixed_point *int_red_Y, - png_fixed_point *int_red_Z, png_fixed_point *int_green_X, - png_fixed_point *int_green_Y, png_fixed_point *int_green_Z, - png_fixed_point *int_blue_X, png_fixed_point *int_blue_Y, - png_fixed_point *int_blue_Z) -{ - png_debug1(1, "in %s retrieval function", "cHRM_XYZ"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_ENDPOINTS) != 0) - { - if (int_red_X != NULL) - *int_red_X = info_ptr->colorspace.end_points_XYZ.red_X; - if (int_red_Y != NULL) - *int_red_Y = info_ptr->colorspace.end_points_XYZ.red_Y; - if (int_red_Z != NULL) - *int_red_Z = info_ptr->colorspace.end_points_XYZ.red_Z; - if (int_green_X != NULL) - *int_green_X = info_ptr->colorspace.end_points_XYZ.green_X; - if (int_green_Y != NULL) - *int_green_Y = info_ptr->colorspace.end_points_XYZ.green_Y; - if (int_green_Z != NULL) - *int_green_Z = info_ptr->colorspace.end_points_XYZ.green_Z; - if (int_blue_X != NULL) - *int_blue_X = info_ptr->colorspace.end_points_XYZ.blue_X; - if (int_blue_Y != NULL) - *int_blue_Y = info_ptr->colorspace.end_points_XYZ.blue_Y; - if (int_blue_Z != NULL) - *int_blue_Z = info_ptr->colorspace.end_points_XYZ.blue_Z; - return PNG_INFO_cHRM; - } - - return 0; -} - -png_uint_32 PNGAPI -png_get_cHRM_fixed(png_const_structrp png_ptr, png_const_inforp info_ptr, - png_fixed_point *white_x, png_fixed_point *white_y, png_fixed_point *red_x, - png_fixed_point *red_y, png_fixed_point *green_x, png_fixed_point *green_y, - png_fixed_point *blue_x, png_fixed_point *blue_y) -{ - png_debug1(1, "in %s retrieval function", "cHRM"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_ENDPOINTS) != 0) - { - if (white_x != NULL) - *white_x = info_ptr->colorspace.end_points_xy.whitex; - if (white_y != NULL) - *white_y = info_ptr->colorspace.end_points_xy.whitey; - if (red_x != NULL) - *red_x = info_ptr->colorspace.end_points_xy.redx; - if (red_y != NULL) - *red_y = info_ptr->colorspace.end_points_xy.redy; - if (green_x != NULL) - *green_x = info_ptr->colorspace.end_points_xy.greenx; - if (green_y != NULL) - *green_y = info_ptr->colorspace.end_points_xy.greeny; - if (blue_x != NULL) - *blue_x = info_ptr->colorspace.end_points_xy.bluex; - if (blue_y != NULL) - *blue_y = info_ptr->colorspace.end_points_xy.bluey; - return PNG_INFO_cHRM; - } - - return 0; -} -# endif -#endif - -#ifdef PNG_gAMA_SUPPORTED -# ifdef PNG_FIXED_POINT_SUPPORTED -png_uint_32 PNGAPI -png_get_gAMA_fixed(png_const_structrp png_ptr, png_const_inforp info_ptr, - png_fixed_point *file_gamma) -{ - png_debug1(1, "in %s retrieval function", "gAMA"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_GAMMA) != 0 && - file_gamma != NULL) - { - *file_gamma = info_ptr->colorspace.gamma; - return PNG_INFO_gAMA; - } - - return 0; -} -# endif - -# ifdef PNG_FLOATING_POINT_SUPPORTED -png_uint_32 PNGAPI -png_get_gAMA(png_const_structrp png_ptr, png_const_inforp info_ptr, - double *file_gamma) -{ - png_debug1(1, "in %s retrieval function", "gAMA(float)"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_GAMMA) != 0 && - file_gamma != NULL) - { - *file_gamma = png_float(png_ptr, info_ptr->colorspace.gamma, - "png_get_gAMA"); - return PNG_INFO_gAMA; - } - - return 0; -} -# endif -#endif - -#ifdef PNG_sRGB_SUPPORTED -png_uint_32 PNGAPI -png_get_sRGB(png_const_structrp png_ptr, png_const_inforp info_ptr, - int *file_srgb_intent) -{ - png_debug1(1, "in %s retrieval function", "sRGB"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_sRGB) != 0 && file_srgb_intent != NULL) - { - *file_srgb_intent = info_ptr->colorspace.rendering_intent; - return PNG_INFO_sRGB; - } - - return 0; -} -#endif - -#ifdef PNG_iCCP_SUPPORTED -png_uint_32 PNGAPI -png_get_iCCP(png_const_structrp png_ptr, png_inforp info_ptr, - png_charpp name, int *compression_type, - png_bytepp profile, png_uint_32 *proflen) -{ - png_debug1(1, "in %s retrieval function", "iCCP"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_iCCP) != 0 && - name != NULL && profile != NULL && proflen != NULL) - { - *name = info_ptr->iccp_name; - *profile = info_ptr->iccp_profile; - *proflen = png_get_uint_32(info_ptr->iccp_profile); - /* This is somewhat irrelevant since the profile data returned has - * actually been uncompressed. - */ - if (compression_type != NULL) - *compression_type = PNG_COMPRESSION_TYPE_BASE; - return PNG_INFO_iCCP; - } - - return 0; - -} -#endif - -#ifdef PNG_sPLT_SUPPORTED -int PNGAPI -png_get_sPLT(png_const_structrp png_ptr, png_inforp info_ptr, - png_sPLT_tpp spalettes) -{ - png_debug1(1, "in %s retrieval function", "sPLT"); - - if (png_ptr != NULL && info_ptr != NULL && spalettes != NULL) - { - *spalettes = info_ptr->splt_palettes; - return info_ptr->splt_palettes_num; - } - - return 0; -} -#endif - -#ifdef PNG_eXIf_SUPPORTED -png_uint_32 PNGAPI -png_get_eXIf(png_const_structrp png_ptr, png_inforp info_ptr, - png_bytep *exif) -{ - png_warning(png_ptr, "png_get_eXIf does not work; use png_get_eXIf_1"); - PNG_UNUSED(info_ptr) - PNG_UNUSED(exif) - return 0; -} - -png_uint_32 PNGAPI -png_get_eXIf_1(png_const_structrp png_ptr, png_const_inforp info_ptr, - png_uint_32 *num_exif, png_bytep *exif) -{ - png_debug1(1, "in %s retrieval function", "eXIf"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_eXIf) != 0 && exif != NULL) - { - *num_exif = info_ptr->num_exif; - *exif = info_ptr->exif; - return PNG_INFO_eXIf; - } - - return 0; -} -#endif - -#ifdef PNG_hIST_SUPPORTED -png_uint_32 PNGAPI -png_get_hIST(png_const_structrp png_ptr, png_inforp info_ptr, - png_uint_16p *hist) -{ - png_debug1(1, "in %s retrieval function", "hIST"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_hIST) != 0 && hist != NULL) - { - *hist = info_ptr->hist; - return PNG_INFO_hIST; - } - - return 0; -} -#endif - -png_uint_32 PNGAPI -png_get_IHDR(png_const_structrp png_ptr, png_const_inforp info_ptr, - png_uint_32 *width, png_uint_32 *height, int *bit_depth, - int *color_type, int *interlace_type, int *compression_type, - int *filter_type) -{ - png_debug1(1, "in %s retrieval function", "IHDR"); - - if (png_ptr == NULL || info_ptr == NULL) - return 0; - - if (width != NULL) - *width = info_ptr->width; - - if (height != NULL) - *height = info_ptr->height; - - if (bit_depth != NULL) - *bit_depth = info_ptr->bit_depth; - - if (color_type != NULL) - *color_type = info_ptr->color_type; - - if (compression_type != NULL) - *compression_type = info_ptr->compression_type; - - if (filter_type != NULL) - *filter_type = info_ptr->filter_type; - - if (interlace_type != NULL) - *interlace_type = info_ptr->interlace_type; - - /* This is redundant if we can be sure that the info_ptr values were all - * assigned in png_set_IHDR(). We do the check anyhow in case an - * application has ignored our advice not to mess with the members - * of info_ptr directly. - */ - png_check_IHDR(png_ptr, info_ptr->width, info_ptr->height, - info_ptr->bit_depth, info_ptr->color_type, info_ptr->interlace_type, - info_ptr->compression_type, info_ptr->filter_type); - - return 1; -} - -#ifdef PNG_oFFs_SUPPORTED -png_uint_32 PNGAPI -png_get_oFFs(png_const_structrp png_ptr, png_const_inforp info_ptr, - png_int_32 *offset_x, png_int_32 *offset_y, int *unit_type) -{ - png_debug1(1, "in %s retrieval function", "oFFs"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_oFFs) != 0 && - offset_x != NULL && offset_y != NULL && unit_type != NULL) - { - *offset_x = info_ptr->x_offset; - *offset_y = info_ptr->y_offset; - *unit_type = (int)info_ptr->offset_unit_type; - return PNG_INFO_oFFs; - } - - return 0; -} -#endif - -#ifdef PNG_pCAL_SUPPORTED -png_uint_32 PNGAPI -png_get_pCAL(png_const_structrp png_ptr, png_inforp info_ptr, - png_charp *purpose, png_int_32 *X0, png_int_32 *X1, int *type, int *nparams, - png_charp *units, png_charpp *params) -{ - png_debug1(1, "in %s retrieval function", "pCAL"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_pCAL) != 0 && - purpose != NULL && X0 != NULL && X1 != NULL && type != NULL && - nparams != NULL && units != NULL && params != NULL) - { - *purpose = info_ptr->pcal_purpose; - *X0 = info_ptr->pcal_X0; - *X1 = info_ptr->pcal_X1; - *type = (int)info_ptr->pcal_type; - *nparams = (int)info_ptr->pcal_nparams; - *units = info_ptr->pcal_units; - *params = info_ptr->pcal_params; - return PNG_INFO_pCAL; - } - - return 0; -} -#endif - -#ifdef PNG_sCAL_SUPPORTED -# ifdef PNG_FIXED_POINT_SUPPORTED -# if defined(PNG_FLOATING_ARITHMETIC_SUPPORTED) || \ - defined(PNG_FLOATING_POINT_SUPPORTED) -png_uint_32 PNGAPI -png_get_sCAL_fixed(png_const_structrp png_ptr, png_const_inforp info_ptr, - int *unit, png_fixed_point *width, png_fixed_point *height) -{ - png_debug1(1, "in %s retrieval function", "sCAL"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_sCAL) != 0) - { - *unit = info_ptr->scal_unit; - /*TODO: make this work without FP support; the API is currently eliminated - * if neither floating point APIs nor internal floating point arithmetic - * are enabled. - */ - *width = png_fixed(png_ptr, atof(info_ptr->scal_s_width), "sCAL width"); - *height = png_fixed(png_ptr, atof(info_ptr->scal_s_height), - "sCAL height"); - return PNG_INFO_sCAL; - } - - return 0; -} -# endif /* FLOATING_ARITHMETIC */ -# endif /* FIXED_POINT */ -# ifdef PNG_FLOATING_POINT_SUPPORTED -png_uint_32 PNGAPI -png_get_sCAL(png_const_structrp png_ptr, png_const_inforp info_ptr, - int *unit, double *width, double *height) -{ - png_debug1(1, "in %s retrieval function", "sCAL(float)"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_sCAL) != 0) - { - *unit = info_ptr->scal_unit; - *width = atof(info_ptr->scal_s_width); - *height = atof(info_ptr->scal_s_height); - return PNG_INFO_sCAL; - } - - return 0; -} -# endif /* FLOATING POINT */ -png_uint_32 PNGAPI -png_get_sCAL_s(png_const_structrp png_ptr, png_const_inforp info_ptr, - int *unit, png_charpp width, png_charpp height) -{ - png_debug1(1, "in %s retrieval function", "sCAL(str)"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_sCAL) != 0) - { - *unit = info_ptr->scal_unit; - *width = info_ptr->scal_s_width; - *height = info_ptr->scal_s_height; - return PNG_INFO_sCAL; - } - - return 0; -} -#endif /* sCAL */ - -#ifdef PNG_pHYs_SUPPORTED -png_uint_32 PNGAPI -png_get_pHYs(png_const_structrp png_ptr, png_const_inforp info_ptr, - png_uint_32 *res_x, png_uint_32 *res_y, int *unit_type) -{ - png_uint_32 retval = 0; - - png_debug1(1, "in %s retrieval function", "pHYs"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_pHYs) != 0) - { - if (res_x != NULL) - { - *res_x = info_ptr->x_pixels_per_unit; - retval |= PNG_INFO_pHYs; - } - - if (res_y != NULL) - { - *res_y = info_ptr->y_pixels_per_unit; - retval |= PNG_INFO_pHYs; - } - - if (unit_type != NULL) - { - *unit_type = (int)info_ptr->phys_unit_type; - retval |= PNG_INFO_pHYs; - } - } - - return retval; -} -#endif /* pHYs */ - -png_uint_32 PNGAPI -png_get_PLTE(png_const_structrp png_ptr, png_inforp info_ptr, - png_colorp *palette, int *num_palette) -{ - png_debug1(1, "in %s retrieval function", "PLTE"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_PLTE) != 0 && palette != NULL) - { - *palette = info_ptr->palette; - *num_palette = info_ptr->num_palette; - png_debug1(3, "num_palette = %d", *num_palette); - return PNG_INFO_PLTE; - } - - return 0; -} - -#ifdef PNG_sBIT_SUPPORTED -png_uint_32 PNGAPI -png_get_sBIT(png_const_structrp png_ptr, png_inforp info_ptr, - png_color_8p *sig_bit) -{ - png_debug1(1, "in %s retrieval function", "sBIT"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_sBIT) != 0 && sig_bit != NULL) - { - *sig_bit = &(info_ptr->sig_bit); - return PNG_INFO_sBIT; - } - - return 0; -} -#endif - -#ifdef PNG_TEXT_SUPPORTED -int PNGAPI -png_get_text(png_const_structrp png_ptr, png_inforp info_ptr, - png_textp *text_ptr, int *num_text) -{ - if (png_ptr != NULL && info_ptr != NULL && info_ptr->num_text > 0) - { - png_debug1(1, "in text retrieval function, chunk typeid = 0x%lx", - (unsigned long)png_ptr->chunk_name); - - if (text_ptr != NULL) - *text_ptr = info_ptr->text; - - if (num_text != NULL) - *num_text = info_ptr->num_text; - - return info_ptr->num_text; - } - - if (num_text != NULL) - *num_text = 0; - - return 0; -} -#endif - -#ifdef PNG_tIME_SUPPORTED -png_uint_32 PNGAPI -png_get_tIME(png_const_structrp png_ptr, png_inforp info_ptr, - png_timep *mod_time) -{ - png_debug1(1, "in %s retrieval function", "tIME"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_tIME) != 0 && mod_time != NULL) - { - *mod_time = &(info_ptr->mod_time); - return PNG_INFO_tIME; - } - - return 0; -} -#endif - -#ifdef PNG_tRNS_SUPPORTED -png_uint_32 PNGAPI -png_get_tRNS(png_const_structrp png_ptr, png_inforp info_ptr, - png_bytep *trans_alpha, int *num_trans, png_color_16p *trans_color) -{ - png_uint_32 retval = 0; - - png_debug1(1, "in %s retrieval function", "tRNS"); - - if (png_ptr != NULL && info_ptr != NULL && - (info_ptr->valid & PNG_INFO_tRNS) != 0) - { - if (info_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - { - if (trans_alpha != NULL) - { - *trans_alpha = info_ptr->trans_alpha; - retval |= PNG_INFO_tRNS; - } - - if (trans_color != NULL) - *trans_color = &(info_ptr->trans_color); - } - - else /* if (info_ptr->color_type != PNG_COLOR_TYPE_PALETTE) */ - { - if (trans_color != NULL) - { - *trans_color = &(info_ptr->trans_color); - retval |= PNG_INFO_tRNS; - } - - if (trans_alpha != NULL) - *trans_alpha = NULL; - } - - if (num_trans != NULL) - { - *num_trans = info_ptr->num_trans; - retval |= PNG_INFO_tRNS; - } - } - - return retval; -} -#endif - -#ifdef PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED -int PNGAPI -png_get_unknown_chunks(png_const_structrp png_ptr, png_inforp info_ptr, - png_unknown_chunkpp unknowns) -{ - if (png_ptr != NULL && info_ptr != NULL && unknowns != NULL) - { - *unknowns = info_ptr->unknown_chunks; - return info_ptr->unknown_chunks_num; - } - - return 0; -} -#endif - -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED -png_byte PNGAPI -png_get_rgb_to_gray_status(png_const_structrp png_ptr) -{ - return (png_byte)(png_ptr ? png_ptr->rgb_to_gray_status : 0); -} -#endif - -#ifdef PNG_USER_CHUNKS_SUPPORTED -png_voidp PNGAPI -png_get_user_chunk_ptr(png_const_structrp png_ptr) -{ - return (png_ptr ? png_ptr->user_chunk_ptr : NULL); -} -#endif - -size_t PNGAPI -png_get_compression_buffer_size(png_const_structrp png_ptr) -{ - if (png_ptr == NULL) - return 0; - -#ifdef PNG_WRITE_SUPPORTED - if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0) -#endif - { -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED - return png_ptr->IDAT_read_size; -#else - return PNG_IDAT_READ_SIZE; -#endif - } - -#ifdef PNG_WRITE_SUPPORTED - else - return png_ptr->zbuffer_size; -#endif -} - -#ifdef PNG_SET_USER_LIMITS_SUPPORTED -/* These functions were added to libpng 1.2.6 and were enabled - * by default in libpng-1.4.0 */ -png_uint_32 PNGAPI -png_get_user_width_max(png_const_structrp png_ptr) -{ - return (png_ptr ? png_ptr->user_width_max : 0); -} - -png_uint_32 PNGAPI -png_get_user_height_max(png_const_structrp png_ptr) -{ - return (png_ptr ? png_ptr->user_height_max : 0); -} - -/* This function was added to libpng 1.4.0 */ -png_uint_32 PNGAPI -png_get_chunk_cache_max(png_const_structrp png_ptr) -{ - return (png_ptr ? png_ptr->user_chunk_cache_max : 0); -} - -/* This function was added to libpng 1.4.1 */ -png_alloc_size_t PNGAPI -png_get_chunk_malloc_max(png_const_structrp png_ptr) -{ - return (png_ptr ? png_ptr->user_chunk_malloc_max : 0); -} -#endif /* SET_USER_LIMITS */ - -/* These functions were added to libpng 1.4.0 */ -#ifdef PNG_IO_STATE_SUPPORTED -png_uint_32 PNGAPI -png_get_io_state(png_const_structrp png_ptr) -{ - return png_ptr->io_state; -} - -png_uint_32 PNGAPI -png_get_io_chunk_type(png_const_structrp png_ptr) -{ - return png_ptr->chunk_name; -} -#endif /* IO_STATE */ - -#ifdef PNG_CHECK_FOR_INVALID_INDEX_SUPPORTED -# ifdef PNG_GET_PALETTE_MAX_SUPPORTED -int PNGAPI -png_get_palette_max(png_const_structp png_ptr, png_const_infop info_ptr) -{ - if (png_ptr != NULL && info_ptr != NULL) - return png_ptr->num_palette_max; - - return -1; -} -# endif -#endif - -#endif /* READ || WRITE */ diff --git a/dep/libpng/src/pnginfo.h b/dep/libpng/src/pnginfo.h deleted file mode 100644 index 1f98dedc4..000000000 --- a/dep/libpng/src/pnginfo.h +++ /dev/null @@ -1,267 +0,0 @@ - -/* pnginfo.h - header file for PNG reference library - * - * Copyright (c) 2018 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2013,2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - - /* png_info is a structure that holds the information in a PNG file so - * that the application can find out the characteristics of the image. - * If you are reading the file, this structure will tell you what is - * in the PNG file. If you are writing the file, fill in the information - * you want to put into the PNG file, using png_set_*() functions, then - * call png_write_info(). - * - * The names chosen should be very close to the PNG specification, so - * consult that document for information about the meaning of each field. - * - * With libpng < 0.95, it was only possible to directly set and read the - * the values in the png_info_struct, which meant that the contents and - * order of the values had to remain fixed. With libpng 0.95 and later, - * however, there are now functions that abstract the contents of - * png_info_struct from the application, so this makes it easier to use - * libpng with dynamic libraries, and even makes it possible to use - * libraries that don't have all of the libpng ancillary chunk-handing - * functionality. In libpng-1.5.0 this was moved into a separate private - * file that is not visible to applications. - * - * The following members may have allocated storage attached that should be - * cleaned up before the structure is discarded: palette, trans, text, - * pcal_purpose, pcal_units, pcal_params, hist, iccp_name, iccp_profile, - * splt_palettes, scal_unit, row_pointers, and unknowns. By default, these - * are automatically freed when the info structure is deallocated, if they were - * allocated internally by libpng. This behavior can be changed by means - * of the png_data_freer() function. - * - * More allocation details: all the chunk-reading functions that - * change these members go through the corresponding png_set_* - * functions. A function to clear these members is available: see - * png_free_data(). The png_set_* functions do not depend on being - * able to point info structure members to any of the storage they are - * passed (they make their own copies), EXCEPT that the png_set_text - * functions use the same storage passed to them in the text_ptr or - * itxt_ptr structure argument, and the png_set_rows and png_set_unknowns - * functions do not make their own copies. - */ -#ifndef PNGINFO_H -#define PNGINFO_H - -struct png_info_def -{ - /* The following are necessary for every PNG file */ - png_uint_32 width; /* width of image in pixels (from IHDR) */ - png_uint_32 height; /* height of image in pixels (from IHDR) */ - png_uint_32 valid; /* valid chunk data (see PNG_INFO_ below) */ - size_t rowbytes; /* bytes needed to hold an untransformed row */ - png_colorp palette; /* array of color values (valid & PNG_INFO_PLTE) */ - png_uint_16 num_palette; /* number of color entries in "palette" (PLTE) */ - png_uint_16 num_trans; /* number of transparent palette color (tRNS) */ - png_byte bit_depth; /* 1, 2, 4, 8, or 16 bits/channel (from IHDR) */ - png_byte color_type; /* see PNG_COLOR_TYPE_ below (from IHDR) */ - /* The following three should have been named *_method not *_type */ - png_byte compression_type; /* must be PNG_COMPRESSION_TYPE_BASE (IHDR) */ - png_byte filter_type; /* must be PNG_FILTER_TYPE_BASE (from IHDR) */ - png_byte interlace_type; /* One of PNG_INTERLACE_NONE, PNG_INTERLACE_ADAM7 */ - - /* The following are set by png_set_IHDR, called from the application on - * write, but the are never actually used by the write code. - */ - png_byte channels; /* number of data channels per pixel (1, 2, 3, 4) */ - png_byte pixel_depth; /* number of bits per pixel */ - png_byte spare_byte; /* to align the data, and for future use */ - -#ifdef PNG_READ_SUPPORTED - /* This is never set during write */ - png_byte signature[8]; /* magic bytes read by libpng from start of file */ -#endif - - /* The rest of the data is optional. If you are reading, check the - * valid field to see if the information in these are valid. If you - * are writing, set the valid field to those chunks you want written, - * and initialize the appropriate fields below. - */ - -#if defined(PNG_COLORSPACE_SUPPORTED) || defined(PNG_GAMMA_SUPPORTED) - /* png_colorspace only contains 'flags' if neither GAMMA or COLORSPACE are - * defined. When COLORSPACE is switched on all the colorspace-defining - * chunks should be enabled, when GAMMA is switched on all the gamma-defining - * chunks should be enabled. If this is not done it becomes possible to read - * inconsistent PNG files and assign a probably incorrect interpretation to - * the information. (In other words, by carefully choosing which chunks to - * recognize the system configuration can select an interpretation for PNG - * files containing ambiguous data and this will result in inconsistent - * behavior between different libpng builds!) - */ - png_colorspace colorspace; -#endif - -#ifdef PNG_iCCP_SUPPORTED - /* iCCP chunk data. */ - png_charp iccp_name; /* profile name */ - png_bytep iccp_profile; /* International Color Consortium profile data */ - png_uint_32 iccp_proflen; /* ICC profile data length */ -#endif - -#ifdef PNG_TEXT_SUPPORTED - /* The tEXt, and zTXt chunks contain human-readable textual data in - * uncompressed, compressed, and optionally compressed forms, respectively. - * The data in "text" is an array of pointers to uncompressed, - * null-terminated C strings. Each chunk has a keyword that describes the - * textual data contained in that chunk. Keywords are not required to be - * unique, and the text string may be empty. Any number of text chunks may - * be in an image. - */ - int num_text; /* number of comments read or comments to write */ - int max_text; /* current size of text array */ - png_textp text; /* array of comments read or comments to write */ -#endif /* TEXT */ - -#ifdef PNG_tIME_SUPPORTED - /* The tIME chunk holds the last time the displayed image data was - * modified. See the png_time struct for the contents of this struct. - */ - png_time mod_time; -#endif - -#ifdef PNG_sBIT_SUPPORTED - /* The sBIT chunk specifies the number of significant high-order bits - * in the pixel data. Values are in the range [1, bit_depth], and are - * only specified for the channels in the pixel data. The contents of - * the low-order bits is not specified. Data is valid if - * (valid & PNG_INFO_sBIT) is non-zero. - */ - png_color_8 sig_bit; /* significant bits in color channels */ -#endif - -#if defined(PNG_tRNS_SUPPORTED) || defined(PNG_READ_EXPAND_SUPPORTED) || \ -defined(PNG_READ_BACKGROUND_SUPPORTED) - /* The tRNS chunk supplies transparency data for paletted images and - * other image types that don't need a full alpha channel. There are - * "num_trans" transparency values for a paletted image, stored in the - * same order as the palette colors, starting from index 0. Values - * for the data are in the range [0, 255], ranging from fully transparent - * to fully opaque, respectively. For non-paletted images, there is a - * single color specified that should be treated as fully transparent. - * Data is valid if (valid & PNG_INFO_tRNS) is non-zero. - */ - png_bytep trans_alpha; /* alpha values for paletted image */ - png_color_16 trans_color; /* transparent color for non-palette image */ -#endif - -#if defined(PNG_bKGD_SUPPORTED) || defined(PNG_READ_BACKGROUND_SUPPORTED) - /* The bKGD chunk gives the suggested image background color if the - * display program does not have its own background color and the image - * is needs to composited onto a background before display. The colors - * in "background" are normally in the same color space/depth as the - * pixel data. Data is valid if (valid & PNG_INFO_bKGD) is non-zero. - */ - png_color_16 background; -#endif - -#ifdef PNG_oFFs_SUPPORTED - /* The oFFs chunk gives the offset in "offset_unit_type" units rightwards - * and downwards from the top-left corner of the display, page, or other - * application-specific co-ordinate space. See the PNG_OFFSET_ defines - * below for the unit types. Valid if (valid & PNG_INFO_oFFs) non-zero. - */ - png_int_32 x_offset; /* x offset on page */ - png_int_32 y_offset; /* y offset on page */ - png_byte offset_unit_type; /* offset units type */ -#endif - -#ifdef PNG_pHYs_SUPPORTED - /* The pHYs chunk gives the physical pixel density of the image for - * display or printing in "phys_unit_type" units (see PNG_RESOLUTION_ - * defines below). Data is valid if (valid & PNG_INFO_pHYs) is non-zero. - */ - png_uint_32 x_pixels_per_unit; /* horizontal pixel density */ - png_uint_32 y_pixels_per_unit; /* vertical pixel density */ - png_byte phys_unit_type; /* resolution type (see PNG_RESOLUTION_ below) */ -#endif - -#ifdef PNG_eXIf_SUPPORTED - int num_exif; /* Added at libpng-1.6.31 */ - png_bytep exif; -# ifdef PNG_READ_eXIf_SUPPORTED - png_bytep eXIf_buf; /* Added at libpng-1.6.32 */ -# endif -#endif - -#ifdef PNG_hIST_SUPPORTED - /* The hIST chunk contains the relative frequency or importance of the - * various palette entries, so that a viewer can intelligently select a - * reduced-color palette, if required. Data is an array of "num_palette" - * values in the range [0,65535]. Data valid if (valid & PNG_INFO_hIST) - * is non-zero. - */ - png_uint_16p hist; -#endif - -#ifdef PNG_pCAL_SUPPORTED - /* The pCAL chunk describes a transformation between the stored pixel - * values and original physical data values used to create the image. - * The integer range [0, 2^bit_depth - 1] maps to the floating-point - * range given by [pcal_X0, pcal_X1], and are further transformed by a - * (possibly non-linear) transformation function given by "pcal_type" - * and "pcal_params" into "pcal_units". Please see the PNG_EQUATION_ - * defines below, and the PNG-Group's PNG extensions document for a - * complete description of the transformations and how they should be - * implemented, and for a description of the ASCII parameter strings. - * Data values are valid if (valid & PNG_INFO_pCAL) non-zero. - */ - png_charp pcal_purpose; /* pCAL chunk description string */ - png_int_32 pcal_X0; /* minimum value */ - png_int_32 pcal_X1; /* maximum value */ - png_charp pcal_units; /* Latin-1 string giving physical units */ - png_charpp pcal_params; /* ASCII strings containing parameter values */ - png_byte pcal_type; /* equation type (see PNG_EQUATION_ below) */ - png_byte pcal_nparams; /* number of parameters given in pcal_params */ -#endif - -/* New members added in libpng-1.0.6 */ - png_uint_32 free_me; /* flags items libpng is responsible for freeing */ - -#ifdef PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED - /* Storage for unknown chunks that the library doesn't recognize. */ - png_unknown_chunkp unknown_chunks; - - /* The type of this field is limited by the type of - * png_struct::user_chunk_cache_max, else overflow can occur. - */ - int unknown_chunks_num; -#endif - -#ifdef PNG_sPLT_SUPPORTED - /* Data on sPLT chunks (there may be more than one). */ - png_sPLT_tp splt_palettes; - int splt_palettes_num; /* Match type returned by png_get API */ -#endif - -#ifdef PNG_sCAL_SUPPORTED - /* The sCAL chunk describes the actual physical dimensions of the - * subject matter of the graphic. The chunk contains a unit specification - * a byte value, and two ASCII strings representing floating-point - * values. The values are width and height corresponding to one pixel - * in the image. Data values are valid if (valid & PNG_INFO_sCAL) is - * non-zero. - */ - png_byte scal_unit; /* unit of physical scale */ - png_charp scal_s_width; /* string containing height */ - png_charp scal_s_height; /* string containing width */ -#endif - -#ifdef PNG_INFO_IMAGE_SUPPORTED - /* Memory has been allocated if (valid & PNG_ALLOCATED_INFO_ROWS) - non-zero */ - /* Data valid if (valid & PNG_INFO_IDAT) non-zero */ - png_bytepp row_pointers; /* the image bits */ -#endif - -}; -#endif /* PNGINFO_H */ diff --git a/dep/libpng/src/pngmem.c b/dep/libpng/src/pngmem.c deleted file mode 100644 index 09ed9c1c9..000000000 --- a/dep/libpng/src/pngmem.c +++ /dev/null @@ -1,284 +0,0 @@ - -/* pngmem.c - stub functions for memory allocation - * - * Copyright (c) 2018 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2014,2016 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - * - * This file provides a location for all memory allocation. Users who - * need special memory handling are expected to supply replacement - * functions for png_malloc() and png_free(), and to use - * png_create_read_struct_2() and png_create_write_struct_2() to - * identify the replacement functions. - */ - -#include "pngpriv.h" - -#if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) -/* Free a png_struct */ -void /* PRIVATE */ -png_destroy_png_struct(png_structrp png_ptr) -{ - if (png_ptr != NULL) - { - /* png_free might call png_error and may certainly call - * png_get_mem_ptr, so fake a temporary png_struct to support this. - */ - png_struct dummy_struct = *png_ptr; - memset(png_ptr, 0, (sizeof *png_ptr)); - png_free(&dummy_struct, png_ptr); - -# ifdef PNG_SETJMP_SUPPORTED - /* We may have a jmp_buf left to deallocate. */ - png_free_jmpbuf(&dummy_struct); -# endif - } -} - -/* Allocate memory. For reasonable files, size should never exceed - * 64K. However, zlib may allocate more than 64K if you don't tell - * it not to. See zconf.h and png.h for more information. zlib does - * need to allocate exactly 64K, so whatever you call here must - * have the ability to do that. - */ -PNG_FUNCTION(png_voidp,PNGAPI -png_calloc,(png_const_structrp png_ptr, png_alloc_size_t size),PNG_ALLOCATED) -{ - png_voidp ret; - - ret = png_malloc(png_ptr, size); - - if (ret != NULL) - memset(ret, 0, size); - - return ret; -} - -/* png_malloc_base, an internal function added at libpng 1.6.0, does the work of - * allocating memory, taking into account limits and PNG_USER_MEM_SUPPORTED. - * Checking and error handling must happen outside this routine; it returns NULL - * if the allocation cannot be done (for any reason.) - */ -PNG_FUNCTION(png_voidp /* PRIVATE */, -png_malloc_base,(png_const_structrp png_ptr, png_alloc_size_t size), - PNG_ALLOCATED) -{ - /* Moved to png_malloc_base from png_malloc_default in 1.6.0; the DOS - * allocators have also been removed in 1.6.0, so any 16-bit system now has - * to implement a user memory handler. This checks to be sure it isn't - * called with big numbers. - */ -#ifndef PNG_USER_MEM_SUPPORTED - PNG_UNUSED(png_ptr) -#endif - - /* Some compilers complain that this is always true. However, it - * can be false when integer overflow happens. - */ - if (size > 0 && size <= PNG_SIZE_MAX -# ifdef PNG_MAX_MALLOC_64K - && size <= 65536U -# endif - ) - { -#ifdef PNG_USER_MEM_SUPPORTED - if (png_ptr != NULL && png_ptr->malloc_fn != NULL) - return png_ptr->malloc_fn(png_constcast(png_structrp,png_ptr), size); - - else -#endif - return malloc((size_t)size); /* checked for truncation above */ - } - - else - return NULL; -} - -#if defined(PNG_TEXT_SUPPORTED) || defined(PNG_sPLT_SUPPORTED) ||\ - defined(PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED) -/* This is really here only to work round a spurious warning in GCC 4.6 and 4.7 - * that arises because of the checks in png_realloc_array that are repeated in - * png_malloc_array. - */ -static png_voidp -png_malloc_array_checked(png_const_structrp png_ptr, int nelements, - size_t element_size) -{ - png_alloc_size_t req = (png_alloc_size_t)nelements; /* known to be > 0 */ - - if (req <= PNG_SIZE_MAX/element_size) - return png_malloc_base(png_ptr, req * element_size); - - /* The failure case when the request is too large */ - return NULL; -} - -PNG_FUNCTION(png_voidp /* PRIVATE */, -png_malloc_array,(png_const_structrp png_ptr, int nelements, - size_t element_size),PNG_ALLOCATED) -{ - if (nelements <= 0 || element_size == 0) - png_error(png_ptr, "internal error: array alloc"); - - return png_malloc_array_checked(png_ptr, nelements, element_size); -} - -PNG_FUNCTION(png_voidp /* PRIVATE */, -png_realloc_array,(png_const_structrp png_ptr, png_const_voidp old_array, - int old_elements, int add_elements, size_t element_size),PNG_ALLOCATED) -{ - /* These are internal errors: */ - if (add_elements <= 0 || element_size == 0 || old_elements < 0 || - (old_array == NULL && old_elements > 0)) - png_error(png_ptr, "internal error: array realloc"); - - /* Check for overflow on the elements count (so the caller does not have to - * check.) - */ - if (add_elements <= INT_MAX - old_elements) - { - png_voidp new_array = png_malloc_array_checked(png_ptr, - old_elements+add_elements, element_size); - - if (new_array != NULL) - { - /* Because png_malloc_array worked the size calculations below cannot - * overflow. - */ - if (old_elements > 0) - memcpy(new_array, old_array, element_size*(unsigned)old_elements); - - memset((char*)new_array + element_size*(unsigned)old_elements, 0, - element_size*(unsigned)add_elements); - - return new_array; - } - } - - return NULL; /* error */ -} -#endif /* TEXT || sPLT || STORE_UNKNOWN_CHUNKS */ - -/* Various functions that have different error handling are derived from this. - * png_malloc always exists, but if PNG_USER_MEM_SUPPORTED is defined a separate - * function png_malloc_default is also provided. - */ -PNG_FUNCTION(png_voidp,PNGAPI -png_malloc,(png_const_structrp png_ptr, png_alloc_size_t size),PNG_ALLOCATED) -{ - png_voidp ret; - - if (png_ptr == NULL) - return NULL; - - ret = png_malloc_base(png_ptr, size); - - if (ret == NULL) - png_error(png_ptr, "Out of memory"); /* 'm' means png_malloc */ - - return ret; -} - -#ifdef PNG_USER_MEM_SUPPORTED -PNG_FUNCTION(png_voidp,PNGAPI -png_malloc_default,(png_const_structrp png_ptr, png_alloc_size_t size), - PNG_ALLOCATED PNG_DEPRECATED) -{ - png_voidp ret; - - if (png_ptr == NULL) - return NULL; - - /* Passing 'NULL' here bypasses the application provided memory handler. */ - ret = png_malloc_base(NULL/*use malloc*/, size); - - if (ret == NULL) - png_error(png_ptr, "Out of Memory"); /* 'M' means png_malloc_default */ - - return ret; -} -#endif /* USER_MEM */ - -/* This function was added at libpng version 1.2.3. The png_malloc_warn() - * function will issue a png_warning and return NULL instead of issuing a - * png_error, if it fails to allocate the requested memory. - */ -PNG_FUNCTION(png_voidp,PNGAPI -png_malloc_warn,(png_const_structrp png_ptr, png_alloc_size_t size), - PNG_ALLOCATED) -{ - if (png_ptr != NULL) - { - png_voidp ret = png_malloc_base(png_ptr, size); - - if (ret != NULL) - return ret; - - png_warning(png_ptr, "Out of memory"); - } - - return NULL; -} - -/* Free a pointer allocated by png_malloc(). If ptr is NULL, return - * without taking any action. - */ -void PNGAPI -png_free(png_const_structrp png_ptr, png_voidp ptr) -{ - if (png_ptr == NULL || ptr == NULL) - return; - -#ifdef PNG_USER_MEM_SUPPORTED - if (png_ptr->free_fn != NULL) - png_ptr->free_fn(png_constcast(png_structrp,png_ptr), ptr); - - else - png_free_default(png_ptr, ptr); -} - -PNG_FUNCTION(void,PNGAPI -png_free_default,(png_const_structrp png_ptr, png_voidp ptr),PNG_DEPRECATED) -{ - if (png_ptr == NULL || ptr == NULL) - return; -#endif /* USER_MEM */ - - free(ptr); -} - -#ifdef PNG_USER_MEM_SUPPORTED -/* This function is called when the application wants to use another method - * of allocating and freeing memory. - */ -void PNGAPI -png_set_mem_fn(png_structrp png_ptr, png_voidp mem_ptr, png_malloc_ptr - malloc_fn, png_free_ptr free_fn) -{ - if (png_ptr != NULL) - { - png_ptr->mem_ptr = mem_ptr; - png_ptr->malloc_fn = malloc_fn; - png_ptr->free_fn = free_fn; - } -} - -/* This function returns a pointer to the mem_ptr associated with the user - * functions. The application should free any memory associated with this - * pointer before png_write_destroy and png_read_destroy are called. - */ -png_voidp PNGAPI -png_get_mem_ptr(png_const_structrp png_ptr) -{ - if (png_ptr == NULL) - return NULL; - - return png_ptr->mem_ptr; -} -#endif /* USER_MEM */ -#endif /* READ || WRITE */ diff --git a/dep/libpng/src/pngpread.c b/dep/libpng/src/pngpread.c deleted file mode 100644 index ffab19c08..000000000 --- a/dep/libpng/src/pngpread.c +++ /dev/null @@ -1,1104 +0,0 @@ - -/* pngpread.c - read a png file in push mode - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -#include "pngpriv.h" - -#ifdef PNG_PROGRESSIVE_READ_SUPPORTED - -/* Push model modes */ -#define PNG_READ_SIG_MODE 0 -#define PNG_READ_CHUNK_MODE 1 -#define PNG_READ_IDAT_MODE 2 -#define PNG_READ_tEXt_MODE 4 -#define PNG_READ_zTXt_MODE 5 -#define PNG_READ_DONE_MODE 6 -#define PNG_READ_iTXt_MODE 7 -#define PNG_ERROR_MODE 8 - -#define PNG_PUSH_SAVE_BUFFER_IF_FULL \ -if (png_ptr->push_length + 4 > png_ptr->buffer_size) \ - { png_push_save_buffer(png_ptr); return; } -#define PNG_PUSH_SAVE_BUFFER_IF_LT(N) \ -if (png_ptr->buffer_size < N) \ - { png_push_save_buffer(png_ptr); return; } - -void PNGAPI -png_process_data(png_structrp png_ptr, png_inforp info_ptr, - png_bytep buffer, size_t buffer_size) -{ - if (png_ptr == NULL || info_ptr == NULL) - return; - - png_push_restore_buffer(png_ptr, buffer, buffer_size); - - while (png_ptr->buffer_size) - { - png_process_some_data(png_ptr, info_ptr); - } -} - -size_t PNGAPI -png_process_data_pause(png_structrp png_ptr, int save) -{ - if (png_ptr != NULL) - { - /* It's easiest for the caller if we do the save; then the caller doesn't - * have to supply the same data again: - */ - if (save != 0) - png_push_save_buffer(png_ptr); - else - { - /* This includes any pending saved bytes: */ - size_t remaining = png_ptr->buffer_size; - png_ptr->buffer_size = 0; - - /* So subtract the saved buffer size, unless all the data - * is actually 'saved', in which case we just return 0 - */ - if (png_ptr->save_buffer_size < remaining) - return remaining - png_ptr->save_buffer_size; - } - } - - return 0; -} - -png_uint_32 PNGAPI -png_process_data_skip(png_structrp png_ptr) -{ -/* TODO: Deprecate and remove this API. - * Somewhere the implementation of this seems to have been lost, - * or abandoned. It was only to support some internal back-door access - * to png_struct) in libpng-1.4.x. - */ - png_app_warning(png_ptr, -"png_process_data_skip is not implemented in any current version of libpng"); - return 0; -} - -/* What we do with the incoming data depends on what we were previously - * doing before we ran out of data... - */ -void /* PRIVATE */ -png_process_some_data(png_structrp png_ptr, png_inforp info_ptr) -{ - if (png_ptr == NULL) - return; - - switch (png_ptr->process_mode) - { - case PNG_READ_SIG_MODE: - { - png_push_read_sig(png_ptr, info_ptr); - break; - } - - case PNG_READ_CHUNK_MODE: - { - png_push_read_chunk(png_ptr, info_ptr); - break; - } - - case PNG_READ_IDAT_MODE: - { - png_push_read_IDAT(png_ptr); - break; - } - - default: - { - png_ptr->buffer_size = 0; - break; - } - } -} - -/* Read any remaining signature bytes from the stream and compare them with - * the correct PNG signature. It is possible that this routine is called - * with bytes already read from the signature, either because they have been - * checked by the calling application, or because of multiple calls to this - * routine. - */ -void /* PRIVATE */ -png_push_read_sig(png_structrp png_ptr, png_inforp info_ptr) -{ - size_t num_checked = png_ptr->sig_bytes; /* SAFE, does not exceed 8 */ - size_t num_to_check = 8 - num_checked; - - if (png_ptr->buffer_size < num_to_check) - { - num_to_check = png_ptr->buffer_size; - } - - png_push_fill_buffer(png_ptr, &(info_ptr->signature[num_checked]), - num_to_check); - png_ptr->sig_bytes = (png_byte)(png_ptr->sig_bytes + num_to_check); - - if (png_sig_cmp(info_ptr->signature, num_checked, num_to_check) != 0) - { - if (num_checked < 4 && - png_sig_cmp(info_ptr->signature, num_checked, num_to_check - 4) != 0) - png_error(png_ptr, "Not a PNG file"); - - else - png_error(png_ptr, "PNG file corrupted by ASCII conversion"); - } - else - { - if (png_ptr->sig_bytes >= 8) - { - png_ptr->process_mode = PNG_READ_CHUNK_MODE; - } - } -} - -void /* PRIVATE */ -png_push_read_chunk(png_structrp png_ptr, png_inforp info_ptr) -{ - png_uint_32 chunk_name; -#ifdef PNG_HANDLE_AS_UNKNOWN_SUPPORTED - int keep; /* unknown handling method */ -#endif - - /* First we make sure we have enough data for the 4-byte chunk name - * and the 4-byte chunk length before proceeding with decoding the - * chunk data. To fully decode each of these chunks, we also make - * sure we have enough data in the buffer for the 4-byte CRC at the - * end of every chunk (except IDAT, which is handled separately). - */ - if ((png_ptr->mode & PNG_HAVE_CHUNK_HEADER) == 0) - { - png_byte chunk_length[4]; - png_byte chunk_tag[4]; - - PNG_PUSH_SAVE_BUFFER_IF_LT(8) - png_push_fill_buffer(png_ptr, chunk_length, 4); - png_ptr->push_length = png_get_uint_31(png_ptr, chunk_length); - png_reset_crc(png_ptr); - png_crc_read(png_ptr, chunk_tag, 4); - png_ptr->chunk_name = PNG_CHUNK_FROM_STRING(chunk_tag); - png_check_chunk_name(png_ptr, png_ptr->chunk_name); - png_check_chunk_length(png_ptr, png_ptr->push_length); - png_ptr->mode |= PNG_HAVE_CHUNK_HEADER; - } - - chunk_name = png_ptr->chunk_name; - - if (chunk_name == png_IDAT) - { - if ((png_ptr->mode & PNG_AFTER_IDAT) != 0) - png_ptr->mode |= PNG_HAVE_CHUNK_AFTER_IDAT; - - /* If we reach an IDAT chunk, this means we have read all of the - * header chunks, and we can start reading the image (or if this - * is called after the image has been read - we have an error). - */ - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_error(png_ptr, "Missing IHDR before IDAT"); - - else if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE && - (png_ptr->mode & PNG_HAVE_PLTE) == 0) - png_error(png_ptr, "Missing PLTE before IDAT"); - - png_ptr->process_mode = PNG_READ_IDAT_MODE; - - if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - if ((png_ptr->mode & PNG_HAVE_CHUNK_AFTER_IDAT) == 0) - if (png_ptr->push_length == 0) - return; - - png_ptr->mode |= PNG_HAVE_IDAT; - - if ((png_ptr->mode & PNG_AFTER_IDAT) != 0) - png_benign_error(png_ptr, "Too many IDATs found"); - } - - if (chunk_name == png_IHDR) - { - if (png_ptr->push_length != 13) - png_error(png_ptr, "Invalid IHDR length"); - - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_IHDR(png_ptr, info_ptr, png_ptr->push_length); - } - - else if (chunk_name == png_IEND) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_IEND(png_ptr, info_ptr, png_ptr->push_length); - - png_ptr->process_mode = PNG_READ_DONE_MODE; - png_push_have_end(png_ptr, info_ptr); - } - -#ifdef PNG_HANDLE_AS_UNKNOWN_SUPPORTED - else if ((keep = png_chunk_unknown_handling(png_ptr, chunk_name)) != 0) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_unknown(png_ptr, info_ptr, png_ptr->push_length, keep); - - if (chunk_name == png_PLTE) - png_ptr->mode |= PNG_HAVE_PLTE; - } -#endif - - else if (chunk_name == png_PLTE) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_PLTE(png_ptr, info_ptr, png_ptr->push_length); - } - - else if (chunk_name == png_IDAT) - { - png_ptr->idat_size = png_ptr->push_length; - png_ptr->process_mode = PNG_READ_IDAT_MODE; - png_push_have_info(png_ptr, info_ptr); - png_ptr->zstream.avail_out = - (uInt) PNG_ROWBYTES(png_ptr->pixel_depth, - png_ptr->iwidth) + 1; - png_ptr->zstream.next_out = png_ptr->row_buf; - return; - } - -#ifdef PNG_READ_gAMA_SUPPORTED - else if (png_ptr->chunk_name == png_gAMA) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_gAMA(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_sBIT_SUPPORTED - else if (png_ptr->chunk_name == png_sBIT) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_sBIT(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_cHRM_SUPPORTED - else if (png_ptr->chunk_name == png_cHRM) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_cHRM(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_eXIf_SUPPORTED - else if (png_ptr->chunk_name == png_eXIf) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_eXIf(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_sRGB_SUPPORTED - else if (chunk_name == png_sRGB) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_sRGB(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_iCCP_SUPPORTED - else if (png_ptr->chunk_name == png_iCCP) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_iCCP(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_sPLT_SUPPORTED - else if (chunk_name == png_sPLT) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_sPLT(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_tRNS_SUPPORTED - else if (chunk_name == png_tRNS) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_tRNS(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_bKGD_SUPPORTED - else if (chunk_name == png_bKGD) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_bKGD(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_hIST_SUPPORTED - else if (chunk_name == png_hIST) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_hIST(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_pHYs_SUPPORTED - else if (chunk_name == png_pHYs) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_pHYs(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_oFFs_SUPPORTED - else if (chunk_name == png_oFFs) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_oFFs(png_ptr, info_ptr, png_ptr->push_length); - } -#endif - -#ifdef PNG_READ_pCAL_SUPPORTED - else if (chunk_name == png_pCAL) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_pCAL(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_sCAL_SUPPORTED - else if (chunk_name == png_sCAL) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_sCAL(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_tIME_SUPPORTED - else if (chunk_name == png_tIME) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_tIME(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_tEXt_SUPPORTED - else if (chunk_name == png_tEXt) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_tEXt(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_zTXt_SUPPORTED - else if (chunk_name == png_zTXt) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_zTXt(png_ptr, info_ptr, png_ptr->push_length); - } - -#endif -#ifdef PNG_READ_iTXt_SUPPORTED - else if (chunk_name == png_iTXt) - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_iTXt(png_ptr, info_ptr, png_ptr->push_length); - } -#endif - - else - { - PNG_PUSH_SAVE_BUFFER_IF_FULL - png_handle_unknown(png_ptr, info_ptr, png_ptr->push_length, - PNG_HANDLE_CHUNK_AS_DEFAULT); - } - - png_ptr->mode &= ~PNG_HAVE_CHUNK_HEADER; -} - -void PNGCBAPI -png_push_fill_buffer(png_structp png_ptr, png_bytep buffer, size_t length) -{ - png_bytep ptr; - - if (png_ptr == NULL) - return; - - ptr = buffer; - if (png_ptr->save_buffer_size != 0) - { - size_t save_size; - - if (length < png_ptr->save_buffer_size) - save_size = length; - - else - save_size = png_ptr->save_buffer_size; - - memcpy(ptr, png_ptr->save_buffer_ptr, save_size); - length -= save_size; - ptr += save_size; - png_ptr->buffer_size -= save_size; - png_ptr->save_buffer_size -= save_size; - png_ptr->save_buffer_ptr += save_size; - } - if (length != 0 && png_ptr->current_buffer_size != 0) - { - size_t save_size; - - if (length < png_ptr->current_buffer_size) - save_size = length; - - else - save_size = png_ptr->current_buffer_size; - - memcpy(ptr, png_ptr->current_buffer_ptr, save_size); - png_ptr->buffer_size -= save_size; - png_ptr->current_buffer_size -= save_size; - png_ptr->current_buffer_ptr += save_size; - } -} - -void /* PRIVATE */ -png_push_save_buffer(png_structrp png_ptr) -{ - if (png_ptr->save_buffer_size != 0) - { - if (png_ptr->save_buffer_ptr != png_ptr->save_buffer) - { - size_t i, istop; - png_bytep sp; - png_bytep dp; - - istop = png_ptr->save_buffer_size; - for (i = 0, sp = png_ptr->save_buffer_ptr, dp = png_ptr->save_buffer; - i < istop; i++, sp++, dp++) - { - *dp = *sp; - } - } - } - if (png_ptr->save_buffer_size + png_ptr->current_buffer_size > - png_ptr->save_buffer_max) - { - size_t new_max; - png_bytep old_buffer; - - if (png_ptr->save_buffer_size > PNG_SIZE_MAX - - (png_ptr->current_buffer_size + 256)) - { - png_error(png_ptr, "Potential overflow of save_buffer"); - } - - new_max = png_ptr->save_buffer_size + png_ptr->current_buffer_size + 256; - old_buffer = png_ptr->save_buffer; - png_ptr->save_buffer = (png_bytep)png_malloc_warn(png_ptr, - (size_t)new_max); - - if (png_ptr->save_buffer == NULL) - { - png_free(png_ptr, old_buffer); - png_error(png_ptr, "Insufficient memory for save_buffer"); - } - - if (old_buffer) - memcpy(png_ptr->save_buffer, old_buffer, png_ptr->save_buffer_size); - else if (png_ptr->save_buffer_size) - png_error(png_ptr, "save_buffer error"); - png_free(png_ptr, old_buffer); - png_ptr->save_buffer_max = new_max; - } - if (png_ptr->current_buffer_size) - { - memcpy(png_ptr->save_buffer + png_ptr->save_buffer_size, - png_ptr->current_buffer_ptr, png_ptr->current_buffer_size); - png_ptr->save_buffer_size += png_ptr->current_buffer_size; - png_ptr->current_buffer_size = 0; - } - png_ptr->save_buffer_ptr = png_ptr->save_buffer; - png_ptr->buffer_size = 0; -} - -void /* PRIVATE */ -png_push_restore_buffer(png_structrp png_ptr, png_bytep buffer, - size_t buffer_length) -{ - png_ptr->current_buffer = buffer; - png_ptr->current_buffer_size = buffer_length; - png_ptr->buffer_size = buffer_length + png_ptr->save_buffer_size; - png_ptr->current_buffer_ptr = png_ptr->current_buffer; -} - -void /* PRIVATE */ -png_push_read_IDAT(png_structrp png_ptr) -{ - if ((png_ptr->mode & PNG_HAVE_CHUNK_HEADER) == 0) - { - png_byte chunk_length[4]; - png_byte chunk_tag[4]; - - /* TODO: this code can be commoned up with the same code in push_read */ - PNG_PUSH_SAVE_BUFFER_IF_LT(8) - png_push_fill_buffer(png_ptr, chunk_length, 4); - png_ptr->push_length = png_get_uint_31(png_ptr, chunk_length); - png_reset_crc(png_ptr); - png_crc_read(png_ptr, chunk_tag, 4); - png_ptr->chunk_name = PNG_CHUNK_FROM_STRING(chunk_tag); - png_ptr->mode |= PNG_HAVE_CHUNK_HEADER; - - if (png_ptr->chunk_name != png_IDAT) - { - png_ptr->process_mode = PNG_READ_CHUNK_MODE; - - if ((png_ptr->flags & PNG_FLAG_ZSTREAM_ENDED) == 0) - png_error(png_ptr, "Not enough compressed data"); - - return; - } - - png_ptr->idat_size = png_ptr->push_length; - } - - if (png_ptr->idat_size != 0 && png_ptr->save_buffer_size != 0) - { - size_t save_size = png_ptr->save_buffer_size; - png_uint_32 idat_size = png_ptr->idat_size; - - /* We want the smaller of 'idat_size' and 'current_buffer_size', but they - * are of different types and we don't know which variable has the fewest - * bits. Carefully select the smaller and cast it to the type of the - * larger - this cannot overflow. Do not cast in the following test - it - * will break on either 16-bit or 64-bit platforms. - */ - if (idat_size < save_size) - save_size = (size_t)idat_size; - - else - idat_size = (png_uint_32)save_size; - - png_calculate_crc(png_ptr, png_ptr->save_buffer_ptr, save_size); - - png_process_IDAT_data(png_ptr, png_ptr->save_buffer_ptr, save_size); - - png_ptr->idat_size -= idat_size; - png_ptr->buffer_size -= save_size; - png_ptr->save_buffer_size -= save_size; - png_ptr->save_buffer_ptr += save_size; - } - - if (png_ptr->idat_size != 0 && png_ptr->current_buffer_size != 0) - { - size_t save_size = png_ptr->current_buffer_size; - png_uint_32 idat_size = png_ptr->idat_size; - - /* We want the smaller of 'idat_size' and 'current_buffer_size', but they - * are of different types and we don't know which variable has the fewest - * bits. Carefully select the smaller and cast it to the type of the - * larger - this cannot overflow. - */ - if (idat_size < save_size) - save_size = (size_t)idat_size; - - else - idat_size = (png_uint_32)save_size; - - png_calculate_crc(png_ptr, png_ptr->current_buffer_ptr, save_size); - - png_process_IDAT_data(png_ptr, png_ptr->current_buffer_ptr, save_size); - - png_ptr->idat_size -= idat_size; - png_ptr->buffer_size -= save_size; - png_ptr->current_buffer_size -= save_size; - png_ptr->current_buffer_ptr += save_size; - } - - if (png_ptr->idat_size == 0) - { - PNG_PUSH_SAVE_BUFFER_IF_LT(4) - png_crc_finish(png_ptr, 0); - png_ptr->mode &= ~PNG_HAVE_CHUNK_HEADER; - png_ptr->mode |= PNG_AFTER_IDAT; - png_ptr->zowner = 0; - } -} - -void /* PRIVATE */ -png_process_IDAT_data(png_structrp png_ptr, png_bytep buffer, - size_t buffer_length) -{ - /* The caller checks for a non-zero buffer length. */ - if (!(buffer_length > 0) || buffer == NULL) - png_error(png_ptr, "No IDAT data (internal error)"); - - /* This routine must process all the data it has been given - * before returning, calling the row callback as required to - * handle the uncompressed results. - */ - png_ptr->zstream.next_in = buffer; - /* TODO: WARNING: TRUNCATION ERROR: DANGER WILL ROBINSON: */ - png_ptr->zstream.avail_in = (uInt)buffer_length; - - /* Keep going until the decompressed data is all processed - * or the stream marked as finished. - */ - while (png_ptr->zstream.avail_in > 0 && - (png_ptr->flags & PNG_FLAG_ZSTREAM_ENDED) == 0) - { - int ret; - - /* We have data for zlib, but we must check that zlib - * has someplace to put the results. It doesn't matter - * if we don't expect any results -- it may be the input - * data is just the LZ end code. - */ - if (!(png_ptr->zstream.avail_out > 0)) - { - /* TODO: WARNING: TRUNCATION ERROR: DANGER WILL ROBINSON: */ - png_ptr->zstream.avail_out = (uInt)(PNG_ROWBYTES(png_ptr->pixel_depth, - png_ptr->iwidth) + 1); - - png_ptr->zstream.next_out = png_ptr->row_buf; - } - - /* Using Z_SYNC_FLUSH here means that an unterminated - * LZ stream (a stream with a missing end code) can still - * be handled, otherwise (Z_NO_FLUSH) a future zlib - * implementation might defer output and therefore - * change the current behavior (see comments in inflate.c - * for why this doesn't happen at present with zlib 1.2.5). - */ - ret = PNG_INFLATE(png_ptr, Z_SYNC_FLUSH); - - /* Check for any failure before proceeding. */ - if (ret != Z_OK && ret != Z_STREAM_END) - { - /* Terminate the decompression. */ - png_ptr->flags |= PNG_FLAG_ZSTREAM_ENDED; - png_ptr->zowner = 0; - - /* This may be a truncated stream (missing or - * damaged end code). Treat that as a warning. - */ - if (png_ptr->row_number >= png_ptr->num_rows || - png_ptr->pass > 6) - png_warning(png_ptr, "Truncated compressed data in IDAT"); - - else - { - if (ret == Z_DATA_ERROR) - png_benign_error(png_ptr, "IDAT: ADLER32 checksum mismatch"); - else - png_error(png_ptr, "Decompression error in IDAT"); - } - - /* Skip the check on unprocessed input */ - return; - } - - /* Did inflate output any data? */ - if (png_ptr->zstream.next_out != png_ptr->row_buf) - { - /* Is this unexpected data after the last row? - * If it is, artificially terminate the LZ output - * here. - */ - if (png_ptr->row_number >= png_ptr->num_rows || - png_ptr->pass > 6) - { - /* Extra data. */ - png_warning(png_ptr, "Extra compressed data in IDAT"); - png_ptr->flags |= PNG_FLAG_ZSTREAM_ENDED; - png_ptr->zowner = 0; - - /* Do no more processing; skip the unprocessed - * input check below. - */ - return; - } - - /* Do we have a complete row? */ - if (png_ptr->zstream.avail_out == 0) - png_push_process_row(png_ptr); - } - - /* And check for the end of the stream. */ - if (ret == Z_STREAM_END) - png_ptr->flags |= PNG_FLAG_ZSTREAM_ENDED; - } - - /* All the data should have been processed, if anything - * is left at this point we have bytes of IDAT data - * after the zlib end code. - */ - if (png_ptr->zstream.avail_in > 0) - png_warning(png_ptr, "Extra compression data in IDAT"); -} - -void /* PRIVATE */ -png_push_process_row(png_structrp png_ptr) -{ - /* 1.5.6: row_info moved out of png_struct to a local here. */ - png_row_info row_info; - - row_info.width = png_ptr->iwidth; /* NOTE: width of current interlaced row */ - row_info.color_type = png_ptr->color_type; - row_info.bit_depth = png_ptr->bit_depth; - row_info.channels = png_ptr->channels; - row_info.pixel_depth = png_ptr->pixel_depth; - row_info.rowbytes = PNG_ROWBYTES(row_info.pixel_depth, row_info.width); - - if (png_ptr->row_buf[0] > PNG_FILTER_VALUE_NONE) - { - if (png_ptr->row_buf[0] < PNG_FILTER_VALUE_LAST) - png_read_filter_row(png_ptr, &row_info, png_ptr->row_buf + 1, - png_ptr->prev_row + 1, png_ptr->row_buf[0]); - else - png_error(png_ptr, "bad adaptive filter value"); - } - - /* libpng 1.5.6: the following line was copying png_ptr->rowbytes before - * 1.5.6, while the buffer really is this big in current versions of libpng - * it may not be in the future, so this was changed just to copy the - * interlaced row count: - */ - memcpy(png_ptr->prev_row, png_ptr->row_buf, row_info.rowbytes + 1); - -#ifdef PNG_READ_TRANSFORMS_SUPPORTED - if (png_ptr->transformations != 0) - png_do_read_transformations(png_ptr, &row_info); -#endif - - /* The transformed pixel depth should match the depth now in row_info. */ - if (png_ptr->transformed_pixel_depth == 0) - { - png_ptr->transformed_pixel_depth = row_info.pixel_depth; - if (row_info.pixel_depth > png_ptr->maximum_pixel_depth) - png_error(png_ptr, "progressive row overflow"); - } - - else if (png_ptr->transformed_pixel_depth != row_info.pixel_depth) - png_error(png_ptr, "internal progressive row size calculation error"); - - -#ifdef PNG_READ_INTERLACING_SUPPORTED - /* Expand interlaced rows to full size */ - if (png_ptr->interlaced != 0 && - (png_ptr->transformations & PNG_INTERLACE) != 0) - { - if (png_ptr->pass < 6) - png_do_read_interlace(&row_info, png_ptr->row_buf + 1, png_ptr->pass, - png_ptr->transformations); - - switch (png_ptr->pass) - { - case 0: - { - int i; - for (i = 0; i < 8 && png_ptr->pass == 0; i++) - { - png_push_have_row(png_ptr, png_ptr->row_buf + 1); - png_read_push_finish_row(png_ptr); /* Updates png_ptr->pass */ - } - - if (png_ptr->pass == 2) /* Pass 1 might be empty */ - { - for (i = 0; i < 4 && png_ptr->pass == 2; i++) - { - png_push_have_row(png_ptr, NULL); - png_read_push_finish_row(png_ptr); - } - } - - if (png_ptr->pass == 4 && png_ptr->height <= 4) - { - for (i = 0; i < 2 && png_ptr->pass == 4; i++) - { - png_push_have_row(png_ptr, NULL); - png_read_push_finish_row(png_ptr); - } - } - - if (png_ptr->pass == 6 && png_ptr->height <= 4) - { - png_push_have_row(png_ptr, NULL); - png_read_push_finish_row(png_ptr); - } - - break; - } - - case 1: - { - int i; - for (i = 0; i < 8 && png_ptr->pass == 1; i++) - { - png_push_have_row(png_ptr, png_ptr->row_buf + 1); - png_read_push_finish_row(png_ptr); - } - - if (png_ptr->pass == 2) /* Skip top 4 generated rows */ - { - for (i = 0; i < 4 && png_ptr->pass == 2; i++) - { - png_push_have_row(png_ptr, NULL); - png_read_push_finish_row(png_ptr); - } - } - - break; - } - - case 2: - { - int i; - - for (i = 0; i < 4 && png_ptr->pass == 2; i++) - { - png_push_have_row(png_ptr, png_ptr->row_buf + 1); - png_read_push_finish_row(png_ptr); - } - - for (i = 0; i < 4 && png_ptr->pass == 2; i++) - { - png_push_have_row(png_ptr, NULL); - png_read_push_finish_row(png_ptr); - } - - if (png_ptr->pass == 4) /* Pass 3 might be empty */ - { - for (i = 0; i < 2 && png_ptr->pass == 4; i++) - { - png_push_have_row(png_ptr, NULL); - png_read_push_finish_row(png_ptr); - } - } - - break; - } - - case 3: - { - int i; - - for (i = 0; i < 4 && png_ptr->pass == 3; i++) - { - png_push_have_row(png_ptr, png_ptr->row_buf + 1); - png_read_push_finish_row(png_ptr); - } - - if (png_ptr->pass == 4) /* Skip top two generated rows */ - { - for (i = 0; i < 2 && png_ptr->pass == 4; i++) - { - png_push_have_row(png_ptr, NULL); - png_read_push_finish_row(png_ptr); - } - } - - break; - } - - case 4: - { - int i; - - for (i = 0; i < 2 && png_ptr->pass == 4; i++) - { - png_push_have_row(png_ptr, png_ptr->row_buf + 1); - png_read_push_finish_row(png_ptr); - } - - for (i = 0; i < 2 && png_ptr->pass == 4; i++) - { - png_push_have_row(png_ptr, NULL); - png_read_push_finish_row(png_ptr); - } - - if (png_ptr->pass == 6) /* Pass 5 might be empty */ - { - png_push_have_row(png_ptr, NULL); - png_read_push_finish_row(png_ptr); - } - - break; - } - - case 5: - { - int i; - - for (i = 0; i < 2 && png_ptr->pass == 5; i++) - { - png_push_have_row(png_ptr, png_ptr->row_buf + 1); - png_read_push_finish_row(png_ptr); - } - - if (png_ptr->pass == 6) /* Skip top generated row */ - { - png_push_have_row(png_ptr, NULL); - png_read_push_finish_row(png_ptr); - } - - break; - } - - default: - case 6: - { - png_push_have_row(png_ptr, png_ptr->row_buf + 1); - png_read_push_finish_row(png_ptr); - - if (png_ptr->pass != 6) - break; - - png_push_have_row(png_ptr, NULL); - png_read_push_finish_row(png_ptr); - } - } - } - else -#endif - { - png_push_have_row(png_ptr, png_ptr->row_buf + 1); - png_read_push_finish_row(png_ptr); - } -} - -void /* PRIVATE */ -png_read_push_finish_row(png_structrp png_ptr) -{ -#ifdef PNG_READ_INTERLACING_SUPPORTED - /* Arrays to facilitate easy interlacing - use pass (0 - 6) as index */ - - /* Start of interlace block */ - static const png_byte png_pass_start[] = {0, 4, 0, 2, 0, 1, 0}; - - /* Offset to next interlace block */ - static const png_byte png_pass_inc[] = {8, 8, 4, 4, 2, 2, 1}; - - /* Start of interlace block in the y direction */ - static const png_byte png_pass_ystart[] = {0, 0, 4, 0, 2, 0, 1}; - - /* Offset to next interlace block in the y direction */ - static const png_byte png_pass_yinc[] = {8, 8, 8, 4, 4, 2, 2}; - - /* Height of interlace block. This is not currently used - if you need - * it, uncomment it here and in png.h - static const png_byte png_pass_height[] = {8, 8, 4, 4, 2, 2, 1}; - */ -#endif - - png_ptr->row_number++; - if (png_ptr->row_number < png_ptr->num_rows) - return; - -#ifdef PNG_READ_INTERLACING_SUPPORTED - if (png_ptr->interlaced != 0) - { - png_ptr->row_number = 0; - memset(png_ptr->prev_row, 0, png_ptr->rowbytes + 1); - - do - { - png_ptr->pass++; - if ((png_ptr->pass == 1 && png_ptr->width < 5) || - (png_ptr->pass == 3 && png_ptr->width < 3) || - (png_ptr->pass == 5 && png_ptr->width < 2)) - png_ptr->pass++; - - if (png_ptr->pass > 7) - png_ptr->pass--; - - if (png_ptr->pass >= 7) - break; - - png_ptr->iwidth = (png_ptr->width + - png_pass_inc[png_ptr->pass] - 1 - - png_pass_start[png_ptr->pass]) / - png_pass_inc[png_ptr->pass]; - - if ((png_ptr->transformations & PNG_INTERLACE) != 0) - break; - - png_ptr->num_rows = (png_ptr->height + - png_pass_yinc[png_ptr->pass] - 1 - - png_pass_ystart[png_ptr->pass]) / - png_pass_yinc[png_ptr->pass]; - - } while (png_ptr->iwidth == 0 || png_ptr->num_rows == 0); - } -#endif /* READ_INTERLACING */ -} - -void /* PRIVATE */ -png_push_have_info(png_structrp png_ptr, png_inforp info_ptr) -{ - if (png_ptr->info_fn != NULL) - (*(png_ptr->info_fn))(png_ptr, info_ptr); -} - -void /* PRIVATE */ -png_push_have_end(png_structrp png_ptr, png_inforp info_ptr) -{ - if (png_ptr->end_fn != NULL) - (*(png_ptr->end_fn))(png_ptr, info_ptr); -} - -void /* PRIVATE */ -png_push_have_row(png_structrp png_ptr, png_bytep row) -{ - if (png_ptr->row_fn != NULL) - (*(png_ptr->row_fn))(png_ptr, row, png_ptr->row_number, - (int)png_ptr->pass); -} - -#ifdef PNG_READ_INTERLACING_SUPPORTED -void PNGAPI -png_progressive_combine_row(png_const_structrp png_ptr, png_bytep old_row, - png_const_bytep new_row) -{ - if (png_ptr == NULL) - return; - - /* new_row is a flag here - if it is NULL then the app callback was called - * from an empty row (see the calls to png_struct::row_fn below), otherwise - * it must be png_ptr->row_buf+1 - */ - if (new_row != NULL) - png_combine_row(png_ptr, old_row, 1/*blocky display*/); -} -#endif /* READ_INTERLACING */ - -void PNGAPI -png_set_progressive_read_fn(png_structrp png_ptr, png_voidp progressive_ptr, - png_progressive_info_ptr info_fn, png_progressive_row_ptr row_fn, - png_progressive_end_ptr end_fn) -{ - if (png_ptr == NULL) - return; - - png_ptr->info_fn = info_fn; - png_ptr->row_fn = row_fn; - png_ptr->end_fn = end_fn; - - png_set_read_fn(png_ptr, progressive_ptr, png_push_fill_buffer); -} - -png_voidp PNGAPI -png_get_progressive_ptr(png_const_structrp png_ptr) -{ - if (png_ptr == NULL) - return NULL; - - return png_ptr->io_ptr; -} -#endif /* PROGRESSIVE_READ */ diff --git a/dep/libpng/src/pngpriv.h b/dep/libpng/src/pngpriv.h deleted file mode 100644 index 9bfdb7134..000000000 --- a/dep/libpng/src/pngpriv.h +++ /dev/null @@ -1,2221 +0,0 @@ - -/* pngpriv.h - private declarations for use inside libpng - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -/* The symbols declared in this file (including the functions declared - * as extern) are PRIVATE. They are not part of the libpng public - * interface, and are not recommended for use by regular applications. - * Some of them may become public in the future; others may stay private, - * change in an incompatible way, or even disappear. - * Although the libpng users are not forbidden to include this header, - * they should be well aware of the issues that may arise from doing so. - */ - -#ifndef PNGPRIV_H -#define PNGPRIV_H - -/* Feature Test Macros. The following are defined here to ensure that correctly - * implemented libraries reveal the APIs libpng needs to build and hide those - * that are not needed and potentially damaging to the compilation. - * - * Feature Test Macros must be defined before any system header is included (see - * POSIX 1003.1 2.8.2 "POSIX Symbols." - * - * These macros only have an effect if the operating system supports either - * POSIX 1003.1 or C99, or both. On other operating systems (particularly - * Windows/Visual Studio) there is no effect; the OS specific tests below are - * still required (as of 2011-05-02.) - */ -#ifndef _POSIX_SOURCE -# define _POSIX_SOURCE 1 /* Just the POSIX 1003.1 and C89 APIs */ -#endif - -#ifndef PNG_VERSION_INFO_ONLY -/* Standard library headers not required by png.h: */ -# include -# include -#endif - -#define PNGLIB_BUILD /*libpng is being built, not used*/ - -/* If HAVE_CONFIG_H is defined during the build then the build system must - * provide an appropriate "config.h" file on the include path. The header file - * must provide definitions as required below (search for "HAVE_CONFIG_H"); - * see configure.ac for more details of the requirements. The macro - * "PNG_NO_CONFIG_H" is provided for maintainers to test for dependencies on - * 'configure'; define this macro to prevent the configure build including the - * configure generated config.h. Libpng is expected to compile without *any* - * special build system support on a reasonably ANSI-C compliant system. - */ -#if defined(HAVE_CONFIG_H) && !defined(PNG_NO_CONFIG_H) -# include - - /* Pick up the definition of 'restrict' from config.h if it was read: */ -# define PNG_RESTRICT restrict -#endif - -/* To support symbol prefixing it is necessary to know *before* including png.h - * whether the fixed point (and maybe other) APIs are exported, because if they - * are not internal definitions may be required. This is handled below just - * before png.h is included, but load the configuration now if it is available. - */ -#ifndef PNGLCONF_H -# include "pnglibconf.h" -#endif - -/* Local renames may change non-exported API functions from png.h */ -#if defined(PNG_PREFIX) && !defined(PNGPREFIX_H) -# include "pngprefix.h" -#endif - -#ifdef PNG_USER_CONFIG -# include "pngusr.h" - /* These should have been defined in pngusr.h */ -# ifndef PNG_USER_PRIVATEBUILD -# define PNG_USER_PRIVATEBUILD "Custom libpng build" -# endif -# ifndef PNG_USER_DLLFNAME_POSTFIX -# define PNG_USER_DLLFNAME_POSTFIX "Cb" -# endif -#endif - -/* Compile time options. - * ===================== - * In a multi-arch build the compiler may compile the code several times for the - * same object module, producing different binaries for different architectures. - * When this happens configure-time setting of the target host options cannot be - * done and this interferes with the handling of the ARM NEON optimizations, and - * possibly other similar optimizations. Put additional tests here; in general - * this is needed when the same option can be changed at both compile time and - * run time depending on the target OS (i.e. iOS vs Android.) - * - * NOTE: symbol prefixing does not pass $(CFLAGS) to the preprocessor, because - * this is not possible with certain compilers (Oracle SUN OS CC), as a result - * it is necessary to ensure that all extern functions that *might* be used - * regardless of $(CFLAGS) get declared in this file. The test on __ARM_NEON__ - * below is one example of this behavior because it is controlled by the - * presence or not of -mfpu=neon on the GCC command line, it is possible to do - * this in $(CC), e.g. "CC=gcc -mfpu=neon", but people who build libpng rarely - * do this. - */ -#ifndef PNG_ARM_NEON_OPT - /* ARM NEON optimizations are being controlled by the compiler settings, - * typically the target FPU. If the FPU has been set to NEON (-mfpu=neon - * with GCC) then the compiler will define __ARM_NEON__ and we can rely - * unconditionally on NEON instructions not crashing, otherwise we must - * disable use of NEON instructions. - * - * NOTE: at present these optimizations depend on 'ALIGNED_MEMORY', so they - * can only be turned on automatically if that is supported too. If - * PNG_ARM_NEON_OPT is set in CPPFLAGS (to >0) then arm/arm_init.c will fail - * to compile with an appropriate #error if ALIGNED_MEMORY has been turned - * off. - * - * Note that gcc-4.9 defines __ARM_NEON instead of the deprecated - * __ARM_NEON__, so we check both variants. - * - * To disable ARM_NEON optimizations entirely, and skip compiling the - * associated assembler code, pass --enable-arm-neon=no to configure - * or put -DPNG_ARM_NEON_OPT=0 in CPPFLAGS. - */ -# if (defined(__ARM_NEON__) || defined(__ARM_NEON)) && \ - defined(PNG_ALIGNED_MEMORY_SUPPORTED) -# define PNG_ARM_NEON_OPT 2 -# else -# define PNG_ARM_NEON_OPT 0 -# endif -#endif - -#if PNG_ARM_NEON_OPT > 0 - /* NEON optimizations are to be at least considered by libpng, so enable the - * callbacks to do this. - */ -# define PNG_FILTER_OPTIMIZATIONS png_init_filter_functions_neon - - /* By default the 'intrinsics' code in arm/filter_neon_intrinsics.c is used - * if possible - if __ARM_NEON__ is set and the compiler version is not known - * to be broken. This is controlled by PNG_ARM_NEON_IMPLEMENTATION which can - * be: - * - * 1 The intrinsics code (the default with __ARM_NEON__) - * 2 The hand coded assembler (the default without __ARM_NEON__) - * - * It is possible to set PNG_ARM_NEON_IMPLEMENTATION in CPPFLAGS, however - * this is *NOT* supported and may cease to work even after a minor revision - * to libpng. It *is* valid to do this for testing purposes, e.g. speed - * testing or a new compiler, but the results should be communicated to the - * libpng implementation list for incorporation in the next minor release. - */ -# ifndef PNG_ARM_NEON_IMPLEMENTATION -# if defined(__ARM_NEON__) || defined(__ARM_NEON) -# if defined(__clang__) - /* At present it is unknown by the libpng developers which versions - * of clang support the intrinsics, however some or perhaps all - * versions do not work with the assembler so this may be - * irrelevant, so just use the default (do nothing here.) - */ -# elif defined(__GNUC__) - /* GCC 4.5.4 NEON support is known to be broken. 4.6.3 is known to - * work, so if this *is* GCC, or G++, look for a version >4.5 - */ -# if __GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 6) -# define PNG_ARM_NEON_IMPLEMENTATION 2 -# endif /* no GNUC support */ -# endif /* __GNUC__ */ -# else /* !defined __ARM_NEON__ */ - /* The 'intrinsics' code simply won't compile without this -mfpu=neon: - */ -# if !defined(__aarch64__) && !defined(_M_ARM64) - /* The assembler code currently does not work on ARM64 */ -# define PNG_ARM_NEON_IMPLEMENTATION 2 -# endif /* __aarch64__ */ -# endif /* __ARM_NEON__ */ -# endif /* !PNG_ARM_NEON_IMPLEMENTATION */ - -# ifndef PNG_ARM_NEON_IMPLEMENTATION - /* Use the intrinsics code by default. */ -# define PNG_ARM_NEON_IMPLEMENTATION 1 -# endif -#else /* PNG_ARM_NEON_OPT == 0 */ -# define PNG_ARM_NEON_IMPLEMENTATION 0 -#endif /* PNG_ARM_NEON_OPT > 0 */ - -#ifndef PNG_MIPS_MSA_OPT -# if defined(__mips_msa) && (__mips_isa_rev >= 5) && \ - defined(PNG_ALIGNED_MEMORY_SUPPORTED) -# define PNG_MIPS_MSA_OPT 2 -# else -# define PNG_MIPS_MSA_OPT 0 -# endif -#endif - -#ifndef PNG_MIPS_MMI_OPT -# ifdef PNG_MIPS_MMI -# if defined(__mips_loongson_mmi) && (_MIPS_SIM == _ABI64) && \ - defined(PNG_ALIGNED_MEMORY_SUPPORTED) -# define PNG_MIPS_MMI_OPT 1 -# else -# define PNG_MIPS_MMI_OPT 0 -# endif -# else -# define PNG_MIPS_MMI_OPT 0 -# endif -#endif - -#ifndef PNG_POWERPC_VSX_OPT -# if defined(__PPC64__) && defined(__ALTIVEC__) && defined(__VSX__) -# define PNG_POWERPC_VSX_OPT 2 -# else -# define PNG_POWERPC_VSX_OPT 0 -# endif -#endif - -#ifndef PNG_LOONGARCH_LSX_OPT -# if defined(__loongarch_sx) -# define PNG_LOONGARCH_LSX_OPT 1 -# else -# define PNG_LOONGARCH_LSX_OPT 0 -# endif -#endif - -#ifndef PNG_INTEL_SSE_OPT -# ifdef PNG_INTEL_SSE - /* Only check for SSE if the build configuration has been modified to - * enable SSE optimizations. This means that these optimizations will - * be off by default. See contrib/intel for more details. - */ -# if defined(__SSE4_1__) || defined(__AVX__) || defined(__SSSE3__) || \ - defined(__SSE2__) || defined(_M_X64) || defined(_M_AMD64) || \ - (defined(_M_IX86_FP) && _M_IX86_FP >= 2) -# define PNG_INTEL_SSE_OPT 1 -# else -# define PNG_INTEL_SSE_OPT 0 -# endif -# else -# define PNG_INTEL_SSE_OPT 0 -# endif -#endif - -#if PNG_INTEL_SSE_OPT > 0 -# ifndef PNG_INTEL_SSE_IMPLEMENTATION -# if defined(__SSE4_1__) || defined(__AVX__) - /* We are not actually using AVX, but checking for AVX is the best - way we can detect SSE4.1 and SSSE3 on MSVC. - */ -# define PNG_INTEL_SSE_IMPLEMENTATION 3 -# elif defined(__SSSE3__) -# define PNG_INTEL_SSE_IMPLEMENTATION 2 -# elif defined(__SSE2__) || defined(_M_X64) || defined(_M_AMD64) || \ - (defined(_M_IX86_FP) && _M_IX86_FP >= 2) -# define PNG_INTEL_SSE_IMPLEMENTATION 1 -# else -# define PNG_INTEL_SSE_IMPLEMENTATION 0 -# endif -# endif - -# if PNG_INTEL_SSE_IMPLEMENTATION > 0 -# define PNG_FILTER_OPTIMIZATIONS png_init_filter_functions_sse2 -# endif -#else -# define PNG_INTEL_SSE_IMPLEMENTATION 0 -#endif - -#if PNG_MIPS_MSA_OPT > 0 -# ifndef PNG_MIPS_MSA_IMPLEMENTATION -# if defined(__mips_msa) -# if defined(__clang__) -# elif defined(__GNUC__) -# if __GNUC__ < 4 || (__GNUC__ == 4 && __GNUC_MINOR__ < 7) -# define PNG_MIPS_MSA_IMPLEMENTATION 2 -# endif /* no GNUC support */ -# endif /* __GNUC__ */ -# else /* !defined __mips_msa */ -# define PNG_MIPS_MSA_IMPLEMENTATION 2 -# endif /* __mips_msa */ -# endif /* !PNG_MIPS_MSA_IMPLEMENTATION */ - -# ifndef PNG_MIPS_MSA_IMPLEMENTATION -# define PNG_MIPS_MSA_IMPLEMENTATION 1 -# define PNG_FILTER_OPTIMIZATIONS png_init_filter_functions_mips -# endif -#else -# define PNG_MIPS_MSA_IMPLEMENTATION 0 -#endif /* PNG_MIPS_MSA_OPT > 0 */ - -#if PNG_MIPS_MMI_OPT > 0 -# ifndef PNG_MIPS_MMI_IMPLEMENTATION -# if defined(__mips_loongson_mmi) && (_MIPS_SIM == _ABI64) -# define PNG_MIPS_MMI_IMPLEMENTATION 2 -# else /* !defined __mips_loongson_mmi || _MIPS_SIM != _ABI64 */ -# define PNG_MIPS_MMI_IMPLEMENTATION 0 -# endif /* __mips_loongson_mmi && _MIPS_SIM == _ABI64 */ -# endif /* !PNG_MIPS_MMI_IMPLEMENTATION */ - -# if PNG_MIPS_MMI_IMPLEMENTATION > 0 -# define PNG_FILTER_OPTIMIZATIONS png_init_filter_functions_mips -# endif -#else -# define PNG_MIPS_MMI_IMPLEMENTATION 0 -#endif /* PNG_MIPS_MMI_OPT > 0 */ - -#if PNG_POWERPC_VSX_OPT > 0 -# define PNG_FILTER_OPTIMIZATIONS png_init_filter_functions_vsx -# define PNG_POWERPC_VSX_IMPLEMENTATION 1 -#else -# define PNG_POWERPC_VSX_IMPLEMENTATION 0 -#endif - -#if PNG_LOONGARCH_LSX_OPT > 0 -# define PNG_FILTER_OPTIMIZATIONS png_init_filter_functions_lsx -# define PNG_LOONGARCH_LSX_IMPLEMENTATION 1 -#else -# define PNG_LOONGARCH_LSX_IMPLEMENTATION 0 -#endif - -/* Is this a build of a DLL where compilation of the object modules requires - * different preprocessor settings to those required for a simple library? If - * so PNG_BUILD_DLL must be set. - * - * If libpng is used inside a DLL but that DLL does not export the libpng APIs - * PNG_BUILD_DLL must not be set. To avoid the code below kicking in build a - * static library of libpng then link the DLL against that. - */ -#ifndef PNG_BUILD_DLL -# ifdef DLL_EXPORT - /* This is set by libtool when files are compiled for a DLL; libtool - * always compiles twice, even on systems where it isn't necessary. Set - * PNG_BUILD_DLL in case it is necessary: - */ -# define PNG_BUILD_DLL -# else -# ifdef _WINDLL - /* This is set by the Microsoft Visual Studio IDE in projects that - * build a DLL. It can't easily be removed from those projects (it - * isn't visible in the Visual Studio UI) so it is a fairly reliable - * indication that PNG_IMPEXP needs to be set to the DLL export - * attributes. - */ -# define PNG_BUILD_DLL -# else -# ifdef __DLL__ - /* This is set by the Borland C system when compiling for a DLL - * (as above.) - */ -# define PNG_BUILD_DLL -# else - /* Add additional compiler cases here. */ -# endif -# endif -# endif -#endif /* Setting PNG_BUILD_DLL if required */ - -/* See pngconf.h for more details: the builder of the library may set this on - * the command line to the right thing for the specific compilation system or it - * may be automagically set above (at present we know of no system where it does - * need to be set on the command line.) - * - * PNG_IMPEXP must be set here when building the library to prevent pngconf.h - * setting it to the "import" setting for a DLL build. - */ -#ifndef PNG_IMPEXP -# ifdef PNG_BUILD_DLL -# define PNG_IMPEXP PNG_DLL_EXPORT -# else - /* Not building a DLL, or the DLL doesn't require specific export - * definitions. - */ -# define PNG_IMPEXP -# endif -#endif - -/* No warnings for private or deprecated functions in the build: */ -#ifndef PNG_DEPRECATED -# define PNG_DEPRECATED -#endif -#ifndef PNG_PRIVATE -# define PNG_PRIVATE -#endif - -/* Symbol preprocessing support. - * - * To enable listing global, but internal, symbols the following macros should - * always be used to declare an extern data or function object in this file. - */ -#ifndef PNG_INTERNAL_DATA -# define PNG_INTERNAL_DATA(type, name, array) PNG_LINKAGE_DATA type name array -#endif - -#ifndef PNG_INTERNAL_FUNCTION -# define PNG_INTERNAL_FUNCTION(type, name, args, attributes)\ - PNG_LINKAGE_FUNCTION PNG_FUNCTION(type, name, args, PNG_EMPTY attributes) -#endif - -#ifndef PNG_INTERNAL_CALLBACK -# define PNG_INTERNAL_CALLBACK(type, name, args, attributes)\ - PNG_LINKAGE_CALLBACK PNG_FUNCTION(type, (PNGCBAPI name), args,\ - PNG_EMPTY attributes) -#endif - -/* If floating or fixed point APIs are disabled they may still be compiled - * internally. To handle this make sure they are declared as the appropriate - * internal extern function (otherwise the symbol prefixing stuff won't work and - * the functions will be used without definitions.) - * - * NOTE: although all the API functions are declared here they are not all - * actually built! Because the declarations are still made it is necessary to - * fake out types that they depend on. - */ -#ifndef PNG_FP_EXPORT -# ifndef PNG_FLOATING_POINT_SUPPORTED -# define PNG_FP_EXPORT(ordinal, type, name, args)\ - PNG_INTERNAL_FUNCTION(type, name, args, PNG_EMPTY); -# ifndef PNG_VERSION_INFO_ONLY - typedef struct png_incomplete png_double; - typedef png_double* png_doublep; - typedef const png_double* png_const_doublep; - typedef png_double** png_doublepp; -# endif -# endif -#endif -#ifndef PNG_FIXED_EXPORT -# ifndef PNG_FIXED_POINT_SUPPORTED -# define PNG_FIXED_EXPORT(ordinal, type, name, args)\ - PNG_INTERNAL_FUNCTION(type, name, args, PNG_EMPTY); -# endif -#endif - -#include "png.h" - -/* pngconf.h does not set PNG_DLL_EXPORT unless it is required, so: */ -#ifndef PNG_DLL_EXPORT -# define PNG_DLL_EXPORT -#endif - -/* This is a global switch to set the compilation for an installed system - * (a release build). It can be set for testing debug builds to ensure that - * they will compile when the build type is switched to RC or STABLE, the - * default is just to use PNG_LIBPNG_BUILD_BASE_TYPE. Set this in CPPFLAGS - * with either: - * - * -DPNG_RELEASE_BUILD Turns on the release compile path - * -DPNG_RELEASE_BUILD=0 Turns it off - * or in your pngusr.h with - * #define PNG_RELEASE_BUILD=1 Turns on the release compile path - * #define PNG_RELEASE_BUILD=0 Turns it off - */ -#ifndef PNG_RELEASE_BUILD -# define PNG_RELEASE_BUILD (PNG_LIBPNG_BUILD_BASE_TYPE >= PNG_LIBPNG_BUILD_RC) -#endif - -/* SECURITY and SAFETY: - * - * libpng is built with support for internal limits on image dimensions and - * memory usage. These are documented in scripts/pnglibconf.dfa of the - * source and recorded in the machine generated header file pnglibconf.h. - */ - -/* If you are running on a machine where you cannot allocate more - * than 64K of memory at once, uncomment this. While libpng will not - * normally need that much memory in a chunk (unless you load up a very - * large file), zlib needs to know how big of a chunk it can use, and - * libpng thus makes sure to check any memory allocation to verify it - * will fit into memory. - * - * zlib provides 'MAXSEG_64K' which, if defined, indicates the - * same limit and pngconf.h (already included) sets the limit - * if certain operating systems are detected. - */ -#if defined(MAXSEG_64K) && !defined(PNG_MAX_MALLOC_64K) -# define PNG_MAX_MALLOC_64K -#endif - -#ifndef PNG_UNUSED -/* Unused formal parameter warnings are silenced using the following macro - * which is expected to have no bad effects on performance (optimizing - * compilers will probably remove it entirely). Note that if you replace - * it with something other than whitespace, you must include the terminating - * semicolon. - */ -# define PNG_UNUSED(param) (void)param; -#endif - -/* Just a little check that someone hasn't tried to define something - * contradictory. - */ -#if (PNG_ZBUF_SIZE > 65536L) && defined(PNG_MAX_MALLOC_64K) -# undef PNG_ZBUF_SIZE -# define PNG_ZBUF_SIZE 65536L -#endif - -/* If warnings or errors are turned off the code is disabled or redirected here. - * From 1.5.4 functions have been added to allow very limited formatting of - * error and warning messages - this code will also be disabled here. - */ -#ifdef PNG_WARNINGS_SUPPORTED -# define PNG_WARNING_PARAMETERS(p) png_warning_parameters p; -#else -# define png_warning_parameter(p,number,string) ((void)0) -# define png_warning_parameter_unsigned(p,number,format,value) ((void)0) -# define png_warning_parameter_signed(p,number,format,value) ((void)0) -# define png_formatted_warning(pp,p,message) ((void)(pp)) -# define PNG_WARNING_PARAMETERS(p) -#endif -#ifndef PNG_ERROR_TEXT_SUPPORTED -# define png_fixed_error(s1,s2) png_err(s1) -#endif - -/* Some fixed point APIs are still required even if not exported because - * they get used by the corresponding floating point APIs. This magic - * deals with this: - */ -#ifdef PNG_FIXED_POINT_SUPPORTED -# define PNGFAPI PNGAPI -#else -# define PNGFAPI /* PRIVATE */ -#endif - -#ifndef PNG_VERSION_INFO_ONLY -/* Other defines specific to compilers can go here. Try to keep - * them inside an appropriate ifdef/endif pair for portability. - */ - -/* C allows up-casts from (void*) to any pointer and (const void*) to any - * pointer to a const object. C++ regards this as a type error and requires an - * explicit, static, cast and provides the static_cast<> rune to ensure that - * const is not cast away. - */ -#ifdef __cplusplus -# define png_voidcast(type, value) static_cast(value) -# define png_constcast(type, value) const_cast(value) -# define png_aligncast(type, value) \ - static_cast(static_cast(value)) -# define png_aligncastconst(type, value) \ - static_cast(static_cast(value)) -#else -# define png_voidcast(type, value) (value) -# define png_constcast(type, value) ((type)(void*)(const void*)(value)) -# define png_aligncast(type, value) ((void*)(value)) -# define png_aligncastconst(type, value) ((const void*)(value)) -#endif /* __cplusplus */ - -#if defined(PNG_FLOATING_POINT_SUPPORTED) ||\ - defined(PNG_FLOATING_ARITHMETIC_SUPPORTED) - /* png.c requires the following ANSI-C constants if the conversion of - * floating point to ASCII is implemented therein: - * - * DBL_DIG Maximum number of decimal digits (can be set to any constant) - * DBL_MIN Smallest normalized fp number (can be set to an arbitrary value) - * DBL_MAX Maximum floating point number (can be set to an arbitrary value) - */ -# include - -# include - -# if defined(_AMIGA) && defined(__SASC) && defined(_M68881) - /* Amiga SAS/C: We must include builtin FPU functions when compiling using - * MATH=68881 - */ -# include -# endif -#endif - -/* This provides the non-ANSI (far) memory allocation routines. */ -#if defined(__TURBOC__) && defined(__MSDOS__) -# include -# include -#endif - -#if defined(_WIN32) || defined(__WIN32__) || defined(__NT__) -# include -#endif -#endif /* PNG_VERSION_INFO_ONLY */ - -/* Moved here around 1.5.0beta36 from pngconf.h */ -/* Users may want to use these so they are not private. Any library - * functions that are passed far data must be model-independent. - */ - -/* Platform-independent functions */ -#ifndef PNG_ABORT -# define PNG_ABORT() abort() -#endif - -/* These macros may need to be architecture dependent. */ -#define PNG_ALIGN_NONE 0 /* do not use data alignment */ -#define PNG_ALIGN_ALWAYS 1 /* assume unaligned accesses are OK */ -#ifdef offsetof -# define PNG_ALIGN_OFFSET 2 /* use offsetof to determine alignment */ -#else -# define PNG_ALIGN_OFFSET -1 /* prevent the use of this */ -#endif -#define PNG_ALIGN_SIZE 3 /* use sizeof to determine alignment */ - -#ifndef PNG_ALIGN_TYPE - /* Default to using aligned access optimizations and requiring alignment to a - * multiple of the data type size. Override in a compiler specific fashion - * if necessary by inserting tests here: - */ -# define PNG_ALIGN_TYPE PNG_ALIGN_SIZE -#endif - -#if PNG_ALIGN_TYPE == PNG_ALIGN_SIZE - /* This is used because in some compiler implementations non-aligned - * structure members are supported, so the offsetof approach below fails. - * Set PNG_ALIGN_SIZE=0 for compiler combinations where unaligned access - * is good for performance. Do not do this unless you have tested the - * result and understand it. - */ -# define png_alignof(type) (sizeof(type)) -#else -# if PNG_ALIGN_TYPE == PNG_ALIGN_OFFSET -# define png_alignof(type) offsetof(struct{char c; type t;}, t) -# else -# if PNG_ALIGN_TYPE == PNG_ALIGN_ALWAYS -# define png_alignof(type) 1 -# endif - /* Else leave png_alignof undefined to prevent use thereof */ -# endif -#endif - -/* This implicitly assumes alignment is always a multiple of 2. */ -#ifdef png_alignof -# define png_isaligned(ptr, type) \ - (((type)(size_t)((const void*)(ptr)) & (type)(png_alignof(type)-1)) == 0) -#else -# define png_isaligned(ptr, type) 0 -#endif - -/* End of memory model/platform independent support */ -/* End of 1.5.0beta36 move from pngconf.h */ - -/* CONSTANTS and UTILITY MACROS - * These are used internally by libpng and not exposed in the API - */ - -/* Various modes of operation. Note that after an init, mode is set to - * zero automatically when the structure is created. Three of these - * are defined in png.h because they need to be visible to applications - * that call png_set_unknown_chunk(). - */ -/* #define PNG_HAVE_IHDR 0x01U (defined in png.h) */ -/* #define PNG_HAVE_PLTE 0x02U (defined in png.h) */ -#define PNG_HAVE_IDAT 0x04U -/* #define PNG_AFTER_IDAT 0x08U (defined in png.h) */ -#define PNG_HAVE_IEND 0x10U - /* 0x20U (unused) */ - /* 0x40U (unused) */ - /* 0x80U (unused) */ -#define PNG_HAVE_CHUNK_HEADER 0x100U -#define PNG_WROTE_tIME 0x200U -#define PNG_WROTE_INFO_BEFORE_PLTE 0x400U -#define PNG_BACKGROUND_IS_GRAY 0x800U -#define PNG_HAVE_PNG_SIGNATURE 0x1000U -#define PNG_HAVE_CHUNK_AFTER_IDAT 0x2000U /* Have another chunk after IDAT */ -#define PNG_WROTE_eXIf 0x4000U -#define PNG_IS_READ_STRUCT 0x8000U /* Else is a write struct */ - -/* Flags for the transformations the PNG library does on the image data */ -#define PNG_BGR 0x0001U -#define PNG_INTERLACE 0x0002U -#define PNG_PACK 0x0004U -#define PNG_SHIFT 0x0008U -#define PNG_SWAP_BYTES 0x0010U -#define PNG_INVERT_MONO 0x0020U -#define PNG_QUANTIZE 0x0040U -#define PNG_COMPOSE 0x0080U /* Was PNG_BACKGROUND */ -#define PNG_BACKGROUND_EXPAND 0x0100U -#define PNG_EXPAND_16 0x0200U /* Added to libpng 1.5.2 */ -#define PNG_16_TO_8 0x0400U /* Becomes 'chop' in 1.5.4 */ -#define PNG_RGBA 0x0800U -#define PNG_EXPAND 0x1000U -#define PNG_GAMMA 0x2000U -#define PNG_GRAY_TO_RGB 0x4000U -#define PNG_FILLER 0x8000U -#define PNG_PACKSWAP 0x10000U -#define PNG_SWAP_ALPHA 0x20000U -#define PNG_STRIP_ALPHA 0x40000U -#define PNG_INVERT_ALPHA 0x80000U -#define PNG_USER_TRANSFORM 0x100000U -#define PNG_RGB_TO_GRAY_ERR 0x200000U -#define PNG_RGB_TO_GRAY_WARN 0x400000U -#define PNG_RGB_TO_GRAY 0x600000U /* two bits, RGB_TO_GRAY_ERR|WARN */ -#define PNG_ENCODE_ALPHA 0x800000U /* Added to libpng-1.5.4 */ -#define PNG_ADD_ALPHA 0x1000000U /* Added to libpng-1.2.7 */ -#define PNG_EXPAND_tRNS 0x2000000U /* Added to libpng-1.2.9 */ -#define PNG_SCALE_16_TO_8 0x4000000U /* Added to libpng-1.5.4 */ - /* 0x8000000U unused */ - /* 0x10000000U unused */ - /* 0x20000000U unused */ - /* 0x40000000U unused */ -/* Flags for png_create_struct */ -#define PNG_STRUCT_PNG 0x0001U -#define PNG_STRUCT_INFO 0x0002U - -/* Flags for the png_ptr->flags rather than declaring a byte for each one */ -#define PNG_FLAG_ZLIB_CUSTOM_STRATEGY 0x0001U -#define PNG_FLAG_ZSTREAM_INITIALIZED 0x0002U /* Added to libpng-1.6.0 */ - /* 0x0004U unused */ -#define PNG_FLAG_ZSTREAM_ENDED 0x0008U /* Added to libpng-1.6.0 */ - /* 0x0010U unused */ - /* 0x0020U unused */ -#define PNG_FLAG_ROW_INIT 0x0040U -#define PNG_FLAG_FILLER_AFTER 0x0080U -#define PNG_FLAG_CRC_ANCILLARY_USE 0x0100U -#define PNG_FLAG_CRC_ANCILLARY_NOWARN 0x0200U -#define PNG_FLAG_CRC_CRITICAL_USE 0x0400U -#define PNG_FLAG_CRC_CRITICAL_IGNORE 0x0800U -#define PNG_FLAG_ASSUME_sRGB 0x1000U /* Added to libpng-1.5.4 */ -#define PNG_FLAG_OPTIMIZE_ALPHA 0x2000U /* Added to libpng-1.5.4 */ -#define PNG_FLAG_DETECT_UNINITIALIZED 0x4000U /* Added to libpng-1.5.4 */ -/* #define PNG_FLAG_KEEP_UNKNOWN_CHUNKS 0x8000U */ -/* #define PNG_FLAG_KEEP_UNSAFE_CHUNKS 0x10000U */ -#define PNG_FLAG_LIBRARY_MISMATCH 0x20000U -#define PNG_FLAG_STRIP_ERROR_NUMBERS 0x40000U -#define PNG_FLAG_STRIP_ERROR_TEXT 0x80000U -#define PNG_FLAG_BENIGN_ERRORS_WARN 0x100000U /* Added to libpng-1.4.0 */ -#define PNG_FLAG_APP_WARNINGS_WARN 0x200000U /* Added to libpng-1.6.0 */ -#define PNG_FLAG_APP_ERRORS_WARN 0x400000U /* Added to libpng-1.6.0 */ - /* 0x800000U unused */ - /* 0x1000000U unused */ - /* 0x2000000U unused */ - /* 0x4000000U unused */ - /* 0x8000000U unused */ - /* 0x10000000U unused */ - /* 0x20000000U unused */ - /* 0x40000000U unused */ - -#define PNG_FLAG_CRC_ANCILLARY_MASK (PNG_FLAG_CRC_ANCILLARY_USE | \ - PNG_FLAG_CRC_ANCILLARY_NOWARN) - -#define PNG_FLAG_CRC_CRITICAL_MASK (PNG_FLAG_CRC_CRITICAL_USE | \ - PNG_FLAG_CRC_CRITICAL_IGNORE) - -#define PNG_FLAG_CRC_MASK (PNG_FLAG_CRC_ANCILLARY_MASK | \ - PNG_FLAG_CRC_CRITICAL_MASK) - -/* Save typing and make code easier to understand */ - -#define PNG_COLOR_DIST(c1, c2) (abs((int)((c1).red) - (int)((c2).red)) + \ - abs((int)((c1).green) - (int)((c2).green)) + \ - abs((int)((c1).blue) - (int)((c2).blue))) - -/* Added to libpng-1.6.0: scale a 16-bit value in the range 0..65535 to 0..255 - * by dividing by 257 *with rounding*. This macro is exact for the given range. - * See the discourse in pngrtran.c png_do_scale_16_to_8. The values in the - * macro were established by experiment (modifying the added value). The macro - * has a second variant that takes a value already scaled by 255 and divides by - * 65535 - this has a maximum error of .502. Over the range 0..65535*65535 it - * only gives off-by-one errors and only for 0.5% (1 in 200) of the values. - */ -#define PNG_DIV65535(v24) (((v24) + 32895) >> 16) -#define PNG_DIV257(v16) PNG_DIV65535((png_uint_32)(v16) * 255) - -/* Added to libpng-1.2.6 JB */ -#define PNG_ROWBYTES(pixel_bits, width) \ - ((pixel_bits) >= 8 ? \ - ((size_t)(width) * (((size_t)(pixel_bits)) >> 3)) : \ - (( ((size_t)(width) * ((size_t)(pixel_bits))) + 7) >> 3) ) - -/* This returns the number of trailing bits in the last byte of a row, 0 if the - * last byte is completely full of pixels. It is, in principle, (pixel_bits x - * width) % 8, but that would overflow for large 'width'. The second macro is - * the same except that it returns the number of unused bits in the last byte; - * (8-TRAILBITS), but 0 when TRAILBITS is 0. - * - * NOTE: these macros are intended to be self-evidently correct and never - * overflow on the assumption that pixel_bits is in the range 0..255. The - * arguments are evaluated only once and they can be signed (e.g. as a result of - * the integral promotions). The result of the expression always has type - * (png_uint_32), however the compiler always knows it is in the range 0..7. - */ -#define PNG_TRAILBITS(pixel_bits, width) \ - (((pixel_bits) * ((width) % (png_uint_32)8)) % 8) - -#define PNG_PADBITS(pixel_bits, width) \ - ((8 - PNG_TRAILBITS(pixel_bits, width)) % 8) - -/* PNG_OUT_OF_RANGE returns true if value is outside the range - * ideal-delta..ideal+delta. Each argument is evaluated twice. - * "ideal" and "delta" should be constants, normally simple - * integers, "value" a variable. Added to libpng-1.2.6 JB - */ -#define PNG_OUT_OF_RANGE(value, ideal, delta) \ - ( (value) < (ideal)-(delta) || (value) > (ideal)+(delta) ) - -/* Conversions between fixed and floating point, only defined if - * required (to make sure the code doesn't accidentally use float - * when it is supposedly disabled.) - */ -#ifdef PNG_FLOATING_POINT_SUPPORTED -/* The floating point conversion can't overflow, though it can and - * does lose accuracy relative to the original fixed point value. - * In practice this doesn't matter because png_fixed_point only - * stores numbers with very low precision. The png_ptr and s - * arguments are unused by default but are there in case error - * checking becomes a requirement. - */ -#define png_float(png_ptr, fixed, s) (.00001 * (fixed)) - -/* The fixed point conversion performs range checking and evaluates - * its argument multiple times, so must be used with care. The - * range checking uses the PNG specification values for a signed - * 32-bit fixed point value except that the values are deliberately - * rounded-to-zero to an integral value - 21474 (21474.83 is roughly - * (2^31-1) * 100000). 's' is a string that describes the value being - * converted. - * - * NOTE: this macro will raise a png_error if the range check fails, - * therefore it is normally only appropriate to use this on values - * that come from API calls or other sources where an out of range - * error indicates a programming error, not a data error! - * - * NOTE: by default this is off - the macro is not used - because the - * function call saves a lot of code. - */ -#ifdef PNG_FIXED_POINT_MACRO_SUPPORTED -#define png_fixed(png_ptr, fp, s) ((fp) <= 21474 && (fp) >= -21474 ?\ - ((png_fixed_point)(100000 * (fp))) : (png_fixed_error(png_ptr, s),0)) -#endif -/* else the corresponding function is defined below, inside the scope of the - * cplusplus test. - */ -#endif - -/* Constants for known chunk types. If you need to add a chunk, define the name - * here. For historical reasons these constants have the form png_; i.e. - * the prefix is lower case. Please use decimal values as the parameters to - * match the ISO PNG specification and to avoid relying on the C locale - * interpretation of character values. - * - * Prior to 1.5.6 these constants were strings, as of 1.5.6 png_uint_32 values - * are computed and a new macro (PNG_STRING_FROM_CHUNK) added to allow a string - * to be generated if required. - * - * PNG_32b correctly produces a value shifted by up to 24 bits, even on - * architectures where (int) is only 16 bits. - */ -#define PNG_32b(b,s) ((png_uint_32)(b) << (s)) -#define PNG_U32(b1,b2,b3,b4) \ - (PNG_32b(b1,24) | PNG_32b(b2,16) | PNG_32b(b3,8) | PNG_32b(b4,0)) - -/* Constants for known chunk types. - * - * MAINTAINERS: If you need to add a chunk, define the name here. - * For historical reasons these constants have the form png_; i.e. - * the prefix is lower case. Please use decimal values as the parameters to - * match the ISO PNG specification and to avoid relying on the C locale - * interpretation of character values. Please keep the list sorted. - * - * Notice that PNG_U32 is used to define a 32-bit value for the 4 byte chunk - * type. In fact the specification does not express chunk types this way, - * however using a 32-bit value means that the chunk type can be read from the - * stream using exactly the same code as used for a 32-bit unsigned value and - * can be examined far more efficiently (using one arithmetic compare). - * - * Prior to 1.5.6 the chunk type constants were expressed as C strings. The - * libpng API still uses strings for 'unknown' chunks and a macro, - * PNG_STRING_FROM_CHUNK, allows a string to be generated if required. Notice - * that for portable code numeric values must still be used; the string "IHDR" - * is not portable and neither is PNG_U32('I', 'H', 'D', 'R'). - * - * In 1.7.0 the definitions will be made public in png.h to avoid having to - * duplicate the same definitions in application code. - */ -#define png_IDAT PNG_U32( 73, 68, 65, 84) -#define png_IEND PNG_U32( 73, 69, 78, 68) -#define png_IHDR PNG_U32( 73, 72, 68, 82) -#define png_PLTE PNG_U32( 80, 76, 84, 69) -#define png_bKGD PNG_U32( 98, 75, 71, 68) -#define png_cHRM PNG_U32( 99, 72, 82, 77) -#define png_eXIf PNG_U32(101, 88, 73, 102) /* registered July 2017 */ -#define png_fRAc PNG_U32(102, 82, 65, 99) /* registered, not defined */ -#define png_gAMA PNG_U32(103, 65, 77, 65) -#define png_gIFg PNG_U32(103, 73, 70, 103) -#define png_gIFt PNG_U32(103, 73, 70, 116) /* deprecated */ -#define png_gIFx PNG_U32(103, 73, 70, 120) -#define png_hIST PNG_U32(104, 73, 83, 84) -#define png_iCCP PNG_U32(105, 67, 67, 80) -#define png_iTXt PNG_U32(105, 84, 88, 116) -#define png_oFFs PNG_U32(111, 70, 70, 115) -#define png_pCAL PNG_U32(112, 67, 65, 76) -#define png_pHYs PNG_U32(112, 72, 89, 115) -#define png_sBIT PNG_U32(115, 66, 73, 84) -#define png_sCAL PNG_U32(115, 67, 65, 76) -#define png_sPLT PNG_U32(115, 80, 76, 84) -#define png_sRGB PNG_U32(115, 82, 71, 66) -#define png_sTER PNG_U32(115, 84, 69, 82) -#define png_tEXt PNG_U32(116, 69, 88, 116) -#define png_tIME PNG_U32(116, 73, 77, 69) -#define png_tRNS PNG_U32(116, 82, 78, 83) -#define png_zTXt PNG_U32(122, 84, 88, 116) - -/* The following will work on (signed char*) strings, whereas the get_uint_32 - * macro will fail on top-bit-set values because of the sign extension. - */ -#define PNG_CHUNK_FROM_STRING(s)\ - PNG_U32(0xff & (s)[0], 0xff & (s)[1], 0xff & (s)[2], 0xff & (s)[3]) - -/* This uses (char), not (png_byte) to avoid warnings on systems where (char) is - * signed and the argument is a (char[]) This macro will fail miserably on - * systems where (char) is more than 8 bits. - */ -#define PNG_STRING_FROM_CHUNK(s,c)\ - (void)(((char*)(s))[0]=(char)(((c)>>24) & 0xff), \ - ((char*)(s))[1]=(char)(((c)>>16) & 0xff),\ - ((char*)(s))[2]=(char)(((c)>>8) & 0xff), \ - ((char*)(s))[3]=(char)((c & 0xff))) - -/* Do the same but terminate with a null character. */ -#define PNG_CSTRING_FROM_CHUNK(s,c)\ - (void)(PNG_STRING_FROM_CHUNK(s,c), ((char*)(s))[4] = 0) - -/* Test on flag values as defined in the spec (section 5.4): */ -#define PNG_CHUNK_ANCILLARY(c) (1 & ((c) >> 29)) -#define PNG_CHUNK_CRITICAL(c) (!PNG_CHUNK_ANCILLARY(c)) -#define PNG_CHUNK_PRIVATE(c) (1 & ((c) >> 21)) -#define PNG_CHUNK_RESERVED(c) (1 & ((c) >> 13)) -#define PNG_CHUNK_SAFE_TO_COPY(c) (1 & ((c) >> 5)) - -/* Gamma values (new at libpng-1.5.4): */ -#define PNG_GAMMA_MAC_OLD 151724 /* Assume '1.8' is really 2.2/1.45! */ -#define PNG_GAMMA_MAC_INVERSE 65909 -#define PNG_GAMMA_sRGB_INVERSE 45455 - -/* Almost everything below is C specific; the #defines above can be used in - * non-C code (so long as it is C-preprocessed) the rest of this stuff cannot. - */ -#ifndef PNG_VERSION_INFO_ONLY - -#include "pngstruct.h" -#include "pnginfo.h" - -/* Validate the include paths - the include path used to generate pnglibconf.h - * must match that used in the build, or we must be using pnglibconf.h.prebuilt: - */ -#if PNG_ZLIB_VERNUM != 0 && PNG_ZLIB_VERNUM != ZLIB_VERNUM -# error ZLIB_VERNUM != PNG_ZLIB_VERNUM \ - "-I (include path) error: see the notes in pngpriv.h" - /* This means that when pnglibconf.h was built the copy of zlib.h that it - * used is not the same as the one being used here. Because the build of - * libpng makes decisions to use inflateInit2 and inflateReset2 based on the - * zlib version number and because this affects handling of certain broken - * PNG files the -I directives must match. - * - * The most likely explanation is that you passed a -I in CFLAGS. This will - * not work; all the preprocessor directives and in particular all the -I - * directives must be in CPPFLAGS. - */ -#endif - -/* This is used for 16-bit gamma tables -- only the top level pointers are - * const; this could be changed: - */ -typedef const png_uint_16p * png_const_uint_16pp; - -/* Added to libpng-1.5.7: sRGB conversion tables */ -#if defined(PNG_SIMPLIFIED_READ_SUPPORTED) ||\ - defined(PNG_SIMPLIFIED_WRITE_SUPPORTED) -#ifdef PNG_SIMPLIFIED_READ_SUPPORTED -PNG_INTERNAL_DATA(const png_uint_16, png_sRGB_table, [256]); - /* Convert from an sRGB encoded value 0..255 to a 16-bit linear value, - * 0..65535. This table gives the closest 16-bit answers (no errors). - */ -#endif - -PNG_INTERNAL_DATA(const png_uint_16, png_sRGB_base, [512]); -PNG_INTERNAL_DATA(const png_byte, png_sRGB_delta, [512]); - -#define PNG_sRGB_FROM_LINEAR(linear) \ - ((png_byte)(0xff & ((png_sRGB_base[(linear)>>15] \ - + ((((linear) & 0x7fff)*png_sRGB_delta[(linear)>>15])>>12)) >> 8))) - /* Given a value 'linear' in the range 0..255*65535 calculate the 8-bit sRGB - * encoded value with maximum error 0.646365. Note that the input is not a - * 16-bit value; it has been multiplied by 255! */ -#endif /* SIMPLIFIED_READ/WRITE */ - - -/* Inhibit C++ name-mangling for libpng functions but not for system calls. */ -#ifdef __cplusplus -extern "C" { -#endif /* __cplusplus */ - -/* Internal functions; these are not exported from a DLL however because they - * are used within several of the C source files they have to be C extern. - * - * All of these functions must be declared with PNG_INTERNAL_FUNCTION. - */ - -/* Zlib support */ -#define PNG_UNEXPECTED_ZLIB_RETURN (-7) -PNG_INTERNAL_FUNCTION(void, png_zstream_error,(png_structrp png_ptr, int ret), - PNG_EMPTY); - /* Used by the zlib handling functions to ensure that z_stream::msg is always - * set before they return. - */ - -#ifdef PNG_WRITE_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_free_buffer_list,(png_structrp png_ptr, - png_compression_bufferp *list),PNG_EMPTY); - /* Free the buffer list used by the compressed write code. */ -#endif - -#if defined(PNG_FLOATING_POINT_SUPPORTED) && \ - !defined(PNG_FIXED_POINT_MACRO_SUPPORTED) && \ - (defined(PNG_gAMA_SUPPORTED) || defined(PNG_cHRM_SUPPORTED) || \ - defined(PNG_sCAL_SUPPORTED) || defined(PNG_READ_BACKGROUND_SUPPORTED) || \ - defined(PNG_READ_RGB_TO_GRAY_SUPPORTED)) || \ - (defined(PNG_sCAL_SUPPORTED) && \ - defined(PNG_FLOATING_ARITHMETIC_SUPPORTED)) -PNG_INTERNAL_FUNCTION(png_fixed_point,png_fixed,(png_const_structrp png_ptr, - double fp, png_const_charp text),PNG_EMPTY); -#endif - -/* Check the user version string for compatibility, returns false if the version - * numbers aren't compatible. - */ -PNG_INTERNAL_FUNCTION(int,png_user_version_check,(png_structrp png_ptr, - png_const_charp user_png_ver),PNG_EMPTY); - -/* Internal base allocator - no messages, NULL on failure to allocate. This - * does, however, call the application provided allocator and that could call - * png_error (although that would be a bug in the application implementation.) - */ -PNG_INTERNAL_FUNCTION(png_voidp,png_malloc_base,(png_const_structrp png_ptr, - png_alloc_size_t size),PNG_ALLOCATED); - -#if defined(PNG_TEXT_SUPPORTED) || defined(PNG_sPLT_SUPPORTED) ||\ - defined(PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED) -/* Internal array allocator, outputs no error or warning messages on failure, - * just returns NULL. - */ -PNG_INTERNAL_FUNCTION(png_voidp,png_malloc_array,(png_const_structrp png_ptr, - int nelements, size_t element_size),PNG_ALLOCATED); - -/* The same but an existing array is extended by add_elements. This function - * also memsets the new elements to 0 and copies the old elements. The old - * array is not freed or altered. - */ -PNG_INTERNAL_FUNCTION(png_voidp,png_realloc_array,(png_const_structrp png_ptr, - png_const_voidp array, int old_elements, int add_elements, - size_t element_size),PNG_ALLOCATED); -#endif /* text, sPLT or unknown chunks */ - -/* Magic to create a struct when there is no struct to call the user supplied - * memory allocators. Because error handling has not been set up the memory - * handlers can't safely call png_error, but this is an obscure and undocumented - * restriction so libpng has to assume that the 'free' handler, at least, might - * call png_error. - */ -PNG_INTERNAL_FUNCTION(png_structp,png_create_png_struct, - (png_const_charp user_png_ver, png_voidp error_ptr, png_error_ptr error_fn, - png_error_ptr warn_fn, png_voidp mem_ptr, png_malloc_ptr malloc_fn, - png_free_ptr free_fn),PNG_ALLOCATED); - -/* Free memory from internal libpng struct */ -PNG_INTERNAL_FUNCTION(void,png_destroy_png_struct,(png_structrp png_ptr), - PNG_EMPTY); - -/* Free an allocated jmp_buf (always succeeds) */ -PNG_INTERNAL_FUNCTION(void,png_free_jmpbuf,(png_structrp png_ptr),PNG_EMPTY); - -/* Function to allocate memory for zlib. PNGAPI is disallowed. */ -PNG_INTERNAL_FUNCTION(voidpf,png_zalloc,(voidpf png_ptr, uInt items, uInt size), - PNG_ALLOCATED); - -/* Function to free memory for zlib. PNGAPI is disallowed. */ -PNG_INTERNAL_FUNCTION(void,png_zfree,(voidpf png_ptr, voidpf ptr),PNG_EMPTY); - -/* Next four functions are used internally as callbacks. PNGCBAPI is required - * but not PNG_EXPORT. PNGAPI added at libpng version 1.2.3, changed to - * PNGCBAPI at 1.5.0 - */ - -PNG_INTERNAL_FUNCTION(void PNGCBAPI,png_default_read_data,(png_structp png_ptr, - png_bytep data, size_t length),PNG_EMPTY); - -#ifdef PNG_PROGRESSIVE_READ_SUPPORTED -PNG_INTERNAL_FUNCTION(void PNGCBAPI,png_push_fill_buffer,(png_structp png_ptr, - png_bytep buffer, size_t length),PNG_EMPTY); -#endif - -PNG_INTERNAL_FUNCTION(void PNGCBAPI,png_default_write_data,(png_structp png_ptr, - png_bytep data, size_t length),PNG_EMPTY); - -#ifdef PNG_WRITE_FLUSH_SUPPORTED -# ifdef PNG_STDIO_SUPPORTED -PNG_INTERNAL_FUNCTION(void PNGCBAPI,png_default_flush,(png_structp png_ptr), - PNG_EMPTY); -# endif -#endif - -/* Reset the CRC variable */ -PNG_INTERNAL_FUNCTION(void,png_reset_crc,(png_structrp png_ptr),PNG_EMPTY); - -/* Write the "data" buffer to whatever output you are using */ -PNG_INTERNAL_FUNCTION(void,png_write_data,(png_structrp png_ptr, - png_const_bytep data, size_t length),PNG_EMPTY); - -/* Read and check the PNG file signature */ -PNG_INTERNAL_FUNCTION(void,png_read_sig,(png_structrp png_ptr, - png_inforp info_ptr),PNG_EMPTY); - -/* Read the chunk header (length + type name) */ -PNG_INTERNAL_FUNCTION(png_uint_32,png_read_chunk_header,(png_structrp png_ptr), - PNG_EMPTY); - -/* Read data from whatever input you are using into the "data" buffer */ -PNG_INTERNAL_FUNCTION(void,png_read_data,(png_structrp png_ptr, png_bytep data, - size_t length),PNG_EMPTY); - -/* Read bytes into buf, and update png_ptr->crc */ -PNG_INTERNAL_FUNCTION(void,png_crc_read,(png_structrp png_ptr, png_bytep buf, - png_uint_32 length),PNG_EMPTY); - -/* Read "skip" bytes, read the file crc, and (optionally) verify png_ptr->crc */ -PNG_INTERNAL_FUNCTION(int,png_crc_finish,(png_structrp png_ptr, - png_uint_32 skip),PNG_EMPTY); - -/* Read the CRC from the file and compare it to the libpng calculated CRC */ -PNG_INTERNAL_FUNCTION(int,png_crc_error,(png_structrp png_ptr),PNG_EMPTY); - -/* Calculate the CRC over a section of data. Note that we are only - * passing a maximum of 64K on systems that have this as a memory limit, - * since this is the maximum buffer size we can specify. - */ -PNG_INTERNAL_FUNCTION(void,png_calculate_crc,(png_structrp png_ptr, - png_const_bytep ptr, size_t length),PNG_EMPTY); - -#ifdef PNG_WRITE_FLUSH_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_flush,(png_structrp png_ptr),PNG_EMPTY); -#endif - -/* Write various chunks */ - -/* Write the IHDR chunk, and update the png_struct with the necessary - * information. - */ -PNG_INTERNAL_FUNCTION(void,png_write_IHDR,(png_structrp png_ptr, - png_uint_32 width, png_uint_32 height, int bit_depth, int color_type, - int compression_method, int filter_method, int interlace_method),PNG_EMPTY); - -PNG_INTERNAL_FUNCTION(void,png_write_PLTE,(png_structrp png_ptr, - png_const_colorp palette, png_uint_32 num_pal),PNG_EMPTY); - -PNG_INTERNAL_FUNCTION(void,png_compress_IDAT,(png_structrp png_ptr, - png_const_bytep row_data, png_alloc_size_t row_data_length, int flush), - PNG_EMPTY); - -PNG_INTERNAL_FUNCTION(void,png_write_IEND,(png_structrp png_ptr),PNG_EMPTY); - -#ifdef PNG_WRITE_gAMA_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_gAMA_fixed,(png_structrp png_ptr, - png_fixed_point file_gamma),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_sBIT_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_sBIT,(png_structrp png_ptr, - png_const_color_8p sbit, int color_type),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_cHRM_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_cHRM_fixed,(png_structrp png_ptr, - const png_xy *xy), PNG_EMPTY); - /* The xy value must have been previously validated */ -#endif - -#ifdef PNG_WRITE_sRGB_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_sRGB,(png_structrp png_ptr, - int intent),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_eXIf_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_eXIf,(png_structrp png_ptr, - png_bytep exif, int num_exif),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_iCCP_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_iCCP,(png_structrp png_ptr, - png_const_charp name, png_const_bytep profile), PNG_EMPTY); - /* The profile must have been previously validated for correctness, the - * length comes from the first four bytes. Only the base, deflate, - * compression is supported. - */ -#endif - -#ifdef PNG_WRITE_sPLT_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_sPLT,(png_structrp png_ptr, - png_const_sPLT_tp palette),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_tRNS_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_tRNS,(png_structrp png_ptr, - png_const_bytep trans, png_const_color_16p values, int number, - int color_type),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_bKGD_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_bKGD,(png_structrp png_ptr, - png_const_color_16p values, int color_type),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_hIST_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_hIST,(png_structrp png_ptr, - png_const_uint_16p hist, int num_hist),PNG_EMPTY); -#endif - -/* Chunks that have keywords */ -#ifdef PNG_WRITE_tEXt_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_tEXt,(png_structrp png_ptr, - png_const_charp key, png_const_charp text, size_t text_len),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_zTXt_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_zTXt,(png_structrp png_ptr, png_const_charp - key, png_const_charp text, int compression),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_iTXt_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_iTXt,(png_structrp png_ptr, - int compression, png_const_charp key, png_const_charp lang, - png_const_charp lang_key, png_const_charp text),PNG_EMPTY); -#endif - -#ifdef PNG_TEXT_SUPPORTED /* Added at version 1.0.14 and 1.2.4 */ -PNG_INTERNAL_FUNCTION(int,png_set_text_2,(png_const_structrp png_ptr, - png_inforp info_ptr, png_const_textp text_ptr, int num_text),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_oFFs_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_oFFs,(png_structrp png_ptr, - png_int_32 x_offset, png_int_32 y_offset, int unit_type),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_pCAL_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_pCAL,(png_structrp png_ptr, - png_charp purpose, png_int_32 X0, png_int_32 X1, int type, int nparams, - png_const_charp units, png_charpp params),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_pHYs_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_pHYs,(png_structrp png_ptr, - png_uint_32 x_pixels_per_unit, png_uint_32 y_pixels_per_unit, - int unit_type),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_tIME_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_tIME,(png_structrp png_ptr, - png_const_timep mod_time),PNG_EMPTY); -#endif - -#ifdef PNG_WRITE_sCAL_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_write_sCAL_s,(png_structrp png_ptr, - int unit, png_const_charp width, png_const_charp height),PNG_EMPTY); -#endif - -/* Called when finished processing a row of data */ -PNG_INTERNAL_FUNCTION(void,png_write_finish_row,(png_structrp png_ptr), - PNG_EMPTY); - -/* Internal use only. Called before first row of data */ -PNG_INTERNAL_FUNCTION(void,png_write_start_row,(png_structrp png_ptr), - PNG_EMPTY); - -/* Combine a row of data, dealing with alpha, etc. if requested. 'row' is an - * array of png_ptr->width pixels. If the image is not interlaced or this - * is the final pass this just does a memcpy, otherwise the "display" flag - * is used to determine whether to copy pixels that are not in the current pass. - * - * Because 'png_do_read_interlace' (below) replicates pixels this allows this - * function to achieve the documented 'blocky' appearance during interlaced read - * if display is 1 and the 'sparkle' appearance, where existing pixels in 'row' - * are not changed if they are not in the current pass, when display is 0. - * - * 'display' must be 0 or 1, otherwise the memcpy will be done regardless. - * - * The API always reads from the png_struct row buffer and always assumes that - * it is full width (png_do_read_interlace has already been called.) - * - * This function is only ever used to write to row buffers provided by the - * caller of the relevant libpng API and the row must have already been - * transformed by the read transformations. - * - * The PNG_USE_COMPILE_TIME_MASKS option causes generation of pre-computed - * bitmasks for use within the code, otherwise runtime generated masks are used. - * The default is compile time masks. - */ -#ifndef PNG_USE_COMPILE_TIME_MASKS -# define PNG_USE_COMPILE_TIME_MASKS 1 -#endif -PNG_INTERNAL_FUNCTION(void,png_combine_row,(png_const_structrp png_ptr, - png_bytep row, int display),PNG_EMPTY); - -#ifdef PNG_READ_INTERLACING_SUPPORTED -/* Expand an interlaced row: the 'row_info' describes the pass data that has - * been read in and must correspond to the pixels in 'row', the pixels are - * expanded (moved apart) in 'row' to match the final layout, when doing this - * the pixels are *replicated* to the intervening space. This is essential for - * the correct operation of png_combine_row, above. - */ -PNG_INTERNAL_FUNCTION(void,png_do_read_interlace,(png_row_infop row_info, - png_bytep row, int pass, png_uint_32 transformations),PNG_EMPTY); -#endif - -/* GRR TO DO (2.0 or whenever): simplify other internal calling interfaces */ - -#ifdef PNG_WRITE_INTERLACING_SUPPORTED -/* Grab pixels out of a row for an interlaced pass */ -PNG_INTERNAL_FUNCTION(void,png_do_write_interlace,(png_row_infop row_info, - png_bytep row, int pass),PNG_EMPTY); -#endif - -/* Unfilter a row: check the filter value before calling this, there is no point - * calling it for PNG_FILTER_VALUE_NONE. - */ -PNG_INTERNAL_FUNCTION(void,png_read_filter_row,(png_structrp pp, png_row_infop - row_info, png_bytep row, png_const_bytep prev_row, int filter),PNG_EMPTY); - -#if PNG_ARM_NEON_OPT > 0 -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_up_neon,(png_row_infop row_info, - png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_neon,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_neon,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg3_neon,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_neon,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_neon,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_neon,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -#endif - -#if PNG_MIPS_MSA_IMPLEMENTATION == 1 -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_up_msa,(png_row_infop row_info, - png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_msa,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_msa,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg3_msa,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_msa,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_msa,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_msa,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -#endif - -#if PNG_MIPS_MMI_IMPLEMENTATION > 0 -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_up_mmi,(png_row_infop row_info, - png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_mmi,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_mmi,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg3_mmi,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_mmi,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_mmi,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_mmi,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -#endif - -#if PNG_POWERPC_VSX_OPT > 0 -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_up_vsx,(png_row_infop row_info, - png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_vsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_vsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg3_vsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_vsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_vsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_vsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -#endif - -#if PNG_INTEL_SSE_IMPLEMENTATION > 0 -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_sse2,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_sse2,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg3_sse2,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_sse2,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_sse2,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_sse2,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -#endif - -#if PNG_LOONGARCH_LSX_IMPLEMENTATION == 1 -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_up_lsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub3_lsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_sub4_lsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg3_lsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_avg4_lsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth3_lsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_filter_row_paeth4_lsx,(png_row_infop - row_info, png_bytep row, png_const_bytep prev_row),PNG_EMPTY); -#endif - -/* Choose the best filter to use and filter the row data */ -PNG_INTERNAL_FUNCTION(void,png_write_find_filter,(png_structrp png_ptr, - png_row_infop row_info),PNG_EMPTY); - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_read_IDAT_data,(png_structrp png_ptr, - png_bytep output, png_alloc_size_t avail_out),PNG_EMPTY); - /* Read 'avail_out' bytes of data from the IDAT stream. If the output buffer - * is NULL the function checks, instead, for the end of the stream. In this - * case a benign error will be issued if the stream end is not found or if - * extra data has to be consumed. - */ -PNG_INTERNAL_FUNCTION(void,png_read_finish_IDAT,(png_structrp png_ptr), - PNG_EMPTY); - /* This cleans up when the IDAT LZ stream does not end when the last image - * byte is read; there is still some pending input. - */ - -PNG_INTERNAL_FUNCTION(void,png_read_finish_row,(png_structrp png_ptr), - PNG_EMPTY); - /* Finish a row while reading, dealing with interlacing passes, etc. */ -#endif /* SEQUENTIAL_READ */ - -/* Initialize the row buffers, etc. */ -PNG_INTERNAL_FUNCTION(void,png_read_start_row,(png_structrp png_ptr),PNG_EMPTY); - -#if ZLIB_VERNUM >= 0x1240 -PNG_INTERNAL_FUNCTION(int,png_zlib_inflate,(png_structrp png_ptr, int flush), - PNG_EMPTY); -# define PNG_INFLATE(pp, flush) png_zlib_inflate(pp, flush) -#else /* Zlib < 1.2.4 */ -# define PNG_INFLATE(pp, flush) inflate(&(pp)->zstream, flush) -#endif /* Zlib < 1.2.4 */ - -#ifdef PNG_READ_TRANSFORMS_SUPPORTED -/* Optional call to update the users info structure */ -PNG_INTERNAL_FUNCTION(void,png_read_transform_info,(png_structrp png_ptr, - png_inforp info_ptr),PNG_EMPTY); -#endif - -/* Shared transform functions, defined in pngtran.c */ -#if defined(PNG_WRITE_FILLER_SUPPORTED) || \ - defined(PNG_READ_STRIP_ALPHA_SUPPORTED) -PNG_INTERNAL_FUNCTION(void,png_do_strip_channel,(png_row_infop row_info, - png_bytep row, int at_start),PNG_EMPTY); -#endif - -#ifdef PNG_16BIT_SUPPORTED -#if defined(PNG_READ_SWAP_SUPPORTED) || defined(PNG_WRITE_SWAP_SUPPORTED) -PNG_INTERNAL_FUNCTION(void,png_do_swap,(png_row_infop row_info, - png_bytep row),PNG_EMPTY); -#endif -#endif - -#if defined(PNG_READ_PACKSWAP_SUPPORTED) || \ - defined(PNG_WRITE_PACKSWAP_SUPPORTED) -PNG_INTERNAL_FUNCTION(void,png_do_packswap,(png_row_infop row_info, - png_bytep row),PNG_EMPTY); -#endif - -#if defined(PNG_READ_INVERT_SUPPORTED) || defined(PNG_WRITE_INVERT_SUPPORTED) -PNG_INTERNAL_FUNCTION(void,png_do_invert,(png_row_infop row_info, - png_bytep row),PNG_EMPTY); -#endif - -#if defined(PNG_READ_BGR_SUPPORTED) || defined(PNG_WRITE_BGR_SUPPORTED) -PNG_INTERNAL_FUNCTION(void,png_do_bgr,(png_row_infop row_info, - png_bytep row),PNG_EMPTY); -#endif - -/* The following decodes the appropriate chunks, and does error correction, - * then calls the appropriate callback for the chunk if it is valid. - */ - -/* Decode the IHDR chunk */ -PNG_INTERNAL_FUNCTION(void,png_handle_IHDR,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_handle_PLTE,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_handle_IEND,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); - -#ifdef PNG_READ_bKGD_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_bKGD,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_cHRM_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_cHRM,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_eXIf_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_eXIf,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_gAMA_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_gAMA,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_hIST_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_hIST,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_iCCP_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_iCCP,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif /* READ_iCCP */ - -#ifdef PNG_READ_iTXt_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_iTXt,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_oFFs_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_oFFs,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_pCAL_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_pCAL,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_pHYs_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_pHYs,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_sBIT_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_sBIT,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_sCAL_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_sCAL,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_sPLT_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_sPLT,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif /* READ_sPLT */ - -#ifdef PNG_READ_sRGB_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_sRGB,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_tEXt_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_tEXt,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_tIME_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_tIME,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_tRNS_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_tRNS,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -#ifdef PNG_READ_zTXt_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_handle_zTXt,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -#endif - -PNG_INTERNAL_FUNCTION(void,png_check_chunk_name,(png_const_structrp png_ptr, - png_uint_32 chunk_name),PNG_EMPTY); - -PNG_INTERNAL_FUNCTION(void,png_check_chunk_length,(png_const_structrp png_ptr, - png_uint_32 chunk_length),PNG_EMPTY); - -PNG_INTERNAL_FUNCTION(void,png_handle_unknown,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length, int keep),PNG_EMPTY); - /* This is the function that gets called for unknown chunks. The 'keep' - * argument is either non-zero for a known chunk that has been set to be - * handled as unknown or zero for an unknown chunk. By default the function - * just skips the chunk or errors out if it is critical. - */ - -#if defined(PNG_READ_UNKNOWN_CHUNKS_SUPPORTED) ||\ - defined(PNG_HANDLE_AS_UNKNOWN_SUPPORTED) -PNG_INTERNAL_FUNCTION(int,png_chunk_unknown_handling, - (png_const_structrp png_ptr, png_uint_32 chunk_name),PNG_EMPTY); - /* Exactly as the API png_handle_as_unknown() except that the argument is a - * 32-bit chunk name, not a string. - */ -#endif /* READ_UNKNOWN_CHUNKS || HANDLE_AS_UNKNOWN */ - -/* Handle the transformations for reading and writing */ -#ifdef PNG_READ_TRANSFORMS_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_do_read_transformations,(png_structrp png_ptr, - png_row_infop row_info),PNG_EMPTY); -#endif -#ifdef PNG_WRITE_TRANSFORMS_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_do_write_transformations,(png_structrp png_ptr, - png_row_infop row_info),PNG_EMPTY); -#endif - -#ifdef PNG_READ_TRANSFORMS_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_init_read_transformations,(png_structrp png_ptr), - PNG_EMPTY); -#endif - -#ifdef PNG_PROGRESSIVE_READ_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_push_read_chunk,(png_structrp png_ptr, - png_inforp info_ptr),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_read_sig,(png_structrp png_ptr, - png_inforp info_ptr),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_check_crc,(png_structrp png_ptr),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_save_buffer,(png_structrp png_ptr), - PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_restore_buffer,(png_structrp png_ptr, - png_bytep buffer, size_t buffer_length),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_read_IDAT,(png_structrp png_ptr),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_process_IDAT_data,(png_structrp png_ptr, - png_bytep buffer, size_t buffer_length),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_process_row,(png_structrp png_ptr), - PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_handle_unknown,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_have_info,(png_structrp png_ptr, - png_inforp info_ptr),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_have_end,(png_structrp png_ptr, - png_inforp info_ptr),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_have_row,(png_structrp png_ptr, - png_bytep row),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_read_end,(png_structrp png_ptr, - png_inforp info_ptr),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_process_some_data,(png_structrp png_ptr, - png_inforp info_ptr),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_read_push_finish_row,(png_structrp png_ptr), - PNG_EMPTY); -# ifdef PNG_READ_tEXt_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_push_handle_tEXt,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_read_tEXt,(png_structrp png_ptr, - png_inforp info_ptr),PNG_EMPTY); -# endif -# ifdef PNG_READ_zTXt_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_push_handle_zTXt,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_read_zTXt,(png_structrp png_ptr, - png_inforp info_ptr),PNG_EMPTY); -# endif -# ifdef PNG_READ_iTXt_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_push_handle_iTXt,(png_structrp png_ptr, - png_inforp info_ptr, png_uint_32 length),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_push_read_iTXt,(png_structrp png_ptr, - png_inforp info_ptr),PNG_EMPTY); -# endif - -#endif /* PROGRESSIVE_READ */ - -/* Added at libpng version 1.6.0 */ -#ifdef PNG_GAMMA_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_colorspace_set_gamma,(png_const_structrp png_ptr, - png_colorspacerp colorspace, png_fixed_point gAMA), PNG_EMPTY); - /* Set the colorspace gamma with a value provided by the application or by - * the gAMA chunk on read. The value will override anything set by an ICC - * profile. - */ - -PNG_INTERNAL_FUNCTION(void,png_colorspace_sync_info,(png_const_structrp png_ptr, - png_inforp info_ptr), PNG_EMPTY); - /* Synchronize the info 'valid' flags with the colorspace */ - -PNG_INTERNAL_FUNCTION(void,png_colorspace_sync,(png_const_structrp png_ptr, - png_inforp info_ptr), PNG_EMPTY); - /* Copy the png_struct colorspace to the info_struct and call the above to - * synchronize the flags. Checks for NULL info_ptr and does nothing. - */ -#endif - -/* Added at libpng version 1.4.0 */ -#ifdef PNG_COLORSPACE_SUPPORTED -/* These internal functions are for maintaining the colorspace structure within - * a png_info or png_struct (or, indeed, both). - */ -PNG_INTERNAL_FUNCTION(int,png_colorspace_set_chromaticities, - (png_const_structrp png_ptr, png_colorspacerp colorspace, const png_xy *xy, - int preferred), PNG_EMPTY); - -PNG_INTERNAL_FUNCTION(int,png_colorspace_set_endpoints, - (png_const_structrp png_ptr, png_colorspacerp colorspace, const png_XYZ *XYZ, - int preferred), PNG_EMPTY); - -#ifdef PNG_sRGB_SUPPORTED -PNG_INTERNAL_FUNCTION(int,png_colorspace_set_sRGB,(png_const_structrp png_ptr, - png_colorspacerp colorspace, int intent), PNG_EMPTY); - /* This does set the colorspace gAMA and cHRM values too, but doesn't set the - * flags to write them, if it returns false there was a problem and an error - * message has already been output (but the colorspace may still need to be - * synced to record the invalid flag). - */ -#endif /* sRGB */ - -#ifdef PNG_iCCP_SUPPORTED -PNG_INTERNAL_FUNCTION(int,png_colorspace_set_ICC,(png_const_structrp png_ptr, - png_colorspacerp colorspace, png_const_charp name, - png_uint_32 profile_length, png_const_bytep profile, int color_type), - PNG_EMPTY); - /* The 'name' is used for information only */ - -/* Routines for checking parts of an ICC profile. */ -#ifdef PNG_READ_iCCP_SUPPORTED -PNG_INTERNAL_FUNCTION(int,png_icc_check_length,(png_const_structrp png_ptr, - png_colorspacerp colorspace, png_const_charp name, - png_uint_32 profile_length), PNG_EMPTY); -#endif /* READ_iCCP */ -PNG_INTERNAL_FUNCTION(int,png_icc_check_header,(png_const_structrp png_ptr, - png_colorspacerp colorspace, png_const_charp name, - png_uint_32 profile_length, - png_const_bytep profile /* first 132 bytes only */, int color_type), - PNG_EMPTY); -PNG_INTERNAL_FUNCTION(int,png_icc_check_tag_table,(png_const_structrp png_ptr, - png_colorspacerp colorspace, png_const_charp name, - png_uint_32 profile_length, - png_const_bytep profile /* header plus whole tag table */), PNG_EMPTY); -#ifdef PNG_sRGB_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_icc_set_sRGB,( - png_const_structrp png_ptr, png_colorspacerp colorspace, - png_const_bytep profile, uLong adler), PNG_EMPTY); - /* 'adler' is the Adler32 checksum of the uncompressed profile data. It may - * be zero to indicate that it is not available. It is used, if provided, - * as a fast check on the profile when checking to see if it is sRGB. - */ -#endif -#endif /* iCCP */ - -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_colorspace_set_rgb_coefficients, - (png_structrp png_ptr), PNG_EMPTY); - /* Set the rgb_to_gray coefficients from the colorspace Y values */ -#endif /* READ_RGB_TO_GRAY */ -#endif /* COLORSPACE */ - -/* Added at libpng version 1.4.0 */ -PNG_INTERNAL_FUNCTION(void,png_check_IHDR,(png_const_structrp png_ptr, - png_uint_32 width, png_uint_32 height, int bit_depth, - int color_type, int interlace_type, int compression_type, - int filter_type),PNG_EMPTY); - -/* Added at libpng version 1.5.10 */ -#if defined(PNG_READ_CHECK_FOR_INVALID_INDEX_SUPPORTED) || \ - defined(PNG_WRITE_CHECK_FOR_INVALID_INDEX_SUPPORTED) -PNG_INTERNAL_FUNCTION(void,png_do_check_palette_indexes, - (png_structrp png_ptr, png_row_infop row_info),PNG_EMPTY); -#endif - -#if defined(PNG_FLOATING_POINT_SUPPORTED) && defined(PNG_ERROR_TEXT_SUPPORTED) -PNG_INTERNAL_FUNCTION(void,png_fixed_error,(png_const_structrp png_ptr, - png_const_charp name),PNG_NORETURN); -#endif - -/* Puts 'string' into 'buffer' at buffer[pos], taking care never to overwrite - * the end. Always leaves the buffer nul terminated. Never errors out (and - * there is no error code.) - */ -PNG_INTERNAL_FUNCTION(size_t,png_safecat,(png_charp buffer, size_t bufsize, - size_t pos, png_const_charp string),PNG_EMPTY); - -/* Various internal functions to handle formatted warning messages, currently - * only implemented for warnings. - */ -#if defined(PNG_WARNINGS_SUPPORTED) || defined(PNG_TIME_RFC1123_SUPPORTED) -/* Utility to dump an unsigned value into a buffer, given a start pointer and - * and end pointer (which should point just *beyond* the end of the buffer!) - * Returns the pointer to the start of the formatted string. This utility only - * does unsigned values. - */ -PNG_INTERNAL_FUNCTION(png_charp,png_format_number,(png_const_charp start, - png_charp end, int format, png_alloc_size_t number),PNG_EMPTY); - -/* Convenience macro that takes an array: */ -#define PNG_FORMAT_NUMBER(buffer,format,number) \ - png_format_number(buffer, buffer + (sizeof buffer), format, number) - -/* Suggested size for a number buffer (enough for 64 bits and a sign!) */ -#define PNG_NUMBER_BUFFER_SIZE 24 - -/* These are the integer formats currently supported, the name is formed from - * the standard printf(3) format string. - */ -#define PNG_NUMBER_FORMAT_u 1 /* chose unsigned API! */ -#define PNG_NUMBER_FORMAT_02u 2 -#define PNG_NUMBER_FORMAT_d 1 /* chose signed API! */ -#define PNG_NUMBER_FORMAT_02d 2 -#define PNG_NUMBER_FORMAT_x 3 -#define PNG_NUMBER_FORMAT_02x 4 -#define PNG_NUMBER_FORMAT_fixed 5 /* choose the signed API */ -#endif - -#ifdef PNG_WARNINGS_SUPPORTED -/* New defines and members adding in libpng-1.5.4 */ -# define PNG_WARNING_PARAMETER_SIZE 32 -# define PNG_WARNING_PARAMETER_COUNT 8 /* Maximum 9; see pngerror.c */ - -/* An l-value of this type has to be passed to the APIs below to cache the - * values of the parameters to a formatted warning message. - */ -typedef char png_warning_parameters[PNG_WARNING_PARAMETER_COUNT][ - PNG_WARNING_PARAMETER_SIZE]; - -PNG_INTERNAL_FUNCTION(void,png_warning_parameter,(png_warning_parameters p, - int number, png_const_charp string),PNG_EMPTY); - /* Parameters are limited in size to PNG_WARNING_PARAMETER_SIZE characters, - * including the trailing '\0'. - */ -PNG_INTERNAL_FUNCTION(void,png_warning_parameter_unsigned, - (png_warning_parameters p, int number, int format, png_alloc_size_t value), - PNG_EMPTY); - /* Use png_alloc_size_t because it is an unsigned type as big as any we - * need to output. Use the following for a signed value. - */ -PNG_INTERNAL_FUNCTION(void,png_warning_parameter_signed, - (png_warning_parameters p, int number, int format, png_int_32 value), - PNG_EMPTY); - -PNG_INTERNAL_FUNCTION(void,png_formatted_warning,(png_const_structrp png_ptr, - png_warning_parameters p, png_const_charp message),PNG_EMPTY); - /* 'message' follows the X/Open approach of using @1, @2 to insert - * parameters previously supplied using the above functions. Errors in - * specifying the parameters will simply result in garbage substitutions. - */ -#endif - -#ifdef PNG_BENIGN_ERRORS_SUPPORTED -/* Application errors (new in 1.6); use these functions (declared below) for - * errors in the parameters or order of API function calls on read. The - * 'warning' should be used for an error that can be handled completely; the - * 'error' for one which can be handled safely but which may lose application - * information or settings. - * - * By default these both result in a png_error call prior to release, while in a - * released version the 'warning' is just a warning. However if the application - * explicitly disables benign errors (explicitly permitting the code to lose - * information) they both turn into warnings. - * - * If benign errors aren't supported they end up as the corresponding base call - * (png_warning or png_error.) - */ -PNG_INTERNAL_FUNCTION(void,png_app_warning,(png_const_structrp png_ptr, - png_const_charp message),PNG_EMPTY); - /* The application provided invalid parameters to an API function or called - * an API function at the wrong time, libpng can completely recover. - */ - -PNG_INTERNAL_FUNCTION(void,png_app_error,(png_const_structrp png_ptr, - png_const_charp message),PNG_EMPTY); - /* As above but libpng will ignore the call, or attempt some other partial - * recovery from the error. - */ -#else -# define png_app_warning(pp,s) png_warning(pp,s) -# define png_app_error(pp,s) png_error(pp,s) -#endif - -PNG_INTERNAL_FUNCTION(void,png_chunk_report,(png_const_structrp png_ptr, - png_const_charp message, int error),PNG_EMPTY); - /* Report a recoverable issue in chunk data. On read this is used to report - * a problem found while reading a particular chunk and the - * png_chunk_benign_error or png_chunk_warning function is used as - * appropriate. On write this is used to report an error that comes from - * data set via an application call to a png_set_ API and png_app_error or - * png_app_warning is used as appropriate. - * - * The 'error' parameter must have one of the following values: - */ -#define PNG_CHUNK_WARNING 0 /* never an error */ -#define PNG_CHUNK_WRITE_ERROR 1 /* an error only on write */ -#define PNG_CHUNK_ERROR 2 /* always an error */ - -/* ASCII to FP interfaces, currently only implemented if sCAL - * support is required. - */ -#if defined(PNG_sCAL_SUPPORTED) -/* MAX_DIGITS is actually the maximum number of characters in an sCAL - * width or height, derived from the precision (number of significant - * digits - a build time settable option) and assumptions about the - * maximum ridiculous exponent. - */ -#define PNG_sCAL_MAX_DIGITS (PNG_sCAL_PRECISION+1/*.*/+1/*E*/+10/*exponent*/) - -#ifdef PNG_FLOATING_POINT_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_ascii_from_fp,(png_const_structrp png_ptr, - png_charp ascii, size_t size, double fp, unsigned int precision), - PNG_EMPTY); -#endif /* FLOATING_POINT */ - -#ifdef PNG_FIXED_POINT_SUPPORTED -PNG_INTERNAL_FUNCTION(void,png_ascii_from_fixed,(png_const_structrp png_ptr, - png_charp ascii, size_t size, png_fixed_point fp),PNG_EMPTY); -#endif /* FIXED_POINT */ -#endif /* sCAL */ - -#if defined(PNG_sCAL_SUPPORTED) || defined(PNG_pCAL_SUPPORTED) -/* An internal API to validate the format of a floating point number. - * The result is the index of the next character. If the number is - * not valid it will be the index of a character in the supposed number. - * - * The format of a number is defined in the PNG extensions specification - * and this API is strictly conformant to that spec, not anyone elses! - * - * The format as a regular expression is: - * - * [+-]?[0-9]+.?([Ee][+-]?[0-9]+)? - * - * or: - * - * [+-]?.[0-9]+(.[0-9]+)?([Ee][+-]?[0-9]+)? - * - * The complexity is that either integer or fraction must be present and the - * fraction is permitted to have no digits only if the integer is present. - * - * NOTE: The dangling E problem. - * There is a PNG valid floating point number in the following: - * - * PNG floating point numbers are not greedy. - * - * Working this out requires *TWO* character lookahead (because of the - * sign), the parser does not do this - it will fail at the 'r' - this - * doesn't matter for PNG sCAL chunk values, but it requires more care - * if the value were ever to be embedded in something more complex. Use - * ANSI-C strtod if you need the lookahead. - */ -/* State table for the parser. */ -#define PNG_FP_INTEGER 0 /* before or in integer */ -#define PNG_FP_FRACTION 1 /* before or in fraction */ -#define PNG_FP_EXPONENT 2 /* before or in exponent */ -#define PNG_FP_STATE 3 /* mask for the above */ -#define PNG_FP_SAW_SIGN 4 /* Saw +/- in current state */ -#define PNG_FP_SAW_DIGIT 8 /* Saw a digit in current state */ -#define PNG_FP_SAW_DOT 16 /* Saw a dot in current state */ -#define PNG_FP_SAW_E 32 /* Saw an E (or e) in current state */ -#define PNG_FP_SAW_ANY 60 /* Saw any of the above 4 */ - -/* These three values don't affect the parser. They are set but not used. - */ -#define PNG_FP_WAS_VALID 64 /* Preceding substring is a valid fp number */ -#define PNG_FP_NEGATIVE 128 /* A negative number, including "-0" */ -#define PNG_FP_NONZERO 256 /* A non-zero value */ -#define PNG_FP_STICKY 448 /* The above three flags */ - -/* This is available for the caller to store in 'state' if required. Do not - * call the parser after setting it (the parser sometimes clears it.) - */ -#define PNG_FP_INVALID 512 /* Available for callers as a distinct value */ - -/* Result codes for the parser (boolean - true means ok, false means - * not ok yet.) - */ -#define PNG_FP_MAYBE 0 /* The number may be valid in the future */ -#define PNG_FP_OK 1 /* The number is valid */ - -/* Tests on the sticky non-zero and negative flags. To pass these checks - * the state must also indicate that the whole number is valid - this is - * achieved by testing PNG_FP_SAW_DIGIT (see the implementation for why this - * is equivalent to PNG_FP_OK above.) - */ -#define PNG_FP_NZ_MASK (PNG_FP_SAW_DIGIT | PNG_FP_NEGATIVE | PNG_FP_NONZERO) - /* NZ_MASK: the string is valid and a non-zero negative value */ -#define PNG_FP_Z_MASK (PNG_FP_SAW_DIGIT | PNG_FP_NONZERO) - /* Z MASK: the string is valid and a non-zero value. */ - /* PNG_FP_SAW_DIGIT: the string is valid. */ -#define PNG_FP_IS_ZERO(state) (((state) & PNG_FP_Z_MASK) == PNG_FP_SAW_DIGIT) -#define PNG_FP_IS_POSITIVE(state) (((state) & PNG_FP_NZ_MASK) == PNG_FP_Z_MASK) -#define PNG_FP_IS_NEGATIVE(state) (((state) & PNG_FP_NZ_MASK) == PNG_FP_NZ_MASK) - -/* The actual parser. This can be called repeatedly. It updates - * the index into the string and the state variable (which must - * be initialized to 0). It returns a result code, as above. There - * is no point calling the parser any more if it fails to advance to - * the end of the string - it is stuck on an invalid character (or - * terminated by '\0'). - * - * Note that the pointer will consume an E or even an E+ and then leave - * a 'maybe' state even though a preceding integer.fraction is valid. - * The PNG_FP_WAS_VALID flag indicates that a preceding substring was - * a valid number. It's possible to recover from this by calling - * the parser again (from the start, with state 0) but with a string - * that omits the last character (i.e. set the size to the index of - * the problem character.) This has not been tested within libpng. - */ -PNG_INTERNAL_FUNCTION(int,png_check_fp_number,(png_const_charp string, - size_t size, int *statep, size_t *whereami),PNG_EMPTY); - -/* This is the same but it checks a complete string and returns true - * only if it just contains a floating point number. As of 1.5.4 this - * function also returns the state at the end of parsing the number if - * it was valid (otherwise it returns 0.) This can be used for testing - * for negative or zero values using the sticky flag. - */ -PNG_INTERNAL_FUNCTION(int,png_check_fp_string,(png_const_charp string, - size_t size),PNG_EMPTY); -#endif /* pCAL || sCAL */ - -#if defined(PNG_GAMMA_SUPPORTED) ||\ - defined(PNG_INCH_CONVERSIONS_SUPPORTED) || defined(PNG_READ_pHYs_SUPPORTED) -/* Added at libpng version 1.5.0 */ -/* This is a utility to provide a*times/div (rounded) and indicate - * if there is an overflow. The result is a boolean - false (0) - * for overflow, true (1) if no overflow, in which case *res - * holds the result. - */ -PNG_INTERNAL_FUNCTION(int,png_muldiv,(png_fixed_point_p res, png_fixed_point a, - png_int_32 multiplied_by, png_int_32 divided_by),PNG_EMPTY); -#endif - -#if defined(PNG_READ_GAMMA_SUPPORTED) || defined(PNG_INCH_CONVERSIONS_SUPPORTED) -/* Same deal, but issue a warning on overflow and return 0. */ -PNG_INTERNAL_FUNCTION(png_fixed_point,png_muldiv_warn, - (png_const_structrp png_ptr, png_fixed_point a, png_int_32 multiplied_by, - png_int_32 divided_by),PNG_EMPTY); -#endif - -#ifdef PNG_GAMMA_SUPPORTED -/* Calculate a reciprocal - used for gamma values. This returns - * 0 if the argument is 0 in order to maintain an undefined value; - * there are no warnings. - */ -PNG_INTERNAL_FUNCTION(png_fixed_point,png_reciprocal,(png_fixed_point a), - PNG_EMPTY); - -#ifdef PNG_READ_GAMMA_SUPPORTED -/* The same but gives a reciprocal of the product of two fixed point - * values. Accuracy is suitable for gamma calculations but this is - * not exact - use png_muldiv for that. Only required at present on read. - */ -PNG_INTERNAL_FUNCTION(png_fixed_point,png_reciprocal2,(png_fixed_point a, - png_fixed_point b),PNG_EMPTY); -#endif - -/* Return true if the gamma value is significantly different from 1.0 */ -PNG_INTERNAL_FUNCTION(int,png_gamma_significant,(png_fixed_point gamma_value), - PNG_EMPTY); -#endif - -#ifdef PNG_READ_GAMMA_SUPPORTED -/* Internal fixed point gamma correction. These APIs are called as - * required to convert single values - they don't need to be fast, - * they are not used when processing image pixel values. - * - * While the input is an 'unsigned' value it must actually be the - * correct bit value - 0..255 or 0..65535 as required. - */ -PNG_INTERNAL_FUNCTION(png_uint_16,png_gamma_correct,(png_structrp png_ptr, - unsigned int value, png_fixed_point gamma_value),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(png_uint_16,png_gamma_16bit_correct,(unsigned int value, - png_fixed_point gamma_value),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(png_byte,png_gamma_8bit_correct,(unsigned int value, - png_fixed_point gamma_value),PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_destroy_gamma_table,(png_structrp png_ptr), - PNG_EMPTY); -PNG_INTERNAL_FUNCTION(void,png_build_gamma_table,(png_structrp png_ptr, - int bit_depth),PNG_EMPTY); -#endif - -/* SIMPLIFIED READ/WRITE SUPPORT */ -#if defined(PNG_SIMPLIFIED_READ_SUPPORTED) ||\ - defined(PNG_SIMPLIFIED_WRITE_SUPPORTED) -/* The internal structure that png_image::opaque points to. */ -typedef struct png_control -{ - png_structp png_ptr; - png_infop info_ptr; - png_voidp error_buf; /* Always a jmp_buf at present. */ - - png_const_bytep memory; /* Memory buffer. */ - size_t size; /* Size of the memory buffer. */ - - unsigned int for_write :1; /* Otherwise it is a read structure */ - unsigned int owned_file :1; /* We own the file in io_ptr */ -} png_control; - -/* Return the pointer to the jmp_buf from a png_control: necessary because C - * does not reveal the type of the elements of jmp_buf. - */ -#ifdef __cplusplus -# define png_control_jmp_buf(pc) (((jmp_buf*)((pc)->error_buf))[0]) -#else -# define png_control_jmp_buf(pc) ((pc)->error_buf) -#endif - -/* Utility to safely execute a piece of libpng code catching and logging any - * errors that might occur. Returns true on success, false on failure (either - * of the function or as a result of a png_error.) - */ -PNG_INTERNAL_CALLBACK(void,png_safe_error,(png_structp png_ptr, - png_const_charp error_message),PNG_NORETURN); - -#ifdef PNG_WARNINGS_SUPPORTED -PNG_INTERNAL_CALLBACK(void,png_safe_warning,(png_structp png_ptr, - png_const_charp warning_message),PNG_EMPTY); -#else -# define png_safe_warning 0/*dummy argument*/ -#endif - -PNG_INTERNAL_FUNCTION(int,png_safe_execute,(png_imagep image, - int (*function)(png_voidp), png_voidp arg),PNG_EMPTY); - -/* Utility to log an error; this also cleans up the png_image; the function - * always returns 0 (false). - */ -PNG_INTERNAL_FUNCTION(int,png_image_error,(png_imagep image, - png_const_charp error_message),PNG_EMPTY); - -#ifndef PNG_SIMPLIFIED_READ_SUPPORTED -/* png_image_free is used by the write code but not exported */ -PNG_INTERNAL_FUNCTION(void, png_image_free, (png_imagep image), PNG_EMPTY); -#endif /* !SIMPLIFIED_READ */ - -#endif /* SIMPLIFIED READ/WRITE */ - -/* These are initialization functions for hardware specific PNG filter - * optimizations; list these here then select the appropriate one at compile - * time using the macro PNG_FILTER_OPTIMIZATIONS. If the macro is not defined - * the generic code is used. - */ -#ifdef PNG_FILTER_OPTIMIZATIONS -PNG_INTERNAL_FUNCTION(void, PNG_FILTER_OPTIMIZATIONS, (png_structp png_ptr, - unsigned int bpp), PNG_EMPTY); - /* Just declare the optimization that will be used */ -#else - /* List *all* the possible optimizations here - this branch is required if - * the builder of libpng passes the definition of PNG_FILTER_OPTIMIZATIONS in - * CFLAGS in place of CPPFLAGS *and* uses symbol prefixing. - */ -# if PNG_ARM_NEON_OPT > 0 -PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_neon, - (png_structp png_ptr, unsigned int bpp), PNG_EMPTY); -#endif - -#if PNG_MIPS_MSA_IMPLEMENTATION == 1 -PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_mips, - (png_structp png_ptr, unsigned int bpp), PNG_EMPTY); -#endif - -# if PNG_MIPS_MMI_IMPLEMENTATION > 0 -PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_mips, - (png_structp png_ptr, unsigned int bpp), PNG_EMPTY); -# endif - -# if PNG_INTEL_SSE_IMPLEMENTATION > 0 -PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_sse2, - (png_structp png_ptr, unsigned int bpp), PNG_EMPTY); -# endif -#endif - -#if PNG_LOONGARCH_LSX_OPT > 0 -PNG_INTERNAL_FUNCTION(void, png_init_filter_functions_lsx, - (png_structp png_ptr, unsigned int bpp), PNG_EMPTY); -#endif - -PNG_INTERNAL_FUNCTION(png_uint_32, png_check_keyword, (png_structrp png_ptr, - png_const_charp key, png_bytep new_key), PNG_EMPTY); - -#if PNG_ARM_NEON_IMPLEMENTATION == 1 -PNG_INTERNAL_FUNCTION(void, - png_riffle_palette_neon, - (png_structrp), - PNG_EMPTY); -PNG_INTERNAL_FUNCTION(int, - png_do_expand_palette_rgba8_neon, - (png_structrp, - png_row_infop, - png_const_bytep, - const png_bytepp, - const png_bytepp), - PNG_EMPTY); -PNG_INTERNAL_FUNCTION(int, - png_do_expand_palette_rgb8_neon, - (png_structrp, - png_row_infop, - png_const_bytep, - const png_bytepp, - const png_bytepp), - PNG_EMPTY); -#endif - -/* Maintainer: Put new private prototypes here ^ */ - -#include "pngdebug.h" - -#ifdef __cplusplus -} -#endif - -#endif /* PNG_VERSION_INFO_ONLY */ -#endif /* PNGPRIV_H */ diff --git a/dep/libpng/src/pngread.c b/dep/libpng/src/pngread.c deleted file mode 100644 index 07a39df6e..000000000 --- a/dep/libpng/src/pngread.c +++ /dev/null @@ -1,4228 +0,0 @@ - -/* pngread.c - read a PNG file - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - * - * This file contains routines that an application calls directly to - * read a PNG file or stream. - */ - -#include "pngpriv.h" -#if defined(PNG_SIMPLIFIED_READ_SUPPORTED) && defined(PNG_STDIO_SUPPORTED) -# include -#endif - -#ifdef PNG_READ_SUPPORTED - -/* Create a PNG structure for reading, and allocate any memory needed. */ -PNG_FUNCTION(png_structp,PNGAPI -png_create_read_struct,(png_const_charp user_png_ver, png_voidp error_ptr, - png_error_ptr error_fn, png_error_ptr warn_fn),PNG_ALLOCATED) -{ -#ifndef PNG_USER_MEM_SUPPORTED - png_structp png_ptr = png_create_png_struct(user_png_ver, error_ptr, - error_fn, warn_fn, NULL, NULL, NULL); -#else - return png_create_read_struct_2(user_png_ver, error_ptr, error_fn, - warn_fn, NULL, NULL, NULL); -} - -/* Alternate create PNG structure for reading, and allocate any memory - * needed. - */ -PNG_FUNCTION(png_structp,PNGAPI -png_create_read_struct_2,(png_const_charp user_png_ver, png_voidp error_ptr, - png_error_ptr error_fn, png_error_ptr warn_fn, png_voidp mem_ptr, - png_malloc_ptr malloc_fn, png_free_ptr free_fn),PNG_ALLOCATED) -{ - png_structp png_ptr = png_create_png_struct(user_png_ver, error_ptr, - error_fn, warn_fn, mem_ptr, malloc_fn, free_fn); -#endif /* USER_MEM */ - - if (png_ptr != NULL) - { - png_ptr->mode = PNG_IS_READ_STRUCT; - - /* Added in libpng-1.6.0; this can be used to detect a read structure if - * required (it will be zero in a write structure.) - */ -# ifdef PNG_SEQUENTIAL_READ_SUPPORTED - png_ptr->IDAT_read_size = PNG_IDAT_READ_SIZE; -# endif - -# ifdef PNG_BENIGN_READ_ERRORS_SUPPORTED - png_ptr->flags |= PNG_FLAG_BENIGN_ERRORS_WARN; - - /* In stable builds only warn if an application error can be completely - * handled. - */ -# if PNG_RELEASE_BUILD - png_ptr->flags |= PNG_FLAG_APP_WARNINGS_WARN; -# endif -# endif - - /* TODO: delay this, it can be done in png_init_io (if the app doesn't - * do it itself) avoiding setting the default function if it is not - * required. - */ - png_set_read_fn(png_ptr, NULL, NULL); - } - - return png_ptr; -} - - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -/* Read the information before the actual image data. This has been - * changed in v0.90 to allow reading a file that already has the magic - * bytes read from the stream. You can tell libpng how many bytes have - * been read from the beginning of the stream (up to the maximum of 8) - * via png_set_sig_bytes(), and we will only check the remaining bytes - * here. The application can then have access to the signature bytes we - * read if it is determined that this isn't a valid PNG file. - */ -void PNGAPI -png_read_info(png_structrp png_ptr, png_inforp info_ptr) -{ -#ifdef PNG_HANDLE_AS_UNKNOWN_SUPPORTED - int keep; -#endif - - png_debug(1, "in png_read_info"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - /* Read and check the PNG file signature. */ - png_read_sig(png_ptr, info_ptr); - - for (;;) - { - png_uint_32 length = png_read_chunk_header(png_ptr); - png_uint_32 chunk_name = png_ptr->chunk_name; - - /* IDAT logic needs to happen here to simplify getting the two flags - * right. - */ - if (chunk_name == png_IDAT) - { - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "Missing IHDR before IDAT"); - - else if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE && - (png_ptr->mode & PNG_HAVE_PLTE) == 0) - png_chunk_error(png_ptr, "Missing PLTE before IDAT"); - - else if ((png_ptr->mode & PNG_AFTER_IDAT) != 0) - png_chunk_benign_error(png_ptr, "Too many IDATs found"); - - png_ptr->mode |= PNG_HAVE_IDAT; - } - - else if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - { - png_ptr->mode |= PNG_HAVE_CHUNK_AFTER_IDAT; - png_ptr->mode |= PNG_AFTER_IDAT; - } - - /* This should be a binary subdivision search or a hash for - * matching the chunk name rather than a linear search. - */ - if (chunk_name == png_IHDR) - png_handle_IHDR(png_ptr, info_ptr, length); - - else if (chunk_name == png_IEND) - png_handle_IEND(png_ptr, info_ptr, length); - -#ifdef PNG_HANDLE_AS_UNKNOWN_SUPPORTED - else if ((keep = png_chunk_unknown_handling(png_ptr, chunk_name)) != 0) - { - png_handle_unknown(png_ptr, info_ptr, length, keep); - - if (chunk_name == png_PLTE) - png_ptr->mode |= PNG_HAVE_PLTE; - - else if (chunk_name == png_IDAT) - { - png_ptr->idat_size = 0; /* It has been consumed */ - break; - } - } -#endif - else if (chunk_name == png_PLTE) - png_handle_PLTE(png_ptr, info_ptr, length); - - else if (chunk_name == png_IDAT) - { - png_ptr->idat_size = length; - break; - } - -#ifdef PNG_READ_bKGD_SUPPORTED - else if (chunk_name == png_bKGD) - png_handle_bKGD(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_cHRM_SUPPORTED - else if (chunk_name == png_cHRM) - png_handle_cHRM(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_eXIf_SUPPORTED - else if (chunk_name == png_eXIf) - png_handle_eXIf(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_gAMA_SUPPORTED - else if (chunk_name == png_gAMA) - png_handle_gAMA(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_hIST_SUPPORTED - else if (chunk_name == png_hIST) - png_handle_hIST(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_oFFs_SUPPORTED - else if (chunk_name == png_oFFs) - png_handle_oFFs(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_pCAL_SUPPORTED - else if (chunk_name == png_pCAL) - png_handle_pCAL(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_sCAL_SUPPORTED - else if (chunk_name == png_sCAL) - png_handle_sCAL(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_pHYs_SUPPORTED - else if (chunk_name == png_pHYs) - png_handle_pHYs(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_sBIT_SUPPORTED - else if (chunk_name == png_sBIT) - png_handle_sBIT(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_sRGB_SUPPORTED - else if (chunk_name == png_sRGB) - png_handle_sRGB(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_iCCP_SUPPORTED - else if (chunk_name == png_iCCP) - png_handle_iCCP(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_sPLT_SUPPORTED - else if (chunk_name == png_sPLT) - png_handle_sPLT(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_tEXt_SUPPORTED - else if (chunk_name == png_tEXt) - png_handle_tEXt(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_tIME_SUPPORTED - else if (chunk_name == png_tIME) - png_handle_tIME(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_tRNS_SUPPORTED - else if (chunk_name == png_tRNS) - png_handle_tRNS(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_zTXt_SUPPORTED - else if (chunk_name == png_zTXt) - png_handle_zTXt(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_iTXt_SUPPORTED - else if (chunk_name == png_iTXt) - png_handle_iTXt(png_ptr, info_ptr, length); -#endif - - else - png_handle_unknown(png_ptr, info_ptr, length, - PNG_HANDLE_CHUNK_AS_DEFAULT); - } -} -#endif /* SEQUENTIAL_READ */ - -/* Optional call to update the users info_ptr structure */ -void PNGAPI -png_read_update_info(png_structrp png_ptr, png_inforp info_ptr) -{ - png_debug(1, "in png_read_update_info"); - - if (png_ptr != NULL) - { - if ((png_ptr->flags & PNG_FLAG_ROW_INIT) == 0) - { - png_read_start_row(png_ptr); - -# ifdef PNG_READ_TRANSFORMS_SUPPORTED - png_read_transform_info(png_ptr, info_ptr); -# else - PNG_UNUSED(info_ptr) -# endif - } - - /* New in 1.6.0 this avoids the bug of doing the initializations twice */ - else - png_app_error(png_ptr, - "png_read_update_info/png_start_read_image: duplicate call"); - } -} - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -/* Initialize palette, background, etc, after transformations - * are set, but before any reading takes place. This allows - * the user to obtain a gamma-corrected palette, for example. - * If the user doesn't call this, we will do it ourselves. - */ -void PNGAPI -png_start_read_image(png_structrp png_ptr) -{ - png_debug(1, "in png_start_read_image"); - - if (png_ptr != NULL) - { - if ((png_ptr->flags & PNG_FLAG_ROW_INIT) == 0) - png_read_start_row(png_ptr); - - /* New in 1.6.0 this avoids the bug of doing the initializations twice */ - else - png_app_error(png_ptr, - "png_start_read_image/png_read_update_info: duplicate call"); - } -} -#endif /* SEQUENTIAL_READ */ - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -#ifdef PNG_MNG_FEATURES_SUPPORTED -/* Undoes intrapixel differencing, - * NOTE: this is apparently only supported in the 'sequential' reader. - */ -static void -png_do_read_intrapixel(png_row_infop row_info, png_bytep row) -{ - png_debug(1, "in png_do_read_intrapixel"); - - if ( - (row_info->color_type & PNG_COLOR_MASK_COLOR) != 0) - { - int bytes_per_pixel; - png_uint_32 row_width = row_info->width; - - if (row_info->bit_depth == 8) - { - png_bytep rp; - png_uint_32 i; - - if (row_info->color_type == PNG_COLOR_TYPE_RGB) - bytes_per_pixel = 3; - - else if (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - bytes_per_pixel = 4; - - else - return; - - for (i = 0, rp = row; i < row_width; i++, rp += bytes_per_pixel) - { - *(rp) = (png_byte)((256 + *rp + *(rp + 1)) & 0xff); - *(rp+2) = (png_byte)((256 + *(rp + 2) + *(rp + 1)) & 0xff); - } - } - else if (row_info->bit_depth == 16) - { - png_bytep rp; - png_uint_32 i; - - if (row_info->color_type == PNG_COLOR_TYPE_RGB) - bytes_per_pixel = 6; - - else if (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - bytes_per_pixel = 8; - - else - return; - - for (i = 0, rp = row; i < row_width; i++, rp += bytes_per_pixel) - { - png_uint_32 s0 = (png_uint_32)(*(rp ) << 8) | *(rp + 1); - png_uint_32 s1 = (png_uint_32)(*(rp + 2) << 8) | *(rp + 3); - png_uint_32 s2 = (png_uint_32)(*(rp + 4) << 8) | *(rp + 5); - png_uint_32 red = (s0 + s1 + 65536) & 0xffff; - png_uint_32 blue = (s2 + s1 + 65536) & 0xffff; - *(rp ) = (png_byte)((red >> 8) & 0xff); - *(rp + 1) = (png_byte)(red & 0xff); - *(rp + 4) = (png_byte)((blue >> 8) & 0xff); - *(rp + 5) = (png_byte)(blue & 0xff); - } - } - } -} -#endif /* MNG_FEATURES */ - -void PNGAPI -png_read_row(png_structrp png_ptr, png_bytep row, png_bytep dsp_row) -{ - png_row_info row_info; - - if (png_ptr == NULL) - return; - - png_debug2(1, "in png_read_row (row %lu, pass %d)", - (unsigned long)png_ptr->row_number, png_ptr->pass); - - /* png_read_start_row sets the information (in particular iwidth) for this - * interlace pass. - */ - if ((png_ptr->flags & PNG_FLAG_ROW_INIT) == 0) - png_read_start_row(png_ptr); - - /* 1.5.6: row_info moved out of png_struct to a local here. */ - row_info.width = png_ptr->iwidth; /* NOTE: width of current interlaced row */ - row_info.color_type = png_ptr->color_type; - row_info.bit_depth = png_ptr->bit_depth; - row_info.channels = png_ptr->channels; - row_info.pixel_depth = png_ptr->pixel_depth; - row_info.rowbytes = PNG_ROWBYTES(row_info.pixel_depth, row_info.width); - -#ifdef PNG_WARNINGS_SUPPORTED - if (png_ptr->row_number == 0 && png_ptr->pass == 0) - { - /* Check for transforms that have been set but were defined out */ -#if defined(PNG_WRITE_INVERT_SUPPORTED) && !defined(PNG_READ_INVERT_SUPPORTED) - if ((png_ptr->transformations & PNG_INVERT_MONO) != 0) - png_warning(png_ptr, "PNG_READ_INVERT_SUPPORTED is not defined"); -#endif - -#if defined(PNG_WRITE_FILLER_SUPPORTED) && !defined(PNG_READ_FILLER_SUPPORTED) - if ((png_ptr->transformations & PNG_FILLER) != 0) - png_warning(png_ptr, "PNG_READ_FILLER_SUPPORTED is not defined"); -#endif - -#if defined(PNG_WRITE_PACKSWAP_SUPPORTED) && \ - !defined(PNG_READ_PACKSWAP_SUPPORTED) - if ((png_ptr->transformations & PNG_PACKSWAP) != 0) - png_warning(png_ptr, "PNG_READ_PACKSWAP_SUPPORTED is not defined"); -#endif - -#if defined(PNG_WRITE_PACK_SUPPORTED) && !defined(PNG_READ_PACK_SUPPORTED) - if ((png_ptr->transformations & PNG_PACK) != 0) - png_warning(png_ptr, "PNG_READ_PACK_SUPPORTED is not defined"); -#endif - -#if defined(PNG_WRITE_SHIFT_SUPPORTED) && !defined(PNG_READ_SHIFT_SUPPORTED) - if ((png_ptr->transformations & PNG_SHIFT) != 0) - png_warning(png_ptr, "PNG_READ_SHIFT_SUPPORTED is not defined"); -#endif - -#if defined(PNG_WRITE_BGR_SUPPORTED) && !defined(PNG_READ_BGR_SUPPORTED) - if ((png_ptr->transformations & PNG_BGR) != 0) - png_warning(png_ptr, "PNG_READ_BGR_SUPPORTED is not defined"); -#endif - -#if defined(PNG_WRITE_SWAP_SUPPORTED) && !defined(PNG_READ_SWAP_SUPPORTED) - if ((png_ptr->transformations & PNG_SWAP_BYTES) != 0) - png_warning(png_ptr, "PNG_READ_SWAP_SUPPORTED is not defined"); -#endif - } -#endif /* WARNINGS */ - -#ifdef PNG_READ_INTERLACING_SUPPORTED - /* If interlaced and we do not need a new row, combine row and return. - * Notice that the pixels we have from previous rows have been transformed - * already; we can only combine like with like (transformed or - * untransformed) and, because of the libpng API for interlaced images, this - * means we must transform before de-interlacing. - */ - if (png_ptr->interlaced != 0 && - (png_ptr->transformations & PNG_INTERLACE) != 0) - { - switch (png_ptr->pass) - { - case 0: - if (png_ptr->row_number & 0x07) - { - if (dsp_row != NULL) - png_combine_row(png_ptr, dsp_row, 1/*display*/); - png_read_finish_row(png_ptr); - return; - } - break; - - case 1: - if ((png_ptr->row_number & 0x07) || png_ptr->width < 5) - { - if (dsp_row != NULL) - png_combine_row(png_ptr, dsp_row, 1/*display*/); - - png_read_finish_row(png_ptr); - return; - } - break; - - case 2: - if ((png_ptr->row_number & 0x07) != 4) - { - if (dsp_row != NULL && (png_ptr->row_number & 4)) - png_combine_row(png_ptr, dsp_row, 1/*display*/); - - png_read_finish_row(png_ptr); - return; - } - break; - - case 3: - if ((png_ptr->row_number & 3) || png_ptr->width < 3) - { - if (dsp_row != NULL) - png_combine_row(png_ptr, dsp_row, 1/*display*/); - - png_read_finish_row(png_ptr); - return; - } - break; - - case 4: - if ((png_ptr->row_number & 3) != 2) - { - if (dsp_row != NULL && (png_ptr->row_number & 2)) - png_combine_row(png_ptr, dsp_row, 1/*display*/); - - png_read_finish_row(png_ptr); - return; - } - break; - - case 5: - if ((png_ptr->row_number & 1) || png_ptr->width < 2) - { - if (dsp_row != NULL) - png_combine_row(png_ptr, dsp_row, 1/*display*/); - - png_read_finish_row(png_ptr); - return; - } - break; - - default: - case 6: - if ((png_ptr->row_number & 1) == 0) - { - png_read_finish_row(png_ptr); - return; - } - break; - } - } -#endif - - if ((png_ptr->mode & PNG_HAVE_IDAT) == 0) - png_error(png_ptr, "Invalid attempt to read row data"); - - /* Fill the row with IDAT data: */ - png_ptr->row_buf[0]=255; /* to force error if no data was found */ - png_read_IDAT_data(png_ptr, png_ptr->row_buf, row_info.rowbytes + 1); - - if (png_ptr->row_buf[0] > PNG_FILTER_VALUE_NONE) - { - if (png_ptr->row_buf[0] < PNG_FILTER_VALUE_LAST) - png_read_filter_row(png_ptr, &row_info, png_ptr->row_buf + 1, - png_ptr->prev_row + 1, png_ptr->row_buf[0]); - else - png_error(png_ptr, "bad adaptive filter value"); - } - - /* libpng 1.5.6: the following line was copying png_ptr->rowbytes before - * 1.5.6, while the buffer really is this big in current versions of libpng - * it may not be in the future, so this was changed just to copy the - * interlaced count: - */ - memcpy(png_ptr->prev_row, png_ptr->row_buf, row_info.rowbytes + 1); - -#ifdef PNG_MNG_FEATURES_SUPPORTED - if ((png_ptr->mng_features_permitted & PNG_FLAG_MNG_FILTER_64) != 0 && - (png_ptr->filter_type == PNG_INTRAPIXEL_DIFFERENCING)) - { - /* Intrapixel differencing */ - png_do_read_intrapixel(&row_info, png_ptr->row_buf + 1); - } -#endif - -#ifdef PNG_READ_TRANSFORMS_SUPPORTED - if (png_ptr->transformations -# ifdef PNG_CHECK_FOR_INVALID_INDEX_SUPPORTED - || png_ptr->num_palette_max >= 0 -# endif - ) - png_do_read_transformations(png_ptr, &row_info); -#endif - - /* The transformed pixel depth should match the depth now in row_info. */ - if (png_ptr->transformed_pixel_depth == 0) - { - png_ptr->transformed_pixel_depth = row_info.pixel_depth; - if (row_info.pixel_depth > png_ptr->maximum_pixel_depth) - png_error(png_ptr, "sequential row overflow"); - } - - else if (png_ptr->transformed_pixel_depth != row_info.pixel_depth) - png_error(png_ptr, "internal sequential row size calculation error"); - -#ifdef PNG_READ_INTERLACING_SUPPORTED - /* Expand interlaced rows to full size */ - if (png_ptr->interlaced != 0 && - (png_ptr->transformations & PNG_INTERLACE) != 0) - { - if (png_ptr->pass < 6) - png_do_read_interlace(&row_info, png_ptr->row_buf + 1, png_ptr->pass, - png_ptr->transformations); - - if (dsp_row != NULL) - png_combine_row(png_ptr, dsp_row, 1/*display*/); - - if (row != NULL) - png_combine_row(png_ptr, row, 0/*row*/); - } - - else -#endif - { - if (row != NULL) - png_combine_row(png_ptr, row, -1/*ignored*/); - - if (dsp_row != NULL) - png_combine_row(png_ptr, dsp_row, -1/*ignored*/); - } - png_read_finish_row(png_ptr); - - if (png_ptr->read_row_fn != NULL) - (*(png_ptr->read_row_fn))(png_ptr, png_ptr->row_number, png_ptr->pass); - -} -#endif /* SEQUENTIAL_READ */ - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -/* Read one or more rows of image data. If the image is interlaced, - * and png_set_interlace_handling() has been called, the rows need to - * contain the contents of the rows from the previous pass. If the - * image has alpha or transparency, and png_handle_alpha()[*] has been - * called, the rows contents must be initialized to the contents of the - * screen. - * - * "row" holds the actual image, and pixels are placed in it - * as they arrive. If the image is displayed after each pass, it will - * appear to "sparkle" in. "display_row" can be used to display a - * "chunky" progressive image, with finer detail added as it becomes - * available. If you do not want this "chunky" display, you may pass - * NULL for display_row. If you do not want the sparkle display, and - * you have not called png_handle_alpha(), you may pass NULL for rows. - * If you have called png_handle_alpha(), and the image has either an - * alpha channel or a transparency chunk, you must provide a buffer for - * rows. In this case, you do not have to provide a display_row buffer - * also, but you may. If the image is not interlaced, or if you have - * not called png_set_interlace_handling(), the display_row buffer will - * be ignored, so pass NULL to it. - * - * [*] png_handle_alpha() does not exist yet, as of this version of libpng - */ - -void PNGAPI -png_read_rows(png_structrp png_ptr, png_bytepp row, - png_bytepp display_row, png_uint_32 num_rows) -{ - png_uint_32 i; - png_bytepp rp; - png_bytepp dp; - - png_debug(1, "in png_read_rows"); - - if (png_ptr == NULL) - return; - - rp = row; - dp = display_row; - if (rp != NULL && dp != NULL) - for (i = 0; i < num_rows; i++) - { - png_bytep rptr = *rp++; - png_bytep dptr = *dp++; - - png_read_row(png_ptr, rptr, dptr); - } - - else if (rp != NULL) - for (i = 0; i < num_rows; i++) - { - png_bytep rptr = *rp; - png_read_row(png_ptr, rptr, NULL); - rp++; - } - - else if (dp != NULL) - for (i = 0; i < num_rows; i++) - { - png_bytep dptr = *dp; - png_read_row(png_ptr, NULL, dptr); - dp++; - } -} -#endif /* SEQUENTIAL_READ */ - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -/* Read the entire image. If the image has an alpha channel or a tRNS - * chunk, and you have called png_handle_alpha()[*], you will need to - * initialize the image to the current image that PNG will be overlaying. - * We set the num_rows again here, in case it was incorrectly set in - * png_read_start_row() by a call to png_read_update_info() or - * png_start_read_image() if png_set_interlace_handling() wasn't called - * prior to either of these functions like it should have been. You can - * only call this function once. If you desire to have an image for - * each pass of a interlaced image, use png_read_rows() instead. - * - * [*] png_handle_alpha() does not exist yet, as of this version of libpng - */ -void PNGAPI -png_read_image(png_structrp png_ptr, png_bytepp image) -{ - png_uint_32 i, image_height; - int pass, j; - png_bytepp rp; - - png_debug(1, "in png_read_image"); - - if (png_ptr == NULL) - return; - -#ifdef PNG_READ_INTERLACING_SUPPORTED - if ((png_ptr->flags & PNG_FLAG_ROW_INIT) == 0) - { - pass = png_set_interlace_handling(png_ptr); - /* And make sure transforms are initialized. */ - png_start_read_image(png_ptr); - } - else - { - if (png_ptr->interlaced != 0 && - (png_ptr->transformations & PNG_INTERLACE) == 0) - { - /* Caller called png_start_read_image or png_read_update_info without - * first turning on the PNG_INTERLACE transform. We can fix this here, - * but the caller should do it! - */ - png_warning(png_ptr, "Interlace handling should be turned on when " - "using png_read_image"); - /* Make sure this is set correctly */ - png_ptr->num_rows = png_ptr->height; - } - - /* Obtain the pass number, which also turns on the PNG_INTERLACE flag in - * the above error case. - */ - pass = png_set_interlace_handling(png_ptr); - } -#else - if (png_ptr->interlaced) - png_error(png_ptr, - "Cannot read interlaced image -- interlace handler disabled"); - - pass = 1; -#endif - - image_height=png_ptr->height; - - for (j = 0; j < pass; j++) - { - rp = image; - for (i = 0; i < image_height; i++) - { - png_read_row(png_ptr, *rp, NULL); - rp++; - } - } -} -#endif /* SEQUENTIAL_READ */ - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -/* Read the end of the PNG file. Will not read past the end of the - * file, will verify the end is accurate, and will read any comments - * or time information at the end of the file, if info is not NULL. - */ -void PNGAPI -png_read_end(png_structrp png_ptr, png_inforp info_ptr) -{ -#ifdef PNG_HANDLE_AS_UNKNOWN_SUPPORTED - int keep; -#endif - - png_debug(1, "in png_read_end"); - - if (png_ptr == NULL) - return; - - /* If png_read_end is called in the middle of reading the rows there may - * still be pending IDAT data and an owned zstream. Deal with this here. - */ -#ifdef PNG_HANDLE_AS_UNKNOWN_SUPPORTED - if (png_chunk_unknown_handling(png_ptr, png_IDAT) == 0) -#endif - png_read_finish_IDAT(png_ptr); - -#ifdef PNG_READ_CHECK_FOR_INVALID_INDEX_SUPPORTED - /* Report invalid palette index; added at libng-1.5.10 */ - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE && - png_ptr->num_palette_max >= png_ptr->num_palette) - png_benign_error(png_ptr, "Read palette index exceeding num_palette"); -#endif - - do - { - png_uint_32 length = png_read_chunk_header(png_ptr); - png_uint_32 chunk_name = png_ptr->chunk_name; - - if (chunk_name != png_IDAT) - png_ptr->mode |= PNG_HAVE_CHUNK_AFTER_IDAT; - - if (chunk_name == png_IEND) - png_handle_IEND(png_ptr, info_ptr, length); - - else if (chunk_name == png_IHDR) - png_handle_IHDR(png_ptr, info_ptr, length); - - else if (info_ptr == NULL) - png_crc_finish(png_ptr, length); - -#ifdef PNG_HANDLE_AS_UNKNOWN_SUPPORTED - else if ((keep = png_chunk_unknown_handling(png_ptr, chunk_name)) != 0) - { - if (chunk_name == png_IDAT) - { - if ((length > 0 && !(png_ptr->flags & PNG_FLAG_ZSTREAM_ENDED)) - || (png_ptr->mode & PNG_HAVE_CHUNK_AFTER_IDAT) != 0) - png_benign_error(png_ptr, ".Too many IDATs found"); - } - png_handle_unknown(png_ptr, info_ptr, length, keep); - if (chunk_name == png_PLTE) - png_ptr->mode |= PNG_HAVE_PLTE; - } -#endif - - else if (chunk_name == png_IDAT) - { - /* Zero length IDATs are legal after the last IDAT has been - * read, but not after other chunks have been read. 1.6 does not - * always read all the deflate data; specifically it cannot be relied - * upon to read the Adler32 at the end. If it doesn't ignore IDAT - * chunks which are longer than zero as well: - */ - if ((length > 0 && !(png_ptr->flags & PNG_FLAG_ZSTREAM_ENDED)) - || (png_ptr->mode & PNG_HAVE_CHUNK_AFTER_IDAT) != 0) - png_benign_error(png_ptr, "..Too many IDATs found"); - - png_crc_finish(png_ptr, length); - } - else if (chunk_name == png_PLTE) - png_handle_PLTE(png_ptr, info_ptr, length); - -#ifdef PNG_READ_bKGD_SUPPORTED - else if (chunk_name == png_bKGD) - png_handle_bKGD(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_cHRM_SUPPORTED - else if (chunk_name == png_cHRM) - png_handle_cHRM(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_eXIf_SUPPORTED - else if (chunk_name == png_eXIf) - png_handle_eXIf(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_gAMA_SUPPORTED - else if (chunk_name == png_gAMA) - png_handle_gAMA(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_hIST_SUPPORTED - else if (chunk_name == png_hIST) - png_handle_hIST(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_oFFs_SUPPORTED - else if (chunk_name == png_oFFs) - png_handle_oFFs(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_pCAL_SUPPORTED - else if (chunk_name == png_pCAL) - png_handle_pCAL(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_sCAL_SUPPORTED - else if (chunk_name == png_sCAL) - png_handle_sCAL(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_pHYs_SUPPORTED - else if (chunk_name == png_pHYs) - png_handle_pHYs(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_sBIT_SUPPORTED - else if (chunk_name == png_sBIT) - png_handle_sBIT(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_sRGB_SUPPORTED - else if (chunk_name == png_sRGB) - png_handle_sRGB(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_iCCP_SUPPORTED - else if (chunk_name == png_iCCP) - png_handle_iCCP(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_sPLT_SUPPORTED - else if (chunk_name == png_sPLT) - png_handle_sPLT(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_tEXt_SUPPORTED - else if (chunk_name == png_tEXt) - png_handle_tEXt(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_tIME_SUPPORTED - else if (chunk_name == png_tIME) - png_handle_tIME(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_tRNS_SUPPORTED - else if (chunk_name == png_tRNS) - png_handle_tRNS(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_zTXt_SUPPORTED - else if (chunk_name == png_zTXt) - png_handle_zTXt(png_ptr, info_ptr, length); -#endif - -#ifdef PNG_READ_iTXt_SUPPORTED - else if (chunk_name == png_iTXt) - png_handle_iTXt(png_ptr, info_ptr, length); -#endif - - else - png_handle_unknown(png_ptr, info_ptr, length, - PNG_HANDLE_CHUNK_AS_DEFAULT); - } while ((png_ptr->mode & PNG_HAVE_IEND) == 0); -} -#endif /* SEQUENTIAL_READ */ - -/* Free all memory used in the read struct */ -static void -png_read_destroy(png_structrp png_ptr) -{ - png_debug(1, "in png_read_destroy"); - -#ifdef PNG_READ_GAMMA_SUPPORTED - png_destroy_gamma_table(png_ptr); -#endif - - png_free(png_ptr, png_ptr->big_row_buf); - png_ptr->big_row_buf = NULL; - png_free(png_ptr, png_ptr->big_prev_row); - png_ptr->big_prev_row = NULL; - png_free(png_ptr, png_ptr->read_buffer); - png_ptr->read_buffer = NULL; - -#ifdef PNG_READ_QUANTIZE_SUPPORTED - png_free(png_ptr, png_ptr->palette_lookup); - png_ptr->palette_lookup = NULL; - png_free(png_ptr, png_ptr->quantize_index); - png_ptr->quantize_index = NULL; -#endif - - if ((png_ptr->free_me & PNG_FREE_PLTE) != 0) - { - png_zfree(png_ptr, png_ptr->palette); - png_ptr->palette = NULL; - } - png_ptr->free_me &= ~PNG_FREE_PLTE; - -#if defined(PNG_tRNS_SUPPORTED) || \ - defined(PNG_READ_EXPAND_SUPPORTED) || defined(PNG_READ_BACKGROUND_SUPPORTED) - if ((png_ptr->free_me & PNG_FREE_TRNS) != 0) - { - png_free(png_ptr, png_ptr->trans_alpha); - png_ptr->trans_alpha = NULL; - } - png_ptr->free_me &= ~PNG_FREE_TRNS; -#endif - - inflateEnd(&png_ptr->zstream); - -#ifdef PNG_PROGRESSIVE_READ_SUPPORTED - png_free(png_ptr, png_ptr->save_buffer); - png_ptr->save_buffer = NULL; -#endif - -#if defined(PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED) && \ - defined(PNG_READ_UNKNOWN_CHUNKS_SUPPORTED) - png_free(png_ptr, png_ptr->unknown_chunk.data); - png_ptr->unknown_chunk.data = NULL; -#endif - -#ifdef PNG_SET_UNKNOWN_CHUNKS_SUPPORTED - png_free(png_ptr, png_ptr->chunk_list); - png_ptr->chunk_list = NULL; -#endif - -#if defined(PNG_READ_EXPAND_SUPPORTED) && \ - defined(PNG_ARM_NEON_IMPLEMENTATION) - png_free(png_ptr, png_ptr->riffled_palette); - png_ptr->riffled_palette = NULL; -#endif - - /* NOTE: the 'setjmp' buffer may still be allocated and the memory and error - * callbacks are still set at this point. They are required to complete the - * destruction of the png_struct itself. - */ -} - -/* Free all memory used by the read */ -void PNGAPI -png_destroy_read_struct(png_structpp png_ptr_ptr, png_infopp info_ptr_ptr, - png_infopp end_info_ptr_ptr) -{ - png_structrp png_ptr = NULL; - - png_debug(1, "in png_destroy_read_struct"); - - if (png_ptr_ptr != NULL) - png_ptr = *png_ptr_ptr; - - if (png_ptr == NULL) - return; - - /* libpng 1.6.0: use the API to destroy info structs to ensure consistent - * behavior. Prior to 1.6.0 libpng did extra 'info' destruction in this API. - * The extra was, apparently, unnecessary yet this hides memory leak bugs. - */ - png_destroy_info_struct(png_ptr, end_info_ptr_ptr); - png_destroy_info_struct(png_ptr, info_ptr_ptr); - - *png_ptr_ptr = NULL; - png_read_destroy(png_ptr); - png_destroy_png_struct(png_ptr); -} - -void PNGAPI -png_set_read_status_fn(png_structrp png_ptr, png_read_status_ptr read_row_fn) -{ - if (png_ptr == NULL) - return; - - png_ptr->read_row_fn = read_row_fn; -} - - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -#ifdef PNG_INFO_IMAGE_SUPPORTED -void PNGAPI -png_read_png(png_structrp png_ptr, png_inforp info_ptr, - int transforms, voidp params) -{ - png_debug(1, "in png_read_png"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - /* png_read_info() gives us all of the information from the - * PNG file before the first IDAT (image data chunk). - */ - png_read_info(png_ptr, info_ptr); - if (info_ptr->height > PNG_UINT_32_MAX/(sizeof (png_bytep))) - png_error(png_ptr, "Image is too high to process with png_read_png()"); - - /* -------------- image transformations start here ------------------- */ - /* libpng 1.6.10: add code to cause a png_app_error if a selected TRANSFORM - * is not implemented. This will only happen in de-configured (non-default) - * libpng builds. The results can be unexpected - png_read_png may return - * short or mal-formed rows because the transform is skipped. - */ - - /* Tell libpng to strip 16-bit/color files down to 8 bits per color. - */ - if ((transforms & PNG_TRANSFORM_SCALE_16) != 0) - /* Added at libpng-1.5.4. "strip_16" produces the same result that it - * did in earlier versions, while "scale_16" is now more accurate. - */ -#ifdef PNG_READ_SCALE_16_TO_8_SUPPORTED - png_set_scale_16(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_SCALE_16 not supported"); -#endif - - /* If both SCALE and STRIP are required pngrtran will effectively cancel the - * latter by doing SCALE first. This is ok and allows apps not to check for - * which is supported to get the right answer. - */ - if ((transforms & PNG_TRANSFORM_STRIP_16) != 0) -#ifdef PNG_READ_STRIP_16_TO_8_SUPPORTED - png_set_strip_16(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_STRIP_16 not supported"); -#endif - - /* Strip alpha bytes from the input data without combining with - * the background (not recommended). - */ - if ((transforms & PNG_TRANSFORM_STRIP_ALPHA) != 0) -#ifdef PNG_READ_STRIP_ALPHA_SUPPORTED - png_set_strip_alpha(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_STRIP_ALPHA not supported"); -#endif - - /* Extract multiple pixels with bit depths of 1, 2, or 4 from a single - * byte into separate bytes (useful for paletted and grayscale images). - */ - if ((transforms & PNG_TRANSFORM_PACKING) != 0) -#ifdef PNG_READ_PACK_SUPPORTED - png_set_packing(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_PACKING not supported"); -#endif - - /* Change the order of packed pixels to least significant bit first - * (not useful if you are using png_set_packing). - */ - if ((transforms & PNG_TRANSFORM_PACKSWAP) != 0) -#ifdef PNG_READ_PACKSWAP_SUPPORTED - png_set_packswap(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_PACKSWAP not supported"); -#endif - - /* Expand paletted colors into true RGB triplets - * Expand grayscale images to full 8 bits from 1, 2, or 4 bits/pixel - * Expand paletted or RGB images with transparency to full alpha - * channels so the data will be available as RGBA quartets. - */ - if ((transforms & PNG_TRANSFORM_EXPAND) != 0) -#ifdef PNG_READ_EXPAND_SUPPORTED - png_set_expand(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_EXPAND not supported"); -#endif - - /* We don't handle background color or gamma transformation or quantizing. - */ - - /* Invert monochrome files to have 0 as white and 1 as black - */ - if ((transforms & PNG_TRANSFORM_INVERT_MONO) != 0) -#ifdef PNG_READ_INVERT_SUPPORTED - png_set_invert_mono(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_INVERT_MONO not supported"); -#endif - - /* If you want to shift the pixel values from the range [0,255] or - * [0,65535] to the original [0,7] or [0,31], or whatever range the - * colors were originally in: - */ - if ((transforms & PNG_TRANSFORM_SHIFT) != 0) -#ifdef PNG_READ_SHIFT_SUPPORTED - if ((info_ptr->valid & PNG_INFO_sBIT) != 0) - png_set_shift(png_ptr, &info_ptr->sig_bit); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_SHIFT not supported"); -#endif - - /* Flip the RGB pixels to BGR (or RGBA to BGRA) */ - if ((transforms & PNG_TRANSFORM_BGR) != 0) -#ifdef PNG_READ_BGR_SUPPORTED - png_set_bgr(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_BGR not supported"); -#endif - - /* Swap the RGBA or GA data to ARGB or AG (or BGRA to ABGR) */ - if ((transforms & PNG_TRANSFORM_SWAP_ALPHA) != 0) -#ifdef PNG_READ_SWAP_ALPHA_SUPPORTED - png_set_swap_alpha(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_SWAP_ALPHA not supported"); -#endif - - /* Swap bytes of 16-bit files to least significant byte first */ - if ((transforms & PNG_TRANSFORM_SWAP_ENDIAN) != 0) -#ifdef PNG_READ_SWAP_SUPPORTED - png_set_swap(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_SWAP_ENDIAN not supported"); -#endif - -/* Added at libpng-1.2.41 */ - /* Invert the alpha channel from opacity to transparency */ - if ((transforms & PNG_TRANSFORM_INVERT_ALPHA) != 0) -#ifdef PNG_READ_INVERT_ALPHA_SUPPORTED - png_set_invert_alpha(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_INVERT_ALPHA not supported"); -#endif - -/* Added at libpng-1.2.41 */ - /* Expand grayscale image to RGB */ - if ((transforms & PNG_TRANSFORM_GRAY_TO_RGB) != 0) -#ifdef PNG_READ_GRAY_TO_RGB_SUPPORTED - png_set_gray_to_rgb(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_GRAY_TO_RGB not supported"); -#endif - -/* Added at libpng-1.5.4 */ - if ((transforms & PNG_TRANSFORM_EXPAND_16) != 0) -#ifdef PNG_READ_EXPAND_16_SUPPORTED - png_set_expand_16(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_EXPAND_16 not supported"); -#endif - - /* We don't handle adding filler bytes */ - - /* We use png_read_image and rely on that for interlace handling, but we also - * call png_read_update_info therefore must turn on interlace handling now: - */ - (void)png_set_interlace_handling(png_ptr); - - /* Optional call to gamma correct and add the background to the palette - * and update info structure. REQUIRED if you are expecting libpng to - * update the palette for you (i.e., you selected such a transform above). - */ - png_read_update_info(png_ptr, info_ptr); - - /* -------------- image transformations end here ------------------- */ - - png_free_data(png_ptr, info_ptr, PNG_FREE_ROWS, 0); - if (info_ptr->row_pointers == NULL) - { - png_uint_32 iptr; - - info_ptr->row_pointers = png_voidcast(png_bytepp, png_malloc(png_ptr, - info_ptr->height * (sizeof (png_bytep)))); - - for (iptr=0; iptrheight; iptr++) - info_ptr->row_pointers[iptr] = NULL; - - info_ptr->free_me |= PNG_FREE_ROWS; - - for (iptr = 0; iptr < info_ptr->height; iptr++) - info_ptr->row_pointers[iptr] = png_voidcast(png_bytep, - png_malloc(png_ptr, info_ptr->rowbytes)); - } - - png_read_image(png_ptr, info_ptr->row_pointers); - info_ptr->valid |= PNG_INFO_IDAT; - - /* Read rest of file, and get additional chunks in info_ptr - REQUIRED */ - png_read_end(png_ptr, info_ptr); - - PNG_UNUSED(params) -} -#endif /* INFO_IMAGE */ -#endif /* SEQUENTIAL_READ */ - -#ifdef PNG_SIMPLIFIED_READ_SUPPORTED -/* SIMPLIFIED READ - * - * This code currently relies on the sequential reader, though it could easily - * be made to work with the progressive one. - */ -/* Arguments to png_image_finish_read: */ - -/* Encoding of PNG data (used by the color-map code) */ -# define P_NOTSET 0 /* File encoding not yet known */ -# define P_sRGB 1 /* 8-bit encoded to sRGB gamma */ -# define P_LINEAR 2 /* 16-bit linear: not encoded, NOT pre-multiplied! */ -# define P_FILE 3 /* 8-bit encoded to file gamma, not sRGB or linear */ -# define P_LINEAR8 4 /* 8-bit linear: only from a file value */ - -/* Color-map processing: after libpng has run on the PNG image further - * processing may be needed to convert the data to color-map indices. - */ -#define PNG_CMAP_NONE 0 -#define PNG_CMAP_GA 1 /* Process GA data to a color-map with alpha */ -#define PNG_CMAP_TRANS 2 /* Process GA data to a background index */ -#define PNG_CMAP_RGB 3 /* Process RGB data */ -#define PNG_CMAP_RGB_ALPHA 4 /* Process RGBA data */ - -/* The following document where the background is for each processing case. */ -#define PNG_CMAP_NONE_BACKGROUND 256 -#define PNG_CMAP_GA_BACKGROUND 231 -#define PNG_CMAP_TRANS_BACKGROUND 254 -#define PNG_CMAP_RGB_BACKGROUND 256 -#define PNG_CMAP_RGB_ALPHA_BACKGROUND 216 - -typedef struct -{ - /* Arguments: */ - png_imagep image; - png_voidp buffer; - png_int_32 row_stride; - png_voidp colormap; - png_const_colorp background; - /* Local variables: */ - png_voidp local_row; - png_voidp first_row; - ptrdiff_t row_bytes; /* step between rows */ - int file_encoding; /* E_ values above */ - png_fixed_point gamma_to_linear; /* For P_FILE, reciprocal of gamma */ - int colormap_processing; /* PNG_CMAP_ values above */ -} png_image_read_control; - -/* Do all the *safe* initialization - 'safe' means that png_error won't be - * called, so setting up the jmp_buf is not required. This means that anything - * called from here must *not* call png_malloc - it has to call png_malloc_warn - * instead so that control is returned safely back to this routine. - */ -static int -png_image_read_init(png_imagep image) -{ - if (image->opaque == NULL) - { - png_structp png_ptr = png_create_read_struct(PNG_LIBPNG_VER_STRING, image, - png_safe_error, png_safe_warning); - - /* And set the rest of the structure to NULL to ensure that the various - * fields are consistent. - */ - memset(image, 0, (sizeof *image)); - image->version = PNG_IMAGE_VERSION; - - if (png_ptr != NULL) - { - png_infop info_ptr = png_create_info_struct(png_ptr); - - if (info_ptr != NULL) - { - png_controlp control = png_voidcast(png_controlp, - png_malloc_warn(png_ptr, (sizeof *control))); - - if (control != NULL) - { - memset(control, 0, (sizeof *control)); - - control->png_ptr = png_ptr; - control->info_ptr = info_ptr; - control->for_write = 0; - - image->opaque = control; - return 1; - } - - /* Error clean up */ - png_destroy_info_struct(png_ptr, &info_ptr); - } - - png_destroy_read_struct(&png_ptr, NULL, NULL); - } - - return png_image_error(image, "png_image_read: out of memory"); - } - - return png_image_error(image, "png_image_read: opaque pointer not NULL"); -} - -/* Utility to find the base format of a PNG file from a png_struct. */ -static png_uint_32 -png_image_format(png_structrp png_ptr) -{ - png_uint_32 format = 0; - - if ((png_ptr->color_type & PNG_COLOR_MASK_COLOR) != 0) - format |= PNG_FORMAT_FLAG_COLOR; - - if ((png_ptr->color_type & PNG_COLOR_MASK_ALPHA) != 0) - format |= PNG_FORMAT_FLAG_ALPHA; - - /* Use png_ptr here, not info_ptr, because by examination png_handle_tRNS - * sets the png_struct fields; that's all we are interested in here. The - * precise interaction with an app call to png_set_tRNS and PNG file reading - * is unclear. - */ - else if (png_ptr->num_trans > 0) - format |= PNG_FORMAT_FLAG_ALPHA; - - if (png_ptr->bit_depth == 16) - format |= PNG_FORMAT_FLAG_LINEAR; - - if ((png_ptr->color_type & PNG_COLOR_MASK_PALETTE) != 0) - format |= PNG_FORMAT_FLAG_COLORMAP; - - return format; -} - -/* Is the given gamma significantly different from sRGB? The test is the same - * one used in pngrtran.c when deciding whether to do gamma correction. The - * arithmetic optimizes the division by using the fact that the inverse of the - * file sRGB gamma is 2.2 - */ -static int -png_gamma_not_sRGB(png_fixed_point g) -{ - if (g < PNG_FP_1) - { - /* An uninitialized gamma is assumed to be sRGB for the simplified API. */ - if (g == 0) - return 0; - - return png_gamma_significant((g * 11 + 2)/5 /* i.e. *2.2, rounded */); - } - - return 1; -} - -/* Do the main body of a 'png_image_begin_read' function; read the PNG file - * header and fill in all the information. This is executed in a safe context, - * unlike the init routine above. - */ -static int -png_image_read_header(png_voidp argument) -{ - png_imagep image = png_voidcast(png_imagep, argument); - png_structrp png_ptr = image->opaque->png_ptr; - png_inforp info_ptr = image->opaque->info_ptr; - -#ifdef PNG_BENIGN_ERRORS_SUPPORTED - png_set_benign_errors(png_ptr, 1/*warn*/); -#endif - png_read_info(png_ptr, info_ptr); - - /* Do this the fast way; just read directly out of png_struct. */ - image->width = png_ptr->width; - image->height = png_ptr->height; - - { - png_uint_32 format = png_image_format(png_ptr); - - image->format = format; - -#ifdef PNG_COLORSPACE_SUPPORTED - /* Does the colorspace match sRGB? If there is no color endpoint - * (colorant) information assume yes, otherwise require the - * 'ENDPOINTS_MATCHP_sRGB' colorspace flag to have been set. If the - * colorspace has been determined to be invalid ignore it. - */ - if ((format & PNG_FORMAT_FLAG_COLOR) != 0 && ((png_ptr->colorspace.flags - & (PNG_COLORSPACE_HAVE_ENDPOINTS|PNG_COLORSPACE_ENDPOINTS_MATCH_sRGB| - PNG_COLORSPACE_INVALID)) == PNG_COLORSPACE_HAVE_ENDPOINTS)) - image->flags |= PNG_IMAGE_FLAG_COLORSPACE_NOT_sRGB; -#endif - } - - /* We need the maximum number of entries regardless of the format the - * application sets here. - */ - { - png_uint_32 cmap_entries; - - switch (png_ptr->color_type) - { - case PNG_COLOR_TYPE_GRAY: - cmap_entries = 1U << png_ptr->bit_depth; - break; - - case PNG_COLOR_TYPE_PALETTE: - cmap_entries = (png_uint_32)png_ptr->num_palette; - break; - - default: - cmap_entries = 256; - break; - } - - if (cmap_entries > 256) - cmap_entries = 256; - - image->colormap_entries = cmap_entries; - } - - return 1; -} - -#ifdef PNG_STDIO_SUPPORTED -int PNGAPI -png_image_begin_read_from_stdio(png_imagep image, FILE* file) -{ - if (image != NULL && image->version == PNG_IMAGE_VERSION) - { - if (file != NULL) - { - if (png_image_read_init(image) != 0) - { - /* This is slightly evil, but png_init_io doesn't do anything other - * than this and we haven't changed the standard IO functions so - * this saves a 'safe' function. - */ - image->opaque->png_ptr->io_ptr = file; - return png_safe_execute(image, png_image_read_header, image); - } - } - - else - return png_image_error(image, - "png_image_begin_read_from_stdio: invalid argument"); - } - - else if (image != NULL) - return png_image_error(image, - "png_image_begin_read_from_stdio: incorrect PNG_IMAGE_VERSION"); - - return 0; -} - -int PNGAPI -png_image_begin_read_from_file(png_imagep image, const char *file_name) -{ - if (image != NULL && image->version == PNG_IMAGE_VERSION) - { - if (file_name != NULL) - { - FILE *fp = fopen(file_name, "rb"); - - if (fp != NULL) - { - if (png_image_read_init(image) != 0) - { - image->opaque->png_ptr->io_ptr = fp; - image->opaque->owned_file = 1; - return png_safe_execute(image, png_image_read_header, image); - } - - /* Clean up: just the opened file. */ - (void)fclose(fp); - } - - else - return png_image_error(image, strerror(errno)); - } - - else - return png_image_error(image, - "png_image_begin_read_from_file: invalid argument"); - } - - else if (image != NULL) - return png_image_error(image, - "png_image_begin_read_from_file: incorrect PNG_IMAGE_VERSION"); - - return 0; -} -#endif /* STDIO */ - -static void PNGCBAPI -png_image_memory_read(png_structp png_ptr, png_bytep out, size_t need) -{ - if (png_ptr != NULL) - { - png_imagep image = png_voidcast(png_imagep, png_ptr->io_ptr); - if (image != NULL) - { - png_controlp cp = image->opaque; - if (cp != NULL) - { - png_const_bytep memory = cp->memory; - size_t size = cp->size; - - if (memory != NULL && size >= need) - { - memcpy(out, memory, need); - cp->memory = memory + need; - cp->size = size - need; - return; - } - - png_error(png_ptr, "read beyond end of data"); - } - } - - png_error(png_ptr, "invalid memory read"); - } -} - -int PNGAPI png_image_begin_read_from_memory(png_imagep image, - png_const_voidp memory, size_t size) -{ - if (image != NULL && image->version == PNG_IMAGE_VERSION) - { - if (memory != NULL && size > 0) - { - if (png_image_read_init(image) != 0) - { - /* Now set the IO functions to read from the memory buffer and - * store it into io_ptr. Again do this in-place to avoid calling a - * libpng function that requires error handling. - */ - image->opaque->memory = png_voidcast(png_const_bytep, memory); - image->opaque->size = size; - image->opaque->png_ptr->io_ptr = image; - image->opaque->png_ptr->read_data_fn = png_image_memory_read; - - return png_safe_execute(image, png_image_read_header, image); - } - } - - else - return png_image_error(image, - "png_image_begin_read_from_memory: invalid argument"); - } - - else if (image != NULL) - return png_image_error(image, - "png_image_begin_read_from_memory: incorrect PNG_IMAGE_VERSION"); - - return 0; -} - -/* Utility function to skip chunks that are not used by the simplified image - * read functions and an appropriate macro to call it. - */ -#ifdef PNG_HANDLE_AS_UNKNOWN_SUPPORTED -static void -png_image_skip_unused_chunks(png_structrp png_ptr) -{ - /* Prepare the reader to ignore all recognized chunks whose data will not - * be used, i.e., all chunks recognized by libpng except for those - * involved in basic image reading: - * - * IHDR, PLTE, IDAT, IEND - * - * Or image data handling: - * - * tRNS, bKGD, gAMA, cHRM, sRGB, [iCCP] and sBIT. - * - * This provides a small performance improvement and eliminates any - * potential vulnerability to security problems in the unused chunks. - * - * At present the iCCP chunk data isn't used, so iCCP chunk can be ignored - * too. This allows the simplified API to be compiled without iCCP support, - * however if the support is there the chunk is still checked to detect - * errors (which are unfortunately quite common.) - */ - { - static const png_byte chunks_to_process[] = { - 98, 75, 71, 68, '\0', /* bKGD */ - 99, 72, 82, 77, '\0', /* cHRM */ - 103, 65, 77, 65, '\0', /* gAMA */ -# ifdef PNG_READ_iCCP_SUPPORTED - 105, 67, 67, 80, '\0', /* iCCP */ -# endif - 115, 66, 73, 84, '\0', /* sBIT */ - 115, 82, 71, 66, '\0', /* sRGB */ - }; - - /* Ignore unknown chunks and all other chunks except for the - * IHDR, PLTE, tRNS, IDAT, and IEND chunks. - */ - png_set_keep_unknown_chunks(png_ptr, PNG_HANDLE_CHUNK_NEVER, - NULL, -1); - - /* But do not ignore image data handling chunks */ - png_set_keep_unknown_chunks(png_ptr, PNG_HANDLE_CHUNK_AS_DEFAULT, - chunks_to_process, (int)/*SAFE*/(sizeof chunks_to_process)/5); - } -} - -# define PNG_SKIP_CHUNKS(p) png_image_skip_unused_chunks(p) -#else -# define PNG_SKIP_CHUNKS(p) ((void)0) -#endif /* HANDLE_AS_UNKNOWN */ - -/* The following macro gives the exact rounded answer for all values in the - * range 0..255 (it actually divides by 51.2, but the rounding still generates - * the correct numbers 0..5 - */ -#define PNG_DIV51(v8) (((v8) * 5 + 130) >> 8) - -/* Utility functions to make particular color-maps */ -static void -set_file_encoding(png_image_read_control *display) -{ - png_fixed_point g = display->image->opaque->png_ptr->colorspace.gamma; - if (png_gamma_significant(g) != 0) - { - if (png_gamma_not_sRGB(g) != 0) - { - display->file_encoding = P_FILE; - display->gamma_to_linear = png_reciprocal(g); - } - - else - display->file_encoding = P_sRGB; - } - - else - display->file_encoding = P_LINEAR8; -} - -static unsigned int -decode_gamma(png_image_read_control *display, png_uint_32 value, int encoding) -{ - if (encoding == P_FILE) /* double check */ - encoding = display->file_encoding; - - if (encoding == P_NOTSET) /* must be the file encoding */ - { - set_file_encoding(display); - encoding = display->file_encoding; - } - - switch (encoding) - { - case P_FILE: - value = png_gamma_16bit_correct(value*257, display->gamma_to_linear); - break; - - case P_sRGB: - value = png_sRGB_table[value]; - break; - - case P_LINEAR: - break; - - case P_LINEAR8: - value *= 257; - break; - -#ifdef __GNUC__ - default: - png_error(display->image->opaque->png_ptr, - "unexpected encoding (internal error)"); -#endif - } - - return value; -} - -static png_uint_32 -png_colormap_compose(png_image_read_control *display, - png_uint_32 foreground, int foreground_encoding, png_uint_32 alpha, - png_uint_32 background, int encoding) -{ - /* The file value is composed on the background, the background has the given - * encoding and so does the result, the file is encoded with P_FILE and the - * file and alpha are 8-bit values. The (output) encoding will always be - * P_LINEAR or P_sRGB. - */ - png_uint_32 f = decode_gamma(display, foreground, foreground_encoding); - png_uint_32 b = decode_gamma(display, background, encoding); - - /* The alpha is always an 8-bit value (it comes from the palette), the value - * scaled by 255 is what PNG_sRGB_FROM_LINEAR requires. - */ - f = f * alpha + b * (255-alpha); - - if (encoding == P_LINEAR) - { - /* Scale to 65535; divide by 255, approximately (in fact this is extremely - * accurate, it divides by 255.00000005937181414556, with no overflow.) - */ - f *= 257; /* Now scaled by 65535 */ - f += f >> 16; - f = (f+32768) >> 16; - } - - else /* P_sRGB */ - f = PNG_sRGB_FROM_LINEAR(f); - - return f; -} - -/* NOTE: P_LINEAR values to this routine must be 16-bit, but P_FILE values must - * be 8-bit. - */ -static void -png_create_colormap_entry(png_image_read_control *display, - png_uint_32 ip, png_uint_32 red, png_uint_32 green, png_uint_32 blue, - png_uint_32 alpha, int encoding) -{ - png_imagep image = display->image; - int output_encoding = (image->format & PNG_FORMAT_FLAG_LINEAR) != 0 ? - P_LINEAR : P_sRGB; - int convert_to_Y = (image->format & PNG_FORMAT_FLAG_COLOR) == 0 && - (red != green || green != blue); - - if (ip > 255) - png_error(image->opaque->png_ptr, "color-map index out of range"); - - /* Update the cache with whether the file gamma is significantly different - * from sRGB. - */ - if (encoding == P_FILE) - { - if (display->file_encoding == P_NOTSET) - set_file_encoding(display); - - /* Note that the cached value may be P_FILE too, but if it is then the - * gamma_to_linear member has been set. - */ - encoding = display->file_encoding; - } - - if (encoding == P_FILE) - { - png_fixed_point g = display->gamma_to_linear; - - red = png_gamma_16bit_correct(red*257, g); - green = png_gamma_16bit_correct(green*257, g); - blue = png_gamma_16bit_correct(blue*257, g); - - if (convert_to_Y != 0 || output_encoding == P_LINEAR) - { - alpha *= 257; - encoding = P_LINEAR; - } - - else - { - red = PNG_sRGB_FROM_LINEAR(red * 255); - green = PNG_sRGB_FROM_LINEAR(green * 255); - blue = PNG_sRGB_FROM_LINEAR(blue * 255); - encoding = P_sRGB; - } - } - - else if (encoding == P_LINEAR8) - { - /* This encoding occurs quite frequently in test cases because PngSuite - * includes a gAMA 1.0 chunk with most images. - */ - red *= 257; - green *= 257; - blue *= 257; - alpha *= 257; - encoding = P_LINEAR; - } - - else if (encoding == P_sRGB && - (convert_to_Y != 0 || output_encoding == P_LINEAR)) - { - /* The values are 8-bit sRGB values, but must be converted to 16-bit - * linear. - */ - red = png_sRGB_table[red]; - green = png_sRGB_table[green]; - blue = png_sRGB_table[blue]; - alpha *= 257; - encoding = P_LINEAR; - } - - /* This is set if the color isn't gray but the output is. */ - if (encoding == P_LINEAR) - { - if (convert_to_Y != 0) - { - /* NOTE: these values are copied from png_do_rgb_to_gray */ - png_uint_32 y = (png_uint_32)6968 * red + (png_uint_32)23434 * green + - (png_uint_32)2366 * blue; - - if (output_encoding == P_LINEAR) - y = (y + 16384) >> 15; - - else - { - /* y is scaled by 32768, we need it scaled by 255: */ - y = (y + 128) >> 8; - y *= 255; - y = PNG_sRGB_FROM_LINEAR((y + 64) >> 7); - alpha = PNG_DIV257(alpha); - encoding = P_sRGB; - } - - blue = red = green = y; - } - - else if (output_encoding == P_sRGB) - { - red = PNG_sRGB_FROM_LINEAR(red * 255); - green = PNG_sRGB_FROM_LINEAR(green * 255); - blue = PNG_sRGB_FROM_LINEAR(blue * 255); - alpha = PNG_DIV257(alpha); - encoding = P_sRGB; - } - } - - if (encoding != output_encoding) - png_error(image->opaque->png_ptr, "bad encoding (internal error)"); - - /* Store the value. */ - { -# ifdef PNG_FORMAT_AFIRST_SUPPORTED - int afirst = (image->format & PNG_FORMAT_FLAG_AFIRST) != 0 && - (image->format & PNG_FORMAT_FLAG_ALPHA) != 0; -# else -# define afirst 0 -# endif -# ifdef PNG_FORMAT_BGR_SUPPORTED - int bgr = (image->format & PNG_FORMAT_FLAG_BGR) != 0 ? 2 : 0; -# else -# define bgr 0 -# endif - - if (output_encoding == P_LINEAR) - { - png_uint_16p entry = png_voidcast(png_uint_16p, display->colormap); - - entry += ip * PNG_IMAGE_SAMPLE_CHANNELS(image->format); - - /* The linear 16-bit values must be pre-multiplied by the alpha channel - * value, if less than 65535 (this is, effectively, composite on black - * if the alpha channel is removed.) - */ - switch (PNG_IMAGE_SAMPLE_CHANNELS(image->format)) - { - case 4: - entry[afirst ? 0 : 3] = (png_uint_16)alpha; - /* FALLTHROUGH */ - - case 3: - if (alpha < 65535) - { - if (alpha > 0) - { - blue = (blue * alpha + 32767U)/65535U; - green = (green * alpha + 32767U)/65535U; - red = (red * alpha + 32767U)/65535U; - } - - else - red = green = blue = 0; - } - entry[afirst + (2 ^ bgr)] = (png_uint_16)blue; - entry[afirst + 1] = (png_uint_16)green; - entry[afirst + bgr] = (png_uint_16)red; - break; - - case 2: - entry[1 ^ afirst] = (png_uint_16)alpha; - /* FALLTHROUGH */ - - case 1: - if (alpha < 65535) - { - if (alpha > 0) - green = (green * alpha + 32767U)/65535U; - - else - green = 0; - } - entry[afirst] = (png_uint_16)green; - break; - - default: - break; - } - } - - else /* output encoding is P_sRGB */ - { - png_bytep entry = png_voidcast(png_bytep, display->colormap); - - entry += ip * PNG_IMAGE_SAMPLE_CHANNELS(image->format); - - switch (PNG_IMAGE_SAMPLE_CHANNELS(image->format)) - { - case 4: - entry[afirst ? 0 : 3] = (png_byte)alpha; - /* FALLTHROUGH */ - case 3: - entry[afirst + (2 ^ bgr)] = (png_byte)blue; - entry[afirst + 1] = (png_byte)green; - entry[afirst + bgr] = (png_byte)red; - break; - - case 2: - entry[1 ^ afirst] = (png_byte)alpha; - /* FALLTHROUGH */ - case 1: - entry[afirst] = (png_byte)green; - break; - - default: - break; - } - } - -# ifdef afirst -# undef afirst -# endif -# ifdef bgr -# undef bgr -# endif - } -} - -static int -make_gray_file_colormap(png_image_read_control *display) -{ - unsigned int i; - - for (i=0; i<256; ++i) - png_create_colormap_entry(display, i, i, i, i, 255, P_FILE); - - return (int)i; -} - -static int -make_gray_colormap(png_image_read_control *display) -{ - unsigned int i; - - for (i=0; i<256; ++i) - png_create_colormap_entry(display, i, i, i, i, 255, P_sRGB); - - return (int)i; -} -#define PNG_GRAY_COLORMAP_ENTRIES 256 - -static int -make_ga_colormap(png_image_read_control *display) -{ - unsigned int i, a; - - /* Alpha is retained, the output will be a color-map with entries - * selected by six levels of alpha. One transparent entry, 6 gray - * levels for all the intermediate alpha values, leaving 230 entries - * for the opaque grays. The color-map entries are the six values - * [0..5]*51, the GA processing uses PNG_DIV51(value) to find the - * relevant entry. - * - * if (alpha > 229) // opaque - * { - * // The 231 entries are selected to make the math below work: - * base = 0; - * entry = (231 * gray + 128) >> 8; - * } - * else if (alpha < 26) // transparent - * { - * base = 231; - * entry = 0; - * } - * else // partially opaque - * { - * base = 226 + 6 * PNG_DIV51(alpha); - * entry = PNG_DIV51(gray); - * } - */ - i = 0; - while (i < 231) - { - unsigned int gray = (i * 256 + 115) / 231; - png_create_colormap_entry(display, i++, gray, gray, gray, 255, P_sRGB); - } - - /* 255 is used here for the component values for consistency with the code - * that undoes premultiplication in pngwrite.c. - */ - png_create_colormap_entry(display, i++, 255, 255, 255, 0, P_sRGB); - - for (a=1; a<5; ++a) - { - unsigned int g; - - for (g=0; g<6; ++g) - png_create_colormap_entry(display, i++, g*51, g*51, g*51, a*51, - P_sRGB); - } - - return (int)i; -} - -#define PNG_GA_COLORMAP_ENTRIES 256 - -static int -make_rgb_colormap(png_image_read_control *display) -{ - unsigned int i, r; - - /* Build a 6x6x6 opaque RGB cube */ - for (i=r=0; r<6; ++r) - { - unsigned int g; - - for (g=0; g<6; ++g) - { - unsigned int b; - - for (b=0; b<6; ++b) - png_create_colormap_entry(display, i++, r*51, g*51, b*51, 255, - P_sRGB); - } - } - - return (int)i; -} - -#define PNG_RGB_COLORMAP_ENTRIES 216 - -/* Return a palette index to the above palette given three 8-bit sRGB values. */ -#define PNG_RGB_INDEX(r,g,b) \ - ((png_byte)(6 * (6 * PNG_DIV51(r) + PNG_DIV51(g)) + PNG_DIV51(b))) - -static int -png_image_read_colormap(png_voidp argument) -{ - png_image_read_control *display = - png_voidcast(png_image_read_control*, argument); - png_imagep image = display->image; - - png_structrp png_ptr = image->opaque->png_ptr; - png_uint_32 output_format = image->format; - int output_encoding = (output_format & PNG_FORMAT_FLAG_LINEAR) != 0 ? - P_LINEAR : P_sRGB; - - unsigned int cmap_entries; - unsigned int output_processing; /* Output processing option */ - unsigned int data_encoding = P_NOTSET; /* Encoding libpng must produce */ - - /* Background information; the background color and the index of this color - * in the color-map if it exists (else 256). - */ - unsigned int background_index = 256; - png_uint_32 back_r, back_g, back_b; - - /* Flags to accumulate things that need to be done to the input. */ - int expand_tRNS = 0; - - /* Exclude the NYI feature of compositing onto a color-mapped buffer; it is - * very difficult to do, the results look awful, and it is difficult to see - * what possible use it is because the application can't control the - * color-map. - */ - if (((png_ptr->color_type & PNG_COLOR_MASK_ALPHA) != 0 || - png_ptr->num_trans > 0) /* alpha in input */ && - ((output_format & PNG_FORMAT_FLAG_ALPHA) == 0) /* no alpha in output */) - { - if (output_encoding == P_LINEAR) /* compose on black */ - back_b = back_g = back_r = 0; - - else if (display->background == NULL /* no way to remove it */) - png_error(png_ptr, - "background color must be supplied to remove alpha/transparency"); - - /* Get a copy of the background color (this avoids repeating the checks - * below.) The encoding is 8-bit sRGB or 16-bit linear, depending on the - * output format. - */ - else - { - back_g = display->background->green; - if ((output_format & PNG_FORMAT_FLAG_COLOR) != 0) - { - back_r = display->background->red; - back_b = display->background->blue; - } - else - back_b = back_r = back_g; - } - } - - else if (output_encoding == P_LINEAR) - back_b = back_r = back_g = 65535; - - else - back_b = back_r = back_g = 255; - - /* Default the input file gamma if required - this is necessary because - * libpng assumes that if no gamma information is present the data is in the - * output format, but the simplified API deduces the gamma from the input - * format. - */ - if ((png_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_GAMMA) == 0) - { - /* Do this directly, not using the png_colorspace functions, to ensure - * that it happens even if the colorspace is invalid (though probably if - * it is the setting will be ignored) Note that the same thing can be - * achieved at the application interface with png_set_gAMA. - */ - if (png_ptr->bit_depth == 16 && - (image->flags & PNG_IMAGE_FLAG_16BIT_sRGB) == 0) - png_ptr->colorspace.gamma = PNG_GAMMA_LINEAR; - - else - png_ptr->colorspace.gamma = PNG_GAMMA_sRGB_INVERSE; - - png_ptr->colorspace.flags |= PNG_COLORSPACE_HAVE_GAMMA; - } - - /* Decide what to do based on the PNG color type of the input data. The - * utility function png_create_colormap_entry deals with most aspects of the - * output transformations; this code works out how to produce bytes of - * color-map entries from the original format. - */ - switch (png_ptr->color_type) - { - case PNG_COLOR_TYPE_GRAY: - if (png_ptr->bit_depth <= 8) - { - /* There at most 256 colors in the output, regardless of - * transparency. - */ - unsigned int step, i, val, trans = 256/*ignore*/, back_alpha = 0; - - cmap_entries = 1U << png_ptr->bit_depth; - if (cmap_entries > image->colormap_entries) - png_error(png_ptr, "gray[8] color-map: too few entries"); - - step = 255 / (cmap_entries - 1); - output_processing = PNG_CMAP_NONE; - - /* If there is a tRNS chunk then this either selects a transparent - * value or, if the output has no alpha, the background color. - */ - if (png_ptr->num_trans > 0) - { - trans = png_ptr->trans_color.gray; - - if ((output_format & PNG_FORMAT_FLAG_ALPHA) == 0) - back_alpha = output_encoding == P_LINEAR ? 65535 : 255; - } - - /* png_create_colormap_entry just takes an RGBA and writes the - * corresponding color-map entry using the format from 'image', - * including the required conversion to sRGB or linear as - * appropriate. The input values are always either sRGB (if the - * gamma correction flag is 0) or 0..255 scaled file encoded values - * (if the function must gamma correct them). - */ - for (i=val=0; ibit_depth < 8) - png_set_packing(png_ptr); - } - - else /* bit depth is 16 */ - { - /* The 16-bit input values can be converted directly to 8-bit gamma - * encoded values; however, if a tRNS chunk is present 257 color-map - * entries are required. This means that the extra entry requires - * special processing; add an alpha channel, sacrifice gray level - * 254 and convert transparent (alpha==0) entries to that. - * - * Use libpng to chop the data to 8 bits. Convert it to sRGB at the - * same time to minimize quality loss. If a tRNS chunk is present - * this means libpng must handle it too; otherwise it is impossible - * to do the exact match on the 16-bit value. - * - * If the output has no alpha channel *and* the background color is - * gray then it is possible to let libpng handle the substitution by - * ensuring that the corresponding gray level matches the background - * color exactly. - */ - data_encoding = P_sRGB; - - if (PNG_GRAY_COLORMAP_ENTRIES > image->colormap_entries) - png_error(png_ptr, "gray[16] color-map: too few entries"); - - cmap_entries = (unsigned int)make_gray_colormap(display); - - if (png_ptr->num_trans > 0) - { - unsigned int back_alpha; - - if ((output_format & PNG_FORMAT_FLAG_ALPHA) != 0) - back_alpha = 0; - - else - { - if (back_r == back_g && back_g == back_b) - { - /* Background is gray; no special processing will be - * required. - */ - png_color_16 c; - png_uint_32 gray = back_g; - - if (output_encoding == P_LINEAR) - { - gray = PNG_sRGB_FROM_LINEAR(gray * 255); - - /* And make sure the corresponding palette entry - * matches. - */ - png_create_colormap_entry(display, gray, back_g, back_g, - back_g, 65535, P_LINEAR); - } - - /* The background passed to libpng, however, must be the - * sRGB value. - */ - c.index = 0; /*unused*/ - c.gray = c.red = c.green = c.blue = (png_uint_16)gray; - - /* NOTE: does this work without expanding tRNS to alpha? - * It should be the color->gray case below apparently - * doesn't. - */ - png_set_background_fixed(png_ptr, &c, - PNG_BACKGROUND_GAMMA_SCREEN, 0/*need_expand*/, - 0/*gamma: not used*/); - - output_processing = PNG_CMAP_NONE; - break; - } -#ifdef __COVERITY__ - /* Coverity claims that output_encoding cannot be 2 (P_LINEAR) - * here. - */ - back_alpha = 255; -#else - back_alpha = output_encoding == P_LINEAR ? 65535 : 255; -#endif - } - - /* output_processing means that the libpng-processed row will be - * 8-bit GA and it has to be processing to single byte color-map - * values. Entry 254 is replaced by either a completely - * transparent entry or by the background color at full - * precision (and the background color is not a simple gray - * level in this case.) - */ - expand_tRNS = 1; - output_processing = PNG_CMAP_TRANS; - background_index = 254; - - /* And set (overwrite) color-map entry 254 to the actual - * background color at full precision. - */ - png_create_colormap_entry(display, 254, back_r, back_g, back_b, - back_alpha, output_encoding); - } - - else - output_processing = PNG_CMAP_NONE; - } - break; - - case PNG_COLOR_TYPE_GRAY_ALPHA: - /* 8-bit or 16-bit PNG with two channels - gray and alpha. A minimum - * of 65536 combinations. If, however, the alpha channel is to be - * removed there are only 256 possibilities if the background is gray. - * (Otherwise there is a subset of the 65536 possibilities defined by - * the triangle between black, white and the background color.) - * - * Reduce 16-bit files to 8-bit and sRGB encode the result. No need to - * worry about tRNS matching - tRNS is ignored if there is an alpha - * channel. - */ - data_encoding = P_sRGB; - - if ((output_format & PNG_FORMAT_FLAG_ALPHA) != 0) - { - if (PNG_GA_COLORMAP_ENTRIES > image->colormap_entries) - png_error(png_ptr, "gray+alpha color-map: too few entries"); - - cmap_entries = (unsigned int)make_ga_colormap(display); - - background_index = PNG_CMAP_GA_BACKGROUND; - output_processing = PNG_CMAP_GA; - } - - else /* alpha is removed */ - { - /* Alpha must be removed as the PNG data is processed when the - * background is a color because the G and A channels are - * independent and the vector addition (non-parallel vectors) is a - * 2-D problem. - * - * This can be reduced to the same algorithm as above by making a - * colormap containing gray levels (for the opaque grays), a - * background entry (for a transparent pixel) and a set of four six - * level color values, one set for each intermediate alpha value. - * See the comments in make_ga_colormap for how this works in the - * per-pixel processing. - * - * If the background is gray, however, we only need a 256 entry gray - * level color map. It is sufficient to make the entry generated - * for the background color be exactly the color specified. - */ - if ((output_format & PNG_FORMAT_FLAG_COLOR) == 0 || - (back_r == back_g && back_g == back_b)) - { - /* Background is gray; no special processing will be required. */ - png_color_16 c; - png_uint_32 gray = back_g; - - if (PNG_GRAY_COLORMAP_ENTRIES > image->colormap_entries) - png_error(png_ptr, "gray-alpha color-map: too few entries"); - - cmap_entries = (unsigned int)make_gray_colormap(display); - - if (output_encoding == P_LINEAR) - { - gray = PNG_sRGB_FROM_LINEAR(gray * 255); - - /* And make sure the corresponding palette entry matches. */ - png_create_colormap_entry(display, gray, back_g, back_g, - back_g, 65535, P_LINEAR); - } - - /* The background passed to libpng, however, must be the sRGB - * value. - */ - c.index = 0; /*unused*/ - c.gray = c.red = c.green = c.blue = (png_uint_16)gray; - - png_set_background_fixed(png_ptr, &c, - PNG_BACKGROUND_GAMMA_SCREEN, 0/*need_expand*/, - 0/*gamma: not used*/); - - output_processing = PNG_CMAP_NONE; - } - - else - { - png_uint_32 i, a; - - /* This is the same as png_make_ga_colormap, above, except that - * the entries are all opaque. - */ - if (PNG_GA_COLORMAP_ENTRIES > image->colormap_entries) - png_error(png_ptr, "ga-alpha color-map: too few entries"); - - i = 0; - while (i < 231) - { - png_uint_32 gray = (i * 256 + 115) / 231; - png_create_colormap_entry(display, i++, gray, gray, gray, - 255, P_sRGB); - } - - /* NOTE: this preserves the full precision of the application - * background color. - */ - background_index = i; - png_create_colormap_entry(display, i++, back_r, back_g, back_b, -#ifdef __COVERITY__ - /* Coverity claims that output_encoding - * cannot be 2 (P_LINEAR) here. - */ 255U, -#else - output_encoding == P_LINEAR ? 65535U : 255U, -#endif - output_encoding); - - /* For non-opaque input composite on the sRGB background - this - * requires inverting the encoding for each component. The input - * is still converted to the sRGB encoding because this is a - * reasonable approximate to the logarithmic curve of human - * visual sensitivity, at least over the narrow range which PNG - * represents. Consequently 'G' is always sRGB encoded, while - * 'A' is linear. We need the linear background colors. - */ - if (output_encoding == P_sRGB) /* else already linear */ - { - /* This may produce a value not exactly matching the - * background, but that's ok because these numbers are only - * used when alpha != 0 - */ - back_r = png_sRGB_table[back_r]; - back_g = png_sRGB_table[back_g]; - back_b = png_sRGB_table[back_b]; - } - - for (a=1; a<5; ++a) - { - unsigned int g; - - /* PNG_sRGB_FROM_LINEAR expects a 16-bit linear value scaled - * by an 8-bit alpha value (0..255). - */ - png_uint_32 alpha = 51 * a; - png_uint_32 back_rx = (255-alpha) * back_r; - png_uint_32 back_gx = (255-alpha) * back_g; - png_uint_32 back_bx = (255-alpha) * back_b; - - for (g=0; g<6; ++g) - { - png_uint_32 gray = png_sRGB_table[g*51] * alpha; - - png_create_colormap_entry(display, i++, - PNG_sRGB_FROM_LINEAR(gray + back_rx), - PNG_sRGB_FROM_LINEAR(gray + back_gx), - PNG_sRGB_FROM_LINEAR(gray + back_bx), 255, P_sRGB); - } - } - - cmap_entries = i; - output_processing = PNG_CMAP_GA; - } - } - break; - - case PNG_COLOR_TYPE_RGB: - case PNG_COLOR_TYPE_RGB_ALPHA: - /* Exclude the case where the output is gray; we can always handle this - * with the cases above. - */ - if ((output_format & PNG_FORMAT_FLAG_COLOR) == 0) - { - /* The color-map will be grayscale, so we may as well convert the - * input RGB values to a simple grayscale and use the grayscale - * code above. - * - * NOTE: calling this apparently damages the recognition of the - * transparent color in background color handling; call - * png_set_tRNS_to_alpha before png_set_background_fixed. - */ - png_set_rgb_to_gray_fixed(png_ptr, PNG_ERROR_ACTION_NONE, -1, - -1); - data_encoding = P_sRGB; - - /* The output will now be one or two 8-bit gray or gray+alpha - * channels. The more complex case arises when the input has alpha. - */ - if ((png_ptr->color_type == PNG_COLOR_TYPE_RGB_ALPHA || - png_ptr->num_trans > 0) && - (output_format & PNG_FORMAT_FLAG_ALPHA) != 0) - { - /* Both input and output have an alpha channel, so no background - * processing is required; just map the GA bytes to the right - * color-map entry. - */ - expand_tRNS = 1; - - if (PNG_GA_COLORMAP_ENTRIES > image->colormap_entries) - png_error(png_ptr, "rgb[ga] color-map: too few entries"); - - cmap_entries = (unsigned int)make_ga_colormap(display); - background_index = PNG_CMAP_GA_BACKGROUND; - output_processing = PNG_CMAP_GA; - } - - else - { - /* Either the input or the output has no alpha channel, so there - * will be no non-opaque pixels in the color-map; it will just be - * grayscale. - */ - if (PNG_GRAY_COLORMAP_ENTRIES > image->colormap_entries) - png_error(png_ptr, "rgb[gray] color-map: too few entries"); - - /* Ideally this code would use libpng to do the gamma correction, - * but if an input alpha channel is to be removed we will hit the - * libpng bug in gamma+compose+rgb-to-gray (the double gamma - * correction bug). Fix this by dropping the gamma correction in - * this case and doing it in the palette; this will result in - * duplicate palette entries, but that's better than the - * alternative of double gamma correction. - */ - if ((png_ptr->color_type == PNG_COLOR_TYPE_RGB_ALPHA || - png_ptr->num_trans > 0) && - png_gamma_not_sRGB(png_ptr->colorspace.gamma) != 0) - { - cmap_entries = (unsigned int)make_gray_file_colormap(display); - data_encoding = P_FILE; - } - - else - cmap_entries = (unsigned int)make_gray_colormap(display); - - /* But if the input has alpha or transparency it must be removed - */ - if (png_ptr->color_type == PNG_COLOR_TYPE_RGB_ALPHA || - png_ptr->num_trans > 0) - { - png_color_16 c; - png_uint_32 gray = back_g; - - /* We need to ensure that the application background exists in - * the colormap and that completely transparent pixels map to - * it. Achieve this simply by ensuring that the entry - * selected for the background really is the background color. - */ - if (data_encoding == P_FILE) /* from the fixup above */ - { - /* The app supplied a gray which is in output_encoding, we - * need to convert it to a value of the input (P_FILE) - * encoding then set this palette entry to the required - * output encoding. - */ - if (output_encoding == P_sRGB) - gray = png_sRGB_table[gray]; /* now P_LINEAR */ - - gray = PNG_DIV257(png_gamma_16bit_correct(gray, - png_ptr->colorspace.gamma)); /* now P_FILE */ - - /* And make sure the corresponding palette entry contains - * exactly the required sRGB value. - */ - png_create_colormap_entry(display, gray, back_g, back_g, - back_g, 0/*unused*/, output_encoding); - } - - else if (output_encoding == P_LINEAR) - { - gray = PNG_sRGB_FROM_LINEAR(gray * 255); - - /* And make sure the corresponding palette entry matches. - */ - png_create_colormap_entry(display, gray, back_g, back_g, - back_g, 0/*unused*/, P_LINEAR); - } - - /* The background passed to libpng, however, must be the - * output (normally sRGB) value. - */ - c.index = 0; /*unused*/ - c.gray = c.red = c.green = c.blue = (png_uint_16)gray; - - /* NOTE: the following is apparently a bug in libpng. Without - * it the transparent color recognition in - * png_set_background_fixed seems to go wrong. - */ - expand_tRNS = 1; - png_set_background_fixed(png_ptr, &c, - PNG_BACKGROUND_GAMMA_SCREEN, 0/*need_expand*/, - 0/*gamma: not used*/); - } - - output_processing = PNG_CMAP_NONE; - } - } - - else /* output is color */ - { - /* We could use png_quantize here so long as there is no transparent - * color or alpha; png_quantize ignores alpha. Easier overall just - * to do it once and using PNG_DIV51 on the 6x6x6 reduced RGB cube. - * Consequently we always want libpng to produce sRGB data. - */ - data_encoding = P_sRGB; - - /* Is there any transparency or alpha? */ - if (png_ptr->color_type == PNG_COLOR_TYPE_RGB_ALPHA || - png_ptr->num_trans > 0) - { - /* Is there alpha in the output too? If so all four channels are - * processed into a special RGB cube with alpha support. - */ - if ((output_format & PNG_FORMAT_FLAG_ALPHA) != 0) - { - png_uint_32 r; - - if (PNG_RGB_COLORMAP_ENTRIES+1+27 > image->colormap_entries) - png_error(png_ptr, "rgb+alpha color-map: too few entries"); - - cmap_entries = (unsigned int)make_rgb_colormap(display); - - /* Add a transparent entry. */ - png_create_colormap_entry(display, cmap_entries, 255, 255, - 255, 0, P_sRGB); - - /* This is stored as the background index for the processing - * algorithm. - */ - background_index = cmap_entries++; - - /* Add 27 r,g,b entries each with alpha 0.5. */ - for (r=0; r<256; r = (r << 1) | 0x7f) - { - png_uint_32 g; - - for (g=0; g<256; g = (g << 1) | 0x7f) - { - png_uint_32 b; - - /* This generates components with the values 0, 127 and - * 255 - */ - for (b=0; b<256; b = (b << 1) | 0x7f) - png_create_colormap_entry(display, cmap_entries++, - r, g, b, 128, P_sRGB); - } - } - - expand_tRNS = 1; - output_processing = PNG_CMAP_RGB_ALPHA; - } - - else - { - /* Alpha/transparency must be removed. The background must - * exist in the color map (achieved by setting adding it after - * the 666 color-map). If the standard processing code will - * pick up this entry automatically that's all that is - * required; libpng can be called to do the background - * processing. - */ - unsigned int sample_size = - PNG_IMAGE_SAMPLE_SIZE(output_format); - png_uint_32 r, g, b; /* sRGB background */ - - if (PNG_RGB_COLORMAP_ENTRIES+1+27 > image->colormap_entries) - png_error(png_ptr, "rgb-alpha color-map: too few entries"); - - cmap_entries = (unsigned int)make_rgb_colormap(display); - - png_create_colormap_entry(display, cmap_entries, back_r, - back_g, back_b, 0/*unused*/, output_encoding); - - if (output_encoding == P_LINEAR) - { - r = PNG_sRGB_FROM_LINEAR(back_r * 255); - g = PNG_sRGB_FROM_LINEAR(back_g * 255); - b = PNG_sRGB_FROM_LINEAR(back_b * 255); - } - - else - { - r = back_r; - g = back_g; - b = back_g; - } - - /* Compare the newly-created color-map entry with the one the - * PNG_CMAP_RGB algorithm will use. If the two entries don't - * match, add the new one and set this as the background - * index. - */ - if (memcmp((png_const_bytep)display->colormap + - sample_size * cmap_entries, - (png_const_bytep)display->colormap + - sample_size * PNG_RGB_INDEX(r,g,b), - sample_size) != 0) - { - /* The background color must be added. */ - background_index = cmap_entries++; - - /* Add 27 r,g,b entries each with created by composing with - * the background at alpha 0.5. - */ - for (r=0; r<256; r = (r << 1) | 0x7f) - { - for (g=0; g<256; g = (g << 1) | 0x7f) - { - /* This generates components with the values 0, 127 - * and 255 - */ - for (b=0; b<256; b = (b << 1) | 0x7f) - png_create_colormap_entry(display, cmap_entries++, - png_colormap_compose(display, r, P_sRGB, 128, - back_r, output_encoding), - png_colormap_compose(display, g, P_sRGB, 128, - back_g, output_encoding), - png_colormap_compose(display, b, P_sRGB, 128, - back_b, output_encoding), - 0/*unused*/, output_encoding); - } - } - - expand_tRNS = 1; - output_processing = PNG_CMAP_RGB_ALPHA; - } - - else /* background color is in the standard color-map */ - { - png_color_16 c; - - c.index = 0; /*unused*/ - c.red = (png_uint_16)back_r; - c.gray = c.green = (png_uint_16)back_g; - c.blue = (png_uint_16)back_b; - - png_set_background_fixed(png_ptr, &c, - PNG_BACKGROUND_GAMMA_SCREEN, 0/*need_expand*/, - 0/*gamma: not used*/); - - output_processing = PNG_CMAP_RGB; - } - } - } - - else /* no alpha or transparency in the input */ - { - /* Alpha in the output is irrelevant, simply map the opaque input - * pixels to the 6x6x6 color-map. - */ - if (PNG_RGB_COLORMAP_ENTRIES > image->colormap_entries) - png_error(png_ptr, "rgb color-map: too few entries"); - - cmap_entries = (unsigned int)make_rgb_colormap(display); - output_processing = PNG_CMAP_RGB; - } - } - break; - - case PNG_COLOR_TYPE_PALETTE: - /* It's already got a color-map. It may be necessary to eliminate the - * tRNS entries though. - */ - { - unsigned int num_trans = png_ptr->num_trans; - png_const_bytep trans = num_trans > 0 ? png_ptr->trans_alpha : NULL; - png_const_colorp colormap = png_ptr->palette; - int do_background = trans != NULL && - (output_format & PNG_FORMAT_FLAG_ALPHA) == 0; - unsigned int i; - - /* Just in case: */ - if (trans == NULL) - num_trans = 0; - - output_processing = PNG_CMAP_NONE; - data_encoding = P_FILE; /* Don't change from color-map indices */ - cmap_entries = (unsigned int)png_ptr->num_palette; - if (cmap_entries > 256) - cmap_entries = 256; - - if (cmap_entries > (unsigned int)image->colormap_entries) - png_error(png_ptr, "palette color-map: too few entries"); - - for (i=0; i < cmap_entries; ++i) - { - if (do_background != 0 && i < num_trans && trans[i] < 255) - { - if (trans[i] == 0) - png_create_colormap_entry(display, i, back_r, back_g, - back_b, 0, output_encoding); - - else - { - /* Must compose the PNG file color in the color-map entry - * on the sRGB color in 'back'. - */ - png_create_colormap_entry(display, i, - png_colormap_compose(display, colormap[i].red, - P_FILE, trans[i], back_r, output_encoding), - png_colormap_compose(display, colormap[i].green, - P_FILE, trans[i], back_g, output_encoding), - png_colormap_compose(display, colormap[i].blue, - P_FILE, trans[i], back_b, output_encoding), - output_encoding == P_LINEAR ? trans[i] * 257U : - trans[i], - output_encoding); - } - } - - else - png_create_colormap_entry(display, i, colormap[i].red, - colormap[i].green, colormap[i].blue, - i < num_trans ? trans[i] : 255U, P_FILE/*8-bit*/); - } - - /* The PNG data may have indices packed in fewer than 8 bits, it - * must be expanded if so. - */ - if (png_ptr->bit_depth < 8) - png_set_packing(png_ptr); - } - break; - - default: - png_error(png_ptr, "invalid PNG color type"); - /*NOT REACHED*/ - } - - /* Now deal with the output processing */ - if (expand_tRNS != 0 && png_ptr->num_trans > 0 && - (png_ptr->color_type & PNG_COLOR_MASK_ALPHA) == 0) - png_set_tRNS_to_alpha(png_ptr); - - switch (data_encoding) - { - case P_sRGB: - /* Change to 8-bit sRGB */ - png_set_alpha_mode_fixed(png_ptr, PNG_ALPHA_PNG, PNG_GAMMA_sRGB); - /* FALLTHROUGH */ - - case P_FILE: - if (png_ptr->bit_depth > 8) - png_set_scale_16(png_ptr); - break; - -#ifdef __GNUC__ - default: - png_error(png_ptr, "bad data option (internal error)"); -#endif - } - - if (cmap_entries > 256 || cmap_entries > image->colormap_entries) - png_error(png_ptr, "color map overflow (BAD internal error)"); - - image->colormap_entries = cmap_entries; - - /* Double check using the recorded background index */ - switch (output_processing) - { - case PNG_CMAP_NONE: - if (background_index != PNG_CMAP_NONE_BACKGROUND) - goto bad_background; - break; - - case PNG_CMAP_GA: - if (background_index != PNG_CMAP_GA_BACKGROUND) - goto bad_background; - break; - - case PNG_CMAP_TRANS: - if (background_index >= cmap_entries || - background_index != PNG_CMAP_TRANS_BACKGROUND) - goto bad_background; - break; - - case PNG_CMAP_RGB: - if (background_index != PNG_CMAP_RGB_BACKGROUND) - goto bad_background; - break; - - case PNG_CMAP_RGB_ALPHA: - if (background_index != PNG_CMAP_RGB_ALPHA_BACKGROUND) - goto bad_background; - break; - - default: - png_error(png_ptr, "bad processing option (internal error)"); - - bad_background: - png_error(png_ptr, "bad background index (internal error)"); - } - - display->colormap_processing = (int)output_processing; - - return 1/*ok*/; -} - -/* The final part of the color-map read called from png_image_finish_read. */ -static int -png_image_read_and_map(png_voidp argument) -{ - png_image_read_control *display = png_voidcast(png_image_read_control*, - argument); - png_imagep image = display->image; - png_structrp png_ptr = image->opaque->png_ptr; - int passes; - - /* Called when the libpng data must be transformed into the color-mapped - * form. There is a local row buffer in display->local and this routine must - * do the interlace handling. - */ - switch (png_ptr->interlaced) - { - case PNG_INTERLACE_NONE: - passes = 1; - break; - - case PNG_INTERLACE_ADAM7: - passes = PNG_INTERLACE_ADAM7_PASSES; - break; - - default: - png_error(png_ptr, "unknown interlace type"); - } - - { - png_uint_32 height = image->height; - png_uint_32 width = image->width; - int proc = display->colormap_processing; - png_bytep first_row = png_voidcast(png_bytep, display->first_row); - ptrdiff_t step_row = display->row_bytes; - int pass; - - for (pass = 0; pass < passes; ++pass) - { - unsigned int startx, stepx, stepy; - png_uint_32 y; - - if (png_ptr->interlaced == PNG_INTERLACE_ADAM7) - { - /* The row may be empty for a short image: */ - if (PNG_PASS_COLS(width, pass) == 0) - continue; - - startx = PNG_PASS_START_COL(pass); - stepx = PNG_PASS_COL_OFFSET(pass); - y = PNG_PASS_START_ROW(pass); - stepy = PNG_PASS_ROW_OFFSET(pass); - } - - else - { - y = 0; - startx = 0; - stepx = stepy = 1; - } - - for (; ylocal_row); - png_bytep outrow = first_row + y * step_row; - png_const_bytep end_row = outrow + width; - - /* Read read the libpng data into the temporary buffer. */ - png_read_row(png_ptr, inrow, NULL); - - /* Now process the row according to the processing option, note - * that the caller verifies that the format of the libpng output - * data is as required. - */ - outrow += startx; - switch (proc) - { - case PNG_CMAP_GA: - for (; outrow < end_row; outrow += stepx) - { - /* The data is always in the PNG order */ - unsigned int gray = *inrow++; - unsigned int alpha = *inrow++; - unsigned int entry; - - /* NOTE: this code is copied as a comment in - * make_ga_colormap above. Please update the - * comment if you change this code! - */ - if (alpha > 229) /* opaque */ - { - entry = (231 * gray + 128) >> 8; - } - else if (alpha < 26) /* transparent */ - { - entry = 231; - } - else /* partially opaque */ - { - entry = 226 + 6 * PNG_DIV51(alpha) + PNG_DIV51(gray); - } - - *outrow = (png_byte)entry; - } - break; - - case PNG_CMAP_TRANS: - for (; outrow < end_row; outrow += stepx) - { - png_byte gray = *inrow++; - png_byte alpha = *inrow++; - - if (alpha == 0) - *outrow = PNG_CMAP_TRANS_BACKGROUND; - - else if (gray != PNG_CMAP_TRANS_BACKGROUND) - *outrow = gray; - - else - *outrow = (png_byte)(PNG_CMAP_TRANS_BACKGROUND+1); - } - break; - - case PNG_CMAP_RGB: - for (; outrow < end_row; outrow += stepx) - { - *outrow = PNG_RGB_INDEX(inrow[0], inrow[1], inrow[2]); - inrow += 3; - } - break; - - case PNG_CMAP_RGB_ALPHA: - for (; outrow < end_row; outrow += stepx) - { - unsigned int alpha = inrow[3]; - - /* Because the alpha entries only hold alpha==0.5 values - * split the processing at alpha==0.25 (64) and 0.75 - * (196). - */ - - if (alpha >= 196) - *outrow = PNG_RGB_INDEX(inrow[0], inrow[1], - inrow[2]); - - else if (alpha < 64) - *outrow = PNG_CMAP_RGB_ALPHA_BACKGROUND; - - else - { - /* Likewise there are three entries for each of r, g - * and b. We could select the entry by popcount on - * the top two bits on those architectures that - * support it, this is what the code below does, - * crudely. - */ - unsigned int back_i = PNG_CMAP_RGB_ALPHA_BACKGROUND+1; - - /* Here are how the values map: - * - * 0x00 .. 0x3f -> 0 - * 0x40 .. 0xbf -> 1 - * 0xc0 .. 0xff -> 2 - * - * So, as above with the explicit alpha checks, the - * breakpoints are at 64 and 196. - */ - if (inrow[0] & 0x80) back_i += 9; /* red */ - if (inrow[0] & 0x40) back_i += 9; - if (inrow[0] & 0x80) back_i += 3; /* green */ - if (inrow[0] & 0x40) back_i += 3; - if (inrow[0] & 0x80) back_i += 1; /* blue */ - if (inrow[0] & 0x40) back_i += 1; - - *outrow = (png_byte)back_i; - } - - inrow += 4; - } - break; - - default: - break; - } - } - } - } - - return 1; -} - -static int -png_image_read_colormapped(png_voidp argument) -{ - png_image_read_control *display = png_voidcast(png_image_read_control*, - argument); - png_imagep image = display->image; - png_controlp control = image->opaque; - png_structrp png_ptr = control->png_ptr; - png_inforp info_ptr = control->info_ptr; - - int passes = 0; /* As a flag */ - - PNG_SKIP_CHUNKS(png_ptr); - - /* Update the 'info' structure and make sure the result is as required; first - * make sure to turn on the interlace handling if it will be required - * (because it can't be turned on *after* the call to png_read_update_info!) - */ - if (display->colormap_processing == PNG_CMAP_NONE) - passes = png_set_interlace_handling(png_ptr); - - png_read_update_info(png_ptr, info_ptr); - - /* The expected output can be deduced from the colormap_processing option. */ - switch (display->colormap_processing) - { - case PNG_CMAP_NONE: - /* Output must be one channel and one byte per pixel, the output - * encoding can be anything. - */ - if ((info_ptr->color_type == PNG_COLOR_TYPE_PALETTE || - info_ptr->color_type == PNG_COLOR_TYPE_GRAY) && - info_ptr->bit_depth == 8) - break; - - goto bad_output; - - case PNG_CMAP_TRANS: - case PNG_CMAP_GA: - /* Output must be two channels and the 'G' one must be sRGB, the latter - * can be checked with an exact number because it should have been set - * to this number above! - */ - if (info_ptr->color_type == PNG_COLOR_TYPE_GRAY_ALPHA && - info_ptr->bit_depth == 8 && - png_ptr->screen_gamma == PNG_GAMMA_sRGB && - image->colormap_entries == 256) - break; - - goto bad_output; - - case PNG_CMAP_RGB: - /* Output must be 8-bit sRGB encoded RGB */ - if (info_ptr->color_type == PNG_COLOR_TYPE_RGB && - info_ptr->bit_depth == 8 && - png_ptr->screen_gamma == PNG_GAMMA_sRGB && - image->colormap_entries == 216) - break; - - goto bad_output; - - case PNG_CMAP_RGB_ALPHA: - /* Output must be 8-bit sRGB encoded RGBA */ - if (info_ptr->color_type == PNG_COLOR_TYPE_RGB_ALPHA && - info_ptr->bit_depth == 8 && - png_ptr->screen_gamma == PNG_GAMMA_sRGB && - image->colormap_entries == 244 /* 216 + 1 + 27 */) - break; - - goto bad_output; - - default: - bad_output: - png_error(png_ptr, "bad color-map processing (internal error)"); - } - - /* Now read the rows. Do this here if it is possible to read directly into - * the output buffer, otherwise allocate a local row buffer of the maximum - * size libpng requires and call the relevant processing routine safely. - */ - { - png_voidp first_row = display->buffer; - ptrdiff_t row_bytes = display->row_stride; - - /* The following expression is designed to work correctly whether it gives - * a signed or an unsigned result. - */ - if (row_bytes < 0) - { - char *ptr = png_voidcast(char*, first_row); - ptr += (image->height-1) * (-row_bytes); - first_row = png_voidcast(png_voidp, ptr); - } - - display->first_row = first_row; - display->row_bytes = row_bytes; - } - - if (passes == 0) - { - int result; - png_voidp row = png_malloc(png_ptr, png_get_rowbytes(png_ptr, info_ptr)); - - display->local_row = row; - result = png_safe_execute(image, png_image_read_and_map, display); - display->local_row = NULL; - png_free(png_ptr, row); - - return result; - } - - else - { - png_alloc_size_t row_bytes = (png_alloc_size_t)display->row_bytes; - - while (--passes >= 0) - { - png_uint_32 y = image->height; - png_bytep row = png_voidcast(png_bytep, display->first_row); - - for (; y > 0; --y) - { - png_read_row(png_ptr, row, NULL); - row += row_bytes; - } - } - - return 1; - } -} - -/* Just the row reading part of png_image_read. */ -static int -png_image_read_composite(png_voidp argument) -{ - png_image_read_control *display = png_voidcast(png_image_read_control*, - argument); - png_imagep image = display->image; - png_structrp png_ptr = image->opaque->png_ptr; - int passes; - - switch (png_ptr->interlaced) - { - case PNG_INTERLACE_NONE: - passes = 1; - break; - - case PNG_INTERLACE_ADAM7: - passes = PNG_INTERLACE_ADAM7_PASSES; - break; - - default: - png_error(png_ptr, "unknown interlace type"); - } - - { - png_uint_32 height = image->height; - png_uint_32 width = image->width; - ptrdiff_t step_row = display->row_bytes; - unsigned int channels = - (image->format & PNG_FORMAT_FLAG_COLOR) != 0 ? 3 : 1; - int pass; - - for (pass = 0; pass < passes; ++pass) - { - unsigned int startx, stepx, stepy; - png_uint_32 y; - - if (png_ptr->interlaced == PNG_INTERLACE_ADAM7) - { - /* The row may be empty for a short image: */ - if (PNG_PASS_COLS(width, pass) == 0) - continue; - - startx = PNG_PASS_START_COL(pass) * channels; - stepx = PNG_PASS_COL_OFFSET(pass) * channels; - y = PNG_PASS_START_ROW(pass); - stepy = PNG_PASS_ROW_OFFSET(pass); - } - - else - { - y = 0; - startx = 0; - stepx = channels; - stepy = 1; - } - - for (; ylocal_row); - png_bytep outrow; - png_const_bytep end_row; - - /* Read the row, which is packed: */ - png_read_row(png_ptr, inrow, NULL); - - outrow = png_voidcast(png_bytep, display->first_row); - outrow += y * step_row; - end_row = outrow + width * channels; - - /* Now do the composition on each pixel in this row. */ - outrow += startx; - for (; outrow < end_row; outrow += stepx) - { - png_byte alpha = inrow[channels]; - - if (alpha > 0) /* else no change to the output */ - { - unsigned int c; - - for (c=0; cimage; - png_structrp png_ptr = image->opaque->png_ptr; - png_inforp info_ptr = image->opaque->info_ptr; - png_uint_32 height = image->height; - png_uint_32 width = image->width; - int pass, passes; - - /* Double check the convoluted logic below. We expect to get here with - * libpng doing rgb to gray and gamma correction but background processing - * left to the png_image_read_background function. The rows libpng produce - * might be 8 or 16-bit but should always have two channels; gray plus alpha. - */ - if ((png_ptr->transformations & PNG_RGB_TO_GRAY) == 0) - png_error(png_ptr, "lost rgb to gray"); - - if ((png_ptr->transformations & PNG_COMPOSE) != 0) - png_error(png_ptr, "unexpected compose"); - - if (png_get_channels(png_ptr, info_ptr) != 2) - png_error(png_ptr, "lost/gained channels"); - - /* Expect the 8-bit case to always remove the alpha channel */ - if ((image->format & PNG_FORMAT_FLAG_LINEAR) == 0 && - (image->format & PNG_FORMAT_FLAG_ALPHA) != 0) - png_error(png_ptr, "unexpected 8-bit transformation"); - - switch (png_ptr->interlaced) - { - case PNG_INTERLACE_NONE: - passes = 1; - break; - - case PNG_INTERLACE_ADAM7: - passes = PNG_INTERLACE_ADAM7_PASSES; - break; - - default: - png_error(png_ptr, "unknown interlace type"); - } - - /* Use direct access to info_ptr here because otherwise the simplified API - * would require PNG_EASY_ACCESS_SUPPORTED (just for this.) Note this is - * checking the value after libpng expansions, not the original value in the - * PNG. - */ - switch (info_ptr->bit_depth) - { - case 8: - /* 8-bit sRGB gray values with an alpha channel; the alpha channel is - * to be removed by composing on a background: either the row if - * display->background is NULL or display->background->green if not. - * Unlike the code above ALPHA_OPTIMIZED has *not* been done. - */ - { - png_bytep first_row = png_voidcast(png_bytep, display->first_row); - ptrdiff_t step_row = display->row_bytes; - - for (pass = 0; pass < passes; ++pass) - { - unsigned int startx, stepx, stepy; - png_uint_32 y; - - if (png_ptr->interlaced == PNG_INTERLACE_ADAM7) - { - /* The row may be empty for a short image: */ - if (PNG_PASS_COLS(width, pass) == 0) - continue; - - startx = PNG_PASS_START_COL(pass); - stepx = PNG_PASS_COL_OFFSET(pass); - y = PNG_PASS_START_ROW(pass); - stepy = PNG_PASS_ROW_OFFSET(pass); - } - - else - { - y = 0; - startx = 0; - stepx = stepy = 1; - } - - if (display->background == NULL) - { - for (; ylocal_row); - png_bytep outrow = first_row + y * step_row; - png_const_bytep end_row = outrow + width; - - /* Read the row, which is packed: */ - png_read_row(png_ptr, inrow, NULL); - - /* Now do the composition on each pixel in this row. */ - outrow += startx; - for (; outrow < end_row; outrow += stepx) - { - png_byte alpha = inrow[1]; - - if (alpha > 0) /* else no change to the output */ - { - png_uint_32 component = inrow[0]; - - if (alpha < 255) /* else just use component */ - { - /* Since PNG_OPTIMIZED_ALPHA was not set it is - * necessary to invert the sRGB transfer - * function and multiply the alpha out. - */ - component = png_sRGB_table[component] * alpha; - component += png_sRGB_table[outrow[0]] * - (255-alpha); - component = PNG_sRGB_FROM_LINEAR(component); - } - - outrow[0] = (png_byte)component; - } - - inrow += 2; /* gray and alpha channel */ - } - } - } - - else /* constant background value */ - { - png_byte background8 = display->background->green; - png_uint_16 background = png_sRGB_table[background8]; - - for (; ylocal_row); - png_bytep outrow = first_row + y * step_row; - png_const_bytep end_row = outrow + width; - - /* Read the row, which is packed: */ - png_read_row(png_ptr, inrow, NULL); - - /* Now do the composition on each pixel in this row. */ - outrow += startx; - for (; outrow < end_row; outrow += stepx) - { - png_byte alpha = inrow[1]; - - if (alpha > 0) /* else use background */ - { - png_uint_32 component = inrow[0]; - - if (alpha < 255) /* else just use component */ - { - component = png_sRGB_table[component] * alpha; - component += background * (255-alpha); - component = PNG_sRGB_FROM_LINEAR(component); - } - - outrow[0] = (png_byte)component; - } - - else - outrow[0] = background8; - - inrow += 2; /* gray and alpha channel */ - } - } - } - } - } - break; - - case 16: - /* 16-bit linear with pre-multiplied alpha; the pre-multiplication must - * still be done and, maybe, the alpha channel removed. This code also - * handles the alpha-first option. - */ - { - png_uint_16p first_row = png_voidcast(png_uint_16p, - display->first_row); - /* The division by two is safe because the caller passed in a - * stride which was multiplied by 2 (below) to get row_bytes. - */ - ptrdiff_t step_row = display->row_bytes / 2; - unsigned int preserve_alpha = (image->format & - PNG_FORMAT_FLAG_ALPHA) != 0; - unsigned int outchannels = 1U+preserve_alpha; - int swap_alpha = 0; - -# ifdef PNG_SIMPLIFIED_READ_AFIRST_SUPPORTED - if (preserve_alpha != 0 && - (image->format & PNG_FORMAT_FLAG_AFIRST) != 0) - swap_alpha = 1; -# endif - - for (pass = 0; pass < passes; ++pass) - { - unsigned int startx, stepx, stepy; - png_uint_32 y; - - /* The 'x' start and step are adjusted to output components here. - */ - if (png_ptr->interlaced == PNG_INTERLACE_ADAM7) - { - /* The row may be empty for a short image: */ - if (PNG_PASS_COLS(width, pass) == 0) - continue; - - startx = PNG_PASS_START_COL(pass) * outchannels; - stepx = PNG_PASS_COL_OFFSET(pass) * outchannels; - y = PNG_PASS_START_ROW(pass); - stepy = PNG_PASS_ROW_OFFSET(pass); - } - - else - { - y = 0; - startx = 0; - stepx = outchannels; - stepy = 1; - } - - for (; ylocal_row), NULL); - inrow = png_voidcast(png_const_uint_16p, display->local_row); - - /* Now do the pre-multiplication on each pixel in this row. - */ - outrow += startx; - for (; outrow < end_row; outrow += stepx) - { - png_uint_32 component = inrow[0]; - png_uint_16 alpha = inrow[1]; - - if (alpha > 0) /* else 0 */ - { - if (alpha < 65535) /* else just use component */ - { - component *= alpha; - component += 32767; - component /= 65535; - } - } - - else - component = 0; - - outrow[swap_alpha] = (png_uint_16)component; - if (preserve_alpha != 0) - outrow[1 ^ swap_alpha] = alpha; - - inrow += 2; /* components and alpha channel */ - } - } - } - } - break; - -#ifdef __GNUC__ - default: - png_error(png_ptr, "unexpected bit depth"); -#endif - } - - return 1; -} - -/* The guts of png_image_finish_read as a png_safe_execute callback. */ -static int -png_image_read_direct(png_voidp argument) -{ - png_image_read_control *display = png_voidcast(png_image_read_control*, - argument); - png_imagep image = display->image; - png_structrp png_ptr = image->opaque->png_ptr; - png_inforp info_ptr = image->opaque->info_ptr; - - png_uint_32 format = image->format; - int linear = (format & PNG_FORMAT_FLAG_LINEAR) != 0; - int do_local_compose = 0; - int do_local_background = 0; /* to avoid double gamma correction bug */ - int passes = 0; - - /* Add transforms to ensure the correct output format is produced then check - * that the required implementation support is there. Always expand; always - * need 8 bits minimum, no palette and expanded tRNS. - */ - png_set_expand(png_ptr); - - /* Now check the format to see if it was modified. */ - { - png_uint_32 base_format = png_image_format(png_ptr) & - ~PNG_FORMAT_FLAG_COLORMAP /* removed by png_set_expand */; - png_uint_32 change = format ^ base_format; - png_fixed_point output_gamma; - int mode; /* alpha mode */ - - /* Do this first so that we have a record if rgb to gray is happening. */ - if ((change & PNG_FORMAT_FLAG_COLOR) != 0) - { - /* gray<->color transformation required. */ - if ((format & PNG_FORMAT_FLAG_COLOR) != 0) - png_set_gray_to_rgb(png_ptr); - - else - { - /* libpng can't do both rgb to gray and - * background/pre-multiplication if there is also significant gamma - * correction, because both operations require linear colors and - * the code only supports one transform doing the gamma correction. - * Handle this by doing the pre-multiplication or background - * operation in this code, if necessary. - * - * TODO: fix this by rewriting pngrtran.c (!) - * - * For the moment (given that fixing this in pngrtran.c is an - * enormous change) 'do_local_background' is used to indicate that - * the problem exists. - */ - if ((base_format & PNG_FORMAT_FLAG_ALPHA) != 0) - do_local_background = 1/*maybe*/; - - png_set_rgb_to_gray_fixed(png_ptr, PNG_ERROR_ACTION_NONE, - PNG_RGB_TO_GRAY_DEFAULT, PNG_RGB_TO_GRAY_DEFAULT); - } - - change &= ~PNG_FORMAT_FLAG_COLOR; - } - - /* Set the gamma appropriately, linear for 16-bit input, sRGB otherwise. - */ - { - png_fixed_point input_gamma_default; - - if ((base_format & PNG_FORMAT_FLAG_LINEAR) != 0 && - (image->flags & PNG_IMAGE_FLAG_16BIT_sRGB) == 0) - input_gamma_default = PNG_GAMMA_LINEAR; - else - input_gamma_default = PNG_DEFAULT_sRGB; - - /* Call png_set_alpha_mode to set the default for the input gamma; the - * output gamma is set by a second call below. - */ - png_set_alpha_mode_fixed(png_ptr, PNG_ALPHA_PNG, input_gamma_default); - } - - if (linear != 0) - { - /* If there *is* an alpha channel in the input it must be multiplied - * out; use PNG_ALPHA_STANDARD, otherwise just use PNG_ALPHA_PNG. - */ - if ((base_format & PNG_FORMAT_FLAG_ALPHA) != 0) - mode = PNG_ALPHA_STANDARD; /* associated alpha */ - - else - mode = PNG_ALPHA_PNG; - - output_gamma = PNG_GAMMA_LINEAR; - } - - else - { - mode = PNG_ALPHA_PNG; - output_gamma = PNG_DEFAULT_sRGB; - } - - if ((change & PNG_FORMAT_FLAG_ASSOCIATED_ALPHA) != 0) - { - mode = PNG_ALPHA_OPTIMIZED; - change &= ~PNG_FORMAT_FLAG_ASSOCIATED_ALPHA; - } - - /* If 'do_local_background' is set check for the presence of gamma - * correction; this is part of the work-round for the libpng bug - * described above. - * - * TODO: fix libpng and remove this. - */ - if (do_local_background != 0) - { - png_fixed_point gtest; - - /* This is 'png_gamma_threshold' from pngrtran.c; the test used for - * gamma correction, the screen gamma hasn't been set on png_struct - * yet; it's set below. png_struct::gamma, however, is set to the - * final value. - */ - if (png_muldiv(>est, output_gamma, png_ptr->colorspace.gamma, - PNG_FP_1) != 0 && png_gamma_significant(gtest) == 0) - do_local_background = 0; - - else if (mode == PNG_ALPHA_STANDARD) - { - do_local_background = 2/*required*/; - mode = PNG_ALPHA_PNG; /* prevent libpng doing it */ - } - - /* else leave as 1 for the checks below */ - } - - /* If the bit-depth changes then handle that here. */ - if ((change & PNG_FORMAT_FLAG_LINEAR) != 0) - { - if (linear != 0 /*16-bit output*/) - png_set_expand_16(png_ptr); - - else /* 8-bit output */ - png_set_scale_16(png_ptr); - - change &= ~PNG_FORMAT_FLAG_LINEAR; - } - - /* Now the background/alpha channel changes. */ - if ((change & PNG_FORMAT_FLAG_ALPHA) != 0) - { - /* Removing an alpha channel requires composition for the 8-bit - * formats; for the 16-bit it is already done, above, by the - * pre-multiplication and the channel just needs to be stripped. - */ - if ((base_format & PNG_FORMAT_FLAG_ALPHA) != 0) - { - /* If RGB->gray is happening the alpha channel must be left and the - * operation completed locally. - * - * TODO: fix libpng and remove this. - */ - if (do_local_background != 0) - do_local_background = 2/*required*/; - - /* 16-bit output: just remove the channel */ - else if (linear != 0) /* compose on black (well, pre-multiply) */ - png_set_strip_alpha(png_ptr); - - /* 8-bit output: do an appropriate compose */ - else if (display->background != NULL) - { - png_color_16 c; - - c.index = 0; /*unused*/ - c.red = display->background->red; - c.green = display->background->green; - c.blue = display->background->blue; - c.gray = display->background->green; - - /* This is always an 8-bit sRGB value, using the 'green' channel - * for gray is much better than calculating the luminance here; - * we can get off-by-one errors in that calculation relative to - * the app expectations and that will show up in transparent - * pixels. - */ - png_set_background_fixed(png_ptr, &c, - PNG_BACKGROUND_GAMMA_SCREEN, 0/*need_expand*/, - 0/*gamma: not used*/); - } - - else /* compose on row: implemented below. */ - { - do_local_compose = 1; - /* This leaves the alpha channel in the output, so it has to be - * removed by the code below. Set the encoding to the 'OPTIMIZE' - * one so the code only has to hack on the pixels that require - * composition. - */ - mode = PNG_ALPHA_OPTIMIZED; - } - } - - else /* output needs an alpha channel */ - { - /* This is tricky because it happens before the swap operation has - * been accomplished; however, the swap does *not* swap the added - * alpha channel (weird API), so it must be added in the correct - * place. - */ - png_uint_32 filler; /* opaque filler */ - int where; - - if (linear != 0) - filler = 65535; - - else - filler = 255; - -#ifdef PNG_FORMAT_AFIRST_SUPPORTED - if ((format & PNG_FORMAT_FLAG_AFIRST) != 0) - { - where = PNG_FILLER_BEFORE; - change &= ~PNG_FORMAT_FLAG_AFIRST; - } - - else -#endif - where = PNG_FILLER_AFTER; - - png_set_add_alpha(png_ptr, filler, where); - } - - /* This stops the (irrelevant) call to swap_alpha below. */ - change &= ~PNG_FORMAT_FLAG_ALPHA; - } - - /* Now set the alpha mode correctly; this is always done, even if there is - * no alpha channel in either the input or the output because it correctly - * sets the output gamma. - */ - png_set_alpha_mode_fixed(png_ptr, mode, output_gamma); - -# ifdef PNG_FORMAT_BGR_SUPPORTED - if ((change & PNG_FORMAT_FLAG_BGR) != 0) - { - /* Check only the output format; PNG is never BGR; don't do this if - * the output is gray, but fix up the 'format' value in that case. - */ - if ((format & PNG_FORMAT_FLAG_COLOR) != 0) - png_set_bgr(png_ptr); - - else - format &= ~PNG_FORMAT_FLAG_BGR; - - change &= ~PNG_FORMAT_FLAG_BGR; - } -# endif - -# ifdef PNG_FORMAT_AFIRST_SUPPORTED - if ((change & PNG_FORMAT_FLAG_AFIRST) != 0) - { - /* Only relevant if there is an alpha channel - it's particularly - * important to handle this correctly because do_local_compose may - * be set above and then libpng will keep the alpha channel for this - * code to remove. - */ - if ((format & PNG_FORMAT_FLAG_ALPHA) != 0) - { - /* Disable this if doing a local background, - * TODO: remove this when local background is no longer required. - */ - if (do_local_background != 2) - png_set_swap_alpha(png_ptr); - } - - else - format &= ~PNG_FORMAT_FLAG_AFIRST; - - change &= ~PNG_FORMAT_FLAG_AFIRST; - } -# endif - - /* If the *output* is 16-bit then we need to check for a byte-swap on this - * architecture. - */ - if (linear != 0) - { - png_uint_16 le = 0x0001; - - if ((*(png_const_bytep) & le) != 0) - png_set_swap(png_ptr); - } - - /* If change is not now 0 some transformation is missing - error out. */ - if (change != 0) - png_error(png_ptr, "png_read_image: unsupported transformation"); - } - - PNG_SKIP_CHUNKS(png_ptr); - - /* Update the 'info' structure and make sure the result is as required; first - * make sure to turn on the interlace handling if it will be required - * (because it can't be turned on *after* the call to png_read_update_info!) - * - * TODO: remove the do_local_background fixup below. - */ - if (do_local_compose == 0 && do_local_background != 2) - passes = png_set_interlace_handling(png_ptr); - - png_read_update_info(png_ptr, info_ptr); - - { - png_uint_32 info_format = 0; - - if ((info_ptr->color_type & PNG_COLOR_MASK_COLOR) != 0) - info_format |= PNG_FORMAT_FLAG_COLOR; - - if ((info_ptr->color_type & PNG_COLOR_MASK_ALPHA) != 0) - { - /* do_local_compose removes this channel below. */ - if (do_local_compose == 0) - { - /* do_local_background does the same if required. */ - if (do_local_background != 2 || - (format & PNG_FORMAT_FLAG_ALPHA) != 0) - info_format |= PNG_FORMAT_FLAG_ALPHA; - } - } - - else if (do_local_compose != 0) /* internal error */ - png_error(png_ptr, "png_image_read: alpha channel lost"); - - if ((format & PNG_FORMAT_FLAG_ASSOCIATED_ALPHA) != 0) { - info_format |= PNG_FORMAT_FLAG_ASSOCIATED_ALPHA; - } - - if (info_ptr->bit_depth == 16) - info_format |= PNG_FORMAT_FLAG_LINEAR; - -#ifdef PNG_FORMAT_BGR_SUPPORTED - if ((png_ptr->transformations & PNG_BGR) != 0) - info_format |= PNG_FORMAT_FLAG_BGR; -#endif - -#ifdef PNG_FORMAT_AFIRST_SUPPORTED - if (do_local_background == 2) - { - if ((format & PNG_FORMAT_FLAG_AFIRST) != 0) - info_format |= PNG_FORMAT_FLAG_AFIRST; - } - - if ((png_ptr->transformations & PNG_SWAP_ALPHA) != 0 || - ((png_ptr->transformations & PNG_ADD_ALPHA) != 0 && - (png_ptr->flags & PNG_FLAG_FILLER_AFTER) == 0)) - { - if (do_local_background == 2) - png_error(png_ptr, "unexpected alpha swap transformation"); - - info_format |= PNG_FORMAT_FLAG_AFIRST; - } -# endif - - /* This is actually an internal error. */ - if (info_format != format) - png_error(png_ptr, "png_read_image: invalid transformations"); - } - - /* Now read the rows. If do_local_compose is set then it is necessary to use - * a local row buffer. The output will be GA, RGBA or BGRA and must be - * converted to G, RGB or BGR as appropriate. The 'local_row' member of the - * display acts as a flag. - */ - { - png_voidp first_row = display->buffer; - ptrdiff_t row_bytes = display->row_stride; - - if (linear != 0) - row_bytes *= 2; - - /* The following expression is designed to work correctly whether it gives - * a signed or an unsigned result. - */ - if (row_bytes < 0) - { - char *ptr = png_voidcast(char*, first_row); - ptr += (image->height-1) * (-row_bytes); - first_row = png_voidcast(png_voidp, ptr); - } - - display->first_row = first_row; - display->row_bytes = row_bytes; - } - - if (do_local_compose != 0) - { - int result; - png_voidp row = png_malloc(png_ptr, png_get_rowbytes(png_ptr, info_ptr)); - - display->local_row = row; - result = png_safe_execute(image, png_image_read_composite, display); - display->local_row = NULL; - png_free(png_ptr, row); - - return result; - } - - else if (do_local_background == 2) - { - int result; - png_voidp row = png_malloc(png_ptr, png_get_rowbytes(png_ptr, info_ptr)); - - display->local_row = row; - result = png_safe_execute(image, png_image_read_background, display); - display->local_row = NULL; - png_free(png_ptr, row); - - return result; - } - - else - { - png_alloc_size_t row_bytes = (png_alloc_size_t)display->row_bytes; - - while (--passes >= 0) - { - png_uint_32 y = image->height; - png_bytep row = png_voidcast(png_bytep, display->first_row); - - for (; y > 0; --y) - { - png_read_row(png_ptr, row, NULL); - row += row_bytes; - } - } - - return 1; - } -} - -int PNGAPI -png_image_finish_read(png_imagep image, png_const_colorp background, - void *buffer, png_int_32 row_stride, void *colormap) -{ - if (image != NULL && image->version == PNG_IMAGE_VERSION) - { - /* Check for row_stride overflow. This check is not performed on the - * original PNG format because it may not occur in the output PNG format - * and libpng deals with the issues of reading the original. - */ - unsigned int channels = PNG_IMAGE_PIXEL_CHANNELS(image->format); - - /* The following checks just the 'row_stride' calculation to ensure it - * fits in a signed 32-bit value. Because channels/components can be - * either 1 or 2 bytes in size the length of a row can still overflow 32 - * bits; this is just to verify that the 'row_stride' argument can be - * represented. - */ - if (image->width <= 0x7fffffffU/channels) /* no overflow */ - { - png_uint_32 check; - png_uint_32 png_row_stride = image->width * channels; - - if (row_stride == 0) - row_stride = (png_int_32)/*SAFE*/png_row_stride; - - if (row_stride < 0) - check = (png_uint_32)(-row_stride); - - else - check = (png_uint_32)row_stride; - - /* This verifies 'check', the absolute value of the actual stride - * passed in and detects overflow in the application calculation (i.e. - * if the app did actually pass in a non-zero 'row_stride'. - */ - if (image->opaque != NULL && buffer != NULL && check >= png_row_stride) - { - /* Now check for overflow of the image buffer calculation; this - * limits the whole image size to 32 bits for API compatibility with - * the current, 32-bit, PNG_IMAGE_BUFFER_SIZE macro. - * - * The PNG_IMAGE_BUFFER_SIZE macro is: - * - * (PNG_IMAGE_PIXEL_COMPONENT_SIZE(fmt)*height*(row_stride)) - * - * And the component size is always 1 or 2, so make sure that the - * number of *bytes* that the application is saying are available - * does actually fit into a 32-bit number. - * - * NOTE: this will be changed in 1.7 because PNG_IMAGE_BUFFER_SIZE - * will be changed to use png_alloc_size_t; bigger images can be - * accommodated on 64-bit systems. - */ - if (image->height <= - 0xffffffffU/PNG_IMAGE_PIXEL_COMPONENT_SIZE(image->format)/check) - { - if ((image->format & PNG_FORMAT_FLAG_COLORMAP) == 0 || - (image->colormap_entries > 0 && colormap != NULL)) - { - int result; - png_image_read_control display; - - memset(&display, 0, (sizeof display)); - display.image = image; - display.buffer = buffer; - display.row_stride = row_stride; - display.colormap = colormap; - display.background = background; - display.local_row = NULL; - - /* Choose the correct 'end' routine; for the color-map case - * all the setup has already been done. - */ - if ((image->format & PNG_FORMAT_FLAG_COLORMAP) != 0) - result = - png_safe_execute(image, - png_image_read_colormap, &display) && - png_safe_execute(image, - png_image_read_colormapped, &display); - - else - result = - png_safe_execute(image, - png_image_read_direct, &display); - - png_image_free(image); - return result; - } - - else - return png_image_error(image, - "png_image_finish_read[color-map]: no color-map"); - } - - else - return png_image_error(image, - "png_image_finish_read: image too large"); - } - - else - return png_image_error(image, - "png_image_finish_read: invalid argument"); - } - - else - return png_image_error(image, - "png_image_finish_read: row_stride too large"); - } - - else if (image != NULL) - return png_image_error(image, - "png_image_finish_read: damaged PNG_IMAGE_VERSION"); - - return 0; -} - -#endif /* SIMPLIFIED_READ */ -#endif /* READ */ diff --git a/dep/libpng/src/pngrio.c b/dep/libpng/src/pngrio.c deleted file mode 100644 index 794635810..000000000 --- a/dep/libpng/src/pngrio.c +++ /dev/null @@ -1,120 +0,0 @@ - -/* pngrio.c - functions for data input - * - * Copyright (c) 2018 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2016,2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - * - * This file provides a location for all input. Users who need - * special handling are expected to write a function that has the same - * arguments as this and performs a similar function, but that possibly - * has a different input method. Note that you shouldn't change this - * function, but rather write a replacement function and then make - * libpng use it at run time with png_set_read_fn(...). - */ - -#include "pngpriv.h" - -#ifdef PNG_READ_SUPPORTED - -/* Read the data from whatever input you are using. The default routine - * reads from a file pointer. Note that this routine sometimes gets called - * with very small lengths, so you should implement some kind of simple - * buffering if you are using unbuffered reads. This should never be asked - * to read more than 64K on a 16-bit machine. - */ -void /* PRIVATE */ -png_read_data(png_structrp png_ptr, png_bytep data, size_t length) -{ - png_debug1(4, "reading %d bytes", (int)length); - - if (png_ptr->read_data_fn != NULL) - (*(png_ptr->read_data_fn))(png_ptr, data, length); - - else - png_error(png_ptr, "Call to NULL read function"); -} - -#ifdef PNG_STDIO_SUPPORTED -/* This is the function that does the actual reading of data. If you are - * not reading from a standard C stream, you should create a replacement - * read_data function and use it at run time with png_set_read_fn(), rather - * than changing the library. - */ -void PNGCBAPI -png_default_read_data(png_structp png_ptr, png_bytep data, size_t length) -{ - size_t check; - - if (png_ptr == NULL) - return; - - /* fread() returns 0 on error, so it is OK to store this in a size_t - * instead of an int, which is what fread() actually returns. - */ - check = fread(data, 1, length, png_voidcast(png_FILE_p, png_ptr->io_ptr)); - - if (check != length) - png_error(png_ptr, "Read Error"); -} -#endif - -/* This function allows the application to supply a new input function - * for libpng if standard C streams aren't being used. - * - * This function takes as its arguments: - * - * png_ptr - pointer to a png input data structure - * - * io_ptr - pointer to user supplied structure containing info about - * the input functions. May be NULL. - * - * read_data_fn - pointer to a new input function that takes as its - * arguments a pointer to a png_struct, a pointer to - * a location where input data can be stored, and a 32-bit - * unsigned int that is the number of bytes to be read. - * To exit and output any fatal error messages the new write - * function should call png_error(png_ptr, "Error msg"). - * May be NULL, in which case libpng's default function will - * be used. - */ -void PNGAPI -png_set_read_fn(png_structrp png_ptr, png_voidp io_ptr, - png_rw_ptr read_data_fn) -{ - if (png_ptr == NULL) - return; - - png_ptr->io_ptr = io_ptr; - -#ifdef PNG_STDIO_SUPPORTED - if (read_data_fn != NULL) - png_ptr->read_data_fn = read_data_fn; - - else - png_ptr->read_data_fn = png_default_read_data; -#else - png_ptr->read_data_fn = read_data_fn; -#endif - -#ifdef PNG_WRITE_SUPPORTED - /* It is an error to write to a read device */ - if (png_ptr->write_data_fn != NULL) - { - png_ptr->write_data_fn = NULL; - png_warning(png_ptr, - "Can't set both read_data_fn and write_data_fn in the" - " same structure"); - } -#endif - -#ifdef PNG_WRITE_FLUSH_SUPPORTED - png_ptr->output_flush_fn = NULL; -#endif -} -#endif /* READ */ diff --git a/dep/libpng/src/pngrtran.c b/dep/libpng/src/pngrtran.c deleted file mode 100644 index 1526123e0..000000000 --- a/dep/libpng/src/pngrtran.c +++ /dev/null @@ -1,5040 +0,0 @@ - -/* pngrtran.c - transforms the data in a row for PNG readers - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - * - * This file contains functions optionally called by an application - * in order to tell libpng how to handle data when reading a PNG. - * Transformations that are used in both reading and writing are - * in pngtrans.c. - */ - -#include "pngpriv.h" - -#ifdef PNG_ARM_NEON_IMPLEMENTATION -# if PNG_ARM_NEON_IMPLEMENTATION == 1 -# define PNG_ARM_NEON_INTRINSICS_AVAILABLE -# if defined(_MSC_VER) && !defined(__clang__) && defined(_M_ARM64) -# include -# else -# include -# endif -# endif -#endif - -#ifdef PNG_READ_SUPPORTED - -/* Set the action on getting a CRC error for an ancillary or critical chunk. */ -void PNGAPI -png_set_crc_action(png_structrp png_ptr, int crit_action, int ancil_action) -{ - png_debug(1, "in png_set_crc_action"); - - if (png_ptr == NULL) - return; - - /* Tell libpng how we react to CRC errors in critical chunks */ - switch (crit_action) - { - case PNG_CRC_NO_CHANGE: /* Leave setting as is */ - break; - - case PNG_CRC_WARN_USE: /* Warn/use data */ - png_ptr->flags &= ~PNG_FLAG_CRC_CRITICAL_MASK; - png_ptr->flags |= PNG_FLAG_CRC_CRITICAL_USE; - break; - - case PNG_CRC_QUIET_USE: /* Quiet/use data */ - png_ptr->flags &= ~PNG_FLAG_CRC_CRITICAL_MASK; - png_ptr->flags |= PNG_FLAG_CRC_CRITICAL_USE | - PNG_FLAG_CRC_CRITICAL_IGNORE; - break; - - case PNG_CRC_WARN_DISCARD: /* Not a valid action for critical data */ - png_warning(png_ptr, - "Can't discard critical data on CRC error"); - /* FALLTHROUGH */ - case PNG_CRC_ERROR_QUIT: /* Error/quit */ - - case PNG_CRC_DEFAULT: - default: - png_ptr->flags &= ~PNG_FLAG_CRC_CRITICAL_MASK; - break; - } - - /* Tell libpng how we react to CRC errors in ancillary chunks */ - switch (ancil_action) - { - case PNG_CRC_NO_CHANGE: /* Leave setting as is */ - break; - - case PNG_CRC_WARN_USE: /* Warn/use data */ - png_ptr->flags &= ~PNG_FLAG_CRC_ANCILLARY_MASK; - png_ptr->flags |= PNG_FLAG_CRC_ANCILLARY_USE; - break; - - case PNG_CRC_QUIET_USE: /* Quiet/use data */ - png_ptr->flags &= ~PNG_FLAG_CRC_ANCILLARY_MASK; - png_ptr->flags |= PNG_FLAG_CRC_ANCILLARY_USE | - PNG_FLAG_CRC_ANCILLARY_NOWARN; - break; - - case PNG_CRC_ERROR_QUIT: /* Error/quit */ - png_ptr->flags &= ~PNG_FLAG_CRC_ANCILLARY_MASK; - png_ptr->flags |= PNG_FLAG_CRC_ANCILLARY_NOWARN; - break; - - case PNG_CRC_WARN_DISCARD: /* Warn/discard data */ - - case PNG_CRC_DEFAULT: - default: - png_ptr->flags &= ~PNG_FLAG_CRC_ANCILLARY_MASK; - break; - } -} - -#ifdef PNG_READ_TRANSFORMS_SUPPORTED -/* Is it OK to set a transformation now? Only if png_start_read_image or - * png_read_update_info have not been called. It is not necessary for the IHDR - * to have been read in all cases; the need_IHDR parameter allows for this - * check too. - */ -static int -png_rtran_ok(png_structrp png_ptr, int need_IHDR) -{ - if (png_ptr != NULL) - { - if ((png_ptr->flags & PNG_FLAG_ROW_INIT) != 0) - png_app_error(png_ptr, - "invalid after png_start_read_image or png_read_update_info"); - - else if (need_IHDR && (png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_app_error(png_ptr, "invalid before the PNG header has been read"); - - else - { - /* Turn on failure to initialize correctly for all transforms. */ - png_ptr->flags |= PNG_FLAG_DETECT_UNINITIALIZED; - - return 1; /* Ok */ - } - } - - return 0; /* no png_error possible! */ -} -#endif - -#ifdef PNG_READ_BACKGROUND_SUPPORTED -/* Handle alpha and tRNS via a background color */ -void PNGFAPI -png_set_background_fixed(png_structrp png_ptr, - png_const_color_16p background_color, int background_gamma_code, - int need_expand, png_fixed_point background_gamma) -{ - png_debug(1, "in png_set_background_fixed"); - - if (png_rtran_ok(png_ptr, 0) == 0 || background_color == NULL) - return; - - if (background_gamma_code == PNG_BACKGROUND_GAMMA_UNKNOWN) - { - png_warning(png_ptr, "Application must supply a known background gamma"); - return; - } - - png_ptr->transformations |= PNG_COMPOSE | PNG_STRIP_ALPHA; - png_ptr->transformations &= ~PNG_ENCODE_ALPHA; - png_ptr->flags &= ~PNG_FLAG_OPTIMIZE_ALPHA; - - png_ptr->background = *background_color; - png_ptr->background_gamma = background_gamma; - png_ptr->background_gamma_type = (png_byte)(background_gamma_code); - if (need_expand != 0) - png_ptr->transformations |= PNG_BACKGROUND_EXPAND; - else - png_ptr->transformations &= ~PNG_BACKGROUND_EXPAND; -} - -# ifdef PNG_FLOATING_POINT_SUPPORTED -void PNGAPI -png_set_background(png_structrp png_ptr, - png_const_color_16p background_color, int background_gamma_code, - int need_expand, double background_gamma) -{ - png_set_background_fixed(png_ptr, background_color, background_gamma_code, - need_expand, png_fixed(png_ptr, background_gamma, "png_set_background")); -} -# endif /* FLOATING_POINT */ -#endif /* READ_BACKGROUND */ - -/* Scale 16-bit depth files to 8-bit depth. If both of these are set then the - * one that pngrtran does first (scale) happens. This is necessary to allow the - * TRANSFORM and API behavior to be somewhat consistent, and it's simpler. - */ -#ifdef PNG_READ_SCALE_16_TO_8_SUPPORTED -void PNGAPI -png_set_scale_16(png_structrp png_ptr) -{ - png_debug(1, "in png_set_scale_16"); - - if (png_rtran_ok(png_ptr, 0) == 0) - return; - - png_ptr->transformations |= PNG_SCALE_16_TO_8; -} -#endif - -#ifdef PNG_READ_STRIP_16_TO_8_SUPPORTED -/* Chop 16-bit depth files to 8-bit depth */ -void PNGAPI -png_set_strip_16(png_structrp png_ptr) -{ - png_debug(1, "in png_set_strip_16"); - - if (png_rtran_ok(png_ptr, 0) == 0) - return; - - png_ptr->transformations |= PNG_16_TO_8; -} -#endif - -#ifdef PNG_READ_STRIP_ALPHA_SUPPORTED -void PNGAPI -png_set_strip_alpha(png_structrp png_ptr) -{ - png_debug(1, "in png_set_strip_alpha"); - - if (png_rtran_ok(png_ptr, 0) == 0) - return; - - png_ptr->transformations |= PNG_STRIP_ALPHA; -} -#endif - -#if defined(PNG_READ_ALPHA_MODE_SUPPORTED) || defined(PNG_READ_GAMMA_SUPPORTED) -static png_fixed_point -translate_gamma_flags(png_structrp png_ptr, png_fixed_point output_gamma, - int is_screen) -{ - /* Check for flag values. The main reason for having the old Mac value as a - * flag is that it is pretty near impossible to work out what the correct - * value is from Apple documentation - a working Mac system is needed to - * discover the value! - */ - if (output_gamma == PNG_DEFAULT_sRGB || - output_gamma == PNG_FP_1 / PNG_DEFAULT_sRGB) - { - /* If there is no sRGB support this just sets the gamma to the standard - * sRGB value. (This is a side effect of using this function!) - */ -# ifdef PNG_READ_sRGB_SUPPORTED - png_ptr->flags |= PNG_FLAG_ASSUME_sRGB; -# else - PNG_UNUSED(png_ptr) -# endif - if (is_screen != 0) - output_gamma = PNG_GAMMA_sRGB; - else - output_gamma = PNG_GAMMA_sRGB_INVERSE; - } - - else if (output_gamma == PNG_GAMMA_MAC_18 || - output_gamma == PNG_FP_1 / PNG_GAMMA_MAC_18) - { - if (is_screen != 0) - output_gamma = PNG_GAMMA_MAC_OLD; - else - output_gamma = PNG_GAMMA_MAC_INVERSE; - } - - return output_gamma; -} - -# ifdef PNG_FLOATING_POINT_SUPPORTED -static png_fixed_point -convert_gamma_value(png_structrp png_ptr, double output_gamma) -{ - /* The following silently ignores cases where fixed point (times 100,000) - * gamma values are passed to the floating point API. This is safe and it - * means the fixed point constants work just fine with the floating point - * API. The alternative would just lead to undetected errors and spurious - * bug reports. Negative values fail inside the _fixed API unless they - * correspond to the flag values. - */ - if (output_gamma > 0 && output_gamma < 128) - output_gamma *= PNG_FP_1; - - /* This preserves -1 and -2 exactly: */ - output_gamma = floor(output_gamma + .5); - - if (output_gamma > PNG_FP_MAX || output_gamma < PNG_FP_MIN) - png_fixed_error(png_ptr, "gamma value"); - - return (png_fixed_point)output_gamma; -} -# endif -#endif /* READ_ALPHA_MODE || READ_GAMMA */ - -#ifdef PNG_READ_ALPHA_MODE_SUPPORTED -void PNGFAPI -png_set_alpha_mode_fixed(png_structrp png_ptr, int mode, - png_fixed_point output_gamma) -{ - int compose = 0; - png_fixed_point file_gamma; - - png_debug(1, "in png_set_alpha_mode_fixed"); - - if (png_rtran_ok(png_ptr, 0) == 0) - return; - - output_gamma = translate_gamma_flags(png_ptr, output_gamma, 1/*screen*/); - - /* Validate the value to ensure it is in a reasonable range. The value - * is expected to be 1 or greater, but this range test allows for some - * viewing correction values. The intent is to weed out the API users - * who might use the inverse of the gamma value accidentally! - * - * In libpng 1.6.0, we changed from 0.07..3 to 0.01..100, to accommodate - * the optimal 16-bit gamma of 36 and its reciprocal. - */ - if (output_gamma < 1000 || output_gamma > 10000000) - png_error(png_ptr, "output gamma out of expected range"); - - /* The default file gamma is the inverse of the output gamma; the output - * gamma may be changed below so get the file value first: - */ - file_gamma = png_reciprocal(output_gamma); - - /* There are really 8 possibilities here, composed of any combination - * of: - * - * premultiply the color channels - * do not encode non-opaque pixels - * encode the alpha as well as the color channels - * - * The differences disappear if the input/output ('screen') gamma is 1.0, - * because then the encoding is a no-op and there is only the choice of - * premultiplying the color channels or not. - * - * png_set_alpha_mode and png_set_background interact because both use - * png_compose to do the work. Calling both is only useful when - * png_set_alpha_mode is used to set the default mode - PNG_ALPHA_PNG - along - * with a default gamma value. Otherwise PNG_COMPOSE must not be set. - */ - switch (mode) - { - case PNG_ALPHA_PNG: /* default: png standard */ - /* No compose, but it may be set by png_set_background! */ - png_ptr->transformations &= ~PNG_ENCODE_ALPHA; - png_ptr->flags &= ~PNG_FLAG_OPTIMIZE_ALPHA; - break; - - case PNG_ALPHA_ASSOCIATED: /* color channels premultiplied */ - compose = 1; - png_ptr->transformations &= ~PNG_ENCODE_ALPHA; - png_ptr->flags &= ~PNG_FLAG_OPTIMIZE_ALPHA; - /* The output is linear: */ - output_gamma = PNG_FP_1; - break; - - case PNG_ALPHA_OPTIMIZED: /* associated, non-opaque pixels linear */ - compose = 1; - png_ptr->transformations &= ~PNG_ENCODE_ALPHA; - png_ptr->flags |= PNG_FLAG_OPTIMIZE_ALPHA; - /* output_gamma records the encoding of opaque pixels! */ - break; - - case PNG_ALPHA_BROKEN: /* associated, non-linear, alpha encoded */ - compose = 1; - png_ptr->transformations |= PNG_ENCODE_ALPHA; - png_ptr->flags &= ~PNG_FLAG_OPTIMIZE_ALPHA; - break; - - default: - png_error(png_ptr, "invalid alpha mode"); - } - - /* Only set the default gamma if the file gamma has not been set (this has - * the side effect that the gamma in a second call to png_set_alpha_mode will - * be ignored.) - */ - if (png_ptr->colorspace.gamma == 0) - { - png_ptr->colorspace.gamma = file_gamma; - png_ptr->colorspace.flags |= PNG_COLORSPACE_HAVE_GAMMA; - } - - /* But always set the output gamma: */ - png_ptr->screen_gamma = output_gamma; - - /* Finally, if pre-multiplying, set the background fields to achieve the - * desired result. - */ - if (compose != 0) - { - /* And obtain alpha pre-multiplication by composing on black: */ - memset(&png_ptr->background, 0, (sizeof png_ptr->background)); - png_ptr->background_gamma = png_ptr->colorspace.gamma; /* just in case */ - png_ptr->background_gamma_type = PNG_BACKGROUND_GAMMA_FILE; - png_ptr->transformations &= ~PNG_BACKGROUND_EXPAND; - - if ((png_ptr->transformations & PNG_COMPOSE) != 0) - png_error(png_ptr, - "conflicting calls to set alpha mode and background"); - - png_ptr->transformations |= PNG_COMPOSE; - } -} - -# ifdef PNG_FLOATING_POINT_SUPPORTED -void PNGAPI -png_set_alpha_mode(png_structrp png_ptr, int mode, double output_gamma) -{ - png_set_alpha_mode_fixed(png_ptr, mode, convert_gamma_value(png_ptr, - output_gamma)); -} -# endif -#endif - -#ifdef PNG_READ_QUANTIZE_SUPPORTED -/* Dither file to 8-bit. Supply a palette, the current number - * of elements in the palette, the maximum number of elements - * allowed, and a histogram if possible. If the current number - * of colors is greater than the maximum number, the palette will be - * modified to fit in the maximum number. "full_quantize" indicates - * whether we need a quantizing cube set up for RGB images, or if we - * simply are reducing the number of colors in a paletted image. - */ - -typedef struct png_dsort_struct -{ - struct png_dsort_struct * next; - png_byte left; - png_byte right; -} png_dsort; -typedef png_dsort * png_dsortp; -typedef png_dsort * * png_dsortpp; - -void PNGAPI -png_set_quantize(png_structrp png_ptr, png_colorp palette, - int num_palette, int maximum_colors, png_const_uint_16p histogram, - int full_quantize) -{ - png_debug(1, "in png_set_quantize"); - - if (png_rtran_ok(png_ptr, 0) == 0) - return; - - png_ptr->transformations |= PNG_QUANTIZE; - - if (full_quantize == 0) - { - int i; - - png_ptr->quantize_index = (png_bytep)png_malloc(png_ptr, - (png_alloc_size_t)num_palette); - for (i = 0; i < num_palette; i++) - png_ptr->quantize_index[i] = (png_byte)i; - } - - if (num_palette > maximum_colors) - { - if (histogram != NULL) - { - /* This is easy enough, just throw out the least used colors. - * Perhaps not the best solution, but good enough. - */ - - int i; - - /* Initialize an array to sort colors */ - png_ptr->quantize_sort = (png_bytep)png_malloc(png_ptr, - (png_alloc_size_t)num_palette); - - /* Initialize the quantize_sort array */ - for (i = 0; i < num_palette; i++) - png_ptr->quantize_sort[i] = (png_byte)i; - - /* Find the least used palette entries by starting a - * bubble sort, and running it until we have sorted - * out enough colors. Note that we don't care about - * sorting all the colors, just finding which are - * least used. - */ - - for (i = num_palette - 1; i >= maximum_colors; i--) - { - int done; /* To stop early if the list is pre-sorted */ - int j; - - done = 1; - for (j = 0; j < i; j++) - { - if (histogram[png_ptr->quantize_sort[j]] - < histogram[png_ptr->quantize_sort[j + 1]]) - { - png_byte t; - - t = png_ptr->quantize_sort[j]; - png_ptr->quantize_sort[j] = png_ptr->quantize_sort[j + 1]; - png_ptr->quantize_sort[j + 1] = t; - done = 0; - } - } - - if (done != 0) - break; - } - - /* Swap the palette around, and set up a table, if necessary */ - if (full_quantize != 0) - { - int j = num_palette; - - /* Put all the useful colors within the max, but don't - * move the others. - */ - for (i = 0; i < maximum_colors; i++) - { - if ((int)png_ptr->quantize_sort[i] >= maximum_colors) - { - do - j--; - while ((int)png_ptr->quantize_sort[j] >= maximum_colors); - - palette[i] = palette[j]; - } - } - } - else - { - int j = num_palette; - - /* Move all the used colors inside the max limit, and - * develop a translation table. - */ - for (i = 0; i < maximum_colors; i++) - { - /* Only move the colors we need to */ - if ((int)png_ptr->quantize_sort[i] >= maximum_colors) - { - png_color tmp_color; - - do - j--; - while ((int)png_ptr->quantize_sort[j] >= maximum_colors); - - tmp_color = palette[j]; - palette[j] = palette[i]; - palette[i] = tmp_color; - /* Indicate where the color went */ - png_ptr->quantize_index[j] = (png_byte)i; - png_ptr->quantize_index[i] = (png_byte)j; - } - } - - /* Find closest color for those colors we are not using */ - for (i = 0; i < num_palette; i++) - { - if ((int)png_ptr->quantize_index[i] >= maximum_colors) - { - int min_d, k, min_k, d_index; - - /* Find the closest color to one we threw out */ - d_index = png_ptr->quantize_index[i]; - min_d = PNG_COLOR_DIST(palette[d_index], palette[0]); - for (k = 1, min_k = 0; k < maximum_colors; k++) - { - int d; - - d = PNG_COLOR_DIST(palette[d_index], palette[k]); - - if (d < min_d) - { - min_d = d; - min_k = k; - } - } - /* Point to closest color */ - png_ptr->quantize_index[i] = (png_byte)min_k; - } - } - } - png_free(png_ptr, png_ptr->quantize_sort); - png_ptr->quantize_sort = NULL; - } - else - { - /* This is much harder to do simply (and quickly). Perhaps - * we need to go through a median cut routine, but those - * don't always behave themselves with only a few colors - * as input. So we will just find the closest two colors, - * and throw out one of them (chosen somewhat randomly). - * [We don't understand this at all, so if someone wants to - * work on improving it, be our guest - AED, GRP] - */ - int i; - int max_d; - int num_new_palette; - png_dsortp t; - png_dsortpp hash; - - t = NULL; - - /* Initialize palette index arrays */ - png_ptr->index_to_palette = (png_bytep)png_malloc(png_ptr, - (png_alloc_size_t)num_palette); - png_ptr->palette_to_index = (png_bytep)png_malloc(png_ptr, - (png_alloc_size_t)num_palette); - - /* Initialize the sort array */ - for (i = 0; i < num_palette; i++) - { - png_ptr->index_to_palette[i] = (png_byte)i; - png_ptr->palette_to_index[i] = (png_byte)i; - } - - hash = (png_dsortpp)png_calloc(png_ptr, (png_alloc_size_t)(769 * - (sizeof (png_dsortp)))); - - num_new_palette = num_palette; - - /* Initial wild guess at how far apart the farthest pixel - * pair we will be eliminating will be. Larger - * numbers mean more areas will be allocated, Smaller - * numbers run the risk of not saving enough data, and - * having to do this all over again. - * - * I have not done extensive checking on this number. - */ - max_d = 96; - - while (num_new_palette > maximum_colors) - { - for (i = 0; i < num_new_palette - 1; i++) - { - int j; - - for (j = i + 1; j < num_new_palette; j++) - { - int d; - - d = PNG_COLOR_DIST(palette[i], palette[j]); - - if (d <= max_d) - { - - t = (png_dsortp)png_malloc_warn(png_ptr, - (png_alloc_size_t)(sizeof (png_dsort))); - - if (t == NULL) - break; - - t->next = hash[d]; - t->left = (png_byte)i; - t->right = (png_byte)j; - hash[d] = t; - } - } - if (t == NULL) - break; - } - - if (t != NULL) - for (i = 0; i <= max_d; i++) - { - if (hash[i] != NULL) - { - png_dsortp p; - - for (p = hash[i]; p; p = p->next) - { - if ((int)png_ptr->index_to_palette[p->left] - < num_new_palette && - (int)png_ptr->index_to_palette[p->right] - < num_new_palette) - { - int j, next_j; - - if (num_new_palette & 0x01) - { - j = p->left; - next_j = p->right; - } - else - { - j = p->right; - next_j = p->left; - } - - num_new_palette--; - palette[png_ptr->index_to_palette[j]] - = palette[num_new_palette]; - if (full_quantize == 0) - { - int k; - - for (k = 0; k < num_palette; k++) - { - if (png_ptr->quantize_index[k] == - png_ptr->index_to_palette[j]) - png_ptr->quantize_index[k] = - png_ptr->index_to_palette[next_j]; - - if ((int)png_ptr->quantize_index[k] == - num_new_palette) - png_ptr->quantize_index[k] = - png_ptr->index_to_palette[j]; - } - } - - png_ptr->index_to_palette[png_ptr->palette_to_index - [num_new_palette]] = png_ptr->index_to_palette[j]; - - png_ptr->palette_to_index[png_ptr->index_to_palette[j]] - = png_ptr->palette_to_index[num_new_palette]; - - png_ptr->index_to_palette[j] = - (png_byte)num_new_palette; - - png_ptr->palette_to_index[num_new_palette] = - (png_byte)j; - } - if (num_new_palette <= maximum_colors) - break; - } - if (num_new_palette <= maximum_colors) - break; - } - } - - for (i = 0; i < 769; i++) - { - if (hash[i] != NULL) - { - png_dsortp p = hash[i]; - while (p) - { - t = p->next; - png_free(png_ptr, p); - p = t; - } - } - hash[i] = 0; - } - max_d += 96; - } - png_free(png_ptr, hash); - png_free(png_ptr, png_ptr->palette_to_index); - png_free(png_ptr, png_ptr->index_to_palette); - png_ptr->palette_to_index = NULL; - png_ptr->index_to_palette = NULL; - } - num_palette = maximum_colors; - } - if (png_ptr->palette == NULL) - { - png_ptr->palette = palette; - } - png_ptr->num_palette = (png_uint_16)num_palette; - - if (full_quantize != 0) - { - int i; - png_bytep distance; - int total_bits = PNG_QUANTIZE_RED_BITS + PNG_QUANTIZE_GREEN_BITS + - PNG_QUANTIZE_BLUE_BITS; - int num_red = (1 << PNG_QUANTIZE_RED_BITS); - int num_green = (1 << PNG_QUANTIZE_GREEN_BITS); - int num_blue = (1 << PNG_QUANTIZE_BLUE_BITS); - size_t num_entries = ((size_t)1 << total_bits); - - png_ptr->palette_lookup = (png_bytep)png_calloc(png_ptr, - (png_alloc_size_t)(num_entries)); - - distance = (png_bytep)png_malloc(png_ptr, (png_alloc_size_t)num_entries); - - memset(distance, 0xff, num_entries); - - for (i = 0; i < num_palette; i++) - { - int ir, ig, ib; - int r = (palette[i].red >> (8 - PNG_QUANTIZE_RED_BITS)); - int g = (palette[i].green >> (8 - PNG_QUANTIZE_GREEN_BITS)); - int b = (palette[i].blue >> (8 - PNG_QUANTIZE_BLUE_BITS)); - - for (ir = 0; ir < num_red; ir++) - { - /* int dr = abs(ir - r); */ - int dr = ((ir > r) ? ir - r : r - ir); - int index_r = (ir << (PNG_QUANTIZE_BLUE_BITS + - PNG_QUANTIZE_GREEN_BITS)); - - for (ig = 0; ig < num_green; ig++) - { - /* int dg = abs(ig - g); */ - int dg = ((ig > g) ? ig - g : g - ig); - int dt = dr + dg; - int dm = ((dr > dg) ? dr : dg); - int index_g = index_r | (ig << PNG_QUANTIZE_BLUE_BITS); - - for (ib = 0; ib < num_blue; ib++) - { - int d_index = index_g | ib; - /* int db = abs(ib - b); */ - int db = ((ib > b) ? ib - b : b - ib); - int dmax = ((dm > db) ? dm : db); - int d = dmax + dt + db; - - if (d < (int)distance[d_index]) - { - distance[d_index] = (png_byte)d; - png_ptr->palette_lookup[d_index] = (png_byte)i; - } - } - } - } - } - - png_free(png_ptr, distance); - } -} -#endif /* READ_QUANTIZE */ - -#ifdef PNG_READ_GAMMA_SUPPORTED -void PNGFAPI -png_set_gamma_fixed(png_structrp png_ptr, png_fixed_point scrn_gamma, - png_fixed_point file_gamma) -{ - png_debug(1, "in png_set_gamma_fixed"); - - if (png_rtran_ok(png_ptr, 0) == 0) - return; - - /* New in libpng-1.5.4 - reserve particular negative values as flags. */ - scrn_gamma = translate_gamma_flags(png_ptr, scrn_gamma, 1/*screen*/); - file_gamma = translate_gamma_flags(png_ptr, file_gamma, 0/*file*/); - - /* Checking the gamma values for being >0 was added in 1.5.4 along with the - * premultiplied alpha support; this actually hides an undocumented feature - * of the previous implementation which allowed gamma processing to be - * disabled in background handling. There is no evidence (so far) that this - * was being used; however, png_set_background itself accepted and must still - * accept '0' for the gamma value it takes, because it isn't always used. - * - * Since this is an API change (albeit a very minor one that removes an - * undocumented API feature) the following checks were only enabled in - * libpng-1.6.0. - */ - if (file_gamma <= 0) - png_error(png_ptr, "invalid file gamma in png_set_gamma"); - - if (scrn_gamma <= 0) - png_error(png_ptr, "invalid screen gamma in png_set_gamma"); - - /* Set the gamma values unconditionally - this overrides the value in the PNG - * file if a gAMA chunk was present. png_set_alpha_mode provides a - * different, easier, way to default the file gamma. - */ - png_ptr->colorspace.gamma = file_gamma; - png_ptr->colorspace.flags |= PNG_COLORSPACE_HAVE_GAMMA; - png_ptr->screen_gamma = scrn_gamma; -} - -# ifdef PNG_FLOATING_POINT_SUPPORTED -void PNGAPI -png_set_gamma(png_structrp png_ptr, double scrn_gamma, double file_gamma) -{ - png_set_gamma_fixed(png_ptr, convert_gamma_value(png_ptr, scrn_gamma), - convert_gamma_value(png_ptr, file_gamma)); -} -# endif /* FLOATING_POINT */ -#endif /* READ_GAMMA */ - -#ifdef PNG_READ_EXPAND_SUPPORTED -/* Expand paletted images to RGB, expand grayscale images of - * less than 8-bit depth to 8-bit depth, and expand tRNS chunks - * to alpha channels. - */ -void PNGAPI -png_set_expand(png_structrp png_ptr) -{ - png_debug(1, "in png_set_expand"); - - if (png_rtran_ok(png_ptr, 0) == 0) - return; - - png_ptr->transformations |= (PNG_EXPAND | PNG_EXPAND_tRNS); -} - -/* GRR 19990627: the following three functions currently are identical - * to png_set_expand(). However, it is entirely reasonable that someone - * might wish to expand an indexed image to RGB but *not* expand a single, - * fully transparent palette entry to a full alpha channel--perhaps instead - * convert tRNS to the grayscale/RGB format (16-bit RGB value), or replace - * the transparent color with a particular RGB value, or drop tRNS entirely. - * IOW, a future version of the library may make the transformations flag - * a bit more fine-grained, with separate bits for each of these three - * functions. - * - * More to the point, these functions make it obvious what libpng will be - * doing, whereas "expand" can (and does) mean any number of things. - * - * GRP 20060307: In libpng-1.2.9, png_set_gray_1_2_4_to_8() was modified - * to expand only the sample depth but not to expand the tRNS to alpha - * and its name was changed to png_set_expand_gray_1_2_4_to_8(). - */ - -/* Expand paletted images to RGB. */ -void PNGAPI -png_set_palette_to_rgb(png_structrp png_ptr) -{ - png_debug(1, "in png_set_palette_to_rgb"); - - if (png_rtran_ok(png_ptr, 0) == 0) - return; - - png_ptr->transformations |= (PNG_EXPAND | PNG_EXPAND_tRNS); -} - -/* Expand grayscale images of less than 8-bit depth to 8 bits. */ -void PNGAPI -png_set_expand_gray_1_2_4_to_8(png_structrp png_ptr) -{ - png_debug(1, "in png_set_expand_gray_1_2_4_to_8"); - - if (png_rtran_ok(png_ptr, 0) == 0) - return; - - png_ptr->transformations |= PNG_EXPAND; -} - -/* Expand tRNS chunks to alpha channels. */ -void PNGAPI -png_set_tRNS_to_alpha(png_structrp png_ptr) -{ - png_debug(1, "in png_set_tRNS_to_alpha"); - - if (png_rtran_ok(png_ptr, 0) == 0) - return; - - png_ptr->transformations |= (PNG_EXPAND | PNG_EXPAND_tRNS); -} -#endif /* READ_EXPAND */ - -#ifdef PNG_READ_EXPAND_16_SUPPORTED -/* Expand to 16-bit channels, expand the tRNS chunk too (because otherwise - * it may not work correctly.) - */ -void PNGAPI -png_set_expand_16(png_structrp png_ptr) -{ - png_debug(1, "in png_set_expand_16"); - - if (png_rtran_ok(png_ptr, 0) == 0) - return; - - png_ptr->transformations |= (PNG_EXPAND_16 | PNG_EXPAND | PNG_EXPAND_tRNS); -} -#endif - -#ifdef PNG_READ_GRAY_TO_RGB_SUPPORTED -void PNGAPI -png_set_gray_to_rgb(png_structrp png_ptr) -{ - png_debug(1, "in png_set_gray_to_rgb"); - - if (png_rtran_ok(png_ptr, 0) == 0) - return; - - /* Because rgb must be 8 bits or more: */ - png_set_expand_gray_1_2_4_to_8(png_ptr); - png_ptr->transformations |= PNG_GRAY_TO_RGB; -} -#endif - -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED -void PNGFAPI -png_set_rgb_to_gray_fixed(png_structrp png_ptr, int error_action, - png_fixed_point red, png_fixed_point green) -{ - png_debug(1, "in png_set_rgb_to_gray_fixed"); - - /* Need the IHDR here because of the check on color_type below. */ - /* TODO: fix this */ - if (png_rtran_ok(png_ptr, 1) == 0) - return; - - switch (error_action) - { - case PNG_ERROR_ACTION_NONE: - png_ptr->transformations |= PNG_RGB_TO_GRAY; - break; - - case PNG_ERROR_ACTION_WARN: - png_ptr->transformations |= PNG_RGB_TO_GRAY_WARN; - break; - - case PNG_ERROR_ACTION_ERROR: - png_ptr->transformations |= PNG_RGB_TO_GRAY_ERR; - break; - - default: - png_error(png_ptr, "invalid error action to rgb_to_gray"); - } - - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) -#ifdef PNG_READ_EXPAND_SUPPORTED - png_ptr->transformations |= PNG_EXPAND; -#else - { - /* Make this an error in 1.6 because otherwise the application may assume - * that it just worked and get a memory overwrite. - */ - png_error(png_ptr, - "Cannot do RGB_TO_GRAY without EXPAND_SUPPORTED"); - - /* png_ptr->transformations &= ~PNG_RGB_TO_GRAY; */ - } -#endif - { - if (red >= 0 && green >= 0 && red + green <= PNG_FP_1) - { - png_uint_16 red_int, green_int; - - /* NOTE: this calculation does not round, but this behavior is retained - * for consistency; the inaccuracy is very small. The code here always - * overwrites the coefficients, regardless of whether they have been - * defaulted or set already. - */ - red_int = (png_uint_16)(((png_uint_32)red*32768)/100000); - green_int = (png_uint_16)(((png_uint_32)green*32768)/100000); - - png_ptr->rgb_to_gray_red_coeff = red_int; - png_ptr->rgb_to_gray_green_coeff = green_int; - png_ptr->rgb_to_gray_coefficients_set = 1; - } - - else - { - if (red >= 0 && green >= 0) - png_app_warning(png_ptr, - "ignoring out of range rgb_to_gray coefficients"); - - /* Use the defaults, from the cHRM chunk if set, else the historical - * values which are close to the sRGB/HDTV/ITU-Rec 709 values. See - * png_do_rgb_to_gray for more discussion of the values. In this case - * the coefficients are not marked as 'set' and are not overwritten if - * something has already provided a default. - */ - if (png_ptr->rgb_to_gray_red_coeff == 0 && - png_ptr->rgb_to_gray_green_coeff == 0) - { - png_ptr->rgb_to_gray_red_coeff = 6968; - png_ptr->rgb_to_gray_green_coeff = 23434; - /* png_ptr->rgb_to_gray_blue_coeff = 2366; */ - } - } - } -} - -#ifdef PNG_FLOATING_POINT_SUPPORTED -/* Convert a RGB image to a grayscale of the same width. This allows us, - * for example, to convert a 24 bpp RGB image into an 8 bpp grayscale image. - */ - -void PNGAPI -png_set_rgb_to_gray(png_structrp png_ptr, int error_action, double red, - double green) -{ - png_set_rgb_to_gray_fixed(png_ptr, error_action, - png_fixed(png_ptr, red, "rgb to gray red coefficient"), - png_fixed(png_ptr, green, "rgb to gray green coefficient")); -} -#endif /* FLOATING POINT */ - -#endif /* RGB_TO_GRAY */ - -#if defined(PNG_READ_USER_TRANSFORM_SUPPORTED) || \ - defined(PNG_WRITE_USER_TRANSFORM_SUPPORTED) -void PNGAPI -png_set_read_user_transform_fn(png_structrp png_ptr, png_user_transform_ptr - read_user_transform_fn) -{ - png_debug(1, "in png_set_read_user_transform_fn"); - -#ifdef PNG_READ_USER_TRANSFORM_SUPPORTED - png_ptr->transformations |= PNG_USER_TRANSFORM; - png_ptr->read_user_transform_fn = read_user_transform_fn; -#endif -} -#endif - -#ifdef PNG_READ_TRANSFORMS_SUPPORTED -#ifdef PNG_READ_GAMMA_SUPPORTED -/* In the case of gamma transformations only do transformations on images where - * the [file] gamma and screen_gamma are not close reciprocals, otherwise it - * slows things down slightly, and also needlessly introduces small errors. - */ -static int /* PRIVATE */ -png_gamma_threshold(png_fixed_point screen_gamma, png_fixed_point file_gamma) -{ - /* PNG_GAMMA_THRESHOLD is the threshold for performing gamma - * correction as a difference of the overall transform from 1.0 - * - * We want to compare the threshold with s*f - 1, if we get - * overflow here it is because of wacky gamma values so we - * turn on processing anyway. - */ - png_fixed_point gtest; - return !png_muldiv(>est, screen_gamma, file_gamma, PNG_FP_1) || - png_gamma_significant(gtest); -} -#endif - -/* Initialize everything needed for the read. This includes modifying - * the palette. - */ - -/* For the moment 'png_init_palette_transformations' and - * 'png_init_rgb_transformations' only do some flag canceling optimizations. - * The intent is that these two routines should have palette or rgb operations - * extracted from 'png_init_read_transformations'. - */ -static void /* PRIVATE */ -png_init_palette_transformations(png_structrp png_ptr) -{ - /* Called to handle the (input) palette case. In png_do_read_transformations - * the first step is to expand the palette if requested, so this code must - * take care to only make changes that are invariant with respect to the - * palette expansion, or only do them if there is no expansion. - * - * STRIP_ALPHA has already been handled in the caller (by setting num_trans - * to 0.) - */ - int input_has_alpha = 0; - int input_has_transparency = 0; - - if (png_ptr->num_trans > 0) - { - int i; - - /* Ignore if all the entries are opaque (unlikely!) */ - for (i=0; inum_trans; ++i) - { - if (png_ptr->trans_alpha[i] == 255) - continue; - else if (png_ptr->trans_alpha[i] == 0) - input_has_transparency = 1; - else - { - input_has_transparency = 1; - input_has_alpha = 1; - break; - } - } - } - - /* If no alpha we can optimize. */ - if (input_has_alpha == 0) - { - /* Any alpha means background and associative alpha processing is - * required, however if the alpha is 0 or 1 throughout OPTIMIZE_ALPHA - * and ENCODE_ALPHA are irrelevant. - */ - png_ptr->transformations &= ~PNG_ENCODE_ALPHA; - png_ptr->flags &= ~PNG_FLAG_OPTIMIZE_ALPHA; - - if (input_has_transparency == 0) - png_ptr->transformations &= ~(PNG_COMPOSE | PNG_BACKGROUND_EXPAND); - } - -#if defined(PNG_READ_EXPAND_SUPPORTED) && defined(PNG_READ_BACKGROUND_SUPPORTED) - /* png_set_background handling - deals with the complexity of whether the - * background color is in the file format or the screen format in the case - * where an 'expand' will happen. - */ - - /* The following code cannot be entered in the alpha pre-multiplication case - * because PNG_BACKGROUND_EXPAND is cancelled below. - */ - if ((png_ptr->transformations & PNG_BACKGROUND_EXPAND) != 0 && - (png_ptr->transformations & PNG_EXPAND) != 0) - { - { - png_ptr->background.red = - png_ptr->palette[png_ptr->background.index].red; - png_ptr->background.green = - png_ptr->palette[png_ptr->background.index].green; - png_ptr->background.blue = - png_ptr->palette[png_ptr->background.index].blue; - -#ifdef PNG_READ_INVERT_ALPHA_SUPPORTED - if ((png_ptr->transformations & PNG_INVERT_ALPHA) != 0) - { - if ((png_ptr->transformations & PNG_EXPAND_tRNS) == 0) - { - /* Invert the alpha channel (in tRNS) unless the pixels are - * going to be expanded, in which case leave it for later - */ - int i, istop = png_ptr->num_trans; - - for (i = 0; i < istop; i++) - png_ptr->trans_alpha[i] = - (png_byte)(255 - png_ptr->trans_alpha[i]); - } - } -#endif /* READ_INVERT_ALPHA */ - } - } /* background expand and (therefore) no alpha association. */ -#endif /* READ_EXPAND && READ_BACKGROUND */ -} - -static void /* PRIVATE */ -png_init_rgb_transformations(png_structrp png_ptr) -{ - /* Added to libpng-1.5.4: check the color type to determine whether there - * is any alpha or transparency in the image and simply cancel the - * background and alpha mode stuff if there isn't. - */ - int input_has_alpha = (png_ptr->color_type & PNG_COLOR_MASK_ALPHA) != 0; - int input_has_transparency = png_ptr->num_trans > 0; - - /* If no alpha we can optimize. */ - if (input_has_alpha == 0) - { - /* Any alpha means background and associative alpha processing is - * required, however if the alpha is 0 or 1 throughout OPTIMIZE_ALPHA - * and ENCODE_ALPHA are irrelevant. - */ -# ifdef PNG_READ_ALPHA_MODE_SUPPORTED - png_ptr->transformations &= ~PNG_ENCODE_ALPHA; - png_ptr->flags &= ~PNG_FLAG_OPTIMIZE_ALPHA; -# endif - - if (input_has_transparency == 0) - png_ptr->transformations &= ~(PNG_COMPOSE | PNG_BACKGROUND_EXPAND); - } - -#if defined(PNG_READ_EXPAND_SUPPORTED) && defined(PNG_READ_BACKGROUND_SUPPORTED) - /* png_set_background handling - deals with the complexity of whether the - * background color is in the file format or the screen format in the case - * where an 'expand' will happen. - */ - - /* The following code cannot be entered in the alpha pre-multiplication case - * because PNG_BACKGROUND_EXPAND is cancelled below. - */ - if ((png_ptr->transformations & PNG_BACKGROUND_EXPAND) != 0 && - (png_ptr->transformations & PNG_EXPAND) != 0 && - (png_ptr->color_type & PNG_COLOR_MASK_COLOR) == 0) - /* i.e., GRAY or GRAY_ALPHA */ - { - { - /* Expand background and tRNS chunks */ - int gray = png_ptr->background.gray; - int trans_gray = png_ptr->trans_color.gray; - - switch (png_ptr->bit_depth) - { - case 1: - gray *= 0xff; - trans_gray *= 0xff; - break; - - case 2: - gray *= 0x55; - trans_gray *= 0x55; - break; - - case 4: - gray *= 0x11; - trans_gray *= 0x11; - break; - - default: - - case 8: - /* FALLTHROUGH */ /* (Already 8 bits) */ - - case 16: - /* Already a full 16 bits */ - break; - } - - png_ptr->background.red = png_ptr->background.green = - png_ptr->background.blue = (png_uint_16)gray; - - if ((png_ptr->transformations & PNG_EXPAND_tRNS) == 0) - { - png_ptr->trans_color.red = png_ptr->trans_color.green = - png_ptr->trans_color.blue = (png_uint_16)trans_gray; - } - } - } /* background expand and (therefore) no alpha association. */ -#endif /* READ_EXPAND && READ_BACKGROUND */ -} - -void /* PRIVATE */ -png_init_read_transformations(png_structrp png_ptr) -{ - png_debug(1, "in png_init_read_transformations"); - - /* This internal function is called from png_read_start_row in pngrutil.c - * and it is called before the 'rowbytes' calculation is done, so the code - * in here can change or update the transformations flags. - * - * First do updates that do not depend on the details of the PNG image data - * being processed. - */ - -#ifdef PNG_READ_GAMMA_SUPPORTED - /* Prior to 1.5.4 these tests were performed from png_set_gamma, 1.5.4 adds - * png_set_alpha_mode and this is another source for a default file gamma so - * the test needs to be performed later - here. In addition prior to 1.5.4 - * the tests were repeated for the PALETTE color type here - this is no - * longer necessary (and doesn't seem to have been necessary before.) - */ - { - /* The following temporary indicates if overall gamma correction is - * required. - */ - int gamma_correction = 0; - - if (png_ptr->colorspace.gamma != 0) /* has been set */ - { - if (png_ptr->screen_gamma != 0) /* screen set too */ - gamma_correction = png_gamma_threshold(png_ptr->colorspace.gamma, - png_ptr->screen_gamma); - - else - /* Assume the output matches the input; a long time default behavior - * of libpng, although the standard has nothing to say about this. - */ - png_ptr->screen_gamma = png_reciprocal(png_ptr->colorspace.gamma); - } - - else if (png_ptr->screen_gamma != 0) - /* The converse - assume the file matches the screen, note that this - * perhaps undesirable default can (from 1.5.4) be changed by calling - * png_set_alpha_mode (even if the alpha handling mode isn't required - * or isn't changed from the default.) - */ - png_ptr->colorspace.gamma = png_reciprocal(png_ptr->screen_gamma); - - else /* neither are set */ - /* Just in case the following prevents any processing - file and screen - * are both assumed to be linear and there is no way to introduce a - * third gamma value other than png_set_background with 'UNIQUE', and, - * prior to 1.5.4 - */ - png_ptr->screen_gamma = png_ptr->colorspace.gamma = PNG_FP_1; - - /* We have a gamma value now. */ - png_ptr->colorspace.flags |= PNG_COLORSPACE_HAVE_GAMMA; - - /* Now turn the gamma transformation on or off as appropriate. Notice - * that PNG_GAMMA just refers to the file->screen correction. Alpha - * composition may independently cause gamma correction because it needs - * linear data (e.g. if the file has a gAMA chunk but the screen gamma - * hasn't been specified.) In any case this flag may get turned off in - * the code immediately below if the transform can be handled outside the - * row loop. - */ - if (gamma_correction != 0) - png_ptr->transformations |= PNG_GAMMA; - - else - png_ptr->transformations &= ~PNG_GAMMA; - } -#endif - - /* Certain transformations have the effect of preventing other - * transformations that happen afterward in png_do_read_transformations; - * resolve the interdependencies here. From the code of - * png_do_read_transformations the order is: - * - * 1) PNG_EXPAND (including PNG_EXPAND_tRNS) - * 2) PNG_STRIP_ALPHA (if no compose) - * 3) PNG_RGB_TO_GRAY - * 4) PNG_GRAY_TO_RGB iff !PNG_BACKGROUND_IS_GRAY - * 5) PNG_COMPOSE - * 6) PNG_GAMMA - * 7) PNG_STRIP_ALPHA (if compose) - * 8) PNG_ENCODE_ALPHA - * 9) PNG_SCALE_16_TO_8 - * 10) PNG_16_TO_8 - * 11) PNG_QUANTIZE (converts to palette) - * 12) PNG_EXPAND_16 - * 13) PNG_GRAY_TO_RGB iff PNG_BACKGROUND_IS_GRAY - * 14) PNG_INVERT_MONO - * 15) PNG_INVERT_ALPHA - * 16) PNG_SHIFT - * 17) PNG_PACK - * 18) PNG_BGR - * 19) PNG_PACKSWAP - * 20) PNG_FILLER (includes PNG_ADD_ALPHA) - * 21) PNG_SWAP_ALPHA - * 22) PNG_SWAP_BYTES - * 23) PNG_USER_TRANSFORM [must be last] - */ -#ifdef PNG_READ_STRIP_ALPHA_SUPPORTED - if ((png_ptr->transformations & PNG_STRIP_ALPHA) != 0 && - (png_ptr->transformations & PNG_COMPOSE) == 0) - { - /* Stripping the alpha channel happens immediately after the 'expand' - * transformations, before all other transformation, so it cancels out - * the alpha handling. It has the side effect negating the effect of - * PNG_EXPAND_tRNS too: - */ - png_ptr->transformations &= ~(PNG_BACKGROUND_EXPAND | PNG_ENCODE_ALPHA | - PNG_EXPAND_tRNS); - png_ptr->flags &= ~PNG_FLAG_OPTIMIZE_ALPHA; - - /* Kill the tRNS chunk itself too. Prior to 1.5.4 this did not happen - * so transparency information would remain just so long as it wasn't - * expanded. This produces unexpected API changes if the set of things - * that do PNG_EXPAND_tRNS changes (perfectly possible given the - * documentation - which says ask for what you want, accept what you - * get.) This makes the behavior consistent from 1.5.4: - */ - png_ptr->num_trans = 0; - } -#endif /* STRIP_ALPHA supported, no COMPOSE */ - -#ifdef PNG_READ_ALPHA_MODE_SUPPORTED - /* If the screen gamma is about 1.0 then the OPTIMIZE_ALPHA and ENCODE_ALPHA - * settings will have no effect. - */ - if (png_gamma_significant(png_ptr->screen_gamma) == 0) - { - png_ptr->transformations &= ~PNG_ENCODE_ALPHA; - png_ptr->flags &= ~PNG_FLAG_OPTIMIZE_ALPHA; - } -#endif - -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED - /* Make sure the coefficients for the rgb to gray conversion are set - * appropriately. - */ - if ((png_ptr->transformations & PNG_RGB_TO_GRAY) != 0) - png_colorspace_set_rgb_coefficients(png_ptr); -#endif - -#ifdef PNG_READ_GRAY_TO_RGB_SUPPORTED -#if defined(PNG_READ_EXPAND_SUPPORTED) && defined(PNG_READ_BACKGROUND_SUPPORTED) - /* Detect gray background and attempt to enable optimization for - * gray --> RGB case. - * - * Note: if PNG_BACKGROUND_EXPAND is set and color_type is either RGB or - * RGB_ALPHA (in which case need_expand is superfluous anyway), the - * background color might actually be gray yet not be flagged as such. - * This is not a problem for the current code, which uses - * PNG_BACKGROUND_IS_GRAY only to decide when to do the - * png_do_gray_to_rgb() transformation. - * - * TODO: this code needs to be revised to avoid the complexity and - * interdependencies. The color type of the background should be recorded in - * png_set_background, along with the bit depth, then the code has a record - * of exactly what color space the background is currently in. - */ - if ((png_ptr->transformations & PNG_BACKGROUND_EXPAND) != 0) - { - /* PNG_BACKGROUND_EXPAND: the background is in the file color space, so if - * the file was grayscale the background value is gray. - */ - if ((png_ptr->color_type & PNG_COLOR_MASK_COLOR) == 0) - png_ptr->mode |= PNG_BACKGROUND_IS_GRAY; - } - - else if ((png_ptr->transformations & PNG_COMPOSE) != 0) - { - /* PNG_COMPOSE: png_set_background was called with need_expand false, - * so the color is in the color space of the output or png_set_alpha_mode - * was called and the color is black. Ignore RGB_TO_GRAY because that - * happens before GRAY_TO_RGB. - */ - if ((png_ptr->transformations & PNG_GRAY_TO_RGB) != 0) - { - if (png_ptr->background.red == png_ptr->background.green && - png_ptr->background.red == png_ptr->background.blue) - { - png_ptr->mode |= PNG_BACKGROUND_IS_GRAY; - png_ptr->background.gray = png_ptr->background.red; - } - } - } -#endif /* READ_EXPAND && READ_BACKGROUND */ -#endif /* READ_GRAY_TO_RGB */ - - /* For indexed PNG data (PNG_COLOR_TYPE_PALETTE) many of the transformations - * can be performed directly on the palette, and some (such as rgb to gray) - * can be optimized inside the palette. This is particularly true of the - * composite (background and alpha) stuff, which can be pretty much all done - * in the palette even if the result is expanded to RGB or gray afterward. - * - * NOTE: this is Not Yet Implemented, the code behaves as in 1.5.1 and - * earlier and the palette stuff is actually handled on the first row. This - * leads to the reported bug that the palette returned by png_get_PLTE is not - * updated. - */ - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - png_init_palette_transformations(png_ptr); - - else - png_init_rgb_transformations(png_ptr); - -#if defined(PNG_READ_BACKGROUND_SUPPORTED) && \ - defined(PNG_READ_EXPAND_16_SUPPORTED) - if ((png_ptr->transformations & PNG_EXPAND_16) != 0 && - (png_ptr->transformations & PNG_COMPOSE) != 0 && - (png_ptr->transformations & PNG_BACKGROUND_EXPAND) == 0 && - png_ptr->bit_depth != 16) - { - /* TODO: fix this. Because the expand_16 operation is after the compose - * handling the background color must be 8, not 16, bits deep, but the - * application will supply a 16-bit value so reduce it here. - * - * The PNG_BACKGROUND_EXPAND code above does not expand to 16 bits at - * present, so that case is ok (until do_expand_16 is moved.) - * - * NOTE: this discards the low 16 bits of the user supplied background - * color, but until expand_16 works properly there is no choice! - */ -# define CHOP(x) (x)=((png_uint_16)PNG_DIV257(x)) - CHOP(png_ptr->background.red); - CHOP(png_ptr->background.green); - CHOP(png_ptr->background.blue); - CHOP(png_ptr->background.gray); -# undef CHOP - } -#endif /* READ_BACKGROUND && READ_EXPAND_16 */ - -#if defined(PNG_READ_BACKGROUND_SUPPORTED) && \ - (defined(PNG_READ_SCALE_16_TO_8_SUPPORTED) || \ - defined(PNG_READ_STRIP_16_TO_8_SUPPORTED)) - if ((png_ptr->transformations & (PNG_16_TO_8|PNG_SCALE_16_TO_8)) != 0 && - (png_ptr->transformations & PNG_COMPOSE) != 0 && - (png_ptr->transformations & PNG_BACKGROUND_EXPAND) == 0 && - png_ptr->bit_depth == 16) - { - /* On the other hand, if a 16-bit file is to be reduced to 8-bits per - * component this will also happen after PNG_COMPOSE and so the background - * color must be pre-expanded here. - * - * TODO: fix this too. - */ - png_ptr->background.red = (png_uint_16)(png_ptr->background.red * 257); - png_ptr->background.green = - (png_uint_16)(png_ptr->background.green * 257); - png_ptr->background.blue = (png_uint_16)(png_ptr->background.blue * 257); - png_ptr->background.gray = (png_uint_16)(png_ptr->background.gray * 257); - } -#endif - - /* NOTE: below 'PNG_READ_ALPHA_MODE_SUPPORTED' is presumed to also enable the - * background support (see the comments in scripts/pnglibconf.dfa), this - * allows pre-multiplication of the alpha channel to be implemented as - * compositing on black. This is probably sub-optimal and has been done in - * 1.5.4 betas simply to enable external critique and testing (i.e. to - * implement the new API quickly, without lots of internal changes.) - */ - -#ifdef PNG_READ_GAMMA_SUPPORTED -# ifdef PNG_READ_BACKGROUND_SUPPORTED - /* Includes ALPHA_MODE */ - png_ptr->background_1 = png_ptr->background; -# endif - - /* This needs to change - in the palette image case a whole set of tables are - * built when it would be quicker to just calculate the correct value for - * each palette entry directly. Also, the test is too tricky - why check - * PNG_RGB_TO_GRAY if PNG_GAMMA is not set? The answer seems to be that - * PNG_GAMMA is cancelled even if the gamma is known? The test excludes the - * PNG_COMPOSE case, so apparently if there is no *overall* gamma correction - * the gamma tables will not be built even if composition is required on a - * gamma encoded value. - * - * In 1.5.4 this is addressed below by an additional check on the individual - * file gamma - if it is not 1.0 both RGB_TO_GRAY and COMPOSE need the - * tables. - */ - if ((png_ptr->transformations & PNG_GAMMA) != 0 || - ((png_ptr->transformations & PNG_RGB_TO_GRAY) != 0 && - (png_gamma_significant(png_ptr->colorspace.gamma) != 0 || - png_gamma_significant(png_ptr->screen_gamma) != 0)) || - ((png_ptr->transformations & PNG_COMPOSE) != 0 && - (png_gamma_significant(png_ptr->colorspace.gamma) != 0 || - png_gamma_significant(png_ptr->screen_gamma) != 0 -# ifdef PNG_READ_BACKGROUND_SUPPORTED - || (png_ptr->background_gamma_type == PNG_BACKGROUND_GAMMA_UNIQUE && - png_gamma_significant(png_ptr->background_gamma) != 0) -# endif - )) || ((png_ptr->transformations & PNG_ENCODE_ALPHA) != 0 && - png_gamma_significant(png_ptr->screen_gamma) != 0)) - { - png_build_gamma_table(png_ptr, png_ptr->bit_depth); - -#ifdef PNG_READ_BACKGROUND_SUPPORTED - if ((png_ptr->transformations & PNG_COMPOSE) != 0) - { - /* Issue a warning about this combination: because RGB_TO_GRAY is - * optimized to do the gamma transform if present yet do_background has - * to do the same thing if both options are set a - * double-gamma-correction happens. This is true in all versions of - * libpng to date. - */ - if ((png_ptr->transformations & PNG_RGB_TO_GRAY) != 0) - png_warning(png_ptr, - "libpng does not support gamma+background+rgb_to_gray"); - - if ((png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) != 0) - { - /* We don't get to here unless there is a tRNS chunk with non-opaque - * entries - see the checking code at the start of this function. - */ - png_color back, back_1; - png_colorp palette = png_ptr->palette; - int num_palette = png_ptr->num_palette; - int i; - if (png_ptr->background_gamma_type == PNG_BACKGROUND_GAMMA_FILE) - { - - back.red = png_ptr->gamma_table[png_ptr->background.red]; - back.green = png_ptr->gamma_table[png_ptr->background.green]; - back.blue = png_ptr->gamma_table[png_ptr->background.blue]; - - back_1.red = png_ptr->gamma_to_1[png_ptr->background.red]; - back_1.green = png_ptr->gamma_to_1[png_ptr->background.green]; - back_1.blue = png_ptr->gamma_to_1[png_ptr->background.blue]; - } - else - { - png_fixed_point g, gs; - - switch (png_ptr->background_gamma_type) - { - case PNG_BACKGROUND_GAMMA_SCREEN: - g = (png_ptr->screen_gamma); - gs = PNG_FP_1; - break; - - case PNG_BACKGROUND_GAMMA_FILE: - g = png_reciprocal(png_ptr->colorspace.gamma); - gs = png_reciprocal2(png_ptr->colorspace.gamma, - png_ptr->screen_gamma); - break; - - case PNG_BACKGROUND_GAMMA_UNIQUE: - g = png_reciprocal(png_ptr->background_gamma); - gs = png_reciprocal2(png_ptr->background_gamma, - png_ptr->screen_gamma); - break; - default: - g = PNG_FP_1; /* back_1 */ - gs = PNG_FP_1; /* back */ - break; - } - - if (png_gamma_significant(gs) != 0) - { - back.red = png_gamma_8bit_correct(png_ptr->background.red, - gs); - back.green = png_gamma_8bit_correct(png_ptr->background.green, - gs); - back.blue = png_gamma_8bit_correct(png_ptr->background.blue, - gs); - } - - else - { - back.red = (png_byte)png_ptr->background.red; - back.green = (png_byte)png_ptr->background.green; - back.blue = (png_byte)png_ptr->background.blue; - } - - if (png_gamma_significant(g) != 0) - { - back_1.red = png_gamma_8bit_correct(png_ptr->background.red, - g); - back_1.green = png_gamma_8bit_correct( - png_ptr->background.green, g); - back_1.blue = png_gamma_8bit_correct(png_ptr->background.blue, - g); - } - - else - { - back_1.red = (png_byte)png_ptr->background.red; - back_1.green = (png_byte)png_ptr->background.green; - back_1.blue = (png_byte)png_ptr->background.blue; - } - } - - for (i = 0; i < num_palette; i++) - { - if (i < (int)png_ptr->num_trans && - png_ptr->trans_alpha[i] != 0xff) - { - if (png_ptr->trans_alpha[i] == 0) - { - palette[i] = back; - } - else /* if (png_ptr->trans_alpha[i] != 0xff) */ - { - png_byte v, w; - - v = png_ptr->gamma_to_1[palette[i].red]; - png_composite(w, v, png_ptr->trans_alpha[i], back_1.red); - palette[i].red = png_ptr->gamma_from_1[w]; - - v = png_ptr->gamma_to_1[palette[i].green]; - png_composite(w, v, png_ptr->trans_alpha[i], back_1.green); - palette[i].green = png_ptr->gamma_from_1[w]; - - v = png_ptr->gamma_to_1[palette[i].blue]; - png_composite(w, v, png_ptr->trans_alpha[i], back_1.blue); - palette[i].blue = png_ptr->gamma_from_1[w]; - } - } - else - { - palette[i].red = png_ptr->gamma_table[palette[i].red]; - palette[i].green = png_ptr->gamma_table[palette[i].green]; - palette[i].blue = png_ptr->gamma_table[palette[i].blue]; - } - } - - /* Prevent the transformations being done again. - * - * NOTE: this is highly dubious; it removes the transformations in - * place. This seems inconsistent with the general treatment of the - * transformations elsewhere. - */ - png_ptr->transformations &= ~(PNG_COMPOSE | PNG_GAMMA); - } /* color_type == PNG_COLOR_TYPE_PALETTE */ - - /* if (png_ptr->background_gamma_type!=PNG_BACKGROUND_GAMMA_UNKNOWN) */ - else /* color_type != PNG_COLOR_TYPE_PALETTE */ - { - int gs_sig, g_sig; - png_fixed_point g = PNG_FP_1; /* Correction to linear */ - png_fixed_point gs = PNG_FP_1; /* Correction to screen */ - - switch (png_ptr->background_gamma_type) - { - case PNG_BACKGROUND_GAMMA_SCREEN: - g = png_ptr->screen_gamma; - /* gs = PNG_FP_1; */ - break; - - case PNG_BACKGROUND_GAMMA_FILE: - g = png_reciprocal(png_ptr->colorspace.gamma); - gs = png_reciprocal2(png_ptr->colorspace.gamma, - png_ptr->screen_gamma); - break; - - case PNG_BACKGROUND_GAMMA_UNIQUE: - g = png_reciprocal(png_ptr->background_gamma); - gs = png_reciprocal2(png_ptr->background_gamma, - png_ptr->screen_gamma); - break; - - default: - png_error(png_ptr, "invalid background gamma type"); - } - - g_sig = png_gamma_significant(g); - gs_sig = png_gamma_significant(gs); - - if (g_sig != 0) - png_ptr->background_1.gray = png_gamma_correct(png_ptr, - png_ptr->background.gray, g); - - if (gs_sig != 0) - png_ptr->background.gray = png_gamma_correct(png_ptr, - png_ptr->background.gray, gs); - - if ((png_ptr->background.red != png_ptr->background.green) || - (png_ptr->background.red != png_ptr->background.blue) || - (png_ptr->background.red != png_ptr->background.gray)) - { - /* RGB or RGBA with color background */ - if (g_sig != 0) - { - png_ptr->background_1.red = png_gamma_correct(png_ptr, - png_ptr->background.red, g); - - png_ptr->background_1.green = png_gamma_correct(png_ptr, - png_ptr->background.green, g); - - png_ptr->background_1.blue = png_gamma_correct(png_ptr, - png_ptr->background.blue, g); - } - - if (gs_sig != 0) - { - png_ptr->background.red = png_gamma_correct(png_ptr, - png_ptr->background.red, gs); - - png_ptr->background.green = png_gamma_correct(png_ptr, - png_ptr->background.green, gs); - - png_ptr->background.blue = png_gamma_correct(png_ptr, - png_ptr->background.blue, gs); - } - } - - else - { - /* GRAY, GRAY ALPHA, RGB, or RGBA with gray background */ - png_ptr->background_1.red = png_ptr->background_1.green - = png_ptr->background_1.blue = png_ptr->background_1.gray; - - png_ptr->background.red = png_ptr->background.green - = png_ptr->background.blue = png_ptr->background.gray; - } - - /* The background is now in screen gamma: */ - png_ptr->background_gamma_type = PNG_BACKGROUND_GAMMA_SCREEN; - } /* color_type != PNG_COLOR_TYPE_PALETTE */ - }/* png_ptr->transformations & PNG_BACKGROUND */ - - else - /* Transformation does not include PNG_BACKGROUND */ -#endif /* READ_BACKGROUND */ - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED - /* RGB_TO_GRAY needs to have non-gamma-corrected values! */ - && ((png_ptr->transformations & PNG_EXPAND) == 0 || - (png_ptr->transformations & PNG_RGB_TO_GRAY) == 0) -#endif - ) - { - png_colorp palette = png_ptr->palette; - int num_palette = png_ptr->num_palette; - int i; - - /* NOTE: there are other transformations that should probably be in - * here too. - */ - for (i = 0; i < num_palette; i++) - { - palette[i].red = png_ptr->gamma_table[palette[i].red]; - palette[i].green = png_ptr->gamma_table[palette[i].green]; - palette[i].blue = png_ptr->gamma_table[palette[i].blue]; - } - - /* Done the gamma correction. */ - png_ptr->transformations &= ~PNG_GAMMA; - } /* color_type == PALETTE && !PNG_BACKGROUND transformation */ - } -#ifdef PNG_READ_BACKGROUND_SUPPORTED - else -#endif -#endif /* READ_GAMMA */ - -#ifdef PNG_READ_BACKGROUND_SUPPORTED - /* No GAMMA transformation (see the hanging else 4 lines above) */ - if ((png_ptr->transformations & PNG_COMPOSE) != 0 && - (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE)) - { - int i; - int istop = (int)png_ptr->num_trans; - png_color back; - png_colorp palette = png_ptr->palette; - - back.red = (png_byte)png_ptr->background.red; - back.green = (png_byte)png_ptr->background.green; - back.blue = (png_byte)png_ptr->background.blue; - - for (i = 0; i < istop; i++) - { - if (png_ptr->trans_alpha[i] == 0) - { - palette[i] = back; - } - - else if (png_ptr->trans_alpha[i] != 0xff) - { - /* The png_composite() macro is defined in png.h */ - png_composite(palette[i].red, palette[i].red, - png_ptr->trans_alpha[i], back.red); - - png_composite(palette[i].green, palette[i].green, - png_ptr->trans_alpha[i], back.green); - - png_composite(palette[i].blue, palette[i].blue, - png_ptr->trans_alpha[i], back.blue); - } - } - - png_ptr->transformations &= ~PNG_COMPOSE; - } -#endif /* READ_BACKGROUND */ - -#ifdef PNG_READ_SHIFT_SUPPORTED - if ((png_ptr->transformations & PNG_SHIFT) != 0 && - (png_ptr->transformations & PNG_EXPAND) == 0 && - (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE)) - { - int i; - int istop = png_ptr->num_palette; - int shift = 8 - png_ptr->sig_bit.red; - - png_ptr->transformations &= ~PNG_SHIFT; - - /* significant bits can be in the range 1 to 7 for a meaningful result, if - * the number of significant bits is 0 then no shift is done (this is an - * error condition which is silently ignored.) - */ - if (shift > 0 && shift < 8) - for (i=0; ipalette[i].red; - - component >>= shift; - png_ptr->palette[i].red = (png_byte)component; - } - - shift = 8 - png_ptr->sig_bit.green; - if (shift > 0 && shift < 8) - for (i=0; ipalette[i].green; - - component >>= shift; - png_ptr->palette[i].green = (png_byte)component; - } - - shift = 8 - png_ptr->sig_bit.blue; - if (shift > 0 && shift < 8) - for (i=0; ipalette[i].blue; - - component >>= shift; - png_ptr->palette[i].blue = (png_byte)component; - } - } -#endif /* READ_SHIFT */ -} - -/* Modify the info structure to reflect the transformations. The - * info should be updated so a PNG file could be written with it, - * assuming the transformations result in valid PNG data. - */ -void /* PRIVATE */ -png_read_transform_info(png_structrp png_ptr, png_inforp info_ptr) -{ - png_debug(1, "in png_read_transform_info"); - -#ifdef PNG_READ_EXPAND_SUPPORTED - if ((png_ptr->transformations & PNG_EXPAND) != 0) - { - if (info_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - { - /* This check must match what actually happens in - * png_do_expand_palette; if it ever checks the tRNS chunk to see if - * it is all opaque we must do the same (at present it does not.) - */ - if (png_ptr->num_trans > 0) - info_ptr->color_type = PNG_COLOR_TYPE_RGB_ALPHA; - - else - info_ptr->color_type = PNG_COLOR_TYPE_RGB; - - info_ptr->bit_depth = 8; - info_ptr->num_trans = 0; - - if (png_ptr->palette == NULL) - png_error (png_ptr, "Palette is NULL in indexed image"); - } - else - { - if (png_ptr->num_trans != 0) - { - if ((png_ptr->transformations & PNG_EXPAND_tRNS) != 0) - info_ptr->color_type |= PNG_COLOR_MASK_ALPHA; - } - if (info_ptr->bit_depth < 8) - info_ptr->bit_depth = 8; - - info_ptr->num_trans = 0; - } - } -#endif - -#if defined(PNG_READ_BACKGROUND_SUPPORTED) ||\ - defined(PNG_READ_ALPHA_MODE_SUPPORTED) - /* The following is almost certainly wrong unless the background value is in - * the screen space! - */ - if ((png_ptr->transformations & PNG_COMPOSE) != 0) - info_ptr->background = png_ptr->background; -#endif - -#ifdef PNG_READ_GAMMA_SUPPORTED - /* The following used to be conditional on PNG_GAMMA (prior to 1.5.4), - * however it seems that the code in png_init_read_transformations, which has - * been called before this from png_read_update_info->png_read_start_row - * sometimes does the gamma transform and cancels the flag. - * - * TODO: this looks wrong; the info_ptr should end up with a gamma equal to - * the screen_gamma value. The following probably results in weirdness if - * the info_ptr is used by the app after the rows have been read. - */ - info_ptr->colorspace.gamma = png_ptr->colorspace.gamma; -#endif - - if (info_ptr->bit_depth == 16) - { -# ifdef PNG_READ_16BIT_SUPPORTED -# ifdef PNG_READ_SCALE_16_TO_8_SUPPORTED - if ((png_ptr->transformations & PNG_SCALE_16_TO_8) != 0) - info_ptr->bit_depth = 8; -# endif - -# ifdef PNG_READ_STRIP_16_TO_8_SUPPORTED - if ((png_ptr->transformations & PNG_16_TO_8) != 0) - info_ptr->bit_depth = 8; -# endif - -# else - /* No 16-bit support: force chopping 16-bit input down to 8, in this case - * the app program can chose if both APIs are available by setting the - * correct scaling to use. - */ -# ifdef PNG_READ_STRIP_16_TO_8_SUPPORTED - /* For compatibility with previous versions use the strip method by - * default. This code works because if PNG_SCALE_16_TO_8 is already - * set the code below will do that in preference to the chop. - */ - png_ptr->transformations |= PNG_16_TO_8; - info_ptr->bit_depth = 8; -# else - -# ifdef PNG_READ_SCALE_16_TO_8_SUPPORTED - png_ptr->transformations |= PNG_SCALE_16_TO_8; - info_ptr->bit_depth = 8; -# else - - CONFIGURATION ERROR: you must enable at least one 16 to 8 method -# endif -# endif -#endif /* !READ_16BIT */ - } - -#ifdef PNG_READ_GRAY_TO_RGB_SUPPORTED - if ((png_ptr->transformations & PNG_GRAY_TO_RGB) != 0) - info_ptr->color_type = (png_byte)(info_ptr->color_type | - PNG_COLOR_MASK_COLOR); -#endif - -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED - if ((png_ptr->transformations & PNG_RGB_TO_GRAY) != 0) - info_ptr->color_type = (png_byte)(info_ptr->color_type & - ~PNG_COLOR_MASK_COLOR); -#endif - -#ifdef PNG_READ_QUANTIZE_SUPPORTED - if ((png_ptr->transformations & PNG_QUANTIZE) != 0) - { - if (((info_ptr->color_type == PNG_COLOR_TYPE_RGB) || - (info_ptr->color_type == PNG_COLOR_TYPE_RGB_ALPHA)) && - png_ptr->palette_lookup != 0 && info_ptr->bit_depth == 8) - { - info_ptr->color_type = PNG_COLOR_TYPE_PALETTE; - } - } -#endif - -#ifdef PNG_READ_EXPAND_16_SUPPORTED - if ((png_ptr->transformations & PNG_EXPAND_16) != 0 && - info_ptr->bit_depth == 8 && - info_ptr->color_type != PNG_COLOR_TYPE_PALETTE) - { - info_ptr->bit_depth = 16; - } -#endif - -#ifdef PNG_READ_PACK_SUPPORTED - if ((png_ptr->transformations & PNG_PACK) != 0 && - (info_ptr->bit_depth < 8)) - info_ptr->bit_depth = 8; -#endif - - if (info_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - info_ptr->channels = 1; - - else if ((info_ptr->color_type & PNG_COLOR_MASK_COLOR) != 0) - info_ptr->channels = 3; - - else - info_ptr->channels = 1; - -#ifdef PNG_READ_STRIP_ALPHA_SUPPORTED - if ((png_ptr->transformations & PNG_STRIP_ALPHA) != 0) - { - info_ptr->color_type = (png_byte)(info_ptr->color_type & - ~PNG_COLOR_MASK_ALPHA); - info_ptr->num_trans = 0; - } -#endif - - if ((info_ptr->color_type & PNG_COLOR_MASK_ALPHA) != 0) - info_ptr->channels++; - -#ifdef PNG_READ_FILLER_SUPPORTED - /* STRIP_ALPHA and FILLER allowed: MASK_ALPHA bit stripped above */ - if ((png_ptr->transformations & PNG_FILLER) != 0 && - (info_ptr->color_type == PNG_COLOR_TYPE_RGB || - info_ptr->color_type == PNG_COLOR_TYPE_GRAY)) - { - info_ptr->channels++; - /* If adding a true alpha channel not just filler */ - if ((png_ptr->transformations & PNG_ADD_ALPHA) != 0) - info_ptr->color_type |= PNG_COLOR_MASK_ALPHA; - } -#endif - -#if defined(PNG_USER_TRANSFORM_PTR_SUPPORTED) && \ -defined(PNG_READ_USER_TRANSFORM_SUPPORTED) - if ((png_ptr->transformations & PNG_USER_TRANSFORM) != 0) - { - if (png_ptr->user_transform_depth != 0) - info_ptr->bit_depth = png_ptr->user_transform_depth; - - if (png_ptr->user_transform_channels != 0) - info_ptr->channels = png_ptr->user_transform_channels; - } -#endif - - info_ptr->pixel_depth = (png_byte)(info_ptr->channels * - info_ptr->bit_depth); - - info_ptr->rowbytes = PNG_ROWBYTES(info_ptr->pixel_depth, info_ptr->width); - - /* Adding in 1.5.4: cache the above value in png_struct so that we can later - * check in png_rowbytes that the user buffer won't get overwritten. Note - * that the field is not always set - if png_read_update_info isn't called - * the application has to either not do any transforms or get the calculation - * right itself. - */ - png_ptr->info_rowbytes = info_ptr->rowbytes; - -#ifndef PNG_READ_EXPAND_SUPPORTED - if (png_ptr != NULL) - return; -#endif -} - -#ifdef PNG_READ_PACK_SUPPORTED -/* Unpack pixels of 1, 2, or 4 bits per pixel into 1 byte per pixel, - * without changing the actual values. Thus, if you had a row with - * a bit depth of 1, you would end up with bytes that only contained - * the numbers 0 or 1. If you would rather they contain 0 and 255, use - * png_do_shift() after this. - */ -static void -png_do_unpack(png_row_infop row_info, png_bytep row) -{ - png_debug(1, "in png_do_unpack"); - - if (row_info->bit_depth < 8) - { - png_uint_32 i; - png_uint_32 row_width=row_info->width; - - switch (row_info->bit_depth) - { - case 1: - { - png_bytep sp = row + (size_t)((row_width - 1) >> 3); - png_bytep dp = row + (size_t)row_width - 1; - png_uint_32 shift = 7U - ((row_width + 7U) & 0x07); - for (i = 0; i < row_width; i++) - { - *dp = (png_byte)((*sp >> shift) & 0x01); - - if (shift == 7) - { - shift = 0; - sp--; - } - - else - shift++; - - dp--; - } - break; - } - - case 2: - { - - png_bytep sp = row + (size_t)((row_width - 1) >> 2); - png_bytep dp = row + (size_t)row_width - 1; - png_uint_32 shift = ((3U - ((row_width + 3U) & 0x03)) << 1); - for (i = 0; i < row_width; i++) - { - *dp = (png_byte)((*sp >> shift) & 0x03); - - if (shift == 6) - { - shift = 0; - sp--; - } - - else - shift += 2; - - dp--; - } - break; - } - - case 4: - { - png_bytep sp = row + (size_t)((row_width - 1) >> 1); - png_bytep dp = row + (size_t)row_width - 1; - png_uint_32 shift = ((1U - ((row_width + 1U) & 0x01)) << 2); - for (i = 0; i < row_width; i++) - { - *dp = (png_byte)((*sp >> shift) & 0x0f); - - if (shift == 4) - { - shift = 0; - sp--; - } - - else - shift = 4; - - dp--; - } - break; - } - - default: - break; - } - row_info->bit_depth = 8; - row_info->pixel_depth = (png_byte)(8 * row_info->channels); - row_info->rowbytes = row_width * row_info->channels; - } -} -#endif - -#ifdef PNG_READ_SHIFT_SUPPORTED -/* Reverse the effects of png_do_shift. This routine merely shifts the - * pixels back to their significant bits values. Thus, if you have - * a row of bit depth 8, but only 5 are significant, this will shift - * the values back to 0 through 31. - */ -static void -png_do_unshift(png_row_infop row_info, png_bytep row, - png_const_color_8p sig_bits) -{ - int color_type; - - png_debug(1, "in png_do_unshift"); - - /* The palette case has already been handled in the _init routine. */ - color_type = row_info->color_type; - - if (color_type != PNG_COLOR_TYPE_PALETTE) - { - int shift[4]; - int channels = 0; - int bit_depth = row_info->bit_depth; - - if ((color_type & PNG_COLOR_MASK_COLOR) != 0) - { - shift[channels++] = bit_depth - sig_bits->red; - shift[channels++] = bit_depth - sig_bits->green; - shift[channels++] = bit_depth - sig_bits->blue; - } - - else - { - shift[channels++] = bit_depth - sig_bits->gray; - } - - if ((color_type & PNG_COLOR_MASK_ALPHA) != 0) - { - shift[channels++] = bit_depth - sig_bits->alpha; - } - - { - int c, have_shift; - - for (c = have_shift = 0; c < channels; ++c) - { - /* A shift of more than the bit depth is an error condition but it - * gets ignored here. - */ - if (shift[c] <= 0 || shift[c] >= bit_depth) - shift[c] = 0; - - else - have_shift = 1; - } - - if (have_shift == 0) - return; - } - - switch (bit_depth) - { - default: - /* Must be 1bpp gray: should not be here! */ - /* NOTREACHED */ - break; - - case 2: - /* Must be 2bpp gray */ - /* assert(channels == 1 && shift[0] == 1) */ - { - png_bytep bp = row; - png_bytep bp_end = bp + row_info->rowbytes; - - while (bp < bp_end) - { - int b = (*bp >> 1) & 0x55; - *bp++ = (png_byte)b; - } - break; - } - - case 4: - /* Must be 4bpp gray */ - /* assert(channels == 1) */ - { - png_bytep bp = row; - png_bytep bp_end = bp + row_info->rowbytes; - int gray_shift = shift[0]; - int mask = 0xf >> gray_shift; - - mask |= mask << 4; - - while (bp < bp_end) - { - int b = (*bp >> gray_shift) & mask; - *bp++ = (png_byte)b; - } - break; - } - - case 8: - /* Single byte components, G, GA, RGB, RGBA */ - { - png_bytep bp = row; - png_bytep bp_end = bp + row_info->rowbytes; - int channel = 0; - - while (bp < bp_end) - { - int b = *bp >> shift[channel]; - if (++channel >= channels) - channel = 0; - *bp++ = (png_byte)b; - } - break; - } - -#ifdef PNG_READ_16BIT_SUPPORTED - case 16: - /* Double byte components, G, GA, RGB, RGBA */ - { - png_bytep bp = row; - png_bytep bp_end = bp + row_info->rowbytes; - int channel = 0; - - while (bp < bp_end) - { - int value = (bp[0] << 8) + bp[1]; - - value >>= shift[channel]; - if (++channel >= channels) - channel = 0; - *bp++ = (png_byte)(value >> 8); - *bp++ = (png_byte)value; - } - break; - } -#endif - } - } -} -#endif - -#ifdef PNG_READ_SCALE_16_TO_8_SUPPORTED -/* Scale rows of bit depth 16 down to 8 accurately */ -static void -png_do_scale_16_to_8(png_row_infop row_info, png_bytep row) -{ - png_debug(1, "in png_do_scale_16_to_8"); - - if (row_info->bit_depth == 16) - { - png_bytep sp = row; /* source */ - png_bytep dp = row; /* destination */ - png_bytep ep = sp + row_info->rowbytes; /* end+1 */ - - while (sp < ep) - { - /* The input is an array of 16-bit components, these must be scaled to - * 8 bits each. For a 16-bit value V the required value (from the PNG - * specification) is: - * - * (V * 255) / 65535 - * - * This reduces to round(V / 257), or floor((V + 128.5)/257) - * - * Represent V as the two byte value vhi.vlo. Make a guess that the - * result is the top byte of V, vhi, then the correction to this value - * is: - * - * error = floor(((V-vhi.vhi) + 128.5) / 257) - * = floor(((vlo-vhi) + 128.5) / 257) - * - * This can be approximated using integer arithmetic (and a signed - * shift): - * - * error = (vlo-vhi+128) >> 8; - * - * The approximate differs from the exact answer only when (vlo-vhi) is - * 128; it then gives a correction of +1 when the exact correction is - * 0. This gives 128 errors. The exact answer (correct for all 16-bit - * input values) is: - * - * error = (vlo-vhi+128)*65535 >> 24; - * - * An alternative arithmetic calculation which also gives no errors is: - * - * (V * 255 + 32895) >> 16 - */ - - png_int_32 tmp = *sp++; /* must be signed! */ - tmp += (((int)*sp++ - tmp + 128) * 65535) >> 24; - *dp++ = (png_byte)tmp; - } - - row_info->bit_depth = 8; - row_info->pixel_depth = (png_byte)(8 * row_info->channels); - row_info->rowbytes = row_info->width * row_info->channels; - } -} -#endif - -#ifdef PNG_READ_STRIP_16_TO_8_SUPPORTED -static void -/* Simply discard the low byte. This was the default behavior prior - * to libpng-1.5.4. - */ -png_do_chop(png_row_infop row_info, png_bytep row) -{ - png_debug(1, "in png_do_chop"); - - if (row_info->bit_depth == 16) - { - png_bytep sp = row; /* source */ - png_bytep dp = row; /* destination */ - png_bytep ep = sp + row_info->rowbytes; /* end+1 */ - - while (sp < ep) - { - *dp++ = *sp; - sp += 2; /* skip low byte */ - } - - row_info->bit_depth = 8; - row_info->pixel_depth = (png_byte)(8 * row_info->channels); - row_info->rowbytes = row_info->width * row_info->channels; - } -} -#endif - -#ifdef PNG_READ_SWAP_ALPHA_SUPPORTED -static void -png_do_read_swap_alpha(png_row_infop row_info, png_bytep row) -{ - png_uint_32 row_width = row_info->width; - - png_debug(1, "in png_do_read_swap_alpha"); - - if (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - { - /* This converts from RGBA to ARGB */ - if (row_info->bit_depth == 8) - { - png_bytep sp = row + row_info->rowbytes; - png_bytep dp = sp; - png_byte save; - png_uint_32 i; - - for (i = 0; i < row_width; i++) - { - save = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = save; - } - } - -#ifdef PNG_READ_16BIT_SUPPORTED - /* This converts from RRGGBBAA to AARRGGBB */ - else - { - png_bytep sp = row + row_info->rowbytes; - png_bytep dp = sp; - png_byte save[2]; - png_uint_32 i; - - for (i = 0; i < row_width; i++) - { - save[0] = *(--sp); - save[1] = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = save[0]; - *(--dp) = save[1]; - } - } -#endif - } - - else if (row_info->color_type == PNG_COLOR_TYPE_GRAY_ALPHA) - { - /* This converts from GA to AG */ - if (row_info->bit_depth == 8) - { - png_bytep sp = row + row_info->rowbytes; - png_bytep dp = sp; - png_byte save; - png_uint_32 i; - - for (i = 0; i < row_width; i++) - { - save = *(--sp); - *(--dp) = *(--sp); - *(--dp) = save; - } - } - -#ifdef PNG_READ_16BIT_SUPPORTED - /* This converts from GGAA to AAGG */ - else - { - png_bytep sp = row + row_info->rowbytes; - png_bytep dp = sp; - png_byte save[2]; - png_uint_32 i; - - for (i = 0; i < row_width; i++) - { - save[0] = *(--sp); - save[1] = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = save[0]; - *(--dp) = save[1]; - } - } -#endif - } -} -#endif - -#ifdef PNG_READ_INVERT_ALPHA_SUPPORTED -static void -png_do_read_invert_alpha(png_row_infop row_info, png_bytep row) -{ - png_uint_32 row_width; - png_debug(1, "in png_do_read_invert_alpha"); - - row_width = row_info->width; - if (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - { - if (row_info->bit_depth == 8) - { - /* This inverts the alpha channel in RGBA */ - png_bytep sp = row + row_info->rowbytes; - png_bytep dp = sp; - png_uint_32 i; - - for (i = 0; i < row_width; i++) - { - *(--dp) = (png_byte)(255 - *(--sp)); - -/* This does nothing: - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - We can replace it with: -*/ - sp-=3; - dp=sp; - } - } - -#ifdef PNG_READ_16BIT_SUPPORTED - /* This inverts the alpha channel in RRGGBBAA */ - else - { - png_bytep sp = row + row_info->rowbytes; - png_bytep dp = sp; - png_uint_32 i; - - for (i = 0; i < row_width; i++) - { - *(--dp) = (png_byte)(255 - *(--sp)); - *(--dp) = (png_byte)(255 - *(--sp)); - -/* This does nothing: - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - We can replace it with: -*/ - sp-=6; - dp=sp; - } - } -#endif - } - else if (row_info->color_type == PNG_COLOR_TYPE_GRAY_ALPHA) - { - if (row_info->bit_depth == 8) - { - /* This inverts the alpha channel in GA */ - png_bytep sp = row + row_info->rowbytes; - png_bytep dp = sp; - png_uint_32 i; - - for (i = 0; i < row_width; i++) - { - *(--dp) = (png_byte)(255 - *(--sp)); - *(--dp) = *(--sp); - } - } - -#ifdef PNG_READ_16BIT_SUPPORTED - else - { - /* This inverts the alpha channel in GGAA */ - png_bytep sp = row + row_info->rowbytes; - png_bytep dp = sp; - png_uint_32 i; - - for (i = 0; i < row_width; i++) - { - *(--dp) = (png_byte)(255 - *(--sp)); - *(--dp) = (png_byte)(255 - *(--sp)); -/* - *(--dp) = *(--sp); - *(--dp) = *(--sp); -*/ - sp-=2; - dp=sp; - } - } -#endif - } -} -#endif - -#ifdef PNG_READ_FILLER_SUPPORTED -/* Add filler channel if we have RGB color */ -static void -png_do_read_filler(png_row_infop row_info, png_bytep row, - png_uint_32 filler, png_uint_32 flags) -{ - png_uint_32 i; - png_uint_32 row_width = row_info->width; - -#ifdef PNG_READ_16BIT_SUPPORTED - png_byte hi_filler = (png_byte)(filler>>8); -#endif - png_byte lo_filler = (png_byte)filler; - - png_debug(1, "in png_do_read_filler"); - - if ( - row_info->color_type == PNG_COLOR_TYPE_GRAY) - { - if (row_info->bit_depth == 8) - { - if ((flags & PNG_FLAG_FILLER_AFTER) != 0) - { - /* This changes the data from G to GX */ - png_bytep sp = row + (size_t)row_width; - png_bytep dp = sp + (size_t)row_width; - for (i = 1; i < row_width; i++) - { - *(--dp) = lo_filler; - *(--dp) = *(--sp); - } - *(--dp) = lo_filler; - row_info->channels = 2; - row_info->pixel_depth = 16; - row_info->rowbytes = row_width * 2; - } - - else - { - /* This changes the data from G to XG */ - png_bytep sp = row + (size_t)row_width; - png_bytep dp = sp + (size_t)row_width; - for (i = 0; i < row_width; i++) - { - *(--dp) = *(--sp); - *(--dp) = lo_filler; - } - row_info->channels = 2; - row_info->pixel_depth = 16; - row_info->rowbytes = row_width * 2; - } - } - -#ifdef PNG_READ_16BIT_SUPPORTED - else if (row_info->bit_depth == 16) - { - if ((flags & PNG_FLAG_FILLER_AFTER) != 0) - { - /* This changes the data from GG to GGXX */ - png_bytep sp = row + (size_t)row_width * 2; - png_bytep dp = sp + (size_t)row_width * 2; - for (i = 1; i < row_width; i++) - { - *(--dp) = lo_filler; - *(--dp) = hi_filler; - *(--dp) = *(--sp); - *(--dp) = *(--sp); - } - *(--dp) = lo_filler; - *(--dp) = hi_filler; - row_info->channels = 2; - row_info->pixel_depth = 32; - row_info->rowbytes = row_width * 4; - } - - else - { - /* This changes the data from GG to XXGG */ - png_bytep sp = row + (size_t)row_width * 2; - png_bytep dp = sp + (size_t)row_width * 2; - for (i = 0; i < row_width; i++) - { - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = lo_filler; - *(--dp) = hi_filler; - } - row_info->channels = 2; - row_info->pixel_depth = 32; - row_info->rowbytes = row_width * 4; - } - } -#endif - } /* COLOR_TYPE == GRAY */ - else if (row_info->color_type == PNG_COLOR_TYPE_RGB) - { - if (row_info->bit_depth == 8) - { - if ((flags & PNG_FLAG_FILLER_AFTER) != 0) - { - /* This changes the data from RGB to RGBX */ - png_bytep sp = row + (size_t)row_width * 3; - png_bytep dp = sp + (size_t)row_width; - for (i = 1; i < row_width; i++) - { - *(--dp) = lo_filler; - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - } - *(--dp) = lo_filler; - row_info->channels = 4; - row_info->pixel_depth = 32; - row_info->rowbytes = row_width * 4; - } - - else - { - /* This changes the data from RGB to XRGB */ - png_bytep sp = row + (size_t)row_width * 3; - png_bytep dp = sp + (size_t)row_width; - for (i = 0; i < row_width; i++) - { - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = lo_filler; - } - row_info->channels = 4; - row_info->pixel_depth = 32; - row_info->rowbytes = row_width * 4; - } - } - -#ifdef PNG_READ_16BIT_SUPPORTED - else if (row_info->bit_depth == 16) - { - if ((flags & PNG_FLAG_FILLER_AFTER) != 0) - { - /* This changes the data from RRGGBB to RRGGBBXX */ - png_bytep sp = row + (size_t)row_width * 6; - png_bytep dp = sp + (size_t)row_width * 2; - for (i = 1; i < row_width; i++) - { - *(--dp) = lo_filler; - *(--dp) = hi_filler; - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - } - *(--dp) = lo_filler; - *(--dp) = hi_filler; - row_info->channels = 4; - row_info->pixel_depth = 64; - row_info->rowbytes = row_width * 8; - } - - else - { - /* This changes the data from RRGGBB to XXRRGGBB */ - png_bytep sp = row + (size_t)row_width * 6; - png_bytep dp = sp + (size_t)row_width * 2; - for (i = 0; i < row_width; i++) - { - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = *(--sp); - *(--dp) = lo_filler; - *(--dp) = hi_filler; - } - - row_info->channels = 4; - row_info->pixel_depth = 64; - row_info->rowbytes = row_width * 8; - } - } -#endif - } /* COLOR_TYPE == RGB */ -} -#endif - -#ifdef PNG_READ_GRAY_TO_RGB_SUPPORTED -/* Expand grayscale files to RGB, with or without alpha */ -static void -png_do_gray_to_rgb(png_row_infop row_info, png_bytep row) -{ - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - png_debug(1, "in png_do_gray_to_rgb"); - - if (row_info->bit_depth >= 8 && - (row_info->color_type & PNG_COLOR_MASK_COLOR) == 0) - { - if (row_info->color_type == PNG_COLOR_TYPE_GRAY) - { - if (row_info->bit_depth == 8) - { - /* This changes G to RGB */ - png_bytep sp = row + (size_t)row_width - 1; - png_bytep dp = sp + (size_t)row_width * 2; - for (i = 0; i < row_width; i++) - { - *(dp--) = *sp; - *(dp--) = *sp; - *(dp--) = *(sp--); - } - } - - else - { - /* This changes GG to RRGGBB */ - png_bytep sp = row + (size_t)row_width * 2 - 1; - png_bytep dp = sp + (size_t)row_width * 4; - for (i = 0; i < row_width; i++) - { - *(dp--) = *sp; - *(dp--) = *(sp - 1); - *(dp--) = *sp; - *(dp--) = *(sp - 1); - *(dp--) = *(sp--); - *(dp--) = *(sp--); - } - } - } - - else if (row_info->color_type == PNG_COLOR_TYPE_GRAY_ALPHA) - { - if (row_info->bit_depth == 8) - { - /* This changes GA to RGBA */ - png_bytep sp = row + (size_t)row_width * 2 - 1; - png_bytep dp = sp + (size_t)row_width * 2; - for (i = 0; i < row_width; i++) - { - *(dp--) = *(sp--); - *(dp--) = *sp; - *(dp--) = *sp; - *(dp--) = *(sp--); - } - } - - else - { - /* This changes GGAA to RRGGBBAA */ - png_bytep sp = row + (size_t)row_width * 4 - 1; - png_bytep dp = sp + (size_t)row_width * 4; - for (i = 0; i < row_width; i++) - { - *(dp--) = *(sp--); - *(dp--) = *(sp--); - *(dp--) = *sp; - *(dp--) = *(sp - 1); - *(dp--) = *sp; - *(dp--) = *(sp - 1); - *(dp--) = *(sp--); - *(dp--) = *(sp--); - } - } - } - row_info->channels = (png_byte)(row_info->channels + 2); - row_info->color_type |= PNG_COLOR_MASK_COLOR; - row_info->pixel_depth = (png_byte)(row_info->channels * - row_info->bit_depth); - row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth, row_width); - } -} -#endif - -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED -/* Reduce RGB files to grayscale, with or without alpha - * using the equation given in Poynton's ColorFAQ of 1998-01-04 at - * (THIS LINK IS DEAD June 2008 but - * versions dated 1998 through November 2002 have been archived at - * https://web.archive.org/web/20000816232553/www.inforamp.net/ - * ~poynton/notes/colour_and_gamma/ColorFAQ.txt ) - * Charles Poynton poynton at poynton.com - * - * Y = 0.212671 * R + 0.715160 * G + 0.072169 * B - * - * which can be expressed with integers as - * - * Y = (6969 * R + 23434 * G + 2365 * B)/32768 - * - * Poynton's current link (as of January 2003 through July 2011): - * - * has changed the numbers slightly: - * - * Y = 0.2126*R + 0.7152*G + 0.0722*B - * - * which can be expressed with integers as - * - * Y = (6966 * R + 23436 * G + 2366 * B)/32768 - * - * Historically, however, libpng uses numbers derived from the ITU-R Rec 709 - * end point chromaticities and the D65 white point. Depending on the - * precision used for the D65 white point this produces a variety of different - * numbers, however if the four decimal place value used in ITU-R Rec 709 is - * used (0.3127,0.3290) the Y calculation would be: - * - * Y = (6968 * R + 23435 * G + 2366 * B)/32768 - * - * While this is correct the rounding results in an overflow for white, because - * the sum of the rounded coefficients is 32769, not 32768. Consequently - * libpng uses, instead, the closest non-overflowing approximation: - * - * Y = (6968 * R + 23434 * G + 2366 * B)/32768 - * - * Starting with libpng-1.5.5, if the image being converted has a cHRM chunk - * (including an sRGB chunk) then the chromaticities are used to calculate the - * coefficients. See the chunk handling in pngrutil.c for more information. - * - * In all cases the calculation is to be done in a linear colorspace. If no - * gamma information is available to correct the encoding of the original RGB - * values this results in an implicit assumption that the original PNG RGB - * values were linear. - * - * Other integer coefficients can be used via png_set_rgb_to_gray(). Because - * the API takes just red and green coefficients the blue coefficient is - * calculated to make the sum 32768. This will result in different rounding - * to that used above. - */ -static int -png_do_rgb_to_gray(png_structrp png_ptr, png_row_infop row_info, png_bytep row) -{ - int rgb_error = 0; - - png_debug(1, "in png_do_rgb_to_gray"); - - if ((row_info->color_type & PNG_COLOR_MASK_PALETTE) == 0 && - (row_info->color_type & PNG_COLOR_MASK_COLOR) != 0) - { - png_uint_32 rc = png_ptr->rgb_to_gray_red_coeff; - png_uint_32 gc = png_ptr->rgb_to_gray_green_coeff; - png_uint_32 bc = 32768 - rc - gc; - png_uint_32 row_width = row_info->width; - int have_alpha = (row_info->color_type & PNG_COLOR_MASK_ALPHA) != 0; - - if (row_info->bit_depth == 8) - { -#ifdef PNG_READ_GAMMA_SUPPORTED - /* Notice that gamma to/from 1 are not necessarily inverses (if - * there is an overall gamma correction). Prior to 1.5.5 this code - * checked the linearized values for equality; this doesn't match - * the documentation, the original values must be checked. - */ - if (png_ptr->gamma_from_1 != NULL && png_ptr->gamma_to_1 != NULL) - { - png_bytep sp = row; - png_bytep dp = row; - png_uint_32 i; - - for (i = 0; i < row_width; i++) - { - png_byte red = *(sp++); - png_byte green = *(sp++); - png_byte blue = *(sp++); - - if (red != green || red != blue) - { - red = png_ptr->gamma_to_1[red]; - green = png_ptr->gamma_to_1[green]; - blue = png_ptr->gamma_to_1[blue]; - - rgb_error |= 1; - *(dp++) = png_ptr->gamma_from_1[ - (rc*red + gc*green + bc*blue + 16384)>>15]; - } - - else - { - /* If there is no overall correction the table will not be - * set. - */ - if (png_ptr->gamma_table != NULL) - red = png_ptr->gamma_table[red]; - - *(dp++) = red; - } - - if (have_alpha != 0) - *(dp++) = *(sp++); - } - } - else -#endif - { - png_bytep sp = row; - png_bytep dp = row; - png_uint_32 i; - - for (i = 0; i < row_width; i++) - { - png_byte red = *(sp++); - png_byte green = *(sp++); - png_byte blue = *(sp++); - - if (red != green || red != blue) - { - rgb_error |= 1; - /* NOTE: this is the historical approach which simply - * truncates the results. - */ - *(dp++) = (png_byte)((rc*red + gc*green + bc*blue)>>15); - } - - else - *(dp++) = red; - - if (have_alpha != 0) - *(dp++) = *(sp++); - } - } - } - - else /* RGB bit_depth == 16 */ - { -#ifdef PNG_READ_GAMMA_SUPPORTED - if (png_ptr->gamma_16_to_1 != NULL && png_ptr->gamma_16_from_1 != NULL) - { - png_bytep sp = row; - png_bytep dp = row; - png_uint_32 i; - - for (i = 0; i < row_width; i++) - { - png_uint_16 red, green, blue, w; - png_byte hi,lo; - - hi=*(sp)++; lo=*(sp)++; red = (png_uint_16)((hi << 8) | (lo)); - hi=*(sp)++; lo=*(sp)++; green = (png_uint_16)((hi << 8) | (lo)); - hi=*(sp)++; lo=*(sp)++; blue = (png_uint_16)((hi << 8) | (lo)); - - if (red == green && red == blue) - { - if (png_ptr->gamma_16_table != NULL) - w = png_ptr->gamma_16_table[(red & 0xff) - >> png_ptr->gamma_shift][red >> 8]; - - else - w = red; - } - - else - { - png_uint_16 red_1 = png_ptr->gamma_16_to_1[(red & 0xff) - >> png_ptr->gamma_shift][red>>8]; - png_uint_16 green_1 = - png_ptr->gamma_16_to_1[(green & 0xff) >> - png_ptr->gamma_shift][green>>8]; - png_uint_16 blue_1 = png_ptr->gamma_16_to_1[(blue & 0xff) - >> png_ptr->gamma_shift][blue>>8]; - png_uint_16 gray16 = (png_uint_16)((rc*red_1 + gc*green_1 - + bc*blue_1 + 16384)>>15); - w = png_ptr->gamma_16_from_1[(gray16 & 0xff) >> - png_ptr->gamma_shift][gray16 >> 8]; - rgb_error |= 1; - } - - *(dp++) = (png_byte)((w>>8) & 0xff); - *(dp++) = (png_byte)(w & 0xff); - - if (have_alpha != 0) - { - *(dp++) = *(sp++); - *(dp++) = *(sp++); - } - } - } - else -#endif - { - png_bytep sp = row; - png_bytep dp = row; - png_uint_32 i; - - for (i = 0; i < row_width; i++) - { - png_uint_16 red, green, blue, gray16; - png_byte hi,lo; - - hi=*(sp)++; lo=*(sp)++; red = (png_uint_16)((hi << 8) | (lo)); - hi=*(sp)++; lo=*(sp)++; green = (png_uint_16)((hi << 8) | (lo)); - hi=*(sp)++; lo=*(sp)++; blue = (png_uint_16)((hi << 8) | (lo)); - - if (red != green || red != blue) - rgb_error |= 1; - - /* From 1.5.5 in the 16-bit case do the accurate conversion even - * in the 'fast' case - this is because this is where the code - * ends up when handling linear 16-bit data. - */ - gray16 = (png_uint_16)((rc*red + gc*green + bc*blue + 16384) >> - 15); - *(dp++) = (png_byte)((gray16 >> 8) & 0xff); - *(dp++) = (png_byte)(gray16 & 0xff); - - if (have_alpha != 0) - { - *(dp++) = *(sp++); - *(dp++) = *(sp++); - } - } - } - } - - row_info->channels = (png_byte)(row_info->channels - 2); - row_info->color_type = (png_byte)(row_info->color_type & - ~PNG_COLOR_MASK_COLOR); - row_info->pixel_depth = (png_byte)(row_info->channels * - row_info->bit_depth); - row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth, row_width); - } - return rgb_error; -} -#endif - -#if defined(PNG_READ_BACKGROUND_SUPPORTED) ||\ - defined(PNG_READ_ALPHA_MODE_SUPPORTED) -/* Replace any alpha or transparency with the supplied background color. - * "background" is already in the screen gamma, while "background_1" is - * at a gamma of 1.0. Paletted files have already been taken care of. - */ -static void -png_do_compose(png_row_infop row_info, png_bytep row, png_structrp png_ptr) -{ -#ifdef PNG_READ_GAMMA_SUPPORTED - png_const_bytep gamma_table = png_ptr->gamma_table; - png_const_bytep gamma_from_1 = png_ptr->gamma_from_1; - png_const_bytep gamma_to_1 = png_ptr->gamma_to_1; - png_const_uint_16pp gamma_16 = png_ptr->gamma_16_table; - png_const_uint_16pp gamma_16_from_1 = png_ptr->gamma_16_from_1; - png_const_uint_16pp gamma_16_to_1 = png_ptr->gamma_16_to_1; - int gamma_shift = png_ptr->gamma_shift; - int optimize = (png_ptr->flags & PNG_FLAG_OPTIMIZE_ALPHA) != 0; -#endif - - png_bytep sp; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - int shift; - - png_debug(1, "in png_do_compose"); - - switch (row_info->color_type) - { - case PNG_COLOR_TYPE_GRAY: - { - switch (row_info->bit_depth) - { - case 1: - { - sp = row; - shift = 7; - for (i = 0; i < row_width; i++) - { - if ((png_uint_16)((*sp >> shift) & 0x01) - == png_ptr->trans_color.gray) - { - unsigned int tmp = *sp & (0x7f7f >> (7 - shift)); - tmp |= - (unsigned int)(png_ptr->background.gray << shift); - *sp = (png_byte)(tmp & 0xff); - } - - if (shift == 0) - { - shift = 7; - sp++; - } - - else - shift--; - } - break; - } - - case 2: - { -#ifdef PNG_READ_GAMMA_SUPPORTED - if (gamma_table != NULL) - { - sp = row; - shift = 6; - for (i = 0; i < row_width; i++) - { - if ((png_uint_16)((*sp >> shift) & 0x03) - == png_ptr->trans_color.gray) - { - unsigned int tmp = *sp & (0x3f3f >> (6 - shift)); - tmp |= - (unsigned int)png_ptr->background.gray << shift; - *sp = (png_byte)(tmp & 0xff); - } - - else - { - unsigned int p = (*sp >> shift) & 0x03; - unsigned int g = (gamma_table [p | (p << 2) | - (p << 4) | (p << 6)] >> 6) & 0x03; - unsigned int tmp = *sp & (0x3f3f >> (6 - shift)); - tmp |= (unsigned int)(g << shift); - *sp = (png_byte)(tmp & 0xff); - } - - if (shift == 0) - { - shift = 6; - sp++; - } - - else - shift -= 2; - } - } - - else -#endif - { - sp = row; - shift = 6; - for (i = 0; i < row_width; i++) - { - if ((png_uint_16)((*sp >> shift) & 0x03) - == png_ptr->trans_color.gray) - { - unsigned int tmp = *sp & (0x3f3f >> (6 - shift)); - tmp |= - (unsigned int)png_ptr->background.gray << shift; - *sp = (png_byte)(tmp & 0xff); - } - - if (shift == 0) - { - shift = 6; - sp++; - } - - else - shift -= 2; - } - } - break; - } - - case 4: - { -#ifdef PNG_READ_GAMMA_SUPPORTED - if (gamma_table != NULL) - { - sp = row; - shift = 4; - for (i = 0; i < row_width; i++) - { - if ((png_uint_16)((*sp >> shift) & 0x0f) - == png_ptr->trans_color.gray) - { - unsigned int tmp = *sp & (0x0f0f >> (4 - shift)); - tmp |= - (unsigned int)(png_ptr->background.gray << shift); - *sp = (png_byte)(tmp & 0xff); - } - - else - { - unsigned int p = (*sp >> shift) & 0x0f; - unsigned int g = (gamma_table[p | (p << 4)] >> 4) & - 0x0f; - unsigned int tmp = *sp & (0x0f0f >> (4 - shift)); - tmp |= (unsigned int)(g << shift); - *sp = (png_byte)(tmp & 0xff); - } - - if (shift == 0) - { - shift = 4; - sp++; - } - - else - shift -= 4; - } - } - - else -#endif - { - sp = row; - shift = 4; - for (i = 0; i < row_width; i++) - { - if ((png_uint_16)((*sp >> shift) & 0x0f) - == png_ptr->trans_color.gray) - { - unsigned int tmp = *sp & (0x0f0f >> (4 - shift)); - tmp |= - (unsigned int)(png_ptr->background.gray << shift); - *sp = (png_byte)(tmp & 0xff); - } - - if (shift == 0) - { - shift = 4; - sp++; - } - - else - shift -= 4; - } - } - break; - } - - case 8: - { -#ifdef PNG_READ_GAMMA_SUPPORTED - if (gamma_table != NULL) - { - sp = row; - for (i = 0; i < row_width; i++, sp++) - { - if (*sp == png_ptr->trans_color.gray) - *sp = (png_byte)png_ptr->background.gray; - - else - *sp = gamma_table[*sp]; - } - } - else -#endif - { - sp = row; - for (i = 0; i < row_width; i++, sp++) - { - if (*sp == png_ptr->trans_color.gray) - *sp = (png_byte)png_ptr->background.gray; - } - } - break; - } - - case 16: - { -#ifdef PNG_READ_GAMMA_SUPPORTED - if (gamma_16 != NULL) - { - sp = row; - for (i = 0; i < row_width; i++, sp += 2) - { - png_uint_16 v; - - v = (png_uint_16)(((*sp) << 8) + *(sp + 1)); - - if (v == png_ptr->trans_color.gray) - { - /* Background is already in screen gamma */ - *sp = (png_byte)((png_ptr->background.gray >> 8) - & 0xff); - *(sp + 1) = (png_byte)(png_ptr->background.gray - & 0xff); - } - - else - { - v = gamma_16[*(sp + 1) >> gamma_shift][*sp]; - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - } - } - } - else -#endif - { - sp = row; - for (i = 0; i < row_width; i++, sp += 2) - { - png_uint_16 v; - - v = (png_uint_16)(((*sp) << 8) + *(sp + 1)); - - if (v == png_ptr->trans_color.gray) - { - *sp = (png_byte)((png_ptr->background.gray >> 8) - & 0xff); - *(sp + 1) = (png_byte)(png_ptr->background.gray - & 0xff); - } - } - } - break; - } - - default: - break; - } - break; - } - - case PNG_COLOR_TYPE_RGB: - { - if (row_info->bit_depth == 8) - { -#ifdef PNG_READ_GAMMA_SUPPORTED - if (gamma_table != NULL) - { - sp = row; - for (i = 0; i < row_width; i++, sp += 3) - { - if (*sp == png_ptr->trans_color.red && - *(sp + 1) == png_ptr->trans_color.green && - *(sp + 2) == png_ptr->trans_color.blue) - { - *sp = (png_byte)png_ptr->background.red; - *(sp + 1) = (png_byte)png_ptr->background.green; - *(sp + 2) = (png_byte)png_ptr->background.blue; - } - - else - { - *sp = gamma_table[*sp]; - *(sp + 1) = gamma_table[*(sp + 1)]; - *(sp + 2) = gamma_table[*(sp + 2)]; - } - } - } - else -#endif - { - sp = row; - for (i = 0; i < row_width; i++, sp += 3) - { - if (*sp == png_ptr->trans_color.red && - *(sp + 1) == png_ptr->trans_color.green && - *(sp + 2) == png_ptr->trans_color.blue) - { - *sp = (png_byte)png_ptr->background.red; - *(sp + 1) = (png_byte)png_ptr->background.green; - *(sp + 2) = (png_byte)png_ptr->background.blue; - } - } - } - } - else /* if (row_info->bit_depth == 16) */ - { -#ifdef PNG_READ_GAMMA_SUPPORTED - if (gamma_16 != NULL) - { - sp = row; - for (i = 0; i < row_width; i++, sp += 6) - { - png_uint_16 r = (png_uint_16)(((*sp) << 8) + *(sp + 1)); - - png_uint_16 g = (png_uint_16)(((*(sp + 2)) << 8) - + *(sp + 3)); - - png_uint_16 b = (png_uint_16)(((*(sp + 4)) << 8) - + *(sp + 5)); - - if (r == png_ptr->trans_color.red && - g == png_ptr->trans_color.green && - b == png_ptr->trans_color.blue) - { - /* Background is already in screen gamma */ - *sp = (png_byte)((png_ptr->background.red >> 8) & 0xff); - *(sp + 1) = (png_byte)(png_ptr->background.red & 0xff); - *(sp + 2) = (png_byte)((png_ptr->background.green >> 8) - & 0xff); - *(sp + 3) = (png_byte)(png_ptr->background.green - & 0xff); - *(sp + 4) = (png_byte)((png_ptr->background.blue >> 8) - & 0xff); - *(sp + 5) = (png_byte)(png_ptr->background.blue & 0xff); - } - - else - { - png_uint_16 v = gamma_16[*(sp + 1) >> gamma_shift][*sp]; - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - - v = gamma_16[*(sp + 3) >> gamma_shift][*(sp + 2)]; - *(sp + 2) = (png_byte)((v >> 8) & 0xff); - *(sp + 3) = (png_byte)(v & 0xff); - - v = gamma_16[*(sp + 5) >> gamma_shift][*(sp + 4)]; - *(sp + 4) = (png_byte)((v >> 8) & 0xff); - *(sp + 5) = (png_byte)(v & 0xff); - } - } - } - - else -#endif - { - sp = row; - for (i = 0; i < row_width; i++, sp += 6) - { - png_uint_16 r = (png_uint_16)(((*sp) << 8) + *(sp + 1)); - - png_uint_16 g = (png_uint_16)(((*(sp + 2)) << 8) - + *(sp + 3)); - - png_uint_16 b = (png_uint_16)(((*(sp + 4)) << 8) - + *(sp + 5)); - - if (r == png_ptr->trans_color.red && - g == png_ptr->trans_color.green && - b == png_ptr->trans_color.blue) - { - *sp = (png_byte)((png_ptr->background.red >> 8) & 0xff); - *(sp + 1) = (png_byte)(png_ptr->background.red & 0xff); - *(sp + 2) = (png_byte)((png_ptr->background.green >> 8) - & 0xff); - *(sp + 3) = (png_byte)(png_ptr->background.green - & 0xff); - *(sp + 4) = (png_byte)((png_ptr->background.blue >> 8) - & 0xff); - *(sp + 5) = (png_byte)(png_ptr->background.blue & 0xff); - } - } - } - } - break; - } - - case PNG_COLOR_TYPE_GRAY_ALPHA: - { - if (row_info->bit_depth == 8) - { -#ifdef PNG_READ_GAMMA_SUPPORTED - if (gamma_to_1 != NULL && gamma_from_1 != NULL && - gamma_table != NULL) - { - sp = row; - for (i = 0; i < row_width; i++, sp += 2) - { - png_uint_16 a = *(sp + 1); - - if (a == 0xff) - *sp = gamma_table[*sp]; - - else if (a == 0) - { - /* Background is already in screen gamma */ - *sp = (png_byte)png_ptr->background.gray; - } - - else - { - png_byte v, w; - - v = gamma_to_1[*sp]; - png_composite(w, v, a, png_ptr->background_1.gray); - if (optimize == 0) - w = gamma_from_1[w]; - *sp = w; - } - } - } - else -#endif - { - sp = row; - for (i = 0; i < row_width; i++, sp += 2) - { - png_byte a = *(sp + 1); - - if (a == 0) - *sp = (png_byte)png_ptr->background.gray; - - else if (a < 0xff) - png_composite(*sp, *sp, a, png_ptr->background.gray); - } - } - } - else /* if (png_ptr->bit_depth == 16) */ - { -#ifdef PNG_READ_GAMMA_SUPPORTED - if (gamma_16 != NULL && gamma_16_from_1 != NULL && - gamma_16_to_1 != NULL) - { - sp = row; - for (i = 0; i < row_width; i++, sp += 4) - { - png_uint_16 a = (png_uint_16)(((*(sp + 2)) << 8) - + *(sp + 3)); - - if (a == (png_uint_16)0xffff) - { - png_uint_16 v; - - v = gamma_16[*(sp + 1) >> gamma_shift][*sp]; - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - } - - else if (a == 0) - { - /* Background is already in screen gamma */ - *sp = (png_byte)((png_ptr->background.gray >> 8) - & 0xff); - *(sp + 1) = (png_byte)(png_ptr->background.gray & 0xff); - } - - else - { - png_uint_16 g, v, w; - - g = gamma_16_to_1[*(sp + 1) >> gamma_shift][*sp]; - png_composite_16(v, g, a, png_ptr->background_1.gray); - if (optimize != 0) - w = v; - else - w = gamma_16_from_1[(v & 0xff) >> - gamma_shift][v >> 8]; - *sp = (png_byte)((w >> 8) & 0xff); - *(sp + 1) = (png_byte)(w & 0xff); - } - } - } - else -#endif - { - sp = row; - for (i = 0; i < row_width; i++, sp += 4) - { - png_uint_16 a = (png_uint_16)(((*(sp + 2)) << 8) - + *(sp + 3)); - - if (a == 0) - { - *sp = (png_byte)((png_ptr->background.gray >> 8) - & 0xff); - *(sp + 1) = (png_byte)(png_ptr->background.gray & 0xff); - } - - else if (a < 0xffff) - { - png_uint_16 g, v; - - g = (png_uint_16)(((*sp) << 8) + *(sp + 1)); - png_composite_16(v, g, a, png_ptr->background.gray); - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - } - } - } - } - break; - } - - case PNG_COLOR_TYPE_RGB_ALPHA: - { - if (row_info->bit_depth == 8) - { -#ifdef PNG_READ_GAMMA_SUPPORTED - if (gamma_to_1 != NULL && gamma_from_1 != NULL && - gamma_table != NULL) - { - sp = row; - for (i = 0; i < row_width; i++, sp += 4) - { - png_byte a = *(sp + 3); - - if (a == 0xff) - { - *sp = gamma_table[*sp]; - *(sp + 1) = gamma_table[*(sp + 1)]; - *(sp + 2) = gamma_table[*(sp + 2)]; - } - - else if (a == 0) - { - /* Background is already in screen gamma */ - *sp = (png_byte)png_ptr->background.red; - *(sp + 1) = (png_byte)png_ptr->background.green; - *(sp + 2) = (png_byte)png_ptr->background.blue; - } - - else - { - png_byte v, w; - - v = gamma_to_1[*sp]; - png_composite(w, v, a, png_ptr->background_1.red); - if (optimize == 0) w = gamma_from_1[w]; - *sp = w; - - v = gamma_to_1[*(sp + 1)]; - png_composite(w, v, a, png_ptr->background_1.green); - if (optimize == 0) w = gamma_from_1[w]; - *(sp + 1) = w; - - v = gamma_to_1[*(sp + 2)]; - png_composite(w, v, a, png_ptr->background_1.blue); - if (optimize == 0) w = gamma_from_1[w]; - *(sp + 2) = w; - } - } - } - else -#endif - { - sp = row; - for (i = 0; i < row_width; i++, sp += 4) - { - png_byte a = *(sp + 3); - - if (a == 0) - { - *sp = (png_byte)png_ptr->background.red; - *(sp + 1) = (png_byte)png_ptr->background.green; - *(sp + 2) = (png_byte)png_ptr->background.blue; - } - - else if (a < 0xff) - { - png_composite(*sp, *sp, a, png_ptr->background.red); - - png_composite(*(sp + 1), *(sp + 1), a, - png_ptr->background.green); - - png_composite(*(sp + 2), *(sp + 2), a, - png_ptr->background.blue); - } - } - } - } - else /* if (row_info->bit_depth == 16) */ - { -#ifdef PNG_READ_GAMMA_SUPPORTED - if (gamma_16 != NULL && gamma_16_from_1 != NULL && - gamma_16_to_1 != NULL) - { - sp = row; - for (i = 0; i < row_width; i++, sp += 8) - { - png_uint_16 a = (png_uint_16)(((png_uint_16)(*(sp + 6)) - << 8) + (png_uint_16)(*(sp + 7))); - - if (a == (png_uint_16)0xffff) - { - png_uint_16 v; - - v = gamma_16[*(sp + 1) >> gamma_shift][*sp]; - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - - v = gamma_16[*(sp + 3) >> gamma_shift][*(sp + 2)]; - *(sp + 2) = (png_byte)((v >> 8) & 0xff); - *(sp + 3) = (png_byte)(v & 0xff); - - v = gamma_16[*(sp + 5) >> gamma_shift][*(sp + 4)]; - *(sp + 4) = (png_byte)((v >> 8) & 0xff); - *(sp + 5) = (png_byte)(v & 0xff); - } - - else if (a == 0) - { - /* Background is already in screen gamma */ - *sp = (png_byte)((png_ptr->background.red >> 8) & 0xff); - *(sp + 1) = (png_byte)(png_ptr->background.red & 0xff); - *(sp + 2) = (png_byte)((png_ptr->background.green >> 8) - & 0xff); - *(sp + 3) = (png_byte)(png_ptr->background.green - & 0xff); - *(sp + 4) = (png_byte)((png_ptr->background.blue >> 8) - & 0xff); - *(sp + 5) = (png_byte)(png_ptr->background.blue & 0xff); - } - - else - { - png_uint_16 v, w; - - v = gamma_16_to_1[*(sp + 1) >> gamma_shift][*sp]; - png_composite_16(w, v, a, png_ptr->background_1.red); - if (optimize == 0) - w = gamma_16_from_1[((w & 0xff) >> gamma_shift)][w >> - 8]; - *sp = (png_byte)((w >> 8) & 0xff); - *(sp + 1) = (png_byte)(w & 0xff); - - v = gamma_16_to_1[*(sp + 3) >> gamma_shift][*(sp + 2)]; - png_composite_16(w, v, a, png_ptr->background_1.green); - if (optimize == 0) - w = gamma_16_from_1[((w & 0xff) >> gamma_shift)][w >> - 8]; - - *(sp + 2) = (png_byte)((w >> 8) & 0xff); - *(sp + 3) = (png_byte)(w & 0xff); - - v = gamma_16_to_1[*(sp + 5) >> gamma_shift][*(sp + 4)]; - png_composite_16(w, v, a, png_ptr->background_1.blue); - if (optimize == 0) - w = gamma_16_from_1[((w & 0xff) >> gamma_shift)][w >> - 8]; - - *(sp + 4) = (png_byte)((w >> 8) & 0xff); - *(sp + 5) = (png_byte)(w & 0xff); - } - } - } - - else -#endif - { - sp = row; - for (i = 0; i < row_width; i++, sp += 8) - { - png_uint_16 a = (png_uint_16)(((png_uint_16)(*(sp + 6)) - << 8) + (png_uint_16)(*(sp + 7))); - - if (a == 0) - { - *sp = (png_byte)((png_ptr->background.red >> 8) & 0xff); - *(sp + 1) = (png_byte)(png_ptr->background.red & 0xff); - *(sp + 2) = (png_byte)((png_ptr->background.green >> 8) - & 0xff); - *(sp + 3) = (png_byte)(png_ptr->background.green - & 0xff); - *(sp + 4) = (png_byte)((png_ptr->background.blue >> 8) - & 0xff); - *(sp + 5) = (png_byte)(png_ptr->background.blue & 0xff); - } - - else if (a < 0xffff) - { - png_uint_16 v; - - png_uint_16 r = (png_uint_16)(((*sp) << 8) + *(sp + 1)); - png_uint_16 g = (png_uint_16)(((*(sp + 2)) << 8) - + *(sp + 3)); - png_uint_16 b = (png_uint_16)(((*(sp + 4)) << 8) - + *(sp + 5)); - - png_composite_16(v, r, a, png_ptr->background.red); - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - - png_composite_16(v, g, a, png_ptr->background.green); - *(sp + 2) = (png_byte)((v >> 8) & 0xff); - *(sp + 3) = (png_byte)(v & 0xff); - - png_composite_16(v, b, a, png_ptr->background.blue); - *(sp + 4) = (png_byte)((v >> 8) & 0xff); - *(sp + 5) = (png_byte)(v & 0xff); - } - } - } - } - break; - } - - default: - break; - } -} -#endif /* READ_BACKGROUND || READ_ALPHA_MODE */ - -#ifdef PNG_READ_GAMMA_SUPPORTED -/* Gamma correct the image, avoiding the alpha channel. Make sure - * you do this after you deal with the transparency issue on grayscale - * or RGB images. If your bit depth is 8, use gamma_table, if it - * is 16, use gamma_16_table and gamma_shift. Build these with - * build_gamma_table(). - */ -static void -png_do_gamma(png_row_infop row_info, png_bytep row, png_structrp png_ptr) -{ - png_const_bytep gamma_table = png_ptr->gamma_table; - png_const_uint_16pp gamma_16_table = png_ptr->gamma_16_table; - int gamma_shift = png_ptr->gamma_shift; - - png_bytep sp; - png_uint_32 i; - png_uint_32 row_width=row_info->width; - - png_debug(1, "in png_do_gamma"); - - if (((row_info->bit_depth <= 8 && gamma_table != NULL) || - (row_info->bit_depth == 16 && gamma_16_table != NULL))) - { - switch (row_info->color_type) - { - case PNG_COLOR_TYPE_RGB: - { - if (row_info->bit_depth == 8) - { - sp = row; - for (i = 0; i < row_width; i++) - { - *sp = gamma_table[*sp]; - sp++; - *sp = gamma_table[*sp]; - sp++; - *sp = gamma_table[*sp]; - sp++; - } - } - - else /* if (row_info->bit_depth == 16) */ - { - sp = row; - for (i = 0; i < row_width; i++) - { - png_uint_16 v; - - v = gamma_16_table[*(sp + 1) >> gamma_shift][*sp]; - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - sp += 2; - - v = gamma_16_table[*(sp + 1) >> gamma_shift][*sp]; - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - sp += 2; - - v = gamma_16_table[*(sp + 1) >> gamma_shift][*sp]; - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - sp += 2; - } - } - break; - } - - case PNG_COLOR_TYPE_RGB_ALPHA: - { - if (row_info->bit_depth == 8) - { - sp = row; - for (i = 0; i < row_width; i++) - { - *sp = gamma_table[*sp]; - sp++; - - *sp = gamma_table[*sp]; - sp++; - - *sp = gamma_table[*sp]; - sp++; - - sp++; - } - } - - else /* if (row_info->bit_depth == 16) */ - { - sp = row; - for (i = 0; i < row_width; i++) - { - png_uint_16 v = gamma_16_table[*(sp + 1) >> gamma_shift][*sp]; - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - sp += 2; - - v = gamma_16_table[*(sp + 1) >> gamma_shift][*sp]; - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - sp += 2; - - v = gamma_16_table[*(sp + 1) >> gamma_shift][*sp]; - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - sp += 4; - } - } - break; - } - - case PNG_COLOR_TYPE_GRAY_ALPHA: - { - if (row_info->bit_depth == 8) - { - sp = row; - for (i = 0; i < row_width; i++) - { - *sp = gamma_table[*sp]; - sp += 2; - } - } - - else /* if (row_info->bit_depth == 16) */ - { - sp = row; - for (i = 0; i < row_width; i++) - { - png_uint_16 v = gamma_16_table[*(sp + 1) >> gamma_shift][*sp]; - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - sp += 4; - } - } - break; - } - - case PNG_COLOR_TYPE_GRAY: - { - if (row_info->bit_depth == 2) - { - sp = row; - for (i = 0; i < row_width; i += 4) - { - int a = *sp & 0xc0; - int b = *sp & 0x30; - int c = *sp & 0x0c; - int d = *sp & 0x03; - - *sp = (png_byte)( - ((((int)gamma_table[a|(a>>2)|(a>>4)|(a>>6)]) ) & 0xc0)| - ((((int)gamma_table[(b<<2)|b|(b>>2)|(b>>4)])>>2) & 0x30)| - ((((int)gamma_table[(c<<4)|(c<<2)|c|(c>>2)])>>4) & 0x0c)| - ((((int)gamma_table[(d<<6)|(d<<4)|(d<<2)|d])>>6) )); - sp++; - } - } - - if (row_info->bit_depth == 4) - { - sp = row; - for (i = 0; i < row_width; i += 2) - { - int msb = *sp & 0xf0; - int lsb = *sp & 0x0f; - - *sp = (png_byte)((((int)gamma_table[msb | (msb >> 4)]) & 0xf0) - | (((int)gamma_table[(lsb << 4) | lsb]) >> 4)); - sp++; - } - } - - else if (row_info->bit_depth == 8) - { - sp = row; - for (i = 0; i < row_width; i++) - { - *sp = gamma_table[*sp]; - sp++; - } - } - - else if (row_info->bit_depth == 16) - { - sp = row; - for (i = 0; i < row_width; i++) - { - png_uint_16 v = gamma_16_table[*(sp + 1) >> gamma_shift][*sp]; - *sp = (png_byte)((v >> 8) & 0xff); - *(sp + 1) = (png_byte)(v & 0xff); - sp += 2; - } - } - break; - } - - default: - break; - } - } -} -#endif - -#ifdef PNG_READ_ALPHA_MODE_SUPPORTED -/* Encode the alpha channel to the output gamma (the input channel is always - * linear.) Called only with color types that have an alpha channel. Needs the - * from_1 tables. - */ -static void -png_do_encode_alpha(png_row_infop row_info, png_bytep row, png_structrp png_ptr) -{ - png_uint_32 row_width = row_info->width; - - png_debug(1, "in png_do_encode_alpha"); - - if ((row_info->color_type & PNG_COLOR_MASK_ALPHA) != 0) - { - if (row_info->bit_depth == 8) - { - png_bytep table = png_ptr->gamma_from_1; - - if (table != NULL) - { - int step = (row_info->color_type & PNG_COLOR_MASK_COLOR) ? 4 : 2; - - /* The alpha channel is the last component: */ - row += step - 1; - - for (; row_width > 0; --row_width, row += step) - *row = table[*row]; - - return; - } - } - - else if (row_info->bit_depth == 16) - { - png_uint_16pp table = png_ptr->gamma_16_from_1; - int gamma_shift = png_ptr->gamma_shift; - - if (table != NULL) - { - int step = (row_info->color_type & PNG_COLOR_MASK_COLOR) ? 8 : 4; - - /* The alpha channel is the last component: */ - row += step - 2; - - for (; row_width > 0; --row_width, row += step) - { - png_uint_16 v; - - v = table[*(row + 1) >> gamma_shift][*row]; - *row = (png_byte)((v >> 8) & 0xff); - *(row + 1) = (png_byte)(v & 0xff); - } - - return; - } - } - } - - /* Only get to here if called with a weird row_info; no harm has been done, - * so just issue a warning. - */ - png_warning(png_ptr, "png_do_encode_alpha: unexpected call"); -} -#endif - -#ifdef PNG_READ_EXPAND_SUPPORTED -/* Expands a palette row to an RGB or RGBA row depending - * upon whether you supply trans and num_trans. - */ -static void -png_do_expand_palette(png_structrp png_ptr, png_row_infop row_info, - png_bytep row, png_const_colorp palette, png_const_bytep trans_alpha, - int num_trans) -{ - int shift, value; - png_bytep sp, dp; - png_uint_32 i; - png_uint_32 row_width=row_info->width; - - png_debug(1, "in png_do_expand_palette"); - - if (row_info->color_type == PNG_COLOR_TYPE_PALETTE) - { - if (row_info->bit_depth < 8) - { - switch (row_info->bit_depth) - { - case 1: - { - sp = row + (size_t)((row_width - 1) >> 3); - dp = row + (size_t)row_width - 1; - shift = 7 - (int)((row_width + 7) & 0x07); - for (i = 0; i < row_width; i++) - { - if ((*sp >> shift) & 0x01) - *dp = 1; - - else - *dp = 0; - - if (shift == 7) - { - shift = 0; - sp--; - } - - else - shift++; - - dp--; - } - break; - } - - case 2: - { - sp = row + (size_t)((row_width - 1) >> 2); - dp = row + (size_t)row_width - 1; - shift = (int)((3 - ((row_width + 3) & 0x03)) << 1); - for (i = 0; i < row_width; i++) - { - value = (*sp >> shift) & 0x03; - *dp = (png_byte)value; - if (shift == 6) - { - shift = 0; - sp--; - } - - else - shift += 2; - - dp--; - } - break; - } - - case 4: - { - sp = row + (size_t)((row_width - 1) >> 1); - dp = row + (size_t)row_width - 1; - shift = (int)((row_width & 0x01) << 2); - for (i = 0; i < row_width; i++) - { - value = (*sp >> shift) & 0x0f; - *dp = (png_byte)value; - if (shift == 4) - { - shift = 0; - sp--; - } - - else - shift += 4; - - dp--; - } - break; - } - - default: - break; - } - row_info->bit_depth = 8; - row_info->pixel_depth = 8; - row_info->rowbytes = row_width; - } - - if (row_info->bit_depth == 8) - { - { - if (num_trans > 0) - { - sp = row + (size_t)row_width - 1; - dp = row + ((size_t)row_width << 2) - 1; - - i = 0; -#ifdef PNG_ARM_NEON_INTRINSICS_AVAILABLE - if (png_ptr->riffled_palette != NULL) - { - /* The RGBA optimization works with png_ptr->bit_depth == 8 - * but sometimes row_info->bit_depth has been changed to 8. - * In these cases, the palette hasn't been riffled. - */ - i = png_do_expand_palette_rgba8_neon(png_ptr, row_info, row, - &sp, &dp); - } -#else - PNG_UNUSED(png_ptr) -#endif - - for (; i < row_width; i++) - { - if ((int)(*sp) >= num_trans) - *dp-- = 0xff; - else - *dp-- = trans_alpha[*sp]; - *dp-- = palette[*sp].blue; - *dp-- = palette[*sp].green; - *dp-- = palette[*sp].red; - sp--; - } - row_info->bit_depth = 8; - row_info->pixel_depth = 32; - row_info->rowbytes = row_width * 4; - row_info->color_type = 6; - row_info->channels = 4; - } - - else - { - sp = row + (size_t)row_width - 1; - dp = row + (size_t)(row_width * 3) - 1; - i = 0; -#ifdef PNG_ARM_NEON_INTRINSICS_AVAILABLE - i = png_do_expand_palette_rgb8_neon(png_ptr, row_info, row, - &sp, &dp); -#else - PNG_UNUSED(png_ptr) -#endif - - for (; i < row_width; i++) - { - *dp-- = palette[*sp].blue; - *dp-- = palette[*sp].green; - *dp-- = palette[*sp].red; - sp--; - } - - row_info->bit_depth = 8; - row_info->pixel_depth = 24; - row_info->rowbytes = row_width * 3; - row_info->color_type = 2; - row_info->channels = 3; - } - } - } - } -} - -/* If the bit depth < 8, it is expanded to 8. Also, if the already - * expanded transparency value is supplied, an alpha channel is built. - */ -static void -png_do_expand(png_row_infop row_info, png_bytep row, - png_const_color_16p trans_color) -{ - int shift, value; - png_bytep sp, dp; - png_uint_32 i; - png_uint_32 row_width=row_info->width; - - png_debug(1, "in png_do_expand"); - - if (row_info->color_type == PNG_COLOR_TYPE_GRAY) - { - unsigned int gray = trans_color != NULL ? trans_color->gray : 0; - - if (row_info->bit_depth < 8) - { - switch (row_info->bit_depth) - { - case 1: - { - gray = (gray & 0x01) * 0xff; - sp = row + (size_t)((row_width - 1) >> 3); - dp = row + (size_t)row_width - 1; - shift = 7 - (int)((row_width + 7) & 0x07); - for (i = 0; i < row_width; i++) - { - if ((*sp >> shift) & 0x01) - *dp = 0xff; - - else - *dp = 0; - - if (shift == 7) - { - shift = 0; - sp--; - } - - else - shift++; - - dp--; - } - break; - } - - case 2: - { - gray = (gray & 0x03) * 0x55; - sp = row + (size_t)((row_width - 1) >> 2); - dp = row + (size_t)row_width - 1; - shift = (int)((3 - ((row_width + 3) & 0x03)) << 1); - for (i = 0; i < row_width; i++) - { - value = (*sp >> shift) & 0x03; - *dp = (png_byte)(value | (value << 2) | (value << 4) | - (value << 6)); - if (shift == 6) - { - shift = 0; - sp--; - } - - else - shift += 2; - - dp--; - } - break; - } - - case 4: - { - gray = (gray & 0x0f) * 0x11; - sp = row + (size_t)((row_width - 1) >> 1); - dp = row + (size_t)row_width - 1; - shift = (int)((1 - ((row_width + 1) & 0x01)) << 2); - for (i = 0; i < row_width; i++) - { - value = (*sp >> shift) & 0x0f; - *dp = (png_byte)(value | (value << 4)); - if (shift == 4) - { - shift = 0; - sp--; - } - - else - shift = 4; - - dp--; - } - break; - } - - default: - break; - } - - row_info->bit_depth = 8; - row_info->pixel_depth = 8; - row_info->rowbytes = row_width; - } - - if (trans_color != NULL) - { - if (row_info->bit_depth == 8) - { - gray = gray & 0xff; - sp = row + (size_t)row_width - 1; - dp = row + ((size_t)row_width << 1) - 1; - - for (i = 0; i < row_width; i++) - { - if ((*sp & 0xffU) == gray) - *dp-- = 0; - - else - *dp-- = 0xff; - - *dp-- = *sp--; - } - } - - else if (row_info->bit_depth == 16) - { - unsigned int gray_high = (gray >> 8) & 0xff; - unsigned int gray_low = gray & 0xff; - sp = row + row_info->rowbytes - 1; - dp = row + (row_info->rowbytes << 1) - 1; - for (i = 0; i < row_width; i++) - { - if ((*(sp - 1) & 0xffU) == gray_high && - (*(sp) & 0xffU) == gray_low) - { - *dp-- = 0; - *dp-- = 0; - } - - else - { - *dp-- = 0xff; - *dp-- = 0xff; - } - - *dp-- = *sp--; - *dp-- = *sp--; - } - } - - row_info->color_type = PNG_COLOR_TYPE_GRAY_ALPHA; - row_info->channels = 2; - row_info->pixel_depth = (png_byte)(row_info->bit_depth << 1); - row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth, - row_width); - } - } - else if (row_info->color_type == PNG_COLOR_TYPE_RGB && - trans_color != NULL) - { - if (row_info->bit_depth == 8) - { - png_byte red = (png_byte)(trans_color->red & 0xff); - png_byte green = (png_byte)(trans_color->green & 0xff); - png_byte blue = (png_byte)(trans_color->blue & 0xff); - sp = row + (size_t)row_info->rowbytes - 1; - dp = row + ((size_t)row_width << 2) - 1; - for (i = 0; i < row_width; i++) - { - if (*(sp - 2) == red && *(sp - 1) == green && *(sp) == blue) - *dp-- = 0; - - else - *dp-- = 0xff; - - *dp-- = *sp--; - *dp-- = *sp--; - *dp-- = *sp--; - } - } - else if (row_info->bit_depth == 16) - { - png_byte red_high = (png_byte)((trans_color->red >> 8) & 0xff); - png_byte green_high = (png_byte)((trans_color->green >> 8) & 0xff); - png_byte blue_high = (png_byte)((trans_color->blue >> 8) & 0xff); - png_byte red_low = (png_byte)(trans_color->red & 0xff); - png_byte green_low = (png_byte)(trans_color->green & 0xff); - png_byte blue_low = (png_byte)(trans_color->blue & 0xff); - sp = row + row_info->rowbytes - 1; - dp = row + ((size_t)row_width << 3) - 1; - for (i = 0; i < row_width; i++) - { - if (*(sp - 5) == red_high && - *(sp - 4) == red_low && - *(sp - 3) == green_high && - *(sp - 2) == green_low && - *(sp - 1) == blue_high && - *(sp ) == blue_low) - { - *dp-- = 0; - *dp-- = 0; - } - - else - { - *dp-- = 0xff; - *dp-- = 0xff; - } - - *dp-- = *sp--; - *dp-- = *sp--; - *dp-- = *sp--; - *dp-- = *sp--; - *dp-- = *sp--; - *dp-- = *sp--; - } - } - row_info->color_type = PNG_COLOR_TYPE_RGB_ALPHA; - row_info->channels = 4; - row_info->pixel_depth = (png_byte)(row_info->bit_depth << 2); - row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth, row_width); - } -} -#endif - -#ifdef PNG_READ_EXPAND_16_SUPPORTED -/* If the bit depth is 8 and the color type is not a palette type expand the - * whole row to 16 bits. Has no effect otherwise. - */ -static void -png_do_expand_16(png_row_infop row_info, png_bytep row) -{ - if (row_info->bit_depth == 8 && - row_info->color_type != PNG_COLOR_TYPE_PALETTE) - { - /* The row have a sequence of bytes containing [0..255] and we need - * to turn it into another row containing [0..65535], to do this we - * calculate: - * - * (input / 255) * 65535 - * - * Which happens to be exactly input * 257 and this can be achieved - * simply by byte replication in place (copying backwards). - */ - png_byte *sp = row + row_info->rowbytes; /* source, last byte + 1 */ - png_byte *dp = sp + row_info->rowbytes; /* destination, end + 1 */ - while (dp > sp) - { - dp[-2] = dp[-1] = *--sp; dp -= 2; - } - - row_info->rowbytes *= 2; - row_info->bit_depth = 16; - row_info->pixel_depth = (png_byte)(row_info->channels * 16); - } -} -#endif - -#ifdef PNG_READ_QUANTIZE_SUPPORTED -static void -png_do_quantize(png_row_infop row_info, png_bytep row, - png_const_bytep palette_lookup, png_const_bytep quantize_lookup) -{ - png_bytep sp, dp; - png_uint_32 i; - png_uint_32 row_width=row_info->width; - - png_debug(1, "in png_do_quantize"); - - if (row_info->bit_depth == 8) - { - if (row_info->color_type == PNG_COLOR_TYPE_RGB && palette_lookup) - { - int r, g, b, p; - sp = row; - dp = row; - for (i = 0; i < row_width; i++) - { - r = *sp++; - g = *sp++; - b = *sp++; - - /* This looks real messy, but the compiler will reduce - * it down to a reasonable formula. For example, with - * 5 bits per color, we get: - * p = (((r >> 3) & 0x1f) << 10) | - * (((g >> 3) & 0x1f) << 5) | - * ((b >> 3) & 0x1f); - */ - p = (((r >> (8 - PNG_QUANTIZE_RED_BITS)) & - ((1 << PNG_QUANTIZE_RED_BITS) - 1)) << - (PNG_QUANTIZE_GREEN_BITS + PNG_QUANTIZE_BLUE_BITS)) | - (((g >> (8 - PNG_QUANTIZE_GREEN_BITS)) & - ((1 << PNG_QUANTIZE_GREEN_BITS) - 1)) << - (PNG_QUANTIZE_BLUE_BITS)) | - ((b >> (8 - PNG_QUANTIZE_BLUE_BITS)) & - ((1 << PNG_QUANTIZE_BLUE_BITS) - 1)); - - *dp++ = palette_lookup[p]; - } - - row_info->color_type = PNG_COLOR_TYPE_PALETTE; - row_info->channels = 1; - row_info->pixel_depth = row_info->bit_depth; - row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth, row_width); - } - - else if (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA && - palette_lookup != NULL) - { - int r, g, b, p; - sp = row; - dp = row; - for (i = 0; i < row_width; i++) - { - r = *sp++; - g = *sp++; - b = *sp++; - sp++; - - p = (((r >> (8 - PNG_QUANTIZE_RED_BITS)) & - ((1 << PNG_QUANTIZE_RED_BITS) - 1)) << - (PNG_QUANTIZE_GREEN_BITS + PNG_QUANTIZE_BLUE_BITS)) | - (((g >> (8 - PNG_QUANTIZE_GREEN_BITS)) & - ((1 << PNG_QUANTIZE_GREEN_BITS) - 1)) << - (PNG_QUANTIZE_BLUE_BITS)) | - ((b >> (8 - PNG_QUANTIZE_BLUE_BITS)) & - ((1 << PNG_QUANTIZE_BLUE_BITS) - 1)); - - *dp++ = palette_lookup[p]; - } - - row_info->color_type = PNG_COLOR_TYPE_PALETTE; - row_info->channels = 1; - row_info->pixel_depth = row_info->bit_depth; - row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth, row_width); - } - - else if (row_info->color_type == PNG_COLOR_TYPE_PALETTE && - quantize_lookup) - { - sp = row; - - for (i = 0; i < row_width; i++, sp++) - { - *sp = quantize_lookup[*sp]; - } - } - } -} -#endif /* READ_QUANTIZE */ - -/* Transform the row. The order of transformations is significant, - * and is very touchy. If you add a transformation, take care to - * decide how it fits in with the other transformations here. - */ -void /* PRIVATE */ -png_do_read_transformations(png_structrp png_ptr, png_row_infop row_info) -{ - png_debug(1, "in png_do_read_transformations"); - - if (png_ptr->row_buf == NULL) - { - /* Prior to 1.5.4 this output row/pass where the NULL pointer is, but this - * error is incredibly rare and incredibly easy to debug without this - * information. - */ - png_error(png_ptr, "NULL row buffer"); - } - - /* The following is debugging; prior to 1.5.4 the code was never compiled in; - * in 1.5.4 PNG_FLAG_DETECT_UNINITIALIZED was added and the macro - * PNG_WARN_UNINITIALIZED_ROW removed. In 1.6 the new flag is set only for - * all transformations, however in practice the ROW_INIT always gets done on - * demand, if necessary. - */ - if ((png_ptr->flags & PNG_FLAG_DETECT_UNINITIALIZED) != 0 && - (png_ptr->flags & PNG_FLAG_ROW_INIT) == 0) - { - /* Application has failed to call either png_read_start_image() or - * png_read_update_info() after setting transforms that expand pixels. - * This check added to libpng-1.2.19 (but not enabled until 1.5.4). - */ - png_error(png_ptr, "Uninitialized row"); - } - -#ifdef PNG_READ_EXPAND_SUPPORTED - if ((png_ptr->transformations & PNG_EXPAND) != 0) - { - if (row_info->color_type == PNG_COLOR_TYPE_PALETTE) - { -#ifdef PNG_ARM_NEON_INTRINSICS_AVAILABLE - if ((png_ptr->num_trans > 0) && (png_ptr->bit_depth == 8)) - { - if (png_ptr->riffled_palette == NULL) - { - /* Initialize the accelerated palette expansion. */ - png_ptr->riffled_palette = - (png_bytep)png_malloc(png_ptr, 256 * 4); - png_riffle_palette_neon(png_ptr); - } - } -#endif - png_do_expand_palette(png_ptr, row_info, png_ptr->row_buf + 1, - png_ptr->palette, png_ptr->trans_alpha, png_ptr->num_trans); - } - - else - { - if (png_ptr->num_trans != 0 && - (png_ptr->transformations & PNG_EXPAND_tRNS) != 0) - png_do_expand(row_info, png_ptr->row_buf + 1, - &(png_ptr->trans_color)); - - else - png_do_expand(row_info, png_ptr->row_buf + 1, NULL); - } - } -#endif - -#ifdef PNG_READ_STRIP_ALPHA_SUPPORTED - if ((png_ptr->transformations & PNG_STRIP_ALPHA) != 0 && - (png_ptr->transformations & PNG_COMPOSE) == 0 && - (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA || - row_info->color_type == PNG_COLOR_TYPE_GRAY_ALPHA)) - png_do_strip_channel(row_info, png_ptr->row_buf + 1, - 0 /* at_start == false, because SWAP_ALPHA happens later */); -#endif - -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED - if ((png_ptr->transformations & PNG_RGB_TO_GRAY) != 0) - { - int rgb_error = - png_do_rgb_to_gray(png_ptr, row_info, - png_ptr->row_buf + 1); - - if (rgb_error != 0) - { - png_ptr->rgb_to_gray_status=1; - if ((png_ptr->transformations & PNG_RGB_TO_GRAY) == - PNG_RGB_TO_GRAY_WARN) - png_warning(png_ptr, "png_do_rgb_to_gray found nongray pixel"); - - if ((png_ptr->transformations & PNG_RGB_TO_GRAY) == - PNG_RGB_TO_GRAY_ERR) - png_error(png_ptr, "png_do_rgb_to_gray found nongray pixel"); - } - } -#endif - -/* From Andreas Dilger e-mail to png-implement, 26 March 1998: - * - * In most cases, the "simple transparency" should be done prior to doing - * gray-to-RGB, or you will have to test 3x as many bytes to check if a - * pixel is transparent. You would also need to make sure that the - * transparency information is upgraded to RGB. - * - * To summarize, the current flow is: - * - Gray + simple transparency -> compare 1 or 2 gray bytes and composite - * with background "in place" if transparent, - * convert to RGB if necessary - * - Gray + alpha -> composite with gray background and remove alpha bytes, - * convert to RGB if necessary - * - * To support RGB backgrounds for gray images we need: - * - Gray + simple transparency -> convert to RGB + simple transparency, - * compare 3 or 6 bytes and composite with - * background "in place" if transparent - * (3x compare/pixel compared to doing - * composite with gray bkgrnd) - * - Gray + alpha -> convert to RGB + alpha, composite with background and - * remove alpha bytes (3x float - * operations/pixel compared with composite - * on gray background) - * - * Greg's change will do this. The reason it wasn't done before is for - * performance, as this increases the per-pixel operations. If we would check - * in advance if the background was gray or RGB, and position the gray-to-RGB - * transform appropriately, then it would save a lot of work/time. - */ - -#ifdef PNG_READ_GRAY_TO_RGB_SUPPORTED - /* If gray -> RGB, do so now only if background is non-gray; else do later - * for performance reasons - */ - if ((png_ptr->transformations & PNG_GRAY_TO_RGB) != 0 && - (png_ptr->mode & PNG_BACKGROUND_IS_GRAY) == 0) - png_do_gray_to_rgb(row_info, png_ptr->row_buf + 1); -#endif - -#if defined(PNG_READ_BACKGROUND_SUPPORTED) ||\ - defined(PNG_READ_ALPHA_MODE_SUPPORTED) - if ((png_ptr->transformations & PNG_COMPOSE) != 0) - png_do_compose(row_info, png_ptr->row_buf + 1, png_ptr); -#endif - -#ifdef PNG_READ_GAMMA_SUPPORTED - if ((png_ptr->transformations & PNG_GAMMA) != 0 && -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED - /* Because RGB_TO_GRAY does the gamma transform. */ - (png_ptr->transformations & PNG_RGB_TO_GRAY) == 0 && -#endif -#if defined(PNG_READ_BACKGROUND_SUPPORTED) ||\ - defined(PNG_READ_ALPHA_MODE_SUPPORTED) - /* Because PNG_COMPOSE does the gamma transform if there is something to - * do (if there is an alpha channel or transparency.) - */ - !((png_ptr->transformations & PNG_COMPOSE) != 0 && - ((png_ptr->num_trans != 0) || - (png_ptr->color_type & PNG_COLOR_MASK_ALPHA) != 0)) && -#endif - /* Because png_init_read_transformations transforms the palette, unless - * RGB_TO_GRAY will do the transform. - */ - (png_ptr->color_type != PNG_COLOR_TYPE_PALETTE)) - png_do_gamma(row_info, png_ptr->row_buf + 1, png_ptr); -#endif - -#ifdef PNG_READ_STRIP_ALPHA_SUPPORTED - if ((png_ptr->transformations & PNG_STRIP_ALPHA) != 0 && - (png_ptr->transformations & PNG_COMPOSE) != 0 && - (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA || - row_info->color_type == PNG_COLOR_TYPE_GRAY_ALPHA)) - png_do_strip_channel(row_info, png_ptr->row_buf + 1, - 0 /* at_start == false, because SWAP_ALPHA happens later */); -#endif - -#ifdef PNG_READ_ALPHA_MODE_SUPPORTED - if ((png_ptr->transformations & PNG_ENCODE_ALPHA) != 0 && - (row_info->color_type & PNG_COLOR_MASK_ALPHA) != 0) - png_do_encode_alpha(row_info, png_ptr->row_buf + 1, png_ptr); -#endif - -#ifdef PNG_READ_SCALE_16_TO_8_SUPPORTED - if ((png_ptr->transformations & PNG_SCALE_16_TO_8) != 0) - png_do_scale_16_to_8(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_READ_STRIP_16_TO_8_SUPPORTED - /* There is no harm in doing both of these because only one has any effect, - * by putting the 'scale' option first if the app asks for scale (either by - * calling the API or in a TRANSFORM flag) this is what happens. - */ - if ((png_ptr->transformations & PNG_16_TO_8) != 0) - png_do_chop(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_READ_QUANTIZE_SUPPORTED - if ((png_ptr->transformations & PNG_QUANTIZE) != 0) - { - png_do_quantize(row_info, png_ptr->row_buf + 1, - png_ptr->palette_lookup, png_ptr->quantize_index); - - if (row_info->rowbytes == 0) - png_error(png_ptr, "png_do_quantize returned rowbytes=0"); - } -#endif /* READ_QUANTIZE */ - -#ifdef PNG_READ_EXPAND_16_SUPPORTED - /* Do the expansion now, after all the arithmetic has been done. Notice - * that previous transformations can handle the PNG_EXPAND_16 flag if this - * is efficient (particularly true in the case of gamma correction, where - * better accuracy results faster!) - */ - if ((png_ptr->transformations & PNG_EXPAND_16) != 0) - png_do_expand_16(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_READ_GRAY_TO_RGB_SUPPORTED - /* NOTE: moved here in 1.5.4 (from much later in this list.) */ - if ((png_ptr->transformations & PNG_GRAY_TO_RGB) != 0 && - (png_ptr->mode & PNG_BACKGROUND_IS_GRAY) != 0) - png_do_gray_to_rgb(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_READ_INVERT_SUPPORTED - if ((png_ptr->transformations & PNG_INVERT_MONO) != 0) - png_do_invert(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_READ_INVERT_ALPHA_SUPPORTED - if ((png_ptr->transformations & PNG_INVERT_ALPHA) != 0) - png_do_read_invert_alpha(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_READ_SHIFT_SUPPORTED - if ((png_ptr->transformations & PNG_SHIFT) != 0) - png_do_unshift(row_info, png_ptr->row_buf + 1, - &(png_ptr->shift)); -#endif - -#ifdef PNG_READ_PACK_SUPPORTED - if ((png_ptr->transformations & PNG_PACK) != 0) - png_do_unpack(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_READ_CHECK_FOR_INVALID_INDEX_SUPPORTED - /* Added at libpng-1.5.10 */ - if (row_info->color_type == PNG_COLOR_TYPE_PALETTE && - png_ptr->num_palette_max >= 0) - png_do_check_palette_indexes(png_ptr, row_info); -#endif - -#ifdef PNG_READ_BGR_SUPPORTED - if ((png_ptr->transformations & PNG_BGR) != 0) - png_do_bgr(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_READ_PACKSWAP_SUPPORTED - if ((png_ptr->transformations & PNG_PACKSWAP) != 0) - png_do_packswap(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_READ_FILLER_SUPPORTED - if ((png_ptr->transformations & PNG_FILLER) != 0) - png_do_read_filler(row_info, png_ptr->row_buf + 1, - (png_uint_32)png_ptr->filler, png_ptr->flags); -#endif - -#ifdef PNG_READ_SWAP_ALPHA_SUPPORTED - if ((png_ptr->transformations & PNG_SWAP_ALPHA) != 0) - png_do_read_swap_alpha(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_READ_16BIT_SUPPORTED -#ifdef PNG_READ_SWAP_SUPPORTED - if ((png_ptr->transformations & PNG_SWAP_BYTES) != 0) - png_do_swap(row_info, png_ptr->row_buf + 1); -#endif -#endif - -#ifdef PNG_READ_USER_TRANSFORM_SUPPORTED - if ((png_ptr->transformations & PNG_USER_TRANSFORM) != 0) - { - if (png_ptr->read_user_transform_fn != NULL) - (*(png_ptr->read_user_transform_fn)) /* User read transform function */ - (png_ptr, /* png_ptr */ - row_info, /* row_info: */ - /* png_uint_32 width; width of row */ - /* size_t rowbytes; number of bytes in row */ - /* png_byte color_type; color type of pixels */ - /* png_byte bit_depth; bit depth of samples */ - /* png_byte channels; number of channels (1-4) */ - /* png_byte pixel_depth; bits per pixel (depth*channels) */ - png_ptr->row_buf + 1); /* start of pixel data for row */ -#ifdef PNG_USER_TRANSFORM_PTR_SUPPORTED - if (png_ptr->user_transform_depth != 0) - row_info->bit_depth = png_ptr->user_transform_depth; - - if (png_ptr->user_transform_channels != 0) - row_info->channels = png_ptr->user_transform_channels; -#endif - row_info->pixel_depth = (png_byte)(row_info->bit_depth * - row_info->channels); - - row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth, row_info->width); - } -#endif -} - -#endif /* READ_TRANSFORMS */ -#endif /* READ */ diff --git a/dep/libpng/src/pngrutil.c b/dep/libpng/src/pngrutil.c deleted file mode 100644 index d31dc21da..000000000 --- a/dep/libpng/src/pngrutil.c +++ /dev/null @@ -1,4680 +0,0 @@ - -/* pngrutil.c - utilities to read a PNG file - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - * - * This file contains routines that are only called from within - * libpng itself during the course of reading an image. - */ - -#include "pngpriv.h" - -#ifdef PNG_READ_SUPPORTED - -png_uint_32 PNGAPI -png_get_uint_31(png_const_structrp png_ptr, png_const_bytep buf) -{ - png_uint_32 uval = png_get_uint_32(buf); - - if (uval > PNG_UINT_31_MAX) - png_error(png_ptr, "PNG unsigned integer out of range"); - - return uval; -} - -#if defined(PNG_READ_gAMA_SUPPORTED) || defined(PNG_READ_cHRM_SUPPORTED) -/* The following is a variation on the above for use with the fixed - * point values used for gAMA and cHRM. Instead of png_error it - * issues a warning and returns (-1) - an invalid value because both - * gAMA and cHRM use *unsigned* integers for fixed point values. - */ -#define PNG_FIXED_ERROR (-1) - -static png_fixed_point /* PRIVATE */ -png_get_fixed_point(png_structrp png_ptr, png_const_bytep buf) -{ - png_uint_32 uval = png_get_uint_32(buf); - - if (uval <= PNG_UINT_31_MAX) - return (png_fixed_point)uval; /* known to be in range */ - - /* The caller can turn off the warning by passing NULL. */ - if (png_ptr != NULL) - png_warning(png_ptr, "PNG fixed point integer out of range"); - - return PNG_FIXED_ERROR; -} -#endif - -#ifdef PNG_READ_INT_FUNCTIONS_SUPPORTED -/* NOTE: the read macros will obscure these definitions, so that if - * PNG_USE_READ_MACROS is set the library will not use them internally, - * but the APIs will still be available externally. - * - * The parentheses around "PNGAPI function_name" in the following three - * functions are necessary because they allow the macros to co-exist with - * these (unused but exported) functions. - */ - -/* Grab an unsigned 32-bit integer from a buffer in big-endian format. */ -png_uint_32 (PNGAPI -png_get_uint_32)(png_const_bytep buf) -{ - png_uint_32 uval = - ((png_uint_32)(*(buf )) << 24) + - ((png_uint_32)(*(buf + 1)) << 16) + - ((png_uint_32)(*(buf + 2)) << 8) + - ((png_uint_32)(*(buf + 3)) ) ; - - return uval; -} - -/* Grab a signed 32-bit integer from a buffer in big-endian format. The - * data is stored in the PNG file in two's complement format and there - * is no guarantee that a 'png_int_32' is exactly 32 bits, therefore - * the following code does a two's complement to native conversion. - */ -png_int_32 (PNGAPI -png_get_int_32)(png_const_bytep buf) -{ - png_uint_32 uval = png_get_uint_32(buf); - if ((uval & 0x80000000) == 0) /* non-negative */ - return (png_int_32)uval; - - uval = (uval ^ 0xffffffff) + 1; /* 2's complement: -x = ~x+1 */ - if ((uval & 0x80000000) == 0) /* no overflow */ - return -(png_int_32)uval; - /* The following has to be safe; this function only gets called on PNG data - * and if we get here that data is invalid. 0 is the most safe value and - * if not then an attacker would surely just generate a PNG with 0 instead. - */ - return 0; -} - -/* Grab an unsigned 16-bit integer from a buffer in big-endian format. */ -png_uint_16 (PNGAPI -png_get_uint_16)(png_const_bytep buf) -{ - /* ANSI-C requires an int value to accommodate at least 16 bits so this - * works and allows the compiler not to worry about possible narrowing - * on 32-bit systems. (Pre-ANSI systems did not make integers smaller - * than 16 bits either.) - */ - unsigned int val = - ((unsigned int)(*buf) << 8) + - ((unsigned int)(*(buf + 1))); - - return (png_uint_16)val; -} - -#endif /* READ_INT_FUNCTIONS */ - -/* Read and check the PNG file signature */ -void /* PRIVATE */ -png_read_sig(png_structrp png_ptr, png_inforp info_ptr) -{ - size_t num_checked, num_to_check; - - /* Exit if the user application does not expect a signature. */ - if (png_ptr->sig_bytes >= 8) - return; - - num_checked = png_ptr->sig_bytes; - num_to_check = 8 - num_checked; - -#ifdef PNG_IO_STATE_SUPPORTED - png_ptr->io_state = PNG_IO_READING | PNG_IO_SIGNATURE; -#endif - - /* The signature must be serialized in a single I/O call. */ - png_read_data(png_ptr, &(info_ptr->signature[num_checked]), num_to_check); - png_ptr->sig_bytes = 8; - - if (png_sig_cmp(info_ptr->signature, num_checked, num_to_check) != 0) - { - if (num_checked < 4 && - png_sig_cmp(info_ptr->signature, num_checked, num_to_check - 4) != 0) - png_error(png_ptr, "Not a PNG file"); - else - png_error(png_ptr, "PNG file corrupted by ASCII conversion"); - } - if (num_checked < 3) - png_ptr->mode |= PNG_HAVE_PNG_SIGNATURE; -} - -/* Read the chunk header (length + type name). - * Put the type name into png_ptr->chunk_name, and return the length. - */ -png_uint_32 /* PRIVATE */ -png_read_chunk_header(png_structrp png_ptr) -{ - png_byte buf[8]; - png_uint_32 length; - -#ifdef PNG_IO_STATE_SUPPORTED - png_ptr->io_state = PNG_IO_READING | PNG_IO_CHUNK_HDR; -#endif - - /* Read the length and the chunk name. - * This must be performed in a single I/O call. - */ - png_read_data(png_ptr, buf, 8); - length = png_get_uint_31(png_ptr, buf); - - /* Put the chunk name into png_ptr->chunk_name. */ - png_ptr->chunk_name = PNG_CHUNK_FROM_STRING(buf+4); - - png_debug2(0, "Reading chunk typeid = 0x%lx, length = %lu", - (unsigned long)png_ptr->chunk_name, (unsigned long)length); - - /* Reset the crc and run it over the chunk name. */ - png_reset_crc(png_ptr); - png_calculate_crc(png_ptr, buf + 4, 4); - - /* Check to see if chunk name is valid. */ - png_check_chunk_name(png_ptr, png_ptr->chunk_name); - - /* Check for too-large chunk length */ - png_check_chunk_length(png_ptr, length); - -#ifdef PNG_IO_STATE_SUPPORTED - png_ptr->io_state = PNG_IO_READING | PNG_IO_CHUNK_DATA; -#endif - - return length; -} - -/* Read data, and (optionally) run it through the CRC. */ -void /* PRIVATE */ -png_crc_read(png_structrp png_ptr, png_bytep buf, png_uint_32 length) -{ - if (png_ptr == NULL) - return; - - png_read_data(png_ptr, buf, length); - png_calculate_crc(png_ptr, buf, length); -} - -/* Optionally skip data and then check the CRC. Depending on whether we - * are reading an ancillary or critical chunk, and how the program has set - * things up, we may calculate the CRC on the data and print a message. - * Returns '1' if there was a CRC error, '0' otherwise. - */ -int /* PRIVATE */ -png_crc_finish(png_structrp png_ptr, png_uint_32 skip) -{ - /* The size of the local buffer for inflate is a good guess as to a - * reasonable size to use for buffering reads from the application. - */ - while (skip > 0) - { - png_uint_32 len; - png_byte tmpbuf[PNG_INFLATE_BUF_SIZE]; - - len = (sizeof tmpbuf); - if (len > skip) - len = skip; - skip -= len; - - png_crc_read(png_ptr, tmpbuf, len); - } - - if (png_crc_error(png_ptr) != 0) - { - if (PNG_CHUNK_ANCILLARY(png_ptr->chunk_name) != 0 ? - (png_ptr->flags & PNG_FLAG_CRC_ANCILLARY_NOWARN) == 0 : - (png_ptr->flags & PNG_FLAG_CRC_CRITICAL_USE) != 0) - { - png_chunk_warning(png_ptr, "CRC error"); - } - - else - png_chunk_error(png_ptr, "CRC error"); - - return 1; - } - - return 0; -} - -/* Compare the CRC stored in the PNG file with that calculated by libpng from - * the data it has read thus far. - */ -int /* PRIVATE */ -png_crc_error(png_structrp png_ptr) -{ - png_byte crc_bytes[4]; - png_uint_32 crc; - int need_crc = 1; - - if (PNG_CHUNK_ANCILLARY(png_ptr->chunk_name) != 0) - { - if ((png_ptr->flags & PNG_FLAG_CRC_ANCILLARY_MASK) == - (PNG_FLAG_CRC_ANCILLARY_USE | PNG_FLAG_CRC_ANCILLARY_NOWARN)) - need_crc = 0; - } - - else /* critical */ - { - if ((png_ptr->flags & PNG_FLAG_CRC_CRITICAL_IGNORE) != 0) - need_crc = 0; - } - -#ifdef PNG_IO_STATE_SUPPORTED - png_ptr->io_state = PNG_IO_READING | PNG_IO_CHUNK_CRC; -#endif - - /* The chunk CRC must be serialized in a single I/O call. */ - png_read_data(png_ptr, crc_bytes, 4); - - if (need_crc != 0) - { - crc = png_get_uint_32(crc_bytes); - return crc != png_ptr->crc; - } - - else - return 0; -} - -#if defined(PNG_READ_iCCP_SUPPORTED) || defined(PNG_READ_iTXt_SUPPORTED) ||\ - defined(PNG_READ_pCAL_SUPPORTED) || defined(PNG_READ_sCAL_SUPPORTED) ||\ - defined(PNG_READ_sPLT_SUPPORTED) || defined(PNG_READ_tEXt_SUPPORTED) ||\ - defined(PNG_READ_zTXt_SUPPORTED) || defined(PNG_SEQUENTIAL_READ_SUPPORTED) -/* Manage the read buffer; this simply reallocates the buffer if it is not small - * enough (or if it is not allocated). The routine returns a pointer to the - * buffer; if an error occurs and 'warn' is set the routine returns NULL, else - * it will call png_error (via png_malloc) on failure. (warn == 2 means - * 'silent'). - */ -static png_bytep -png_read_buffer(png_structrp png_ptr, png_alloc_size_t new_size, int warn) -{ - png_bytep buffer = png_ptr->read_buffer; - - if (buffer != NULL && new_size > png_ptr->read_buffer_size) - { - png_ptr->read_buffer = NULL; - png_ptr->read_buffer_size = 0; - png_free(png_ptr, buffer); - buffer = NULL; - } - - if (buffer == NULL) - { - buffer = png_voidcast(png_bytep, png_malloc_base(png_ptr, new_size)); - - if (buffer != NULL) - { - memset(buffer, 0, new_size); /* just in case */ - png_ptr->read_buffer = buffer; - png_ptr->read_buffer_size = new_size; - } - - else if (warn < 2) /* else silent */ - { - if (warn != 0) - png_chunk_warning(png_ptr, "insufficient memory to read chunk"); - - else - png_chunk_error(png_ptr, "insufficient memory to read chunk"); - } - } - - return buffer; -} -#endif /* READ_iCCP|iTXt|pCAL|sCAL|sPLT|tEXt|zTXt|SEQUENTIAL_READ */ - -/* png_inflate_claim: claim the zstream for some nefarious purpose that involves - * decompression. Returns Z_OK on success, else a zlib error code. It checks - * the owner but, in final release builds, just issues a warning if some other - * chunk apparently owns the stream. Prior to release it does a png_error. - */ -static int -png_inflate_claim(png_structrp png_ptr, png_uint_32 owner) -{ - if (png_ptr->zowner != 0) - { - char msg[64]; - - PNG_STRING_FROM_CHUNK(msg, png_ptr->zowner); - /* So the message that results is " using zstream"; this is an - * internal error, but is very useful for debugging. i18n requirements - * are minimal. - */ - (void)png_safecat(msg, (sizeof msg), 4, " using zstream"); -#if PNG_RELEASE_BUILD - png_chunk_warning(png_ptr, msg); - png_ptr->zowner = 0; -#else - png_chunk_error(png_ptr, msg); -#endif - } - - /* Implementation note: unlike 'png_deflate_claim' this internal function - * does not take the size of the data as an argument. Some efficiency could - * be gained by using this when it is known *if* the zlib stream itself does - * not record the number; however, this is an illusion: the original writer - * of the PNG may have selected a lower window size, and we really must - * follow that because, for systems with with limited capabilities, we - * would otherwise reject the application's attempts to use a smaller window - * size (zlib doesn't have an interface to say "this or lower"!). - * - * inflateReset2 was added to zlib 1.2.4; before this the window could not be - * reset, therefore it is necessary to always allocate the maximum window - * size with earlier zlibs just in case later compressed chunks need it. - */ - { - int ret; /* zlib return code */ -#if ZLIB_VERNUM >= 0x1240 - int window_bits = 0; - -# if defined(PNG_SET_OPTION_SUPPORTED) && defined(PNG_MAXIMUM_INFLATE_WINDOW) - if (((png_ptr->options >> PNG_MAXIMUM_INFLATE_WINDOW) & 3) == - PNG_OPTION_ON) - { - window_bits = 15; - png_ptr->zstream_start = 0; /* fixed window size */ - } - - else - { - png_ptr->zstream_start = 1; - } -# endif - -#endif /* ZLIB_VERNUM >= 0x1240 */ - - /* Set this for safety, just in case the previous owner left pointers to - * memory allocations. - */ - png_ptr->zstream.next_in = NULL; - png_ptr->zstream.avail_in = 0; - png_ptr->zstream.next_out = NULL; - png_ptr->zstream.avail_out = 0; - - if ((png_ptr->flags & PNG_FLAG_ZSTREAM_INITIALIZED) != 0) - { -#if ZLIB_VERNUM >= 0x1240 - ret = inflateReset2(&png_ptr->zstream, window_bits); -#else - ret = inflateReset(&png_ptr->zstream); -#endif - } - - else - { -#if ZLIB_VERNUM >= 0x1240 - ret = inflateInit2(&png_ptr->zstream, window_bits); -#else - ret = inflateInit(&png_ptr->zstream); -#endif - - if (ret == Z_OK) - png_ptr->flags |= PNG_FLAG_ZSTREAM_INITIALIZED; - } - -#ifdef PNG_DISABLE_ADLER32_CHECK_SUPPORTED - if (((png_ptr->options >> PNG_IGNORE_ADLER32) & 3) == PNG_OPTION_ON) - /* Turn off validation of the ADLER32 checksum in IDAT chunks */ - ret = inflateValidate(&png_ptr->zstream, 0); -#endif - - if (ret == Z_OK) - png_ptr->zowner = owner; - - else - png_zstream_error(png_ptr, ret); - - return ret; - } - -#ifdef window_bits -# undef window_bits -#endif -} - -#if ZLIB_VERNUM >= 0x1240 -/* Handle the start of the inflate stream if we called inflateInit2(strm,0); - * in this case some zlib versions skip validation of the CINFO field and, in - * certain circumstances, libpng may end up displaying an invalid image, in - * contrast to implementations that call zlib in the normal way (e.g. libpng - * 1.5). - */ -int /* PRIVATE */ -png_zlib_inflate(png_structrp png_ptr, int flush) -{ - if (png_ptr->zstream_start && png_ptr->zstream.avail_in > 0) - { - if ((*png_ptr->zstream.next_in >> 4) > 7) - { - png_ptr->zstream.msg = "invalid window size (libpng)"; - return Z_DATA_ERROR; - } - - png_ptr->zstream_start = 0; - } - - return inflate(&png_ptr->zstream, flush); -} -#endif /* Zlib >= 1.2.4 */ - -#ifdef PNG_READ_COMPRESSED_TEXT_SUPPORTED -#if defined(PNG_READ_zTXt_SUPPORTED) || defined (PNG_READ_iTXt_SUPPORTED) -/* png_inflate now returns zlib error codes including Z_OK and Z_STREAM_END to - * allow the caller to do multiple calls if required. If the 'finish' flag is - * set Z_FINISH will be passed to the final inflate() call and Z_STREAM_END must - * be returned or there has been a problem, otherwise Z_SYNC_FLUSH is used and - * Z_OK or Z_STREAM_END will be returned on success. - * - * The input and output sizes are updated to the actual amounts of data consumed - * or written, not the amount available (as in a z_stream). The data pointers - * are not changed, so the next input is (data+input_size) and the next - * available output is (output+output_size). - */ -static int -png_inflate(png_structrp png_ptr, png_uint_32 owner, int finish, - /* INPUT: */ png_const_bytep input, png_uint_32p input_size_ptr, - /* OUTPUT: */ png_bytep output, png_alloc_size_t *output_size_ptr) -{ - if (png_ptr->zowner == owner) /* Else not claimed */ - { - int ret; - png_alloc_size_t avail_out = *output_size_ptr; - png_uint_32 avail_in = *input_size_ptr; - - /* zlib can't necessarily handle more than 65535 bytes at once (i.e. it - * can't even necessarily handle 65536 bytes) because the type uInt is - * "16 bits or more". Consequently it is necessary to chunk the input to - * zlib. This code uses ZLIB_IO_MAX, from pngpriv.h, as the maximum (the - * maximum value that can be stored in a uInt.) It is possible to set - * ZLIB_IO_MAX to a lower value in pngpriv.h and this may sometimes have - * a performance advantage, because it reduces the amount of data accessed - * at each step and that may give the OS more time to page it in. - */ - png_ptr->zstream.next_in = PNGZ_INPUT_CAST(input); - /* avail_in and avail_out are set below from 'size' */ - png_ptr->zstream.avail_in = 0; - png_ptr->zstream.avail_out = 0; - - /* Read directly into the output if it is available (this is set to - * a local buffer below if output is NULL). - */ - if (output != NULL) - png_ptr->zstream.next_out = output; - - do - { - uInt avail; - Byte local_buffer[PNG_INFLATE_BUF_SIZE]; - - /* zlib INPUT BUFFER */ - /* The setting of 'avail_in' used to be outside the loop; by setting it - * inside it is possible to chunk the input to zlib and simply rely on - * zlib to advance the 'next_in' pointer. This allows arbitrary - * amounts of data to be passed through zlib at the unavoidable cost of - * requiring a window save (memcpy of up to 32768 output bytes) - * every ZLIB_IO_MAX input bytes. - */ - avail_in += png_ptr->zstream.avail_in; /* not consumed last time */ - - avail = ZLIB_IO_MAX; - - if (avail_in < avail) - avail = (uInt)avail_in; /* safe: < than ZLIB_IO_MAX */ - - avail_in -= avail; - png_ptr->zstream.avail_in = avail; - - /* zlib OUTPUT BUFFER */ - avail_out += png_ptr->zstream.avail_out; /* not written last time */ - - avail = ZLIB_IO_MAX; /* maximum zlib can process */ - - if (output == NULL) - { - /* Reset the output buffer each time round if output is NULL and - * make available the full buffer, up to 'remaining_space' - */ - png_ptr->zstream.next_out = local_buffer; - if ((sizeof local_buffer) < avail) - avail = (sizeof local_buffer); - } - - if (avail_out < avail) - avail = (uInt)avail_out; /* safe: < ZLIB_IO_MAX */ - - png_ptr->zstream.avail_out = avail; - avail_out -= avail; - - /* zlib inflate call */ - /* In fact 'avail_out' may be 0 at this point, that happens at the end - * of the read when the final LZ end code was not passed at the end of - * the previous chunk of input data. Tell zlib if we have reached the - * end of the output buffer. - */ - ret = PNG_INFLATE(png_ptr, avail_out > 0 ? Z_NO_FLUSH : - (finish ? Z_FINISH : Z_SYNC_FLUSH)); - } while (ret == Z_OK); - - /* For safety kill the local buffer pointer now */ - if (output == NULL) - png_ptr->zstream.next_out = NULL; - - /* Claw back the 'size' and 'remaining_space' byte counts. */ - avail_in += png_ptr->zstream.avail_in; - avail_out += png_ptr->zstream.avail_out; - - /* Update the input and output sizes; the updated values are the amount - * consumed or written, effectively the inverse of what zlib uses. - */ - if (avail_out > 0) - *output_size_ptr -= avail_out; - - if (avail_in > 0) - *input_size_ptr -= avail_in; - - /* Ensure png_ptr->zstream.msg is set (even in the success case!) */ - png_zstream_error(png_ptr, ret); - return ret; - } - - else - { - /* This is a bad internal error. The recovery assigns to the zstream msg - * pointer, which is not owned by the caller, but this is safe; it's only - * used on errors! - */ - png_ptr->zstream.msg = PNGZ_MSG_CAST("zstream unclaimed"); - return Z_STREAM_ERROR; - } -} - -/* - * Decompress trailing data in a chunk. The assumption is that read_buffer - * points at an allocated area holding the contents of a chunk with a - * trailing compressed part. What we get back is an allocated area - * holding the original prefix part and an uncompressed version of the - * trailing part (the malloc area passed in is freed). - */ -static int -png_decompress_chunk(png_structrp png_ptr, - png_uint_32 chunklength, png_uint_32 prefix_size, - png_alloc_size_t *newlength /* must be initialized to the maximum! */, - int terminate /*add a '\0' to the end of the uncompressed data*/) -{ - /* TODO: implement different limits for different types of chunk. - * - * The caller supplies *newlength set to the maximum length of the - * uncompressed data, but this routine allocates space for the prefix and - * maybe a '\0' terminator too. We have to assume that 'prefix_size' is - * limited only by the maximum chunk size. - */ - png_alloc_size_t limit = PNG_SIZE_MAX; - -# ifdef PNG_SET_USER_LIMITS_SUPPORTED - if (png_ptr->user_chunk_malloc_max > 0 && - png_ptr->user_chunk_malloc_max < limit) - limit = png_ptr->user_chunk_malloc_max; -# elif PNG_USER_CHUNK_MALLOC_MAX > 0 - if (PNG_USER_CHUNK_MALLOC_MAX < limit) - limit = PNG_USER_CHUNK_MALLOC_MAX; -# endif - - if (limit >= prefix_size + (terminate != 0)) - { - int ret; - - limit -= prefix_size + (terminate != 0); - - if (limit < *newlength) - *newlength = limit; - - /* Now try to claim the stream. */ - ret = png_inflate_claim(png_ptr, png_ptr->chunk_name); - - if (ret == Z_OK) - { - png_uint_32 lzsize = chunklength - prefix_size; - - ret = png_inflate(png_ptr, png_ptr->chunk_name, 1/*finish*/, - /* input: */ png_ptr->read_buffer + prefix_size, &lzsize, - /* output: */ NULL, newlength); - - if (ret == Z_STREAM_END) - { - /* Use 'inflateReset' here, not 'inflateReset2' because this - * preserves the previously decided window size (otherwise it would - * be necessary to store the previous window size.) In practice - * this doesn't matter anyway, because png_inflate will call inflate - * with Z_FINISH in almost all cases, so the window will not be - * maintained. - */ - if (inflateReset(&png_ptr->zstream) == Z_OK) - { - /* Because of the limit checks above we know that the new, - * expanded, size will fit in a size_t (let alone an - * png_alloc_size_t). Use png_malloc_base here to avoid an - * extra OOM message. - */ - png_alloc_size_t new_size = *newlength; - png_alloc_size_t buffer_size = prefix_size + new_size + - (terminate != 0); - png_bytep text = png_voidcast(png_bytep, png_malloc_base(png_ptr, - buffer_size)); - - if (text != NULL) - { - memset(text, 0, buffer_size); - - ret = png_inflate(png_ptr, png_ptr->chunk_name, 1/*finish*/, - png_ptr->read_buffer + prefix_size, &lzsize, - text + prefix_size, newlength); - - if (ret == Z_STREAM_END) - { - if (new_size == *newlength) - { - if (terminate != 0) - text[prefix_size + *newlength] = 0; - - if (prefix_size > 0) - memcpy(text, png_ptr->read_buffer, prefix_size); - - { - png_bytep old_ptr = png_ptr->read_buffer; - - png_ptr->read_buffer = text; - png_ptr->read_buffer_size = buffer_size; - text = old_ptr; /* freed below */ - } - } - - else - { - /* The size changed on the second read, there can be no - * guarantee that anything is correct at this point. - * The 'msg' pointer has been set to "unexpected end of - * LZ stream", which is fine, but return an error code - * that the caller won't accept. - */ - ret = PNG_UNEXPECTED_ZLIB_RETURN; - } - } - - else if (ret == Z_OK) - ret = PNG_UNEXPECTED_ZLIB_RETURN; /* for safety */ - - /* Free the text pointer (this is the old read_buffer on - * success) - */ - png_free(png_ptr, text); - - /* This really is very benign, but it's still an error because - * the extra space may otherwise be used as a Trojan Horse. - */ - if (ret == Z_STREAM_END && - chunklength - prefix_size != lzsize) - png_chunk_benign_error(png_ptr, "extra compressed data"); - } - - else - { - /* Out of memory allocating the buffer */ - ret = Z_MEM_ERROR; - png_zstream_error(png_ptr, Z_MEM_ERROR); - } - } - - else - { - /* inflateReset failed, store the error message */ - png_zstream_error(png_ptr, ret); - ret = PNG_UNEXPECTED_ZLIB_RETURN; - } - } - - else if (ret == Z_OK) - ret = PNG_UNEXPECTED_ZLIB_RETURN; - - /* Release the claimed stream */ - png_ptr->zowner = 0; - } - - else /* the claim failed */ if (ret == Z_STREAM_END) /* impossible! */ - ret = PNG_UNEXPECTED_ZLIB_RETURN; - - return ret; - } - - else - { - /* Application/configuration limits exceeded */ - png_zstream_error(png_ptr, Z_MEM_ERROR); - return Z_MEM_ERROR; - } -} -#endif /* READ_zTXt || READ_iTXt */ -#endif /* READ_COMPRESSED_TEXT */ - -#ifdef PNG_READ_iCCP_SUPPORTED -/* Perform a partial read and decompress, producing 'avail_out' bytes and - * reading from the current chunk as required. - */ -static int -png_inflate_read(png_structrp png_ptr, png_bytep read_buffer, uInt read_size, - png_uint_32p chunk_bytes, png_bytep next_out, png_alloc_size_t *out_size, - int finish) -{ - if (png_ptr->zowner == png_ptr->chunk_name) - { - int ret; - - /* next_in and avail_in must have been initialized by the caller. */ - png_ptr->zstream.next_out = next_out; - png_ptr->zstream.avail_out = 0; /* set in the loop */ - - do - { - if (png_ptr->zstream.avail_in == 0) - { - if (read_size > *chunk_bytes) - read_size = (uInt)*chunk_bytes; - *chunk_bytes -= read_size; - - if (read_size > 0) - png_crc_read(png_ptr, read_buffer, read_size); - - png_ptr->zstream.next_in = read_buffer; - png_ptr->zstream.avail_in = read_size; - } - - if (png_ptr->zstream.avail_out == 0) - { - uInt avail = ZLIB_IO_MAX; - if (avail > *out_size) - avail = (uInt)*out_size; - *out_size -= avail; - - png_ptr->zstream.avail_out = avail; - } - - /* Use Z_SYNC_FLUSH when there is no more chunk data to ensure that all - * the available output is produced; this allows reading of truncated - * streams. - */ - ret = PNG_INFLATE(png_ptr, *chunk_bytes > 0 ? - Z_NO_FLUSH : (finish ? Z_FINISH : Z_SYNC_FLUSH)); - } - while (ret == Z_OK && (*out_size > 0 || png_ptr->zstream.avail_out > 0)); - - *out_size += png_ptr->zstream.avail_out; - png_ptr->zstream.avail_out = 0; /* Should not be required, but is safe */ - - /* Ensure the error message pointer is always set: */ - png_zstream_error(png_ptr, ret); - return ret; - } - - else - { - png_ptr->zstream.msg = PNGZ_MSG_CAST("zstream unclaimed"); - return Z_STREAM_ERROR; - } -} -#endif /* READ_iCCP */ - -/* Read and check the IDHR chunk */ - -void /* PRIVATE */ -png_handle_IHDR(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_byte buf[13]; - png_uint_32 width, height; - int bit_depth, color_type, compression_type, filter_type; - int interlace_type; - - png_debug(1, "in png_handle_IHDR"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) != 0) - png_chunk_error(png_ptr, "out of place"); - - /* Check the length */ - if (length != 13) - png_chunk_error(png_ptr, "invalid"); - - png_ptr->mode |= PNG_HAVE_IHDR; - - png_crc_read(png_ptr, buf, 13); - png_crc_finish(png_ptr, 0); - - width = png_get_uint_31(png_ptr, buf); - height = png_get_uint_31(png_ptr, buf + 4); - bit_depth = buf[8]; - color_type = buf[9]; - compression_type = buf[10]; - filter_type = buf[11]; - interlace_type = buf[12]; - - /* Set internal variables */ - png_ptr->width = width; - png_ptr->height = height; - png_ptr->bit_depth = (png_byte)bit_depth; - png_ptr->interlaced = (png_byte)interlace_type; - png_ptr->color_type = (png_byte)color_type; -#ifdef PNG_MNG_FEATURES_SUPPORTED - png_ptr->filter_type = (png_byte)filter_type; -#endif - png_ptr->compression_type = (png_byte)compression_type; - - /* Find number of channels */ - switch (png_ptr->color_type) - { - default: /* invalid, png_set_IHDR calls png_error */ - case PNG_COLOR_TYPE_GRAY: - case PNG_COLOR_TYPE_PALETTE: - png_ptr->channels = 1; - break; - - case PNG_COLOR_TYPE_RGB: - png_ptr->channels = 3; - break; - - case PNG_COLOR_TYPE_GRAY_ALPHA: - png_ptr->channels = 2; - break; - - case PNG_COLOR_TYPE_RGB_ALPHA: - png_ptr->channels = 4; - break; - } - - /* Set up other useful info */ - png_ptr->pixel_depth = (png_byte)(png_ptr->bit_depth * png_ptr->channels); - png_ptr->rowbytes = PNG_ROWBYTES(png_ptr->pixel_depth, png_ptr->width); - png_debug1(3, "bit_depth = %d", png_ptr->bit_depth); - png_debug1(3, "channels = %d", png_ptr->channels); - png_debug1(3, "rowbytes = %lu", (unsigned long)png_ptr->rowbytes); - png_set_IHDR(png_ptr, info_ptr, width, height, bit_depth, - color_type, interlace_type, compression_type, filter_type); -} - -/* Read and check the palette */ -void /* PRIVATE */ -png_handle_PLTE(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_color palette[PNG_MAX_PALETTE_LENGTH]; - int max_palette_length, num, i; -#ifdef PNG_POINTER_INDEXING_SUPPORTED - png_colorp pal_ptr; -#endif - - png_debug(1, "in png_handle_PLTE"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - /* Moved to before the 'after IDAT' check below because otherwise duplicate - * PLTE chunks are potentially ignored (the spec says there shall not be more - * than one PLTE, the error is not treated as benign, so this check trumps - * the requirement that PLTE appears before IDAT.) - */ - else if ((png_ptr->mode & PNG_HAVE_PLTE) != 0) - png_chunk_error(png_ptr, "duplicate"); - - else if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - { - /* This is benign because the non-benign error happened before, when an - * IDAT was encountered in a color-mapped image with no PLTE. - */ - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - png_ptr->mode |= PNG_HAVE_PLTE; - - if ((png_ptr->color_type & PNG_COLOR_MASK_COLOR) == 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "ignored in grayscale PNG"); - return; - } - -#ifndef PNG_READ_OPT_PLTE_SUPPORTED - if (png_ptr->color_type != PNG_COLOR_TYPE_PALETTE) - { - png_crc_finish(png_ptr, length); - return; - } -#endif - - if (length > 3*PNG_MAX_PALETTE_LENGTH || length % 3) - { - png_crc_finish(png_ptr, length); - - if (png_ptr->color_type != PNG_COLOR_TYPE_PALETTE) - png_chunk_benign_error(png_ptr, "invalid"); - - else - png_chunk_error(png_ptr, "invalid"); - - return; - } - - /* The cast is safe because 'length' is less than 3*PNG_MAX_PALETTE_LENGTH */ - num = (int)length / 3; - - /* If the palette has 256 or fewer entries but is too large for the bit - * depth, we don't issue an error, to preserve the behavior of previous - * libpng versions. We silently truncate the unused extra palette entries - * here. - */ - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - max_palette_length = (1 << png_ptr->bit_depth); - else - max_palette_length = PNG_MAX_PALETTE_LENGTH; - - if (num > max_palette_length) - num = max_palette_length; - -#ifdef PNG_POINTER_INDEXING_SUPPORTED - for (i = 0, pal_ptr = palette; i < num; i++, pal_ptr++) - { - png_byte buf[3]; - - png_crc_read(png_ptr, buf, 3); - pal_ptr->red = buf[0]; - pal_ptr->green = buf[1]; - pal_ptr->blue = buf[2]; - } -#else - for (i = 0; i < num; i++) - { - png_byte buf[3]; - - png_crc_read(png_ptr, buf, 3); - /* Don't depend upon png_color being any order */ - palette[i].red = buf[0]; - palette[i].green = buf[1]; - palette[i].blue = buf[2]; - } -#endif - - /* If we actually need the PLTE chunk (ie for a paletted image), we do - * whatever the normal CRC configuration tells us. However, if we - * have an RGB image, the PLTE can be considered ancillary, so - * we will act as though it is. - */ -#ifndef PNG_READ_OPT_PLTE_SUPPORTED - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) -#endif - { - png_crc_finish(png_ptr, (png_uint_32) (length - (unsigned int)num * 3)); - } - -#ifndef PNG_READ_OPT_PLTE_SUPPORTED - else if (png_crc_error(png_ptr) != 0) /* Only if we have a CRC error */ - { - /* If we don't want to use the data from an ancillary chunk, - * we have two options: an error abort, or a warning and we - * ignore the data in this chunk (which should be OK, since - * it's considered ancillary for a RGB or RGBA image). - * - * IMPLEMENTATION NOTE: this is only here because png_crc_finish uses the - * chunk type to determine whether to check the ancillary or the critical - * flags. - */ - if ((png_ptr->flags & PNG_FLAG_CRC_ANCILLARY_USE) == 0) - { - if ((png_ptr->flags & PNG_FLAG_CRC_ANCILLARY_NOWARN) != 0) - return; - - else - png_chunk_error(png_ptr, "CRC error"); - } - - /* Otherwise, we (optionally) emit a warning and use the chunk. */ - else if ((png_ptr->flags & PNG_FLAG_CRC_ANCILLARY_NOWARN) == 0) - png_chunk_warning(png_ptr, "CRC error"); - } -#endif - - /* TODO: png_set_PLTE has the side effect of setting png_ptr->palette to its - * own copy of the palette. This has the side effect that when png_start_row - * is called (this happens after any call to png_read_update_info) the - * info_ptr palette gets changed. This is extremely unexpected and - * confusing. - * - * Fix this by not sharing the palette in this way. - */ - png_set_PLTE(png_ptr, info_ptr, palette, num); - - /* The three chunks, bKGD, hIST and tRNS *must* appear after PLTE and before - * IDAT. Prior to 1.6.0 this was not checked; instead the code merely - * checked the apparent validity of a tRNS chunk inserted before PLTE on a - * palette PNG. 1.6.0 attempts to rigorously follow the standard and - * therefore does a benign error if the erroneous condition is detected *and* - * cancels the tRNS if the benign error returns. The alternative is to - * amend the standard since it would be rather hypocritical of the standards - * maintainers to ignore it. - */ -#ifdef PNG_READ_tRNS_SUPPORTED - if (png_ptr->num_trans > 0 || - (info_ptr != NULL && (info_ptr->valid & PNG_INFO_tRNS) != 0)) - { - /* Cancel this because otherwise it would be used if the transforms - * require it. Don't cancel the 'valid' flag because this would prevent - * detection of duplicate chunks. - */ - png_ptr->num_trans = 0; - - if (info_ptr != NULL) - info_ptr->num_trans = 0; - - png_chunk_benign_error(png_ptr, "tRNS must be after"); - } -#endif - -#ifdef PNG_READ_hIST_SUPPORTED - if (info_ptr != NULL && (info_ptr->valid & PNG_INFO_hIST) != 0) - png_chunk_benign_error(png_ptr, "hIST must be after"); -#endif - -#ifdef PNG_READ_bKGD_SUPPORTED - if (info_ptr != NULL && (info_ptr->valid & PNG_INFO_bKGD) != 0) - png_chunk_benign_error(png_ptr, "bKGD must be after"); -#endif -} - -void /* PRIVATE */ -png_handle_IEND(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_debug(1, "in png_handle_IEND"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0 || - (png_ptr->mode & PNG_HAVE_IDAT) == 0) - png_chunk_error(png_ptr, "out of place"); - - png_ptr->mode |= (PNG_AFTER_IDAT | PNG_HAVE_IEND); - - png_crc_finish(png_ptr, length); - - if (length != 0) - png_chunk_benign_error(png_ptr, "invalid"); - - PNG_UNUSED(info_ptr) -} - -#ifdef PNG_READ_gAMA_SUPPORTED -void /* PRIVATE */ -png_handle_gAMA(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_fixed_point igamma; - png_byte buf[4]; - - png_debug(1, "in png_handle_gAMA"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & (PNG_HAVE_IDAT|PNG_HAVE_PLTE)) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - if (length != 4) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - png_crc_read(png_ptr, buf, 4); - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - igamma = png_get_fixed_point(NULL, buf); - - png_colorspace_set_gamma(png_ptr, &png_ptr->colorspace, igamma); - png_colorspace_sync(png_ptr, info_ptr); -} -#endif - -#ifdef PNG_READ_sBIT_SUPPORTED -void /* PRIVATE */ -png_handle_sBIT(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - unsigned int truelen, i; - png_byte sample_depth; - png_byte buf[4]; - - png_debug(1, "in png_handle_sBIT"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & (PNG_HAVE_IDAT|PNG_HAVE_PLTE)) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - if (info_ptr != NULL && (info_ptr->valid & PNG_INFO_sBIT) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "duplicate"); - return; - } - - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - { - truelen = 3; - sample_depth = 8; - } - - else - { - truelen = png_ptr->channels; - sample_depth = png_ptr->bit_depth; - } - - if (length != truelen || length > 4) - { - png_chunk_benign_error(png_ptr, "invalid"); - png_crc_finish(png_ptr, length); - return; - } - - buf[0] = buf[1] = buf[2] = buf[3] = sample_depth; - png_crc_read(png_ptr, buf, truelen); - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - for (i=0; i sample_depth) - { - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - } - - if ((png_ptr->color_type & PNG_COLOR_MASK_COLOR) != 0) - { - png_ptr->sig_bit.red = buf[0]; - png_ptr->sig_bit.green = buf[1]; - png_ptr->sig_bit.blue = buf[2]; - png_ptr->sig_bit.alpha = buf[3]; - } - - else - { - png_ptr->sig_bit.gray = buf[0]; - png_ptr->sig_bit.red = buf[0]; - png_ptr->sig_bit.green = buf[0]; - png_ptr->sig_bit.blue = buf[0]; - png_ptr->sig_bit.alpha = buf[1]; - } - - png_set_sBIT(png_ptr, info_ptr, &(png_ptr->sig_bit)); -} -#endif - -#ifdef PNG_READ_cHRM_SUPPORTED -void /* PRIVATE */ -png_handle_cHRM(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_byte buf[32]; - png_xy xy; - - png_debug(1, "in png_handle_cHRM"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & (PNG_HAVE_IDAT|PNG_HAVE_PLTE)) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - if (length != 32) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - png_crc_read(png_ptr, buf, 32); - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - xy.whitex = png_get_fixed_point(NULL, buf); - xy.whitey = png_get_fixed_point(NULL, buf + 4); - xy.redx = png_get_fixed_point(NULL, buf + 8); - xy.redy = png_get_fixed_point(NULL, buf + 12); - xy.greenx = png_get_fixed_point(NULL, buf + 16); - xy.greeny = png_get_fixed_point(NULL, buf + 20); - xy.bluex = png_get_fixed_point(NULL, buf + 24); - xy.bluey = png_get_fixed_point(NULL, buf + 28); - - if (xy.whitex == PNG_FIXED_ERROR || - xy.whitey == PNG_FIXED_ERROR || - xy.redx == PNG_FIXED_ERROR || - xy.redy == PNG_FIXED_ERROR || - xy.greenx == PNG_FIXED_ERROR || - xy.greeny == PNG_FIXED_ERROR || - xy.bluex == PNG_FIXED_ERROR || - xy.bluey == PNG_FIXED_ERROR) - { - png_chunk_benign_error(png_ptr, "invalid values"); - return; - } - - /* If a colorspace error has already been output skip this chunk */ - if ((png_ptr->colorspace.flags & PNG_COLORSPACE_INVALID) != 0) - return; - - if ((png_ptr->colorspace.flags & PNG_COLORSPACE_FROM_cHRM) != 0) - { - png_ptr->colorspace.flags |= PNG_COLORSPACE_INVALID; - png_colorspace_sync(png_ptr, info_ptr); - png_chunk_benign_error(png_ptr, "duplicate"); - return; - } - - png_ptr->colorspace.flags |= PNG_COLORSPACE_FROM_cHRM; - (void)png_colorspace_set_chromaticities(png_ptr, &png_ptr->colorspace, &xy, - 1/*prefer cHRM values*/); - png_colorspace_sync(png_ptr, info_ptr); -} -#endif - -#ifdef PNG_READ_sRGB_SUPPORTED -void /* PRIVATE */ -png_handle_sRGB(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_byte intent; - - png_debug(1, "in png_handle_sRGB"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & (PNG_HAVE_IDAT|PNG_HAVE_PLTE)) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - if (length != 1) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - png_crc_read(png_ptr, &intent, 1); - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - /* If a colorspace error has already been output skip this chunk */ - if ((png_ptr->colorspace.flags & PNG_COLORSPACE_INVALID) != 0) - return; - - /* Only one sRGB or iCCP chunk is allowed, use the HAVE_INTENT flag to detect - * this. - */ - if ((png_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_INTENT) != 0) - { - png_ptr->colorspace.flags |= PNG_COLORSPACE_INVALID; - png_colorspace_sync(png_ptr, info_ptr); - png_chunk_benign_error(png_ptr, "too many profiles"); - return; - } - - (void)png_colorspace_set_sRGB(png_ptr, &png_ptr->colorspace, intent); - png_colorspace_sync(png_ptr, info_ptr); -} -#endif /* READ_sRGB */ - -#ifdef PNG_READ_iCCP_SUPPORTED -void /* PRIVATE */ -png_handle_iCCP(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -/* Note: this does not properly handle profiles that are > 64K under DOS */ -{ - png_const_charp errmsg = NULL; /* error message output, or no error */ - int finished = 0; /* crc checked */ - - png_debug(1, "in png_handle_iCCP"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & (PNG_HAVE_IDAT|PNG_HAVE_PLTE)) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - /* Consistent with all the above colorspace handling an obviously *invalid* - * chunk is just ignored, so does not invalidate the color space. An - * alternative is to set the 'invalid' flags at the start of this routine - * and only clear them in they were not set before and all the tests pass. - */ - - /* The keyword must be at least one character and there is a - * terminator (0) byte and the compression method byte, and the - * 'zlib' datastream is at least 11 bytes. - */ - if (length < 14) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "too short"); - return; - } - - /* If a colorspace error has already been output skip this chunk */ - if ((png_ptr->colorspace.flags & PNG_COLORSPACE_INVALID) != 0) - { - png_crc_finish(png_ptr, length); - return; - } - - /* Only one sRGB or iCCP chunk is allowed, use the HAVE_INTENT flag to detect - * this. - */ - if ((png_ptr->colorspace.flags & PNG_COLORSPACE_HAVE_INTENT) == 0) - { - uInt read_length, keyword_length; - char keyword[81]; - - /* Find the keyword; the keyword plus separator and compression method - * bytes can be at most 81 characters long. - */ - read_length = 81; /* maximum */ - if (read_length > length) - read_length = (uInt)length; - - png_crc_read(png_ptr, (png_bytep)keyword, read_length); - length -= read_length; - - /* The minimum 'zlib' stream is assumed to be just the 2 byte header, - * 5 bytes minimum 'deflate' stream, and the 4 byte checksum. - */ - if (length < 11) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "too short"); - return; - } - - keyword_length = 0; - while (keyword_length < 80 && keyword_length < read_length && - keyword[keyword_length] != 0) - ++keyword_length; - - /* TODO: make the keyword checking common */ - if (keyword_length >= 1 && keyword_length <= 79) - { - /* We only understand '0' compression - deflate - so if we get a - * different value we can't safely decode the chunk. - */ - if (keyword_length+1 < read_length && - keyword[keyword_length+1] == PNG_COMPRESSION_TYPE_BASE) - { - read_length -= keyword_length+2; - - if (png_inflate_claim(png_ptr, png_iCCP) == Z_OK) - { - Byte profile_header[132]={0}; - Byte local_buffer[PNG_INFLATE_BUF_SIZE]; - png_alloc_size_t size = (sizeof profile_header); - - png_ptr->zstream.next_in = (Bytef*)keyword + (keyword_length+2); - png_ptr->zstream.avail_in = read_length; - (void)png_inflate_read(png_ptr, local_buffer, - (sizeof local_buffer), &length, profile_header, &size, - 0/*finish: don't, because the output is too small*/); - - if (size == 0) - { - /* We have the ICC profile header; do the basic header checks. - */ - png_uint_32 profile_length = png_get_uint_32(profile_header); - - if (png_icc_check_length(png_ptr, &png_ptr->colorspace, - keyword, profile_length) != 0) - { - /* The length is apparently ok, so we can check the 132 - * byte header. - */ - if (png_icc_check_header(png_ptr, &png_ptr->colorspace, - keyword, profile_length, profile_header, - png_ptr->color_type) != 0) - { - /* Now read the tag table; a variable size buffer is - * needed at this point, allocate one for the whole - * profile. The header check has already validated - * that none of this stuff will overflow. - */ - png_uint_32 tag_count = - png_get_uint_32(profile_header + 128); - png_bytep profile = png_read_buffer(png_ptr, - profile_length, 2/*silent*/); - - if (profile != NULL) - { - memcpy(profile, profile_header, - (sizeof profile_header)); - - size = 12 * tag_count; - - (void)png_inflate_read(png_ptr, local_buffer, - (sizeof local_buffer), &length, - profile + (sizeof profile_header), &size, 0); - - /* Still expect a buffer error because we expect - * there to be some tag data! - */ - if (size == 0) - { - if (png_icc_check_tag_table(png_ptr, - &png_ptr->colorspace, keyword, profile_length, - profile) != 0) - { - /* The profile has been validated for basic - * security issues, so read the whole thing in. - */ - size = profile_length - (sizeof profile_header) - - 12 * tag_count; - - (void)png_inflate_read(png_ptr, local_buffer, - (sizeof local_buffer), &length, - profile + (sizeof profile_header) + - 12 * tag_count, &size, 1/*finish*/); - - if (length > 0 && !(png_ptr->flags & - PNG_FLAG_BENIGN_ERRORS_WARN)) - errmsg = "extra compressed data"; - - /* But otherwise allow extra data: */ - else if (size == 0) - { - if (length > 0) - { - /* This can be handled completely, so - * keep going. - */ - png_chunk_warning(png_ptr, - "extra compressed data"); - } - - png_crc_finish(png_ptr, length); - finished = 1; - -# if defined(PNG_sRGB_SUPPORTED) && PNG_sRGB_PROFILE_CHECKS >= 0 - /* Check for a match against sRGB */ - png_icc_set_sRGB(png_ptr, - &png_ptr->colorspace, profile, - png_ptr->zstream.adler); -# endif - - /* Steal the profile for info_ptr. */ - if (info_ptr != NULL) - { - png_free_data(png_ptr, info_ptr, - PNG_FREE_ICCP, 0); - - info_ptr->iccp_name = png_voidcast(char*, - png_malloc_base(png_ptr, - keyword_length+1)); - if (info_ptr->iccp_name != NULL) - { - memcpy(info_ptr->iccp_name, keyword, - keyword_length+1); - info_ptr->iccp_proflen = - profile_length; - info_ptr->iccp_profile = profile; - png_ptr->read_buffer = NULL; /*steal*/ - info_ptr->free_me |= PNG_FREE_ICCP; - info_ptr->valid |= PNG_INFO_iCCP; - } - - else - { - png_ptr->colorspace.flags |= - PNG_COLORSPACE_INVALID; - errmsg = "out of memory"; - } - } - - /* else the profile remains in the read - * buffer which gets reused for subsequent - * chunks. - */ - - if (info_ptr != NULL) - png_colorspace_sync(png_ptr, info_ptr); - - if (errmsg == NULL) - { - png_ptr->zowner = 0; - return; - } - } - if (errmsg == NULL) - errmsg = png_ptr->zstream.msg; - } - /* else png_icc_check_tag_table output an error */ - } - else /* profile truncated */ - errmsg = png_ptr->zstream.msg; - } - - else - errmsg = "out of memory"; - } - - /* else png_icc_check_header output an error */ - } - - /* else png_icc_check_length output an error */ - } - - else /* profile truncated */ - errmsg = png_ptr->zstream.msg; - - /* Release the stream */ - png_ptr->zowner = 0; - } - - else /* png_inflate_claim failed */ - errmsg = png_ptr->zstream.msg; - } - - else - errmsg = "bad compression method"; /* or missing */ - } - - else - errmsg = "bad keyword"; - } - - else - errmsg = "too many profiles"; - - /* Failure: the reason is in 'errmsg' */ - if (finished == 0) - png_crc_finish(png_ptr, length); - - png_ptr->colorspace.flags |= PNG_COLORSPACE_INVALID; - png_colorspace_sync(png_ptr, info_ptr); - if (errmsg != NULL) /* else already output */ - png_chunk_benign_error(png_ptr, errmsg); -} -#endif /* READ_iCCP */ - -#ifdef PNG_READ_sPLT_SUPPORTED -void /* PRIVATE */ -png_handle_sPLT(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -/* Note: this does not properly handle chunks that are > 64K under DOS */ -{ - png_bytep entry_start, buffer; - png_sPLT_t new_palette; - png_sPLT_entryp pp; - png_uint_32 data_length; - int entry_size, i; - png_uint_32 skip = 0; - png_uint_32 dl; - size_t max_dl; - - png_debug(1, "in png_handle_sPLT"); - -#ifdef PNG_USER_LIMITS_SUPPORTED - if (png_ptr->user_chunk_cache_max != 0) - { - if (png_ptr->user_chunk_cache_max == 1) - { - png_crc_finish(png_ptr, length); - return; - } - - if (--png_ptr->user_chunk_cache_max == 1) - { - png_warning(png_ptr, "No space in chunk cache for sPLT"); - png_crc_finish(png_ptr, length); - return; - } - } -#endif - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - -#ifdef PNG_MAX_MALLOC_64K - if (length > 65535U) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "too large to fit in memory"); - return; - } -#endif - - buffer = png_read_buffer(png_ptr, length+1, 2/*silent*/); - if (buffer == NULL) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of memory"); - return; - } - - - /* WARNING: this may break if size_t is less than 32 bits; it is assumed - * that the PNG_MAX_MALLOC_64K test is enabled in this case, but this is a - * potential breakage point if the types in pngconf.h aren't exactly right. - */ - png_crc_read(png_ptr, buffer, length); - - if (png_crc_finish(png_ptr, skip) != 0) - return; - - buffer[length] = 0; - - for (entry_start = buffer; *entry_start; entry_start++) - /* Empty loop to find end of name */ ; - - ++entry_start; - - /* A sample depth should follow the separator, and we should be on it */ - if (length < 2U || entry_start > buffer + (length - 2U)) - { - png_warning(png_ptr, "malformed sPLT chunk"); - return; - } - - new_palette.depth = *entry_start++; - entry_size = (new_palette.depth == 8 ? 6 : 10); - /* This must fit in a png_uint_32 because it is derived from the original - * chunk data length. - */ - data_length = length - (png_uint_32)(entry_start - buffer); - - /* Integrity-check the data length */ - if ((data_length % (unsigned int)entry_size) != 0) - { - png_warning(png_ptr, "sPLT chunk has bad length"); - return; - } - - dl = (png_uint_32)(data_length / (unsigned int)entry_size); - max_dl = PNG_SIZE_MAX / (sizeof (png_sPLT_entry)); - - if (dl > max_dl) - { - png_warning(png_ptr, "sPLT chunk too long"); - return; - } - - new_palette.nentries = (png_int_32)(data_length / (unsigned int)entry_size); - - new_palette.entries = (png_sPLT_entryp)png_malloc_warn(png_ptr, - (png_alloc_size_t) new_palette.nentries * (sizeof (png_sPLT_entry))); - - if (new_palette.entries == NULL) - { - png_warning(png_ptr, "sPLT chunk requires too much memory"); - return; - } - -#ifdef PNG_POINTER_INDEXING_SUPPORTED - for (i = 0; i < new_palette.nentries; i++) - { - pp = new_palette.entries + i; - - if (new_palette.depth == 8) - { - pp->red = *entry_start++; - pp->green = *entry_start++; - pp->blue = *entry_start++; - pp->alpha = *entry_start++; - } - - else - { - pp->red = png_get_uint_16(entry_start); entry_start += 2; - pp->green = png_get_uint_16(entry_start); entry_start += 2; - pp->blue = png_get_uint_16(entry_start); entry_start += 2; - pp->alpha = png_get_uint_16(entry_start); entry_start += 2; - } - - pp->frequency = png_get_uint_16(entry_start); entry_start += 2; - } -#else - pp = new_palette.entries; - - for (i = 0; i < new_palette.nentries; i++) - { - - if (new_palette.depth == 8) - { - pp[i].red = *entry_start++; - pp[i].green = *entry_start++; - pp[i].blue = *entry_start++; - pp[i].alpha = *entry_start++; - } - - else - { - pp[i].red = png_get_uint_16(entry_start); entry_start += 2; - pp[i].green = png_get_uint_16(entry_start); entry_start += 2; - pp[i].blue = png_get_uint_16(entry_start); entry_start += 2; - pp[i].alpha = png_get_uint_16(entry_start); entry_start += 2; - } - - pp[i].frequency = png_get_uint_16(entry_start); entry_start += 2; - } -#endif - - /* Discard all chunk data except the name and stash that */ - new_palette.name = (png_charp)buffer; - - png_set_sPLT(png_ptr, info_ptr, &new_palette, 1); - - png_free(png_ptr, new_palette.entries); -} -#endif /* READ_sPLT */ - -#ifdef PNG_READ_tRNS_SUPPORTED -void /* PRIVATE */ -png_handle_tRNS(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_byte readbuf[PNG_MAX_PALETTE_LENGTH]; - - png_debug(1, "in png_handle_tRNS"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - else if (info_ptr != NULL && (info_ptr->valid & PNG_INFO_tRNS) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "duplicate"); - return; - } - - if (png_ptr->color_type == PNG_COLOR_TYPE_GRAY) - { - png_byte buf[2]; - - if (length != 2) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - png_crc_read(png_ptr, buf, 2); - png_ptr->num_trans = 1; - png_ptr->trans_color.gray = png_get_uint_16(buf); - } - - else if (png_ptr->color_type == PNG_COLOR_TYPE_RGB) - { - png_byte buf[6]; - - if (length != 6) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - png_crc_read(png_ptr, buf, length); - png_ptr->num_trans = 1; - png_ptr->trans_color.red = png_get_uint_16(buf); - png_ptr->trans_color.green = png_get_uint_16(buf + 2); - png_ptr->trans_color.blue = png_get_uint_16(buf + 4); - } - - else if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - { - if ((png_ptr->mode & PNG_HAVE_PLTE) == 0) - { - /* TODO: is this actually an error in the ISO spec? */ - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - if (length > (unsigned int) png_ptr->num_palette || - length > (unsigned int) PNG_MAX_PALETTE_LENGTH || - length == 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - png_crc_read(png_ptr, readbuf, length); - png_ptr->num_trans = (png_uint_16)length; - } - - else - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid with alpha channel"); - return; - } - - if (png_crc_finish(png_ptr, 0) != 0) - { - png_ptr->num_trans = 0; - return; - } - - /* TODO: this is a horrible side effect in the palette case because the - * png_struct ends up with a pointer to the tRNS buffer owned by the - * png_info. Fix this. - */ - png_set_tRNS(png_ptr, info_ptr, readbuf, png_ptr->num_trans, - &(png_ptr->trans_color)); -} -#endif - -#ifdef PNG_READ_bKGD_SUPPORTED -void /* PRIVATE */ -png_handle_bKGD(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - unsigned int truelen; - png_byte buf[6]; - png_color_16 background; - - png_debug(1, "in png_handle_bKGD"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & PNG_HAVE_IDAT) != 0 || - (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE && - (png_ptr->mode & PNG_HAVE_PLTE) == 0)) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - else if (info_ptr != NULL && (info_ptr->valid & PNG_INFO_bKGD) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "duplicate"); - return; - } - - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - truelen = 1; - - else if ((png_ptr->color_type & PNG_COLOR_MASK_COLOR) != 0) - truelen = 6; - - else - truelen = 2; - - if (length != truelen) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - png_crc_read(png_ptr, buf, truelen); - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - /* We convert the index value into RGB components so that we can allow - * arbitrary RGB values for background when we have transparency, and - * so it is easy to determine the RGB values of the background color - * from the info_ptr struct. - */ - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - { - background.index = buf[0]; - - if (info_ptr != NULL && info_ptr->num_palette != 0) - { - if (buf[0] >= info_ptr->num_palette) - { - png_chunk_benign_error(png_ptr, "invalid index"); - return; - } - - background.red = (png_uint_16)png_ptr->palette[buf[0]].red; - background.green = (png_uint_16)png_ptr->palette[buf[0]].green; - background.blue = (png_uint_16)png_ptr->palette[buf[0]].blue; - } - - else - background.red = background.green = background.blue = 0; - - background.gray = 0; - } - - else if ((png_ptr->color_type & PNG_COLOR_MASK_COLOR) == 0) /* GRAY */ - { - if (png_ptr->bit_depth <= 8) - { - if (buf[0] != 0 || buf[1] >= (unsigned int)(1 << png_ptr->bit_depth)) - { - png_chunk_benign_error(png_ptr, "invalid gray level"); - return; - } - } - - background.index = 0; - background.red = - background.green = - background.blue = - background.gray = png_get_uint_16(buf); - } - - else - { - if (png_ptr->bit_depth <= 8) - { - if (buf[0] != 0 || buf[2] != 0 || buf[4] != 0) - { - png_chunk_benign_error(png_ptr, "invalid color"); - return; - } - } - - background.index = 0; - background.red = png_get_uint_16(buf); - background.green = png_get_uint_16(buf + 2); - background.blue = png_get_uint_16(buf + 4); - background.gray = 0; - } - - png_set_bKGD(png_ptr, info_ptr, &background); -} -#endif - -#ifdef PNG_READ_eXIf_SUPPORTED -void /* PRIVATE */ -png_handle_eXIf(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - unsigned int i; - - png_debug(1, "in png_handle_eXIf"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - if (length < 2) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "too short"); - return; - } - - else if (info_ptr == NULL || (info_ptr->valid & PNG_INFO_eXIf) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "duplicate"); - return; - } - - info_ptr->free_me |= PNG_FREE_EXIF; - - info_ptr->eXIf_buf = png_voidcast(png_bytep, - png_malloc_warn(png_ptr, length)); - - if (info_ptr->eXIf_buf == NULL) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of memory"); - return; - } - - for (i = 0; i < length; i++) - { - png_byte buf[1]; - png_crc_read(png_ptr, buf, 1); - info_ptr->eXIf_buf[i] = buf[0]; - if (i == 1) - { - if ((buf[0] != 'M' && buf[0] != 'I') || - (info_ptr->eXIf_buf[0] != buf[0])) - { - png_crc_finish(png_ptr, length - 2); - png_chunk_benign_error(png_ptr, "incorrect byte-order specifier"); - png_free(png_ptr, info_ptr->eXIf_buf); - info_ptr->eXIf_buf = NULL; - return; - } - } - } - - if (png_crc_finish(png_ptr, 0) == 0) - png_set_eXIf_1(png_ptr, info_ptr, length, info_ptr->eXIf_buf); - - png_free(png_ptr, info_ptr->eXIf_buf); - info_ptr->eXIf_buf = NULL; -} -#endif - -#ifdef PNG_READ_hIST_SUPPORTED -void /* PRIVATE */ -png_handle_hIST(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - unsigned int num, i; - png_uint_16 readbuf[PNG_MAX_PALETTE_LENGTH]; - - png_debug(1, "in png_handle_hIST"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & PNG_HAVE_IDAT) != 0 || - (png_ptr->mode & PNG_HAVE_PLTE) == 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - else if (info_ptr != NULL && (info_ptr->valid & PNG_INFO_hIST) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "duplicate"); - return; - } - - num = length / 2 ; - - if (length != num * 2 || - num != (unsigned int)png_ptr->num_palette || - num > (unsigned int)PNG_MAX_PALETTE_LENGTH) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - for (i = 0; i < num; i++) - { - png_byte buf[2]; - - png_crc_read(png_ptr, buf, 2); - readbuf[i] = png_get_uint_16(buf); - } - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - png_set_hIST(png_ptr, info_ptr, readbuf); -} -#endif - -#ifdef PNG_READ_pHYs_SUPPORTED -void /* PRIVATE */ -png_handle_pHYs(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_byte buf[9]; - png_uint_32 res_x, res_y; - int unit_type; - - png_debug(1, "in png_handle_pHYs"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - else if (info_ptr != NULL && (info_ptr->valid & PNG_INFO_pHYs) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "duplicate"); - return; - } - - if (length != 9) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - png_crc_read(png_ptr, buf, 9); - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - res_x = png_get_uint_32(buf); - res_y = png_get_uint_32(buf + 4); - unit_type = buf[8]; - png_set_pHYs(png_ptr, info_ptr, res_x, res_y, unit_type); -} -#endif - -#ifdef PNG_READ_oFFs_SUPPORTED -void /* PRIVATE */ -png_handle_oFFs(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_byte buf[9]; - png_int_32 offset_x, offset_y; - int unit_type; - - png_debug(1, "in png_handle_oFFs"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - else if (info_ptr != NULL && (info_ptr->valid & PNG_INFO_oFFs) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "duplicate"); - return; - } - - if (length != 9) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - png_crc_read(png_ptr, buf, 9); - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - offset_x = png_get_int_32(buf); - offset_y = png_get_int_32(buf + 4); - unit_type = buf[8]; - png_set_oFFs(png_ptr, info_ptr, offset_x, offset_y, unit_type); -} -#endif - -#ifdef PNG_READ_pCAL_SUPPORTED -/* Read the pCAL chunk (described in the PNG Extensions document) */ -void /* PRIVATE */ -png_handle_pCAL(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_int_32 X0, X1; - png_byte type, nparams; - png_bytep buffer, buf, units, endptr; - png_charpp params; - int i; - - png_debug(1, "in png_handle_pCAL"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - else if (info_ptr != NULL && (info_ptr->valid & PNG_INFO_pCAL) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "duplicate"); - return; - } - - png_debug1(2, "Allocating and reading pCAL chunk data (%u bytes)", - length + 1); - - buffer = png_read_buffer(png_ptr, length+1, 2/*silent*/); - - if (buffer == NULL) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of memory"); - return; - } - - png_crc_read(png_ptr, buffer, length); - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - buffer[length] = 0; /* Null terminate the last string */ - - png_debug(3, "Finding end of pCAL purpose string"); - for (buf = buffer; *buf; buf++) - /* Empty loop */ ; - - endptr = buffer + length; - - /* We need to have at least 12 bytes after the purpose string - * in order to get the parameter information. - */ - if (endptr - buf <= 12) - { - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - png_debug(3, "Reading pCAL X0, X1, type, nparams, and units"); - X0 = png_get_int_32((png_bytep)buf+1); - X1 = png_get_int_32((png_bytep)buf+5); - type = buf[9]; - nparams = buf[10]; - units = buf + 11; - - png_debug(3, "Checking pCAL equation type and number of parameters"); - /* Check that we have the right number of parameters for known - * equation types. - */ - if ((type == PNG_EQUATION_LINEAR && nparams != 2) || - (type == PNG_EQUATION_BASE_E && nparams != 3) || - (type == PNG_EQUATION_ARBITRARY && nparams != 3) || - (type == PNG_EQUATION_HYPERBOLIC && nparams != 4)) - { - png_chunk_benign_error(png_ptr, "invalid parameter count"); - return; - } - - else if (type >= PNG_EQUATION_LAST) - { - png_chunk_benign_error(png_ptr, "unrecognized equation type"); - } - - for (buf = units; *buf; buf++) - /* Empty loop to move past the units string. */ ; - - png_debug(3, "Allocating pCAL parameters array"); - - params = png_voidcast(png_charpp, png_malloc_warn(png_ptr, - nparams * (sizeof (png_charp)))); - - if (params == NULL) - { - png_chunk_benign_error(png_ptr, "out of memory"); - return; - } - - /* Get pointers to the start of each parameter string. */ - for (i = 0; i < nparams; i++) - { - buf++; /* Skip the null string terminator from previous parameter. */ - - png_debug1(3, "Reading pCAL parameter %d", i); - - for (params[i] = (png_charp)buf; buf <= endptr && *buf != 0; buf++) - /* Empty loop to move past each parameter string */ ; - - /* Make sure we haven't run out of data yet */ - if (buf > endptr) - { - png_free(png_ptr, params); - png_chunk_benign_error(png_ptr, "invalid data"); - return; - } - } - - png_set_pCAL(png_ptr, info_ptr, (png_charp)buffer, X0, X1, type, nparams, - (png_charp)units, params); - - png_free(png_ptr, params); -} -#endif - -#ifdef PNG_READ_sCAL_SUPPORTED -/* Read the sCAL chunk */ -void /* PRIVATE */ -png_handle_sCAL(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_bytep buffer; - size_t i; - int state; - - png_debug(1, "in png_handle_sCAL"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of place"); - return; - } - - else if (info_ptr != NULL && (info_ptr->valid & PNG_INFO_sCAL) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "duplicate"); - return; - } - - /* Need unit type, width, \0, height: minimum 4 bytes */ - else if (length < 4) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - png_debug1(2, "Allocating and reading sCAL chunk data (%u bytes)", - length + 1); - - buffer = png_read_buffer(png_ptr, length+1, 2/*silent*/); - - if (buffer == NULL) - { - png_chunk_benign_error(png_ptr, "out of memory"); - png_crc_finish(png_ptr, length); - return; - } - - png_crc_read(png_ptr, buffer, length); - buffer[length] = 0; /* Null terminate the last string */ - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - /* Validate the unit. */ - if (buffer[0] != 1 && buffer[0] != 2) - { - png_chunk_benign_error(png_ptr, "invalid unit"); - return; - } - - /* Validate the ASCII numbers, need two ASCII numbers separated by - * a '\0' and they need to fit exactly in the chunk data. - */ - i = 1; - state = 0; - - if (png_check_fp_number((png_const_charp)buffer, length, &state, &i) == 0 || - i >= length || buffer[i++] != 0) - png_chunk_benign_error(png_ptr, "bad width format"); - - else if (PNG_FP_IS_POSITIVE(state) == 0) - png_chunk_benign_error(png_ptr, "non-positive width"); - - else - { - size_t heighti = i; - - state = 0; - if (png_check_fp_number((png_const_charp)buffer, length, - &state, &i) == 0 || i != length) - png_chunk_benign_error(png_ptr, "bad height format"); - - else if (PNG_FP_IS_POSITIVE(state) == 0) - png_chunk_benign_error(png_ptr, "non-positive height"); - - else - /* This is the (only) success case. */ - png_set_sCAL_s(png_ptr, info_ptr, buffer[0], - (png_charp)buffer+1, (png_charp)buffer+heighti); - } -} -#endif - -#ifdef PNG_READ_tIME_SUPPORTED -void /* PRIVATE */ -png_handle_tIME(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_byte buf[7]; - png_time mod_time; - - png_debug(1, "in png_handle_tIME"); - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - else if (info_ptr != NULL && (info_ptr->valid & PNG_INFO_tIME) != 0) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "duplicate"); - return; - } - - if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - png_ptr->mode |= PNG_AFTER_IDAT; - - if (length != 7) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "invalid"); - return; - } - - png_crc_read(png_ptr, buf, 7); - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - mod_time.second = buf[6]; - mod_time.minute = buf[5]; - mod_time.hour = buf[4]; - mod_time.day = buf[3]; - mod_time.month = buf[2]; - mod_time.year = png_get_uint_16(buf); - - png_set_tIME(png_ptr, info_ptr, &mod_time); -} -#endif - -#ifdef PNG_READ_tEXt_SUPPORTED -/* Note: this does not properly handle chunks that are > 64K under DOS */ -void /* PRIVATE */ -png_handle_tEXt(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_text text_info; - png_bytep buffer; - png_charp key; - png_charp text; - png_uint_32 skip = 0; - - png_debug(1, "in png_handle_tEXt"); - -#ifdef PNG_USER_LIMITS_SUPPORTED - if (png_ptr->user_chunk_cache_max != 0) - { - if (png_ptr->user_chunk_cache_max == 1) - { - png_crc_finish(png_ptr, length); - return; - } - - if (--png_ptr->user_chunk_cache_max == 1) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "no space in chunk cache"); - return; - } - } -#endif - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - png_ptr->mode |= PNG_AFTER_IDAT; - -#ifdef PNG_MAX_MALLOC_64K - if (length > 65535U) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "too large to fit in memory"); - return; - } -#endif - - buffer = png_read_buffer(png_ptr, length+1, 1/*warn*/); - - if (buffer == NULL) - { - png_chunk_benign_error(png_ptr, "out of memory"); - return; - } - - png_crc_read(png_ptr, buffer, length); - - if (png_crc_finish(png_ptr, skip) != 0) - return; - - key = (png_charp)buffer; - key[length] = 0; - - for (text = key; *text; text++) - /* Empty loop to find end of key */ ; - - if (text != key + length) - text++; - - text_info.compression = PNG_TEXT_COMPRESSION_NONE; - text_info.key = key; - text_info.lang = NULL; - text_info.lang_key = NULL; - text_info.itxt_length = 0; - text_info.text = text; - text_info.text_length = strlen(text); - - if (png_set_text_2(png_ptr, info_ptr, &text_info, 1) != 0) - png_warning(png_ptr, "Insufficient memory to process text chunk"); -} -#endif - -#ifdef PNG_READ_zTXt_SUPPORTED -/* Note: this does not correctly handle chunks that are > 64K under DOS */ -void /* PRIVATE */ -png_handle_zTXt(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_const_charp errmsg = NULL; - png_bytep buffer; - png_uint_32 keyword_length; - - png_debug(1, "in png_handle_zTXt"); - -#ifdef PNG_USER_LIMITS_SUPPORTED - if (png_ptr->user_chunk_cache_max != 0) - { - if (png_ptr->user_chunk_cache_max == 1) - { - png_crc_finish(png_ptr, length); - return; - } - - if (--png_ptr->user_chunk_cache_max == 1) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "no space in chunk cache"); - return; - } - } -#endif - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - png_ptr->mode |= PNG_AFTER_IDAT; - - /* Note, "length" is sufficient here; we won't be adding - * a null terminator later. - */ - buffer = png_read_buffer(png_ptr, length, 2/*silent*/); - - if (buffer == NULL) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of memory"); - return; - } - - png_crc_read(png_ptr, buffer, length); - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - /* TODO: also check that the keyword contents match the spec! */ - for (keyword_length = 0; - keyword_length < length && buffer[keyword_length] != 0; - ++keyword_length) - /* Empty loop to find end of name */ ; - - if (keyword_length > 79 || keyword_length < 1) - errmsg = "bad keyword"; - - /* zTXt must have some LZ data after the keyword, although it may expand to - * zero bytes; we need a '\0' at the end of the keyword, the compression type - * then the LZ data: - */ - else if (keyword_length + 3 > length) - errmsg = "truncated"; - - else if (buffer[keyword_length+1] != PNG_COMPRESSION_TYPE_BASE) - errmsg = "unknown compression type"; - - else - { - png_alloc_size_t uncompressed_length = PNG_SIZE_MAX; - - /* TODO: at present png_decompress_chunk imposes a single application - * level memory limit, this should be split to different values for iCCP - * and text chunks. - */ - if (png_decompress_chunk(png_ptr, length, keyword_length+2, - &uncompressed_length, 1/*terminate*/) == Z_STREAM_END) - { - png_text text; - - if (png_ptr->read_buffer == NULL) - errmsg="Read failure in png_handle_zTXt"; - else - { - /* It worked; png_ptr->read_buffer now looks like a tEXt chunk - * except for the extra compression type byte and the fact that - * it isn't necessarily '\0' terminated. - */ - buffer = png_ptr->read_buffer; - buffer[uncompressed_length+(keyword_length+2)] = 0; - - text.compression = PNG_TEXT_COMPRESSION_zTXt; - text.key = (png_charp)buffer; - text.text = (png_charp)(buffer + keyword_length+2); - text.text_length = uncompressed_length; - text.itxt_length = 0; - text.lang = NULL; - text.lang_key = NULL; - - if (png_set_text_2(png_ptr, info_ptr, &text, 1) != 0) - errmsg = "insufficient memory"; - } - } - - else - errmsg = png_ptr->zstream.msg; - } - - if (errmsg != NULL) - png_chunk_benign_error(png_ptr, errmsg); -} -#endif - -#ifdef PNG_READ_iTXt_SUPPORTED -/* Note: this does not correctly handle chunks that are > 64K under DOS */ -void /* PRIVATE */ -png_handle_iTXt(png_structrp png_ptr, png_inforp info_ptr, png_uint_32 length) -{ - png_const_charp errmsg = NULL; - png_bytep buffer; - png_uint_32 prefix_length; - - png_debug(1, "in png_handle_iTXt"); - -#ifdef PNG_USER_LIMITS_SUPPORTED - if (png_ptr->user_chunk_cache_max != 0) - { - if (png_ptr->user_chunk_cache_max == 1) - { - png_crc_finish(png_ptr, length); - return; - } - - if (--png_ptr->user_chunk_cache_max == 1) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "no space in chunk cache"); - return; - } - } -#endif - - if ((png_ptr->mode & PNG_HAVE_IHDR) == 0) - png_chunk_error(png_ptr, "missing IHDR"); - - if ((png_ptr->mode & PNG_HAVE_IDAT) != 0) - png_ptr->mode |= PNG_AFTER_IDAT; - - buffer = png_read_buffer(png_ptr, length+1, 1/*warn*/); - - if (buffer == NULL) - { - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "out of memory"); - return; - } - - png_crc_read(png_ptr, buffer, length); - - if (png_crc_finish(png_ptr, 0) != 0) - return; - - /* First the keyword. */ - for (prefix_length=0; - prefix_length < length && buffer[prefix_length] != 0; - ++prefix_length) - /* Empty loop */ ; - - /* Perform a basic check on the keyword length here. */ - if (prefix_length > 79 || prefix_length < 1) - errmsg = "bad keyword"; - - /* Expect keyword, compression flag, compression type, language, translated - * keyword (both may be empty but are 0 terminated) then the text, which may - * be empty. - */ - else if (prefix_length + 5 > length) - errmsg = "truncated"; - - else if (buffer[prefix_length+1] == 0 || - (buffer[prefix_length+1] == 1 && - buffer[prefix_length+2] == PNG_COMPRESSION_TYPE_BASE)) - { - int compressed = buffer[prefix_length+1] != 0; - png_uint_32 language_offset, translated_keyword_offset; - png_alloc_size_t uncompressed_length = 0; - - /* Now the language tag */ - prefix_length += 3; - language_offset = prefix_length; - - for (; prefix_length < length && buffer[prefix_length] != 0; - ++prefix_length) - /* Empty loop */ ; - - /* WARNING: the length may be invalid here, this is checked below. */ - translated_keyword_offset = ++prefix_length; - - for (; prefix_length < length && buffer[prefix_length] != 0; - ++prefix_length) - /* Empty loop */ ; - - /* prefix_length should now be at the trailing '\0' of the translated - * keyword, but it may already be over the end. None of this arithmetic - * can overflow because chunks are at most 2^31 bytes long, but on 16-bit - * systems the available allocation may overflow. - */ - ++prefix_length; - - if (compressed == 0 && prefix_length <= length) - uncompressed_length = length - prefix_length; - - else if (compressed != 0 && prefix_length < length) - { - uncompressed_length = PNG_SIZE_MAX; - - /* TODO: at present png_decompress_chunk imposes a single application - * level memory limit, this should be split to different values for - * iCCP and text chunks. - */ - if (png_decompress_chunk(png_ptr, length, prefix_length, - &uncompressed_length, 1/*terminate*/) == Z_STREAM_END) - buffer = png_ptr->read_buffer; - - else - errmsg = png_ptr->zstream.msg; - } - - else - errmsg = "truncated"; - - if (errmsg == NULL) - { - png_text text; - - buffer[uncompressed_length+prefix_length] = 0; - - if (compressed == 0) - text.compression = PNG_ITXT_COMPRESSION_NONE; - - else - text.compression = PNG_ITXT_COMPRESSION_zTXt; - - text.key = (png_charp)buffer; - text.lang = (png_charp)buffer + language_offset; - text.lang_key = (png_charp)buffer + translated_keyword_offset; - text.text = (png_charp)buffer + prefix_length; - text.text_length = 0; - text.itxt_length = uncompressed_length; - - if (png_set_text_2(png_ptr, info_ptr, &text, 1) != 0) - errmsg = "insufficient memory"; - } - } - - else - errmsg = "bad compression info"; - - if (errmsg != NULL) - png_chunk_benign_error(png_ptr, errmsg); -} -#endif - -#ifdef PNG_READ_UNKNOWN_CHUNKS_SUPPORTED -/* Utility function for png_handle_unknown; set up png_ptr::unknown_chunk */ -static int -png_cache_unknown_chunk(png_structrp png_ptr, png_uint_32 length) -{ - png_alloc_size_t limit = PNG_SIZE_MAX; - - if (png_ptr->unknown_chunk.data != NULL) - { - png_free(png_ptr, png_ptr->unknown_chunk.data); - png_ptr->unknown_chunk.data = NULL; - } - -# ifdef PNG_SET_USER_LIMITS_SUPPORTED - if (png_ptr->user_chunk_malloc_max > 0 && - png_ptr->user_chunk_malloc_max < limit) - limit = png_ptr->user_chunk_malloc_max; - -# elif PNG_USER_CHUNK_MALLOC_MAX > 0 - if (PNG_USER_CHUNK_MALLOC_MAX < limit) - limit = PNG_USER_CHUNK_MALLOC_MAX; -# endif - - if (length <= limit) - { - PNG_CSTRING_FROM_CHUNK(png_ptr->unknown_chunk.name, png_ptr->chunk_name); - /* The following is safe because of the PNG_SIZE_MAX init above */ - png_ptr->unknown_chunk.size = (size_t)length/*SAFE*/; - /* 'mode' is a flag array, only the bottom four bits matter here */ - png_ptr->unknown_chunk.location = (png_byte)png_ptr->mode/*SAFE*/; - - if (length == 0) - png_ptr->unknown_chunk.data = NULL; - - else - { - /* Do a 'warn' here - it is handled below. */ - png_ptr->unknown_chunk.data = png_voidcast(png_bytep, - png_malloc_warn(png_ptr, length)); - } - } - - if (png_ptr->unknown_chunk.data == NULL && length > 0) - { - /* This is benign because we clean up correctly */ - png_crc_finish(png_ptr, length); - png_chunk_benign_error(png_ptr, "unknown chunk exceeds memory limits"); - return 0; - } - - else - { - if (length > 0) - png_crc_read(png_ptr, png_ptr->unknown_chunk.data, length); - png_crc_finish(png_ptr, 0); - return 1; - } -} -#endif /* READ_UNKNOWN_CHUNKS */ - -/* Handle an unknown, or known but disabled, chunk */ -void /* PRIVATE */ -png_handle_unknown(png_structrp png_ptr, png_inforp info_ptr, - png_uint_32 length, int keep) -{ - int handled = 0; /* the chunk was handled */ - - png_debug(1, "in png_handle_unknown"); - -#ifdef PNG_READ_UNKNOWN_CHUNKS_SUPPORTED - /* NOTE: this code is based on the code in libpng-1.4.12 except for fixing - * the bug which meant that setting a non-default behavior for a specific - * chunk would be ignored (the default was always used unless a user - * callback was installed). - * - * 'keep' is the value from the png_chunk_unknown_handling, the setting for - * this specific chunk_name, if PNG_HANDLE_AS_UNKNOWN_SUPPORTED, if not it - * will always be PNG_HANDLE_CHUNK_AS_DEFAULT and it needs to be set here. - * This is just an optimization to avoid multiple calls to the lookup - * function. - */ -# ifndef PNG_HANDLE_AS_UNKNOWN_SUPPORTED -# ifdef PNG_SET_UNKNOWN_CHUNKS_SUPPORTED - keep = png_chunk_unknown_handling(png_ptr, png_ptr->chunk_name); -# endif -# endif - - /* One of the following methods will read the chunk or skip it (at least one - * of these is always defined because this is the only way to switch on - * PNG_READ_UNKNOWN_CHUNKS_SUPPORTED) - */ -# ifdef PNG_READ_USER_CHUNKS_SUPPORTED - /* The user callback takes precedence over the chunk keep value, but the - * keep value is still required to validate a save of a critical chunk. - */ - if (png_ptr->read_user_chunk_fn != NULL) - { - if (png_cache_unknown_chunk(png_ptr, length) != 0) - { - /* Callback to user unknown chunk handler */ - int ret = (*(png_ptr->read_user_chunk_fn))(png_ptr, - &png_ptr->unknown_chunk); - - /* ret is: - * negative: An error occurred; png_chunk_error will be called. - * zero: The chunk was not handled, the chunk will be discarded - * unless png_set_keep_unknown_chunks has been used to set - * a 'keep' behavior for this particular chunk, in which - * case that will be used. A critical chunk will cause an - * error at this point unless it is to be saved. - * positive: The chunk was handled, libpng will ignore/discard it. - */ - if (ret < 0) - png_chunk_error(png_ptr, "error in user chunk"); - - else if (ret == 0) - { - /* If the keep value is 'default' or 'never' override it, but - * still error out on critical chunks unless the keep value is - * 'always' While this is weird it is the behavior in 1.4.12. - * A possible improvement would be to obey the value set for the - * chunk, but this would be an API change that would probably - * damage some applications. - * - * The png_app_warning below catches the case that matters, where - * the application has not set specific save or ignore for this - * chunk or global save or ignore. - */ - if (keep < PNG_HANDLE_CHUNK_IF_SAFE) - { -# ifdef PNG_SET_UNKNOWN_CHUNKS_SUPPORTED - if (png_ptr->unknown_default < PNG_HANDLE_CHUNK_IF_SAFE) - { - png_chunk_warning(png_ptr, "Saving unknown chunk:"); - png_app_warning(png_ptr, - "forcing save of an unhandled chunk;" - " please call png_set_keep_unknown_chunks"); - /* with keep = PNG_HANDLE_CHUNK_IF_SAFE */ - } -# endif - keep = PNG_HANDLE_CHUNK_IF_SAFE; - } - } - - else /* chunk was handled */ - { - handled = 1; - /* Critical chunks can be safely discarded at this point. */ - keep = PNG_HANDLE_CHUNK_NEVER; - } - } - - else - keep = PNG_HANDLE_CHUNK_NEVER; /* insufficient memory */ - } - - else - /* Use the SAVE_UNKNOWN_CHUNKS code or skip the chunk */ -# endif /* READ_USER_CHUNKS */ - -# ifdef PNG_SAVE_UNKNOWN_CHUNKS_SUPPORTED - { - /* keep is currently just the per-chunk setting, if there was no - * setting change it to the global default now (not that this may - * still be AS_DEFAULT) then obtain the cache of the chunk if required, - * if not simply skip the chunk. - */ - if (keep == PNG_HANDLE_CHUNK_AS_DEFAULT) - keep = png_ptr->unknown_default; - - if (keep == PNG_HANDLE_CHUNK_ALWAYS || - (keep == PNG_HANDLE_CHUNK_IF_SAFE && - PNG_CHUNK_ANCILLARY(png_ptr->chunk_name))) - { - if (png_cache_unknown_chunk(png_ptr, length) == 0) - keep = PNG_HANDLE_CHUNK_NEVER; - } - - else - png_crc_finish(png_ptr, length); - } -# else -# ifndef PNG_READ_USER_CHUNKS_SUPPORTED -# error no method to support READ_UNKNOWN_CHUNKS -# endif - - { - /* If here there is no read callback pointer set and no support is - * compiled in to just save the unknown chunks, so simply skip this - * chunk. If 'keep' is something other than AS_DEFAULT or NEVER then - * the app has erroneously asked for unknown chunk saving when there - * is no support. - */ - if (keep > PNG_HANDLE_CHUNK_NEVER) - png_app_error(png_ptr, "no unknown chunk support available"); - - png_crc_finish(png_ptr, length); - } -# endif - -# ifdef PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED - /* Now store the chunk in the chunk list if appropriate, and if the limits - * permit it. - */ - if (keep == PNG_HANDLE_CHUNK_ALWAYS || - (keep == PNG_HANDLE_CHUNK_IF_SAFE && - PNG_CHUNK_ANCILLARY(png_ptr->chunk_name))) - { -# ifdef PNG_USER_LIMITS_SUPPORTED - switch (png_ptr->user_chunk_cache_max) - { - case 2: - png_ptr->user_chunk_cache_max = 1; - png_chunk_benign_error(png_ptr, "no space in chunk cache"); - /* FALLTHROUGH */ - case 1: - /* NOTE: prior to 1.6.0 this case resulted in an unknown critical - * chunk being skipped, now there will be a hard error below. - */ - break; - - default: /* not at limit */ - --(png_ptr->user_chunk_cache_max); - /* FALLTHROUGH */ - case 0: /* no limit */ -# endif /* USER_LIMITS */ - /* Here when the limit isn't reached or when limits are compiled - * out; store the chunk. - */ - png_set_unknown_chunks(png_ptr, info_ptr, - &png_ptr->unknown_chunk, 1); - handled = 1; -# ifdef PNG_USER_LIMITS_SUPPORTED - break; - } -# endif - } -# else /* no store support: the chunk must be handled by the user callback */ - PNG_UNUSED(info_ptr) -# endif - - /* Regardless of the error handling below the cached data (if any) can be - * freed now. Notice that the data is not freed if there is a png_error, but - * it will be freed by destroy_read_struct. - */ - if (png_ptr->unknown_chunk.data != NULL) - png_free(png_ptr, png_ptr->unknown_chunk.data); - png_ptr->unknown_chunk.data = NULL; - -#else /* !PNG_READ_UNKNOWN_CHUNKS_SUPPORTED */ - /* There is no support to read an unknown chunk, so just skip it. */ - png_crc_finish(png_ptr, length); - PNG_UNUSED(info_ptr) - PNG_UNUSED(keep) -#endif /* !READ_UNKNOWN_CHUNKS */ - - /* Check for unhandled critical chunks */ - if (handled == 0 && PNG_CHUNK_CRITICAL(png_ptr->chunk_name)) - png_chunk_error(png_ptr, "unhandled critical chunk"); -} - -/* This function is called to verify that a chunk name is valid. - * This function can't have the "critical chunk check" incorporated - * into it, since in the future we will need to be able to call user - * functions to handle unknown critical chunks after we check that - * the chunk name itself is valid. - */ - -/* Bit hacking: the test for an invalid byte in the 4 byte chunk name is: - * - * ((c) < 65 || (c) > 122 || ((c) > 90 && (c) < 97)) - */ - -void /* PRIVATE */ -png_check_chunk_name(png_const_structrp png_ptr, png_uint_32 chunk_name) -{ - int i; - png_uint_32 cn=chunk_name; - - png_debug(1, "in png_check_chunk_name"); - - for (i=1; i<=4; ++i) - { - int c = cn & 0xff; - - if (c < 65 || c > 122 || (c > 90 && c < 97)) - png_chunk_error(png_ptr, "invalid chunk type"); - - cn >>= 8; - } -} - -void /* PRIVATE */ -png_check_chunk_length(png_const_structrp png_ptr, png_uint_32 length) -{ - png_alloc_size_t limit = PNG_UINT_31_MAX; - -# ifdef PNG_SET_USER_LIMITS_SUPPORTED - if (png_ptr->user_chunk_malloc_max > 0 && - png_ptr->user_chunk_malloc_max < limit) - limit = png_ptr->user_chunk_malloc_max; -# elif PNG_USER_CHUNK_MALLOC_MAX > 0 - if (PNG_USER_CHUNK_MALLOC_MAX < limit) - limit = PNG_USER_CHUNK_MALLOC_MAX; -# endif - if (png_ptr->chunk_name == png_IDAT) - { - png_alloc_size_t idat_limit = PNG_UINT_31_MAX; - size_t row_factor = - (size_t)png_ptr->width - * (size_t)png_ptr->channels - * (png_ptr->bit_depth > 8? 2: 1) - + 1 - + (png_ptr->interlaced? 6: 0); - if (png_ptr->height > PNG_UINT_32_MAX/row_factor) - idat_limit = PNG_UINT_31_MAX; - else - idat_limit = png_ptr->height * row_factor; - row_factor = row_factor > 32566? 32566 : row_factor; - idat_limit += 6 + 5*(idat_limit/row_factor+1); /* zlib+deflate overhead */ - idat_limit=idat_limit < PNG_UINT_31_MAX? idat_limit : PNG_UINT_31_MAX; - limit = limit < idat_limit? idat_limit : limit; - } - - if (length > limit) - { - png_debug2(0," length = %lu, limit = %lu", - (unsigned long)length,(unsigned long)limit); - png_benign_error(png_ptr, "chunk data is too large"); - } -} - -/* Combines the row recently read in with the existing pixels in the row. This - * routine takes care of alpha and transparency if requested. This routine also - * handles the two methods of progressive display of interlaced images, - * depending on the 'display' value; if 'display' is true then the whole row - * (dp) is filled from the start by replicating the available pixels. If - * 'display' is false only those pixels present in the pass are filled in. - */ -void /* PRIVATE */ -png_combine_row(png_const_structrp png_ptr, png_bytep dp, int display) -{ - unsigned int pixel_depth = png_ptr->transformed_pixel_depth; - png_const_bytep sp = png_ptr->row_buf + 1; - png_alloc_size_t row_width = png_ptr->width; - unsigned int pass = png_ptr->pass; - png_bytep end_ptr = 0; - png_byte end_byte = 0; - unsigned int end_mask; - - png_debug(1, "in png_combine_row"); - - /* Added in 1.5.6: it should not be possible to enter this routine until at - * least one row has been read from the PNG data and transformed. - */ - if (pixel_depth == 0) - png_error(png_ptr, "internal row logic error"); - - /* Added in 1.5.4: the pixel depth should match the information returned by - * any call to png_read_update_info at this point. Do not continue if we got - * this wrong. - */ - if (png_ptr->info_rowbytes != 0 && png_ptr->info_rowbytes != - PNG_ROWBYTES(pixel_depth, row_width)) - png_error(png_ptr, "internal row size calculation error"); - - /* Don't expect this to ever happen: */ - if (row_width == 0) - png_error(png_ptr, "internal row width error"); - - /* Preserve the last byte in cases where only part of it will be overwritten, - * the multiply below may overflow, we don't care because ANSI-C guarantees - * we get the low bits. - */ - end_mask = (pixel_depth * row_width) & 7; - if (end_mask != 0) - { - /* end_ptr == NULL is a flag to say do nothing */ - end_ptr = dp + PNG_ROWBYTES(pixel_depth, row_width) - 1; - end_byte = *end_ptr; -# ifdef PNG_READ_PACKSWAP_SUPPORTED - if ((png_ptr->transformations & PNG_PACKSWAP) != 0) - /* little-endian byte */ - end_mask = (unsigned int)(0xff << end_mask); - - else /* big-endian byte */ -# endif - end_mask = 0xff >> end_mask; - /* end_mask is now the bits to *keep* from the destination row */ - } - - /* For non-interlaced images this reduces to a memcpy(). A memcpy() - * will also happen if interlacing isn't supported or if the application - * does not call png_set_interlace_handling(). In the latter cases the - * caller just gets a sequence of the unexpanded rows from each interlace - * pass. - */ -#ifdef PNG_READ_INTERLACING_SUPPORTED - if (png_ptr->interlaced != 0 && - (png_ptr->transformations & PNG_INTERLACE) != 0 && - pass < 6 && (display == 0 || - /* The following copies everything for 'display' on passes 0, 2 and 4. */ - (display == 1 && (pass & 1) != 0))) - { - /* Narrow images may have no bits in a pass; the caller should handle - * this, but this test is cheap: - */ - if (row_width <= PNG_PASS_START_COL(pass)) - return; - - if (pixel_depth < 8) - { - /* For pixel depths up to 4 bpp the 8-pixel mask can be expanded to fit - * into 32 bits, then a single loop over the bytes using the four byte - * values in the 32-bit mask can be used. For the 'display' option the - * expanded mask may also not require any masking within a byte. To - * make this work the PACKSWAP option must be taken into account - it - * simply requires the pixels to be reversed in each byte. - * - * The 'regular' case requires a mask for each of the first 6 passes, - * the 'display' case does a copy for the even passes in the range - * 0..6. This has already been handled in the test above. - * - * The masks are arranged as four bytes with the first byte to use in - * the lowest bits (little-endian) regardless of the order (PACKSWAP or - * not) of the pixels in each byte. - * - * NOTE: the whole of this logic depends on the caller of this function - * only calling it on rows appropriate to the pass. This function only - * understands the 'x' logic; the 'y' logic is handled by the caller. - * - * The following defines allow generation of compile time constant bit - * masks for each pixel depth and each possibility of swapped or not - * swapped bytes. Pass 'p' is in the range 0..6; 'x', a pixel index, - * is in the range 0..7; and the result is 1 if the pixel is to be - * copied in the pass, 0 if not. 'S' is for the sparkle method, 'B' - * for the block method. - * - * With some compilers a compile time expression of the general form: - * - * (shift >= 32) ? (a >> (shift-32)) : (b >> shift) - * - * Produces warnings with values of 'shift' in the range 33 to 63 - * because the right hand side of the ?: expression is evaluated by - * the compiler even though it isn't used. Microsoft Visual C (various - * versions) and the Intel C compiler are known to do this. To avoid - * this the following macros are used in 1.5.6. This is a temporary - * solution to avoid destabilizing the code during the release process. - */ -# if PNG_USE_COMPILE_TIME_MASKS -# define PNG_LSR(x,s) ((x)>>((s) & 0x1f)) -# define PNG_LSL(x,s) ((x)<<((s) & 0x1f)) -# else -# define PNG_LSR(x,s) ((x)>>(s)) -# define PNG_LSL(x,s) ((x)<<(s)) -# endif -# define S_COPY(p,x) (((p)<4 ? PNG_LSR(0x80088822,(3-(p))*8+(7-(x))) :\ - PNG_LSR(0xaa55ff00,(7-(p))*8+(7-(x)))) & 1) -# define B_COPY(p,x) (((p)<4 ? PNG_LSR(0xff0fff33,(3-(p))*8+(7-(x))) :\ - PNG_LSR(0xff55ff00,(7-(p))*8+(7-(x)))) & 1) - - /* Return a mask for pass 'p' pixel 'x' at depth 'd'. The mask is - * little endian - the first pixel is at bit 0 - however the extra - * parameter 's' can be set to cause the mask position to be swapped - * within each byte, to match the PNG format. This is done by XOR of - * the shift with 7, 6 or 4 for bit depths 1, 2 and 4. - */ -# define PIXEL_MASK(p,x,d,s) \ - (PNG_LSL(((PNG_LSL(1U,(d)))-1),(((x)*(d))^((s)?8-(d):0)))) - - /* Hence generate the appropriate 'block' or 'sparkle' pixel copy mask. - */ -# define S_MASKx(p,x,d,s) (S_COPY(p,x)?PIXEL_MASK(p,x,d,s):0) -# define B_MASKx(p,x,d,s) (B_COPY(p,x)?PIXEL_MASK(p,x,d,s):0) - - /* Combine 8 of these to get the full mask. For the 1-bpp and 2-bpp - * cases the result needs replicating, for the 4-bpp case the above - * generates a full 32 bits. - */ -# define MASK_EXPAND(m,d) ((m)*((d)==1?0x01010101:((d)==2?0x00010001:1))) - -# define S_MASK(p,d,s) MASK_EXPAND(S_MASKx(p,0,d,s) + S_MASKx(p,1,d,s) +\ - S_MASKx(p,2,d,s) + S_MASKx(p,3,d,s) + S_MASKx(p,4,d,s) +\ - S_MASKx(p,5,d,s) + S_MASKx(p,6,d,s) + S_MASKx(p,7,d,s), d) - -# define B_MASK(p,d,s) MASK_EXPAND(B_MASKx(p,0,d,s) + B_MASKx(p,1,d,s) +\ - B_MASKx(p,2,d,s) + B_MASKx(p,3,d,s) + B_MASKx(p,4,d,s) +\ - B_MASKx(p,5,d,s) + B_MASKx(p,6,d,s) + B_MASKx(p,7,d,s), d) - -#if PNG_USE_COMPILE_TIME_MASKS - /* Utility macros to construct all the masks for a depth/swap - * combination. The 's' parameter says whether the format is PNG - * (big endian bytes) or not. Only the three odd-numbered passes are - * required for the display/block algorithm. - */ -# define S_MASKS(d,s) { S_MASK(0,d,s), S_MASK(1,d,s), S_MASK(2,d,s),\ - S_MASK(3,d,s), S_MASK(4,d,s), S_MASK(5,d,s) } - -# define B_MASKS(d,s) { B_MASK(1,d,s), B_MASK(3,d,s), B_MASK(5,d,s) } - -# define DEPTH_INDEX(d) ((d)==1?0:((d)==2?1:2)) - - /* Hence the pre-compiled masks indexed by PACKSWAP (or not), depth and - * then pass: - */ - static const png_uint_32 row_mask[2/*PACKSWAP*/][3/*depth*/][6] = - { - /* Little-endian byte masks for PACKSWAP */ - { S_MASKS(1,0), S_MASKS(2,0), S_MASKS(4,0) }, - /* Normal (big-endian byte) masks - PNG format */ - { S_MASKS(1,1), S_MASKS(2,1), S_MASKS(4,1) } - }; - - /* display_mask has only three entries for the odd passes, so index by - * pass>>1. - */ - static const png_uint_32 display_mask[2][3][3] = - { - /* Little-endian byte masks for PACKSWAP */ - { B_MASKS(1,0), B_MASKS(2,0), B_MASKS(4,0) }, - /* Normal (big-endian byte) masks - PNG format */ - { B_MASKS(1,1), B_MASKS(2,1), B_MASKS(4,1) } - }; - -# define MASK(pass,depth,display,png)\ - ((display)?display_mask[png][DEPTH_INDEX(depth)][pass>>1]:\ - row_mask[png][DEPTH_INDEX(depth)][pass]) - -#else /* !PNG_USE_COMPILE_TIME_MASKS */ - /* This is the runtime alternative: it seems unlikely that this will - * ever be either smaller or faster than the compile time approach. - */ -# define MASK(pass,depth,display,png)\ - ((display)?B_MASK(pass,depth,png):S_MASK(pass,depth,png)) -#endif /* !USE_COMPILE_TIME_MASKS */ - - /* Use the appropriate mask to copy the required bits. In some cases - * the byte mask will be 0 or 0xff; optimize these cases. row_width is - * the number of pixels, but the code copies bytes, so it is necessary - * to special case the end. - */ - png_uint_32 pixels_per_byte = 8 / pixel_depth; - png_uint_32 mask; - -# ifdef PNG_READ_PACKSWAP_SUPPORTED - if ((png_ptr->transformations & PNG_PACKSWAP) != 0) - mask = MASK(pass, pixel_depth, display, 0); - - else -# endif - mask = MASK(pass, pixel_depth, display, 1); - - for (;;) - { - png_uint_32 m; - - /* It doesn't matter in the following if png_uint_32 has more than - * 32 bits because the high bits always match those in m<<24; it is, - * however, essential to use OR here, not +, because of this. - */ - m = mask; - mask = (m >> 8) | (m << 24); /* rotate right to good compilers */ - m &= 0xff; - - if (m != 0) /* something to copy */ - { - if (m != 0xff) - *dp = (png_byte)((*dp & ~m) | (*sp & m)); - else - *dp = *sp; - } - - /* NOTE: this may overwrite the last byte with garbage if the image - * is not an exact number of bytes wide; libpng has always done - * this. - */ - if (row_width <= pixels_per_byte) - break; /* May need to restore part of the last byte */ - - row_width -= pixels_per_byte; - ++dp; - ++sp; - } - } - - else /* pixel_depth >= 8 */ - { - unsigned int bytes_to_copy, bytes_to_jump; - - /* Validate the depth - it must be a multiple of 8 */ - if (pixel_depth & 7) - png_error(png_ptr, "invalid user transform pixel depth"); - - pixel_depth >>= 3; /* now in bytes */ - row_width *= pixel_depth; - - /* Regardless of pass number the Adam 7 interlace always results in a - * fixed number of pixels to copy then to skip. There may be a - * different number of pixels to skip at the start though. - */ - { - unsigned int offset = PNG_PASS_START_COL(pass) * pixel_depth; - - row_width -= offset; - dp += offset; - sp += offset; - } - - /* Work out the bytes to copy. */ - if (display != 0) - { - /* When doing the 'block' algorithm the pixel in the pass gets - * replicated to adjacent pixels. This is why the even (0,2,4,6) - * passes are skipped above - the entire expanded row is copied. - */ - bytes_to_copy = (1<<((6-pass)>>1)) * pixel_depth; - - /* But don't allow this number to exceed the actual row width. */ - if (bytes_to_copy > row_width) - bytes_to_copy = (unsigned int)/*SAFE*/row_width; - } - - else /* normal row; Adam7 only ever gives us one pixel to copy. */ - bytes_to_copy = pixel_depth; - - /* In Adam7 there is a constant offset between where the pixels go. */ - bytes_to_jump = PNG_PASS_COL_OFFSET(pass) * pixel_depth; - - /* And simply copy these bytes. Some optimization is possible here, - * depending on the value of 'bytes_to_copy'. Special case the low - * byte counts, which we know to be frequent. - * - * Notice that these cases all 'return' rather than 'break' - this - * avoids an unnecessary test on whether to restore the last byte - * below. - */ - switch (bytes_to_copy) - { - case 1: - for (;;) - { - *dp = *sp; - - if (row_width <= bytes_to_jump) - return; - - dp += bytes_to_jump; - sp += bytes_to_jump; - row_width -= bytes_to_jump; - } - - case 2: - /* There is a possibility of a partial copy at the end here; this - * slows the code down somewhat. - */ - do - { - dp[0] = sp[0]; dp[1] = sp[1]; - - if (row_width <= bytes_to_jump) - return; - - sp += bytes_to_jump; - dp += bytes_to_jump; - row_width -= bytes_to_jump; - } - while (row_width > 1); - - /* And there can only be one byte left at this point: */ - *dp = *sp; - return; - - case 3: - /* This can only be the RGB case, so each copy is exactly one - * pixel and it is not necessary to check for a partial copy. - */ - for (;;) - { - dp[0] = sp[0]; dp[1] = sp[1]; dp[2] = sp[2]; - - if (row_width <= bytes_to_jump) - return; - - sp += bytes_to_jump; - dp += bytes_to_jump; - row_width -= bytes_to_jump; - } - - default: -#if PNG_ALIGN_TYPE != PNG_ALIGN_NONE - /* Check for double byte alignment and, if possible, use a - * 16-bit copy. Don't attempt this for narrow images - ones that - * are less than an interlace panel wide. Don't attempt it for - * wide bytes_to_copy either - use the memcpy there. - */ - if (bytes_to_copy < 16 /*else use memcpy*/ && - png_isaligned(dp, png_uint_16) && - png_isaligned(sp, png_uint_16) && - bytes_to_copy % (sizeof (png_uint_16)) == 0 && - bytes_to_jump % (sizeof (png_uint_16)) == 0) - { - /* Everything is aligned for png_uint_16 copies, but try for - * png_uint_32 first. - */ - if (png_isaligned(dp, png_uint_32) && - png_isaligned(sp, png_uint_32) && - bytes_to_copy % (sizeof (png_uint_32)) == 0 && - bytes_to_jump % (sizeof (png_uint_32)) == 0) - { - png_uint_32p dp32 = png_aligncast(png_uint_32p,dp); - png_const_uint_32p sp32 = png_aligncastconst( - png_const_uint_32p, sp); - size_t skip = (bytes_to_jump-bytes_to_copy) / - (sizeof (png_uint_32)); - - do - { - size_t c = bytes_to_copy; - do - { - *dp32++ = *sp32++; - c -= (sizeof (png_uint_32)); - } - while (c > 0); - - if (row_width <= bytes_to_jump) - return; - - dp32 += skip; - sp32 += skip; - row_width -= bytes_to_jump; - } - while (bytes_to_copy <= row_width); - - /* Get to here when the row_width truncates the final copy. - * There will be 1-3 bytes left to copy, so don't try the - * 16-bit loop below. - */ - dp = (png_bytep)dp32; - sp = (png_const_bytep)sp32; - do - *dp++ = *sp++; - while (--row_width > 0); - return; - } - - /* Else do it in 16-bit quantities, but only if the size is - * not too large. - */ - else - { - png_uint_16p dp16 = png_aligncast(png_uint_16p, dp); - png_const_uint_16p sp16 = png_aligncastconst( - png_const_uint_16p, sp); - size_t skip = (bytes_to_jump-bytes_to_copy) / - (sizeof (png_uint_16)); - - do - { - size_t c = bytes_to_copy; - do - { - *dp16++ = *sp16++; - c -= (sizeof (png_uint_16)); - } - while (c > 0); - - if (row_width <= bytes_to_jump) - return; - - dp16 += skip; - sp16 += skip; - row_width -= bytes_to_jump; - } - while (bytes_to_copy <= row_width); - - /* End of row - 1 byte left, bytes_to_copy > row_width: */ - dp = (png_bytep)dp16; - sp = (png_const_bytep)sp16; - do - *dp++ = *sp++; - while (--row_width > 0); - return; - } - } -#endif /* ALIGN_TYPE code */ - - /* The true default - use a memcpy: */ - for (;;) - { - memcpy(dp, sp, bytes_to_copy); - - if (row_width <= bytes_to_jump) - return; - - sp += bytes_to_jump; - dp += bytes_to_jump; - row_width -= bytes_to_jump; - if (bytes_to_copy > row_width) - bytes_to_copy = (unsigned int)/*SAFE*/row_width; - } - } - - /* NOT REACHED*/ - } /* pixel_depth >= 8 */ - - /* Here if pixel_depth < 8 to check 'end_ptr' below. */ - } - else -#endif /* READ_INTERLACING */ - - /* If here then the switch above wasn't used so just memcpy the whole row - * from the temporary row buffer (notice that this overwrites the end of the - * destination row if it is a partial byte.) - */ - memcpy(dp, sp, PNG_ROWBYTES(pixel_depth, row_width)); - - /* Restore the overwritten bits from the last byte if necessary. */ - if (end_ptr != NULL) - *end_ptr = (png_byte)((end_byte & end_mask) | (*end_ptr & ~end_mask)); -} - -#ifdef PNG_READ_INTERLACING_SUPPORTED -void /* PRIVATE */ -png_do_read_interlace(png_row_infop row_info, png_bytep row, int pass, - png_uint_32 transformations /* Because these may affect the byte layout */) -{ - /* Arrays to facilitate easy interlacing - use pass (0 - 6) as index */ - /* Offset to next interlace block */ - static const unsigned int png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; - - png_debug(1, "in png_do_read_interlace"); - if (row != NULL && row_info != NULL) - { - png_uint_32 final_width; - - final_width = row_info->width * png_pass_inc[pass]; - - switch (row_info->pixel_depth) - { - case 1: - { - png_bytep sp = row + (size_t)((row_info->width - 1) >> 3); - png_bytep dp = row + (size_t)((final_width - 1) >> 3); - unsigned int sshift, dshift; - unsigned int s_start, s_end; - int s_inc; - int jstop = (int)png_pass_inc[pass]; - png_byte v; - png_uint_32 i; - int j; - -#ifdef PNG_READ_PACKSWAP_SUPPORTED - if ((transformations & PNG_PACKSWAP) != 0) - { - sshift = ((row_info->width + 7) & 0x07); - dshift = ((final_width + 7) & 0x07); - s_start = 7; - s_end = 0; - s_inc = -1; - } - - else -#endif - { - sshift = 7 - ((row_info->width + 7) & 0x07); - dshift = 7 - ((final_width + 7) & 0x07); - s_start = 0; - s_end = 7; - s_inc = 1; - } - - for (i = 0; i < row_info->width; i++) - { - v = (png_byte)((*sp >> sshift) & 0x01); - for (j = 0; j < jstop; j++) - { - unsigned int tmp = *dp & (0x7f7f >> (7 - dshift)); - tmp |= (unsigned int)(v << dshift); - *dp = (png_byte)(tmp & 0xff); - - if (dshift == s_end) - { - dshift = s_start; - dp--; - } - - else - dshift = (unsigned int)((int)dshift + s_inc); - } - - if (sshift == s_end) - { - sshift = s_start; - sp--; - } - - else - sshift = (unsigned int)((int)sshift + s_inc); - } - break; - } - - case 2: - { - png_bytep sp = row + (png_uint_32)((row_info->width - 1) >> 2); - png_bytep dp = row + (png_uint_32)((final_width - 1) >> 2); - unsigned int sshift, dshift; - unsigned int s_start, s_end; - int s_inc; - int jstop = (int)png_pass_inc[pass]; - png_uint_32 i; - -#ifdef PNG_READ_PACKSWAP_SUPPORTED - if ((transformations & PNG_PACKSWAP) != 0) - { - sshift = (((row_info->width + 3) & 0x03) << 1); - dshift = (((final_width + 3) & 0x03) << 1); - s_start = 6; - s_end = 0; - s_inc = -2; - } - - else -#endif - { - sshift = ((3 - ((row_info->width + 3) & 0x03)) << 1); - dshift = ((3 - ((final_width + 3) & 0x03)) << 1); - s_start = 0; - s_end = 6; - s_inc = 2; - } - - for (i = 0; i < row_info->width; i++) - { - png_byte v; - int j; - - v = (png_byte)((*sp >> sshift) & 0x03); - for (j = 0; j < jstop; j++) - { - unsigned int tmp = *dp & (0x3f3f >> (6 - dshift)); - tmp |= (unsigned int)(v << dshift); - *dp = (png_byte)(tmp & 0xff); - - if (dshift == s_end) - { - dshift = s_start; - dp--; - } - - else - dshift = (unsigned int)((int)dshift + s_inc); - } - - if (sshift == s_end) - { - sshift = s_start; - sp--; - } - - else - sshift = (unsigned int)((int)sshift + s_inc); - } - break; - } - - case 4: - { - png_bytep sp = row + (size_t)((row_info->width - 1) >> 1); - png_bytep dp = row + (size_t)((final_width - 1) >> 1); - unsigned int sshift, dshift; - unsigned int s_start, s_end; - int s_inc; - png_uint_32 i; - int jstop = (int)png_pass_inc[pass]; - -#ifdef PNG_READ_PACKSWAP_SUPPORTED - if ((transformations & PNG_PACKSWAP) != 0) - { - sshift = (((row_info->width + 1) & 0x01) << 2); - dshift = (((final_width + 1) & 0x01) << 2); - s_start = 4; - s_end = 0; - s_inc = -4; - } - - else -#endif - { - sshift = ((1 - ((row_info->width + 1) & 0x01)) << 2); - dshift = ((1 - ((final_width + 1) & 0x01)) << 2); - s_start = 0; - s_end = 4; - s_inc = 4; - } - - for (i = 0; i < row_info->width; i++) - { - png_byte v = (png_byte)((*sp >> sshift) & 0x0f); - int j; - - for (j = 0; j < jstop; j++) - { - unsigned int tmp = *dp & (0xf0f >> (4 - dshift)); - tmp |= (unsigned int)(v << dshift); - *dp = (png_byte)(tmp & 0xff); - - if (dshift == s_end) - { - dshift = s_start; - dp--; - } - - else - dshift = (unsigned int)((int)dshift + s_inc); - } - - if (sshift == s_end) - { - sshift = s_start; - sp--; - } - - else - sshift = (unsigned int)((int)sshift + s_inc); - } - break; - } - - default: - { - size_t pixel_bytes = (row_info->pixel_depth >> 3); - - png_bytep sp = row + (size_t)(row_info->width - 1) - * pixel_bytes; - - png_bytep dp = row + (size_t)(final_width - 1) * pixel_bytes; - - int jstop = (int)png_pass_inc[pass]; - png_uint_32 i; - - for (i = 0; i < row_info->width; i++) - { - png_byte v[8]; /* SAFE; pixel_depth does not exceed 64 */ - int j; - - memcpy(v, sp, pixel_bytes); - - for (j = 0; j < jstop; j++) - { - memcpy(dp, v, pixel_bytes); - dp -= pixel_bytes; - } - - sp -= pixel_bytes; - } - break; - } - } - - row_info->width = final_width; - row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth, final_width); - } -#ifndef PNG_READ_PACKSWAP_SUPPORTED - PNG_UNUSED(transformations) /* Silence compiler warning */ -#endif -} -#endif /* READ_INTERLACING */ - -static void -png_read_filter_row_sub(png_row_infop row_info, png_bytep row, - png_const_bytep prev_row) -{ - size_t i; - size_t istop = row_info->rowbytes; - unsigned int bpp = (row_info->pixel_depth + 7) >> 3; - png_bytep rp = row + bpp; - - PNG_UNUSED(prev_row) - - for (i = bpp; i < istop; i++) - { - *rp = (png_byte)(((int)(*rp) + (int)(*(rp-bpp))) & 0xff); - rp++; - } -} - -static void -png_read_filter_row_up(png_row_infop row_info, png_bytep row, - png_const_bytep prev_row) -{ - size_t i; - size_t istop = row_info->rowbytes; - png_bytep rp = row; - png_const_bytep pp = prev_row; - - for (i = 0; i < istop; i++) - { - *rp = (png_byte)(((int)(*rp) + (int)(*pp++)) & 0xff); - rp++; - } -} - -static void -png_read_filter_row_avg(png_row_infop row_info, png_bytep row, - png_const_bytep prev_row) -{ - size_t i; - png_bytep rp = row; - png_const_bytep pp = prev_row; - unsigned int bpp = (row_info->pixel_depth + 7) >> 3; - size_t istop = row_info->rowbytes - bpp; - - for (i = 0; i < bpp; i++) - { - *rp = (png_byte)(((int)(*rp) + - ((int)(*pp++) / 2 )) & 0xff); - - rp++; - } - - for (i = 0; i < istop; i++) - { - *rp = (png_byte)(((int)(*rp) + - (int)(*pp++ + *(rp-bpp)) / 2 ) & 0xff); - - rp++; - } -} - -static void -png_read_filter_row_paeth_1byte_pixel(png_row_infop row_info, png_bytep row, - png_const_bytep prev_row) -{ - png_bytep rp_end = row + row_info->rowbytes; - int a, c; - - /* First pixel/byte */ - c = *prev_row++; - a = *row + c; - *row++ = (png_byte)a; - - /* Remainder */ - while (row < rp_end) - { - int b, pa, pb, pc, p; - - a &= 0xff; /* From previous iteration or start */ - b = *prev_row++; - - p = b - c; - pc = a - c; - -#ifdef PNG_USE_ABS - pa = abs(p); - pb = abs(pc); - pc = abs(p + pc); -#else - pa = p < 0 ? -p : p; - pb = pc < 0 ? -pc : pc; - pc = (p + pc) < 0 ? -(p + pc) : p + pc; -#endif - - /* Find the best predictor, the least of pa, pb, pc favoring the earlier - * ones in the case of a tie. - */ - if (pb < pa) - { - pa = pb; a = b; - } - if (pc < pa) a = c; - - /* Calculate the current pixel in a, and move the previous row pixel to c - * for the next time round the loop - */ - c = b; - a += *row; - *row++ = (png_byte)a; - } -} - -static void -png_read_filter_row_paeth_multibyte_pixel(png_row_infop row_info, png_bytep row, - png_const_bytep prev_row) -{ - unsigned int bpp = (row_info->pixel_depth + 7) >> 3; - png_bytep rp_end = row + bpp; - - /* Process the first pixel in the row completely (this is the same as 'up' - * because there is only one candidate predictor for the first row). - */ - while (row < rp_end) - { - int a = *row + *prev_row++; - *row++ = (png_byte)a; - } - - /* Remainder */ - rp_end = rp_end + (row_info->rowbytes - bpp); - - while (row < rp_end) - { - int a, b, c, pa, pb, pc, p; - - c = *(prev_row - bpp); - a = *(row - bpp); - b = *prev_row++; - - p = b - c; - pc = a - c; - -#ifdef PNG_USE_ABS - pa = abs(p); - pb = abs(pc); - pc = abs(p + pc); -#else - pa = p < 0 ? -p : p; - pb = pc < 0 ? -pc : pc; - pc = (p + pc) < 0 ? -(p + pc) : p + pc; -#endif - - if (pb < pa) - { - pa = pb; a = b; - } - if (pc < pa) a = c; - - a += *row; - *row++ = (png_byte)a; - } -} - -static void -png_init_filter_functions(png_structrp pp) - /* This function is called once for every PNG image (except for PNG images - * that only use PNG_FILTER_VALUE_NONE for all rows) to set the - * implementations required to reverse the filtering of PNG rows. Reversing - * the filter is the first transformation performed on the row data. It is - * performed in place, therefore an implementation can be selected based on - * the image pixel format. If the implementation depends on image width then - * take care to ensure that it works correctly if the image is interlaced - - * interlacing causes the actual row width to vary. - */ -{ - unsigned int bpp = (pp->pixel_depth + 7) >> 3; - - pp->read_filter[PNG_FILTER_VALUE_SUB-1] = png_read_filter_row_sub; - pp->read_filter[PNG_FILTER_VALUE_UP-1] = png_read_filter_row_up; - pp->read_filter[PNG_FILTER_VALUE_AVG-1] = png_read_filter_row_avg; - if (bpp == 1) - pp->read_filter[PNG_FILTER_VALUE_PAETH-1] = - png_read_filter_row_paeth_1byte_pixel; - else - pp->read_filter[PNG_FILTER_VALUE_PAETH-1] = - png_read_filter_row_paeth_multibyte_pixel; - -#ifdef PNG_FILTER_OPTIMIZATIONS - /* To use this define PNG_FILTER_OPTIMIZATIONS as the name of a function to - * call to install hardware optimizations for the above functions; simply - * replace whatever elements of the pp->read_filter[] array with a hardware - * specific (or, for that matter, generic) optimization. - * - * To see an example of this examine what configure.ac does when - * --enable-arm-neon is specified on the command line. - */ - PNG_FILTER_OPTIMIZATIONS(pp, bpp); -#endif -} - -void /* PRIVATE */ -png_read_filter_row(png_structrp pp, png_row_infop row_info, png_bytep row, - png_const_bytep prev_row, int filter) -{ - /* OPTIMIZATION: DO NOT MODIFY THIS FUNCTION, instead #define - * PNG_FILTER_OPTIMIZATIONS to a function that overrides the generic - * implementations. See png_init_filter_functions above. - */ - if (filter > PNG_FILTER_VALUE_NONE && filter < PNG_FILTER_VALUE_LAST) - { - if (pp->read_filter[0] == NULL) - png_init_filter_functions(pp); - - pp->read_filter[filter-1](row_info, row, prev_row); - } -} - -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED -void /* PRIVATE */ -png_read_IDAT_data(png_structrp png_ptr, png_bytep output, - png_alloc_size_t avail_out) -{ - /* Loop reading IDATs and decompressing the result into output[avail_out] */ - png_ptr->zstream.next_out = output; - png_ptr->zstream.avail_out = 0; /* safety: set below */ - - if (output == NULL) - avail_out = 0; - - do - { - int ret; - png_byte tmpbuf[PNG_INFLATE_BUF_SIZE]; - - if (png_ptr->zstream.avail_in == 0) - { - uInt avail_in; - png_bytep buffer; - - while (png_ptr->idat_size == 0) - { - png_crc_finish(png_ptr, 0); - - png_ptr->idat_size = png_read_chunk_header(png_ptr); - /* This is an error even in the 'check' case because the code just - * consumed a non-IDAT header. - */ - if (png_ptr->chunk_name != png_IDAT) - png_error(png_ptr, "Not enough image data"); - } - - avail_in = png_ptr->IDAT_read_size; - - if (avail_in > png_ptr->idat_size) - avail_in = (uInt)png_ptr->idat_size; - - /* A PNG with a gradually increasing IDAT size will defeat this attempt - * to minimize memory usage by causing lots of re-allocs, but - * realistically doing IDAT_read_size re-allocs is not likely to be a - * big problem. - */ - buffer = png_read_buffer(png_ptr, avail_in, 0/*error*/); - - png_crc_read(png_ptr, buffer, avail_in); - png_ptr->idat_size -= avail_in; - - png_ptr->zstream.next_in = buffer; - png_ptr->zstream.avail_in = avail_in; - } - - /* And set up the output side. */ - if (output != NULL) /* standard read */ - { - uInt out = ZLIB_IO_MAX; - - if (out > avail_out) - out = (uInt)avail_out; - - avail_out -= out; - png_ptr->zstream.avail_out = out; - } - - else /* after last row, checking for end */ - { - png_ptr->zstream.next_out = tmpbuf; - png_ptr->zstream.avail_out = (sizeof tmpbuf); - } - - /* Use NO_FLUSH; this gives zlib the maximum opportunity to optimize the - * process. If the LZ stream is truncated the sequential reader will - * terminally damage the stream, above, by reading the chunk header of the - * following chunk (it then exits with png_error). - * - * TODO: deal more elegantly with truncated IDAT lists. - */ - ret = PNG_INFLATE(png_ptr, Z_NO_FLUSH); - - /* Take the unconsumed output back. */ - if (output != NULL) - avail_out += png_ptr->zstream.avail_out; - - else /* avail_out counts the extra bytes */ - avail_out += (sizeof tmpbuf) - png_ptr->zstream.avail_out; - - png_ptr->zstream.avail_out = 0; - - if (ret == Z_STREAM_END) - { - /* Do this for safety; we won't read any more into this row. */ - png_ptr->zstream.next_out = NULL; - - png_ptr->mode |= PNG_AFTER_IDAT; - png_ptr->flags |= PNG_FLAG_ZSTREAM_ENDED; - - if (png_ptr->zstream.avail_in > 0 || png_ptr->idat_size > 0) - png_chunk_benign_error(png_ptr, "Extra compressed data"); - break; - } - - if (ret != Z_OK) - { - png_zstream_error(png_ptr, ret); - - if (output != NULL) - png_chunk_error(png_ptr, png_ptr->zstream.msg); - - else /* checking */ - { - png_chunk_benign_error(png_ptr, png_ptr->zstream.msg); - return; - } - } - } while (avail_out > 0); - - if (avail_out > 0) - { - /* The stream ended before the image; this is the same as too few IDATs so - * should be handled the same way. - */ - if (output != NULL) - png_error(png_ptr, "Not enough image data"); - - else /* the deflate stream contained extra data */ - png_chunk_benign_error(png_ptr, "Too much image data"); - } -} - -void /* PRIVATE */ -png_read_finish_IDAT(png_structrp png_ptr) -{ - /* We don't need any more data and the stream should have ended, however the - * LZ end code may actually not have been processed. In this case we must - * read it otherwise stray unread IDAT data or, more likely, an IDAT chunk - * may still remain to be consumed. - */ - if ((png_ptr->flags & PNG_FLAG_ZSTREAM_ENDED) == 0) - { - /* The NULL causes png_read_IDAT_data to swallow any remaining bytes in - * the compressed stream, but the stream may be damaged too, so even after - * this call we may need to terminate the zstream ownership. - */ - png_read_IDAT_data(png_ptr, NULL, 0); - png_ptr->zstream.next_out = NULL; /* safety */ - - /* Now clear everything out for safety; the following may not have been - * done. - */ - if ((png_ptr->flags & PNG_FLAG_ZSTREAM_ENDED) == 0) - { - png_ptr->mode |= PNG_AFTER_IDAT; - png_ptr->flags |= PNG_FLAG_ZSTREAM_ENDED; - } - } - - /* If the zstream has not been released do it now *and* terminate the reading - * of the final IDAT chunk. - */ - if (png_ptr->zowner == png_IDAT) - { - /* Always do this; the pointers otherwise point into the read buffer. */ - png_ptr->zstream.next_in = NULL; - png_ptr->zstream.avail_in = 0; - - /* Now we no longer own the zstream. */ - png_ptr->zowner = 0; - - /* The slightly weird semantics of the sequential IDAT reading is that we - * are always in or at the end of an IDAT chunk, so we always need to do a - * crc_finish here. If idat_size is non-zero we also need to read the - * spurious bytes at the end of the chunk now. - */ - (void)png_crc_finish(png_ptr, png_ptr->idat_size); - } -} - -void /* PRIVATE */ -png_read_finish_row(png_structrp png_ptr) -{ - /* Arrays to facilitate easy interlacing - use pass (0 - 6) as index */ - - /* Start of interlace block */ - static const png_byte png_pass_start[7] = {0, 4, 0, 2, 0, 1, 0}; - - /* Offset to next interlace block */ - static const png_byte png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; - - /* Start of interlace block in the y direction */ - static const png_byte png_pass_ystart[7] = {0, 0, 4, 0, 2, 0, 1}; - - /* Offset to next interlace block in the y direction */ - static const png_byte png_pass_yinc[7] = {8, 8, 8, 4, 4, 2, 2}; - - png_debug(1, "in png_read_finish_row"); - png_ptr->row_number++; - if (png_ptr->row_number < png_ptr->num_rows) - return; - - if (png_ptr->interlaced != 0) - { - png_ptr->row_number = 0; - - /* TO DO: don't do this if prev_row isn't needed (requires - * read-ahead of the next row's filter byte. - */ - memset(png_ptr->prev_row, 0, png_ptr->rowbytes + 1); - - do - { - png_ptr->pass++; - - if (png_ptr->pass >= 7) - break; - - png_ptr->iwidth = (png_ptr->width + - png_pass_inc[png_ptr->pass] - 1 - - png_pass_start[png_ptr->pass]) / - png_pass_inc[png_ptr->pass]; - - if ((png_ptr->transformations & PNG_INTERLACE) == 0) - { - png_ptr->num_rows = (png_ptr->height + - png_pass_yinc[png_ptr->pass] - 1 - - png_pass_ystart[png_ptr->pass]) / - png_pass_yinc[png_ptr->pass]; - } - - else /* if (png_ptr->transformations & PNG_INTERLACE) */ - break; /* libpng deinterlacing sees every row */ - - } while (png_ptr->num_rows == 0 || png_ptr->iwidth == 0); - - if (png_ptr->pass < 7) - return; - } - - /* Here after at the end of the last row of the last pass. */ - png_read_finish_IDAT(png_ptr); -} -#endif /* SEQUENTIAL_READ */ - -void /* PRIVATE */ -png_read_start_row(png_structrp png_ptr) -{ - /* Arrays to facilitate easy interlacing - use pass (0 - 6) as index */ - - /* Start of interlace block */ - static const png_byte png_pass_start[7] = {0, 4, 0, 2, 0, 1, 0}; - - /* Offset to next interlace block */ - static const png_byte png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; - - /* Start of interlace block in the y direction */ - static const png_byte png_pass_ystart[7] = {0, 0, 4, 0, 2, 0, 1}; - - /* Offset to next interlace block in the y direction */ - static const png_byte png_pass_yinc[7] = {8, 8, 8, 4, 4, 2, 2}; - - unsigned int max_pixel_depth; - size_t row_bytes; - - png_debug(1, "in png_read_start_row"); - -#ifdef PNG_READ_TRANSFORMS_SUPPORTED - png_init_read_transformations(png_ptr); -#endif - if (png_ptr->interlaced != 0) - { - if ((png_ptr->transformations & PNG_INTERLACE) == 0) - png_ptr->num_rows = (png_ptr->height + png_pass_yinc[0] - 1 - - png_pass_ystart[0]) / png_pass_yinc[0]; - - else - png_ptr->num_rows = png_ptr->height; - - png_ptr->iwidth = (png_ptr->width + - png_pass_inc[png_ptr->pass] - 1 - - png_pass_start[png_ptr->pass]) / - png_pass_inc[png_ptr->pass]; - } - - else - { - png_ptr->num_rows = png_ptr->height; - png_ptr->iwidth = png_ptr->width; - } - - max_pixel_depth = (unsigned int)png_ptr->pixel_depth; - - /* WARNING: * png_read_transform_info (pngrtran.c) performs a simpler set of - * calculations to calculate the final pixel depth, then - * png_do_read_transforms actually does the transforms. This means that the - * code which effectively calculates this value is actually repeated in three - * separate places. They must all match. Innocent changes to the order of - * transformations can and will break libpng in a way that causes memory - * overwrites. - * - * TODO: fix this. - */ -#ifdef PNG_READ_PACK_SUPPORTED - if ((png_ptr->transformations & PNG_PACK) != 0 && png_ptr->bit_depth < 8) - max_pixel_depth = 8; -#endif - -#ifdef PNG_READ_EXPAND_SUPPORTED - if ((png_ptr->transformations & PNG_EXPAND) != 0) - { - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - { - if (png_ptr->num_trans != 0) - max_pixel_depth = 32; - - else - max_pixel_depth = 24; - } - - else if (png_ptr->color_type == PNG_COLOR_TYPE_GRAY) - { - if (max_pixel_depth < 8) - max_pixel_depth = 8; - - if (png_ptr->num_trans != 0) - max_pixel_depth *= 2; - } - - else if (png_ptr->color_type == PNG_COLOR_TYPE_RGB) - { - if (png_ptr->num_trans != 0) - { - max_pixel_depth *= 4; - max_pixel_depth /= 3; - } - } - } -#endif - -#ifdef PNG_READ_EXPAND_16_SUPPORTED - if ((png_ptr->transformations & PNG_EXPAND_16) != 0) - { -# ifdef PNG_READ_EXPAND_SUPPORTED - /* In fact it is an error if it isn't supported, but checking is - * the safe way. - */ - if ((png_ptr->transformations & PNG_EXPAND) != 0) - { - if (png_ptr->bit_depth < 16) - max_pixel_depth *= 2; - } - else -# endif - png_ptr->transformations &= ~PNG_EXPAND_16; - } -#endif - -#ifdef PNG_READ_FILLER_SUPPORTED - if ((png_ptr->transformations & (PNG_FILLER)) != 0) - { - if (png_ptr->color_type == PNG_COLOR_TYPE_GRAY) - { - if (max_pixel_depth <= 8) - max_pixel_depth = 16; - - else - max_pixel_depth = 32; - } - - else if (png_ptr->color_type == PNG_COLOR_TYPE_RGB || - png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - { - if (max_pixel_depth <= 32) - max_pixel_depth = 32; - - else - max_pixel_depth = 64; - } - } -#endif - -#ifdef PNG_READ_GRAY_TO_RGB_SUPPORTED - if ((png_ptr->transformations & PNG_GRAY_TO_RGB) != 0) - { - if ( -#ifdef PNG_READ_EXPAND_SUPPORTED - (png_ptr->num_trans != 0 && - (png_ptr->transformations & PNG_EXPAND) != 0) || -#endif -#ifdef PNG_READ_FILLER_SUPPORTED - (png_ptr->transformations & (PNG_FILLER)) != 0 || -#endif - png_ptr->color_type == PNG_COLOR_TYPE_GRAY_ALPHA) - { - if (max_pixel_depth <= 16) - max_pixel_depth = 32; - - else - max_pixel_depth = 64; - } - - else - { - if (max_pixel_depth <= 8) - { - if (png_ptr->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - max_pixel_depth = 32; - - else - max_pixel_depth = 24; - } - - else if (png_ptr->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - max_pixel_depth = 64; - - else - max_pixel_depth = 48; - } - } -#endif - -#if defined(PNG_READ_USER_TRANSFORM_SUPPORTED) && \ -defined(PNG_USER_TRANSFORM_PTR_SUPPORTED) - if ((png_ptr->transformations & PNG_USER_TRANSFORM) != 0) - { - unsigned int user_pixel_depth = png_ptr->user_transform_depth * - png_ptr->user_transform_channels; - - if (user_pixel_depth > max_pixel_depth) - max_pixel_depth = user_pixel_depth; - } -#endif - - /* This value is stored in png_struct and double checked in the row read - * code. - */ - png_ptr->maximum_pixel_depth = (png_byte)max_pixel_depth; - png_ptr->transformed_pixel_depth = 0; /* calculated on demand */ - - /* Align the width on the next larger 8 pixels. Mainly used - * for interlacing - */ - row_bytes = ((png_ptr->width + 7) & ~((png_uint_32)7)); - /* Calculate the maximum bytes needed, adding a byte and a pixel - * for safety's sake - */ - row_bytes = PNG_ROWBYTES(max_pixel_depth, row_bytes) + - 1 + ((max_pixel_depth + 7) >> 3U); - -#ifdef PNG_MAX_MALLOC_64K - if (row_bytes > (png_uint_32)65536L) - png_error(png_ptr, "This image requires a row greater than 64KB"); -#endif - - if (row_bytes + 48 > png_ptr->old_big_row_buf_size) - { - png_free(png_ptr, png_ptr->big_row_buf); - png_free(png_ptr, png_ptr->big_prev_row); - - if (png_ptr->interlaced != 0) - png_ptr->big_row_buf = (png_bytep)png_calloc(png_ptr, - row_bytes + 48); - - else - png_ptr->big_row_buf = (png_bytep)png_malloc(png_ptr, row_bytes + 48); - - png_ptr->big_prev_row = (png_bytep)png_malloc(png_ptr, row_bytes + 48); - -#ifdef PNG_ALIGNED_MEMORY_SUPPORTED - /* Use 16-byte aligned memory for row_buf with at least 16 bytes - * of padding before and after row_buf; treat prev_row similarly. - * NOTE: the alignment is to the start of the pixels, one beyond the start - * of the buffer, because of the filter byte. Prior to libpng 1.5.6 this - * was incorrect; the filter byte was aligned, which had the exact - * opposite effect of that intended. - */ - { - png_bytep temp = png_ptr->big_row_buf + 32; - size_t extra = (size_t)temp & 0x0f; - png_ptr->row_buf = temp - extra - 1/*filter byte*/; - - temp = png_ptr->big_prev_row + 32; - extra = (size_t)temp & 0x0f; - png_ptr->prev_row = temp - extra - 1/*filter byte*/; - } -#else - /* Use 31 bytes of padding before and 17 bytes after row_buf. */ - png_ptr->row_buf = png_ptr->big_row_buf + 31; - png_ptr->prev_row = png_ptr->big_prev_row + 31; -#endif - png_ptr->old_big_row_buf_size = row_bytes + 48; - } - -#ifdef PNG_MAX_MALLOC_64K - if (png_ptr->rowbytes > 65535) - png_error(png_ptr, "This image requires a row greater than 64KB"); - -#endif - if (png_ptr->rowbytes > (PNG_SIZE_MAX - 1)) - png_error(png_ptr, "Row has too many bytes to allocate in memory"); - - memset(png_ptr->prev_row, 0, png_ptr->rowbytes + 1); - - png_debug1(3, "width = %u,", png_ptr->width); - png_debug1(3, "height = %u,", png_ptr->height); - png_debug1(3, "iwidth = %u,", png_ptr->iwidth); - png_debug1(3, "num_rows = %u,", png_ptr->num_rows); - png_debug1(3, "rowbytes = %lu,", (unsigned long)png_ptr->rowbytes); - png_debug1(3, "irowbytes = %lu", - (unsigned long)PNG_ROWBYTES(png_ptr->pixel_depth, png_ptr->iwidth) + 1); - - /* The sequential reader needs a buffer for IDAT, but the progressive reader - * does not, so free the read buffer now regardless; the sequential reader - * reallocates it on demand. - */ - if (png_ptr->read_buffer != NULL) - { - png_bytep buffer = png_ptr->read_buffer; - - png_ptr->read_buffer_size = 0; - png_ptr->read_buffer = NULL; - png_free(png_ptr, buffer); - } - - /* Finally claim the zstream for the inflate of the IDAT data, use the bits - * value from the stream (note that this will result in a fatal error if the - * IDAT stream has a bogus deflate header window_bits value, but this should - * not be happening any longer!) - */ - if (png_inflate_claim(png_ptr, png_IDAT) != Z_OK) - png_error(png_ptr, png_ptr->zstream.msg); - - png_ptr->flags |= PNG_FLAG_ROW_INIT; -} -#endif /* READ */ diff --git a/dep/libpng/src/pngset.c b/dep/libpng/src/pngset.c deleted file mode 100644 index eb1c8c7a3..000000000 --- a/dep/libpng/src/pngset.c +++ /dev/null @@ -1,1803 +0,0 @@ - -/* pngset.c - storage of image information into info struct - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - * - * The functions here are used during reads to store data from the file - * into the info struct, and during writes to store application data - * into the info struct for writing into the file. This abstracts the - * info struct and allows us to change the structure in the future. - */ - -#include "pngpriv.h" - -#if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) - -#ifdef PNG_bKGD_SUPPORTED -void PNGAPI -png_set_bKGD(png_const_structrp png_ptr, png_inforp info_ptr, - png_const_color_16p background) -{ - png_debug1(1, "in %s storage function", "bKGD"); - - if (png_ptr == NULL || info_ptr == NULL || background == NULL) - return; - - info_ptr->background = *background; - info_ptr->valid |= PNG_INFO_bKGD; -} -#endif - -#ifdef PNG_cHRM_SUPPORTED -void PNGFAPI -png_set_cHRM_fixed(png_const_structrp png_ptr, png_inforp info_ptr, - png_fixed_point white_x, png_fixed_point white_y, png_fixed_point red_x, - png_fixed_point red_y, png_fixed_point green_x, png_fixed_point green_y, - png_fixed_point blue_x, png_fixed_point blue_y) -{ - png_xy xy; - - png_debug1(1, "in %s storage function", "cHRM fixed"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - xy.redx = red_x; - xy.redy = red_y; - xy.greenx = green_x; - xy.greeny = green_y; - xy.bluex = blue_x; - xy.bluey = blue_y; - xy.whitex = white_x; - xy.whitey = white_y; - - if (png_colorspace_set_chromaticities(png_ptr, &info_ptr->colorspace, &xy, - 2/* override with app values*/) != 0) - info_ptr->colorspace.flags |= PNG_COLORSPACE_FROM_cHRM; - - png_colorspace_sync_info(png_ptr, info_ptr); -} - -void PNGFAPI -png_set_cHRM_XYZ_fixed(png_const_structrp png_ptr, png_inforp info_ptr, - png_fixed_point int_red_X, png_fixed_point int_red_Y, - png_fixed_point int_red_Z, png_fixed_point int_green_X, - png_fixed_point int_green_Y, png_fixed_point int_green_Z, - png_fixed_point int_blue_X, png_fixed_point int_blue_Y, - png_fixed_point int_blue_Z) -{ - png_XYZ XYZ; - - png_debug1(1, "in %s storage function", "cHRM XYZ fixed"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - XYZ.red_X = int_red_X; - XYZ.red_Y = int_red_Y; - XYZ.red_Z = int_red_Z; - XYZ.green_X = int_green_X; - XYZ.green_Y = int_green_Y; - XYZ.green_Z = int_green_Z; - XYZ.blue_X = int_blue_X; - XYZ.blue_Y = int_blue_Y; - XYZ.blue_Z = int_blue_Z; - - if (png_colorspace_set_endpoints(png_ptr, &info_ptr->colorspace, - &XYZ, 2) != 0) - info_ptr->colorspace.flags |= PNG_COLORSPACE_FROM_cHRM; - - png_colorspace_sync_info(png_ptr, info_ptr); -} - -# ifdef PNG_FLOATING_POINT_SUPPORTED -void PNGAPI -png_set_cHRM(png_const_structrp png_ptr, png_inforp info_ptr, - double white_x, double white_y, double red_x, double red_y, - double green_x, double green_y, double blue_x, double blue_y) -{ - png_set_cHRM_fixed(png_ptr, info_ptr, - png_fixed(png_ptr, white_x, "cHRM White X"), - png_fixed(png_ptr, white_y, "cHRM White Y"), - png_fixed(png_ptr, red_x, "cHRM Red X"), - png_fixed(png_ptr, red_y, "cHRM Red Y"), - png_fixed(png_ptr, green_x, "cHRM Green X"), - png_fixed(png_ptr, green_y, "cHRM Green Y"), - png_fixed(png_ptr, blue_x, "cHRM Blue X"), - png_fixed(png_ptr, blue_y, "cHRM Blue Y")); -} - -void PNGAPI -png_set_cHRM_XYZ(png_const_structrp png_ptr, png_inforp info_ptr, double red_X, - double red_Y, double red_Z, double green_X, double green_Y, double green_Z, - double blue_X, double blue_Y, double blue_Z) -{ - png_set_cHRM_XYZ_fixed(png_ptr, info_ptr, - png_fixed(png_ptr, red_X, "cHRM Red X"), - png_fixed(png_ptr, red_Y, "cHRM Red Y"), - png_fixed(png_ptr, red_Z, "cHRM Red Z"), - png_fixed(png_ptr, green_X, "cHRM Green X"), - png_fixed(png_ptr, green_Y, "cHRM Green Y"), - png_fixed(png_ptr, green_Z, "cHRM Green Z"), - png_fixed(png_ptr, blue_X, "cHRM Blue X"), - png_fixed(png_ptr, blue_Y, "cHRM Blue Y"), - png_fixed(png_ptr, blue_Z, "cHRM Blue Z")); -} -# endif /* FLOATING_POINT */ - -#endif /* cHRM */ - -#ifdef PNG_eXIf_SUPPORTED -void PNGAPI -png_set_eXIf(png_const_structrp png_ptr, png_inforp info_ptr, - png_bytep exif) -{ - png_warning(png_ptr, "png_set_eXIf does not work; use png_set_eXIf_1"); - PNG_UNUSED(info_ptr) - PNG_UNUSED(exif) -} - -void PNGAPI -png_set_eXIf_1(png_const_structrp png_ptr, png_inforp info_ptr, - png_uint_32 num_exif, png_bytep exif) -{ - png_bytep new_exif; - - png_debug1(1, "in %s storage function", "eXIf"); - - if (png_ptr == NULL || info_ptr == NULL || - (png_ptr->mode & PNG_WROTE_eXIf) != 0) - return; - - new_exif = png_voidcast(png_bytep, png_malloc_warn(png_ptr, num_exif)); - - if (new_exif == NULL) - { - png_warning(png_ptr, "Insufficient memory for eXIf chunk data"); - return; - } - - memcpy(new_exif, exif, (size_t)num_exif); - - png_free_data(png_ptr, info_ptr, PNG_FREE_EXIF, 0); - - info_ptr->num_exif = num_exif; - info_ptr->exif = new_exif; - info_ptr->free_me |= PNG_FREE_EXIF; - info_ptr->valid |= PNG_INFO_eXIf; -} -#endif /* eXIf */ - -#ifdef PNG_gAMA_SUPPORTED -void PNGFAPI -png_set_gAMA_fixed(png_const_structrp png_ptr, png_inforp info_ptr, - png_fixed_point file_gamma) -{ - png_debug1(1, "in %s storage function", "gAMA"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - png_colorspace_set_gamma(png_ptr, &info_ptr->colorspace, file_gamma); - png_colorspace_sync_info(png_ptr, info_ptr); -} - -# ifdef PNG_FLOATING_POINT_SUPPORTED -void PNGAPI -png_set_gAMA(png_const_structrp png_ptr, png_inforp info_ptr, double file_gamma) -{ - png_set_gAMA_fixed(png_ptr, info_ptr, png_fixed(png_ptr, file_gamma, - "png_set_gAMA")); -} -# endif -#endif - -#ifdef PNG_hIST_SUPPORTED -void PNGAPI -png_set_hIST(png_const_structrp png_ptr, png_inforp info_ptr, - png_const_uint_16p hist) -{ - int i; - - png_debug1(1, "in %s storage function", "hIST"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - if (info_ptr->num_palette == 0 || info_ptr->num_palette - > PNG_MAX_PALETTE_LENGTH) - { - png_warning(png_ptr, - "Invalid palette size, hIST allocation skipped"); - - return; - } - - png_free_data(png_ptr, info_ptr, PNG_FREE_HIST, 0); - - /* Changed from info->num_palette to PNG_MAX_PALETTE_LENGTH in - * version 1.2.1 - */ - info_ptr->hist = png_voidcast(png_uint_16p, png_malloc_warn(png_ptr, - PNG_MAX_PALETTE_LENGTH * (sizeof (png_uint_16)))); - - if (info_ptr->hist == NULL) - { - png_warning(png_ptr, "Insufficient memory for hIST chunk data"); - return; - } - - for (i = 0; i < info_ptr->num_palette; i++) - info_ptr->hist[i] = hist[i]; - - info_ptr->free_me |= PNG_FREE_HIST; - info_ptr->valid |= PNG_INFO_hIST; -} -#endif - -void PNGAPI -png_set_IHDR(png_const_structrp png_ptr, png_inforp info_ptr, - png_uint_32 width, png_uint_32 height, int bit_depth, - int color_type, int interlace_type, int compression_type, - int filter_type) -{ - png_debug1(1, "in %s storage function", "IHDR"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - info_ptr->width = width; - info_ptr->height = height; - info_ptr->bit_depth = (png_byte)bit_depth; - info_ptr->color_type = (png_byte)color_type; - info_ptr->compression_type = (png_byte)compression_type; - info_ptr->filter_type = (png_byte)filter_type; - info_ptr->interlace_type = (png_byte)interlace_type; - - png_check_IHDR (png_ptr, info_ptr->width, info_ptr->height, - info_ptr->bit_depth, info_ptr->color_type, info_ptr->interlace_type, - info_ptr->compression_type, info_ptr->filter_type); - - if (info_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - info_ptr->channels = 1; - - else if ((info_ptr->color_type & PNG_COLOR_MASK_COLOR) != 0) - info_ptr->channels = 3; - - else - info_ptr->channels = 1; - - if ((info_ptr->color_type & PNG_COLOR_MASK_ALPHA) != 0) - info_ptr->channels++; - - info_ptr->pixel_depth = (png_byte)(info_ptr->channels * info_ptr->bit_depth); - - info_ptr->rowbytes = PNG_ROWBYTES(info_ptr->pixel_depth, width); -} - -#ifdef PNG_oFFs_SUPPORTED -void PNGAPI -png_set_oFFs(png_const_structrp png_ptr, png_inforp info_ptr, - png_int_32 offset_x, png_int_32 offset_y, int unit_type) -{ - png_debug1(1, "in %s storage function", "oFFs"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - info_ptr->x_offset = offset_x; - info_ptr->y_offset = offset_y; - info_ptr->offset_unit_type = (png_byte)unit_type; - info_ptr->valid |= PNG_INFO_oFFs; -} -#endif - -#ifdef PNG_pCAL_SUPPORTED -void PNGAPI -png_set_pCAL(png_const_structrp png_ptr, png_inforp info_ptr, - png_const_charp purpose, png_int_32 X0, png_int_32 X1, int type, - int nparams, png_const_charp units, png_charpp params) -{ - size_t length; - int i; - - png_debug1(1, "in %s storage function", "pCAL"); - - if (png_ptr == NULL || info_ptr == NULL || purpose == NULL || units == NULL - || (nparams > 0 && params == NULL)) - return; - - length = strlen(purpose) + 1; - png_debug1(3, "allocating purpose for info (%lu bytes)", - (unsigned long)length); - - /* TODO: validate format of calibration name and unit name */ - - /* Check that the type matches the specification. */ - if (type < 0 || type > 3) - { - png_chunk_report(png_ptr, "Invalid pCAL equation type", - PNG_CHUNK_WRITE_ERROR); - return; - } - - if (nparams < 0 || nparams > 255) - { - png_chunk_report(png_ptr, "Invalid pCAL parameter count", - PNG_CHUNK_WRITE_ERROR); - return; - } - - /* Validate params[nparams] */ - for (i=0; ipcal_purpose = png_voidcast(png_charp, - png_malloc_warn(png_ptr, length)); - - if (info_ptr->pcal_purpose == NULL) - { - png_chunk_report(png_ptr, "Insufficient memory for pCAL purpose", - PNG_CHUNK_WRITE_ERROR); - return; - } - - memcpy(info_ptr->pcal_purpose, purpose, length); - - info_ptr->free_me |= PNG_FREE_PCAL; - - png_debug(3, "storing X0, X1, type, and nparams in info"); - info_ptr->pcal_X0 = X0; - info_ptr->pcal_X1 = X1; - info_ptr->pcal_type = (png_byte)type; - info_ptr->pcal_nparams = (png_byte)nparams; - - length = strlen(units) + 1; - png_debug1(3, "allocating units for info (%lu bytes)", - (unsigned long)length); - - info_ptr->pcal_units = png_voidcast(png_charp, - png_malloc_warn(png_ptr, length)); - - if (info_ptr->pcal_units == NULL) - { - png_warning(png_ptr, "Insufficient memory for pCAL units"); - return; - } - - memcpy(info_ptr->pcal_units, units, length); - - info_ptr->pcal_params = png_voidcast(png_charpp, png_malloc_warn(png_ptr, - (size_t)(((unsigned int)nparams + 1) * (sizeof (png_charp))))); - - if (info_ptr->pcal_params == NULL) - { - png_warning(png_ptr, "Insufficient memory for pCAL params"); - return; - } - - memset(info_ptr->pcal_params, 0, ((unsigned int)nparams + 1) * - (sizeof (png_charp))); - - for (i = 0; i < nparams; i++) - { - length = strlen(params[i]) + 1; - png_debug2(3, "allocating parameter %d for info (%lu bytes)", i, - (unsigned long)length); - - info_ptr->pcal_params[i] = (png_charp)png_malloc_warn(png_ptr, length); - - if (info_ptr->pcal_params[i] == NULL) - { - png_warning(png_ptr, "Insufficient memory for pCAL parameter"); - return; - } - - memcpy(info_ptr->pcal_params[i], params[i], length); - } - - info_ptr->valid |= PNG_INFO_pCAL; -} -#endif - -#ifdef PNG_sCAL_SUPPORTED -void PNGAPI -png_set_sCAL_s(png_const_structrp png_ptr, png_inforp info_ptr, - int unit, png_const_charp swidth, png_const_charp sheight) -{ - size_t lengthw = 0, lengthh = 0; - - png_debug1(1, "in %s storage function", "sCAL"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - /* Double check the unit (should never get here with an invalid - * unit unless this is an API call.) - */ - if (unit != 1 && unit != 2) - png_error(png_ptr, "Invalid sCAL unit"); - - if (swidth == NULL || (lengthw = strlen(swidth)) == 0 || - swidth[0] == 45 /* '-' */ || !png_check_fp_string(swidth, lengthw)) - png_error(png_ptr, "Invalid sCAL width"); - - if (sheight == NULL || (lengthh = strlen(sheight)) == 0 || - sheight[0] == 45 /* '-' */ || !png_check_fp_string(sheight, lengthh)) - png_error(png_ptr, "Invalid sCAL height"); - - info_ptr->scal_unit = (png_byte)unit; - - ++lengthw; - - png_debug1(3, "allocating unit for info (%u bytes)", (unsigned int)lengthw); - - info_ptr->scal_s_width = png_voidcast(png_charp, - png_malloc_warn(png_ptr, lengthw)); - - if (info_ptr->scal_s_width == NULL) - { - png_warning(png_ptr, "Memory allocation failed while processing sCAL"); - - return; - } - - memcpy(info_ptr->scal_s_width, swidth, lengthw); - - ++lengthh; - - png_debug1(3, "allocating unit for info (%u bytes)", (unsigned int)lengthh); - - info_ptr->scal_s_height = png_voidcast(png_charp, - png_malloc_warn(png_ptr, lengthh)); - - if (info_ptr->scal_s_height == NULL) - { - png_free(png_ptr, info_ptr->scal_s_width); - info_ptr->scal_s_width = NULL; - - png_warning(png_ptr, "Memory allocation failed while processing sCAL"); - return; - } - - memcpy(info_ptr->scal_s_height, sheight, lengthh); - - info_ptr->free_me |= PNG_FREE_SCAL; - info_ptr->valid |= PNG_INFO_sCAL; -} - -# ifdef PNG_FLOATING_POINT_SUPPORTED -void PNGAPI -png_set_sCAL(png_const_structrp png_ptr, png_inforp info_ptr, int unit, - double width, double height) -{ - png_debug1(1, "in %s storage function", "sCAL"); - - /* Check the arguments. */ - if (width <= 0) - png_warning(png_ptr, "Invalid sCAL width ignored"); - - else if (height <= 0) - png_warning(png_ptr, "Invalid sCAL height ignored"); - - else - { - /* Convert 'width' and 'height' to ASCII. */ - char swidth[PNG_sCAL_MAX_DIGITS+1]; - char sheight[PNG_sCAL_MAX_DIGITS+1]; - - png_ascii_from_fp(png_ptr, swidth, (sizeof swidth), width, - PNG_sCAL_PRECISION); - png_ascii_from_fp(png_ptr, sheight, (sizeof sheight), height, - PNG_sCAL_PRECISION); - - png_set_sCAL_s(png_ptr, info_ptr, unit, swidth, sheight); - } -} -# endif - -# ifdef PNG_FIXED_POINT_SUPPORTED -void PNGAPI -png_set_sCAL_fixed(png_const_structrp png_ptr, png_inforp info_ptr, int unit, - png_fixed_point width, png_fixed_point height) -{ - png_debug1(1, "in %s storage function", "sCAL"); - - /* Check the arguments. */ - if (width <= 0) - png_warning(png_ptr, "Invalid sCAL width ignored"); - - else if (height <= 0) - png_warning(png_ptr, "Invalid sCAL height ignored"); - - else - { - /* Convert 'width' and 'height' to ASCII. */ - char swidth[PNG_sCAL_MAX_DIGITS+1]; - char sheight[PNG_sCAL_MAX_DIGITS+1]; - - png_ascii_from_fixed(png_ptr, swidth, (sizeof swidth), width); - png_ascii_from_fixed(png_ptr, sheight, (sizeof sheight), height); - - png_set_sCAL_s(png_ptr, info_ptr, unit, swidth, sheight); - } -} -# endif -#endif - -#ifdef PNG_pHYs_SUPPORTED -void PNGAPI -png_set_pHYs(png_const_structrp png_ptr, png_inforp info_ptr, - png_uint_32 res_x, png_uint_32 res_y, int unit_type) -{ - png_debug1(1, "in %s storage function", "pHYs"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - info_ptr->x_pixels_per_unit = res_x; - info_ptr->y_pixels_per_unit = res_y; - info_ptr->phys_unit_type = (png_byte)unit_type; - info_ptr->valid |= PNG_INFO_pHYs; -} -#endif - -void PNGAPI -png_set_PLTE(png_structrp png_ptr, png_inforp info_ptr, - png_const_colorp palette, int num_palette) -{ - - png_uint_32 max_palette_length; - - png_debug1(1, "in %s storage function", "PLTE"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - max_palette_length = (info_ptr->color_type == PNG_COLOR_TYPE_PALETTE) ? - (1 << info_ptr->bit_depth) : PNG_MAX_PALETTE_LENGTH; - - if (num_palette < 0 || num_palette > (int) max_palette_length) - { - if (info_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - png_error(png_ptr, "Invalid palette length"); - - else - { - png_warning(png_ptr, "Invalid palette length"); - - return; - } - } - - if ((num_palette > 0 && palette == NULL) || - (num_palette == 0 -# ifdef PNG_MNG_FEATURES_SUPPORTED - && (png_ptr->mng_features_permitted & PNG_FLAG_MNG_EMPTY_PLTE) == 0 -# endif - )) - { - png_error(png_ptr, "Invalid palette"); - } - - /* It may not actually be necessary to set png_ptr->palette here; - * we do it for backward compatibility with the way the png_handle_tRNS - * function used to do the allocation. - * - * 1.6.0: the above statement appears to be incorrect; something has to set - * the palette inside png_struct on read. - */ - png_free_data(png_ptr, info_ptr, PNG_FREE_PLTE, 0); - - /* Changed in libpng-1.2.1 to allocate PNG_MAX_PALETTE_LENGTH instead - * of num_palette entries, in case of an invalid PNG file or incorrect - * call to png_set_PLTE() with too-large sample values. - */ - png_ptr->palette = png_voidcast(png_colorp, png_calloc(png_ptr, - PNG_MAX_PALETTE_LENGTH * (sizeof (png_color)))); - - if (num_palette > 0) - memcpy(png_ptr->palette, palette, (unsigned int)num_palette * - (sizeof (png_color))); - - info_ptr->palette = png_ptr->palette; - info_ptr->num_palette = png_ptr->num_palette = (png_uint_16)num_palette; - info_ptr->free_me |= PNG_FREE_PLTE; - info_ptr->valid |= PNG_INFO_PLTE; -} - -#ifdef PNG_sBIT_SUPPORTED -void PNGAPI -png_set_sBIT(png_const_structrp png_ptr, png_inforp info_ptr, - png_const_color_8p sig_bit) -{ - png_debug1(1, "in %s storage function", "sBIT"); - - if (png_ptr == NULL || info_ptr == NULL || sig_bit == NULL) - return; - - info_ptr->sig_bit = *sig_bit; - info_ptr->valid |= PNG_INFO_sBIT; -} -#endif - -#ifdef PNG_sRGB_SUPPORTED -void PNGAPI -png_set_sRGB(png_const_structrp png_ptr, png_inforp info_ptr, int srgb_intent) -{ - png_debug1(1, "in %s storage function", "sRGB"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - (void)png_colorspace_set_sRGB(png_ptr, &info_ptr->colorspace, srgb_intent); - png_colorspace_sync_info(png_ptr, info_ptr); -} - -void PNGAPI -png_set_sRGB_gAMA_and_cHRM(png_const_structrp png_ptr, png_inforp info_ptr, - int srgb_intent) -{ - png_debug1(1, "in %s storage function", "sRGB_gAMA_and_cHRM"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - if (png_colorspace_set_sRGB(png_ptr, &info_ptr->colorspace, - srgb_intent) != 0) - { - /* This causes the gAMA and cHRM to be written too */ - info_ptr->colorspace.flags |= - PNG_COLORSPACE_FROM_gAMA|PNG_COLORSPACE_FROM_cHRM; - } - - png_colorspace_sync_info(png_ptr, info_ptr); -} -#endif /* sRGB */ - - -#ifdef PNG_iCCP_SUPPORTED -void PNGAPI -png_set_iCCP(png_const_structrp png_ptr, png_inforp info_ptr, - png_const_charp name, int compression_type, - png_const_bytep profile, png_uint_32 proflen) -{ - png_charp new_iccp_name; - png_bytep new_iccp_profile; - size_t length; - - png_debug1(1, "in %s storage function", "iCCP"); - - if (png_ptr == NULL || info_ptr == NULL || name == NULL || profile == NULL) - return; - - if (compression_type != PNG_COMPRESSION_TYPE_BASE) - png_app_error(png_ptr, "Invalid iCCP compression method"); - - /* Set the colorspace first because this validates the profile; do not - * override previously set app cHRM or gAMA here (because likely as not the - * application knows better than libpng what the correct values are.) Pass - * the info_ptr color_type field to png_colorspace_set_ICC because in the - * write case it has not yet been stored in png_ptr. - */ - { - int result = png_colorspace_set_ICC(png_ptr, &info_ptr->colorspace, name, - proflen, profile, info_ptr->color_type); - - png_colorspace_sync_info(png_ptr, info_ptr); - - /* Don't do any of the copying if the profile was bad, or inconsistent. */ - if (result == 0) - return; - - /* But do write the gAMA and cHRM chunks from the profile. */ - info_ptr->colorspace.flags |= - PNG_COLORSPACE_FROM_gAMA|PNG_COLORSPACE_FROM_cHRM; - } - - length = strlen(name)+1; - new_iccp_name = png_voidcast(png_charp, png_malloc_warn(png_ptr, length)); - - if (new_iccp_name == NULL) - { - png_benign_error(png_ptr, "Insufficient memory to process iCCP chunk"); - - return; - } - - memcpy(new_iccp_name, name, length); - new_iccp_profile = png_voidcast(png_bytep, - png_malloc_warn(png_ptr, proflen)); - - if (new_iccp_profile == NULL) - { - png_free(png_ptr, new_iccp_name); - png_benign_error(png_ptr, - "Insufficient memory to process iCCP profile"); - - return; - } - - memcpy(new_iccp_profile, profile, proflen); - - png_free_data(png_ptr, info_ptr, PNG_FREE_ICCP, 0); - - info_ptr->iccp_proflen = proflen; - info_ptr->iccp_name = new_iccp_name; - info_ptr->iccp_profile = new_iccp_profile; - info_ptr->free_me |= PNG_FREE_ICCP; - info_ptr->valid |= PNG_INFO_iCCP; -} -#endif - -#ifdef PNG_TEXT_SUPPORTED -void PNGAPI -png_set_text(png_const_structrp png_ptr, png_inforp info_ptr, - png_const_textp text_ptr, int num_text) -{ - int ret; - ret = png_set_text_2(png_ptr, info_ptr, text_ptr, num_text); - - if (ret != 0) - png_error(png_ptr, "Insufficient memory to store text"); -} - -int /* PRIVATE */ -png_set_text_2(png_const_structrp png_ptr, png_inforp info_ptr, - png_const_textp text_ptr, int num_text) -{ - int i; - - png_debug1(1, "in text storage function, chunk typeid = 0x%lx", - png_ptr == NULL ? 0xabadca11UL : (unsigned long)png_ptr->chunk_name); - - if (png_ptr == NULL || info_ptr == NULL || num_text <= 0 || text_ptr == NULL) - return 0; - - /* Make sure we have enough space in the "text" array in info_struct - * to hold all of the incoming text_ptr objects. This compare can't overflow - * because max_text >= num_text (anyway, subtract of two positive integers - * can't overflow in any case.) - */ - if (num_text > info_ptr->max_text - info_ptr->num_text) - { - int old_num_text = info_ptr->num_text; - int max_text; - png_textp new_text = NULL; - - /* Calculate an appropriate max_text, checking for overflow. */ - max_text = old_num_text; - if (num_text <= INT_MAX - max_text) - { - max_text += num_text; - - /* Round up to a multiple of 8 */ - if (max_text < INT_MAX-8) - max_text = (max_text + 8) & ~0x7; - - else - max_text = INT_MAX; - - /* Now allocate a new array and copy the old members in; this does all - * the overflow checks. - */ - new_text = png_voidcast(png_textp,png_realloc_array(png_ptr, - info_ptr->text, old_num_text, max_text-old_num_text, - sizeof *new_text)); - } - - if (new_text == NULL) - { - png_chunk_report(png_ptr, "too many text chunks", - PNG_CHUNK_WRITE_ERROR); - - return 1; - } - - png_free(png_ptr, info_ptr->text); - - info_ptr->text = new_text; - info_ptr->free_me |= PNG_FREE_TEXT; - info_ptr->max_text = max_text; - /* num_text is adjusted below as the entries are copied in */ - - png_debug1(3, "allocated %d entries for info_ptr->text", max_text); - } - - for (i = 0; i < num_text; i++) - { - size_t text_length, key_len; - size_t lang_len, lang_key_len; - png_textp textp = &(info_ptr->text[info_ptr->num_text]); - - if (text_ptr[i].key == NULL) - continue; - - if (text_ptr[i].compression < PNG_TEXT_COMPRESSION_NONE || - text_ptr[i].compression >= PNG_TEXT_COMPRESSION_LAST) - { - png_chunk_report(png_ptr, "text compression mode is out of range", - PNG_CHUNK_WRITE_ERROR); - continue; - } - - key_len = strlen(text_ptr[i].key); - - if (text_ptr[i].compression <= 0) - { - lang_len = 0; - lang_key_len = 0; - } - - else -# ifdef PNG_iTXt_SUPPORTED - { - /* Set iTXt data */ - - if (text_ptr[i].lang != NULL) - lang_len = strlen(text_ptr[i].lang); - - else - lang_len = 0; - - if (text_ptr[i].lang_key != NULL) - lang_key_len = strlen(text_ptr[i].lang_key); - - else - lang_key_len = 0; - } -# else /* iTXt */ - { - png_chunk_report(png_ptr, "iTXt chunk not supported", - PNG_CHUNK_WRITE_ERROR); - continue; - } -# endif - - if (text_ptr[i].text == NULL || text_ptr[i].text[0] == '\0') - { - text_length = 0; -# ifdef PNG_iTXt_SUPPORTED - if (text_ptr[i].compression > 0) - textp->compression = PNG_ITXT_COMPRESSION_NONE; - - else -# endif - textp->compression = PNG_TEXT_COMPRESSION_NONE; - } - - else - { - text_length = strlen(text_ptr[i].text); - textp->compression = text_ptr[i].compression; - } - - textp->key = png_voidcast(png_charp,png_malloc_base(png_ptr, - key_len + text_length + lang_len + lang_key_len + 4)); - - if (textp->key == NULL) - { - png_chunk_report(png_ptr, "text chunk: out of memory", - PNG_CHUNK_WRITE_ERROR); - - return 1; - } - - png_debug2(2, "Allocated %lu bytes at %p in png_set_text", - (unsigned long)(png_uint_32) - (key_len + lang_len + lang_key_len + text_length + 4), - textp->key); - - memcpy(textp->key, text_ptr[i].key, key_len); - *(textp->key + key_len) = '\0'; - - if (text_ptr[i].compression > 0) - { - textp->lang = textp->key + key_len + 1; - memcpy(textp->lang, text_ptr[i].lang, lang_len); - *(textp->lang + lang_len) = '\0'; - textp->lang_key = textp->lang + lang_len + 1; - memcpy(textp->lang_key, text_ptr[i].lang_key, lang_key_len); - *(textp->lang_key + lang_key_len) = '\0'; - textp->text = textp->lang_key + lang_key_len + 1; - } - - else - { - textp->lang=NULL; - textp->lang_key=NULL; - textp->text = textp->key + key_len + 1; - } - - if (text_length != 0) - memcpy(textp->text, text_ptr[i].text, text_length); - - *(textp->text + text_length) = '\0'; - -# ifdef PNG_iTXt_SUPPORTED - if (textp->compression > 0) - { - textp->text_length = 0; - textp->itxt_length = text_length; - } - - else -# endif - { - textp->text_length = text_length; - textp->itxt_length = 0; - } - - info_ptr->num_text++; - png_debug1(3, "transferred text chunk %d", info_ptr->num_text); - } - - return 0; -} -#endif - -#ifdef PNG_tIME_SUPPORTED -void PNGAPI -png_set_tIME(png_const_structrp png_ptr, png_inforp info_ptr, - png_const_timep mod_time) -{ - png_debug1(1, "in %s storage function", "tIME"); - - if (png_ptr == NULL || info_ptr == NULL || mod_time == NULL || - (png_ptr->mode & PNG_WROTE_tIME) != 0) - return; - - if (mod_time->month == 0 || mod_time->month > 12 || - mod_time->day == 0 || mod_time->day > 31 || - mod_time->hour > 23 || mod_time->minute > 59 || - mod_time->second > 60) - { - png_warning(png_ptr, "Ignoring invalid time value"); - - return; - } - - info_ptr->mod_time = *mod_time; - info_ptr->valid |= PNG_INFO_tIME; -} -#endif - -#ifdef PNG_tRNS_SUPPORTED -void PNGAPI -png_set_tRNS(png_structrp png_ptr, png_inforp info_ptr, - png_const_bytep trans_alpha, int num_trans, png_const_color_16p trans_color) -{ - png_debug1(1, "in %s storage function", "tRNS"); - - if (png_ptr == NULL || info_ptr == NULL) - - return; - - if (trans_alpha != NULL) - { - /* It may not actually be necessary to set png_ptr->trans_alpha here; - * we do it for backward compatibility with the way the png_handle_tRNS - * function used to do the allocation. - * - * 1.6.0: The above statement is incorrect; png_handle_tRNS effectively - * relies on png_set_tRNS storing the information in png_struct - * (otherwise it won't be there for the code in pngrtran.c). - */ - - png_free_data(png_ptr, info_ptr, PNG_FREE_TRNS, 0); - - if (num_trans > 0 && num_trans <= PNG_MAX_PALETTE_LENGTH) - { - /* Changed from num_trans to PNG_MAX_PALETTE_LENGTH in version 1.2.1 */ - info_ptr->trans_alpha = png_voidcast(png_bytep, - png_malloc(png_ptr, PNG_MAX_PALETTE_LENGTH)); - memcpy(info_ptr->trans_alpha, trans_alpha, (size_t)num_trans); - - info_ptr->free_me |= PNG_FREE_TRNS; - info_ptr->valid |= PNG_INFO_tRNS; - } - png_ptr->trans_alpha = info_ptr->trans_alpha; - } - - if (trans_color != NULL) - { -#ifdef PNG_WARNINGS_SUPPORTED - if (info_ptr->bit_depth < 16) - { - int sample_max = (1 << info_ptr->bit_depth) - 1; - - if ((info_ptr->color_type == PNG_COLOR_TYPE_GRAY && - trans_color->gray > sample_max) || - (info_ptr->color_type == PNG_COLOR_TYPE_RGB && - (trans_color->red > sample_max || - trans_color->green > sample_max || - trans_color->blue > sample_max))) - png_warning(png_ptr, - "tRNS chunk has out-of-range samples for bit_depth"); - } -#endif - - info_ptr->trans_color = *trans_color; - - if (num_trans == 0) - num_trans = 1; - } - - info_ptr->num_trans = (png_uint_16)num_trans; - - if (num_trans != 0) - { - info_ptr->free_me |= PNG_FREE_TRNS; - info_ptr->valid |= PNG_INFO_tRNS; - } -} -#endif - -#ifdef PNG_sPLT_SUPPORTED -void PNGAPI -png_set_sPLT(png_const_structrp png_ptr, - png_inforp info_ptr, png_const_sPLT_tp entries, int nentries) -/* - * entries - array of png_sPLT_t structures - * to be added to the list of palettes - * in the info structure. - * - * nentries - number of palette structures to be - * added. - */ -{ - png_sPLT_tp np; - - png_debug1(1, "in %s storage function", "sPLT"); - - if (png_ptr == NULL || info_ptr == NULL || nentries <= 0 || entries == NULL) - return; - - /* Use the internal realloc function, which checks for all the possible - * overflows. Notice that the parameters are (int) and (size_t) - */ - np = png_voidcast(png_sPLT_tp,png_realloc_array(png_ptr, - info_ptr->splt_palettes, info_ptr->splt_palettes_num, nentries, - sizeof *np)); - - if (np == NULL) - { - /* Out of memory or too many chunks */ - png_chunk_report(png_ptr, "too many sPLT chunks", PNG_CHUNK_WRITE_ERROR); - return; - } - - png_free(png_ptr, info_ptr->splt_palettes); - - info_ptr->splt_palettes = np; - info_ptr->free_me |= PNG_FREE_SPLT; - - np += info_ptr->splt_palettes_num; - - do - { - size_t length; - - /* Skip invalid input entries */ - if (entries->name == NULL || entries->entries == NULL) - { - /* png_handle_sPLT doesn't do this, so this is an app error */ - png_app_error(png_ptr, "png_set_sPLT: invalid sPLT"); - /* Just skip the invalid entry */ - continue; - } - - np->depth = entries->depth; - - /* In the event of out-of-memory just return - there's no point keeping - * on trying to add sPLT chunks. - */ - length = strlen(entries->name) + 1; - np->name = png_voidcast(png_charp, png_malloc_base(png_ptr, length)); - - if (np->name == NULL) - break; - - memcpy(np->name, entries->name, length); - - /* IMPORTANT: we have memory now that won't get freed if something else - * goes wrong; this code must free it. png_malloc_array produces no - * warnings; use a png_chunk_report (below) if there is an error. - */ - np->entries = png_voidcast(png_sPLT_entryp, png_malloc_array(png_ptr, - entries->nentries, sizeof (png_sPLT_entry))); - - if (np->entries == NULL) - { - png_free(png_ptr, np->name); - np->name = NULL; - break; - } - - np->nentries = entries->nentries; - /* This multiply can't overflow because png_malloc_array has already - * checked it when doing the allocation. - */ - memcpy(np->entries, entries->entries, - (unsigned int)entries->nentries * sizeof (png_sPLT_entry)); - - /* Note that 'continue' skips the advance of the out pointer and out - * count, so an invalid entry is not added. - */ - info_ptr->valid |= PNG_INFO_sPLT; - ++(info_ptr->splt_palettes_num); - ++np; - ++entries; - } - while (--nentries); - - if (nentries > 0) - png_chunk_report(png_ptr, "sPLT out of memory", PNG_CHUNK_WRITE_ERROR); -} -#endif /* sPLT */ - -#ifdef PNG_STORE_UNKNOWN_CHUNKS_SUPPORTED -static png_byte -check_location(png_const_structrp png_ptr, int location) -{ - location &= (PNG_HAVE_IHDR|PNG_HAVE_PLTE|PNG_AFTER_IDAT); - - /* New in 1.6.0; copy the location and check it. This is an API - * change; previously the app had to use the - * png_set_unknown_chunk_location API below for each chunk. - */ - if (location == 0 && (png_ptr->mode & PNG_IS_READ_STRUCT) == 0) - { - /* Write struct, so unknown chunks come from the app */ - png_app_warning(png_ptr, - "png_set_unknown_chunks now expects a valid location"); - /* Use the old behavior */ - location = (png_byte)(png_ptr->mode & - (PNG_HAVE_IHDR|PNG_HAVE_PLTE|PNG_AFTER_IDAT)); - } - - /* This need not be an internal error - if the app calls - * png_set_unknown_chunks on a read pointer it must get the location right. - */ - if (location == 0) - png_error(png_ptr, "invalid location in png_set_unknown_chunks"); - - /* Now reduce the location to the top-most set bit by removing each least - * significant bit in turn. - */ - while (location != (location & -location)) - location &= ~(location & -location); - - /* The cast is safe because 'location' is a bit mask and only the low four - * bits are significant. - */ - return (png_byte)location; -} - -void PNGAPI -png_set_unknown_chunks(png_const_structrp png_ptr, - png_inforp info_ptr, png_const_unknown_chunkp unknowns, int num_unknowns) -{ - png_unknown_chunkp np; - - if (png_ptr == NULL || info_ptr == NULL || num_unknowns <= 0 || - unknowns == NULL) - return; - - /* Check for the failure cases where support has been disabled at compile - * time. This code is hardly ever compiled - it's here because - * STORE_UNKNOWN_CHUNKS is set by both read and write code (compiling in this - * code) but may be meaningless if the read or write handling of unknown - * chunks is not compiled in. - */ -# if !defined(PNG_READ_UNKNOWN_CHUNKS_SUPPORTED) && \ - defined(PNG_READ_SUPPORTED) - if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0) - { - png_app_error(png_ptr, "no unknown chunk support on read"); - - return; - } -# endif -# if !defined(PNG_WRITE_UNKNOWN_CHUNKS_SUPPORTED) && \ - defined(PNG_WRITE_SUPPORTED) - if ((png_ptr->mode & PNG_IS_READ_STRUCT) == 0) - { - png_app_error(png_ptr, "no unknown chunk support on write"); - - return; - } -# endif - - /* Prior to 1.6.0 this code used png_malloc_warn; however, this meant that - * unknown critical chunks could be lost with just a warning resulting in - * undefined behavior. Now png_chunk_report is used to provide behavior - * appropriate to read or write. - */ - np = png_voidcast(png_unknown_chunkp, png_realloc_array(png_ptr, - info_ptr->unknown_chunks, info_ptr->unknown_chunks_num, num_unknowns, - sizeof *np)); - - if (np == NULL) - { - png_chunk_report(png_ptr, "too many unknown chunks", - PNG_CHUNK_WRITE_ERROR); - return; - } - - png_free(png_ptr, info_ptr->unknown_chunks); - - info_ptr->unknown_chunks = np; /* safe because it is initialized */ - info_ptr->free_me |= PNG_FREE_UNKN; - - np += info_ptr->unknown_chunks_num; - - /* Increment unknown_chunks_num each time round the loop to protect the - * just-allocated chunk data. - */ - for (; num_unknowns > 0; --num_unknowns, ++unknowns) - { - memcpy(np->name, unknowns->name, (sizeof np->name)); - np->name[(sizeof np->name)-1] = '\0'; - np->location = check_location(png_ptr, unknowns->location); - - if (unknowns->size == 0) - { - np->data = NULL; - np->size = 0; - } - - else - { - np->data = png_voidcast(png_bytep, - png_malloc_base(png_ptr, unknowns->size)); - - if (np->data == NULL) - { - png_chunk_report(png_ptr, "unknown chunk: out of memory", - PNG_CHUNK_WRITE_ERROR); - /* But just skip storing the unknown chunk */ - continue; - } - - memcpy(np->data, unknowns->data, unknowns->size); - np->size = unknowns->size; - } - - /* These increments are skipped on out-of-memory for the data - the - * unknown chunk entry gets overwritten if the png_chunk_report returns. - * This is correct in the read case (the chunk is just dropped.) - */ - ++np; - ++(info_ptr->unknown_chunks_num); - } -} - -void PNGAPI -png_set_unknown_chunk_location(png_const_structrp png_ptr, png_inforp info_ptr, - int chunk, int location) -{ - /* This API is pretty pointless in 1.6.0 because the location can be set - * before the call to png_set_unknown_chunks. - * - * TODO: add a png_app_warning in 1.7 - */ - if (png_ptr != NULL && info_ptr != NULL && chunk >= 0 && - chunk < info_ptr->unknown_chunks_num) - { - if ((location & (PNG_HAVE_IHDR|PNG_HAVE_PLTE|PNG_AFTER_IDAT)) == 0) - { - png_app_error(png_ptr, "invalid unknown chunk location"); - /* Fake out the pre 1.6.0 behavior: */ - if (((unsigned int)location & PNG_HAVE_IDAT) != 0) /* undocumented! */ - location = PNG_AFTER_IDAT; - - else - location = PNG_HAVE_IHDR; /* also undocumented */ - } - - info_ptr->unknown_chunks[chunk].location = - check_location(png_ptr, location); - } -} -#endif /* STORE_UNKNOWN_CHUNKS */ - -#ifdef PNG_MNG_FEATURES_SUPPORTED -png_uint_32 PNGAPI -png_permit_mng_features(png_structrp png_ptr, png_uint_32 mng_features) -{ - png_debug(1, "in png_permit_mng_features"); - - if (png_ptr == NULL) - return 0; - - png_ptr->mng_features_permitted = mng_features & PNG_ALL_MNG_FEATURES; - - return png_ptr->mng_features_permitted; -} -#endif - -#ifdef PNG_HANDLE_AS_UNKNOWN_SUPPORTED -static unsigned int -add_one_chunk(png_bytep list, unsigned int count, png_const_bytep add, int keep) -{ - unsigned int i; - - /* Utility function: update the 'keep' state of a chunk if it is already in - * the list, otherwise add it to the list. - */ - for (i=0; i= PNG_HANDLE_CHUNK_LAST) - { - png_app_error(png_ptr, "png_set_keep_unknown_chunks: invalid keep"); - - return; - } - - if (num_chunks_in <= 0) - { - png_ptr->unknown_default = keep; - - /* '0' means just set the flags, so stop here */ - if (num_chunks_in == 0) - return; - } - - if (num_chunks_in < 0) - { - /* Ignore all unknown chunks and all chunks recognized by - * libpng except for IHDR, PLTE, tRNS, IDAT, and IEND - */ - static const png_byte chunks_to_ignore[] = { - 98, 75, 71, 68, '\0', /* bKGD */ - 99, 72, 82, 77, '\0', /* cHRM */ - 101, 88, 73, 102, '\0', /* eXIf */ - 103, 65, 77, 65, '\0', /* gAMA */ - 104, 73, 83, 84, '\0', /* hIST */ - 105, 67, 67, 80, '\0', /* iCCP */ - 105, 84, 88, 116, '\0', /* iTXt */ - 111, 70, 70, 115, '\0', /* oFFs */ - 112, 67, 65, 76, '\0', /* pCAL */ - 112, 72, 89, 115, '\0', /* pHYs */ - 115, 66, 73, 84, '\0', /* sBIT */ - 115, 67, 65, 76, '\0', /* sCAL */ - 115, 80, 76, 84, '\0', /* sPLT */ - 115, 84, 69, 82, '\0', /* sTER */ - 115, 82, 71, 66, '\0', /* sRGB */ - 116, 69, 88, 116, '\0', /* tEXt */ - 116, 73, 77, 69, '\0', /* tIME */ - 122, 84, 88, 116, '\0' /* zTXt */ - }; - - chunk_list = chunks_to_ignore; - num_chunks = (unsigned int)/*SAFE*/(sizeof chunks_to_ignore)/5U; - } - - else /* num_chunks_in > 0 */ - { - if (chunk_list == NULL) - { - /* Prior to 1.6.0 this was silently ignored, now it is an app_error - * which can be switched off. - */ - png_app_error(png_ptr, "png_set_keep_unknown_chunks: no chunk list"); - - return; - } - - num_chunks = (unsigned int)num_chunks_in; - } - - old_num_chunks = png_ptr->num_chunk_list; - if (png_ptr->chunk_list == NULL) - old_num_chunks = 0; - - /* Since num_chunks is always restricted to UINT_MAX/5 this can't overflow. - */ - if (num_chunks + old_num_chunks > UINT_MAX/5) - { - png_app_error(png_ptr, "png_set_keep_unknown_chunks: too many chunks"); - - return; - } - - /* If these chunks are being reset to the default then no more memory is - * required because add_one_chunk above doesn't extend the list if the 'keep' - * parameter is the default. - */ - if (keep != 0) - { - new_list = png_voidcast(png_bytep, png_malloc(png_ptr, - 5 * (num_chunks + old_num_chunks))); - - if (old_num_chunks > 0) - memcpy(new_list, png_ptr->chunk_list, 5*old_num_chunks); - } - - else if (old_num_chunks > 0) - new_list = png_ptr->chunk_list; - - else - new_list = NULL; - - /* Add the new chunks together with each one's handling code. If the chunk - * already exists the code is updated, otherwise the chunk is added to the - * end. (In libpng 1.6.0 order no longer matters because this code enforces - * the earlier convention that the last setting is the one that is used.) - */ - if (new_list != NULL) - { - png_const_bytep inlist; - png_bytep outlist; - unsigned int i; - - for (i=0; ichunk_list != new_list) - png_free(png_ptr, new_list); - - new_list = NULL; - } - } - - else - num_chunks = 0; - - png_ptr->num_chunk_list = num_chunks; - - if (png_ptr->chunk_list != new_list) - { - if (png_ptr->chunk_list != NULL) - png_free(png_ptr, png_ptr->chunk_list); - - png_ptr->chunk_list = new_list; - } -} -#endif - -#ifdef PNG_READ_USER_CHUNKS_SUPPORTED -void PNGAPI -png_set_read_user_chunk_fn(png_structrp png_ptr, png_voidp user_chunk_ptr, - png_user_chunk_ptr read_user_chunk_fn) -{ - png_debug(1, "in png_set_read_user_chunk_fn"); - - if (png_ptr == NULL) - return; - - png_ptr->read_user_chunk_fn = read_user_chunk_fn; - png_ptr->user_chunk_ptr = user_chunk_ptr; -} -#endif - -#ifdef PNG_INFO_IMAGE_SUPPORTED -void PNGAPI -png_set_rows(png_const_structrp png_ptr, png_inforp info_ptr, - png_bytepp row_pointers) -{ - png_debug(1, "in png_set_rows"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - if (info_ptr->row_pointers != NULL && - (info_ptr->row_pointers != row_pointers)) - png_free_data(png_ptr, info_ptr, PNG_FREE_ROWS, 0); - - info_ptr->row_pointers = row_pointers; - - if (row_pointers != NULL) - info_ptr->valid |= PNG_INFO_IDAT; -} -#endif - -void PNGAPI -png_set_compression_buffer_size(png_structrp png_ptr, size_t size) -{ - png_debug(1, "in png_set_compression_buffer_size"); - - if (png_ptr == NULL) - return; - - if (size == 0 || size > PNG_UINT_31_MAX) - png_error(png_ptr, "invalid compression buffer size"); - -# ifdef PNG_SEQUENTIAL_READ_SUPPORTED - if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0) - { - png_ptr->IDAT_read_size = (png_uint_32)size; /* checked above */ - return; - } -# endif - -# ifdef PNG_WRITE_SUPPORTED - if ((png_ptr->mode & PNG_IS_READ_STRUCT) == 0) - { - if (png_ptr->zowner != 0) - { - png_warning(png_ptr, - "Compression buffer size cannot be changed because it is in use"); - - return; - } - -#ifndef __COVERITY__ - /* Some compilers complain that this is always false. However, it - * can be true when integer overflow happens. - */ - if (size > ZLIB_IO_MAX) - { - png_warning(png_ptr, - "Compression buffer size limited to system maximum"); - size = ZLIB_IO_MAX; /* must fit */ - } -#endif - - if (size < 6) - { - /* Deflate will potentially go into an infinite loop on a SYNC_FLUSH - * if this is permitted. - */ - png_warning(png_ptr, - "Compression buffer size cannot be reduced below 6"); - - return; - } - - if (png_ptr->zbuffer_size != size) - { - png_free_buffer_list(png_ptr, &png_ptr->zbuffer_list); - png_ptr->zbuffer_size = (uInt)size; - } - } -# endif -} - -void PNGAPI -png_set_invalid(png_const_structrp png_ptr, png_inforp info_ptr, int mask) -{ - if (png_ptr != NULL && info_ptr != NULL) - info_ptr->valid &= (unsigned int)(~mask); -} - - -#ifdef PNG_SET_USER_LIMITS_SUPPORTED -/* This function was added to libpng 1.2.6 */ -void PNGAPI -png_set_user_limits(png_structrp png_ptr, png_uint_32 user_width_max, - png_uint_32 user_height_max) -{ - png_debug(1, "in png_set_user_limits"); - - /* Images with dimensions larger than these limits will be - * rejected by png_set_IHDR(). To accept any PNG datastream - * regardless of dimensions, set both limits to 0x7fffffff. - */ - if (png_ptr == NULL) - return; - - png_ptr->user_width_max = user_width_max; - png_ptr->user_height_max = user_height_max; -} - -/* This function was added to libpng 1.4.0 */ -void PNGAPI -png_set_chunk_cache_max(png_structrp png_ptr, png_uint_32 user_chunk_cache_max) -{ - png_debug(1, "in png_set_chunk_cache_max"); - - if (png_ptr != NULL) - png_ptr->user_chunk_cache_max = user_chunk_cache_max; -} - -/* This function was added to libpng 1.4.1 */ -void PNGAPI -png_set_chunk_malloc_max(png_structrp png_ptr, - png_alloc_size_t user_chunk_malloc_max) -{ - png_debug(1, "in png_set_chunk_malloc_max"); - - if (png_ptr != NULL) - png_ptr->user_chunk_malloc_max = user_chunk_malloc_max; -} -#endif /* ?SET_USER_LIMITS */ - - -#ifdef PNG_BENIGN_ERRORS_SUPPORTED -void PNGAPI -png_set_benign_errors(png_structrp png_ptr, int allowed) -{ - png_debug(1, "in png_set_benign_errors"); - - /* If allowed is 1, png_benign_error() is treated as a warning. - * - * If allowed is 0, png_benign_error() is treated as an error (which - * is the default behavior if png_set_benign_errors() is not called). - */ - - if (allowed != 0) - png_ptr->flags |= PNG_FLAG_BENIGN_ERRORS_WARN | - PNG_FLAG_APP_WARNINGS_WARN | PNG_FLAG_APP_ERRORS_WARN; - - else - png_ptr->flags &= ~(PNG_FLAG_BENIGN_ERRORS_WARN | - PNG_FLAG_APP_WARNINGS_WARN | PNG_FLAG_APP_ERRORS_WARN); -} -#endif /* BENIGN_ERRORS */ - -#ifdef PNG_CHECK_FOR_INVALID_INDEX_SUPPORTED - /* Whether to report invalid palette index; added at libng-1.5.10. - * It is possible for an indexed (color-type==3) PNG file to contain - * pixels with invalid (out-of-range) indexes if the PLTE chunk has - * fewer entries than the image's bit-depth would allow. We recover - * from this gracefully by filling any incomplete palette with zeros - * (opaque black). By default, when this occurs libpng will issue - * a benign error. This API can be used to override that behavior. - */ -void PNGAPI -png_set_check_for_invalid_index(png_structrp png_ptr, int allowed) -{ - png_debug(1, "in png_set_check_for_invalid_index"); - - if (allowed > 0) - png_ptr->num_palette_max = 0; - - else - png_ptr->num_palette_max = -1; -} -#endif - -#if defined(PNG_TEXT_SUPPORTED) || defined(PNG_pCAL_SUPPORTED) || \ - defined(PNG_iCCP_SUPPORTED) || defined(PNG_sPLT_SUPPORTED) -/* Check that the tEXt or zTXt keyword is valid per PNG 1.0 specification, - * and if invalid, correct the keyword rather than discarding the entire - * chunk. The PNG 1.0 specification requires keywords 1-79 characters in - * length, forbids leading or trailing whitespace, multiple internal spaces, - * and the non-break space (0x80) from ISO 8859-1. Returns keyword length. - * - * The 'new_key' buffer must be 80 characters in size (for the keyword plus a - * trailing '\0'). If this routine returns 0 then there was no keyword, or a - * valid one could not be generated, and the caller must png_error. - */ -png_uint_32 /* PRIVATE */ -png_check_keyword(png_structrp png_ptr, png_const_charp key, png_bytep new_key) -{ -#ifdef PNG_WARNINGS_SUPPORTED - png_const_charp orig_key = key; -#endif - png_uint_32 key_len = 0; - int bad_character = 0; - int space = 1; - - png_debug(1, "in png_check_keyword"); - - if (key == NULL) - { - *new_key = 0; - return 0; - } - - while (*key && key_len < 79) - { - png_byte ch = (png_byte)*key++; - - if ((ch > 32 && ch <= 126) || (ch >= 161 /*&& ch <= 255*/)) - { - *new_key++ = ch; ++key_len; space = 0; - } - - else if (space == 0) - { - /* A space or an invalid character when one wasn't seen immediately - * before; output just a space. - */ - *new_key++ = 32; ++key_len; space = 1; - - /* If the character was not a space then it is invalid. */ - if (ch != 32) - bad_character = ch; - } - - else if (bad_character == 0) - bad_character = ch; /* just skip it, record the first error */ - } - - if (key_len > 0 && space != 0) /* trailing space */ - { - --key_len; --new_key; - if (bad_character == 0) - bad_character = 32; - } - - /* Terminate the keyword */ - *new_key = 0; - - if (key_len == 0) - return 0; - -#ifdef PNG_WARNINGS_SUPPORTED - /* Try to only output one warning per keyword: */ - if (*key != 0) /* keyword too long */ - png_warning(png_ptr, "keyword truncated"); - - else if (bad_character != 0) - { - PNG_WARNING_PARAMETERS(p) - - png_warning_parameter(p, 1, orig_key); - png_warning_parameter_signed(p, 2, PNG_NUMBER_FORMAT_02x, bad_character); - - png_formatted_warning(png_ptr, p, "keyword \"@1\": bad character '0x@2'"); - } -#else /* !WARNINGS */ - PNG_UNUSED(png_ptr) -#endif /* !WARNINGS */ - - return key_len; -} -#endif /* TEXT || pCAL || iCCP || sPLT */ -#endif /* READ || WRITE */ diff --git a/dep/libpng/src/pngstruct.h b/dep/libpng/src/pngstruct.h deleted file mode 100644 index e591d94d5..000000000 --- a/dep/libpng/src/pngstruct.h +++ /dev/null @@ -1,479 +0,0 @@ - -/* pngstruct.h - header file for PNG reference library - * - * Copyright (c) 2018-2022 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -/* The structure that holds the information to read and write PNG files. - * The only people who need to care about what is inside of this are the - * people who will be modifying the library for their own special needs. - * It should NOT be accessed directly by an application. - */ - -#ifndef PNGSTRUCT_H -#define PNGSTRUCT_H -/* zlib.h defines the structure z_stream, an instance of which is included - * in this structure and is required for decompressing the LZ compressed - * data in PNG files. - */ -#ifndef ZLIB_CONST - /* We must ensure that zlib uses 'const' in declarations. */ -# define ZLIB_CONST -#endif -#include "zlib.h" -#ifdef const - /* zlib.h sometimes #defines const to nothing, undo this. */ -# undef const -#endif - -/* zlib.h has mediocre z_const use before 1.2.6, this stuff is for compatibility - * with older builds. - */ -#if ZLIB_VERNUM < 0x1260 -# define PNGZ_MSG_CAST(s) png_constcast(char*,s) -# define PNGZ_INPUT_CAST(b) png_constcast(png_bytep,b) -#else -# define PNGZ_MSG_CAST(s) (s) -# define PNGZ_INPUT_CAST(b) (b) -#endif - -/* zlib.h declares a magic type 'uInt' that limits the amount of data that zlib - * can handle at once. This type need be no larger than 16 bits (so maximum of - * 65535), this define allows us to discover how big it is, but limited by the - * maximum for size_t. The value can be overridden in a library build - * (pngusr.h, or set it in CPPFLAGS) and it works to set it to a considerably - * lower value (e.g. 255 works). A lower value may help memory usage (slightly) - * and may even improve performance on some systems (and degrade it on others.) - */ -#ifndef ZLIB_IO_MAX -# define ZLIB_IO_MAX ((uInt)-1) -#endif - -#ifdef PNG_WRITE_SUPPORTED -/* The type of a compression buffer list used by the write code. */ -typedef struct png_compression_buffer -{ - struct png_compression_buffer *next; - png_byte output[1]; /* actually zbuf_size */ -} png_compression_buffer, *png_compression_bufferp; - -#define PNG_COMPRESSION_BUFFER_SIZE(pp)\ - (offsetof(png_compression_buffer, output) + (pp)->zbuffer_size) -#endif - -/* Colorspace support; structures used in png_struct, png_info and in internal - * functions to hold and communicate information about the color space. - * - * PNG_COLORSPACE_SUPPORTED is only required if the application will perform - * colorspace corrections, otherwise all the colorspace information can be - * skipped and the size of libpng can be reduced (significantly) by compiling - * out the colorspace support. - */ -#ifdef PNG_COLORSPACE_SUPPORTED -/* The chromaticities of the red, green and blue colorants and the chromaticity - * of the corresponding white point (i.e. of rgb(1.0,1.0,1.0)). - */ -typedef struct png_xy -{ - png_fixed_point redx, redy; - png_fixed_point greenx, greeny; - png_fixed_point bluex, bluey; - png_fixed_point whitex, whitey; -} png_xy; - -/* The same data as above but encoded as CIE XYZ values. When this data comes - * from chromaticities the sum of the Y values is assumed to be 1.0 - */ -typedef struct png_XYZ -{ - png_fixed_point red_X, red_Y, red_Z; - png_fixed_point green_X, green_Y, green_Z; - png_fixed_point blue_X, blue_Y, blue_Z; -} png_XYZ; -#endif /* COLORSPACE */ - -#if defined(PNG_COLORSPACE_SUPPORTED) || defined(PNG_GAMMA_SUPPORTED) -/* A colorspace is all the above plus, potentially, profile information; - * however at present libpng does not use the profile internally so it is only - * stored in the png_info struct (if iCCP is supported.) The rendering intent - * is retained here and is checked. - * - * The file gamma encoding information is also stored here and gamma correction - * is done by libpng, whereas color correction must currently be done by the - * application. - */ -typedef struct png_colorspace -{ -#ifdef PNG_GAMMA_SUPPORTED - png_fixed_point gamma; /* File gamma */ -#endif - -#ifdef PNG_COLORSPACE_SUPPORTED - png_xy end_points_xy; /* End points as chromaticities */ - png_XYZ end_points_XYZ; /* End points as CIE XYZ colorant values */ - png_uint_16 rendering_intent; /* Rendering intent of a profile */ -#endif - - /* Flags are always defined to simplify the code. */ - png_uint_16 flags; /* As defined below */ -} png_colorspace, * PNG_RESTRICT png_colorspacerp; - -typedef const png_colorspace * PNG_RESTRICT png_const_colorspacerp; - -/* General flags for the 'flags' field */ -#define PNG_COLORSPACE_HAVE_GAMMA 0x0001 -#define PNG_COLORSPACE_HAVE_ENDPOINTS 0x0002 -#define PNG_COLORSPACE_HAVE_INTENT 0x0004 -#define PNG_COLORSPACE_FROM_gAMA 0x0008 -#define PNG_COLORSPACE_FROM_cHRM 0x0010 -#define PNG_COLORSPACE_FROM_sRGB 0x0020 -#define PNG_COLORSPACE_ENDPOINTS_MATCH_sRGB 0x0040 -#define PNG_COLORSPACE_MATCHES_sRGB 0x0080 /* exact match on profile */ -#define PNG_COLORSPACE_INVALID 0x8000 -#define PNG_COLORSPACE_CANCEL(flags) (0xffff ^ (flags)) -#endif /* COLORSPACE || GAMMA */ - -struct png_struct_def -{ -#ifdef PNG_SETJMP_SUPPORTED - jmp_buf jmp_buf_local; /* New name in 1.6.0 for jmp_buf in png_struct */ - png_longjmp_ptr longjmp_fn;/* setjmp non-local goto function. */ - jmp_buf *jmp_buf_ptr; /* passed to longjmp_fn */ - size_t jmp_buf_size; /* size of the above, if allocated */ -#endif - png_error_ptr error_fn; /* function for printing errors and aborting */ -#ifdef PNG_WARNINGS_SUPPORTED - png_error_ptr warning_fn; /* function for printing warnings */ -#endif - png_voidp error_ptr; /* user supplied struct for error functions */ - png_rw_ptr write_data_fn; /* function for writing output data */ - png_rw_ptr read_data_fn; /* function for reading input data */ - png_voidp io_ptr; /* ptr to application struct for I/O functions */ - -#ifdef PNG_READ_USER_TRANSFORM_SUPPORTED - png_user_transform_ptr read_user_transform_fn; /* user read transform */ -#endif - -#ifdef PNG_WRITE_USER_TRANSFORM_SUPPORTED - png_user_transform_ptr write_user_transform_fn; /* user write transform */ -#endif - -/* These were added in libpng-1.0.2 */ -#ifdef PNG_USER_TRANSFORM_PTR_SUPPORTED -#if defined(PNG_READ_USER_TRANSFORM_SUPPORTED) || \ - defined(PNG_WRITE_USER_TRANSFORM_SUPPORTED) - png_voidp user_transform_ptr; /* user supplied struct for user transform */ - png_byte user_transform_depth; /* bit depth of user transformed pixels */ - png_byte user_transform_channels; /* channels in user transformed pixels */ -#endif -#endif - - png_uint_32 mode; /* tells us where we are in the PNG file */ - png_uint_32 flags; /* flags indicating various things to libpng */ - png_uint_32 transformations; /* which transformations to perform */ - - png_uint_32 zowner; /* ID (chunk type) of zstream owner, 0 if none */ - z_stream zstream; /* decompression structure */ - -#ifdef PNG_WRITE_SUPPORTED - png_compression_bufferp zbuffer_list; /* Created on demand during write */ - uInt zbuffer_size; /* size of the actual buffer */ - - int zlib_level; /* holds zlib compression level */ - int zlib_method; /* holds zlib compression method */ - int zlib_window_bits; /* holds zlib compression window bits */ - int zlib_mem_level; /* holds zlib compression memory level */ - int zlib_strategy; /* holds zlib compression strategy */ -#endif -/* Added at libpng 1.5.4 */ -#ifdef PNG_WRITE_CUSTOMIZE_ZTXT_COMPRESSION_SUPPORTED - int zlib_text_level; /* holds zlib compression level */ - int zlib_text_method; /* holds zlib compression method */ - int zlib_text_window_bits; /* holds zlib compression window bits */ - int zlib_text_mem_level; /* holds zlib compression memory level */ - int zlib_text_strategy; /* holds zlib compression strategy */ -#endif -/* End of material added at libpng 1.5.4 */ -/* Added at libpng 1.6.0 */ -#ifdef PNG_WRITE_SUPPORTED - int zlib_set_level; /* Actual values set into the zstream on write */ - int zlib_set_method; - int zlib_set_window_bits; - int zlib_set_mem_level; - int zlib_set_strategy; -#endif - - png_uint_32 width; /* width of image in pixels */ - png_uint_32 height; /* height of image in pixels */ - png_uint_32 num_rows; /* number of rows in current pass */ - png_uint_32 usr_width; /* width of row at start of write */ - size_t rowbytes; /* size of row in bytes */ - png_uint_32 iwidth; /* width of current interlaced row in pixels */ - png_uint_32 row_number; /* current row in interlace pass */ - png_uint_32 chunk_name; /* PNG_CHUNK() id of current chunk */ - png_bytep prev_row; /* buffer to save previous (unfiltered) row. - * While reading this is a pointer into - * big_prev_row; while writing it is separately - * allocated if needed. - */ - png_bytep row_buf; /* buffer to save current (unfiltered) row. - * While reading, this is a pointer into - * big_row_buf; while writing it is separately - * allocated. - */ -#ifdef PNG_WRITE_FILTER_SUPPORTED - png_bytep try_row; /* buffer to save trial row when filtering */ - png_bytep tst_row; /* buffer to save best trial row when filtering */ -#endif - size_t info_rowbytes; /* Added in 1.5.4: cache of updated row bytes */ - - png_uint_32 idat_size; /* current IDAT size for read */ - png_uint_32 crc; /* current chunk CRC value */ - png_colorp palette; /* palette from the input file */ - png_uint_16 num_palette; /* number of color entries in palette */ - -/* Added at libpng-1.5.10 */ -#ifdef PNG_CHECK_FOR_INVALID_INDEX_SUPPORTED - int num_palette_max; /* maximum palette index found in IDAT */ -#endif - - png_uint_16 num_trans; /* number of transparency values */ - png_byte compression; /* file compression type (always 0) */ - png_byte filter; /* file filter type (always 0) */ - png_byte interlaced; /* PNG_INTERLACE_NONE, PNG_INTERLACE_ADAM7 */ - png_byte pass; /* current interlace pass (0 - 6) */ - png_byte do_filter; /* row filter flags (see PNG_FILTER_ in png.h ) */ - png_byte color_type; /* color type of file */ - png_byte bit_depth; /* bit depth of file */ - png_byte usr_bit_depth; /* bit depth of users row: write only */ - png_byte pixel_depth; /* number of bits per pixel */ - png_byte channels; /* number of channels in file */ -#ifdef PNG_WRITE_SUPPORTED - png_byte usr_channels; /* channels at start of write: write only */ -#endif - png_byte sig_bytes; /* magic bytes read/written from start of file */ - png_byte maximum_pixel_depth; - /* pixel depth used for the row buffers */ - png_byte transformed_pixel_depth; - /* pixel depth after read/write transforms */ -#if ZLIB_VERNUM >= 0x1240 - png_byte zstream_start; /* at start of an input zlib stream */ -#endif /* Zlib >= 1.2.4 */ -#if defined(PNG_READ_FILLER_SUPPORTED) || defined(PNG_WRITE_FILLER_SUPPORTED) - png_uint_16 filler; /* filler bytes for pixel expansion */ -#endif - -#if defined(PNG_bKGD_SUPPORTED) || defined(PNG_READ_BACKGROUND_SUPPORTED) ||\ - defined(PNG_READ_ALPHA_MODE_SUPPORTED) - png_byte background_gamma_type; - png_fixed_point background_gamma; - png_color_16 background; /* background color in screen gamma space */ -#ifdef PNG_READ_GAMMA_SUPPORTED - png_color_16 background_1; /* background normalized to gamma 1.0 */ -#endif -#endif /* bKGD */ - -#ifdef PNG_WRITE_FLUSH_SUPPORTED - png_flush_ptr output_flush_fn; /* Function for flushing output */ - png_uint_32 flush_dist; /* how many rows apart to flush, 0 - no flush */ - png_uint_32 flush_rows; /* number of rows written since last flush */ -#endif - -#ifdef PNG_READ_GAMMA_SUPPORTED - int gamma_shift; /* number of "insignificant" bits in 16-bit gamma */ - png_fixed_point screen_gamma; /* screen gamma value (display_exponent) */ - - png_bytep gamma_table; /* gamma table for 8-bit depth files */ - png_uint_16pp gamma_16_table; /* gamma table for 16-bit depth files */ -#if defined(PNG_READ_BACKGROUND_SUPPORTED) || \ - defined(PNG_READ_ALPHA_MODE_SUPPORTED) || \ - defined(PNG_READ_RGB_TO_GRAY_SUPPORTED) - png_bytep gamma_from_1; /* converts from 1.0 to screen */ - png_bytep gamma_to_1; /* converts from file to 1.0 */ - png_uint_16pp gamma_16_from_1; /* converts from 1.0 to screen */ - png_uint_16pp gamma_16_to_1; /* converts from file to 1.0 */ -#endif /* READ_BACKGROUND || READ_ALPHA_MODE || RGB_TO_GRAY */ -#endif - -#if defined(PNG_READ_GAMMA_SUPPORTED) || defined(PNG_sBIT_SUPPORTED) - png_color_8 sig_bit; /* significant bits in each available channel */ -#endif - -#if defined(PNG_READ_SHIFT_SUPPORTED) || defined(PNG_WRITE_SHIFT_SUPPORTED) - png_color_8 shift; /* shift for significant bit transformation */ -#endif - -#if defined(PNG_tRNS_SUPPORTED) || defined(PNG_READ_BACKGROUND_SUPPORTED) \ - || defined(PNG_READ_EXPAND_SUPPORTED) || defined(PNG_READ_BACKGROUND_SUPPORTED) - png_bytep trans_alpha; /* alpha values for paletted files */ - png_color_16 trans_color; /* transparent color for non-paletted files */ -#endif - - png_read_status_ptr read_row_fn; /* called after each row is decoded */ - png_write_status_ptr write_row_fn; /* called after each row is encoded */ -#ifdef PNG_PROGRESSIVE_READ_SUPPORTED - png_progressive_info_ptr info_fn; /* called after header data fully read */ - png_progressive_row_ptr row_fn; /* called after a prog. row is decoded */ - png_progressive_end_ptr end_fn; /* called after image is complete */ - png_bytep save_buffer_ptr; /* current location in save_buffer */ - png_bytep save_buffer; /* buffer for previously read data */ - png_bytep current_buffer_ptr; /* current location in current_buffer */ - png_bytep current_buffer; /* buffer for recently used data */ - png_uint_32 push_length; /* size of current input chunk */ - png_uint_32 skip_length; /* bytes to skip in input data */ - size_t save_buffer_size; /* amount of data now in save_buffer */ - size_t save_buffer_max; /* total size of save_buffer */ - size_t buffer_size; /* total amount of available input data */ - size_t current_buffer_size; /* amount of data now in current_buffer */ - int process_mode; /* what push library is currently doing */ - int cur_palette; /* current push library palette index */ -#endif /* PROGRESSIVE_READ */ - -#ifdef PNG_READ_QUANTIZE_SUPPORTED - png_bytep palette_lookup; /* lookup table for quantizing */ - png_bytep quantize_index; /* index translation for palette files */ -#endif - -/* Options */ -#ifdef PNG_SET_OPTION_SUPPORTED - png_uint_32 options; /* On/off state (up to 16 options) */ -#endif - -#if PNG_LIBPNG_VER < 10700 -/* To do: remove this from libpng-1.7 */ -#ifdef PNG_TIME_RFC1123_SUPPORTED - char time_buffer[29]; /* String to hold RFC 1123 time text */ -#endif -#endif - -/* New members added in libpng-1.0.6 */ - - png_uint_32 free_me; /* flags items libpng is responsible for freeing */ - -#ifdef PNG_USER_CHUNKS_SUPPORTED - png_voidp user_chunk_ptr; -#ifdef PNG_READ_USER_CHUNKS_SUPPORTED - png_user_chunk_ptr read_user_chunk_fn; /* user read chunk handler */ -#endif -#endif - -#ifdef PNG_SET_UNKNOWN_CHUNKS_SUPPORTED - int unknown_default; /* As PNG_HANDLE_* */ - unsigned int num_chunk_list; /* Number of entries in the list */ - png_bytep chunk_list; /* List of png_byte[5]; the textual chunk name - * followed by a PNG_HANDLE_* byte */ -#endif - -/* New members added in libpng-1.0.3 */ -#ifdef PNG_READ_RGB_TO_GRAY_SUPPORTED - png_byte rgb_to_gray_status; - /* Added in libpng 1.5.5 to record setting of coefficients: */ - png_byte rgb_to_gray_coefficients_set; - /* These were changed from png_byte in libpng-1.0.6 */ - png_uint_16 rgb_to_gray_red_coeff; - png_uint_16 rgb_to_gray_green_coeff; - /* deleted in 1.5.5: rgb_to_gray_blue_coeff; */ -#endif - -/* New member added in libpng-1.6.36 */ -#if defined(PNG_READ_EXPAND_SUPPORTED) && \ - defined(PNG_ARM_NEON_IMPLEMENTATION) - png_bytep riffled_palette; /* buffer for accelerated palette expansion */ -#endif - -/* New member added in libpng-1.0.4 (renamed in 1.0.9) */ -#if defined(PNG_MNG_FEATURES_SUPPORTED) -/* Changed from png_byte to png_uint_32 at version 1.2.0 */ - png_uint_32 mng_features_permitted; -#endif - -/* New member added in libpng-1.0.9, ifdef'ed out in 1.0.12, enabled in 1.2.0 */ -#ifdef PNG_MNG_FEATURES_SUPPORTED - png_byte filter_type; -#endif - -/* New members added in libpng-1.2.0 */ - -/* New members added in libpng-1.0.2 but first enabled by default in 1.2.0 */ -#ifdef PNG_USER_MEM_SUPPORTED - png_voidp mem_ptr; /* user supplied struct for mem functions */ - png_malloc_ptr malloc_fn; /* function for allocating memory */ - png_free_ptr free_fn; /* function for freeing memory */ -#endif - -/* New member added in libpng-1.0.13 and 1.2.0 */ - png_bytep big_row_buf; /* buffer to save current (unfiltered) row */ - -#ifdef PNG_READ_QUANTIZE_SUPPORTED -/* The following three members were added at version 1.0.14 and 1.2.4 */ - png_bytep quantize_sort; /* working sort array */ - png_bytep index_to_palette; /* where the original index currently is - in the palette */ - png_bytep palette_to_index; /* which original index points to this - palette color */ -#endif - -/* New members added in libpng-1.0.16 and 1.2.6 */ - png_byte compression_type; - -#ifdef PNG_USER_LIMITS_SUPPORTED - png_uint_32 user_width_max; - png_uint_32 user_height_max; - - /* Added in libpng-1.4.0: Total number of sPLT, text, and unknown - * chunks that can be stored (0 means unlimited). - */ - png_uint_32 user_chunk_cache_max; - - /* Total memory that a zTXt, sPLT, iTXt, iCCP, or unknown chunk - * can occupy when decompressed. 0 means unlimited. - */ - png_alloc_size_t user_chunk_malloc_max; -#endif - -/* New member added in libpng-1.0.25 and 1.2.17 */ -#ifdef PNG_READ_UNKNOWN_CHUNKS_SUPPORTED - /* Temporary storage for unknown chunk that the library doesn't recognize, - * used while reading the chunk. - */ - png_unknown_chunk unknown_chunk; -#endif - -/* New member added in libpng-1.2.26 */ - size_t old_big_row_buf_size; - -#ifdef PNG_READ_SUPPORTED -/* New member added in libpng-1.2.30 */ - png_bytep read_buffer; /* buffer for reading chunk data */ - png_alloc_size_t read_buffer_size; /* current size of the buffer */ -#endif -#ifdef PNG_SEQUENTIAL_READ_SUPPORTED - uInt IDAT_read_size; /* limit on read buffer size for IDAT */ -#endif - -#ifdef PNG_IO_STATE_SUPPORTED -/* New member added in libpng-1.4.0 */ - png_uint_32 io_state; -#endif - -/* New member added in libpng-1.5.6 */ - png_bytep big_prev_row; - -/* New member added in libpng-1.5.7 */ - void (*read_filter[PNG_FILTER_VALUE_LAST-1])(png_row_infop row_info, - png_bytep row, png_const_bytep prev_row); - -#ifdef PNG_READ_SUPPORTED -#if defined(PNG_COLORSPACE_SUPPORTED) || defined(PNG_GAMMA_SUPPORTED) - png_colorspace colorspace; -#endif -#endif -}; -#endif /* PNGSTRUCT_H */ diff --git a/dep/libpng/src/pngtrans.c b/dep/libpng/src/pngtrans.c deleted file mode 100644 index 62cb21edf..000000000 --- a/dep/libpng/src/pngtrans.c +++ /dev/null @@ -1,868 +0,0 @@ - -/* pngtrans.c - transforms the data in a row (used by both readers and writers) - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -#include "pngpriv.h" - -#if defined(PNG_READ_SUPPORTED) || defined(PNG_WRITE_SUPPORTED) - -#if defined(PNG_READ_BGR_SUPPORTED) || defined(PNG_WRITE_BGR_SUPPORTED) -/* Turn on BGR-to-RGB mapping */ -void PNGAPI -png_set_bgr(png_structrp png_ptr) -{ - png_debug(1, "in png_set_bgr"); - - if (png_ptr == NULL) - return; - - png_ptr->transformations |= PNG_BGR; -} -#endif - -#if defined(PNG_READ_SWAP_SUPPORTED) || defined(PNG_WRITE_SWAP_SUPPORTED) -/* Turn on 16-bit byte swapping */ -void PNGAPI -png_set_swap(png_structrp png_ptr) -{ - png_debug(1, "in png_set_swap"); - - if (png_ptr == NULL) - return; - - if (png_ptr->bit_depth == 16) - png_ptr->transformations |= PNG_SWAP_BYTES; -} -#endif - -#if defined(PNG_READ_PACK_SUPPORTED) || defined(PNG_WRITE_PACK_SUPPORTED) -/* Turn on pixel packing */ -void PNGAPI -png_set_packing(png_structrp png_ptr) -{ - png_debug(1, "in png_set_packing"); - - if (png_ptr == NULL) - return; - - if (png_ptr->bit_depth < 8) - { - png_ptr->transformations |= PNG_PACK; -# ifdef PNG_WRITE_SUPPORTED - png_ptr->usr_bit_depth = 8; -# endif - } -} -#endif - -#if defined(PNG_READ_PACKSWAP_SUPPORTED)||defined(PNG_WRITE_PACKSWAP_SUPPORTED) -/* Turn on packed pixel swapping */ -void PNGAPI -png_set_packswap(png_structrp png_ptr) -{ - png_debug(1, "in png_set_packswap"); - - if (png_ptr == NULL) - return; - - if (png_ptr->bit_depth < 8) - png_ptr->transformations |= PNG_PACKSWAP; -} -#endif - -#if defined(PNG_READ_SHIFT_SUPPORTED) || defined(PNG_WRITE_SHIFT_SUPPORTED) -void PNGAPI -png_set_shift(png_structrp png_ptr, png_const_color_8p true_bits) -{ - png_debug(1, "in png_set_shift"); - - if (png_ptr == NULL) - return; - - png_ptr->transformations |= PNG_SHIFT; - png_ptr->shift = *true_bits; -} -#endif - -#if defined(PNG_READ_INTERLACING_SUPPORTED) || \ - defined(PNG_WRITE_INTERLACING_SUPPORTED) -int PNGAPI -png_set_interlace_handling(png_structrp png_ptr) -{ - png_debug(1, "in png_set_interlace handling"); - - if (png_ptr != 0 && png_ptr->interlaced != 0) - { - png_ptr->transformations |= PNG_INTERLACE; - return 7; - } - - return 1; -} -#endif - -#if defined(PNG_READ_FILLER_SUPPORTED) || defined(PNG_WRITE_FILLER_SUPPORTED) -/* Add a filler byte on read, or remove a filler or alpha byte on write. - * The filler type has changed in v0.95 to allow future 2-byte fillers - * for 48-bit input data, as well as to avoid problems with some compilers - * that don't like bytes as parameters. - */ -void PNGAPI -png_set_filler(png_structrp png_ptr, png_uint_32 filler, int filler_loc) -{ - png_debug(1, "in png_set_filler"); - - if (png_ptr == NULL) - return; - - /* In libpng 1.6 it is possible to determine whether this is a read or write - * operation and therefore to do more checking here for a valid call. - */ - if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0) - { -# ifdef PNG_READ_FILLER_SUPPORTED - /* On read png_set_filler is always valid, regardless of the base PNG - * format, because other transformations can give a format where the - * filler code can execute (basically an 8 or 16-bit component RGB or G - * format.) - * - * NOTE: usr_channels is not used by the read code! (This has led to - * confusion in the past.) The filler is only used in the read code. - */ - png_ptr->filler = (png_uint_16)filler; -# else - png_app_error(png_ptr, "png_set_filler not supported on read"); - PNG_UNUSED(filler) /* not used in the write case */ - return; -# endif - } - - else /* write */ - { -# ifdef PNG_WRITE_FILLER_SUPPORTED - /* On write the usr_channels parameter must be set correctly at the - * start to record the number of channels in the app-supplied data. - */ - switch (png_ptr->color_type) - { - case PNG_COLOR_TYPE_RGB: - png_ptr->usr_channels = 4; - break; - - case PNG_COLOR_TYPE_GRAY: - if (png_ptr->bit_depth >= 8) - { - png_ptr->usr_channels = 2; - break; - } - - else - { - /* There simply isn't any code in libpng to strip out bits - * from bytes when the components are less than a byte in - * size! - */ - png_app_error(png_ptr, - "png_set_filler is invalid for" - " low bit depth gray output"); - return; - } - - default: - png_app_error(png_ptr, - "png_set_filler: inappropriate color type"); - return; - } -# else - png_app_error(png_ptr, "png_set_filler not supported on write"); - return; -# endif - } - - /* Here on success - libpng supports the operation, set the transformation - * and the flag to say where the filler channel is. - */ - png_ptr->transformations |= PNG_FILLER; - - if (filler_loc == PNG_FILLER_AFTER) - png_ptr->flags |= PNG_FLAG_FILLER_AFTER; - - else - png_ptr->flags &= ~PNG_FLAG_FILLER_AFTER; -} - -/* Added to libpng-1.2.7 */ -void PNGAPI -png_set_add_alpha(png_structrp png_ptr, png_uint_32 filler, int filler_loc) -{ - png_debug(1, "in png_set_add_alpha"); - - if (png_ptr == NULL) - return; - - png_set_filler(png_ptr, filler, filler_loc); - /* The above may fail to do anything. */ - if ((png_ptr->transformations & PNG_FILLER) != 0) - png_ptr->transformations |= PNG_ADD_ALPHA; -} - -#endif - -#if defined(PNG_READ_SWAP_ALPHA_SUPPORTED) || \ - defined(PNG_WRITE_SWAP_ALPHA_SUPPORTED) -void PNGAPI -png_set_swap_alpha(png_structrp png_ptr) -{ - png_debug(1, "in png_set_swap_alpha"); - - if (png_ptr == NULL) - return; - - png_ptr->transformations |= PNG_SWAP_ALPHA; -} -#endif - -#if defined(PNG_READ_INVERT_ALPHA_SUPPORTED) || \ - defined(PNG_WRITE_INVERT_ALPHA_SUPPORTED) -void PNGAPI -png_set_invert_alpha(png_structrp png_ptr) -{ - png_debug(1, "in png_set_invert_alpha"); - - if (png_ptr == NULL) - return; - - png_ptr->transformations |= PNG_INVERT_ALPHA; -} -#endif - -#if defined(PNG_READ_INVERT_SUPPORTED) || defined(PNG_WRITE_INVERT_SUPPORTED) -void PNGAPI -png_set_invert_mono(png_structrp png_ptr) -{ - png_debug(1, "in png_set_invert_mono"); - - if (png_ptr == NULL) - return; - - png_ptr->transformations |= PNG_INVERT_MONO; -} - -/* Invert monochrome grayscale data */ -void /* PRIVATE */ -png_do_invert(png_row_infop row_info, png_bytep row) -{ - png_debug(1, "in png_do_invert"); - - /* This test removed from libpng version 1.0.13 and 1.2.0: - * if (row_info->bit_depth == 1 && - */ - if (row_info->color_type == PNG_COLOR_TYPE_GRAY) - { - png_bytep rp = row; - size_t i; - size_t istop = row_info->rowbytes; - - for (i = 0; i < istop; i++) - { - *rp = (png_byte)(~(*rp)); - rp++; - } - } - - else if (row_info->color_type == PNG_COLOR_TYPE_GRAY_ALPHA && - row_info->bit_depth == 8) - { - png_bytep rp = row; - size_t i; - size_t istop = row_info->rowbytes; - - for (i = 0; i < istop; i += 2) - { - *rp = (png_byte)(~(*rp)); - rp += 2; - } - } - -#ifdef PNG_16BIT_SUPPORTED - else if (row_info->color_type == PNG_COLOR_TYPE_GRAY_ALPHA && - row_info->bit_depth == 16) - { - png_bytep rp = row; - size_t i; - size_t istop = row_info->rowbytes; - - for (i = 0; i < istop; i += 4) - { - *rp = (png_byte)(~(*rp)); - *(rp + 1) = (png_byte)(~(*(rp + 1))); - rp += 4; - } - } -#endif -} -#endif - -#ifdef PNG_16BIT_SUPPORTED -#if defined(PNG_READ_SWAP_SUPPORTED) || defined(PNG_WRITE_SWAP_SUPPORTED) -/* Swaps byte order on 16-bit depth images */ -void /* PRIVATE */ -png_do_swap(png_row_infop row_info, png_bytep row) -{ - png_debug(1, "in png_do_swap"); - - if (row_info->bit_depth == 16) - { - png_bytep rp = row; - png_uint_32 i; - png_uint_32 istop= row_info->width * row_info->channels; - - for (i = 0; i < istop; i++, rp += 2) - { -#ifdef PNG_BUILTIN_BSWAP16_SUPPORTED - /* Feature added to libpng-1.6.11 for testing purposes, not - * enabled by default. - */ - *(png_uint_16*)rp = __builtin_bswap16(*(png_uint_16*)rp); -#else - png_byte t = *rp; - *rp = *(rp + 1); - *(rp + 1) = t; -#endif - } - } -} -#endif -#endif - -#if defined(PNG_READ_PACKSWAP_SUPPORTED)||defined(PNG_WRITE_PACKSWAP_SUPPORTED) -static const png_byte onebppswaptable[256] = { - 0x00, 0x80, 0x40, 0xC0, 0x20, 0xA0, 0x60, 0xE0, - 0x10, 0x90, 0x50, 0xD0, 0x30, 0xB0, 0x70, 0xF0, - 0x08, 0x88, 0x48, 0xC8, 0x28, 0xA8, 0x68, 0xE8, - 0x18, 0x98, 0x58, 0xD8, 0x38, 0xB8, 0x78, 0xF8, - 0x04, 0x84, 0x44, 0xC4, 0x24, 0xA4, 0x64, 0xE4, - 0x14, 0x94, 0x54, 0xD4, 0x34, 0xB4, 0x74, 0xF4, - 0x0C, 0x8C, 0x4C, 0xCC, 0x2C, 0xAC, 0x6C, 0xEC, - 0x1C, 0x9C, 0x5C, 0xDC, 0x3C, 0xBC, 0x7C, 0xFC, - 0x02, 0x82, 0x42, 0xC2, 0x22, 0xA2, 0x62, 0xE2, - 0x12, 0x92, 0x52, 0xD2, 0x32, 0xB2, 0x72, 0xF2, - 0x0A, 0x8A, 0x4A, 0xCA, 0x2A, 0xAA, 0x6A, 0xEA, - 0x1A, 0x9A, 0x5A, 0xDA, 0x3A, 0xBA, 0x7A, 0xFA, - 0x06, 0x86, 0x46, 0xC6, 0x26, 0xA6, 0x66, 0xE6, - 0x16, 0x96, 0x56, 0xD6, 0x36, 0xB6, 0x76, 0xF6, - 0x0E, 0x8E, 0x4E, 0xCE, 0x2E, 0xAE, 0x6E, 0xEE, - 0x1E, 0x9E, 0x5E, 0xDE, 0x3E, 0xBE, 0x7E, 0xFE, - 0x01, 0x81, 0x41, 0xC1, 0x21, 0xA1, 0x61, 0xE1, - 0x11, 0x91, 0x51, 0xD1, 0x31, 0xB1, 0x71, 0xF1, - 0x09, 0x89, 0x49, 0xC9, 0x29, 0xA9, 0x69, 0xE9, - 0x19, 0x99, 0x59, 0xD9, 0x39, 0xB9, 0x79, 0xF9, - 0x05, 0x85, 0x45, 0xC5, 0x25, 0xA5, 0x65, 0xE5, - 0x15, 0x95, 0x55, 0xD5, 0x35, 0xB5, 0x75, 0xF5, - 0x0D, 0x8D, 0x4D, 0xCD, 0x2D, 0xAD, 0x6D, 0xED, - 0x1D, 0x9D, 0x5D, 0xDD, 0x3D, 0xBD, 0x7D, 0xFD, - 0x03, 0x83, 0x43, 0xC3, 0x23, 0xA3, 0x63, 0xE3, - 0x13, 0x93, 0x53, 0xD3, 0x33, 0xB3, 0x73, 0xF3, - 0x0B, 0x8B, 0x4B, 0xCB, 0x2B, 0xAB, 0x6B, 0xEB, - 0x1B, 0x9B, 0x5B, 0xDB, 0x3B, 0xBB, 0x7B, 0xFB, - 0x07, 0x87, 0x47, 0xC7, 0x27, 0xA7, 0x67, 0xE7, - 0x17, 0x97, 0x57, 0xD7, 0x37, 0xB7, 0x77, 0xF7, - 0x0F, 0x8F, 0x4F, 0xCF, 0x2F, 0xAF, 0x6F, 0xEF, - 0x1F, 0x9F, 0x5F, 0xDF, 0x3F, 0xBF, 0x7F, 0xFF -}; - -static const png_byte twobppswaptable[256] = { - 0x00, 0x40, 0x80, 0xC0, 0x10, 0x50, 0x90, 0xD0, - 0x20, 0x60, 0xA0, 0xE0, 0x30, 0x70, 0xB0, 0xF0, - 0x04, 0x44, 0x84, 0xC4, 0x14, 0x54, 0x94, 0xD4, - 0x24, 0x64, 0xA4, 0xE4, 0x34, 0x74, 0xB4, 0xF4, - 0x08, 0x48, 0x88, 0xC8, 0x18, 0x58, 0x98, 0xD8, - 0x28, 0x68, 0xA8, 0xE8, 0x38, 0x78, 0xB8, 0xF8, - 0x0C, 0x4C, 0x8C, 0xCC, 0x1C, 0x5C, 0x9C, 0xDC, - 0x2C, 0x6C, 0xAC, 0xEC, 0x3C, 0x7C, 0xBC, 0xFC, - 0x01, 0x41, 0x81, 0xC1, 0x11, 0x51, 0x91, 0xD1, - 0x21, 0x61, 0xA1, 0xE1, 0x31, 0x71, 0xB1, 0xF1, - 0x05, 0x45, 0x85, 0xC5, 0x15, 0x55, 0x95, 0xD5, - 0x25, 0x65, 0xA5, 0xE5, 0x35, 0x75, 0xB5, 0xF5, - 0x09, 0x49, 0x89, 0xC9, 0x19, 0x59, 0x99, 0xD9, - 0x29, 0x69, 0xA9, 0xE9, 0x39, 0x79, 0xB9, 0xF9, - 0x0D, 0x4D, 0x8D, 0xCD, 0x1D, 0x5D, 0x9D, 0xDD, - 0x2D, 0x6D, 0xAD, 0xED, 0x3D, 0x7D, 0xBD, 0xFD, - 0x02, 0x42, 0x82, 0xC2, 0x12, 0x52, 0x92, 0xD2, - 0x22, 0x62, 0xA2, 0xE2, 0x32, 0x72, 0xB2, 0xF2, - 0x06, 0x46, 0x86, 0xC6, 0x16, 0x56, 0x96, 0xD6, - 0x26, 0x66, 0xA6, 0xE6, 0x36, 0x76, 0xB6, 0xF6, - 0x0A, 0x4A, 0x8A, 0xCA, 0x1A, 0x5A, 0x9A, 0xDA, - 0x2A, 0x6A, 0xAA, 0xEA, 0x3A, 0x7A, 0xBA, 0xFA, - 0x0E, 0x4E, 0x8E, 0xCE, 0x1E, 0x5E, 0x9E, 0xDE, - 0x2E, 0x6E, 0xAE, 0xEE, 0x3E, 0x7E, 0xBE, 0xFE, - 0x03, 0x43, 0x83, 0xC3, 0x13, 0x53, 0x93, 0xD3, - 0x23, 0x63, 0xA3, 0xE3, 0x33, 0x73, 0xB3, 0xF3, - 0x07, 0x47, 0x87, 0xC7, 0x17, 0x57, 0x97, 0xD7, - 0x27, 0x67, 0xA7, 0xE7, 0x37, 0x77, 0xB7, 0xF7, - 0x0B, 0x4B, 0x8B, 0xCB, 0x1B, 0x5B, 0x9B, 0xDB, - 0x2B, 0x6B, 0xAB, 0xEB, 0x3B, 0x7B, 0xBB, 0xFB, - 0x0F, 0x4F, 0x8F, 0xCF, 0x1F, 0x5F, 0x9F, 0xDF, - 0x2F, 0x6F, 0xAF, 0xEF, 0x3F, 0x7F, 0xBF, 0xFF -}; - -static const png_byte fourbppswaptable[256] = { - 0x00, 0x10, 0x20, 0x30, 0x40, 0x50, 0x60, 0x70, - 0x80, 0x90, 0xA0, 0xB0, 0xC0, 0xD0, 0xE0, 0xF0, - 0x01, 0x11, 0x21, 0x31, 0x41, 0x51, 0x61, 0x71, - 0x81, 0x91, 0xA1, 0xB1, 0xC1, 0xD1, 0xE1, 0xF1, - 0x02, 0x12, 0x22, 0x32, 0x42, 0x52, 0x62, 0x72, - 0x82, 0x92, 0xA2, 0xB2, 0xC2, 0xD2, 0xE2, 0xF2, - 0x03, 0x13, 0x23, 0x33, 0x43, 0x53, 0x63, 0x73, - 0x83, 0x93, 0xA3, 0xB3, 0xC3, 0xD3, 0xE3, 0xF3, - 0x04, 0x14, 0x24, 0x34, 0x44, 0x54, 0x64, 0x74, - 0x84, 0x94, 0xA4, 0xB4, 0xC4, 0xD4, 0xE4, 0xF4, - 0x05, 0x15, 0x25, 0x35, 0x45, 0x55, 0x65, 0x75, - 0x85, 0x95, 0xA5, 0xB5, 0xC5, 0xD5, 0xE5, 0xF5, - 0x06, 0x16, 0x26, 0x36, 0x46, 0x56, 0x66, 0x76, - 0x86, 0x96, 0xA6, 0xB6, 0xC6, 0xD6, 0xE6, 0xF6, - 0x07, 0x17, 0x27, 0x37, 0x47, 0x57, 0x67, 0x77, - 0x87, 0x97, 0xA7, 0xB7, 0xC7, 0xD7, 0xE7, 0xF7, - 0x08, 0x18, 0x28, 0x38, 0x48, 0x58, 0x68, 0x78, - 0x88, 0x98, 0xA8, 0xB8, 0xC8, 0xD8, 0xE8, 0xF8, - 0x09, 0x19, 0x29, 0x39, 0x49, 0x59, 0x69, 0x79, - 0x89, 0x99, 0xA9, 0xB9, 0xC9, 0xD9, 0xE9, 0xF9, - 0x0A, 0x1A, 0x2A, 0x3A, 0x4A, 0x5A, 0x6A, 0x7A, - 0x8A, 0x9A, 0xAA, 0xBA, 0xCA, 0xDA, 0xEA, 0xFA, - 0x0B, 0x1B, 0x2B, 0x3B, 0x4B, 0x5B, 0x6B, 0x7B, - 0x8B, 0x9B, 0xAB, 0xBB, 0xCB, 0xDB, 0xEB, 0xFB, - 0x0C, 0x1C, 0x2C, 0x3C, 0x4C, 0x5C, 0x6C, 0x7C, - 0x8C, 0x9C, 0xAC, 0xBC, 0xCC, 0xDC, 0xEC, 0xFC, - 0x0D, 0x1D, 0x2D, 0x3D, 0x4D, 0x5D, 0x6D, 0x7D, - 0x8D, 0x9D, 0xAD, 0xBD, 0xCD, 0xDD, 0xED, 0xFD, - 0x0E, 0x1E, 0x2E, 0x3E, 0x4E, 0x5E, 0x6E, 0x7E, - 0x8E, 0x9E, 0xAE, 0xBE, 0xCE, 0xDE, 0xEE, 0xFE, - 0x0F, 0x1F, 0x2F, 0x3F, 0x4F, 0x5F, 0x6F, 0x7F, - 0x8F, 0x9F, 0xAF, 0xBF, 0xCF, 0xDF, 0xEF, 0xFF -}; - -/* Swaps pixel packing order within bytes */ -void /* PRIVATE */ -png_do_packswap(png_row_infop row_info, png_bytep row) -{ - png_debug(1, "in png_do_packswap"); - - if (row_info->bit_depth < 8) - { - png_bytep rp; - png_const_bytep end, table; - - end = row + row_info->rowbytes; - - if (row_info->bit_depth == 1) - table = onebppswaptable; - - else if (row_info->bit_depth == 2) - table = twobppswaptable; - - else if (row_info->bit_depth == 4) - table = fourbppswaptable; - - else - return; - - for (rp = row; rp < end; rp++) - *rp = table[*rp]; - } -} -#endif /* PACKSWAP || WRITE_PACKSWAP */ - -#if defined(PNG_WRITE_FILLER_SUPPORTED) || \ - defined(PNG_READ_STRIP_ALPHA_SUPPORTED) -/* Remove a channel - this used to be 'png_do_strip_filler' but it used a - * somewhat weird combination of flags to determine what to do. All the calls - * to png_do_strip_filler are changed in 1.5.2 to call this instead with the - * correct arguments. - * - * The routine isn't general - the channel must be the channel at the start or - * end (not in the middle) of each pixel. - */ -void /* PRIVATE */ -png_do_strip_channel(png_row_infop row_info, png_bytep row, int at_start) -{ - png_bytep sp = row; /* source pointer */ - png_bytep dp = row; /* destination pointer */ - png_bytep ep = row + row_info->rowbytes; /* One beyond end of row */ - - png_debug(1, "in png_do_strip_channel"); - - /* At the start sp will point to the first byte to copy and dp to where - * it is copied to. ep always points just beyond the end of the row, so - * the loop simply copies (channels-1) channels until sp reaches ep. - * - * at_start: 0 -- convert AG, XG, ARGB, XRGB, AAGG, XXGG, etc. - * nonzero -- convert GA, GX, RGBA, RGBX, GGAA, RRGGBBXX, etc. - */ - - /* GA, GX, XG cases */ - if (row_info->channels == 2) - { - if (row_info->bit_depth == 8) - { - if (at_start != 0) /* Skip initial filler */ - ++sp; - else /* Skip initial channel and, for sp, the filler */ - { - sp += 2; ++dp; - } - - /* For a 1 pixel wide image there is nothing to do */ - while (sp < ep) - { - *dp++ = *sp; sp += 2; - } - - row_info->pixel_depth = 8; - } - - else if (row_info->bit_depth == 16) - { - if (at_start != 0) /* Skip initial filler */ - sp += 2; - else /* Skip initial channel and, for sp, the filler */ - { - sp += 4; dp += 2; - } - - while (sp < ep) - { - *dp++ = *sp++; *dp++ = *sp; sp += 3; - } - - row_info->pixel_depth = 16; - } - - else - return; /* bad bit depth */ - - row_info->channels = 1; - - /* Finally fix the color type if it records an alpha channel */ - if (row_info->color_type == PNG_COLOR_TYPE_GRAY_ALPHA) - row_info->color_type = PNG_COLOR_TYPE_GRAY; - } - - /* RGBA, RGBX, XRGB cases */ - else if (row_info->channels == 4) - { - if (row_info->bit_depth == 8) - { - if (at_start != 0) /* Skip initial filler */ - ++sp; - else /* Skip initial channels and, for sp, the filler */ - { - sp += 4; dp += 3; - } - - /* Note that the loop adds 3 to dp and 4 to sp each time. */ - while (sp < ep) - { - *dp++ = *sp++; *dp++ = *sp++; *dp++ = *sp; sp += 2; - } - - row_info->pixel_depth = 24; - } - - else if (row_info->bit_depth == 16) - { - if (at_start != 0) /* Skip initial filler */ - sp += 2; - else /* Skip initial channels and, for sp, the filler */ - { - sp += 8; dp += 6; - } - - while (sp < ep) - { - /* Copy 6 bytes, skip 2 */ - *dp++ = *sp++; *dp++ = *sp++; - *dp++ = *sp++; *dp++ = *sp++; - *dp++ = *sp++; *dp++ = *sp; sp += 3; - } - - row_info->pixel_depth = 48; - } - - else - return; /* bad bit depth */ - - row_info->channels = 3; - - /* Finally fix the color type if it records an alpha channel */ - if (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - row_info->color_type = PNG_COLOR_TYPE_RGB; - } - - else - return; /* The filler channel has gone already */ - - /* Fix the rowbytes value. */ - row_info->rowbytes = (size_t)(dp-row); -} -#endif - -#if defined(PNG_READ_BGR_SUPPORTED) || defined(PNG_WRITE_BGR_SUPPORTED) -/* Swaps red and blue bytes within a pixel */ -void /* PRIVATE */ -png_do_bgr(png_row_infop row_info, png_bytep row) -{ - png_debug(1, "in png_do_bgr"); - - if ((row_info->color_type & PNG_COLOR_MASK_COLOR) != 0) - { - png_uint_32 row_width = row_info->width; - if (row_info->bit_depth == 8) - { - if (row_info->color_type == PNG_COLOR_TYPE_RGB) - { - png_bytep rp; - png_uint_32 i; - - for (i = 0, rp = row; i < row_width; i++, rp += 3) - { - png_byte save = *rp; - *rp = *(rp + 2); - *(rp + 2) = save; - } - } - - else if (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - { - png_bytep rp; - png_uint_32 i; - - for (i = 0, rp = row; i < row_width; i++, rp += 4) - { - png_byte save = *rp; - *rp = *(rp + 2); - *(rp + 2) = save; - } - } - } - -#ifdef PNG_16BIT_SUPPORTED - else if (row_info->bit_depth == 16) - { - if (row_info->color_type == PNG_COLOR_TYPE_RGB) - { - png_bytep rp; - png_uint_32 i; - - for (i = 0, rp = row; i < row_width; i++, rp += 6) - { - png_byte save = *rp; - *rp = *(rp + 4); - *(rp + 4) = save; - save = *(rp + 1); - *(rp + 1) = *(rp + 5); - *(rp + 5) = save; - } - } - - else if (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - { - png_bytep rp; - png_uint_32 i; - - for (i = 0, rp = row; i < row_width; i++, rp += 8) - { - png_byte save = *rp; - *rp = *(rp + 4); - *(rp + 4) = save; - save = *(rp + 1); - *(rp + 1) = *(rp + 5); - *(rp + 5) = save; - } - } - } -#endif - } -} -#endif /* READ_BGR || WRITE_BGR */ - -#if defined(PNG_READ_CHECK_FOR_INVALID_INDEX_SUPPORTED) || \ - defined(PNG_WRITE_CHECK_FOR_INVALID_INDEX_SUPPORTED) -/* Added at libpng-1.5.10 */ -void /* PRIVATE */ -png_do_check_palette_indexes(png_structrp png_ptr, png_row_infop row_info) -{ - png_debug(1, "in png_do_check_palette_indexes"); - - if (png_ptr->num_palette < (1 << row_info->bit_depth) && - png_ptr->num_palette > 0) /* num_palette can be 0 in MNG files */ - { - /* Calculations moved outside switch in an attempt to stop different - * compiler warnings. 'padding' is in *bits* within the last byte, it is - * an 'int' because pixel_depth becomes an 'int' in the expression below, - * and this calculation is used because it avoids warnings that other - * forms produced on either GCC or MSVC. - */ - int padding = PNG_PADBITS(row_info->pixel_depth, row_info->width); - png_bytep rp = png_ptr->row_buf + row_info->rowbytes; - - switch (row_info->bit_depth) - { - case 1: - { - /* in this case, all bytes must be 0 so we don't need - * to unpack the pixels except for the rightmost one. - */ - for (; rp > png_ptr->row_buf; rp--) - { - if ((*rp >> padding) != 0) - png_ptr->num_palette_max = 1; - padding = 0; - } - - break; - } - - case 2: - { - for (; rp > png_ptr->row_buf; rp--) - { - int i = ((*rp >> padding) & 0x03); - - if (i > png_ptr->num_palette_max) - png_ptr->num_palette_max = i; - - i = (((*rp >> padding) >> 2) & 0x03); - - if (i > png_ptr->num_palette_max) - png_ptr->num_palette_max = i; - - i = (((*rp >> padding) >> 4) & 0x03); - - if (i > png_ptr->num_palette_max) - png_ptr->num_palette_max = i; - - i = (((*rp >> padding) >> 6) & 0x03); - - if (i > png_ptr->num_palette_max) - png_ptr->num_palette_max = i; - - padding = 0; - } - - break; - } - - case 4: - { - for (; rp > png_ptr->row_buf; rp--) - { - int i = ((*rp >> padding) & 0x0f); - - if (i > png_ptr->num_palette_max) - png_ptr->num_palette_max = i; - - i = (((*rp >> padding) >> 4) & 0x0f); - - if (i > png_ptr->num_palette_max) - png_ptr->num_palette_max = i; - - padding = 0; - } - - break; - } - - case 8: - { - for (; rp > png_ptr->row_buf; rp--) - { - if (*rp > png_ptr->num_palette_max) - png_ptr->num_palette_max = (int) *rp; - } - - break; - } - - default: - break; - } - } -} -#endif /* CHECK_FOR_INVALID_INDEX */ - -#if defined(PNG_READ_USER_TRANSFORM_SUPPORTED) || \ - defined(PNG_WRITE_USER_TRANSFORM_SUPPORTED) -#ifdef PNG_USER_TRANSFORM_PTR_SUPPORTED -void PNGAPI -png_set_user_transform_info(png_structrp png_ptr, png_voidp - user_transform_ptr, int user_transform_depth, int user_transform_channels) -{ - png_debug(1, "in png_set_user_transform_info"); - - if (png_ptr == NULL) - return; - -#ifdef PNG_READ_USER_TRANSFORM_SUPPORTED - if ((png_ptr->mode & PNG_IS_READ_STRUCT) != 0 && - (png_ptr->flags & PNG_FLAG_ROW_INIT) != 0) - { - png_app_error(png_ptr, - "info change after png_start_read_image or png_read_update_info"); - return; - } -#endif - - png_ptr->user_transform_ptr = user_transform_ptr; - png_ptr->user_transform_depth = (png_byte)user_transform_depth; - png_ptr->user_transform_channels = (png_byte)user_transform_channels; -} -#endif - -/* This function returns a pointer to the user_transform_ptr associated with - * the user transform functions. The application should free any memory - * associated with this pointer before png_write_destroy and png_read_destroy - * are called. - */ -#ifdef PNG_USER_TRANSFORM_PTR_SUPPORTED -png_voidp PNGAPI -png_get_user_transform_ptr(png_const_structrp png_ptr) -{ - if (png_ptr == NULL) - return NULL; - - return png_ptr->user_transform_ptr; -} -#endif - -#ifdef PNG_USER_TRANSFORM_INFO_SUPPORTED -png_uint_32 PNGAPI -png_get_current_row_number(png_const_structrp png_ptr) -{ - /* See the comments in png.h - this is the sub-image row when reading an - * interlaced image. - */ - if (png_ptr != NULL) - return png_ptr->row_number; - - return PNG_UINT_32_MAX; /* help the app not to fail silently */ -} - -png_byte PNGAPI -png_get_current_pass_number(png_const_structrp png_ptr) -{ - if (png_ptr != NULL) - return png_ptr->pass; - return 8; /* invalid */ -} -#endif /* USER_TRANSFORM_INFO */ -#endif /* READ_USER_TRANSFORM || WRITE_USER_TRANSFORM */ -#endif /* READ || WRITE */ diff --git a/dep/libpng/src/pngwio.c b/dep/libpng/src/pngwio.c deleted file mode 100644 index 10e919dd0..000000000 --- a/dep/libpng/src/pngwio.c +++ /dev/null @@ -1,168 +0,0 @@ - -/* pngwio.c - functions for data output - * - * Copyright (c) 2018 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2014,2016,2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - * - * This file provides a location for all output. Users who need - * special handling are expected to write functions that have the same - * arguments as these and perform similar functions, but that possibly - * use different output methods. Note that you shouldn't change these - * functions, but rather write replacement functions and then change - * them at run time with png_set_write_fn(...). - */ - -#include "pngpriv.h" - -#ifdef PNG_WRITE_SUPPORTED - -/* Write the data to whatever output you are using. The default routine - * writes to a file pointer. Note that this routine sometimes gets called - * with very small lengths, so you should implement some kind of simple - * buffering if you are using unbuffered writes. This should never be asked - * to write more than 64K on a 16-bit machine. - */ - -void /* PRIVATE */ -png_write_data(png_structrp png_ptr, png_const_bytep data, size_t length) -{ - /* NOTE: write_data_fn must not change the buffer! */ - if (png_ptr->write_data_fn != NULL ) - (*(png_ptr->write_data_fn))(png_ptr, png_constcast(png_bytep,data), - length); - - else - png_error(png_ptr, "Call to NULL write function"); -} - -#ifdef PNG_STDIO_SUPPORTED -/* This is the function that does the actual writing of data. If you are - * not writing to a standard C stream, you should create a replacement - * write_data function and use it at run time with png_set_write_fn(), rather - * than changing the library. - */ -void PNGCBAPI -png_default_write_data(png_structp png_ptr, png_bytep data, size_t length) -{ - size_t check; - - if (png_ptr == NULL) - return; - - check = fwrite(data, 1, length, (png_FILE_p)(png_ptr->io_ptr)); - - if (check != length) - png_error(png_ptr, "Write Error"); -} -#endif - -/* This function is called to output any data pending writing (normally - * to disk). After png_flush is called, there should be no data pending - * writing in any buffers. - */ -#ifdef PNG_WRITE_FLUSH_SUPPORTED -void /* PRIVATE */ -png_flush(png_structrp png_ptr) -{ - if (png_ptr->output_flush_fn != NULL) - (*(png_ptr->output_flush_fn))(png_ptr); -} - -# ifdef PNG_STDIO_SUPPORTED -void PNGCBAPI -png_default_flush(png_structp png_ptr) -{ - png_FILE_p io_ptr; - - if (png_ptr == NULL) - return; - - io_ptr = png_voidcast(png_FILE_p, (png_ptr->io_ptr)); - fflush(io_ptr); -} -# endif -#endif - -/* This function allows the application to supply new output functions for - * libpng if standard C streams aren't being used. - * - * This function takes as its arguments: - * png_ptr - pointer to a png output data structure - * io_ptr - pointer to user supplied structure containing info about - * the output functions. May be NULL. - * write_data_fn - pointer to a new output function that takes as its - * arguments a pointer to a png_struct, a pointer to - * data to be written, and a 32-bit unsigned int that is - * the number of bytes to be written. The new write - * function should call png_error(png_ptr, "Error msg") - * to exit and output any fatal error messages. May be - * NULL, in which case libpng's default function will - * be used. - * flush_data_fn - pointer to a new flush function that takes as its - * arguments a pointer to a png_struct. After a call to - * the flush function, there should be no data in any buffers - * or pending transmission. If the output method doesn't do - * any buffering of output, a function prototype must still be - * supplied although it doesn't have to do anything. If - * PNG_WRITE_FLUSH_SUPPORTED is not defined at libpng compile - * time, output_flush_fn will be ignored, although it must be - * supplied for compatibility. May be NULL, in which case - * libpng's default function will be used, if - * PNG_WRITE_FLUSH_SUPPORTED is defined. This is not - * a good idea if io_ptr does not point to a standard - * *FILE structure. - */ -void PNGAPI -png_set_write_fn(png_structrp png_ptr, png_voidp io_ptr, - png_rw_ptr write_data_fn, png_flush_ptr output_flush_fn) -{ - if (png_ptr == NULL) - return; - - png_ptr->io_ptr = io_ptr; - -#ifdef PNG_STDIO_SUPPORTED - if (write_data_fn != NULL) - png_ptr->write_data_fn = write_data_fn; - - else - png_ptr->write_data_fn = png_default_write_data; -#else - png_ptr->write_data_fn = write_data_fn; -#endif - -#ifdef PNG_WRITE_FLUSH_SUPPORTED -# ifdef PNG_STDIO_SUPPORTED - - if (output_flush_fn != NULL) - png_ptr->output_flush_fn = output_flush_fn; - - else - png_ptr->output_flush_fn = png_default_flush; - -# else - png_ptr->output_flush_fn = output_flush_fn; -# endif -#else - PNG_UNUSED(output_flush_fn) -#endif /* WRITE_FLUSH */ - -#ifdef PNG_READ_SUPPORTED - /* It is an error to read while writing a png file */ - if (png_ptr->read_data_fn != NULL) - { - png_ptr->read_data_fn = NULL; - - png_warning(png_ptr, - "Can't set both read_data_fn and write_data_fn in the" - " same structure"); - } -#endif -} -#endif /* WRITE */ diff --git a/dep/libpng/src/pngwrite.c b/dep/libpng/src/pngwrite.c deleted file mode 100644 index 77e412f43..000000000 --- a/dep/libpng/src/pngwrite.c +++ /dev/null @@ -1,2418 +0,0 @@ - -/* pngwrite.c - general routines to write a PNG file - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -#include "pngpriv.h" -#ifdef PNG_SIMPLIFIED_WRITE_STDIO_SUPPORTED -# include -#endif /* SIMPLIFIED_WRITE_STDIO */ - -#ifdef PNG_WRITE_SUPPORTED - -#ifdef PNG_WRITE_UNKNOWN_CHUNKS_SUPPORTED -/* Write out all the unknown chunks for the current given location */ -static void -write_unknown_chunks(png_structrp png_ptr, png_const_inforp info_ptr, - unsigned int where) -{ - if (info_ptr->unknown_chunks_num != 0) - { - png_const_unknown_chunkp up; - - png_debug(5, "writing extra chunks"); - - for (up = info_ptr->unknown_chunks; - up < info_ptr->unknown_chunks + info_ptr->unknown_chunks_num; - ++up) - if ((up->location & where) != 0) - { - /* If per-chunk unknown chunk handling is enabled use it, otherwise - * just write the chunks the application has set. - */ -#ifdef PNG_SET_UNKNOWN_CHUNKS_SUPPORTED - int keep = png_handle_as_unknown(png_ptr, up->name); - - /* NOTE: this code is radically different from the read side in the - * matter of handling an ancillary unknown chunk. In the read side - * the default behavior is to discard it, in the code below the default - * behavior is to write it. Critical chunks are, however, only - * written if explicitly listed or if the default is set to write all - * unknown chunks. - * - * The default handling is also slightly weird - it is not possible to - * stop the writing of all unsafe-to-copy chunks! - * - * TODO: REVIEW: this would seem to be a bug. - */ - if (keep != PNG_HANDLE_CHUNK_NEVER && - ((up->name[3] & 0x20) /* safe-to-copy overrides everything */ || - keep == PNG_HANDLE_CHUNK_ALWAYS || - (keep == PNG_HANDLE_CHUNK_AS_DEFAULT && - png_ptr->unknown_default == PNG_HANDLE_CHUNK_ALWAYS))) -#endif - { - /* TODO: review, what is wrong with a zero length unknown chunk? */ - if (up->size == 0) - png_warning(png_ptr, "Writing zero-length unknown chunk"); - - png_write_chunk(png_ptr, up->name, up->data, up->size); - } - } - } -} -#endif /* WRITE_UNKNOWN_CHUNKS */ - -/* Writes all the PNG information. This is the suggested way to use the - * library. If you have a new chunk to add, make a function to write it, - * and put it in the correct location here. If you want the chunk written - * after the image data, put it in png_write_end(). I strongly encourage - * you to supply a PNG_INFO_ flag, and check info_ptr->valid before - * writing the chunk, as that will keep the code from breaking if you want - * to just write a plain PNG file. If you have long comments, I suggest - * writing them in png_write_end(), and compressing them. - */ -void PNGAPI -png_write_info_before_PLTE(png_structrp png_ptr, png_const_inforp info_ptr) -{ - png_debug(1, "in png_write_info_before_PLTE"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - if ((png_ptr->mode & PNG_WROTE_INFO_BEFORE_PLTE) == 0) - { - /* Write PNG signature */ - png_write_sig(png_ptr); - -#ifdef PNG_MNG_FEATURES_SUPPORTED - if ((png_ptr->mode & PNG_HAVE_PNG_SIGNATURE) != 0 && \ - png_ptr->mng_features_permitted != 0) - { - png_warning(png_ptr, - "MNG features are not allowed in a PNG datastream"); - png_ptr->mng_features_permitted = 0; - } -#endif - - /* Write IHDR information. */ - png_write_IHDR(png_ptr, info_ptr->width, info_ptr->height, - info_ptr->bit_depth, info_ptr->color_type, info_ptr->compression_type, - info_ptr->filter_type, -#ifdef PNG_WRITE_INTERLACING_SUPPORTED - info_ptr->interlace_type -#else - 0 -#endif - ); - - /* The rest of these check to see if the valid field has the appropriate - * flag set, and if it does, writes the chunk. - * - * 1.6.0: COLORSPACE support controls the writing of these chunks too, and - * the chunks will be written if the WRITE routine is there and - * information * is available in the COLORSPACE. (See - * png_colorspace_sync_info in png.c for where the valid flags get set.) - * - * Under certain circumstances the colorspace can be invalidated without - * syncing the info_struct 'valid' flags; this happens if libpng detects - * an error and calls png_error while the color space is being set, yet - * the application continues writing the PNG. So check the 'invalid' - * flag here too. - */ -#ifdef PNG_GAMMA_SUPPORTED -# ifdef PNG_WRITE_gAMA_SUPPORTED - if ((info_ptr->colorspace.flags & PNG_COLORSPACE_INVALID) == 0 && - (info_ptr->colorspace.flags & PNG_COLORSPACE_FROM_gAMA) != 0 && - (info_ptr->valid & PNG_INFO_gAMA) != 0) - png_write_gAMA_fixed(png_ptr, info_ptr->colorspace.gamma); -# endif -#endif - -#ifdef PNG_COLORSPACE_SUPPORTED - /* Write only one of sRGB or an ICC profile. If a profile was supplied - * and it matches one of the known sRGB ones issue a warning. - */ -# ifdef PNG_WRITE_iCCP_SUPPORTED - if ((info_ptr->colorspace.flags & PNG_COLORSPACE_INVALID) == 0 && - (info_ptr->valid & PNG_INFO_iCCP) != 0) - { -# ifdef PNG_WRITE_sRGB_SUPPORTED - if ((info_ptr->valid & PNG_INFO_sRGB) != 0) - png_app_warning(png_ptr, - "profile matches sRGB but writing iCCP instead"); -# endif - - png_write_iCCP(png_ptr, info_ptr->iccp_name, - info_ptr->iccp_profile); - } -# ifdef PNG_WRITE_sRGB_SUPPORTED - else -# endif -# endif - -# ifdef PNG_WRITE_sRGB_SUPPORTED - if ((info_ptr->colorspace.flags & PNG_COLORSPACE_INVALID) == 0 && - (info_ptr->valid & PNG_INFO_sRGB) != 0) - png_write_sRGB(png_ptr, info_ptr->colorspace.rendering_intent); -# endif /* WRITE_sRGB */ -#endif /* COLORSPACE */ - -#ifdef PNG_WRITE_sBIT_SUPPORTED - if ((info_ptr->valid & PNG_INFO_sBIT) != 0) - png_write_sBIT(png_ptr, &(info_ptr->sig_bit), info_ptr->color_type); -#endif - -#ifdef PNG_COLORSPACE_SUPPORTED -# ifdef PNG_WRITE_cHRM_SUPPORTED - if ((info_ptr->colorspace.flags & PNG_COLORSPACE_INVALID) == 0 && - (info_ptr->colorspace.flags & PNG_COLORSPACE_FROM_cHRM) != 0 && - (info_ptr->valid & PNG_INFO_cHRM) != 0) - png_write_cHRM_fixed(png_ptr, &info_ptr->colorspace.end_points_xy); -# endif -#endif - -#ifdef PNG_WRITE_UNKNOWN_CHUNKS_SUPPORTED - write_unknown_chunks(png_ptr, info_ptr, PNG_HAVE_IHDR); -#endif - - png_ptr->mode |= PNG_WROTE_INFO_BEFORE_PLTE; - } -} - -void PNGAPI -png_write_info(png_structrp png_ptr, png_const_inforp info_ptr) -{ -#if defined(PNG_WRITE_TEXT_SUPPORTED) || defined(PNG_WRITE_sPLT_SUPPORTED) - int i; -#endif - - png_debug(1, "in png_write_info"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - png_write_info_before_PLTE(png_ptr, info_ptr); - - if ((info_ptr->valid & PNG_INFO_PLTE) != 0) - png_write_PLTE(png_ptr, info_ptr->palette, - (png_uint_32)info_ptr->num_palette); - - else if (info_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - png_error(png_ptr, "Valid palette required for paletted images"); - -#ifdef PNG_WRITE_tRNS_SUPPORTED - if ((info_ptr->valid & PNG_INFO_tRNS) !=0) - { -#ifdef PNG_WRITE_INVERT_ALPHA_SUPPORTED - /* Invert the alpha channel (in tRNS) */ - if ((png_ptr->transformations & PNG_INVERT_ALPHA) != 0 && - info_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - { - int j, jend; - - jend = info_ptr->num_trans; - if (jend > PNG_MAX_PALETTE_LENGTH) - jend = PNG_MAX_PALETTE_LENGTH; - - for (j = 0; jtrans_alpha[j] = - (png_byte)(255 - info_ptr->trans_alpha[j]); - } -#endif - png_write_tRNS(png_ptr, info_ptr->trans_alpha, &(info_ptr->trans_color), - info_ptr->num_trans, info_ptr->color_type); - } -#endif -#ifdef PNG_WRITE_bKGD_SUPPORTED - if ((info_ptr->valid & PNG_INFO_bKGD) != 0) - png_write_bKGD(png_ptr, &(info_ptr->background), info_ptr->color_type); -#endif - -#ifdef PNG_WRITE_eXIf_SUPPORTED - if ((info_ptr->valid & PNG_INFO_eXIf) != 0) - { - png_write_eXIf(png_ptr, info_ptr->exif, info_ptr->num_exif); - png_ptr->mode |= PNG_WROTE_eXIf; - } -#endif - -#ifdef PNG_WRITE_hIST_SUPPORTED - if ((info_ptr->valid & PNG_INFO_hIST) != 0) - png_write_hIST(png_ptr, info_ptr->hist, info_ptr->num_palette); -#endif - -#ifdef PNG_WRITE_oFFs_SUPPORTED - if ((info_ptr->valid & PNG_INFO_oFFs) != 0) - png_write_oFFs(png_ptr, info_ptr->x_offset, info_ptr->y_offset, - info_ptr->offset_unit_type); -#endif - -#ifdef PNG_WRITE_pCAL_SUPPORTED - if ((info_ptr->valid & PNG_INFO_pCAL) != 0) - png_write_pCAL(png_ptr, info_ptr->pcal_purpose, info_ptr->pcal_X0, - info_ptr->pcal_X1, info_ptr->pcal_type, info_ptr->pcal_nparams, - info_ptr->pcal_units, info_ptr->pcal_params); -#endif - -#ifdef PNG_WRITE_sCAL_SUPPORTED - if ((info_ptr->valid & PNG_INFO_sCAL) != 0) - png_write_sCAL_s(png_ptr, (int)info_ptr->scal_unit, - info_ptr->scal_s_width, info_ptr->scal_s_height); -#endif /* sCAL */ - -#ifdef PNG_WRITE_pHYs_SUPPORTED - if ((info_ptr->valid & PNG_INFO_pHYs) != 0) - png_write_pHYs(png_ptr, info_ptr->x_pixels_per_unit, - info_ptr->y_pixels_per_unit, info_ptr->phys_unit_type); -#endif /* pHYs */ - -#ifdef PNG_WRITE_tIME_SUPPORTED - if ((info_ptr->valid & PNG_INFO_tIME) != 0) - { - png_write_tIME(png_ptr, &(info_ptr->mod_time)); - png_ptr->mode |= PNG_WROTE_tIME; - } -#endif /* tIME */ - -#ifdef PNG_WRITE_sPLT_SUPPORTED - if ((info_ptr->valid & PNG_INFO_sPLT) != 0) - for (i = 0; i < (int)info_ptr->splt_palettes_num; i++) - png_write_sPLT(png_ptr, info_ptr->splt_palettes + i); -#endif /* sPLT */ - -#ifdef PNG_WRITE_TEXT_SUPPORTED - /* Check to see if we need to write text chunks */ - for (i = 0; i < info_ptr->num_text; i++) - { - png_debug2(2, "Writing header text chunk %d, type %d", i, - info_ptr->text[i].compression); - /* An internationalized chunk? */ - if (info_ptr->text[i].compression > 0) - { -#ifdef PNG_WRITE_iTXt_SUPPORTED - /* Write international chunk */ - png_write_iTXt(png_ptr, - info_ptr->text[i].compression, - info_ptr->text[i].key, - info_ptr->text[i].lang, - info_ptr->text[i].lang_key, - info_ptr->text[i].text); - /* Mark this chunk as written */ - if (info_ptr->text[i].compression == PNG_TEXT_COMPRESSION_NONE) - info_ptr->text[i].compression = PNG_TEXT_COMPRESSION_NONE_WR; - else - info_ptr->text[i].compression = PNG_TEXT_COMPRESSION_zTXt_WR; -#else - png_warning(png_ptr, "Unable to write international text"); -#endif - } - - /* If we want a compressed text chunk */ - else if (info_ptr->text[i].compression == PNG_TEXT_COMPRESSION_zTXt) - { -#ifdef PNG_WRITE_zTXt_SUPPORTED - /* Write compressed chunk */ - png_write_zTXt(png_ptr, info_ptr->text[i].key, - info_ptr->text[i].text, info_ptr->text[i].compression); - /* Mark this chunk as written */ - info_ptr->text[i].compression = PNG_TEXT_COMPRESSION_zTXt_WR; -#else - png_warning(png_ptr, "Unable to write compressed text"); -#endif - } - - else if (info_ptr->text[i].compression == PNG_TEXT_COMPRESSION_NONE) - { -#ifdef PNG_WRITE_tEXt_SUPPORTED - /* Write uncompressed chunk */ - png_write_tEXt(png_ptr, info_ptr->text[i].key, - info_ptr->text[i].text, - 0); - /* Mark this chunk as written */ - info_ptr->text[i].compression = PNG_TEXT_COMPRESSION_NONE_WR; -#else - /* Can't get here */ - png_warning(png_ptr, "Unable to write uncompressed text"); -#endif - } - } -#endif /* tEXt */ - -#ifdef PNG_WRITE_UNKNOWN_CHUNKS_SUPPORTED - write_unknown_chunks(png_ptr, info_ptr, PNG_HAVE_PLTE); -#endif -} - -/* Writes the end of the PNG file. If you don't want to write comments or - * time information, you can pass NULL for info. If you already wrote these - * in png_write_info(), do not write them again here. If you have long - * comments, I suggest writing them here, and compressing them. - */ -void PNGAPI -png_write_end(png_structrp png_ptr, png_inforp info_ptr) -{ - png_debug(1, "in png_write_end"); - - if (png_ptr == NULL) - return; - - if ((png_ptr->mode & PNG_HAVE_IDAT) == 0) - png_error(png_ptr, "No IDATs written into file"); - -#ifdef PNG_WRITE_CHECK_FOR_INVALID_INDEX_SUPPORTED - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE && - png_ptr->num_palette_max >= png_ptr->num_palette) - png_benign_error(png_ptr, "Wrote palette index exceeding num_palette"); -#endif - - /* See if user wants us to write information chunks */ - if (info_ptr != NULL) - { -#ifdef PNG_WRITE_TEXT_SUPPORTED - int i; /* local index variable */ -#endif -#ifdef PNG_WRITE_tIME_SUPPORTED - /* Check to see if user has supplied a time chunk */ - if ((info_ptr->valid & PNG_INFO_tIME) != 0 && - (png_ptr->mode & PNG_WROTE_tIME) == 0) - png_write_tIME(png_ptr, &(info_ptr->mod_time)); - -#endif -#ifdef PNG_WRITE_TEXT_SUPPORTED - /* Loop through comment chunks */ - for (i = 0; i < info_ptr->num_text; i++) - { - png_debug2(2, "Writing trailer text chunk %d, type %d", i, - info_ptr->text[i].compression); - /* An internationalized chunk? */ - if (info_ptr->text[i].compression > 0) - { -#ifdef PNG_WRITE_iTXt_SUPPORTED - /* Write international chunk */ - png_write_iTXt(png_ptr, - info_ptr->text[i].compression, - info_ptr->text[i].key, - info_ptr->text[i].lang, - info_ptr->text[i].lang_key, - info_ptr->text[i].text); - /* Mark this chunk as written */ - if (info_ptr->text[i].compression == PNG_TEXT_COMPRESSION_NONE) - info_ptr->text[i].compression = PNG_TEXT_COMPRESSION_NONE_WR; - else - info_ptr->text[i].compression = PNG_TEXT_COMPRESSION_zTXt_WR; -#else - png_warning(png_ptr, "Unable to write international text"); -#endif - } - - else if (info_ptr->text[i].compression >= PNG_TEXT_COMPRESSION_zTXt) - { -#ifdef PNG_WRITE_zTXt_SUPPORTED - /* Write compressed chunk */ - png_write_zTXt(png_ptr, info_ptr->text[i].key, - info_ptr->text[i].text, info_ptr->text[i].compression); - /* Mark this chunk as written */ - info_ptr->text[i].compression = PNG_TEXT_COMPRESSION_zTXt_WR; -#else - png_warning(png_ptr, "Unable to write compressed text"); -#endif - } - - else if (info_ptr->text[i].compression == PNG_TEXT_COMPRESSION_NONE) - { -#ifdef PNG_WRITE_tEXt_SUPPORTED - /* Write uncompressed chunk */ - png_write_tEXt(png_ptr, info_ptr->text[i].key, - info_ptr->text[i].text, 0); - /* Mark this chunk as written */ - info_ptr->text[i].compression = PNG_TEXT_COMPRESSION_NONE_WR; -#else - png_warning(png_ptr, "Unable to write uncompressed text"); -#endif - } - } -#endif - -#ifdef PNG_WRITE_eXIf_SUPPORTED - if ((info_ptr->valid & PNG_INFO_eXIf) != 0 && - (png_ptr->mode & PNG_WROTE_eXIf) == 0) - png_write_eXIf(png_ptr, info_ptr->exif, info_ptr->num_exif); -#endif - -#ifdef PNG_WRITE_UNKNOWN_CHUNKS_SUPPORTED - write_unknown_chunks(png_ptr, info_ptr, PNG_AFTER_IDAT); -#endif - } - - png_ptr->mode |= PNG_AFTER_IDAT; - - /* Write end of PNG file */ - png_write_IEND(png_ptr); - - /* This flush, added in libpng-1.0.8, removed from libpng-1.0.9beta03, - * and restored again in libpng-1.2.30, may cause some applications that - * do not set png_ptr->output_flush_fn to crash. If your application - * experiences a problem, please try building libpng with - * PNG_WRITE_FLUSH_AFTER_IEND_SUPPORTED defined, and report the event to - * png-mng-implement at lists.sf.net . - */ -#ifdef PNG_WRITE_FLUSH_SUPPORTED -# ifdef PNG_WRITE_FLUSH_AFTER_IEND_SUPPORTED - png_flush(png_ptr); -# endif -#endif -} - -#ifdef PNG_CONVERT_tIME_SUPPORTED -void PNGAPI -png_convert_from_struct_tm(png_timep ptime, const struct tm * ttime) -{ - png_debug(1, "in png_convert_from_struct_tm"); - - ptime->year = (png_uint_16)(1900 + ttime->tm_year); - ptime->month = (png_byte)(ttime->tm_mon + 1); - ptime->day = (png_byte)ttime->tm_mday; - ptime->hour = (png_byte)ttime->tm_hour; - ptime->minute = (png_byte)ttime->tm_min; - ptime->second = (png_byte)ttime->tm_sec; -} - -void PNGAPI -png_convert_from_time_t(png_timep ptime, time_t ttime) -{ - struct tm *tbuf; - - png_debug(1, "in png_convert_from_time_t"); - - tbuf = gmtime(&ttime); - if (tbuf == NULL) - { - /* TODO: add a safe function which takes a png_ptr argument and raises - * a png_error if the ttime argument is invalid and the call to gmtime - * fails as a consequence. - */ - memset(ptime, 0, sizeof(*ptime)); - return; - } - - png_convert_from_struct_tm(ptime, tbuf); -} -#endif - -/* Initialize png_ptr structure, and allocate any memory needed */ -PNG_FUNCTION(png_structp,PNGAPI -png_create_write_struct,(png_const_charp user_png_ver, png_voidp error_ptr, - png_error_ptr error_fn, png_error_ptr warn_fn),PNG_ALLOCATED) -{ -#ifndef PNG_USER_MEM_SUPPORTED - png_structrp png_ptr = png_create_png_struct(user_png_ver, error_ptr, - error_fn, warn_fn, NULL, NULL, NULL); -#else - return png_create_write_struct_2(user_png_ver, error_ptr, error_fn, - warn_fn, NULL, NULL, NULL); -} - -/* Alternate initialize png_ptr structure, and allocate any memory needed */ -PNG_FUNCTION(png_structp,PNGAPI -png_create_write_struct_2,(png_const_charp user_png_ver, png_voidp error_ptr, - png_error_ptr error_fn, png_error_ptr warn_fn, png_voidp mem_ptr, - png_malloc_ptr malloc_fn, png_free_ptr free_fn),PNG_ALLOCATED) -{ - png_structrp png_ptr = png_create_png_struct(user_png_ver, error_ptr, - error_fn, warn_fn, mem_ptr, malloc_fn, free_fn); -#endif /* USER_MEM */ - if (png_ptr != NULL) - { - /* Set the zlib control values to defaults; they can be overridden by the - * application after the struct has been created. - */ - png_ptr->zbuffer_size = PNG_ZBUF_SIZE; - - /* The 'zlib_strategy' setting is irrelevant because png_default_claim in - * pngwutil.c defaults it according to whether or not filters will be - * used, and ignores this setting. - */ - png_ptr->zlib_strategy = PNG_Z_DEFAULT_STRATEGY; - png_ptr->zlib_level = PNG_Z_DEFAULT_COMPRESSION; - png_ptr->zlib_mem_level = 8; - png_ptr->zlib_window_bits = 15; - png_ptr->zlib_method = 8; - -#ifdef PNG_WRITE_COMPRESSED_TEXT_SUPPORTED - png_ptr->zlib_text_strategy = PNG_TEXT_Z_DEFAULT_STRATEGY; - png_ptr->zlib_text_level = PNG_TEXT_Z_DEFAULT_COMPRESSION; - png_ptr->zlib_text_mem_level = 8; - png_ptr->zlib_text_window_bits = 15; - png_ptr->zlib_text_method = 8; -#endif /* WRITE_COMPRESSED_TEXT */ - - /* This is a highly dubious configuration option; by default it is off, - * but it may be appropriate for private builds that are testing - * extensions not conformant to the current specification, or of - * applications that must not fail to write at all costs! - */ -#ifdef PNG_BENIGN_WRITE_ERRORS_SUPPORTED - /* In stable builds only warn if an application error can be completely - * handled. - */ - png_ptr->flags |= PNG_FLAG_BENIGN_ERRORS_WARN; -#endif - - /* App warnings are warnings in release (or release candidate) builds but - * are errors during development. - */ -#if PNG_RELEASE_BUILD - png_ptr->flags |= PNG_FLAG_APP_WARNINGS_WARN; -#endif - - /* TODO: delay this, it can be done in png_init_io() (if the app doesn't - * do it itself) avoiding setting the default function if it is not - * required. - */ - png_set_write_fn(png_ptr, NULL, NULL, NULL); - } - - return png_ptr; -} - - -/* Write a few rows of image data. If the image is interlaced, - * either you will have to write the 7 sub images, or, if you - * have called png_set_interlace_handling(), you will have to - * "write" the image seven times. - */ -void PNGAPI -png_write_rows(png_structrp png_ptr, png_bytepp row, - png_uint_32 num_rows) -{ - png_uint_32 i; /* row counter */ - png_bytepp rp; /* row pointer */ - - png_debug(1, "in png_write_rows"); - - if (png_ptr == NULL) - return; - - /* Loop through the rows */ - for (i = 0, rp = row; i < num_rows; i++, rp++) - { - png_write_row(png_ptr, *rp); - } -} - -/* Write the image. You only need to call this function once, even - * if you are writing an interlaced image. - */ -void PNGAPI -png_write_image(png_structrp png_ptr, png_bytepp image) -{ - png_uint_32 i; /* row index */ - int pass, num_pass; /* pass variables */ - png_bytepp rp; /* points to current row */ - - if (png_ptr == NULL) - return; - - png_debug(1, "in png_write_image"); - -#ifdef PNG_WRITE_INTERLACING_SUPPORTED - /* Initialize interlace handling. If image is not interlaced, - * this will set pass to 1 - */ - num_pass = png_set_interlace_handling(png_ptr); -#else - num_pass = 1; -#endif - /* Loop through passes */ - for (pass = 0; pass < num_pass; pass++) - { - /* Loop through image */ - for (i = 0, rp = image; i < png_ptr->height; i++, rp++) - { - png_write_row(png_ptr, *rp); - } - } -} - -#ifdef PNG_MNG_FEATURES_SUPPORTED -/* Performs intrapixel differencing */ -static void -png_do_write_intrapixel(png_row_infop row_info, png_bytep row) -{ - png_debug(1, "in png_do_write_intrapixel"); - - if ((row_info->color_type & PNG_COLOR_MASK_COLOR) != 0) - { - int bytes_per_pixel; - png_uint_32 row_width = row_info->width; - if (row_info->bit_depth == 8) - { - png_bytep rp; - png_uint_32 i; - - if (row_info->color_type == PNG_COLOR_TYPE_RGB) - bytes_per_pixel = 3; - - else if (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - bytes_per_pixel = 4; - - else - return; - - for (i = 0, rp = row; i < row_width; i++, rp += bytes_per_pixel) - { - *(rp) = (png_byte)(*rp - *(rp + 1)); - *(rp + 2) = (png_byte)(*(rp + 2) - *(rp + 1)); - } - } - -#ifdef PNG_WRITE_16BIT_SUPPORTED - else if (row_info->bit_depth == 16) - { - png_bytep rp; - png_uint_32 i; - - if (row_info->color_type == PNG_COLOR_TYPE_RGB) - bytes_per_pixel = 6; - - else if (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - bytes_per_pixel = 8; - - else - return; - - for (i = 0, rp = row; i < row_width; i++, rp += bytes_per_pixel) - { - png_uint_32 s0 = (png_uint_32)(*(rp ) << 8) | *(rp + 1); - png_uint_32 s1 = (png_uint_32)(*(rp + 2) << 8) | *(rp + 3); - png_uint_32 s2 = (png_uint_32)(*(rp + 4) << 8) | *(rp + 5); - png_uint_32 red = (png_uint_32)((s0 - s1) & 0xffffL); - png_uint_32 blue = (png_uint_32)((s2 - s1) & 0xffffL); - *(rp ) = (png_byte)(red >> 8); - *(rp + 1) = (png_byte)red; - *(rp + 4) = (png_byte)(blue >> 8); - *(rp + 5) = (png_byte)blue; - } - } -#endif /* WRITE_16BIT */ - } -} -#endif /* MNG_FEATURES */ - -/* Called by user to write a row of image data */ -void PNGAPI -png_write_row(png_structrp png_ptr, png_const_bytep row) -{ - /* 1.5.6: moved from png_struct to be a local structure: */ - png_row_info row_info; - - png_debug2(1, "in png_write_row (row %u, pass %d)", - png_ptr->row_number, png_ptr->pass); - - if (png_ptr == NULL) - return; - - /* Initialize transformations and other stuff if first time */ - if (png_ptr->row_number == 0 && png_ptr->pass == 0) - { - /* Make sure we wrote the header info */ - if ((png_ptr->mode & PNG_WROTE_INFO_BEFORE_PLTE) == 0) - png_error(png_ptr, - "png_write_info was never called before png_write_row"); - - /* Check for transforms that have been set but were defined out */ -#if !defined(PNG_WRITE_INVERT_SUPPORTED) && defined(PNG_READ_INVERT_SUPPORTED) - if ((png_ptr->transformations & PNG_INVERT_MONO) != 0) - png_warning(png_ptr, "PNG_WRITE_INVERT_SUPPORTED is not defined"); -#endif - -#if !defined(PNG_WRITE_FILLER_SUPPORTED) && defined(PNG_READ_FILLER_SUPPORTED) - if ((png_ptr->transformations & PNG_FILLER) != 0) - png_warning(png_ptr, "PNG_WRITE_FILLER_SUPPORTED is not defined"); -#endif -#if !defined(PNG_WRITE_PACKSWAP_SUPPORTED) && \ - defined(PNG_READ_PACKSWAP_SUPPORTED) - if ((png_ptr->transformations & PNG_PACKSWAP) != 0) - png_warning(png_ptr, - "PNG_WRITE_PACKSWAP_SUPPORTED is not defined"); -#endif - -#if !defined(PNG_WRITE_PACK_SUPPORTED) && defined(PNG_READ_PACK_SUPPORTED) - if ((png_ptr->transformations & PNG_PACK) != 0) - png_warning(png_ptr, "PNG_WRITE_PACK_SUPPORTED is not defined"); -#endif - -#if !defined(PNG_WRITE_SHIFT_SUPPORTED) && defined(PNG_READ_SHIFT_SUPPORTED) - if ((png_ptr->transformations & PNG_SHIFT) != 0) - png_warning(png_ptr, "PNG_WRITE_SHIFT_SUPPORTED is not defined"); -#endif - -#if !defined(PNG_WRITE_BGR_SUPPORTED) && defined(PNG_READ_BGR_SUPPORTED) - if ((png_ptr->transformations & PNG_BGR) != 0) - png_warning(png_ptr, "PNG_WRITE_BGR_SUPPORTED is not defined"); -#endif - -#if !defined(PNG_WRITE_SWAP_SUPPORTED) && defined(PNG_READ_SWAP_SUPPORTED) - if ((png_ptr->transformations & PNG_SWAP_BYTES) != 0) - png_warning(png_ptr, "PNG_WRITE_SWAP_SUPPORTED is not defined"); -#endif - - png_write_start_row(png_ptr); - } - -#ifdef PNG_WRITE_INTERLACING_SUPPORTED - /* If interlaced and not interested in row, return */ - if (png_ptr->interlaced != 0 && - (png_ptr->transformations & PNG_INTERLACE) != 0) - { - switch (png_ptr->pass) - { - case 0: - if ((png_ptr->row_number & 0x07) != 0) - { - png_write_finish_row(png_ptr); - return; - } - break; - - case 1: - if ((png_ptr->row_number & 0x07) != 0 || png_ptr->width < 5) - { - png_write_finish_row(png_ptr); - return; - } - break; - - case 2: - if ((png_ptr->row_number & 0x07) != 4) - { - png_write_finish_row(png_ptr); - return; - } - break; - - case 3: - if ((png_ptr->row_number & 0x03) != 0 || png_ptr->width < 3) - { - png_write_finish_row(png_ptr); - return; - } - break; - - case 4: - if ((png_ptr->row_number & 0x03) != 2) - { - png_write_finish_row(png_ptr); - return; - } - break; - - case 5: - if ((png_ptr->row_number & 0x01) != 0 || png_ptr->width < 2) - { - png_write_finish_row(png_ptr); - return; - } - break; - - case 6: - if ((png_ptr->row_number & 0x01) == 0) - { - png_write_finish_row(png_ptr); - return; - } - break; - - default: /* error: ignore it */ - break; - } - } -#endif - - /* Set up row info for transformations */ - row_info.color_type = png_ptr->color_type; - row_info.width = png_ptr->usr_width; - row_info.channels = png_ptr->usr_channels; - row_info.bit_depth = png_ptr->usr_bit_depth; - row_info.pixel_depth = (png_byte)(row_info.bit_depth * row_info.channels); - row_info.rowbytes = PNG_ROWBYTES(row_info.pixel_depth, row_info.width); - - png_debug1(3, "row_info->color_type = %d", row_info.color_type); - png_debug1(3, "row_info->width = %u", row_info.width); - png_debug1(3, "row_info->channels = %d", row_info.channels); - png_debug1(3, "row_info->bit_depth = %d", row_info.bit_depth); - png_debug1(3, "row_info->pixel_depth = %d", row_info.pixel_depth); - png_debug1(3, "row_info->rowbytes = %lu", (unsigned long)row_info.rowbytes); - - /* Copy user's row into buffer, leaving room for filter byte. */ - memcpy(png_ptr->row_buf + 1, row, row_info.rowbytes); - -#ifdef PNG_WRITE_INTERLACING_SUPPORTED - /* Handle interlacing */ - if (png_ptr->interlaced && png_ptr->pass < 6 && - (png_ptr->transformations & PNG_INTERLACE) != 0) - { - png_do_write_interlace(&row_info, png_ptr->row_buf + 1, png_ptr->pass); - /* This should always get caught above, but still ... */ - if (row_info.width == 0) - { - png_write_finish_row(png_ptr); - return; - } - } -#endif - -#ifdef PNG_WRITE_TRANSFORMS_SUPPORTED - /* Handle other transformations */ - if (png_ptr->transformations != 0) - png_do_write_transformations(png_ptr, &row_info); -#endif - - /* At this point the row_info pixel depth must match the 'transformed' depth, - * which is also the output depth. - */ - if (row_info.pixel_depth != png_ptr->pixel_depth || - row_info.pixel_depth != png_ptr->transformed_pixel_depth) - png_error(png_ptr, "internal write transform logic error"); - -#ifdef PNG_MNG_FEATURES_SUPPORTED - /* Write filter_method 64 (intrapixel differencing) only if - * 1. Libpng was compiled with PNG_MNG_FEATURES_SUPPORTED and - * 2. Libpng did not write a PNG signature (this filter_method is only - * used in PNG datastreams that are embedded in MNG datastreams) and - * 3. The application called png_permit_mng_features with a mask that - * included PNG_FLAG_MNG_FILTER_64 and - * 4. The filter_method is 64 and - * 5. The color_type is RGB or RGBA - */ - if ((png_ptr->mng_features_permitted & PNG_FLAG_MNG_FILTER_64) != 0 && - (png_ptr->filter_type == PNG_INTRAPIXEL_DIFFERENCING)) - { - /* Intrapixel differencing */ - png_do_write_intrapixel(&row_info, png_ptr->row_buf + 1); - } -#endif - -/* Added at libpng-1.5.10 */ -#ifdef PNG_WRITE_CHECK_FOR_INVALID_INDEX_SUPPORTED - /* Check for out-of-range palette index */ - if (row_info.color_type == PNG_COLOR_TYPE_PALETTE && - png_ptr->num_palette_max >= 0) - png_do_check_palette_indexes(png_ptr, &row_info); -#endif - - /* Find a filter if necessary, filter the row and write it out. */ - png_write_find_filter(png_ptr, &row_info); - - if (png_ptr->write_row_fn != NULL) - (*(png_ptr->write_row_fn))(png_ptr, png_ptr->row_number, png_ptr->pass); -} - -#ifdef PNG_WRITE_FLUSH_SUPPORTED -/* Set the automatic flush interval or 0 to turn flushing off */ -void PNGAPI -png_set_flush(png_structrp png_ptr, int nrows) -{ - png_debug(1, "in png_set_flush"); - - if (png_ptr == NULL) - return; - - png_ptr->flush_dist = (nrows < 0 ? 0 : (png_uint_32)nrows); -} - -/* Flush the current output buffers now */ -void PNGAPI -png_write_flush(png_structrp png_ptr) -{ - png_debug(1, "in png_write_flush"); - - if (png_ptr == NULL) - return; - - /* We have already written out all of the data */ - if (png_ptr->row_number >= png_ptr->num_rows) - return; - - png_compress_IDAT(png_ptr, NULL, 0, Z_SYNC_FLUSH); - png_ptr->flush_rows = 0; - png_flush(png_ptr); -} -#endif /* WRITE_FLUSH */ - -/* Free any memory used in png_ptr struct without freeing the struct itself. */ -static void -png_write_destroy(png_structrp png_ptr) -{ - png_debug(1, "in png_write_destroy"); - - /* Free any memory zlib uses */ - if ((png_ptr->flags & PNG_FLAG_ZSTREAM_INITIALIZED) != 0) - deflateEnd(&png_ptr->zstream); - - /* Free our memory. png_free checks NULL for us. */ - png_free_buffer_list(png_ptr, &png_ptr->zbuffer_list); - png_free(png_ptr, png_ptr->row_buf); - png_ptr->row_buf = NULL; -#ifdef PNG_WRITE_FILTER_SUPPORTED - png_free(png_ptr, png_ptr->prev_row); - png_free(png_ptr, png_ptr->try_row); - png_free(png_ptr, png_ptr->tst_row); - png_ptr->prev_row = NULL; - png_ptr->try_row = NULL; - png_ptr->tst_row = NULL; -#endif - -#ifdef PNG_SET_UNKNOWN_CHUNKS_SUPPORTED - png_free(png_ptr, png_ptr->chunk_list); - png_ptr->chunk_list = NULL; -#endif - - /* The error handling and memory handling information is left intact at this - * point: the jmp_buf may still have to be freed. See png_destroy_png_struct - * for how this happens. - */ -} - -/* Free all memory used by the write. - * In libpng 1.6.0 this API changed quietly to no longer accept a NULL value for - * *png_ptr_ptr. Prior to 1.6.0 it would accept such a value and it would free - * the passed in info_structs but it would quietly fail to free any of the data - * inside them. In 1.6.0 it quietly does nothing (it has to be quiet because it - * has no png_ptr.) - */ -void PNGAPI -png_destroy_write_struct(png_structpp png_ptr_ptr, png_infopp info_ptr_ptr) -{ - png_debug(1, "in png_destroy_write_struct"); - - if (png_ptr_ptr != NULL) - { - png_structrp png_ptr = *png_ptr_ptr; - - if (png_ptr != NULL) /* added in libpng 1.6.0 */ - { - png_destroy_info_struct(png_ptr, info_ptr_ptr); - - *png_ptr_ptr = NULL; - png_write_destroy(png_ptr); - png_destroy_png_struct(png_ptr); - } - } -} - -/* Allow the application to select one or more row filters to use. */ -void PNGAPI -png_set_filter(png_structrp png_ptr, int method, int filters) -{ - png_debug(1, "in png_set_filter"); - - if (png_ptr == NULL) - return; - -#ifdef PNG_MNG_FEATURES_SUPPORTED - if ((png_ptr->mng_features_permitted & PNG_FLAG_MNG_FILTER_64) != 0 && - (method == PNG_INTRAPIXEL_DIFFERENCING)) - method = PNG_FILTER_TYPE_BASE; - -#endif - if (method == PNG_FILTER_TYPE_BASE) - { - switch (filters & (PNG_ALL_FILTERS | 0x07)) - { -#ifdef PNG_WRITE_FILTER_SUPPORTED - case 5: - case 6: - case 7: png_app_error(png_ptr, "Unknown row filter for method 0"); -#endif /* WRITE_FILTER */ - /* FALLTHROUGH */ - case PNG_FILTER_VALUE_NONE: - png_ptr->do_filter = PNG_FILTER_NONE; break; - -#ifdef PNG_WRITE_FILTER_SUPPORTED - case PNG_FILTER_VALUE_SUB: - png_ptr->do_filter = PNG_FILTER_SUB; break; - - case PNG_FILTER_VALUE_UP: - png_ptr->do_filter = PNG_FILTER_UP; break; - - case PNG_FILTER_VALUE_AVG: - png_ptr->do_filter = PNG_FILTER_AVG; break; - - case PNG_FILTER_VALUE_PAETH: - png_ptr->do_filter = PNG_FILTER_PAETH; break; - - default: - png_ptr->do_filter = (png_byte)filters; break; -#else - default: - png_app_error(png_ptr, "Unknown row filter for method 0"); -#endif /* WRITE_FILTER */ - } - -#ifdef PNG_WRITE_FILTER_SUPPORTED - /* If we have allocated the row_buf, this means we have already started - * with the image and we should have allocated all of the filter buffers - * that have been selected. If prev_row isn't already allocated, then - * it is too late to start using the filters that need it, since we - * will be missing the data in the previous row. If an application - * wants to start and stop using particular filters during compression, - * it should start out with all of the filters, and then remove them - * or add them back after the start of compression. - * - * NOTE: this is a nasty constraint on the code, because it means that the - * prev_row buffer must be maintained even if there are currently no - * 'prev_row' requiring filters active. - */ - if (png_ptr->row_buf != NULL) - { - int num_filters; - png_alloc_size_t buf_size; - - /* Repeat the checks in png_write_start_row; 1 pixel high or wide - * images cannot benefit from certain filters. If this isn't done here - * the check below will fire on 1 pixel high images. - */ - if (png_ptr->height == 1) - filters &= ~(PNG_FILTER_UP|PNG_FILTER_AVG|PNG_FILTER_PAETH); - - if (png_ptr->width == 1) - filters &= ~(PNG_FILTER_SUB|PNG_FILTER_AVG|PNG_FILTER_PAETH); - - if ((filters & (PNG_FILTER_UP|PNG_FILTER_AVG|PNG_FILTER_PAETH)) != 0 - && png_ptr->prev_row == NULL) - { - /* This is the error case, however it is benign - the previous row - * is not available so the filter can't be used. Just warn here. - */ - png_app_warning(png_ptr, - "png_set_filter: UP/AVG/PAETH cannot be added after start"); - filters &= ~(PNG_FILTER_UP|PNG_FILTER_AVG|PNG_FILTER_PAETH); - } - - num_filters = 0; - - if (filters & PNG_FILTER_SUB) - num_filters++; - - if (filters & PNG_FILTER_UP) - num_filters++; - - if (filters & PNG_FILTER_AVG) - num_filters++; - - if (filters & PNG_FILTER_PAETH) - num_filters++; - - /* Allocate needed row buffers if they have not already been - * allocated. - */ - buf_size = PNG_ROWBYTES(png_ptr->usr_channels * png_ptr->usr_bit_depth, - png_ptr->width) + 1; - - if (png_ptr->try_row == NULL) - png_ptr->try_row = png_voidcast(png_bytep, - png_malloc(png_ptr, buf_size)); - - if (num_filters > 1) - { - if (png_ptr->tst_row == NULL) - png_ptr->tst_row = png_voidcast(png_bytep, - png_malloc(png_ptr, buf_size)); - } - } - png_ptr->do_filter = (png_byte)filters; -#endif - } - else - png_error(png_ptr, "Unknown custom filter method"); -} - -#ifdef PNG_WRITE_WEIGHTED_FILTER_SUPPORTED /* DEPRECATED */ -/* Provide floating and fixed point APIs */ -#ifdef PNG_FLOATING_POINT_SUPPORTED -void PNGAPI -png_set_filter_heuristics(png_structrp png_ptr, int heuristic_method, - int num_weights, png_const_doublep filter_weights, - png_const_doublep filter_costs) -{ - PNG_UNUSED(png_ptr) - PNG_UNUSED(heuristic_method) - PNG_UNUSED(num_weights) - PNG_UNUSED(filter_weights) - PNG_UNUSED(filter_costs) -} -#endif /* FLOATING_POINT */ - -#ifdef PNG_FIXED_POINT_SUPPORTED -void PNGAPI -png_set_filter_heuristics_fixed(png_structrp png_ptr, int heuristic_method, - int num_weights, png_const_fixed_point_p filter_weights, - png_const_fixed_point_p filter_costs) -{ - PNG_UNUSED(png_ptr) - PNG_UNUSED(heuristic_method) - PNG_UNUSED(num_weights) - PNG_UNUSED(filter_weights) - PNG_UNUSED(filter_costs) -} -#endif /* FIXED_POINT */ -#endif /* WRITE_WEIGHTED_FILTER */ - -#ifdef PNG_WRITE_CUSTOMIZE_COMPRESSION_SUPPORTED -void PNGAPI -png_set_compression_level(png_structrp png_ptr, int level) -{ - png_debug(1, "in png_set_compression_level"); - - if (png_ptr == NULL) - return; - - png_ptr->zlib_level = level; -} - -void PNGAPI -png_set_compression_mem_level(png_structrp png_ptr, int mem_level) -{ - png_debug(1, "in png_set_compression_mem_level"); - - if (png_ptr == NULL) - return; - - png_ptr->zlib_mem_level = mem_level; -} - -void PNGAPI -png_set_compression_strategy(png_structrp png_ptr, int strategy) -{ - png_debug(1, "in png_set_compression_strategy"); - - if (png_ptr == NULL) - return; - - /* The flag setting here prevents the libpng dynamic selection of strategy. - */ - png_ptr->flags |= PNG_FLAG_ZLIB_CUSTOM_STRATEGY; - png_ptr->zlib_strategy = strategy; -} - -/* If PNG_WRITE_OPTIMIZE_CMF_SUPPORTED is defined, libpng will use a - * smaller value of window_bits if it can do so safely. - */ -void PNGAPI -png_set_compression_window_bits(png_structrp png_ptr, int window_bits) -{ - png_debug(1, "in png_set_compression_window_bits"); - - if (png_ptr == NULL) - return; - - /* Prior to 1.6.0 this would warn but then set the window_bits value. This - * meant that negative window bits values could be selected that would cause - * libpng to write a non-standard PNG file with raw deflate or gzip - * compressed IDAT or ancillary chunks. Such files can be read and there is - * no warning on read, so this seems like a very bad idea. - */ - if (window_bits > 15) - { - png_warning(png_ptr, "Only compression windows <= 32k supported by PNG"); - window_bits = 15; - } - - else if (window_bits < 8) - { - png_warning(png_ptr, "Only compression windows >= 256 supported by PNG"); - window_bits = 8; - } - - png_ptr->zlib_window_bits = window_bits; -} - -void PNGAPI -png_set_compression_method(png_structrp png_ptr, int method) -{ - png_debug(1, "in png_set_compression_method"); - - if (png_ptr == NULL) - return; - - /* This would produce an invalid PNG file if it worked, but it doesn't and - * deflate will fault it, so it is harmless to just warn here. - */ - if (method != 8) - png_warning(png_ptr, "Only compression method 8 is supported by PNG"); - - png_ptr->zlib_method = method; -} -#endif /* WRITE_CUSTOMIZE_COMPRESSION */ - -/* The following were added to libpng-1.5.4 */ -#ifdef PNG_WRITE_CUSTOMIZE_ZTXT_COMPRESSION_SUPPORTED -void PNGAPI -png_set_text_compression_level(png_structrp png_ptr, int level) -{ - png_debug(1, "in png_set_text_compression_level"); - - if (png_ptr == NULL) - return; - - png_ptr->zlib_text_level = level; -} - -void PNGAPI -png_set_text_compression_mem_level(png_structrp png_ptr, int mem_level) -{ - png_debug(1, "in png_set_text_compression_mem_level"); - - if (png_ptr == NULL) - return; - - png_ptr->zlib_text_mem_level = mem_level; -} - -void PNGAPI -png_set_text_compression_strategy(png_structrp png_ptr, int strategy) -{ - png_debug(1, "in png_set_text_compression_strategy"); - - if (png_ptr == NULL) - return; - - png_ptr->zlib_text_strategy = strategy; -} - -/* If PNG_WRITE_OPTIMIZE_CMF_SUPPORTED is defined, libpng will use a - * smaller value of window_bits if it can do so safely. - */ -void PNGAPI -png_set_text_compression_window_bits(png_structrp png_ptr, int window_bits) -{ - png_debug(1, "in png_set_text_compression_window_bits"); - - if (png_ptr == NULL) - return; - - if (window_bits > 15) - { - png_warning(png_ptr, "Only compression windows <= 32k supported by PNG"); - window_bits = 15; - } - - else if (window_bits < 8) - { - png_warning(png_ptr, "Only compression windows >= 256 supported by PNG"); - window_bits = 8; - } - - png_ptr->zlib_text_window_bits = window_bits; -} - -void PNGAPI -png_set_text_compression_method(png_structrp png_ptr, int method) -{ - png_debug(1, "in png_set_text_compression_method"); - - if (png_ptr == NULL) - return; - - if (method != 8) - png_warning(png_ptr, "Only compression method 8 is supported by PNG"); - - png_ptr->zlib_text_method = method; -} -#endif /* WRITE_CUSTOMIZE_ZTXT_COMPRESSION */ -/* end of API added to libpng-1.5.4 */ - -void PNGAPI -png_set_write_status_fn(png_structrp png_ptr, png_write_status_ptr write_row_fn) -{ - png_debug(1, "in png_set_write_status_fn"); - - if (png_ptr == NULL) - return; - - png_ptr->write_row_fn = write_row_fn; -} - -#ifdef PNG_WRITE_USER_TRANSFORM_SUPPORTED -void PNGAPI -png_set_write_user_transform_fn(png_structrp png_ptr, png_user_transform_ptr - write_user_transform_fn) -{ - png_debug(1, "in png_set_write_user_transform_fn"); - - if (png_ptr == NULL) - return; - - png_ptr->transformations |= PNG_USER_TRANSFORM; - png_ptr->write_user_transform_fn = write_user_transform_fn; -} -#endif - - -#ifdef PNG_INFO_IMAGE_SUPPORTED -void PNGAPI -png_write_png(png_structrp png_ptr, png_inforp info_ptr, - int transforms, voidp params) -{ - png_debug(1, "in png_write_png"); - - if (png_ptr == NULL || info_ptr == NULL) - return; - - if ((info_ptr->valid & PNG_INFO_IDAT) == 0) - { - png_app_error(png_ptr, "no rows for png_write_image to write"); - return; - } - - /* Write the file header information. */ - png_write_info(png_ptr, info_ptr); - - /* ------ these transformations don't touch the info structure ------- */ - - /* Invert monochrome pixels */ - if ((transforms & PNG_TRANSFORM_INVERT_MONO) != 0) -#ifdef PNG_WRITE_INVERT_SUPPORTED - png_set_invert_mono(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_INVERT_MONO not supported"); -#endif - - /* Shift the pixels up to a legal bit depth and fill in - * as appropriate to correctly scale the image. - */ - if ((transforms & PNG_TRANSFORM_SHIFT) != 0) -#ifdef PNG_WRITE_SHIFT_SUPPORTED - if ((info_ptr->valid & PNG_INFO_sBIT) != 0) - png_set_shift(png_ptr, &info_ptr->sig_bit); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_SHIFT not supported"); -#endif - - /* Pack pixels into bytes */ - if ((transforms & PNG_TRANSFORM_PACKING) != 0) -#ifdef PNG_WRITE_PACK_SUPPORTED - png_set_packing(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_PACKING not supported"); -#endif - - /* Swap location of alpha bytes from ARGB to RGBA */ - if ((transforms & PNG_TRANSFORM_SWAP_ALPHA) != 0) -#ifdef PNG_WRITE_SWAP_ALPHA_SUPPORTED - png_set_swap_alpha(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_SWAP_ALPHA not supported"); -#endif - - /* Remove a filler (X) from XRGB/RGBX/AG/GA into to convert it into - * RGB, note that the code expects the input color type to be G or RGB; no - * alpha channel. - */ - if ((transforms & (PNG_TRANSFORM_STRIP_FILLER_AFTER| - PNG_TRANSFORM_STRIP_FILLER_BEFORE)) != 0) - { -#ifdef PNG_WRITE_FILLER_SUPPORTED - if ((transforms & PNG_TRANSFORM_STRIP_FILLER_AFTER) != 0) - { - if ((transforms & PNG_TRANSFORM_STRIP_FILLER_BEFORE) != 0) - png_app_error(png_ptr, - "PNG_TRANSFORM_STRIP_FILLER: BEFORE+AFTER not supported"); - - /* Continue if ignored - this is the pre-1.6.10 behavior */ - png_set_filler(png_ptr, 0, PNG_FILLER_AFTER); - } - - else if ((transforms & PNG_TRANSFORM_STRIP_FILLER_BEFORE) != 0) - png_set_filler(png_ptr, 0, PNG_FILLER_BEFORE); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_STRIP_FILLER not supported"); -#endif - } - - /* Flip BGR pixels to RGB */ - if ((transforms & PNG_TRANSFORM_BGR) != 0) -#ifdef PNG_WRITE_BGR_SUPPORTED - png_set_bgr(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_BGR not supported"); -#endif - - /* Swap bytes of 16-bit files to most significant byte first */ - if ((transforms & PNG_TRANSFORM_SWAP_ENDIAN) != 0) -#ifdef PNG_WRITE_SWAP_SUPPORTED - png_set_swap(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_SWAP_ENDIAN not supported"); -#endif - - /* Swap bits of 1-bit, 2-bit, 4-bit packed pixel formats */ - if ((transforms & PNG_TRANSFORM_PACKSWAP) != 0) -#ifdef PNG_WRITE_PACKSWAP_SUPPORTED - png_set_packswap(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_PACKSWAP not supported"); -#endif - - /* Invert the alpha channel from opacity to transparency */ - if ((transforms & PNG_TRANSFORM_INVERT_ALPHA) != 0) -#ifdef PNG_WRITE_INVERT_ALPHA_SUPPORTED - png_set_invert_alpha(png_ptr); -#else - png_app_error(png_ptr, "PNG_TRANSFORM_INVERT_ALPHA not supported"); -#endif - - /* ----------------------- end of transformations ------------------- */ - - /* Write the bits */ - png_write_image(png_ptr, info_ptr->row_pointers); - - /* It is REQUIRED to call this to finish writing the rest of the file */ - png_write_end(png_ptr, info_ptr); - - PNG_UNUSED(params) -} -#endif - - -#ifdef PNG_SIMPLIFIED_WRITE_SUPPORTED -/* Initialize the write structure - general purpose utility. */ -static int -png_image_write_init(png_imagep image) -{ - png_structp png_ptr = png_create_write_struct(PNG_LIBPNG_VER_STRING, image, - png_safe_error, png_safe_warning); - - if (png_ptr != NULL) - { - png_infop info_ptr = png_create_info_struct(png_ptr); - - if (info_ptr != NULL) - { - png_controlp control = png_voidcast(png_controlp, - png_malloc_warn(png_ptr, (sizeof *control))); - - if (control != NULL) - { - memset(control, 0, (sizeof *control)); - - control->png_ptr = png_ptr; - control->info_ptr = info_ptr; - control->for_write = 1; - - image->opaque = control; - return 1; - } - - /* Error clean up */ - png_destroy_info_struct(png_ptr, &info_ptr); - } - - png_destroy_write_struct(&png_ptr, NULL); - } - - return png_image_error(image, "png_image_write_: out of memory"); -} - -/* Arguments to png_image_write_main: */ -typedef struct -{ - /* Arguments: */ - png_imagep image; - png_const_voidp buffer; - png_int_32 row_stride; - png_const_voidp colormap; - int convert_to_8bit; - /* Local variables: */ - png_const_voidp first_row; - ptrdiff_t row_bytes; - png_voidp local_row; - /* Byte count for memory writing */ - png_bytep memory; - png_alloc_size_t memory_bytes; /* not used for STDIO */ - png_alloc_size_t output_bytes; /* running total */ -} png_image_write_control; - -/* Write png_uint_16 input to a 16-bit PNG; the png_ptr has already been set to - * do any necessary byte swapping. The component order is defined by the - * png_image format value. - */ -static int -png_write_image_16bit(png_voidp argument) -{ - png_image_write_control *display = png_voidcast(png_image_write_control*, - argument); - png_imagep image = display->image; - png_structrp png_ptr = image->opaque->png_ptr; - - png_const_uint_16p input_row = png_voidcast(png_const_uint_16p, - display->first_row); - png_uint_16p output_row = png_voidcast(png_uint_16p, display->local_row); - png_uint_16p row_end; - unsigned int channels = (image->format & PNG_FORMAT_FLAG_COLOR) != 0 ? - 3 : 1; - int aindex = 0; - png_uint_32 y = image->height; - - if ((image->format & PNG_FORMAT_FLAG_ALPHA) != 0) - { -# ifdef PNG_SIMPLIFIED_WRITE_AFIRST_SUPPORTED - if ((image->format & PNG_FORMAT_FLAG_AFIRST) != 0) - { - aindex = -1; - ++input_row; /* To point to the first component */ - ++output_row; - } - else - aindex = (int)channels; -# else - aindex = (int)channels; -# endif - } - - else - png_error(png_ptr, "png_write_image: internal call error"); - - /* Work out the output row end and count over this, note that the increment - * above to 'row' means that row_end can actually be beyond the end of the - * row; this is correct. - */ - row_end = output_row + image->width * (channels+1); - - for (; y > 0; --y) - { - png_const_uint_16p in_ptr = input_row; - png_uint_16p out_ptr = output_row; - - while (out_ptr < row_end) - { - png_uint_16 alpha = in_ptr[aindex]; - png_uint_32 reciprocal = 0; - int c; - - out_ptr[aindex] = alpha; - - /* Calculate a reciprocal. The correct calculation is simply - * component/alpha*65535 << 15. (I.e. 15 bits of precision); this - * allows correct rounding by adding .5 before the shift. 'reciprocal' - * is only initialized when required. - */ - if (alpha > 0 && alpha < 65535) - reciprocal = ((0xffff<<15)+(alpha>>1))/alpha; - - c = (int)channels; - do /* always at least one channel */ - { - png_uint_16 component = *in_ptr++; - - /* The following gives 65535 for an alpha of 0, which is fine, - * otherwise if 0/0 is represented as some other value there is more - * likely to be a discontinuity which will probably damage - * compression when moving from a fully transparent area to a - * nearly transparent one. (The assumption here is that opaque - * areas tend not to be 0 intensity.) - */ - if (component >= alpha) - component = 65535; - - /* component 0 && alpha < 65535) - { - png_uint_32 calc = component * reciprocal; - calc += 16384; /* round to nearest */ - component = (png_uint_16)(calc >> 15); - } - - *out_ptr++ = component; - } - while (--c > 0); - - /* Skip to next component (skip the intervening alpha channel) */ - ++in_ptr; - ++out_ptr; - } - - png_write_row(png_ptr, png_voidcast(png_const_bytep, display->local_row)); - input_row += (png_uint_16)display->row_bytes/(sizeof (png_uint_16)); - } - - return 1; -} - -/* Given 16-bit input (1 to 4 channels) write 8-bit output. If an alpha channel - * is present it must be removed from the components, the components are then - * written in sRGB encoding. No components are added or removed. - * - * Calculate an alpha reciprocal to reverse pre-multiplication. As above the - * calculation can be done to 15 bits of accuracy; however, the output needs to - * be scaled in the range 0..255*65535, so include that scaling here. - */ -# define UNP_RECIPROCAL(alpha) ((((0xffff*0xff)<<7)+((alpha)>>1))/(alpha)) - -static png_byte -png_unpremultiply(png_uint_32 component, png_uint_32 alpha, - png_uint_32 reciprocal/*from the above macro*/) -{ - /* The following gives 1.0 for an alpha of 0, which is fine, otherwise if 0/0 - * is represented as some other value there is more likely to be a - * discontinuity which will probably damage compression when moving from a - * fully transparent area to a nearly transparent one. (The assumption here - * is that opaque areas tend not to be 0 intensity.) - * - * There is a rounding problem here; if alpha is less than 128 it will end up - * as 0 when scaled to 8 bits. To avoid introducing spurious colors into the - * output change for this too. - */ - if (component >= alpha || alpha < 128) - return 255; - - /* component 0) - { - /* The test is that alpha/257 (rounded) is less than 255, the first value - * that becomes 255 is 65407. - * NOTE: this must agree with the PNG_DIV257 macro (which must, therefore, - * be exact!) [Could also test reciprocal != 0] - */ - if (alpha < 65407) - { - component *= reciprocal; - component += 64; /* round to nearest */ - component >>= 7; - } - - else - component *= 255; - - /* Convert the component to sRGB. */ - return (png_byte)PNG_sRGB_FROM_LINEAR(component); - } - - else - return 0; -} - -static int -png_write_image_8bit(png_voidp argument) -{ - png_image_write_control *display = png_voidcast(png_image_write_control*, - argument); - png_imagep image = display->image; - png_structrp png_ptr = image->opaque->png_ptr; - - png_const_uint_16p input_row = png_voidcast(png_const_uint_16p, - display->first_row); - png_bytep output_row = png_voidcast(png_bytep, display->local_row); - png_uint_32 y = image->height; - unsigned int channels = (image->format & PNG_FORMAT_FLAG_COLOR) != 0 ? - 3 : 1; - - if ((image->format & PNG_FORMAT_FLAG_ALPHA) != 0) - { - png_bytep row_end; - int aindex; - -# ifdef PNG_SIMPLIFIED_WRITE_AFIRST_SUPPORTED - if ((image->format & PNG_FORMAT_FLAG_AFIRST) != 0) - { - aindex = -1; - ++input_row; /* To point to the first component */ - ++output_row; - } - - else -# endif - aindex = (int)channels; - - /* Use row_end in place of a loop counter: */ - row_end = output_row + image->width * (channels+1); - - for (; y > 0; --y) - { - png_const_uint_16p in_ptr = input_row; - png_bytep out_ptr = output_row; - - while (out_ptr < row_end) - { - png_uint_16 alpha = in_ptr[aindex]; - png_byte alphabyte = (png_byte)PNG_DIV257(alpha); - png_uint_32 reciprocal = 0; - int c; - - /* Scale and write the alpha channel. */ - out_ptr[aindex] = alphabyte; - - if (alphabyte > 0 && alphabyte < 255) - reciprocal = UNP_RECIPROCAL(alpha); - - c = (int)channels; - do /* always at least one channel */ - *out_ptr++ = png_unpremultiply(*in_ptr++, alpha, reciprocal); - while (--c > 0); - - /* Skip to next component (skip the intervening alpha channel) */ - ++in_ptr; - ++out_ptr; - } /* while out_ptr < row_end */ - - png_write_row(png_ptr, png_voidcast(png_const_bytep, - display->local_row)); - input_row += (png_uint_16)display->row_bytes/(sizeof (png_uint_16)); - } /* while y */ - } - - else - { - /* No alpha channel, so the row_end really is the end of the row and it - * is sufficient to loop over the components one by one. - */ - png_bytep row_end = output_row + image->width * channels; - - for (; y > 0; --y) - { - png_const_uint_16p in_ptr = input_row; - png_bytep out_ptr = output_row; - - while (out_ptr < row_end) - { - png_uint_32 component = *in_ptr++; - - component *= 255; - *out_ptr++ = (png_byte)PNG_sRGB_FROM_LINEAR(component); - } - - png_write_row(png_ptr, output_row); - input_row += (png_uint_16)display->row_bytes/(sizeof (png_uint_16)); - } - } - - return 1; -} - -static void -png_image_set_PLTE(png_image_write_control *display) -{ - png_imagep image = display->image; - const void *cmap = display->colormap; - int entries = image->colormap_entries > 256 ? 256 : - (int)image->colormap_entries; - - /* NOTE: the caller must check for cmap != NULL and entries != 0 */ - png_uint_32 format = image->format; - unsigned int channels = PNG_IMAGE_SAMPLE_CHANNELS(format); - -# if defined(PNG_FORMAT_BGR_SUPPORTED) &&\ - defined(PNG_SIMPLIFIED_WRITE_AFIRST_SUPPORTED) - int afirst = (format & PNG_FORMAT_FLAG_AFIRST) != 0 && - (format & PNG_FORMAT_FLAG_ALPHA) != 0; -# else -# define afirst 0 -# endif - -# ifdef PNG_FORMAT_BGR_SUPPORTED - int bgr = (format & PNG_FORMAT_FLAG_BGR) != 0 ? 2 : 0; -# else -# define bgr 0 -# endif - - int i, num_trans; - png_color palette[256]; - png_byte tRNS[256]; - - memset(tRNS, 255, (sizeof tRNS)); - memset(palette, 0, (sizeof palette)); - - for (i=num_trans=0; i= 3) /* RGB */ - { - palette[i].blue = (png_byte)PNG_sRGB_FROM_LINEAR(255 * - entry[(2 ^ bgr)]); - palette[i].green = (png_byte)PNG_sRGB_FROM_LINEAR(255 * - entry[1]); - palette[i].red = (png_byte)PNG_sRGB_FROM_LINEAR(255 * - entry[bgr]); - } - - else /* Gray */ - palette[i].blue = palette[i].red = palette[i].green = - (png_byte)PNG_sRGB_FROM_LINEAR(255 * *entry); - } - - else /* alpha */ - { - png_uint_16 alpha = entry[afirst ? 0 : channels-1]; - png_byte alphabyte = (png_byte)PNG_DIV257(alpha); - png_uint_32 reciprocal = 0; - - /* Calculate a reciprocal, as in the png_write_image_8bit code above - * this is designed to produce a value scaled to 255*65535 when - * divided by 128 (i.e. asr 7). - */ - if (alphabyte > 0 && alphabyte < 255) - reciprocal = (((0xffff*0xff)<<7)+(alpha>>1))/alpha; - - tRNS[i] = alphabyte; - if (alphabyte < 255) - num_trans = i+1; - - if (channels >= 3) /* RGB */ - { - palette[i].blue = png_unpremultiply(entry[afirst + (2 ^ bgr)], - alpha, reciprocal); - palette[i].green = png_unpremultiply(entry[afirst + 1], alpha, - reciprocal); - palette[i].red = png_unpremultiply(entry[afirst + bgr], alpha, - reciprocal); - } - - else /* gray */ - palette[i].blue = palette[i].red = palette[i].green = - png_unpremultiply(entry[afirst], alpha, reciprocal); - } - } - - else /* Color-map has sRGB values */ - { - png_const_bytep entry = png_voidcast(png_const_bytep, cmap); - - entry += (unsigned int)i * channels; - - switch (channels) - { - case 4: - tRNS[i] = entry[afirst ? 0 : 3]; - if (tRNS[i] < 255) - num_trans = i+1; - /* FALLTHROUGH */ - case 3: - palette[i].blue = entry[afirst + (2 ^ bgr)]; - palette[i].green = entry[afirst + 1]; - palette[i].red = entry[afirst + bgr]; - break; - - case 2: - tRNS[i] = entry[1 ^ afirst]; - if (tRNS[i] < 255) - num_trans = i+1; - /* FALLTHROUGH */ - case 1: - palette[i].blue = palette[i].red = palette[i].green = - entry[afirst]; - break; - - default: - break; - } - } - } - -# ifdef afirst -# undef afirst -# endif -# ifdef bgr -# undef bgr -# endif - - png_set_PLTE(image->opaque->png_ptr, image->opaque->info_ptr, palette, - entries); - - if (num_trans > 0) - png_set_tRNS(image->opaque->png_ptr, image->opaque->info_ptr, tRNS, - num_trans, NULL); - - image->colormap_entries = (png_uint_32)entries; -} - -static int -png_image_write_main(png_voidp argument) -{ - png_image_write_control *display = png_voidcast(png_image_write_control*, - argument); - png_imagep image = display->image; - png_structrp png_ptr = image->opaque->png_ptr; - png_inforp info_ptr = image->opaque->info_ptr; - png_uint_32 format = image->format; - - /* The following four ints are actually booleans */ - int colormap = (format & PNG_FORMAT_FLAG_COLORMAP); - int linear = !colormap && (format & PNG_FORMAT_FLAG_LINEAR); /* input */ - int alpha = !colormap && (format & PNG_FORMAT_FLAG_ALPHA); - int write_16bit = linear && (display->convert_to_8bit == 0); - -# ifdef PNG_BENIGN_ERRORS_SUPPORTED - /* Make sure we error out on any bad situation */ - png_set_benign_errors(png_ptr, 0/*error*/); -# endif - - /* Default the 'row_stride' parameter if required, also check the row stride - * and total image size to ensure that they are within the system limits. - */ - { - unsigned int channels = PNG_IMAGE_PIXEL_CHANNELS(image->format); - - if (image->width <= 0x7fffffffU/channels) /* no overflow */ - { - png_uint_32 check; - png_uint_32 png_row_stride = image->width * channels; - - if (display->row_stride == 0) - display->row_stride = (png_int_32)/*SAFE*/png_row_stride; - - if (display->row_stride < 0) - check = (png_uint_32)(-display->row_stride); - - else - check = (png_uint_32)display->row_stride; - - if (check >= png_row_stride) - { - /* Now check for overflow of the image buffer calculation; this - * limits the whole image size to 32 bits for API compatibility with - * the current, 32-bit, PNG_IMAGE_BUFFER_SIZE macro. - */ - if (image->height > 0xffffffffU/png_row_stride) - png_error(image->opaque->png_ptr, "memory image too large"); - } - - else - png_error(image->opaque->png_ptr, "supplied row stride too small"); - } - - else - png_error(image->opaque->png_ptr, "image row stride too large"); - } - - /* Set the required transforms then write the rows in the correct order. */ - if ((format & PNG_FORMAT_FLAG_COLORMAP) != 0) - { - if (display->colormap != NULL && image->colormap_entries > 0) - { - png_uint_32 entries = image->colormap_entries; - - png_set_IHDR(png_ptr, info_ptr, image->width, image->height, - entries > 16 ? 8 : (entries > 4 ? 4 : (entries > 2 ? 2 : 1)), - PNG_COLOR_TYPE_PALETTE, PNG_INTERLACE_NONE, - PNG_COMPRESSION_TYPE_BASE, PNG_FILTER_TYPE_BASE); - - png_image_set_PLTE(display); - } - - else - png_error(image->opaque->png_ptr, - "no color-map for color-mapped image"); - } - - else - png_set_IHDR(png_ptr, info_ptr, image->width, image->height, - write_16bit ? 16 : 8, - ((format & PNG_FORMAT_FLAG_COLOR) ? PNG_COLOR_MASK_COLOR : 0) + - ((format & PNG_FORMAT_FLAG_ALPHA) ? PNG_COLOR_MASK_ALPHA : 0), - PNG_INTERLACE_NONE, PNG_COMPRESSION_TYPE_BASE, PNG_FILTER_TYPE_BASE); - - /* Counter-intuitively the data transformations must be called *after* - * png_write_info, not before as in the read code, but the 'set' functions - * must still be called before. Just set the color space information, never - * write an interlaced image. - */ - - if (write_16bit != 0) - { - /* The gamma here is 1.0 (linear) and the cHRM chunk matches sRGB. */ - png_set_gAMA_fixed(png_ptr, info_ptr, PNG_GAMMA_LINEAR); - - if ((image->flags & PNG_IMAGE_FLAG_COLORSPACE_NOT_sRGB) == 0) - png_set_cHRM_fixed(png_ptr, info_ptr, - /* color x y */ - /* white */ 31270, 32900, - /* red */ 64000, 33000, - /* green */ 30000, 60000, - /* blue */ 15000, 6000 - ); - } - - else if ((image->flags & PNG_IMAGE_FLAG_COLORSPACE_NOT_sRGB) == 0) - png_set_sRGB(png_ptr, info_ptr, PNG_sRGB_INTENT_PERCEPTUAL); - - /* Else writing an 8-bit file and the *colors* aren't sRGB, but the 8-bit - * space must still be gamma encoded. - */ - else - png_set_gAMA_fixed(png_ptr, info_ptr, PNG_GAMMA_sRGB_INVERSE); - - /* Write the file header. */ - png_write_info(png_ptr, info_ptr); - - /* Now set up the data transformations (*after* the header is written), - * remove the handled transformations from the 'format' flags for checking. - * - * First check for a little endian system if writing 16-bit files. - */ - if (write_16bit != 0) - { - png_uint_16 le = 0x0001; - - if ((*(png_const_bytep) & le) != 0) - png_set_swap(png_ptr); - } - -# ifdef PNG_SIMPLIFIED_WRITE_BGR_SUPPORTED - if ((format & PNG_FORMAT_FLAG_BGR) != 0) - { - if (colormap == 0 && (format & PNG_FORMAT_FLAG_COLOR) != 0) - png_set_bgr(png_ptr); - format &= ~PNG_FORMAT_FLAG_BGR; - } -# endif - -# ifdef PNG_SIMPLIFIED_WRITE_AFIRST_SUPPORTED - if ((format & PNG_FORMAT_FLAG_AFIRST) != 0) - { - if (colormap == 0 && (format & PNG_FORMAT_FLAG_ALPHA) != 0) - png_set_swap_alpha(png_ptr); - format &= ~PNG_FORMAT_FLAG_AFIRST; - } -# endif - - /* If there are 16 or fewer color-map entries we wrote a lower bit depth - * above, but the application data is still byte packed. - */ - if (colormap != 0 && image->colormap_entries <= 16) - png_set_packing(png_ptr); - - /* That should have handled all (both) the transforms. */ - if ((format & ~(png_uint_32)(PNG_FORMAT_FLAG_COLOR | PNG_FORMAT_FLAG_LINEAR | - PNG_FORMAT_FLAG_ALPHA | PNG_FORMAT_FLAG_COLORMAP)) != 0) - png_error(png_ptr, "png_write_image: unsupported transformation"); - - { - png_const_bytep row = png_voidcast(png_const_bytep, display->buffer); - ptrdiff_t row_bytes = display->row_stride; - - if (linear != 0) - row_bytes *= (sizeof (png_uint_16)); - - if (row_bytes < 0) - row += (image->height-1) * (-row_bytes); - - display->first_row = row; - display->row_bytes = row_bytes; - } - - /* Apply 'fast' options if the flag is set. */ - if ((image->flags & PNG_IMAGE_FLAG_FAST) != 0) - { - png_set_filter(png_ptr, PNG_FILTER_TYPE_BASE, PNG_NO_FILTERS); - /* NOTE: determined by experiment using pngstest, this reflects some - * balance between the time to write the image once and the time to read - * it about 50 times. The speed-up in pngstest was about 10-20% of the - * total (user) time on a heavily loaded system. - */ -# ifdef PNG_WRITE_CUSTOMIZE_COMPRESSION_SUPPORTED - png_set_compression_level(png_ptr, 3); -# endif - } - - /* Check for the cases that currently require a pre-transform on the row - * before it is written. This only applies when the input is 16-bit and - * either there is an alpha channel or it is converted to 8-bit. - */ - if ((linear != 0 && alpha != 0 ) || - (colormap == 0 && display->convert_to_8bit != 0)) - { - png_bytep row = png_voidcast(png_bytep, png_malloc(png_ptr, - png_get_rowbytes(png_ptr, info_ptr))); - int result; - - display->local_row = row; - if (write_16bit != 0) - result = png_safe_execute(image, png_write_image_16bit, display); - else - result = png_safe_execute(image, png_write_image_8bit, display); - display->local_row = NULL; - - png_free(png_ptr, row); - - /* Skip the 'write_end' on error: */ - if (result == 0) - return 0; - } - - /* Otherwise this is the case where the input is in a format currently - * supported by the rest of the libpng write code; call it directly. - */ - else - { - png_const_bytep row = png_voidcast(png_const_bytep, display->first_row); - ptrdiff_t row_bytes = display->row_bytes; - png_uint_32 y = image->height; - - for (; y > 0; --y) - { - png_write_row(png_ptr, row); - row += row_bytes; - } - } - - png_write_end(png_ptr, info_ptr); - return 1; -} - - -static void (PNGCBAPI -image_memory_write)(png_structp png_ptr, png_bytep/*const*/ data, size_t size) -{ - png_image_write_control *display = png_voidcast(png_image_write_control*, - png_ptr->io_ptr/*backdoor: png_get_io_ptr(png_ptr)*/); - png_alloc_size_t ob = display->output_bytes; - - /* Check for overflow; this should never happen: */ - if (size <= ((png_alloc_size_t)-1) - ob) - { - /* I don't think libpng ever does this, but just in case: */ - if (size > 0) - { - if (display->memory_bytes >= ob+size) /* writing */ - memcpy(display->memory+ob, data, size); - - /* Always update the size: */ - display->output_bytes = ob+size; - } - } - - else - png_error(png_ptr, "png_image_write_to_memory: PNG too big"); -} - -static void (PNGCBAPI -image_memory_flush)(png_structp png_ptr) -{ - PNG_UNUSED(png_ptr) -} - -static int -png_image_write_memory(png_voidp argument) -{ - png_image_write_control *display = png_voidcast(png_image_write_control*, - argument); - - /* The rest of the memory-specific init and write_main in an error protected - * environment. This case needs to use callbacks for the write operations - * since libpng has no built in support for writing to memory. - */ - png_set_write_fn(display->image->opaque->png_ptr, display/*io_ptr*/, - image_memory_write, image_memory_flush); - - return png_image_write_main(display); -} - -int PNGAPI -png_image_write_to_memory(png_imagep image, void *memory, - png_alloc_size_t * PNG_RESTRICT memory_bytes, int convert_to_8bit, - const void *buffer, png_int_32 row_stride, const void *colormap) -{ - /* Write the image to the given buffer, or count the bytes if it is NULL */ - if (image != NULL && image->version == PNG_IMAGE_VERSION) - { - if (memory_bytes != NULL && buffer != NULL) - { - /* This is to give the caller an easier error detection in the NULL - * case and guard against uninitialized variable problems: - */ - if (memory == NULL) - *memory_bytes = 0; - - if (png_image_write_init(image) != 0) - { - png_image_write_control display; - int result; - - memset(&display, 0, (sizeof display)); - display.image = image; - display.buffer = buffer; - display.row_stride = row_stride; - display.colormap = colormap; - display.convert_to_8bit = convert_to_8bit; - display.memory = png_voidcast(png_bytep, memory); - display.memory_bytes = *memory_bytes; - display.output_bytes = 0; - - result = png_safe_execute(image, png_image_write_memory, &display); - png_image_free(image); - - /* write_memory returns true even if we ran out of buffer. */ - if (result) - { - /* On out-of-buffer this function returns '0' but still updates - * memory_bytes: - */ - if (memory != NULL && display.output_bytes > *memory_bytes) - result = 0; - - *memory_bytes = display.output_bytes; - } - - return result; - } - - else - return 0; - } - - else - return png_image_error(image, - "png_image_write_to_memory: invalid argument"); - } - - else if (image != NULL) - return png_image_error(image, - "png_image_write_to_memory: incorrect PNG_IMAGE_VERSION"); - - else - return 0; -} - -#ifdef PNG_SIMPLIFIED_WRITE_STDIO_SUPPORTED -int PNGAPI -png_image_write_to_stdio(png_imagep image, FILE *file, int convert_to_8bit, - const void *buffer, png_int_32 row_stride, const void *colormap) -{ - /* Write the image to the given (FILE*). */ - if (image != NULL && image->version == PNG_IMAGE_VERSION) - { - if (file != NULL && buffer != NULL) - { - if (png_image_write_init(image) != 0) - { - png_image_write_control display; - int result; - - /* This is slightly evil, but png_init_io doesn't do anything other - * than this and we haven't changed the standard IO functions so - * this saves a 'safe' function. - */ - image->opaque->png_ptr->io_ptr = file; - - memset(&display, 0, (sizeof display)); - display.image = image; - display.buffer = buffer; - display.row_stride = row_stride; - display.colormap = colormap; - display.convert_to_8bit = convert_to_8bit; - - result = png_safe_execute(image, png_image_write_main, &display); - png_image_free(image); - return result; - } - - else - return 0; - } - - else - return png_image_error(image, - "png_image_write_to_stdio: invalid argument"); - } - - else if (image != NULL) - return png_image_error(image, - "png_image_write_to_stdio: incorrect PNG_IMAGE_VERSION"); - - else - return 0; -} - -int PNGAPI -png_image_write_to_file(png_imagep image, const char *file_name, - int convert_to_8bit, const void *buffer, png_int_32 row_stride, - const void *colormap) -{ - /* Write the image to the named file. */ - if (image != NULL && image->version == PNG_IMAGE_VERSION) - { - if (file_name != NULL && buffer != NULL) - { - FILE *fp = fopen(file_name, "wb"); - - if (fp != NULL) - { - if (png_image_write_to_stdio(image, fp, convert_to_8bit, buffer, - row_stride, colormap) != 0) - { - int error; /* from fflush/fclose */ - - /* Make sure the file is flushed correctly. */ - if (fflush(fp) == 0 && ferror(fp) == 0) - { - if (fclose(fp) == 0) - return 1; - - error = errno; /* from fclose */ - } - - else - { - error = errno; /* from fflush or ferror */ - (void)fclose(fp); - } - - (void)remove(file_name); - /* The image has already been cleaned up; this is just used to - * set the error (because the original write succeeded). - */ - return png_image_error(image, strerror(error)); - } - - else - { - /* Clean up: just the opened file. */ - (void)fclose(fp); - (void)remove(file_name); - return 0; - } - } - - else - return png_image_error(image, strerror(errno)); - } - - else - return png_image_error(image, - "png_image_write_to_file: invalid argument"); - } - - else if (image != NULL) - return png_image_error(image, - "png_image_write_to_file: incorrect PNG_IMAGE_VERSION"); - - else - return 0; -} -#endif /* SIMPLIFIED_WRITE_STDIO */ -#endif /* SIMPLIFIED_WRITE */ -#endif /* WRITE */ diff --git a/dep/libpng/src/pngwtran.c b/dep/libpng/src/pngwtran.c deleted file mode 100644 index 49a13c1e9..000000000 --- a/dep/libpng/src/pngwtran.c +++ /dev/null @@ -1,575 +0,0 @@ - -/* pngwtran.c - transforms the data in a row for PNG writers - * - * Copyright (c) 2018 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2016,2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -#include "pngpriv.h" - -#ifdef PNG_WRITE_SUPPORTED -#ifdef PNG_WRITE_TRANSFORMS_SUPPORTED - -#ifdef PNG_WRITE_PACK_SUPPORTED -/* Pack pixels into bytes. Pass the true bit depth in bit_depth. The - * row_info bit depth should be 8 (one pixel per byte). The channels - * should be 1 (this only happens on grayscale and paletted images). - */ -static void -png_do_pack(png_row_infop row_info, png_bytep row, png_uint_32 bit_depth) -{ - png_debug(1, "in png_do_pack"); - - if (row_info->bit_depth == 8 && - row_info->channels == 1) - { - switch ((int)bit_depth) - { - case 1: - { - png_bytep sp, dp; - int mask, v; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - sp = row; - dp = row; - mask = 0x80; - v = 0; - - for (i = 0; i < row_width; i++) - { - if (*sp != 0) - v |= mask; - - sp++; - - if (mask > 1) - mask >>= 1; - - else - { - mask = 0x80; - *dp = (png_byte)v; - dp++; - v = 0; - } - } - - if (mask != 0x80) - *dp = (png_byte)v; - - break; - } - - case 2: - { - png_bytep sp, dp; - unsigned int shift; - int v; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - sp = row; - dp = row; - shift = 6; - v = 0; - - for (i = 0; i < row_width; i++) - { - png_byte value; - - value = (png_byte)(*sp & 0x03); - v |= (value << shift); - - if (shift == 0) - { - shift = 6; - *dp = (png_byte)v; - dp++; - v = 0; - } - - else - shift -= 2; - - sp++; - } - - if (shift != 6) - *dp = (png_byte)v; - - break; - } - - case 4: - { - png_bytep sp, dp; - unsigned int shift; - int v; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - sp = row; - dp = row; - shift = 4; - v = 0; - - for (i = 0; i < row_width; i++) - { - png_byte value; - - value = (png_byte)(*sp & 0x0f); - v |= (value << shift); - - if (shift == 0) - { - shift = 4; - *dp = (png_byte)v; - dp++; - v = 0; - } - - else - shift -= 4; - - sp++; - } - - if (shift != 4) - *dp = (png_byte)v; - - break; - } - - default: - break; - } - - row_info->bit_depth = (png_byte)bit_depth; - row_info->pixel_depth = (png_byte)(bit_depth * row_info->channels); - row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth, - row_info->width); - } -} -#endif - -#ifdef PNG_WRITE_SHIFT_SUPPORTED -/* Shift pixel values to take advantage of whole range. Pass the - * true number of bits in bit_depth. The row should be packed - * according to row_info->bit_depth. Thus, if you had a row of - * bit depth 4, but the pixels only had values from 0 to 7, you - * would pass 3 as bit_depth, and this routine would translate the - * data to 0 to 15. - */ -static void -png_do_shift(png_row_infop row_info, png_bytep row, - png_const_color_8p bit_depth) -{ - png_debug(1, "in png_do_shift"); - - if (row_info->color_type != PNG_COLOR_TYPE_PALETTE) - { - int shift_start[4], shift_dec[4]; - unsigned int channels = 0; - - if ((row_info->color_type & PNG_COLOR_MASK_COLOR) != 0) - { - shift_start[channels] = row_info->bit_depth - bit_depth->red; - shift_dec[channels] = bit_depth->red; - channels++; - - shift_start[channels] = row_info->bit_depth - bit_depth->green; - shift_dec[channels] = bit_depth->green; - channels++; - - shift_start[channels] = row_info->bit_depth - bit_depth->blue; - shift_dec[channels] = bit_depth->blue; - channels++; - } - - else - { - shift_start[channels] = row_info->bit_depth - bit_depth->gray; - shift_dec[channels] = bit_depth->gray; - channels++; - } - - if ((row_info->color_type & PNG_COLOR_MASK_ALPHA) != 0) - { - shift_start[channels] = row_info->bit_depth - bit_depth->alpha; - shift_dec[channels] = bit_depth->alpha; - channels++; - } - - /* With low row depths, could only be grayscale, so one channel */ - if (row_info->bit_depth < 8) - { - png_bytep bp = row; - size_t i; - unsigned int mask; - size_t row_bytes = row_info->rowbytes; - - if (bit_depth->gray == 1 && row_info->bit_depth == 2) - mask = 0x55; - - else if (row_info->bit_depth == 4 && bit_depth->gray == 3) - mask = 0x11; - - else - mask = 0xff; - - for (i = 0; i < row_bytes; i++, bp++) - { - int j; - unsigned int v, out; - - v = *bp; - out = 0; - - for (j = shift_start[0]; j > -shift_dec[0]; j -= shift_dec[0]) - { - if (j > 0) - out |= v << j; - - else - out |= (v >> (-j)) & mask; - } - - *bp = (png_byte)(out & 0xff); - } - } - - else if (row_info->bit_depth == 8) - { - png_bytep bp = row; - png_uint_32 i; - png_uint_32 istop = channels * row_info->width; - - for (i = 0; i < istop; i++, bp++) - { - unsigned int c = i%channels; - int j; - unsigned int v, out; - - v = *bp; - out = 0; - - for (j = shift_start[c]; j > -shift_dec[c]; j -= shift_dec[c]) - { - if (j > 0) - out |= v << j; - - else - out |= v >> (-j); - } - - *bp = (png_byte)(out & 0xff); - } - } - - else - { - png_bytep bp; - png_uint_32 i; - png_uint_32 istop = channels * row_info->width; - - for (bp = row, i = 0; i < istop; i++) - { - unsigned int c = i%channels; - int j; - unsigned int value, v; - - v = png_get_uint_16(bp); - value = 0; - - for (j = shift_start[c]; j > -shift_dec[c]; j -= shift_dec[c]) - { - if (j > 0) - value |= v << j; - - else - value |= v >> (-j); - } - *bp++ = (png_byte)((value >> 8) & 0xff); - *bp++ = (png_byte)(value & 0xff); - } - } - } -} -#endif - -#ifdef PNG_WRITE_SWAP_ALPHA_SUPPORTED -static void -png_do_write_swap_alpha(png_row_infop row_info, png_bytep row) -{ - png_debug(1, "in png_do_write_swap_alpha"); - - { - if (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - { - if (row_info->bit_depth == 8) - { - /* This converts from ARGB to RGBA */ - png_bytep sp, dp; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - for (i = 0, sp = dp = row; i < row_width; i++) - { - png_byte save = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = save; - } - } - -#ifdef PNG_WRITE_16BIT_SUPPORTED - else - { - /* This converts from AARRGGBB to RRGGBBAA */ - png_bytep sp, dp; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - for (i = 0, sp = dp = row; i < row_width; i++) - { - png_byte save[2]; - save[0] = *(sp++); - save[1] = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = save[0]; - *(dp++) = save[1]; - } - } -#endif /* WRITE_16BIT */ - } - - else if (row_info->color_type == PNG_COLOR_TYPE_GRAY_ALPHA) - { - if (row_info->bit_depth == 8) - { - /* This converts from AG to GA */ - png_bytep sp, dp; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - for (i = 0, sp = dp = row; i < row_width; i++) - { - png_byte save = *(sp++); - *(dp++) = *(sp++); - *(dp++) = save; - } - } - -#ifdef PNG_WRITE_16BIT_SUPPORTED - else - { - /* This converts from AAGG to GGAA */ - png_bytep sp, dp; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - for (i = 0, sp = dp = row; i < row_width; i++) - { - png_byte save[2]; - save[0] = *(sp++); - save[1] = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = save[0]; - *(dp++) = save[1]; - } - } -#endif /* WRITE_16BIT */ - } - } -} -#endif - -#ifdef PNG_WRITE_INVERT_ALPHA_SUPPORTED -static void -png_do_write_invert_alpha(png_row_infop row_info, png_bytep row) -{ - png_debug(1, "in png_do_write_invert_alpha"); - - { - if (row_info->color_type == PNG_COLOR_TYPE_RGB_ALPHA) - { - if (row_info->bit_depth == 8) - { - /* This inverts the alpha channel in RGBA */ - png_bytep sp, dp; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - for (i = 0, sp = dp = row; i < row_width; i++) - { - /* Does nothing - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - */ - sp+=3; dp = sp; - *dp = (png_byte)(255 - *(sp++)); - } - } - -#ifdef PNG_WRITE_16BIT_SUPPORTED - else - { - /* This inverts the alpha channel in RRGGBBAA */ - png_bytep sp, dp; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - for (i = 0, sp = dp = row; i < row_width; i++) - { - /* Does nothing - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - *(dp++) = *(sp++); - */ - sp+=6; dp = sp; - *(dp++) = (png_byte)(255 - *(sp++)); - *dp = (png_byte)(255 - *(sp++)); - } - } -#endif /* WRITE_16BIT */ - } - - else if (row_info->color_type == PNG_COLOR_TYPE_GRAY_ALPHA) - { - if (row_info->bit_depth == 8) - { - /* This inverts the alpha channel in GA */ - png_bytep sp, dp; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - for (i = 0, sp = dp = row; i < row_width; i++) - { - *(dp++) = *(sp++); - *(dp++) = (png_byte)(255 - *(sp++)); - } - } - -#ifdef PNG_WRITE_16BIT_SUPPORTED - else - { - /* This inverts the alpha channel in GGAA */ - png_bytep sp, dp; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - for (i = 0, sp = dp = row; i < row_width; i++) - { - /* Does nothing - *(dp++) = *(sp++); - *(dp++) = *(sp++); - */ - sp+=2; dp = sp; - *(dp++) = (png_byte)(255 - *(sp++)); - *dp = (png_byte)(255 - *(sp++)); - } - } -#endif /* WRITE_16BIT */ - } - } -} -#endif - -/* Transform the data according to the user's wishes. The order of - * transformations is significant. - */ -void /* PRIVATE */ -png_do_write_transformations(png_structrp png_ptr, png_row_infop row_info) -{ - png_debug(1, "in png_do_write_transformations"); - - if (png_ptr == NULL) - return; - -#ifdef PNG_WRITE_USER_TRANSFORM_SUPPORTED - if ((png_ptr->transformations & PNG_USER_TRANSFORM) != 0) - if (png_ptr->write_user_transform_fn != NULL) - (*(png_ptr->write_user_transform_fn)) /* User write transform - function */ - (png_ptr, /* png_ptr */ - row_info, /* row_info: */ - /* png_uint_32 width; width of row */ - /* size_t rowbytes; number of bytes in row */ - /* png_byte color_type; color type of pixels */ - /* png_byte bit_depth; bit depth of samples */ - /* png_byte channels; number of channels (1-4) */ - /* png_byte pixel_depth; bits per pixel (depth*channels) */ - png_ptr->row_buf + 1); /* start of pixel data for row */ -#endif - -#ifdef PNG_WRITE_FILLER_SUPPORTED - if ((png_ptr->transformations & PNG_FILLER) != 0) - png_do_strip_channel(row_info, png_ptr->row_buf + 1, - !(png_ptr->flags & PNG_FLAG_FILLER_AFTER)); -#endif - -#ifdef PNG_WRITE_PACKSWAP_SUPPORTED - if ((png_ptr->transformations & PNG_PACKSWAP) != 0) - png_do_packswap(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_WRITE_PACK_SUPPORTED - if ((png_ptr->transformations & PNG_PACK) != 0) - png_do_pack(row_info, png_ptr->row_buf + 1, - (png_uint_32)png_ptr->bit_depth); -#endif - -#ifdef PNG_WRITE_SWAP_SUPPORTED -# ifdef PNG_16BIT_SUPPORTED - if ((png_ptr->transformations & PNG_SWAP_BYTES) != 0) - png_do_swap(row_info, png_ptr->row_buf + 1); -# endif -#endif - -#ifdef PNG_WRITE_SHIFT_SUPPORTED - if ((png_ptr->transformations & PNG_SHIFT) != 0) - png_do_shift(row_info, png_ptr->row_buf + 1, - &(png_ptr->shift)); -#endif - -#ifdef PNG_WRITE_SWAP_ALPHA_SUPPORTED - if ((png_ptr->transformations & PNG_SWAP_ALPHA) != 0) - png_do_write_swap_alpha(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_WRITE_INVERT_ALPHA_SUPPORTED - if ((png_ptr->transformations & PNG_INVERT_ALPHA) != 0) - png_do_write_invert_alpha(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_WRITE_BGR_SUPPORTED - if ((png_ptr->transformations & PNG_BGR) != 0) - png_do_bgr(row_info, png_ptr->row_buf + 1); -#endif - -#ifdef PNG_WRITE_INVERT_SUPPORTED - if ((png_ptr->transformations & PNG_INVERT_MONO) != 0) - png_do_invert(row_info, png_ptr->row_buf + 1); -#endif -} -#endif /* WRITE_TRANSFORMS */ -#endif /* WRITE */ diff --git a/dep/libpng/src/pngwutil.c b/dep/libpng/src/pngwutil.c deleted file mode 100644 index 14cc4ce36..000000000 --- a/dep/libpng/src/pngwutil.c +++ /dev/null @@ -1,2781 +0,0 @@ - -/* pngwutil.c - utilities to write a PNG file - * - * Copyright (c) 2018-2024 Cosmin Truta - * Copyright (c) 1998-2002,2004,2006-2018 Glenn Randers-Pehrson - * Copyright (c) 1996-1997 Andreas Dilger - * Copyright (c) 1995-1996 Guy Eric Schalnat, Group 42, Inc. - * - * This code is released under the libpng license. - * For conditions of distribution and use, see the disclaimer - * and license in png.h - */ - -#include "pngpriv.h" - -#ifdef PNG_WRITE_SUPPORTED - -#ifdef PNG_WRITE_INT_FUNCTIONS_SUPPORTED -/* Place a 32-bit number into a buffer in PNG byte order. We work - * with unsigned numbers for convenience, although one supported - * ancillary chunk uses signed (two's complement) numbers. - */ -void PNGAPI -png_save_uint_32(png_bytep buf, png_uint_32 i) -{ - buf[0] = (png_byte)((i >> 24) & 0xffU); - buf[1] = (png_byte)((i >> 16) & 0xffU); - buf[2] = (png_byte)((i >> 8) & 0xffU); - buf[3] = (png_byte)( i & 0xffU); -} - -/* Place a 16-bit number into a buffer in PNG byte order. - * The parameter is declared unsigned int, not png_uint_16, - * just to avoid potential problems on pre-ANSI C compilers. - */ -void PNGAPI -png_save_uint_16(png_bytep buf, unsigned int i) -{ - buf[0] = (png_byte)((i >> 8) & 0xffU); - buf[1] = (png_byte)( i & 0xffU); -} -#endif - -/* Simple function to write the signature. If we have already written - * the magic bytes of the signature, or more likely, the PNG stream is - * being embedded into another stream and doesn't need its own signature, - * we should call png_set_sig_bytes() to tell libpng how many of the - * bytes have already been written. - */ -void PNGAPI -png_write_sig(png_structrp png_ptr) -{ - png_byte png_signature[8] = {137, 80, 78, 71, 13, 10, 26, 10}; - -#ifdef PNG_IO_STATE_SUPPORTED - /* Inform the I/O callback that the signature is being written */ - png_ptr->io_state = PNG_IO_WRITING | PNG_IO_SIGNATURE; -#endif - - /* Write the rest of the 8 byte signature */ - png_write_data(png_ptr, &png_signature[png_ptr->sig_bytes], - (size_t)(8 - png_ptr->sig_bytes)); - - if (png_ptr->sig_bytes < 3) - png_ptr->mode |= PNG_HAVE_PNG_SIGNATURE; -} - -/* Write the start of a PNG chunk. The type is the chunk type. - * The total_length is the sum of the lengths of all the data you will be - * passing in png_write_chunk_data(). - */ -static void -png_write_chunk_header(png_structrp png_ptr, png_uint_32 chunk_name, - png_uint_32 length) -{ - png_byte buf[8]; - -#if defined(PNG_DEBUG) && (PNG_DEBUG > 0) - PNG_CSTRING_FROM_CHUNK(buf, chunk_name); - png_debug2(0, "Writing %s chunk, length = %lu", buf, (unsigned long)length); -#endif - - if (png_ptr == NULL) - return; - -#ifdef PNG_IO_STATE_SUPPORTED - /* Inform the I/O callback that the chunk header is being written. - * PNG_IO_CHUNK_HDR requires a single I/O call. - */ - png_ptr->io_state = PNG_IO_WRITING | PNG_IO_CHUNK_HDR; -#endif - - /* Write the length and the chunk name */ - png_save_uint_32(buf, length); - png_save_uint_32(buf + 4, chunk_name); - png_write_data(png_ptr, buf, 8); - - /* Put the chunk name into png_ptr->chunk_name */ - png_ptr->chunk_name = chunk_name; - - /* Reset the crc and run it over the chunk name */ - png_reset_crc(png_ptr); - - png_calculate_crc(png_ptr, buf + 4, 4); - -#ifdef PNG_IO_STATE_SUPPORTED - /* Inform the I/O callback that chunk data will (possibly) be written. - * PNG_IO_CHUNK_DATA does NOT require a specific number of I/O calls. - */ - png_ptr->io_state = PNG_IO_WRITING | PNG_IO_CHUNK_DATA; -#endif -} - -void PNGAPI -png_write_chunk_start(png_structrp png_ptr, png_const_bytep chunk_string, - png_uint_32 length) -{ - png_write_chunk_header(png_ptr, PNG_CHUNK_FROM_STRING(chunk_string), length); -} - -/* Write the data of a PNG chunk started with png_write_chunk_header(). - * Note that multiple calls to this function are allowed, and that the - * sum of the lengths from these calls *must* add up to the total_length - * given to png_write_chunk_header(). - */ -void PNGAPI -png_write_chunk_data(png_structrp png_ptr, png_const_bytep data, size_t length) -{ - /* Write the data, and run the CRC over it */ - if (png_ptr == NULL) - return; - - if (data != NULL && length > 0) - { - png_write_data(png_ptr, data, length); - - /* Update the CRC after writing the data, - * in case the user I/O routine alters it. - */ - png_calculate_crc(png_ptr, data, length); - } -} - -/* Finish a chunk started with png_write_chunk_header(). */ -void PNGAPI -png_write_chunk_end(png_structrp png_ptr) -{ - png_byte buf[4]; - - if (png_ptr == NULL) return; - -#ifdef PNG_IO_STATE_SUPPORTED - /* Inform the I/O callback that the chunk CRC is being written. - * PNG_IO_CHUNK_CRC requires a single I/O function call. - */ - png_ptr->io_state = PNG_IO_WRITING | PNG_IO_CHUNK_CRC; -#endif - - /* Write the crc in a single operation */ - png_save_uint_32(buf, png_ptr->crc); - - png_write_data(png_ptr, buf, 4); -} - -/* Write a PNG chunk all at once. The type is an array of ASCII characters - * representing the chunk name. The array must be at least 4 bytes in - * length, and does not need to be null terminated. To be safe, pass the - * pre-defined chunk names here, and if you need a new one, define it - * where the others are defined. The length is the length of the data. - * All the data must be present. If that is not possible, use the - * png_write_chunk_start(), png_write_chunk_data(), and png_write_chunk_end() - * functions instead. - */ -static void -png_write_complete_chunk(png_structrp png_ptr, png_uint_32 chunk_name, - png_const_bytep data, size_t length) -{ - if (png_ptr == NULL) - return; - - /* On 64-bit architectures 'length' may not fit in a png_uint_32. */ - if (length > PNG_UINT_31_MAX) - png_error(png_ptr, "length exceeds PNG maximum"); - - png_write_chunk_header(png_ptr, chunk_name, (png_uint_32)length); - png_write_chunk_data(png_ptr, data, length); - png_write_chunk_end(png_ptr); -} - -/* This is the API that calls the internal function above. */ -void PNGAPI -png_write_chunk(png_structrp png_ptr, png_const_bytep chunk_string, - png_const_bytep data, size_t length) -{ - png_write_complete_chunk(png_ptr, PNG_CHUNK_FROM_STRING(chunk_string), data, - length); -} - -/* This is used below to find the size of an image to pass to png_deflate_claim, - * so it only needs to be accurate if the size is less than 16384 bytes (the - * point at which a lower LZ window size can be used.) - */ -static png_alloc_size_t -png_image_size(png_structrp png_ptr) -{ - /* Only return sizes up to the maximum of a png_uint_32; do this by limiting - * the width and height used to 15 bits. - */ - png_uint_32 h = png_ptr->height; - - if (png_ptr->rowbytes < 32768 && h < 32768) - { - if (png_ptr->interlaced != 0) - { - /* Interlacing makes the image larger because of the replication of - * both the filter byte and the padding to a byte boundary. - */ - png_uint_32 w = png_ptr->width; - unsigned int pd = png_ptr->pixel_depth; - png_alloc_size_t cb_base; - int pass; - - for (cb_base=0, pass=0; pass<=6; ++pass) - { - png_uint_32 pw = PNG_PASS_COLS(w, pass); - - if (pw > 0) - cb_base += (PNG_ROWBYTES(pd, pw)+1) * PNG_PASS_ROWS(h, pass); - } - - return cb_base; - } - - else - return (png_ptr->rowbytes+1) * h; - } - - else - return 0xffffffffU; -} - -#ifdef PNG_WRITE_OPTIMIZE_CMF_SUPPORTED - /* This is the code to hack the first two bytes of the deflate stream (the - * deflate header) to correct the windowBits value to match the actual data - * size. Note that the second argument is the *uncompressed* size but the - * first argument is the *compressed* data (and it must be deflate - * compressed.) - */ -static void -optimize_cmf(png_bytep data, png_alloc_size_t data_size) -{ - /* Optimize the CMF field in the zlib stream. The resultant zlib stream is - * still compliant to the stream specification. - */ - if (data_size <= 16384) /* else windowBits must be 15 */ - { - unsigned int z_cmf = data[0]; /* zlib compression method and flags */ - - if ((z_cmf & 0x0f) == 8 && (z_cmf & 0xf0) <= 0x70) - { - unsigned int z_cinfo; - unsigned int half_z_window_size; - - z_cinfo = z_cmf >> 4; - half_z_window_size = 1U << (z_cinfo + 7); - - if (data_size <= half_z_window_size) /* else no change */ - { - unsigned int tmp; - - do - { - half_z_window_size >>= 1; - --z_cinfo; - } - while (z_cinfo > 0 && data_size <= half_z_window_size); - - z_cmf = (z_cmf & 0x0f) | (z_cinfo << 4); - - data[0] = (png_byte)z_cmf; - tmp = data[1] & 0xe0; - tmp += 0x1f - ((z_cmf << 8) + tmp) % 0x1f; - data[1] = (png_byte)tmp; - } - } - } -} -#endif /* WRITE_OPTIMIZE_CMF */ - -/* Initialize the compressor for the appropriate type of compression. */ -static int -png_deflate_claim(png_structrp png_ptr, png_uint_32 owner, - png_alloc_size_t data_size) -{ - if (png_ptr->zowner != 0) - { -#if defined(PNG_WARNINGS_SUPPORTED) || defined(PNG_ERROR_TEXT_SUPPORTED) - char msg[64]; - - PNG_STRING_FROM_CHUNK(msg, owner); - msg[4] = ':'; - msg[5] = ' '; - PNG_STRING_FROM_CHUNK(msg+6, png_ptr->zowner); - /* So the message that results is " using zstream"; this is an - * internal error, but is very useful for debugging. i18n requirements - * are minimal. - */ - (void)png_safecat(msg, (sizeof msg), 10, " using zstream"); -#endif -#if PNG_RELEASE_BUILD - png_warning(png_ptr, msg); - - /* Attempt sane error recovery */ - if (png_ptr->zowner == png_IDAT) /* don't steal from IDAT */ - { - png_ptr->zstream.msg = PNGZ_MSG_CAST("in use by IDAT"); - return Z_STREAM_ERROR; - } - - png_ptr->zowner = 0; -#else - png_error(png_ptr, msg); -#endif - } - - { - int level = png_ptr->zlib_level; - int method = png_ptr->zlib_method; - int windowBits = png_ptr->zlib_window_bits; - int memLevel = png_ptr->zlib_mem_level; - int strategy; /* set below */ - int ret; /* zlib return code */ - - if (owner == png_IDAT) - { - if ((png_ptr->flags & PNG_FLAG_ZLIB_CUSTOM_STRATEGY) != 0) - strategy = png_ptr->zlib_strategy; - - else if (png_ptr->do_filter != PNG_FILTER_NONE) - strategy = PNG_Z_DEFAULT_STRATEGY; - - else - strategy = PNG_Z_DEFAULT_NOFILTER_STRATEGY; - } - - else - { -#ifdef PNG_WRITE_CUSTOMIZE_ZTXT_COMPRESSION_SUPPORTED - level = png_ptr->zlib_text_level; - method = png_ptr->zlib_text_method; - windowBits = png_ptr->zlib_text_window_bits; - memLevel = png_ptr->zlib_text_mem_level; - strategy = png_ptr->zlib_text_strategy; -#else - /* If customization is not supported the values all come from the - * IDAT values except for the strategy, which is fixed to the - * default. (This is the pre-1.6.0 behavior too, although it was - * implemented in a very different way.) - */ - strategy = Z_DEFAULT_STRATEGY; -#endif - } - - /* Adjust 'windowBits' down if larger than 'data_size'; to stop this - * happening just pass 32768 as the data_size parameter. Notice that zlib - * requires an extra 262 bytes in the window in addition to the data to be - * able to see the whole of the data, so if data_size+262 takes us to the - * next windowBits size we need to fix up the value later. (Because even - * though deflate needs the extra window, inflate does not!) - */ - if (data_size <= 16384) - { - /* IMPLEMENTATION NOTE: this 'half_window_size' stuff is only here to - * work round a Microsoft Visual C misbehavior which, contrary to C-90, - * widens the result of the following shift to 64-bits if (and, - * apparently, only if) it is used in a test. - */ - unsigned int half_window_size = 1U << (windowBits-1); - - while (data_size + 262 <= half_window_size) - { - half_window_size >>= 1; - --windowBits; - } - } - - /* Check against the previous initialized values, if any. */ - if ((png_ptr->flags & PNG_FLAG_ZSTREAM_INITIALIZED) != 0 && - (png_ptr->zlib_set_level != level || - png_ptr->zlib_set_method != method || - png_ptr->zlib_set_window_bits != windowBits || - png_ptr->zlib_set_mem_level != memLevel || - png_ptr->zlib_set_strategy != strategy)) - { - if (deflateEnd(&png_ptr->zstream) != Z_OK) - png_warning(png_ptr, "deflateEnd failed (ignored)"); - - png_ptr->flags &= ~PNG_FLAG_ZSTREAM_INITIALIZED; - } - - /* For safety clear out the input and output pointers (currently zlib - * doesn't use them on Init, but it might in the future). - */ - png_ptr->zstream.next_in = NULL; - png_ptr->zstream.avail_in = 0; - png_ptr->zstream.next_out = NULL; - png_ptr->zstream.avail_out = 0; - - /* Now initialize if required, setting the new parameters, otherwise just - * do a simple reset to the previous parameters. - */ - if ((png_ptr->flags & PNG_FLAG_ZSTREAM_INITIALIZED) != 0) - ret = deflateReset(&png_ptr->zstream); - - else - { - ret = deflateInit2(&png_ptr->zstream, level, method, windowBits, - memLevel, strategy); - - if (ret == Z_OK) - png_ptr->flags |= PNG_FLAG_ZSTREAM_INITIALIZED; - } - - /* The return code is from either deflateReset or deflateInit2; they have - * pretty much the same set of error codes. - */ - if (ret == Z_OK) - png_ptr->zowner = owner; - - else - png_zstream_error(png_ptr, ret); - - return ret; - } -} - -/* Clean up (or trim) a linked list of compression buffers. */ -void /* PRIVATE */ -png_free_buffer_list(png_structrp png_ptr, png_compression_bufferp *listp) -{ - png_compression_bufferp list = *listp; - - if (list != NULL) - { - *listp = NULL; - - do - { - png_compression_bufferp next = list->next; - - png_free(png_ptr, list); - list = next; - } - while (list != NULL); - } -} - -#ifdef PNG_WRITE_COMPRESSED_TEXT_SUPPORTED -/* This pair of functions encapsulates the operation of (a) compressing a - * text string, and (b) issuing it later as a series of chunk data writes. - * The compression_state structure is shared context for these functions - * set up by the caller to allow access to the relevant local variables. - * - * compression_buffer (new in 1.6.0) is just a linked list of zbuffer_size - * temporary buffers. From 1.6.0 it is retained in png_struct so that it will - * be correctly freed in the event of a write error (previous implementations - * just leaked memory.) - */ -typedef struct -{ - png_const_bytep input; /* The uncompressed input data */ - png_alloc_size_t input_len; /* Its length */ - png_uint_32 output_len; /* Final compressed length */ - png_byte output[1024]; /* First block of output */ -} compression_state; - -static void -png_text_compress_init(compression_state *comp, png_const_bytep input, - png_alloc_size_t input_len) -{ - comp->input = input; - comp->input_len = input_len; - comp->output_len = 0; -} - -/* Compress the data in the compression state input */ -static int -png_text_compress(png_structrp png_ptr, png_uint_32 chunk_name, - compression_state *comp, png_uint_32 prefix_len) -{ - int ret; - - /* To find the length of the output it is necessary to first compress the - * input. The result is buffered rather than using the two-pass algorithm - * that is used on the inflate side; deflate is assumed to be slower and a - * PNG writer is assumed to have more memory available than a PNG reader. - * - * IMPLEMENTATION NOTE: the zlib API deflateBound() can be used to find an - * upper limit on the output size, but it is always bigger than the input - * size so it is likely to be more efficient to use this linked-list - * approach. - */ - ret = png_deflate_claim(png_ptr, chunk_name, comp->input_len); - - if (ret != Z_OK) - return ret; - - /* Set up the compression buffers, we need a loop here to avoid overflowing a - * uInt. Use ZLIB_IO_MAX to limit the input. The output is always limited - * by the output buffer size, so there is no need to check that. Since this - * is ANSI-C we know that an 'int', hence a uInt, is always at least 16 bits - * in size. - */ - { - png_compression_bufferp *end = &png_ptr->zbuffer_list; - png_alloc_size_t input_len = comp->input_len; /* may be zero! */ - png_uint_32 output_len; - - /* zlib updates these for us: */ - png_ptr->zstream.next_in = PNGZ_INPUT_CAST(comp->input); - png_ptr->zstream.avail_in = 0; /* Set below */ - png_ptr->zstream.next_out = comp->output; - png_ptr->zstream.avail_out = (sizeof comp->output); - - output_len = png_ptr->zstream.avail_out; - - do - { - uInt avail_in = ZLIB_IO_MAX; - - if (avail_in > input_len) - avail_in = (uInt)input_len; - - input_len -= avail_in; - - png_ptr->zstream.avail_in = avail_in; - - if (png_ptr->zstream.avail_out == 0) - { - png_compression_buffer *next; - - /* Chunk data is limited to 2^31 bytes in length, so the prefix - * length must be counted here. - */ - if (output_len + prefix_len > PNG_UINT_31_MAX) - { - ret = Z_MEM_ERROR; - break; - } - - /* Need a new (malloc'ed) buffer, but there may be one present - * already. - */ - next = *end; - if (next == NULL) - { - next = png_voidcast(png_compression_bufferp, png_malloc_base - (png_ptr, PNG_COMPRESSION_BUFFER_SIZE(png_ptr))); - - if (next == NULL) - { - ret = Z_MEM_ERROR; - break; - } - - /* Link in this buffer (so that it will be freed later) */ - next->next = NULL; - *end = next; - } - - png_ptr->zstream.next_out = next->output; - png_ptr->zstream.avail_out = png_ptr->zbuffer_size; - output_len += png_ptr->zstream.avail_out; - - /* Move 'end' to the next buffer pointer. */ - end = &next->next; - } - - /* Compress the data */ - ret = deflate(&png_ptr->zstream, - input_len > 0 ? Z_NO_FLUSH : Z_FINISH); - - /* Claw back input data that was not consumed (because avail_in is - * reset above every time round the loop). - */ - input_len += png_ptr->zstream.avail_in; - png_ptr->zstream.avail_in = 0; /* safety */ - } - while (ret == Z_OK); - - /* There may be some space left in the last output buffer. This needs to - * be subtracted from output_len. - */ - output_len -= png_ptr->zstream.avail_out; - png_ptr->zstream.avail_out = 0; /* safety */ - comp->output_len = output_len; - - /* Now double check the output length, put in a custom message if it is - * too long. Otherwise ensure the z_stream::msg pointer is set to - * something. - */ - if (output_len + prefix_len >= PNG_UINT_31_MAX) - { - png_ptr->zstream.msg = PNGZ_MSG_CAST("compressed data too long"); - ret = Z_MEM_ERROR; - } - - else - png_zstream_error(png_ptr, ret); - - /* Reset zlib for another zTXt/iTXt or image data */ - png_ptr->zowner = 0; - - /* The only success case is Z_STREAM_END, input_len must be 0; if not this - * is an internal error. - */ - if (ret == Z_STREAM_END && input_len == 0) - { -#ifdef PNG_WRITE_OPTIMIZE_CMF_SUPPORTED - /* Fix up the deflate header, if required */ - optimize_cmf(comp->output, comp->input_len); -#endif - /* But Z_OK is returned, not Z_STREAM_END; this allows the claim - * function above to return Z_STREAM_END on an error (though it never - * does in the current versions of zlib.) - */ - return Z_OK; - } - - else - return ret; - } -} - -/* Ship the compressed text out via chunk writes */ -static void -png_write_compressed_data_out(png_structrp png_ptr, compression_state *comp) -{ - png_uint_32 output_len = comp->output_len; - png_const_bytep output = comp->output; - png_uint_32 avail = (sizeof comp->output); - png_compression_buffer *next = png_ptr->zbuffer_list; - - for (;;) - { - if (avail > output_len) - avail = output_len; - - png_write_chunk_data(png_ptr, output, avail); - - output_len -= avail; - - if (output_len == 0 || next == NULL) - break; - - avail = png_ptr->zbuffer_size; - output = next->output; - next = next->next; - } - - /* This is an internal error; 'next' must have been NULL! */ - if (output_len > 0) - png_error(png_ptr, "error writing ancillary chunked compressed data"); -} -#endif /* WRITE_COMPRESSED_TEXT */ - -/* Write the IHDR chunk, and update the png_struct with the necessary - * information. Note that the rest of this code depends upon this - * information being correct. - */ -void /* PRIVATE */ -png_write_IHDR(png_structrp png_ptr, png_uint_32 width, png_uint_32 height, - int bit_depth, int color_type, int compression_type, int filter_type, - int interlace_type) -{ - png_byte buf[13]; /* Buffer to store the IHDR info */ - int is_invalid_depth; - - png_debug(1, "in png_write_IHDR"); - - /* Check that we have valid input data from the application info */ - switch (color_type) - { - case PNG_COLOR_TYPE_GRAY: - switch (bit_depth) - { - case 1: - case 2: - case 4: - case 8: -#ifdef PNG_WRITE_16BIT_SUPPORTED - case 16: -#endif - png_ptr->channels = 1; break; - - default: - png_error(png_ptr, - "Invalid bit depth for grayscale image"); - } - break; - - case PNG_COLOR_TYPE_RGB: - is_invalid_depth = (bit_depth != 8); -#ifdef PNG_WRITE_16BIT_SUPPORTED - is_invalid_depth = (is_invalid_depth && bit_depth != 16); -#endif - if (is_invalid_depth) - png_error(png_ptr, "Invalid bit depth for RGB image"); - - png_ptr->channels = 3; - break; - - case PNG_COLOR_TYPE_PALETTE: - switch (bit_depth) - { - case 1: - case 2: - case 4: - case 8: - png_ptr->channels = 1; - break; - - default: - png_error(png_ptr, "Invalid bit depth for paletted image"); - } - break; - - case PNG_COLOR_TYPE_GRAY_ALPHA: - is_invalid_depth = (bit_depth != 8); -#ifdef PNG_WRITE_16BIT_SUPPORTED - is_invalid_depth = (is_invalid_depth && bit_depth != 16); -#endif - if (is_invalid_depth) - png_error(png_ptr, "Invalid bit depth for grayscale+alpha image"); - - png_ptr->channels = 2; - break; - - case PNG_COLOR_TYPE_RGB_ALPHA: - is_invalid_depth = (bit_depth != 8); -#ifdef PNG_WRITE_16BIT_SUPPORTED - is_invalid_depth = (is_invalid_depth && bit_depth != 16); -#endif - if (is_invalid_depth) - png_error(png_ptr, "Invalid bit depth for RGBA image"); - - png_ptr->channels = 4; - break; - - default: - png_error(png_ptr, "Invalid image color type specified"); - } - - if (compression_type != PNG_COMPRESSION_TYPE_BASE) - { - png_warning(png_ptr, "Invalid compression type specified"); - compression_type = PNG_COMPRESSION_TYPE_BASE; - } - - /* Write filter_method 64 (intrapixel differencing) only if - * 1. Libpng was compiled with PNG_MNG_FEATURES_SUPPORTED and - * 2. Libpng did not write a PNG signature (this filter_method is only - * used in PNG datastreams that are embedded in MNG datastreams) and - * 3. The application called png_permit_mng_features with a mask that - * included PNG_FLAG_MNG_FILTER_64 and - * 4. The filter_method is 64 and - * 5. The color_type is RGB or RGBA - */ - if ( -#ifdef PNG_MNG_FEATURES_SUPPORTED - !((png_ptr->mng_features_permitted & PNG_FLAG_MNG_FILTER_64) != 0 && - ((png_ptr->mode & PNG_HAVE_PNG_SIGNATURE) == 0) && - (color_type == PNG_COLOR_TYPE_RGB || - color_type == PNG_COLOR_TYPE_RGB_ALPHA) && - (filter_type == PNG_INTRAPIXEL_DIFFERENCING)) && -#endif - filter_type != PNG_FILTER_TYPE_BASE) - { - png_warning(png_ptr, "Invalid filter type specified"); - filter_type = PNG_FILTER_TYPE_BASE; - } - -#ifdef PNG_WRITE_INTERLACING_SUPPORTED - if (interlace_type != PNG_INTERLACE_NONE && - interlace_type != PNG_INTERLACE_ADAM7) - { - png_warning(png_ptr, "Invalid interlace type specified"); - interlace_type = PNG_INTERLACE_ADAM7; - } -#else - interlace_type=PNG_INTERLACE_NONE; -#endif - - /* Save the relevant information */ - png_ptr->bit_depth = (png_byte)bit_depth; - png_ptr->color_type = (png_byte)color_type; - png_ptr->interlaced = (png_byte)interlace_type; -#ifdef PNG_MNG_FEATURES_SUPPORTED - png_ptr->filter_type = (png_byte)filter_type; -#endif - png_ptr->compression_type = (png_byte)compression_type; - png_ptr->width = width; - png_ptr->height = height; - - png_ptr->pixel_depth = (png_byte)(bit_depth * png_ptr->channels); - png_ptr->rowbytes = PNG_ROWBYTES(png_ptr->pixel_depth, width); - /* Set the usr info, so any transformations can modify it */ - png_ptr->usr_width = png_ptr->width; - png_ptr->usr_bit_depth = png_ptr->bit_depth; - png_ptr->usr_channels = png_ptr->channels; - - /* Pack the header information into the buffer */ - png_save_uint_32(buf, width); - png_save_uint_32(buf + 4, height); - buf[8] = (png_byte)bit_depth; - buf[9] = (png_byte)color_type; - buf[10] = (png_byte)compression_type; - buf[11] = (png_byte)filter_type; - buf[12] = (png_byte)interlace_type; - - /* Write the chunk */ - png_write_complete_chunk(png_ptr, png_IHDR, buf, 13); - - if ((png_ptr->do_filter) == PNG_NO_FILTERS) - { - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE || - png_ptr->bit_depth < 8) - png_ptr->do_filter = PNG_FILTER_NONE; - - else - png_ptr->do_filter = PNG_ALL_FILTERS; - } - - png_ptr->mode = PNG_HAVE_IHDR; /* not READY_FOR_ZTXT */ -} - -/* Write the palette. We are careful not to trust png_color to be in the - * correct order for PNG, so people can redefine it to any convenient - * structure. - */ -void /* PRIVATE */ -png_write_PLTE(png_structrp png_ptr, png_const_colorp palette, - png_uint_32 num_pal) -{ - png_uint_32 max_palette_length, i; - png_const_colorp pal_ptr; - png_byte buf[3]; - - png_debug(1, "in png_write_PLTE"); - - max_palette_length = (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) ? - (1 << png_ptr->bit_depth) : PNG_MAX_PALETTE_LENGTH; - - if (( -#ifdef PNG_MNG_FEATURES_SUPPORTED - (png_ptr->mng_features_permitted & PNG_FLAG_MNG_EMPTY_PLTE) == 0 && -#endif - num_pal == 0) || num_pal > max_palette_length) - { - if (png_ptr->color_type == PNG_COLOR_TYPE_PALETTE) - { - png_error(png_ptr, "Invalid number of colors in palette"); - } - - else - { - png_warning(png_ptr, "Invalid number of colors in palette"); - return; - } - } - - if ((png_ptr->color_type & PNG_COLOR_MASK_COLOR) == 0) - { - png_warning(png_ptr, - "Ignoring request to write a PLTE chunk in grayscale PNG"); - - return; - } - - png_ptr->num_palette = (png_uint_16)num_pal; - png_debug1(3, "num_palette = %d", png_ptr->num_palette); - - png_write_chunk_header(png_ptr, png_PLTE, (png_uint_32)(num_pal * 3)); -#ifdef PNG_POINTER_INDEXING_SUPPORTED - - for (i = 0, pal_ptr = palette; i < num_pal; i++, pal_ptr++) - { - buf[0] = pal_ptr->red; - buf[1] = pal_ptr->green; - buf[2] = pal_ptr->blue; - png_write_chunk_data(png_ptr, buf, 3); - } - -#else - /* This is a little slower but some buggy compilers need to do this - * instead - */ - pal_ptr=palette; - - for (i = 0; i < num_pal; i++) - { - buf[0] = pal_ptr[i].red; - buf[1] = pal_ptr[i].green; - buf[2] = pal_ptr[i].blue; - png_write_chunk_data(png_ptr, buf, 3); - } - -#endif - png_write_chunk_end(png_ptr); - png_ptr->mode |= PNG_HAVE_PLTE; -} - -/* This is similar to png_text_compress, above, except that it does not require - * all of the data at once and, instead of buffering the compressed result, - * writes it as IDAT chunks. Unlike png_text_compress it *can* png_error out - * because it calls the write interface. As a result it does its own error - * reporting and does not return an error code. In the event of error it will - * just call png_error. The input data length may exceed 32-bits. The 'flush' - * parameter is exactly the same as that to deflate, with the following - * meanings: - * - * Z_NO_FLUSH: normal incremental output of compressed data - * Z_SYNC_FLUSH: do a SYNC_FLUSH, used by png_write_flush - * Z_FINISH: this is the end of the input, do a Z_FINISH and clean up - * - * The routine manages the acquire and release of the png_ptr->zstream by - * checking and (at the end) clearing png_ptr->zowner; it does some sanity - * checks on the 'mode' flags while doing this. - */ -void /* PRIVATE */ -png_compress_IDAT(png_structrp png_ptr, png_const_bytep input, - png_alloc_size_t input_len, int flush) -{ - if (png_ptr->zowner != png_IDAT) - { - /* First time. Ensure we have a temporary buffer for compression and - * trim the buffer list if it has more than one entry to free memory. - * If 'WRITE_COMPRESSED_TEXT' is not set the list will never have been - * created at this point, but the check here is quick and safe. - */ - if (png_ptr->zbuffer_list == NULL) - { - png_ptr->zbuffer_list = png_voidcast(png_compression_bufferp, - png_malloc(png_ptr, PNG_COMPRESSION_BUFFER_SIZE(png_ptr))); - png_ptr->zbuffer_list->next = NULL; - } - - else - png_free_buffer_list(png_ptr, &png_ptr->zbuffer_list->next); - - /* It is a terminal error if we can't claim the zstream. */ - if (png_deflate_claim(png_ptr, png_IDAT, png_image_size(png_ptr)) != Z_OK) - png_error(png_ptr, png_ptr->zstream.msg); - - /* The output state is maintained in png_ptr->zstream, so it must be - * initialized here after the claim. - */ - png_ptr->zstream.next_out = png_ptr->zbuffer_list->output; - png_ptr->zstream.avail_out = png_ptr->zbuffer_size; - } - - /* Now loop reading and writing until all the input is consumed or an error - * terminates the operation. The _out values are maintained across calls to - * this function, but the input must be reset each time. - */ - png_ptr->zstream.next_in = PNGZ_INPUT_CAST(input); - png_ptr->zstream.avail_in = 0; /* set below */ - for (;;) - { - int ret; - - /* INPUT: from the row data */ - uInt avail = ZLIB_IO_MAX; - - if (avail > input_len) - avail = (uInt)input_len; /* safe because of the check */ - - png_ptr->zstream.avail_in = avail; - input_len -= avail; - - ret = deflate(&png_ptr->zstream, input_len > 0 ? Z_NO_FLUSH : flush); - - /* Include as-yet unconsumed input */ - input_len += png_ptr->zstream.avail_in; - png_ptr->zstream.avail_in = 0; - - /* OUTPUT: write complete IDAT chunks when avail_out drops to zero. Note - * that these two zstream fields are preserved across the calls, therefore - * there is no need to set these up on entry to the loop. - */ - if (png_ptr->zstream.avail_out == 0) - { - png_bytep data = png_ptr->zbuffer_list->output; - uInt size = png_ptr->zbuffer_size; - - /* Write an IDAT containing the data then reset the buffer. The - * first IDAT may need deflate header optimization. - */ -#ifdef PNG_WRITE_OPTIMIZE_CMF_SUPPORTED - if ((png_ptr->mode & PNG_HAVE_IDAT) == 0 && - png_ptr->compression_type == PNG_COMPRESSION_TYPE_BASE) - optimize_cmf(data, png_image_size(png_ptr)); -#endif - - if (size > 0) - png_write_complete_chunk(png_ptr, png_IDAT, data, size); - png_ptr->mode |= PNG_HAVE_IDAT; - - png_ptr->zstream.next_out = data; - png_ptr->zstream.avail_out = size; - - /* For SYNC_FLUSH or FINISH it is essential to keep calling zlib with - * the same flush parameter until it has finished output, for NO_FLUSH - * it doesn't matter. - */ - if (ret == Z_OK && flush != Z_NO_FLUSH) - continue; - } - - /* The order of these checks doesn't matter much; it just affects which - * possible error might be detected if multiple things go wrong at once. - */ - if (ret == Z_OK) /* most likely return code! */ - { - /* If all the input has been consumed then just return. If Z_FINISH - * was used as the flush parameter something has gone wrong if we get - * here. - */ - if (input_len == 0) - { - if (flush == Z_FINISH) - png_error(png_ptr, "Z_OK on Z_FINISH with output space"); - - return; - } - } - - else if (ret == Z_STREAM_END && flush == Z_FINISH) - { - /* This is the end of the IDAT data; any pending output must be - * flushed. For small PNG files we may still be at the beginning. - */ - png_bytep data = png_ptr->zbuffer_list->output; - uInt size = png_ptr->zbuffer_size - png_ptr->zstream.avail_out; - -#ifdef PNG_WRITE_OPTIMIZE_CMF_SUPPORTED - if ((png_ptr->mode & PNG_HAVE_IDAT) == 0 && - png_ptr->compression_type == PNG_COMPRESSION_TYPE_BASE) - optimize_cmf(data, png_image_size(png_ptr)); -#endif - - if (size > 0) - png_write_complete_chunk(png_ptr, png_IDAT, data, size); - png_ptr->zstream.avail_out = 0; - png_ptr->zstream.next_out = NULL; - png_ptr->mode |= PNG_HAVE_IDAT | PNG_AFTER_IDAT; - - png_ptr->zowner = 0; /* Release the stream */ - return; - } - - else - { - /* This is an error condition. */ - png_zstream_error(png_ptr, ret); - png_error(png_ptr, png_ptr->zstream.msg); - } - } -} - -/* Write an IEND chunk */ -void /* PRIVATE */ -png_write_IEND(png_structrp png_ptr) -{ - png_debug(1, "in png_write_IEND"); - - png_write_complete_chunk(png_ptr, png_IEND, NULL, 0); - png_ptr->mode |= PNG_HAVE_IEND; -} - -#ifdef PNG_WRITE_gAMA_SUPPORTED -/* Write a gAMA chunk */ -void /* PRIVATE */ -png_write_gAMA_fixed(png_structrp png_ptr, png_fixed_point file_gamma) -{ - png_byte buf[4]; - - png_debug(1, "in png_write_gAMA"); - - /* file_gamma is saved in 1/100,000ths */ - png_save_uint_32(buf, (png_uint_32)file_gamma); - png_write_complete_chunk(png_ptr, png_gAMA, buf, 4); -} -#endif - -#ifdef PNG_WRITE_sRGB_SUPPORTED -/* Write a sRGB chunk */ -void /* PRIVATE */ -png_write_sRGB(png_structrp png_ptr, int srgb_intent) -{ - png_byte buf[1]; - - png_debug(1, "in png_write_sRGB"); - - if (srgb_intent >= PNG_sRGB_INTENT_LAST) - png_warning(png_ptr, - "Invalid sRGB rendering intent specified"); - - buf[0]=(png_byte)srgb_intent; - png_write_complete_chunk(png_ptr, png_sRGB, buf, 1); -} -#endif - -#ifdef PNG_WRITE_iCCP_SUPPORTED -/* Write an iCCP chunk */ -void /* PRIVATE */ -png_write_iCCP(png_structrp png_ptr, png_const_charp name, - png_const_bytep profile) -{ - png_uint_32 name_len; - png_uint_32 profile_len; - png_byte new_name[81]; /* 1 byte for the compression byte */ - compression_state comp; - png_uint_32 temp; - - png_debug(1, "in png_write_iCCP"); - - /* These are all internal problems: the profile should have been checked - * before when it was stored. - */ - if (profile == NULL) - png_error(png_ptr, "No profile for iCCP chunk"); /* internal error */ - - profile_len = png_get_uint_32(profile); - - if (profile_len < 132) - png_error(png_ptr, "ICC profile too short"); - - temp = (png_uint_32) (*(profile+8)); - if (temp > 3 && (profile_len & 0x03)) - png_error(png_ptr, "ICC profile length invalid (not a multiple of 4)"); - - { - png_uint_32 embedded_profile_len = png_get_uint_32(profile); - - if (profile_len != embedded_profile_len) - png_error(png_ptr, "Profile length does not match profile"); - } - - name_len = png_check_keyword(png_ptr, name, new_name); - - if (name_len == 0) - png_error(png_ptr, "iCCP: invalid keyword"); - - new_name[++name_len] = PNG_COMPRESSION_TYPE_BASE; - - /* Make sure we include the NULL after the name and the compression type */ - ++name_len; - - png_text_compress_init(&comp, profile, profile_len); - - /* Allow for keyword terminator and compression byte */ - if (png_text_compress(png_ptr, png_iCCP, &comp, name_len) != Z_OK) - png_error(png_ptr, png_ptr->zstream.msg); - - png_write_chunk_header(png_ptr, png_iCCP, name_len + comp.output_len); - - png_write_chunk_data(png_ptr, new_name, name_len); - - png_write_compressed_data_out(png_ptr, &comp); - - png_write_chunk_end(png_ptr); -} -#endif - -#ifdef PNG_WRITE_sPLT_SUPPORTED -/* Write a sPLT chunk */ -void /* PRIVATE */ -png_write_sPLT(png_structrp png_ptr, png_const_sPLT_tp spalette) -{ - png_uint_32 name_len; - png_byte new_name[80]; - png_byte entrybuf[10]; - size_t entry_size = (spalette->depth == 8 ? 6 : 10); - size_t palette_size = entry_size * (size_t)spalette->nentries; - png_sPLT_entryp ep; -#ifndef PNG_POINTER_INDEXING_SUPPORTED - int i; -#endif - - png_debug(1, "in png_write_sPLT"); - - name_len = png_check_keyword(png_ptr, spalette->name, new_name); - - if (name_len == 0) - png_error(png_ptr, "sPLT: invalid keyword"); - - /* Make sure we include the NULL after the name */ - png_write_chunk_header(png_ptr, png_sPLT, - (png_uint_32)(name_len + 2 + palette_size)); - - png_write_chunk_data(png_ptr, (png_bytep)new_name, (size_t)(name_len + 1)); - - png_write_chunk_data(png_ptr, &spalette->depth, 1); - - /* Loop through each palette entry, writing appropriately */ -#ifdef PNG_POINTER_INDEXING_SUPPORTED - for (ep = spalette->entries; epentries + spalette->nentries; ep++) - { - if (spalette->depth == 8) - { - entrybuf[0] = (png_byte)ep->red; - entrybuf[1] = (png_byte)ep->green; - entrybuf[2] = (png_byte)ep->blue; - entrybuf[3] = (png_byte)ep->alpha; - png_save_uint_16(entrybuf + 4, ep->frequency); - } - - else - { - png_save_uint_16(entrybuf + 0, ep->red); - png_save_uint_16(entrybuf + 2, ep->green); - png_save_uint_16(entrybuf + 4, ep->blue); - png_save_uint_16(entrybuf + 6, ep->alpha); - png_save_uint_16(entrybuf + 8, ep->frequency); - } - - png_write_chunk_data(png_ptr, entrybuf, entry_size); - } -#else - ep=spalette->entries; - for (i = 0; i>spalette->nentries; i++) - { - if (spalette->depth == 8) - { - entrybuf[0] = (png_byte)ep[i].red; - entrybuf[1] = (png_byte)ep[i].green; - entrybuf[2] = (png_byte)ep[i].blue; - entrybuf[3] = (png_byte)ep[i].alpha; - png_save_uint_16(entrybuf + 4, ep[i].frequency); - } - - else - { - png_save_uint_16(entrybuf + 0, ep[i].red); - png_save_uint_16(entrybuf + 2, ep[i].green); - png_save_uint_16(entrybuf + 4, ep[i].blue); - png_save_uint_16(entrybuf + 6, ep[i].alpha); - png_save_uint_16(entrybuf + 8, ep[i].frequency); - } - - png_write_chunk_data(png_ptr, entrybuf, entry_size); - } -#endif - - png_write_chunk_end(png_ptr); -} -#endif - -#ifdef PNG_WRITE_sBIT_SUPPORTED -/* Write the sBIT chunk */ -void /* PRIVATE */ -png_write_sBIT(png_structrp png_ptr, png_const_color_8p sbit, int color_type) -{ - png_byte buf[4]; - size_t size; - - png_debug(1, "in png_write_sBIT"); - - /* Make sure we don't depend upon the order of PNG_COLOR_8 */ - if ((color_type & PNG_COLOR_MASK_COLOR) != 0) - { - png_byte maxbits; - - maxbits = (png_byte)(color_type==PNG_COLOR_TYPE_PALETTE ? 8 : - png_ptr->usr_bit_depth); - - if (sbit->red == 0 || sbit->red > maxbits || - sbit->green == 0 || sbit->green > maxbits || - sbit->blue == 0 || sbit->blue > maxbits) - { - png_warning(png_ptr, "Invalid sBIT depth specified"); - return; - } - - buf[0] = sbit->red; - buf[1] = sbit->green; - buf[2] = sbit->blue; - size = 3; - } - - else - { - if (sbit->gray == 0 || sbit->gray > png_ptr->usr_bit_depth) - { - png_warning(png_ptr, "Invalid sBIT depth specified"); - return; - } - - buf[0] = sbit->gray; - size = 1; - } - - if ((color_type & PNG_COLOR_MASK_ALPHA) != 0) - { - if (sbit->alpha == 0 || sbit->alpha > png_ptr->usr_bit_depth) - { - png_warning(png_ptr, "Invalid sBIT depth specified"); - return; - } - - buf[size++] = sbit->alpha; - } - - png_write_complete_chunk(png_ptr, png_sBIT, buf, size); -} -#endif - -#ifdef PNG_WRITE_cHRM_SUPPORTED -/* Write the cHRM chunk */ -void /* PRIVATE */ -png_write_cHRM_fixed(png_structrp png_ptr, const png_xy *xy) -{ - png_byte buf[32]; - - png_debug(1, "in png_write_cHRM"); - - /* Each value is saved in 1/100,000ths */ - png_save_int_32(buf, xy->whitex); - png_save_int_32(buf + 4, xy->whitey); - - png_save_int_32(buf + 8, xy->redx); - png_save_int_32(buf + 12, xy->redy); - - png_save_int_32(buf + 16, xy->greenx); - png_save_int_32(buf + 20, xy->greeny); - - png_save_int_32(buf + 24, xy->bluex); - png_save_int_32(buf + 28, xy->bluey); - - png_write_complete_chunk(png_ptr, png_cHRM, buf, 32); -} -#endif - -#ifdef PNG_WRITE_tRNS_SUPPORTED -/* Write the tRNS chunk */ -void /* PRIVATE */ -png_write_tRNS(png_structrp png_ptr, png_const_bytep trans_alpha, - png_const_color_16p tran, int num_trans, int color_type) -{ - png_byte buf[6]; - - png_debug(1, "in png_write_tRNS"); - - if (color_type == PNG_COLOR_TYPE_PALETTE) - { - if (num_trans <= 0 || num_trans > (int)png_ptr->num_palette) - { - png_app_warning(png_ptr, - "Invalid number of transparent colors specified"); - return; - } - - /* Write the chunk out as it is */ - png_write_complete_chunk(png_ptr, png_tRNS, trans_alpha, - (size_t)num_trans); - } - - else if (color_type == PNG_COLOR_TYPE_GRAY) - { - /* One 16-bit value */ - if (tran->gray >= (1 << png_ptr->bit_depth)) - { - png_app_warning(png_ptr, - "Ignoring attempt to write tRNS chunk out-of-range for bit_depth"); - - return; - } - - png_save_uint_16(buf, tran->gray); - png_write_complete_chunk(png_ptr, png_tRNS, buf, 2); - } - - else if (color_type == PNG_COLOR_TYPE_RGB) - { - /* Three 16-bit values */ - png_save_uint_16(buf, tran->red); - png_save_uint_16(buf + 2, tran->green); - png_save_uint_16(buf + 4, tran->blue); -#ifdef PNG_WRITE_16BIT_SUPPORTED - if (png_ptr->bit_depth == 8 && (buf[0] | buf[2] | buf[4]) != 0) -#else - if ((buf[0] | buf[2] | buf[4]) != 0) -#endif - { - png_app_warning(png_ptr, - "Ignoring attempt to write 16-bit tRNS chunk when bit_depth is 8"); - return; - } - - png_write_complete_chunk(png_ptr, png_tRNS, buf, 6); - } - - else - { - png_app_warning(png_ptr, "Can't write tRNS with an alpha channel"); - } -} -#endif - -#ifdef PNG_WRITE_bKGD_SUPPORTED -/* Write the background chunk */ -void /* PRIVATE */ -png_write_bKGD(png_structrp png_ptr, png_const_color_16p back, int color_type) -{ - png_byte buf[6]; - - png_debug(1, "in png_write_bKGD"); - - if (color_type == PNG_COLOR_TYPE_PALETTE) - { - if ( -#ifdef PNG_MNG_FEATURES_SUPPORTED - (png_ptr->num_palette != 0 || - (png_ptr->mng_features_permitted & PNG_FLAG_MNG_EMPTY_PLTE) == 0) && -#endif - back->index >= png_ptr->num_palette) - { - png_warning(png_ptr, "Invalid background palette index"); - return; - } - - buf[0] = back->index; - png_write_complete_chunk(png_ptr, png_bKGD, buf, 1); - } - - else if ((color_type & PNG_COLOR_MASK_COLOR) != 0) - { - png_save_uint_16(buf, back->red); - png_save_uint_16(buf + 2, back->green); - png_save_uint_16(buf + 4, back->blue); -#ifdef PNG_WRITE_16BIT_SUPPORTED - if (png_ptr->bit_depth == 8 && (buf[0] | buf[2] | buf[4]) != 0) -#else - if ((buf[0] | buf[2] | buf[4]) != 0) -#endif - { - png_warning(png_ptr, - "Ignoring attempt to write 16-bit bKGD chunk " - "when bit_depth is 8"); - - return; - } - - png_write_complete_chunk(png_ptr, png_bKGD, buf, 6); - } - - else - { - if (back->gray >= (1 << png_ptr->bit_depth)) - { - png_warning(png_ptr, - "Ignoring attempt to write bKGD chunk out-of-range for bit_depth"); - - return; - } - - png_save_uint_16(buf, back->gray); - png_write_complete_chunk(png_ptr, png_bKGD, buf, 2); - } -} -#endif - -#ifdef PNG_WRITE_eXIf_SUPPORTED -/* Write the Exif data */ -void /* PRIVATE */ -png_write_eXIf(png_structrp png_ptr, png_bytep exif, int num_exif) -{ - int i; - png_byte buf[1]; - - png_debug(1, "in png_write_eXIf"); - - png_write_chunk_header(png_ptr, png_eXIf, (png_uint_32)(num_exif)); - - for (i = 0; i < num_exif; i++) - { - buf[0] = exif[i]; - png_write_chunk_data(png_ptr, buf, 1); - } - - png_write_chunk_end(png_ptr); -} -#endif - -#ifdef PNG_WRITE_hIST_SUPPORTED -/* Write the histogram */ -void /* PRIVATE */ -png_write_hIST(png_structrp png_ptr, png_const_uint_16p hist, int num_hist) -{ - int i; - png_byte buf[3]; - - png_debug(1, "in png_write_hIST"); - - if (num_hist > (int)png_ptr->num_palette) - { - png_debug2(3, "num_hist = %d, num_palette = %d", num_hist, - png_ptr->num_palette); - - png_warning(png_ptr, "Invalid number of histogram entries specified"); - return; - } - - png_write_chunk_header(png_ptr, png_hIST, (png_uint_32)(num_hist * 2)); - - for (i = 0; i < num_hist; i++) - { - png_save_uint_16(buf, hist[i]); - png_write_chunk_data(png_ptr, buf, 2); - } - - png_write_chunk_end(png_ptr); -} -#endif - -#ifdef PNG_WRITE_tEXt_SUPPORTED -/* Write a tEXt chunk */ -void /* PRIVATE */ -png_write_tEXt(png_structrp png_ptr, png_const_charp key, png_const_charp text, - size_t text_len) -{ - png_uint_32 key_len; - png_byte new_key[80]; - - png_debug(1, "in png_write_tEXt"); - - key_len = png_check_keyword(png_ptr, key, new_key); - - if (key_len == 0) - png_error(png_ptr, "tEXt: invalid keyword"); - - if (text == NULL || *text == '\0') - text_len = 0; - - else - text_len = strlen(text); - - if (text_len > PNG_UINT_31_MAX - (key_len+1)) - png_error(png_ptr, "tEXt: text too long"); - - /* Make sure we include the 0 after the key */ - png_write_chunk_header(png_ptr, png_tEXt, - (png_uint_32)/*checked above*/(key_len + text_len + 1)); - /* - * We leave it to the application to meet PNG-1.0 requirements on the - * contents of the text. PNG-1.0 through PNG-1.2 discourage the use of - * any non-Latin-1 characters except for NEWLINE. ISO PNG will forbid them. - * The NUL character is forbidden by PNG-1.0 through PNG-1.2 and ISO PNG. - */ - png_write_chunk_data(png_ptr, new_key, key_len + 1); - - if (text_len != 0) - png_write_chunk_data(png_ptr, (png_const_bytep)text, text_len); - - png_write_chunk_end(png_ptr); -} -#endif - -#ifdef PNG_WRITE_zTXt_SUPPORTED -/* Write a compressed text chunk */ -void /* PRIVATE */ -png_write_zTXt(png_structrp png_ptr, png_const_charp key, png_const_charp text, - int compression) -{ - png_uint_32 key_len; - png_byte new_key[81]; - compression_state comp; - - png_debug(1, "in png_write_zTXt"); - - if (compression == PNG_TEXT_COMPRESSION_NONE) - { - png_write_tEXt(png_ptr, key, text, 0); - return; - } - - if (compression != PNG_TEXT_COMPRESSION_zTXt) - png_error(png_ptr, "zTXt: invalid compression type"); - - key_len = png_check_keyword(png_ptr, key, new_key); - - if (key_len == 0) - png_error(png_ptr, "zTXt: invalid keyword"); - - /* Add the compression method and 1 for the keyword separator. */ - new_key[++key_len] = PNG_COMPRESSION_TYPE_BASE; - ++key_len; - - /* Compute the compressed data; do it now for the length */ - png_text_compress_init(&comp, (png_const_bytep)text, - text == NULL ? 0 : strlen(text)); - - if (png_text_compress(png_ptr, png_zTXt, &comp, key_len) != Z_OK) - png_error(png_ptr, png_ptr->zstream.msg); - - /* Write start of chunk */ - png_write_chunk_header(png_ptr, png_zTXt, key_len + comp.output_len); - - /* Write key */ - png_write_chunk_data(png_ptr, new_key, key_len); - - /* Write the compressed data */ - png_write_compressed_data_out(png_ptr, &comp); - - /* Close the chunk */ - png_write_chunk_end(png_ptr); -} -#endif - -#ifdef PNG_WRITE_iTXt_SUPPORTED -/* Write an iTXt chunk */ -void /* PRIVATE */ -png_write_iTXt(png_structrp png_ptr, int compression, png_const_charp key, - png_const_charp lang, png_const_charp lang_key, png_const_charp text) -{ - png_uint_32 key_len, prefix_len; - size_t lang_len, lang_key_len; - png_byte new_key[82]; - compression_state comp; - - png_debug(1, "in png_write_iTXt"); - - key_len = png_check_keyword(png_ptr, key, new_key); - - if (key_len == 0) - png_error(png_ptr, "iTXt: invalid keyword"); - - /* Set the compression flag */ - switch (compression) - { - case PNG_ITXT_COMPRESSION_NONE: - case PNG_TEXT_COMPRESSION_NONE: - compression = new_key[++key_len] = 0; /* no compression */ - break; - - case PNG_TEXT_COMPRESSION_zTXt: - case PNG_ITXT_COMPRESSION_zTXt: - compression = new_key[++key_len] = 1; /* compressed */ - break; - - default: - png_error(png_ptr, "iTXt: invalid compression"); - } - - new_key[++key_len] = PNG_COMPRESSION_TYPE_BASE; - ++key_len; /* for the keywod separator */ - - /* We leave it to the application to meet PNG-1.0 requirements on the - * contents of the text. PNG-1.0 through PNG-1.2 discourage the use of - * any non-Latin-1 characters except for NEWLINE. ISO PNG, however, - * specifies that the text is UTF-8 and this really doesn't require any - * checking. - * - * The NUL character is forbidden by PNG-1.0 through PNG-1.2 and ISO PNG. - * - * TODO: validate the language tag correctly (see the spec.) - */ - if (lang == NULL) lang = ""; /* empty language is valid */ - lang_len = strlen(lang)+1; - if (lang_key == NULL) lang_key = ""; /* may be empty */ - lang_key_len = strlen(lang_key)+1; - if (text == NULL) text = ""; /* may be empty */ - - prefix_len = key_len; - if (lang_len > PNG_UINT_31_MAX-prefix_len) - prefix_len = PNG_UINT_31_MAX; - else - prefix_len = (png_uint_32)(prefix_len + lang_len); - - if (lang_key_len > PNG_UINT_31_MAX-prefix_len) - prefix_len = PNG_UINT_31_MAX; - else - prefix_len = (png_uint_32)(prefix_len + lang_key_len); - - png_text_compress_init(&comp, (png_const_bytep)text, strlen(text)); - - if (compression != 0) - { - if (png_text_compress(png_ptr, png_iTXt, &comp, prefix_len) != Z_OK) - png_error(png_ptr, png_ptr->zstream.msg); - } - - else - { - if (comp.input_len > PNG_UINT_31_MAX-prefix_len) - png_error(png_ptr, "iTXt: uncompressed text too long"); - - /* So the string will fit in a chunk: */ - comp.output_len = (png_uint_32)/*SAFE*/comp.input_len; - } - - png_write_chunk_header(png_ptr, png_iTXt, comp.output_len + prefix_len); - - png_write_chunk_data(png_ptr, new_key, key_len); - - png_write_chunk_data(png_ptr, (png_const_bytep)lang, lang_len); - - png_write_chunk_data(png_ptr, (png_const_bytep)lang_key, lang_key_len); - - if (compression != 0) - png_write_compressed_data_out(png_ptr, &comp); - - else - png_write_chunk_data(png_ptr, (png_const_bytep)text, comp.output_len); - - png_write_chunk_end(png_ptr); -} -#endif - -#ifdef PNG_WRITE_oFFs_SUPPORTED -/* Write the oFFs chunk */ -void /* PRIVATE */ -png_write_oFFs(png_structrp png_ptr, png_int_32 x_offset, png_int_32 y_offset, - int unit_type) -{ - png_byte buf[9]; - - png_debug(1, "in png_write_oFFs"); - - if (unit_type >= PNG_OFFSET_LAST) - png_warning(png_ptr, "Unrecognized unit type for oFFs chunk"); - - png_save_int_32(buf, x_offset); - png_save_int_32(buf + 4, y_offset); - buf[8] = (png_byte)unit_type; - - png_write_complete_chunk(png_ptr, png_oFFs, buf, 9); -} -#endif -#ifdef PNG_WRITE_pCAL_SUPPORTED -/* Write the pCAL chunk (described in the PNG extensions document) */ -void /* PRIVATE */ -png_write_pCAL(png_structrp png_ptr, png_charp purpose, png_int_32 X0, - png_int_32 X1, int type, int nparams, png_const_charp units, - png_charpp params) -{ - png_uint_32 purpose_len; - size_t units_len, total_len; - size_t *params_len; - png_byte buf[10]; - png_byte new_purpose[80]; - int i; - - png_debug1(1, "in png_write_pCAL (%d parameters)", nparams); - - if (type >= PNG_EQUATION_LAST) - png_error(png_ptr, "Unrecognized equation type for pCAL chunk"); - - purpose_len = png_check_keyword(png_ptr, purpose, new_purpose); - - if (purpose_len == 0) - png_error(png_ptr, "pCAL: invalid keyword"); - - ++purpose_len; /* terminator */ - - png_debug1(3, "pCAL purpose length = %d", (int)purpose_len); - units_len = strlen(units) + (nparams == 0 ? 0 : 1); - png_debug1(3, "pCAL units length = %d", (int)units_len); - total_len = purpose_len + units_len + 10; - - params_len = (size_t *)png_malloc(png_ptr, - (png_alloc_size_t)((png_alloc_size_t)nparams * (sizeof (size_t)))); - - /* Find the length of each parameter, making sure we don't count the - * null terminator for the last parameter. - */ - for (i = 0; i < nparams; i++) - { - params_len[i] = strlen(params[i]) + (i == nparams - 1 ? 0 : 1); - png_debug2(3, "pCAL parameter %d length = %lu", i, - (unsigned long)params_len[i]); - total_len += params_len[i]; - } - - png_debug1(3, "pCAL total length = %d", (int)total_len); - png_write_chunk_header(png_ptr, png_pCAL, (png_uint_32)total_len); - png_write_chunk_data(png_ptr, new_purpose, purpose_len); - png_save_int_32(buf, X0); - png_save_int_32(buf + 4, X1); - buf[8] = (png_byte)type; - buf[9] = (png_byte)nparams; - png_write_chunk_data(png_ptr, buf, 10); - png_write_chunk_data(png_ptr, (png_const_bytep)units, (size_t)units_len); - - for (i = 0; i < nparams; i++) - { - png_write_chunk_data(png_ptr, (png_const_bytep)params[i], params_len[i]); - } - - png_free(png_ptr, params_len); - png_write_chunk_end(png_ptr); -} -#endif - -#ifdef PNG_WRITE_sCAL_SUPPORTED -/* Write the sCAL chunk */ -void /* PRIVATE */ -png_write_sCAL_s(png_structrp png_ptr, int unit, png_const_charp width, - png_const_charp height) -{ - png_byte buf[64]; - size_t wlen, hlen, total_len; - - png_debug(1, "in png_write_sCAL_s"); - - wlen = strlen(width); - hlen = strlen(height); - total_len = wlen + hlen + 2; - - if (total_len > 64) - { - png_warning(png_ptr, "Can't write sCAL (buffer too small)"); - return; - } - - buf[0] = (png_byte)unit; - memcpy(buf + 1, width, wlen + 1); /* Append the '\0' here */ - memcpy(buf + wlen + 2, height, hlen); /* Do NOT append the '\0' here */ - - png_debug1(3, "sCAL total length = %u", (unsigned int)total_len); - png_write_complete_chunk(png_ptr, png_sCAL, buf, total_len); -} -#endif - -#ifdef PNG_WRITE_pHYs_SUPPORTED -/* Write the pHYs chunk */ -void /* PRIVATE */ -png_write_pHYs(png_structrp png_ptr, png_uint_32 x_pixels_per_unit, - png_uint_32 y_pixels_per_unit, - int unit_type) -{ - png_byte buf[9]; - - png_debug(1, "in png_write_pHYs"); - - if (unit_type >= PNG_RESOLUTION_LAST) - png_warning(png_ptr, "Unrecognized unit type for pHYs chunk"); - - png_save_uint_32(buf, x_pixels_per_unit); - png_save_uint_32(buf + 4, y_pixels_per_unit); - buf[8] = (png_byte)unit_type; - - png_write_complete_chunk(png_ptr, png_pHYs, buf, 9); -} -#endif - -#ifdef PNG_WRITE_tIME_SUPPORTED -/* Write the tIME chunk. Use either png_convert_from_struct_tm() - * or png_convert_from_time_t(), or fill in the structure yourself. - */ -void /* PRIVATE */ -png_write_tIME(png_structrp png_ptr, png_const_timep mod_time) -{ - png_byte buf[7]; - - png_debug(1, "in png_write_tIME"); - - if (mod_time->month > 12 || mod_time->month < 1 || - mod_time->day > 31 || mod_time->day < 1 || - mod_time->hour > 23 || mod_time->second > 60) - { - png_warning(png_ptr, "Invalid time specified for tIME chunk"); - return; - } - - png_save_uint_16(buf, mod_time->year); - buf[2] = mod_time->month; - buf[3] = mod_time->day; - buf[4] = mod_time->hour; - buf[5] = mod_time->minute; - buf[6] = mod_time->second; - - png_write_complete_chunk(png_ptr, png_tIME, buf, 7); -} -#endif - -/* Initializes the row writing capability of libpng */ -void /* PRIVATE */ -png_write_start_row(png_structrp png_ptr) -{ -#ifdef PNG_WRITE_INTERLACING_SUPPORTED - /* Arrays to facilitate easy interlacing - use pass (0 - 6) as index */ - - /* Start of interlace block */ - static const png_byte png_pass_start[7] = {0, 4, 0, 2, 0, 1, 0}; - - /* Offset to next interlace block */ - static const png_byte png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; - - /* Start of interlace block in the y direction */ - static const png_byte png_pass_ystart[7] = {0, 0, 4, 0, 2, 0, 1}; - - /* Offset to next interlace block in the y direction */ - static const png_byte png_pass_yinc[7] = {8, 8, 8, 4, 4, 2, 2}; -#endif - - png_alloc_size_t buf_size; - int usr_pixel_depth; - -#ifdef PNG_WRITE_FILTER_SUPPORTED - png_byte filters; -#endif - - png_debug(1, "in png_write_start_row"); - - usr_pixel_depth = png_ptr->usr_channels * png_ptr->usr_bit_depth; - buf_size = PNG_ROWBYTES(usr_pixel_depth, png_ptr->width) + 1; - - /* 1.5.6: added to allow checking in the row write code. */ - png_ptr->transformed_pixel_depth = png_ptr->pixel_depth; - png_ptr->maximum_pixel_depth = (png_byte)usr_pixel_depth; - - /* Set up row buffer */ - png_ptr->row_buf = png_voidcast(png_bytep, png_malloc(png_ptr, buf_size)); - - png_ptr->row_buf[0] = PNG_FILTER_VALUE_NONE; - -#ifdef PNG_WRITE_FILTER_SUPPORTED - filters = png_ptr->do_filter; - - if (png_ptr->height == 1) - filters &= 0xff & ~(PNG_FILTER_UP|PNG_FILTER_AVG|PNG_FILTER_PAETH); - - if (png_ptr->width == 1) - filters &= 0xff & ~(PNG_FILTER_SUB|PNG_FILTER_AVG|PNG_FILTER_PAETH); - - if (filters == 0) - filters = PNG_FILTER_NONE; - - png_ptr->do_filter = filters; - - if (((filters & (PNG_FILTER_SUB | PNG_FILTER_UP | PNG_FILTER_AVG | - PNG_FILTER_PAETH)) != 0) && png_ptr->try_row == NULL) - { - int num_filters = 0; - - png_ptr->try_row = png_voidcast(png_bytep, png_malloc(png_ptr, buf_size)); - - if (filters & PNG_FILTER_SUB) - num_filters++; - - if (filters & PNG_FILTER_UP) - num_filters++; - - if (filters & PNG_FILTER_AVG) - num_filters++; - - if (filters & PNG_FILTER_PAETH) - num_filters++; - - if (num_filters > 1) - png_ptr->tst_row = png_voidcast(png_bytep, png_malloc(png_ptr, - buf_size)); - } - - /* We only need to keep the previous row if we are using one of the following - * filters. - */ - if ((filters & (PNG_FILTER_AVG | PNG_FILTER_UP | PNG_FILTER_PAETH)) != 0) - png_ptr->prev_row = png_voidcast(png_bytep, - png_calloc(png_ptr, buf_size)); -#endif /* WRITE_FILTER */ - -#ifdef PNG_WRITE_INTERLACING_SUPPORTED - /* If interlaced, we need to set up width and height of pass */ - if (png_ptr->interlaced != 0) - { - if ((png_ptr->transformations & PNG_INTERLACE) == 0) - { - png_ptr->num_rows = (png_ptr->height + png_pass_yinc[0] - 1 - - png_pass_ystart[0]) / png_pass_yinc[0]; - - png_ptr->usr_width = (png_ptr->width + png_pass_inc[0] - 1 - - png_pass_start[0]) / png_pass_inc[0]; - } - - else - { - png_ptr->num_rows = png_ptr->height; - png_ptr->usr_width = png_ptr->width; - } - } - - else -#endif - { - png_ptr->num_rows = png_ptr->height; - png_ptr->usr_width = png_ptr->width; - } -} - -/* Internal use only. Called when finished processing a row of data. */ -void /* PRIVATE */ -png_write_finish_row(png_structrp png_ptr) -{ -#ifdef PNG_WRITE_INTERLACING_SUPPORTED - /* Arrays to facilitate easy interlacing - use pass (0 - 6) as index */ - - /* Start of interlace block */ - static const png_byte png_pass_start[7] = {0, 4, 0, 2, 0, 1, 0}; - - /* Offset to next interlace block */ - static const png_byte png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; - - /* Start of interlace block in the y direction */ - static const png_byte png_pass_ystart[7] = {0, 0, 4, 0, 2, 0, 1}; - - /* Offset to next interlace block in the y direction */ - static const png_byte png_pass_yinc[7] = {8, 8, 8, 4, 4, 2, 2}; -#endif - - png_debug(1, "in png_write_finish_row"); - - /* Next row */ - png_ptr->row_number++; - - /* See if we are done */ - if (png_ptr->row_number < png_ptr->num_rows) - return; - -#ifdef PNG_WRITE_INTERLACING_SUPPORTED - /* If interlaced, go to next pass */ - if (png_ptr->interlaced != 0) - { - png_ptr->row_number = 0; - if ((png_ptr->transformations & PNG_INTERLACE) != 0) - { - png_ptr->pass++; - } - - else - { - /* Loop until we find a non-zero width or height pass */ - do - { - png_ptr->pass++; - - if (png_ptr->pass >= 7) - break; - - png_ptr->usr_width = (png_ptr->width + - png_pass_inc[png_ptr->pass] - 1 - - png_pass_start[png_ptr->pass]) / - png_pass_inc[png_ptr->pass]; - - png_ptr->num_rows = (png_ptr->height + - png_pass_yinc[png_ptr->pass] - 1 - - png_pass_ystart[png_ptr->pass]) / - png_pass_yinc[png_ptr->pass]; - - if ((png_ptr->transformations & PNG_INTERLACE) != 0) - break; - - } while (png_ptr->usr_width == 0 || png_ptr->num_rows == 0); - - } - - /* Reset the row above the image for the next pass */ - if (png_ptr->pass < 7) - { - if (png_ptr->prev_row != NULL) - memset(png_ptr->prev_row, 0, - PNG_ROWBYTES(png_ptr->usr_channels * - png_ptr->usr_bit_depth, png_ptr->width) + 1); - - return; - } - } -#endif - - /* If we get here, we've just written the last row, so we need - to flush the compressor */ - png_compress_IDAT(png_ptr, NULL, 0, Z_FINISH); -} - -#ifdef PNG_WRITE_INTERLACING_SUPPORTED -/* Pick out the correct pixels for the interlace pass. - * The basic idea here is to go through the row with a source - * pointer and a destination pointer (sp and dp), and copy the - * correct pixels for the pass. As the row gets compacted, - * sp will always be >= dp, so we should never overwrite anything. - * See the default: case for the easiest code to understand. - */ -void /* PRIVATE */ -png_do_write_interlace(png_row_infop row_info, png_bytep row, int pass) -{ - /* Arrays to facilitate easy interlacing - use pass (0 - 6) as index */ - - /* Start of interlace block */ - static const png_byte png_pass_start[7] = {0, 4, 0, 2, 0, 1, 0}; - - /* Offset to next interlace block */ - static const png_byte png_pass_inc[7] = {8, 8, 4, 4, 2, 2, 1}; - - png_debug(1, "in png_do_write_interlace"); - - /* We don't have to do anything on the last pass (6) */ - if (pass < 6) - { - /* Each pixel depth is handled separately */ - switch (row_info->pixel_depth) - { - case 1: - { - png_bytep sp; - png_bytep dp; - unsigned int shift; - int d; - int value; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - dp = row; - d = 0; - shift = 7; - - for (i = png_pass_start[pass]; i < row_width; - i += png_pass_inc[pass]) - { - sp = row + (size_t)(i >> 3); - value = (int)(*sp >> (7 - (int)(i & 0x07))) & 0x01; - d |= (value << shift); - - if (shift == 0) - { - shift = 7; - *dp++ = (png_byte)d; - d = 0; - } - - else - shift--; - - } - if (shift != 7) - *dp = (png_byte)d; - - break; - } - - case 2: - { - png_bytep sp; - png_bytep dp; - unsigned int shift; - int d; - int value; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - dp = row; - shift = 6; - d = 0; - - for (i = png_pass_start[pass]; i < row_width; - i += png_pass_inc[pass]) - { - sp = row + (size_t)(i >> 2); - value = (*sp >> ((3 - (int)(i & 0x03)) << 1)) & 0x03; - d |= (value << shift); - - if (shift == 0) - { - shift = 6; - *dp++ = (png_byte)d; - d = 0; - } - - else - shift -= 2; - } - if (shift != 6) - *dp = (png_byte)d; - - break; - } - - case 4: - { - png_bytep sp; - png_bytep dp; - unsigned int shift; - int d; - int value; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - - dp = row; - shift = 4; - d = 0; - for (i = png_pass_start[pass]; i < row_width; - i += png_pass_inc[pass]) - { - sp = row + (size_t)(i >> 1); - value = (*sp >> ((1 - (int)(i & 0x01)) << 2)) & 0x0f; - d |= (value << shift); - - if (shift == 0) - { - shift = 4; - *dp++ = (png_byte)d; - d = 0; - } - - else - shift -= 4; - } - if (shift != 4) - *dp = (png_byte)d; - - break; - } - - default: - { - png_bytep sp; - png_bytep dp; - png_uint_32 i; - png_uint_32 row_width = row_info->width; - size_t pixel_bytes; - - /* Start at the beginning */ - dp = row; - - /* Find out how many bytes each pixel takes up */ - pixel_bytes = (row_info->pixel_depth >> 3); - - /* Loop through the row, only looking at the pixels that matter */ - for (i = png_pass_start[pass]; i < row_width; - i += png_pass_inc[pass]) - { - /* Find out where the original pixel is */ - sp = row + (size_t)i * pixel_bytes; - - /* Move the pixel */ - if (dp != sp) - memcpy(dp, sp, pixel_bytes); - - /* Next pixel */ - dp += pixel_bytes; - } - break; - } - } - /* Set new row width */ - row_info->width = (row_info->width + - png_pass_inc[pass] - 1 - - png_pass_start[pass]) / - png_pass_inc[pass]; - - row_info->rowbytes = PNG_ROWBYTES(row_info->pixel_depth, - row_info->width); - } -} -#endif - - -/* This filters the row, chooses which filter to use, if it has not already - * been specified by the application, and then writes the row out with the - * chosen filter. - */ -static void /* PRIVATE */ -png_write_filtered_row(png_structrp png_ptr, png_bytep filtered_row, - size_t row_bytes); - -#ifdef PNG_WRITE_FILTER_SUPPORTED -static size_t /* PRIVATE */ -png_setup_sub_row(png_structrp png_ptr, png_uint_32 bpp, - size_t row_bytes, size_t lmins) -{ - png_bytep rp, dp, lp; - size_t i; - size_t sum = 0; - unsigned int v; - - png_ptr->try_row[0] = PNG_FILTER_VALUE_SUB; - - for (i = 0, rp = png_ptr->row_buf + 1, dp = png_ptr->try_row + 1; i < bpp; - i++, rp++, dp++) - { - v = *dp = *rp; -#ifdef PNG_USE_ABS - sum += 128 - abs((int)v - 128); -#else - sum += (v < 128) ? v : 256 - v; -#endif - } - - for (lp = png_ptr->row_buf + 1; i < row_bytes; - i++, rp++, lp++, dp++) - { - v = *dp = (png_byte)(((int)*rp - (int)*lp) & 0xff); -#ifdef PNG_USE_ABS - sum += 128 - abs((int)v - 128); -#else - sum += (v < 128) ? v : 256 - v; -#endif - - if (sum > lmins) /* We are already worse, don't continue. */ - break; - } - - return sum; -} - -static void /* PRIVATE */ -png_setup_sub_row_only(png_structrp png_ptr, png_uint_32 bpp, - size_t row_bytes) -{ - png_bytep rp, dp, lp; - size_t i; - - png_ptr->try_row[0] = PNG_FILTER_VALUE_SUB; - - for (i = 0, rp = png_ptr->row_buf + 1, dp = png_ptr->try_row + 1; i < bpp; - i++, rp++, dp++) - { - *dp = *rp; - } - - for (lp = png_ptr->row_buf + 1; i < row_bytes; - i++, rp++, lp++, dp++) - { - *dp = (png_byte)(((int)*rp - (int)*lp) & 0xff); - } -} - -static size_t /* PRIVATE */ -png_setup_up_row(png_structrp png_ptr, size_t row_bytes, size_t lmins) -{ - png_bytep rp, dp, pp; - size_t i; - size_t sum = 0; - unsigned int v; - - png_ptr->try_row[0] = PNG_FILTER_VALUE_UP; - - for (i = 0, rp = png_ptr->row_buf + 1, dp = png_ptr->try_row + 1, - pp = png_ptr->prev_row + 1; i < row_bytes; - i++, rp++, pp++, dp++) - { - v = *dp = (png_byte)(((int)*rp - (int)*pp) & 0xff); -#ifdef PNG_USE_ABS - sum += 128 - abs((int)v - 128); -#else - sum += (v < 128) ? v : 256 - v; -#endif - - if (sum > lmins) /* We are already worse, don't continue. */ - break; - } - - return sum; -} -static void /* PRIVATE */ -png_setup_up_row_only(png_structrp png_ptr, size_t row_bytes) -{ - png_bytep rp, dp, pp; - size_t i; - - png_ptr->try_row[0] = PNG_FILTER_VALUE_UP; - - for (i = 0, rp = png_ptr->row_buf + 1, dp = png_ptr->try_row + 1, - pp = png_ptr->prev_row + 1; i < row_bytes; - i++, rp++, pp++, dp++) - { - *dp = (png_byte)(((int)*rp - (int)*pp) & 0xff); - } -} - -static size_t /* PRIVATE */ -png_setup_avg_row(png_structrp png_ptr, png_uint_32 bpp, - size_t row_bytes, size_t lmins) -{ - png_bytep rp, dp, pp, lp; - png_uint_32 i; - size_t sum = 0; - unsigned int v; - - png_ptr->try_row[0] = PNG_FILTER_VALUE_AVG; - - for (i = 0, rp = png_ptr->row_buf + 1, dp = png_ptr->try_row + 1, - pp = png_ptr->prev_row + 1; i < bpp; i++) - { - v = *dp++ = (png_byte)(((int)*rp++ - ((int)*pp++ / 2)) & 0xff); - -#ifdef PNG_USE_ABS - sum += 128 - abs((int)v - 128); -#else - sum += (v < 128) ? v : 256 - v; -#endif - } - - for (lp = png_ptr->row_buf + 1; i < row_bytes; i++) - { - v = *dp++ = (png_byte)(((int)*rp++ - (((int)*pp++ + (int)*lp++) / 2)) - & 0xff); - -#ifdef PNG_USE_ABS - sum += 128 - abs((int)v - 128); -#else - sum += (v < 128) ? v : 256 - v; -#endif - - if (sum > lmins) /* We are already worse, don't continue. */ - break; - } - - return sum; -} -static void /* PRIVATE */ -png_setup_avg_row_only(png_structrp png_ptr, png_uint_32 bpp, - size_t row_bytes) -{ - png_bytep rp, dp, pp, lp; - png_uint_32 i; - - png_ptr->try_row[0] = PNG_FILTER_VALUE_AVG; - - for (i = 0, rp = png_ptr->row_buf + 1, dp = png_ptr->try_row + 1, - pp = png_ptr->prev_row + 1; i < bpp; i++) - { - *dp++ = (png_byte)(((int)*rp++ - ((int)*pp++ / 2)) & 0xff); - } - - for (lp = png_ptr->row_buf + 1; i < row_bytes; i++) - { - *dp++ = (png_byte)(((int)*rp++ - (((int)*pp++ + (int)*lp++) / 2)) - & 0xff); - } -} - -static size_t /* PRIVATE */ -png_setup_paeth_row(png_structrp png_ptr, png_uint_32 bpp, - size_t row_bytes, size_t lmins) -{ - png_bytep rp, dp, pp, cp, lp; - size_t i; - size_t sum = 0; - unsigned int v; - - png_ptr->try_row[0] = PNG_FILTER_VALUE_PAETH; - - for (i = 0, rp = png_ptr->row_buf + 1, dp = png_ptr->try_row + 1, - pp = png_ptr->prev_row + 1; i < bpp; i++) - { - v = *dp++ = (png_byte)(((int)*rp++ - (int)*pp++) & 0xff); - -#ifdef PNG_USE_ABS - sum += 128 - abs((int)v - 128); -#else - sum += (v < 128) ? v : 256 - v; -#endif - } - - for (lp = png_ptr->row_buf + 1, cp = png_ptr->prev_row + 1; i < row_bytes; - i++) - { - int a, b, c, pa, pb, pc, p; - - b = *pp++; - c = *cp++; - a = *lp++; - - p = b - c; - pc = a - c; - -#ifdef PNG_USE_ABS - pa = abs(p); - pb = abs(pc); - pc = abs(p + pc); -#else - pa = p < 0 ? -p : p; - pb = pc < 0 ? -pc : pc; - pc = (p + pc) < 0 ? -(p + pc) : p + pc; -#endif - - p = (pa <= pb && pa <=pc) ? a : (pb <= pc) ? b : c; - - v = *dp++ = (png_byte)(((int)*rp++ - p) & 0xff); - -#ifdef PNG_USE_ABS - sum += 128 - abs((int)v - 128); -#else - sum += (v < 128) ? v : 256 - v; -#endif - - if (sum > lmins) /* We are already worse, don't continue. */ - break; - } - - return sum; -} -static void /* PRIVATE */ -png_setup_paeth_row_only(png_structrp png_ptr, png_uint_32 bpp, - size_t row_bytes) -{ - png_bytep rp, dp, pp, cp, lp; - size_t i; - - png_ptr->try_row[0] = PNG_FILTER_VALUE_PAETH; - - for (i = 0, rp = png_ptr->row_buf + 1, dp = png_ptr->try_row + 1, - pp = png_ptr->prev_row + 1; i < bpp; i++) - { - *dp++ = (png_byte)(((int)*rp++ - (int)*pp++) & 0xff); - } - - for (lp = png_ptr->row_buf + 1, cp = png_ptr->prev_row + 1; i < row_bytes; - i++) - { - int a, b, c, pa, pb, pc, p; - - b = *pp++; - c = *cp++; - a = *lp++; - - p = b - c; - pc = a - c; - -#ifdef PNG_USE_ABS - pa = abs(p); - pb = abs(pc); - pc = abs(p + pc); -#else - pa = p < 0 ? -p : p; - pb = pc < 0 ? -pc : pc; - pc = (p + pc) < 0 ? -(p + pc) : p + pc; -#endif - - p = (pa <= pb && pa <=pc) ? a : (pb <= pc) ? b : c; - - *dp++ = (png_byte)(((int)*rp++ - p) & 0xff); - } -} -#endif /* WRITE_FILTER */ - -void /* PRIVATE */ -png_write_find_filter(png_structrp png_ptr, png_row_infop row_info) -{ -#ifndef PNG_WRITE_FILTER_SUPPORTED - png_write_filtered_row(png_ptr, png_ptr->row_buf, row_info->rowbytes+1); -#else - unsigned int filter_to_do = png_ptr->do_filter; - png_bytep row_buf; - png_bytep best_row; - png_uint_32 bpp; - size_t mins; - size_t row_bytes = row_info->rowbytes; - - png_debug(1, "in png_write_find_filter"); - - /* Find out how many bytes offset each pixel is */ - bpp = (row_info->pixel_depth + 7) >> 3; - - row_buf = png_ptr->row_buf; - mins = PNG_SIZE_MAX - 256/* so we can detect potential overflow of the - running sum */; - - /* The prediction method we use is to find which method provides the - * smallest value when summing the absolute values of the distances - * from zero, using anything >= 128 as negative numbers. This is known - * as the "minimum sum of absolute differences" heuristic. Other - * heuristics are the "weighted minimum sum of absolute differences" - * (experimental and can in theory improve compression), and the "zlib - * predictive" method (not implemented yet), which does test compressions - * of lines using different filter methods, and then chooses the - * (series of) filter(s) that give minimum compressed data size (VERY - * computationally expensive). - * - * GRR 980525: consider also - * - * (1) minimum sum of absolute differences from running average (i.e., - * keep running sum of non-absolute differences & count of bytes) - * [track dispersion, too? restart average if dispersion too large?] - * - * (1b) minimum sum of absolute differences from sliding average, probably - * with window size <= deflate window (usually 32K) - * - * (2) minimum sum of squared differences from zero or running average - * (i.e., ~ root-mean-square approach) - */ - - - /* We don't need to test the 'no filter' case if this is the only filter - * that has been chosen, as it doesn't actually do anything to the data. - */ - best_row = png_ptr->row_buf; - - if (PNG_SIZE_MAX/128 <= row_bytes) - { - /* Overflow can occur in the calculation, just select the lowest set - * filter. - */ - filter_to_do &= 0U-filter_to_do; - } - else if ((filter_to_do & PNG_FILTER_NONE) != 0 && - filter_to_do != PNG_FILTER_NONE) - { - /* Overflow not possible and multiple filters in the list, including the - * 'none' filter. - */ - png_bytep rp; - size_t sum = 0; - size_t i; - unsigned int v; - - { - for (i = 0, rp = row_buf + 1; i < row_bytes; i++, rp++) - { - v = *rp; -#ifdef PNG_USE_ABS - sum += 128 - abs((int)v - 128); -#else - sum += (v < 128) ? v : 256 - v; -#endif - } - } - - mins = sum; - } - - /* Sub filter */ - if (filter_to_do == PNG_FILTER_SUB) - /* It's the only filter so no testing is needed */ - { - png_setup_sub_row_only(png_ptr, bpp, row_bytes); - best_row = png_ptr->try_row; - } - - else if ((filter_to_do & PNG_FILTER_SUB) != 0) - { - size_t sum; - size_t lmins = mins; - - sum = png_setup_sub_row(png_ptr, bpp, row_bytes, lmins); - - if (sum < mins) - { - mins = sum; - best_row = png_ptr->try_row; - if (png_ptr->tst_row != NULL) - { - png_ptr->try_row = png_ptr->tst_row; - png_ptr->tst_row = best_row; - } - } - } - - /* Up filter */ - if (filter_to_do == PNG_FILTER_UP) - { - png_setup_up_row_only(png_ptr, row_bytes); - best_row = png_ptr->try_row; - } - - else if ((filter_to_do & PNG_FILTER_UP) != 0) - { - size_t sum; - size_t lmins = mins; - - sum = png_setup_up_row(png_ptr, row_bytes, lmins); - - if (sum < mins) - { - mins = sum; - best_row = png_ptr->try_row; - if (png_ptr->tst_row != NULL) - { - png_ptr->try_row = png_ptr->tst_row; - png_ptr->tst_row = best_row; - } - } - } - - /* Avg filter */ - if (filter_to_do == PNG_FILTER_AVG) - { - png_setup_avg_row_only(png_ptr, bpp, row_bytes); - best_row = png_ptr->try_row; - } - - else if ((filter_to_do & PNG_FILTER_AVG) != 0) - { - size_t sum; - size_t lmins = mins; - - sum= png_setup_avg_row(png_ptr, bpp, row_bytes, lmins); - - if (sum < mins) - { - mins = sum; - best_row = png_ptr->try_row; - if (png_ptr->tst_row != NULL) - { - png_ptr->try_row = png_ptr->tst_row; - png_ptr->tst_row = best_row; - } - } - } - - /* Paeth filter */ - if (filter_to_do == PNG_FILTER_PAETH) - { - png_setup_paeth_row_only(png_ptr, bpp, row_bytes); - best_row = png_ptr->try_row; - } - - else if ((filter_to_do & PNG_FILTER_PAETH) != 0) - { - size_t sum; - size_t lmins = mins; - - sum = png_setup_paeth_row(png_ptr, bpp, row_bytes, lmins); - - if (sum < mins) - { - best_row = png_ptr->try_row; - if (png_ptr->tst_row != NULL) - { - png_ptr->try_row = png_ptr->tst_row; - png_ptr->tst_row = best_row; - } - } - } - - /* Do the actual writing of the filtered row data from the chosen filter. */ - png_write_filtered_row(png_ptr, best_row, row_info->rowbytes+1); - -#endif /* WRITE_FILTER */ -} - - -/* Do the actual writing of a previously filtered row. */ -static void -png_write_filtered_row(png_structrp png_ptr, png_bytep filtered_row, - size_t full_row_length/*includes filter byte*/) -{ - png_debug(1, "in png_write_filtered_row"); - - png_debug1(2, "filter = %d", filtered_row[0]); - - png_compress_IDAT(png_ptr, filtered_row, full_row_length, Z_NO_FLUSH); - -#ifdef PNG_WRITE_FILTER_SUPPORTED - /* Swap the current and previous rows */ - if (png_ptr->prev_row != NULL) - { - png_bytep tptr; - - tptr = png_ptr->prev_row; - png_ptr->prev_row = png_ptr->row_buf; - png_ptr->row_buf = tptr; - } -#endif /* WRITE_FILTER */ - - /* Finish row - updates counters and flushes zlib if last row */ - png_write_finish_row(png_ptr); - -#ifdef PNG_WRITE_FLUSH_SUPPORTED - png_ptr->flush_rows++; - - if (png_ptr->flush_dist > 0 && - png_ptr->flush_rows >= png_ptr->flush_dist) - { - png_write_flush(png_ptr); - } -#endif /* WRITE_FLUSH */ -} -#endif /* WRITE */ diff --git a/dep/zlib/CMakeLists.txt b/dep/zlib/CMakeLists.txt deleted file mode 100644 index 10f512ccc..000000000 --- a/dep/zlib/CMakeLists.txt +++ /dev/null @@ -1,33 +0,0 @@ -add_library(zlib - include/zconf.h - include/zlib.h - src/adler32.c - src/compress.c - src/crc32.c - src/crc32.h - src/deflate.c - src/deflate.h - src/gzclose.c - src/gzguts.h - src/gzlib.c - src/gzread.c - src/gzwrite.c - src/infback.c - src/inffast.c - src/inffast.h - src/inffixed.h - src/inflate.c - src/inflate.h - src/inftrees.c - src/inftrees.h - src/trees.c - src/trees.h - src/uncompr.c - src/zutil.c - src/zutil.h -) - -target_include_directories(zlib PRIVATE "${CMAKE_CURRENT_SOURCE_DIR}/include") -target_include_directories(zlib INTERFACE "${CMAKE_CURRENT_SOURCE_DIR}/include") -disable_compiler_warnings_for_target(zlib) -add_library(ZLIB::ZLIB ALIAS zlib) diff --git a/dep/zlib/ChangeLog b/dep/zlib/ChangeLog deleted file mode 100644 index 30199a65a..000000000 --- a/dep/zlib/ChangeLog +++ /dev/null @@ -1,1515 +0,0 @@ - - ChangeLog file for zlib - -Changes in 1.2.11 (15 Jan 2017) -- Fix deflate stored bug when pulling last block from window -- Permit immediate deflateParams changes before any deflate input - -Changes in 1.2.10 (2 Jan 2017) -- Avoid warnings on snprintf() return value -- Fix bug in deflate_stored() for zero-length input -- Fix bug in gzwrite.c that produced corrupt gzip files -- Remove files to be installed before copying them in Makefile.in -- Add warnings when compiling with assembler code - -Changes in 1.2.9 (31 Dec 2016) -- Fix contrib/minizip to permit unzipping with desktop API [Zouzou] -- Improve contrib/blast to return unused bytes -- Assure that gzoffset() is correct when appending -- Improve compress() and uncompress() to support large lengths -- Fix bug in test/example.c where error code not saved -- Remedy Coverity warning [Randers-Pehrson] -- Improve speed of gzprintf() in transparent mode -- Fix inflateInit2() bug when windowBits is 16 or 32 -- Change DEBUG macro to ZLIB_DEBUG -- Avoid uninitialized access by gzclose_w() -- Allow building zlib outside of the source directory -- Fix bug that accepted invalid zlib header when windowBits is zero -- Fix gzseek() problem on MinGW due to buggy _lseeki64 there -- Loop on write() calls in gzwrite.c in case of non-blocking I/O -- Add --warn (-w) option to ./configure for more compiler warnings -- Reject a window size of 256 bytes if not using the zlib wrapper -- Fix bug when level 0 used with Z_HUFFMAN or Z_RLE -- Add --debug (-d) option to ./configure to define ZLIB_DEBUG -- Fix bugs in creating a very large gzip header -- Add uncompress2() function, which returns the input size used -- Assure that deflateParams() will not switch functions mid-block -- Dramatically speed up deflation for level 0 (storing) -- Add gzfread(), duplicating the interface of fread() -- Add gzfwrite(), duplicating the interface of fwrite() -- Add deflateGetDictionary() function -- Use snprintf() for later versions of Microsoft C -- Fix *Init macros to use z_ prefix when requested -- Replace as400 with os400 for OS/400 support [Monnerat] -- Add crc32_z() and adler32_z() functions with size_t lengths -- Update Visual Studio project files [AraHaan] - -Changes in 1.2.8 (28 Apr 2013) -- Update contrib/minizip/iowin32.c for Windows RT [Vollant] -- Do not force Z_CONST for C++ -- Clean up contrib/vstudio [Roß] -- Correct spelling error in zlib.h -- Fix mixed line endings in contrib/vstudio - -Changes in 1.2.7.3 (13 Apr 2013) -- Fix version numbers and DLL names in contrib/vstudio/*/zlib.rc - -Changes in 1.2.7.2 (13 Apr 2013) -- Change check for a four-byte type back to hexadecimal -- Fix typo in win32/Makefile.msc -- Add casts in gzwrite.c for pointer differences - -Changes in 1.2.7.1 (24 Mar 2013) -- Replace use of unsafe string functions with snprintf if available -- Avoid including stddef.h on Windows for Z_SOLO compile [Niessink] -- Fix gzgetc undefine when Z_PREFIX set [Turk] -- Eliminate use of mktemp in Makefile (not always available) -- Fix bug in 'F' mode for gzopen() -- Add inflateGetDictionary() function -- Correct comment in deflate.h -- Use _snprintf for snprintf in Microsoft C -- On Darwin, only use /usr/bin/libtool if libtool is not Apple -- Delete "--version" file if created by "ar --version" [Richard G.] -- Fix configure check for veracity of compiler error return codes -- Fix CMake compilation of static lib for MSVC2010 x64 -- Remove unused variable in infback9.c -- Fix argument checks in gzlog_compress() and gzlog_write() -- Clean up the usage of z_const and respect const usage within zlib -- Clean up examples/gzlog.[ch] comparisons of different types -- Avoid shift equal to bits in type (caused endless loop) -- Fix uninitialized value bug in gzputc() introduced by const patches -- Fix memory allocation error in examples/zran.c [Nor] -- Fix bug where gzopen(), gzclose() would write an empty file -- Fix bug in gzclose() when gzwrite() runs out of memory -- Check for input buffer malloc failure in examples/gzappend.c -- Add note to contrib/blast to use binary mode in stdio -- Fix comparisons of differently signed integers in contrib/blast -- Check for invalid code length codes in contrib/puff -- Fix serious but very rare decompression bug in inftrees.c -- Update inflateBack() comments, since inflate() can be faster -- Use underscored I/O function names for WINAPI_FAMILY -- Add _tr_flush_bits to the external symbols prefixed by --zprefix -- Add contrib/vstudio/vc10 pre-build step for static only -- Quote --version-script argument in CMakeLists.txt -- Don't specify --version-script on Apple platforms in CMakeLists.txt -- Fix casting error in contrib/testzlib/testzlib.c -- Fix types in contrib/minizip to match result of get_crc_table() -- Simplify contrib/vstudio/vc10 with 'd' suffix -- Add TOP support to win32/Makefile.msc -- Suport i686 and amd64 assembler builds in CMakeLists.txt -- Fix typos in the use of _LARGEFILE64_SOURCE in zconf.h -- Add vc11 and vc12 build files to contrib/vstudio -- Add gzvprintf() as an undocumented function in zlib -- Fix configure for Sun shell -- Remove runtime check in configure for four-byte integer type -- Add casts and consts to ease user conversion to C++ -- Add man pages for minizip and miniunzip -- In Makefile uninstall, don't rm if preceding cd fails -- Do not return Z_BUF_ERROR if deflateParam() has nothing to write - -Changes in 1.2.7 (2 May 2012) -- Replace use of memmove() with a simple copy for portability -- Test for existence of strerror -- Restore gzgetc_ for backward compatibility with 1.2.6 -- Fix build with non-GNU make on Solaris -- Require gcc 4.0 or later on Mac OS X to use the hidden attribute -- Include unistd.h for Watcom C -- Use __WATCOMC__ instead of __WATCOM__ -- Do not use the visibility attribute if NO_VIZ defined -- Improve the detection of no hidden visibility attribute -- Avoid using __int64 for gcc or solo compilation -- Cast to char * in gzprintf to avoid warnings [Zinser] -- Fix make_vms.com for VAX [Zinser] -- Don't use library or built-in byte swaps -- Simplify test and use of gcc hidden attribute -- Fix bug in gzclose_w() when gzwrite() fails to allocate memory -- Add "x" (O_EXCL) and "e" (O_CLOEXEC) modes support to gzopen() -- Fix bug in test/minigzip.c for configure --solo -- Fix contrib/vstudio project link errors [Mohanathas] -- Add ability to choose the builder in make_vms.com [Schweda] -- Add DESTDIR support to mingw32 win32/Makefile.gcc -- Fix comments in win32/Makefile.gcc for proper usage -- Allow overriding the default install locations for cmake -- Generate and install the pkg-config file with cmake -- Build both a static and a shared version of zlib with cmake -- Include version symbols for cmake builds -- If using cmake with MSVC, add the source directory to the includes -- Remove unneeded EXTRA_CFLAGS from win32/Makefile.gcc [Truta] -- Move obsolete emx makefile to old [Truta] -- Allow the use of -Wundef when compiling or using zlib -- Avoid the use of the -u option with mktemp -- Improve inflate() documentation on the use of Z_FINISH -- Recognize clang as gcc -- Add gzopen_w() in Windows for wide character path names -- Rename zconf.h in CMakeLists.txt to move it out of the way -- Add source directory in CMakeLists.txt for building examples -- Look in build directory for zlib.pc in CMakeLists.txt -- Remove gzflags from zlibvc.def in vc9 and vc10 -- Fix contrib/minizip compilation in the MinGW environment -- Update ./configure for Solaris, support --64 [Mooney] -- Remove -R. from Solaris shared build (possible security issue) -- Avoid race condition for parallel make (-j) running example -- Fix type mismatch between get_crc_table() and crc_table -- Fix parsing of version with "-" in CMakeLists.txt [Snider, Ziegler] -- Fix the path to zlib.map in CMakeLists.txt -- Force the native libtool in Mac OS X to avoid GNU libtool [Beebe] -- Add instructions to win32/Makefile.gcc for shared install [Torri] - -Changes in 1.2.6.1 (12 Feb 2012) -- Avoid the use of the Objective-C reserved name "id" -- Include io.h in gzguts.h for Microsoft compilers -- Fix problem with ./configure --prefix and gzgetc macro -- Include gz_header definition when compiling zlib solo -- Put gzflags() functionality back in zutil.c -- Avoid library header include in crc32.c for Z_SOLO -- Use name in GCC_CLASSIC as C compiler for coverage testing, if set -- Minor cleanup in contrib/minizip/zip.c [Vollant] -- Update make_vms.com [Zinser] -- Remove unnecessary gzgetc_ function -- Use optimized byte swap operations for Microsoft and GNU [Snyder] -- Fix minor typo in zlib.h comments [Rzesniowiecki] - -Changes in 1.2.6 (29 Jan 2012) -- Update the Pascal interface in contrib/pascal -- Fix function numbers for gzgetc_ in zlibvc.def files -- Fix configure.ac for contrib/minizip [Schiffer] -- Fix large-entry detection in minizip on 64-bit systems [Schiffer] -- Have ./configure use the compiler return code for error indication -- Fix CMakeLists.txt for cross compilation [McClure] -- Fix contrib/minizip/zip.c for 64-bit architectures [Dalsnes] -- Fix compilation of contrib/minizip on FreeBSD [Marquez] -- Correct suggested usages in win32/Makefile.msc [Shachar, Horvath] -- Include io.h for Turbo C / Borland C on all platforms [Truta] -- Make version explicit in contrib/minizip/configure.ac [Bosmans] -- Avoid warning for no encryption in contrib/minizip/zip.c [Vollant] -- Minor cleanup up contrib/minizip/unzip.c [Vollant] -- Fix bug when compiling minizip with C++ [Vollant] -- Protect for long name and extra fields in contrib/minizip [Vollant] -- Avoid some warnings in contrib/minizip [Vollant] -- Add -I../.. -L../.. to CFLAGS for minizip and miniunzip -- Add missing libs to minizip linker command -- Add support for VPATH builds in contrib/minizip -- Add an --enable-demos option to contrib/minizip/configure -- Add the generation of configure.log by ./configure -- Exit when required parameters not provided to win32/Makefile.gcc -- Have gzputc return the character written instead of the argument -- Use the -m option on ldconfig for BSD systems [Tobias] -- Correct in zlib.map when deflateResetKeep was added - -Changes in 1.2.5.3 (15 Jan 2012) -- Restore gzgetc function for binary compatibility -- Do not use _lseeki64 under Borland C++ [Truta] -- Update win32/Makefile.msc to build test/*.c [Truta] -- Remove old/visualc6 given CMakefile and other alternatives -- Update AS400 build files and documentation [Monnerat] -- Update win32/Makefile.gcc to build test/*.c [Truta] -- Permit stronger flushes after Z_BLOCK flushes -- Avoid extraneous empty blocks when doing empty flushes -- Permit Z_NULL arguments to deflatePending -- Allow deflatePrime() to insert bits in the middle of a stream -- Remove second empty static block for Z_PARTIAL_FLUSH -- Write out all of the available bits when using Z_BLOCK -- Insert the first two strings in the hash table after a flush - -Changes in 1.2.5.2 (17 Dec 2011) -- fix ld error: unable to find version dependency 'ZLIB_1.2.5' -- use relative symlinks for shared libs -- Avoid searching past window for Z_RLE strategy -- Assure that high-water mark initialization is always applied in deflate -- Add assertions to fill_window() in deflate.c to match comments -- Update python link in README -- Correct spelling error in gzread.c -- Fix bug in gzgets() for a concatenated empty gzip stream -- Correct error in comment for gz_make() -- Change gzread() and related to ignore junk after gzip streams -- Allow gzread() and related to continue after gzclearerr() -- Allow gzrewind() and gzseek() after a premature end-of-file -- Simplify gzseek() now that raw after gzip is ignored -- Change gzgetc() to a macro for speed (~40% speedup in testing) -- Fix gzclose() to return the actual error last encountered -- Always add large file support for windows -- Include zconf.h for windows large file support -- Include zconf.h.cmakein for windows large file support -- Update zconf.h.cmakein on make distclean -- Merge vestigial vsnprintf determination from zutil.h to gzguts.h -- Clarify how gzopen() appends in zlib.h comments -- Correct documentation of gzdirect() since junk at end now ignored -- Add a transparent write mode to gzopen() when 'T' is in the mode -- Update python link in zlib man page -- Get inffixed.h and MAKEFIXED result to match -- Add a ./config --solo option to make zlib subset with no library use -- Add undocumented inflateResetKeep() function for CAB file decoding -- Add --cover option to ./configure for gcc coverage testing -- Add #define ZLIB_CONST option to use const in the z_stream interface -- Add comment to gzdopen() in zlib.h to use dup() when using fileno() -- Note behavior of uncompress() to provide as much data as it can -- Add files in contrib/minizip to aid in building libminizip -- Split off AR options in Makefile.in and configure -- Change ON macro to Z_ARG to avoid application conflicts -- Facilitate compilation with Borland C++ for pragmas and vsnprintf -- Include io.h for Turbo C / Borland C++ -- Move example.c and minigzip.c to test/ -- Simplify incomplete code table filling in inflate_table() -- Remove code from inflate.c and infback.c that is impossible to execute -- Test the inflate code with full coverage -- Allow deflateSetDictionary, inflateSetDictionary at any time (in raw) -- Add deflateResetKeep and fix inflateResetKeep to retain dictionary -- Fix gzwrite.c to accommodate reduced memory zlib compilation -- Have inflate() with Z_FINISH avoid the allocation of a window -- Do not set strm->adler when doing raw inflate -- Fix gzeof() to behave just like feof() when read is not past end of file -- Fix bug in gzread.c when end-of-file is reached -- Avoid use of Z_BUF_ERROR in gz* functions except for premature EOF -- Document gzread() capability to read concurrently written files -- Remove hard-coding of resource compiler in CMakeLists.txt [Blammo] - -Changes in 1.2.5.1 (10 Sep 2011) -- Update FAQ entry on shared builds (#13) -- Avoid symbolic argument to chmod in Makefile.in -- Fix bug and add consts in contrib/puff [Oberhumer] -- Update contrib/puff/zeros.raw test file to have all block types -- Add full coverage test for puff in contrib/puff/Makefile -- Fix static-only-build install in Makefile.in -- Fix bug in unzGetCurrentFileInfo() in contrib/minizip [Kuno] -- Add libz.a dependency to shared in Makefile.in for parallel builds -- Spell out "number" (instead of "nb") in zlib.h for total_in, total_out -- Replace $(...) with `...` in configure for non-bash sh [Bowler] -- Add darwin* to Darwin* and solaris* to SunOS\ 5* in configure [Groffen] -- Add solaris* to Linux* in configure to allow gcc use [Groffen] -- Add *bsd* to Linux* case in configure [Bar-Lev] -- Add inffast.obj to dependencies in win32/Makefile.msc -- Correct spelling error in deflate.h [Kohler] -- Change libzdll.a again to libz.dll.a (!) in win32/Makefile.gcc -- Add test to configure for GNU C looking for gcc in output of $cc -v -- Add zlib.pc generation to win32/Makefile.gcc [Weigelt] -- Fix bug in zlib.h for _FILE_OFFSET_BITS set and _LARGEFILE64_SOURCE not -- Add comment in zlib.h that adler32_combine with len2 < 0 makes no sense -- Make NO_DIVIDE option in adler32.c much faster (thanks to John Reiser) -- Make stronger test in zconf.h to include unistd.h for LFS -- Apply Darwin patches for 64-bit file offsets to contrib/minizip [Slack] -- Fix zlib.h LFS support when Z_PREFIX used -- Add updated as400 support (removed from old) [Monnerat] -- Avoid deflate sensitivity to volatile input data -- Avoid division in adler32_combine for NO_DIVIDE -- Clarify the use of Z_FINISH with deflateBound() amount of space -- Set binary for output file in puff.c -- Use u4 type for crc_table to avoid conversion warnings -- Apply casts in zlib.h to avoid conversion warnings -- Add OF to prototypes for adler32_combine_ and crc32_combine_ [Miller] -- Improve inflateSync() documentation to note indeterminancy -- Add deflatePending() function to return the amount of pending output -- Correct the spelling of "specification" in FAQ [Randers-Pehrson] -- Add a check in configure for stdarg.h, use for gzprintf() -- Check that pointers fit in ints when gzprint() compiled old style -- Add dummy name before $(SHAREDLIBV) in Makefile [Bar-Lev, Bowler] -- Delete line in configure that adds -L. libz.a to LDFLAGS [Weigelt] -- Add debug records in assmebler code [Londer] -- Update RFC references to use http://tools.ietf.org/html/... [Li] -- Add --archs option, use of libtool to configure for Mac OS X [Borstel] - -Changes in 1.2.5 (19 Apr 2010) -- Disable visibility attribute in win32/Makefile.gcc [Bar-Lev] -- Default to libdir as sharedlibdir in configure [Nieder] -- Update copyright dates on modified source files -- Update trees.c to be able to generate modified trees.h -- Exit configure for MinGW, suggesting win32/Makefile.gcc -- Check for NULL path in gz_open [Homurlu] - -Changes in 1.2.4.5 (18 Apr 2010) -- Set sharedlibdir in configure [Torok] -- Set LDFLAGS in Makefile.in [Bar-Lev] -- Avoid mkdir objs race condition in Makefile.in [Bowler] -- Add ZLIB_INTERNAL in front of internal inter-module functions and arrays -- Define ZLIB_INTERNAL to hide internal functions and arrays for GNU C -- Don't use hidden attribute when it is a warning generator (e.g. Solaris) - -Changes in 1.2.4.4 (18 Apr 2010) -- Fix CROSS_PREFIX executable testing, CHOST extract, mingw* [Torok] -- Undefine _LARGEFILE64_SOURCE in zconf.h if it is zero, but not if empty -- Try to use bash or ksh regardless of functionality of /bin/sh -- Fix configure incompatibility with NetBSD sh -- Remove attempt to run under bash or ksh since have better NetBSD fix -- Fix win32/Makefile.gcc for MinGW [Bar-Lev] -- Add diagnostic messages when using CROSS_PREFIX in configure -- Added --sharedlibdir option to configure [Weigelt] -- Use hidden visibility attribute when available [Frysinger] - -Changes in 1.2.4.3 (10 Apr 2010) -- Only use CROSS_PREFIX in configure for ar and ranlib if they exist -- Use CROSS_PREFIX for nm [Bar-Lev] -- Assume _LARGEFILE64_SOURCE defined is equivalent to true -- Avoid use of undefined symbols in #if with && and || -- Make *64 prototypes in gzguts.h consistent with functions -- Add -shared load option for MinGW in configure [Bowler] -- Move z_off64_t to public interface, use instead of off64_t -- Remove ! from shell test in configure (not portable to Solaris) -- Change +0 macro tests to -0 for possibly increased portability - -Changes in 1.2.4.2 (9 Apr 2010) -- Add consistent carriage returns to readme.txt's in masmx86 and masmx64 -- Really provide prototypes for *64 functions when building without LFS -- Only define unlink() in minigzip.c if unistd.h not included -- Update README to point to contrib/vstudio project files -- Move projects/vc6 to old/ and remove projects/ -- Include stdlib.h in minigzip.c for setmode() definition under WinCE -- Clean up assembler builds in win32/Makefile.msc [Rowe] -- Include sys/types.h for Microsoft for off_t definition -- Fix memory leak on error in gz_open() -- Symbolize nm as $NM in configure [Weigelt] -- Use TEST_LDSHARED instead of LDSHARED to link test programs [Weigelt] -- Add +0 to _FILE_OFFSET_BITS and _LFS64_LARGEFILE in case not defined -- Fix bug in gzeof() to take into account unused input data -- Avoid initialization of structures with variables in puff.c -- Updated win32/README-WIN32.txt [Rowe] - -Changes in 1.2.4.1 (28 Mar 2010) -- Remove the use of [a-z] constructs for sed in configure [gentoo 310225] -- Remove $(SHAREDLIB) from LIBS in Makefile.in [Creech] -- Restore "for debugging" comment on sprintf() in gzlib.c -- Remove fdopen for MVS from gzguts.h -- Put new README-WIN32.txt in win32 [Rowe] -- Add check for shell to configure and invoke another shell if needed -- Fix big fat stinking bug in gzseek() on uncompressed files -- Remove vestigial F_OPEN64 define in zutil.h -- Set and check the value of _LARGEFILE_SOURCE and _LARGEFILE64_SOURCE -- Avoid errors on non-LFS systems when applications define LFS macros -- Set EXE to ".exe" in configure for MINGW [Kahle] -- Match crc32() in crc32.c exactly to the prototype in zlib.h [Sherrill] -- Add prefix for cross-compilation in win32/makefile.gcc [Bar-Lev] -- Add DLL install in win32/makefile.gcc [Bar-Lev] -- Allow Linux* or linux* from uname in configure [Bar-Lev] -- Allow ldconfig to be redefined in configure and Makefile.in [Bar-Lev] -- Add cross-compilation prefixes to configure [Bar-Lev] -- Match type exactly in gz_load() invocation in gzread.c -- Match type exactly of zcalloc() in zutil.c to zlib.h alloc_func -- Provide prototypes for *64 functions when building zlib without LFS -- Don't use -lc when linking shared library on MinGW -- Remove errno.h check in configure and vestigial errno code in zutil.h - -Changes in 1.2.4 (14 Mar 2010) -- Fix VER3 extraction in configure for no fourth subversion -- Update zlib.3, add docs to Makefile.in to make .pdf out of it -- Add zlib.3.pdf to distribution -- Don't set error code in gzerror() if passed pointer is NULL -- Apply destination directory fixes to CMakeLists.txt [Lowman] -- Move #cmakedefine's to a new zconf.in.cmakein -- Restore zconf.h for builds that don't use configure or cmake -- Add distclean to dummy Makefile for convenience -- Update and improve INDEX, README, and FAQ -- Update CMakeLists.txt for the return of zconf.h [Lowman] -- Update contrib/vstudio/vc9 and vc10 [Vollant] -- Change libz.dll.a back to libzdll.a in win32/Makefile.gcc -- Apply license and readme changes to contrib/asm686 [Raiter] -- Check file name lengths and add -c option in minigzip.c [Li] -- Update contrib/amd64 and contrib/masmx86/ [Vollant] -- Avoid use of "eof" parameter in trees.c to not shadow library variable -- Update make_vms.com for removal of zlibdefs.h [Zinser] -- Update assembler code and vstudio projects in contrib [Vollant] -- Remove outdated assembler code contrib/masm686 and contrib/asm586 -- Remove old vc7 and vc8 from contrib/vstudio -- Update win32/Makefile.msc, add ZLIB_VER_SUBREVISION [Rowe] -- Fix memory leaks in gzclose_r() and gzclose_w(), file leak in gz_open() -- Add contrib/gcc_gvmat64 for longest_match and inflate_fast [Vollant] -- Remove *64 functions from win32/zlib.def (they're not 64-bit yet) -- Fix bug in void-returning vsprintf() case in gzwrite.c -- Fix name change from inflate.h in contrib/inflate86/inffas86.c -- Check if temporary file exists before removing in make_vms.com [Zinser] -- Fix make install and uninstall for --static option -- Fix usage of _MSC_VER in gzguts.h and zutil.h [Truta] -- Update readme.txt in contrib/masmx64 and masmx86 to assemble - -Changes in 1.2.3.9 (21 Feb 2010) -- Expunge gzio.c -- Move as400 build information to old -- Fix updates in contrib/minizip and contrib/vstudio -- Add const to vsnprintf test in configure to avoid warnings [Weigelt] -- Delete zconf.h (made by configure) [Weigelt] -- Change zconf.in.h to zconf.h.in per convention [Weigelt] -- Check for NULL buf in gzgets() -- Return empty string for gzgets() with len == 1 (like fgets()) -- Fix description of gzgets() in zlib.h for end-of-file, NULL return -- Update minizip to 1.1 [Vollant] -- Avoid MSVC loss of data warnings in gzread.c, gzwrite.c -- Note in zlib.h that gzerror() should be used to distinguish from EOF -- Remove use of snprintf() from gzlib.c -- Fix bug in gzseek() -- Update contrib/vstudio, adding vc9 and vc10 [Kuno, Vollant] -- Fix zconf.h generation in CMakeLists.txt [Lowman] -- Improve comments in zconf.h where modified by configure - -Changes in 1.2.3.8 (13 Feb 2010) -- Clean up text files (tabs, trailing whitespace, etc.) [Oberhumer] -- Use z_off64_t in gz_zero() and gz_skip() to match state->skip -- Avoid comparison problem when sizeof(int) == sizeof(z_off64_t) -- Revert to Makefile.in from 1.2.3.6 (live with the clutter) -- Fix missing error return in gzflush(), add zlib.h note -- Add *64 functions to zlib.map [Levin] -- Fix signed/unsigned comparison in gz_comp() -- Use SFLAGS when testing shared linking in configure -- Add --64 option to ./configure to use -m64 with gcc -- Fix ./configure --help to correctly name options -- Have make fail if a test fails [Levin] -- Avoid buffer overrun in contrib/masmx64/gvmat64.asm [Simpson] -- Remove assembler object files from contrib - -Changes in 1.2.3.7 (24 Jan 2010) -- Always gzopen() with O_LARGEFILE if available -- Fix gzdirect() to work immediately after gzopen() or gzdopen() -- Make gzdirect() more precise when the state changes while reading -- Improve zlib.h documentation in many places -- Catch memory allocation failure in gz_open() -- Complete close operation if seek forward in gzclose_w() fails -- Return Z_ERRNO from gzclose_r() if close() fails -- Return Z_STREAM_ERROR instead of EOF for gzclose() being passed NULL -- Return zero for gzwrite() errors to match zlib.h description -- Return -1 on gzputs() error to match zlib.h description -- Add zconf.in.h to allow recovery from configure modification [Weigelt] -- Fix static library permissions in Makefile.in [Weigelt] -- Avoid warnings in configure tests that hide functionality [Weigelt] -- Add *BSD and DragonFly to Linux case in configure [gentoo 123571] -- Change libzdll.a to libz.dll.a in win32/Makefile.gcc [gentoo 288212] -- Avoid access of uninitialized data for first inflateReset2 call [Gomes] -- Keep object files in subdirectories to reduce the clutter somewhat -- Remove default Makefile and zlibdefs.h, add dummy Makefile -- Add new external functions to Z_PREFIX, remove duplicates, z_z_ -> z_ -- Remove zlibdefs.h completely -- modify zconf.h instead - -Changes in 1.2.3.6 (17 Jan 2010) -- Avoid void * arithmetic in gzread.c and gzwrite.c -- Make compilers happier with const char * for gz_error message -- Avoid unused parameter warning in inflate.c -- Avoid signed-unsigned comparison warning in inflate.c -- Indent #pragma's for traditional C -- Fix usage of strwinerror() in glib.c, change to gz_strwinerror() -- Correct email address in configure for system options -- Update make_vms.com and add make_vms.com to contrib/minizip [Zinser] -- Update zlib.map [Brown] -- Fix Makefile.in for Solaris 10 make of example64 and minizip64 [Torok] -- Apply various fixes to CMakeLists.txt [Lowman] -- Add checks on len in gzread() and gzwrite() -- Add error message for no more room for gzungetc() -- Remove zlib version check in gzwrite() -- Defer compression of gzprintf() result until need to -- Use snprintf() in gzdopen() if available -- Remove USE_MMAP configuration determination (only used by minigzip) -- Remove examples/pigz.c (available separately) -- Update examples/gun.c to 1.6 - -Changes in 1.2.3.5 (8 Jan 2010) -- Add space after #if in zutil.h for some compilers -- Fix relatively harmless bug in deflate_fast() [Exarevsky] -- Fix same problem in deflate_slow() -- Add $(SHAREDLIBV) to LIBS in Makefile.in [Brown] -- Add deflate_rle() for faster Z_RLE strategy run-length encoding -- Add deflate_huff() for faster Z_HUFFMAN_ONLY encoding -- Change name of "write" variable in inffast.c to avoid library collisions -- Fix premature EOF from gzread() in gzio.c [Brown] -- Use zlib header window size if windowBits is 0 in inflateInit2() -- Remove compressBound() call in deflate.c to avoid linking compress.o -- Replace use of errno in gz* with functions, support WinCE [Alves] -- Provide alternative to perror() in minigzip.c for WinCE [Alves] -- Don't use _vsnprintf on later versions of MSVC [Lowman] -- Add CMake build script and input file [Lowman] -- Update contrib/minizip to 1.1 [Svensson, Vollant] -- Moved nintendods directory from contrib to . -- Replace gzio.c with a new set of routines with the same functionality -- Add gzbuffer(), gzoffset(), gzclose_r(), gzclose_w() as part of above -- Update contrib/minizip to 1.1b -- Change gzeof() to return 0 on error instead of -1 to agree with zlib.h - -Changes in 1.2.3.4 (21 Dec 2009) -- Use old school .SUFFIXES in Makefile.in for FreeBSD compatibility -- Update comments in configure and Makefile.in for default --shared -- Fix test -z's in configure [Marquess] -- Build examplesh and minigzipsh when not testing -- Change NULL's to Z_NULL's in deflate.c and in comments in zlib.h -- Import LDFLAGS from the environment in configure -- Fix configure to populate SFLAGS with discovered CFLAGS options -- Adapt make_vms.com to the new Makefile.in [Zinser] -- Add zlib2ansi script for C++ compilation [Marquess] -- Add _FILE_OFFSET_BITS=64 test to make test (when applicable) -- Add AMD64 assembler code for longest match to contrib [Teterin] -- Include options from $SFLAGS when doing $LDSHARED -- Simplify 64-bit file support by introducing z_off64_t type -- Make shared object files in objs directory to work around old Sun cc -- Use only three-part version number for Darwin shared compiles -- Add rc option to ar in Makefile.in for when ./configure not run -- Add -WI,-rpath,. to LDFLAGS for OSF 1 V4* -- Set LD_LIBRARYN32_PATH for SGI IRIX shared compile -- Protect against _FILE_OFFSET_BITS being defined when compiling zlib -- Rename Makefile.in targets allstatic to static and allshared to shared -- Fix static and shared Makefile.in targets to be independent -- Correct error return bug in gz_open() by setting state [Brown] -- Put spaces before ;;'s in configure for better sh compatibility -- Add pigz.c (parallel implementation of gzip) to examples/ -- Correct constant in crc32.c to UL [Leventhal] -- Reject negative lengths in crc32_combine() -- Add inflateReset2() function to work like inflateEnd()/inflateInit2() -- Include sys/types.h for _LARGEFILE64_SOURCE [Brown] -- Correct typo in doc/algorithm.txt [Janik] -- Fix bug in adler32_combine() [Zhu] -- Catch missing-end-of-block-code error in all inflates and in puff - Assures that random input to inflate eventually results in an error -- Added enough.c (calculation of ENOUGH for inftrees.h) to examples/ -- Update ENOUGH and its usage to reflect discovered bounds -- Fix gzerror() error report on empty input file [Brown] -- Add ush casts in trees.c to avoid pedantic runtime errors -- Fix typo in zlib.h uncompress() description [Reiss] -- Correct inflate() comments with regard to automatic header detection -- Remove deprecation comment on Z_PARTIAL_FLUSH (it stays) -- Put new version of gzlog (2.0) in examples with interruption recovery -- Add puff compile option to permit invalid distance-too-far streams -- Add puff TEST command options, ability to read piped input -- Prototype the *64 functions in zlib.h when _FILE_OFFSET_BITS == 64, but - _LARGEFILE64_SOURCE not defined -- Fix Z_FULL_FLUSH to truly erase the past by resetting s->strstart -- Fix deflateSetDictionary() to use all 32K for output consistency -- Remove extraneous #define MIN_LOOKAHEAD in deflate.c (in deflate.h) -- Clear bytes after deflate lookahead to avoid use of uninitialized data -- Change a limit in inftrees.c to be more transparent to Coverity Prevent -- Update win32/zlib.def with exported symbols from zlib.h -- Correct spelling errors in zlib.h [Willem, Sobrado] -- Allow Z_BLOCK for deflate() to force a new block -- Allow negative bits in inflatePrime() to delete existing bit buffer -- Add Z_TREES flush option to inflate() to return at end of trees -- Add inflateMark() to return current state information for random access -- Add Makefile for NintendoDS to contrib [Costa] -- Add -w in configure compile tests to avoid spurious warnings [Beucler] -- Fix typos in zlib.h comments for deflateSetDictionary() -- Fix EOF detection in transparent gzread() [Maier] - -Changes in 1.2.3.3 (2 October 2006) -- Make --shared the default for configure, add a --static option -- Add compile option to permit invalid distance-too-far streams -- Add inflateUndermine() function which is required to enable above -- Remove use of "this" variable name for C++ compatibility [Marquess] -- Add testing of shared library in make test, if shared library built -- Use ftello() and fseeko() if available instead of ftell() and fseek() -- Provide two versions of all functions that use the z_off_t type for - binary compatibility -- a normal version and a 64-bit offset version, - per the Large File Support Extension when _LARGEFILE64_SOURCE is - defined; use the 64-bit versions by default when _FILE_OFFSET_BITS - is defined to be 64 -- Add a --uname= option to configure to perhaps help with cross-compiling - -Changes in 1.2.3.2 (3 September 2006) -- Turn off silly Borland warnings [Hay] -- Use off64_t and define _LARGEFILE64_SOURCE when present -- Fix missing dependency on inffixed.h in Makefile.in -- Rig configure --shared to build both shared and static [Teredesai, Truta] -- Remove zconf.in.h and instead create a new zlibdefs.h file -- Fix contrib/minizip/unzip.c non-encrypted after encrypted [Vollant] -- Add treebuild.xml (see http://treebuild.metux.de/) [Weigelt] - -Changes in 1.2.3.1 (16 August 2006) -- Add watcom directory with OpenWatcom make files [Daniel] -- Remove #undef of FAR in zconf.in.h for MVS [Fedtke] -- Update make_vms.com [Zinser] -- Use -fPIC for shared build in configure [Teredesai, Nicholson] -- Use only major version number for libz.so on IRIX and OSF1 [Reinholdtsen] -- Use fdopen() (not _fdopen()) for Interix in zutil.h [Bäck] -- Add some FAQ entries about the contrib directory -- Update the MVS question in the FAQ -- Avoid extraneous reads after EOF in gzio.c [Brown] -- Correct spelling of "successfully" in gzio.c [Randers-Pehrson] -- Add comments to zlib.h about gzerror() usage [Brown] -- Set extra flags in gzip header in gzopen() like deflate() does -- Make configure options more compatible with double-dash conventions - [Weigelt] -- Clean up compilation under Solaris SunStudio cc [Rowe, Reinholdtsen] -- Fix uninstall target in Makefile.in [Truta] -- Add pkgconfig support [Weigelt] -- Use $(DESTDIR) macro in Makefile.in [Reinholdtsen, Weigelt] -- Replace set_data_type() with a more accurate detect_data_type() in - trees.c, according to the txtvsbin.txt document [Truta] -- Swap the order of #include and #include "zlib.h" in - gzio.c, example.c and minigzip.c [Truta] -- Shut up annoying VS2005 warnings about standard C deprecation [Rowe, - Truta] (where?) -- Fix target "clean" from win32/Makefile.bor [Truta] -- Create .pdb and .manifest files in win32/makefile.msc [Ziegler, Rowe] -- Update zlib www home address in win32/DLL_FAQ.txt [Truta] -- Update contrib/masmx86/inffas32.asm for VS2005 [Vollant, Van Wassenhove] -- Enable browse info in the "Debug" and "ASM Debug" configurations in - the Visual C++ 6 project, and set (non-ASM) "Debug" as default [Truta] -- Add pkgconfig support [Weigelt] -- Add ZLIB_VER_MAJOR, ZLIB_VER_MINOR and ZLIB_VER_REVISION in zlib.h, - for use in win32/zlib1.rc [Polushin, Rowe, Truta] -- Add a document that explains the new text detection scheme to - doc/txtvsbin.txt [Truta] -- Add rfc1950.txt, rfc1951.txt and rfc1952.txt to doc/ [Truta] -- Move algorithm.txt into doc/ [Truta] -- Synchronize FAQ with website -- Fix compressBound(), was low for some pathological cases [Fearnley] -- Take into account wrapper variations in deflateBound() -- Set examples/zpipe.c input and output to binary mode for Windows -- Update examples/zlib_how.html with new zpipe.c (also web site) -- Fix some warnings in examples/gzlog.c and examples/zran.c (it seems - that gcc became pickier in 4.0) -- Add zlib.map for Linux: "All symbols from zlib-1.1.4 remain - un-versioned, the patch adds versioning only for symbols introduced in - zlib-1.2.0 or later. It also declares as local those symbols which are - not designed to be exported." [Levin] -- Update Z_PREFIX list in zconf.in.h, add --zprefix option to configure -- Do not initialize global static by default in trees.c, add a response - NO_INIT_GLOBAL_POINTERS to initialize them if needed [Marquess] -- Don't use strerror() in gzio.c under WinCE [Yakimov] -- Don't use errno.h in zutil.h under WinCE [Yakimov] -- Move arguments for AR to its usage to allow replacing ar [Marot] -- Add HAVE_VISIBILITY_PRAGMA in zconf.in.h for Mozilla [Randers-Pehrson] -- Improve inflateInit() and inflateInit2() documentation -- Fix structure size comment in inflate.h -- Change configure help option from --h* to --help [Santos] - -Changes in 1.2.3 (18 July 2005) -- Apply security vulnerability fixes to contrib/infback9 as well -- Clean up some text files (carriage returns, trailing space) -- Update testzlib, vstudio, masmx64, and masmx86 in contrib [Vollant] - -Changes in 1.2.2.4 (11 July 2005) -- Add inflatePrime() function for starting inflation at bit boundary -- Avoid some Visual C warnings in deflate.c -- Avoid more silly Visual C warnings in inflate.c and inftrees.c for 64-bit - compile -- Fix some spelling errors in comments [Betts] -- Correct inflateInit2() error return documentation in zlib.h -- Add zran.c example of compressed data random access to examples - directory, shows use of inflatePrime() -- Fix cast for assignments to strm->state in inflate.c and infback.c -- Fix zlibCompileFlags() in zutil.c to use 1L for long shifts [Oberhumer] -- Move declarations of gf2 functions to right place in crc32.c [Oberhumer] -- Add cast in trees.c t avoid a warning [Oberhumer] -- Avoid some warnings in fitblk.c, gun.c, gzjoin.c in examples [Oberhumer] -- Update make_vms.com [Zinser] -- Initialize state->write in inflateReset() since copied in inflate_fast() -- Be more strict on incomplete code sets in inflate_table() and increase - ENOUGH and MAXD -- this repairs a possible security vulnerability for - invalid inflate input. Thanks to Tavis Ormandy and Markus Oberhumer for - discovering the vulnerability and providing test cases. -- Add ia64 support to configure for HP-UX [Smith] -- Add error return to gzread() for format or i/o error [Levin] -- Use malloc.h for OS/2 [Necasek] - -Changes in 1.2.2.3 (27 May 2005) -- Replace 1U constants in inflate.c and inftrees.c for 64-bit compile -- Typecast fread() return values in gzio.c [Vollant] -- Remove trailing space in minigzip.c outmode (VC++ can't deal with it) -- Fix crc check bug in gzread() after gzungetc() [Heiner] -- Add the deflateTune() function to adjust internal compression parameters -- Add a fast gzip decompressor, gun.c, to examples (use of inflateBack) -- Remove an incorrect assertion in examples/zpipe.c -- Add C++ wrapper in infback9.h [Donais] -- Fix bug in inflateCopy() when decoding fixed codes -- Note in zlib.h how much deflateSetDictionary() actually uses -- Remove USE_DICT_HEAD in deflate.c (would mess up inflate if used) -- Add _WIN32_WCE to define WIN32 in zconf.in.h [Spencer] -- Don't include stderr.h or errno.h for _WIN32_WCE in zutil.h [Spencer] -- Add gzdirect() function to indicate transparent reads -- Update contrib/minizip [Vollant] -- Fix compilation of deflate.c when both ASMV and FASTEST [Oberhumer] -- Add casts in crc32.c to avoid warnings [Oberhumer] -- Add contrib/masmx64 [Vollant] -- Update contrib/asm586, asm686, masmx86, testzlib, vstudio [Vollant] - -Changes in 1.2.2.2 (30 December 2004) -- Replace structure assignments in deflate.c and inflate.c with zmemcpy to - avoid implicit memcpy calls (portability for no-library compilation) -- Increase sprintf() buffer size in gzdopen() to allow for large numbers -- Add INFLATE_STRICT to check distances against zlib header -- Improve WinCE errno handling and comments [Chang] -- Remove comment about no gzip header processing in FAQ -- Add Z_FIXED strategy option to deflateInit2() to force fixed trees -- Add updated make_vms.com [Coghlan], update README -- Create a new "examples" directory, move gzappend.c there, add zpipe.c, - fitblk.c, gzlog.[ch], gzjoin.c, and zlib_how.html. -- Add FAQ entry and comments in deflate.c on uninitialized memory access -- Add Solaris 9 make options in configure [Gilbert] -- Allow strerror() usage in gzio.c for STDC -- Fix DecompressBuf in contrib/delphi/ZLib.pas [ManChesTer] -- Update contrib/masmx86/inffas32.asm and gvmat32.asm [Vollant] -- Use z_off_t for adler32_combine() and crc32_combine() lengths -- Make adler32() much faster for small len -- Use OS_CODE in deflate() default gzip header - -Changes in 1.2.2.1 (31 October 2004) -- Allow inflateSetDictionary() call for raw inflate -- Fix inflate header crc check bug for file names and comments -- Add deflateSetHeader() and gz_header structure for custom gzip headers -- Add inflateGetheader() to retrieve gzip headers -- Add crc32_combine() and adler32_combine() functions -- Add alloc_func, free_func, in_func, out_func to Z_PREFIX list -- Use zstreamp consistently in zlib.h (inflate_back functions) -- Remove GUNZIP condition from definition of inflate_mode in inflate.h - and in contrib/inflate86/inffast.S [Truta, Anderson] -- Add support for AMD64 in contrib/inflate86/inffas86.c [Anderson] -- Update projects/README.projects and projects/visualc6 [Truta] -- Update win32/DLL_FAQ.txt [Truta] -- Avoid warning under NO_GZCOMPRESS in gzio.c; fix typo [Truta] -- Deprecate Z_ASCII; use Z_TEXT instead [Truta] -- Use a new algorithm for setting strm->data_type in trees.c [Truta] -- Do not define an exit() prototype in zutil.c unless DEBUG defined -- Remove prototype of exit() from zutil.c, example.c, minigzip.c [Truta] -- Add comment in zlib.h for Z_NO_FLUSH parameter to deflate() -- Fix Darwin build version identification [Peterson] - -Changes in 1.2.2 (3 October 2004) -- Update zlib.h comments on gzip in-memory processing -- Set adler to 1 in inflateReset() to support Java test suite [Walles] -- Add contrib/dotzlib [Ravn] -- Update win32/DLL_FAQ.txt [Truta] -- Update contrib/minizip [Vollant] -- Move contrib/visual-basic.txt to old/ [Truta] -- Fix assembler builds in projects/visualc6/ [Truta] - -Changes in 1.2.1.2 (9 September 2004) -- Update INDEX file -- Fix trees.c to update strm->data_type (no one ever noticed!) -- Fix bug in error case in inflate.c, infback.c, and infback9.c [Brown] -- Add "volatile" to crc table flag declaration (for DYNAMIC_CRC_TABLE) -- Add limited multitasking protection to DYNAMIC_CRC_TABLE -- Add NO_vsnprintf for VMS in zutil.h [Mozilla] -- Don't declare strerror() under VMS [Mozilla] -- Add comment to DYNAMIC_CRC_TABLE to use get_crc_table() to initialize -- Update contrib/ada [Anisimkov] -- Update contrib/minizip [Vollant] -- Fix configure to not hardcode directories for Darwin [Peterson] -- Fix gzio.c to not return error on empty files [Brown] -- Fix indentation; update version in contrib/delphi/ZLib.pas and - contrib/pascal/zlibpas.pas [Truta] -- Update mkasm.bat in contrib/masmx86 [Truta] -- Update contrib/untgz [Truta] -- Add projects/README.projects [Truta] -- Add project for MS Visual C++ 6.0 in projects/visualc6 [Cadieux, Truta] -- Update win32/DLL_FAQ.txt [Truta] -- Update list of Z_PREFIX symbols in zconf.h [Randers-Pehrson, Truta] -- Remove an unnecessary assignment to curr in inftrees.c [Truta] -- Add OS/2 to exe builds in configure [Poltorak] -- Remove err dummy parameter in zlib.h [Kientzle] - -Changes in 1.2.1.1 (9 January 2004) -- Update email address in README -- Several FAQ updates -- Fix a big fat bug in inftrees.c that prevented decoding valid - dynamic blocks with only literals and no distance codes -- - Thanks to "Hot Emu" for the bug report and sample file -- Add a note to puff.c on no distance codes case. - -Changes in 1.2.1 (17 November 2003) -- Remove a tab in contrib/gzappend/gzappend.c -- Update some interfaces in contrib for new zlib functions -- Update zlib version number in some contrib entries -- Add Windows CE definition for ptrdiff_t in zutil.h [Mai, Truta] -- Support shared libraries on Hurd and KFreeBSD [Brown] -- Fix error in NO_DIVIDE option of adler32.c - -Changes in 1.2.0.8 (4 November 2003) -- Update version in contrib/delphi/ZLib.pas and contrib/pascal/zlibpas.pas -- Add experimental NO_DIVIDE #define in adler32.c - - Possibly faster on some processors (let me know if it is) -- Correct Z_BLOCK to not return on first inflate call if no wrap -- Fix strm->data_type on inflate() return to correctly indicate EOB -- Add deflatePrime() function for appending in the middle of a byte -- Add contrib/gzappend for an example of appending to a stream -- Update win32/DLL_FAQ.txt [Truta] -- Delete Turbo C comment in README [Truta] -- Improve some indentation in zconf.h [Truta] -- Fix infinite loop on bad input in configure script [Church] -- Fix gzeof() for concatenated gzip files [Johnson] -- Add example to contrib/visual-basic.txt [Michael B.] -- Add -p to mkdir's in Makefile.in [vda] -- Fix configure to properly detect presence or lack of printf functions -- Add AS400 support [Monnerat] -- Add a little Cygwin support [Wilson] - -Changes in 1.2.0.7 (21 September 2003) -- Correct some debug formats in contrib/infback9 -- Cast a type in a debug statement in trees.c -- Change search and replace delimiter in configure from % to # [Beebe] -- Update contrib/untgz to 0.2 with various fixes [Truta] -- Add build support for Amiga [Nikl] -- Remove some directories in old that have been updated to 1.2 -- Add dylib building for Mac OS X in configure and Makefile.in -- Remove old distribution stuff from Makefile -- Update README to point to DLL_FAQ.txt, and add comment on Mac OS X -- Update links in README - -Changes in 1.2.0.6 (13 September 2003) -- Minor FAQ updates -- Update contrib/minizip to 1.00 [Vollant] -- Remove test of gz functions in example.c when GZ_COMPRESS defined [Truta] -- Update POSTINC comment for 68060 [Nikl] -- Add contrib/infback9 with deflate64 decoding (unsupported) -- For MVS define NO_vsnprintf and undefine FAR [van Burik] -- Add pragma for fdopen on MVS [van Burik] - -Changes in 1.2.0.5 (8 September 2003) -- Add OF to inflateBackEnd() declaration in zlib.h -- Remember start when using gzdopen in the middle of a file -- Use internal off_t counters in gz* functions to properly handle seeks -- Perform more rigorous check for distance-too-far in inffast.c -- Add Z_BLOCK flush option to return from inflate at block boundary -- Set strm->data_type on return from inflate - - Indicate bits unused, if at block boundary, and if in last block -- Replace size_t with ptrdiff_t in crc32.c, and check for correct size -- Add condition so old NO_DEFLATE define still works for compatibility -- FAQ update regarding the Windows DLL [Truta] -- INDEX update: add qnx entry, remove aix entry [Truta] -- Install zlib.3 into mandir [Wilson] -- Move contrib/zlib_dll_FAQ.txt to win32/DLL_FAQ.txt; update [Truta] -- Adapt the zlib interface to the new DLL convention guidelines [Truta] -- Introduce ZLIB_WINAPI macro to allow the export of functions using - the WINAPI calling convention, for Visual Basic [Vollant, Truta] -- Update msdos and win32 scripts and makefiles [Truta] -- Export symbols by name, not by ordinal, in win32/zlib.def [Truta] -- Add contrib/ada [Anisimkov] -- Move asm files from contrib/vstudio/vc70_32 to contrib/asm386 [Truta] -- Rename contrib/asm386 to contrib/masmx86 [Truta, Vollant] -- Add contrib/masm686 [Truta] -- Fix offsets in contrib/inflate86 and contrib/masmx86/inffas32.asm - [Truta, Vollant] -- Update contrib/delphi; rename to contrib/pascal; add example [Truta] -- Remove contrib/delphi2; add a new contrib/delphi [Truta] -- Avoid inclusion of the nonstandard in contrib/iostream, - and fix some method prototypes [Truta] -- Fix the ZCR_SEED2 constant to avoid warnings in contrib/minizip - [Truta] -- Avoid the use of backslash (\) in contrib/minizip [Vollant] -- Fix file time handling in contrib/untgz; update makefiles [Truta] -- Update contrib/vstudio/vc70_32 to comply with the new DLL guidelines - [Vollant] -- Remove contrib/vstudio/vc15_16 [Vollant] -- Rename contrib/vstudio/vc70_32 to contrib/vstudio/vc7 [Truta] -- Update README.contrib [Truta] -- Invert the assignment order of match_head and s->prev[...] in - INSERT_STRING [Truta] -- Compare TOO_FAR with 32767 instead of 32768, to avoid 16-bit warnings - [Truta] -- Compare function pointers with 0, not with NULL or Z_NULL [Truta] -- Fix prototype of syncsearch in inflate.c [Truta] -- Introduce ASMINF macro to be enabled when using an ASM implementation - of inflate_fast [Truta] -- Change NO_DEFLATE to NO_GZCOMPRESS [Truta] -- Modify test_gzio in example.c to take a single file name as a - parameter [Truta] -- Exit the example.c program if gzopen fails [Truta] -- Add type casts around strlen in example.c [Truta] -- Remove casting to sizeof in minigzip.c; give a proper type - to the variable compared with SUFFIX_LEN [Truta] -- Update definitions of STDC and STDC99 in zconf.h [Truta] -- Synchronize zconf.h with the new Windows DLL interface [Truta] -- Use SYS16BIT instead of __32BIT__ to distinguish between - 16- and 32-bit platforms [Truta] -- Use far memory allocators in small 16-bit memory models for - Turbo C [Truta] -- Add info about the use of ASMV, ASMINF and ZLIB_WINAPI in - zlibCompileFlags [Truta] -- Cygwin has vsnprintf [Wilson] -- In Windows16, OS_CODE is 0, as in MSDOS [Truta] -- In Cygwin, OS_CODE is 3 (Unix), not 11 (Windows32) [Wilson] - -Changes in 1.2.0.4 (10 August 2003) -- Minor FAQ updates -- Be more strict when checking inflateInit2's windowBits parameter -- Change NO_GUNZIP compile option to NO_GZIP to cover deflate as well -- Add gzip wrapper option to deflateInit2 using windowBits -- Add updated QNX rule in configure and qnx directory [Bonnefoy] -- Make inflate distance-too-far checks more rigorous -- Clean up FAR usage in inflate -- Add casting to sizeof() in gzio.c and minigzip.c - -Changes in 1.2.0.3 (19 July 2003) -- Fix silly error in gzungetc() implementation [Vollant] -- Update contrib/minizip and contrib/vstudio [Vollant] -- Fix printf format in example.c -- Correct cdecl support in zconf.in.h [Anisimkov] -- Minor FAQ updates - -Changes in 1.2.0.2 (13 July 2003) -- Add ZLIB_VERNUM in zlib.h for numerical preprocessor comparisons -- Attempt to avoid warnings in crc32.c for pointer-int conversion -- Add AIX to configure, remove aix directory [Bakker] -- Add some casts to minigzip.c -- Improve checking after insecure sprintf() or vsprintf() calls -- Remove #elif's from crc32.c -- Change leave label to inf_leave in inflate.c and infback.c to avoid - library conflicts -- Remove inflate gzip decoding by default--only enable gzip decoding by - special request for stricter backward compatibility -- Add zlibCompileFlags() function to return compilation information -- More typecasting in deflate.c to avoid warnings -- Remove leading underscore from _Capital #defines [Truta] -- Fix configure to link shared library when testing -- Add some Windows CE target adjustments [Mai] -- Remove #define ZLIB_DLL in zconf.h [Vollant] -- Add zlib.3 [Rodgers] -- Update RFC URL in deflate.c and algorithm.txt [Mai] -- Add zlib_dll_FAQ.txt to contrib [Truta] -- Add UL to some constants [Truta] -- Update minizip and vstudio [Vollant] -- Remove vestigial NEED_DUMMY_RETURN from zconf.in.h -- Expand use of NO_DUMMY_DECL to avoid all dummy structures -- Added iostream3 to contrib [Schwardt] -- Replace rewind() with fseek() for WinCE [Truta] -- Improve setting of zlib format compression level flags - - Report 0 for huffman and rle strategies and for level == 0 or 1 - - Report 2 only for level == 6 -- Only deal with 64K limit when necessary at compile time [Truta] -- Allow TOO_FAR check to be turned off at compile time [Truta] -- Add gzclearerr() function [Souza] -- Add gzungetc() function - -Changes in 1.2.0.1 (17 March 2003) -- Add Z_RLE strategy for run-length encoding [Truta] - - When Z_RLE requested, restrict matches to distance one - - Update zlib.h, minigzip.c, gzopen(), gzdopen() for Z_RLE -- Correct FASTEST compilation to allow level == 0 -- Clean up what gets compiled for FASTEST -- Incorporate changes to zconf.in.h [Vollant] - - Refine detection of Turbo C need for dummy returns - - Refine ZLIB_DLL compilation - - Include additional header file on VMS for off_t typedef -- Try to use _vsnprintf where it supplants vsprintf [Vollant] -- Add some casts in inffast.c -- Enchance comments in zlib.h on what happens if gzprintf() tries to - write more than 4095 bytes before compression -- Remove unused state from inflateBackEnd() -- Remove exit(0) from minigzip.c, example.c -- Get rid of all those darn tabs -- Add "check" target to Makefile.in that does the same thing as "test" -- Add "mostlyclean" and "maintainer-clean" targets to Makefile.in -- Update contrib/inflate86 [Anderson] -- Update contrib/testzlib, contrib/vstudio, contrib/minizip [Vollant] -- Add msdos and win32 directories with makefiles [Truta] -- More additions and improvements to the FAQ - -Changes in 1.2.0 (9 March 2003) -- New and improved inflate code - - About 20% faster - - Does not allocate 32K window unless and until needed - - Automatically detects and decompresses gzip streams - - Raw inflate no longer needs an extra dummy byte at end - - Added inflateBack functions using a callback interface--even faster - than inflate, useful for file utilities (gzip, zip) - - Added inflateCopy() function to record state for random access on - externally generated deflate streams (e.g. in gzip files) - - More readable code (I hope) -- New and improved crc32() - - About 50% faster, thanks to suggestions from Rodney Brown -- Add deflateBound() and compressBound() functions -- Fix memory leak in deflateInit2() -- Permit setting dictionary for raw deflate (for parallel deflate) -- Fix const declaration for gzwrite() -- Check for some malloc() failures in gzio.c -- Fix bug in gzopen() on single-byte file 0x1f -- Fix bug in gzread() on concatenated file with 0x1f at end of buffer - and next buffer doesn't start with 0x8b -- Fix uncompress() to return Z_DATA_ERROR on truncated input -- Free memory at end of example.c -- Remove MAX #define in trees.c (conflicted with some libraries) -- Fix static const's in deflate.c, gzio.c, and zutil.[ch] -- Declare malloc() and free() in gzio.c if STDC not defined -- Use malloc() instead of calloc() in zutil.c if int big enough -- Define STDC for AIX -- Add aix/ with approach for compiling shared library on AIX -- Add HP-UX support for shared libraries in configure -- Add OpenUNIX support for shared libraries in configure -- Use $cc instead of gcc to build shared library -- Make prefix directory if needed when installing -- Correct Macintosh avoidance of typedef Byte in zconf.h -- Correct Turbo C memory allocation when under Linux -- Use libz.a instead of -lz in Makefile (assure use of compiled library) -- Update configure to check for snprintf or vsnprintf functions and their - return value, warn during make if using an insecure function -- Fix configure problem with compile-time knowledge of HAVE_UNISTD_H that - is lost when library is used--resolution is to build new zconf.h -- Documentation improvements (in zlib.h): - - Document raw deflate and inflate - - Update RFCs URL - - Point out that zlib and gzip formats are different - - Note that Z_BUF_ERROR is not fatal - - Document string limit for gzprintf() and possible buffer overflow - - Note requirement on avail_out when flushing - - Note permitted values of flush parameter of inflate() -- Add some FAQs (and even answers) to the FAQ -- Add contrib/inflate86/ for x86 faster inflate -- Add contrib/blast/ for PKWare Data Compression Library decompression -- Add contrib/puff/ simple inflate for deflate format description - -Changes in 1.1.4 (11 March 2002) -- ZFREE was repeated on same allocation on some error conditions. - This creates a security problem described in - http://www.zlib.org/advisory-2002-03-11.txt -- Returned incorrect error (Z_MEM_ERROR) on some invalid data -- Avoid accesses before window for invalid distances with inflate window - less than 32K. -- force windowBits > 8 to avoid a bug in the encoder for a window size - of 256 bytes. (A complete fix will be available in 1.1.5). - -Changes in 1.1.3 (9 July 1998) -- fix "an inflate input buffer bug that shows up on rare but persistent - occasions" (Mark) -- fix gzread and gztell for concatenated .gz files (Didier Le Botlan) -- fix gzseek(..., SEEK_SET) in write mode -- fix crc check after a gzeek (Frank Faubert) -- fix miniunzip when the last entry in a zip file is itself a zip file - (J Lillge) -- add contrib/asm586 and contrib/asm686 (Brian Raiter) - See http://www.muppetlabs.com/~breadbox/software/assembly.html -- add support for Delphi 3 in contrib/delphi (Bob Dellaca) -- add support for C++Builder 3 and Delphi 3 in contrib/delphi2 (Davide Moretti) -- do not exit prematurely in untgz if 0 at start of block (Magnus Holmgren) -- use macro EXTERN instead of extern to support DLL for BeOS (Sander Stoks) -- added a FAQ file - -- Support gzdopen on Mac with Metrowerks (Jason Linhart) -- Do not redefine Byte on Mac (Brad Pettit & Jason Linhart) -- define SEEK_END too if SEEK_SET is not defined (Albert Chin-A-Young) -- avoid some warnings with Borland C (Tom Tanner) -- fix a problem in contrib/minizip/zip.c for 16-bit MSDOS (Gilles Vollant) -- emulate utime() for WIN32 in contrib/untgz (Gilles Vollant) -- allow several arguments to configure (Tim Mooney, Frodo Looijaard) -- use libdir and includedir in Makefile.in (Tim Mooney) -- support shared libraries on OSF1 V4 (Tim Mooney) -- remove so_locations in "make clean" (Tim Mooney) -- fix maketree.c compilation error (Glenn, Mark) -- Python interface to zlib now in Python 1.5 (Jeremy Hylton) -- new Makefile.riscos (Rich Walker) -- initialize static descriptors in trees.c for embedded targets (Nick Smith) -- use "foo-gz" in example.c for RISCOS and VMS (Nick Smith) -- add the OS/2 files in Makefile.in too (Andrew Zabolotny) -- fix fdopen and halloc macros for Microsoft C 6.0 (Tom Lane) -- fix maketree.c to allow clean compilation of inffixed.h (Mark) -- fix parameter check in deflateCopy (Gunther Nikl) -- cleanup trees.c, use compressed_len only in debug mode (Christian Spieler) -- Many portability patches by Christian Spieler: - . zutil.c, zutil.h: added "const" for zmem* - . Make_vms.com: fixed some typos - . Make_vms.com: msdos/Makefile.*: removed zutil.h from some dependency lists - . msdos/Makefile.msc: remove "default rtl link library" info from obj files - . msdos/Makefile.*: use model-dependent name for the built zlib library - . msdos/Makefile.emx, nt/Makefile.emx, nt/Makefile.gcc: - new makefiles, for emx (DOS/OS2), emx&rsxnt and mingw32 (Windows 9x / NT) -- use define instead of typedef for Bytef also for MSC small/medium (Tom Lane) -- replace __far with _far for better portability (Christian Spieler, Tom Lane) -- fix test for errno.h in configure (Tim Newsham) - -Changes in 1.1.2 (19 March 98) -- added contrib/minzip, mini zip and unzip based on zlib (Gilles Vollant) - See http://www.winimage.com/zLibDll/unzip.html -- preinitialize the inflate tables for fixed codes, to make the code - completely thread safe (Mark) -- some simplifications and slight speed-up to the inflate code (Mark) -- fix gzeof on non-compressed files (Allan Schrum) -- add -std1 option in configure for OSF1 to fix gzprintf (Martin Mokrejs) -- use default value of 4K for Z_BUFSIZE for 16-bit MSDOS (Tim Wegner + Glenn) -- added os2/Makefile.def and os2/zlib.def (Andrew Zabolotny) -- add shared lib support for UNIX_SV4.2MP (MATSUURA Takanori) -- do not wrap extern "C" around system includes (Tom Lane) -- mention zlib binding for TCL in README (Andreas Kupries) -- added amiga/Makefile.pup for Amiga powerUP SAS/C PPC (Andreas Kleinert) -- allow "make install prefix=..." even after configure (Glenn Randers-Pehrson) -- allow "configure --prefix $HOME" (Tim Mooney) -- remove warnings in example.c and gzio.c (Glenn Randers-Pehrson) -- move Makefile.sas to amiga/Makefile.sas - -Changes in 1.1.1 (27 Feb 98) -- fix macros _tr_tally_* in deflate.h for debug mode (Glenn Randers-Pehrson) -- remove block truncation heuristic which had very marginal effect for zlib - (smaller lit_bufsize than in gzip 1.2.4) and degraded a little the - compression ratio on some files. This also allows inlining _tr_tally for - matches in deflate_slow. -- added msdos/Makefile.w32 for WIN32 Microsoft Visual C++ (Bob Frazier) - -Changes in 1.1.0 (24 Feb 98) -- do not return STREAM_END prematurely in inflate (John Bowler) -- revert to the zlib 1.0.8 inflate to avoid the gcc 2.8.0 bug (Jeremy Buhler) -- compile with -DFASTEST to get compression code optimized for speed only -- in minigzip, try mmap'ing the input file first (Miguel Albrecht) -- increase size of I/O buffers in minigzip.c and gzio.c (not a big gain - on Sun but significant on HP) - -- add a pointer to experimental unzip library in README (Gilles Vollant) -- initialize variable gcc in configure (Chris Herborth) - -Changes in 1.0.9 (17 Feb 1998) -- added gzputs and gzgets functions -- do not clear eof flag in gzseek (Mark Diekhans) -- fix gzseek for files in transparent mode (Mark Diekhans) -- do not assume that vsprintf returns the number of bytes written (Jens Krinke) -- replace EXPORT with ZEXPORT to avoid conflict with other programs -- added compress2 in zconf.h, zlib.def, zlib.dnt -- new asm code from Gilles Vollant in contrib/asm386 -- simplify the inflate code (Mark): - . Replace ZALLOC's in huft_build() with single ZALLOC in inflate_blocks_new() - . ZALLOC the length list in inflate_trees_fixed() instead of using stack - . ZALLOC the value area for huft_build() instead of using stack - . Simplify Z_FINISH check in inflate() - -- Avoid gcc 2.8.0 comparison bug a little differently than zlib 1.0.8 -- in inftrees.c, avoid cc -O bug on HP (Farshid Elahi) -- in zconf.h move the ZLIB_DLL stuff earlier to avoid problems with - the declaration of FAR (Gilles VOllant) -- install libz.so* with mode 755 (executable) instead of 644 (Marc Lehmann) -- read_buf buf parameter of type Bytef* instead of charf* -- zmemcpy parameters are of type Bytef*, not charf* (Joseph Strout) -- do not redeclare unlink in minigzip.c for WIN32 (John Bowler) -- fix check for presence of directories in "make install" (Ian Willis) - -Changes in 1.0.8 (27 Jan 1998) -- fixed offsets in contrib/asm386/gvmat32.asm (Gilles Vollant) -- fix gzgetc and gzputc for big endian systems (Markus Oberhumer) -- added compress2() to allow setting the compression level -- include sys/types.h to get off_t on some systems (Marc Lehmann & QingLong) -- use constant arrays for the static trees in trees.c instead of computing - them at run time (thanks to Ken Raeburn for this suggestion). To create - trees.h, compile with GEN_TREES_H and run "make test". -- check return code of example in "make test" and display result -- pass minigzip command line options to file_compress -- simplifying code of inflateSync to avoid gcc 2.8 bug - -- support CC="gcc -Wall" in configure -s (QingLong) -- avoid a flush caused by ftell in gzopen for write mode (Ken Raeburn) -- fix test for shared library support to avoid compiler warnings -- zlib.lib -> zlib.dll in msdos/zlib.rc (Gilles Vollant) -- check for TARGET_OS_MAC in addition to MACOS (Brad Pettit) -- do not use fdopen for Metrowerks on Mac (Brad Pettit)) -- add checks for gzputc and gzputc in example.c -- avoid warnings in gzio.c and deflate.c (Andreas Kleinert) -- use const for the CRC table (Ken Raeburn) -- fixed "make uninstall" for shared libraries -- use Tracev instead of Trace in infblock.c -- in example.c use correct compressed length for test_sync -- suppress +vnocompatwarnings in configure for HPUX (not always supported) - -Changes in 1.0.7 (20 Jan 1998) -- fix gzseek which was broken in write mode -- return error for gzseek to negative absolute position -- fix configure for Linux (Chun-Chung Chen) -- increase stack space for MSC (Tim Wegner) -- get_crc_table and inflateSyncPoint are EXPORTed (Gilles Vollant) -- define EXPORTVA for gzprintf (Gilles Vollant) -- added man page zlib.3 (Rick Rodgers) -- for contrib/untgz, fix makedir() and improve Makefile - -- check gzseek in write mode in example.c -- allocate extra buffer for seeks only if gzseek is actually called -- avoid signed/unsigned comparisons (Tim Wegner, Gilles Vollant) -- add inflateSyncPoint in zconf.h -- fix list of exported functions in nt/zlib.dnt and mdsos/zlib.def - -Changes in 1.0.6 (19 Jan 1998) -- add functions gzprintf, gzputc, gzgetc, gztell, gzeof, gzseek, gzrewind and - gzsetparams (thanks to Roland Giersig and Kevin Ruland for some of this code) -- Fix a deflate bug occurring only with compression level 0 (thanks to - Andy Buckler for finding this one). -- In minigzip, pass transparently also the first byte for .Z files. -- return Z_BUF_ERROR instead of Z_OK if output buffer full in uncompress() -- check Z_FINISH in inflate (thanks to Marc Schluper) -- Implement deflateCopy (thanks to Adam Costello) -- make static libraries by default in configure, add --shared option. -- move MSDOS or Windows specific files to directory msdos -- suppress the notion of partial flush to simplify the interface - (but the symbol Z_PARTIAL_FLUSH is kept for compatibility with 1.0.4) -- suppress history buffer provided by application to simplify the interface - (this feature was not implemented anyway in 1.0.4) -- next_in and avail_in must be initialized before calling inflateInit or - inflateInit2 -- add EXPORT in all exported functions (for Windows DLL) -- added Makefile.nt (thanks to Stephen Williams) -- added the unsupported "contrib" directory: - contrib/asm386/ by Gilles Vollant - 386 asm code replacing longest_match(). - contrib/iostream/ by Kevin Ruland - A C++ I/O streams interface to the zlib gz* functions - contrib/iostream2/ by Tyge Løvset - Another C++ I/O streams interface - contrib/untgz/ by "Pedro A. Aranda Guti\irrez" - A very simple tar.gz file extractor using zlib - contrib/visual-basic.txt by Carlos Rios - How to use compress(), uncompress() and the gz* functions from VB. -- pass params -f (filtered data), -h (huffman only), -1 to -9 (compression - level) in minigzip (thanks to Tom Lane) - -- use const for rommable constants in deflate -- added test for gzseek and gztell in example.c -- add undocumented function inflateSyncPoint() (hack for Paul Mackerras) -- add undocumented function zError to convert error code to string - (for Tim Smithers) -- Allow compilation of gzio with -DNO_DEFLATE to avoid the compression code. -- Use default memcpy for Symantec MSDOS compiler. -- Add EXPORT keyword for check_func (needed for Windows DLL) -- add current directory to LD_LIBRARY_PATH for "make test" -- create also a link for libz.so.1 -- added support for FUJITSU UXP/DS (thanks to Toshiaki Nomura) -- use $(SHAREDLIB) instead of libz.so in Makefile.in (for HPUX) -- added -soname for Linux in configure (Chun-Chung Chen, -- assign numbers to the exported functions in zlib.def (for Windows DLL) -- add advice in zlib.h for best usage of deflateSetDictionary -- work around compiler bug on Atari (cast Z_NULL in call of s->checkfn) -- allow compilation with ANSI keywords only enabled for TurboC in large model -- avoid "versionString"[0] (Borland bug) -- add NEED_DUMMY_RETURN for Borland -- use variable z_verbose for tracing in debug mode (L. Peter Deutsch). -- allow compilation with CC -- defined STDC for OS/2 (David Charlap) -- limit external names to 8 chars for MVS (Thomas Lund) -- in minigzip.c, use static buffers only for 16-bit systems -- fix suffix check for "minigzip -d foo.gz" -- do not return an error for the 2nd of two consecutive gzflush() (Felix Lee) -- use _fdopen instead of fdopen for MSC >= 6.0 (Thomas Fanslau) -- added makelcc.bat for lcc-win32 (Tom St Denis) -- in Makefile.dj2, use copy and del instead of install and rm (Frank Donahoe) -- Avoid expanded $Id$. Use "rcs -kb" or "cvs admin -kb" to avoid Id expansion. -- check for unistd.h in configure (for off_t) -- remove useless check parameter in inflate_blocks_free -- avoid useless assignment of s->check to itself in inflate_blocks_new -- do not flush twice in gzclose (thanks to Ken Raeburn) -- rename FOPEN as F_OPEN to avoid clash with /usr/include/sys/file.h -- use NO_ERRNO_H instead of enumeration of operating systems with errno.h -- work around buggy fclose on pipes for HP/UX -- support zlib DLL with BORLAND C++ 5.0 (thanks to Glenn Randers-Pehrson) -- fix configure if CC is already equal to gcc - -Changes in 1.0.5 (3 Jan 98) -- Fix inflate to terminate gracefully when fed corrupted or invalid data -- Use const for rommable constants in inflate -- Eliminate memory leaks on error conditions in inflate -- Removed some vestigial code in inflate -- Update web address in README - -Changes in 1.0.4 (24 Jul 96) -- In very rare conditions, deflate(s, Z_FINISH) could fail to produce an EOF - bit, so the decompressor could decompress all the correct data but went - on to attempt decompressing extra garbage data. This affected minigzip too. -- zlibVersion and gzerror return const char* (needed for DLL) -- port to RISCOS (no fdopen, no multiple dots, no unlink, no fileno) -- use z_error only for DEBUG (avoid problem with DLLs) - -Changes in 1.0.3 (2 Jul 96) -- use z_streamp instead of z_stream *, which is now a far pointer in MSDOS - small and medium models; this makes the library incompatible with previous - versions for these models. (No effect in large model or on other systems.) -- return OK instead of BUF_ERROR if previous deflate call returned with - avail_out as zero but there is nothing to do -- added memcmp for non STDC compilers -- define NO_DUMMY_DECL for more Mac compilers (.h files merged incorrectly) -- define __32BIT__ if __386__ or i386 is defined (pb. with Watcom and SCO) -- better check for 16-bit mode MSC (avoids problem with Symantec) - -Changes in 1.0.2 (23 May 96) -- added Windows DLL support -- added a function zlibVersion (for the DLL support) -- fixed declarations using Bytef in infutil.c (pb with MSDOS medium model) -- Bytef is define's instead of typedef'd only for Borland C -- avoid reading uninitialized memory in example.c -- mention in README that the zlib format is now RFC1950 -- updated Makefile.dj2 -- added algorithm.doc - -Changes in 1.0.1 (20 May 96) [1.0 skipped to avoid confusion] -- fix array overlay in deflate.c which sometimes caused bad compressed data -- fix inflate bug with empty stored block -- fix MSDOS medium model which was broken in 0.99 -- fix deflateParams() which could generate bad compressed data. -- Bytef is define'd instead of typedef'ed (work around Borland bug) -- added an INDEX file -- new makefiles for DJGPP (Makefile.dj2), 32-bit Borland (Makefile.b32), - Watcom (Makefile.wat), Amiga SAS/C (Makefile.sas) -- speed up adler32 for modern machines without auto-increment -- added -ansi for IRIX in configure -- static_init_done in trees.c is an int -- define unlink as delete for VMS -- fix configure for QNX -- add configure branch for SCO and HPUX -- avoid many warnings (unused variables, dead assignments, etc...) -- no fdopen for BeOS -- fix the Watcom fix for 32 bit mode (define FAR as empty) -- removed redefinition of Byte for MKWERKS -- work around an MWKERKS bug (incorrect merge of all .h files) - -Changes in 0.99 (27 Jan 96) -- allow preset dictionary shared between compressor and decompressor -- allow compression level 0 (no compression) -- add deflateParams in zlib.h: allow dynamic change of compression level - and compression strategy. -- test large buffers and deflateParams in example.c -- add optional "configure" to build zlib as a shared library -- suppress Makefile.qnx, use configure instead -- fixed deflate for 64-bit systems (detected on Cray) -- fixed inflate_blocks for 64-bit systems (detected on Alpha) -- declare Z_DEFLATED in zlib.h (possible parameter for deflateInit2) -- always return Z_BUF_ERROR when deflate() has nothing to do -- deflateInit and inflateInit are now macros to allow version checking -- prefix all global functions and types with z_ with -DZ_PREFIX -- make falloc completely reentrant (inftrees.c) -- fixed very unlikely race condition in ct_static_init -- free in reverse order of allocation to help memory manager -- use zlib-1.0/* instead of zlib/* inside the tar.gz -- make zlib warning-free with "gcc -O3 -Wall -Wwrite-strings -Wpointer-arith - -Wconversion -Wstrict-prototypes -Wmissing-prototypes" -- allow gzread on concatenated .gz files -- deflateEnd now returns Z_DATA_ERROR if it was premature -- deflate is finally (?) fully deterministic (no matches beyond end of input) -- Document Z_SYNC_FLUSH -- add uninstall in Makefile -- Check for __cpluplus in zlib.h -- Better test in ct_align for partial flush -- avoid harmless warnings for Borland C++ -- initialize hash_head in deflate.c -- avoid warning on fdopen (gzio.c) for HP cc -Aa -- include stdlib.h for STDC compilers -- include errno.h for Cray -- ignore error if ranlib doesn't exist -- call ranlib twice for NeXTSTEP -- use exec_prefix instead of prefix for libz.a -- renamed ct_* as _tr_* to avoid conflict with applications -- clear z->msg in inflateInit2 before any error return -- initialize opaque in example.c, gzio.c, deflate.c and inflate.c -- fixed typo in zconf.h (_GNUC__ => __GNUC__) -- check for WIN32 in zconf.h and zutil.c (avoid farmalloc in 32-bit mode) -- fix typo in Make_vms.com (f$trnlnm -> f$getsyi) -- in fcalloc, normalize pointer if size > 65520 bytes -- don't use special fcalloc for 32 bit Borland C++ -- use STDC instead of __GO32__ to avoid redeclaring exit, calloc, etc... -- use Z_BINARY instead of BINARY -- document that gzclose after gzdopen will close the file -- allow "a" as mode in gzopen. -- fix error checking in gzread -- allow skipping .gz extra-field on pipes -- added reference to Perl interface in README -- put the crc table in FAR data (I dislike more and more the medium model :) -- added get_crc_table -- added a dimension to all arrays (Borland C can't count). -- workaround Borland C bug in declaration of inflate_codes_new & inflate_fast -- guard against multiple inclusion of *.h (for precompiled header on Mac) -- Watcom C pretends to be Microsoft C small model even in 32 bit mode. -- don't use unsized arrays to avoid silly warnings by Visual C++: - warning C4746: 'inflate_mask' : unsized array treated as '__far' - (what's wrong with far data in far model?). -- define enum out of inflate_blocks_state to allow compilation with C++ - -Changes in 0.95 (16 Aug 95) -- fix MSDOS small and medium model (now easier to adapt to any compiler) -- inlined send_bits -- fix the final (:-) bug for deflate with flush (output was correct but - not completely flushed in rare occasions). -- default window size is same for compression and decompression - (it's now sufficient to set MAX_WBITS in zconf.h). -- voidp -> voidpf and voidnp -> voidp (for consistency with other - typedefs and because voidnp was not near in large model). - -Changes in 0.94 (13 Aug 95) -- support MSDOS medium model -- fix deflate with flush (could sometimes generate bad output) -- fix deflateReset (zlib header was incorrectly suppressed) -- added support for VMS -- allow a compression level in gzopen() -- gzflush now calls fflush -- For deflate with flush, flush even if no more input is provided. -- rename libgz.a as libz.a -- avoid complex expression in infcodes.c triggering Turbo C bug -- work around a problem with gcc on Alpha (in INSERT_STRING) -- don't use inline functions (problem with some gcc versions) -- allow renaming of Byte, uInt, etc... with #define. -- avoid warning about (unused) pointer before start of array in deflate.c -- avoid various warnings in gzio.c, example.c, infblock.c, adler32.c, zutil.c -- avoid reserved word 'new' in trees.c - -Changes in 0.93 (25 June 95) -- temporarily disable inline functions -- make deflate deterministic -- give enough lookahead for PARTIAL_FLUSH -- Set binary mode for stdin/stdout in minigzip.c for OS/2 -- don't even use signed char in inflate (not portable enough) -- fix inflate memory leak for segmented architectures - -Changes in 0.92 (3 May 95) -- don't assume that char is signed (problem on SGI) -- Clear bit buffer when starting a stored block -- no memcpy on Pyramid -- suppressed inftest.c -- optimized fill_window, put longest_match inline for gcc -- optimized inflate on stored blocks. -- untabify all sources to simplify patches - -Changes in 0.91 (2 May 95) -- Default MEM_LEVEL is 8 (not 9 for Unix) as documented in zlib.h -- Document the memory requirements in zconf.h -- added "make install" -- fix sync search logic in inflateSync -- deflate(Z_FULL_FLUSH) now works even if output buffer too short -- after inflateSync, don't scare people with just "lo world" -- added support for DJGPP - -Changes in 0.9 (1 May 95) -- don't assume that zalloc clears the allocated memory (the TurboC bug - was Mark's bug after all :) -- let again gzread copy uncompressed data unchanged (was working in 0.71) -- deflate(Z_FULL_FLUSH), inflateReset and inflateSync are now fully implemented -- added a test of inflateSync in example.c -- moved MAX_WBITS to zconf.h because users might want to change that. -- document explicitly that zalloc(64K) on MSDOS must return a normalized - pointer (zero offset) -- added Makefiles for Microsoft C, Turbo C, Borland C++ -- faster crc32() - -Changes in 0.8 (29 April 95) -- added fast inflate (inffast.c) -- deflate(Z_FINISH) now returns Z_STREAM_END when done. Warning: this - is incompatible with previous versions of zlib which returned Z_OK. -- work around a TurboC compiler bug (bad code for b << 0, see infutil.h) - (actually that was not a compiler bug, see 0.81 above) -- gzread no longer reads one extra byte in certain cases -- In gzio destroy(), don't reference a freed structure -- avoid many warnings for MSDOS -- avoid the ERROR symbol which is used by MS Windows - -Changes in 0.71 (14 April 95) -- Fixed more MSDOS compilation problems :( There is still a bug with - TurboC large model. - -Changes in 0.7 (14 April 95) -- Added full inflate support. -- Simplified the crc32() interface. The pre- and post-conditioning - (one's complement) is now done inside crc32(). WARNING: this is - incompatible with previous versions; see zlib.h for the new usage. - -Changes in 0.61 (12 April 95) -- workaround for a bug in TurboC. example and minigzip now work on MSDOS. - -Changes in 0.6 (11 April 95) -- added minigzip.c -- added gzdopen to reopen a file descriptor as gzFile -- added transparent reading of non-gziped files in gzread. -- fixed bug in gzread (don't read crc as data) -- fixed bug in destroy (gzio.c) (don't return Z_STREAM_END for gzclose). -- don't allocate big arrays in the stack (for MSDOS) -- fix some MSDOS compilation problems - -Changes in 0.5: -- do real compression in deflate.c. Z_PARTIAL_FLUSH is supported but - not yet Z_FULL_FLUSH. -- support decompression but only in a single step (forced Z_FINISH) -- added opaque object for zalloc and zfree. -- added deflateReset and inflateReset -- added a variable zlib_version for consistency checking. -- renamed the 'filter' parameter of deflateInit2 as 'strategy'. - Added Z_FILTERED and Z_HUFFMAN_ONLY constants. - -Changes in 0.4: -- avoid "zip" everywhere, use zlib instead of ziplib. -- suppress Z_BLOCK_FLUSH, interpret Z_PARTIAL_FLUSH as block flush - if compression method == 8. -- added adler32 and crc32 -- renamed deflateOptions as deflateInit2, call one or the other but not both -- added the method parameter for deflateInit2. -- added inflateInit2 -- simplied considerably deflateInit and inflateInit by not supporting - user-provided history buffer. This is supported only in deflateInit2 - and inflateInit2. - -Changes in 0.3: -- prefix all macro names with Z_ -- use Z_FINISH instead of deflateEnd to finish compression. -- added Z_HUFFMAN_ONLY -- added gzerror() diff --git a/dep/zlib/FAQ b/dep/zlib/FAQ deleted file mode 100644 index 99b7cf92e..000000000 --- a/dep/zlib/FAQ +++ /dev/null @@ -1,368 +0,0 @@ - - Frequently Asked Questions about zlib - - -If your question is not there, please check the zlib home page -http://zlib.net/ which may have more recent information. -The lastest zlib FAQ is at http://zlib.net/zlib_faq.html - - - 1. Is zlib Y2K-compliant? - - Yes. zlib doesn't handle dates. - - 2. Where can I get a Windows DLL version? - - The zlib sources can be compiled without change to produce a DLL. See the - file win32/DLL_FAQ.txt in the zlib distribution. Pointers to the - precompiled DLL are found in the zlib web site at http://zlib.net/ . - - 3. Where can I get a Visual Basic interface to zlib? - - See - * http://marknelson.us/1997/01/01/zlib-engine/ - * win32/DLL_FAQ.txt in the zlib distribution - - 4. compress() returns Z_BUF_ERROR. - - Make sure that before the call of compress(), the length of the compressed - buffer is equal to the available size of the compressed buffer and not - zero. For Visual Basic, check that this parameter is passed by reference - ("as any"), not by value ("as long"). - - 5. deflate() or inflate() returns Z_BUF_ERROR. - - Before making the call, make sure that avail_in and avail_out are not zero. - When setting the parameter flush equal to Z_FINISH, also make sure that - avail_out is big enough to allow processing all pending input. Note that a - Z_BUF_ERROR is not fatal--another call to deflate() or inflate() can be - made with more input or output space. A Z_BUF_ERROR may in fact be - unavoidable depending on how the functions are used, since it is not - possible to tell whether or not there is more output pending when - strm.avail_out returns with zero. See http://zlib.net/zlib_how.html for a - heavily annotated example. - - 6. Where's the zlib documentation (man pages, etc.)? - - It's in zlib.h . Examples of zlib usage are in the files test/example.c - and test/minigzip.c, with more in examples/ . - - 7. Why don't you use GNU autoconf or libtool or ...? - - Because we would like to keep zlib as a very small and simple package. - zlib is rather portable and doesn't need much configuration. - - 8. I found a bug in zlib. - - Most of the time, such problems are due to an incorrect usage of zlib. - Please try to reproduce the problem with a small program and send the - corresponding source to us at zlib@gzip.org . Do not send multi-megabyte - data files without prior agreement. - - 9. Why do I get "undefined reference to gzputc"? - - If "make test" produces something like - - example.o(.text+0x154): undefined reference to `gzputc' - - check that you don't have old files libz.* in /usr/lib, /usr/local/lib or - /usr/X11R6/lib. Remove any old versions, then do "make install". - -10. I need a Delphi interface to zlib. - - See the contrib/delphi directory in the zlib distribution. - -11. Can zlib handle .zip archives? - - Not by itself, no. See the directory contrib/minizip in the zlib - distribution. - -12. Can zlib handle .Z files? - - No, sorry. You have to spawn an uncompress or gunzip subprocess, or adapt - the code of uncompress on your own. - -13. How can I make a Unix shared library? - - By default a shared (and a static) library is built for Unix. So: - - make distclean - ./configure - make - -14. How do I install a shared zlib library on Unix? - - After the above, then: - - make install - - However, many flavors of Unix come with a shared zlib already installed. - Before going to the trouble of compiling a shared version of zlib and - trying to install it, you may want to check if it's already there! If you - can #include , it's there. The -lz option will probably link to - it. You can check the version at the top of zlib.h or with the - ZLIB_VERSION symbol defined in zlib.h . - -15. I have a question about OttoPDF. - - We are not the authors of OttoPDF. The real author is on the OttoPDF web - site: Joel Hainley, jhainley@myndkryme.com. - -16. Can zlib decode Flate data in an Adobe PDF file? - - Yes. See http://www.pdflib.com/ . To modify PDF forms, see - http://sourceforge.net/projects/acroformtool/ . - -17. Why am I getting this "register_frame_info not found" error on Solaris? - - After installing zlib 1.1.4 on Solaris 2.6, running applications using zlib - generates an error such as: - - ld.so.1: rpm: fatal: relocation error: file /usr/local/lib/libz.so: - symbol __register_frame_info: referenced symbol not found - - The symbol __register_frame_info is not part of zlib, it is generated by - the C compiler (cc or gcc). You must recompile applications using zlib - which have this problem. This problem is specific to Solaris. See - http://www.sunfreeware.com for Solaris versions of zlib and applications - using zlib. - -18. Why does gzip give an error on a file I make with compress/deflate? - - The compress and deflate functions produce data in the zlib format, which - is different and incompatible with the gzip format. The gz* functions in - zlib on the other hand use the gzip format. Both the zlib and gzip formats - use the same compressed data format internally, but have different headers - and trailers around the compressed data. - -19. Ok, so why are there two different formats? - - The gzip format was designed to retain the directory information about a - single file, such as the name and last modification date. The zlib format - on the other hand was designed for in-memory and communication channel - applications, and has a much more compact header and trailer and uses a - faster integrity check than gzip. - -20. Well that's nice, but how do I make a gzip file in memory? - - You can request that deflate write the gzip format instead of the zlib - format using deflateInit2(). You can also request that inflate decode the - gzip format using inflateInit2(). Read zlib.h for more details. - -21. Is zlib thread-safe? - - Yes. However any library routines that zlib uses and any application- - provided memory allocation routines must also be thread-safe. zlib's gz* - functions use stdio library routines, and most of zlib's functions use the - library memory allocation routines by default. zlib's *Init* functions - allow for the application to provide custom memory allocation routines. - - Of course, you should only operate on any given zlib or gzip stream from a - single thread at a time. - -22. Can I use zlib in my commercial application? - - Yes. Please read the license in zlib.h. - -23. Is zlib under the GNU license? - - No. Please read the license in zlib.h. - -24. The license says that altered source versions must be "plainly marked". So - what exactly do I need to do to meet that requirement? - - You need to change the ZLIB_VERSION and ZLIB_VERNUM #defines in zlib.h. In - particular, the final version number needs to be changed to "f", and an - identification string should be appended to ZLIB_VERSION. Version numbers - x.x.x.f are reserved for modifications to zlib by others than the zlib - maintainers. For example, if the version of the base zlib you are altering - is "1.2.3.4", then in zlib.h you should change ZLIB_VERNUM to 0x123f, and - ZLIB_VERSION to something like "1.2.3.f-zachary-mods-v3". You can also - update the version strings in deflate.c and inftrees.c. - - For altered source distributions, you should also note the origin and - nature of the changes in zlib.h, as well as in ChangeLog and README, along - with the dates of the alterations. The origin should include at least your - name (or your company's name), and an email address to contact for help or - issues with the library. - - Note that distributing a compiled zlib library along with zlib.h and - zconf.h is also a source distribution, and so you should change - ZLIB_VERSION and ZLIB_VERNUM and note the origin and nature of the changes - in zlib.h as you would for a full source distribution. - -25. Will zlib work on a big-endian or little-endian architecture, and can I - exchange compressed data between them? - - Yes and yes. - -26. Will zlib work on a 64-bit machine? - - Yes. It has been tested on 64-bit machines, and has no dependence on any - data types being limited to 32-bits in length. If you have any - difficulties, please provide a complete problem report to zlib@gzip.org - -27. Will zlib decompress data from the PKWare Data Compression Library? - - No. The PKWare DCL uses a completely different compressed data format than - does PKZIP and zlib. However, you can look in zlib's contrib/blast - directory for a possible solution to your problem. - -28. Can I access data randomly in a compressed stream? - - No, not without some preparation. If when compressing you periodically use - Z_FULL_FLUSH, carefully write all the pending data at those points, and - keep an index of those locations, then you can start decompression at those - points. You have to be careful to not use Z_FULL_FLUSH too often, since it - can significantly degrade compression. Alternatively, you can scan a - deflate stream once to generate an index, and then use that index for - random access. See examples/zran.c . - -29. Does zlib work on MVS, OS/390, CICS, etc.? - - It has in the past, but we have not heard of any recent evidence. There - were working ports of zlib 1.1.4 to MVS, but those links no longer work. - If you know of recent, successful applications of zlib on these operating - systems, please let us know. Thanks. - -30. Is there some simpler, easier to read version of inflate I can look at to - understand the deflate format? - - First off, you should read RFC 1951. Second, yes. Look in zlib's - contrib/puff directory. - -31. Does zlib infringe on any patents? - - As far as we know, no. In fact, that was originally the whole point behind - zlib. Look here for some more information: - - http://www.gzip.org/#faq11 - -32. Can zlib work with greater than 4 GB of data? - - Yes. inflate() and deflate() will process any amount of data correctly. - Each call of inflate() or deflate() is limited to input and output chunks - of the maximum value that can be stored in the compiler's "unsigned int" - type, but there is no limit to the number of chunks. Note however that the - strm.total_in and strm_total_out counters may be limited to 4 GB. These - counters are provided as a convenience and are not used internally by - inflate() or deflate(). The application can easily set up its own counters - updated after each call of inflate() or deflate() to count beyond 4 GB. - compress() and uncompress() may be limited to 4 GB, since they operate in a - single call. gzseek() and gztell() may be limited to 4 GB depending on how - zlib is compiled. See the zlibCompileFlags() function in zlib.h. - - The word "may" appears several times above since there is a 4 GB limit only - if the compiler's "long" type is 32 bits. If the compiler's "long" type is - 64 bits, then the limit is 16 exabytes. - -33. Does zlib have any security vulnerabilities? - - The only one that we are aware of is potentially in gzprintf(). If zlib is - compiled to use sprintf() or vsprintf(), then there is no protection - against a buffer overflow of an 8K string space (or other value as set by - gzbuffer()), other than the caller of gzprintf() assuring that the output - will not exceed 8K. On the other hand, if zlib is compiled to use - snprintf() or vsnprintf(), which should normally be the case, then there is - no vulnerability. The ./configure script will display warnings if an - insecure variation of sprintf() will be used by gzprintf(). Also the - zlibCompileFlags() function will return information on what variant of - sprintf() is used by gzprintf(). - - If you don't have snprintf() or vsnprintf() and would like one, you can - find a portable implementation here: - - http://www.ijs.si/software/snprintf/ - - Note that you should be using the most recent version of zlib. Versions - 1.1.3 and before were subject to a double-free vulnerability, and versions - 1.2.1 and 1.2.2 were subject to an access exception when decompressing - invalid compressed data. - -34. Is there a Java version of zlib? - - Probably what you want is to use zlib in Java. zlib is already included - as part of the Java SDK in the java.util.zip package. If you really want - a version of zlib written in the Java language, look on the zlib home - page for links: http://zlib.net/ . - -35. I get this or that compiler or source-code scanner warning when I crank it - up to maximally-pedantic. Can't you guys write proper code? - - Many years ago, we gave up attempting to avoid warnings on every compiler - in the universe. It just got to be a waste of time, and some compilers - were downright silly as well as contradicted each other. So now, we simply - make sure that the code always works. - -36. Valgrind (or some similar memory access checker) says that deflate is - performing a conditional jump that depends on an uninitialized value. - Isn't that a bug? - - No. That is intentional for performance reasons, and the output of deflate - is not affected. This only started showing up recently since zlib 1.2.x - uses malloc() by default for allocations, whereas earlier versions used - calloc(), which zeros out the allocated memory. Even though the code was - correct, versions 1.2.4 and later was changed to not stimulate these - checkers. - -37. Will zlib read the (insert any ancient or arcane format here) compressed - data format? - - Probably not. Look in the comp.compression FAQ for pointers to various - formats and associated software. - -38. How can I encrypt/decrypt zip files with zlib? - - zlib doesn't support encryption. The original PKZIP encryption is very - weak and can be broken with freely available programs. To get strong - encryption, use GnuPG, http://www.gnupg.org/ , which already includes zlib - compression. For PKZIP compatible "encryption", look at - http://www.info-zip.org/ - -39. What's the difference between the "gzip" and "deflate" HTTP 1.1 encodings? - - "gzip" is the gzip format, and "deflate" is the zlib format. They should - probably have called the second one "zlib" instead to avoid confusion with - the raw deflate compressed data format. While the HTTP 1.1 RFC 2616 - correctly points to the zlib specification in RFC 1950 for the "deflate" - transfer encoding, there have been reports of servers and browsers that - incorrectly produce or expect raw deflate data per the deflate - specification in RFC 1951, most notably Microsoft. So even though the - "deflate" transfer encoding using the zlib format would be the more - efficient approach (and in fact exactly what the zlib format was designed - for), using the "gzip" transfer encoding is probably more reliable due to - an unfortunate choice of name on the part of the HTTP 1.1 authors. - - Bottom line: use the gzip format for HTTP 1.1 encoding. - -40. Does zlib support the new "Deflate64" format introduced by PKWare? - - No. PKWare has apparently decided to keep that format proprietary, since - they have not documented it as they have previous compression formats. In - any case, the compression improvements are so modest compared to other more - modern approaches, that it's not worth the effort to implement. - -41. I'm having a problem with the zip functions in zlib, can you help? - - There are no zip functions in zlib. You are probably using minizip by - Giles Vollant, which is found in the contrib directory of zlib. It is not - part of zlib. In fact none of the stuff in contrib is part of zlib. The - files in there are not supported by the zlib authors. You need to contact - the authors of the respective contribution for help. - -42. The match.asm code in contrib is under the GNU General Public License. - Since it's part of zlib, doesn't that mean that all of zlib falls under the - GNU GPL? - - No. The files in contrib are not part of zlib. They were contributed by - other authors and are provided as a convenience to the user within the zlib - distribution. Each item in contrib has its own license. - -43. Is zlib subject to export controls? What is its ECCN? - - zlib is not subject to export controls, and so is classified as EAR99. - -44. Can you please sign these lengthy legal documents and fax them back to us - so that we can use your software in our product? - - No. Go away. Shoo. diff --git a/dep/zlib/README b/dep/zlib/README deleted file mode 100644 index 51106de47..000000000 --- a/dep/zlib/README +++ /dev/null @@ -1,115 +0,0 @@ -ZLIB DATA COMPRESSION LIBRARY - -zlib 1.2.11 is a general purpose data compression library. All the code is -thread safe. The data format used by the zlib library is described by RFCs -(Request for Comments) 1950 to 1952 in the files -http://tools.ietf.org/html/rfc1950 (zlib format), rfc1951 (deflate format) and -rfc1952 (gzip format). - -All functions of the compression library are documented in the file zlib.h -(volunteer to write man pages welcome, contact zlib@gzip.org). A usage example -of the library is given in the file test/example.c which also tests that -the library is working correctly. Another example is given in the file -test/minigzip.c. The compression library itself is composed of all source -files in the root directory. - -To compile all files and run the test program, follow the instructions given at -the top of Makefile.in. In short "./configure; make test", and if that goes -well, "make install" should work for most flavors of Unix. For Windows, use -one of the special makefiles in win32/ or contrib/vstudio/ . For VMS, use -make_vms.com. - -Questions about zlib should be sent to , or to Gilles Vollant - for the Windows DLL version. The zlib home page is -http://zlib.net/ . Before reporting a problem, please check this site to -verify that you have the latest version of zlib; otherwise get the latest -version and check whether the problem still exists or not. - -PLEASE read the zlib FAQ http://zlib.net/zlib_faq.html before asking for help. - -Mark Nelson wrote an article about zlib for the Jan. 1997 -issue of Dr. Dobb's Journal; a copy of the article is available at -http://marknelson.us/1997/01/01/zlib-engine/ . - -The changes made in version 1.2.11 are documented in the file ChangeLog. - -Unsupported third party contributions are provided in directory contrib/ . - -zlib is available in Java using the java.util.zip package, documented at -http://java.sun.com/developer/technicalArticles/Programming/compression/ . - -A Perl interface to zlib written by Paul Marquess is available -at CPAN (Comprehensive Perl Archive Network) sites, including -http://search.cpan.org/~pmqs/IO-Compress-Zlib/ . - -A Python interface to zlib written by A.M. Kuchling is -available in Python 1.5 and later versions, see -http://docs.python.org/library/zlib.html . - -zlib is built into tcl: http://wiki.tcl.tk/4610 . - -An experimental package to read and write files in .zip format, written on top -of zlib by Gilles Vollant , is available in the -contrib/minizip directory of zlib. - - -Notes for some targets: - -- For Windows DLL versions, please see win32/DLL_FAQ.txt - -- For 64-bit Irix, deflate.c must be compiled without any optimization. With - -O, one libpng test fails. The test works in 32 bit mode (with the -n32 - compiler flag). The compiler bug has been reported to SGI. - -- zlib doesn't work with gcc 2.6.3 on a DEC 3000/300LX under OSF/1 2.1 it works - when compiled with cc. - -- On Digital Unix 4.0D (formely OSF/1) on AlphaServer, the cc option -std1 is - necessary to get gzprintf working correctly. This is done by configure. - -- zlib doesn't work on HP-UX 9.05 with some versions of /bin/cc. It works with - other compilers. Use "make test" to check your compiler. - -- gzdopen is not supported on RISCOS or BEOS. - -- For PalmOs, see http://palmzlib.sourceforge.net/ - - -Acknowledgments: - - The deflate format used by zlib was defined by Phil Katz. The deflate and - zlib specifications were written by L. Peter Deutsch. Thanks to all the - people who reported problems and suggested various improvements in zlib; they - are too numerous to cite here. - -Copyright notice: - - (C) 1995-2017 Jean-loup Gailly and Mark Adler - - This software is provided 'as-is', without any express or implied - warranty. In no event will the authors be held liable for any damages - arising from the use of this software. - - Permission is granted to anyone to use this software for any purpose, - including commercial applications, and to alter it and redistribute it - freely, subject to the following restrictions: - - 1. The origin of this software must not be misrepresented; you must not - claim that you wrote the original software. If you use this software - in a product, an acknowledgment in the product documentation would be - appreciated but is not required. - 2. Altered source versions must be plainly marked as such, and must not be - misrepresented as being the original software. - 3. This notice may not be removed or altered from any source distribution. - - Jean-loup Gailly Mark Adler - jloup@gzip.org madler@alumni.caltech.edu - -If you use the zlib library in a product, we would appreciate *not* receiving -lengthy legal documents to sign. The sources are provided for free but without -warranty of any kind. The library has been entirely written by Jean-loup -Gailly and Mark Adler; it does not include third-party code. - -If you redistribute modified sources, we would appreciate that you include in -the file ChangeLog history information documenting your changes. Please read -the FAQ for more information on the distribution of modified source versions. diff --git a/dep/zlib/include/zconf.h b/dep/zlib/include/zconf.h deleted file mode 100644 index 5e1d68a00..000000000 --- a/dep/zlib/include/zconf.h +++ /dev/null @@ -1,534 +0,0 @@ -/* zconf.h -- configuration of the zlib compression library - * Copyright (C) 1995-2016 Jean-loup Gailly, Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* @(#) $Id$ */ - -#ifndef ZCONF_H -#define ZCONF_H - -/* - * If you *really* need a unique prefix for all types and library functions, - * compile with -DZ_PREFIX. The "standard" zlib should be compiled without it. - * Even better than compiling with -DZ_PREFIX would be to use configure to set - * this permanently in zconf.h using "./configure --zprefix". - */ -#ifdef Z_PREFIX /* may be set to #if 1 by ./configure */ -# define Z_PREFIX_SET - -/* all linked symbols and init macros */ -# define _dist_code z__dist_code -# define _length_code z__length_code -# define _tr_align z__tr_align -# define _tr_flush_bits z__tr_flush_bits -# define _tr_flush_block z__tr_flush_block -# define _tr_init z__tr_init -# define _tr_stored_block z__tr_stored_block -# define _tr_tally z__tr_tally -# define adler32 z_adler32 -# define adler32_combine z_adler32_combine -# define adler32_combine64 z_adler32_combine64 -# define adler32_z z_adler32_z -# ifndef Z_SOLO -# define compress z_compress -# define compress2 z_compress2 -# define compressBound z_compressBound -# endif -# define crc32 z_crc32 -# define crc32_combine z_crc32_combine -# define crc32_combine64 z_crc32_combine64 -# define crc32_z z_crc32_z -# define deflate z_deflate -# define deflateBound z_deflateBound -# define deflateCopy z_deflateCopy -# define deflateEnd z_deflateEnd -# define deflateGetDictionary z_deflateGetDictionary -# define deflateInit z_deflateInit -# define deflateInit2 z_deflateInit2 -# define deflateInit2_ z_deflateInit2_ -# define deflateInit_ z_deflateInit_ -# define deflateParams z_deflateParams -# define deflatePending z_deflatePending -# define deflatePrime z_deflatePrime -# define deflateReset z_deflateReset -# define deflateResetKeep z_deflateResetKeep -# define deflateSetDictionary z_deflateSetDictionary -# define deflateSetHeader z_deflateSetHeader -# define deflateTune z_deflateTune -# define deflate_copyright z_deflate_copyright -# define get_crc_table z_get_crc_table -# ifndef Z_SOLO -# define gz_error z_gz_error -# define gz_intmax z_gz_intmax -# define gz_strwinerror z_gz_strwinerror -# define gzbuffer z_gzbuffer -# define gzclearerr z_gzclearerr -# define gzclose z_gzclose -# define gzclose_r z_gzclose_r -# define gzclose_w z_gzclose_w -# define gzdirect z_gzdirect -# define gzdopen z_gzdopen -# define gzeof z_gzeof -# define gzerror z_gzerror -# define gzflush z_gzflush -# define gzfread z_gzfread -# define gzfwrite z_gzfwrite -# define gzgetc z_gzgetc -# define gzgetc_ z_gzgetc_ -# define gzgets z_gzgets -# define gzoffset z_gzoffset -# define gzoffset64 z_gzoffset64 -# define gzopen z_gzopen -# define gzopen64 z_gzopen64 -# ifdef _WIN32 -# define gzopen_w z_gzopen_w -# endif -# define gzprintf z_gzprintf -# define gzputc z_gzputc -# define gzputs z_gzputs -# define gzread z_gzread -# define gzrewind z_gzrewind -# define gzseek z_gzseek -# define gzseek64 z_gzseek64 -# define gzsetparams z_gzsetparams -# define gztell z_gztell -# define gztell64 z_gztell64 -# define gzungetc z_gzungetc -# define gzvprintf z_gzvprintf -# define gzwrite z_gzwrite -# endif -# define inflate z_inflate -# define inflateBack z_inflateBack -# define inflateBackEnd z_inflateBackEnd -# define inflateBackInit z_inflateBackInit -# define inflateBackInit_ z_inflateBackInit_ -# define inflateCodesUsed z_inflateCodesUsed -# define inflateCopy z_inflateCopy -# define inflateEnd z_inflateEnd -# define inflateGetDictionary z_inflateGetDictionary -# define inflateGetHeader z_inflateGetHeader -# define inflateInit z_inflateInit -# define inflateInit2 z_inflateInit2 -# define inflateInit2_ z_inflateInit2_ -# define inflateInit_ z_inflateInit_ -# define inflateMark z_inflateMark -# define inflatePrime z_inflatePrime -# define inflateReset z_inflateReset -# define inflateReset2 z_inflateReset2 -# define inflateResetKeep z_inflateResetKeep -# define inflateSetDictionary z_inflateSetDictionary -# define inflateSync z_inflateSync -# define inflateSyncPoint z_inflateSyncPoint -# define inflateUndermine z_inflateUndermine -# define inflateValidate z_inflateValidate -# define inflate_copyright z_inflate_copyright -# define inflate_fast z_inflate_fast -# define inflate_table z_inflate_table -# ifndef Z_SOLO -# define uncompress z_uncompress -# define uncompress2 z_uncompress2 -# endif -# define zError z_zError -# ifndef Z_SOLO -# define zcalloc z_zcalloc -# define zcfree z_zcfree -# endif -# define zlibCompileFlags z_zlibCompileFlags -# define zlibVersion z_zlibVersion - -/* all zlib typedefs in zlib.h and zconf.h */ -# define Byte z_Byte -# define Bytef z_Bytef -# define alloc_func z_alloc_func -# define charf z_charf -# define free_func z_free_func -# ifndef Z_SOLO -# define gzFile z_gzFile -# endif -# define gz_header z_gz_header -# define gz_headerp z_gz_headerp -# define in_func z_in_func -# define intf z_intf -# define out_func z_out_func -# define uInt z_uInt -# define uIntf z_uIntf -# define uLong z_uLong -# define uLongf z_uLongf -# define voidp z_voidp -# define voidpc z_voidpc -# define voidpf z_voidpf - -/* all zlib structs in zlib.h and zconf.h */ -# define gz_header_s z_gz_header_s -# define internal_state z_internal_state - -#endif - -#if defined(__MSDOS__) && !defined(MSDOS) -# define MSDOS -#endif -#if (defined(OS_2) || defined(__OS2__)) && !defined(OS2) -# define OS2 -#endif -#if defined(_WINDOWS) && !defined(WINDOWS) -# define WINDOWS -#endif -#if defined(_WIN32) || defined(_WIN32_WCE) || defined(__WIN32__) -# ifndef WIN32 -# define WIN32 -# endif -#endif -#if (defined(MSDOS) || defined(OS2) || defined(WINDOWS)) && !defined(WIN32) -# if !defined(__GNUC__) && !defined(__FLAT__) && !defined(__386__) -# ifndef SYS16BIT -# define SYS16BIT -# endif -# endif -#endif - -/* - * Compile with -DMAXSEG_64K if the alloc function cannot allocate more - * than 64k bytes at a time (needed on systems with 16-bit int). - */ -#ifdef SYS16BIT -# define MAXSEG_64K -#endif -#ifdef MSDOS -# define UNALIGNED_OK -#endif - -#ifdef __STDC_VERSION__ -# ifndef STDC -# define STDC -# endif -# if __STDC_VERSION__ >= 199901L -# ifndef STDC99 -# define STDC99 -# endif -# endif -#endif -#if !defined(STDC) && (defined(__STDC__) || defined(__cplusplus)) -# define STDC -#endif -#if !defined(STDC) && (defined(__GNUC__) || defined(__BORLANDC__)) -# define STDC -#endif -#if !defined(STDC) && (defined(MSDOS) || defined(WINDOWS) || defined(WIN32)) -# define STDC -#endif -#if !defined(STDC) && (defined(OS2) || defined(__HOS_AIX__)) -# define STDC -#endif - -#if defined(__OS400__) && !defined(STDC) /* iSeries (formerly AS/400). */ -# define STDC -#endif - -#ifndef STDC -# ifndef const /* cannot use !defined(STDC) && !defined(const) on Mac */ -# define const /* note: need a more gentle solution here */ -# endif -#endif - -#if defined(ZLIB_CONST) && !defined(z_const) -# define z_const const -#else -# define z_const -#endif - -#ifdef Z_SOLO - typedef unsigned long z_size_t; -#else -# define z_longlong long long -# if defined(NO_SIZE_T) - typedef unsigned NO_SIZE_T z_size_t; -# elif defined(STDC) -# include - typedef size_t z_size_t; -# else - typedef unsigned long z_size_t; -# endif -# undef z_longlong -#endif - -/* Maximum value for memLevel in deflateInit2 */ -#ifndef MAX_MEM_LEVEL -# ifdef MAXSEG_64K -# define MAX_MEM_LEVEL 8 -# else -# define MAX_MEM_LEVEL 9 -# endif -#endif - -/* Maximum value for windowBits in deflateInit2 and inflateInit2. - * WARNING: reducing MAX_WBITS makes minigzip unable to extract .gz files - * created by gzip. (Files created by minigzip can still be extracted by - * gzip.) - */ -#ifndef MAX_WBITS -# define MAX_WBITS 15 /* 32K LZ77 window */ -#endif - -/* The memory requirements for deflate are (in bytes): - (1 << (windowBits+2)) + (1 << (memLevel+9)) - that is: 128K for windowBits=15 + 128K for memLevel = 8 (default values) - plus a few kilobytes for small objects. For example, if you want to reduce - the default memory requirements from 256K to 128K, compile with - make CFLAGS="-O -DMAX_WBITS=14 -DMAX_MEM_LEVEL=7" - Of course this will generally degrade compression (there's no free lunch). - - The memory requirements for inflate are (in bytes) 1 << windowBits - that is, 32K for windowBits=15 (default value) plus about 7 kilobytes - for small objects. -*/ - - /* Type declarations */ - -#ifndef OF /* function prototypes */ -# ifdef STDC -# define OF(args) args -# else -# define OF(args) () -# endif -#endif - -#ifndef Z_ARG /* function prototypes for stdarg */ -# if defined(STDC) || defined(Z_HAVE_STDARG_H) -# define Z_ARG(args) args -# else -# define Z_ARG(args) () -# endif -#endif - -/* The following definitions for FAR are needed only for MSDOS mixed - * model programming (small or medium model with some far allocations). - * This was tested only with MSC; for other MSDOS compilers you may have - * to define NO_MEMCPY in zutil.h. If you don't need the mixed model, - * just define FAR to be empty. - */ -#ifdef SYS16BIT -# if defined(M_I86SM) || defined(M_I86MM) - /* MSC small or medium model */ -# define SMALL_MEDIUM -# ifdef _MSC_VER -# define FAR _far -# else -# define FAR far -# endif -# endif -# if (defined(__SMALL__) || defined(__MEDIUM__)) - /* Turbo C small or medium model */ -# define SMALL_MEDIUM -# ifdef __BORLANDC__ -# define FAR _far -# else -# define FAR far -# endif -# endif -#endif - -#if defined(WINDOWS) || defined(WIN32) - /* If building or using zlib as a DLL, define ZLIB_DLL. - * This is not mandatory, but it offers a little performance increase. - */ -# ifdef ZLIB_DLL -# if defined(WIN32) && (!defined(__BORLANDC__) || (__BORLANDC__ >= 0x500)) -# ifdef ZLIB_INTERNAL -# define ZEXTERN extern __declspec(dllexport) -# else -# define ZEXTERN extern __declspec(dllimport) -# endif -# endif -# endif /* ZLIB_DLL */ - /* If building or using zlib with the WINAPI/WINAPIV calling convention, - * define ZLIB_WINAPI. - * Caution: the standard ZLIB1.DLL is NOT compiled using ZLIB_WINAPI. - */ -# ifdef ZLIB_WINAPI -# ifdef FAR -# undef FAR -# endif -# include - /* No need for _export, use ZLIB.DEF instead. */ - /* For complete Windows compatibility, use WINAPI, not __stdcall. */ -# define ZEXPORT WINAPI -# ifdef WIN32 -# define ZEXPORTVA WINAPIV -# else -# define ZEXPORTVA FAR CDECL -# endif -# endif -#endif - -#if defined (__BEOS__) -# ifdef ZLIB_DLL -# ifdef ZLIB_INTERNAL -# define ZEXPORT __declspec(dllexport) -# define ZEXPORTVA __declspec(dllexport) -# else -# define ZEXPORT __declspec(dllimport) -# define ZEXPORTVA __declspec(dllimport) -# endif -# endif -#endif - -#ifndef ZEXTERN -# define ZEXTERN extern -#endif -#ifndef ZEXPORT -# define ZEXPORT -#endif -#ifndef ZEXPORTVA -# define ZEXPORTVA -#endif - -#ifndef FAR -# define FAR -#endif - -#if !defined(__MACTYPES__) -typedef unsigned char Byte; /* 8 bits */ -#endif -typedef unsigned int uInt; /* 16 bits or more */ -typedef unsigned long uLong; /* 32 bits or more */ - -#ifdef SMALL_MEDIUM - /* Borland C/C++ and some old MSC versions ignore FAR inside typedef */ -# define Bytef Byte FAR -#else - typedef Byte FAR Bytef; -#endif -typedef char FAR charf; -typedef int FAR intf; -typedef uInt FAR uIntf; -typedef uLong FAR uLongf; - -#ifdef STDC - typedef void const *voidpc; - typedef void FAR *voidpf; - typedef void *voidp; -#else - typedef Byte const *voidpc; - typedef Byte FAR *voidpf; - typedef Byte *voidp; -#endif - -#if !defined(Z_U4) && !defined(Z_SOLO) && defined(STDC) -# include -# if (UINT_MAX == 0xffffffffUL) -# define Z_U4 unsigned -# elif (ULONG_MAX == 0xffffffffUL) -# define Z_U4 unsigned long -# elif (USHRT_MAX == 0xffffffffUL) -# define Z_U4 unsigned short -# endif -#endif - -#ifdef Z_U4 - typedef Z_U4 z_crc_t; -#else - typedef unsigned long z_crc_t; -#endif - -#ifdef HAVE_UNISTD_H /* may be set to #if 1 by ./configure */ -# define Z_HAVE_UNISTD_H -#endif - -#ifdef HAVE_STDARG_H /* may be set to #if 1 by ./configure */ -# define Z_HAVE_STDARG_H -#endif - -#ifdef STDC -# ifndef Z_SOLO -# include /* for off_t */ -# endif -#endif - -#if defined(STDC) || defined(Z_HAVE_STDARG_H) -# ifndef Z_SOLO -# include /* for va_list */ -# endif -#endif - -#ifdef _WIN32 -# ifndef Z_SOLO -# include /* for wchar_t */ -# endif -#endif - -/* a little trick to accommodate both "#define _LARGEFILE64_SOURCE" and - * "#define _LARGEFILE64_SOURCE 1" as requesting 64-bit operations, (even - * though the former does not conform to the LFS document), but considering - * both "#undef _LARGEFILE64_SOURCE" and "#define _LARGEFILE64_SOURCE 0" as - * equivalently requesting no 64-bit operations - */ -#if defined(_LARGEFILE64_SOURCE) && -_LARGEFILE64_SOURCE - -1 == 1 -# undef _LARGEFILE64_SOURCE -#endif - -#if defined(__WATCOMC__) && !defined(Z_HAVE_UNISTD_H) -# define Z_HAVE_UNISTD_H -#endif -#ifndef Z_SOLO -# if defined(Z_HAVE_UNISTD_H) || defined(_LARGEFILE64_SOURCE) -# include /* for SEEK_*, off_t, and _LFS64_LARGEFILE */ -# ifdef VMS -# include /* for off_t */ -# endif -# ifndef z_off_t -# define z_off_t off_t -# endif -# endif -#endif - -#if defined(_LFS64_LARGEFILE) && _LFS64_LARGEFILE-0 -# define Z_LFS64 -#endif - -#if defined(_LARGEFILE64_SOURCE) && defined(Z_LFS64) -# define Z_LARGE64 -#endif - -#if defined(_FILE_OFFSET_BITS) && _FILE_OFFSET_BITS-0 == 64 && defined(Z_LFS64) -# define Z_WANT64 -#endif - -#if !defined(SEEK_SET) && !defined(Z_SOLO) -# define SEEK_SET 0 /* Seek from beginning of file. */ -# define SEEK_CUR 1 /* Seek from current position. */ -# define SEEK_END 2 /* Set file pointer to EOF plus "offset" */ -#endif - -#ifndef z_off_t -# define z_off_t long -#endif - -#if !defined(_WIN32) && defined(Z_LARGE64) -# define z_off64_t off64_t -#else -# if defined(_WIN32) && !defined(__GNUC__) && !defined(Z_SOLO) -# define z_off64_t __int64 -# else -# define z_off64_t z_off_t -# endif -#endif - -/* MVS linker does not support external names larger than 8 bytes */ -#if defined(__MVS__) - #pragma map(deflateInit_,"DEIN") - #pragma map(deflateInit2_,"DEIN2") - #pragma map(deflateEnd,"DEEND") - #pragma map(deflateBound,"DEBND") - #pragma map(inflateInit_,"ININ") - #pragma map(inflateInit2_,"ININ2") - #pragma map(inflateEnd,"INEND") - #pragma map(inflateSync,"INSY") - #pragma map(inflateSetDictionary,"INSEDI") - #pragma map(compressBound,"CMBND") - #pragma map(inflate_table,"INTABL") - #pragma map(inflate_fast,"INFA") - #pragma map(inflate_copyright,"INCOPY") -#endif - -#endif /* ZCONF_H */ diff --git a/dep/zlib/include/zlib.h b/dep/zlib/include/zlib.h deleted file mode 100644 index f09cdaf1e..000000000 --- a/dep/zlib/include/zlib.h +++ /dev/null @@ -1,1912 +0,0 @@ -/* zlib.h -- interface of the 'zlib' general purpose compression library - version 1.2.11, January 15th, 2017 - - Copyright (C) 1995-2017 Jean-loup Gailly and Mark Adler - - This software is provided 'as-is', without any express or implied - warranty. In no event will the authors be held liable for any damages - arising from the use of this software. - - Permission is granted to anyone to use this software for any purpose, - including commercial applications, and to alter it and redistribute it - freely, subject to the following restrictions: - - 1. The origin of this software must not be misrepresented; you must not - claim that you wrote the original software. If you use this software - in a product, an acknowledgment in the product documentation would be - appreciated but is not required. - 2. Altered source versions must be plainly marked as such, and must not be - misrepresented as being the original software. - 3. This notice may not be removed or altered from any source distribution. - - Jean-loup Gailly Mark Adler - jloup@gzip.org madler@alumni.caltech.edu - - - The data format used by the zlib library is described by RFCs (Request for - Comments) 1950 to 1952 in the files http://tools.ietf.org/html/rfc1950 - (zlib format), rfc1951 (deflate format) and rfc1952 (gzip format). -*/ - -#ifndef ZLIB_H -#define ZLIB_H - -#include "zconf.h" - -#ifdef __cplusplus -extern "C" { -#endif - -#define ZLIB_VERSION "1.2.11" -#define ZLIB_VERNUM 0x12b0 -#define ZLIB_VER_MAJOR 1 -#define ZLIB_VER_MINOR 2 -#define ZLIB_VER_REVISION 11 -#define ZLIB_VER_SUBREVISION 0 - -/* - The 'zlib' compression library provides in-memory compression and - decompression functions, including integrity checks of the uncompressed data. - This version of the library supports only one compression method (deflation) - but other algorithms will be added later and will have the same stream - interface. - - Compression can be done in a single step if the buffers are large enough, - or can be done by repeated calls of the compression function. In the latter - case, the application must provide more input and/or consume the output - (providing more output space) before each call. - - The compressed data format used by default by the in-memory functions is - the zlib format, which is a zlib wrapper documented in RFC 1950, wrapped - around a deflate stream, which is itself documented in RFC 1951. - - The library also supports reading and writing files in gzip (.gz) format - with an interface similar to that of stdio using the functions that start - with "gz". The gzip format is different from the zlib format. gzip is a - gzip wrapper, documented in RFC 1952, wrapped around a deflate stream. - - This library can optionally read and write gzip and raw deflate streams in - memory as well. - - The zlib format was designed to be compact and fast for use in memory - and on communications channels. The gzip format was designed for single- - file compression on file systems, has a larger header than zlib to maintain - directory information, and uses a different, slower check method than zlib. - - The library does not install any signal handler. The decoder checks - the consistency of the compressed data, so the library should never crash - even in the case of corrupted input. -*/ - -typedef voidpf (*alloc_func) OF((voidpf opaque, uInt items, uInt size)); -typedef void (*free_func) OF((voidpf opaque, voidpf address)); - -struct internal_state; - -typedef struct z_stream_s { - z_const Bytef *next_in; /* next input byte */ - uInt avail_in; /* number of bytes available at next_in */ - uLong total_in; /* total number of input bytes read so far */ - - Bytef *next_out; /* next output byte will go here */ - uInt avail_out; /* remaining free space at next_out */ - uLong total_out; /* total number of bytes output so far */ - - z_const char *msg; /* last error message, NULL if no error */ - struct internal_state FAR *state; /* not visible by applications */ - - alloc_func zalloc; /* used to allocate the internal state */ - free_func zfree; /* used to free the internal state */ - voidpf opaque; /* private data object passed to zalloc and zfree */ - - int data_type; /* best guess about the data type: binary or text - for deflate, or the decoding state for inflate */ - uLong adler; /* Adler-32 or CRC-32 value of the uncompressed data */ - uLong reserved; /* reserved for future use */ -} z_stream; - -typedef z_stream FAR *z_streamp; - -/* - gzip header information passed to and from zlib routines. See RFC 1952 - for more details on the meanings of these fields. -*/ -typedef struct gz_header_s { - int text; /* true if compressed data believed to be text */ - uLong time; /* modification time */ - int xflags; /* extra flags (not used when writing a gzip file) */ - int os; /* operating system */ - Bytef *extra; /* pointer to extra field or Z_NULL if none */ - uInt extra_len; /* extra field length (valid if extra != Z_NULL) */ - uInt extra_max; /* space at extra (only when reading header) */ - Bytef *name; /* pointer to zero-terminated file name or Z_NULL */ - uInt name_max; /* space at name (only when reading header) */ - Bytef *comment; /* pointer to zero-terminated comment or Z_NULL */ - uInt comm_max; /* space at comment (only when reading header) */ - int hcrc; /* true if there was or will be a header crc */ - int done; /* true when done reading gzip header (not used - when writing a gzip file) */ -} gz_header; - -typedef gz_header FAR *gz_headerp; - -/* - The application must update next_in and avail_in when avail_in has dropped - to zero. It must update next_out and avail_out when avail_out has dropped - to zero. The application must initialize zalloc, zfree and opaque before - calling the init function. All other fields are set by the compression - library and must not be updated by the application. - - The opaque value provided by the application will be passed as the first - parameter for calls of zalloc and zfree. This can be useful for custom - memory management. The compression library attaches no meaning to the - opaque value. - - zalloc must return Z_NULL if there is not enough memory for the object. - If zlib is used in a multi-threaded application, zalloc and zfree must be - thread safe. In that case, zlib is thread-safe. When zalloc and zfree are - Z_NULL on entry to the initialization function, they are set to internal - routines that use the standard library functions malloc() and free(). - - On 16-bit systems, the functions zalloc and zfree must be able to allocate - exactly 65536 bytes, but will not be required to allocate more than this if - the symbol MAXSEG_64K is defined (see zconf.h). WARNING: On MSDOS, pointers - returned by zalloc for objects of exactly 65536 bytes *must* have their - offset normalized to zero. The default allocation function provided by this - library ensures this (see zutil.c). To reduce memory requirements and avoid - any allocation of 64K objects, at the expense of compression ratio, compile - the library with -DMAX_WBITS=14 (see zconf.h). - - The fields total_in and total_out can be used for statistics or progress - reports. After compression, total_in holds the total size of the - uncompressed data and may be saved for use by the decompressor (particularly - if the decompressor wants to decompress everything in a single step). -*/ - - /* constants */ - -#define Z_NO_FLUSH 0 -#define Z_PARTIAL_FLUSH 1 -#define Z_SYNC_FLUSH 2 -#define Z_FULL_FLUSH 3 -#define Z_FINISH 4 -#define Z_BLOCK 5 -#define Z_TREES 6 -/* Allowed flush values; see deflate() and inflate() below for details */ - -#define Z_OK 0 -#define Z_STREAM_END 1 -#define Z_NEED_DICT 2 -#define Z_ERRNO (-1) -#define Z_STREAM_ERROR (-2) -#define Z_DATA_ERROR (-3) -#define Z_MEM_ERROR (-4) -#define Z_BUF_ERROR (-5) -#define Z_VERSION_ERROR (-6) -/* Return codes for the compression/decompression functions. Negative values - * are errors, positive values are used for special but normal events. - */ - -#define Z_NO_COMPRESSION 0 -#define Z_BEST_SPEED 1 -#define Z_BEST_COMPRESSION 9 -#define Z_DEFAULT_COMPRESSION (-1) -/* compression levels */ - -#define Z_FILTERED 1 -#define Z_HUFFMAN_ONLY 2 -#define Z_RLE 3 -#define Z_FIXED 4 -#define Z_DEFAULT_STRATEGY 0 -/* compression strategy; see deflateInit2() below for details */ - -#define Z_BINARY 0 -#define Z_TEXT 1 -#define Z_ASCII Z_TEXT /* for compatibility with 1.2.2 and earlier */ -#define Z_UNKNOWN 2 -/* Possible values of the data_type field for deflate() */ - -#define Z_DEFLATED 8 -/* The deflate compression method (the only one supported in this version) */ - -#define Z_NULL 0 /* for initializing zalloc, zfree, opaque */ - -#define zlib_version zlibVersion() -/* for compatibility with versions < 1.0.2 */ - - - /* basic functions */ - -ZEXTERN const char * ZEXPORT zlibVersion OF((void)); -/* The application can compare zlibVersion and ZLIB_VERSION for consistency. - If the first character differs, the library code actually used is not - compatible with the zlib.h header file used by the application. This check - is automatically made by deflateInit and inflateInit. - */ - -/* -ZEXTERN int ZEXPORT deflateInit OF((z_streamp strm, int level)); - - Initializes the internal stream state for compression. The fields - zalloc, zfree and opaque must be initialized before by the caller. If - zalloc and zfree are set to Z_NULL, deflateInit updates them to use default - allocation functions. - - The compression level must be Z_DEFAULT_COMPRESSION, or between 0 and 9: - 1 gives best speed, 9 gives best compression, 0 gives no compression at all - (the input data is simply copied a block at a time). Z_DEFAULT_COMPRESSION - requests a default compromise between speed and compression (currently - equivalent to level 6). - - deflateInit returns Z_OK if success, Z_MEM_ERROR if there was not enough - memory, Z_STREAM_ERROR if level is not a valid compression level, or - Z_VERSION_ERROR if the zlib library version (zlib_version) is incompatible - with the version assumed by the caller (ZLIB_VERSION). msg is set to null - if there is no error message. deflateInit does not perform any compression: - this will be done by deflate(). -*/ - - -ZEXTERN int ZEXPORT deflate OF((z_streamp strm, int flush)); -/* - deflate compresses as much data as possible, and stops when the input - buffer becomes empty or the output buffer becomes full. It may introduce - some output latency (reading input without producing any output) except when - forced to flush. - - The detailed semantics are as follows. deflate performs one or both of the - following actions: - - - Compress more input starting at next_in and update next_in and avail_in - accordingly. If not all input can be processed (because there is not - enough room in the output buffer), next_in and avail_in are updated and - processing will resume at this point for the next call of deflate(). - - - Generate more output starting at next_out and update next_out and avail_out - accordingly. This action is forced if the parameter flush is non zero. - Forcing flush frequently degrades the compression ratio, so this parameter - should be set only when necessary. Some output may be provided even if - flush is zero. - - Before the call of deflate(), the application should ensure that at least - one of the actions is possible, by providing more input and/or consuming more - output, and updating avail_in or avail_out accordingly; avail_out should - never be zero before the call. The application can consume the compressed - output when it wants, for example when the output buffer is full (avail_out - == 0), or after each call of deflate(). If deflate returns Z_OK and with - zero avail_out, it must be called again after making room in the output - buffer because there might be more output pending. See deflatePending(), - which can be used if desired to determine whether or not there is more ouput - in that case. - - Normally the parameter flush is set to Z_NO_FLUSH, which allows deflate to - decide how much data to accumulate before producing output, in order to - maximize compression. - - If the parameter flush is set to Z_SYNC_FLUSH, all pending output is - flushed to the output buffer and the output is aligned on a byte boundary, so - that the decompressor can get all input data available so far. (In - particular avail_in is zero after the call if enough output space has been - provided before the call.) Flushing may degrade compression for some - compression algorithms and so it should be used only when necessary. This - completes the current deflate block and follows it with an empty stored block - that is three bits plus filler bits to the next byte, followed by four bytes - (00 00 ff ff). - - If flush is set to Z_PARTIAL_FLUSH, all pending output is flushed to the - output buffer, but the output is not aligned to a byte boundary. All of the - input data so far will be available to the decompressor, as for Z_SYNC_FLUSH. - This completes the current deflate block and follows it with an empty fixed - codes block that is 10 bits long. This assures that enough bytes are output - in order for the decompressor to finish the block before the empty fixed - codes block. - - If flush is set to Z_BLOCK, a deflate block is completed and emitted, as - for Z_SYNC_FLUSH, but the output is not aligned on a byte boundary, and up to - seven bits of the current block are held to be written as the next byte after - the next deflate block is completed. In this case, the decompressor may not - be provided enough bits at this point in order to complete decompression of - the data provided so far to the compressor. It may need to wait for the next - block to be emitted. This is for advanced applications that need to control - the emission of deflate blocks. - - If flush is set to Z_FULL_FLUSH, all output is flushed as with - Z_SYNC_FLUSH, and the compression state is reset so that decompression can - restart from this point if previous compressed data has been damaged or if - random access is desired. Using Z_FULL_FLUSH too often can seriously degrade - compression. - - If deflate returns with avail_out == 0, this function must be called again - with the same value of the flush parameter and more output space (updated - avail_out), until the flush is complete (deflate returns with non-zero - avail_out). In the case of a Z_FULL_FLUSH or Z_SYNC_FLUSH, make sure that - avail_out is greater than six to avoid repeated flush markers due to - avail_out == 0 on return. - - If the parameter flush is set to Z_FINISH, pending input is processed, - pending output is flushed and deflate returns with Z_STREAM_END if there was - enough output space. If deflate returns with Z_OK or Z_BUF_ERROR, this - function must be called again with Z_FINISH and more output space (updated - avail_out) but no more input data, until it returns with Z_STREAM_END or an - error. After deflate has returned Z_STREAM_END, the only possible operations - on the stream are deflateReset or deflateEnd. - - Z_FINISH can be used in the first deflate call after deflateInit if all the - compression is to be done in a single step. In order to complete in one - call, avail_out must be at least the value returned by deflateBound (see - below). Then deflate is guaranteed to return Z_STREAM_END. If not enough - output space is provided, deflate will not return Z_STREAM_END, and it must - be called again as described above. - - deflate() sets strm->adler to the Adler-32 checksum of all input read - so far (that is, total_in bytes). If a gzip stream is being generated, then - strm->adler will be the CRC-32 checksum of the input read so far. (See - deflateInit2 below.) - - deflate() may update strm->data_type if it can make a good guess about - the input data type (Z_BINARY or Z_TEXT). If in doubt, the data is - considered binary. This field is only for information purposes and does not - affect the compression algorithm in any manner. - - deflate() returns Z_OK if some progress has been made (more input - processed or more output produced), Z_STREAM_END if all input has been - consumed and all output has been produced (only when flush is set to - Z_FINISH), Z_STREAM_ERROR if the stream state was inconsistent (for example - if next_in or next_out was Z_NULL or the state was inadvertently written over - by the application), or Z_BUF_ERROR if no progress is possible (for example - avail_in or avail_out was zero). Note that Z_BUF_ERROR is not fatal, and - deflate() can be called again with more input and more output space to - continue compressing. -*/ - - -ZEXTERN int ZEXPORT deflateEnd OF((z_streamp strm)); -/* - All dynamically allocated data structures for this stream are freed. - This function discards any unprocessed input and does not flush any pending - output. - - deflateEnd returns Z_OK if success, Z_STREAM_ERROR if the - stream state was inconsistent, Z_DATA_ERROR if the stream was freed - prematurely (some input or output was discarded). In the error case, msg - may be set but then points to a static string (which must not be - deallocated). -*/ - - -/* -ZEXTERN int ZEXPORT inflateInit OF((z_streamp strm)); - - Initializes the internal stream state for decompression. The fields - next_in, avail_in, zalloc, zfree and opaque must be initialized before by - the caller. In the current version of inflate, the provided input is not - read or consumed. The allocation of a sliding window will be deferred to - the first call of inflate (if the decompression does not complete on the - first call). If zalloc and zfree are set to Z_NULL, inflateInit updates - them to use default allocation functions. - - inflateInit returns Z_OK if success, Z_MEM_ERROR if there was not enough - memory, Z_VERSION_ERROR if the zlib library version is incompatible with the - version assumed by the caller, or Z_STREAM_ERROR if the parameters are - invalid, such as a null pointer to the structure. msg is set to null if - there is no error message. inflateInit does not perform any decompression. - Actual decompression will be done by inflate(). So next_in, and avail_in, - next_out, and avail_out are unused and unchanged. The current - implementation of inflateInit() does not process any header information -- - that is deferred until inflate() is called. -*/ - - -ZEXTERN int ZEXPORT inflate OF((z_streamp strm, int flush)); -/* - inflate decompresses as much data as possible, and stops when the input - buffer becomes empty or the output buffer becomes full. It may introduce - some output latency (reading input without producing any output) except when - forced to flush. - - The detailed semantics are as follows. inflate performs one or both of the - following actions: - - - Decompress more input starting at next_in and update next_in and avail_in - accordingly. If not all input can be processed (because there is not - enough room in the output buffer), then next_in and avail_in are updated - accordingly, and processing will resume at this point for the next call of - inflate(). - - - Generate more output starting at next_out and update next_out and avail_out - accordingly. inflate() provides as much output as possible, until there is - no more input data or no more space in the output buffer (see below about - the flush parameter). - - Before the call of inflate(), the application should ensure that at least - one of the actions is possible, by providing more input and/or consuming more - output, and updating the next_* and avail_* values accordingly. If the - caller of inflate() does not provide both available input and available - output space, it is possible that there will be no progress made. The - application can consume the uncompressed output when it wants, for example - when the output buffer is full (avail_out == 0), or after each call of - inflate(). If inflate returns Z_OK and with zero avail_out, it must be - called again after making room in the output buffer because there might be - more output pending. - - The flush parameter of inflate() can be Z_NO_FLUSH, Z_SYNC_FLUSH, Z_FINISH, - Z_BLOCK, or Z_TREES. Z_SYNC_FLUSH requests that inflate() flush as much - output as possible to the output buffer. Z_BLOCK requests that inflate() - stop if and when it gets to the next deflate block boundary. When decoding - the zlib or gzip format, this will cause inflate() to return immediately - after the header and before the first block. When doing a raw inflate, - inflate() will go ahead and process the first block, and will return when it - gets to the end of that block, or when it runs out of data. - - The Z_BLOCK option assists in appending to or combining deflate streams. - To assist in this, on return inflate() always sets strm->data_type to the - number of unused bits in the last byte taken from strm->next_in, plus 64 if - inflate() is currently decoding the last block in the deflate stream, plus - 128 if inflate() returned immediately after decoding an end-of-block code or - decoding the complete header up to just before the first byte of the deflate - stream. The end-of-block will not be indicated until all of the uncompressed - data from that block has been written to strm->next_out. The number of - unused bits may in general be greater than seven, except when bit 7 of - data_type is set, in which case the number of unused bits will be less than - eight. data_type is set as noted here every time inflate() returns for all - flush options, and so can be used to determine the amount of currently - consumed input in bits. - - The Z_TREES option behaves as Z_BLOCK does, but it also returns when the - end of each deflate block header is reached, before any actual data in that - block is decoded. This allows the caller to determine the length of the - deflate block header for later use in random access within a deflate block. - 256 is added to the value of strm->data_type when inflate() returns - immediately after reaching the end of the deflate block header. - - inflate() should normally be called until it returns Z_STREAM_END or an - error. However if all decompression is to be performed in a single step (a - single call of inflate), the parameter flush should be set to Z_FINISH. In - this case all pending input is processed and all pending output is flushed; - avail_out must be large enough to hold all of the uncompressed data for the - operation to complete. (The size of the uncompressed data may have been - saved by the compressor for this purpose.) The use of Z_FINISH is not - required to perform an inflation in one step. However it may be used to - inform inflate that a faster approach can be used for the single inflate() - call. Z_FINISH also informs inflate to not maintain a sliding window if the - stream completes, which reduces inflate's memory footprint. If the stream - does not complete, either because not all of the stream is provided or not - enough output space is provided, then a sliding window will be allocated and - inflate() can be called again to continue the operation as if Z_NO_FLUSH had - been used. - - In this implementation, inflate() always flushes as much output as - possible to the output buffer, and always uses the faster approach on the - first call. So the effects of the flush parameter in this implementation are - on the return value of inflate() as noted below, when inflate() returns early - when Z_BLOCK or Z_TREES is used, and when inflate() avoids the allocation of - memory for a sliding window when Z_FINISH is used. - - If a preset dictionary is needed after this call (see inflateSetDictionary - below), inflate sets strm->adler to the Adler-32 checksum of the dictionary - chosen by the compressor and returns Z_NEED_DICT; otherwise it sets - strm->adler to the Adler-32 checksum of all output produced so far (that is, - total_out bytes) and returns Z_OK, Z_STREAM_END or an error code as described - below. At the end of the stream, inflate() checks that its computed Adler-32 - checksum is equal to that saved by the compressor and returns Z_STREAM_END - only if the checksum is correct. - - inflate() can decompress and check either zlib-wrapped or gzip-wrapped - deflate data. The header type is detected automatically, if requested when - initializing with inflateInit2(). Any information contained in the gzip - header is not retained unless inflateGetHeader() is used. When processing - gzip-wrapped deflate data, strm->adler32 is set to the CRC-32 of the output - produced so far. The CRC-32 is checked against the gzip trailer, as is the - uncompressed length, modulo 2^32. - - inflate() returns Z_OK if some progress has been made (more input processed - or more output produced), Z_STREAM_END if the end of the compressed data has - been reached and all uncompressed output has been produced, Z_NEED_DICT if a - preset dictionary is needed at this point, Z_DATA_ERROR if the input data was - corrupted (input stream not conforming to the zlib format or incorrect check - value, in which case strm->msg points to a string with a more specific - error), Z_STREAM_ERROR if the stream structure was inconsistent (for example - next_in or next_out was Z_NULL, or the state was inadvertently written over - by the application), Z_MEM_ERROR if there was not enough memory, Z_BUF_ERROR - if no progress was possible or if there was not enough room in the output - buffer when Z_FINISH is used. Note that Z_BUF_ERROR is not fatal, and - inflate() can be called again with more input and more output space to - continue decompressing. If Z_DATA_ERROR is returned, the application may - then call inflateSync() to look for a good compression block if a partial - recovery of the data is to be attempted. -*/ - - -ZEXTERN int ZEXPORT inflateEnd OF((z_streamp strm)); -/* - All dynamically allocated data structures for this stream are freed. - This function discards any unprocessed input and does not flush any pending - output. - - inflateEnd returns Z_OK if success, or Z_STREAM_ERROR if the stream state - was inconsistent. -*/ - - - /* Advanced functions */ - -/* - The following functions are needed only in some special applications. -*/ - -/* -ZEXTERN int ZEXPORT deflateInit2 OF((z_streamp strm, - int level, - int method, - int windowBits, - int memLevel, - int strategy)); - - This is another version of deflateInit with more compression options. The - fields next_in, zalloc, zfree and opaque must be initialized before by the - caller. - - The method parameter is the compression method. It must be Z_DEFLATED in - this version of the library. - - The windowBits parameter is the base two logarithm of the window size - (the size of the history buffer). It should be in the range 8..15 for this - version of the library. Larger values of this parameter result in better - compression at the expense of memory usage. The default value is 15 if - deflateInit is used instead. - - For the current implementation of deflate(), a windowBits value of 8 (a - window size of 256 bytes) is not supported. As a result, a request for 8 - will result in 9 (a 512-byte window). In that case, providing 8 to - inflateInit2() will result in an error when the zlib header with 9 is - checked against the initialization of inflate(). The remedy is to not use 8 - with deflateInit2() with this initialization, or at least in that case use 9 - with inflateInit2(). - - windowBits can also be -8..-15 for raw deflate. In this case, -windowBits - determines the window size. deflate() will then generate raw deflate data - with no zlib header or trailer, and will not compute a check value. - - windowBits can also be greater than 15 for optional gzip encoding. Add - 16 to windowBits to write a simple gzip header and trailer around the - compressed data instead of a zlib wrapper. The gzip header will have no - file name, no extra data, no comment, no modification time (set to zero), no - header crc, and the operating system will be set to the appropriate value, - if the operating system was determined at compile time. If a gzip stream is - being written, strm->adler is a CRC-32 instead of an Adler-32. - - For raw deflate or gzip encoding, a request for a 256-byte window is - rejected as invalid, since only the zlib header provides a means of - transmitting the window size to the decompressor. - - The memLevel parameter specifies how much memory should be allocated - for the internal compression state. memLevel=1 uses minimum memory but is - slow and reduces compression ratio; memLevel=9 uses maximum memory for - optimal speed. The default value is 8. See zconf.h for total memory usage - as a function of windowBits and memLevel. - - The strategy parameter is used to tune the compression algorithm. Use the - value Z_DEFAULT_STRATEGY for normal data, Z_FILTERED for data produced by a - filter (or predictor), Z_HUFFMAN_ONLY to force Huffman encoding only (no - string match), or Z_RLE to limit match distances to one (run-length - encoding). Filtered data consists mostly of small values with a somewhat - random distribution. In this case, the compression algorithm is tuned to - compress them better. The effect of Z_FILTERED is to force more Huffman - coding and less string matching; it is somewhat intermediate between - Z_DEFAULT_STRATEGY and Z_HUFFMAN_ONLY. Z_RLE is designed to be almost as - fast as Z_HUFFMAN_ONLY, but give better compression for PNG image data. The - strategy parameter only affects the compression ratio but not the - correctness of the compressed output even if it is not set appropriately. - Z_FIXED prevents the use of dynamic Huffman codes, allowing for a simpler - decoder for special applications. - - deflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough - memory, Z_STREAM_ERROR if any parameter is invalid (such as an invalid - method), or Z_VERSION_ERROR if the zlib library version (zlib_version) is - incompatible with the version assumed by the caller (ZLIB_VERSION). msg is - set to null if there is no error message. deflateInit2 does not perform any - compression: this will be done by deflate(). -*/ - -ZEXTERN int ZEXPORT deflateSetDictionary OF((z_streamp strm, - const Bytef *dictionary, - uInt dictLength)); -/* - Initializes the compression dictionary from the given byte sequence - without producing any compressed output. When using the zlib format, this - function must be called immediately after deflateInit, deflateInit2 or - deflateReset, and before any call of deflate. When doing raw deflate, this - function must be called either before any call of deflate, or immediately - after the completion of a deflate block, i.e. after all input has been - consumed and all output has been delivered when using any of the flush - options Z_BLOCK, Z_PARTIAL_FLUSH, Z_SYNC_FLUSH, or Z_FULL_FLUSH. The - compressor and decompressor must use exactly the same dictionary (see - inflateSetDictionary). - - The dictionary should consist of strings (byte sequences) that are likely - to be encountered later in the data to be compressed, with the most commonly - used strings preferably put towards the end of the dictionary. Using a - dictionary is most useful when the data to be compressed is short and can be - predicted with good accuracy; the data can then be compressed better than - with the default empty dictionary. - - Depending on the size of the compression data structures selected by - deflateInit or deflateInit2, a part of the dictionary may in effect be - discarded, for example if the dictionary is larger than the window size - provided in deflateInit or deflateInit2. Thus the strings most likely to be - useful should be put at the end of the dictionary, not at the front. In - addition, the current implementation of deflate will use at most the window - size minus 262 bytes of the provided dictionary. - - Upon return of this function, strm->adler is set to the Adler-32 value - of the dictionary; the decompressor may later use this value to determine - which dictionary has been used by the compressor. (The Adler-32 value - applies to the whole dictionary even if only a subset of the dictionary is - actually used by the compressor.) If a raw deflate was requested, then the - Adler-32 value is not computed and strm->adler is not set. - - deflateSetDictionary returns Z_OK if success, or Z_STREAM_ERROR if a - parameter is invalid (e.g. dictionary being Z_NULL) or the stream state is - inconsistent (for example if deflate has already been called for this stream - or if not at a block boundary for raw deflate). deflateSetDictionary does - not perform any compression: this will be done by deflate(). -*/ - -ZEXTERN int ZEXPORT deflateGetDictionary OF((z_streamp strm, - Bytef *dictionary, - uInt *dictLength)); -/* - Returns the sliding dictionary being maintained by deflate. dictLength is - set to the number of bytes in the dictionary, and that many bytes are copied - to dictionary. dictionary must have enough space, where 32768 bytes is - always enough. If deflateGetDictionary() is called with dictionary equal to - Z_NULL, then only the dictionary length is returned, and nothing is copied. - Similary, if dictLength is Z_NULL, then it is not set. - - deflateGetDictionary() may return a length less than the window size, even - when more than the window size in input has been provided. It may return up - to 258 bytes less in that case, due to how zlib's implementation of deflate - manages the sliding window and lookahead for matches, where matches can be - up to 258 bytes long. If the application needs the last window-size bytes of - input, then that would need to be saved by the application outside of zlib. - - deflateGetDictionary returns Z_OK on success, or Z_STREAM_ERROR if the - stream state is inconsistent. -*/ - -ZEXTERN int ZEXPORT deflateCopy OF((z_streamp dest, - z_streamp source)); -/* - Sets the destination stream as a complete copy of the source stream. - - This function can be useful when several compression strategies will be - tried, for example when there are several ways of pre-processing the input - data with a filter. The streams that will be discarded should then be freed - by calling deflateEnd. Note that deflateCopy duplicates the internal - compression state which can be quite large, so this strategy is slow and can - consume lots of memory. - - deflateCopy returns Z_OK if success, Z_MEM_ERROR if there was not - enough memory, Z_STREAM_ERROR if the source stream state was inconsistent - (such as zalloc being Z_NULL). msg is left unchanged in both source and - destination. -*/ - -ZEXTERN int ZEXPORT deflateReset OF((z_streamp strm)); -/* - This function is equivalent to deflateEnd followed by deflateInit, but - does not free and reallocate the internal compression state. The stream - will leave the compression level and any other attributes that may have been - set unchanged. - - deflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source - stream state was inconsistent (such as zalloc or state being Z_NULL). -*/ - -ZEXTERN int ZEXPORT deflateParams OF((z_streamp strm, - int level, - int strategy)); -/* - Dynamically update the compression level and compression strategy. The - interpretation of level and strategy is as in deflateInit2(). This can be - used to switch between compression and straight copy of the input data, or - to switch to a different kind of input data requiring a different strategy. - If the compression approach (which is a function of the level) or the - strategy is changed, and if any input has been consumed in a previous - deflate() call, then the input available so far is compressed with the old - level and strategy using deflate(strm, Z_BLOCK). There are three approaches - for the compression levels 0, 1..3, and 4..9 respectively. The new level - and strategy will take effect at the next call of deflate(). - - If a deflate(strm, Z_BLOCK) is performed by deflateParams(), and it does - not have enough output space to complete, then the parameter change will not - take effect. In this case, deflateParams() can be called again with the - same parameters and more output space to try again. - - In order to assure a change in the parameters on the first try, the - deflate stream should be flushed using deflate() with Z_BLOCK or other flush - request until strm.avail_out is not zero, before calling deflateParams(). - Then no more input data should be provided before the deflateParams() call. - If this is done, the old level and strategy will be applied to the data - compressed before deflateParams(), and the new level and strategy will be - applied to the the data compressed after deflateParams(). - - deflateParams returns Z_OK on success, Z_STREAM_ERROR if the source stream - state was inconsistent or if a parameter was invalid, or Z_BUF_ERROR if - there was not enough output space to complete the compression of the - available input data before a change in the strategy or approach. Note that - in the case of a Z_BUF_ERROR, the parameters are not changed. A return - value of Z_BUF_ERROR is not fatal, in which case deflateParams() can be - retried with more output space. -*/ - -ZEXTERN int ZEXPORT deflateTune OF((z_streamp strm, - int good_length, - int max_lazy, - int nice_length, - int max_chain)); -/* - Fine tune deflate's internal compression parameters. This should only be - used by someone who understands the algorithm used by zlib's deflate for - searching for the best matching string, and even then only by the most - fanatic optimizer trying to squeeze out the last compressed bit for their - specific input data. Read the deflate.c source code for the meaning of the - max_lazy, good_length, nice_length, and max_chain parameters. - - deflateTune() can be called after deflateInit() or deflateInit2(), and - returns Z_OK on success, or Z_STREAM_ERROR for an invalid deflate stream. - */ - -ZEXTERN uLong ZEXPORT deflateBound OF((z_streamp strm, - uLong sourceLen)); -/* - deflateBound() returns an upper bound on the compressed size after - deflation of sourceLen bytes. It must be called after deflateInit() or - deflateInit2(), and after deflateSetHeader(), if used. This would be used - to allocate an output buffer for deflation in a single pass, and so would be - called before deflate(). If that first deflate() call is provided the - sourceLen input bytes, an output buffer allocated to the size returned by - deflateBound(), and the flush value Z_FINISH, then deflate() is guaranteed - to return Z_STREAM_END. Note that it is possible for the compressed size to - be larger than the value returned by deflateBound() if flush options other - than Z_FINISH or Z_NO_FLUSH are used. -*/ - -ZEXTERN int ZEXPORT deflatePending OF((z_streamp strm, - unsigned *pending, - int *bits)); -/* - deflatePending() returns the number of bytes and bits of output that have - been generated, but not yet provided in the available output. The bytes not - provided would be due to the available output space having being consumed. - The number of bits of output not provided are between 0 and 7, where they - await more bits to join them in order to fill out a full byte. If pending - or bits are Z_NULL, then those values are not set. - - deflatePending returns Z_OK if success, or Z_STREAM_ERROR if the source - stream state was inconsistent. - */ - -ZEXTERN int ZEXPORT deflatePrime OF((z_streamp strm, - int bits, - int value)); -/* - deflatePrime() inserts bits in the deflate output stream. The intent - is that this function is used to start off the deflate output with the bits - leftover from a previous deflate stream when appending to it. As such, this - function can only be used for raw deflate, and must be used before the first - deflate() call after a deflateInit2() or deflateReset(). bits must be less - than or equal to 16, and that many of the least significant bits of value - will be inserted in the output. - - deflatePrime returns Z_OK if success, Z_BUF_ERROR if there was not enough - room in the internal buffer to insert the bits, or Z_STREAM_ERROR if the - source stream state was inconsistent. -*/ - -ZEXTERN int ZEXPORT deflateSetHeader OF((z_streamp strm, - gz_headerp head)); -/* - deflateSetHeader() provides gzip header information for when a gzip - stream is requested by deflateInit2(). deflateSetHeader() may be called - after deflateInit2() or deflateReset() and before the first call of - deflate(). The text, time, os, extra field, name, and comment information - in the provided gz_header structure are written to the gzip header (xflag is - ignored -- the extra flags are set according to the compression level). The - caller must assure that, if not Z_NULL, name and comment are terminated with - a zero byte, and that if extra is not Z_NULL, that extra_len bytes are - available there. If hcrc is true, a gzip header crc is included. Note that - the current versions of the command-line version of gzip (up through version - 1.3.x) do not support header crc's, and will report that it is a "multi-part - gzip file" and give up. - - If deflateSetHeader is not used, the default gzip header has text false, - the time set to zero, and os set to 255, with no extra, name, or comment - fields. The gzip header is returned to the default state by deflateReset(). - - deflateSetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source - stream state was inconsistent. -*/ - -/* -ZEXTERN int ZEXPORT inflateInit2 OF((z_streamp strm, - int windowBits)); - - This is another version of inflateInit with an extra parameter. The - fields next_in, avail_in, zalloc, zfree and opaque must be initialized - before by the caller. - - The windowBits parameter is the base two logarithm of the maximum window - size (the size of the history buffer). It should be in the range 8..15 for - this version of the library. The default value is 15 if inflateInit is used - instead. windowBits must be greater than or equal to the windowBits value - provided to deflateInit2() while compressing, or it must be equal to 15 if - deflateInit2() was not used. If a compressed stream with a larger window - size is given as input, inflate() will return with the error code - Z_DATA_ERROR instead of trying to allocate a larger window. - - windowBits can also be zero to request that inflate use the window size in - the zlib header of the compressed stream. - - windowBits can also be -8..-15 for raw inflate. In this case, -windowBits - determines the window size. inflate() will then process raw deflate data, - not looking for a zlib or gzip header, not generating a check value, and not - looking for any check values for comparison at the end of the stream. This - is for use with other formats that use the deflate compressed data format - such as zip. Those formats provide their own check values. If a custom - format is developed using the raw deflate format for compressed data, it is - recommended that a check value such as an Adler-32 or a CRC-32 be applied to - the uncompressed data as is done in the zlib, gzip, and zip formats. For - most applications, the zlib format should be used as is. Note that comments - above on the use in deflateInit2() applies to the magnitude of windowBits. - - windowBits can also be greater than 15 for optional gzip decoding. Add - 32 to windowBits to enable zlib and gzip decoding with automatic header - detection, or add 16 to decode only the gzip format (the zlib format will - return a Z_DATA_ERROR). If a gzip stream is being decoded, strm->adler is a - CRC-32 instead of an Adler-32. Unlike the gunzip utility and gzread() (see - below), inflate() will not automatically decode concatenated gzip streams. - inflate() will return Z_STREAM_END at the end of the gzip stream. The state - would need to be reset to continue decoding a subsequent gzip stream. - - inflateInit2 returns Z_OK if success, Z_MEM_ERROR if there was not enough - memory, Z_VERSION_ERROR if the zlib library version is incompatible with the - version assumed by the caller, or Z_STREAM_ERROR if the parameters are - invalid, such as a null pointer to the structure. msg is set to null if - there is no error message. inflateInit2 does not perform any decompression - apart from possibly reading the zlib header if present: actual decompression - will be done by inflate(). (So next_in and avail_in may be modified, but - next_out and avail_out are unused and unchanged.) The current implementation - of inflateInit2() does not process any header information -- that is - deferred until inflate() is called. -*/ - -ZEXTERN int ZEXPORT inflateSetDictionary OF((z_streamp strm, - const Bytef *dictionary, - uInt dictLength)); -/* - Initializes the decompression dictionary from the given uncompressed byte - sequence. This function must be called immediately after a call of inflate, - if that call returned Z_NEED_DICT. The dictionary chosen by the compressor - can be determined from the Adler-32 value returned by that call of inflate. - The compressor and decompressor must use exactly the same dictionary (see - deflateSetDictionary). For raw inflate, this function can be called at any - time to set the dictionary. If the provided dictionary is smaller than the - window and there is already data in the window, then the provided dictionary - will amend what's there. The application must insure that the dictionary - that was used for compression is provided. - - inflateSetDictionary returns Z_OK if success, Z_STREAM_ERROR if a - parameter is invalid (e.g. dictionary being Z_NULL) or the stream state is - inconsistent, Z_DATA_ERROR if the given dictionary doesn't match the - expected one (incorrect Adler-32 value). inflateSetDictionary does not - perform any decompression: this will be done by subsequent calls of - inflate(). -*/ - -ZEXTERN int ZEXPORT inflateGetDictionary OF((z_streamp strm, - Bytef *dictionary, - uInt *dictLength)); -/* - Returns the sliding dictionary being maintained by inflate. dictLength is - set to the number of bytes in the dictionary, and that many bytes are copied - to dictionary. dictionary must have enough space, where 32768 bytes is - always enough. If inflateGetDictionary() is called with dictionary equal to - Z_NULL, then only the dictionary length is returned, and nothing is copied. - Similary, if dictLength is Z_NULL, then it is not set. - - inflateGetDictionary returns Z_OK on success, or Z_STREAM_ERROR if the - stream state is inconsistent. -*/ - -ZEXTERN int ZEXPORT inflateSync OF((z_streamp strm)); -/* - Skips invalid compressed data until a possible full flush point (see above - for the description of deflate with Z_FULL_FLUSH) can be found, or until all - available input is skipped. No output is provided. - - inflateSync searches for a 00 00 FF FF pattern in the compressed data. - All full flush points have this pattern, but not all occurrences of this - pattern are full flush points. - - inflateSync returns Z_OK if a possible full flush point has been found, - Z_BUF_ERROR if no more input was provided, Z_DATA_ERROR if no flush point - has been found, or Z_STREAM_ERROR if the stream structure was inconsistent. - In the success case, the application may save the current current value of - total_in which indicates where valid compressed data was found. In the - error case, the application may repeatedly call inflateSync, providing more - input each time, until success or end of the input data. -*/ - -ZEXTERN int ZEXPORT inflateCopy OF((z_streamp dest, - z_streamp source)); -/* - Sets the destination stream as a complete copy of the source stream. - - This function can be useful when randomly accessing a large stream. The - first pass through the stream can periodically record the inflate state, - allowing restarting inflate at those points when randomly accessing the - stream. - - inflateCopy returns Z_OK if success, Z_MEM_ERROR if there was not - enough memory, Z_STREAM_ERROR if the source stream state was inconsistent - (such as zalloc being Z_NULL). msg is left unchanged in both source and - destination. -*/ - -ZEXTERN int ZEXPORT inflateReset OF((z_streamp strm)); -/* - This function is equivalent to inflateEnd followed by inflateInit, - but does not free and reallocate the internal decompression state. The - stream will keep attributes that may have been set by inflateInit2. - - inflateReset returns Z_OK if success, or Z_STREAM_ERROR if the source - stream state was inconsistent (such as zalloc or state being Z_NULL). -*/ - -ZEXTERN int ZEXPORT inflateReset2 OF((z_streamp strm, - int windowBits)); -/* - This function is the same as inflateReset, but it also permits changing - the wrap and window size requests. The windowBits parameter is interpreted - the same as it is for inflateInit2. If the window size is changed, then the - memory allocated for the window is freed, and the window will be reallocated - by inflate() if needed. - - inflateReset2 returns Z_OK if success, or Z_STREAM_ERROR if the source - stream state was inconsistent (such as zalloc or state being Z_NULL), or if - the windowBits parameter is invalid. -*/ - -ZEXTERN int ZEXPORT inflatePrime OF((z_streamp strm, - int bits, - int value)); -/* - This function inserts bits in the inflate input stream. The intent is - that this function is used to start inflating at a bit position in the - middle of a byte. The provided bits will be used before any bytes are used - from next_in. This function should only be used with raw inflate, and - should be used before the first inflate() call after inflateInit2() or - inflateReset(). bits must be less than or equal to 16, and that many of the - least significant bits of value will be inserted in the input. - - If bits is negative, then the input stream bit buffer is emptied. Then - inflatePrime() can be called again to put bits in the buffer. This is used - to clear out bits leftover after feeding inflate a block description prior - to feeding inflate codes. - - inflatePrime returns Z_OK if success, or Z_STREAM_ERROR if the source - stream state was inconsistent. -*/ - -ZEXTERN long ZEXPORT inflateMark OF((z_streamp strm)); -/* - This function returns two values, one in the lower 16 bits of the return - value, and the other in the remaining upper bits, obtained by shifting the - return value down 16 bits. If the upper value is -1 and the lower value is - zero, then inflate() is currently decoding information outside of a block. - If the upper value is -1 and the lower value is non-zero, then inflate is in - the middle of a stored block, with the lower value equaling the number of - bytes from the input remaining to copy. If the upper value is not -1, then - it is the number of bits back from the current bit position in the input of - the code (literal or length/distance pair) currently being processed. In - that case the lower value is the number of bytes already emitted for that - code. - - A code is being processed if inflate is waiting for more input to complete - decoding of the code, or if it has completed decoding but is waiting for - more output space to write the literal or match data. - - inflateMark() is used to mark locations in the input data for random - access, which may be at bit positions, and to note those cases where the - output of a code may span boundaries of random access blocks. The current - location in the input stream can be determined from avail_in and data_type - as noted in the description for the Z_BLOCK flush parameter for inflate. - - inflateMark returns the value noted above, or -65536 if the provided - source stream state was inconsistent. -*/ - -ZEXTERN int ZEXPORT inflateGetHeader OF((z_streamp strm, - gz_headerp head)); -/* - inflateGetHeader() requests that gzip header information be stored in the - provided gz_header structure. inflateGetHeader() may be called after - inflateInit2() or inflateReset(), and before the first call of inflate(). - As inflate() processes the gzip stream, head->done is zero until the header - is completed, at which time head->done is set to one. If a zlib stream is - being decoded, then head->done is set to -1 to indicate that there will be - no gzip header information forthcoming. Note that Z_BLOCK or Z_TREES can be - used to force inflate() to return immediately after header processing is - complete and before any actual data is decompressed. - - The text, time, xflags, and os fields are filled in with the gzip header - contents. hcrc is set to true if there is a header CRC. (The header CRC - was valid if done is set to one.) If extra is not Z_NULL, then extra_max - contains the maximum number of bytes to write to extra. Once done is true, - extra_len contains the actual extra field length, and extra contains the - extra field, or that field truncated if extra_max is less than extra_len. - If name is not Z_NULL, then up to name_max characters are written there, - terminated with a zero unless the length is greater than name_max. If - comment is not Z_NULL, then up to comm_max characters are written there, - terminated with a zero unless the length is greater than comm_max. When any - of extra, name, or comment are not Z_NULL and the respective field is not - present in the header, then that field is set to Z_NULL to signal its - absence. This allows the use of deflateSetHeader() with the returned - structure to duplicate the header. However if those fields are set to - allocated memory, then the application will need to save those pointers - elsewhere so that they can be eventually freed. - - If inflateGetHeader is not used, then the header information is simply - discarded. The header is always checked for validity, including the header - CRC if present. inflateReset() will reset the process to discard the header - information. The application would need to call inflateGetHeader() again to - retrieve the header from the next gzip stream. - - inflateGetHeader returns Z_OK if success, or Z_STREAM_ERROR if the source - stream state was inconsistent. -*/ - -/* -ZEXTERN int ZEXPORT inflateBackInit OF((z_streamp strm, int windowBits, - unsigned char FAR *window)); - - Initialize the internal stream state for decompression using inflateBack() - calls. The fields zalloc, zfree and opaque in strm must be initialized - before the call. If zalloc and zfree are Z_NULL, then the default library- - derived memory allocation routines are used. windowBits is the base two - logarithm of the window size, in the range 8..15. window is a caller - supplied buffer of that size. Except for special applications where it is - assured that deflate was used with small window sizes, windowBits must be 15 - and a 32K byte window must be supplied to be able to decompress general - deflate streams. - - See inflateBack() for the usage of these routines. - - inflateBackInit will return Z_OK on success, Z_STREAM_ERROR if any of - the parameters are invalid, Z_MEM_ERROR if the internal state could not be - allocated, or Z_VERSION_ERROR if the version of the library does not match - the version of the header file. -*/ - -typedef unsigned (*in_func) OF((void FAR *, - z_const unsigned char FAR * FAR *)); -typedef int (*out_func) OF((void FAR *, unsigned char FAR *, unsigned)); - -ZEXTERN int ZEXPORT inflateBack OF((z_streamp strm, - in_func in, void FAR *in_desc, - out_func out, void FAR *out_desc)); -/* - inflateBack() does a raw inflate with a single call using a call-back - interface for input and output. This is potentially more efficient than - inflate() for file i/o applications, in that it avoids copying between the - output and the sliding window by simply making the window itself the output - buffer. inflate() can be faster on modern CPUs when used with large - buffers. inflateBack() trusts the application to not change the output - buffer passed by the output function, at least until inflateBack() returns. - - inflateBackInit() must be called first to allocate the internal state - and to initialize the state with the user-provided window buffer. - inflateBack() may then be used multiple times to inflate a complete, raw - deflate stream with each call. inflateBackEnd() is then called to free the - allocated state. - - A raw deflate stream is one with no zlib or gzip header or trailer. - This routine would normally be used in a utility that reads zip or gzip - files and writes out uncompressed files. The utility would decode the - header and process the trailer on its own, hence this routine expects only - the raw deflate stream to decompress. This is different from the default - behavior of inflate(), which expects a zlib header and trailer around the - deflate stream. - - inflateBack() uses two subroutines supplied by the caller that are then - called by inflateBack() for input and output. inflateBack() calls those - routines until it reads a complete deflate stream and writes out all of the - uncompressed data, or until it encounters an error. The function's - parameters and return types are defined above in the in_func and out_func - typedefs. inflateBack() will call in(in_desc, &buf) which should return the - number of bytes of provided input, and a pointer to that input in buf. If - there is no input available, in() must return zero -- buf is ignored in that - case -- and inflateBack() will return a buffer error. inflateBack() will - call out(out_desc, buf, len) to write the uncompressed data buf[0..len-1]. - out() should return zero on success, or non-zero on failure. If out() - returns non-zero, inflateBack() will return with an error. Neither in() nor - out() are permitted to change the contents of the window provided to - inflateBackInit(), which is also the buffer that out() uses to write from. - The length written by out() will be at most the window size. Any non-zero - amount of input may be provided by in(). - - For convenience, inflateBack() can be provided input on the first call by - setting strm->next_in and strm->avail_in. If that input is exhausted, then - in() will be called. Therefore strm->next_in must be initialized before - calling inflateBack(). If strm->next_in is Z_NULL, then in() will be called - immediately for input. If strm->next_in is not Z_NULL, then strm->avail_in - must also be initialized, and then if strm->avail_in is not zero, input will - initially be taken from strm->next_in[0 .. strm->avail_in - 1]. - - The in_desc and out_desc parameters of inflateBack() is passed as the - first parameter of in() and out() respectively when they are called. These - descriptors can be optionally used to pass any information that the caller- - supplied in() and out() functions need to do their job. - - On return, inflateBack() will set strm->next_in and strm->avail_in to - pass back any unused input that was provided by the last in() call. The - return values of inflateBack() can be Z_STREAM_END on success, Z_BUF_ERROR - if in() or out() returned an error, Z_DATA_ERROR if there was a format error - in the deflate stream (in which case strm->msg is set to indicate the nature - of the error), or Z_STREAM_ERROR if the stream was not properly initialized. - In the case of Z_BUF_ERROR, an input or output error can be distinguished - using strm->next_in which will be Z_NULL only if in() returned an error. If - strm->next_in is not Z_NULL, then the Z_BUF_ERROR was due to out() returning - non-zero. (in() will always be called before out(), so strm->next_in is - assured to be defined if out() returns non-zero.) Note that inflateBack() - cannot return Z_OK. -*/ - -ZEXTERN int ZEXPORT inflateBackEnd OF((z_streamp strm)); -/* - All memory allocated by inflateBackInit() is freed. - - inflateBackEnd() returns Z_OK on success, or Z_STREAM_ERROR if the stream - state was inconsistent. -*/ - -ZEXTERN uLong ZEXPORT zlibCompileFlags OF((void)); -/* Return flags indicating compile-time options. - - Type sizes, two bits each, 00 = 16 bits, 01 = 32, 10 = 64, 11 = other: - 1.0: size of uInt - 3.2: size of uLong - 5.4: size of voidpf (pointer) - 7.6: size of z_off_t - - Compiler, assembler, and debug options: - 8: ZLIB_DEBUG - 9: ASMV or ASMINF -- use ASM code - 10: ZLIB_WINAPI -- exported functions use the WINAPI calling convention - 11: 0 (reserved) - - One-time table building (smaller code, but not thread-safe if true): - 12: BUILDFIXED -- build static block decoding tables when needed - 13: DYNAMIC_CRC_TABLE -- build CRC calculation tables when needed - 14,15: 0 (reserved) - - Library content (indicates missing functionality): - 16: NO_GZCOMPRESS -- gz* functions cannot compress (to avoid linking - deflate code when not needed) - 17: NO_GZIP -- deflate can't write gzip streams, and inflate can't detect - and decode gzip streams (to avoid linking crc code) - 18-19: 0 (reserved) - - Operation variations (changes in library functionality): - 20: PKZIP_BUG_WORKAROUND -- slightly more permissive inflate - 21: FASTEST -- deflate algorithm with only one, lowest compression level - 22,23: 0 (reserved) - - The sprintf variant used by gzprintf (zero is best): - 24: 0 = vs*, 1 = s* -- 1 means limited to 20 arguments after the format - 25: 0 = *nprintf, 1 = *printf -- 1 means gzprintf() not secure! - 26: 0 = returns value, 1 = void -- 1 means inferred string length returned - - Remainder: - 27-31: 0 (reserved) - */ - -#ifndef Z_SOLO - - /* utility functions */ - -/* - The following utility functions are implemented on top of the basic - stream-oriented functions. To simplify the interface, some default options - are assumed (compression level and memory usage, standard memory allocation - functions). The source code of these utility functions can be modified if - you need special options. -*/ - -ZEXTERN int ZEXPORT compress OF((Bytef *dest, uLongf *destLen, - const Bytef *source, uLong sourceLen)); -/* - Compresses the source buffer into the destination buffer. sourceLen is - the byte length of the source buffer. Upon entry, destLen is the total size - of the destination buffer, which must be at least the value returned by - compressBound(sourceLen). Upon exit, destLen is the actual size of the - compressed data. compress() is equivalent to compress2() with a level - parameter of Z_DEFAULT_COMPRESSION. - - compress returns Z_OK if success, Z_MEM_ERROR if there was not - enough memory, Z_BUF_ERROR if there was not enough room in the output - buffer. -*/ - -ZEXTERN int ZEXPORT compress2 OF((Bytef *dest, uLongf *destLen, - const Bytef *source, uLong sourceLen, - int level)); -/* - Compresses the source buffer into the destination buffer. The level - parameter has the same meaning as in deflateInit. sourceLen is the byte - length of the source buffer. Upon entry, destLen is the total size of the - destination buffer, which must be at least the value returned by - compressBound(sourceLen). Upon exit, destLen is the actual size of the - compressed data. - - compress2 returns Z_OK if success, Z_MEM_ERROR if there was not enough - memory, Z_BUF_ERROR if there was not enough room in the output buffer, - Z_STREAM_ERROR if the level parameter is invalid. -*/ - -ZEXTERN uLong ZEXPORT compressBound OF((uLong sourceLen)); -/* - compressBound() returns an upper bound on the compressed size after - compress() or compress2() on sourceLen bytes. It would be used before a - compress() or compress2() call to allocate the destination buffer. -*/ - -ZEXTERN int ZEXPORT uncompress OF((Bytef *dest, uLongf *destLen, - const Bytef *source, uLong sourceLen)); -/* - Decompresses the source buffer into the destination buffer. sourceLen is - the byte length of the source buffer. Upon entry, destLen is the total size - of the destination buffer, which must be large enough to hold the entire - uncompressed data. (The size of the uncompressed data must have been saved - previously by the compressor and transmitted to the decompressor by some - mechanism outside the scope of this compression library.) Upon exit, destLen - is the actual size of the uncompressed data. - - uncompress returns Z_OK if success, Z_MEM_ERROR if there was not - enough memory, Z_BUF_ERROR if there was not enough room in the output - buffer, or Z_DATA_ERROR if the input data was corrupted or incomplete. In - the case where there is not enough room, uncompress() will fill the output - buffer with the uncompressed data up to that point. -*/ - -ZEXTERN int ZEXPORT uncompress2 OF((Bytef *dest, uLongf *destLen, - const Bytef *source, uLong *sourceLen)); -/* - Same as uncompress, except that sourceLen is a pointer, where the - length of the source is *sourceLen. On return, *sourceLen is the number of - source bytes consumed. -*/ - - /* gzip file access functions */ - -/* - This library supports reading and writing files in gzip (.gz) format with - an interface similar to that of stdio, using the functions that start with - "gz". The gzip format is different from the zlib format. gzip is a gzip - wrapper, documented in RFC 1952, wrapped around a deflate stream. -*/ - -typedef struct gzFile_s *gzFile; /* semi-opaque gzip file descriptor */ - -/* -ZEXTERN gzFile ZEXPORT gzopen OF((const char *path, const char *mode)); - - Opens a gzip (.gz) file for reading or writing. The mode parameter is as - in fopen ("rb" or "wb") but can also include a compression level ("wb9") or - a strategy: 'f' for filtered data as in "wb6f", 'h' for Huffman-only - compression as in "wb1h", 'R' for run-length encoding as in "wb1R", or 'F' - for fixed code compression as in "wb9F". (See the description of - deflateInit2 for more information about the strategy parameter.) 'T' will - request transparent writing or appending with no compression and not using - the gzip format. - - "a" can be used instead of "w" to request that the gzip stream that will - be written be appended to the file. "+" will result in an error, since - reading and writing to the same gzip file is not supported. The addition of - "x" when writing will create the file exclusively, which fails if the file - already exists. On systems that support it, the addition of "e" when - reading or writing will set the flag to close the file on an execve() call. - - These functions, as well as gzip, will read and decode a sequence of gzip - streams in a file. The append function of gzopen() can be used to create - such a file. (Also see gzflush() for another way to do this.) When - appending, gzopen does not test whether the file begins with a gzip stream, - nor does it look for the end of the gzip streams to begin appending. gzopen - will simply append a gzip stream to the existing file. - - gzopen can be used to read a file which is not in gzip format; in this - case gzread will directly read from the file without decompression. When - reading, this will be detected automatically by looking for the magic two- - byte gzip header. - - gzopen returns NULL if the file could not be opened, if there was - insufficient memory to allocate the gzFile state, or if an invalid mode was - specified (an 'r', 'w', or 'a' was not provided, or '+' was provided). - errno can be checked to determine if the reason gzopen failed was that the - file could not be opened. -*/ - -ZEXTERN gzFile ZEXPORT gzdopen OF((int fd, const char *mode)); -/* - gzdopen associates a gzFile with the file descriptor fd. File descriptors - are obtained from calls like open, dup, creat, pipe or fileno (if the file - has been previously opened with fopen). The mode parameter is as in gzopen. - - The next call of gzclose on the returned gzFile will also close the file - descriptor fd, just like fclose(fdopen(fd, mode)) closes the file descriptor - fd. If you want to keep fd open, use fd = dup(fd_keep); gz = gzdopen(fd, - mode);. The duplicated descriptor should be saved to avoid a leak, since - gzdopen does not close fd if it fails. If you are using fileno() to get the - file descriptor from a FILE *, then you will have to use dup() to avoid - double-close()ing the file descriptor. Both gzclose() and fclose() will - close the associated file descriptor, so they need to have different file - descriptors. - - gzdopen returns NULL if there was insufficient memory to allocate the - gzFile state, if an invalid mode was specified (an 'r', 'w', or 'a' was not - provided, or '+' was provided), or if fd is -1. The file descriptor is not - used until the next gz* read, write, seek, or close operation, so gzdopen - will not detect if fd is invalid (unless fd is -1). -*/ - -ZEXTERN int ZEXPORT gzbuffer OF((gzFile file, unsigned size)); -/* - Set the internal buffer size used by this library's functions. The - default buffer size is 8192 bytes. This function must be called after - gzopen() or gzdopen(), and before any other calls that read or write the - file. The buffer memory allocation is always deferred to the first read or - write. Three times that size in buffer space is allocated. A larger buffer - size of, for example, 64K or 128K bytes will noticeably increase the speed - of decompression (reading). - - The new buffer size also affects the maximum length for gzprintf(). - - gzbuffer() returns 0 on success, or -1 on failure, such as being called - too late. -*/ - -ZEXTERN int ZEXPORT gzsetparams OF((gzFile file, int level, int strategy)); -/* - Dynamically update the compression level or strategy. See the description - of deflateInit2 for the meaning of these parameters. Previously provided - data is flushed before the parameter change. - - gzsetparams returns Z_OK if success, Z_STREAM_ERROR if the file was not - opened for writing, Z_ERRNO if there is an error writing the flushed data, - or Z_MEM_ERROR if there is a memory allocation error. -*/ - -ZEXTERN int ZEXPORT gzread OF((gzFile file, voidp buf, unsigned len)); -/* - Reads the given number of uncompressed bytes from the compressed file. If - the input file is not in gzip format, gzread copies the given number of - bytes into the buffer directly from the file. - - After reaching the end of a gzip stream in the input, gzread will continue - to read, looking for another gzip stream. Any number of gzip streams may be - concatenated in the input file, and will all be decompressed by gzread(). - If something other than a gzip stream is encountered after a gzip stream, - that remaining trailing garbage is ignored (and no error is returned). - - gzread can be used to read a gzip file that is being concurrently written. - Upon reaching the end of the input, gzread will return with the available - data. If the error code returned by gzerror is Z_OK or Z_BUF_ERROR, then - gzclearerr can be used to clear the end of file indicator in order to permit - gzread to be tried again. Z_OK indicates that a gzip stream was completed - on the last gzread. Z_BUF_ERROR indicates that the input file ended in the - middle of a gzip stream. Note that gzread does not return -1 in the event - of an incomplete gzip stream. This error is deferred until gzclose(), which - will return Z_BUF_ERROR if the last gzread ended in the middle of a gzip - stream. Alternatively, gzerror can be used before gzclose to detect this - case. - - gzread returns the number of uncompressed bytes actually read, less than - len for end of file, or -1 for error. If len is too large to fit in an int, - then nothing is read, -1 is returned, and the error state is set to - Z_STREAM_ERROR. -*/ - -ZEXTERN z_size_t ZEXPORT gzfread OF((voidp buf, z_size_t size, z_size_t nitems, - gzFile file)); -/* - Read up to nitems items of size size from file to buf, otherwise operating - as gzread() does. This duplicates the interface of stdio's fread(), with - size_t request and return types. If the library defines size_t, then - z_size_t is identical to size_t. If not, then z_size_t is an unsigned - integer type that can contain a pointer. - - gzfread() returns the number of full items read of size size, or zero if - the end of the file was reached and a full item could not be read, or if - there was an error. gzerror() must be consulted if zero is returned in - order to determine if there was an error. If the multiplication of size and - nitems overflows, i.e. the product does not fit in a z_size_t, then nothing - is read, zero is returned, and the error state is set to Z_STREAM_ERROR. - - In the event that the end of file is reached and only a partial item is - available at the end, i.e. the remaining uncompressed data length is not a - multiple of size, then the final partial item is nevetheless read into buf - and the end-of-file flag is set. The length of the partial item read is not - provided, but could be inferred from the result of gztell(). This behavior - is the same as the behavior of fread() implementations in common libraries, - but it prevents the direct use of gzfread() to read a concurrently written - file, reseting and retrying on end-of-file, when size is not 1. -*/ - -ZEXTERN int ZEXPORT gzwrite OF((gzFile file, - voidpc buf, unsigned len)); -/* - Writes the given number of uncompressed bytes into the compressed file. - gzwrite returns the number of uncompressed bytes written or 0 in case of - error. -*/ - -ZEXTERN z_size_t ZEXPORT gzfwrite OF((voidpc buf, z_size_t size, - z_size_t nitems, gzFile file)); -/* - gzfwrite() writes nitems items of size size from buf to file, duplicating - the interface of stdio's fwrite(), with size_t request and return types. If - the library defines size_t, then z_size_t is identical to size_t. If not, - then z_size_t is an unsigned integer type that can contain a pointer. - - gzfwrite() returns the number of full items written of size size, or zero - if there was an error. If the multiplication of size and nitems overflows, - i.e. the product does not fit in a z_size_t, then nothing is written, zero - is returned, and the error state is set to Z_STREAM_ERROR. -*/ - -ZEXTERN int ZEXPORTVA gzprintf Z_ARG((gzFile file, const char *format, ...)); -/* - Converts, formats, and writes the arguments to the compressed file under - control of the format string, as in fprintf. gzprintf returns the number of - uncompressed bytes actually written, or a negative zlib error code in case - of error. The number of uncompressed bytes written is limited to 8191, or - one less than the buffer size given to gzbuffer(). The caller should assure - that this limit is not exceeded. If it is exceeded, then gzprintf() will - return an error (0) with nothing written. In this case, there may also be a - buffer overflow with unpredictable consequences, which is possible only if - zlib was compiled with the insecure functions sprintf() or vsprintf() - because the secure snprintf() or vsnprintf() functions were not available. - This can be determined using zlibCompileFlags(). -*/ - -ZEXTERN int ZEXPORT gzputs OF((gzFile file, const char *s)); -/* - Writes the given null-terminated string to the compressed file, excluding - the terminating null character. - - gzputs returns the number of characters written, or -1 in case of error. -*/ - -ZEXTERN char * ZEXPORT gzgets OF((gzFile file, char *buf, int len)); -/* - Reads bytes from the compressed file until len-1 characters are read, or a - newline character is read and transferred to buf, or an end-of-file - condition is encountered. If any characters are read or if len == 1, the - string is terminated with a null character. If no characters are read due - to an end-of-file or len < 1, then the buffer is left untouched. - - gzgets returns buf which is a null-terminated string, or it returns NULL - for end-of-file or in case of error. If there was an error, the contents at - buf are indeterminate. -*/ - -ZEXTERN int ZEXPORT gzputc OF((gzFile file, int c)); -/* - Writes c, converted to an unsigned char, into the compressed file. gzputc - returns the value that was written, or -1 in case of error. -*/ - -ZEXTERN int ZEXPORT gzgetc OF((gzFile file)); -/* - Reads one byte from the compressed file. gzgetc returns this byte or -1 - in case of end of file or error. This is implemented as a macro for speed. - As such, it does not do all of the checking the other functions do. I.e. - it does not check to see if file is NULL, nor whether the structure file - points to has been clobbered or not. -*/ - -ZEXTERN int ZEXPORT gzungetc OF((int c, gzFile file)); -/* - Push one character back onto the stream to be read as the first character - on the next read. At least one character of push-back is allowed. - gzungetc() returns the character pushed, or -1 on failure. gzungetc() will - fail if c is -1, and may fail if a character has been pushed but not read - yet. If gzungetc is used immediately after gzopen or gzdopen, at least the - output buffer size of pushed characters is allowed. (See gzbuffer above.) - The pushed character will be discarded if the stream is repositioned with - gzseek() or gzrewind(). -*/ - -ZEXTERN int ZEXPORT gzflush OF((gzFile file, int flush)); -/* - Flushes all pending output into the compressed file. The parameter flush - is as in the deflate() function. The return value is the zlib error number - (see function gzerror below). gzflush is only permitted when writing. - - If the flush parameter is Z_FINISH, the remaining data is written and the - gzip stream is completed in the output. If gzwrite() is called again, a new - gzip stream will be started in the output. gzread() is able to read such - concatenated gzip streams. - - gzflush should be called only when strictly necessary because it will - degrade compression if called too often. -*/ - -/* -ZEXTERN z_off_t ZEXPORT gzseek OF((gzFile file, - z_off_t offset, int whence)); - - Sets the starting position for the next gzread or gzwrite on the given - compressed file. The offset represents a number of bytes in the - uncompressed data stream. The whence parameter is defined as in lseek(2); - the value SEEK_END is not supported. - - If the file is opened for reading, this function is emulated but can be - extremely slow. If the file is opened for writing, only forward seeks are - supported; gzseek then compresses a sequence of zeroes up to the new - starting position. - - gzseek returns the resulting offset location as measured in bytes from - the beginning of the uncompressed stream, or -1 in case of error, in - particular if the file is opened for writing and the new starting position - would be before the current position. -*/ - -ZEXTERN int ZEXPORT gzrewind OF((gzFile file)); -/* - Rewinds the given file. This function is supported only for reading. - - gzrewind(file) is equivalent to (int)gzseek(file, 0L, SEEK_SET) -*/ - -/* -ZEXTERN z_off_t ZEXPORT gztell OF((gzFile file)); - - Returns the starting position for the next gzread or gzwrite on the given - compressed file. This position represents a number of bytes in the - uncompressed data stream, and is zero when starting, even if appending or - reading a gzip stream from the middle of a file using gzdopen(). - - gztell(file) is equivalent to gzseek(file, 0L, SEEK_CUR) -*/ - -/* -ZEXTERN z_off_t ZEXPORT gzoffset OF((gzFile file)); - - Returns the current offset in the file being read or written. This offset - includes the count of bytes that precede the gzip stream, for example when - appending or when using gzdopen() for reading. When reading, the offset - does not include as yet unused buffered input. This information can be used - for a progress indicator. On error, gzoffset() returns -1. -*/ - -ZEXTERN int ZEXPORT gzeof OF((gzFile file)); -/* - Returns true (1) if the end-of-file indicator has been set while reading, - false (0) otherwise. Note that the end-of-file indicator is set only if the - read tried to go past the end of the input, but came up short. Therefore, - just like feof(), gzeof() may return false even if there is no more data to - read, in the event that the last read request was for the exact number of - bytes remaining in the input file. This will happen if the input file size - is an exact multiple of the buffer size. - - If gzeof() returns true, then the read functions will return no more data, - unless the end-of-file indicator is reset by gzclearerr() and the input file - has grown since the previous end of file was detected. -*/ - -ZEXTERN int ZEXPORT gzdirect OF((gzFile file)); -/* - Returns true (1) if file is being copied directly while reading, or false - (0) if file is a gzip stream being decompressed. - - If the input file is empty, gzdirect() will return true, since the input - does not contain a gzip stream. - - If gzdirect() is used immediately after gzopen() or gzdopen() it will - cause buffers to be allocated to allow reading the file to determine if it - is a gzip file. Therefore if gzbuffer() is used, it should be called before - gzdirect(). - - When writing, gzdirect() returns true (1) if transparent writing was - requested ("wT" for the gzopen() mode), or false (0) otherwise. (Note: - gzdirect() is not needed when writing. Transparent writing must be - explicitly requested, so the application already knows the answer. When - linking statically, using gzdirect() will include all of the zlib code for - gzip file reading and decompression, which may not be desired.) -*/ - -ZEXTERN int ZEXPORT gzclose OF((gzFile file)); -/* - Flushes all pending output if necessary, closes the compressed file and - deallocates the (de)compression state. Note that once file is closed, you - cannot call gzerror with file, since its structures have been deallocated. - gzclose must not be called more than once on the same file, just as free - must not be called more than once on the same allocation. - - gzclose will return Z_STREAM_ERROR if file is not valid, Z_ERRNO on a - file operation error, Z_MEM_ERROR if out of memory, Z_BUF_ERROR if the - last read ended in the middle of a gzip stream, or Z_OK on success. -*/ - -ZEXTERN int ZEXPORT gzclose_r OF((gzFile file)); -ZEXTERN int ZEXPORT gzclose_w OF((gzFile file)); -/* - Same as gzclose(), but gzclose_r() is only for use when reading, and - gzclose_w() is only for use when writing or appending. The advantage to - using these instead of gzclose() is that they avoid linking in zlib - compression or decompression code that is not used when only reading or only - writing respectively. If gzclose() is used, then both compression and - decompression code will be included the application when linking to a static - zlib library. -*/ - -ZEXTERN const char * ZEXPORT gzerror OF((gzFile file, int *errnum)); -/* - Returns the error message for the last error which occurred on the given - compressed file. errnum is set to zlib error number. If an error occurred - in the file system and not in the compression library, errnum is set to - Z_ERRNO and the application may consult errno to get the exact error code. - - The application must not modify the returned string. Future calls to - this function may invalidate the previously returned string. If file is - closed, then the string previously returned by gzerror will no longer be - available. - - gzerror() should be used to distinguish errors from end-of-file for those - functions above that do not distinguish those cases in their return values. -*/ - -ZEXTERN void ZEXPORT gzclearerr OF((gzFile file)); -/* - Clears the error and end-of-file flags for file. This is analogous to the - clearerr() function in stdio. This is useful for continuing to read a gzip - file that is being written concurrently. -*/ - -#endif /* !Z_SOLO */ - - /* checksum functions */ - -/* - These functions are not related to compression but are exported - anyway because they might be useful in applications using the compression - library. -*/ - -ZEXTERN uLong ZEXPORT adler32 OF((uLong adler, const Bytef *buf, uInt len)); -/* - Update a running Adler-32 checksum with the bytes buf[0..len-1] and - return the updated checksum. If buf is Z_NULL, this function returns the - required initial value for the checksum. - - An Adler-32 checksum is almost as reliable as a CRC-32 but can be computed - much faster. - - Usage example: - - uLong adler = adler32(0L, Z_NULL, 0); - - while (read_buffer(buffer, length) != EOF) { - adler = adler32(adler, buffer, length); - } - if (adler != original_adler) error(); -*/ - -ZEXTERN uLong ZEXPORT adler32_z OF((uLong adler, const Bytef *buf, - z_size_t len)); -/* - Same as adler32(), but with a size_t length. -*/ - -/* -ZEXTERN uLong ZEXPORT adler32_combine OF((uLong adler1, uLong adler2, - z_off_t len2)); - - Combine two Adler-32 checksums into one. For two sequences of bytes, seq1 - and seq2 with lengths len1 and len2, Adler-32 checksums were calculated for - each, adler1 and adler2. adler32_combine() returns the Adler-32 checksum of - seq1 and seq2 concatenated, requiring only adler1, adler2, and len2. Note - that the z_off_t type (like off_t) is a signed integer. If len2 is - negative, the result has no meaning or utility. -*/ - -ZEXTERN uLong ZEXPORT crc32 OF((uLong crc, const Bytef *buf, uInt len)); -/* - Update a running CRC-32 with the bytes buf[0..len-1] and return the - updated CRC-32. If buf is Z_NULL, this function returns the required - initial value for the crc. Pre- and post-conditioning (one's complement) is - performed within this function so it shouldn't be done by the application. - - Usage example: - - uLong crc = crc32(0L, Z_NULL, 0); - - while (read_buffer(buffer, length) != EOF) { - crc = crc32(crc, buffer, length); - } - if (crc != original_crc) error(); -*/ - -ZEXTERN uLong ZEXPORT crc32_z OF((uLong adler, const Bytef *buf, - z_size_t len)); -/* - Same as crc32(), but with a size_t length. -*/ - -/* -ZEXTERN uLong ZEXPORT crc32_combine OF((uLong crc1, uLong crc2, z_off_t len2)); - - Combine two CRC-32 check values into one. For two sequences of bytes, - seq1 and seq2 with lengths len1 and len2, CRC-32 check values were - calculated for each, crc1 and crc2. crc32_combine() returns the CRC-32 - check value of seq1 and seq2 concatenated, requiring only crc1, crc2, and - len2. -*/ - - - /* various hacks, don't look :) */ - -/* deflateInit and inflateInit are macros to allow checking the zlib version - * and the compiler's view of z_stream: - */ -ZEXTERN int ZEXPORT deflateInit_ OF((z_streamp strm, int level, - const char *version, int stream_size)); -ZEXTERN int ZEXPORT inflateInit_ OF((z_streamp strm, - const char *version, int stream_size)); -ZEXTERN int ZEXPORT deflateInit2_ OF((z_streamp strm, int level, int method, - int windowBits, int memLevel, - int strategy, const char *version, - int stream_size)); -ZEXTERN int ZEXPORT inflateInit2_ OF((z_streamp strm, int windowBits, - const char *version, int stream_size)); -ZEXTERN int ZEXPORT inflateBackInit_ OF((z_streamp strm, int windowBits, - unsigned char FAR *window, - const char *version, - int stream_size)); -#ifdef Z_PREFIX_SET -# define z_deflateInit(strm, level) \ - deflateInit_((strm), (level), ZLIB_VERSION, (int)sizeof(z_stream)) -# define z_inflateInit(strm) \ - inflateInit_((strm), ZLIB_VERSION, (int)sizeof(z_stream)) -# define z_deflateInit2(strm, level, method, windowBits, memLevel, strategy) \ - deflateInit2_((strm),(level),(method),(windowBits),(memLevel),\ - (strategy), ZLIB_VERSION, (int)sizeof(z_stream)) -# define z_inflateInit2(strm, windowBits) \ - inflateInit2_((strm), (windowBits), ZLIB_VERSION, \ - (int)sizeof(z_stream)) -# define z_inflateBackInit(strm, windowBits, window) \ - inflateBackInit_((strm), (windowBits), (window), \ - ZLIB_VERSION, (int)sizeof(z_stream)) -#else -# define deflateInit(strm, level) \ - deflateInit_((strm), (level), ZLIB_VERSION, (int)sizeof(z_stream)) -# define inflateInit(strm) \ - inflateInit_((strm), ZLIB_VERSION, (int)sizeof(z_stream)) -# define deflateInit2(strm, level, method, windowBits, memLevel, strategy) \ - deflateInit2_((strm),(level),(method),(windowBits),(memLevel),\ - (strategy), ZLIB_VERSION, (int)sizeof(z_stream)) -# define inflateInit2(strm, windowBits) \ - inflateInit2_((strm), (windowBits), ZLIB_VERSION, \ - (int)sizeof(z_stream)) -# define inflateBackInit(strm, windowBits, window) \ - inflateBackInit_((strm), (windowBits), (window), \ - ZLIB_VERSION, (int)sizeof(z_stream)) -#endif - -#ifndef Z_SOLO - -/* gzgetc() macro and its supporting function and exposed data structure. Note - * that the real internal state is much larger than the exposed structure. - * This abbreviated structure exposes just enough for the gzgetc() macro. The - * user should not mess with these exposed elements, since their names or - * behavior could change in the future, perhaps even capriciously. They can - * only be used by the gzgetc() macro. You have been warned. - */ -struct gzFile_s { - unsigned have; - unsigned char *next; - z_off64_t pos; -}; -ZEXTERN int ZEXPORT gzgetc_ OF((gzFile file)); /* backward compatibility */ -#ifdef Z_PREFIX_SET -# undef z_gzgetc -# define z_gzgetc(g) \ - ((g)->have ? ((g)->have--, (g)->pos++, *((g)->next)++) : (gzgetc)(g)) -#else -# define gzgetc(g) \ - ((g)->have ? ((g)->have--, (g)->pos++, *((g)->next)++) : (gzgetc)(g)) -#endif - -/* provide 64-bit offset functions if _LARGEFILE64_SOURCE defined, and/or - * change the regular functions to 64 bits if _FILE_OFFSET_BITS is 64 (if - * both are true, the application gets the *64 functions, and the regular - * functions are changed to 64 bits) -- in case these are set on systems - * without large file support, _LFS64_LARGEFILE must also be true - */ -#ifdef Z_LARGE64 - ZEXTERN gzFile ZEXPORT gzopen64 OF((const char *, const char *)); - ZEXTERN z_off64_t ZEXPORT gzseek64 OF((gzFile, z_off64_t, int)); - ZEXTERN z_off64_t ZEXPORT gztell64 OF((gzFile)); - ZEXTERN z_off64_t ZEXPORT gzoffset64 OF((gzFile)); - ZEXTERN uLong ZEXPORT adler32_combine64 OF((uLong, uLong, z_off64_t)); - ZEXTERN uLong ZEXPORT crc32_combine64 OF((uLong, uLong, z_off64_t)); -#endif - -#if !defined(ZLIB_INTERNAL) && defined(Z_WANT64) -# ifdef Z_PREFIX_SET -# define z_gzopen z_gzopen64 -# define z_gzseek z_gzseek64 -# define z_gztell z_gztell64 -# define z_gzoffset z_gzoffset64 -# define z_adler32_combine z_adler32_combine64 -# define z_crc32_combine z_crc32_combine64 -# else -# define gzopen gzopen64 -# define gzseek gzseek64 -# define gztell gztell64 -# define gzoffset gzoffset64 -# define adler32_combine adler32_combine64 -# define crc32_combine crc32_combine64 -# endif -# ifndef Z_LARGE64 - ZEXTERN gzFile ZEXPORT gzopen64 OF((const char *, const char *)); - ZEXTERN z_off_t ZEXPORT gzseek64 OF((gzFile, z_off_t, int)); - ZEXTERN z_off_t ZEXPORT gztell64 OF((gzFile)); - ZEXTERN z_off_t ZEXPORT gzoffset64 OF((gzFile)); - ZEXTERN uLong ZEXPORT adler32_combine64 OF((uLong, uLong, z_off_t)); - ZEXTERN uLong ZEXPORT crc32_combine64 OF((uLong, uLong, z_off_t)); -# endif -#else - ZEXTERN gzFile ZEXPORT gzopen OF((const char *, const char *)); - ZEXTERN z_off_t ZEXPORT gzseek OF((gzFile, z_off_t, int)); - ZEXTERN z_off_t ZEXPORT gztell OF((gzFile)); - ZEXTERN z_off_t ZEXPORT gzoffset OF((gzFile)); - ZEXTERN uLong ZEXPORT adler32_combine OF((uLong, uLong, z_off_t)); - ZEXTERN uLong ZEXPORT crc32_combine OF((uLong, uLong, z_off_t)); -#endif - -#else /* Z_SOLO */ - - ZEXTERN uLong ZEXPORT adler32_combine OF((uLong, uLong, z_off_t)); - ZEXTERN uLong ZEXPORT crc32_combine OF((uLong, uLong, z_off_t)); - -#endif /* !Z_SOLO */ - -/* undocumented functions */ -ZEXTERN const char * ZEXPORT zError OF((int)); -ZEXTERN int ZEXPORT inflateSyncPoint OF((z_streamp)); -ZEXTERN const z_crc_t FAR * ZEXPORT get_crc_table OF((void)); -ZEXTERN int ZEXPORT inflateUndermine OF((z_streamp, int)); -ZEXTERN int ZEXPORT inflateValidate OF((z_streamp, int)); -ZEXTERN unsigned long ZEXPORT inflateCodesUsed OF ((z_streamp)); -ZEXTERN int ZEXPORT inflateResetKeep OF((z_streamp)); -ZEXTERN int ZEXPORT deflateResetKeep OF((z_streamp)); -#if (defined(_WIN32) || defined(__CYGWIN__)) && !defined(Z_SOLO) -ZEXTERN gzFile ZEXPORT gzopen_w OF((const wchar_t *path, - const char *mode)); -#endif -#if defined(STDC) || defined(Z_HAVE_STDARG_H) -# ifndef Z_SOLO -ZEXTERN int ZEXPORTVA gzvprintf Z_ARG((gzFile file, - const char *format, - va_list va)); -# endif -#endif - -#ifdef __cplusplus -} -#endif - -#endif /* ZLIB_H */ diff --git a/dep/zlib/src/adler32.c b/dep/zlib/src/adler32.c deleted file mode 100644 index d0be4380a..000000000 --- a/dep/zlib/src/adler32.c +++ /dev/null @@ -1,186 +0,0 @@ -/* adler32.c -- compute the Adler-32 checksum of a data stream - * Copyright (C) 1995-2011, 2016 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* @(#) $Id$ */ - -#include "zutil.h" - -local uLong adler32_combine_ OF((uLong adler1, uLong adler2, z_off64_t len2)); - -#define BASE 65521U /* largest prime smaller than 65536 */ -#define NMAX 5552 -/* NMAX is the largest n such that 255n(n+1)/2 + (n+1)(BASE-1) <= 2^32-1 */ - -#define DO1(buf,i) {adler += (buf)[i]; sum2 += adler;} -#define DO2(buf,i) DO1(buf,i); DO1(buf,i+1); -#define DO4(buf,i) DO2(buf,i); DO2(buf,i+2); -#define DO8(buf,i) DO4(buf,i); DO4(buf,i+4); -#define DO16(buf) DO8(buf,0); DO8(buf,8); - -/* use NO_DIVIDE if your processor does not do division in hardware -- - try it both ways to see which is faster */ -#ifdef NO_DIVIDE -/* note that this assumes BASE is 65521, where 65536 % 65521 == 15 - (thank you to John Reiser for pointing this out) */ -# define CHOP(a) \ - do { \ - unsigned long tmp = a >> 16; \ - a &= 0xffffUL; \ - a += (tmp << 4) - tmp; \ - } while (0) -# define MOD28(a) \ - do { \ - CHOP(a); \ - if (a >= BASE) a -= BASE; \ - } while (0) -# define MOD(a) \ - do { \ - CHOP(a); \ - MOD28(a); \ - } while (0) -# define MOD63(a) \ - do { /* this assumes a is not negative */ \ - z_off64_t tmp = a >> 32; \ - a &= 0xffffffffL; \ - a += (tmp << 8) - (tmp << 5) + tmp; \ - tmp = a >> 16; \ - a &= 0xffffL; \ - a += (tmp << 4) - tmp; \ - tmp = a >> 16; \ - a &= 0xffffL; \ - a += (tmp << 4) - tmp; \ - if (a >= BASE) a -= BASE; \ - } while (0) -#else -# define MOD(a) a %= BASE -# define MOD28(a) a %= BASE -# define MOD63(a) a %= BASE -#endif - -/* ========================================================================= */ -uLong ZEXPORT adler32_z(adler, buf, len) - uLong adler; - const Bytef *buf; - z_size_t len; -{ - unsigned long sum2; - unsigned n; - - /* split Adler-32 into component sums */ - sum2 = (adler >> 16) & 0xffff; - adler &= 0xffff; - - /* in case user likes doing a byte at a time, keep it fast */ - if (len == 1) { - adler += buf[0]; - if (adler >= BASE) - adler -= BASE; - sum2 += adler; - if (sum2 >= BASE) - sum2 -= BASE; - return adler | (sum2 << 16); - } - - /* initial Adler-32 value (deferred check for len == 1 speed) */ - if (buf == Z_NULL) - return 1L; - - /* in case short lengths are provided, keep it somewhat fast */ - if (len < 16) { - while (len--) { - adler += *buf++; - sum2 += adler; - } - if (adler >= BASE) - adler -= BASE; - MOD28(sum2); /* only added so many BASE's */ - return adler | (sum2 << 16); - } - - /* do length NMAX blocks -- requires just one modulo operation */ - while (len >= NMAX) { - len -= NMAX; - n = NMAX / 16; /* NMAX is divisible by 16 */ - do { - DO16(buf); /* 16 sums unrolled */ - buf += 16; - } while (--n); - MOD(adler); - MOD(sum2); - } - - /* do remaining bytes (less than NMAX, still just one modulo) */ - if (len) { /* avoid modulos if none remaining */ - while (len >= 16) { - len -= 16; - DO16(buf); - buf += 16; - } - while (len--) { - adler += *buf++; - sum2 += adler; - } - MOD(adler); - MOD(sum2); - } - - /* return recombined sums */ - return adler | (sum2 << 16); -} - -/* ========================================================================= */ -uLong ZEXPORT adler32(adler, buf, len) - uLong adler; - const Bytef *buf; - uInt len; -{ - return adler32_z(adler, buf, len); -} - -/* ========================================================================= */ -local uLong adler32_combine_(adler1, adler2, len2) - uLong adler1; - uLong adler2; - z_off64_t len2; -{ - unsigned long sum1; - unsigned long sum2; - unsigned rem; - - /* for negative len, return invalid adler32 as a clue for debugging */ - if (len2 < 0) - return 0xffffffffUL; - - /* the derivation of this formula is left as an exercise for the reader */ - MOD63(len2); /* assumes len2 >= 0 */ - rem = (unsigned)len2; - sum1 = adler1 & 0xffff; - sum2 = rem * sum1; - MOD(sum2); - sum1 += (adler2 & 0xffff) + BASE - 1; - sum2 += ((adler1 >> 16) & 0xffff) + ((adler2 >> 16) & 0xffff) + BASE - rem; - if (sum1 >= BASE) sum1 -= BASE; - if (sum1 >= BASE) sum1 -= BASE; - if (sum2 >= ((unsigned long)BASE << 1)) sum2 -= ((unsigned long)BASE << 1); - if (sum2 >= BASE) sum2 -= BASE; - return sum1 | (sum2 << 16); -} - -/* ========================================================================= */ -uLong ZEXPORT adler32_combine(adler1, adler2, len2) - uLong adler1; - uLong adler2; - z_off_t len2; -{ - return adler32_combine_(adler1, adler2, len2); -} - -uLong ZEXPORT adler32_combine64(adler1, adler2, len2) - uLong adler1; - uLong adler2; - z_off64_t len2; -{ - return adler32_combine_(adler1, adler2, len2); -} diff --git a/dep/zlib/src/compress.c b/dep/zlib/src/compress.c deleted file mode 100644 index e2db404ab..000000000 --- a/dep/zlib/src/compress.c +++ /dev/null @@ -1,86 +0,0 @@ -/* compress.c -- compress a memory buffer - * Copyright (C) 1995-2005, 2014, 2016 Jean-loup Gailly, Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* @(#) $Id$ */ - -#define ZLIB_INTERNAL -#include "zlib.h" - -/* =========================================================================== - Compresses the source buffer into the destination buffer. The level - parameter has the same meaning as in deflateInit. sourceLen is the byte - length of the source buffer. Upon entry, destLen is the total size of the - destination buffer, which must be at least 0.1% larger than sourceLen plus - 12 bytes. Upon exit, destLen is the actual size of the compressed buffer. - - compress2 returns Z_OK if success, Z_MEM_ERROR if there was not enough - memory, Z_BUF_ERROR if there was not enough room in the output buffer, - Z_STREAM_ERROR if the level parameter is invalid. -*/ -int ZEXPORT compress2 (dest, destLen, source, sourceLen, level) - Bytef *dest; - uLongf *destLen; - const Bytef *source; - uLong sourceLen; - int level; -{ - z_stream stream; - int err; - const uInt max = (uInt)-1; - uLong left; - - left = *destLen; - *destLen = 0; - - stream.zalloc = (alloc_func)0; - stream.zfree = (free_func)0; - stream.opaque = (voidpf)0; - - err = deflateInit(&stream, level); - if (err != Z_OK) return err; - - stream.next_out = dest; - stream.avail_out = 0; - stream.next_in = (z_const Bytef *)source; - stream.avail_in = 0; - - do { - if (stream.avail_out == 0) { - stream.avail_out = left > (uLong)max ? max : (uInt)left; - left -= stream.avail_out; - } - if (stream.avail_in == 0) { - stream.avail_in = sourceLen > (uLong)max ? max : (uInt)sourceLen; - sourceLen -= stream.avail_in; - } - err = deflate(&stream, sourceLen ? Z_NO_FLUSH : Z_FINISH); - } while (err == Z_OK); - - *destLen = stream.total_out; - deflateEnd(&stream); - return err == Z_STREAM_END ? Z_OK : err; -} - -/* =========================================================================== - */ -int ZEXPORT compress (dest, destLen, source, sourceLen) - Bytef *dest; - uLongf *destLen; - const Bytef *source; - uLong sourceLen; -{ - return compress2(dest, destLen, source, sourceLen, Z_DEFAULT_COMPRESSION); -} - -/* =========================================================================== - If the default memLevel or windowBits for deflateInit() is changed, then - this function needs to be updated. - */ -uLong ZEXPORT compressBound (sourceLen) - uLong sourceLen; -{ - return sourceLen + (sourceLen >> 12) + (sourceLen >> 14) + - (sourceLen >> 25) + 13; -} diff --git a/dep/zlib/src/crc32.c b/dep/zlib/src/crc32.c deleted file mode 100644 index 9580440c0..000000000 --- a/dep/zlib/src/crc32.c +++ /dev/null @@ -1,442 +0,0 @@ -/* crc32.c -- compute the CRC-32 of a data stream - * Copyright (C) 1995-2006, 2010, 2011, 2012, 2016 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - * - * Thanks to Rodney Brown for his contribution of faster - * CRC methods: exclusive-oring 32 bits of data at a time, and pre-computing - * tables for updating the shift register in one step with three exclusive-ors - * instead of four steps with four exclusive-ors. This results in about a - * factor of two increase in speed on a Power PC G4 (PPC7455) using gcc -O3. - */ - -/* @(#) $Id$ */ - -/* - Note on the use of DYNAMIC_CRC_TABLE: there is no mutex or semaphore - protection on the static variables used to control the first-use generation - of the crc tables. Therefore, if you #define DYNAMIC_CRC_TABLE, you should - first call get_crc_table() to initialize the tables before allowing more than - one thread to use crc32(). - - DYNAMIC_CRC_TABLE and MAKECRCH can be #defined to write out crc32.h. - */ - -#ifdef MAKECRCH -# include -# ifndef DYNAMIC_CRC_TABLE -# define DYNAMIC_CRC_TABLE -# endif /* !DYNAMIC_CRC_TABLE */ -#endif /* MAKECRCH */ - -#include "zutil.h" /* for STDC and FAR definitions */ - -/* Definitions for doing the crc four data bytes at a time. */ -#if !defined(NOBYFOUR) && defined(Z_U4) -# define BYFOUR -#endif -#ifdef BYFOUR - local unsigned long crc32_little OF((unsigned long, - const unsigned char FAR *, z_size_t)); - local unsigned long crc32_big OF((unsigned long, - const unsigned char FAR *, z_size_t)); -# define TBLS 8 -#else -# define TBLS 1 -#endif /* BYFOUR */ - -/* Local functions for crc concatenation */ -local unsigned long gf2_matrix_times OF((unsigned long *mat, - unsigned long vec)); -local void gf2_matrix_square OF((unsigned long *square, unsigned long *mat)); -local uLong crc32_combine_ OF((uLong crc1, uLong crc2, z_off64_t len2)); - - -#ifdef DYNAMIC_CRC_TABLE - -local volatile int crc_table_empty = 1; -local z_crc_t FAR crc_table[TBLS][256]; -local void make_crc_table OF((void)); -#ifdef MAKECRCH - local void write_table OF((FILE *, const z_crc_t FAR *)); -#endif /* MAKECRCH */ -/* - Generate tables for a byte-wise 32-bit CRC calculation on the polynomial: - x^32+x^26+x^23+x^22+x^16+x^12+x^11+x^10+x^8+x^7+x^5+x^4+x^2+x+1. - - Polynomials over GF(2) are represented in binary, one bit per coefficient, - with the lowest powers in the most significant bit. Then adding polynomials - is just exclusive-or, and multiplying a polynomial by x is a right shift by - one. If we call the above polynomial p, and represent a byte as the - polynomial q, also with the lowest power in the most significant bit (so the - byte 0xb1 is the polynomial x^7+x^3+x+1), then the CRC is (q*x^32) mod p, - where a mod b means the remainder after dividing a by b. - - This calculation is done using the shift-register method of multiplying and - taking the remainder. The register is initialized to zero, and for each - incoming bit, x^32 is added mod p to the register if the bit is a one (where - x^32 mod p is p+x^32 = x^26+...+1), and the register is multiplied mod p by - x (which is shifting right by one and adding x^32 mod p if the bit shifted - out is a one). We start with the highest power (least significant bit) of - q and repeat for all eight bits of q. - - The first table is simply the CRC of all possible eight bit values. This is - all the information needed to generate CRCs on data a byte at a time for all - combinations of CRC register values and incoming bytes. The remaining tables - allow for word-at-a-time CRC calculation for both big-endian and little- - endian machines, where a word is four bytes. -*/ -local void make_crc_table() -{ - z_crc_t c; - int n, k; - z_crc_t poly; /* polynomial exclusive-or pattern */ - /* terms of polynomial defining this crc (except x^32): */ - static volatile int first = 1; /* flag to limit concurrent making */ - static const unsigned char p[] = {0,1,2,4,5,7,8,10,11,12,16,22,23,26}; - - /* See if another task is already doing this (not thread-safe, but better - than nothing -- significantly reduces duration of vulnerability in - case the advice about DYNAMIC_CRC_TABLE is ignored) */ - if (first) { - first = 0; - - /* make exclusive-or pattern from polynomial (0xedb88320UL) */ - poly = 0; - for (n = 0; n < (int)(sizeof(p)/sizeof(unsigned char)); n++) - poly |= (z_crc_t)1 << (31 - p[n]); - - /* generate a crc for every 8-bit value */ - for (n = 0; n < 256; n++) { - c = (z_crc_t)n; - for (k = 0; k < 8; k++) - c = c & 1 ? poly ^ (c >> 1) : c >> 1; - crc_table[0][n] = c; - } - -#ifdef BYFOUR - /* generate crc for each value followed by one, two, and three zeros, - and then the byte reversal of those as well as the first table */ - for (n = 0; n < 256; n++) { - c = crc_table[0][n]; - crc_table[4][n] = ZSWAP32(c); - for (k = 1; k < 4; k++) { - c = crc_table[0][c & 0xff] ^ (c >> 8); - crc_table[k][n] = c; - crc_table[k + 4][n] = ZSWAP32(c); - } - } -#endif /* BYFOUR */ - - crc_table_empty = 0; - } - else { /* not first */ - /* wait for the other guy to finish (not efficient, but rare) */ - while (crc_table_empty) - ; - } - -#ifdef MAKECRCH - /* write out CRC tables to crc32.h */ - { - FILE *out; - - out = fopen("crc32.h", "w"); - if (out == NULL) return; - fprintf(out, "/* crc32.h -- tables for rapid CRC calculation\n"); - fprintf(out, " * Generated automatically by crc32.c\n */\n\n"); - fprintf(out, "local const z_crc_t FAR "); - fprintf(out, "crc_table[TBLS][256] =\n{\n {\n"); - write_table(out, crc_table[0]); -# ifdef BYFOUR - fprintf(out, "#ifdef BYFOUR\n"); - for (k = 1; k < 8; k++) { - fprintf(out, " },\n {\n"); - write_table(out, crc_table[k]); - } - fprintf(out, "#endif\n"); -# endif /* BYFOUR */ - fprintf(out, " }\n};\n"); - fclose(out); - } -#endif /* MAKECRCH */ -} - -#ifdef MAKECRCH -local void write_table(out, table) - FILE *out; - const z_crc_t FAR *table; -{ - int n; - - for (n = 0; n < 256; n++) - fprintf(out, "%s0x%08lxUL%s", n % 5 ? "" : " ", - (unsigned long)(table[n]), - n == 255 ? "\n" : (n % 5 == 4 ? ",\n" : ", ")); -} -#endif /* MAKECRCH */ - -#else /* !DYNAMIC_CRC_TABLE */ -/* ======================================================================== - * Tables of CRC-32s of all single-byte values, made by make_crc_table(). - */ -#include "crc32.h" -#endif /* DYNAMIC_CRC_TABLE */ - -/* ========================================================================= - * This function can be used by asm versions of crc32() - */ -const z_crc_t FAR * ZEXPORT get_crc_table() -{ -#ifdef DYNAMIC_CRC_TABLE - if (crc_table_empty) - make_crc_table(); -#endif /* DYNAMIC_CRC_TABLE */ - return (const z_crc_t FAR *)crc_table; -} - -/* ========================================================================= */ -#define DO1 crc = crc_table[0][((int)crc ^ (*buf++)) & 0xff] ^ (crc >> 8) -#define DO8 DO1; DO1; DO1; DO1; DO1; DO1; DO1; DO1 - -/* ========================================================================= */ -unsigned long ZEXPORT crc32_z(crc, buf, len) - unsigned long crc; - const unsigned char FAR *buf; - z_size_t len; -{ - if (buf == Z_NULL) return 0UL; - -#ifdef DYNAMIC_CRC_TABLE - if (crc_table_empty) - make_crc_table(); -#endif /* DYNAMIC_CRC_TABLE */ - -#ifdef BYFOUR - if (sizeof(void *) == sizeof(ptrdiff_t)) { - z_crc_t endian; - - endian = 1; - if (*((unsigned char *)(&endian))) - return crc32_little(crc, buf, len); - else - return crc32_big(crc, buf, len); - } -#endif /* BYFOUR */ - crc = crc ^ 0xffffffffUL; - while (len >= 8) { - DO8; - len -= 8; - } - if (len) do { - DO1; - } while (--len); - return crc ^ 0xffffffffUL; -} - -/* ========================================================================= */ -unsigned long ZEXPORT crc32(crc, buf, len) - unsigned long crc; - const unsigned char FAR *buf; - uInt len; -{ - return crc32_z(crc, buf, len); -} - -#ifdef BYFOUR - -/* - This BYFOUR code accesses the passed unsigned char * buffer with a 32-bit - integer pointer type. This violates the strict aliasing rule, where a - compiler can assume, for optimization purposes, that two pointers to - fundamentally different types won't ever point to the same memory. This can - manifest as a problem only if one of the pointers is written to. This code - only reads from those pointers. So long as this code remains isolated in - this compilation unit, there won't be a problem. For this reason, this code - should not be copied and pasted into a compilation unit in which other code - writes to the buffer that is passed to these routines. - */ - -/* ========================================================================= */ -#define DOLIT4 c ^= *buf4++; \ - c = crc_table[3][c & 0xff] ^ crc_table[2][(c >> 8) & 0xff] ^ \ - crc_table[1][(c >> 16) & 0xff] ^ crc_table[0][c >> 24] -#define DOLIT32 DOLIT4; DOLIT4; DOLIT4; DOLIT4; DOLIT4; DOLIT4; DOLIT4; DOLIT4 - -/* ========================================================================= */ -local unsigned long crc32_little(crc, buf, len) - unsigned long crc; - const unsigned char FAR *buf; - z_size_t len; -{ - register z_crc_t c; - register const z_crc_t FAR *buf4; - - c = (z_crc_t)crc; - c = ~c; - while (len && ((ptrdiff_t)buf & 3)) { - c = crc_table[0][(c ^ *buf++) & 0xff] ^ (c >> 8); - len--; - } - - buf4 = (const z_crc_t FAR *)(const void FAR *)buf; - while (len >= 32) { - DOLIT32; - len -= 32; - } - while (len >= 4) { - DOLIT4; - len -= 4; - } - buf = (const unsigned char FAR *)buf4; - - if (len) do { - c = crc_table[0][(c ^ *buf++) & 0xff] ^ (c >> 8); - } while (--len); - c = ~c; - return (unsigned long)c; -} - -/* ========================================================================= */ -#define DOBIG4 c ^= *buf4++; \ - c = crc_table[4][c & 0xff] ^ crc_table[5][(c >> 8) & 0xff] ^ \ - crc_table[6][(c >> 16) & 0xff] ^ crc_table[7][c >> 24] -#define DOBIG32 DOBIG4; DOBIG4; DOBIG4; DOBIG4; DOBIG4; DOBIG4; DOBIG4; DOBIG4 - -/* ========================================================================= */ -local unsigned long crc32_big(crc, buf, len) - unsigned long crc; - const unsigned char FAR *buf; - z_size_t len; -{ - register z_crc_t c; - register const z_crc_t FAR *buf4; - - c = ZSWAP32((z_crc_t)crc); - c = ~c; - while (len && ((ptrdiff_t)buf & 3)) { - c = crc_table[4][(c >> 24) ^ *buf++] ^ (c << 8); - len--; - } - - buf4 = (const z_crc_t FAR *)(const void FAR *)buf; - while (len >= 32) { - DOBIG32; - len -= 32; - } - while (len >= 4) { - DOBIG4; - len -= 4; - } - buf = (const unsigned char FAR *)buf4; - - if (len) do { - c = crc_table[4][(c >> 24) ^ *buf++] ^ (c << 8); - } while (--len); - c = ~c; - return (unsigned long)(ZSWAP32(c)); -} - -#endif /* BYFOUR */ - -#define GF2_DIM 32 /* dimension of GF(2) vectors (length of CRC) */ - -/* ========================================================================= */ -local unsigned long gf2_matrix_times(mat, vec) - unsigned long *mat; - unsigned long vec; -{ - unsigned long sum; - - sum = 0; - while (vec) { - if (vec & 1) - sum ^= *mat; - vec >>= 1; - mat++; - } - return sum; -} - -/* ========================================================================= */ -local void gf2_matrix_square(square, mat) - unsigned long *square; - unsigned long *mat; -{ - int n; - - for (n = 0; n < GF2_DIM; n++) - square[n] = gf2_matrix_times(mat, mat[n]); -} - -/* ========================================================================= */ -local uLong crc32_combine_(crc1, crc2, len2) - uLong crc1; - uLong crc2; - z_off64_t len2; -{ - int n; - unsigned long row; - unsigned long even[GF2_DIM]; /* even-power-of-two zeros operator */ - unsigned long odd[GF2_DIM]; /* odd-power-of-two zeros operator */ - - /* degenerate case (also disallow negative lengths) */ - if (len2 <= 0) - return crc1; - - /* put operator for one zero bit in odd */ - odd[0] = 0xedb88320UL; /* CRC-32 polynomial */ - row = 1; - for (n = 1; n < GF2_DIM; n++) { - odd[n] = row; - row <<= 1; - } - - /* put operator for two zero bits in even */ - gf2_matrix_square(even, odd); - - /* put operator for four zero bits in odd */ - gf2_matrix_square(odd, even); - - /* apply len2 zeros to crc1 (first square will put the operator for one - zero byte, eight zero bits, in even) */ - do { - /* apply zeros operator for this bit of len2 */ - gf2_matrix_square(even, odd); - if (len2 & 1) - crc1 = gf2_matrix_times(even, crc1); - len2 >>= 1; - - /* if no more bits set, then done */ - if (len2 == 0) - break; - - /* another iteration of the loop with odd and even swapped */ - gf2_matrix_square(odd, even); - if (len2 & 1) - crc1 = gf2_matrix_times(odd, crc1); - len2 >>= 1; - - /* if no more bits set, then done */ - } while (len2 != 0); - - /* return combined crc */ - crc1 ^= crc2; - return crc1; -} - -/* ========================================================================= */ -uLong ZEXPORT crc32_combine(crc1, crc2, len2) - uLong crc1; - uLong crc2; - z_off_t len2; -{ - return crc32_combine_(crc1, crc2, len2); -} - -uLong ZEXPORT crc32_combine64(crc1, crc2, len2) - uLong crc1; - uLong crc2; - z_off64_t len2; -{ - return crc32_combine_(crc1, crc2, len2); -} diff --git a/dep/zlib/src/crc32.h b/dep/zlib/src/crc32.h deleted file mode 100644 index 9e0c77810..000000000 --- a/dep/zlib/src/crc32.h +++ /dev/null @@ -1,441 +0,0 @@ -/* crc32.h -- tables for rapid CRC calculation - * Generated automatically by crc32.c - */ - -local const z_crc_t FAR crc_table[TBLS][256] = -{ - { - 0x00000000UL, 0x77073096UL, 0xee0e612cUL, 0x990951baUL, 0x076dc419UL, - 0x706af48fUL, 0xe963a535UL, 0x9e6495a3UL, 0x0edb8832UL, 0x79dcb8a4UL, - 0xe0d5e91eUL, 0x97d2d988UL, 0x09b64c2bUL, 0x7eb17cbdUL, 0xe7b82d07UL, - 0x90bf1d91UL, 0x1db71064UL, 0x6ab020f2UL, 0xf3b97148UL, 0x84be41deUL, - 0x1adad47dUL, 0x6ddde4ebUL, 0xf4d4b551UL, 0x83d385c7UL, 0x136c9856UL, - 0x646ba8c0UL, 0xfd62f97aUL, 0x8a65c9ecUL, 0x14015c4fUL, 0x63066cd9UL, - 0xfa0f3d63UL, 0x8d080df5UL, 0x3b6e20c8UL, 0x4c69105eUL, 0xd56041e4UL, - 0xa2677172UL, 0x3c03e4d1UL, 0x4b04d447UL, 0xd20d85fdUL, 0xa50ab56bUL, - 0x35b5a8faUL, 0x42b2986cUL, 0xdbbbc9d6UL, 0xacbcf940UL, 0x32d86ce3UL, - 0x45df5c75UL, 0xdcd60dcfUL, 0xabd13d59UL, 0x26d930acUL, 0x51de003aUL, - 0xc8d75180UL, 0xbfd06116UL, 0x21b4f4b5UL, 0x56b3c423UL, 0xcfba9599UL, - 0xb8bda50fUL, 0x2802b89eUL, 0x5f058808UL, 0xc60cd9b2UL, 0xb10be924UL, - 0x2f6f7c87UL, 0x58684c11UL, 0xc1611dabUL, 0xb6662d3dUL, 0x76dc4190UL, - 0x01db7106UL, 0x98d220bcUL, 0xefd5102aUL, 0x71b18589UL, 0x06b6b51fUL, - 0x9fbfe4a5UL, 0xe8b8d433UL, 0x7807c9a2UL, 0x0f00f934UL, 0x9609a88eUL, - 0xe10e9818UL, 0x7f6a0dbbUL, 0x086d3d2dUL, 0x91646c97UL, 0xe6635c01UL, - 0x6b6b51f4UL, 0x1c6c6162UL, 0x856530d8UL, 0xf262004eUL, 0x6c0695edUL, - 0x1b01a57bUL, 0x8208f4c1UL, 0xf50fc457UL, 0x65b0d9c6UL, 0x12b7e950UL, - 0x8bbeb8eaUL, 0xfcb9887cUL, 0x62dd1ddfUL, 0x15da2d49UL, 0x8cd37cf3UL, - 0xfbd44c65UL, 0x4db26158UL, 0x3ab551ceUL, 0xa3bc0074UL, 0xd4bb30e2UL, - 0x4adfa541UL, 0x3dd895d7UL, 0xa4d1c46dUL, 0xd3d6f4fbUL, 0x4369e96aUL, - 0x346ed9fcUL, 0xad678846UL, 0xda60b8d0UL, 0x44042d73UL, 0x33031de5UL, - 0xaa0a4c5fUL, 0xdd0d7cc9UL, 0x5005713cUL, 0x270241aaUL, 0xbe0b1010UL, - 0xc90c2086UL, 0x5768b525UL, 0x206f85b3UL, 0xb966d409UL, 0xce61e49fUL, - 0x5edef90eUL, 0x29d9c998UL, 0xb0d09822UL, 0xc7d7a8b4UL, 0x59b33d17UL, - 0x2eb40d81UL, 0xb7bd5c3bUL, 0xc0ba6cadUL, 0xedb88320UL, 0x9abfb3b6UL, - 0x03b6e20cUL, 0x74b1d29aUL, 0xead54739UL, 0x9dd277afUL, 0x04db2615UL, - 0x73dc1683UL, 0xe3630b12UL, 0x94643b84UL, 0x0d6d6a3eUL, 0x7a6a5aa8UL, - 0xe40ecf0bUL, 0x9309ff9dUL, 0x0a00ae27UL, 0x7d079eb1UL, 0xf00f9344UL, - 0x8708a3d2UL, 0x1e01f268UL, 0x6906c2feUL, 0xf762575dUL, 0x806567cbUL, - 0x196c3671UL, 0x6e6b06e7UL, 0xfed41b76UL, 0x89d32be0UL, 0x10da7a5aUL, - 0x67dd4accUL, 0xf9b9df6fUL, 0x8ebeeff9UL, 0x17b7be43UL, 0x60b08ed5UL, - 0xd6d6a3e8UL, 0xa1d1937eUL, 0x38d8c2c4UL, 0x4fdff252UL, 0xd1bb67f1UL, - 0xa6bc5767UL, 0x3fb506ddUL, 0x48b2364bUL, 0xd80d2bdaUL, 0xaf0a1b4cUL, - 0x36034af6UL, 0x41047a60UL, 0xdf60efc3UL, 0xa867df55UL, 0x316e8eefUL, - 0x4669be79UL, 0xcb61b38cUL, 0xbc66831aUL, 0x256fd2a0UL, 0x5268e236UL, - 0xcc0c7795UL, 0xbb0b4703UL, 0x220216b9UL, 0x5505262fUL, 0xc5ba3bbeUL, - 0xb2bd0b28UL, 0x2bb45a92UL, 0x5cb36a04UL, 0xc2d7ffa7UL, 0xb5d0cf31UL, - 0x2cd99e8bUL, 0x5bdeae1dUL, 0x9b64c2b0UL, 0xec63f226UL, 0x756aa39cUL, - 0x026d930aUL, 0x9c0906a9UL, 0xeb0e363fUL, 0x72076785UL, 0x05005713UL, - 0x95bf4a82UL, 0xe2b87a14UL, 0x7bb12baeUL, 0x0cb61b38UL, 0x92d28e9bUL, - 0xe5d5be0dUL, 0x7cdcefb7UL, 0x0bdbdf21UL, 0x86d3d2d4UL, 0xf1d4e242UL, - 0x68ddb3f8UL, 0x1fda836eUL, 0x81be16cdUL, 0xf6b9265bUL, 0x6fb077e1UL, - 0x18b74777UL, 0x88085ae6UL, 0xff0f6a70UL, 0x66063bcaUL, 0x11010b5cUL, - 0x8f659effUL, 0xf862ae69UL, 0x616bffd3UL, 0x166ccf45UL, 0xa00ae278UL, - 0xd70dd2eeUL, 0x4e048354UL, 0x3903b3c2UL, 0xa7672661UL, 0xd06016f7UL, - 0x4969474dUL, 0x3e6e77dbUL, 0xaed16a4aUL, 0xd9d65adcUL, 0x40df0b66UL, - 0x37d83bf0UL, 0xa9bcae53UL, 0xdebb9ec5UL, 0x47b2cf7fUL, 0x30b5ffe9UL, - 0xbdbdf21cUL, 0xcabac28aUL, 0x53b39330UL, 0x24b4a3a6UL, 0xbad03605UL, - 0xcdd70693UL, 0x54de5729UL, 0x23d967bfUL, 0xb3667a2eUL, 0xc4614ab8UL, - 0x5d681b02UL, 0x2a6f2b94UL, 0xb40bbe37UL, 0xc30c8ea1UL, 0x5a05df1bUL, - 0x2d02ef8dUL -#ifdef BYFOUR - }, - { - 0x00000000UL, 0x191b3141UL, 0x32366282UL, 0x2b2d53c3UL, 0x646cc504UL, - 0x7d77f445UL, 0x565aa786UL, 0x4f4196c7UL, 0xc8d98a08UL, 0xd1c2bb49UL, - 0xfaefe88aUL, 0xe3f4d9cbUL, 0xacb54f0cUL, 0xb5ae7e4dUL, 0x9e832d8eUL, - 0x87981ccfUL, 0x4ac21251UL, 0x53d92310UL, 0x78f470d3UL, 0x61ef4192UL, - 0x2eaed755UL, 0x37b5e614UL, 0x1c98b5d7UL, 0x05838496UL, 0x821b9859UL, - 0x9b00a918UL, 0xb02dfadbUL, 0xa936cb9aUL, 0xe6775d5dUL, 0xff6c6c1cUL, - 0xd4413fdfUL, 0xcd5a0e9eUL, 0x958424a2UL, 0x8c9f15e3UL, 0xa7b24620UL, - 0xbea97761UL, 0xf1e8e1a6UL, 0xe8f3d0e7UL, 0xc3de8324UL, 0xdac5b265UL, - 0x5d5daeaaUL, 0x44469febUL, 0x6f6bcc28UL, 0x7670fd69UL, 0x39316baeUL, - 0x202a5aefUL, 0x0b07092cUL, 0x121c386dUL, 0xdf4636f3UL, 0xc65d07b2UL, - 0xed705471UL, 0xf46b6530UL, 0xbb2af3f7UL, 0xa231c2b6UL, 0x891c9175UL, - 0x9007a034UL, 0x179fbcfbUL, 0x0e848dbaUL, 0x25a9de79UL, 0x3cb2ef38UL, - 0x73f379ffUL, 0x6ae848beUL, 0x41c51b7dUL, 0x58de2a3cUL, 0xf0794f05UL, - 0xe9627e44UL, 0xc24f2d87UL, 0xdb541cc6UL, 0x94158a01UL, 0x8d0ebb40UL, - 0xa623e883UL, 0xbf38d9c2UL, 0x38a0c50dUL, 0x21bbf44cUL, 0x0a96a78fUL, - 0x138d96ceUL, 0x5ccc0009UL, 0x45d73148UL, 0x6efa628bUL, 0x77e153caUL, - 0xbabb5d54UL, 0xa3a06c15UL, 0x888d3fd6UL, 0x91960e97UL, 0xded79850UL, - 0xc7cca911UL, 0xece1fad2UL, 0xf5facb93UL, 0x7262d75cUL, 0x6b79e61dUL, - 0x4054b5deUL, 0x594f849fUL, 0x160e1258UL, 0x0f152319UL, 0x243870daUL, - 0x3d23419bUL, 0x65fd6ba7UL, 0x7ce65ae6UL, 0x57cb0925UL, 0x4ed03864UL, - 0x0191aea3UL, 0x188a9fe2UL, 0x33a7cc21UL, 0x2abcfd60UL, 0xad24e1afUL, - 0xb43fd0eeUL, 0x9f12832dUL, 0x8609b26cUL, 0xc94824abUL, 0xd05315eaUL, - 0xfb7e4629UL, 0xe2657768UL, 0x2f3f79f6UL, 0x362448b7UL, 0x1d091b74UL, - 0x04122a35UL, 0x4b53bcf2UL, 0x52488db3UL, 0x7965de70UL, 0x607eef31UL, - 0xe7e6f3feUL, 0xfefdc2bfUL, 0xd5d0917cUL, 0xcccba03dUL, 0x838a36faUL, - 0x9a9107bbUL, 0xb1bc5478UL, 0xa8a76539UL, 0x3b83984bUL, 0x2298a90aUL, - 0x09b5fac9UL, 0x10aecb88UL, 0x5fef5d4fUL, 0x46f46c0eUL, 0x6dd93fcdUL, - 0x74c20e8cUL, 0xf35a1243UL, 0xea412302UL, 0xc16c70c1UL, 0xd8774180UL, - 0x9736d747UL, 0x8e2de606UL, 0xa500b5c5UL, 0xbc1b8484UL, 0x71418a1aUL, - 0x685abb5bUL, 0x4377e898UL, 0x5a6cd9d9UL, 0x152d4f1eUL, 0x0c367e5fUL, - 0x271b2d9cUL, 0x3e001cddUL, 0xb9980012UL, 0xa0833153UL, 0x8bae6290UL, - 0x92b553d1UL, 0xddf4c516UL, 0xc4eff457UL, 0xefc2a794UL, 0xf6d996d5UL, - 0xae07bce9UL, 0xb71c8da8UL, 0x9c31de6bUL, 0x852aef2aUL, 0xca6b79edUL, - 0xd37048acUL, 0xf85d1b6fUL, 0xe1462a2eUL, 0x66de36e1UL, 0x7fc507a0UL, - 0x54e85463UL, 0x4df36522UL, 0x02b2f3e5UL, 0x1ba9c2a4UL, 0x30849167UL, - 0x299fa026UL, 0xe4c5aeb8UL, 0xfdde9ff9UL, 0xd6f3cc3aUL, 0xcfe8fd7bUL, - 0x80a96bbcUL, 0x99b25afdUL, 0xb29f093eUL, 0xab84387fUL, 0x2c1c24b0UL, - 0x350715f1UL, 0x1e2a4632UL, 0x07317773UL, 0x4870e1b4UL, 0x516bd0f5UL, - 0x7a468336UL, 0x635db277UL, 0xcbfad74eUL, 0xd2e1e60fUL, 0xf9ccb5ccUL, - 0xe0d7848dUL, 0xaf96124aUL, 0xb68d230bUL, 0x9da070c8UL, 0x84bb4189UL, - 0x03235d46UL, 0x1a386c07UL, 0x31153fc4UL, 0x280e0e85UL, 0x674f9842UL, - 0x7e54a903UL, 0x5579fac0UL, 0x4c62cb81UL, 0x8138c51fUL, 0x9823f45eUL, - 0xb30ea79dUL, 0xaa1596dcUL, 0xe554001bUL, 0xfc4f315aUL, 0xd7626299UL, - 0xce7953d8UL, 0x49e14f17UL, 0x50fa7e56UL, 0x7bd72d95UL, 0x62cc1cd4UL, - 0x2d8d8a13UL, 0x3496bb52UL, 0x1fbbe891UL, 0x06a0d9d0UL, 0x5e7ef3ecUL, - 0x4765c2adUL, 0x6c48916eUL, 0x7553a02fUL, 0x3a1236e8UL, 0x230907a9UL, - 0x0824546aUL, 0x113f652bUL, 0x96a779e4UL, 0x8fbc48a5UL, 0xa4911b66UL, - 0xbd8a2a27UL, 0xf2cbbce0UL, 0xebd08da1UL, 0xc0fdde62UL, 0xd9e6ef23UL, - 0x14bce1bdUL, 0x0da7d0fcUL, 0x268a833fUL, 0x3f91b27eUL, 0x70d024b9UL, - 0x69cb15f8UL, 0x42e6463bUL, 0x5bfd777aUL, 0xdc656bb5UL, 0xc57e5af4UL, - 0xee530937UL, 0xf7483876UL, 0xb809aeb1UL, 0xa1129ff0UL, 0x8a3fcc33UL, - 0x9324fd72UL - }, - { - 0x00000000UL, 0x01c26a37UL, 0x0384d46eUL, 0x0246be59UL, 0x0709a8dcUL, - 0x06cbc2ebUL, 0x048d7cb2UL, 0x054f1685UL, 0x0e1351b8UL, 0x0fd13b8fUL, - 0x0d9785d6UL, 0x0c55efe1UL, 0x091af964UL, 0x08d89353UL, 0x0a9e2d0aUL, - 0x0b5c473dUL, 0x1c26a370UL, 0x1de4c947UL, 0x1fa2771eUL, 0x1e601d29UL, - 0x1b2f0bacUL, 0x1aed619bUL, 0x18abdfc2UL, 0x1969b5f5UL, 0x1235f2c8UL, - 0x13f798ffUL, 0x11b126a6UL, 0x10734c91UL, 0x153c5a14UL, 0x14fe3023UL, - 0x16b88e7aUL, 0x177ae44dUL, 0x384d46e0UL, 0x398f2cd7UL, 0x3bc9928eUL, - 0x3a0bf8b9UL, 0x3f44ee3cUL, 0x3e86840bUL, 0x3cc03a52UL, 0x3d025065UL, - 0x365e1758UL, 0x379c7d6fUL, 0x35dac336UL, 0x3418a901UL, 0x3157bf84UL, - 0x3095d5b3UL, 0x32d36beaUL, 0x331101ddUL, 0x246be590UL, 0x25a98fa7UL, - 0x27ef31feUL, 0x262d5bc9UL, 0x23624d4cUL, 0x22a0277bUL, 0x20e69922UL, - 0x2124f315UL, 0x2a78b428UL, 0x2bbade1fUL, 0x29fc6046UL, 0x283e0a71UL, - 0x2d711cf4UL, 0x2cb376c3UL, 0x2ef5c89aUL, 0x2f37a2adUL, 0x709a8dc0UL, - 0x7158e7f7UL, 0x731e59aeUL, 0x72dc3399UL, 0x7793251cUL, 0x76514f2bUL, - 0x7417f172UL, 0x75d59b45UL, 0x7e89dc78UL, 0x7f4bb64fUL, 0x7d0d0816UL, - 0x7ccf6221UL, 0x798074a4UL, 0x78421e93UL, 0x7a04a0caUL, 0x7bc6cafdUL, - 0x6cbc2eb0UL, 0x6d7e4487UL, 0x6f38fadeUL, 0x6efa90e9UL, 0x6bb5866cUL, - 0x6a77ec5bUL, 0x68315202UL, 0x69f33835UL, 0x62af7f08UL, 0x636d153fUL, - 0x612bab66UL, 0x60e9c151UL, 0x65a6d7d4UL, 0x6464bde3UL, 0x662203baUL, - 0x67e0698dUL, 0x48d7cb20UL, 0x4915a117UL, 0x4b531f4eUL, 0x4a917579UL, - 0x4fde63fcUL, 0x4e1c09cbUL, 0x4c5ab792UL, 0x4d98dda5UL, 0x46c49a98UL, - 0x4706f0afUL, 0x45404ef6UL, 0x448224c1UL, 0x41cd3244UL, 0x400f5873UL, - 0x4249e62aUL, 0x438b8c1dUL, 0x54f16850UL, 0x55330267UL, 0x5775bc3eUL, - 0x56b7d609UL, 0x53f8c08cUL, 0x523aaabbUL, 0x507c14e2UL, 0x51be7ed5UL, - 0x5ae239e8UL, 0x5b2053dfUL, 0x5966ed86UL, 0x58a487b1UL, 0x5deb9134UL, - 0x5c29fb03UL, 0x5e6f455aUL, 0x5fad2f6dUL, 0xe1351b80UL, 0xe0f771b7UL, - 0xe2b1cfeeUL, 0xe373a5d9UL, 0xe63cb35cUL, 0xe7fed96bUL, 0xe5b86732UL, - 0xe47a0d05UL, 0xef264a38UL, 0xeee4200fUL, 0xeca29e56UL, 0xed60f461UL, - 0xe82fe2e4UL, 0xe9ed88d3UL, 0xebab368aUL, 0xea695cbdUL, 0xfd13b8f0UL, - 0xfcd1d2c7UL, 0xfe976c9eUL, 0xff5506a9UL, 0xfa1a102cUL, 0xfbd87a1bUL, - 0xf99ec442UL, 0xf85cae75UL, 0xf300e948UL, 0xf2c2837fUL, 0xf0843d26UL, - 0xf1465711UL, 0xf4094194UL, 0xf5cb2ba3UL, 0xf78d95faUL, 0xf64fffcdUL, - 0xd9785d60UL, 0xd8ba3757UL, 0xdafc890eUL, 0xdb3ee339UL, 0xde71f5bcUL, - 0xdfb39f8bUL, 0xddf521d2UL, 0xdc374be5UL, 0xd76b0cd8UL, 0xd6a966efUL, - 0xd4efd8b6UL, 0xd52db281UL, 0xd062a404UL, 0xd1a0ce33UL, 0xd3e6706aUL, - 0xd2241a5dUL, 0xc55efe10UL, 0xc49c9427UL, 0xc6da2a7eUL, 0xc7184049UL, - 0xc25756ccUL, 0xc3953cfbUL, 0xc1d382a2UL, 0xc011e895UL, 0xcb4dafa8UL, - 0xca8fc59fUL, 0xc8c97bc6UL, 0xc90b11f1UL, 0xcc440774UL, 0xcd866d43UL, - 0xcfc0d31aUL, 0xce02b92dUL, 0x91af9640UL, 0x906dfc77UL, 0x922b422eUL, - 0x93e92819UL, 0x96a63e9cUL, 0x976454abUL, 0x9522eaf2UL, 0x94e080c5UL, - 0x9fbcc7f8UL, 0x9e7eadcfUL, 0x9c381396UL, 0x9dfa79a1UL, 0x98b56f24UL, - 0x99770513UL, 0x9b31bb4aUL, 0x9af3d17dUL, 0x8d893530UL, 0x8c4b5f07UL, - 0x8e0de15eUL, 0x8fcf8b69UL, 0x8a809decUL, 0x8b42f7dbUL, 0x89044982UL, - 0x88c623b5UL, 0x839a6488UL, 0x82580ebfUL, 0x801eb0e6UL, 0x81dcdad1UL, - 0x8493cc54UL, 0x8551a663UL, 0x8717183aUL, 0x86d5720dUL, 0xa9e2d0a0UL, - 0xa820ba97UL, 0xaa6604ceUL, 0xaba46ef9UL, 0xaeeb787cUL, 0xaf29124bUL, - 0xad6fac12UL, 0xacadc625UL, 0xa7f18118UL, 0xa633eb2fUL, 0xa4755576UL, - 0xa5b73f41UL, 0xa0f829c4UL, 0xa13a43f3UL, 0xa37cfdaaUL, 0xa2be979dUL, - 0xb5c473d0UL, 0xb40619e7UL, 0xb640a7beUL, 0xb782cd89UL, 0xb2cddb0cUL, - 0xb30fb13bUL, 0xb1490f62UL, 0xb08b6555UL, 0xbbd72268UL, 0xba15485fUL, - 0xb853f606UL, 0xb9919c31UL, 0xbcde8ab4UL, 0xbd1ce083UL, 0xbf5a5edaUL, - 0xbe9834edUL - }, - { - 0x00000000UL, 0xb8bc6765UL, 0xaa09c88bUL, 0x12b5afeeUL, 0x8f629757UL, - 0x37def032UL, 0x256b5fdcUL, 0x9dd738b9UL, 0xc5b428efUL, 0x7d084f8aUL, - 0x6fbde064UL, 0xd7018701UL, 0x4ad6bfb8UL, 0xf26ad8ddUL, 0xe0df7733UL, - 0x58631056UL, 0x5019579fUL, 0xe8a530faUL, 0xfa109f14UL, 0x42acf871UL, - 0xdf7bc0c8UL, 0x67c7a7adUL, 0x75720843UL, 0xcdce6f26UL, 0x95ad7f70UL, - 0x2d111815UL, 0x3fa4b7fbUL, 0x8718d09eUL, 0x1acfe827UL, 0xa2738f42UL, - 0xb0c620acUL, 0x087a47c9UL, 0xa032af3eUL, 0x188ec85bUL, 0x0a3b67b5UL, - 0xb28700d0UL, 0x2f503869UL, 0x97ec5f0cUL, 0x8559f0e2UL, 0x3de59787UL, - 0x658687d1UL, 0xdd3ae0b4UL, 0xcf8f4f5aUL, 0x7733283fUL, 0xeae41086UL, - 0x525877e3UL, 0x40edd80dUL, 0xf851bf68UL, 0xf02bf8a1UL, 0x48979fc4UL, - 0x5a22302aUL, 0xe29e574fUL, 0x7f496ff6UL, 0xc7f50893UL, 0xd540a77dUL, - 0x6dfcc018UL, 0x359fd04eUL, 0x8d23b72bUL, 0x9f9618c5UL, 0x272a7fa0UL, - 0xbafd4719UL, 0x0241207cUL, 0x10f48f92UL, 0xa848e8f7UL, 0x9b14583dUL, - 0x23a83f58UL, 0x311d90b6UL, 0x89a1f7d3UL, 0x1476cf6aUL, 0xaccaa80fUL, - 0xbe7f07e1UL, 0x06c36084UL, 0x5ea070d2UL, 0xe61c17b7UL, 0xf4a9b859UL, - 0x4c15df3cUL, 0xd1c2e785UL, 0x697e80e0UL, 0x7bcb2f0eUL, 0xc377486bUL, - 0xcb0d0fa2UL, 0x73b168c7UL, 0x6104c729UL, 0xd9b8a04cUL, 0x446f98f5UL, - 0xfcd3ff90UL, 0xee66507eUL, 0x56da371bUL, 0x0eb9274dUL, 0xb6054028UL, - 0xa4b0efc6UL, 0x1c0c88a3UL, 0x81dbb01aUL, 0x3967d77fUL, 0x2bd27891UL, - 0x936e1ff4UL, 0x3b26f703UL, 0x839a9066UL, 0x912f3f88UL, 0x299358edUL, - 0xb4446054UL, 0x0cf80731UL, 0x1e4da8dfUL, 0xa6f1cfbaUL, 0xfe92dfecUL, - 0x462eb889UL, 0x549b1767UL, 0xec277002UL, 0x71f048bbUL, 0xc94c2fdeUL, - 0xdbf98030UL, 0x6345e755UL, 0x6b3fa09cUL, 0xd383c7f9UL, 0xc1366817UL, - 0x798a0f72UL, 0xe45d37cbUL, 0x5ce150aeUL, 0x4e54ff40UL, 0xf6e89825UL, - 0xae8b8873UL, 0x1637ef16UL, 0x048240f8UL, 0xbc3e279dUL, 0x21e91f24UL, - 0x99557841UL, 0x8be0d7afUL, 0x335cb0caUL, 0xed59b63bUL, 0x55e5d15eUL, - 0x47507eb0UL, 0xffec19d5UL, 0x623b216cUL, 0xda874609UL, 0xc832e9e7UL, - 0x708e8e82UL, 0x28ed9ed4UL, 0x9051f9b1UL, 0x82e4565fUL, 0x3a58313aUL, - 0xa78f0983UL, 0x1f336ee6UL, 0x0d86c108UL, 0xb53aa66dUL, 0xbd40e1a4UL, - 0x05fc86c1UL, 0x1749292fUL, 0xaff54e4aUL, 0x322276f3UL, 0x8a9e1196UL, - 0x982bbe78UL, 0x2097d91dUL, 0x78f4c94bUL, 0xc048ae2eUL, 0xd2fd01c0UL, - 0x6a4166a5UL, 0xf7965e1cUL, 0x4f2a3979UL, 0x5d9f9697UL, 0xe523f1f2UL, - 0x4d6b1905UL, 0xf5d77e60UL, 0xe762d18eUL, 0x5fdeb6ebUL, 0xc2098e52UL, - 0x7ab5e937UL, 0x680046d9UL, 0xd0bc21bcUL, 0x88df31eaUL, 0x3063568fUL, - 0x22d6f961UL, 0x9a6a9e04UL, 0x07bda6bdUL, 0xbf01c1d8UL, 0xadb46e36UL, - 0x15080953UL, 0x1d724e9aUL, 0xa5ce29ffUL, 0xb77b8611UL, 0x0fc7e174UL, - 0x9210d9cdUL, 0x2aacbea8UL, 0x38191146UL, 0x80a57623UL, 0xd8c66675UL, - 0x607a0110UL, 0x72cfaefeUL, 0xca73c99bUL, 0x57a4f122UL, 0xef189647UL, - 0xfdad39a9UL, 0x45115eccUL, 0x764dee06UL, 0xcef18963UL, 0xdc44268dUL, - 0x64f841e8UL, 0xf92f7951UL, 0x41931e34UL, 0x5326b1daUL, 0xeb9ad6bfUL, - 0xb3f9c6e9UL, 0x0b45a18cUL, 0x19f00e62UL, 0xa14c6907UL, 0x3c9b51beUL, - 0x842736dbUL, 0x96929935UL, 0x2e2efe50UL, 0x2654b999UL, 0x9ee8defcUL, - 0x8c5d7112UL, 0x34e11677UL, 0xa9362eceUL, 0x118a49abUL, 0x033fe645UL, - 0xbb838120UL, 0xe3e09176UL, 0x5b5cf613UL, 0x49e959fdUL, 0xf1553e98UL, - 0x6c820621UL, 0xd43e6144UL, 0xc68bceaaUL, 0x7e37a9cfUL, 0xd67f4138UL, - 0x6ec3265dUL, 0x7c7689b3UL, 0xc4caeed6UL, 0x591dd66fUL, 0xe1a1b10aUL, - 0xf3141ee4UL, 0x4ba87981UL, 0x13cb69d7UL, 0xab770eb2UL, 0xb9c2a15cUL, - 0x017ec639UL, 0x9ca9fe80UL, 0x241599e5UL, 0x36a0360bUL, 0x8e1c516eUL, - 0x866616a7UL, 0x3eda71c2UL, 0x2c6fde2cUL, 0x94d3b949UL, 0x090481f0UL, - 0xb1b8e695UL, 0xa30d497bUL, 0x1bb12e1eUL, 0x43d23e48UL, 0xfb6e592dUL, - 0xe9dbf6c3UL, 0x516791a6UL, 0xccb0a91fUL, 0x740cce7aUL, 0x66b96194UL, - 0xde0506f1UL - }, - { - 0x00000000UL, 0x96300777UL, 0x2c610eeeUL, 0xba510999UL, 0x19c46d07UL, - 0x8ff46a70UL, 0x35a563e9UL, 0xa395649eUL, 0x3288db0eUL, 0xa4b8dc79UL, - 0x1ee9d5e0UL, 0x88d9d297UL, 0x2b4cb609UL, 0xbd7cb17eUL, 0x072db8e7UL, - 0x911dbf90UL, 0x6410b71dUL, 0xf220b06aUL, 0x4871b9f3UL, 0xde41be84UL, - 0x7dd4da1aUL, 0xebe4dd6dUL, 0x51b5d4f4UL, 0xc785d383UL, 0x56986c13UL, - 0xc0a86b64UL, 0x7af962fdUL, 0xecc9658aUL, 0x4f5c0114UL, 0xd96c0663UL, - 0x633d0ffaUL, 0xf50d088dUL, 0xc8206e3bUL, 0x5e10694cUL, 0xe44160d5UL, - 0x727167a2UL, 0xd1e4033cUL, 0x47d4044bUL, 0xfd850dd2UL, 0x6bb50aa5UL, - 0xfaa8b535UL, 0x6c98b242UL, 0xd6c9bbdbUL, 0x40f9bcacUL, 0xe36cd832UL, - 0x755cdf45UL, 0xcf0dd6dcUL, 0x593dd1abUL, 0xac30d926UL, 0x3a00de51UL, - 0x8051d7c8UL, 0x1661d0bfUL, 0xb5f4b421UL, 0x23c4b356UL, 0x9995bacfUL, - 0x0fa5bdb8UL, 0x9eb80228UL, 0x0888055fUL, 0xb2d90cc6UL, 0x24e90bb1UL, - 0x877c6f2fUL, 0x114c6858UL, 0xab1d61c1UL, 0x3d2d66b6UL, 0x9041dc76UL, - 0x0671db01UL, 0xbc20d298UL, 0x2a10d5efUL, 0x8985b171UL, 0x1fb5b606UL, - 0xa5e4bf9fUL, 0x33d4b8e8UL, 0xa2c90778UL, 0x34f9000fUL, 0x8ea80996UL, - 0x18980ee1UL, 0xbb0d6a7fUL, 0x2d3d6d08UL, 0x976c6491UL, 0x015c63e6UL, - 0xf4516b6bUL, 0x62616c1cUL, 0xd8306585UL, 0x4e0062f2UL, 0xed95066cUL, - 0x7ba5011bUL, 0xc1f40882UL, 0x57c40ff5UL, 0xc6d9b065UL, 0x50e9b712UL, - 0xeab8be8bUL, 0x7c88b9fcUL, 0xdf1ddd62UL, 0x492dda15UL, 0xf37cd38cUL, - 0x654cd4fbUL, 0x5861b24dUL, 0xce51b53aUL, 0x7400bca3UL, 0xe230bbd4UL, - 0x41a5df4aUL, 0xd795d83dUL, 0x6dc4d1a4UL, 0xfbf4d6d3UL, 0x6ae96943UL, - 0xfcd96e34UL, 0x468867adUL, 0xd0b860daUL, 0x732d0444UL, 0xe51d0333UL, - 0x5f4c0aaaUL, 0xc97c0dddUL, 0x3c710550UL, 0xaa410227UL, 0x10100bbeUL, - 0x86200cc9UL, 0x25b56857UL, 0xb3856f20UL, 0x09d466b9UL, 0x9fe461ceUL, - 0x0ef9de5eUL, 0x98c9d929UL, 0x2298d0b0UL, 0xb4a8d7c7UL, 0x173db359UL, - 0x810db42eUL, 0x3b5cbdb7UL, 0xad6cbac0UL, 0x2083b8edUL, 0xb6b3bf9aUL, - 0x0ce2b603UL, 0x9ad2b174UL, 0x3947d5eaUL, 0xaf77d29dUL, 0x1526db04UL, - 0x8316dc73UL, 0x120b63e3UL, 0x843b6494UL, 0x3e6a6d0dUL, 0xa85a6a7aUL, - 0x0bcf0ee4UL, 0x9dff0993UL, 0x27ae000aUL, 0xb19e077dUL, 0x44930ff0UL, - 0xd2a30887UL, 0x68f2011eUL, 0xfec20669UL, 0x5d5762f7UL, 0xcb676580UL, - 0x71366c19UL, 0xe7066b6eUL, 0x761bd4feUL, 0xe02bd389UL, 0x5a7ada10UL, - 0xcc4add67UL, 0x6fdfb9f9UL, 0xf9efbe8eUL, 0x43beb717UL, 0xd58eb060UL, - 0xe8a3d6d6UL, 0x7e93d1a1UL, 0xc4c2d838UL, 0x52f2df4fUL, 0xf167bbd1UL, - 0x6757bca6UL, 0xdd06b53fUL, 0x4b36b248UL, 0xda2b0dd8UL, 0x4c1b0aafUL, - 0xf64a0336UL, 0x607a0441UL, 0xc3ef60dfUL, 0x55df67a8UL, 0xef8e6e31UL, - 0x79be6946UL, 0x8cb361cbUL, 0x1a8366bcUL, 0xa0d26f25UL, 0x36e26852UL, - 0x95770cccUL, 0x03470bbbUL, 0xb9160222UL, 0x2f260555UL, 0xbe3bbac5UL, - 0x280bbdb2UL, 0x925ab42bUL, 0x046ab35cUL, 0xa7ffd7c2UL, 0x31cfd0b5UL, - 0x8b9ed92cUL, 0x1daede5bUL, 0xb0c2649bUL, 0x26f263ecUL, 0x9ca36a75UL, - 0x0a936d02UL, 0xa906099cUL, 0x3f360eebUL, 0x85670772UL, 0x13570005UL, - 0x824abf95UL, 0x147ab8e2UL, 0xae2bb17bUL, 0x381bb60cUL, 0x9b8ed292UL, - 0x0dbed5e5UL, 0xb7efdc7cUL, 0x21dfdb0bUL, 0xd4d2d386UL, 0x42e2d4f1UL, - 0xf8b3dd68UL, 0x6e83da1fUL, 0xcd16be81UL, 0x5b26b9f6UL, 0xe177b06fUL, - 0x7747b718UL, 0xe65a0888UL, 0x706a0fffUL, 0xca3b0666UL, 0x5c0b0111UL, - 0xff9e658fUL, 0x69ae62f8UL, 0xd3ff6b61UL, 0x45cf6c16UL, 0x78e20aa0UL, - 0xeed20dd7UL, 0x5483044eUL, 0xc2b30339UL, 0x612667a7UL, 0xf71660d0UL, - 0x4d476949UL, 0xdb776e3eUL, 0x4a6ad1aeUL, 0xdc5ad6d9UL, 0x660bdf40UL, - 0xf03bd837UL, 0x53aebca9UL, 0xc59ebbdeUL, 0x7fcfb247UL, 0xe9ffb530UL, - 0x1cf2bdbdUL, 0x8ac2bacaUL, 0x3093b353UL, 0xa6a3b424UL, 0x0536d0baUL, - 0x9306d7cdUL, 0x2957de54UL, 0xbf67d923UL, 0x2e7a66b3UL, 0xb84a61c4UL, - 0x021b685dUL, 0x942b6f2aUL, 0x37be0bb4UL, 0xa18e0cc3UL, 0x1bdf055aUL, - 0x8def022dUL - }, - { - 0x00000000UL, 0x41311b19UL, 0x82623632UL, 0xc3532d2bUL, 0x04c56c64UL, - 0x45f4777dUL, 0x86a75a56UL, 0xc796414fUL, 0x088ad9c8UL, 0x49bbc2d1UL, - 0x8ae8effaUL, 0xcbd9f4e3UL, 0x0c4fb5acUL, 0x4d7eaeb5UL, 0x8e2d839eUL, - 0xcf1c9887UL, 0x5112c24aUL, 0x1023d953UL, 0xd370f478UL, 0x9241ef61UL, - 0x55d7ae2eUL, 0x14e6b537UL, 0xd7b5981cUL, 0x96848305UL, 0x59981b82UL, - 0x18a9009bUL, 0xdbfa2db0UL, 0x9acb36a9UL, 0x5d5d77e6UL, 0x1c6c6cffUL, - 0xdf3f41d4UL, 0x9e0e5acdUL, 0xa2248495UL, 0xe3159f8cUL, 0x2046b2a7UL, - 0x6177a9beUL, 0xa6e1e8f1UL, 0xe7d0f3e8UL, 0x2483dec3UL, 0x65b2c5daUL, - 0xaaae5d5dUL, 0xeb9f4644UL, 0x28cc6b6fUL, 0x69fd7076UL, 0xae6b3139UL, - 0xef5a2a20UL, 0x2c09070bUL, 0x6d381c12UL, 0xf33646dfUL, 0xb2075dc6UL, - 0x715470edUL, 0x30656bf4UL, 0xf7f32abbUL, 0xb6c231a2UL, 0x75911c89UL, - 0x34a00790UL, 0xfbbc9f17UL, 0xba8d840eUL, 0x79dea925UL, 0x38efb23cUL, - 0xff79f373UL, 0xbe48e86aUL, 0x7d1bc541UL, 0x3c2ade58UL, 0x054f79f0UL, - 0x447e62e9UL, 0x872d4fc2UL, 0xc61c54dbUL, 0x018a1594UL, 0x40bb0e8dUL, - 0x83e823a6UL, 0xc2d938bfUL, 0x0dc5a038UL, 0x4cf4bb21UL, 0x8fa7960aUL, - 0xce968d13UL, 0x0900cc5cUL, 0x4831d745UL, 0x8b62fa6eUL, 0xca53e177UL, - 0x545dbbbaUL, 0x156ca0a3UL, 0xd63f8d88UL, 0x970e9691UL, 0x5098d7deUL, - 0x11a9ccc7UL, 0xd2fae1ecUL, 0x93cbfaf5UL, 0x5cd76272UL, 0x1de6796bUL, - 0xdeb55440UL, 0x9f844f59UL, 0x58120e16UL, 0x1923150fUL, 0xda703824UL, - 0x9b41233dUL, 0xa76bfd65UL, 0xe65ae67cUL, 0x2509cb57UL, 0x6438d04eUL, - 0xa3ae9101UL, 0xe29f8a18UL, 0x21cca733UL, 0x60fdbc2aUL, 0xafe124adUL, - 0xeed03fb4UL, 0x2d83129fUL, 0x6cb20986UL, 0xab2448c9UL, 0xea1553d0UL, - 0x29467efbUL, 0x687765e2UL, 0xf6793f2fUL, 0xb7482436UL, 0x741b091dUL, - 0x352a1204UL, 0xf2bc534bUL, 0xb38d4852UL, 0x70de6579UL, 0x31ef7e60UL, - 0xfef3e6e7UL, 0xbfc2fdfeUL, 0x7c91d0d5UL, 0x3da0cbccUL, 0xfa368a83UL, - 0xbb07919aUL, 0x7854bcb1UL, 0x3965a7a8UL, 0x4b98833bUL, 0x0aa99822UL, - 0xc9fab509UL, 0x88cbae10UL, 0x4f5def5fUL, 0x0e6cf446UL, 0xcd3fd96dUL, - 0x8c0ec274UL, 0x43125af3UL, 0x022341eaUL, 0xc1706cc1UL, 0x804177d8UL, - 0x47d73697UL, 0x06e62d8eUL, 0xc5b500a5UL, 0x84841bbcUL, 0x1a8a4171UL, - 0x5bbb5a68UL, 0x98e87743UL, 0xd9d96c5aUL, 0x1e4f2d15UL, 0x5f7e360cUL, - 0x9c2d1b27UL, 0xdd1c003eUL, 0x120098b9UL, 0x533183a0UL, 0x9062ae8bUL, - 0xd153b592UL, 0x16c5f4ddUL, 0x57f4efc4UL, 0x94a7c2efUL, 0xd596d9f6UL, - 0xe9bc07aeUL, 0xa88d1cb7UL, 0x6bde319cUL, 0x2aef2a85UL, 0xed796bcaUL, - 0xac4870d3UL, 0x6f1b5df8UL, 0x2e2a46e1UL, 0xe136de66UL, 0xa007c57fUL, - 0x6354e854UL, 0x2265f34dUL, 0xe5f3b202UL, 0xa4c2a91bUL, 0x67918430UL, - 0x26a09f29UL, 0xb8aec5e4UL, 0xf99fdefdUL, 0x3accf3d6UL, 0x7bfde8cfUL, - 0xbc6ba980UL, 0xfd5ab299UL, 0x3e099fb2UL, 0x7f3884abUL, 0xb0241c2cUL, - 0xf1150735UL, 0x32462a1eUL, 0x73773107UL, 0xb4e17048UL, 0xf5d06b51UL, - 0x3683467aUL, 0x77b25d63UL, 0x4ed7facbUL, 0x0fe6e1d2UL, 0xccb5ccf9UL, - 0x8d84d7e0UL, 0x4a1296afUL, 0x0b238db6UL, 0xc870a09dUL, 0x8941bb84UL, - 0x465d2303UL, 0x076c381aUL, 0xc43f1531UL, 0x850e0e28UL, 0x42984f67UL, - 0x03a9547eUL, 0xc0fa7955UL, 0x81cb624cUL, 0x1fc53881UL, 0x5ef42398UL, - 0x9da70eb3UL, 0xdc9615aaUL, 0x1b0054e5UL, 0x5a314ffcUL, 0x996262d7UL, - 0xd85379ceUL, 0x174fe149UL, 0x567efa50UL, 0x952dd77bUL, 0xd41ccc62UL, - 0x138a8d2dUL, 0x52bb9634UL, 0x91e8bb1fUL, 0xd0d9a006UL, 0xecf37e5eUL, - 0xadc26547UL, 0x6e91486cUL, 0x2fa05375UL, 0xe836123aUL, 0xa9070923UL, - 0x6a542408UL, 0x2b653f11UL, 0xe479a796UL, 0xa548bc8fUL, 0x661b91a4UL, - 0x272a8abdUL, 0xe0bccbf2UL, 0xa18dd0ebUL, 0x62defdc0UL, 0x23efe6d9UL, - 0xbde1bc14UL, 0xfcd0a70dUL, 0x3f838a26UL, 0x7eb2913fUL, 0xb924d070UL, - 0xf815cb69UL, 0x3b46e642UL, 0x7a77fd5bUL, 0xb56b65dcUL, 0xf45a7ec5UL, - 0x370953eeUL, 0x763848f7UL, 0xb1ae09b8UL, 0xf09f12a1UL, 0x33cc3f8aUL, - 0x72fd2493UL - }, - { - 0x00000000UL, 0x376ac201UL, 0x6ed48403UL, 0x59be4602UL, 0xdca80907UL, - 0xebc2cb06UL, 0xb27c8d04UL, 0x85164f05UL, 0xb851130eUL, 0x8f3bd10fUL, - 0xd685970dUL, 0xe1ef550cUL, 0x64f91a09UL, 0x5393d808UL, 0x0a2d9e0aUL, - 0x3d475c0bUL, 0x70a3261cUL, 0x47c9e41dUL, 0x1e77a21fUL, 0x291d601eUL, - 0xac0b2f1bUL, 0x9b61ed1aUL, 0xc2dfab18UL, 0xf5b56919UL, 0xc8f23512UL, - 0xff98f713UL, 0xa626b111UL, 0x914c7310UL, 0x145a3c15UL, 0x2330fe14UL, - 0x7a8eb816UL, 0x4de47a17UL, 0xe0464d38UL, 0xd72c8f39UL, 0x8e92c93bUL, - 0xb9f80b3aUL, 0x3cee443fUL, 0x0b84863eUL, 0x523ac03cUL, 0x6550023dUL, - 0x58175e36UL, 0x6f7d9c37UL, 0x36c3da35UL, 0x01a91834UL, 0x84bf5731UL, - 0xb3d59530UL, 0xea6bd332UL, 0xdd011133UL, 0x90e56b24UL, 0xa78fa925UL, - 0xfe31ef27UL, 0xc95b2d26UL, 0x4c4d6223UL, 0x7b27a022UL, 0x2299e620UL, - 0x15f32421UL, 0x28b4782aUL, 0x1fdeba2bUL, 0x4660fc29UL, 0x710a3e28UL, - 0xf41c712dUL, 0xc376b32cUL, 0x9ac8f52eUL, 0xada2372fUL, 0xc08d9a70UL, - 0xf7e75871UL, 0xae591e73UL, 0x9933dc72UL, 0x1c259377UL, 0x2b4f5176UL, - 0x72f11774UL, 0x459bd575UL, 0x78dc897eUL, 0x4fb64b7fUL, 0x16080d7dUL, - 0x2162cf7cUL, 0xa4748079UL, 0x931e4278UL, 0xcaa0047aUL, 0xfdcac67bUL, - 0xb02ebc6cUL, 0x87447e6dUL, 0xdefa386fUL, 0xe990fa6eUL, 0x6c86b56bUL, - 0x5bec776aUL, 0x02523168UL, 0x3538f369UL, 0x087faf62UL, 0x3f156d63UL, - 0x66ab2b61UL, 0x51c1e960UL, 0xd4d7a665UL, 0xe3bd6464UL, 0xba032266UL, - 0x8d69e067UL, 0x20cbd748UL, 0x17a11549UL, 0x4e1f534bUL, 0x7975914aUL, - 0xfc63de4fUL, 0xcb091c4eUL, 0x92b75a4cUL, 0xa5dd984dUL, 0x989ac446UL, - 0xaff00647UL, 0xf64e4045UL, 0xc1248244UL, 0x4432cd41UL, 0x73580f40UL, - 0x2ae64942UL, 0x1d8c8b43UL, 0x5068f154UL, 0x67023355UL, 0x3ebc7557UL, - 0x09d6b756UL, 0x8cc0f853UL, 0xbbaa3a52UL, 0xe2147c50UL, 0xd57ebe51UL, - 0xe839e25aUL, 0xdf53205bUL, 0x86ed6659UL, 0xb187a458UL, 0x3491eb5dUL, - 0x03fb295cUL, 0x5a456f5eUL, 0x6d2fad5fUL, 0x801b35e1UL, 0xb771f7e0UL, - 0xeecfb1e2UL, 0xd9a573e3UL, 0x5cb33ce6UL, 0x6bd9fee7UL, 0x3267b8e5UL, - 0x050d7ae4UL, 0x384a26efUL, 0x0f20e4eeUL, 0x569ea2ecUL, 0x61f460edUL, - 0xe4e22fe8UL, 0xd388ede9UL, 0x8a36abebUL, 0xbd5c69eaUL, 0xf0b813fdUL, - 0xc7d2d1fcUL, 0x9e6c97feUL, 0xa90655ffUL, 0x2c101afaUL, 0x1b7ad8fbUL, - 0x42c49ef9UL, 0x75ae5cf8UL, 0x48e900f3UL, 0x7f83c2f2UL, 0x263d84f0UL, - 0x115746f1UL, 0x944109f4UL, 0xa32bcbf5UL, 0xfa958df7UL, 0xcdff4ff6UL, - 0x605d78d9UL, 0x5737bad8UL, 0x0e89fcdaUL, 0x39e33edbUL, 0xbcf571deUL, - 0x8b9fb3dfUL, 0xd221f5ddUL, 0xe54b37dcUL, 0xd80c6bd7UL, 0xef66a9d6UL, - 0xb6d8efd4UL, 0x81b22dd5UL, 0x04a462d0UL, 0x33cea0d1UL, 0x6a70e6d3UL, - 0x5d1a24d2UL, 0x10fe5ec5UL, 0x27949cc4UL, 0x7e2adac6UL, 0x494018c7UL, - 0xcc5657c2UL, 0xfb3c95c3UL, 0xa282d3c1UL, 0x95e811c0UL, 0xa8af4dcbUL, - 0x9fc58fcaUL, 0xc67bc9c8UL, 0xf1110bc9UL, 0x740744ccUL, 0x436d86cdUL, - 0x1ad3c0cfUL, 0x2db902ceUL, 0x4096af91UL, 0x77fc6d90UL, 0x2e422b92UL, - 0x1928e993UL, 0x9c3ea696UL, 0xab546497UL, 0xf2ea2295UL, 0xc580e094UL, - 0xf8c7bc9fUL, 0xcfad7e9eUL, 0x9613389cUL, 0xa179fa9dUL, 0x246fb598UL, - 0x13057799UL, 0x4abb319bUL, 0x7dd1f39aUL, 0x3035898dUL, 0x075f4b8cUL, - 0x5ee10d8eUL, 0x698bcf8fUL, 0xec9d808aUL, 0xdbf7428bUL, 0x82490489UL, - 0xb523c688UL, 0x88649a83UL, 0xbf0e5882UL, 0xe6b01e80UL, 0xd1dadc81UL, - 0x54cc9384UL, 0x63a65185UL, 0x3a181787UL, 0x0d72d586UL, 0xa0d0e2a9UL, - 0x97ba20a8UL, 0xce0466aaUL, 0xf96ea4abUL, 0x7c78ebaeUL, 0x4b1229afUL, - 0x12ac6fadUL, 0x25c6adacUL, 0x1881f1a7UL, 0x2feb33a6UL, 0x765575a4UL, - 0x413fb7a5UL, 0xc429f8a0UL, 0xf3433aa1UL, 0xaafd7ca3UL, 0x9d97bea2UL, - 0xd073c4b5UL, 0xe71906b4UL, 0xbea740b6UL, 0x89cd82b7UL, 0x0cdbcdb2UL, - 0x3bb10fb3UL, 0x620f49b1UL, 0x55658bb0UL, 0x6822d7bbUL, 0x5f4815baUL, - 0x06f653b8UL, 0x319c91b9UL, 0xb48adebcUL, 0x83e01cbdUL, 0xda5e5abfUL, - 0xed3498beUL - }, - { - 0x00000000UL, 0x6567bcb8UL, 0x8bc809aaUL, 0xeeafb512UL, 0x5797628fUL, - 0x32f0de37UL, 0xdc5f6b25UL, 0xb938d79dUL, 0xef28b4c5UL, 0x8a4f087dUL, - 0x64e0bd6fUL, 0x018701d7UL, 0xb8bfd64aUL, 0xddd86af2UL, 0x3377dfe0UL, - 0x56106358UL, 0x9f571950UL, 0xfa30a5e8UL, 0x149f10faUL, 0x71f8ac42UL, - 0xc8c07bdfUL, 0xada7c767UL, 0x43087275UL, 0x266fcecdUL, 0x707fad95UL, - 0x1518112dUL, 0xfbb7a43fUL, 0x9ed01887UL, 0x27e8cf1aUL, 0x428f73a2UL, - 0xac20c6b0UL, 0xc9477a08UL, 0x3eaf32a0UL, 0x5bc88e18UL, 0xb5673b0aUL, - 0xd00087b2UL, 0x6938502fUL, 0x0c5fec97UL, 0xe2f05985UL, 0x8797e53dUL, - 0xd1878665UL, 0xb4e03addUL, 0x5a4f8fcfUL, 0x3f283377UL, 0x8610e4eaUL, - 0xe3775852UL, 0x0dd8ed40UL, 0x68bf51f8UL, 0xa1f82bf0UL, 0xc49f9748UL, - 0x2a30225aUL, 0x4f579ee2UL, 0xf66f497fUL, 0x9308f5c7UL, 0x7da740d5UL, - 0x18c0fc6dUL, 0x4ed09f35UL, 0x2bb7238dUL, 0xc518969fUL, 0xa07f2a27UL, - 0x1947fdbaUL, 0x7c204102UL, 0x928ff410UL, 0xf7e848a8UL, 0x3d58149bUL, - 0x583fa823UL, 0xb6901d31UL, 0xd3f7a189UL, 0x6acf7614UL, 0x0fa8caacUL, - 0xe1077fbeUL, 0x8460c306UL, 0xd270a05eUL, 0xb7171ce6UL, 0x59b8a9f4UL, - 0x3cdf154cUL, 0x85e7c2d1UL, 0xe0807e69UL, 0x0e2fcb7bUL, 0x6b4877c3UL, - 0xa20f0dcbUL, 0xc768b173UL, 0x29c70461UL, 0x4ca0b8d9UL, 0xf5986f44UL, - 0x90ffd3fcUL, 0x7e5066eeUL, 0x1b37da56UL, 0x4d27b90eUL, 0x284005b6UL, - 0xc6efb0a4UL, 0xa3880c1cUL, 0x1ab0db81UL, 0x7fd76739UL, 0x9178d22bUL, - 0xf41f6e93UL, 0x03f7263bUL, 0x66909a83UL, 0x883f2f91UL, 0xed589329UL, - 0x546044b4UL, 0x3107f80cUL, 0xdfa84d1eUL, 0xbacff1a6UL, 0xecdf92feUL, - 0x89b82e46UL, 0x67179b54UL, 0x027027ecUL, 0xbb48f071UL, 0xde2f4cc9UL, - 0x3080f9dbUL, 0x55e74563UL, 0x9ca03f6bUL, 0xf9c783d3UL, 0x176836c1UL, - 0x720f8a79UL, 0xcb375de4UL, 0xae50e15cUL, 0x40ff544eUL, 0x2598e8f6UL, - 0x73888baeUL, 0x16ef3716UL, 0xf8408204UL, 0x9d273ebcUL, 0x241fe921UL, - 0x41785599UL, 0xafd7e08bUL, 0xcab05c33UL, 0x3bb659edUL, 0x5ed1e555UL, - 0xb07e5047UL, 0xd519ecffUL, 0x6c213b62UL, 0x094687daUL, 0xe7e932c8UL, - 0x828e8e70UL, 0xd49eed28UL, 0xb1f95190UL, 0x5f56e482UL, 0x3a31583aUL, - 0x83098fa7UL, 0xe66e331fUL, 0x08c1860dUL, 0x6da63ab5UL, 0xa4e140bdUL, - 0xc186fc05UL, 0x2f294917UL, 0x4a4ef5afUL, 0xf3762232UL, 0x96119e8aUL, - 0x78be2b98UL, 0x1dd99720UL, 0x4bc9f478UL, 0x2eae48c0UL, 0xc001fdd2UL, - 0xa566416aUL, 0x1c5e96f7UL, 0x79392a4fUL, 0x97969f5dUL, 0xf2f123e5UL, - 0x05196b4dUL, 0x607ed7f5UL, 0x8ed162e7UL, 0xebb6de5fUL, 0x528e09c2UL, - 0x37e9b57aUL, 0xd9460068UL, 0xbc21bcd0UL, 0xea31df88UL, 0x8f566330UL, - 0x61f9d622UL, 0x049e6a9aUL, 0xbda6bd07UL, 0xd8c101bfUL, 0x366eb4adUL, - 0x53090815UL, 0x9a4e721dUL, 0xff29cea5UL, 0x11867bb7UL, 0x74e1c70fUL, - 0xcdd91092UL, 0xa8beac2aUL, 0x46111938UL, 0x2376a580UL, 0x7566c6d8UL, - 0x10017a60UL, 0xfeaecf72UL, 0x9bc973caUL, 0x22f1a457UL, 0x479618efUL, - 0xa939adfdUL, 0xcc5e1145UL, 0x06ee4d76UL, 0x6389f1ceUL, 0x8d2644dcUL, - 0xe841f864UL, 0x51792ff9UL, 0x341e9341UL, 0xdab12653UL, 0xbfd69aebUL, - 0xe9c6f9b3UL, 0x8ca1450bUL, 0x620ef019UL, 0x07694ca1UL, 0xbe519b3cUL, - 0xdb362784UL, 0x35999296UL, 0x50fe2e2eUL, 0x99b95426UL, 0xfcdee89eUL, - 0x12715d8cUL, 0x7716e134UL, 0xce2e36a9UL, 0xab498a11UL, 0x45e63f03UL, - 0x208183bbUL, 0x7691e0e3UL, 0x13f65c5bUL, 0xfd59e949UL, 0x983e55f1UL, - 0x2106826cUL, 0x44613ed4UL, 0xaace8bc6UL, 0xcfa9377eUL, 0x38417fd6UL, - 0x5d26c36eUL, 0xb389767cUL, 0xd6eecac4UL, 0x6fd61d59UL, 0x0ab1a1e1UL, - 0xe41e14f3UL, 0x8179a84bUL, 0xd769cb13UL, 0xb20e77abUL, 0x5ca1c2b9UL, - 0x39c67e01UL, 0x80fea99cUL, 0xe5991524UL, 0x0b36a036UL, 0x6e511c8eUL, - 0xa7166686UL, 0xc271da3eUL, 0x2cde6f2cUL, 0x49b9d394UL, 0xf0810409UL, - 0x95e6b8b1UL, 0x7b490da3UL, 0x1e2eb11bUL, 0x483ed243UL, 0x2d596efbUL, - 0xc3f6dbe9UL, 0xa6916751UL, 0x1fa9b0ccUL, 0x7ace0c74UL, 0x9461b966UL, - 0xf10605deUL -#endif - } -}; diff --git a/dep/zlib/src/deflate.c b/dep/zlib/src/deflate.c deleted file mode 100644 index 1ec761448..000000000 --- a/dep/zlib/src/deflate.c +++ /dev/null @@ -1,2163 +0,0 @@ -/* deflate.c -- compress data using the deflation algorithm - * Copyright (C) 1995-2017 Jean-loup Gailly and Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* - * ALGORITHM - * - * The "deflation" process depends on being able to identify portions - * of the input text which are identical to earlier input (within a - * sliding window trailing behind the input currently being processed). - * - * The most straightforward technique turns out to be the fastest for - * most input files: try all possible matches and select the longest. - * The key feature of this algorithm is that insertions into the string - * dictionary are very simple and thus fast, and deletions are avoided - * completely. Insertions are performed at each input character, whereas - * string matches are performed only when the previous match ends. So it - * is preferable to spend more time in matches to allow very fast string - * insertions and avoid deletions. The matching algorithm for small - * strings is inspired from that of Rabin & Karp. A brute force approach - * is used to find longer strings when a small match has been found. - * A similar algorithm is used in comic (by Jan-Mark Wams) and freeze - * (by Leonid Broukhis). - * A previous version of this file used a more sophisticated algorithm - * (by Fiala and Greene) which is guaranteed to run in linear amortized - * time, but has a larger average cost, uses more memory and is patented. - * However the F&G algorithm may be faster for some highly redundant - * files if the parameter max_chain_length (described below) is too large. - * - * ACKNOWLEDGEMENTS - * - * The idea of lazy evaluation of matches is due to Jan-Mark Wams, and - * I found it in 'freeze' written by Leonid Broukhis. - * Thanks to many people for bug reports and testing. - * - * REFERENCES - * - * Deutsch, L.P.,"DEFLATE Compressed Data Format Specification". - * Available in http://tools.ietf.org/html/rfc1951 - * - * A description of the Rabin and Karp algorithm is given in the book - * "Algorithms" by R. Sedgewick, Addison-Wesley, p252. - * - * Fiala,E.R., and Greene,D.H. - * Data Compression with Finite Windows, Comm.ACM, 32,4 (1989) 490-595 - * - */ - -/* @(#) $Id$ */ - -#include "deflate.h" - -const char deflate_copyright[] = - " deflate 1.2.11 Copyright 1995-2017 Jean-loup Gailly and Mark Adler "; -/* - If you use the zlib library in a product, an acknowledgment is welcome - in the documentation of your product. If for some reason you cannot - include such an acknowledgment, I would appreciate that you keep this - copyright string in the executable of your product. - */ - -/* =========================================================================== - * Function prototypes. - */ -typedef enum { - need_more, /* block not completed, need more input or more output */ - block_done, /* block flush performed */ - finish_started, /* finish started, need only more output at next deflate */ - finish_done /* finish done, accept no more input or output */ -} block_state; - -typedef block_state (*compress_func) OF((deflate_state *s, int flush)); -/* Compression function. Returns the block state after the call. */ - -local int deflateStateCheck OF((z_streamp strm)); -local void slide_hash OF((deflate_state *s)); -local void fill_window OF((deflate_state *s)); -local block_state deflate_stored OF((deflate_state *s, int flush)); -local block_state deflate_fast OF((deflate_state *s, int flush)); -#ifndef FASTEST -local block_state deflate_slow OF((deflate_state *s, int flush)); -#endif -local block_state deflate_rle OF((deflate_state *s, int flush)); -local block_state deflate_huff OF((deflate_state *s, int flush)); -local void lm_init OF((deflate_state *s)); -local void putShortMSB OF((deflate_state *s, uInt b)); -local void flush_pending OF((z_streamp strm)); -local unsigned read_buf OF((z_streamp strm, Bytef *buf, unsigned size)); -#ifdef ASMV -# pragma message("Assembler code may have bugs -- use at your own risk") - void match_init OF((void)); /* asm code initialization */ - uInt longest_match OF((deflate_state *s, IPos cur_match)); -#else -local uInt longest_match OF((deflate_state *s, IPos cur_match)); -#endif - -#ifdef ZLIB_DEBUG -local void check_match OF((deflate_state *s, IPos start, IPos match, - int length)); -#endif - -/* =========================================================================== - * Local data - */ - -#define NIL 0 -/* Tail of hash chains */ - -#ifndef TOO_FAR -# define TOO_FAR 4096 -#endif -/* Matches of length 3 are discarded if their distance exceeds TOO_FAR */ - -/* Values for max_lazy_match, good_match and max_chain_length, depending on - * the desired pack level (0..9). The values given below have been tuned to - * exclude worst case performance for pathological files. Better values may be - * found for specific files. - */ -typedef struct config_s { - ush good_length; /* reduce lazy search above this match length */ - ush max_lazy; /* do not perform lazy search above this match length */ - ush nice_length; /* quit search above this match length */ - ush max_chain; - compress_func func; -} config; - -#ifdef FASTEST -local const config configuration_table[2] = { -/* good lazy nice chain */ -/* 0 */ {0, 0, 0, 0, deflate_stored}, /* store only */ -/* 1 */ {4, 4, 8, 4, deflate_fast}}; /* max speed, no lazy matches */ -#else -local const config configuration_table[10] = { -/* good lazy nice chain */ -/* 0 */ {0, 0, 0, 0, deflate_stored}, /* store only */ -/* 1 */ {4, 4, 8, 4, deflate_fast}, /* max speed, no lazy matches */ -/* 2 */ {4, 5, 16, 8, deflate_fast}, -/* 3 */ {4, 6, 32, 32, deflate_fast}, - -/* 4 */ {4, 4, 16, 16, deflate_slow}, /* lazy matches */ -/* 5 */ {8, 16, 32, 32, deflate_slow}, -/* 6 */ {8, 16, 128, 128, deflate_slow}, -/* 7 */ {8, 32, 128, 256, deflate_slow}, -/* 8 */ {32, 128, 258, 1024, deflate_slow}, -/* 9 */ {32, 258, 258, 4096, deflate_slow}}; /* max compression */ -#endif - -/* Note: the deflate() code requires max_lazy >= MIN_MATCH and max_chain >= 4 - * For deflate_fast() (levels <= 3) good is ignored and lazy has a different - * meaning. - */ - -/* rank Z_BLOCK between Z_NO_FLUSH and Z_PARTIAL_FLUSH */ -#define RANK(f) (((f) * 2) - ((f) > 4 ? 9 : 0)) - -/* =========================================================================== - * Update a hash value with the given input byte - * IN assertion: all calls to UPDATE_HASH are made with consecutive input - * characters, so that a running hash key can be computed from the previous - * key instead of complete recalculation each time. - */ -#define UPDATE_HASH(s,h,c) (h = (((h)<hash_shift) ^ (c)) & s->hash_mask) - - -/* =========================================================================== - * Insert string str in the dictionary and set match_head to the previous head - * of the hash chain (the most recent string with same hash key). Return - * the previous length of the hash chain. - * If this file is compiled with -DFASTEST, the compression level is forced - * to 1, and no hash chains are maintained. - * IN assertion: all calls to INSERT_STRING are made with consecutive input - * characters and the first MIN_MATCH bytes of str are valid (except for - * the last MIN_MATCH-1 bytes of the input file). - */ -#ifdef FASTEST -#define INSERT_STRING(s, str, match_head) \ - (UPDATE_HASH(s, s->ins_h, s->window[(str) + (MIN_MATCH-1)]), \ - match_head = s->head[s->ins_h], \ - s->head[s->ins_h] = (Pos)(str)) -#else -#define INSERT_STRING(s, str, match_head) \ - (UPDATE_HASH(s, s->ins_h, s->window[(str) + (MIN_MATCH-1)]), \ - match_head = s->prev[(str) & s->w_mask] = s->head[s->ins_h], \ - s->head[s->ins_h] = (Pos)(str)) -#endif - -/* =========================================================================== - * Initialize the hash table (avoiding 64K overflow for 16 bit systems). - * prev[] will be initialized on the fly. - */ -#define CLEAR_HASH(s) \ - s->head[s->hash_size-1] = NIL; \ - zmemzero((Bytef *)s->head, (unsigned)(s->hash_size-1)*sizeof(*s->head)); - -/* =========================================================================== - * Slide the hash table when sliding the window down (could be avoided with 32 - * bit values at the expense of memory usage). We slide even when level == 0 to - * keep the hash table consistent if we switch back to level > 0 later. - */ -local void slide_hash(s) - deflate_state *s; -{ - unsigned n, m; - Posf *p; - uInt wsize = s->w_size; - - n = s->hash_size; - p = &s->head[n]; - do { - m = *--p; - *p = (Pos)(m >= wsize ? m - wsize : NIL); - } while (--n); - n = wsize; -#ifndef FASTEST - p = &s->prev[n]; - do { - m = *--p; - *p = (Pos)(m >= wsize ? m - wsize : NIL); - /* If n is not on any hash chain, prev[n] is garbage but - * its value will never be used. - */ - } while (--n); -#endif -} - -/* ========================================================================= */ -int ZEXPORT deflateInit_(strm, level, version, stream_size) - z_streamp strm; - int level; - const char *version; - int stream_size; -{ - return deflateInit2_(strm, level, Z_DEFLATED, MAX_WBITS, DEF_MEM_LEVEL, - Z_DEFAULT_STRATEGY, version, stream_size); - /* To do: ignore strm->next_in if we use it as window */ -} - -/* ========================================================================= */ -int ZEXPORT deflateInit2_(strm, level, method, windowBits, memLevel, strategy, - version, stream_size) - z_streamp strm; - int level; - int method; - int windowBits; - int memLevel; - int strategy; - const char *version; - int stream_size; -{ - deflate_state *s; - int wrap = 1; - static const char my_version[] = ZLIB_VERSION; - - ushf *overlay; - /* We overlay pending_buf and d_buf+l_buf. This works since the average - * output size for (length,distance) codes is <= 24 bits. - */ - - if (version == Z_NULL || version[0] != my_version[0] || - stream_size != sizeof(z_stream)) { - return Z_VERSION_ERROR; - } - if (strm == Z_NULL) return Z_STREAM_ERROR; - - strm->msg = Z_NULL; - if (strm->zalloc == (alloc_func)0) { -#ifdef Z_SOLO - return Z_STREAM_ERROR; -#else - strm->zalloc = zcalloc; - strm->opaque = (voidpf)0; -#endif - } - if (strm->zfree == (free_func)0) -#ifdef Z_SOLO - return Z_STREAM_ERROR; -#else - strm->zfree = zcfree; -#endif - -#ifdef FASTEST - if (level != 0) level = 1; -#else - if (level == Z_DEFAULT_COMPRESSION) level = 6; -#endif - - if (windowBits < 0) { /* suppress zlib wrapper */ - wrap = 0; - windowBits = -windowBits; - } -#ifdef GZIP - else if (windowBits > 15) { - wrap = 2; /* write gzip wrapper instead */ - windowBits -= 16; - } -#endif - if (memLevel < 1 || memLevel > MAX_MEM_LEVEL || method != Z_DEFLATED || - windowBits < 8 || windowBits > 15 || level < 0 || level > 9 || - strategy < 0 || strategy > Z_FIXED || (windowBits == 8 && wrap != 1)) { - return Z_STREAM_ERROR; - } - if (windowBits == 8) windowBits = 9; /* until 256-byte window bug fixed */ - s = (deflate_state *) ZALLOC(strm, 1, sizeof(deflate_state)); - if (s == Z_NULL) return Z_MEM_ERROR; - strm->state = (struct internal_state FAR *)s; - s->strm = strm; - s->status = INIT_STATE; /* to pass state test in deflateReset() */ - - s->wrap = wrap; - s->gzhead = Z_NULL; - s->w_bits = (uInt)windowBits; - s->w_size = 1 << s->w_bits; - s->w_mask = s->w_size - 1; - - s->hash_bits = (uInt)memLevel + 7; - s->hash_size = 1 << s->hash_bits; - s->hash_mask = s->hash_size - 1; - s->hash_shift = ((s->hash_bits+MIN_MATCH-1)/MIN_MATCH); - - s->window = (Bytef *) ZALLOC(strm, s->w_size, 2*sizeof(Byte)); - s->prev = (Posf *) ZALLOC(strm, s->w_size, sizeof(Pos)); - s->head = (Posf *) ZALLOC(strm, s->hash_size, sizeof(Pos)); - - s->high_water = 0; /* nothing written to s->window yet */ - - s->lit_bufsize = 1 << (memLevel + 6); /* 16K elements by default */ - - overlay = (ushf *) ZALLOC(strm, s->lit_bufsize, sizeof(ush)+2); - s->pending_buf = (uchf *) overlay; - s->pending_buf_size = (ulg)s->lit_bufsize * (sizeof(ush)+2L); - - if (s->window == Z_NULL || s->prev == Z_NULL || s->head == Z_NULL || - s->pending_buf == Z_NULL) { - s->status = FINISH_STATE; - strm->msg = ERR_MSG(Z_MEM_ERROR); - deflateEnd (strm); - return Z_MEM_ERROR; - } - s->d_buf = overlay + s->lit_bufsize/sizeof(ush); - s->l_buf = s->pending_buf + (1+sizeof(ush))*s->lit_bufsize; - - s->level = level; - s->strategy = strategy; - s->method = (Byte)method; - - return deflateReset(strm); -} - -/* ========================================================================= - * Check for a valid deflate stream state. Return 0 if ok, 1 if not. - */ -local int deflateStateCheck (strm) - z_streamp strm; -{ - deflate_state *s; - if (strm == Z_NULL || - strm->zalloc == (alloc_func)0 || strm->zfree == (free_func)0) - return 1; - s = strm->state; - if (s == Z_NULL || s->strm != strm || (s->status != INIT_STATE && -#ifdef GZIP - s->status != GZIP_STATE && -#endif - s->status != EXTRA_STATE && - s->status != NAME_STATE && - s->status != COMMENT_STATE && - s->status != HCRC_STATE && - s->status != BUSY_STATE && - s->status != FINISH_STATE)) - return 1; - return 0; -} - -/* ========================================================================= */ -int ZEXPORT deflateSetDictionary (strm, dictionary, dictLength) - z_streamp strm; - const Bytef *dictionary; - uInt dictLength; -{ - deflate_state *s; - uInt str, n; - int wrap; - unsigned avail; - z_const unsigned char *next; - - if (deflateStateCheck(strm) || dictionary == Z_NULL) - return Z_STREAM_ERROR; - s = strm->state; - wrap = s->wrap; - if (wrap == 2 || (wrap == 1 && s->status != INIT_STATE) || s->lookahead) - return Z_STREAM_ERROR; - - /* when using zlib wrappers, compute Adler-32 for provided dictionary */ - if (wrap == 1) - strm->adler = adler32(strm->adler, dictionary, dictLength); - s->wrap = 0; /* avoid computing Adler-32 in read_buf */ - - /* if dictionary would fill window, just replace the history */ - if (dictLength >= s->w_size) { - if (wrap == 0) { /* already empty otherwise */ - CLEAR_HASH(s); - s->strstart = 0; - s->block_start = 0L; - s->insert = 0; - } - dictionary += dictLength - s->w_size; /* use the tail */ - dictLength = s->w_size; - } - - /* insert dictionary into window and hash */ - avail = strm->avail_in; - next = strm->next_in; - strm->avail_in = dictLength; - strm->next_in = (z_const Bytef *)dictionary; - fill_window(s); - while (s->lookahead >= MIN_MATCH) { - str = s->strstart; - n = s->lookahead - (MIN_MATCH-1); - do { - UPDATE_HASH(s, s->ins_h, s->window[str + MIN_MATCH-1]); -#ifndef FASTEST - s->prev[str & s->w_mask] = s->head[s->ins_h]; -#endif - s->head[s->ins_h] = (Pos)str; - str++; - } while (--n); - s->strstart = str; - s->lookahead = MIN_MATCH-1; - fill_window(s); - } - s->strstart += s->lookahead; - s->block_start = (long)s->strstart; - s->insert = s->lookahead; - s->lookahead = 0; - s->match_length = s->prev_length = MIN_MATCH-1; - s->match_available = 0; - strm->next_in = next; - strm->avail_in = avail; - s->wrap = wrap; - return Z_OK; -} - -/* ========================================================================= */ -int ZEXPORT deflateGetDictionary (strm, dictionary, dictLength) - z_streamp strm; - Bytef *dictionary; - uInt *dictLength; -{ - deflate_state *s; - uInt len; - - if (deflateStateCheck(strm)) - return Z_STREAM_ERROR; - s = strm->state; - len = s->strstart + s->lookahead; - if (len > s->w_size) - len = s->w_size; - if (dictionary != Z_NULL && len) - zmemcpy(dictionary, s->window + s->strstart + s->lookahead - len, len); - if (dictLength != Z_NULL) - *dictLength = len; - return Z_OK; -} - -/* ========================================================================= */ -int ZEXPORT deflateResetKeep (strm) - z_streamp strm; -{ - deflate_state *s; - - if (deflateStateCheck(strm)) { - return Z_STREAM_ERROR; - } - - strm->total_in = strm->total_out = 0; - strm->msg = Z_NULL; /* use zfree if we ever allocate msg dynamically */ - strm->data_type = Z_UNKNOWN; - - s = (deflate_state *)strm->state; - s->pending = 0; - s->pending_out = s->pending_buf; - - if (s->wrap < 0) { - s->wrap = -s->wrap; /* was made negative by deflate(..., Z_FINISH); */ - } - s->status = -#ifdef GZIP - s->wrap == 2 ? GZIP_STATE : -#endif - s->wrap ? INIT_STATE : BUSY_STATE; - strm->adler = -#ifdef GZIP - s->wrap == 2 ? crc32(0L, Z_NULL, 0) : -#endif - adler32(0L, Z_NULL, 0); - s->last_flush = Z_NO_FLUSH; - - _tr_init(s); - - return Z_OK; -} - -/* ========================================================================= */ -int ZEXPORT deflateReset (strm) - z_streamp strm; -{ - int ret; - - ret = deflateResetKeep(strm); - if (ret == Z_OK) - lm_init(strm->state); - return ret; -} - -/* ========================================================================= */ -int ZEXPORT deflateSetHeader (strm, head) - z_streamp strm; - gz_headerp head; -{ - if (deflateStateCheck(strm) || strm->state->wrap != 2) - return Z_STREAM_ERROR; - strm->state->gzhead = head; - return Z_OK; -} - -/* ========================================================================= */ -int ZEXPORT deflatePending (strm, pending, bits) - unsigned *pending; - int *bits; - z_streamp strm; -{ - if (deflateStateCheck(strm)) return Z_STREAM_ERROR; - if (pending != Z_NULL) - *pending = strm->state->pending; - if (bits != Z_NULL) - *bits = strm->state->bi_valid; - return Z_OK; -} - -/* ========================================================================= */ -int ZEXPORT deflatePrime (strm, bits, value) - z_streamp strm; - int bits; - int value; -{ - deflate_state *s; - int put; - - if (deflateStateCheck(strm)) return Z_STREAM_ERROR; - s = strm->state; - if ((Bytef *)(s->d_buf) < s->pending_out + ((Buf_size + 7) >> 3)) - return Z_BUF_ERROR; - do { - put = Buf_size - s->bi_valid; - if (put > bits) - put = bits; - s->bi_buf |= (ush)((value & ((1 << put) - 1)) << s->bi_valid); - s->bi_valid += put; - _tr_flush_bits(s); - value >>= put; - bits -= put; - } while (bits); - return Z_OK; -} - -/* ========================================================================= */ -int ZEXPORT deflateParams(strm, level, strategy) - z_streamp strm; - int level; - int strategy; -{ - deflate_state *s; - compress_func func; - - if (deflateStateCheck(strm)) return Z_STREAM_ERROR; - s = strm->state; - -#ifdef FASTEST - if (level != 0) level = 1; -#else - if (level == Z_DEFAULT_COMPRESSION) level = 6; -#endif - if (level < 0 || level > 9 || strategy < 0 || strategy > Z_FIXED) { - return Z_STREAM_ERROR; - } - func = configuration_table[s->level].func; - - if ((strategy != s->strategy || func != configuration_table[level].func) && - s->high_water) { - /* Flush the last buffer: */ - int err = deflate(strm, Z_BLOCK); - if (err == Z_STREAM_ERROR) - return err; - if (strm->avail_out == 0) - return Z_BUF_ERROR; - } - if (s->level != level) { - if (s->level == 0 && s->matches != 0) { - if (s->matches == 1) - slide_hash(s); - else - CLEAR_HASH(s); - s->matches = 0; - } - s->level = level; - s->max_lazy_match = configuration_table[level].max_lazy; - s->good_match = configuration_table[level].good_length; - s->nice_match = configuration_table[level].nice_length; - s->max_chain_length = configuration_table[level].max_chain; - } - s->strategy = strategy; - return Z_OK; -} - -/* ========================================================================= */ -int ZEXPORT deflateTune(strm, good_length, max_lazy, nice_length, max_chain) - z_streamp strm; - int good_length; - int max_lazy; - int nice_length; - int max_chain; -{ - deflate_state *s; - - if (deflateStateCheck(strm)) return Z_STREAM_ERROR; - s = strm->state; - s->good_match = (uInt)good_length; - s->max_lazy_match = (uInt)max_lazy; - s->nice_match = nice_length; - s->max_chain_length = (uInt)max_chain; - return Z_OK; -} - -/* ========================================================================= - * For the default windowBits of 15 and memLevel of 8, this function returns - * a close to exact, as well as small, upper bound on the compressed size. - * They are coded as constants here for a reason--if the #define's are - * changed, then this function needs to be changed as well. The return - * value for 15 and 8 only works for those exact settings. - * - * For any setting other than those defaults for windowBits and memLevel, - * the value returned is a conservative worst case for the maximum expansion - * resulting from using fixed blocks instead of stored blocks, which deflate - * can emit on compressed data for some combinations of the parameters. - * - * This function could be more sophisticated to provide closer upper bounds for - * every combination of windowBits and memLevel. But even the conservative - * upper bound of about 14% expansion does not seem onerous for output buffer - * allocation. - */ -uLong ZEXPORT deflateBound(strm, sourceLen) - z_streamp strm; - uLong sourceLen; -{ - deflate_state *s; - uLong complen, wraplen; - - /* conservative upper bound for compressed data */ - complen = sourceLen + - ((sourceLen + 7) >> 3) + ((sourceLen + 63) >> 6) + 5; - - /* if can't get parameters, return conservative bound plus zlib wrapper */ - if (deflateStateCheck(strm)) - return complen + 6; - - /* compute wrapper length */ - s = strm->state; - switch (s->wrap) { - case 0: /* raw deflate */ - wraplen = 0; - break; - case 1: /* zlib wrapper */ - wraplen = 6 + (s->strstart ? 4 : 0); - break; -#ifdef GZIP - case 2: /* gzip wrapper */ - wraplen = 18; - if (s->gzhead != Z_NULL) { /* user-supplied gzip header */ - Bytef *str; - if (s->gzhead->extra != Z_NULL) - wraplen += 2 + s->gzhead->extra_len; - str = s->gzhead->name; - if (str != Z_NULL) - do { - wraplen++; - } while (*str++); - str = s->gzhead->comment; - if (str != Z_NULL) - do { - wraplen++; - } while (*str++); - if (s->gzhead->hcrc) - wraplen += 2; - } - break; -#endif - default: /* for compiler happiness */ - wraplen = 6; - } - - /* if not default parameters, return conservative bound */ - if (s->w_bits != 15 || s->hash_bits != 8 + 7) - return complen + wraplen; - - /* default settings: return tight bound for that case */ - return sourceLen + (sourceLen >> 12) + (sourceLen >> 14) + - (sourceLen >> 25) + 13 - 6 + wraplen; -} - -/* ========================================================================= - * Put a short in the pending buffer. The 16-bit value is put in MSB order. - * IN assertion: the stream state is correct and there is enough room in - * pending_buf. - */ -local void putShortMSB (s, b) - deflate_state *s; - uInt b; -{ - put_byte(s, (Byte)(b >> 8)); - put_byte(s, (Byte)(b & 0xff)); -} - -/* ========================================================================= - * Flush as much pending output as possible. All deflate() output, except for - * some deflate_stored() output, goes through this function so some - * applications may wish to modify it to avoid allocating a large - * strm->next_out buffer and copying into it. (See also read_buf()). - */ -local void flush_pending(strm) - z_streamp strm; -{ - unsigned len; - deflate_state *s = strm->state; - - _tr_flush_bits(s); - len = s->pending; - if (len > strm->avail_out) len = strm->avail_out; - if (len == 0) return; - - zmemcpy(strm->next_out, s->pending_out, len); - strm->next_out += len; - s->pending_out += len; - strm->total_out += len; - strm->avail_out -= len; - s->pending -= len; - if (s->pending == 0) { - s->pending_out = s->pending_buf; - } -} - -/* =========================================================================== - * Update the header CRC with the bytes s->pending_buf[beg..s->pending - 1]. - */ -#define HCRC_UPDATE(beg) \ - do { \ - if (s->gzhead->hcrc && s->pending > (beg)) \ - strm->adler = crc32(strm->adler, s->pending_buf + (beg), \ - s->pending - (beg)); \ - } while (0) - -/* ========================================================================= */ -int ZEXPORT deflate (strm, flush) - z_streamp strm; - int flush; -{ - int old_flush; /* value of flush param for previous deflate call */ - deflate_state *s; - - if (deflateStateCheck(strm) || flush > Z_BLOCK || flush < 0) { - return Z_STREAM_ERROR; - } - s = strm->state; - - if (strm->next_out == Z_NULL || - (strm->avail_in != 0 && strm->next_in == Z_NULL) || - (s->status == FINISH_STATE && flush != Z_FINISH)) { - ERR_RETURN(strm, Z_STREAM_ERROR); - } - if (strm->avail_out == 0) ERR_RETURN(strm, Z_BUF_ERROR); - - old_flush = s->last_flush; - s->last_flush = flush; - - /* Flush as much pending output as possible */ - if (s->pending != 0) { - flush_pending(strm); - if (strm->avail_out == 0) { - /* Since avail_out is 0, deflate will be called again with - * more output space, but possibly with both pending and - * avail_in equal to zero. There won't be anything to do, - * but this is not an error situation so make sure we - * return OK instead of BUF_ERROR at next call of deflate: - */ - s->last_flush = -1; - return Z_OK; - } - - /* Make sure there is something to do and avoid duplicate consecutive - * flushes. For repeated and useless calls with Z_FINISH, we keep - * returning Z_STREAM_END instead of Z_BUF_ERROR. - */ - } else if (strm->avail_in == 0 && RANK(flush) <= RANK(old_flush) && - flush != Z_FINISH) { - ERR_RETURN(strm, Z_BUF_ERROR); - } - - /* User must not provide more input after the first FINISH: */ - if (s->status == FINISH_STATE && strm->avail_in != 0) { - ERR_RETURN(strm, Z_BUF_ERROR); - } - - /* Write the header */ - if (s->status == INIT_STATE) { - /* zlib header */ - uInt header = (Z_DEFLATED + ((s->w_bits-8)<<4)) << 8; - uInt level_flags; - - if (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2) - level_flags = 0; - else if (s->level < 6) - level_flags = 1; - else if (s->level == 6) - level_flags = 2; - else - level_flags = 3; - header |= (level_flags << 6); - if (s->strstart != 0) header |= PRESET_DICT; - header += 31 - (header % 31); - - putShortMSB(s, header); - - /* Save the adler32 of the preset dictionary: */ - if (s->strstart != 0) { - putShortMSB(s, (uInt)(strm->adler >> 16)); - putShortMSB(s, (uInt)(strm->adler & 0xffff)); - } - strm->adler = adler32(0L, Z_NULL, 0); - s->status = BUSY_STATE; - - /* Compression must start with an empty pending buffer */ - flush_pending(strm); - if (s->pending != 0) { - s->last_flush = -1; - return Z_OK; - } - } -#ifdef GZIP - if (s->status == GZIP_STATE) { - /* gzip header */ - strm->adler = crc32(0L, Z_NULL, 0); - put_byte(s, 31); - put_byte(s, 139); - put_byte(s, 8); - if (s->gzhead == Z_NULL) { - put_byte(s, 0); - put_byte(s, 0); - put_byte(s, 0); - put_byte(s, 0); - put_byte(s, 0); - put_byte(s, s->level == 9 ? 2 : - (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2 ? - 4 : 0)); - put_byte(s, OS_CODE); - s->status = BUSY_STATE; - - /* Compression must start with an empty pending buffer */ - flush_pending(strm); - if (s->pending != 0) { - s->last_flush = -1; - return Z_OK; - } - } - else { - put_byte(s, (s->gzhead->text ? 1 : 0) + - (s->gzhead->hcrc ? 2 : 0) + - (s->gzhead->extra == Z_NULL ? 0 : 4) + - (s->gzhead->name == Z_NULL ? 0 : 8) + - (s->gzhead->comment == Z_NULL ? 0 : 16) - ); - put_byte(s, (Byte)(s->gzhead->time & 0xff)); - put_byte(s, (Byte)((s->gzhead->time >> 8) & 0xff)); - put_byte(s, (Byte)((s->gzhead->time >> 16) & 0xff)); - put_byte(s, (Byte)((s->gzhead->time >> 24) & 0xff)); - put_byte(s, s->level == 9 ? 2 : - (s->strategy >= Z_HUFFMAN_ONLY || s->level < 2 ? - 4 : 0)); - put_byte(s, s->gzhead->os & 0xff); - if (s->gzhead->extra != Z_NULL) { - put_byte(s, s->gzhead->extra_len & 0xff); - put_byte(s, (s->gzhead->extra_len >> 8) & 0xff); - } - if (s->gzhead->hcrc) - strm->adler = crc32(strm->adler, s->pending_buf, - s->pending); - s->gzindex = 0; - s->status = EXTRA_STATE; - } - } - if (s->status == EXTRA_STATE) { - if (s->gzhead->extra != Z_NULL) { - ulg beg = s->pending; /* start of bytes to update crc */ - uInt left = (s->gzhead->extra_len & 0xffff) - s->gzindex; - while (s->pending + left > s->pending_buf_size) { - uInt copy = s->pending_buf_size - s->pending; - zmemcpy(s->pending_buf + s->pending, - s->gzhead->extra + s->gzindex, copy); - s->pending = s->pending_buf_size; - HCRC_UPDATE(beg); - s->gzindex += copy; - flush_pending(strm); - if (s->pending != 0) { - s->last_flush = -1; - return Z_OK; - } - beg = 0; - left -= copy; - } - zmemcpy(s->pending_buf + s->pending, - s->gzhead->extra + s->gzindex, left); - s->pending += left; - HCRC_UPDATE(beg); - s->gzindex = 0; - } - s->status = NAME_STATE; - } - if (s->status == NAME_STATE) { - if (s->gzhead->name != Z_NULL) { - ulg beg = s->pending; /* start of bytes to update crc */ - int val; - do { - if (s->pending == s->pending_buf_size) { - HCRC_UPDATE(beg); - flush_pending(strm); - if (s->pending != 0) { - s->last_flush = -1; - return Z_OK; - } - beg = 0; - } - val = s->gzhead->name[s->gzindex++]; - put_byte(s, val); - } while (val != 0); - HCRC_UPDATE(beg); - s->gzindex = 0; - } - s->status = COMMENT_STATE; - } - if (s->status == COMMENT_STATE) { - if (s->gzhead->comment != Z_NULL) { - ulg beg = s->pending; /* start of bytes to update crc */ - int val; - do { - if (s->pending == s->pending_buf_size) { - HCRC_UPDATE(beg); - flush_pending(strm); - if (s->pending != 0) { - s->last_flush = -1; - return Z_OK; - } - beg = 0; - } - val = s->gzhead->comment[s->gzindex++]; - put_byte(s, val); - } while (val != 0); - HCRC_UPDATE(beg); - } - s->status = HCRC_STATE; - } - if (s->status == HCRC_STATE) { - if (s->gzhead->hcrc) { - if (s->pending + 2 > s->pending_buf_size) { - flush_pending(strm); - if (s->pending != 0) { - s->last_flush = -1; - return Z_OK; - } - } - put_byte(s, (Byte)(strm->adler & 0xff)); - put_byte(s, (Byte)((strm->adler >> 8) & 0xff)); - strm->adler = crc32(0L, Z_NULL, 0); - } - s->status = BUSY_STATE; - - /* Compression must start with an empty pending buffer */ - flush_pending(strm); - if (s->pending != 0) { - s->last_flush = -1; - return Z_OK; - } - } -#endif - - /* Start a new block or continue the current one. - */ - if (strm->avail_in != 0 || s->lookahead != 0 || - (flush != Z_NO_FLUSH && s->status != FINISH_STATE)) { - block_state bstate; - - bstate = s->level == 0 ? deflate_stored(s, flush) : - s->strategy == Z_HUFFMAN_ONLY ? deflate_huff(s, flush) : - s->strategy == Z_RLE ? deflate_rle(s, flush) : - (*(configuration_table[s->level].func))(s, flush); - - if (bstate == finish_started || bstate == finish_done) { - s->status = FINISH_STATE; - } - if (bstate == need_more || bstate == finish_started) { - if (strm->avail_out == 0) { - s->last_flush = -1; /* avoid BUF_ERROR next call, see above */ - } - return Z_OK; - /* If flush != Z_NO_FLUSH && avail_out == 0, the next call - * of deflate should use the same flush parameter to make sure - * that the flush is complete. So we don't have to output an - * empty block here, this will be done at next call. This also - * ensures that for a very small output buffer, we emit at most - * one empty block. - */ - } - if (bstate == block_done) { - if (flush == Z_PARTIAL_FLUSH) { - _tr_align(s); - } else if (flush != Z_BLOCK) { /* FULL_FLUSH or SYNC_FLUSH */ - _tr_stored_block(s, (char*)0, 0L, 0); - /* For a full flush, this empty block will be recognized - * as a special marker by inflate_sync(). - */ - if (flush == Z_FULL_FLUSH) { - CLEAR_HASH(s); /* forget history */ - if (s->lookahead == 0) { - s->strstart = 0; - s->block_start = 0L; - s->insert = 0; - } - } - } - flush_pending(strm); - if (strm->avail_out == 0) { - s->last_flush = -1; /* avoid BUF_ERROR at next call, see above */ - return Z_OK; - } - } - } - - if (flush != Z_FINISH) return Z_OK; - if (s->wrap <= 0) return Z_STREAM_END; - - /* Write the trailer */ -#ifdef GZIP - if (s->wrap == 2) { - put_byte(s, (Byte)(strm->adler & 0xff)); - put_byte(s, (Byte)((strm->adler >> 8) & 0xff)); - put_byte(s, (Byte)((strm->adler >> 16) & 0xff)); - put_byte(s, (Byte)((strm->adler >> 24) & 0xff)); - put_byte(s, (Byte)(strm->total_in & 0xff)); - put_byte(s, (Byte)((strm->total_in >> 8) & 0xff)); - put_byte(s, (Byte)((strm->total_in >> 16) & 0xff)); - put_byte(s, (Byte)((strm->total_in >> 24) & 0xff)); - } - else -#endif - { - putShortMSB(s, (uInt)(strm->adler >> 16)); - putShortMSB(s, (uInt)(strm->adler & 0xffff)); - } - flush_pending(strm); - /* If avail_out is zero, the application will call deflate again - * to flush the rest. - */ - if (s->wrap > 0) s->wrap = -s->wrap; /* write the trailer only once! */ - return s->pending != 0 ? Z_OK : Z_STREAM_END; -} - -/* ========================================================================= */ -int ZEXPORT deflateEnd (strm) - z_streamp strm; -{ - int status; - - if (deflateStateCheck(strm)) return Z_STREAM_ERROR; - - status = strm->state->status; - - /* Deallocate in reverse order of allocations: */ - TRY_FREE(strm, strm->state->pending_buf); - TRY_FREE(strm, strm->state->head); - TRY_FREE(strm, strm->state->prev); - TRY_FREE(strm, strm->state->window); - - ZFREE(strm, strm->state); - strm->state = Z_NULL; - - return status == BUSY_STATE ? Z_DATA_ERROR : Z_OK; -} - -/* ========================================================================= - * Copy the source state to the destination state. - * To simplify the source, this is not supported for 16-bit MSDOS (which - * doesn't have enough memory anyway to duplicate compression states). - */ -int ZEXPORT deflateCopy (dest, source) - z_streamp dest; - z_streamp source; -{ -#ifdef MAXSEG_64K - return Z_STREAM_ERROR; -#else - deflate_state *ds; - deflate_state *ss; - ushf *overlay; - - - if (deflateStateCheck(source) || dest == Z_NULL) { - return Z_STREAM_ERROR; - } - - ss = source->state; - - zmemcpy((voidpf)dest, (voidpf)source, sizeof(z_stream)); - - ds = (deflate_state *) ZALLOC(dest, 1, sizeof(deflate_state)); - if (ds == Z_NULL) return Z_MEM_ERROR; - dest->state = (struct internal_state FAR *) ds; - zmemcpy((voidpf)ds, (voidpf)ss, sizeof(deflate_state)); - ds->strm = dest; - - ds->window = (Bytef *) ZALLOC(dest, ds->w_size, 2*sizeof(Byte)); - ds->prev = (Posf *) ZALLOC(dest, ds->w_size, sizeof(Pos)); - ds->head = (Posf *) ZALLOC(dest, ds->hash_size, sizeof(Pos)); - overlay = (ushf *) ZALLOC(dest, ds->lit_bufsize, sizeof(ush)+2); - ds->pending_buf = (uchf *) overlay; - - if (ds->window == Z_NULL || ds->prev == Z_NULL || ds->head == Z_NULL || - ds->pending_buf == Z_NULL) { - deflateEnd (dest); - return Z_MEM_ERROR; - } - /* following zmemcpy do not work for 16-bit MSDOS */ - zmemcpy(ds->window, ss->window, ds->w_size * 2 * sizeof(Byte)); - zmemcpy((voidpf)ds->prev, (voidpf)ss->prev, ds->w_size * sizeof(Pos)); - zmemcpy((voidpf)ds->head, (voidpf)ss->head, ds->hash_size * sizeof(Pos)); - zmemcpy(ds->pending_buf, ss->pending_buf, (uInt)ds->pending_buf_size); - - ds->pending_out = ds->pending_buf + (ss->pending_out - ss->pending_buf); - ds->d_buf = overlay + ds->lit_bufsize/sizeof(ush); - ds->l_buf = ds->pending_buf + (1+sizeof(ush))*ds->lit_bufsize; - - ds->l_desc.dyn_tree = ds->dyn_ltree; - ds->d_desc.dyn_tree = ds->dyn_dtree; - ds->bl_desc.dyn_tree = ds->bl_tree; - - return Z_OK; -#endif /* MAXSEG_64K */ -} - -/* =========================================================================== - * Read a new buffer from the current input stream, update the adler32 - * and total number of bytes read. All deflate() input goes through - * this function so some applications may wish to modify it to avoid - * allocating a large strm->next_in buffer and copying from it. - * (See also flush_pending()). - */ -local unsigned read_buf(strm, buf, size) - z_streamp strm; - Bytef *buf; - unsigned size; -{ - unsigned len = strm->avail_in; - - if (len > size) len = size; - if (len == 0) return 0; - - strm->avail_in -= len; - - zmemcpy(buf, strm->next_in, len); - if (strm->state->wrap == 1) { - strm->adler = adler32(strm->adler, buf, len); - } -#ifdef GZIP - else if (strm->state->wrap == 2) { - strm->adler = crc32(strm->adler, buf, len); - } -#endif - strm->next_in += len; - strm->total_in += len; - - return len; -} - -/* =========================================================================== - * Initialize the "longest match" routines for a new zlib stream - */ -local void lm_init (s) - deflate_state *s; -{ - s->window_size = (ulg)2L*s->w_size; - - CLEAR_HASH(s); - - /* Set the default configuration parameters: - */ - s->max_lazy_match = configuration_table[s->level].max_lazy; - s->good_match = configuration_table[s->level].good_length; - s->nice_match = configuration_table[s->level].nice_length; - s->max_chain_length = configuration_table[s->level].max_chain; - - s->strstart = 0; - s->block_start = 0L; - s->lookahead = 0; - s->insert = 0; - s->match_length = s->prev_length = MIN_MATCH-1; - s->match_available = 0; - s->ins_h = 0; -#ifndef FASTEST -#ifdef ASMV - match_init(); /* initialize the asm code */ -#endif -#endif -} - -#ifndef FASTEST -/* =========================================================================== - * Set match_start to the longest match starting at the given string and - * return its length. Matches shorter or equal to prev_length are discarded, - * in which case the result is equal to prev_length and match_start is - * garbage. - * IN assertions: cur_match is the head of the hash chain for the current - * string (strstart) and its distance is <= MAX_DIST, and prev_length >= 1 - * OUT assertion: the match length is not greater than s->lookahead. - */ -#ifndef ASMV -/* For 80x86 and 680x0, an optimized version will be provided in match.asm or - * match.S. The code will be functionally equivalent. - */ -local uInt longest_match(s, cur_match) - deflate_state *s; - IPos cur_match; /* current match */ -{ - unsigned chain_length = s->max_chain_length;/* max hash chain length */ - register Bytef *scan = s->window + s->strstart; /* current string */ - register Bytef *match; /* matched string */ - register int len; /* length of current match */ - int best_len = (int)s->prev_length; /* best match length so far */ - int nice_match = s->nice_match; /* stop if match long enough */ - IPos limit = s->strstart > (IPos)MAX_DIST(s) ? - s->strstart - (IPos)MAX_DIST(s) : NIL; - /* Stop when cur_match becomes <= limit. To simplify the code, - * we prevent matches with the string of window index 0. - */ - Posf *prev = s->prev; - uInt wmask = s->w_mask; - -#ifdef UNALIGNED_OK - /* Compare two bytes at a time. Note: this is not always beneficial. - * Try with and without -DUNALIGNED_OK to check. - */ - register Bytef *strend = s->window + s->strstart + MAX_MATCH - 1; - register ush scan_start = *(ushf*)scan; - register ush scan_end = *(ushf*)(scan+best_len-1); -#else - register Bytef *strend = s->window + s->strstart + MAX_MATCH; - register Byte scan_end1 = scan[best_len-1]; - register Byte scan_end = scan[best_len]; -#endif - - /* The code is optimized for HASH_BITS >= 8 and MAX_MATCH-2 multiple of 16. - * It is easy to get rid of this optimization if necessary. - */ - Assert(s->hash_bits >= 8 && MAX_MATCH == 258, "Code too clever"); - - /* Do not waste too much time if we already have a good match: */ - if (s->prev_length >= s->good_match) { - chain_length >>= 2; - } - /* Do not look for matches beyond the end of the input. This is necessary - * to make deflate deterministic. - */ - if ((uInt)nice_match > s->lookahead) nice_match = (int)s->lookahead; - - Assert((ulg)s->strstart <= s->window_size-MIN_LOOKAHEAD, "need lookahead"); - - do { - Assert(cur_match < s->strstart, "no future"); - match = s->window + cur_match; - - /* Skip to next match if the match length cannot increase - * or if the match length is less than 2. Note that the checks below - * for insufficient lookahead only occur occasionally for performance - * reasons. Therefore uninitialized memory will be accessed, and - * conditional jumps will be made that depend on those values. - * However the length of the match is limited to the lookahead, so - * the output of deflate is not affected by the uninitialized values. - */ -#if (defined(UNALIGNED_OK) && MAX_MATCH == 258) - /* This code assumes sizeof(unsigned short) == 2. Do not use - * UNALIGNED_OK if your compiler uses a different size. - */ - if (*(ushf*)(match+best_len-1) != scan_end || - *(ushf*)match != scan_start) continue; - - /* It is not necessary to compare scan[2] and match[2] since they are - * always equal when the other bytes match, given that the hash keys - * are equal and that HASH_BITS >= 8. Compare 2 bytes at a time at - * strstart+3, +5, ... up to strstart+257. We check for insufficient - * lookahead only every 4th comparison; the 128th check will be made - * at strstart+257. If MAX_MATCH-2 is not a multiple of 8, it is - * necessary to put more guard bytes at the end of the window, or - * to check more often for insufficient lookahead. - */ - Assert(scan[2] == match[2], "scan[2]?"); - scan++, match++; - do { - } while (*(ushf*)(scan+=2) == *(ushf*)(match+=2) && - *(ushf*)(scan+=2) == *(ushf*)(match+=2) && - *(ushf*)(scan+=2) == *(ushf*)(match+=2) && - *(ushf*)(scan+=2) == *(ushf*)(match+=2) && - scan < strend); - /* The funny "do {}" generates better code on most compilers */ - - /* Here, scan <= window+strstart+257 */ - Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan"); - if (*scan == *match) scan++; - - len = (MAX_MATCH - 1) - (int)(strend-scan); - scan = strend - (MAX_MATCH-1); - -#else /* UNALIGNED_OK */ - - if (match[best_len] != scan_end || - match[best_len-1] != scan_end1 || - *match != *scan || - *++match != scan[1]) continue; - - /* The check at best_len-1 can be removed because it will be made - * again later. (This heuristic is not always a win.) - * It is not necessary to compare scan[2] and match[2] since they - * are always equal when the other bytes match, given that - * the hash keys are equal and that HASH_BITS >= 8. - */ - scan += 2, match++; - Assert(*scan == *match, "match[2]?"); - - /* We check for insufficient lookahead only every 8th comparison; - * the 256th check will be made at strstart+258. - */ - do { - } while (*++scan == *++match && *++scan == *++match && - *++scan == *++match && *++scan == *++match && - *++scan == *++match && *++scan == *++match && - *++scan == *++match && *++scan == *++match && - scan < strend); - - Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan"); - - len = MAX_MATCH - (int)(strend - scan); - scan = strend - MAX_MATCH; - -#endif /* UNALIGNED_OK */ - - if (len > best_len) { - s->match_start = cur_match; - best_len = len; - if (len >= nice_match) break; -#ifdef UNALIGNED_OK - scan_end = *(ushf*)(scan+best_len-1); -#else - scan_end1 = scan[best_len-1]; - scan_end = scan[best_len]; -#endif - } - } while ((cur_match = prev[cur_match & wmask]) > limit - && --chain_length != 0); - - if ((uInt)best_len <= s->lookahead) return (uInt)best_len; - return s->lookahead; -} -#endif /* ASMV */ - -#else /* FASTEST */ - -/* --------------------------------------------------------------------------- - * Optimized version for FASTEST only - */ -local uInt longest_match(s, cur_match) - deflate_state *s; - IPos cur_match; /* current match */ -{ - register Bytef *scan = s->window + s->strstart; /* current string */ - register Bytef *match; /* matched string */ - register int len; /* length of current match */ - register Bytef *strend = s->window + s->strstart + MAX_MATCH; - - /* The code is optimized for HASH_BITS >= 8 and MAX_MATCH-2 multiple of 16. - * It is easy to get rid of this optimization if necessary. - */ - Assert(s->hash_bits >= 8 && MAX_MATCH == 258, "Code too clever"); - - Assert((ulg)s->strstart <= s->window_size-MIN_LOOKAHEAD, "need lookahead"); - - Assert(cur_match < s->strstart, "no future"); - - match = s->window + cur_match; - - /* Return failure if the match length is less than 2: - */ - if (match[0] != scan[0] || match[1] != scan[1]) return MIN_MATCH-1; - - /* The check at best_len-1 can be removed because it will be made - * again later. (This heuristic is not always a win.) - * It is not necessary to compare scan[2] and match[2] since they - * are always equal when the other bytes match, given that - * the hash keys are equal and that HASH_BITS >= 8. - */ - scan += 2, match += 2; - Assert(*scan == *match, "match[2]?"); - - /* We check for insufficient lookahead only every 8th comparison; - * the 256th check will be made at strstart+258. - */ - do { - } while (*++scan == *++match && *++scan == *++match && - *++scan == *++match && *++scan == *++match && - *++scan == *++match && *++scan == *++match && - *++scan == *++match && *++scan == *++match && - scan < strend); - - Assert(scan <= s->window+(unsigned)(s->window_size-1), "wild scan"); - - len = MAX_MATCH - (int)(strend - scan); - - if (len < MIN_MATCH) return MIN_MATCH - 1; - - s->match_start = cur_match; - return (uInt)len <= s->lookahead ? (uInt)len : s->lookahead; -} - -#endif /* FASTEST */ - -#ifdef ZLIB_DEBUG - -#define EQUAL 0 -/* result of memcmp for equal strings */ - -/* =========================================================================== - * Check that the match at match_start is indeed a match. - */ -local void check_match(s, start, match, length) - deflate_state *s; - IPos start, match; - int length; -{ - /* check that the match is indeed a match */ - if (zmemcmp(s->window + match, - s->window + start, length) != EQUAL) { - fprintf(stderr, " start %u, match %u, length %d\n", - start, match, length); - do { - fprintf(stderr, "%c%c", s->window[match++], s->window[start++]); - } while (--length != 0); - z_error("invalid match"); - } - if (z_verbose > 1) { - fprintf(stderr,"\\[%d,%d]", start-match, length); - do { putc(s->window[start++], stderr); } while (--length != 0); - } -} -#else -# define check_match(s, start, match, length) -#endif /* ZLIB_DEBUG */ - -/* =========================================================================== - * Fill the window when the lookahead becomes insufficient. - * Updates strstart and lookahead. - * - * IN assertion: lookahead < MIN_LOOKAHEAD - * OUT assertions: strstart <= window_size-MIN_LOOKAHEAD - * At least one byte has been read, or avail_in == 0; reads are - * performed for at least two bytes (required for the zip translate_eol - * option -- not supported here). - */ -local void fill_window(s) - deflate_state *s; -{ - unsigned n; - unsigned more; /* Amount of free space at the end of the window. */ - uInt wsize = s->w_size; - - Assert(s->lookahead < MIN_LOOKAHEAD, "already enough lookahead"); - - do { - more = (unsigned)(s->window_size -(ulg)s->lookahead -(ulg)s->strstart); - - /* Deal with !@#$% 64K limit: */ - if (sizeof(int) <= 2) { - if (more == 0 && s->strstart == 0 && s->lookahead == 0) { - more = wsize; - - } else if (more == (unsigned)(-1)) { - /* Very unlikely, but possible on 16 bit machine if - * strstart == 0 && lookahead == 1 (input done a byte at time) - */ - more--; - } - } - - /* If the window is almost full and there is insufficient lookahead, - * move the upper half to the lower one to make room in the upper half. - */ - if (s->strstart >= wsize+MAX_DIST(s)) { - - zmemcpy(s->window, s->window+wsize, (unsigned)wsize - more); - s->match_start -= wsize; - s->strstart -= wsize; /* we now have strstart >= MAX_DIST */ - s->block_start -= (long) wsize; - slide_hash(s); - more += wsize; - } - if (s->strm->avail_in == 0) break; - - /* If there was no sliding: - * strstart <= WSIZE+MAX_DIST-1 && lookahead <= MIN_LOOKAHEAD - 1 && - * more == window_size - lookahead - strstart - * => more >= window_size - (MIN_LOOKAHEAD-1 + WSIZE + MAX_DIST-1) - * => more >= window_size - 2*WSIZE + 2 - * In the BIG_MEM or MMAP case (not yet supported), - * window_size == input_size + MIN_LOOKAHEAD && - * strstart + s->lookahead <= input_size => more >= MIN_LOOKAHEAD. - * Otherwise, window_size == 2*WSIZE so more >= 2. - * If there was sliding, more >= WSIZE. So in all cases, more >= 2. - */ - Assert(more >= 2, "more < 2"); - - n = read_buf(s->strm, s->window + s->strstart + s->lookahead, more); - s->lookahead += n; - - /* Initialize the hash value now that we have some input: */ - if (s->lookahead + s->insert >= MIN_MATCH) { - uInt str = s->strstart - s->insert; - s->ins_h = s->window[str]; - UPDATE_HASH(s, s->ins_h, s->window[str + 1]); -#if MIN_MATCH != 3 - Call UPDATE_HASH() MIN_MATCH-3 more times -#endif - while (s->insert) { - UPDATE_HASH(s, s->ins_h, s->window[str + MIN_MATCH-1]); -#ifndef FASTEST - s->prev[str & s->w_mask] = s->head[s->ins_h]; -#endif - s->head[s->ins_h] = (Pos)str; - str++; - s->insert--; - if (s->lookahead + s->insert < MIN_MATCH) - break; - } - } - /* If the whole input has less than MIN_MATCH bytes, ins_h is garbage, - * but this is not important since only literal bytes will be emitted. - */ - - } while (s->lookahead < MIN_LOOKAHEAD && s->strm->avail_in != 0); - - /* If the WIN_INIT bytes after the end of the current data have never been - * written, then zero those bytes in order to avoid memory check reports of - * the use of uninitialized (or uninitialised as Julian writes) bytes by - * the longest match routines. Update the high water mark for the next - * time through here. WIN_INIT is set to MAX_MATCH since the longest match - * routines allow scanning to strstart + MAX_MATCH, ignoring lookahead. - */ - if (s->high_water < s->window_size) { - ulg curr = s->strstart + (ulg)(s->lookahead); - ulg init; - - if (s->high_water < curr) { - /* Previous high water mark below current data -- zero WIN_INIT - * bytes or up to end of window, whichever is less. - */ - init = s->window_size - curr; - if (init > WIN_INIT) - init = WIN_INIT; - zmemzero(s->window + curr, (unsigned)init); - s->high_water = curr + init; - } - else if (s->high_water < (ulg)curr + WIN_INIT) { - /* High water mark at or above current data, but below current data - * plus WIN_INIT -- zero out to current data plus WIN_INIT, or up - * to end of window, whichever is less. - */ - init = (ulg)curr + WIN_INIT - s->high_water; - if (init > s->window_size - s->high_water) - init = s->window_size - s->high_water; - zmemzero(s->window + s->high_water, (unsigned)init); - s->high_water += init; - } - } - - Assert((ulg)s->strstart <= s->window_size - MIN_LOOKAHEAD, - "not enough room for search"); -} - -/* =========================================================================== - * Flush the current block, with given end-of-file flag. - * IN assertion: strstart is set to the end of the current match. - */ -#define FLUSH_BLOCK_ONLY(s, last) { \ - _tr_flush_block(s, (s->block_start >= 0L ? \ - (charf *)&s->window[(unsigned)s->block_start] : \ - (charf *)Z_NULL), \ - (ulg)((long)s->strstart - s->block_start), \ - (last)); \ - s->block_start = s->strstart; \ - flush_pending(s->strm); \ - Tracev((stderr,"[FLUSH]")); \ -} - -/* Same but force premature exit if necessary. */ -#define FLUSH_BLOCK(s, last) { \ - FLUSH_BLOCK_ONLY(s, last); \ - if (s->strm->avail_out == 0) return (last) ? finish_started : need_more; \ -} - -/* Maximum stored block length in deflate format (not including header). */ -#define MAX_STORED 65535 - -/* Minimum of a and b. */ -#define MIN(a, b) ((a) > (b) ? (b) : (a)) - -/* =========================================================================== - * Copy without compression as much as possible from the input stream, return - * the current block state. - * - * In case deflateParams() is used to later switch to a non-zero compression - * level, s->matches (otherwise unused when storing) keeps track of the number - * of hash table slides to perform. If s->matches is 1, then one hash table - * slide will be done when switching. If s->matches is 2, the maximum value - * allowed here, then the hash table will be cleared, since two or more slides - * is the same as a clear. - * - * deflate_stored() is written to minimize the number of times an input byte is - * copied. It is most efficient with large input and output buffers, which - * maximizes the opportunites to have a single copy from next_in to next_out. - */ -local block_state deflate_stored(s, flush) - deflate_state *s; - int flush; -{ - /* Smallest worthy block size when not flushing or finishing. By default - * this is 32K. This can be as small as 507 bytes for memLevel == 1. For - * large input and output buffers, the stored block size will be larger. - */ - unsigned min_block = MIN(s->pending_buf_size - 5, s->w_size); - - /* Copy as many min_block or larger stored blocks directly to next_out as - * possible. If flushing, copy the remaining available input to next_out as - * stored blocks, if there is enough space. - */ - unsigned len, left, have, last = 0; - unsigned used = s->strm->avail_in; - do { - /* Set len to the maximum size block that we can copy directly with the - * available input data and output space. Set left to how much of that - * would be copied from what's left in the window. - */ - len = MAX_STORED; /* maximum deflate stored block length */ - have = (s->bi_valid + 42) >> 3; /* number of header bytes */ - if (s->strm->avail_out < have) /* need room for header */ - break; - /* maximum stored block length that will fit in avail_out: */ - have = s->strm->avail_out - have; - left = s->strstart - s->block_start; /* bytes left in window */ - if (len > (ulg)left + s->strm->avail_in) - len = left + s->strm->avail_in; /* limit len to the input */ - if (len > have) - len = have; /* limit len to the output */ - - /* If the stored block would be less than min_block in length, or if - * unable to copy all of the available input when flushing, then try - * copying to the window and the pending buffer instead. Also don't - * write an empty block when flushing -- deflate() does that. - */ - if (len < min_block && ((len == 0 && flush != Z_FINISH) || - flush == Z_NO_FLUSH || - len != left + s->strm->avail_in)) - break; - - /* Make a dummy stored block in pending to get the header bytes, - * including any pending bits. This also updates the debugging counts. - */ - last = flush == Z_FINISH && len == left + s->strm->avail_in ? 1 : 0; - _tr_stored_block(s, (char *)0, 0L, last); - - /* Replace the lengths in the dummy stored block with len. */ - s->pending_buf[s->pending - 4] = len; - s->pending_buf[s->pending - 3] = len >> 8; - s->pending_buf[s->pending - 2] = ~len; - s->pending_buf[s->pending - 1] = ~len >> 8; - - /* Write the stored block header bytes. */ - flush_pending(s->strm); - -#ifdef ZLIB_DEBUG - /* Update debugging counts for the data about to be copied. */ - s->compressed_len += len << 3; - s->bits_sent += len << 3; -#endif - - /* Copy uncompressed bytes from the window to next_out. */ - if (left) { - if (left > len) - left = len; - zmemcpy(s->strm->next_out, s->window + s->block_start, left); - s->strm->next_out += left; - s->strm->avail_out -= left; - s->strm->total_out += left; - s->block_start += left; - len -= left; - } - - /* Copy uncompressed bytes directly from next_in to next_out, updating - * the check value. - */ - if (len) { - read_buf(s->strm, s->strm->next_out, len); - s->strm->next_out += len; - s->strm->avail_out -= len; - s->strm->total_out += len; - } - } while (last == 0); - - /* Update the sliding window with the last s->w_size bytes of the copied - * data, or append all of the copied data to the existing window if less - * than s->w_size bytes were copied. Also update the number of bytes to - * insert in the hash tables, in the event that deflateParams() switches to - * a non-zero compression level. - */ - used -= s->strm->avail_in; /* number of input bytes directly copied */ - if (used) { - /* If any input was used, then no unused input remains in the window, - * therefore s->block_start == s->strstart. - */ - if (used >= s->w_size) { /* supplant the previous history */ - s->matches = 2; /* clear hash */ - zmemcpy(s->window, s->strm->next_in - s->w_size, s->w_size); - s->strstart = s->w_size; - } - else { - if (s->window_size - s->strstart <= used) { - /* Slide the window down. */ - s->strstart -= s->w_size; - zmemcpy(s->window, s->window + s->w_size, s->strstart); - if (s->matches < 2) - s->matches++; /* add a pending slide_hash() */ - } - zmemcpy(s->window + s->strstart, s->strm->next_in - used, used); - s->strstart += used; - } - s->block_start = s->strstart; - s->insert += MIN(used, s->w_size - s->insert); - } - if (s->high_water < s->strstart) - s->high_water = s->strstart; - - /* If the last block was written to next_out, then done. */ - if (last) - return finish_done; - - /* If flushing and all input has been consumed, then done. */ - if (flush != Z_NO_FLUSH && flush != Z_FINISH && - s->strm->avail_in == 0 && (long)s->strstart == s->block_start) - return block_done; - - /* Fill the window with any remaining input. */ - have = s->window_size - s->strstart - 1; - if (s->strm->avail_in > have && s->block_start >= (long)s->w_size) { - /* Slide the window down. */ - s->block_start -= s->w_size; - s->strstart -= s->w_size; - zmemcpy(s->window, s->window + s->w_size, s->strstart); - if (s->matches < 2) - s->matches++; /* add a pending slide_hash() */ - have += s->w_size; /* more space now */ - } - if (have > s->strm->avail_in) - have = s->strm->avail_in; - if (have) { - read_buf(s->strm, s->window + s->strstart, have); - s->strstart += have; - } - if (s->high_water < s->strstart) - s->high_water = s->strstart; - - /* There was not enough avail_out to write a complete worthy or flushed - * stored block to next_out. Write a stored block to pending instead, if we - * have enough input for a worthy block, or if flushing and there is enough - * room for the remaining input as a stored block in the pending buffer. - */ - have = (s->bi_valid + 42) >> 3; /* number of header bytes */ - /* maximum stored block length that will fit in pending: */ - have = MIN(s->pending_buf_size - have, MAX_STORED); - min_block = MIN(have, s->w_size); - left = s->strstart - s->block_start; - if (left >= min_block || - ((left || flush == Z_FINISH) && flush != Z_NO_FLUSH && - s->strm->avail_in == 0 && left <= have)) { - len = MIN(left, have); - last = flush == Z_FINISH && s->strm->avail_in == 0 && - len == left ? 1 : 0; - _tr_stored_block(s, (charf *)s->window + s->block_start, len, last); - s->block_start += len; - flush_pending(s->strm); - } - - /* We've done all we can with the available input and output. */ - return last ? finish_started : need_more; -} - -/* =========================================================================== - * Compress as much as possible from the input stream, return the current - * block state. - * This function does not perform lazy evaluation of matches and inserts - * new strings in the dictionary only for unmatched strings or for short - * matches. It is used only for the fast compression options. - */ -local block_state deflate_fast(s, flush) - deflate_state *s; - int flush; -{ - IPos hash_head; /* head of the hash chain */ - int bflush; /* set if current block must be flushed */ - - for (;;) { - /* Make sure that we always have enough lookahead, except - * at the end of the input file. We need MAX_MATCH bytes - * for the next match, plus MIN_MATCH bytes to insert the - * string following the next match. - */ - if (s->lookahead < MIN_LOOKAHEAD) { - fill_window(s); - if (s->lookahead < MIN_LOOKAHEAD && flush == Z_NO_FLUSH) { - return need_more; - } - if (s->lookahead == 0) break; /* flush the current block */ - } - - /* Insert the string window[strstart .. strstart+2] in the - * dictionary, and set hash_head to the head of the hash chain: - */ - hash_head = NIL; - if (s->lookahead >= MIN_MATCH) { - INSERT_STRING(s, s->strstart, hash_head); - } - - /* Find the longest match, discarding those <= prev_length. - * At this point we have always match_length < MIN_MATCH - */ - if (hash_head != NIL && s->strstart - hash_head <= MAX_DIST(s)) { - /* To simplify the code, we prevent matches with the string - * of window index 0 (in particular we have to avoid a match - * of the string with itself at the start of the input file). - */ - s->match_length = longest_match (s, hash_head); - /* longest_match() sets match_start */ - } - if (s->match_length >= MIN_MATCH) { - check_match(s, s->strstart, s->match_start, s->match_length); - - _tr_tally_dist(s, s->strstart - s->match_start, - s->match_length - MIN_MATCH, bflush); - - s->lookahead -= s->match_length; - - /* Insert new strings in the hash table only if the match length - * is not too large. This saves time but degrades compression. - */ -#ifndef FASTEST - if (s->match_length <= s->max_insert_length && - s->lookahead >= MIN_MATCH) { - s->match_length--; /* string at strstart already in table */ - do { - s->strstart++; - INSERT_STRING(s, s->strstart, hash_head); - /* strstart never exceeds WSIZE-MAX_MATCH, so there are - * always MIN_MATCH bytes ahead. - */ - } while (--s->match_length != 0); - s->strstart++; - } else -#endif - { - s->strstart += s->match_length; - s->match_length = 0; - s->ins_h = s->window[s->strstart]; - UPDATE_HASH(s, s->ins_h, s->window[s->strstart+1]); -#if MIN_MATCH != 3 - Call UPDATE_HASH() MIN_MATCH-3 more times -#endif - /* If lookahead < MIN_MATCH, ins_h is garbage, but it does not - * matter since it will be recomputed at next deflate call. - */ - } - } else { - /* No match, output a literal byte */ - Tracevv((stderr,"%c", s->window[s->strstart])); - _tr_tally_lit (s, s->window[s->strstart], bflush); - s->lookahead--; - s->strstart++; - } - if (bflush) FLUSH_BLOCK(s, 0); - } - s->insert = s->strstart < MIN_MATCH-1 ? s->strstart : MIN_MATCH-1; - if (flush == Z_FINISH) { - FLUSH_BLOCK(s, 1); - return finish_done; - } - if (s->last_lit) - FLUSH_BLOCK(s, 0); - return block_done; -} - -#ifndef FASTEST -/* =========================================================================== - * Same as above, but achieves better compression. We use a lazy - * evaluation for matches: a match is finally adopted only if there is - * no better match at the next window position. - */ -local block_state deflate_slow(s, flush) - deflate_state *s; - int flush; -{ - IPos hash_head; /* head of hash chain */ - int bflush; /* set if current block must be flushed */ - - /* Process the input block. */ - for (;;) { - /* Make sure that we always have enough lookahead, except - * at the end of the input file. We need MAX_MATCH bytes - * for the next match, plus MIN_MATCH bytes to insert the - * string following the next match. - */ - if (s->lookahead < MIN_LOOKAHEAD) { - fill_window(s); - if (s->lookahead < MIN_LOOKAHEAD && flush == Z_NO_FLUSH) { - return need_more; - } - if (s->lookahead == 0) break; /* flush the current block */ - } - - /* Insert the string window[strstart .. strstart+2] in the - * dictionary, and set hash_head to the head of the hash chain: - */ - hash_head = NIL; - if (s->lookahead >= MIN_MATCH) { - INSERT_STRING(s, s->strstart, hash_head); - } - - /* Find the longest match, discarding those <= prev_length. - */ - s->prev_length = s->match_length, s->prev_match = s->match_start; - s->match_length = MIN_MATCH-1; - - if (hash_head != NIL && s->prev_length < s->max_lazy_match && - s->strstart - hash_head <= MAX_DIST(s)) { - /* To simplify the code, we prevent matches with the string - * of window index 0 (in particular we have to avoid a match - * of the string with itself at the start of the input file). - */ - s->match_length = longest_match (s, hash_head); - /* longest_match() sets match_start */ - - if (s->match_length <= 5 && (s->strategy == Z_FILTERED -#if TOO_FAR <= 32767 - || (s->match_length == MIN_MATCH && - s->strstart - s->match_start > TOO_FAR) -#endif - )) { - - /* If prev_match is also MIN_MATCH, match_start is garbage - * but we will ignore the current match anyway. - */ - s->match_length = MIN_MATCH-1; - } - } - /* If there was a match at the previous step and the current - * match is not better, output the previous match: - */ - if (s->prev_length >= MIN_MATCH && s->match_length <= s->prev_length) { - uInt max_insert = s->strstart + s->lookahead - MIN_MATCH; - /* Do not insert strings in hash table beyond this. */ - - check_match(s, s->strstart-1, s->prev_match, s->prev_length); - - _tr_tally_dist(s, s->strstart -1 - s->prev_match, - s->prev_length - MIN_MATCH, bflush); - - /* Insert in hash table all strings up to the end of the match. - * strstart-1 and strstart are already inserted. If there is not - * enough lookahead, the last two strings are not inserted in - * the hash table. - */ - s->lookahead -= s->prev_length-1; - s->prev_length -= 2; - do { - if (++s->strstart <= max_insert) { - INSERT_STRING(s, s->strstart, hash_head); - } - } while (--s->prev_length != 0); - s->match_available = 0; - s->match_length = MIN_MATCH-1; - s->strstart++; - - if (bflush) FLUSH_BLOCK(s, 0); - - } else if (s->match_available) { - /* If there was no match at the previous position, output a - * single literal. If there was a match but the current match - * is longer, truncate the previous match to a single literal. - */ - Tracevv((stderr,"%c", s->window[s->strstart-1])); - _tr_tally_lit(s, s->window[s->strstart-1], bflush); - if (bflush) { - FLUSH_BLOCK_ONLY(s, 0); - } - s->strstart++; - s->lookahead--; - if (s->strm->avail_out == 0) return need_more; - } else { - /* There is no previous match to compare with, wait for - * the next step to decide. - */ - s->match_available = 1; - s->strstart++; - s->lookahead--; - } - } - Assert (flush != Z_NO_FLUSH, "no flush?"); - if (s->match_available) { - Tracevv((stderr,"%c", s->window[s->strstart-1])); - _tr_tally_lit(s, s->window[s->strstart-1], bflush); - s->match_available = 0; - } - s->insert = s->strstart < MIN_MATCH-1 ? s->strstart : MIN_MATCH-1; - if (flush == Z_FINISH) { - FLUSH_BLOCK(s, 1); - return finish_done; - } - if (s->last_lit) - FLUSH_BLOCK(s, 0); - return block_done; -} -#endif /* FASTEST */ - -/* =========================================================================== - * For Z_RLE, simply look for runs of bytes, generate matches only of distance - * one. Do not maintain a hash table. (It will be regenerated if this run of - * deflate switches away from Z_RLE.) - */ -local block_state deflate_rle(s, flush) - deflate_state *s; - int flush; -{ - int bflush; /* set if current block must be flushed */ - uInt prev; /* byte at distance one to match */ - Bytef *scan, *strend; /* scan goes up to strend for length of run */ - - for (;;) { - /* Make sure that we always have enough lookahead, except - * at the end of the input file. We need MAX_MATCH bytes - * for the longest run, plus one for the unrolled loop. - */ - if (s->lookahead <= MAX_MATCH) { - fill_window(s); - if (s->lookahead <= MAX_MATCH && flush == Z_NO_FLUSH) { - return need_more; - } - if (s->lookahead == 0) break; /* flush the current block */ - } - - /* See how many times the previous byte repeats */ - s->match_length = 0; - if (s->lookahead >= MIN_MATCH && s->strstart > 0) { - scan = s->window + s->strstart - 1; - prev = *scan; - if (prev == *++scan && prev == *++scan && prev == *++scan) { - strend = s->window + s->strstart + MAX_MATCH; - do { - } while (prev == *++scan && prev == *++scan && - prev == *++scan && prev == *++scan && - prev == *++scan && prev == *++scan && - prev == *++scan && prev == *++scan && - scan < strend); - s->match_length = MAX_MATCH - (uInt)(strend - scan); - if (s->match_length > s->lookahead) - s->match_length = s->lookahead; - } - Assert(scan <= s->window+(uInt)(s->window_size-1), "wild scan"); - } - - /* Emit match if have run of MIN_MATCH or longer, else emit literal */ - if (s->match_length >= MIN_MATCH) { - check_match(s, s->strstart, s->strstart - 1, s->match_length); - - _tr_tally_dist(s, 1, s->match_length - MIN_MATCH, bflush); - - s->lookahead -= s->match_length; - s->strstart += s->match_length; - s->match_length = 0; - } else { - /* No match, output a literal byte */ - Tracevv((stderr,"%c", s->window[s->strstart])); - _tr_tally_lit (s, s->window[s->strstart], bflush); - s->lookahead--; - s->strstart++; - } - if (bflush) FLUSH_BLOCK(s, 0); - } - s->insert = 0; - if (flush == Z_FINISH) { - FLUSH_BLOCK(s, 1); - return finish_done; - } - if (s->last_lit) - FLUSH_BLOCK(s, 0); - return block_done; -} - -/* =========================================================================== - * For Z_HUFFMAN_ONLY, do not look for matches. Do not maintain a hash table. - * (It will be regenerated if this run of deflate switches away from Huffman.) - */ -local block_state deflate_huff(s, flush) - deflate_state *s; - int flush; -{ - int bflush; /* set if current block must be flushed */ - - for (;;) { - /* Make sure that we have a literal to write. */ - if (s->lookahead == 0) { - fill_window(s); - if (s->lookahead == 0) { - if (flush == Z_NO_FLUSH) - return need_more; - break; /* flush the current block */ - } - } - - /* Output a literal byte */ - s->match_length = 0; - Tracevv((stderr,"%c", s->window[s->strstart])); - _tr_tally_lit (s, s->window[s->strstart], bflush); - s->lookahead--; - s->strstart++; - if (bflush) FLUSH_BLOCK(s, 0); - } - s->insert = 0; - if (flush == Z_FINISH) { - FLUSH_BLOCK(s, 1); - return finish_done; - } - if (s->last_lit) - FLUSH_BLOCK(s, 0); - return block_done; -} diff --git a/dep/zlib/src/deflate.h b/dep/zlib/src/deflate.h deleted file mode 100644 index 23ecdd312..000000000 --- a/dep/zlib/src/deflate.h +++ /dev/null @@ -1,349 +0,0 @@ -/* deflate.h -- internal compression state - * Copyright (C) 1995-2016 Jean-loup Gailly - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* WARNING: this file should *not* be used by applications. It is - part of the implementation of the compression library and is - subject to change. Applications should only use zlib.h. - */ - -/* @(#) $Id$ */ - -#ifndef DEFLATE_H -#define DEFLATE_H - -#include "zutil.h" - -/* define NO_GZIP when compiling if you want to disable gzip header and - trailer creation by deflate(). NO_GZIP would be used to avoid linking in - the crc code when it is not needed. For shared libraries, gzip encoding - should be left enabled. */ -#ifndef NO_GZIP -# define GZIP -#endif - -/* =========================================================================== - * Internal compression state. - */ - -#define LENGTH_CODES 29 -/* number of length codes, not counting the special END_BLOCK code */ - -#define LITERALS 256 -/* number of literal bytes 0..255 */ - -#define L_CODES (LITERALS+1+LENGTH_CODES) -/* number of Literal or Length codes, including the END_BLOCK code */ - -#define D_CODES 30 -/* number of distance codes */ - -#define BL_CODES 19 -/* number of codes used to transfer the bit lengths */ - -#define HEAP_SIZE (2*L_CODES+1) -/* maximum heap size */ - -#define MAX_BITS 15 -/* All codes must not exceed MAX_BITS bits */ - -#define Buf_size 16 -/* size of bit buffer in bi_buf */ - -#define INIT_STATE 42 /* zlib header -> BUSY_STATE */ -#ifdef GZIP -# define GZIP_STATE 57 /* gzip header -> BUSY_STATE | EXTRA_STATE */ -#endif -#define EXTRA_STATE 69 /* gzip extra block -> NAME_STATE */ -#define NAME_STATE 73 /* gzip file name -> COMMENT_STATE */ -#define COMMENT_STATE 91 /* gzip comment -> HCRC_STATE */ -#define HCRC_STATE 103 /* gzip header CRC -> BUSY_STATE */ -#define BUSY_STATE 113 /* deflate -> FINISH_STATE */ -#define FINISH_STATE 666 /* stream complete */ -/* Stream status */ - - -/* Data structure describing a single value and its code string. */ -typedef struct ct_data_s { - union { - ush freq; /* frequency count */ - ush code; /* bit string */ - } fc; - union { - ush dad; /* father node in Huffman tree */ - ush len; /* length of bit string */ - } dl; -} FAR ct_data; - -#define Freq fc.freq -#define Code fc.code -#define Dad dl.dad -#define Len dl.len - -typedef struct static_tree_desc_s static_tree_desc; - -typedef struct tree_desc_s { - ct_data *dyn_tree; /* the dynamic tree */ - int max_code; /* largest code with non zero frequency */ - const static_tree_desc *stat_desc; /* the corresponding static tree */ -} FAR tree_desc; - -typedef ush Pos; -typedef Pos FAR Posf; -typedef unsigned IPos; - -/* A Pos is an index in the character window. We use short instead of int to - * save space in the various tables. IPos is used only for parameter passing. - */ - -typedef struct internal_state { - z_streamp strm; /* pointer back to this zlib stream */ - int status; /* as the name implies */ - Bytef *pending_buf; /* output still pending */ - ulg pending_buf_size; /* size of pending_buf */ - Bytef *pending_out; /* next pending byte to output to the stream */ - ulg pending; /* nb of bytes in the pending buffer */ - int wrap; /* bit 0 true for zlib, bit 1 true for gzip */ - gz_headerp gzhead; /* gzip header information to write */ - ulg gzindex; /* where in extra, name, or comment */ - Byte method; /* can only be DEFLATED */ - int last_flush; /* value of flush param for previous deflate call */ - - /* used by deflate.c: */ - - uInt w_size; /* LZ77 window size (32K by default) */ - uInt w_bits; /* log2(w_size) (8..16) */ - uInt w_mask; /* w_size - 1 */ - - Bytef *window; - /* Sliding window. Input bytes are read into the second half of the window, - * and move to the first half later to keep a dictionary of at least wSize - * bytes. With this organization, matches are limited to a distance of - * wSize-MAX_MATCH bytes, but this ensures that IO is always - * performed with a length multiple of the block size. Also, it limits - * the window size to 64K, which is quite useful on MSDOS. - * To do: use the user input buffer as sliding window. - */ - - ulg window_size; - /* Actual size of window: 2*wSize, except when the user input buffer - * is directly used as sliding window. - */ - - Posf *prev; - /* Link to older string with same hash index. To limit the size of this - * array to 64K, this link is maintained only for the last 32K strings. - * An index in this array is thus a window index modulo 32K. - */ - - Posf *head; /* Heads of the hash chains or NIL. */ - - uInt ins_h; /* hash index of string to be inserted */ - uInt hash_size; /* number of elements in hash table */ - uInt hash_bits; /* log2(hash_size) */ - uInt hash_mask; /* hash_size-1 */ - - uInt hash_shift; - /* Number of bits by which ins_h must be shifted at each input - * step. It must be such that after MIN_MATCH steps, the oldest - * byte no longer takes part in the hash key, that is: - * hash_shift * MIN_MATCH >= hash_bits - */ - - long block_start; - /* Window position at the beginning of the current output block. Gets - * negative when the window is moved backwards. - */ - - uInt match_length; /* length of best match */ - IPos prev_match; /* previous match */ - int match_available; /* set if previous match exists */ - uInt strstart; /* start of string to insert */ - uInt match_start; /* start of matching string */ - uInt lookahead; /* number of valid bytes ahead in window */ - - uInt prev_length; - /* Length of the best match at previous step. Matches not greater than this - * are discarded. This is used in the lazy match evaluation. - */ - - uInt max_chain_length; - /* To speed up deflation, hash chains are never searched beyond this - * length. A higher limit improves compression ratio but degrades the - * speed. - */ - - uInt max_lazy_match; - /* Attempt to find a better match only when the current match is strictly - * smaller than this value. This mechanism is used only for compression - * levels >= 4. - */ -# define max_insert_length max_lazy_match - /* Insert new strings in the hash table only if the match length is not - * greater than this length. This saves time but degrades compression. - * max_insert_length is used only for compression levels <= 3. - */ - - int level; /* compression level (1..9) */ - int strategy; /* favor or force Huffman coding*/ - - uInt good_match; - /* Use a faster search when the previous match is longer than this */ - - int nice_match; /* Stop searching when current match exceeds this */ - - /* used by trees.c: */ - /* Didn't use ct_data typedef below to suppress compiler warning */ - struct ct_data_s dyn_ltree[HEAP_SIZE]; /* literal and length tree */ - struct ct_data_s dyn_dtree[2*D_CODES+1]; /* distance tree */ - struct ct_data_s bl_tree[2*BL_CODES+1]; /* Huffman tree for bit lengths */ - - struct tree_desc_s l_desc; /* desc. for literal tree */ - struct tree_desc_s d_desc; /* desc. for distance tree */ - struct tree_desc_s bl_desc; /* desc. for bit length tree */ - - ush bl_count[MAX_BITS+1]; - /* number of codes at each bit length for an optimal tree */ - - int heap[2*L_CODES+1]; /* heap used to build the Huffman trees */ - int heap_len; /* number of elements in the heap */ - int heap_max; /* element of largest frequency */ - /* The sons of heap[n] are heap[2*n] and heap[2*n+1]. heap[0] is not used. - * The same heap array is used to build all trees. - */ - - uch depth[2*L_CODES+1]; - /* Depth of each subtree used as tie breaker for trees of equal frequency - */ - - uchf *l_buf; /* buffer for literals or lengths */ - - uInt lit_bufsize; - /* Size of match buffer for literals/lengths. There are 4 reasons for - * limiting lit_bufsize to 64K: - * - frequencies can be kept in 16 bit counters - * - if compression is not successful for the first block, all input - * data is still in the window so we can still emit a stored block even - * when input comes from standard input. (This can also be done for - * all blocks if lit_bufsize is not greater than 32K.) - * - if compression is not successful for a file smaller than 64K, we can - * even emit a stored file instead of a stored block (saving 5 bytes). - * This is applicable only for zip (not gzip or zlib). - * - creating new Huffman trees less frequently may not provide fast - * adaptation to changes in the input data statistics. (Take for - * example a binary file with poorly compressible code followed by - * a highly compressible string table.) Smaller buffer sizes give - * fast adaptation but have of course the overhead of transmitting - * trees more frequently. - * - I can't count above 4 - */ - - uInt last_lit; /* running index in l_buf */ - - ushf *d_buf; - /* Buffer for distances. To simplify the code, d_buf and l_buf have - * the same number of elements. To use different lengths, an extra flag - * array would be necessary. - */ - - ulg opt_len; /* bit length of current block with optimal trees */ - ulg static_len; /* bit length of current block with static trees */ - uInt matches; /* number of string matches in current block */ - uInt insert; /* bytes at end of window left to insert */ - -#ifdef ZLIB_DEBUG - ulg compressed_len; /* total bit length of compressed file mod 2^32 */ - ulg bits_sent; /* bit length of compressed data sent mod 2^32 */ -#endif - - ush bi_buf; - /* Output buffer. bits are inserted starting at the bottom (least - * significant bits). - */ - int bi_valid; - /* Number of valid bits in bi_buf. All bits above the last valid bit - * are always zero. - */ - - ulg high_water; - /* High water mark offset in window for initialized bytes -- bytes above - * this are set to zero in order to avoid memory check warnings when - * longest match routines access bytes past the input. This is then - * updated to the new high water mark. - */ - -} FAR deflate_state; - -/* Output a byte on the stream. - * IN assertion: there is enough room in pending_buf. - */ -#define put_byte(s, c) {s->pending_buf[s->pending++] = (Bytef)(c);} - - -#define MIN_LOOKAHEAD (MAX_MATCH+MIN_MATCH+1) -/* Minimum amount of lookahead, except at the end of the input file. - * See deflate.c for comments about the MIN_MATCH+1. - */ - -#define MAX_DIST(s) ((s)->w_size-MIN_LOOKAHEAD) -/* In order to simplify the code, particularly on 16 bit machines, match - * distances are limited to MAX_DIST instead of WSIZE. - */ - -#define WIN_INIT MAX_MATCH -/* Number of bytes after end of data in window to initialize in order to avoid - memory checker errors from longest match routines */ - - /* in trees.c */ -void ZLIB_INTERNAL _tr_init OF((deflate_state *s)); -int ZLIB_INTERNAL _tr_tally OF((deflate_state *s, unsigned dist, unsigned lc)); -void ZLIB_INTERNAL _tr_flush_block OF((deflate_state *s, charf *buf, - ulg stored_len, int last)); -void ZLIB_INTERNAL _tr_flush_bits OF((deflate_state *s)); -void ZLIB_INTERNAL _tr_align OF((deflate_state *s)); -void ZLIB_INTERNAL _tr_stored_block OF((deflate_state *s, charf *buf, - ulg stored_len, int last)); - -#define d_code(dist) \ - ((dist) < 256 ? _dist_code[dist] : _dist_code[256+((dist)>>7)]) -/* Mapping from a distance to a distance code. dist is the distance - 1 and - * must not have side effects. _dist_code[256] and _dist_code[257] are never - * used. - */ - -#ifndef ZLIB_DEBUG -/* Inline versions of _tr_tally for speed: */ - -#if defined(GEN_TREES_H) || !defined(STDC) - extern uch ZLIB_INTERNAL _length_code[]; - extern uch ZLIB_INTERNAL _dist_code[]; -#else - extern const uch ZLIB_INTERNAL _length_code[]; - extern const uch ZLIB_INTERNAL _dist_code[]; -#endif - -# define _tr_tally_lit(s, c, flush) \ - { uch cc = (c); \ - s->d_buf[s->last_lit] = 0; \ - s->l_buf[s->last_lit++] = cc; \ - s->dyn_ltree[cc].Freq++; \ - flush = (s->last_lit == s->lit_bufsize-1); \ - } -# define _tr_tally_dist(s, distance, length, flush) \ - { uch len = (uch)(length); \ - ush dist = (ush)(distance); \ - s->d_buf[s->last_lit] = dist; \ - s->l_buf[s->last_lit++] = len; \ - dist--; \ - s->dyn_ltree[_length_code[len]+LITERALS+1].Freq++; \ - s->dyn_dtree[d_code(dist)].Freq++; \ - flush = (s->last_lit == s->lit_bufsize-1); \ - } -#else -# define _tr_tally_lit(s, c, flush) flush = _tr_tally(s, 0, c) -# define _tr_tally_dist(s, distance, length, flush) \ - flush = _tr_tally(s, distance, length) -#endif - -#endif /* DEFLATE_H */ diff --git a/dep/zlib/src/gzclose.c b/dep/zlib/src/gzclose.c deleted file mode 100644 index caeb99a31..000000000 --- a/dep/zlib/src/gzclose.c +++ /dev/null @@ -1,25 +0,0 @@ -/* gzclose.c -- zlib gzclose() function - * Copyright (C) 2004, 2010 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -#include "gzguts.h" - -/* gzclose() is in a separate file so that it is linked in only if it is used. - That way the other gzclose functions can be used instead to avoid linking in - unneeded compression or decompression routines. */ -int ZEXPORT gzclose(file) - gzFile file; -{ -#ifndef NO_GZCOMPRESS - gz_statep state; - - if (file == NULL) - return Z_STREAM_ERROR; - state = (gz_statep)file; - - return state->mode == GZ_READ ? gzclose_r(file) : gzclose_w(file); -#else - return gzclose_r(file); -#endif -} diff --git a/dep/zlib/src/gzguts.h b/dep/zlib/src/gzguts.h deleted file mode 100644 index 1b56c48a5..000000000 --- a/dep/zlib/src/gzguts.h +++ /dev/null @@ -1,220 +0,0 @@ -/* gzguts.h -- zlib internal header definitions for gz* operations - * Copyright (C) 2004, 2005, 2010, 2011, 2012, 2013, 2016 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -#ifdef _LARGEFILE64_SOURCE -# ifndef _LARGEFILE_SOURCE -# define _LARGEFILE_SOURCE 1 -# endif -# ifdef _FILE_OFFSET_BITS -# undef _FILE_OFFSET_BITS -# endif -#endif - -#ifdef HAVE_HIDDEN -# define ZLIB_INTERNAL __attribute__((visibility ("hidden"))) -#else -# define ZLIB_INTERNAL -#endif - -#include -#include "zlib.h" -#ifdef STDC -# include -# include -# include -#endif - -#ifndef _POSIX_SOURCE -# define _POSIX_SOURCE -#endif -#include - -#ifdef _WIN32 -# include -#endif - -#if defined(__TURBOC__) || defined(_MSC_VER) || defined(_WIN32) -# include -#else -# include -#endif - -#if defined(_WIN32) || defined(__CYGWIN__) -# define WIDECHAR -#endif - -#ifdef WINAPI_FAMILY -# define open _open -# define read _read -# define write _write -# define close _close -#endif - -#ifdef NO_DEFLATE /* for compatibility with old definition */ -# define NO_GZCOMPRESS -#endif - -#if defined(STDC99) || (defined(__TURBOC__) && __TURBOC__ >= 0x550) -# ifndef HAVE_VSNPRINTF -# define HAVE_VSNPRINTF -# endif -#endif - -#if defined(__CYGWIN__) -# ifndef HAVE_VSNPRINTF -# define HAVE_VSNPRINTF -# endif -#endif - -#if defined(MSDOS) && defined(__BORLANDC__) && (BORLANDC > 0x410) -# ifndef HAVE_VSNPRINTF -# define HAVE_VSNPRINTF -# endif -#endif - -#ifndef HAVE_VSNPRINTF -# ifdef MSDOS -/* vsnprintf may exist on some MS-DOS compilers (DJGPP?), - but for now we just assume it doesn't. */ -# define NO_vsnprintf -# endif -# ifdef __TURBOC__ -# define NO_vsnprintf -# endif -# ifdef WIN32 -/* In Win32, vsnprintf is available as the "non-ANSI" _vsnprintf. */ -# if !defined(vsnprintf) && !defined(NO_vsnprintf) -# if !defined(_MSC_VER) || ( defined(_MSC_VER) && _MSC_VER < 1500 ) -# define vsnprintf _vsnprintf -# endif -# endif -# endif -# ifdef __SASC -# define NO_vsnprintf -# endif -# ifdef VMS -# define NO_vsnprintf -# endif -# ifdef __OS400__ -# define NO_vsnprintf -# endif -# ifdef __MVS__ -# define NO_vsnprintf -# endif -#endif - -/* unlike snprintf (which is required in C99), _snprintf does not guarantee - null termination of the result -- however this is only used in gzlib.c where - the result is assured to fit in the space provided */ -#if defined(_MSC_VER) && _MSC_VER < 1900 -# define snprintf _snprintf -#endif - -#ifndef local -# define local static -#endif -/* since "static" is used to mean two completely different things in C, we - define "local" for the non-static meaning of "static", for readability - (compile with -Dlocal if your debugger can't find static symbols) */ - -/* gz* functions always use library allocation functions */ -#ifndef STDC - extern voidp malloc OF((uInt size)); - extern void free OF((voidpf ptr)); -#endif - -/* get errno and strerror definition */ -#if defined UNDER_CE -# include -# define zstrerror() gz_strwinerror((DWORD)GetLastError()) -#else -# ifndef NO_STRERROR -# include -# define zstrerror() strerror(errno) -# else -# define zstrerror() "stdio error (consult errno)" -# endif -#endif - -/* provide prototypes for these when building zlib without LFS */ -#if !defined(_LARGEFILE64_SOURCE) || _LFS64_LARGEFILE-0 == 0 - ZEXTERN gzFile ZEXPORT gzopen64 OF((const char *, const char *)); - ZEXTERN z_off64_t ZEXPORT gzseek64 OF((gzFile, z_off64_t, int)); - ZEXTERN z_off64_t ZEXPORT gztell64 OF((gzFile)); - ZEXTERN z_off64_t ZEXPORT gzoffset64 OF((gzFile)); -#endif - -/* default memLevel */ -#if MAX_MEM_LEVEL >= 8 -# define DEF_MEM_LEVEL 8 -#else -# define DEF_MEM_LEVEL MAX_MEM_LEVEL -#endif - -/* default i/o buffer size -- double this for output when reading (this and - twice this must be able to fit in an unsigned type) */ -#define GZBUFSIZE 8192 - -/* gzip modes, also provide a little integrity check on the passed structure */ -#define GZ_NONE 0 -#define GZ_READ 7247 -#define GZ_WRITE 31153 -#define GZ_APPEND 1 /* mode set to GZ_WRITE after the file is opened */ - -/* values for gz_state how */ -#define LOOK 0 /* look for a gzip header */ -#define COPY 1 /* copy input directly */ -#define GZIP 2 /* decompress a gzip stream */ - -/* internal gzip file state data structure */ -typedef struct { - /* exposed contents for gzgetc() macro */ - struct gzFile_s x; /* "x" for exposed */ - /* x.have: number of bytes available at x.next */ - /* x.next: next output data to deliver or write */ - /* x.pos: current position in uncompressed data */ - /* used for both reading and writing */ - int mode; /* see gzip modes above */ - int fd; /* file descriptor */ - char *path; /* path or fd for error messages */ - unsigned size; /* buffer size, zero if not allocated yet */ - unsigned want; /* requested buffer size, default is GZBUFSIZE */ - unsigned char *in; /* input buffer (double-sized when writing) */ - unsigned char *out; /* output buffer (double-sized when reading) */ - int direct; /* 0 if processing gzip, 1 if transparent */ - /* just for reading */ - int how; /* 0: get header, 1: copy, 2: decompress */ - z_off64_t start; /* where the gzip data started, for rewinding */ - int eof; /* true if end of input file reached */ - int past; /* true if read requested past end */ - /* just for writing */ - int level; /* compression level */ - int strategy; /* compression strategy */ - /* seek request */ - z_off64_t skip; /* amount to skip (already rewound if backwards) */ - int seek; /* true if seek request pending */ - /* error information */ - int err; /* error code */ - char *msg; /* error message */ - /* zlib inflate or deflate stream */ - z_stream strm; /* stream structure in-place (not a pointer) */ -} gz_state; -typedef gz_state FAR *gz_statep; - -/* shared functions */ -void ZLIB_INTERNAL gz_error OF((gz_statep, int, const char *)); -#if defined UNDER_CE -char ZLIB_INTERNAL *gz_strwinerror OF((DWORD error)); -#endif - -/* GT_OFF(x), where x is an unsigned value, is true if x > maximum z_off64_t - value -- needed when comparing unsigned to z_off64_t, which is signed - (possible z_off64_t types off_t, off64_t, and long are all signed) */ -#ifdef INT_MAX -# define GT_OFF(x) (sizeof(int) == sizeof(z_off64_t) && (x) > INT_MAX) -#else -unsigned ZLIB_INTERNAL gz_intmax OF((void)); -# define GT_OFF(x) (sizeof(int) == sizeof(z_off64_t) && (x) > gz_intmax()) -#endif diff --git a/dep/zlib/src/gzlib.c b/dep/zlib/src/gzlib.c deleted file mode 100644 index 4105e6aff..000000000 --- a/dep/zlib/src/gzlib.c +++ /dev/null @@ -1,637 +0,0 @@ -/* gzlib.c -- zlib functions common to reading and writing gzip files - * Copyright (C) 2004-2017 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -#include "gzguts.h" - -#if defined(_WIN32) && !defined(__BORLANDC__) && !defined(__MINGW32__) -# define LSEEK _lseeki64 -#else -#if defined(_LARGEFILE64_SOURCE) && _LFS64_LARGEFILE-0 -# define LSEEK lseek64 -#else -# define LSEEK lseek -#endif -#endif - -/* Local functions */ -local void gz_reset OF((gz_statep)); -local gzFile gz_open OF((const void *, int, const char *)); - -#if defined UNDER_CE - -/* Map the Windows error number in ERROR to a locale-dependent error message - string and return a pointer to it. Typically, the values for ERROR come - from GetLastError. - - The string pointed to shall not be modified by the application, but may be - overwritten by a subsequent call to gz_strwinerror - - The gz_strwinerror function does not change the current setting of - GetLastError. */ -char ZLIB_INTERNAL *gz_strwinerror (error) - DWORD error; -{ - static char buf[1024]; - - wchar_t *msgbuf; - DWORD lasterr = GetLastError(); - DWORD chars = FormatMessage(FORMAT_MESSAGE_FROM_SYSTEM - | FORMAT_MESSAGE_ALLOCATE_BUFFER, - NULL, - error, - 0, /* Default language */ - (LPVOID)&msgbuf, - 0, - NULL); - if (chars != 0) { - /* If there is an \r\n appended, zap it. */ - if (chars >= 2 - && msgbuf[chars - 2] == '\r' && msgbuf[chars - 1] == '\n') { - chars -= 2; - msgbuf[chars] = 0; - } - - if (chars > sizeof (buf) - 1) { - chars = sizeof (buf) - 1; - msgbuf[chars] = 0; - } - - wcstombs(buf, msgbuf, chars + 1); - LocalFree(msgbuf); - } - else { - sprintf(buf, "unknown win32 error (%ld)", error); - } - - SetLastError(lasterr); - return buf; -} - -#endif /* UNDER_CE */ - -/* Reset gzip file state */ -local void gz_reset(state) - gz_statep state; -{ - state->x.have = 0; /* no output data available */ - if (state->mode == GZ_READ) { /* for reading ... */ - state->eof = 0; /* not at end of file */ - state->past = 0; /* have not read past end yet */ - state->how = LOOK; /* look for gzip header */ - } - state->seek = 0; /* no seek request pending */ - gz_error(state, Z_OK, NULL); /* clear error */ - state->x.pos = 0; /* no uncompressed data yet */ - state->strm.avail_in = 0; /* no input data yet */ -} - -/* Open a gzip file either by name or file descriptor. */ -local gzFile gz_open(path, fd, mode) - const void *path; - int fd; - const char *mode; -{ - gz_statep state; - z_size_t len; - int oflag; -#ifdef O_CLOEXEC - int cloexec = 0; -#endif -#ifdef O_EXCL - int exclusive = 0; -#endif - - /* check input */ - if (path == NULL) - return NULL; - - /* allocate gzFile structure to return */ - state = (gz_statep)malloc(sizeof(gz_state)); - if (state == NULL) - return NULL; - state->size = 0; /* no buffers allocated yet */ - state->want = GZBUFSIZE; /* requested buffer size */ - state->msg = NULL; /* no error message yet */ - - /* interpret mode */ - state->mode = GZ_NONE; - state->level = Z_DEFAULT_COMPRESSION; - state->strategy = Z_DEFAULT_STRATEGY; - state->direct = 0; - while (*mode) { - if (*mode >= '0' && *mode <= '9') - state->level = *mode - '0'; - else - switch (*mode) { - case 'r': - state->mode = GZ_READ; - break; -#ifndef NO_GZCOMPRESS - case 'w': - state->mode = GZ_WRITE; - break; - case 'a': - state->mode = GZ_APPEND; - break; -#endif - case '+': /* can't read and write at the same time */ - free(state); - return NULL; - case 'b': /* ignore -- will request binary anyway */ - break; -#ifdef O_CLOEXEC - case 'e': - cloexec = 1; - break; -#endif -#ifdef O_EXCL - case 'x': - exclusive = 1; - break; -#endif - case 'f': - state->strategy = Z_FILTERED; - break; - case 'h': - state->strategy = Z_HUFFMAN_ONLY; - break; - case 'R': - state->strategy = Z_RLE; - break; - case 'F': - state->strategy = Z_FIXED; - break; - case 'T': - state->direct = 1; - break; - default: /* could consider as an error, but just ignore */ - ; - } - mode++; - } - - /* must provide an "r", "w", or "a" */ - if (state->mode == GZ_NONE) { - free(state); - return NULL; - } - - /* can't force transparent read */ - if (state->mode == GZ_READ) { - if (state->direct) { - free(state); - return NULL; - } - state->direct = 1; /* for empty file */ - } - - /* save the path name for error messages */ -#ifdef WIDECHAR - if (fd == -2) { - len = wcstombs(NULL, path, 0); - if (len == (z_size_t)-1) - len = 0; - } - else -#endif - len = strlen((const char *)path); - state->path = (char *)malloc(len + 1); - if (state->path == NULL) { - free(state); - return NULL; - } -#ifdef WIDECHAR - if (fd == -2) - if (len) - wcstombs(state->path, path, len + 1); - else - *(state->path) = 0; - else -#endif -#if !defined(NO_snprintf) && !defined(NO_vsnprintf) - (void)snprintf(state->path, len + 1, "%s", (const char *)path); -#else - strcpy(state->path, path); -#endif - - /* compute the flags for open() */ - oflag = -#ifdef O_LARGEFILE - O_LARGEFILE | -#endif -#ifdef O_BINARY - O_BINARY | -#endif -#ifdef O_CLOEXEC - (cloexec ? O_CLOEXEC : 0) | -#endif - (state->mode == GZ_READ ? - O_RDONLY : - (O_WRONLY | O_CREAT | -#ifdef O_EXCL - (exclusive ? O_EXCL : 0) | -#endif - (state->mode == GZ_WRITE ? - O_TRUNC : - O_APPEND))); - - /* open the file with the appropriate flags (or just use fd) */ - state->fd = fd > -1 ? fd : ( -#ifdef WIDECHAR - fd == -2 ? _wopen(path, oflag, 0666) : -#endif - open((const char *)path, oflag, 0666)); - if (state->fd == -1) { - free(state->path); - free(state); - return NULL; - } - if (state->mode == GZ_APPEND) { - LSEEK(state->fd, 0, SEEK_END); /* so gzoffset() is correct */ - state->mode = GZ_WRITE; /* simplify later checks */ - } - - /* save the current position for rewinding (only if reading) */ - if (state->mode == GZ_READ) { - state->start = LSEEK(state->fd, 0, SEEK_CUR); - if (state->start == -1) state->start = 0; - } - - /* initialize stream */ - gz_reset(state); - - /* return stream */ - return (gzFile)state; -} - -/* -- see zlib.h -- */ -gzFile ZEXPORT gzopen(path, mode) - const char *path; - const char *mode; -{ - return gz_open(path, -1, mode); -} - -/* -- see zlib.h -- */ -gzFile ZEXPORT gzopen64(path, mode) - const char *path; - const char *mode; -{ - return gz_open(path, -1, mode); -} - -/* -- see zlib.h -- */ -gzFile ZEXPORT gzdopen(fd, mode) - int fd; - const char *mode; -{ - char *path; /* identifier for error messages */ - gzFile gz; - - if (fd == -1 || (path = (char *)malloc(7 + 3 * sizeof(int))) == NULL) - return NULL; -#if !defined(NO_snprintf) && !defined(NO_vsnprintf) - (void)snprintf(path, 7 + 3 * sizeof(int), "", fd); -#else - sprintf(path, "", fd); /* for debugging */ -#endif - gz = gz_open(path, fd, mode); - free(path); - return gz; -} - -/* -- see zlib.h -- */ -#ifdef WIDECHAR -gzFile ZEXPORT gzopen_w(path, mode) - const wchar_t *path; - const char *mode; -{ - return gz_open(path, -2, mode); -} -#endif - -/* -- see zlib.h -- */ -int ZEXPORT gzbuffer(file, size) - gzFile file; - unsigned size; -{ - gz_statep state; - - /* get internal structure and check integrity */ - if (file == NULL) - return -1; - state = (gz_statep)file; - if (state->mode != GZ_READ && state->mode != GZ_WRITE) - return -1; - - /* make sure we haven't already allocated memory */ - if (state->size != 0) - return -1; - - /* check and set requested size */ - if ((size << 1) < size) - return -1; /* need to be able to double it */ - if (size < 2) - size = 2; /* need two bytes to check magic header */ - state->want = size; - return 0; -} - -/* -- see zlib.h -- */ -int ZEXPORT gzrewind(file) - gzFile file; -{ - gz_statep state; - - /* get internal structure */ - if (file == NULL) - return -1; - state = (gz_statep)file; - - /* check that we're reading and that there's no error */ - if (state->mode != GZ_READ || - (state->err != Z_OK && state->err != Z_BUF_ERROR)) - return -1; - - /* back up and start over */ - if (LSEEK(state->fd, state->start, SEEK_SET) == -1) - return -1; - gz_reset(state); - return 0; -} - -/* -- see zlib.h -- */ -z_off64_t ZEXPORT gzseek64(file, offset, whence) - gzFile file; - z_off64_t offset; - int whence; -{ - unsigned n; - z_off64_t ret; - gz_statep state; - - /* get internal structure and check integrity */ - if (file == NULL) - return -1; - state = (gz_statep)file; - if (state->mode != GZ_READ && state->mode != GZ_WRITE) - return -1; - - /* check that there's no error */ - if (state->err != Z_OK && state->err != Z_BUF_ERROR) - return -1; - - /* can only seek from start or relative to current position */ - if (whence != SEEK_SET && whence != SEEK_CUR) - return -1; - - /* normalize offset to a SEEK_CUR specification */ - if (whence == SEEK_SET) - offset -= state->x.pos; - else if (state->seek) - offset += state->skip; - state->seek = 0; - - /* if within raw area while reading, just go there */ - if (state->mode == GZ_READ && state->how == COPY && - state->x.pos + offset >= 0) { - ret = LSEEK(state->fd, offset - state->x.have, SEEK_CUR); - if (ret == -1) - return -1; - state->x.have = 0; - state->eof = 0; - state->past = 0; - state->seek = 0; - gz_error(state, Z_OK, NULL); - state->strm.avail_in = 0; - state->x.pos += offset; - return state->x.pos; - } - - /* calculate skip amount, rewinding if needed for back seek when reading */ - if (offset < 0) { - if (state->mode != GZ_READ) /* writing -- can't go backwards */ - return -1; - offset += state->x.pos; - if (offset < 0) /* before start of file! */ - return -1; - if (gzrewind(file) == -1) /* rewind, then skip to offset */ - return -1; - } - - /* if reading, skip what's in output buffer (one less gzgetc() check) */ - if (state->mode == GZ_READ) { - n = GT_OFF(state->x.have) || (z_off64_t)state->x.have > offset ? - (unsigned)offset : state->x.have; - state->x.have -= n; - state->x.next += n; - state->x.pos += n; - offset -= n; - } - - /* request skip (if not zero) */ - if (offset) { - state->seek = 1; - state->skip = offset; - } - return state->x.pos + offset; -} - -/* -- see zlib.h -- */ -z_off_t ZEXPORT gzseek(file, offset, whence) - gzFile file; - z_off_t offset; - int whence; -{ - z_off64_t ret; - - ret = gzseek64(file, (z_off64_t)offset, whence); - return ret == (z_off_t)ret ? (z_off_t)ret : -1; -} - -/* -- see zlib.h -- */ -z_off64_t ZEXPORT gztell64(file) - gzFile file; -{ - gz_statep state; - - /* get internal structure and check integrity */ - if (file == NULL) - return -1; - state = (gz_statep)file; - if (state->mode != GZ_READ && state->mode != GZ_WRITE) - return -1; - - /* return position */ - return state->x.pos + (state->seek ? state->skip : 0); -} - -/* -- see zlib.h -- */ -z_off_t ZEXPORT gztell(file) - gzFile file; -{ - z_off64_t ret; - - ret = gztell64(file); - return ret == (z_off_t)ret ? (z_off_t)ret : -1; -} - -/* -- see zlib.h -- */ -z_off64_t ZEXPORT gzoffset64(file) - gzFile file; -{ - z_off64_t offset; - gz_statep state; - - /* get internal structure and check integrity */ - if (file == NULL) - return -1; - state = (gz_statep)file; - if (state->mode != GZ_READ && state->mode != GZ_WRITE) - return -1; - - /* compute and return effective offset in file */ - offset = LSEEK(state->fd, 0, SEEK_CUR); - if (offset == -1) - return -1; - if (state->mode == GZ_READ) /* reading */ - offset -= state->strm.avail_in; /* don't count buffered input */ - return offset; -} - -/* -- see zlib.h -- */ -z_off_t ZEXPORT gzoffset(file) - gzFile file; -{ - z_off64_t ret; - - ret = gzoffset64(file); - return ret == (z_off_t)ret ? (z_off_t)ret : -1; -} - -/* -- see zlib.h -- */ -int ZEXPORT gzeof(file) - gzFile file; -{ - gz_statep state; - - /* get internal structure and check integrity */ - if (file == NULL) - return 0; - state = (gz_statep)file; - if (state->mode != GZ_READ && state->mode != GZ_WRITE) - return 0; - - /* return end-of-file state */ - return state->mode == GZ_READ ? state->past : 0; -} - -/* -- see zlib.h -- */ -const char * ZEXPORT gzerror(file, errnum) - gzFile file; - int *errnum; -{ - gz_statep state; - - /* get internal structure and check integrity */ - if (file == NULL) - return NULL; - state = (gz_statep)file; - if (state->mode != GZ_READ && state->mode != GZ_WRITE) - return NULL; - - /* return error information */ - if (errnum != NULL) - *errnum = state->err; - return state->err == Z_MEM_ERROR ? "out of memory" : - (state->msg == NULL ? "" : state->msg); -} - -/* -- see zlib.h -- */ -void ZEXPORT gzclearerr(file) - gzFile file; -{ - gz_statep state; - - /* get internal structure and check integrity */ - if (file == NULL) - return; - state = (gz_statep)file; - if (state->mode != GZ_READ && state->mode != GZ_WRITE) - return; - - /* clear error and end-of-file */ - if (state->mode == GZ_READ) { - state->eof = 0; - state->past = 0; - } - gz_error(state, Z_OK, NULL); -} - -/* Create an error message in allocated memory and set state->err and - state->msg accordingly. Free any previous error message already there. Do - not try to free or allocate space if the error is Z_MEM_ERROR (out of - memory). Simply save the error message as a static string. If there is an - allocation failure constructing the error message, then convert the error to - out of memory. */ -void ZLIB_INTERNAL gz_error(state, err, msg) - gz_statep state; - int err; - const char *msg; -{ - /* free previously allocated message and clear */ - if (state->msg != NULL) { - if (state->err != Z_MEM_ERROR) - free(state->msg); - state->msg = NULL; - } - - /* if fatal, set state->x.have to 0 so that the gzgetc() macro fails */ - if (err != Z_OK && err != Z_BUF_ERROR) - state->x.have = 0; - - /* set error code, and if no message, then done */ - state->err = err; - if (msg == NULL) - return; - - /* for an out of memory error, return literal string when requested */ - if (err == Z_MEM_ERROR) - return; - - /* construct error message with path */ - if ((state->msg = (char *)malloc(strlen(state->path) + strlen(msg) + 3)) == - NULL) { - state->err = Z_MEM_ERROR; - return; - } -#if !defined(NO_snprintf) && !defined(NO_vsnprintf) - (void)snprintf(state->msg, strlen(state->path) + strlen(msg) + 3, - "%s%s%s", state->path, ": ", msg); -#else - strcpy(state->msg, state->path); - strcat(state->msg, ": "); - strcat(state->msg, msg); -#endif -} - -#ifndef INT_MAX -/* portably return maximum value for an int (when limits.h presumed not - available) -- we need to do this to cover cases where 2's complement not - used, since C standard permits 1's complement and sign-bit representations, - otherwise we could just use ((unsigned)-1) >> 1 */ -unsigned ZLIB_INTERNAL gz_intmax() -{ - unsigned p, q; - - p = 1; - do { - q = p; - p <<= 1; - p++; - } while (p > q); - return q >> 1; -} -#endif diff --git a/dep/zlib/src/gzread.c b/dep/zlib/src/gzread.c deleted file mode 100644 index 956b91ea7..000000000 --- a/dep/zlib/src/gzread.c +++ /dev/null @@ -1,654 +0,0 @@ -/* gzread.c -- zlib functions for reading gzip files - * Copyright (C) 2004, 2005, 2010, 2011, 2012, 2013, 2016 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -#include "gzguts.h" - -/* Local functions */ -local int gz_load OF((gz_statep, unsigned char *, unsigned, unsigned *)); -local int gz_avail OF((gz_statep)); -local int gz_look OF((gz_statep)); -local int gz_decomp OF((gz_statep)); -local int gz_fetch OF((gz_statep)); -local int gz_skip OF((gz_statep, z_off64_t)); -local z_size_t gz_read OF((gz_statep, voidp, z_size_t)); - -/* Use read() to load a buffer -- return -1 on error, otherwise 0. Read from - state->fd, and update state->eof, state->err, and state->msg as appropriate. - This function needs to loop on read(), since read() is not guaranteed to - read the number of bytes requested, depending on the type of descriptor. */ -local int gz_load(state, buf, len, have) - gz_statep state; - unsigned char *buf; - unsigned len; - unsigned *have; -{ - int ret; - unsigned get, max = ((unsigned)-1 >> 2) + 1; - - *have = 0; - do { - get = len - *have; - if (get > max) - get = max; - ret = read(state->fd, buf + *have, get); - if (ret <= 0) - break; - *have += (unsigned)ret; - } while (*have < len); - if (ret < 0) { - gz_error(state, Z_ERRNO, zstrerror()); - return -1; - } - if (ret == 0) - state->eof = 1; - return 0; -} - -/* Load up input buffer and set eof flag if last data loaded -- return -1 on - error, 0 otherwise. Note that the eof flag is set when the end of the input - file is reached, even though there may be unused data in the buffer. Once - that data has been used, no more attempts will be made to read the file. - If strm->avail_in != 0, then the current data is moved to the beginning of - the input buffer, and then the remainder of the buffer is loaded with the - available data from the input file. */ -local int gz_avail(state) - gz_statep state; -{ - unsigned got; - z_streamp strm = &(state->strm); - - if (state->err != Z_OK && state->err != Z_BUF_ERROR) - return -1; - if (state->eof == 0) { - if (strm->avail_in) { /* copy what's there to the start */ - unsigned char *p = state->in; - unsigned const char *q = strm->next_in; - unsigned n = strm->avail_in; - do { - *p++ = *q++; - } while (--n); - } - if (gz_load(state, state->in + strm->avail_in, - state->size - strm->avail_in, &got) == -1) - return -1; - strm->avail_in += got; - strm->next_in = state->in; - } - return 0; -} - -/* Look for gzip header, set up for inflate or copy. state->x.have must be 0. - If this is the first time in, allocate required memory. state->how will be - left unchanged if there is no more input data available, will be set to COPY - if there is no gzip header and direct copying will be performed, or it will - be set to GZIP for decompression. If direct copying, then leftover input - data from the input buffer will be copied to the output buffer. In that - case, all further file reads will be directly to either the output buffer or - a user buffer. If decompressing, the inflate state will be initialized. - gz_look() will return 0 on success or -1 on failure. */ -local int gz_look(state) - gz_statep state; -{ - z_streamp strm = &(state->strm); - - /* allocate read buffers and inflate memory */ - if (state->size == 0) { - /* allocate buffers */ - state->in = (unsigned char *)malloc(state->want); - state->out = (unsigned char *)malloc(state->want << 1); - if (state->in == NULL || state->out == NULL) { - free(state->out); - free(state->in); - gz_error(state, Z_MEM_ERROR, "out of memory"); - return -1; - } - state->size = state->want; - - /* allocate inflate memory */ - state->strm.zalloc = Z_NULL; - state->strm.zfree = Z_NULL; - state->strm.opaque = Z_NULL; - state->strm.avail_in = 0; - state->strm.next_in = Z_NULL; - if (inflateInit2(&(state->strm), 15 + 16) != Z_OK) { /* gunzip */ - free(state->out); - free(state->in); - state->size = 0; - gz_error(state, Z_MEM_ERROR, "out of memory"); - return -1; - } - } - - /* get at least the magic bytes in the input buffer */ - if (strm->avail_in < 2) { - if (gz_avail(state) == -1) - return -1; - if (strm->avail_in == 0) - return 0; - } - - /* look for gzip magic bytes -- if there, do gzip decoding (note: there is - a logical dilemma here when considering the case of a partially written - gzip file, to wit, if a single 31 byte is written, then we cannot tell - whether this is a single-byte file, or just a partially written gzip - file -- for here we assume that if a gzip file is being written, then - the header will be written in a single operation, so that reading a - single byte is sufficient indication that it is not a gzip file) */ - if (strm->avail_in > 1 && - strm->next_in[0] == 31 && strm->next_in[1] == 139) { - inflateReset(strm); - state->how = GZIP; - state->direct = 0; - return 0; - } - - /* no gzip header -- if we were decoding gzip before, then this is trailing - garbage. Ignore the trailing garbage and finish. */ - if (state->direct == 0) { - strm->avail_in = 0; - state->eof = 1; - state->x.have = 0; - return 0; - } - - /* doing raw i/o, copy any leftover input to output -- this assumes that - the output buffer is larger than the input buffer, which also assures - space for gzungetc() */ - state->x.next = state->out; - if (strm->avail_in) { - memcpy(state->x.next, strm->next_in, strm->avail_in); - state->x.have = strm->avail_in; - strm->avail_in = 0; - } - state->how = COPY; - state->direct = 1; - return 0; -} - -/* Decompress from input to the provided next_out and avail_out in the state. - On return, state->x.have and state->x.next point to the just decompressed - data. If the gzip stream completes, state->how is reset to LOOK to look for - the next gzip stream or raw data, once state->x.have is depleted. Returns 0 - on success, -1 on failure. */ -local int gz_decomp(state) - gz_statep state; -{ - int ret = Z_OK; - unsigned had; - z_streamp strm = &(state->strm); - - /* fill output buffer up to end of deflate stream */ - had = strm->avail_out; - do { - /* get more input for inflate() */ - if (strm->avail_in == 0 && gz_avail(state) == -1) - return -1; - if (strm->avail_in == 0) { - gz_error(state, Z_BUF_ERROR, "unexpected end of file"); - break; - } - - /* decompress and handle errors */ - ret = inflate(strm, Z_NO_FLUSH); - if (ret == Z_STREAM_ERROR || ret == Z_NEED_DICT) { - gz_error(state, Z_STREAM_ERROR, - "internal error: inflate stream corrupt"); - return -1; - } - if (ret == Z_MEM_ERROR) { - gz_error(state, Z_MEM_ERROR, "out of memory"); - return -1; - } - if (ret == Z_DATA_ERROR) { /* deflate stream invalid */ - gz_error(state, Z_DATA_ERROR, - strm->msg == NULL ? "compressed data error" : strm->msg); - return -1; - } - } while (strm->avail_out && ret != Z_STREAM_END); - - /* update available output */ - state->x.have = had - strm->avail_out; - state->x.next = strm->next_out - state->x.have; - - /* if the gzip stream completed successfully, look for another */ - if (ret == Z_STREAM_END) - state->how = LOOK; - - /* good decompression */ - return 0; -} - -/* Fetch data and put it in the output buffer. Assumes state->x.have is 0. - Data is either copied from the input file or decompressed from the input - file depending on state->how. If state->how is LOOK, then a gzip header is - looked for to determine whether to copy or decompress. Returns -1 on error, - otherwise 0. gz_fetch() will leave state->how as COPY or GZIP unless the - end of the input file has been reached and all data has been processed. */ -local int gz_fetch(state) - gz_statep state; -{ - z_streamp strm = &(state->strm); - - do { - switch(state->how) { - case LOOK: /* -> LOOK, COPY (only if never GZIP), or GZIP */ - if (gz_look(state) == -1) - return -1; - if (state->how == LOOK) - return 0; - break; - case COPY: /* -> COPY */ - if (gz_load(state, state->out, state->size << 1, &(state->x.have)) - == -1) - return -1; - state->x.next = state->out; - return 0; - case GZIP: /* -> GZIP or LOOK (if end of gzip stream) */ - strm->avail_out = state->size << 1; - strm->next_out = state->out; - if (gz_decomp(state) == -1) - return -1; - } - } while (state->x.have == 0 && (!state->eof || strm->avail_in)); - return 0; -} - -/* Skip len uncompressed bytes of output. Return -1 on error, 0 on success. */ -local int gz_skip(state, len) - gz_statep state; - z_off64_t len; -{ - unsigned n; - - /* skip over len bytes or reach end-of-file, whichever comes first */ - while (len) - /* skip over whatever is in output buffer */ - if (state->x.have) { - n = GT_OFF(state->x.have) || (z_off64_t)state->x.have > len ? - (unsigned)len : state->x.have; - state->x.have -= n; - state->x.next += n; - state->x.pos += n; - len -= n; - } - - /* output buffer empty -- return if we're at the end of the input */ - else if (state->eof && state->strm.avail_in == 0) - break; - - /* need more data to skip -- load up output buffer */ - else { - /* get more output, looking for header if required */ - if (gz_fetch(state) == -1) - return -1; - } - return 0; -} - -/* Read len bytes into buf from file, or less than len up to the end of the - input. Return the number of bytes read. If zero is returned, either the - end of file was reached, or there was an error. state->err must be - consulted in that case to determine which. */ -local z_size_t gz_read(state, buf, len) - gz_statep state; - voidp buf; - z_size_t len; -{ - z_size_t got; - unsigned n; - - /* if len is zero, avoid unnecessary operations */ - if (len == 0) - return 0; - - /* process a skip request */ - if (state->seek) { - state->seek = 0; - if (gz_skip(state, state->skip) == -1) - return 0; - } - - /* get len bytes to buf, or less than len if at the end */ - got = 0; - do { - /* set n to the maximum amount of len that fits in an unsigned int */ - n = -1; - if (n > len) - n = len; - - /* first just try copying data from the output buffer */ - if (state->x.have) { - if (state->x.have < n) - n = state->x.have; - memcpy(buf, state->x.next, n); - state->x.next += n; - state->x.have -= n; - } - - /* output buffer empty -- return if we're at the end of the input */ - else if (state->eof && state->strm.avail_in == 0) { - state->past = 1; /* tried to read past end */ - break; - } - - /* need output data -- for small len or new stream load up our output - buffer */ - else if (state->how == LOOK || n < (state->size << 1)) { - /* get more output, looking for header if required */ - if (gz_fetch(state) == -1) - return 0; - continue; /* no progress yet -- go back to copy above */ - /* the copy above assures that we will leave with space in the - output buffer, allowing at least one gzungetc() to succeed */ - } - - /* large len -- read directly into user buffer */ - else if (state->how == COPY) { /* read directly */ - if (gz_load(state, (unsigned char *)buf, n, &n) == -1) - return 0; - } - - /* large len -- decompress directly into user buffer */ - else { /* state->how == GZIP */ - state->strm.avail_out = n; - state->strm.next_out = (unsigned char *)buf; - if (gz_decomp(state) == -1) - return 0; - n = state->x.have; - state->x.have = 0; - } - - /* update progress */ - len -= n; - buf = (char *)buf + n; - got += n; - state->x.pos += n; - } while (len); - - /* return number of bytes read into user buffer */ - return got; -} - -/* -- see zlib.h -- */ -int ZEXPORT gzread(file, buf, len) - gzFile file; - voidp buf; - unsigned len; -{ - gz_statep state; - - /* get internal structure */ - if (file == NULL) - return -1; - state = (gz_statep)file; - - /* check that we're reading and that there's no (serious) error */ - if (state->mode != GZ_READ || - (state->err != Z_OK && state->err != Z_BUF_ERROR)) - return -1; - - /* since an int is returned, make sure len fits in one, otherwise return - with an error (this avoids a flaw in the interface) */ - if ((int)len < 0) { - gz_error(state, Z_STREAM_ERROR, "request does not fit in an int"); - return -1; - } - - /* read len or fewer bytes to buf */ - len = gz_read(state, buf, len); - - /* check for an error */ - if (len == 0 && state->err != Z_OK && state->err != Z_BUF_ERROR) - return -1; - - /* return the number of bytes read (this is assured to fit in an int) */ - return (int)len; -} - -/* -- see zlib.h -- */ -z_size_t ZEXPORT gzfread(buf, size, nitems, file) - voidp buf; - z_size_t size; - z_size_t nitems; - gzFile file; -{ - z_size_t len; - gz_statep state; - - /* get internal structure */ - if (file == NULL) - return 0; - state = (gz_statep)file; - - /* check that we're reading and that there's no (serious) error */ - if (state->mode != GZ_READ || - (state->err != Z_OK && state->err != Z_BUF_ERROR)) - return 0; - - /* compute bytes to read -- error on overflow */ - len = nitems * size; - if (size && len / size != nitems) { - gz_error(state, Z_STREAM_ERROR, "request does not fit in a size_t"); - return 0; - } - - /* read len or fewer bytes to buf, return the number of full items read */ - return len ? gz_read(state, buf, len) / size : 0; -} - -/* -- see zlib.h -- */ -#ifdef Z_PREFIX_SET -# undef z_gzgetc -#else -# undef gzgetc -#endif -int ZEXPORT gzgetc(file) - gzFile file; -{ - int ret; - unsigned char buf[1]; - gz_statep state; - - /* get internal structure */ - if (file == NULL) - return -1; - state = (gz_statep)file; - - /* check that we're reading and that there's no (serious) error */ - if (state->mode != GZ_READ || - (state->err != Z_OK && state->err != Z_BUF_ERROR)) - return -1; - - /* try output buffer (no need to check for skip request) */ - if (state->x.have) { - state->x.have--; - state->x.pos++; - return *(state->x.next)++; - } - - /* nothing there -- try gz_read() */ - ret = gz_read(state, buf, 1); - return ret < 1 ? -1 : buf[0]; -} - -int ZEXPORT gzgetc_(file) -gzFile file; -{ - return gzgetc(file); -} - -/* -- see zlib.h -- */ -int ZEXPORT gzungetc(c, file) - int c; - gzFile file; -{ - gz_statep state; - - /* get internal structure */ - if (file == NULL) - return -1; - state = (gz_statep)file; - - /* check that we're reading and that there's no (serious) error */ - if (state->mode != GZ_READ || - (state->err != Z_OK && state->err != Z_BUF_ERROR)) - return -1; - - /* process a skip request */ - if (state->seek) { - state->seek = 0; - if (gz_skip(state, state->skip) == -1) - return -1; - } - - /* can't push EOF */ - if (c < 0) - return -1; - - /* if output buffer empty, put byte at end (allows more pushing) */ - if (state->x.have == 0) { - state->x.have = 1; - state->x.next = state->out + (state->size << 1) - 1; - state->x.next[0] = (unsigned char)c; - state->x.pos--; - state->past = 0; - return c; - } - - /* if no room, give up (must have already done a gzungetc()) */ - if (state->x.have == (state->size << 1)) { - gz_error(state, Z_DATA_ERROR, "out of room to push characters"); - return -1; - } - - /* slide output data if needed and insert byte before existing data */ - if (state->x.next == state->out) { - unsigned char *src = state->out + state->x.have; - unsigned char *dest = state->out + (state->size << 1); - while (src > state->out) - *--dest = *--src; - state->x.next = dest; - } - state->x.have++; - state->x.next--; - state->x.next[0] = (unsigned char)c; - state->x.pos--; - state->past = 0; - return c; -} - -/* -- see zlib.h -- */ -char * ZEXPORT gzgets(file, buf, len) - gzFile file; - char *buf; - int len; -{ - unsigned left, n; - char *str; - unsigned char *eol; - gz_statep state; - - /* check parameters and get internal structure */ - if (file == NULL || buf == NULL || len < 1) - return NULL; - state = (gz_statep)file; - - /* check that we're reading and that there's no (serious) error */ - if (state->mode != GZ_READ || - (state->err != Z_OK && state->err != Z_BUF_ERROR)) - return NULL; - - /* process a skip request */ - if (state->seek) { - state->seek = 0; - if (gz_skip(state, state->skip) == -1) - return NULL; - } - - /* copy output bytes up to new line or len - 1, whichever comes first -- - append a terminating zero to the string (we don't check for a zero in - the contents, let the user worry about that) */ - str = buf; - left = (unsigned)len - 1; - if (left) do { - /* assure that something is in the output buffer */ - if (state->x.have == 0 && gz_fetch(state) == -1) - return NULL; /* error */ - if (state->x.have == 0) { /* end of file */ - state->past = 1; /* read past end */ - break; /* return what we have */ - } - - /* look for end-of-line in current output buffer */ - n = state->x.have > left ? left : state->x.have; - eol = (unsigned char *)memchr(state->x.next, '\n', n); - if (eol != NULL) - n = (unsigned)(eol - state->x.next) + 1; - - /* copy through end-of-line, or remainder if not found */ - memcpy(buf, state->x.next, n); - state->x.have -= n; - state->x.next += n; - state->x.pos += n; - left -= n; - buf += n; - } while (left && eol == NULL); - - /* return terminated string, or if nothing, end of file */ - if (buf == str) - return NULL; - buf[0] = 0; - return str; -} - -/* -- see zlib.h -- */ -int ZEXPORT gzdirect(file) - gzFile file; -{ - gz_statep state; - - /* get internal structure */ - if (file == NULL) - return 0; - state = (gz_statep)file; - - /* if the state is not known, but we can find out, then do so (this is - mainly for right after a gzopen() or gzdopen()) */ - if (state->mode == GZ_READ && state->how == LOOK && state->x.have == 0) - (void)gz_look(state); - - /* return 1 if transparent, 0 if processing a gzip stream */ - return state->direct; -} - -/* -- see zlib.h -- */ -int ZEXPORT gzclose_r(file) - gzFile file; -{ - int ret, err; - gz_statep state; - - /* get internal structure */ - if (file == NULL) - return Z_STREAM_ERROR; - state = (gz_statep)file; - - /* check that we're reading */ - if (state->mode != GZ_READ) - return Z_STREAM_ERROR; - - /* free memory and close file */ - if (state->size) { - inflateEnd(&(state->strm)); - free(state->out); - free(state->in); - } - err = state->err == Z_BUF_ERROR ? Z_BUF_ERROR : Z_OK; - gz_error(state, Z_OK, NULL); - free(state->path); - ret = close(state->fd); - free(state); - return ret ? Z_ERRNO : err; -} diff --git a/dep/zlib/src/gzwrite.c b/dep/zlib/src/gzwrite.c deleted file mode 100644 index c7b5651d7..000000000 --- a/dep/zlib/src/gzwrite.c +++ /dev/null @@ -1,665 +0,0 @@ -/* gzwrite.c -- zlib functions for writing gzip files - * Copyright (C) 2004-2017 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -#include "gzguts.h" - -/* Local functions */ -local int gz_init OF((gz_statep)); -local int gz_comp OF((gz_statep, int)); -local int gz_zero OF((gz_statep, z_off64_t)); -local z_size_t gz_write OF((gz_statep, voidpc, z_size_t)); - -/* Initialize state for writing a gzip file. Mark initialization by setting - state->size to non-zero. Return -1 on a memory allocation failure, or 0 on - success. */ -local int gz_init(state) - gz_statep state; -{ - int ret; - z_streamp strm = &(state->strm); - - /* allocate input buffer (double size for gzprintf) */ - state->in = (unsigned char *)malloc(state->want << 1); - if (state->in == NULL) { - gz_error(state, Z_MEM_ERROR, "out of memory"); - return -1; - } - - /* only need output buffer and deflate state if compressing */ - if (!state->direct) { - /* allocate output buffer */ - state->out = (unsigned char *)malloc(state->want); - if (state->out == NULL) { - free(state->in); - gz_error(state, Z_MEM_ERROR, "out of memory"); - return -1; - } - - /* allocate deflate memory, set up for gzip compression */ - strm->zalloc = Z_NULL; - strm->zfree = Z_NULL; - strm->opaque = Z_NULL; - ret = deflateInit2(strm, state->level, Z_DEFLATED, - MAX_WBITS + 16, DEF_MEM_LEVEL, state->strategy); - if (ret != Z_OK) { - free(state->out); - free(state->in); - gz_error(state, Z_MEM_ERROR, "out of memory"); - return -1; - } - strm->next_in = NULL; - } - - /* mark state as initialized */ - state->size = state->want; - - /* initialize write buffer if compressing */ - if (!state->direct) { - strm->avail_out = state->size; - strm->next_out = state->out; - state->x.next = strm->next_out; - } - return 0; -} - -/* Compress whatever is at avail_in and next_in and write to the output file. - Return -1 if there is an error writing to the output file or if gz_init() - fails to allocate memory, otherwise 0. flush is assumed to be a valid - deflate() flush value. If flush is Z_FINISH, then the deflate() state is - reset to start a new gzip stream. If gz->direct is true, then simply write - to the output file without compressing, and ignore flush. */ -local int gz_comp(state, flush) - gz_statep state; - int flush; -{ - int ret, writ; - unsigned have, put, max = ((unsigned)-1 >> 2) + 1; - z_streamp strm = &(state->strm); - - /* allocate memory if this is the first time through */ - if (state->size == 0 && gz_init(state) == -1) - return -1; - - /* write directly if requested */ - if (state->direct) { - while (strm->avail_in) { - put = strm->avail_in > max ? max : strm->avail_in; - writ = write(state->fd, strm->next_in, put); - if (writ < 0) { - gz_error(state, Z_ERRNO, zstrerror()); - return -1; - } - strm->avail_in -= (unsigned)writ; - strm->next_in += writ; - } - return 0; - } - - /* run deflate() on provided input until it produces no more output */ - ret = Z_OK; - do { - /* write out current buffer contents if full, or if flushing, but if - doing Z_FINISH then don't write until we get to Z_STREAM_END */ - if (strm->avail_out == 0 || (flush != Z_NO_FLUSH && - (flush != Z_FINISH || ret == Z_STREAM_END))) { - while (strm->next_out > state->x.next) { - put = strm->next_out - state->x.next > (int)max ? max : - (unsigned)(strm->next_out - state->x.next); - writ = write(state->fd, state->x.next, put); - if (writ < 0) { - gz_error(state, Z_ERRNO, zstrerror()); - return -1; - } - state->x.next += writ; - } - if (strm->avail_out == 0) { - strm->avail_out = state->size; - strm->next_out = state->out; - state->x.next = state->out; - } - } - - /* compress */ - have = strm->avail_out; - ret = deflate(strm, flush); - if (ret == Z_STREAM_ERROR) { - gz_error(state, Z_STREAM_ERROR, - "internal error: deflate stream corrupt"); - return -1; - } - have -= strm->avail_out; - } while (have); - - /* if that completed a deflate stream, allow another to start */ - if (flush == Z_FINISH) - deflateReset(strm); - - /* all done, no errors */ - return 0; -} - -/* Compress len zeros to output. Return -1 on a write error or memory - allocation failure by gz_comp(), or 0 on success. */ -local int gz_zero(state, len) - gz_statep state; - z_off64_t len; -{ - int first; - unsigned n; - z_streamp strm = &(state->strm); - - /* consume whatever's left in the input buffer */ - if (strm->avail_in && gz_comp(state, Z_NO_FLUSH) == -1) - return -1; - - /* compress len zeros (len guaranteed > 0) */ - first = 1; - while (len) { - n = GT_OFF(state->size) || (z_off64_t)state->size > len ? - (unsigned)len : state->size; - if (first) { - memset(state->in, 0, n); - first = 0; - } - strm->avail_in = n; - strm->next_in = state->in; - state->x.pos += n; - if (gz_comp(state, Z_NO_FLUSH) == -1) - return -1; - len -= n; - } - return 0; -} - -/* Write len bytes from buf to file. Return the number of bytes written. If - the returned value is less than len, then there was an error. */ -local z_size_t gz_write(state, buf, len) - gz_statep state; - voidpc buf; - z_size_t len; -{ - z_size_t put = len; - - /* if len is zero, avoid unnecessary operations */ - if (len == 0) - return 0; - - /* allocate memory if this is the first time through */ - if (state->size == 0 && gz_init(state) == -1) - return 0; - - /* check for seek request */ - if (state->seek) { - state->seek = 0; - if (gz_zero(state, state->skip) == -1) - return 0; - } - - /* for small len, copy to input buffer, otherwise compress directly */ - if (len < state->size) { - /* copy to input buffer, compress when full */ - do { - unsigned have, copy; - - if (state->strm.avail_in == 0) - state->strm.next_in = state->in; - have = (unsigned)((state->strm.next_in + state->strm.avail_in) - - state->in); - copy = state->size - have; - if (copy > len) - copy = len; - memcpy(state->in + have, buf, copy); - state->strm.avail_in += copy; - state->x.pos += copy; - buf = (const char *)buf + copy; - len -= copy; - if (len && gz_comp(state, Z_NO_FLUSH) == -1) - return 0; - } while (len); - } - else { - /* consume whatever's left in the input buffer */ - if (state->strm.avail_in && gz_comp(state, Z_NO_FLUSH) == -1) - return 0; - - /* directly compress user buffer to file */ - state->strm.next_in = (z_const Bytef *)buf; - do { - unsigned n = (unsigned)-1; - if (n > len) - n = len; - state->strm.avail_in = n; - state->x.pos += n; - if (gz_comp(state, Z_NO_FLUSH) == -1) - return 0; - len -= n; - } while (len); - } - - /* input was all buffered or compressed */ - return put; -} - -/* -- see zlib.h -- */ -int ZEXPORT gzwrite(file, buf, len) - gzFile file; - voidpc buf; - unsigned len; -{ - gz_statep state; - - /* get internal structure */ - if (file == NULL) - return 0; - state = (gz_statep)file; - - /* check that we're writing and that there's no error */ - if (state->mode != GZ_WRITE || state->err != Z_OK) - return 0; - - /* since an int is returned, make sure len fits in one, otherwise return - with an error (this avoids a flaw in the interface) */ - if ((int)len < 0) { - gz_error(state, Z_DATA_ERROR, "requested length does not fit in int"); - return 0; - } - - /* write len bytes from buf (the return value will fit in an int) */ - return (int)gz_write(state, buf, len); -} - -/* -- see zlib.h -- */ -z_size_t ZEXPORT gzfwrite(buf, size, nitems, file) - voidpc buf; - z_size_t size; - z_size_t nitems; - gzFile file; -{ - z_size_t len; - gz_statep state; - - /* get internal structure */ - if (file == NULL) - return 0; - state = (gz_statep)file; - - /* check that we're writing and that there's no error */ - if (state->mode != GZ_WRITE || state->err != Z_OK) - return 0; - - /* compute bytes to read -- error on overflow */ - len = nitems * size; - if (size && len / size != nitems) { - gz_error(state, Z_STREAM_ERROR, "request does not fit in a size_t"); - return 0; - } - - /* write len bytes to buf, return the number of full items written */ - return len ? gz_write(state, buf, len) / size : 0; -} - -/* -- see zlib.h -- */ -int ZEXPORT gzputc(file, c) - gzFile file; - int c; -{ - unsigned have; - unsigned char buf[1]; - gz_statep state; - z_streamp strm; - - /* get internal structure */ - if (file == NULL) - return -1; - state = (gz_statep)file; - strm = &(state->strm); - - /* check that we're writing and that there's no error */ - if (state->mode != GZ_WRITE || state->err != Z_OK) - return -1; - - /* check for seek request */ - if (state->seek) { - state->seek = 0; - if (gz_zero(state, state->skip) == -1) - return -1; - } - - /* try writing to input buffer for speed (state->size == 0 if buffer not - initialized) */ - if (state->size) { - if (strm->avail_in == 0) - strm->next_in = state->in; - have = (unsigned)((strm->next_in + strm->avail_in) - state->in); - if (have < state->size) { - state->in[have] = (unsigned char)c; - strm->avail_in++; - state->x.pos++; - return c & 0xff; - } - } - - /* no room in buffer or not initialized, use gz_write() */ - buf[0] = (unsigned char)c; - if (gz_write(state, buf, 1) != 1) - return -1; - return c & 0xff; -} - -/* -- see zlib.h -- */ -int ZEXPORT gzputs(file, str) - gzFile file; - const char *str; -{ - int ret; - z_size_t len; - gz_statep state; - - /* get internal structure */ - if (file == NULL) - return -1; - state = (gz_statep)file; - - /* check that we're writing and that there's no error */ - if (state->mode != GZ_WRITE || state->err != Z_OK) - return -1; - - /* write string */ - len = strlen(str); - ret = gz_write(state, str, len); - return ret == 0 && len != 0 ? -1 : ret; -} - -#if defined(STDC) || defined(Z_HAVE_STDARG_H) -#include - -/* -- see zlib.h -- */ -int ZEXPORTVA gzvprintf(gzFile file, const char *format, va_list va) -{ - int len; - unsigned left; - char *next; - gz_statep state; - z_streamp strm; - - /* get internal structure */ - if (file == NULL) - return Z_STREAM_ERROR; - state = (gz_statep)file; - strm = &(state->strm); - - /* check that we're writing and that there's no error */ - if (state->mode != GZ_WRITE || state->err != Z_OK) - return Z_STREAM_ERROR; - - /* make sure we have some buffer space */ - if (state->size == 0 && gz_init(state) == -1) - return state->err; - - /* check for seek request */ - if (state->seek) { - state->seek = 0; - if (gz_zero(state, state->skip) == -1) - return state->err; - } - - /* do the printf() into the input buffer, put length in len -- the input - buffer is double-sized just for this function, so there is guaranteed to - be state->size bytes available after the current contents */ - if (strm->avail_in == 0) - strm->next_in = state->in; - next = (char *)(state->in + (strm->next_in - state->in) + strm->avail_in); - next[state->size - 1] = 0; -#ifdef NO_vsnprintf -# ifdef HAS_vsprintf_void - (void)vsprintf(next, format, va); - for (len = 0; len < state->size; len++) - if (next[len] == 0) break; -# else - len = vsprintf(next, format, va); -# endif -#else -# ifdef HAS_vsnprintf_void - (void)vsnprintf(next, state->size, format, va); - len = strlen(next); -# else - len = vsnprintf(next, state->size, format, va); -# endif -#endif - - /* check that printf() results fit in buffer */ - if (len == 0 || (unsigned)len >= state->size || next[state->size - 1] != 0) - return 0; - - /* update buffer and position, compress first half if past that */ - strm->avail_in += (unsigned)len; - state->x.pos += len; - if (strm->avail_in >= state->size) { - left = strm->avail_in - state->size; - strm->avail_in = state->size; - if (gz_comp(state, Z_NO_FLUSH) == -1) - return state->err; - memcpy(state->in, state->in + state->size, left); - strm->next_in = state->in; - strm->avail_in = left; - } - return len; -} - -int ZEXPORTVA gzprintf(gzFile file, const char *format, ...) -{ - va_list va; - int ret; - - va_start(va, format); - ret = gzvprintf(file, format, va); - va_end(va); - return ret; -} - -#else /* !STDC && !Z_HAVE_STDARG_H */ - -/* -- see zlib.h -- */ -int ZEXPORTVA gzprintf (file, format, a1, a2, a3, a4, a5, a6, a7, a8, a9, a10, - a11, a12, a13, a14, a15, a16, a17, a18, a19, a20) - gzFile file; - const char *format; - int a1, a2, a3, a4, a5, a6, a7, a8, a9, a10, - a11, a12, a13, a14, a15, a16, a17, a18, a19, a20; -{ - unsigned len, left; - char *next; - gz_statep state; - z_streamp strm; - - /* get internal structure */ - if (file == NULL) - return Z_STREAM_ERROR; - state = (gz_statep)file; - strm = &(state->strm); - - /* check that can really pass pointer in ints */ - if (sizeof(int) != sizeof(void *)) - return Z_STREAM_ERROR; - - /* check that we're writing and that there's no error */ - if (state->mode != GZ_WRITE || state->err != Z_OK) - return Z_STREAM_ERROR; - - /* make sure we have some buffer space */ - if (state->size == 0 && gz_init(state) == -1) - return state->error; - - /* check for seek request */ - if (state->seek) { - state->seek = 0; - if (gz_zero(state, state->skip) == -1) - return state->error; - } - - /* do the printf() into the input buffer, put length in len -- the input - buffer is double-sized just for this function, so there is guaranteed to - be state->size bytes available after the current contents */ - if (strm->avail_in == 0) - strm->next_in = state->in; - next = (char *)(strm->next_in + strm->avail_in); - next[state->size - 1] = 0; -#ifdef NO_snprintf -# ifdef HAS_sprintf_void - sprintf(next, format, a1, a2, a3, a4, a5, a6, a7, a8, a9, a10, a11, a12, - a13, a14, a15, a16, a17, a18, a19, a20); - for (len = 0; len < size; len++) - if (next[len] == 0) - break; -# else - len = sprintf(next, format, a1, a2, a3, a4, a5, a6, a7, a8, a9, a10, a11, - a12, a13, a14, a15, a16, a17, a18, a19, a20); -# endif -#else -# ifdef HAS_snprintf_void - snprintf(next, state->size, format, a1, a2, a3, a4, a5, a6, a7, a8, a9, - a10, a11, a12, a13, a14, a15, a16, a17, a18, a19, a20); - len = strlen(next); -# else - len = snprintf(next, state->size, format, a1, a2, a3, a4, a5, a6, a7, a8, - a9, a10, a11, a12, a13, a14, a15, a16, a17, a18, a19, a20); -# endif -#endif - - /* check that printf() results fit in buffer */ - if (len == 0 || len >= state->size || next[state->size - 1] != 0) - return 0; - - /* update buffer and position, compress first half if past that */ - strm->avail_in += len; - state->x.pos += len; - if (strm->avail_in >= state->size) { - left = strm->avail_in - state->size; - strm->avail_in = state->size; - if (gz_comp(state, Z_NO_FLUSH) == -1) - return state->err; - memcpy(state->in, state->in + state->size, left); - strm->next_in = state->in; - strm->avail_in = left; - } - return (int)len; -} - -#endif - -/* -- see zlib.h -- */ -int ZEXPORT gzflush(file, flush) - gzFile file; - int flush; -{ - gz_statep state; - - /* get internal structure */ - if (file == NULL) - return Z_STREAM_ERROR; - state = (gz_statep)file; - - /* check that we're writing and that there's no error */ - if (state->mode != GZ_WRITE || state->err != Z_OK) - return Z_STREAM_ERROR; - - /* check flush parameter */ - if (flush < 0 || flush > Z_FINISH) - return Z_STREAM_ERROR; - - /* check for seek request */ - if (state->seek) { - state->seek = 0; - if (gz_zero(state, state->skip) == -1) - return state->err; - } - - /* compress remaining data with requested flush */ - (void)gz_comp(state, flush); - return state->err; -} - -/* -- see zlib.h -- */ -int ZEXPORT gzsetparams(file, level, strategy) - gzFile file; - int level; - int strategy; -{ - gz_statep state; - z_streamp strm; - - /* get internal structure */ - if (file == NULL) - return Z_STREAM_ERROR; - state = (gz_statep)file; - strm = &(state->strm); - - /* check that we're writing and that there's no error */ - if (state->mode != GZ_WRITE || state->err != Z_OK) - return Z_STREAM_ERROR; - - /* if no change is requested, then do nothing */ - if (level == state->level && strategy == state->strategy) - return Z_OK; - - /* check for seek request */ - if (state->seek) { - state->seek = 0; - if (gz_zero(state, state->skip) == -1) - return state->err; - } - - /* change compression parameters for subsequent input */ - if (state->size) { - /* flush previous input with previous parameters before changing */ - if (strm->avail_in && gz_comp(state, Z_BLOCK) == -1) - return state->err; - deflateParams(strm, level, strategy); - } - state->level = level; - state->strategy = strategy; - return Z_OK; -} - -/* -- see zlib.h -- */ -int ZEXPORT gzclose_w(file) - gzFile file; -{ - int ret = Z_OK; - gz_statep state; - - /* get internal structure */ - if (file == NULL) - return Z_STREAM_ERROR; - state = (gz_statep)file; - - /* check that we're writing */ - if (state->mode != GZ_WRITE) - return Z_STREAM_ERROR; - - /* check for seek request */ - if (state->seek) { - state->seek = 0; - if (gz_zero(state, state->skip) == -1) - ret = state->err; - } - - /* flush, free memory, and close file */ - if (gz_comp(state, Z_FINISH) == -1) - ret = state->err; - if (state->size) { - if (!state->direct) { - (void)deflateEnd(&(state->strm)); - free(state->out); - } - free(state->in); - } - gz_error(state, Z_OK, NULL); - free(state->path); - if (close(state->fd) == -1) - ret = Z_ERRNO; - free(state); - return ret; -} diff --git a/dep/zlib/src/infback.c b/dep/zlib/src/infback.c deleted file mode 100644 index 59679ecbf..000000000 --- a/dep/zlib/src/infback.c +++ /dev/null @@ -1,640 +0,0 @@ -/* infback.c -- inflate using a call-back interface - * Copyright (C) 1995-2016 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* - This code is largely copied from inflate.c. Normally either infback.o or - inflate.o would be linked into an application--not both. The interface - with inffast.c is retained so that optimized assembler-coded versions of - inflate_fast() can be used with either inflate.c or infback.c. - */ - -#include "zutil.h" -#include "inftrees.h" -#include "inflate.h" -#include "inffast.h" - -/* function prototypes */ -local void fixedtables OF((struct inflate_state FAR *state)); - -/* - strm provides memory allocation functions in zalloc and zfree, or - Z_NULL to use the library memory allocation functions. - - windowBits is in the range 8..15, and window is a user-supplied - window and output buffer that is 2**windowBits bytes. - */ -int ZEXPORT inflateBackInit_(strm, windowBits, window, version, stream_size) -z_streamp strm; -int windowBits; -unsigned char FAR *window; -const char *version; -int stream_size; -{ - struct inflate_state FAR *state; - - if (version == Z_NULL || version[0] != ZLIB_VERSION[0] || - stream_size != (int)(sizeof(z_stream))) - return Z_VERSION_ERROR; - if (strm == Z_NULL || window == Z_NULL || - windowBits < 8 || windowBits > 15) - return Z_STREAM_ERROR; - strm->msg = Z_NULL; /* in case we return an error */ - if (strm->zalloc == (alloc_func)0) { -#ifdef Z_SOLO - return Z_STREAM_ERROR; -#else - strm->zalloc = zcalloc; - strm->opaque = (voidpf)0; -#endif - } - if (strm->zfree == (free_func)0) -#ifdef Z_SOLO - return Z_STREAM_ERROR; -#else - strm->zfree = zcfree; -#endif - state = (struct inflate_state FAR *)ZALLOC(strm, 1, - sizeof(struct inflate_state)); - if (state == Z_NULL) return Z_MEM_ERROR; - Tracev((stderr, "inflate: allocated\n")); - strm->state = (struct internal_state FAR *)state; - state->dmax = 32768U; - state->wbits = (uInt)windowBits; - state->wsize = 1U << windowBits; - state->window = window; - state->wnext = 0; - state->whave = 0; - return Z_OK; -} - -/* - Return state with length and distance decoding tables and index sizes set to - fixed code decoding. Normally this returns fixed tables from inffixed.h. - If BUILDFIXED is defined, then instead this routine builds the tables the - first time it's called, and returns those tables the first time and - thereafter. This reduces the size of the code by about 2K bytes, in - exchange for a little execution time. However, BUILDFIXED should not be - used for threaded applications, since the rewriting of the tables and virgin - may not be thread-safe. - */ -local void fixedtables(state) -struct inflate_state FAR *state; -{ -#ifdef BUILDFIXED - static int virgin = 1; - static code *lenfix, *distfix; - static code fixed[544]; - - /* build fixed huffman tables if first call (may not be thread safe) */ - if (virgin) { - unsigned sym, bits; - static code *next; - - /* literal/length table */ - sym = 0; - while (sym < 144) state->lens[sym++] = 8; - while (sym < 256) state->lens[sym++] = 9; - while (sym < 280) state->lens[sym++] = 7; - while (sym < 288) state->lens[sym++] = 8; - next = fixed; - lenfix = next; - bits = 9; - inflate_table(LENS, state->lens, 288, &(next), &(bits), state->work); - - /* distance table */ - sym = 0; - while (sym < 32) state->lens[sym++] = 5; - distfix = next; - bits = 5; - inflate_table(DISTS, state->lens, 32, &(next), &(bits), state->work); - - /* do this just once */ - virgin = 0; - } -#else /* !BUILDFIXED */ -# include "inffixed.h" -#endif /* BUILDFIXED */ - state->lencode = lenfix; - state->lenbits = 9; - state->distcode = distfix; - state->distbits = 5; -} - -/* Macros for inflateBack(): */ - -/* Load returned state from inflate_fast() */ -#define LOAD() \ - do { \ - put = strm->next_out; \ - left = strm->avail_out; \ - next = strm->next_in; \ - have = strm->avail_in; \ - hold = state->hold; \ - bits = state->bits; \ - } while (0) - -/* Set state from registers for inflate_fast() */ -#define RESTORE() \ - do { \ - strm->next_out = put; \ - strm->avail_out = left; \ - strm->next_in = next; \ - strm->avail_in = have; \ - state->hold = hold; \ - state->bits = bits; \ - } while (0) - -/* Clear the input bit accumulator */ -#define INITBITS() \ - do { \ - hold = 0; \ - bits = 0; \ - } while (0) - -/* Assure that some input is available. If input is requested, but denied, - then return a Z_BUF_ERROR from inflateBack(). */ -#define PULL() \ - do { \ - if (have == 0) { \ - have = in(in_desc, &next); \ - if (have == 0) { \ - next = Z_NULL; \ - ret = Z_BUF_ERROR; \ - goto inf_leave; \ - } \ - } \ - } while (0) - -/* Get a byte of input into the bit accumulator, or return from inflateBack() - with an error if there is no input available. */ -#define PULLBYTE() \ - do { \ - PULL(); \ - have--; \ - hold += (unsigned long)(*next++) << bits; \ - bits += 8; \ - } while (0) - -/* Assure that there are at least n bits in the bit accumulator. If there is - not enough available input to do that, then return from inflateBack() with - an error. */ -#define NEEDBITS(n) \ - do { \ - while (bits < (unsigned)(n)) \ - PULLBYTE(); \ - } while (0) - -/* Return the low n bits of the bit accumulator (n < 16) */ -#define BITS(n) \ - ((unsigned)hold & ((1U << (n)) - 1)) - -/* Remove n bits from the bit accumulator */ -#define DROPBITS(n) \ - do { \ - hold >>= (n); \ - bits -= (unsigned)(n); \ - } while (0) - -/* Remove zero to seven bits as needed to go to a byte boundary */ -#define BYTEBITS() \ - do { \ - hold >>= bits & 7; \ - bits -= bits & 7; \ - } while (0) - -/* Assure that some output space is available, by writing out the window - if it's full. If the write fails, return from inflateBack() with a - Z_BUF_ERROR. */ -#define ROOM() \ - do { \ - if (left == 0) { \ - put = state->window; \ - left = state->wsize; \ - state->whave = left; \ - if (out(out_desc, put, left)) { \ - ret = Z_BUF_ERROR; \ - goto inf_leave; \ - } \ - } \ - } while (0) - -/* - strm provides the memory allocation functions and window buffer on input, - and provides information on the unused input on return. For Z_DATA_ERROR - returns, strm will also provide an error message. - - in() and out() are the call-back input and output functions. When - inflateBack() needs more input, it calls in(). When inflateBack() has - filled the window with output, or when it completes with data in the - window, it calls out() to write out the data. The application must not - change the provided input until in() is called again or inflateBack() - returns. The application must not change the window/output buffer until - inflateBack() returns. - - in() and out() are called with a descriptor parameter provided in the - inflateBack() call. This parameter can be a structure that provides the - information required to do the read or write, as well as accumulated - information on the input and output such as totals and check values. - - in() should return zero on failure. out() should return non-zero on - failure. If either in() or out() fails, than inflateBack() returns a - Z_BUF_ERROR. strm->next_in can be checked for Z_NULL to see whether it - was in() or out() that caused in the error. Otherwise, inflateBack() - returns Z_STREAM_END on success, Z_DATA_ERROR for an deflate format - error, or Z_MEM_ERROR if it could not allocate memory for the state. - inflateBack() can also return Z_STREAM_ERROR if the input parameters - are not correct, i.e. strm is Z_NULL or the state was not initialized. - */ -int ZEXPORT inflateBack(strm, in, in_desc, out, out_desc) -z_streamp strm; -in_func in; -void FAR *in_desc; -out_func out; -void FAR *out_desc; -{ - struct inflate_state FAR *state; - z_const unsigned char FAR *next; /* next input */ - unsigned char FAR *put; /* next output */ - unsigned have, left; /* available input and output */ - unsigned long hold; /* bit buffer */ - unsigned bits; /* bits in bit buffer */ - unsigned copy; /* number of stored or match bytes to copy */ - unsigned char FAR *from; /* where to copy match bytes from */ - code here; /* current decoding table entry */ - code last; /* parent table entry */ - unsigned len; /* length to copy for repeats, bits to drop */ - int ret; /* return code */ - static const unsigned short order[19] = /* permutation of code lengths */ - {16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15}; - - /* Check that the strm exists and that the state was initialized */ - if (strm == Z_NULL || strm->state == Z_NULL) - return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; - - /* Reset the state */ - strm->msg = Z_NULL; - state->mode = TYPE; - state->last = 0; - state->whave = 0; - next = strm->next_in; - have = next != Z_NULL ? strm->avail_in : 0; - hold = 0; - bits = 0; - put = state->window; - left = state->wsize; - - /* Inflate until end of block marked as last */ - for (;;) - switch (state->mode) { - case TYPE: - /* determine and dispatch block type */ - if (state->last) { - BYTEBITS(); - state->mode = DONE; - break; - } - NEEDBITS(3); - state->last = BITS(1); - DROPBITS(1); - switch (BITS(2)) { - case 0: /* stored block */ - Tracev((stderr, "inflate: stored block%s\n", - state->last ? " (last)" : "")); - state->mode = STORED; - break; - case 1: /* fixed block */ - fixedtables(state); - Tracev((stderr, "inflate: fixed codes block%s\n", - state->last ? " (last)" : "")); - state->mode = LEN; /* decode codes */ - break; - case 2: /* dynamic block */ - Tracev((stderr, "inflate: dynamic codes block%s\n", - state->last ? " (last)" : "")); - state->mode = TABLE; - break; - case 3: - strm->msg = (char *)"invalid block type"; - state->mode = BAD; - } - DROPBITS(2); - break; - - case STORED: - /* get and verify stored block length */ - BYTEBITS(); /* go to byte boundary */ - NEEDBITS(32); - if ((hold & 0xffff) != ((hold >> 16) ^ 0xffff)) { - strm->msg = (char *)"invalid stored block lengths"; - state->mode = BAD; - break; - } - state->length = (unsigned)hold & 0xffff; - Tracev((stderr, "inflate: stored length %u\n", - state->length)); - INITBITS(); - - /* copy stored block from input to output */ - while (state->length != 0) { - copy = state->length; - PULL(); - ROOM(); - if (copy > have) copy = have; - if (copy > left) copy = left; - zmemcpy(put, next, copy); - have -= copy; - next += copy; - left -= copy; - put += copy; - state->length -= copy; - } - Tracev((stderr, "inflate: stored end\n")); - state->mode = TYPE; - break; - - case TABLE: - /* get dynamic table entries descriptor */ - NEEDBITS(14); - state->nlen = BITS(5) + 257; - DROPBITS(5); - state->ndist = BITS(5) + 1; - DROPBITS(5); - state->ncode = BITS(4) + 4; - DROPBITS(4); -#ifndef PKZIP_BUG_WORKAROUND - if (state->nlen > 286 || state->ndist > 30) { - strm->msg = (char *)"too many length or distance symbols"; - state->mode = BAD; - break; - } -#endif - Tracev((stderr, "inflate: table sizes ok\n")); - - /* get code length code lengths (not a typo) */ - state->have = 0; - while (state->have < state->ncode) { - NEEDBITS(3); - state->lens[order[state->have++]] = (unsigned short)BITS(3); - DROPBITS(3); - } - while (state->have < 19) - state->lens[order[state->have++]] = 0; - state->next = state->codes; - state->lencode = (code const FAR *)(state->next); - state->lenbits = 7; - ret = inflate_table(CODES, state->lens, 19, &(state->next), - &(state->lenbits), state->work); - if (ret) { - strm->msg = (char *)"invalid code lengths set"; - state->mode = BAD; - break; - } - Tracev((stderr, "inflate: code lengths ok\n")); - - /* get length and distance code code lengths */ - state->have = 0; - while (state->have < state->nlen + state->ndist) { - for (;;) { - here = state->lencode[BITS(state->lenbits)]; - if ((unsigned)(here.bits) <= bits) break; - PULLBYTE(); - } - if (here.val < 16) { - DROPBITS(here.bits); - state->lens[state->have++] = here.val; - } - else { - if (here.val == 16) { - NEEDBITS(here.bits + 2); - DROPBITS(here.bits); - if (state->have == 0) { - strm->msg = (char *)"invalid bit length repeat"; - state->mode = BAD; - break; - } - len = (unsigned)(state->lens[state->have - 1]); - copy = 3 + BITS(2); - DROPBITS(2); - } - else if (here.val == 17) { - NEEDBITS(here.bits + 3); - DROPBITS(here.bits); - len = 0; - copy = 3 + BITS(3); - DROPBITS(3); - } - else { - NEEDBITS(here.bits + 7); - DROPBITS(here.bits); - len = 0; - copy = 11 + BITS(7); - DROPBITS(7); - } - if (state->have + copy > state->nlen + state->ndist) { - strm->msg = (char *)"invalid bit length repeat"; - state->mode = BAD; - break; - } - while (copy--) - state->lens[state->have++] = (unsigned short)len; - } - } - - /* handle error breaks in while */ - if (state->mode == BAD) break; - - /* check for end-of-block code (better have one) */ - if (state->lens[256] == 0) { - strm->msg = (char *)"invalid code -- missing end-of-block"; - state->mode = BAD; - break; - } - - /* build code tables -- note: do not change the lenbits or distbits - values here (9 and 6) without reading the comments in inftrees.h - concerning the ENOUGH constants, which depend on those values */ - state->next = state->codes; - state->lencode = (code const FAR *)(state->next); - state->lenbits = 9; - ret = inflate_table(LENS, state->lens, state->nlen, &(state->next), - &(state->lenbits), state->work); - if (ret) { - strm->msg = (char *)"invalid literal/lengths set"; - state->mode = BAD; - break; - } - state->distcode = (code const FAR *)(state->next); - state->distbits = 6; - ret = inflate_table(DISTS, state->lens + state->nlen, state->ndist, - &(state->next), &(state->distbits), state->work); - if (ret) { - strm->msg = (char *)"invalid distances set"; - state->mode = BAD; - break; - } - Tracev((stderr, "inflate: codes ok\n")); - state->mode = LEN; - - case LEN: - /* use inflate_fast() if we have enough input and output */ - if (have >= 6 && left >= 258) { - RESTORE(); - if (state->whave < state->wsize) - state->whave = state->wsize - left; - inflate_fast(strm, state->wsize); - LOAD(); - break; - } - - /* get a literal, length, or end-of-block code */ - for (;;) { - here = state->lencode[BITS(state->lenbits)]; - if ((unsigned)(here.bits) <= bits) break; - PULLBYTE(); - } - if (here.op && (here.op & 0xf0) == 0) { - last = here; - for (;;) { - here = state->lencode[last.val + - (BITS(last.bits + last.op) >> last.bits)]; - if ((unsigned)(last.bits + here.bits) <= bits) break; - PULLBYTE(); - } - DROPBITS(last.bits); - } - DROPBITS(here.bits); - state->length = (unsigned)here.val; - - /* process literal */ - if (here.op == 0) { - Tracevv((stderr, here.val >= 0x20 && here.val < 0x7f ? - "inflate: literal '%c'\n" : - "inflate: literal 0x%02x\n", here.val)); - ROOM(); - *put++ = (unsigned char)(state->length); - left--; - state->mode = LEN; - break; - } - - /* process end of block */ - if (here.op & 32) { - Tracevv((stderr, "inflate: end of block\n")); - state->mode = TYPE; - break; - } - - /* invalid code */ - if (here.op & 64) { - strm->msg = (char *)"invalid literal/length code"; - state->mode = BAD; - break; - } - - /* length code -- get extra bits, if any */ - state->extra = (unsigned)(here.op) & 15; - if (state->extra != 0) { - NEEDBITS(state->extra); - state->length += BITS(state->extra); - DROPBITS(state->extra); - } - Tracevv((stderr, "inflate: length %u\n", state->length)); - - /* get distance code */ - for (;;) { - here = state->distcode[BITS(state->distbits)]; - if ((unsigned)(here.bits) <= bits) break; - PULLBYTE(); - } - if ((here.op & 0xf0) == 0) { - last = here; - for (;;) { - here = state->distcode[last.val + - (BITS(last.bits + last.op) >> last.bits)]; - if ((unsigned)(last.bits + here.bits) <= bits) break; - PULLBYTE(); - } - DROPBITS(last.bits); - } - DROPBITS(here.bits); - if (here.op & 64) { - strm->msg = (char *)"invalid distance code"; - state->mode = BAD; - break; - } - state->offset = (unsigned)here.val; - - /* get distance extra bits, if any */ - state->extra = (unsigned)(here.op) & 15; - if (state->extra != 0) { - NEEDBITS(state->extra); - state->offset += BITS(state->extra); - DROPBITS(state->extra); - } - if (state->offset > state->wsize - (state->whave < state->wsize ? - left : 0)) { - strm->msg = (char *)"invalid distance too far back"; - state->mode = BAD; - break; - } - Tracevv((stderr, "inflate: distance %u\n", state->offset)); - - /* copy match from window to output */ - do { - ROOM(); - copy = state->wsize - state->offset; - if (copy < left) { - from = put + copy; - copy = left - copy; - } - else { - from = put - state->offset; - copy = left; - } - if (copy > state->length) copy = state->length; - state->length -= copy; - left -= copy; - do { - *put++ = *from++; - } while (--copy); - } while (state->length != 0); - break; - - case DONE: - /* inflate stream terminated properly -- write leftover output */ - ret = Z_STREAM_END; - if (left < state->wsize) { - if (out(out_desc, state->window, state->wsize - left)) - ret = Z_BUF_ERROR; - } - goto inf_leave; - - case BAD: - ret = Z_DATA_ERROR; - goto inf_leave; - - default: /* can't happen, but makes compilers happy */ - ret = Z_STREAM_ERROR; - goto inf_leave; - } - - /* Return unused input */ - inf_leave: - strm->next_in = next; - strm->avail_in = have; - return ret; -} - -int ZEXPORT inflateBackEnd(strm) -z_streamp strm; -{ - if (strm == Z_NULL || strm->state == Z_NULL || strm->zfree == (free_func)0) - return Z_STREAM_ERROR; - ZFREE(strm, strm->state); - strm->state = Z_NULL; - Tracev((stderr, "inflate: end\n")); - return Z_OK; -} diff --git a/dep/zlib/src/inffast.c b/dep/zlib/src/inffast.c deleted file mode 100644 index 0dbd1dbc0..000000000 --- a/dep/zlib/src/inffast.c +++ /dev/null @@ -1,323 +0,0 @@ -/* inffast.c -- fast decoding - * Copyright (C) 1995-2017 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -#include "zutil.h" -#include "inftrees.h" -#include "inflate.h" -#include "inffast.h" - -#ifdef ASMINF -# pragma message("Assembler code may have bugs -- use at your own risk") -#else - -/* - Decode literal, length, and distance codes and write out the resulting - literal and match bytes until either not enough input or output is - available, an end-of-block is encountered, or a data error is encountered. - When large enough input and output buffers are supplied to inflate(), for - example, a 16K input buffer and a 64K output buffer, more than 95% of the - inflate execution time is spent in this routine. - - Entry assumptions: - - state->mode == LEN - strm->avail_in >= 6 - strm->avail_out >= 258 - start >= strm->avail_out - state->bits < 8 - - On return, state->mode is one of: - - LEN -- ran out of enough output space or enough available input - TYPE -- reached end of block code, inflate() to interpret next block - BAD -- error in block data - - Notes: - - - The maximum input bits used by a length/distance pair is 15 bits for the - length code, 5 bits for the length extra, 15 bits for the distance code, - and 13 bits for the distance extra. This totals 48 bits, or six bytes. - Therefore if strm->avail_in >= 6, then there is enough input to avoid - checking for available input while decoding. - - - The maximum bytes that a single length/distance pair can output is 258 - bytes, which is the maximum length that can be coded. inflate_fast() - requires strm->avail_out >= 258 for each loop to avoid checking for - output space. - */ -void ZLIB_INTERNAL inflate_fast(strm, start) -z_streamp strm; -unsigned start; /* inflate()'s starting value for strm->avail_out */ -{ - struct inflate_state FAR *state; - z_const unsigned char FAR *in; /* local strm->next_in */ - z_const unsigned char FAR *last; /* have enough input while in < last */ - unsigned char FAR *out; /* local strm->next_out */ - unsigned char FAR *beg; /* inflate()'s initial strm->next_out */ - unsigned char FAR *end; /* while out < end, enough space available */ -#ifdef INFLATE_STRICT - unsigned dmax; /* maximum distance from zlib header */ -#endif - unsigned wsize; /* window size or zero if not using window */ - unsigned whave; /* valid bytes in the window */ - unsigned wnext; /* window write index */ - unsigned char FAR *window; /* allocated sliding window, if wsize != 0 */ - unsigned long hold; /* local strm->hold */ - unsigned bits; /* local strm->bits */ - code const FAR *lcode; /* local strm->lencode */ - code const FAR *dcode; /* local strm->distcode */ - unsigned lmask; /* mask for first level of length codes */ - unsigned dmask; /* mask for first level of distance codes */ - code here; /* retrieved table entry */ - unsigned op; /* code bits, operation, extra bits, or */ - /* window position, window bytes to copy */ - unsigned len; /* match length, unused bytes */ - unsigned dist; /* match distance */ - unsigned char FAR *from; /* where to copy match from */ - - /* copy state to local variables */ - state = (struct inflate_state FAR *)strm->state; - in = strm->next_in; - last = in + (strm->avail_in - 5); - out = strm->next_out; - beg = out - (start - strm->avail_out); - end = out + (strm->avail_out - 257); -#ifdef INFLATE_STRICT - dmax = state->dmax; -#endif - wsize = state->wsize; - whave = state->whave; - wnext = state->wnext; - window = state->window; - hold = state->hold; - bits = state->bits; - lcode = state->lencode; - dcode = state->distcode; - lmask = (1U << state->lenbits) - 1; - dmask = (1U << state->distbits) - 1; - - /* decode literals and length/distances until end-of-block or not enough - input data or output space */ - do { - if (bits < 15) { - hold += (unsigned long)(*in++) << bits; - bits += 8; - hold += (unsigned long)(*in++) << bits; - bits += 8; - } - here = lcode[hold & lmask]; - dolen: - op = (unsigned)(here.bits); - hold >>= op; - bits -= op; - op = (unsigned)(here.op); - if (op == 0) { /* literal */ - Tracevv((stderr, here.val >= 0x20 && here.val < 0x7f ? - "inflate: literal '%c'\n" : - "inflate: literal 0x%02x\n", here.val)); - *out++ = (unsigned char)(here.val); - } - else if (op & 16) { /* length base */ - len = (unsigned)(here.val); - op &= 15; /* number of extra bits */ - if (op) { - if (bits < op) { - hold += (unsigned long)(*in++) << bits; - bits += 8; - } - len += (unsigned)hold & ((1U << op) - 1); - hold >>= op; - bits -= op; - } - Tracevv((stderr, "inflate: length %u\n", len)); - if (bits < 15) { - hold += (unsigned long)(*in++) << bits; - bits += 8; - hold += (unsigned long)(*in++) << bits; - bits += 8; - } - here = dcode[hold & dmask]; - dodist: - op = (unsigned)(here.bits); - hold >>= op; - bits -= op; - op = (unsigned)(here.op); - if (op & 16) { /* distance base */ - dist = (unsigned)(here.val); - op &= 15; /* number of extra bits */ - if (bits < op) { - hold += (unsigned long)(*in++) << bits; - bits += 8; - if (bits < op) { - hold += (unsigned long)(*in++) << bits; - bits += 8; - } - } - dist += (unsigned)hold & ((1U << op) - 1); -#ifdef INFLATE_STRICT - if (dist > dmax) { - strm->msg = (char *)"invalid distance too far back"; - state->mode = BAD; - break; - } -#endif - hold >>= op; - bits -= op; - Tracevv((stderr, "inflate: distance %u\n", dist)); - op = (unsigned)(out - beg); /* max distance in output */ - if (dist > op) { /* see if copy from window */ - op = dist - op; /* distance back in window */ - if (op > whave) { - if (state->sane) { - strm->msg = - (char *)"invalid distance too far back"; - state->mode = BAD; - break; - } -#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR - if (len <= op - whave) { - do { - *out++ = 0; - } while (--len); - continue; - } - len -= op - whave; - do { - *out++ = 0; - } while (--op > whave); - if (op == 0) { - from = out - dist; - do { - *out++ = *from++; - } while (--len); - continue; - } -#endif - } - from = window; - if (wnext == 0) { /* very common case */ - from += wsize - op; - if (op < len) { /* some from window */ - len -= op; - do { - *out++ = *from++; - } while (--op); - from = out - dist; /* rest from output */ - } - } - else if (wnext < op) { /* wrap around window */ - from += wsize + wnext - op; - op -= wnext; - if (op < len) { /* some from end of window */ - len -= op; - do { - *out++ = *from++; - } while (--op); - from = window; - if (wnext < len) { /* some from start of window */ - op = wnext; - len -= op; - do { - *out++ = *from++; - } while (--op); - from = out - dist; /* rest from output */ - } - } - } - else { /* contiguous in window */ - from += wnext - op; - if (op < len) { /* some from window */ - len -= op; - do { - *out++ = *from++; - } while (--op); - from = out - dist; /* rest from output */ - } - } - while (len > 2) { - *out++ = *from++; - *out++ = *from++; - *out++ = *from++; - len -= 3; - } - if (len) { - *out++ = *from++; - if (len > 1) - *out++ = *from++; - } - } - else { - from = out - dist; /* copy direct from output */ - do { /* minimum length is three */ - *out++ = *from++; - *out++ = *from++; - *out++ = *from++; - len -= 3; - } while (len > 2); - if (len) { - *out++ = *from++; - if (len > 1) - *out++ = *from++; - } - } - } - else if ((op & 64) == 0) { /* 2nd level distance code */ - here = dcode[here.val + (hold & ((1U << op) - 1))]; - goto dodist; - } - else { - strm->msg = (char *)"invalid distance code"; - state->mode = BAD; - break; - } - } - else if ((op & 64) == 0) { /* 2nd level length code */ - here = lcode[here.val + (hold & ((1U << op) - 1))]; - goto dolen; - } - else if (op & 32) { /* end-of-block */ - Tracevv((stderr, "inflate: end of block\n")); - state->mode = TYPE; - break; - } - else { - strm->msg = (char *)"invalid literal/length code"; - state->mode = BAD; - break; - } - } while (in < last && out < end); - - /* return unused bytes (on entry, bits < 8, so in won't go too far back) */ - len = bits >> 3; - in -= len; - bits -= len << 3; - hold &= (1U << bits) - 1; - - /* update state and return */ - strm->next_in = in; - strm->next_out = out; - strm->avail_in = (unsigned)(in < last ? 5 + (last - in) : 5 - (in - last)); - strm->avail_out = (unsigned)(out < end ? - 257 + (end - out) : 257 - (out - end)); - state->hold = hold; - state->bits = bits; - return; -} - -/* - inflate_fast() speedups that turned out slower (on a PowerPC G3 750CXe): - - Using bit fields for code structure - - Different op definition to avoid & for extra bits (do & for table bits) - - Three separate decoding do-loops for direct, window, and wnext == 0 - - Special case for distance > 1 copies to do overlapped load and store copy - - Explicit branch predictions (based on measured branch probabilities) - - Deferring match copy and interspersed it with decoding subsequent codes - - Swapping literal/length else - - Swapping window/direct else - - Larger unrolled copy loops (three is about right) - - Moving len -= 3 statement into middle of loop - */ - -#endif /* !ASMINF */ diff --git a/dep/zlib/src/inffast.h b/dep/zlib/src/inffast.h deleted file mode 100644 index e5c1aa4ca..000000000 --- a/dep/zlib/src/inffast.h +++ /dev/null @@ -1,11 +0,0 @@ -/* inffast.h -- header to use inffast.c - * Copyright (C) 1995-2003, 2010 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* WARNING: this file should *not* be used by applications. It is - part of the implementation of the compression library and is - subject to change. Applications should only use zlib.h. - */ - -void ZLIB_INTERNAL inflate_fast OF((z_streamp strm, unsigned start)); diff --git a/dep/zlib/src/inffixed.h b/dep/zlib/src/inffixed.h deleted file mode 100644 index d62832776..000000000 --- a/dep/zlib/src/inffixed.h +++ /dev/null @@ -1,94 +0,0 @@ - /* inffixed.h -- table for decoding fixed codes - * Generated automatically by makefixed(). - */ - - /* WARNING: this file should *not* be used by applications. - It is part of the implementation of this library and is - subject to change. Applications should only use zlib.h. - */ - - static const code lenfix[512] = { - {96,7,0},{0,8,80},{0,8,16},{20,8,115},{18,7,31},{0,8,112},{0,8,48}, - {0,9,192},{16,7,10},{0,8,96},{0,8,32},{0,9,160},{0,8,0},{0,8,128}, - {0,8,64},{0,9,224},{16,7,6},{0,8,88},{0,8,24},{0,9,144},{19,7,59}, - {0,8,120},{0,8,56},{0,9,208},{17,7,17},{0,8,104},{0,8,40},{0,9,176}, - {0,8,8},{0,8,136},{0,8,72},{0,9,240},{16,7,4},{0,8,84},{0,8,20}, - {21,8,227},{19,7,43},{0,8,116},{0,8,52},{0,9,200},{17,7,13},{0,8,100}, - {0,8,36},{0,9,168},{0,8,4},{0,8,132},{0,8,68},{0,9,232},{16,7,8}, - {0,8,92},{0,8,28},{0,9,152},{20,7,83},{0,8,124},{0,8,60},{0,9,216}, - {18,7,23},{0,8,108},{0,8,44},{0,9,184},{0,8,12},{0,8,140},{0,8,76}, - {0,9,248},{16,7,3},{0,8,82},{0,8,18},{21,8,163},{19,7,35},{0,8,114}, - {0,8,50},{0,9,196},{17,7,11},{0,8,98},{0,8,34},{0,9,164},{0,8,2}, - {0,8,130},{0,8,66},{0,9,228},{16,7,7},{0,8,90},{0,8,26},{0,9,148}, - {20,7,67},{0,8,122},{0,8,58},{0,9,212},{18,7,19},{0,8,106},{0,8,42}, - {0,9,180},{0,8,10},{0,8,138},{0,8,74},{0,9,244},{16,7,5},{0,8,86}, - {0,8,22},{64,8,0},{19,7,51},{0,8,118},{0,8,54},{0,9,204},{17,7,15}, - {0,8,102},{0,8,38},{0,9,172},{0,8,6},{0,8,134},{0,8,70},{0,9,236}, - {16,7,9},{0,8,94},{0,8,30},{0,9,156},{20,7,99},{0,8,126},{0,8,62}, - {0,9,220},{18,7,27},{0,8,110},{0,8,46},{0,9,188},{0,8,14},{0,8,142}, - {0,8,78},{0,9,252},{96,7,0},{0,8,81},{0,8,17},{21,8,131},{18,7,31}, - {0,8,113},{0,8,49},{0,9,194},{16,7,10},{0,8,97},{0,8,33},{0,9,162}, - {0,8,1},{0,8,129},{0,8,65},{0,9,226},{16,7,6},{0,8,89},{0,8,25}, - {0,9,146},{19,7,59},{0,8,121},{0,8,57},{0,9,210},{17,7,17},{0,8,105}, - {0,8,41},{0,9,178},{0,8,9},{0,8,137},{0,8,73},{0,9,242},{16,7,4}, - {0,8,85},{0,8,21},{16,8,258},{19,7,43},{0,8,117},{0,8,53},{0,9,202}, - {17,7,13},{0,8,101},{0,8,37},{0,9,170},{0,8,5},{0,8,133},{0,8,69}, - {0,9,234},{16,7,8},{0,8,93},{0,8,29},{0,9,154},{20,7,83},{0,8,125}, - {0,8,61},{0,9,218},{18,7,23},{0,8,109},{0,8,45},{0,9,186},{0,8,13}, - {0,8,141},{0,8,77},{0,9,250},{16,7,3},{0,8,83},{0,8,19},{21,8,195}, - {19,7,35},{0,8,115},{0,8,51},{0,9,198},{17,7,11},{0,8,99},{0,8,35}, - {0,9,166},{0,8,3},{0,8,131},{0,8,67},{0,9,230},{16,7,7},{0,8,91}, - {0,8,27},{0,9,150},{20,7,67},{0,8,123},{0,8,59},{0,9,214},{18,7,19}, - {0,8,107},{0,8,43},{0,9,182},{0,8,11},{0,8,139},{0,8,75},{0,9,246}, - {16,7,5},{0,8,87},{0,8,23},{64,8,0},{19,7,51},{0,8,119},{0,8,55}, - {0,9,206},{17,7,15},{0,8,103},{0,8,39},{0,9,174},{0,8,7},{0,8,135}, - {0,8,71},{0,9,238},{16,7,9},{0,8,95},{0,8,31},{0,9,158},{20,7,99}, - {0,8,127},{0,8,63},{0,9,222},{18,7,27},{0,8,111},{0,8,47},{0,9,190}, - {0,8,15},{0,8,143},{0,8,79},{0,9,254},{96,7,0},{0,8,80},{0,8,16}, - {20,8,115},{18,7,31},{0,8,112},{0,8,48},{0,9,193},{16,7,10},{0,8,96}, - {0,8,32},{0,9,161},{0,8,0},{0,8,128},{0,8,64},{0,9,225},{16,7,6}, - {0,8,88},{0,8,24},{0,9,145},{19,7,59},{0,8,120},{0,8,56},{0,9,209}, - {17,7,17},{0,8,104},{0,8,40},{0,9,177},{0,8,8},{0,8,136},{0,8,72}, - {0,9,241},{16,7,4},{0,8,84},{0,8,20},{21,8,227},{19,7,43},{0,8,116}, - {0,8,52},{0,9,201},{17,7,13},{0,8,100},{0,8,36},{0,9,169},{0,8,4}, - {0,8,132},{0,8,68},{0,9,233},{16,7,8},{0,8,92},{0,8,28},{0,9,153}, - {20,7,83},{0,8,124},{0,8,60},{0,9,217},{18,7,23},{0,8,108},{0,8,44}, - {0,9,185},{0,8,12},{0,8,140},{0,8,76},{0,9,249},{16,7,3},{0,8,82}, - {0,8,18},{21,8,163},{19,7,35},{0,8,114},{0,8,50},{0,9,197},{17,7,11}, - {0,8,98},{0,8,34},{0,9,165},{0,8,2},{0,8,130},{0,8,66},{0,9,229}, - {16,7,7},{0,8,90},{0,8,26},{0,9,149},{20,7,67},{0,8,122},{0,8,58}, - {0,9,213},{18,7,19},{0,8,106},{0,8,42},{0,9,181},{0,8,10},{0,8,138}, - {0,8,74},{0,9,245},{16,7,5},{0,8,86},{0,8,22},{64,8,0},{19,7,51}, - {0,8,118},{0,8,54},{0,9,205},{17,7,15},{0,8,102},{0,8,38},{0,9,173}, - {0,8,6},{0,8,134},{0,8,70},{0,9,237},{16,7,9},{0,8,94},{0,8,30}, - {0,9,157},{20,7,99},{0,8,126},{0,8,62},{0,9,221},{18,7,27},{0,8,110}, - {0,8,46},{0,9,189},{0,8,14},{0,8,142},{0,8,78},{0,9,253},{96,7,0}, - {0,8,81},{0,8,17},{21,8,131},{18,7,31},{0,8,113},{0,8,49},{0,9,195}, - {16,7,10},{0,8,97},{0,8,33},{0,9,163},{0,8,1},{0,8,129},{0,8,65}, - {0,9,227},{16,7,6},{0,8,89},{0,8,25},{0,9,147},{19,7,59},{0,8,121}, - {0,8,57},{0,9,211},{17,7,17},{0,8,105},{0,8,41},{0,9,179},{0,8,9}, - {0,8,137},{0,8,73},{0,9,243},{16,7,4},{0,8,85},{0,8,21},{16,8,258}, - {19,7,43},{0,8,117},{0,8,53},{0,9,203},{17,7,13},{0,8,101},{0,8,37}, - {0,9,171},{0,8,5},{0,8,133},{0,8,69},{0,9,235},{16,7,8},{0,8,93}, - {0,8,29},{0,9,155},{20,7,83},{0,8,125},{0,8,61},{0,9,219},{18,7,23}, - {0,8,109},{0,8,45},{0,9,187},{0,8,13},{0,8,141},{0,8,77},{0,9,251}, - {16,7,3},{0,8,83},{0,8,19},{21,8,195},{19,7,35},{0,8,115},{0,8,51}, - {0,9,199},{17,7,11},{0,8,99},{0,8,35},{0,9,167},{0,8,3},{0,8,131}, - {0,8,67},{0,9,231},{16,7,7},{0,8,91},{0,8,27},{0,9,151},{20,7,67}, - {0,8,123},{0,8,59},{0,9,215},{18,7,19},{0,8,107},{0,8,43},{0,9,183}, - {0,8,11},{0,8,139},{0,8,75},{0,9,247},{16,7,5},{0,8,87},{0,8,23}, - {64,8,0},{19,7,51},{0,8,119},{0,8,55},{0,9,207},{17,7,15},{0,8,103}, - {0,8,39},{0,9,175},{0,8,7},{0,8,135},{0,8,71},{0,9,239},{16,7,9}, - {0,8,95},{0,8,31},{0,9,159},{20,7,99},{0,8,127},{0,8,63},{0,9,223}, - {18,7,27},{0,8,111},{0,8,47},{0,9,191},{0,8,15},{0,8,143},{0,8,79}, - {0,9,255} - }; - - static const code distfix[32] = { - {16,5,1},{23,5,257},{19,5,17},{27,5,4097},{17,5,5},{25,5,1025}, - {21,5,65},{29,5,16385},{16,5,3},{24,5,513},{20,5,33},{28,5,8193}, - {18,5,9},{26,5,2049},{22,5,129},{64,5,0},{16,5,2},{23,5,385}, - {19,5,25},{27,5,6145},{17,5,7},{25,5,1537},{21,5,97},{29,5,24577}, - {16,5,4},{24,5,769},{20,5,49},{28,5,12289},{18,5,13},{26,5,3073}, - {22,5,193},{64,5,0} - }; diff --git a/dep/zlib/src/inflate.c b/dep/zlib/src/inflate.c deleted file mode 100644 index ac333e8c2..000000000 --- a/dep/zlib/src/inflate.c +++ /dev/null @@ -1,1561 +0,0 @@ -/* inflate.c -- zlib decompression - * Copyright (C) 1995-2016 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* - * Change history: - * - * 1.2.beta0 24 Nov 2002 - * - First version -- complete rewrite of inflate to simplify code, avoid - * creation of window when not needed, minimize use of window when it is - * needed, make inffast.c even faster, implement gzip decoding, and to - * improve code readability and style over the previous zlib inflate code - * - * 1.2.beta1 25 Nov 2002 - * - Use pointers for available input and output checking in inffast.c - * - Remove input and output counters in inffast.c - * - Change inffast.c entry and loop from avail_in >= 7 to >= 6 - * - Remove unnecessary second byte pull from length extra in inffast.c - * - Unroll direct copy to three copies per loop in inffast.c - * - * 1.2.beta2 4 Dec 2002 - * - Change external routine names to reduce potential conflicts - * - Correct filename to inffixed.h for fixed tables in inflate.c - * - Make hbuf[] unsigned char to match parameter type in inflate.c - * - Change strm->next_out[-state->offset] to *(strm->next_out - state->offset) - * to avoid negation problem on Alphas (64 bit) in inflate.c - * - * 1.2.beta3 22 Dec 2002 - * - Add comments on state->bits assertion in inffast.c - * - Add comments on op field in inftrees.h - * - Fix bug in reuse of allocated window after inflateReset() - * - Remove bit fields--back to byte structure for speed - * - Remove distance extra == 0 check in inflate_fast()--only helps for lengths - * - Change post-increments to pre-increments in inflate_fast(), PPC biased? - * - Add compile time option, POSTINC, to use post-increments instead (Intel?) - * - Make MATCH copy in inflate() much faster for when inflate_fast() not used - * - Use local copies of stream next and avail values, as well as local bit - * buffer and bit count in inflate()--for speed when inflate_fast() not used - * - * 1.2.beta4 1 Jan 2003 - * - Split ptr - 257 statements in inflate_table() to avoid compiler warnings - * - Move a comment on output buffer sizes from inffast.c to inflate.c - * - Add comments in inffast.c to introduce the inflate_fast() routine - * - Rearrange window copies in inflate_fast() for speed and simplification - * - Unroll last copy for window match in inflate_fast() - * - Use local copies of window variables in inflate_fast() for speed - * - Pull out common wnext == 0 case for speed in inflate_fast() - * - Make op and len in inflate_fast() unsigned for consistency - * - Add FAR to lcode and dcode declarations in inflate_fast() - * - Simplified bad distance check in inflate_fast() - * - Added inflateBackInit(), inflateBack(), and inflateBackEnd() in new - * source file infback.c to provide a call-back interface to inflate for - * programs like gzip and unzip -- uses window as output buffer to avoid - * window copying - * - * 1.2.beta5 1 Jan 2003 - * - Improved inflateBack() interface to allow the caller to provide initial - * input in strm. - * - Fixed stored blocks bug in inflateBack() - * - * 1.2.beta6 4 Jan 2003 - * - Added comments in inffast.c on effectiveness of POSTINC - * - Typecasting all around to reduce compiler warnings - * - Changed loops from while (1) or do {} while (1) to for (;;), again to - * make compilers happy - * - Changed type of window in inflateBackInit() to unsigned char * - * - * 1.2.beta7 27 Jan 2003 - * - Changed many types to unsigned or unsigned short to avoid warnings - * - Added inflateCopy() function - * - * 1.2.0 9 Mar 2003 - * - Changed inflateBack() interface to provide separate opaque descriptors - * for the in() and out() functions - * - Changed inflateBack() argument and in_func typedef to swap the length - * and buffer address return values for the input function - * - Check next_in and next_out for Z_NULL on entry to inflate() - * - * The history for versions after 1.2.0 are in ChangeLog in zlib distribution. - */ - -#include "zutil.h" -#include "inftrees.h" -#include "inflate.h" -#include "inffast.h" - -#ifdef MAKEFIXED -# ifndef BUILDFIXED -# define BUILDFIXED -# endif -#endif - -/* function prototypes */ -local int inflateStateCheck OF((z_streamp strm)); -local void fixedtables OF((struct inflate_state FAR *state)); -local int updatewindow OF((z_streamp strm, const unsigned char FAR *end, - unsigned copy)); -#ifdef BUILDFIXED - void makefixed OF((void)); -#endif -local unsigned syncsearch OF((unsigned FAR *have, const unsigned char FAR *buf, - unsigned len)); - -local int inflateStateCheck(strm) -z_streamp strm; -{ - struct inflate_state FAR *state; - if (strm == Z_NULL || - strm->zalloc == (alloc_func)0 || strm->zfree == (free_func)0) - return 1; - state = (struct inflate_state FAR *)strm->state; - if (state == Z_NULL || state->strm != strm || - state->mode < HEAD || state->mode > SYNC) - return 1; - return 0; -} - -int ZEXPORT inflateResetKeep(strm) -z_streamp strm; -{ - struct inflate_state FAR *state; - - if (inflateStateCheck(strm)) return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; - strm->total_in = strm->total_out = state->total = 0; - strm->msg = Z_NULL; - if (state->wrap) /* to support ill-conceived Java test suite */ - strm->adler = state->wrap & 1; - state->mode = HEAD; - state->last = 0; - state->havedict = 0; - state->dmax = 32768U; - state->head = Z_NULL; - state->hold = 0; - state->bits = 0; - state->lencode = state->distcode = state->next = state->codes; - state->sane = 1; - state->back = -1; - Tracev((stderr, "inflate: reset\n")); - return Z_OK; -} - -int ZEXPORT inflateReset(strm) -z_streamp strm; -{ - struct inflate_state FAR *state; - - if (inflateStateCheck(strm)) return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; - state->wsize = 0; - state->whave = 0; - state->wnext = 0; - return inflateResetKeep(strm); -} - -int ZEXPORT inflateReset2(strm, windowBits) -z_streamp strm; -int windowBits; -{ - int wrap; - struct inflate_state FAR *state; - - /* get the state */ - if (inflateStateCheck(strm)) return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; - - /* extract wrap request from windowBits parameter */ - if (windowBits < 0) { - wrap = 0; - windowBits = -windowBits; - } - else { - wrap = (windowBits >> 4) + 5; -#ifdef GUNZIP - if (windowBits < 48) - windowBits &= 15; -#endif - } - - /* set number of window bits, free window if different */ - if (windowBits && (windowBits < 8 || windowBits > 15)) - return Z_STREAM_ERROR; - if (state->window != Z_NULL && state->wbits != (unsigned)windowBits) { - ZFREE(strm, state->window); - state->window = Z_NULL; - } - - /* update state and reset the rest of it */ - state->wrap = wrap; - state->wbits = (unsigned)windowBits; - return inflateReset(strm); -} - -int ZEXPORT inflateInit2_(strm, windowBits, version, stream_size) -z_streamp strm; -int windowBits; -const char *version; -int stream_size; -{ - int ret; - struct inflate_state FAR *state; - - if (version == Z_NULL || version[0] != ZLIB_VERSION[0] || - stream_size != (int)(sizeof(z_stream))) - return Z_VERSION_ERROR; - if (strm == Z_NULL) return Z_STREAM_ERROR; - strm->msg = Z_NULL; /* in case we return an error */ - if (strm->zalloc == (alloc_func)0) { -#ifdef Z_SOLO - return Z_STREAM_ERROR; -#else - strm->zalloc = zcalloc; - strm->opaque = (voidpf)0; -#endif - } - if (strm->zfree == (free_func)0) -#ifdef Z_SOLO - return Z_STREAM_ERROR; -#else - strm->zfree = zcfree; -#endif - state = (struct inflate_state FAR *) - ZALLOC(strm, 1, sizeof(struct inflate_state)); - if (state == Z_NULL) return Z_MEM_ERROR; - Tracev((stderr, "inflate: allocated\n")); - strm->state = (struct internal_state FAR *)state; - state->strm = strm; - state->window = Z_NULL; - state->mode = HEAD; /* to pass state test in inflateReset2() */ - ret = inflateReset2(strm, windowBits); - if (ret != Z_OK) { - ZFREE(strm, state); - strm->state = Z_NULL; - } - return ret; -} - -int ZEXPORT inflateInit_(strm, version, stream_size) -z_streamp strm; -const char *version; -int stream_size; -{ - return inflateInit2_(strm, DEF_WBITS, version, stream_size); -} - -int ZEXPORT inflatePrime(strm, bits, value) -z_streamp strm; -int bits; -int value; -{ - struct inflate_state FAR *state; - - if (inflateStateCheck(strm)) return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; - if (bits < 0) { - state->hold = 0; - state->bits = 0; - return Z_OK; - } - if (bits > 16 || state->bits + (uInt)bits > 32) return Z_STREAM_ERROR; - value &= (1L << bits) - 1; - state->hold += (unsigned)value << state->bits; - state->bits += (uInt)bits; - return Z_OK; -} - -/* - Return state with length and distance decoding tables and index sizes set to - fixed code decoding. Normally this returns fixed tables from inffixed.h. - If BUILDFIXED is defined, then instead this routine builds the tables the - first time it's called, and returns those tables the first time and - thereafter. This reduces the size of the code by about 2K bytes, in - exchange for a little execution time. However, BUILDFIXED should not be - used for threaded applications, since the rewriting of the tables and virgin - may not be thread-safe. - */ -local void fixedtables(state) -struct inflate_state FAR *state; -{ -#ifdef BUILDFIXED - static int virgin = 1; - static code *lenfix, *distfix; - static code fixed[544]; - - /* build fixed huffman tables if first call (may not be thread safe) */ - if (virgin) { - unsigned sym, bits; - static code *next; - - /* literal/length table */ - sym = 0; - while (sym < 144) state->lens[sym++] = 8; - while (sym < 256) state->lens[sym++] = 9; - while (sym < 280) state->lens[sym++] = 7; - while (sym < 288) state->lens[sym++] = 8; - next = fixed; - lenfix = next; - bits = 9; - inflate_table(LENS, state->lens, 288, &(next), &(bits), state->work); - - /* distance table */ - sym = 0; - while (sym < 32) state->lens[sym++] = 5; - distfix = next; - bits = 5; - inflate_table(DISTS, state->lens, 32, &(next), &(bits), state->work); - - /* do this just once */ - virgin = 0; - } -#else /* !BUILDFIXED */ -# include "inffixed.h" -#endif /* BUILDFIXED */ - state->lencode = lenfix; - state->lenbits = 9; - state->distcode = distfix; - state->distbits = 5; -} - -#ifdef MAKEFIXED -#include - -/* - Write out the inffixed.h that is #include'd above. Defining MAKEFIXED also - defines BUILDFIXED, so the tables are built on the fly. makefixed() writes - those tables to stdout, which would be piped to inffixed.h. A small program - can simply call makefixed to do this: - - void makefixed(void); - - int main(void) - { - makefixed(); - return 0; - } - - Then that can be linked with zlib built with MAKEFIXED defined and run: - - a.out > inffixed.h - */ -void makefixed() -{ - unsigned low, size; - struct inflate_state state; - - fixedtables(&state); - puts(" /* inffixed.h -- table for decoding fixed codes"); - puts(" * Generated automatically by makefixed()."); - puts(" */"); - puts(""); - puts(" /* WARNING: this file should *not* be used by applications."); - puts(" It is part of the implementation of this library and is"); - puts(" subject to change. Applications should only use zlib.h."); - puts(" */"); - puts(""); - size = 1U << 9; - printf(" static const code lenfix[%u] = {", size); - low = 0; - for (;;) { - if ((low % 7) == 0) printf("\n "); - printf("{%u,%u,%d}", (low & 127) == 99 ? 64 : state.lencode[low].op, - state.lencode[low].bits, state.lencode[low].val); - if (++low == size) break; - putchar(','); - } - puts("\n };"); - size = 1U << 5; - printf("\n static const code distfix[%u] = {", size); - low = 0; - for (;;) { - if ((low % 6) == 0) printf("\n "); - printf("{%u,%u,%d}", state.distcode[low].op, state.distcode[low].bits, - state.distcode[low].val); - if (++low == size) break; - putchar(','); - } - puts("\n };"); -} -#endif /* MAKEFIXED */ - -/* - Update the window with the last wsize (normally 32K) bytes written before - returning. If window does not exist yet, create it. This is only called - when a window is already in use, or when output has been written during this - inflate call, but the end of the deflate stream has not been reached yet. - It is also called to create a window for dictionary data when a dictionary - is loaded. - - Providing output buffers larger than 32K to inflate() should provide a speed - advantage, since only the last 32K of output is copied to the sliding window - upon return from inflate(), and since all distances after the first 32K of - output will fall in the output data, making match copies simpler and faster. - The advantage may be dependent on the size of the processor's data caches. - */ -local int updatewindow(strm, end, copy) -z_streamp strm; -const Bytef *end; -unsigned copy; -{ - struct inflate_state FAR *state; - unsigned dist; - - state = (struct inflate_state FAR *)strm->state; - - /* if it hasn't been done already, allocate space for the window */ - if (state->window == Z_NULL) { - state->window = (unsigned char FAR *) - ZALLOC(strm, 1U << state->wbits, - sizeof(unsigned char)); - if (state->window == Z_NULL) return 1; - } - - /* if window not in use yet, initialize */ - if (state->wsize == 0) { - state->wsize = 1U << state->wbits; - state->wnext = 0; - state->whave = 0; - } - - /* copy state->wsize or less output bytes into the circular window */ - if (copy >= state->wsize) { - zmemcpy(state->window, end - state->wsize, state->wsize); - state->wnext = 0; - state->whave = state->wsize; - } - else { - dist = state->wsize - state->wnext; - if (dist > copy) dist = copy; - zmemcpy(state->window + state->wnext, end - copy, dist); - copy -= dist; - if (copy) { - zmemcpy(state->window, end - copy, copy); - state->wnext = copy; - state->whave = state->wsize; - } - else { - state->wnext += dist; - if (state->wnext == state->wsize) state->wnext = 0; - if (state->whave < state->wsize) state->whave += dist; - } - } - return 0; -} - -/* Macros for inflate(): */ - -/* check function to use adler32() for zlib or crc32() for gzip */ -#ifdef GUNZIP -# define UPDATE(check, buf, len) \ - (state->flags ? crc32(check, buf, len) : adler32(check, buf, len)) -#else -# define UPDATE(check, buf, len) adler32(check, buf, len) -#endif - -/* check macros for header crc */ -#ifdef GUNZIP -# define CRC2(check, word) \ - do { \ - hbuf[0] = (unsigned char)(word); \ - hbuf[1] = (unsigned char)((word) >> 8); \ - check = crc32(check, hbuf, 2); \ - } while (0) - -# define CRC4(check, word) \ - do { \ - hbuf[0] = (unsigned char)(word); \ - hbuf[1] = (unsigned char)((word) >> 8); \ - hbuf[2] = (unsigned char)((word) >> 16); \ - hbuf[3] = (unsigned char)((word) >> 24); \ - check = crc32(check, hbuf, 4); \ - } while (0) -#endif - -/* Load registers with state in inflate() for speed */ -#define LOAD() \ - do { \ - put = strm->next_out; \ - left = strm->avail_out; \ - next = strm->next_in; \ - have = strm->avail_in; \ - hold = state->hold; \ - bits = state->bits; \ - } while (0) - -/* Restore state from registers in inflate() */ -#define RESTORE() \ - do { \ - strm->next_out = put; \ - strm->avail_out = left; \ - strm->next_in = next; \ - strm->avail_in = have; \ - state->hold = hold; \ - state->bits = bits; \ - } while (0) - -/* Clear the input bit accumulator */ -#define INITBITS() \ - do { \ - hold = 0; \ - bits = 0; \ - } while (0) - -/* Get a byte of input into the bit accumulator, or return from inflate() - if there is no input available. */ -#define PULLBYTE() \ - do { \ - if (have == 0) goto inf_leave; \ - have--; \ - hold += (unsigned long)(*next++) << bits; \ - bits += 8; \ - } while (0) - -/* Assure that there are at least n bits in the bit accumulator. If there is - not enough available input to do that, then return from inflate(). */ -#define NEEDBITS(n) \ - do { \ - while (bits < (unsigned)(n)) \ - PULLBYTE(); \ - } while (0) - -/* Return the low n bits of the bit accumulator (n < 16) */ -#define BITS(n) \ - ((unsigned)hold & ((1U << (n)) - 1)) - -/* Remove n bits from the bit accumulator */ -#define DROPBITS(n) \ - do { \ - hold >>= (n); \ - bits -= (unsigned)(n); \ - } while (0) - -/* Remove zero to seven bits as needed to go to a byte boundary */ -#define BYTEBITS() \ - do { \ - hold >>= bits & 7; \ - bits -= bits & 7; \ - } while (0) - -/* - inflate() uses a state machine to process as much input data and generate as - much output data as possible before returning. The state machine is - structured roughly as follows: - - for (;;) switch (state) { - ... - case STATEn: - if (not enough input data or output space to make progress) - return; - ... make progress ... - state = STATEm; - break; - ... - } - - so when inflate() is called again, the same case is attempted again, and - if the appropriate resources are provided, the machine proceeds to the - next state. The NEEDBITS() macro is usually the way the state evaluates - whether it can proceed or should return. NEEDBITS() does the return if - the requested bits are not available. The typical use of the BITS macros - is: - - NEEDBITS(n); - ... do something with BITS(n) ... - DROPBITS(n); - - where NEEDBITS(n) either returns from inflate() if there isn't enough - input left to load n bits into the accumulator, or it continues. BITS(n) - gives the low n bits in the accumulator. When done, DROPBITS(n) drops - the low n bits off the accumulator. INITBITS() clears the accumulator - and sets the number of available bits to zero. BYTEBITS() discards just - enough bits to put the accumulator on a byte boundary. After BYTEBITS() - and a NEEDBITS(8), then BITS(8) would return the next byte in the stream. - - NEEDBITS(n) uses PULLBYTE() to get an available byte of input, or to return - if there is no input available. The decoding of variable length codes uses - PULLBYTE() directly in order to pull just enough bytes to decode the next - code, and no more. - - Some states loop until they get enough input, making sure that enough - state information is maintained to continue the loop where it left off - if NEEDBITS() returns in the loop. For example, want, need, and keep - would all have to actually be part of the saved state in case NEEDBITS() - returns: - - case STATEw: - while (want < need) { - NEEDBITS(n); - keep[want++] = BITS(n); - DROPBITS(n); - } - state = STATEx; - case STATEx: - - As shown above, if the next state is also the next case, then the break - is omitted. - - A state may also return if there is not enough output space available to - complete that state. Those states are copying stored data, writing a - literal byte, and copying a matching string. - - When returning, a "goto inf_leave" is used to update the total counters, - update the check value, and determine whether any progress has been made - during that inflate() call in order to return the proper return code. - Progress is defined as a change in either strm->avail_in or strm->avail_out. - When there is a window, goto inf_leave will update the window with the last - output written. If a goto inf_leave occurs in the middle of decompression - and there is no window currently, goto inf_leave will create one and copy - output to the window for the next call of inflate(). - - In this implementation, the flush parameter of inflate() only affects the - return code (per zlib.h). inflate() always writes as much as possible to - strm->next_out, given the space available and the provided input--the effect - documented in zlib.h of Z_SYNC_FLUSH. Furthermore, inflate() always defers - the allocation of and copying into a sliding window until necessary, which - provides the effect documented in zlib.h for Z_FINISH when the entire input - stream available. So the only thing the flush parameter actually does is: - when flush is set to Z_FINISH, inflate() cannot return Z_OK. Instead it - will return Z_BUF_ERROR if it has not reached the end of the stream. - */ - -int ZEXPORT inflate(strm, flush) -z_streamp strm; -int flush; -{ - struct inflate_state FAR *state; - z_const unsigned char FAR *next; /* next input */ - unsigned char FAR *put; /* next output */ - unsigned have, left; /* available input and output */ - unsigned long hold; /* bit buffer */ - unsigned bits; /* bits in bit buffer */ - unsigned in, out; /* save starting available input and output */ - unsigned copy; /* number of stored or match bytes to copy */ - unsigned char FAR *from; /* where to copy match bytes from */ - code here; /* current decoding table entry */ - code last; /* parent table entry */ - unsigned len; /* length to copy for repeats, bits to drop */ - int ret; /* return code */ -#ifdef GUNZIP - unsigned char hbuf[4]; /* buffer for gzip header crc calculation */ -#endif - static const unsigned short order[19] = /* permutation of code lengths */ - {16, 17, 18, 0, 8, 7, 9, 6, 10, 5, 11, 4, 12, 3, 13, 2, 14, 1, 15}; - - if (inflateStateCheck(strm) || strm->next_out == Z_NULL || - (strm->next_in == Z_NULL && strm->avail_in != 0)) - return Z_STREAM_ERROR; - - state = (struct inflate_state FAR *)strm->state; - if (state->mode == TYPE) state->mode = TYPEDO; /* skip check */ - LOAD(); - in = have; - out = left; - ret = Z_OK; - for (;;) - switch (state->mode) { - case HEAD: - if (state->wrap == 0) { - state->mode = TYPEDO; - break; - } - NEEDBITS(16); -#ifdef GUNZIP - if ((state->wrap & 2) && hold == 0x8b1f) { /* gzip header */ - if (state->wbits == 0) - state->wbits = 15; - state->check = crc32(0L, Z_NULL, 0); - CRC2(state->check, hold); - INITBITS(); - state->mode = FLAGS; - break; - } - state->flags = 0; /* expect zlib header */ - if (state->head != Z_NULL) - state->head->done = -1; - if (!(state->wrap & 1) || /* check if zlib header allowed */ -#else - if ( -#endif - ((BITS(8) << 8) + (hold >> 8)) % 31) { - strm->msg = (char *)"incorrect header check"; - state->mode = BAD; - break; - } - if (BITS(4) != Z_DEFLATED) { - strm->msg = (char *)"unknown compression method"; - state->mode = BAD; - break; - } - DROPBITS(4); - len = BITS(4) + 8; - if (state->wbits == 0) - state->wbits = len; - if (len > 15 || len > state->wbits) { - strm->msg = (char *)"invalid window size"; - state->mode = BAD; - break; - } - state->dmax = 1U << len; - Tracev((stderr, "inflate: zlib header ok\n")); - strm->adler = state->check = adler32(0L, Z_NULL, 0); - state->mode = hold & 0x200 ? DICTID : TYPE; - INITBITS(); - break; -#ifdef GUNZIP - case FLAGS: - NEEDBITS(16); - state->flags = (int)(hold); - if ((state->flags & 0xff) != Z_DEFLATED) { - strm->msg = (char *)"unknown compression method"; - state->mode = BAD; - break; - } - if (state->flags & 0xe000) { - strm->msg = (char *)"unknown header flags set"; - state->mode = BAD; - break; - } - if (state->head != Z_NULL) - state->head->text = (int)((hold >> 8) & 1); - if ((state->flags & 0x0200) && (state->wrap & 4)) - CRC2(state->check, hold); - INITBITS(); - state->mode = TIME; - case TIME: - NEEDBITS(32); - if (state->head != Z_NULL) - state->head->time = hold; - if ((state->flags & 0x0200) && (state->wrap & 4)) - CRC4(state->check, hold); - INITBITS(); - state->mode = OS; - case OS: - NEEDBITS(16); - if (state->head != Z_NULL) { - state->head->xflags = (int)(hold & 0xff); - state->head->os = (int)(hold >> 8); - } - if ((state->flags & 0x0200) && (state->wrap & 4)) - CRC2(state->check, hold); - INITBITS(); - state->mode = EXLEN; - case EXLEN: - if (state->flags & 0x0400) { - NEEDBITS(16); - state->length = (unsigned)(hold); - if (state->head != Z_NULL) - state->head->extra_len = (unsigned)hold; - if ((state->flags & 0x0200) && (state->wrap & 4)) - CRC2(state->check, hold); - INITBITS(); - } - else if (state->head != Z_NULL) - state->head->extra = Z_NULL; - state->mode = EXTRA; - case EXTRA: - if (state->flags & 0x0400) { - copy = state->length; - if (copy > have) copy = have; - if (copy) { - if (state->head != Z_NULL && - state->head->extra != Z_NULL) { - len = state->head->extra_len - state->length; - zmemcpy(state->head->extra + len, next, - len + copy > state->head->extra_max ? - state->head->extra_max - len : copy); - } - if ((state->flags & 0x0200) && (state->wrap & 4)) - state->check = crc32(state->check, next, copy); - have -= copy; - next += copy; - state->length -= copy; - } - if (state->length) goto inf_leave; - } - state->length = 0; - state->mode = NAME; - case NAME: - if (state->flags & 0x0800) { - if (have == 0) goto inf_leave; - copy = 0; - do { - len = (unsigned)(next[copy++]); - if (state->head != Z_NULL && - state->head->name != Z_NULL && - state->length < state->head->name_max) - state->head->name[state->length++] = (Bytef)len; - } while (len && copy < have); - if ((state->flags & 0x0200) && (state->wrap & 4)) - state->check = crc32(state->check, next, copy); - have -= copy; - next += copy; - if (len) goto inf_leave; - } - else if (state->head != Z_NULL) - state->head->name = Z_NULL; - state->length = 0; - state->mode = COMMENT; - case COMMENT: - if (state->flags & 0x1000) { - if (have == 0) goto inf_leave; - copy = 0; - do { - len = (unsigned)(next[copy++]); - if (state->head != Z_NULL && - state->head->comment != Z_NULL && - state->length < state->head->comm_max) - state->head->comment[state->length++] = (Bytef)len; - } while (len && copy < have); - if ((state->flags & 0x0200) && (state->wrap & 4)) - state->check = crc32(state->check, next, copy); - have -= copy; - next += copy; - if (len) goto inf_leave; - } - else if (state->head != Z_NULL) - state->head->comment = Z_NULL; - state->mode = HCRC; - case HCRC: - if (state->flags & 0x0200) { - NEEDBITS(16); - if ((state->wrap & 4) && hold != (state->check & 0xffff)) { - strm->msg = (char *)"header crc mismatch"; - state->mode = BAD; - break; - } - INITBITS(); - } - if (state->head != Z_NULL) { - state->head->hcrc = (int)((state->flags >> 9) & 1); - state->head->done = 1; - } - strm->adler = state->check = crc32(0L, Z_NULL, 0); - state->mode = TYPE; - break; -#endif - case DICTID: - NEEDBITS(32); - strm->adler = state->check = ZSWAP32(hold); - INITBITS(); - state->mode = DICT; - case DICT: - if (state->havedict == 0) { - RESTORE(); - return Z_NEED_DICT; - } - strm->adler = state->check = adler32(0L, Z_NULL, 0); - state->mode = TYPE; - case TYPE: - if (flush == Z_BLOCK || flush == Z_TREES) goto inf_leave; - case TYPEDO: - if (state->last) { - BYTEBITS(); - state->mode = CHECK; - break; - } - NEEDBITS(3); - state->last = BITS(1); - DROPBITS(1); - switch (BITS(2)) { - case 0: /* stored block */ - Tracev((stderr, "inflate: stored block%s\n", - state->last ? " (last)" : "")); - state->mode = STORED; - break; - case 1: /* fixed block */ - fixedtables(state); - Tracev((stderr, "inflate: fixed codes block%s\n", - state->last ? " (last)" : "")); - state->mode = LEN_; /* decode codes */ - if (flush == Z_TREES) { - DROPBITS(2); - goto inf_leave; - } - break; - case 2: /* dynamic block */ - Tracev((stderr, "inflate: dynamic codes block%s\n", - state->last ? " (last)" : "")); - state->mode = TABLE; - break; - case 3: - strm->msg = (char *)"invalid block type"; - state->mode = BAD; - } - DROPBITS(2); - break; - case STORED: - BYTEBITS(); /* go to byte boundary */ - NEEDBITS(32); - if ((hold & 0xffff) != ((hold >> 16) ^ 0xffff)) { - strm->msg = (char *)"invalid stored block lengths"; - state->mode = BAD; - break; - } - state->length = (unsigned)hold & 0xffff; - Tracev((stderr, "inflate: stored length %u\n", - state->length)); - INITBITS(); - state->mode = COPY_; - if (flush == Z_TREES) goto inf_leave; - case COPY_: - state->mode = COPY; - case COPY: - copy = state->length; - if (copy) { - if (copy > have) copy = have; - if (copy > left) copy = left; - if (copy == 0) goto inf_leave; - zmemcpy(put, next, copy); - have -= copy; - next += copy; - left -= copy; - put += copy; - state->length -= copy; - break; - } - Tracev((stderr, "inflate: stored end\n")); - state->mode = TYPE; - break; - case TABLE: - NEEDBITS(14); - state->nlen = BITS(5) + 257; - DROPBITS(5); - state->ndist = BITS(5) + 1; - DROPBITS(5); - state->ncode = BITS(4) + 4; - DROPBITS(4); -#ifndef PKZIP_BUG_WORKAROUND - if (state->nlen > 286 || state->ndist > 30) { - strm->msg = (char *)"too many length or distance symbols"; - state->mode = BAD; - break; - } -#endif - Tracev((stderr, "inflate: table sizes ok\n")); - state->have = 0; - state->mode = LENLENS; - case LENLENS: - while (state->have < state->ncode) { - NEEDBITS(3); - state->lens[order[state->have++]] = (unsigned short)BITS(3); - DROPBITS(3); - } - while (state->have < 19) - state->lens[order[state->have++]] = 0; - state->next = state->codes; - state->lencode = (const code FAR *)(state->next); - state->lenbits = 7; - ret = inflate_table(CODES, state->lens, 19, &(state->next), - &(state->lenbits), state->work); - if (ret) { - strm->msg = (char *)"invalid code lengths set"; - state->mode = BAD; - break; - } - Tracev((stderr, "inflate: code lengths ok\n")); - state->have = 0; - state->mode = CODELENS; - case CODELENS: - while (state->have < state->nlen + state->ndist) { - for (;;) { - here = state->lencode[BITS(state->lenbits)]; - if ((unsigned)(here.bits) <= bits) break; - PULLBYTE(); - } - if (here.val < 16) { - DROPBITS(here.bits); - state->lens[state->have++] = here.val; - } - else { - if (here.val == 16) { - NEEDBITS(here.bits + 2); - DROPBITS(here.bits); - if (state->have == 0) { - strm->msg = (char *)"invalid bit length repeat"; - state->mode = BAD; - break; - } - len = state->lens[state->have - 1]; - copy = 3 + BITS(2); - DROPBITS(2); - } - else if (here.val == 17) { - NEEDBITS(here.bits + 3); - DROPBITS(here.bits); - len = 0; - copy = 3 + BITS(3); - DROPBITS(3); - } - else { - NEEDBITS(here.bits + 7); - DROPBITS(here.bits); - len = 0; - copy = 11 + BITS(7); - DROPBITS(7); - } - if (state->have + copy > state->nlen + state->ndist) { - strm->msg = (char *)"invalid bit length repeat"; - state->mode = BAD; - break; - } - while (copy--) - state->lens[state->have++] = (unsigned short)len; - } - } - - /* handle error breaks in while */ - if (state->mode == BAD) break; - - /* check for end-of-block code (better have one) */ - if (state->lens[256] == 0) { - strm->msg = (char *)"invalid code -- missing end-of-block"; - state->mode = BAD; - break; - } - - /* build code tables -- note: do not change the lenbits or distbits - values here (9 and 6) without reading the comments in inftrees.h - concerning the ENOUGH constants, which depend on those values */ - state->next = state->codes; - state->lencode = (const code FAR *)(state->next); - state->lenbits = 9; - ret = inflate_table(LENS, state->lens, state->nlen, &(state->next), - &(state->lenbits), state->work); - if (ret) { - strm->msg = (char *)"invalid literal/lengths set"; - state->mode = BAD; - break; - } - state->distcode = (const code FAR *)(state->next); - state->distbits = 6; - ret = inflate_table(DISTS, state->lens + state->nlen, state->ndist, - &(state->next), &(state->distbits), state->work); - if (ret) { - strm->msg = (char *)"invalid distances set"; - state->mode = BAD; - break; - } - Tracev((stderr, "inflate: codes ok\n")); - state->mode = LEN_; - if (flush == Z_TREES) goto inf_leave; - case LEN_: - state->mode = LEN; - case LEN: - if (have >= 6 && left >= 258) { - RESTORE(); - inflate_fast(strm, out); - LOAD(); - if (state->mode == TYPE) - state->back = -1; - break; - } - state->back = 0; - for (;;) { - here = state->lencode[BITS(state->lenbits)]; - if ((unsigned)(here.bits) <= bits) break; - PULLBYTE(); - } - if (here.op && (here.op & 0xf0) == 0) { - last = here; - for (;;) { - here = state->lencode[last.val + - (BITS(last.bits + last.op) >> last.bits)]; - if ((unsigned)(last.bits + here.bits) <= bits) break; - PULLBYTE(); - } - DROPBITS(last.bits); - state->back += last.bits; - } - DROPBITS(here.bits); - state->back += here.bits; - state->length = (unsigned)here.val; - if ((int)(here.op) == 0) { - Tracevv((stderr, here.val >= 0x20 && here.val < 0x7f ? - "inflate: literal '%c'\n" : - "inflate: literal 0x%02x\n", here.val)); - state->mode = LIT; - break; - } - if (here.op & 32) { - Tracevv((stderr, "inflate: end of block\n")); - state->back = -1; - state->mode = TYPE; - break; - } - if (here.op & 64) { - strm->msg = (char *)"invalid literal/length code"; - state->mode = BAD; - break; - } - state->extra = (unsigned)(here.op) & 15; - state->mode = LENEXT; - case LENEXT: - if (state->extra) { - NEEDBITS(state->extra); - state->length += BITS(state->extra); - DROPBITS(state->extra); - state->back += state->extra; - } - Tracevv((stderr, "inflate: length %u\n", state->length)); - state->was = state->length; - state->mode = DIST; - case DIST: - for (;;) { - here = state->distcode[BITS(state->distbits)]; - if ((unsigned)(here.bits) <= bits) break; - PULLBYTE(); - } - if ((here.op & 0xf0) == 0) { - last = here; - for (;;) { - here = state->distcode[last.val + - (BITS(last.bits + last.op) >> last.bits)]; - if ((unsigned)(last.bits + here.bits) <= bits) break; - PULLBYTE(); - } - DROPBITS(last.bits); - state->back += last.bits; - } - DROPBITS(here.bits); - state->back += here.bits; - if (here.op & 64) { - strm->msg = (char *)"invalid distance code"; - state->mode = BAD; - break; - } - state->offset = (unsigned)here.val; - state->extra = (unsigned)(here.op) & 15; - state->mode = DISTEXT; - case DISTEXT: - if (state->extra) { - NEEDBITS(state->extra); - state->offset += BITS(state->extra); - DROPBITS(state->extra); - state->back += state->extra; - } -#ifdef INFLATE_STRICT - if (state->offset > state->dmax) { - strm->msg = (char *)"invalid distance too far back"; - state->mode = BAD; - break; - } -#endif - Tracevv((stderr, "inflate: distance %u\n", state->offset)); - state->mode = MATCH; - case MATCH: - if (left == 0) goto inf_leave; - copy = out - left; - if (state->offset > copy) { /* copy from window */ - copy = state->offset - copy; - if (copy > state->whave) { - if (state->sane) { - strm->msg = (char *)"invalid distance too far back"; - state->mode = BAD; - break; - } -#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR - Trace((stderr, "inflate.c too far\n")); - copy -= state->whave; - if (copy > state->length) copy = state->length; - if (copy > left) copy = left; - left -= copy; - state->length -= copy; - do { - *put++ = 0; - } while (--copy); - if (state->length == 0) state->mode = LEN; - break; -#endif - } - if (copy > state->wnext) { - copy -= state->wnext; - from = state->window + (state->wsize - copy); - } - else - from = state->window + (state->wnext - copy); - if (copy > state->length) copy = state->length; - } - else { /* copy from output */ - from = put - state->offset; - copy = state->length; - } - if (copy > left) copy = left; - left -= copy; - state->length -= copy; - do { - *put++ = *from++; - } while (--copy); - if (state->length == 0) state->mode = LEN; - break; - case LIT: - if (left == 0) goto inf_leave; - *put++ = (unsigned char)(state->length); - left--; - state->mode = LEN; - break; - case CHECK: - if (state->wrap) { - NEEDBITS(32); - out -= left; - strm->total_out += out; - state->total += out; - if ((state->wrap & 4) && out) - strm->adler = state->check = - UPDATE(state->check, put - out, out); - out = left; - if ((state->wrap & 4) && ( -#ifdef GUNZIP - state->flags ? hold : -#endif - ZSWAP32(hold)) != state->check) { - strm->msg = (char *)"incorrect data check"; - state->mode = BAD; - break; - } - INITBITS(); - Tracev((stderr, "inflate: check matches trailer\n")); - } -#ifdef GUNZIP - state->mode = LENGTH; - case LENGTH: - if (state->wrap && state->flags) { - NEEDBITS(32); - if (hold != (state->total & 0xffffffffUL)) { - strm->msg = (char *)"incorrect length check"; - state->mode = BAD; - break; - } - INITBITS(); - Tracev((stderr, "inflate: length matches trailer\n")); - } -#endif - state->mode = DONE; - case DONE: - ret = Z_STREAM_END; - goto inf_leave; - case BAD: - ret = Z_DATA_ERROR; - goto inf_leave; - case MEM: - return Z_MEM_ERROR; - case SYNC: - default: - return Z_STREAM_ERROR; - } - - /* - Return from inflate(), updating the total counts and the check value. - If there was no progress during the inflate() call, return a buffer - error. Call updatewindow() to create and/or update the window state. - Note: a memory error from inflate() is non-recoverable. - */ - inf_leave: - RESTORE(); - if (state->wsize || (out != strm->avail_out && state->mode < BAD && - (state->mode < CHECK || flush != Z_FINISH))) - if (updatewindow(strm, strm->next_out, out - strm->avail_out)) { - state->mode = MEM; - return Z_MEM_ERROR; - } - in -= strm->avail_in; - out -= strm->avail_out; - strm->total_in += in; - strm->total_out += out; - state->total += out; - if ((state->wrap & 4) && out) - strm->adler = state->check = - UPDATE(state->check, strm->next_out - out, out); - strm->data_type = (int)state->bits + (state->last ? 64 : 0) + - (state->mode == TYPE ? 128 : 0) + - (state->mode == LEN_ || state->mode == COPY_ ? 256 : 0); - if (((in == 0 && out == 0) || flush == Z_FINISH) && ret == Z_OK) - ret = Z_BUF_ERROR; - return ret; -} - -int ZEXPORT inflateEnd(strm) -z_streamp strm; -{ - struct inflate_state FAR *state; - if (inflateStateCheck(strm)) - return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; - if (state->window != Z_NULL) ZFREE(strm, state->window); - ZFREE(strm, strm->state); - strm->state = Z_NULL; - Tracev((stderr, "inflate: end\n")); - return Z_OK; -} - -int ZEXPORT inflateGetDictionary(strm, dictionary, dictLength) -z_streamp strm; -Bytef *dictionary; -uInt *dictLength; -{ - struct inflate_state FAR *state; - - /* check state */ - if (inflateStateCheck(strm)) return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; - - /* copy dictionary */ - if (state->whave && dictionary != Z_NULL) { - zmemcpy(dictionary, state->window + state->wnext, - state->whave - state->wnext); - zmemcpy(dictionary + state->whave - state->wnext, - state->window, state->wnext); - } - if (dictLength != Z_NULL) - *dictLength = state->whave; - return Z_OK; -} - -int ZEXPORT inflateSetDictionary(strm, dictionary, dictLength) -z_streamp strm; -const Bytef *dictionary; -uInt dictLength; -{ - struct inflate_state FAR *state; - unsigned long dictid; - int ret; - - /* check state */ - if (inflateStateCheck(strm)) return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; - if (state->wrap != 0 && state->mode != DICT) - return Z_STREAM_ERROR; - - /* check for correct dictionary identifier */ - if (state->mode == DICT) { - dictid = adler32(0L, Z_NULL, 0); - dictid = adler32(dictid, dictionary, dictLength); - if (dictid != state->check) - return Z_DATA_ERROR; - } - - /* copy dictionary to window using updatewindow(), which will amend the - existing dictionary if appropriate */ - ret = updatewindow(strm, dictionary + dictLength, dictLength); - if (ret) { - state->mode = MEM; - return Z_MEM_ERROR; - } - state->havedict = 1; - Tracev((stderr, "inflate: dictionary set\n")); - return Z_OK; -} - -int ZEXPORT inflateGetHeader(strm, head) -z_streamp strm; -gz_headerp head; -{ - struct inflate_state FAR *state; - - /* check state */ - if (inflateStateCheck(strm)) return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; - if ((state->wrap & 2) == 0) return Z_STREAM_ERROR; - - /* save header structure */ - state->head = head; - head->done = 0; - return Z_OK; -} - -/* - Search buf[0..len-1] for the pattern: 0, 0, 0xff, 0xff. Return when found - or when out of input. When called, *have is the number of pattern bytes - found in order so far, in 0..3. On return *have is updated to the new - state. If on return *have equals four, then the pattern was found and the - return value is how many bytes were read including the last byte of the - pattern. If *have is less than four, then the pattern has not been found - yet and the return value is len. In the latter case, syncsearch() can be - called again with more data and the *have state. *have is initialized to - zero for the first call. - */ -local unsigned syncsearch(have, buf, len) -unsigned FAR *have; -const unsigned char FAR *buf; -unsigned len; -{ - unsigned got; - unsigned next; - - got = *have; - next = 0; - while (next < len && got < 4) { - if ((int)(buf[next]) == (got < 2 ? 0 : 0xff)) - got++; - else if (buf[next]) - got = 0; - else - got = 4 - got; - next++; - } - *have = got; - return next; -} - -int ZEXPORT inflateSync(strm) -z_streamp strm; -{ - unsigned len; /* number of bytes to look at or looked at */ - unsigned long in, out; /* temporary to save total_in and total_out */ - unsigned char buf[4]; /* to restore bit buffer to byte string */ - struct inflate_state FAR *state; - - /* check parameters */ - if (inflateStateCheck(strm)) return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; - if (strm->avail_in == 0 && state->bits < 8) return Z_BUF_ERROR; - - /* if first time, start search in bit buffer */ - if (state->mode != SYNC) { - state->mode = SYNC; - state->hold <<= state->bits & 7; - state->bits -= state->bits & 7; - len = 0; - while (state->bits >= 8) { - buf[len++] = (unsigned char)(state->hold); - state->hold >>= 8; - state->bits -= 8; - } - state->have = 0; - syncsearch(&(state->have), buf, len); - } - - /* search available input */ - len = syncsearch(&(state->have), strm->next_in, strm->avail_in); - strm->avail_in -= len; - strm->next_in += len; - strm->total_in += len; - - /* return no joy or set up to restart inflate() on a new block */ - if (state->have != 4) return Z_DATA_ERROR; - in = strm->total_in; out = strm->total_out; - inflateReset(strm); - strm->total_in = in; strm->total_out = out; - state->mode = TYPE; - return Z_OK; -} - -/* - Returns true if inflate is currently at the end of a block generated by - Z_SYNC_FLUSH or Z_FULL_FLUSH. This function is used by one PPP - implementation to provide an additional safety check. PPP uses - Z_SYNC_FLUSH but removes the length bytes of the resulting empty stored - block. When decompressing, PPP checks that at the end of input packet, - inflate is waiting for these length bytes. - */ -int ZEXPORT inflateSyncPoint(strm) -z_streamp strm; -{ - struct inflate_state FAR *state; - - if (inflateStateCheck(strm)) return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; - return state->mode == STORED && state->bits == 0; -} - -int ZEXPORT inflateCopy(dest, source) -z_streamp dest; -z_streamp source; -{ - struct inflate_state FAR *state; - struct inflate_state FAR *copy; - unsigned char FAR *window; - unsigned wsize; - - /* check input */ - if (inflateStateCheck(source) || dest == Z_NULL) - return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)source->state; - - /* allocate space */ - copy = (struct inflate_state FAR *) - ZALLOC(source, 1, sizeof(struct inflate_state)); - if (copy == Z_NULL) return Z_MEM_ERROR; - window = Z_NULL; - if (state->window != Z_NULL) { - window = (unsigned char FAR *) - ZALLOC(source, 1U << state->wbits, sizeof(unsigned char)); - if (window == Z_NULL) { - ZFREE(source, copy); - return Z_MEM_ERROR; - } - } - - /* copy state */ - zmemcpy((voidpf)dest, (voidpf)source, sizeof(z_stream)); - zmemcpy((voidpf)copy, (voidpf)state, sizeof(struct inflate_state)); - copy->strm = dest; - if (state->lencode >= state->codes && - state->lencode <= state->codes + ENOUGH - 1) { - copy->lencode = copy->codes + (state->lencode - state->codes); - copy->distcode = copy->codes + (state->distcode - state->codes); - } - copy->next = copy->codes + (state->next - state->codes); - if (window != Z_NULL) { - wsize = 1U << state->wbits; - zmemcpy(window, state->window, wsize); - } - copy->window = window; - dest->state = (struct internal_state FAR *)copy; - return Z_OK; -} - -int ZEXPORT inflateUndermine(strm, subvert) -z_streamp strm; -int subvert; -{ - struct inflate_state FAR *state; - - if (inflateStateCheck(strm)) return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; -#ifdef INFLATE_ALLOW_INVALID_DISTANCE_TOOFAR_ARRR - state->sane = !subvert; - return Z_OK; -#else - (void)subvert; - state->sane = 1; - return Z_DATA_ERROR; -#endif -} - -int ZEXPORT inflateValidate(strm, check) -z_streamp strm; -int check; -{ - struct inflate_state FAR *state; - - if (inflateStateCheck(strm)) return Z_STREAM_ERROR; - state = (struct inflate_state FAR *)strm->state; - if (check) - state->wrap |= 4; - else - state->wrap &= ~4; - return Z_OK; -} - -long ZEXPORT inflateMark(strm) -z_streamp strm; -{ - struct inflate_state FAR *state; - - if (inflateStateCheck(strm)) - return -(1L << 16); - state = (struct inflate_state FAR *)strm->state; - return (long)(((unsigned long)((long)state->back)) << 16) + - (state->mode == COPY ? state->length : - (state->mode == MATCH ? state->was - state->length : 0)); -} - -unsigned long ZEXPORT inflateCodesUsed(strm) -z_streamp strm; -{ - struct inflate_state FAR *state; - if (inflateStateCheck(strm)) return (unsigned long)-1; - state = (struct inflate_state FAR *)strm->state; - return (unsigned long)(state->next - state->codes); -} diff --git a/dep/zlib/src/inflate.h b/dep/zlib/src/inflate.h deleted file mode 100644 index a46cce6b6..000000000 --- a/dep/zlib/src/inflate.h +++ /dev/null @@ -1,125 +0,0 @@ -/* inflate.h -- internal inflate state definition - * Copyright (C) 1995-2016 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* WARNING: this file should *not* be used by applications. It is - part of the implementation of the compression library and is - subject to change. Applications should only use zlib.h. - */ - -/* define NO_GZIP when compiling if you want to disable gzip header and - trailer decoding by inflate(). NO_GZIP would be used to avoid linking in - the crc code when it is not needed. For shared libraries, gzip decoding - should be left enabled. */ -#ifndef NO_GZIP -# define GUNZIP -#endif - -/* Possible inflate modes between inflate() calls */ -typedef enum { - HEAD = 16180, /* i: waiting for magic header */ - FLAGS, /* i: waiting for method and flags (gzip) */ - TIME, /* i: waiting for modification time (gzip) */ - OS, /* i: waiting for extra flags and operating system (gzip) */ - EXLEN, /* i: waiting for extra length (gzip) */ - EXTRA, /* i: waiting for extra bytes (gzip) */ - NAME, /* i: waiting for end of file name (gzip) */ - COMMENT, /* i: waiting for end of comment (gzip) */ - HCRC, /* i: waiting for header crc (gzip) */ - DICTID, /* i: waiting for dictionary check value */ - DICT, /* waiting for inflateSetDictionary() call */ - TYPE, /* i: waiting for type bits, including last-flag bit */ - TYPEDO, /* i: same, but skip check to exit inflate on new block */ - STORED, /* i: waiting for stored size (length and complement) */ - COPY_, /* i/o: same as COPY below, but only first time in */ - COPY, /* i/o: waiting for input or output to copy stored block */ - TABLE, /* i: waiting for dynamic block table lengths */ - LENLENS, /* i: waiting for code length code lengths */ - CODELENS, /* i: waiting for length/lit and distance code lengths */ - LEN_, /* i: same as LEN below, but only first time in */ - LEN, /* i: waiting for length/lit/eob code */ - LENEXT, /* i: waiting for length extra bits */ - DIST, /* i: waiting for distance code */ - DISTEXT, /* i: waiting for distance extra bits */ - MATCH, /* o: waiting for output space to copy string */ - LIT, /* o: waiting for output space to write literal */ - CHECK, /* i: waiting for 32-bit check value */ - LENGTH, /* i: waiting for 32-bit length (gzip) */ - DONE, /* finished check, done -- remain here until reset */ - BAD, /* got a data error -- remain here until reset */ - MEM, /* got an inflate() memory error -- remain here until reset */ - SYNC /* looking for synchronization bytes to restart inflate() */ -} inflate_mode; - -/* - State transitions between above modes - - - (most modes can go to BAD or MEM on error -- not shown for clarity) - - Process header: - HEAD -> (gzip) or (zlib) or (raw) - (gzip) -> FLAGS -> TIME -> OS -> EXLEN -> EXTRA -> NAME -> COMMENT -> - HCRC -> TYPE - (zlib) -> DICTID or TYPE - DICTID -> DICT -> TYPE - (raw) -> TYPEDO - Read deflate blocks: - TYPE -> TYPEDO -> STORED or TABLE or LEN_ or CHECK - STORED -> COPY_ -> COPY -> TYPE - TABLE -> LENLENS -> CODELENS -> LEN_ - LEN_ -> LEN - Read deflate codes in fixed or dynamic block: - LEN -> LENEXT or LIT or TYPE - LENEXT -> DIST -> DISTEXT -> MATCH -> LEN - LIT -> LEN - Process trailer: - CHECK -> LENGTH -> DONE - */ - -/* State maintained between inflate() calls -- approximately 7K bytes, not - including the allocated sliding window, which is up to 32K bytes. */ -struct inflate_state { - z_streamp strm; /* pointer back to this zlib stream */ - inflate_mode mode; /* current inflate mode */ - int last; /* true if processing last block */ - int wrap; /* bit 0 true for zlib, bit 1 true for gzip, - bit 2 true to validate check value */ - int havedict; /* true if dictionary provided */ - int flags; /* gzip header method and flags (0 if zlib) */ - unsigned dmax; /* zlib header max distance (INFLATE_STRICT) */ - unsigned long check; /* protected copy of check value */ - unsigned long total; /* protected copy of output count */ - gz_headerp head; /* where to save gzip header information */ - /* sliding window */ - unsigned wbits; /* log base 2 of requested window size */ - unsigned wsize; /* window size or zero if not using window */ - unsigned whave; /* valid bytes in the window */ - unsigned wnext; /* window write index */ - unsigned char FAR *window; /* allocated sliding window, if needed */ - /* bit accumulator */ - unsigned long hold; /* input bit accumulator */ - unsigned bits; /* number of bits in "in" */ - /* for string and stored block copying */ - unsigned length; /* literal or length of data to copy */ - unsigned offset; /* distance back to copy string from */ - /* for table and code decoding */ - unsigned extra; /* extra bits needed */ - /* fixed and dynamic code tables */ - code const FAR *lencode; /* starting table for length/literal codes */ - code const FAR *distcode; /* starting table for distance codes */ - unsigned lenbits; /* index bits for lencode */ - unsigned distbits; /* index bits for distcode */ - /* dynamic table building */ - unsigned ncode; /* number of code length code lengths */ - unsigned nlen; /* number of length code lengths */ - unsigned ndist; /* number of distance code lengths */ - unsigned have; /* number of code lengths in lens[] */ - code FAR *next; /* next available space in codes[] */ - unsigned short lens[320]; /* temporary storage for code lengths */ - unsigned short work[288]; /* work area for code table building */ - code codes[ENOUGH]; /* space for code tables */ - int sane; /* if false, allow invalid distance too far */ - int back; /* bits back of last unprocessed length/lit */ - unsigned was; /* initial length of match */ -}; diff --git a/dep/zlib/src/inftrees.c b/dep/zlib/src/inftrees.c deleted file mode 100644 index 2ea08fc13..000000000 --- a/dep/zlib/src/inftrees.c +++ /dev/null @@ -1,304 +0,0 @@ -/* inftrees.c -- generate Huffman trees for efficient decoding - * Copyright (C) 1995-2017 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -#include "zutil.h" -#include "inftrees.h" - -#define MAXBITS 15 - -const char inflate_copyright[] = - " inflate 1.2.11 Copyright 1995-2017 Mark Adler "; -/* - If you use the zlib library in a product, an acknowledgment is welcome - in the documentation of your product. If for some reason you cannot - include such an acknowledgment, I would appreciate that you keep this - copyright string in the executable of your product. - */ - -/* - Build a set of tables to decode the provided canonical Huffman code. - The code lengths are lens[0..codes-1]. The result starts at *table, - whose indices are 0..2^bits-1. work is a writable array of at least - lens shorts, which is used as a work area. type is the type of code - to be generated, CODES, LENS, or DISTS. On return, zero is success, - -1 is an invalid code, and +1 means that ENOUGH isn't enough. table - on return points to the next available entry's address. bits is the - requested root table index bits, and on return it is the actual root - table index bits. It will differ if the request is greater than the - longest code or if it is less than the shortest code. - */ -int ZLIB_INTERNAL inflate_table(type, lens, codes, table, bits, work) -codetype type; -unsigned short FAR *lens; -unsigned codes; -code FAR * FAR *table; -unsigned FAR *bits; -unsigned short FAR *work; -{ - unsigned len; /* a code's length in bits */ - unsigned sym; /* index of code symbols */ - unsigned min, max; /* minimum and maximum code lengths */ - unsigned root; /* number of index bits for root table */ - unsigned curr; /* number of index bits for current table */ - unsigned drop; /* code bits to drop for sub-table */ - int left; /* number of prefix codes available */ - unsigned used; /* code entries in table used */ - unsigned huff; /* Huffman code */ - unsigned incr; /* for incrementing code, index */ - unsigned fill; /* index for replicating entries */ - unsigned low; /* low bits for current root entry */ - unsigned mask; /* mask for low root bits */ - code here; /* table entry for duplication */ - code FAR *next; /* next available space in table */ - const unsigned short FAR *base; /* base value table to use */ - const unsigned short FAR *extra; /* extra bits table to use */ - unsigned match; /* use base and extra for symbol >= match */ - unsigned short count[MAXBITS+1]; /* number of codes of each length */ - unsigned short offs[MAXBITS+1]; /* offsets in table for each length */ - static const unsigned short lbase[31] = { /* Length codes 257..285 base */ - 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 15, 17, 19, 23, 27, 31, - 35, 43, 51, 59, 67, 83, 99, 115, 131, 163, 195, 227, 258, 0, 0}; - static const unsigned short lext[31] = { /* Length codes 257..285 extra */ - 16, 16, 16, 16, 16, 16, 16, 16, 17, 17, 17, 17, 18, 18, 18, 18, - 19, 19, 19, 19, 20, 20, 20, 20, 21, 21, 21, 21, 16, 77, 202}; - static const unsigned short dbase[32] = { /* Distance codes 0..29 base */ - 1, 2, 3, 4, 5, 7, 9, 13, 17, 25, 33, 49, 65, 97, 129, 193, - 257, 385, 513, 769, 1025, 1537, 2049, 3073, 4097, 6145, - 8193, 12289, 16385, 24577, 0, 0}; - static const unsigned short dext[32] = { /* Distance codes 0..29 extra */ - 16, 16, 16, 16, 17, 17, 18, 18, 19, 19, 20, 20, 21, 21, 22, 22, - 23, 23, 24, 24, 25, 25, 26, 26, 27, 27, - 28, 28, 29, 29, 64, 64}; - - /* - Process a set of code lengths to create a canonical Huffman code. The - code lengths are lens[0..codes-1]. Each length corresponds to the - symbols 0..codes-1. The Huffman code is generated by first sorting the - symbols by length from short to long, and retaining the symbol order - for codes with equal lengths. Then the code starts with all zero bits - for the first code of the shortest length, and the codes are integer - increments for the same length, and zeros are appended as the length - increases. For the deflate format, these bits are stored backwards - from their more natural integer increment ordering, and so when the - decoding tables are built in the large loop below, the integer codes - are incremented backwards. - - This routine assumes, but does not check, that all of the entries in - lens[] are in the range 0..MAXBITS. The caller must assure this. - 1..MAXBITS is interpreted as that code length. zero means that that - symbol does not occur in this code. - - The codes are sorted by computing a count of codes for each length, - creating from that a table of starting indices for each length in the - sorted table, and then entering the symbols in order in the sorted - table. The sorted table is work[], with that space being provided by - the caller. - - The length counts are used for other purposes as well, i.e. finding - the minimum and maximum length codes, determining if there are any - codes at all, checking for a valid set of lengths, and looking ahead - at length counts to determine sub-table sizes when building the - decoding tables. - */ - - /* accumulate lengths for codes (assumes lens[] all in 0..MAXBITS) */ - for (len = 0; len <= MAXBITS; len++) - count[len] = 0; - for (sym = 0; sym < codes; sym++) - count[lens[sym]]++; - - /* bound code lengths, force root to be within code lengths */ - root = *bits; - for (max = MAXBITS; max >= 1; max--) - if (count[max] != 0) break; - if (root > max) root = max; - if (max == 0) { /* no symbols to code at all */ - here.op = (unsigned char)64; /* invalid code marker */ - here.bits = (unsigned char)1; - here.val = (unsigned short)0; - *(*table)++ = here; /* make a table to force an error */ - *(*table)++ = here; - *bits = 1; - return 0; /* no symbols, but wait for decoding to report error */ - } - for (min = 1; min < max; min++) - if (count[min] != 0) break; - if (root < min) root = min; - - /* check for an over-subscribed or incomplete set of lengths */ - left = 1; - for (len = 1; len <= MAXBITS; len++) { - left <<= 1; - left -= count[len]; - if (left < 0) return -1; /* over-subscribed */ - } - if (left > 0 && (type == CODES || max != 1)) - return -1; /* incomplete set */ - - /* generate offsets into symbol table for each length for sorting */ - offs[1] = 0; - for (len = 1; len < MAXBITS; len++) - offs[len + 1] = offs[len] + count[len]; - - /* sort symbols by length, by symbol order within each length */ - for (sym = 0; sym < codes; sym++) - if (lens[sym] != 0) work[offs[lens[sym]]++] = (unsigned short)sym; - - /* - Create and fill in decoding tables. In this loop, the table being - filled is at next and has curr index bits. The code being used is huff - with length len. That code is converted to an index by dropping drop - bits off of the bottom. For codes where len is less than drop + curr, - those top drop + curr - len bits are incremented through all values to - fill the table with replicated entries. - - root is the number of index bits for the root table. When len exceeds - root, sub-tables are created pointed to by the root entry with an index - of the low root bits of huff. This is saved in low to check for when a - new sub-table should be started. drop is zero when the root table is - being filled, and drop is root when sub-tables are being filled. - - When a new sub-table is needed, it is necessary to look ahead in the - code lengths to determine what size sub-table is needed. The length - counts are used for this, and so count[] is decremented as codes are - entered in the tables. - - used keeps track of how many table entries have been allocated from the - provided *table space. It is checked for LENS and DIST tables against - the constants ENOUGH_LENS and ENOUGH_DISTS to guard against changes in - the initial root table size constants. See the comments in inftrees.h - for more information. - - sym increments through all symbols, and the loop terminates when - all codes of length max, i.e. all codes, have been processed. This - routine permits incomplete codes, so another loop after this one fills - in the rest of the decoding tables with invalid code markers. - */ - - /* set up for code type */ - switch (type) { - case CODES: - base = extra = work; /* dummy value--not used */ - match = 20; - break; - case LENS: - base = lbase; - extra = lext; - match = 257; - break; - default: /* DISTS */ - base = dbase; - extra = dext; - match = 0; - } - - /* initialize state for loop */ - huff = 0; /* starting code */ - sym = 0; /* starting code symbol */ - len = min; /* starting code length */ - next = *table; /* current table to fill in */ - curr = root; /* current table index bits */ - drop = 0; /* current bits to drop from code for index */ - low = (unsigned)(-1); /* trigger new sub-table when len > root */ - used = 1U << root; /* use root table entries */ - mask = used - 1; /* mask for comparing low */ - - /* check available table space */ - if ((type == LENS && used > ENOUGH_LENS) || - (type == DISTS && used > ENOUGH_DISTS)) - return 1; - - /* process all codes and make table entries */ - for (;;) { - /* create table entry */ - here.bits = (unsigned char)(len - drop); - if (work[sym] + 1U < match) { - here.op = (unsigned char)0; - here.val = work[sym]; - } - else if (work[sym] >= match) { - here.op = (unsigned char)(extra[work[sym] - match]); - here.val = base[work[sym] - match]; - } - else { - here.op = (unsigned char)(32 + 64); /* end of block */ - here.val = 0; - } - - /* replicate for those indices with low len bits equal to huff */ - incr = 1U << (len - drop); - fill = 1U << curr; - min = fill; /* save offset to next table */ - do { - fill -= incr; - next[(huff >> drop) + fill] = here; - } while (fill != 0); - - /* backwards increment the len-bit code huff */ - incr = 1U << (len - 1); - while (huff & incr) - incr >>= 1; - if (incr != 0) { - huff &= incr - 1; - huff += incr; - } - else - huff = 0; - - /* go to next symbol, update count, len */ - sym++; - if (--(count[len]) == 0) { - if (len == max) break; - len = lens[work[sym]]; - } - - /* create new sub-table if needed */ - if (len > root && (huff & mask) != low) { - /* if first time, transition to sub-tables */ - if (drop == 0) - drop = root; - - /* increment past last table */ - next += min; /* here min is 1 << curr */ - - /* determine length of next table */ - curr = len - drop; - left = (int)(1 << curr); - while (curr + drop < max) { - left -= count[curr + drop]; - if (left <= 0) break; - curr++; - left <<= 1; - } - - /* check for enough space */ - used += 1U << curr; - if ((type == LENS && used > ENOUGH_LENS) || - (type == DISTS && used > ENOUGH_DISTS)) - return 1; - - /* point entry in root table to sub-table */ - low = huff & mask; - (*table)[low].op = (unsigned char)curr; - (*table)[low].bits = (unsigned char)root; - (*table)[low].val = (unsigned short)(next - *table); - } - } - - /* fill in remaining table entry if code is incomplete (guaranteed to have - at most one remaining entry, since if the code is incomplete, the - maximum code length that was allowed to get this far is one bit) */ - if (huff != 0) { - here.op = (unsigned char)64; /* invalid code marker */ - here.bits = (unsigned char)(len - drop); - here.val = (unsigned short)0; - next[huff] = here; - } - - /* set return parameters */ - *table += used; - *bits = root; - return 0; -} diff --git a/dep/zlib/src/inftrees.h b/dep/zlib/src/inftrees.h deleted file mode 100644 index baa53a0b1..000000000 --- a/dep/zlib/src/inftrees.h +++ /dev/null @@ -1,62 +0,0 @@ -/* inftrees.h -- header to use inftrees.c - * Copyright (C) 1995-2005, 2010 Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* WARNING: this file should *not* be used by applications. It is - part of the implementation of the compression library and is - subject to change. Applications should only use zlib.h. - */ - -/* Structure for decoding tables. Each entry provides either the - information needed to do the operation requested by the code that - indexed that table entry, or it provides a pointer to another - table that indexes more bits of the code. op indicates whether - the entry is a pointer to another table, a literal, a length or - distance, an end-of-block, or an invalid code. For a table - pointer, the low four bits of op is the number of index bits of - that table. For a length or distance, the low four bits of op - is the number of extra bits to get after the code. bits is - the number of bits in this code or part of the code to drop off - of the bit buffer. val is the actual byte to output in the case - of a literal, the base length or distance, or the offset from - the current table to the next table. Each entry is four bytes. */ -typedef struct { - unsigned char op; /* operation, extra bits, table bits */ - unsigned char bits; /* bits in this part of the code */ - unsigned short val; /* offset in table or code value */ -} code; - -/* op values as set by inflate_table(): - 00000000 - literal - 0000tttt - table link, tttt != 0 is the number of table index bits - 0001eeee - length or distance, eeee is the number of extra bits - 01100000 - end of block - 01000000 - invalid code - */ - -/* Maximum size of the dynamic table. The maximum number of code structures is - 1444, which is the sum of 852 for literal/length codes and 592 for distance - codes. These values were found by exhaustive searches using the program - examples/enough.c found in the zlib distribtution. The arguments to that - program are the number of symbols, the initial root table size, and the - maximum bit length of a code. "enough 286 9 15" for literal/length codes - returns returns 852, and "enough 30 6 15" for distance codes returns 592. - The initial root table size (9 or 6) is found in the fifth argument of the - inflate_table() calls in inflate.c and infback.c. If the root table size is - changed, then these maximum sizes would be need to be recalculated and - updated. */ -#define ENOUGH_LENS 852 -#define ENOUGH_DISTS 592 -#define ENOUGH (ENOUGH_LENS+ENOUGH_DISTS) - -/* Type of code to build for inflate_table() */ -typedef enum { - CODES, - LENS, - DISTS -} codetype; - -int ZLIB_INTERNAL inflate_table OF((codetype type, unsigned short FAR *lens, - unsigned codes, code FAR * FAR *table, - unsigned FAR *bits, unsigned short FAR *work)); diff --git a/dep/zlib/src/trees.c b/dep/zlib/src/trees.c deleted file mode 100644 index 50cf4b457..000000000 --- a/dep/zlib/src/trees.c +++ /dev/null @@ -1,1203 +0,0 @@ -/* trees.c -- output deflated data using Huffman coding - * Copyright (C) 1995-2017 Jean-loup Gailly - * detect_data_type() function provided freely by Cosmin Truta, 2006 - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* - * ALGORITHM - * - * The "deflation" process uses several Huffman trees. The more - * common source values are represented by shorter bit sequences. - * - * Each code tree is stored in a compressed form which is itself - * a Huffman encoding of the lengths of all the code strings (in - * ascending order by source values). The actual code strings are - * reconstructed from the lengths in the inflate process, as described - * in the deflate specification. - * - * REFERENCES - * - * Deutsch, L.P.,"'Deflate' Compressed Data Format Specification". - * Available in ftp.uu.net:/pub/archiving/zip/doc/deflate-1.1.doc - * - * Storer, James A. - * Data Compression: Methods and Theory, pp. 49-50. - * Computer Science Press, 1988. ISBN 0-7167-8156-5. - * - * Sedgewick, R. - * Algorithms, p290. - * Addison-Wesley, 1983. ISBN 0-201-06672-6. - */ - -/* @(#) $Id$ */ - -/* #define GEN_TREES_H */ - -#include "deflate.h" - -#ifdef ZLIB_DEBUG -# include -#endif - -/* =========================================================================== - * Constants - */ - -#define MAX_BL_BITS 7 -/* Bit length codes must not exceed MAX_BL_BITS bits */ - -#define END_BLOCK 256 -/* end of block literal code */ - -#define REP_3_6 16 -/* repeat previous bit length 3-6 times (2 bits of repeat count) */ - -#define REPZ_3_10 17 -/* repeat a zero length 3-10 times (3 bits of repeat count) */ - -#define REPZ_11_138 18 -/* repeat a zero length 11-138 times (7 bits of repeat count) */ - -local const int extra_lbits[LENGTH_CODES] /* extra bits for each length code */ - = {0,0,0,0,0,0,0,0,1,1,1,1,2,2,2,2,3,3,3,3,4,4,4,4,5,5,5,5,0}; - -local const int extra_dbits[D_CODES] /* extra bits for each distance code */ - = {0,0,0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7,8,8,9,9,10,10,11,11,12,12,13,13}; - -local const int extra_blbits[BL_CODES]/* extra bits for each bit length code */ - = {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,2,3,7}; - -local const uch bl_order[BL_CODES] - = {16,17,18,0,8,7,9,6,10,5,11,4,12,3,13,2,14,1,15}; -/* The lengths of the bit length codes are sent in order of decreasing - * probability, to avoid transmitting the lengths for unused bit length codes. - */ - -/* =========================================================================== - * Local data. These are initialized only once. - */ - -#define DIST_CODE_LEN 512 /* see definition of array dist_code below */ - -#if defined(GEN_TREES_H) || !defined(STDC) -/* non ANSI compilers may not accept trees.h */ - -local ct_data static_ltree[L_CODES+2]; -/* The static literal tree. Since the bit lengths are imposed, there is no - * need for the L_CODES extra codes used during heap construction. However - * The codes 286 and 287 are needed to build a canonical tree (see _tr_init - * below). - */ - -local ct_data static_dtree[D_CODES]; -/* The static distance tree. (Actually a trivial tree since all codes use - * 5 bits.) - */ - -uch _dist_code[DIST_CODE_LEN]; -/* Distance codes. The first 256 values correspond to the distances - * 3 .. 258, the last 256 values correspond to the top 8 bits of - * the 15 bit distances. - */ - -uch _length_code[MAX_MATCH-MIN_MATCH+1]; -/* length code for each normalized match length (0 == MIN_MATCH) */ - -local int base_length[LENGTH_CODES]; -/* First normalized length for each code (0 = MIN_MATCH) */ - -local int base_dist[D_CODES]; -/* First normalized distance for each code (0 = distance of 1) */ - -#else -# include "trees.h" -#endif /* GEN_TREES_H */ - -struct static_tree_desc_s { - const ct_data *static_tree; /* static tree or NULL */ - const intf *extra_bits; /* extra bits for each code or NULL */ - int extra_base; /* base index for extra_bits */ - int elems; /* max number of elements in the tree */ - int max_length; /* max bit length for the codes */ -}; - -local const static_tree_desc static_l_desc = -{static_ltree, extra_lbits, LITERALS+1, L_CODES, MAX_BITS}; - -local const static_tree_desc static_d_desc = -{static_dtree, extra_dbits, 0, D_CODES, MAX_BITS}; - -local const static_tree_desc static_bl_desc = -{(const ct_data *)0, extra_blbits, 0, BL_CODES, MAX_BL_BITS}; - -/* =========================================================================== - * Local (static) routines in this file. - */ - -local void tr_static_init OF((void)); -local void init_block OF((deflate_state *s)); -local void pqdownheap OF((deflate_state *s, ct_data *tree, int k)); -local void gen_bitlen OF((deflate_state *s, tree_desc *desc)); -local void gen_codes OF((ct_data *tree, int max_code, ushf *bl_count)); -local void build_tree OF((deflate_state *s, tree_desc *desc)); -local void scan_tree OF((deflate_state *s, ct_data *tree, int max_code)); -local void send_tree OF((deflate_state *s, ct_data *tree, int max_code)); -local int build_bl_tree OF((deflate_state *s)); -local void send_all_trees OF((deflate_state *s, int lcodes, int dcodes, - int blcodes)); -local void compress_block OF((deflate_state *s, const ct_data *ltree, - const ct_data *dtree)); -local int detect_data_type OF((deflate_state *s)); -local unsigned bi_reverse OF((unsigned value, int length)); -local void bi_windup OF((deflate_state *s)); -local void bi_flush OF((deflate_state *s)); - -#ifdef GEN_TREES_H -local void gen_trees_header OF((void)); -#endif - -#ifndef ZLIB_DEBUG -# define send_code(s, c, tree) send_bits(s, tree[c].Code, tree[c].Len) - /* Send a code of the given tree. c and tree must not have side effects */ - -#else /* !ZLIB_DEBUG */ -# define send_code(s, c, tree) \ - { if (z_verbose>2) fprintf(stderr,"\ncd %3d ",(c)); \ - send_bits(s, tree[c].Code, tree[c].Len); } -#endif - -/* =========================================================================== - * Output a short LSB first on the stream. - * IN assertion: there is enough room in pendingBuf. - */ -#define put_short(s, w) { \ - put_byte(s, (uch)((w) & 0xff)); \ - put_byte(s, (uch)((ush)(w) >> 8)); \ -} - -/* =========================================================================== - * Send a value on a given number of bits. - * IN assertion: length <= 16 and value fits in length bits. - */ -#ifdef ZLIB_DEBUG -local void send_bits OF((deflate_state *s, int value, int length)); - -local void send_bits(s, value, length) - deflate_state *s; - int value; /* value to send */ - int length; /* number of bits */ -{ - Tracevv((stderr," l %2d v %4x ", length, value)); - Assert(length > 0 && length <= 15, "invalid length"); - s->bits_sent += (ulg)length; - - /* If not enough room in bi_buf, use (valid) bits from bi_buf and - * (16 - bi_valid) bits from value, leaving (width - (16-bi_valid)) - * unused bits in value. - */ - if (s->bi_valid > (int)Buf_size - length) { - s->bi_buf |= (ush)value << s->bi_valid; - put_short(s, s->bi_buf); - s->bi_buf = (ush)value >> (Buf_size - s->bi_valid); - s->bi_valid += length - Buf_size; - } else { - s->bi_buf |= (ush)value << s->bi_valid; - s->bi_valid += length; - } -} -#else /* !ZLIB_DEBUG */ - -#define send_bits(s, value, length) \ -{ int len = length;\ - if (s->bi_valid > (int)Buf_size - len) {\ - int val = (int)value;\ - s->bi_buf |= (ush)val << s->bi_valid;\ - put_short(s, s->bi_buf);\ - s->bi_buf = (ush)val >> (Buf_size - s->bi_valid);\ - s->bi_valid += len - Buf_size;\ - } else {\ - s->bi_buf |= (ush)(value) << s->bi_valid;\ - s->bi_valid += len;\ - }\ -} -#endif /* ZLIB_DEBUG */ - - -/* the arguments must not have side effects */ - -/* =========================================================================== - * Initialize the various 'constant' tables. - */ -local void tr_static_init() -{ -#if defined(GEN_TREES_H) || !defined(STDC) - static int static_init_done = 0; - int n; /* iterates over tree elements */ - int bits; /* bit counter */ - int length; /* length value */ - int code; /* code value */ - int dist; /* distance index */ - ush bl_count[MAX_BITS+1]; - /* number of codes at each bit length for an optimal tree */ - - if (static_init_done) return; - - /* For some embedded targets, global variables are not initialized: */ -#ifdef NO_INIT_GLOBAL_POINTERS - static_l_desc.static_tree = static_ltree; - static_l_desc.extra_bits = extra_lbits; - static_d_desc.static_tree = static_dtree; - static_d_desc.extra_bits = extra_dbits; - static_bl_desc.extra_bits = extra_blbits; -#endif - - /* Initialize the mapping length (0..255) -> length code (0..28) */ - length = 0; - for (code = 0; code < LENGTH_CODES-1; code++) { - base_length[code] = length; - for (n = 0; n < (1< dist code (0..29) */ - dist = 0; - for (code = 0 ; code < 16; code++) { - base_dist[code] = dist; - for (n = 0; n < (1<>= 7; /* from now on, all distances are divided by 128 */ - for ( ; code < D_CODES; code++) { - base_dist[code] = dist << 7; - for (n = 0; n < (1<<(extra_dbits[code]-7)); n++) { - _dist_code[256 + dist++] = (uch)code; - } - } - Assert (dist == 256, "tr_static_init: 256+dist != 512"); - - /* Construct the codes of the static literal tree */ - for (bits = 0; bits <= MAX_BITS; bits++) bl_count[bits] = 0; - n = 0; - while (n <= 143) static_ltree[n++].Len = 8, bl_count[8]++; - while (n <= 255) static_ltree[n++].Len = 9, bl_count[9]++; - while (n <= 279) static_ltree[n++].Len = 7, bl_count[7]++; - while (n <= 287) static_ltree[n++].Len = 8, bl_count[8]++; - /* Codes 286 and 287 do not exist, but we must include them in the - * tree construction to get a canonical Huffman tree (longest code - * all ones) - */ - gen_codes((ct_data *)static_ltree, L_CODES+1, bl_count); - - /* The static distance tree is trivial: */ - for (n = 0; n < D_CODES; n++) { - static_dtree[n].Len = 5; - static_dtree[n].Code = bi_reverse((unsigned)n, 5); - } - static_init_done = 1; - -# ifdef GEN_TREES_H - gen_trees_header(); -# endif -#endif /* defined(GEN_TREES_H) || !defined(STDC) */ -} - -/* =========================================================================== - * Genererate the file trees.h describing the static trees. - */ -#ifdef GEN_TREES_H -# ifndef ZLIB_DEBUG -# include -# endif - -# define SEPARATOR(i, last, width) \ - ((i) == (last)? "\n};\n\n" : \ - ((i) % (width) == (width)-1 ? ",\n" : ", ")) - -void gen_trees_header() -{ - FILE *header = fopen("trees.h", "w"); - int i; - - Assert (header != NULL, "Can't open trees.h"); - fprintf(header, - "/* header created automatically with -DGEN_TREES_H */\n\n"); - - fprintf(header, "local const ct_data static_ltree[L_CODES+2] = {\n"); - for (i = 0; i < L_CODES+2; i++) { - fprintf(header, "{{%3u},{%3u}}%s", static_ltree[i].Code, - static_ltree[i].Len, SEPARATOR(i, L_CODES+1, 5)); - } - - fprintf(header, "local const ct_data static_dtree[D_CODES] = {\n"); - for (i = 0; i < D_CODES; i++) { - fprintf(header, "{{%2u},{%2u}}%s", static_dtree[i].Code, - static_dtree[i].Len, SEPARATOR(i, D_CODES-1, 5)); - } - - fprintf(header, "const uch ZLIB_INTERNAL _dist_code[DIST_CODE_LEN] = {\n"); - for (i = 0; i < DIST_CODE_LEN; i++) { - fprintf(header, "%2u%s", _dist_code[i], - SEPARATOR(i, DIST_CODE_LEN-1, 20)); - } - - fprintf(header, - "const uch ZLIB_INTERNAL _length_code[MAX_MATCH-MIN_MATCH+1]= {\n"); - for (i = 0; i < MAX_MATCH-MIN_MATCH+1; i++) { - fprintf(header, "%2u%s", _length_code[i], - SEPARATOR(i, MAX_MATCH-MIN_MATCH, 20)); - } - - fprintf(header, "local const int base_length[LENGTH_CODES] = {\n"); - for (i = 0; i < LENGTH_CODES; i++) { - fprintf(header, "%1u%s", base_length[i], - SEPARATOR(i, LENGTH_CODES-1, 20)); - } - - fprintf(header, "local const int base_dist[D_CODES] = {\n"); - for (i = 0; i < D_CODES; i++) { - fprintf(header, "%5u%s", base_dist[i], - SEPARATOR(i, D_CODES-1, 10)); - } - - fclose(header); -} -#endif /* GEN_TREES_H */ - -/* =========================================================================== - * Initialize the tree data structures for a new zlib stream. - */ -void ZLIB_INTERNAL _tr_init(s) - deflate_state *s; -{ - tr_static_init(); - - s->l_desc.dyn_tree = s->dyn_ltree; - s->l_desc.stat_desc = &static_l_desc; - - s->d_desc.dyn_tree = s->dyn_dtree; - s->d_desc.stat_desc = &static_d_desc; - - s->bl_desc.dyn_tree = s->bl_tree; - s->bl_desc.stat_desc = &static_bl_desc; - - s->bi_buf = 0; - s->bi_valid = 0; -#ifdef ZLIB_DEBUG - s->compressed_len = 0L; - s->bits_sent = 0L; -#endif - - /* Initialize the first block of the first file: */ - init_block(s); -} - -/* =========================================================================== - * Initialize a new block. - */ -local void init_block(s) - deflate_state *s; -{ - int n; /* iterates over tree elements */ - - /* Initialize the trees. */ - for (n = 0; n < L_CODES; n++) s->dyn_ltree[n].Freq = 0; - for (n = 0; n < D_CODES; n++) s->dyn_dtree[n].Freq = 0; - for (n = 0; n < BL_CODES; n++) s->bl_tree[n].Freq = 0; - - s->dyn_ltree[END_BLOCK].Freq = 1; - s->opt_len = s->static_len = 0L; - s->last_lit = s->matches = 0; -} - -#define SMALLEST 1 -/* Index within the heap array of least frequent node in the Huffman tree */ - - -/* =========================================================================== - * Remove the smallest element from the heap and recreate the heap with - * one less element. Updates heap and heap_len. - */ -#define pqremove(s, tree, top) \ -{\ - top = s->heap[SMALLEST]; \ - s->heap[SMALLEST] = s->heap[s->heap_len--]; \ - pqdownheap(s, tree, SMALLEST); \ -} - -/* =========================================================================== - * Compares to subtrees, using the tree depth as tie breaker when - * the subtrees have equal frequency. This minimizes the worst case length. - */ -#define smaller(tree, n, m, depth) \ - (tree[n].Freq < tree[m].Freq || \ - (tree[n].Freq == tree[m].Freq && depth[n] <= depth[m])) - -/* =========================================================================== - * Restore the heap property by moving down the tree starting at node k, - * exchanging a node with the smallest of its two sons if necessary, stopping - * when the heap property is re-established (each father smaller than its - * two sons). - */ -local void pqdownheap(s, tree, k) - deflate_state *s; - ct_data *tree; /* the tree to restore */ - int k; /* node to move down */ -{ - int v = s->heap[k]; - int j = k << 1; /* left son of k */ - while (j <= s->heap_len) { - /* Set j to the smallest of the two sons: */ - if (j < s->heap_len && - smaller(tree, s->heap[j+1], s->heap[j], s->depth)) { - j++; - } - /* Exit if v is smaller than both sons */ - if (smaller(tree, v, s->heap[j], s->depth)) break; - - /* Exchange v with the smallest son */ - s->heap[k] = s->heap[j]; k = j; - - /* And continue down the tree, setting j to the left son of k */ - j <<= 1; - } - s->heap[k] = v; -} - -/* =========================================================================== - * Compute the optimal bit lengths for a tree and update the total bit length - * for the current block. - * IN assertion: the fields freq and dad are set, heap[heap_max] and - * above are the tree nodes sorted by increasing frequency. - * OUT assertions: the field len is set to the optimal bit length, the - * array bl_count contains the frequencies for each bit length. - * The length opt_len is updated; static_len is also updated if stree is - * not null. - */ -local void gen_bitlen(s, desc) - deflate_state *s; - tree_desc *desc; /* the tree descriptor */ -{ - ct_data *tree = desc->dyn_tree; - int max_code = desc->max_code; - const ct_data *stree = desc->stat_desc->static_tree; - const intf *extra = desc->stat_desc->extra_bits; - int base = desc->stat_desc->extra_base; - int max_length = desc->stat_desc->max_length; - int h; /* heap index */ - int n, m; /* iterate over the tree elements */ - int bits; /* bit length */ - int xbits; /* extra bits */ - ush f; /* frequency */ - int overflow = 0; /* number of elements with bit length too large */ - - for (bits = 0; bits <= MAX_BITS; bits++) s->bl_count[bits] = 0; - - /* In a first pass, compute the optimal bit lengths (which may - * overflow in the case of the bit length tree). - */ - tree[s->heap[s->heap_max]].Len = 0; /* root of the heap */ - - for (h = s->heap_max+1; h < HEAP_SIZE; h++) { - n = s->heap[h]; - bits = tree[tree[n].Dad].Len + 1; - if (bits > max_length) bits = max_length, overflow++; - tree[n].Len = (ush)bits; - /* We overwrite tree[n].Dad which is no longer needed */ - - if (n > max_code) continue; /* not a leaf node */ - - s->bl_count[bits]++; - xbits = 0; - if (n >= base) xbits = extra[n-base]; - f = tree[n].Freq; - s->opt_len += (ulg)f * (unsigned)(bits + xbits); - if (stree) s->static_len += (ulg)f * (unsigned)(stree[n].Len + xbits); - } - if (overflow == 0) return; - - Tracev((stderr,"\nbit length overflow\n")); - /* This happens for example on obj2 and pic of the Calgary corpus */ - - /* Find the first bit length which could increase: */ - do { - bits = max_length-1; - while (s->bl_count[bits] == 0) bits--; - s->bl_count[bits]--; /* move one leaf down the tree */ - s->bl_count[bits+1] += 2; /* move one overflow item as its brother */ - s->bl_count[max_length]--; - /* The brother of the overflow item also moves one step up, - * but this does not affect bl_count[max_length] - */ - overflow -= 2; - } while (overflow > 0); - - /* Now recompute all bit lengths, scanning in increasing frequency. - * h is still equal to HEAP_SIZE. (It is simpler to reconstruct all - * lengths instead of fixing only the wrong ones. This idea is taken - * from 'ar' written by Haruhiko Okumura.) - */ - for (bits = max_length; bits != 0; bits--) { - n = s->bl_count[bits]; - while (n != 0) { - m = s->heap[--h]; - if (m > max_code) continue; - if ((unsigned) tree[m].Len != (unsigned) bits) { - Tracev((stderr,"code %d bits %d->%d\n", m, tree[m].Len, bits)); - s->opt_len += ((ulg)bits - tree[m].Len) * tree[m].Freq; - tree[m].Len = (ush)bits; - } - n--; - } - } -} - -/* =========================================================================== - * Generate the codes for a given tree and bit counts (which need not be - * optimal). - * IN assertion: the array bl_count contains the bit length statistics for - * the given tree and the field len is set for all tree elements. - * OUT assertion: the field code is set for all tree elements of non - * zero code length. - */ -local void gen_codes (tree, max_code, bl_count) - ct_data *tree; /* the tree to decorate */ - int max_code; /* largest code with non zero frequency */ - ushf *bl_count; /* number of codes at each bit length */ -{ - ush next_code[MAX_BITS+1]; /* next code value for each bit length */ - unsigned code = 0; /* running code value */ - int bits; /* bit index */ - int n; /* code index */ - - /* The distribution counts are first used to generate the code values - * without bit reversal. - */ - for (bits = 1; bits <= MAX_BITS; bits++) { - code = (code + bl_count[bits-1]) << 1; - next_code[bits] = (ush)code; - } - /* Check that the bit counts in bl_count are consistent. The last code - * must be all ones. - */ - Assert (code + bl_count[MAX_BITS]-1 == (1<dyn_tree; - const ct_data *stree = desc->stat_desc->static_tree; - int elems = desc->stat_desc->elems; - int n, m; /* iterate over heap elements */ - int max_code = -1; /* largest code with non zero frequency */ - int node; /* new node being created */ - - /* Construct the initial heap, with least frequent element in - * heap[SMALLEST]. The sons of heap[n] are heap[2*n] and heap[2*n+1]. - * heap[0] is not used. - */ - s->heap_len = 0, s->heap_max = HEAP_SIZE; - - for (n = 0; n < elems; n++) { - if (tree[n].Freq != 0) { - s->heap[++(s->heap_len)] = max_code = n; - s->depth[n] = 0; - } else { - tree[n].Len = 0; - } - } - - /* The pkzip format requires that at least one distance code exists, - * and that at least one bit should be sent even if there is only one - * possible code. So to avoid special checks later on we force at least - * two codes of non zero frequency. - */ - while (s->heap_len < 2) { - node = s->heap[++(s->heap_len)] = (max_code < 2 ? ++max_code : 0); - tree[node].Freq = 1; - s->depth[node] = 0; - s->opt_len--; if (stree) s->static_len -= stree[node].Len; - /* node is 0 or 1 so it does not have extra bits */ - } - desc->max_code = max_code; - - /* The elements heap[heap_len/2+1 .. heap_len] are leaves of the tree, - * establish sub-heaps of increasing lengths: - */ - for (n = s->heap_len/2; n >= 1; n--) pqdownheap(s, tree, n); - - /* Construct the Huffman tree by repeatedly combining the least two - * frequent nodes. - */ - node = elems; /* next internal node of the tree */ - do { - pqremove(s, tree, n); /* n = node of least frequency */ - m = s->heap[SMALLEST]; /* m = node of next least frequency */ - - s->heap[--(s->heap_max)] = n; /* keep the nodes sorted by frequency */ - s->heap[--(s->heap_max)] = m; - - /* Create a new node father of n and m */ - tree[node].Freq = tree[n].Freq + tree[m].Freq; - s->depth[node] = (uch)((s->depth[n] >= s->depth[m] ? - s->depth[n] : s->depth[m]) + 1); - tree[n].Dad = tree[m].Dad = (ush)node; -#ifdef DUMP_BL_TREE - if (tree == s->bl_tree) { - fprintf(stderr,"\nnode %d(%d), sons %d(%d) %d(%d)", - node, tree[node].Freq, n, tree[n].Freq, m, tree[m].Freq); - } -#endif - /* and insert the new node in the heap */ - s->heap[SMALLEST] = node++; - pqdownheap(s, tree, SMALLEST); - - } while (s->heap_len >= 2); - - s->heap[--(s->heap_max)] = s->heap[SMALLEST]; - - /* At this point, the fields freq and dad are set. We can now - * generate the bit lengths. - */ - gen_bitlen(s, (tree_desc *)desc); - - /* The field len is now set, we can generate the bit codes */ - gen_codes ((ct_data *)tree, max_code, s->bl_count); -} - -/* =========================================================================== - * Scan a literal or distance tree to determine the frequencies of the codes - * in the bit length tree. - */ -local void scan_tree (s, tree, max_code) - deflate_state *s; - ct_data *tree; /* the tree to be scanned */ - int max_code; /* and its largest code of non zero frequency */ -{ - int n; /* iterates over all tree elements */ - int prevlen = -1; /* last emitted length */ - int curlen; /* length of current code */ - int nextlen = tree[0].Len; /* length of next code */ - int count = 0; /* repeat count of the current code */ - int max_count = 7; /* max repeat count */ - int min_count = 4; /* min repeat count */ - - if (nextlen == 0) max_count = 138, min_count = 3; - tree[max_code+1].Len = (ush)0xffff; /* guard */ - - for (n = 0; n <= max_code; n++) { - curlen = nextlen; nextlen = tree[n+1].Len; - if (++count < max_count && curlen == nextlen) { - continue; - } else if (count < min_count) { - s->bl_tree[curlen].Freq += count; - } else if (curlen != 0) { - if (curlen != prevlen) s->bl_tree[curlen].Freq++; - s->bl_tree[REP_3_6].Freq++; - } else if (count <= 10) { - s->bl_tree[REPZ_3_10].Freq++; - } else { - s->bl_tree[REPZ_11_138].Freq++; - } - count = 0; prevlen = curlen; - if (nextlen == 0) { - max_count = 138, min_count = 3; - } else if (curlen == nextlen) { - max_count = 6, min_count = 3; - } else { - max_count = 7, min_count = 4; - } - } -} - -/* =========================================================================== - * Send a literal or distance tree in compressed form, using the codes in - * bl_tree. - */ -local void send_tree (s, tree, max_code) - deflate_state *s; - ct_data *tree; /* the tree to be scanned */ - int max_code; /* and its largest code of non zero frequency */ -{ - int n; /* iterates over all tree elements */ - int prevlen = -1; /* last emitted length */ - int curlen; /* length of current code */ - int nextlen = tree[0].Len; /* length of next code */ - int count = 0; /* repeat count of the current code */ - int max_count = 7; /* max repeat count */ - int min_count = 4; /* min repeat count */ - - /* tree[max_code+1].Len = -1; */ /* guard already set */ - if (nextlen == 0) max_count = 138, min_count = 3; - - for (n = 0; n <= max_code; n++) { - curlen = nextlen; nextlen = tree[n+1].Len; - if (++count < max_count && curlen == nextlen) { - continue; - } else if (count < min_count) { - do { send_code(s, curlen, s->bl_tree); } while (--count != 0); - - } else if (curlen != 0) { - if (curlen != prevlen) { - send_code(s, curlen, s->bl_tree); count--; - } - Assert(count >= 3 && count <= 6, " 3_6?"); - send_code(s, REP_3_6, s->bl_tree); send_bits(s, count-3, 2); - - } else if (count <= 10) { - send_code(s, REPZ_3_10, s->bl_tree); send_bits(s, count-3, 3); - - } else { - send_code(s, REPZ_11_138, s->bl_tree); send_bits(s, count-11, 7); - } - count = 0; prevlen = curlen; - if (nextlen == 0) { - max_count = 138, min_count = 3; - } else if (curlen == nextlen) { - max_count = 6, min_count = 3; - } else { - max_count = 7, min_count = 4; - } - } -} - -/* =========================================================================== - * Construct the Huffman tree for the bit lengths and return the index in - * bl_order of the last bit length code to send. - */ -local int build_bl_tree(s) - deflate_state *s; -{ - int max_blindex; /* index of last bit length code of non zero freq */ - - /* Determine the bit length frequencies for literal and distance trees */ - scan_tree(s, (ct_data *)s->dyn_ltree, s->l_desc.max_code); - scan_tree(s, (ct_data *)s->dyn_dtree, s->d_desc.max_code); - - /* Build the bit length tree: */ - build_tree(s, (tree_desc *)(&(s->bl_desc))); - /* opt_len now includes the length of the tree representations, except - * the lengths of the bit lengths codes and the 5+5+4 bits for the counts. - */ - - /* Determine the number of bit length codes to send. The pkzip format - * requires that at least 4 bit length codes be sent. (appnote.txt says - * 3 but the actual value used is 4.) - */ - for (max_blindex = BL_CODES-1; max_blindex >= 3; max_blindex--) { - if (s->bl_tree[bl_order[max_blindex]].Len != 0) break; - } - /* Update opt_len to include the bit length tree and counts */ - s->opt_len += 3*((ulg)max_blindex+1) + 5+5+4; - Tracev((stderr, "\ndyn trees: dyn %ld, stat %ld", - s->opt_len, s->static_len)); - - return max_blindex; -} - -/* =========================================================================== - * Send the header for a block using dynamic Huffman trees: the counts, the - * lengths of the bit length codes, the literal tree and the distance tree. - * IN assertion: lcodes >= 257, dcodes >= 1, blcodes >= 4. - */ -local void send_all_trees(s, lcodes, dcodes, blcodes) - deflate_state *s; - int lcodes, dcodes, blcodes; /* number of codes for each tree */ -{ - int rank; /* index in bl_order */ - - Assert (lcodes >= 257 && dcodes >= 1 && blcodes >= 4, "not enough codes"); - Assert (lcodes <= L_CODES && dcodes <= D_CODES && blcodes <= BL_CODES, - "too many codes"); - Tracev((stderr, "\nbl counts: ")); - send_bits(s, lcodes-257, 5); /* not +255 as stated in appnote.txt */ - send_bits(s, dcodes-1, 5); - send_bits(s, blcodes-4, 4); /* not -3 as stated in appnote.txt */ - for (rank = 0; rank < blcodes; rank++) { - Tracev((stderr, "\nbl code %2d ", bl_order[rank])); - send_bits(s, s->bl_tree[bl_order[rank]].Len, 3); - } - Tracev((stderr, "\nbl tree: sent %ld", s->bits_sent)); - - send_tree(s, (ct_data *)s->dyn_ltree, lcodes-1); /* literal tree */ - Tracev((stderr, "\nlit tree: sent %ld", s->bits_sent)); - - send_tree(s, (ct_data *)s->dyn_dtree, dcodes-1); /* distance tree */ - Tracev((stderr, "\ndist tree: sent %ld", s->bits_sent)); -} - -/* =========================================================================== - * Send a stored block - */ -void ZLIB_INTERNAL _tr_stored_block(s, buf, stored_len, last) - deflate_state *s; - charf *buf; /* input block */ - ulg stored_len; /* length of input block */ - int last; /* one if this is the last block for a file */ -{ - send_bits(s, (STORED_BLOCK<<1)+last, 3); /* send block type */ - bi_windup(s); /* align on byte boundary */ - put_short(s, (ush)stored_len); - put_short(s, (ush)~stored_len); - zmemcpy(s->pending_buf + s->pending, (Bytef *)buf, stored_len); - s->pending += stored_len; -#ifdef ZLIB_DEBUG - s->compressed_len = (s->compressed_len + 3 + 7) & (ulg)~7L; - s->compressed_len += (stored_len + 4) << 3; - s->bits_sent += 2*16; - s->bits_sent += stored_len<<3; -#endif -} - -/* =========================================================================== - * Flush the bits in the bit buffer to pending output (leaves at most 7 bits) - */ -void ZLIB_INTERNAL _tr_flush_bits(s) - deflate_state *s; -{ - bi_flush(s); -} - -/* =========================================================================== - * Send one empty static block to give enough lookahead for inflate. - * This takes 10 bits, of which 7 may remain in the bit buffer. - */ -void ZLIB_INTERNAL _tr_align(s) - deflate_state *s; -{ - send_bits(s, STATIC_TREES<<1, 3); - send_code(s, END_BLOCK, static_ltree); -#ifdef ZLIB_DEBUG - s->compressed_len += 10L; /* 3 for block type, 7 for EOB */ -#endif - bi_flush(s); -} - -/* =========================================================================== - * Determine the best encoding for the current block: dynamic trees, static - * trees or store, and write out the encoded block. - */ -void ZLIB_INTERNAL _tr_flush_block(s, buf, stored_len, last) - deflate_state *s; - charf *buf; /* input block, or NULL if too old */ - ulg stored_len; /* length of input block */ - int last; /* one if this is the last block for a file */ -{ - ulg opt_lenb, static_lenb; /* opt_len and static_len in bytes */ - int max_blindex = 0; /* index of last bit length code of non zero freq */ - - /* Build the Huffman trees unless a stored block is forced */ - if (s->level > 0) { - - /* Check if the file is binary or text */ - if (s->strm->data_type == Z_UNKNOWN) - s->strm->data_type = detect_data_type(s); - - /* Construct the literal and distance trees */ - build_tree(s, (tree_desc *)(&(s->l_desc))); - Tracev((stderr, "\nlit data: dyn %ld, stat %ld", s->opt_len, - s->static_len)); - - build_tree(s, (tree_desc *)(&(s->d_desc))); - Tracev((stderr, "\ndist data: dyn %ld, stat %ld", s->opt_len, - s->static_len)); - /* At this point, opt_len and static_len are the total bit lengths of - * the compressed block data, excluding the tree representations. - */ - - /* Build the bit length tree for the above two trees, and get the index - * in bl_order of the last bit length code to send. - */ - max_blindex = build_bl_tree(s); - - /* Determine the best encoding. Compute the block lengths in bytes. */ - opt_lenb = (s->opt_len+3+7)>>3; - static_lenb = (s->static_len+3+7)>>3; - - Tracev((stderr, "\nopt %lu(%lu) stat %lu(%lu) stored %lu lit %u ", - opt_lenb, s->opt_len, static_lenb, s->static_len, stored_len, - s->last_lit)); - - if (static_lenb <= opt_lenb) opt_lenb = static_lenb; - - } else { - Assert(buf != (char*)0, "lost buf"); - opt_lenb = static_lenb = stored_len + 5; /* force a stored block */ - } - -#ifdef FORCE_STORED - if (buf != (char*)0) { /* force stored block */ -#else - if (stored_len+4 <= opt_lenb && buf != (char*)0) { - /* 4: two words for the lengths */ -#endif - /* The test buf != NULL is only necessary if LIT_BUFSIZE > WSIZE. - * Otherwise we can't have processed more than WSIZE input bytes since - * the last block flush, because compression would have been - * successful. If LIT_BUFSIZE <= WSIZE, it is never too late to - * transform a block into a stored block. - */ - _tr_stored_block(s, buf, stored_len, last); - -#ifdef FORCE_STATIC - } else if (static_lenb >= 0) { /* force static trees */ -#else - } else if (s->strategy == Z_FIXED || static_lenb == opt_lenb) { -#endif - send_bits(s, (STATIC_TREES<<1)+last, 3); - compress_block(s, (const ct_data *)static_ltree, - (const ct_data *)static_dtree); -#ifdef ZLIB_DEBUG - s->compressed_len += 3 + s->static_len; -#endif - } else { - send_bits(s, (DYN_TREES<<1)+last, 3); - send_all_trees(s, s->l_desc.max_code+1, s->d_desc.max_code+1, - max_blindex+1); - compress_block(s, (const ct_data *)s->dyn_ltree, - (const ct_data *)s->dyn_dtree); -#ifdef ZLIB_DEBUG - s->compressed_len += 3 + s->opt_len; -#endif - } - Assert (s->compressed_len == s->bits_sent, "bad compressed size"); - /* The above check is made mod 2^32, for files larger than 512 MB - * and uLong implemented on 32 bits. - */ - init_block(s); - - if (last) { - bi_windup(s); -#ifdef ZLIB_DEBUG - s->compressed_len += 7; /* align on byte boundary */ -#endif - } - Tracev((stderr,"\ncomprlen %lu(%lu) ", s->compressed_len>>3, - s->compressed_len-7*last)); -} - -/* =========================================================================== - * Save the match info and tally the frequency counts. Return true if - * the current block must be flushed. - */ -int ZLIB_INTERNAL _tr_tally (s, dist, lc) - deflate_state *s; - unsigned dist; /* distance of matched string */ - unsigned lc; /* match length-MIN_MATCH or unmatched char (if dist==0) */ -{ - s->d_buf[s->last_lit] = (ush)dist; - s->l_buf[s->last_lit++] = (uch)lc; - if (dist == 0) { - /* lc is the unmatched char */ - s->dyn_ltree[lc].Freq++; - } else { - s->matches++; - /* Here, lc is the match length - MIN_MATCH */ - dist--; /* dist = match distance - 1 */ - Assert((ush)dist < (ush)MAX_DIST(s) && - (ush)lc <= (ush)(MAX_MATCH-MIN_MATCH) && - (ush)d_code(dist) < (ush)D_CODES, "_tr_tally: bad match"); - - s->dyn_ltree[_length_code[lc]+LITERALS+1].Freq++; - s->dyn_dtree[d_code(dist)].Freq++; - } - -#ifdef TRUNCATE_BLOCK - /* Try to guess if it is profitable to stop the current block here */ - if ((s->last_lit & 0x1fff) == 0 && s->level > 2) { - /* Compute an upper bound for the compressed length */ - ulg out_length = (ulg)s->last_lit*8L; - ulg in_length = (ulg)((long)s->strstart - s->block_start); - int dcode; - for (dcode = 0; dcode < D_CODES; dcode++) { - out_length += (ulg)s->dyn_dtree[dcode].Freq * - (5L+extra_dbits[dcode]); - } - out_length >>= 3; - Tracev((stderr,"\nlast_lit %u, in %ld, out ~%ld(%ld%%) ", - s->last_lit, in_length, out_length, - 100L - out_length*100L/in_length)); - if (s->matches < s->last_lit/2 && out_length < in_length/2) return 1; - } -#endif - return (s->last_lit == s->lit_bufsize-1); - /* We avoid equality with lit_bufsize because of wraparound at 64K - * on 16 bit machines and because stored blocks are restricted to - * 64K-1 bytes. - */ -} - -/* =========================================================================== - * Send the block data compressed using the given Huffman trees - */ -local void compress_block(s, ltree, dtree) - deflate_state *s; - const ct_data *ltree; /* literal tree */ - const ct_data *dtree; /* distance tree */ -{ - unsigned dist; /* distance of matched string */ - int lc; /* match length or unmatched char (if dist == 0) */ - unsigned lx = 0; /* running index in l_buf */ - unsigned code; /* the code to send */ - int extra; /* number of extra bits to send */ - - if (s->last_lit != 0) do { - dist = s->d_buf[lx]; - lc = s->l_buf[lx++]; - if (dist == 0) { - send_code(s, lc, ltree); /* send a literal byte */ - Tracecv(isgraph(lc), (stderr," '%c' ", lc)); - } else { - /* Here, lc is the match length - MIN_MATCH */ - code = _length_code[lc]; - send_code(s, code+LITERALS+1, ltree); /* send the length code */ - extra = extra_lbits[code]; - if (extra != 0) { - lc -= base_length[code]; - send_bits(s, lc, extra); /* send the extra length bits */ - } - dist--; /* dist is now the match distance - 1 */ - code = d_code(dist); - Assert (code < D_CODES, "bad d_code"); - - send_code(s, code, dtree); /* send the distance code */ - extra = extra_dbits[code]; - if (extra != 0) { - dist -= (unsigned)base_dist[code]; - send_bits(s, dist, extra); /* send the extra distance bits */ - } - } /* literal or match pair ? */ - - /* Check that the overlay between pending_buf and d_buf+l_buf is ok: */ - Assert((uInt)(s->pending) < s->lit_bufsize + 2*lx, - "pendingBuf overflow"); - - } while (lx < s->last_lit); - - send_code(s, END_BLOCK, ltree); -} - -/* =========================================================================== - * Check if the data type is TEXT or BINARY, using the following algorithm: - * - TEXT if the two conditions below are satisfied: - * a) There are no non-portable control characters belonging to the - * "black list" (0..6, 14..25, 28..31). - * b) There is at least one printable character belonging to the - * "white list" (9 {TAB}, 10 {LF}, 13 {CR}, 32..255). - * - BINARY otherwise. - * - The following partially-portable control characters form a - * "gray list" that is ignored in this detection algorithm: - * (7 {BEL}, 8 {BS}, 11 {VT}, 12 {FF}, 26 {SUB}, 27 {ESC}). - * IN assertion: the fields Freq of dyn_ltree are set. - */ -local int detect_data_type(s) - deflate_state *s; -{ - /* black_mask is the bit mask of black-listed bytes - * set bits 0..6, 14..25, and 28..31 - * 0xf3ffc07f = binary 11110011111111111100000001111111 - */ - unsigned long black_mask = 0xf3ffc07fUL; - int n; - - /* Check for non-textual ("black-listed") bytes. */ - for (n = 0; n <= 31; n++, black_mask >>= 1) - if ((black_mask & 1) && (s->dyn_ltree[n].Freq != 0)) - return Z_BINARY; - - /* Check for textual ("white-listed") bytes. */ - if (s->dyn_ltree[9].Freq != 0 || s->dyn_ltree[10].Freq != 0 - || s->dyn_ltree[13].Freq != 0) - return Z_TEXT; - for (n = 32; n < LITERALS; n++) - if (s->dyn_ltree[n].Freq != 0) - return Z_TEXT; - - /* There are no "black-listed" or "white-listed" bytes: - * this stream either is empty or has tolerated ("gray-listed") bytes only. - */ - return Z_BINARY; -} - -/* =========================================================================== - * Reverse the first len bits of a code, using straightforward code (a faster - * method would use a table) - * IN assertion: 1 <= len <= 15 - */ -local unsigned bi_reverse(code, len) - unsigned code; /* the value to invert */ - int len; /* its bit length */ -{ - register unsigned res = 0; - do { - res |= code & 1; - code >>= 1, res <<= 1; - } while (--len > 0); - return res >> 1; -} - -/* =========================================================================== - * Flush the bit buffer, keeping at most 7 bits in it. - */ -local void bi_flush(s) - deflate_state *s; -{ - if (s->bi_valid == 16) { - put_short(s, s->bi_buf); - s->bi_buf = 0; - s->bi_valid = 0; - } else if (s->bi_valid >= 8) { - put_byte(s, (Byte)s->bi_buf); - s->bi_buf >>= 8; - s->bi_valid -= 8; - } -} - -/* =========================================================================== - * Flush the bit buffer and align the output on a byte boundary - */ -local void bi_windup(s) - deflate_state *s; -{ - if (s->bi_valid > 8) { - put_short(s, s->bi_buf); - } else if (s->bi_valid > 0) { - put_byte(s, (Byte)s->bi_buf); - } - s->bi_buf = 0; - s->bi_valid = 0; -#ifdef ZLIB_DEBUG - s->bits_sent = (s->bits_sent+7) & ~7; -#endif -} diff --git a/dep/zlib/src/trees.h b/dep/zlib/src/trees.h deleted file mode 100644 index d35639d82..000000000 --- a/dep/zlib/src/trees.h +++ /dev/null @@ -1,128 +0,0 @@ -/* header created automatically with -DGEN_TREES_H */ - -local const ct_data static_ltree[L_CODES+2] = { -{{ 12},{ 8}}, {{140},{ 8}}, {{ 76},{ 8}}, {{204},{ 8}}, {{ 44},{ 8}}, -{{172},{ 8}}, {{108},{ 8}}, {{236},{ 8}}, {{ 28},{ 8}}, {{156},{ 8}}, -{{ 92},{ 8}}, {{220},{ 8}}, {{ 60},{ 8}}, {{188},{ 8}}, {{124},{ 8}}, -{{252},{ 8}}, {{ 2},{ 8}}, {{130},{ 8}}, {{ 66},{ 8}}, {{194},{ 8}}, -{{ 34},{ 8}}, {{162},{ 8}}, {{ 98},{ 8}}, {{226},{ 8}}, {{ 18},{ 8}}, -{{146},{ 8}}, {{ 82},{ 8}}, {{210},{ 8}}, {{ 50},{ 8}}, {{178},{ 8}}, -{{114},{ 8}}, {{242},{ 8}}, {{ 10},{ 8}}, {{138},{ 8}}, {{ 74},{ 8}}, -{{202},{ 8}}, {{ 42},{ 8}}, {{170},{ 8}}, {{106},{ 8}}, {{234},{ 8}}, -{{ 26},{ 8}}, {{154},{ 8}}, {{ 90},{ 8}}, {{218},{ 8}}, {{ 58},{ 8}}, -{{186},{ 8}}, {{122},{ 8}}, {{250},{ 8}}, {{ 6},{ 8}}, {{134},{ 8}}, -{{ 70},{ 8}}, {{198},{ 8}}, {{ 38},{ 8}}, {{166},{ 8}}, {{102},{ 8}}, -{{230},{ 8}}, {{ 22},{ 8}}, {{150},{ 8}}, {{ 86},{ 8}}, {{214},{ 8}}, -{{ 54},{ 8}}, {{182},{ 8}}, {{118},{ 8}}, {{246},{ 8}}, {{ 14},{ 8}}, -{{142},{ 8}}, {{ 78},{ 8}}, {{206},{ 8}}, {{ 46},{ 8}}, {{174},{ 8}}, -{{110},{ 8}}, {{238},{ 8}}, {{ 30},{ 8}}, {{158},{ 8}}, {{ 94},{ 8}}, -{{222},{ 8}}, {{ 62},{ 8}}, {{190},{ 8}}, {{126},{ 8}}, {{254},{ 8}}, -{{ 1},{ 8}}, {{129},{ 8}}, {{ 65},{ 8}}, {{193},{ 8}}, {{ 33},{ 8}}, -{{161},{ 8}}, {{ 97},{ 8}}, {{225},{ 8}}, {{ 17},{ 8}}, {{145},{ 8}}, -{{ 81},{ 8}}, {{209},{ 8}}, {{ 49},{ 8}}, {{177},{ 8}}, {{113},{ 8}}, -{{241},{ 8}}, {{ 9},{ 8}}, {{137},{ 8}}, {{ 73},{ 8}}, {{201},{ 8}}, -{{ 41},{ 8}}, {{169},{ 8}}, {{105},{ 8}}, {{233},{ 8}}, {{ 25},{ 8}}, -{{153},{ 8}}, {{ 89},{ 8}}, {{217},{ 8}}, {{ 57},{ 8}}, {{185},{ 8}}, -{{121},{ 8}}, {{249},{ 8}}, {{ 5},{ 8}}, {{133},{ 8}}, {{ 69},{ 8}}, -{{197},{ 8}}, {{ 37},{ 8}}, {{165},{ 8}}, {{101},{ 8}}, {{229},{ 8}}, -{{ 21},{ 8}}, {{149},{ 8}}, {{ 85},{ 8}}, {{213},{ 8}}, {{ 53},{ 8}}, -{{181},{ 8}}, {{117},{ 8}}, {{245},{ 8}}, {{ 13},{ 8}}, {{141},{ 8}}, -{{ 77},{ 8}}, {{205},{ 8}}, {{ 45},{ 8}}, {{173},{ 8}}, {{109},{ 8}}, -{{237},{ 8}}, {{ 29},{ 8}}, {{157},{ 8}}, {{ 93},{ 8}}, {{221},{ 8}}, -{{ 61},{ 8}}, {{189},{ 8}}, {{125},{ 8}}, {{253},{ 8}}, {{ 19},{ 9}}, -{{275},{ 9}}, {{147},{ 9}}, {{403},{ 9}}, {{ 83},{ 9}}, {{339},{ 9}}, -{{211},{ 9}}, {{467},{ 9}}, {{ 51},{ 9}}, {{307},{ 9}}, {{179},{ 9}}, -{{435},{ 9}}, {{115},{ 9}}, {{371},{ 9}}, {{243},{ 9}}, {{499},{ 9}}, -{{ 11},{ 9}}, {{267},{ 9}}, {{139},{ 9}}, {{395},{ 9}}, {{ 75},{ 9}}, -{{331},{ 9}}, {{203},{ 9}}, {{459},{ 9}}, {{ 43},{ 9}}, {{299},{ 9}}, -{{171},{ 9}}, {{427},{ 9}}, {{107},{ 9}}, {{363},{ 9}}, {{235},{ 9}}, -{{491},{ 9}}, {{ 27},{ 9}}, {{283},{ 9}}, {{155},{ 9}}, {{411},{ 9}}, -{{ 91},{ 9}}, {{347},{ 9}}, {{219},{ 9}}, {{475},{ 9}}, {{ 59},{ 9}}, -{{315},{ 9}}, {{187},{ 9}}, {{443},{ 9}}, {{123},{ 9}}, {{379},{ 9}}, -{{251},{ 9}}, {{507},{ 9}}, {{ 7},{ 9}}, {{263},{ 9}}, {{135},{ 9}}, -{{391},{ 9}}, {{ 71},{ 9}}, {{327},{ 9}}, {{199},{ 9}}, {{455},{ 9}}, -{{ 39},{ 9}}, {{295},{ 9}}, {{167},{ 9}}, {{423},{ 9}}, {{103},{ 9}}, -{{359},{ 9}}, {{231},{ 9}}, {{487},{ 9}}, {{ 23},{ 9}}, {{279},{ 9}}, -{{151},{ 9}}, {{407},{ 9}}, {{ 87},{ 9}}, {{343},{ 9}}, {{215},{ 9}}, -{{471},{ 9}}, {{ 55},{ 9}}, {{311},{ 9}}, {{183},{ 9}}, {{439},{ 9}}, -{{119},{ 9}}, {{375},{ 9}}, {{247},{ 9}}, {{503},{ 9}}, {{ 15},{ 9}}, -{{271},{ 9}}, {{143},{ 9}}, {{399},{ 9}}, {{ 79},{ 9}}, {{335},{ 9}}, -{{207},{ 9}}, {{463},{ 9}}, {{ 47},{ 9}}, {{303},{ 9}}, {{175},{ 9}}, -{{431},{ 9}}, {{111},{ 9}}, {{367},{ 9}}, {{239},{ 9}}, {{495},{ 9}}, -{{ 31},{ 9}}, {{287},{ 9}}, {{159},{ 9}}, {{415},{ 9}}, {{ 95},{ 9}}, -{{351},{ 9}}, {{223},{ 9}}, {{479},{ 9}}, {{ 63},{ 9}}, {{319},{ 9}}, -{{191},{ 9}}, {{447},{ 9}}, {{127},{ 9}}, {{383},{ 9}}, {{255},{ 9}}, -{{511},{ 9}}, {{ 0},{ 7}}, {{ 64},{ 7}}, {{ 32},{ 7}}, {{ 96},{ 7}}, -{{ 16},{ 7}}, {{ 80},{ 7}}, {{ 48},{ 7}}, {{112},{ 7}}, {{ 8},{ 7}}, -{{ 72},{ 7}}, {{ 40},{ 7}}, {{104},{ 7}}, {{ 24},{ 7}}, {{ 88},{ 7}}, -{{ 56},{ 7}}, {{120},{ 7}}, {{ 4},{ 7}}, {{ 68},{ 7}}, {{ 36},{ 7}}, -{{100},{ 7}}, {{ 20},{ 7}}, {{ 84},{ 7}}, {{ 52},{ 7}}, {{116},{ 7}}, -{{ 3},{ 8}}, {{131},{ 8}}, {{ 67},{ 8}}, {{195},{ 8}}, {{ 35},{ 8}}, -{{163},{ 8}}, {{ 99},{ 8}}, {{227},{ 8}} -}; - -local const ct_data static_dtree[D_CODES] = { -{{ 0},{ 5}}, {{16},{ 5}}, {{ 8},{ 5}}, {{24},{ 5}}, {{ 4},{ 5}}, -{{20},{ 5}}, {{12},{ 5}}, {{28},{ 5}}, {{ 2},{ 5}}, {{18},{ 5}}, -{{10},{ 5}}, {{26},{ 5}}, {{ 6},{ 5}}, {{22},{ 5}}, {{14},{ 5}}, -{{30},{ 5}}, {{ 1},{ 5}}, {{17},{ 5}}, {{ 9},{ 5}}, {{25},{ 5}}, -{{ 5},{ 5}}, {{21},{ 5}}, {{13},{ 5}}, {{29},{ 5}}, {{ 3},{ 5}}, -{{19},{ 5}}, {{11},{ 5}}, {{27},{ 5}}, {{ 7},{ 5}}, {{23},{ 5}} -}; - -const uch ZLIB_INTERNAL _dist_code[DIST_CODE_LEN] = { - 0, 1, 2, 3, 4, 4, 5, 5, 6, 6, 6, 6, 7, 7, 7, 7, 8, 8, 8, 8, - 8, 8, 8, 8, 9, 9, 9, 9, 9, 9, 9, 9, 10, 10, 10, 10, 10, 10, 10, 10, -10, 10, 10, 10, 10, 10, 10, 10, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, 11, -11, 11, 11, 11, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, -12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 12, 13, 13, 13, 13, -13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, 13, -13, 13, 13, 13, 13, 13, 13, 13, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, -14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, -14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, -14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 14, 15, 15, 15, 15, 15, 15, 15, 15, -15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, -15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, -15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 15, 0, 0, 16, 17, -18, 18, 19, 19, 20, 20, 20, 20, 21, 21, 21, 21, 22, 22, 22, 22, 22, 22, 22, 22, -23, 23, 23, 23, 23, 23, 23, 23, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, -24, 24, 24, 24, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, -26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, -26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 27, 27, 27, 27, 27, 27, 27, 27, -27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, -27, 27, 27, 27, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, -28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, -28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, 28, -28, 28, 28, 28, 28, 28, 28, 28, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, -29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, -29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, -29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29, 29 -}; - -const uch ZLIB_INTERNAL _length_code[MAX_MATCH-MIN_MATCH+1]= { - 0, 1, 2, 3, 4, 5, 6, 7, 8, 8, 9, 9, 10, 10, 11, 11, 12, 12, 12, 12, -13, 13, 13, 13, 14, 14, 14, 14, 15, 15, 15, 15, 16, 16, 16, 16, 16, 16, 16, 16, -17, 17, 17, 17, 17, 17, 17, 17, 18, 18, 18, 18, 18, 18, 18, 18, 19, 19, 19, 19, -19, 19, 19, 19, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, 20, -21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 21, 22, 22, 22, 22, -22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 22, 23, 23, 23, 23, 23, 23, 23, 23, -23, 23, 23, 23, 23, 23, 23, 23, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, -24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, 24, -25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, -25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25, 26, 26, 26, 26, 26, 26, 26, 26, -26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, 26, -26, 26, 26, 26, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, -27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 27, 28 -}; - -local const int base_length[LENGTH_CODES] = { -0, 1, 2, 3, 4, 5, 6, 7, 8, 10, 12, 14, 16, 20, 24, 28, 32, 40, 48, 56, -64, 80, 96, 112, 128, 160, 192, 224, 0 -}; - -local const int base_dist[D_CODES] = { - 0, 1, 2, 3, 4, 6, 8, 12, 16, 24, - 32, 48, 64, 96, 128, 192, 256, 384, 512, 768, - 1024, 1536, 2048, 3072, 4096, 6144, 8192, 12288, 16384, 24576 -}; - diff --git a/dep/zlib/src/uncompr.c b/dep/zlib/src/uncompr.c deleted file mode 100644 index f03a1a865..000000000 --- a/dep/zlib/src/uncompr.c +++ /dev/null @@ -1,93 +0,0 @@ -/* uncompr.c -- decompress a memory buffer - * Copyright (C) 1995-2003, 2010, 2014, 2016 Jean-loup Gailly, Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* @(#) $Id$ */ - -#define ZLIB_INTERNAL -#include "zlib.h" - -/* =========================================================================== - Decompresses the source buffer into the destination buffer. *sourceLen is - the byte length of the source buffer. Upon entry, *destLen is the total size - of the destination buffer, which must be large enough to hold the entire - uncompressed data. (The size of the uncompressed data must have been saved - previously by the compressor and transmitted to the decompressor by some - mechanism outside the scope of this compression library.) Upon exit, - *destLen is the size of the decompressed data and *sourceLen is the number - of source bytes consumed. Upon return, source + *sourceLen points to the - first unused input byte. - - uncompress returns Z_OK if success, Z_MEM_ERROR if there was not enough - memory, Z_BUF_ERROR if there was not enough room in the output buffer, or - Z_DATA_ERROR if the input data was corrupted, including if the input data is - an incomplete zlib stream. -*/ -int ZEXPORT uncompress2 (dest, destLen, source, sourceLen) - Bytef *dest; - uLongf *destLen; - const Bytef *source; - uLong *sourceLen; -{ - z_stream stream; - int err; - const uInt max = (uInt)-1; - uLong len, left; - Byte buf[1]; /* for detection of incomplete stream when *destLen == 0 */ - - len = *sourceLen; - if (*destLen) { - left = *destLen; - *destLen = 0; - } - else { - left = 1; - dest = buf; - } - - stream.next_in = (z_const Bytef *)source; - stream.avail_in = 0; - stream.zalloc = (alloc_func)0; - stream.zfree = (free_func)0; - stream.opaque = (voidpf)0; - - err = inflateInit(&stream); - if (err != Z_OK) return err; - - stream.next_out = dest; - stream.avail_out = 0; - - do { - if (stream.avail_out == 0) { - stream.avail_out = left > (uLong)max ? max : (uInt)left; - left -= stream.avail_out; - } - if (stream.avail_in == 0) { - stream.avail_in = len > (uLong)max ? max : (uInt)len; - len -= stream.avail_in; - } - err = inflate(&stream, Z_NO_FLUSH); - } while (err == Z_OK); - - *sourceLen -= len + stream.avail_in; - if (dest != buf) - *destLen = stream.total_out; - else if (stream.total_out && err == Z_BUF_ERROR) - left = 1; - - inflateEnd(&stream); - return err == Z_STREAM_END ? Z_OK : - err == Z_NEED_DICT ? Z_DATA_ERROR : - err == Z_BUF_ERROR && left + stream.avail_out ? Z_DATA_ERROR : - err; -} - -int ZEXPORT uncompress (dest, destLen, source, sourceLen) - Bytef *dest; - uLongf *destLen; - const Bytef *source; - uLong sourceLen; -{ - return uncompress2(dest, destLen, source, &sourceLen); -} diff --git a/dep/zlib/src/zutil.c b/dep/zlib/src/zutil.c deleted file mode 100644 index a76c6b0c7..000000000 --- a/dep/zlib/src/zutil.c +++ /dev/null @@ -1,325 +0,0 @@ -/* zutil.c -- target dependent utility functions for the compression library - * Copyright (C) 1995-2017 Jean-loup Gailly - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* @(#) $Id$ */ - -#include "zutil.h" -#ifndef Z_SOLO -# include "gzguts.h" -#endif - -z_const char * const z_errmsg[10] = { - (z_const char *)"need dictionary", /* Z_NEED_DICT 2 */ - (z_const char *)"stream end", /* Z_STREAM_END 1 */ - (z_const char *)"", /* Z_OK 0 */ - (z_const char *)"file error", /* Z_ERRNO (-1) */ - (z_const char *)"stream error", /* Z_STREAM_ERROR (-2) */ - (z_const char *)"data error", /* Z_DATA_ERROR (-3) */ - (z_const char *)"insufficient memory", /* Z_MEM_ERROR (-4) */ - (z_const char *)"buffer error", /* Z_BUF_ERROR (-5) */ - (z_const char *)"incompatible version",/* Z_VERSION_ERROR (-6) */ - (z_const char *)"" -}; - - -const char * ZEXPORT zlibVersion() -{ - return ZLIB_VERSION; -} - -uLong ZEXPORT zlibCompileFlags() -{ - uLong flags; - - flags = 0; - switch ((int)(sizeof(uInt))) { - case 2: break; - case 4: flags += 1; break; - case 8: flags += 2; break; - default: flags += 3; - } - switch ((int)(sizeof(uLong))) { - case 2: break; - case 4: flags += 1 << 2; break; - case 8: flags += 2 << 2; break; - default: flags += 3 << 2; - } - switch ((int)(sizeof(voidpf))) { - case 2: break; - case 4: flags += 1 << 4; break; - case 8: flags += 2 << 4; break; - default: flags += 3 << 4; - } - switch ((int)(sizeof(z_off_t))) { - case 2: break; - case 4: flags += 1 << 6; break; - case 8: flags += 2 << 6; break; - default: flags += 3 << 6; - } -#ifdef ZLIB_DEBUG - flags += 1 << 8; -#endif -#if defined(ASMV) || defined(ASMINF) - flags += 1 << 9; -#endif -#ifdef ZLIB_WINAPI - flags += 1 << 10; -#endif -#ifdef BUILDFIXED - flags += 1 << 12; -#endif -#ifdef DYNAMIC_CRC_TABLE - flags += 1 << 13; -#endif -#ifdef NO_GZCOMPRESS - flags += 1L << 16; -#endif -#ifdef NO_GZIP - flags += 1L << 17; -#endif -#ifdef PKZIP_BUG_WORKAROUND - flags += 1L << 20; -#endif -#ifdef FASTEST - flags += 1L << 21; -#endif -#if defined(STDC) || defined(Z_HAVE_STDARG_H) -# ifdef NO_vsnprintf - flags += 1L << 25; -# ifdef HAS_vsprintf_void - flags += 1L << 26; -# endif -# else -# ifdef HAS_vsnprintf_void - flags += 1L << 26; -# endif -# endif -#else - flags += 1L << 24; -# ifdef NO_snprintf - flags += 1L << 25; -# ifdef HAS_sprintf_void - flags += 1L << 26; -# endif -# else -# ifdef HAS_snprintf_void - flags += 1L << 26; -# endif -# endif -#endif - return flags; -} - -#ifdef ZLIB_DEBUG -#include -# ifndef verbose -# define verbose 0 -# endif -int ZLIB_INTERNAL z_verbose = verbose; - -void ZLIB_INTERNAL z_error (m) - char *m; -{ - fprintf(stderr, "%s\n", m); - exit(1); -} -#endif - -/* exported to allow conversion of error code to string for compress() and - * uncompress() - */ -const char * ZEXPORT zError(err) - int err; -{ - return ERR_MSG(err); -} - -#if defined(_WIN32_WCE) - /* The Microsoft C Run-Time Library for Windows CE doesn't have - * errno. We define it as a global variable to simplify porting. - * Its value is always 0 and should not be used. - */ - int errno = 0; -#endif - -#ifndef HAVE_MEMCPY - -void ZLIB_INTERNAL zmemcpy(dest, source, len) - Bytef* dest; - const Bytef* source; - uInt len; -{ - if (len == 0) return; - do { - *dest++ = *source++; /* ??? to be unrolled */ - } while (--len != 0); -} - -int ZLIB_INTERNAL zmemcmp(s1, s2, len) - const Bytef* s1; - const Bytef* s2; - uInt len; -{ - uInt j; - - for (j = 0; j < len; j++) { - if (s1[j] != s2[j]) return 2*(s1[j] > s2[j])-1; - } - return 0; -} - -void ZLIB_INTERNAL zmemzero(dest, len) - Bytef* dest; - uInt len; -{ - if (len == 0) return; - do { - *dest++ = 0; /* ??? to be unrolled */ - } while (--len != 0); -} -#endif - -#ifndef Z_SOLO - -#ifdef SYS16BIT - -#ifdef __TURBOC__ -/* Turbo C in 16-bit mode */ - -# define MY_ZCALLOC - -/* Turbo C malloc() does not allow dynamic allocation of 64K bytes - * and farmalloc(64K) returns a pointer with an offset of 8, so we - * must fix the pointer. Warning: the pointer must be put back to its - * original form in order to free it, use zcfree(). - */ - -#define MAX_PTR 10 -/* 10*64K = 640K */ - -local int next_ptr = 0; - -typedef struct ptr_table_s { - voidpf org_ptr; - voidpf new_ptr; -} ptr_table; - -local ptr_table table[MAX_PTR]; -/* This table is used to remember the original form of pointers - * to large buffers (64K). Such pointers are normalized with a zero offset. - * Since MSDOS is not a preemptive multitasking OS, this table is not - * protected from concurrent access. This hack doesn't work anyway on - * a protected system like OS/2. Use Microsoft C instead. - */ - -voidpf ZLIB_INTERNAL zcalloc (voidpf opaque, unsigned items, unsigned size) -{ - voidpf buf; - ulg bsize = (ulg)items*size; - - (void)opaque; - - /* If we allocate less than 65520 bytes, we assume that farmalloc - * will return a usable pointer which doesn't have to be normalized. - */ - if (bsize < 65520L) { - buf = farmalloc(bsize); - if (*(ush*)&buf != 0) return buf; - } else { - buf = farmalloc(bsize + 16L); - } - if (buf == NULL || next_ptr >= MAX_PTR) return NULL; - table[next_ptr].org_ptr = buf; - - /* Normalize the pointer to seg:0 */ - *((ush*)&buf+1) += ((ush)((uch*)buf-0) + 15) >> 4; - *(ush*)&buf = 0; - table[next_ptr++].new_ptr = buf; - return buf; -} - -void ZLIB_INTERNAL zcfree (voidpf opaque, voidpf ptr) -{ - int n; - - (void)opaque; - - if (*(ush*)&ptr != 0) { /* object < 64K */ - farfree(ptr); - return; - } - /* Find the original pointer */ - for (n = 0; n < next_ptr; n++) { - if (ptr != table[n].new_ptr) continue; - - farfree(table[n].org_ptr); - while (++n < next_ptr) { - table[n-1] = table[n]; - } - next_ptr--; - return; - } - Assert(0, "zcfree: ptr not found"); -} - -#endif /* __TURBOC__ */ - - -#ifdef M_I86 -/* Microsoft C in 16-bit mode */ - -# define MY_ZCALLOC - -#if (!defined(_MSC_VER) || (_MSC_VER <= 600)) -# define _halloc halloc -# define _hfree hfree -#endif - -voidpf ZLIB_INTERNAL zcalloc (voidpf opaque, uInt items, uInt size) -{ - (void)opaque; - return _halloc((long)items, size); -} - -void ZLIB_INTERNAL zcfree (voidpf opaque, voidpf ptr) -{ - (void)opaque; - _hfree(ptr); -} - -#endif /* M_I86 */ - -#endif /* SYS16BIT */ - - -#ifndef MY_ZCALLOC /* Any system without a special alloc function */ - -#ifndef STDC -extern voidp malloc OF((uInt size)); -extern voidp calloc OF((uInt items, uInt size)); -extern void free OF((voidpf ptr)); -#endif - -voidpf ZLIB_INTERNAL zcalloc (opaque, items, size) - voidpf opaque; - unsigned items; - unsigned size; -{ - (void)opaque; - return sizeof(uInt) > 2 ? (voidpf)malloc(items * size) : - (voidpf)calloc(items, size); -} - -void ZLIB_INTERNAL zcfree (opaque, ptr) - voidpf opaque; - voidpf ptr; -{ - (void)opaque; - free(ptr); -} - -#endif /* MY_ZCALLOC */ - -#endif /* !Z_SOLO */ diff --git a/dep/zlib/src/zutil.h b/dep/zlib/src/zutil.h deleted file mode 100644 index b079ea6a8..000000000 --- a/dep/zlib/src/zutil.h +++ /dev/null @@ -1,271 +0,0 @@ -/* zutil.h -- internal interface and configuration of the compression library - * Copyright (C) 1995-2016 Jean-loup Gailly, Mark Adler - * For conditions of distribution and use, see copyright notice in zlib.h - */ - -/* WARNING: this file should *not* be used by applications. It is - part of the implementation of the compression library and is - subject to change. Applications should only use zlib.h. - */ - -/* @(#) $Id$ */ - -#ifndef ZUTIL_H -#define ZUTIL_H - -#ifdef HAVE_HIDDEN -# define ZLIB_INTERNAL __attribute__((visibility ("hidden"))) -#else -# define ZLIB_INTERNAL -#endif - -#include "zlib.h" - -#if defined(STDC) && !defined(Z_SOLO) -# if !(defined(_WIN32_WCE) && defined(_MSC_VER)) -# include -# endif -# include -# include -#endif - -#ifdef Z_SOLO - typedef long ptrdiff_t; /* guess -- will be caught if guess is wrong */ -#endif - -#ifndef local -# define local static -#endif -/* since "static" is used to mean two completely different things in C, we - define "local" for the non-static meaning of "static", for readability - (compile with -Dlocal if your debugger can't find static symbols) */ - -typedef unsigned char uch; -typedef uch FAR uchf; -typedef unsigned short ush; -typedef ush FAR ushf; -typedef unsigned long ulg; - -extern z_const char * const z_errmsg[10]; /* indexed by 2-zlib_error */ -/* (size given to avoid silly warnings with Visual C++) */ - -#define ERR_MSG(err) z_errmsg[Z_NEED_DICT-(err)] - -#define ERR_RETURN(strm,err) \ - return (strm->msg = ERR_MSG(err), (err)) -/* To be used only when the state is known to be valid */ - - /* common constants */ - -#ifndef DEF_WBITS -# define DEF_WBITS MAX_WBITS -#endif -/* default windowBits for decompression. MAX_WBITS is for compression only */ - -#if MAX_MEM_LEVEL >= 8 -# define DEF_MEM_LEVEL 8 -#else -# define DEF_MEM_LEVEL MAX_MEM_LEVEL -#endif -/* default memLevel */ - -#define STORED_BLOCK 0 -#define STATIC_TREES 1 -#define DYN_TREES 2 -/* The three kinds of block type */ - -#define MIN_MATCH 3 -#define MAX_MATCH 258 -/* The minimum and maximum match lengths */ - -#define PRESET_DICT 0x20 /* preset dictionary flag in zlib header */ - - /* target dependencies */ - -#if defined(MSDOS) || (defined(WINDOWS) && !defined(WIN32)) -# define OS_CODE 0x00 -# ifndef Z_SOLO -# if defined(__TURBOC__) || defined(__BORLANDC__) -# if (__STDC__ == 1) && (defined(__LARGE__) || defined(__COMPACT__)) - /* Allow compilation with ANSI keywords only enabled */ - void _Cdecl farfree( void *block ); - void *_Cdecl farmalloc( unsigned long nbytes ); -# else -# include -# endif -# else /* MSC or DJGPP */ -# include -# endif -# endif -#endif - -#ifdef AMIGA -# define OS_CODE 1 -#endif - -#if defined(VAXC) || defined(VMS) -# define OS_CODE 2 -# define F_OPEN(name, mode) \ - fopen((name), (mode), "mbc=60", "ctx=stm", "rfm=fix", "mrs=512") -#endif - -#ifdef __370__ -# if __TARGET_LIB__ < 0x20000000 -# define OS_CODE 4 -# elif __TARGET_LIB__ < 0x40000000 -# define OS_CODE 11 -# else -# define OS_CODE 8 -# endif -#endif - -#if defined(ATARI) || defined(atarist) -# define OS_CODE 5 -#endif - -#ifdef OS2 -# define OS_CODE 6 -# if defined(M_I86) && !defined(Z_SOLO) -# include -# endif -#endif - -#if defined(MACOS) || defined(TARGET_OS_MAC) -# define OS_CODE 7 -# ifndef Z_SOLO -# if defined(__MWERKS__) && __dest_os != __be_os && __dest_os != __win32_os -# include /* for fdopen */ -# else -# ifndef fdopen -# define fdopen(fd,mode) NULL /* No fdopen() */ -# endif -# endif -# endif -#endif - -#ifdef __acorn -# define OS_CODE 13 -#endif - -#if defined(WIN32) && !defined(__CYGWIN__) -# define OS_CODE 10 -#endif - -#ifdef _BEOS_ -# define OS_CODE 16 -#endif - -#ifdef __TOS_OS400__ -# define OS_CODE 18 -#endif - -#ifdef __APPLE__ -# define OS_CODE 19 -#endif - -#if defined(_BEOS_) || defined(RISCOS) -# define fdopen(fd,mode) NULL /* No fdopen() */ -#endif - -#if (defined(_MSC_VER) && (_MSC_VER > 600)) && !defined __INTERIX -# if defined(_WIN32_WCE) -# define fdopen(fd,mode) NULL /* No fdopen() */ -# ifndef _PTRDIFF_T_DEFINED - typedef int ptrdiff_t; -# define _PTRDIFF_T_DEFINED -# endif -# else -# define fdopen(fd,type) _fdopen(fd,type) -# endif -#endif - -#if defined(__BORLANDC__) && !defined(MSDOS) - #pragma warn -8004 - #pragma warn -8008 - #pragma warn -8066 -#endif - -/* provide prototypes for these when building zlib without LFS */ -#if !defined(_WIN32) && \ - (!defined(_LARGEFILE64_SOURCE) || _LFS64_LARGEFILE-0 == 0) - ZEXTERN uLong ZEXPORT adler32_combine64 OF((uLong, uLong, z_off_t)); - ZEXTERN uLong ZEXPORT crc32_combine64 OF((uLong, uLong, z_off_t)); -#endif - - /* common defaults */ - -#ifndef OS_CODE -# define OS_CODE 3 /* assume Unix */ -#endif - -#ifndef F_OPEN -# define F_OPEN(name, mode) fopen((name), (mode)) -#endif - - /* functions */ - -#if defined(pyr) || defined(Z_SOLO) -# define NO_MEMCPY -#endif -#if defined(SMALL_MEDIUM) && !defined(_MSC_VER) && !defined(__SC__) - /* Use our own functions for small and medium model with MSC <= 5.0. - * You may have to use the same strategy for Borland C (untested). - * The __SC__ check is for Symantec. - */ -# define NO_MEMCPY -#endif -#if defined(STDC) && !defined(HAVE_MEMCPY) && !defined(NO_MEMCPY) -# define HAVE_MEMCPY -#endif -#ifdef HAVE_MEMCPY -# ifdef SMALL_MEDIUM /* MSDOS small or medium model */ -# define zmemcpy _fmemcpy -# define zmemcmp _fmemcmp -# define zmemzero(dest, len) _fmemset(dest, 0, len) -# else -# define zmemcpy memcpy -# define zmemcmp memcmp -# define zmemzero(dest, len) memset(dest, 0, len) -# endif -#else - void ZLIB_INTERNAL zmemcpy OF((Bytef* dest, const Bytef* source, uInt len)); - int ZLIB_INTERNAL zmemcmp OF((const Bytef* s1, const Bytef* s2, uInt len)); - void ZLIB_INTERNAL zmemzero OF((Bytef* dest, uInt len)); -#endif - -/* Diagnostic functions */ -#ifdef ZLIB_DEBUG -# include - extern int ZLIB_INTERNAL z_verbose; - extern void ZLIB_INTERNAL z_error OF((char *m)); -# define Assert(cond,msg) {if(!(cond)) z_error(msg);} -# define Trace(x) {if (z_verbose>=0) fprintf x ;} -# define Tracev(x) {if (z_verbose>0) fprintf x ;} -# define Tracevv(x) {if (z_verbose>1) fprintf x ;} -# define Tracec(c,x) {if (z_verbose>0 && (c)) fprintf x ;} -# define Tracecv(c,x) {if (z_verbose>1 && (c)) fprintf x ;} -#else -# define Assert(cond,msg) -# define Trace(x) -# define Tracev(x) -# define Tracevv(x) -# define Tracec(c,x) -# define Tracecv(c,x) -#endif - -#ifndef Z_SOLO - voidpf ZLIB_INTERNAL zcalloc OF((voidpf opaque, unsigned items, - unsigned size)); - void ZLIB_INTERNAL zcfree OF((voidpf opaque, voidpf ptr)); -#endif - -#define ZALLOC(strm, items, size) \ - (*((strm)->zalloc))((strm)->opaque, (items), (size)) -#define ZFREE(strm, addr) (*((strm)->zfree))((strm)->opaque, (voidpf)(addr)) -#define TRY_FREE(s, p) {if (p) ZFREE(s, p);} - -/* Reverse the bytes in a 32-bit value */ -#define ZSWAP32(q) ((((q) >> 24) & 0xff) + (((q) >> 8) & 0xff00) + \ - (((q) & 0xff00) << 8) + (((q) & 0xff) << 24)) - -#endif /* ZUTIL_H */ diff --git a/dep/zlib/zlib.vcxproj b/dep/zlib/zlib.vcxproj deleted file mode 100644 index 1bfd57f46..000000000 --- a/dep/zlib/zlib.vcxproj +++ /dev/null @@ -1,50 +0,0 @@ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - {7FF9FDB9-D504-47DB-A16A-B08071999620} - - - - - - - TurnOffAllWarnings - $(ProjectDir)include;%(AdditionalIncludeDirectories) - - - - - \ No newline at end of file diff --git a/dep/zlib/zlib.vcxproj.filters b/dep/zlib/zlib.vcxproj.filters deleted file mode 100644 index 93a8b89a4..000000000 --- a/dep/zlib/zlib.vcxproj.filters +++ /dev/null @@ -1,33 +0,0 @@ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - \ No newline at end of file diff --git a/dep/zstd/CMakeLists.txt b/dep/zstd/CMakeLists.txt deleted file mode 100644 index 12a41904f..000000000 --- a/dep/zstd/CMakeLists.txt +++ /dev/null @@ -1,38 +0,0 @@ -enable_language(C ASM) - -add_library(zstd - lib/common/debug.c - lib/common/entropy_common.c - lib/common/error_private.c - lib/common/fse_decompress.c - lib/common/pool.c - lib/common/threading.c - lib/common/xxhash.c - lib/common/zstd_common.c - lib/compress/fse_compress.c - lib/compress/hist.c - lib/compress/huf_compress.c - lib/compress/zstd_compress.c - lib/compress/zstd_compress_literals.c - lib/compress/zstd_compress_sequences.c - lib/compress/zstd_compress_superblock.c - lib/compress/zstd_double_fast.c - lib/compress/zstd_fast.c - lib/compress/zstd_lazy.c - lib/compress/zstd_ldm.c - lib/compress/zstdmt_compress.c - lib/compress/zstd_opt.c - lib/decompress/huf_decompress.c - lib/decompress/zstd_ddict.c - lib/decompress/zstd_decompress_block.c - lib/decompress/zstd_decompress.c -) - -if(NOT MSVC AND CPU_ARCH_X64) - target_sources(zstd PRIVATE lib/decompress/huf_decompress_amd64.S) -endif() - -target_include_directories(zstd PUBLIC "${CMAKE_CURRENT_SOURCE_DIR}/lib") -disable_compiler_warnings_for_target(zstd) - -add_library(Zstd::Zstd ALIAS zstd) diff --git a/dep/zstd/LICENSE b/dep/zstd/LICENSE deleted file mode 100644 index a793a8028..000000000 --- a/dep/zstd/LICENSE +++ /dev/null @@ -1,30 +0,0 @@ -BSD License - -For Zstandard software - -Copyright (c) 2016-present, Facebook, Inc. All rights reserved. - -Redistribution and use in source and binary forms, with or without modification, -are permitted provided that the following conditions are met: - - * Redistributions of source code must retain the above copyright notice, this - list of conditions and the following disclaimer. - - * Redistributions in binary form must reproduce the above copyright notice, - this list of conditions and the following disclaimer in the documentation - and/or other materials provided with the distribution. - - * Neither the name Facebook nor the names of its contributors may be used to - endorse or promote products derived from this software without specific - prior written permission. - -THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND -ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED -WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE -DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE FOR -ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES -(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; -LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON -ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT -(INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS -SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. diff --git a/dep/zstd/lib/common/bitstream.h b/dep/zstd/lib/common/bitstream.h deleted file mode 100644 index 84b6062ff..000000000 --- a/dep/zstd/lib/common/bitstream.h +++ /dev/null @@ -1,478 +0,0 @@ -/* ****************************************************************** - * bitstream - * Part of FSE library - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -****************************************************************** */ -#ifndef BITSTREAM_H_MODULE -#define BITSTREAM_H_MODULE - -#if defined (__cplusplus) -extern "C" { -#endif -/* -* This API consists of small unitary functions, which must be inlined for best performance. -* Since link-time-optimization is not available for all compilers, -* these functions are defined into a .h to be included. -*/ - -/*-**************************************** -* Dependencies -******************************************/ -#include "mem.h" /* unaligned access routines */ -#include "compiler.h" /* UNLIKELY() */ -#include "debug.h" /* assert(), DEBUGLOG(), RAWLOG() */ -#include "error_private.h" /* error codes and messages */ - - -/*========================================= -* Target specific -=========================================*/ -#ifndef ZSTD_NO_INTRINSICS -# if defined(__BMI__) && defined(__GNUC__) -# include /* support for bextr (experimental) */ -# elif defined(__ICCARM__) -# include -# endif -#endif - -#define STREAM_ACCUMULATOR_MIN_32 25 -#define STREAM_ACCUMULATOR_MIN_64 57 -#define STREAM_ACCUMULATOR_MIN ((U32)(MEM_32bits() ? STREAM_ACCUMULATOR_MIN_32 : STREAM_ACCUMULATOR_MIN_64)) - - -/*-****************************************** -* bitStream encoding API (write forward) -********************************************/ -/* bitStream can mix input from multiple sources. - * A critical property of these streams is that they encode and decode in **reverse** direction. - * So the first bit sequence you add will be the last to be read, like a LIFO stack. - */ -typedef struct { - size_t bitContainer; - unsigned bitPos; - char* startPtr; - char* ptr; - char* endPtr; -} BIT_CStream_t; - -MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC, void* dstBuffer, size_t dstCapacity); -MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC, size_t value, unsigned nbBits); -MEM_STATIC void BIT_flushBits(BIT_CStream_t* bitC); -MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC); - -/* Start with initCStream, providing the size of buffer to write into. -* bitStream will never write outside of this buffer. -* `dstCapacity` must be >= sizeof(bitD->bitContainer), otherwise @return will be an error code. -* -* bits are first added to a local register. -* Local register is size_t, hence 64-bits on 64-bits systems, or 32-bits on 32-bits systems. -* Writing data into memory is an explicit operation, performed by the flushBits function. -* Hence keep track how many bits are potentially stored into local register to avoid register overflow. -* After a flushBits, a maximum of 7 bits might still be stored into local register. -* -* Avoid storing elements of more than 24 bits if you want compatibility with 32-bits bitstream readers. -* -* Last operation is to close the bitStream. -* The function returns the final size of CStream in bytes. -* If data couldn't fit into `dstBuffer`, it will return a 0 ( == not storable) -*/ - - -/*-******************************************** -* bitStream decoding API (read backward) -**********************************************/ -typedef struct { - size_t bitContainer; - unsigned bitsConsumed; - const char* ptr; - const char* start; - const char* limitPtr; -} BIT_DStream_t; - -typedef enum { BIT_DStream_unfinished = 0, - BIT_DStream_endOfBuffer = 1, - BIT_DStream_completed = 2, - BIT_DStream_overflow = 3 } BIT_DStream_status; /* result of BIT_reloadDStream() */ - /* 1,2,4,8 would be better for bitmap combinations, but slows down performance a bit ... :( */ - -MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize); -MEM_STATIC size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits); -MEM_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD); -MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* bitD); - - -/* Start by invoking BIT_initDStream(). -* A chunk of the bitStream is then stored into a local register. -* Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t). -* You can then retrieve bitFields stored into the local register, **in reverse order**. -* Local register is explicitly reloaded from memory by the BIT_reloadDStream() method. -* A reload guarantee a minimum of ((8*sizeof(bitD->bitContainer))-7) bits when its result is BIT_DStream_unfinished. -* Otherwise, it can be less than that, so proceed accordingly. -* Checking if DStream has reached its end can be performed with BIT_endOfDStream(). -*/ - - -/*-**************************************** -* unsafe API -******************************************/ -MEM_STATIC void BIT_addBitsFast(BIT_CStream_t* bitC, size_t value, unsigned nbBits); -/* faster, but works only if value is "clean", meaning all high bits above nbBits are 0 */ - -MEM_STATIC void BIT_flushBitsFast(BIT_CStream_t* bitC); -/* unsafe version; does not check buffer overflow */ - -MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits); -/* faster, but works only if nbBits >= 1 */ - - - -/*-************************************************************** -* Internal functions -****************************************************************/ -MEM_STATIC unsigned BIT_highbit32 (U32 val) -{ - assert(val != 0); - { -# if defined(_MSC_VER) /* Visual */ -# if STATIC_BMI2 == 1 - return _lzcnt_u32(val) ^ 31; -# else - if (val != 0) { - unsigned long r; - _BitScanReverse(&r, val); - return (unsigned)r; - } else { - /* Should not reach this code path */ - __assume(0); - } -# endif -# elif defined(__GNUC__) && (__GNUC__ >= 3) /* Use GCC Intrinsic */ - return __builtin_clz (val) ^ 31; -# elif defined(__ICCARM__) /* IAR Intrinsic */ - return 31 - __CLZ(val); -# else /* Software version */ - static const unsigned DeBruijnClz[32] = { 0, 9, 1, 10, 13, 21, 2, 29, - 11, 14, 16, 18, 22, 25, 3, 30, - 8, 12, 20, 28, 15, 17, 24, 7, - 19, 27, 23, 6, 26, 5, 4, 31 }; - U32 v = val; - v |= v >> 1; - v |= v >> 2; - v |= v >> 4; - v |= v >> 8; - v |= v >> 16; - return DeBruijnClz[ (U32) (v * 0x07C4ACDDU) >> 27]; -# endif - } -} - -/*===== Local Constants =====*/ -static const unsigned BIT_mask[] = { - 0, 1, 3, 7, 0xF, 0x1F, - 0x3F, 0x7F, 0xFF, 0x1FF, 0x3FF, 0x7FF, - 0xFFF, 0x1FFF, 0x3FFF, 0x7FFF, 0xFFFF, 0x1FFFF, - 0x3FFFF, 0x7FFFF, 0xFFFFF, 0x1FFFFF, 0x3FFFFF, 0x7FFFFF, - 0xFFFFFF, 0x1FFFFFF, 0x3FFFFFF, 0x7FFFFFF, 0xFFFFFFF, 0x1FFFFFFF, - 0x3FFFFFFF, 0x7FFFFFFF}; /* up to 31 bits */ -#define BIT_MASK_SIZE (sizeof(BIT_mask) / sizeof(BIT_mask[0])) - -/*-************************************************************** -* bitStream encoding -****************************************************************/ -/*! BIT_initCStream() : - * `dstCapacity` must be > sizeof(size_t) - * @return : 0 if success, - * otherwise an error code (can be tested using ERR_isError()) */ -MEM_STATIC size_t BIT_initCStream(BIT_CStream_t* bitC, - void* startPtr, size_t dstCapacity) -{ - bitC->bitContainer = 0; - bitC->bitPos = 0; - bitC->startPtr = (char*)startPtr; - bitC->ptr = bitC->startPtr; - bitC->endPtr = bitC->startPtr + dstCapacity - sizeof(bitC->bitContainer); - if (dstCapacity <= sizeof(bitC->bitContainer)) return ERROR(dstSize_tooSmall); - return 0; -} - -/*! BIT_addBits() : - * can add up to 31 bits into `bitC`. - * Note : does not check for register overflow ! */ -MEM_STATIC void BIT_addBits(BIT_CStream_t* bitC, - size_t value, unsigned nbBits) -{ - DEBUG_STATIC_ASSERT(BIT_MASK_SIZE == 32); - assert(nbBits < BIT_MASK_SIZE); - assert(nbBits + bitC->bitPos < sizeof(bitC->bitContainer) * 8); - bitC->bitContainer |= (value & BIT_mask[nbBits]) << bitC->bitPos; - bitC->bitPos += nbBits; -} - -/*! BIT_addBitsFast() : - * works only if `value` is _clean_, - * meaning all high bits above nbBits are 0 */ -MEM_STATIC void BIT_addBitsFast(BIT_CStream_t* bitC, - size_t value, unsigned nbBits) -{ - assert((value>>nbBits) == 0); - assert(nbBits + bitC->bitPos < sizeof(bitC->bitContainer) * 8); - bitC->bitContainer |= value << bitC->bitPos; - bitC->bitPos += nbBits; -} - -/*! BIT_flushBitsFast() : - * assumption : bitContainer has not overflowed - * unsafe version; does not check buffer overflow */ -MEM_STATIC void BIT_flushBitsFast(BIT_CStream_t* bitC) -{ - size_t const nbBytes = bitC->bitPos >> 3; - assert(bitC->bitPos < sizeof(bitC->bitContainer) * 8); - assert(bitC->ptr <= bitC->endPtr); - MEM_writeLEST(bitC->ptr, bitC->bitContainer); - bitC->ptr += nbBytes; - bitC->bitPos &= 7; - bitC->bitContainer >>= nbBytes*8; -} - -/*! BIT_flushBits() : - * assumption : bitContainer has not overflowed - * safe version; check for buffer overflow, and prevents it. - * note : does not signal buffer overflow. - * overflow will be revealed later on using BIT_closeCStream() */ -MEM_STATIC void BIT_flushBits(BIT_CStream_t* bitC) -{ - size_t const nbBytes = bitC->bitPos >> 3; - assert(bitC->bitPos < sizeof(bitC->bitContainer) * 8); - assert(bitC->ptr <= bitC->endPtr); - MEM_writeLEST(bitC->ptr, bitC->bitContainer); - bitC->ptr += nbBytes; - if (bitC->ptr > bitC->endPtr) bitC->ptr = bitC->endPtr; - bitC->bitPos &= 7; - bitC->bitContainer >>= nbBytes*8; -} - -/*! BIT_closeCStream() : - * @return : size of CStream, in bytes, - * or 0 if it could not fit into dstBuffer */ -MEM_STATIC size_t BIT_closeCStream(BIT_CStream_t* bitC) -{ - BIT_addBitsFast(bitC, 1, 1); /* endMark */ - BIT_flushBits(bitC); - if (bitC->ptr >= bitC->endPtr) return 0; /* overflow detected */ - return (bitC->ptr - bitC->startPtr) + (bitC->bitPos > 0); -} - - -/*-******************************************************** -* bitStream decoding -**********************************************************/ -/*! BIT_initDStream() : - * Initialize a BIT_DStream_t. - * `bitD` : a pointer to an already allocated BIT_DStream_t structure. - * `srcSize` must be the *exact* size of the bitStream, in bytes. - * @return : size of stream (== srcSize), or an errorCode if a problem is detected - */ -MEM_STATIC size_t BIT_initDStream(BIT_DStream_t* bitD, const void* srcBuffer, size_t srcSize) -{ - if (srcSize < 1) { ZSTD_memset(bitD, 0, sizeof(*bitD)); return ERROR(srcSize_wrong); } - - bitD->start = (const char*)srcBuffer; - bitD->limitPtr = bitD->start + sizeof(bitD->bitContainer); - - if (srcSize >= sizeof(bitD->bitContainer)) { /* normal case */ - bitD->ptr = (const char*)srcBuffer + srcSize - sizeof(bitD->bitContainer); - bitD->bitContainer = MEM_readLEST(bitD->ptr); - { BYTE const lastByte = ((const BYTE*)srcBuffer)[srcSize-1]; - bitD->bitsConsumed = lastByte ? 8 - BIT_highbit32(lastByte) : 0; /* ensures bitsConsumed is always set */ - if (lastByte == 0) return ERROR(GENERIC); /* endMark not present */ } - } else { - bitD->ptr = bitD->start; - bitD->bitContainer = *(const BYTE*)(bitD->start); - switch(srcSize) - { - case 7: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[6]) << (sizeof(bitD->bitContainer)*8 - 16); - ZSTD_FALLTHROUGH; - - case 6: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[5]) << (sizeof(bitD->bitContainer)*8 - 24); - ZSTD_FALLTHROUGH; - - case 5: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[4]) << (sizeof(bitD->bitContainer)*8 - 32); - ZSTD_FALLTHROUGH; - - case 4: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[3]) << 24; - ZSTD_FALLTHROUGH; - - case 3: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[2]) << 16; - ZSTD_FALLTHROUGH; - - case 2: bitD->bitContainer += (size_t)(((const BYTE*)(srcBuffer))[1]) << 8; - ZSTD_FALLTHROUGH; - - default: break; - } - { BYTE const lastByte = ((const BYTE*)srcBuffer)[srcSize-1]; - bitD->bitsConsumed = lastByte ? 8 - BIT_highbit32(lastByte) : 0; - if (lastByte == 0) return ERROR(corruption_detected); /* endMark not present */ - } - bitD->bitsConsumed += (U32)(sizeof(bitD->bitContainer) - srcSize)*8; - } - - return srcSize; -} - -MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getUpperBits(size_t bitContainer, U32 const start) -{ - return bitContainer >> start; -} - -MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getMiddleBits(size_t bitContainer, U32 const start, U32 const nbBits) -{ - U32 const regMask = sizeof(bitContainer)*8 - 1; - /* if start > regMask, bitstream is corrupted, and result is undefined */ - assert(nbBits < BIT_MASK_SIZE); - /* x86 transform & ((1 << nbBits) - 1) to bzhi instruction, it is better - * than accessing memory. When bmi2 instruction is not present, we consider - * such cpus old (pre-Haswell, 2013) and their performance is not of that - * importance. - */ -#if defined(__x86_64__) || defined(_M_X86) - return (bitContainer >> (start & regMask)) & ((((U64)1) << nbBits) - 1); -#else - return (bitContainer >> (start & regMask)) & BIT_mask[nbBits]; -#endif -} - -MEM_STATIC FORCE_INLINE_ATTR size_t BIT_getLowerBits(size_t bitContainer, U32 const nbBits) -{ -#if defined(STATIC_BMI2) && STATIC_BMI2 == 1 - return _bzhi_u64(bitContainer, nbBits); -#else - assert(nbBits < BIT_MASK_SIZE); - return bitContainer & BIT_mask[nbBits]; -#endif -} - -/*! BIT_lookBits() : - * Provides next n bits from local register. - * local register is not modified. - * On 32-bits, maxNbBits==24. - * On 64-bits, maxNbBits==56. - * @return : value extracted */ -MEM_STATIC FORCE_INLINE_ATTR size_t BIT_lookBits(const BIT_DStream_t* bitD, U32 nbBits) -{ - /* arbitrate between double-shift and shift+mask */ -#if 1 - /* if bitD->bitsConsumed + nbBits > sizeof(bitD->bitContainer)*8, - * bitstream is likely corrupted, and result is undefined */ - return BIT_getMiddleBits(bitD->bitContainer, (sizeof(bitD->bitContainer)*8) - bitD->bitsConsumed - nbBits, nbBits); -#else - /* this code path is slower on my os-x laptop */ - U32 const regMask = sizeof(bitD->bitContainer)*8 - 1; - return ((bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> 1) >> ((regMask-nbBits) & regMask); -#endif -} - -/*! BIT_lookBitsFast() : - * unsafe version; only works if nbBits >= 1 */ -MEM_STATIC size_t BIT_lookBitsFast(const BIT_DStream_t* bitD, U32 nbBits) -{ - U32 const regMask = sizeof(bitD->bitContainer)*8 - 1; - assert(nbBits >= 1); - return (bitD->bitContainer << (bitD->bitsConsumed & regMask)) >> (((regMask+1)-nbBits) & regMask); -} - -MEM_STATIC FORCE_INLINE_ATTR void BIT_skipBits(BIT_DStream_t* bitD, U32 nbBits) -{ - bitD->bitsConsumed += nbBits; -} - -/*! BIT_readBits() : - * Read (consume) next n bits from local register and update. - * Pay attention to not read more than nbBits contained into local register. - * @return : extracted value. */ -MEM_STATIC FORCE_INLINE_ATTR size_t BIT_readBits(BIT_DStream_t* bitD, unsigned nbBits) -{ - size_t const value = BIT_lookBits(bitD, nbBits); - BIT_skipBits(bitD, nbBits); - return value; -} - -/*! BIT_readBitsFast() : - * unsafe version; only works only if nbBits >= 1 */ -MEM_STATIC size_t BIT_readBitsFast(BIT_DStream_t* bitD, unsigned nbBits) -{ - size_t const value = BIT_lookBitsFast(bitD, nbBits); - assert(nbBits >= 1); - BIT_skipBits(bitD, nbBits); - return value; -} - -/*! BIT_reloadDStreamFast() : - * Similar to BIT_reloadDStream(), but with two differences: - * 1. bitsConsumed <= sizeof(bitD->bitContainer)*8 must hold! - * 2. Returns BIT_DStream_overflow when bitD->ptr < bitD->limitPtr, at this - * point you must use BIT_reloadDStream() to reload. - */ -MEM_STATIC BIT_DStream_status BIT_reloadDStreamFast(BIT_DStream_t* bitD) -{ - if (UNLIKELY(bitD->ptr < bitD->limitPtr)) - return BIT_DStream_overflow; - assert(bitD->bitsConsumed <= sizeof(bitD->bitContainer)*8); - bitD->ptr -= bitD->bitsConsumed >> 3; - bitD->bitsConsumed &= 7; - bitD->bitContainer = MEM_readLEST(bitD->ptr); - return BIT_DStream_unfinished; -} - -/*! BIT_reloadDStream() : - * Refill `bitD` from buffer previously set in BIT_initDStream() . - * This function is safe, it guarantees it will not read beyond src buffer. - * @return : status of `BIT_DStream_t` internal register. - * when status == BIT_DStream_unfinished, internal register is filled with at least 25 or 57 bits */ -MEM_STATIC BIT_DStream_status BIT_reloadDStream(BIT_DStream_t* bitD) -{ - if (bitD->bitsConsumed > (sizeof(bitD->bitContainer)*8)) /* overflow detected, like end of stream */ - return BIT_DStream_overflow; - - if (bitD->ptr >= bitD->limitPtr) { - return BIT_reloadDStreamFast(bitD); - } - if (bitD->ptr == bitD->start) { - if (bitD->bitsConsumed < sizeof(bitD->bitContainer)*8) return BIT_DStream_endOfBuffer; - return BIT_DStream_completed; - } - /* start < ptr < limitPtr */ - { U32 nbBytes = bitD->bitsConsumed >> 3; - BIT_DStream_status result = BIT_DStream_unfinished; - if (bitD->ptr - nbBytes < bitD->start) { - nbBytes = (U32)(bitD->ptr - bitD->start); /* ptr > start */ - result = BIT_DStream_endOfBuffer; - } - bitD->ptr -= nbBytes; - bitD->bitsConsumed -= nbBytes*8; - bitD->bitContainer = MEM_readLEST(bitD->ptr); /* reminder : srcSize > sizeof(bitD->bitContainer), otherwise bitD->ptr == bitD->start */ - return result; - } -} - -/*! BIT_endOfDStream() : - * @return : 1 if DStream has _exactly_ reached its end (all bits consumed). - */ -MEM_STATIC unsigned BIT_endOfDStream(const BIT_DStream_t* DStream) -{ - return ((DStream->ptr == DStream->start) && (DStream->bitsConsumed == sizeof(DStream->bitContainer)*8)); -} - -#if defined (__cplusplus) -} -#endif - -#endif /* BITSTREAM_H_MODULE */ diff --git a/dep/zstd/lib/common/compiler.h b/dep/zstd/lib/common/compiler.h deleted file mode 100644 index 516930c01..000000000 --- a/dep/zstd/lib/common/compiler.h +++ /dev/null @@ -1,335 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_COMPILER_H -#define ZSTD_COMPILER_H - -#include "portability_macros.h" - -/*-******************************************************* -* Compiler specifics -*********************************************************/ -/* force inlining */ - -#if !defined(ZSTD_NO_INLINE) -#if (defined(__GNUC__) && !defined(__STRICT_ANSI__)) || defined(__cplusplus) || defined(__STDC_VERSION__) && __STDC_VERSION__ >= 199901L /* C99 */ -# define INLINE_KEYWORD inline -#else -# define INLINE_KEYWORD -#endif - -#if defined(__GNUC__) || defined(__ICCARM__) -# define FORCE_INLINE_ATTR __attribute__((always_inline)) -#elif defined(_MSC_VER) -# define FORCE_INLINE_ATTR __forceinline -#else -# define FORCE_INLINE_ATTR -#endif - -#else - -#define INLINE_KEYWORD -#define FORCE_INLINE_ATTR - -#endif - -/** - On MSVC qsort requires that functions passed into it use the __cdecl calling conversion(CC). - This explicitly marks such functions as __cdecl so that the code will still compile - if a CC other than __cdecl has been made the default. -*/ -#if defined(_MSC_VER) -# define WIN_CDECL __cdecl -#else -# define WIN_CDECL -#endif - -/** - * FORCE_INLINE_TEMPLATE is used to define C "templates", which take constant - * parameters. They must be inlined for the compiler to eliminate the constant - * branches. - */ -#define FORCE_INLINE_TEMPLATE static INLINE_KEYWORD FORCE_INLINE_ATTR -/** - * HINT_INLINE is used to help the compiler generate better code. It is *not* - * used for "templates", so it can be tweaked based on the compilers - * performance. - * - * gcc-4.8 and gcc-4.9 have been shown to benefit from leaving off the - * always_inline attribute. - * - * clang up to 5.0.0 (trunk) benefit tremendously from the always_inline - * attribute. - */ -#if !defined(__clang__) && defined(__GNUC__) && __GNUC__ >= 4 && __GNUC_MINOR__ >= 8 && __GNUC__ < 5 -# define HINT_INLINE static INLINE_KEYWORD -#else -# define HINT_INLINE static INLINE_KEYWORD FORCE_INLINE_ATTR -#endif - -/* UNUSED_ATTR tells the compiler it is okay if the function is unused. */ -#if defined(__GNUC__) -# define UNUSED_ATTR __attribute__((unused)) -#else -# define UNUSED_ATTR -#endif - -/* force no inlining */ -#ifdef _MSC_VER -# define FORCE_NOINLINE static __declspec(noinline) -#else -# if defined(__GNUC__) || defined(__ICCARM__) -# define FORCE_NOINLINE static __attribute__((__noinline__)) -# else -# define FORCE_NOINLINE static -# endif -#endif - - -/* target attribute */ -#if defined(__GNUC__) || defined(__ICCARM__) -# define TARGET_ATTRIBUTE(target) __attribute__((__target__(target))) -#else -# define TARGET_ATTRIBUTE(target) -#endif - -/* Target attribute for BMI2 dynamic dispatch. - * Enable lzcnt, bmi, and bmi2. - * We test for bmi1 & bmi2. lzcnt is included in bmi1. - */ -#define BMI2_TARGET_ATTRIBUTE TARGET_ATTRIBUTE("lzcnt,bmi,bmi2") - -/* prefetch - * can be disabled, by declaring NO_PREFETCH build macro */ -#if defined(NO_PREFETCH) -# define PREFETCH_L1(ptr) (void)(ptr) /* disabled */ -# define PREFETCH_L2(ptr) (void)(ptr) /* disabled */ -#else -# if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86)) /* _mm_prefetch() is not defined outside of x86/x64 */ -# include /* https://msdn.microsoft.com/fr-fr/library/84szxsww(v=vs.90).aspx */ -# define PREFETCH_L1(ptr) _mm_prefetch((const char*)(ptr), _MM_HINT_T0) -# define PREFETCH_L2(ptr) _mm_prefetch((const char*)(ptr), _MM_HINT_T1) -# elif defined(__GNUC__) && ( (__GNUC__ >= 4) || ( (__GNUC__ == 3) && (__GNUC_MINOR__ >= 1) ) ) -# define PREFETCH_L1(ptr) __builtin_prefetch((ptr), 0 /* rw==read */, 3 /* locality */) -# define PREFETCH_L2(ptr) __builtin_prefetch((ptr), 0 /* rw==read */, 2 /* locality */) -# elif defined(__aarch64__) -# define PREFETCH_L1(ptr) __asm__ __volatile__("prfm pldl1keep, %0" ::"Q"(*(ptr))) -# define PREFETCH_L2(ptr) __asm__ __volatile__("prfm pldl2keep, %0" ::"Q"(*(ptr))) -# else -# define PREFETCH_L1(ptr) (void)(ptr) /* disabled */ -# define PREFETCH_L2(ptr) (void)(ptr) /* disabled */ -# endif -#endif /* NO_PREFETCH */ - -#define CACHELINE_SIZE 64 - -#define PREFETCH_AREA(p, s) { \ - const char* const _ptr = (const char*)(p); \ - size_t const _size = (size_t)(s); \ - size_t _pos; \ - for (_pos=0; _pos<_size; _pos+=CACHELINE_SIZE) { \ - PREFETCH_L2(_ptr + _pos); \ - } \ -} - -/* vectorization - * older GCC (pre gcc-4.3 picked as the cutoff) uses a different syntax, - * and some compilers, like Intel ICC and MCST LCC, do not support it at all. */ -#if !defined(__INTEL_COMPILER) && !defined(__clang__) && defined(__GNUC__) && !defined(__LCC__) -# if (__GNUC__ == 4 && __GNUC_MINOR__ > 3) || (__GNUC__ >= 5) -# define DONT_VECTORIZE __attribute__((optimize("no-tree-vectorize"))) -# else -# define DONT_VECTORIZE _Pragma("GCC optimize(\"no-tree-vectorize\")") -# endif -#else -# define DONT_VECTORIZE -#endif - -/* Tell the compiler that a branch is likely or unlikely. - * Only use these macros if it causes the compiler to generate better code. - * If you can remove a LIKELY/UNLIKELY annotation without speed changes in gcc - * and clang, please do. - */ -#if defined(__GNUC__) -#define LIKELY(x) (__builtin_expect((x), 1)) -#define UNLIKELY(x) (__builtin_expect((x), 0)) -#else -#define LIKELY(x) (x) -#define UNLIKELY(x) (x) -#endif - -/* disable warnings */ -#ifdef _MSC_VER /* Visual Studio */ -# include /* For Visual 2005 */ -# pragma warning(disable : 4100) /* disable: C4100: unreferenced formal parameter */ -# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */ -# pragma warning(disable : 4204) /* disable: C4204: non-constant aggregate initializer */ -# pragma warning(disable : 4214) /* disable: C4214: non-int bitfields */ -# pragma warning(disable : 4324) /* disable: C4324: padded structure */ -#endif - -/*Like DYNAMIC_BMI2 but for compile time determination of BMI2 support*/ -#ifndef STATIC_BMI2 -# if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_I86)) -# ifdef __AVX2__ //MSVC does not have a BMI2 specific flag, but every CPU that supports AVX2 also supports BMI2 -# define STATIC_BMI2 1 -# endif -# endif -#endif - -#ifndef STATIC_BMI2 - #define STATIC_BMI2 0 -#endif - -/* compile time determination of SIMD support */ -#if !defined(ZSTD_NO_INTRINSICS) -# if defined(__SSE2__) || defined(_M_AMD64) || (defined (_M_IX86) && defined(_M_IX86_FP) && (_M_IX86_FP >= 2)) -# define ZSTD_ARCH_X86_SSE2 -# endif -# if defined(__ARM_NEON) || defined(_M_ARM64) -# define ZSTD_ARCH_ARM_NEON -# endif -# -# if defined(ZSTD_ARCH_X86_SSE2) -# include -# elif defined(ZSTD_ARCH_ARM_NEON) -# include -# endif -#endif - -/* C-language Attributes are added in C23. */ -#if defined(__STDC_VERSION__) && (__STDC_VERSION__ > 201710L) && defined(__has_c_attribute) -# define ZSTD_HAS_C_ATTRIBUTE(x) __has_c_attribute(x) -#else -# define ZSTD_HAS_C_ATTRIBUTE(x) 0 -#endif - -/* Only use C++ attributes in C++. Some compilers report support for C++ - * attributes when compiling with C. - */ -#if defined(__cplusplus) && defined(__has_cpp_attribute) -# define ZSTD_HAS_CPP_ATTRIBUTE(x) __has_cpp_attribute(x) -#else -# define ZSTD_HAS_CPP_ATTRIBUTE(x) 0 -#endif - -/* Define ZSTD_FALLTHROUGH macro for annotating switch case with the 'fallthrough' attribute. - * - C23: https://en.cppreference.com/w/c/language/attributes/fallthrough - * - CPP17: https://en.cppreference.com/w/cpp/language/attributes/fallthrough - * - Else: __attribute__((__fallthrough__)) - */ -#ifndef ZSTD_FALLTHROUGH -# if ZSTD_HAS_C_ATTRIBUTE(fallthrough) -# define ZSTD_FALLTHROUGH [[fallthrough]] -# elif ZSTD_HAS_CPP_ATTRIBUTE(fallthrough) -# define ZSTD_FALLTHROUGH [[fallthrough]] -# elif __has_attribute(__fallthrough__) -/* Leading semicolon is to satisfy gcc-11 with -pedantic. Without the semicolon - * gcc complains about: a label can only be part of a statement and a declaration is not a statement. - */ -# define ZSTD_FALLTHROUGH ; __attribute__((__fallthrough__)) -# else -# define ZSTD_FALLTHROUGH -# endif -#endif - -/*-************************************************************** -* Alignment check -*****************************************************************/ - -/* this test was initially positioned in mem.h, - * but this file is removed (or replaced) for linux kernel - * so it's now hosted in compiler.h, - * which remains valid for both user & kernel spaces. - */ - -#ifndef ZSTD_ALIGNOF -# if defined(__GNUC__) || defined(_MSC_VER) -/* covers gcc, clang & MSVC */ -/* note : this section must come first, before C11, - * due to a limitation in the kernel source generator */ -# define ZSTD_ALIGNOF(T) __alignof(T) - -# elif defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L) -/* C11 support */ -# include -# define ZSTD_ALIGNOF(T) alignof(T) - -# else -/* No known support for alignof() - imperfect backup */ -# define ZSTD_ALIGNOF(T) (sizeof(void*) < sizeof(T) ? sizeof(void*) : sizeof(T)) - -# endif -#endif /* ZSTD_ALIGNOF */ - -/*-************************************************************** -* Sanitizer -*****************************************************************/ - -#if ZSTD_MEMORY_SANITIZER -/* Not all platforms that support msan provide sanitizers/msan_interface.h. - * We therefore declare the functions we need ourselves, rather than trying to - * include the header file... */ -#include /* size_t */ -#define ZSTD_DEPS_NEED_STDINT -#include "zstd_deps.h" /* intptr_t */ - -/* Make memory region fully initialized (without changing its contents). */ -void __msan_unpoison(const volatile void *a, size_t size); - -/* Make memory region fully uninitialized (without changing its contents). - This is a legacy interface that does not update origin information. Use - __msan_allocated_memory() instead. */ -void __msan_poison(const volatile void *a, size_t size); - -/* Returns the offset of the first (at least partially) poisoned byte in the - memory range, or -1 if the whole range is good. */ -intptr_t __msan_test_shadow(const volatile void *x, size_t size); -#endif - -#if ZSTD_ADDRESS_SANITIZER -/* Not all platforms that support asan provide sanitizers/asan_interface.h. - * We therefore declare the functions we need ourselves, rather than trying to - * include the header file... */ -#include /* size_t */ - -/** - * Marks a memory region ([addr, addr+size)) as unaddressable. - * - * This memory must be previously allocated by your program. Instrumented - * code is forbidden from accessing addresses in this region until it is - * unpoisoned. This function is not guaranteed to poison the entire region - - * it could poison only a subregion of [addr, addr+size) due to ASan - * alignment restrictions. - * - * \note This function is not thread-safe because no two threads can poison or - * unpoison memory in the same memory region simultaneously. - * - * \param addr Start of memory region. - * \param size Size of memory region. */ -void __asan_poison_memory_region(void const volatile *addr, size_t size); - -/** - * Marks a memory region ([addr, addr+size)) as addressable. - * - * This memory must be previously allocated by your program. Accessing - * addresses in this region is allowed until this region is poisoned again. - * This function could unpoison a super-region of [addr, addr+size) due - * to ASan alignment restrictions. - * - * \note This function is not thread-safe because no two threads can - * poison or unpoison memory in the same memory region simultaneously. - * - * \param addr Start of memory region. - * \param size Size of memory region. */ -void __asan_unpoison_memory_region(void const volatile *addr, size_t size); -#endif - -#endif /* ZSTD_COMPILER_H */ diff --git a/dep/zstd/lib/common/cpu.h b/dep/zstd/lib/common/cpu.h deleted file mode 100644 index 8acd33be3..000000000 --- a/dep/zstd/lib/common/cpu.h +++ /dev/null @@ -1,213 +0,0 @@ -/* - * Copyright (c) Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_COMMON_CPU_H -#define ZSTD_COMMON_CPU_H - -/** - * Implementation taken from folly/CpuId.h - * https://github.com/facebook/folly/blob/master/folly/CpuId.h - */ - -#include "mem.h" - -#ifdef _MSC_VER -#include -#endif - -typedef struct { - U32 f1c; - U32 f1d; - U32 f7b; - U32 f7c; -} ZSTD_cpuid_t; - -MEM_STATIC ZSTD_cpuid_t ZSTD_cpuid(void) { - U32 f1c = 0; - U32 f1d = 0; - U32 f7b = 0; - U32 f7c = 0; -#if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_IX86)) - int reg[4]; - __cpuid((int*)reg, 0); - { - int const n = reg[0]; - if (n >= 1) { - __cpuid((int*)reg, 1); - f1c = (U32)reg[2]; - f1d = (U32)reg[3]; - } - if (n >= 7) { - __cpuidex((int*)reg, 7, 0); - f7b = (U32)reg[1]; - f7c = (U32)reg[2]; - } - } -#elif defined(__i386__) && defined(__PIC__) && !defined(__clang__) && defined(__GNUC__) - /* The following block like the normal cpuid branch below, but gcc - * reserves ebx for use of its pic register so we must specially - * handle the save and restore to avoid clobbering the register - */ - U32 n; - __asm__( - "pushl %%ebx\n\t" - "cpuid\n\t" - "popl %%ebx\n\t" - : "=a"(n) - : "a"(0) - : "ecx", "edx"); - if (n >= 1) { - U32 f1a; - __asm__( - "pushl %%ebx\n\t" - "cpuid\n\t" - "popl %%ebx\n\t" - : "=a"(f1a), "=c"(f1c), "=d"(f1d) - : "a"(1)); - } - if (n >= 7) { - __asm__( - "pushl %%ebx\n\t" - "cpuid\n\t" - "movl %%ebx, %%eax\n\t" - "popl %%ebx" - : "=a"(f7b), "=c"(f7c) - : "a"(7), "c"(0) - : "edx"); - } -#elif defined(__x86_64__) || defined(_M_X64) || defined(__i386__) - U32 n; - __asm__("cpuid" : "=a"(n) : "a"(0) : "ebx", "ecx", "edx"); - if (n >= 1) { - U32 f1a; - __asm__("cpuid" : "=a"(f1a), "=c"(f1c), "=d"(f1d) : "a"(1) : "ebx"); - } - if (n >= 7) { - U32 f7a; - __asm__("cpuid" - : "=a"(f7a), "=b"(f7b), "=c"(f7c) - : "a"(7), "c"(0) - : "edx"); - } -#endif - { - ZSTD_cpuid_t cpuid; - cpuid.f1c = f1c; - cpuid.f1d = f1d; - cpuid.f7b = f7b; - cpuid.f7c = f7c; - return cpuid; - } -} - -#define X(name, r, bit) \ - MEM_STATIC int ZSTD_cpuid_##name(ZSTD_cpuid_t const cpuid) { \ - return ((cpuid.r) & (1U << bit)) != 0; \ - } - -/* cpuid(1): Processor Info and Feature Bits. */ -#define C(name, bit) X(name, f1c, bit) - C(sse3, 0) - C(pclmuldq, 1) - C(dtes64, 2) - C(monitor, 3) - C(dscpl, 4) - C(vmx, 5) - C(smx, 6) - C(eist, 7) - C(tm2, 8) - C(ssse3, 9) - C(cnxtid, 10) - C(fma, 12) - C(cx16, 13) - C(xtpr, 14) - C(pdcm, 15) - C(pcid, 17) - C(dca, 18) - C(sse41, 19) - C(sse42, 20) - C(x2apic, 21) - C(movbe, 22) - C(popcnt, 23) - C(tscdeadline, 24) - C(aes, 25) - C(xsave, 26) - C(osxsave, 27) - C(avx, 28) - C(f16c, 29) - C(rdrand, 30) -#undef C -#define D(name, bit) X(name, f1d, bit) - D(fpu, 0) - D(vme, 1) - D(de, 2) - D(pse, 3) - D(tsc, 4) - D(msr, 5) - D(pae, 6) - D(mce, 7) - D(cx8, 8) - D(apic, 9) - D(sep, 11) - D(mtrr, 12) - D(pge, 13) - D(mca, 14) - D(cmov, 15) - D(pat, 16) - D(pse36, 17) - D(psn, 18) - D(clfsh, 19) - D(ds, 21) - D(acpi, 22) - D(mmx, 23) - D(fxsr, 24) - D(sse, 25) - D(sse2, 26) - D(ss, 27) - D(htt, 28) - D(tm, 29) - D(pbe, 31) -#undef D - -/* cpuid(7): Extended Features. */ -#define B(name, bit) X(name, f7b, bit) - B(bmi1, 3) - B(hle, 4) - B(avx2, 5) - B(smep, 7) - B(bmi2, 8) - B(erms, 9) - B(invpcid, 10) - B(rtm, 11) - B(mpx, 14) - B(avx512f, 16) - B(avx512dq, 17) - B(rdseed, 18) - B(adx, 19) - B(smap, 20) - B(avx512ifma, 21) - B(pcommit, 22) - B(clflushopt, 23) - B(clwb, 24) - B(avx512pf, 26) - B(avx512er, 27) - B(avx512cd, 28) - B(sha, 29) - B(avx512bw, 30) - B(avx512vl, 31) -#undef B -#define C(name, bit) X(name, f7c, bit) - C(prefetchwt1, 0) - C(avx512vbmi, 1) -#undef C - -#undef X - -#endif /* ZSTD_COMMON_CPU_H */ diff --git a/dep/zstd/lib/common/debug.c b/dep/zstd/lib/common/debug.c deleted file mode 100644 index bb863c9ea..000000000 --- a/dep/zstd/lib/common/debug.c +++ /dev/null @@ -1,24 +0,0 @@ -/* ****************************************************************** - * debug - * Part of FSE library - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -****************************************************************** */ - - -/* - * This module only hosts one global variable - * which can be used to dynamically influence the verbosity of traces, - * such as DEBUGLOG and RAWLOG - */ - -#include "debug.h" - -int g_debuglevel = DEBUGLEVEL; diff --git a/dep/zstd/lib/common/debug.h b/dep/zstd/lib/common/debug.h deleted file mode 100644 index 3b2a320a1..000000000 --- a/dep/zstd/lib/common/debug.h +++ /dev/null @@ -1,107 +0,0 @@ -/* ****************************************************************** - * debug - * Part of FSE library - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -****************************************************************** */ - - -/* - * The purpose of this header is to enable debug functions. - * They regroup assert(), DEBUGLOG() and RAWLOG() for run-time, - * and DEBUG_STATIC_ASSERT() for compile-time. - * - * By default, DEBUGLEVEL==0, which means run-time debug is disabled. - * - * Level 1 enables assert() only. - * Starting level 2, traces can be generated and pushed to stderr. - * The higher the level, the more verbose the traces. - * - * It's possible to dynamically adjust level using variable g_debug_level, - * which is only declared if DEBUGLEVEL>=2, - * and is a global variable, not multi-thread protected (use with care) - */ - -#ifndef DEBUG_H_12987983217 -#define DEBUG_H_12987983217 - -#if defined (__cplusplus) -extern "C" { -#endif - - -/* static assert is triggered at compile time, leaving no runtime artefact. - * static assert only works with compile-time constants. - * Also, this variant can only be used inside a function. */ -#define DEBUG_STATIC_ASSERT(c) (void)sizeof(char[(c) ? 1 : -1]) - - -/* DEBUGLEVEL is expected to be defined externally, - * typically through compiler command line. - * Value must be a number. */ -#ifndef DEBUGLEVEL -# define DEBUGLEVEL 0 -#endif - - -/* recommended values for DEBUGLEVEL : - * 0 : release mode, no debug, all run-time checks disabled - * 1 : enables assert() only, no display - * 2 : reserved, for currently active debug path - * 3 : events once per object lifetime (CCtx, CDict, etc.) - * 4 : events once per frame - * 5 : events once per block - * 6 : events once per sequence (verbose) - * 7+: events at every position (*very* verbose) - * - * It's generally inconvenient to output traces > 5. - * In which case, it's possible to selectively trigger high verbosity levels - * by modifying g_debug_level. - */ - -#if (DEBUGLEVEL>=1) -# define ZSTD_DEPS_NEED_ASSERT -# include "zstd_deps.h" -#else -# ifndef assert /* assert may be already defined, due to prior #include */ -# define assert(condition) ((void)0) /* disable assert (default) */ -# endif -#endif - -#if (DEBUGLEVEL>=2) -# define ZSTD_DEPS_NEED_IO -# include "zstd_deps.h" -extern int g_debuglevel; /* the variable is only declared, - it actually lives in debug.c, - and is shared by the whole process. - It's not thread-safe. - It's useful when enabling very verbose levels - on selective conditions (such as position in src) */ - -# define RAWLOG(l, ...) { \ - if (l<=g_debuglevel) { \ - ZSTD_DEBUG_PRINT(__VA_ARGS__); \ - } } -# define DEBUGLOG(l, ...) { \ - if (l<=g_debuglevel) { \ - ZSTD_DEBUG_PRINT(__FILE__ ": " __VA_ARGS__); \ - ZSTD_DEBUG_PRINT(" \n"); \ - } } -#else -# define RAWLOG(l, ...) {} /* disabled */ -# define DEBUGLOG(l, ...) {} /* disabled */ -#endif - - -#if defined (__cplusplus) -} -#endif - -#endif /* DEBUG_H_12987983217 */ diff --git a/dep/zstd/lib/common/entropy_common.c b/dep/zstd/lib/common/entropy_common.c deleted file mode 100644 index 4229b40c5..000000000 --- a/dep/zstd/lib/common/entropy_common.c +++ /dev/null @@ -1,368 +0,0 @@ -/* ****************************************************************** - * Common functions of New Generation Entropy library - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy - * - Public forum : https://groups.google.com/forum/#!forum/lz4c - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -****************************************************************** */ - -/* ************************************* -* Dependencies -***************************************/ -#include "mem.h" -#include "error_private.h" /* ERR_*, ERROR */ -#define FSE_STATIC_LINKING_ONLY /* FSE_MIN_TABLELOG */ -#include "fse.h" -#define HUF_STATIC_LINKING_ONLY /* HUF_TABLELOG_ABSOLUTEMAX */ -#include "huf.h" - - -/*=== Version ===*/ -unsigned FSE_versionNumber(void) { return FSE_VERSION_NUMBER; } - - -/*=== Error Management ===*/ -unsigned FSE_isError(size_t code) { return ERR_isError(code); } -const char* FSE_getErrorName(size_t code) { return ERR_getErrorName(code); } - -unsigned HUF_isError(size_t code) { return ERR_isError(code); } -const char* HUF_getErrorName(size_t code) { return ERR_getErrorName(code); } - - -/*-************************************************************** -* FSE NCount encoding-decoding -****************************************************************/ -static U32 FSE_ctz(U32 val) -{ - assert(val != 0); - { -# if defined(_MSC_VER) /* Visual */ - if (val != 0) { - unsigned long r; - _BitScanForward(&r, val); - return (unsigned)r; - } else { - /* Should not reach this code path */ - __assume(0); - } -# elif defined(__GNUC__) && (__GNUC__ >= 3) /* GCC Intrinsic */ - return __builtin_ctz(val); -# elif defined(__ICCARM__) /* IAR Intrinsic */ - return __CTZ(val); -# else /* Software version */ - U32 count = 0; - while ((val & 1) == 0) { - val >>= 1; - ++count; - } - return count; -# endif - } -} - -FORCE_INLINE_TEMPLATE -size_t FSE_readNCount_body(short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, - const void* headerBuffer, size_t hbSize) -{ - const BYTE* const istart = (const BYTE*) headerBuffer; - const BYTE* const iend = istart + hbSize; - const BYTE* ip = istart; - int nbBits; - int remaining; - int threshold; - U32 bitStream; - int bitCount; - unsigned charnum = 0; - unsigned const maxSV1 = *maxSVPtr + 1; - int previous0 = 0; - - if (hbSize < 8) { - /* This function only works when hbSize >= 8 */ - char buffer[8] = {0}; - ZSTD_memcpy(buffer, headerBuffer, hbSize); - { size_t const countSize = FSE_readNCount(normalizedCounter, maxSVPtr, tableLogPtr, - buffer, sizeof(buffer)); - if (FSE_isError(countSize)) return countSize; - if (countSize > hbSize) return ERROR(corruption_detected); - return countSize; - } } - assert(hbSize >= 8); - - /* init */ - ZSTD_memset(normalizedCounter, 0, (*maxSVPtr+1) * sizeof(normalizedCounter[0])); /* all symbols not present in NCount have a frequency of 0 */ - bitStream = MEM_readLE32(ip); - nbBits = (bitStream & 0xF) + FSE_MIN_TABLELOG; /* extract tableLog */ - if (nbBits > FSE_TABLELOG_ABSOLUTE_MAX) return ERROR(tableLog_tooLarge); - bitStream >>= 4; - bitCount = 4; - *tableLogPtr = nbBits; - remaining = (1<> 1; - while (repeats >= 12) { - charnum += 3 * 12; - if (LIKELY(ip <= iend-7)) { - ip += 3; - } else { - bitCount -= (int)(8 * (iend - 7 - ip)); - bitCount &= 31; - ip = iend - 4; - } - bitStream = MEM_readLE32(ip) >> bitCount; - repeats = FSE_ctz(~bitStream | 0x80000000) >> 1; - } - charnum += 3 * repeats; - bitStream >>= 2 * repeats; - bitCount += 2 * repeats; - - /* Add the final repeat which isn't 0b11. */ - assert((bitStream & 3) < 3); - charnum += bitStream & 3; - bitCount += 2; - - /* This is an error, but break and return an error - * at the end, because returning out of a loop makes - * it harder for the compiler to optimize. - */ - if (charnum >= maxSV1) break; - - /* We don't need to set the normalized count to 0 - * because we already memset the whole buffer to 0. - */ - - if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) { - assert((bitCount >> 3) <= 3); /* For first condition to work */ - ip += bitCount>>3; - bitCount &= 7; - } else { - bitCount -= (int)(8 * (iend - 4 - ip)); - bitCount &= 31; - ip = iend - 4; - } - bitStream = MEM_readLE32(ip) >> bitCount; - } - { - int const max = (2*threshold-1) - remaining; - int count; - - if ((bitStream & (threshold-1)) < (U32)max) { - count = bitStream & (threshold-1); - bitCount += nbBits-1; - } else { - count = bitStream & (2*threshold-1); - if (count >= threshold) count -= max; - bitCount += nbBits; - } - - count--; /* extra accuracy */ - /* When it matters (small blocks), this is a - * predictable branch, because we don't use -1. - */ - if (count >= 0) { - remaining -= count; - } else { - assert(count == -1); - remaining += count; - } - normalizedCounter[charnum++] = (short)count; - previous0 = !count; - - assert(threshold > 1); - if (remaining < threshold) { - /* This branch can be folded into the - * threshold update condition because we - * know that threshold > 1. - */ - if (remaining <= 1) break; - nbBits = BIT_highbit32(remaining) + 1; - threshold = 1 << (nbBits - 1); - } - if (charnum >= maxSV1) break; - - if (LIKELY(ip <= iend-7) || (ip + (bitCount>>3) <= iend-4)) { - ip += bitCount>>3; - bitCount &= 7; - } else { - bitCount -= (int)(8 * (iend - 4 - ip)); - bitCount &= 31; - ip = iend - 4; - } - bitStream = MEM_readLE32(ip) >> bitCount; - } } - if (remaining != 1) return ERROR(corruption_detected); - /* Only possible when there are too many zeros. */ - if (charnum > maxSV1) return ERROR(maxSymbolValue_tooSmall); - if (bitCount > 32) return ERROR(corruption_detected); - *maxSVPtr = charnum-1; - - ip += (bitCount+7)>>3; - return ip-istart; -} - -/* Avoids the FORCE_INLINE of the _body() function. */ -static size_t FSE_readNCount_body_default( - short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, - const void* headerBuffer, size_t hbSize) -{ - return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize); -} - -#if DYNAMIC_BMI2 -BMI2_TARGET_ATTRIBUTE static size_t FSE_readNCount_body_bmi2( - short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, - const void* headerBuffer, size_t hbSize) -{ - return FSE_readNCount_body(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize); -} -#endif - -size_t FSE_readNCount_bmi2( - short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, - const void* headerBuffer, size_t hbSize, int bmi2) -{ -#if DYNAMIC_BMI2 - if (bmi2) { - return FSE_readNCount_body_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize); - } -#endif - (void)bmi2; - return FSE_readNCount_body_default(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize); -} - -size_t FSE_readNCount( - short* normalizedCounter, unsigned* maxSVPtr, unsigned* tableLogPtr, - const void* headerBuffer, size_t hbSize) -{ - return FSE_readNCount_bmi2(normalizedCounter, maxSVPtr, tableLogPtr, headerBuffer, hbSize, /* bmi2 */ 0); -} - - -/*! HUF_readStats() : - Read compact Huffman tree, saved by HUF_writeCTable(). - `huffWeight` is destination buffer. - `rankStats` is assumed to be a table of at least HUF_TABLELOG_MAX U32. - @return : size read from `src` , or an error Code . - Note : Needed by HUF_readCTable() and HUF_readDTableX?() . -*/ -size_t HUF_readStats(BYTE* huffWeight, size_t hwSize, U32* rankStats, - U32* nbSymbolsPtr, U32* tableLogPtr, - const void* src, size_t srcSize) -{ - U32 wksp[HUF_READ_STATS_WORKSPACE_SIZE_U32]; - return HUF_readStats_wksp(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, wksp, sizeof(wksp), /* bmi2 */ 0); -} - -FORCE_INLINE_TEMPLATE size_t -HUF_readStats_body(BYTE* huffWeight, size_t hwSize, U32* rankStats, - U32* nbSymbolsPtr, U32* tableLogPtr, - const void* src, size_t srcSize, - void* workSpace, size_t wkspSize, - int bmi2) -{ - U32 weightTotal; - const BYTE* ip = (const BYTE*) src; - size_t iSize; - size_t oSize; - - if (!srcSize) return ERROR(srcSize_wrong); - iSize = ip[0]; - /* ZSTD_memset(huffWeight, 0, hwSize); *//* is not necessary, even though some analyzer complain ... */ - - if (iSize >= 128) { /* special header */ - oSize = iSize - 127; - iSize = ((oSize+1)/2); - if (iSize+1 > srcSize) return ERROR(srcSize_wrong); - if (oSize >= hwSize) return ERROR(corruption_detected); - ip += 1; - { U32 n; - for (n=0; n> 4; - huffWeight[n+1] = ip[n/2] & 15; - } } } - else { /* header compressed with FSE (normal case) */ - if (iSize+1 > srcSize) return ERROR(srcSize_wrong); - /* max (hwSize-1) values decoded, as last one is implied */ - oSize = FSE_decompress_wksp_bmi2(huffWeight, hwSize-1, ip+1, iSize, 6, workSpace, wkspSize, bmi2); - if (FSE_isError(oSize)) return oSize; - } - - /* collect weight stats */ - ZSTD_memset(rankStats, 0, (HUF_TABLELOG_MAX + 1) * sizeof(U32)); - weightTotal = 0; - { U32 n; for (n=0; n HUF_TABLELOG_MAX) return ERROR(corruption_detected); - rankStats[huffWeight[n]]++; - weightTotal += (1 << huffWeight[n]) >> 1; - } } - if (weightTotal == 0) return ERROR(corruption_detected); - - /* get last non-null symbol weight (implied, total must be 2^n) */ - { U32 const tableLog = BIT_highbit32(weightTotal) + 1; - if (tableLog > HUF_TABLELOG_MAX) return ERROR(corruption_detected); - *tableLogPtr = tableLog; - /* determine last weight */ - { U32 const total = 1 << tableLog; - U32 const rest = total - weightTotal; - U32 const verif = 1 << BIT_highbit32(rest); - U32 const lastWeight = BIT_highbit32(rest) + 1; - if (verif != rest) return ERROR(corruption_detected); /* last value must be a clean power of 2 */ - huffWeight[oSize] = (BYTE)lastWeight; - rankStats[lastWeight]++; - } } - - /* check tree construction validity */ - if ((rankStats[1] < 2) || (rankStats[1] & 1)) return ERROR(corruption_detected); /* by construction : at least 2 elts of rank 1, must be even */ - - /* results */ - *nbSymbolsPtr = (U32)(oSize+1); - return iSize+1; -} - -/* Avoids the FORCE_INLINE of the _body() function. */ -static size_t HUF_readStats_body_default(BYTE* huffWeight, size_t hwSize, U32* rankStats, - U32* nbSymbolsPtr, U32* tableLogPtr, - const void* src, size_t srcSize, - void* workSpace, size_t wkspSize) -{ - return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 0); -} - -#if DYNAMIC_BMI2 -static BMI2_TARGET_ATTRIBUTE size_t HUF_readStats_body_bmi2(BYTE* huffWeight, size_t hwSize, U32* rankStats, - U32* nbSymbolsPtr, U32* tableLogPtr, - const void* src, size_t srcSize, - void* workSpace, size_t wkspSize) -{ - return HUF_readStats_body(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize, 1); -} -#endif - -size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize, U32* rankStats, - U32* nbSymbolsPtr, U32* tableLogPtr, - const void* src, size_t srcSize, - void* workSpace, size_t wkspSize, - int bmi2) -{ -#if DYNAMIC_BMI2 - if (bmi2) { - return HUF_readStats_body_bmi2(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize); - } -#endif - (void)bmi2; - return HUF_readStats_body_default(huffWeight, hwSize, rankStats, nbSymbolsPtr, tableLogPtr, src, srcSize, workSpace, wkspSize); -} diff --git a/dep/zstd/lib/common/error_private.c b/dep/zstd/lib/common/error_private.c deleted file mode 100644 index 6d1135f8c..000000000 --- a/dep/zstd/lib/common/error_private.c +++ /dev/null @@ -1,56 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -/* The purpose of this file is to have a single list of error strings embedded in binary */ - -#include "error_private.h" - -const char* ERR_getErrorString(ERR_enum code) -{ -#ifdef ZSTD_STRIP_ERROR_STRINGS - (void)code; - return "Error strings stripped"; -#else - static const char* const notErrorCode = "Unspecified error code"; - switch( code ) - { - case PREFIX(no_error): return "No error detected"; - case PREFIX(GENERIC): return "Error (generic)"; - case PREFIX(prefix_unknown): return "Unknown frame descriptor"; - case PREFIX(version_unsupported): return "Version not supported"; - case PREFIX(frameParameter_unsupported): return "Unsupported frame parameter"; - case PREFIX(frameParameter_windowTooLarge): return "Frame requires too much memory for decoding"; - case PREFIX(corruption_detected): return "Corrupted block detected"; - case PREFIX(checksum_wrong): return "Restored data doesn't match checksum"; - case PREFIX(parameter_unsupported): return "Unsupported parameter"; - case PREFIX(parameter_outOfBound): return "Parameter is out of bound"; - case PREFIX(init_missing): return "Context should be init first"; - case PREFIX(memory_allocation): return "Allocation error : not enough memory"; - case PREFIX(workSpace_tooSmall): return "workSpace buffer is not large enough"; - case PREFIX(stage_wrong): return "Operation not authorized at current processing stage"; - case PREFIX(tableLog_tooLarge): return "tableLog requires too much memory : unsupported"; - case PREFIX(maxSymbolValue_tooLarge): return "Unsupported max Symbol Value : too large"; - case PREFIX(maxSymbolValue_tooSmall): return "Specified maxSymbolValue is too small"; - case PREFIX(dictionary_corrupted): return "Dictionary is corrupted"; - case PREFIX(dictionary_wrong): return "Dictionary mismatch"; - case PREFIX(dictionaryCreation_failed): return "Cannot create Dictionary from provided samples"; - case PREFIX(dstSize_tooSmall): return "Destination buffer is too small"; - case PREFIX(srcSize_wrong): return "Src size is incorrect"; - case PREFIX(dstBuffer_null): return "Operation on NULL destination buffer"; - /* following error codes are not stable and may be removed or changed in a future version */ - case PREFIX(frameIndex_tooLarge): return "Frame index is too large"; - case PREFIX(seekableIO): return "An I/O error occurred when reading/seeking"; - case PREFIX(dstBuffer_wrong): return "Destination buffer is wrong"; - case PREFIX(srcBuffer_wrong): return "Source buffer is wrong"; - case PREFIX(maxCode): - default: return notErrorCode; - } -#endif -} diff --git a/dep/zstd/lib/common/error_private.h b/dep/zstd/lib/common/error_private.h deleted file mode 100644 index 007d81066..000000000 --- a/dep/zstd/lib/common/error_private.h +++ /dev/null @@ -1,159 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -/* Note : this module is expected to remain private, do not expose it */ - -#ifndef ERROR_H_MODULE -#define ERROR_H_MODULE - -#if defined (__cplusplus) -extern "C" { -#endif - - -/* **************************************** -* Dependencies -******************************************/ -#include "../zstd_errors.h" /* enum list */ -#include "compiler.h" -#include "debug.h" -#include "zstd_deps.h" /* size_t */ - - -/* **************************************** -* Compiler-specific -******************************************/ -#if defined(__GNUC__) -# define ERR_STATIC static __attribute__((unused)) -#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) -# define ERR_STATIC static inline -#elif defined(_MSC_VER) -# define ERR_STATIC static __inline -#else -# define ERR_STATIC static /* this version may generate warnings for unused static functions; disable the relevant warning */ -#endif - - -/*-**************************************** -* Customization (error_public.h) -******************************************/ -typedef ZSTD_ErrorCode ERR_enum; -#define PREFIX(name) ZSTD_error_##name - - -/*-**************************************** -* Error codes handling -******************************************/ -#undef ERROR /* already defined on Visual Studio */ -#define ERROR(name) ZSTD_ERROR(name) -#define ZSTD_ERROR(name) ((size_t)-PREFIX(name)) - -ERR_STATIC unsigned ERR_isError(size_t code) { return (code > ERROR(maxCode)); } - -ERR_STATIC ERR_enum ERR_getErrorCode(size_t code) { if (!ERR_isError(code)) return (ERR_enum)0; return (ERR_enum) (0-code); } - -/* check and forward error code */ -#define CHECK_V_F(e, f) size_t const e = f; if (ERR_isError(e)) return e -#define CHECK_F(f) { CHECK_V_F(_var_err__, f); } - - -/*-**************************************** -* Error Strings -******************************************/ - -const char* ERR_getErrorString(ERR_enum code); /* error_private.c */ - -ERR_STATIC const char* ERR_getErrorName(size_t code) -{ - return ERR_getErrorString(ERR_getErrorCode(code)); -} - -/** - * Ignore: this is an internal helper. - * - * This is a helper function to help force C99-correctness during compilation. - * Under strict compilation modes, variadic macro arguments can't be empty. - * However, variadic function arguments can be. Using a function therefore lets - * us statically check that at least one (string) argument was passed, - * independent of the compilation flags. - */ -static INLINE_KEYWORD UNUSED_ATTR -void _force_has_format_string(const char *format, ...) { - (void)format; -} - -/** - * Ignore: this is an internal helper. - * - * We want to force this function invocation to be syntactically correct, but - * we don't want to force runtime evaluation of its arguments. - */ -#define _FORCE_HAS_FORMAT_STRING(...) \ - if (0) { \ - _force_has_format_string(__VA_ARGS__); \ - } - -#define ERR_QUOTE(str) #str - -/** - * Return the specified error if the condition evaluates to true. - * - * In debug modes, prints additional information. - * In order to do that (particularly, printing the conditional that failed), - * this can't just wrap RETURN_ERROR(). - */ -#define RETURN_ERROR_IF(cond, err, ...) \ - if (cond) { \ - RAWLOG(3, "%s:%d: ERROR!: check %s failed, returning %s", \ - __FILE__, __LINE__, ERR_QUOTE(cond), ERR_QUOTE(ERROR(err))); \ - _FORCE_HAS_FORMAT_STRING(__VA_ARGS__); \ - RAWLOG(3, ": " __VA_ARGS__); \ - RAWLOG(3, "\n"); \ - return ERROR(err); \ - } - -/** - * Unconditionally return the specified error. - * - * In debug modes, prints additional information. - */ -#define RETURN_ERROR(err, ...) \ - do { \ - RAWLOG(3, "%s:%d: ERROR!: unconditional check failed, returning %s", \ - __FILE__, __LINE__, ERR_QUOTE(ERROR(err))); \ - _FORCE_HAS_FORMAT_STRING(__VA_ARGS__); \ - RAWLOG(3, ": " __VA_ARGS__); \ - RAWLOG(3, "\n"); \ - return ERROR(err); \ - } while(0); - -/** - * If the provided expression evaluates to an error code, returns that error code. - * - * In debug modes, prints additional information. - */ -#define FORWARD_IF_ERROR(err, ...) \ - do { \ - size_t const err_code = (err); \ - if (ERR_isError(err_code)) { \ - RAWLOG(3, "%s:%d: ERROR!: forwarding error in %s: %s", \ - __FILE__, __LINE__, ERR_QUOTE(err), ERR_getErrorName(err_code)); \ - _FORCE_HAS_FORMAT_STRING(__VA_ARGS__); \ - RAWLOG(3, ": " __VA_ARGS__); \ - RAWLOG(3, "\n"); \ - return err_code; \ - } \ - } while(0); - -#if defined (__cplusplus) -} -#endif - -#endif /* ERROR_H_MODULE */ diff --git a/dep/zstd/lib/common/fse.h b/dep/zstd/lib/common/fse.h deleted file mode 100644 index 714bfd3e7..000000000 --- a/dep/zstd/lib/common/fse.h +++ /dev/null @@ -1,717 +0,0 @@ -/* ****************************************************************** - * FSE : Finite State Entropy codec - * Public Prototypes declaration - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -****************************************************************** */ - -#if defined (__cplusplus) -extern "C" { -#endif - -#ifndef FSE_H -#define FSE_H - - -/*-***************************************** -* Dependencies -******************************************/ -#include "zstd_deps.h" /* size_t, ptrdiff_t */ - - -/*-***************************************** -* FSE_PUBLIC_API : control library symbols visibility -******************************************/ -#if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4) -# define FSE_PUBLIC_API __attribute__ ((visibility ("default"))) -#elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) /* Visual expected */ -# define FSE_PUBLIC_API __declspec(dllexport) -#elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1) -# define FSE_PUBLIC_API __declspec(dllimport) /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/ -#else -# define FSE_PUBLIC_API -#endif - -/*------ Version ------*/ -#define FSE_VERSION_MAJOR 0 -#define FSE_VERSION_MINOR 9 -#define FSE_VERSION_RELEASE 0 - -#define FSE_LIB_VERSION FSE_VERSION_MAJOR.FSE_VERSION_MINOR.FSE_VERSION_RELEASE -#define FSE_QUOTE(str) #str -#define FSE_EXPAND_AND_QUOTE(str) FSE_QUOTE(str) -#define FSE_VERSION_STRING FSE_EXPAND_AND_QUOTE(FSE_LIB_VERSION) - -#define FSE_VERSION_NUMBER (FSE_VERSION_MAJOR *100*100 + FSE_VERSION_MINOR *100 + FSE_VERSION_RELEASE) -FSE_PUBLIC_API unsigned FSE_versionNumber(void); /**< library version number; to be used when checking dll version */ - - -/*-**************************************** -* FSE simple functions -******************************************/ -/*! FSE_compress() : - Compress content of buffer 'src', of size 'srcSize', into destination buffer 'dst'. - 'dst' buffer must be already allocated. Compression runs faster is dstCapacity >= FSE_compressBound(srcSize). - @return : size of compressed data (<= dstCapacity). - Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!! - if return == 1, srcData is a single byte symbol * srcSize times. Use RLE compression instead. - if FSE_isError(return), compression failed (more details using FSE_getErrorName()) -*/ -FSE_PUBLIC_API size_t FSE_compress(void* dst, size_t dstCapacity, - const void* src, size_t srcSize); - -/*! FSE_decompress(): - Decompress FSE data from buffer 'cSrc', of size 'cSrcSize', - into already allocated destination buffer 'dst', of size 'dstCapacity'. - @return : size of regenerated data (<= maxDstSize), - or an error code, which can be tested using FSE_isError() . - - ** Important ** : FSE_decompress() does not decompress non-compressible nor RLE data !!! - Why ? : making this distinction requires a header. - Header management is intentionally delegated to the user layer, which can better manage special cases. -*/ -FSE_PUBLIC_API size_t FSE_decompress(void* dst, size_t dstCapacity, - const void* cSrc, size_t cSrcSize); - - -/*-***************************************** -* Tool functions -******************************************/ -FSE_PUBLIC_API size_t FSE_compressBound(size_t size); /* maximum compressed size */ - -/* Error Management */ -FSE_PUBLIC_API unsigned FSE_isError(size_t code); /* tells if a return value is an error code */ -FSE_PUBLIC_API const char* FSE_getErrorName(size_t code); /* provides error code string (useful for debugging) */ - - -/*-***************************************** -* FSE advanced functions -******************************************/ -/*! FSE_compress2() : - Same as FSE_compress(), but allows the selection of 'maxSymbolValue' and 'tableLog' - Both parameters can be defined as '0' to mean : use default value - @return : size of compressed data - Special values : if return == 0, srcData is not compressible => Nothing is stored within cSrc !!! - if return == 1, srcData is a single byte symbol * srcSize times. Use RLE compression. - if FSE_isError(return), it's an error code. -*/ -FSE_PUBLIC_API size_t FSE_compress2 (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog); - - -/*-***************************************** -* FSE detailed API -******************************************/ -/*! -FSE_compress() does the following: -1. count symbol occurrence from source[] into table count[] (see hist.h) -2. normalize counters so that sum(count[]) == Power_of_2 (2^tableLog) -3. save normalized counters to memory buffer using writeNCount() -4. build encoding table 'CTable' from normalized counters -5. encode the data stream using encoding table 'CTable' - -FSE_decompress() does the following: -1. read normalized counters with readNCount() -2. build decoding table 'DTable' from normalized counters -3. decode the data stream using decoding table 'DTable' - -The following API allows targeting specific sub-functions for advanced tasks. -For example, it's possible to compress several blocks using the same 'CTable', -or to save and provide normalized distribution using external method. -*/ - -/* *** COMPRESSION *** */ - -/*! FSE_optimalTableLog(): - dynamically downsize 'tableLog' when conditions are met. - It saves CPU time, by using smaller tables, while preserving or even improving compression ratio. - @return : recommended tableLog (necessarily <= 'maxTableLog') */ -FSE_PUBLIC_API unsigned FSE_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue); - -/*! FSE_normalizeCount(): - normalize counts so that sum(count[]) == Power_of_2 (2^tableLog) - 'normalizedCounter' is a table of short, of minimum size (maxSymbolValue+1). - useLowProbCount is a boolean parameter which trades off compressed size for - faster header decoding. When it is set to 1, the compressed data will be slightly - smaller. And when it is set to 0, FSE_readNCount() and FSE_buildDTable() will be - faster. If you are compressing a small amount of data (< 2 KB) then useLowProbCount=0 - is a good default, since header deserialization makes a big speed difference. - Otherwise, useLowProbCount=1 is a good default, since the speed difference is small. - @return : tableLog, - or an errorCode, which can be tested using FSE_isError() */ -FSE_PUBLIC_API size_t FSE_normalizeCount(short* normalizedCounter, unsigned tableLog, - const unsigned* count, size_t srcSize, unsigned maxSymbolValue, unsigned useLowProbCount); - -/*! FSE_NCountWriteBound(): - Provides the maximum possible size of an FSE normalized table, given 'maxSymbolValue' and 'tableLog'. - Typically useful for allocation purpose. */ -FSE_PUBLIC_API size_t FSE_NCountWriteBound(unsigned maxSymbolValue, unsigned tableLog); - -/*! FSE_writeNCount(): - Compactly save 'normalizedCounter' into 'buffer'. - @return : size of the compressed table, - or an errorCode, which can be tested using FSE_isError(). */ -FSE_PUBLIC_API size_t FSE_writeNCount (void* buffer, size_t bufferSize, - const short* normalizedCounter, - unsigned maxSymbolValue, unsigned tableLog); - -/*! Constructor and Destructor of FSE_CTable. - Note that FSE_CTable size depends on 'tableLog' and 'maxSymbolValue' */ -typedef unsigned FSE_CTable; /* don't allocate that. It's only meant to be more restrictive than void* */ -FSE_PUBLIC_API FSE_CTable* FSE_createCTable (unsigned maxSymbolValue, unsigned tableLog); -FSE_PUBLIC_API void FSE_freeCTable (FSE_CTable* ct); - -/*! FSE_buildCTable(): - Builds `ct`, which must be already allocated, using FSE_createCTable(). - @return : 0, or an errorCode, which can be tested using FSE_isError() */ -FSE_PUBLIC_API size_t FSE_buildCTable(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog); - -/*! FSE_compress_usingCTable(): - Compress `src` using `ct` into `dst` which must be already allocated. - @return : size of compressed data (<= `dstCapacity`), - or 0 if compressed data could not fit into `dst`, - or an errorCode, which can be tested using FSE_isError() */ -FSE_PUBLIC_API size_t FSE_compress_usingCTable (void* dst, size_t dstCapacity, const void* src, size_t srcSize, const FSE_CTable* ct); - -/*! -Tutorial : ----------- -The first step is to count all symbols. FSE_count() does this job very fast. -Result will be saved into 'count', a table of unsigned int, which must be already allocated, and have 'maxSymbolValuePtr[0]+1' cells. -'src' is a table of bytes of size 'srcSize'. All values within 'src' MUST be <= maxSymbolValuePtr[0] -maxSymbolValuePtr[0] will be updated, with its real value (necessarily <= original value) -FSE_count() will return the number of occurrence of the most frequent symbol. -This can be used to know if there is a single symbol within 'src', and to quickly evaluate its compressibility. -If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError()). - -The next step is to normalize the frequencies. -FSE_normalizeCount() will ensure that sum of frequencies is == 2 ^'tableLog'. -It also guarantees a minimum of 1 to any Symbol with frequency >= 1. -You can use 'tableLog'==0 to mean "use default tableLog value". -If you are unsure of which tableLog value to use, you can ask FSE_optimalTableLog(), -which will provide the optimal valid tableLog given sourceSize, maxSymbolValue, and a user-defined maximum (0 means "default"). - -The result of FSE_normalizeCount() will be saved into a table, -called 'normalizedCounter', which is a table of signed short. -'normalizedCounter' must be already allocated, and have at least 'maxSymbolValue+1' cells. -The return value is tableLog if everything proceeded as expected. -It is 0 if there is a single symbol within distribution. -If there is an error (ex: invalid tableLog value), the function will return an ErrorCode (which can be tested using FSE_isError()). - -'normalizedCounter' can be saved in a compact manner to a memory area using FSE_writeNCount(). -'buffer' must be already allocated. -For guaranteed success, buffer size must be at least FSE_headerBound(). -The result of the function is the number of bytes written into 'buffer'. -If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError(); ex : buffer size too small). - -'normalizedCounter' can then be used to create the compression table 'CTable'. -The space required by 'CTable' must be already allocated, using FSE_createCTable(). -You can then use FSE_buildCTable() to fill 'CTable'. -If there is an error, both functions will return an ErrorCode (which can be tested using FSE_isError()). - -'CTable' can then be used to compress 'src', with FSE_compress_usingCTable(). -Similar to FSE_count(), the convention is that 'src' is assumed to be a table of char of size 'srcSize' -The function returns the size of compressed data (without header), necessarily <= `dstCapacity`. -If it returns '0', compressed data could not fit into 'dst'. -If there is an error, the function will return an ErrorCode (which can be tested using FSE_isError()). -*/ - - -/* *** DECOMPRESSION *** */ - -/*! FSE_readNCount(): - Read compactly saved 'normalizedCounter' from 'rBuffer'. - @return : size read from 'rBuffer', - or an errorCode, which can be tested using FSE_isError(). - maxSymbolValuePtr[0] and tableLogPtr[0] will also be updated with their respective values */ -FSE_PUBLIC_API size_t FSE_readNCount (short* normalizedCounter, - unsigned* maxSymbolValuePtr, unsigned* tableLogPtr, - const void* rBuffer, size_t rBuffSize); - -/*! FSE_readNCount_bmi2(): - * Same as FSE_readNCount() but pass bmi2=1 when your CPU supports BMI2 and 0 otherwise. - */ -FSE_PUBLIC_API size_t FSE_readNCount_bmi2(short* normalizedCounter, - unsigned* maxSymbolValuePtr, unsigned* tableLogPtr, - const void* rBuffer, size_t rBuffSize, int bmi2); - -/*! Constructor and Destructor of FSE_DTable. - Note that its size depends on 'tableLog' */ -typedef unsigned FSE_DTable; /* don't allocate that. It's just a way to be more restrictive than void* */ -FSE_PUBLIC_API FSE_DTable* FSE_createDTable(unsigned tableLog); -FSE_PUBLIC_API void FSE_freeDTable(FSE_DTable* dt); - -/*! FSE_buildDTable(): - Builds 'dt', which must be already allocated, using FSE_createDTable(). - return : 0, or an errorCode, which can be tested using FSE_isError() */ -FSE_PUBLIC_API size_t FSE_buildDTable (FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog); - -/*! FSE_decompress_usingDTable(): - Decompress compressed source `cSrc` of size `cSrcSize` using `dt` - into `dst` which must be already allocated. - @return : size of regenerated data (necessarily <= `dstCapacity`), - or an errorCode, which can be tested using FSE_isError() */ -FSE_PUBLIC_API size_t FSE_decompress_usingDTable(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, const FSE_DTable* dt); - -/*! -Tutorial : ----------- -(Note : these functions only decompress FSE-compressed blocks. - If block is uncompressed, use memcpy() instead - If block is a single repeated byte, use memset() instead ) - -The first step is to obtain the normalized frequencies of symbols. -This can be performed by FSE_readNCount() if it was saved using FSE_writeNCount(). -'normalizedCounter' must be already allocated, and have at least 'maxSymbolValuePtr[0]+1' cells of signed short. -In practice, that means it's necessary to know 'maxSymbolValue' beforehand, -or size the table to handle worst case situations (typically 256). -FSE_readNCount() will provide 'tableLog' and 'maxSymbolValue'. -The result of FSE_readNCount() is the number of bytes read from 'rBuffer'. -Note that 'rBufferSize' must be at least 4 bytes, even if useful information is less than that. -If there is an error, the function will return an error code, which can be tested using FSE_isError(). - -The next step is to build the decompression tables 'FSE_DTable' from 'normalizedCounter'. -This is performed by the function FSE_buildDTable(). -The space required by 'FSE_DTable' must be already allocated using FSE_createDTable(). -If there is an error, the function will return an error code, which can be tested using FSE_isError(). - -`FSE_DTable` can then be used to decompress `cSrc`, with FSE_decompress_usingDTable(). -`cSrcSize` must be strictly correct, otherwise decompression will fail. -FSE_decompress_usingDTable() result will tell how many bytes were regenerated (<=`dstCapacity`). -If there is an error, the function will return an error code, which can be tested using FSE_isError(). (ex: dst buffer too small) -*/ - -#endif /* FSE_H */ - -#if defined(FSE_STATIC_LINKING_ONLY) && !defined(FSE_H_FSE_STATIC_LINKING_ONLY) -#define FSE_H_FSE_STATIC_LINKING_ONLY - -/* *** Dependency *** */ -#include "bitstream.h" - - -/* ***************************************** -* Static allocation -*******************************************/ -/* FSE buffer bounds */ -#define FSE_NCOUNTBOUND 512 -#define FSE_BLOCKBOUND(size) ((size) + ((size)>>7) + 4 /* fse states */ + sizeof(size_t) /* bitContainer */) -#define FSE_COMPRESSBOUND(size) (FSE_NCOUNTBOUND + FSE_BLOCKBOUND(size)) /* Macro version, useful for static allocation */ - -/* It is possible to statically allocate FSE CTable/DTable as a table of FSE_CTable/FSE_DTable using below macros */ -#define FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) (1 + (1<<((maxTableLog)-1)) + (((maxSymbolValue)+1)*2)) -#define FSE_DTABLE_SIZE_U32(maxTableLog) (1 + (1<<(maxTableLog))) - -/* or use the size to malloc() space directly. Pay attention to alignment restrictions though */ -#define FSE_CTABLE_SIZE(maxTableLog, maxSymbolValue) (FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(FSE_CTable)) -#define FSE_DTABLE_SIZE(maxTableLog) (FSE_DTABLE_SIZE_U32(maxTableLog) * sizeof(FSE_DTable)) - - -/* ***************************************** - * FSE advanced API - ***************************************** */ - -unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, unsigned minus); -/**< same as FSE_optimalTableLog(), which used `minus==2` */ - -/* FSE_compress_wksp() : - * Same as FSE_compress2(), but using an externally allocated scratch buffer (`workSpace`). - * FSE_COMPRESS_WKSP_SIZE_U32() provides the minimum size required for `workSpace` as a table of FSE_CTable. - */ -#define FSE_COMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) ( FSE_CTABLE_SIZE_U32(maxTableLog, maxSymbolValue) + ((maxTableLog > 12) ? (1 << (maxTableLog - 2)) : 1024) ) -size_t FSE_compress_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); - -size_t FSE_buildCTable_raw (FSE_CTable* ct, unsigned nbBits); -/**< build a fake FSE_CTable, designed for a flat distribution, where each symbol uses nbBits */ - -size_t FSE_buildCTable_rle (FSE_CTable* ct, unsigned char symbolValue); -/**< build a fake FSE_CTable, designed to compress always the same symbolValue */ - -/* FSE_buildCTable_wksp() : - * Same as FSE_buildCTable(), but using an externally allocated scratch buffer (`workSpace`). - * `wkspSize` must be >= `FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog)` of `unsigned`. - * See FSE_buildCTable_wksp() for breakdown of workspace usage. - */ -#define FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog) (((maxSymbolValue + 2) + (1ull << (tableLog)))/2 + sizeof(U64)/sizeof(U32) /* additional 8 bytes for potential table overwrite */) -#define FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) (sizeof(unsigned) * FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(maxSymbolValue, tableLog)) -size_t FSE_buildCTable_wksp(FSE_CTable* ct, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); - -#define FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) (sizeof(short) * (maxSymbolValue + 1) + (1ULL << maxTableLog) + 8) -#define FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) ((FSE_BUILD_DTABLE_WKSP_SIZE(maxTableLog, maxSymbolValue) + sizeof(unsigned) - 1) / sizeof(unsigned)) -FSE_PUBLIC_API size_t FSE_buildDTable_wksp(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); -/**< Same as FSE_buildDTable(), using an externally allocated `workspace` produced with `FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxSymbolValue)` */ - -size_t FSE_buildDTable_raw (FSE_DTable* dt, unsigned nbBits); -/**< build a fake FSE_DTable, designed to read a flat distribution where each symbol uses nbBits */ - -size_t FSE_buildDTable_rle (FSE_DTable* dt, unsigned char symbolValue); -/**< build a fake FSE_DTable, designed to always generate the same symbolValue */ - -#define FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) (FSE_DTABLE_SIZE_U32(maxTableLog) + FSE_BUILD_DTABLE_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) + (FSE_MAX_SYMBOL_VALUE + 1) / 2 + 1) -#define FSE_DECOMPRESS_WKSP_SIZE(maxTableLog, maxSymbolValue) (FSE_DECOMPRESS_WKSP_SIZE_U32(maxTableLog, maxSymbolValue) * sizeof(unsigned)) -size_t FSE_decompress_wksp(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize); -/**< same as FSE_decompress(), using an externally allocated `workSpace` produced with `FSE_DECOMPRESS_WKSP_SIZE_U32(maxLog, maxSymbolValue)` */ - -size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2); -/**< Same as FSE_decompress_wksp() but with dynamic BMI2 support. Pass 1 if your CPU supports BMI2 or 0 if it doesn't. */ - -typedef enum { - FSE_repeat_none, /**< Cannot use the previous table */ - FSE_repeat_check, /**< Can use the previous table but it must be checked */ - FSE_repeat_valid /**< Can use the previous table and it is assumed to be valid */ - } FSE_repeat; - -/* ***************************************** -* FSE symbol compression API -*******************************************/ -/*! - This API consists of small unitary functions, which highly benefit from being inlined. - Hence their body are included in next section. -*/ -typedef struct { - ptrdiff_t value; - const void* stateTable; - const void* symbolTT; - unsigned stateLog; -} FSE_CState_t; - -static void FSE_initCState(FSE_CState_t* CStatePtr, const FSE_CTable* ct); - -static void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* CStatePtr, unsigned symbol); - -static void FSE_flushCState(BIT_CStream_t* bitC, const FSE_CState_t* CStatePtr); - -/**< -These functions are inner components of FSE_compress_usingCTable(). -They allow the creation of custom streams, mixing multiple tables and bit sources. - -A key property to keep in mind is that encoding and decoding are done **in reverse direction**. -So the first symbol you will encode is the last you will decode, like a LIFO stack. - -You will need a few variables to track your CStream. They are : - -FSE_CTable ct; // Provided by FSE_buildCTable() -BIT_CStream_t bitStream; // bitStream tracking structure -FSE_CState_t state; // State tracking structure (can have several) - - -The first thing to do is to init bitStream and state. - size_t errorCode = BIT_initCStream(&bitStream, dstBuffer, maxDstSize); - FSE_initCState(&state, ct); - -Note that BIT_initCStream() can produce an error code, so its result should be tested, using FSE_isError(); -You can then encode your input data, byte after byte. -FSE_encodeSymbol() outputs a maximum of 'tableLog' bits at a time. -Remember decoding will be done in reverse direction. - FSE_encodeByte(&bitStream, &state, symbol); - -At any time, you can also add any bit sequence. -Note : maximum allowed nbBits is 25, for compatibility with 32-bits decoders - BIT_addBits(&bitStream, bitField, nbBits); - -The above methods don't commit data to memory, they just store it into local register, for speed. -Local register size is 64-bits on 64-bits systems, 32-bits on 32-bits systems (size_t). -Writing data to memory is a manual operation, performed by the flushBits function. - BIT_flushBits(&bitStream); - -Your last FSE encoding operation shall be to flush your last state value(s). - FSE_flushState(&bitStream, &state); - -Finally, you must close the bitStream. -The function returns the size of CStream in bytes. -If data couldn't fit into dstBuffer, it will return a 0 ( == not compressible) -If there is an error, it returns an errorCode (which can be tested using FSE_isError()). - size_t size = BIT_closeCStream(&bitStream); -*/ - - -/* ***************************************** -* FSE symbol decompression API -*******************************************/ -typedef struct { - size_t state; - const void* table; /* precise table may vary, depending on U16 */ -} FSE_DState_t; - - -static void FSE_initDState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD, const FSE_DTable* dt); - -static unsigned char FSE_decodeSymbol(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD); - -static unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr); - -/**< -Let's now decompose FSE_decompress_usingDTable() into its unitary components. -You will decode FSE-encoded symbols from the bitStream, -and also any other bitFields you put in, **in reverse order**. - -You will need a few variables to track your bitStream. They are : - -BIT_DStream_t DStream; // Stream context -FSE_DState_t DState; // State context. Multiple ones are possible -FSE_DTable* DTablePtr; // Decoding table, provided by FSE_buildDTable() - -The first thing to do is to init the bitStream. - errorCode = BIT_initDStream(&DStream, srcBuffer, srcSize); - -You should then retrieve your initial state(s) -(in reverse flushing order if you have several ones) : - errorCode = FSE_initDState(&DState, &DStream, DTablePtr); - -You can then decode your data, symbol after symbol. -For information the maximum number of bits read by FSE_decodeSymbol() is 'tableLog'. -Keep in mind that symbols are decoded in reverse order, like a LIFO stack (last in, first out). - unsigned char symbol = FSE_decodeSymbol(&DState, &DStream); - -You can retrieve any bitfield you eventually stored into the bitStream (in reverse order) -Note : maximum allowed nbBits is 25, for 32-bits compatibility - size_t bitField = BIT_readBits(&DStream, nbBits); - -All above operations only read from local register (which size depends on size_t). -Refueling the register from memory is manually performed by the reload method. - endSignal = FSE_reloadDStream(&DStream); - -BIT_reloadDStream() result tells if there is still some more data to read from DStream. -BIT_DStream_unfinished : there is still some data left into the DStream. -BIT_DStream_endOfBuffer : Dstream reached end of buffer. Its container may no longer be completely filled. -BIT_DStream_completed : Dstream reached its exact end, corresponding in general to decompression completed. -BIT_DStream_tooFar : Dstream went too far. Decompression result is corrupted. - -When reaching end of buffer (BIT_DStream_endOfBuffer), progress slowly, notably if you decode multiple symbols per loop, -to properly detect the exact end of stream. -After each decoded symbol, check if DStream is fully consumed using this simple test : - BIT_reloadDStream(&DStream) >= BIT_DStream_completed - -When it's done, verify decompression is fully completed, by checking both DStream and the relevant states. -Checking if DStream has reached its end is performed by : - BIT_endOfDStream(&DStream); -Check also the states. There might be some symbols left there, if some high probability ones (>50%) are possible. - FSE_endOfDState(&DState); -*/ - - -/* ***************************************** -* FSE unsafe API -*******************************************/ -static unsigned char FSE_decodeSymbolFast(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD); -/* faster, but works only if nbBits is always >= 1 (otherwise, result will be corrupted) */ - - -/* ***************************************** -* Implementation of inlined functions -*******************************************/ -typedef struct { - int deltaFindState; - U32 deltaNbBits; -} FSE_symbolCompressionTransform; /* total 8 bytes */ - -MEM_STATIC void FSE_initCState(FSE_CState_t* statePtr, const FSE_CTable* ct) -{ - const void* ptr = ct; - const U16* u16ptr = (const U16*) ptr; - const U32 tableLog = MEM_read16(ptr); - statePtr->value = (ptrdiff_t)1<stateTable = u16ptr+2; - statePtr->symbolTT = ct + 1 + (tableLog ? (1<<(tableLog-1)) : 1); - statePtr->stateLog = tableLog; -} - - -/*! FSE_initCState2() : -* Same as FSE_initCState(), but the first symbol to include (which will be the last to be read) -* uses the smallest state value possible, saving the cost of this symbol */ -MEM_STATIC void FSE_initCState2(FSE_CState_t* statePtr, const FSE_CTable* ct, U32 symbol) -{ - FSE_initCState(statePtr, ct); - { const FSE_symbolCompressionTransform symbolTT = ((const FSE_symbolCompressionTransform*)(statePtr->symbolTT))[symbol]; - const U16* stateTable = (const U16*)(statePtr->stateTable); - U32 nbBitsOut = (U32)((symbolTT.deltaNbBits + (1<<15)) >> 16); - statePtr->value = (nbBitsOut << 16) - symbolTT.deltaNbBits; - statePtr->value = stateTable[(statePtr->value >> nbBitsOut) + symbolTT.deltaFindState]; - } -} - -MEM_STATIC void FSE_encodeSymbol(BIT_CStream_t* bitC, FSE_CState_t* statePtr, unsigned symbol) -{ - FSE_symbolCompressionTransform const symbolTT = ((const FSE_symbolCompressionTransform*)(statePtr->symbolTT))[symbol]; - const U16* const stateTable = (const U16*)(statePtr->stateTable); - U32 const nbBitsOut = (U32)((statePtr->value + symbolTT.deltaNbBits) >> 16); - BIT_addBits(bitC, statePtr->value, nbBitsOut); - statePtr->value = stateTable[ (statePtr->value >> nbBitsOut) + symbolTT.deltaFindState]; -} - -MEM_STATIC void FSE_flushCState(BIT_CStream_t* bitC, const FSE_CState_t* statePtr) -{ - BIT_addBits(bitC, statePtr->value, statePtr->stateLog); - BIT_flushBits(bitC); -} - - -/* FSE_getMaxNbBits() : - * Approximate maximum cost of a symbol, in bits. - * Fractional get rounded up (i.e : a symbol with a normalized frequency of 3 gives the same result as a frequency of 2) - * note 1 : assume symbolValue is valid (<= maxSymbolValue) - * note 2 : if freq[symbolValue]==0, @return a fake cost of tableLog+1 bits */ -MEM_STATIC U32 FSE_getMaxNbBits(const void* symbolTTPtr, U32 symbolValue) -{ - const FSE_symbolCompressionTransform* symbolTT = (const FSE_symbolCompressionTransform*) symbolTTPtr; - return (symbolTT[symbolValue].deltaNbBits + ((1<<16)-1)) >> 16; -} - -/* FSE_bitCost() : - * Approximate symbol cost, as fractional value, using fixed-point format (accuracyLog fractional bits) - * note 1 : assume symbolValue is valid (<= maxSymbolValue) - * note 2 : if freq[symbolValue]==0, @return a fake cost of tableLog+1 bits */ -MEM_STATIC U32 FSE_bitCost(const void* symbolTTPtr, U32 tableLog, U32 symbolValue, U32 accuracyLog) -{ - const FSE_symbolCompressionTransform* symbolTT = (const FSE_symbolCompressionTransform*) symbolTTPtr; - U32 const minNbBits = symbolTT[symbolValue].deltaNbBits >> 16; - U32 const threshold = (minNbBits+1) << 16; - assert(tableLog < 16); - assert(accuracyLog < 31-tableLog); /* ensure enough room for renormalization double shift */ - { U32 const tableSize = 1 << tableLog; - U32 const deltaFromThreshold = threshold - (symbolTT[symbolValue].deltaNbBits + tableSize); - U32 const normalizedDeltaFromThreshold = (deltaFromThreshold << accuracyLog) >> tableLog; /* linear interpolation (very approximate) */ - U32 const bitMultiplier = 1 << accuracyLog; - assert(symbolTT[symbolValue].deltaNbBits + tableSize <= threshold); - assert(normalizedDeltaFromThreshold <= bitMultiplier); - return (minNbBits+1)*bitMultiplier - normalizedDeltaFromThreshold; - } -} - - -/* ====== Decompression ====== */ - -typedef struct { - U16 tableLog; - U16 fastMode; -} FSE_DTableHeader; /* sizeof U32 */ - -typedef struct -{ - unsigned short newState; - unsigned char symbol; - unsigned char nbBits; -} FSE_decode_t; /* size == U32 */ - -MEM_STATIC void FSE_initDState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD, const FSE_DTable* dt) -{ - const void* ptr = dt; - const FSE_DTableHeader* const DTableH = (const FSE_DTableHeader*)ptr; - DStatePtr->state = BIT_readBits(bitD, DTableH->tableLog); - BIT_reloadDStream(bitD); - DStatePtr->table = dt + 1; -} - -MEM_STATIC BYTE FSE_peekSymbol(const FSE_DState_t* DStatePtr) -{ - FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state]; - return DInfo.symbol; -} - -MEM_STATIC void FSE_updateState(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD) -{ - FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state]; - U32 const nbBits = DInfo.nbBits; - size_t const lowBits = BIT_readBits(bitD, nbBits); - DStatePtr->state = DInfo.newState + lowBits; -} - -MEM_STATIC BYTE FSE_decodeSymbol(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD) -{ - FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state]; - U32 const nbBits = DInfo.nbBits; - BYTE const symbol = DInfo.symbol; - size_t const lowBits = BIT_readBits(bitD, nbBits); - - DStatePtr->state = DInfo.newState + lowBits; - return symbol; -} - -/*! FSE_decodeSymbolFast() : - unsafe, only works if no symbol has a probability > 50% */ -MEM_STATIC BYTE FSE_decodeSymbolFast(FSE_DState_t* DStatePtr, BIT_DStream_t* bitD) -{ - FSE_decode_t const DInfo = ((const FSE_decode_t*)(DStatePtr->table))[DStatePtr->state]; - U32 const nbBits = DInfo.nbBits; - BYTE const symbol = DInfo.symbol; - size_t const lowBits = BIT_readBitsFast(bitD, nbBits); - - DStatePtr->state = DInfo.newState + lowBits; - return symbol; -} - -MEM_STATIC unsigned FSE_endOfDState(const FSE_DState_t* DStatePtr) -{ - return DStatePtr->state == 0; -} - - - -#ifndef FSE_COMMONDEFS_ONLY - -/* ************************************************************** -* Tuning parameters -****************************************************************/ -/*!MEMORY_USAGE : -* Memory usage formula : N->2^N Bytes (examples : 10 -> 1KB; 12 -> 4KB ; 16 -> 64KB; 20 -> 1MB; etc.) -* Increasing memory usage improves compression ratio -* Reduced memory usage can improve speed, due to cache effect -* Recommended max value is 14, for 16KB, which nicely fits into Intel x86 L1 cache */ -#ifndef FSE_MAX_MEMORY_USAGE -# define FSE_MAX_MEMORY_USAGE 14 -#endif -#ifndef FSE_DEFAULT_MEMORY_USAGE -# define FSE_DEFAULT_MEMORY_USAGE 13 -#endif -#if (FSE_DEFAULT_MEMORY_USAGE > FSE_MAX_MEMORY_USAGE) -# error "FSE_DEFAULT_MEMORY_USAGE must be <= FSE_MAX_MEMORY_USAGE" -#endif - -/*!FSE_MAX_SYMBOL_VALUE : -* Maximum symbol value authorized. -* Required for proper stack allocation */ -#ifndef FSE_MAX_SYMBOL_VALUE -# define FSE_MAX_SYMBOL_VALUE 255 -#endif - -/* ************************************************************** -* template functions type & suffix -****************************************************************/ -#define FSE_FUNCTION_TYPE BYTE -#define FSE_FUNCTION_EXTENSION -#define FSE_DECODE_TYPE FSE_decode_t - - -#endif /* !FSE_COMMONDEFS_ONLY */ - - -/* *************************************************************** -* Constants -*****************************************************************/ -#define FSE_MAX_TABLELOG (FSE_MAX_MEMORY_USAGE-2) -#define FSE_MAX_TABLESIZE (1U< FSE_TABLELOG_ABSOLUTE_MAX -# error "FSE_MAX_TABLELOG > FSE_TABLELOG_ABSOLUTE_MAX is not supported" -#endif - -#define FSE_TABLESTEP(tableSize) (((tableSize)>>1) + ((tableSize)>>3) + 3) - - -#endif /* FSE_STATIC_LINKING_ONLY */ - - -#if defined (__cplusplus) -} -#endif diff --git a/dep/zstd/lib/common/fse_decompress.c b/dep/zstd/lib/common/fse_decompress.c deleted file mode 100644 index a5a358015..000000000 --- a/dep/zstd/lib/common/fse_decompress.c +++ /dev/null @@ -1,403 +0,0 @@ -/* ****************************************************************** - * FSE : Finite State Entropy decoder - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy - * - Public forum : https://groups.google.com/forum/#!forum/lz4c - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -****************************************************************** */ - - -/* ************************************************************** -* Includes -****************************************************************/ -#include "debug.h" /* assert */ -#include "bitstream.h" -#include "compiler.h" -#define FSE_STATIC_LINKING_ONLY -#include "fse.h" -#include "error_private.h" -#define ZSTD_DEPS_NEED_MALLOC -#include "zstd_deps.h" - - -/* ************************************************************** -* Error Management -****************************************************************/ -#define FSE_isError ERR_isError -#define FSE_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c) /* use only *after* variable declarations */ - - -/* ************************************************************** -* Templates -****************************************************************/ -/* - designed to be included - for type-specific functions (template emulation in C) - Objective is to write these functions only once, for improved maintenance -*/ - -/* safety checks */ -#ifndef FSE_FUNCTION_EXTENSION -# error "FSE_FUNCTION_EXTENSION must be defined" -#endif -#ifndef FSE_FUNCTION_TYPE -# error "FSE_FUNCTION_TYPE must be defined" -#endif - -/* Function names */ -#define FSE_CAT(X,Y) X##Y -#define FSE_FUNCTION_NAME(X,Y) FSE_CAT(X,Y) -#define FSE_TYPE_NAME(X,Y) FSE_CAT(X,Y) - - -/* Function templates */ -FSE_DTable* FSE_createDTable (unsigned tableLog) -{ - if (tableLog > FSE_TABLELOG_ABSOLUTE_MAX) tableLog = FSE_TABLELOG_ABSOLUTE_MAX; - return (FSE_DTable*)ZSTD_malloc( FSE_DTABLE_SIZE_U32(tableLog) * sizeof (U32) ); -} - -void FSE_freeDTable (FSE_DTable* dt) -{ - ZSTD_free(dt); -} - -static size_t FSE_buildDTable_internal(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize) -{ - void* const tdPtr = dt+1; /* because *dt is unsigned, 32-bits aligned on 32-bits */ - FSE_DECODE_TYPE* const tableDecode = (FSE_DECODE_TYPE*) (tdPtr); - U16* symbolNext = (U16*)workSpace; - BYTE* spread = (BYTE*)(symbolNext + maxSymbolValue + 1); - - U32 const maxSV1 = maxSymbolValue + 1; - U32 const tableSize = 1 << tableLog; - U32 highThreshold = tableSize-1; - - /* Sanity Checks */ - if (FSE_BUILD_DTABLE_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(maxSymbolValue_tooLarge); - if (maxSymbolValue > FSE_MAX_SYMBOL_VALUE) return ERROR(maxSymbolValue_tooLarge); - if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge); - - /* Init, lay down lowprob symbols */ - { FSE_DTableHeader DTableH; - DTableH.tableLog = (U16)tableLog; - DTableH.fastMode = 1; - { S16 const largeLimit= (S16)(1 << (tableLog-1)); - U32 s; - for (s=0; s= largeLimit) DTableH.fastMode=0; - symbolNext[s] = normalizedCounter[s]; - } } } - ZSTD_memcpy(dt, &DTableH, sizeof(DTableH)); - } - - /* Spread symbols */ - if (highThreshold == tableSize - 1) { - size_t const tableMask = tableSize-1; - size_t const step = FSE_TABLESTEP(tableSize); - /* First lay down the symbols in order. - * We use a uint64_t to lay down 8 bytes at a time. This reduces branch - * misses since small blocks generally have small table logs, so nearly - * all symbols have counts <= 8. We ensure we have 8 bytes at the end of - * our buffer to handle the over-write. - */ - { - U64 const add = 0x0101010101010101ull; - size_t pos = 0; - U64 sv = 0; - U32 s; - for (s=0; s highThreshold) position = (position + step) & tableMask; /* lowprob area */ - } } - if (position!=0) return ERROR(GENERIC); /* position must reach all cells once, otherwise normalizedCounter is incorrect */ - } - - /* Build Decoding table */ - { U32 u; - for (u=0; utableLog = 0; - DTableH->fastMode = 0; - - cell->newState = 0; - cell->symbol = symbolValue; - cell->nbBits = 0; - - return 0; -} - - -size_t FSE_buildDTable_raw (FSE_DTable* dt, unsigned nbBits) -{ - void* ptr = dt; - FSE_DTableHeader* const DTableH = (FSE_DTableHeader*)ptr; - void* dPtr = dt + 1; - FSE_decode_t* const dinfo = (FSE_decode_t*)dPtr; - const unsigned tableSize = 1 << nbBits; - const unsigned tableMask = tableSize - 1; - const unsigned maxSV1 = tableMask+1; - unsigned s; - - /* Sanity checks */ - if (nbBits < 1) return ERROR(GENERIC); /* min size */ - - /* Build Decoding Table */ - DTableH->tableLog = (U16)nbBits; - DTableH->fastMode = 1; - for (s=0; s sizeof(bitD.bitContainer)*8) /* This test must be static */ - BIT_reloadDStream(&bitD); - - op[1] = FSE_GETSYMBOL(&state2); - - if (FSE_MAX_TABLELOG*4+7 > sizeof(bitD.bitContainer)*8) /* This test must be static */ - { if (BIT_reloadDStream(&bitD) > BIT_DStream_unfinished) { op+=2; break; } } - - op[2] = FSE_GETSYMBOL(&state1); - - if (FSE_MAX_TABLELOG*2+7 > sizeof(bitD.bitContainer)*8) /* This test must be static */ - BIT_reloadDStream(&bitD); - - op[3] = FSE_GETSYMBOL(&state2); - } - - /* tail */ - /* note : BIT_reloadDStream(&bitD) >= FSE_DStream_partiallyFilled; Ends at exactly BIT_DStream_completed */ - while (1) { - if (op>(omax-2)) return ERROR(dstSize_tooSmall); - *op++ = FSE_GETSYMBOL(&state1); - if (BIT_reloadDStream(&bitD)==BIT_DStream_overflow) { - *op++ = FSE_GETSYMBOL(&state2); - break; - } - - if (op>(omax-2)) return ERROR(dstSize_tooSmall); - *op++ = FSE_GETSYMBOL(&state2); - if (BIT_reloadDStream(&bitD)==BIT_DStream_overflow) { - *op++ = FSE_GETSYMBOL(&state1); - break; - } } - - return op-ostart; -} - - -size_t FSE_decompress_usingDTable(void* dst, size_t originalSize, - const void* cSrc, size_t cSrcSize, - const FSE_DTable* dt) -{ - const void* ptr = dt; - const FSE_DTableHeader* DTableH = (const FSE_DTableHeader*)ptr; - const U32 fastMode = DTableH->fastMode; - - /* select fast mode (static) */ - if (fastMode) return FSE_decompress_usingDTable_generic(dst, originalSize, cSrc, cSrcSize, dt, 1); - return FSE_decompress_usingDTable_generic(dst, originalSize, cSrc, cSrcSize, dt, 0); -} - - -size_t FSE_decompress_wksp(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize) -{ - return FSE_decompress_wksp_bmi2(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, /* bmi2 */ 0); -} - -typedef struct { - short ncount[FSE_MAX_SYMBOL_VALUE + 1]; - FSE_DTable dtable[1]; /* Dynamically sized */ -} FSE_DecompressWksp; - - -FORCE_INLINE_TEMPLATE size_t FSE_decompress_wksp_body( - void* dst, size_t dstCapacity, - const void* cSrc, size_t cSrcSize, - unsigned maxLog, void* workSpace, size_t wkspSize, - int bmi2) -{ - const BYTE* const istart = (const BYTE*)cSrc; - const BYTE* ip = istart; - unsigned tableLog; - unsigned maxSymbolValue = FSE_MAX_SYMBOL_VALUE; - FSE_DecompressWksp* const wksp = (FSE_DecompressWksp*)workSpace; - - DEBUG_STATIC_ASSERT((FSE_MAX_SYMBOL_VALUE + 1) % 2 == 0); - if (wkspSize < sizeof(*wksp)) return ERROR(GENERIC); - - /* normal FSE decoding mode */ - { - size_t const NCountLength = FSE_readNCount_bmi2(wksp->ncount, &maxSymbolValue, &tableLog, istart, cSrcSize, bmi2); - if (FSE_isError(NCountLength)) return NCountLength; - if (tableLog > maxLog) return ERROR(tableLog_tooLarge); - assert(NCountLength <= cSrcSize); - ip += NCountLength; - cSrcSize -= NCountLength; - } - - if (FSE_DECOMPRESS_WKSP_SIZE(tableLog, maxSymbolValue) > wkspSize) return ERROR(tableLog_tooLarge); - workSpace = wksp->dtable + FSE_DTABLE_SIZE_U32(tableLog); - wkspSize -= sizeof(*wksp) + FSE_DTABLE_SIZE(tableLog); - - CHECK_F( FSE_buildDTable_internal(wksp->dtable, wksp->ncount, maxSymbolValue, tableLog, workSpace, wkspSize) ); - - { - const void* ptr = wksp->dtable; - const FSE_DTableHeader* DTableH = (const FSE_DTableHeader*)ptr; - const U32 fastMode = DTableH->fastMode; - - /* select fast mode (static) */ - if (fastMode) return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, wksp->dtable, 1); - return FSE_decompress_usingDTable_generic(dst, dstCapacity, ip, cSrcSize, wksp->dtable, 0); - } -} - -/* Avoids the FORCE_INLINE of the _body() function. */ -static size_t FSE_decompress_wksp_body_default(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize) -{ - return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 0); -} - -#if DYNAMIC_BMI2 -BMI2_TARGET_ATTRIBUTE static size_t FSE_decompress_wksp_body_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize) -{ - return FSE_decompress_wksp_body(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize, 1); -} -#endif - -size_t FSE_decompress_wksp_bmi2(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize, unsigned maxLog, void* workSpace, size_t wkspSize, int bmi2) -{ -#if DYNAMIC_BMI2 - if (bmi2) { - return FSE_decompress_wksp_body_bmi2(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize); - } -#endif - (void)bmi2; - return FSE_decompress_wksp_body_default(dst, dstCapacity, cSrc, cSrcSize, maxLog, workSpace, wkspSize); -} - - -typedef FSE_DTable DTable_max_t[FSE_DTABLE_SIZE_U32(FSE_MAX_TABLELOG)]; - -#ifndef ZSTD_NO_UNUSED_FUNCTIONS -size_t FSE_buildDTable(FSE_DTable* dt, const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog) { - U32 wksp[FSE_BUILD_DTABLE_WKSP_SIZE_U32(FSE_TABLELOG_ABSOLUTE_MAX, FSE_MAX_SYMBOL_VALUE)]; - return FSE_buildDTable_wksp(dt, normalizedCounter, maxSymbolValue, tableLog, wksp, sizeof(wksp)); -} - -size_t FSE_decompress(void* dst, size_t dstCapacity, const void* cSrc, size_t cSrcSize) -{ - /* Static analyzer seems unable to understand this table will be properly initialized later */ - U32 wksp[FSE_DECOMPRESS_WKSP_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)]; - return FSE_decompress_wksp(dst, dstCapacity, cSrc, cSrcSize, FSE_MAX_TABLELOG, wksp, sizeof(wksp)); -} -#endif - - -#endif /* FSE_COMMONDEFS_ONLY */ diff --git a/dep/zstd/lib/common/huf.h b/dep/zstd/lib/common/huf.h deleted file mode 100644 index 85518481e..000000000 --- a/dep/zstd/lib/common/huf.h +++ /dev/null @@ -1,364 +0,0 @@ -/* ****************************************************************** - * huff0 huffman codec, - * part of Finite State Entropy library - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - Source repository : https://github.com/Cyan4973/FiniteStateEntropy - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -****************************************************************** */ - -#if defined (__cplusplus) -extern "C" { -#endif - -#ifndef HUF_H_298734234 -#define HUF_H_298734234 - -/* *** Dependencies *** */ -#include "zstd_deps.h" /* size_t */ - - -/* *** library symbols visibility *** */ -/* Note : when linking with -fvisibility=hidden on gcc, or by default on Visual, - * HUF symbols remain "private" (internal symbols for library only). - * Set macro FSE_DLL_EXPORT to 1 if you want HUF symbols visible on DLL interface */ -#if defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) && defined(__GNUC__) && (__GNUC__ >= 4) -# define HUF_PUBLIC_API __attribute__ ((visibility ("default"))) -#elif defined(FSE_DLL_EXPORT) && (FSE_DLL_EXPORT==1) /* Visual expected */ -# define HUF_PUBLIC_API __declspec(dllexport) -#elif defined(FSE_DLL_IMPORT) && (FSE_DLL_IMPORT==1) -# define HUF_PUBLIC_API __declspec(dllimport) /* not required, just to generate faster code (saves a function pointer load from IAT and an indirect jump) */ -#else -# define HUF_PUBLIC_API -#endif - - -/* ========================== */ -/* *** simple functions *** */ -/* ========================== */ - -/** HUF_compress() : - * Compress content from buffer 'src', of size 'srcSize', into buffer 'dst'. - * 'dst' buffer must be already allocated. - * Compression runs faster if `dstCapacity` >= HUF_compressBound(srcSize). - * `srcSize` must be <= `HUF_BLOCKSIZE_MAX` == 128 KB. - * @return : size of compressed data (<= `dstCapacity`). - * Special values : if return == 0, srcData is not compressible => Nothing is stored within dst !!! - * if HUF_isError(return), compression failed (more details using HUF_getErrorName()) - */ -HUF_PUBLIC_API size_t HUF_compress(void* dst, size_t dstCapacity, - const void* src, size_t srcSize); - -/** HUF_decompress() : - * Decompress HUF data from buffer 'cSrc', of size 'cSrcSize', - * into already allocated buffer 'dst', of minimum size 'dstSize'. - * `originalSize` : **must** be the ***exact*** size of original (uncompressed) data. - * Note : in contrast with FSE, HUF_decompress can regenerate - * RLE (cSrcSize==1) and uncompressed (cSrcSize==dstSize) data, - * because it knows size to regenerate (originalSize). - * @return : size of regenerated data (== originalSize), - * or an error code, which can be tested using HUF_isError() - */ -HUF_PUBLIC_API size_t HUF_decompress(void* dst, size_t originalSize, - const void* cSrc, size_t cSrcSize); - - -/* *** Tool functions *** */ -#define HUF_BLOCKSIZE_MAX (128 * 1024) /**< maximum input size for a single block compressed with HUF_compress */ -HUF_PUBLIC_API size_t HUF_compressBound(size_t size); /**< maximum compressed size (worst case) */ - -/* Error Management */ -HUF_PUBLIC_API unsigned HUF_isError(size_t code); /**< tells if a return value is an error code */ -HUF_PUBLIC_API const char* HUF_getErrorName(size_t code); /**< provides error code string (useful for debugging) */ - - -/* *** Advanced function *** */ - -/** HUF_compress2() : - * Same as HUF_compress(), but offers control over `maxSymbolValue` and `tableLog`. - * `maxSymbolValue` must be <= HUF_SYMBOLVALUE_MAX . - * `tableLog` must be `<= HUF_TABLELOG_MAX` . */ -HUF_PUBLIC_API size_t HUF_compress2 (void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - unsigned maxSymbolValue, unsigned tableLog); - -/** HUF_compress4X_wksp() : - * Same as HUF_compress2(), but uses externally allocated `workSpace`. - * `workspace` must be at least as large as HUF_WORKSPACE_SIZE */ -#define HUF_WORKSPACE_SIZE ((8 << 10) + 512 /* sorting scratch space */) -#define HUF_WORKSPACE_SIZE_U64 (HUF_WORKSPACE_SIZE / sizeof(U64)) -HUF_PUBLIC_API size_t HUF_compress4X_wksp (void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - unsigned maxSymbolValue, unsigned tableLog, - void* workSpace, size_t wkspSize); - -#endif /* HUF_H_298734234 */ - -/* ****************************************************************** - * WARNING !! - * The following section contains advanced and experimental definitions - * which shall never be used in the context of a dynamic library, - * because they are not guaranteed to remain stable in the future. - * Only consider them in association with static linking. - * *****************************************************************/ -#if defined(HUF_STATIC_LINKING_ONLY) && !defined(HUF_H_HUF_STATIC_LINKING_ONLY) -#define HUF_H_HUF_STATIC_LINKING_ONLY - -/* *** Dependencies *** */ -#include "mem.h" /* U32 */ -#define FSE_STATIC_LINKING_ONLY -#include "fse.h" - - -/* *** Constants *** */ -#define HUF_TABLELOG_MAX 12 /* max runtime value of tableLog (due to static allocation); can be modified up to HUF_TABLELOG_ABSOLUTEMAX */ -#define HUF_TABLELOG_DEFAULT 11 /* default tableLog value when none specified */ -#define HUF_SYMBOLVALUE_MAX 255 - -#define HUF_TABLELOG_ABSOLUTEMAX 12 /* absolute limit of HUF_MAX_TABLELOG. Beyond that value, code does not work */ -#if (HUF_TABLELOG_MAX > HUF_TABLELOG_ABSOLUTEMAX) -# error "HUF_TABLELOG_MAX is too large !" -#endif - - -/* **************************************** -* Static allocation -******************************************/ -/* HUF buffer bounds */ -#define HUF_CTABLEBOUND 129 -#define HUF_BLOCKBOUND(size) (size + (size>>8) + 8) /* only true when incompressible is pre-filtered with fast heuristic */ -#define HUF_COMPRESSBOUND(size) (HUF_CTABLEBOUND + HUF_BLOCKBOUND(size)) /* Macro version, useful for static allocation */ - -/* static allocation of HUF's Compression Table */ -/* this is a private definition, just exposed for allocation and strict aliasing purpose. never EVER access its members directly */ -typedef size_t HUF_CElt; /* consider it an incomplete type */ -#define HUF_CTABLE_SIZE_ST(maxSymbolValue) ((maxSymbolValue)+2) /* Use tables of size_t, for proper alignment */ -#define HUF_CTABLE_SIZE(maxSymbolValue) (HUF_CTABLE_SIZE_ST(maxSymbolValue) * sizeof(size_t)) -#define HUF_CREATE_STATIC_CTABLE(name, maxSymbolValue) \ - HUF_CElt name[HUF_CTABLE_SIZE_ST(maxSymbolValue)] /* no final ; */ - -/* static allocation of HUF's DTable */ -typedef U32 HUF_DTable; -#define HUF_DTABLE_SIZE(maxTableLog) (1 + (1<<(maxTableLog))) -#define HUF_CREATE_STATIC_DTABLEX1(DTable, maxTableLog) \ - HUF_DTable DTable[HUF_DTABLE_SIZE((maxTableLog)-1)] = { ((U32)((maxTableLog)-1) * 0x01000001) } -#define HUF_CREATE_STATIC_DTABLEX2(DTable, maxTableLog) \ - HUF_DTable DTable[HUF_DTABLE_SIZE(maxTableLog)] = { ((U32)(maxTableLog) * 0x01000001) } - - -/* **************************************** -* Advanced decompression functions -******************************************/ -size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< single-symbol decoder */ -#ifndef HUF_FORCE_DECOMPRESS_X1 -size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< double-symbols decoder */ -#endif - -size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< decodes RLE and uncompressed */ -size_t HUF_decompress4X_hufOnly(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< considers RLE and uncompressed as errors */ -size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< considers RLE and uncompressed as errors */ -size_t HUF_decompress4X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< single-symbol decoder */ -size_t HUF_decompress4X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< single-symbol decoder */ -#ifndef HUF_FORCE_DECOMPRESS_X1 -size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< double-symbols decoder */ -size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< double-symbols decoder */ -#endif - - -/* **************************************** - * HUF detailed API - * ****************************************/ - -/*! HUF_compress() does the following: - * 1. count symbol occurrence from source[] into table count[] using FSE_count() (exposed within "fse.h") - * 2. (optional) refine tableLog using HUF_optimalTableLog() - * 3. build Huffman table from count using HUF_buildCTable() - * 4. save Huffman table to memory buffer using HUF_writeCTable() - * 5. encode the data stream using HUF_compress4X_usingCTable() - * - * The following API allows targeting specific sub-functions for advanced tasks. - * For example, it's possible to compress several blocks using the same 'CTable', - * or to save and regenerate 'CTable' using external methods. - */ -unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue); -size_t HUF_buildCTable (HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue, unsigned maxNbBits); /* @return : maxNbBits; CTable and count can overlap. In which case, CTable will overwrite count content */ -size_t HUF_writeCTable (void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog); -size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, void* workspace, size_t workspaceSize); -size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable); -size_t HUF_compress4X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2); -size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue); -int HUF_validateCTable(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue); - -typedef enum { - HUF_repeat_none, /**< Cannot use the previous table */ - HUF_repeat_check, /**< Can use the previous table but it must be checked. Note : The previous table must have been constructed by HUF_compress{1, 4}X_repeat */ - HUF_repeat_valid /**< Can use the previous table and it is assumed to be valid */ - } HUF_repeat; -/** HUF_compress4X_repeat() : - * Same as HUF_compress4X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none. - * If it uses hufTable it does not modify hufTable or repeat. - * If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used. - * If preferRepeat then the old table will always be used if valid. - * If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */ -size_t HUF_compress4X_repeat(void* dst, size_t dstSize, - const void* src, size_t srcSize, - unsigned maxSymbolValue, unsigned tableLog, - void* workSpace, size_t wkspSize, /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */ - HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2, unsigned suspectUncompressible); - -/** HUF_buildCTable_wksp() : - * Same as HUF_buildCTable(), but using externally allocated scratch buffer. - * `workSpace` must be aligned on 4-bytes boundaries, and its size must be >= HUF_CTABLE_WORKSPACE_SIZE. - */ -#define HUF_CTABLE_WORKSPACE_SIZE_U32 (2*HUF_SYMBOLVALUE_MAX +1 +1) -#define HUF_CTABLE_WORKSPACE_SIZE (HUF_CTABLE_WORKSPACE_SIZE_U32 * sizeof(unsigned)) -size_t HUF_buildCTable_wksp (HUF_CElt* tree, - const unsigned* count, U32 maxSymbolValue, U32 maxNbBits, - void* workSpace, size_t wkspSize); - -/*! HUF_readStats() : - * Read compact Huffman tree, saved by HUF_writeCTable(). - * `huffWeight` is destination buffer. - * @return : size read from `src` , or an error Code . - * Note : Needed by HUF_readCTable() and HUF_readDTableXn() . */ -size_t HUF_readStats(BYTE* huffWeight, size_t hwSize, - U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr, - const void* src, size_t srcSize); - -/*! HUF_readStats_wksp() : - * Same as HUF_readStats() but takes an external workspace which must be - * 4-byte aligned and its size must be >= HUF_READ_STATS_WORKSPACE_SIZE. - * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0. - */ -#define HUF_READ_STATS_WORKSPACE_SIZE_U32 FSE_DECOMPRESS_WKSP_SIZE_U32(6, HUF_TABLELOG_MAX-1) -#define HUF_READ_STATS_WORKSPACE_SIZE (HUF_READ_STATS_WORKSPACE_SIZE_U32 * sizeof(unsigned)) -size_t HUF_readStats_wksp(BYTE* huffWeight, size_t hwSize, - U32* rankStats, U32* nbSymbolsPtr, U32* tableLogPtr, - const void* src, size_t srcSize, - void* workspace, size_t wkspSize, - int bmi2); - -/** HUF_readCTable() : - * Loading a CTable saved with HUF_writeCTable() */ -size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned *hasZeroWeights); - -/** HUF_getNbBitsFromCTable() : - * Read nbBits from CTable symbolTable, for symbol `symbolValue` presumed <= HUF_SYMBOLVALUE_MAX - * Note 1 : is not inlined, as HUF_CElt definition is private */ -U32 HUF_getNbBitsFromCTable(const HUF_CElt* symbolTable, U32 symbolValue); - -/* - * HUF_decompress() does the following: - * 1. select the decompression algorithm (X1, X2) based on pre-computed heuristics - * 2. build Huffman table from save, using HUF_readDTableX?() - * 3. decode 1 or 4 segments in parallel using HUF_decompress?X?_usingDTable() - */ - -/** HUF_selectDecoder() : - * Tells which decoder is likely to decode faster, - * based on a set of pre-computed metrics. - * @return : 0==HUF_decompress4X1, 1==HUF_decompress4X2 . - * Assumption : 0 < dstSize <= 128 KB */ -U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize); - -/** - * The minimum workspace size for the `workSpace` used in - * HUF_readDTableX1_wksp() and HUF_readDTableX2_wksp(). - * - * The space used depends on HUF_TABLELOG_MAX, ranging from ~1500 bytes when - * HUF_TABLE_LOG_MAX=12 to ~1850 bytes when HUF_TABLE_LOG_MAX=15. - * Buffer overflow errors may potentially occur if code modifications result in - * a required workspace size greater than that specified in the following - * macro. - */ -#define HUF_DECOMPRESS_WORKSPACE_SIZE ((2 << 10) + (1 << 9)) -#define HUF_DECOMPRESS_WORKSPACE_SIZE_U32 (HUF_DECOMPRESS_WORKSPACE_SIZE / sizeof(U32)) - -#ifndef HUF_FORCE_DECOMPRESS_X2 -size_t HUF_readDTableX1 (HUF_DTable* DTable, const void* src, size_t srcSize); -size_t HUF_readDTableX1_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize); -#endif -#ifndef HUF_FORCE_DECOMPRESS_X1 -size_t HUF_readDTableX2 (HUF_DTable* DTable, const void* src, size_t srcSize); -size_t HUF_readDTableX2_wksp (HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize); -#endif - -size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); -#ifndef HUF_FORCE_DECOMPRESS_X2 -size_t HUF_decompress4X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); -#endif -#ifndef HUF_FORCE_DECOMPRESS_X1 -size_t HUF_decompress4X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); -#endif - - -/* ====================== */ -/* single stream variants */ -/* ====================== */ - -size_t HUF_compress1X (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog); -size_t HUF_compress1X_wksp (void* dst, size_t dstSize, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog, void* workSpace, size_t wkspSize); /**< `workSpace` must be a table of at least HUF_WORKSPACE_SIZE_U64 U64 */ -size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable); -size_t HUF_compress1X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2); -/** HUF_compress1X_repeat() : - * Same as HUF_compress1X_wksp(), but considers using hufTable if *repeat != HUF_repeat_none. - * If it uses hufTable it does not modify hufTable or repeat. - * If it doesn't, it sets *repeat = HUF_repeat_none, and it sets hufTable to the table used. - * If preferRepeat then the old table will always be used if valid. - * If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */ -size_t HUF_compress1X_repeat(void* dst, size_t dstSize, - const void* src, size_t srcSize, - unsigned maxSymbolValue, unsigned tableLog, - void* workSpace, size_t wkspSize, /**< `workSpace` must be aligned on 4-bytes boundaries, `wkspSize` must be >= HUF_WORKSPACE_SIZE */ - HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2, unsigned suspectUncompressible); - -size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* single-symbol decoder */ -#ifndef HUF_FORCE_DECOMPRESS_X1 -size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /* double-symbol decoder */ -#endif - -size_t HUF_decompress1X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); -size_t HUF_decompress1X_DCtx_wksp (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); -#ifndef HUF_FORCE_DECOMPRESS_X2 -size_t HUF_decompress1X1_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< single-symbol decoder */ -size_t HUF_decompress1X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< single-symbol decoder */ -#endif -#ifndef HUF_FORCE_DECOMPRESS_X1 -size_t HUF_decompress1X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); /**< double-symbols decoder */ -size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize); /**< double-symbols decoder */ -#endif - -size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); /**< automatic selection of sing or double symbol decoder, based on DTable */ -#ifndef HUF_FORCE_DECOMPRESS_X2 -size_t HUF_decompress1X1_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); -#endif -#ifndef HUF_FORCE_DECOMPRESS_X1 -size_t HUF_decompress1X2_usingDTable(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable); -#endif - -/* BMI2 variants. - * If the CPU has BMI2 support, pass bmi2=1, otherwise pass bmi2=0. - */ -size_t HUF_decompress1X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2); -#ifndef HUF_FORCE_DECOMPRESS_X2 -size_t HUF_decompress1X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2); -#endif -size_t HUF_decompress4X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2); -size_t HUF_decompress4X_hufOnly_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2); -#ifndef HUF_FORCE_DECOMPRESS_X2 -size_t HUF_readDTableX1_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2); -#endif -#ifndef HUF_FORCE_DECOMPRESS_X1 -size_t HUF_readDTableX2_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2); -#endif - -#endif /* HUF_STATIC_LINKING_ONLY */ - -#if defined (__cplusplus) -} -#endif diff --git a/dep/zstd/lib/common/mem.h b/dep/zstd/lib/common/mem.h deleted file mode 100644 index 85581c384..000000000 --- a/dep/zstd/lib/common/mem.h +++ /dev/null @@ -1,442 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef MEM_H_MODULE -#define MEM_H_MODULE - -#if defined (__cplusplus) -extern "C" { -#endif - -/*-**************************************** -* Dependencies -******************************************/ -#include /* size_t, ptrdiff_t */ -#include "compiler.h" /* __has_builtin */ -#include "debug.h" /* DEBUG_STATIC_ASSERT */ -#include "zstd_deps.h" /* ZSTD_memcpy */ - - -/*-**************************************** -* Compiler specifics -******************************************/ -#if defined(_MSC_VER) /* Visual Studio */ -# include /* _byteswap_ulong */ -# include /* _byteswap_* */ -#endif -#if defined(__GNUC__) -# define MEM_STATIC static __inline __attribute__((unused)) -#elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) -# define MEM_STATIC static inline -#elif defined(_MSC_VER) -# define MEM_STATIC static __inline -#else -# define MEM_STATIC static /* this version may generate warnings for unused static functions; disable the relevant warning */ -#endif - -/*-************************************************************** -* Basic Types -*****************************************************************/ -#if !defined (__VMS) && (defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) ) -# if defined(_AIX) -# include -# else -# include /* intptr_t */ -# endif - typedef uint8_t BYTE; - typedef uint8_t U8; - typedef int8_t S8; - typedef uint16_t U16; - typedef int16_t S16; - typedef uint32_t U32; - typedef int32_t S32; - typedef uint64_t U64; - typedef int64_t S64; -#else -# include -#if CHAR_BIT != 8 -# error "this implementation requires char to be exactly 8-bit type" -#endif - typedef unsigned char BYTE; - typedef unsigned char U8; - typedef signed char S8; -#if USHRT_MAX != 65535 -# error "this implementation requires short to be exactly 16-bit type" -#endif - typedef unsigned short U16; - typedef signed short S16; -#if UINT_MAX != 4294967295 -# error "this implementation requires int to be exactly 32-bit type" -#endif - typedef unsigned int U32; - typedef signed int S32; -/* note : there are no limits defined for long long type in C90. - * limits exist in C99, however, in such case, is preferred */ - typedef unsigned long long U64; - typedef signed long long S64; -#endif - - -/*-************************************************************** -* Memory I/O API -*****************************************************************/ -/*=== Static platform detection ===*/ -MEM_STATIC unsigned MEM_32bits(void); -MEM_STATIC unsigned MEM_64bits(void); -MEM_STATIC unsigned MEM_isLittleEndian(void); - -/*=== Native unaligned read/write ===*/ -MEM_STATIC U16 MEM_read16(const void* memPtr); -MEM_STATIC U32 MEM_read32(const void* memPtr); -MEM_STATIC U64 MEM_read64(const void* memPtr); -MEM_STATIC size_t MEM_readST(const void* memPtr); - -MEM_STATIC void MEM_write16(void* memPtr, U16 value); -MEM_STATIC void MEM_write32(void* memPtr, U32 value); -MEM_STATIC void MEM_write64(void* memPtr, U64 value); - -/*=== Little endian unaligned read/write ===*/ -MEM_STATIC U16 MEM_readLE16(const void* memPtr); -MEM_STATIC U32 MEM_readLE24(const void* memPtr); -MEM_STATIC U32 MEM_readLE32(const void* memPtr); -MEM_STATIC U64 MEM_readLE64(const void* memPtr); -MEM_STATIC size_t MEM_readLEST(const void* memPtr); - -MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val); -MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val); -MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32); -MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64); -MEM_STATIC void MEM_writeLEST(void* memPtr, size_t val); - -/*=== Big endian unaligned read/write ===*/ -MEM_STATIC U32 MEM_readBE32(const void* memPtr); -MEM_STATIC U64 MEM_readBE64(const void* memPtr); -MEM_STATIC size_t MEM_readBEST(const void* memPtr); - -MEM_STATIC void MEM_writeBE32(void* memPtr, U32 val32); -MEM_STATIC void MEM_writeBE64(void* memPtr, U64 val64); -MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val); - -/*=== Byteswap ===*/ -MEM_STATIC U32 MEM_swap32(U32 in); -MEM_STATIC U64 MEM_swap64(U64 in); -MEM_STATIC size_t MEM_swapST(size_t in); - - -/*-************************************************************** -* Memory I/O Implementation -*****************************************************************/ -/* MEM_FORCE_MEMORY_ACCESS : - * By default, access to unaligned memory is controlled by `memcpy()`, which is safe and portable. - * Unfortunately, on some target/compiler combinations, the generated assembly is sub-optimal. - * The below switch allow to select different access method for improved performance. - * Method 0 (default) : use `memcpy()`. Safe and portable. - * Method 1 : `__packed` statement. It depends on compiler extension (i.e., not portable). - * This method is safe if your compiler supports it, and *generally* as fast or faster than `memcpy`. - * Method 2 : direct access. This method is portable but violate C standard. - * It can generate buggy code on targets depending on alignment. - * In some circumstances, it's the only known way to get the most performance (i.e. GCC + ARMv6) - * See http://fastcompression.blogspot.fr/2015/08/accessing-unaligned-memory.html for details. - * Prefer these methods in priority order (0 > 1 > 2) - */ -#ifndef MEM_FORCE_MEMORY_ACCESS /* can be defined externally, on command line for example */ -# if defined(__INTEL_COMPILER) || defined(__GNUC__) || defined(__ICCARM__) -# define MEM_FORCE_MEMORY_ACCESS 1 -# endif -#endif - -MEM_STATIC unsigned MEM_32bits(void) { return sizeof(size_t)==4; } -MEM_STATIC unsigned MEM_64bits(void) { return sizeof(size_t)==8; } - -MEM_STATIC unsigned MEM_isLittleEndian(void) -{ -#if defined(__BYTE_ORDER__) && defined(__ORDER_LITTLE_ENDIAN__) && (__BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__) - return 1; -#elif defined(__BYTE_ORDER__) && defined(__ORDER_BIG_ENDIAN__) && (__BYTE_ORDER__ == __ORDER_BIG_ENDIAN__) - return 0; -#elif defined(__clang__) && __LITTLE_ENDIAN__ - return 1; -#elif defined(__clang__) && __BIG_ENDIAN__ - return 0; -#elif defined(_MSC_VER) && (_M_AMD64 || _M_IX86) - return 1; -#elif defined(__DMC__) && defined(_M_IX86) - return 1; -#else - const union { U32 u; BYTE c[4]; } one = { 1 }; /* don't use static : performance detrimental */ - return one.c[0]; -#endif -} - -#if defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==2) - -/* violates C standard, by lying on structure alignment. -Only use if no other choice to achieve best performance on target platform */ -MEM_STATIC U16 MEM_read16(const void* memPtr) { return *(const U16*) memPtr; } -MEM_STATIC U32 MEM_read32(const void* memPtr) { return *(const U32*) memPtr; } -MEM_STATIC U64 MEM_read64(const void* memPtr) { return *(const U64*) memPtr; } -MEM_STATIC size_t MEM_readST(const void* memPtr) { return *(const size_t*) memPtr; } - -MEM_STATIC void MEM_write16(void* memPtr, U16 value) { *(U16*)memPtr = value; } -MEM_STATIC void MEM_write32(void* memPtr, U32 value) { *(U32*)memPtr = value; } -MEM_STATIC void MEM_write64(void* memPtr, U64 value) { *(U64*)memPtr = value; } - -#elif defined(MEM_FORCE_MEMORY_ACCESS) && (MEM_FORCE_MEMORY_ACCESS==1) - -/* __pack instructions are safer, but compiler specific, hence potentially problematic for some compilers */ -/* currently only defined for gcc and icc */ -#if defined(_MSC_VER) || (defined(__INTEL_COMPILER) && defined(WIN32)) - __pragma( pack(push, 1) ) - typedef struct { U16 v; } unalign16; - typedef struct { U32 v; } unalign32; - typedef struct { U64 v; } unalign64; - typedef struct { size_t v; } unalignArch; - __pragma( pack(pop) ) -#else - typedef struct { U16 v; } __attribute__((packed)) unalign16; - typedef struct { U32 v; } __attribute__((packed)) unalign32; - typedef struct { U64 v; } __attribute__((packed)) unalign64; - typedef struct { size_t v; } __attribute__((packed)) unalignArch; -#endif - -MEM_STATIC U16 MEM_read16(const void* ptr) { return ((const unalign16*)ptr)->v; } -MEM_STATIC U32 MEM_read32(const void* ptr) { return ((const unalign32*)ptr)->v; } -MEM_STATIC U64 MEM_read64(const void* ptr) { return ((const unalign64*)ptr)->v; } -MEM_STATIC size_t MEM_readST(const void* ptr) { return ((const unalignArch*)ptr)->v; } - -MEM_STATIC void MEM_write16(void* memPtr, U16 value) { ((unalign16*)memPtr)->v = value; } -MEM_STATIC void MEM_write32(void* memPtr, U32 value) { ((unalign32*)memPtr)->v = value; } -MEM_STATIC void MEM_write64(void* memPtr, U64 value) { ((unalign64*)memPtr)->v = value; } - -#else - -/* default method, safe and standard. - can sometimes prove slower */ - -MEM_STATIC U16 MEM_read16(const void* memPtr) -{ - U16 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val; -} - -MEM_STATIC U32 MEM_read32(const void* memPtr) -{ - U32 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val; -} - -MEM_STATIC U64 MEM_read64(const void* memPtr) -{ - U64 val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val; -} - -MEM_STATIC size_t MEM_readST(const void* memPtr) -{ - size_t val; ZSTD_memcpy(&val, memPtr, sizeof(val)); return val; -} - -MEM_STATIC void MEM_write16(void* memPtr, U16 value) -{ - ZSTD_memcpy(memPtr, &value, sizeof(value)); -} - -MEM_STATIC void MEM_write32(void* memPtr, U32 value) -{ - ZSTD_memcpy(memPtr, &value, sizeof(value)); -} - -MEM_STATIC void MEM_write64(void* memPtr, U64 value) -{ - ZSTD_memcpy(memPtr, &value, sizeof(value)); -} - -#endif /* MEM_FORCE_MEMORY_ACCESS */ - -MEM_STATIC U32 MEM_swap32(U32 in) -{ -#if defined(_MSC_VER) /* Visual Studio */ - return _byteswap_ulong(in); -#elif (defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)) \ - || (defined(__clang__) && __has_builtin(__builtin_bswap32)) - return __builtin_bswap32(in); -#else - return ((in << 24) & 0xff000000 ) | - ((in << 8) & 0x00ff0000 ) | - ((in >> 8) & 0x0000ff00 ) | - ((in >> 24) & 0x000000ff ); -#endif -} - -MEM_STATIC U64 MEM_swap64(U64 in) -{ -#if defined(_MSC_VER) /* Visual Studio */ - return _byteswap_uint64(in); -#elif (defined (__GNUC__) && (__GNUC__ * 100 + __GNUC_MINOR__ >= 403)) \ - || (defined(__clang__) && __has_builtin(__builtin_bswap64)) - return __builtin_bswap64(in); -#else - return ((in << 56) & 0xff00000000000000ULL) | - ((in << 40) & 0x00ff000000000000ULL) | - ((in << 24) & 0x0000ff0000000000ULL) | - ((in << 8) & 0x000000ff00000000ULL) | - ((in >> 8) & 0x00000000ff000000ULL) | - ((in >> 24) & 0x0000000000ff0000ULL) | - ((in >> 40) & 0x000000000000ff00ULL) | - ((in >> 56) & 0x00000000000000ffULL); -#endif -} - -MEM_STATIC size_t MEM_swapST(size_t in) -{ - if (MEM_32bits()) - return (size_t)MEM_swap32((U32)in); - else - return (size_t)MEM_swap64((U64)in); -} - -/*=== Little endian r/w ===*/ - -MEM_STATIC U16 MEM_readLE16(const void* memPtr) -{ - if (MEM_isLittleEndian()) - return MEM_read16(memPtr); - else { - const BYTE* p = (const BYTE*)memPtr; - return (U16)(p[0] + (p[1]<<8)); - } -} - -MEM_STATIC void MEM_writeLE16(void* memPtr, U16 val) -{ - if (MEM_isLittleEndian()) { - MEM_write16(memPtr, val); - } else { - BYTE* p = (BYTE*)memPtr; - p[0] = (BYTE)val; - p[1] = (BYTE)(val>>8); - } -} - -MEM_STATIC U32 MEM_readLE24(const void* memPtr) -{ - return (U32)MEM_readLE16(memPtr) + ((U32)(((const BYTE*)memPtr)[2]) << 16); -} - -MEM_STATIC void MEM_writeLE24(void* memPtr, U32 val) -{ - MEM_writeLE16(memPtr, (U16)val); - ((BYTE*)memPtr)[2] = (BYTE)(val>>16); -} - -MEM_STATIC U32 MEM_readLE32(const void* memPtr) -{ - if (MEM_isLittleEndian()) - return MEM_read32(memPtr); - else - return MEM_swap32(MEM_read32(memPtr)); -} - -MEM_STATIC void MEM_writeLE32(void* memPtr, U32 val32) -{ - if (MEM_isLittleEndian()) - MEM_write32(memPtr, val32); - else - MEM_write32(memPtr, MEM_swap32(val32)); -} - -MEM_STATIC U64 MEM_readLE64(const void* memPtr) -{ - if (MEM_isLittleEndian()) - return MEM_read64(memPtr); - else - return MEM_swap64(MEM_read64(memPtr)); -} - -MEM_STATIC void MEM_writeLE64(void* memPtr, U64 val64) -{ - if (MEM_isLittleEndian()) - MEM_write64(memPtr, val64); - else - MEM_write64(memPtr, MEM_swap64(val64)); -} - -MEM_STATIC size_t MEM_readLEST(const void* memPtr) -{ - if (MEM_32bits()) - return (size_t)MEM_readLE32(memPtr); - else - return (size_t)MEM_readLE64(memPtr); -} - -MEM_STATIC void MEM_writeLEST(void* memPtr, size_t val) -{ - if (MEM_32bits()) - MEM_writeLE32(memPtr, (U32)val); - else - MEM_writeLE64(memPtr, (U64)val); -} - -/*=== Big endian r/w ===*/ - -MEM_STATIC U32 MEM_readBE32(const void* memPtr) -{ - if (MEM_isLittleEndian()) - return MEM_swap32(MEM_read32(memPtr)); - else - return MEM_read32(memPtr); -} - -MEM_STATIC void MEM_writeBE32(void* memPtr, U32 val32) -{ - if (MEM_isLittleEndian()) - MEM_write32(memPtr, MEM_swap32(val32)); - else - MEM_write32(memPtr, val32); -} - -MEM_STATIC U64 MEM_readBE64(const void* memPtr) -{ - if (MEM_isLittleEndian()) - return MEM_swap64(MEM_read64(memPtr)); - else - return MEM_read64(memPtr); -} - -MEM_STATIC void MEM_writeBE64(void* memPtr, U64 val64) -{ - if (MEM_isLittleEndian()) - MEM_write64(memPtr, MEM_swap64(val64)); - else - MEM_write64(memPtr, val64); -} - -MEM_STATIC size_t MEM_readBEST(const void* memPtr) -{ - if (MEM_32bits()) - return (size_t)MEM_readBE32(memPtr); - else - return (size_t)MEM_readBE64(memPtr); -} - -MEM_STATIC void MEM_writeBEST(void* memPtr, size_t val) -{ - if (MEM_32bits()) - MEM_writeBE32(memPtr, (U32)val); - else - MEM_writeBE64(memPtr, (U64)val); -} - -/* code only tested on 32 and 64 bits systems */ -MEM_STATIC void MEM_check(void) { DEBUG_STATIC_ASSERT((sizeof(size_t)==4) || (sizeof(size_t)==8)); } - - -#if defined (__cplusplus) -} -#endif - -#endif /* MEM_H_MODULE */ diff --git a/dep/zstd/lib/common/pool.c b/dep/zstd/lib/common/pool.c deleted file mode 100644 index 2e37cdd73..000000000 --- a/dep/zstd/lib/common/pool.c +++ /dev/null @@ -1,355 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -/* ====== Dependencies ======= */ -#include "zstd_deps.h" /* size_t */ -#include "debug.h" /* assert */ -#include "zstd_internal.h" /* ZSTD_customMalloc, ZSTD_customFree */ -#include "pool.h" - -/* ====== Compiler specifics ====== */ -#if defined(_MSC_VER) -# pragma warning(disable : 4204) /* disable: C4204: non-constant aggregate initializer */ -#endif - - -#ifdef ZSTD_MULTITHREAD - -#include "threading.h" /* pthread adaptation */ - -/* A job is a function and an opaque argument */ -typedef struct POOL_job_s { - POOL_function function; - void *opaque; -} POOL_job; - -struct POOL_ctx_s { - ZSTD_customMem customMem; - /* Keep track of the threads */ - ZSTD_pthread_t* threads; - size_t threadCapacity; - size_t threadLimit; - - /* The queue is a circular buffer */ - POOL_job *queue; - size_t queueHead; - size_t queueTail; - size_t queueSize; - - /* The number of threads working on jobs */ - size_t numThreadsBusy; - /* Indicates if the queue is empty */ - int queueEmpty; - - /* The mutex protects the queue */ - ZSTD_pthread_mutex_t queueMutex; - /* Condition variable for pushers to wait on when the queue is full */ - ZSTD_pthread_cond_t queuePushCond; - /* Condition variables for poppers to wait on when the queue is empty */ - ZSTD_pthread_cond_t queuePopCond; - /* Indicates if the queue is shutting down */ - int shutdown; -}; - -/* POOL_thread() : - * Work thread for the thread pool. - * Waits for jobs and executes them. - * @returns : NULL on failure else non-null. - */ -static void* POOL_thread(void* opaque) { - POOL_ctx* const ctx = (POOL_ctx*)opaque; - if (!ctx) { return NULL; } - for (;;) { - /* Lock the mutex and wait for a non-empty queue or until shutdown */ - ZSTD_pthread_mutex_lock(&ctx->queueMutex); - - while ( ctx->queueEmpty - || (ctx->numThreadsBusy >= ctx->threadLimit) ) { - if (ctx->shutdown) { - /* even if !queueEmpty, (possible if numThreadsBusy >= threadLimit), - * a few threads will be shutdown while !queueEmpty, - * but enough threads will remain active to finish the queue */ - ZSTD_pthread_mutex_unlock(&ctx->queueMutex); - return opaque; - } - ZSTD_pthread_cond_wait(&ctx->queuePopCond, &ctx->queueMutex); - } - /* Pop a job off the queue */ - { POOL_job const job = ctx->queue[ctx->queueHead]; - ctx->queueHead = (ctx->queueHead + 1) % ctx->queueSize; - ctx->numThreadsBusy++; - ctx->queueEmpty = (ctx->queueHead == ctx->queueTail); - /* Unlock the mutex, signal a pusher, and run the job */ - ZSTD_pthread_cond_signal(&ctx->queuePushCond); - ZSTD_pthread_mutex_unlock(&ctx->queueMutex); - - job.function(job.opaque); - - /* If the intended queue size was 0, signal after finishing job */ - ZSTD_pthread_mutex_lock(&ctx->queueMutex); - ctx->numThreadsBusy--; - if (ctx->queueSize == 1) { - ZSTD_pthread_cond_signal(&ctx->queuePushCond); - } - ZSTD_pthread_mutex_unlock(&ctx->queueMutex); - } - } /* for (;;) */ - assert(0); /* Unreachable */ -} - -/* ZSTD_createThreadPool() : public access point */ -POOL_ctx* ZSTD_createThreadPool(size_t numThreads) { - return POOL_create (numThreads, 0); -} - -POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) { - return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem); -} - -POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize, - ZSTD_customMem customMem) -{ - POOL_ctx* ctx; - /* Check parameters */ - if (!numThreads) { return NULL; } - /* Allocate the context and zero initialize */ - ctx = (POOL_ctx*)ZSTD_customCalloc(sizeof(POOL_ctx), customMem); - if (!ctx) { return NULL; } - /* Initialize the job queue. - * It needs one extra space since one space is wasted to differentiate - * empty and full queues. - */ - ctx->queueSize = queueSize + 1; - ctx->queue = (POOL_job*)ZSTD_customMalloc(ctx->queueSize * sizeof(POOL_job), customMem); - ctx->queueHead = 0; - ctx->queueTail = 0; - ctx->numThreadsBusy = 0; - ctx->queueEmpty = 1; - { - int error = 0; - error |= ZSTD_pthread_mutex_init(&ctx->queueMutex, NULL); - error |= ZSTD_pthread_cond_init(&ctx->queuePushCond, NULL); - error |= ZSTD_pthread_cond_init(&ctx->queuePopCond, NULL); - if (error) { POOL_free(ctx); return NULL; } - } - ctx->shutdown = 0; - /* Allocate space for the thread handles */ - ctx->threads = (ZSTD_pthread_t*)ZSTD_customMalloc(numThreads * sizeof(ZSTD_pthread_t), customMem); - ctx->threadCapacity = 0; - ctx->customMem = customMem; - /* Check for errors */ - if (!ctx->threads || !ctx->queue) { POOL_free(ctx); return NULL; } - /* Initialize the threads */ - { size_t i; - for (i = 0; i < numThreads; ++i) { - if (ZSTD_pthread_create(&ctx->threads[i], NULL, &POOL_thread, ctx)) { - ctx->threadCapacity = i; - POOL_free(ctx); - return NULL; - } } - ctx->threadCapacity = numThreads; - ctx->threadLimit = numThreads; - } - return ctx; -} - -/*! POOL_join() : - Shutdown the queue, wake any sleeping threads, and join all of the threads. -*/ -static void POOL_join(POOL_ctx* ctx) { - /* Shut down the queue */ - ZSTD_pthread_mutex_lock(&ctx->queueMutex); - ctx->shutdown = 1; - ZSTD_pthread_mutex_unlock(&ctx->queueMutex); - /* Wake up sleeping threads */ - ZSTD_pthread_cond_broadcast(&ctx->queuePushCond); - ZSTD_pthread_cond_broadcast(&ctx->queuePopCond); - /* Join all of the threads */ - { size_t i; - for (i = 0; i < ctx->threadCapacity; ++i) { - ZSTD_pthread_join(ctx->threads[i], NULL); /* note : could fail */ - } } -} - -void POOL_free(POOL_ctx *ctx) { - if (!ctx) { return; } - POOL_join(ctx); - ZSTD_pthread_mutex_destroy(&ctx->queueMutex); - ZSTD_pthread_cond_destroy(&ctx->queuePushCond); - ZSTD_pthread_cond_destroy(&ctx->queuePopCond); - ZSTD_customFree(ctx->queue, ctx->customMem); - ZSTD_customFree(ctx->threads, ctx->customMem); - ZSTD_customFree(ctx, ctx->customMem); -} - -void ZSTD_freeThreadPool (ZSTD_threadPool* pool) { - POOL_free (pool); -} - -size_t POOL_sizeof(const POOL_ctx* ctx) { - if (ctx==NULL) return 0; /* supports sizeof NULL */ - return sizeof(*ctx) - + ctx->queueSize * sizeof(POOL_job) - + ctx->threadCapacity * sizeof(ZSTD_pthread_t); -} - - -/* @return : 0 on success, 1 on error */ -static int POOL_resize_internal(POOL_ctx* ctx, size_t numThreads) -{ - if (numThreads <= ctx->threadCapacity) { - if (!numThreads) return 1; - ctx->threadLimit = numThreads; - return 0; - } - /* numThreads > threadCapacity */ - { ZSTD_pthread_t* const threadPool = (ZSTD_pthread_t*)ZSTD_customMalloc(numThreads * sizeof(ZSTD_pthread_t), ctx->customMem); - if (!threadPool) return 1; - /* replace existing thread pool */ - ZSTD_memcpy(threadPool, ctx->threads, ctx->threadCapacity * sizeof(*threadPool)); - ZSTD_customFree(ctx->threads, ctx->customMem); - ctx->threads = threadPool; - /* Initialize additional threads */ - { size_t threadId; - for (threadId = ctx->threadCapacity; threadId < numThreads; ++threadId) { - if (ZSTD_pthread_create(&threadPool[threadId], NULL, &POOL_thread, ctx)) { - ctx->threadCapacity = threadId; - return 1; - } } - } } - /* successfully expanded */ - ctx->threadCapacity = numThreads; - ctx->threadLimit = numThreads; - return 0; -} - -/* @return : 0 on success, 1 on error */ -int POOL_resize(POOL_ctx* ctx, size_t numThreads) -{ - int result; - if (ctx==NULL) return 1; - ZSTD_pthread_mutex_lock(&ctx->queueMutex); - result = POOL_resize_internal(ctx, numThreads); - ZSTD_pthread_cond_broadcast(&ctx->queuePopCond); - ZSTD_pthread_mutex_unlock(&ctx->queueMutex); - return result; -} - -/** - * Returns 1 if the queue is full and 0 otherwise. - * - * When queueSize is 1 (pool was created with an intended queueSize of 0), - * then a queue is empty if there is a thread free _and_ no job is waiting. - */ -static int isQueueFull(POOL_ctx const* ctx) { - if (ctx->queueSize > 1) { - return ctx->queueHead == ((ctx->queueTail + 1) % ctx->queueSize); - } else { - return (ctx->numThreadsBusy == ctx->threadLimit) || - !ctx->queueEmpty; - } -} - - -static void -POOL_add_internal(POOL_ctx* ctx, POOL_function function, void *opaque) -{ - POOL_job const job = {function, opaque}; - assert(ctx != NULL); - if (ctx->shutdown) return; - - ctx->queueEmpty = 0; - ctx->queue[ctx->queueTail] = job; - ctx->queueTail = (ctx->queueTail + 1) % ctx->queueSize; - ZSTD_pthread_cond_signal(&ctx->queuePopCond); -} - -void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque) -{ - assert(ctx != NULL); - ZSTD_pthread_mutex_lock(&ctx->queueMutex); - /* Wait until there is space in the queue for the new job */ - while (isQueueFull(ctx) && (!ctx->shutdown)) { - ZSTD_pthread_cond_wait(&ctx->queuePushCond, &ctx->queueMutex); - } - POOL_add_internal(ctx, function, opaque); - ZSTD_pthread_mutex_unlock(&ctx->queueMutex); -} - - -int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque) -{ - assert(ctx != NULL); - ZSTD_pthread_mutex_lock(&ctx->queueMutex); - if (isQueueFull(ctx)) { - ZSTD_pthread_mutex_unlock(&ctx->queueMutex); - return 0; - } - POOL_add_internal(ctx, function, opaque); - ZSTD_pthread_mutex_unlock(&ctx->queueMutex); - return 1; -} - - -#else /* ZSTD_MULTITHREAD not defined */ - -/* ========================== */ -/* No multi-threading support */ -/* ========================== */ - - -/* We don't need any data, but if it is empty, malloc() might return NULL. */ -struct POOL_ctx_s { - int dummy; -}; -static POOL_ctx g_poolCtx; - -POOL_ctx* POOL_create(size_t numThreads, size_t queueSize) { - return POOL_create_advanced(numThreads, queueSize, ZSTD_defaultCMem); -} - -POOL_ctx* -POOL_create_advanced(size_t numThreads, size_t queueSize, ZSTD_customMem customMem) -{ - (void)numThreads; - (void)queueSize; - (void)customMem; - return &g_poolCtx; -} - -void POOL_free(POOL_ctx* ctx) { - assert(!ctx || ctx == &g_poolCtx); - (void)ctx; -} - -int POOL_resize(POOL_ctx* ctx, size_t numThreads) { - (void)ctx; (void)numThreads; - return 0; -} - -void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque) { - (void)ctx; - function(opaque); -} - -int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque) { - (void)ctx; - function(opaque); - return 1; -} - -size_t POOL_sizeof(const POOL_ctx* ctx) { - if (ctx==NULL) return 0; /* supports sizeof NULL */ - assert(ctx == &g_poolCtx); - return sizeof(*ctx); -} - -#endif /* ZSTD_MULTITHREAD */ diff --git a/dep/zstd/lib/common/pool.h b/dep/zstd/lib/common/pool.h deleted file mode 100644 index 0ebde1805..000000000 --- a/dep/zstd/lib/common/pool.h +++ /dev/null @@ -1,84 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef POOL_H -#define POOL_H - -#if defined (__cplusplus) -extern "C" { -#endif - - -#include "zstd_deps.h" -#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_customMem */ -#include "../zstd.h" - -typedef struct POOL_ctx_s POOL_ctx; - -/*! POOL_create() : - * Create a thread pool with at most `numThreads` threads. - * `numThreads` must be at least 1. - * The maximum number of queued jobs before blocking is `queueSize`. - * @return : POOL_ctx pointer on success, else NULL. -*/ -POOL_ctx* POOL_create(size_t numThreads, size_t queueSize); - -POOL_ctx* POOL_create_advanced(size_t numThreads, size_t queueSize, - ZSTD_customMem customMem); - -/*! POOL_free() : - * Free a thread pool returned by POOL_create(). - */ -void POOL_free(POOL_ctx* ctx); - -/*! POOL_resize() : - * Expands or shrinks pool's number of threads. - * This is more efficient than releasing + creating a new context, - * since it tries to preserve and re-use existing threads. - * `numThreads` must be at least 1. - * @return : 0 when resize was successful, - * !0 (typically 1) if there is an error. - * note : only numThreads can be resized, queueSize remains unchanged. - */ -int POOL_resize(POOL_ctx* ctx, size_t numThreads); - -/*! POOL_sizeof() : - * @return threadpool memory usage - * note : compatible with NULL (returns 0 in this case) - */ -size_t POOL_sizeof(const POOL_ctx* ctx); - -/*! POOL_function : - * The function type that can be added to a thread pool. - */ -typedef void (*POOL_function)(void*); - -/*! POOL_add() : - * Add the job `function(opaque)` to the thread pool. `ctx` must be valid. - * Possibly blocks until there is room in the queue. - * Note : The function may be executed asynchronously, - * therefore, `opaque` must live until function has been completed. - */ -void POOL_add(POOL_ctx* ctx, POOL_function function, void* opaque); - - -/*! POOL_tryAdd() : - * Add the job `function(opaque)` to thread pool _if_ a queue slot is available. - * Returns immediately even if not (does not block). - * @return : 1 if successful, 0 if not. - */ -int POOL_tryAdd(POOL_ctx* ctx, POOL_function function, void* opaque); - - -#if defined (__cplusplus) -} -#endif - -#endif diff --git a/dep/zstd/lib/common/portability_macros.h b/dep/zstd/lib/common/portability_macros.h deleted file mode 100644 index 2143817f5..000000000 --- a/dep/zstd/lib/common/portability_macros.h +++ /dev/null @@ -1,137 +0,0 @@ -/* - * Copyright (c) Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_PORTABILITY_MACROS_H -#define ZSTD_PORTABILITY_MACROS_H - -/** - * This header file contains macro defintions to support portability. - * This header is shared between C and ASM code, so it MUST only - * contain macro definitions. It MUST not contain any C code. - * - * This header ONLY defines macros to detect platforms/feature support. - * - */ - - -/* compat. with non-clang compilers */ -#ifndef __has_attribute - #define __has_attribute(x) 0 -#endif - -/* compat. with non-clang compilers */ -#ifndef __has_builtin -# define __has_builtin(x) 0 -#endif - -/* compat. with non-clang compilers */ -#ifndef __has_feature -# define __has_feature(x) 0 -#endif - -/* detects whether we are being compiled under msan */ -#ifndef ZSTD_MEMORY_SANITIZER -# if __has_feature(memory_sanitizer) -# define ZSTD_MEMORY_SANITIZER 1 -# else -# define ZSTD_MEMORY_SANITIZER 0 -# endif -#endif - -/* detects whether we are being compiled under asan */ -#ifndef ZSTD_ADDRESS_SANITIZER -# if __has_feature(address_sanitizer) -# define ZSTD_ADDRESS_SANITIZER 1 -# elif defined(__SANITIZE_ADDRESS__) -# define ZSTD_ADDRESS_SANITIZER 1 -# else -# define ZSTD_ADDRESS_SANITIZER 0 -# endif -#endif - -/* detects whether we are being compiled under dfsan */ -#ifndef ZSTD_DATAFLOW_SANITIZER -# if __has_feature(dataflow_sanitizer) -# define ZSTD_DATAFLOW_SANITIZER 1 -# else -# define ZSTD_DATAFLOW_SANITIZER 0 -# endif -#endif - -/* Mark the internal assembly functions as hidden */ -#ifdef __ELF__ -# define ZSTD_HIDE_ASM_FUNCTION(func) .hidden func -#else -# define ZSTD_HIDE_ASM_FUNCTION(func) -#endif - -/* Enable runtime BMI2 dispatch based on the CPU. - * Enabled for clang & gcc >=4.8 on x86 when BMI2 isn't enabled by default. - */ -#ifndef DYNAMIC_BMI2 - #if ((defined(__clang__) && __has_attribute(__target__)) \ - || (defined(__GNUC__) \ - && (__GNUC__ >= 5 || (__GNUC__ == 4 && __GNUC_MINOR__ >= 8)))) \ - && (defined(__x86_64__) || defined(_M_X64)) \ - && !defined(__BMI2__) - # define DYNAMIC_BMI2 1 - #else - # define DYNAMIC_BMI2 0 - #endif -#endif - -/** - * Only enable assembly for GNUC comptabile compilers, - * because other platforms may not support GAS assembly syntax. - * - * Only enable assembly for Linux / MacOS, other platforms may - * work, but they haven't been tested. This could likely be - * extended to BSD systems. - * - * Disable assembly when MSAN is enabled, because MSAN requires - * 100% of code to be instrumented to work. - */ -#if defined(__GNUC__) -# if defined(__linux__) || defined(__linux) || defined(__APPLE__) -# if ZSTD_MEMORY_SANITIZER -# define ZSTD_ASM_SUPPORTED 0 -# elif ZSTD_DATAFLOW_SANITIZER -# define ZSTD_ASM_SUPPORTED 0 -# else -# define ZSTD_ASM_SUPPORTED 1 -# endif -# else -# define ZSTD_ASM_SUPPORTED 0 -# endif -#else -# define ZSTD_ASM_SUPPORTED 0 -#endif - -/** - * Determines whether we should enable assembly for x86-64 - * with BMI2. - * - * Enable if all of the following conditions hold: - * - ASM hasn't been explicitly disabled by defining ZSTD_DISABLE_ASM - * - Assembly is supported - * - We are compiling for x86-64 and either: - * - DYNAMIC_BMI2 is enabled - * - BMI2 is supported at compile time - */ -#if !defined(ZSTD_DISABLE_ASM) && \ - ZSTD_ASM_SUPPORTED && \ - defined(__x86_64__) && \ - (DYNAMIC_BMI2 || defined(__BMI2__)) -# define ZSTD_ENABLE_ASM_X86_64_BMI2 1 -#else -# define ZSTD_ENABLE_ASM_X86_64_BMI2 0 -#endif - -#endif /* ZSTD_PORTABILITY_MACROS_H */ diff --git a/dep/zstd/lib/common/threading.c b/dep/zstd/lib/common/threading.c deleted file mode 100644 index 92cf57c19..000000000 --- a/dep/zstd/lib/common/threading.c +++ /dev/null @@ -1,122 +0,0 @@ -/** - * Copyright (c) 2016 Tino Reichardt - * All rights reserved. - * - * You can contact the author at: - * - zstdmt source repository: https://github.com/mcmilk/zstdmt - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -/** - * This file will hold wrapper for systems, which do not support pthreads - */ - -#include "threading.h" - -/* create fake symbol to avoid empty translation unit warning */ -int g_ZSTD_threading_useless_symbol; - -#if defined(ZSTD_MULTITHREAD) && defined(_WIN32) - -/** - * Windows minimalist Pthread Wrapper, based on : - * http://www.cse.wustl.edu/~schmidt/win32-cv-1.html - */ - - -/* === Dependencies === */ -#include -#include - - -/* === Implementation === */ - -static unsigned __stdcall worker(void *arg) -{ - ZSTD_pthread_t* const thread = (ZSTD_pthread_t*) arg; - thread->arg = thread->start_routine(thread->arg); - return 0; -} - -int ZSTD_pthread_create(ZSTD_pthread_t* thread, const void* unused, - void* (*start_routine) (void*), void* arg) -{ - (void)unused; - thread->arg = arg; - thread->start_routine = start_routine; - thread->handle = (HANDLE) _beginthreadex(NULL, 0, worker, thread, 0, NULL); - - if (!thread->handle) - return errno; - else - return 0; -} - -int ZSTD_pthread_join(ZSTD_pthread_t thread, void **value_ptr) -{ - DWORD result; - - if (!thread.handle) return 0; - - result = WaitForSingleObject(thread.handle, INFINITE); - switch (result) { - case WAIT_OBJECT_0: - if (value_ptr) *value_ptr = thread.arg; - return 0; - case WAIT_ABANDONED: - return EINVAL; - default: - return GetLastError(); - } -} - -#endif /* ZSTD_MULTITHREAD */ - -#if defined(ZSTD_MULTITHREAD) && DEBUGLEVEL >= 1 && !defined(_WIN32) - -#define ZSTD_DEPS_NEED_MALLOC -#include "zstd_deps.h" - -int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t const* attr) -{ - *mutex = (pthread_mutex_t*)ZSTD_malloc(sizeof(pthread_mutex_t)); - if (!*mutex) - return 1; - return pthread_mutex_init(*mutex, attr); -} - -int ZSTD_pthread_mutex_destroy(ZSTD_pthread_mutex_t* mutex) -{ - if (!*mutex) - return 0; - { - int const ret = pthread_mutex_destroy(*mutex); - ZSTD_free(*mutex); - return ret; - } -} - -int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const* attr) -{ - *cond = (pthread_cond_t*)ZSTD_malloc(sizeof(pthread_cond_t)); - if (!*cond) - return 1; - return pthread_cond_init(*cond, attr); -} - -int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond) -{ - if (!*cond) - return 0; - { - int const ret = pthread_cond_destroy(*cond); - ZSTD_free(*cond); - return ret; - } -} - -#endif diff --git a/dep/zstd/lib/common/threading.h b/dep/zstd/lib/common/threading.h deleted file mode 100644 index fd0060d5a..000000000 --- a/dep/zstd/lib/common/threading.h +++ /dev/null @@ -1,155 +0,0 @@ -/** - * Copyright (c) 2016 Tino Reichardt - * All rights reserved. - * - * You can contact the author at: - * - zstdmt source repository: https://github.com/mcmilk/zstdmt - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef THREADING_H_938743 -#define THREADING_H_938743 - -#include "debug.h" - -#if defined (__cplusplus) -extern "C" { -#endif - -#if defined(ZSTD_MULTITHREAD) && defined(_WIN32) - -/** - * Windows minimalist Pthread Wrapper, based on : - * http://www.cse.wustl.edu/~schmidt/win32-cv-1.html - */ -#ifdef WINVER -# undef WINVER -#endif -#define WINVER 0x0600 - -#ifdef _WIN32_WINNT -# undef _WIN32_WINNT -#endif -#define _WIN32_WINNT 0x0600 - -#ifndef WIN32_LEAN_AND_MEAN -# define WIN32_LEAN_AND_MEAN -#endif - -#undef ERROR /* reported already defined on VS 2015 (Rich Geldreich) */ -#include -#undef ERROR -#define ERROR(name) ZSTD_ERROR(name) - - -/* mutex */ -#define ZSTD_pthread_mutex_t CRITICAL_SECTION -#define ZSTD_pthread_mutex_init(a, b) ((void)(b), InitializeCriticalSection((a)), 0) -#define ZSTD_pthread_mutex_destroy(a) DeleteCriticalSection((a)) -#define ZSTD_pthread_mutex_lock(a) EnterCriticalSection((a)) -#define ZSTD_pthread_mutex_unlock(a) LeaveCriticalSection((a)) - -/* condition variable */ -#define ZSTD_pthread_cond_t CONDITION_VARIABLE -#define ZSTD_pthread_cond_init(a, b) ((void)(b), InitializeConditionVariable((a)), 0) -#define ZSTD_pthread_cond_destroy(a) ((void)(a)) -#define ZSTD_pthread_cond_wait(a, b) SleepConditionVariableCS((a), (b), INFINITE) -#define ZSTD_pthread_cond_signal(a) WakeConditionVariable((a)) -#define ZSTD_pthread_cond_broadcast(a) WakeAllConditionVariable((a)) - -/* ZSTD_pthread_create() and ZSTD_pthread_join() */ -typedef struct { - HANDLE handle; - void* (*start_routine)(void*); - void* arg; -} ZSTD_pthread_t; - -int ZSTD_pthread_create(ZSTD_pthread_t* thread, const void* unused, - void* (*start_routine) (void*), void* arg); - -int ZSTD_pthread_join(ZSTD_pthread_t thread, void** value_ptr); - -/** - * add here more wrappers as required - */ - - -#elif defined(ZSTD_MULTITHREAD) /* posix assumed ; need a better detection method */ -/* === POSIX Systems === */ -# include - -#if DEBUGLEVEL < 1 - -#define ZSTD_pthread_mutex_t pthread_mutex_t -#define ZSTD_pthread_mutex_init(a, b) pthread_mutex_init((a), (b)) -#define ZSTD_pthread_mutex_destroy(a) pthread_mutex_destroy((a)) -#define ZSTD_pthread_mutex_lock(a) pthread_mutex_lock((a)) -#define ZSTD_pthread_mutex_unlock(a) pthread_mutex_unlock((a)) - -#define ZSTD_pthread_cond_t pthread_cond_t -#define ZSTD_pthread_cond_init(a, b) pthread_cond_init((a), (b)) -#define ZSTD_pthread_cond_destroy(a) pthread_cond_destroy((a)) -#define ZSTD_pthread_cond_wait(a, b) pthread_cond_wait((a), (b)) -#define ZSTD_pthread_cond_signal(a) pthread_cond_signal((a)) -#define ZSTD_pthread_cond_broadcast(a) pthread_cond_broadcast((a)) - -#define ZSTD_pthread_t pthread_t -#define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d)) -#define ZSTD_pthread_join(a, b) pthread_join((a),(b)) - -#else /* DEBUGLEVEL >= 1 */ - -/* Debug implementation of threading. - * In this implementation we use pointers for mutexes and condition variables. - * This way, if we forget to init/destroy them the program will crash or ASAN - * will report leaks. - */ - -#define ZSTD_pthread_mutex_t pthread_mutex_t* -int ZSTD_pthread_mutex_init(ZSTD_pthread_mutex_t* mutex, pthread_mutexattr_t const* attr); -int ZSTD_pthread_mutex_destroy(ZSTD_pthread_mutex_t* mutex); -#define ZSTD_pthread_mutex_lock(a) pthread_mutex_lock(*(a)) -#define ZSTD_pthread_mutex_unlock(a) pthread_mutex_unlock(*(a)) - -#define ZSTD_pthread_cond_t pthread_cond_t* -int ZSTD_pthread_cond_init(ZSTD_pthread_cond_t* cond, pthread_condattr_t const* attr); -int ZSTD_pthread_cond_destroy(ZSTD_pthread_cond_t* cond); -#define ZSTD_pthread_cond_wait(a, b) pthread_cond_wait(*(a), *(b)) -#define ZSTD_pthread_cond_signal(a) pthread_cond_signal(*(a)) -#define ZSTD_pthread_cond_broadcast(a) pthread_cond_broadcast(*(a)) - -#define ZSTD_pthread_t pthread_t -#define ZSTD_pthread_create(a, b, c, d) pthread_create((a), (b), (c), (d)) -#define ZSTD_pthread_join(a, b) pthread_join((a),(b)) - -#endif - -#else /* ZSTD_MULTITHREAD not defined */ -/* No multithreading support */ - -typedef int ZSTD_pthread_mutex_t; -#define ZSTD_pthread_mutex_init(a, b) ((void)(a), (void)(b), 0) -#define ZSTD_pthread_mutex_destroy(a) ((void)(a)) -#define ZSTD_pthread_mutex_lock(a) ((void)(a)) -#define ZSTD_pthread_mutex_unlock(a) ((void)(a)) - -typedef int ZSTD_pthread_cond_t; -#define ZSTD_pthread_cond_init(a, b) ((void)(a), (void)(b), 0) -#define ZSTD_pthread_cond_destroy(a) ((void)(a)) -#define ZSTD_pthread_cond_wait(a, b) ((void)(a), (void)(b)) -#define ZSTD_pthread_cond_signal(a) ((void)(a)) -#define ZSTD_pthread_cond_broadcast(a) ((void)(a)) - -/* do not use ZSTD_pthread_t */ - -#endif /* ZSTD_MULTITHREAD */ - -#if defined (__cplusplus) -} -#endif - -#endif /* THREADING_H_938743 */ diff --git a/dep/zstd/lib/common/xxhash.c b/dep/zstd/lib/common/xxhash.c deleted file mode 100644 index d49497cf1..000000000 --- a/dep/zstd/lib/common/xxhash.c +++ /dev/null @@ -1,24 +0,0 @@ -/* - * xxHash - Fast Hash algorithm - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - xxHash homepage: http://www.xxhash.com - * - xxHash source repository : https://github.com/Cyan4973/xxHash - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -*/ - - - -/* - * xxhash.c instantiates functions defined in xxhash.h - */ - -#define XXH_STATIC_LINKING_ONLY /* access advanced declarations */ -#define XXH_IMPLEMENTATION /* access definitions */ - -#include "xxhash.h" diff --git a/dep/zstd/lib/common/xxhash.h b/dep/zstd/lib/common/xxhash.h deleted file mode 100644 index 8ebbfdd62..000000000 --- a/dep/zstd/lib/common/xxhash.h +++ /dev/null @@ -1,5686 +0,0 @@ -/* - * xxHash - Fast Hash algorithm - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - xxHash homepage: http://www.xxhash.com - * - xxHash source repository : https://github.com/Cyan4973/xxHash - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -*/ - - -#ifndef XXH_NO_XXH3 -# define XXH_NO_XXH3 -#endif - -#ifndef XXH_NAMESPACE -# define XXH_NAMESPACE ZSTD_ -#endif - -/*! - * @mainpage xxHash - * - * @file xxhash.h - * xxHash prototypes and implementation - */ -/* TODO: update */ -/* Notice extracted from xxHash homepage: - -xxHash is an extremely fast hash algorithm, running at RAM speed limits. -It also successfully passes all tests from the SMHasher suite. - -Comparison (single thread, Windows Seven 32 bits, using SMHasher on a Core 2 Duo @3GHz) - -Name Speed Q.Score Author -xxHash 5.4 GB/s 10 -CrapWow 3.2 GB/s 2 Andrew -MurmurHash 3a 2.7 GB/s 10 Austin Appleby -SpookyHash 2.0 GB/s 10 Bob Jenkins -SBox 1.4 GB/s 9 Bret Mulvey -Lookup3 1.2 GB/s 9 Bob Jenkins -SuperFastHash 1.2 GB/s 1 Paul Hsieh -CityHash64 1.05 GB/s 10 Pike & Alakuijala -FNV 0.55 GB/s 5 Fowler, Noll, Vo -CRC32 0.43 GB/s 9 -MD5-32 0.33 GB/s 10 Ronald L. Rivest -SHA1-32 0.28 GB/s 10 - -Q.Score is a measure of quality of the hash function. -It depends on successfully passing SMHasher test set. -10 is a perfect score. - -Note: SMHasher's CRC32 implementation is not the fastest one. -Other speed-oriented implementations can be faster, -especially in combination with PCLMUL instruction: -https://fastcompression.blogspot.com/2019/03/presenting-xxh3.html?showComment=1552696407071#c3490092340461170735 - -A 64-bit version, named XXH64, is available since r35. -It offers much better speed, but for 64-bit applications only. -Name Speed on 64 bits Speed on 32 bits -XXH64 13.8 GB/s 1.9 GB/s -XXH32 6.8 GB/s 6.0 GB/s -*/ - -#if defined (__cplusplus) -extern "C" { -#endif - -/* **************************** - * INLINE mode - ******************************/ -/*! - * XXH_INLINE_ALL (and XXH_PRIVATE_API) - * Use these build macros to inline xxhash into the target unit. - * Inlining improves performance on small inputs, especially when the length is - * expressed as a compile-time constant: - * - * https://fastcompression.blogspot.com/2018/03/xxhash-for-small-keys-impressive-power.html - * - * It also keeps xxHash symbols private to the unit, so they are not exported. - * - * Usage: - * #define XXH_INLINE_ALL - * #include "xxhash.h" - * - * Do not compile and link xxhash.o as a separate object, as it is not useful. - */ -#if (defined(XXH_INLINE_ALL) || defined(XXH_PRIVATE_API)) \ - && !defined(XXH_INLINE_ALL_31684351384) - /* this section should be traversed only once */ -# define XXH_INLINE_ALL_31684351384 - /* give access to the advanced API, required to compile implementations */ -# undef XXH_STATIC_LINKING_ONLY /* avoid macro redef */ -# define XXH_STATIC_LINKING_ONLY - /* make all functions private */ -# undef XXH_PUBLIC_API -# if defined(__GNUC__) -# define XXH_PUBLIC_API static __inline __attribute__((unused)) -# elif defined (__cplusplus) || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) -# define XXH_PUBLIC_API static inline -# elif defined(_MSC_VER) -# define XXH_PUBLIC_API static __inline -# else - /* note: this version may generate warnings for unused static functions */ -# define XXH_PUBLIC_API static -# endif - - /* - * This part deals with the special case where a unit wants to inline xxHash, - * but "xxhash.h" has previously been included without XXH_INLINE_ALL, - * such as part of some previously included *.h header file. - * Without further action, the new include would just be ignored, - * and functions would effectively _not_ be inlined (silent failure). - * The following macros solve this situation by prefixing all inlined names, - * avoiding naming collision with previous inclusions. - */ - /* Before that, we unconditionally #undef all symbols, - * in case they were already defined with XXH_NAMESPACE. - * They will then be redefined for XXH_INLINE_ALL - */ -# undef XXH_versionNumber - /* XXH32 */ -# undef XXH32 -# undef XXH32_createState -# undef XXH32_freeState -# undef XXH32_reset -# undef XXH32_update -# undef XXH32_digest -# undef XXH32_copyState -# undef XXH32_canonicalFromHash -# undef XXH32_hashFromCanonical - /* XXH64 */ -# undef XXH64 -# undef XXH64_createState -# undef XXH64_freeState -# undef XXH64_reset -# undef XXH64_update -# undef XXH64_digest -# undef XXH64_copyState -# undef XXH64_canonicalFromHash -# undef XXH64_hashFromCanonical - /* XXH3_64bits */ -# undef XXH3_64bits -# undef XXH3_64bits_withSecret -# undef XXH3_64bits_withSeed -# undef XXH3_64bits_withSecretandSeed -# undef XXH3_createState -# undef XXH3_freeState -# undef XXH3_copyState -# undef XXH3_64bits_reset -# undef XXH3_64bits_reset_withSeed -# undef XXH3_64bits_reset_withSecret -# undef XXH3_64bits_update -# undef XXH3_64bits_digest -# undef XXH3_generateSecret - /* XXH3_128bits */ -# undef XXH128 -# undef XXH3_128bits -# undef XXH3_128bits_withSeed -# undef XXH3_128bits_withSecret -# undef XXH3_128bits_reset -# undef XXH3_128bits_reset_withSeed -# undef XXH3_128bits_reset_withSecret -# undef XXH3_128bits_reset_withSecretandSeed -# undef XXH3_128bits_update -# undef XXH3_128bits_digest -# undef XXH128_isEqual -# undef XXH128_cmp -# undef XXH128_canonicalFromHash -# undef XXH128_hashFromCanonical - /* Finally, free the namespace itself */ -# undef XXH_NAMESPACE - - /* employ the namespace for XXH_INLINE_ALL */ -# define XXH_NAMESPACE XXH_INLINE_ - /* - * Some identifiers (enums, type names) are not symbols, - * but they must nonetheless be renamed to avoid redeclaration. - * Alternative solution: do not redeclare them. - * However, this requires some #ifdefs, and has a more dispersed impact. - * Meanwhile, renaming can be achieved in a single place. - */ -# define XXH_IPREF(Id) XXH_NAMESPACE ## Id -# define XXH_OK XXH_IPREF(XXH_OK) -# define XXH_ERROR XXH_IPREF(XXH_ERROR) -# define XXH_errorcode XXH_IPREF(XXH_errorcode) -# define XXH32_canonical_t XXH_IPREF(XXH32_canonical_t) -# define XXH64_canonical_t XXH_IPREF(XXH64_canonical_t) -# define XXH128_canonical_t XXH_IPREF(XXH128_canonical_t) -# define XXH32_state_s XXH_IPREF(XXH32_state_s) -# define XXH32_state_t XXH_IPREF(XXH32_state_t) -# define XXH64_state_s XXH_IPREF(XXH64_state_s) -# define XXH64_state_t XXH_IPREF(XXH64_state_t) -# define XXH3_state_s XXH_IPREF(XXH3_state_s) -# define XXH3_state_t XXH_IPREF(XXH3_state_t) -# define XXH128_hash_t XXH_IPREF(XXH128_hash_t) - /* Ensure the header is parsed again, even if it was previously included */ -# undef XXHASH_H_5627135585666179 -# undef XXHASH_H_STATIC_13879238742 -#endif /* XXH_INLINE_ALL || XXH_PRIVATE_API */ - - - -/* **************************************************************** - * Stable API - *****************************************************************/ -#ifndef XXHASH_H_5627135585666179 -#define XXHASH_H_5627135585666179 1 - - -/*! - * @defgroup public Public API - * Contains details on the public xxHash functions. - * @{ - */ -/* specific declaration modes for Windows */ -#if !defined(XXH_INLINE_ALL) && !defined(XXH_PRIVATE_API) -# if defined(WIN32) && defined(_MSC_VER) && (defined(XXH_IMPORT) || defined(XXH_EXPORT)) -# ifdef XXH_EXPORT -# define XXH_PUBLIC_API __declspec(dllexport) -# elif XXH_IMPORT -# define XXH_PUBLIC_API __declspec(dllimport) -# endif -# else -# define XXH_PUBLIC_API /* do nothing */ -# endif -#endif - -#ifdef XXH_DOXYGEN -/*! - * @brief Emulate a namespace by transparently prefixing all symbols. - * - * If you want to include _and expose_ xxHash functions from within your own - * library, but also want to avoid symbol collisions with other libraries which - * may also include xxHash, you can use XXH_NAMESPACE to automatically prefix - * any public symbol from xxhash library with the value of XXH_NAMESPACE - * (therefore, avoid empty or numeric values). - * - * Note that no change is required within the calling program as long as it - * includes `xxhash.h`: Regular symbol names will be automatically translated - * by this header. - */ -# define XXH_NAMESPACE /* YOUR NAME HERE */ -# undef XXH_NAMESPACE -#endif - -#ifdef XXH_NAMESPACE -# define XXH_CAT(A,B) A##B -# define XXH_NAME2(A,B) XXH_CAT(A,B) -# define XXH_versionNumber XXH_NAME2(XXH_NAMESPACE, XXH_versionNumber) -/* XXH32 */ -# define XXH32 XXH_NAME2(XXH_NAMESPACE, XXH32) -# define XXH32_createState XXH_NAME2(XXH_NAMESPACE, XXH32_createState) -# define XXH32_freeState XXH_NAME2(XXH_NAMESPACE, XXH32_freeState) -# define XXH32_reset XXH_NAME2(XXH_NAMESPACE, XXH32_reset) -# define XXH32_update XXH_NAME2(XXH_NAMESPACE, XXH32_update) -# define XXH32_digest XXH_NAME2(XXH_NAMESPACE, XXH32_digest) -# define XXH32_copyState XXH_NAME2(XXH_NAMESPACE, XXH32_copyState) -# define XXH32_canonicalFromHash XXH_NAME2(XXH_NAMESPACE, XXH32_canonicalFromHash) -# define XXH32_hashFromCanonical XXH_NAME2(XXH_NAMESPACE, XXH32_hashFromCanonical) -/* XXH64 */ -# define XXH64 XXH_NAME2(XXH_NAMESPACE, XXH64) -# define XXH64_createState XXH_NAME2(XXH_NAMESPACE, XXH64_createState) -# define XXH64_freeState XXH_NAME2(XXH_NAMESPACE, XXH64_freeState) -# define XXH64_reset XXH_NAME2(XXH_NAMESPACE, XXH64_reset) -# define XXH64_update XXH_NAME2(XXH_NAMESPACE, XXH64_update) -# define XXH64_digest XXH_NAME2(XXH_NAMESPACE, XXH64_digest) -# define XXH64_copyState XXH_NAME2(XXH_NAMESPACE, XXH64_copyState) -# define XXH64_canonicalFromHash XXH_NAME2(XXH_NAMESPACE, XXH64_canonicalFromHash) -# define XXH64_hashFromCanonical XXH_NAME2(XXH_NAMESPACE, XXH64_hashFromCanonical) -/* XXH3_64bits */ -# define XXH3_64bits XXH_NAME2(XXH_NAMESPACE, XXH3_64bits) -# define XXH3_64bits_withSecret XXH_NAME2(XXH_NAMESPACE, XXH3_64bits_withSecret) -# define XXH3_64bits_withSeed XXH_NAME2(XXH_NAMESPACE, XXH3_64bits_withSeed) -# define XXH3_64bits_withSecretandSeed XXH_NAME2(XXH_NAMESPACE, XXH3_64bits_withSecretandSeed) -# define XXH3_createState XXH_NAME2(XXH_NAMESPACE, XXH3_createState) -# define XXH3_freeState XXH_NAME2(XXH_NAMESPACE, XXH3_freeState) -# define XXH3_copyState XXH_NAME2(XXH_NAMESPACE, XXH3_copyState) -# define XXH3_64bits_reset XXH_NAME2(XXH_NAMESPACE, XXH3_64bits_reset) -# define XXH3_64bits_reset_withSeed XXH_NAME2(XXH_NAMESPACE, XXH3_64bits_reset_withSeed) -# define XXH3_64bits_reset_withSecret XXH_NAME2(XXH_NAMESPACE, XXH3_64bits_reset_withSecret) -# define XXH3_64bits_reset_withSecretandSeed XXH_NAME2(XXH_NAMESPACE, XXH3_64bits_reset_withSecretandSeed) -# define XXH3_64bits_update XXH_NAME2(XXH_NAMESPACE, XXH3_64bits_update) -# define XXH3_64bits_digest XXH_NAME2(XXH_NAMESPACE, XXH3_64bits_digest) -# define XXH3_generateSecret XXH_NAME2(XXH_NAMESPACE, XXH3_generateSecret) -# define XXH3_generateSecret_fromSeed XXH_NAME2(XXH_NAMESPACE, XXH3_generateSecret_fromSeed) -/* XXH3_128bits */ -# define XXH128 XXH_NAME2(XXH_NAMESPACE, XXH128) -# define XXH3_128bits XXH_NAME2(XXH_NAMESPACE, XXH3_128bits) -# define XXH3_128bits_withSeed XXH_NAME2(XXH_NAMESPACE, XXH3_128bits_withSeed) -# define XXH3_128bits_withSecret XXH_NAME2(XXH_NAMESPACE, XXH3_128bits_withSecret) -# define XXH3_128bits_withSecretandSeed XXH_NAME2(XXH_NAMESPACE, XXH3_128bits_withSecretandSeed) -# define XXH3_128bits_reset XXH_NAME2(XXH_NAMESPACE, XXH3_128bits_reset) -# define XXH3_128bits_reset_withSeed XXH_NAME2(XXH_NAMESPACE, XXH3_128bits_reset_withSeed) -# define XXH3_128bits_reset_withSecret XXH_NAME2(XXH_NAMESPACE, XXH3_128bits_reset_withSecret) -# define XXH3_128bits_reset_withSecretandSeed XXH_NAME2(XXH_NAMESPACE, XXH3_128bits_reset_withSecretandSeed) -# define XXH3_128bits_update XXH_NAME2(XXH_NAMESPACE, XXH3_128bits_update) -# define XXH3_128bits_digest XXH_NAME2(XXH_NAMESPACE, XXH3_128bits_digest) -# define XXH128_isEqual XXH_NAME2(XXH_NAMESPACE, XXH128_isEqual) -# define XXH128_cmp XXH_NAME2(XXH_NAMESPACE, XXH128_cmp) -# define XXH128_canonicalFromHash XXH_NAME2(XXH_NAMESPACE, XXH128_canonicalFromHash) -# define XXH128_hashFromCanonical XXH_NAME2(XXH_NAMESPACE, XXH128_hashFromCanonical) -#endif - - -/* ************************************* -* Version -***************************************/ -#define XXH_VERSION_MAJOR 0 -#define XXH_VERSION_MINOR 8 -#define XXH_VERSION_RELEASE 1 -#define XXH_VERSION_NUMBER (XXH_VERSION_MAJOR *100*100 + XXH_VERSION_MINOR *100 + XXH_VERSION_RELEASE) - -/*! - * @brief Obtains the xxHash version. - * - * This is mostly useful when xxHash is compiled as a shared library, - * since the returned value comes from the library, as opposed to header file. - * - * @return `XXH_VERSION_NUMBER` of the invoked library. - */ -XXH_PUBLIC_API unsigned XXH_versionNumber (void); - - -/* **************************** -* Common basic types -******************************/ -#include /* size_t */ -typedef enum { XXH_OK=0, XXH_ERROR } XXH_errorcode; - - -/*-********************************************************************** -* 32-bit hash -************************************************************************/ -#if defined(XXH_DOXYGEN) /* Don't show include */ -/*! - * @brief An unsigned 32-bit integer. - * - * Not necessarily defined to `uint32_t` but functionally equivalent. - */ -typedef uint32_t XXH32_hash_t; - -#elif !defined (__VMS) \ - && (defined (__cplusplus) \ - || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) ) -# include - typedef uint32_t XXH32_hash_t; - -#else -# include -# if UINT_MAX == 0xFFFFFFFFUL - typedef unsigned int XXH32_hash_t; -# else -# if ULONG_MAX == 0xFFFFFFFFUL - typedef unsigned long XXH32_hash_t; -# else -# error "unsupported platform: need a 32-bit type" -# endif -# endif -#endif - -/*! - * @} - * - * @defgroup xxh32_family XXH32 family - * @ingroup public - * Contains functions used in the classic 32-bit xxHash algorithm. - * - * @note - * XXH32 is useful for older platforms, with no or poor 64-bit performance. - * Note that @ref xxh3_family provides competitive speed - * for both 32-bit and 64-bit systems, and offers true 64/128 bit hash results. - * - * @see @ref xxh64_family, @ref xxh3_family : Other xxHash families - * @see @ref xxh32_impl for implementation details - * @{ - */ - -/*! - * @brief Calculates the 32-bit hash of @p input using xxHash32. - * - * Speed on Core 2 Duo @ 3 GHz (single thread, SMHasher benchmark): 5.4 GB/s - * - * @param input The block of data to be hashed, at least @p length bytes in size. - * @param length The length of @p input, in bytes. - * @param seed The 32-bit seed to alter the hash's output predictably. - * - * @pre - * The memory between @p input and @p input + @p length must be valid, - * readable, contiguous memory. However, if @p length is `0`, @p input may be - * `NULL`. In C++, this also must be *TriviallyCopyable*. - * - * @return The calculated 32-bit hash value. - * - * @see - * XXH64(), XXH3_64bits_withSeed(), XXH3_128bits_withSeed(), XXH128(): - * Direct equivalents for the other variants of xxHash. - * @see - * XXH32_createState(), XXH32_update(), XXH32_digest(): Streaming version. - */ -XXH_PUBLIC_API XXH32_hash_t XXH32 (const void* input, size_t length, XXH32_hash_t seed); - -/*! - * Streaming functions generate the xxHash value from an incremental input. - * This method is slower than single-call functions, due to state management. - * For small inputs, prefer `XXH32()` and `XXH64()`, which are better optimized. - * - * An XXH state must first be allocated using `XXH*_createState()`. - * - * Start a new hash by initializing the state with a seed using `XXH*_reset()`. - * - * Then, feed the hash state by calling `XXH*_update()` as many times as necessary. - * - * The function returns an error code, with 0 meaning OK, and any other value - * meaning there is an error. - * - * Finally, a hash value can be produced anytime, by using `XXH*_digest()`. - * This function returns the nn-bits hash as an int or long long. - * - * It's still possible to continue inserting input into the hash state after a - * digest, and generate new hash values later on by invoking `XXH*_digest()`. - * - * When done, release the state using `XXH*_freeState()`. - * - * Example code for incrementally hashing a file: - * @code{.c} - * #include - * #include - * #define BUFFER_SIZE 256 - * - * // Note: XXH64 and XXH3 use the same interface. - * XXH32_hash_t - * hashFile(FILE* stream) - * { - * XXH32_state_t* state; - * unsigned char buf[BUFFER_SIZE]; - * size_t amt; - * XXH32_hash_t hash; - * - * state = XXH32_createState(); // Create a state - * assert(state != NULL); // Error check here - * XXH32_reset(state, 0xbaad5eed); // Reset state with our seed - * while ((amt = fread(buf, 1, sizeof(buf), stream)) != 0) { - * XXH32_update(state, buf, amt); // Hash the file in chunks - * } - * hash = XXH32_digest(state); // Finalize the hash - * XXH32_freeState(state); // Clean up - * return hash; - * } - * @endcode - */ - -/*! - * @typedef struct XXH32_state_s XXH32_state_t - * @brief The opaque state struct for the XXH32 streaming API. - * - * @see XXH32_state_s for details. - */ -typedef struct XXH32_state_s XXH32_state_t; - -/*! - * @brief Allocates an @ref XXH32_state_t. - * - * Must be freed with XXH32_freeState(). - * @return An allocated XXH32_state_t on success, `NULL` on failure. - */ -XXH_PUBLIC_API XXH32_state_t* XXH32_createState(void); -/*! - * @brief Frees an @ref XXH32_state_t. - * - * Must be allocated with XXH32_createState(). - * @param statePtr A pointer to an @ref XXH32_state_t allocated with @ref XXH32_createState(). - * @return XXH_OK. - */ -XXH_PUBLIC_API XXH_errorcode XXH32_freeState(XXH32_state_t* statePtr); -/*! - * @brief Copies one @ref XXH32_state_t to another. - * - * @param dst_state The state to copy to. - * @param src_state The state to copy from. - * @pre - * @p dst_state and @p src_state must not be `NULL` and must not overlap. - */ -XXH_PUBLIC_API void XXH32_copyState(XXH32_state_t* dst_state, const XXH32_state_t* src_state); - -/*! - * @brief Resets an @ref XXH32_state_t to begin a new hash. - * - * This function resets and seeds a state. Call it before @ref XXH32_update(). - * - * @param statePtr The state struct to reset. - * @param seed The 32-bit seed to alter the hash result predictably. - * - * @pre - * @p statePtr must not be `NULL`. - * - * @return @ref XXH_OK on success, @ref XXH_ERROR on failure. - */ -XXH_PUBLIC_API XXH_errorcode XXH32_reset (XXH32_state_t* statePtr, XXH32_hash_t seed); - -/*! - * @brief Consumes a block of @p input to an @ref XXH32_state_t. - * - * Call this to incrementally consume blocks of data. - * - * @param statePtr The state struct to update. - * @param input The block of data to be hashed, at least @p length bytes in size. - * @param length The length of @p input, in bytes. - * - * @pre - * @p statePtr must not be `NULL`. - * @pre - * The memory between @p input and @p input + @p length must be valid, - * readable, contiguous memory. However, if @p length is `0`, @p input may be - * `NULL`. In C++, this also must be *TriviallyCopyable*. - * - * @return @ref XXH_OK on success, @ref XXH_ERROR on failure. - */ -XXH_PUBLIC_API XXH_errorcode XXH32_update (XXH32_state_t* statePtr, const void* input, size_t length); - -/*! - * @brief Returns the calculated hash value from an @ref XXH32_state_t. - * - * @note - * Calling XXH32_digest() will not affect @p statePtr, so you can update, - * digest, and update again. - * - * @param statePtr The state struct to calculate the hash from. - * - * @pre - * @p statePtr must not be `NULL`. - * - * @return The calculated xxHash32 value from that state. - */ -XXH_PUBLIC_API XXH32_hash_t XXH32_digest (const XXH32_state_t* statePtr); - -/******* Canonical representation *******/ - -/* - * The default return values from XXH functions are unsigned 32 and 64 bit - * integers. - * This the simplest and fastest format for further post-processing. - * - * However, this leaves open the question of what is the order on the byte level, - * since little and big endian conventions will store the same number differently. - * - * The canonical representation settles this issue by mandating big-endian - * convention, the same convention as human-readable numbers (large digits first). - * - * When writing hash values to storage, sending them over a network, or printing - * them, it's highly recommended to use the canonical representation to ensure - * portability across a wider range of systems, present and future. - * - * The following functions allow transformation of hash values to and from - * canonical format. - */ - -/*! - * @brief Canonical (big endian) representation of @ref XXH32_hash_t. - */ -typedef struct { - unsigned char digest[4]; /*!< Hash bytes, big endian */ -} XXH32_canonical_t; - -/*! - * @brief Converts an @ref XXH32_hash_t to a big endian @ref XXH32_canonical_t. - * - * @param dst The @ref XXH32_canonical_t pointer to be stored to. - * @param hash The @ref XXH32_hash_t to be converted. - * - * @pre - * @p dst must not be `NULL`. - */ -XXH_PUBLIC_API void XXH32_canonicalFromHash(XXH32_canonical_t* dst, XXH32_hash_t hash); - -/*! - * @brief Converts an @ref XXH32_canonical_t to a native @ref XXH32_hash_t. - * - * @param src The @ref XXH32_canonical_t to convert. - * - * @pre - * @p src must not be `NULL`. - * - * @return The converted hash. - */ -XXH_PUBLIC_API XXH32_hash_t XXH32_hashFromCanonical(const XXH32_canonical_t* src); - - -#ifdef __has_attribute -# define XXH_HAS_ATTRIBUTE(x) __has_attribute(x) -#else -# define XXH_HAS_ATTRIBUTE(x) 0 -#endif - -/* C-language Attributes are added in C23. */ -#if defined(__STDC_VERSION__) && (__STDC_VERSION__ > 201710L) && defined(__has_c_attribute) -# define XXH_HAS_C_ATTRIBUTE(x) __has_c_attribute(x) -#else -# define XXH_HAS_C_ATTRIBUTE(x) 0 -#endif - -#if defined(__cplusplus) && defined(__has_cpp_attribute) -# define XXH_HAS_CPP_ATTRIBUTE(x) __has_cpp_attribute(x) -#else -# define XXH_HAS_CPP_ATTRIBUTE(x) 0 -#endif - -/* -Define XXH_FALLTHROUGH macro for annotating switch case with the 'fallthrough' attribute -introduced in CPP17 and C23. -CPP17 : https://en.cppreference.com/w/cpp/language/attributes/fallthrough -C23 : https://en.cppreference.com/w/c/language/attributes/fallthrough -*/ -#if XXH_HAS_C_ATTRIBUTE(x) -# define XXH_FALLTHROUGH [[fallthrough]] -#elif XXH_HAS_CPP_ATTRIBUTE(x) -# define XXH_FALLTHROUGH [[fallthrough]] -#elif XXH_HAS_ATTRIBUTE(__fallthrough__) -# define XXH_FALLTHROUGH __attribute__ ((fallthrough)) -#else -# define XXH_FALLTHROUGH -#endif - -/*! - * @} - * @ingroup public - * @{ - */ - -#ifndef XXH_NO_LONG_LONG -/*-********************************************************************** -* 64-bit hash -************************************************************************/ -#if defined(XXH_DOXYGEN) /* don't include */ -/*! - * @brief An unsigned 64-bit integer. - * - * Not necessarily defined to `uint64_t` but functionally equivalent. - */ -typedef uint64_t XXH64_hash_t; -#elif !defined (__VMS) \ - && (defined (__cplusplus) \ - || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) ) -# include - typedef uint64_t XXH64_hash_t; -#else -# include -# if defined(__LP64__) && ULONG_MAX == 0xFFFFFFFFFFFFFFFFULL - /* LP64 ABI says uint64_t is unsigned long */ - typedef unsigned long XXH64_hash_t; -# else - /* the following type must have a width of 64-bit */ - typedef unsigned long long XXH64_hash_t; -# endif -#endif - -/*! - * @} - * - * @defgroup xxh64_family XXH64 family - * @ingroup public - * @{ - * Contains functions used in the classic 64-bit xxHash algorithm. - * - * @note - * XXH3 provides competitive speed for both 32-bit and 64-bit systems, - * and offers true 64/128 bit hash results. - * It provides better speed for systems with vector processing capabilities. - */ - - -/*! - * @brief Calculates the 64-bit hash of @p input using xxHash64. - * - * This function usually runs faster on 64-bit systems, but slower on 32-bit - * systems (see benchmark). - * - * @param input The block of data to be hashed, at least @p length bytes in size. - * @param length The length of @p input, in bytes. - * @param seed The 64-bit seed to alter the hash's output predictably. - * - * @pre - * The memory between @p input and @p input + @p length must be valid, - * readable, contiguous memory. However, if @p length is `0`, @p input may be - * `NULL`. In C++, this also must be *TriviallyCopyable*. - * - * @return The calculated 64-bit hash. - * - * @see - * XXH32(), XXH3_64bits_withSeed(), XXH3_128bits_withSeed(), XXH128(): - * Direct equivalents for the other variants of xxHash. - * @see - * XXH64_createState(), XXH64_update(), XXH64_digest(): Streaming version. - */ -XXH_PUBLIC_API XXH64_hash_t XXH64(const void* input, size_t length, XXH64_hash_t seed); - -/******* Streaming *******/ -/*! - * @brief The opaque state struct for the XXH64 streaming API. - * - * @see XXH64_state_s for details. - */ -typedef struct XXH64_state_s XXH64_state_t; /* incomplete type */ -XXH_PUBLIC_API XXH64_state_t* XXH64_createState(void); -XXH_PUBLIC_API XXH_errorcode XXH64_freeState(XXH64_state_t* statePtr); -XXH_PUBLIC_API void XXH64_copyState(XXH64_state_t* dst_state, const XXH64_state_t* src_state); - -XXH_PUBLIC_API XXH_errorcode XXH64_reset (XXH64_state_t* statePtr, XXH64_hash_t seed); -XXH_PUBLIC_API XXH_errorcode XXH64_update (XXH64_state_t* statePtr, const void* input, size_t length); -XXH_PUBLIC_API XXH64_hash_t XXH64_digest (const XXH64_state_t* statePtr); - -/******* Canonical representation *******/ -typedef struct { unsigned char digest[sizeof(XXH64_hash_t)]; } XXH64_canonical_t; -XXH_PUBLIC_API void XXH64_canonicalFromHash(XXH64_canonical_t* dst, XXH64_hash_t hash); -XXH_PUBLIC_API XXH64_hash_t XXH64_hashFromCanonical(const XXH64_canonical_t* src); - -#ifndef XXH_NO_XXH3 -/*! - * @} - * ************************************************************************ - * @defgroup xxh3_family XXH3 family - * @ingroup public - * @{ - * - * XXH3 is a more recent hash algorithm featuring: - * - Improved speed for both small and large inputs - * - True 64-bit and 128-bit outputs - * - SIMD acceleration - * - Improved 32-bit viability - * - * Speed analysis methodology is explained here: - * - * https://fastcompression.blogspot.com/2019/03/presenting-xxh3.html - * - * Compared to XXH64, expect XXH3 to run approximately - * ~2x faster on large inputs and >3x faster on small ones, - * exact differences vary depending on platform. - * - * XXH3's speed benefits greatly from SIMD and 64-bit arithmetic, - * but does not require it. - * Any 32-bit and 64-bit targets that can run XXH32 smoothly - * can run XXH3 at competitive speeds, even without vector support. - * Further details are explained in the implementation. - * - * Optimized implementations are provided for AVX512, AVX2, SSE2, NEON, POWER8, - * ZVector and scalar targets. This can be controlled via the XXH_VECTOR macro. - * - * XXH3 implementation is portable: - * it has a generic C90 formulation that can be compiled on any platform, - * all implementations generage exactly the same hash value on all platforms. - * Starting from v0.8.0, it's also labelled "stable", meaning that - * any future version will also generate the same hash value. - * - * XXH3 offers 2 variants, _64bits and _128bits. - * - * When only 64 bits are needed, prefer invoking the _64bits variant, as it - * reduces the amount of mixing, resulting in faster speed on small inputs. - * It's also generally simpler to manipulate a scalar return type than a struct. - * - * The API supports one-shot hashing, streaming mode, and custom secrets. - */ - -/*-********************************************************************** -* XXH3 64-bit variant -************************************************************************/ - -/* XXH3_64bits(): - * default 64-bit variant, using default secret and default seed of 0. - * It's the fastest variant. */ -XXH_PUBLIC_API XXH64_hash_t XXH3_64bits(const void* data, size_t len); - -/* - * XXH3_64bits_withSeed(): - * This variant generates a custom secret on the fly - * based on default secret altered using the `seed` value. - * While this operation is decently fast, note that it's not completely free. - * Note: seed==0 produces the same results as XXH3_64bits(). - */ -XXH_PUBLIC_API XXH64_hash_t XXH3_64bits_withSeed(const void* data, size_t len, XXH64_hash_t seed); - -/*! - * The bare minimum size for a custom secret. - * - * @see - * XXH3_64bits_withSecret(), XXH3_64bits_reset_withSecret(), - * XXH3_128bits_withSecret(), XXH3_128bits_reset_withSecret(). - */ -#define XXH3_SECRET_SIZE_MIN 136 - -/* - * XXH3_64bits_withSecret(): - * It's possible to provide any blob of bytes as a "secret" to generate the hash. - * This makes it more difficult for an external actor to prepare an intentional collision. - * The main condition is that secretSize *must* be large enough (>= XXH3_SECRET_SIZE_MIN). - * However, the quality of the secret impacts the dispersion of the hash algorithm. - * Therefore, the secret _must_ look like a bunch of random bytes. - * Avoid "trivial" or structured data such as repeated sequences or a text document. - * Whenever in doubt about the "randomness" of the blob of bytes, - * consider employing "XXH3_generateSecret()" instead (see below). - * It will generate a proper high entropy secret derived from the blob of bytes. - * Another advantage of using XXH3_generateSecret() is that - * it guarantees that all bits within the initial blob of bytes - * will impact every bit of the output. - * This is not necessarily the case when using the blob of bytes directly - * because, when hashing _small_ inputs, only a portion of the secret is employed. - */ -XXH_PUBLIC_API XXH64_hash_t XXH3_64bits_withSecret(const void* data, size_t len, const void* secret, size_t secretSize); - - -/******* Streaming *******/ -/* - * Streaming requires state maintenance. - * This operation costs memory and CPU. - * As a consequence, streaming is slower than one-shot hashing. - * For better performance, prefer one-shot functions whenever applicable. - */ - -/*! - * @brief The state struct for the XXH3 streaming API. - * - * @see XXH3_state_s for details. - */ -typedef struct XXH3_state_s XXH3_state_t; -XXH_PUBLIC_API XXH3_state_t* XXH3_createState(void); -XXH_PUBLIC_API XXH_errorcode XXH3_freeState(XXH3_state_t* statePtr); -XXH_PUBLIC_API void XXH3_copyState(XXH3_state_t* dst_state, const XXH3_state_t* src_state); - -/* - * XXH3_64bits_reset(): - * Initialize with default parameters. - * digest will be equivalent to `XXH3_64bits()`. - */ -XXH_PUBLIC_API XXH_errorcode XXH3_64bits_reset(XXH3_state_t* statePtr); -/* - * XXH3_64bits_reset_withSeed(): - * Generate a custom secret from `seed`, and store it into `statePtr`. - * digest will be equivalent to `XXH3_64bits_withSeed()`. - */ -XXH_PUBLIC_API XXH_errorcode XXH3_64bits_reset_withSeed(XXH3_state_t* statePtr, XXH64_hash_t seed); -/* - * XXH3_64bits_reset_withSecret(): - * `secret` is referenced, it _must outlive_ the hash streaming session. - * Similar to one-shot API, `secretSize` must be >= `XXH3_SECRET_SIZE_MIN`, - * and the quality of produced hash values depends on secret's entropy - * (secret's content should look like a bunch of random bytes). - * When in doubt about the randomness of a candidate `secret`, - * consider employing `XXH3_generateSecret()` instead (see below). - */ -XXH_PUBLIC_API XXH_errorcode XXH3_64bits_reset_withSecret(XXH3_state_t* statePtr, const void* secret, size_t secretSize); - -XXH_PUBLIC_API XXH_errorcode XXH3_64bits_update (XXH3_state_t* statePtr, const void* input, size_t length); -XXH_PUBLIC_API XXH64_hash_t XXH3_64bits_digest (const XXH3_state_t* statePtr); - -/* note : canonical representation of XXH3 is the same as XXH64 - * since they both produce XXH64_hash_t values */ - - -/*-********************************************************************** -* XXH3 128-bit variant -************************************************************************/ - -/*! - * @brief The return value from 128-bit hashes. - * - * Stored in little endian order, although the fields themselves are in native - * endianness. - */ -typedef struct { - XXH64_hash_t low64; /*!< `value & 0xFFFFFFFFFFFFFFFF` */ - XXH64_hash_t high64; /*!< `value >> 64` */ -} XXH128_hash_t; - -XXH_PUBLIC_API XXH128_hash_t XXH3_128bits(const void* data, size_t len); -XXH_PUBLIC_API XXH128_hash_t XXH3_128bits_withSeed(const void* data, size_t len, XXH64_hash_t seed); -XXH_PUBLIC_API XXH128_hash_t XXH3_128bits_withSecret(const void* data, size_t len, const void* secret, size_t secretSize); - -/******* Streaming *******/ -/* - * Streaming requires state maintenance. - * This operation costs memory and CPU. - * As a consequence, streaming is slower than one-shot hashing. - * For better performance, prefer one-shot functions whenever applicable. - * - * XXH3_128bits uses the same XXH3_state_t as XXH3_64bits(). - * Use already declared XXH3_createState() and XXH3_freeState(). - * - * All reset and streaming functions have same meaning as their 64-bit counterpart. - */ - -XXH_PUBLIC_API XXH_errorcode XXH3_128bits_reset(XXH3_state_t* statePtr); -XXH_PUBLIC_API XXH_errorcode XXH3_128bits_reset_withSeed(XXH3_state_t* statePtr, XXH64_hash_t seed); -XXH_PUBLIC_API XXH_errorcode XXH3_128bits_reset_withSecret(XXH3_state_t* statePtr, const void* secret, size_t secretSize); - -XXH_PUBLIC_API XXH_errorcode XXH3_128bits_update (XXH3_state_t* statePtr, const void* input, size_t length); -XXH_PUBLIC_API XXH128_hash_t XXH3_128bits_digest (const XXH3_state_t* statePtr); - -/* Following helper functions make it possible to compare XXH128_hast_t values. - * Since XXH128_hash_t is a structure, this capability is not offered by the language. - * Note: For better performance, these functions can be inlined using XXH_INLINE_ALL */ - -/*! - * XXH128_isEqual(): - * Return: 1 if `h1` and `h2` are equal, 0 if they are not. - */ -XXH_PUBLIC_API int XXH128_isEqual(XXH128_hash_t h1, XXH128_hash_t h2); - -/*! - * XXH128_cmp(): - * - * This comparator is compatible with stdlib's `qsort()`/`bsearch()`. - * - * return: >0 if *h128_1 > *h128_2 - * =0 if *h128_1 == *h128_2 - * <0 if *h128_1 < *h128_2 - */ -XXH_PUBLIC_API int XXH128_cmp(const void* h128_1, const void* h128_2); - - -/******* Canonical representation *******/ -typedef struct { unsigned char digest[sizeof(XXH128_hash_t)]; } XXH128_canonical_t; -XXH_PUBLIC_API void XXH128_canonicalFromHash(XXH128_canonical_t* dst, XXH128_hash_t hash); -XXH_PUBLIC_API XXH128_hash_t XXH128_hashFromCanonical(const XXH128_canonical_t* src); - - -#endif /* !XXH_NO_XXH3 */ -#endif /* XXH_NO_LONG_LONG */ - -/*! - * @} - */ -#endif /* XXHASH_H_5627135585666179 */ - - - -#if defined(XXH_STATIC_LINKING_ONLY) && !defined(XXHASH_H_STATIC_13879238742) -#define XXHASH_H_STATIC_13879238742 -/* **************************************************************************** - * This section contains declarations which are not guaranteed to remain stable. - * They may change in future versions, becoming incompatible with a different - * version of the library. - * These declarations should only be used with static linking. - * Never use them in association with dynamic linking! - ***************************************************************************** */ - -/* - * These definitions are only present to allow static allocation - * of XXH states, on stack or in a struct, for example. - * Never **ever** access their members directly. - */ - -/*! - * @internal - * @brief Structure for XXH32 streaming API. - * - * @note This is only defined when @ref XXH_STATIC_LINKING_ONLY, - * @ref XXH_INLINE_ALL, or @ref XXH_IMPLEMENTATION is defined. Otherwise it is - * an opaque type. This allows fields to safely be changed. - * - * Typedef'd to @ref XXH32_state_t. - * Do not access the members of this struct directly. - * @see XXH64_state_s, XXH3_state_s - */ -struct XXH32_state_s { - XXH32_hash_t total_len_32; /*!< Total length hashed, modulo 2^32 */ - XXH32_hash_t large_len; /*!< Whether the hash is >= 16 (handles @ref total_len_32 overflow) */ - XXH32_hash_t v[4]; /*!< Accumulator lanes */ - XXH32_hash_t mem32[4]; /*!< Internal buffer for partial reads. Treated as unsigned char[16]. */ - XXH32_hash_t memsize; /*!< Amount of data in @ref mem32 */ - XXH32_hash_t reserved; /*!< Reserved field. Do not read nor write to it. */ -}; /* typedef'd to XXH32_state_t */ - - -#ifndef XXH_NO_LONG_LONG /* defined when there is no 64-bit support */ - -/*! - * @internal - * @brief Structure for XXH64 streaming API. - * - * @note This is only defined when @ref XXH_STATIC_LINKING_ONLY, - * @ref XXH_INLINE_ALL, or @ref XXH_IMPLEMENTATION is defined. Otherwise it is - * an opaque type. This allows fields to safely be changed. - * - * Typedef'd to @ref XXH64_state_t. - * Do not access the members of this struct directly. - * @see XXH32_state_s, XXH3_state_s - */ -struct XXH64_state_s { - XXH64_hash_t total_len; /*!< Total length hashed. This is always 64-bit. */ - XXH64_hash_t v[4]; /*!< Accumulator lanes */ - XXH64_hash_t mem64[4]; /*!< Internal buffer for partial reads. Treated as unsigned char[32]. */ - XXH32_hash_t memsize; /*!< Amount of data in @ref mem64 */ - XXH32_hash_t reserved32; /*!< Reserved field, needed for padding anyways*/ - XXH64_hash_t reserved64; /*!< Reserved field. Do not read or write to it. */ -}; /* typedef'd to XXH64_state_t */ - - -#ifndef XXH_NO_XXH3 - -#if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L) /* >= C11 */ -# include -# define XXH_ALIGN(n) alignas(n) -#elif defined(__cplusplus) && (__cplusplus >= 201103L) /* >= C++11 */ -/* In C++ alignas() is a keyword */ -# define XXH_ALIGN(n) alignas(n) -#elif defined(__GNUC__) -# define XXH_ALIGN(n) __attribute__ ((aligned(n))) -#elif defined(_MSC_VER) -# define XXH_ALIGN(n) __declspec(align(n)) -#else -# define XXH_ALIGN(n) /* disabled */ -#endif - -/* Old GCC versions only accept the attribute after the type in structures. */ -#if !(defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L)) /* C11+ */ \ - && ! (defined(__cplusplus) && (__cplusplus >= 201103L)) /* >= C++11 */ \ - && defined(__GNUC__) -# define XXH_ALIGN_MEMBER(align, type) type XXH_ALIGN(align) -#else -# define XXH_ALIGN_MEMBER(align, type) XXH_ALIGN(align) type -#endif - -/*! - * @brief The size of the internal XXH3 buffer. - * - * This is the optimal update size for incremental hashing. - * - * @see XXH3_64b_update(), XXH3_128b_update(). - */ -#define XXH3_INTERNALBUFFER_SIZE 256 - -/*! - * @brief Default size of the secret buffer (and @ref XXH3_kSecret). - * - * This is the size used in @ref XXH3_kSecret and the seeded functions. - * - * Not to be confused with @ref XXH3_SECRET_SIZE_MIN. - */ -#define XXH3_SECRET_DEFAULT_SIZE 192 - -/*! - * @internal - * @brief Structure for XXH3 streaming API. - * - * @note This is only defined when @ref XXH_STATIC_LINKING_ONLY, - * @ref XXH_INLINE_ALL, or @ref XXH_IMPLEMENTATION is defined. - * Otherwise it is an opaque type. - * Never use this definition in combination with dynamic library. - * This allows fields to safely be changed in the future. - * - * @note ** This structure has a strict alignment requirement of 64 bytes!! ** - * Do not allocate this with `malloc()` or `new`, - * it will not be sufficiently aligned. - * Use @ref XXH3_createState() and @ref XXH3_freeState(), or stack allocation. - * - * Typedef'd to @ref XXH3_state_t. - * Do never access the members of this struct directly. - * - * @see XXH3_INITSTATE() for stack initialization. - * @see XXH3_createState(), XXH3_freeState(). - * @see XXH32_state_s, XXH64_state_s - */ -struct XXH3_state_s { - XXH_ALIGN_MEMBER(64, XXH64_hash_t acc[8]); - /*!< The 8 accumulators. Similar to `vN` in @ref XXH32_state_s::v1 and @ref XXH64_state_s */ - XXH_ALIGN_MEMBER(64, unsigned char customSecret[XXH3_SECRET_DEFAULT_SIZE]); - /*!< Used to store a custom secret generated from a seed. */ - XXH_ALIGN_MEMBER(64, unsigned char buffer[XXH3_INTERNALBUFFER_SIZE]); - /*!< The internal buffer. @see XXH32_state_s::mem32 */ - XXH32_hash_t bufferedSize; - /*!< The amount of memory in @ref buffer, @see XXH32_state_s::memsize */ - XXH32_hash_t useSeed; - /*!< Reserved field. Needed for padding on 64-bit. */ - size_t nbStripesSoFar; - /*!< Number or stripes processed. */ - XXH64_hash_t totalLen; - /*!< Total length hashed. 64-bit even on 32-bit targets. */ - size_t nbStripesPerBlock; - /*!< Number of stripes per block. */ - size_t secretLimit; - /*!< Size of @ref customSecret or @ref extSecret */ - XXH64_hash_t seed; - /*!< Seed for _withSeed variants. Must be zero otherwise, @see XXH3_INITSTATE() */ - XXH64_hash_t reserved64; - /*!< Reserved field. */ - const unsigned char* extSecret; - /*!< Reference to an external secret for the _withSecret variants, NULL - * for other variants. */ - /* note: there may be some padding at the end due to alignment on 64 bytes */ -}; /* typedef'd to XXH3_state_t */ - -#undef XXH_ALIGN_MEMBER - -/*! - * @brief Initializes a stack-allocated `XXH3_state_s`. - * - * When the @ref XXH3_state_t structure is merely emplaced on stack, - * it should be initialized with XXH3_INITSTATE() or a memset() - * in case its first reset uses XXH3_NNbits_reset_withSeed(). - * This init can be omitted if the first reset uses default or _withSecret mode. - * This operation isn't necessary when the state is created with XXH3_createState(). - * Note that this doesn't prepare the state for a streaming operation, - * it's still necessary to use XXH3_NNbits_reset*() afterwards. - */ -#define XXH3_INITSTATE(XXH3_state_ptr) { (XXH3_state_ptr)->seed = 0; } - - -/* XXH128() : - * simple alias to pre-selected XXH3_128bits variant - */ -XXH_PUBLIC_API XXH128_hash_t XXH128(const void* data, size_t len, XXH64_hash_t seed); - - -/* === Experimental API === */ -/* Symbols defined below must be considered tied to a specific library version. */ - -/* - * XXH3_generateSecret(): - * - * Derive a high-entropy secret from any user-defined content, named customSeed. - * The generated secret can be used in combination with `*_withSecret()` functions. - * The `_withSecret()` variants are useful to provide a higher level of protection than 64-bit seed, - * as it becomes much more difficult for an external actor to guess how to impact the calculation logic. - * - * The function accepts as input a custom seed of any length and any content, - * and derives from it a high-entropy secret of length @secretSize - * into an already allocated buffer @secretBuffer. - * @secretSize must be >= XXH3_SECRET_SIZE_MIN - * - * The generated secret can then be used with any `*_withSecret()` variant. - * Functions `XXH3_128bits_withSecret()`, `XXH3_64bits_withSecret()`, - * `XXH3_128bits_reset_withSecret()` and `XXH3_64bits_reset_withSecret()` - * are part of this list. They all accept a `secret` parameter - * which must be large enough for implementation reasons (>= XXH3_SECRET_SIZE_MIN) - * _and_ feature very high entropy (consist of random-looking bytes). - * These conditions can be a high bar to meet, so - * XXH3_generateSecret() can be employed to ensure proper quality. - * - * customSeed can be anything. It can have any size, even small ones, - * and its content can be anything, even "poor entropy" sources such as a bunch of zeroes. - * The resulting `secret` will nonetheless provide all required qualities. - * - * When customSeedSize > 0, supplying NULL as customSeed is undefined behavior. - */ -XXH_PUBLIC_API XXH_errorcode XXH3_generateSecret(void* secretBuffer, size_t secretSize, const void* customSeed, size_t customSeedSize); - - -/* - * XXH3_generateSecret_fromSeed(): - * - * Generate the same secret as the _withSeed() variants. - * - * The resulting secret has a length of XXH3_SECRET_DEFAULT_SIZE (necessarily). - * @secretBuffer must be already allocated, of size at least XXH3_SECRET_DEFAULT_SIZE bytes. - * - * The generated secret can be used in combination with - *`*_withSecret()` and `_withSecretandSeed()` variants. - * This generator is notably useful in combination with `_withSecretandSeed()`, - * as a way to emulate a faster `_withSeed()` variant. - */ -XXH_PUBLIC_API void XXH3_generateSecret_fromSeed(void* secretBuffer, XXH64_hash_t seed); - -/* - * *_withSecretandSeed() : - * These variants generate hash values using either - * @seed for "short" keys (< XXH3_MIDSIZE_MAX = 240 bytes) - * or @secret for "large" keys (>= XXH3_MIDSIZE_MAX). - * - * This generally benefits speed, compared to `_withSeed()` or `_withSecret()`. - * `_withSeed()` has to generate the secret on the fly for "large" keys. - * It's fast, but can be perceptible for "not so large" keys (< 1 KB). - * `_withSecret()` has to generate the masks on the fly for "small" keys, - * which requires more instructions than _withSeed() variants. - * Therefore, _withSecretandSeed variant combines the best of both worlds. - * - * When @secret has been generated by XXH3_generateSecret_fromSeed(), - * this variant produces *exactly* the same results as `_withSeed()` variant, - * hence offering only a pure speed benefit on "large" input, - * by skipping the need to regenerate the secret for every large input. - * - * Another usage scenario is to hash the secret to a 64-bit hash value, - * for example with XXH3_64bits(), which then becomes the seed, - * and then employ both the seed and the secret in _withSecretandSeed(). - * On top of speed, an added benefit is that each bit in the secret - * has a 50% chance to swap each bit in the output, - * via its impact to the seed. - * This is not guaranteed when using the secret directly in "small data" scenarios, - * because only portions of the secret are employed for small data. - */ -XXH_PUBLIC_API XXH64_hash_t -XXH3_64bits_withSecretandSeed(const void* data, size_t len, - const void* secret, size_t secretSize, - XXH64_hash_t seed); - -XXH_PUBLIC_API XXH128_hash_t -XXH3_128bits_withSecretandSeed(const void* data, size_t len, - const void* secret, size_t secretSize, - XXH64_hash_t seed64); - -XXH_PUBLIC_API XXH_errorcode -XXH3_64bits_reset_withSecretandSeed(XXH3_state_t* statePtr, - const void* secret, size_t secretSize, - XXH64_hash_t seed64); - -XXH_PUBLIC_API XXH_errorcode -XXH3_128bits_reset_withSecretandSeed(XXH3_state_t* statePtr, - const void* secret, size_t secretSize, - XXH64_hash_t seed64); - - -#endif /* XXH_NO_XXH3 */ -#endif /* XXH_NO_LONG_LONG */ -#if defined(XXH_INLINE_ALL) || defined(XXH_PRIVATE_API) -# define XXH_IMPLEMENTATION -#endif - -#endif /* defined(XXH_STATIC_LINKING_ONLY) && !defined(XXHASH_H_STATIC_13879238742) */ - - -/* ======================================================================== */ -/* ======================================================================== */ -/* ======================================================================== */ - - -/*-********************************************************************** - * xxHash implementation - *-********************************************************************** - * xxHash's implementation used to be hosted inside xxhash.c. - * - * However, inlining requires implementation to be visible to the compiler, - * hence be included alongside the header. - * Previously, implementation was hosted inside xxhash.c, - * which was then #included when inlining was activated. - * This construction created issues with a few build and install systems, - * as it required xxhash.c to be stored in /include directory. - * - * xxHash implementation is now directly integrated within xxhash.h. - * As a consequence, xxhash.c is no longer needed in /include. - * - * xxhash.c is still available and is still useful. - * In a "normal" setup, when xxhash is not inlined, - * xxhash.h only exposes the prototypes and public symbols, - * while xxhash.c can be built into an object file xxhash.o - * which can then be linked into the final binary. - ************************************************************************/ - -#if ( defined(XXH_INLINE_ALL) || defined(XXH_PRIVATE_API) \ - || defined(XXH_IMPLEMENTATION) ) && !defined(XXH_IMPLEM_13a8737387) -# define XXH_IMPLEM_13a8737387 - -/* ************************************* -* Tuning parameters -***************************************/ - -/*! - * @defgroup tuning Tuning parameters - * @{ - * - * Various macros to control xxHash's behavior. - */ -#ifdef XXH_DOXYGEN -/*! - * @brief Define this to disable 64-bit code. - * - * Useful if only using the @ref xxh32_family and you have a strict C90 compiler. - */ -# define XXH_NO_LONG_LONG -# undef XXH_NO_LONG_LONG /* don't actually */ -/*! - * @brief Controls how unaligned memory is accessed. - * - * By default, access to unaligned memory is controlled by `memcpy()`, which is - * safe and portable. - * - * Unfortunately, on some target/compiler combinations, the generated assembly - * is sub-optimal. - * - * The below switch allow selection of a different access method - * in the search for improved performance. - * - * @par Possible options: - * - * - `XXH_FORCE_MEMORY_ACCESS=0` (default): `memcpy` - * @par - * Use `memcpy()`. Safe and portable. Note that most modern compilers will - * eliminate the function call and treat it as an unaligned access. - * - * - `XXH_FORCE_MEMORY_ACCESS=1`: `__attribute__((packed))` - * @par - * Depends on compiler extensions and is therefore not portable. - * This method is safe _if_ your compiler supports it, - * and *generally* as fast or faster than `memcpy`. - * - * - `XXH_FORCE_MEMORY_ACCESS=2`: Direct cast - * @par - * Casts directly and dereferences. This method doesn't depend on the - * compiler, but it violates the C standard as it directly dereferences an - * unaligned pointer. It can generate buggy code on targets which do not - * support unaligned memory accesses, but in some circumstances, it's the - * only known way to get the most performance. - * - * - `XXH_FORCE_MEMORY_ACCESS=3`: Byteshift - * @par - * Also portable. This can generate the best code on old compilers which don't - * inline small `memcpy()` calls, and it might also be faster on big-endian - * systems which lack a native byteswap instruction. However, some compilers - * will emit literal byteshifts even if the target supports unaligned access. - * . - * - * @warning - * Methods 1 and 2 rely on implementation-defined behavior. Use these with - * care, as what works on one compiler/platform/optimization level may cause - * another to read garbage data or even crash. - * - * See http://fastcompression.blogspot.com/2015/08/accessing-unaligned-memory.html for details. - * - * Prefer these methods in priority order (0 > 3 > 1 > 2) - */ -# define XXH_FORCE_MEMORY_ACCESS 0 - -/*! - * @def XXH_FORCE_ALIGN_CHECK - * @brief If defined to non-zero, adds a special path for aligned inputs (XXH32() - * and XXH64() only). - * - * This is an important performance trick for architectures without decent - * unaligned memory access performance. - * - * It checks for input alignment, and when conditions are met, uses a "fast - * path" employing direct 32-bit/64-bit reads, resulting in _dramatically - * faster_ read speed. - * - * The check costs one initial branch per hash, which is generally negligible, - * but not zero. - * - * Moreover, it's not useful to generate an additional code path if memory - * access uses the same instruction for both aligned and unaligned - * addresses (e.g. x86 and aarch64). - * - * In these cases, the alignment check can be removed by setting this macro to 0. - * Then the code will always use unaligned memory access. - * Align check is automatically disabled on x86, x64 & arm64, - * which are platforms known to offer good unaligned memory accesses performance. - * - * This option does not affect XXH3 (only XXH32 and XXH64). - */ -# define XXH_FORCE_ALIGN_CHECK 0 - -/*! - * @def XXH_NO_INLINE_HINTS - * @brief When non-zero, sets all functions to `static`. - * - * By default, xxHash tries to force the compiler to inline almost all internal - * functions. - * - * This can usually improve performance due to reduced jumping and improved - * constant folding, but significantly increases the size of the binary which - * might not be favorable. - * - * Additionally, sometimes the forced inlining can be detrimental to performance, - * depending on the architecture. - * - * XXH_NO_INLINE_HINTS marks all internal functions as static, giving the - * compiler full control on whether to inline or not. - * - * When not optimizing (-O0), optimizing for size (-Os, -Oz), or using - * -fno-inline with GCC or Clang, this will automatically be defined. - */ -# define XXH_NO_INLINE_HINTS 0 - -/*! - * @def XXH32_ENDJMP - * @brief Whether to use a jump for `XXH32_finalize`. - * - * For performance, `XXH32_finalize` uses multiple branches in the finalizer. - * This is generally preferable for performance, - * but depending on exact architecture, a jmp may be preferable. - * - * This setting is only possibly making a difference for very small inputs. - */ -# define XXH32_ENDJMP 0 - -/*! - * @internal - * @brief Redefines old internal names. - * - * For compatibility with code that uses xxHash's internals before the names - * were changed to improve namespacing. There is no other reason to use this. - */ -# define XXH_OLD_NAMES -# undef XXH_OLD_NAMES /* don't actually use, it is ugly. */ -#endif /* XXH_DOXYGEN */ -/*! - * @} - */ - -#ifndef XXH_FORCE_MEMORY_ACCESS /* can be defined externally, on command line for example */ - /* prefer __packed__ structures (method 1) for gcc on armv7+ and mips */ -# if !defined(__clang__) && \ -( \ - (defined(__INTEL_COMPILER) && !defined(_WIN32)) || \ - ( \ - defined(__GNUC__) && ( \ - (defined(__ARM_ARCH) && __ARM_ARCH >= 7) || \ - ( \ - defined(__mips__) && \ - (__mips <= 5 || __mips_isa_rev < 6) && \ - (!defined(__mips16) || defined(__mips_mips16e2)) \ - ) \ - ) \ - ) \ -) -# define XXH_FORCE_MEMORY_ACCESS 1 -# endif -#endif - -#ifndef XXH_FORCE_ALIGN_CHECK /* can be defined externally */ -# if defined(__i386) || defined(__x86_64__) || defined(__aarch64__) \ - || defined(_M_IX86) || defined(_M_X64) || defined(_M_ARM64) /* visual */ -# define XXH_FORCE_ALIGN_CHECK 0 -# else -# define XXH_FORCE_ALIGN_CHECK 1 -# endif -#endif - -#ifndef XXH_NO_INLINE_HINTS -# if defined(__OPTIMIZE_SIZE__) /* -Os, -Oz */ \ - || defined(__NO_INLINE__) /* -O0, -fno-inline */ -# define XXH_NO_INLINE_HINTS 1 -# else -# define XXH_NO_INLINE_HINTS 0 -# endif -#endif - -#ifndef XXH32_ENDJMP -/* generally preferable for performance */ -# define XXH32_ENDJMP 0 -#endif - -/*! - * @defgroup impl Implementation - * @{ - */ - - -/* ************************************* -* Includes & Memory related functions -***************************************/ -/* Modify the local functions below should you wish to use some other memory routines */ -/* for ZSTD_malloc(), ZSTD_free() */ -#define ZSTD_DEPS_NEED_MALLOC -#include "zstd_deps.h" /* size_t, ZSTD_malloc, ZSTD_free, ZSTD_memcpy */ -static void* XXH_malloc(size_t s) { return ZSTD_malloc(s); } -static void XXH_free (void* p) { ZSTD_free(p); } -static void* XXH_memcpy(void* dest, const void* src, size_t size) { return ZSTD_memcpy(dest,src,size); } - - -/* ************************************* -* Compiler Specific Options -***************************************/ -#ifdef _MSC_VER /* Visual Studio warning fix */ -# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */ -#endif - -#if XXH_NO_INLINE_HINTS /* disable inlining hints */ -# if defined(__GNUC__) || defined(__clang__) -# define XXH_FORCE_INLINE static __attribute__((unused)) -# else -# define XXH_FORCE_INLINE static -# endif -# define XXH_NO_INLINE static -/* enable inlining hints */ -#elif defined(__GNUC__) || defined(__clang__) -# define XXH_FORCE_INLINE static __inline__ __attribute__((always_inline, unused)) -# define XXH_NO_INLINE static __attribute__((noinline)) -#elif defined(_MSC_VER) /* Visual Studio */ -# define XXH_FORCE_INLINE static __forceinline -# define XXH_NO_INLINE static __declspec(noinline) -#elif defined (__cplusplus) \ - || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L)) /* C99 */ -# define XXH_FORCE_INLINE static inline -# define XXH_NO_INLINE static -#else -# define XXH_FORCE_INLINE static -# define XXH_NO_INLINE static -#endif - - - -/* ************************************* -* Debug -***************************************/ -/*! - * @ingroup tuning - * @def XXH_DEBUGLEVEL - * @brief Sets the debugging level. - * - * XXH_DEBUGLEVEL is expected to be defined externally, typically via the - * compiler's command line options. The value must be a number. - */ -#ifndef XXH_DEBUGLEVEL -# ifdef DEBUGLEVEL /* backwards compat */ -# define XXH_DEBUGLEVEL DEBUGLEVEL -# else -# define XXH_DEBUGLEVEL 0 -# endif -#endif - -#if (XXH_DEBUGLEVEL>=1) -# include /* note: can still be disabled with NDEBUG */ -# define XXH_ASSERT(c) assert(c) -#else -# define XXH_ASSERT(c) ((void)0) -#endif - -/* note: use after variable declarations */ -#ifndef XXH_STATIC_ASSERT -# if defined(__STDC_VERSION__) && (__STDC_VERSION__ >= 201112L) /* C11 */ -# include -# define XXH_STATIC_ASSERT_WITH_MESSAGE(c,m) do { static_assert((c),m); } while(0) -# elif defined(__cplusplus) && (__cplusplus >= 201103L) /* C++11 */ -# define XXH_STATIC_ASSERT_WITH_MESSAGE(c,m) do { static_assert((c),m); } while(0) -# else -# define XXH_STATIC_ASSERT_WITH_MESSAGE(c,m) do { struct xxh_sa { char x[(c) ? 1 : -1]; }; } while(0) -# endif -# define XXH_STATIC_ASSERT(c) XXH_STATIC_ASSERT_WITH_MESSAGE((c),#c) -#endif - -/*! - * @internal - * @def XXH_COMPILER_GUARD(var) - * @brief Used to prevent unwanted optimizations for @p var. - * - * It uses an empty GCC inline assembly statement with a register constraint - * which forces @p var into a general purpose register (eg eax, ebx, ecx - * on x86) and marks it as modified. - * - * This is used in a few places to avoid unwanted autovectorization (e.g. - * XXH32_round()). All vectorization we want is explicit via intrinsics, - * and _usually_ isn't wanted elsewhere. - * - * We also use it to prevent unwanted constant folding for AArch64 in - * XXH3_initCustomSecret_scalar(). - */ -#if defined(__GNUC__) || defined(__clang__) -# define XXH_COMPILER_GUARD(var) __asm__ __volatile__("" : "+r" (var)) -#else -# define XXH_COMPILER_GUARD(var) ((void)0) -#endif - -/* ************************************* -* Basic Types -***************************************/ -#if !defined (__VMS) \ - && (defined (__cplusplus) \ - || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) ) -# include - typedef uint8_t xxh_u8; -#else - typedef unsigned char xxh_u8; -#endif -typedef XXH32_hash_t xxh_u32; - -#ifdef XXH_OLD_NAMES -# define BYTE xxh_u8 -# define U8 xxh_u8 -# define U32 xxh_u32 -#endif - -/* *** Memory access *** */ - -/*! - * @internal - * @fn xxh_u32 XXH_read32(const void* ptr) - * @brief Reads an unaligned 32-bit integer from @p ptr in native endianness. - * - * Affected by @ref XXH_FORCE_MEMORY_ACCESS. - * - * @param ptr The pointer to read from. - * @return The 32-bit native endian integer from the bytes at @p ptr. - */ - -/*! - * @internal - * @fn xxh_u32 XXH_readLE32(const void* ptr) - * @brief Reads an unaligned 32-bit little endian integer from @p ptr. - * - * Affected by @ref XXH_FORCE_MEMORY_ACCESS. - * - * @param ptr The pointer to read from. - * @return The 32-bit little endian integer from the bytes at @p ptr. - */ - -/*! - * @internal - * @fn xxh_u32 XXH_readBE32(const void* ptr) - * @brief Reads an unaligned 32-bit big endian integer from @p ptr. - * - * Affected by @ref XXH_FORCE_MEMORY_ACCESS. - * - * @param ptr The pointer to read from. - * @return The 32-bit big endian integer from the bytes at @p ptr. - */ - -/*! - * @internal - * @fn xxh_u32 XXH_readLE32_align(const void* ptr, XXH_alignment align) - * @brief Like @ref XXH_readLE32(), but has an option for aligned reads. - * - * Affected by @ref XXH_FORCE_MEMORY_ACCESS. - * Note that when @ref XXH_FORCE_ALIGN_CHECK == 0, the @p align parameter is - * always @ref XXH_alignment::XXH_unaligned. - * - * @param ptr The pointer to read from. - * @param align Whether @p ptr is aligned. - * @pre - * If @p align == @ref XXH_alignment::XXH_aligned, @p ptr must be 4 byte - * aligned. - * @return The 32-bit little endian integer from the bytes at @p ptr. - */ - -#if (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==3)) -/* - * Manual byteshift. Best for old compilers which don't inline memcpy. - * We actually directly use XXH_readLE32 and XXH_readBE32. - */ -#elif (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==2)) - -/* - * Force direct memory access. Only works on CPU which support unaligned memory - * access in hardware. - */ -static xxh_u32 XXH_read32(const void* memPtr) { return *(const xxh_u32*) memPtr; } - -#elif (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==1)) - -/* - * __pack instructions are safer but compiler specific, hence potentially - * problematic for some compilers. - * - * Currently only defined for GCC and ICC. - */ -#ifdef XXH_OLD_NAMES -typedef union { xxh_u32 u32; } __attribute__((packed)) unalign; -#endif -static xxh_u32 XXH_read32(const void* ptr) -{ - typedef union { xxh_u32 u32; } __attribute__((packed)) xxh_unalign; - return ((const xxh_unalign*)ptr)->u32; -} - -#else - -/* - * Portable and safe solution. Generally efficient. - * see: http://fastcompression.blogspot.com/2015/08/accessing-unaligned-memory.html - */ -static xxh_u32 XXH_read32(const void* memPtr) -{ - xxh_u32 val; - XXH_memcpy(&val, memPtr, sizeof(val)); - return val; -} - -#endif /* XXH_FORCE_DIRECT_MEMORY_ACCESS */ - - -/* *** Endianness *** */ - -/*! - * @ingroup tuning - * @def XXH_CPU_LITTLE_ENDIAN - * @brief Whether the target is little endian. - * - * Defined to 1 if the target is little endian, or 0 if it is big endian. - * It can be defined externally, for example on the compiler command line. - * - * If it is not defined, - * a runtime check (which is usually constant folded) is used instead. - * - * @note - * This is not necessarily defined to an integer constant. - * - * @see XXH_isLittleEndian() for the runtime check. - */ -#ifndef XXH_CPU_LITTLE_ENDIAN -/* - * Try to detect endianness automatically, to avoid the nonstandard behavior - * in `XXH_isLittleEndian()` - */ -# if defined(_WIN32) /* Windows is always little endian */ \ - || defined(__LITTLE_ENDIAN__) \ - || (defined(__BYTE_ORDER__) && __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__) -# define XXH_CPU_LITTLE_ENDIAN 1 -# elif defined(__BIG_ENDIAN__) \ - || (defined(__BYTE_ORDER__) && __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__) -# define XXH_CPU_LITTLE_ENDIAN 0 -# else -/*! - * @internal - * @brief Runtime check for @ref XXH_CPU_LITTLE_ENDIAN. - * - * Most compilers will constant fold this. - */ -static int XXH_isLittleEndian(void) -{ - /* - * Portable and well-defined behavior. - * Don't use static: it is detrimental to performance. - */ - const union { xxh_u32 u; xxh_u8 c[4]; } one = { 1 }; - return one.c[0]; -} -# define XXH_CPU_LITTLE_ENDIAN XXH_isLittleEndian() -# endif -#endif - - - - -/* **************************************** -* Compiler-specific Functions and Macros -******************************************/ -#define XXH_GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__) - -#ifdef __has_builtin -# define XXH_HAS_BUILTIN(x) __has_builtin(x) -#else -# define XXH_HAS_BUILTIN(x) 0 -#endif - -/*! - * @internal - * @def XXH_rotl32(x,r) - * @brief 32-bit rotate left. - * - * @param x The 32-bit integer to be rotated. - * @param r The number of bits to rotate. - * @pre - * @p r > 0 && @p r < 32 - * @note - * @p x and @p r may be evaluated multiple times. - * @return The rotated result. - */ -#if !defined(NO_CLANG_BUILTIN) && XXH_HAS_BUILTIN(__builtin_rotateleft32) \ - && XXH_HAS_BUILTIN(__builtin_rotateleft64) -# define XXH_rotl32 __builtin_rotateleft32 -# define XXH_rotl64 __builtin_rotateleft64 -/* Note: although _rotl exists for minGW (GCC under windows), performance seems poor */ -#elif defined(_MSC_VER) -# define XXH_rotl32(x,r) _rotl(x,r) -# define XXH_rotl64(x,r) _rotl64(x,r) -#else -# define XXH_rotl32(x,r) (((x) << (r)) | ((x) >> (32 - (r)))) -# define XXH_rotl64(x,r) (((x) << (r)) | ((x) >> (64 - (r)))) -#endif - -/*! - * @internal - * @fn xxh_u32 XXH_swap32(xxh_u32 x) - * @brief A 32-bit byteswap. - * - * @param x The 32-bit integer to byteswap. - * @return @p x, byteswapped. - */ -#if defined(_MSC_VER) /* Visual Studio */ -# define XXH_swap32 _byteswap_ulong -#elif XXH_GCC_VERSION >= 403 -# define XXH_swap32 __builtin_bswap32 -#else -static xxh_u32 XXH_swap32 (xxh_u32 x) -{ - return ((x << 24) & 0xff000000 ) | - ((x << 8) & 0x00ff0000 ) | - ((x >> 8) & 0x0000ff00 ) | - ((x >> 24) & 0x000000ff ); -} -#endif - - -/* *************************** -* Memory reads -*****************************/ - -/*! - * @internal - * @brief Enum to indicate whether a pointer is aligned. - */ -typedef enum { - XXH_aligned, /*!< Aligned */ - XXH_unaligned /*!< Possibly unaligned */ -} XXH_alignment; - -/* - * XXH_FORCE_MEMORY_ACCESS==3 is an endian-independent byteshift load. - * - * This is ideal for older compilers which don't inline memcpy. - */ -#if (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==3)) - -XXH_FORCE_INLINE xxh_u32 XXH_readLE32(const void* memPtr) -{ - const xxh_u8* bytePtr = (const xxh_u8 *)memPtr; - return bytePtr[0] - | ((xxh_u32)bytePtr[1] << 8) - | ((xxh_u32)bytePtr[2] << 16) - | ((xxh_u32)bytePtr[3] << 24); -} - -XXH_FORCE_INLINE xxh_u32 XXH_readBE32(const void* memPtr) -{ - const xxh_u8* bytePtr = (const xxh_u8 *)memPtr; - return bytePtr[3] - | ((xxh_u32)bytePtr[2] << 8) - | ((xxh_u32)bytePtr[1] << 16) - | ((xxh_u32)bytePtr[0] << 24); -} - -#else -XXH_FORCE_INLINE xxh_u32 XXH_readLE32(const void* ptr) -{ - return XXH_CPU_LITTLE_ENDIAN ? XXH_read32(ptr) : XXH_swap32(XXH_read32(ptr)); -} - -static xxh_u32 XXH_readBE32(const void* ptr) -{ - return XXH_CPU_LITTLE_ENDIAN ? XXH_swap32(XXH_read32(ptr)) : XXH_read32(ptr); -} -#endif - -XXH_FORCE_INLINE xxh_u32 -XXH_readLE32_align(const void* ptr, XXH_alignment align) -{ - if (align==XXH_unaligned) { - return XXH_readLE32(ptr); - } else { - return XXH_CPU_LITTLE_ENDIAN ? *(const xxh_u32*)ptr : XXH_swap32(*(const xxh_u32*)ptr); - } -} - - -/* ************************************* -* Misc -***************************************/ -/*! @ingroup public */ -XXH_PUBLIC_API unsigned XXH_versionNumber (void) { return XXH_VERSION_NUMBER; } - - -/* ******************************************************************* -* 32-bit hash functions -*********************************************************************/ -/*! - * @} - * @defgroup xxh32_impl XXH32 implementation - * @ingroup impl - * @{ - */ - /* #define instead of static const, to be used as initializers */ -#define XXH_PRIME32_1 0x9E3779B1U /*!< 0b10011110001101110111100110110001 */ -#define XXH_PRIME32_2 0x85EBCA77U /*!< 0b10000101111010111100101001110111 */ -#define XXH_PRIME32_3 0xC2B2AE3DU /*!< 0b11000010101100101010111000111101 */ -#define XXH_PRIME32_4 0x27D4EB2FU /*!< 0b00100111110101001110101100101111 */ -#define XXH_PRIME32_5 0x165667B1U /*!< 0b00010110010101100110011110110001 */ - -#ifdef XXH_OLD_NAMES -# define PRIME32_1 XXH_PRIME32_1 -# define PRIME32_2 XXH_PRIME32_2 -# define PRIME32_3 XXH_PRIME32_3 -# define PRIME32_4 XXH_PRIME32_4 -# define PRIME32_5 XXH_PRIME32_5 -#endif - -/*! - * @internal - * @brief Normal stripe processing routine. - * - * This shuffles the bits so that any bit from @p input impacts several bits in - * @p acc. - * - * @param acc The accumulator lane. - * @param input The stripe of input to mix. - * @return The mixed accumulator lane. - */ -static xxh_u32 XXH32_round(xxh_u32 acc, xxh_u32 input) -{ - acc += input * XXH_PRIME32_2; - acc = XXH_rotl32(acc, 13); - acc *= XXH_PRIME32_1; -#if (defined(__SSE4_1__) || defined(__aarch64__)) && !defined(XXH_ENABLE_AUTOVECTORIZE) - /* - * UGLY HACK: - * A compiler fence is the only thing that prevents GCC and Clang from - * autovectorizing the XXH32 loop (pragmas and attributes don't work for some - * reason) without globally disabling SSE4.1. - * - * The reason we want to avoid vectorization is because despite working on - * 4 integers at a time, there are multiple factors slowing XXH32 down on - * SSE4: - * - There's a ridiculous amount of lag from pmulld (10 cycles of latency on - * newer chips!) making it slightly slower to multiply four integers at - * once compared to four integers independently. Even when pmulld was - * fastest, Sandy/Ivy Bridge, it is still not worth it to go into SSE - * just to multiply unless doing a long operation. - * - * - Four instructions are required to rotate, - * movqda tmp, v // not required with VEX encoding - * pslld tmp, 13 // tmp <<= 13 - * psrld v, 19 // x >>= 19 - * por v, tmp // x |= tmp - * compared to one for scalar: - * roll v, 13 // reliably fast across the board - * shldl v, v, 13 // Sandy Bridge and later prefer this for some reason - * - * - Instruction level parallelism is actually more beneficial here because - * the SIMD actually serializes this operation: While v1 is rotating, v2 - * can load data, while v3 can multiply. SSE forces them to operate - * together. - * - * This is also enabled on AArch64, as Clang autovectorizes it incorrectly - * and it is pointless writing a NEON implementation that is basically the - * same speed as scalar for XXH32. - */ - XXH_COMPILER_GUARD(acc); -#endif - return acc; -} - -/*! - * @internal - * @brief Mixes all bits to finalize the hash. - * - * The final mix ensures that all input bits have a chance to impact any bit in - * the output digest, resulting in an unbiased distribution. - * - * @param h32 The hash to avalanche. - * @return The avalanched hash. - */ -static xxh_u32 XXH32_avalanche(xxh_u32 h32) -{ - h32 ^= h32 >> 15; - h32 *= XXH_PRIME32_2; - h32 ^= h32 >> 13; - h32 *= XXH_PRIME32_3; - h32 ^= h32 >> 16; - return(h32); -} - -#define XXH_get32bits(p) XXH_readLE32_align(p, align) - -/*! - * @internal - * @brief Processes the last 0-15 bytes of @p ptr. - * - * There may be up to 15 bytes remaining to consume from the input. - * This final stage will digest them to ensure that all input bytes are present - * in the final mix. - * - * @param h32 The hash to finalize. - * @param ptr The pointer to the remaining input. - * @param len The remaining length, modulo 16. - * @param align Whether @p ptr is aligned. - * @return The finalized hash. - */ -static xxh_u32 -XXH32_finalize(xxh_u32 h32, const xxh_u8* ptr, size_t len, XXH_alignment align) -{ -#define XXH_PROCESS1 do { \ - h32 += (*ptr++) * XXH_PRIME32_5; \ - h32 = XXH_rotl32(h32, 11) * XXH_PRIME32_1; \ -} while (0) - -#define XXH_PROCESS4 do { \ - h32 += XXH_get32bits(ptr) * XXH_PRIME32_3; \ - ptr += 4; \ - h32 = XXH_rotl32(h32, 17) * XXH_PRIME32_4; \ -} while (0) - - if (ptr==NULL) XXH_ASSERT(len == 0); - - /* Compact rerolled version; generally faster */ - if (!XXH32_ENDJMP) { - len &= 15; - while (len >= 4) { - XXH_PROCESS4; - len -= 4; - } - while (len > 0) { - XXH_PROCESS1; - --len; - } - return XXH32_avalanche(h32); - } else { - switch(len&15) /* or switch(bEnd - p) */ { - case 12: XXH_PROCESS4; - XXH_FALLTHROUGH; - case 8: XXH_PROCESS4; - XXH_FALLTHROUGH; - case 4: XXH_PROCESS4; - return XXH32_avalanche(h32); - - case 13: XXH_PROCESS4; - XXH_FALLTHROUGH; - case 9: XXH_PROCESS4; - XXH_FALLTHROUGH; - case 5: XXH_PROCESS4; - XXH_PROCESS1; - return XXH32_avalanche(h32); - - case 14: XXH_PROCESS4; - XXH_FALLTHROUGH; - case 10: XXH_PROCESS4; - XXH_FALLTHROUGH; - case 6: XXH_PROCESS4; - XXH_PROCESS1; - XXH_PROCESS1; - return XXH32_avalanche(h32); - - case 15: XXH_PROCESS4; - XXH_FALLTHROUGH; - case 11: XXH_PROCESS4; - XXH_FALLTHROUGH; - case 7: XXH_PROCESS4; - XXH_FALLTHROUGH; - case 3: XXH_PROCESS1; - XXH_FALLTHROUGH; - case 2: XXH_PROCESS1; - XXH_FALLTHROUGH; - case 1: XXH_PROCESS1; - XXH_FALLTHROUGH; - case 0: return XXH32_avalanche(h32); - } - XXH_ASSERT(0); - return h32; /* reaching this point is deemed impossible */ - } -} - -#ifdef XXH_OLD_NAMES -# define PROCESS1 XXH_PROCESS1 -# define PROCESS4 XXH_PROCESS4 -#else -# undef XXH_PROCESS1 -# undef XXH_PROCESS4 -#endif - -/*! - * @internal - * @brief The implementation for @ref XXH32(). - * - * @param input , len , seed Directly passed from @ref XXH32(). - * @param align Whether @p input is aligned. - * @return The calculated hash. - */ -XXH_FORCE_INLINE xxh_u32 -XXH32_endian_align(const xxh_u8* input, size_t len, xxh_u32 seed, XXH_alignment align) -{ - xxh_u32 h32; - - if (input==NULL) XXH_ASSERT(len == 0); - - if (len>=16) { - const xxh_u8* const bEnd = input + len; - const xxh_u8* const limit = bEnd - 15; - xxh_u32 v1 = seed + XXH_PRIME32_1 + XXH_PRIME32_2; - xxh_u32 v2 = seed + XXH_PRIME32_2; - xxh_u32 v3 = seed + 0; - xxh_u32 v4 = seed - XXH_PRIME32_1; - - do { - v1 = XXH32_round(v1, XXH_get32bits(input)); input += 4; - v2 = XXH32_round(v2, XXH_get32bits(input)); input += 4; - v3 = XXH32_round(v3, XXH_get32bits(input)); input += 4; - v4 = XXH32_round(v4, XXH_get32bits(input)); input += 4; - } while (input < limit); - - h32 = XXH_rotl32(v1, 1) + XXH_rotl32(v2, 7) - + XXH_rotl32(v3, 12) + XXH_rotl32(v4, 18); - } else { - h32 = seed + XXH_PRIME32_5; - } - - h32 += (xxh_u32)len; - - return XXH32_finalize(h32, input, len&15, align); -} - -/*! @ingroup xxh32_family */ -XXH_PUBLIC_API XXH32_hash_t XXH32 (const void* input, size_t len, XXH32_hash_t seed) -{ -#if 0 - /* Simple version, good for code maintenance, but unfortunately slow for small inputs */ - XXH32_state_t state; - XXH32_reset(&state, seed); - XXH32_update(&state, (const xxh_u8*)input, len); - return XXH32_digest(&state); -#else - if (XXH_FORCE_ALIGN_CHECK) { - if ((((size_t)input) & 3) == 0) { /* Input is 4-bytes aligned, leverage the speed benefit */ - return XXH32_endian_align((const xxh_u8*)input, len, seed, XXH_aligned); - } } - - return XXH32_endian_align((const xxh_u8*)input, len, seed, XXH_unaligned); -#endif -} - - - -/******* Hash streaming *******/ -/*! - * @ingroup xxh32_family - */ -XXH_PUBLIC_API XXH32_state_t* XXH32_createState(void) -{ - return (XXH32_state_t*)XXH_malloc(sizeof(XXH32_state_t)); -} -/*! @ingroup xxh32_family */ -XXH_PUBLIC_API XXH_errorcode XXH32_freeState(XXH32_state_t* statePtr) -{ - XXH_free(statePtr); - return XXH_OK; -} - -/*! @ingroup xxh32_family */ -XXH_PUBLIC_API void XXH32_copyState(XXH32_state_t* dstState, const XXH32_state_t* srcState) -{ - XXH_memcpy(dstState, srcState, sizeof(*dstState)); -} - -/*! @ingroup xxh32_family */ -XXH_PUBLIC_API XXH_errorcode XXH32_reset(XXH32_state_t* statePtr, XXH32_hash_t seed) -{ - XXH_ASSERT(statePtr != NULL); - memset(statePtr, 0, sizeof(*statePtr)); - statePtr->v[0] = seed + XXH_PRIME32_1 + XXH_PRIME32_2; - statePtr->v[1] = seed + XXH_PRIME32_2; - statePtr->v[2] = seed + 0; - statePtr->v[3] = seed - XXH_PRIME32_1; - return XXH_OK; -} - - -/*! @ingroup xxh32_family */ -XXH_PUBLIC_API XXH_errorcode -XXH32_update(XXH32_state_t* state, const void* input, size_t len) -{ - if (input==NULL) { - XXH_ASSERT(len == 0); - return XXH_OK; - } - - { const xxh_u8* p = (const xxh_u8*)input; - const xxh_u8* const bEnd = p + len; - - state->total_len_32 += (XXH32_hash_t)len; - state->large_len |= (XXH32_hash_t)((len>=16) | (state->total_len_32>=16)); - - if (state->memsize + len < 16) { /* fill in tmp buffer */ - XXH_memcpy((xxh_u8*)(state->mem32) + state->memsize, input, len); - state->memsize += (XXH32_hash_t)len; - return XXH_OK; - } - - if (state->memsize) { /* some data left from previous update */ - XXH_memcpy((xxh_u8*)(state->mem32) + state->memsize, input, 16-state->memsize); - { const xxh_u32* p32 = state->mem32; - state->v[0] = XXH32_round(state->v[0], XXH_readLE32(p32)); p32++; - state->v[1] = XXH32_round(state->v[1], XXH_readLE32(p32)); p32++; - state->v[2] = XXH32_round(state->v[2], XXH_readLE32(p32)); p32++; - state->v[3] = XXH32_round(state->v[3], XXH_readLE32(p32)); - } - p += 16-state->memsize; - state->memsize = 0; - } - - if (p <= bEnd-16) { - const xxh_u8* const limit = bEnd - 16; - - do { - state->v[0] = XXH32_round(state->v[0], XXH_readLE32(p)); p+=4; - state->v[1] = XXH32_round(state->v[1], XXH_readLE32(p)); p+=4; - state->v[2] = XXH32_round(state->v[2], XXH_readLE32(p)); p+=4; - state->v[3] = XXH32_round(state->v[3], XXH_readLE32(p)); p+=4; - } while (p<=limit); - - } - - if (p < bEnd) { - XXH_memcpy(state->mem32, p, (size_t)(bEnd-p)); - state->memsize = (unsigned)(bEnd-p); - } - } - - return XXH_OK; -} - - -/*! @ingroup xxh32_family */ -XXH_PUBLIC_API XXH32_hash_t XXH32_digest(const XXH32_state_t* state) -{ - xxh_u32 h32; - - if (state->large_len) { - h32 = XXH_rotl32(state->v[0], 1) - + XXH_rotl32(state->v[1], 7) - + XXH_rotl32(state->v[2], 12) - + XXH_rotl32(state->v[3], 18); - } else { - h32 = state->v[2] /* == seed */ + XXH_PRIME32_5; - } - - h32 += state->total_len_32; - - return XXH32_finalize(h32, (const xxh_u8*)state->mem32, state->memsize, XXH_aligned); -} - - -/******* Canonical representation *******/ - -/*! - * @ingroup xxh32_family - * The default return values from XXH functions are unsigned 32 and 64 bit - * integers. - * - * The canonical representation uses big endian convention, the same convention - * as human-readable numbers (large digits first). - * - * This way, hash values can be written into a file or buffer, remaining - * comparable across different systems. - * - * The following functions allow transformation of hash values to and from their - * canonical format. - */ -XXH_PUBLIC_API void XXH32_canonicalFromHash(XXH32_canonical_t* dst, XXH32_hash_t hash) -{ - /* XXH_STATIC_ASSERT(sizeof(XXH32_canonical_t) == sizeof(XXH32_hash_t)); */ - if (XXH_CPU_LITTLE_ENDIAN) hash = XXH_swap32(hash); - XXH_memcpy(dst, &hash, sizeof(*dst)); -} -/*! @ingroup xxh32_family */ -XXH_PUBLIC_API XXH32_hash_t XXH32_hashFromCanonical(const XXH32_canonical_t* src) -{ - return XXH_readBE32(src); -} - - -#ifndef XXH_NO_LONG_LONG - -/* ******************************************************************* -* 64-bit hash functions -*********************************************************************/ -/*! - * @} - * @ingroup impl - * @{ - */ -/******* Memory access *******/ - -typedef XXH64_hash_t xxh_u64; - -#ifdef XXH_OLD_NAMES -# define U64 xxh_u64 -#endif - -#if (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==3)) -/* - * Manual byteshift. Best for old compilers which don't inline memcpy. - * We actually directly use XXH_readLE64 and XXH_readBE64. - */ -#elif (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==2)) - -/* Force direct memory access. Only works on CPU which support unaligned memory access in hardware */ -static xxh_u64 XXH_read64(const void* memPtr) -{ - return *(const xxh_u64*) memPtr; -} - -#elif (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==1)) - -/* - * __pack instructions are safer, but compiler specific, hence potentially - * problematic for some compilers. - * - * Currently only defined for GCC and ICC. - */ -#ifdef XXH_OLD_NAMES -typedef union { xxh_u32 u32; xxh_u64 u64; } __attribute__((packed)) unalign64; -#endif -static xxh_u64 XXH_read64(const void* ptr) -{ - typedef union { xxh_u32 u32; xxh_u64 u64; } __attribute__((packed)) xxh_unalign64; - return ((const xxh_unalign64*)ptr)->u64; -} - -#else - -/* - * Portable and safe solution. Generally efficient. - * see: http://fastcompression.blogspot.com/2015/08/accessing-unaligned-memory.html - */ -static xxh_u64 XXH_read64(const void* memPtr) -{ - xxh_u64 val; - XXH_memcpy(&val, memPtr, sizeof(val)); - return val; -} - -#endif /* XXH_FORCE_DIRECT_MEMORY_ACCESS */ - -#if defined(_MSC_VER) /* Visual Studio */ -# define XXH_swap64 _byteswap_uint64 -#elif XXH_GCC_VERSION >= 403 -# define XXH_swap64 __builtin_bswap64 -#else -static xxh_u64 XXH_swap64(xxh_u64 x) -{ - return ((x << 56) & 0xff00000000000000ULL) | - ((x << 40) & 0x00ff000000000000ULL) | - ((x << 24) & 0x0000ff0000000000ULL) | - ((x << 8) & 0x000000ff00000000ULL) | - ((x >> 8) & 0x00000000ff000000ULL) | - ((x >> 24) & 0x0000000000ff0000ULL) | - ((x >> 40) & 0x000000000000ff00ULL) | - ((x >> 56) & 0x00000000000000ffULL); -} -#endif - - -/* XXH_FORCE_MEMORY_ACCESS==3 is an endian-independent byteshift load. */ -#if (defined(XXH_FORCE_MEMORY_ACCESS) && (XXH_FORCE_MEMORY_ACCESS==3)) - -XXH_FORCE_INLINE xxh_u64 XXH_readLE64(const void* memPtr) -{ - const xxh_u8* bytePtr = (const xxh_u8 *)memPtr; - return bytePtr[0] - | ((xxh_u64)bytePtr[1] << 8) - | ((xxh_u64)bytePtr[2] << 16) - | ((xxh_u64)bytePtr[3] << 24) - | ((xxh_u64)bytePtr[4] << 32) - | ((xxh_u64)bytePtr[5] << 40) - | ((xxh_u64)bytePtr[6] << 48) - | ((xxh_u64)bytePtr[7] << 56); -} - -XXH_FORCE_INLINE xxh_u64 XXH_readBE64(const void* memPtr) -{ - const xxh_u8* bytePtr = (const xxh_u8 *)memPtr; - return bytePtr[7] - | ((xxh_u64)bytePtr[6] << 8) - | ((xxh_u64)bytePtr[5] << 16) - | ((xxh_u64)bytePtr[4] << 24) - | ((xxh_u64)bytePtr[3] << 32) - | ((xxh_u64)bytePtr[2] << 40) - | ((xxh_u64)bytePtr[1] << 48) - | ((xxh_u64)bytePtr[0] << 56); -} - -#else -XXH_FORCE_INLINE xxh_u64 XXH_readLE64(const void* ptr) -{ - return XXH_CPU_LITTLE_ENDIAN ? XXH_read64(ptr) : XXH_swap64(XXH_read64(ptr)); -} - -static xxh_u64 XXH_readBE64(const void* ptr) -{ - return XXH_CPU_LITTLE_ENDIAN ? XXH_swap64(XXH_read64(ptr)) : XXH_read64(ptr); -} -#endif - -XXH_FORCE_INLINE xxh_u64 -XXH_readLE64_align(const void* ptr, XXH_alignment align) -{ - if (align==XXH_unaligned) - return XXH_readLE64(ptr); - else - return XXH_CPU_LITTLE_ENDIAN ? *(const xxh_u64*)ptr : XXH_swap64(*(const xxh_u64*)ptr); -} - - -/******* xxh64 *******/ -/*! - * @} - * @defgroup xxh64_impl XXH64 implementation - * @ingroup impl - * @{ - */ -/* #define rather that static const, to be used as initializers */ -#define XXH_PRIME64_1 0x9E3779B185EBCA87ULL /*!< 0b1001111000110111011110011011000110000101111010111100101010000111 */ -#define XXH_PRIME64_2 0xC2B2AE3D27D4EB4FULL /*!< 0b1100001010110010101011100011110100100111110101001110101101001111 */ -#define XXH_PRIME64_3 0x165667B19E3779F9ULL /*!< 0b0001011001010110011001111011000110011110001101110111100111111001 */ -#define XXH_PRIME64_4 0x85EBCA77C2B2AE63ULL /*!< 0b1000010111101011110010100111011111000010101100101010111001100011 */ -#define XXH_PRIME64_5 0x27D4EB2F165667C5ULL /*!< 0b0010011111010100111010110010111100010110010101100110011111000101 */ - -#ifdef XXH_OLD_NAMES -# define PRIME64_1 XXH_PRIME64_1 -# define PRIME64_2 XXH_PRIME64_2 -# define PRIME64_3 XXH_PRIME64_3 -# define PRIME64_4 XXH_PRIME64_4 -# define PRIME64_5 XXH_PRIME64_5 -#endif - -static xxh_u64 XXH64_round(xxh_u64 acc, xxh_u64 input) -{ - acc += input * XXH_PRIME64_2; - acc = XXH_rotl64(acc, 31); - acc *= XXH_PRIME64_1; - return acc; -} - -static xxh_u64 XXH64_mergeRound(xxh_u64 acc, xxh_u64 val) -{ - val = XXH64_round(0, val); - acc ^= val; - acc = acc * XXH_PRIME64_1 + XXH_PRIME64_4; - return acc; -} - -static xxh_u64 XXH64_avalanche(xxh_u64 h64) -{ - h64 ^= h64 >> 33; - h64 *= XXH_PRIME64_2; - h64 ^= h64 >> 29; - h64 *= XXH_PRIME64_3; - h64 ^= h64 >> 32; - return h64; -} - - -#define XXH_get64bits(p) XXH_readLE64_align(p, align) - -static xxh_u64 -XXH64_finalize(xxh_u64 h64, const xxh_u8* ptr, size_t len, XXH_alignment align) -{ - if (ptr==NULL) XXH_ASSERT(len == 0); - len &= 31; - while (len >= 8) { - xxh_u64 const k1 = XXH64_round(0, XXH_get64bits(ptr)); - ptr += 8; - h64 ^= k1; - h64 = XXH_rotl64(h64,27) * XXH_PRIME64_1 + XXH_PRIME64_4; - len -= 8; - } - if (len >= 4) { - h64 ^= (xxh_u64)(XXH_get32bits(ptr)) * XXH_PRIME64_1; - ptr += 4; - h64 = XXH_rotl64(h64, 23) * XXH_PRIME64_2 + XXH_PRIME64_3; - len -= 4; - } - while (len > 0) { - h64 ^= (*ptr++) * XXH_PRIME64_5; - h64 = XXH_rotl64(h64, 11) * XXH_PRIME64_1; - --len; - } - return XXH64_avalanche(h64); -} - -#ifdef XXH_OLD_NAMES -# define PROCESS1_64 XXH_PROCESS1_64 -# define PROCESS4_64 XXH_PROCESS4_64 -# define PROCESS8_64 XXH_PROCESS8_64 -#else -# undef XXH_PROCESS1_64 -# undef XXH_PROCESS4_64 -# undef XXH_PROCESS8_64 -#endif - -XXH_FORCE_INLINE xxh_u64 -XXH64_endian_align(const xxh_u8* input, size_t len, xxh_u64 seed, XXH_alignment align) -{ - xxh_u64 h64; - if (input==NULL) XXH_ASSERT(len == 0); - - if (len>=32) { - const xxh_u8* const bEnd = input + len; - const xxh_u8* const limit = bEnd - 31; - xxh_u64 v1 = seed + XXH_PRIME64_1 + XXH_PRIME64_2; - xxh_u64 v2 = seed + XXH_PRIME64_2; - xxh_u64 v3 = seed + 0; - xxh_u64 v4 = seed - XXH_PRIME64_1; - - do { - v1 = XXH64_round(v1, XXH_get64bits(input)); input+=8; - v2 = XXH64_round(v2, XXH_get64bits(input)); input+=8; - v3 = XXH64_round(v3, XXH_get64bits(input)); input+=8; - v4 = XXH64_round(v4, XXH_get64bits(input)); input+=8; - } while (inputv[0] = seed + XXH_PRIME64_1 + XXH_PRIME64_2; - statePtr->v[1] = seed + XXH_PRIME64_2; - statePtr->v[2] = seed + 0; - statePtr->v[3] = seed - XXH_PRIME64_1; - return XXH_OK; -} - -/*! @ingroup xxh64_family */ -XXH_PUBLIC_API XXH_errorcode -XXH64_update (XXH64_state_t* state, const void* input, size_t len) -{ - if (input==NULL) { - XXH_ASSERT(len == 0); - return XXH_OK; - } - - { const xxh_u8* p = (const xxh_u8*)input; - const xxh_u8* const bEnd = p + len; - - state->total_len += len; - - if (state->memsize + len < 32) { /* fill in tmp buffer */ - XXH_memcpy(((xxh_u8*)state->mem64) + state->memsize, input, len); - state->memsize += (xxh_u32)len; - return XXH_OK; - } - - if (state->memsize) { /* tmp buffer is full */ - XXH_memcpy(((xxh_u8*)state->mem64) + state->memsize, input, 32-state->memsize); - state->v[0] = XXH64_round(state->v[0], XXH_readLE64(state->mem64+0)); - state->v[1] = XXH64_round(state->v[1], XXH_readLE64(state->mem64+1)); - state->v[2] = XXH64_round(state->v[2], XXH_readLE64(state->mem64+2)); - state->v[3] = XXH64_round(state->v[3], XXH_readLE64(state->mem64+3)); - p += 32 - state->memsize; - state->memsize = 0; - } - - if (p+32 <= bEnd) { - const xxh_u8* const limit = bEnd - 32; - - do { - state->v[0] = XXH64_round(state->v[0], XXH_readLE64(p)); p+=8; - state->v[1] = XXH64_round(state->v[1], XXH_readLE64(p)); p+=8; - state->v[2] = XXH64_round(state->v[2], XXH_readLE64(p)); p+=8; - state->v[3] = XXH64_round(state->v[3], XXH_readLE64(p)); p+=8; - } while (p<=limit); - - } - - if (p < bEnd) { - XXH_memcpy(state->mem64, p, (size_t)(bEnd-p)); - state->memsize = (unsigned)(bEnd-p); - } - } - - return XXH_OK; -} - - -/*! @ingroup xxh64_family */ -XXH_PUBLIC_API XXH64_hash_t XXH64_digest(const XXH64_state_t* state) -{ - xxh_u64 h64; - - if (state->total_len >= 32) { - h64 = XXH_rotl64(state->v[0], 1) + XXH_rotl64(state->v[1], 7) + XXH_rotl64(state->v[2], 12) + XXH_rotl64(state->v[3], 18); - h64 = XXH64_mergeRound(h64, state->v[0]); - h64 = XXH64_mergeRound(h64, state->v[1]); - h64 = XXH64_mergeRound(h64, state->v[2]); - h64 = XXH64_mergeRound(h64, state->v[3]); - } else { - h64 = state->v[2] /*seed*/ + XXH_PRIME64_5; - } - - h64 += (xxh_u64) state->total_len; - - return XXH64_finalize(h64, (const xxh_u8*)state->mem64, (size_t)state->total_len, XXH_aligned); -} - - -/******* Canonical representation *******/ - -/*! @ingroup xxh64_family */ -XXH_PUBLIC_API void XXH64_canonicalFromHash(XXH64_canonical_t* dst, XXH64_hash_t hash) -{ - /* XXH_STATIC_ASSERT(sizeof(XXH64_canonical_t) == sizeof(XXH64_hash_t)); */ - if (XXH_CPU_LITTLE_ENDIAN) hash = XXH_swap64(hash); - XXH_memcpy(dst, &hash, sizeof(*dst)); -} - -/*! @ingroup xxh64_family */ -XXH_PUBLIC_API XXH64_hash_t XXH64_hashFromCanonical(const XXH64_canonical_t* src) -{ - return XXH_readBE64(src); -} - -#ifndef XXH_NO_XXH3 - -/* ********************************************************************* -* XXH3 -* New generation hash designed for speed on small keys and vectorization -************************************************************************ */ -/*! - * @} - * @defgroup xxh3_impl XXH3 implementation - * @ingroup impl - * @{ - */ - -/* === Compiler specifics === */ - -#if ((defined(sun) || defined(__sun)) && __cplusplus) /* Solaris includes __STDC_VERSION__ with C++. Tested with GCC 5.5 */ -# define XXH_RESTRICT /* disable */ -#elif defined (__STDC_VERSION__) && __STDC_VERSION__ >= 199901L /* >= C99 */ -# define XXH_RESTRICT restrict -#else -/* Note: it might be useful to define __restrict or __restrict__ for some C++ compilers */ -# define XXH_RESTRICT /* disable */ -#endif - -#if (defined(__GNUC__) && (__GNUC__ >= 3)) \ - || (defined(__INTEL_COMPILER) && (__INTEL_COMPILER >= 800)) \ - || defined(__clang__) -# define XXH_likely(x) __builtin_expect(x, 1) -# define XXH_unlikely(x) __builtin_expect(x, 0) -#else -# define XXH_likely(x) (x) -# define XXH_unlikely(x) (x) -#endif - -#if defined(__GNUC__) || defined(__clang__) -# if defined(__ARM_NEON__) || defined(__ARM_NEON) \ - || defined(__aarch64__) || defined(_M_ARM) \ - || defined(_M_ARM64) || defined(_M_ARM64EC) -# define inline __inline__ /* circumvent a clang bug */ -# include -# undef inline -# elif defined(__AVX2__) -# include -# elif defined(__SSE2__) -# include -# endif -#endif - -#if defined(_MSC_VER) -# include -#endif - -/* - * One goal of XXH3 is to make it fast on both 32-bit and 64-bit, while - * remaining a true 64-bit/128-bit hash function. - * - * This is done by prioritizing a subset of 64-bit operations that can be - * emulated without too many steps on the average 32-bit machine. - * - * For example, these two lines seem similar, and run equally fast on 64-bit: - * - * xxh_u64 x; - * x ^= (x >> 47); // good - * x ^= (x >> 13); // bad - * - * However, to a 32-bit machine, there is a major difference. - * - * x ^= (x >> 47) looks like this: - * - * x.lo ^= (x.hi >> (47 - 32)); - * - * while x ^= (x >> 13) looks like this: - * - * // note: funnel shifts are not usually cheap. - * x.lo ^= (x.lo >> 13) | (x.hi << (32 - 13)); - * x.hi ^= (x.hi >> 13); - * - * The first one is significantly faster than the second, simply because the - * shift is larger than 32. This means: - * - All the bits we need are in the upper 32 bits, so we can ignore the lower - * 32 bits in the shift. - * - The shift result will always fit in the lower 32 bits, and therefore, - * we can ignore the upper 32 bits in the xor. - * - * Thanks to this optimization, XXH3 only requires these features to be efficient: - * - * - Usable unaligned access - * - A 32-bit or 64-bit ALU - * - If 32-bit, a decent ADC instruction - * - A 32 or 64-bit multiply with a 64-bit result - * - For the 128-bit variant, a decent byteswap helps short inputs. - * - * The first two are already required by XXH32, and almost all 32-bit and 64-bit - * platforms which can run XXH32 can run XXH3 efficiently. - * - * Thumb-1, the classic 16-bit only subset of ARM's instruction set, is one - * notable exception. - * - * First of all, Thumb-1 lacks support for the UMULL instruction which - * performs the important long multiply. This means numerous __aeabi_lmul - * calls. - * - * Second of all, the 8 functional registers are just not enough. - * Setup for __aeabi_lmul, byteshift loads, pointers, and all arithmetic need - * Lo registers, and this shuffling results in thousands more MOVs than A32. - * - * A32 and T32 don't have this limitation. They can access all 14 registers, - * do a 32->64 multiply with UMULL, and the flexible operand allowing free - * shifts is helpful, too. - * - * Therefore, we do a quick sanity check. - * - * If compiling Thumb-1 for a target which supports ARM instructions, we will - * emit a warning, as it is not a "sane" platform to compile for. - * - * Usually, if this happens, it is because of an accident and you probably need - * to specify -march, as you likely meant to compile for a newer architecture. - * - * Credit: large sections of the vectorial and asm source code paths - * have been contributed by @easyaspi314 - */ -#if defined(__thumb__) && !defined(__thumb2__) && defined(__ARM_ARCH_ISA_ARM) -# warning "XXH3 is highly inefficient without ARM or Thumb-2." -#endif - -/* ========================================== - * Vectorization detection - * ========================================== */ - -#ifdef XXH_DOXYGEN -/*! - * @ingroup tuning - * @brief Overrides the vectorization implementation chosen for XXH3. - * - * Can be defined to 0 to disable SIMD or any of the values mentioned in - * @ref XXH_VECTOR_TYPE. - * - * If this is not defined, it uses predefined macros to determine the best - * implementation. - */ -# define XXH_VECTOR XXH_SCALAR -/*! - * @ingroup tuning - * @brief Possible values for @ref XXH_VECTOR. - * - * Note that these are actually implemented as macros. - * - * If this is not defined, it is detected automatically. - * @ref XXH_X86DISPATCH overrides this. - */ -enum XXH_VECTOR_TYPE /* fake enum */ { - XXH_SCALAR = 0, /*!< Portable scalar version */ - XXH_SSE2 = 1, /*!< - * SSE2 for Pentium 4, Opteron, all x86_64. - * - * @note SSE2 is also guaranteed on Windows 10, macOS, and - * Android x86. - */ - XXH_AVX2 = 2, /*!< AVX2 for Haswell and Bulldozer */ - XXH_AVX512 = 3, /*!< AVX512 for Skylake and Icelake */ - XXH_NEON = 4, /*!< NEON for most ARMv7-A and all AArch64 */ - XXH_VSX = 5, /*!< VSX and ZVector for POWER8/z13 (64-bit) */ -}; -/*! - * @ingroup tuning - * @brief Selects the minimum alignment for XXH3's accumulators. - * - * When using SIMD, this should match the alignment reqired for said vector - * type, so, for example, 32 for AVX2. - * - * Default: Auto detected. - */ -# define XXH_ACC_ALIGN 8 -#endif - -/* Actual definition */ -#ifndef XXH_DOXYGEN -# define XXH_SCALAR 0 -# define XXH_SSE2 1 -# define XXH_AVX2 2 -# define XXH_AVX512 3 -# define XXH_NEON 4 -# define XXH_VSX 5 -#endif - -#ifndef XXH_VECTOR /* can be defined on command line */ -# if ( \ - defined(__ARM_NEON__) || defined(__ARM_NEON) /* gcc */ \ - || defined(_M_ARM) || defined(_M_ARM64) || defined(_M_ARM64EC) /* msvc */ \ - ) && ( \ - defined(_WIN32) || defined(__LITTLE_ENDIAN__) /* little endian only */ \ - || (defined(__BYTE_ORDER__) && __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__) \ - ) -# define XXH_VECTOR XXH_NEON -# elif defined(__AVX512F__) -# define XXH_VECTOR XXH_AVX512 -# elif defined(__AVX2__) -# define XXH_VECTOR XXH_AVX2 -# elif defined(__SSE2__) || defined(_M_AMD64) || defined(_M_X64) || (defined(_M_IX86_FP) && (_M_IX86_FP == 2)) -# define XXH_VECTOR XXH_SSE2 -# elif (defined(__PPC64__) && defined(__POWER8_VECTOR__)) \ - || (defined(__s390x__) && defined(__VEC__)) \ - && defined(__GNUC__) /* TODO: IBM XL */ -# define XXH_VECTOR XXH_VSX -# else -# define XXH_VECTOR XXH_SCALAR -# endif -#endif - -/* - * Controls the alignment of the accumulator, - * for compatibility with aligned vector loads, which are usually faster. - */ -#ifndef XXH_ACC_ALIGN -# if defined(XXH_X86DISPATCH) -# define XXH_ACC_ALIGN 64 /* for compatibility with avx512 */ -# elif XXH_VECTOR == XXH_SCALAR /* scalar */ -# define XXH_ACC_ALIGN 8 -# elif XXH_VECTOR == XXH_SSE2 /* sse2 */ -# define XXH_ACC_ALIGN 16 -# elif XXH_VECTOR == XXH_AVX2 /* avx2 */ -# define XXH_ACC_ALIGN 32 -# elif XXH_VECTOR == XXH_NEON /* neon */ -# define XXH_ACC_ALIGN 16 -# elif XXH_VECTOR == XXH_VSX /* vsx */ -# define XXH_ACC_ALIGN 16 -# elif XXH_VECTOR == XXH_AVX512 /* avx512 */ -# define XXH_ACC_ALIGN 64 -# endif -#endif - -#if defined(XXH_X86DISPATCH) || XXH_VECTOR == XXH_SSE2 \ - || XXH_VECTOR == XXH_AVX2 || XXH_VECTOR == XXH_AVX512 -# define XXH_SEC_ALIGN XXH_ACC_ALIGN -#else -# define XXH_SEC_ALIGN 8 -#endif - -/* - * UGLY HACK: - * GCC usually generates the best code with -O3 for xxHash. - * - * However, when targeting AVX2, it is overzealous in its unrolling resulting - * in code roughly 3/4 the speed of Clang. - * - * There are other issues, such as GCC splitting _mm256_loadu_si256 into - * _mm_loadu_si128 + _mm256_inserti128_si256. This is an optimization which - * only applies to Sandy and Ivy Bridge... which don't even support AVX2. - * - * That is why when compiling the AVX2 version, it is recommended to use either - * -O2 -mavx2 -march=haswell - * or - * -O2 -mavx2 -mno-avx256-split-unaligned-load - * for decent performance, or to use Clang instead. - * - * Fortunately, we can control the first one with a pragma that forces GCC into - * -O2, but the other one we can't control without "failed to inline always - * inline function due to target mismatch" warnings. - */ -#if XXH_VECTOR == XXH_AVX2 /* AVX2 */ \ - && defined(__GNUC__) && !defined(__clang__) /* GCC, not Clang */ \ - && defined(__OPTIMIZE__) && !defined(__OPTIMIZE_SIZE__) /* respect -O0 and -Os */ -# pragma GCC push_options -# pragma GCC optimize("-O2") -#endif - - -#if XXH_VECTOR == XXH_NEON -/* - * NEON's setup for vmlal_u32 is a little more complicated than it is on - * SSE2, AVX2, and VSX. - * - * While PMULUDQ and VMULEUW both perform a mask, VMLAL.U32 performs an upcast. - * - * To do the same operation, the 128-bit 'Q' register needs to be split into - * two 64-bit 'D' registers, performing this operation:: - * - * [ a | b ] - * | '---------. .--------' | - * | x | - * | .---------' '--------. | - * [ a & 0xFFFFFFFF | b & 0xFFFFFFFF ],[ a >> 32 | b >> 32 ] - * - * Due to significant changes in aarch64, the fastest method for aarch64 is - * completely different than the fastest method for ARMv7-A. - * - * ARMv7-A treats D registers as unions overlaying Q registers, so modifying - * D11 will modify the high half of Q5. This is similar to how modifying AH - * will only affect bits 8-15 of AX on x86. - * - * VZIP takes two registers, and puts even lanes in one register and odd lanes - * in the other. - * - * On ARMv7-A, this strangely modifies both parameters in place instead of - * taking the usual 3-operand form. - * - * Therefore, if we want to do this, we can simply use a D-form VZIP.32 on the - * lower and upper halves of the Q register to end up with the high and low - * halves where we want - all in one instruction. - * - * vzip.32 d10, d11 @ d10 = { d10[0], d11[0] }; d11 = { d10[1], d11[1] } - * - * Unfortunately we need inline assembly for this: Instructions modifying two - * registers at once is not possible in GCC or Clang's IR, and they have to - * create a copy. - * - * aarch64 requires a different approach. - * - * In order to make it easier to write a decent compiler for aarch64, many - * quirks were removed, such as conditional execution. - * - * NEON was also affected by this. - * - * aarch64 cannot access the high bits of a Q-form register, and writes to a - * D-form register zero the high bits, similar to how writes to W-form scalar - * registers (or DWORD registers on x86_64) work. - * - * The formerly free vget_high intrinsics now require a vext (with a few - * exceptions) - * - * Additionally, VZIP was replaced by ZIP1 and ZIP2, which are the equivalent - * of PUNPCKL* and PUNPCKH* in SSE, respectively, in order to only modify one - * operand. - * - * The equivalent of the VZIP.32 on the lower and upper halves would be this - * mess: - * - * ext v2.4s, v0.4s, v0.4s, #2 // v2 = { v0[2], v0[3], v0[0], v0[1] } - * zip1 v1.2s, v0.2s, v2.2s // v1 = { v0[0], v2[0] } - * zip2 v0.2s, v0.2s, v1.2s // v0 = { v0[1], v2[1] } - * - * Instead, we use a literal downcast, vmovn_u64 (XTN), and vshrn_n_u64 (SHRN): - * - * shrn v1.2s, v0.2d, #32 // v1 = (uint32x2_t)(v0 >> 32); - * xtn v0.2s, v0.2d // v0 = (uint32x2_t)(v0 & 0xFFFFFFFF); - * - * This is available on ARMv7-A, but is less efficient than a single VZIP.32. - */ - -/*! - * Function-like macro: - * void XXH_SPLIT_IN_PLACE(uint64x2_t &in, uint32x2_t &outLo, uint32x2_t &outHi) - * { - * outLo = (uint32x2_t)(in & 0xFFFFFFFF); - * outHi = (uint32x2_t)(in >> 32); - * in = UNDEFINED; - * } - */ -# if !defined(XXH_NO_VZIP_HACK) /* define to disable */ \ - && (defined(__GNUC__) || defined(__clang__)) \ - && (defined(__arm__) || defined(__thumb__) || defined(_M_ARM)) -# define XXH_SPLIT_IN_PLACE(in, outLo, outHi) \ - do { \ - /* Undocumented GCC/Clang operand modifier: %e0 = lower D half, %f0 = upper D half */ \ - /* https://github.com/gcc-mirror/gcc/blob/38cf91e5/gcc/config/arm/arm.c#L22486 */ \ - /* https://github.com/llvm-mirror/llvm/blob/2c4ca683/lib/Target/ARM/ARMAsmPrinter.cpp#L399 */ \ - __asm__("vzip.32 %e0, %f0" : "+w" (in)); \ - (outLo) = vget_low_u32 (vreinterpretq_u32_u64(in)); \ - (outHi) = vget_high_u32(vreinterpretq_u32_u64(in)); \ - } while (0) -# else -# define XXH_SPLIT_IN_PLACE(in, outLo, outHi) \ - do { \ - (outLo) = vmovn_u64 (in); \ - (outHi) = vshrn_n_u64 ((in), 32); \ - } while (0) -# endif - -/*! - * @ingroup tuning - * @brief Controls the NEON to scalar ratio for XXH3 - * - * On AArch64 when not optimizing for size, XXH3 will run 6 lanes using NEON and - * 2 lanes on scalar by default. - * - * This can be set to 2, 4, 6, or 8. ARMv7 will default to all 8 NEON lanes, as the - * emulated 64-bit arithmetic is too slow. - * - * Modern ARM CPUs are _very_ sensitive to how their pipelines are used. - * - * For example, the Cortex-A73 can dispatch 3 micro-ops per cycle, but it can't - * have more than 2 NEON (F0/F1) micro-ops. If you are only using NEON instructions, - * you are only using 2/3 of the CPU bandwidth. - * - * This is even more noticable on the more advanced cores like the A76 which - * can dispatch 8 micro-ops per cycle, but still only 2 NEON micro-ops at once. - * - * Therefore, @ref XXH3_NEON_LANES lanes will be processed using NEON, and the - * remaining lanes will use scalar instructions. This improves the bandwidth - * and also gives the integer pipelines something to do besides twiddling loop - * counters and pointers. - * - * This change benefits CPUs with large micro-op buffers without negatively affecting - * other CPUs: - * - * | Chipset | Dispatch type | NEON only | 6:2 hybrid | Diff. | - * |:----------------------|:--------------------|----------:|-----------:|------:| - * | Snapdragon 730 (A76) | 2 NEON/8 micro-ops | 8.8 GB/s | 10.1 GB/s | ~16% | - * | Snapdragon 835 (A73) | 2 NEON/3 micro-ops | 5.1 GB/s | 5.3 GB/s | ~5% | - * | Marvell PXA1928 (A53) | In-order dual-issue | 1.9 GB/s | 1.9 GB/s | 0% | - * - * It also seems to fix some bad codegen on GCC, making it almost as fast as clang. - * - * @see XXH3_accumulate_512_neon() - */ -# ifndef XXH3_NEON_LANES -# if (defined(__aarch64__) || defined(__arm64__) || defined(_M_ARM64) || defined(_M_ARM64EC)) \ - && !defined(__OPTIMIZE_SIZE__) -# define XXH3_NEON_LANES 6 -# else -# define XXH3_NEON_LANES XXH_ACC_NB -# endif -# endif -#endif /* XXH_VECTOR == XXH_NEON */ - -/* - * VSX and Z Vector helpers. - * - * This is very messy, and any pull requests to clean this up are welcome. - * - * There are a lot of problems with supporting VSX and s390x, due to - * inconsistent intrinsics, spotty coverage, and multiple endiannesses. - */ -#if XXH_VECTOR == XXH_VSX -# if defined(__s390x__) -# include -# else -/* gcc's altivec.h can have the unwanted consequence to unconditionally - * #define bool, vector, and pixel keywords, - * with bad consequences for programs already using these keywords for other purposes. - * The paragraph defining these macros is skipped when __APPLE_ALTIVEC__ is defined. - * __APPLE_ALTIVEC__ is _generally_ defined automatically by the compiler, - * but it seems that, in some cases, it isn't. - * Force the build macro to be defined, so that keywords are not altered. - */ -# if defined(__GNUC__) && !defined(__APPLE_ALTIVEC__) -# define __APPLE_ALTIVEC__ -# endif -# include -# endif - -typedef __vector unsigned long long xxh_u64x2; -typedef __vector unsigned char xxh_u8x16; -typedef __vector unsigned xxh_u32x4; - -# ifndef XXH_VSX_BE -# if defined(__BIG_ENDIAN__) \ - || (defined(__BYTE_ORDER__) && __BYTE_ORDER__ == __ORDER_BIG_ENDIAN__) -# define XXH_VSX_BE 1 -# elif defined(__VEC_ELEMENT_REG_ORDER__) && __VEC_ELEMENT_REG_ORDER__ == __ORDER_BIG_ENDIAN__ -# warning "-maltivec=be is not recommended. Please use native endianness." -# define XXH_VSX_BE 1 -# else -# define XXH_VSX_BE 0 -# endif -# endif /* !defined(XXH_VSX_BE) */ - -# if XXH_VSX_BE -# if defined(__POWER9_VECTOR__) || (defined(__clang__) && defined(__s390x__)) -# define XXH_vec_revb vec_revb -# else -/*! - * A polyfill for POWER9's vec_revb(). - */ -XXH_FORCE_INLINE xxh_u64x2 XXH_vec_revb(xxh_u64x2 val) -{ - xxh_u8x16 const vByteSwap = { 0x07, 0x06, 0x05, 0x04, 0x03, 0x02, 0x01, 0x00, - 0x0F, 0x0E, 0x0D, 0x0C, 0x0B, 0x0A, 0x09, 0x08 }; - return vec_perm(val, val, vByteSwap); -} -# endif -# endif /* XXH_VSX_BE */ - -/*! - * Performs an unaligned vector load and byte swaps it on big endian. - */ -XXH_FORCE_INLINE xxh_u64x2 XXH_vec_loadu(const void *ptr) -{ - xxh_u64x2 ret; - XXH_memcpy(&ret, ptr, sizeof(xxh_u64x2)); -# if XXH_VSX_BE - ret = XXH_vec_revb(ret); -# endif - return ret; -} - -/* - * vec_mulo and vec_mule are very problematic intrinsics on PowerPC - * - * These intrinsics weren't added until GCC 8, despite existing for a while, - * and they are endian dependent. Also, their meaning swap depending on version. - * */ -# if defined(__s390x__) - /* s390x is always big endian, no issue on this platform */ -# define XXH_vec_mulo vec_mulo -# define XXH_vec_mule vec_mule -# elif defined(__clang__) && XXH_HAS_BUILTIN(__builtin_altivec_vmuleuw) -/* Clang has a better way to control this, we can just use the builtin which doesn't swap. */ -# define XXH_vec_mulo __builtin_altivec_vmulouw -# define XXH_vec_mule __builtin_altivec_vmuleuw -# else -/* gcc needs inline assembly */ -/* Adapted from https://github.com/google/highwayhash/blob/master/highwayhash/hh_vsx.h. */ -XXH_FORCE_INLINE xxh_u64x2 XXH_vec_mulo(xxh_u32x4 a, xxh_u32x4 b) -{ - xxh_u64x2 result; - __asm__("vmulouw %0, %1, %2" : "=v" (result) : "v" (a), "v" (b)); - return result; -} -XXH_FORCE_INLINE xxh_u64x2 XXH_vec_mule(xxh_u32x4 a, xxh_u32x4 b) -{ - xxh_u64x2 result; - __asm__("vmuleuw %0, %1, %2" : "=v" (result) : "v" (a), "v" (b)); - return result; -} -# endif /* XXH_vec_mulo, XXH_vec_mule */ -#endif /* XXH_VECTOR == XXH_VSX */ - - -/* prefetch - * can be disabled, by declaring XXH_NO_PREFETCH build macro */ -#if defined(XXH_NO_PREFETCH) -# define XXH_PREFETCH(ptr) (void)(ptr) /* disabled */ -#else -# if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_IX86)) /* _mm_prefetch() not defined outside of x86/x64 */ -# include /* https://msdn.microsoft.com/fr-fr/library/84szxsww(v=vs.90).aspx */ -# define XXH_PREFETCH(ptr) _mm_prefetch((const char*)(ptr), _MM_HINT_T0) -# elif defined(__GNUC__) && ( (__GNUC__ >= 4) || ( (__GNUC__ == 3) && (__GNUC_MINOR__ >= 1) ) ) -# define XXH_PREFETCH(ptr) __builtin_prefetch((ptr), 0 /* rw==read */, 3 /* locality */) -# else -# define XXH_PREFETCH(ptr) (void)(ptr) /* disabled */ -# endif -#endif /* XXH_NO_PREFETCH */ - - -/* ========================================== - * XXH3 default settings - * ========================================== */ - -#define XXH_SECRET_DEFAULT_SIZE 192 /* minimum XXH3_SECRET_SIZE_MIN */ - -#if (XXH_SECRET_DEFAULT_SIZE < XXH3_SECRET_SIZE_MIN) -# error "default keyset is not large enough" -#endif - -/*! Pseudorandom secret taken directly from FARSH. */ -XXH_ALIGN(64) static const xxh_u8 XXH3_kSecret[XXH_SECRET_DEFAULT_SIZE] = { - 0xb8, 0xfe, 0x6c, 0x39, 0x23, 0xa4, 0x4b, 0xbe, 0x7c, 0x01, 0x81, 0x2c, 0xf7, 0x21, 0xad, 0x1c, - 0xde, 0xd4, 0x6d, 0xe9, 0x83, 0x90, 0x97, 0xdb, 0x72, 0x40, 0xa4, 0xa4, 0xb7, 0xb3, 0x67, 0x1f, - 0xcb, 0x79, 0xe6, 0x4e, 0xcc, 0xc0, 0xe5, 0x78, 0x82, 0x5a, 0xd0, 0x7d, 0xcc, 0xff, 0x72, 0x21, - 0xb8, 0x08, 0x46, 0x74, 0xf7, 0x43, 0x24, 0x8e, 0xe0, 0x35, 0x90, 0xe6, 0x81, 0x3a, 0x26, 0x4c, - 0x3c, 0x28, 0x52, 0xbb, 0x91, 0xc3, 0x00, 0xcb, 0x88, 0xd0, 0x65, 0x8b, 0x1b, 0x53, 0x2e, 0xa3, - 0x71, 0x64, 0x48, 0x97, 0xa2, 0x0d, 0xf9, 0x4e, 0x38, 0x19, 0xef, 0x46, 0xa9, 0xde, 0xac, 0xd8, - 0xa8, 0xfa, 0x76, 0x3f, 0xe3, 0x9c, 0x34, 0x3f, 0xf9, 0xdc, 0xbb, 0xc7, 0xc7, 0x0b, 0x4f, 0x1d, - 0x8a, 0x51, 0xe0, 0x4b, 0xcd, 0xb4, 0x59, 0x31, 0xc8, 0x9f, 0x7e, 0xc9, 0xd9, 0x78, 0x73, 0x64, - 0xea, 0xc5, 0xac, 0x83, 0x34, 0xd3, 0xeb, 0xc3, 0xc5, 0x81, 0xa0, 0xff, 0xfa, 0x13, 0x63, 0xeb, - 0x17, 0x0d, 0xdd, 0x51, 0xb7, 0xf0, 0xda, 0x49, 0xd3, 0x16, 0x55, 0x26, 0x29, 0xd4, 0x68, 0x9e, - 0x2b, 0x16, 0xbe, 0x58, 0x7d, 0x47, 0xa1, 0xfc, 0x8f, 0xf8, 0xb8, 0xd1, 0x7a, 0xd0, 0x31, 0xce, - 0x45, 0xcb, 0x3a, 0x8f, 0x95, 0x16, 0x04, 0x28, 0xaf, 0xd7, 0xfb, 0xca, 0xbb, 0x4b, 0x40, 0x7e, -}; - - -#ifdef XXH_OLD_NAMES -# define kSecret XXH3_kSecret -#endif - -#ifdef XXH_DOXYGEN -/*! - * @brief Calculates a 32-bit to 64-bit long multiply. - * - * Implemented as a macro. - * - * Wraps `__emulu` on MSVC x86 because it tends to call `__allmul` when it doesn't - * need to (but it shouldn't need to anyways, it is about 7 instructions to do - * a 64x64 multiply...). Since we know that this will _always_ emit `MULL`, we - * use that instead of the normal method. - * - * If you are compiling for platforms like Thumb-1 and don't have a better option, - * you may also want to write your own long multiply routine here. - * - * @param x, y Numbers to be multiplied - * @return 64-bit product of the low 32 bits of @p x and @p y. - */ -XXH_FORCE_INLINE xxh_u64 -XXH_mult32to64(xxh_u64 x, xxh_u64 y) -{ - return (x & 0xFFFFFFFF) * (y & 0xFFFFFFFF); -} -#elif defined(_MSC_VER) && defined(_M_IX86) -# define XXH_mult32to64(x, y) __emulu((unsigned)(x), (unsigned)(y)) -#else -/* - * Downcast + upcast is usually better than masking on older compilers like - * GCC 4.2 (especially 32-bit ones), all without affecting newer compilers. - * - * The other method, (x & 0xFFFFFFFF) * (y & 0xFFFFFFFF), will AND both operands - * and perform a full 64x64 multiply -- entirely redundant on 32-bit. - */ -# define XXH_mult32to64(x, y) ((xxh_u64)(xxh_u32)(x) * (xxh_u64)(xxh_u32)(y)) -#endif - -/*! - * @brief Calculates a 64->128-bit long multiply. - * - * Uses `__uint128_t` and `_umul128` if available, otherwise uses a scalar - * version. - * - * @param lhs , rhs The 64-bit integers to be multiplied - * @return The 128-bit result represented in an @ref XXH128_hash_t. - */ -static XXH128_hash_t -XXH_mult64to128(xxh_u64 lhs, xxh_u64 rhs) -{ - /* - * GCC/Clang __uint128_t method. - * - * On most 64-bit targets, GCC and Clang define a __uint128_t type. - * This is usually the best way as it usually uses a native long 64-bit - * multiply, such as MULQ on x86_64 or MUL + UMULH on aarch64. - * - * Usually. - * - * Despite being a 32-bit platform, Clang (and emscripten) define this type - * despite not having the arithmetic for it. This results in a laggy - * compiler builtin call which calculates a full 128-bit multiply. - * In that case it is best to use the portable one. - * https://github.com/Cyan4973/xxHash/issues/211#issuecomment-515575677 - */ -#if (defined(__GNUC__) || defined(__clang__)) && !defined(__wasm__) \ - && defined(__SIZEOF_INT128__) \ - || (defined(_INTEGRAL_MAX_BITS) && _INTEGRAL_MAX_BITS >= 128) - - __uint128_t const product = (__uint128_t)lhs * (__uint128_t)rhs; - XXH128_hash_t r128; - r128.low64 = (xxh_u64)(product); - r128.high64 = (xxh_u64)(product >> 64); - return r128; - - /* - * MSVC for x64's _umul128 method. - * - * xxh_u64 _umul128(xxh_u64 Multiplier, xxh_u64 Multiplicand, xxh_u64 *HighProduct); - * - * This compiles to single operand MUL on x64. - */ -#elif (defined(_M_X64) || defined(_M_IA64)) && !defined(_M_ARM64EC) - -#ifndef _MSC_VER -# pragma intrinsic(_umul128) -#endif - xxh_u64 product_high; - xxh_u64 const product_low = _umul128(lhs, rhs, &product_high); - XXH128_hash_t r128; - r128.low64 = product_low; - r128.high64 = product_high; - return r128; - - /* - * MSVC for ARM64's __umulh method. - * - * This compiles to the same MUL + UMULH as GCC/Clang's __uint128_t method. - */ -#elif defined(_M_ARM64) || defined(_M_ARM64EC) - -#ifndef _MSC_VER -# pragma intrinsic(__umulh) -#endif - XXH128_hash_t r128; - r128.low64 = lhs * rhs; - r128.high64 = __umulh(lhs, rhs); - return r128; - -#else - /* - * Portable scalar method. Optimized for 32-bit and 64-bit ALUs. - * - * This is a fast and simple grade school multiply, which is shown below - * with base 10 arithmetic instead of base 0x100000000. - * - * 9 3 // D2 lhs = 93 - * x 7 5 // D2 rhs = 75 - * ---------- - * 1 5 // D2 lo_lo = (93 % 10) * (75 % 10) = 15 - * 4 5 | // D2 hi_lo = (93 / 10) * (75 % 10) = 45 - * 2 1 | // D2 lo_hi = (93 % 10) * (75 / 10) = 21 - * + 6 3 | | // D2 hi_hi = (93 / 10) * (75 / 10) = 63 - * --------- - * 2 7 | // D2 cross = (15 / 10) + (45 % 10) + 21 = 27 - * + 6 7 | | // D2 upper = (27 / 10) + (45 / 10) + 63 = 67 - * --------- - * 6 9 7 5 // D4 res = (27 * 10) + (15 % 10) + (67 * 100) = 6975 - * - * The reasons for adding the products like this are: - * 1. It avoids manual carry tracking. Just like how - * (9 * 9) + 9 + 9 = 99, the same applies with this for UINT64_MAX. - * This avoids a lot of complexity. - * - * 2. It hints for, and on Clang, compiles to, the powerful UMAAL - * instruction available in ARM's Digital Signal Processing extension - * in 32-bit ARMv6 and later, which is shown below: - * - * void UMAAL(xxh_u32 *RdLo, xxh_u32 *RdHi, xxh_u32 Rn, xxh_u32 Rm) - * { - * xxh_u64 product = (xxh_u64)*RdLo * (xxh_u64)*RdHi + Rn + Rm; - * *RdLo = (xxh_u32)(product & 0xFFFFFFFF); - * *RdHi = (xxh_u32)(product >> 32); - * } - * - * This instruction was designed for efficient long multiplication, and - * allows this to be calculated in only 4 instructions at speeds - * comparable to some 64-bit ALUs. - * - * 3. It isn't terrible on other platforms. Usually this will be a couple - * of 32-bit ADD/ADCs. - */ - - /* First calculate all of the cross products. */ - xxh_u64 const lo_lo = XXH_mult32to64(lhs & 0xFFFFFFFF, rhs & 0xFFFFFFFF); - xxh_u64 const hi_lo = XXH_mult32to64(lhs >> 32, rhs & 0xFFFFFFFF); - xxh_u64 const lo_hi = XXH_mult32to64(lhs & 0xFFFFFFFF, rhs >> 32); - xxh_u64 const hi_hi = XXH_mult32to64(lhs >> 32, rhs >> 32); - - /* Now add the products together. These will never overflow. */ - xxh_u64 const cross = (lo_lo >> 32) + (hi_lo & 0xFFFFFFFF) + lo_hi; - xxh_u64 const upper = (hi_lo >> 32) + (cross >> 32) + hi_hi; - xxh_u64 const lower = (cross << 32) | (lo_lo & 0xFFFFFFFF); - - XXH128_hash_t r128; - r128.low64 = lower; - r128.high64 = upper; - return r128; -#endif -} - -/*! - * @brief Calculates a 64-bit to 128-bit multiply, then XOR folds it. - * - * The reason for the separate function is to prevent passing too many structs - * around by value. This will hopefully inline the multiply, but we don't force it. - * - * @param lhs , rhs The 64-bit integers to multiply - * @return The low 64 bits of the product XOR'd by the high 64 bits. - * @see XXH_mult64to128() - */ -static xxh_u64 -XXH3_mul128_fold64(xxh_u64 lhs, xxh_u64 rhs) -{ - XXH128_hash_t product = XXH_mult64to128(lhs, rhs); - return product.low64 ^ product.high64; -} - -/*! Seems to produce slightly better code on GCC for some reason. */ -XXH_FORCE_INLINE xxh_u64 XXH_xorshift64(xxh_u64 v64, int shift) -{ - XXH_ASSERT(0 <= shift && shift < 64); - return v64 ^ (v64 >> shift); -} - -/* - * This is a fast avalanche stage, - * suitable when input bits are already partially mixed - */ -static XXH64_hash_t XXH3_avalanche(xxh_u64 h64) -{ - h64 = XXH_xorshift64(h64, 37); - h64 *= 0x165667919E3779F9ULL; - h64 = XXH_xorshift64(h64, 32); - return h64; -} - -/* - * This is a stronger avalanche, - * inspired by Pelle Evensen's rrmxmx - * preferable when input has not been previously mixed - */ -static XXH64_hash_t XXH3_rrmxmx(xxh_u64 h64, xxh_u64 len) -{ - /* this mix is inspired by Pelle Evensen's rrmxmx */ - h64 ^= XXH_rotl64(h64, 49) ^ XXH_rotl64(h64, 24); - h64 *= 0x9FB21C651E98DF25ULL; - h64 ^= (h64 >> 35) + len ; - h64 *= 0x9FB21C651E98DF25ULL; - return XXH_xorshift64(h64, 28); -} - - -/* ========================================== - * Short keys - * ========================================== - * One of the shortcomings of XXH32 and XXH64 was that their performance was - * sub-optimal on short lengths. It used an iterative algorithm which strongly - * favored lengths that were a multiple of 4 or 8. - * - * Instead of iterating over individual inputs, we use a set of single shot - * functions which piece together a range of lengths and operate in constant time. - * - * Additionally, the number of multiplies has been significantly reduced. This - * reduces latency, especially when emulating 64-bit multiplies on 32-bit. - * - * Depending on the platform, this may or may not be faster than XXH32, but it - * is almost guaranteed to be faster than XXH64. - */ - -/* - * At very short lengths, there isn't enough input to fully hide secrets, or use - * the entire secret. - * - * There is also only a limited amount of mixing we can do before significantly - * impacting performance. - * - * Therefore, we use different sections of the secret and always mix two secret - * samples with an XOR. This should have no effect on performance on the - * seedless or withSeed variants because everything _should_ be constant folded - * by modern compilers. - * - * The XOR mixing hides individual parts of the secret and increases entropy. - * - * This adds an extra layer of strength for custom secrets. - */ -XXH_FORCE_INLINE XXH64_hash_t -XXH3_len_1to3_64b(const xxh_u8* input, size_t len, const xxh_u8* secret, XXH64_hash_t seed) -{ - XXH_ASSERT(input != NULL); - XXH_ASSERT(1 <= len && len <= 3); - XXH_ASSERT(secret != NULL); - /* - * len = 1: combined = { input[0], 0x01, input[0], input[0] } - * len = 2: combined = { input[1], 0x02, input[0], input[1] } - * len = 3: combined = { input[2], 0x03, input[0], input[1] } - */ - { xxh_u8 const c1 = input[0]; - xxh_u8 const c2 = input[len >> 1]; - xxh_u8 const c3 = input[len - 1]; - xxh_u32 const combined = ((xxh_u32)c1 << 16) | ((xxh_u32)c2 << 24) - | ((xxh_u32)c3 << 0) | ((xxh_u32)len << 8); - xxh_u64 const bitflip = (XXH_readLE32(secret) ^ XXH_readLE32(secret+4)) + seed; - xxh_u64 const keyed = (xxh_u64)combined ^ bitflip; - return XXH64_avalanche(keyed); - } -} - -XXH_FORCE_INLINE XXH64_hash_t -XXH3_len_4to8_64b(const xxh_u8* input, size_t len, const xxh_u8* secret, XXH64_hash_t seed) -{ - XXH_ASSERT(input != NULL); - XXH_ASSERT(secret != NULL); - XXH_ASSERT(4 <= len && len <= 8); - seed ^= (xxh_u64)XXH_swap32((xxh_u32)seed) << 32; - { xxh_u32 const input1 = XXH_readLE32(input); - xxh_u32 const input2 = XXH_readLE32(input + len - 4); - xxh_u64 const bitflip = (XXH_readLE64(secret+8) ^ XXH_readLE64(secret+16)) - seed; - xxh_u64 const input64 = input2 + (((xxh_u64)input1) << 32); - xxh_u64 const keyed = input64 ^ bitflip; - return XXH3_rrmxmx(keyed, len); - } -} - -XXH_FORCE_INLINE XXH64_hash_t -XXH3_len_9to16_64b(const xxh_u8* input, size_t len, const xxh_u8* secret, XXH64_hash_t seed) -{ - XXH_ASSERT(input != NULL); - XXH_ASSERT(secret != NULL); - XXH_ASSERT(9 <= len && len <= 16); - { xxh_u64 const bitflip1 = (XXH_readLE64(secret+24) ^ XXH_readLE64(secret+32)) + seed; - xxh_u64 const bitflip2 = (XXH_readLE64(secret+40) ^ XXH_readLE64(secret+48)) - seed; - xxh_u64 const input_lo = XXH_readLE64(input) ^ bitflip1; - xxh_u64 const input_hi = XXH_readLE64(input + len - 8) ^ bitflip2; - xxh_u64 const acc = len - + XXH_swap64(input_lo) + input_hi - + XXH3_mul128_fold64(input_lo, input_hi); - return XXH3_avalanche(acc); - } -} - -XXH_FORCE_INLINE XXH64_hash_t -XXH3_len_0to16_64b(const xxh_u8* input, size_t len, const xxh_u8* secret, XXH64_hash_t seed) -{ - XXH_ASSERT(len <= 16); - { if (XXH_likely(len > 8)) return XXH3_len_9to16_64b(input, len, secret, seed); - if (XXH_likely(len >= 4)) return XXH3_len_4to8_64b(input, len, secret, seed); - if (len) return XXH3_len_1to3_64b(input, len, secret, seed); - return XXH64_avalanche(seed ^ (XXH_readLE64(secret+56) ^ XXH_readLE64(secret+64))); - } -} - -/* - * DISCLAIMER: There are known *seed-dependent* multicollisions here due to - * multiplication by zero, affecting hashes of lengths 17 to 240. - * - * However, they are very unlikely. - * - * Keep this in mind when using the unseeded XXH3_64bits() variant: As with all - * unseeded non-cryptographic hashes, it does not attempt to defend itself - * against specially crafted inputs, only random inputs. - * - * Compared to classic UMAC where a 1 in 2^31 chance of 4 consecutive bytes - * cancelling out the secret is taken an arbitrary number of times (addressed - * in XXH3_accumulate_512), this collision is very unlikely with random inputs - * and/or proper seeding: - * - * This only has a 1 in 2^63 chance of 8 consecutive bytes cancelling out, in a - * function that is only called up to 16 times per hash with up to 240 bytes of - * input. - * - * This is not too bad for a non-cryptographic hash function, especially with - * only 64 bit outputs. - * - * The 128-bit variant (which trades some speed for strength) is NOT affected - * by this, although it is always a good idea to use a proper seed if you care - * about strength. - */ -XXH_FORCE_INLINE xxh_u64 XXH3_mix16B(const xxh_u8* XXH_RESTRICT input, - const xxh_u8* XXH_RESTRICT secret, xxh_u64 seed64) -{ -#if defined(__GNUC__) && !defined(__clang__) /* GCC, not Clang */ \ - && defined(__i386__) && defined(__SSE2__) /* x86 + SSE2 */ \ - && !defined(XXH_ENABLE_AUTOVECTORIZE) /* Define to disable like XXH32 hack */ - /* - * UGLY HACK: - * GCC for x86 tends to autovectorize the 128-bit multiply, resulting in - * slower code. - * - * By forcing seed64 into a register, we disrupt the cost model and - * cause it to scalarize. See `XXH32_round()` - * - * FIXME: Clang's output is still _much_ faster -- On an AMD Ryzen 3600, - * XXH3_64bits @ len=240 runs at 4.6 GB/s with Clang 9, but 3.3 GB/s on - * GCC 9.2, despite both emitting scalar code. - * - * GCC generates much better scalar code than Clang for the rest of XXH3, - * which is why finding a more optimal codepath is an interest. - */ - XXH_COMPILER_GUARD(seed64); -#endif - { xxh_u64 const input_lo = XXH_readLE64(input); - xxh_u64 const input_hi = XXH_readLE64(input+8); - return XXH3_mul128_fold64( - input_lo ^ (XXH_readLE64(secret) + seed64), - input_hi ^ (XXH_readLE64(secret+8) - seed64) - ); - } -} - -/* For mid range keys, XXH3 uses a Mum-hash variant. */ -XXH_FORCE_INLINE XXH64_hash_t -XXH3_len_17to128_64b(const xxh_u8* XXH_RESTRICT input, size_t len, - const xxh_u8* XXH_RESTRICT secret, size_t secretSize, - XXH64_hash_t seed) -{ - XXH_ASSERT(secretSize >= XXH3_SECRET_SIZE_MIN); (void)secretSize; - XXH_ASSERT(16 < len && len <= 128); - - { xxh_u64 acc = len * XXH_PRIME64_1; - if (len > 32) { - if (len > 64) { - if (len > 96) { - acc += XXH3_mix16B(input+48, secret+96, seed); - acc += XXH3_mix16B(input+len-64, secret+112, seed); - } - acc += XXH3_mix16B(input+32, secret+64, seed); - acc += XXH3_mix16B(input+len-48, secret+80, seed); - } - acc += XXH3_mix16B(input+16, secret+32, seed); - acc += XXH3_mix16B(input+len-32, secret+48, seed); - } - acc += XXH3_mix16B(input+0, secret+0, seed); - acc += XXH3_mix16B(input+len-16, secret+16, seed); - - return XXH3_avalanche(acc); - } -} - -#define XXH3_MIDSIZE_MAX 240 - -XXH_NO_INLINE XXH64_hash_t -XXH3_len_129to240_64b(const xxh_u8* XXH_RESTRICT input, size_t len, - const xxh_u8* XXH_RESTRICT secret, size_t secretSize, - XXH64_hash_t seed) -{ - XXH_ASSERT(secretSize >= XXH3_SECRET_SIZE_MIN); (void)secretSize; - XXH_ASSERT(128 < len && len <= XXH3_MIDSIZE_MAX); - - #define XXH3_MIDSIZE_STARTOFFSET 3 - #define XXH3_MIDSIZE_LASTOFFSET 17 - - { xxh_u64 acc = len * XXH_PRIME64_1; - int const nbRounds = (int)len / 16; - int i; - for (i=0; i<8; i++) { - acc += XXH3_mix16B(input+(16*i), secret+(16*i), seed); - } - acc = XXH3_avalanche(acc); - XXH_ASSERT(nbRounds >= 8); -#if defined(__clang__) /* Clang */ \ - && (defined(__ARM_NEON) || defined(__ARM_NEON__)) /* NEON */ \ - && !defined(XXH_ENABLE_AUTOVECTORIZE) /* Define to disable */ - /* - * UGLY HACK: - * Clang for ARMv7-A tries to vectorize this loop, similar to GCC x86. - * In everywhere else, it uses scalar code. - * - * For 64->128-bit multiplies, even if the NEON was 100% optimal, it - * would still be slower than UMAAL (see XXH_mult64to128). - * - * Unfortunately, Clang doesn't handle the long multiplies properly and - * converts them to the nonexistent "vmulq_u64" intrinsic, which is then - * scalarized into an ugly mess of VMOV.32 instructions. - * - * This mess is difficult to avoid without turning autovectorization - * off completely, but they are usually relatively minor and/or not - * worth it to fix. - * - * This loop is the easiest to fix, as unlike XXH32, this pragma - * _actually works_ because it is a loop vectorization instead of an - * SLP vectorization. - */ - #pragma clang loop vectorize(disable) -#endif - for (i=8 ; i < nbRounds; i++) { - acc += XXH3_mix16B(input+(16*i), secret+(16*(i-8)) + XXH3_MIDSIZE_STARTOFFSET, seed); - } - /* last bytes */ - acc += XXH3_mix16B(input + len - 16, secret + XXH3_SECRET_SIZE_MIN - XXH3_MIDSIZE_LASTOFFSET, seed); - return XXH3_avalanche(acc); - } -} - - -/* ======= Long Keys ======= */ - -#define XXH_STRIPE_LEN 64 -#define XXH_SECRET_CONSUME_RATE 8 /* nb of secret bytes consumed at each accumulation */ -#define XXH_ACC_NB (XXH_STRIPE_LEN / sizeof(xxh_u64)) - -#ifdef XXH_OLD_NAMES -# define STRIPE_LEN XXH_STRIPE_LEN -# define ACC_NB XXH_ACC_NB -#endif - -XXH_FORCE_INLINE void XXH_writeLE64(void* dst, xxh_u64 v64) -{ - if (!XXH_CPU_LITTLE_ENDIAN) v64 = XXH_swap64(v64); - XXH_memcpy(dst, &v64, sizeof(v64)); -} - -/* Several intrinsic functions below are supposed to accept __int64 as argument, - * as documented in https://software.intel.com/sites/landingpage/IntrinsicsGuide/ . - * However, several environments do not define __int64 type, - * requiring a workaround. - */ -#if !defined (__VMS) \ - && (defined (__cplusplus) \ - || (defined (__STDC_VERSION__) && (__STDC_VERSION__ >= 199901L) /* C99 */) ) - typedef int64_t xxh_i64; -#else - /* the following type must have a width of 64-bit */ - typedef long long xxh_i64; -#endif - - -/* - * XXH3_accumulate_512 is the tightest loop for long inputs, and it is the most optimized. - * - * It is a hardened version of UMAC, based off of FARSH's implementation. - * - * This was chosen because it adapts quite well to 32-bit, 64-bit, and SIMD - * implementations, and it is ridiculously fast. - * - * We harden it by mixing the original input to the accumulators as well as the product. - * - * This means that in the (relatively likely) case of a multiply by zero, the - * original input is preserved. - * - * On 128-bit inputs, we swap 64-bit pairs when we add the input to improve - * cross-pollination, as otherwise the upper and lower halves would be - * essentially independent. - * - * This doesn't matter on 64-bit hashes since they all get merged together in - * the end, so we skip the extra step. - * - * Both XXH3_64bits and XXH3_128bits use this subroutine. - */ - -#if (XXH_VECTOR == XXH_AVX512) \ - || (defined(XXH_DISPATCH_AVX512) && XXH_DISPATCH_AVX512 != 0) - -#ifndef XXH_TARGET_AVX512 -# define XXH_TARGET_AVX512 /* disable attribute target */ -#endif - -XXH_FORCE_INLINE XXH_TARGET_AVX512 void -XXH3_accumulate_512_avx512(void* XXH_RESTRICT acc, - const void* XXH_RESTRICT input, - const void* XXH_RESTRICT secret) -{ - __m512i* const xacc = (__m512i *) acc; - XXH_ASSERT((((size_t)acc) & 63) == 0); - XXH_STATIC_ASSERT(XXH_STRIPE_LEN == sizeof(__m512i)); - - { - /* data_vec = input[0]; */ - __m512i const data_vec = _mm512_loadu_si512 (input); - /* key_vec = secret[0]; */ - __m512i const key_vec = _mm512_loadu_si512 (secret); - /* data_key = data_vec ^ key_vec; */ - __m512i const data_key = _mm512_xor_si512 (data_vec, key_vec); - /* data_key_lo = data_key >> 32; */ - __m512i const data_key_lo = _mm512_shuffle_epi32 (data_key, (_MM_PERM_ENUM)_MM_SHUFFLE(0, 3, 0, 1)); - /* product = (data_key & 0xffffffff) * (data_key_lo & 0xffffffff); */ - __m512i const product = _mm512_mul_epu32 (data_key, data_key_lo); - /* xacc[0] += swap(data_vec); */ - __m512i const data_swap = _mm512_shuffle_epi32(data_vec, (_MM_PERM_ENUM)_MM_SHUFFLE(1, 0, 3, 2)); - __m512i const sum = _mm512_add_epi64(*xacc, data_swap); - /* xacc[0] += product; */ - *xacc = _mm512_add_epi64(product, sum); - } -} - -/* - * XXH3_scrambleAcc: Scrambles the accumulators to improve mixing. - * - * Multiplication isn't perfect, as explained by Google in HighwayHash: - * - * // Multiplication mixes/scrambles bytes 0-7 of the 64-bit result to - * // varying degrees. In descending order of goodness, bytes - * // 3 4 2 5 1 6 0 7 have quality 228 224 164 160 100 96 36 32. - * // As expected, the upper and lower bytes are much worse. - * - * Source: https://github.com/google/highwayhash/blob/0aaf66b/highwayhash/hh_avx2.h#L291 - * - * Since our algorithm uses a pseudorandom secret to add some variance into the - * mix, we don't need to (or want to) mix as often or as much as HighwayHash does. - * - * This isn't as tight as XXH3_accumulate, but still written in SIMD to avoid - * extraction. - * - * Both XXH3_64bits and XXH3_128bits use this subroutine. - */ - -XXH_FORCE_INLINE XXH_TARGET_AVX512 void -XXH3_scrambleAcc_avx512(void* XXH_RESTRICT acc, const void* XXH_RESTRICT secret) -{ - XXH_ASSERT((((size_t)acc) & 63) == 0); - XXH_STATIC_ASSERT(XXH_STRIPE_LEN == sizeof(__m512i)); - { __m512i* const xacc = (__m512i*) acc; - const __m512i prime32 = _mm512_set1_epi32((int)XXH_PRIME32_1); - - /* xacc[0] ^= (xacc[0] >> 47) */ - __m512i const acc_vec = *xacc; - __m512i const shifted = _mm512_srli_epi64 (acc_vec, 47); - __m512i const data_vec = _mm512_xor_si512 (acc_vec, shifted); - /* xacc[0] ^= secret; */ - __m512i const key_vec = _mm512_loadu_si512 (secret); - __m512i const data_key = _mm512_xor_si512 (data_vec, key_vec); - - /* xacc[0] *= XXH_PRIME32_1; */ - __m512i const data_key_hi = _mm512_shuffle_epi32 (data_key, (_MM_PERM_ENUM)_MM_SHUFFLE(0, 3, 0, 1)); - __m512i const prod_lo = _mm512_mul_epu32 (data_key, prime32); - __m512i const prod_hi = _mm512_mul_epu32 (data_key_hi, prime32); - *xacc = _mm512_add_epi64(prod_lo, _mm512_slli_epi64(prod_hi, 32)); - } -} - -XXH_FORCE_INLINE XXH_TARGET_AVX512 void -XXH3_initCustomSecret_avx512(void* XXH_RESTRICT customSecret, xxh_u64 seed64) -{ - XXH_STATIC_ASSERT((XXH_SECRET_DEFAULT_SIZE & 63) == 0); - XXH_STATIC_ASSERT(XXH_SEC_ALIGN == 64); - XXH_ASSERT(((size_t)customSecret & 63) == 0); - (void)(&XXH_writeLE64); - { int const nbRounds = XXH_SECRET_DEFAULT_SIZE / sizeof(__m512i); - __m512i const seed = _mm512_mask_set1_epi64(_mm512_set1_epi64((xxh_i64)seed64), 0xAA, (xxh_i64)(0U - seed64)); - - const __m512i* const src = (const __m512i*) ((const void*) XXH3_kSecret); - __m512i* const dest = ( __m512i*) customSecret; - int i; - XXH_ASSERT(((size_t)src & 63) == 0); /* control alignment */ - XXH_ASSERT(((size_t)dest & 63) == 0); - for (i=0; i < nbRounds; ++i) { - /* GCC has a bug, _mm512_stream_load_si512 accepts 'void*', not 'void const*', - * this will warn "discards 'const' qualifier". */ - union { - const __m512i* cp; - void* p; - } remote_const_void; - remote_const_void.cp = src + i; - dest[i] = _mm512_add_epi64(_mm512_stream_load_si512(remote_const_void.p), seed); - } } -} - -#endif - -#if (XXH_VECTOR == XXH_AVX2) \ - || (defined(XXH_DISPATCH_AVX2) && XXH_DISPATCH_AVX2 != 0) - -#ifndef XXH_TARGET_AVX2 -# define XXH_TARGET_AVX2 /* disable attribute target */ -#endif - -XXH_FORCE_INLINE XXH_TARGET_AVX2 void -XXH3_accumulate_512_avx2( void* XXH_RESTRICT acc, - const void* XXH_RESTRICT input, - const void* XXH_RESTRICT secret) -{ - XXH_ASSERT((((size_t)acc) & 31) == 0); - { __m256i* const xacc = (__m256i *) acc; - /* Unaligned. This is mainly for pointer arithmetic, and because - * _mm256_loadu_si256 requires a const __m256i * pointer for some reason. */ - const __m256i* const xinput = (const __m256i *) input; - /* Unaligned. This is mainly for pointer arithmetic, and because - * _mm256_loadu_si256 requires a const __m256i * pointer for some reason. */ - const __m256i* const xsecret = (const __m256i *) secret; - - size_t i; - for (i=0; i < XXH_STRIPE_LEN/sizeof(__m256i); i++) { - /* data_vec = xinput[i]; */ - __m256i const data_vec = _mm256_loadu_si256 (xinput+i); - /* key_vec = xsecret[i]; */ - __m256i const key_vec = _mm256_loadu_si256 (xsecret+i); - /* data_key = data_vec ^ key_vec; */ - __m256i const data_key = _mm256_xor_si256 (data_vec, key_vec); - /* data_key_lo = data_key >> 32; */ - __m256i const data_key_lo = _mm256_shuffle_epi32 (data_key, _MM_SHUFFLE(0, 3, 0, 1)); - /* product = (data_key & 0xffffffff) * (data_key_lo & 0xffffffff); */ - __m256i const product = _mm256_mul_epu32 (data_key, data_key_lo); - /* xacc[i] += swap(data_vec); */ - __m256i const data_swap = _mm256_shuffle_epi32(data_vec, _MM_SHUFFLE(1, 0, 3, 2)); - __m256i const sum = _mm256_add_epi64(xacc[i], data_swap); - /* xacc[i] += product; */ - xacc[i] = _mm256_add_epi64(product, sum); - } } -} - -XXH_FORCE_INLINE XXH_TARGET_AVX2 void -XXH3_scrambleAcc_avx2(void* XXH_RESTRICT acc, const void* XXH_RESTRICT secret) -{ - XXH_ASSERT((((size_t)acc) & 31) == 0); - { __m256i* const xacc = (__m256i*) acc; - /* Unaligned. This is mainly for pointer arithmetic, and because - * _mm256_loadu_si256 requires a const __m256i * pointer for some reason. */ - const __m256i* const xsecret = (const __m256i *) secret; - const __m256i prime32 = _mm256_set1_epi32((int)XXH_PRIME32_1); - - size_t i; - for (i=0; i < XXH_STRIPE_LEN/sizeof(__m256i); i++) { - /* xacc[i] ^= (xacc[i] >> 47) */ - __m256i const acc_vec = xacc[i]; - __m256i const shifted = _mm256_srli_epi64 (acc_vec, 47); - __m256i const data_vec = _mm256_xor_si256 (acc_vec, shifted); - /* xacc[i] ^= xsecret; */ - __m256i const key_vec = _mm256_loadu_si256 (xsecret+i); - __m256i const data_key = _mm256_xor_si256 (data_vec, key_vec); - - /* xacc[i] *= XXH_PRIME32_1; */ - __m256i const data_key_hi = _mm256_shuffle_epi32 (data_key, _MM_SHUFFLE(0, 3, 0, 1)); - __m256i const prod_lo = _mm256_mul_epu32 (data_key, prime32); - __m256i const prod_hi = _mm256_mul_epu32 (data_key_hi, prime32); - xacc[i] = _mm256_add_epi64(prod_lo, _mm256_slli_epi64(prod_hi, 32)); - } - } -} - -XXH_FORCE_INLINE XXH_TARGET_AVX2 void XXH3_initCustomSecret_avx2(void* XXH_RESTRICT customSecret, xxh_u64 seed64) -{ - XXH_STATIC_ASSERT((XXH_SECRET_DEFAULT_SIZE & 31) == 0); - XXH_STATIC_ASSERT((XXH_SECRET_DEFAULT_SIZE / sizeof(__m256i)) == 6); - XXH_STATIC_ASSERT(XXH_SEC_ALIGN <= 64); - (void)(&XXH_writeLE64); - XXH_PREFETCH(customSecret); - { __m256i const seed = _mm256_set_epi64x((xxh_i64)(0U - seed64), (xxh_i64)seed64, (xxh_i64)(0U - seed64), (xxh_i64)seed64); - - const __m256i* const src = (const __m256i*) ((const void*) XXH3_kSecret); - __m256i* dest = ( __m256i*) customSecret; - -# if defined(__GNUC__) || defined(__clang__) - /* - * On GCC & Clang, marking 'dest' as modified will cause the compiler: - * - do not extract the secret from sse registers in the internal loop - * - use less common registers, and avoid pushing these reg into stack - */ - XXH_COMPILER_GUARD(dest); -# endif - XXH_ASSERT(((size_t)src & 31) == 0); /* control alignment */ - XXH_ASSERT(((size_t)dest & 31) == 0); - - /* GCC -O2 need unroll loop manually */ - dest[0] = _mm256_add_epi64(_mm256_stream_load_si256(src+0), seed); - dest[1] = _mm256_add_epi64(_mm256_stream_load_si256(src+1), seed); - dest[2] = _mm256_add_epi64(_mm256_stream_load_si256(src+2), seed); - dest[3] = _mm256_add_epi64(_mm256_stream_load_si256(src+3), seed); - dest[4] = _mm256_add_epi64(_mm256_stream_load_si256(src+4), seed); - dest[5] = _mm256_add_epi64(_mm256_stream_load_si256(src+5), seed); - } -} - -#endif - -/* x86dispatch always generates SSE2 */ -#if (XXH_VECTOR == XXH_SSE2) || defined(XXH_X86DISPATCH) - -#ifndef XXH_TARGET_SSE2 -# define XXH_TARGET_SSE2 /* disable attribute target */ -#endif - -XXH_FORCE_INLINE XXH_TARGET_SSE2 void -XXH3_accumulate_512_sse2( void* XXH_RESTRICT acc, - const void* XXH_RESTRICT input, - const void* XXH_RESTRICT secret) -{ - /* SSE2 is just a half-scale version of the AVX2 version. */ - XXH_ASSERT((((size_t)acc) & 15) == 0); - { __m128i* const xacc = (__m128i *) acc; - /* Unaligned. This is mainly for pointer arithmetic, and because - * _mm_loadu_si128 requires a const __m128i * pointer for some reason. */ - const __m128i* const xinput = (const __m128i *) input; - /* Unaligned. This is mainly for pointer arithmetic, and because - * _mm_loadu_si128 requires a const __m128i * pointer for some reason. */ - const __m128i* const xsecret = (const __m128i *) secret; - - size_t i; - for (i=0; i < XXH_STRIPE_LEN/sizeof(__m128i); i++) { - /* data_vec = xinput[i]; */ - __m128i const data_vec = _mm_loadu_si128 (xinput+i); - /* key_vec = xsecret[i]; */ - __m128i const key_vec = _mm_loadu_si128 (xsecret+i); - /* data_key = data_vec ^ key_vec; */ - __m128i const data_key = _mm_xor_si128 (data_vec, key_vec); - /* data_key_lo = data_key >> 32; */ - __m128i const data_key_lo = _mm_shuffle_epi32 (data_key, _MM_SHUFFLE(0, 3, 0, 1)); - /* product = (data_key & 0xffffffff) * (data_key_lo & 0xffffffff); */ - __m128i const product = _mm_mul_epu32 (data_key, data_key_lo); - /* xacc[i] += swap(data_vec); */ - __m128i const data_swap = _mm_shuffle_epi32(data_vec, _MM_SHUFFLE(1,0,3,2)); - __m128i const sum = _mm_add_epi64(xacc[i], data_swap); - /* xacc[i] += product; */ - xacc[i] = _mm_add_epi64(product, sum); - } } -} - -XXH_FORCE_INLINE XXH_TARGET_SSE2 void -XXH3_scrambleAcc_sse2(void* XXH_RESTRICT acc, const void* XXH_RESTRICT secret) -{ - XXH_ASSERT((((size_t)acc) & 15) == 0); - { __m128i* const xacc = (__m128i*) acc; - /* Unaligned. This is mainly for pointer arithmetic, and because - * _mm_loadu_si128 requires a const __m128i * pointer for some reason. */ - const __m128i* const xsecret = (const __m128i *) secret; - const __m128i prime32 = _mm_set1_epi32((int)XXH_PRIME32_1); - - size_t i; - for (i=0; i < XXH_STRIPE_LEN/sizeof(__m128i); i++) { - /* xacc[i] ^= (xacc[i] >> 47) */ - __m128i const acc_vec = xacc[i]; - __m128i const shifted = _mm_srli_epi64 (acc_vec, 47); - __m128i const data_vec = _mm_xor_si128 (acc_vec, shifted); - /* xacc[i] ^= xsecret[i]; */ - __m128i const key_vec = _mm_loadu_si128 (xsecret+i); - __m128i const data_key = _mm_xor_si128 (data_vec, key_vec); - - /* xacc[i] *= XXH_PRIME32_1; */ - __m128i const data_key_hi = _mm_shuffle_epi32 (data_key, _MM_SHUFFLE(0, 3, 0, 1)); - __m128i const prod_lo = _mm_mul_epu32 (data_key, prime32); - __m128i const prod_hi = _mm_mul_epu32 (data_key_hi, prime32); - xacc[i] = _mm_add_epi64(prod_lo, _mm_slli_epi64(prod_hi, 32)); - } - } -} - -XXH_FORCE_INLINE XXH_TARGET_SSE2 void XXH3_initCustomSecret_sse2(void* XXH_RESTRICT customSecret, xxh_u64 seed64) -{ - XXH_STATIC_ASSERT((XXH_SECRET_DEFAULT_SIZE & 15) == 0); - (void)(&XXH_writeLE64); - { int const nbRounds = XXH_SECRET_DEFAULT_SIZE / sizeof(__m128i); - -# if defined(_MSC_VER) && defined(_M_IX86) && _MSC_VER < 1900 - /* MSVC 32bit mode does not support _mm_set_epi64x before 2015 */ - XXH_ALIGN(16) const xxh_i64 seed64x2[2] = { (xxh_i64)seed64, (xxh_i64)(0U - seed64) }; - __m128i const seed = _mm_load_si128((__m128i const*)seed64x2); -# else - __m128i const seed = _mm_set_epi64x((xxh_i64)(0U - seed64), (xxh_i64)seed64); -# endif - int i; - - const void* const src16 = XXH3_kSecret; - __m128i* dst16 = (__m128i*) customSecret; -# if defined(__GNUC__) || defined(__clang__) - /* - * On GCC & Clang, marking 'dest' as modified will cause the compiler: - * - do not extract the secret from sse registers in the internal loop - * - use less common registers, and avoid pushing these reg into stack - */ - XXH_COMPILER_GUARD(dst16); -# endif - XXH_ASSERT(((size_t)src16 & 15) == 0); /* control alignment */ - XXH_ASSERT(((size_t)dst16 & 15) == 0); - - for (i=0; i < nbRounds; ++i) { - dst16[i] = _mm_add_epi64(_mm_load_si128((const __m128i *)src16+i), seed); - } } -} - -#endif - -#if (XXH_VECTOR == XXH_NEON) - -/* forward declarations for the scalar routines */ -XXH_FORCE_INLINE void -XXH3_scalarRound(void* XXH_RESTRICT acc, void const* XXH_RESTRICT input, - void const* XXH_RESTRICT secret, size_t lane); - -XXH_FORCE_INLINE void -XXH3_scalarScrambleRound(void* XXH_RESTRICT acc, - void const* XXH_RESTRICT secret, size_t lane); - -/*! - * @internal - * @brief The bulk processing loop for NEON. - * - * The NEON code path is actually partially scalar when running on AArch64. This - * is to optimize the pipelining and can have up to 15% speedup depending on the - * CPU, and it also mitigates some GCC codegen issues. - * - * @see XXH3_NEON_LANES for configuring this and details about this optimization. - */ -XXH_FORCE_INLINE void -XXH3_accumulate_512_neon( void* XXH_RESTRICT acc, - const void* XXH_RESTRICT input, - const void* XXH_RESTRICT secret) -{ - XXH_ASSERT((((size_t)acc) & 15) == 0); - XXH_STATIC_ASSERT(XXH3_NEON_LANES > 0 && XXH3_NEON_LANES <= XXH_ACC_NB && XXH3_NEON_LANES % 2 == 0); - { - uint64x2_t* const xacc = (uint64x2_t *) acc; - /* We don't use a uint32x4_t pointer because it causes bus errors on ARMv7. */ - uint8_t const* const xinput = (const uint8_t *) input; - uint8_t const* const xsecret = (const uint8_t *) secret; - - size_t i; - /* NEON for the first few lanes (these loops are normally interleaved) */ - for (i=0; i < XXH3_NEON_LANES / 2; i++) { - /* data_vec = xinput[i]; */ - uint8x16_t data_vec = vld1q_u8(xinput + (i * 16)); - /* key_vec = xsecret[i]; */ - uint8x16_t key_vec = vld1q_u8(xsecret + (i * 16)); - uint64x2_t data_key; - uint32x2_t data_key_lo, data_key_hi; - /* xacc[i] += swap(data_vec); */ - uint64x2_t const data64 = vreinterpretq_u64_u8(data_vec); - uint64x2_t const swapped = vextq_u64(data64, data64, 1); - xacc[i] = vaddq_u64 (xacc[i], swapped); - /* data_key = data_vec ^ key_vec; */ - data_key = vreinterpretq_u64_u8(veorq_u8(data_vec, key_vec)); - /* data_key_lo = (uint32x2_t) (data_key & 0xFFFFFFFF); - * data_key_hi = (uint32x2_t) (data_key >> 32); - * data_key = UNDEFINED; */ - XXH_SPLIT_IN_PLACE(data_key, data_key_lo, data_key_hi); - /* xacc[i] += (uint64x2_t) data_key_lo * (uint64x2_t) data_key_hi; */ - xacc[i] = vmlal_u32 (xacc[i], data_key_lo, data_key_hi); - - } - /* Scalar for the remainder. This may be a zero iteration loop. */ - for (i = XXH3_NEON_LANES; i < XXH_ACC_NB; i++) { - XXH3_scalarRound(acc, input, secret, i); - } - } -} - -XXH_FORCE_INLINE void -XXH3_scrambleAcc_neon(void* XXH_RESTRICT acc, const void* XXH_RESTRICT secret) -{ - XXH_ASSERT((((size_t)acc) & 15) == 0); - - { uint64x2_t* xacc = (uint64x2_t*) acc; - uint8_t const* xsecret = (uint8_t const*) secret; - uint32x2_t prime = vdup_n_u32 (XXH_PRIME32_1); - - size_t i; - /* NEON for the first few lanes (these loops are normally interleaved) */ - for (i=0; i < XXH3_NEON_LANES / 2; i++) { - /* xacc[i] ^= (xacc[i] >> 47); */ - uint64x2_t acc_vec = xacc[i]; - uint64x2_t shifted = vshrq_n_u64 (acc_vec, 47); - uint64x2_t data_vec = veorq_u64 (acc_vec, shifted); - - /* xacc[i] ^= xsecret[i]; */ - uint8x16_t key_vec = vld1q_u8 (xsecret + (i * 16)); - uint64x2_t data_key = veorq_u64 (data_vec, vreinterpretq_u64_u8(key_vec)); - - /* xacc[i] *= XXH_PRIME32_1 */ - uint32x2_t data_key_lo, data_key_hi; - /* data_key_lo = (uint32x2_t) (xacc[i] & 0xFFFFFFFF); - * data_key_hi = (uint32x2_t) (xacc[i] >> 32); - * xacc[i] = UNDEFINED; */ - XXH_SPLIT_IN_PLACE(data_key, data_key_lo, data_key_hi); - { /* - * prod_hi = (data_key >> 32) * XXH_PRIME32_1; - * - * Avoid vmul_u32 + vshll_n_u32 since Clang 6 and 7 will - * incorrectly "optimize" this: - * tmp = vmul_u32(vmovn_u64(a), vmovn_u64(b)); - * shifted = vshll_n_u32(tmp, 32); - * to this: - * tmp = "vmulq_u64"(a, b); // no such thing! - * shifted = vshlq_n_u64(tmp, 32); - * - * However, unlike SSE, Clang lacks a 64-bit multiply routine - * for NEON, and it scalarizes two 64-bit multiplies instead. - * - * vmull_u32 has the same timing as vmul_u32, and it avoids - * this bug completely. - * See https://bugs.llvm.org/show_bug.cgi?id=39967 - */ - uint64x2_t prod_hi = vmull_u32 (data_key_hi, prime); - /* xacc[i] = prod_hi << 32; */ - xacc[i] = vshlq_n_u64(prod_hi, 32); - /* xacc[i] += (prod_hi & 0xFFFFFFFF) * XXH_PRIME32_1; */ - xacc[i] = vmlal_u32(xacc[i], data_key_lo, prime); - } - } - /* Scalar for the remainder. This may be a zero iteration loop. */ - for (i = XXH3_NEON_LANES; i < XXH_ACC_NB; i++) { - XXH3_scalarScrambleRound(acc, secret, i); - } - } -} - -#endif - -#if (XXH_VECTOR == XXH_VSX) - -XXH_FORCE_INLINE void -XXH3_accumulate_512_vsx( void* XXH_RESTRICT acc, - const void* XXH_RESTRICT input, - const void* XXH_RESTRICT secret) -{ - /* presumed aligned */ - unsigned int* const xacc = (unsigned int*) acc; - xxh_u64x2 const* const xinput = (xxh_u64x2 const*) input; /* no alignment restriction */ - xxh_u64x2 const* const xsecret = (xxh_u64x2 const*) secret; /* no alignment restriction */ - xxh_u64x2 const v32 = { 32, 32 }; - size_t i; - for (i = 0; i < XXH_STRIPE_LEN / sizeof(xxh_u64x2); i++) { - /* data_vec = xinput[i]; */ - xxh_u64x2 const data_vec = XXH_vec_loadu(xinput + i); - /* key_vec = xsecret[i]; */ - xxh_u64x2 const key_vec = XXH_vec_loadu(xsecret + i); - xxh_u64x2 const data_key = data_vec ^ key_vec; - /* shuffled = (data_key << 32) | (data_key >> 32); */ - xxh_u32x4 const shuffled = (xxh_u32x4)vec_rl(data_key, v32); - /* product = ((xxh_u64x2)data_key & 0xFFFFFFFF) * ((xxh_u64x2)shuffled & 0xFFFFFFFF); */ - xxh_u64x2 const product = XXH_vec_mulo((xxh_u32x4)data_key, shuffled); - /* acc_vec = xacc[i]; */ - xxh_u64x2 acc_vec = (xxh_u64x2)vec_xl(0, xacc + 4 * i); - acc_vec += product; - - /* swap high and low halves */ -#ifdef __s390x__ - acc_vec += vec_permi(data_vec, data_vec, 2); -#else - acc_vec += vec_xxpermdi(data_vec, data_vec, 2); -#endif - /* xacc[i] = acc_vec; */ - vec_xst((xxh_u32x4)acc_vec, 0, xacc + 4 * i); - } -} - -XXH_FORCE_INLINE void -XXH3_scrambleAcc_vsx(void* XXH_RESTRICT acc, const void* XXH_RESTRICT secret) -{ - XXH_ASSERT((((size_t)acc) & 15) == 0); - - { xxh_u64x2* const xacc = (xxh_u64x2*) acc; - const xxh_u64x2* const xsecret = (const xxh_u64x2*) secret; - /* constants */ - xxh_u64x2 const v32 = { 32, 32 }; - xxh_u64x2 const v47 = { 47, 47 }; - xxh_u32x4 const prime = { XXH_PRIME32_1, XXH_PRIME32_1, XXH_PRIME32_1, XXH_PRIME32_1 }; - size_t i; - for (i = 0; i < XXH_STRIPE_LEN / sizeof(xxh_u64x2); i++) { - /* xacc[i] ^= (xacc[i] >> 47); */ - xxh_u64x2 const acc_vec = xacc[i]; - xxh_u64x2 const data_vec = acc_vec ^ (acc_vec >> v47); - - /* xacc[i] ^= xsecret[i]; */ - xxh_u64x2 const key_vec = XXH_vec_loadu(xsecret + i); - xxh_u64x2 const data_key = data_vec ^ key_vec; - - /* xacc[i] *= XXH_PRIME32_1 */ - /* prod_lo = ((xxh_u64x2)data_key & 0xFFFFFFFF) * ((xxh_u64x2)prime & 0xFFFFFFFF); */ - xxh_u64x2 const prod_even = XXH_vec_mule((xxh_u32x4)data_key, prime); - /* prod_hi = ((xxh_u64x2)data_key >> 32) * ((xxh_u64x2)prime >> 32); */ - xxh_u64x2 const prod_odd = XXH_vec_mulo((xxh_u32x4)data_key, prime); - xacc[i] = prod_odd + (prod_even << v32); - } } -} - -#endif - -/* scalar variants - universal */ - -/*! - * @internal - * @brief Scalar round for @ref XXH3_accumulate_512_scalar(). - * - * This is extracted to its own function because the NEON path uses a combination - * of NEON and scalar. - */ -XXH_FORCE_INLINE void -XXH3_scalarRound(void* XXH_RESTRICT acc, - void const* XXH_RESTRICT input, - void const* XXH_RESTRICT secret, - size_t lane) -{ - xxh_u64* xacc = (xxh_u64*) acc; - xxh_u8 const* xinput = (xxh_u8 const*) input; - xxh_u8 const* xsecret = (xxh_u8 const*) secret; - XXH_ASSERT(lane < XXH_ACC_NB); - XXH_ASSERT(((size_t)acc & (XXH_ACC_ALIGN-1)) == 0); - { - xxh_u64 const data_val = XXH_readLE64(xinput + lane * 8); - xxh_u64 const data_key = data_val ^ XXH_readLE64(xsecret + lane * 8); - xacc[lane ^ 1] += data_val; /* swap adjacent lanes */ - xacc[lane] += XXH_mult32to64(data_key & 0xFFFFFFFF, data_key >> 32); - } -} - -/*! - * @internal - * @brief Processes a 64 byte block of data using the scalar path. - */ -XXH_FORCE_INLINE void -XXH3_accumulate_512_scalar(void* XXH_RESTRICT acc, - const void* XXH_RESTRICT input, - const void* XXH_RESTRICT secret) -{ - size_t i; - for (i=0; i < XXH_ACC_NB; i++) { - XXH3_scalarRound(acc, input, secret, i); - } -} - -/*! - * @internal - * @brief Scalar scramble step for @ref XXH3_scrambleAcc_scalar(). - * - * This is extracted to its own function because the NEON path uses a combination - * of NEON and scalar. - */ -XXH_FORCE_INLINE void -XXH3_scalarScrambleRound(void* XXH_RESTRICT acc, - void const* XXH_RESTRICT secret, - size_t lane) -{ - xxh_u64* const xacc = (xxh_u64*) acc; /* presumed aligned */ - const xxh_u8* const xsecret = (const xxh_u8*) secret; /* no alignment restriction */ - XXH_ASSERT((((size_t)acc) & (XXH_ACC_ALIGN-1)) == 0); - XXH_ASSERT(lane < XXH_ACC_NB); - { - xxh_u64 const key64 = XXH_readLE64(xsecret + lane * 8); - xxh_u64 acc64 = xacc[lane]; - acc64 = XXH_xorshift64(acc64, 47); - acc64 ^= key64; - acc64 *= XXH_PRIME32_1; - xacc[lane] = acc64; - } -} - -/*! - * @internal - * @brief Scrambles the accumulators after a large chunk has been read - */ -XXH_FORCE_INLINE void -XXH3_scrambleAcc_scalar(void* XXH_RESTRICT acc, const void* XXH_RESTRICT secret) -{ - size_t i; - for (i=0; i < XXH_ACC_NB; i++) { - XXH3_scalarScrambleRound(acc, secret, i); - } -} - -XXH_FORCE_INLINE void -XXH3_initCustomSecret_scalar(void* XXH_RESTRICT customSecret, xxh_u64 seed64) -{ - /* - * We need a separate pointer for the hack below, - * which requires a non-const pointer. - * Any decent compiler will optimize this out otherwise. - */ - const xxh_u8* kSecretPtr = XXH3_kSecret; - XXH_STATIC_ASSERT((XXH_SECRET_DEFAULT_SIZE & 15) == 0); - -#if defined(__clang__) && defined(__aarch64__) - /* - * UGLY HACK: - * Clang generates a bunch of MOV/MOVK pairs for aarch64, and they are - * placed sequentially, in order, at the top of the unrolled loop. - * - * While MOVK is great for generating constants (2 cycles for a 64-bit - * constant compared to 4 cycles for LDR), it fights for bandwidth with - * the arithmetic instructions. - * - * I L S - * MOVK - * MOVK - * MOVK - * MOVK - * ADD - * SUB STR - * STR - * By forcing loads from memory (as the asm line causes Clang to assume - * that XXH3_kSecretPtr has been changed), the pipelines are used more - * efficiently: - * I L S - * LDR - * ADD LDR - * SUB STR - * STR - * - * See XXH3_NEON_LANES for details on the pipsline. - * - * XXH3_64bits_withSeed, len == 256, Snapdragon 835 - * without hack: 2654.4 MB/s - * with hack: 3202.9 MB/s - */ - XXH_COMPILER_GUARD(kSecretPtr); -#endif - /* - * Note: in debug mode, this overrides the asm optimization - * and Clang will emit MOVK chains again. - */ - XXH_ASSERT(kSecretPtr == XXH3_kSecret); - - { int const nbRounds = XXH_SECRET_DEFAULT_SIZE / 16; - int i; - for (i=0; i < nbRounds; i++) { - /* - * The asm hack causes Clang to assume that kSecretPtr aliases with - * customSecret, and on aarch64, this prevented LDP from merging two - * loads together for free. Putting the loads together before the stores - * properly generates LDP. - */ - xxh_u64 lo = XXH_readLE64(kSecretPtr + 16*i) + seed64; - xxh_u64 hi = XXH_readLE64(kSecretPtr + 16*i + 8) - seed64; - XXH_writeLE64((xxh_u8*)customSecret + 16*i, lo); - XXH_writeLE64((xxh_u8*)customSecret + 16*i + 8, hi); - } } -} - - -typedef void (*XXH3_f_accumulate_512)(void* XXH_RESTRICT, const void*, const void*); -typedef void (*XXH3_f_scrambleAcc)(void* XXH_RESTRICT, const void*); -typedef void (*XXH3_f_initCustomSecret)(void* XXH_RESTRICT, xxh_u64); - - -#if (XXH_VECTOR == XXH_AVX512) - -#define XXH3_accumulate_512 XXH3_accumulate_512_avx512 -#define XXH3_scrambleAcc XXH3_scrambleAcc_avx512 -#define XXH3_initCustomSecret XXH3_initCustomSecret_avx512 - -#elif (XXH_VECTOR == XXH_AVX2) - -#define XXH3_accumulate_512 XXH3_accumulate_512_avx2 -#define XXH3_scrambleAcc XXH3_scrambleAcc_avx2 -#define XXH3_initCustomSecret XXH3_initCustomSecret_avx2 - -#elif (XXH_VECTOR == XXH_SSE2) - -#define XXH3_accumulate_512 XXH3_accumulate_512_sse2 -#define XXH3_scrambleAcc XXH3_scrambleAcc_sse2 -#define XXH3_initCustomSecret XXH3_initCustomSecret_sse2 - -#elif (XXH_VECTOR == XXH_NEON) - -#define XXH3_accumulate_512 XXH3_accumulate_512_neon -#define XXH3_scrambleAcc XXH3_scrambleAcc_neon -#define XXH3_initCustomSecret XXH3_initCustomSecret_scalar - -#elif (XXH_VECTOR == XXH_VSX) - -#define XXH3_accumulate_512 XXH3_accumulate_512_vsx -#define XXH3_scrambleAcc XXH3_scrambleAcc_vsx -#define XXH3_initCustomSecret XXH3_initCustomSecret_scalar - -#else /* scalar */ - -#define XXH3_accumulate_512 XXH3_accumulate_512_scalar -#define XXH3_scrambleAcc XXH3_scrambleAcc_scalar -#define XXH3_initCustomSecret XXH3_initCustomSecret_scalar - -#endif - - - -#ifndef XXH_PREFETCH_DIST -# ifdef __clang__ -# define XXH_PREFETCH_DIST 320 -# else -# if (XXH_VECTOR == XXH_AVX512) -# define XXH_PREFETCH_DIST 512 -# else -# define XXH_PREFETCH_DIST 384 -# endif -# endif /* __clang__ */ -#endif /* XXH_PREFETCH_DIST */ - -/* - * XXH3_accumulate() - * Loops over XXH3_accumulate_512(). - * Assumption: nbStripes will not overflow the secret size - */ -XXH_FORCE_INLINE void -XXH3_accumulate( xxh_u64* XXH_RESTRICT acc, - const xxh_u8* XXH_RESTRICT input, - const xxh_u8* XXH_RESTRICT secret, - size_t nbStripes, - XXH3_f_accumulate_512 f_acc512) -{ - size_t n; - for (n = 0; n < nbStripes; n++ ) { - const xxh_u8* const in = input + n*XXH_STRIPE_LEN; - XXH_PREFETCH(in + XXH_PREFETCH_DIST); - f_acc512(acc, - in, - secret + n*XXH_SECRET_CONSUME_RATE); - } -} - -XXH_FORCE_INLINE void -XXH3_hashLong_internal_loop(xxh_u64* XXH_RESTRICT acc, - const xxh_u8* XXH_RESTRICT input, size_t len, - const xxh_u8* XXH_RESTRICT secret, size_t secretSize, - XXH3_f_accumulate_512 f_acc512, - XXH3_f_scrambleAcc f_scramble) -{ - size_t const nbStripesPerBlock = (secretSize - XXH_STRIPE_LEN) / XXH_SECRET_CONSUME_RATE; - size_t const block_len = XXH_STRIPE_LEN * nbStripesPerBlock; - size_t const nb_blocks = (len - 1) / block_len; - - size_t n; - - XXH_ASSERT(secretSize >= XXH3_SECRET_SIZE_MIN); - - for (n = 0; n < nb_blocks; n++) { - XXH3_accumulate(acc, input + n*block_len, secret, nbStripesPerBlock, f_acc512); - f_scramble(acc, secret + secretSize - XXH_STRIPE_LEN); - } - - /* last partial block */ - XXH_ASSERT(len > XXH_STRIPE_LEN); - { size_t const nbStripes = ((len - 1) - (block_len * nb_blocks)) / XXH_STRIPE_LEN; - XXH_ASSERT(nbStripes <= (secretSize / XXH_SECRET_CONSUME_RATE)); - XXH3_accumulate(acc, input + nb_blocks*block_len, secret, nbStripes, f_acc512); - - /* last stripe */ - { const xxh_u8* const p = input + len - XXH_STRIPE_LEN; -#define XXH_SECRET_LASTACC_START 7 /* not aligned on 8, last secret is different from acc & scrambler */ - f_acc512(acc, p, secret + secretSize - XXH_STRIPE_LEN - XXH_SECRET_LASTACC_START); - } } -} - -XXH_FORCE_INLINE xxh_u64 -XXH3_mix2Accs(const xxh_u64* XXH_RESTRICT acc, const xxh_u8* XXH_RESTRICT secret) -{ - return XXH3_mul128_fold64( - acc[0] ^ XXH_readLE64(secret), - acc[1] ^ XXH_readLE64(secret+8) ); -} - -static XXH64_hash_t -XXH3_mergeAccs(const xxh_u64* XXH_RESTRICT acc, const xxh_u8* XXH_RESTRICT secret, xxh_u64 start) -{ - xxh_u64 result64 = start; - size_t i = 0; - - for (i = 0; i < 4; i++) { - result64 += XXH3_mix2Accs(acc+2*i, secret + 16*i); -#if defined(__clang__) /* Clang */ \ - && (defined(__arm__) || defined(__thumb__)) /* ARMv7 */ \ - && (defined(__ARM_NEON) || defined(__ARM_NEON__)) /* NEON */ \ - && !defined(XXH_ENABLE_AUTOVECTORIZE) /* Define to disable */ - /* - * UGLY HACK: - * Prevent autovectorization on Clang ARMv7-a. Exact same problem as - * the one in XXH3_len_129to240_64b. Speeds up shorter keys > 240b. - * XXH3_64bits, len == 256, Snapdragon 835: - * without hack: 2063.7 MB/s - * with hack: 2560.7 MB/s - */ - XXH_COMPILER_GUARD(result64); -#endif - } - - return XXH3_avalanche(result64); -} - -#define XXH3_INIT_ACC { XXH_PRIME32_3, XXH_PRIME64_1, XXH_PRIME64_2, XXH_PRIME64_3, \ - XXH_PRIME64_4, XXH_PRIME32_2, XXH_PRIME64_5, XXH_PRIME32_1 } - -XXH_FORCE_INLINE XXH64_hash_t -XXH3_hashLong_64b_internal(const void* XXH_RESTRICT input, size_t len, - const void* XXH_RESTRICT secret, size_t secretSize, - XXH3_f_accumulate_512 f_acc512, - XXH3_f_scrambleAcc f_scramble) -{ - XXH_ALIGN(XXH_ACC_ALIGN) xxh_u64 acc[XXH_ACC_NB] = XXH3_INIT_ACC; - - XXH3_hashLong_internal_loop(acc, (const xxh_u8*)input, len, (const xxh_u8*)secret, secretSize, f_acc512, f_scramble); - - /* converge into final hash */ - XXH_STATIC_ASSERT(sizeof(acc) == 64); - /* do not align on 8, so that the secret is different from the accumulator */ -#define XXH_SECRET_MERGEACCS_START 11 - XXH_ASSERT(secretSize >= sizeof(acc) + XXH_SECRET_MERGEACCS_START); - return XXH3_mergeAccs(acc, (const xxh_u8*)secret + XXH_SECRET_MERGEACCS_START, (xxh_u64)len * XXH_PRIME64_1); -} - -/* - * It's important for performance to transmit secret's size (when it's static) - * so that the compiler can properly optimize the vectorized loop. - * This makes a big performance difference for "medium" keys (<1 KB) when using AVX instruction set. - */ -XXH_FORCE_INLINE XXH64_hash_t -XXH3_hashLong_64b_withSecret(const void* XXH_RESTRICT input, size_t len, - XXH64_hash_t seed64, const xxh_u8* XXH_RESTRICT secret, size_t secretLen) -{ - (void)seed64; - return XXH3_hashLong_64b_internal(input, len, secret, secretLen, XXH3_accumulate_512, XXH3_scrambleAcc); -} - -/* - * It's preferable for performance that XXH3_hashLong is not inlined, - * as it results in a smaller function for small data, easier to the instruction cache. - * Note that inside this no_inline function, we do inline the internal loop, - * and provide a statically defined secret size to allow optimization of vector loop. - */ -XXH_NO_INLINE XXH64_hash_t -XXH3_hashLong_64b_default(const void* XXH_RESTRICT input, size_t len, - XXH64_hash_t seed64, const xxh_u8* XXH_RESTRICT secret, size_t secretLen) -{ - (void)seed64; (void)secret; (void)secretLen; - return XXH3_hashLong_64b_internal(input, len, XXH3_kSecret, sizeof(XXH3_kSecret), XXH3_accumulate_512, XXH3_scrambleAcc); -} - -/* - * XXH3_hashLong_64b_withSeed(): - * Generate a custom key based on alteration of default XXH3_kSecret with the seed, - * and then use this key for long mode hashing. - * - * This operation is decently fast but nonetheless costs a little bit of time. - * Try to avoid it whenever possible (typically when seed==0). - * - * It's important for performance that XXH3_hashLong is not inlined. Not sure - * why (uop cache maybe?), but the difference is large and easily measurable. - */ -XXH_FORCE_INLINE XXH64_hash_t -XXH3_hashLong_64b_withSeed_internal(const void* input, size_t len, - XXH64_hash_t seed, - XXH3_f_accumulate_512 f_acc512, - XXH3_f_scrambleAcc f_scramble, - XXH3_f_initCustomSecret f_initSec) -{ - if (seed == 0) - return XXH3_hashLong_64b_internal(input, len, - XXH3_kSecret, sizeof(XXH3_kSecret), - f_acc512, f_scramble); - { XXH_ALIGN(XXH_SEC_ALIGN) xxh_u8 secret[XXH_SECRET_DEFAULT_SIZE]; - f_initSec(secret, seed); - return XXH3_hashLong_64b_internal(input, len, secret, sizeof(secret), - f_acc512, f_scramble); - } -} - -/* - * It's important for performance that XXH3_hashLong is not inlined. - */ -XXH_NO_INLINE XXH64_hash_t -XXH3_hashLong_64b_withSeed(const void* input, size_t len, - XXH64_hash_t seed, const xxh_u8* secret, size_t secretLen) -{ - (void)secret; (void)secretLen; - return XXH3_hashLong_64b_withSeed_internal(input, len, seed, - XXH3_accumulate_512, XXH3_scrambleAcc, XXH3_initCustomSecret); -} - - -typedef XXH64_hash_t (*XXH3_hashLong64_f)(const void* XXH_RESTRICT, size_t, - XXH64_hash_t, const xxh_u8* XXH_RESTRICT, size_t); - -XXH_FORCE_INLINE XXH64_hash_t -XXH3_64bits_internal(const void* XXH_RESTRICT input, size_t len, - XXH64_hash_t seed64, const void* XXH_RESTRICT secret, size_t secretLen, - XXH3_hashLong64_f f_hashLong) -{ - XXH_ASSERT(secretLen >= XXH3_SECRET_SIZE_MIN); - /* - * If an action is to be taken if `secretLen` condition is not respected, - * it should be done here. - * For now, it's a contract pre-condition. - * Adding a check and a branch here would cost performance at every hash. - * Also, note that function signature doesn't offer room to return an error. - */ - if (len <= 16) - return XXH3_len_0to16_64b((const xxh_u8*)input, len, (const xxh_u8*)secret, seed64); - if (len <= 128) - return XXH3_len_17to128_64b((const xxh_u8*)input, len, (const xxh_u8*)secret, secretLen, seed64); - if (len <= XXH3_MIDSIZE_MAX) - return XXH3_len_129to240_64b((const xxh_u8*)input, len, (const xxh_u8*)secret, secretLen, seed64); - return f_hashLong(input, len, seed64, (const xxh_u8*)secret, secretLen); -} - - -/* === Public entry point === */ - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH64_hash_t XXH3_64bits(const void* input, size_t len) -{ - return XXH3_64bits_internal(input, len, 0, XXH3_kSecret, sizeof(XXH3_kSecret), XXH3_hashLong_64b_default); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH64_hash_t -XXH3_64bits_withSecret(const void* input, size_t len, const void* secret, size_t secretSize) -{ - return XXH3_64bits_internal(input, len, 0, secret, secretSize, XXH3_hashLong_64b_withSecret); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH64_hash_t -XXH3_64bits_withSeed(const void* input, size_t len, XXH64_hash_t seed) -{ - return XXH3_64bits_internal(input, len, seed, XXH3_kSecret, sizeof(XXH3_kSecret), XXH3_hashLong_64b_withSeed); -} - -XXH_PUBLIC_API XXH64_hash_t -XXH3_64bits_withSecretandSeed(const void* input, size_t len, const void* secret, size_t secretSize, XXH64_hash_t seed) -{ - if (len <= XXH3_MIDSIZE_MAX) - return XXH3_64bits_internal(input, len, seed, XXH3_kSecret, sizeof(XXH3_kSecret), NULL); - return XXH3_hashLong_64b_withSecret(input, len, seed, (const xxh_u8*)secret, secretSize); -} - - -/* === XXH3 streaming === */ - -/* - * Malloc's a pointer that is always aligned to align. - * - * This must be freed with `XXH_alignedFree()`. - * - * malloc typically guarantees 16 byte alignment on 64-bit systems and 8 byte - * alignment on 32-bit. This isn't enough for the 32 byte aligned loads in AVX2 - * or on 32-bit, the 16 byte aligned loads in SSE2 and NEON. - * - * This underalignment previously caused a rather obvious crash which went - * completely unnoticed due to XXH3_createState() not actually being tested. - * Credit to RedSpah for noticing this bug. - * - * The alignment is done manually: Functions like posix_memalign or _mm_malloc - * are avoided: To maintain portability, we would have to write a fallback - * like this anyways, and besides, testing for the existence of library - * functions without relying on external build tools is impossible. - * - * The method is simple: Overallocate, manually align, and store the offset - * to the original behind the returned pointer. - * - * Align must be a power of 2 and 8 <= align <= 128. - */ -static void* XXH_alignedMalloc(size_t s, size_t align) -{ - XXH_ASSERT(align <= 128 && align >= 8); /* range check */ - XXH_ASSERT((align & (align-1)) == 0); /* power of 2 */ - XXH_ASSERT(s != 0 && s < (s + align)); /* empty/overflow */ - { /* Overallocate to make room for manual realignment and an offset byte */ - xxh_u8* base = (xxh_u8*)XXH_malloc(s + align); - if (base != NULL) { - /* - * Get the offset needed to align this pointer. - * - * Even if the returned pointer is aligned, there will always be - * at least one byte to store the offset to the original pointer. - */ - size_t offset = align - ((size_t)base & (align - 1)); /* base % align */ - /* Add the offset for the now-aligned pointer */ - xxh_u8* ptr = base + offset; - - XXH_ASSERT((size_t)ptr % align == 0); - - /* Store the offset immediately before the returned pointer. */ - ptr[-1] = (xxh_u8)offset; - return ptr; - } - return NULL; - } -} -/* - * Frees an aligned pointer allocated by XXH_alignedMalloc(). Don't pass - * normal malloc'd pointers, XXH_alignedMalloc has a specific data layout. - */ -static void XXH_alignedFree(void* p) -{ - if (p != NULL) { - xxh_u8* ptr = (xxh_u8*)p; - /* Get the offset byte we added in XXH_malloc. */ - xxh_u8 offset = ptr[-1]; - /* Free the original malloc'd pointer */ - xxh_u8* base = ptr - offset; - XXH_free(base); - } -} -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH3_state_t* XXH3_createState(void) -{ - XXH3_state_t* const state = (XXH3_state_t*)XXH_alignedMalloc(sizeof(XXH3_state_t), 64); - if (state==NULL) return NULL; - XXH3_INITSTATE(state); - return state; -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH_errorcode XXH3_freeState(XXH3_state_t* statePtr) -{ - XXH_alignedFree(statePtr); - return XXH_OK; -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API void -XXH3_copyState(XXH3_state_t* dst_state, const XXH3_state_t* src_state) -{ - XXH_memcpy(dst_state, src_state, sizeof(*dst_state)); -} - -static void -XXH3_reset_internal(XXH3_state_t* statePtr, - XXH64_hash_t seed, - const void* secret, size_t secretSize) -{ - size_t const initStart = offsetof(XXH3_state_t, bufferedSize); - size_t const initLength = offsetof(XXH3_state_t, nbStripesPerBlock) - initStart; - XXH_ASSERT(offsetof(XXH3_state_t, nbStripesPerBlock) > initStart); - XXH_ASSERT(statePtr != NULL); - /* set members from bufferedSize to nbStripesPerBlock (excluded) to 0 */ - memset((char*)statePtr + initStart, 0, initLength); - statePtr->acc[0] = XXH_PRIME32_3; - statePtr->acc[1] = XXH_PRIME64_1; - statePtr->acc[2] = XXH_PRIME64_2; - statePtr->acc[3] = XXH_PRIME64_3; - statePtr->acc[4] = XXH_PRIME64_4; - statePtr->acc[5] = XXH_PRIME32_2; - statePtr->acc[6] = XXH_PRIME64_5; - statePtr->acc[7] = XXH_PRIME32_1; - statePtr->seed = seed; - statePtr->useSeed = (seed != 0); - statePtr->extSecret = (const unsigned char*)secret; - XXH_ASSERT(secretSize >= XXH3_SECRET_SIZE_MIN); - statePtr->secretLimit = secretSize - XXH_STRIPE_LEN; - statePtr->nbStripesPerBlock = statePtr->secretLimit / XXH_SECRET_CONSUME_RATE; -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH_errorcode -XXH3_64bits_reset(XXH3_state_t* statePtr) -{ - if (statePtr == NULL) return XXH_ERROR; - XXH3_reset_internal(statePtr, 0, XXH3_kSecret, XXH_SECRET_DEFAULT_SIZE); - return XXH_OK; -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH_errorcode -XXH3_64bits_reset_withSecret(XXH3_state_t* statePtr, const void* secret, size_t secretSize) -{ - if (statePtr == NULL) return XXH_ERROR; - XXH3_reset_internal(statePtr, 0, secret, secretSize); - if (secret == NULL) return XXH_ERROR; - if (secretSize < XXH3_SECRET_SIZE_MIN) return XXH_ERROR; - return XXH_OK; -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH_errorcode -XXH3_64bits_reset_withSeed(XXH3_state_t* statePtr, XXH64_hash_t seed) -{ - if (statePtr == NULL) return XXH_ERROR; - if (seed==0) return XXH3_64bits_reset(statePtr); - if ((seed != statePtr->seed) || (statePtr->extSecret != NULL)) - XXH3_initCustomSecret(statePtr->customSecret, seed); - XXH3_reset_internal(statePtr, seed, NULL, XXH_SECRET_DEFAULT_SIZE); - return XXH_OK; -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH_errorcode -XXH3_64bits_reset_withSecretandSeed(XXH3_state_t* statePtr, const void* secret, size_t secretSize, XXH64_hash_t seed64) -{ - if (statePtr == NULL) return XXH_ERROR; - if (secret == NULL) return XXH_ERROR; - if (secretSize < XXH3_SECRET_SIZE_MIN) return XXH_ERROR; - XXH3_reset_internal(statePtr, seed64, secret, secretSize); - statePtr->useSeed = 1; /* always, even if seed64==0 */ - return XXH_OK; -} - -/* Note : when XXH3_consumeStripes() is invoked, - * there must be a guarantee that at least one more byte must be consumed from input - * so that the function can blindly consume all stripes using the "normal" secret segment */ -XXH_FORCE_INLINE void -XXH3_consumeStripes(xxh_u64* XXH_RESTRICT acc, - size_t* XXH_RESTRICT nbStripesSoFarPtr, size_t nbStripesPerBlock, - const xxh_u8* XXH_RESTRICT input, size_t nbStripes, - const xxh_u8* XXH_RESTRICT secret, size_t secretLimit, - XXH3_f_accumulate_512 f_acc512, - XXH3_f_scrambleAcc f_scramble) -{ - XXH_ASSERT(nbStripes <= nbStripesPerBlock); /* can handle max 1 scramble per invocation */ - XXH_ASSERT(*nbStripesSoFarPtr < nbStripesPerBlock); - if (nbStripesPerBlock - *nbStripesSoFarPtr <= nbStripes) { - /* need a scrambling operation */ - size_t const nbStripesToEndofBlock = nbStripesPerBlock - *nbStripesSoFarPtr; - size_t const nbStripesAfterBlock = nbStripes - nbStripesToEndofBlock; - XXH3_accumulate(acc, input, secret + nbStripesSoFarPtr[0] * XXH_SECRET_CONSUME_RATE, nbStripesToEndofBlock, f_acc512); - f_scramble(acc, secret + secretLimit); - XXH3_accumulate(acc, input + nbStripesToEndofBlock * XXH_STRIPE_LEN, secret, nbStripesAfterBlock, f_acc512); - *nbStripesSoFarPtr = nbStripesAfterBlock; - } else { - XXH3_accumulate(acc, input, secret + nbStripesSoFarPtr[0] * XXH_SECRET_CONSUME_RATE, nbStripes, f_acc512); - *nbStripesSoFarPtr += nbStripes; - } -} - -#ifndef XXH3_STREAM_USE_STACK -# ifndef __clang__ /* clang doesn't need additional stack space */ -# define XXH3_STREAM_USE_STACK 1 -# endif -#endif -/* - * Both XXH3_64bits_update and XXH3_128bits_update use this routine. - */ -XXH_FORCE_INLINE XXH_errorcode -XXH3_update(XXH3_state_t* XXH_RESTRICT const state, - const xxh_u8* XXH_RESTRICT input, size_t len, - XXH3_f_accumulate_512 f_acc512, - XXH3_f_scrambleAcc f_scramble) -{ - if (input==NULL) { - XXH_ASSERT(len == 0); - return XXH_OK; - } - - XXH_ASSERT(state != NULL); - { const xxh_u8* const bEnd = input + len; - const unsigned char* const secret = (state->extSecret == NULL) ? state->customSecret : state->extSecret; -#if defined(XXH3_STREAM_USE_STACK) && XXH3_STREAM_USE_STACK >= 1 - /* For some reason, gcc and MSVC seem to suffer greatly - * when operating accumulators directly into state. - * Operating into stack space seems to enable proper optimization. - * clang, on the other hand, doesn't seem to need this trick */ - XXH_ALIGN(XXH_ACC_ALIGN) xxh_u64 acc[8]; memcpy(acc, state->acc, sizeof(acc)); -#else - xxh_u64* XXH_RESTRICT const acc = state->acc; -#endif - state->totalLen += len; - XXH_ASSERT(state->bufferedSize <= XXH3_INTERNALBUFFER_SIZE); - - /* small input : just fill in tmp buffer */ - if (state->bufferedSize + len <= XXH3_INTERNALBUFFER_SIZE) { - XXH_memcpy(state->buffer + state->bufferedSize, input, len); - state->bufferedSize += (XXH32_hash_t)len; - return XXH_OK; - } - - /* total input is now > XXH3_INTERNALBUFFER_SIZE */ - #define XXH3_INTERNALBUFFER_STRIPES (XXH3_INTERNALBUFFER_SIZE / XXH_STRIPE_LEN) - XXH_STATIC_ASSERT(XXH3_INTERNALBUFFER_SIZE % XXH_STRIPE_LEN == 0); /* clean multiple */ - - /* - * Internal buffer is partially filled (always, except at beginning) - * Complete it, then consume it. - */ - if (state->bufferedSize) { - size_t const loadSize = XXH3_INTERNALBUFFER_SIZE - state->bufferedSize; - XXH_memcpy(state->buffer + state->bufferedSize, input, loadSize); - input += loadSize; - XXH3_consumeStripes(acc, - &state->nbStripesSoFar, state->nbStripesPerBlock, - state->buffer, XXH3_INTERNALBUFFER_STRIPES, - secret, state->secretLimit, - f_acc512, f_scramble); - state->bufferedSize = 0; - } - XXH_ASSERT(input < bEnd); - - /* large input to consume : ingest per full block */ - if ((size_t)(bEnd - input) > state->nbStripesPerBlock * XXH_STRIPE_LEN) { - size_t nbStripes = (size_t)(bEnd - 1 - input) / XXH_STRIPE_LEN; - XXH_ASSERT(state->nbStripesPerBlock >= state->nbStripesSoFar); - /* join to current block's end */ - { size_t const nbStripesToEnd = state->nbStripesPerBlock - state->nbStripesSoFar; - XXH_ASSERT(nbStripesToEnd <= nbStripes); - XXH3_accumulate(acc, input, secret + state->nbStripesSoFar * XXH_SECRET_CONSUME_RATE, nbStripesToEnd, f_acc512); - f_scramble(acc, secret + state->secretLimit); - state->nbStripesSoFar = 0; - input += nbStripesToEnd * XXH_STRIPE_LEN; - nbStripes -= nbStripesToEnd; - } - /* consume per entire blocks */ - while(nbStripes >= state->nbStripesPerBlock) { - XXH3_accumulate(acc, input, secret, state->nbStripesPerBlock, f_acc512); - f_scramble(acc, secret + state->secretLimit); - input += state->nbStripesPerBlock * XXH_STRIPE_LEN; - nbStripes -= state->nbStripesPerBlock; - } - /* consume last partial block */ - XXH3_accumulate(acc, input, secret, nbStripes, f_acc512); - input += nbStripes * XXH_STRIPE_LEN; - XXH_ASSERT(input < bEnd); /* at least some bytes left */ - state->nbStripesSoFar = nbStripes; - /* buffer predecessor of last partial stripe */ - XXH_memcpy(state->buffer + sizeof(state->buffer) - XXH_STRIPE_LEN, input - XXH_STRIPE_LEN, XXH_STRIPE_LEN); - XXH_ASSERT(bEnd - input <= XXH_STRIPE_LEN); - } else { - /* content to consume <= block size */ - /* Consume input by a multiple of internal buffer size */ - if (bEnd - input > XXH3_INTERNALBUFFER_SIZE) { - const xxh_u8* const limit = bEnd - XXH3_INTERNALBUFFER_SIZE; - do { - XXH3_consumeStripes(acc, - &state->nbStripesSoFar, state->nbStripesPerBlock, - input, XXH3_INTERNALBUFFER_STRIPES, - secret, state->secretLimit, - f_acc512, f_scramble); - input += XXH3_INTERNALBUFFER_SIZE; - } while (inputbuffer + sizeof(state->buffer) - XXH_STRIPE_LEN, input - XXH_STRIPE_LEN, XXH_STRIPE_LEN); - } - } - - /* Some remaining input (always) : buffer it */ - XXH_ASSERT(input < bEnd); - XXH_ASSERT(bEnd - input <= XXH3_INTERNALBUFFER_SIZE); - XXH_ASSERT(state->bufferedSize == 0); - XXH_memcpy(state->buffer, input, (size_t)(bEnd-input)); - state->bufferedSize = (XXH32_hash_t)(bEnd-input); -#if defined(XXH3_STREAM_USE_STACK) && XXH3_STREAM_USE_STACK >= 1 - /* save stack accumulators into state */ - memcpy(state->acc, acc, sizeof(acc)); -#endif - } - - return XXH_OK; -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH_errorcode -XXH3_64bits_update(XXH3_state_t* state, const void* input, size_t len) -{ - return XXH3_update(state, (const xxh_u8*)input, len, - XXH3_accumulate_512, XXH3_scrambleAcc); -} - - -XXH_FORCE_INLINE void -XXH3_digest_long (XXH64_hash_t* acc, - const XXH3_state_t* state, - const unsigned char* secret) -{ - /* - * Digest on a local copy. This way, the state remains unaltered, and it can - * continue ingesting more input afterwards. - */ - XXH_memcpy(acc, state->acc, sizeof(state->acc)); - if (state->bufferedSize >= XXH_STRIPE_LEN) { - size_t const nbStripes = (state->bufferedSize - 1) / XXH_STRIPE_LEN; - size_t nbStripesSoFar = state->nbStripesSoFar; - XXH3_consumeStripes(acc, - &nbStripesSoFar, state->nbStripesPerBlock, - state->buffer, nbStripes, - secret, state->secretLimit, - XXH3_accumulate_512, XXH3_scrambleAcc); - /* last stripe */ - XXH3_accumulate_512(acc, - state->buffer + state->bufferedSize - XXH_STRIPE_LEN, - secret + state->secretLimit - XXH_SECRET_LASTACC_START); - } else { /* bufferedSize < XXH_STRIPE_LEN */ - xxh_u8 lastStripe[XXH_STRIPE_LEN]; - size_t const catchupSize = XXH_STRIPE_LEN - state->bufferedSize; - XXH_ASSERT(state->bufferedSize > 0); /* there is always some input buffered */ - XXH_memcpy(lastStripe, state->buffer + sizeof(state->buffer) - catchupSize, catchupSize); - XXH_memcpy(lastStripe + catchupSize, state->buffer, state->bufferedSize); - XXH3_accumulate_512(acc, - lastStripe, - secret + state->secretLimit - XXH_SECRET_LASTACC_START); - } -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH64_hash_t XXH3_64bits_digest (const XXH3_state_t* state) -{ - const unsigned char* const secret = (state->extSecret == NULL) ? state->customSecret : state->extSecret; - if (state->totalLen > XXH3_MIDSIZE_MAX) { - XXH_ALIGN(XXH_ACC_ALIGN) XXH64_hash_t acc[XXH_ACC_NB]; - XXH3_digest_long(acc, state, secret); - return XXH3_mergeAccs(acc, - secret + XXH_SECRET_MERGEACCS_START, - (xxh_u64)state->totalLen * XXH_PRIME64_1); - } - /* totalLen <= XXH3_MIDSIZE_MAX: digesting a short input */ - if (state->useSeed) - return XXH3_64bits_withSeed(state->buffer, (size_t)state->totalLen, state->seed); - return XXH3_64bits_withSecret(state->buffer, (size_t)(state->totalLen), - secret, state->secretLimit + XXH_STRIPE_LEN); -} - - - -/* ========================================== - * XXH3 128 bits (a.k.a XXH128) - * ========================================== - * XXH3's 128-bit variant has better mixing and strength than the 64-bit variant, - * even without counting the significantly larger output size. - * - * For example, extra steps are taken to avoid the seed-dependent collisions - * in 17-240 byte inputs (See XXH3_mix16B and XXH128_mix32B). - * - * This strength naturally comes at the cost of some speed, especially on short - * lengths. Note that longer hashes are about as fast as the 64-bit version - * due to it using only a slight modification of the 64-bit loop. - * - * XXH128 is also more oriented towards 64-bit machines. It is still extremely - * fast for a _128-bit_ hash on 32-bit (it usually clears XXH64). - */ - -XXH_FORCE_INLINE XXH128_hash_t -XXH3_len_1to3_128b(const xxh_u8* input, size_t len, const xxh_u8* secret, XXH64_hash_t seed) -{ - /* A doubled version of 1to3_64b with different constants. */ - XXH_ASSERT(input != NULL); - XXH_ASSERT(1 <= len && len <= 3); - XXH_ASSERT(secret != NULL); - /* - * len = 1: combinedl = { input[0], 0x01, input[0], input[0] } - * len = 2: combinedl = { input[1], 0x02, input[0], input[1] } - * len = 3: combinedl = { input[2], 0x03, input[0], input[1] } - */ - { xxh_u8 const c1 = input[0]; - xxh_u8 const c2 = input[len >> 1]; - xxh_u8 const c3 = input[len - 1]; - xxh_u32 const combinedl = ((xxh_u32)c1 <<16) | ((xxh_u32)c2 << 24) - | ((xxh_u32)c3 << 0) | ((xxh_u32)len << 8); - xxh_u32 const combinedh = XXH_rotl32(XXH_swap32(combinedl), 13); - xxh_u64 const bitflipl = (XXH_readLE32(secret) ^ XXH_readLE32(secret+4)) + seed; - xxh_u64 const bitfliph = (XXH_readLE32(secret+8) ^ XXH_readLE32(secret+12)) - seed; - xxh_u64 const keyed_lo = (xxh_u64)combinedl ^ bitflipl; - xxh_u64 const keyed_hi = (xxh_u64)combinedh ^ bitfliph; - XXH128_hash_t h128; - h128.low64 = XXH64_avalanche(keyed_lo); - h128.high64 = XXH64_avalanche(keyed_hi); - return h128; - } -} - -XXH_FORCE_INLINE XXH128_hash_t -XXH3_len_4to8_128b(const xxh_u8* input, size_t len, const xxh_u8* secret, XXH64_hash_t seed) -{ - XXH_ASSERT(input != NULL); - XXH_ASSERT(secret != NULL); - XXH_ASSERT(4 <= len && len <= 8); - seed ^= (xxh_u64)XXH_swap32((xxh_u32)seed) << 32; - { xxh_u32 const input_lo = XXH_readLE32(input); - xxh_u32 const input_hi = XXH_readLE32(input + len - 4); - xxh_u64 const input_64 = input_lo + ((xxh_u64)input_hi << 32); - xxh_u64 const bitflip = (XXH_readLE64(secret+16) ^ XXH_readLE64(secret+24)) + seed; - xxh_u64 const keyed = input_64 ^ bitflip; - - /* Shift len to the left to ensure it is even, this avoids even multiplies. */ - XXH128_hash_t m128 = XXH_mult64to128(keyed, XXH_PRIME64_1 + (len << 2)); - - m128.high64 += (m128.low64 << 1); - m128.low64 ^= (m128.high64 >> 3); - - m128.low64 = XXH_xorshift64(m128.low64, 35); - m128.low64 *= 0x9FB21C651E98DF25ULL; - m128.low64 = XXH_xorshift64(m128.low64, 28); - m128.high64 = XXH3_avalanche(m128.high64); - return m128; - } -} - -XXH_FORCE_INLINE XXH128_hash_t -XXH3_len_9to16_128b(const xxh_u8* input, size_t len, const xxh_u8* secret, XXH64_hash_t seed) -{ - XXH_ASSERT(input != NULL); - XXH_ASSERT(secret != NULL); - XXH_ASSERT(9 <= len && len <= 16); - { xxh_u64 const bitflipl = (XXH_readLE64(secret+32) ^ XXH_readLE64(secret+40)) - seed; - xxh_u64 const bitfliph = (XXH_readLE64(secret+48) ^ XXH_readLE64(secret+56)) + seed; - xxh_u64 const input_lo = XXH_readLE64(input); - xxh_u64 input_hi = XXH_readLE64(input + len - 8); - XXH128_hash_t m128 = XXH_mult64to128(input_lo ^ input_hi ^ bitflipl, XXH_PRIME64_1); - /* - * Put len in the middle of m128 to ensure that the length gets mixed to - * both the low and high bits in the 128x64 multiply below. - */ - m128.low64 += (xxh_u64)(len - 1) << 54; - input_hi ^= bitfliph; - /* - * Add the high 32 bits of input_hi to the high 32 bits of m128, then - * add the long product of the low 32 bits of input_hi and XXH_PRIME32_2 to - * the high 64 bits of m128. - * - * The best approach to this operation is different on 32-bit and 64-bit. - */ - if (sizeof(void *) < sizeof(xxh_u64)) { /* 32-bit */ - /* - * 32-bit optimized version, which is more readable. - * - * On 32-bit, it removes an ADC and delays a dependency between the two - * halves of m128.high64, but it generates an extra mask on 64-bit. - */ - m128.high64 += (input_hi & 0xFFFFFFFF00000000ULL) + XXH_mult32to64((xxh_u32)input_hi, XXH_PRIME32_2); - } else { - /* - * 64-bit optimized (albeit more confusing) version. - * - * Uses some properties of addition and multiplication to remove the mask: - * - * Let: - * a = input_hi.lo = (input_hi & 0x00000000FFFFFFFF) - * b = input_hi.hi = (input_hi & 0xFFFFFFFF00000000) - * c = XXH_PRIME32_2 - * - * a + (b * c) - * Inverse Property: x + y - x == y - * a + (b * (1 + c - 1)) - * Distributive Property: x * (y + z) == (x * y) + (x * z) - * a + (b * 1) + (b * (c - 1)) - * Identity Property: x * 1 == x - * a + b + (b * (c - 1)) - * - * Substitute a, b, and c: - * input_hi.hi + input_hi.lo + ((xxh_u64)input_hi.lo * (XXH_PRIME32_2 - 1)) - * - * Since input_hi.hi + input_hi.lo == input_hi, we get this: - * input_hi + ((xxh_u64)input_hi.lo * (XXH_PRIME32_2 - 1)) - */ - m128.high64 += input_hi + XXH_mult32to64((xxh_u32)input_hi, XXH_PRIME32_2 - 1); - } - /* m128 ^= XXH_swap64(m128 >> 64); */ - m128.low64 ^= XXH_swap64(m128.high64); - - { /* 128x64 multiply: h128 = m128 * XXH_PRIME64_2; */ - XXH128_hash_t h128 = XXH_mult64to128(m128.low64, XXH_PRIME64_2); - h128.high64 += m128.high64 * XXH_PRIME64_2; - - h128.low64 = XXH3_avalanche(h128.low64); - h128.high64 = XXH3_avalanche(h128.high64); - return h128; - } } -} - -/* - * Assumption: `secret` size is >= XXH3_SECRET_SIZE_MIN - */ -XXH_FORCE_INLINE XXH128_hash_t -XXH3_len_0to16_128b(const xxh_u8* input, size_t len, const xxh_u8* secret, XXH64_hash_t seed) -{ - XXH_ASSERT(len <= 16); - { if (len > 8) return XXH3_len_9to16_128b(input, len, secret, seed); - if (len >= 4) return XXH3_len_4to8_128b(input, len, secret, seed); - if (len) return XXH3_len_1to3_128b(input, len, secret, seed); - { XXH128_hash_t h128; - xxh_u64 const bitflipl = XXH_readLE64(secret+64) ^ XXH_readLE64(secret+72); - xxh_u64 const bitfliph = XXH_readLE64(secret+80) ^ XXH_readLE64(secret+88); - h128.low64 = XXH64_avalanche(seed ^ bitflipl); - h128.high64 = XXH64_avalanche( seed ^ bitfliph); - return h128; - } } -} - -/* - * A bit slower than XXH3_mix16B, but handles multiply by zero better. - */ -XXH_FORCE_INLINE XXH128_hash_t -XXH128_mix32B(XXH128_hash_t acc, const xxh_u8* input_1, const xxh_u8* input_2, - const xxh_u8* secret, XXH64_hash_t seed) -{ - acc.low64 += XXH3_mix16B (input_1, secret+0, seed); - acc.low64 ^= XXH_readLE64(input_2) + XXH_readLE64(input_2 + 8); - acc.high64 += XXH3_mix16B (input_2, secret+16, seed); - acc.high64 ^= XXH_readLE64(input_1) + XXH_readLE64(input_1 + 8); - return acc; -} - - -XXH_FORCE_INLINE XXH128_hash_t -XXH3_len_17to128_128b(const xxh_u8* XXH_RESTRICT input, size_t len, - const xxh_u8* XXH_RESTRICT secret, size_t secretSize, - XXH64_hash_t seed) -{ - XXH_ASSERT(secretSize >= XXH3_SECRET_SIZE_MIN); (void)secretSize; - XXH_ASSERT(16 < len && len <= 128); - - { XXH128_hash_t acc; - acc.low64 = len * XXH_PRIME64_1; - acc.high64 = 0; - if (len > 32) { - if (len > 64) { - if (len > 96) { - acc = XXH128_mix32B(acc, input+48, input+len-64, secret+96, seed); - } - acc = XXH128_mix32B(acc, input+32, input+len-48, secret+64, seed); - } - acc = XXH128_mix32B(acc, input+16, input+len-32, secret+32, seed); - } - acc = XXH128_mix32B(acc, input, input+len-16, secret, seed); - { XXH128_hash_t h128; - h128.low64 = acc.low64 + acc.high64; - h128.high64 = (acc.low64 * XXH_PRIME64_1) - + (acc.high64 * XXH_PRIME64_4) - + ((len - seed) * XXH_PRIME64_2); - h128.low64 = XXH3_avalanche(h128.low64); - h128.high64 = (XXH64_hash_t)0 - XXH3_avalanche(h128.high64); - return h128; - } - } -} - -XXH_NO_INLINE XXH128_hash_t -XXH3_len_129to240_128b(const xxh_u8* XXH_RESTRICT input, size_t len, - const xxh_u8* XXH_RESTRICT secret, size_t secretSize, - XXH64_hash_t seed) -{ - XXH_ASSERT(secretSize >= XXH3_SECRET_SIZE_MIN); (void)secretSize; - XXH_ASSERT(128 < len && len <= XXH3_MIDSIZE_MAX); - - { XXH128_hash_t acc; - int const nbRounds = (int)len / 32; - int i; - acc.low64 = len * XXH_PRIME64_1; - acc.high64 = 0; - for (i=0; i<4; i++) { - acc = XXH128_mix32B(acc, - input + (32 * i), - input + (32 * i) + 16, - secret + (32 * i), - seed); - } - acc.low64 = XXH3_avalanche(acc.low64); - acc.high64 = XXH3_avalanche(acc.high64); - XXH_ASSERT(nbRounds >= 4); - for (i=4 ; i < nbRounds; i++) { - acc = XXH128_mix32B(acc, - input + (32 * i), - input + (32 * i) + 16, - secret + XXH3_MIDSIZE_STARTOFFSET + (32 * (i - 4)), - seed); - } - /* last bytes */ - acc = XXH128_mix32B(acc, - input + len - 16, - input + len - 32, - secret + XXH3_SECRET_SIZE_MIN - XXH3_MIDSIZE_LASTOFFSET - 16, - 0ULL - seed); - - { XXH128_hash_t h128; - h128.low64 = acc.low64 + acc.high64; - h128.high64 = (acc.low64 * XXH_PRIME64_1) - + (acc.high64 * XXH_PRIME64_4) - + ((len - seed) * XXH_PRIME64_2); - h128.low64 = XXH3_avalanche(h128.low64); - h128.high64 = (XXH64_hash_t)0 - XXH3_avalanche(h128.high64); - return h128; - } - } -} - -XXH_FORCE_INLINE XXH128_hash_t -XXH3_hashLong_128b_internal(const void* XXH_RESTRICT input, size_t len, - const xxh_u8* XXH_RESTRICT secret, size_t secretSize, - XXH3_f_accumulate_512 f_acc512, - XXH3_f_scrambleAcc f_scramble) -{ - XXH_ALIGN(XXH_ACC_ALIGN) xxh_u64 acc[XXH_ACC_NB] = XXH3_INIT_ACC; - - XXH3_hashLong_internal_loop(acc, (const xxh_u8*)input, len, secret, secretSize, f_acc512, f_scramble); - - /* converge into final hash */ - XXH_STATIC_ASSERT(sizeof(acc) == 64); - XXH_ASSERT(secretSize >= sizeof(acc) + XXH_SECRET_MERGEACCS_START); - { XXH128_hash_t h128; - h128.low64 = XXH3_mergeAccs(acc, - secret + XXH_SECRET_MERGEACCS_START, - (xxh_u64)len * XXH_PRIME64_1); - h128.high64 = XXH3_mergeAccs(acc, - secret + secretSize - - sizeof(acc) - XXH_SECRET_MERGEACCS_START, - ~((xxh_u64)len * XXH_PRIME64_2)); - return h128; - } -} - -/* - * It's important for performance that XXH3_hashLong is not inlined. - */ -XXH_NO_INLINE XXH128_hash_t -XXH3_hashLong_128b_default(const void* XXH_RESTRICT input, size_t len, - XXH64_hash_t seed64, - const void* XXH_RESTRICT secret, size_t secretLen) -{ - (void)seed64; (void)secret; (void)secretLen; - return XXH3_hashLong_128b_internal(input, len, XXH3_kSecret, sizeof(XXH3_kSecret), - XXH3_accumulate_512, XXH3_scrambleAcc); -} - -/* - * It's important for performance to pass @secretLen (when it's static) - * to the compiler, so that it can properly optimize the vectorized loop. - */ -XXH_FORCE_INLINE XXH128_hash_t -XXH3_hashLong_128b_withSecret(const void* XXH_RESTRICT input, size_t len, - XXH64_hash_t seed64, - const void* XXH_RESTRICT secret, size_t secretLen) -{ - (void)seed64; - return XXH3_hashLong_128b_internal(input, len, (const xxh_u8*)secret, secretLen, - XXH3_accumulate_512, XXH3_scrambleAcc); -} - -XXH_FORCE_INLINE XXH128_hash_t -XXH3_hashLong_128b_withSeed_internal(const void* XXH_RESTRICT input, size_t len, - XXH64_hash_t seed64, - XXH3_f_accumulate_512 f_acc512, - XXH3_f_scrambleAcc f_scramble, - XXH3_f_initCustomSecret f_initSec) -{ - if (seed64 == 0) - return XXH3_hashLong_128b_internal(input, len, - XXH3_kSecret, sizeof(XXH3_kSecret), - f_acc512, f_scramble); - { XXH_ALIGN(XXH_SEC_ALIGN) xxh_u8 secret[XXH_SECRET_DEFAULT_SIZE]; - f_initSec(secret, seed64); - return XXH3_hashLong_128b_internal(input, len, (const xxh_u8*)secret, sizeof(secret), - f_acc512, f_scramble); - } -} - -/* - * It's important for performance that XXH3_hashLong is not inlined. - */ -XXH_NO_INLINE XXH128_hash_t -XXH3_hashLong_128b_withSeed(const void* input, size_t len, - XXH64_hash_t seed64, const void* XXH_RESTRICT secret, size_t secretLen) -{ - (void)secret; (void)secretLen; - return XXH3_hashLong_128b_withSeed_internal(input, len, seed64, - XXH3_accumulate_512, XXH3_scrambleAcc, XXH3_initCustomSecret); -} - -typedef XXH128_hash_t (*XXH3_hashLong128_f)(const void* XXH_RESTRICT, size_t, - XXH64_hash_t, const void* XXH_RESTRICT, size_t); - -XXH_FORCE_INLINE XXH128_hash_t -XXH3_128bits_internal(const void* input, size_t len, - XXH64_hash_t seed64, const void* XXH_RESTRICT secret, size_t secretLen, - XXH3_hashLong128_f f_hl128) -{ - XXH_ASSERT(secretLen >= XXH3_SECRET_SIZE_MIN); - /* - * If an action is to be taken if `secret` conditions are not respected, - * it should be done here. - * For now, it's a contract pre-condition. - * Adding a check and a branch here would cost performance at every hash. - */ - if (len <= 16) - return XXH3_len_0to16_128b((const xxh_u8*)input, len, (const xxh_u8*)secret, seed64); - if (len <= 128) - return XXH3_len_17to128_128b((const xxh_u8*)input, len, (const xxh_u8*)secret, secretLen, seed64); - if (len <= XXH3_MIDSIZE_MAX) - return XXH3_len_129to240_128b((const xxh_u8*)input, len, (const xxh_u8*)secret, secretLen, seed64); - return f_hl128(input, len, seed64, secret, secretLen); -} - - -/* === Public XXH128 API === */ - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH128_hash_t XXH3_128bits(const void* input, size_t len) -{ - return XXH3_128bits_internal(input, len, 0, - XXH3_kSecret, sizeof(XXH3_kSecret), - XXH3_hashLong_128b_default); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH128_hash_t -XXH3_128bits_withSecret(const void* input, size_t len, const void* secret, size_t secretSize) -{ - return XXH3_128bits_internal(input, len, 0, - (const xxh_u8*)secret, secretSize, - XXH3_hashLong_128b_withSecret); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH128_hash_t -XXH3_128bits_withSeed(const void* input, size_t len, XXH64_hash_t seed) -{ - return XXH3_128bits_internal(input, len, seed, - XXH3_kSecret, sizeof(XXH3_kSecret), - XXH3_hashLong_128b_withSeed); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH128_hash_t -XXH3_128bits_withSecretandSeed(const void* input, size_t len, const void* secret, size_t secretSize, XXH64_hash_t seed) -{ - if (len <= XXH3_MIDSIZE_MAX) - return XXH3_128bits_internal(input, len, seed, XXH3_kSecret, sizeof(XXH3_kSecret), NULL); - return XXH3_hashLong_128b_withSecret(input, len, seed, secret, secretSize); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH128_hash_t -XXH128(const void* input, size_t len, XXH64_hash_t seed) -{ - return XXH3_128bits_withSeed(input, len, seed); -} - - -/* === XXH3 128-bit streaming === */ - -/* - * All initialization and update functions are identical to 64-bit streaming variant. - * The only difference is the finalization routine. - */ - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH_errorcode -XXH3_128bits_reset(XXH3_state_t* statePtr) -{ - return XXH3_64bits_reset(statePtr); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH_errorcode -XXH3_128bits_reset_withSecret(XXH3_state_t* statePtr, const void* secret, size_t secretSize) -{ - return XXH3_64bits_reset_withSecret(statePtr, secret, secretSize); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH_errorcode -XXH3_128bits_reset_withSeed(XXH3_state_t* statePtr, XXH64_hash_t seed) -{ - return XXH3_64bits_reset_withSeed(statePtr, seed); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH_errorcode -XXH3_128bits_reset_withSecretandSeed(XXH3_state_t* statePtr, const void* secret, size_t secretSize, XXH64_hash_t seed) -{ - return XXH3_64bits_reset_withSecretandSeed(statePtr, secret, secretSize, seed); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH_errorcode -XXH3_128bits_update(XXH3_state_t* state, const void* input, size_t len) -{ - return XXH3_update(state, (const xxh_u8*)input, len, - XXH3_accumulate_512, XXH3_scrambleAcc); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH128_hash_t XXH3_128bits_digest (const XXH3_state_t* state) -{ - const unsigned char* const secret = (state->extSecret == NULL) ? state->customSecret : state->extSecret; - if (state->totalLen > XXH3_MIDSIZE_MAX) { - XXH_ALIGN(XXH_ACC_ALIGN) XXH64_hash_t acc[XXH_ACC_NB]; - XXH3_digest_long(acc, state, secret); - XXH_ASSERT(state->secretLimit + XXH_STRIPE_LEN >= sizeof(acc) + XXH_SECRET_MERGEACCS_START); - { XXH128_hash_t h128; - h128.low64 = XXH3_mergeAccs(acc, - secret + XXH_SECRET_MERGEACCS_START, - (xxh_u64)state->totalLen * XXH_PRIME64_1); - h128.high64 = XXH3_mergeAccs(acc, - secret + state->secretLimit + XXH_STRIPE_LEN - - sizeof(acc) - XXH_SECRET_MERGEACCS_START, - ~((xxh_u64)state->totalLen * XXH_PRIME64_2)); - return h128; - } - } - /* len <= XXH3_MIDSIZE_MAX : short code */ - if (state->seed) - return XXH3_128bits_withSeed(state->buffer, (size_t)state->totalLen, state->seed); - return XXH3_128bits_withSecret(state->buffer, (size_t)(state->totalLen), - secret, state->secretLimit + XXH_STRIPE_LEN); -} - -/* 128-bit utility functions */ - -#include /* memcmp, memcpy */ - -/* return : 1 is equal, 0 if different */ -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API int XXH128_isEqual(XXH128_hash_t h1, XXH128_hash_t h2) -{ - /* note : XXH128_hash_t is compact, it has no padding byte */ - return !(memcmp(&h1, &h2, sizeof(h1))); -} - -/* This prototype is compatible with stdlib's qsort(). - * return : >0 if *h128_1 > *h128_2 - * <0 if *h128_1 < *h128_2 - * =0 if *h128_1 == *h128_2 */ -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API int XXH128_cmp(const void* h128_1, const void* h128_2) -{ - XXH128_hash_t const h1 = *(const XXH128_hash_t*)h128_1; - XXH128_hash_t const h2 = *(const XXH128_hash_t*)h128_2; - int const hcmp = (h1.high64 > h2.high64) - (h2.high64 > h1.high64); - /* note : bets that, in most cases, hash values are different */ - if (hcmp) return hcmp; - return (h1.low64 > h2.low64) - (h2.low64 > h1.low64); -} - - -/*====== Canonical representation ======*/ -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API void -XXH128_canonicalFromHash(XXH128_canonical_t* dst, XXH128_hash_t hash) -{ - XXH_STATIC_ASSERT(sizeof(XXH128_canonical_t) == sizeof(XXH128_hash_t)); - if (XXH_CPU_LITTLE_ENDIAN) { - hash.high64 = XXH_swap64(hash.high64); - hash.low64 = XXH_swap64(hash.low64); - } - XXH_memcpy(dst, &hash.high64, sizeof(hash.high64)); - XXH_memcpy((char*)dst + sizeof(hash.high64), &hash.low64, sizeof(hash.low64)); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH128_hash_t -XXH128_hashFromCanonical(const XXH128_canonical_t* src) -{ - XXH128_hash_t h; - h.high64 = XXH_readBE64(src); - h.low64 = XXH_readBE64(src->digest + 8); - return h; -} - - - -/* ========================================== - * Secret generators - * ========================================== - */ -#define XXH_MIN(x, y) (((x) > (y)) ? (y) : (x)) - -XXH_FORCE_INLINE void XXH3_combine16(void* dst, XXH128_hash_t h128) -{ - XXH_writeLE64( dst, XXH_readLE64(dst) ^ h128.low64 ); - XXH_writeLE64( (char*)dst+8, XXH_readLE64((char*)dst+8) ^ h128.high64 ); -} - -/*! @ingroup xxh3_family */ -XXH_PUBLIC_API XXH_errorcode -XXH3_generateSecret(void* secretBuffer, size_t secretSize, const void* customSeed, size_t customSeedSize) -{ -#if (XXH_DEBUGLEVEL >= 1) - XXH_ASSERT(secretBuffer != NULL); - XXH_ASSERT(secretSize >= XXH3_SECRET_SIZE_MIN); -#else - /* production mode, assert() are disabled */ - if (secretBuffer == NULL) return XXH_ERROR; - if (secretSize < XXH3_SECRET_SIZE_MIN) return XXH_ERROR; -#endif - - if (customSeedSize == 0) { - customSeed = XXH3_kSecret; - customSeedSize = XXH_SECRET_DEFAULT_SIZE; - } -#if (XXH_DEBUGLEVEL >= 1) - XXH_ASSERT(customSeed != NULL); -#else - if (customSeed == NULL) return XXH_ERROR; -#endif - - /* Fill secretBuffer with a copy of customSeed - repeat as needed */ - { size_t pos = 0; - while (pos < secretSize) { - size_t const toCopy = XXH_MIN((secretSize - pos), customSeedSize); - memcpy((char*)secretBuffer + pos, customSeed, toCopy); - pos += toCopy; - } } - - { size_t const nbSeg16 = secretSize / 16; - size_t n; - XXH128_canonical_t scrambler; - XXH128_canonicalFromHash(&scrambler, XXH128(customSeed, customSeedSize, 0)); - for (n=0; n -#include -#include - -#if defined(__GNUC__) && __GNUC__ >= 4 -# define ZSTD_memcpy(d,s,l) __builtin_memcpy((d),(s),(l)) -# define ZSTD_memmove(d,s,l) __builtin_memmove((d),(s),(l)) -# define ZSTD_memset(p,v,l) __builtin_memset((p),(v),(l)) -#else -# define ZSTD_memcpy(d,s,l) memcpy((d),(s),(l)) -# define ZSTD_memmove(d,s,l) memmove((d),(s),(l)) -# define ZSTD_memset(p,v,l) memset((p),(v),(l)) -#endif - -#endif /* ZSTD_DEPS_COMMON */ - -/* Need: - * ZSTD_malloc() - * ZSTD_free() - * ZSTD_calloc() - */ -#ifdef ZSTD_DEPS_NEED_MALLOC -#ifndef ZSTD_DEPS_MALLOC -#define ZSTD_DEPS_MALLOC - -#include - -#define ZSTD_malloc(s) malloc(s) -#define ZSTD_calloc(n,s) calloc((n), (s)) -#define ZSTD_free(p) free((p)) - -#endif /* ZSTD_DEPS_MALLOC */ -#endif /* ZSTD_DEPS_NEED_MALLOC */ - -/* - * Provides 64-bit math support. - * Need: - * U64 ZSTD_div64(U64 dividend, U32 divisor) - */ -#ifdef ZSTD_DEPS_NEED_MATH64 -#ifndef ZSTD_DEPS_MATH64 -#define ZSTD_DEPS_MATH64 - -#define ZSTD_div64(dividend, divisor) ((dividend) / (divisor)) - -#endif /* ZSTD_DEPS_MATH64 */ -#endif /* ZSTD_DEPS_NEED_MATH64 */ - -/* Need: - * assert() - */ -#ifdef ZSTD_DEPS_NEED_ASSERT -#ifndef ZSTD_DEPS_ASSERT -#define ZSTD_DEPS_ASSERT - -#include - -#endif /* ZSTD_DEPS_ASSERT */ -#endif /* ZSTD_DEPS_NEED_ASSERT */ - -/* Need: - * ZSTD_DEBUG_PRINT() - */ -#ifdef ZSTD_DEPS_NEED_IO -#ifndef ZSTD_DEPS_IO -#define ZSTD_DEPS_IO - -#include -#define ZSTD_DEBUG_PRINT(...) fprintf(stderr, __VA_ARGS__) - -#endif /* ZSTD_DEPS_IO */ -#endif /* ZSTD_DEPS_NEED_IO */ - -/* Only requested when is known to be present. - * Need: - * intptr_t - */ -#ifdef ZSTD_DEPS_NEED_STDINT -#ifndef ZSTD_DEPS_STDINT -#define ZSTD_DEPS_STDINT - -#include - -#endif /* ZSTD_DEPS_STDINT */ -#endif /* ZSTD_DEPS_NEED_STDINT */ diff --git a/dep/zstd/lib/common/zstd_internal.h b/dep/zstd/lib/common/zstd_internal.h deleted file mode 100644 index e4d36ce09..000000000 --- a/dep/zstd/lib/common/zstd_internal.h +++ /dev/null @@ -1,493 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_CCOMMON_H_MODULE -#define ZSTD_CCOMMON_H_MODULE - -/* this module contains definitions which must be identical - * across compression, decompression and dictBuilder. - * It also contains a few functions useful to at least 2 of them - * and which benefit from being inlined */ - -/*-************************************* -* Dependencies -***************************************/ -#include "compiler.h" -#include "cpu.h" -#include "mem.h" -#include "debug.h" /* assert, DEBUGLOG, RAWLOG, g_debuglevel */ -#include "error_private.h" -#define ZSTD_STATIC_LINKING_ONLY -#include "../zstd.h" -#define FSE_STATIC_LINKING_ONLY -#include "fse.h" -#define HUF_STATIC_LINKING_ONLY -#include "huf.h" -#ifndef XXH_STATIC_LINKING_ONLY -# define XXH_STATIC_LINKING_ONLY /* XXH64_state_t */ -#endif -#include "xxhash.h" /* XXH_reset, update, digest */ -#ifndef ZSTD_NO_TRACE -# include "zstd_trace.h" -#else -# define ZSTD_TRACE 0 -#endif - -#if defined (__cplusplus) -extern "C" { -#endif - -/* ---- static assert (debug) --- */ -#define ZSTD_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c) -#define ZSTD_isError ERR_isError /* for inlining */ -#define FSE_isError ERR_isError -#define HUF_isError ERR_isError - - -/*-************************************* -* shared macros -***************************************/ -#undef MIN -#undef MAX -#define MIN(a,b) ((a)<(b) ? (a) : (b)) -#define MAX(a,b) ((a)>(b) ? (a) : (b)) -#define BOUNDED(min,val,max) (MAX(min,MIN(val,max))) - - -/*-************************************* -* Common constants -***************************************/ -#define ZSTD_OPT_NUM (1<<12) - -#define ZSTD_REP_NUM 3 /* number of repcodes */ -static UNUSED_ATTR const U32 repStartValue[ZSTD_REP_NUM] = { 1, 4, 8 }; - -#define KB *(1 <<10) -#define MB *(1 <<20) -#define GB *(1U<<30) - -#define BIT7 128 -#define BIT6 64 -#define BIT5 32 -#define BIT4 16 -#define BIT1 2 -#define BIT0 1 - -#define ZSTD_WINDOWLOG_ABSOLUTEMIN 10 -static UNUSED_ATTR const size_t ZSTD_fcs_fieldSize[4] = { 0, 2, 4, 8 }; -static UNUSED_ATTR const size_t ZSTD_did_fieldSize[4] = { 0, 1, 2, 4 }; - -#define ZSTD_FRAMEIDSIZE 4 /* magic number size */ - -#define ZSTD_BLOCKHEADERSIZE 3 /* C standard doesn't allow `static const` variable to be init using another `static const` variable */ -static UNUSED_ATTR const size_t ZSTD_blockHeaderSize = ZSTD_BLOCKHEADERSIZE; -typedef enum { bt_raw, bt_rle, bt_compressed, bt_reserved } blockType_e; - -#define ZSTD_FRAMECHECKSUMSIZE 4 - -#define MIN_SEQUENCES_SIZE 1 /* nbSeq==0 */ -#define MIN_CBLOCK_SIZE (1 /*litCSize*/ + 1 /* RLE or RAW */ + MIN_SEQUENCES_SIZE /* nbSeq==0 */) /* for a non-null block */ - -#define HufLog 12 -typedef enum { set_basic, set_rle, set_compressed, set_repeat } symbolEncodingType_e; - -#define LONGNBSEQ 0x7F00 - -#define MINMATCH 3 - -#define Litbits 8 -#define MaxLit ((1<= WILDCOPY_VECLEN || diff <= -WILDCOPY_VECLEN); - /* Separate out the first COPY16() call because the copy length is - * almost certain to be short, so the branches have different - * probabilities. Since it is almost certain to be short, only do - * one COPY16() in the first call. Then, do two calls per loop since - * at that point it is more likely to have a high trip count. - */ -#ifdef __aarch64__ - do { - COPY16(op, ip); - } - while (op < oend); -#else - ZSTD_copy16(op, ip); - if (16 >= length) return; - op += 16; - ip += 16; - do { - COPY16(op, ip); - COPY16(op, ip); - } - while (op < oend); -#endif - } -} - -MEM_STATIC size_t ZSTD_limitCopy(void* dst, size_t dstCapacity, const void* src, size_t srcSize) -{ - size_t const length = MIN(dstCapacity, srcSize); - if (length > 0) { - ZSTD_memcpy(dst, src, length); - } - return length; -} - -/* define "workspace is too large" as this number of times larger than needed */ -#define ZSTD_WORKSPACETOOLARGE_FACTOR 3 - -/* when workspace is continuously too large - * during at least this number of times, - * context's memory usage is considered wasteful, - * because it's sized to handle a worst case scenario which rarely happens. - * In which case, resize it down to free some memory */ -#define ZSTD_WORKSPACETOOLARGE_MAXDURATION 128 - -/* Controls whether the input/output buffer is buffered or stable. */ -typedef enum { - ZSTD_bm_buffered = 0, /* Buffer the input/output */ - ZSTD_bm_stable = 1 /* ZSTD_inBuffer/ZSTD_outBuffer is stable */ -} ZSTD_bufferMode_e; - - -/*-******************************************* -* Private declarations -*********************************************/ -typedef struct seqDef_s { - U32 offBase; /* offBase == Offset + ZSTD_REP_NUM, or repcode 1,2,3 */ - U16 litLength; - U16 mlBase; /* mlBase == matchLength - MINMATCH */ -} seqDef; - -/* Controls whether seqStore has a single "long" litLength or matchLength. See seqStore_t. */ -typedef enum { - ZSTD_llt_none = 0, /* no longLengthType */ - ZSTD_llt_literalLength = 1, /* represents a long literal */ - ZSTD_llt_matchLength = 2 /* represents a long match */ -} ZSTD_longLengthType_e; - -typedef struct { - seqDef* sequencesStart; - seqDef* sequences; /* ptr to end of sequences */ - BYTE* litStart; - BYTE* lit; /* ptr to end of literals */ - BYTE* llCode; - BYTE* mlCode; - BYTE* ofCode; - size_t maxNbSeq; - size_t maxNbLit; - - /* longLengthPos and longLengthType to allow us to represent either a single litLength or matchLength - * in the seqStore that has a value larger than U16 (if it exists). To do so, we increment - * the existing value of the litLength or matchLength by 0x10000. - */ - ZSTD_longLengthType_e longLengthType; - U32 longLengthPos; /* Index of the sequence to apply long length modification to */ -} seqStore_t; - -typedef struct { - U32 litLength; - U32 matchLength; -} ZSTD_sequenceLength; - -/** - * Returns the ZSTD_sequenceLength for the given sequences. It handles the decoding of long sequences - * indicated by longLengthPos and longLengthType, and adds MINMATCH back to matchLength. - */ -MEM_STATIC ZSTD_sequenceLength ZSTD_getSequenceLength(seqStore_t const* seqStore, seqDef const* seq) -{ - ZSTD_sequenceLength seqLen; - seqLen.litLength = seq->litLength; - seqLen.matchLength = seq->mlBase + MINMATCH; - if (seqStore->longLengthPos == (U32)(seq - seqStore->sequencesStart)) { - if (seqStore->longLengthType == ZSTD_llt_literalLength) { - seqLen.litLength += 0xFFFF; - } - if (seqStore->longLengthType == ZSTD_llt_matchLength) { - seqLen.matchLength += 0xFFFF; - } - } - return seqLen; -} - -/** - * Contains the compressed frame size and an upper-bound for the decompressed frame size. - * Note: before using `compressedSize`, check for errors using ZSTD_isError(). - * similarly, before using `decompressedBound`, check for errors using: - * `decompressedBound != ZSTD_CONTENTSIZE_ERROR` - */ -typedef struct { - size_t compressedSize; - unsigned long long decompressedBound; -} ZSTD_frameSizeInfo; /* decompress & legacy */ - -const seqStore_t* ZSTD_getSeqStore(const ZSTD_CCtx* ctx); /* compress & dictBuilder */ -void ZSTD_seqToCodes(const seqStore_t* seqStorePtr); /* compress, dictBuilder, decodeCorpus (shouldn't get its definition from here) */ - -/* custom memory allocation functions */ -void* ZSTD_customMalloc(size_t size, ZSTD_customMem customMem); -void* ZSTD_customCalloc(size_t size, ZSTD_customMem customMem); -void ZSTD_customFree(void* ptr, ZSTD_customMem customMem); - - -MEM_STATIC U32 ZSTD_highbit32(U32 val) /* compress, dictBuilder, decodeCorpus */ -{ - assert(val != 0); - { -# if defined(_MSC_VER) /* Visual */ -# if STATIC_BMI2 == 1 - return _lzcnt_u32(val)^31; -# else - if (val != 0) { - unsigned long r; - _BitScanReverse(&r, val); - return (unsigned)r; - } else { - /* Should not reach this code path */ - __assume(0); - } -# endif -# elif defined(__GNUC__) && (__GNUC__ >= 3) /* GCC Intrinsic */ - return __builtin_clz (val) ^ 31; -# elif defined(__ICCARM__) /* IAR Intrinsic */ - return 31 - __CLZ(val); -# else /* Software version */ - static const U32 DeBruijnClz[32] = { 0, 9, 1, 10, 13, 21, 2, 29, 11, 14, 16, 18, 22, 25, 3, 30, 8, 12, 20, 28, 15, 17, 24, 7, 19, 27, 23, 6, 26, 5, 4, 31 }; - U32 v = val; - v |= v >> 1; - v |= v >> 2; - v |= v >> 4; - v |= v >> 8; - v |= v >> 16; - return DeBruijnClz[(v * 0x07C4ACDDU) >> 27]; -# endif - } -} - -/** - * Counts the number of trailing zeros of a `size_t`. - * Most compilers should support CTZ as a builtin. A backup - * implementation is provided if the builtin isn't supported, but - * it may not be terribly efficient. - */ -MEM_STATIC unsigned ZSTD_countTrailingZeros(size_t val) -{ - if (MEM_64bits()) { -# if defined(_MSC_VER) && defined(_WIN64) -# if STATIC_BMI2 - return _tzcnt_u64(val); -# else - if (val != 0) { - unsigned long r; - _BitScanForward64(&r, (U64)val); - return (unsigned)r; - } else { - /* Should not reach this code path */ - __assume(0); - } -# endif -# elif defined(__GNUC__) && (__GNUC__ >= 4) - return __builtin_ctzll((U64)val); -# else - static const int DeBruijnBytePos[64] = { 0, 1, 2, 7, 3, 13, 8, 19, - 4, 25, 14, 28, 9, 34, 20, 56, - 5, 17, 26, 54, 15, 41, 29, 43, - 10, 31, 38, 35, 21, 45, 49, 57, - 63, 6, 12, 18, 24, 27, 33, 55, - 16, 53, 40, 42, 30, 37, 44, 48, - 62, 11, 23, 32, 52, 39, 36, 47, - 61, 22, 51, 46, 60, 50, 59, 58 }; - return DeBruijnBytePos[((U64)((val & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58]; -# endif - } else { /* 32 bits */ -# if defined(_MSC_VER) - if (val != 0) { - unsigned long r; - _BitScanForward(&r, (U32)val); - return (unsigned)r; - } else { - /* Should not reach this code path */ - __assume(0); - } -# elif defined(__GNUC__) && (__GNUC__ >= 3) - return __builtin_ctz((U32)val); -# else - static const int DeBruijnBytePos[32] = { 0, 1, 28, 2, 29, 14, 24, 3, - 30, 22, 20, 15, 25, 17, 4, 8, - 31, 27, 13, 23, 21, 19, 16, 7, - 26, 12, 18, 6, 11, 5, 10, 9 }; - return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27]; -# endif - } -} - - -/* ZSTD_invalidateRepCodes() : - * ensures next compression will not use repcodes from previous block. - * Note : only works with regular variant; - * do not use with extDict variant ! */ -void ZSTD_invalidateRepCodes(ZSTD_CCtx* cctx); /* zstdmt, adaptive_compression (shouldn't get this definition from here) */ - - -typedef struct { - blockType_e blockType; - U32 lastBlock; - U32 origSize; -} blockProperties_t; /* declared here for decompress and fullbench */ - -/*! ZSTD_getcBlockSize() : - * Provides the size of compressed block from block header `src` */ -/* Used by: decompress, fullbench (does not get its definition from here) */ -size_t ZSTD_getcBlockSize(const void* src, size_t srcSize, - blockProperties_t* bpPtr); - -/*! ZSTD_decodeSeqHeaders() : - * decode sequence header from src */ -/* Used by: decompress, fullbench (does not get its definition from here) */ -size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr, - const void* src, size_t srcSize); - -/** - * @returns true iff the CPU supports dynamic BMI2 dispatch. - */ -MEM_STATIC int ZSTD_cpuSupportsBmi2(void) -{ - ZSTD_cpuid_t cpuid = ZSTD_cpuid(); - return ZSTD_cpuid_bmi1(cpuid) && ZSTD_cpuid_bmi2(cpuid); -} - -#if defined (__cplusplus) -} -#endif - -#endif /* ZSTD_CCOMMON_H_MODULE */ diff --git a/dep/zstd/lib/common/zstd_trace.h b/dep/zstd/lib/common/zstd_trace.h deleted file mode 100644 index f9121f7d8..000000000 --- a/dep/zstd/lib/common/zstd_trace.h +++ /dev/null @@ -1,163 +0,0 @@ -/* - * Copyright (c) Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_TRACE_H -#define ZSTD_TRACE_H - -#if defined (__cplusplus) -extern "C" { -#endif - -#include - -/* weak symbol support - * For now, enable conservatively: - * - Only GNUC - * - Only ELF - * - Only x86-64 and i386 - * Also, explicitly disable on platforms known not to work so they aren't - * forgotten in the future. - */ -#if !defined(ZSTD_HAVE_WEAK_SYMBOLS) && \ - defined(__GNUC__) && defined(__ELF__) && \ - (defined(__x86_64__) || defined(_M_X64) || defined(__i386__) || defined(_M_IX86)) && \ - !defined(__APPLE__) && !defined(_WIN32) && !defined(__MINGW32__) && \ - !defined(__CYGWIN__) && !defined(_AIX) -# define ZSTD_HAVE_WEAK_SYMBOLS 1 -#else -# define ZSTD_HAVE_WEAK_SYMBOLS 0 -#endif -#if ZSTD_HAVE_WEAK_SYMBOLS -# define ZSTD_WEAK_ATTR __attribute__((__weak__)) -#else -# define ZSTD_WEAK_ATTR -#endif - -/* Only enable tracing when weak symbols are available. */ -#ifndef ZSTD_TRACE -# define ZSTD_TRACE ZSTD_HAVE_WEAK_SYMBOLS -#endif - -#if ZSTD_TRACE - -struct ZSTD_CCtx_s; -struct ZSTD_DCtx_s; -struct ZSTD_CCtx_params_s; - -typedef struct { - /** - * ZSTD_VERSION_NUMBER - * - * This is guaranteed to be the first member of ZSTD_trace. - * Otherwise, this struct is not stable between versions. If - * the version number does not match your expectation, you - * should not interpret the rest of the struct. - */ - unsigned version; - /** - * Non-zero if streaming (de)compression is used. - */ - unsigned streaming; - /** - * The dictionary ID. - */ - unsigned dictionaryID; - /** - * Is the dictionary cold? - * Only set on decompression. - */ - unsigned dictionaryIsCold; - /** - * The dictionary size or zero if no dictionary. - */ - size_t dictionarySize; - /** - * The uncompressed size of the data. - */ - size_t uncompressedSize; - /** - * The compressed size of the data. - */ - size_t compressedSize; - /** - * The fully resolved CCtx parameters (NULL on decompression). - */ - struct ZSTD_CCtx_params_s const* params; - /** - * The ZSTD_CCtx pointer (NULL on decompression). - */ - struct ZSTD_CCtx_s const* cctx; - /** - * The ZSTD_DCtx pointer (NULL on compression). - */ - struct ZSTD_DCtx_s const* dctx; -} ZSTD_Trace; - -/** - * A tracing context. It must be 0 when tracing is disabled. - * Otherwise, any non-zero value returned by a tracing begin() - * function is presented to any subsequent calls to end(). - * - * Any non-zero value is treated as tracing is enabled and not - * interpreted by the library. - * - * Two possible uses are: - * * A timestamp for when the begin() function was called. - * * A unique key identifying the (de)compression, like the - * address of the [dc]ctx pointer if you need to track - * more information than just a timestamp. - */ -typedef unsigned long long ZSTD_TraceCtx; - -/** - * Trace the beginning of a compression call. - * @param cctx The dctx pointer for the compression. - * It can be used as a key to map begin() to end(). - * @returns Non-zero if tracing is enabled. The return value is - * passed to ZSTD_trace_compress_end(). - */ -ZSTD_WEAK_ATTR ZSTD_TraceCtx ZSTD_trace_compress_begin( - struct ZSTD_CCtx_s const* cctx); - -/** - * Trace the end of a compression call. - * @param ctx The return value of ZSTD_trace_compress_begin(). - * @param trace The zstd tracing info. - */ -ZSTD_WEAK_ATTR void ZSTD_trace_compress_end( - ZSTD_TraceCtx ctx, - ZSTD_Trace const* trace); - -/** - * Trace the beginning of a decompression call. - * @param dctx The dctx pointer for the decompression. - * It can be used as a key to map begin() to end(). - * @returns Non-zero if tracing is enabled. The return value is - * passed to ZSTD_trace_compress_end(). - */ -ZSTD_WEAK_ATTR ZSTD_TraceCtx ZSTD_trace_decompress_begin( - struct ZSTD_DCtx_s const* dctx); - -/** - * Trace the end of a decompression call. - * @param ctx The return value of ZSTD_trace_decompress_begin(). - * @param trace The zstd tracing info. - */ -ZSTD_WEAK_ATTR void ZSTD_trace_decompress_end( - ZSTD_TraceCtx ctx, - ZSTD_Trace const* trace); - -#endif /* ZSTD_TRACE */ - -#if defined (__cplusplus) -} -#endif - -#endif /* ZSTD_TRACE_H */ diff --git a/dep/zstd/lib/compress/clevels.h b/dep/zstd/lib/compress/clevels.h deleted file mode 100644 index 7ed2e0049..000000000 --- a/dep/zstd/lib/compress/clevels.h +++ /dev/null @@ -1,134 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_CLEVELS_H -#define ZSTD_CLEVELS_H - -#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_compressionParameters */ -#include "../zstd.h" - -/*-===== Pre-defined compression levels =====-*/ - -#define ZSTD_MAX_CLEVEL 22 - -#ifdef __GNUC__ -__attribute__((__unused__)) -#endif - -static const ZSTD_compressionParameters ZSTD_defaultCParameters[4][ZSTD_MAX_CLEVEL+1] = { -{ /* "default" - for any srcSize > 256 KB */ - /* W, C, H, S, L, TL, strat */ - { 19, 12, 13, 1, 6, 1, ZSTD_fast }, /* base for negative levels */ - { 19, 13, 14, 1, 7, 0, ZSTD_fast }, /* level 1 */ - { 20, 15, 16, 1, 6, 0, ZSTD_fast }, /* level 2 */ - { 21, 16, 17, 1, 5, 0, ZSTD_dfast }, /* level 3 */ - { 21, 18, 18, 1, 5, 0, ZSTD_dfast }, /* level 4 */ - { 21, 18, 19, 3, 5, 2, ZSTD_greedy }, /* level 5 */ - { 21, 18, 19, 3, 5, 4, ZSTD_lazy }, /* level 6 */ - { 21, 19, 20, 4, 5, 8, ZSTD_lazy }, /* level 7 */ - { 21, 19, 20, 4, 5, 16, ZSTD_lazy2 }, /* level 8 */ - { 22, 20, 21, 4, 5, 16, ZSTD_lazy2 }, /* level 9 */ - { 22, 21, 22, 5, 5, 16, ZSTD_lazy2 }, /* level 10 */ - { 22, 21, 22, 6, 5, 16, ZSTD_lazy2 }, /* level 11 */ - { 22, 22, 23, 6, 5, 32, ZSTD_lazy2 }, /* level 12 */ - { 22, 22, 22, 4, 5, 32, ZSTD_btlazy2 }, /* level 13 */ - { 22, 22, 23, 5, 5, 32, ZSTD_btlazy2 }, /* level 14 */ - { 22, 23, 23, 6, 5, 32, ZSTD_btlazy2 }, /* level 15 */ - { 22, 22, 22, 5, 5, 48, ZSTD_btopt }, /* level 16 */ - { 23, 23, 22, 5, 4, 64, ZSTD_btopt }, /* level 17 */ - { 23, 23, 22, 6, 3, 64, ZSTD_btultra }, /* level 18 */ - { 23, 24, 22, 7, 3,256, ZSTD_btultra2}, /* level 19 */ - { 25, 25, 23, 7, 3,256, ZSTD_btultra2}, /* level 20 */ - { 26, 26, 24, 7, 3,512, ZSTD_btultra2}, /* level 21 */ - { 27, 27, 25, 9, 3,999, ZSTD_btultra2}, /* level 22 */ -}, -{ /* for srcSize <= 256 KB */ - /* W, C, H, S, L, T, strat */ - { 18, 12, 13, 1, 5, 1, ZSTD_fast }, /* base for negative levels */ - { 18, 13, 14, 1, 6, 0, ZSTD_fast }, /* level 1 */ - { 18, 14, 14, 1, 5, 0, ZSTD_dfast }, /* level 2 */ - { 18, 16, 16, 1, 4, 0, ZSTD_dfast }, /* level 3 */ - { 18, 16, 17, 3, 5, 2, ZSTD_greedy }, /* level 4.*/ - { 18, 17, 18, 5, 5, 2, ZSTD_greedy }, /* level 5.*/ - { 18, 18, 19, 3, 5, 4, ZSTD_lazy }, /* level 6.*/ - { 18, 18, 19, 4, 4, 4, ZSTD_lazy }, /* level 7 */ - { 18, 18, 19, 4, 4, 8, ZSTD_lazy2 }, /* level 8 */ - { 18, 18, 19, 5, 4, 8, ZSTD_lazy2 }, /* level 9 */ - { 18, 18, 19, 6, 4, 8, ZSTD_lazy2 }, /* level 10 */ - { 18, 18, 19, 5, 4, 12, ZSTD_btlazy2 }, /* level 11.*/ - { 18, 19, 19, 7, 4, 12, ZSTD_btlazy2 }, /* level 12.*/ - { 18, 18, 19, 4, 4, 16, ZSTD_btopt }, /* level 13 */ - { 18, 18, 19, 4, 3, 32, ZSTD_btopt }, /* level 14.*/ - { 18, 18, 19, 6, 3,128, ZSTD_btopt }, /* level 15.*/ - { 18, 19, 19, 6, 3,128, ZSTD_btultra }, /* level 16.*/ - { 18, 19, 19, 8, 3,256, ZSTD_btultra }, /* level 17.*/ - { 18, 19, 19, 6, 3,128, ZSTD_btultra2}, /* level 18.*/ - { 18, 19, 19, 8, 3,256, ZSTD_btultra2}, /* level 19.*/ - { 18, 19, 19, 10, 3,512, ZSTD_btultra2}, /* level 20.*/ - { 18, 19, 19, 12, 3,512, ZSTD_btultra2}, /* level 21.*/ - { 18, 19, 19, 13, 3,999, ZSTD_btultra2}, /* level 22.*/ -}, -{ /* for srcSize <= 128 KB */ - /* W, C, H, S, L, T, strat */ - { 17, 12, 12, 1, 5, 1, ZSTD_fast }, /* base for negative levels */ - { 17, 12, 13, 1, 6, 0, ZSTD_fast }, /* level 1 */ - { 17, 13, 15, 1, 5, 0, ZSTD_fast }, /* level 2 */ - { 17, 15, 16, 2, 5, 0, ZSTD_dfast }, /* level 3 */ - { 17, 17, 17, 2, 4, 0, ZSTD_dfast }, /* level 4 */ - { 17, 16, 17, 3, 4, 2, ZSTD_greedy }, /* level 5 */ - { 17, 16, 17, 3, 4, 4, ZSTD_lazy }, /* level 6 */ - { 17, 16, 17, 3, 4, 8, ZSTD_lazy2 }, /* level 7 */ - { 17, 16, 17, 4, 4, 8, ZSTD_lazy2 }, /* level 8 */ - { 17, 16, 17, 5, 4, 8, ZSTD_lazy2 }, /* level 9 */ - { 17, 16, 17, 6, 4, 8, ZSTD_lazy2 }, /* level 10 */ - { 17, 17, 17, 5, 4, 8, ZSTD_btlazy2 }, /* level 11 */ - { 17, 18, 17, 7, 4, 12, ZSTD_btlazy2 }, /* level 12 */ - { 17, 18, 17, 3, 4, 12, ZSTD_btopt }, /* level 13.*/ - { 17, 18, 17, 4, 3, 32, ZSTD_btopt }, /* level 14.*/ - { 17, 18, 17, 6, 3,256, ZSTD_btopt }, /* level 15.*/ - { 17, 18, 17, 6, 3,128, ZSTD_btultra }, /* level 16.*/ - { 17, 18, 17, 8, 3,256, ZSTD_btultra }, /* level 17.*/ - { 17, 18, 17, 10, 3,512, ZSTD_btultra }, /* level 18.*/ - { 17, 18, 17, 5, 3,256, ZSTD_btultra2}, /* level 19.*/ - { 17, 18, 17, 7, 3,512, ZSTD_btultra2}, /* level 20.*/ - { 17, 18, 17, 9, 3,512, ZSTD_btultra2}, /* level 21.*/ - { 17, 18, 17, 11, 3,999, ZSTD_btultra2}, /* level 22.*/ -}, -{ /* for srcSize <= 16 KB */ - /* W, C, H, S, L, T, strat */ - { 14, 12, 13, 1, 5, 1, ZSTD_fast }, /* base for negative levels */ - { 14, 14, 15, 1, 5, 0, ZSTD_fast }, /* level 1 */ - { 14, 14, 15, 1, 4, 0, ZSTD_fast }, /* level 2 */ - { 14, 14, 15, 2, 4, 0, ZSTD_dfast }, /* level 3 */ - { 14, 14, 14, 4, 4, 2, ZSTD_greedy }, /* level 4 */ - { 14, 14, 14, 3, 4, 4, ZSTD_lazy }, /* level 5.*/ - { 14, 14, 14, 4, 4, 8, ZSTD_lazy2 }, /* level 6 */ - { 14, 14, 14, 6, 4, 8, ZSTD_lazy2 }, /* level 7 */ - { 14, 14, 14, 8, 4, 8, ZSTD_lazy2 }, /* level 8.*/ - { 14, 15, 14, 5, 4, 8, ZSTD_btlazy2 }, /* level 9.*/ - { 14, 15, 14, 9, 4, 8, ZSTD_btlazy2 }, /* level 10.*/ - { 14, 15, 14, 3, 4, 12, ZSTD_btopt }, /* level 11.*/ - { 14, 15, 14, 4, 3, 24, ZSTD_btopt }, /* level 12.*/ - { 14, 15, 14, 5, 3, 32, ZSTD_btultra }, /* level 13.*/ - { 14, 15, 15, 6, 3, 64, ZSTD_btultra }, /* level 14.*/ - { 14, 15, 15, 7, 3,256, ZSTD_btultra }, /* level 15.*/ - { 14, 15, 15, 5, 3, 48, ZSTD_btultra2}, /* level 16.*/ - { 14, 15, 15, 6, 3,128, ZSTD_btultra2}, /* level 17.*/ - { 14, 15, 15, 7, 3,256, ZSTD_btultra2}, /* level 18.*/ - { 14, 15, 15, 8, 3,256, ZSTD_btultra2}, /* level 19.*/ - { 14, 15, 15, 8, 3,512, ZSTD_btultra2}, /* level 20.*/ - { 14, 15, 15, 9, 3,512, ZSTD_btultra2}, /* level 21.*/ - { 14, 15, 15, 10, 3,999, ZSTD_btultra2}, /* level 22.*/ -}, -}; - - - -#endif /* ZSTD_CLEVELS_H */ diff --git a/dep/zstd/lib/compress/fse_compress.c b/dep/zstd/lib/compress/fse_compress.c deleted file mode 100644 index 5547b4ac0..000000000 --- a/dep/zstd/lib/compress/fse_compress.c +++ /dev/null @@ -1,741 +0,0 @@ -/* ****************************************************************** - * FSE : Finite State Entropy encoder - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy - * - Public forum : https://groups.google.com/forum/#!forum/lz4c - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -****************************************************************** */ - -/* ************************************************************** -* Includes -****************************************************************/ -#include "../common/compiler.h" -#include "../common/mem.h" /* U32, U16, etc. */ -#include "../common/debug.h" /* assert, DEBUGLOG */ -#include "hist.h" /* HIST_count_wksp */ -#include "../common/bitstream.h" -#define FSE_STATIC_LINKING_ONLY -#include "../common/fse.h" -#include "../common/error_private.h" -#define ZSTD_DEPS_NEED_MALLOC -#define ZSTD_DEPS_NEED_MATH64 -#include "../common/zstd_deps.h" /* ZSTD_malloc, ZSTD_free, ZSTD_memcpy, ZSTD_memset */ - - -/* ************************************************************** -* Error Management -****************************************************************/ -#define FSE_isError ERR_isError - - -/* ************************************************************** -* Templates -****************************************************************/ -/* - designed to be included - for type-specific functions (template emulation in C) - Objective is to write these functions only once, for improved maintenance -*/ - -/* safety checks */ -#ifndef FSE_FUNCTION_EXTENSION -# error "FSE_FUNCTION_EXTENSION must be defined" -#endif -#ifndef FSE_FUNCTION_TYPE -# error "FSE_FUNCTION_TYPE must be defined" -#endif - -/* Function names */ -#define FSE_CAT(X,Y) X##Y -#define FSE_FUNCTION_NAME(X,Y) FSE_CAT(X,Y) -#define FSE_TYPE_NAME(X,Y) FSE_CAT(X,Y) - - -/* Function templates */ - -/* FSE_buildCTable_wksp() : - * Same as FSE_buildCTable(), but using an externally allocated scratch buffer (`workSpace`). - * wkspSize should be sized to handle worst case situation, which is `1<>1 : 1) ; - FSE_symbolCompressionTransform* const symbolTT = (FSE_symbolCompressionTransform*) (FSCT); - U32 const step = FSE_TABLESTEP(tableSize); - U32 const maxSV1 = maxSymbolValue+1; - - U16* cumul = (U16*)workSpace; /* size = maxSV1 */ - FSE_FUNCTION_TYPE* const tableSymbol = (FSE_FUNCTION_TYPE*)(cumul + (maxSV1+1)); /* size = tableSize */ - - U32 highThreshold = tableSize-1; - - assert(((size_t)workSpace & 1) == 0); /* Must be 2 bytes-aligned */ - if (FSE_BUILD_CTABLE_WORKSPACE_SIZE(maxSymbolValue, tableLog) > wkspSize) return ERROR(tableLog_tooLarge); - /* CTable header */ - tableU16[-2] = (U16) tableLog; - tableU16[-1] = (U16) maxSymbolValue; - assert(tableLog < 16); /* required for threshold strategy to work */ - - /* For explanations on how to distribute symbol values over the table : - * http://fastcompression.blogspot.fr/2014/02/fse-distributing-symbol-values.html */ - - #ifdef __clang_analyzer__ - ZSTD_memset(tableSymbol, 0, sizeof(*tableSymbol) * tableSize); /* useless initialization, just to keep scan-build happy */ - #endif - - /* symbol start positions */ - { U32 u; - cumul[0] = 0; - for (u=1; u <= maxSV1; u++) { - if (normalizedCounter[u-1]==-1) { /* Low proba symbol */ - cumul[u] = cumul[u-1] + 1; - tableSymbol[highThreshold--] = (FSE_FUNCTION_TYPE)(u-1); - } else { - assert(normalizedCounter[u-1] >= 0); - cumul[u] = cumul[u-1] + (U16)normalizedCounter[u-1]; - assert(cumul[u] >= cumul[u-1]); /* no overflow */ - } } - cumul[maxSV1] = (U16)(tableSize+1); - } - - /* Spread symbols */ - if (highThreshold == tableSize - 1) { - /* Case for no low prob count symbols. Lay down 8 bytes at a time - * to reduce branch misses since we are operating on a small block - */ - BYTE* const spread = tableSymbol + tableSize; /* size = tableSize + 8 (may write beyond tableSize) */ - { U64 const add = 0x0101010101010101ull; - size_t pos = 0; - U64 sv = 0; - U32 s; - for (s=0; s=0); - pos += (size_t)n; - } - } - /* Spread symbols across the table. Lack of lowprob symbols means that - * we don't need variable sized inner loop, so we can unroll the loop and - * reduce branch misses. - */ - { size_t position = 0; - size_t s; - size_t const unroll = 2; /* Experimentally determined optimal unroll */ - assert(tableSize % unroll == 0); /* FSE_MIN_TABLELOG is 5 */ - for (s = 0; s < (size_t)tableSize; s += unroll) { - size_t u; - for (u = 0; u < unroll; ++u) { - size_t const uPosition = (position + (u * step)) & tableMask; - tableSymbol[uPosition] = spread[s + u]; - } - position = (position + (unroll * step)) & tableMask; - } - assert(position == 0); /* Must have initialized all positions */ - } - } else { - U32 position = 0; - U32 symbol; - for (symbol=0; symbol highThreshold) - position = (position + step) & tableMask; /* Low proba area */ - } } - assert(position==0); /* Must have initialized all positions */ - } - - /* Build table */ - { U32 u; for (u=0; u 1); - { U32 const maxBitsOut = tableLog - BIT_highbit32 ((U32)normalizedCounter[s]-1); - U32 const minStatePlus = (U32)normalizedCounter[s] << maxBitsOut; - symbolTT[s].deltaNbBits = (maxBitsOut << 16) - minStatePlus; - symbolTT[s].deltaFindState = (int)(total - (unsigned)normalizedCounter[s]); - total += (unsigned)normalizedCounter[s]; - } } } } - -#if 0 /* debug : symbol costs */ - DEBUGLOG(5, "\n --- table statistics : "); - { U32 symbol; - for (symbol=0; symbol<=maxSymbolValue; symbol++) { - DEBUGLOG(5, "%3u: w=%3i, maxBits=%u, fracBits=%.2f", - symbol, normalizedCounter[symbol], - FSE_getMaxNbBits(symbolTT, symbol), - (double)FSE_bitCost(symbolTT, tableLog, symbol, 8) / 256); - } } -#endif - - return 0; -} - - - -#ifndef FSE_COMMONDEFS_ONLY - -/*-************************************************************** -* FSE NCount encoding -****************************************************************/ -size_t FSE_NCountWriteBound(unsigned maxSymbolValue, unsigned tableLog) -{ - size_t const maxHeaderSize = (((maxSymbolValue+1) * tableLog - + 4 /* bitCount initialized at 4 */ - + 2 /* first two symbols may use one additional bit each */) / 8) - + 1 /* round up to whole nb bytes */ - + 2 /* additional two bytes for bitstream flush */; - return maxSymbolValue ? maxHeaderSize : FSE_NCOUNTBOUND; /* maxSymbolValue==0 ? use default */ -} - -static size_t -FSE_writeNCount_generic (void* header, size_t headerBufferSize, - const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog, - unsigned writeIsSafe) -{ - BYTE* const ostart = (BYTE*) header; - BYTE* out = ostart; - BYTE* const oend = ostart + headerBufferSize; - int nbBits; - const int tableSize = 1 << tableLog; - int remaining; - int threshold; - U32 bitStream = 0; - int bitCount = 0; - unsigned symbol = 0; - unsigned const alphabetSize = maxSymbolValue + 1; - int previousIs0 = 0; - - /* Table Size */ - bitStream += (tableLog-FSE_MIN_TABLELOG) << bitCount; - bitCount += 4; - - /* Init */ - remaining = tableSize+1; /* +1 for extra accuracy */ - threshold = tableSize; - nbBits = tableLog+1; - - while ((symbol < alphabetSize) && (remaining>1)) { /* stops at 1 */ - if (previousIs0) { - unsigned start = symbol; - while ((symbol < alphabetSize) && !normalizedCounter[symbol]) symbol++; - if (symbol == alphabetSize) break; /* incorrect distribution */ - while (symbol >= start+24) { - start+=24; - bitStream += 0xFFFFU << bitCount; - if ((!writeIsSafe) && (out > oend-2)) - return ERROR(dstSize_tooSmall); /* Buffer overflow */ - out[0] = (BYTE) bitStream; - out[1] = (BYTE)(bitStream>>8); - out+=2; - bitStream>>=16; - } - while (symbol >= start+3) { - start+=3; - bitStream += 3 << bitCount; - bitCount += 2; - } - bitStream += (symbol-start) << bitCount; - bitCount += 2; - if (bitCount>16) { - if ((!writeIsSafe) && (out > oend - 2)) - return ERROR(dstSize_tooSmall); /* Buffer overflow */ - out[0] = (BYTE)bitStream; - out[1] = (BYTE)(bitStream>>8); - out += 2; - bitStream >>= 16; - bitCount -= 16; - } } - { int count = normalizedCounter[symbol++]; - int const max = (2*threshold-1) - remaining; - remaining -= count < 0 ? -count : count; - count++; /* +1 for extra accuracy */ - if (count>=threshold) - count += max; /* [0..max[ [max..threshold[ (...) [threshold+max 2*threshold[ */ - bitStream += count << bitCount; - bitCount += nbBits; - bitCount -= (count>=1; } - } - if (bitCount>16) { - if ((!writeIsSafe) && (out > oend - 2)) - return ERROR(dstSize_tooSmall); /* Buffer overflow */ - out[0] = (BYTE)bitStream; - out[1] = (BYTE)(bitStream>>8); - out += 2; - bitStream >>= 16; - bitCount -= 16; - } } - - if (remaining != 1) - return ERROR(GENERIC); /* incorrect normalized distribution */ - assert(symbol <= alphabetSize); - - /* flush remaining bitStream */ - if ((!writeIsSafe) && (out > oend - 2)) - return ERROR(dstSize_tooSmall); /* Buffer overflow */ - out[0] = (BYTE)bitStream; - out[1] = (BYTE)(bitStream>>8); - out+= (bitCount+7) /8; - - return (out-ostart); -} - - -size_t FSE_writeNCount (void* buffer, size_t bufferSize, - const short* normalizedCounter, unsigned maxSymbolValue, unsigned tableLog) -{ - if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge); /* Unsupported */ - if (tableLog < FSE_MIN_TABLELOG) return ERROR(GENERIC); /* Unsupported */ - - if (bufferSize < FSE_NCountWriteBound(maxSymbolValue, tableLog)) - return FSE_writeNCount_generic(buffer, bufferSize, normalizedCounter, maxSymbolValue, tableLog, 0); - - return FSE_writeNCount_generic(buffer, bufferSize, normalizedCounter, maxSymbolValue, tableLog, 1 /* write in buffer is safe */); -} - - -/*-************************************************************** -* FSE Compression Code -****************************************************************/ - -FSE_CTable* FSE_createCTable (unsigned maxSymbolValue, unsigned tableLog) -{ - size_t size; - if (tableLog > FSE_TABLELOG_ABSOLUTE_MAX) tableLog = FSE_TABLELOG_ABSOLUTE_MAX; - size = FSE_CTABLE_SIZE_U32 (tableLog, maxSymbolValue) * sizeof(U32); - return (FSE_CTable*)ZSTD_malloc(size); -} - -void FSE_freeCTable (FSE_CTable* ct) { ZSTD_free(ct); } - -/* provides the minimum logSize to safely represent a distribution */ -static unsigned FSE_minTableLog(size_t srcSize, unsigned maxSymbolValue) -{ - U32 minBitsSrc = BIT_highbit32((U32)(srcSize)) + 1; - U32 minBitsSymbols = BIT_highbit32(maxSymbolValue) + 2; - U32 minBits = minBitsSrc < minBitsSymbols ? minBitsSrc : minBitsSymbols; - assert(srcSize > 1); /* Not supported, RLE should be used instead */ - return minBits; -} - -unsigned FSE_optimalTableLog_internal(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue, unsigned minus) -{ - U32 maxBitsSrc = BIT_highbit32((U32)(srcSize - 1)) - minus; - U32 tableLog = maxTableLog; - U32 minBits = FSE_minTableLog(srcSize, maxSymbolValue); - assert(srcSize > 1); /* Not supported, RLE should be used instead */ - if (tableLog==0) tableLog = FSE_DEFAULT_TABLELOG; - if (maxBitsSrc < tableLog) tableLog = maxBitsSrc; /* Accuracy can be reduced */ - if (minBits > tableLog) tableLog = minBits; /* Need a minimum to safely represent all symbol values */ - if (tableLog < FSE_MIN_TABLELOG) tableLog = FSE_MIN_TABLELOG; - if (tableLog > FSE_MAX_TABLELOG) tableLog = FSE_MAX_TABLELOG; - return tableLog; -} - -unsigned FSE_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue) -{ - return FSE_optimalTableLog_internal(maxTableLog, srcSize, maxSymbolValue, 2); -} - -/* Secondary normalization method. - To be used when primary method fails. */ - -static size_t FSE_normalizeM2(short* norm, U32 tableLog, const unsigned* count, size_t total, U32 maxSymbolValue, short lowProbCount) -{ - short const NOT_YET_ASSIGNED = -2; - U32 s; - U32 distributed = 0; - U32 ToDistribute; - - /* Init */ - U32 const lowThreshold = (U32)(total >> tableLog); - U32 lowOne = (U32)((total * 3) >> (tableLog + 1)); - - for (s=0; s<=maxSymbolValue; s++) { - if (count[s] == 0) { - norm[s]=0; - continue; - } - if (count[s] <= lowThreshold) { - norm[s] = lowProbCount; - distributed++; - total -= count[s]; - continue; - } - if (count[s] <= lowOne) { - norm[s] = 1; - distributed++; - total -= count[s]; - continue; - } - - norm[s]=NOT_YET_ASSIGNED; - } - ToDistribute = (1 << tableLog) - distributed; - - if (ToDistribute == 0) - return 0; - - if ((total / ToDistribute) > lowOne) { - /* risk of rounding to zero */ - lowOne = (U32)((total * 3) / (ToDistribute * 2)); - for (s=0; s<=maxSymbolValue; s++) { - if ((norm[s] == NOT_YET_ASSIGNED) && (count[s] <= lowOne)) { - norm[s] = 1; - distributed++; - total -= count[s]; - continue; - } } - ToDistribute = (1 << tableLog) - distributed; - } - - if (distributed == maxSymbolValue+1) { - /* all values are pretty poor; - probably incompressible data (should have already been detected); - find max, then give all remaining points to max */ - U32 maxV = 0, maxC = 0; - for (s=0; s<=maxSymbolValue; s++) - if (count[s] > maxC) { maxV=s; maxC=count[s]; } - norm[maxV] += (short)ToDistribute; - return 0; - } - - if (total == 0) { - /* all of the symbols were low enough for the lowOne or lowThreshold */ - for (s=0; ToDistribute > 0; s = (s+1)%(maxSymbolValue+1)) - if (norm[s] > 0) { ToDistribute--; norm[s]++; } - return 0; - } - - { U64 const vStepLog = 62 - tableLog; - U64 const mid = (1ULL << (vStepLog-1)) - 1; - U64 const rStep = ZSTD_div64((((U64)1<> vStepLog); - U32 const sEnd = (U32)(end >> vStepLog); - U32 const weight = sEnd - sStart; - if (weight < 1) - return ERROR(GENERIC); - norm[s] = (short)weight; - tmpTotal = end; - } } } - - return 0; -} - -size_t FSE_normalizeCount (short* normalizedCounter, unsigned tableLog, - const unsigned* count, size_t total, - unsigned maxSymbolValue, unsigned useLowProbCount) -{ - /* Sanity checks */ - if (tableLog==0) tableLog = FSE_DEFAULT_TABLELOG; - if (tableLog < FSE_MIN_TABLELOG) return ERROR(GENERIC); /* Unsupported size */ - if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge); /* Unsupported size */ - if (tableLog < FSE_minTableLog(total, maxSymbolValue)) return ERROR(GENERIC); /* Too small tableLog, compression potentially impossible */ - - { static U32 const rtbTable[] = { 0, 473195, 504333, 520860, 550000, 700000, 750000, 830000 }; - short const lowProbCount = useLowProbCount ? -1 : 1; - U64 const scale = 62 - tableLog; - U64 const step = ZSTD_div64((U64)1<<62, (U32)total); /* <== here, one division ! */ - U64 const vStep = 1ULL<<(scale-20); - int stillToDistribute = 1<> tableLog); - - for (s=0; s<=maxSymbolValue; s++) { - if (count[s] == total) return 0; /* rle special case */ - if (count[s] == 0) { normalizedCounter[s]=0; continue; } - if (count[s] <= lowThreshold) { - normalizedCounter[s] = lowProbCount; - stillToDistribute--; - } else { - short proba = (short)((count[s]*step) >> scale); - if (proba<8) { - U64 restToBeat = vStep * rtbTable[proba]; - proba += (count[s]*step) - ((U64)proba< restToBeat; - } - if (proba > largestP) { largestP=proba; largest=s; } - normalizedCounter[s] = proba; - stillToDistribute -= proba; - } } - if (-stillToDistribute >= (normalizedCounter[largest] >> 1)) { - /* corner case, need another normalization method */ - size_t const errorCode = FSE_normalizeM2(normalizedCounter, tableLog, count, total, maxSymbolValue, lowProbCount); - if (FSE_isError(errorCode)) return errorCode; - } - else normalizedCounter[largest] += (short)stillToDistribute; - } - -#if 0 - { /* Print Table (debug) */ - U32 s; - U32 nTotal = 0; - for (s=0; s<=maxSymbolValue; s++) - RAWLOG(2, "%3i: %4i \n", s, normalizedCounter[s]); - for (s=0; s<=maxSymbolValue; s++) - nTotal += abs(normalizedCounter[s]); - if (nTotal != (1U<>1); /* assumption : tableLog >= 1 */ - FSE_symbolCompressionTransform* const symbolTT = (FSE_symbolCompressionTransform*) (FSCT); - unsigned s; - - /* Sanity checks */ - if (nbBits < 1) return ERROR(GENERIC); /* min size */ - - /* header */ - tableU16[-2] = (U16) nbBits; - tableU16[-1] = (U16) maxSymbolValue; - - /* Build table */ - for (s=0; s FSE_MAX_TABLELOG*4+7 ) && (srcSize & 2)) { /* test bit 2 */ - FSE_encodeSymbol(&bitC, &CState2, *--ip); - FSE_encodeSymbol(&bitC, &CState1, *--ip); - FSE_FLUSHBITS(&bitC); - } - - /* 2 or 4 encoding per loop */ - while ( ip>istart ) { - - FSE_encodeSymbol(&bitC, &CState2, *--ip); - - if (sizeof(bitC.bitContainer)*8 < FSE_MAX_TABLELOG*2+7 ) /* this test must be static */ - FSE_FLUSHBITS(&bitC); - - FSE_encodeSymbol(&bitC, &CState1, *--ip); - - if (sizeof(bitC.bitContainer)*8 > FSE_MAX_TABLELOG*4+7 ) { /* this test must be static */ - FSE_encodeSymbol(&bitC, &CState2, *--ip); - FSE_encodeSymbol(&bitC, &CState1, *--ip); - } - - FSE_FLUSHBITS(&bitC); - } - - FSE_flushCState(&bitC, &CState2); - FSE_flushCState(&bitC, &CState1); - return BIT_closeCStream(&bitC); -} - -size_t FSE_compress_usingCTable (void* dst, size_t dstSize, - const void* src, size_t srcSize, - const FSE_CTable* ct) -{ - unsigned const fast = (dstSize >= FSE_BLOCKBOUND(srcSize)); - - if (fast) - return FSE_compress_usingCTable_generic(dst, dstSize, src, srcSize, ct, 1); - else - return FSE_compress_usingCTable_generic(dst, dstSize, src, srcSize, ct, 0); -} - - -size_t FSE_compressBound(size_t size) { return FSE_COMPRESSBOUND(size); } - -#ifndef ZSTD_NO_UNUSED_FUNCTIONS -/* FSE_compress_wksp() : - * Same as FSE_compress2(), but using an externally allocated scratch buffer (`workSpace`). - * `wkspSize` size must be `(1< not compressible */ - if (maxCount < (srcSize >> 7)) return 0; /* Heuristic : not compressible enough */ - } - - tableLog = FSE_optimalTableLog(tableLog, srcSize, maxSymbolValue); - CHECK_F( FSE_normalizeCount(norm, tableLog, count, srcSize, maxSymbolValue, /* useLowProbCount */ srcSize >= 2048) ); - - /* Write table description header */ - { CHECK_V_F(nc_err, FSE_writeNCount(op, oend-op, norm, maxSymbolValue, tableLog) ); - op += nc_err; - } - - /* Compress */ - CHECK_F( FSE_buildCTable_wksp(CTable, norm, maxSymbolValue, tableLog, scratchBuffer, scratchBufferSize) ); - { CHECK_V_F(cSize, FSE_compress_usingCTable(op, oend - op, src, srcSize, CTable) ); - if (cSize == 0) return 0; /* not enough space for compressed data */ - op += cSize; - } - - /* check compressibility */ - if ( (size_t)(op-ostart) >= srcSize-1 ) return 0; - - return op-ostart; -} - -typedef struct { - FSE_CTable CTable_max[FSE_CTABLE_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)]; - union { - U32 hist_wksp[HIST_WKSP_SIZE_U32]; - BYTE scratchBuffer[1 << FSE_MAX_TABLELOG]; - } workspace; -} fseWkspMax_t; - -size_t FSE_compress2 (void* dst, size_t dstCapacity, const void* src, size_t srcSize, unsigned maxSymbolValue, unsigned tableLog) -{ - fseWkspMax_t scratchBuffer; - DEBUG_STATIC_ASSERT(sizeof(scratchBuffer) >= FSE_COMPRESS_WKSP_SIZE_U32(FSE_MAX_TABLELOG, FSE_MAX_SYMBOL_VALUE)); /* compilation failures here means scratchBuffer is not large enough */ - if (tableLog > FSE_MAX_TABLELOG) return ERROR(tableLog_tooLarge); - return FSE_compress_wksp(dst, dstCapacity, src, srcSize, maxSymbolValue, tableLog, &scratchBuffer, sizeof(scratchBuffer)); -} - -size_t FSE_compress (void* dst, size_t dstCapacity, const void* src, size_t srcSize) -{ - return FSE_compress2(dst, dstCapacity, src, srcSize, FSE_MAX_SYMBOL_VALUE, FSE_DEFAULT_TABLELOG); -} -#endif - -#endif /* FSE_COMMONDEFS_ONLY */ diff --git a/dep/zstd/lib/compress/hist.c b/dep/zstd/lib/compress/hist.c deleted file mode 100644 index 073c57e75..000000000 --- a/dep/zstd/lib/compress/hist.c +++ /dev/null @@ -1,181 +0,0 @@ -/* ****************************************************************** - * hist : Histogram functions - * part of Finite State Entropy project - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy - * - Public forum : https://groups.google.com/forum/#!forum/lz4c - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -****************************************************************** */ - -/* --- dependencies --- */ -#include "../common/mem.h" /* U32, BYTE, etc. */ -#include "../common/debug.h" /* assert, DEBUGLOG */ -#include "../common/error_private.h" /* ERROR */ -#include "hist.h" - - -/* --- Error management --- */ -unsigned HIST_isError(size_t code) { return ERR_isError(code); } - -/*-************************************************************** - * Histogram functions - ****************************************************************/ -unsigned HIST_count_simple(unsigned* count, unsigned* maxSymbolValuePtr, - const void* src, size_t srcSize) -{ - const BYTE* ip = (const BYTE*)src; - const BYTE* const end = ip + srcSize; - unsigned maxSymbolValue = *maxSymbolValuePtr; - unsigned largestCount=0; - - ZSTD_memset(count, 0, (maxSymbolValue+1) * sizeof(*count)); - if (srcSize==0) { *maxSymbolValuePtr = 0; return 0; } - - while (ip largestCount) largestCount = count[s]; - } - - return largestCount; -} - -typedef enum { trustInput, checkMaxSymbolValue } HIST_checkInput_e; - -/* HIST_count_parallel_wksp() : - * store histogram into 4 intermediate tables, recombined at the end. - * this design makes better use of OoO cpus, - * and is noticeably faster when some values are heavily repeated. - * But it needs some additional workspace for intermediate tables. - * `workSpace` must be a U32 table of size >= HIST_WKSP_SIZE_U32. - * @return : largest histogram frequency, - * or an error code (notably when histogram's alphabet is larger than *maxSymbolValuePtr) */ -static size_t HIST_count_parallel_wksp( - unsigned* count, unsigned* maxSymbolValuePtr, - const void* source, size_t sourceSize, - HIST_checkInput_e check, - U32* const workSpace) -{ - const BYTE* ip = (const BYTE*)source; - const BYTE* const iend = ip+sourceSize; - size_t const countSize = (*maxSymbolValuePtr + 1) * sizeof(*count); - unsigned max=0; - U32* const Counting1 = workSpace; - U32* const Counting2 = Counting1 + 256; - U32* const Counting3 = Counting2 + 256; - U32* const Counting4 = Counting3 + 256; - - /* safety checks */ - assert(*maxSymbolValuePtr <= 255); - if (!sourceSize) { - ZSTD_memset(count, 0, countSize); - *maxSymbolValuePtr = 0; - return 0; - } - ZSTD_memset(workSpace, 0, 4*256*sizeof(unsigned)); - - /* by stripes of 16 bytes */ - { U32 cached = MEM_read32(ip); ip += 4; - while (ip < iend-15) { - U32 c = cached; cached = MEM_read32(ip); ip += 4; - Counting1[(BYTE) c ]++; - Counting2[(BYTE)(c>>8) ]++; - Counting3[(BYTE)(c>>16)]++; - Counting4[ c>>24 ]++; - c = cached; cached = MEM_read32(ip); ip += 4; - Counting1[(BYTE) c ]++; - Counting2[(BYTE)(c>>8) ]++; - Counting3[(BYTE)(c>>16)]++; - Counting4[ c>>24 ]++; - c = cached; cached = MEM_read32(ip); ip += 4; - Counting1[(BYTE) c ]++; - Counting2[(BYTE)(c>>8) ]++; - Counting3[(BYTE)(c>>16)]++; - Counting4[ c>>24 ]++; - c = cached; cached = MEM_read32(ip); ip += 4; - Counting1[(BYTE) c ]++; - Counting2[(BYTE)(c>>8) ]++; - Counting3[(BYTE)(c>>16)]++; - Counting4[ c>>24 ]++; - } - ip-=4; - } - - /* finish last symbols */ - while (ip max) max = Counting1[s]; - } } - - { unsigned maxSymbolValue = 255; - while (!Counting1[maxSymbolValue]) maxSymbolValue--; - if (check && maxSymbolValue > *maxSymbolValuePtr) return ERROR(maxSymbolValue_tooSmall); - *maxSymbolValuePtr = maxSymbolValue; - ZSTD_memmove(count, Counting1, countSize); /* in case count & Counting1 are overlapping */ - } - return (size_t)max; -} - -/* HIST_countFast_wksp() : - * Same as HIST_countFast(), but using an externally provided scratch buffer. - * `workSpace` is a writable buffer which must be 4-bytes aligned, - * `workSpaceSize` must be >= HIST_WKSP_SIZE - */ -size_t HIST_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr, - const void* source, size_t sourceSize, - void* workSpace, size_t workSpaceSize) -{ - if (sourceSize < 1500) /* heuristic threshold */ - return HIST_count_simple(count, maxSymbolValuePtr, source, sourceSize); - if ((size_t)workSpace & 3) return ERROR(GENERIC); /* must be aligned on 4-bytes boundaries */ - if (workSpaceSize < HIST_WKSP_SIZE) return ERROR(workSpace_tooSmall); - return HIST_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, trustInput, (U32*)workSpace); -} - -/* HIST_count_wksp() : - * Same as HIST_count(), but using an externally provided scratch buffer. - * `workSpace` size must be table of >= HIST_WKSP_SIZE_U32 unsigned */ -size_t HIST_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr, - const void* source, size_t sourceSize, - void* workSpace, size_t workSpaceSize) -{ - if ((size_t)workSpace & 3) return ERROR(GENERIC); /* must be aligned on 4-bytes boundaries */ - if (workSpaceSize < HIST_WKSP_SIZE) return ERROR(workSpace_tooSmall); - if (*maxSymbolValuePtr < 255) - return HIST_count_parallel_wksp(count, maxSymbolValuePtr, source, sourceSize, checkMaxSymbolValue, (U32*)workSpace); - *maxSymbolValuePtr = 255; - return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, workSpace, workSpaceSize); -} - -#ifndef ZSTD_NO_UNUSED_FUNCTIONS -/* fast variant (unsafe : won't check if src contains values beyond count[] limit) */ -size_t HIST_countFast(unsigned* count, unsigned* maxSymbolValuePtr, - const void* source, size_t sourceSize) -{ - unsigned tmpCounters[HIST_WKSP_SIZE_U32]; - return HIST_countFast_wksp(count, maxSymbolValuePtr, source, sourceSize, tmpCounters, sizeof(tmpCounters)); -} - -size_t HIST_count(unsigned* count, unsigned* maxSymbolValuePtr, - const void* src, size_t srcSize) -{ - unsigned tmpCounters[HIST_WKSP_SIZE_U32]; - return HIST_count_wksp(count, maxSymbolValuePtr, src, srcSize, tmpCounters, sizeof(tmpCounters)); -} -#endif diff --git a/dep/zstd/lib/compress/hist.h b/dep/zstd/lib/compress/hist.h deleted file mode 100644 index 228ed48a7..000000000 --- a/dep/zstd/lib/compress/hist.h +++ /dev/null @@ -1,75 +0,0 @@ -/* ****************************************************************** - * hist : Histogram functions - * part of Finite State Entropy project - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - FSE source repository : https://github.com/Cyan4973/FiniteStateEntropy - * - Public forum : https://groups.google.com/forum/#!forum/lz4c - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -****************************************************************** */ - -/* --- dependencies --- */ -#include "../common/zstd_deps.h" /* size_t */ - - -/* --- simple histogram functions --- */ - -/*! HIST_count(): - * Provides the precise count of each byte within a table 'count'. - * 'count' is a table of unsigned int, of minimum size (*maxSymbolValuePtr+1). - * Updates *maxSymbolValuePtr with actual largest symbol value detected. - * @return : count of the most frequent symbol (which isn't identified). - * or an error code, which can be tested using HIST_isError(). - * note : if return == srcSize, there is only one symbol. - */ -size_t HIST_count(unsigned* count, unsigned* maxSymbolValuePtr, - const void* src, size_t srcSize); - -unsigned HIST_isError(size_t code); /**< tells if a return value is an error code */ - - -/* --- advanced histogram functions --- */ - -#define HIST_WKSP_SIZE_U32 1024 -#define HIST_WKSP_SIZE (HIST_WKSP_SIZE_U32 * sizeof(unsigned)) -/** HIST_count_wksp() : - * Same as HIST_count(), but using an externally provided scratch buffer. - * Benefit is this function will use very little stack space. - * `workSpace` is a writable buffer which must be 4-bytes aligned, - * `workSpaceSize` must be >= HIST_WKSP_SIZE - */ -size_t HIST_count_wksp(unsigned* count, unsigned* maxSymbolValuePtr, - const void* src, size_t srcSize, - void* workSpace, size_t workSpaceSize); - -/** HIST_countFast() : - * same as HIST_count(), but blindly trusts that all byte values within src are <= *maxSymbolValuePtr. - * This function is unsafe, and will segfault if any value within `src` is `> *maxSymbolValuePtr` - */ -size_t HIST_countFast(unsigned* count, unsigned* maxSymbolValuePtr, - const void* src, size_t srcSize); - -/** HIST_countFast_wksp() : - * Same as HIST_countFast(), but using an externally provided scratch buffer. - * `workSpace` is a writable buffer which must be 4-bytes aligned, - * `workSpaceSize` must be >= HIST_WKSP_SIZE - */ -size_t HIST_countFast_wksp(unsigned* count, unsigned* maxSymbolValuePtr, - const void* src, size_t srcSize, - void* workSpace, size_t workSpaceSize); - -/*! HIST_count_simple() : - * Same as HIST_countFast(), this function is unsafe, - * and will segfault if any value within `src` is `> *maxSymbolValuePtr`. - * It is also a bit slower for large inputs. - * However, it does not need any additional memory (not even on stack). - * @return : count of the most frequent symbol. - * Note this function doesn't produce any error (i.e. it must succeed). - */ -unsigned HIST_count_simple(unsigned* count, unsigned* maxSymbolValuePtr, - const void* src, size_t srcSize); diff --git a/dep/zstd/lib/compress/huf_compress.c b/dep/zstd/lib/compress/huf_compress.c deleted file mode 100644 index 2b3d6adc2..000000000 --- a/dep/zstd/lib/compress/huf_compress.c +++ /dev/null @@ -1,1370 +0,0 @@ -/* ****************************************************************** - * Huffman encoder, part of New Generation Entropy library - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy - * - Public forum : https://groups.google.com/forum/#!forum/lz4c - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -****************************************************************** */ - -/* ************************************************************** -* Compiler specifics -****************************************************************/ -#ifdef _MSC_VER /* Visual Studio */ -# pragma warning(disable : 4127) /* disable: C4127: conditional expression is constant */ -#endif - - -/* ************************************************************** -* Includes -****************************************************************/ -#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memset */ -#include "../common/compiler.h" -#include "../common/bitstream.h" -#include "hist.h" -#define FSE_STATIC_LINKING_ONLY /* FSE_optimalTableLog_internal */ -#include "../common/fse.h" /* header compression */ -#define HUF_STATIC_LINKING_ONLY -#include "../common/huf.h" -#include "../common/error_private.h" - - -/* ************************************************************** -* Error Management -****************************************************************/ -#define HUF_isError ERR_isError -#define HUF_STATIC_ASSERT(c) DEBUG_STATIC_ASSERT(c) /* use only *after* variable declarations */ - - -/* ************************************************************** -* Utils -****************************************************************/ -unsigned HUF_optimalTableLog(unsigned maxTableLog, size_t srcSize, unsigned maxSymbolValue) -{ - return FSE_optimalTableLog_internal(maxTableLog, srcSize, maxSymbolValue, 1); -} - - -/* ******************************************************* -* HUF : Huffman block compression -*********************************************************/ -#define HUF_WORKSPACE_MAX_ALIGNMENT 8 - -static void* HUF_alignUpWorkspace(void* workspace, size_t* workspaceSizePtr, size_t align) -{ - size_t const mask = align - 1; - size_t const rem = (size_t)workspace & mask; - size_t const add = (align - rem) & mask; - BYTE* const aligned = (BYTE*)workspace + add; - assert((align & (align - 1)) == 0); /* pow 2 */ - assert(align <= HUF_WORKSPACE_MAX_ALIGNMENT); - if (*workspaceSizePtr >= add) { - assert(add < align); - assert(((size_t)aligned & mask) == 0); - *workspaceSizePtr -= add; - return aligned; - } else { - *workspaceSizePtr = 0; - return NULL; - } -} - - -/* HUF_compressWeights() : - * Same as FSE_compress(), but dedicated to huff0's weights compression. - * The use case needs much less stack memory. - * Note : all elements within weightTable are supposed to be <= HUF_TABLELOG_MAX. - */ -#define MAX_FSE_TABLELOG_FOR_HUFF_HEADER 6 - -typedef struct { - FSE_CTable CTable[FSE_CTABLE_SIZE_U32(MAX_FSE_TABLELOG_FOR_HUFF_HEADER, HUF_TABLELOG_MAX)]; - U32 scratchBuffer[FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(HUF_TABLELOG_MAX, MAX_FSE_TABLELOG_FOR_HUFF_HEADER)]; - unsigned count[HUF_TABLELOG_MAX+1]; - S16 norm[HUF_TABLELOG_MAX+1]; -} HUF_CompressWeightsWksp; - -static size_t HUF_compressWeights(void* dst, size_t dstSize, const void* weightTable, size_t wtSize, void* workspace, size_t workspaceSize) -{ - BYTE* const ostart = (BYTE*) dst; - BYTE* op = ostart; - BYTE* const oend = ostart + dstSize; - - unsigned maxSymbolValue = HUF_TABLELOG_MAX; - U32 tableLog = MAX_FSE_TABLELOG_FOR_HUFF_HEADER; - HUF_CompressWeightsWksp* wksp = (HUF_CompressWeightsWksp*)HUF_alignUpWorkspace(workspace, &workspaceSize, ZSTD_ALIGNOF(U32)); - - if (workspaceSize < sizeof(HUF_CompressWeightsWksp)) return ERROR(GENERIC); - - /* init conditions */ - if (wtSize <= 1) return 0; /* Not compressible */ - - /* Scan input and build symbol stats */ - { unsigned const maxCount = HIST_count_simple(wksp->count, &maxSymbolValue, weightTable, wtSize); /* never fails */ - if (maxCount == wtSize) return 1; /* only a single symbol in src : rle */ - if (maxCount == 1) return 0; /* each symbol present maximum once => not compressible */ - } - - tableLog = FSE_optimalTableLog(tableLog, wtSize, maxSymbolValue); - CHECK_F( FSE_normalizeCount(wksp->norm, tableLog, wksp->count, wtSize, maxSymbolValue, /* useLowProbCount */ 0) ); - - /* Write table description header */ - { CHECK_V_F(hSize, FSE_writeNCount(op, (size_t)(oend-op), wksp->norm, maxSymbolValue, tableLog) ); - op += hSize; - } - - /* Compress */ - CHECK_F( FSE_buildCTable_wksp(wksp->CTable, wksp->norm, maxSymbolValue, tableLog, wksp->scratchBuffer, sizeof(wksp->scratchBuffer)) ); - { CHECK_V_F(cSize, FSE_compress_usingCTable(op, (size_t)(oend - op), weightTable, wtSize, wksp->CTable) ); - if (cSize == 0) return 0; /* not enough space for compressed data */ - op += cSize; - } - - return (size_t)(op-ostart); -} - -static size_t HUF_getNbBits(HUF_CElt elt) -{ - return elt & 0xFF; -} - -static size_t HUF_getNbBitsFast(HUF_CElt elt) -{ - return elt; -} - -static size_t HUF_getValue(HUF_CElt elt) -{ - return elt & ~0xFF; -} - -static size_t HUF_getValueFast(HUF_CElt elt) -{ - return elt; -} - -static void HUF_setNbBits(HUF_CElt* elt, size_t nbBits) -{ - assert(nbBits <= HUF_TABLELOG_ABSOLUTEMAX); - *elt = nbBits; -} - -static void HUF_setValue(HUF_CElt* elt, size_t value) -{ - size_t const nbBits = HUF_getNbBits(*elt); - if (nbBits > 0) { - assert((value >> nbBits) == 0); - *elt |= value << (sizeof(HUF_CElt) * 8 - nbBits); - } -} - -typedef struct { - HUF_CompressWeightsWksp wksp; - BYTE bitsToWeight[HUF_TABLELOG_MAX + 1]; /* precomputed conversion table */ - BYTE huffWeight[HUF_SYMBOLVALUE_MAX]; -} HUF_WriteCTableWksp; - -size_t HUF_writeCTable_wksp(void* dst, size_t maxDstSize, - const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog, - void* workspace, size_t workspaceSize) -{ - HUF_CElt const* const ct = CTable + 1; - BYTE* op = (BYTE*)dst; - U32 n; - HUF_WriteCTableWksp* wksp = (HUF_WriteCTableWksp*)HUF_alignUpWorkspace(workspace, &workspaceSize, ZSTD_ALIGNOF(U32)); - - /* check conditions */ - if (workspaceSize < sizeof(HUF_WriteCTableWksp)) return ERROR(GENERIC); - if (maxSymbolValue > HUF_SYMBOLVALUE_MAX) return ERROR(maxSymbolValue_tooLarge); - - /* convert to weight */ - wksp->bitsToWeight[0] = 0; - for (n=1; nbitsToWeight[n] = (BYTE)(huffLog + 1 - n); - for (n=0; nhuffWeight[n] = wksp->bitsToWeight[HUF_getNbBits(ct[n])]; - - /* attempt weights compression by FSE */ - if (maxDstSize < 1) return ERROR(dstSize_tooSmall); - { CHECK_V_F(hSize, HUF_compressWeights(op+1, maxDstSize-1, wksp->huffWeight, maxSymbolValue, &wksp->wksp, sizeof(wksp->wksp)) ); - if ((hSize>1) & (hSize < maxSymbolValue/2)) { /* FSE compressed */ - op[0] = (BYTE)hSize; - return hSize+1; - } } - - /* write raw values as 4-bits (max : 15) */ - if (maxSymbolValue > (256-128)) return ERROR(GENERIC); /* should not happen : likely means source cannot be compressed */ - if (((maxSymbolValue+1)/2) + 1 > maxDstSize) return ERROR(dstSize_tooSmall); /* not enough space within dst buffer */ - op[0] = (BYTE)(128 /*special case*/ + (maxSymbolValue-1)); - wksp->huffWeight[maxSymbolValue] = 0; /* to be sure it doesn't cause msan issue in final combination */ - for (n=0; nhuffWeight[n] << 4) + wksp->huffWeight[n+1]); - return ((maxSymbolValue+1)/2) + 1; -} - -/*! HUF_writeCTable() : - `CTable` : Huffman tree to save, using huf representation. - @return : size of saved CTable */ -size_t HUF_writeCTable (void* dst, size_t maxDstSize, - const HUF_CElt* CTable, unsigned maxSymbolValue, unsigned huffLog) -{ - HUF_WriteCTableWksp wksp; - return HUF_writeCTable_wksp(dst, maxDstSize, CTable, maxSymbolValue, huffLog, &wksp, sizeof(wksp)); -} - - -size_t HUF_readCTable (HUF_CElt* CTable, unsigned* maxSymbolValuePtr, const void* src, size_t srcSize, unsigned* hasZeroWeights) -{ - BYTE huffWeight[HUF_SYMBOLVALUE_MAX + 1]; /* init not required, even though some static analyzer may complain */ - U32 rankVal[HUF_TABLELOG_ABSOLUTEMAX + 1]; /* large enough for values from 0 to 16 */ - U32 tableLog = 0; - U32 nbSymbols = 0; - HUF_CElt* const ct = CTable + 1; - - /* get symbol weights */ - CHECK_V_F(readSize, HUF_readStats(huffWeight, HUF_SYMBOLVALUE_MAX+1, rankVal, &nbSymbols, &tableLog, src, srcSize)); - *hasZeroWeights = (rankVal[0] > 0); - - /* check result */ - if (tableLog > HUF_TABLELOG_MAX) return ERROR(tableLog_tooLarge); - if (nbSymbols > *maxSymbolValuePtr+1) return ERROR(maxSymbolValue_tooSmall); - - CTable[0] = tableLog; - - /* Prepare base value per rank */ - { U32 n, nextRankStart = 0; - for (n=1; n<=tableLog; n++) { - U32 curr = nextRankStart; - nextRankStart += (rankVal[n] << (n-1)); - rankVal[n] = curr; - } } - - /* fill nbBits */ - { U32 n; for (n=0; nn=tableLog+1 */ - U16 valPerRank[HUF_TABLELOG_MAX+2] = {0}; - { U32 n; for (n=0; n0; n--) { /* start at n=tablelog <-> w=1 */ - valPerRank[n] = min; /* get starting value within each rank */ - min += nbPerRank[n]; - min >>= 1; - } } - /* assign value within rank, symbol order */ - { U32 n; for (n=0; n maxNbBits to be maxNbBits. Then it adjusts - * the tree to so that it is a valid canonical Huffman tree. - * - * @pre The sum of the ranks of each symbol == 2^largestBits, - * where largestBits == huffNode[lastNonNull].nbBits. - * @post The sum of the ranks of each symbol == 2^largestBits, - * where largestBits is the return value <= maxNbBits. - * - * @param huffNode The Huffman tree modified in place to enforce maxNbBits. - * @param lastNonNull The symbol with the lowest count in the Huffman tree. - * @param maxNbBits The maximum allowed number of bits, which the Huffman tree - * may not respect. After this function the Huffman tree will - * respect maxNbBits. - * @return The maximum number of bits of the Huffman tree after adjustment, - * necessarily no more than maxNbBits. - */ -static U32 HUF_setMaxHeight(nodeElt* huffNode, U32 lastNonNull, U32 maxNbBits) -{ - const U32 largestBits = huffNode[lastNonNull].nbBits; - /* early exit : no elt > maxNbBits, so the tree is already valid. */ - if (largestBits <= maxNbBits) return largestBits; - - /* there are several too large elements (at least >= 2) */ - { int totalCost = 0; - const U32 baseCost = 1 << (largestBits - maxNbBits); - int n = (int)lastNonNull; - - /* Adjust any ranks > maxNbBits to maxNbBits. - * Compute totalCost, which is how far the sum of the ranks is - * we are over 2^largestBits after adjust the offending ranks. - */ - while (huffNode[n].nbBits > maxNbBits) { - totalCost += baseCost - (1 << (largestBits - huffNode[n].nbBits)); - huffNode[n].nbBits = (BYTE)maxNbBits; - n--; - } - /* n stops at huffNode[n].nbBits <= maxNbBits */ - assert(huffNode[n].nbBits <= maxNbBits); - /* n end at index of smallest symbol using < maxNbBits */ - while (huffNode[n].nbBits == maxNbBits) --n; - - /* renorm totalCost from 2^largestBits to 2^maxNbBits - * note : totalCost is necessarily a multiple of baseCost */ - assert((totalCost & (baseCost - 1)) == 0); - totalCost >>= (largestBits - maxNbBits); - assert(totalCost > 0); - - /* repay normalized cost */ - { U32 const noSymbol = 0xF0F0F0F0; - U32 rankLast[HUF_TABLELOG_MAX+2]; - - /* Get pos of last (smallest = lowest cum. count) symbol per rank */ - ZSTD_memset(rankLast, 0xF0, sizeof(rankLast)); - { U32 currentNbBits = maxNbBits; - int pos; - for (pos=n ; pos >= 0; pos--) { - if (huffNode[pos].nbBits >= currentNbBits) continue; - currentNbBits = huffNode[pos].nbBits; /* < maxNbBits */ - rankLast[maxNbBits-currentNbBits] = (U32)pos; - } } - - while (totalCost > 0) { - /* Try to reduce the next power of 2 above totalCost because we - * gain back half the rank. - */ - U32 nBitsToDecrease = BIT_highbit32((U32)totalCost) + 1; - for ( ; nBitsToDecrease > 1; nBitsToDecrease--) { - U32 const highPos = rankLast[nBitsToDecrease]; - U32 const lowPos = rankLast[nBitsToDecrease-1]; - if (highPos == noSymbol) continue; - /* Decrease highPos if no symbols of lowPos or if it is - * not cheaper to remove 2 lowPos than highPos. - */ - if (lowPos == noSymbol) break; - { U32 const highTotal = huffNode[highPos].count; - U32 const lowTotal = 2 * huffNode[lowPos].count; - if (highTotal <= lowTotal) break; - } } - /* only triggered when no more rank 1 symbol left => find closest one (note : there is necessarily at least one !) */ - assert(rankLast[nBitsToDecrease] != noSymbol || nBitsToDecrease == 1); - /* HUF_MAX_TABLELOG test just to please gcc 5+; but it should not be necessary */ - while ((nBitsToDecrease<=HUF_TABLELOG_MAX) && (rankLast[nBitsToDecrease] == noSymbol)) - nBitsToDecrease++; - assert(rankLast[nBitsToDecrease] != noSymbol); - /* Increase the number of bits to gain back half the rank cost. */ - totalCost -= 1 << (nBitsToDecrease-1); - huffNode[rankLast[nBitsToDecrease]].nbBits++; - - /* Fix up the new rank. - * If the new rank was empty, this symbol is now its smallest. - * Otherwise, this symbol will be the largest in the new rank so no adjustment. - */ - if (rankLast[nBitsToDecrease-1] == noSymbol) - rankLast[nBitsToDecrease-1] = rankLast[nBitsToDecrease]; - /* Fix up the old rank. - * If the symbol was at position 0, meaning it was the highest weight symbol in the tree, - * it must be the only symbol in its rank, so the old rank now has no symbols. - * Otherwise, since the Huffman nodes are sorted by count, the previous position is now - * the smallest node in the rank. If the previous position belongs to a different rank, - * then the rank is now empty. - */ - if (rankLast[nBitsToDecrease] == 0) /* special case, reached largest symbol */ - rankLast[nBitsToDecrease] = noSymbol; - else { - rankLast[nBitsToDecrease]--; - if (huffNode[rankLast[nBitsToDecrease]].nbBits != maxNbBits-nBitsToDecrease) - rankLast[nBitsToDecrease] = noSymbol; /* this rank is now empty */ - } - } /* while (totalCost > 0) */ - - /* If we've removed too much weight, then we have to add it back. - * To avoid overshooting again, we only adjust the smallest rank. - * We take the largest nodes from the lowest rank 0 and move them - * to rank 1. There's guaranteed to be enough rank 0 symbols because - * TODO. - */ - while (totalCost < 0) { /* Sometimes, cost correction overshoot */ - /* special case : no rank 1 symbol (using maxNbBits-1); - * let's create one from largest rank 0 (using maxNbBits). - */ - if (rankLast[1] == noSymbol) { - while (huffNode[n].nbBits == maxNbBits) n--; - huffNode[n+1].nbBits--; - assert(n >= 0); - rankLast[1] = (U32)(n+1); - totalCost++; - continue; - } - huffNode[ rankLast[1] + 1 ].nbBits--; - rankLast[1]++; - totalCost ++; - } - } /* repay normalized cost */ - } /* there are several too large elements (at least >= 2) */ - - return maxNbBits; -} - -typedef struct { - U16 base; - U16 curr; -} rankPos; - -typedef nodeElt huffNodeTable[HUF_CTABLE_WORKSPACE_SIZE_U32]; - -/* Number of buckets available for HUF_sort() */ -#define RANK_POSITION_TABLE_SIZE 192 - -typedef struct { - huffNodeTable huffNodeTbl; - rankPos rankPosition[RANK_POSITION_TABLE_SIZE]; -} HUF_buildCTable_wksp_tables; - -/* RANK_POSITION_DISTINCT_COUNT_CUTOFF == Cutoff point in HUF_sort() buckets for which we use log2 bucketing. - * Strategy is to use as many buckets as possible for representing distinct - * counts while using the remainder to represent all "large" counts. - * - * To satisfy this requirement for 192 buckets, we can do the following: - * Let buckets 0-166 represent distinct counts of [0, 166] - * Let buckets 166 to 192 represent all remaining counts up to RANK_POSITION_MAX_COUNT_LOG using log2 bucketing. - */ -#define RANK_POSITION_MAX_COUNT_LOG 32 -#define RANK_POSITION_LOG_BUCKETS_BEGIN (RANK_POSITION_TABLE_SIZE - 1) - RANK_POSITION_MAX_COUNT_LOG - 1 /* == 158 */ -#define RANK_POSITION_DISTINCT_COUNT_CUTOFF RANK_POSITION_LOG_BUCKETS_BEGIN + BIT_highbit32(RANK_POSITION_LOG_BUCKETS_BEGIN) /* == 166 */ - -/* Return the appropriate bucket index for a given count. See definition of - * RANK_POSITION_DISTINCT_COUNT_CUTOFF for explanation of bucketing strategy. - */ -static U32 HUF_getIndex(U32 const count) { - return (count < RANK_POSITION_DISTINCT_COUNT_CUTOFF) - ? count - : BIT_highbit32(count) + RANK_POSITION_LOG_BUCKETS_BEGIN; -} - -/* Helper swap function for HUF_quickSortPartition() */ -static void HUF_swapNodes(nodeElt* a, nodeElt* b) { - nodeElt tmp = *a; - *a = *b; - *b = tmp; -} - -/* Returns 0 if the huffNode array is not sorted by descending count */ -MEM_STATIC int HUF_isSorted(nodeElt huffNode[], U32 const maxSymbolValue1) { - U32 i; - for (i = 1; i < maxSymbolValue1; ++i) { - if (huffNode[i].count > huffNode[i-1].count) { - return 0; - } - } - return 1; -} - -/* Insertion sort by descending order */ -HINT_INLINE void HUF_insertionSort(nodeElt huffNode[], int const low, int const high) { - int i; - int const size = high-low+1; - huffNode += low; - for (i = 1; i < size; ++i) { - nodeElt const key = huffNode[i]; - int j = i - 1; - while (j >= 0 && huffNode[j].count < key.count) { - huffNode[j + 1] = huffNode[j]; - j--; - } - huffNode[j + 1] = key; - } -} - -/* Pivot helper function for quicksort. */ -static int HUF_quickSortPartition(nodeElt arr[], int const low, int const high) { - /* Simply select rightmost element as pivot. "Better" selectors like - * median-of-three don't experimentally appear to have any benefit. - */ - U32 const pivot = arr[high].count; - int i = low - 1; - int j = low; - for ( ; j < high; j++) { - if (arr[j].count > pivot) { - i++; - HUF_swapNodes(&arr[i], &arr[j]); - } - } - HUF_swapNodes(&arr[i + 1], &arr[high]); - return i + 1; -} - -/* Classic quicksort by descending with partially iterative calls - * to reduce worst case callstack size. - */ -static void HUF_simpleQuickSort(nodeElt arr[], int low, int high) { - int const kInsertionSortThreshold = 8; - if (high - low < kInsertionSortThreshold) { - HUF_insertionSort(arr, low, high); - return; - } - while (low < high) { - int const idx = HUF_quickSortPartition(arr, low, high); - if (idx - low < high - idx) { - HUF_simpleQuickSort(arr, low, idx - 1); - low = idx + 1; - } else { - HUF_simpleQuickSort(arr, idx + 1, high); - high = idx - 1; - } - } -} - -/** - * HUF_sort(): - * Sorts the symbols [0, maxSymbolValue] by count[symbol] in decreasing order. - * This is a typical bucket sorting strategy that uses either quicksort or insertion sort to sort each bucket. - * - * @param[out] huffNode Sorted symbols by decreasing count. Only members `.count` and `.byte` are filled. - * Must have (maxSymbolValue + 1) entries. - * @param[in] count Histogram of the symbols. - * @param[in] maxSymbolValue Maximum symbol value. - * @param rankPosition This is a scratch workspace. Must have RANK_POSITION_TABLE_SIZE entries. - */ -static void HUF_sort(nodeElt huffNode[], const unsigned count[], U32 const maxSymbolValue, rankPos rankPosition[]) { - U32 n; - U32 const maxSymbolValue1 = maxSymbolValue+1; - - /* Compute base and set curr to base. - * For symbol s let lowerRank = HUF_getIndex(count[n]) and rank = lowerRank + 1. - * See HUF_getIndex to see bucketing strategy. - * We attribute each symbol to lowerRank's base value, because we want to know where - * each rank begins in the output, so for rank R we want to count ranks R+1 and above. - */ - ZSTD_memset(rankPosition, 0, sizeof(*rankPosition) * RANK_POSITION_TABLE_SIZE); - for (n = 0; n < maxSymbolValue1; ++n) { - U32 lowerRank = HUF_getIndex(count[n]); - assert(lowerRank < RANK_POSITION_TABLE_SIZE - 1); - rankPosition[lowerRank].base++; - } - - assert(rankPosition[RANK_POSITION_TABLE_SIZE - 1].base == 0); - /* Set up the rankPosition table */ - for (n = RANK_POSITION_TABLE_SIZE - 1; n > 0; --n) { - rankPosition[n-1].base += rankPosition[n].base; - rankPosition[n-1].curr = rankPosition[n-1].base; - } - - /* Insert each symbol into their appropriate bucket, setting up rankPosition table. */ - for (n = 0; n < maxSymbolValue1; ++n) { - U32 const c = count[n]; - U32 const r = HUF_getIndex(c) + 1; - U32 const pos = rankPosition[r].curr++; - assert(pos < maxSymbolValue1); - huffNode[pos].count = c; - huffNode[pos].byte = (BYTE)n; - } - - /* Sort each bucket. */ - for (n = RANK_POSITION_DISTINCT_COUNT_CUTOFF; n < RANK_POSITION_TABLE_SIZE - 1; ++n) { - U32 const bucketSize = rankPosition[n].curr-rankPosition[n].base; - U32 const bucketStartIdx = rankPosition[n].base; - if (bucketSize > 1) { - assert(bucketStartIdx < maxSymbolValue1); - HUF_simpleQuickSort(huffNode + bucketStartIdx, 0, bucketSize-1); - } - } - - assert(HUF_isSorted(huffNode, maxSymbolValue1)); -} - -/** HUF_buildCTable_wksp() : - * Same as HUF_buildCTable(), but using externally allocated scratch buffer. - * `workSpace` must be aligned on 4-bytes boundaries, and be at least as large as sizeof(HUF_buildCTable_wksp_tables). - */ -#define STARTNODE (HUF_SYMBOLVALUE_MAX+1) - -/* HUF_buildTree(): - * Takes the huffNode array sorted by HUF_sort() and builds an unlimited-depth Huffman tree. - * - * @param huffNode The array sorted by HUF_sort(). Builds the Huffman tree in this array. - * @param maxSymbolValue The maximum symbol value. - * @return The smallest node in the Huffman tree (by count). - */ -static int HUF_buildTree(nodeElt* huffNode, U32 maxSymbolValue) -{ - nodeElt* const huffNode0 = huffNode - 1; - int nonNullRank; - int lowS, lowN; - int nodeNb = STARTNODE; - int n, nodeRoot; - /* init for parents */ - nonNullRank = (int)maxSymbolValue; - while(huffNode[nonNullRank].count == 0) nonNullRank--; - lowS = nonNullRank; nodeRoot = nodeNb + lowS - 1; lowN = nodeNb; - huffNode[nodeNb].count = huffNode[lowS].count + huffNode[lowS-1].count; - huffNode[lowS].parent = huffNode[lowS-1].parent = (U16)nodeNb; - nodeNb++; lowS-=2; - for (n=nodeNb; n<=nodeRoot; n++) huffNode[n].count = (U32)(1U<<30); - huffNode0[0].count = (U32)(1U<<31); /* fake entry, strong barrier */ - - /* create parents */ - while (nodeNb <= nodeRoot) { - int const n1 = (huffNode[lowS].count < huffNode[lowN].count) ? lowS-- : lowN++; - int const n2 = (huffNode[lowS].count < huffNode[lowN].count) ? lowS-- : lowN++; - huffNode[nodeNb].count = huffNode[n1].count + huffNode[n2].count; - huffNode[n1].parent = huffNode[n2].parent = (U16)nodeNb; - nodeNb++; - } - - /* distribute weights (unlimited tree height) */ - huffNode[nodeRoot].nbBits = 0; - for (n=nodeRoot-1; n>=STARTNODE; n--) - huffNode[n].nbBits = huffNode[ huffNode[n].parent ].nbBits + 1; - for (n=0; n<=nonNullRank; n++) - huffNode[n].nbBits = huffNode[ huffNode[n].parent ].nbBits + 1; - - return nonNullRank; -} - -/** - * HUF_buildCTableFromTree(): - * Build the CTable given the Huffman tree in huffNode. - * - * @param[out] CTable The output Huffman CTable. - * @param huffNode The Huffman tree. - * @param nonNullRank The last and smallest node in the Huffman tree. - * @param maxSymbolValue The maximum symbol value. - * @param maxNbBits The exact maximum number of bits used in the Huffman tree. - */ -static void HUF_buildCTableFromTree(HUF_CElt* CTable, nodeElt const* huffNode, int nonNullRank, U32 maxSymbolValue, U32 maxNbBits) -{ - HUF_CElt* const ct = CTable + 1; - /* fill result into ctable (val, nbBits) */ - int n; - U16 nbPerRank[HUF_TABLELOG_MAX+1] = {0}; - U16 valPerRank[HUF_TABLELOG_MAX+1] = {0}; - int const alphabetSize = (int)(maxSymbolValue + 1); - for (n=0; n<=nonNullRank; n++) - nbPerRank[huffNode[n].nbBits]++; - /* determine starting value per rank */ - { U16 min = 0; - for (n=(int)maxNbBits; n>0; n--) { - valPerRank[n] = min; /* get starting value within each rank */ - min += nbPerRank[n]; - min >>= 1; - } } - for (n=0; nhuffNodeTbl; - nodeElt* const huffNode = huffNode0+1; - int nonNullRank; - - /* safety checks */ - if (wkspSize < sizeof(HUF_buildCTable_wksp_tables)) - return ERROR(workSpace_tooSmall); - if (maxNbBits == 0) maxNbBits = HUF_TABLELOG_DEFAULT; - if (maxSymbolValue > HUF_SYMBOLVALUE_MAX) - return ERROR(maxSymbolValue_tooLarge); - ZSTD_memset(huffNode0, 0, sizeof(huffNodeTable)); - - /* sort, decreasing order */ - HUF_sort(huffNode, count, maxSymbolValue, wksp_tables->rankPosition); - - /* build tree */ - nonNullRank = HUF_buildTree(huffNode, maxSymbolValue); - - /* enforce maxTableLog */ - maxNbBits = HUF_setMaxHeight(huffNode, (U32)nonNullRank, maxNbBits); - if (maxNbBits > HUF_TABLELOG_MAX) return ERROR(GENERIC); /* check fit into table */ - - HUF_buildCTableFromTree(CTable, huffNode, nonNullRank, maxSymbolValue, maxNbBits); - - return maxNbBits; -} - -size_t HUF_estimateCompressedSize(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue) -{ - HUF_CElt const* ct = CTable + 1; - size_t nbBits = 0; - int s; - for (s = 0; s <= (int)maxSymbolValue; ++s) { - nbBits += HUF_getNbBits(ct[s]) * count[s]; - } - return nbBits >> 3; -} - -int HUF_validateCTable(const HUF_CElt* CTable, const unsigned* count, unsigned maxSymbolValue) { - HUF_CElt const* ct = CTable + 1; - int bad = 0; - int s; - for (s = 0; s <= (int)maxSymbolValue; ++s) { - bad |= (count[s] != 0) & (HUF_getNbBits(ct[s]) == 0); - } - return !bad; -} - -size_t HUF_compressBound(size_t size) { return HUF_COMPRESSBOUND(size); } - -/** HUF_CStream_t: - * Huffman uses its own BIT_CStream_t implementation. - * There are three major differences from BIT_CStream_t: - * 1. HUF_addBits() takes a HUF_CElt (size_t) which is - * the pair (nbBits, value) in the format: - * format: - * - Bits [0, 4) = nbBits - * - Bits [4, 64 - nbBits) = 0 - * - Bits [64 - nbBits, 64) = value - * 2. The bitContainer is built from the upper bits and - * right shifted. E.g. to add a new value of N bits - * you right shift the bitContainer by N, then or in - * the new value into the N upper bits. - * 3. The bitstream has two bit containers. You can add - * bits to the second container and merge them into - * the first container. - */ - -#define HUF_BITS_IN_CONTAINER (sizeof(size_t) * 8) - -typedef struct { - size_t bitContainer[2]; - size_t bitPos[2]; - - BYTE* startPtr; - BYTE* ptr; - BYTE* endPtr; -} HUF_CStream_t; - -/**! HUF_initCStream(): - * Initializes the bitstream. - * @returns 0 or an error code. - */ -static size_t HUF_initCStream(HUF_CStream_t* bitC, - void* startPtr, size_t dstCapacity) -{ - ZSTD_memset(bitC, 0, sizeof(*bitC)); - bitC->startPtr = (BYTE*)startPtr; - bitC->ptr = bitC->startPtr; - bitC->endPtr = bitC->startPtr + dstCapacity - sizeof(bitC->bitContainer[0]); - if (dstCapacity <= sizeof(bitC->bitContainer[0])) return ERROR(dstSize_tooSmall); - return 0; -} - -/*! HUF_addBits(): - * Adds the symbol stored in HUF_CElt elt to the bitstream. - * - * @param elt The element we're adding. This is a (nbBits, value) pair. - * See the HUF_CStream_t docs for the format. - * @param idx Insert into the bitstream at this idx. - * @param kFast This is a template parameter. If the bitstream is guaranteed - * to have at least 4 unused bits after this call it may be 1, - * otherwise it must be 0. HUF_addBits() is faster when fast is set. - */ -FORCE_INLINE_TEMPLATE void HUF_addBits(HUF_CStream_t* bitC, HUF_CElt elt, int idx, int kFast) -{ - assert(idx <= 1); - assert(HUF_getNbBits(elt) <= HUF_TABLELOG_ABSOLUTEMAX); - /* This is efficient on x86-64 with BMI2 because shrx - * only reads the low 6 bits of the register. The compiler - * knows this and elides the mask. When fast is set, - * every operation can use the same value loaded from elt. - */ - bitC->bitContainer[idx] >>= HUF_getNbBits(elt); - bitC->bitContainer[idx] |= kFast ? HUF_getValueFast(elt) : HUF_getValue(elt); - /* We only read the low 8 bits of bitC->bitPos[idx] so it - * doesn't matter that the high bits have noise from the value. - */ - bitC->bitPos[idx] += HUF_getNbBitsFast(elt); - assert((bitC->bitPos[idx] & 0xFF) <= HUF_BITS_IN_CONTAINER); - /* The last 4-bits of elt are dirty if fast is set, - * so we must not be overwriting bits that have already been - * inserted into the bit container. - */ -#if DEBUGLEVEL >= 1 - { - size_t const nbBits = HUF_getNbBits(elt); - size_t const dirtyBits = nbBits == 0 ? 0 : BIT_highbit32((U32)nbBits) + 1; - (void)dirtyBits; - /* Middle bits are 0. */ - assert(((elt >> dirtyBits) << (dirtyBits + nbBits)) == 0); - /* We didn't overwrite any bits in the bit container. */ - assert(!kFast || (bitC->bitPos[idx] & 0xFF) <= HUF_BITS_IN_CONTAINER); - (void)dirtyBits; - } -#endif -} - -FORCE_INLINE_TEMPLATE void HUF_zeroIndex1(HUF_CStream_t* bitC) -{ - bitC->bitContainer[1] = 0; - bitC->bitPos[1] = 0; -} - -/*! HUF_mergeIndex1() : - * Merges the bit container @ index 1 into the bit container @ index 0 - * and zeros the bit container @ index 1. - */ -FORCE_INLINE_TEMPLATE void HUF_mergeIndex1(HUF_CStream_t* bitC) -{ - assert((bitC->bitPos[1] & 0xFF) < HUF_BITS_IN_CONTAINER); - bitC->bitContainer[0] >>= (bitC->bitPos[1] & 0xFF); - bitC->bitContainer[0] |= bitC->bitContainer[1]; - bitC->bitPos[0] += bitC->bitPos[1]; - assert((bitC->bitPos[0] & 0xFF) <= HUF_BITS_IN_CONTAINER); -} - -/*! HUF_flushBits() : -* Flushes the bits in the bit container @ index 0. -* -* @post bitPos will be < 8. -* @param kFast If kFast is set then we must know a-priori that -* the bit container will not overflow. -*/ -FORCE_INLINE_TEMPLATE void HUF_flushBits(HUF_CStream_t* bitC, int kFast) -{ - /* The upper bits of bitPos are noisy, so we must mask by 0xFF. */ - size_t const nbBits = bitC->bitPos[0] & 0xFF; - size_t const nbBytes = nbBits >> 3; - /* The top nbBits bits of bitContainer are the ones we need. */ - size_t const bitContainer = bitC->bitContainer[0] >> (HUF_BITS_IN_CONTAINER - nbBits); - /* Mask bitPos to account for the bytes we consumed. */ - bitC->bitPos[0] &= 7; - assert(nbBits > 0); - assert(nbBits <= sizeof(bitC->bitContainer[0]) * 8); - assert(bitC->ptr <= bitC->endPtr); - MEM_writeLEST(bitC->ptr, bitContainer); - bitC->ptr += nbBytes; - assert(!kFast || bitC->ptr <= bitC->endPtr); - if (!kFast && bitC->ptr > bitC->endPtr) bitC->ptr = bitC->endPtr; - /* bitContainer doesn't need to be modified because the leftover - * bits are already the top bitPos bits. And we don't care about - * noise in the lower values. - */ -} - -/*! HUF_endMark() - * @returns The Huffman stream end mark: A 1-bit value = 1. - */ -static HUF_CElt HUF_endMark(void) -{ - HUF_CElt endMark; - HUF_setNbBits(&endMark, 1); - HUF_setValue(&endMark, 1); - return endMark; -} - -/*! HUF_closeCStream() : - * @return Size of CStream, in bytes, - * or 0 if it could not fit into dstBuffer */ -static size_t HUF_closeCStream(HUF_CStream_t* bitC) -{ - HUF_addBits(bitC, HUF_endMark(), /* idx */ 0, /* kFast */ 0); - HUF_flushBits(bitC, /* kFast */ 0); - { - size_t const nbBits = bitC->bitPos[0] & 0xFF; - if (bitC->ptr >= bitC->endPtr) return 0; /* overflow detected */ - return (bitC->ptr - bitC->startPtr) + (nbBits > 0); - } -} - -FORCE_INLINE_TEMPLATE void -HUF_encodeSymbol(HUF_CStream_t* bitCPtr, U32 symbol, const HUF_CElt* CTable, int idx, int fast) -{ - HUF_addBits(bitCPtr, CTable[symbol], idx, fast); -} - -FORCE_INLINE_TEMPLATE void -HUF_compress1X_usingCTable_internal_body_loop(HUF_CStream_t* bitC, - const BYTE* ip, size_t srcSize, - const HUF_CElt* ct, - int kUnroll, int kFastFlush, int kLastFast) -{ - /* Join to kUnroll */ - int n = (int)srcSize; - int rem = n % kUnroll; - if (rem > 0) { - for (; rem > 0; --rem) { - HUF_encodeSymbol(bitC, ip[--n], ct, 0, /* fast */ 0); - } - HUF_flushBits(bitC, kFastFlush); - } - assert(n % kUnroll == 0); - - /* Join to 2 * kUnroll */ - if (n % (2 * kUnroll)) { - int u; - for (u = 1; u < kUnroll; ++u) { - HUF_encodeSymbol(bitC, ip[n - u], ct, 0, 1); - } - HUF_encodeSymbol(bitC, ip[n - kUnroll], ct, 0, kLastFast); - HUF_flushBits(bitC, kFastFlush); - n -= kUnroll; - } - assert(n % (2 * kUnroll) == 0); - - for (; n>0; n-= 2 * kUnroll) { - /* Encode kUnroll symbols into the bitstream @ index 0. */ - int u; - for (u = 1; u < kUnroll; ++u) { - HUF_encodeSymbol(bitC, ip[n - u], ct, /* idx */ 0, /* fast */ 1); - } - HUF_encodeSymbol(bitC, ip[n - kUnroll], ct, /* idx */ 0, /* fast */ kLastFast); - HUF_flushBits(bitC, kFastFlush); - /* Encode kUnroll symbols into the bitstream @ index 1. - * This allows us to start filling the bit container - * without any data dependencies. - */ - HUF_zeroIndex1(bitC); - for (u = 1; u < kUnroll; ++u) { - HUF_encodeSymbol(bitC, ip[n - kUnroll - u], ct, /* idx */ 1, /* fast */ 1); - } - HUF_encodeSymbol(bitC, ip[n - kUnroll - kUnroll], ct, /* idx */ 1, /* fast */ kLastFast); - /* Merge bitstream @ index 1 into the bitstream @ index 0 */ - HUF_mergeIndex1(bitC); - HUF_flushBits(bitC, kFastFlush); - } - assert(n == 0); - -} - -/** - * Returns a tight upper bound on the output space needed by Huffman - * with 8 bytes buffer to handle over-writes. If the output is at least - * this large we don't need to do bounds checks during Huffman encoding. - */ -static size_t HUF_tightCompressBound(size_t srcSize, size_t tableLog) -{ - return ((srcSize * tableLog) >> 3) + 8; -} - - -FORCE_INLINE_TEMPLATE size_t -HUF_compress1X_usingCTable_internal_body(void* dst, size_t dstSize, - const void* src, size_t srcSize, - const HUF_CElt* CTable) -{ - U32 const tableLog = (U32)CTable[0]; - HUF_CElt const* ct = CTable + 1; - const BYTE* ip = (const BYTE*) src; - BYTE* const ostart = (BYTE*)dst; - BYTE* const oend = ostart + dstSize; - BYTE* op = ostart; - HUF_CStream_t bitC; - - /* init */ - if (dstSize < 8) return 0; /* not enough space to compress */ - { size_t const initErr = HUF_initCStream(&bitC, op, (size_t)(oend-op)); - if (HUF_isError(initErr)) return 0; } - - if (dstSize < HUF_tightCompressBound(srcSize, (size_t)tableLog) || tableLog > 11) - HUF_compress1X_usingCTable_internal_body_loop(&bitC, ip, srcSize, ct, /* kUnroll */ MEM_32bits() ? 2 : 4, /* kFast */ 0, /* kLastFast */ 0); - else { - if (MEM_32bits()) { - switch (tableLog) { - case 11: - HUF_compress1X_usingCTable_internal_body_loop(&bitC, ip, srcSize, ct, /* kUnroll */ 2, /* kFastFlush */ 1, /* kLastFast */ 0); - break; - case 10: ZSTD_FALLTHROUGH; - case 9: ZSTD_FALLTHROUGH; - case 8: - HUF_compress1X_usingCTable_internal_body_loop(&bitC, ip, srcSize, ct, /* kUnroll */ 2, /* kFastFlush */ 1, /* kLastFast */ 1); - break; - case 7: ZSTD_FALLTHROUGH; - default: - HUF_compress1X_usingCTable_internal_body_loop(&bitC, ip, srcSize, ct, /* kUnroll */ 3, /* kFastFlush */ 1, /* kLastFast */ 1); - break; - } - } else { - switch (tableLog) { - case 11: - HUF_compress1X_usingCTable_internal_body_loop(&bitC, ip, srcSize, ct, /* kUnroll */ 5, /* kFastFlush */ 1, /* kLastFast */ 0); - break; - case 10: - HUF_compress1X_usingCTable_internal_body_loop(&bitC, ip, srcSize, ct, /* kUnroll */ 5, /* kFastFlush */ 1, /* kLastFast */ 1); - break; - case 9: - HUF_compress1X_usingCTable_internal_body_loop(&bitC, ip, srcSize, ct, /* kUnroll */ 6, /* kFastFlush */ 1, /* kLastFast */ 0); - break; - case 8: - HUF_compress1X_usingCTable_internal_body_loop(&bitC, ip, srcSize, ct, /* kUnroll */ 7, /* kFastFlush */ 1, /* kLastFast */ 0); - break; - case 7: - HUF_compress1X_usingCTable_internal_body_loop(&bitC, ip, srcSize, ct, /* kUnroll */ 8, /* kFastFlush */ 1, /* kLastFast */ 0); - break; - case 6: ZSTD_FALLTHROUGH; - default: - HUF_compress1X_usingCTable_internal_body_loop(&bitC, ip, srcSize, ct, /* kUnroll */ 9, /* kFastFlush */ 1, /* kLastFast */ 1); - break; - } - } - } - assert(bitC.ptr <= bitC.endPtr); - - return HUF_closeCStream(&bitC); -} - -#if DYNAMIC_BMI2 - -static BMI2_TARGET_ATTRIBUTE size_t -HUF_compress1X_usingCTable_internal_bmi2(void* dst, size_t dstSize, - const void* src, size_t srcSize, - const HUF_CElt* CTable) -{ - return HUF_compress1X_usingCTable_internal_body(dst, dstSize, src, srcSize, CTable); -} - -static size_t -HUF_compress1X_usingCTable_internal_default(void* dst, size_t dstSize, - const void* src, size_t srcSize, - const HUF_CElt* CTable) -{ - return HUF_compress1X_usingCTable_internal_body(dst, dstSize, src, srcSize, CTable); -} - -static size_t -HUF_compress1X_usingCTable_internal(void* dst, size_t dstSize, - const void* src, size_t srcSize, - const HUF_CElt* CTable, const int bmi2) -{ - if (bmi2) { - return HUF_compress1X_usingCTable_internal_bmi2(dst, dstSize, src, srcSize, CTable); - } - return HUF_compress1X_usingCTable_internal_default(dst, dstSize, src, srcSize, CTable); -} - -#else - -static size_t -HUF_compress1X_usingCTable_internal(void* dst, size_t dstSize, - const void* src, size_t srcSize, - const HUF_CElt* CTable, const int bmi2) -{ - (void)bmi2; - return HUF_compress1X_usingCTable_internal_body(dst, dstSize, src, srcSize, CTable); -} - -#endif - -size_t HUF_compress1X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable) -{ - return HUF_compress1X_usingCTable_bmi2(dst, dstSize, src, srcSize, CTable, /* bmi2 */ 0); -} - -size_t HUF_compress1X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2) -{ - return HUF_compress1X_usingCTable_internal(dst, dstSize, src, srcSize, CTable, bmi2); -} - -static size_t -HUF_compress4X_usingCTable_internal(void* dst, size_t dstSize, - const void* src, size_t srcSize, - const HUF_CElt* CTable, int bmi2) -{ - size_t const segmentSize = (srcSize+3)/4; /* first 3 segments */ - const BYTE* ip = (const BYTE*) src; - const BYTE* const iend = ip + srcSize; - BYTE* const ostart = (BYTE*) dst; - BYTE* const oend = ostart + dstSize; - BYTE* op = ostart; - - if (dstSize < 6 + 1 + 1 + 1 + 8) return 0; /* minimum space to compress successfully */ - if (srcSize < 12) return 0; /* no saving possible : too small input */ - op += 6; /* jumpTable */ - - assert(op <= oend); - { CHECK_V_F(cSize, HUF_compress1X_usingCTable_internal(op, (size_t)(oend-op), ip, segmentSize, CTable, bmi2) ); - if (cSize == 0 || cSize > 65535) return 0; - MEM_writeLE16(ostart, (U16)cSize); - op += cSize; - } - - ip += segmentSize; - assert(op <= oend); - { CHECK_V_F(cSize, HUF_compress1X_usingCTable_internal(op, (size_t)(oend-op), ip, segmentSize, CTable, bmi2) ); - if (cSize == 0 || cSize > 65535) return 0; - MEM_writeLE16(ostart+2, (U16)cSize); - op += cSize; - } - - ip += segmentSize; - assert(op <= oend); - { CHECK_V_F(cSize, HUF_compress1X_usingCTable_internal(op, (size_t)(oend-op), ip, segmentSize, CTable, bmi2) ); - if (cSize == 0 || cSize > 65535) return 0; - MEM_writeLE16(ostart+4, (U16)cSize); - op += cSize; - } - - ip += segmentSize; - assert(op <= oend); - assert(ip <= iend); - { CHECK_V_F(cSize, HUF_compress1X_usingCTable_internal(op, (size_t)(oend-op), ip, (size_t)(iend-ip), CTable, bmi2) ); - if (cSize == 0 || cSize > 65535) return 0; - op += cSize; - } - - return (size_t)(op-ostart); -} - -size_t HUF_compress4X_usingCTable(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable) -{ - return HUF_compress4X_usingCTable_bmi2(dst, dstSize, src, srcSize, CTable, /* bmi2 */ 0); -} - -size_t HUF_compress4X_usingCTable_bmi2(void* dst, size_t dstSize, const void* src, size_t srcSize, const HUF_CElt* CTable, int bmi2) -{ - return HUF_compress4X_usingCTable_internal(dst, dstSize, src, srcSize, CTable, bmi2); -} - -typedef enum { HUF_singleStream, HUF_fourStreams } HUF_nbStreams_e; - -static size_t HUF_compressCTable_internal( - BYTE* const ostart, BYTE* op, BYTE* const oend, - const void* src, size_t srcSize, - HUF_nbStreams_e nbStreams, const HUF_CElt* CTable, const int bmi2) -{ - size_t const cSize = (nbStreams==HUF_singleStream) ? - HUF_compress1X_usingCTable_internal(op, (size_t)(oend - op), src, srcSize, CTable, bmi2) : - HUF_compress4X_usingCTable_internal(op, (size_t)(oend - op), src, srcSize, CTable, bmi2); - if (HUF_isError(cSize)) { return cSize; } - if (cSize==0) { return 0; } /* uncompressible */ - op += cSize; - /* check compressibility */ - assert(op >= ostart); - if ((size_t)(op-ostart) >= srcSize-1) { return 0; } - return (size_t)(op-ostart); -} - -typedef struct { - unsigned count[HUF_SYMBOLVALUE_MAX + 1]; - HUF_CElt CTable[HUF_CTABLE_SIZE_ST(HUF_SYMBOLVALUE_MAX)]; - union { - HUF_buildCTable_wksp_tables buildCTable_wksp; - HUF_WriteCTableWksp writeCTable_wksp; - U32 hist_wksp[HIST_WKSP_SIZE_U32]; - } wksps; -} HUF_compress_tables_t; - -#define SUSPECT_INCOMPRESSIBLE_SAMPLE_SIZE 4096 -#define SUSPECT_INCOMPRESSIBLE_SAMPLE_RATIO 10 /* Must be >= 2 */ - -/* HUF_compress_internal() : - * `workSpace_align4` must be aligned on 4-bytes boundaries, - * and occupies the same space as a table of HUF_WORKSPACE_SIZE_U64 unsigned */ -static size_t -HUF_compress_internal (void* dst, size_t dstSize, - const void* src, size_t srcSize, - unsigned maxSymbolValue, unsigned huffLog, - HUF_nbStreams_e nbStreams, - void* workSpace, size_t wkspSize, - HUF_CElt* oldHufTable, HUF_repeat* repeat, int preferRepeat, - const int bmi2, unsigned suspectUncompressible) -{ - HUF_compress_tables_t* const table = (HUF_compress_tables_t*)HUF_alignUpWorkspace(workSpace, &wkspSize, ZSTD_ALIGNOF(size_t)); - BYTE* const ostart = (BYTE*)dst; - BYTE* const oend = ostart + dstSize; - BYTE* op = ostart; - - HUF_STATIC_ASSERT(sizeof(*table) + HUF_WORKSPACE_MAX_ALIGNMENT <= HUF_WORKSPACE_SIZE); - - /* checks & inits */ - if (wkspSize < sizeof(*table)) return ERROR(workSpace_tooSmall); - if (!srcSize) return 0; /* Uncompressed */ - if (!dstSize) return 0; /* cannot fit anything within dst budget */ - if (srcSize > HUF_BLOCKSIZE_MAX) return ERROR(srcSize_wrong); /* current block size limit */ - if (huffLog > HUF_TABLELOG_MAX) return ERROR(tableLog_tooLarge); - if (maxSymbolValue > HUF_SYMBOLVALUE_MAX) return ERROR(maxSymbolValue_tooLarge); - if (!maxSymbolValue) maxSymbolValue = HUF_SYMBOLVALUE_MAX; - if (!huffLog) huffLog = HUF_TABLELOG_DEFAULT; - - /* Heuristic : If old table is valid, use it for small inputs */ - if (preferRepeat && repeat && *repeat == HUF_repeat_valid) { - return HUF_compressCTable_internal(ostart, op, oend, - src, srcSize, - nbStreams, oldHufTable, bmi2); - } - - /* If uncompressible data is suspected, do a smaller sampling first */ - DEBUG_STATIC_ASSERT(SUSPECT_INCOMPRESSIBLE_SAMPLE_RATIO >= 2); - if (suspectUncompressible && srcSize >= (SUSPECT_INCOMPRESSIBLE_SAMPLE_SIZE * SUSPECT_INCOMPRESSIBLE_SAMPLE_RATIO)) { - size_t largestTotal = 0; - { unsigned maxSymbolValueBegin = maxSymbolValue; - CHECK_V_F(largestBegin, HIST_count_simple (table->count, &maxSymbolValueBegin, (const BYTE*)src, SUSPECT_INCOMPRESSIBLE_SAMPLE_SIZE) ); - largestTotal += largestBegin; - } - { unsigned maxSymbolValueEnd = maxSymbolValue; - CHECK_V_F(largestEnd, HIST_count_simple (table->count, &maxSymbolValueEnd, (const BYTE*)src + srcSize - SUSPECT_INCOMPRESSIBLE_SAMPLE_SIZE, SUSPECT_INCOMPRESSIBLE_SAMPLE_SIZE) ); - largestTotal += largestEnd; - } - if (largestTotal <= ((2 * SUSPECT_INCOMPRESSIBLE_SAMPLE_SIZE) >> 7)+4) return 0; /* heuristic : probably not compressible enough */ - } - - /* Scan input and build symbol stats */ - { CHECK_V_F(largest, HIST_count_wksp (table->count, &maxSymbolValue, (const BYTE*)src, srcSize, table->wksps.hist_wksp, sizeof(table->wksps.hist_wksp)) ); - if (largest == srcSize) { *ostart = ((const BYTE*)src)[0]; return 1; } /* single symbol, rle */ - if (largest <= (srcSize >> 7)+4) return 0; /* heuristic : probably not compressible enough */ - } - - /* Check validity of previous table */ - if ( repeat - && *repeat == HUF_repeat_check - && !HUF_validateCTable(oldHufTable, table->count, maxSymbolValue)) { - *repeat = HUF_repeat_none; - } - /* Heuristic : use existing table for small inputs */ - if (preferRepeat && repeat && *repeat != HUF_repeat_none) { - return HUF_compressCTable_internal(ostart, op, oend, - src, srcSize, - nbStreams, oldHufTable, bmi2); - } - - /* Build Huffman Tree */ - huffLog = HUF_optimalTableLog(huffLog, srcSize, maxSymbolValue); - { size_t const maxBits = HUF_buildCTable_wksp(table->CTable, table->count, - maxSymbolValue, huffLog, - &table->wksps.buildCTable_wksp, sizeof(table->wksps.buildCTable_wksp)); - CHECK_F(maxBits); - huffLog = (U32)maxBits; - } - /* Zero unused symbols in CTable, so we can check it for validity */ - { - size_t const ctableSize = HUF_CTABLE_SIZE_ST(maxSymbolValue); - size_t const unusedSize = sizeof(table->CTable) - ctableSize * sizeof(HUF_CElt); - ZSTD_memset(table->CTable + ctableSize, 0, unusedSize); - } - - /* Write table description header */ - { CHECK_V_F(hSize, HUF_writeCTable_wksp(op, dstSize, table->CTable, maxSymbolValue, huffLog, - &table->wksps.writeCTable_wksp, sizeof(table->wksps.writeCTable_wksp)) ); - /* Check if using previous huffman table is beneficial */ - if (repeat && *repeat != HUF_repeat_none) { - size_t const oldSize = HUF_estimateCompressedSize(oldHufTable, table->count, maxSymbolValue); - size_t const newSize = HUF_estimateCompressedSize(table->CTable, table->count, maxSymbolValue); - if (oldSize <= hSize + newSize || hSize + 12 >= srcSize) { - return HUF_compressCTable_internal(ostart, op, oend, - src, srcSize, - nbStreams, oldHufTable, bmi2); - } } - - /* Use the new huffman table */ - if (hSize + 12ul >= srcSize) { return 0; } - op += hSize; - if (repeat) { *repeat = HUF_repeat_none; } - if (oldHufTable) - ZSTD_memcpy(oldHufTable, table->CTable, sizeof(table->CTable)); /* Save new table */ - } - return HUF_compressCTable_internal(ostart, op, oend, - src, srcSize, - nbStreams, table->CTable, bmi2); -} - - -size_t HUF_compress1X_wksp (void* dst, size_t dstSize, - const void* src, size_t srcSize, - unsigned maxSymbolValue, unsigned huffLog, - void* workSpace, size_t wkspSize) -{ - return HUF_compress_internal(dst, dstSize, src, srcSize, - maxSymbolValue, huffLog, HUF_singleStream, - workSpace, wkspSize, - NULL, NULL, 0, 0 /*bmi2*/, 0); -} - -size_t HUF_compress1X_repeat (void* dst, size_t dstSize, - const void* src, size_t srcSize, - unsigned maxSymbolValue, unsigned huffLog, - void* workSpace, size_t wkspSize, - HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, - int bmi2, unsigned suspectUncompressible) -{ - return HUF_compress_internal(dst, dstSize, src, srcSize, - maxSymbolValue, huffLog, HUF_singleStream, - workSpace, wkspSize, hufTable, - repeat, preferRepeat, bmi2, suspectUncompressible); -} - -/* HUF_compress4X_repeat(): - * compress input using 4 streams. - * provide workspace to generate compression tables */ -size_t HUF_compress4X_wksp (void* dst, size_t dstSize, - const void* src, size_t srcSize, - unsigned maxSymbolValue, unsigned huffLog, - void* workSpace, size_t wkspSize) -{ - return HUF_compress_internal(dst, dstSize, src, srcSize, - maxSymbolValue, huffLog, HUF_fourStreams, - workSpace, wkspSize, - NULL, NULL, 0, 0 /*bmi2*/, 0); -} - -/* HUF_compress4X_repeat(): - * compress input using 4 streams. - * consider skipping quickly - * re-use an existing huffman compression table */ -size_t HUF_compress4X_repeat (void* dst, size_t dstSize, - const void* src, size_t srcSize, - unsigned maxSymbolValue, unsigned huffLog, - void* workSpace, size_t wkspSize, - HUF_CElt* hufTable, HUF_repeat* repeat, int preferRepeat, int bmi2, unsigned suspectUncompressible) -{ - return HUF_compress_internal(dst, dstSize, src, srcSize, - maxSymbolValue, huffLog, HUF_fourStreams, - workSpace, wkspSize, - hufTable, repeat, preferRepeat, bmi2, suspectUncompressible); -} - -#ifndef ZSTD_NO_UNUSED_FUNCTIONS -/** HUF_buildCTable() : - * @return : maxNbBits - * Note : count is used before tree is written, so they can safely overlap - */ -size_t HUF_buildCTable (HUF_CElt* tree, const unsigned* count, unsigned maxSymbolValue, unsigned maxNbBits) -{ - HUF_buildCTable_wksp_tables workspace; - return HUF_buildCTable_wksp(tree, count, maxSymbolValue, maxNbBits, &workspace, sizeof(workspace)); -} - -size_t HUF_compress1X (void* dst, size_t dstSize, - const void* src, size_t srcSize, - unsigned maxSymbolValue, unsigned huffLog) -{ - U64 workSpace[HUF_WORKSPACE_SIZE_U64]; - return HUF_compress1X_wksp(dst, dstSize, src, srcSize, maxSymbolValue, huffLog, workSpace, sizeof(workSpace)); -} - -size_t HUF_compress2 (void* dst, size_t dstSize, - const void* src, size_t srcSize, - unsigned maxSymbolValue, unsigned huffLog) -{ - U64 workSpace[HUF_WORKSPACE_SIZE_U64]; - return HUF_compress4X_wksp(dst, dstSize, src, srcSize, maxSymbolValue, huffLog, workSpace, sizeof(workSpace)); -} - -size_t HUF_compress (void* dst, size_t maxDstSize, const void* src, size_t srcSize) -{ - return HUF_compress2(dst, maxDstSize, src, srcSize, 255, HUF_TABLELOG_DEFAULT); -} -#endif diff --git a/dep/zstd/lib/compress/zstd_compress.c b/dep/zstd/lib/compress/zstd_compress.c deleted file mode 100644 index f06456af9..000000000 --- a/dep/zstd/lib/compress/zstd_compress.c +++ /dev/null @@ -1,6327 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -/*-************************************* -* Dependencies -***************************************/ -#include "../common/zstd_deps.h" /* INT_MAX, ZSTD_memset, ZSTD_memcpy */ -#include "../common/mem.h" -#include "hist.h" /* HIST_countFast_wksp */ -#define FSE_STATIC_LINKING_ONLY /* FSE_encodeSymbol */ -#include "../common/fse.h" -#define HUF_STATIC_LINKING_ONLY -#include "../common/huf.h" -#include "zstd_compress_internal.h" -#include "zstd_compress_sequences.h" -#include "zstd_compress_literals.h" -#include "zstd_fast.h" -#include "zstd_double_fast.h" -#include "zstd_lazy.h" -#include "zstd_opt.h" -#include "zstd_ldm.h" -#include "zstd_compress_superblock.h" - -/* *************************************************************** -* Tuning parameters -*****************************************************************/ -/*! - * COMPRESS_HEAPMODE : - * Select how default decompression function ZSTD_compress() allocates its context, - * on stack (0, default), or into heap (1). - * Note that functions with explicit context such as ZSTD_compressCCtx() are unaffected. - */ -#ifndef ZSTD_COMPRESS_HEAPMODE -# define ZSTD_COMPRESS_HEAPMODE 0 -#endif - -/*! - * ZSTD_HASHLOG3_MAX : - * Maximum size of the hash table dedicated to find 3-bytes matches, - * in log format, aka 17 => 1 << 17 == 128Ki positions. - * This structure is only used in zstd_opt. - * Since allocation is centralized for all strategies, it has to be known here. - * The actual (selected) size of the hash table is then stored in ZSTD_matchState_t.hashLog3, - * so that zstd_opt.c doesn't need to know about this constant. - */ -#ifndef ZSTD_HASHLOG3_MAX -# define ZSTD_HASHLOG3_MAX 17 -#endif - -/*-************************************* -* Helper functions -***************************************/ -/* ZSTD_compressBound() - * Note that the result from this function is only compatible with the "normal" - * full-block strategy. - * When there are a lot of small blocks due to frequent flush in streaming mode - * the overhead of headers can make the compressed data to be larger than the - * return value of ZSTD_compressBound(). - */ -size_t ZSTD_compressBound(size_t srcSize) { - return ZSTD_COMPRESSBOUND(srcSize); -} - - -/*-************************************* -* Context memory management -***************************************/ -struct ZSTD_CDict_s { - const void* dictContent; - size_t dictContentSize; - ZSTD_dictContentType_e dictContentType; /* The dictContentType the CDict was created with */ - U32* entropyWorkspace; /* entropy workspace of HUF_WORKSPACE_SIZE bytes */ - ZSTD_cwksp workspace; - ZSTD_matchState_t matchState; - ZSTD_compressedBlockState_t cBlockState; - ZSTD_customMem customMem; - U32 dictID; - int compressionLevel; /* 0 indicates that advanced API was used to select CDict params */ - ZSTD_paramSwitch_e useRowMatchFinder; /* Indicates whether the CDict was created with params that would use - * row-based matchfinder. Unless the cdict is reloaded, we will use - * the same greedy/lazy matchfinder at compression time. - */ -}; /* typedef'd to ZSTD_CDict within "zstd.h" */ - -ZSTD_CCtx* ZSTD_createCCtx(void) -{ - return ZSTD_createCCtx_advanced(ZSTD_defaultCMem); -} - -static void ZSTD_initCCtx(ZSTD_CCtx* cctx, ZSTD_customMem memManager) -{ - assert(cctx != NULL); - ZSTD_memset(cctx, 0, sizeof(*cctx)); - cctx->customMem = memManager; - cctx->bmi2 = ZSTD_cpuSupportsBmi2(); - { size_t const err = ZSTD_CCtx_reset(cctx, ZSTD_reset_parameters); - assert(!ZSTD_isError(err)); - (void)err; - } -} - -ZSTD_CCtx* ZSTD_createCCtx_advanced(ZSTD_customMem customMem) -{ - ZSTD_STATIC_ASSERT(zcss_init==0); - ZSTD_STATIC_ASSERT(ZSTD_CONTENTSIZE_UNKNOWN==(0ULL - 1)); - if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; - { ZSTD_CCtx* const cctx = (ZSTD_CCtx*)ZSTD_customMalloc(sizeof(ZSTD_CCtx), customMem); - if (!cctx) return NULL; - ZSTD_initCCtx(cctx, customMem); - return cctx; - } -} - -ZSTD_CCtx* ZSTD_initStaticCCtx(void* workspace, size_t workspaceSize) -{ - ZSTD_cwksp ws; - ZSTD_CCtx* cctx; - if (workspaceSize <= sizeof(ZSTD_CCtx)) return NULL; /* minimum size */ - if ((size_t)workspace & 7) return NULL; /* must be 8-aligned */ - ZSTD_cwksp_init(&ws, workspace, workspaceSize, ZSTD_cwksp_static_alloc); - - cctx = (ZSTD_CCtx*)ZSTD_cwksp_reserve_object(&ws, sizeof(ZSTD_CCtx)); - if (cctx == NULL) return NULL; - - ZSTD_memset(cctx, 0, sizeof(ZSTD_CCtx)); - ZSTD_cwksp_move(&cctx->workspace, &ws); - cctx->staticSize = workspaceSize; - - /* statically sized space. entropyWorkspace never moves (but prev/next block swap places) */ - if (!ZSTD_cwksp_check_available(&cctx->workspace, ENTROPY_WORKSPACE_SIZE + 2 * sizeof(ZSTD_compressedBlockState_t))) return NULL; - cctx->blockState.prevCBlock = (ZSTD_compressedBlockState_t*)ZSTD_cwksp_reserve_object(&cctx->workspace, sizeof(ZSTD_compressedBlockState_t)); - cctx->blockState.nextCBlock = (ZSTD_compressedBlockState_t*)ZSTD_cwksp_reserve_object(&cctx->workspace, sizeof(ZSTD_compressedBlockState_t)); - cctx->entropyWorkspace = (U32*)ZSTD_cwksp_reserve_object(&cctx->workspace, ENTROPY_WORKSPACE_SIZE); - cctx->bmi2 = ZSTD_cpuid_bmi2(ZSTD_cpuid()); - return cctx; -} - -/** - * Clears and frees all of the dictionaries in the CCtx. - */ -static void ZSTD_clearAllDicts(ZSTD_CCtx* cctx) -{ - ZSTD_customFree(cctx->localDict.dictBuffer, cctx->customMem); - ZSTD_freeCDict(cctx->localDict.cdict); - ZSTD_memset(&cctx->localDict, 0, sizeof(cctx->localDict)); - ZSTD_memset(&cctx->prefixDict, 0, sizeof(cctx->prefixDict)); - cctx->cdict = NULL; -} - -static size_t ZSTD_sizeof_localDict(ZSTD_localDict dict) -{ - size_t const bufferSize = dict.dictBuffer != NULL ? dict.dictSize : 0; - size_t const cdictSize = ZSTD_sizeof_CDict(dict.cdict); - return bufferSize + cdictSize; -} - -static void ZSTD_freeCCtxContent(ZSTD_CCtx* cctx) -{ - assert(cctx != NULL); - assert(cctx->staticSize == 0); - ZSTD_clearAllDicts(cctx); -#ifdef ZSTD_MULTITHREAD - ZSTDMT_freeCCtx(cctx->mtctx); cctx->mtctx = NULL; -#endif - ZSTD_cwksp_free(&cctx->workspace, cctx->customMem); -} - -size_t ZSTD_freeCCtx(ZSTD_CCtx* cctx) -{ - if (cctx==NULL) return 0; /* support free on NULL */ - RETURN_ERROR_IF(cctx->staticSize, memory_allocation, - "not compatible with static CCtx"); - { - int cctxInWorkspace = ZSTD_cwksp_owns_buffer(&cctx->workspace, cctx); - ZSTD_freeCCtxContent(cctx); - if (!cctxInWorkspace) { - ZSTD_customFree(cctx, cctx->customMem); - } - } - return 0; -} - - -static size_t ZSTD_sizeof_mtctx(const ZSTD_CCtx* cctx) -{ -#ifdef ZSTD_MULTITHREAD - return ZSTDMT_sizeof_CCtx(cctx->mtctx); -#else - (void)cctx; - return 0; -#endif -} - - -size_t ZSTD_sizeof_CCtx(const ZSTD_CCtx* cctx) -{ - if (cctx==NULL) return 0; /* support sizeof on NULL */ - /* cctx may be in the workspace */ - return (cctx->workspace.workspace == cctx ? 0 : sizeof(*cctx)) - + ZSTD_cwksp_sizeof(&cctx->workspace) - + ZSTD_sizeof_localDict(cctx->localDict) - + ZSTD_sizeof_mtctx(cctx); -} - -size_t ZSTD_sizeof_CStream(const ZSTD_CStream* zcs) -{ - return ZSTD_sizeof_CCtx(zcs); /* same object */ -} - -/* private API call, for dictBuilder only */ -const seqStore_t* ZSTD_getSeqStore(const ZSTD_CCtx* ctx) { return &(ctx->seqStore); } - -/* Returns true if the strategy supports using a row based matchfinder */ -static int ZSTD_rowMatchFinderSupported(const ZSTD_strategy strategy) { - return (strategy >= ZSTD_greedy && strategy <= ZSTD_lazy2); -} - -/* Returns true if the strategy and useRowMatchFinder mode indicate that we will use the row based matchfinder - * for this compression. - */ -static int ZSTD_rowMatchFinderUsed(const ZSTD_strategy strategy, const ZSTD_paramSwitch_e mode) { - assert(mode != ZSTD_ps_auto); - return ZSTD_rowMatchFinderSupported(strategy) && (mode == ZSTD_ps_enable); -} - -/* Returns row matchfinder usage given an initial mode and cParams */ -static ZSTD_paramSwitch_e ZSTD_resolveRowMatchFinderMode(ZSTD_paramSwitch_e mode, - const ZSTD_compressionParameters* const cParams) { -#if defined(ZSTD_ARCH_X86_SSE2) || defined(ZSTD_ARCH_ARM_NEON) - int const kHasSIMD128 = 1; -#else - int const kHasSIMD128 = 0; -#endif - if (mode != ZSTD_ps_auto) return mode; /* if requested enabled, but no SIMD, we still will use row matchfinder */ - mode = ZSTD_ps_disable; - if (!ZSTD_rowMatchFinderSupported(cParams->strategy)) return mode; - if (kHasSIMD128) { - if (cParams->windowLog > 14) mode = ZSTD_ps_enable; - } else { - if (cParams->windowLog > 17) mode = ZSTD_ps_enable; - } - return mode; -} - -/* Returns block splitter usage (generally speaking, when using slower/stronger compression modes) */ -static ZSTD_paramSwitch_e ZSTD_resolveBlockSplitterMode(ZSTD_paramSwitch_e mode, - const ZSTD_compressionParameters* const cParams) { - if (mode != ZSTD_ps_auto) return mode; - return (cParams->strategy >= ZSTD_btopt && cParams->windowLog >= 17) ? ZSTD_ps_enable : ZSTD_ps_disable; -} - -/* Returns 1 if the arguments indicate that we should allocate a chainTable, 0 otherwise */ -static int ZSTD_allocateChainTable(const ZSTD_strategy strategy, - const ZSTD_paramSwitch_e useRowMatchFinder, - const U32 forDDSDict) { - assert(useRowMatchFinder != ZSTD_ps_auto); - /* We always should allocate a chaintable if we are allocating a matchstate for a DDS dictionary matchstate. - * We do not allocate a chaintable if we are using ZSTD_fast, or are using the row-based matchfinder. - */ - return forDDSDict || ((strategy != ZSTD_fast) && !ZSTD_rowMatchFinderUsed(strategy, useRowMatchFinder)); -} - -/* Returns 1 if compression parameters are such that we should - * enable long distance matching (wlog >= 27, strategy >= btopt). - * Returns 0 otherwise. - */ -static ZSTD_paramSwitch_e ZSTD_resolveEnableLdm(ZSTD_paramSwitch_e mode, - const ZSTD_compressionParameters* const cParams) { - if (mode != ZSTD_ps_auto) return mode; - return (cParams->strategy >= ZSTD_btopt && cParams->windowLog >= 27) ? ZSTD_ps_enable : ZSTD_ps_disable; -} - -static ZSTD_CCtx_params ZSTD_makeCCtxParamsFromCParams( - ZSTD_compressionParameters cParams) -{ - ZSTD_CCtx_params cctxParams; - /* should not matter, as all cParams are presumed properly defined */ - ZSTD_CCtxParams_init(&cctxParams, ZSTD_CLEVEL_DEFAULT); - cctxParams.cParams = cParams; - - /* Adjust advanced params according to cParams */ - cctxParams.ldmParams.enableLdm = ZSTD_resolveEnableLdm(cctxParams.ldmParams.enableLdm, &cParams); - if (cctxParams.ldmParams.enableLdm == ZSTD_ps_enable) { - ZSTD_ldm_adjustParameters(&cctxParams.ldmParams, &cParams); - assert(cctxParams.ldmParams.hashLog >= cctxParams.ldmParams.bucketSizeLog); - assert(cctxParams.ldmParams.hashRateLog < 32); - } - cctxParams.useBlockSplitter = ZSTD_resolveBlockSplitterMode(cctxParams.useBlockSplitter, &cParams); - cctxParams.useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(cctxParams.useRowMatchFinder, &cParams); - assert(!ZSTD_checkCParams(cParams)); - return cctxParams; -} - -static ZSTD_CCtx_params* ZSTD_createCCtxParams_advanced( - ZSTD_customMem customMem) -{ - ZSTD_CCtx_params* params; - if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; - params = (ZSTD_CCtx_params*)ZSTD_customCalloc( - sizeof(ZSTD_CCtx_params), customMem); - if (!params) { return NULL; } - ZSTD_CCtxParams_init(params, ZSTD_CLEVEL_DEFAULT); - params->customMem = customMem; - return params; -} - -ZSTD_CCtx_params* ZSTD_createCCtxParams(void) -{ - return ZSTD_createCCtxParams_advanced(ZSTD_defaultCMem); -} - -size_t ZSTD_freeCCtxParams(ZSTD_CCtx_params* params) -{ - if (params == NULL) { return 0; } - ZSTD_customFree(params, params->customMem); - return 0; -} - -size_t ZSTD_CCtxParams_reset(ZSTD_CCtx_params* params) -{ - return ZSTD_CCtxParams_init(params, ZSTD_CLEVEL_DEFAULT); -} - -size_t ZSTD_CCtxParams_init(ZSTD_CCtx_params* cctxParams, int compressionLevel) { - RETURN_ERROR_IF(!cctxParams, GENERIC, "NULL pointer!"); - ZSTD_memset(cctxParams, 0, sizeof(*cctxParams)); - cctxParams->compressionLevel = compressionLevel; - cctxParams->fParams.contentSizeFlag = 1; - return 0; -} - -#define ZSTD_NO_CLEVEL 0 - -/** - * Initializes the cctxParams from params and compressionLevel. - * @param compressionLevel If params are derived from a compression level then that compression level, otherwise ZSTD_NO_CLEVEL. - */ -static void ZSTD_CCtxParams_init_internal(ZSTD_CCtx_params* cctxParams, ZSTD_parameters const* params, int compressionLevel) -{ - assert(!ZSTD_checkCParams(params->cParams)); - ZSTD_memset(cctxParams, 0, sizeof(*cctxParams)); - cctxParams->cParams = params->cParams; - cctxParams->fParams = params->fParams; - /* Should not matter, as all cParams are presumed properly defined. - * But, set it for tracing anyway. - */ - cctxParams->compressionLevel = compressionLevel; - cctxParams->useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(cctxParams->useRowMatchFinder, ¶ms->cParams); - cctxParams->useBlockSplitter = ZSTD_resolveBlockSplitterMode(cctxParams->useBlockSplitter, ¶ms->cParams); - cctxParams->ldmParams.enableLdm = ZSTD_resolveEnableLdm(cctxParams->ldmParams.enableLdm, ¶ms->cParams); - DEBUGLOG(4, "ZSTD_CCtxParams_init_internal: useRowMatchFinder=%d, useBlockSplitter=%d ldm=%d", - cctxParams->useRowMatchFinder, cctxParams->useBlockSplitter, cctxParams->ldmParams.enableLdm); -} - -size_t ZSTD_CCtxParams_init_advanced(ZSTD_CCtx_params* cctxParams, ZSTD_parameters params) -{ - RETURN_ERROR_IF(!cctxParams, GENERIC, "NULL pointer!"); - FORWARD_IF_ERROR( ZSTD_checkCParams(params.cParams) , ""); - ZSTD_CCtxParams_init_internal(cctxParams, ¶ms, ZSTD_NO_CLEVEL); - return 0; -} - -/** - * Sets cctxParams' cParams and fParams from params, but otherwise leaves them alone. - * @param param Validated zstd parameters. - */ -static void ZSTD_CCtxParams_setZstdParams( - ZSTD_CCtx_params* cctxParams, const ZSTD_parameters* params) -{ - assert(!ZSTD_checkCParams(params->cParams)); - cctxParams->cParams = params->cParams; - cctxParams->fParams = params->fParams; - /* Should not matter, as all cParams are presumed properly defined. - * But, set it for tracing anyway. - */ - cctxParams->compressionLevel = ZSTD_NO_CLEVEL; -} - -ZSTD_bounds ZSTD_cParam_getBounds(ZSTD_cParameter param) -{ - ZSTD_bounds bounds = { 0, 0, 0 }; - - switch(param) - { - case ZSTD_c_compressionLevel: - bounds.lowerBound = ZSTD_minCLevel(); - bounds.upperBound = ZSTD_maxCLevel(); - return bounds; - - case ZSTD_c_windowLog: - bounds.lowerBound = ZSTD_WINDOWLOG_MIN; - bounds.upperBound = ZSTD_WINDOWLOG_MAX; - return bounds; - - case ZSTD_c_hashLog: - bounds.lowerBound = ZSTD_HASHLOG_MIN; - bounds.upperBound = ZSTD_HASHLOG_MAX; - return bounds; - - case ZSTD_c_chainLog: - bounds.lowerBound = ZSTD_CHAINLOG_MIN; - bounds.upperBound = ZSTD_CHAINLOG_MAX; - return bounds; - - case ZSTD_c_searchLog: - bounds.lowerBound = ZSTD_SEARCHLOG_MIN; - bounds.upperBound = ZSTD_SEARCHLOG_MAX; - return bounds; - - case ZSTD_c_minMatch: - bounds.lowerBound = ZSTD_MINMATCH_MIN; - bounds.upperBound = ZSTD_MINMATCH_MAX; - return bounds; - - case ZSTD_c_targetLength: - bounds.lowerBound = ZSTD_TARGETLENGTH_MIN; - bounds.upperBound = ZSTD_TARGETLENGTH_MAX; - return bounds; - - case ZSTD_c_strategy: - bounds.lowerBound = ZSTD_STRATEGY_MIN; - bounds.upperBound = ZSTD_STRATEGY_MAX; - return bounds; - - case ZSTD_c_contentSizeFlag: - bounds.lowerBound = 0; - bounds.upperBound = 1; - return bounds; - - case ZSTD_c_checksumFlag: - bounds.lowerBound = 0; - bounds.upperBound = 1; - return bounds; - - case ZSTD_c_dictIDFlag: - bounds.lowerBound = 0; - bounds.upperBound = 1; - return bounds; - - case ZSTD_c_nbWorkers: - bounds.lowerBound = 0; -#ifdef ZSTD_MULTITHREAD - bounds.upperBound = ZSTDMT_NBWORKERS_MAX; -#else - bounds.upperBound = 0; -#endif - return bounds; - - case ZSTD_c_jobSize: - bounds.lowerBound = 0; -#ifdef ZSTD_MULTITHREAD - bounds.upperBound = ZSTDMT_JOBSIZE_MAX; -#else - bounds.upperBound = 0; -#endif - return bounds; - - case ZSTD_c_overlapLog: -#ifdef ZSTD_MULTITHREAD - bounds.lowerBound = ZSTD_OVERLAPLOG_MIN; - bounds.upperBound = ZSTD_OVERLAPLOG_MAX; -#else - bounds.lowerBound = 0; - bounds.upperBound = 0; -#endif - return bounds; - - case ZSTD_c_enableDedicatedDictSearch: - bounds.lowerBound = 0; - bounds.upperBound = 1; - return bounds; - - case ZSTD_c_enableLongDistanceMatching: - bounds.lowerBound = 0; - bounds.upperBound = 1; - return bounds; - - case ZSTD_c_ldmHashLog: - bounds.lowerBound = ZSTD_LDM_HASHLOG_MIN; - bounds.upperBound = ZSTD_LDM_HASHLOG_MAX; - return bounds; - - case ZSTD_c_ldmMinMatch: - bounds.lowerBound = ZSTD_LDM_MINMATCH_MIN; - bounds.upperBound = ZSTD_LDM_MINMATCH_MAX; - return bounds; - - case ZSTD_c_ldmBucketSizeLog: - bounds.lowerBound = ZSTD_LDM_BUCKETSIZELOG_MIN; - bounds.upperBound = ZSTD_LDM_BUCKETSIZELOG_MAX; - return bounds; - - case ZSTD_c_ldmHashRateLog: - bounds.lowerBound = ZSTD_LDM_HASHRATELOG_MIN; - bounds.upperBound = ZSTD_LDM_HASHRATELOG_MAX; - return bounds; - - /* experimental parameters */ - case ZSTD_c_rsyncable: - bounds.lowerBound = 0; - bounds.upperBound = 1; - return bounds; - - case ZSTD_c_forceMaxWindow : - bounds.lowerBound = 0; - bounds.upperBound = 1; - return bounds; - - case ZSTD_c_format: - ZSTD_STATIC_ASSERT(ZSTD_f_zstd1 < ZSTD_f_zstd1_magicless); - bounds.lowerBound = ZSTD_f_zstd1; - bounds.upperBound = ZSTD_f_zstd1_magicless; /* note : how to ensure at compile time that this is the highest value enum ? */ - return bounds; - - case ZSTD_c_forceAttachDict: - ZSTD_STATIC_ASSERT(ZSTD_dictDefaultAttach < ZSTD_dictForceLoad); - bounds.lowerBound = ZSTD_dictDefaultAttach; - bounds.upperBound = ZSTD_dictForceLoad; /* note : how to ensure at compile time that this is the highest value enum ? */ - return bounds; - - case ZSTD_c_literalCompressionMode: - ZSTD_STATIC_ASSERT(ZSTD_ps_auto < ZSTD_ps_enable && ZSTD_ps_enable < ZSTD_ps_disable); - bounds.lowerBound = (int)ZSTD_ps_auto; - bounds.upperBound = (int)ZSTD_ps_disable; - return bounds; - - case ZSTD_c_targetCBlockSize: - bounds.lowerBound = ZSTD_TARGETCBLOCKSIZE_MIN; - bounds.upperBound = ZSTD_TARGETCBLOCKSIZE_MAX; - return bounds; - - case ZSTD_c_srcSizeHint: - bounds.lowerBound = ZSTD_SRCSIZEHINT_MIN; - bounds.upperBound = ZSTD_SRCSIZEHINT_MAX; - return bounds; - - case ZSTD_c_stableInBuffer: - case ZSTD_c_stableOutBuffer: - bounds.lowerBound = (int)ZSTD_bm_buffered; - bounds.upperBound = (int)ZSTD_bm_stable; - return bounds; - - case ZSTD_c_blockDelimiters: - bounds.lowerBound = (int)ZSTD_sf_noBlockDelimiters; - bounds.upperBound = (int)ZSTD_sf_explicitBlockDelimiters; - return bounds; - - case ZSTD_c_validateSequences: - bounds.lowerBound = 0; - bounds.upperBound = 1; - return bounds; - - case ZSTD_c_useBlockSplitter: - bounds.lowerBound = (int)ZSTD_ps_auto; - bounds.upperBound = (int)ZSTD_ps_disable; - return bounds; - - case ZSTD_c_useRowMatchFinder: - bounds.lowerBound = (int)ZSTD_ps_auto; - bounds.upperBound = (int)ZSTD_ps_disable; - return bounds; - - case ZSTD_c_deterministicRefPrefix: - bounds.lowerBound = 0; - bounds.upperBound = 1; - return bounds; - - default: - bounds.error = ERROR(parameter_unsupported); - return bounds; - } -} - -/* ZSTD_cParam_clampBounds: - * Clamps the value into the bounded range. - */ -static size_t ZSTD_cParam_clampBounds(ZSTD_cParameter cParam, int* value) -{ - ZSTD_bounds const bounds = ZSTD_cParam_getBounds(cParam); - if (ZSTD_isError(bounds.error)) return bounds.error; - if (*value < bounds.lowerBound) *value = bounds.lowerBound; - if (*value > bounds.upperBound) *value = bounds.upperBound; - return 0; -} - -#define BOUNDCHECK(cParam, val) { \ - RETURN_ERROR_IF(!ZSTD_cParam_withinBounds(cParam,val), \ - parameter_outOfBound, "Param out of bounds"); \ -} - - -static int ZSTD_isUpdateAuthorized(ZSTD_cParameter param) -{ - switch(param) - { - case ZSTD_c_compressionLevel: - case ZSTD_c_hashLog: - case ZSTD_c_chainLog: - case ZSTD_c_searchLog: - case ZSTD_c_minMatch: - case ZSTD_c_targetLength: - case ZSTD_c_strategy: - return 1; - - case ZSTD_c_format: - case ZSTD_c_windowLog: - case ZSTD_c_contentSizeFlag: - case ZSTD_c_checksumFlag: - case ZSTD_c_dictIDFlag: - case ZSTD_c_forceMaxWindow : - case ZSTD_c_nbWorkers: - case ZSTD_c_jobSize: - case ZSTD_c_overlapLog: - case ZSTD_c_rsyncable: - case ZSTD_c_enableDedicatedDictSearch: - case ZSTD_c_enableLongDistanceMatching: - case ZSTD_c_ldmHashLog: - case ZSTD_c_ldmMinMatch: - case ZSTD_c_ldmBucketSizeLog: - case ZSTD_c_ldmHashRateLog: - case ZSTD_c_forceAttachDict: - case ZSTD_c_literalCompressionMode: - case ZSTD_c_targetCBlockSize: - case ZSTD_c_srcSizeHint: - case ZSTD_c_stableInBuffer: - case ZSTD_c_stableOutBuffer: - case ZSTD_c_blockDelimiters: - case ZSTD_c_validateSequences: - case ZSTD_c_useBlockSplitter: - case ZSTD_c_useRowMatchFinder: - case ZSTD_c_deterministicRefPrefix: - default: - return 0; - } -} - -size_t ZSTD_CCtx_setParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int value) -{ - DEBUGLOG(4, "ZSTD_CCtx_setParameter (%i, %i)", (int)param, value); - if (cctx->streamStage != zcss_init) { - if (ZSTD_isUpdateAuthorized(param)) { - cctx->cParamsChanged = 1; - } else { - RETURN_ERROR(stage_wrong, "can only set params in ctx init stage"); - } } - - switch(param) - { - case ZSTD_c_nbWorkers: - RETURN_ERROR_IF((value!=0) && cctx->staticSize, parameter_unsupported, - "MT not compatible with static alloc"); - break; - - case ZSTD_c_compressionLevel: - case ZSTD_c_windowLog: - case ZSTD_c_hashLog: - case ZSTD_c_chainLog: - case ZSTD_c_searchLog: - case ZSTD_c_minMatch: - case ZSTD_c_targetLength: - case ZSTD_c_strategy: - case ZSTD_c_ldmHashRateLog: - case ZSTD_c_format: - case ZSTD_c_contentSizeFlag: - case ZSTD_c_checksumFlag: - case ZSTD_c_dictIDFlag: - case ZSTD_c_forceMaxWindow: - case ZSTD_c_forceAttachDict: - case ZSTD_c_literalCompressionMode: - case ZSTD_c_jobSize: - case ZSTD_c_overlapLog: - case ZSTD_c_rsyncable: - case ZSTD_c_enableDedicatedDictSearch: - case ZSTD_c_enableLongDistanceMatching: - case ZSTD_c_ldmHashLog: - case ZSTD_c_ldmMinMatch: - case ZSTD_c_ldmBucketSizeLog: - case ZSTD_c_targetCBlockSize: - case ZSTD_c_srcSizeHint: - case ZSTD_c_stableInBuffer: - case ZSTD_c_stableOutBuffer: - case ZSTD_c_blockDelimiters: - case ZSTD_c_validateSequences: - case ZSTD_c_useBlockSplitter: - case ZSTD_c_useRowMatchFinder: - case ZSTD_c_deterministicRefPrefix: - break; - - default: RETURN_ERROR(parameter_unsupported, "unknown parameter"); - } - return ZSTD_CCtxParams_setParameter(&cctx->requestedParams, param, value); -} - -size_t ZSTD_CCtxParams_setParameter(ZSTD_CCtx_params* CCtxParams, - ZSTD_cParameter param, int value) -{ - DEBUGLOG(4, "ZSTD_CCtxParams_setParameter (%i, %i)", (int)param, value); - switch(param) - { - case ZSTD_c_format : - BOUNDCHECK(ZSTD_c_format, value); - CCtxParams->format = (ZSTD_format_e)value; - return (size_t)CCtxParams->format; - - case ZSTD_c_compressionLevel : { - FORWARD_IF_ERROR(ZSTD_cParam_clampBounds(param, &value), ""); - if (value == 0) - CCtxParams->compressionLevel = ZSTD_CLEVEL_DEFAULT; /* 0 == default */ - else - CCtxParams->compressionLevel = value; - if (CCtxParams->compressionLevel >= 0) return (size_t)CCtxParams->compressionLevel; - return 0; /* return type (size_t) cannot represent negative values */ - } - - case ZSTD_c_windowLog : - if (value!=0) /* 0 => use default */ - BOUNDCHECK(ZSTD_c_windowLog, value); - CCtxParams->cParams.windowLog = (U32)value; - return CCtxParams->cParams.windowLog; - - case ZSTD_c_hashLog : - if (value!=0) /* 0 => use default */ - BOUNDCHECK(ZSTD_c_hashLog, value); - CCtxParams->cParams.hashLog = (U32)value; - return CCtxParams->cParams.hashLog; - - case ZSTD_c_chainLog : - if (value!=0) /* 0 => use default */ - BOUNDCHECK(ZSTD_c_chainLog, value); - CCtxParams->cParams.chainLog = (U32)value; - return CCtxParams->cParams.chainLog; - - case ZSTD_c_searchLog : - if (value!=0) /* 0 => use default */ - BOUNDCHECK(ZSTD_c_searchLog, value); - CCtxParams->cParams.searchLog = (U32)value; - return (size_t)value; - - case ZSTD_c_minMatch : - if (value!=0) /* 0 => use default */ - BOUNDCHECK(ZSTD_c_minMatch, value); - CCtxParams->cParams.minMatch = value; - return CCtxParams->cParams.minMatch; - - case ZSTD_c_targetLength : - BOUNDCHECK(ZSTD_c_targetLength, value); - CCtxParams->cParams.targetLength = value; - return CCtxParams->cParams.targetLength; - - case ZSTD_c_strategy : - if (value!=0) /* 0 => use default */ - BOUNDCHECK(ZSTD_c_strategy, value); - CCtxParams->cParams.strategy = (ZSTD_strategy)value; - return (size_t)CCtxParams->cParams.strategy; - - case ZSTD_c_contentSizeFlag : - /* Content size written in frame header _when known_ (default:1) */ - DEBUGLOG(4, "set content size flag = %u", (value!=0)); - CCtxParams->fParams.contentSizeFlag = value != 0; - return CCtxParams->fParams.contentSizeFlag; - - case ZSTD_c_checksumFlag : - /* A 32-bits content checksum will be calculated and written at end of frame (default:0) */ - CCtxParams->fParams.checksumFlag = value != 0; - return CCtxParams->fParams.checksumFlag; - - case ZSTD_c_dictIDFlag : /* When applicable, dictionary's dictID is provided in frame header (default:1) */ - DEBUGLOG(4, "set dictIDFlag = %u", (value!=0)); - CCtxParams->fParams.noDictIDFlag = !value; - return !CCtxParams->fParams.noDictIDFlag; - - case ZSTD_c_forceMaxWindow : - CCtxParams->forceWindow = (value != 0); - return CCtxParams->forceWindow; - - case ZSTD_c_forceAttachDict : { - const ZSTD_dictAttachPref_e pref = (ZSTD_dictAttachPref_e)value; - BOUNDCHECK(ZSTD_c_forceAttachDict, pref); - CCtxParams->attachDictPref = pref; - return CCtxParams->attachDictPref; - } - - case ZSTD_c_literalCompressionMode : { - const ZSTD_paramSwitch_e lcm = (ZSTD_paramSwitch_e)value; - BOUNDCHECK(ZSTD_c_literalCompressionMode, lcm); - CCtxParams->literalCompressionMode = lcm; - return CCtxParams->literalCompressionMode; - } - - case ZSTD_c_nbWorkers : -#ifndef ZSTD_MULTITHREAD - RETURN_ERROR_IF(value!=0, parameter_unsupported, "not compiled with multithreading"); - return 0; -#else - FORWARD_IF_ERROR(ZSTD_cParam_clampBounds(param, &value), ""); - CCtxParams->nbWorkers = value; - return CCtxParams->nbWorkers; -#endif - - case ZSTD_c_jobSize : -#ifndef ZSTD_MULTITHREAD - RETURN_ERROR_IF(value!=0, parameter_unsupported, "not compiled with multithreading"); - return 0; -#else - /* Adjust to the minimum non-default value. */ - if (value != 0 && value < ZSTDMT_JOBSIZE_MIN) - value = ZSTDMT_JOBSIZE_MIN; - FORWARD_IF_ERROR(ZSTD_cParam_clampBounds(param, &value), ""); - assert(value >= 0); - CCtxParams->jobSize = value; - return CCtxParams->jobSize; -#endif - - case ZSTD_c_overlapLog : -#ifndef ZSTD_MULTITHREAD - RETURN_ERROR_IF(value!=0, parameter_unsupported, "not compiled with multithreading"); - return 0; -#else - FORWARD_IF_ERROR(ZSTD_cParam_clampBounds(ZSTD_c_overlapLog, &value), ""); - CCtxParams->overlapLog = value; - return CCtxParams->overlapLog; -#endif - - case ZSTD_c_rsyncable : -#ifndef ZSTD_MULTITHREAD - RETURN_ERROR_IF(value!=0, parameter_unsupported, "not compiled with multithreading"); - return 0; -#else - FORWARD_IF_ERROR(ZSTD_cParam_clampBounds(ZSTD_c_overlapLog, &value), ""); - CCtxParams->rsyncable = value; - return CCtxParams->rsyncable; -#endif - - case ZSTD_c_enableDedicatedDictSearch : - CCtxParams->enableDedicatedDictSearch = (value!=0); - return CCtxParams->enableDedicatedDictSearch; - - case ZSTD_c_enableLongDistanceMatching : - CCtxParams->ldmParams.enableLdm = (ZSTD_paramSwitch_e)value; - return CCtxParams->ldmParams.enableLdm; - - case ZSTD_c_ldmHashLog : - if (value!=0) /* 0 ==> auto */ - BOUNDCHECK(ZSTD_c_ldmHashLog, value); - CCtxParams->ldmParams.hashLog = value; - return CCtxParams->ldmParams.hashLog; - - case ZSTD_c_ldmMinMatch : - if (value!=0) /* 0 ==> default */ - BOUNDCHECK(ZSTD_c_ldmMinMatch, value); - CCtxParams->ldmParams.minMatchLength = value; - return CCtxParams->ldmParams.minMatchLength; - - case ZSTD_c_ldmBucketSizeLog : - if (value!=0) /* 0 ==> default */ - BOUNDCHECK(ZSTD_c_ldmBucketSizeLog, value); - CCtxParams->ldmParams.bucketSizeLog = value; - return CCtxParams->ldmParams.bucketSizeLog; - - case ZSTD_c_ldmHashRateLog : - if (value!=0) /* 0 ==> default */ - BOUNDCHECK(ZSTD_c_ldmHashRateLog, value); - CCtxParams->ldmParams.hashRateLog = value; - return CCtxParams->ldmParams.hashRateLog; - - case ZSTD_c_targetCBlockSize : - if (value!=0) /* 0 ==> default */ - BOUNDCHECK(ZSTD_c_targetCBlockSize, value); - CCtxParams->targetCBlockSize = value; - return CCtxParams->targetCBlockSize; - - case ZSTD_c_srcSizeHint : - if (value!=0) /* 0 ==> default */ - BOUNDCHECK(ZSTD_c_srcSizeHint, value); - CCtxParams->srcSizeHint = value; - return CCtxParams->srcSizeHint; - - case ZSTD_c_stableInBuffer: - BOUNDCHECK(ZSTD_c_stableInBuffer, value); - CCtxParams->inBufferMode = (ZSTD_bufferMode_e)value; - return CCtxParams->inBufferMode; - - case ZSTD_c_stableOutBuffer: - BOUNDCHECK(ZSTD_c_stableOutBuffer, value); - CCtxParams->outBufferMode = (ZSTD_bufferMode_e)value; - return CCtxParams->outBufferMode; - - case ZSTD_c_blockDelimiters: - BOUNDCHECK(ZSTD_c_blockDelimiters, value); - CCtxParams->blockDelimiters = (ZSTD_sequenceFormat_e)value; - return CCtxParams->blockDelimiters; - - case ZSTD_c_validateSequences: - BOUNDCHECK(ZSTD_c_validateSequences, value); - CCtxParams->validateSequences = value; - return CCtxParams->validateSequences; - - case ZSTD_c_useBlockSplitter: - BOUNDCHECK(ZSTD_c_useBlockSplitter, value); - CCtxParams->useBlockSplitter = (ZSTD_paramSwitch_e)value; - return CCtxParams->useBlockSplitter; - - case ZSTD_c_useRowMatchFinder: - BOUNDCHECK(ZSTD_c_useRowMatchFinder, value); - CCtxParams->useRowMatchFinder = (ZSTD_paramSwitch_e)value; - return CCtxParams->useRowMatchFinder; - - case ZSTD_c_deterministicRefPrefix: - BOUNDCHECK(ZSTD_c_deterministicRefPrefix, value); - CCtxParams->deterministicRefPrefix = !!value; - return CCtxParams->deterministicRefPrefix; - - default: RETURN_ERROR(parameter_unsupported, "unknown parameter"); - } -} - -size_t ZSTD_CCtx_getParameter(ZSTD_CCtx const* cctx, ZSTD_cParameter param, int* value) -{ - return ZSTD_CCtxParams_getParameter(&cctx->requestedParams, param, value); -} - -size_t ZSTD_CCtxParams_getParameter( - ZSTD_CCtx_params const* CCtxParams, ZSTD_cParameter param, int* value) -{ - switch(param) - { - case ZSTD_c_format : - *value = CCtxParams->format; - break; - case ZSTD_c_compressionLevel : - *value = CCtxParams->compressionLevel; - break; - case ZSTD_c_windowLog : - *value = (int)CCtxParams->cParams.windowLog; - break; - case ZSTD_c_hashLog : - *value = (int)CCtxParams->cParams.hashLog; - break; - case ZSTD_c_chainLog : - *value = (int)CCtxParams->cParams.chainLog; - break; - case ZSTD_c_searchLog : - *value = CCtxParams->cParams.searchLog; - break; - case ZSTD_c_minMatch : - *value = CCtxParams->cParams.minMatch; - break; - case ZSTD_c_targetLength : - *value = CCtxParams->cParams.targetLength; - break; - case ZSTD_c_strategy : - *value = (unsigned)CCtxParams->cParams.strategy; - break; - case ZSTD_c_contentSizeFlag : - *value = CCtxParams->fParams.contentSizeFlag; - break; - case ZSTD_c_checksumFlag : - *value = CCtxParams->fParams.checksumFlag; - break; - case ZSTD_c_dictIDFlag : - *value = !CCtxParams->fParams.noDictIDFlag; - break; - case ZSTD_c_forceMaxWindow : - *value = CCtxParams->forceWindow; - break; - case ZSTD_c_forceAttachDict : - *value = CCtxParams->attachDictPref; - break; - case ZSTD_c_literalCompressionMode : - *value = CCtxParams->literalCompressionMode; - break; - case ZSTD_c_nbWorkers : -#ifndef ZSTD_MULTITHREAD - assert(CCtxParams->nbWorkers == 0); -#endif - *value = CCtxParams->nbWorkers; - break; - case ZSTD_c_jobSize : -#ifndef ZSTD_MULTITHREAD - RETURN_ERROR(parameter_unsupported, "not compiled with multithreading"); -#else - assert(CCtxParams->jobSize <= INT_MAX); - *value = (int)CCtxParams->jobSize; - break; -#endif - case ZSTD_c_overlapLog : -#ifndef ZSTD_MULTITHREAD - RETURN_ERROR(parameter_unsupported, "not compiled with multithreading"); -#else - *value = CCtxParams->overlapLog; - break; -#endif - case ZSTD_c_rsyncable : -#ifndef ZSTD_MULTITHREAD - RETURN_ERROR(parameter_unsupported, "not compiled with multithreading"); -#else - *value = CCtxParams->rsyncable; - break; -#endif - case ZSTD_c_enableDedicatedDictSearch : - *value = CCtxParams->enableDedicatedDictSearch; - break; - case ZSTD_c_enableLongDistanceMatching : - *value = CCtxParams->ldmParams.enableLdm; - break; - case ZSTD_c_ldmHashLog : - *value = CCtxParams->ldmParams.hashLog; - break; - case ZSTD_c_ldmMinMatch : - *value = CCtxParams->ldmParams.minMatchLength; - break; - case ZSTD_c_ldmBucketSizeLog : - *value = CCtxParams->ldmParams.bucketSizeLog; - break; - case ZSTD_c_ldmHashRateLog : - *value = CCtxParams->ldmParams.hashRateLog; - break; - case ZSTD_c_targetCBlockSize : - *value = (int)CCtxParams->targetCBlockSize; - break; - case ZSTD_c_srcSizeHint : - *value = (int)CCtxParams->srcSizeHint; - break; - case ZSTD_c_stableInBuffer : - *value = (int)CCtxParams->inBufferMode; - break; - case ZSTD_c_stableOutBuffer : - *value = (int)CCtxParams->outBufferMode; - break; - case ZSTD_c_blockDelimiters : - *value = (int)CCtxParams->blockDelimiters; - break; - case ZSTD_c_validateSequences : - *value = (int)CCtxParams->validateSequences; - break; - case ZSTD_c_useBlockSplitter : - *value = (int)CCtxParams->useBlockSplitter; - break; - case ZSTD_c_useRowMatchFinder : - *value = (int)CCtxParams->useRowMatchFinder; - break; - case ZSTD_c_deterministicRefPrefix: - *value = (int)CCtxParams->deterministicRefPrefix; - break; - default: RETURN_ERROR(parameter_unsupported, "unknown parameter"); - } - return 0; -} - -/** ZSTD_CCtx_setParametersUsingCCtxParams() : - * just applies `params` into `cctx` - * no action is performed, parameters are merely stored. - * If ZSTDMT is enabled, parameters are pushed to cctx->mtctx. - * This is possible even if a compression is ongoing. - * In which case, new parameters will be applied on the fly, starting with next compression job. - */ -size_t ZSTD_CCtx_setParametersUsingCCtxParams( - ZSTD_CCtx* cctx, const ZSTD_CCtx_params* params) -{ - DEBUGLOG(4, "ZSTD_CCtx_setParametersUsingCCtxParams"); - RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong, - "The context is in the wrong stage!"); - RETURN_ERROR_IF(cctx->cdict, stage_wrong, - "Can't override parameters with cdict attached (some must " - "be inherited from the cdict)."); - - cctx->requestedParams = *params; - return 0; -} - -size_t ZSTD_CCtx_setPledgedSrcSize(ZSTD_CCtx* cctx, unsigned long long pledgedSrcSize) -{ - DEBUGLOG(4, "ZSTD_CCtx_setPledgedSrcSize to %u bytes", (U32)pledgedSrcSize); - RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong, - "Can't set pledgedSrcSize when not in init stage."); - cctx->pledgedSrcSizePlusOne = pledgedSrcSize+1; - return 0; -} - -static ZSTD_compressionParameters ZSTD_dedicatedDictSearch_getCParams( - int const compressionLevel, - size_t const dictSize); -static int ZSTD_dedicatedDictSearch_isSupported( - const ZSTD_compressionParameters* cParams); -static void ZSTD_dedicatedDictSearch_revertCParams( - ZSTD_compressionParameters* cParams); - -/** - * Initializes the local dict using the requested parameters. - * NOTE: This does not use the pledged src size, because it may be used for more - * than one compression. - */ -static size_t ZSTD_initLocalDict(ZSTD_CCtx* cctx) -{ - ZSTD_localDict* const dl = &cctx->localDict; - if (dl->dict == NULL) { - /* No local dictionary. */ - assert(dl->dictBuffer == NULL); - assert(dl->cdict == NULL); - assert(dl->dictSize == 0); - return 0; - } - if (dl->cdict != NULL) { - assert(cctx->cdict == dl->cdict); - /* Local dictionary already initialized. */ - return 0; - } - assert(dl->dictSize > 0); - assert(cctx->cdict == NULL); - assert(cctx->prefixDict.dict == NULL); - - dl->cdict = ZSTD_createCDict_advanced2( - dl->dict, - dl->dictSize, - ZSTD_dlm_byRef, - dl->dictContentType, - &cctx->requestedParams, - cctx->customMem); - RETURN_ERROR_IF(!dl->cdict, memory_allocation, "ZSTD_createCDict_advanced failed"); - cctx->cdict = dl->cdict; - return 0; -} - -size_t ZSTD_CCtx_loadDictionary_advanced( - ZSTD_CCtx* cctx, const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType) -{ - RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong, - "Can't load a dictionary when ctx is not in init stage."); - DEBUGLOG(4, "ZSTD_CCtx_loadDictionary_advanced (size: %u)", (U32)dictSize); - ZSTD_clearAllDicts(cctx); /* in case one already exists */ - if (dict == NULL || dictSize == 0) /* no dictionary mode */ - return 0; - if (dictLoadMethod == ZSTD_dlm_byRef) { - cctx->localDict.dict = dict; - } else { - void* dictBuffer; - RETURN_ERROR_IF(cctx->staticSize, memory_allocation, - "no malloc for static CCtx"); - dictBuffer = ZSTD_customMalloc(dictSize, cctx->customMem); - RETURN_ERROR_IF(!dictBuffer, memory_allocation, "NULL pointer!"); - ZSTD_memcpy(dictBuffer, dict, dictSize); - cctx->localDict.dictBuffer = dictBuffer; - cctx->localDict.dict = dictBuffer; - } - cctx->localDict.dictSize = dictSize; - cctx->localDict.dictContentType = dictContentType; - return 0; -} - -size_t ZSTD_CCtx_loadDictionary_byReference( - ZSTD_CCtx* cctx, const void* dict, size_t dictSize) -{ - return ZSTD_CCtx_loadDictionary_advanced( - cctx, dict, dictSize, ZSTD_dlm_byRef, ZSTD_dct_auto); -} - -size_t ZSTD_CCtx_loadDictionary(ZSTD_CCtx* cctx, const void* dict, size_t dictSize) -{ - return ZSTD_CCtx_loadDictionary_advanced( - cctx, dict, dictSize, ZSTD_dlm_byCopy, ZSTD_dct_auto); -} - - -size_t ZSTD_CCtx_refCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict) -{ - RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong, - "Can't ref a dict when ctx not in init stage."); - /* Free the existing local cdict (if any) to save memory. */ - ZSTD_clearAllDicts(cctx); - cctx->cdict = cdict; - return 0; -} - -size_t ZSTD_CCtx_refThreadPool(ZSTD_CCtx* cctx, ZSTD_threadPool* pool) -{ - RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong, - "Can't ref a pool when ctx not in init stage."); - cctx->pool = pool; - return 0; -} - -size_t ZSTD_CCtx_refPrefix(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize) -{ - return ZSTD_CCtx_refPrefix_advanced(cctx, prefix, prefixSize, ZSTD_dct_rawContent); -} - -size_t ZSTD_CCtx_refPrefix_advanced( - ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType) -{ - RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong, - "Can't ref a prefix when ctx not in init stage."); - ZSTD_clearAllDicts(cctx); - if (prefix != NULL && prefixSize > 0) { - cctx->prefixDict.dict = prefix; - cctx->prefixDict.dictSize = prefixSize; - cctx->prefixDict.dictContentType = dictContentType; - } - return 0; -} - -/*! ZSTD_CCtx_reset() : - * Also dumps dictionary */ -size_t ZSTD_CCtx_reset(ZSTD_CCtx* cctx, ZSTD_ResetDirective reset) -{ - if ( (reset == ZSTD_reset_session_only) - || (reset == ZSTD_reset_session_and_parameters) ) { - cctx->streamStage = zcss_init; - cctx->pledgedSrcSizePlusOne = 0; - } - if ( (reset == ZSTD_reset_parameters) - || (reset == ZSTD_reset_session_and_parameters) ) { - RETURN_ERROR_IF(cctx->streamStage != zcss_init, stage_wrong, - "Can't reset parameters only when not in init stage."); - ZSTD_clearAllDicts(cctx); - return ZSTD_CCtxParams_reset(&cctx->requestedParams); - } - return 0; -} - - -/** ZSTD_checkCParams() : - control CParam values remain within authorized range. - @return : 0, or an error code if one value is beyond authorized range */ -size_t ZSTD_checkCParams(ZSTD_compressionParameters cParams) -{ - BOUNDCHECK(ZSTD_c_windowLog, (int)cParams.windowLog); - BOUNDCHECK(ZSTD_c_chainLog, (int)cParams.chainLog); - BOUNDCHECK(ZSTD_c_hashLog, (int)cParams.hashLog); - BOUNDCHECK(ZSTD_c_searchLog, (int)cParams.searchLog); - BOUNDCHECK(ZSTD_c_minMatch, (int)cParams.minMatch); - BOUNDCHECK(ZSTD_c_targetLength,(int)cParams.targetLength); - BOUNDCHECK(ZSTD_c_strategy, cParams.strategy); - return 0; -} - -/** ZSTD_clampCParams() : - * make CParam values within valid range. - * @return : valid CParams */ -static ZSTD_compressionParameters -ZSTD_clampCParams(ZSTD_compressionParameters cParams) -{ -# define CLAMP_TYPE(cParam, val, type) { \ - ZSTD_bounds const bounds = ZSTD_cParam_getBounds(cParam); \ - if ((int)valbounds.upperBound) val=(type)bounds.upperBound; \ - } -# define CLAMP(cParam, val) CLAMP_TYPE(cParam, val, unsigned) - CLAMP(ZSTD_c_windowLog, cParams.windowLog); - CLAMP(ZSTD_c_chainLog, cParams.chainLog); - CLAMP(ZSTD_c_hashLog, cParams.hashLog); - CLAMP(ZSTD_c_searchLog, cParams.searchLog); - CLAMP(ZSTD_c_minMatch, cParams.minMatch); - CLAMP(ZSTD_c_targetLength,cParams.targetLength); - CLAMP_TYPE(ZSTD_c_strategy,cParams.strategy, ZSTD_strategy); - return cParams; -} - -/** ZSTD_cycleLog() : - * condition for correct operation : hashLog > 1 */ -U32 ZSTD_cycleLog(U32 hashLog, ZSTD_strategy strat) -{ - U32 const btScale = ((U32)strat >= (U32)ZSTD_btlazy2); - return hashLog - btScale; -} - -/** ZSTD_dictAndWindowLog() : - * Returns an adjusted window log that is large enough to fit the source and the dictionary. - * The zstd format says that the entire dictionary is valid if one byte of the dictionary - * is within the window. So the hashLog and chainLog should be large enough to reference both - * the dictionary and the window. So we must use this adjusted dictAndWindowLog when downsizing - * the hashLog and windowLog. - * NOTE: srcSize must not be ZSTD_CONTENTSIZE_UNKNOWN. - */ -static U32 ZSTD_dictAndWindowLog(U32 windowLog, U64 srcSize, U64 dictSize) -{ - const U64 maxWindowSize = 1ULL << ZSTD_WINDOWLOG_MAX; - /* No dictionary ==> No change */ - if (dictSize == 0) { - return windowLog; - } - assert(windowLog <= ZSTD_WINDOWLOG_MAX); - assert(srcSize != ZSTD_CONTENTSIZE_UNKNOWN); /* Handled in ZSTD_adjustCParams_internal() */ - { - U64 const windowSize = 1ULL << windowLog; - U64 const dictAndWindowSize = dictSize + windowSize; - /* If the window size is already large enough to fit both the source and the dictionary - * then just use the window size. Otherwise adjust so that it fits the dictionary and - * the window. - */ - if (windowSize >= dictSize + srcSize) { - return windowLog; /* Window size large enough already */ - } else if (dictAndWindowSize >= maxWindowSize) { - return ZSTD_WINDOWLOG_MAX; /* Larger than max window log */ - } else { - return ZSTD_highbit32((U32)dictAndWindowSize - 1) + 1; - } - } -} - -/** ZSTD_adjustCParams_internal() : - * optimize `cPar` for a specified input (`srcSize` and `dictSize`). - * mostly downsize to reduce memory consumption and initialization latency. - * `srcSize` can be ZSTD_CONTENTSIZE_UNKNOWN when not known. - * `mode` is the mode for parameter adjustment. See docs for `ZSTD_cParamMode_e`. - * note : `srcSize==0` means 0! - * condition : cPar is presumed validated (can be checked using ZSTD_checkCParams()). */ -static ZSTD_compressionParameters -ZSTD_adjustCParams_internal(ZSTD_compressionParameters cPar, - unsigned long long srcSize, - size_t dictSize, - ZSTD_cParamMode_e mode) -{ - const U64 minSrcSize = 513; /* (1<<9) + 1 */ - const U64 maxWindowResize = 1ULL << (ZSTD_WINDOWLOG_MAX-1); - assert(ZSTD_checkCParams(cPar)==0); - - switch (mode) { - case ZSTD_cpm_unknown: - case ZSTD_cpm_noAttachDict: - /* If we don't know the source size, don't make any - * assumptions about it. We will already have selected - * smaller parameters if a dictionary is in use. - */ - break; - case ZSTD_cpm_createCDict: - /* Assume a small source size when creating a dictionary - * with an unknown source size. - */ - if (dictSize && srcSize == ZSTD_CONTENTSIZE_UNKNOWN) - srcSize = minSrcSize; - break; - case ZSTD_cpm_attachDict: - /* Dictionary has its own dedicated parameters which have - * already been selected. We are selecting parameters - * for only the source. - */ - dictSize = 0; - break; - default: - assert(0); - break; - } - - /* resize windowLog if input is small enough, to use less memory */ - if ( (srcSize < maxWindowResize) - && (dictSize < maxWindowResize) ) { - U32 const tSize = (U32)(srcSize + dictSize); - static U32 const hashSizeMin = 1 << ZSTD_HASHLOG_MIN; - U32 const srcLog = (tSize < hashSizeMin) ? ZSTD_HASHLOG_MIN : - ZSTD_highbit32(tSize-1) + 1; - if (cPar.windowLog > srcLog) cPar.windowLog = srcLog; - } - if (srcSize != ZSTD_CONTENTSIZE_UNKNOWN) { - U32 const dictAndWindowLog = ZSTD_dictAndWindowLog(cPar.windowLog, (U64)srcSize, (U64)dictSize); - U32 const cycleLog = ZSTD_cycleLog(cPar.chainLog, cPar.strategy); - if (cPar.hashLog > dictAndWindowLog+1) cPar.hashLog = dictAndWindowLog+1; - if (cycleLog > dictAndWindowLog) - cPar.chainLog -= (cycleLog - dictAndWindowLog); - } - - if (cPar.windowLog < ZSTD_WINDOWLOG_ABSOLUTEMIN) - cPar.windowLog = ZSTD_WINDOWLOG_ABSOLUTEMIN; /* minimum wlog required for valid frame header */ - - return cPar; -} - -ZSTD_compressionParameters -ZSTD_adjustCParams(ZSTD_compressionParameters cPar, - unsigned long long srcSize, - size_t dictSize) -{ - cPar = ZSTD_clampCParams(cPar); /* resulting cPar is necessarily valid (all parameters within range) */ - if (srcSize == 0) srcSize = ZSTD_CONTENTSIZE_UNKNOWN; - return ZSTD_adjustCParams_internal(cPar, srcSize, dictSize, ZSTD_cpm_unknown); -} - -static ZSTD_compressionParameters ZSTD_getCParams_internal(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode); -static ZSTD_parameters ZSTD_getParams_internal(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode); - -static void ZSTD_overrideCParams( - ZSTD_compressionParameters* cParams, - const ZSTD_compressionParameters* overrides) -{ - if (overrides->windowLog) cParams->windowLog = overrides->windowLog; - if (overrides->hashLog) cParams->hashLog = overrides->hashLog; - if (overrides->chainLog) cParams->chainLog = overrides->chainLog; - if (overrides->searchLog) cParams->searchLog = overrides->searchLog; - if (overrides->minMatch) cParams->minMatch = overrides->minMatch; - if (overrides->targetLength) cParams->targetLength = overrides->targetLength; - if (overrides->strategy) cParams->strategy = overrides->strategy; -} - -ZSTD_compressionParameters ZSTD_getCParamsFromCCtxParams( - const ZSTD_CCtx_params* CCtxParams, U64 srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode) -{ - ZSTD_compressionParameters cParams; - if (srcSizeHint == ZSTD_CONTENTSIZE_UNKNOWN && CCtxParams->srcSizeHint > 0) { - srcSizeHint = CCtxParams->srcSizeHint; - } - cParams = ZSTD_getCParams_internal(CCtxParams->compressionLevel, srcSizeHint, dictSize, mode); - if (CCtxParams->ldmParams.enableLdm == ZSTD_ps_enable) cParams.windowLog = ZSTD_LDM_DEFAULT_WINDOW_LOG; - ZSTD_overrideCParams(&cParams, &CCtxParams->cParams); - assert(!ZSTD_checkCParams(cParams)); - /* srcSizeHint == 0 means 0 */ - return ZSTD_adjustCParams_internal(cParams, srcSizeHint, dictSize, mode); -} - -static size_t -ZSTD_sizeof_matchState(const ZSTD_compressionParameters* const cParams, - const ZSTD_paramSwitch_e useRowMatchFinder, - const U32 enableDedicatedDictSearch, - const U32 forCCtx) -{ - /* chain table size should be 0 for fast or row-hash strategies */ - size_t const chainSize = ZSTD_allocateChainTable(cParams->strategy, useRowMatchFinder, enableDedicatedDictSearch && !forCCtx) - ? ((size_t)1 << cParams->chainLog) - : 0; - size_t const hSize = ((size_t)1) << cParams->hashLog; - U32 const hashLog3 = (forCCtx && cParams->minMatch==3) ? MIN(ZSTD_HASHLOG3_MAX, cParams->windowLog) : 0; - size_t const h3Size = hashLog3 ? ((size_t)1) << hashLog3 : 0; - /* We don't use ZSTD_cwksp_alloc_size() here because the tables aren't - * surrounded by redzones in ASAN. */ - size_t const tableSpace = chainSize * sizeof(U32) - + hSize * sizeof(U32) - + h3Size * sizeof(U32); - size_t const optPotentialSpace = - ZSTD_cwksp_aligned_alloc_size((MaxML+1) * sizeof(U32)) - + ZSTD_cwksp_aligned_alloc_size((MaxLL+1) * sizeof(U32)) - + ZSTD_cwksp_aligned_alloc_size((MaxOff+1) * sizeof(U32)) - + ZSTD_cwksp_aligned_alloc_size((1<strategy, useRowMatchFinder) - ? ZSTD_cwksp_aligned_alloc_size(hSize*sizeof(U16)) - : 0; - size_t const optSpace = (forCCtx && (cParams->strategy >= ZSTD_btopt)) - ? optPotentialSpace - : 0; - size_t const slackSpace = ZSTD_cwksp_slack_space_required(); - - /* tables are guaranteed to be sized in multiples of 64 bytes (or 16 uint32_t) */ - ZSTD_STATIC_ASSERT(ZSTD_HASHLOG_MIN >= 4 && ZSTD_WINDOWLOG_MIN >= 4 && ZSTD_CHAINLOG_MIN >= 4); - assert(useRowMatchFinder != ZSTD_ps_auto); - - DEBUGLOG(4, "chainSize: %u - hSize: %u - h3Size: %u", - (U32)chainSize, (U32)hSize, (U32)h3Size); - return tableSpace + optSpace + slackSpace + lazyAdditionalSpace; -} - -static size_t ZSTD_estimateCCtxSize_usingCCtxParams_internal( - const ZSTD_compressionParameters* cParams, - const ldmParams_t* ldmParams, - const int isStatic, - const ZSTD_paramSwitch_e useRowMatchFinder, - const size_t buffInSize, - const size_t buffOutSize, - const U64 pledgedSrcSize) -{ - size_t const windowSize = (size_t) BOUNDED(1ULL, 1ULL << cParams->windowLog, pledgedSrcSize); - size_t const blockSize = MIN(ZSTD_BLOCKSIZE_MAX, windowSize); - U32 const divider = (cParams->minMatch==3) ? 3 : 4; - size_t const maxNbSeq = blockSize / divider; - size_t const tokenSpace = ZSTD_cwksp_alloc_size(WILDCOPY_OVERLENGTH + blockSize) - + ZSTD_cwksp_aligned_alloc_size(maxNbSeq * sizeof(seqDef)) - + 3 * ZSTD_cwksp_alloc_size(maxNbSeq * sizeof(BYTE)); - size_t const entropySpace = ZSTD_cwksp_alloc_size(ENTROPY_WORKSPACE_SIZE); - size_t const blockStateSpace = 2 * ZSTD_cwksp_alloc_size(sizeof(ZSTD_compressedBlockState_t)); - size_t const matchStateSize = ZSTD_sizeof_matchState(cParams, useRowMatchFinder, /* enableDedicatedDictSearch */ 0, /* forCCtx */ 1); - - size_t const ldmSpace = ZSTD_ldm_getTableSize(*ldmParams); - size_t const maxNbLdmSeq = ZSTD_ldm_getMaxNbSeq(*ldmParams, blockSize); - size_t const ldmSeqSpace = ldmParams->enableLdm == ZSTD_ps_enable ? - ZSTD_cwksp_aligned_alloc_size(maxNbLdmSeq * sizeof(rawSeq)) : 0; - - - size_t const bufferSpace = ZSTD_cwksp_alloc_size(buffInSize) - + ZSTD_cwksp_alloc_size(buffOutSize); - - size_t const cctxSpace = isStatic ? ZSTD_cwksp_alloc_size(sizeof(ZSTD_CCtx)) : 0; - - size_t const neededSpace = - cctxSpace + - entropySpace + - blockStateSpace + - ldmSpace + - ldmSeqSpace + - matchStateSize + - tokenSpace + - bufferSpace; - - DEBUGLOG(5, "estimate workspace : %u", (U32)neededSpace); - return neededSpace; -} - -size_t ZSTD_estimateCCtxSize_usingCCtxParams(const ZSTD_CCtx_params* params) -{ - ZSTD_compressionParameters const cParams = - ZSTD_getCParamsFromCCtxParams(params, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict); - ZSTD_paramSwitch_e const useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(params->useRowMatchFinder, - &cParams); - - RETURN_ERROR_IF(params->nbWorkers > 0, GENERIC, "Estimate CCtx size is supported for single-threaded compression only."); - /* estimateCCtxSize is for one-shot compression. So no buffers should - * be needed. However, we still allocate two 0-sized buffers, which can - * take space under ASAN. */ - return ZSTD_estimateCCtxSize_usingCCtxParams_internal( - &cParams, ¶ms->ldmParams, 1, useRowMatchFinder, 0, 0, ZSTD_CONTENTSIZE_UNKNOWN); -} - -size_t ZSTD_estimateCCtxSize_usingCParams(ZSTD_compressionParameters cParams) -{ - ZSTD_CCtx_params initialParams = ZSTD_makeCCtxParamsFromCParams(cParams); - if (ZSTD_rowMatchFinderSupported(cParams.strategy)) { - /* Pick bigger of not using and using row-based matchfinder for greedy and lazy strategies */ - size_t noRowCCtxSize; - size_t rowCCtxSize; - initialParams.useRowMatchFinder = ZSTD_ps_disable; - noRowCCtxSize = ZSTD_estimateCCtxSize_usingCCtxParams(&initialParams); - initialParams.useRowMatchFinder = ZSTD_ps_enable; - rowCCtxSize = ZSTD_estimateCCtxSize_usingCCtxParams(&initialParams); - return MAX(noRowCCtxSize, rowCCtxSize); - } else { - return ZSTD_estimateCCtxSize_usingCCtxParams(&initialParams); - } -} - -static size_t ZSTD_estimateCCtxSize_internal(int compressionLevel) -{ - int tier = 0; - size_t largestSize = 0; - static const unsigned long long srcSizeTiers[4] = {16 KB, 128 KB, 256 KB, ZSTD_CONTENTSIZE_UNKNOWN}; - for (; tier < 4; ++tier) { - /* Choose the set of cParams for a given level across all srcSizes that give the largest cctxSize */ - ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, srcSizeTiers[tier], 0, ZSTD_cpm_noAttachDict); - largestSize = MAX(ZSTD_estimateCCtxSize_usingCParams(cParams), largestSize); - } - return largestSize; -} - -size_t ZSTD_estimateCCtxSize(int compressionLevel) -{ - int level; - size_t memBudget = 0; - for (level=MIN(compressionLevel, 1); level<=compressionLevel; level++) { - /* Ensure monotonically increasing memory usage as compression level increases */ - size_t const newMB = ZSTD_estimateCCtxSize_internal(level); - if (newMB > memBudget) memBudget = newMB; - } - return memBudget; -} - -size_t ZSTD_estimateCStreamSize_usingCCtxParams(const ZSTD_CCtx_params* params) -{ - RETURN_ERROR_IF(params->nbWorkers > 0, GENERIC, "Estimate CCtx size is supported for single-threaded compression only."); - { ZSTD_compressionParameters const cParams = - ZSTD_getCParamsFromCCtxParams(params, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict); - size_t const blockSize = MIN(ZSTD_BLOCKSIZE_MAX, (size_t)1 << cParams.windowLog); - size_t const inBuffSize = (params->inBufferMode == ZSTD_bm_buffered) - ? ((size_t)1 << cParams.windowLog) + blockSize - : 0; - size_t const outBuffSize = (params->outBufferMode == ZSTD_bm_buffered) - ? ZSTD_compressBound(blockSize) + 1 - : 0; - ZSTD_paramSwitch_e const useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(params->useRowMatchFinder, ¶ms->cParams); - - return ZSTD_estimateCCtxSize_usingCCtxParams_internal( - &cParams, ¶ms->ldmParams, 1, useRowMatchFinder, inBuffSize, outBuffSize, - ZSTD_CONTENTSIZE_UNKNOWN); - } -} - -size_t ZSTD_estimateCStreamSize_usingCParams(ZSTD_compressionParameters cParams) -{ - ZSTD_CCtx_params initialParams = ZSTD_makeCCtxParamsFromCParams(cParams); - if (ZSTD_rowMatchFinderSupported(cParams.strategy)) { - /* Pick bigger of not using and using row-based matchfinder for greedy and lazy strategies */ - size_t noRowCCtxSize; - size_t rowCCtxSize; - initialParams.useRowMatchFinder = ZSTD_ps_disable; - noRowCCtxSize = ZSTD_estimateCStreamSize_usingCCtxParams(&initialParams); - initialParams.useRowMatchFinder = ZSTD_ps_enable; - rowCCtxSize = ZSTD_estimateCStreamSize_usingCCtxParams(&initialParams); - return MAX(noRowCCtxSize, rowCCtxSize); - } else { - return ZSTD_estimateCStreamSize_usingCCtxParams(&initialParams); - } -} - -static size_t ZSTD_estimateCStreamSize_internal(int compressionLevel) -{ - ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict); - return ZSTD_estimateCStreamSize_usingCParams(cParams); -} - -size_t ZSTD_estimateCStreamSize(int compressionLevel) -{ - int level; - size_t memBudget = 0; - for (level=MIN(compressionLevel, 1); level<=compressionLevel; level++) { - size_t const newMB = ZSTD_estimateCStreamSize_internal(level); - if (newMB > memBudget) memBudget = newMB; - } - return memBudget; -} - -/* ZSTD_getFrameProgression(): - * tells how much data has been consumed (input) and produced (output) for current frame. - * able to count progression inside worker threads (non-blocking mode). - */ -ZSTD_frameProgression ZSTD_getFrameProgression(const ZSTD_CCtx* cctx) -{ -#ifdef ZSTD_MULTITHREAD - if (cctx->appliedParams.nbWorkers > 0) { - return ZSTDMT_getFrameProgression(cctx->mtctx); - } -#endif - { ZSTD_frameProgression fp; - size_t const buffered = (cctx->inBuff == NULL) ? 0 : - cctx->inBuffPos - cctx->inToCompress; - if (buffered) assert(cctx->inBuffPos >= cctx->inToCompress); - assert(buffered <= ZSTD_BLOCKSIZE_MAX); - fp.ingested = cctx->consumedSrcSize + buffered; - fp.consumed = cctx->consumedSrcSize; - fp.produced = cctx->producedCSize; - fp.flushed = cctx->producedCSize; /* simplified; some data might still be left within streaming output buffer */ - fp.currentJobID = 0; - fp.nbActiveWorkers = 0; - return fp; -} } - -/*! ZSTD_toFlushNow() - * Only useful for multithreading scenarios currently (nbWorkers >= 1). - */ -size_t ZSTD_toFlushNow(ZSTD_CCtx* cctx) -{ -#ifdef ZSTD_MULTITHREAD - if (cctx->appliedParams.nbWorkers > 0) { - return ZSTDMT_toFlushNow(cctx->mtctx); - } -#endif - (void)cctx; - return 0; /* over-simplification; could also check if context is currently running in streaming mode, and in which case, report how many bytes are left to be flushed within output buffer */ -} - -static void ZSTD_assertEqualCParams(ZSTD_compressionParameters cParams1, - ZSTD_compressionParameters cParams2) -{ - (void)cParams1; - (void)cParams2; - assert(cParams1.windowLog == cParams2.windowLog); - assert(cParams1.chainLog == cParams2.chainLog); - assert(cParams1.hashLog == cParams2.hashLog); - assert(cParams1.searchLog == cParams2.searchLog); - assert(cParams1.minMatch == cParams2.minMatch); - assert(cParams1.targetLength == cParams2.targetLength); - assert(cParams1.strategy == cParams2.strategy); -} - -void ZSTD_reset_compressedBlockState(ZSTD_compressedBlockState_t* bs) -{ - int i; - for (i = 0; i < ZSTD_REP_NUM; ++i) - bs->rep[i] = repStartValue[i]; - bs->entropy.huf.repeatMode = HUF_repeat_none; - bs->entropy.fse.offcode_repeatMode = FSE_repeat_none; - bs->entropy.fse.matchlength_repeatMode = FSE_repeat_none; - bs->entropy.fse.litlength_repeatMode = FSE_repeat_none; -} - -/*! ZSTD_invalidateMatchState() - * Invalidate all the matches in the match finder tables. - * Requires nextSrc and base to be set (can be NULL). - */ -static void ZSTD_invalidateMatchState(ZSTD_matchState_t* ms) -{ - ZSTD_window_clear(&ms->window); - - ms->nextToUpdate = ms->window.dictLimit; - ms->loadedDictEnd = 0; - ms->opt.litLengthSum = 0; /* force reset of btopt stats */ - ms->dictMatchState = NULL; -} - -/** - * Controls, for this matchState reset, whether the tables need to be cleared / - * prepared for the coming compression (ZSTDcrp_makeClean), or whether the - * tables can be left unclean (ZSTDcrp_leaveDirty), because we know that a - * subsequent operation will overwrite the table space anyways (e.g., copying - * the matchState contents in from a CDict). - */ -typedef enum { - ZSTDcrp_makeClean, - ZSTDcrp_leaveDirty -} ZSTD_compResetPolicy_e; - -/** - * Controls, for this matchState reset, whether indexing can continue where it - * left off (ZSTDirp_continue), or whether it needs to be restarted from zero - * (ZSTDirp_reset). - */ -typedef enum { - ZSTDirp_continue, - ZSTDirp_reset -} ZSTD_indexResetPolicy_e; - -typedef enum { - ZSTD_resetTarget_CDict, - ZSTD_resetTarget_CCtx -} ZSTD_resetTarget_e; - - -static size_t -ZSTD_reset_matchState(ZSTD_matchState_t* ms, - ZSTD_cwksp* ws, - const ZSTD_compressionParameters* cParams, - const ZSTD_paramSwitch_e useRowMatchFinder, - const ZSTD_compResetPolicy_e crp, - const ZSTD_indexResetPolicy_e forceResetIndex, - const ZSTD_resetTarget_e forWho) -{ - /* disable chain table allocation for fast or row-based strategies */ - size_t const chainSize = ZSTD_allocateChainTable(cParams->strategy, useRowMatchFinder, - ms->dedicatedDictSearch && (forWho == ZSTD_resetTarget_CDict)) - ? ((size_t)1 << cParams->chainLog) - : 0; - size_t const hSize = ((size_t)1) << cParams->hashLog; - U32 const hashLog3 = ((forWho == ZSTD_resetTarget_CCtx) && cParams->minMatch==3) ? MIN(ZSTD_HASHLOG3_MAX, cParams->windowLog) : 0; - size_t const h3Size = hashLog3 ? ((size_t)1) << hashLog3 : 0; - - DEBUGLOG(4, "reset indices : %u", forceResetIndex == ZSTDirp_reset); - assert(useRowMatchFinder != ZSTD_ps_auto); - if (forceResetIndex == ZSTDirp_reset) { - ZSTD_window_init(&ms->window); - ZSTD_cwksp_mark_tables_dirty(ws); - } - - ms->hashLog3 = hashLog3; - - ZSTD_invalidateMatchState(ms); - - assert(!ZSTD_cwksp_reserve_failed(ws)); /* check that allocation hasn't already failed */ - - ZSTD_cwksp_clear_tables(ws); - - DEBUGLOG(5, "reserving table space"); - /* table Space */ - ms->hashTable = (U32*)ZSTD_cwksp_reserve_table(ws, hSize * sizeof(U32)); - ms->chainTable = (U32*)ZSTD_cwksp_reserve_table(ws, chainSize * sizeof(U32)); - ms->hashTable3 = (U32*)ZSTD_cwksp_reserve_table(ws, h3Size * sizeof(U32)); - RETURN_ERROR_IF(ZSTD_cwksp_reserve_failed(ws), memory_allocation, - "failed a workspace allocation in ZSTD_reset_matchState"); - - DEBUGLOG(4, "reset table : %u", crp!=ZSTDcrp_leaveDirty); - if (crp!=ZSTDcrp_leaveDirty) { - /* reset tables only */ - ZSTD_cwksp_clean_tables(ws); - } - - /* opt parser space */ - if ((forWho == ZSTD_resetTarget_CCtx) && (cParams->strategy >= ZSTD_btopt)) { - DEBUGLOG(4, "reserving optimal parser space"); - ms->opt.litFreq = (unsigned*)ZSTD_cwksp_reserve_aligned(ws, (1<opt.litLengthFreq = (unsigned*)ZSTD_cwksp_reserve_aligned(ws, (MaxLL+1) * sizeof(unsigned)); - ms->opt.matchLengthFreq = (unsigned*)ZSTD_cwksp_reserve_aligned(ws, (MaxML+1) * sizeof(unsigned)); - ms->opt.offCodeFreq = (unsigned*)ZSTD_cwksp_reserve_aligned(ws, (MaxOff+1) * sizeof(unsigned)); - ms->opt.matchTable = (ZSTD_match_t*)ZSTD_cwksp_reserve_aligned(ws, (ZSTD_OPT_NUM+1) * sizeof(ZSTD_match_t)); - ms->opt.priceTable = (ZSTD_optimal_t*)ZSTD_cwksp_reserve_aligned(ws, (ZSTD_OPT_NUM+1) * sizeof(ZSTD_optimal_t)); - } - - if (ZSTD_rowMatchFinderUsed(cParams->strategy, useRowMatchFinder)) { - { /* Row match finder needs an additional table of hashes ("tags") */ - size_t const tagTableSize = hSize*sizeof(U16); - ms->tagTable = (U16*)ZSTD_cwksp_reserve_aligned(ws, tagTableSize); - if (ms->tagTable) ZSTD_memset(ms->tagTable, 0, tagTableSize); - } - { /* Switch to 32-entry rows if searchLog is 5 (or more) */ - U32 const rowLog = BOUNDED(4, cParams->searchLog, 6); - assert(cParams->hashLog >= rowLog); - ms->rowHashLog = cParams->hashLog - rowLog; - } - } - - ms->cParams = *cParams; - - RETURN_ERROR_IF(ZSTD_cwksp_reserve_failed(ws), memory_allocation, - "failed a workspace allocation in ZSTD_reset_matchState"); - return 0; -} - -/* ZSTD_indexTooCloseToMax() : - * minor optimization : prefer memset() rather than reduceIndex() - * which is measurably slow in some circumstances (reported for Visual Studio). - * Works when re-using a context for a lot of smallish inputs : - * if all inputs are smaller than ZSTD_INDEXOVERFLOW_MARGIN, - * memset() will be triggered before reduceIndex(). - */ -#define ZSTD_INDEXOVERFLOW_MARGIN (16 MB) -static int ZSTD_indexTooCloseToMax(ZSTD_window_t w) -{ - return (size_t)(w.nextSrc - w.base) > (ZSTD_CURRENT_MAX - ZSTD_INDEXOVERFLOW_MARGIN); -} - -/** ZSTD_dictTooBig(): - * When dictionaries are larger than ZSTD_CHUNKSIZE_MAX they can't be loaded in - * one go generically. So we ensure that in that case we reset the tables to zero, - * so that we can load as much of the dictionary as possible. - */ -static int ZSTD_dictTooBig(size_t const loadedDictSize) -{ - return loadedDictSize > ZSTD_CHUNKSIZE_MAX; -} - -/*! ZSTD_resetCCtx_internal() : - * @param loadedDictSize The size of the dictionary to be loaded - * into the context, if any. If no dictionary is used, or the - * dictionary is being attached / copied, then pass 0. - * note : `params` are assumed fully validated at this stage. - */ -static size_t ZSTD_resetCCtx_internal(ZSTD_CCtx* zc, - ZSTD_CCtx_params const* params, - U64 const pledgedSrcSize, - size_t const loadedDictSize, - ZSTD_compResetPolicy_e const crp, - ZSTD_buffered_policy_e const zbuff) -{ - ZSTD_cwksp* const ws = &zc->workspace; - DEBUGLOG(4, "ZSTD_resetCCtx_internal: pledgedSrcSize=%u, wlog=%u, useRowMatchFinder=%d useBlockSplitter=%d", - (U32)pledgedSrcSize, params->cParams.windowLog, (int)params->useRowMatchFinder, (int)params->useBlockSplitter); - assert(!ZSTD_isError(ZSTD_checkCParams(params->cParams))); - - zc->isFirstBlock = 1; - - /* Set applied params early so we can modify them for LDM, - * and point params at the applied params. - */ - zc->appliedParams = *params; - params = &zc->appliedParams; - - assert(params->useRowMatchFinder != ZSTD_ps_auto); - assert(params->useBlockSplitter != ZSTD_ps_auto); - assert(params->ldmParams.enableLdm != ZSTD_ps_auto); - if (params->ldmParams.enableLdm == ZSTD_ps_enable) { - /* Adjust long distance matching parameters */ - ZSTD_ldm_adjustParameters(&zc->appliedParams.ldmParams, ¶ms->cParams); - assert(params->ldmParams.hashLog >= params->ldmParams.bucketSizeLog); - assert(params->ldmParams.hashRateLog < 32); - } - - { size_t const windowSize = MAX(1, (size_t)MIN(((U64)1 << params->cParams.windowLog), pledgedSrcSize)); - size_t const blockSize = MIN(ZSTD_BLOCKSIZE_MAX, windowSize); - U32 const divider = (params->cParams.minMatch==3) ? 3 : 4; - size_t const maxNbSeq = blockSize / divider; - size_t const buffOutSize = (zbuff == ZSTDb_buffered && params->outBufferMode == ZSTD_bm_buffered) - ? ZSTD_compressBound(blockSize) + 1 - : 0; - size_t const buffInSize = (zbuff == ZSTDb_buffered && params->inBufferMode == ZSTD_bm_buffered) - ? windowSize + blockSize - : 0; - size_t const maxNbLdmSeq = ZSTD_ldm_getMaxNbSeq(params->ldmParams, blockSize); - - int const indexTooClose = ZSTD_indexTooCloseToMax(zc->blockState.matchState.window); - int const dictTooBig = ZSTD_dictTooBig(loadedDictSize); - ZSTD_indexResetPolicy_e needsIndexReset = - (indexTooClose || dictTooBig || !zc->initialized) ? ZSTDirp_reset : ZSTDirp_continue; - - size_t const neededSpace = - ZSTD_estimateCCtxSize_usingCCtxParams_internal( - ¶ms->cParams, ¶ms->ldmParams, zc->staticSize != 0, params->useRowMatchFinder, - buffInSize, buffOutSize, pledgedSrcSize); - int resizeWorkspace; - - FORWARD_IF_ERROR(neededSpace, "cctx size estimate failed!"); - - if (!zc->staticSize) ZSTD_cwksp_bump_oversized_duration(ws, 0); - - { /* Check if workspace is large enough, alloc a new one if needed */ - int const workspaceTooSmall = ZSTD_cwksp_sizeof(ws) < neededSpace; - int const workspaceWasteful = ZSTD_cwksp_check_wasteful(ws, neededSpace); - resizeWorkspace = workspaceTooSmall || workspaceWasteful; - DEBUGLOG(4, "Need %zu B workspace", neededSpace); - DEBUGLOG(4, "windowSize: %zu - blockSize: %zu", windowSize, blockSize); - - if (resizeWorkspace) { - DEBUGLOG(4, "Resize workspaceSize from %zuKB to %zuKB", - ZSTD_cwksp_sizeof(ws) >> 10, - neededSpace >> 10); - - RETURN_ERROR_IF(zc->staticSize, memory_allocation, "static cctx : no resize"); - - needsIndexReset = ZSTDirp_reset; - - ZSTD_cwksp_free(ws, zc->customMem); - FORWARD_IF_ERROR(ZSTD_cwksp_create(ws, neededSpace, zc->customMem), ""); - - DEBUGLOG(5, "reserving object space"); - /* Statically sized space. - * entropyWorkspace never moves, - * though prev/next block swap places */ - assert(ZSTD_cwksp_check_available(ws, 2 * sizeof(ZSTD_compressedBlockState_t))); - zc->blockState.prevCBlock = (ZSTD_compressedBlockState_t*) ZSTD_cwksp_reserve_object(ws, sizeof(ZSTD_compressedBlockState_t)); - RETURN_ERROR_IF(zc->blockState.prevCBlock == NULL, memory_allocation, "couldn't allocate prevCBlock"); - zc->blockState.nextCBlock = (ZSTD_compressedBlockState_t*) ZSTD_cwksp_reserve_object(ws, sizeof(ZSTD_compressedBlockState_t)); - RETURN_ERROR_IF(zc->blockState.nextCBlock == NULL, memory_allocation, "couldn't allocate nextCBlock"); - zc->entropyWorkspace = (U32*) ZSTD_cwksp_reserve_object(ws, ENTROPY_WORKSPACE_SIZE); - RETURN_ERROR_IF(zc->entropyWorkspace == NULL, memory_allocation, "couldn't allocate entropyWorkspace"); - } } - - ZSTD_cwksp_clear(ws); - - /* init params */ - zc->blockState.matchState.cParams = params->cParams; - zc->pledgedSrcSizePlusOne = pledgedSrcSize+1; - zc->consumedSrcSize = 0; - zc->producedCSize = 0; - if (pledgedSrcSize == ZSTD_CONTENTSIZE_UNKNOWN) - zc->appliedParams.fParams.contentSizeFlag = 0; - DEBUGLOG(4, "pledged content size : %u ; flag : %u", - (unsigned)pledgedSrcSize, zc->appliedParams.fParams.contentSizeFlag); - zc->blockSize = blockSize; - - XXH64_reset(&zc->xxhState, 0); - zc->stage = ZSTDcs_init; - zc->dictID = 0; - zc->dictContentSize = 0; - - ZSTD_reset_compressedBlockState(zc->blockState.prevCBlock); - - /* ZSTD_wildcopy() is used to copy into the literals buffer, - * so we have to oversize the buffer by WILDCOPY_OVERLENGTH bytes. - */ - zc->seqStore.litStart = ZSTD_cwksp_reserve_buffer(ws, blockSize + WILDCOPY_OVERLENGTH); - zc->seqStore.maxNbLit = blockSize; - - /* buffers */ - zc->bufferedPolicy = zbuff; - zc->inBuffSize = buffInSize; - zc->inBuff = (char*)ZSTD_cwksp_reserve_buffer(ws, buffInSize); - zc->outBuffSize = buffOutSize; - zc->outBuff = (char*)ZSTD_cwksp_reserve_buffer(ws, buffOutSize); - - /* ldm bucketOffsets table */ - if (params->ldmParams.enableLdm == ZSTD_ps_enable) { - /* TODO: avoid memset? */ - size_t const numBuckets = - ((size_t)1) << (params->ldmParams.hashLog - - params->ldmParams.bucketSizeLog); - zc->ldmState.bucketOffsets = ZSTD_cwksp_reserve_buffer(ws, numBuckets); - ZSTD_memset(zc->ldmState.bucketOffsets, 0, numBuckets); - } - - /* sequences storage */ - ZSTD_referenceExternalSequences(zc, NULL, 0); - zc->seqStore.maxNbSeq = maxNbSeq; - zc->seqStore.llCode = ZSTD_cwksp_reserve_buffer(ws, maxNbSeq * sizeof(BYTE)); - zc->seqStore.mlCode = ZSTD_cwksp_reserve_buffer(ws, maxNbSeq * sizeof(BYTE)); - zc->seqStore.ofCode = ZSTD_cwksp_reserve_buffer(ws, maxNbSeq * sizeof(BYTE)); - zc->seqStore.sequencesStart = (seqDef*)ZSTD_cwksp_reserve_aligned(ws, maxNbSeq * sizeof(seqDef)); - - FORWARD_IF_ERROR(ZSTD_reset_matchState( - &zc->blockState.matchState, - ws, - ¶ms->cParams, - params->useRowMatchFinder, - crp, - needsIndexReset, - ZSTD_resetTarget_CCtx), ""); - - /* ldm hash table */ - if (params->ldmParams.enableLdm == ZSTD_ps_enable) { - /* TODO: avoid memset? */ - size_t const ldmHSize = ((size_t)1) << params->ldmParams.hashLog; - zc->ldmState.hashTable = (ldmEntry_t*)ZSTD_cwksp_reserve_aligned(ws, ldmHSize * sizeof(ldmEntry_t)); - ZSTD_memset(zc->ldmState.hashTable, 0, ldmHSize * sizeof(ldmEntry_t)); - zc->ldmSequences = (rawSeq*)ZSTD_cwksp_reserve_aligned(ws, maxNbLdmSeq * sizeof(rawSeq)); - zc->maxNbLdmSequences = maxNbLdmSeq; - - ZSTD_window_init(&zc->ldmState.window); - zc->ldmState.loadedDictEnd = 0; - } - - DEBUGLOG(3, "wksp: finished allocating, %zd bytes remain available", ZSTD_cwksp_available_space(ws)); - assert(ZSTD_cwksp_estimated_space_within_bounds(ws, neededSpace, resizeWorkspace)); - - zc->initialized = 1; - - return 0; - } -} - -/* ZSTD_invalidateRepCodes() : - * ensures next compression will not use repcodes from previous block. - * Note : only works with regular variant; - * do not use with extDict variant ! */ -void ZSTD_invalidateRepCodes(ZSTD_CCtx* cctx) { - int i; - for (i=0; iblockState.prevCBlock->rep[i] = 0; - assert(!ZSTD_window_hasExtDict(cctx->blockState.matchState.window)); -} - -/* These are the approximate sizes for each strategy past which copying the - * dictionary tables into the working context is faster than using them - * in-place. - */ -static const size_t attachDictSizeCutoffs[ZSTD_STRATEGY_MAX+1] = { - 8 KB, /* unused */ - 8 KB, /* ZSTD_fast */ - 16 KB, /* ZSTD_dfast */ - 32 KB, /* ZSTD_greedy */ - 32 KB, /* ZSTD_lazy */ - 32 KB, /* ZSTD_lazy2 */ - 32 KB, /* ZSTD_btlazy2 */ - 32 KB, /* ZSTD_btopt */ - 8 KB, /* ZSTD_btultra */ - 8 KB /* ZSTD_btultra2 */ -}; - -static int ZSTD_shouldAttachDict(const ZSTD_CDict* cdict, - const ZSTD_CCtx_params* params, - U64 pledgedSrcSize) -{ - size_t cutoff = attachDictSizeCutoffs[cdict->matchState.cParams.strategy]; - int const dedicatedDictSearch = cdict->matchState.dedicatedDictSearch; - return dedicatedDictSearch - || ( ( pledgedSrcSize <= cutoff - || pledgedSrcSize == ZSTD_CONTENTSIZE_UNKNOWN - || params->attachDictPref == ZSTD_dictForceAttach ) - && params->attachDictPref != ZSTD_dictForceCopy - && !params->forceWindow ); /* dictMatchState isn't correctly - * handled in _enforceMaxDist */ -} - -static size_t -ZSTD_resetCCtx_byAttachingCDict(ZSTD_CCtx* cctx, - const ZSTD_CDict* cdict, - ZSTD_CCtx_params params, - U64 pledgedSrcSize, - ZSTD_buffered_policy_e zbuff) -{ - DEBUGLOG(4, "ZSTD_resetCCtx_byAttachingCDict() pledgedSrcSize=%llu", - (unsigned long long)pledgedSrcSize); - { - ZSTD_compressionParameters adjusted_cdict_cParams = cdict->matchState.cParams; - unsigned const windowLog = params.cParams.windowLog; - assert(windowLog != 0); - /* Resize working context table params for input only, since the dict - * has its own tables. */ - /* pledgedSrcSize == 0 means 0! */ - - if (cdict->matchState.dedicatedDictSearch) { - ZSTD_dedicatedDictSearch_revertCParams(&adjusted_cdict_cParams); - } - - params.cParams = ZSTD_adjustCParams_internal(adjusted_cdict_cParams, pledgedSrcSize, - cdict->dictContentSize, ZSTD_cpm_attachDict); - params.cParams.windowLog = windowLog; - params.useRowMatchFinder = cdict->useRowMatchFinder; /* cdict overrides */ - FORWARD_IF_ERROR(ZSTD_resetCCtx_internal(cctx, ¶ms, pledgedSrcSize, - /* loadedDictSize */ 0, - ZSTDcrp_makeClean, zbuff), ""); - assert(cctx->appliedParams.cParams.strategy == adjusted_cdict_cParams.strategy); - } - - { const U32 cdictEnd = (U32)( cdict->matchState.window.nextSrc - - cdict->matchState.window.base); - const U32 cdictLen = cdictEnd - cdict->matchState.window.dictLimit; - if (cdictLen == 0) { - /* don't even attach dictionaries with no contents */ - DEBUGLOG(4, "skipping attaching empty dictionary"); - } else { - DEBUGLOG(4, "attaching dictionary into context"); - cctx->blockState.matchState.dictMatchState = &cdict->matchState; - - /* prep working match state so dict matches never have negative indices - * when they are translated to the working context's index space. */ - if (cctx->blockState.matchState.window.dictLimit < cdictEnd) { - cctx->blockState.matchState.window.nextSrc = - cctx->blockState.matchState.window.base + cdictEnd; - ZSTD_window_clear(&cctx->blockState.matchState.window); - } - /* loadedDictEnd is expressed within the referential of the active context */ - cctx->blockState.matchState.loadedDictEnd = cctx->blockState.matchState.window.dictLimit; - } } - - cctx->dictID = cdict->dictID; - cctx->dictContentSize = cdict->dictContentSize; - - /* copy block state */ - ZSTD_memcpy(cctx->blockState.prevCBlock, &cdict->cBlockState, sizeof(cdict->cBlockState)); - - return 0; -} - -static size_t ZSTD_resetCCtx_byCopyingCDict(ZSTD_CCtx* cctx, - const ZSTD_CDict* cdict, - ZSTD_CCtx_params params, - U64 pledgedSrcSize, - ZSTD_buffered_policy_e zbuff) -{ - const ZSTD_compressionParameters *cdict_cParams = &cdict->matchState.cParams; - - assert(!cdict->matchState.dedicatedDictSearch); - DEBUGLOG(4, "ZSTD_resetCCtx_byCopyingCDict() pledgedSrcSize=%llu", - (unsigned long long)pledgedSrcSize); - - { unsigned const windowLog = params.cParams.windowLog; - assert(windowLog != 0); - /* Copy only compression parameters related to tables. */ - params.cParams = *cdict_cParams; - params.cParams.windowLog = windowLog; - params.useRowMatchFinder = cdict->useRowMatchFinder; - FORWARD_IF_ERROR(ZSTD_resetCCtx_internal(cctx, ¶ms, pledgedSrcSize, - /* loadedDictSize */ 0, - ZSTDcrp_leaveDirty, zbuff), ""); - assert(cctx->appliedParams.cParams.strategy == cdict_cParams->strategy); - assert(cctx->appliedParams.cParams.hashLog == cdict_cParams->hashLog); - assert(cctx->appliedParams.cParams.chainLog == cdict_cParams->chainLog); - } - - ZSTD_cwksp_mark_tables_dirty(&cctx->workspace); - assert(params.useRowMatchFinder != ZSTD_ps_auto); - - /* copy tables */ - { size_t const chainSize = ZSTD_allocateChainTable(cdict_cParams->strategy, cdict->useRowMatchFinder, 0 /* DDS guaranteed disabled */) - ? ((size_t)1 << cdict_cParams->chainLog) - : 0; - size_t const hSize = (size_t)1 << cdict_cParams->hashLog; - - ZSTD_memcpy(cctx->blockState.matchState.hashTable, - cdict->matchState.hashTable, - hSize * sizeof(U32)); - /* Do not copy cdict's chainTable if cctx has parameters such that it would not use chainTable */ - if (ZSTD_allocateChainTable(cctx->appliedParams.cParams.strategy, cctx->appliedParams.useRowMatchFinder, 0 /* forDDSDict */)) { - ZSTD_memcpy(cctx->blockState.matchState.chainTable, - cdict->matchState.chainTable, - chainSize * sizeof(U32)); - } - /* copy tag table */ - if (ZSTD_rowMatchFinderUsed(cdict_cParams->strategy, cdict->useRowMatchFinder)) { - size_t const tagTableSize = hSize*sizeof(U16); - ZSTD_memcpy(cctx->blockState.matchState.tagTable, - cdict->matchState.tagTable, - tagTableSize); - } - } - - /* Zero the hashTable3, since the cdict never fills it */ - { int const h3log = cctx->blockState.matchState.hashLog3; - size_t const h3Size = h3log ? ((size_t)1 << h3log) : 0; - assert(cdict->matchState.hashLog3 == 0); - ZSTD_memset(cctx->blockState.matchState.hashTable3, 0, h3Size * sizeof(U32)); - } - - ZSTD_cwksp_mark_tables_clean(&cctx->workspace); - - /* copy dictionary offsets */ - { ZSTD_matchState_t const* srcMatchState = &cdict->matchState; - ZSTD_matchState_t* dstMatchState = &cctx->blockState.matchState; - dstMatchState->window = srcMatchState->window; - dstMatchState->nextToUpdate = srcMatchState->nextToUpdate; - dstMatchState->loadedDictEnd= srcMatchState->loadedDictEnd; - } - - cctx->dictID = cdict->dictID; - cctx->dictContentSize = cdict->dictContentSize; - - /* copy block state */ - ZSTD_memcpy(cctx->blockState.prevCBlock, &cdict->cBlockState, sizeof(cdict->cBlockState)); - - return 0; -} - -/* We have a choice between copying the dictionary context into the working - * context, or referencing the dictionary context from the working context - * in-place. We decide here which strategy to use. */ -static size_t ZSTD_resetCCtx_usingCDict(ZSTD_CCtx* cctx, - const ZSTD_CDict* cdict, - const ZSTD_CCtx_params* params, - U64 pledgedSrcSize, - ZSTD_buffered_policy_e zbuff) -{ - - DEBUGLOG(4, "ZSTD_resetCCtx_usingCDict (pledgedSrcSize=%u)", - (unsigned)pledgedSrcSize); - - if (ZSTD_shouldAttachDict(cdict, params, pledgedSrcSize)) { - return ZSTD_resetCCtx_byAttachingCDict( - cctx, cdict, *params, pledgedSrcSize, zbuff); - } else { - return ZSTD_resetCCtx_byCopyingCDict( - cctx, cdict, *params, pledgedSrcSize, zbuff); - } -} - -/*! ZSTD_copyCCtx_internal() : - * Duplicate an existing context `srcCCtx` into another one `dstCCtx`. - * Only works during stage ZSTDcs_init (i.e. after creation, but before first call to ZSTD_compressContinue()). - * The "context", in this case, refers to the hash and chain tables, - * entropy tables, and dictionary references. - * `windowLog` value is enforced if != 0, otherwise value is copied from srcCCtx. - * @return : 0, or an error code */ -static size_t ZSTD_copyCCtx_internal(ZSTD_CCtx* dstCCtx, - const ZSTD_CCtx* srcCCtx, - ZSTD_frameParameters fParams, - U64 pledgedSrcSize, - ZSTD_buffered_policy_e zbuff) -{ - RETURN_ERROR_IF(srcCCtx->stage!=ZSTDcs_init, stage_wrong, - "Can't copy a ctx that's not in init stage."); - DEBUGLOG(5, "ZSTD_copyCCtx_internal"); - ZSTD_memcpy(&dstCCtx->customMem, &srcCCtx->customMem, sizeof(ZSTD_customMem)); - { ZSTD_CCtx_params params = dstCCtx->requestedParams; - /* Copy only compression parameters related to tables. */ - params.cParams = srcCCtx->appliedParams.cParams; - assert(srcCCtx->appliedParams.useRowMatchFinder != ZSTD_ps_auto); - assert(srcCCtx->appliedParams.useBlockSplitter != ZSTD_ps_auto); - assert(srcCCtx->appliedParams.ldmParams.enableLdm != ZSTD_ps_auto); - params.useRowMatchFinder = srcCCtx->appliedParams.useRowMatchFinder; - params.useBlockSplitter = srcCCtx->appliedParams.useBlockSplitter; - params.ldmParams = srcCCtx->appliedParams.ldmParams; - params.fParams = fParams; - ZSTD_resetCCtx_internal(dstCCtx, ¶ms, pledgedSrcSize, - /* loadedDictSize */ 0, - ZSTDcrp_leaveDirty, zbuff); - assert(dstCCtx->appliedParams.cParams.windowLog == srcCCtx->appliedParams.cParams.windowLog); - assert(dstCCtx->appliedParams.cParams.strategy == srcCCtx->appliedParams.cParams.strategy); - assert(dstCCtx->appliedParams.cParams.hashLog == srcCCtx->appliedParams.cParams.hashLog); - assert(dstCCtx->appliedParams.cParams.chainLog == srcCCtx->appliedParams.cParams.chainLog); - assert(dstCCtx->blockState.matchState.hashLog3 == srcCCtx->blockState.matchState.hashLog3); - } - - ZSTD_cwksp_mark_tables_dirty(&dstCCtx->workspace); - - /* copy tables */ - { size_t const chainSize = ZSTD_allocateChainTable(srcCCtx->appliedParams.cParams.strategy, - srcCCtx->appliedParams.useRowMatchFinder, - 0 /* forDDSDict */) - ? ((size_t)1 << srcCCtx->appliedParams.cParams.chainLog) - : 0; - size_t const hSize = (size_t)1 << srcCCtx->appliedParams.cParams.hashLog; - int const h3log = srcCCtx->blockState.matchState.hashLog3; - size_t const h3Size = h3log ? ((size_t)1 << h3log) : 0; - - ZSTD_memcpy(dstCCtx->blockState.matchState.hashTable, - srcCCtx->blockState.matchState.hashTable, - hSize * sizeof(U32)); - ZSTD_memcpy(dstCCtx->blockState.matchState.chainTable, - srcCCtx->blockState.matchState.chainTable, - chainSize * sizeof(U32)); - ZSTD_memcpy(dstCCtx->blockState.matchState.hashTable3, - srcCCtx->blockState.matchState.hashTable3, - h3Size * sizeof(U32)); - } - - ZSTD_cwksp_mark_tables_clean(&dstCCtx->workspace); - - /* copy dictionary offsets */ - { - const ZSTD_matchState_t* srcMatchState = &srcCCtx->blockState.matchState; - ZSTD_matchState_t* dstMatchState = &dstCCtx->blockState.matchState; - dstMatchState->window = srcMatchState->window; - dstMatchState->nextToUpdate = srcMatchState->nextToUpdate; - dstMatchState->loadedDictEnd= srcMatchState->loadedDictEnd; - } - dstCCtx->dictID = srcCCtx->dictID; - dstCCtx->dictContentSize = srcCCtx->dictContentSize; - - /* copy block state */ - ZSTD_memcpy(dstCCtx->blockState.prevCBlock, srcCCtx->blockState.prevCBlock, sizeof(*srcCCtx->blockState.prevCBlock)); - - return 0; -} - -/*! ZSTD_copyCCtx() : - * Duplicate an existing context `srcCCtx` into another one `dstCCtx`. - * Only works during stage ZSTDcs_init (i.e. after creation, but before first call to ZSTD_compressContinue()). - * pledgedSrcSize==0 means "unknown". -* @return : 0, or an error code */ -size_t ZSTD_copyCCtx(ZSTD_CCtx* dstCCtx, const ZSTD_CCtx* srcCCtx, unsigned long long pledgedSrcSize) -{ - ZSTD_frameParameters fParams = { 1 /*content*/, 0 /*checksum*/, 0 /*noDictID*/ }; - ZSTD_buffered_policy_e const zbuff = srcCCtx->bufferedPolicy; - ZSTD_STATIC_ASSERT((U32)ZSTDb_buffered==1); - if (pledgedSrcSize==0) pledgedSrcSize = ZSTD_CONTENTSIZE_UNKNOWN; - fParams.contentSizeFlag = (pledgedSrcSize != ZSTD_CONTENTSIZE_UNKNOWN); - - return ZSTD_copyCCtx_internal(dstCCtx, srcCCtx, - fParams, pledgedSrcSize, - zbuff); -} - - -#define ZSTD_ROWSIZE 16 -/*! ZSTD_reduceTable() : - * reduce table indexes by `reducerValue`, or squash to zero. - * PreserveMark preserves "unsorted mark" for btlazy2 strategy. - * It must be set to a clear 0/1 value, to remove branch during inlining. - * Presume table size is a multiple of ZSTD_ROWSIZE - * to help auto-vectorization */ -FORCE_INLINE_TEMPLATE void -ZSTD_reduceTable_internal (U32* const table, U32 const size, U32 const reducerValue, int const preserveMark) -{ - int const nbRows = (int)size / ZSTD_ROWSIZE; - int cellNb = 0; - int rowNb; - /* Protect special index values < ZSTD_WINDOW_START_INDEX. */ - U32 const reducerThreshold = reducerValue + ZSTD_WINDOW_START_INDEX; - assert((size & (ZSTD_ROWSIZE-1)) == 0); /* multiple of ZSTD_ROWSIZE */ - assert(size < (1U<<31)); /* can be casted to int */ - -#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE) - /* To validate that the table re-use logic is sound, and that we don't - * access table space that we haven't cleaned, we re-"poison" the table - * space every time we mark it dirty. - * - * This function however is intended to operate on those dirty tables and - * re-clean them. So when this function is used correctly, we can unpoison - * the memory it operated on. This introduces a blind spot though, since - * if we now try to operate on __actually__ poisoned memory, we will not - * detect that. */ - __msan_unpoison(table, size * sizeof(U32)); -#endif - - for (rowNb=0 ; rowNb < nbRows ; rowNb++) { - int column; - for (column=0; columncParams.hashLog; - ZSTD_reduceTable(ms->hashTable, hSize, reducerValue); - } - - if (ZSTD_allocateChainTable(params->cParams.strategy, params->useRowMatchFinder, (U32)ms->dedicatedDictSearch)) { - U32 const chainSize = (U32)1 << params->cParams.chainLog; - if (params->cParams.strategy == ZSTD_btlazy2) - ZSTD_reduceTable_btlazy2(ms->chainTable, chainSize, reducerValue); - else - ZSTD_reduceTable(ms->chainTable, chainSize, reducerValue); - } - - if (ms->hashLog3) { - U32 const h3Size = (U32)1 << ms->hashLog3; - ZSTD_reduceTable(ms->hashTable3, h3Size, reducerValue); - } -} - - -/*-******************************************************* -* Block entropic compression -*********************************************************/ - -/* See doc/zstd_compression_format.md for detailed format description */ - -void ZSTD_seqToCodes(const seqStore_t* seqStorePtr) -{ - const seqDef* const sequences = seqStorePtr->sequencesStart; - BYTE* const llCodeTable = seqStorePtr->llCode; - BYTE* const ofCodeTable = seqStorePtr->ofCode; - BYTE* const mlCodeTable = seqStorePtr->mlCode; - U32 const nbSeq = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart); - U32 u; - assert(nbSeq <= seqStorePtr->maxNbSeq); - for (u=0; ulongLengthType==ZSTD_llt_literalLength) - llCodeTable[seqStorePtr->longLengthPos] = MaxLL; - if (seqStorePtr->longLengthType==ZSTD_llt_matchLength) - mlCodeTable[seqStorePtr->longLengthPos] = MaxML; -} - -/* ZSTD_useTargetCBlockSize(): - * Returns if target compressed block size param is being used. - * If used, compression will do best effort to make a compressed block size to be around targetCBlockSize. - * Returns 1 if true, 0 otherwise. */ -static int ZSTD_useTargetCBlockSize(const ZSTD_CCtx_params* cctxParams) -{ - DEBUGLOG(5, "ZSTD_useTargetCBlockSize (targetCBlockSize=%zu)", cctxParams->targetCBlockSize); - return (cctxParams->targetCBlockSize != 0); -} - -/* ZSTD_blockSplitterEnabled(): - * Returns if block splitting param is being used - * If used, compression will do best effort to split a block in order to improve compression ratio. - * At the time this function is called, the parameter must be finalized. - * Returns 1 if true, 0 otherwise. */ -static int ZSTD_blockSplitterEnabled(ZSTD_CCtx_params* cctxParams) -{ - DEBUGLOG(5, "ZSTD_blockSplitterEnabled (useBlockSplitter=%d)", cctxParams->useBlockSplitter); - assert(cctxParams->useBlockSplitter != ZSTD_ps_auto); - return (cctxParams->useBlockSplitter == ZSTD_ps_enable); -} - -/* Type returned by ZSTD_buildSequencesStatistics containing finalized symbol encoding types - * and size of the sequences statistics - */ -typedef struct { - U32 LLtype; - U32 Offtype; - U32 MLtype; - size_t size; - size_t lastCountSize; /* Accounts for bug in 1.3.4. More detail in ZSTD_entropyCompressSeqStore_internal() */ -} ZSTD_symbolEncodingTypeStats_t; - -/* ZSTD_buildSequencesStatistics(): - * Returns a ZSTD_symbolEncodingTypeStats_t, or a zstd error code in the `size` field. - * Modifies `nextEntropy` to have the appropriate values as a side effect. - * nbSeq must be greater than 0. - * - * entropyWkspSize must be of size at least ENTROPY_WORKSPACE_SIZE - (MaxSeq + 1)*sizeof(U32) - */ -static ZSTD_symbolEncodingTypeStats_t -ZSTD_buildSequencesStatistics(seqStore_t* seqStorePtr, size_t nbSeq, - const ZSTD_fseCTables_t* prevEntropy, ZSTD_fseCTables_t* nextEntropy, - BYTE* dst, const BYTE* const dstEnd, - ZSTD_strategy strategy, unsigned* countWorkspace, - void* entropyWorkspace, size_t entropyWkspSize) { - BYTE* const ostart = dst; - const BYTE* const oend = dstEnd; - BYTE* op = ostart; - FSE_CTable* CTable_LitLength = nextEntropy->litlengthCTable; - FSE_CTable* CTable_OffsetBits = nextEntropy->offcodeCTable; - FSE_CTable* CTable_MatchLength = nextEntropy->matchlengthCTable; - const BYTE* const ofCodeTable = seqStorePtr->ofCode; - const BYTE* const llCodeTable = seqStorePtr->llCode; - const BYTE* const mlCodeTable = seqStorePtr->mlCode; - ZSTD_symbolEncodingTypeStats_t stats; - - stats.lastCountSize = 0; - /* convert length/distances into codes */ - ZSTD_seqToCodes(seqStorePtr); - assert(op <= oend); - assert(nbSeq != 0); /* ZSTD_selectEncodingType() divides by nbSeq */ - /* build CTable for Literal Lengths */ - { unsigned max = MaxLL; - size_t const mostFrequent = HIST_countFast_wksp(countWorkspace, &max, llCodeTable, nbSeq, entropyWorkspace, entropyWkspSize); /* can't fail */ - DEBUGLOG(5, "Building LL table"); - nextEntropy->litlength_repeatMode = prevEntropy->litlength_repeatMode; - stats.LLtype = ZSTD_selectEncodingType(&nextEntropy->litlength_repeatMode, - countWorkspace, max, mostFrequent, nbSeq, - LLFSELog, prevEntropy->litlengthCTable, - LL_defaultNorm, LL_defaultNormLog, - ZSTD_defaultAllowed, strategy); - assert(set_basic < set_compressed && set_rle < set_compressed); - assert(!(stats.LLtype < set_compressed && nextEntropy->litlength_repeatMode != FSE_repeat_none)); /* We don't copy tables */ - { size_t const countSize = ZSTD_buildCTable( - op, (size_t)(oend - op), - CTable_LitLength, LLFSELog, (symbolEncodingType_e)stats.LLtype, - countWorkspace, max, llCodeTable, nbSeq, - LL_defaultNorm, LL_defaultNormLog, MaxLL, - prevEntropy->litlengthCTable, - sizeof(prevEntropy->litlengthCTable), - entropyWorkspace, entropyWkspSize); - if (ZSTD_isError(countSize)) { - DEBUGLOG(3, "ZSTD_buildCTable for LitLens failed"); - stats.size = countSize; - return stats; - } - if (stats.LLtype == set_compressed) - stats.lastCountSize = countSize; - op += countSize; - assert(op <= oend); - } } - /* build CTable for Offsets */ - { unsigned max = MaxOff; - size_t const mostFrequent = HIST_countFast_wksp( - countWorkspace, &max, ofCodeTable, nbSeq, entropyWorkspace, entropyWkspSize); /* can't fail */ - /* We can only use the basic table if max <= DefaultMaxOff, otherwise the offsets are too large */ - ZSTD_defaultPolicy_e const defaultPolicy = (max <= DefaultMaxOff) ? ZSTD_defaultAllowed : ZSTD_defaultDisallowed; - DEBUGLOG(5, "Building OF table"); - nextEntropy->offcode_repeatMode = prevEntropy->offcode_repeatMode; - stats.Offtype = ZSTD_selectEncodingType(&nextEntropy->offcode_repeatMode, - countWorkspace, max, mostFrequent, nbSeq, - OffFSELog, prevEntropy->offcodeCTable, - OF_defaultNorm, OF_defaultNormLog, - defaultPolicy, strategy); - assert(!(stats.Offtype < set_compressed && nextEntropy->offcode_repeatMode != FSE_repeat_none)); /* We don't copy tables */ - { size_t const countSize = ZSTD_buildCTable( - op, (size_t)(oend - op), - CTable_OffsetBits, OffFSELog, (symbolEncodingType_e)stats.Offtype, - countWorkspace, max, ofCodeTable, nbSeq, - OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff, - prevEntropy->offcodeCTable, - sizeof(prevEntropy->offcodeCTable), - entropyWorkspace, entropyWkspSize); - if (ZSTD_isError(countSize)) { - DEBUGLOG(3, "ZSTD_buildCTable for Offsets failed"); - stats.size = countSize; - return stats; - } - if (stats.Offtype == set_compressed) - stats.lastCountSize = countSize; - op += countSize; - assert(op <= oend); - } } - /* build CTable for MatchLengths */ - { unsigned max = MaxML; - size_t const mostFrequent = HIST_countFast_wksp( - countWorkspace, &max, mlCodeTable, nbSeq, entropyWorkspace, entropyWkspSize); /* can't fail */ - DEBUGLOG(5, "Building ML table (remaining space : %i)", (int)(oend-op)); - nextEntropy->matchlength_repeatMode = prevEntropy->matchlength_repeatMode; - stats.MLtype = ZSTD_selectEncodingType(&nextEntropy->matchlength_repeatMode, - countWorkspace, max, mostFrequent, nbSeq, - MLFSELog, prevEntropy->matchlengthCTable, - ML_defaultNorm, ML_defaultNormLog, - ZSTD_defaultAllowed, strategy); - assert(!(stats.MLtype < set_compressed && nextEntropy->matchlength_repeatMode != FSE_repeat_none)); /* We don't copy tables */ - { size_t const countSize = ZSTD_buildCTable( - op, (size_t)(oend - op), - CTable_MatchLength, MLFSELog, (symbolEncodingType_e)stats.MLtype, - countWorkspace, max, mlCodeTable, nbSeq, - ML_defaultNorm, ML_defaultNormLog, MaxML, - prevEntropy->matchlengthCTable, - sizeof(prevEntropy->matchlengthCTable), - entropyWorkspace, entropyWkspSize); - if (ZSTD_isError(countSize)) { - DEBUGLOG(3, "ZSTD_buildCTable for MatchLengths failed"); - stats.size = countSize; - return stats; - } - if (stats.MLtype == set_compressed) - stats.lastCountSize = countSize; - op += countSize; - assert(op <= oend); - } } - stats.size = (size_t)(op-ostart); - return stats; -} - -/* ZSTD_entropyCompressSeqStore_internal(): - * compresses both literals and sequences - * Returns compressed size of block, or a zstd error. - */ -#define SUSPECT_UNCOMPRESSIBLE_LITERAL_RATIO 20 -MEM_STATIC size_t -ZSTD_entropyCompressSeqStore_internal(seqStore_t* seqStorePtr, - const ZSTD_entropyCTables_t* prevEntropy, - ZSTD_entropyCTables_t* nextEntropy, - const ZSTD_CCtx_params* cctxParams, - void* dst, size_t dstCapacity, - void* entropyWorkspace, size_t entropyWkspSize, - const int bmi2) -{ - const int longOffsets = cctxParams->cParams.windowLog > STREAM_ACCUMULATOR_MIN; - ZSTD_strategy const strategy = cctxParams->cParams.strategy; - unsigned* count = (unsigned*)entropyWorkspace; - FSE_CTable* CTable_LitLength = nextEntropy->fse.litlengthCTable; - FSE_CTable* CTable_OffsetBits = nextEntropy->fse.offcodeCTable; - FSE_CTable* CTable_MatchLength = nextEntropy->fse.matchlengthCTable; - const seqDef* const sequences = seqStorePtr->sequencesStart; - const size_t nbSeq = seqStorePtr->sequences - seqStorePtr->sequencesStart; - const BYTE* const ofCodeTable = seqStorePtr->ofCode; - const BYTE* const llCodeTable = seqStorePtr->llCode; - const BYTE* const mlCodeTable = seqStorePtr->mlCode; - BYTE* const ostart = (BYTE*)dst; - BYTE* const oend = ostart + dstCapacity; - BYTE* op = ostart; - size_t lastCountSize; - - entropyWorkspace = count + (MaxSeq + 1); - entropyWkspSize -= (MaxSeq + 1) * sizeof(*count); - - DEBUGLOG(4, "ZSTD_entropyCompressSeqStore_internal (nbSeq=%zu)", nbSeq); - ZSTD_STATIC_ASSERT(HUF_WORKSPACE_SIZE >= (1<= HUF_WORKSPACE_SIZE); - - /* Compress literals */ - { const BYTE* const literals = seqStorePtr->litStart; - size_t const numSequences = seqStorePtr->sequences - seqStorePtr->sequencesStart; - size_t const numLiterals = seqStorePtr->lit - seqStorePtr->litStart; - /* Base suspicion of uncompressibility on ratio of literals to sequences */ - unsigned const suspectUncompressible = (numSequences == 0) || (numLiterals / numSequences >= SUSPECT_UNCOMPRESSIBLE_LITERAL_RATIO); - size_t const litSize = (size_t)(seqStorePtr->lit - literals); - size_t const cSize = ZSTD_compressLiterals( - &prevEntropy->huf, &nextEntropy->huf, - cctxParams->cParams.strategy, - ZSTD_literalsCompressionIsDisabled(cctxParams), - op, dstCapacity, - literals, litSize, - entropyWorkspace, entropyWkspSize, - bmi2, suspectUncompressible); - FORWARD_IF_ERROR(cSize, "ZSTD_compressLiterals failed"); - assert(cSize <= dstCapacity); - op += cSize; - } - - /* Sequences Header */ - RETURN_ERROR_IF((oend-op) < 3 /*max nbSeq Size*/ + 1 /*seqHead*/, - dstSize_tooSmall, "Can't fit seq hdr in output buf!"); - if (nbSeq < 128) { - *op++ = (BYTE)nbSeq; - } else if (nbSeq < LONGNBSEQ) { - op[0] = (BYTE)((nbSeq>>8) + 0x80); - op[1] = (BYTE)nbSeq; - op+=2; - } else { - op[0]=0xFF; - MEM_writeLE16(op+1, (U16)(nbSeq - LONGNBSEQ)); - op+=3; - } - assert(op <= oend); - if (nbSeq==0) { - /* Copy the old tables over as if we repeated them */ - ZSTD_memcpy(&nextEntropy->fse, &prevEntropy->fse, sizeof(prevEntropy->fse)); - return (size_t)(op - ostart); - } - { - ZSTD_symbolEncodingTypeStats_t stats; - BYTE* seqHead = op++; - /* build stats for sequences */ - stats = ZSTD_buildSequencesStatistics(seqStorePtr, nbSeq, - &prevEntropy->fse, &nextEntropy->fse, - op, oend, - strategy, count, - entropyWorkspace, entropyWkspSize); - FORWARD_IF_ERROR(stats.size, "ZSTD_buildSequencesStatistics failed!"); - *seqHead = (BYTE)((stats.LLtype<<6) + (stats.Offtype<<4) + (stats.MLtype<<2)); - lastCountSize = stats.lastCountSize; - op += stats.size; - } - - { size_t const bitstreamSize = ZSTD_encodeSequences( - op, (size_t)(oend - op), - CTable_MatchLength, mlCodeTable, - CTable_OffsetBits, ofCodeTable, - CTable_LitLength, llCodeTable, - sequences, nbSeq, - longOffsets, bmi2); - FORWARD_IF_ERROR(bitstreamSize, "ZSTD_encodeSequences failed"); - op += bitstreamSize; - assert(op <= oend); - /* zstd versions <= 1.3.4 mistakenly report corruption when - * FSE_readNCount() receives a buffer < 4 bytes. - * Fixed by https://github.com/facebook/zstd/pull/1146. - * This can happen when the last set_compressed table present is 2 - * bytes and the bitstream is only one byte. - * In this exceedingly rare case, we will simply emit an uncompressed - * block, since it isn't worth optimizing. - */ - if (lastCountSize && (lastCountSize + bitstreamSize) < 4) { - /* lastCountSize >= 2 && bitstreamSize > 0 ==> lastCountSize == 3 */ - assert(lastCountSize + bitstreamSize == 3); - DEBUGLOG(5, "Avoiding bug in zstd decoder in versions <= 1.3.4 by " - "emitting an uncompressed block."); - return 0; - } - } - - DEBUGLOG(5, "compressed block size : %u", (unsigned)(op - ostart)); - return (size_t)(op - ostart); -} - -MEM_STATIC size_t -ZSTD_entropyCompressSeqStore(seqStore_t* seqStorePtr, - const ZSTD_entropyCTables_t* prevEntropy, - ZSTD_entropyCTables_t* nextEntropy, - const ZSTD_CCtx_params* cctxParams, - void* dst, size_t dstCapacity, - size_t srcSize, - void* entropyWorkspace, size_t entropyWkspSize, - int bmi2) -{ - size_t const cSize = ZSTD_entropyCompressSeqStore_internal( - seqStorePtr, prevEntropy, nextEntropy, cctxParams, - dst, dstCapacity, - entropyWorkspace, entropyWkspSize, bmi2); - if (cSize == 0) return 0; - /* When srcSize <= dstCapacity, there is enough space to write a raw uncompressed block. - * Since we ran out of space, block must be not compressible, so fall back to raw uncompressed block. - */ - if ((cSize == ERROR(dstSize_tooSmall)) & (srcSize <= dstCapacity)) - return 0; /* block not compressed */ - FORWARD_IF_ERROR(cSize, "ZSTD_entropyCompressSeqStore_internal failed"); - - /* Check compressibility */ - { size_t const maxCSize = srcSize - ZSTD_minGain(srcSize, cctxParams->cParams.strategy); - if (cSize >= maxCSize) return 0; /* block not compressed */ - } - DEBUGLOG(4, "ZSTD_entropyCompressSeqStore() cSize: %zu", cSize); - return cSize; -} - -/* ZSTD_selectBlockCompressor() : - * Not static, but internal use only (used by long distance matcher) - * assumption : strat is a valid strategy */ -ZSTD_blockCompressor ZSTD_selectBlockCompressor(ZSTD_strategy strat, ZSTD_paramSwitch_e useRowMatchFinder, ZSTD_dictMode_e dictMode) -{ - static const ZSTD_blockCompressor blockCompressor[4][ZSTD_STRATEGY_MAX+1] = { - { ZSTD_compressBlock_fast /* default for 0 */, - ZSTD_compressBlock_fast, - ZSTD_compressBlock_doubleFast, - ZSTD_compressBlock_greedy, - ZSTD_compressBlock_lazy, - ZSTD_compressBlock_lazy2, - ZSTD_compressBlock_btlazy2, - ZSTD_compressBlock_btopt, - ZSTD_compressBlock_btultra, - ZSTD_compressBlock_btultra2 }, - { ZSTD_compressBlock_fast_extDict /* default for 0 */, - ZSTD_compressBlock_fast_extDict, - ZSTD_compressBlock_doubleFast_extDict, - ZSTD_compressBlock_greedy_extDict, - ZSTD_compressBlock_lazy_extDict, - ZSTD_compressBlock_lazy2_extDict, - ZSTD_compressBlock_btlazy2_extDict, - ZSTD_compressBlock_btopt_extDict, - ZSTD_compressBlock_btultra_extDict, - ZSTD_compressBlock_btultra_extDict }, - { ZSTD_compressBlock_fast_dictMatchState /* default for 0 */, - ZSTD_compressBlock_fast_dictMatchState, - ZSTD_compressBlock_doubleFast_dictMatchState, - ZSTD_compressBlock_greedy_dictMatchState, - ZSTD_compressBlock_lazy_dictMatchState, - ZSTD_compressBlock_lazy2_dictMatchState, - ZSTD_compressBlock_btlazy2_dictMatchState, - ZSTD_compressBlock_btopt_dictMatchState, - ZSTD_compressBlock_btultra_dictMatchState, - ZSTD_compressBlock_btultra_dictMatchState }, - { NULL /* default for 0 */, - NULL, - NULL, - ZSTD_compressBlock_greedy_dedicatedDictSearch, - ZSTD_compressBlock_lazy_dedicatedDictSearch, - ZSTD_compressBlock_lazy2_dedicatedDictSearch, - NULL, - NULL, - NULL, - NULL } - }; - ZSTD_blockCompressor selectedCompressor; - ZSTD_STATIC_ASSERT((unsigned)ZSTD_fast == 1); - - assert(ZSTD_cParam_withinBounds(ZSTD_c_strategy, strat)); - DEBUGLOG(4, "Selected block compressor: dictMode=%d strat=%d rowMatchfinder=%d", (int)dictMode, (int)strat, (int)useRowMatchFinder); - if (ZSTD_rowMatchFinderUsed(strat, useRowMatchFinder)) { - static const ZSTD_blockCompressor rowBasedBlockCompressors[4][3] = { - { ZSTD_compressBlock_greedy_row, - ZSTD_compressBlock_lazy_row, - ZSTD_compressBlock_lazy2_row }, - { ZSTD_compressBlock_greedy_extDict_row, - ZSTD_compressBlock_lazy_extDict_row, - ZSTD_compressBlock_lazy2_extDict_row }, - { ZSTD_compressBlock_greedy_dictMatchState_row, - ZSTD_compressBlock_lazy_dictMatchState_row, - ZSTD_compressBlock_lazy2_dictMatchState_row }, - { ZSTD_compressBlock_greedy_dedicatedDictSearch_row, - ZSTD_compressBlock_lazy_dedicatedDictSearch_row, - ZSTD_compressBlock_lazy2_dedicatedDictSearch_row } - }; - DEBUGLOG(4, "Selecting a row-based matchfinder"); - assert(useRowMatchFinder != ZSTD_ps_auto); - selectedCompressor = rowBasedBlockCompressors[(int)dictMode][(int)strat - (int)ZSTD_greedy]; - } else { - selectedCompressor = blockCompressor[(int)dictMode][(int)strat]; - } - assert(selectedCompressor != NULL); - return selectedCompressor; -} - -static void ZSTD_storeLastLiterals(seqStore_t* seqStorePtr, - const BYTE* anchor, size_t lastLLSize) -{ - ZSTD_memcpy(seqStorePtr->lit, anchor, lastLLSize); - seqStorePtr->lit += lastLLSize; -} - -void ZSTD_resetSeqStore(seqStore_t* ssPtr) -{ - ssPtr->lit = ssPtr->litStart; - ssPtr->sequences = ssPtr->sequencesStart; - ssPtr->longLengthType = ZSTD_llt_none; -} - -typedef enum { ZSTDbss_compress, ZSTDbss_noCompress } ZSTD_buildSeqStore_e; - -static size_t ZSTD_buildSeqStore(ZSTD_CCtx* zc, const void* src, size_t srcSize) -{ - ZSTD_matchState_t* const ms = &zc->blockState.matchState; - DEBUGLOG(5, "ZSTD_buildSeqStore (srcSize=%zu)", srcSize); - assert(srcSize <= ZSTD_BLOCKSIZE_MAX); - /* Assert that we have correctly flushed the ctx params into the ms's copy */ - ZSTD_assertEqualCParams(zc->appliedParams.cParams, ms->cParams); - if (srcSize < MIN_CBLOCK_SIZE+ZSTD_blockHeaderSize+1) { - if (zc->appliedParams.cParams.strategy >= ZSTD_btopt) { - ZSTD_ldm_skipRawSeqStoreBytes(&zc->externSeqStore, srcSize); - } else { - ZSTD_ldm_skipSequences(&zc->externSeqStore, srcSize, zc->appliedParams.cParams.minMatch); - } - return ZSTDbss_noCompress; /* don't even attempt compression below a certain srcSize */ - } - ZSTD_resetSeqStore(&(zc->seqStore)); - /* required for optimal parser to read stats from dictionary */ - ms->opt.symbolCosts = &zc->blockState.prevCBlock->entropy; - /* tell the optimal parser how we expect to compress literals */ - ms->opt.literalCompressionMode = zc->appliedParams.literalCompressionMode; - /* a gap between an attached dict and the current window is not safe, - * they must remain adjacent, - * and when that stops being the case, the dict must be unset */ - assert(ms->dictMatchState == NULL || ms->loadedDictEnd == ms->window.dictLimit); - - /* limited update after a very long match */ - { const BYTE* const base = ms->window.base; - const BYTE* const istart = (const BYTE*)src; - const U32 curr = (U32)(istart-base); - if (sizeof(ptrdiff_t)==8) assert(istart - base < (ptrdiff_t)(U32)(-1)); /* ensure no overflow */ - if (curr > ms->nextToUpdate + 384) - ms->nextToUpdate = curr - MIN(192, (U32)(curr - ms->nextToUpdate - 384)); - } - - /* select and store sequences */ - { ZSTD_dictMode_e const dictMode = ZSTD_matchState_dictMode(ms); - size_t lastLLSize; - { int i; - for (i = 0; i < ZSTD_REP_NUM; ++i) - zc->blockState.nextCBlock->rep[i] = zc->blockState.prevCBlock->rep[i]; - } - if (zc->externSeqStore.pos < zc->externSeqStore.size) { - assert(zc->appliedParams.ldmParams.enableLdm == ZSTD_ps_disable); - /* Updates ldmSeqStore.pos */ - lastLLSize = - ZSTD_ldm_blockCompress(&zc->externSeqStore, - ms, &zc->seqStore, - zc->blockState.nextCBlock->rep, - zc->appliedParams.useRowMatchFinder, - src, srcSize); - assert(zc->externSeqStore.pos <= zc->externSeqStore.size); - } else if (zc->appliedParams.ldmParams.enableLdm == ZSTD_ps_enable) { - rawSeqStore_t ldmSeqStore = kNullRawSeqStore; - - ldmSeqStore.seq = zc->ldmSequences; - ldmSeqStore.capacity = zc->maxNbLdmSequences; - /* Updates ldmSeqStore.size */ - FORWARD_IF_ERROR(ZSTD_ldm_generateSequences(&zc->ldmState, &ldmSeqStore, - &zc->appliedParams.ldmParams, - src, srcSize), ""); - /* Updates ldmSeqStore.pos */ - lastLLSize = - ZSTD_ldm_blockCompress(&ldmSeqStore, - ms, &zc->seqStore, - zc->blockState.nextCBlock->rep, - zc->appliedParams.useRowMatchFinder, - src, srcSize); - assert(ldmSeqStore.pos == ldmSeqStore.size); - } else { /* not long range mode */ - ZSTD_blockCompressor const blockCompressor = ZSTD_selectBlockCompressor(zc->appliedParams.cParams.strategy, - zc->appliedParams.useRowMatchFinder, - dictMode); - ms->ldmSeqStore = NULL; - lastLLSize = blockCompressor(ms, &zc->seqStore, zc->blockState.nextCBlock->rep, src, srcSize); - } - { const BYTE* const lastLiterals = (const BYTE*)src + srcSize - lastLLSize; - ZSTD_storeLastLiterals(&zc->seqStore, lastLiterals, lastLLSize); - } } - return ZSTDbss_compress; -} - -static void ZSTD_copyBlockSequences(ZSTD_CCtx* zc) -{ - const seqStore_t* seqStore = ZSTD_getSeqStore(zc); - const seqDef* seqStoreSeqs = seqStore->sequencesStart; - size_t seqStoreSeqSize = seqStore->sequences - seqStoreSeqs; - size_t seqStoreLiteralsSize = (size_t)(seqStore->lit - seqStore->litStart); - size_t literalsRead = 0; - size_t lastLLSize; - - ZSTD_Sequence* outSeqs = &zc->seqCollector.seqStart[zc->seqCollector.seqIndex]; - size_t i; - repcodes_t updatedRepcodes; - - assert(zc->seqCollector.seqIndex + 1 < zc->seqCollector.maxSequences); - /* Ensure we have enough space for last literals "sequence" */ - assert(zc->seqCollector.maxSequences >= seqStoreSeqSize + 1); - ZSTD_memcpy(updatedRepcodes.rep, zc->blockState.prevCBlock->rep, sizeof(repcodes_t)); - for (i = 0; i < seqStoreSeqSize; ++i) { - U32 rawOffset = seqStoreSeqs[i].offBase - ZSTD_REP_NUM; - outSeqs[i].litLength = seqStoreSeqs[i].litLength; - outSeqs[i].matchLength = seqStoreSeqs[i].mlBase + MINMATCH; - outSeqs[i].rep = 0; - - if (i == seqStore->longLengthPos) { - if (seqStore->longLengthType == ZSTD_llt_literalLength) { - outSeqs[i].litLength += 0x10000; - } else if (seqStore->longLengthType == ZSTD_llt_matchLength) { - outSeqs[i].matchLength += 0x10000; - } - } - - if (seqStoreSeqs[i].offBase <= ZSTD_REP_NUM) { - /* Derive the correct offset corresponding to a repcode */ - outSeqs[i].rep = seqStoreSeqs[i].offBase; - if (outSeqs[i].litLength != 0) { - rawOffset = updatedRepcodes.rep[outSeqs[i].rep - 1]; - } else { - if (outSeqs[i].rep == 3) { - rawOffset = updatedRepcodes.rep[0] - 1; - } else { - rawOffset = updatedRepcodes.rep[outSeqs[i].rep]; - } - } - } - outSeqs[i].offset = rawOffset; - /* seqStoreSeqs[i].offset == offCode+1, and ZSTD_updateRep() expects offCode - so we provide seqStoreSeqs[i].offset - 1 */ - ZSTD_updateRep(updatedRepcodes.rep, - seqStoreSeqs[i].offBase - 1, - seqStoreSeqs[i].litLength == 0); - literalsRead += outSeqs[i].litLength; - } - /* Insert last literals (if any exist) in the block as a sequence with ml == off == 0. - * If there are no last literals, then we'll emit (of: 0, ml: 0, ll: 0), which is a marker - * for the block boundary, according to the API. - */ - assert(seqStoreLiteralsSize >= literalsRead); - lastLLSize = seqStoreLiteralsSize - literalsRead; - outSeqs[i].litLength = (U32)lastLLSize; - outSeqs[i].matchLength = outSeqs[i].offset = outSeqs[i].rep = 0; - seqStoreSeqSize++; - zc->seqCollector.seqIndex += seqStoreSeqSize; -} - -size_t ZSTD_generateSequences(ZSTD_CCtx* zc, ZSTD_Sequence* outSeqs, - size_t outSeqsSize, const void* src, size_t srcSize) -{ - const size_t dstCapacity = ZSTD_compressBound(srcSize); - void* dst = ZSTD_customMalloc(dstCapacity, ZSTD_defaultCMem); - SeqCollector seqCollector; - - RETURN_ERROR_IF(dst == NULL, memory_allocation, "NULL pointer!"); - - seqCollector.collectSequences = 1; - seqCollector.seqStart = outSeqs; - seqCollector.seqIndex = 0; - seqCollector.maxSequences = outSeqsSize; - zc->seqCollector = seqCollector; - - ZSTD_compress2(zc, dst, dstCapacity, src, srcSize); - ZSTD_customFree(dst, ZSTD_defaultCMem); - return zc->seqCollector.seqIndex; -} - -size_t ZSTD_mergeBlockDelimiters(ZSTD_Sequence* sequences, size_t seqsSize) { - size_t in = 0; - size_t out = 0; - for (; in < seqsSize; ++in) { - if (sequences[in].offset == 0 && sequences[in].matchLength == 0) { - if (in != seqsSize - 1) { - sequences[in+1].litLength += sequences[in].litLength; - } - } else { - sequences[out] = sequences[in]; - ++out; - } - } - return out; -} - -/* Unrolled loop to read four size_ts of input at a time. Returns 1 if is RLE, 0 if not. */ -static int ZSTD_isRLE(const BYTE* src, size_t length) { - const BYTE* ip = src; - const BYTE value = ip[0]; - const size_t valueST = (size_t)((U64)value * 0x0101010101010101ULL); - const size_t unrollSize = sizeof(size_t) * 4; - const size_t unrollMask = unrollSize - 1; - const size_t prefixLength = length & unrollMask; - size_t i; - size_t u; - if (length == 1) return 1; - /* Check if prefix is RLE first before using unrolled loop */ - if (prefixLength && ZSTD_count(ip+1, ip, ip+prefixLength) != prefixLength-1) { - return 0; - } - for (i = prefixLength; i != length; i += unrollSize) { - for (u = 0; u < unrollSize; u += sizeof(size_t)) { - if (MEM_readST(ip + i + u) != valueST) { - return 0; - } - } - } - return 1; -} - -/* Returns true if the given block may be RLE. - * This is just a heuristic based on the compressibility. - * It may return both false positives and false negatives. - */ -static int ZSTD_maybeRLE(seqStore_t const* seqStore) -{ - size_t const nbSeqs = (size_t)(seqStore->sequences - seqStore->sequencesStart); - size_t const nbLits = (size_t)(seqStore->lit - seqStore->litStart); - - return nbSeqs < 4 && nbLits < 10; -} - -static void ZSTD_blockState_confirmRepcodesAndEntropyTables(ZSTD_blockState_t* const bs) -{ - ZSTD_compressedBlockState_t* const tmp = bs->prevCBlock; - bs->prevCBlock = bs->nextCBlock; - bs->nextCBlock = tmp; -} - -/* Writes the block header */ -static void writeBlockHeader(void* op, size_t cSize, size_t blockSize, U32 lastBlock) { - U32 const cBlockHeader = cSize == 1 ? - lastBlock + (((U32)bt_rle)<<1) + (U32)(blockSize << 3) : - lastBlock + (((U32)bt_compressed)<<1) + (U32)(cSize << 3); - MEM_writeLE24(op, cBlockHeader); - DEBUGLOG(3, "writeBlockHeader: cSize: %zu blockSize: %zu lastBlock: %u", cSize, blockSize, lastBlock); -} - -/** ZSTD_buildBlockEntropyStats_literals() : - * Builds entropy for the literals. - * Stores literals block type (raw, rle, compressed, repeat) and - * huffman description table to hufMetadata. - * Requires ENTROPY_WORKSPACE_SIZE workspace - * @return : size of huffman description table or error code */ -static size_t ZSTD_buildBlockEntropyStats_literals(void* const src, size_t srcSize, - const ZSTD_hufCTables_t* prevHuf, - ZSTD_hufCTables_t* nextHuf, - ZSTD_hufCTablesMetadata_t* hufMetadata, - const int literalsCompressionIsDisabled, - void* workspace, size_t wkspSize) -{ - BYTE* const wkspStart = (BYTE*)workspace; - BYTE* const wkspEnd = wkspStart + wkspSize; - BYTE* const countWkspStart = wkspStart; - unsigned* const countWksp = (unsigned*)workspace; - const size_t countWkspSize = (HUF_SYMBOLVALUE_MAX + 1) * sizeof(unsigned); - BYTE* const nodeWksp = countWkspStart + countWkspSize; - const size_t nodeWkspSize = wkspEnd-nodeWksp; - unsigned maxSymbolValue = HUF_SYMBOLVALUE_MAX; - unsigned huffLog = HUF_TABLELOG_DEFAULT; - HUF_repeat repeat = prevHuf->repeatMode; - DEBUGLOG(5, "ZSTD_buildBlockEntropyStats_literals (srcSize=%zu)", srcSize); - - /* Prepare nextEntropy assuming reusing the existing table */ - ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); - - if (literalsCompressionIsDisabled) { - DEBUGLOG(5, "set_basic - disabled"); - hufMetadata->hType = set_basic; - return 0; - } - - /* small ? don't even attempt compression (speed opt) */ -#ifndef COMPRESS_LITERALS_SIZE_MIN -#define COMPRESS_LITERALS_SIZE_MIN 63 -#endif - { size_t const minLitSize = (prevHuf->repeatMode == HUF_repeat_valid) ? 6 : COMPRESS_LITERALS_SIZE_MIN; - if (srcSize <= minLitSize) { - DEBUGLOG(5, "set_basic - too small"); - hufMetadata->hType = set_basic; - return 0; - } - } - - /* Scan input and build symbol stats */ - { size_t const largest = HIST_count_wksp (countWksp, &maxSymbolValue, (const BYTE*)src, srcSize, workspace, wkspSize); - FORWARD_IF_ERROR(largest, "HIST_count_wksp failed"); - if (largest == srcSize) { - DEBUGLOG(5, "set_rle"); - hufMetadata->hType = set_rle; - return 0; - } - if (largest <= (srcSize >> 7)+4) { - DEBUGLOG(5, "set_basic - no gain"); - hufMetadata->hType = set_basic; - return 0; - } - } - - /* Validate the previous Huffman table */ - if (repeat == HUF_repeat_check && !HUF_validateCTable((HUF_CElt const*)prevHuf->CTable, countWksp, maxSymbolValue)) { - repeat = HUF_repeat_none; - } - - /* Build Huffman Tree */ - ZSTD_memset(nextHuf->CTable, 0, sizeof(nextHuf->CTable)); - huffLog = HUF_optimalTableLog(huffLog, srcSize, maxSymbolValue); - { size_t const maxBits = HUF_buildCTable_wksp((HUF_CElt*)nextHuf->CTable, countWksp, - maxSymbolValue, huffLog, - nodeWksp, nodeWkspSize); - FORWARD_IF_ERROR(maxBits, "HUF_buildCTable_wksp"); - huffLog = (U32)maxBits; - { /* Build and write the CTable */ - size_t const newCSize = HUF_estimateCompressedSize( - (HUF_CElt*)nextHuf->CTable, countWksp, maxSymbolValue); - size_t const hSize = HUF_writeCTable_wksp( - hufMetadata->hufDesBuffer, sizeof(hufMetadata->hufDesBuffer), - (HUF_CElt*)nextHuf->CTable, maxSymbolValue, huffLog, - nodeWksp, nodeWkspSize); - /* Check against repeating the previous CTable */ - if (repeat != HUF_repeat_none) { - size_t const oldCSize = HUF_estimateCompressedSize( - (HUF_CElt const*)prevHuf->CTable, countWksp, maxSymbolValue); - if (oldCSize < srcSize && (oldCSize <= hSize + newCSize || hSize + 12 >= srcSize)) { - DEBUGLOG(5, "set_repeat - smaller"); - ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); - hufMetadata->hType = set_repeat; - return 0; - } - } - if (newCSize + hSize >= srcSize) { - DEBUGLOG(5, "set_basic - no gains"); - ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); - hufMetadata->hType = set_basic; - return 0; - } - DEBUGLOG(5, "set_compressed (hSize=%u)", (U32)hSize); - hufMetadata->hType = set_compressed; - nextHuf->repeatMode = HUF_repeat_check; - return hSize; - } - } -} - - -/* ZSTD_buildDummySequencesStatistics(): - * Returns a ZSTD_symbolEncodingTypeStats_t with all encoding types as set_basic, - * and updates nextEntropy to the appropriate repeatMode. - */ -static ZSTD_symbolEncodingTypeStats_t -ZSTD_buildDummySequencesStatistics(ZSTD_fseCTables_t* nextEntropy) { - ZSTD_symbolEncodingTypeStats_t stats = {set_basic, set_basic, set_basic, 0, 0}; - nextEntropy->litlength_repeatMode = FSE_repeat_none; - nextEntropy->offcode_repeatMode = FSE_repeat_none; - nextEntropy->matchlength_repeatMode = FSE_repeat_none; - return stats; -} - -/** ZSTD_buildBlockEntropyStats_sequences() : - * Builds entropy for the sequences. - * Stores symbol compression modes and fse table to fseMetadata. - * Requires ENTROPY_WORKSPACE_SIZE wksp. - * @return : size of fse tables or error code */ -static size_t ZSTD_buildBlockEntropyStats_sequences(seqStore_t* seqStorePtr, - const ZSTD_fseCTables_t* prevEntropy, - ZSTD_fseCTables_t* nextEntropy, - const ZSTD_CCtx_params* cctxParams, - ZSTD_fseCTablesMetadata_t* fseMetadata, - void* workspace, size_t wkspSize) -{ - ZSTD_strategy const strategy = cctxParams->cParams.strategy; - size_t const nbSeq = seqStorePtr->sequences - seqStorePtr->sequencesStart; - BYTE* const ostart = fseMetadata->fseTablesBuffer; - BYTE* const oend = ostart + sizeof(fseMetadata->fseTablesBuffer); - BYTE* op = ostart; - unsigned* countWorkspace = (unsigned*)workspace; - unsigned* entropyWorkspace = countWorkspace + (MaxSeq + 1); - size_t entropyWorkspaceSize = wkspSize - (MaxSeq + 1) * sizeof(*countWorkspace); - ZSTD_symbolEncodingTypeStats_t stats; - - DEBUGLOG(5, "ZSTD_buildBlockEntropyStats_sequences (nbSeq=%zu)", nbSeq); - stats = nbSeq != 0 ? ZSTD_buildSequencesStatistics(seqStorePtr, nbSeq, - prevEntropy, nextEntropy, op, oend, - strategy, countWorkspace, - entropyWorkspace, entropyWorkspaceSize) - : ZSTD_buildDummySequencesStatistics(nextEntropy); - FORWARD_IF_ERROR(stats.size, "ZSTD_buildSequencesStatistics failed!"); - fseMetadata->llType = (symbolEncodingType_e) stats.LLtype; - fseMetadata->ofType = (symbolEncodingType_e) stats.Offtype; - fseMetadata->mlType = (symbolEncodingType_e) stats.MLtype; - fseMetadata->lastCountSize = stats.lastCountSize; - return stats.size; -} - - -/** ZSTD_buildBlockEntropyStats() : - * Builds entropy for the block. - * Requires workspace size ENTROPY_WORKSPACE_SIZE - * - * @return : 0 on success or error code - */ -size_t ZSTD_buildBlockEntropyStats(seqStore_t* seqStorePtr, - const ZSTD_entropyCTables_t* prevEntropy, - ZSTD_entropyCTables_t* nextEntropy, - const ZSTD_CCtx_params* cctxParams, - ZSTD_entropyCTablesMetadata_t* entropyMetadata, - void* workspace, size_t wkspSize) -{ - size_t const litSize = seqStorePtr->lit - seqStorePtr->litStart; - entropyMetadata->hufMetadata.hufDesSize = - ZSTD_buildBlockEntropyStats_literals(seqStorePtr->litStart, litSize, - &prevEntropy->huf, &nextEntropy->huf, - &entropyMetadata->hufMetadata, - ZSTD_literalsCompressionIsDisabled(cctxParams), - workspace, wkspSize); - FORWARD_IF_ERROR(entropyMetadata->hufMetadata.hufDesSize, "ZSTD_buildBlockEntropyStats_literals failed"); - entropyMetadata->fseMetadata.fseTablesSize = - ZSTD_buildBlockEntropyStats_sequences(seqStorePtr, - &prevEntropy->fse, &nextEntropy->fse, - cctxParams, - &entropyMetadata->fseMetadata, - workspace, wkspSize); - FORWARD_IF_ERROR(entropyMetadata->fseMetadata.fseTablesSize, "ZSTD_buildBlockEntropyStats_sequences failed"); - return 0; -} - -/* Returns the size estimate for the literals section (header + content) of a block */ -static size_t ZSTD_estimateBlockSize_literal(const BYTE* literals, size_t litSize, - const ZSTD_hufCTables_t* huf, - const ZSTD_hufCTablesMetadata_t* hufMetadata, - void* workspace, size_t wkspSize, - int writeEntropy) -{ - unsigned* const countWksp = (unsigned*)workspace; - unsigned maxSymbolValue = HUF_SYMBOLVALUE_MAX; - size_t literalSectionHeaderSize = 3 + (litSize >= 1 KB) + (litSize >= 16 KB); - U32 singleStream = litSize < 256; - - if (hufMetadata->hType == set_basic) return litSize; - else if (hufMetadata->hType == set_rle) return 1; - else if (hufMetadata->hType == set_compressed || hufMetadata->hType == set_repeat) { - size_t const largest = HIST_count_wksp (countWksp, &maxSymbolValue, (const BYTE*)literals, litSize, workspace, wkspSize); - if (ZSTD_isError(largest)) return litSize; - { size_t cLitSizeEstimate = HUF_estimateCompressedSize((const HUF_CElt*)huf->CTable, countWksp, maxSymbolValue); - if (writeEntropy) cLitSizeEstimate += hufMetadata->hufDesSize; - if (!singleStream) cLitSizeEstimate += 6; /* multi-stream huffman uses 6-byte jump table */ - return cLitSizeEstimate + literalSectionHeaderSize; - } } - assert(0); /* impossible */ - return 0; -} - -/* Returns the size estimate for the FSE-compressed symbols (of, ml, ll) of a block */ -static size_t ZSTD_estimateBlockSize_symbolType(symbolEncodingType_e type, - const BYTE* codeTable, size_t nbSeq, unsigned maxCode, - const FSE_CTable* fseCTable, - const U8* additionalBits, - short const* defaultNorm, U32 defaultNormLog, U32 defaultMax, - void* workspace, size_t wkspSize) -{ - unsigned* const countWksp = (unsigned*)workspace; - const BYTE* ctp = codeTable; - const BYTE* const ctStart = ctp; - const BYTE* const ctEnd = ctStart + nbSeq; - size_t cSymbolTypeSizeEstimateInBits = 0; - unsigned max = maxCode; - - HIST_countFast_wksp(countWksp, &max, codeTable, nbSeq, workspace, wkspSize); /* can't fail */ - if (type == set_basic) { - /* We selected this encoding type, so it must be valid. */ - assert(max <= defaultMax); - (void)defaultMax; - cSymbolTypeSizeEstimateInBits = ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, countWksp, max); - } else if (type == set_rle) { - cSymbolTypeSizeEstimateInBits = 0; - } else if (type == set_compressed || type == set_repeat) { - cSymbolTypeSizeEstimateInBits = ZSTD_fseBitCost(fseCTable, countWksp, max); - } - if (ZSTD_isError(cSymbolTypeSizeEstimateInBits)) { - return nbSeq * 10; - } - while (ctp < ctEnd) { - if (additionalBits) cSymbolTypeSizeEstimateInBits += additionalBits[*ctp]; - else cSymbolTypeSizeEstimateInBits += *ctp; /* for offset, offset code is also the number of additional bits */ - ctp++; - } - return cSymbolTypeSizeEstimateInBits >> 3; -} - -/* Returns the size estimate for the sequences section (header + content) of a block */ -static size_t ZSTD_estimateBlockSize_sequences(const BYTE* ofCodeTable, - const BYTE* llCodeTable, - const BYTE* mlCodeTable, - size_t nbSeq, - const ZSTD_fseCTables_t* fseTables, - const ZSTD_fseCTablesMetadata_t* fseMetadata, - void* workspace, size_t wkspSize, - int writeEntropy) -{ - size_t sequencesSectionHeaderSize = 1 /* seqHead */ + 1 /* min seqSize size */ + (nbSeq >= 128) + (nbSeq >= LONGNBSEQ); - size_t cSeqSizeEstimate = 0; - cSeqSizeEstimate += ZSTD_estimateBlockSize_symbolType(fseMetadata->ofType, ofCodeTable, nbSeq, MaxOff, - fseTables->offcodeCTable, NULL, - OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff, - workspace, wkspSize); - cSeqSizeEstimate += ZSTD_estimateBlockSize_symbolType(fseMetadata->llType, llCodeTable, nbSeq, MaxLL, - fseTables->litlengthCTable, LL_bits, - LL_defaultNorm, LL_defaultNormLog, MaxLL, - workspace, wkspSize); - cSeqSizeEstimate += ZSTD_estimateBlockSize_symbolType(fseMetadata->mlType, mlCodeTable, nbSeq, MaxML, - fseTables->matchlengthCTable, ML_bits, - ML_defaultNorm, ML_defaultNormLog, MaxML, - workspace, wkspSize); - if (writeEntropy) cSeqSizeEstimate += fseMetadata->fseTablesSize; - return cSeqSizeEstimate + sequencesSectionHeaderSize; -} - -/* Returns the size estimate for a given stream of literals, of, ll, ml */ -static size_t ZSTD_estimateBlockSize(const BYTE* literals, size_t litSize, - const BYTE* ofCodeTable, - const BYTE* llCodeTable, - const BYTE* mlCodeTable, - size_t nbSeq, - const ZSTD_entropyCTables_t* entropy, - const ZSTD_entropyCTablesMetadata_t* entropyMetadata, - void* workspace, size_t wkspSize, - int writeLitEntropy, int writeSeqEntropy) { - size_t const literalsSize = ZSTD_estimateBlockSize_literal(literals, litSize, - &entropy->huf, &entropyMetadata->hufMetadata, - workspace, wkspSize, writeLitEntropy); - size_t const seqSize = ZSTD_estimateBlockSize_sequences(ofCodeTable, llCodeTable, mlCodeTable, - nbSeq, &entropy->fse, &entropyMetadata->fseMetadata, - workspace, wkspSize, writeSeqEntropy); - return seqSize + literalsSize + ZSTD_blockHeaderSize; -} - -/* Builds entropy statistics and uses them for blocksize estimation. - * - * Returns the estimated compressed size of the seqStore, or a zstd error. - */ -static size_t ZSTD_buildEntropyStatisticsAndEstimateSubBlockSize(seqStore_t* seqStore, ZSTD_CCtx* zc) { - ZSTD_entropyCTablesMetadata_t* entropyMetadata = &zc->blockSplitCtx.entropyMetadata; - DEBUGLOG(6, "ZSTD_buildEntropyStatisticsAndEstimateSubBlockSize()"); - FORWARD_IF_ERROR(ZSTD_buildBlockEntropyStats(seqStore, - &zc->blockState.prevCBlock->entropy, - &zc->blockState.nextCBlock->entropy, - &zc->appliedParams, - entropyMetadata, - zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */), ""); - return ZSTD_estimateBlockSize(seqStore->litStart, (size_t)(seqStore->lit - seqStore->litStart), - seqStore->ofCode, seqStore->llCode, seqStore->mlCode, - (size_t)(seqStore->sequences - seqStore->sequencesStart), - &zc->blockState.nextCBlock->entropy, entropyMetadata, zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE, - (int)(entropyMetadata->hufMetadata.hType == set_compressed), 1); -} - -/* Returns literals bytes represented in a seqStore */ -static size_t ZSTD_countSeqStoreLiteralsBytes(const seqStore_t* const seqStore) { - size_t literalsBytes = 0; - size_t const nbSeqs = seqStore->sequences - seqStore->sequencesStart; - size_t i; - for (i = 0; i < nbSeqs; ++i) { - seqDef seq = seqStore->sequencesStart[i]; - literalsBytes += seq.litLength; - if (i == seqStore->longLengthPos && seqStore->longLengthType == ZSTD_llt_literalLength) { - literalsBytes += 0x10000; - } - } - return literalsBytes; -} - -/* Returns match bytes represented in a seqStore */ -static size_t ZSTD_countSeqStoreMatchBytes(const seqStore_t* const seqStore) { - size_t matchBytes = 0; - size_t const nbSeqs = seqStore->sequences - seqStore->sequencesStart; - size_t i; - for (i = 0; i < nbSeqs; ++i) { - seqDef seq = seqStore->sequencesStart[i]; - matchBytes += seq.mlBase + MINMATCH; - if (i == seqStore->longLengthPos && seqStore->longLengthType == ZSTD_llt_matchLength) { - matchBytes += 0x10000; - } - } - return matchBytes; -} - -/* Derives the seqStore that is a chunk of the originalSeqStore from [startIdx, endIdx). - * Stores the result in resultSeqStore. - */ -static void ZSTD_deriveSeqStoreChunk(seqStore_t* resultSeqStore, - const seqStore_t* originalSeqStore, - size_t startIdx, size_t endIdx) { - BYTE* const litEnd = originalSeqStore->lit; - size_t literalsBytes; - size_t literalsBytesPreceding = 0; - - *resultSeqStore = *originalSeqStore; - if (startIdx > 0) { - resultSeqStore->sequences = originalSeqStore->sequencesStart + startIdx; - literalsBytesPreceding = ZSTD_countSeqStoreLiteralsBytes(resultSeqStore); - } - - /* Move longLengthPos into the correct position if necessary */ - if (originalSeqStore->longLengthType != ZSTD_llt_none) { - if (originalSeqStore->longLengthPos < startIdx || originalSeqStore->longLengthPos > endIdx) { - resultSeqStore->longLengthType = ZSTD_llt_none; - } else { - resultSeqStore->longLengthPos -= (U32)startIdx; - } - } - resultSeqStore->sequencesStart = originalSeqStore->sequencesStart + startIdx; - resultSeqStore->sequences = originalSeqStore->sequencesStart + endIdx; - literalsBytes = ZSTD_countSeqStoreLiteralsBytes(resultSeqStore); - resultSeqStore->litStart += literalsBytesPreceding; - if (endIdx == (size_t)(originalSeqStore->sequences - originalSeqStore->sequencesStart)) { - /* This accounts for possible last literals if the derived chunk reaches the end of the block */ - resultSeqStore->lit = litEnd; - } else { - resultSeqStore->lit = resultSeqStore->litStart+literalsBytes; - } - resultSeqStore->llCode += startIdx; - resultSeqStore->mlCode += startIdx; - resultSeqStore->ofCode += startIdx; -} - -/** - * Returns the raw offset represented by the combination of offCode, ll0, and repcode history. - * offCode must represent a repcode in the numeric representation of ZSTD_storeSeq(). - */ -static U32 -ZSTD_resolveRepcodeToRawOffset(const U32 rep[ZSTD_REP_NUM], const U32 offCode, const U32 ll0) -{ - U32 const adjustedOffCode = STORED_REPCODE(offCode) - 1 + ll0; /* [ 0 - 3 ] */ - assert(STORED_IS_REPCODE(offCode)); - if (adjustedOffCode == ZSTD_REP_NUM) { - /* litlength == 0 and offCode == 2 implies selection of first repcode - 1 */ - assert(rep[0] > 0); - return rep[0] - 1; - } - return rep[adjustedOffCode]; -} - -/** - * ZSTD_seqStore_resolveOffCodes() reconciles any possible divergences in offset history that may arise - * due to emission of RLE/raw blocks that disturb the offset history, - * and replaces any repcodes within the seqStore that may be invalid. - * - * dRepcodes are updated as would be on the decompression side. - * cRepcodes are updated exactly in accordance with the seqStore. - * - * Note : this function assumes seq->offBase respects the following numbering scheme : - * 0 : invalid - * 1-3 : repcode 1-3 - * 4+ : real_offset+3 - */ -static void ZSTD_seqStore_resolveOffCodes(repcodes_t* const dRepcodes, repcodes_t* const cRepcodes, - seqStore_t* const seqStore, U32 const nbSeq) { - U32 idx = 0; - for (; idx < nbSeq; ++idx) { - seqDef* const seq = seqStore->sequencesStart + idx; - U32 const ll0 = (seq->litLength == 0); - U32 const offCode = OFFBASE_TO_STORED(seq->offBase); - assert(seq->offBase > 0); - if (STORED_IS_REPCODE(offCode)) { - U32 const dRawOffset = ZSTD_resolveRepcodeToRawOffset(dRepcodes->rep, offCode, ll0); - U32 const cRawOffset = ZSTD_resolveRepcodeToRawOffset(cRepcodes->rep, offCode, ll0); - /* Adjust simulated decompression repcode history if we come across a mismatch. Replace - * the repcode with the offset it actually references, determined by the compression - * repcode history. - */ - if (dRawOffset != cRawOffset) { - seq->offBase = cRawOffset + ZSTD_REP_NUM; - } - } - /* Compression repcode history is always updated with values directly from the unmodified seqStore. - * Decompression repcode history may use modified seq->offset value taken from compression repcode history. - */ - ZSTD_updateRep(dRepcodes->rep, OFFBASE_TO_STORED(seq->offBase), ll0); - ZSTD_updateRep(cRepcodes->rep, offCode, ll0); - } -} - -/* ZSTD_compressSeqStore_singleBlock(): - * Compresses a seqStore into a block with a block header, into the buffer dst. - * - * Returns the total size of that block (including header) or a ZSTD error code. - */ -static size_t -ZSTD_compressSeqStore_singleBlock(ZSTD_CCtx* zc, seqStore_t* const seqStore, - repcodes_t* const dRep, repcodes_t* const cRep, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - U32 lastBlock, U32 isPartition) -{ - const U32 rleMaxLength = 25; - BYTE* op = (BYTE*)dst; - const BYTE* ip = (const BYTE*)src; - size_t cSize; - size_t cSeqsSize; - - /* In case of an RLE or raw block, the simulated decompression repcode history must be reset */ - repcodes_t const dRepOriginal = *dRep; - DEBUGLOG(5, "ZSTD_compressSeqStore_singleBlock"); - if (isPartition) - ZSTD_seqStore_resolveOffCodes(dRep, cRep, seqStore, (U32)(seqStore->sequences - seqStore->sequencesStart)); - - RETURN_ERROR_IF(dstCapacity < ZSTD_blockHeaderSize, dstSize_tooSmall, "Block header doesn't fit"); - cSeqsSize = ZSTD_entropyCompressSeqStore(seqStore, - &zc->blockState.prevCBlock->entropy, &zc->blockState.nextCBlock->entropy, - &zc->appliedParams, - op + ZSTD_blockHeaderSize, dstCapacity - ZSTD_blockHeaderSize, - srcSize, - zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */, - zc->bmi2); - FORWARD_IF_ERROR(cSeqsSize, "ZSTD_entropyCompressSeqStore failed!"); - - if (!zc->isFirstBlock && - cSeqsSize < rleMaxLength && - ZSTD_isRLE((BYTE const*)src, srcSize)) { - /* We don't want to emit our first block as a RLE even if it qualifies because - * doing so will cause the decoder (cli only) to throw a "should consume all input error." - * This is only an issue for zstd <= v1.4.3 - */ - cSeqsSize = 1; - } - - if (zc->seqCollector.collectSequences) { - ZSTD_copyBlockSequences(zc); - ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); - return 0; - } - - if (cSeqsSize == 0) { - cSize = ZSTD_noCompressBlock(op, dstCapacity, ip, srcSize, lastBlock); - FORWARD_IF_ERROR(cSize, "Nocompress block failed"); - DEBUGLOG(4, "Writing out nocompress block, size: %zu", cSize); - *dRep = dRepOriginal; /* reset simulated decompression repcode history */ - } else if (cSeqsSize == 1) { - cSize = ZSTD_rleCompressBlock(op, dstCapacity, *ip, srcSize, lastBlock); - FORWARD_IF_ERROR(cSize, "RLE compress block failed"); - DEBUGLOG(4, "Writing out RLE block, size: %zu", cSize); - *dRep = dRepOriginal; /* reset simulated decompression repcode history */ - } else { - ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); - writeBlockHeader(op, cSeqsSize, srcSize, lastBlock); - cSize = ZSTD_blockHeaderSize + cSeqsSize; - DEBUGLOG(4, "Writing out compressed block, size: %zu", cSize); - } - - if (zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode == FSE_repeat_valid) - zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode = FSE_repeat_check; - - return cSize; -} - -/* Struct to keep track of where we are in our recursive calls. */ -typedef struct { - U32* splitLocations; /* Array of split indices */ - size_t idx; /* The current index within splitLocations being worked on */ -} seqStoreSplits; - -#define MIN_SEQUENCES_BLOCK_SPLITTING 300 - -/* Helper function to perform the recursive search for block splits. - * Estimates the cost of seqStore prior to split, and estimates the cost of splitting the sequences in half. - * If advantageous to split, then we recurse down the two sub-blocks. If not, or if an error occurred in estimation, then - * we do not recurse. - * - * Note: The recursion depth is capped by a heuristic minimum number of sequences, defined by MIN_SEQUENCES_BLOCK_SPLITTING. - * In theory, this means the absolute largest recursion depth is 10 == log2(maxNbSeqInBlock/MIN_SEQUENCES_BLOCK_SPLITTING). - * In practice, recursion depth usually doesn't go beyond 4. - * - * Furthermore, the number of splits is capped by ZSTD_MAX_NB_BLOCK_SPLITS. At ZSTD_MAX_NB_BLOCK_SPLITS == 196 with the current existing blockSize - * maximum of 128 KB, this value is actually impossible to reach. - */ -static void -ZSTD_deriveBlockSplitsHelper(seqStoreSplits* splits, size_t startIdx, size_t endIdx, - ZSTD_CCtx* zc, const seqStore_t* origSeqStore) -{ - seqStore_t* fullSeqStoreChunk = &zc->blockSplitCtx.fullSeqStoreChunk; - seqStore_t* firstHalfSeqStore = &zc->blockSplitCtx.firstHalfSeqStore; - seqStore_t* secondHalfSeqStore = &zc->blockSplitCtx.secondHalfSeqStore; - size_t estimatedOriginalSize; - size_t estimatedFirstHalfSize; - size_t estimatedSecondHalfSize; - size_t midIdx = (startIdx + endIdx)/2; - - if (endIdx - startIdx < MIN_SEQUENCES_BLOCK_SPLITTING || splits->idx >= ZSTD_MAX_NB_BLOCK_SPLITS) { - DEBUGLOG(6, "ZSTD_deriveBlockSplitsHelper: Too few sequences"); - return; - } - DEBUGLOG(4, "ZSTD_deriveBlockSplitsHelper: startIdx=%zu endIdx=%zu", startIdx, endIdx); - ZSTD_deriveSeqStoreChunk(fullSeqStoreChunk, origSeqStore, startIdx, endIdx); - ZSTD_deriveSeqStoreChunk(firstHalfSeqStore, origSeqStore, startIdx, midIdx); - ZSTD_deriveSeqStoreChunk(secondHalfSeqStore, origSeqStore, midIdx, endIdx); - estimatedOriginalSize = ZSTD_buildEntropyStatisticsAndEstimateSubBlockSize(fullSeqStoreChunk, zc); - estimatedFirstHalfSize = ZSTD_buildEntropyStatisticsAndEstimateSubBlockSize(firstHalfSeqStore, zc); - estimatedSecondHalfSize = ZSTD_buildEntropyStatisticsAndEstimateSubBlockSize(secondHalfSeqStore, zc); - DEBUGLOG(4, "Estimated original block size: %zu -- First half split: %zu -- Second half split: %zu", - estimatedOriginalSize, estimatedFirstHalfSize, estimatedSecondHalfSize); - if (ZSTD_isError(estimatedOriginalSize) || ZSTD_isError(estimatedFirstHalfSize) || ZSTD_isError(estimatedSecondHalfSize)) { - return; - } - if (estimatedFirstHalfSize + estimatedSecondHalfSize < estimatedOriginalSize) { - ZSTD_deriveBlockSplitsHelper(splits, startIdx, midIdx, zc, origSeqStore); - splits->splitLocations[splits->idx] = (U32)midIdx; - splits->idx++; - ZSTD_deriveBlockSplitsHelper(splits, midIdx, endIdx, zc, origSeqStore); - } -} - -/* Base recursive function. Populates a table with intra-block partition indices that can improve compression ratio. - * - * Returns the number of splits made (which equals the size of the partition table - 1). - */ -static size_t ZSTD_deriveBlockSplits(ZSTD_CCtx* zc, U32 partitions[], U32 nbSeq) { - seqStoreSplits splits = {partitions, 0}; - if (nbSeq <= 4) { - DEBUGLOG(4, "ZSTD_deriveBlockSplits: Too few sequences to split"); - /* Refuse to try and split anything with less than 4 sequences */ - return 0; - } - ZSTD_deriveBlockSplitsHelper(&splits, 0, nbSeq, zc, &zc->seqStore); - splits.splitLocations[splits.idx] = nbSeq; - DEBUGLOG(5, "ZSTD_deriveBlockSplits: final nb partitions: %zu", splits.idx+1); - return splits.idx; -} - -/* ZSTD_compressBlock_splitBlock(): - * Attempts to split a given block into multiple blocks to improve compression ratio. - * - * Returns combined size of all blocks (which includes headers), or a ZSTD error code. - */ -static size_t -ZSTD_compressBlock_splitBlock_internal(ZSTD_CCtx* zc, void* dst, size_t dstCapacity, - const void* src, size_t blockSize, U32 lastBlock, U32 nbSeq) -{ - size_t cSize = 0; - const BYTE* ip = (const BYTE*)src; - BYTE* op = (BYTE*)dst; - size_t i = 0; - size_t srcBytesTotal = 0; - U32* partitions = zc->blockSplitCtx.partitions; /* size == ZSTD_MAX_NB_BLOCK_SPLITS */ - seqStore_t* nextSeqStore = &zc->blockSplitCtx.nextSeqStore; - seqStore_t* currSeqStore = &zc->blockSplitCtx.currSeqStore; - size_t numSplits = ZSTD_deriveBlockSplits(zc, partitions, nbSeq); - - /* If a block is split and some partitions are emitted as RLE/uncompressed, then repcode history - * may become invalid. In order to reconcile potentially invalid repcodes, we keep track of two - * separate repcode histories that simulate repcode history on compression and decompression side, - * and use the histories to determine whether we must replace a particular repcode with its raw offset. - * - * 1) cRep gets updated for each partition, regardless of whether the block was emitted as uncompressed - * or RLE. This allows us to retrieve the offset value that an invalid repcode references within - * a nocompress/RLE block. - * 2) dRep gets updated only for compressed partitions, and when a repcode gets replaced, will use - * the replacement offset value rather than the original repcode to update the repcode history. - * dRep also will be the final repcode history sent to the next block. - * - * See ZSTD_seqStore_resolveOffCodes() for more details. - */ - repcodes_t dRep; - repcodes_t cRep; - ZSTD_memcpy(dRep.rep, zc->blockState.prevCBlock->rep, sizeof(repcodes_t)); - ZSTD_memcpy(cRep.rep, zc->blockState.prevCBlock->rep, sizeof(repcodes_t)); - ZSTD_memset(nextSeqStore, 0, sizeof(seqStore_t)); - - DEBUGLOG(4, "ZSTD_compressBlock_splitBlock_internal (dstCapacity=%u, dictLimit=%u, nextToUpdate=%u)", - (unsigned)dstCapacity, (unsigned)zc->blockState.matchState.window.dictLimit, - (unsigned)zc->blockState.matchState.nextToUpdate); - - if (numSplits == 0) { - size_t cSizeSingleBlock = ZSTD_compressSeqStore_singleBlock(zc, &zc->seqStore, - &dRep, &cRep, - op, dstCapacity, - ip, blockSize, - lastBlock, 0 /* isPartition */); - FORWARD_IF_ERROR(cSizeSingleBlock, "Compressing single block from splitBlock_internal() failed!"); - DEBUGLOG(5, "ZSTD_compressBlock_splitBlock_internal: No splits"); - assert(cSizeSingleBlock <= ZSTD_BLOCKSIZE_MAX + ZSTD_blockHeaderSize); - return cSizeSingleBlock; - } - - ZSTD_deriveSeqStoreChunk(currSeqStore, &zc->seqStore, 0, partitions[0]); - for (i = 0; i <= numSplits; ++i) { - size_t srcBytes; - size_t cSizeChunk; - U32 const lastPartition = (i == numSplits); - U32 lastBlockEntireSrc = 0; - - srcBytes = ZSTD_countSeqStoreLiteralsBytes(currSeqStore) + ZSTD_countSeqStoreMatchBytes(currSeqStore); - srcBytesTotal += srcBytes; - if (lastPartition) { - /* This is the final partition, need to account for possible last literals */ - srcBytes += blockSize - srcBytesTotal; - lastBlockEntireSrc = lastBlock; - } else { - ZSTD_deriveSeqStoreChunk(nextSeqStore, &zc->seqStore, partitions[i], partitions[i+1]); - } - - cSizeChunk = ZSTD_compressSeqStore_singleBlock(zc, currSeqStore, - &dRep, &cRep, - op, dstCapacity, - ip, srcBytes, - lastBlockEntireSrc, 1 /* isPartition */); - DEBUGLOG(5, "Estimated size: %zu actual size: %zu", ZSTD_buildEntropyStatisticsAndEstimateSubBlockSize(currSeqStore, zc), cSizeChunk); - FORWARD_IF_ERROR(cSizeChunk, "Compressing chunk failed!"); - - ip += srcBytes; - op += cSizeChunk; - dstCapacity -= cSizeChunk; - cSize += cSizeChunk; - *currSeqStore = *nextSeqStore; - assert(cSizeChunk <= ZSTD_BLOCKSIZE_MAX + ZSTD_blockHeaderSize); - } - /* cRep and dRep may have diverged during the compression. If so, we use the dRep repcodes - * for the next block. - */ - ZSTD_memcpy(zc->blockState.prevCBlock->rep, dRep.rep, sizeof(repcodes_t)); - return cSize; -} - -static size_t -ZSTD_compressBlock_splitBlock(ZSTD_CCtx* zc, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, U32 lastBlock) -{ - const BYTE* ip = (const BYTE*)src; - BYTE* op = (BYTE*)dst; - U32 nbSeq; - size_t cSize; - DEBUGLOG(4, "ZSTD_compressBlock_splitBlock"); - assert(zc->appliedParams.useBlockSplitter == ZSTD_ps_enable); - - { const size_t bss = ZSTD_buildSeqStore(zc, src, srcSize); - FORWARD_IF_ERROR(bss, "ZSTD_buildSeqStore failed"); - if (bss == ZSTDbss_noCompress) { - if (zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode == FSE_repeat_valid) - zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode = FSE_repeat_check; - cSize = ZSTD_noCompressBlock(op, dstCapacity, ip, srcSize, lastBlock); - FORWARD_IF_ERROR(cSize, "ZSTD_noCompressBlock failed"); - DEBUGLOG(4, "ZSTD_compressBlock_splitBlock: Nocompress block"); - return cSize; - } - nbSeq = (U32)(zc->seqStore.sequences - zc->seqStore.sequencesStart); - } - - cSize = ZSTD_compressBlock_splitBlock_internal(zc, dst, dstCapacity, src, srcSize, lastBlock, nbSeq); - FORWARD_IF_ERROR(cSize, "Splitting blocks failed!"); - return cSize; -} - -static size_t -ZSTD_compressBlock_internal(ZSTD_CCtx* zc, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, U32 frame) -{ - /* This the upper bound for the length of an rle block. - * This isn't the actual upper bound. Finding the real threshold - * needs further investigation. - */ - const U32 rleMaxLength = 25; - size_t cSize; - const BYTE* ip = (const BYTE*)src; - BYTE* op = (BYTE*)dst; - DEBUGLOG(5, "ZSTD_compressBlock_internal (dstCapacity=%u, dictLimit=%u, nextToUpdate=%u)", - (unsigned)dstCapacity, (unsigned)zc->blockState.matchState.window.dictLimit, - (unsigned)zc->blockState.matchState.nextToUpdate); - - { const size_t bss = ZSTD_buildSeqStore(zc, src, srcSize); - FORWARD_IF_ERROR(bss, "ZSTD_buildSeqStore failed"); - if (bss == ZSTDbss_noCompress) { cSize = 0; goto out; } - } - - if (zc->seqCollector.collectSequences) { - ZSTD_copyBlockSequences(zc); - ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); - return 0; - } - - /* encode sequences and literals */ - cSize = ZSTD_entropyCompressSeqStore(&zc->seqStore, - &zc->blockState.prevCBlock->entropy, &zc->blockState.nextCBlock->entropy, - &zc->appliedParams, - dst, dstCapacity, - srcSize, - zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */, - zc->bmi2); - - if (frame && - /* We don't want to emit our first block as a RLE even if it qualifies because - * doing so will cause the decoder (cli only) to throw a "should consume all input error." - * This is only an issue for zstd <= v1.4.3 - */ - !zc->isFirstBlock && - cSize < rleMaxLength && - ZSTD_isRLE(ip, srcSize)) - { - cSize = 1; - op[0] = ip[0]; - } - -out: - if (!ZSTD_isError(cSize) && cSize > 1) { - ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); - } - /* We check that dictionaries have offset codes available for the first - * block. After the first block, the offcode table might not have large - * enough codes to represent the offsets in the data. - */ - if (zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode == FSE_repeat_valid) - zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode = FSE_repeat_check; - - return cSize; -} - -static size_t ZSTD_compressBlock_targetCBlockSize_body(ZSTD_CCtx* zc, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const size_t bss, U32 lastBlock) -{ - DEBUGLOG(6, "Attempting ZSTD_compressSuperBlock()"); - if (bss == ZSTDbss_compress) { - if (/* We don't want to emit our first block as a RLE even if it qualifies because - * doing so will cause the decoder (cli only) to throw a "should consume all input error." - * This is only an issue for zstd <= v1.4.3 - */ - !zc->isFirstBlock && - ZSTD_maybeRLE(&zc->seqStore) && - ZSTD_isRLE((BYTE const*)src, srcSize)) - { - return ZSTD_rleCompressBlock(dst, dstCapacity, *(BYTE const*)src, srcSize, lastBlock); - } - /* Attempt superblock compression. - * - * Note that compressed size of ZSTD_compressSuperBlock() is not bound by the - * standard ZSTD_compressBound(). This is a problem, because even if we have - * space now, taking an extra byte now could cause us to run out of space later - * and violate ZSTD_compressBound(). - * - * Define blockBound(blockSize) = blockSize + ZSTD_blockHeaderSize. - * - * In order to respect ZSTD_compressBound() we must attempt to emit a raw - * uncompressed block in these cases: - * * cSize == 0: Return code for an uncompressed block. - * * cSize == dstSize_tooSmall: We may have expanded beyond blockBound(srcSize). - * ZSTD_noCompressBlock() will return dstSize_tooSmall if we are really out of - * output space. - * * cSize >= blockBound(srcSize): We have expanded the block too much so - * emit an uncompressed block. - */ - { - size_t const cSize = ZSTD_compressSuperBlock(zc, dst, dstCapacity, src, srcSize, lastBlock); - if (cSize != ERROR(dstSize_tooSmall)) { - size_t const maxCSize = srcSize - ZSTD_minGain(srcSize, zc->appliedParams.cParams.strategy); - FORWARD_IF_ERROR(cSize, "ZSTD_compressSuperBlock failed"); - if (cSize != 0 && cSize < maxCSize + ZSTD_blockHeaderSize) { - ZSTD_blockState_confirmRepcodesAndEntropyTables(&zc->blockState); - return cSize; - } - } - } - } - - DEBUGLOG(6, "Resorting to ZSTD_noCompressBlock()"); - /* Superblock compression failed, attempt to emit a single no compress block. - * The decoder will be able to stream this block since it is uncompressed. - */ - return ZSTD_noCompressBlock(dst, dstCapacity, src, srcSize, lastBlock); -} - -static size_t ZSTD_compressBlock_targetCBlockSize(ZSTD_CCtx* zc, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - U32 lastBlock) -{ - size_t cSize = 0; - const size_t bss = ZSTD_buildSeqStore(zc, src, srcSize); - DEBUGLOG(5, "ZSTD_compressBlock_targetCBlockSize (dstCapacity=%u, dictLimit=%u, nextToUpdate=%u, srcSize=%zu)", - (unsigned)dstCapacity, (unsigned)zc->blockState.matchState.window.dictLimit, (unsigned)zc->blockState.matchState.nextToUpdate, srcSize); - FORWARD_IF_ERROR(bss, "ZSTD_buildSeqStore failed"); - - cSize = ZSTD_compressBlock_targetCBlockSize_body(zc, dst, dstCapacity, src, srcSize, bss, lastBlock); - FORWARD_IF_ERROR(cSize, "ZSTD_compressBlock_targetCBlockSize_body failed"); - - if (zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode == FSE_repeat_valid) - zc->blockState.prevCBlock->entropy.fse.offcode_repeatMode = FSE_repeat_check; - - return cSize; -} - -static void ZSTD_overflowCorrectIfNeeded(ZSTD_matchState_t* ms, - ZSTD_cwksp* ws, - ZSTD_CCtx_params const* params, - void const* ip, - void const* iend) -{ - U32 const cycleLog = ZSTD_cycleLog(params->cParams.chainLog, params->cParams.strategy); - U32 const maxDist = (U32)1 << params->cParams.windowLog; - if (ZSTD_window_needOverflowCorrection(ms->window, cycleLog, maxDist, ms->loadedDictEnd, ip, iend)) { - U32 const correction = ZSTD_window_correctOverflow(&ms->window, cycleLog, maxDist, ip); - ZSTD_STATIC_ASSERT(ZSTD_CHAINLOG_MAX <= 30); - ZSTD_STATIC_ASSERT(ZSTD_WINDOWLOG_MAX_32 <= 30); - ZSTD_STATIC_ASSERT(ZSTD_WINDOWLOG_MAX <= 31); - ZSTD_cwksp_mark_tables_dirty(ws); - ZSTD_reduceIndex(ms, params, correction); - ZSTD_cwksp_mark_tables_clean(ws); - if (ms->nextToUpdate < correction) ms->nextToUpdate = 0; - else ms->nextToUpdate -= correction; - /* invalidate dictionaries on overflow correction */ - ms->loadedDictEnd = 0; - ms->dictMatchState = NULL; - } -} - -/*! ZSTD_compress_frameChunk() : -* Compress a chunk of data into one or multiple blocks. -* All blocks will be terminated, all input will be consumed. -* Function will issue an error if there is not enough `dstCapacity` to hold the compressed content. -* Frame is supposed already started (header already produced) -* @return : compressed size, or an error code -*/ -static size_t ZSTD_compress_frameChunk(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - U32 lastFrameChunk) -{ - size_t blockSize = cctx->blockSize; - size_t remaining = srcSize; - const BYTE* ip = (const BYTE*)src; - BYTE* const ostart = (BYTE*)dst; - BYTE* op = ostart; - U32 const maxDist = (U32)1 << cctx->appliedParams.cParams.windowLog; - - assert(cctx->appliedParams.cParams.windowLog <= ZSTD_WINDOWLOG_MAX); - - DEBUGLOG(4, "ZSTD_compress_frameChunk (blockSize=%u)", (unsigned)blockSize); - if (cctx->appliedParams.fParams.checksumFlag && srcSize) - XXH64_update(&cctx->xxhState, src, srcSize); - - while (remaining) { - ZSTD_matchState_t* const ms = &cctx->blockState.matchState; - U32 const lastBlock = lastFrameChunk & (blockSize >= remaining); - - RETURN_ERROR_IF(dstCapacity < ZSTD_blockHeaderSize + MIN_CBLOCK_SIZE, - dstSize_tooSmall, - "not enough space to store compressed block"); - if (remaining < blockSize) blockSize = remaining; - - ZSTD_overflowCorrectIfNeeded( - ms, &cctx->workspace, &cctx->appliedParams, ip, ip + blockSize); - ZSTD_checkDictValidity(&ms->window, ip + blockSize, maxDist, &ms->loadedDictEnd, &ms->dictMatchState); - ZSTD_window_enforceMaxDist(&ms->window, ip, maxDist, &ms->loadedDictEnd, &ms->dictMatchState); - - /* Ensure hash/chain table insertion resumes no sooner than lowlimit */ - if (ms->nextToUpdate < ms->window.lowLimit) ms->nextToUpdate = ms->window.lowLimit; - - { size_t cSize; - if (ZSTD_useTargetCBlockSize(&cctx->appliedParams)) { - cSize = ZSTD_compressBlock_targetCBlockSize(cctx, op, dstCapacity, ip, blockSize, lastBlock); - FORWARD_IF_ERROR(cSize, "ZSTD_compressBlock_targetCBlockSize failed"); - assert(cSize > 0); - assert(cSize <= blockSize + ZSTD_blockHeaderSize); - } else if (ZSTD_blockSplitterEnabled(&cctx->appliedParams)) { - cSize = ZSTD_compressBlock_splitBlock(cctx, op, dstCapacity, ip, blockSize, lastBlock); - FORWARD_IF_ERROR(cSize, "ZSTD_compressBlock_splitBlock failed"); - assert(cSize > 0 || cctx->seqCollector.collectSequences == 1); - } else { - cSize = ZSTD_compressBlock_internal(cctx, - op+ZSTD_blockHeaderSize, dstCapacity-ZSTD_blockHeaderSize, - ip, blockSize, 1 /* frame */); - FORWARD_IF_ERROR(cSize, "ZSTD_compressBlock_internal failed"); - - if (cSize == 0) { /* block is not compressible */ - cSize = ZSTD_noCompressBlock(op, dstCapacity, ip, blockSize, lastBlock); - FORWARD_IF_ERROR(cSize, "ZSTD_noCompressBlock failed"); - } else { - U32 const cBlockHeader = cSize == 1 ? - lastBlock + (((U32)bt_rle)<<1) + (U32)(blockSize << 3) : - lastBlock + (((U32)bt_compressed)<<1) + (U32)(cSize << 3); - MEM_writeLE24(op, cBlockHeader); - cSize += ZSTD_blockHeaderSize; - } - } - - - ip += blockSize; - assert(remaining >= blockSize); - remaining -= blockSize; - op += cSize; - assert(dstCapacity >= cSize); - dstCapacity -= cSize; - cctx->isFirstBlock = 0; - DEBUGLOG(5, "ZSTD_compress_frameChunk: adding a block of size %u", - (unsigned)cSize); - } } - - if (lastFrameChunk && (op>ostart)) cctx->stage = ZSTDcs_ending; - return (size_t)(op-ostart); -} - - -static size_t ZSTD_writeFrameHeader(void* dst, size_t dstCapacity, - const ZSTD_CCtx_params* params, U64 pledgedSrcSize, U32 dictID) -{ BYTE* const op = (BYTE*)dst; - U32 const dictIDSizeCodeLength = (dictID>0) + (dictID>=256) + (dictID>=65536); /* 0-3 */ - U32 const dictIDSizeCode = params->fParams.noDictIDFlag ? 0 : dictIDSizeCodeLength; /* 0-3 */ - U32 const checksumFlag = params->fParams.checksumFlag>0; - U32 const windowSize = (U32)1 << params->cParams.windowLog; - U32 const singleSegment = params->fParams.contentSizeFlag && (windowSize >= pledgedSrcSize); - BYTE const windowLogByte = (BYTE)((params->cParams.windowLog - ZSTD_WINDOWLOG_ABSOLUTEMIN) << 3); - U32 const fcsCode = params->fParams.contentSizeFlag ? - (pledgedSrcSize>=256) + (pledgedSrcSize>=65536+256) + (pledgedSrcSize>=0xFFFFFFFFU) : 0; /* 0-3 */ - BYTE const frameHeaderDescriptionByte = (BYTE)(dictIDSizeCode + (checksumFlag<<2) + (singleSegment<<5) + (fcsCode<<6) ); - size_t pos=0; - - assert(!(params->fParams.contentSizeFlag && pledgedSrcSize == ZSTD_CONTENTSIZE_UNKNOWN)); - RETURN_ERROR_IF(dstCapacity < ZSTD_FRAMEHEADERSIZE_MAX, dstSize_tooSmall, - "dst buf is too small to fit worst-case frame header size."); - DEBUGLOG(4, "ZSTD_writeFrameHeader : dictIDFlag : %u ; dictID : %u ; dictIDSizeCode : %u", - !params->fParams.noDictIDFlag, (unsigned)dictID, (unsigned)dictIDSizeCode); - if (params->format == ZSTD_f_zstd1) { - MEM_writeLE32(dst, ZSTD_MAGICNUMBER); - pos = 4; - } - op[pos++] = frameHeaderDescriptionByte; - if (!singleSegment) op[pos++] = windowLogByte; - switch(dictIDSizeCode) - { - default: - assert(0); /* impossible */ - ZSTD_FALLTHROUGH; - case 0 : break; - case 1 : op[pos] = (BYTE)(dictID); pos++; break; - case 2 : MEM_writeLE16(op+pos, (U16)dictID); pos+=2; break; - case 3 : MEM_writeLE32(op+pos, dictID); pos+=4; break; - } - switch(fcsCode) - { - default: - assert(0); /* impossible */ - ZSTD_FALLTHROUGH; - case 0 : if (singleSegment) op[pos++] = (BYTE)(pledgedSrcSize); break; - case 1 : MEM_writeLE16(op+pos, (U16)(pledgedSrcSize-256)); pos+=2; break; - case 2 : MEM_writeLE32(op+pos, (U32)(pledgedSrcSize)); pos+=4; break; - case 3 : MEM_writeLE64(op+pos, (U64)(pledgedSrcSize)); pos+=8; break; - } - return pos; -} - -/* ZSTD_writeSkippableFrame_advanced() : - * Writes out a skippable frame with the specified magic number variant (16 are supported), - * from ZSTD_MAGIC_SKIPPABLE_START to ZSTD_MAGIC_SKIPPABLE_START+15, and the desired source data. - * - * Returns the total number of bytes written, or a ZSTD error code. - */ -size_t ZSTD_writeSkippableFrame(void* dst, size_t dstCapacity, - const void* src, size_t srcSize, unsigned magicVariant) { - BYTE* op = (BYTE*)dst; - RETURN_ERROR_IF(dstCapacity < srcSize + ZSTD_SKIPPABLEHEADERSIZE /* Skippable frame overhead */, - dstSize_tooSmall, "Not enough room for skippable frame"); - RETURN_ERROR_IF(srcSize > (unsigned)0xFFFFFFFF, srcSize_wrong, "Src size too large for skippable frame"); - RETURN_ERROR_IF(magicVariant > 15, parameter_outOfBound, "Skippable frame magic number variant not supported"); - - MEM_writeLE32(op, (U32)(ZSTD_MAGIC_SKIPPABLE_START + magicVariant)); - MEM_writeLE32(op+4, (U32)srcSize); - ZSTD_memcpy(op+8, src, srcSize); - return srcSize + ZSTD_SKIPPABLEHEADERSIZE; -} - -/* ZSTD_writeLastEmptyBlock() : - * output an empty Block with end-of-frame mark to complete a frame - * @return : size of data written into `dst` (== ZSTD_blockHeaderSize (defined in zstd_internal.h)) - * or an error code if `dstCapacity` is too small (stage != ZSTDcs_init, stage_wrong, - "wrong cctx stage"); - RETURN_ERROR_IF(cctx->appliedParams.ldmParams.enableLdm == ZSTD_ps_enable, - parameter_unsupported, - "incompatible with ldm"); - cctx->externSeqStore.seq = seq; - cctx->externSeqStore.size = nbSeq; - cctx->externSeqStore.capacity = nbSeq; - cctx->externSeqStore.pos = 0; - cctx->externSeqStore.posInSequence = 0; - return 0; -} - - -static size_t ZSTD_compressContinue_internal (ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - U32 frame, U32 lastFrameChunk) -{ - ZSTD_matchState_t* const ms = &cctx->blockState.matchState; - size_t fhSize = 0; - - DEBUGLOG(5, "ZSTD_compressContinue_internal, stage: %u, srcSize: %u", - cctx->stage, (unsigned)srcSize); - RETURN_ERROR_IF(cctx->stage==ZSTDcs_created, stage_wrong, - "missing init (ZSTD_compressBegin)"); - - if (frame && (cctx->stage==ZSTDcs_init)) { - fhSize = ZSTD_writeFrameHeader(dst, dstCapacity, &cctx->appliedParams, - cctx->pledgedSrcSizePlusOne-1, cctx->dictID); - FORWARD_IF_ERROR(fhSize, "ZSTD_writeFrameHeader failed"); - assert(fhSize <= dstCapacity); - dstCapacity -= fhSize; - dst = (char*)dst + fhSize; - cctx->stage = ZSTDcs_ongoing; - } - - if (!srcSize) return fhSize; /* do not generate an empty block if no input */ - - if (!ZSTD_window_update(&ms->window, src, srcSize, ms->forceNonContiguous)) { - ms->forceNonContiguous = 0; - ms->nextToUpdate = ms->window.dictLimit; - } - if (cctx->appliedParams.ldmParams.enableLdm == ZSTD_ps_enable) { - ZSTD_window_update(&cctx->ldmState.window, src, srcSize, /* forceNonContiguous */ 0); - } - - if (!frame) { - /* overflow check and correction for block mode */ - ZSTD_overflowCorrectIfNeeded( - ms, &cctx->workspace, &cctx->appliedParams, - src, (BYTE const*)src + srcSize); - } - - DEBUGLOG(5, "ZSTD_compressContinue_internal (blockSize=%u)", (unsigned)cctx->blockSize); - { size_t const cSize = frame ? - ZSTD_compress_frameChunk (cctx, dst, dstCapacity, src, srcSize, lastFrameChunk) : - ZSTD_compressBlock_internal (cctx, dst, dstCapacity, src, srcSize, 0 /* frame */); - FORWARD_IF_ERROR(cSize, "%s", frame ? "ZSTD_compress_frameChunk failed" : "ZSTD_compressBlock_internal failed"); - cctx->consumedSrcSize += srcSize; - cctx->producedCSize += (cSize + fhSize); - assert(!(cctx->appliedParams.fParams.contentSizeFlag && cctx->pledgedSrcSizePlusOne == 0)); - if (cctx->pledgedSrcSizePlusOne != 0) { /* control src size */ - ZSTD_STATIC_ASSERT(ZSTD_CONTENTSIZE_UNKNOWN == (unsigned long long)-1); - RETURN_ERROR_IF( - cctx->consumedSrcSize+1 > cctx->pledgedSrcSizePlusOne, - srcSize_wrong, - "error : pledgedSrcSize = %u, while realSrcSize >= %u", - (unsigned)cctx->pledgedSrcSizePlusOne-1, - (unsigned)cctx->consumedSrcSize); - } - return cSize + fhSize; - } -} - -size_t ZSTD_compressContinue (ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize) -{ - DEBUGLOG(5, "ZSTD_compressContinue (srcSize=%u)", (unsigned)srcSize); - return ZSTD_compressContinue_internal(cctx, dst, dstCapacity, src, srcSize, 1 /* frame mode */, 0 /* last chunk */); -} - - -size_t ZSTD_getBlockSize(const ZSTD_CCtx* cctx) -{ - ZSTD_compressionParameters const cParams = cctx->appliedParams.cParams; - assert(!ZSTD_checkCParams(cParams)); - return MIN (ZSTD_BLOCKSIZE_MAX, (U32)1 << cParams.windowLog); -} - -size_t ZSTD_compressBlock(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize) -{ - DEBUGLOG(5, "ZSTD_compressBlock: srcSize = %u", (unsigned)srcSize); - { size_t const blockSizeMax = ZSTD_getBlockSize(cctx); - RETURN_ERROR_IF(srcSize > blockSizeMax, srcSize_wrong, "input is larger than a block"); } - - return ZSTD_compressContinue_internal(cctx, dst, dstCapacity, src, srcSize, 0 /* frame mode */, 0 /* last chunk */); -} - -/*! ZSTD_loadDictionaryContent() : - * @return : 0, or an error code - */ -static size_t ZSTD_loadDictionaryContent(ZSTD_matchState_t* ms, - ldmState_t* ls, - ZSTD_cwksp* ws, - ZSTD_CCtx_params const* params, - const void* src, size_t srcSize, - ZSTD_dictTableLoadMethod_e dtlm) -{ - const BYTE* ip = (const BYTE*) src; - const BYTE* const iend = ip + srcSize; - int const loadLdmDict = params->ldmParams.enableLdm == ZSTD_ps_enable && ls != NULL; - - /* Assert that we the ms params match the params we're being given */ - ZSTD_assertEqualCParams(params->cParams, ms->cParams); - - if (srcSize > ZSTD_CHUNKSIZE_MAX) { - /* Allow the dictionary to set indices up to exactly ZSTD_CURRENT_MAX. - * Dictionaries right at the edge will immediately trigger overflow - * correction, but I don't want to insert extra constraints here. - */ - U32 const maxDictSize = ZSTD_CURRENT_MAX - 1; - /* We must have cleared our windows when our source is this large. */ - assert(ZSTD_window_isEmpty(ms->window)); - if (loadLdmDict) - assert(ZSTD_window_isEmpty(ls->window)); - /* If the dictionary is too large, only load the suffix of the dictionary. */ - if (srcSize > maxDictSize) { - ip = iend - maxDictSize; - src = ip; - srcSize = maxDictSize; - } - } - - DEBUGLOG(4, "ZSTD_loadDictionaryContent(): useRowMatchFinder=%d", (int)params->useRowMatchFinder); - ZSTD_window_update(&ms->window, src, srcSize, /* forceNonContiguous */ 0); - ms->loadedDictEnd = params->forceWindow ? 0 : (U32)(iend - ms->window.base); - ms->forceNonContiguous = params->deterministicRefPrefix; - - if (loadLdmDict) { - ZSTD_window_update(&ls->window, src, srcSize, /* forceNonContiguous */ 0); - ls->loadedDictEnd = params->forceWindow ? 0 : (U32)(iend - ls->window.base); - } - - if (srcSize <= HASH_READ_SIZE) return 0; - - ZSTD_overflowCorrectIfNeeded(ms, ws, params, ip, iend); - - if (loadLdmDict) - ZSTD_ldm_fillHashTable(ls, ip, iend, ¶ms->ldmParams); - - switch(params->cParams.strategy) - { - case ZSTD_fast: - ZSTD_fillHashTable(ms, iend, dtlm); - break; - case ZSTD_dfast: - ZSTD_fillDoubleHashTable(ms, iend, dtlm); - break; - - case ZSTD_greedy: - case ZSTD_lazy: - case ZSTD_lazy2: - assert(srcSize >= HASH_READ_SIZE); - if (ms->dedicatedDictSearch) { - assert(ms->chainTable != NULL); - ZSTD_dedicatedDictSearch_lazy_loadDictionary(ms, iend-HASH_READ_SIZE); - } else { - assert(params->useRowMatchFinder != ZSTD_ps_auto); - if (params->useRowMatchFinder == ZSTD_ps_enable) { - size_t const tagTableSize = ((size_t)1 << params->cParams.hashLog) * sizeof(U16); - ZSTD_memset(ms->tagTable, 0, tagTableSize); - ZSTD_row_update(ms, iend-HASH_READ_SIZE); - DEBUGLOG(4, "Using row-based hash table for lazy dict"); - } else { - ZSTD_insertAndFindFirstIndex(ms, iend-HASH_READ_SIZE); - DEBUGLOG(4, "Using chain-based hash table for lazy dict"); - } - } - break; - - case ZSTD_btlazy2: /* we want the dictionary table fully sorted */ - case ZSTD_btopt: - case ZSTD_btultra: - case ZSTD_btultra2: - assert(srcSize >= HASH_READ_SIZE); - ZSTD_updateTree(ms, iend-HASH_READ_SIZE, iend); - break; - - default: - assert(0); /* not possible : not a valid strategy id */ - } - - ms->nextToUpdate = (U32)(iend - ms->window.base); - return 0; -} - - -/* Dictionaries that assign zero probability to symbols that show up causes problems - * when FSE encoding. Mark dictionaries with zero probability symbols as FSE_repeat_check - * and only dictionaries with 100% valid symbols can be assumed valid. - */ -static FSE_repeat ZSTD_dictNCountRepeat(short* normalizedCounter, unsigned dictMaxSymbolValue, unsigned maxSymbolValue) -{ - U32 s; - if (dictMaxSymbolValue < maxSymbolValue) { - return FSE_repeat_check; - } - for (s = 0; s <= maxSymbolValue; ++s) { - if (normalizedCounter[s] == 0) { - return FSE_repeat_check; - } - } - return FSE_repeat_valid; -} - -size_t ZSTD_loadCEntropy(ZSTD_compressedBlockState_t* bs, void* workspace, - const void* const dict, size_t dictSize) -{ - short offcodeNCount[MaxOff+1]; - unsigned offcodeMaxValue = MaxOff; - const BYTE* dictPtr = (const BYTE*)dict; /* skip magic num and dict ID */ - const BYTE* const dictEnd = dictPtr + dictSize; - dictPtr += 8; - bs->entropy.huf.repeatMode = HUF_repeat_check; - - { unsigned maxSymbolValue = 255; - unsigned hasZeroWeights = 1; - size_t const hufHeaderSize = HUF_readCTable((HUF_CElt*)bs->entropy.huf.CTable, &maxSymbolValue, dictPtr, - dictEnd-dictPtr, &hasZeroWeights); - - /* We only set the loaded table as valid if it contains all non-zero - * weights. Otherwise, we set it to check */ - if (!hasZeroWeights) - bs->entropy.huf.repeatMode = HUF_repeat_valid; - - RETURN_ERROR_IF(HUF_isError(hufHeaderSize), dictionary_corrupted, ""); - RETURN_ERROR_IF(maxSymbolValue < 255, dictionary_corrupted, ""); - dictPtr += hufHeaderSize; - } - - { unsigned offcodeLog; - size_t const offcodeHeaderSize = FSE_readNCount(offcodeNCount, &offcodeMaxValue, &offcodeLog, dictPtr, dictEnd-dictPtr); - RETURN_ERROR_IF(FSE_isError(offcodeHeaderSize), dictionary_corrupted, ""); - RETURN_ERROR_IF(offcodeLog > OffFSELog, dictionary_corrupted, ""); - /* fill all offset symbols to avoid garbage at end of table */ - RETURN_ERROR_IF(FSE_isError(FSE_buildCTable_wksp( - bs->entropy.fse.offcodeCTable, - offcodeNCount, MaxOff, offcodeLog, - workspace, HUF_WORKSPACE_SIZE)), - dictionary_corrupted, ""); - /* Defer checking offcodeMaxValue because we need to know the size of the dictionary content */ - dictPtr += offcodeHeaderSize; - } - - { short matchlengthNCount[MaxML+1]; - unsigned matchlengthMaxValue = MaxML, matchlengthLog; - size_t const matchlengthHeaderSize = FSE_readNCount(matchlengthNCount, &matchlengthMaxValue, &matchlengthLog, dictPtr, dictEnd-dictPtr); - RETURN_ERROR_IF(FSE_isError(matchlengthHeaderSize), dictionary_corrupted, ""); - RETURN_ERROR_IF(matchlengthLog > MLFSELog, dictionary_corrupted, ""); - RETURN_ERROR_IF(FSE_isError(FSE_buildCTable_wksp( - bs->entropy.fse.matchlengthCTable, - matchlengthNCount, matchlengthMaxValue, matchlengthLog, - workspace, HUF_WORKSPACE_SIZE)), - dictionary_corrupted, ""); - bs->entropy.fse.matchlength_repeatMode = ZSTD_dictNCountRepeat(matchlengthNCount, matchlengthMaxValue, MaxML); - dictPtr += matchlengthHeaderSize; - } - - { short litlengthNCount[MaxLL+1]; - unsigned litlengthMaxValue = MaxLL, litlengthLog; - size_t const litlengthHeaderSize = FSE_readNCount(litlengthNCount, &litlengthMaxValue, &litlengthLog, dictPtr, dictEnd-dictPtr); - RETURN_ERROR_IF(FSE_isError(litlengthHeaderSize), dictionary_corrupted, ""); - RETURN_ERROR_IF(litlengthLog > LLFSELog, dictionary_corrupted, ""); - RETURN_ERROR_IF(FSE_isError(FSE_buildCTable_wksp( - bs->entropy.fse.litlengthCTable, - litlengthNCount, litlengthMaxValue, litlengthLog, - workspace, HUF_WORKSPACE_SIZE)), - dictionary_corrupted, ""); - bs->entropy.fse.litlength_repeatMode = ZSTD_dictNCountRepeat(litlengthNCount, litlengthMaxValue, MaxLL); - dictPtr += litlengthHeaderSize; - } - - RETURN_ERROR_IF(dictPtr+12 > dictEnd, dictionary_corrupted, ""); - bs->rep[0] = MEM_readLE32(dictPtr+0); - bs->rep[1] = MEM_readLE32(dictPtr+4); - bs->rep[2] = MEM_readLE32(dictPtr+8); - dictPtr += 12; - - { size_t const dictContentSize = (size_t)(dictEnd - dictPtr); - U32 offcodeMax = MaxOff; - if (dictContentSize <= ((U32)-1) - 128 KB) { - U32 const maxOffset = (U32)dictContentSize + 128 KB; /* The maximum offset that must be supported */ - offcodeMax = ZSTD_highbit32(maxOffset); /* Calculate minimum offset code required to represent maxOffset */ - } - /* All offset values <= dictContentSize + 128 KB must be representable for a valid table */ - bs->entropy.fse.offcode_repeatMode = ZSTD_dictNCountRepeat(offcodeNCount, offcodeMaxValue, MIN(offcodeMax, MaxOff)); - - /* All repCodes must be <= dictContentSize and != 0 */ - { U32 u; - for (u=0; u<3; u++) { - RETURN_ERROR_IF(bs->rep[u] == 0, dictionary_corrupted, ""); - RETURN_ERROR_IF(bs->rep[u] > dictContentSize, dictionary_corrupted, ""); - } } } - - return dictPtr - (const BYTE*)dict; -} - -/* Dictionary format : - * See : - * https://github.com/facebook/zstd/blob/release/doc/zstd_compression_format.md#dictionary-format - */ -/*! ZSTD_loadZstdDictionary() : - * @return : dictID, or an error code - * assumptions : magic number supposed already checked - * dictSize supposed >= 8 - */ -static size_t ZSTD_loadZstdDictionary(ZSTD_compressedBlockState_t* bs, - ZSTD_matchState_t* ms, - ZSTD_cwksp* ws, - ZSTD_CCtx_params const* params, - const void* dict, size_t dictSize, - ZSTD_dictTableLoadMethod_e dtlm, - void* workspace) -{ - const BYTE* dictPtr = (const BYTE*)dict; - const BYTE* const dictEnd = dictPtr + dictSize; - size_t dictID; - size_t eSize; - ZSTD_STATIC_ASSERT(HUF_WORKSPACE_SIZE >= (1<= 8); - assert(MEM_readLE32(dictPtr) == ZSTD_MAGIC_DICTIONARY); - - dictID = params->fParams.noDictIDFlag ? 0 : MEM_readLE32(dictPtr + 4 /* skip magic number */ ); - eSize = ZSTD_loadCEntropy(bs, workspace, dict, dictSize); - FORWARD_IF_ERROR(eSize, "ZSTD_loadCEntropy failed"); - dictPtr += eSize; - - { - size_t const dictContentSize = (size_t)(dictEnd - dictPtr); - FORWARD_IF_ERROR(ZSTD_loadDictionaryContent( - ms, NULL, ws, params, dictPtr, dictContentSize, dtlm), ""); - } - return dictID; -} - -/** ZSTD_compress_insertDictionary() : -* @return : dictID, or an error code */ -static size_t -ZSTD_compress_insertDictionary(ZSTD_compressedBlockState_t* bs, - ZSTD_matchState_t* ms, - ldmState_t* ls, - ZSTD_cwksp* ws, - const ZSTD_CCtx_params* params, - const void* dict, size_t dictSize, - ZSTD_dictContentType_e dictContentType, - ZSTD_dictTableLoadMethod_e dtlm, - void* workspace) -{ - DEBUGLOG(4, "ZSTD_compress_insertDictionary (dictSize=%u)", (U32)dictSize); - if ((dict==NULL) || (dictSize<8)) { - RETURN_ERROR_IF(dictContentType == ZSTD_dct_fullDict, dictionary_wrong, ""); - return 0; - } - - ZSTD_reset_compressedBlockState(bs); - - /* dict restricted modes */ - if (dictContentType == ZSTD_dct_rawContent) - return ZSTD_loadDictionaryContent(ms, ls, ws, params, dict, dictSize, dtlm); - - if (MEM_readLE32(dict) != ZSTD_MAGIC_DICTIONARY) { - if (dictContentType == ZSTD_dct_auto) { - DEBUGLOG(4, "raw content dictionary detected"); - return ZSTD_loadDictionaryContent( - ms, ls, ws, params, dict, dictSize, dtlm); - } - RETURN_ERROR_IF(dictContentType == ZSTD_dct_fullDict, dictionary_wrong, ""); - assert(0); /* impossible */ - } - - /* dict as full zstd dictionary */ - return ZSTD_loadZstdDictionary( - bs, ms, ws, params, dict, dictSize, dtlm, workspace); -} - -#define ZSTD_USE_CDICT_PARAMS_SRCSIZE_CUTOFF (128 KB) -#define ZSTD_USE_CDICT_PARAMS_DICTSIZE_MULTIPLIER (6ULL) - -/*! ZSTD_compressBegin_internal() : - * @return : 0, or an error code */ -static size_t ZSTD_compressBegin_internal(ZSTD_CCtx* cctx, - const void* dict, size_t dictSize, - ZSTD_dictContentType_e dictContentType, - ZSTD_dictTableLoadMethod_e dtlm, - const ZSTD_CDict* cdict, - const ZSTD_CCtx_params* params, U64 pledgedSrcSize, - ZSTD_buffered_policy_e zbuff) -{ - size_t const dictContentSize = cdict ? cdict->dictContentSize : dictSize; -#if ZSTD_TRACE - cctx->traceCtx = (ZSTD_trace_compress_begin != NULL) ? ZSTD_trace_compress_begin(cctx) : 0; -#endif - DEBUGLOG(4, "ZSTD_compressBegin_internal: wlog=%u", params->cParams.windowLog); - /* params are supposed to be fully validated at this point */ - assert(!ZSTD_isError(ZSTD_checkCParams(params->cParams))); - assert(!((dict) && (cdict))); /* either dict or cdict, not both */ - if ( (cdict) - && (cdict->dictContentSize > 0) - && ( pledgedSrcSize < ZSTD_USE_CDICT_PARAMS_SRCSIZE_CUTOFF - || pledgedSrcSize < cdict->dictContentSize * ZSTD_USE_CDICT_PARAMS_DICTSIZE_MULTIPLIER - || pledgedSrcSize == ZSTD_CONTENTSIZE_UNKNOWN - || cdict->compressionLevel == 0) - && (params->attachDictPref != ZSTD_dictForceLoad) ) { - return ZSTD_resetCCtx_usingCDict(cctx, cdict, params, pledgedSrcSize, zbuff); - } - - FORWARD_IF_ERROR( ZSTD_resetCCtx_internal(cctx, params, pledgedSrcSize, - dictContentSize, - ZSTDcrp_makeClean, zbuff) , ""); - { size_t const dictID = cdict ? - ZSTD_compress_insertDictionary( - cctx->blockState.prevCBlock, &cctx->blockState.matchState, - &cctx->ldmState, &cctx->workspace, &cctx->appliedParams, cdict->dictContent, - cdict->dictContentSize, cdict->dictContentType, dtlm, - cctx->entropyWorkspace) - : ZSTD_compress_insertDictionary( - cctx->blockState.prevCBlock, &cctx->blockState.matchState, - &cctx->ldmState, &cctx->workspace, &cctx->appliedParams, dict, dictSize, - dictContentType, dtlm, cctx->entropyWorkspace); - FORWARD_IF_ERROR(dictID, "ZSTD_compress_insertDictionary failed"); - assert(dictID <= UINT_MAX); - cctx->dictID = (U32)dictID; - cctx->dictContentSize = dictContentSize; - } - return 0; -} - -size_t ZSTD_compressBegin_advanced_internal(ZSTD_CCtx* cctx, - const void* dict, size_t dictSize, - ZSTD_dictContentType_e dictContentType, - ZSTD_dictTableLoadMethod_e dtlm, - const ZSTD_CDict* cdict, - const ZSTD_CCtx_params* params, - unsigned long long pledgedSrcSize) -{ - DEBUGLOG(4, "ZSTD_compressBegin_advanced_internal: wlog=%u", params->cParams.windowLog); - /* compression parameters verification and optimization */ - FORWARD_IF_ERROR( ZSTD_checkCParams(params->cParams) , ""); - return ZSTD_compressBegin_internal(cctx, - dict, dictSize, dictContentType, dtlm, - cdict, - params, pledgedSrcSize, - ZSTDb_not_buffered); -} - -/*! ZSTD_compressBegin_advanced() : -* @return : 0, or an error code */ -size_t ZSTD_compressBegin_advanced(ZSTD_CCtx* cctx, - const void* dict, size_t dictSize, - ZSTD_parameters params, unsigned long long pledgedSrcSize) -{ - ZSTD_CCtx_params cctxParams; - ZSTD_CCtxParams_init_internal(&cctxParams, ¶ms, ZSTD_NO_CLEVEL); - return ZSTD_compressBegin_advanced_internal(cctx, - dict, dictSize, ZSTD_dct_auto, ZSTD_dtlm_fast, - NULL /*cdict*/, - &cctxParams, pledgedSrcSize); -} - -size_t ZSTD_compressBegin_usingDict(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, int compressionLevel) -{ - ZSTD_CCtx_params cctxParams; - { - ZSTD_parameters const params = ZSTD_getParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_noAttachDict); - ZSTD_CCtxParams_init_internal(&cctxParams, ¶ms, (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : compressionLevel); - } - DEBUGLOG(4, "ZSTD_compressBegin_usingDict (dictSize=%u)", (unsigned)dictSize); - return ZSTD_compressBegin_internal(cctx, dict, dictSize, ZSTD_dct_auto, ZSTD_dtlm_fast, NULL, - &cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, ZSTDb_not_buffered); -} - -size_t ZSTD_compressBegin(ZSTD_CCtx* cctx, int compressionLevel) -{ - return ZSTD_compressBegin_usingDict(cctx, NULL, 0, compressionLevel); -} - - -/*! ZSTD_writeEpilogue() : -* Ends a frame. -* @return : nb of bytes written into dst (or an error code) */ -static size_t ZSTD_writeEpilogue(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity) -{ - BYTE* const ostart = (BYTE*)dst; - BYTE* op = ostart; - size_t fhSize = 0; - - DEBUGLOG(4, "ZSTD_writeEpilogue"); - RETURN_ERROR_IF(cctx->stage == ZSTDcs_created, stage_wrong, "init missing"); - - /* special case : empty frame */ - if (cctx->stage == ZSTDcs_init) { - fhSize = ZSTD_writeFrameHeader(dst, dstCapacity, &cctx->appliedParams, 0, 0); - FORWARD_IF_ERROR(fhSize, "ZSTD_writeFrameHeader failed"); - dstCapacity -= fhSize; - op += fhSize; - cctx->stage = ZSTDcs_ongoing; - } - - if (cctx->stage != ZSTDcs_ending) { - /* write one last empty block, make it the "last" block */ - U32 const cBlockHeader24 = 1 /* last block */ + (((U32)bt_raw)<<1) + 0; - RETURN_ERROR_IF(dstCapacity<4, dstSize_tooSmall, "no room for epilogue"); - MEM_writeLE32(op, cBlockHeader24); - op += ZSTD_blockHeaderSize; - dstCapacity -= ZSTD_blockHeaderSize; - } - - if (cctx->appliedParams.fParams.checksumFlag) { - U32 const checksum = (U32) XXH64_digest(&cctx->xxhState); - RETURN_ERROR_IF(dstCapacity<4, dstSize_tooSmall, "no room for checksum"); - DEBUGLOG(4, "ZSTD_writeEpilogue: write checksum : %08X", (unsigned)checksum); - MEM_writeLE32(op, checksum); - op += 4; - } - - cctx->stage = ZSTDcs_created; /* return to "created but no init" status */ - return op-ostart; -} - -void ZSTD_CCtx_trace(ZSTD_CCtx* cctx, size_t extraCSize) -{ -#if ZSTD_TRACE - if (cctx->traceCtx && ZSTD_trace_compress_end != NULL) { - int const streaming = cctx->inBuffSize > 0 || cctx->outBuffSize > 0 || cctx->appliedParams.nbWorkers > 0; - ZSTD_Trace trace; - ZSTD_memset(&trace, 0, sizeof(trace)); - trace.version = ZSTD_VERSION_NUMBER; - trace.streaming = streaming; - trace.dictionaryID = cctx->dictID; - trace.dictionarySize = cctx->dictContentSize; - trace.uncompressedSize = cctx->consumedSrcSize; - trace.compressedSize = cctx->producedCSize + extraCSize; - trace.params = &cctx->appliedParams; - trace.cctx = cctx; - ZSTD_trace_compress_end(cctx->traceCtx, &trace); - } - cctx->traceCtx = 0; -#else - (void)cctx; - (void)extraCSize; -#endif -} - -size_t ZSTD_compressEnd (ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize) -{ - size_t endResult; - size_t const cSize = ZSTD_compressContinue_internal(cctx, - dst, dstCapacity, src, srcSize, - 1 /* frame mode */, 1 /* last chunk */); - FORWARD_IF_ERROR(cSize, "ZSTD_compressContinue_internal failed"); - endResult = ZSTD_writeEpilogue(cctx, (char*)dst + cSize, dstCapacity-cSize); - FORWARD_IF_ERROR(endResult, "ZSTD_writeEpilogue failed"); - assert(!(cctx->appliedParams.fParams.contentSizeFlag && cctx->pledgedSrcSizePlusOne == 0)); - if (cctx->pledgedSrcSizePlusOne != 0) { /* control src size */ - ZSTD_STATIC_ASSERT(ZSTD_CONTENTSIZE_UNKNOWN == (unsigned long long)-1); - DEBUGLOG(4, "end of frame : controlling src size"); - RETURN_ERROR_IF( - cctx->pledgedSrcSizePlusOne != cctx->consumedSrcSize+1, - srcSize_wrong, - "error : pledgedSrcSize = %u, while realSrcSize = %u", - (unsigned)cctx->pledgedSrcSizePlusOne-1, - (unsigned)cctx->consumedSrcSize); - } - ZSTD_CCtx_trace(cctx, endResult); - return cSize + endResult; -} - -size_t ZSTD_compress_advanced (ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const void* dict,size_t dictSize, - ZSTD_parameters params) -{ - DEBUGLOG(4, "ZSTD_compress_advanced"); - FORWARD_IF_ERROR(ZSTD_checkCParams(params.cParams), ""); - ZSTD_CCtxParams_init_internal(&cctx->simpleApiParams, ¶ms, ZSTD_NO_CLEVEL); - return ZSTD_compress_advanced_internal(cctx, - dst, dstCapacity, - src, srcSize, - dict, dictSize, - &cctx->simpleApiParams); -} - -/* Internal */ -size_t ZSTD_compress_advanced_internal( - ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const void* dict,size_t dictSize, - const ZSTD_CCtx_params* params) -{ - DEBUGLOG(4, "ZSTD_compress_advanced_internal (srcSize:%u)", (unsigned)srcSize); - FORWARD_IF_ERROR( ZSTD_compressBegin_internal(cctx, - dict, dictSize, ZSTD_dct_auto, ZSTD_dtlm_fast, NULL, - params, srcSize, ZSTDb_not_buffered) , ""); - return ZSTD_compressEnd(cctx, dst, dstCapacity, src, srcSize); -} - -size_t ZSTD_compress_usingDict(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const void* dict, size_t dictSize, - int compressionLevel) -{ - { - ZSTD_parameters const params = ZSTD_getParams_internal(compressionLevel, srcSize, dict ? dictSize : 0, ZSTD_cpm_noAttachDict); - assert(params.fParams.contentSizeFlag == 1); - ZSTD_CCtxParams_init_internal(&cctx->simpleApiParams, ¶ms, (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT: compressionLevel); - } - DEBUGLOG(4, "ZSTD_compress_usingDict (srcSize=%u)", (unsigned)srcSize); - return ZSTD_compress_advanced_internal(cctx, dst, dstCapacity, src, srcSize, dict, dictSize, &cctx->simpleApiParams); -} - -size_t ZSTD_compressCCtx(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - int compressionLevel) -{ - DEBUGLOG(4, "ZSTD_compressCCtx (srcSize=%u)", (unsigned)srcSize); - assert(cctx != NULL); - return ZSTD_compress_usingDict(cctx, dst, dstCapacity, src, srcSize, NULL, 0, compressionLevel); -} - -size_t ZSTD_compress(void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - int compressionLevel) -{ - size_t result; -#if ZSTD_COMPRESS_HEAPMODE - ZSTD_CCtx* cctx = ZSTD_createCCtx(); - RETURN_ERROR_IF(!cctx, memory_allocation, "ZSTD_createCCtx failed"); - result = ZSTD_compressCCtx(cctx, dst, dstCapacity, src, srcSize, compressionLevel); - ZSTD_freeCCtx(cctx); -#else - ZSTD_CCtx ctxBody; - ZSTD_initCCtx(&ctxBody, ZSTD_defaultCMem); - result = ZSTD_compressCCtx(&ctxBody, dst, dstCapacity, src, srcSize, compressionLevel); - ZSTD_freeCCtxContent(&ctxBody); /* can't free ctxBody itself, as it's on stack; free only heap content */ -#endif - return result; -} - - -/* ===== Dictionary API ===== */ - -/*! ZSTD_estimateCDictSize_advanced() : - * Estimate amount of memory that will be needed to create a dictionary with following arguments */ -size_t ZSTD_estimateCDictSize_advanced( - size_t dictSize, ZSTD_compressionParameters cParams, - ZSTD_dictLoadMethod_e dictLoadMethod) -{ - DEBUGLOG(5, "sizeof(ZSTD_CDict) : %u", (unsigned)sizeof(ZSTD_CDict)); - return ZSTD_cwksp_alloc_size(sizeof(ZSTD_CDict)) - + ZSTD_cwksp_alloc_size(HUF_WORKSPACE_SIZE) - /* enableDedicatedDictSearch == 1 ensures that CDict estimation will not be too small - * in case we are using DDS with row-hash. */ - + ZSTD_sizeof_matchState(&cParams, ZSTD_resolveRowMatchFinderMode(ZSTD_ps_auto, &cParams), - /* enableDedicatedDictSearch */ 1, /* forCCtx */ 0) - + (dictLoadMethod == ZSTD_dlm_byRef ? 0 - : ZSTD_cwksp_alloc_size(ZSTD_cwksp_align(dictSize, sizeof(void *)))); -} - -size_t ZSTD_estimateCDictSize(size_t dictSize, int compressionLevel) -{ - ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); - return ZSTD_estimateCDictSize_advanced(dictSize, cParams, ZSTD_dlm_byCopy); -} - -size_t ZSTD_sizeof_CDict(const ZSTD_CDict* cdict) -{ - if (cdict==NULL) return 0; /* support sizeof on NULL */ - DEBUGLOG(5, "sizeof(*cdict) : %u", (unsigned)sizeof(*cdict)); - /* cdict may be in the workspace */ - return (cdict->workspace.workspace == cdict ? 0 : sizeof(*cdict)) - + ZSTD_cwksp_sizeof(&cdict->workspace); -} - -static size_t ZSTD_initCDict_internal( - ZSTD_CDict* cdict, - const void* dictBuffer, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - ZSTD_CCtx_params params) -{ - DEBUGLOG(3, "ZSTD_initCDict_internal (dictContentType:%u)", (unsigned)dictContentType); - assert(!ZSTD_checkCParams(params.cParams)); - cdict->matchState.cParams = params.cParams; - cdict->matchState.dedicatedDictSearch = params.enableDedicatedDictSearch; - if ((dictLoadMethod == ZSTD_dlm_byRef) || (!dictBuffer) || (!dictSize)) { - cdict->dictContent = dictBuffer; - } else { - void *internalBuffer = ZSTD_cwksp_reserve_object(&cdict->workspace, ZSTD_cwksp_align(dictSize, sizeof(void*))); - RETURN_ERROR_IF(!internalBuffer, memory_allocation, "NULL pointer!"); - cdict->dictContent = internalBuffer; - ZSTD_memcpy(internalBuffer, dictBuffer, dictSize); - } - cdict->dictContentSize = dictSize; - cdict->dictContentType = dictContentType; - - cdict->entropyWorkspace = (U32*)ZSTD_cwksp_reserve_object(&cdict->workspace, HUF_WORKSPACE_SIZE); - - - /* Reset the state to no dictionary */ - ZSTD_reset_compressedBlockState(&cdict->cBlockState); - FORWARD_IF_ERROR(ZSTD_reset_matchState( - &cdict->matchState, - &cdict->workspace, - ¶ms.cParams, - params.useRowMatchFinder, - ZSTDcrp_makeClean, - ZSTDirp_reset, - ZSTD_resetTarget_CDict), ""); - /* (Maybe) load the dictionary - * Skips loading the dictionary if it is < 8 bytes. - */ - { params.compressionLevel = ZSTD_CLEVEL_DEFAULT; - params.fParams.contentSizeFlag = 1; - { size_t const dictID = ZSTD_compress_insertDictionary( - &cdict->cBlockState, &cdict->matchState, NULL, &cdict->workspace, - ¶ms, cdict->dictContent, cdict->dictContentSize, - dictContentType, ZSTD_dtlm_full, cdict->entropyWorkspace); - FORWARD_IF_ERROR(dictID, "ZSTD_compress_insertDictionary failed"); - assert(dictID <= (size_t)(U32)-1); - cdict->dictID = (U32)dictID; - } - } - - return 0; -} - -static ZSTD_CDict* ZSTD_createCDict_advanced_internal(size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_compressionParameters cParams, - ZSTD_paramSwitch_e useRowMatchFinder, - U32 enableDedicatedDictSearch, - ZSTD_customMem customMem) -{ - if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; - - { size_t const workspaceSize = - ZSTD_cwksp_alloc_size(sizeof(ZSTD_CDict)) + - ZSTD_cwksp_alloc_size(HUF_WORKSPACE_SIZE) + - ZSTD_sizeof_matchState(&cParams, useRowMatchFinder, enableDedicatedDictSearch, /* forCCtx */ 0) + - (dictLoadMethod == ZSTD_dlm_byRef ? 0 - : ZSTD_cwksp_alloc_size(ZSTD_cwksp_align(dictSize, sizeof(void*)))); - void* const workspace = ZSTD_customMalloc(workspaceSize, customMem); - ZSTD_cwksp ws; - ZSTD_CDict* cdict; - - if (!workspace) { - ZSTD_customFree(workspace, customMem); - return NULL; - } - - ZSTD_cwksp_init(&ws, workspace, workspaceSize, ZSTD_cwksp_dynamic_alloc); - - cdict = (ZSTD_CDict*)ZSTD_cwksp_reserve_object(&ws, sizeof(ZSTD_CDict)); - assert(cdict != NULL); - ZSTD_cwksp_move(&cdict->workspace, &ws); - cdict->customMem = customMem; - cdict->compressionLevel = ZSTD_NO_CLEVEL; /* signals advanced API usage */ - cdict->useRowMatchFinder = useRowMatchFinder; - return cdict; - } -} - -ZSTD_CDict* ZSTD_createCDict_advanced(const void* dictBuffer, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - ZSTD_compressionParameters cParams, - ZSTD_customMem customMem) -{ - ZSTD_CCtx_params cctxParams; - ZSTD_memset(&cctxParams, 0, sizeof(cctxParams)); - ZSTD_CCtxParams_init(&cctxParams, 0); - cctxParams.cParams = cParams; - cctxParams.customMem = customMem; - return ZSTD_createCDict_advanced2( - dictBuffer, dictSize, - dictLoadMethod, dictContentType, - &cctxParams, customMem); -} - -ZSTD_CDict* ZSTD_createCDict_advanced2( - const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - const ZSTD_CCtx_params* originalCctxParams, - ZSTD_customMem customMem) -{ - ZSTD_CCtx_params cctxParams = *originalCctxParams; - ZSTD_compressionParameters cParams; - ZSTD_CDict* cdict; - - DEBUGLOG(3, "ZSTD_createCDict_advanced2, mode %u", (unsigned)dictContentType); - if (!customMem.customAlloc ^ !customMem.customFree) return NULL; - - if (cctxParams.enableDedicatedDictSearch) { - cParams = ZSTD_dedicatedDictSearch_getCParams( - cctxParams.compressionLevel, dictSize); - ZSTD_overrideCParams(&cParams, &cctxParams.cParams); - } else { - cParams = ZSTD_getCParamsFromCCtxParams( - &cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); - } - - if (!ZSTD_dedicatedDictSearch_isSupported(&cParams)) { - /* Fall back to non-DDSS params */ - cctxParams.enableDedicatedDictSearch = 0; - cParams = ZSTD_getCParamsFromCCtxParams( - &cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); - } - - DEBUGLOG(3, "ZSTD_createCDict_advanced2: DDS: %u", cctxParams.enableDedicatedDictSearch); - cctxParams.cParams = cParams; - cctxParams.useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(cctxParams.useRowMatchFinder, &cParams); - - cdict = ZSTD_createCDict_advanced_internal(dictSize, - dictLoadMethod, cctxParams.cParams, - cctxParams.useRowMatchFinder, cctxParams.enableDedicatedDictSearch, - customMem); - - if (ZSTD_isError( ZSTD_initCDict_internal(cdict, - dict, dictSize, - dictLoadMethod, dictContentType, - cctxParams) )) { - ZSTD_freeCDict(cdict); - return NULL; - } - - return cdict; -} - -ZSTD_CDict* ZSTD_createCDict(const void* dict, size_t dictSize, int compressionLevel) -{ - ZSTD_compressionParameters cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); - ZSTD_CDict* const cdict = ZSTD_createCDict_advanced(dict, dictSize, - ZSTD_dlm_byCopy, ZSTD_dct_auto, - cParams, ZSTD_defaultCMem); - if (cdict) - cdict->compressionLevel = (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : compressionLevel; - return cdict; -} - -ZSTD_CDict* ZSTD_createCDict_byReference(const void* dict, size_t dictSize, int compressionLevel) -{ - ZSTD_compressionParameters cParams = ZSTD_getCParams_internal(compressionLevel, ZSTD_CONTENTSIZE_UNKNOWN, dictSize, ZSTD_cpm_createCDict); - ZSTD_CDict* const cdict = ZSTD_createCDict_advanced(dict, dictSize, - ZSTD_dlm_byRef, ZSTD_dct_auto, - cParams, ZSTD_defaultCMem); - if (cdict) - cdict->compressionLevel = (compressionLevel == 0) ? ZSTD_CLEVEL_DEFAULT : compressionLevel; - return cdict; -} - -size_t ZSTD_freeCDict(ZSTD_CDict* cdict) -{ - if (cdict==NULL) return 0; /* support free on NULL */ - { ZSTD_customMem const cMem = cdict->customMem; - int cdictInWorkspace = ZSTD_cwksp_owns_buffer(&cdict->workspace, cdict); - ZSTD_cwksp_free(&cdict->workspace, cMem); - if (!cdictInWorkspace) { - ZSTD_customFree(cdict, cMem); - } - return 0; - } -} - -/*! ZSTD_initStaticCDict_advanced() : - * Generate a digested dictionary in provided memory area. - * workspace: The memory area to emplace the dictionary into. - * Provided pointer must 8-bytes aligned. - * It must outlive dictionary usage. - * workspaceSize: Use ZSTD_estimateCDictSize() - * to determine how large workspace must be. - * cParams : use ZSTD_getCParams() to transform a compression level - * into its relevants cParams. - * @return : pointer to ZSTD_CDict*, or NULL if error (size too small) - * Note : there is no corresponding "free" function. - * Since workspace was allocated externally, it must be freed externally. - */ -const ZSTD_CDict* ZSTD_initStaticCDict( - void* workspace, size_t workspaceSize, - const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - ZSTD_compressionParameters cParams) -{ - ZSTD_paramSwitch_e const useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(ZSTD_ps_auto, &cParams); - /* enableDedicatedDictSearch == 1 ensures matchstate is not too small in case this CDict will be used for DDS + row hash */ - size_t const matchStateSize = ZSTD_sizeof_matchState(&cParams, useRowMatchFinder, /* enableDedicatedDictSearch */ 1, /* forCCtx */ 0); - size_t const neededSize = ZSTD_cwksp_alloc_size(sizeof(ZSTD_CDict)) - + (dictLoadMethod == ZSTD_dlm_byRef ? 0 - : ZSTD_cwksp_alloc_size(ZSTD_cwksp_align(dictSize, sizeof(void*)))) - + ZSTD_cwksp_alloc_size(HUF_WORKSPACE_SIZE) - + matchStateSize; - ZSTD_CDict* cdict; - ZSTD_CCtx_params params; - - if ((size_t)workspace & 7) return NULL; /* 8-aligned */ - - { - ZSTD_cwksp ws; - ZSTD_cwksp_init(&ws, workspace, workspaceSize, ZSTD_cwksp_static_alloc); - cdict = (ZSTD_CDict*)ZSTD_cwksp_reserve_object(&ws, sizeof(ZSTD_CDict)); - if (cdict == NULL) return NULL; - ZSTD_cwksp_move(&cdict->workspace, &ws); - } - - DEBUGLOG(4, "(workspaceSize < neededSize) : (%u < %u) => %u", - (unsigned)workspaceSize, (unsigned)neededSize, (unsigned)(workspaceSize < neededSize)); - if (workspaceSize < neededSize) return NULL; - - ZSTD_CCtxParams_init(¶ms, 0); - params.cParams = cParams; - params.useRowMatchFinder = useRowMatchFinder; - cdict->useRowMatchFinder = useRowMatchFinder; - - if (ZSTD_isError( ZSTD_initCDict_internal(cdict, - dict, dictSize, - dictLoadMethod, dictContentType, - params) )) - return NULL; - - return cdict; -} - -ZSTD_compressionParameters ZSTD_getCParamsFromCDict(const ZSTD_CDict* cdict) -{ - assert(cdict != NULL); - return cdict->matchState.cParams; -} - -/*! ZSTD_getDictID_fromCDict() : - * Provides the dictID of the dictionary loaded into `cdict`. - * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty. - * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */ -unsigned ZSTD_getDictID_fromCDict(const ZSTD_CDict* cdict) -{ - if (cdict==NULL) return 0; - return cdict->dictID; -} - -/* ZSTD_compressBegin_usingCDict_internal() : - * Implementation of various ZSTD_compressBegin_usingCDict* functions. - */ -static size_t ZSTD_compressBegin_usingCDict_internal( - ZSTD_CCtx* const cctx, const ZSTD_CDict* const cdict, - ZSTD_frameParameters const fParams, unsigned long long const pledgedSrcSize) -{ - ZSTD_CCtx_params cctxParams; - DEBUGLOG(4, "ZSTD_compressBegin_usingCDict_internal"); - RETURN_ERROR_IF(cdict==NULL, dictionary_wrong, "NULL pointer!"); - /* Initialize the cctxParams from the cdict */ - { - ZSTD_parameters params; - params.fParams = fParams; - params.cParams = ( pledgedSrcSize < ZSTD_USE_CDICT_PARAMS_SRCSIZE_CUTOFF - || pledgedSrcSize < cdict->dictContentSize * ZSTD_USE_CDICT_PARAMS_DICTSIZE_MULTIPLIER - || pledgedSrcSize == ZSTD_CONTENTSIZE_UNKNOWN - || cdict->compressionLevel == 0 ) ? - ZSTD_getCParamsFromCDict(cdict) - : ZSTD_getCParams(cdict->compressionLevel, - pledgedSrcSize, - cdict->dictContentSize); - ZSTD_CCtxParams_init_internal(&cctxParams, ¶ms, cdict->compressionLevel); - } - /* Increase window log to fit the entire dictionary and source if the - * source size is known. Limit the increase to 19, which is the - * window log for compression level 1 with the largest source size. - */ - if (pledgedSrcSize != ZSTD_CONTENTSIZE_UNKNOWN) { - U32 const limitedSrcSize = (U32)MIN(pledgedSrcSize, 1U << 19); - U32 const limitedSrcLog = limitedSrcSize > 1 ? ZSTD_highbit32(limitedSrcSize - 1) + 1 : 1; - cctxParams.cParams.windowLog = MAX(cctxParams.cParams.windowLog, limitedSrcLog); - } - return ZSTD_compressBegin_internal(cctx, - NULL, 0, ZSTD_dct_auto, ZSTD_dtlm_fast, - cdict, - &cctxParams, pledgedSrcSize, - ZSTDb_not_buffered); -} - - -/* ZSTD_compressBegin_usingCDict_advanced() : - * This function is DEPRECATED. - * cdict must be != NULL */ -size_t ZSTD_compressBegin_usingCDict_advanced( - ZSTD_CCtx* const cctx, const ZSTD_CDict* const cdict, - ZSTD_frameParameters const fParams, unsigned long long const pledgedSrcSize) -{ - return ZSTD_compressBegin_usingCDict_internal(cctx, cdict, fParams, pledgedSrcSize); -} - -/* ZSTD_compressBegin_usingCDict() : - * cdict must be != NULL */ -size_t ZSTD_compressBegin_usingCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict) -{ - ZSTD_frameParameters const fParams = { 0 /*content*/, 0 /*checksum*/, 0 /*noDictID*/ }; - return ZSTD_compressBegin_usingCDict_internal(cctx, cdict, fParams, ZSTD_CONTENTSIZE_UNKNOWN); -} - -/*! ZSTD_compress_usingCDict_internal(): - * Implementation of various ZSTD_compress_usingCDict* functions. - */ -static size_t ZSTD_compress_usingCDict_internal(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const ZSTD_CDict* cdict, ZSTD_frameParameters fParams) -{ - FORWARD_IF_ERROR(ZSTD_compressBegin_usingCDict_internal(cctx, cdict, fParams, srcSize), ""); /* will check if cdict != NULL */ - return ZSTD_compressEnd(cctx, dst, dstCapacity, src, srcSize); -} - -/*! ZSTD_compress_usingCDict_advanced(): - * This function is DEPRECATED. - */ -size_t ZSTD_compress_usingCDict_advanced(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const ZSTD_CDict* cdict, ZSTD_frameParameters fParams) -{ - return ZSTD_compress_usingCDict_internal(cctx, dst, dstCapacity, src, srcSize, cdict, fParams); -} - -/*! ZSTD_compress_usingCDict() : - * Compression using a digested Dictionary. - * Faster startup than ZSTD_compress_usingDict(), recommended when same dictionary is used multiple times. - * Note that compression parameters are decided at CDict creation time - * while frame parameters are hardcoded */ -size_t ZSTD_compress_usingCDict(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const ZSTD_CDict* cdict) -{ - ZSTD_frameParameters const fParams = { 1 /*content*/, 0 /*checksum*/, 0 /*noDictID*/ }; - return ZSTD_compress_usingCDict_internal(cctx, dst, dstCapacity, src, srcSize, cdict, fParams); -} - - - -/* ****************************************************************** -* Streaming -********************************************************************/ - -ZSTD_CStream* ZSTD_createCStream(void) -{ - DEBUGLOG(3, "ZSTD_createCStream"); - return ZSTD_createCStream_advanced(ZSTD_defaultCMem); -} - -ZSTD_CStream* ZSTD_initStaticCStream(void *workspace, size_t workspaceSize) -{ - return ZSTD_initStaticCCtx(workspace, workspaceSize); -} - -ZSTD_CStream* ZSTD_createCStream_advanced(ZSTD_customMem customMem) -{ /* CStream and CCtx are now same object */ - return ZSTD_createCCtx_advanced(customMem); -} - -size_t ZSTD_freeCStream(ZSTD_CStream* zcs) -{ - return ZSTD_freeCCtx(zcs); /* same object */ -} - - - -/*====== Initialization ======*/ - -size_t ZSTD_CStreamInSize(void) { return ZSTD_BLOCKSIZE_MAX; } - -size_t ZSTD_CStreamOutSize(void) -{ - return ZSTD_compressBound(ZSTD_BLOCKSIZE_MAX) + ZSTD_blockHeaderSize + 4 /* 32-bits hash */ ; -} - -static ZSTD_cParamMode_e ZSTD_getCParamMode(ZSTD_CDict const* cdict, ZSTD_CCtx_params const* params, U64 pledgedSrcSize) -{ - if (cdict != NULL && ZSTD_shouldAttachDict(cdict, params, pledgedSrcSize)) - return ZSTD_cpm_attachDict; - else - return ZSTD_cpm_noAttachDict; -} - -/* ZSTD_resetCStream(): - * pledgedSrcSize == 0 means "unknown" */ -size_t ZSTD_resetCStream(ZSTD_CStream* zcs, unsigned long long pss) -{ - /* temporary : 0 interpreted as "unknown" during transition period. - * Users willing to specify "unknown" **must** use ZSTD_CONTENTSIZE_UNKNOWN. - * 0 will be interpreted as "empty" in the future. - */ - U64 const pledgedSrcSize = (pss==0) ? ZSTD_CONTENTSIZE_UNKNOWN : pss; - DEBUGLOG(4, "ZSTD_resetCStream: pledgedSrcSize = %u", (unsigned)pledgedSrcSize); - FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) , ""); - FORWARD_IF_ERROR( ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize) , ""); - return 0; -} - -/*! ZSTD_initCStream_internal() : - * Note : for lib/compress only. Used by zstdmt_compress.c. - * Assumption 1 : params are valid - * Assumption 2 : either dict, or cdict, is defined, not both */ -size_t ZSTD_initCStream_internal(ZSTD_CStream* zcs, - const void* dict, size_t dictSize, const ZSTD_CDict* cdict, - const ZSTD_CCtx_params* params, - unsigned long long pledgedSrcSize) -{ - DEBUGLOG(4, "ZSTD_initCStream_internal"); - FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) , ""); - FORWARD_IF_ERROR( ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize) , ""); - assert(!ZSTD_isError(ZSTD_checkCParams(params->cParams))); - zcs->requestedParams = *params; - assert(!((dict) && (cdict))); /* either dict or cdict, not both */ - if (dict) { - FORWARD_IF_ERROR( ZSTD_CCtx_loadDictionary(zcs, dict, dictSize) , ""); - } else { - /* Dictionary is cleared if !cdict */ - FORWARD_IF_ERROR( ZSTD_CCtx_refCDict(zcs, cdict) , ""); - } - return 0; -} - -/* ZSTD_initCStream_usingCDict_advanced() : - * same as ZSTD_initCStream_usingCDict(), with control over frame parameters */ -size_t ZSTD_initCStream_usingCDict_advanced(ZSTD_CStream* zcs, - const ZSTD_CDict* cdict, - ZSTD_frameParameters fParams, - unsigned long long pledgedSrcSize) -{ - DEBUGLOG(4, "ZSTD_initCStream_usingCDict_advanced"); - FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) , ""); - FORWARD_IF_ERROR( ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize) , ""); - zcs->requestedParams.fParams = fParams; - FORWARD_IF_ERROR( ZSTD_CCtx_refCDict(zcs, cdict) , ""); - return 0; -} - -/* note : cdict must outlive compression session */ -size_t ZSTD_initCStream_usingCDict(ZSTD_CStream* zcs, const ZSTD_CDict* cdict) -{ - DEBUGLOG(4, "ZSTD_initCStream_usingCDict"); - FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) , ""); - FORWARD_IF_ERROR( ZSTD_CCtx_refCDict(zcs, cdict) , ""); - return 0; -} - - -/* ZSTD_initCStream_advanced() : - * pledgedSrcSize must be exact. - * if srcSize is not known at init time, use value ZSTD_CONTENTSIZE_UNKNOWN. - * dict is loaded with default parameters ZSTD_dct_auto and ZSTD_dlm_byCopy. */ -size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs, - const void* dict, size_t dictSize, - ZSTD_parameters params, unsigned long long pss) -{ - /* for compatibility with older programs relying on this behavior. - * Users should now specify ZSTD_CONTENTSIZE_UNKNOWN. - * This line will be removed in the future. - */ - U64 const pledgedSrcSize = (pss==0 && params.fParams.contentSizeFlag==0) ? ZSTD_CONTENTSIZE_UNKNOWN : pss; - DEBUGLOG(4, "ZSTD_initCStream_advanced"); - FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) , ""); - FORWARD_IF_ERROR( ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize) , ""); - FORWARD_IF_ERROR( ZSTD_checkCParams(params.cParams) , ""); - ZSTD_CCtxParams_setZstdParams(&zcs->requestedParams, ¶ms); - FORWARD_IF_ERROR( ZSTD_CCtx_loadDictionary(zcs, dict, dictSize) , ""); - return 0; -} - -size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, const void* dict, size_t dictSize, int compressionLevel) -{ - DEBUGLOG(4, "ZSTD_initCStream_usingDict"); - FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) , ""); - FORWARD_IF_ERROR( ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel) , ""); - FORWARD_IF_ERROR( ZSTD_CCtx_loadDictionary(zcs, dict, dictSize) , ""); - return 0; -} - -size_t ZSTD_initCStream_srcSize(ZSTD_CStream* zcs, int compressionLevel, unsigned long long pss) -{ - /* temporary : 0 interpreted as "unknown" during transition period. - * Users willing to specify "unknown" **must** use ZSTD_CONTENTSIZE_UNKNOWN. - * 0 will be interpreted as "empty" in the future. - */ - U64 const pledgedSrcSize = (pss==0) ? ZSTD_CONTENTSIZE_UNKNOWN : pss; - DEBUGLOG(4, "ZSTD_initCStream_srcSize"); - FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) , ""); - FORWARD_IF_ERROR( ZSTD_CCtx_refCDict(zcs, NULL) , ""); - FORWARD_IF_ERROR( ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel) , ""); - FORWARD_IF_ERROR( ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize) , ""); - return 0; -} - -size_t ZSTD_initCStream(ZSTD_CStream* zcs, int compressionLevel) -{ - DEBUGLOG(4, "ZSTD_initCStream"); - FORWARD_IF_ERROR( ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only) , ""); - FORWARD_IF_ERROR( ZSTD_CCtx_refCDict(zcs, NULL) , ""); - FORWARD_IF_ERROR( ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel) , ""); - return 0; -} - -/*====== Compression ======*/ - -static size_t ZSTD_nextInputSizeHint(const ZSTD_CCtx* cctx) -{ - size_t hintInSize = cctx->inBuffTarget - cctx->inBuffPos; - if (hintInSize==0) hintInSize = cctx->blockSize; - return hintInSize; -} - -/** ZSTD_compressStream_generic(): - * internal function for all *compressStream*() variants - * non-static, because can be called from zstdmt_compress.c - * @return : hint size for next input */ -static size_t ZSTD_compressStream_generic(ZSTD_CStream* zcs, - ZSTD_outBuffer* output, - ZSTD_inBuffer* input, - ZSTD_EndDirective const flushMode) -{ - const char* const istart = (const char*)input->src; - const char* const iend = input->size != 0 ? istart + input->size : istart; - const char* ip = input->pos != 0 ? istart + input->pos : istart; - char* const ostart = (char*)output->dst; - char* const oend = output->size != 0 ? ostart + output->size : ostart; - char* op = output->pos != 0 ? ostart + output->pos : ostart; - U32 someMoreWork = 1; - - /* check expectations */ - DEBUGLOG(5, "ZSTD_compressStream_generic, flush=%u", (unsigned)flushMode); - if (zcs->appliedParams.inBufferMode == ZSTD_bm_buffered) { - assert(zcs->inBuff != NULL); - assert(zcs->inBuffSize > 0); - } - if (zcs->appliedParams.outBufferMode == ZSTD_bm_buffered) { - assert(zcs->outBuff != NULL); - assert(zcs->outBuffSize > 0); - } - assert(output->pos <= output->size); - assert(input->pos <= input->size); - assert((U32)flushMode <= (U32)ZSTD_e_end); - - while (someMoreWork) { - switch(zcs->streamStage) - { - case zcss_init: - RETURN_ERROR(init_missing, "call ZSTD_initCStream() first!"); - - case zcss_load: - if ( (flushMode == ZSTD_e_end) - && ( (size_t)(oend-op) >= ZSTD_compressBound(iend-ip) /* Enough output space */ - || zcs->appliedParams.outBufferMode == ZSTD_bm_stable) /* OR we are allowed to return dstSizeTooSmall */ - && (zcs->inBuffPos == 0) ) { - /* shortcut to compression pass directly into output buffer */ - size_t const cSize = ZSTD_compressEnd(zcs, - op, oend-op, ip, iend-ip); - DEBUGLOG(4, "ZSTD_compressEnd : cSize=%u", (unsigned)cSize); - FORWARD_IF_ERROR(cSize, "ZSTD_compressEnd failed"); - ip = iend; - op += cSize; - zcs->frameEnded = 1; - ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); - someMoreWork = 0; break; - } - /* complete loading into inBuffer in buffered mode */ - if (zcs->appliedParams.inBufferMode == ZSTD_bm_buffered) { - size_t const toLoad = zcs->inBuffTarget - zcs->inBuffPos; - size_t const loaded = ZSTD_limitCopy( - zcs->inBuff + zcs->inBuffPos, toLoad, - ip, iend-ip); - zcs->inBuffPos += loaded; - if (loaded != 0) - ip += loaded; - if ( (flushMode == ZSTD_e_continue) - && (zcs->inBuffPos < zcs->inBuffTarget) ) { - /* not enough input to fill full block : stop here */ - someMoreWork = 0; break; - } - if ( (flushMode == ZSTD_e_flush) - && (zcs->inBuffPos == zcs->inToCompress) ) { - /* empty */ - someMoreWork = 0; break; - } - } - /* compress current block (note : this stage cannot be stopped in the middle) */ - DEBUGLOG(5, "stream compression stage (flushMode==%u)", flushMode); - { int const inputBuffered = (zcs->appliedParams.inBufferMode == ZSTD_bm_buffered); - void* cDst; - size_t cSize; - size_t oSize = oend-op; - size_t const iSize = inputBuffered - ? zcs->inBuffPos - zcs->inToCompress - : MIN((size_t)(iend - ip), zcs->blockSize); - if (oSize >= ZSTD_compressBound(iSize) || zcs->appliedParams.outBufferMode == ZSTD_bm_stable) - cDst = op; /* compress into output buffer, to skip flush stage */ - else - cDst = zcs->outBuff, oSize = zcs->outBuffSize; - if (inputBuffered) { - unsigned const lastBlock = (flushMode == ZSTD_e_end) && (ip==iend); - cSize = lastBlock ? - ZSTD_compressEnd(zcs, cDst, oSize, - zcs->inBuff + zcs->inToCompress, iSize) : - ZSTD_compressContinue(zcs, cDst, oSize, - zcs->inBuff + zcs->inToCompress, iSize); - FORWARD_IF_ERROR(cSize, "%s", lastBlock ? "ZSTD_compressEnd failed" : "ZSTD_compressContinue failed"); - zcs->frameEnded = lastBlock; - /* prepare next block */ - zcs->inBuffTarget = zcs->inBuffPos + zcs->blockSize; - if (zcs->inBuffTarget > zcs->inBuffSize) - zcs->inBuffPos = 0, zcs->inBuffTarget = zcs->blockSize; - DEBUGLOG(5, "inBuffTarget:%u / inBuffSize:%u", - (unsigned)zcs->inBuffTarget, (unsigned)zcs->inBuffSize); - if (!lastBlock) - assert(zcs->inBuffTarget <= zcs->inBuffSize); - zcs->inToCompress = zcs->inBuffPos; - } else { - unsigned const lastBlock = (ip + iSize == iend); - assert(flushMode == ZSTD_e_end /* Already validated */); - cSize = lastBlock ? - ZSTD_compressEnd(zcs, cDst, oSize, ip, iSize) : - ZSTD_compressContinue(zcs, cDst, oSize, ip, iSize); - /* Consume the input prior to error checking to mirror buffered mode. */ - if (iSize > 0) - ip += iSize; - FORWARD_IF_ERROR(cSize, "%s", lastBlock ? "ZSTD_compressEnd failed" : "ZSTD_compressContinue failed"); - zcs->frameEnded = lastBlock; - if (lastBlock) - assert(ip == iend); - } - if (cDst == op) { /* no need to flush */ - op += cSize; - if (zcs->frameEnded) { - DEBUGLOG(5, "Frame completed directly in outBuffer"); - someMoreWork = 0; - ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); - } - break; - } - zcs->outBuffContentSize = cSize; - zcs->outBuffFlushedSize = 0; - zcs->streamStage = zcss_flush; /* pass-through to flush stage */ - } - ZSTD_FALLTHROUGH; - case zcss_flush: - DEBUGLOG(5, "flush stage"); - assert(zcs->appliedParams.outBufferMode == ZSTD_bm_buffered); - { size_t const toFlush = zcs->outBuffContentSize - zcs->outBuffFlushedSize; - size_t const flushed = ZSTD_limitCopy(op, (size_t)(oend-op), - zcs->outBuff + zcs->outBuffFlushedSize, toFlush); - DEBUGLOG(5, "toFlush: %u into %u ==> flushed: %u", - (unsigned)toFlush, (unsigned)(oend-op), (unsigned)flushed); - if (flushed) - op += flushed; - zcs->outBuffFlushedSize += flushed; - if (toFlush!=flushed) { - /* flush not fully completed, presumably because dst is too small */ - assert(op==oend); - someMoreWork = 0; - break; - } - zcs->outBuffContentSize = zcs->outBuffFlushedSize = 0; - if (zcs->frameEnded) { - DEBUGLOG(5, "Frame completed on flush"); - someMoreWork = 0; - ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); - break; - } - zcs->streamStage = zcss_load; - break; - } - - default: /* impossible */ - assert(0); - } - } - - input->pos = ip - istart; - output->pos = op - ostart; - if (zcs->frameEnded) return 0; - return ZSTD_nextInputSizeHint(zcs); -} - -static size_t ZSTD_nextInputSizeHint_MTorST(const ZSTD_CCtx* cctx) -{ -#ifdef ZSTD_MULTITHREAD - if (cctx->appliedParams.nbWorkers >= 1) { - assert(cctx->mtctx != NULL); - return ZSTDMT_nextInputSizeHint(cctx->mtctx); - } -#endif - return ZSTD_nextInputSizeHint(cctx); - -} - -size_t ZSTD_compressStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output, ZSTD_inBuffer* input) -{ - FORWARD_IF_ERROR( ZSTD_compressStream2(zcs, output, input, ZSTD_e_continue) , ""); - return ZSTD_nextInputSizeHint_MTorST(zcs); -} - -/* After a compression call set the expected input/output buffer. - * This is validated at the start of the next compression call. - */ -static void ZSTD_setBufferExpectations(ZSTD_CCtx* cctx, ZSTD_outBuffer const* output, ZSTD_inBuffer const* input) -{ - if (cctx->appliedParams.inBufferMode == ZSTD_bm_stable) { - cctx->expectedInBuffer = *input; - } - if (cctx->appliedParams.outBufferMode == ZSTD_bm_stable) { - cctx->expectedOutBufferSize = output->size - output->pos; - } -} - -/* Validate that the input/output buffers match the expectations set by - * ZSTD_setBufferExpectations. - */ -static size_t ZSTD_checkBufferStability(ZSTD_CCtx const* cctx, - ZSTD_outBuffer const* output, - ZSTD_inBuffer const* input, - ZSTD_EndDirective endOp) -{ - if (cctx->appliedParams.inBufferMode == ZSTD_bm_stable) { - ZSTD_inBuffer const expect = cctx->expectedInBuffer; - if (expect.src != input->src || expect.pos != input->pos || expect.size != input->size) - RETURN_ERROR(srcBuffer_wrong, "ZSTD_c_stableInBuffer enabled but input differs!"); - if (endOp != ZSTD_e_end) - RETURN_ERROR(srcBuffer_wrong, "ZSTD_c_stableInBuffer can only be used with ZSTD_e_end!"); - } - if (cctx->appliedParams.outBufferMode == ZSTD_bm_stable) { - size_t const outBufferSize = output->size - output->pos; - if (cctx->expectedOutBufferSize != outBufferSize) - RETURN_ERROR(dstBuffer_wrong, "ZSTD_c_stableOutBuffer enabled but output size differs!"); - } - return 0; -} - -static size_t ZSTD_CCtx_init_compressStream2(ZSTD_CCtx* cctx, - ZSTD_EndDirective endOp, - size_t inSize) { - ZSTD_CCtx_params params = cctx->requestedParams; - ZSTD_prefixDict const prefixDict = cctx->prefixDict; - FORWARD_IF_ERROR( ZSTD_initLocalDict(cctx) , ""); /* Init the local dict if present. */ - ZSTD_memset(&cctx->prefixDict, 0, sizeof(cctx->prefixDict)); /* single usage */ - assert(prefixDict.dict==NULL || cctx->cdict==NULL); /* only one can be set */ - if (cctx->cdict && !cctx->localDict.cdict) { - /* Let the cdict's compression level take priority over the requested params. - * But do not take the cdict's compression level if the "cdict" is actually a localDict - * generated from ZSTD_initLocalDict(). - */ - params.compressionLevel = cctx->cdict->compressionLevel; - } - DEBUGLOG(4, "ZSTD_compressStream2 : transparent init stage"); - if (endOp == ZSTD_e_end) cctx->pledgedSrcSizePlusOne = inSize + 1; /* auto-fix pledgedSrcSize */ - { - size_t const dictSize = prefixDict.dict - ? prefixDict.dictSize - : (cctx->cdict ? cctx->cdict->dictContentSize : 0); - ZSTD_cParamMode_e const mode = ZSTD_getCParamMode(cctx->cdict, ¶ms, cctx->pledgedSrcSizePlusOne - 1); - params.cParams = ZSTD_getCParamsFromCCtxParams( - ¶ms, cctx->pledgedSrcSizePlusOne-1, - dictSize, mode); - } - - params.useBlockSplitter = ZSTD_resolveBlockSplitterMode(params.useBlockSplitter, ¶ms.cParams); - params.ldmParams.enableLdm = ZSTD_resolveEnableLdm(params.ldmParams.enableLdm, ¶ms.cParams); - params.useRowMatchFinder = ZSTD_resolveRowMatchFinderMode(params.useRowMatchFinder, ¶ms.cParams); - -#ifdef ZSTD_MULTITHREAD - if ((cctx->pledgedSrcSizePlusOne-1) <= ZSTDMT_JOBSIZE_MIN) { - params.nbWorkers = 0; /* do not invoke multi-threading when src size is too small */ - } - if (params.nbWorkers > 0) { -#if ZSTD_TRACE - cctx->traceCtx = (ZSTD_trace_compress_begin != NULL) ? ZSTD_trace_compress_begin(cctx) : 0; -#endif - /* mt context creation */ - if (cctx->mtctx == NULL) { - DEBUGLOG(4, "ZSTD_compressStream2: creating new mtctx for nbWorkers=%u", - params.nbWorkers); - cctx->mtctx = ZSTDMT_createCCtx_advanced((U32)params.nbWorkers, cctx->customMem, cctx->pool); - RETURN_ERROR_IF(cctx->mtctx == NULL, memory_allocation, "NULL pointer!"); - } - /* mt compression */ - DEBUGLOG(4, "call ZSTDMT_initCStream_internal as nbWorkers=%u", params.nbWorkers); - FORWARD_IF_ERROR( ZSTDMT_initCStream_internal( - cctx->mtctx, - prefixDict.dict, prefixDict.dictSize, prefixDict.dictContentType, - cctx->cdict, params, cctx->pledgedSrcSizePlusOne-1) , ""); - cctx->dictID = cctx->cdict ? cctx->cdict->dictID : 0; - cctx->dictContentSize = cctx->cdict ? cctx->cdict->dictContentSize : prefixDict.dictSize; - cctx->consumedSrcSize = 0; - cctx->producedCSize = 0; - cctx->streamStage = zcss_load; - cctx->appliedParams = params; - } else -#endif - { U64 const pledgedSrcSize = cctx->pledgedSrcSizePlusOne - 1; - assert(!ZSTD_isError(ZSTD_checkCParams(params.cParams))); - FORWARD_IF_ERROR( ZSTD_compressBegin_internal(cctx, - prefixDict.dict, prefixDict.dictSize, prefixDict.dictContentType, ZSTD_dtlm_fast, - cctx->cdict, - ¶ms, pledgedSrcSize, - ZSTDb_buffered) , ""); - assert(cctx->appliedParams.nbWorkers == 0); - cctx->inToCompress = 0; - cctx->inBuffPos = 0; - if (cctx->appliedParams.inBufferMode == ZSTD_bm_buffered) { - /* for small input: avoid automatic flush on reaching end of block, since - * it would require to add a 3-bytes null block to end frame - */ - cctx->inBuffTarget = cctx->blockSize + (cctx->blockSize == pledgedSrcSize); - } else { - cctx->inBuffTarget = 0; - } - cctx->outBuffContentSize = cctx->outBuffFlushedSize = 0; - cctx->streamStage = zcss_load; - cctx->frameEnded = 0; - } - return 0; -} - -size_t ZSTD_compressStream2( ZSTD_CCtx* cctx, - ZSTD_outBuffer* output, - ZSTD_inBuffer* input, - ZSTD_EndDirective endOp) -{ - DEBUGLOG(5, "ZSTD_compressStream2, endOp=%u ", (unsigned)endOp); - /* check conditions */ - RETURN_ERROR_IF(output->pos > output->size, dstSize_tooSmall, "invalid output buffer"); - RETURN_ERROR_IF(input->pos > input->size, srcSize_wrong, "invalid input buffer"); - RETURN_ERROR_IF((U32)endOp > (U32)ZSTD_e_end, parameter_outOfBound, "invalid endDirective"); - assert(cctx != NULL); - - /* transparent initialization stage */ - if (cctx->streamStage == zcss_init) { - FORWARD_IF_ERROR(ZSTD_CCtx_init_compressStream2(cctx, endOp, input->size), "CompressStream2 initialization failed"); - ZSTD_setBufferExpectations(cctx, output, input); /* Set initial buffer expectations now that we've initialized */ - } - /* end of transparent initialization stage */ - - FORWARD_IF_ERROR(ZSTD_checkBufferStability(cctx, output, input, endOp), "invalid buffers"); - /* compression stage */ -#ifdef ZSTD_MULTITHREAD - if (cctx->appliedParams.nbWorkers > 0) { - size_t flushMin; - if (cctx->cParamsChanged) { - ZSTDMT_updateCParams_whileCompressing(cctx->mtctx, &cctx->requestedParams); - cctx->cParamsChanged = 0; - } - for (;;) { - size_t const ipos = input->pos; - size_t const opos = output->pos; - flushMin = ZSTDMT_compressStream_generic(cctx->mtctx, output, input, endOp); - cctx->consumedSrcSize += (U64)(input->pos - ipos); - cctx->producedCSize += (U64)(output->pos - opos); - if ( ZSTD_isError(flushMin) - || (endOp == ZSTD_e_end && flushMin == 0) ) { /* compression completed */ - if (flushMin == 0) - ZSTD_CCtx_trace(cctx, 0); - ZSTD_CCtx_reset(cctx, ZSTD_reset_session_only); - } - FORWARD_IF_ERROR(flushMin, "ZSTDMT_compressStream_generic failed"); - - if (endOp == ZSTD_e_continue) { - /* We only require some progress with ZSTD_e_continue, not maximal progress. - * We're done if we've consumed or produced any bytes, or either buffer is - * full. - */ - if (input->pos != ipos || output->pos != opos || input->pos == input->size || output->pos == output->size) - break; - } else { - assert(endOp == ZSTD_e_flush || endOp == ZSTD_e_end); - /* We require maximal progress. We're done when the flush is complete or the - * output buffer is full. - */ - if (flushMin == 0 || output->pos == output->size) - break; - } - } - DEBUGLOG(5, "completed ZSTD_compressStream2 delegating to ZSTDMT_compressStream_generic"); - /* Either we don't require maximum forward progress, we've finished the - * flush, or we are out of output space. - */ - assert(endOp == ZSTD_e_continue || flushMin == 0 || output->pos == output->size); - ZSTD_setBufferExpectations(cctx, output, input); - return flushMin; - } -#endif - FORWARD_IF_ERROR( ZSTD_compressStream_generic(cctx, output, input, endOp) , ""); - DEBUGLOG(5, "completed ZSTD_compressStream2"); - ZSTD_setBufferExpectations(cctx, output, input); - return cctx->outBuffContentSize - cctx->outBuffFlushedSize; /* remaining to flush */ -} - -size_t ZSTD_compressStream2_simpleArgs ( - ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, size_t* dstPos, - const void* src, size_t srcSize, size_t* srcPos, - ZSTD_EndDirective endOp) -{ - ZSTD_outBuffer output = { dst, dstCapacity, *dstPos }; - ZSTD_inBuffer input = { src, srcSize, *srcPos }; - /* ZSTD_compressStream2() will check validity of dstPos and srcPos */ - size_t const cErr = ZSTD_compressStream2(cctx, &output, &input, endOp); - *dstPos = output.pos; - *srcPos = input.pos; - return cErr; -} - -size_t ZSTD_compress2(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize) -{ - ZSTD_bufferMode_e const originalInBufferMode = cctx->requestedParams.inBufferMode; - ZSTD_bufferMode_e const originalOutBufferMode = cctx->requestedParams.outBufferMode; - DEBUGLOG(4, "ZSTD_compress2 (srcSize=%u)", (unsigned)srcSize); - ZSTD_CCtx_reset(cctx, ZSTD_reset_session_only); - /* Enable stable input/output buffers. */ - cctx->requestedParams.inBufferMode = ZSTD_bm_stable; - cctx->requestedParams.outBufferMode = ZSTD_bm_stable; - { size_t oPos = 0; - size_t iPos = 0; - size_t const result = ZSTD_compressStream2_simpleArgs(cctx, - dst, dstCapacity, &oPos, - src, srcSize, &iPos, - ZSTD_e_end); - /* Reset to the original values. */ - cctx->requestedParams.inBufferMode = originalInBufferMode; - cctx->requestedParams.outBufferMode = originalOutBufferMode; - FORWARD_IF_ERROR(result, "ZSTD_compressStream2_simpleArgs failed"); - if (result != 0) { /* compression not completed, due to lack of output space */ - assert(oPos == dstCapacity); - RETURN_ERROR(dstSize_tooSmall, ""); - } - assert(iPos == srcSize); /* all input is expected consumed */ - return oPos; - } -} - -typedef struct { - U32 idx; /* Index in array of ZSTD_Sequence */ - U32 posInSequence; /* Position within sequence at idx */ - size_t posInSrc; /* Number of bytes given by sequences provided so far */ -} ZSTD_sequencePosition; - -/* ZSTD_validateSequence() : - * @offCode : is presumed to follow format required by ZSTD_storeSeq() - * @returns a ZSTD error code if sequence is not valid - */ -static size_t -ZSTD_validateSequence(U32 offCode, U32 matchLength, - size_t posInSrc, U32 windowLog, size_t dictSize) -{ - U32 const windowSize = 1 << windowLog; - /* posInSrc represents the amount of data the the decoder would decode up to this point. - * As long as the amount of data decoded is less than or equal to window size, offsets may be - * larger than the total length of output decoded in order to reference the dict, even larger than - * window size. After output surpasses windowSize, we're limited to windowSize offsets again. - */ - size_t const offsetBound = posInSrc > windowSize ? (size_t)windowSize : posInSrc + (size_t)dictSize; - RETURN_ERROR_IF(offCode > STORE_OFFSET(offsetBound), corruption_detected, "Offset too large!"); - RETURN_ERROR_IF(matchLength < MINMATCH, corruption_detected, "Matchlength too small"); - return 0; -} - -/* Returns an offset code, given a sequence's raw offset, the ongoing repcode array, and whether litLength == 0 */ -static U32 ZSTD_finalizeOffCode(U32 rawOffset, const U32 rep[ZSTD_REP_NUM], U32 ll0) -{ - U32 offCode = STORE_OFFSET(rawOffset); - - if (!ll0 && rawOffset == rep[0]) { - offCode = STORE_REPCODE_1; - } else if (rawOffset == rep[1]) { - offCode = STORE_REPCODE(2 - ll0); - } else if (rawOffset == rep[2]) { - offCode = STORE_REPCODE(3 - ll0); - } else if (ll0 && rawOffset == rep[0] - 1) { - offCode = STORE_REPCODE_3; - } - return offCode; -} - -/* Returns 0 on success, and a ZSTD_error otherwise. This function scans through an array of - * ZSTD_Sequence, storing the sequences it finds, until it reaches a block delimiter. - */ -static size_t -ZSTD_copySequencesToSeqStoreExplicitBlockDelim(ZSTD_CCtx* cctx, - ZSTD_sequencePosition* seqPos, - const ZSTD_Sequence* const inSeqs, size_t inSeqsSize, - const void* src, size_t blockSize) -{ - U32 idx = seqPos->idx; - BYTE const* ip = (BYTE const*)(src); - const BYTE* const iend = ip + blockSize; - repcodes_t updatedRepcodes; - U32 dictSize; - - if (cctx->cdict) { - dictSize = (U32)cctx->cdict->dictContentSize; - } else if (cctx->prefixDict.dict) { - dictSize = (U32)cctx->prefixDict.dictSize; - } else { - dictSize = 0; - } - ZSTD_memcpy(updatedRepcodes.rep, cctx->blockState.prevCBlock->rep, sizeof(repcodes_t)); - for (; (inSeqs[idx].matchLength != 0 || inSeqs[idx].offset != 0) && idx < inSeqsSize; ++idx) { - U32 const litLength = inSeqs[idx].litLength; - U32 const ll0 = (litLength == 0); - U32 const matchLength = inSeqs[idx].matchLength; - U32 const offCode = ZSTD_finalizeOffCode(inSeqs[idx].offset, updatedRepcodes.rep, ll0); - ZSTD_updateRep(updatedRepcodes.rep, offCode, ll0); - - DEBUGLOG(6, "Storing sequence: (of: %u, ml: %u, ll: %u)", offCode, matchLength, litLength); - if (cctx->appliedParams.validateSequences) { - seqPos->posInSrc += litLength + matchLength; - FORWARD_IF_ERROR(ZSTD_validateSequence(offCode, matchLength, seqPos->posInSrc, - cctx->appliedParams.cParams.windowLog, dictSize), - "Sequence validation failed"); - } - RETURN_ERROR_IF(idx - seqPos->idx > cctx->seqStore.maxNbSeq, memory_allocation, - "Not enough memory allocated. Try adjusting ZSTD_c_minMatch."); - ZSTD_storeSeq(&cctx->seqStore, litLength, ip, iend, offCode, matchLength); - ip += matchLength + litLength; - } - ZSTD_memcpy(cctx->blockState.nextCBlock->rep, updatedRepcodes.rep, sizeof(repcodes_t)); - - if (inSeqs[idx].litLength) { - DEBUGLOG(6, "Storing last literals of size: %u", inSeqs[idx].litLength); - ZSTD_storeLastLiterals(&cctx->seqStore, ip, inSeqs[idx].litLength); - ip += inSeqs[idx].litLength; - seqPos->posInSrc += inSeqs[idx].litLength; - } - RETURN_ERROR_IF(ip != iend, corruption_detected, "Blocksize doesn't agree with block delimiter!"); - seqPos->idx = idx+1; - return 0; -} - -/* Returns the number of bytes to move the current read position back by. Only non-zero - * if we ended up splitting a sequence. Otherwise, it may return a ZSTD error if something - * went wrong. - * - * This function will attempt to scan through blockSize bytes represented by the sequences - * in inSeqs, storing any (partial) sequences. - * - * Occasionally, we may want to change the actual number of bytes we consumed from inSeqs to - * avoid splitting a match, or to avoid splitting a match such that it would produce a match - * smaller than MINMATCH. In this case, we return the number of bytes that we didn't read from this block. - */ -static size_t -ZSTD_copySequencesToSeqStoreNoBlockDelim(ZSTD_CCtx* cctx, ZSTD_sequencePosition* seqPos, - const ZSTD_Sequence* const inSeqs, size_t inSeqsSize, - const void* src, size_t blockSize) -{ - U32 idx = seqPos->idx; - U32 startPosInSequence = seqPos->posInSequence; - U32 endPosInSequence = seqPos->posInSequence + (U32)blockSize; - size_t dictSize; - BYTE const* ip = (BYTE const*)(src); - BYTE const* iend = ip + blockSize; /* May be adjusted if we decide to process fewer than blockSize bytes */ - repcodes_t updatedRepcodes; - U32 bytesAdjustment = 0; - U32 finalMatchSplit = 0; - - if (cctx->cdict) { - dictSize = cctx->cdict->dictContentSize; - } else if (cctx->prefixDict.dict) { - dictSize = cctx->prefixDict.dictSize; - } else { - dictSize = 0; - } - DEBUGLOG(5, "ZSTD_copySequencesToSeqStore: idx: %u PIS: %u blockSize: %zu", idx, startPosInSequence, blockSize); - DEBUGLOG(5, "Start seq: idx: %u (of: %u ml: %u ll: %u)", idx, inSeqs[idx].offset, inSeqs[idx].matchLength, inSeqs[idx].litLength); - ZSTD_memcpy(updatedRepcodes.rep, cctx->blockState.prevCBlock->rep, sizeof(repcodes_t)); - while (endPosInSequence && idx < inSeqsSize && !finalMatchSplit) { - const ZSTD_Sequence currSeq = inSeqs[idx]; - U32 litLength = currSeq.litLength; - U32 matchLength = currSeq.matchLength; - U32 const rawOffset = currSeq.offset; - U32 offCode; - - /* Modify the sequence depending on where endPosInSequence lies */ - if (endPosInSequence >= currSeq.litLength + currSeq.matchLength) { - if (startPosInSequence >= litLength) { - startPosInSequence -= litLength; - litLength = 0; - matchLength -= startPosInSequence; - } else { - litLength -= startPosInSequence; - } - /* Move to the next sequence */ - endPosInSequence -= currSeq.litLength + currSeq.matchLength; - startPosInSequence = 0; - idx++; - } else { - /* This is the final (partial) sequence we're adding from inSeqs, and endPosInSequence - does not reach the end of the match. So, we have to split the sequence */ - DEBUGLOG(6, "Require a split: diff: %u, idx: %u PIS: %u", - currSeq.litLength + currSeq.matchLength - endPosInSequence, idx, endPosInSequence); - if (endPosInSequence > litLength) { - U32 firstHalfMatchLength; - litLength = startPosInSequence >= litLength ? 0 : litLength - startPosInSequence; - firstHalfMatchLength = endPosInSequence - startPosInSequence - litLength; - if (matchLength > blockSize && firstHalfMatchLength >= cctx->appliedParams.cParams.minMatch) { - /* Only ever split the match if it is larger than the block size */ - U32 secondHalfMatchLength = currSeq.matchLength + currSeq.litLength - endPosInSequence; - if (secondHalfMatchLength < cctx->appliedParams.cParams.minMatch) { - /* Move the endPosInSequence backward so that it creates match of minMatch length */ - endPosInSequence -= cctx->appliedParams.cParams.minMatch - secondHalfMatchLength; - bytesAdjustment = cctx->appliedParams.cParams.minMatch - secondHalfMatchLength; - firstHalfMatchLength -= bytesAdjustment; - } - matchLength = firstHalfMatchLength; - /* Flag that we split the last match - after storing the sequence, exit the loop, - but keep the value of endPosInSequence */ - finalMatchSplit = 1; - } else { - /* Move the position in sequence backwards so that we don't split match, and break to store - * the last literals. We use the original currSeq.litLength as a marker for where endPosInSequence - * should go. We prefer to do this whenever it is not necessary to split the match, or if doing so - * would cause the first half of the match to be too small - */ - bytesAdjustment = endPosInSequence - currSeq.litLength; - endPosInSequence = currSeq.litLength; - break; - } - } else { - /* This sequence ends inside the literals, break to store the last literals */ - break; - } - } - /* Check if this offset can be represented with a repcode */ - { U32 const ll0 = (litLength == 0); - offCode = ZSTD_finalizeOffCode(rawOffset, updatedRepcodes.rep, ll0); - ZSTD_updateRep(updatedRepcodes.rep, offCode, ll0); - } - - if (cctx->appliedParams.validateSequences) { - seqPos->posInSrc += litLength + matchLength; - FORWARD_IF_ERROR(ZSTD_validateSequence(offCode, matchLength, seqPos->posInSrc, - cctx->appliedParams.cParams.windowLog, dictSize), - "Sequence validation failed"); - } - DEBUGLOG(6, "Storing sequence: (of: %u, ml: %u, ll: %u)", offCode, matchLength, litLength); - RETURN_ERROR_IF(idx - seqPos->idx > cctx->seqStore.maxNbSeq, memory_allocation, - "Not enough memory allocated. Try adjusting ZSTD_c_minMatch."); - ZSTD_storeSeq(&cctx->seqStore, litLength, ip, iend, offCode, matchLength); - ip += matchLength + litLength; - } - DEBUGLOG(5, "Ending seq: idx: %u (of: %u ml: %u ll: %u)", idx, inSeqs[idx].offset, inSeqs[idx].matchLength, inSeqs[idx].litLength); - assert(idx == inSeqsSize || endPosInSequence <= inSeqs[idx].litLength + inSeqs[idx].matchLength); - seqPos->idx = idx; - seqPos->posInSequence = endPosInSequence; - ZSTD_memcpy(cctx->blockState.nextCBlock->rep, updatedRepcodes.rep, sizeof(repcodes_t)); - - iend -= bytesAdjustment; - if (ip != iend) { - /* Store any last literals */ - U32 lastLLSize = (U32)(iend - ip); - assert(ip <= iend); - DEBUGLOG(6, "Storing last literals of size: %u", lastLLSize); - ZSTD_storeLastLiterals(&cctx->seqStore, ip, lastLLSize); - seqPos->posInSrc += lastLLSize; - } - - return bytesAdjustment; -} - -typedef size_t (*ZSTD_sequenceCopier) (ZSTD_CCtx* cctx, ZSTD_sequencePosition* seqPos, - const ZSTD_Sequence* const inSeqs, size_t inSeqsSize, - const void* src, size_t blockSize); -static ZSTD_sequenceCopier ZSTD_selectSequenceCopier(ZSTD_sequenceFormat_e mode) -{ - ZSTD_sequenceCopier sequenceCopier = NULL; - assert(ZSTD_cParam_withinBounds(ZSTD_c_blockDelimiters, mode)); - if (mode == ZSTD_sf_explicitBlockDelimiters) { - return ZSTD_copySequencesToSeqStoreExplicitBlockDelim; - } else if (mode == ZSTD_sf_noBlockDelimiters) { - return ZSTD_copySequencesToSeqStoreNoBlockDelim; - } - assert(sequenceCopier != NULL); - return sequenceCopier; -} - -/* Compress, block-by-block, all of the sequences given. - * - * Returns the cumulative size of all compressed blocks (including their headers), - * otherwise a ZSTD error. - */ -static size_t -ZSTD_compressSequences_internal(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const ZSTD_Sequence* inSeqs, size_t inSeqsSize, - const void* src, size_t srcSize) -{ - size_t cSize = 0; - U32 lastBlock; - size_t blockSize; - size_t compressedSeqsSize; - size_t remaining = srcSize; - ZSTD_sequencePosition seqPos = {0, 0, 0}; - - BYTE const* ip = (BYTE const*)src; - BYTE* op = (BYTE*)dst; - ZSTD_sequenceCopier const sequenceCopier = ZSTD_selectSequenceCopier(cctx->appliedParams.blockDelimiters); - - DEBUGLOG(4, "ZSTD_compressSequences_internal srcSize: %zu, inSeqsSize: %zu", srcSize, inSeqsSize); - /* Special case: empty frame */ - if (remaining == 0) { - U32 const cBlockHeader24 = 1 /* last block */ + (((U32)bt_raw)<<1); - RETURN_ERROR_IF(dstCapacity<4, dstSize_tooSmall, "No room for empty frame block header"); - MEM_writeLE32(op, cBlockHeader24); - op += ZSTD_blockHeaderSize; - dstCapacity -= ZSTD_blockHeaderSize; - cSize += ZSTD_blockHeaderSize; - } - - while (remaining) { - size_t cBlockSize; - size_t additionalByteAdjustment; - lastBlock = remaining <= cctx->blockSize; - blockSize = lastBlock ? (U32)remaining : (U32)cctx->blockSize; - ZSTD_resetSeqStore(&cctx->seqStore); - DEBUGLOG(4, "Working on new block. Blocksize: %zu", blockSize); - - additionalByteAdjustment = sequenceCopier(cctx, &seqPos, inSeqs, inSeqsSize, ip, blockSize); - FORWARD_IF_ERROR(additionalByteAdjustment, "Bad sequence copy"); - blockSize -= additionalByteAdjustment; - - /* If blocks are too small, emit as a nocompress block */ - if (blockSize < MIN_CBLOCK_SIZE+ZSTD_blockHeaderSize+1) { - cBlockSize = ZSTD_noCompressBlock(op, dstCapacity, ip, blockSize, lastBlock); - FORWARD_IF_ERROR(cBlockSize, "Nocompress block failed"); - DEBUGLOG(4, "Block too small, writing out nocompress block: cSize: %zu", cBlockSize); - cSize += cBlockSize; - ip += blockSize; - op += cBlockSize; - remaining -= blockSize; - dstCapacity -= cBlockSize; - continue; - } - - compressedSeqsSize = ZSTD_entropyCompressSeqStore(&cctx->seqStore, - &cctx->blockState.prevCBlock->entropy, &cctx->blockState.nextCBlock->entropy, - &cctx->appliedParams, - op + ZSTD_blockHeaderSize /* Leave space for block header */, dstCapacity - ZSTD_blockHeaderSize, - blockSize, - cctx->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */, - cctx->bmi2); - FORWARD_IF_ERROR(compressedSeqsSize, "Compressing sequences of block failed"); - DEBUGLOG(4, "Compressed sequences size: %zu", compressedSeqsSize); - - if (!cctx->isFirstBlock && - ZSTD_maybeRLE(&cctx->seqStore) && - ZSTD_isRLE((BYTE const*)src, srcSize)) { - /* We don't want to emit our first block as a RLE even if it qualifies because - * doing so will cause the decoder (cli only) to throw a "should consume all input error." - * This is only an issue for zstd <= v1.4.3 - */ - compressedSeqsSize = 1; - } - - if (compressedSeqsSize == 0) { - /* ZSTD_noCompressBlock writes the block header as well */ - cBlockSize = ZSTD_noCompressBlock(op, dstCapacity, ip, blockSize, lastBlock); - FORWARD_IF_ERROR(cBlockSize, "Nocompress block failed"); - DEBUGLOG(4, "Writing out nocompress block, size: %zu", cBlockSize); - } else if (compressedSeqsSize == 1) { - cBlockSize = ZSTD_rleCompressBlock(op, dstCapacity, *ip, blockSize, lastBlock); - FORWARD_IF_ERROR(cBlockSize, "RLE compress block failed"); - DEBUGLOG(4, "Writing out RLE block, size: %zu", cBlockSize); - } else { - U32 cBlockHeader; - /* Error checking and repcodes update */ - ZSTD_blockState_confirmRepcodesAndEntropyTables(&cctx->blockState); - if (cctx->blockState.prevCBlock->entropy.fse.offcode_repeatMode == FSE_repeat_valid) - cctx->blockState.prevCBlock->entropy.fse.offcode_repeatMode = FSE_repeat_check; - - /* Write block header into beginning of block*/ - cBlockHeader = lastBlock + (((U32)bt_compressed)<<1) + (U32)(compressedSeqsSize << 3); - MEM_writeLE24(op, cBlockHeader); - cBlockSize = ZSTD_blockHeaderSize + compressedSeqsSize; - DEBUGLOG(4, "Writing out compressed block, size: %zu", cBlockSize); - } - - cSize += cBlockSize; - DEBUGLOG(4, "cSize running total: %zu", cSize); - - if (lastBlock) { - break; - } else { - ip += blockSize; - op += cBlockSize; - remaining -= blockSize; - dstCapacity -= cBlockSize; - cctx->isFirstBlock = 0; - } - } - - return cSize; -} - -size_t ZSTD_compressSequences(ZSTD_CCtx* const cctx, void* dst, size_t dstCapacity, - const ZSTD_Sequence* inSeqs, size_t inSeqsSize, - const void* src, size_t srcSize) -{ - BYTE* op = (BYTE*)dst; - size_t cSize = 0; - size_t compressedBlocksSize = 0; - size_t frameHeaderSize = 0; - - /* Transparent initialization stage, same as compressStream2() */ - DEBUGLOG(3, "ZSTD_compressSequences()"); - assert(cctx != NULL); - FORWARD_IF_ERROR(ZSTD_CCtx_init_compressStream2(cctx, ZSTD_e_end, srcSize), "CCtx initialization failed"); - /* Begin writing output, starting with frame header */ - frameHeaderSize = ZSTD_writeFrameHeader(op, dstCapacity, &cctx->appliedParams, srcSize, cctx->dictID); - op += frameHeaderSize; - dstCapacity -= frameHeaderSize; - cSize += frameHeaderSize; - if (cctx->appliedParams.fParams.checksumFlag && srcSize) { - XXH64_update(&cctx->xxhState, src, srcSize); - } - /* cSize includes block header size and compressed sequences size */ - compressedBlocksSize = ZSTD_compressSequences_internal(cctx, - op, dstCapacity, - inSeqs, inSeqsSize, - src, srcSize); - FORWARD_IF_ERROR(compressedBlocksSize, "Compressing blocks failed!"); - cSize += compressedBlocksSize; - dstCapacity -= compressedBlocksSize; - - if (cctx->appliedParams.fParams.checksumFlag) { - U32 const checksum = (U32) XXH64_digest(&cctx->xxhState); - RETURN_ERROR_IF(dstCapacity<4, dstSize_tooSmall, "no room for checksum"); - DEBUGLOG(4, "Write checksum : %08X", (unsigned)checksum); - MEM_writeLE32((char*)dst + cSize, checksum); - cSize += 4; - } - - DEBUGLOG(3, "Final compressed size: %zu", cSize); - return cSize; -} - -/*====== Finalize ======*/ - -/*! ZSTD_flushStream() : - * @return : amount of data remaining to flush */ -size_t ZSTD_flushStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output) -{ - ZSTD_inBuffer input = { NULL, 0, 0 }; - return ZSTD_compressStream2(zcs, output, &input, ZSTD_e_flush); -} - - -size_t ZSTD_endStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output) -{ - ZSTD_inBuffer input = { NULL, 0, 0 }; - size_t const remainingToFlush = ZSTD_compressStream2(zcs, output, &input, ZSTD_e_end); - FORWARD_IF_ERROR( remainingToFlush , "ZSTD_compressStream2 failed"); - if (zcs->appliedParams.nbWorkers > 0) return remainingToFlush; /* minimal estimation */ - /* single thread mode : attempt to calculate remaining to flush more precisely */ - { size_t const lastBlockSize = zcs->frameEnded ? 0 : ZSTD_BLOCKHEADERSIZE; - size_t const checksumSize = (size_t)(zcs->frameEnded ? 0 : zcs->appliedParams.fParams.checksumFlag * 4); - size_t const toFlush = remainingToFlush + lastBlockSize + checksumSize; - DEBUGLOG(4, "ZSTD_endStream : remaining to flush : %u", (unsigned)toFlush); - return toFlush; - } -} - - -/*-===== Pre-defined compression levels =====-*/ -#include "clevels.h" - -int ZSTD_maxCLevel(void) { return ZSTD_MAX_CLEVEL; } -int ZSTD_minCLevel(void) { return (int)-ZSTD_TARGETLENGTH_MAX; } -int ZSTD_defaultCLevel(void) { return ZSTD_CLEVEL_DEFAULT; } - -static ZSTD_compressionParameters ZSTD_dedicatedDictSearch_getCParams(int const compressionLevel, size_t const dictSize) -{ - ZSTD_compressionParameters cParams = ZSTD_getCParams_internal(compressionLevel, 0, dictSize, ZSTD_cpm_createCDict); - switch (cParams.strategy) { - case ZSTD_fast: - case ZSTD_dfast: - break; - case ZSTD_greedy: - case ZSTD_lazy: - case ZSTD_lazy2: - cParams.hashLog += ZSTD_LAZY_DDSS_BUCKET_LOG; - break; - case ZSTD_btlazy2: - case ZSTD_btopt: - case ZSTD_btultra: - case ZSTD_btultra2: - break; - } - return cParams; -} - -static int ZSTD_dedicatedDictSearch_isSupported( - ZSTD_compressionParameters const* cParams) -{ - return (cParams->strategy >= ZSTD_greedy) - && (cParams->strategy <= ZSTD_lazy2) - && (cParams->hashLog > cParams->chainLog) - && (cParams->chainLog <= 24); -} - -/** - * Reverses the adjustment applied to cparams when enabling dedicated dict - * search. This is used to recover the params set to be used in the working - * context. (Otherwise, those tables would also grow.) - */ -static void ZSTD_dedicatedDictSearch_revertCParams( - ZSTD_compressionParameters* cParams) { - switch (cParams->strategy) { - case ZSTD_fast: - case ZSTD_dfast: - break; - case ZSTD_greedy: - case ZSTD_lazy: - case ZSTD_lazy2: - cParams->hashLog -= ZSTD_LAZY_DDSS_BUCKET_LOG; - if (cParams->hashLog < ZSTD_HASHLOG_MIN) { - cParams->hashLog = ZSTD_HASHLOG_MIN; - } - break; - case ZSTD_btlazy2: - case ZSTD_btopt: - case ZSTD_btultra: - case ZSTD_btultra2: - break; - } -} - -static U64 ZSTD_getCParamRowSize(U64 srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode) -{ - switch (mode) { - case ZSTD_cpm_unknown: - case ZSTD_cpm_noAttachDict: - case ZSTD_cpm_createCDict: - break; - case ZSTD_cpm_attachDict: - dictSize = 0; - break; - default: - assert(0); - break; - } - { int const unknown = srcSizeHint == ZSTD_CONTENTSIZE_UNKNOWN; - size_t const addedSize = unknown && dictSize > 0 ? 500 : 0; - return unknown && dictSize == 0 ? ZSTD_CONTENTSIZE_UNKNOWN : srcSizeHint+dictSize+addedSize; - } -} - -/*! ZSTD_getCParams_internal() : - * @return ZSTD_compressionParameters structure for a selected compression level, srcSize and dictSize. - * Note: srcSizeHint 0 means 0, use ZSTD_CONTENTSIZE_UNKNOWN for unknown. - * Use dictSize == 0 for unknown or unused. - * Note: `mode` controls how we treat the `dictSize`. See docs for `ZSTD_cParamMode_e`. */ -static ZSTD_compressionParameters ZSTD_getCParams_internal(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode) -{ - U64 const rSize = ZSTD_getCParamRowSize(srcSizeHint, dictSize, mode); - U32 const tableID = (rSize <= 256 KB) + (rSize <= 128 KB) + (rSize <= 16 KB); - int row; - DEBUGLOG(5, "ZSTD_getCParams_internal (cLevel=%i)", compressionLevel); - - /* row */ - if (compressionLevel == 0) row = ZSTD_CLEVEL_DEFAULT; /* 0 == default */ - else if (compressionLevel < 0) row = 0; /* entry 0 is baseline for fast mode */ - else if (compressionLevel > ZSTD_MAX_CLEVEL) row = ZSTD_MAX_CLEVEL; - else row = compressionLevel; - - { ZSTD_compressionParameters cp = ZSTD_defaultCParameters[tableID][row]; - DEBUGLOG(5, "ZSTD_getCParams_internal selected tableID: %u row: %u strat: %u", tableID, row, (U32)cp.strategy); - /* acceleration factor */ - if (compressionLevel < 0) { - int const clampedCompressionLevel = MAX(ZSTD_minCLevel(), compressionLevel); - cp.targetLength = (unsigned)(-clampedCompressionLevel); - } - /* refine parameters based on srcSize & dictSize */ - return ZSTD_adjustCParams_internal(cp, srcSizeHint, dictSize, mode); - } -} - -/*! ZSTD_getCParams() : - * @return ZSTD_compressionParameters structure for a selected compression level, srcSize and dictSize. - * Size values are optional, provide 0 if not known or unused */ -ZSTD_compressionParameters ZSTD_getCParams(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize) -{ - if (srcSizeHint == 0) srcSizeHint = ZSTD_CONTENTSIZE_UNKNOWN; - return ZSTD_getCParams_internal(compressionLevel, srcSizeHint, dictSize, ZSTD_cpm_unknown); -} - -/*! ZSTD_getParams() : - * same idea as ZSTD_getCParams() - * @return a `ZSTD_parameters` structure (instead of `ZSTD_compressionParameters`). - * Fields of `ZSTD_frameParameters` are set to default values */ -static ZSTD_parameters ZSTD_getParams_internal(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode) { - ZSTD_parameters params; - ZSTD_compressionParameters const cParams = ZSTD_getCParams_internal(compressionLevel, srcSizeHint, dictSize, mode); - DEBUGLOG(5, "ZSTD_getParams (cLevel=%i)", compressionLevel); - ZSTD_memset(¶ms, 0, sizeof(params)); - params.cParams = cParams; - params.fParams.contentSizeFlag = 1; - return params; -} - -/*! ZSTD_getParams() : - * same idea as ZSTD_getCParams() - * @return a `ZSTD_parameters` structure (instead of `ZSTD_compressionParameters`). - * Fields of `ZSTD_frameParameters` are set to default values */ -ZSTD_parameters ZSTD_getParams(int compressionLevel, unsigned long long srcSizeHint, size_t dictSize) { - if (srcSizeHint == 0) srcSizeHint = ZSTD_CONTENTSIZE_UNKNOWN; - return ZSTD_getParams_internal(compressionLevel, srcSizeHint, dictSize, ZSTD_cpm_unknown); -} diff --git a/dep/zstd/lib/compress/zstd_compress_internal.h b/dep/zstd/lib/compress/zstd_compress_internal.h deleted file mode 100644 index c406e794b..000000000 --- a/dep/zstd/lib/compress/zstd_compress_internal.h +++ /dev/null @@ -1,1458 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -/* This header contains definitions - * that shall **only** be used by modules within lib/compress. - */ - -#ifndef ZSTD_COMPRESS_H -#define ZSTD_COMPRESS_H - -/*-************************************* -* Dependencies -***************************************/ -#include "../common/zstd_internal.h" -#include "zstd_cwksp.h" -#ifdef ZSTD_MULTITHREAD -# include "zstdmt_compress.h" -#endif - -#if defined (__cplusplus) -extern "C" { -#endif - -/*-************************************* -* Constants -***************************************/ -#define kSearchStrength 8 -#define HASH_READ_SIZE 8 -#define ZSTD_DUBT_UNSORTED_MARK 1 /* For btlazy2 strategy, index ZSTD_DUBT_UNSORTED_MARK==1 means "unsorted". - It could be confused for a real successor at index "1", if sorted as larger than its predecessor. - It's not a big deal though : candidate will just be sorted again. - Additionally, candidate position 1 will be lost. - But candidate 1 cannot hide a large tree of candidates, so it's a minimal loss. - The benefit is that ZSTD_DUBT_UNSORTED_MARK cannot be mishandled after table re-use with a different strategy. - This constant is required by ZSTD_compressBlock_btlazy2() and ZSTD_reduceTable_internal() */ - - -/*-************************************* -* Context memory management -***************************************/ -typedef enum { ZSTDcs_created=0, ZSTDcs_init, ZSTDcs_ongoing, ZSTDcs_ending } ZSTD_compressionStage_e; -typedef enum { zcss_init=0, zcss_load, zcss_flush } ZSTD_cStreamStage; - -typedef struct ZSTD_prefixDict_s { - const void* dict; - size_t dictSize; - ZSTD_dictContentType_e dictContentType; -} ZSTD_prefixDict; - -typedef struct { - void* dictBuffer; - void const* dict; - size_t dictSize; - ZSTD_dictContentType_e dictContentType; - ZSTD_CDict* cdict; -} ZSTD_localDict; - -typedef struct { - HUF_CElt CTable[HUF_CTABLE_SIZE_ST(255)]; - HUF_repeat repeatMode; -} ZSTD_hufCTables_t; - -typedef struct { - FSE_CTable offcodeCTable[FSE_CTABLE_SIZE_U32(OffFSELog, MaxOff)]; - FSE_CTable matchlengthCTable[FSE_CTABLE_SIZE_U32(MLFSELog, MaxML)]; - FSE_CTable litlengthCTable[FSE_CTABLE_SIZE_U32(LLFSELog, MaxLL)]; - FSE_repeat offcode_repeatMode; - FSE_repeat matchlength_repeatMode; - FSE_repeat litlength_repeatMode; -} ZSTD_fseCTables_t; - -typedef struct { - ZSTD_hufCTables_t huf; - ZSTD_fseCTables_t fse; -} ZSTD_entropyCTables_t; - -/*********************************************** -* Entropy buffer statistics structs and funcs * -***********************************************/ -/** ZSTD_hufCTablesMetadata_t : - * Stores Literals Block Type for a super-block in hType, and - * huffman tree description in hufDesBuffer. - * hufDesSize refers to the size of huffman tree description in bytes. - * This metadata is populated in ZSTD_buildBlockEntropyStats_literals() */ -typedef struct { - symbolEncodingType_e hType; - BYTE hufDesBuffer[ZSTD_MAX_HUF_HEADER_SIZE]; - size_t hufDesSize; -} ZSTD_hufCTablesMetadata_t; - -/** ZSTD_fseCTablesMetadata_t : - * Stores symbol compression modes for a super-block in {ll, ol, ml}Type, and - * fse tables in fseTablesBuffer. - * fseTablesSize refers to the size of fse tables in bytes. - * This metadata is populated in ZSTD_buildBlockEntropyStats_sequences() */ -typedef struct { - symbolEncodingType_e llType; - symbolEncodingType_e ofType; - symbolEncodingType_e mlType; - BYTE fseTablesBuffer[ZSTD_MAX_FSE_HEADERS_SIZE]; - size_t fseTablesSize; - size_t lastCountSize; /* This is to account for bug in 1.3.4. More detail in ZSTD_entropyCompressSeqStore_internal() */ -} ZSTD_fseCTablesMetadata_t; - -typedef struct { - ZSTD_hufCTablesMetadata_t hufMetadata; - ZSTD_fseCTablesMetadata_t fseMetadata; -} ZSTD_entropyCTablesMetadata_t; - -/** ZSTD_buildBlockEntropyStats() : - * Builds entropy for the block. - * @return : 0 on success or error code */ -size_t ZSTD_buildBlockEntropyStats(seqStore_t* seqStorePtr, - const ZSTD_entropyCTables_t* prevEntropy, - ZSTD_entropyCTables_t* nextEntropy, - const ZSTD_CCtx_params* cctxParams, - ZSTD_entropyCTablesMetadata_t* entropyMetadata, - void* workspace, size_t wkspSize); - -/********************************* -* Compression internals structs * -*********************************/ - -typedef struct { - U32 off; /* Offset sumtype code for the match, using ZSTD_storeSeq() format */ - U32 len; /* Raw length of match */ -} ZSTD_match_t; - -typedef struct { - U32 offset; /* Offset of sequence */ - U32 litLength; /* Length of literals prior to match */ - U32 matchLength; /* Raw length of match */ -} rawSeq; - -typedef struct { - rawSeq* seq; /* The start of the sequences */ - size_t pos; /* The index in seq where reading stopped. pos <= size. */ - size_t posInSequence; /* The position within the sequence at seq[pos] where reading - stopped. posInSequence <= seq[pos].litLength + seq[pos].matchLength */ - size_t size; /* The number of sequences. <= capacity. */ - size_t capacity; /* The capacity starting from `seq` pointer */ -} rawSeqStore_t; - -UNUSED_ATTR static const rawSeqStore_t kNullRawSeqStore = {NULL, 0, 0, 0, 0}; - -typedef struct { - int price; - U32 off; - U32 mlen; - U32 litlen; - U32 rep[ZSTD_REP_NUM]; -} ZSTD_optimal_t; - -typedef enum { zop_dynamic=0, zop_predef } ZSTD_OptPrice_e; - -typedef struct { - /* All tables are allocated inside cctx->workspace by ZSTD_resetCCtx_internal() */ - unsigned* litFreq; /* table of literals statistics, of size 256 */ - unsigned* litLengthFreq; /* table of litLength statistics, of size (MaxLL+1) */ - unsigned* matchLengthFreq; /* table of matchLength statistics, of size (MaxML+1) */ - unsigned* offCodeFreq; /* table of offCode statistics, of size (MaxOff+1) */ - ZSTD_match_t* matchTable; /* list of found matches, of size ZSTD_OPT_NUM+1 */ - ZSTD_optimal_t* priceTable; /* All positions tracked by optimal parser, of size ZSTD_OPT_NUM+1 */ - - U32 litSum; /* nb of literals */ - U32 litLengthSum; /* nb of litLength codes */ - U32 matchLengthSum; /* nb of matchLength codes */ - U32 offCodeSum; /* nb of offset codes */ - U32 litSumBasePrice; /* to compare to log2(litfreq) */ - U32 litLengthSumBasePrice; /* to compare to log2(llfreq) */ - U32 matchLengthSumBasePrice;/* to compare to log2(mlfreq) */ - U32 offCodeSumBasePrice; /* to compare to log2(offreq) */ - ZSTD_OptPrice_e priceType; /* prices can be determined dynamically, or follow a pre-defined cost structure */ - const ZSTD_entropyCTables_t* symbolCosts; /* pre-calculated dictionary statistics */ - ZSTD_paramSwitch_e literalCompressionMode; -} optState_t; - -typedef struct { - ZSTD_entropyCTables_t entropy; - U32 rep[ZSTD_REP_NUM]; -} ZSTD_compressedBlockState_t; - -typedef struct { - BYTE const* nextSrc; /* next block here to continue on current prefix */ - BYTE const* base; /* All regular indexes relative to this position */ - BYTE const* dictBase; /* extDict indexes relative to this position */ - U32 dictLimit; /* below that point, need extDict */ - U32 lowLimit; /* below that point, no more valid data */ - U32 nbOverflowCorrections; /* Number of times overflow correction has run since - * ZSTD_window_init(). Useful for debugging coredumps - * and for ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY. - */ -} ZSTD_window_t; - -#define ZSTD_WINDOW_START_INDEX 2 - -typedef struct ZSTD_matchState_t ZSTD_matchState_t; - -#define ZSTD_ROW_HASH_CACHE_SIZE 8 /* Size of prefetching hash cache for row-based matchfinder */ - -struct ZSTD_matchState_t { - ZSTD_window_t window; /* State for window round buffer management */ - U32 loadedDictEnd; /* index of end of dictionary, within context's referential. - * When loadedDictEnd != 0, a dictionary is in use, and still valid. - * This relies on a mechanism to set loadedDictEnd=0 when dictionary is no longer within distance. - * Such mechanism is provided within ZSTD_window_enforceMaxDist() and ZSTD_checkDictValidity(). - * When dict referential is copied into active context (i.e. not attached), - * loadedDictEnd == dictSize, since referential starts from zero. - */ - U32 nextToUpdate; /* index from which to continue table update */ - U32 hashLog3; /* dispatch table for matches of len==3 : larger == faster, more memory */ - - U32 rowHashLog; /* For row-based matchfinder: Hashlog based on nb of rows in the hashTable.*/ - U16* tagTable; /* For row-based matchFinder: A row-based table containing the hashes and head index. */ - U32 hashCache[ZSTD_ROW_HASH_CACHE_SIZE]; /* For row-based matchFinder: a cache of hashes to improve speed */ - - U32* hashTable; - U32* hashTable3; - U32* chainTable; - - U32 forceNonContiguous; /* Non-zero if we should force non-contiguous load for the next window update. */ - - int dedicatedDictSearch; /* Indicates whether this matchState is using the - * dedicated dictionary search structure. - */ - optState_t opt; /* optimal parser state */ - const ZSTD_matchState_t* dictMatchState; - ZSTD_compressionParameters cParams; - const rawSeqStore_t* ldmSeqStore; -}; - -typedef struct { - ZSTD_compressedBlockState_t* prevCBlock; - ZSTD_compressedBlockState_t* nextCBlock; - ZSTD_matchState_t matchState; -} ZSTD_blockState_t; - -typedef struct { - U32 offset; - U32 checksum; -} ldmEntry_t; - -typedef struct { - BYTE const* split; - U32 hash; - U32 checksum; - ldmEntry_t* bucket; -} ldmMatchCandidate_t; - -#define LDM_BATCH_SIZE 64 - -typedef struct { - ZSTD_window_t window; /* State for the window round buffer management */ - ldmEntry_t* hashTable; - U32 loadedDictEnd; - BYTE* bucketOffsets; /* Next position in bucket to insert entry */ - size_t splitIndices[LDM_BATCH_SIZE]; - ldmMatchCandidate_t matchCandidates[LDM_BATCH_SIZE]; -} ldmState_t; - -typedef struct { - ZSTD_paramSwitch_e enableLdm; /* ZSTD_ps_enable to enable LDM. ZSTD_ps_auto by default */ - U32 hashLog; /* Log size of hashTable */ - U32 bucketSizeLog; /* Log bucket size for collision resolution, at most 8 */ - U32 minMatchLength; /* Minimum match length */ - U32 hashRateLog; /* Log number of entries to skip */ - U32 windowLog; /* Window log for the LDM */ -} ldmParams_t; - -typedef struct { - int collectSequences; - ZSTD_Sequence* seqStart; - size_t seqIndex; - size_t maxSequences; -} SeqCollector; - -struct ZSTD_CCtx_params_s { - ZSTD_format_e format; - ZSTD_compressionParameters cParams; - ZSTD_frameParameters fParams; - - int compressionLevel; - int forceWindow; /* force back-references to respect limit of - * 1< 63) ? ZSTD_highbit32(litLength) + LL_deltaCode : LL_Code[litLength]; -} - -/* ZSTD_MLcode() : - * note : mlBase = matchLength - MINMATCH; - * because it's the format it's stored in seqStore->sequences */ -MEM_STATIC U32 ZSTD_MLcode(U32 mlBase) -{ - static const BYTE ML_Code[128] = { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, - 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, - 32, 32, 33, 33, 34, 34, 35, 35, 36, 36, 36, 36, 37, 37, 37, 37, - 38, 38, 38, 38, 38, 38, 38, 38, 39, 39, 39, 39, 39, 39, 39, 39, - 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, 40, - 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, 41, - 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, - 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42, 42 }; - static const U32 ML_deltaCode = 36; - return (mlBase > 127) ? ZSTD_highbit32(mlBase) + ML_deltaCode : ML_Code[mlBase]; -} - -/* ZSTD_cParam_withinBounds: - * @return 1 if value is within cParam bounds, - * 0 otherwise */ -MEM_STATIC int ZSTD_cParam_withinBounds(ZSTD_cParameter cParam, int value) -{ - ZSTD_bounds const bounds = ZSTD_cParam_getBounds(cParam); - if (ZSTD_isError(bounds.error)) return 0; - if (value < bounds.lowerBound) return 0; - if (value > bounds.upperBound) return 0; - return 1; -} - -/* ZSTD_noCompressBlock() : - * Writes uncompressed block to dst buffer from given src. - * Returns the size of the block */ -MEM_STATIC size_t ZSTD_noCompressBlock (void* dst, size_t dstCapacity, const void* src, size_t srcSize, U32 lastBlock) -{ - U32 const cBlockHeader24 = lastBlock + (((U32)bt_raw)<<1) + (U32)(srcSize << 3); - RETURN_ERROR_IF(srcSize + ZSTD_blockHeaderSize > dstCapacity, - dstSize_tooSmall, "dst buf too small for uncompressed block"); - MEM_writeLE24(dst, cBlockHeader24); - ZSTD_memcpy((BYTE*)dst + ZSTD_blockHeaderSize, src, srcSize); - return ZSTD_blockHeaderSize + srcSize; -} - -MEM_STATIC size_t ZSTD_rleCompressBlock (void* dst, size_t dstCapacity, BYTE src, size_t srcSize, U32 lastBlock) -{ - BYTE* const op = (BYTE*)dst; - U32 const cBlockHeader = lastBlock + (((U32)bt_rle)<<1) + (U32)(srcSize << 3); - RETURN_ERROR_IF(dstCapacity < 4, dstSize_tooSmall, ""); - MEM_writeLE24(op, cBlockHeader); - op[3] = src; - return 4; -} - - -/* ZSTD_minGain() : - * minimum compression required - * to generate a compress block or a compressed literals section. - * note : use same formula for both situations */ -MEM_STATIC size_t ZSTD_minGain(size_t srcSize, ZSTD_strategy strat) -{ - U32 const minlog = (strat>=ZSTD_btultra) ? (U32)(strat) - 1 : 6; - ZSTD_STATIC_ASSERT(ZSTD_btultra == 8); - assert(ZSTD_cParam_withinBounds(ZSTD_c_strategy, strat)); - return (srcSize >> minlog) + 2; -} - -MEM_STATIC int ZSTD_literalsCompressionIsDisabled(const ZSTD_CCtx_params* cctxParams) -{ - switch (cctxParams->literalCompressionMode) { - case ZSTD_ps_enable: - return 0; - case ZSTD_ps_disable: - return 1; - default: - assert(0 /* impossible: pre-validated */); - ZSTD_FALLTHROUGH; - case ZSTD_ps_auto: - return (cctxParams->cParams.strategy == ZSTD_fast) && (cctxParams->cParams.targetLength > 0); - } -} - -/*! ZSTD_safecopyLiterals() : - * memcpy() function that won't read beyond more than WILDCOPY_OVERLENGTH bytes past ilimit_w. - * Only called when the sequence ends past ilimit_w, so it only needs to be optimized for single - * large copies. - */ -static void -ZSTD_safecopyLiterals(BYTE* op, BYTE const* ip, BYTE const* const iend, BYTE const* ilimit_w) -{ - assert(iend > ilimit_w); - if (ip <= ilimit_w) { - ZSTD_wildcopy(op, ip, ilimit_w - ip, ZSTD_no_overlap); - op += ilimit_w - ip; - ip = ilimit_w; - } - while (ip < iend) *op++ = *ip++; -} - -#define ZSTD_REP_MOVE (ZSTD_REP_NUM-1) -#define STORE_REPCODE_1 STORE_REPCODE(1) -#define STORE_REPCODE_2 STORE_REPCODE(2) -#define STORE_REPCODE_3 STORE_REPCODE(3) -#define STORE_REPCODE(r) (assert((r)>=1), assert((r)<=3), (r)-1) -#define STORE_OFFSET(o) (assert((o)>0), o + ZSTD_REP_MOVE) -#define STORED_IS_OFFSET(o) ((o) > ZSTD_REP_MOVE) -#define STORED_IS_REPCODE(o) ((o) <= ZSTD_REP_MOVE) -#define STORED_OFFSET(o) (assert(STORED_IS_OFFSET(o)), (o)-ZSTD_REP_MOVE) -#define STORED_REPCODE(o) (assert(STORED_IS_REPCODE(o)), (o)+1) /* returns ID 1,2,3 */ -#define STORED_TO_OFFBASE(o) ((o)+1) -#define OFFBASE_TO_STORED(o) ((o)-1) - -/*! ZSTD_storeSeq() : - * Store a sequence (litlen, litPtr, offCode and matchLength) into seqStore_t. - * @offBase_minus1 : Users should use employ macros STORE_REPCODE_X and STORE_OFFSET(). - * @matchLength : must be >= MINMATCH - * Allowed to overread literals up to litLimit. -*/ -HINT_INLINE UNUSED_ATTR void -ZSTD_storeSeq(seqStore_t* seqStorePtr, - size_t litLength, const BYTE* literals, const BYTE* litLimit, - U32 offBase_minus1, - size_t matchLength) -{ - BYTE const* const litLimit_w = litLimit - WILDCOPY_OVERLENGTH; - BYTE const* const litEnd = literals + litLength; -#if defined(DEBUGLEVEL) && (DEBUGLEVEL >= 6) - static const BYTE* g_start = NULL; - if (g_start==NULL) g_start = (const BYTE*)literals; /* note : index only works for compression within a single segment */ - { U32 const pos = (U32)((const BYTE*)literals - g_start); - DEBUGLOG(6, "Cpos%7u :%3u literals, match%4u bytes at offCode%7u", - pos, (U32)litLength, (U32)matchLength, (U32)offBase_minus1); - } -#endif - assert((size_t)(seqStorePtr->sequences - seqStorePtr->sequencesStart) < seqStorePtr->maxNbSeq); - /* copy Literals */ - assert(seqStorePtr->maxNbLit <= 128 KB); - assert(seqStorePtr->lit + litLength <= seqStorePtr->litStart + seqStorePtr->maxNbLit); - assert(literals + litLength <= litLimit); - if (litEnd <= litLimit_w) { - /* Common case we can use wildcopy. - * First copy 16 bytes, because literals are likely short. - */ - assert(WILDCOPY_OVERLENGTH >= 16); - ZSTD_copy16(seqStorePtr->lit, literals); - if (litLength > 16) { - ZSTD_wildcopy(seqStorePtr->lit+16, literals+16, (ptrdiff_t)litLength-16, ZSTD_no_overlap); - } - } else { - ZSTD_safecopyLiterals(seqStorePtr->lit, literals, litEnd, litLimit_w); - } - seqStorePtr->lit += litLength; - - /* literal Length */ - if (litLength>0xFFFF) { - assert(seqStorePtr->longLengthType == ZSTD_llt_none); /* there can only be a single long length */ - seqStorePtr->longLengthType = ZSTD_llt_literalLength; - seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart); - } - seqStorePtr->sequences[0].litLength = (U16)litLength; - - /* match offset */ - seqStorePtr->sequences[0].offBase = STORED_TO_OFFBASE(offBase_minus1); - - /* match Length */ - assert(matchLength >= MINMATCH); - { size_t const mlBase = matchLength - MINMATCH; - if (mlBase>0xFFFF) { - assert(seqStorePtr->longLengthType == ZSTD_llt_none); /* there can only be a single long length */ - seqStorePtr->longLengthType = ZSTD_llt_matchLength; - seqStorePtr->longLengthPos = (U32)(seqStorePtr->sequences - seqStorePtr->sequencesStart); - } - seqStorePtr->sequences[0].mlBase = (U16)mlBase; - } - - seqStorePtr->sequences++; -} - -/* ZSTD_updateRep() : - * updates in-place @rep (array of repeat offsets) - * @offBase_minus1 : sum-type, with same numeric representation as ZSTD_storeSeq() - */ -MEM_STATIC void -ZSTD_updateRep(U32 rep[ZSTD_REP_NUM], U32 const offBase_minus1, U32 const ll0) -{ - if (STORED_IS_OFFSET(offBase_minus1)) { /* full offset */ - rep[2] = rep[1]; - rep[1] = rep[0]; - rep[0] = STORED_OFFSET(offBase_minus1); - } else { /* repcode */ - U32 const repCode = STORED_REPCODE(offBase_minus1) - 1 + ll0; - if (repCode > 0) { /* note : if repCode==0, no change */ - U32 const currentOffset = (repCode==ZSTD_REP_NUM) ? (rep[0] - 1) : rep[repCode]; - rep[2] = (repCode >= 2) ? rep[1] : rep[2]; - rep[1] = rep[0]; - rep[0] = currentOffset; - } else { /* repCode == 0 */ - /* nothing to do */ - } - } -} - -typedef struct repcodes_s { - U32 rep[3]; -} repcodes_t; - -MEM_STATIC repcodes_t -ZSTD_newRep(U32 const rep[ZSTD_REP_NUM], U32 const offBase_minus1, U32 const ll0) -{ - repcodes_t newReps; - ZSTD_memcpy(&newReps, rep, sizeof(newReps)); - ZSTD_updateRep(newReps.rep, offBase_minus1, ll0); - return newReps; -} - - -/*-************************************* -* Match length counter -***************************************/ -static unsigned ZSTD_NbCommonBytes (size_t val) -{ - if (MEM_isLittleEndian()) { - if (MEM_64bits()) { -# if defined(_MSC_VER) && defined(_WIN64) -# if STATIC_BMI2 - return _tzcnt_u64(val) >> 3; -# else - if (val != 0) { - unsigned long r; - _BitScanForward64(&r, (U64)val); - return (unsigned)(r >> 3); - } else { - /* Should not reach this code path */ - __assume(0); - } -# endif -# elif defined(__GNUC__) && (__GNUC__ >= 4) - return (__builtin_ctzll((U64)val) >> 3); -# else - static const int DeBruijnBytePos[64] = { 0, 0, 0, 0, 0, 1, 1, 2, - 0, 3, 1, 3, 1, 4, 2, 7, - 0, 2, 3, 6, 1, 5, 3, 5, - 1, 3, 4, 4, 2, 5, 6, 7, - 7, 0, 1, 2, 3, 3, 4, 6, - 2, 6, 5, 5, 3, 4, 5, 6, - 7, 1, 2, 4, 6, 4, 4, 5, - 7, 2, 6, 5, 7, 6, 7, 7 }; - return DeBruijnBytePos[((U64)((val & -(long long)val) * 0x0218A392CDABBD3FULL)) >> 58]; -# endif - } else { /* 32 bits */ -# if defined(_MSC_VER) - if (val != 0) { - unsigned long r; - _BitScanForward(&r, (U32)val); - return (unsigned)(r >> 3); - } else { - /* Should not reach this code path */ - __assume(0); - } -# elif defined(__GNUC__) && (__GNUC__ >= 3) - return (__builtin_ctz((U32)val) >> 3); -# else - static const int DeBruijnBytePos[32] = { 0, 0, 3, 0, 3, 1, 3, 0, - 3, 2, 2, 1, 3, 2, 0, 1, - 3, 3, 1, 2, 2, 2, 2, 0, - 3, 1, 2, 0, 1, 0, 1, 1 }; - return DeBruijnBytePos[((U32)((val & -(S32)val) * 0x077CB531U)) >> 27]; -# endif - } - } else { /* Big Endian CPU */ - if (MEM_64bits()) { -# if defined(_MSC_VER) && defined(_WIN64) -# if STATIC_BMI2 - return _lzcnt_u64(val) >> 3; -# else - if (val != 0) { - unsigned long r; - _BitScanReverse64(&r, (U64)val); - return (unsigned)(r >> 3); - } else { - /* Should not reach this code path */ - __assume(0); - } -# endif -# elif defined(__GNUC__) && (__GNUC__ >= 4) - return (__builtin_clzll(val) >> 3); -# else - unsigned r; - const unsigned n32 = sizeof(size_t)*4; /* calculate this way due to compiler complaining in 32-bits mode */ - if (!(val>>n32)) { r=4; } else { r=0; val>>=n32; } - if (!(val>>16)) { r+=2; val>>=8; } else { val>>=24; } - r += (!val); - return r; -# endif - } else { /* 32 bits */ -# if defined(_MSC_VER) - if (val != 0) { - unsigned long r; - _BitScanReverse(&r, (unsigned long)val); - return (unsigned)(r >> 3); - } else { - /* Should not reach this code path */ - __assume(0); - } -# elif defined(__GNUC__) && (__GNUC__ >= 3) - return (__builtin_clz((U32)val) >> 3); -# else - unsigned r; - if (!(val>>16)) { r=2; val>>=8; } else { r=0; val>>=24; } - r += (!val); - return r; -# endif - } } -} - - -MEM_STATIC size_t ZSTD_count(const BYTE* pIn, const BYTE* pMatch, const BYTE* const pInLimit) -{ - const BYTE* const pStart = pIn; - const BYTE* const pInLoopLimit = pInLimit - (sizeof(size_t)-1); - - if (pIn < pInLoopLimit) { - { size_t const diff = MEM_readST(pMatch) ^ MEM_readST(pIn); - if (diff) return ZSTD_NbCommonBytes(diff); } - pIn+=sizeof(size_t); pMatch+=sizeof(size_t); - while (pIn < pInLoopLimit) { - size_t const diff = MEM_readST(pMatch) ^ MEM_readST(pIn); - if (!diff) { pIn+=sizeof(size_t); pMatch+=sizeof(size_t); continue; } - pIn += ZSTD_NbCommonBytes(diff); - return (size_t)(pIn - pStart); - } } - if (MEM_64bits() && (pIn<(pInLimit-3)) && (MEM_read32(pMatch) == MEM_read32(pIn))) { pIn+=4; pMatch+=4; } - if ((pIn<(pInLimit-1)) && (MEM_read16(pMatch) == MEM_read16(pIn))) { pIn+=2; pMatch+=2; } - if ((pIn> (32-h) ; } -MEM_STATIC size_t ZSTD_hash3Ptr(const void* ptr, U32 h) { return ZSTD_hash3(MEM_readLE32(ptr), h); } /* only in zstd_opt.h */ - -static const U32 prime4bytes = 2654435761U; -static U32 ZSTD_hash4(U32 u, U32 h) { return (u * prime4bytes) >> (32-h) ; } -static size_t ZSTD_hash4Ptr(const void* ptr, U32 h) { return ZSTD_hash4(MEM_read32(ptr), h); } - -static const U64 prime5bytes = 889523592379ULL; -static size_t ZSTD_hash5(U64 u, U32 h) { return (size_t)(((u << (64-40)) * prime5bytes) >> (64-h)) ; } -static size_t ZSTD_hash5Ptr(const void* p, U32 h) { return ZSTD_hash5(MEM_readLE64(p), h); } - -static const U64 prime6bytes = 227718039650203ULL; -static size_t ZSTD_hash6(U64 u, U32 h) { return (size_t)(((u << (64-48)) * prime6bytes) >> (64-h)) ; } -static size_t ZSTD_hash6Ptr(const void* p, U32 h) { return ZSTD_hash6(MEM_readLE64(p), h); } - -static const U64 prime7bytes = 58295818150454627ULL; -static size_t ZSTD_hash7(U64 u, U32 h) { return (size_t)(((u << (64-56)) * prime7bytes) >> (64-h)) ; } -static size_t ZSTD_hash7Ptr(const void* p, U32 h) { return ZSTD_hash7(MEM_readLE64(p), h); } - -static const U64 prime8bytes = 0xCF1BBCDCB7A56463ULL; -static size_t ZSTD_hash8(U64 u, U32 h) { return (size_t)(((u) * prime8bytes) >> (64-h)) ; } -static size_t ZSTD_hash8Ptr(const void* p, U32 h) { return ZSTD_hash8(MEM_readLE64(p), h); } - -MEM_STATIC FORCE_INLINE_ATTR -size_t ZSTD_hashPtr(const void* p, U32 hBits, U32 mls) -{ - switch(mls) - { - default: - case 4: return ZSTD_hash4Ptr(p, hBits); - case 5: return ZSTD_hash5Ptr(p, hBits); - case 6: return ZSTD_hash6Ptr(p, hBits); - case 7: return ZSTD_hash7Ptr(p, hBits); - case 8: return ZSTD_hash8Ptr(p, hBits); - } -} - -/** ZSTD_ipow() : - * Return base^exponent. - */ -static U64 ZSTD_ipow(U64 base, U64 exponent) -{ - U64 power = 1; - while (exponent) { - if (exponent & 1) power *= base; - exponent >>= 1; - base *= base; - } - return power; -} - -#define ZSTD_ROLL_HASH_CHAR_OFFSET 10 - -/** ZSTD_rollingHash_append() : - * Add the buffer to the hash value. - */ -static U64 ZSTD_rollingHash_append(U64 hash, void const* buf, size_t size) -{ - BYTE const* istart = (BYTE const*)buf; - size_t pos; - for (pos = 0; pos < size; ++pos) { - hash *= prime8bytes; - hash += istart[pos] + ZSTD_ROLL_HASH_CHAR_OFFSET; - } - return hash; -} - -/** ZSTD_rollingHash_compute() : - * Compute the rolling hash value of the buffer. - */ -MEM_STATIC U64 ZSTD_rollingHash_compute(void const* buf, size_t size) -{ - return ZSTD_rollingHash_append(0, buf, size); -} - -/** ZSTD_rollingHash_primePower() : - * Compute the primePower to be passed to ZSTD_rollingHash_rotate() for a hash - * over a window of length bytes. - */ -MEM_STATIC U64 ZSTD_rollingHash_primePower(U32 length) -{ - return ZSTD_ipow(prime8bytes, length - 1); -} - -/** ZSTD_rollingHash_rotate() : - * Rotate the rolling hash by one byte. - */ -MEM_STATIC U64 ZSTD_rollingHash_rotate(U64 hash, BYTE toRemove, BYTE toAdd, U64 primePower) -{ - hash -= (toRemove + ZSTD_ROLL_HASH_CHAR_OFFSET) * primePower; - hash *= prime8bytes; - hash += toAdd + ZSTD_ROLL_HASH_CHAR_OFFSET; - return hash; -} - -/*-************************************* -* Round buffer management -***************************************/ -#if (ZSTD_WINDOWLOG_MAX_64 > 31) -# error "ZSTD_WINDOWLOG_MAX is too large : would overflow ZSTD_CURRENT_MAX" -#endif -/* Max current allowed */ -#define ZSTD_CURRENT_MAX ((3U << 29) + (1U << ZSTD_WINDOWLOG_MAX)) -/* Maximum chunk size before overflow correction needs to be called again */ -#define ZSTD_CHUNKSIZE_MAX \ - ( ((U32)-1) /* Maximum ending current index */ \ - - ZSTD_CURRENT_MAX) /* Maximum beginning lowLimit */ - -/** - * ZSTD_window_clear(): - * Clears the window containing the history by simply setting it to empty. - */ -MEM_STATIC void ZSTD_window_clear(ZSTD_window_t* window) -{ - size_t const endT = (size_t)(window->nextSrc - window->base); - U32 const end = (U32)endT; - - window->lowLimit = end; - window->dictLimit = end; -} - -MEM_STATIC U32 ZSTD_window_isEmpty(ZSTD_window_t const window) -{ - return window.dictLimit == ZSTD_WINDOW_START_INDEX && - window.lowLimit == ZSTD_WINDOW_START_INDEX && - (window.nextSrc - window.base) == ZSTD_WINDOW_START_INDEX; -} - -/** - * ZSTD_window_hasExtDict(): - * Returns non-zero if the window has a non-empty extDict. - */ -MEM_STATIC U32 ZSTD_window_hasExtDict(ZSTD_window_t const window) -{ - return window.lowLimit < window.dictLimit; -} - -/** - * ZSTD_matchState_dictMode(): - * Inspects the provided matchState and figures out what dictMode should be - * passed to the compressor. - */ -MEM_STATIC ZSTD_dictMode_e ZSTD_matchState_dictMode(const ZSTD_matchState_t *ms) -{ - return ZSTD_window_hasExtDict(ms->window) ? - ZSTD_extDict : - ms->dictMatchState != NULL ? - (ms->dictMatchState->dedicatedDictSearch ? ZSTD_dedicatedDictSearch : ZSTD_dictMatchState) : - ZSTD_noDict; -} - -/* Defining this macro to non-zero tells zstd to run the overflow correction - * code much more frequently. This is very inefficient, and should only be - * used for tests and fuzzers. - */ -#ifndef ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY -# ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION -# define ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY 1 -# else -# define ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY 0 -# endif -#endif - -/** - * ZSTD_window_canOverflowCorrect(): - * Returns non-zero if the indices are large enough for overflow correction - * to work correctly without impacting compression ratio. - */ -MEM_STATIC U32 ZSTD_window_canOverflowCorrect(ZSTD_window_t const window, - U32 cycleLog, - U32 maxDist, - U32 loadedDictEnd, - void const* src) -{ - U32 const cycleSize = 1u << cycleLog; - U32 const curr = (U32)((BYTE const*)src - window.base); - U32 const minIndexToOverflowCorrect = cycleSize - + MAX(maxDist, cycleSize) - + ZSTD_WINDOW_START_INDEX; - - /* Adjust the min index to backoff the overflow correction frequency, - * so we don't waste too much CPU in overflow correction. If this - * computation overflows we don't really care, we just need to make - * sure it is at least minIndexToOverflowCorrect. - */ - U32 const adjustment = window.nbOverflowCorrections + 1; - U32 const adjustedIndex = MAX(minIndexToOverflowCorrect * adjustment, - minIndexToOverflowCorrect); - U32 const indexLargeEnough = curr > adjustedIndex; - - /* Only overflow correct early if the dictionary is invalidated already, - * so we don't hurt compression ratio. - */ - U32 const dictionaryInvalidated = curr > maxDist + loadedDictEnd; - - return indexLargeEnough && dictionaryInvalidated; -} - -/** - * ZSTD_window_needOverflowCorrection(): - * Returns non-zero if the indices are getting too large and need overflow - * protection. - */ -MEM_STATIC U32 ZSTD_window_needOverflowCorrection(ZSTD_window_t const window, - U32 cycleLog, - U32 maxDist, - U32 loadedDictEnd, - void const* src, - void const* srcEnd) -{ - U32 const curr = (U32)((BYTE const*)srcEnd - window.base); - if (ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY) { - if (ZSTD_window_canOverflowCorrect(window, cycleLog, maxDist, loadedDictEnd, src)) { - return 1; - } - } - return curr > ZSTD_CURRENT_MAX; -} - -/** - * ZSTD_window_correctOverflow(): - * Reduces the indices to protect from index overflow. - * Returns the correction made to the indices, which must be applied to every - * stored index. - * - * The least significant cycleLog bits of the indices must remain the same, - * which may be 0. Every index up to maxDist in the past must be valid. - */ -MEM_STATIC U32 ZSTD_window_correctOverflow(ZSTD_window_t* window, U32 cycleLog, - U32 maxDist, void const* src) -{ - /* preemptive overflow correction: - * 1. correction is large enough: - * lowLimit > (3<<29) ==> current > 3<<29 + 1< (3<<29 + 1< (3<<29) - (1< (3<<29) - (1<<30) (NOTE: chainLog <= 30) - * > 1<<29 - * - * 2. (ip+ZSTD_CHUNKSIZE_MAX - cctx->base) doesn't overflow: - * After correction, current is less than (1<base < 1<<32. - * 3. (cctx->lowLimit + 1< 3<<29 + 1<base); - U32 const currentCycle = curr & cycleMask; - /* Ensure newCurrent - maxDist >= ZSTD_WINDOW_START_INDEX. */ - U32 const currentCycleCorrection = currentCycle < ZSTD_WINDOW_START_INDEX - ? MAX(cycleSize, ZSTD_WINDOW_START_INDEX) - : 0; - U32 const newCurrent = currentCycle - + currentCycleCorrection - + MAX(maxDist, cycleSize); - U32 const correction = curr - newCurrent; - /* maxDist must be a power of two so that: - * (newCurrent & cycleMask) == (curr & cycleMask) - * This is required to not corrupt the chains / binary tree. - */ - assert((maxDist & (maxDist - 1)) == 0); - assert((curr & cycleMask) == (newCurrent & cycleMask)); - assert(curr > newCurrent); - if (!ZSTD_WINDOW_OVERFLOW_CORRECT_FREQUENTLY) { - /* Loose bound, should be around 1<<29 (see above) */ - assert(correction > 1<<28); - } - - window->base += correction; - window->dictBase += correction; - if (window->lowLimit < correction + ZSTD_WINDOW_START_INDEX) { - window->lowLimit = ZSTD_WINDOW_START_INDEX; - } else { - window->lowLimit -= correction; - } - if (window->dictLimit < correction + ZSTD_WINDOW_START_INDEX) { - window->dictLimit = ZSTD_WINDOW_START_INDEX; - } else { - window->dictLimit -= correction; - } - - /* Ensure we can still reference the full window. */ - assert(newCurrent >= maxDist); - assert(newCurrent - maxDist >= ZSTD_WINDOW_START_INDEX); - /* Ensure that lowLimit and dictLimit didn't underflow. */ - assert(window->lowLimit <= newCurrent); - assert(window->dictLimit <= newCurrent); - - ++window->nbOverflowCorrections; - - DEBUGLOG(4, "Correction of 0x%x bytes to lowLimit=0x%x", correction, - window->lowLimit); - return correction; -} - -/** - * ZSTD_window_enforceMaxDist(): - * Updates lowLimit so that: - * (srcEnd - base) - lowLimit == maxDist + loadedDictEnd - * - * It ensures index is valid as long as index >= lowLimit. - * This must be called before a block compression call. - * - * loadedDictEnd is only defined if a dictionary is in use for current compression. - * As the name implies, loadedDictEnd represents the index at end of dictionary. - * The value lies within context's referential, it can be directly compared to blockEndIdx. - * - * If loadedDictEndPtr is NULL, no dictionary is in use, and we use loadedDictEnd == 0. - * If loadedDictEndPtr is not NULL, we set it to zero after updating lowLimit. - * This is because dictionaries are allowed to be referenced fully - * as long as the last byte of the dictionary is in the window. - * Once input has progressed beyond window size, dictionary cannot be referenced anymore. - * - * In normal dict mode, the dictionary lies between lowLimit and dictLimit. - * In dictMatchState mode, lowLimit and dictLimit are the same, - * and the dictionary is below them. - * forceWindow and dictMatchState are therefore incompatible. - */ -MEM_STATIC void -ZSTD_window_enforceMaxDist(ZSTD_window_t* window, - const void* blockEnd, - U32 maxDist, - U32* loadedDictEndPtr, - const ZSTD_matchState_t** dictMatchStatePtr) -{ - U32 const blockEndIdx = (U32)((BYTE const*)blockEnd - window->base); - U32 const loadedDictEnd = (loadedDictEndPtr != NULL) ? *loadedDictEndPtr : 0; - DEBUGLOG(5, "ZSTD_window_enforceMaxDist: blockEndIdx=%u, maxDist=%u, loadedDictEnd=%u", - (unsigned)blockEndIdx, (unsigned)maxDist, (unsigned)loadedDictEnd); - - /* - When there is no dictionary : loadedDictEnd == 0. - In which case, the test (blockEndIdx > maxDist) is merely to avoid - overflowing next operation `newLowLimit = blockEndIdx - maxDist`. - - When there is a standard dictionary : - Index referential is copied from the dictionary, - which means it starts from 0. - In which case, loadedDictEnd == dictSize, - and it makes sense to compare `blockEndIdx > maxDist + dictSize` - since `blockEndIdx` also starts from zero. - - When there is an attached dictionary : - loadedDictEnd is expressed within the referential of the context, - so it can be directly compared against blockEndIdx. - */ - if (blockEndIdx > maxDist + loadedDictEnd) { - U32 const newLowLimit = blockEndIdx - maxDist; - if (window->lowLimit < newLowLimit) window->lowLimit = newLowLimit; - if (window->dictLimit < window->lowLimit) { - DEBUGLOG(5, "Update dictLimit to match lowLimit, from %u to %u", - (unsigned)window->dictLimit, (unsigned)window->lowLimit); - window->dictLimit = window->lowLimit; - } - /* On reaching window size, dictionaries are invalidated */ - if (loadedDictEndPtr) *loadedDictEndPtr = 0; - if (dictMatchStatePtr) *dictMatchStatePtr = NULL; - } -} - -/* Similar to ZSTD_window_enforceMaxDist(), - * but only invalidates dictionary - * when input progresses beyond window size. - * assumption : loadedDictEndPtr and dictMatchStatePtr are valid (non NULL) - * loadedDictEnd uses same referential as window->base - * maxDist is the window size */ -MEM_STATIC void -ZSTD_checkDictValidity(const ZSTD_window_t* window, - const void* blockEnd, - U32 maxDist, - U32* loadedDictEndPtr, - const ZSTD_matchState_t** dictMatchStatePtr) -{ - assert(loadedDictEndPtr != NULL); - assert(dictMatchStatePtr != NULL); - { U32 const blockEndIdx = (U32)((BYTE const*)blockEnd - window->base); - U32 const loadedDictEnd = *loadedDictEndPtr; - DEBUGLOG(5, "ZSTD_checkDictValidity: blockEndIdx=%u, maxDist=%u, loadedDictEnd=%u", - (unsigned)blockEndIdx, (unsigned)maxDist, (unsigned)loadedDictEnd); - assert(blockEndIdx >= loadedDictEnd); - - if (blockEndIdx > loadedDictEnd + maxDist) { - /* On reaching window size, dictionaries are invalidated. - * For simplification, if window size is reached anywhere within next block, - * the dictionary is invalidated for the full block. - */ - DEBUGLOG(6, "invalidating dictionary for current block (distance > windowSize)"); - *loadedDictEndPtr = 0; - *dictMatchStatePtr = NULL; - } else { - if (*loadedDictEndPtr != 0) { - DEBUGLOG(6, "dictionary considered valid for current block"); - } } } -} - -MEM_STATIC void ZSTD_window_init(ZSTD_window_t* window) { - ZSTD_memset(window, 0, sizeof(*window)); - window->base = (BYTE const*)" "; - window->dictBase = (BYTE const*)" "; - ZSTD_STATIC_ASSERT(ZSTD_DUBT_UNSORTED_MARK < ZSTD_WINDOW_START_INDEX); /* Start above ZSTD_DUBT_UNSORTED_MARK */ - window->dictLimit = ZSTD_WINDOW_START_INDEX; /* start from >0, so that 1st position is valid */ - window->lowLimit = ZSTD_WINDOW_START_INDEX; /* it ensures first and later CCtx usages compress the same */ - window->nextSrc = window->base + ZSTD_WINDOW_START_INDEX; /* see issue #1241 */ - window->nbOverflowCorrections = 0; -} - -/** - * ZSTD_window_update(): - * Updates the window by appending [src, src + srcSize) to the window. - * If it is not contiguous, the current prefix becomes the extDict, and we - * forget about the extDict. Handles overlap of the prefix and extDict. - * Returns non-zero if the segment is contiguous. - */ -MEM_STATIC U32 ZSTD_window_update(ZSTD_window_t* window, - void const* src, size_t srcSize, - int forceNonContiguous) -{ - BYTE const* const ip = (BYTE const*)src; - U32 contiguous = 1; - DEBUGLOG(5, "ZSTD_window_update"); - if (srcSize == 0) - return contiguous; - assert(window->base != NULL); - assert(window->dictBase != NULL); - /* Check if blocks follow each other */ - if (src != window->nextSrc || forceNonContiguous) { - /* not contiguous */ - size_t const distanceFromBase = (size_t)(window->nextSrc - window->base); - DEBUGLOG(5, "Non contiguous blocks, new segment starts at %u", window->dictLimit); - window->lowLimit = window->dictLimit; - assert(distanceFromBase == (size_t)(U32)distanceFromBase); /* should never overflow */ - window->dictLimit = (U32)distanceFromBase; - window->dictBase = window->base; - window->base = ip - distanceFromBase; - /* ms->nextToUpdate = window->dictLimit; */ - if (window->dictLimit - window->lowLimit < HASH_READ_SIZE) window->lowLimit = window->dictLimit; /* too small extDict */ - contiguous = 0; - } - window->nextSrc = ip + srcSize; - /* if input and dictionary overlap : reduce dictionary (area presumed modified by input) */ - if ( (ip+srcSize > window->dictBase + window->lowLimit) - & (ip < window->dictBase + window->dictLimit)) { - ptrdiff_t const highInputIdx = (ip + srcSize) - window->dictBase; - U32 const lowLimitMax = (highInputIdx > (ptrdiff_t)window->dictLimit) ? window->dictLimit : (U32)highInputIdx; - window->lowLimit = lowLimitMax; - DEBUGLOG(5, "Overlapping extDict and input : new lowLimit = %u", window->lowLimit); - } - return contiguous; -} - -/** - * Returns the lowest allowed match index. It may either be in the ext-dict or the prefix. - */ -MEM_STATIC U32 ZSTD_getLowestMatchIndex(const ZSTD_matchState_t* ms, U32 curr, unsigned windowLog) -{ - U32 const maxDistance = 1U << windowLog; - U32 const lowestValid = ms->window.lowLimit; - U32 const withinWindow = (curr - lowestValid > maxDistance) ? curr - maxDistance : lowestValid; - U32 const isDictionary = (ms->loadedDictEnd != 0); - /* When using a dictionary the entire dictionary is valid if a single byte of the dictionary - * is within the window. We invalidate the dictionary (and set loadedDictEnd to 0) when it isn't - * valid for the entire block. So this check is sufficient to find the lowest valid match index. - */ - U32 const matchLowest = isDictionary ? lowestValid : withinWindow; - return matchLowest; -} - -/** - * Returns the lowest allowed match index in the prefix. - */ -MEM_STATIC U32 ZSTD_getLowestPrefixIndex(const ZSTD_matchState_t* ms, U32 curr, unsigned windowLog) -{ - U32 const maxDistance = 1U << windowLog; - U32 const lowestValid = ms->window.dictLimit; - U32 const withinWindow = (curr - lowestValid > maxDistance) ? curr - maxDistance : lowestValid; - U32 const isDictionary = (ms->loadedDictEnd != 0); - /* When computing the lowest prefix index we need to take the dictionary into account to handle - * the edge case where the dictionary and the source are contiguous in memory. - */ - U32 const matchLowest = isDictionary ? lowestValid : withinWindow; - return matchLowest; -} - - - -/* debug functions */ -#if (DEBUGLEVEL>=2) - -MEM_STATIC double ZSTD_fWeight(U32 rawStat) -{ - U32 const fp_accuracy = 8; - U32 const fp_multiplier = (1 << fp_accuracy); - U32 const newStat = rawStat + 1; - U32 const hb = ZSTD_highbit32(newStat); - U32 const BWeight = hb * fp_multiplier; - U32 const FWeight = (newStat << fp_accuracy) >> hb; - U32 const weight = BWeight + FWeight; - assert(hb + fp_accuracy < 31); - return (double)weight / fp_multiplier; -} - -/* display a table content, - * listing each element, its frequency, and its predicted bit cost */ -MEM_STATIC void ZSTD_debugTable(const U32* table, U32 max) -{ - unsigned u, sum; - for (u=0, sum=0; u<=max; u++) sum += table[u]; - DEBUGLOG(2, "total nb elts: %u", sum); - for (u=0; u<=max; u++) { - DEBUGLOG(2, "%2u: %5u (%.2f)", - u, table[u], ZSTD_fWeight(sum) - ZSTD_fWeight(table[u]) ); - } -} - -#endif - - -#if defined (__cplusplus) -} -#endif - -/* =============================================================== - * Shared internal declarations - * These prototypes may be called from sources not in lib/compress - * =============================================================== */ - -/* ZSTD_loadCEntropy() : - * dict : must point at beginning of a valid zstd dictionary. - * return : size of dictionary header (size of magic number + dict ID + entropy tables) - * assumptions : magic number supposed already checked - * and dictSize >= 8 */ -size_t ZSTD_loadCEntropy(ZSTD_compressedBlockState_t* bs, void* workspace, - const void* const dict, size_t dictSize); - -void ZSTD_reset_compressedBlockState(ZSTD_compressedBlockState_t* bs); - -/* ============================================================== - * Private declarations - * These prototypes shall only be called from within lib/compress - * ============================================================== */ - -/* ZSTD_getCParamsFromCCtxParams() : - * cParams are built depending on compressionLevel, src size hints, - * LDM and manually set compression parameters. - * Note: srcSizeHint == 0 means 0! - */ -ZSTD_compressionParameters ZSTD_getCParamsFromCCtxParams( - const ZSTD_CCtx_params* CCtxParams, U64 srcSizeHint, size_t dictSize, ZSTD_cParamMode_e mode); - -/*! ZSTD_initCStream_internal() : - * Private use only. Init streaming operation. - * expects params to be valid. - * must receive dict, or cdict, or none, but not both. - * @return : 0, or an error code */ -size_t ZSTD_initCStream_internal(ZSTD_CStream* zcs, - const void* dict, size_t dictSize, - const ZSTD_CDict* cdict, - const ZSTD_CCtx_params* params, unsigned long long pledgedSrcSize); - -void ZSTD_resetSeqStore(seqStore_t* ssPtr); - -/*! ZSTD_getCParamsFromCDict() : - * as the name implies */ -ZSTD_compressionParameters ZSTD_getCParamsFromCDict(const ZSTD_CDict* cdict); - -/* ZSTD_compressBegin_advanced_internal() : - * Private use only. To be called from zstdmt_compress.c. */ -size_t ZSTD_compressBegin_advanced_internal(ZSTD_CCtx* cctx, - const void* dict, size_t dictSize, - ZSTD_dictContentType_e dictContentType, - ZSTD_dictTableLoadMethod_e dtlm, - const ZSTD_CDict* cdict, - const ZSTD_CCtx_params* params, - unsigned long long pledgedSrcSize); - -/* ZSTD_compress_advanced_internal() : - * Private use only. To be called from zstdmt_compress.c. */ -size_t ZSTD_compress_advanced_internal(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const void* dict,size_t dictSize, - const ZSTD_CCtx_params* params); - - -/* ZSTD_writeLastEmptyBlock() : - * output an empty Block with end-of-frame mark to complete a frame - * @return : size of data written into `dst` (== ZSTD_blockHeaderSize (defined in zstd_internal.h)) - * or an error code if `dstCapacity` is too small ( 1 */ -U32 ZSTD_cycleLog(U32 hashLog, ZSTD_strategy strat); - -/** ZSTD_CCtx_trace() : - * Trace the end of a compression call. - */ -void ZSTD_CCtx_trace(ZSTD_CCtx* cctx, size_t extraCSize); - -#endif /* ZSTD_COMPRESS_H */ diff --git a/dep/zstd/lib/compress/zstd_compress_literals.c b/dep/zstd/lib/compress/zstd_compress_literals.c deleted file mode 100644 index 52b0a8059..000000000 --- a/dep/zstd/lib/compress/zstd_compress_literals.c +++ /dev/null @@ -1,159 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - /*-************************************* - * Dependencies - ***************************************/ -#include "zstd_compress_literals.h" - -size_t ZSTD_noCompressLiterals (void* dst, size_t dstCapacity, const void* src, size_t srcSize) -{ - BYTE* const ostart = (BYTE*)dst; - U32 const flSize = 1 + (srcSize>31) + (srcSize>4095); - - RETURN_ERROR_IF(srcSize + flSize > dstCapacity, dstSize_tooSmall, ""); - - switch(flSize) - { - case 1: /* 2 - 1 - 5 */ - ostart[0] = (BYTE)((U32)set_basic + (srcSize<<3)); - break; - case 2: /* 2 - 2 - 12 */ - MEM_writeLE16(ostart, (U16)((U32)set_basic + (1<<2) + (srcSize<<4))); - break; - case 3: /* 2 - 2 - 20 */ - MEM_writeLE32(ostart, (U32)((U32)set_basic + (3<<2) + (srcSize<<4))); - break; - default: /* not necessary : flSize is {1,2,3} */ - assert(0); - } - - ZSTD_memcpy(ostart + flSize, src, srcSize); - DEBUGLOG(5, "Raw literals: %u -> %u", (U32)srcSize, (U32)(srcSize + flSize)); - return srcSize + flSize; -} - -size_t ZSTD_compressRleLiteralsBlock (void* dst, size_t dstCapacity, const void* src, size_t srcSize) -{ - BYTE* const ostart = (BYTE*)dst; - U32 const flSize = 1 + (srcSize>31) + (srcSize>4095); - - (void)dstCapacity; /* dstCapacity already guaranteed to be >=4, hence large enough */ - - switch(flSize) - { - case 1: /* 2 - 1 - 5 */ - ostart[0] = (BYTE)((U32)set_rle + (srcSize<<3)); - break; - case 2: /* 2 - 2 - 12 */ - MEM_writeLE16(ostart, (U16)((U32)set_rle + (1<<2) + (srcSize<<4))); - break; - case 3: /* 2 - 2 - 20 */ - MEM_writeLE32(ostart, (U32)((U32)set_rle + (3<<2) + (srcSize<<4))); - break; - default: /* not necessary : flSize is {1,2,3} */ - assert(0); - } - - ostart[flSize] = *(const BYTE*)src; - DEBUGLOG(5, "RLE literals: %u -> %u", (U32)srcSize, (U32)flSize + 1); - return flSize+1; -} - -size_t ZSTD_compressLiterals (ZSTD_hufCTables_t const* prevHuf, - ZSTD_hufCTables_t* nextHuf, - ZSTD_strategy strategy, int disableLiteralCompression, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - void* entropyWorkspace, size_t entropyWorkspaceSize, - const int bmi2, - unsigned suspectUncompressible) -{ - size_t const minGain = ZSTD_minGain(srcSize, strategy); - size_t const lhSize = 3 + (srcSize >= 1 KB) + (srcSize >= 16 KB); - BYTE* const ostart = (BYTE*)dst; - U32 singleStream = srcSize < 256; - symbolEncodingType_e hType = set_compressed; - size_t cLitSize; - - DEBUGLOG(5,"ZSTD_compressLiterals (disableLiteralCompression=%i srcSize=%u)", - disableLiteralCompression, (U32)srcSize); - - /* Prepare nextEntropy assuming reusing the existing table */ - ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); - - if (disableLiteralCompression) - return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize); - - /* small ? don't even attempt compression (speed opt) */ -# define COMPRESS_LITERALS_SIZE_MIN 63 - { size_t const minLitSize = (prevHuf->repeatMode == HUF_repeat_valid) ? 6 : COMPRESS_LITERALS_SIZE_MIN; - if (srcSize <= minLitSize) return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize); - } - - RETURN_ERROR_IF(dstCapacity < lhSize+1, dstSize_tooSmall, "not enough space for compression"); - { HUF_repeat repeat = prevHuf->repeatMode; - int const preferRepeat = strategy < ZSTD_lazy ? srcSize <= 1024 : 0; - if (repeat == HUF_repeat_valid && lhSize == 3) singleStream = 1; - cLitSize = singleStream ? - HUF_compress1X_repeat( - ostart+lhSize, dstCapacity-lhSize, src, srcSize, - HUF_SYMBOLVALUE_MAX, HUF_TABLELOG_DEFAULT, entropyWorkspace, entropyWorkspaceSize, - (HUF_CElt*)nextHuf->CTable, &repeat, preferRepeat, bmi2, suspectUncompressible) : - HUF_compress4X_repeat( - ostart+lhSize, dstCapacity-lhSize, src, srcSize, - HUF_SYMBOLVALUE_MAX, HUF_TABLELOG_DEFAULT, entropyWorkspace, entropyWorkspaceSize, - (HUF_CElt*)nextHuf->CTable, &repeat, preferRepeat, bmi2, suspectUncompressible); - if (repeat != HUF_repeat_none) { - /* reused the existing table */ - DEBUGLOG(5, "Reusing previous huffman table"); - hType = set_repeat; - } - } - - if ((cLitSize==0) || (cLitSize >= srcSize - minGain) || ERR_isError(cLitSize)) { - ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); - return ZSTD_noCompressLiterals(dst, dstCapacity, src, srcSize); - } - if (cLitSize==1) { - ZSTD_memcpy(nextHuf, prevHuf, sizeof(*prevHuf)); - return ZSTD_compressRleLiteralsBlock(dst, dstCapacity, src, srcSize); - } - - if (hType == set_compressed) { - /* using a newly constructed table */ - nextHuf->repeatMode = HUF_repeat_check; - } - - /* Build header */ - switch(lhSize) - { - case 3: /* 2 - 2 - 10 - 10 */ - { U32 const lhc = hType + ((!singleStream) << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<14); - MEM_writeLE24(ostart, lhc); - break; - } - case 4: /* 2 - 2 - 14 - 14 */ - { U32 const lhc = hType + (2 << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<18); - MEM_writeLE32(ostart, lhc); - break; - } - case 5: /* 2 - 2 - 18 - 18 */ - { U32 const lhc = hType + (3 << 2) + ((U32)srcSize<<4) + ((U32)cLitSize<<22); - MEM_writeLE32(ostart, lhc); - ostart[4] = (BYTE)(cLitSize >> 10); - break; - } - default: /* not possible : lhSize is {3,4,5} */ - assert(0); - } - DEBUGLOG(5, "Compressed literals: %u -> %u", (U32)srcSize, (U32)(lhSize+cLitSize)); - return lhSize+cLitSize; -} diff --git a/dep/zstd/lib/compress/zstd_compress_literals.h b/dep/zstd/lib/compress/zstd_compress_literals.h deleted file mode 100644 index 9775fb97c..000000000 --- a/dep/zstd/lib/compress/zstd_compress_literals.h +++ /dev/null @@ -1,31 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_COMPRESS_LITERALS_H -#define ZSTD_COMPRESS_LITERALS_H - -#include "zstd_compress_internal.h" /* ZSTD_hufCTables_t, ZSTD_minGain() */ - - -size_t ZSTD_noCompressLiterals (void* dst, size_t dstCapacity, const void* src, size_t srcSize); - -size_t ZSTD_compressRleLiteralsBlock (void* dst, size_t dstCapacity, const void* src, size_t srcSize); - -/* If suspectUncompressible then some sampling checks will be run to potentially skip huffman coding */ -size_t ZSTD_compressLiterals (ZSTD_hufCTables_t const* prevHuf, - ZSTD_hufCTables_t* nextHuf, - ZSTD_strategy strategy, int disableLiteralCompression, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - void* entropyWorkspace, size_t entropyWorkspaceSize, - const int bmi2, - unsigned suspectUncompressible); - -#endif /* ZSTD_COMPRESS_LITERALS_H */ diff --git a/dep/zstd/lib/compress/zstd_compress_sequences.c b/dep/zstd/lib/compress/zstd_compress_sequences.c deleted file mode 100644 index f1e40af2e..000000000 --- a/dep/zstd/lib/compress/zstd_compress_sequences.c +++ /dev/null @@ -1,442 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - /*-************************************* - * Dependencies - ***************************************/ -#include "zstd_compress_sequences.h" - -/** - * -log2(x / 256) lookup table for x in [0, 256). - * If x == 0: Return 0 - * Else: Return floor(-log2(x / 256) * 256) - */ -static unsigned const kInverseProbabilityLog256[256] = { - 0, 2048, 1792, 1642, 1536, 1453, 1386, 1329, 1280, 1236, 1197, 1162, - 1130, 1100, 1073, 1047, 1024, 1001, 980, 960, 941, 923, 906, 889, - 874, 859, 844, 830, 817, 804, 791, 779, 768, 756, 745, 734, - 724, 714, 704, 694, 685, 676, 667, 658, 650, 642, 633, 626, - 618, 610, 603, 595, 588, 581, 574, 567, 561, 554, 548, 542, - 535, 529, 523, 517, 512, 506, 500, 495, 489, 484, 478, 473, - 468, 463, 458, 453, 448, 443, 438, 434, 429, 424, 420, 415, - 411, 407, 402, 398, 394, 390, 386, 382, 377, 373, 370, 366, - 362, 358, 354, 350, 347, 343, 339, 336, 332, 329, 325, 322, - 318, 315, 311, 308, 305, 302, 298, 295, 292, 289, 286, 282, - 279, 276, 273, 270, 267, 264, 261, 258, 256, 253, 250, 247, - 244, 241, 239, 236, 233, 230, 228, 225, 222, 220, 217, 215, - 212, 209, 207, 204, 202, 199, 197, 194, 192, 190, 187, 185, - 182, 180, 178, 175, 173, 171, 168, 166, 164, 162, 159, 157, - 155, 153, 151, 149, 146, 144, 142, 140, 138, 136, 134, 132, - 130, 128, 126, 123, 121, 119, 117, 115, 114, 112, 110, 108, - 106, 104, 102, 100, 98, 96, 94, 93, 91, 89, 87, 85, - 83, 82, 80, 78, 76, 74, 73, 71, 69, 67, 66, 64, - 62, 61, 59, 57, 55, 54, 52, 50, 49, 47, 46, 44, - 42, 41, 39, 37, 36, 34, 33, 31, 30, 28, 26, 25, - 23, 22, 20, 19, 17, 16, 14, 13, 11, 10, 8, 7, - 5, 4, 2, 1, -}; - -static unsigned ZSTD_getFSEMaxSymbolValue(FSE_CTable const* ctable) { - void const* ptr = ctable; - U16 const* u16ptr = (U16 const*)ptr; - U32 const maxSymbolValue = MEM_read16(u16ptr + 1); - return maxSymbolValue; -} - -/** - * Returns true if we should use ncount=-1 else we should - * use ncount=1 for low probability symbols instead. - */ -static unsigned ZSTD_useLowProbCount(size_t const nbSeq) -{ - /* Heuristic: This should cover most blocks <= 16K and - * start to fade out after 16K to about 32K depending on - * comprssibility. - */ - return nbSeq >= 2048; -} - -/** - * Returns the cost in bytes of encoding the normalized count header. - * Returns an error if any of the helper functions return an error. - */ -static size_t ZSTD_NCountCost(unsigned const* count, unsigned const max, - size_t const nbSeq, unsigned const FSELog) -{ - BYTE wksp[FSE_NCOUNTBOUND]; - S16 norm[MaxSeq + 1]; - const U32 tableLog = FSE_optimalTableLog(FSELog, nbSeq, max); - FORWARD_IF_ERROR(FSE_normalizeCount(norm, tableLog, count, nbSeq, max, ZSTD_useLowProbCount(nbSeq)), ""); - return FSE_writeNCount(wksp, sizeof(wksp), norm, max, tableLog); -} - -/** - * Returns the cost in bits of encoding the distribution described by count - * using the entropy bound. - */ -static size_t ZSTD_entropyCost(unsigned const* count, unsigned const max, size_t const total) -{ - unsigned cost = 0; - unsigned s; - - assert(total > 0); - for (s = 0; s <= max; ++s) { - unsigned norm = (unsigned)((256 * count[s]) / total); - if (count[s] != 0 && norm == 0) - norm = 1; - assert(count[s] < total); - cost += count[s] * kInverseProbabilityLog256[norm]; - } - return cost >> 8; -} - -/** - * Returns the cost in bits of encoding the distribution in count using ctable. - * Returns an error if ctable cannot represent all the symbols in count. - */ -size_t ZSTD_fseBitCost( - FSE_CTable const* ctable, - unsigned const* count, - unsigned const max) -{ - unsigned const kAccuracyLog = 8; - size_t cost = 0; - unsigned s; - FSE_CState_t cstate; - FSE_initCState(&cstate, ctable); - if (ZSTD_getFSEMaxSymbolValue(ctable) < max) { - DEBUGLOG(5, "Repeat FSE_CTable has maxSymbolValue %u < %u", - ZSTD_getFSEMaxSymbolValue(ctable), max); - return ERROR(GENERIC); - } - for (s = 0; s <= max; ++s) { - unsigned const tableLog = cstate.stateLog; - unsigned const badCost = (tableLog + 1) << kAccuracyLog; - unsigned const bitCost = FSE_bitCost(cstate.symbolTT, tableLog, s, kAccuracyLog); - if (count[s] == 0) - continue; - if (bitCost >= badCost) { - DEBUGLOG(5, "Repeat FSE_CTable has Prob[%u] == 0", s); - return ERROR(GENERIC); - } - cost += (size_t)count[s] * bitCost; - } - return cost >> kAccuracyLog; -} - -/** - * Returns the cost in bits of encoding the distribution in count using the - * table described by norm. The max symbol support by norm is assumed >= max. - * norm must be valid for every symbol with non-zero probability in count. - */ -size_t ZSTD_crossEntropyCost(short const* norm, unsigned accuracyLog, - unsigned const* count, unsigned const max) -{ - unsigned const shift = 8 - accuracyLog; - size_t cost = 0; - unsigned s; - assert(accuracyLog <= 8); - for (s = 0; s <= max; ++s) { - unsigned const normAcc = (norm[s] != -1) ? (unsigned)norm[s] : 1; - unsigned const norm256 = normAcc << shift; - assert(norm256 > 0); - assert(norm256 < 256); - cost += count[s] * kInverseProbabilityLog256[norm256]; - } - return cost >> 8; -} - -symbolEncodingType_e -ZSTD_selectEncodingType( - FSE_repeat* repeatMode, unsigned const* count, unsigned const max, - size_t const mostFrequent, size_t nbSeq, unsigned const FSELog, - FSE_CTable const* prevCTable, - short const* defaultNorm, U32 defaultNormLog, - ZSTD_defaultPolicy_e const isDefaultAllowed, - ZSTD_strategy const strategy) -{ - ZSTD_STATIC_ASSERT(ZSTD_defaultDisallowed == 0 && ZSTD_defaultAllowed != 0); - if (mostFrequent == nbSeq) { - *repeatMode = FSE_repeat_none; - if (isDefaultAllowed && nbSeq <= 2) { - /* Prefer set_basic over set_rle when there are 2 or less symbols, - * since RLE uses 1 byte, but set_basic uses 5-6 bits per symbol. - * If basic encoding isn't possible, always choose RLE. - */ - DEBUGLOG(5, "Selected set_basic"); - return set_basic; - } - DEBUGLOG(5, "Selected set_rle"); - return set_rle; - } - if (strategy < ZSTD_lazy) { - if (isDefaultAllowed) { - size_t const staticFse_nbSeq_max = 1000; - size_t const mult = 10 - strategy; - size_t const baseLog = 3; - size_t const dynamicFse_nbSeq_min = (((size_t)1 << defaultNormLog) * mult) >> baseLog; /* 28-36 for offset, 56-72 for lengths */ - assert(defaultNormLog >= 5 && defaultNormLog <= 6); /* xx_DEFAULTNORMLOG */ - assert(mult <= 9 && mult >= 7); - if ( (*repeatMode == FSE_repeat_valid) - && (nbSeq < staticFse_nbSeq_max) ) { - DEBUGLOG(5, "Selected set_repeat"); - return set_repeat; - } - if ( (nbSeq < dynamicFse_nbSeq_min) - || (mostFrequent < (nbSeq >> (defaultNormLog-1))) ) { - DEBUGLOG(5, "Selected set_basic"); - /* The format allows default tables to be repeated, but it isn't useful. - * When using simple heuristics to select encoding type, we don't want - * to confuse these tables with dictionaries. When running more careful - * analysis, we don't need to waste time checking both repeating tables - * and default tables. - */ - *repeatMode = FSE_repeat_none; - return set_basic; - } - } - } else { - size_t const basicCost = isDefaultAllowed ? ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, count, max) : ERROR(GENERIC); - size_t const repeatCost = *repeatMode != FSE_repeat_none ? ZSTD_fseBitCost(prevCTable, count, max) : ERROR(GENERIC); - size_t const NCountCost = ZSTD_NCountCost(count, max, nbSeq, FSELog); - size_t const compressedCost = (NCountCost << 3) + ZSTD_entropyCost(count, max, nbSeq); - - if (isDefaultAllowed) { - assert(!ZSTD_isError(basicCost)); - assert(!(*repeatMode == FSE_repeat_valid && ZSTD_isError(repeatCost))); - } - assert(!ZSTD_isError(NCountCost)); - assert(compressedCost < ERROR(maxCode)); - DEBUGLOG(5, "Estimated bit costs: basic=%u\trepeat=%u\tcompressed=%u", - (unsigned)basicCost, (unsigned)repeatCost, (unsigned)compressedCost); - if (basicCost <= repeatCost && basicCost <= compressedCost) { - DEBUGLOG(5, "Selected set_basic"); - assert(isDefaultAllowed); - *repeatMode = FSE_repeat_none; - return set_basic; - } - if (repeatCost <= compressedCost) { - DEBUGLOG(5, "Selected set_repeat"); - assert(!ZSTD_isError(repeatCost)); - return set_repeat; - } - assert(compressedCost < basicCost && compressedCost < repeatCost); - } - DEBUGLOG(5, "Selected set_compressed"); - *repeatMode = FSE_repeat_check; - return set_compressed; -} - -typedef struct { - S16 norm[MaxSeq + 1]; - U32 wksp[FSE_BUILD_CTABLE_WORKSPACE_SIZE_U32(MaxSeq, MaxFSELog)]; -} ZSTD_BuildCTableWksp; - -size_t -ZSTD_buildCTable(void* dst, size_t dstCapacity, - FSE_CTable* nextCTable, U32 FSELog, symbolEncodingType_e type, - unsigned* count, U32 max, - const BYTE* codeTable, size_t nbSeq, - const S16* defaultNorm, U32 defaultNormLog, U32 defaultMax, - const FSE_CTable* prevCTable, size_t prevCTableSize, - void* entropyWorkspace, size_t entropyWorkspaceSize) -{ - BYTE* op = (BYTE*)dst; - const BYTE* const oend = op + dstCapacity; - DEBUGLOG(6, "ZSTD_buildCTable (dstCapacity=%u)", (unsigned)dstCapacity); - - switch (type) { - case set_rle: - FORWARD_IF_ERROR(FSE_buildCTable_rle(nextCTable, (BYTE)max), ""); - RETURN_ERROR_IF(dstCapacity==0, dstSize_tooSmall, "not enough space"); - *op = codeTable[0]; - return 1; - case set_repeat: - ZSTD_memcpy(nextCTable, prevCTable, prevCTableSize); - return 0; - case set_basic: - FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, defaultNorm, defaultMax, defaultNormLog, entropyWorkspace, entropyWorkspaceSize), ""); /* note : could be pre-calculated */ - return 0; - case set_compressed: { - ZSTD_BuildCTableWksp* wksp = (ZSTD_BuildCTableWksp*)entropyWorkspace; - size_t nbSeq_1 = nbSeq; - const U32 tableLog = FSE_optimalTableLog(FSELog, nbSeq, max); - if (count[codeTable[nbSeq-1]] > 1) { - count[codeTable[nbSeq-1]]--; - nbSeq_1--; - } - assert(nbSeq_1 > 1); - assert(entropyWorkspaceSize >= sizeof(ZSTD_BuildCTableWksp)); - (void)entropyWorkspaceSize; - FORWARD_IF_ERROR(FSE_normalizeCount(wksp->norm, tableLog, count, nbSeq_1, max, ZSTD_useLowProbCount(nbSeq_1)), "FSE_normalizeCount failed"); - assert(oend >= op); - { size_t const NCountSize = FSE_writeNCount(op, (size_t)(oend - op), wksp->norm, max, tableLog); /* overflow protected */ - FORWARD_IF_ERROR(NCountSize, "FSE_writeNCount failed"); - FORWARD_IF_ERROR(FSE_buildCTable_wksp(nextCTable, wksp->norm, max, tableLog, wksp->wksp, sizeof(wksp->wksp)), "FSE_buildCTable_wksp failed"); - return NCountSize; - } - } - default: assert(0); RETURN_ERROR(GENERIC, "impossible to reach"); - } -} - -FORCE_INLINE_TEMPLATE size_t -ZSTD_encodeSequences_body( - void* dst, size_t dstCapacity, - FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable, - FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable, - FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable, - seqDef const* sequences, size_t nbSeq, int longOffsets) -{ - BIT_CStream_t blockStream; - FSE_CState_t stateMatchLength; - FSE_CState_t stateOffsetBits; - FSE_CState_t stateLitLength; - - RETURN_ERROR_IF( - ERR_isError(BIT_initCStream(&blockStream, dst, dstCapacity)), - dstSize_tooSmall, "not enough space remaining"); - DEBUGLOG(6, "available space for bitstream : %i (dstCapacity=%u)", - (int)(blockStream.endPtr - blockStream.startPtr), - (unsigned)dstCapacity); - - /* first symbols */ - FSE_initCState2(&stateMatchLength, CTable_MatchLength, mlCodeTable[nbSeq-1]); - FSE_initCState2(&stateOffsetBits, CTable_OffsetBits, ofCodeTable[nbSeq-1]); - FSE_initCState2(&stateLitLength, CTable_LitLength, llCodeTable[nbSeq-1]); - BIT_addBits(&blockStream, sequences[nbSeq-1].litLength, LL_bits[llCodeTable[nbSeq-1]]); - if (MEM_32bits()) BIT_flushBits(&blockStream); - BIT_addBits(&blockStream, sequences[nbSeq-1].mlBase, ML_bits[mlCodeTable[nbSeq-1]]); - if (MEM_32bits()) BIT_flushBits(&blockStream); - if (longOffsets) { - U32 const ofBits = ofCodeTable[nbSeq-1]; - unsigned const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN-1); - if (extraBits) { - BIT_addBits(&blockStream, sequences[nbSeq-1].offBase, extraBits); - BIT_flushBits(&blockStream); - } - BIT_addBits(&blockStream, sequences[nbSeq-1].offBase >> extraBits, - ofBits - extraBits); - } else { - BIT_addBits(&blockStream, sequences[nbSeq-1].offBase, ofCodeTable[nbSeq-1]); - } - BIT_flushBits(&blockStream); - - { size_t n; - for (n=nbSeq-2 ; n= 64-7-(LLFSELog+MLFSELog+OffFSELog))) - BIT_flushBits(&blockStream); /* (7)*/ - BIT_addBits(&blockStream, sequences[n].litLength, llBits); - if (MEM_32bits() && ((llBits+mlBits)>24)) BIT_flushBits(&blockStream); - BIT_addBits(&blockStream, sequences[n].mlBase, mlBits); - if (MEM_32bits() || (ofBits+mlBits+llBits > 56)) BIT_flushBits(&blockStream); - if (longOffsets) { - unsigned const extraBits = ofBits - MIN(ofBits, STREAM_ACCUMULATOR_MIN-1); - if (extraBits) { - BIT_addBits(&blockStream, sequences[n].offBase, extraBits); - BIT_flushBits(&blockStream); /* (7)*/ - } - BIT_addBits(&blockStream, sequences[n].offBase >> extraBits, - ofBits - extraBits); /* 31 */ - } else { - BIT_addBits(&blockStream, sequences[n].offBase, ofBits); /* 31 */ - } - BIT_flushBits(&blockStream); /* (7)*/ - DEBUGLOG(7, "remaining space : %i", (int)(blockStream.endPtr - blockStream.ptr)); - } } - - DEBUGLOG(6, "ZSTD_encodeSequences: flushing ML state with %u bits", stateMatchLength.stateLog); - FSE_flushCState(&blockStream, &stateMatchLength); - DEBUGLOG(6, "ZSTD_encodeSequences: flushing Off state with %u bits", stateOffsetBits.stateLog); - FSE_flushCState(&blockStream, &stateOffsetBits); - DEBUGLOG(6, "ZSTD_encodeSequences: flushing LL state with %u bits", stateLitLength.stateLog); - FSE_flushCState(&blockStream, &stateLitLength); - - { size_t const streamSize = BIT_closeCStream(&blockStream); - RETURN_ERROR_IF(streamSize==0, dstSize_tooSmall, "not enough space"); - return streamSize; - } -} - -static size_t -ZSTD_encodeSequences_default( - void* dst, size_t dstCapacity, - FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable, - FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable, - FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable, - seqDef const* sequences, size_t nbSeq, int longOffsets) -{ - return ZSTD_encodeSequences_body(dst, dstCapacity, - CTable_MatchLength, mlCodeTable, - CTable_OffsetBits, ofCodeTable, - CTable_LitLength, llCodeTable, - sequences, nbSeq, longOffsets); -} - - -#if DYNAMIC_BMI2 - -static BMI2_TARGET_ATTRIBUTE size_t -ZSTD_encodeSequences_bmi2( - void* dst, size_t dstCapacity, - FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable, - FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable, - FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable, - seqDef const* sequences, size_t nbSeq, int longOffsets) -{ - return ZSTD_encodeSequences_body(dst, dstCapacity, - CTable_MatchLength, mlCodeTable, - CTable_OffsetBits, ofCodeTable, - CTable_LitLength, llCodeTable, - sequences, nbSeq, longOffsets); -} - -#endif - -size_t ZSTD_encodeSequences( - void* dst, size_t dstCapacity, - FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable, - FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable, - FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable, - seqDef const* sequences, size_t nbSeq, int longOffsets, int bmi2) -{ - DEBUGLOG(5, "ZSTD_encodeSequences: dstCapacity = %u", (unsigned)dstCapacity); -#if DYNAMIC_BMI2 - if (bmi2) { - return ZSTD_encodeSequences_bmi2(dst, dstCapacity, - CTable_MatchLength, mlCodeTable, - CTable_OffsetBits, ofCodeTable, - CTable_LitLength, llCodeTable, - sequences, nbSeq, longOffsets); - } -#endif - (void)bmi2; - return ZSTD_encodeSequences_default(dst, dstCapacity, - CTable_MatchLength, mlCodeTable, - CTable_OffsetBits, ofCodeTable, - CTable_LitLength, llCodeTable, - sequences, nbSeq, longOffsets); -} diff --git a/dep/zstd/lib/compress/zstd_compress_sequences.h b/dep/zstd/lib/compress/zstd_compress_sequences.h deleted file mode 100644 index 7991364c2..000000000 --- a/dep/zstd/lib/compress/zstd_compress_sequences.h +++ /dev/null @@ -1,54 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_COMPRESS_SEQUENCES_H -#define ZSTD_COMPRESS_SEQUENCES_H - -#include "../common/fse.h" /* FSE_repeat, FSE_CTable */ -#include "../common/zstd_internal.h" /* symbolEncodingType_e, ZSTD_strategy */ - -typedef enum { - ZSTD_defaultDisallowed = 0, - ZSTD_defaultAllowed = 1 -} ZSTD_defaultPolicy_e; - -symbolEncodingType_e -ZSTD_selectEncodingType( - FSE_repeat* repeatMode, unsigned const* count, unsigned const max, - size_t const mostFrequent, size_t nbSeq, unsigned const FSELog, - FSE_CTable const* prevCTable, - short const* defaultNorm, U32 defaultNormLog, - ZSTD_defaultPolicy_e const isDefaultAllowed, - ZSTD_strategy const strategy); - -size_t -ZSTD_buildCTable(void* dst, size_t dstCapacity, - FSE_CTable* nextCTable, U32 FSELog, symbolEncodingType_e type, - unsigned* count, U32 max, - const BYTE* codeTable, size_t nbSeq, - const S16* defaultNorm, U32 defaultNormLog, U32 defaultMax, - const FSE_CTable* prevCTable, size_t prevCTableSize, - void* entropyWorkspace, size_t entropyWorkspaceSize); - -size_t ZSTD_encodeSequences( - void* dst, size_t dstCapacity, - FSE_CTable const* CTable_MatchLength, BYTE const* mlCodeTable, - FSE_CTable const* CTable_OffsetBits, BYTE const* ofCodeTable, - FSE_CTable const* CTable_LitLength, BYTE const* llCodeTable, - seqDef const* sequences, size_t nbSeq, int longOffsets, int bmi2); - -size_t ZSTD_fseBitCost( - FSE_CTable const* ctable, - unsigned const* count, - unsigned const max); - -size_t ZSTD_crossEntropyCost(short const* norm, unsigned accuracyLog, - unsigned const* count, unsigned const max); -#endif /* ZSTD_COMPRESS_SEQUENCES_H */ diff --git a/dep/zstd/lib/compress/zstd_compress_superblock.c b/dep/zstd/lib/compress/zstd_compress_superblock.c deleted file mode 100644 index 10e337857..000000000 --- a/dep/zstd/lib/compress/zstd_compress_superblock.c +++ /dev/null @@ -1,573 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - /*-************************************* - * Dependencies - ***************************************/ -#include "zstd_compress_superblock.h" - -#include "../common/zstd_internal.h" /* ZSTD_getSequenceLength */ -#include "hist.h" /* HIST_countFast_wksp */ -#include "zstd_compress_internal.h" /* ZSTD_[huf|fse|entropy]CTablesMetadata_t */ -#include "zstd_compress_sequences.h" -#include "zstd_compress_literals.h" - -/** ZSTD_compressSubBlock_literal() : - * Compresses literals section for a sub-block. - * When we have to write the Huffman table we will sometimes choose a header - * size larger than necessary. This is because we have to pick the header size - * before we know the table size + compressed size, so we have a bound on the - * table size. If we guessed incorrectly, we fall back to uncompressed literals. - * - * We write the header when writeEntropy=1 and set entropyWritten=1 when we succeeded - * in writing the header, otherwise it is set to 0. - * - * hufMetadata->hType has literals block type info. - * If it is set_basic, all sub-blocks literals section will be Raw_Literals_Block. - * If it is set_rle, all sub-blocks literals section will be RLE_Literals_Block. - * If it is set_compressed, first sub-block's literals section will be Compressed_Literals_Block - * If it is set_compressed, first sub-block's literals section will be Treeless_Literals_Block - * and the following sub-blocks' literals sections will be Treeless_Literals_Block. - * @return : compressed size of literals section of a sub-block - * Or 0 if it unable to compress. - * Or error code */ -static size_t ZSTD_compressSubBlock_literal(const HUF_CElt* hufTable, - const ZSTD_hufCTablesMetadata_t* hufMetadata, - const BYTE* literals, size_t litSize, - void* dst, size_t dstSize, - const int bmi2, int writeEntropy, int* entropyWritten) -{ - size_t const header = writeEntropy ? 200 : 0; - size_t const lhSize = 3 + (litSize >= (1 KB - header)) + (litSize >= (16 KB - header)); - BYTE* const ostart = (BYTE*)dst; - BYTE* const oend = ostart + dstSize; - BYTE* op = ostart + lhSize; - U32 const singleStream = lhSize == 3; - symbolEncodingType_e hType = writeEntropy ? hufMetadata->hType : set_repeat; - size_t cLitSize = 0; - - (void)bmi2; /* TODO bmi2... */ - - DEBUGLOG(5, "ZSTD_compressSubBlock_literal (litSize=%zu, lhSize=%zu, writeEntropy=%d)", litSize, lhSize, writeEntropy); - - *entropyWritten = 0; - if (litSize == 0 || hufMetadata->hType == set_basic) { - DEBUGLOG(5, "ZSTD_compressSubBlock_literal using raw literal"); - return ZSTD_noCompressLiterals(dst, dstSize, literals, litSize); - } else if (hufMetadata->hType == set_rle) { - DEBUGLOG(5, "ZSTD_compressSubBlock_literal using rle literal"); - return ZSTD_compressRleLiteralsBlock(dst, dstSize, literals, litSize); - } - - assert(litSize > 0); - assert(hufMetadata->hType == set_compressed || hufMetadata->hType == set_repeat); - - if (writeEntropy && hufMetadata->hType == set_compressed) { - ZSTD_memcpy(op, hufMetadata->hufDesBuffer, hufMetadata->hufDesSize); - op += hufMetadata->hufDesSize; - cLitSize += hufMetadata->hufDesSize; - DEBUGLOG(5, "ZSTD_compressSubBlock_literal (hSize=%zu)", hufMetadata->hufDesSize); - } - - /* TODO bmi2 */ - { const size_t cSize = singleStream ? HUF_compress1X_usingCTable(op, oend-op, literals, litSize, hufTable) - : HUF_compress4X_usingCTable(op, oend-op, literals, litSize, hufTable); - op += cSize; - cLitSize += cSize; - if (cSize == 0 || ERR_isError(cSize)) { - DEBUGLOG(5, "Failed to write entropy tables %s", ZSTD_getErrorName(cSize)); - return 0; - } - /* If we expand and we aren't writing a header then emit uncompressed */ - if (!writeEntropy && cLitSize >= litSize) { - DEBUGLOG(5, "ZSTD_compressSubBlock_literal using raw literal because uncompressible"); - return ZSTD_noCompressLiterals(dst, dstSize, literals, litSize); - } - /* If we are writing headers then allow expansion that doesn't change our header size. */ - if (lhSize < (size_t)(3 + (cLitSize >= 1 KB) + (cLitSize >= 16 KB))) { - assert(cLitSize > litSize); - DEBUGLOG(5, "Literals expanded beyond allowed header size"); - return ZSTD_noCompressLiterals(dst, dstSize, literals, litSize); - } - DEBUGLOG(5, "ZSTD_compressSubBlock_literal (cSize=%zu)", cSize); - } - - /* Build header */ - switch(lhSize) - { - case 3: /* 2 - 2 - 10 - 10 */ - { U32 const lhc = hType + ((!singleStream) << 2) + ((U32)litSize<<4) + ((U32)cLitSize<<14); - MEM_writeLE24(ostart, lhc); - break; - } - case 4: /* 2 - 2 - 14 - 14 */ - { U32 const lhc = hType + (2 << 2) + ((U32)litSize<<4) + ((U32)cLitSize<<18); - MEM_writeLE32(ostart, lhc); - break; - } - case 5: /* 2 - 2 - 18 - 18 */ - { U32 const lhc = hType + (3 << 2) + ((U32)litSize<<4) + ((U32)cLitSize<<22); - MEM_writeLE32(ostart, lhc); - ostart[4] = (BYTE)(cLitSize >> 10); - break; - } - default: /* not possible : lhSize is {3,4,5} */ - assert(0); - } - *entropyWritten = 1; - DEBUGLOG(5, "Compressed literals: %u -> %u", (U32)litSize, (U32)(op-ostart)); - return op-ostart; -} - -static size_t ZSTD_seqDecompressedSize(seqStore_t const* seqStore, const seqDef* sequences, size_t nbSeq, size_t litSize, int lastSequence) { - const seqDef* const sstart = sequences; - const seqDef* const send = sequences + nbSeq; - const seqDef* sp = sstart; - size_t matchLengthSum = 0; - size_t litLengthSum = 0; - (void)(litLengthSum); /* suppress unused variable warning on some environments */ - while (send-sp > 0) { - ZSTD_sequenceLength const seqLen = ZSTD_getSequenceLength(seqStore, sp); - litLengthSum += seqLen.litLength; - matchLengthSum += seqLen.matchLength; - sp++; - } - assert(litLengthSum <= litSize); - if (!lastSequence) { - assert(litLengthSum == litSize); - } - return matchLengthSum + litSize; -} - -/** ZSTD_compressSubBlock_sequences() : - * Compresses sequences section for a sub-block. - * fseMetadata->llType, fseMetadata->ofType, and fseMetadata->mlType have - * symbol compression modes for the super-block. - * The first successfully compressed block will have these in its header. - * We set entropyWritten=1 when we succeed in compressing the sequences. - * The following sub-blocks will always have repeat mode. - * @return : compressed size of sequences section of a sub-block - * Or 0 if it is unable to compress - * Or error code. */ -static size_t ZSTD_compressSubBlock_sequences(const ZSTD_fseCTables_t* fseTables, - const ZSTD_fseCTablesMetadata_t* fseMetadata, - const seqDef* sequences, size_t nbSeq, - const BYTE* llCode, const BYTE* mlCode, const BYTE* ofCode, - const ZSTD_CCtx_params* cctxParams, - void* dst, size_t dstCapacity, - const int bmi2, int writeEntropy, int* entropyWritten) -{ - const int longOffsets = cctxParams->cParams.windowLog > STREAM_ACCUMULATOR_MIN; - BYTE* const ostart = (BYTE*)dst; - BYTE* const oend = ostart + dstCapacity; - BYTE* op = ostart; - BYTE* seqHead; - - DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (nbSeq=%zu, writeEntropy=%d, longOffsets=%d)", nbSeq, writeEntropy, longOffsets); - - *entropyWritten = 0; - /* Sequences Header */ - RETURN_ERROR_IF((oend-op) < 3 /*max nbSeq Size*/ + 1 /*seqHead*/, - dstSize_tooSmall, ""); - if (nbSeq < 0x7F) - *op++ = (BYTE)nbSeq; - else if (nbSeq < LONGNBSEQ) - op[0] = (BYTE)((nbSeq>>8) + 0x80), op[1] = (BYTE)nbSeq, op+=2; - else - op[0]=0xFF, MEM_writeLE16(op+1, (U16)(nbSeq - LONGNBSEQ)), op+=3; - if (nbSeq==0) { - return op - ostart; - } - - /* seqHead : flags for FSE encoding type */ - seqHead = op++; - - DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (seqHeadSize=%u)", (unsigned)(op-ostart)); - - if (writeEntropy) { - const U32 LLtype = fseMetadata->llType; - const U32 Offtype = fseMetadata->ofType; - const U32 MLtype = fseMetadata->mlType; - DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (fseTablesSize=%zu)", fseMetadata->fseTablesSize); - *seqHead = (BYTE)((LLtype<<6) + (Offtype<<4) + (MLtype<<2)); - ZSTD_memcpy(op, fseMetadata->fseTablesBuffer, fseMetadata->fseTablesSize); - op += fseMetadata->fseTablesSize; - } else { - const U32 repeat = set_repeat; - *seqHead = (BYTE)((repeat<<6) + (repeat<<4) + (repeat<<2)); - } - - { size_t const bitstreamSize = ZSTD_encodeSequences( - op, oend - op, - fseTables->matchlengthCTable, mlCode, - fseTables->offcodeCTable, ofCode, - fseTables->litlengthCTable, llCode, - sequences, nbSeq, - longOffsets, bmi2); - FORWARD_IF_ERROR(bitstreamSize, "ZSTD_encodeSequences failed"); - op += bitstreamSize; - /* zstd versions <= 1.3.4 mistakenly report corruption when - * FSE_readNCount() receives a buffer < 4 bytes. - * Fixed by https://github.com/facebook/zstd/pull/1146. - * This can happen when the last set_compressed table present is 2 - * bytes and the bitstream is only one byte. - * In this exceedingly rare case, we will simply emit an uncompressed - * block, since it isn't worth optimizing. - */ -#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION - if (writeEntropy && fseMetadata->lastCountSize && fseMetadata->lastCountSize + bitstreamSize < 4) { - /* NCountSize >= 2 && bitstreamSize > 0 ==> lastCountSize == 3 */ - assert(fseMetadata->lastCountSize + bitstreamSize == 3); - DEBUGLOG(5, "Avoiding bug in zstd decoder in versions <= 1.3.4 by " - "emitting an uncompressed block."); - return 0; - } -#endif - DEBUGLOG(5, "ZSTD_compressSubBlock_sequences (bitstreamSize=%zu)", bitstreamSize); - } - - /* zstd versions <= 1.4.0 mistakenly report error when - * sequences section body size is less than 3 bytes. - * Fixed by https://github.com/facebook/zstd/pull/1664. - * This can happen when the previous sequences section block is compressed - * with rle mode and the current block's sequences section is compressed - * with repeat mode where sequences section body size can be 1 byte. - */ -#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION - if (op-seqHead < 4) { - DEBUGLOG(5, "Avoiding bug in zstd decoder in versions <= 1.4.0 by emitting " - "an uncompressed block when sequences are < 4 bytes"); - return 0; - } -#endif - - *entropyWritten = 1; - return op - ostart; -} - -/** ZSTD_compressSubBlock() : - * Compresses a single sub-block. - * @return : compressed size of the sub-block - * Or 0 if it failed to compress. */ -static size_t ZSTD_compressSubBlock(const ZSTD_entropyCTables_t* entropy, - const ZSTD_entropyCTablesMetadata_t* entropyMetadata, - const seqDef* sequences, size_t nbSeq, - const BYTE* literals, size_t litSize, - const BYTE* llCode, const BYTE* mlCode, const BYTE* ofCode, - const ZSTD_CCtx_params* cctxParams, - void* dst, size_t dstCapacity, - const int bmi2, - int writeLitEntropy, int writeSeqEntropy, - int* litEntropyWritten, int* seqEntropyWritten, - U32 lastBlock) -{ - BYTE* const ostart = (BYTE*)dst; - BYTE* const oend = ostart + dstCapacity; - BYTE* op = ostart + ZSTD_blockHeaderSize; - DEBUGLOG(5, "ZSTD_compressSubBlock (litSize=%zu, nbSeq=%zu, writeLitEntropy=%d, writeSeqEntropy=%d, lastBlock=%d)", - litSize, nbSeq, writeLitEntropy, writeSeqEntropy, lastBlock); - { size_t cLitSize = ZSTD_compressSubBlock_literal((const HUF_CElt*)entropy->huf.CTable, - &entropyMetadata->hufMetadata, literals, litSize, - op, oend-op, bmi2, writeLitEntropy, litEntropyWritten); - FORWARD_IF_ERROR(cLitSize, "ZSTD_compressSubBlock_literal failed"); - if (cLitSize == 0) return 0; - op += cLitSize; - } - { size_t cSeqSize = ZSTD_compressSubBlock_sequences(&entropy->fse, - &entropyMetadata->fseMetadata, - sequences, nbSeq, - llCode, mlCode, ofCode, - cctxParams, - op, oend-op, - bmi2, writeSeqEntropy, seqEntropyWritten); - FORWARD_IF_ERROR(cSeqSize, "ZSTD_compressSubBlock_sequences failed"); - if (cSeqSize == 0) return 0; - op += cSeqSize; - } - /* Write block header */ - { size_t cSize = (op-ostart)-ZSTD_blockHeaderSize; - U32 const cBlockHeader24 = lastBlock + (((U32)bt_compressed)<<1) + (U32)(cSize << 3); - MEM_writeLE24(ostart, cBlockHeader24); - } - return op-ostart; -} - -static size_t ZSTD_estimateSubBlockSize_literal(const BYTE* literals, size_t litSize, - const ZSTD_hufCTables_t* huf, - const ZSTD_hufCTablesMetadata_t* hufMetadata, - void* workspace, size_t wkspSize, - int writeEntropy) -{ - unsigned* const countWksp = (unsigned*)workspace; - unsigned maxSymbolValue = 255; - size_t literalSectionHeaderSize = 3; /* Use hard coded size of 3 bytes */ - - if (hufMetadata->hType == set_basic) return litSize; - else if (hufMetadata->hType == set_rle) return 1; - else if (hufMetadata->hType == set_compressed || hufMetadata->hType == set_repeat) { - size_t const largest = HIST_count_wksp (countWksp, &maxSymbolValue, (const BYTE*)literals, litSize, workspace, wkspSize); - if (ZSTD_isError(largest)) return litSize; - { size_t cLitSizeEstimate = HUF_estimateCompressedSize((const HUF_CElt*)huf->CTable, countWksp, maxSymbolValue); - if (writeEntropy) cLitSizeEstimate += hufMetadata->hufDesSize; - return cLitSizeEstimate + literalSectionHeaderSize; - } } - assert(0); /* impossible */ - return 0; -} - -static size_t ZSTD_estimateSubBlockSize_symbolType(symbolEncodingType_e type, - const BYTE* codeTable, unsigned maxCode, - size_t nbSeq, const FSE_CTable* fseCTable, - const U8* additionalBits, - short const* defaultNorm, U32 defaultNormLog, U32 defaultMax, - void* workspace, size_t wkspSize) -{ - unsigned* const countWksp = (unsigned*)workspace; - const BYTE* ctp = codeTable; - const BYTE* const ctStart = ctp; - const BYTE* const ctEnd = ctStart + nbSeq; - size_t cSymbolTypeSizeEstimateInBits = 0; - unsigned max = maxCode; - - HIST_countFast_wksp(countWksp, &max, codeTable, nbSeq, workspace, wkspSize); /* can't fail */ - if (type == set_basic) { - /* We selected this encoding type, so it must be valid. */ - assert(max <= defaultMax); - cSymbolTypeSizeEstimateInBits = max <= defaultMax - ? ZSTD_crossEntropyCost(defaultNorm, defaultNormLog, countWksp, max) - : ERROR(GENERIC); - } else if (type == set_rle) { - cSymbolTypeSizeEstimateInBits = 0; - } else if (type == set_compressed || type == set_repeat) { - cSymbolTypeSizeEstimateInBits = ZSTD_fseBitCost(fseCTable, countWksp, max); - } - if (ZSTD_isError(cSymbolTypeSizeEstimateInBits)) return nbSeq * 10; - while (ctp < ctEnd) { - if (additionalBits) cSymbolTypeSizeEstimateInBits += additionalBits[*ctp]; - else cSymbolTypeSizeEstimateInBits += *ctp; /* for offset, offset code is also the number of additional bits */ - ctp++; - } - return cSymbolTypeSizeEstimateInBits / 8; -} - -static size_t ZSTD_estimateSubBlockSize_sequences(const BYTE* ofCodeTable, - const BYTE* llCodeTable, - const BYTE* mlCodeTable, - size_t nbSeq, - const ZSTD_fseCTables_t* fseTables, - const ZSTD_fseCTablesMetadata_t* fseMetadata, - void* workspace, size_t wkspSize, - int writeEntropy) -{ - size_t const sequencesSectionHeaderSize = 3; /* Use hard coded size of 3 bytes */ - size_t cSeqSizeEstimate = 0; - if (nbSeq == 0) return sequencesSectionHeaderSize; - cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->ofType, ofCodeTable, MaxOff, - nbSeq, fseTables->offcodeCTable, NULL, - OF_defaultNorm, OF_defaultNormLog, DefaultMaxOff, - workspace, wkspSize); - cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->llType, llCodeTable, MaxLL, - nbSeq, fseTables->litlengthCTable, LL_bits, - LL_defaultNorm, LL_defaultNormLog, MaxLL, - workspace, wkspSize); - cSeqSizeEstimate += ZSTD_estimateSubBlockSize_symbolType(fseMetadata->mlType, mlCodeTable, MaxML, - nbSeq, fseTables->matchlengthCTable, ML_bits, - ML_defaultNorm, ML_defaultNormLog, MaxML, - workspace, wkspSize); - if (writeEntropy) cSeqSizeEstimate += fseMetadata->fseTablesSize; - return cSeqSizeEstimate + sequencesSectionHeaderSize; -} - -static size_t ZSTD_estimateSubBlockSize(const BYTE* literals, size_t litSize, - const BYTE* ofCodeTable, - const BYTE* llCodeTable, - const BYTE* mlCodeTable, - size_t nbSeq, - const ZSTD_entropyCTables_t* entropy, - const ZSTD_entropyCTablesMetadata_t* entropyMetadata, - void* workspace, size_t wkspSize, - int writeLitEntropy, int writeSeqEntropy) { - size_t cSizeEstimate = 0; - cSizeEstimate += ZSTD_estimateSubBlockSize_literal(literals, litSize, - &entropy->huf, &entropyMetadata->hufMetadata, - workspace, wkspSize, writeLitEntropy); - cSizeEstimate += ZSTD_estimateSubBlockSize_sequences(ofCodeTable, llCodeTable, mlCodeTable, - nbSeq, &entropy->fse, &entropyMetadata->fseMetadata, - workspace, wkspSize, writeSeqEntropy); - return cSizeEstimate + ZSTD_blockHeaderSize; -} - -static int ZSTD_needSequenceEntropyTables(ZSTD_fseCTablesMetadata_t const* fseMetadata) -{ - if (fseMetadata->llType == set_compressed || fseMetadata->llType == set_rle) - return 1; - if (fseMetadata->mlType == set_compressed || fseMetadata->mlType == set_rle) - return 1; - if (fseMetadata->ofType == set_compressed || fseMetadata->ofType == set_rle) - return 1; - return 0; -} - -/** ZSTD_compressSubBlock_multi() : - * Breaks super-block into multiple sub-blocks and compresses them. - * Entropy will be written to the first block. - * The following blocks will use repeat mode to compress. - * All sub-blocks are compressed blocks (no raw or rle blocks). - * @return : compressed size of the super block (which is multiple ZSTD blocks) - * Or 0 if it failed to compress. */ -static size_t ZSTD_compressSubBlock_multi(const seqStore_t* seqStorePtr, - const ZSTD_compressedBlockState_t* prevCBlock, - ZSTD_compressedBlockState_t* nextCBlock, - const ZSTD_entropyCTablesMetadata_t* entropyMetadata, - const ZSTD_CCtx_params* cctxParams, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const int bmi2, U32 lastBlock, - void* workspace, size_t wkspSize) -{ - const seqDef* const sstart = seqStorePtr->sequencesStart; - const seqDef* const send = seqStorePtr->sequences; - const seqDef* sp = sstart; - const BYTE* const lstart = seqStorePtr->litStart; - const BYTE* const lend = seqStorePtr->lit; - const BYTE* lp = lstart; - BYTE const* ip = (BYTE const*)src; - BYTE const* const iend = ip + srcSize; - BYTE* const ostart = (BYTE*)dst; - BYTE* const oend = ostart + dstCapacity; - BYTE* op = ostart; - const BYTE* llCodePtr = seqStorePtr->llCode; - const BYTE* mlCodePtr = seqStorePtr->mlCode; - const BYTE* ofCodePtr = seqStorePtr->ofCode; - size_t targetCBlockSize = cctxParams->targetCBlockSize; - size_t litSize, seqCount; - int writeLitEntropy = entropyMetadata->hufMetadata.hType == set_compressed; - int writeSeqEntropy = 1; - int lastSequence = 0; - - DEBUGLOG(5, "ZSTD_compressSubBlock_multi (litSize=%u, nbSeq=%u)", - (unsigned)(lend-lp), (unsigned)(send-sstart)); - - litSize = 0; - seqCount = 0; - do { - size_t cBlockSizeEstimate = 0; - if (sstart == send) { - lastSequence = 1; - } else { - const seqDef* const sequence = sp + seqCount; - lastSequence = sequence == send - 1; - litSize += ZSTD_getSequenceLength(seqStorePtr, sequence).litLength; - seqCount++; - } - if (lastSequence) { - assert(lp <= lend); - assert(litSize <= (size_t)(lend - lp)); - litSize = (size_t)(lend - lp); - } - /* I think there is an optimization opportunity here. - * Calling ZSTD_estimateSubBlockSize for every sequence can be wasteful - * since it recalculates estimate from scratch. - * For example, it would recount literal distribution and symbol codes every time. - */ - cBlockSizeEstimate = ZSTD_estimateSubBlockSize(lp, litSize, ofCodePtr, llCodePtr, mlCodePtr, seqCount, - &nextCBlock->entropy, entropyMetadata, - workspace, wkspSize, writeLitEntropy, writeSeqEntropy); - if (cBlockSizeEstimate > targetCBlockSize || lastSequence) { - int litEntropyWritten = 0; - int seqEntropyWritten = 0; - const size_t decompressedSize = ZSTD_seqDecompressedSize(seqStorePtr, sp, seqCount, litSize, lastSequence); - const size_t cSize = ZSTD_compressSubBlock(&nextCBlock->entropy, entropyMetadata, - sp, seqCount, - lp, litSize, - llCodePtr, mlCodePtr, ofCodePtr, - cctxParams, - op, oend-op, - bmi2, writeLitEntropy, writeSeqEntropy, - &litEntropyWritten, &seqEntropyWritten, - lastBlock && lastSequence); - FORWARD_IF_ERROR(cSize, "ZSTD_compressSubBlock failed"); - if (cSize > 0 && cSize < decompressedSize) { - DEBUGLOG(5, "Committed the sub-block"); - assert(ip + decompressedSize <= iend); - ip += decompressedSize; - sp += seqCount; - lp += litSize; - op += cSize; - llCodePtr += seqCount; - mlCodePtr += seqCount; - ofCodePtr += seqCount; - litSize = 0; - seqCount = 0; - /* Entropy only needs to be written once */ - if (litEntropyWritten) { - writeLitEntropy = 0; - } - if (seqEntropyWritten) { - writeSeqEntropy = 0; - } - } - } - } while (!lastSequence); - if (writeLitEntropy) { - DEBUGLOG(5, "ZSTD_compressSubBlock_multi has literal entropy tables unwritten"); - ZSTD_memcpy(&nextCBlock->entropy.huf, &prevCBlock->entropy.huf, sizeof(prevCBlock->entropy.huf)); - } - if (writeSeqEntropy && ZSTD_needSequenceEntropyTables(&entropyMetadata->fseMetadata)) { - /* If we haven't written our entropy tables, then we've violated our contract and - * must emit an uncompressed block. - */ - DEBUGLOG(5, "ZSTD_compressSubBlock_multi has sequence entropy tables unwritten"); - return 0; - } - if (ip < iend) { - size_t const cSize = ZSTD_noCompressBlock(op, oend - op, ip, iend - ip, lastBlock); - DEBUGLOG(5, "ZSTD_compressSubBlock_multi last sub-block uncompressed, %zu bytes", (size_t)(iend - ip)); - FORWARD_IF_ERROR(cSize, "ZSTD_noCompressBlock failed"); - assert(cSize != 0); - op += cSize; - /* We have to regenerate the repcodes because we've skipped some sequences */ - if (sp < send) { - seqDef const* seq; - repcodes_t rep; - ZSTD_memcpy(&rep, prevCBlock->rep, sizeof(rep)); - for (seq = sstart; seq < sp; ++seq) { - ZSTD_updateRep(rep.rep, seq->offBase - 1, ZSTD_getSequenceLength(seqStorePtr, seq).litLength == 0); - } - ZSTD_memcpy(nextCBlock->rep, &rep, sizeof(rep)); - } - } - DEBUGLOG(5, "ZSTD_compressSubBlock_multi compressed"); - return op-ostart; -} - -size_t ZSTD_compressSuperBlock(ZSTD_CCtx* zc, - void* dst, size_t dstCapacity, - void const* src, size_t srcSize, - unsigned lastBlock) { - ZSTD_entropyCTablesMetadata_t entropyMetadata; - - FORWARD_IF_ERROR(ZSTD_buildBlockEntropyStats(&zc->seqStore, - &zc->blockState.prevCBlock->entropy, - &zc->blockState.nextCBlock->entropy, - &zc->appliedParams, - &entropyMetadata, - zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */), ""); - - return ZSTD_compressSubBlock_multi(&zc->seqStore, - zc->blockState.prevCBlock, - zc->blockState.nextCBlock, - &entropyMetadata, - &zc->appliedParams, - dst, dstCapacity, - src, srcSize, - zc->bmi2, lastBlock, - zc->entropyWorkspace, ENTROPY_WORKSPACE_SIZE /* statically allocated in resetCCtx */); -} diff --git a/dep/zstd/lib/compress/zstd_compress_superblock.h b/dep/zstd/lib/compress/zstd_compress_superblock.h deleted file mode 100644 index 176f9b106..000000000 --- a/dep/zstd/lib/compress/zstd_compress_superblock.h +++ /dev/null @@ -1,32 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_COMPRESS_ADVANCED_H -#define ZSTD_COMPRESS_ADVANCED_H - -/*-************************************* -* Dependencies -***************************************/ - -#include "../zstd.h" /* ZSTD_CCtx */ - -/*-************************************* -* Target Compressed Block Size -***************************************/ - -/* ZSTD_compressSuperBlock() : - * Used to compress a super block when targetCBlockSize is being used. - * The given block will be compressed into multiple sub blocks that are around targetCBlockSize. */ -size_t ZSTD_compressSuperBlock(ZSTD_CCtx* zc, - void* dst, size_t dstCapacity, - void const* src, size_t srcSize, - unsigned lastBlock); - -#endif /* ZSTD_COMPRESS_ADVANCED_H */ diff --git a/dep/zstd/lib/compress/zstd_cwksp.h b/dep/zstd/lib/compress/zstd_cwksp.h deleted file mode 100644 index dc3f40c80..000000000 --- a/dep/zstd/lib/compress/zstd_cwksp.h +++ /dev/null @@ -1,676 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_CWKSP_H -#define ZSTD_CWKSP_H - -/*-************************************* -* Dependencies -***************************************/ -#include "../common/zstd_internal.h" - -#if defined (__cplusplus) -extern "C" { -#endif - -/*-************************************* -* Constants -***************************************/ - -/* Since the workspace is effectively its own little malloc implementation / - * arena, when we run under ASAN, we should similarly insert redzones between - * each internal element of the workspace, so ASAN will catch overruns that - * reach outside an object but that stay inside the workspace. - * - * This defines the size of that redzone. - */ -#ifndef ZSTD_CWKSP_ASAN_REDZONE_SIZE -#define ZSTD_CWKSP_ASAN_REDZONE_SIZE 128 -#endif - - -/* Set our tables and aligneds to align by 64 bytes */ -#define ZSTD_CWKSP_ALIGNMENT_BYTES 64 - -/*-************************************* -* Structures -***************************************/ -typedef enum { - ZSTD_cwksp_alloc_objects, - ZSTD_cwksp_alloc_buffers, - ZSTD_cwksp_alloc_aligned -} ZSTD_cwksp_alloc_phase_e; - -/** - * Used to describe whether the workspace is statically allocated (and will not - * necessarily ever be freed), or if it's dynamically allocated and we can - * expect a well-formed caller to free this. - */ -typedef enum { - ZSTD_cwksp_dynamic_alloc, - ZSTD_cwksp_static_alloc -} ZSTD_cwksp_static_alloc_e; - -/** - * Zstd fits all its internal datastructures into a single continuous buffer, - * so that it only needs to perform a single OS allocation (or so that a buffer - * can be provided to it and it can perform no allocations at all). This buffer - * is called the workspace. - * - * Several optimizations complicate that process of allocating memory ranges - * from this workspace for each internal datastructure: - * - * - These different internal datastructures have different setup requirements: - * - * - The static objects need to be cleared once and can then be trivially - * reused for each compression. - * - * - Various buffers don't need to be initialized at all--they are always - * written into before they're read. - * - * - The matchstate tables have a unique requirement that they don't need - * their memory to be totally cleared, but they do need the memory to have - * some bound, i.e., a guarantee that all values in the memory they've been - * allocated is less than some maximum value (which is the starting value - * for the indices that they will then use for compression). When this - * guarantee is provided to them, they can use the memory without any setup - * work. When it can't, they have to clear the area. - * - * - These buffers also have different alignment requirements. - * - * - We would like to reuse the objects in the workspace for multiple - * compressions without having to perform any expensive reallocation or - * reinitialization work. - * - * - We would like to be able to efficiently reuse the workspace across - * multiple compressions **even when the compression parameters change** and - * we need to resize some of the objects (where possible). - * - * To attempt to manage this buffer, given these constraints, the ZSTD_cwksp - * abstraction was created. It works as follows: - * - * Workspace Layout: - * - * [ ... workspace ... ] - * [objects][tables ... ->] free space [<- ... aligned][<- ... buffers] - * - * The various objects that live in the workspace are divided into the - * following categories, and are allocated separately: - * - * - Static objects: this is optionally the enclosing ZSTD_CCtx or ZSTD_CDict, - * so that literally everything fits in a single buffer. Note: if present, - * this must be the first object in the workspace, since ZSTD_customFree{CCtx, - * CDict}() rely on a pointer comparison to see whether one or two frees are - * required. - * - * - Fixed size objects: these are fixed-size, fixed-count objects that are - * nonetheless "dynamically" allocated in the workspace so that we can - * control how they're initialized separately from the broader ZSTD_CCtx. - * Examples: - * - Entropy Workspace - * - 2 x ZSTD_compressedBlockState_t - * - CDict dictionary contents - * - * - Tables: these are any of several different datastructures (hash tables, - * chain tables, binary trees) that all respect a common format: they are - * uint32_t arrays, all of whose values are between 0 and (nextSrc - base). - * Their sizes depend on the cparams. These tables are 64-byte aligned. - * - * - Aligned: these buffers are used for various purposes that require 4 byte - * alignment, but don't require any initialization before they're used. These - * buffers are each aligned to 64 bytes. - * - * - Buffers: these buffers are used for various purposes that don't require - * any alignment or initialization before they're used. This means they can - * be moved around at no cost for a new compression. - * - * Allocating Memory: - * - * The various types of objects must be allocated in order, so they can be - * correctly packed into the workspace buffer. That order is: - * - * 1. Objects - * 2. Buffers - * 3. Aligned/Tables - * - * Attempts to reserve objects of different types out of order will fail. - */ -typedef struct { - void* workspace; - void* workspaceEnd; - - void* objectEnd; - void* tableEnd; - void* tableValidEnd; - void* allocStart; - - BYTE allocFailed; - int workspaceOversizedDuration; - ZSTD_cwksp_alloc_phase_e phase; - ZSTD_cwksp_static_alloc_e isStatic; -} ZSTD_cwksp; - -/*-************************************* -* Functions -***************************************/ - -MEM_STATIC size_t ZSTD_cwksp_available_space(ZSTD_cwksp* ws); - -MEM_STATIC void ZSTD_cwksp_assert_internal_consistency(ZSTD_cwksp* ws) { - (void)ws; - assert(ws->workspace <= ws->objectEnd); - assert(ws->objectEnd <= ws->tableEnd); - assert(ws->objectEnd <= ws->tableValidEnd); - assert(ws->tableEnd <= ws->allocStart); - assert(ws->tableValidEnd <= ws->allocStart); - assert(ws->allocStart <= ws->workspaceEnd); -} - -/** - * Align must be a power of 2. - */ -MEM_STATIC size_t ZSTD_cwksp_align(size_t size, size_t const align) { - size_t const mask = align - 1; - assert((align & mask) == 0); - return (size + mask) & ~mask; -} - -/** - * Use this to determine how much space in the workspace we will consume to - * allocate this object. (Normally it should be exactly the size of the object, - * but under special conditions, like ASAN, where we pad each object, it might - * be larger.) - * - * Since tables aren't currently redzoned, you don't need to call through this - * to figure out how much space you need for the matchState tables. Everything - * else is though. - * - * Do not use for sizing aligned buffers. Instead, use ZSTD_cwksp_aligned_alloc_size(). - */ -MEM_STATIC size_t ZSTD_cwksp_alloc_size(size_t size) { - if (size == 0) - return 0; -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) - return size + 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE; -#else - return size; -#endif -} - -/** - * Returns an adjusted alloc size that is the nearest larger multiple of 64 bytes. - * Used to determine the number of bytes required for a given "aligned". - */ -MEM_STATIC size_t ZSTD_cwksp_aligned_alloc_size(size_t size) { - return ZSTD_cwksp_alloc_size(ZSTD_cwksp_align(size, ZSTD_CWKSP_ALIGNMENT_BYTES)); -} - -/** - * Returns the amount of additional space the cwksp must allocate - * for internal purposes (currently only alignment). - */ -MEM_STATIC size_t ZSTD_cwksp_slack_space_required(void) { - /* For alignment, the wksp will always allocate an additional n_1=[1, 64] bytes - * to align the beginning of tables section, as well as another n_2=[0, 63] bytes - * to align the beginning of the aligned section. - * - * n_1 + n_2 == 64 bytes if the cwksp is freshly allocated, due to tables and - * aligneds being sized in multiples of 64 bytes. - */ - size_t const slackSpace = ZSTD_CWKSP_ALIGNMENT_BYTES; - return slackSpace; -} - - -/** - * Return the number of additional bytes required to align a pointer to the given number of bytes. - * alignBytes must be a power of two. - */ -MEM_STATIC size_t ZSTD_cwksp_bytes_to_align_ptr(void* ptr, const size_t alignBytes) { - size_t const alignBytesMask = alignBytes - 1; - size_t const bytes = (alignBytes - ((size_t)ptr & (alignBytesMask))) & alignBytesMask; - assert((alignBytes & alignBytesMask) == 0); - assert(bytes != ZSTD_CWKSP_ALIGNMENT_BYTES); - return bytes; -} - -/** - * Internal function. Do not use directly. - * Reserves the given number of bytes within the aligned/buffer segment of the wksp, - * which counts from the end of the wksp (as opposed to the object/table segment). - * - * Returns a pointer to the beginning of that space. - */ -MEM_STATIC void* -ZSTD_cwksp_reserve_internal_buffer_space(ZSTD_cwksp* ws, size_t const bytes) -{ - void* const alloc = (BYTE*)ws->allocStart - bytes; - void* const bottom = ws->tableEnd; - DEBUGLOG(5, "cwksp: reserving %p %zd bytes, %zd bytes remaining", - alloc, bytes, ZSTD_cwksp_available_space(ws) - bytes); - ZSTD_cwksp_assert_internal_consistency(ws); - assert(alloc >= bottom); - if (alloc < bottom) { - DEBUGLOG(4, "cwksp: alloc failed!"); - ws->allocFailed = 1; - return NULL; - } - /* the area is reserved from the end of wksp. - * If it overlaps with tableValidEnd, it voids guarantees on values' range */ - if (alloc < ws->tableValidEnd) { - ws->tableValidEnd = alloc; - } - ws->allocStart = alloc; - return alloc; -} - -/** - * Moves the cwksp to the next phase, and does any necessary allocations. - * cwksp initialization must necessarily go through each phase in order. - * Returns a 0 on success, or zstd error - */ -MEM_STATIC size_t -ZSTD_cwksp_internal_advance_phase(ZSTD_cwksp* ws, ZSTD_cwksp_alloc_phase_e phase) -{ - assert(phase >= ws->phase); - if (phase > ws->phase) { - /* Going from allocating objects to allocating buffers */ - if (ws->phase < ZSTD_cwksp_alloc_buffers && - phase >= ZSTD_cwksp_alloc_buffers) { - ws->tableValidEnd = ws->objectEnd; - } - - /* Going from allocating buffers to allocating aligneds/tables */ - if (ws->phase < ZSTD_cwksp_alloc_aligned && - phase >= ZSTD_cwksp_alloc_aligned) { - { /* Align the start of the "aligned" to 64 bytes. Use [1, 64] bytes. */ - size_t const bytesToAlign = - ZSTD_CWKSP_ALIGNMENT_BYTES - ZSTD_cwksp_bytes_to_align_ptr(ws->allocStart, ZSTD_CWKSP_ALIGNMENT_BYTES); - DEBUGLOG(5, "reserving aligned alignment addtl space: %zu", bytesToAlign); - ZSTD_STATIC_ASSERT((ZSTD_CWKSP_ALIGNMENT_BYTES & (ZSTD_CWKSP_ALIGNMENT_BYTES - 1)) == 0); /* power of 2 */ - RETURN_ERROR_IF(!ZSTD_cwksp_reserve_internal_buffer_space(ws, bytesToAlign), - memory_allocation, "aligned phase - alignment initial allocation failed!"); - } - { /* Align the start of the tables to 64 bytes. Use [0, 63] bytes */ - void* const alloc = ws->objectEnd; - size_t const bytesToAlign = ZSTD_cwksp_bytes_to_align_ptr(alloc, ZSTD_CWKSP_ALIGNMENT_BYTES); - void* const objectEnd = (BYTE*)alloc + bytesToAlign; - DEBUGLOG(5, "reserving table alignment addtl space: %zu", bytesToAlign); - RETURN_ERROR_IF(objectEnd > ws->workspaceEnd, memory_allocation, - "table phase - alignment initial allocation failed!"); - ws->objectEnd = objectEnd; - ws->tableEnd = objectEnd; /* table area starts being empty */ - if (ws->tableValidEnd < ws->tableEnd) { - ws->tableValidEnd = ws->tableEnd; - } } } - ws->phase = phase; - ZSTD_cwksp_assert_internal_consistency(ws); - } - return 0; -} - -/** - * Returns whether this object/buffer/etc was allocated in this workspace. - */ -MEM_STATIC int ZSTD_cwksp_owns_buffer(const ZSTD_cwksp* ws, const void* ptr) -{ - return (ptr != NULL) && (ws->workspace <= ptr) && (ptr <= ws->workspaceEnd); -} - -/** - * Internal function. Do not use directly. - */ -MEM_STATIC void* -ZSTD_cwksp_reserve_internal(ZSTD_cwksp* ws, size_t bytes, ZSTD_cwksp_alloc_phase_e phase) -{ - void* alloc; - if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase)) || bytes == 0) { - return NULL; - } - -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) - /* over-reserve space */ - bytes += 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE; -#endif - - alloc = ZSTD_cwksp_reserve_internal_buffer_space(ws, bytes); - -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) - /* Move alloc so there's ZSTD_CWKSP_ASAN_REDZONE_SIZE unused space on - * either size. */ - if (alloc) { - alloc = (BYTE *)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE; - if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { - __asan_unpoison_memory_region(alloc, bytes); - } - } -#endif - - return alloc; -} - -/** - * Reserves and returns unaligned memory. - */ -MEM_STATIC BYTE* ZSTD_cwksp_reserve_buffer(ZSTD_cwksp* ws, size_t bytes) -{ - return (BYTE*)ZSTD_cwksp_reserve_internal(ws, bytes, ZSTD_cwksp_alloc_buffers); -} - -/** - * Reserves and returns memory sized on and aligned on ZSTD_CWKSP_ALIGNMENT_BYTES (64 bytes). - */ -MEM_STATIC void* ZSTD_cwksp_reserve_aligned(ZSTD_cwksp* ws, size_t bytes) -{ - void* ptr = ZSTD_cwksp_reserve_internal(ws, ZSTD_cwksp_align(bytes, ZSTD_CWKSP_ALIGNMENT_BYTES), - ZSTD_cwksp_alloc_aligned); - assert(((size_t)ptr & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0); - return ptr; -} - -/** - * Aligned on 64 bytes. These buffers have the special property that - * their values remain constrained, allowing us to re-use them without - * memset()-ing them. - */ -MEM_STATIC void* ZSTD_cwksp_reserve_table(ZSTD_cwksp* ws, size_t bytes) -{ - const ZSTD_cwksp_alloc_phase_e phase = ZSTD_cwksp_alloc_aligned; - void* alloc; - void* end; - void* top; - - if (ZSTD_isError(ZSTD_cwksp_internal_advance_phase(ws, phase))) { - return NULL; - } - alloc = ws->tableEnd; - end = (BYTE *)alloc + bytes; - top = ws->allocStart; - - DEBUGLOG(5, "cwksp: reserving %p table %zd bytes, %zd bytes remaining", - alloc, bytes, ZSTD_cwksp_available_space(ws) - bytes); - assert((bytes & (sizeof(U32)-1)) == 0); - ZSTD_cwksp_assert_internal_consistency(ws); - assert(end <= top); - if (end > top) { - DEBUGLOG(4, "cwksp: table alloc failed!"); - ws->allocFailed = 1; - return NULL; - } - ws->tableEnd = end; - -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) - if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { - __asan_unpoison_memory_region(alloc, bytes); - } -#endif - - assert((bytes & (ZSTD_CWKSP_ALIGNMENT_BYTES-1)) == 0); - assert(((size_t)alloc & (ZSTD_CWKSP_ALIGNMENT_BYTES-1))== 0); - return alloc; -} - -/** - * Aligned on sizeof(void*). - * Note : should happen only once, at workspace first initialization - */ -MEM_STATIC void* ZSTD_cwksp_reserve_object(ZSTD_cwksp* ws, size_t bytes) -{ - size_t const roundedBytes = ZSTD_cwksp_align(bytes, sizeof(void*)); - void* alloc = ws->objectEnd; - void* end = (BYTE*)alloc + roundedBytes; - -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) - /* over-reserve space */ - end = (BYTE *)end + 2 * ZSTD_CWKSP_ASAN_REDZONE_SIZE; -#endif - - DEBUGLOG(4, - "cwksp: reserving %p object %zd bytes (rounded to %zd), %zd bytes remaining", - alloc, bytes, roundedBytes, ZSTD_cwksp_available_space(ws) - roundedBytes); - assert((size_t)alloc % ZSTD_ALIGNOF(void*) == 0); - assert(bytes % ZSTD_ALIGNOF(void*) == 0); - ZSTD_cwksp_assert_internal_consistency(ws); - /* we must be in the first phase, no advance is possible */ - if (ws->phase != ZSTD_cwksp_alloc_objects || end > ws->workspaceEnd) { - DEBUGLOG(3, "cwksp: object alloc failed!"); - ws->allocFailed = 1; - return NULL; - } - ws->objectEnd = end; - ws->tableEnd = end; - ws->tableValidEnd = end; - -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) - /* Move alloc so there's ZSTD_CWKSP_ASAN_REDZONE_SIZE unused space on - * either size. */ - alloc = (BYTE*)alloc + ZSTD_CWKSP_ASAN_REDZONE_SIZE; - if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { - __asan_unpoison_memory_region(alloc, bytes); - } -#endif - - return alloc; -} - -MEM_STATIC void ZSTD_cwksp_mark_tables_dirty(ZSTD_cwksp* ws) -{ - DEBUGLOG(4, "cwksp: ZSTD_cwksp_mark_tables_dirty"); - -#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE) - /* To validate that the table re-use logic is sound, and that we don't - * access table space that we haven't cleaned, we re-"poison" the table - * space every time we mark it dirty. */ - { - size_t size = (BYTE*)ws->tableValidEnd - (BYTE*)ws->objectEnd; - assert(__msan_test_shadow(ws->objectEnd, size) == -1); - __msan_poison(ws->objectEnd, size); - } -#endif - - assert(ws->tableValidEnd >= ws->objectEnd); - assert(ws->tableValidEnd <= ws->allocStart); - ws->tableValidEnd = ws->objectEnd; - ZSTD_cwksp_assert_internal_consistency(ws); -} - -MEM_STATIC void ZSTD_cwksp_mark_tables_clean(ZSTD_cwksp* ws) { - DEBUGLOG(4, "cwksp: ZSTD_cwksp_mark_tables_clean"); - assert(ws->tableValidEnd >= ws->objectEnd); - assert(ws->tableValidEnd <= ws->allocStart); - if (ws->tableValidEnd < ws->tableEnd) { - ws->tableValidEnd = ws->tableEnd; - } - ZSTD_cwksp_assert_internal_consistency(ws); -} - -/** - * Zero the part of the allocated tables not already marked clean. - */ -MEM_STATIC void ZSTD_cwksp_clean_tables(ZSTD_cwksp* ws) { - DEBUGLOG(4, "cwksp: ZSTD_cwksp_clean_tables"); - assert(ws->tableValidEnd >= ws->objectEnd); - assert(ws->tableValidEnd <= ws->allocStart); - if (ws->tableValidEnd < ws->tableEnd) { - ZSTD_memset(ws->tableValidEnd, 0, (BYTE*)ws->tableEnd - (BYTE*)ws->tableValidEnd); - } - ZSTD_cwksp_mark_tables_clean(ws); -} - -/** - * Invalidates table allocations. - * All other allocations remain valid. - */ -MEM_STATIC void ZSTD_cwksp_clear_tables(ZSTD_cwksp* ws) { - DEBUGLOG(4, "cwksp: clearing tables!"); - -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) - /* We don't do this when the workspace is statically allocated, because - * when that is the case, we have no capability to hook into the end of the - * workspace's lifecycle to unpoison the memory. - */ - if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { - size_t size = (BYTE*)ws->tableValidEnd - (BYTE*)ws->objectEnd; - __asan_poison_memory_region(ws->objectEnd, size); - } -#endif - - ws->tableEnd = ws->objectEnd; - ZSTD_cwksp_assert_internal_consistency(ws); -} - -/** - * Invalidates all buffer, aligned, and table allocations. - * Object allocations remain valid. - */ -MEM_STATIC void ZSTD_cwksp_clear(ZSTD_cwksp* ws) { - DEBUGLOG(4, "cwksp: clearing!"); - -#if ZSTD_MEMORY_SANITIZER && !defined (ZSTD_MSAN_DONT_POISON_WORKSPACE) - /* To validate that the context re-use logic is sound, and that we don't - * access stuff that this compression hasn't initialized, we re-"poison" - * the workspace (or at least the non-static, non-table parts of it) - * every time we start a new compression. */ - { - size_t size = (BYTE*)ws->workspaceEnd - (BYTE*)ws->tableValidEnd; - __msan_poison(ws->tableValidEnd, size); - } -#endif - -#if ZSTD_ADDRESS_SANITIZER && !defined (ZSTD_ASAN_DONT_POISON_WORKSPACE) - /* We don't do this when the workspace is statically allocated, because - * when that is the case, we have no capability to hook into the end of the - * workspace's lifecycle to unpoison the memory. - */ - if (ws->isStatic == ZSTD_cwksp_dynamic_alloc) { - size_t size = (BYTE*)ws->workspaceEnd - (BYTE*)ws->objectEnd; - __asan_poison_memory_region(ws->objectEnd, size); - } -#endif - - ws->tableEnd = ws->objectEnd; - ws->allocStart = ws->workspaceEnd; - ws->allocFailed = 0; - if (ws->phase > ZSTD_cwksp_alloc_buffers) { - ws->phase = ZSTD_cwksp_alloc_buffers; - } - ZSTD_cwksp_assert_internal_consistency(ws); -} - -/** - * The provided workspace takes ownership of the buffer [start, start+size). - * Any existing values in the workspace are ignored (the previously managed - * buffer, if present, must be separately freed). - */ -MEM_STATIC void ZSTD_cwksp_init(ZSTD_cwksp* ws, void* start, size_t size, ZSTD_cwksp_static_alloc_e isStatic) { - DEBUGLOG(4, "cwksp: init'ing workspace with %zd bytes", size); - assert(((size_t)start & (sizeof(void*)-1)) == 0); /* ensure correct alignment */ - ws->workspace = start; - ws->workspaceEnd = (BYTE*)start + size; - ws->objectEnd = ws->workspace; - ws->tableValidEnd = ws->objectEnd; - ws->phase = ZSTD_cwksp_alloc_objects; - ws->isStatic = isStatic; - ZSTD_cwksp_clear(ws); - ws->workspaceOversizedDuration = 0; - ZSTD_cwksp_assert_internal_consistency(ws); -} - -MEM_STATIC size_t ZSTD_cwksp_create(ZSTD_cwksp* ws, size_t size, ZSTD_customMem customMem) { - void* workspace = ZSTD_customMalloc(size, customMem); - DEBUGLOG(4, "cwksp: creating new workspace with %zd bytes", size); - RETURN_ERROR_IF(workspace == NULL, memory_allocation, "NULL pointer!"); - ZSTD_cwksp_init(ws, workspace, size, ZSTD_cwksp_dynamic_alloc); - return 0; -} - -MEM_STATIC void ZSTD_cwksp_free(ZSTD_cwksp* ws, ZSTD_customMem customMem) { - void *ptr = ws->workspace; - DEBUGLOG(4, "cwksp: freeing workspace"); - ZSTD_memset(ws, 0, sizeof(ZSTD_cwksp)); - ZSTD_customFree(ptr, customMem); -} - -/** - * Moves the management of a workspace from one cwksp to another. The src cwksp - * is left in an invalid state (src must be re-init()'ed before it's used again). - */ -MEM_STATIC void ZSTD_cwksp_move(ZSTD_cwksp* dst, ZSTD_cwksp* src) { - *dst = *src; - ZSTD_memset(src, 0, sizeof(ZSTD_cwksp)); -} - -MEM_STATIC size_t ZSTD_cwksp_sizeof(const ZSTD_cwksp* ws) { - return (size_t)((BYTE*)ws->workspaceEnd - (BYTE*)ws->workspace); -} - -MEM_STATIC size_t ZSTD_cwksp_used(const ZSTD_cwksp* ws) { - return (size_t)((BYTE*)ws->tableEnd - (BYTE*)ws->workspace) - + (size_t)((BYTE*)ws->workspaceEnd - (BYTE*)ws->allocStart); -} - -MEM_STATIC int ZSTD_cwksp_reserve_failed(const ZSTD_cwksp* ws) { - return ws->allocFailed; -} - -/*-************************************* -* Functions Checking Free Space -***************************************/ - -/* ZSTD_alignmentSpaceWithinBounds() : - * Returns if the estimated space needed for a wksp is within an acceptable limit of the - * actual amount of space used. - */ -MEM_STATIC int ZSTD_cwksp_estimated_space_within_bounds(const ZSTD_cwksp* const ws, - size_t const estimatedSpace, int resizedWorkspace) { - if (resizedWorkspace) { - /* Resized/newly allocated wksp should have exact bounds */ - return ZSTD_cwksp_used(ws) == estimatedSpace; - } else { - /* Due to alignment, when reusing a workspace, we can actually consume 63 fewer or more bytes - * than estimatedSpace. See the comments in zstd_cwksp.h for details. - */ - return (ZSTD_cwksp_used(ws) >= estimatedSpace - 63) && (ZSTD_cwksp_used(ws) <= estimatedSpace + 63); - } -} - - -MEM_STATIC size_t ZSTD_cwksp_available_space(ZSTD_cwksp* ws) { - return (size_t)((BYTE*)ws->allocStart - (BYTE*)ws->tableEnd); -} - -MEM_STATIC int ZSTD_cwksp_check_available(ZSTD_cwksp* ws, size_t additionalNeededSpace) { - return ZSTD_cwksp_available_space(ws) >= additionalNeededSpace; -} - -MEM_STATIC int ZSTD_cwksp_check_too_large(ZSTD_cwksp* ws, size_t additionalNeededSpace) { - return ZSTD_cwksp_check_available( - ws, additionalNeededSpace * ZSTD_WORKSPACETOOLARGE_FACTOR); -} - -MEM_STATIC int ZSTD_cwksp_check_wasteful(ZSTD_cwksp* ws, size_t additionalNeededSpace) { - return ZSTD_cwksp_check_too_large(ws, additionalNeededSpace) - && ws->workspaceOversizedDuration > ZSTD_WORKSPACETOOLARGE_MAXDURATION; -} - -MEM_STATIC void ZSTD_cwksp_bump_oversized_duration( - ZSTD_cwksp* ws, size_t additionalNeededSpace) { - if (ZSTD_cwksp_check_too_large(ws, additionalNeededSpace)) { - ws->workspaceOversizedDuration++; - } else { - ws->workspaceOversizedDuration = 0; - } -} - -#if defined (__cplusplus) -} -#endif - -#endif /* ZSTD_CWKSP_H */ diff --git a/dep/zstd/lib/compress/zstd_double_fast.c b/dep/zstd/lib/compress/zstd_double_fast.c deleted file mode 100644 index 76933dea2..000000000 --- a/dep/zstd/lib/compress/zstd_double_fast.c +++ /dev/null @@ -1,696 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#include "zstd_compress_internal.h" -#include "zstd_double_fast.h" - - -void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms, - void const* end, ZSTD_dictTableLoadMethod_e dtlm) -{ - const ZSTD_compressionParameters* const cParams = &ms->cParams; - U32* const hashLarge = ms->hashTable; - U32 const hBitsL = cParams->hashLog; - U32 const mls = cParams->minMatch; - U32* const hashSmall = ms->chainTable; - U32 const hBitsS = cParams->chainLog; - const BYTE* const base = ms->window.base; - const BYTE* ip = base + ms->nextToUpdate; - const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE; - const U32 fastHashFillStep = 3; - - /* Always insert every fastHashFillStep position into the hash tables. - * Insert the other positions into the large hash table if their entry - * is empty. - */ - for (; ip + fastHashFillStep - 1 <= iend; ip += fastHashFillStep) { - U32 const curr = (U32)(ip - base); - U32 i; - for (i = 0; i < fastHashFillStep; ++i) { - size_t const smHash = ZSTD_hashPtr(ip + i, hBitsS, mls); - size_t const lgHash = ZSTD_hashPtr(ip + i, hBitsL, 8); - if (i == 0) - hashSmall[smHash] = curr + i; - if (i == 0 || hashLarge[lgHash] == 0) - hashLarge[lgHash] = curr + i; - /* Only load extra positions for ZSTD_dtlm_full */ - if (dtlm == ZSTD_dtlm_fast) - break; - } } -} - - -FORCE_INLINE_TEMPLATE -size_t ZSTD_compressBlock_doubleFast_noDict_generic( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize, U32 const mls /* template */) -{ - ZSTD_compressionParameters const* cParams = &ms->cParams; - U32* const hashLong = ms->hashTable; - const U32 hBitsL = cParams->hashLog; - U32* const hashSmall = ms->chainTable; - const U32 hBitsS = cParams->chainLog; - const BYTE* const base = ms->window.base; - const BYTE* const istart = (const BYTE*)src; - const BYTE* anchor = istart; - const U32 endIndex = (U32)((size_t)(istart - base) + srcSize); - /* presumes that, if there is a dictionary, it must be using Attach mode */ - const U32 prefixLowestIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog); - const BYTE* const prefixLowest = base + prefixLowestIndex; - const BYTE* const iend = istart + srcSize; - const BYTE* const ilimit = iend - HASH_READ_SIZE; - U32 offset_1=rep[0], offset_2=rep[1]; - U32 offsetSaved = 0; - - size_t mLength; - U32 offset; - U32 curr; - - /* how many positions to search before increasing step size */ - const size_t kStepIncr = 1 << kSearchStrength; - /* the position at which to increment the step size if no match is found */ - const BYTE* nextStep; - size_t step; /* the current step size */ - - size_t hl0; /* the long hash at ip */ - size_t hl1; /* the long hash at ip1 */ - - U32 idxl0; /* the long match index for ip */ - U32 idxl1; /* the long match index for ip1 */ - - const BYTE* matchl0; /* the long match for ip */ - const BYTE* matchs0; /* the short match for ip */ - const BYTE* matchl1; /* the long match for ip1 */ - - const BYTE* ip = istart; /* the current position */ - const BYTE* ip1; /* the next position */ - - DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_noDict_generic"); - - /* init */ - ip += ((ip - prefixLowest) == 0); - { - U32 const current = (U32)(ip - base); - U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, current, cParams->windowLog); - U32 const maxRep = current - windowLow; - if (offset_2 > maxRep) offsetSaved = offset_2, offset_2 = 0; - if (offset_1 > maxRep) offsetSaved = offset_1, offset_1 = 0; - } - - /* Outer Loop: one iteration per match found and stored */ - while (1) { - step = 1; - nextStep = ip + kStepIncr; - ip1 = ip + step; - - if (ip1 > ilimit) { - goto _cleanup; - } - - hl0 = ZSTD_hashPtr(ip, hBitsL, 8); - idxl0 = hashLong[hl0]; - matchl0 = base + idxl0; - - /* Inner Loop: one iteration per search / position */ - do { - const size_t hs0 = ZSTD_hashPtr(ip, hBitsS, mls); - const U32 idxs0 = hashSmall[hs0]; - curr = (U32)(ip-base); - matchs0 = base + idxs0; - - hashLong[hl0] = hashSmall[hs0] = curr; /* update hash tables */ - - /* check noDict repcode */ - if ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1))) { - mLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4; - ip++; - ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_REPCODE_1, mLength); - goto _match_stored; - } - - hl1 = ZSTD_hashPtr(ip1, hBitsL, 8); - - if (idxl0 > prefixLowestIndex) { - /* check prefix long match */ - if (MEM_read64(matchl0) == MEM_read64(ip)) { - mLength = ZSTD_count(ip+8, matchl0+8, iend) + 8; - offset = (U32)(ip-matchl0); - while (((ip>anchor) & (matchl0>prefixLowest)) && (ip[-1] == matchl0[-1])) { ip--; matchl0--; mLength++; } /* catch up */ - goto _match_found; - } - } - - idxl1 = hashLong[hl1]; - matchl1 = base + idxl1; - - if (idxs0 > prefixLowestIndex) { - /* check prefix short match */ - if (MEM_read32(matchs0) == MEM_read32(ip)) { - goto _search_next_long; - } - } - - if (ip1 >= nextStep) { - PREFETCH_L1(ip1 + 64); - PREFETCH_L1(ip1 + 128); - step++; - nextStep += kStepIncr; - } - ip = ip1; - ip1 += step; - - hl0 = hl1; - idxl0 = idxl1; - matchl0 = matchl1; - #if defined(__aarch64__) - PREFETCH_L1(ip+256); - #endif - } while (ip1 <= ilimit); - -_cleanup: - /* save reps for next block */ - rep[0] = offset_1 ? offset_1 : offsetSaved; - rep[1] = offset_2 ? offset_2 : offsetSaved; - - /* Return the last literals size */ - return (size_t)(iend - anchor); - -_search_next_long: - - /* check prefix long +1 match */ - if (idxl1 > prefixLowestIndex) { - if (MEM_read64(matchl1) == MEM_read64(ip1)) { - ip = ip1; - mLength = ZSTD_count(ip+8, matchl1+8, iend) + 8; - offset = (U32)(ip-matchl1); - while (((ip>anchor) & (matchl1>prefixLowest)) && (ip[-1] == matchl1[-1])) { ip--; matchl1--; mLength++; } /* catch up */ - goto _match_found; - } - } - - /* if no long +1 match, explore the short match we found */ - mLength = ZSTD_count(ip+4, matchs0+4, iend) + 4; - offset = (U32)(ip - matchs0); - while (((ip>anchor) & (matchs0>prefixLowest)) && (ip[-1] == matchs0[-1])) { ip--; matchs0--; mLength++; } /* catch up */ - - /* fall-through */ - -_match_found: /* requires ip, offset, mLength */ - offset_2 = offset_1; - offset_1 = offset; - - if (step < 4) { - /* It is unsafe to write this value back to the hashtable when ip1 is - * greater than or equal to the new ip we will have after we're done - * processing this match. Rather than perform that test directly - * (ip1 >= ip + mLength), which costs speed in practice, we do a simpler - * more predictable test. The minmatch even if we take a short match is - * 4 bytes, so as long as step, the distance between ip and ip1 - * (initially) is less than 4, we know ip1 < new ip. */ - hashLong[hl1] = (U32)(ip1 - base); - } - - ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength); - -_match_stored: - /* match found */ - ip += mLength; - anchor = ip; - - if (ip <= ilimit) { - /* Complementary insertion */ - /* done after iLimit test, as candidates could be > iend-8 */ - { U32 const indexToInsert = curr+2; - hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert; - hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base); - hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert; - hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base); - } - - /* check immediate repcode */ - while ( (ip <= ilimit) - && ( (offset_2>0) - & (MEM_read32(ip) == MEM_read32(ip - offset_2)) )) { - /* store sequence */ - size_t const rLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4; - U32 const tmpOff = offset_2; offset_2 = offset_1; offset_1 = tmpOff; /* swap offset_2 <=> offset_1 */ - hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = (U32)(ip-base); - hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = (U32)(ip-base); - ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, rLength); - ip += rLength; - anchor = ip; - continue; /* faster when present ... (?) */ - } - } - } -} - - -FORCE_INLINE_TEMPLATE -size_t ZSTD_compressBlock_doubleFast_dictMatchState_generic( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize, - U32 const mls /* template */) -{ - ZSTD_compressionParameters const* cParams = &ms->cParams; - U32* const hashLong = ms->hashTable; - const U32 hBitsL = cParams->hashLog; - U32* const hashSmall = ms->chainTable; - const U32 hBitsS = cParams->chainLog; - const BYTE* const base = ms->window.base; - const BYTE* const istart = (const BYTE*)src; - const BYTE* ip = istart; - const BYTE* anchor = istart; - const U32 endIndex = (U32)((size_t)(istart - base) + srcSize); - /* presumes that, if there is a dictionary, it must be using Attach mode */ - const U32 prefixLowestIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog); - const BYTE* const prefixLowest = base + prefixLowestIndex; - const BYTE* const iend = istart + srcSize; - const BYTE* const ilimit = iend - HASH_READ_SIZE; - U32 offset_1=rep[0], offset_2=rep[1]; - U32 offsetSaved = 0; - - const ZSTD_matchState_t* const dms = ms->dictMatchState; - const ZSTD_compressionParameters* const dictCParams = &dms->cParams; - const U32* const dictHashLong = dms->hashTable; - const U32* const dictHashSmall = dms->chainTable; - const U32 dictStartIndex = dms->window.dictLimit; - const BYTE* const dictBase = dms->window.base; - const BYTE* const dictStart = dictBase + dictStartIndex; - const BYTE* const dictEnd = dms->window.nextSrc; - const U32 dictIndexDelta = prefixLowestIndex - (U32)(dictEnd - dictBase); - const U32 dictHBitsL = dictCParams->hashLog; - const U32 dictHBitsS = dictCParams->chainLog; - const U32 dictAndPrefixLength = (U32)((ip - prefixLowest) + (dictEnd - dictStart)); - - DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_dictMatchState_generic"); - - /* if a dictionary is attached, it must be within window range */ - assert(ms->window.dictLimit + (1U << cParams->windowLog) >= endIndex); - - /* init */ - ip += (dictAndPrefixLength == 0); - - /* dictMatchState repCode checks don't currently handle repCode == 0 - * disabling. */ - assert(offset_1 <= dictAndPrefixLength); - assert(offset_2 <= dictAndPrefixLength); - - /* Main Search Loop */ - while (ip < ilimit) { /* < instead of <=, because repcode check at (ip+1) */ - size_t mLength; - U32 offset; - size_t const h2 = ZSTD_hashPtr(ip, hBitsL, 8); - size_t const h = ZSTD_hashPtr(ip, hBitsS, mls); - size_t const dictHL = ZSTD_hashPtr(ip, dictHBitsL, 8); - size_t const dictHS = ZSTD_hashPtr(ip, dictHBitsS, mls); - U32 const curr = (U32)(ip-base); - U32 const matchIndexL = hashLong[h2]; - U32 matchIndexS = hashSmall[h]; - const BYTE* matchLong = base + matchIndexL; - const BYTE* match = base + matchIndexS; - const U32 repIndex = curr + 1 - offset_1; - const BYTE* repMatch = (repIndex < prefixLowestIndex) ? - dictBase + (repIndex - dictIndexDelta) : - base + repIndex; - hashLong[h2] = hashSmall[h] = curr; /* update hash tables */ - - /* check repcode */ - if (((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */) - && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) { - const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend; - mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4; - ip++; - ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_REPCODE_1, mLength); - goto _match_stored; - } - - if (matchIndexL > prefixLowestIndex) { - /* check prefix long match */ - if (MEM_read64(matchLong) == MEM_read64(ip)) { - mLength = ZSTD_count(ip+8, matchLong+8, iend) + 8; - offset = (U32)(ip-matchLong); - while (((ip>anchor) & (matchLong>prefixLowest)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */ - goto _match_found; - } - } else { - /* check dictMatchState long match */ - U32 const dictMatchIndexL = dictHashLong[dictHL]; - const BYTE* dictMatchL = dictBase + dictMatchIndexL; - assert(dictMatchL < dictEnd); - - if (dictMatchL > dictStart && MEM_read64(dictMatchL) == MEM_read64(ip)) { - mLength = ZSTD_count_2segments(ip+8, dictMatchL+8, iend, dictEnd, prefixLowest) + 8; - offset = (U32)(curr - dictMatchIndexL - dictIndexDelta); - while (((ip>anchor) & (dictMatchL>dictStart)) && (ip[-1] == dictMatchL[-1])) { ip--; dictMatchL--; mLength++; } /* catch up */ - goto _match_found; - } } - - if (matchIndexS > prefixLowestIndex) { - /* check prefix short match */ - if (MEM_read32(match) == MEM_read32(ip)) { - goto _search_next_long; - } - } else { - /* check dictMatchState short match */ - U32 const dictMatchIndexS = dictHashSmall[dictHS]; - match = dictBase + dictMatchIndexS; - matchIndexS = dictMatchIndexS + dictIndexDelta; - - if (match > dictStart && MEM_read32(match) == MEM_read32(ip)) { - goto _search_next_long; - } } - - ip += ((ip-anchor) >> kSearchStrength) + 1; -#if defined(__aarch64__) - PREFETCH_L1(ip+256); -#endif - continue; - -_search_next_long: - - { size_t const hl3 = ZSTD_hashPtr(ip+1, hBitsL, 8); - size_t const dictHLNext = ZSTD_hashPtr(ip+1, dictHBitsL, 8); - U32 const matchIndexL3 = hashLong[hl3]; - const BYTE* matchL3 = base + matchIndexL3; - hashLong[hl3] = curr + 1; - - /* check prefix long +1 match */ - if (matchIndexL3 > prefixLowestIndex) { - if (MEM_read64(matchL3) == MEM_read64(ip+1)) { - mLength = ZSTD_count(ip+9, matchL3+8, iend) + 8; - ip++; - offset = (U32)(ip-matchL3); - while (((ip>anchor) & (matchL3>prefixLowest)) && (ip[-1] == matchL3[-1])) { ip--; matchL3--; mLength++; } /* catch up */ - goto _match_found; - } - } else { - /* check dict long +1 match */ - U32 const dictMatchIndexL3 = dictHashLong[dictHLNext]; - const BYTE* dictMatchL3 = dictBase + dictMatchIndexL3; - assert(dictMatchL3 < dictEnd); - if (dictMatchL3 > dictStart && MEM_read64(dictMatchL3) == MEM_read64(ip+1)) { - mLength = ZSTD_count_2segments(ip+1+8, dictMatchL3+8, iend, dictEnd, prefixLowest) + 8; - ip++; - offset = (U32)(curr + 1 - dictMatchIndexL3 - dictIndexDelta); - while (((ip>anchor) & (dictMatchL3>dictStart)) && (ip[-1] == dictMatchL3[-1])) { ip--; dictMatchL3--; mLength++; } /* catch up */ - goto _match_found; - } } } - - /* if no long +1 match, explore the short match we found */ - if (matchIndexS < prefixLowestIndex) { - mLength = ZSTD_count_2segments(ip+4, match+4, iend, dictEnd, prefixLowest) + 4; - offset = (U32)(curr - matchIndexS); - while (((ip>anchor) & (match>dictStart)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */ - } else { - mLength = ZSTD_count(ip+4, match+4, iend) + 4; - offset = (U32)(ip - match); - while (((ip>anchor) & (match>prefixLowest)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */ - } - -_match_found: - offset_2 = offset_1; - offset_1 = offset; - - ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength); - -_match_stored: - /* match found */ - ip += mLength; - anchor = ip; - - if (ip <= ilimit) { - /* Complementary insertion */ - /* done after iLimit test, as candidates could be > iend-8 */ - { U32 const indexToInsert = curr+2; - hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert; - hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base); - hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert; - hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base); - } - - /* check immediate repcode */ - while (ip <= ilimit) { - U32 const current2 = (U32)(ip-base); - U32 const repIndex2 = current2 - offset_2; - const BYTE* repMatch2 = repIndex2 < prefixLowestIndex ? - dictBase + repIndex2 - dictIndexDelta : - base + repIndex2; - if ( ((U32)((prefixLowestIndex-1) - (U32)repIndex2) >= 3 /* intentional overflow */) - && (MEM_read32(repMatch2) == MEM_read32(ip)) ) { - const BYTE* const repEnd2 = repIndex2 < prefixLowestIndex ? dictEnd : iend; - size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixLowest) + 4; - U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */ - ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, repLength2); - hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2; - hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2; - ip += repLength2; - anchor = ip; - continue; - } - break; - } - } - } /* while (ip < ilimit) */ - - /* save reps for next block */ - rep[0] = offset_1 ? offset_1 : offsetSaved; - rep[1] = offset_2 ? offset_2 : offsetSaved; - - /* Return the last literals size */ - return (size_t)(iend - anchor); -} - -#define ZSTD_GEN_DFAST_FN(dictMode, mls) \ - static size_t ZSTD_compressBlock_doubleFast_##dictMode##_##mls( \ - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], \ - void const* src, size_t srcSize) \ - { \ - return ZSTD_compressBlock_doubleFast_##dictMode##_generic(ms, seqStore, rep, src, srcSize, mls); \ - } - -ZSTD_GEN_DFAST_FN(noDict, 4) -ZSTD_GEN_DFAST_FN(noDict, 5) -ZSTD_GEN_DFAST_FN(noDict, 6) -ZSTD_GEN_DFAST_FN(noDict, 7) - -ZSTD_GEN_DFAST_FN(dictMatchState, 4) -ZSTD_GEN_DFAST_FN(dictMatchState, 5) -ZSTD_GEN_DFAST_FN(dictMatchState, 6) -ZSTD_GEN_DFAST_FN(dictMatchState, 7) - - -size_t ZSTD_compressBlock_doubleFast( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - const U32 mls = ms->cParams.minMatch; - switch(mls) - { - default: /* includes case 3 */ - case 4 : - return ZSTD_compressBlock_doubleFast_noDict_4(ms, seqStore, rep, src, srcSize); - case 5 : - return ZSTD_compressBlock_doubleFast_noDict_5(ms, seqStore, rep, src, srcSize); - case 6 : - return ZSTD_compressBlock_doubleFast_noDict_6(ms, seqStore, rep, src, srcSize); - case 7 : - return ZSTD_compressBlock_doubleFast_noDict_7(ms, seqStore, rep, src, srcSize); - } -} - - -size_t ZSTD_compressBlock_doubleFast_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - const U32 mls = ms->cParams.minMatch; - switch(mls) - { - default: /* includes case 3 */ - case 4 : - return ZSTD_compressBlock_doubleFast_dictMatchState_4(ms, seqStore, rep, src, srcSize); - case 5 : - return ZSTD_compressBlock_doubleFast_dictMatchState_5(ms, seqStore, rep, src, srcSize); - case 6 : - return ZSTD_compressBlock_doubleFast_dictMatchState_6(ms, seqStore, rep, src, srcSize); - case 7 : - return ZSTD_compressBlock_doubleFast_dictMatchState_7(ms, seqStore, rep, src, srcSize); - } -} - - -static size_t ZSTD_compressBlock_doubleFast_extDict_generic( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize, - U32 const mls /* template */) -{ - ZSTD_compressionParameters const* cParams = &ms->cParams; - U32* const hashLong = ms->hashTable; - U32 const hBitsL = cParams->hashLog; - U32* const hashSmall = ms->chainTable; - U32 const hBitsS = cParams->chainLog; - const BYTE* const istart = (const BYTE*)src; - const BYTE* ip = istart; - const BYTE* anchor = istart; - const BYTE* const iend = istart + srcSize; - const BYTE* const ilimit = iend - 8; - const BYTE* const base = ms->window.base; - const U32 endIndex = (U32)((size_t)(istart - base) + srcSize); - const U32 lowLimit = ZSTD_getLowestMatchIndex(ms, endIndex, cParams->windowLog); - const U32 dictStartIndex = lowLimit; - const U32 dictLimit = ms->window.dictLimit; - const U32 prefixStartIndex = (dictLimit > lowLimit) ? dictLimit : lowLimit; - const BYTE* const prefixStart = base + prefixStartIndex; - const BYTE* const dictBase = ms->window.dictBase; - const BYTE* const dictStart = dictBase + dictStartIndex; - const BYTE* const dictEnd = dictBase + prefixStartIndex; - U32 offset_1=rep[0], offset_2=rep[1]; - - DEBUGLOG(5, "ZSTD_compressBlock_doubleFast_extDict_generic (srcSize=%zu)", srcSize); - - /* if extDict is invalidated due to maxDistance, switch to "regular" variant */ - if (prefixStartIndex == dictStartIndex) - return ZSTD_compressBlock_doubleFast(ms, seqStore, rep, src, srcSize); - - /* Search Loop */ - while (ip < ilimit) { /* < instead of <=, because (ip+1) */ - const size_t hSmall = ZSTD_hashPtr(ip, hBitsS, mls); - const U32 matchIndex = hashSmall[hSmall]; - const BYTE* const matchBase = matchIndex < prefixStartIndex ? dictBase : base; - const BYTE* match = matchBase + matchIndex; - - const size_t hLong = ZSTD_hashPtr(ip, hBitsL, 8); - const U32 matchLongIndex = hashLong[hLong]; - const BYTE* const matchLongBase = matchLongIndex < prefixStartIndex ? dictBase : base; - const BYTE* matchLong = matchLongBase + matchLongIndex; - - const U32 curr = (U32)(ip-base); - const U32 repIndex = curr + 1 - offset_1; /* offset_1 expected <= curr +1 */ - const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base; - const BYTE* const repMatch = repBase + repIndex; - size_t mLength; - hashSmall[hSmall] = hashLong[hLong] = curr; /* update hash table */ - - if ((((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow : ensure repIndex doesn't overlap dict + prefix */ - & (offset_1 <= curr+1 - dictStartIndex)) /* note: we are searching at curr+1 */ - && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) { - const BYTE* repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend; - mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4; - ip++; - ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_REPCODE_1, mLength); - } else { - if ((matchLongIndex > dictStartIndex) && (MEM_read64(matchLong) == MEM_read64(ip))) { - const BYTE* const matchEnd = matchLongIndex < prefixStartIndex ? dictEnd : iend; - const BYTE* const lowMatchPtr = matchLongIndex < prefixStartIndex ? dictStart : prefixStart; - U32 offset; - mLength = ZSTD_count_2segments(ip+8, matchLong+8, iend, matchEnd, prefixStart) + 8; - offset = curr - matchLongIndex; - while (((ip>anchor) & (matchLong>lowMatchPtr)) && (ip[-1] == matchLong[-1])) { ip--; matchLong--; mLength++; } /* catch up */ - offset_2 = offset_1; - offset_1 = offset; - ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength); - - } else if ((matchIndex > dictStartIndex) && (MEM_read32(match) == MEM_read32(ip))) { - size_t const h3 = ZSTD_hashPtr(ip+1, hBitsL, 8); - U32 const matchIndex3 = hashLong[h3]; - const BYTE* const match3Base = matchIndex3 < prefixStartIndex ? dictBase : base; - const BYTE* match3 = match3Base + matchIndex3; - U32 offset; - hashLong[h3] = curr + 1; - if ( (matchIndex3 > dictStartIndex) && (MEM_read64(match3) == MEM_read64(ip+1)) ) { - const BYTE* const matchEnd = matchIndex3 < prefixStartIndex ? dictEnd : iend; - const BYTE* const lowMatchPtr = matchIndex3 < prefixStartIndex ? dictStart : prefixStart; - mLength = ZSTD_count_2segments(ip+9, match3+8, iend, matchEnd, prefixStart) + 8; - ip++; - offset = curr+1 - matchIndex3; - while (((ip>anchor) & (match3>lowMatchPtr)) && (ip[-1] == match3[-1])) { ip--; match3--; mLength++; } /* catch up */ - } else { - const BYTE* const matchEnd = matchIndex < prefixStartIndex ? dictEnd : iend; - const BYTE* const lowMatchPtr = matchIndex < prefixStartIndex ? dictStart : prefixStart; - mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4; - offset = curr - matchIndex; - while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */ - } - offset_2 = offset_1; - offset_1 = offset; - ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength); - - } else { - ip += ((ip-anchor) >> kSearchStrength) + 1; - continue; - } } - - /* move to next sequence start */ - ip += mLength; - anchor = ip; - - if (ip <= ilimit) { - /* Complementary insertion */ - /* done after iLimit test, as candidates could be > iend-8 */ - { U32 const indexToInsert = curr+2; - hashLong[ZSTD_hashPtr(base+indexToInsert, hBitsL, 8)] = indexToInsert; - hashLong[ZSTD_hashPtr(ip-2, hBitsL, 8)] = (U32)(ip-2-base); - hashSmall[ZSTD_hashPtr(base+indexToInsert, hBitsS, mls)] = indexToInsert; - hashSmall[ZSTD_hashPtr(ip-1, hBitsS, mls)] = (U32)(ip-1-base); - } - - /* check immediate repcode */ - while (ip <= ilimit) { - U32 const current2 = (U32)(ip-base); - U32 const repIndex2 = current2 - offset_2; - const BYTE* repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2; - if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3) /* intentional overflow : ensure repIndex2 doesn't overlap dict + prefix */ - & (offset_2 <= current2 - dictStartIndex)) - && (MEM_read32(repMatch2) == MEM_read32(ip)) ) { - const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend; - size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4; - U32 const tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */ - ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, repLength2); - hashSmall[ZSTD_hashPtr(ip, hBitsS, mls)] = current2; - hashLong[ZSTD_hashPtr(ip, hBitsL, 8)] = current2; - ip += repLength2; - anchor = ip; - continue; - } - break; - } } } - - /* save reps for next block */ - rep[0] = offset_1; - rep[1] = offset_2; - - /* Return the last literals size */ - return (size_t)(iend - anchor); -} - -ZSTD_GEN_DFAST_FN(extDict, 4) -ZSTD_GEN_DFAST_FN(extDict, 5) -ZSTD_GEN_DFAST_FN(extDict, 6) -ZSTD_GEN_DFAST_FN(extDict, 7) - -size_t ZSTD_compressBlock_doubleFast_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - U32 const mls = ms->cParams.minMatch; - switch(mls) - { - default: /* includes case 3 */ - case 4 : - return ZSTD_compressBlock_doubleFast_extDict_4(ms, seqStore, rep, src, srcSize); - case 5 : - return ZSTD_compressBlock_doubleFast_extDict_5(ms, seqStore, rep, src, srcSize); - case 6 : - return ZSTD_compressBlock_doubleFast_extDict_6(ms, seqStore, rep, src, srcSize); - case 7 : - return ZSTD_compressBlock_doubleFast_extDict_7(ms, seqStore, rep, src, srcSize); - } -} diff --git a/dep/zstd/lib/compress/zstd_double_fast.h b/dep/zstd/lib/compress/zstd_double_fast.h deleted file mode 100644 index e16b7b03a..000000000 --- a/dep/zstd/lib/compress/zstd_double_fast.h +++ /dev/null @@ -1,38 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_DOUBLE_FAST_H -#define ZSTD_DOUBLE_FAST_H - -#if defined (__cplusplus) -extern "C" { -#endif - -#include "../common/mem.h" /* U32 */ -#include "zstd_compress_internal.h" /* ZSTD_CCtx, size_t */ - -void ZSTD_fillDoubleHashTable(ZSTD_matchState_t* ms, - void const* end, ZSTD_dictTableLoadMethod_e dtlm); -size_t ZSTD_compressBlock_doubleFast( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_doubleFast_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_doubleFast_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); - - -#if defined (__cplusplus) -} -#endif - -#endif /* ZSTD_DOUBLE_FAST_H */ diff --git a/dep/zstd/lib/compress/zstd_fast.c b/dep/zstd/lib/compress/zstd_fast.c deleted file mode 100644 index 802fc3157..000000000 --- a/dep/zstd/lib/compress/zstd_fast.c +++ /dev/null @@ -1,675 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#include "zstd_compress_internal.h" /* ZSTD_hashPtr, ZSTD_count, ZSTD_storeSeq */ -#include "zstd_fast.h" - - -void ZSTD_fillHashTable(ZSTD_matchState_t* ms, - const void* const end, - ZSTD_dictTableLoadMethod_e dtlm) -{ - const ZSTD_compressionParameters* const cParams = &ms->cParams; - U32* const hashTable = ms->hashTable; - U32 const hBits = cParams->hashLog; - U32 const mls = cParams->minMatch; - const BYTE* const base = ms->window.base; - const BYTE* ip = base + ms->nextToUpdate; - const BYTE* const iend = ((const BYTE*)end) - HASH_READ_SIZE; - const U32 fastHashFillStep = 3; - - /* Always insert every fastHashFillStep position into the hash table. - * Insert the other positions if their hash entry is empty. - */ - for ( ; ip + fastHashFillStep < iend + 2; ip += fastHashFillStep) { - U32 const curr = (U32)(ip - base); - size_t const hash0 = ZSTD_hashPtr(ip, hBits, mls); - hashTable[hash0] = curr; - if (dtlm == ZSTD_dtlm_fast) continue; - /* Only load extra positions for ZSTD_dtlm_full */ - { U32 p; - for (p = 1; p < fastHashFillStep; ++p) { - size_t const hash = ZSTD_hashPtr(ip + p, hBits, mls); - if (hashTable[hash] == 0) { /* not yet filled */ - hashTable[hash] = curr + p; - } } } } -} - - -/** - * If you squint hard enough (and ignore repcodes), the search operation at any - * given position is broken into 4 stages: - * - * 1. Hash (map position to hash value via input read) - * 2. Lookup (map hash val to index via hashtable read) - * 3. Load (map index to value at that position via input read) - * 4. Compare - * - * Each of these steps involves a memory read at an address which is computed - * from the previous step. This means these steps must be sequenced and their - * latencies are cumulative. - * - * Rather than do 1->2->3->4 sequentially for a single position before moving - * onto the next, this implementation interleaves these operations across the - * next few positions: - * - * R = Repcode Read & Compare - * H = Hash - * T = Table Lookup - * M = Match Read & Compare - * - * Pos | Time --> - * ----+------------------- - * N | ... M - * N+1 | ... TM - * N+2 | R H T M - * N+3 | H TM - * N+4 | R H T M - * N+5 | H ... - * N+6 | R ... - * - * This is very much analogous to the pipelining of execution in a CPU. And just - * like a CPU, we have to dump the pipeline when we find a match (i.e., take a - * branch). - * - * When this happens, we throw away our current state, and do the following prep - * to re-enter the loop: - * - * Pos | Time --> - * ----+------------------- - * N | H T - * N+1 | H - * - * This is also the work we do at the beginning to enter the loop initially. - */ -FORCE_INLINE_TEMPLATE size_t -ZSTD_compressBlock_fast_noDict_generic( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize, - U32 const mls, U32 const hasStep) -{ - const ZSTD_compressionParameters* const cParams = &ms->cParams; - U32* const hashTable = ms->hashTable; - U32 const hlog = cParams->hashLog; - /* support stepSize of 0 */ - size_t const stepSize = hasStep ? (cParams->targetLength + !(cParams->targetLength) + 1) : 2; - const BYTE* const base = ms->window.base; - const BYTE* const istart = (const BYTE*)src; - const U32 endIndex = (U32)((size_t)(istart - base) + srcSize); - const U32 prefixStartIndex = ZSTD_getLowestPrefixIndex(ms, endIndex, cParams->windowLog); - const BYTE* const prefixStart = base + prefixStartIndex; - const BYTE* const iend = istart + srcSize; - const BYTE* const ilimit = iend - HASH_READ_SIZE; - - const BYTE* anchor = istart; - const BYTE* ip0 = istart; - const BYTE* ip1; - const BYTE* ip2; - const BYTE* ip3; - U32 current0; - - U32 rep_offset1 = rep[0]; - U32 rep_offset2 = rep[1]; - U32 offsetSaved = 0; - - size_t hash0; /* hash for ip0 */ - size_t hash1; /* hash for ip1 */ - U32 idx; /* match idx for ip0 */ - U32 mval; /* src value at match idx */ - - U32 offcode; - const BYTE* match0; - size_t mLength; - - /* ip0 and ip1 are always adjacent. The targetLength skipping and - * uncompressibility acceleration is applied to every other position, - * matching the behavior of #1562. step therefore represents the gap - * between pairs of positions, from ip0 to ip2 or ip1 to ip3. */ - size_t step; - const BYTE* nextStep; - const size_t kStepIncr = (1 << (kSearchStrength - 1)); - - DEBUGLOG(5, "ZSTD_compressBlock_fast_generic"); - ip0 += (ip0 == prefixStart); - { U32 const curr = (U32)(ip0 - base); - U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, curr, cParams->windowLog); - U32 const maxRep = curr - windowLow; - if (rep_offset2 > maxRep) offsetSaved = rep_offset2, rep_offset2 = 0; - if (rep_offset1 > maxRep) offsetSaved = rep_offset1, rep_offset1 = 0; - } - - /* start each op */ -_start: /* Requires: ip0 */ - - step = stepSize; - nextStep = ip0 + kStepIncr; - - /* calculate positions, ip0 - anchor == 0, so we skip step calc */ - ip1 = ip0 + 1; - ip2 = ip0 + step; - ip3 = ip2 + 1; - - if (ip3 >= ilimit) { - goto _cleanup; - } - - hash0 = ZSTD_hashPtr(ip0, hlog, mls); - hash1 = ZSTD_hashPtr(ip1, hlog, mls); - - idx = hashTable[hash0]; - - do { - /* load repcode match for ip[2]*/ - const U32 rval = MEM_read32(ip2 - rep_offset1); - - /* write back hash table entry */ - current0 = (U32)(ip0 - base); - hashTable[hash0] = current0; - - /* check repcode at ip[2] */ - if ((MEM_read32(ip2) == rval) & (rep_offset1 > 0)) { - ip0 = ip2; - match0 = ip0 - rep_offset1; - mLength = ip0[-1] == match0[-1]; - ip0 -= mLength; - match0 -= mLength; - offcode = STORE_REPCODE_1; - mLength += 4; - goto _match; - } - - /* load match for ip[0] */ - if (idx >= prefixStartIndex) { - mval = MEM_read32(base + idx); - } else { - mval = MEM_read32(ip0) ^ 1; /* guaranteed to not match. */ - } - - /* check match at ip[0] */ - if (MEM_read32(ip0) == mval) { - /* found a match! */ - goto _offset; - } - - /* lookup ip[1] */ - idx = hashTable[hash1]; - - /* hash ip[2] */ - hash0 = hash1; - hash1 = ZSTD_hashPtr(ip2, hlog, mls); - - /* advance to next positions */ - ip0 = ip1; - ip1 = ip2; - ip2 = ip3; - - /* write back hash table entry */ - current0 = (U32)(ip0 - base); - hashTable[hash0] = current0; - - /* load match for ip[0] */ - if (idx >= prefixStartIndex) { - mval = MEM_read32(base + idx); - } else { - mval = MEM_read32(ip0) ^ 1; /* guaranteed to not match. */ - } - - /* check match at ip[0] */ - if (MEM_read32(ip0) == mval) { - /* found a match! */ - goto _offset; - } - - /* lookup ip[1] */ - idx = hashTable[hash1]; - - /* hash ip[2] */ - hash0 = hash1; - hash1 = ZSTD_hashPtr(ip2, hlog, mls); - - /* advance to next positions */ - ip0 = ip1; - ip1 = ip2; - ip2 = ip0 + step; - ip3 = ip1 + step; - - /* calculate step */ - if (ip2 >= nextStep) { - step++; - PREFETCH_L1(ip1 + 64); - PREFETCH_L1(ip1 + 128); - nextStep += kStepIncr; - } - } while (ip3 < ilimit); - -_cleanup: - /* Note that there are probably still a couple positions we could search. - * However, it seems to be a meaningful performance hit to try to search - * them. So let's not. */ - - /* save reps for next block */ - rep[0] = rep_offset1 ? rep_offset1 : offsetSaved; - rep[1] = rep_offset2 ? rep_offset2 : offsetSaved; - - /* Return the last literals size */ - return (size_t)(iend - anchor); - -_offset: /* Requires: ip0, idx */ - - /* Compute the offset code. */ - match0 = base + idx; - rep_offset2 = rep_offset1; - rep_offset1 = (U32)(ip0-match0); - offcode = STORE_OFFSET(rep_offset1); - mLength = 4; - - /* Count the backwards match length. */ - while (((ip0>anchor) & (match0>prefixStart)) && (ip0[-1] == match0[-1])) { - ip0--; - match0--; - mLength++; - } - -_match: /* Requires: ip0, match0, offcode */ - - /* Count the forward length. */ - mLength += ZSTD_count(ip0 + mLength, match0 + mLength, iend); - - ZSTD_storeSeq(seqStore, (size_t)(ip0 - anchor), anchor, iend, offcode, mLength); - - ip0 += mLength; - anchor = ip0; - - /* write next hash table entry */ - if (ip1 < ip0) { - hashTable[hash1] = (U32)(ip1 - base); - } - - /* Fill table and check for immediate repcode. */ - if (ip0 <= ilimit) { - /* Fill Table */ - assert(base+current0+2 > istart); /* check base overflow */ - hashTable[ZSTD_hashPtr(base+current0+2, hlog, mls)] = current0+2; /* here because current+2 could be > iend-8 */ - hashTable[ZSTD_hashPtr(ip0-2, hlog, mls)] = (U32)(ip0-2-base); - - if (rep_offset2 > 0) { /* rep_offset2==0 means rep_offset2 is invalidated */ - while ( (ip0 <= ilimit) && (MEM_read32(ip0) == MEM_read32(ip0 - rep_offset2)) ) { - /* store sequence */ - size_t const rLength = ZSTD_count(ip0+4, ip0+4-rep_offset2, iend) + 4; - { U32 const tmpOff = rep_offset2; rep_offset2 = rep_offset1; rep_offset1 = tmpOff; } /* swap rep_offset2 <=> rep_offset1 */ - hashTable[ZSTD_hashPtr(ip0, hlog, mls)] = (U32)(ip0-base); - ip0 += rLength; - ZSTD_storeSeq(seqStore, 0 /*litLen*/, anchor, iend, STORE_REPCODE_1, rLength); - anchor = ip0; - continue; /* faster when present (confirmed on gcc-8) ... (?) */ - } } } - - goto _start; -} - -#define ZSTD_GEN_FAST_FN(dictMode, mls, step) \ - static size_t ZSTD_compressBlock_fast_##dictMode##_##mls##_##step( \ - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], \ - void const* src, size_t srcSize) \ - { \ - return ZSTD_compressBlock_fast_##dictMode##_generic(ms, seqStore, rep, src, srcSize, mls, step); \ - } - -ZSTD_GEN_FAST_FN(noDict, 4, 1) -ZSTD_GEN_FAST_FN(noDict, 5, 1) -ZSTD_GEN_FAST_FN(noDict, 6, 1) -ZSTD_GEN_FAST_FN(noDict, 7, 1) - -ZSTD_GEN_FAST_FN(noDict, 4, 0) -ZSTD_GEN_FAST_FN(noDict, 5, 0) -ZSTD_GEN_FAST_FN(noDict, 6, 0) -ZSTD_GEN_FAST_FN(noDict, 7, 0) - -size_t ZSTD_compressBlock_fast( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - U32 const mls = ms->cParams.minMatch; - assert(ms->dictMatchState == NULL); - if (ms->cParams.targetLength > 1) { - switch(mls) - { - default: /* includes case 3 */ - case 4 : - return ZSTD_compressBlock_fast_noDict_4_1(ms, seqStore, rep, src, srcSize); - case 5 : - return ZSTD_compressBlock_fast_noDict_5_1(ms, seqStore, rep, src, srcSize); - case 6 : - return ZSTD_compressBlock_fast_noDict_6_1(ms, seqStore, rep, src, srcSize); - case 7 : - return ZSTD_compressBlock_fast_noDict_7_1(ms, seqStore, rep, src, srcSize); - } - } else { - switch(mls) - { - default: /* includes case 3 */ - case 4 : - return ZSTD_compressBlock_fast_noDict_4_0(ms, seqStore, rep, src, srcSize); - case 5 : - return ZSTD_compressBlock_fast_noDict_5_0(ms, seqStore, rep, src, srcSize); - case 6 : - return ZSTD_compressBlock_fast_noDict_6_0(ms, seqStore, rep, src, srcSize); - case 7 : - return ZSTD_compressBlock_fast_noDict_7_0(ms, seqStore, rep, src, srcSize); - } - - } -} - -FORCE_INLINE_TEMPLATE -size_t ZSTD_compressBlock_fast_dictMatchState_generic( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize, U32 const mls, U32 const hasStep) -{ - const ZSTD_compressionParameters* const cParams = &ms->cParams; - U32* const hashTable = ms->hashTable; - U32 const hlog = cParams->hashLog; - /* support stepSize of 0 */ - U32 const stepSize = cParams->targetLength + !(cParams->targetLength); - const BYTE* const base = ms->window.base; - const BYTE* const istart = (const BYTE*)src; - const BYTE* ip = istart; - const BYTE* anchor = istart; - const U32 prefixStartIndex = ms->window.dictLimit; - const BYTE* const prefixStart = base + prefixStartIndex; - const BYTE* const iend = istart + srcSize; - const BYTE* const ilimit = iend - HASH_READ_SIZE; - U32 offset_1=rep[0], offset_2=rep[1]; - U32 offsetSaved = 0; - - const ZSTD_matchState_t* const dms = ms->dictMatchState; - const ZSTD_compressionParameters* const dictCParams = &dms->cParams ; - const U32* const dictHashTable = dms->hashTable; - const U32 dictStartIndex = dms->window.dictLimit; - const BYTE* const dictBase = dms->window.base; - const BYTE* const dictStart = dictBase + dictStartIndex; - const BYTE* const dictEnd = dms->window.nextSrc; - const U32 dictIndexDelta = prefixStartIndex - (U32)(dictEnd - dictBase); - const U32 dictAndPrefixLength = (U32)(ip - prefixStart + dictEnd - dictStart); - const U32 dictHLog = dictCParams->hashLog; - - /* if a dictionary is still attached, it necessarily means that - * it is within window size. So we just check it. */ - const U32 maxDistance = 1U << cParams->windowLog; - const U32 endIndex = (U32)((size_t)(ip - base) + srcSize); - assert(endIndex - prefixStartIndex <= maxDistance); - (void)maxDistance; (void)endIndex; /* these variables are not used when assert() is disabled */ - - (void)hasStep; /* not currently specialized on whether it's accelerated */ - - /* ensure there will be no underflow - * when translating a dict index into a local index */ - assert(prefixStartIndex >= (U32)(dictEnd - dictBase)); - - /* init */ - DEBUGLOG(5, "ZSTD_compressBlock_fast_dictMatchState_generic"); - ip += (dictAndPrefixLength == 0); - /* dictMatchState repCode checks don't currently handle repCode == 0 - * disabling. */ - assert(offset_1 <= dictAndPrefixLength); - assert(offset_2 <= dictAndPrefixLength); - - /* Main Search Loop */ - while (ip < ilimit) { /* < instead of <=, because repcode check at (ip+1) */ - size_t mLength; - size_t const h = ZSTD_hashPtr(ip, hlog, mls); - U32 const curr = (U32)(ip-base); - U32 const matchIndex = hashTable[h]; - const BYTE* match = base + matchIndex; - const U32 repIndex = curr + 1 - offset_1; - const BYTE* repMatch = (repIndex < prefixStartIndex) ? - dictBase + (repIndex - dictIndexDelta) : - base + repIndex; - hashTable[h] = curr; /* update hash table */ - - if ( ((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow : ensure repIndex isn't overlapping dict + prefix */ - && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) { - const BYTE* const repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend; - mLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixStart) + 4; - ip++; - ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_REPCODE_1, mLength); - } else if ( (matchIndex <= prefixStartIndex) ) { - size_t const dictHash = ZSTD_hashPtr(ip, dictHLog, mls); - U32 const dictMatchIndex = dictHashTable[dictHash]; - const BYTE* dictMatch = dictBase + dictMatchIndex; - if (dictMatchIndex <= dictStartIndex || - MEM_read32(dictMatch) != MEM_read32(ip)) { - assert(stepSize >= 1); - ip += ((ip-anchor) >> kSearchStrength) + stepSize; - continue; - } else { - /* found a dict match */ - U32 const offset = (U32)(curr-dictMatchIndex-dictIndexDelta); - mLength = ZSTD_count_2segments(ip+4, dictMatch+4, iend, dictEnd, prefixStart) + 4; - while (((ip>anchor) & (dictMatch>dictStart)) - && (ip[-1] == dictMatch[-1])) { - ip--; dictMatch--; mLength++; - } /* catch up */ - offset_2 = offset_1; - offset_1 = offset; - ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength); - } - } else if (MEM_read32(match) != MEM_read32(ip)) { - /* it's not a match, and we're not going to check the dictionary */ - assert(stepSize >= 1); - ip += ((ip-anchor) >> kSearchStrength) + stepSize; - continue; - } else { - /* found a regular match */ - U32 const offset = (U32)(ip-match); - mLength = ZSTD_count(ip+4, match+4, iend) + 4; - while (((ip>anchor) & (match>prefixStart)) - && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */ - offset_2 = offset_1; - offset_1 = offset; - ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength); - } - - /* match found */ - ip += mLength; - anchor = ip; - - if (ip <= ilimit) { - /* Fill Table */ - assert(base+curr+2 > istart); /* check base overflow */ - hashTable[ZSTD_hashPtr(base+curr+2, hlog, mls)] = curr+2; /* here because curr+2 could be > iend-8 */ - hashTable[ZSTD_hashPtr(ip-2, hlog, mls)] = (U32)(ip-2-base); - - /* check immediate repcode */ - while (ip <= ilimit) { - U32 const current2 = (U32)(ip-base); - U32 const repIndex2 = current2 - offset_2; - const BYTE* repMatch2 = repIndex2 < prefixStartIndex ? - dictBase - dictIndexDelta + repIndex2 : - base + repIndex2; - if ( ((U32)((prefixStartIndex-1) - (U32)repIndex2) >= 3 /* intentional overflow */) - && (MEM_read32(repMatch2) == MEM_read32(ip)) ) { - const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend; - size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4; - U32 tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; /* swap offset_2 <=> offset_1 */ - ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, repLength2); - hashTable[ZSTD_hashPtr(ip, hlog, mls)] = current2; - ip += repLength2; - anchor = ip; - continue; - } - break; - } - } - } - - /* save reps for next block */ - rep[0] = offset_1 ? offset_1 : offsetSaved; - rep[1] = offset_2 ? offset_2 : offsetSaved; - - /* Return the last literals size */ - return (size_t)(iend - anchor); -} - - -ZSTD_GEN_FAST_FN(dictMatchState, 4, 0) -ZSTD_GEN_FAST_FN(dictMatchState, 5, 0) -ZSTD_GEN_FAST_FN(dictMatchState, 6, 0) -ZSTD_GEN_FAST_FN(dictMatchState, 7, 0) - -size_t ZSTD_compressBlock_fast_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - U32 const mls = ms->cParams.minMatch; - assert(ms->dictMatchState != NULL); - switch(mls) - { - default: /* includes case 3 */ - case 4 : - return ZSTD_compressBlock_fast_dictMatchState_4_0(ms, seqStore, rep, src, srcSize); - case 5 : - return ZSTD_compressBlock_fast_dictMatchState_5_0(ms, seqStore, rep, src, srcSize); - case 6 : - return ZSTD_compressBlock_fast_dictMatchState_6_0(ms, seqStore, rep, src, srcSize); - case 7 : - return ZSTD_compressBlock_fast_dictMatchState_7_0(ms, seqStore, rep, src, srcSize); - } -} - - -static size_t ZSTD_compressBlock_fast_extDict_generic( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize, U32 const mls, U32 const hasStep) -{ - const ZSTD_compressionParameters* const cParams = &ms->cParams; - U32* const hashTable = ms->hashTable; - U32 const hlog = cParams->hashLog; - /* support stepSize of 0 */ - U32 const stepSize = cParams->targetLength + !(cParams->targetLength); - const BYTE* const base = ms->window.base; - const BYTE* const dictBase = ms->window.dictBase; - const BYTE* const istart = (const BYTE*)src; - const BYTE* ip = istart; - const BYTE* anchor = istart; - const U32 endIndex = (U32)((size_t)(istart - base) + srcSize); - const U32 lowLimit = ZSTD_getLowestMatchIndex(ms, endIndex, cParams->windowLog); - const U32 dictStartIndex = lowLimit; - const BYTE* const dictStart = dictBase + dictStartIndex; - const U32 dictLimit = ms->window.dictLimit; - const U32 prefixStartIndex = dictLimit < lowLimit ? lowLimit : dictLimit; - const BYTE* const prefixStart = base + prefixStartIndex; - const BYTE* const dictEnd = dictBase + prefixStartIndex; - const BYTE* const iend = istart + srcSize; - const BYTE* const ilimit = iend - 8; - U32 offset_1=rep[0], offset_2=rep[1]; - - (void)hasStep; /* not currently specialized on whether it's accelerated */ - - DEBUGLOG(5, "ZSTD_compressBlock_fast_extDict_generic (offset_1=%u)", offset_1); - - /* switch to "regular" variant if extDict is invalidated due to maxDistance */ - if (prefixStartIndex == dictStartIndex) - return ZSTD_compressBlock_fast(ms, seqStore, rep, src, srcSize); - - /* Search Loop */ - while (ip < ilimit) { /* < instead of <=, because (ip+1) */ - const size_t h = ZSTD_hashPtr(ip, hlog, mls); - const U32 matchIndex = hashTable[h]; - const BYTE* const matchBase = matchIndex < prefixStartIndex ? dictBase : base; - const BYTE* match = matchBase + matchIndex; - const U32 curr = (U32)(ip-base); - const U32 repIndex = curr + 1 - offset_1; - const BYTE* const repBase = repIndex < prefixStartIndex ? dictBase : base; - const BYTE* const repMatch = repBase + repIndex; - hashTable[h] = curr; /* update hash table */ - DEBUGLOG(7, "offset_1 = %u , curr = %u", offset_1, curr); - - if ( ( ((U32)((prefixStartIndex-1) - repIndex) >= 3) /* intentional underflow */ - & (offset_1 <= curr+1 - dictStartIndex) ) /* note: we are searching at curr+1 */ - && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) { - const BYTE* const repMatchEnd = repIndex < prefixStartIndex ? dictEnd : iend; - size_t const rLength = ZSTD_count_2segments(ip+1 +4, repMatch +4, iend, repMatchEnd, prefixStart) + 4; - ip++; - ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_REPCODE_1, rLength); - ip += rLength; - anchor = ip; - } else { - if ( (matchIndex < dictStartIndex) || - (MEM_read32(match) != MEM_read32(ip)) ) { - assert(stepSize >= 1); - ip += ((ip-anchor) >> kSearchStrength) + stepSize; - continue; - } - { const BYTE* const matchEnd = matchIndex < prefixStartIndex ? dictEnd : iend; - const BYTE* const lowMatchPtr = matchIndex < prefixStartIndex ? dictStart : prefixStart; - U32 const offset = curr - matchIndex; - size_t mLength = ZSTD_count_2segments(ip+4, match+4, iend, matchEnd, prefixStart) + 4; - while (((ip>anchor) & (match>lowMatchPtr)) && (ip[-1] == match[-1])) { ip--; match--; mLength++; } /* catch up */ - offset_2 = offset_1; offset_1 = offset; /* update offset history */ - ZSTD_storeSeq(seqStore, (size_t)(ip-anchor), anchor, iend, STORE_OFFSET(offset), mLength); - ip += mLength; - anchor = ip; - } } - - if (ip <= ilimit) { - /* Fill Table */ - hashTable[ZSTD_hashPtr(base+curr+2, hlog, mls)] = curr+2; - hashTable[ZSTD_hashPtr(ip-2, hlog, mls)] = (U32)(ip-2-base); - /* check immediate repcode */ - while (ip <= ilimit) { - U32 const current2 = (U32)(ip-base); - U32 const repIndex2 = current2 - offset_2; - const BYTE* const repMatch2 = repIndex2 < prefixStartIndex ? dictBase + repIndex2 : base + repIndex2; - if ( (((U32)((prefixStartIndex-1) - repIndex2) >= 3) & (offset_2 <= curr - dictStartIndex)) /* intentional overflow */ - && (MEM_read32(repMatch2) == MEM_read32(ip)) ) { - const BYTE* const repEnd2 = repIndex2 < prefixStartIndex ? dictEnd : iend; - size_t const repLength2 = ZSTD_count_2segments(ip+4, repMatch2+4, iend, repEnd2, prefixStart) + 4; - { U32 const tmpOffset = offset_2; offset_2 = offset_1; offset_1 = tmpOffset; } /* swap offset_2 <=> offset_1 */ - ZSTD_storeSeq(seqStore, 0 /*litlen*/, anchor, iend, STORE_REPCODE_1, repLength2); - hashTable[ZSTD_hashPtr(ip, hlog, mls)] = current2; - ip += repLength2; - anchor = ip; - continue; - } - break; - } } } - - /* save reps for next block */ - rep[0] = offset_1; - rep[1] = offset_2; - - /* Return the last literals size */ - return (size_t)(iend - anchor); -} - -ZSTD_GEN_FAST_FN(extDict, 4, 0) -ZSTD_GEN_FAST_FN(extDict, 5, 0) -ZSTD_GEN_FAST_FN(extDict, 6, 0) -ZSTD_GEN_FAST_FN(extDict, 7, 0) - -size_t ZSTD_compressBlock_fast_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - U32 const mls = ms->cParams.minMatch; - switch(mls) - { - default: /* includes case 3 */ - case 4 : - return ZSTD_compressBlock_fast_extDict_4_0(ms, seqStore, rep, src, srcSize); - case 5 : - return ZSTD_compressBlock_fast_extDict_5_0(ms, seqStore, rep, src, srcSize); - case 6 : - return ZSTD_compressBlock_fast_extDict_6_0(ms, seqStore, rep, src, srcSize); - case 7 : - return ZSTD_compressBlock_fast_extDict_7_0(ms, seqStore, rep, src, srcSize); - } -} diff --git a/dep/zstd/lib/compress/zstd_fast.h b/dep/zstd/lib/compress/zstd_fast.h deleted file mode 100644 index 0d4a0c109..000000000 --- a/dep/zstd/lib/compress/zstd_fast.h +++ /dev/null @@ -1,37 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_FAST_H -#define ZSTD_FAST_H - -#if defined (__cplusplus) -extern "C" { -#endif - -#include "../common/mem.h" /* U32 */ -#include "zstd_compress_internal.h" - -void ZSTD_fillHashTable(ZSTD_matchState_t* ms, - void const* end, ZSTD_dictTableLoadMethod_e dtlm); -size_t ZSTD_compressBlock_fast( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_fast_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_fast_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); - -#if defined (__cplusplus) -} -#endif - -#endif /* ZSTD_FAST_H */ diff --git a/dep/zstd/lib/compress/zstd_lazy.c b/dep/zstd/lib/compress/zstd_lazy.c deleted file mode 100644 index 2e38dcb46..000000000 --- a/dep/zstd/lib/compress/zstd_lazy.c +++ /dev/null @@ -1,2104 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#include "zstd_compress_internal.h" -#include "zstd_lazy.h" - - -/*-************************************* -* Binary Tree search -***************************************/ - -static void -ZSTD_updateDUBT(ZSTD_matchState_t* ms, - const BYTE* ip, const BYTE* iend, - U32 mls) -{ - const ZSTD_compressionParameters* const cParams = &ms->cParams; - U32* const hashTable = ms->hashTable; - U32 const hashLog = cParams->hashLog; - - U32* const bt = ms->chainTable; - U32 const btLog = cParams->chainLog - 1; - U32 const btMask = (1 << btLog) - 1; - - const BYTE* const base = ms->window.base; - U32 const target = (U32)(ip - base); - U32 idx = ms->nextToUpdate; - - if (idx != target) - DEBUGLOG(7, "ZSTD_updateDUBT, from %u to %u (dictLimit:%u)", - idx, target, ms->window.dictLimit); - assert(ip + 8 <= iend); /* condition for ZSTD_hashPtr */ - (void)iend; - - assert(idx >= ms->window.dictLimit); /* condition for valid base+idx */ - for ( ; idx < target ; idx++) { - size_t const h = ZSTD_hashPtr(base + idx, hashLog, mls); /* assumption : ip + 8 <= iend */ - U32 const matchIndex = hashTable[h]; - - U32* const nextCandidatePtr = bt + 2*(idx&btMask); - U32* const sortMarkPtr = nextCandidatePtr + 1; - - DEBUGLOG(8, "ZSTD_updateDUBT: insert %u", idx); - hashTable[h] = idx; /* Update Hash Table */ - *nextCandidatePtr = matchIndex; /* update BT like a chain */ - *sortMarkPtr = ZSTD_DUBT_UNSORTED_MARK; - } - ms->nextToUpdate = target; -} - - -/** ZSTD_insertDUBT1() : - * sort one already inserted but unsorted position - * assumption : curr >= btlow == (curr - btmask) - * doesn't fail */ -static void -ZSTD_insertDUBT1(const ZSTD_matchState_t* ms, - U32 curr, const BYTE* inputEnd, - U32 nbCompares, U32 btLow, - const ZSTD_dictMode_e dictMode) -{ - const ZSTD_compressionParameters* const cParams = &ms->cParams; - U32* const bt = ms->chainTable; - U32 const btLog = cParams->chainLog - 1; - U32 const btMask = (1 << btLog) - 1; - size_t commonLengthSmaller=0, commonLengthLarger=0; - const BYTE* const base = ms->window.base; - const BYTE* const dictBase = ms->window.dictBase; - const U32 dictLimit = ms->window.dictLimit; - const BYTE* const ip = (curr>=dictLimit) ? base + curr : dictBase + curr; - const BYTE* const iend = (curr>=dictLimit) ? inputEnd : dictBase + dictLimit; - const BYTE* const dictEnd = dictBase + dictLimit; - const BYTE* const prefixStart = base + dictLimit; - const BYTE* match; - U32* smallerPtr = bt + 2*(curr&btMask); - U32* largerPtr = smallerPtr + 1; - U32 matchIndex = *smallerPtr; /* this candidate is unsorted : next sorted candidate is reached through *smallerPtr, while *largerPtr contains previous unsorted candidate (which is already saved and can be overwritten) */ - U32 dummy32; /* to be nullified at the end */ - U32 const windowValid = ms->window.lowLimit; - U32 const maxDistance = 1U << cParams->windowLog; - U32 const windowLow = (curr - windowValid > maxDistance) ? curr - maxDistance : windowValid; - - - DEBUGLOG(8, "ZSTD_insertDUBT1(%u) (dictLimit=%u, lowLimit=%u)", - curr, dictLimit, windowLow); - assert(curr >= btLow); - assert(ip < iend); /* condition for ZSTD_count */ - - for (; nbCompares && (matchIndex > windowLow); --nbCompares) { - U32* const nextPtr = bt + 2*(matchIndex & btMask); - size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */ - assert(matchIndex < curr); - /* note : all candidates are now supposed sorted, - * but it's still possible to have nextPtr[1] == ZSTD_DUBT_UNSORTED_MARK - * when a real index has the same value as ZSTD_DUBT_UNSORTED_MARK */ - - if ( (dictMode != ZSTD_extDict) - || (matchIndex+matchLength >= dictLimit) /* both in current segment*/ - || (curr < dictLimit) /* both in extDict */) { - const BYTE* const mBase = ( (dictMode != ZSTD_extDict) - || (matchIndex+matchLength >= dictLimit)) ? - base : dictBase; - assert( (matchIndex+matchLength >= dictLimit) /* might be wrong if extDict is incorrectly set to 0 */ - || (curr < dictLimit) ); - match = mBase + matchIndex; - matchLength += ZSTD_count(ip+matchLength, match+matchLength, iend); - } else { - match = dictBase + matchIndex; - matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iend, dictEnd, prefixStart); - if (matchIndex+matchLength >= dictLimit) - match = base + matchIndex; /* preparation for next read of match[matchLength] */ - } - - DEBUGLOG(8, "ZSTD_insertDUBT1: comparing %u with %u : found %u common bytes ", - curr, matchIndex, (U32)matchLength); - - if (ip+matchLength == iend) { /* equal : no way to know if inf or sup */ - break; /* drop , to guarantee consistency ; miss a bit of compression, but other solutions can corrupt tree */ - } - - if (match[matchLength] < ip[matchLength]) { /* necessarily within buffer */ - /* match is smaller than current */ - *smallerPtr = matchIndex; /* update smaller idx */ - commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */ - if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop searching */ - DEBUGLOG(8, "ZSTD_insertDUBT1: %u (>btLow=%u) is smaller : next => %u", - matchIndex, btLow, nextPtr[1]); - smallerPtr = nextPtr+1; /* new "candidate" => larger than match, which was smaller than target */ - matchIndex = nextPtr[1]; /* new matchIndex, larger than previous and closer to current */ - } else { - /* match is larger than current */ - *largerPtr = matchIndex; - commonLengthLarger = matchLength; - if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop searching */ - DEBUGLOG(8, "ZSTD_insertDUBT1: %u (>btLow=%u) is larger => %u", - matchIndex, btLow, nextPtr[0]); - largerPtr = nextPtr; - matchIndex = nextPtr[0]; - } } - - *smallerPtr = *largerPtr = 0; -} - - -static size_t -ZSTD_DUBT_findBetterDictMatch ( - const ZSTD_matchState_t* ms, - const BYTE* const ip, const BYTE* const iend, - size_t* offsetPtr, - size_t bestLength, - U32 nbCompares, - U32 const mls, - const ZSTD_dictMode_e dictMode) -{ - const ZSTD_matchState_t * const dms = ms->dictMatchState; - const ZSTD_compressionParameters* const dmsCParams = &dms->cParams; - const U32 * const dictHashTable = dms->hashTable; - U32 const hashLog = dmsCParams->hashLog; - size_t const h = ZSTD_hashPtr(ip, hashLog, mls); - U32 dictMatchIndex = dictHashTable[h]; - - const BYTE* const base = ms->window.base; - const BYTE* const prefixStart = base + ms->window.dictLimit; - U32 const curr = (U32)(ip-base); - const BYTE* const dictBase = dms->window.base; - const BYTE* const dictEnd = dms->window.nextSrc; - U32 const dictHighLimit = (U32)(dms->window.nextSrc - dms->window.base); - U32 const dictLowLimit = dms->window.lowLimit; - U32 const dictIndexDelta = ms->window.lowLimit - dictHighLimit; - - U32* const dictBt = dms->chainTable; - U32 const btLog = dmsCParams->chainLog - 1; - U32 const btMask = (1 << btLog) - 1; - U32 const btLow = (btMask >= dictHighLimit - dictLowLimit) ? dictLowLimit : dictHighLimit - btMask; - - size_t commonLengthSmaller=0, commonLengthLarger=0; - - (void)dictMode; - assert(dictMode == ZSTD_dictMatchState); - - for (; nbCompares && (dictMatchIndex > dictLowLimit); --nbCompares) { - U32* const nextPtr = dictBt + 2*(dictMatchIndex & btMask); - size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */ - const BYTE* match = dictBase + dictMatchIndex; - matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iend, dictEnd, prefixStart); - if (dictMatchIndex+matchLength >= dictHighLimit) - match = base + dictMatchIndex + dictIndexDelta; /* to prepare for next usage of match[matchLength] */ - - if (matchLength > bestLength) { - U32 matchIndex = dictMatchIndex + dictIndexDelta; - if ( (4*(int)(matchLength-bestLength)) > (int)(ZSTD_highbit32(curr-matchIndex+1) - ZSTD_highbit32((U32)offsetPtr[0]+1)) ) { - DEBUGLOG(9, "ZSTD_DUBT_findBetterDictMatch(%u) : found better match length %u -> %u and offsetCode %u -> %u (dictMatchIndex %u, matchIndex %u)", - curr, (U32)bestLength, (U32)matchLength, (U32)*offsetPtr, STORE_OFFSET(curr - matchIndex), dictMatchIndex, matchIndex); - bestLength = matchLength, *offsetPtr = STORE_OFFSET(curr - matchIndex); - } - if (ip+matchLength == iend) { /* reached end of input : ip[matchLength] is not valid, no way to know if it's larger or smaller than match */ - break; /* drop, to guarantee consistency (miss a little bit of compression) */ - } - } - - if (match[matchLength] < ip[matchLength]) { - if (dictMatchIndex <= btLow) { break; } /* beyond tree size, stop the search */ - commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */ - dictMatchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */ - } else { - /* match is larger than current */ - if (dictMatchIndex <= btLow) { break; } /* beyond tree size, stop the search */ - commonLengthLarger = matchLength; - dictMatchIndex = nextPtr[0]; - } - } - - if (bestLength >= MINMATCH) { - U32 const mIndex = curr - (U32)STORED_OFFSET(*offsetPtr); (void)mIndex; - DEBUGLOG(8, "ZSTD_DUBT_findBetterDictMatch(%u) : found match of length %u and offsetCode %u (pos %u)", - curr, (U32)bestLength, (U32)*offsetPtr, mIndex); - } - return bestLength; - -} - - -static size_t -ZSTD_DUBT_findBestMatch(ZSTD_matchState_t* ms, - const BYTE* const ip, const BYTE* const iend, - size_t* offsetPtr, - U32 const mls, - const ZSTD_dictMode_e dictMode) -{ - const ZSTD_compressionParameters* const cParams = &ms->cParams; - U32* const hashTable = ms->hashTable; - U32 const hashLog = cParams->hashLog; - size_t const h = ZSTD_hashPtr(ip, hashLog, mls); - U32 matchIndex = hashTable[h]; - - const BYTE* const base = ms->window.base; - U32 const curr = (U32)(ip-base); - U32 const windowLow = ZSTD_getLowestMatchIndex(ms, curr, cParams->windowLog); - - U32* const bt = ms->chainTable; - U32 const btLog = cParams->chainLog - 1; - U32 const btMask = (1 << btLog) - 1; - U32 const btLow = (btMask >= curr) ? 0 : curr - btMask; - U32 const unsortLimit = MAX(btLow, windowLow); - - U32* nextCandidate = bt + 2*(matchIndex&btMask); - U32* unsortedMark = bt + 2*(matchIndex&btMask) + 1; - U32 nbCompares = 1U << cParams->searchLog; - U32 nbCandidates = nbCompares; - U32 previousCandidate = 0; - - DEBUGLOG(7, "ZSTD_DUBT_findBestMatch (%u) ", curr); - assert(ip <= iend-8); /* required for h calculation */ - assert(dictMode != ZSTD_dedicatedDictSearch); - - /* reach end of unsorted candidates list */ - while ( (matchIndex > unsortLimit) - && (*unsortedMark == ZSTD_DUBT_UNSORTED_MARK) - && (nbCandidates > 1) ) { - DEBUGLOG(8, "ZSTD_DUBT_findBestMatch: candidate %u is unsorted", - matchIndex); - *unsortedMark = previousCandidate; /* the unsortedMark becomes a reversed chain, to move up back to original position */ - previousCandidate = matchIndex; - matchIndex = *nextCandidate; - nextCandidate = bt + 2*(matchIndex&btMask); - unsortedMark = bt + 2*(matchIndex&btMask) + 1; - nbCandidates --; - } - - /* nullify last candidate if it's still unsorted - * simplification, detrimental to compression ratio, beneficial for speed */ - if ( (matchIndex > unsortLimit) - && (*unsortedMark==ZSTD_DUBT_UNSORTED_MARK) ) { - DEBUGLOG(7, "ZSTD_DUBT_findBestMatch: nullify last unsorted candidate %u", - matchIndex); - *nextCandidate = *unsortedMark = 0; - } - - /* batch sort stacked candidates */ - matchIndex = previousCandidate; - while (matchIndex) { /* will end on matchIndex == 0 */ - U32* const nextCandidateIdxPtr = bt + 2*(matchIndex&btMask) + 1; - U32 const nextCandidateIdx = *nextCandidateIdxPtr; - ZSTD_insertDUBT1(ms, matchIndex, iend, - nbCandidates, unsortLimit, dictMode); - matchIndex = nextCandidateIdx; - nbCandidates++; - } - - /* find longest match */ - { size_t commonLengthSmaller = 0, commonLengthLarger = 0; - const BYTE* const dictBase = ms->window.dictBase; - const U32 dictLimit = ms->window.dictLimit; - const BYTE* const dictEnd = dictBase + dictLimit; - const BYTE* const prefixStart = base + dictLimit; - U32* smallerPtr = bt + 2*(curr&btMask); - U32* largerPtr = bt + 2*(curr&btMask) + 1; - U32 matchEndIdx = curr + 8 + 1; - U32 dummy32; /* to be nullified at the end */ - size_t bestLength = 0; - - matchIndex = hashTable[h]; - hashTable[h] = curr; /* Update Hash Table */ - - for (; nbCompares && (matchIndex > windowLow); --nbCompares) { - U32* const nextPtr = bt + 2*(matchIndex & btMask); - size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */ - const BYTE* match; - - if ((dictMode != ZSTD_extDict) || (matchIndex+matchLength >= dictLimit)) { - match = base + matchIndex; - matchLength += ZSTD_count(ip+matchLength, match+matchLength, iend); - } else { - match = dictBase + matchIndex; - matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iend, dictEnd, prefixStart); - if (matchIndex+matchLength >= dictLimit) - match = base + matchIndex; /* to prepare for next usage of match[matchLength] */ - } - - if (matchLength > bestLength) { - if (matchLength > matchEndIdx - matchIndex) - matchEndIdx = matchIndex + (U32)matchLength; - if ( (4*(int)(matchLength-bestLength)) > (int)(ZSTD_highbit32(curr-matchIndex+1) - ZSTD_highbit32((U32)offsetPtr[0]+1)) ) - bestLength = matchLength, *offsetPtr = STORE_OFFSET(curr - matchIndex); - if (ip+matchLength == iend) { /* equal : no way to know if inf or sup */ - if (dictMode == ZSTD_dictMatchState) { - nbCompares = 0; /* in addition to avoiding checking any - * further in this loop, make sure we - * skip checking in the dictionary. */ - } - break; /* drop, to guarantee consistency (miss a little bit of compression) */ - } - } - - if (match[matchLength] < ip[matchLength]) { - /* match is smaller than current */ - *smallerPtr = matchIndex; /* update smaller idx */ - commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */ - if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop the search */ - smallerPtr = nextPtr+1; /* new "smaller" => larger of match */ - matchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */ - } else { - /* match is larger than current */ - *largerPtr = matchIndex; - commonLengthLarger = matchLength; - if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop the search */ - largerPtr = nextPtr; - matchIndex = nextPtr[0]; - } } - - *smallerPtr = *largerPtr = 0; - - assert(nbCompares <= (1U << ZSTD_SEARCHLOG_MAX)); /* Check we haven't underflowed. */ - if (dictMode == ZSTD_dictMatchState && nbCompares) { - bestLength = ZSTD_DUBT_findBetterDictMatch( - ms, ip, iend, - offsetPtr, bestLength, nbCompares, - mls, dictMode); - } - - assert(matchEndIdx > curr+8); /* ensure nextToUpdate is increased */ - ms->nextToUpdate = matchEndIdx - 8; /* skip repetitive patterns */ - if (bestLength >= MINMATCH) { - U32 const mIndex = curr - (U32)STORED_OFFSET(*offsetPtr); (void)mIndex; - DEBUGLOG(8, "ZSTD_DUBT_findBestMatch(%u) : found match of length %u and offsetCode %u (pos %u)", - curr, (U32)bestLength, (U32)*offsetPtr, mIndex); - } - return bestLength; - } -} - - -/** ZSTD_BtFindBestMatch() : Tree updater, providing best match */ -FORCE_INLINE_TEMPLATE size_t -ZSTD_BtFindBestMatch( ZSTD_matchState_t* ms, - const BYTE* const ip, const BYTE* const iLimit, - size_t* offsetPtr, - const U32 mls /* template */, - const ZSTD_dictMode_e dictMode) -{ - DEBUGLOG(7, "ZSTD_BtFindBestMatch"); - if (ip < ms->window.base + ms->nextToUpdate) return 0; /* skipped area */ - ZSTD_updateDUBT(ms, ip, iLimit, mls); - return ZSTD_DUBT_findBestMatch(ms, ip, iLimit, offsetPtr, mls, dictMode); -} - -/*********************************** -* Dedicated dict search -***********************************/ - -void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_matchState_t* ms, const BYTE* const ip) -{ - const BYTE* const base = ms->window.base; - U32 const target = (U32)(ip - base); - U32* const hashTable = ms->hashTable; - U32* const chainTable = ms->chainTable; - U32 const chainSize = 1 << ms->cParams.chainLog; - U32 idx = ms->nextToUpdate; - U32 const minChain = chainSize < target - idx ? target - chainSize : idx; - U32 const bucketSize = 1 << ZSTD_LAZY_DDSS_BUCKET_LOG; - U32 const cacheSize = bucketSize - 1; - U32 const chainAttempts = (1 << ms->cParams.searchLog) - cacheSize; - U32 const chainLimit = chainAttempts > 255 ? 255 : chainAttempts; - - /* We know the hashtable is oversized by a factor of `bucketSize`. - * We are going to temporarily pretend `bucketSize == 1`, keeping only a - * single entry. We will use the rest of the space to construct a temporary - * chaintable. - */ - U32 const hashLog = ms->cParams.hashLog - ZSTD_LAZY_DDSS_BUCKET_LOG; - U32* const tmpHashTable = hashTable; - U32* const tmpChainTable = hashTable + ((size_t)1 << hashLog); - U32 const tmpChainSize = (U32)((1 << ZSTD_LAZY_DDSS_BUCKET_LOG) - 1) << hashLog; - U32 const tmpMinChain = tmpChainSize < target ? target - tmpChainSize : idx; - U32 hashIdx; - - assert(ms->cParams.chainLog <= 24); - assert(ms->cParams.hashLog > ms->cParams.chainLog); - assert(idx != 0); - assert(tmpMinChain <= minChain); - - /* fill conventional hash table and conventional chain table */ - for ( ; idx < target; idx++) { - U32 const h = (U32)ZSTD_hashPtr(base + idx, hashLog, ms->cParams.minMatch); - if (idx >= tmpMinChain) { - tmpChainTable[idx - tmpMinChain] = hashTable[h]; - } - tmpHashTable[h] = idx; - } - - /* sort chains into ddss chain table */ - { - U32 chainPos = 0; - for (hashIdx = 0; hashIdx < (1U << hashLog); hashIdx++) { - U32 count; - U32 countBeyondMinChain = 0; - U32 i = tmpHashTable[hashIdx]; - for (count = 0; i >= tmpMinChain && count < cacheSize; count++) { - /* skip through the chain to the first position that won't be - * in the hash cache bucket */ - if (i < minChain) { - countBeyondMinChain++; - } - i = tmpChainTable[i - tmpMinChain]; - } - if (count == cacheSize) { - for (count = 0; count < chainLimit;) { - if (i < minChain) { - if (!i || ++countBeyondMinChain > cacheSize) { - /* only allow pulling `cacheSize` number of entries - * into the cache or chainTable beyond `minChain`, - * to replace the entries pulled out of the - * chainTable into the cache. This lets us reach - * back further without increasing the total number - * of entries in the chainTable, guaranteeing the - * DDSS chain table will fit into the space - * allocated for the regular one. */ - break; - } - } - chainTable[chainPos++] = i; - count++; - if (i < tmpMinChain) { - break; - } - i = tmpChainTable[i - tmpMinChain]; - } - } else { - count = 0; - } - if (count) { - tmpHashTable[hashIdx] = ((chainPos - count) << 8) + count; - } else { - tmpHashTable[hashIdx] = 0; - } - } - assert(chainPos <= chainSize); /* I believe this is guaranteed... */ - } - - /* move chain pointers into the last entry of each hash bucket */ - for (hashIdx = (1 << hashLog); hashIdx; ) { - U32 const bucketIdx = --hashIdx << ZSTD_LAZY_DDSS_BUCKET_LOG; - U32 const chainPackedPointer = tmpHashTable[hashIdx]; - U32 i; - for (i = 0; i < cacheSize; i++) { - hashTable[bucketIdx + i] = 0; - } - hashTable[bucketIdx + bucketSize - 1] = chainPackedPointer; - } - - /* fill the buckets of the hash table */ - for (idx = ms->nextToUpdate; idx < target; idx++) { - U32 const h = (U32)ZSTD_hashPtr(base + idx, hashLog, ms->cParams.minMatch) - << ZSTD_LAZY_DDSS_BUCKET_LOG; - U32 i; - /* Shift hash cache down 1. */ - for (i = cacheSize - 1; i; i--) - hashTable[h + i] = hashTable[h + i - 1]; - hashTable[h] = idx; - } - - ms->nextToUpdate = target; -} - -/* Returns the longest match length found in the dedicated dict search structure. - * If none are longer than the argument ml, then ml will be returned. - */ -FORCE_INLINE_TEMPLATE -size_t ZSTD_dedicatedDictSearch_lazy_search(size_t* offsetPtr, size_t ml, U32 nbAttempts, - const ZSTD_matchState_t* const dms, - const BYTE* const ip, const BYTE* const iLimit, - const BYTE* const prefixStart, const U32 curr, - const U32 dictLimit, const size_t ddsIdx) { - const U32 ddsLowestIndex = dms->window.dictLimit; - const BYTE* const ddsBase = dms->window.base; - const BYTE* const ddsEnd = dms->window.nextSrc; - const U32 ddsSize = (U32)(ddsEnd - ddsBase); - const U32 ddsIndexDelta = dictLimit - ddsSize; - const U32 bucketSize = (1 << ZSTD_LAZY_DDSS_BUCKET_LOG); - const U32 bucketLimit = nbAttempts < bucketSize - 1 ? nbAttempts : bucketSize - 1; - U32 ddsAttempt; - U32 matchIndex; - - for (ddsAttempt = 0; ddsAttempt < bucketSize - 1; ddsAttempt++) { - PREFETCH_L1(ddsBase + dms->hashTable[ddsIdx + ddsAttempt]); - } - - { - U32 const chainPackedPointer = dms->hashTable[ddsIdx + bucketSize - 1]; - U32 const chainIndex = chainPackedPointer >> 8; - - PREFETCH_L1(&dms->chainTable[chainIndex]); - } - - for (ddsAttempt = 0; ddsAttempt < bucketLimit; ddsAttempt++) { - size_t currentMl=0; - const BYTE* match; - matchIndex = dms->hashTable[ddsIdx + ddsAttempt]; - match = ddsBase + matchIndex; - - if (!matchIndex) { - return ml; - } - - /* guaranteed by table construction */ - (void)ddsLowestIndex; - assert(matchIndex >= ddsLowestIndex); - assert(match+4 <= ddsEnd); - if (MEM_read32(match) == MEM_read32(ip)) { - /* assumption : matchIndex <= dictLimit-4 (by table construction) */ - currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, ddsEnd, prefixStart) + 4; - } - - /* save best solution */ - if (currentMl > ml) { - ml = currentMl; - *offsetPtr = STORE_OFFSET(curr - (matchIndex + ddsIndexDelta)); - if (ip+currentMl == iLimit) { - /* best possible, avoids read overflow on next attempt */ - return ml; - } - } - } - - { - U32 const chainPackedPointer = dms->hashTable[ddsIdx + bucketSize - 1]; - U32 chainIndex = chainPackedPointer >> 8; - U32 const chainLength = chainPackedPointer & 0xFF; - U32 const chainAttempts = nbAttempts - ddsAttempt; - U32 const chainLimit = chainAttempts > chainLength ? chainLength : chainAttempts; - U32 chainAttempt; - - for (chainAttempt = 0 ; chainAttempt < chainLimit; chainAttempt++) { - PREFETCH_L1(ddsBase + dms->chainTable[chainIndex + chainAttempt]); - } - - for (chainAttempt = 0 ; chainAttempt < chainLimit; chainAttempt++, chainIndex++) { - size_t currentMl=0; - const BYTE* match; - matchIndex = dms->chainTable[chainIndex]; - match = ddsBase + matchIndex; - - /* guaranteed by table construction */ - assert(matchIndex >= ddsLowestIndex); - assert(match+4 <= ddsEnd); - if (MEM_read32(match) == MEM_read32(ip)) { - /* assumption : matchIndex <= dictLimit-4 (by table construction) */ - currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, ddsEnd, prefixStart) + 4; - } - - /* save best solution */ - if (currentMl > ml) { - ml = currentMl; - *offsetPtr = STORE_OFFSET(curr - (matchIndex + ddsIndexDelta)); - if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */ - } - } - } - return ml; -} - - -/* ********************************* -* Hash Chain -***********************************/ -#define NEXT_IN_CHAIN(d, mask) chainTable[(d) & (mask)] - -/* Update chains up to ip (excluded) - Assumption : always within prefix (i.e. not within extDict) */ -FORCE_INLINE_TEMPLATE U32 ZSTD_insertAndFindFirstIndex_internal( - ZSTD_matchState_t* ms, - const ZSTD_compressionParameters* const cParams, - const BYTE* ip, U32 const mls) -{ - U32* const hashTable = ms->hashTable; - const U32 hashLog = cParams->hashLog; - U32* const chainTable = ms->chainTable; - const U32 chainMask = (1 << cParams->chainLog) - 1; - const BYTE* const base = ms->window.base; - const U32 target = (U32)(ip - base); - U32 idx = ms->nextToUpdate; - - while(idx < target) { /* catch up */ - size_t const h = ZSTD_hashPtr(base+idx, hashLog, mls); - NEXT_IN_CHAIN(idx, chainMask) = hashTable[h]; - hashTable[h] = idx; - idx++; - } - - ms->nextToUpdate = target; - return hashTable[ZSTD_hashPtr(ip, hashLog, mls)]; -} - -U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip) { - const ZSTD_compressionParameters* const cParams = &ms->cParams; - return ZSTD_insertAndFindFirstIndex_internal(ms, cParams, ip, ms->cParams.minMatch); -} - -/* inlining is important to hardwire a hot branch (template emulation) */ -FORCE_INLINE_TEMPLATE -size_t ZSTD_HcFindBestMatch( - ZSTD_matchState_t* ms, - const BYTE* const ip, const BYTE* const iLimit, - size_t* offsetPtr, - const U32 mls, const ZSTD_dictMode_e dictMode) -{ - const ZSTD_compressionParameters* const cParams = &ms->cParams; - U32* const chainTable = ms->chainTable; - const U32 chainSize = (1 << cParams->chainLog); - const U32 chainMask = chainSize-1; - const BYTE* const base = ms->window.base; - const BYTE* const dictBase = ms->window.dictBase; - const U32 dictLimit = ms->window.dictLimit; - const BYTE* const prefixStart = base + dictLimit; - const BYTE* const dictEnd = dictBase + dictLimit; - const U32 curr = (U32)(ip-base); - const U32 maxDistance = 1U << cParams->windowLog; - const U32 lowestValid = ms->window.lowLimit; - const U32 withinMaxDistance = (curr - lowestValid > maxDistance) ? curr - maxDistance : lowestValid; - const U32 isDictionary = (ms->loadedDictEnd != 0); - const U32 lowLimit = isDictionary ? lowestValid : withinMaxDistance; - const U32 minChain = curr > chainSize ? curr - chainSize : 0; - U32 nbAttempts = 1U << cParams->searchLog; - size_t ml=4-1; - - const ZSTD_matchState_t* const dms = ms->dictMatchState; - const U32 ddsHashLog = dictMode == ZSTD_dedicatedDictSearch - ? dms->cParams.hashLog - ZSTD_LAZY_DDSS_BUCKET_LOG : 0; - const size_t ddsIdx = dictMode == ZSTD_dedicatedDictSearch - ? ZSTD_hashPtr(ip, ddsHashLog, mls) << ZSTD_LAZY_DDSS_BUCKET_LOG : 0; - - U32 matchIndex; - - if (dictMode == ZSTD_dedicatedDictSearch) { - const U32* entry = &dms->hashTable[ddsIdx]; - PREFETCH_L1(entry); - } - - /* HC4 match finder */ - matchIndex = ZSTD_insertAndFindFirstIndex_internal(ms, cParams, ip, mls); - - for ( ; (matchIndex>=lowLimit) & (nbAttempts>0) ; nbAttempts--) { - size_t currentMl=0; - if ((dictMode != ZSTD_extDict) || matchIndex >= dictLimit) { - const BYTE* const match = base + matchIndex; - assert(matchIndex >= dictLimit); /* ensures this is true if dictMode != ZSTD_extDict */ - if (match[ml] == ip[ml]) /* potentially better */ - currentMl = ZSTD_count(ip, match, iLimit); - } else { - const BYTE* const match = dictBase + matchIndex; - assert(match+4 <= dictEnd); - if (MEM_read32(match) == MEM_read32(ip)) /* assumption : matchIndex <= dictLimit-4 (by table construction) */ - currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, dictEnd, prefixStart) + 4; - } - - /* save best solution */ - if (currentMl > ml) { - ml = currentMl; - *offsetPtr = STORE_OFFSET(curr - matchIndex); - if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */ - } - - if (matchIndex <= minChain) break; - matchIndex = NEXT_IN_CHAIN(matchIndex, chainMask); - } - - assert(nbAttempts <= (1U << ZSTD_SEARCHLOG_MAX)); /* Check we haven't underflowed. */ - if (dictMode == ZSTD_dedicatedDictSearch) { - ml = ZSTD_dedicatedDictSearch_lazy_search(offsetPtr, ml, nbAttempts, dms, - ip, iLimit, prefixStart, curr, dictLimit, ddsIdx); - } else if (dictMode == ZSTD_dictMatchState) { - const U32* const dmsChainTable = dms->chainTable; - const U32 dmsChainSize = (1 << dms->cParams.chainLog); - const U32 dmsChainMask = dmsChainSize - 1; - const U32 dmsLowestIndex = dms->window.dictLimit; - const BYTE* const dmsBase = dms->window.base; - const BYTE* const dmsEnd = dms->window.nextSrc; - const U32 dmsSize = (U32)(dmsEnd - dmsBase); - const U32 dmsIndexDelta = dictLimit - dmsSize; - const U32 dmsMinChain = dmsSize > dmsChainSize ? dmsSize - dmsChainSize : 0; - - matchIndex = dms->hashTable[ZSTD_hashPtr(ip, dms->cParams.hashLog, mls)]; - - for ( ; (matchIndex>=dmsLowestIndex) & (nbAttempts>0) ; nbAttempts--) { - size_t currentMl=0; - const BYTE* const match = dmsBase + matchIndex; - assert(match+4 <= dmsEnd); - if (MEM_read32(match) == MEM_read32(ip)) /* assumption : matchIndex <= dictLimit-4 (by table construction) */ - currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, dmsEnd, prefixStart) + 4; - - /* save best solution */ - if (currentMl > ml) { - ml = currentMl; - assert(curr > matchIndex + dmsIndexDelta); - *offsetPtr = STORE_OFFSET(curr - (matchIndex + dmsIndexDelta)); - if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */ - } - - if (matchIndex <= dmsMinChain) break; - - matchIndex = dmsChainTable[matchIndex & dmsChainMask]; - } - } - - return ml; -} - -/* ********************************* -* (SIMD) Row-based matchfinder -***********************************/ -/* Constants for row-based hash */ -#define ZSTD_ROW_HASH_TAG_OFFSET 16 /* byte offset of hashes in the match state's tagTable from the beginning of a row */ -#define ZSTD_ROW_HASH_TAG_BITS 8 /* nb bits to use for the tag */ -#define ZSTD_ROW_HASH_TAG_MASK ((1u << ZSTD_ROW_HASH_TAG_BITS) - 1) -#define ZSTD_ROW_HASH_MAX_ENTRIES 64 /* absolute maximum number of entries per row, for all configurations */ - -#define ZSTD_ROW_HASH_CACHE_MASK (ZSTD_ROW_HASH_CACHE_SIZE - 1) - -typedef U64 ZSTD_VecMask; /* Clarifies when we are interacting with a U64 representing a mask of matches */ - -/* ZSTD_VecMask_next(): - * Starting from the LSB, returns the idx of the next non-zero bit. - * Basically counting the nb of trailing zeroes. - */ -static U32 ZSTD_VecMask_next(ZSTD_VecMask val) { - assert(val != 0); -# if defined(_MSC_VER) && defined(_WIN64) - if (val != 0) { - unsigned long r; - _BitScanForward64(&r, val); - return (U32)(r); - } else { - /* Should not reach this code path */ - __assume(0); - } -# elif (defined(__GNUC__) && ((__GNUC__ > 3) || ((__GNUC__ == 3) && (__GNUC_MINOR__ >= 4)))) - if (sizeof(size_t) == 4) { - U32 mostSignificantWord = (U32)(val >> 32); - U32 leastSignificantWord = (U32)val; - if (leastSignificantWord == 0) { - return 32 + (U32)__builtin_ctz(mostSignificantWord); - } else { - return (U32)__builtin_ctz(leastSignificantWord); - } - } else { - return (U32)__builtin_ctzll(val); - } -# else - /* Software ctz version: http://aggregate.org/MAGIC/#Trailing%20Zero%20Count - * and: https://stackoverflow.com/questions/2709430/count-number-of-bits-in-a-64-bit-long-big-integer - */ - val = ~val & (val - 1ULL); /* Lowest set bit mask */ - val = val - ((val >> 1) & 0x5555555555555555); - val = (val & 0x3333333333333333ULL) + ((val >> 2) & 0x3333333333333333ULL); - return (U32)((((val + (val >> 4)) & 0xF0F0F0F0F0F0F0FULL) * 0x101010101010101ULL) >> 56); -# endif -} - -/* ZSTD_rotateRight_*(): - * Rotates a bitfield to the right by "count" bits. - * https://en.wikipedia.org/w/index.php?title=Circular_shift&oldid=991635599#Implementing_circular_shifts - */ -FORCE_INLINE_TEMPLATE -U64 ZSTD_rotateRight_U64(U64 const value, U32 count) { - assert(count < 64); - count &= 0x3F; /* for fickle pattern recognition */ - return (value >> count) | (U64)(value << ((0U - count) & 0x3F)); -} - -FORCE_INLINE_TEMPLATE -U32 ZSTD_rotateRight_U32(U32 const value, U32 count) { - assert(count < 32); - count &= 0x1F; /* for fickle pattern recognition */ - return (value >> count) | (U32)(value << ((0U - count) & 0x1F)); -} - -FORCE_INLINE_TEMPLATE -U16 ZSTD_rotateRight_U16(U16 const value, U32 count) { - assert(count < 16); - count &= 0x0F; /* for fickle pattern recognition */ - return (value >> count) | (U16)(value << ((0U - count) & 0x0F)); -} - -/* ZSTD_row_nextIndex(): - * Returns the next index to insert at within a tagTable row, and updates the "head" - * value to reflect the update. Essentially cycles backwards from [0, {entries per row}) - */ -FORCE_INLINE_TEMPLATE U32 ZSTD_row_nextIndex(BYTE* const tagRow, U32 const rowMask) { - U32 const next = (*tagRow - 1) & rowMask; - *tagRow = (BYTE)next; - return next; -} - -/* ZSTD_isAligned(): - * Checks that a pointer is aligned to "align" bytes which must be a power of 2. - */ -MEM_STATIC int ZSTD_isAligned(void const* ptr, size_t align) { - assert((align & (align - 1)) == 0); - return (((size_t)ptr) & (align - 1)) == 0; -} - -/* ZSTD_row_prefetch(): - * Performs prefetching for the hashTable and tagTable at a given row. - */ -FORCE_INLINE_TEMPLATE void ZSTD_row_prefetch(U32 const* hashTable, U16 const* tagTable, U32 const relRow, U32 const rowLog) { - PREFETCH_L1(hashTable + relRow); - if (rowLog >= 5) { - PREFETCH_L1(hashTable + relRow + 16); - /* Note: prefetching more of the hash table does not appear to be beneficial for 128-entry rows */ - } - PREFETCH_L1(tagTable + relRow); - if (rowLog == 6) { - PREFETCH_L1(tagTable + relRow + 32); - } - assert(rowLog == 4 || rowLog == 5 || rowLog == 6); - assert(ZSTD_isAligned(hashTable + relRow, 64)); /* prefetched hash row always 64-byte aligned */ - assert(ZSTD_isAligned(tagTable + relRow, (size_t)1 << rowLog)); /* prefetched tagRow sits on correct multiple of bytes (32,64,128) */ -} - -/* ZSTD_row_fillHashCache(): - * Fill up the hash cache starting at idx, prefetching up to ZSTD_ROW_HASH_CACHE_SIZE entries, - * but not beyond iLimit. - */ -FORCE_INLINE_TEMPLATE void ZSTD_row_fillHashCache(ZSTD_matchState_t* ms, const BYTE* base, - U32 const rowLog, U32 const mls, - U32 idx, const BYTE* const iLimit) -{ - U32 const* const hashTable = ms->hashTable; - U16 const* const tagTable = ms->tagTable; - U32 const hashLog = ms->rowHashLog; - U32 const maxElemsToPrefetch = (base + idx) > iLimit ? 0 : (U32)(iLimit - (base + idx) + 1); - U32 const lim = idx + MIN(ZSTD_ROW_HASH_CACHE_SIZE, maxElemsToPrefetch); - - for (; idx < lim; ++idx) { - U32 const hash = (U32)ZSTD_hashPtr(base + idx, hashLog + ZSTD_ROW_HASH_TAG_BITS, mls); - U32 const row = (hash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; - ZSTD_row_prefetch(hashTable, tagTable, row, rowLog); - ms->hashCache[idx & ZSTD_ROW_HASH_CACHE_MASK] = hash; - } - - DEBUGLOG(6, "ZSTD_row_fillHashCache(): [%u %u %u %u %u %u %u %u]", ms->hashCache[0], ms->hashCache[1], - ms->hashCache[2], ms->hashCache[3], ms->hashCache[4], - ms->hashCache[5], ms->hashCache[6], ms->hashCache[7]); -} - -/* ZSTD_row_nextCachedHash(): - * Returns the hash of base + idx, and replaces the hash in the hash cache with the byte at - * base + idx + ZSTD_ROW_HASH_CACHE_SIZE. Also prefetches the appropriate rows from hashTable and tagTable. - */ -FORCE_INLINE_TEMPLATE U32 ZSTD_row_nextCachedHash(U32* cache, U32 const* hashTable, - U16 const* tagTable, BYTE const* base, - U32 idx, U32 const hashLog, - U32 const rowLog, U32 const mls) -{ - U32 const newHash = (U32)ZSTD_hashPtr(base+idx+ZSTD_ROW_HASH_CACHE_SIZE, hashLog + ZSTD_ROW_HASH_TAG_BITS, mls); - U32 const row = (newHash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; - ZSTD_row_prefetch(hashTable, tagTable, row, rowLog); - { U32 const hash = cache[idx & ZSTD_ROW_HASH_CACHE_MASK]; - cache[idx & ZSTD_ROW_HASH_CACHE_MASK] = newHash; - return hash; - } -} - -/* ZSTD_row_update_internalImpl(): - * Updates the hash table with positions starting from updateStartIdx until updateEndIdx. - */ -FORCE_INLINE_TEMPLATE void ZSTD_row_update_internalImpl(ZSTD_matchState_t* ms, - U32 updateStartIdx, U32 const updateEndIdx, - U32 const mls, U32 const rowLog, - U32 const rowMask, U32 const useCache) -{ - U32* const hashTable = ms->hashTable; - U16* const tagTable = ms->tagTable; - U32 const hashLog = ms->rowHashLog; - const BYTE* const base = ms->window.base; - - DEBUGLOG(6, "ZSTD_row_update_internalImpl(): updateStartIdx=%u, updateEndIdx=%u", updateStartIdx, updateEndIdx); - for (; updateStartIdx < updateEndIdx; ++updateStartIdx) { - U32 const hash = useCache ? ZSTD_row_nextCachedHash(ms->hashCache, hashTable, tagTable, base, updateStartIdx, hashLog, rowLog, mls) - : (U32)ZSTD_hashPtr(base + updateStartIdx, hashLog + ZSTD_ROW_HASH_TAG_BITS, mls); - U32 const relRow = (hash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; - U32* const row = hashTable + relRow; - BYTE* tagRow = (BYTE*)(tagTable + relRow); /* Though tagTable is laid out as a table of U16, each tag is only 1 byte. - Explicit cast allows us to get exact desired position within each row */ - U32 const pos = ZSTD_row_nextIndex(tagRow, rowMask); - - assert(hash == ZSTD_hashPtr(base + updateStartIdx, hashLog + ZSTD_ROW_HASH_TAG_BITS, mls)); - ((BYTE*)tagRow)[pos + ZSTD_ROW_HASH_TAG_OFFSET] = hash & ZSTD_ROW_HASH_TAG_MASK; - row[pos] = updateStartIdx; - } -} - -/* ZSTD_row_update_internal(): - * Inserts the byte at ip into the appropriate position in the hash table, and updates ms->nextToUpdate. - * Skips sections of long matches as is necessary. - */ -FORCE_INLINE_TEMPLATE void ZSTD_row_update_internal(ZSTD_matchState_t* ms, const BYTE* ip, - U32 const mls, U32 const rowLog, - U32 const rowMask, U32 const useCache) -{ - U32 idx = ms->nextToUpdate; - const BYTE* const base = ms->window.base; - const U32 target = (U32)(ip - base); - const U32 kSkipThreshold = 384; - const U32 kMaxMatchStartPositionsToUpdate = 96; - const U32 kMaxMatchEndPositionsToUpdate = 32; - - if (useCache) { - /* Only skip positions when using hash cache, i.e. - * if we are loading a dict, don't skip anything. - * If we decide to skip, then we only update a set number - * of positions at the beginning and end of the match. - */ - if (UNLIKELY(target - idx > kSkipThreshold)) { - U32 const bound = idx + kMaxMatchStartPositionsToUpdate; - ZSTD_row_update_internalImpl(ms, idx, bound, mls, rowLog, rowMask, useCache); - idx = target - kMaxMatchEndPositionsToUpdate; - ZSTD_row_fillHashCache(ms, base, rowLog, mls, idx, ip+1); - } - } - assert(target >= idx); - ZSTD_row_update_internalImpl(ms, idx, target, mls, rowLog, rowMask, useCache); - ms->nextToUpdate = target; -} - -/* ZSTD_row_update(): - * External wrapper for ZSTD_row_update_internal(). Used for filling the hashtable during dictionary - * processing. - */ -void ZSTD_row_update(ZSTD_matchState_t* const ms, const BYTE* ip) { - const U32 rowLog = BOUNDED(4, ms->cParams.searchLog, 6); - const U32 rowMask = (1u << rowLog) - 1; - const U32 mls = MIN(ms->cParams.minMatch, 6 /* mls caps out at 6 */); - - DEBUGLOG(5, "ZSTD_row_update(), rowLog=%u", rowLog); - ZSTD_row_update_internal(ms, ip, mls, rowLog, rowMask, 0 /* dont use cache */); -} - -#if defined(ZSTD_ARCH_X86_SSE2) -FORCE_INLINE_TEMPLATE ZSTD_VecMask -ZSTD_row_getSSEMask(int nbChunks, const BYTE* const src, const BYTE tag, const U32 head) -{ - const __m128i comparisonMask = _mm_set1_epi8((char)tag); - int matches[4] = {0}; - int i; - assert(nbChunks == 1 || nbChunks == 2 || nbChunks == 4); - for (i=0; i> chunkSize; - do { - size_t chunk = MEM_readST(&src[i]); - chunk ^= splatChar; - chunk = (((chunk | x80) - x01) | chunk) & x80; - matches <<= chunkSize; - matches |= (chunk * extractMagic) >> shiftAmount; - i -= chunkSize; - } while (i >= 0); - } else { /* big endian: reverse bits during extraction */ - const size_t msb = xFF ^ (xFF >> 1); - const size_t extractMagic = (msb / 0x1FF) | msb; - do { - size_t chunk = MEM_readST(&src[i]); - chunk ^= splatChar; - chunk = (((chunk | x80) - x01) | chunk) & x80; - matches <<= chunkSize; - matches |= ((chunk >> 7) * extractMagic) >> shiftAmount; - i -= chunkSize; - } while (i >= 0); - } - matches = ~matches; - if (rowEntries == 16) { - return ZSTD_rotateRight_U16((U16)matches, head); - } else if (rowEntries == 32) { - return ZSTD_rotateRight_U32((U32)matches, head); - } else { - return ZSTD_rotateRight_U64((U64)matches, head); - } - } -#endif -} - -/* The high-level approach of the SIMD row based match finder is as follows: - * - Figure out where to insert the new entry: - * - Generate a hash from a byte along with an additional 1-byte "short hash". The additional byte is our "tag" - * - The hashTable is effectively split into groups or "rows" of 16 or 32 entries of U32, and the hash determines - * which row to insert into. - * - Determine the correct position within the row to insert the entry into. Each row of 16 or 32 can - * be considered as a circular buffer with a "head" index that resides in the tagTable. - * - Also insert the "tag" into the equivalent row and position in the tagTable. - * - Note: The tagTable has 17 or 33 1-byte entries per row, due to 16 or 32 tags, and 1 "head" entry. - * The 17 or 33 entry rows are spaced out to occur every 32 or 64 bytes, respectively, - * for alignment/performance reasons, leaving some bytes unused. - * - Use SIMD to efficiently compare the tags in the tagTable to the 1-byte "short hash" and - * generate a bitfield that we can cycle through to check the collisions in the hash table. - * - Pick the longest match. - */ -FORCE_INLINE_TEMPLATE -size_t ZSTD_RowFindBestMatch( - ZSTD_matchState_t* ms, - const BYTE* const ip, const BYTE* const iLimit, - size_t* offsetPtr, - const U32 mls, const ZSTD_dictMode_e dictMode, - const U32 rowLog) -{ - U32* const hashTable = ms->hashTable; - U16* const tagTable = ms->tagTable; - U32* const hashCache = ms->hashCache; - const U32 hashLog = ms->rowHashLog; - const ZSTD_compressionParameters* const cParams = &ms->cParams; - const BYTE* const base = ms->window.base; - const BYTE* const dictBase = ms->window.dictBase; - const U32 dictLimit = ms->window.dictLimit; - const BYTE* const prefixStart = base + dictLimit; - const BYTE* const dictEnd = dictBase + dictLimit; - const U32 curr = (U32)(ip-base); - const U32 maxDistance = 1U << cParams->windowLog; - const U32 lowestValid = ms->window.lowLimit; - const U32 withinMaxDistance = (curr - lowestValid > maxDistance) ? curr - maxDistance : lowestValid; - const U32 isDictionary = (ms->loadedDictEnd != 0); - const U32 lowLimit = isDictionary ? lowestValid : withinMaxDistance; - const U32 rowEntries = (1U << rowLog); - const U32 rowMask = rowEntries - 1; - const U32 cappedSearchLog = MIN(cParams->searchLog, rowLog); /* nb of searches is capped at nb entries per row */ - U32 nbAttempts = 1U << cappedSearchLog; - size_t ml=4-1; - - /* DMS/DDS variables that may be referenced laster */ - const ZSTD_matchState_t* const dms = ms->dictMatchState; - - /* Initialize the following variables to satisfy static analyzer */ - size_t ddsIdx = 0; - U32 ddsExtraAttempts = 0; /* cctx hash tables are limited in searches, but allow extra searches into DDS */ - U32 dmsTag = 0; - U32* dmsRow = NULL; - BYTE* dmsTagRow = NULL; - - if (dictMode == ZSTD_dedicatedDictSearch) { - const U32 ddsHashLog = dms->cParams.hashLog - ZSTD_LAZY_DDSS_BUCKET_LOG; - { /* Prefetch DDS hashtable entry */ - ddsIdx = ZSTD_hashPtr(ip, ddsHashLog, mls) << ZSTD_LAZY_DDSS_BUCKET_LOG; - PREFETCH_L1(&dms->hashTable[ddsIdx]); - } - ddsExtraAttempts = cParams->searchLog > rowLog ? 1U << (cParams->searchLog - rowLog) : 0; - } - - if (dictMode == ZSTD_dictMatchState) { - /* Prefetch DMS rows */ - U32* const dmsHashTable = dms->hashTable; - U16* const dmsTagTable = dms->tagTable; - U32 const dmsHash = (U32)ZSTD_hashPtr(ip, dms->rowHashLog + ZSTD_ROW_HASH_TAG_BITS, mls); - U32 const dmsRelRow = (dmsHash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; - dmsTag = dmsHash & ZSTD_ROW_HASH_TAG_MASK; - dmsTagRow = (BYTE*)(dmsTagTable + dmsRelRow); - dmsRow = dmsHashTable + dmsRelRow; - ZSTD_row_prefetch(dmsHashTable, dmsTagTable, dmsRelRow, rowLog); - } - - /* Update the hashTable and tagTable up to (but not including) ip */ - ZSTD_row_update_internal(ms, ip, mls, rowLog, rowMask, 1 /* useCache */); - { /* Get the hash for ip, compute the appropriate row */ - U32 const hash = ZSTD_row_nextCachedHash(hashCache, hashTable, tagTable, base, curr, hashLog, rowLog, mls); - U32 const relRow = (hash >> ZSTD_ROW_HASH_TAG_BITS) << rowLog; - U32 const tag = hash & ZSTD_ROW_HASH_TAG_MASK; - U32* const row = hashTable + relRow; - BYTE* tagRow = (BYTE*)(tagTable + relRow); - U32 const head = *tagRow & rowMask; - U32 matchBuffer[ZSTD_ROW_HASH_MAX_ENTRIES]; - size_t numMatches = 0; - size_t currMatch = 0; - ZSTD_VecMask matches = ZSTD_row_getMatchMask(tagRow, (BYTE)tag, head, rowEntries); - - /* Cycle through the matches and prefetch */ - for (; (matches > 0) && (nbAttempts > 0); --nbAttempts, matches &= (matches - 1)) { - U32 const matchPos = (head + ZSTD_VecMask_next(matches)) & rowMask; - U32 const matchIndex = row[matchPos]; - assert(numMatches < rowEntries); - if (matchIndex < lowLimit) - break; - if ((dictMode != ZSTD_extDict) || matchIndex >= dictLimit) { - PREFETCH_L1(base + matchIndex); - } else { - PREFETCH_L1(dictBase + matchIndex); - } - matchBuffer[numMatches++] = matchIndex; - } - - /* Speed opt: insert current byte into hashtable too. This allows us to avoid one iteration of the loop - in ZSTD_row_update_internal() at the next search. */ - { - U32 const pos = ZSTD_row_nextIndex(tagRow, rowMask); - tagRow[pos + ZSTD_ROW_HASH_TAG_OFFSET] = (BYTE)tag; - row[pos] = ms->nextToUpdate++; - } - - /* Return the longest match */ - for (; currMatch < numMatches; ++currMatch) { - U32 const matchIndex = matchBuffer[currMatch]; - size_t currentMl=0; - assert(matchIndex < curr); - assert(matchIndex >= lowLimit); - - if ((dictMode != ZSTD_extDict) || matchIndex >= dictLimit) { - const BYTE* const match = base + matchIndex; - assert(matchIndex >= dictLimit); /* ensures this is true if dictMode != ZSTD_extDict */ - if (match[ml] == ip[ml]) /* potentially better */ - currentMl = ZSTD_count(ip, match, iLimit); - } else { - const BYTE* const match = dictBase + matchIndex; - assert(match+4 <= dictEnd); - if (MEM_read32(match) == MEM_read32(ip)) /* assumption : matchIndex <= dictLimit-4 (by table construction) */ - currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, dictEnd, prefixStart) + 4; - } - - /* Save best solution */ - if (currentMl > ml) { - ml = currentMl; - *offsetPtr = STORE_OFFSET(curr - matchIndex); - if (ip+currentMl == iLimit) break; /* best possible, avoids read overflow on next attempt */ - } - } - } - - assert(nbAttempts <= (1U << ZSTD_SEARCHLOG_MAX)); /* Check we haven't underflowed. */ - if (dictMode == ZSTD_dedicatedDictSearch) { - ml = ZSTD_dedicatedDictSearch_lazy_search(offsetPtr, ml, nbAttempts + ddsExtraAttempts, dms, - ip, iLimit, prefixStart, curr, dictLimit, ddsIdx); - } else if (dictMode == ZSTD_dictMatchState) { - /* TODO: Measure and potentially add prefetching to DMS */ - const U32 dmsLowestIndex = dms->window.dictLimit; - const BYTE* const dmsBase = dms->window.base; - const BYTE* const dmsEnd = dms->window.nextSrc; - const U32 dmsSize = (U32)(dmsEnd - dmsBase); - const U32 dmsIndexDelta = dictLimit - dmsSize; - - { U32 const head = *dmsTagRow & rowMask; - U32 matchBuffer[ZSTD_ROW_HASH_MAX_ENTRIES]; - size_t numMatches = 0; - size_t currMatch = 0; - ZSTD_VecMask matches = ZSTD_row_getMatchMask(dmsTagRow, (BYTE)dmsTag, head, rowEntries); - - for (; (matches > 0) && (nbAttempts > 0); --nbAttempts, matches &= (matches - 1)) { - U32 const matchPos = (head + ZSTD_VecMask_next(matches)) & rowMask; - U32 const matchIndex = dmsRow[matchPos]; - if (matchIndex < dmsLowestIndex) - break; - PREFETCH_L1(dmsBase + matchIndex); - matchBuffer[numMatches++] = matchIndex; - } - - /* Return the longest match */ - for (; currMatch < numMatches; ++currMatch) { - U32 const matchIndex = matchBuffer[currMatch]; - size_t currentMl=0; - assert(matchIndex >= dmsLowestIndex); - assert(matchIndex < curr); - - { const BYTE* const match = dmsBase + matchIndex; - assert(match+4 <= dmsEnd); - if (MEM_read32(match) == MEM_read32(ip)) - currentMl = ZSTD_count_2segments(ip+4, match+4, iLimit, dmsEnd, prefixStart) + 4; - } - - if (currentMl > ml) { - ml = currentMl; - assert(curr > matchIndex + dmsIndexDelta); - *offsetPtr = STORE_OFFSET(curr - (matchIndex + dmsIndexDelta)); - if (ip+currentMl == iLimit) break; - } - } - } - } - return ml; -} - - -typedef size_t (*searchMax_f)( - ZSTD_matchState_t* ms, - const BYTE* ip, const BYTE* iLimit, size_t* offsetPtr); - -/** - * This struct contains the functions necessary for lazy to search. - * Currently, that is only searchMax. However, it is still valuable to have the - * VTable because this makes it easier to add more functions to the VTable later. - * - * TODO: The start of the search function involves loading and calculating a - * bunch of constants from the ZSTD_matchState_t. These computations could be - * done in an initialization function, and saved somewhere in the match state. - * Then we could pass a pointer to the saved state instead of the match state, - * and avoid duplicate computations. - * - * TODO: Move the match re-winding into searchMax. This improves compression - * ratio, and unlocks further simplifications with the next TODO. - * - * TODO: Try moving the repcode search into searchMax. After the re-winding - * and repcode search are in searchMax, there is no more logic in the match - * finder loop that requires knowledge about the dictMode. So we should be - * able to avoid force inlining it, and we can join the extDict loop with - * the single segment loop. It should go in searchMax instead of its own - * function to avoid having multiple virtual function calls per search. - */ -typedef struct { - searchMax_f searchMax; -} ZSTD_LazyVTable; - -#define GEN_ZSTD_BT_VTABLE(dictMode, mls) \ - static size_t ZSTD_BtFindBestMatch_##dictMode##_##mls( \ - ZSTD_matchState_t* ms, \ - const BYTE* ip, const BYTE* const iLimit, \ - size_t* offsetPtr) \ - { \ - assert(MAX(4, MIN(6, ms->cParams.minMatch)) == mls); \ - return ZSTD_BtFindBestMatch(ms, ip, iLimit, offsetPtr, mls, ZSTD_##dictMode); \ - } \ - static const ZSTD_LazyVTable ZSTD_BtVTable_##dictMode##_##mls = { \ - ZSTD_BtFindBestMatch_##dictMode##_##mls \ - }; - -#define GEN_ZSTD_HC_VTABLE(dictMode, mls) \ - static size_t ZSTD_HcFindBestMatch_##dictMode##_##mls( \ - ZSTD_matchState_t* ms, \ - const BYTE* ip, const BYTE* const iLimit, \ - size_t* offsetPtr) \ - { \ - assert(MAX(4, MIN(6, ms->cParams.minMatch)) == mls); \ - return ZSTD_HcFindBestMatch(ms, ip, iLimit, offsetPtr, mls, ZSTD_##dictMode); \ - } \ - static const ZSTD_LazyVTable ZSTD_HcVTable_##dictMode##_##mls = { \ - ZSTD_HcFindBestMatch_##dictMode##_##mls \ - }; - -#define GEN_ZSTD_ROW_VTABLE(dictMode, mls, rowLog) \ - static size_t ZSTD_RowFindBestMatch_##dictMode##_##mls##_##rowLog( \ - ZSTD_matchState_t* ms, \ - const BYTE* ip, const BYTE* const iLimit, \ - size_t* offsetPtr) \ - { \ - assert(MAX(4, MIN(6, ms->cParams.minMatch)) == mls); \ - assert(MAX(4, MIN(6, ms->cParams.searchLog)) == rowLog); \ - return ZSTD_RowFindBestMatch(ms, ip, iLimit, offsetPtr, mls, ZSTD_##dictMode, rowLog); \ - } \ - static const ZSTD_LazyVTable ZSTD_RowVTable_##dictMode##_##mls##_##rowLog = { \ - ZSTD_RowFindBestMatch_##dictMode##_##mls##_##rowLog \ - }; - -#define ZSTD_FOR_EACH_ROWLOG(X, dictMode, mls) \ - X(dictMode, mls, 4) \ - X(dictMode, mls, 5) \ - X(dictMode, mls, 6) - -#define ZSTD_FOR_EACH_MLS_ROWLOG(X, dictMode) \ - ZSTD_FOR_EACH_ROWLOG(X, dictMode, 4) \ - ZSTD_FOR_EACH_ROWLOG(X, dictMode, 5) \ - ZSTD_FOR_EACH_ROWLOG(X, dictMode, 6) - -#define ZSTD_FOR_EACH_MLS(X, dictMode) \ - X(dictMode, 4) \ - X(dictMode, 5) \ - X(dictMode, 6) - -#define ZSTD_FOR_EACH_DICT_MODE(X, ...) \ - X(__VA_ARGS__, noDict) \ - X(__VA_ARGS__, extDict) \ - X(__VA_ARGS__, dictMatchState) \ - X(__VA_ARGS__, dedicatedDictSearch) - -/* Generate Row VTables for each combination of (dictMode, mls, rowLog) */ -ZSTD_FOR_EACH_DICT_MODE(ZSTD_FOR_EACH_MLS_ROWLOG, GEN_ZSTD_ROW_VTABLE) -/* Generate Binary Tree VTables for each combination of (dictMode, mls) */ -ZSTD_FOR_EACH_DICT_MODE(ZSTD_FOR_EACH_MLS, GEN_ZSTD_BT_VTABLE) -/* Generate Hash Chain VTables for each combination of (dictMode, mls) */ -ZSTD_FOR_EACH_DICT_MODE(ZSTD_FOR_EACH_MLS, GEN_ZSTD_HC_VTABLE) - -#define GEN_ZSTD_BT_VTABLE_ARRAY(dictMode) \ - { \ - &ZSTD_BtVTable_##dictMode##_4, \ - &ZSTD_BtVTable_##dictMode##_5, \ - &ZSTD_BtVTable_##dictMode##_6 \ - } - -#define GEN_ZSTD_HC_VTABLE_ARRAY(dictMode) \ - { \ - &ZSTD_HcVTable_##dictMode##_4, \ - &ZSTD_HcVTable_##dictMode##_5, \ - &ZSTD_HcVTable_##dictMode##_6 \ - } - -#define GEN_ZSTD_ROW_VTABLE_ARRAY_(dictMode, mls) \ - { \ - &ZSTD_RowVTable_##dictMode##_##mls##_4, \ - &ZSTD_RowVTable_##dictMode##_##mls##_5, \ - &ZSTD_RowVTable_##dictMode##_##mls##_6 \ - } - -#define GEN_ZSTD_ROW_VTABLE_ARRAY(dictMode) \ - { \ - GEN_ZSTD_ROW_VTABLE_ARRAY_(dictMode, 4), \ - GEN_ZSTD_ROW_VTABLE_ARRAY_(dictMode, 5), \ - GEN_ZSTD_ROW_VTABLE_ARRAY_(dictMode, 6) \ - } - -#define GEN_ZSTD_VTABLE_ARRAY(X) \ - { \ - X(noDict), \ - X(extDict), \ - X(dictMatchState), \ - X(dedicatedDictSearch) \ - } - -/* ******************************* -* Common parser - lazy strategy -*********************************/ -typedef enum { search_hashChain=0, search_binaryTree=1, search_rowHash=2 } searchMethod_e; - -/** - * This table is indexed first by the four ZSTD_dictMode_e values, and then - * by the two searchMethod_e values. NULLs are placed for configurations - * that should never occur (extDict modes go to the other implementation - * below and there is no DDSS for binary tree search yet). - */ - -static ZSTD_LazyVTable const* -ZSTD_selectLazyVTable(ZSTD_matchState_t const* ms, searchMethod_e searchMethod, ZSTD_dictMode_e dictMode) -{ - /* Fill the Hc/Bt VTable arrays with the right functions for the (dictMode, mls) combination. */ - ZSTD_LazyVTable const* const hcVTables[4][3] = GEN_ZSTD_VTABLE_ARRAY(GEN_ZSTD_HC_VTABLE_ARRAY); - ZSTD_LazyVTable const* const btVTables[4][3] = GEN_ZSTD_VTABLE_ARRAY(GEN_ZSTD_BT_VTABLE_ARRAY); - /* Fill the Row VTable array with the right functions for the (dictMode, mls, rowLog) combination. */ - ZSTD_LazyVTable const* const rowVTables[4][3][3] = GEN_ZSTD_VTABLE_ARRAY(GEN_ZSTD_ROW_VTABLE_ARRAY); - - U32 const mls = MAX(4, MIN(6, ms->cParams.minMatch)); - U32 const rowLog = MAX(4, MIN(6, ms->cParams.searchLog)); - switch (searchMethod) { - case search_hashChain: - return hcVTables[dictMode][mls - 4]; - case search_binaryTree: - return btVTables[dictMode][mls - 4]; - case search_rowHash: - return rowVTables[dictMode][mls - 4][rowLog - 4]; - default: - return NULL; - } -} - -FORCE_INLINE_TEMPLATE size_t -ZSTD_compressBlock_lazy_generic( - ZSTD_matchState_t* ms, seqStore_t* seqStore, - U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize, - const searchMethod_e searchMethod, const U32 depth, - ZSTD_dictMode_e const dictMode) -{ - const BYTE* const istart = (const BYTE*)src; - const BYTE* ip = istart; - const BYTE* anchor = istart; - const BYTE* const iend = istart + srcSize; - const BYTE* const ilimit = (searchMethod == search_rowHash) ? iend - 8 - ZSTD_ROW_HASH_CACHE_SIZE : iend - 8; - const BYTE* const base = ms->window.base; - const U32 prefixLowestIndex = ms->window.dictLimit; - const BYTE* const prefixLowest = base + prefixLowestIndex; - - searchMax_f const searchMax = ZSTD_selectLazyVTable(ms, searchMethod, dictMode)->searchMax; - U32 offset_1 = rep[0], offset_2 = rep[1], savedOffset=0; - - const int isDMS = dictMode == ZSTD_dictMatchState; - const int isDDS = dictMode == ZSTD_dedicatedDictSearch; - const int isDxS = isDMS || isDDS; - const ZSTD_matchState_t* const dms = ms->dictMatchState; - const U32 dictLowestIndex = isDxS ? dms->window.dictLimit : 0; - const BYTE* const dictBase = isDxS ? dms->window.base : NULL; - const BYTE* const dictLowest = isDxS ? dictBase + dictLowestIndex : NULL; - const BYTE* const dictEnd = isDxS ? dms->window.nextSrc : NULL; - const U32 dictIndexDelta = isDxS ? - prefixLowestIndex - (U32)(dictEnd - dictBase) : - 0; - const U32 dictAndPrefixLength = (U32)((ip - prefixLowest) + (dictEnd - dictLowest)); - - assert(searchMax != NULL); - - DEBUGLOG(5, "ZSTD_compressBlock_lazy_generic (dictMode=%u) (searchFunc=%u)", (U32)dictMode, (U32)searchMethod); - ip += (dictAndPrefixLength == 0); - if (dictMode == ZSTD_noDict) { - U32 const curr = (U32)(ip - base); - U32 const windowLow = ZSTD_getLowestPrefixIndex(ms, curr, ms->cParams.windowLog); - U32 const maxRep = curr - windowLow; - if (offset_2 > maxRep) savedOffset = offset_2, offset_2 = 0; - if (offset_1 > maxRep) savedOffset = offset_1, offset_1 = 0; - } - if (isDxS) { - /* dictMatchState repCode checks don't currently handle repCode == 0 - * disabling. */ - assert(offset_1 <= dictAndPrefixLength); - assert(offset_2 <= dictAndPrefixLength); - } - - if (searchMethod == search_rowHash) { - const U32 rowLog = MAX(4, MIN(6, ms->cParams.searchLog)); - ZSTD_row_fillHashCache(ms, base, rowLog, - MIN(ms->cParams.minMatch, 6 /* mls caps out at 6 */), - ms->nextToUpdate, ilimit); - } - - /* Match Loop */ -#if defined(__GNUC__) && defined(__x86_64__) - /* I've measured random a 5% speed loss on levels 5 & 6 (greedy) when the - * code alignment is perturbed. To fix the instability align the loop on 32-bytes. - */ - __asm__(".p2align 5"); -#endif - while (ip < ilimit) { - size_t matchLength=0; - size_t offcode=STORE_REPCODE_1; - const BYTE* start=ip+1; - DEBUGLOG(7, "search baseline (depth 0)"); - - /* check repCode */ - if (isDxS) { - const U32 repIndex = (U32)(ip - base) + 1 - offset_1; - const BYTE* repMatch = ((dictMode == ZSTD_dictMatchState || dictMode == ZSTD_dedicatedDictSearch) - && repIndex < prefixLowestIndex) ? - dictBase + (repIndex - dictIndexDelta) : - base + repIndex; - if (((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */) - && (MEM_read32(repMatch) == MEM_read32(ip+1)) ) { - const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend; - matchLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4; - if (depth==0) goto _storeSequence; - } - } - if ( dictMode == ZSTD_noDict - && ((offset_1 > 0) & (MEM_read32(ip+1-offset_1) == MEM_read32(ip+1)))) { - matchLength = ZSTD_count(ip+1+4, ip+1+4-offset_1, iend) + 4; - if (depth==0) goto _storeSequence; - } - - /* first search (depth 0) */ - { size_t offsetFound = 999999999; - size_t const ml2 = searchMax(ms, ip, iend, &offsetFound); - if (ml2 > matchLength) - matchLength = ml2, start = ip, offcode=offsetFound; - } - - if (matchLength < 4) { - ip += ((ip-anchor) >> kSearchStrength) + 1; /* jump faster over incompressible sections */ - continue; - } - - /* let's try to find a better solution */ - if (depth>=1) - while (ip0) & (MEM_read32(ip) == MEM_read32(ip - offset_1)))) { - size_t const mlRep = ZSTD_count(ip+4, ip+4-offset_1, iend) + 4; - int const gain2 = (int)(mlRep * 3); - int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 1); - if ((mlRep >= 4) && (gain2 > gain1)) - matchLength = mlRep, offcode = STORE_REPCODE_1, start = ip; - } - if (isDxS) { - const U32 repIndex = (U32)(ip - base) - offset_1; - const BYTE* repMatch = repIndex < prefixLowestIndex ? - dictBase + (repIndex - dictIndexDelta) : - base + repIndex; - if (((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */) - && (MEM_read32(repMatch) == MEM_read32(ip)) ) { - const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend; - size_t const mlRep = ZSTD_count_2segments(ip+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4; - int const gain2 = (int)(mlRep * 3); - int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 1); - if ((mlRep >= 4) && (gain2 > gain1)) - matchLength = mlRep, offcode = STORE_REPCODE_1, start = ip; - } - } - { size_t offset2=999999999; - size_t const ml2 = searchMax(ms, ip, iend, &offset2); - int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offset2))); /* raw approx */ - int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 4); - if ((ml2 >= 4) && (gain2 > gain1)) { - matchLength = ml2, offcode = offset2, start = ip; - continue; /* search a better one */ - } } - - /* let's find an even better one */ - if ((depth==2) && (ip0) & (MEM_read32(ip) == MEM_read32(ip - offset_1)))) { - size_t const mlRep = ZSTD_count(ip+4, ip+4-offset_1, iend) + 4; - int const gain2 = (int)(mlRep * 4); - int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 1); - if ((mlRep >= 4) && (gain2 > gain1)) - matchLength = mlRep, offcode = STORE_REPCODE_1, start = ip; - } - if (isDxS) { - const U32 repIndex = (U32)(ip - base) - offset_1; - const BYTE* repMatch = repIndex < prefixLowestIndex ? - dictBase + (repIndex - dictIndexDelta) : - base + repIndex; - if (((U32)((prefixLowestIndex-1) - repIndex) >= 3 /* intentional underflow */) - && (MEM_read32(repMatch) == MEM_read32(ip)) ) { - const BYTE* repMatchEnd = repIndex < prefixLowestIndex ? dictEnd : iend; - size_t const mlRep = ZSTD_count_2segments(ip+4, repMatch+4, iend, repMatchEnd, prefixLowest) + 4; - int const gain2 = (int)(mlRep * 4); - int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 1); - if ((mlRep >= 4) && (gain2 > gain1)) - matchLength = mlRep, offcode = STORE_REPCODE_1, start = ip; - } - } - { size_t offset2=999999999; - size_t const ml2 = searchMax(ms, ip, iend, &offset2); - int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offset2))); /* raw approx */ - int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 7); - if ((ml2 >= 4) && (gain2 > gain1)) { - matchLength = ml2, offcode = offset2, start = ip; - continue; - } } } - break; /* nothing found : store previous solution */ - } - - /* NOTE: - * Pay attention that `start[-value]` can lead to strange undefined behavior - * notably if `value` is unsigned, resulting in a large positive `-value`. - */ - /* catch up */ - if (STORED_IS_OFFSET(offcode)) { - if (dictMode == ZSTD_noDict) { - while ( ((start > anchor) & (start - STORED_OFFSET(offcode) > prefixLowest)) - && (start[-1] == (start-STORED_OFFSET(offcode))[-1]) ) /* only search for offset within prefix */ - { start--; matchLength++; } - } - if (isDxS) { - U32 const matchIndex = (U32)((size_t)(start-base) - STORED_OFFSET(offcode)); - const BYTE* match = (matchIndex < prefixLowestIndex) ? dictBase + matchIndex - dictIndexDelta : base + matchIndex; - const BYTE* const mStart = (matchIndex < prefixLowestIndex) ? dictLowest : prefixLowest; - while ((start>anchor) && (match>mStart) && (start[-1] == match[-1])) { start--; match--; matchLength++; } /* catch up */ - } - offset_2 = offset_1; offset_1 = (U32)STORED_OFFSET(offcode); - } - /* store sequence */ -_storeSequence: - { size_t const litLength = (size_t)(start - anchor); - ZSTD_storeSeq(seqStore, litLength, anchor, iend, (U32)offcode, matchLength); - anchor = ip = start + matchLength; - } - - /* check immediate repcode */ - if (isDxS) { - while (ip <= ilimit) { - U32 const current2 = (U32)(ip-base); - U32 const repIndex = current2 - offset_2; - const BYTE* repMatch = repIndex < prefixLowestIndex ? - dictBase - dictIndexDelta + repIndex : - base + repIndex; - if ( ((U32)((prefixLowestIndex-1) - (U32)repIndex) >= 3 /* intentional overflow */) - && (MEM_read32(repMatch) == MEM_read32(ip)) ) { - const BYTE* const repEnd2 = repIndex < prefixLowestIndex ? dictEnd : iend; - matchLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd2, prefixLowest) + 4; - offcode = offset_2; offset_2 = offset_1; offset_1 = (U32)offcode; /* swap offset_2 <=> offset_1 */ - ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, matchLength); - ip += matchLength; - anchor = ip; - continue; - } - break; - } - } - - if (dictMode == ZSTD_noDict) { - while ( ((ip <= ilimit) & (offset_2>0)) - && (MEM_read32(ip) == MEM_read32(ip - offset_2)) ) { - /* store sequence */ - matchLength = ZSTD_count(ip+4, ip+4-offset_2, iend) + 4; - offcode = offset_2; offset_2 = offset_1; offset_1 = (U32)offcode; /* swap repcodes */ - ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, matchLength); - ip += matchLength; - anchor = ip; - continue; /* faster when present ... (?) */ - } } } - - /* Save reps for next block */ - rep[0] = offset_1 ? offset_1 : savedOffset; - rep[1] = offset_2 ? offset_2 : savedOffset; - - /* Return the last literals size */ - return (size_t)(iend - anchor); -} - - -size_t ZSTD_compressBlock_btlazy2( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_binaryTree, 2, ZSTD_noDict); -} - -size_t ZSTD_compressBlock_lazy2( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 2, ZSTD_noDict); -} - -size_t ZSTD_compressBlock_lazy( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 1, ZSTD_noDict); -} - -size_t ZSTD_compressBlock_greedy( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 0, ZSTD_noDict); -} - -size_t ZSTD_compressBlock_btlazy2_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_binaryTree, 2, ZSTD_dictMatchState); -} - -size_t ZSTD_compressBlock_lazy2_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 2, ZSTD_dictMatchState); -} - -size_t ZSTD_compressBlock_lazy_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 1, ZSTD_dictMatchState); -} - -size_t ZSTD_compressBlock_greedy_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 0, ZSTD_dictMatchState); -} - - -size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 2, ZSTD_dedicatedDictSearch); -} - -size_t ZSTD_compressBlock_lazy_dedicatedDictSearch( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 1, ZSTD_dedicatedDictSearch); -} - -size_t ZSTD_compressBlock_greedy_dedicatedDictSearch( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 0, ZSTD_dedicatedDictSearch); -} - -/* Row-based matchfinder */ -size_t ZSTD_compressBlock_lazy2_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2, ZSTD_noDict); -} - -size_t ZSTD_compressBlock_lazy_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1, ZSTD_noDict); -} - -size_t ZSTD_compressBlock_greedy_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0, ZSTD_noDict); -} - -size_t ZSTD_compressBlock_lazy2_dictMatchState_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2, ZSTD_dictMatchState); -} - -size_t ZSTD_compressBlock_lazy_dictMatchState_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1, ZSTD_dictMatchState); -} - -size_t ZSTD_compressBlock_greedy_dictMatchState_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0, ZSTD_dictMatchState); -} - - -size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2, ZSTD_dedicatedDictSearch); -} - -size_t ZSTD_compressBlock_lazy_dedicatedDictSearch_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1, ZSTD_dedicatedDictSearch); -} - -size_t ZSTD_compressBlock_greedy_dedicatedDictSearch_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0, ZSTD_dedicatedDictSearch); -} - -FORCE_INLINE_TEMPLATE -size_t ZSTD_compressBlock_lazy_extDict_generic( - ZSTD_matchState_t* ms, seqStore_t* seqStore, - U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize, - const searchMethod_e searchMethod, const U32 depth) -{ - const BYTE* const istart = (const BYTE*)src; - const BYTE* ip = istart; - const BYTE* anchor = istart; - const BYTE* const iend = istart + srcSize; - const BYTE* const ilimit = searchMethod == search_rowHash ? iend - 8 - ZSTD_ROW_HASH_CACHE_SIZE : iend - 8; - const BYTE* const base = ms->window.base; - const U32 dictLimit = ms->window.dictLimit; - const BYTE* const prefixStart = base + dictLimit; - const BYTE* const dictBase = ms->window.dictBase; - const BYTE* const dictEnd = dictBase + dictLimit; - const BYTE* const dictStart = dictBase + ms->window.lowLimit; - const U32 windowLog = ms->cParams.windowLog; - const U32 rowLog = ms->cParams.searchLog < 5 ? 4 : 5; - - searchMax_f const searchMax = ZSTD_selectLazyVTable(ms, searchMethod, ZSTD_extDict)->searchMax; - U32 offset_1 = rep[0], offset_2 = rep[1]; - - DEBUGLOG(5, "ZSTD_compressBlock_lazy_extDict_generic (searchFunc=%u)", (U32)searchMethod); - - /* init */ - ip += (ip == prefixStart); - if (searchMethod == search_rowHash) { - ZSTD_row_fillHashCache(ms, base, rowLog, - MIN(ms->cParams.minMatch, 6 /* mls caps out at 6 */), - ms->nextToUpdate, ilimit); - } - - /* Match Loop */ -#if defined(__GNUC__) && defined(__x86_64__) - /* I've measured random a 5% speed loss on levels 5 & 6 (greedy) when the - * code alignment is perturbed. To fix the instability align the loop on 32-bytes. - */ - __asm__(".p2align 5"); -#endif - while (ip < ilimit) { - size_t matchLength=0; - size_t offcode=STORE_REPCODE_1; - const BYTE* start=ip+1; - U32 curr = (U32)(ip-base); - - /* check repCode */ - { const U32 windowLow = ZSTD_getLowestMatchIndex(ms, curr+1, windowLog); - const U32 repIndex = (U32)(curr+1 - offset_1); - const BYTE* const repBase = repIndex < dictLimit ? dictBase : base; - const BYTE* const repMatch = repBase + repIndex; - if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow */ - & (offset_1 <= curr+1 - windowLow) ) /* note: we are searching at curr+1 */ - if (MEM_read32(ip+1) == MEM_read32(repMatch)) { - /* repcode detected we should take it */ - const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend; - matchLength = ZSTD_count_2segments(ip+1+4, repMatch+4, iend, repEnd, prefixStart) + 4; - if (depth==0) goto _storeSequence; - } } - - /* first search (depth 0) */ - { size_t offsetFound = 999999999; - size_t const ml2 = searchMax(ms, ip, iend, &offsetFound); - if (ml2 > matchLength) - matchLength = ml2, start = ip, offcode=offsetFound; - } - - if (matchLength < 4) { - ip += ((ip-anchor) >> kSearchStrength) + 1; /* jump faster over incompressible sections */ - continue; - } - - /* let's try to find a better solution */ - if (depth>=1) - while (ip= 3) /* intentional overflow : do not test positions overlapping 2 memory segments */ - & (offset_1 <= curr - windowLow) ) /* equivalent to `curr > repIndex >= windowLow` */ - if (MEM_read32(ip) == MEM_read32(repMatch)) { - /* repcode detected */ - const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend; - size_t const repLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd, prefixStart) + 4; - int const gain2 = (int)(repLength * 3); - int const gain1 = (int)(matchLength*3 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 1); - if ((repLength >= 4) && (gain2 > gain1)) - matchLength = repLength, offcode = STORE_REPCODE_1, start = ip; - } } - - /* search match, depth 1 */ - { size_t offset2=999999999; - size_t const ml2 = searchMax(ms, ip, iend, &offset2); - int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offset2))); /* raw approx */ - int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 4); - if ((ml2 >= 4) && (gain2 > gain1)) { - matchLength = ml2, offcode = offset2, start = ip; - continue; /* search a better one */ - } } - - /* let's find an even better one */ - if ((depth==2) && (ip= 3) /* intentional overflow : do not test positions overlapping 2 memory segments */ - & (offset_1 <= curr - windowLow) ) /* equivalent to `curr > repIndex >= windowLow` */ - if (MEM_read32(ip) == MEM_read32(repMatch)) { - /* repcode detected */ - const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend; - size_t const repLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd, prefixStart) + 4; - int const gain2 = (int)(repLength * 4); - int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 1); - if ((repLength >= 4) && (gain2 > gain1)) - matchLength = repLength, offcode = STORE_REPCODE_1, start = ip; - } } - - /* search match, depth 2 */ - { size_t offset2=999999999; - size_t const ml2 = searchMax(ms, ip, iend, &offset2); - int const gain2 = (int)(ml2*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offset2))); /* raw approx */ - int const gain1 = (int)(matchLength*4 - ZSTD_highbit32((U32)STORED_TO_OFFBASE(offcode)) + 7); - if ((ml2 >= 4) && (gain2 > gain1)) { - matchLength = ml2, offcode = offset2, start = ip; - continue; - } } } - break; /* nothing found : store previous solution */ - } - - /* catch up */ - if (STORED_IS_OFFSET(offcode)) { - U32 const matchIndex = (U32)((size_t)(start-base) - STORED_OFFSET(offcode)); - const BYTE* match = (matchIndex < dictLimit) ? dictBase + matchIndex : base + matchIndex; - const BYTE* const mStart = (matchIndex < dictLimit) ? dictStart : prefixStart; - while ((start>anchor) && (match>mStart) && (start[-1] == match[-1])) { start--; match--; matchLength++; } /* catch up */ - offset_2 = offset_1; offset_1 = (U32)STORED_OFFSET(offcode); - } - - /* store sequence */ -_storeSequence: - { size_t const litLength = (size_t)(start - anchor); - ZSTD_storeSeq(seqStore, litLength, anchor, iend, (U32)offcode, matchLength); - anchor = ip = start + matchLength; - } - - /* check immediate repcode */ - while (ip <= ilimit) { - const U32 repCurrent = (U32)(ip-base); - const U32 windowLow = ZSTD_getLowestMatchIndex(ms, repCurrent, windowLog); - const U32 repIndex = repCurrent - offset_2; - const BYTE* const repBase = repIndex < dictLimit ? dictBase : base; - const BYTE* const repMatch = repBase + repIndex; - if ( ((U32)((dictLimit-1) - repIndex) >= 3) /* intentional overflow : do not test positions overlapping 2 memory segments */ - & (offset_2 <= repCurrent - windowLow) ) /* equivalent to `curr > repIndex >= windowLow` */ - if (MEM_read32(ip) == MEM_read32(repMatch)) { - /* repcode detected we should take it */ - const BYTE* const repEnd = repIndex < dictLimit ? dictEnd : iend; - matchLength = ZSTD_count_2segments(ip+4, repMatch+4, iend, repEnd, prefixStart) + 4; - offcode = offset_2; offset_2 = offset_1; offset_1 = (U32)offcode; /* swap offset history */ - ZSTD_storeSeq(seqStore, 0, anchor, iend, STORE_REPCODE_1, matchLength); - ip += matchLength; - anchor = ip; - continue; /* faster when present ... (?) */ - } - break; - } } - - /* Save reps for next block */ - rep[0] = offset_1; - rep[1] = offset_2; - - /* Return the last literals size */ - return (size_t)(iend - anchor); -} - - -size_t ZSTD_compressBlock_greedy_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 0); -} - -size_t ZSTD_compressBlock_lazy_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) - -{ - return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 1); -} - -size_t ZSTD_compressBlock_lazy2_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) - -{ - return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_hashChain, 2); -} - -size_t ZSTD_compressBlock_btlazy2_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) - -{ - return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_binaryTree, 2); -} - -size_t ZSTD_compressBlock_greedy_extDict_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) -{ - return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 0); -} - -size_t ZSTD_compressBlock_lazy_extDict_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) - -{ - return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 1); -} - -size_t ZSTD_compressBlock_lazy2_extDict_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize) - -{ - return ZSTD_compressBlock_lazy_extDict_generic(ms, seqStore, rep, src, srcSize, search_rowHash, 2); -} diff --git a/dep/zstd/lib/compress/zstd_lazy.h b/dep/zstd/lib/compress/zstd_lazy.h deleted file mode 100644 index 150f7b390..000000000 --- a/dep/zstd/lib/compress/zstd_lazy.h +++ /dev/null @@ -1,125 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_LAZY_H -#define ZSTD_LAZY_H - -#if defined (__cplusplus) -extern "C" { -#endif - -#include "zstd_compress_internal.h" - -/** - * Dedicated Dictionary Search Structure bucket log. In the - * ZSTD_dedicatedDictSearch mode, the hashTable has - * 2 ** ZSTD_LAZY_DDSS_BUCKET_LOG entries in each bucket, rather than just - * one. - */ -#define ZSTD_LAZY_DDSS_BUCKET_LOG 2 - -U32 ZSTD_insertAndFindFirstIndex(ZSTD_matchState_t* ms, const BYTE* ip); -void ZSTD_row_update(ZSTD_matchState_t* const ms, const BYTE* ip); - -void ZSTD_dedicatedDictSearch_lazy_loadDictionary(ZSTD_matchState_t* ms, const BYTE* const ip); - -void ZSTD_preserveUnsortedMark (U32* const table, U32 const size, U32 const reducerValue); /*! used in ZSTD_reduceIndex(). preemptively increase value of ZSTD_DUBT_UNSORTED_MARK */ - -size_t ZSTD_compressBlock_btlazy2( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy2( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_greedy( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy2_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_greedy_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); - -size_t ZSTD_compressBlock_btlazy2_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy2_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_greedy_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy2_dictMatchState_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy_dictMatchState_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_greedy_dictMatchState_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); - -size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy_dedicatedDictSearch( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_greedy_dedicatedDictSearch( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy2_dedicatedDictSearch_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy_dedicatedDictSearch_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_greedy_dedicatedDictSearch_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); - -size_t ZSTD_compressBlock_greedy_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy2_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_greedy_extDict_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy_extDict_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_lazy2_extDict_row( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_btlazy2_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); - - -#if defined (__cplusplus) -} -#endif - -#endif /* ZSTD_LAZY_H */ diff --git a/dep/zstd/lib/compress/zstd_ldm.c b/dep/zstd/lib/compress/zstd_ldm.c deleted file mode 100644 index f662b2546..000000000 --- a/dep/zstd/lib/compress/zstd_ldm.c +++ /dev/null @@ -1,724 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#include "zstd_ldm.h" - -#include "../common/debug.h" -#include "../common/xxhash.h" -#include "zstd_fast.h" /* ZSTD_fillHashTable() */ -#include "zstd_double_fast.h" /* ZSTD_fillDoubleHashTable() */ -#include "zstd_ldm_geartab.h" - -#define LDM_BUCKET_SIZE_LOG 3 -#define LDM_MIN_MATCH_LENGTH 64 -#define LDM_HASH_RLOG 7 - -typedef struct { - U64 rolling; - U64 stopMask; -} ldmRollingHashState_t; - -/** ZSTD_ldm_gear_init(): - * - * Initializes the rolling hash state such that it will honor the - * settings in params. */ -static void ZSTD_ldm_gear_init(ldmRollingHashState_t* state, ldmParams_t const* params) -{ - unsigned maxBitsInMask = MIN(params->minMatchLength, 64); - unsigned hashRateLog = params->hashRateLog; - - state->rolling = ~(U32)0; - - /* The choice of the splitting criterion is subject to two conditions: - * 1. it has to trigger on average every 2^(hashRateLog) bytes; - * 2. ideally, it has to depend on a window of minMatchLength bytes. - * - * In the gear hash algorithm, bit n depends on the last n bytes; - * so in order to obtain a good quality splitting criterion it is - * preferable to use bits with high weight. - * - * To match condition 1 we use a mask with hashRateLog bits set - * and, because of the previous remark, we make sure these bits - * have the highest possible weight while still respecting - * condition 2. - */ - if (hashRateLog > 0 && hashRateLog <= maxBitsInMask) { - state->stopMask = (((U64)1 << hashRateLog) - 1) << (maxBitsInMask - hashRateLog); - } else { - /* In this degenerate case we simply honor the hash rate. */ - state->stopMask = ((U64)1 << hashRateLog) - 1; - } -} - -/** ZSTD_ldm_gear_reset() - * Feeds [data, data + minMatchLength) into the hash without registering any - * splits. This effectively resets the hash state. This is used when skipping - * over data, either at the beginning of a block, or skipping sections. - */ -static void ZSTD_ldm_gear_reset(ldmRollingHashState_t* state, - BYTE const* data, size_t minMatchLength) -{ - U64 hash = state->rolling; - size_t n = 0; - -#define GEAR_ITER_ONCE() do { \ - hash = (hash << 1) + ZSTD_ldm_gearTab[data[n] & 0xff]; \ - n += 1; \ - } while (0) - while (n + 3 < minMatchLength) { - GEAR_ITER_ONCE(); - GEAR_ITER_ONCE(); - GEAR_ITER_ONCE(); - GEAR_ITER_ONCE(); - } - while (n < minMatchLength) { - GEAR_ITER_ONCE(); - } -#undef GEAR_ITER_ONCE -} - -/** ZSTD_ldm_gear_feed(): - * - * Registers in the splits array all the split points found in the first - * size bytes following the data pointer. This function terminates when - * either all the data has been processed or LDM_BATCH_SIZE splits are - * present in the splits array. - * - * Precondition: The splits array must not be full. - * Returns: The number of bytes processed. */ -static size_t ZSTD_ldm_gear_feed(ldmRollingHashState_t* state, - BYTE const* data, size_t size, - size_t* splits, unsigned* numSplits) -{ - size_t n; - U64 hash, mask; - - hash = state->rolling; - mask = state->stopMask; - n = 0; - -#define GEAR_ITER_ONCE() do { \ - hash = (hash << 1) + ZSTD_ldm_gearTab[data[n] & 0xff]; \ - n += 1; \ - if (UNLIKELY((hash & mask) == 0)) { \ - splits[*numSplits] = n; \ - *numSplits += 1; \ - if (*numSplits == LDM_BATCH_SIZE) \ - goto done; \ - } \ - } while (0) - - while (n + 3 < size) { - GEAR_ITER_ONCE(); - GEAR_ITER_ONCE(); - GEAR_ITER_ONCE(); - GEAR_ITER_ONCE(); - } - while (n < size) { - GEAR_ITER_ONCE(); - } - -#undef GEAR_ITER_ONCE - -done: - state->rolling = hash; - return n; -} - -void ZSTD_ldm_adjustParameters(ldmParams_t* params, - ZSTD_compressionParameters const* cParams) -{ - params->windowLog = cParams->windowLog; - ZSTD_STATIC_ASSERT(LDM_BUCKET_SIZE_LOG <= ZSTD_LDM_BUCKETSIZELOG_MAX); - DEBUGLOG(4, "ZSTD_ldm_adjustParameters"); - if (!params->bucketSizeLog) params->bucketSizeLog = LDM_BUCKET_SIZE_LOG; - if (!params->minMatchLength) params->minMatchLength = LDM_MIN_MATCH_LENGTH; - if (params->hashLog == 0) { - params->hashLog = MAX(ZSTD_HASHLOG_MIN, params->windowLog - LDM_HASH_RLOG); - assert(params->hashLog <= ZSTD_HASHLOG_MAX); - } - if (params->hashRateLog == 0) { - params->hashRateLog = params->windowLog < params->hashLog - ? 0 - : params->windowLog - params->hashLog; - } - params->bucketSizeLog = MIN(params->bucketSizeLog, params->hashLog); -} - -size_t ZSTD_ldm_getTableSize(ldmParams_t params) -{ - size_t const ldmHSize = ((size_t)1) << params.hashLog; - size_t const ldmBucketSizeLog = MIN(params.bucketSizeLog, params.hashLog); - size_t const ldmBucketSize = ((size_t)1) << (params.hashLog - ldmBucketSizeLog); - size_t const totalSize = ZSTD_cwksp_alloc_size(ldmBucketSize) - + ZSTD_cwksp_alloc_size(ldmHSize * sizeof(ldmEntry_t)); - return params.enableLdm == ZSTD_ps_enable ? totalSize : 0; -} - -size_t ZSTD_ldm_getMaxNbSeq(ldmParams_t params, size_t maxChunkSize) -{ - return params.enableLdm == ZSTD_ps_enable ? (maxChunkSize / params.minMatchLength) : 0; -} - -/** ZSTD_ldm_getBucket() : - * Returns a pointer to the start of the bucket associated with hash. */ -static ldmEntry_t* ZSTD_ldm_getBucket( - ldmState_t* ldmState, size_t hash, ldmParams_t const ldmParams) -{ - return ldmState->hashTable + (hash << ldmParams.bucketSizeLog); -} - -/** ZSTD_ldm_insertEntry() : - * Insert the entry with corresponding hash into the hash table */ -static void ZSTD_ldm_insertEntry(ldmState_t* ldmState, - size_t const hash, const ldmEntry_t entry, - ldmParams_t const ldmParams) -{ - BYTE* const pOffset = ldmState->bucketOffsets + hash; - unsigned const offset = *pOffset; - - *(ZSTD_ldm_getBucket(ldmState, hash, ldmParams) + offset) = entry; - *pOffset = (BYTE)((offset + 1) & ((1u << ldmParams.bucketSizeLog) - 1)); - -} - -/** ZSTD_ldm_countBackwardsMatch() : - * Returns the number of bytes that match backwards before pIn and pMatch. - * - * We count only bytes where pMatch >= pBase and pIn >= pAnchor. */ -static size_t ZSTD_ldm_countBackwardsMatch( - const BYTE* pIn, const BYTE* pAnchor, - const BYTE* pMatch, const BYTE* pMatchBase) -{ - size_t matchLength = 0; - while (pIn > pAnchor && pMatch > pMatchBase && pIn[-1] == pMatch[-1]) { - pIn--; - pMatch--; - matchLength++; - } - return matchLength; -} - -/** ZSTD_ldm_countBackwardsMatch_2segments() : - * Returns the number of bytes that match backwards from pMatch, - * even with the backwards match spanning 2 different segments. - * - * On reaching `pMatchBase`, start counting from mEnd */ -static size_t ZSTD_ldm_countBackwardsMatch_2segments( - const BYTE* pIn, const BYTE* pAnchor, - const BYTE* pMatch, const BYTE* pMatchBase, - const BYTE* pExtDictStart, const BYTE* pExtDictEnd) -{ - size_t matchLength = ZSTD_ldm_countBackwardsMatch(pIn, pAnchor, pMatch, pMatchBase); - if (pMatch - matchLength != pMatchBase || pMatchBase == pExtDictStart) { - /* If backwards match is entirely in the extDict or prefix, immediately return */ - return matchLength; - } - DEBUGLOG(7, "ZSTD_ldm_countBackwardsMatch_2segments: found 2-parts backwards match (length in prefix==%zu)", matchLength); - matchLength += ZSTD_ldm_countBackwardsMatch(pIn - matchLength, pAnchor, pExtDictEnd, pExtDictStart); - DEBUGLOG(7, "final backwards match length = %zu", matchLength); - return matchLength; -} - -/** ZSTD_ldm_fillFastTables() : - * - * Fills the relevant tables for the ZSTD_fast and ZSTD_dfast strategies. - * This is similar to ZSTD_loadDictionaryContent. - * - * The tables for the other strategies are filled within their - * block compressors. */ -static size_t ZSTD_ldm_fillFastTables(ZSTD_matchState_t* ms, - void const* end) -{ - const BYTE* const iend = (const BYTE*)end; - - switch(ms->cParams.strategy) - { - case ZSTD_fast: - ZSTD_fillHashTable(ms, iend, ZSTD_dtlm_fast); - break; - - case ZSTD_dfast: - ZSTD_fillDoubleHashTable(ms, iend, ZSTD_dtlm_fast); - break; - - case ZSTD_greedy: - case ZSTD_lazy: - case ZSTD_lazy2: - case ZSTD_btlazy2: - case ZSTD_btopt: - case ZSTD_btultra: - case ZSTD_btultra2: - break; - default: - assert(0); /* not possible : not a valid strategy id */ - } - - return 0; -} - -void ZSTD_ldm_fillHashTable( - ldmState_t* ldmState, const BYTE* ip, - const BYTE* iend, ldmParams_t const* params) -{ - U32 const minMatchLength = params->minMatchLength; - U32 const hBits = params->hashLog - params->bucketSizeLog; - BYTE const* const base = ldmState->window.base; - BYTE const* const istart = ip; - ldmRollingHashState_t hashState; - size_t* const splits = ldmState->splitIndices; - unsigned numSplits; - - DEBUGLOG(5, "ZSTD_ldm_fillHashTable"); - - ZSTD_ldm_gear_init(&hashState, params); - while (ip < iend) { - size_t hashed; - unsigned n; - - numSplits = 0; - hashed = ZSTD_ldm_gear_feed(&hashState, ip, iend - ip, splits, &numSplits); - - for (n = 0; n < numSplits; n++) { - if (ip + splits[n] >= istart + minMatchLength) { - BYTE const* const split = ip + splits[n] - minMatchLength; - U64 const xxhash = XXH64(split, minMatchLength, 0); - U32 const hash = (U32)(xxhash & (((U32)1 << hBits) - 1)); - ldmEntry_t entry; - - entry.offset = (U32)(split - base); - entry.checksum = (U32)(xxhash >> 32); - ZSTD_ldm_insertEntry(ldmState, hash, entry, *params); - } - } - - ip += hashed; - } -} - - -/** ZSTD_ldm_limitTableUpdate() : - * - * Sets cctx->nextToUpdate to a position corresponding closer to anchor - * if it is far way - * (after a long match, only update tables a limited amount). */ -static void ZSTD_ldm_limitTableUpdate(ZSTD_matchState_t* ms, const BYTE* anchor) -{ - U32 const curr = (U32)(anchor - ms->window.base); - if (curr > ms->nextToUpdate + 1024) { - ms->nextToUpdate = - curr - MIN(512, curr - ms->nextToUpdate - 1024); - } -} - -static size_t ZSTD_ldm_generateSequences_internal( - ldmState_t* ldmState, rawSeqStore_t* rawSeqStore, - ldmParams_t const* params, void const* src, size_t srcSize) -{ - /* LDM parameters */ - int const extDict = ZSTD_window_hasExtDict(ldmState->window); - U32 const minMatchLength = params->minMatchLength; - U32 const entsPerBucket = 1U << params->bucketSizeLog; - U32 const hBits = params->hashLog - params->bucketSizeLog; - /* Prefix and extDict parameters */ - U32 const dictLimit = ldmState->window.dictLimit; - U32 const lowestIndex = extDict ? ldmState->window.lowLimit : dictLimit; - BYTE const* const base = ldmState->window.base; - BYTE const* const dictBase = extDict ? ldmState->window.dictBase : NULL; - BYTE const* const dictStart = extDict ? dictBase + lowestIndex : NULL; - BYTE const* const dictEnd = extDict ? dictBase + dictLimit : NULL; - BYTE const* const lowPrefixPtr = base + dictLimit; - /* Input bounds */ - BYTE const* const istart = (BYTE const*)src; - BYTE const* const iend = istart + srcSize; - BYTE const* const ilimit = iend - HASH_READ_SIZE; - /* Input positions */ - BYTE const* anchor = istart; - BYTE const* ip = istart; - /* Rolling hash state */ - ldmRollingHashState_t hashState; - /* Arrays for staged-processing */ - size_t* const splits = ldmState->splitIndices; - ldmMatchCandidate_t* const candidates = ldmState->matchCandidates; - unsigned numSplits; - - if (srcSize < minMatchLength) - return iend - anchor; - - /* Initialize the rolling hash state with the first minMatchLength bytes */ - ZSTD_ldm_gear_init(&hashState, params); - ZSTD_ldm_gear_reset(&hashState, ip, minMatchLength); - ip += minMatchLength; - - while (ip < ilimit) { - size_t hashed; - unsigned n; - - numSplits = 0; - hashed = ZSTD_ldm_gear_feed(&hashState, ip, ilimit - ip, - splits, &numSplits); - - for (n = 0; n < numSplits; n++) { - BYTE const* const split = ip + splits[n] - minMatchLength; - U64 const xxhash = XXH64(split, minMatchLength, 0); - U32 const hash = (U32)(xxhash & (((U32)1 << hBits) - 1)); - - candidates[n].split = split; - candidates[n].hash = hash; - candidates[n].checksum = (U32)(xxhash >> 32); - candidates[n].bucket = ZSTD_ldm_getBucket(ldmState, hash, *params); - PREFETCH_L1(candidates[n].bucket); - } - - for (n = 0; n < numSplits; n++) { - size_t forwardMatchLength = 0, backwardMatchLength = 0, - bestMatchLength = 0, mLength; - U32 offset; - BYTE const* const split = candidates[n].split; - U32 const checksum = candidates[n].checksum; - U32 const hash = candidates[n].hash; - ldmEntry_t* const bucket = candidates[n].bucket; - ldmEntry_t const* cur; - ldmEntry_t const* bestEntry = NULL; - ldmEntry_t newEntry; - - newEntry.offset = (U32)(split - base); - newEntry.checksum = checksum; - - /* If a split point would generate a sequence overlapping with - * the previous one, we merely register it in the hash table and - * move on */ - if (split < anchor) { - ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params); - continue; - } - - for (cur = bucket; cur < bucket + entsPerBucket; cur++) { - size_t curForwardMatchLength, curBackwardMatchLength, - curTotalMatchLength; - if (cur->checksum != checksum || cur->offset <= lowestIndex) { - continue; - } - if (extDict) { - BYTE const* const curMatchBase = - cur->offset < dictLimit ? dictBase : base; - BYTE const* const pMatch = curMatchBase + cur->offset; - BYTE const* const matchEnd = - cur->offset < dictLimit ? dictEnd : iend; - BYTE const* const lowMatchPtr = - cur->offset < dictLimit ? dictStart : lowPrefixPtr; - curForwardMatchLength = - ZSTD_count_2segments(split, pMatch, iend, matchEnd, lowPrefixPtr); - if (curForwardMatchLength < minMatchLength) { - continue; - } - curBackwardMatchLength = ZSTD_ldm_countBackwardsMatch_2segments( - split, anchor, pMatch, lowMatchPtr, dictStart, dictEnd); - } else { /* !extDict */ - BYTE const* const pMatch = base + cur->offset; - curForwardMatchLength = ZSTD_count(split, pMatch, iend); - if (curForwardMatchLength < minMatchLength) { - continue; - } - curBackwardMatchLength = - ZSTD_ldm_countBackwardsMatch(split, anchor, pMatch, lowPrefixPtr); - } - curTotalMatchLength = curForwardMatchLength + curBackwardMatchLength; - - if (curTotalMatchLength > bestMatchLength) { - bestMatchLength = curTotalMatchLength; - forwardMatchLength = curForwardMatchLength; - backwardMatchLength = curBackwardMatchLength; - bestEntry = cur; - } - } - - /* No match found -- insert an entry into the hash table - * and process the next candidate match */ - if (bestEntry == NULL) { - ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params); - continue; - } - - /* Match found */ - offset = (U32)(split - base) - bestEntry->offset; - mLength = forwardMatchLength + backwardMatchLength; - { - rawSeq* const seq = rawSeqStore->seq + rawSeqStore->size; - - /* Out of sequence storage */ - if (rawSeqStore->size == rawSeqStore->capacity) - return ERROR(dstSize_tooSmall); - seq->litLength = (U32)(split - backwardMatchLength - anchor); - seq->matchLength = (U32)mLength; - seq->offset = offset; - rawSeqStore->size++; - } - - /* Insert the current entry into the hash table --- it must be - * done after the previous block to avoid clobbering bestEntry */ - ZSTD_ldm_insertEntry(ldmState, hash, newEntry, *params); - - anchor = split + forwardMatchLength; - - /* If we find a match that ends after the data that we've hashed - * then we have a repeating, overlapping, pattern. E.g. all zeros. - * If one repetition of the pattern matches our `stopMask` then all - * repetitions will. We don't need to insert them all into out table, - * only the first one. So skip over overlapping matches. - * This is a major speed boost (20x) for compressing a single byte - * repeated, when that byte ends up in the table. - */ - if (anchor > ip + hashed) { - ZSTD_ldm_gear_reset(&hashState, anchor - minMatchLength, minMatchLength); - /* Continue the outer loop at anchor (ip + hashed == anchor). */ - ip = anchor - hashed; - break; - } - } - - ip += hashed; - } - - return iend - anchor; -} - -/*! ZSTD_ldm_reduceTable() : - * reduce table indexes by `reducerValue` */ -static void ZSTD_ldm_reduceTable(ldmEntry_t* const table, U32 const size, - U32 const reducerValue) -{ - U32 u; - for (u = 0; u < size; u++) { - if (table[u].offset < reducerValue) table[u].offset = 0; - else table[u].offset -= reducerValue; - } -} - -size_t ZSTD_ldm_generateSequences( - ldmState_t* ldmState, rawSeqStore_t* sequences, - ldmParams_t const* params, void const* src, size_t srcSize) -{ - U32 const maxDist = 1U << params->windowLog; - BYTE const* const istart = (BYTE const*)src; - BYTE const* const iend = istart + srcSize; - size_t const kMaxChunkSize = 1 << 20; - size_t const nbChunks = (srcSize / kMaxChunkSize) + ((srcSize % kMaxChunkSize) != 0); - size_t chunk; - size_t leftoverSize = 0; - - assert(ZSTD_CHUNKSIZE_MAX >= kMaxChunkSize); - /* Check that ZSTD_window_update() has been called for this chunk prior - * to passing it to this function. - */ - assert(ldmState->window.nextSrc >= (BYTE const*)src + srcSize); - /* The input could be very large (in zstdmt), so it must be broken up into - * chunks to enforce the maximum distance and handle overflow correction. - */ - assert(sequences->pos <= sequences->size); - assert(sequences->size <= sequences->capacity); - for (chunk = 0; chunk < nbChunks && sequences->size < sequences->capacity; ++chunk) { - BYTE const* const chunkStart = istart + chunk * kMaxChunkSize; - size_t const remaining = (size_t)(iend - chunkStart); - BYTE const *const chunkEnd = - (remaining < kMaxChunkSize) ? iend : chunkStart + kMaxChunkSize; - size_t const chunkSize = chunkEnd - chunkStart; - size_t newLeftoverSize; - size_t const prevSize = sequences->size; - - assert(chunkStart < iend); - /* 1. Perform overflow correction if necessary. */ - if (ZSTD_window_needOverflowCorrection(ldmState->window, 0, maxDist, ldmState->loadedDictEnd, chunkStart, chunkEnd)) { - U32 const ldmHSize = 1U << params->hashLog; - U32 const correction = ZSTD_window_correctOverflow( - &ldmState->window, /* cycleLog */ 0, maxDist, chunkStart); - ZSTD_ldm_reduceTable(ldmState->hashTable, ldmHSize, correction); - /* invalidate dictionaries on overflow correction */ - ldmState->loadedDictEnd = 0; - } - /* 2. We enforce the maximum offset allowed. - * - * kMaxChunkSize should be small enough that we don't lose too much of - * the window through early invalidation. - * TODO: * Test the chunk size. - * * Try invalidation after the sequence generation and test the - * the offset against maxDist directly. - * - * NOTE: Because of dictionaries + sequence splitting we MUST make sure - * that any offset used is valid at the END of the sequence, since it may - * be split into two sequences. This condition holds when using - * ZSTD_window_enforceMaxDist(), but if we move to checking offsets - * against maxDist directly, we'll have to carefully handle that case. - */ - ZSTD_window_enforceMaxDist(&ldmState->window, chunkEnd, maxDist, &ldmState->loadedDictEnd, NULL); - /* 3. Generate the sequences for the chunk, and get newLeftoverSize. */ - newLeftoverSize = ZSTD_ldm_generateSequences_internal( - ldmState, sequences, params, chunkStart, chunkSize); - if (ZSTD_isError(newLeftoverSize)) - return newLeftoverSize; - /* 4. We add the leftover literals from previous iterations to the first - * newly generated sequence, or add the `newLeftoverSize` if none are - * generated. - */ - /* Prepend the leftover literals from the last call */ - if (prevSize < sequences->size) { - sequences->seq[prevSize].litLength += (U32)leftoverSize; - leftoverSize = newLeftoverSize; - } else { - assert(newLeftoverSize == chunkSize); - leftoverSize += chunkSize; - } - } - return 0; -} - -void -ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize, U32 const minMatch) -{ - while (srcSize > 0 && rawSeqStore->pos < rawSeqStore->size) { - rawSeq* seq = rawSeqStore->seq + rawSeqStore->pos; - if (srcSize <= seq->litLength) { - /* Skip past srcSize literals */ - seq->litLength -= (U32)srcSize; - return; - } - srcSize -= seq->litLength; - seq->litLength = 0; - if (srcSize < seq->matchLength) { - /* Skip past the first srcSize of the match */ - seq->matchLength -= (U32)srcSize; - if (seq->matchLength < minMatch) { - /* The match is too short, omit it */ - if (rawSeqStore->pos + 1 < rawSeqStore->size) { - seq[1].litLength += seq[0].matchLength; - } - rawSeqStore->pos++; - } - return; - } - srcSize -= seq->matchLength; - seq->matchLength = 0; - rawSeqStore->pos++; - } -} - -/** - * If the sequence length is longer than remaining then the sequence is split - * between this block and the next. - * - * Returns the current sequence to handle, or if the rest of the block should - * be literals, it returns a sequence with offset == 0. - */ -static rawSeq maybeSplitSequence(rawSeqStore_t* rawSeqStore, - U32 const remaining, U32 const minMatch) -{ - rawSeq sequence = rawSeqStore->seq[rawSeqStore->pos]; - assert(sequence.offset > 0); - /* Likely: No partial sequence */ - if (remaining >= sequence.litLength + sequence.matchLength) { - rawSeqStore->pos++; - return sequence; - } - /* Cut the sequence short (offset == 0 ==> rest is literals). */ - if (remaining <= sequence.litLength) { - sequence.offset = 0; - } else if (remaining < sequence.litLength + sequence.matchLength) { - sequence.matchLength = remaining - sequence.litLength; - if (sequence.matchLength < minMatch) { - sequence.offset = 0; - } - } - /* Skip past `remaining` bytes for the future sequences. */ - ZSTD_ldm_skipSequences(rawSeqStore, remaining, minMatch); - return sequence; -} - -void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes) { - U32 currPos = (U32)(rawSeqStore->posInSequence + nbBytes); - while (currPos && rawSeqStore->pos < rawSeqStore->size) { - rawSeq currSeq = rawSeqStore->seq[rawSeqStore->pos]; - if (currPos >= currSeq.litLength + currSeq.matchLength) { - currPos -= currSeq.litLength + currSeq.matchLength; - rawSeqStore->pos++; - } else { - rawSeqStore->posInSequence = currPos; - break; - } - } - if (currPos == 0 || rawSeqStore->pos == rawSeqStore->size) { - rawSeqStore->posInSequence = 0; - } -} - -size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore, - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - ZSTD_paramSwitch_e useRowMatchFinder, - void const* src, size_t srcSize) -{ - const ZSTD_compressionParameters* const cParams = &ms->cParams; - unsigned const minMatch = cParams->minMatch; - ZSTD_blockCompressor const blockCompressor = - ZSTD_selectBlockCompressor(cParams->strategy, useRowMatchFinder, ZSTD_matchState_dictMode(ms)); - /* Input bounds */ - BYTE const* const istart = (BYTE const*)src; - BYTE const* const iend = istart + srcSize; - /* Input positions */ - BYTE const* ip = istart; - - DEBUGLOG(5, "ZSTD_ldm_blockCompress: srcSize=%zu", srcSize); - /* If using opt parser, use LDMs only as candidates rather than always accepting them */ - if (cParams->strategy >= ZSTD_btopt) { - size_t lastLLSize; - ms->ldmSeqStore = rawSeqStore; - lastLLSize = blockCompressor(ms, seqStore, rep, src, srcSize); - ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore, srcSize); - return lastLLSize; - } - - assert(rawSeqStore->pos <= rawSeqStore->size); - assert(rawSeqStore->size <= rawSeqStore->capacity); - /* Loop through each sequence and apply the block compressor to the literals */ - while (rawSeqStore->pos < rawSeqStore->size && ip < iend) { - /* maybeSplitSequence updates rawSeqStore->pos */ - rawSeq const sequence = maybeSplitSequence(rawSeqStore, - (U32)(iend - ip), minMatch); - int i; - /* End signal */ - if (sequence.offset == 0) - break; - - assert(ip + sequence.litLength + sequence.matchLength <= iend); - - /* Fill tables for block compressor */ - ZSTD_ldm_limitTableUpdate(ms, ip); - ZSTD_ldm_fillFastTables(ms, ip); - /* Run the block compressor */ - DEBUGLOG(5, "pos %u : calling block compressor on segment of size %u", (unsigned)(ip-istart), sequence.litLength); - { - size_t const newLitLength = - blockCompressor(ms, seqStore, rep, ip, sequence.litLength); - ip += sequence.litLength; - /* Update the repcodes */ - for (i = ZSTD_REP_NUM - 1; i > 0; i--) - rep[i] = rep[i-1]; - rep[0] = sequence.offset; - /* Store the sequence */ - ZSTD_storeSeq(seqStore, newLitLength, ip - newLitLength, iend, - STORE_OFFSET(sequence.offset), - sequence.matchLength); - ip += sequence.matchLength; - } - } - /* Fill the tables for the block compressor */ - ZSTD_ldm_limitTableUpdate(ms, ip); - ZSTD_ldm_fillFastTables(ms, ip); - /* Compress the last literals */ - return blockCompressor(ms, seqStore, rep, ip, iend - ip); -} diff --git a/dep/zstd/lib/compress/zstd_ldm.h b/dep/zstd/lib/compress/zstd_ldm.h deleted file mode 100644 index 4e68dbf52..000000000 --- a/dep/zstd/lib/compress/zstd_ldm.h +++ /dev/null @@ -1,117 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_LDM_H -#define ZSTD_LDM_H - -#if defined (__cplusplus) -extern "C" { -#endif - -#include "zstd_compress_internal.h" /* ldmParams_t, U32 */ -#include "../zstd.h" /* ZSTD_CCtx, size_t */ - -/*-************************************* -* Long distance matching -***************************************/ - -#define ZSTD_LDM_DEFAULT_WINDOW_LOG ZSTD_WINDOWLOG_LIMIT_DEFAULT - -void ZSTD_ldm_fillHashTable( - ldmState_t* state, const BYTE* ip, - const BYTE* iend, ldmParams_t const* params); - -/** - * ZSTD_ldm_generateSequences(): - * - * Generates the sequences using the long distance match finder. - * Generates long range matching sequences in `sequences`, which parse a prefix - * of the source. `sequences` must be large enough to store every sequence, - * which can be checked with `ZSTD_ldm_getMaxNbSeq()`. - * @returns 0 or an error code. - * - * NOTE: The user must have called ZSTD_window_update() for all of the input - * they have, even if they pass it to ZSTD_ldm_generateSequences() in chunks. - * NOTE: This function returns an error if it runs out of space to store - * sequences. - */ -size_t ZSTD_ldm_generateSequences( - ldmState_t* ldms, rawSeqStore_t* sequences, - ldmParams_t const* params, void const* src, size_t srcSize); - -/** - * ZSTD_ldm_blockCompress(): - * - * Compresses a block using the predefined sequences, along with a secondary - * block compressor. The literals section of every sequence is passed to the - * secondary block compressor, and those sequences are interspersed with the - * predefined sequences. Returns the length of the last literals. - * Updates `rawSeqStore.pos` to indicate how many sequences have been consumed. - * `rawSeqStore.seq` may also be updated to split the last sequence between two - * blocks. - * @return The length of the last literals. - * - * NOTE: The source must be at most the maximum block size, but the predefined - * sequences can be any size, and may be longer than the block. In the case that - * they are longer than the block, the last sequences may need to be split into - * two. We handle that case correctly, and update `rawSeqStore` appropriately. - * NOTE: This function does not return any errors. - */ -size_t ZSTD_ldm_blockCompress(rawSeqStore_t* rawSeqStore, - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - ZSTD_paramSwitch_e useRowMatchFinder, - void const* src, size_t srcSize); - -/** - * ZSTD_ldm_skipSequences(): - * - * Skip past `srcSize` bytes worth of sequences in `rawSeqStore`. - * Avoids emitting matches less than `minMatch` bytes. - * Must be called for data that is not passed to ZSTD_ldm_blockCompress(). - */ -void ZSTD_ldm_skipSequences(rawSeqStore_t* rawSeqStore, size_t srcSize, - U32 const minMatch); - -/* ZSTD_ldm_skipRawSeqStoreBytes(): - * Moves forward in rawSeqStore by nbBytes, updating fields 'pos' and 'posInSequence'. - * Not to be used in conjunction with ZSTD_ldm_skipSequences(). - * Must be called for data with is not passed to ZSTD_ldm_blockCompress(). - */ -void ZSTD_ldm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes); - -/** ZSTD_ldm_getTableSize() : - * Estimate the space needed for long distance matching tables or 0 if LDM is - * disabled. - */ -size_t ZSTD_ldm_getTableSize(ldmParams_t params); - -/** ZSTD_ldm_getSeqSpace() : - * Return an upper bound on the number of sequences that can be produced by - * the long distance matcher, or 0 if LDM is disabled. - */ -size_t ZSTD_ldm_getMaxNbSeq(ldmParams_t params, size_t maxChunkSize); - -/** ZSTD_ldm_adjustParameters() : - * If the params->hashRateLog is not set, set it to its default value based on - * windowLog and params->hashLog. - * - * Ensures that params->bucketSizeLog is <= params->hashLog (setting it to - * params->hashLog if it is not). - * - * Ensures that the minMatchLength >= targetLength during optimal parsing. - */ -void ZSTD_ldm_adjustParameters(ldmParams_t* params, - ZSTD_compressionParameters const* cParams); - -#if defined (__cplusplus) -} -#endif - -#endif /* ZSTD_FAST_H */ diff --git a/dep/zstd/lib/compress/zstd_ldm_geartab.h b/dep/zstd/lib/compress/zstd_ldm_geartab.h deleted file mode 100644 index 647f865be..000000000 --- a/dep/zstd/lib/compress/zstd_ldm_geartab.h +++ /dev/null @@ -1,106 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_LDM_GEARTAB_H -#define ZSTD_LDM_GEARTAB_H - -#include "../common/compiler.h" /* UNUSED_ATTR */ -#include "../common/mem.h" /* U64 */ - -static UNUSED_ATTR const U64 ZSTD_ldm_gearTab[256] = { - 0xf5b8f72c5f77775c, 0x84935f266b7ac412, 0xb647ada9ca730ccc, - 0xb065bb4b114fb1de, 0x34584e7e8c3a9fd0, 0x4e97e17c6ae26b05, - 0x3a03d743bc99a604, 0xcecd042422c4044f, 0x76de76c58524259e, - 0x9c8528f65badeaca, 0x86563706e2097529, 0x2902475fa375d889, - 0xafb32a9739a5ebe6, 0xce2714da3883e639, 0x21eaf821722e69e, - 0x37b628620b628, 0x49a8d455d88caf5, 0x8556d711e6958140, - 0x4f7ae74fc605c1f, 0x829f0c3468bd3a20, 0x4ffdc885c625179e, - 0x8473de048a3daf1b, 0x51008822b05646b2, 0x69d75d12b2d1cc5f, - 0x8c9d4a19159154bc, 0xc3cc10f4abbd4003, 0xd06ddc1cecb97391, - 0xbe48e6e7ed80302e, 0x3481db31cee03547, 0xacc3f67cdaa1d210, - 0x65cb771d8c7f96cc, 0x8eb27177055723dd, 0xc789950d44cd94be, - 0x934feadc3700b12b, 0x5e485f11edbdf182, 0x1e2e2a46fd64767a, - 0x2969ca71d82efa7c, 0x9d46e9935ebbba2e, 0xe056b67e05e6822b, - 0x94d73f55739d03a0, 0xcd7010bdb69b5a03, 0x455ef9fcd79b82f4, - 0x869cb54a8749c161, 0x38d1a4fa6185d225, 0xb475166f94bbe9bb, - 0xa4143548720959f1, 0x7aed4780ba6b26ba, 0xd0ce264439e02312, - 0x84366d746078d508, 0xa8ce973c72ed17be, 0x21c323a29a430b01, - 0x9962d617e3af80ee, 0xab0ce91d9c8cf75b, 0x530e8ee6d19a4dbc, - 0x2ef68c0cf53f5d72, 0xc03a681640a85506, 0x496e4e9f9c310967, - 0x78580472b59b14a0, 0x273824c23b388577, 0x66bf923ad45cb553, - 0x47ae1a5a2492ba86, 0x35e304569e229659, 0x4765182a46870b6f, - 0x6cbab625e9099412, 0xddac9a2e598522c1, 0x7172086e666624f2, - 0xdf5003ca503b7837, 0x88c0c1db78563d09, 0x58d51865acfc289d, - 0x177671aec65224f1, 0xfb79d8a241e967d7, 0x2be1e101cad9a49a, - 0x6625682f6e29186b, 0x399553457ac06e50, 0x35dffb4c23abb74, - 0x429db2591f54aade, 0xc52802a8037d1009, 0x6acb27381f0b25f3, - 0xf45e2551ee4f823b, 0x8b0ea2d99580c2f7, 0x3bed519cbcb4e1e1, - 0xff452823dbb010a, 0x9d42ed614f3dd267, 0x5b9313c06257c57b, - 0xa114b8008b5e1442, 0xc1fe311c11c13d4b, 0x66e8763ea34c5568, - 0x8b982af1c262f05d, 0xee8876faaa75fbb7, 0x8a62a4d0d172bb2a, - 0xc13d94a3b7449a97, 0x6dbbba9dc15d037c, 0xc786101f1d92e0f1, - 0xd78681a907a0b79b, 0xf61aaf2962c9abb9, 0x2cfd16fcd3cb7ad9, - 0x868c5b6744624d21, 0x25e650899c74ddd7, 0xba042af4a7c37463, - 0x4eb1a539465a3eca, 0xbe09dbf03b05d5ca, 0x774e5a362b5472ba, - 0x47a1221229d183cd, 0x504b0ca18ef5a2df, 0xdffbdfbde2456eb9, - 0x46cd2b2fbee34634, 0xf2aef8fe819d98c3, 0x357f5276d4599d61, - 0x24a5483879c453e3, 0x88026889192b4b9, 0x28da96671782dbec, - 0x4ef37c40588e9aaa, 0x8837b90651bc9fb3, 0xc164f741d3f0e5d6, - 0xbc135a0a704b70ba, 0x69cd868f7622ada, 0xbc37ba89e0b9c0ab, - 0x47c14a01323552f6, 0x4f00794bacee98bb, 0x7107de7d637a69d5, - 0x88af793bb6f2255e, 0xf3c6466b8799b598, 0xc288c616aa7f3b59, - 0x81ca63cf42fca3fd, 0x88d85ace36a2674b, 0xd056bd3792389e7, - 0xe55c396c4e9dd32d, 0xbefb504571e6c0a6, 0x96ab32115e91e8cc, - 0xbf8acb18de8f38d1, 0x66dae58801672606, 0x833b6017872317fb, - 0xb87c16f2d1c92864, 0xdb766a74e58b669c, 0x89659f85c61417be, - 0xc8daad856011ea0c, 0x76a4b565b6fe7eae, 0xa469d085f6237312, - 0xaaf0365683a3e96c, 0x4dbb746f8424f7b8, 0x638755af4e4acc1, - 0x3d7807f5bde64486, 0x17be6d8f5bbb7639, 0x903f0cd44dc35dc, - 0x67b672eafdf1196c, 0xa676ff93ed4c82f1, 0x521d1004c5053d9d, - 0x37ba9ad09ccc9202, 0x84e54d297aacfb51, 0xa0b4b776a143445, - 0x820d471e20b348e, 0x1874383cb83d46dc, 0x97edeec7a1efe11c, - 0xb330e50b1bdc42aa, 0x1dd91955ce70e032, 0xa514cdb88f2939d5, - 0x2791233fd90db9d3, 0x7b670a4cc50f7a9b, 0x77c07d2a05c6dfa5, - 0xe3778b6646d0a6fa, 0xb39c8eda47b56749, 0x933ed448addbef28, - 0xaf846af6ab7d0bf4, 0xe5af208eb666e49, 0x5e6622f73534cd6a, - 0x297daeca42ef5b6e, 0x862daef3d35539a6, 0xe68722498f8e1ea9, - 0x981c53093dc0d572, 0xfa09b0bfbf86fbf5, 0x30b1e96166219f15, - 0x70e7d466bdc4fb83, 0x5a66736e35f2a8e9, 0xcddb59d2b7c1baef, - 0xd6c7d247d26d8996, 0xea4e39eac8de1ba3, 0x539c8bb19fa3aff2, - 0x9f90e4c5fd508d8, 0xa34e5956fbaf3385, 0x2e2f8e151d3ef375, - 0x173691e9b83faec1, 0xb85a8d56bf016379, 0x8382381267408ae3, - 0xb90f901bbdc0096d, 0x7c6ad32933bcec65, 0x76bb5e2f2c8ad595, - 0x390f851a6cf46d28, 0xc3e6064da1c2da72, 0xc52a0c101cfa5389, - 0xd78eaf84a3fbc530, 0x3781b9e2288b997e, 0x73c2f6dea83d05c4, - 0x4228e364c5b5ed7, 0x9d7a3edf0da43911, 0x8edcfeda24686756, - 0x5e7667a7b7a9b3a1, 0x4c4f389fa143791d, 0xb08bc1023da7cddc, - 0x7ab4be3ae529b1cc, 0x754e6132dbe74ff9, 0x71635442a839df45, - 0x2f6fb1643fbe52de, 0x961e0a42cf7a8177, 0xf3b45d83d89ef2ea, - 0xee3de4cf4a6e3e9b, 0xcd6848542c3295e7, 0xe4cee1664c78662f, - 0x9947548b474c68c4, 0x25d73777a5ed8b0b, 0xc915b1d636b7fc, - 0x21c2ba75d9b0d2da, 0x5f6b5dcf608a64a1, 0xdcf333255ff9570c, - 0x633b922418ced4ee, 0xc136dde0b004b34a, 0x58cc83b05d4b2f5a, - 0x5eb424dda28e42d2, 0x62df47369739cd98, 0xb4e0b42485e4ce17, - 0x16e1f0c1f9a8d1e7, 0x8ec3916707560ebf, 0x62ba6e2df2cc9db3, - 0xcbf9f4ff77d83a16, 0x78d9d7d07d2bbcc4, 0xef554ce1e02c41f4, - 0x8d7581127eccf94d, 0xa9b53336cb3c8a05, 0x38c42c0bf45c4f91, - 0x640893cdf4488863, 0x80ec34bc575ea568, 0x39f324f5b48eaa40, - 0xe9d9ed1f8eff527f, 0x9224fc058cc5a214, 0xbaba00b04cfe7741, - 0x309a9f120fcf52af, 0xa558f3ec65626212, 0x424bec8b7adabe2f, - 0x41622513a6aea433, 0xb88da2d5324ca798, 0xd287733b245528a4, - 0x9a44697e6d68aec3, 0x7b1093be2f49bb28, 0x50bbec632e3d8aad, - 0x6cd90723e1ea8283, 0x897b9e7431b02bf3, 0x219efdcb338a7047, - 0x3b0311f0a27c0656, 0xdb17bf91c0db96e7, 0x8cd4fd6b4e85a5b2, - 0xfab071054ba6409d, 0x40d6fe831fa9dfd9, 0xaf358debad7d791e, - 0xeb8d0e25a65e3e58, 0xbbcbd3df14e08580, 0xcf751f27ecdab2b, - 0x2b4da14f2613d8f4 -}; - -#endif /* ZSTD_LDM_GEARTAB_H */ diff --git a/dep/zstd/lib/compress/zstd_opt.c b/dep/zstd/lib/compress/zstd_opt.c deleted file mode 100644 index 1b1ddad42..000000000 --- a/dep/zstd/lib/compress/zstd_opt.c +++ /dev/null @@ -1,1446 +0,0 @@ -/* - * Copyright (c) Przemyslaw Skibinski, Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#include "zstd_compress_internal.h" -#include "hist.h" -#include "zstd_opt.h" - - -#define ZSTD_LITFREQ_ADD 2 /* scaling factor for litFreq, so that frequencies adapt faster to new stats */ -#define ZSTD_MAX_PRICE (1<<30) - -#define ZSTD_PREDEF_THRESHOLD 1024 /* if srcSize < ZSTD_PREDEF_THRESHOLD, symbols' cost is assumed static, directly determined by pre-defined distributions */ - - -/*-************************************* -* Price functions for optimal parser -***************************************/ - -#if 0 /* approximation at bit level (for tests) */ -# define BITCOST_ACCURACY 0 -# define BITCOST_MULTIPLIER (1 << BITCOST_ACCURACY) -# define WEIGHT(stat, opt) ((void)opt, ZSTD_bitWeight(stat)) -#elif 0 /* fractional bit accuracy (for tests) */ -# define BITCOST_ACCURACY 8 -# define BITCOST_MULTIPLIER (1 << BITCOST_ACCURACY) -# define WEIGHT(stat,opt) ((void)opt, ZSTD_fracWeight(stat)) -#else /* opt==approx, ultra==accurate */ -# define BITCOST_ACCURACY 8 -# define BITCOST_MULTIPLIER (1 << BITCOST_ACCURACY) -# define WEIGHT(stat,opt) (opt ? ZSTD_fracWeight(stat) : ZSTD_bitWeight(stat)) -#endif - -MEM_STATIC U32 ZSTD_bitWeight(U32 stat) -{ - return (ZSTD_highbit32(stat+1) * BITCOST_MULTIPLIER); -} - -MEM_STATIC U32 ZSTD_fracWeight(U32 rawStat) -{ - U32 const stat = rawStat + 1; - U32 const hb = ZSTD_highbit32(stat); - U32 const BWeight = hb * BITCOST_MULTIPLIER; - U32 const FWeight = (stat << BITCOST_ACCURACY) >> hb; - U32 const weight = BWeight + FWeight; - assert(hb + BITCOST_ACCURACY < 31); - return weight; -} - -#if (DEBUGLEVEL>=2) -/* debugging function, - * @return price in bytes as fractional value - * for debug messages only */ -MEM_STATIC double ZSTD_fCost(U32 price) -{ - return (double)price / (BITCOST_MULTIPLIER*8); -} -#endif - -static int ZSTD_compressedLiterals(optState_t const* const optPtr) -{ - return optPtr->literalCompressionMode != ZSTD_ps_disable; -} - -static void ZSTD_setBasePrices(optState_t* optPtr, int optLevel) -{ - if (ZSTD_compressedLiterals(optPtr)) - optPtr->litSumBasePrice = WEIGHT(optPtr->litSum, optLevel); - optPtr->litLengthSumBasePrice = WEIGHT(optPtr->litLengthSum, optLevel); - optPtr->matchLengthSumBasePrice = WEIGHT(optPtr->matchLengthSum, optLevel); - optPtr->offCodeSumBasePrice = WEIGHT(optPtr->offCodeSum, optLevel); -} - - -static U32 sum_u32(const unsigned table[], size_t nbElts) -{ - size_t n; - U32 total = 0; - for (n=0; n> shift); - sum += table[s]; - } - return sum; -} - -/* ZSTD_scaleStats() : - * reduce all elements in table is sum too large - * return the resulting sum of elements */ -static U32 ZSTD_scaleStats(unsigned* table, U32 lastEltIndex, U32 logTarget) -{ - U32 const prevsum = sum_u32(table, lastEltIndex+1); - U32 const factor = prevsum >> logTarget; - DEBUGLOG(5, "ZSTD_scaleStats (nbElts=%u, target=%u)", (unsigned)lastEltIndex+1, (unsigned)logTarget); - assert(logTarget < 30); - if (factor <= 1) return prevsum; - return ZSTD_downscaleStats(table, lastEltIndex, ZSTD_highbit32(factor)); -} - -/* ZSTD_rescaleFreqs() : - * if first block (detected by optPtr->litLengthSum == 0) : init statistics - * take hints from dictionary if there is one - * and init from zero if there is none, - * using src for literals stats, and baseline stats for sequence symbols - * otherwise downscale existing stats, to be used as seed for next block. - */ -static void -ZSTD_rescaleFreqs(optState_t* const optPtr, - const BYTE* const src, size_t const srcSize, - int const optLevel) -{ - int const compressedLiterals = ZSTD_compressedLiterals(optPtr); - DEBUGLOG(5, "ZSTD_rescaleFreqs (srcSize=%u)", (unsigned)srcSize); - optPtr->priceType = zop_dynamic; - - if (optPtr->litLengthSum == 0) { /* first block : init */ - if (srcSize <= ZSTD_PREDEF_THRESHOLD) { /* heuristic */ - DEBUGLOG(5, "(srcSize <= ZSTD_PREDEF_THRESHOLD) => zop_predef"); - optPtr->priceType = zop_predef; - } - - assert(optPtr->symbolCosts != NULL); - if (optPtr->symbolCosts->huf.repeatMode == HUF_repeat_valid) { - /* huffman table presumed generated by dictionary */ - optPtr->priceType = zop_dynamic; - - if (compressedLiterals) { - unsigned lit; - assert(optPtr->litFreq != NULL); - optPtr->litSum = 0; - for (lit=0; lit<=MaxLit; lit++) { - U32 const scaleLog = 11; /* scale to 2K */ - U32 const bitCost = HUF_getNbBitsFromCTable(optPtr->symbolCosts->huf.CTable, lit); - assert(bitCost <= scaleLog); - optPtr->litFreq[lit] = bitCost ? 1 << (scaleLog-bitCost) : 1 /*minimum to calculate cost*/; - optPtr->litSum += optPtr->litFreq[lit]; - } } - - { unsigned ll; - FSE_CState_t llstate; - FSE_initCState(&llstate, optPtr->symbolCosts->fse.litlengthCTable); - optPtr->litLengthSum = 0; - for (ll=0; ll<=MaxLL; ll++) { - U32 const scaleLog = 10; /* scale to 1K */ - U32 const bitCost = FSE_getMaxNbBits(llstate.symbolTT, ll); - assert(bitCost < scaleLog); - optPtr->litLengthFreq[ll] = bitCost ? 1 << (scaleLog-bitCost) : 1 /*minimum to calculate cost*/; - optPtr->litLengthSum += optPtr->litLengthFreq[ll]; - } } - - { unsigned ml; - FSE_CState_t mlstate; - FSE_initCState(&mlstate, optPtr->symbolCosts->fse.matchlengthCTable); - optPtr->matchLengthSum = 0; - for (ml=0; ml<=MaxML; ml++) { - U32 const scaleLog = 10; - U32 const bitCost = FSE_getMaxNbBits(mlstate.symbolTT, ml); - assert(bitCost < scaleLog); - optPtr->matchLengthFreq[ml] = bitCost ? 1 << (scaleLog-bitCost) : 1 /*minimum to calculate cost*/; - optPtr->matchLengthSum += optPtr->matchLengthFreq[ml]; - } } - - { unsigned of; - FSE_CState_t ofstate; - FSE_initCState(&ofstate, optPtr->symbolCosts->fse.offcodeCTable); - optPtr->offCodeSum = 0; - for (of=0; of<=MaxOff; of++) { - U32 const scaleLog = 10; - U32 const bitCost = FSE_getMaxNbBits(ofstate.symbolTT, of); - assert(bitCost < scaleLog); - optPtr->offCodeFreq[of] = bitCost ? 1 << (scaleLog-bitCost) : 1 /*minimum to calculate cost*/; - optPtr->offCodeSum += optPtr->offCodeFreq[of]; - } } - - } else { /* not a dictionary */ - - assert(optPtr->litFreq != NULL); - if (compressedLiterals) { - unsigned lit = MaxLit; - HIST_count_simple(optPtr->litFreq, &lit, src, srcSize); /* use raw first block to init statistics */ - optPtr->litSum = ZSTD_downscaleStats(optPtr->litFreq, MaxLit, 8); - } - - { unsigned const baseLLfreqs[MaxLL+1] = { - 4, 2, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1 - }; - ZSTD_memcpy(optPtr->litLengthFreq, baseLLfreqs, sizeof(baseLLfreqs)); - optPtr->litLengthSum = sum_u32(baseLLfreqs, MaxLL+1); - } - - { unsigned ml; - for (ml=0; ml<=MaxML; ml++) - optPtr->matchLengthFreq[ml] = 1; - } - optPtr->matchLengthSum = MaxML+1; - - { unsigned const baseOFCfreqs[MaxOff+1] = { - 6, 2, 1, 1, 2, 3, 4, 4, - 4, 3, 2, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1, - 1, 1, 1, 1, 1, 1, 1, 1 - }; - ZSTD_memcpy(optPtr->offCodeFreq, baseOFCfreqs, sizeof(baseOFCfreqs)); - optPtr->offCodeSum = sum_u32(baseOFCfreqs, MaxOff+1); - } - - - } - - } else { /* new block : re-use previous statistics, scaled down */ - - if (compressedLiterals) - optPtr->litSum = ZSTD_scaleStats(optPtr->litFreq, MaxLit, 12); - optPtr->litLengthSum = ZSTD_scaleStats(optPtr->litLengthFreq, MaxLL, 11); - optPtr->matchLengthSum = ZSTD_scaleStats(optPtr->matchLengthFreq, MaxML, 11); - optPtr->offCodeSum = ZSTD_scaleStats(optPtr->offCodeFreq, MaxOff, 11); - } - - ZSTD_setBasePrices(optPtr, optLevel); -} - -/* ZSTD_rawLiteralsCost() : - * price of literals (only) in specified segment (which length can be 0). - * does not include price of literalLength symbol */ -static U32 ZSTD_rawLiteralsCost(const BYTE* const literals, U32 const litLength, - const optState_t* const optPtr, - int optLevel) -{ - if (litLength == 0) return 0; - - if (!ZSTD_compressedLiterals(optPtr)) - return (litLength << 3) * BITCOST_MULTIPLIER; /* Uncompressed - 8 bytes per literal. */ - - if (optPtr->priceType == zop_predef) - return (litLength*6) * BITCOST_MULTIPLIER; /* 6 bit per literal - no statistic used */ - - /* dynamic statistics */ - { U32 price = litLength * optPtr->litSumBasePrice; - U32 u; - for (u=0; u < litLength; u++) { - assert(WEIGHT(optPtr->litFreq[literals[u]], optLevel) <= optPtr->litSumBasePrice); /* literal cost should never be negative */ - price -= WEIGHT(optPtr->litFreq[literals[u]], optLevel); - } - return price; - } -} - -/* ZSTD_litLengthPrice() : - * cost of literalLength symbol */ -static U32 ZSTD_litLengthPrice(U32 const litLength, const optState_t* const optPtr, int optLevel) -{ - assert(litLength <= ZSTD_BLOCKSIZE_MAX); - if (optPtr->priceType == zop_predef) - return WEIGHT(litLength, optLevel); - /* We can't compute the litLength price for sizes >= ZSTD_BLOCKSIZE_MAX - * because it isn't representable in the zstd format. So instead just - * call it 1 bit more than ZSTD_BLOCKSIZE_MAX - 1. In this case the block - * would be all literals. - */ - if (litLength == ZSTD_BLOCKSIZE_MAX) - return BITCOST_MULTIPLIER + ZSTD_litLengthPrice(ZSTD_BLOCKSIZE_MAX - 1, optPtr, optLevel); - - /* dynamic statistics */ - { U32 const llCode = ZSTD_LLcode(litLength); - return (LL_bits[llCode] * BITCOST_MULTIPLIER) - + optPtr->litLengthSumBasePrice - - WEIGHT(optPtr->litLengthFreq[llCode], optLevel); - } -} - -/* ZSTD_getMatchPrice() : - * Provides the cost of the match part (offset + matchLength) of a sequence - * Must be combined with ZSTD_fullLiteralsCost() to get the full cost of a sequence. - * @offcode : expects a scale where 0,1,2 are repcodes 1-3, and 3+ are real_offsets+2 - * @optLevel: when <2, favors small offset for decompression speed (improved cache efficiency) - */ -FORCE_INLINE_TEMPLATE U32 -ZSTD_getMatchPrice(U32 const offcode, - U32 const matchLength, - const optState_t* const optPtr, - int const optLevel) -{ - U32 price; - U32 const offCode = ZSTD_highbit32(STORED_TO_OFFBASE(offcode)); - U32 const mlBase = matchLength - MINMATCH; - assert(matchLength >= MINMATCH); - - if (optPtr->priceType == zop_predef) /* fixed scheme, do not use statistics */ - return WEIGHT(mlBase, optLevel) + ((16 + offCode) * BITCOST_MULTIPLIER); - - /* dynamic statistics */ - price = (offCode * BITCOST_MULTIPLIER) + (optPtr->offCodeSumBasePrice - WEIGHT(optPtr->offCodeFreq[offCode], optLevel)); - if ((optLevel<2) /*static*/ && offCode >= 20) - price += (offCode-19)*2 * BITCOST_MULTIPLIER; /* handicap for long distance offsets, favor decompression speed */ - - /* match Length */ - { U32 const mlCode = ZSTD_MLcode(mlBase); - price += (ML_bits[mlCode] * BITCOST_MULTIPLIER) + (optPtr->matchLengthSumBasePrice - WEIGHT(optPtr->matchLengthFreq[mlCode], optLevel)); - } - - price += BITCOST_MULTIPLIER / 5; /* heuristic : make matches a bit more costly to favor less sequences -> faster decompression speed */ - - DEBUGLOG(8, "ZSTD_getMatchPrice(ml:%u) = %u", matchLength, price); - return price; -} - -/* ZSTD_updateStats() : - * assumption : literals + litLengtn <= iend */ -static void ZSTD_updateStats(optState_t* const optPtr, - U32 litLength, const BYTE* literals, - U32 offsetCode, U32 matchLength) -{ - /* literals */ - if (ZSTD_compressedLiterals(optPtr)) { - U32 u; - for (u=0; u < litLength; u++) - optPtr->litFreq[literals[u]] += ZSTD_LITFREQ_ADD; - optPtr->litSum += litLength*ZSTD_LITFREQ_ADD; - } - - /* literal Length */ - { U32 const llCode = ZSTD_LLcode(litLength); - optPtr->litLengthFreq[llCode]++; - optPtr->litLengthSum++; - } - - /* offset code : expected to follow storeSeq() numeric representation */ - { U32 const offCode = ZSTD_highbit32(STORED_TO_OFFBASE(offsetCode)); - assert(offCode <= MaxOff); - optPtr->offCodeFreq[offCode]++; - optPtr->offCodeSum++; - } - - /* match Length */ - { U32 const mlBase = matchLength - MINMATCH; - U32 const mlCode = ZSTD_MLcode(mlBase); - optPtr->matchLengthFreq[mlCode]++; - optPtr->matchLengthSum++; - } -} - - -/* ZSTD_readMINMATCH() : - * function safe only for comparisons - * assumption : memPtr must be at least 4 bytes before end of buffer */ -MEM_STATIC U32 ZSTD_readMINMATCH(const void* memPtr, U32 length) -{ - switch (length) - { - default : - case 4 : return MEM_read32(memPtr); - case 3 : if (MEM_isLittleEndian()) - return MEM_read32(memPtr)<<8; - else - return MEM_read32(memPtr)>>8; - } -} - - -/* Update hashTable3 up to ip (excluded) - Assumption : always within prefix (i.e. not within extDict) */ -static U32 ZSTD_insertAndFindFirstIndexHash3 (const ZSTD_matchState_t* ms, - U32* nextToUpdate3, - const BYTE* const ip) -{ - U32* const hashTable3 = ms->hashTable3; - U32 const hashLog3 = ms->hashLog3; - const BYTE* const base = ms->window.base; - U32 idx = *nextToUpdate3; - U32 const target = (U32)(ip - base); - size_t const hash3 = ZSTD_hash3Ptr(ip, hashLog3); - assert(hashLog3 > 0); - - while(idx < target) { - hashTable3[ZSTD_hash3Ptr(base+idx, hashLog3)] = idx; - idx++; - } - - *nextToUpdate3 = target; - return hashTable3[hash3]; -} - - -/*-************************************* -* Binary Tree search -***************************************/ -/** ZSTD_insertBt1() : add one or multiple positions to tree. - * @param ip assumed <= iend-8 . - * @param target The target of ZSTD_updateTree_internal() - we are filling to this position - * @return : nb of positions added */ -static U32 ZSTD_insertBt1( - const ZSTD_matchState_t* ms, - const BYTE* const ip, const BYTE* const iend, - U32 const target, - U32 const mls, const int extDict) -{ - const ZSTD_compressionParameters* const cParams = &ms->cParams; - U32* const hashTable = ms->hashTable; - U32 const hashLog = cParams->hashLog; - size_t const h = ZSTD_hashPtr(ip, hashLog, mls); - U32* const bt = ms->chainTable; - U32 const btLog = cParams->chainLog - 1; - U32 const btMask = (1 << btLog) - 1; - U32 matchIndex = hashTable[h]; - size_t commonLengthSmaller=0, commonLengthLarger=0; - const BYTE* const base = ms->window.base; - const BYTE* const dictBase = ms->window.dictBase; - const U32 dictLimit = ms->window.dictLimit; - const BYTE* const dictEnd = dictBase + dictLimit; - const BYTE* const prefixStart = base + dictLimit; - const BYTE* match; - const U32 curr = (U32)(ip-base); - const U32 btLow = btMask >= curr ? 0 : curr - btMask; - U32* smallerPtr = bt + 2*(curr&btMask); - U32* largerPtr = smallerPtr + 1; - U32 dummy32; /* to be nullified at the end */ - /* windowLow is based on target because - * we only need positions that will be in the window at the end of the tree update. - */ - U32 const windowLow = ZSTD_getLowestMatchIndex(ms, target, cParams->windowLog); - U32 matchEndIdx = curr+8+1; - size_t bestLength = 8; - U32 nbCompares = 1U << cParams->searchLog; -#ifdef ZSTD_C_PREDICT - U32 predictedSmall = *(bt + 2*((curr-1)&btMask) + 0); - U32 predictedLarge = *(bt + 2*((curr-1)&btMask) + 1); - predictedSmall += (predictedSmall>0); - predictedLarge += (predictedLarge>0); -#endif /* ZSTD_C_PREDICT */ - - DEBUGLOG(8, "ZSTD_insertBt1 (%u)", curr); - - assert(curr <= target); - assert(ip <= iend-8); /* required for h calculation */ - hashTable[h] = curr; /* Update Hash Table */ - - assert(windowLow > 0); - for (; nbCompares && (matchIndex >= windowLow); --nbCompares) { - U32* const nextPtr = bt + 2*(matchIndex & btMask); - size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */ - assert(matchIndex < curr); - -#ifdef ZSTD_C_PREDICT /* note : can create issues when hlog small <= 11 */ - const U32* predictPtr = bt + 2*((matchIndex-1) & btMask); /* written this way, as bt is a roll buffer */ - if (matchIndex == predictedSmall) { - /* no need to check length, result known */ - *smallerPtr = matchIndex; - if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop the search */ - smallerPtr = nextPtr+1; /* new "smaller" => larger of match */ - matchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */ - predictedSmall = predictPtr[1] + (predictPtr[1]>0); - continue; - } - if (matchIndex == predictedLarge) { - *largerPtr = matchIndex; - if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop the search */ - largerPtr = nextPtr; - matchIndex = nextPtr[0]; - predictedLarge = predictPtr[0] + (predictPtr[0]>0); - continue; - } -#endif - - if (!extDict || (matchIndex+matchLength >= dictLimit)) { - assert(matchIndex+matchLength >= dictLimit); /* might be wrong if actually extDict */ - match = base + matchIndex; - matchLength += ZSTD_count(ip+matchLength, match+matchLength, iend); - } else { - match = dictBase + matchIndex; - matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iend, dictEnd, prefixStart); - if (matchIndex+matchLength >= dictLimit) - match = base + matchIndex; /* to prepare for next usage of match[matchLength] */ - } - - if (matchLength > bestLength) { - bestLength = matchLength; - if (matchLength > matchEndIdx - matchIndex) - matchEndIdx = matchIndex + (U32)matchLength; - } - - if (ip+matchLength == iend) { /* equal : no way to know if inf or sup */ - break; /* drop , to guarantee consistency ; miss a bit of compression, but other solutions can corrupt tree */ - } - - if (match[matchLength] < ip[matchLength]) { /* necessarily within buffer */ - /* match is smaller than current */ - *smallerPtr = matchIndex; /* update smaller idx */ - commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */ - if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop searching */ - smallerPtr = nextPtr+1; /* new "candidate" => larger than match, which was smaller than target */ - matchIndex = nextPtr[1]; /* new matchIndex, larger than previous and closer to current */ - } else { - /* match is larger than current */ - *largerPtr = matchIndex; - commonLengthLarger = matchLength; - if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop searching */ - largerPtr = nextPtr; - matchIndex = nextPtr[0]; - } } - - *smallerPtr = *largerPtr = 0; - { U32 positions = 0; - if (bestLength > 384) positions = MIN(192, (U32)(bestLength - 384)); /* speed optimization */ - assert(matchEndIdx > curr + 8); - return MAX(positions, matchEndIdx - (curr + 8)); - } -} - -FORCE_INLINE_TEMPLATE -void ZSTD_updateTree_internal( - ZSTD_matchState_t* ms, - const BYTE* const ip, const BYTE* const iend, - const U32 mls, const ZSTD_dictMode_e dictMode) -{ - const BYTE* const base = ms->window.base; - U32 const target = (U32)(ip - base); - U32 idx = ms->nextToUpdate; - DEBUGLOG(6, "ZSTD_updateTree_internal, from %u to %u (dictMode:%u)", - idx, target, dictMode); - - while(idx < target) { - U32 const forward = ZSTD_insertBt1(ms, base+idx, iend, target, mls, dictMode == ZSTD_extDict); - assert(idx < (U32)(idx + forward)); - idx += forward; - } - assert((size_t)(ip - base) <= (size_t)(U32)(-1)); - assert((size_t)(iend - base) <= (size_t)(U32)(-1)); - ms->nextToUpdate = target; -} - -void ZSTD_updateTree(ZSTD_matchState_t* ms, const BYTE* ip, const BYTE* iend) { - ZSTD_updateTree_internal(ms, ip, iend, ms->cParams.minMatch, ZSTD_noDict); -} - -FORCE_INLINE_TEMPLATE -U32 ZSTD_insertBtAndGetAllMatches ( - ZSTD_match_t* matches, /* store result (found matches) in this table (presumed large enough) */ - ZSTD_matchState_t* ms, - U32* nextToUpdate3, - const BYTE* const ip, const BYTE* const iLimit, const ZSTD_dictMode_e dictMode, - const U32 rep[ZSTD_REP_NUM], - U32 const ll0, /* tells if associated literal length is 0 or not. This value must be 0 or 1 */ - const U32 lengthToBeat, - U32 const mls /* template */) -{ - const ZSTD_compressionParameters* const cParams = &ms->cParams; - U32 const sufficient_len = MIN(cParams->targetLength, ZSTD_OPT_NUM -1); - const BYTE* const base = ms->window.base; - U32 const curr = (U32)(ip-base); - U32 const hashLog = cParams->hashLog; - U32 const minMatch = (mls==3) ? 3 : 4; - U32* const hashTable = ms->hashTable; - size_t const h = ZSTD_hashPtr(ip, hashLog, mls); - U32 matchIndex = hashTable[h]; - U32* const bt = ms->chainTable; - U32 const btLog = cParams->chainLog - 1; - U32 const btMask= (1U << btLog) - 1; - size_t commonLengthSmaller=0, commonLengthLarger=0; - const BYTE* const dictBase = ms->window.dictBase; - U32 const dictLimit = ms->window.dictLimit; - const BYTE* const dictEnd = dictBase + dictLimit; - const BYTE* const prefixStart = base + dictLimit; - U32 const btLow = (btMask >= curr) ? 0 : curr - btMask; - U32 const windowLow = ZSTD_getLowestMatchIndex(ms, curr, cParams->windowLog); - U32 const matchLow = windowLow ? windowLow : 1; - U32* smallerPtr = bt + 2*(curr&btMask); - U32* largerPtr = bt + 2*(curr&btMask) + 1; - U32 matchEndIdx = curr+8+1; /* farthest referenced position of any match => detects repetitive patterns */ - U32 dummy32; /* to be nullified at the end */ - U32 mnum = 0; - U32 nbCompares = 1U << cParams->searchLog; - - const ZSTD_matchState_t* dms = dictMode == ZSTD_dictMatchState ? ms->dictMatchState : NULL; - const ZSTD_compressionParameters* const dmsCParams = - dictMode == ZSTD_dictMatchState ? &dms->cParams : NULL; - const BYTE* const dmsBase = dictMode == ZSTD_dictMatchState ? dms->window.base : NULL; - const BYTE* const dmsEnd = dictMode == ZSTD_dictMatchState ? dms->window.nextSrc : NULL; - U32 const dmsHighLimit = dictMode == ZSTD_dictMatchState ? (U32)(dmsEnd - dmsBase) : 0; - U32 const dmsLowLimit = dictMode == ZSTD_dictMatchState ? dms->window.lowLimit : 0; - U32 const dmsIndexDelta = dictMode == ZSTD_dictMatchState ? windowLow - dmsHighLimit : 0; - U32 const dmsHashLog = dictMode == ZSTD_dictMatchState ? dmsCParams->hashLog : hashLog; - U32 const dmsBtLog = dictMode == ZSTD_dictMatchState ? dmsCParams->chainLog - 1 : btLog; - U32 const dmsBtMask = dictMode == ZSTD_dictMatchState ? (1U << dmsBtLog) - 1 : 0; - U32 const dmsBtLow = dictMode == ZSTD_dictMatchState && dmsBtMask < dmsHighLimit - dmsLowLimit ? dmsHighLimit - dmsBtMask : dmsLowLimit; - - size_t bestLength = lengthToBeat-1; - DEBUGLOG(8, "ZSTD_insertBtAndGetAllMatches: current=%u", curr); - - /* check repCode */ - assert(ll0 <= 1); /* necessarily 1 or 0 */ - { U32 const lastR = ZSTD_REP_NUM + ll0; - U32 repCode; - for (repCode = ll0; repCode < lastR; repCode++) { - U32 const repOffset = (repCode==ZSTD_REP_NUM) ? (rep[0] - 1) : rep[repCode]; - U32 const repIndex = curr - repOffset; - U32 repLen = 0; - assert(curr >= dictLimit); - if (repOffset-1 /* intentional overflow, discards 0 and -1 */ < curr-dictLimit) { /* equivalent to `curr > repIndex >= dictLimit` */ - /* We must validate the repcode offset because when we're using a dictionary the - * valid offset range shrinks when the dictionary goes out of bounds. - */ - if ((repIndex >= windowLow) & (ZSTD_readMINMATCH(ip, minMatch) == ZSTD_readMINMATCH(ip - repOffset, minMatch))) { - repLen = (U32)ZSTD_count(ip+minMatch, ip+minMatch-repOffset, iLimit) + minMatch; - } - } else { /* repIndex < dictLimit || repIndex >= curr */ - const BYTE* const repMatch = dictMode == ZSTD_dictMatchState ? - dmsBase + repIndex - dmsIndexDelta : - dictBase + repIndex; - assert(curr >= windowLow); - if ( dictMode == ZSTD_extDict - && ( ((repOffset-1) /*intentional overflow*/ < curr - windowLow) /* equivalent to `curr > repIndex >= windowLow` */ - & (((U32)((dictLimit-1) - repIndex) >= 3) ) /* intentional overflow : do not test positions overlapping 2 memory segments */) - && (ZSTD_readMINMATCH(ip, minMatch) == ZSTD_readMINMATCH(repMatch, minMatch)) ) { - repLen = (U32)ZSTD_count_2segments(ip+minMatch, repMatch+minMatch, iLimit, dictEnd, prefixStart) + minMatch; - } - if (dictMode == ZSTD_dictMatchState - && ( ((repOffset-1) /*intentional overflow*/ < curr - (dmsLowLimit + dmsIndexDelta)) /* equivalent to `curr > repIndex >= dmsLowLimit` */ - & ((U32)((dictLimit-1) - repIndex) >= 3) ) /* intentional overflow : do not test positions overlapping 2 memory segments */ - && (ZSTD_readMINMATCH(ip, minMatch) == ZSTD_readMINMATCH(repMatch, minMatch)) ) { - repLen = (U32)ZSTD_count_2segments(ip+minMatch, repMatch+minMatch, iLimit, dmsEnd, prefixStart) + minMatch; - } } - /* save longer solution */ - if (repLen > bestLength) { - DEBUGLOG(8, "found repCode %u (ll0:%u, offset:%u) of length %u", - repCode, ll0, repOffset, repLen); - bestLength = repLen; - matches[mnum].off = STORE_REPCODE(repCode - ll0 + 1); /* expect value between 1 and 3 */ - matches[mnum].len = (U32)repLen; - mnum++; - if ( (repLen > sufficient_len) - | (ip+repLen == iLimit) ) { /* best possible */ - return mnum; - } } } } - - /* HC3 match finder */ - if ((mls == 3) /*static*/ && (bestLength < mls)) { - U32 const matchIndex3 = ZSTD_insertAndFindFirstIndexHash3(ms, nextToUpdate3, ip); - if ((matchIndex3 >= matchLow) - & (curr - matchIndex3 < (1<<18)) /*heuristic : longer distance likely too expensive*/ ) { - size_t mlen; - if ((dictMode == ZSTD_noDict) /*static*/ || (dictMode == ZSTD_dictMatchState) /*static*/ || (matchIndex3 >= dictLimit)) { - const BYTE* const match = base + matchIndex3; - mlen = ZSTD_count(ip, match, iLimit); - } else { - const BYTE* const match = dictBase + matchIndex3; - mlen = ZSTD_count_2segments(ip, match, iLimit, dictEnd, prefixStart); - } - - /* save best solution */ - if (mlen >= mls /* == 3 > bestLength */) { - DEBUGLOG(8, "found small match with hlog3, of length %u", - (U32)mlen); - bestLength = mlen; - assert(curr > matchIndex3); - assert(mnum==0); /* no prior solution */ - matches[0].off = STORE_OFFSET(curr - matchIndex3); - matches[0].len = (U32)mlen; - mnum = 1; - if ( (mlen > sufficient_len) | - (ip+mlen == iLimit) ) { /* best possible length */ - ms->nextToUpdate = curr+1; /* skip insertion */ - return 1; - } } } - /* no dictMatchState lookup: dicts don't have a populated HC3 table */ - } /* if (mls == 3) */ - - hashTable[h] = curr; /* Update Hash Table */ - - for (; nbCompares && (matchIndex >= matchLow); --nbCompares) { - U32* const nextPtr = bt + 2*(matchIndex & btMask); - const BYTE* match; - size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */ - assert(curr > matchIndex); - - if ((dictMode == ZSTD_noDict) || (dictMode == ZSTD_dictMatchState) || (matchIndex+matchLength >= dictLimit)) { - assert(matchIndex+matchLength >= dictLimit); /* ensure the condition is correct when !extDict */ - match = base + matchIndex; - if (matchIndex >= dictLimit) assert(memcmp(match, ip, matchLength) == 0); /* ensure early section of match is equal as expected */ - matchLength += ZSTD_count(ip+matchLength, match+matchLength, iLimit); - } else { - match = dictBase + matchIndex; - assert(memcmp(match, ip, matchLength) == 0); /* ensure early section of match is equal as expected */ - matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iLimit, dictEnd, prefixStart); - if (matchIndex+matchLength >= dictLimit) - match = base + matchIndex; /* prepare for match[matchLength] read */ - } - - if (matchLength > bestLength) { - DEBUGLOG(8, "found match of length %u at distance %u (offCode=%u)", - (U32)matchLength, curr - matchIndex, STORE_OFFSET(curr - matchIndex)); - assert(matchEndIdx > matchIndex); - if (matchLength > matchEndIdx - matchIndex) - matchEndIdx = matchIndex + (U32)matchLength; - bestLength = matchLength; - matches[mnum].off = STORE_OFFSET(curr - matchIndex); - matches[mnum].len = (U32)matchLength; - mnum++; - if ( (matchLength > ZSTD_OPT_NUM) - | (ip+matchLength == iLimit) /* equal : no way to know if inf or sup */) { - if (dictMode == ZSTD_dictMatchState) nbCompares = 0; /* break should also skip searching dms */ - break; /* drop, to preserve bt consistency (miss a little bit of compression) */ - } } - - if (match[matchLength] < ip[matchLength]) { - /* match smaller than current */ - *smallerPtr = matchIndex; /* update smaller idx */ - commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */ - if (matchIndex <= btLow) { smallerPtr=&dummy32; break; } /* beyond tree size, stop the search */ - smallerPtr = nextPtr+1; /* new candidate => larger than match, which was smaller than current */ - matchIndex = nextPtr[1]; /* new matchIndex, larger than previous, closer to current */ - } else { - *largerPtr = matchIndex; - commonLengthLarger = matchLength; - if (matchIndex <= btLow) { largerPtr=&dummy32; break; } /* beyond tree size, stop the search */ - largerPtr = nextPtr; - matchIndex = nextPtr[0]; - } } - - *smallerPtr = *largerPtr = 0; - - assert(nbCompares <= (1U << ZSTD_SEARCHLOG_MAX)); /* Check we haven't underflowed. */ - if (dictMode == ZSTD_dictMatchState && nbCompares) { - size_t const dmsH = ZSTD_hashPtr(ip, dmsHashLog, mls); - U32 dictMatchIndex = dms->hashTable[dmsH]; - const U32* const dmsBt = dms->chainTable; - commonLengthSmaller = commonLengthLarger = 0; - for (; nbCompares && (dictMatchIndex > dmsLowLimit); --nbCompares) { - const U32* const nextPtr = dmsBt + 2*(dictMatchIndex & dmsBtMask); - size_t matchLength = MIN(commonLengthSmaller, commonLengthLarger); /* guaranteed minimum nb of common bytes */ - const BYTE* match = dmsBase + dictMatchIndex; - matchLength += ZSTD_count_2segments(ip+matchLength, match+matchLength, iLimit, dmsEnd, prefixStart); - if (dictMatchIndex+matchLength >= dmsHighLimit) - match = base + dictMatchIndex + dmsIndexDelta; /* to prepare for next usage of match[matchLength] */ - - if (matchLength > bestLength) { - matchIndex = dictMatchIndex + dmsIndexDelta; - DEBUGLOG(8, "found dms match of length %u at distance %u (offCode=%u)", - (U32)matchLength, curr - matchIndex, STORE_OFFSET(curr - matchIndex)); - if (matchLength > matchEndIdx - matchIndex) - matchEndIdx = matchIndex + (U32)matchLength; - bestLength = matchLength; - matches[mnum].off = STORE_OFFSET(curr - matchIndex); - matches[mnum].len = (U32)matchLength; - mnum++; - if ( (matchLength > ZSTD_OPT_NUM) - | (ip+matchLength == iLimit) /* equal : no way to know if inf or sup */) { - break; /* drop, to guarantee consistency (miss a little bit of compression) */ - } } - - if (dictMatchIndex <= dmsBtLow) { break; } /* beyond tree size, stop the search */ - if (match[matchLength] < ip[matchLength]) { - commonLengthSmaller = matchLength; /* all smaller will now have at least this guaranteed common length */ - dictMatchIndex = nextPtr[1]; /* new matchIndex larger than previous (closer to current) */ - } else { - /* match is larger than current */ - commonLengthLarger = matchLength; - dictMatchIndex = nextPtr[0]; - } } } /* if (dictMode == ZSTD_dictMatchState) */ - - assert(matchEndIdx > curr+8); - ms->nextToUpdate = matchEndIdx - 8; /* skip repetitive patterns */ - return mnum; -} - -typedef U32 (*ZSTD_getAllMatchesFn)( - ZSTD_match_t*, - ZSTD_matchState_t*, - U32*, - const BYTE*, - const BYTE*, - const U32 rep[ZSTD_REP_NUM], - U32 const ll0, - U32 const lengthToBeat); - -FORCE_INLINE_TEMPLATE U32 ZSTD_btGetAllMatches_internal( - ZSTD_match_t* matches, - ZSTD_matchState_t* ms, - U32* nextToUpdate3, - const BYTE* ip, - const BYTE* const iHighLimit, - const U32 rep[ZSTD_REP_NUM], - U32 const ll0, - U32 const lengthToBeat, - const ZSTD_dictMode_e dictMode, - const U32 mls) -{ - assert(BOUNDED(3, ms->cParams.minMatch, 6) == mls); - DEBUGLOG(8, "ZSTD_BtGetAllMatches(dictMode=%d, mls=%u)", (int)dictMode, mls); - if (ip < ms->window.base + ms->nextToUpdate) - return 0; /* skipped area */ - ZSTD_updateTree_internal(ms, ip, iHighLimit, mls, dictMode); - return ZSTD_insertBtAndGetAllMatches(matches, ms, nextToUpdate3, ip, iHighLimit, dictMode, rep, ll0, lengthToBeat, mls); -} - -#define ZSTD_BT_GET_ALL_MATCHES_FN(dictMode, mls) ZSTD_btGetAllMatches_##dictMode##_##mls - -#define GEN_ZSTD_BT_GET_ALL_MATCHES_(dictMode, mls) \ - static U32 ZSTD_BT_GET_ALL_MATCHES_FN(dictMode, mls)( \ - ZSTD_match_t* matches, \ - ZSTD_matchState_t* ms, \ - U32* nextToUpdate3, \ - const BYTE* ip, \ - const BYTE* const iHighLimit, \ - const U32 rep[ZSTD_REP_NUM], \ - U32 const ll0, \ - U32 const lengthToBeat) \ - { \ - return ZSTD_btGetAllMatches_internal( \ - matches, ms, nextToUpdate3, ip, iHighLimit, \ - rep, ll0, lengthToBeat, ZSTD_##dictMode, mls); \ - } - -#define GEN_ZSTD_BT_GET_ALL_MATCHES(dictMode) \ - GEN_ZSTD_BT_GET_ALL_MATCHES_(dictMode, 3) \ - GEN_ZSTD_BT_GET_ALL_MATCHES_(dictMode, 4) \ - GEN_ZSTD_BT_GET_ALL_MATCHES_(dictMode, 5) \ - GEN_ZSTD_BT_GET_ALL_MATCHES_(dictMode, 6) - -GEN_ZSTD_BT_GET_ALL_MATCHES(noDict) -GEN_ZSTD_BT_GET_ALL_MATCHES(extDict) -GEN_ZSTD_BT_GET_ALL_MATCHES(dictMatchState) - -#define ZSTD_BT_GET_ALL_MATCHES_ARRAY(dictMode) \ - { \ - ZSTD_BT_GET_ALL_MATCHES_FN(dictMode, 3), \ - ZSTD_BT_GET_ALL_MATCHES_FN(dictMode, 4), \ - ZSTD_BT_GET_ALL_MATCHES_FN(dictMode, 5), \ - ZSTD_BT_GET_ALL_MATCHES_FN(dictMode, 6) \ - } - -static ZSTD_getAllMatchesFn -ZSTD_selectBtGetAllMatches(ZSTD_matchState_t const* ms, ZSTD_dictMode_e const dictMode) -{ - ZSTD_getAllMatchesFn const getAllMatchesFns[3][4] = { - ZSTD_BT_GET_ALL_MATCHES_ARRAY(noDict), - ZSTD_BT_GET_ALL_MATCHES_ARRAY(extDict), - ZSTD_BT_GET_ALL_MATCHES_ARRAY(dictMatchState) - }; - U32 const mls = BOUNDED(3, ms->cParams.minMatch, 6); - assert((U32)dictMode < 3); - assert(mls - 3 < 4); - return getAllMatchesFns[(int)dictMode][mls - 3]; -} - -/************************* -* LDM helper functions * -*************************/ - -/* Struct containing info needed to make decision about ldm inclusion */ -typedef struct { - rawSeqStore_t seqStore; /* External match candidates store for this block */ - U32 startPosInBlock; /* Start position of the current match candidate */ - U32 endPosInBlock; /* End position of the current match candidate */ - U32 offset; /* Offset of the match candidate */ -} ZSTD_optLdm_t; - -/* ZSTD_optLdm_skipRawSeqStoreBytes(): - * Moves forward in @rawSeqStore by @nbBytes, - * which will update the fields 'pos' and 'posInSequence'. - */ -static void ZSTD_optLdm_skipRawSeqStoreBytes(rawSeqStore_t* rawSeqStore, size_t nbBytes) -{ - U32 currPos = (U32)(rawSeqStore->posInSequence + nbBytes); - while (currPos && rawSeqStore->pos < rawSeqStore->size) { - rawSeq currSeq = rawSeqStore->seq[rawSeqStore->pos]; - if (currPos >= currSeq.litLength + currSeq.matchLength) { - currPos -= currSeq.litLength + currSeq.matchLength; - rawSeqStore->pos++; - } else { - rawSeqStore->posInSequence = currPos; - break; - } - } - if (currPos == 0 || rawSeqStore->pos == rawSeqStore->size) { - rawSeqStore->posInSequence = 0; - } -} - -/* ZSTD_opt_getNextMatchAndUpdateSeqStore(): - * Calculates the beginning and end of the next match in the current block. - * Updates 'pos' and 'posInSequence' of the ldmSeqStore. - */ -static void -ZSTD_opt_getNextMatchAndUpdateSeqStore(ZSTD_optLdm_t* optLdm, U32 currPosInBlock, - U32 blockBytesRemaining) -{ - rawSeq currSeq; - U32 currBlockEndPos; - U32 literalsBytesRemaining; - U32 matchBytesRemaining; - - /* Setting match end position to MAX to ensure we never use an LDM during this block */ - if (optLdm->seqStore.size == 0 || optLdm->seqStore.pos >= optLdm->seqStore.size) { - optLdm->startPosInBlock = UINT_MAX; - optLdm->endPosInBlock = UINT_MAX; - return; - } - /* Calculate appropriate bytes left in matchLength and litLength - * after adjusting based on ldmSeqStore->posInSequence */ - currSeq = optLdm->seqStore.seq[optLdm->seqStore.pos]; - assert(optLdm->seqStore.posInSequence <= currSeq.litLength + currSeq.matchLength); - currBlockEndPos = currPosInBlock + blockBytesRemaining; - literalsBytesRemaining = (optLdm->seqStore.posInSequence < currSeq.litLength) ? - currSeq.litLength - (U32)optLdm->seqStore.posInSequence : - 0; - matchBytesRemaining = (literalsBytesRemaining == 0) ? - currSeq.matchLength - ((U32)optLdm->seqStore.posInSequence - currSeq.litLength) : - currSeq.matchLength; - - /* If there are more literal bytes than bytes remaining in block, no ldm is possible */ - if (literalsBytesRemaining >= blockBytesRemaining) { - optLdm->startPosInBlock = UINT_MAX; - optLdm->endPosInBlock = UINT_MAX; - ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, blockBytesRemaining); - return; - } - - /* Matches may be < MINMATCH by this process. In that case, we will reject them - when we are deciding whether or not to add the ldm */ - optLdm->startPosInBlock = currPosInBlock + literalsBytesRemaining; - optLdm->endPosInBlock = optLdm->startPosInBlock + matchBytesRemaining; - optLdm->offset = currSeq.offset; - - if (optLdm->endPosInBlock > currBlockEndPos) { - /* Match ends after the block ends, we can't use the whole match */ - optLdm->endPosInBlock = currBlockEndPos; - ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, currBlockEndPos - currPosInBlock); - } else { - /* Consume nb of bytes equal to size of sequence left */ - ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, literalsBytesRemaining + matchBytesRemaining); - } -} - -/* ZSTD_optLdm_maybeAddMatch(): - * Adds a match if it's long enough, - * based on it's 'matchStartPosInBlock' and 'matchEndPosInBlock', - * into 'matches'. Maintains the correct ordering of 'matches'. - */ -static void ZSTD_optLdm_maybeAddMatch(ZSTD_match_t* matches, U32* nbMatches, - const ZSTD_optLdm_t* optLdm, U32 currPosInBlock) -{ - U32 const posDiff = currPosInBlock - optLdm->startPosInBlock; - /* Note: ZSTD_match_t actually contains offCode and matchLength (before subtracting MINMATCH) */ - U32 const candidateMatchLength = optLdm->endPosInBlock - optLdm->startPosInBlock - posDiff; - - /* Ensure that current block position is not outside of the match */ - if (currPosInBlock < optLdm->startPosInBlock - || currPosInBlock >= optLdm->endPosInBlock - || candidateMatchLength < MINMATCH) { - return; - } - - if (*nbMatches == 0 || ((candidateMatchLength > matches[*nbMatches-1].len) && *nbMatches < ZSTD_OPT_NUM)) { - U32 const candidateOffCode = STORE_OFFSET(optLdm->offset); - DEBUGLOG(6, "ZSTD_optLdm_maybeAddMatch(): Adding ldm candidate match (offCode: %u matchLength %u) at block position=%u", - candidateOffCode, candidateMatchLength, currPosInBlock); - matches[*nbMatches].len = candidateMatchLength; - matches[*nbMatches].off = candidateOffCode; - (*nbMatches)++; - } -} - -/* ZSTD_optLdm_processMatchCandidate(): - * Wrapper function to update ldm seq store and call ldm functions as necessary. - */ -static void -ZSTD_optLdm_processMatchCandidate(ZSTD_optLdm_t* optLdm, - ZSTD_match_t* matches, U32* nbMatches, - U32 currPosInBlock, U32 remainingBytes) -{ - if (optLdm->seqStore.size == 0 || optLdm->seqStore.pos >= optLdm->seqStore.size) { - return; - } - - if (currPosInBlock >= optLdm->endPosInBlock) { - if (currPosInBlock > optLdm->endPosInBlock) { - /* The position at which ZSTD_optLdm_processMatchCandidate() is called is not necessarily - * at the end of a match from the ldm seq store, and will often be some bytes - * over beyond matchEndPosInBlock. As such, we need to correct for these "overshoots" - */ - U32 const posOvershoot = currPosInBlock - optLdm->endPosInBlock; - ZSTD_optLdm_skipRawSeqStoreBytes(&optLdm->seqStore, posOvershoot); - } - ZSTD_opt_getNextMatchAndUpdateSeqStore(optLdm, currPosInBlock, remainingBytes); - } - ZSTD_optLdm_maybeAddMatch(matches, nbMatches, optLdm, currPosInBlock); -} - - -/*-******************************* -* Optimal parser -*********************************/ - -static U32 ZSTD_totalLen(ZSTD_optimal_t sol) -{ - return sol.litlen + sol.mlen; -} - -#if 0 /* debug */ - -static void -listStats(const U32* table, int lastEltID) -{ - int const nbElts = lastEltID + 1; - int enb; - for (enb=0; enb < nbElts; enb++) { - (void)table; - /* RAWLOG(2, "%3i:%3i, ", enb, table[enb]); */ - RAWLOG(2, "%4i,", table[enb]); - } - RAWLOG(2, " \n"); -} - -#endif - -FORCE_INLINE_TEMPLATE size_t -ZSTD_compressBlock_opt_generic(ZSTD_matchState_t* ms, - seqStore_t* seqStore, - U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize, - const int optLevel, - const ZSTD_dictMode_e dictMode) -{ - optState_t* const optStatePtr = &ms->opt; - const BYTE* const istart = (const BYTE*)src; - const BYTE* ip = istart; - const BYTE* anchor = istart; - const BYTE* const iend = istart + srcSize; - const BYTE* const ilimit = iend - 8; - const BYTE* const base = ms->window.base; - const BYTE* const prefixStart = base + ms->window.dictLimit; - const ZSTD_compressionParameters* const cParams = &ms->cParams; - - ZSTD_getAllMatchesFn getAllMatches = ZSTD_selectBtGetAllMatches(ms, dictMode); - - U32 const sufficient_len = MIN(cParams->targetLength, ZSTD_OPT_NUM -1); - U32 const minMatch = (cParams->minMatch == 3) ? 3 : 4; - U32 nextToUpdate3 = ms->nextToUpdate; - - ZSTD_optimal_t* const opt = optStatePtr->priceTable; - ZSTD_match_t* const matches = optStatePtr->matchTable; - ZSTD_optimal_t lastSequence; - ZSTD_optLdm_t optLdm; - - optLdm.seqStore = ms->ldmSeqStore ? *ms->ldmSeqStore : kNullRawSeqStore; - optLdm.endPosInBlock = optLdm.startPosInBlock = optLdm.offset = 0; - ZSTD_opt_getNextMatchAndUpdateSeqStore(&optLdm, (U32)(ip-istart), (U32)(iend-ip)); - - /* init */ - DEBUGLOG(5, "ZSTD_compressBlock_opt_generic: current=%u, prefix=%u, nextToUpdate=%u", - (U32)(ip - base), ms->window.dictLimit, ms->nextToUpdate); - assert(optLevel <= 2); - ZSTD_rescaleFreqs(optStatePtr, (const BYTE*)src, srcSize, optLevel); - ip += (ip==prefixStart); - - /* Match Loop */ - while (ip < ilimit) { - U32 cur, last_pos = 0; - - /* find first match */ - { U32 const litlen = (U32)(ip - anchor); - U32 const ll0 = !litlen; - U32 nbMatches = getAllMatches(matches, ms, &nextToUpdate3, ip, iend, rep, ll0, minMatch); - ZSTD_optLdm_processMatchCandidate(&optLdm, matches, &nbMatches, - (U32)(ip-istart), (U32)(iend - ip)); - if (!nbMatches) { ip++; continue; } - - /* initialize opt[0] */ - { U32 i ; for (i=0; i immediate encoding */ - { U32 const maxML = matches[nbMatches-1].len; - U32 const maxOffcode = matches[nbMatches-1].off; - DEBUGLOG(6, "found %u matches of maxLength=%u and maxOffCode=%u at cPos=%u => start new series", - nbMatches, maxML, maxOffcode, (U32)(ip-prefixStart)); - - if (maxML > sufficient_len) { - lastSequence.litlen = litlen; - lastSequence.mlen = maxML; - lastSequence.off = maxOffcode; - DEBUGLOG(6, "large match (%u>%u), immediate encoding", - maxML, sufficient_len); - cur = 0; - last_pos = ZSTD_totalLen(lastSequence); - goto _shortestPath; - } } - - /* set prices for first matches starting position == 0 */ - assert(opt[0].price >= 0); - { U32 const literalsPrice = (U32)opt[0].price + ZSTD_litLengthPrice(0, optStatePtr, optLevel); - U32 pos; - U32 matchNb; - for (pos = 1; pos < minMatch; pos++) { - opt[pos].price = ZSTD_MAX_PRICE; /* mlen, litlen and price will be fixed during forward scanning */ - } - for (matchNb = 0; matchNb < nbMatches; matchNb++) { - U32 const offcode = matches[matchNb].off; - U32 const end = matches[matchNb].len; - for ( ; pos <= end ; pos++ ) { - U32 const matchPrice = ZSTD_getMatchPrice(offcode, pos, optStatePtr, optLevel); - U32 const sequencePrice = literalsPrice + matchPrice; - DEBUGLOG(7, "rPos:%u => set initial price : %.2f", - pos, ZSTD_fCost(sequencePrice)); - opt[pos].mlen = pos; - opt[pos].off = offcode; - opt[pos].litlen = litlen; - opt[pos].price = (int)sequencePrice; - } } - last_pos = pos-1; - } - } - - /* check further positions */ - for (cur = 1; cur <= last_pos; cur++) { - const BYTE* const inr = ip + cur; - assert(cur < ZSTD_OPT_NUM); - DEBUGLOG(7, "cPos:%zi==rPos:%u", inr-istart, cur) - - /* Fix current position with one literal if cheaper */ - { U32 const litlen = (opt[cur-1].mlen == 0) ? opt[cur-1].litlen + 1 : 1; - int const price = opt[cur-1].price - + (int)ZSTD_rawLiteralsCost(ip+cur-1, 1, optStatePtr, optLevel) - + (int)ZSTD_litLengthPrice(litlen, optStatePtr, optLevel) - - (int)ZSTD_litLengthPrice(litlen-1, optStatePtr, optLevel); - assert(price < 1000000000); /* overflow check */ - if (price <= opt[cur].price) { - DEBUGLOG(7, "cPos:%zi==rPos:%u : better price (%.2f<=%.2f) using literal (ll==%u) (hist:%u,%u,%u)", - inr-istart, cur, ZSTD_fCost(price), ZSTD_fCost(opt[cur].price), litlen, - opt[cur-1].rep[0], opt[cur-1].rep[1], opt[cur-1].rep[2]); - opt[cur].mlen = 0; - opt[cur].off = 0; - opt[cur].litlen = litlen; - opt[cur].price = price; - } else { - DEBUGLOG(7, "cPos:%zi==rPos:%u : literal would cost more (%.2f>%.2f) (hist:%u,%u,%u)", - inr-istart, cur, ZSTD_fCost(price), ZSTD_fCost(opt[cur].price), - opt[cur].rep[0], opt[cur].rep[1], opt[cur].rep[2]); - } - } - - /* Set the repcodes of the current position. We must do it here - * because we rely on the repcodes of the 2nd to last sequence being - * correct to set the next chunks repcodes during the backward - * traversal. - */ - ZSTD_STATIC_ASSERT(sizeof(opt[cur].rep) == sizeof(repcodes_t)); - assert(cur >= opt[cur].mlen); - if (opt[cur].mlen != 0) { - U32 const prev = cur - opt[cur].mlen; - repcodes_t const newReps = ZSTD_newRep(opt[prev].rep, opt[cur].off, opt[cur].litlen==0); - ZSTD_memcpy(opt[cur].rep, &newReps, sizeof(repcodes_t)); - } else { - ZSTD_memcpy(opt[cur].rep, opt[cur - 1].rep, sizeof(repcodes_t)); - } - - /* last match must start at a minimum distance of 8 from oend */ - if (inr > ilimit) continue; - - if (cur == last_pos) break; - - if ( (optLevel==0) /*static_test*/ - && (opt[cur+1].price <= opt[cur].price + (BITCOST_MULTIPLIER/2)) ) { - DEBUGLOG(7, "move to next rPos:%u : price is <=", cur+1); - continue; /* skip unpromising positions; about ~+6% speed, -0.01 ratio */ - } - - assert(opt[cur].price >= 0); - { U32 const ll0 = (opt[cur].mlen != 0); - U32 const litlen = (opt[cur].mlen == 0) ? opt[cur].litlen : 0; - U32 const previousPrice = (U32)opt[cur].price; - U32 const basePrice = previousPrice + ZSTD_litLengthPrice(0, optStatePtr, optLevel); - U32 nbMatches = getAllMatches(matches, ms, &nextToUpdate3, inr, iend, opt[cur].rep, ll0, minMatch); - U32 matchNb; - - ZSTD_optLdm_processMatchCandidate(&optLdm, matches, &nbMatches, - (U32)(inr-istart), (U32)(iend-inr)); - - if (!nbMatches) { - DEBUGLOG(7, "rPos:%u : no match found", cur); - continue; - } - - { U32 const maxML = matches[nbMatches-1].len; - DEBUGLOG(7, "cPos:%zi==rPos:%u, found %u matches, of maxLength=%u", - inr-istart, cur, nbMatches, maxML); - - if ( (maxML > sufficient_len) - || (cur + maxML >= ZSTD_OPT_NUM) ) { - lastSequence.mlen = maxML; - lastSequence.off = matches[nbMatches-1].off; - lastSequence.litlen = litlen; - cur -= (opt[cur].mlen==0) ? opt[cur].litlen : 0; /* last sequence is actually only literals, fix cur to last match - note : may underflow, in which case, it's first sequence, and it's okay */ - last_pos = cur + ZSTD_totalLen(lastSequence); - if (cur > ZSTD_OPT_NUM) cur = 0; /* underflow => first match */ - goto _shortestPath; - } } - - /* set prices using matches found at position == cur */ - for (matchNb = 0; matchNb < nbMatches; matchNb++) { - U32 const offset = matches[matchNb].off; - U32 const lastML = matches[matchNb].len; - U32 const startML = (matchNb>0) ? matches[matchNb-1].len+1 : minMatch; - U32 mlen; - - DEBUGLOG(7, "testing match %u => offCode=%4u, mlen=%2u, llen=%2u", - matchNb, matches[matchNb].off, lastML, litlen); - - for (mlen = lastML; mlen >= startML; mlen--) { /* scan downward */ - U32 const pos = cur + mlen; - int const price = (int)basePrice + (int)ZSTD_getMatchPrice(offset, mlen, optStatePtr, optLevel); - - if ((pos > last_pos) || (price < opt[pos].price)) { - DEBUGLOG(7, "rPos:%u (ml=%2u) => new better price (%.2f<%.2f)", - pos, mlen, ZSTD_fCost(price), ZSTD_fCost(opt[pos].price)); - while (last_pos < pos) { opt[last_pos+1].price = ZSTD_MAX_PRICE; last_pos++; } /* fill empty positions */ - opt[pos].mlen = mlen; - opt[pos].off = offset; - opt[pos].litlen = litlen; - opt[pos].price = price; - } else { - DEBUGLOG(7, "rPos:%u (ml=%2u) => new price is worse (%.2f>=%.2f)", - pos, mlen, ZSTD_fCost(price), ZSTD_fCost(opt[pos].price)); - if (optLevel==0) break; /* early update abort; gets ~+10% speed for about -0.01 ratio loss */ - } - } } } - } /* for (cur = 1; cur <= last_pos; cur++) */ - - lastSequence = opt[last_pos]; - cur = last_pos > ZSTD_totalLen(lastSequence) ? last_pos - ZSTD_totalLen(lastSequence) : 0; /* single sequence, and it starts before `ip` */ - assert(cur < ZSTD_OPT_NUM); /* control overflow*/ - -_shortestPath: /* cur, last_pos, best_mlen, best_off have to be set */ - assert(opt[0].mlen == 0); - - /* Set the next chunk's repcodes based on the repcodes of the beginning - * of the last match, and the last sequence. This avoids us having to - * update them while traversing the sequences. - */ - if (lastSequence.mlen != 0) { - repcodes_t const reps = ZSTD_newRep(opt[cur].rep, lastSequence.off, lastSequence.litlen==0); - ZSTD_memcpy(rep, &reps, sizeof(reps)); - } else { - ZSTD_memcpy(rep, opt[cur].rep, sizeof(repcodes_t)); - } - - { U32 const storeEnd = cur + 1; - U32 storeStart = storeEnd; - U32 seqPos = cur; - - DEBUGLOG(6, "start reverse traversal (last_pos:%u, cur:%u)", - last_pos, cur); (void)last_pos; - assert(storeEnd < ZSTD_OPT_NUM); - DEBUGLOG(6, "last sequence copied into pos=%u (llen=%u,mlen=%u,ofc=%u)", - storeEnd, lastSequence.litlen, lastSequence.mlen, lastSequence.off); - opt[storeEnd] = lastSequence; - while (seqPos > 0) { - U32 const backDist = ZSTD_totalLen(opt[seqPos]); - storeStart--; - DEBUGLOG(6, "sequence from rPos=%u copied into pos=%u (llen=%u,mlen=%u,ofc=%u)", - seqPos, storeStart, opt[seqPos].litlen, opt[seqPos].mlen, opt[seqPos].off); - opt[storeStart] = opt[seqPos]; - seqPos = (seqPos > backDist) ? seqPos - backDist : 0; - } - - /* save sequences */ - DEBUGLOG(6, "sending selected sequences into seqStore") - { U32 storePos; - for (storePos=storeStart; storePos <= storeEnd; storePos++) { - U32 const llen = opt[storePos].litlen; - U32 const mlen = opt[storePos].mlen; - U32 const offCode = opt[storePos].off; - U32 const advance = llen + mlen; - DEBUGLOG(6, "considering seq starting at %zi, llen=%u, mlen=%u", - anchor - istart, (unsigned)llen, (unsigned)mlen); - - if (mlen==0) { /* only literals => must be last "sequence", actually starting a new stream of sequences */ - assert(storePos == storeEnd); /* must be last sequence */ - ip = anchor + llen; /* last "sequence" is a bunch of literals => don't progress anchor */ - continue; /* will finish */ - } - - assert(anchor + llen <= iend); - ZSTD_updateStats(optStatePtr, llen, anchor, offCode, mlen); - ZSTD_storeSeq(seqStore, llen, anchor, iend, offCode, mlen); - anchor += advance; - ip = anchor; - } } - ZSTD_setBasePrices(optStatePtr, optLevel); - } - } /* while (ip < ilimit) */ - - /* Return the last literals size */ - return (size_t)(iend - anchor); -} - -static size_t ZSTD_compressBlock_opt0( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize, const ZSTD_dictMode_e dictMode) -{ - return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, src, srcSize, 0 /* optLevel */, dictMode); -} - -static size_t ZSTD_compressBlock_opt2( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize, const ZSTD_dictMode_e dictMode) -{ - return ZSTD_compressBlock_opt_generic(ms, seqStore, rep, src, srcSize, 2 /* optLevel */, dictMode); -} - -size_t ZSTD_compressBlock_btopt( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize) -{ - DEBUGLOG(5, "ZSTD_compressBlock_btopt"); - return ZSTD_compressBlock_opt0(ms, seqStore, rep, src, srcSize, ZSTD_noDict); -} - - - - -/* ZSTD_initStats_ultra(): - * make a first compression pass, just to seed stats with more accurate starting values. - * only works on first block, with no dictionary and no ldm. - * this function cannot error, hence its contract must be respected. - */ -static void -ZSTD_initStats_ultra(ZSTD_matchState_t* ms, - seqStore_t* seqStore, - U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize) -{ - U32 tmpRep[ZSTD_REP_NUM]; /* updated rep codes will sink here */ - ZSTD_memcpy(tmpRep, rep, sizeof(tmpRep)); - - DEBUGLOG(4, "ZSTD_initStats_ultra (srcSize=%zu)", srcSize); - assert(ms->opt.litLengthSum == 0); /* first block */ - assert(seqStore->sequences == seqStore->sequencesStart); /* no ldm */ - assert(ms->window.dictLimit == ms->window.lowLimit); /* no dictionary */ - assert(ms->window.dictLimit - ms->nextToUpdate <= 1); /* no prefix (note: intentional overflow, defined as 2-complement) */ - - ZSTD_compressBlock_opt2(ms, seqStore, tmpRep, src, srcSize, ZSTD_noDict); /* generate stats into ms->opt*/ - - /* invalidate first scan from history */ - ZSTD_resetSeqStore(seqStore); - ms->window.base -= srcSize; - ms->window.dictLimit += (U32)srcSize; - ms->window.lowLimit = ms->window.dictLimit; - ms->nextToUpdate = ms->window.dictLimit; - -} - -size_t ZSTD_compressBlock_btultra( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize) -{ - DEBUGLOG(5, "ZSTD_compressBlock_btultra (srcSize=%zu)", srcSize); - return ZSTD_compressBlock_opt2(ms, seqStore, rep, src, srcSize, ZSTD_noDict); -} - -size_t ZSTD_compressBlock_btultra2( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize) -{ - U32 const curr = (U32)((const BYTE*)src - ms->window.base); - DEBUGLOG(5, "ZSTD_compressBlock_btultra2 (srcSize=%zu)", srcSize); - - /* 2-pass strategy: - * this strategy makes a first pass over first block to collect statistics - * and seed next round's statistics with it. - * After 1st pass, function forgets everything, and starts a new block. - * Consequently, this can only work if no data has been previously loaded in tables, - * aka, no dictionary, no prefix, no ldm preprocessing. - * The compression ratio gain is generally small (~0.5% on first block), - * the cost is 2x cpu time on first block. */ - assert(srcSize <= ZSTD_BLOCKSIZE_MAX); - if ( (ms->opt.litLengthSum==0) /* first block */ - && (seqStore->sequences == seqStore->sequencesStart) /* no ldm */ - && (ms->window.dictLimit == ms->window.lowLimit) /* no dictionary */ - && (curr == ms->window.dictLimit) /* start of frame, nothing already loaded nor skipped */ - && (srcSize > ZSTD_PREDEF_THRESHOLD) - ) { - ZSTD_initStats_ultra(ms, seqStore, rep, src, srcSize); - } - - return ZSTD_compressBlock_opt2(ms, seqStore, rep, src, srcSize, ZSTD_noDict); -} - -size_t ZSTD_compressBlock_btopt_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize) -{ - return ZSTD_compressBlock_opt0(ms, seqStore, rep, src, srcSize, ZSTD_dictMatchState); -} - -size_t ZSTD_compressBlock_btultra_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize) -{ - return ZSTD_compressBlock_opt2(ms, seqStore, rep, src, srcSize, ZSTD_dictMatchState); -} - -size_t ZSTD_compressBlock_btopt_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize) -{ - return ZSTD_compressBlock_opt0(ms, seqStore, rep, src, srcSize, ZSTD_extDict); -} - -size_t ZSTD_compressBlock_btultra_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - const void* src, size_t srcSize) -{ - return ZSTD_compressBlock_opt2(ms, seqStore, rep, src, srcSize, ZSTD_extDict); -} - -/* note : no btultra2 variant for extDict nor dictMatchState, - * because btultra2 is not meant to work with dictionaries - * and is only specific for the first block (no prefix) */ diff --git a/dep/zstd/lib/compress/zstd_opt.h b/dep/zstd/lib/compress/zstd_opt.h deleted file mode 100644 index 627255f53..000000000 --- a/dep/zstd/lib/compress/zstd_opt.h +++ /dev/null @@ -1,56 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_OPT_H -#define ZSTD_OPT_H - -#if defined (__cplusplus) -extern "C" { -#endif - -#include "zstd_compress_internal.h" - -/* used in ZSTD_loadDictionaryContent() */ -void ZSTD_updateTree(ZSTD_matchState_t* ms, const BYTE* ip, const BYTE* iend); - -size_t ZSTD_compressBlock_btopt( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_btultra( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_btultra2( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); - - -size_t ZSTD_compressBlock_btopt_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_btultra_dictMatchState( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); - -size_t ZSTD_compressBlock_btopt_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); -size_t ZSTD_compressBlock_btultra_extDict( - ZSTD_matchState_t* ms, seqStore_t* seqStore, U32 rep[ZSTD_REP_NUM], - void const* src, size_t srcSize); - - /* note : no btultra2 variant for extDict nor dictMatchState, - * because btultra2 is not meant to work with dictionaries - * and is only specific for the first block (no prefix) */ - -#if defined (__cplusplus) -} -#endif - -#endif /* ZSTD_OPT_H */ diff --git a/dep/zstd/lib/compress/zstdmt_compress.c b/dep/zstd/lib/compress/zstdmt_compress.c deleted file mode 100644 index 6bc14b035..000000000 --- a/dep/zstd/lib/compress/zstdmt_compress.c +++ /dev/null @@ -1,1859 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -/* ====== Compiler specifics ====== */ -#if defined(_MSC_VER) -# pragma warning(disable : 4204) /* disable: C4204: non-constant aggregate initializer */ -#endif - - -/* ====== Constants ====== */ -#define ZSTDMT_OVERLAPLOG_DEFAULT 0 - - -/* ====== Dependencies ====== */ -#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memset, INT_MAX, UINT_MAX */ -#include "../common/mem.h" /* MEM_STATIC */ -#include "../common/pool.h" /* threadpool */ -#include "../common/threading.h" /* mutex */ -#include "zstd_compress_internal.h" /* MIN, ERROR, ZSTD_*, ZSTD_highbit32 */ -#include "zstd_ldm.h" -#include "zstdmt_compress.h" - -/* Guards code to support resizing the SeqPool. - * We will want to resize the SeqPool to save memory in the future. - * Until then, comment the code out since it is unused. - */ -#define ZSTD_RESIZE_SEQPOOL 0 - -/* ====== Debug ====== */ -#if defined(DEBUGLEVEL) && (DEBUGLEVEL>=2) \ - && !defined(_MSC_VER) \ - && !defined(__MINGW32__) - -# include -# include -# include - -# define DEBUG_PRINTHEX(l,p,n) { \ - unsigned debug_u; \ - for (debug_u=0; debug_u<(n); debug_u++) \ - RAWLOG(l, "%02X ", ((const unsigned char*)(p))[debug_u]); \ - RAWLOG(l, " \n"); \ -} - -static unsigned long long GetCurrentClockTimeMicroseconds(void) -{ - static clock_t _ticksPerSecond = 0; - if (_ticksPerSecond <= 0) _ticksPerSecond = sysconf(_SC_CLK_TCK); - - { struct tms junk; clock_t newTicks = (clock_t) times(&junk); - return ((((unsigned long long)newTicks)*(1000000))/_ticksPerSecond); -} } - -#define MUTEX_WAIT_TIME_DLEVEL 6 -#define ZSTD_PTHREAD_MUTEX_LOCK(mutex) { \ - if (DEBUGLEVEL >= MUTEX_WAIT_TIME_DLEVEL) { \ - unsigned long long const beforeTime = GetCurrentClockTimeMicroseconds(); \ - ZSTD_pthread_mutex_lock(mutex); \ - { unsigned long long const afterTime = GetCurrentClockTimeMicroseconds(); \ - unsigned long long const elapsedTime = (afterTime-beforeTime); \ - if (elapsedTime > 1000) { /* or whatever threshold you like; I'm using 1 millisecond here */ \ - DEBUGLOG(MUTEX_WAIT_TIME_DLEVEL, "Thread took %llu microseconds to acquire mutex %s \n", \ - elapsedTime, #mutex); \ - } } \ - } else { \ - ZSTD_pthread_mutex_lock(mutex); \ - } \ -} - -#else - -# define ZSTD_PTHREAD_MUTEX_LOCK(m) ZSTD_pthread_mutex_lock(m) -# define DEBUG_PRINTHEX(l,p,n) {} - -#endif - - -/* ===== Buffer Pool ===== */ -/* a single Buffer Pool can be invoked from multiple threads in parallel */ - -typedef struct buffer_s { - void* start; - size_t capacity; -} buffer_t; - -static const buffer_t g_nullBuffer = { NULL, 0 }; - -typedef struct ZSTDMT_bufferPool_s { - ZSTD_pthread_mutex_t poolMutex; - size_t bufferSize; - unsigned totalBuffers; - unsigned nbBuffers; - ZSTD_customMem cMem; - buffer_t bTable[1]; /* variable size */ -} ZSTDMT_bufferPool; - -static ZSTDMT_bufferPool* ZSTDMT_createBufferPool(unsigned maxNbBuffers, ZSTD_customMem cMem) -{ - ZSTDMT_bufferPool* const bufPool = (ZSTDMT_bufferPool*)ZSTD_customCalloc( - sizeof(ZSTDMT_bufferPool) + (maxNbBuffers-1) * sizeof(buffer_t), cMem); - if (bufPool==NULL) return NULL; - if (ZSTD_pthread_mutex_init(&bufPool->poolMutex, NULL)) { - ZSTD_customFree(bufPool, cMem); - return NULL; - } - bufPool->bufferSize = 64 KB; - bufPool->totalBuffers = maxNbBuffers; - bufPool->nbBuffers = 0; - bufPool->cMem = cMem; - return bufPool; -} - -static void ZSTDMT_freeBufferPool(ZSTDMT_bufferPool* bufPool) -{ - unsigned u; - DEBUGLOG(3, "ZSTDMT_freeBufferPool (address:%08X)", (U32)(size_t)bufPool); - if (!bufPool) return; /* compatibility with free on NULL */ - for (u=0; utotalBuffers; u++) { - DEBUGLOG(4, "free buffer %2u (address:%08X)", u, (U32)(size_t)bufPool->bTable[u].start); - ZSTD_customFree(bufPool->bTable[u].start, bufPool->cMem); - } - ZSTD_pthread_mutex_destroy(&bufPool->poolMutex); - ZSTD_customFree(bufPool, bufPool->cMem); -} - -/* only works at initialization, not during compression */ -static size_t ZSTDMT_sizeof_bufferPool(ZSTDMT_bufferPool* bufPool) -{ - size_t const poolSize = sizeof(*bufPool) - + (bufPool->totalBuffers - 1) * sizeof(buffer_t); - unsigned u; - size_t totalBufferSize = 0; - ZSTD_pthread_mutex_lock(&bufPool->poolMutex); - for (u=0; utotalBuffers; u++) - totalBufferSize += bufPool->bTable[u].capacity; - ZSTD_pthread_mutex_unlock(&bufPool->poolMutex); - - return poolSize + totalBufferSize; -} - -/* ZSTDMT_setBufferSize() : - * all future buffers provided by this buffer pool will have _at least_ this size - * note : it's better for all buffers to have same size, - * as they become freely interchangeable, reducing malloc/free usages and memory fragmentation */ -static void ZSTDMT_setBufferSize(ZSTDMT_bufferPool* const bufPool, size_t const bSize) -{ - ZSTD_pthread_mutex_lock(&bufPool->poolMutex); - DEBUGLOG(4, "ZSTDMT_setBufferSize: bSize = %u", (U32)bSize); - bufPool->bufferSize = bSize; - ZSTD_pthread_mutex_unlock(&bufPool->poolMutex); -} - - -static ZSTDMT_bufferPool* ZSTDMT_expandBufferPool(ZSTDMT_bufferPool* srcBufPool, unsigned maxNbBuffers) -{ - if (srcBufPool==NULL) return NULL; - if (srcBufPool->totalBuffers >= maxNbBuffers) /* good enough */ - return srcBufPool; - /* need a larger buffer pool */ - { ZSTD_customMem const cMem = srcBufPool->cMem; - size_t const bSize = srcBufPool->bufferSize; /* forward parameters */ - ZSTDMT_bufferPool* newBufPool; - ZSTDMT_freeBufferPool(srcBufPool); - newBufPool = ZSTDMT_createBufferPool(maxNbBuffers, cMem); - if (newBufPool==NULL) return newBufPool; - ZSTDMT_setBufferSize(newBufPool, bSize); - return newBufPool; - } -} - -/** ZSTDMT_getBuffer() : - * assumption : bufPool must be valid - * @return : a buffer, with start pointer and size - * note: allocation may fail, in this case, start==NULL and size==0 */ -static buffer_t ZSTDMT_getBuffer(ZSTDMT_bufferPool* bufPool) -{ - size_t const bSize = bufPool->bufferSize; - DEBUGLOG(5, "ZSTDMT_getBuffer: bSize = %u", (U32)bufPool->bufferSize); - ZSTD_pthread_mutex_lock(&bufPool->poolMutex); - if (bufPool->nbBuffers) { /* try to use an existing buffer */ - buffer_t const buf = bufPool->bTable[--(bufPool->nbBuffers)]; - size_t const availBufferSize = buf.capacity; - bufPool->bTable[bufPool->nbBuffers] = g_nullBuffer; - if ((availBufferSize >= bSize) & ((availBufferSize>>3) <= bSize)) { - /* large enough, but not too much */ - DEBUGLOG(5, "ZSTDMT_getBuffer: provide buffer %u of size %u", - bufPool->nbBuffers, (U32)buf.capacity); - ZSTD_pthread_mutex_unlock(&bufPool->poolMutex); - return buf; - } - /* size conditions not respected : scratch this buffer, create new one */ - DEBUGLOG(5, "ZSTDMT_getBuffer: existing buffer does not meet size conditions => freeing"); - ZSTD_customFree(buf.start, bufPool->cMem); - } - ZSTD_pthread_mutex_unlock(&bufPool->poolMutex); - /* create new buffer */ - DEBUGLOG(5, "ZSTDMT_getBuffer: create a new buffer"); - { buffer_t buffer; - void* const start = ZSTD_customMalloc(bSize, bufPool->cMem); - buffer.start = start; /* note : start can be NULL if malloc fails ! */ - buffer.capacity = (start==NULL) ? 0 : bSize; - if (start==NULL) { - DEBUGLOG(5, "ZSTDMT_getBuffer: buffer allocation failure !!"); - } else { - DEBUGLOG(5, "ZSTDMT_getBuffer: created buffer of size %u", (U32)bSize); - } - return buffer; - } -} - -#if ZSTD_RESIZE_SEQPOOL -/** ZSTDMT_resizeBuffer() : - * assumption : bufPool must be valid - * @return : a buffer that is at least the buffer pool buffer size. - * If a reallocation happens, the data in the input buffer is copied. - */ -static buffer_t ZSTDMT_resizeBuffer(ZSTDMT_bufferPool* bufPool, buffer_t buffer) -{ - size_t const bSize = bufPool->bufferSize; - if (buffer.capacity < bSize) { - void* const start = ZSTD_customMalloc(bSize, bufPool->cMem); - buffer_t newBuffer; - newBuffer.start = start; - newBuffer.capacity = start == NULL ? 0 : bSize; - if (start != NULL) { - assert(newBuffer.capacity >= buffer.capacity); - ZSTD_memcpy(newBuffer.start, buffer.start, buffer.capacity); - DEBUGLOG(5, "ZSTDMT_resizeBuffer: created buffer of size %u", (U32)bSize); - return newBuffer; - } - DEBUGLOG(5, "ZSTDMT_resizeBuffer: buffer allocation failure !!"); - } - return buffer; -} -#endif - -/* store buffer for later re-use, up to pool capacity */ -static void ZSTDMT_releaseBuffer(ZSTDMT_bufferPool* bufPool, buffer_t buf) -{ - DEBUGLOG(5, "ZSTDMT_releaseBuffer"); - if (buf.start == NULL) return; /* compatible with release on NULL */ - ZSTD_pthread_mutex_lock(&bufPool->poolMutex); - if (bufPool->nbBuffers < bufPool->totalBuffers) { - bufPool->bTable[bufPool->nbBuffers++] = buf; /* stored for later use */ - DEBUGLOG(5, "ZSTDMT_releaseBuffer: stored buffer of size %u in slot %u", - (U32)buf.capacity, (U32)(bufPool->nbBuffers-1)); - ZSTD_pthread_mutex_unlock(&bufPool->poolMutex); - return; - } - ZSTD_pthread_mutex_unlock(&bufPool->poolMutex); - /* Reached bufferPool capacity (should not happen) */ - DEBUGLOG(5, "ZSTDMT_releaseBuffer: pool capacity reached => freeing "); - ZSTD_customFree(buf.start, bufPool->cMem); -} - -/* We need 2 output buffers per worker since each dstBuff must be flushed after it is released. - * The 3 additional buffers are as follows: - * 1 buffer for input loading - * 1 buffer for "next input" when submitting current one - * 1 buffer stuck in queue */ -#define BUF_POOL_MAX_NB_BUFFERS(nbWorkers) 2*nbWorkers + 3 - -/* After a worker releases its rawSeqStore, it is immediately ready for reuse. - * So we only need one seq buffer per worker. */ -#define SEQ_POOL_MAX_NB_BUFFERS(nbWorkers) nbWorkers - -/* ===== Seq Pool Wrapper ====== */ - -typedef ZSTDMT_bufferPool ZSTDMT_seqPool; - -static size_t ZSTDMT_sizeof_seqPool(ZSTDMT_seqPool* seqPool) -{ - return ZSTDMT_sizeof_bufferPool(seqPool); -} - -static rawSeqStore_t bufferToSeq(buffer_t buffer) -{ - rawSeqStore_t seq = kNullRawSeqStore; - seq.seq = (rawSeq*)buffer.start; - seq.capacity = buffer.capacity / sizeof(rawSeq); - return seq; -} - -static buffer_t seqToBuffer(rawSeqStore_t seq) -{ - buffer_t buffer; - buffer.start = seq.seq; - buffer.capacity = seq.capacity * sizeof(rawSeq); - return buffer; -} - -static rawSeqStore_t ZSTDMT_getSeq(ZSTDMT_seqPool* seqPool) -{ - if (seqPool->bufferSize == 0) { - return kNullRawSeqStore; - } - return bufferToSeq(ZSTDMT_getBuffer(seqPool)); -} - -#if ZSTD_RESIZE_SEQPOOL -static rawSeqStore_t ZSTDMT_resizeSeq(ZSTDMT_seqPool* seqPool, rawSeqStore_t seq) -{ - return bufferToSeq(ZSTDMT_resizeBuffer(seqPool, seqToBuffer(seq))); -} -#endif - -static void ZSTDMT_releaseSeq(ZSTDMT_seqPool* seqPool, rawSeqStore_t seq) -{ - ZSTDMT_releaseBuffer(seqPool, seqToBuffer(seq)); -} - -static void ZSTDMT_setNbSeq(ZSTDMT_seqPool* const seqPool, size_t const nbSeq) -{ - ZSTDMT_setBufferSize(seqPool, nbSeq * sizeof(rawSeq)); -} - -static ZSTDMT_seqPool* ZSTDMT_createSeqPool(unsigned nbWorkers, ZSTD_customMem cMem) -{ - ZSTDMT_seqPool* const seqPool = ZSTDMT_createBufferPool(SEQ_POOL_MAX_NB_BUFFERS(nbWorkers), cMem); - if (seqPool == NULL) return NULL; - ZSTDMT_setNbSeq(seqPool, 0); - return seqPool; -} - -static void ZSTDMT_freeSeqPool(ZSTDMT_seqPool* seqPool) -{ - ZSTDMT_freeBufferPool(seqPool); -} - -static ZSTDMT_seqPool* ZSTDMT_expandSeqPool(ZSTDMT_seqPool* pool, U32 nbWorkers) -{ - return ZSTDMT_expandBufferPool(pool, SEQ_POOL_MAX_NB_BUFFERS(nbWorkers)); -} - - -/* ===== CCtx Pool ===== */ -/* a single CCtx Pool can be invoked from multiple threads in parallel */ - -typedef struct { - ZSTD_pthread_mutex_t poolMutex; - int totalCCtx; - int availCCtx; - ZSTD_customMem cMem; - ZSTD_CCtx* cctx[1]; /* variable size */ -} ZSTDMT_CCtxPool; - -/* note : all CCtx borrowed from the pool should be released back to the pool _before_ freeing the pool */ -static void ZSTDMT_freeCCtxPool(ZSTDMT_CCtxPool* pool) -{ - int cid; - for (cid=0; cidtotalCCtx; cid++) - ZSTD_freeCCtx(pool->cctx[cid]); /* note : compatible with free on NULL */ - ZSTD_pthread_mutex_destroy(&pool->poolMutex); - ZSTD_customFree(pool, pool->cMem); -} - -/* ZSTDMT_createCCtxPool() : - * implies nbWorkers >= 1 , checked by caller ZSTDMT_createCCtx() */ -static ZSTDMT_CCtxPool* ZSTDMT_createCCtxPool(int nbWorkers, - ZSTD_customMem cMem) -{ - ZSTDMT_CCtxPool* const cctxPool = (ZSTDMT_CCtxPool*) ZSTD_customCalloc( - sizeof(ZSTDMT_CCtxPool) + (nbWorkers-1)*sizeof(ZSTD_CCtx*), cMem); - assert(nbWorkers > 0); - if (!cctxPool) return NULL; - if (ZSTD_pthread_mutex_init(&cctxPool->poolMutex, NULL)) { - ZSTD_customFree(cctxPool, cMem); - return NULL; - } - cctxPool->cMem = cMem; - cctxPool->totalCCtx = nbWorkers; - cctxPool->availCCtx = 1; /* at least one cctx for single-thread mode */ - cctxPool->cctx[0] = ZSTD_createCCtx_advanced(cMem); - if (!cctxPool->cctx[0]) { ZSTDMT_freeCCtxPool(cctxPool); return NULL; } - DEBUGLOG(3, "cctxPool created, with %u workers", nbWorkers); - return cctxPool; -} - -static ZSTDMT_CCtxPool* ZSTDMT_expandCCtxPool(ZSTDMT_CCtxPool* srcPool, - int nbWorkers) -{ - if (srcPool==NULL) return NULL; - if (nbWorkers <= srcPool->totalCCtx) return srcPool; /* good enough */ - /* need a larger cctx pool */ - { ZSTD_customMem const cMem = srcPool->cMem; - ZSTDMT_freeCCtxPool(srcPool); - return ZSTDMT_createCCtxPool(nbWorkers, cMem); - } -} - -/* only works during initialization phase, not during compression */ -static size_t ZSTDMT_sizeof_CCtxPool(ZSTDMT_CCtxPool* cctxPool) -{ - ZSTD_pthread_mutex_lock(&cctxPool->poolMutex); - { unsigned const nbWorkers = cctxPool->totalCCtx; - size_t const poolSize = sizeof(*cctxPool) - + (nbWorkers-1) * sizeof(ZSTD_CCtx*); - unsigned u; - size_t totalCCtxSize = 0; - for (u=0; ucctx[u]); - } - ZSTD_pthread_mutex_unlock(&cctxPool->poolMutex); - assert(nbWorkers > 0); - return poolSize + totalCCtxSize; - } -} - -static ZSTD_CCtx* ZSTDMT_getCCtx(ZSTDMT_CCtxPool* cctxPool) -{ - DEBUGLOG(5, "ZSTDMT_getCCtx"); - ZSTD_pthread_mutex_lock(&cctxPool->poolMutex); - if (cctxPool->availCCtx) { - cctxPool->availCCtx--; - { ZSTD_CCtx* const cctx = cctxPool->cctx[cctxPool->availCCtx]; - ZSTD_pthread_mutex_unlock(&cctxPool->poolMutex); - return cctx; - } } - ZSTD_pthread_mutex_unlock(&cctxPool->poolMutex); - DEBUGLOG(5, "create one more CCtx"); - return ZSTD_createCCtx_advanced(cctxPool->cMem); /* note : can be NULL, when creation fails ! */ -} - -static void ZSTDMT_releaseCCtx(ZSTDMT_CCtxPool* pool, ZSTD_CCtx* cctx) -{ - if (cctx==NULL) return; /* compatibility with release on NULL */ - ZSTD_pthread_mutex_lock(&pool->poolMutex); - if (pool->availCCtx < pool->totalCCtx) - pool->cctx[pool->availCCtx++] = cctx; - else { - /* pool overflow : should not happen, since totalCCtx==nbWorkers */ - DEBUGLOG(4, "CCtx pool overflow : free cctx"); - ZSTD_freeCCtx(cctx); - } - ZSTD_pthread_mutex_unlock(&pool->poolMutex); -} - -/* ==== Serial State ==== */ - -typedef struct { - void const* start; - size_t size; -} range_t; - -typedef struct { - /* All variables in the struct are protected by mutex. */ - ZSTD_pthread_mutex_t mutex; - ZSTD_pthread_cond_t cond; - ZSTD_CCtx_params params; - ldmState_t ldmState; - XXH64_state_t xxhState; - unsigned nextJobID; - /* Protects ldmWindow. - * Must be acquired after the main mutex when acquiring both. - */ - ZSTD_pthread_mutex_t ldmWindowMutex; - ZSTD_pthread_cond_t ldmWindowCond; /* Signaled when ldmWindow is updated */ - ZSTD_window_t ldmWindow; /* A thread-safe copy of ldmState.window */ -} serialState_t; - -static int -ZSTDMT_serialState_reset(serialState_t* serialState, - ZSTDMT_seqPool* seqPool, - ZSTD_CCtx_params params, - size_t jobSize, - const void* dict, size_t const dictSize, - ZSTD_dictContentType_e dictContentType) -{ - /* Adjust parameters */ - if (params.ldmParams.enableLdm == ZSTD_ps_enable) { - DEBUGLOG(4, "LDM window size = %u KB", (1U << params.cParams.windowLog) >> 10); - ZSTD_ldm_adjustParameters(¶ms.ldmParams, ¶ms.cParams); - assert(params.ldmParams.hashLog >= params.ldmParams.bucketSizeLog); - assert(params.ldmParams.hashRateLog < 32); - } else { - ZSTD_memset(¶ms.ldmParams, 0, sizeof(params.ldmParams)); - } - serialState->nextJobID = 0; - if (params.fParams.checksumFlag) - XXH64_reset(&serialState->xxhState, 0); - if (params.ldmParams.enableLdm == ZSTD_ps_enable) { - ZSTD_customMem cMem = params.customMem; - unsigned const hashLog = params.ldmParams.hashLog; - size_t const hashSize = ((size_t)1 << hashLog) * sizeof(ldmEntry_t); - unsigned const bucketLog = - params.ldmParams.hashLog - params.ldmParams.bucketSizeLog; - unsigned const prevBucketLog = - serialState->params.ldmParams.hashLog - - serialState->params.ldmParams.bucketSizeLog; - size_t const numBuckets = (size_t)1 << bucketLog; - /* Size the seq pool tables */ - ZSTDMT_setNbSeq(seqPool, ZSTD_ldm_getMaxNbSeq(params.ldmParams, jobSize)); - /* Reset the window */ - ZSTD_window_init(&serialState->ldmState.window); - /* Resize tables and output space if necessary. */ - if (serialState->ldmState.hashTable == NULL || serialState->params.ldmParams.hashLog < hashLog) { - ZSTD_customFree(serialState->ldmState.hashTable, cMem); - serialState->ldmState.hashTable = (ldmEntry_t*)ZSTD_customMalloc(hashSize, cMem); - } - if (serialState->ldmState.bucketOffsets == NULL || prevBucketLog < bucketLog) { - ZSTD_customFree(serialState->ldmState.bucketOffsets, cMem); - serialState->ldmState.bucketOffsets = (BYTE*)ZSTD_customMalloc(numBuckets, cMem); - } - if (!serialState->ldmState.hashTable || !serialState->ldmState.bucketOffsets) - return 1; - /* Zero the tables */ - ZSTD_memset(serialState->ldmState.hashTable, 0, hashSize); - ZSTD_memset(serialState->ldmState.bucketOffsets, 0, numBuckets); - - /* Update window state and fill hash table with dict */ - serialState->ldmState.loadedDictEnd = 0; - if (dictSize > 0) { - if (dictContentType == ZSTD_dct_rawContent) { - BYTE const* const dictEnd = (const BYTE*)dict + dictSize; - ZSTD_window_update(&serialState->ldmState.window, dict, dictSize, /* forceNonContiguous */ 0); - ZSTD_ldm_fillHashTable(&serialState->ldmState, (const BYTE*)dict, dictEnd, ¶ms.ldmParams); - serialState->ldmState.loadedDictEnd = params.forceWindow ? 0 : (U32)(dictEnd - serialState->ldmState.window.base); - } else { - /* don't even load anything */ - } - } - - /* Initialize serialState's copy of ldmWindow. */ - serialState->ldmWindow = serialState->ldmState.window; - } - - serialState->params = params; - serialState->params.jobSize = (U32)jobSize; - return 0; -} - -static int ZSTDMT_serialState_init(serialState_t* serialState) -{ - int initError = 0; - ZSTD_memset(serialState, 0, sizeof(*serialState)); - initError |= ZSTD_pthread_mutex_init(&serialState->mutex, NULL); - initError |= ZSTD_pthread_cond_init(&serialState->cond, NULL); - initError |= ZSTD_pthread_mutex_init(&serialState->ldmWindowMutex, NULL); - initError |= ZSTD_pthread_cond_init(&serialState->ldmWindowCond, NULL); - return initError; -} - -static void ZSTDMT_serialState_free(serialState_t* serialState) -{ - ZSTD_customMem cMem = serialState->params.customMem; - ZSTD_pthread_mutex_destroy(&serialState->mutex); - ZSTD_pthread_cond_destroy(&serialState->cond); - ZSTD_pthread_mutex_destroy(&serialState->ldmWindowMutex); - ZSTD_pthread_cond_destroy(&serialState->ldmWindowCond); - ZSTD_customFree(serialState->ldmState.hashTable, cMem); - ZSTD_customFree(serialState->ldmState.bucketOffsets, cMem); -} - -static void ZSTDMT_serialState_update(serialState_t* serialState, - ZSTD_CCtx* jobCCtx, rawSeqStore_t seqStore, - range_t src, unsigned jobID) -{ - /* Wait for our turn */ - ZSTD_PTHREAD_MUTEX_LOCK(&serialState->mutex); - while (serialState->nextJobID < jobID) { - DEBUGLOG(5, "wait for serialState->cond"); - ZSTD_pthread_cond_wait(&serialState->cond, &serialState->mutex); - } - /* A future job may error and skip our job */ - if (serialState->nextJobID == jobID) { - /* It is now our turn, do any processing necessary */ - if (serialState->params.ldmParams.enableLdm == ZSTD_ps_enable) { - size_t error; - assert(seqStore.seq != NULL && seqStore.pos == 0 && - seqStore.size == 0 && seqStore.capacity > 0); - assert(src.size <= serialState->params.jobSize); - ZSTD_window_update(&serialState->ldmState.window, src.start, src.size, /* forceNonContiguous */ 0); - error = ZSTD_ldm_generateSequences( - &serialState->ldmState, &seqStore, - &serialState->params.ldmParams, src.start, src.size); - /* We provide a large enough buffer to never fail. */ - assert(!ZSTD_isError(error)); (void)error; - /* Update ldmWindow to match the ldmState.window and signal the main - * thread if it is waiting for a buffer. - */ - ZSTD_PTHREAD_MUTEX_LOCK(&serialState->ldmWindowMutex); - serialState->ldmWindow = serialState->ldmState.window; - ZSTD_pthread_cond_signal(&serialState->ldmWindowCond); - ZSTD_pthread_mutex_unlock(&serialState->ldmWindowMutex); - } - if (serialState->params.fParams.checksumFlag && src.size > 0) - XXH64_update(&serialState->xxhState, src.start, src.size); - } - /* Now it is the next jobs turn */ - serialState->nextJobID++; - ZSTD_pthread_cond_broadcast(&serialState->cond); - ZSTD_pthread_mutex_unlock(&serialState->mutex); - - if (seqStore.size > 0) { - size_t const err = ZSTD_referenceExternalSequences( - jobCCtx, seqStore.seq, seqStore.size); - assert(serialState->params.ldmParams.enableLdm == ZSTD_ps_enable); - assert(!ZSTD_isError(err)); - (void)err; - } -} - -static void ZSTDMT_serialState_ensureFinished(serialState_t* serialState, - unsigned jobID, size_t cSize) -{ - ZSTD_PTHREAD_MUTEX_LOCK(&serialState->mutex); - if (serialState->nextJobID <= jobID) { - assert(ZSTD_isError(cSize)); (void)cSize; - DEBUGLOG(5, "Skipping past job %u because of error", jobID); - serialState->nextJobID = jobID + 1; - ZSTD_pthread_cond_broadcast(&serialState->cond); - - ZSTD_PTHREAD_MUTEX_LOCK(&serialState->ldmWindowMutex); - ZSTD_window_clear(&serialState->ldmWindow); - ZSTD_pthread_cond_signal(&serialState->ldmWindowCond); - ZSTD_pthread_mutex_unlock(&serialState->ldmWindowMutex); - } - ZSTD_pthread_mutex_unlock(&serialState->mutex); - -} - - -/* ------------------------------------------ */ -/* ===== Worker thread ===== */ -/* ------------------------------------------ */ - -static const range_t kNullRange = { NULL, 0 }; - -typedef struct { - size_t consumed; /* SHARED - set0 by mtctx, then modified by worker AND read by mtctx */ - size_t cSize; /* SHARED - set0 by mtctx, then modified by worker AND read by mtctx, then set0 by mtctx */ - ZSTD_pthread_mutex_t job_mutex; /* Thread-safe - used by mtctx and worker */ - ZSTD_pthread_cond_t job_cond; /* Thread-safe - used by mtctx and worker */ - ZSTDMT_CCtxPool* cctxPool; /* Thread-safe - used by mtctx and (all) workers */ - ZSTDMT_bufferPool* bufPool; /* Thread-safe - used by mtctx and (all) workers */ - ZSTDMT_seqPool* seqPool; /* Thread-safe - used by mtctx and (all) workers */ - serialState_t* serial; /* Thread-safe - used by mtctx and (all) workers */ - buffer_t dstBuff; /* set by worker (or mtctx), then read by worker & mtctx, then modified by mtctx => no barrier */ - range_t prefix; /* set by mtctx, then read by worker & mtctx => no barrier */ - range_t src; /* set by mtctx, then read by worker & mtctx => no barrier */ - unsigned jobID; /* set by mtctx, then read by worker => no barrier */ - unsigned firstJob; /* set by mtctx, then read by worker => no barrier */ - unsigned lastJob; /* set by mtctx, then read by worker => no barrier */ - ZSTD_CCtx_params params; /* set by mtctx, then read by worker => no barrier */ - const ZSTD_CDict* cdict; /* set by mtctx, then read by worker => no barrier */ - unsigned long long fullFrameSize; /* set by mtctx, then read by worker => no barrier */ - size_t dstFlushed; /* used only by mtctx */ - unsigned frameChecksumNeeded; /* used only by mtctx */ -} ZSTDMT_jobDescription; - -#define JOB_ERROR(e) { \ - ZSTD_PTHREAD_MUTEX_LOCK(&job->job_mutex); \ - job->cSize = e; \ - ZSTD_pthread_mutex_unlock(&job->job_mutex); \ - goto _endJob; \ -} - -/* ZSTDMT_compressionJob() is a POOL_function type */ -static void ZSTDMT_compressionJob(void* jobDescription) -{ - ZSTDMT_jobDescription* const job = (ZSTDMT_jobDescription*)jobDescription; - ZSTD_CCtx_params jobParams = job->params; /* do not modify job->params ! copy it, modify the copy */ - ZSTD_CCtx* const cctx = ZSTDMT_getCCtx(job->cctxPool); - rawSeqStore_t rawSeqStore = ZSTDMT_getSeq(job->seqPool); - buffer_t dstBuff = job->dstBuff; - size_t lastCBlockSize = 0; - - /* resources */ - if (cctx==NULL) JOB_ERROR(ERROR(memory_allocation)); - if (dstBuff.start == NULL) { /* streaming job : doesn't provide a dstBuffer */ - dstBuff = ZSTDMT_getBuffer(job->bufPool); - if (dstBuff.start==NULL) JOB_ERROR(ERROR(memory_allocation)); - job->dstBuff = dstBuff; /* this value can be read in ZSTDMT_flush, when it copies the whole job */ - } - if (jobParams.ldmParams.enableLdm == ZSTD_ps_enable && rawSeqStore.seq == NULL) - JOB_ERROR(ERROR(memory_allocation)); - - /* Don't compute the checksum for chunks, since we compute it externally, - * but write it in the header. - */ - if (job->jobID != 0) jobParams.fParams.checksumFlag = 0; - /* Don't run LDM for the chunks, since we handle it externally */ - jobParams.ldmParams.enableLdm = ZSTD_ps_disable; - /* Correct nbWorkers to 0. */ - jobParams.nbWorkers = 0; - - - /* init */ - if (job->cdict) { - size_t const initError = ZSTD_compressBegin_advanced_internal(cctx, NULL, 0, ZSTD_dct_auto, ZSTD_dtlm_fast, job->cdict, &jobParams, job->fullFrameSize); - assert(job->firstJob); /* only allowed for first job */ - if (ZSTD_isError(initError)) JOB_ERROR(initError); - } else { /* srcStart points at reloaded section */ - U64 const pledgedSrcSize = job->firstJob ? job->fullFrameSize : job->src.size; - { size_t const forceWindowError = ZSTD_CCtxParams_setParameter(&jobParams, ZSTD_c_forceMaxWindow, !job->firstJob); - if (ZSTD_isError(forceWindowError)) JOB_ERROR(forceWindowError); - } - if (!job->firstJob) { - size_t const err = ZSTD_CCtxParams_setParameter(&jobParams, ZSTD_c_deterministicRefPrefix, 0); - if (ZSTD_isError(err)) JOB_ERROR(err); - } - { size_t const initError = ZSTD_compressBegin_advanced_internal(cctx, - job->prefix.start, job->prefix.size, ZSTD_dct_rawContent, /* load dictionary in "content-only" mode (no header analysis) */ - ZSTD_dtlm_fast, - NULL, /*cdict*/ - &jobParams, pledgedSrcSize); - if (ZSTD_isError(initError)) JOB_ERROR(initError); - } } - - /* Perform serial step as early as possible, but after CCtx initialization */ - ZSTDMT_serialState_update(job->serial, cctx, rawSeqStore, job->src, job->jobID); - - if (!job->firstJob) { /* flush and overwrite frame header when it's not first job */ - size_t const hSize = ZSTD_compressContinue(cctx, dstBuff.start, dstBuff.capacity, job->src.start, 0); - if (ZSTD_isError(hSize)) JOB_ERROR(hSize); - DEBUGLOG(5, "ZSTDMT_compressionJob: flush and overwrite %u bytes of frame header (not first job)", (U32)hSize); - ZSTD_invalidateRepCodes(cctx); - } - - /* compress */ - { size_t const chunkSize = 4*ZSTD_BLOCKSIZE_MAX; - int const nbChunks = (int)((job->src.size + (chunkSize-1)) / chunkSize); - const BYTE* ip = (const BYTE*) job->src.start; - BYTE* const ostart = (BYTE*)dstBuff.start; - BYTE* op = ostart; - BYTE* oend = op + dstBuff.capacity; - int chunkNb; - if (sizeof(size_t) > sizeof(int)) assert(job->src.size < ((size_t)INT_MAX) * chunkSize); /* check overflow */ - DEBUGLOG(5, "ZSTDMT_compressionJob: compress %u bytes in %i blocks", (U32)job->src.size, nbChunks); - assert(job->cSize == 0); - for (chunkNb = 1; chunkNb < nbChunks; chunkNb++) { - size_t const cSize = ZSTD_compressContinue(cctx, op, oend-op, ip, chunkSize); - if (ZSTD_isError(cSize)) JOB_ERROR(cSize); - ip += chunkSize; - op += cSize; assert(op < oend); - /* stats */ - ZSTD_PTHREAD_MUTEX_LOCK(&job->job_mutex); - job->cSize += cSize; - job->consumed = chunkSize * chunkNb; - DEBUGLOG(5, "ZSTDMT_compressionJob: compress new block : cSize==%u bytes (total: %u)", - (U32)cSize, (U32)job->cSize); - ZSTD_pthread_cond_signal(&job->job_cond); /* warns some more data is ready to be flushed */ - ZSTD_pthread_mutex_unlock(&job->job_mutex); - } - /* last block */ - assert(chunkSize > 0); - assert((chunkSize & (chunkSize - 1)) == 0); /* chunkSize must be power of 2 for mask==(chunkSize-1) to work */ - if ((nbChunks > 0) | job->lastJob /*must output a "last block" flag*/ ) { - size_t const lastBlockSize1 = job->src.size & (chunkSize-1); - size_t const lastBlockSize = ((lastBlockSize1==0) & (job->src.size>=chunkSize)) ? chunkSize : lastBlockSize1; - size_t const cSize = (job->lastJob) ? - ZSTD_compressEnd (cctx, op, oend-op, ip, lastBlockSize) : - ZSTD_compressContinue(cctx, op, oend-op, ip, lastBlockSize); - if (ZSTD_isError(cSize)) JOB_ERROR(cSize); - lastCBlockSize = cSize; - } } - if (!job->firstJob) { - /* Double check that we don't have an ext-dict, because then our - * repcode invalidation doesn't work. - */ - assert(!ZSTD_window_hasExtDict(cctx->blockState.matchState.window)); - } - ZSTD_CCtx_trace(cctx, 0); - -_endJob: - ZSTDMT_serialState_ensureFinished(job->serial, job->jobID, job->cSize); - if (job->prefix.size > 0) - DEBUGLOG(5, "Finished with prefix: %zx", (size_t)job->prefix.start); - DEBUGLOG(5, "Finished with source: %zx", (size_t)job->src.start); - /* release resources */ - ZSTDMT_releaseSeq(job->seqPool, rawSeqStore); - ZSTDMT_releaseCCtx(job->cctxPool, cctx); - /* report */ - ZSTD_PTHREAD_MUTEX_LOCK(&job->job_mutex); - if (ZSTD_isError(job->cSize)) assert(lastCBlockSize == 0); - job->cSize += lastCBlockSize; - job->consumed = job->src.size; /* when job->consumed == job->src.size , compression job is presumed completed */ - ZSTD_pthread_cond_signal(&job->job_cond); - ZSTD_pthread_mutex_unlock(&job->job_mutex); -} - - -/* ------------------------------------------ */ -/* ===== Multi-threaded compression ===== */ -/* ------------------------------------------ */ - -typedef struct { - range_t prefix; /* read-only non-owned prefix buffer */ - buffer_t buffer; - size_t filled; -} inBuff_t; - -typedef struct { - BYTE* buffer; /* The round input buffer. All jobs get references - * to pieces of the buffer. ZSTDMT_tryGetInputRange() - * handles handing out job input buffers, and makes - * sure it doesn't overlap with any pieces still in use. - */ - size_t capacity; /* The capacity of buffer. */ - size_t pos; /* The position of the current inBuff in the round - * buffer. Updated past the end if the inBuff once - * the inBuff is sent to the worker thread. - * pos <= capacity. - */ -} roundBuff_t; - -static const roundBuff_t kNullRoundBuff = {NULL, 0, 0}; - -#define RSYNC_LENGTH 32 -/* Don't create chunks smaller than the zstd block size. - * This stops us from regressing compression ratio too much, - * and ensures our output fits in ZSTD_compressBound(). - * - * If this is shrunk < ZSTD_BLOCKSIZELOG_MIN then - * ZSTD_COMPRESSBOUND() will need to be updated. - */ -#define RSYNC_MIN_BLOCK_LOG ZSTD_BLOCKSIZELOG_MAX -#define RSYNC_MIN_BLOCK_SIZE (1< one job is already prepared, but pool has shortage of workers. Don't create a new job. */ - inBuff_t inBuff; - roundBuff_t roundBuff; - serialState_t serial; - rsyncState_t rsync; - unsigned jobIDMask; - unsigned doneJobID; - unsigned nextJobID; - unsigned frameEnded; - unsigned allJobsCompleted; - unsigned long long frameContentSize; - unsigned long long consumed; - unsigned long long produced; - ZSTD_customMem cMem; - ZSTD_CDict* cdictLocal; - const ZSTD_CDict* cdict; - unsigned providedFactory: 1; -}; - -static void ZSTDMT_freeJobsTable(ZSTDMT_jobDescription* jobTable, U32 nbJobs, ZSTD_customMem cMem) -{ - U32 jobNb; - if (jobTable == NULL) return; - for (jobNb=0; jobNb mtctx->jobIDMask+1) { /* need more job capacity */ - ZSTDMT_freeJobsTable(mtctx->jobs, mtctx->jobIDMask+1, mtctx->cMem); - mtctx->jobIDMask = 0; - mtctx->jobs = ZSTDMT_createJobsTable(&nbJobs, mtctx->cMem); - if (mtctx->jobs==NULL) return ERROR(memory_allocation); - assert((nbJobs != 0) && ((nbJobs & (nbJobs - 1)) == 0)); /* ensure nbJobs is a power of 2 */ - mtctx->jobIDMask = nbJobs - 1; - } - return 0; -} - - -/* ZSTDMT_CCtxParam_setNbWorkers(): - * Internal use only */ -static size_t ZSTDMT_CCtxParam_setNbWorkers(ZSTD_CCtx_params* params, unsigned nbWorkers) -{ - return ZSTD_CCtxParams_setParameter(params, ZSTD_c_nbWorkers, (int)nbWorkers); -} - -MEM_STATIC ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced_internal(unsigned nbWorkers, ZSTD_customMem cMem, ZSTD_threadPool* pool) -{ - ZSTDMT_CCtx* mtctx; - U32 nbJobs = nbWorkers + 2; - int initError; - DEBUGLOG(3, "ZSTDMT_createCCtx_advanced (nbWorkers = %u)", nbWorkers); - - if (nbWorkers < 1) return NULL; - nbWorkers = MIN(nbWorkers , ZSTDMT_NBWORKERS_MAX); - if ((cMem.customAlloc!=NULL) ^ (cMem.customFree!=NULL)) - /* invalid custom allocator */ - return NULL; - - mtctx = (ZSTDMT_CCtx*) ZSTD_customCalloc(sizeof(ZSTDMT_CCtx), cMem); - if (!mtctx) return NULL; - ZSTDMT_CCtxParam_setNbWorkers(&mtctx->params, nbWorkers); - mtctx->cMem = cMem; - mtctx->allJobsCompleted = 1; - if (pool != NULL) { - mtctx->factory = pool; - mtctx->providedFactory = 1; - } - else { - mtctx->factory = POOL_create_advanced(nbWorkers, 0, cMem); - mtctx->providedFactory = 0; - } - mtctx->jobs = ZSTDMT_createJobsTable(&nbJobs, cMem); - assert(nbJobs > 0); assert((nbJobs & (nbJobs - 1)) == 0); /* ensure nbJobs is a power of 2 */ - mtctx->jobIDMask = nbJobs - 1; - mtctx->bufPool = ZSTDMT_createBufferPool(BUF_POOL_MAX_NB_BUFFERS(nbWorkers), cMem); - mtctx->cctxPool = ZSTDMT_createCCtxPool(nbWorkers, cMem); - mtctx->seqPool = ZSTDMT_createSeqPool(nbWorkers, cMem); - initError = ZSTDMT_serialState_init(&mtctx->serial); - mtctx->roundBuff = kNullRoundBuff; - if (!mtctx->factory | !mtctx->jobs | !mtctx->bufPool | !mtctx->cctxPool | !mtctx->seqPool | initError) { - ZSTDMT_freeCCtx(mtctx); - return NULL; - } - DEBUGLOG(3, "mt_cctx created, for %u threads", nbWorkers); - return mtctx; -} - -ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced(unsigned nbWorkers, ZSTD_customMem cMem, ZSTD_threadPool* pool) -{ -#ifdef ZSTD_MULTITHREAD - return ZSTDMT_createCCtx_advanced_internal(nbWorkers, cMem, pool); -#else - (void)nbWorkers; - (void)cMem; - (void)pool; - return NULL; -#endif -} - - -/* ZSTDMT_releaseAllJobResources() : - * note : ensure all workers are killed first ! */ -static void ZSTDMT_releaseAllJobResources(ZSTDMT_CCtx* mtctx) -{ - unsigned jobID; - DEBUGLOG(3, "ZSTDMT_releaseAllJobResources"); - for (jobID=0; jobID <= mtctx->jobIDMask; jobID++) { - /* Copy the mutex/cond out */ - ZSTD_pthread_mutex_t const mutex = mtctx->jobs[jobID].job_mutex; - ZSTD_pthread_cond_t const cond = mtctx->jobs[jobID].job_cond; - - DEBUGLOG(4, "job%02u: release dst address %08X", jobID, (U32)(size_t)mtctx->jobs[jobID].dstBuff.start); - ZSTDMT_releaseBuffer(mtctx->bufPool, mtctx->jobs[jobID].dstBuff); - - /* Clear the job description, but keep the mutex/cond */ - ZSTD_memset(&mtctx->jobs[jobID], 0, sizeof(mtctx->jobs[jobID])); - mtctx->jobs[jobID].job_mutex = mutex; - mtctx->jobs[jobID].job_cond = cond; - } - mtctx->inBuff.buffer = g_nullBuffer; - mtctx->inBuff.filled = 0; - mtctx->allJobsCompleted = 1; -} - -static void ZSTDMT_waitForAllJobsCompleted(ZSTDMT_CCtx* mtctx) -{ - DEBUGLOG(4, "ZSTDMT_waitForAllJobsCompleted"); - while (mtctx->doneJobID < mtctx->nextJobID) { - unsigned const jobID = mtctx->doneJobID & mtctx->jobIDMask; - ZSTD_PTHREAD_MUTEX_LOCK(&mtctx->jobs[jobID].job_mutex); - while (mtctx->jobs[jobID].consumed < mtctx->jobs[jobID].src.size) { - DEBUGLOG(4, "waiting for jobCompleted signal from job %u", mtctx->doneJobID); /* we want to block when waiting for data to flush */ - ZSTD_pthread_cond_wait(&mtctx->jobs[jobID].job_cond, &mtctx->jobs[jobID].job_mutex); - } - ZSTD_pthread_mutex_unlock(&mtctx->jobs[jobID].job_mutex); - mtctx->doneJobID++; - } -} - -size_t ZSTDMT_freeCCtx(ZSTDMT_CCtx* mtctx) -{ - if (mtctx==NULL) return 0; /* compatible with free on NULL */ - if (!mtctx->providedFactory) - POOL_free(mtctx->factory); /* stop and free worker threads */ - ZSTDMT_releaseAllJobResources(mtctx); /* release job resources into pools first */ - ZSTDMT_freeJobsTable(mtctx->jobs, mtctx->jobIDMask+1, mtctx->cMem); - ZSTDMT_freeBufferPool(mtctx->bufPool); - ZSTDMT_freeCCtxPool(mtctx->cctxPool); - ZSTDMT_freeSeqPool(mtctx->seqPool); - ZSTDMT_serialState_free(&mtctx->serial); - ZSTD_freeCDict(mtctx->cdictLocal); - if (mtctx->roundBuff.buffer) - ZSTD_customFree(mtctx->roundBuff.buffer, mtctx->cMem); - ZSTD_customFree(mtctx, mtctx->cMem); - return 0; -} - -size_t ZSTDMT_sizeof_CCtx(ZSTDMT_CCtx* mtctx) -{ - if (mtctx == NULL) return 0; /* supports sizeof NULL */ - return sizeof(*mtctx) - + POOL_sizeof(mtctx->factory) - + ZSTDMT_sizeof_bufferPool(mtctx->bufPool) - + (mtctx->jobIDMask+1) * sizeof(ZSTDMT_jobDescription) - + ZSTDMT_sizeof_CCtxPool(mtctx->cctxPool) - + ZSTDMT_sizeof_seqPool(mtctx->seqPool) - + ZSTD_sizeof_CDict(mtctx->cdictLocal) - + mtctx->roundBuff.capacity; -} - - -/* ZSTDMT_resize() : - * @return : error code if fails, 0 on success */ -static size_t ZSTDMT_resize(ZSTDMT_CCtx* mtctx, unsigned nbWorkers) -{ - if (POOL_resize(mtctx->factory, nbWorkers)) return ERROR(memory_allocation); - FORWARD_IF_ERROR( ZSTDMT_expandJobsTable(mtctx, nbWorkers) , ""); - mtctx->bufPool = ZSTDMT_expandBufferPool(mtctx->bufPool, BUF_POOL_MAX_NB_BUFFERS(nbWorkers)); - if (mtctx->bufPool == NULL) return ERROR(memory_allocation); - mtctx->cctxPool = ZSTDMT_expandCCtxPool(mtctx->cctxPool, nbWorkers); - if (mtctx->cctxPool == NULL) return ERROR(memory_allocation); - mtctx->seqPool = ZSTDMT_expandSeqPool(mtctx->seqPool, nbWorkers); - if (mtctx->seqPool == NULL) return ERROR(memory_allocation); - ZSTDMT_CCtxParam_setNbWorkers(&mtctx->params, nbWorkers); - return 0; -} - - -/*! ZSTDMT_updateCParams_whileCompressing() : - * Updates a selected set of compression parameters, remaining compatible with currently active frame. - * New parameters will be applied to next compression job. */ -void ZSTDMT_updateCParams_whileCompressing(ZSTDMT_CCtx* mtctx, const ZSTD_CCtx_params* cctxParams) -{ - U32 const saved_wlog = mtctx->params.cParams.windowLog; /* Do not modify windowLog while compressing */ - int const compressionLevel = cctxParams->compressionLevel; - DEBUGLOG(5, "ZSTDMT_updateCParams_whileCompressing (level:%i)", - compressionLevel); - mtctx->params.compressionLevel = compressionLevel; - { ZSTD_compressionParameters cParams = ZSTD_getCParamsFromCCtxParams(cctxParams, ZSTD_CONTENTSIZE_UNKNOWN, 0, ZSTD_cpm_noAttachDict); - cParams.windowLog = saved_wlog; - mtctx->params.cParams = cParams; - } -} - -/* ZSTDMT_getFrameProgression(): - * tells how much data has been consumed (input) and produced (output) for current frame. - * able to count progression inside worker threads. - * Note : mutex will be acquired during statistics collection inside workers. */ -ZSTD_frameProgression ZSTDMT_getFrameProgression(ZSTDMT_CCtx* mtctx) -{ - ZSTD_frameProgression fps; - DEBUGLOG(5, "ZSTDMT_getFrameProgression"); - fps.ingested = mtctx->consumed + mtctx->inBuff.filled; - fps.consumed = mtctx->consumed; - fps.produced = fps.flushed = mtctx->produced; - fps.currentJobID = mtctx->nextJobID; - fps.nbActiveWorkers = 0; - { unsigned jobNb; - unsigned lastJobNb = mtctx->nextJobID + mtctx->jobReady; assert(mtctx->jobReady <= 1); - DEBUGLOG(6, "ZSTDMT_getFrameProgression: jobs: from %u to <%u (jobReady:%u)", - mtctx->doneJobID, lastJobNb, mtctx->jobReady) - for (jobNb = mtctx->doneJobID ; jobNb < lastJobNb ; jobNb++) { - unsigned const wJobID = jobNb & mtctx->jobIDMask; - ZSTDMT_jobDescription* jobPtr = &mtctx->jobs[wJobID]; - ZSTD_pthread_mutex_lock(&jobPtr->job_mutex); - { size_t const cResult = jobPtr->cSize; - size_t const produced = ZSTD_isError(cResult) ? 0 : cResult; - size_t const flushed = ZSTD_isError(cResult) ? 0 : jobPtr->dstFlushed; - assert(flushed <= produced); - fps.ingested += jobPtr->src.size; - fps.consumed += jobPtr->consumed; - fps.produced += produced; - fps.flushed += flushed; - fps.nbActiveWorkers += (jobPtr->consumed < jobPtr->src.size); - } - ZSTD_pthread_mutex_unlock(&mtctx->jobs[wJobID].job_mutex); - } - } - return fps; -} - - -size_t ZSTDMT_toFlushNow(ZSTDMT_CCtx* mtctx) -{ - size_t toFlush; - unsigned const jobID = mtctx->doneJobID; - assert(jobID <= mtctx->nextJobID); - if (jobID == mtctx->nextJobID) return 0; /* no active job => nothing to flush */ - - /* look into oldest non-fully-flushed job */ - { unsigned const wJobID = jobID & mtctx->jobIDMask; - ZSTDMT_jobDescription* const jobPtr = &mtctx->jobs[wJobID]; - ZSTD_pthread_mutex_lock(&jobPtr->job_mutex); - { size_t const cResult = jobPtr->cSize; - size_t const produced = ZSTD_isError(cResult) ? 0 : cResult; - size_t const flushed = ZSTD_isError(cResult) ? 0 : jobPtr->dstFlushed; - assert(flushed <= produced); - assert(jobPtr->consumed <= jobPtr->src.size); - toFlush = produced - flushed; - /* if toFlush==0, nothing is available to flush. - * However, jobID is expected to still be active: - * if jobID was already completed and fully flushed, - * ZSTDMT_flushProduced() should have already moved onto next job. - * Therefore, some input has not yet been consumed. */ - if (toFlush==0) { - assert(jobPtr->consumed < jobPtr->src.size); - } - } - ZSTD_pthread_mutex_unlock(&mtctx->jobs[wJobID].job_mutex); - } - - return toFlush; -} - - -/* ------------------------------------------ */ -/* ===== Multi-threaded compression ===== */ -/* ------------------------------------------ */ - -static unsigned ZSTDMT_computeTargetJobLog(const ZSTD_CCtx_params* params) -{ - unsigned jobLog; - if (params->ldmParams.enableLdm == ZSTD_ps_enable) { - /* In Long Range Mode, the windowLog is typically oversized. - * In which case, it's preferable to determine the jobSize - * based on cycleLog instead. */ - jobLog = MAX(21, ZSTD_cycleLog(params->cParams.chainLog, params->cParams.strategy) + 3); - } else { - jobLog = MAX(20, params->cParams.windowLog + 2); - } - return MIN(jobLog, (unsigned)ZSTDMT_JOBLOG_MAX); -} - -static int ZSTDMT_overlapLog_default(ZSTD_strategy strat) -{ - switch(strat) - { - case ZSTD_btultra2: - return 9; - case ZSTD_btultra: - case ZSTD_btopt: - return 8; - case ZSTD_btlazy2: - case ZSTD_lazy2: - return 7; - case ZSTD_lazy: - case ZSTD_greedy: - case ZSTD_dfast: - case ZSTD_fast: - default:; - } - return 6; -} - -static int ZSTDMT_overlapLog(int ovlog, ZSTD_strategy strat) -{ - assert(0 <= ovlog && ovlog <= 9); - if (ovlog == 0) return ZSTDMT_overlapLog_default(strat); - return ovlog; -} - -static size_t ZSTDMT_computeOverlapSize(const ZSTD_CCtx_params* params) -{ - int const overlapRLog = 9 - ZSTDMT_overlapLog(params->overlapLog, params->cParams.strategy); - int ovLog = (overlapRLog >= 8) ? 0 : (params->cParams.windowLog - overlapRLog); - assert(0 <= overlapRLog && overlapRLog <= 8); - if (params->ldmParams.enableLdm == ZSTD_ps_enable) { - /* In Long Range Mode, the windowLog is typically oversized. - * In which case, it's preferable to determine the jobSize - * based on chainLog instead. - * Then, ovLog becomes a fraction of the jobSize, rather than windowSize */ - ovLog = MIN(params->cParams.windowLog, ZSTDMT_computeTargetJobLog(params) - 2) - - overlapRLog; - } - assert(0 <= ovLog && ovLog <= ZSTD_WINDOWLOG_MAX); - DEBUGLOG(4, "overlapLog : %i", params->overlapLog); - DEBUGLOG(4, "overlap size : %i", 1 << ovLog); - return (ovLog==0) ? 0 : (size_t)1 << ovLog; -} - -/* ====================================== */ -/* ======= Streaming API ======= */ -/* ====================================== */ - -size_t ZSTDMT_initCStream_internal( - ZSTDMT_CCtx* mtctx, - const void* dict, size_t dictSize, ZSTD_dictContentType_e dictContentType, - const ZSTD_CDict* cdict, ZSTD_CCtx_params params, - unsigned long long pledgedSrcSize) -{ - DEBUGLOG(4, "ZSTDMT_initCStream_internal (pledgedSrcSize=%u, nbWorkers=%u, cctxPool=%u)", - (U32)pledgedSrcSize, params.nbWorkers, mtctx->cctxPool->totalCCtx); - - /* params supposed partially fully validated at this point */ - assert(!ZSTD_isError(ZSTD_checkCParams(params.cParams))); - assert(!((dict) && (cdict))); /* either dict or cdict, not both */ - - /* init */ - if (params.nbWorkers != mtctx->params.nbWorkers) - FORWARD_IF_ERROR( ZSTDMT_resize(mtctx, params.nbWorkers) , ""); - - if (params.jobSize != 0 && params.jobSize < ZSTDMT_JOBSIZE_MIN) params.jobSize = ZSTDMT_JOBSIZE_MIN; - if (params.jobSize > (size_t)ZSTDMT_JOBSIZE_MAX) params.jobSize = (size_t)ZSTDMT_JOBSIZE_MAX; - - DEBUGLOG(4, "ZSTDMT_initCStream_internal: %u workers", params.nbWorkers); - - if (mtctx->allJobsCompleted == 0) { /* previous compression not correctly finished */ - ZSTDMT_waitForAllJobsCompleted(mtctx); - ZSTDMT_releaseAllJobResources(mtctx); - mtctx->allJobsCompleted = 1; - } - - mtctx->params = params; - mtctx->frameContentSize = pledgedSrcSize; - if (dict) { - ZSTD_freeCDict(mtctx->cdictLocal); - mtctx->cdictLocal = ZSTD_createCDict_advanced(dict, dictSize, - ZSTD_dlm_byCopy, dictContentType, /* note : a loadPrefix becomes an internal CDict */ - params.cParams, mtctx->cMem); - mtctx->cdict = mtctx->cdictLocal; - if (mtctx->cdictLocal == NULL) return ERROR(memory_allocation); - } else { - ZSTD_freeCDict(mtctx->cdictLocal); - mtctx->cdictLocal = NULL; - mtctx->cdict = cdict; - } - - mtctx->targetPrefixSize = ZSTDMT_computeOverlapSize(¶ms); - DEBUGLOG(4, "overlapLog=%i => %u KB", params.overlapLog, (U32)(mtctx->targetPrefixSize>>10)); - mtctx->targetSectionSize = params.jobSize; - if (mtctx->targetSectionSize == 0) { - mtctx->targetSectionSize = 1ULL << ZSTDMT_computeTargetJobLog(¶ms); - } - assert(mtctx->targetSectionSize <= (size_t)ZSTDMT_JOBSIZE_MAX); - - if (params.rsyncable) { - /* Aim for the targetsectionSize as the average job size. */ - U32 const jobSizeKB = (U32)(mtctx->targetSectionSize >> 10); - U32 const rsyncBits = (assert(jobSizeKB >= 1), ZSTD_highbit32(jobSizeKB) + 10); - /* We refuse to create jobs < RSYNC_MIN_BLOCK_SIZE bytes, so make sure our - * expected job size is at least 4x larger. */ - assert(rsyncBits >= RSYNC_MIN_BLOCK_LOG + 2); - DEBUGLOG(4, "rsyncLog = %u", rsyncBits); - mtctx->rsync.hash = 0; - mtctx->rsync.hitMask = (1ULL << rsyncBits) - 1; - mtctx->rsync.primePower = ZSTD_rollingHash_primePower(RSYNC_LENGTH); - } - if (mtctx->targetSectionSize < mtctx->targetPrefixSize) mtctx->targetSectionSize = mtctx->targetPrefixSize; /* job size must be >= overlap size */ - DEBUGLOG(4, "Job Size : %u KB (note : set to %u)", (U32)(mtctx->targetSectionSize>>10), (U32)params.jobSize); - DEBUGLOG(4, "inBuff Size : %u KB", (U32)(mtctx->targetSectionSize>>10)); - ZSTDMT_setBufferSize(mtctx->bufPool, ZSTD_compressBound(mtctx->targetSectionSize)); - { - /* If ldm is enabled we need windowSize space. */ - size_t const windowSize = mtctx->params.ldmParams.enableLdm == ZSTD_ps_enable ? (1U << mtctx->params.cParams.windowLog) : 0; - /* Two buffers of slack, plus extra space for the overlap - * This is the minimum slack that LDM works with. One extra because - * flush might waste up to targetSectionSize-1 bytes. Another extra - * for the overlap (if > 0), then one to fill which doesn't overlap - * with the LDM window. - */ - size_t const nbSlackBuffers = 2 + (mtctx->targetPrefixSize > 0); - size_t const slackSize = mtctx->targetSectionSize * nbSlackBuffers; - /* Compute the total size, and always have enough slack */ - size_t const nbWorkers = MAX(mtctx->params.nbWorkers, 1); - size_t const sectionsSize = mtctx->targetSectionSize * nbWorkers; - size_t const capacity = MAX(windowSize, sectionsSize) + slackSize; - if (mtctx->roundBuff.capacity < capacity) { - if (mtctx->roundBuff.buffer) - ZSTD_customFree(mtctx->roundBuff.buffer, mtctx->cMem); - mtctx->roundBuff.buffer = (BYTE*)ZSTD_customMalloc(capacity, mtctx->cMem); - if (mtctx->roundBuff.buffer == NULL) { - mtctx->roundBuff.capacity = 0; - return ERROR(memory_allocation); - } - mtctx->roundBuff.capacity = capacity; - } - } - DEBUGLOG(4, "roundBuff capacity : %u KB", (U32)(mtctx->roundBuff.capacity>>10)); - mtctx->roundBuff.pos = 0; - mtctx->inBuff.buffer = g_nullBuffer; - mtctx->inBuff.filled = 0; - mtctx->inBuff.prefix = kNullRange; - mtctx->doneJobID = 0; - mtctx->nextJobID = 0; - mtctx->frameEnded = 0; - mtctx->allJobsCompleted = 0; - mtctx->consumed = 0; - mtctx->produced = 0; - if (ZSTDMT_serialState_reset(&mtctx->serial, mtctx->seqPool, params, mtctx->targetSectionSize, - dict, dictSize, dictContentType)) - return ERROR(memory_allocation); - return 0; -} - - -/* ZSTDMT_writeLastEmptyBlock() - * Write a single empty block with an end-of-frame to finish a frame. - * Job must be created from streaming variant. - * This function is always successful if expected conditions are fulfilled. - */ -static void ZSTDMT_writeLastEmptyBlock(ZSTDMT_jobDescription* job) -{ - assert(job->lastJob == 1); - assert(job->src.size == 0); /* last job is empty -> will be simplified into a last empty block */ - assert(job->firstJob == 0); /* cannot be first job, as it also needs to create frame header */ - assert(job->dstBuff.start == NULL); /* invoked from streaming variant only (otherwise, dstBuff might be user's output) */ - job->dstBuff = ZSTDMT_getBuffer(job->bufPool); - if (job->dstBuff.start == NULL) { - job->cSize = ERROR(memory_allocation); - return; - } - assert(job->dstBuff.capacity >= ZSTD_blockHeaderSize); /* no buffer should ever be that small */ - job->src = kNullRange; - job->cSize = ZSTD_writeLastEmptyBlock(job->dstBuff.start, job->dstBuff.capacity); - assert(!ZSTD_isError(job->cSize)); - assert(job->consumed == 0); -} - -static size_t ZSTDMT_createCompressionJob(ZSTDMT_CCtx* mtctx, size_t srcSize, ZSTD_EndDirective endOp) -{ - unsigned const jobID = mtctx->nextJobID & mtctx->jobIDMask; - int const endFrame = (endOp == ZSTD_e_end); - - if (mtctx->nextJobID > mtctx->doneJobID + mtctx->jobIDMask) { - DEBUGLOG(5, "ZSTDMT_createCompressionJob: will not create new job : table is full"); - assert((mtctx->nextJobID & mtctx->jobIDMask) == (mtctx->doneJobID & mtctx->jobIDMask)); - return 0; - } - - if (!mtctx->jobReady) { - BYTE const* src = (BYTE const*)mtctx->inBuff.buffer.start; - DEBUGLOG(5, "ZSTDMT_createCompressionJob: preparing job %u to compress %u bytes with %u preload ", - mtctx->nextJobID, (U32)srcSize, (U32)mtctx->inBuff.prefix.size); - mtctx->jobs[jobID].src.start = src; - mtctx->jobs[jobID].src.size = srcSize; - assert(mtctx->inBuff.filled >= srcSize); - mtctx->jobs[jobID].prefix = mtctx->inBuff.prefix; - mtctx->jobs[jobID].consumed = 0; - mtctx->jobs[jobID].cSize = 0; - mtctx->jobs[jobID].params = mtctx->params; - mtctx->jobs[jobID].cdict = mtctx->nextJobID==0 ? mtctx->cdict : NULL; - mtctx->jobs[jobID].fullFrameSize = mtctx->frameContentSize; - mtctx->jobs[jobID].dstBuff = g_nullBuffer; - mtctx->jobs[jobID].cctxPool = mtctx->cctxPool; - mtctx->jobs[jobID].bufPool = mtctx->bufPool; - mtctx->jobs[jobID].seqPool = mtctx->seqPool; - mtctx->jobs[jobID].serial = &mtctx->serial; - mtctx->jobs[jobID].jobID = mtctx->nextJobID; - mtctx->jobs[jobID].firstJob = (mtctx->nextJobID==0); - mtctx->jobs[jobID].lastJob = endFrame; - mtctx->jobs[jobID].frameChecksumNeeded = mtctx->params.fParams.checksumFlag && endFrame && (mtctx->nextJobID>0); - mtctx->jobs[jobID].dstFlushed = 0; - - /* Update the round buffer pos and clear the input buffer to be reset */ - mtctx->roundBuff.pos += srcSize; - mtctx->inBuff.buffer = g_nullBuffer; - mtctx->inBuff.filled = 0; - /* Set the prefix */ - if (!endFrame) { - size_t const newPrefixSize = MIN(srcSize, mtctx->targetPrefixSize); - mtctx->inBuff.prefix.start = src + srcSize - newPrefixSize; - mtctx->inBuff.prefix.size = newPrefixSize; - } else { /* endFrame==1 => no need for another input buffer */ - mtctx->inBuff.prefix = kNullRange; - mtctx->frameEnded = endFrame; - if (mtctx->nextJobID == 0) { - /* single job exception : checksum is already calculated directly within worker thread */ - mtctx->params.fParams.checksumFlag = 0; - } } - - if ( (srcSize == 0) - && (mtctx->nextJobID>0)/*single job must also write frame header*/ ) { - DEBUGLOG(5, "ZSTDMT_createCompressionJob: creating a last empty block to end frame"); - assert(endOp == ZSTD_e_end); /* only possible case : need to end the frame with an empty last block */ - ZSTDMT_writeLastEmptyBlock(mtctx->jobs + jobID); - mtctx->nextJobID++; - return 0; - } - } - - DEBUGLOG(5, "ZSTDMT_createCompressionJob: posting job %u : %u bytes (end:%u, jobNb == %u (mod:%u))", - mtctx->nextJobID, - (U32)mtctx->jobs[jobID].src.size, - mtctx->jobs[jobID].lastJob, - mtctx->nextJobID, - jobID); - if (POOL_tryAdd(mtctx->factory, ZSTDMT_compressionJob, &mtctx->jobs[jobID])) { - mtctx->nextJobID++; - mtctx->jobReady = 0; - } else { - DEBUGLOG(5, "ZSTDMT_createCompressionJob: no worker available for job %u", mtctx->nextJobID); - mtctx->jobReady = 1; - } - return 0; -} - - -/*! ZSTDMT_flushProduced() : - * flush whatever data has been produced but not yet flushed in current job. - * move to next job if current one is fully flushed. - * `output` : `pos` will be updated with amount of data flushed . - * `blockToFlush` : if >0, the function will block and wait if there is no data available to flush . - * @return : amount of data remaining within internal buffer, 0 if no more, 1 if unknown but > 0, or an error code */ -static size_t ZSTDMT_flushProduced(ZSTDMT_CCtx* mtctx, ZSTD_outBuffer* output, unsigned blockToFlush, ZSTD_EndDirective end) -{ - unsigned const wJobID = mtctx->doneJobID & mtctx->jobIDMask; - DEBUGLOG(5, "ZSTDMT_flushProduced (blocking:%u , job %u <= %u)", - blockToFlush, mtctx->doneJobID, mtctx->nextJobID); - assert(output->size >= output->pos); - - ZSTD_PTHREAD_MUTEX_LOCK(&mtctx->jobs[wJobID].job_mutex); - if ( blockToFlush - && (mtctx->doneJobID < mtctx->nextJobID) ) { - assert(mtctx->jobs[wJobID].dstFlushed <= mtctx->jobs[wJobID].cSize); - while (mtctx->jobs[wJobID].dstFlushed == mtctx->jobs[wJobID].cSize) { /* nothing to flush */ - if (mtctx->jobs[wJobID].consumed == mtctx->jobs[wJobID].src.size) { - DEBUGLOG(5, "job %u is completely consumed (%u == %u) => don't wait for cond, there will be none", - mtctx->doneJobID, (U32)mtctx->jobs[wJobID].consumed, (U32)mtctx->jobs[wJobID].src.size); - break; - } - DEBUGLOG(5, "waiting for something to flush from job %u (currently flushed: %u bytes)", - mtctx->doneJobID, (U32)mtctx->jobs[wJobID].dstFlushed); - ZSTD_pthread_cond_wait(&mtctx->jobs[wJobID].job_cond, &mtctx->jobs[wJobID].job_mutex); /* block when nothing to flush but some to come */ - } } - - /* try to flush something */ - { size_t cSize = mtctx->jobs[wJobID].cSize; /* shared */ - size_t const srcConsumed = mtctx->jobs[wJobID].consumed; /* shared */ - size_t const srcSize = mtctx->jobs[wJobID].src.size; /* read-only, could be done after mutex lock, but no-declaration-after-statement */ - ZSTD_pthread_mutex_unlock(&mtctx->jobs[wJobID].job_mutex); - if (ZSTD_isError(cSize)) { - DEBUGLOG(5, "ZSTDMT_flushProduced: job %u : compression error detected : %s", - mtctx->doneJobID, ZSTD_getErrorName(cSize)); - ZSTDMT_waitForAllJobsCompleted(mtctx); - ZSTDMT_releaseAllJobResources(mtctx); - return cSize; - } - /* add frame checksum if necessary (can only happen once) */ - assert(srcConsumed <= srcSize); - if ( (srcConsumed == srcSize) /* job completed -> worker no longer active */ - && mtctx->jobs[wJobID].frameChecksumNeeded ) { - U32 const checksum = (U32)XXH64_digest(&mtctx->serial.xxhState); - DEBUGLOG(4, "ZSTDMT_flushProduced: writing checksum : %08X \n", checksum); - MEM_writeLE32((char*)mtctx->jobs[wJobID].dstBuff.start + mtctx->jobs[wJobID].cSize, checksum); - cSize += 4; - mtctx->jobs[wJobID].cSize += 4; /* can write this shared value, as worker is no longer active */ - mtctx->jobs[wJobID].frameChecksumNeeded = 0; - } - - if (cSize > 0) { /* compression is ongoing or completed */ - size_t const toFlush = MIN(cSize - mtctx->jobs[wJobID].dstFlushed, output->size - output->pos); - DEBUGLOG(5, "ZSTDMT_flushProduced: Flushing %u bytes from job %u (completion:%u/%u, generated:%u)", - (U32)toFlush, mtctx->doneJobID, (U32)srcConsumed, (U32)srcSize, (U32)cSize); - assert(mtctx->doneJobID < mtctx->nextJobID); - assert(cSize >= mtctx->jobs[wJobID].dstFlushed); - assert(mtctx->jobs[wJobID].dstBuff.start != NULL); - if (toFlush > 0) { - ZSTD_memcpy((char*)output->dst + output->pos, - (const char*)mtctx->jobs[wJobID].dstBuff.start + mtctx->jobs[wJobID].dstFlushed, - toFlush); - } - output->pos += toFlush; - mtctx->jobs[wJobID].dstFlushed += toFlush; /* can write : this value is only used by mtctx */ - - if ( (srcConsumed == srcSize) /* job is completed */ - && (mtctx->jobs[wJobID].dstFlushed == cSize) ) { /* output buffer fully flushed => free this job position */ - DEBUGLOG(5, "Job %u completed (%u bytes), moving to next one", - mtctx->doneJobID, (U32)mtctx->jobs[wJobID].dstFlushed); - ZSTDMT_releaseBuffer(mtctx->bufPool, mtctx->jobs[wJobID].dstBuff); - DEBUGLOG(5, "dstBuffer released"); - mtctx->jobs[wJobID].dstBuff = g_nullBuffer; - mtctx->jobs[wJobID].cSize = 0; /* ensure this job slot is considered "not started" in future check */ - mtctx->consumed += srcSize; - mtctx->produced += cSize; - mtctx->doneJobID++; - } } - - /* return value : how many bytes left in buffer ; fake it to 1 when unknown but >0 */ - if (cSize > mtctx->jobs[wJobID].dstFlushed) return (cSize - mtctx->jobs[wJobID].dstFlushed); - if (srcSize > srcConsumed) return 1; /* current job not completely compressed */ - } - if (mtctx->doneJobID < mtctx->nextJobID) return 1; /* some more jobs ongoing */ - if (mtctx->jobReady) return 1; /* one job is ready to push, just not yet in the list */ - if (mtctx->inBuff.filled > 0) return 1; /* input is not empty, and still needs to be converted into a job */ - mtctx->allJobsCompleted = mtctx->frameEnded; /* all jobs are entirely flushed => if this one is last one, frame is completed */ - if (end == ZSTD_e_end) return !mtctx->frameEnded; /* for ZSTD_e_end, question becomes : is frame completed ? instead of : are internal buffers fully flushed ? */ - return 0; /* internal buffers fully flushed */ -} - -/** - * Returns the range of data used by the earliest job that is not yet complete. - * If the data of the first job is broken up into two segments, we cover both - * sections. - */ -static range_t ZSTDMT_getInputDataInUse(ZSTDMT_CCtx* mtctx) -{ - unsigned const firstJobID = mtctx->doneJobID; - unsigned const lastJobID = mtctx->nextJobID; - unsigned jobID; - - for (jobID = firstJobID; jobID < lastJobID; ++jobID) { - unsigned const wJobID = jobID & mtctx->jobIDMask; - size_t consumed; - - ZSTD_PTHREAD_MUTEX_LOCK(&mtctx->jobs[wJobID].job_mutex); - consumed = mtctx->jobs[wJobID].consumed; - ZSTD_pthread_mutex_unlock(&mtctx->jobs[wJobID].job_mutex); - - if (consumed < mtctx->jobs[wJobID].src.size) { - range_t range = mtctx->jobs[wJobID].prefix; - if (range.size == 0) { - /* Empty prefix */ - range = mtctx->jobs[wJobID].src; - } - /* Job source in multiple segments not supported yet */ - assert(range.start <= mtctx->jobs[wJobID].src.start); - return range; - } - } - return kNullRange; -} - -/** - * Returns non-zero iff buffer and range overlap. - */ -static int ZSTDMT_isOverlapped(buffer_t buffer, range_t range) -{ - BYTE const* const bufferStart = (BYTE const*)buffer.start; - BYTE const* const rangeStart = (BYTE const*)range.start; - - if (rangeStart == NULL || bufferStart == NULL) - return 0; - - { - BYTE const* const bufferEnd = bufferStart + buffer.capacity; - BYTE const* const rangeEnd = rangeStart + range.size; - - /* Empty ranges cannot overlap */ - if (bufferStart == bufferEnd || rangeStart == rangeEnd) - return 0; - - return bufferStart < rangeEnd && rangeStart < bufferEnd; - } -} - -static int ZSTDMT_doesOverlapWindow(buffer_t buffer, ZSTD_window_t window) -{ - range_t extDict; - range_t prefix; - - DEBUGLOG(5, "ZSTDMT_doesOverlapWindow"); - extDict.start = window.dictBase + window.lowLimit; - extDict.size = window.dictLimit - window.lowLimit; - - prefix.start = window.base + window.dictLimit; - prefix.size = window.nextSrc - (window.base + window.dictLimit); - DEBUGLOG(5, "extDict [0x%zx, 0x%zx)", - (size_t)extDict.start, - (size_t)extDict.start + extDict.size); - DEBUGLOG(5, "prefix [0x%zx, 0x%zx)", - (size_t)prefix.start, - (size_t)prefix.start + prefix.size); - - return ZSTDMT_isOverlapped(buffer, extDict) - || ZSTDMT_isOverlapped(buffer, prefix); -} - -static void ZSTDMT_waitForLdmComplete(ZSTDMT_CCtx* mtctx, buffer_t buffer) -{ - if (mtctx->params.ldmParams.enableLdm == ZSTD_ps_enable) { - ZSTD_pthread_mutex_t* mutex = &mtctx->serial.ldmWindowMutex; - DEBUGLOG(5, "ZSTDMT_waitForLdmComplete"); - DEBUGLOG(5, "source [0x%zx, 0x%zx)", - (size_t)buffer.start, - (size_t)buffer.start + buffer.capacity); - ZSTD_PTHREAD_MUTEX_LOCK(mutex); - while (ZSTDMT_doesOverlapWindow(buffer, mtctx->serial.ldmWindow)) { - DEBUGLOG(5, "Waiting for LDM to finish..."); - ZSTD_pthread_cond_wait(&mtctx->serial.ldmWindowCond, mutex); - } - DEBUGLOG(6, "Done waiting for LDM to finish"); - ZSTD_pthread_mutex_unlock(mutex); - } -} - -/** - * Attempts to set the inBuff to the next section to fill. - * If any part of the new section is still in use we give up. - * Returns non-zero if the buffer is filled. - */ -static int ZSTDMT_tryGetInputRange(ZSTDMT_CCtx* mtctx) -{ - range_t const inUse = ZSTDMT_getInputDataInUse(mtctx); - size_t const spaceLeft = mtctx->roundBuff.capacity - mtctx->roundBuff.pos; - size_t const target = mtctx->targetSectionSize; - buffer_t buffer; - - DEBUGLOG(5, "ZSTDMT_tryGetInputRange"); - assert(mtctx->inBuff.buffer.start == NULL); - assert(mtctx->roundBuff.capacity >= target); - - if (spaceLeft < target) { - /* ZSTD_invalidateRepCodes() doesn't work for extDict variants. - * Simply copy the prefix to the beginning in that case. - */ - BYTE* const start = (BYTE*)mtctx->roundBuff.buffer; - size_t const prefixSize = mtctx->inBuff.prefix.size; - - buffer.start = start; - buffer.capacity = prefixSize; - if (ZSTDMT_isOverlapped(buffer, inUse)) { - DEBUGLOG(5, "Waiting for buffer..."); - return 0; - } - ZSTDMT_waitForLdmComplete(mtctx, buffer); - ZSTD_memmove(start, mtctx->inBuff.prefix.start, prefixSize); - mtctx->inBuff.prefix.start = start; - mtctx->roundBuff.pos = prefixSize; - } - buffer.start = mtctx->roundBuff.buffer + mtctx->roundBuff.pos; - buffer.capacity = target; - - if (ZSTDMT_isOverlapped(buffer, inUse)) { - DEBUGLOG(5, "Waiting for buffer..."); - return 0; - } - assert(!ZSTDMT_isOverlapped(buffer, mtctx->inBuff.prefix)); - - ZSTDMT_waitForLdmComplete(mtctx, buffer); - - DEBUGLOG(5, "Using prefix range [%zx, %zx)", - (size_t)mtctx->inBuff.prefix.start, - (size_t)mtctx->inBuff.prefix.start + mtctx->inBuff.prefix.size); - DEBUGLOG(5, "Using source range [%zx, %zx)", - (size_t)buffer.start, - (size_t)buffer.start + buffer.capacity); - - - mtctx->inBuff.buffer = buffer; - mtctx->inBuff.filled = 0; - assert(mtctx->roundBuff.pos + buffer.capacity <= mtctx->roundBuff.capacity); - return 1; -} - -typedef struct { - size_t toLoad; /* The number of bytes to load from the input. */ - int flush; /* Boolean declaring if we must flush because we found a synchronization point. */ -} syncPoint_t; - -/** - * Searches through the input for a synchronization point. If one is found, we - * will instruct the caller to flush, and return the number of bytes to load. - * Otherwise, we will load as many bytes as possible and instruct the caller - * to continue as normal. - */ -static syncPoint_t -findSynchronizationPoint(ZSTDMT_CCtx const* mtctx, ZSTD_inBuffer const input) -{ - BYTE const* const istart = (BYTE const*)input.src + input.pos; - U64 const primePower = mtctx->rsync.primePower; - U64 const hitMask = mtctx->rsync.hitMask; - - syncPoint_t syncPoint; - U64 hash; - BYTE const* prev; - size_t pos; - - syncPoint.toLoad = MIN(input.size - input.pos, mtctx->targetSectionSize - mtctx->inBuff.filled); - syncPoint.flush = 0; - if (!mtctx->params.rsyncable) - /* Rsync is disabled. */ - return syncPoint; - if (mtctx->inBuff.filled + input.size - input.pos < RSYNC_MIN_BLOCK_SIZE) - /* We don't emit synchronization points if it would produce too small blocks. - * We don't have enough input to find a synchronization point, so don't look. - */ - return syncPoint; - if (mtctx->inBuff.filled + syncPoint.toLoad < RSYNC_LENGTH) - /* Not enough to compute the hash. - * We will miss any synchronization points in this RSYNC_LENGTH byte - * window. However, since it depends only in the internal buffers, if the - * state is already synchronized, we will remain synchronized. - * Additionally, the probability that we miss a synchronization point is - * low: RSYNC_LENGTH / targetSectionSize. - */ - return syncPoint; - /* Initialize the loop variables. */ - if (mtctx->inBuff.filled < RSYNC_MIN_BLOCK_SIZE) { - /* We don't need to scan the first RSYNC_MIN_BLOCK_SIZE positions - * because they can't possibly be a sync point. So we can start - * part way through the input buffer. - */ - pos = RSYNC_MIN_BLOCK_SIZE - mtctx->inBuff.filled; - if (pos >= RSYNC_LENGTH) { - prev = istart + pos - RSYNC_LENGTH; - hash = ZSTD_rollingHash_compute(prev, RSYNC_LENGTH); - } else { - assert(mtctx->inBuff.filled >= RSYNC_LENGTH); - prev = (BYTE const*)mtctx->inBuff.buffer.start + mtctx->inBuff.filled - RSYNC_LENGTH; - hash = ZSTD_rollingHash_compute(prev + pos, (RSYNC_LENGTH - pos)); - hash = ZSTD_rollingHash_append(hash, istart, pos); - } - } else { - /* We have enough bytes buffered to initialize the hash, - * and are have processed enough bytes to find a sync point. - * Start scanning at the beginning of the input. - */ - assert(mtctx->inBuff.filled >= RSYNC_MIN_BLOCK_SIZE); - assert(RSYNC_MIN_BLOCK_SIZE >= RSYNC_LENGTH); - pos = 0; - prev = (BYTE const*)mtctx->inBuff.buffer.start + mtctx->inBuff.filled - RSYNC_LENGTH; - hash = ZSTD_rollingHash_compute(prev, RSYNC_LENGTH); - if ((hash & hitMask) == hitMask) { - /* We're already at a sync point so don't load any more until - * we're able to flush this sync point. - * This likely happened because the job table was full so we - * couldn't add our job. - */ - syncPoint.toLoad = 0; - syncPoint.flush = 1; - return syncPoint; - } - } - /* Starting with the hash of the previous RSYNC_LENGTH bytes, roll - * through the input. If we hit a synchronization point, then cut the - * job off, and tell the compressor to flush the job. Otherwise, load - * all the bytes and continue as normal. - * If we go too long without a synchronization point (targetSectionSize) - * then a block will be emitted anyways, but this is okay, since if we - * are already synchronized we will remain synchronized. - */ - for (; pos < syncPoint.toLoad; ++pos) { - BYTE const toRemove = pos < RSYNC_LENGTH ? prev[pos] : istart[pos - RSYNC_LENGTH]; - assert(pos < RSYNC_LENGTH || ZSTD_rollingHash_compute(istart + pos - RSYNC_LENGTH, RSYNC_LENGTH) == hash); - hash = ZSTD_rollingHash_rotate(hash, toRemove, istart[pos], primePower); - assert(mtctx->inBuff.filled + pos >= RSYNC_MIN_BLOCK_SIZE); - if ((hash & hitMask) == hitMask) { - syncPoint.toLoad = pos + 1; - syncPoint.flush = 1; - break; - } - } - return syncPoint; -} - -size_t ZSTDMT_nextInputSizeHint(const ZSTDMT_CCtx* mtctx) -{ - size_t hintInSize = mtctx->targetSectionSize - mtctx->inBuff.filled; - if (hintInSize==0) hintInSize = mtctx->targetSectionSize; - return hintInSize; -} - -/** ZSTDMT_compressStream_generic() : - * internal use only - exposed to be invoked from zstd_compress.c - * assumption : output and input are valid (pos <= size) - * @return : minimum amount of data remaining to flush, 0 if none */ -size_t ZSTDMT_compressStream_generic(ZSTDMT_CCtx* mtctx, - ZSTD_outBuffer* output, - ZSTD_inBuffer* input, - ZSTD_EndDirective endOp) -{ - unsigned forwardInputProgress = 0; - DEBUGLOG(5, "ZSTDMT_compressStream_generic (endOp=%u, srcSize=%u)", - (U32)endOp, (U32)(input->size - input->pos)); - assert(output->pos <= output->size); - assert(input->pos <= input->size); - - if ((mtctx->frameEnded) && (endOp==ZSTD_e_continue)) { - /* current frame being ended. Only flush/end are allowed */ - return ERROR(stage_wrong); - } - - /* fill input buffer */ - if ( (!mtctx->jobReady) - && (input->size > input->pos) ) { /* support NULL input */ - if (mtctx->inBuff.buffer.start == NULL) { - assert(mtctx->inBuff.filled == 0); /* Can't fill an empty buffer */ - if (!ZSTDMT_tryGetInputRange(mtctx)) { - /* It is only possible for this operation to fail if there are - * still compression jobs ongoing. - */ - DEBUGLOG(5, "ZSTDMT_tryGetInputRange failed"); - assert(mtctx->doneJobID != mtctx->nextJobID); - } else - DEBUGLOG(5, "ZSTDMT_tryGetInputRange completed successfully : mtctx->inBuff.buffer.start = %p", mtctx->inBuff.buffer.start); - } - if (mtctx->inBuff.buffer.start != NULL) { - syncPoint_t const syncPoint = findSynchronizationPoint(mtctx, *input); - if (syncPoint.flush && endOp == ZSTD_e_continue) { - endOp = ZSTD_e_flush; - } - assert(mtctx->inBuff.buffer.capacity >= mtctx->targetSectionSize); - DEBUGLOG(5, "ZSTDMT_compressStream_generic: adding %u bytes on top of %u to buffer of size %u", - (U32)syncPoint.toLoad, (U32)mtctx->inBuff.filled, (U32)mtctx->targetSectionSize); - ZSTD_memcpy((char*)mtctx->inBuff.buffer.start + mtctx->inBuff.filled, (const char*)input->src + input->pos, syncPoint.toLoad); - input->pos += syncPoint.toLoad; - mtctx->inBuff.filled += syncPoint.toLoad; - forwardInputProgress = syncPoint.toLoad>0; - } - } - if ((input->pos < input->size) && (endOp == ZSTD_e_end)) { - /* Can't end yet because the input is not fully consumed. - * We are in one of these cases: - * - mtctx->inBuff is NULL & empty: we couldn't get an input buffer so don't create a new job. - * - We filled the input buffer: flush this job but don't end the frame. - * - We hit a synchronization point: flush this job but don't end the frame. - */ - assert(mtctx->inBuff.filled == 0 || mtctx->inBuff.filled == mtctx->targetSectionSize || mtctx->params.rsyncable); - endOp = ZSTD_e_flush; - } - - if ( (mtctx->jobReady) - || (mtctx->inBuff.filled >= mtctx->targetSectionSize) /* filled enough : let's compress */ - || ((endOp != ZSTD_e_continue) && (mtctx->inBuff.filled > 0)) /* something to flush : let's go */ - || ((endOp == ZSTD_e_end) && (!mtctx->frameEnded)) ) { /* must finish the frame with a zero-size block */ - size_t const jobSize = mtctx->inBuff.filled; - assert(mtctx->inBuff.filled <= mtctx->targetSectionSize); - FORWARD_IF_ERROR( ZSTDMT_createCompressionJob(mtctx, jobSize, endOp) , ""); - } - - /* check for potential compressed data ready to be flushed */ - { size_t const remainingToFlush = ZSTDMT_flushProduced(mtctx, output, !forwardInputProgress, endOp); /* block if there was no forward input progress */ - if (input->pos < input->size) return MAX(remainingToFlush, 1); /* input not consumed : do not end flush yet */ - DEBUGLOG(5, "end of ZSTDMT_compressStream_generic: remainingToFlush = %u", (U32)remainingToFlush); - return remainingToFlush; - } -} diff --git a/dep/zstd/lib/compress/zstdmt_compress.h b/dep/zstd/lib/compress/zstdmt_compress.h deleted file mode 100644 index 271eb1ac7..000000000 --- a/dep/zstd/lib/compress/zstdmt_compress.h +++ /dev/null @@ -1,113 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - #ifndef ZSTDMT_COMPRESS_H - #define ZSTDMT_COMPRESS_H - - #if defined (__cplusplus) - extern "C" { - #endif - - -/* Note : This is an internal API. - * These APIs used to be exposed with ZSTDLIB_API, - * because it used to be the only way to invoke MT compression. - * Now, you must use ZSTD_compress2 and ZSTD_compressStream2() instead. - * - * This API requires ZSTD_MULTITHREAD to be defined during compilation, - * otherwise ZSTDMT_createCCtx*() will fail. - */ - -/* === Dependencies === */ -#include "../common/zstd_deps.h" /* size_t */ -#define ZSTD_STATIC_LINKING_ONLY /* ZSTD_parameters */ -#include "../zstd.h" /* ZSTD_inBuffer, ZSTD_outBuffer, ZSTDLIB_API */ - - -/* === Constants === */ -#ifndef ZSTDMT_NBWORKERS_MAX /* a different value can be selected at compile time */ -# define ZSTDMT_NBWORKERS_MAX ((sizeof(void*)==4) /*32-bit*/ ? 64 : 256) -#endif -#ifndef ZSTDMT_JOBSIZE_MIN /* a different value can be selected at compile time */ -# define ZSTDMT_JOBSIZE_MIN (512 KB) -#endif -#define ZSTDMT_JOBLOG_MAX (MEM_32bits() ? 29 : 30) -#define ZSTDMT_JOBSIZE_MAX (MEM_32bits() ? (512 MB) : (1024 MB)) - - -/* ======================================================== - * === Private interface, for use by ZSTD_compress.c === - * === Not exposed in libzstd. Never invoke directly === - * ======================================================== */ - -/* === Memory management === */ -typedef struct ZSTDMT_CCtx_s ZSTDMT_CCtx; -/* Requires ZSTD_MULTITHREAD to be defined during compilation, otherwise it will return NULL. */ -ZSTDMT_CCtx* ZSTDMT_createCCtx_advanced(unsigned nbWorkers, - ZSTD_customMem cMem, - ZSTD_threadPool *pool); -size_t ZSTDMT_freeCCtx(ZSTDMT_CCtx* mtctx); - -size_t ZSTDMT_sizeof_CCtx(ZSTDMT_CCtx* mtctx); - -/* === Streaming functions === */ - -size_t ZSTDMT_nextInputSizeHint(const ZSTDMT_CCtx* mtctx); - -/*! ZSTDMT_initCStream_internal() : - * Private use only. Init streaming operation. - * expects params to be valid. - * must receive dict, or cdict, or none, but not both. - * mtctx can be freshly constructed or reused from a prior compression. - * If mtctx is reused, memory allocations from the prior compression may not be freed, - * even if they are not needed for the current compression. - * @return : 0, or an error code */ -size_t ZSTDMT_initCStream_internal(ZSTDMT_CCtx* mtctx, - const void* dict, size_t dictSize, ZSTD_dictContentType_e dictContentType, - const ZSTD_CDict* cdict, - ZSTD_CCtx_params params, unsigned long long pledgedSrcSize); - -/*! ZSTDMT_compressStream_generic() : - * Combines ZSTDMT_compressStream() with optional ZSTDMT_flushStream() or ZSTDMT_endStream() - * depending on flush directive. - * @return : minimum amount of data still to be flushed - * 0 if fully flushed - * or an error code - * note : needs to be init using any ZSTD_initCStream*() variant */ -size_t ZSTDMT_compressStream_generic(ZSTDMT_CCtx* mtctx, - ZSTD_outBuffer* output, - ZSTD_inBuffer* input, - ZSTD_EndDirective endOp); - - /*! ZSTDMT_toFlushNow() - * Tell how many bytes are ready to be flushed immediately. - * Probe the oldest active job (not yet entirely flushed) and check its output buffer. - * If return 0, it means there is no active job, - * or, it means oldest job is still active, but everything produced has been flushed so far, - * therefore flushing is limited by speed of oldest job. */ -size_t ZSTDMT_toFlushNow(ZSTDMT_CCtx* mtctx); - -/*! ZSTDMT_updateCParams_whileCompressing() : - * Updates only a selected set of compression parameters, to remain compatible with current frame. - * New parameters will be applied to next compression job. */ -void ZSTDMT_updateCParams_whileCompressing(ZSTDMT_CCtx* mtctx, const ZSTD_CCtx_params* cctxParams); - -/*! ZSTDMT_getFrameProgression(): - * tells how much data has been consumed (input) and produced (output) for current frame. - * able to count progression inside worker threads. - */ -ZSTD_frameProgression ZSTDMT_getFrameProgression(ZSTDMT_CCtx* mtctx); - - -#if defined (__cplusplus) -} -#endif - -#endif /* ZSTDMT_COMPRESS_H */ diff --git a/dep/zstd/lib/decompress/huf_decompress.c b/dep/zstd/lib/decompress/huf_decompress.c deleted file mode 100644 index 202718825..000000000 --- a/dep/zstd/lib/decompress/huf_decompress.c +++ /dev/null @@ -1,1889 +0,0 @@ -/* ****************************************************************** - * huff0 huffman decoder, - * part of Finite State Entropy library - * Copyright (c) Yann Collet, Facebook, Inc. - * - * You can contact the author at : - * - FSE+HUF source repository : https://github.com/Cyan4973/FiniteStateEntropy - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. -****************************************************************** */ - -/* ************************************************************** -* Dependencies -****************************************************************/ -#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memset */ -#include "../common/compiler.h" -#include "../common/bitstream.h" /* BIT_* */ -#include "../common/fse.h" /* to compress headers */ -#define HUF_STATIC_LINKING_ONLY -#include "../common/huf.h" -#include "../common/error_private.h" -#include "../common/zstd_internal.h" - -/* ************************************************************** -* Constants -****************************************************************/ - -#define HUF_DECODER_FAST_TABLELOG 11 - -/* ************************************************************** -* Macros -****************************************************************/ - -/* These two optional macros force the use one way or another of the two - * Huffman decompression implementations. You can't force in both directions - * at the same time. - */ -#if defined(HUF_FORCE_DECOMPRESS_X1) && \ - defined(HUF_FORCE_DECOMPRESS_X2) -#error "Cannot force the use of the X1 and X2 decoders at the same time!" -#endif - -#if ZSTD_ENABLE_ASM_X86_64_BMI2 && DYNAMIC_BMI2 -# define HUF_ASM_X86_64_BMI2_ATTRS BMI2_TARGET_ATTRIBUTE -#else -# define HUF_ASM_X86_64_BMI2_ATTRS -#endif - -#ifdef __cplusplus -# define HUF_EXTERN_C extern "C" -#else -# define HUF_EXTERN_C -#endif -#define HUF_ASM_DECL HUF_EXTERN_C - -#if DYNAMIC_BMI2 || (ZSTD_ENABLE_ASM_X86_64_BMI2 && defined(__BMI2__)) -# define HUF_NEED_BMI2_FUNCTION 1 -#else -# define HUF_NEED_BMI2_FUNCTION 0 -#endif - -#if !(ZSTD_ENABLE_ASM_X86_64_BMI2 && defined(__BMI2__)) -# define HUF_NEED_DEFAULT_FUNCTION 1 -#else -# define HUF_NEED_DEFAULT_FUNCTION 0 -#endif - -/* ************************************************************** -* Error Management -****************************************************************/ -#define HUF_isError ERR_isError - - -/* ************************************************************** -* Byte alignment for workSpace management -****************************************************************/ -#define HUF_ALIGN(x, a) HUF_ALIGN_MASK((x), (a) - 1) -#define HUF_ALIGN_MASK(x, mask) (((x) + (mask)) & ~(mask)) - - -/* ************************************************************** -* BMI2 Variant Wrappers -****************************************************************/ -#if DYNAMIC_BMI2 - -#define HUF_DGEN(fn) \ - \ - static size_t fn##_default( \ - void* dst, size_t dstSize, \ - const void* cSrc, size_t cSrcSize, \ - const HUF_DTable* DTable) \ - { \ - return fn##_body(dst, dstSize, cSrc, cSrcSize, DTable); \ - } \ - \ - static BMI2_TARGET_ATTRIBUTE size_t fn##_bmi2( \ - void* dst, size_t dstSize, \ - const void* cSrc, size_t cSrcSize, \ - const HUF_DTable* DTable) \ - { \ - return fn##_body(dst, dstSize, cSrc, cSrcSize, DTable); \ - } \ - \ - static size_t fn(void* dst, size_t dstSize, void const* cSrc, \ - size_t cSrcSize, HUF_DTable const* DTable, int bmi2) \ - { \ - if (bmi2) { \ - return fn##_bmi2(dst, dstSize, cSrc, cSrcSize, DTable); \ - } \ - return fn##_default(dst, dstSize, cSrc, cSrcSize, DTable); \ - } - -#else - -#define HUF_DGEN(fn) \ - static size_t fn(void* dst, size_t dstSize, void const* cSrc, \ - size_t cSrcSize, HUF_DTable const* DTable, int bmi2) \ - { \ - (void)bmi2; \ - return fn##_body(dst, dstSize, cSrc, cSrcSize, DTable); \ - } - -#endif - - -/*-***************************/ -/* generic DTableDesc */ -/*-***************************/ -typedef struct { BYTE maxTableLog; BYTE tableType; BYTE tableLog; BYTE reserved; } DTableDesc; - -static DTableDesc HUF_getDTableDesc(const HUF_DTable* table) -{ - DTableDesc dtd; - ZSTD_memcpy(&dtd, table, sizeof(dtd)); - return dtd; -} - -#if ZSTD_ENABLE_ASM_X86_64_BMI2 - -static size_t HUF_initDStream(BYTE const* ip) { - BYTE const lastByte = ip[7]; - size_t const bitsConsumed = lastByte ? 8 - BIT_highbit32(lastByte) : 0; - size_t const value = MEM_readLEST(ip) | 1; - assert(bitsConsumed <= 8); - return value << bitsConsumed; -} -typedef struct { - BYTE const* ip[4]; - BYTE* op[4]; - U64 bits[4]; - void const* dt; - BYTE const* ilimit; - BYTE* oend; - BYTE const* iend[4]; -} HUF_DecompressAsmArgs; - -/** - * Initializes args for the asm decoding loop. - * @returns 0 on success - * 1 if the fallback implementation should be used. - * Or an error code on failure. - */ -static size_t HUF_DecompressAsmArgs_init(HUF_DecompressAsmArgs* args, void* dst, size_t dstSize, void const* src, size_t srcSize, const HUF_DTable* DTable) -{ - void const* dt = DTable + 1; - U32 const dtLog = HUF_getDTableDesc(DTable).tableLog; - - const BYTE* const ilimit = (const BYTE*)src + 6 + 8; - - BYTE* const oend = (BYTE*)dst + dstSize; - - /* The following condition is false on x32 platform, - * but HUF_asm is not compatible with this ABI */ - if (!(MEM_isLittleEndian() && !MEM_32bits())) return 1; - - /* strict minimum : jump table + 1 byte per stream */ - if (srcSize < 10) - return ERROR(corruption_detected); - - /* Must have at least 8 bytes per stream because we don't handle initializing smaller bit containers. - * If table log is not correct at this point, fallback to the old decoder. - * On small inputs we don't have enough data to trigger the fast loop, so use the old decoder. - */ - if (dtLog != HUF_DECODER_FAST_TABLELOG) - return 1; - - /* Read the jump table. */ - { - const BYTE* const istart = (const BYTE*)src; - size_t const length1 = MEM_readLE16(istart); - size_t const length2 = MEM_readLE16(istart+2); - size_t const length3 = MEM_readLE16(istart+4); - size_t const length4 = srcSize - (length1 + length2 + length3 + 6); - args->iend[0] = istart + 6; /* jumpTable */ - args->iend[1] = args->iend[0] + length1; - args->iend[2] = args->iend[1] + length2; - args->iend[3] = args->iend[2] + length3; - - /* HUF_initDStream() requires this, and this small of an input - * won't benefit from the ASM loop anyways. - * length1 must be >= 16 so that ip[0] >= ilimit before the loop - * starts. - */ - if (length1 < 16 || length2 < 8 || length3 < 8 || length4 < 8) - return 1; - if (length4 > srcSize) return ERROR(corruption_detected); /* overflow */ - } - /* ip[] contains the position that is currently loaded into bits[]. */ - args->ip[0] = args->iend[1] - sizeof(U64); - args->ip[1] = args->iend[2] - sizeof(U64); - args->ip[2] = args->iend[3] - sizeof(U64); - args->ip[3] = (BYTE const*)src + srcSize - sizeof(U64); - - /* op[] contains the output pointers. */ - args->op[0] = (BYTE*)dst; - args->op[1] = args->op[0] + (dstSize+3)/4; - args->op[2] = args->op[1] + (dstSize+3)/4; - args->op[3] = args->op[2] + (dstSize+3)/4; - - /* No point to call the ASM loop for tiny outputs. */ - if (args->op[3] >= oend) - return 1; - - /* bits[] is the bit container. - * It is read from the MSB down to the LSB. - * It is shifted left as it is read, and zeros are - * shifted in. After the lowest valid bit a 1 is - * set, so that CountTrailingZeros(bits[]) can be used - * to count how many bits we've consumed. - */ - args->bits[0] = HUF_initDStream(args->ip[0]); - args->bits[1] = HUF_initDStream(args->ip[1]); - args->bits[2] = HUF_initDStream(args->ip[2]); - args->bits[3] = HUF_initDStream(args->ip[3]); - - /* If ip[] >= ilimit, it is guaranteed to be safe to - * reload bits[]. It may be beyond its section, but is - * guaranteed to be valid (>= istart). - */ - args->ilimit = ilimit; - - args->oend = oend; - args->dt = dt; - - return 0; -} - -static size_t HUF_initRemainingDStream(BIT_DStream_t* bit, HUF_DecompressAsmArgs const* args, int stream, BYTE* segmentEnd) -{ - /* Validate that we haven't overwritten. */ - if (args->op[stream] > segmentEnd) - return ERROR(corruption_detected); - /* Validate that we haven't read beyond iend[]. - * Note that ip[] may be < iend[] because the MSB is - * the next bit to read, and we may have consumed 100% - * of the stream, so down to iend[i] - 8 is valid. - */ - if (args->ip[stream] < args->iend[stream] - 8) - return ERROR(corruption_detected); - - /* Construct the BIT_DStream_t. */ - bit->bitContainer = MEM_readLE64(args->ip[stream]); - bit->bitsConsumed = ZSTD_countTrailingZeros((size_t)args->bits[stream]); - bit->start = (const char*)args->iend[0]; - bit->limitPtr = bit->start + sizeof(size_t); - bit->ptr = (const char*)args->ip[stream]; - - return 0; -} -#endif - - -#ifndef HUF_FORCE_DECOMPRESS_X2 - -/*-***************************/ -/* single-symbol decoding */ -/*-***************************/ -typedef struct { BYTE nbBits; BYTE byte; } HUF_DEltX1; /* single-symbol decoding */ - -/** - * Packs 4 HUF_DEltX1 structs into a U64. This is used to lay down 4 entries at - * a time. - */ -static U64 HUF_DEltX1_set4(BYTE symbol, BYTE nbBits) { - U64 D4; - if (MEM_isLittleEndian()) { - D4 = (symbol << 8) + nbBits; - } else { - D4 = symbol + (nbBits << 8); - } - D4 *= 0x0001000100010001ULL; - return D4; -} - -/** - * Increase the tableLog to targetTableLog and rescales the stats. - * If tableLog > targetTableLog this is a no-op. - * @returns New tableLog - */ -static U32 HUF_rescaleStats(BYTE* huffWeight, U32* rankVal, U32 nbSymbols, U32 tableLog, U32 targetTableLog) -{ - if (tableLog > targetTableLog) - return tableLog; - if (tableLog < targetTableLog) { - U32 const scale = targetTableLog - tableLog; - U32 s; - /* Increase the weight for all non-zero probability symbols by scale. */ - for (s = 0; s < nbSymbols; ++s) { - huffWeight[s] += (BYTE)((huffWeight[s] == 0) ? 0 : scale); - } - /* Update rankVal to reflect the new weights. - * All weights except 0 get moved to weight + scale. - * Weights [1, scale] are empty. - */ - for (s = targetTableLog; s > scale; --s) { - rankVal[s] = rankVal[s - scale]; - } - for (s = scale; s > 0; --s) { - rankVal[s] = 0; - } - } - return targetTableLog; -} - -typedef struct { - U32 rankVal[HUF_TABLELOG_ABSOLUTEMAX + 1]; - U32 rankStart[HUF_TABLELOG_ABSOLUTEMAX + 1]; - U32 statsWksp[HUF_READ_STATS_WORKSPACE_SIZE_U32]; - BYTE symbols[HUF_SYMBOLVALUE_MAX + 1]; - BYTE huffWeight[HUF_SYMBOLVALUE_MAX + 1]; -} HUF_ReadDTableX1_Workspace; - - -size_t HUF_readDTableX1_wksp(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize) -{ - return HUF_readDTableX1_wksp_bmi2(DTable, src, srcSize, workSpace, wkspSize, /* bmi2 */ 0); -} - -size_t HUF_readDTableX1_wksp_bmi2(HUF_DTable* DTable, const void* src, size_t srcSize, void* workSpace, size_t wkspSize, int bmi2) -{ - U32 tableLog = 0; - U32 nbSymbols = 0; - size_t iSize; - void* const dtPtr = DTable + 1; - HUF_DEltX1* const dt = (HUF_DEltX1*)dtPtr; - HUF_ReadDTableX1_Workspace* wksp = (HUF_ReadDTableX1_Workspace*)workSpace; - - DEBUG_STATIC_ASSERT(HUF_DECOMPRESS_WORKSPACE_SIZE >= sizeof(*wksp)); - if (sizeof(*wksp) > wkspSize) return ERROR(tableLog_tooLarge); - - DEBUG_STATIC_ASSERT(sizeof(DTableDesc) == sizeof(HUF_DTable)); - /* ZSTD_memset(huffWeight, 0, sizeof(huffWeight)); */ /* is not necessary, even though some analyzer complain ... */ - - iSize = HUF_readStats_wksp(wksp->huffWeight, HUF_SYMBOLVALUE_MAX + 1, wksp->rankVal, &nbSymbols, &tableLog, src, srcSize, wksp->statsWksp, sizeof(wksp->statsWksp), bmi2); - if (HUF_isError(iSize)) return iSize; - - - /* Table header */ - { DTableDesc dtd = HUF_getDTableDesc(DTable); - U32 const maxTableLog = dtd.maxTableLog + 1; - U32 const targetTableLog = MIN(maxTableLog, HUF_DECODER_FAST_TABLELOG); - tableLog = HUF_rescaleStats(wksp->huffWeight, wksp->rankVal, nbSymbols, tableLog, targetTableLog); - if (tableLog > (U32)(dtd.maxTableLog+1)) return ERROR(tableLog_tooLarge); /* DTable too small, Huffman tree cannot fit in */ - dtd.tableType = 0; - dtd.tableLog = (BYTE)tableLog; - ZSTD_memcpy(DTable, &dtd, sizeof(dtd)); - } - - /* Compute symbols and rankStart given rankVal: - * - * rankVal already contains the number of values of each weight. - * - * symbols contains the symbols ordered by weight. First are the rankVal[0] - * weight 0 symbols, followed by the rankVal[1] weight 1 symbols, and so on. - * symbols[0] is filled (but unused) to avoid a branch. - * - * rankStart contains the offset where each rank belongs in the DTable. - * rankStart[0] is not filled because there are no entries in the table for - * weight 0. - */ - { - int n; - int nextRankStart = 0; - int const unroll = 4; - int const nLimit = (int)nbSymbols - unroll + 1; - for (n=0; n<(int)tableLog+1; n++) { - U32 const curr = nextRankStart; - nextRankStart += wksp->rankVal[n]; - wksp->rankStart[n] = curr; - } - for (n=0; n < nLimit; n += unroll) { - int u; - for (u=0; u < unroll; ++u) { - size_t const w = wksp->huffWeight[n+u]; - wksp->symbols[wksp->rankStart[w]++] = (BYTE)(n+u); - } - } - for (; n < (int)nbSymbols; ++n) { - size_t const w = wksp->huffWeight[n]; - wksp->symbols[wksp->rankStart[w]++] = (BYTE)n; - } - } - - /* fill DTable - * We fill all entries of each weight in order. - * That way length is a constant for each iteration of the outer loop. - * We can switch based on the length to a different inner loop which is - * optimized for that particular case. - */ - { - U32 w; - int symbol=wksp->rankVal[0]; - int rankStart=0; - for (w=1; wrankVal[w]; - int const length = (1 << w) >> 1; - int uStart = rankStart; - BYTE const nbBits = (BYTE)(tableLog + 1 - w); - int s; - int u; - switch (length) { - case 1: - for (s=0; ssymbols[symbol + s]; - D.nbBits = nbBits; - dt[uStart] = D; - uStart += 1; - } - break; - case 2: - for (s=0; ssymbols[symbol + s]; - D.nbBits = nbBits; - dt[uStart+0] = D; - dt[uStart+1] = D; - uStart += 2; - } - break; - case 4: - for (s=0; ssymbols[symbol + s], nbBits); - MEM_write64(dt + uStart, D4); - uStart += 4; - } - break; - case 8: - for (s=0; ssymbols[symbol + s], nbBits); - MEM_write64(dt + uStart, D4); - MEM_write64(dt + uStart + 4, D4); - uStart += 8; - } - break; - default: - for (s=0; ssymbols[symbol + s], nbBits); - for (u=0; u < length; u += 16) { - MEM_write64(dt + uStart + u + 0, D4); - MEM_write64(dt + uStart + u + 4, D4); - MEM_write64(dt + uStart + u + 8, D4); - MEM_write64(dt + uStart + u + 12, D4); - } - assert(u == length); - uStart += length; - } - break; - } - symbol += symbolCount; - rankStart += symbolCount * length; - } - } - return iSize; -} - -FORCE_INLINE_TEMPLATE BYTE -HUF_decodeSymbolX1(BIT_DStream_t* Dstream, const HUF_DEltX1* dt, const U32 dtLog) -{ - size_t const val = BIT_lookBitsFast(Dstream, dtLog); /* note : dtLog >= 1 */ - BYTE const c = dt[val].byte; - BIT_skipBits(Dstream, dt[val].nbBits); - return c; -} - -#define HUF_DECODE_SYMBOLX1_0(ptr, DStreamPtr) \ - *ptr++ = HUF_decodeSymbolX1(DStreamPtr, dt, dtLog) - -#define HUF_DECODE_SYMBOLX1_1(ptr, DStreamPtr) \ - if (MEM_64bits() || (HUF_TABLELOG_MAX<=12)) \ - HUF_DECODE_SYMBOLX1_0(ptr, DStreamPtr) - -#define HUF_DECODE_SYMBOLX1_2(ptr, DStreamPtr) \ - if (MEM_64bits()) \ - HUF_DECODE_SYMBOLX1_0(ptr, DStreamPtr) - -HINT_INLINE size_t -HUF_decodeStreamX1(BYTE* p, BIT_DStream_t* const bitDPtr, BYTE* const pEnd, const HUF_DEltX1* const dt, const U32 dtLog) -{ - BYTE* const pStart = p; - - /* up to 4 symbols at a time */ - if ((pEnd - p) > 3) { - while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p < pEnd-3)) { - HUF_DECODE_SYMBOLX1_2(p, bitDPtr); - HUF_DECODE_SYMBOLX1_1(p, bitDPtr); - HUF_DECODE_SYMBOLX1_2(p, bitDPtr); - HUF_DECODE_SYMBOLX1_0(p, bitDPtr); - } - } else { - BIT_reloadDStream(bitDPtr); - } - - /* [0-3] symbols remaining */ - if (MEM_32bits()) - while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p < pEnd)) - HUF_DECODE_SYMBOLX1_0(p, bitDPtr); - - /* no more data to retrieve from bitstream, no need to reload */ - while (p < pEnd) - HUF_DECODE_SYMBOLX1_0(p, bitDPtr); - - return pEnd-pStart; -} - -FORCE_INLINE_TEMPLATE size_t -HUF_decompress1X1_usingDTable_internal_body( - void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - const HUF_DTable* DTable) -{ - BYTE* op = (BYTE*)dst; - BYTE* const oend = op + dstSize; - const void* dtPtr = DTable + 1; - const HUF_DEltX1* const dt = (const HUF_DEltX1*)dtPtr; - BIT_DStream_t bitD; - DTableDesc const dtd = HUF_getDTableDesc(DTable); - U32 const dtLog = dtd.tableLog; - - CHECK_F( BIT_initDStream(&bitD, cSrc, cSrcSize) ); - - HUF_decodeStreamX1(op, &bitD, oend, dt, dtLog); - - if (!BIT_endOfDStream(&bitD)) return ERROR(corruption_detected); - - return dstSize; -} - -FORCE_INLINE_TEMPLATE size_t -HUF_decompress4X1_usingDTable_internal_body( - void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - const HUF_DTable* DTable) -{ - /* Check */ - if (cSrcSize < 10) return ERROR(corruption_detected); /* strict minimum : jump table + 1 byte per stream */ - - { const BYTE* const istart = (const BYTE*) cSrc; - BYTE* const ostart = (BYTE*) dst; - BYTE* const oend = ostart + dstSize; - BYTE* const olimit = oend - 3; - const void* const dtPtr = DTable + 1; - const HUF_DEltX1* const dt = (const HUF_DEltX1*)dtPtr; - - /* Init */ - BIT_DStream_t bitD1; - BIT_DStream_t bitD2; - BIT_DStream_t bitD3; - BIT_DStream_t bitD4; - size_t const length1 = MEM_readLE16(istart); - size_t const length2 = MEM_readLE16(istart+2); - size_t const length3 = MEM_readLE16(istart+4); - size_t const length4 = cSrcSize - (length1 + length2 + length3 + 6); - const BYTE* const istart1 = istart + 6; /* jumpTable */ - const BYTE* const istart2 = istart1 + length1; - const BYTE* const istart3 = istart2 + length2; - const BYTE* const istart4 = istart3 + length3; - const size_t segmentSize = (dstSize+3) / 4; - BYTE* const opStart2 = ostart + segmentSize; - BYTE* const opStart3 = opStart2 + segmentSize; - BYTE* const opStart4 = opStart3 + segmentSize; - BYTE* op1 = ostart; - BYTE* op2 = opStart2; - BYTE* op3 = opStart3; - BYTE* op4 = opStart4; - DTableDesc const dtd = HUF_getDTableDesc(DTable); - U32 const dtLog = dtd.tableLog; - U32 endSignal = 1; - - if (length4 > cSrcSize) return ERROR(corruption_detected); /* overflow */ - if (opStart4 > oend) return ERROR(corruption_detected); /* overflow */ - CHECK_F( BIT_initDStream(&bitD1, istart1, length1) ); - CHECK_F( BIT_initDStream(&bitD2, istart2, length2) ); - CHECK_F( BIT_initDStream(&bitD3, istart3, length3) ); - CHECK_F( BIT_initDStream(&bitD4, istart4, length4) ); - - /* up to 16 symbols per loop (4 symbols per stream) in 64-bit mode */ - if ((size_t)(oend - op4) >= sizeof(size_t)) { - for ( ; (endSignal) & (op4 < olimit) ; ) { - HUF_DECODE_SYMBOLX1_2(op1, &bitD1); - HUF_DECODE_SYMBOLX1_2(op2, &bitD2); - HUF_DECODE_SYMBOLX1_2(op3, &bitD3); - HUF_DECODE_SYMBOLX1_2(op4, &bitD4); - HUF_DECODE_SYMBOLX1_1(op1, &bitD1); - HUF_DECODE_SYMBOLX1_1(op2, &bitD2); - HUF_DECODE_SYMBOLX1_1(op3, &bitD3); - HUF_DECODE_SYMBOLX1_1(op4, &bitD4); - HUF_DECODE_SYMBOLX1_2(op1, &bitD1); - HUF_DECODE_SYMBOLX1_2(op2, &bitD2); - HUF_DECODE_SYMBOLX1_2(op3, &bitD3); - HUF_DECODE_SYMBOLX1_2(op4, &bitD4); - HUF_DECODE_SYMBOLX1_0(op1, &bitD1); - HUF_DECODE_SYMBOLX1_0(op2, &bitD2); - HUF_DECODE_SYMBOLX1_0(op3, &bitD3); - HUF_DECODE_SYMBOLX1_0(op4, &bitD4); - endSignal &= BIT_reloadDStreamFast(&bitD1) == BIT_DStream_unfinished; - endSignal &= BIT_reloadDStreamFast(&bitD2) == BIT_DStream_unfinished; - endSignal &= BIT_reloadDStreamFast(&bitD3) == BIT_DStream_unfinished; - endSignal &= BIT_reloadDStreamFast(&bitD4) == BIT_DStream_unfinished; - } - } - - /* check corruption */ - /* note : should not be necessary : op# advance in lock step, and we control op4. - * but curiously, binary generated by gcc 7.2 & 7.3 with -mbmi2 runs faster when >=1 test is present */ - if (op1 > opStart2) return ERROR(corruption_detected); - if (op2 > opStart3) return ERROR(corruption_detected); - if (op3 > opStart4) return ERROR(corruption_detected); - /* note : op4 supposed already verified within main loop */ - - /* finish bitStreams one by one */ - HUF_decodeStreamX1(op1, &bitD1, opStart2, dt, dtLog); - HUF_decodeStreamX1(op2, &bitD2, opStart3, dt, dtLog); - HUF_decodeStreamX1(op3, &bitD3, opStart4, dt, dtLog); - HUF_decodeStreamX1(op4, &bitD4, oend, dt, dtLog); - - /* check */ - { U32 const endCheck = BIT_endOfDStream(&bitD1) & BIT_endOfDStream(&bitD2) & BIT_endOfDStream(&bitD3) & BIT_endOfDStream(&bitD4); - if (!endCheck) return ERROR(corruption_detected); } - - /* decoded size */ - return dstSize; - } -} - -#if HUF_NEED_BMI2_FUNCTION -static BMI2_TARGET_ATTRIBUTE -size_t HUF_decompress4X1_usingDTable_internal_bmi2(void* dst, size_t dstSize, void const* cSrc, - size_t cSrcSize, HUF_DTable const* DTable) { - return HUF_decompress4X1_usingDTable_internal_body(dst, dstSize, cSrc, cSrcSize, DTable); -} -#endif - -#if HUF_NEED_DEFAULT_FUNCTION -static -size_t HUF_decompress4X1_usingDTable_internal_default(void* dst, size_t dstSize, void const* cSrc, - size_t cSrcSize, HUF_DTable const* DTable) { - return HUF_decompress4X1_usingDTable_internal_body(dst, dstSize, cSrc, cSrcSize, DTable); -} -#endif - -#if ZSTD_ENABLE_ASM_X86_64_BMI2 - -HUF_ASM_DECL void HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop(HUF_DecompressAsmArgs* args) ZSTDLIB_HIDDEN; - -static HUF_ASM_X86_64_BMI2_ATTRS -size_t -HUF_decompress4X1_usingDTable_internal_bmi2_asm( - void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - const HUF_DTable* DTable) -{ - void const* dt = DTable + 1; - const BYTE* const iend = (const BYTE*)cSrc + 6; - BYTE* const oend = (BYTE*)dst + dstSize; - HUF_DecompressAsmArgs args; - { - size_t const ret = HUF_DecompressAsmArgs_init(&args, dst, dstSize, cSrc, cSrcSize, DTable); - FORWARD_IF_ERROR(ret, "Failed to init asm args"); - if (ret != 0) - return HUF_decompress4X1_usingDTable_internal_bmi2(dst, dstSize, cSrc, cSrcSize, DTable); - } - - assert(args.ip[0] >= args.ilimit); - HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop(&args); - - /* Our loop guarantees that ip[] >= ilimit and that we haven't - * overwritten any op[]. - */ - assert(args.ip[0] >= iend); - assert(args.ip[1] >= iend); - assert(args.ip[2] >= iend); - assert(args.ip[3] >= iend); - assert(args.op[3] <= oend); - (void)iend; - - /* finish bit streams one by one. */ - { - size_t const segmentSize = (dstSize+3) / 4; - BYTE* segmentEnd = (BYTE*)dst; - int i; - for (i = 0; i < 4; ++i) { - BIT_DStream_t bit; - if (segmentSize <= (size_t)(oend - segmentEnd)) - segmentEnd += segmentSize; - else - segmentEnd = oend; - FORWARD_IF_ERROR(HUF_initRemainingDStream(&bit, &args, i, segmentEnd), "corruption"); - /* Decompress and validate that we've produced exactly the expected length. */ - args.op[i] += HUF_decodeStreamX1(args.op[i], &bit, segmentEnd, (HUF_DEltX1 const*)dt, HUF_DECODER_FAST_TABLELOG); - if (args.op[i] != segmentEnd) return ERROR(corruption_detected); - } - } - - /* decoded size */ - return dstSize; -} -#endif /* ZSTD_ENABLE_ASM_X86_64_BMI2 */ - -typedef size_t (*HUF_decompress_usingDTable_t)(void *dst, size_t dstSize, - const void *cSrc, - size_t cSrcSize, - const HUF_DTable *DTable); - -HUF_DGEN(HUF_decompress1X1_usingDTable_internal) - -static size_t HUF_decompress4X1_usingDTable_internal(void* dst, size_t dstSize, void const* cSrc, - size_t cSrcSize, HUF_DTable const* DTable, int bmi2) -{ -#if DYNAMIC_BMI2 - if (bmi2) { -# if ZSTD_ENABLE_ASM_X86_64_BMI2 - return HUF_decompress4X1_usingDTable_internal_bmi2_asm(dst, dstSize, cSrc, cSrcSize, DTable); -# else - return HUF_decompress4X1_usingDTable_internal_bmi2(dst, dstSize, cSrc, cSrcSize, DTable); -# endif - } -#else - (void)bmi2; -#endif - -#if ZSTD_ENABLE_ASM_X86_64_BMI2 && defined(__BMI2__) - return HUF_decompress4X1_usingDTable_internal_bmi2_asm(dst, dstSize, cSrc, cSrcSize, DTable); -#else - return HUF_decompress4X1_usingDTable_internal_default(dst, dstSize, cSrc, cSrcSize, DTable); -#endif -} - - -size_t HUF_decompress1X1_usingDTable( - void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - const HUF_DTable* DTable) -{ - DTableDesc dtd = HUF_getDTableDesc(DTable); - if (dtd.tableType != 0) return ERROR(GENERIC); - return HUF_decompress1X1_usingDTable_internal(dst, dstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0); -} - -size_t HUF_decompress1X1_DCtx_wksp(HUF_DTable* DCtx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - void* workSpace, size_t wkspSize) -{ - const BYTE* ip = (const BYTE*) cSrc; - - size_t const hSize = HUF_readDTableX1_wksp(DCtx, cSrc, cSrcSize, workSpace, wkspSize); - if (HUF_isError(hSize)) return hSize; - if (hSize >= cSrcSize) return ERROR(srcSize_wrong); - ip += hSize; cSrcSize -= hSize; - - return HUF_decompress1X1_usingDTable_internal(dst, dstSize, ip, cSrcSize, DCtx, /* bmi2 */ 0); -} - - -size_t HUF_decompress4X1_usingDTable( - void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - const HUF_DTable* DTable) -{ - DTableDesc dtd = HUF_getDTableDesc(DTable); - if (dtd.tableType != 0) return ERROR(GENERIC); - return HUF_decompress4X1_usingDTable_internal(dst, dstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0); -} - -static size_t HUF_decompress4X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - void* workSpace, size_t wkspSize, int bmi2) -{ - const BYTE* ip = (const BYTE*) cSrc; - - size_t const hSize = HUF_readDTableX1_wksp_bmi2(dctx, cSrc, cSrcSize, workSpace, wkspSize, bmi2); - if (HUF_isError(hSize)) return hSize; - if (hSize >= cSrcSize) return ERROR(srcSize_wrong); - ip += hSize; cSrcSize -= hSize; - - return HUF_decompress4X1_usingDTable_internal(dst, dstSize, ip, cSrcSize, dctx, bmi2); -} - -size_t HUF_decompress4X1_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - void* workSpace, size_t wkspSize) -{ - return HUF_decompress4X1_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, 0); -} - - -#endif /* HUF_FORCE_DECOMPRESS_X2 */ - - -#ifndef HUF_FORCE_DECOMPRESS_X1 - -/* *************************/ -/* double-symbols decoding */ -/* *************************/ - -typedef struct { U16 sequence; BYTE nbBits; BYTE length; } HUF_DEltX2; /* double-symbols decoding */ -typedef struct { BYTE symbol; } sortedSymbol_t; -typedef U32 rankValCol_t[HUF_TABLELOG_MAX + 1]; -typedef rankValCol_t rankVal_t[HUF_TABLELOG_MAX]; - -/** - * Constructs a HUF_DEltX2 in a U32. - */ -static U32 HUF_buildDEltX2U32(U32 symbol, U32 nbBits, U32 baseSeq, int level) -{ - U32 seq; - DEBUG_STATIC_ASSERT(offsetof(HUF_DEltX2, sequence) == 0); - DEBUG_STATIC_ASSERT(offsetof(HUF_DEltX2, nbBits) == 2); - DEBUG_STATIC_ASSERT(offsetof(HUF_DEltX2, length) == 3); - DEBUG_STATIC_ASSERT(sizeof(HUF_DEltX2) == sizeof(U32)); - if (MEM_isLittleEndian()) { - seq = level == 1 ? symbol : (baseSeq + (symbol << 8)); - return seq + (nbBits << 16) + ((U32)level << 24); - } else { - seq = level == 1 ? (symbol << 8) : ((baseSeq << 8) + symbol); - return (seq << 16) + (nbBits << 8) + (U32)level; - } -} - -/** - * Constructs a HUF_DEltX2. - */ -static HUF_DEltX2 HUF_buildDEltX2(U32 symbol, U32 nbBits, U32 baseSeq, int level) -{ - HUF_DEltX2 DElt; - U32 const val = HUF_buildDEltX2U32(symbol, nbBits, baseSeq, level); - DEBUG_STATIC_ASSERT(sizeof(DElt) == sizeof(val)); - ZSTD_memcpy(&DElt, &val, sizeof(val)); - return DElt; -} - -/** - * Constructs 2 HUF_DEltX2s and packs them into a U64. - */ -static U64 HUF_buildDEltX2U64(U32 symbol, U32 nbBits, U16 baseSeq, int level) -{ - U32 DElt = HUF_buildDEltX2U32(symbol, nbBits, baseSeq, level); - return (U64)DElt + ((U64)DElt << 32); -} - -/** - * Fills the DTable rank with all the symbols from [begin, end) that are each - * nbBits long. - * - * @param DTableRank The start of the rank in the DTable. - * @param begin The first symbol to fill (inclusive). - * @param end The last symbol to fill (exclusive). - * @param nbBits Each symbol is nbBits long. - * @param tableLog The table log. - * @param baseSeq If level == 1 { 0 } else { the first level symbol } - * @param level The level in the table. Must be 1 or 2. - */ -static void HUF_fillDTableX2ForWeight( - HUF_DEltX2* DTableRank, - sortedSymbol_t const* begin, sortedSymbol_t const* end, - U32 nbBits, U32 tableLog, - U16 baseSeq, int const level) -{ - U32 const length = 1U << ((tableLog - nbBits) & 0x1F /* quiet static-analyzer */); - const sortedSymbol_t* ptr; - assert(level >= 1 && level <= 2); - switch (length) { - case 1: - for (ptr = begin; ptr != end; ++ptr) { - HUF_DEltX2 const DElt = HUF_buildDEltX2(ptr->symbol, nbBits, baseSeq, level); - *DTableRank++ = DElt; - } - break; - case 2: - for (ptr = begin; ptr != end; ++ptr) { - HUF_DEltX2 const DElt = HUF_buildDEltX2(ptr->symbol, nbBits, baseSeq, level); - DTableRank[0] = DElt; - DTableRank[1] = DElt; - DTableRank += 2; - } - break; - case 4: - for (ptr = begin; ptr != end; ++ptr) { - U64 const DEltX2 = HUF_buildDEltX2U64(ptr->symbol, nbBits, baseSeq, level); - ZSTD_memcpy(DTableRank + 0, &DEltX2, sizeof(DEltX2)); - ZSTD_memcpy(DTableRank + 2, &DEltX2, sizeof(DEltX2)); - DTableRank += 4; - } - break; - case 8: - for (ptr = begin; ptr != end; ++ptr) { - U64 const DEltX2 = HUF_buildDEltX2U64(ptr->symbol, nbBits, baseSeq, level); - ZSTD_memcpy(DTableRank + 0, &DEltX2, sizeof(DEltX2)); - ZSTD_memcpy(DTableRank + 2, &DEltX2, sizeof(DEltX2)); - ZSTD_memcpy(DTableRank + 4, &DEltX2, sizeof(DEltX2)); - ZSTD_memcpy(DTableRank + 6, &DEltX2, sizeof(DEltX2)); - DTableRank += 8; - } - break; - default: - for (ptr = begin; ptr != end; ++ptr) { - U64 const DEltX2 = HUF_buildDEltX2U64(ptr->symbol, nbBits, baseSeq, level); - HUF_DEltX2* const DTableRankEnd = DTableRank + length; - for (; DTableRank != DTableRankEnd; DTableRank += 8) { - ZSTD_memcpy(DTableRank + 0, &DEltX2, sizeof(DEltX2)); - ZSTD_memcpy(DTableRank + 2, &DEltX2, sizeof(DEltX2)); - ZSTD_memcpy(DTableRank + 4, &DEltX2, sizeof(DEltX2)); - ZSTD_memcpy(DTableRank + 6, &DEltX2, sizeof(DEltX2)); - } - } - break; - } -} - -/* HUF_fillDTableX2Level2() : - * `rankValOrigin` must be a table of at least (HUF_TABLELOG_MAX + 1) U32 */ -static void HUF_fillDTableX2Level2(HUF_DEltX2* DTable, U32 targetLog, const U32 consumedBits, - const U32* rankVal, const int minWeight, const int maxWeight1, - const sortedSymbol_t* sortedSymbols, U32 const* rankStart, - U32 nbBitsBaseline, U16 baseSeq) -{ - /* Fill skipped values (all positions up to rankVal[minWeight]). - * These are positions only get a single symbol because the combined weight - * is too large. - */ - if (minWeight>1) { - U32 const length = 1U << ((targetLog - consumedBits) & 0x1F /* quiet static-analyzer */); - U64 const DEltX2 = HUF_buildDEltX2U64(baseSeq, consumedBits, /* baseSeq */ 0, /* level */ 1); - int const skipSize = rankVal[minWeight]; - assert(length > 1); - assert((U32)skipSize < length); - switch (length) { - case 2: - assert(skipSize == 1); - ZSTD_memcpy(DTable, &DEltX2, sizeof(DEltX2)); - break; - case 4: - assert(skipSize <= 4); - ZSTD_memcpy(DTable + 0, &DEltX2, sizeof(DEltX2)); - ZSTD_memcpy(DTable + 2, &DEltX2, sizeof(DEltX2)); - break; - default: - { - int i; - for (i = 0; i < skipSize; i += 8) { - ZSTD_memcpy(DTable + i + 0, &DEltX2, sizeof(DEltX2)); - ZSTD_memcpy(DTable + i + 2, &DEltX2, sizeof(DEltX2)); - ZSTD_memcpy(DTable + i + 4, &DEltX2, sizeof(DEltX2)); - ZSTD_memcpy(DTable + i + 6, &DEltX2, sizeof(DEltX2)); - } - } - } - } - - /* Fill each of the second level symbols by weight. */ - { - int w; - for (w = minWeight; w < maxWeight1; ++w) { - int const begin = rankStart[w]; - int const end = rankStart[w+1]; - U32 const nbBits = nbBitsBaseline - w; - U32 const totalBits = nbBits + consumedBits; - HUF_fillDTableX2ForWeight( - DTable + rankVal[w], - sortedSymbols + begin, sortedSymbols + end, - totalBits, targetLog, - baseSeq, /* level */ 2); - } - } -} - -static void HUF_fillDTableX2(HUF_DEltX2* DTable, const U32 targetLog, - const sortedSymbol_t* sortedList, - const U32* rankStart, rankVal_t rankValOrigin, const U32 maxWeight, - const U32 nbBitsBaseline) -{ - U32* const rankVal = rankValOrigin[0]; - const int scaleLog = nbBitsBaseline - targetLog; /* note : targetLog >= srcLog, hence scaleLog <= 1 */ - const U32 minBits = nbBitsBaseline - maxWeight; - int w; - int const wEnd = (int)maxWeight + 1; - - /* Fill DTable in order of weight. */ - for (w = 1; w < wEnd; ++w) { - int const begin = (int)rankStart[w]; - int const end = (int)rankStart[w+1]; - U32 const nbBits = nbBitsBaseline - w; - - if (targetLog-nbBits >= minBits) { - /* Enough room for a second symbol. */ - int start = rankVal[w]; - U32 const length = 1U << ((targetLog - nbBits) & 0x1F /* quiet static-analyzer */); - int minWeight = nbBits + scaleLog; - int s; - if (minWeight < 1) minWeight = 1; - /* Fill the DTable for every symbol of weight w. - * These symbols get at least 1 second symbol. - */ - for (s = begin; s != end; ++s) { - HUF_fillDTableX2Level2( - DTable + start, targetLog, nbBits, - rankValOrigin[nbBits], minWeight, wEnd, - sortedList, rankStart, - nbBitsBaseline, sortedList[s].symbol); - start += length; - } - } else { - /* Only a single symbol. */ - HUF_fillDTableX2ForWeight( - DTable + rankVal[w], - sortedList + begin, sortedList + end, - nbBits, targetLog, - /* baseSeq */ 0, /* level */ 1); - } - } -} - -typedef struct { - rankValCol_t rankVal[HUF_TABLELOG_MAX]; - U32 rankStats[HUF_TABLELOG_MAX + 1]; - U32 rankStart0[HUF_TABLELOG_MAX + 3]; - sortedSymbol_t sortedSymbol[HUF_SYMBOLVALUE_MAX + 1]; - BYTE weightList[HUF_SYMBOLVALUE_MAX + 1]; - U32 calleeWksp[HUF_READ_STATS_WORKSPACE_SIZE_U32]; -} HUF_ReadDTableX2_Workspace; - -size_t HUF_readDTableX2_wksp(HUF_DTable* DTable, - const void* src, size_t srcSize, - void* workSpace, size_t wkspSize) -{ - return HUF_readDTableX2_wksp_bmi2(DTable, src, srcSize, workSpace, wkspSize, /* bmi2 */ 0); -} - -size_t HUF_readDTableX2_wksp_bmi2(HUF_DTable* DTable, - const void* src, size_t srcSize, - void* workSpace, size_t wkspSize, int bmi2) -{ - U32 tableLog, maxW, nbSymbols; - DTableDesc dtd = HUF_getDTableDesc(DTable); - U32 maxTableLog = dtd.maxTableLog; - size_t iSize; - void* dtPtr = DTable+1; /* force compiler to avoid strict-aliasing */ - HUF_DEltX2* const dt = (HUF_DEltX2*)dtPtr; - U32 *rankStart; - - HUF_ReadDTableX2_Workspace* const wksp = (HUF_ReadDTableX2_Workspace*)workSpace; - - if (sizeof(*wksp) > wkspSize) return ERROR(GENERIC); - - rankStart = wksp->rankStart0 + 1; - ZSTD_memset(wksp->rankStats, 0, sizeof(wksp->rankStats)); - ZSTD_memset(wksp->rankStart0, 0, sizeof(wksp->rankStart0)); - - DEBUG_STATIC_ASSERT(sizeof(HUF_DEltX2) == sizeof(HUF_DTable)); /* if compiler fails here, assertion is wrong */ - if (maxTableLog > HUF_TABLELOG_MAX) return ERROR(tableLog_tooLarge); - /* ZSTD_memset(weightList, 0, sizeof(weightList)); */ /* is not necessary, even though some analyzer complain ... */ - - iSize = HUF_readStats_wksp(wksp->weightList, HUF_SYMBOLVALUE_MAX + 1, wksp->rankStats, &nbSymbols, &tableLog, src, srcSize, wksp->calleeWksp, sizeof(wksp->calleeWksp), bmi2); - if (HUF_isError(iSize)) return iSize; - - /* check result */ - if (tableLog > maxTableLog) return ERROR(tableLog_tooLarge); /* DTable can't fit code depth */ - if (tableLog <= HUF_DECODER_FAST_TABLELOG && maxTableLog > HUF_DECODER_FAST_TABLELOG) maxTableLog = HUF_DECODER_FAST_TABLELOG; - - /* find maxWeight */ - for (maxW = tableLog; wksp->rankStats[maxW]==0; maxW--) {} /* necessarily finds a solution before 0 */ - - /* Get start index of each weight */ - { U32 w, nextRankStart = 0; - for (w=1; wrankStats[w]; - rankStart[w] = curr; - } - rankStart[0] = nextRankStart; /* put all 0w symbols at the end of sorted list*/ - rankStart[maxW+1] = nextRankStart; - } - - /* sort symbols by weight */ - { U32 s; - for (s=0; sweightList[s]; - U32 const r = rankStart[w]++; - wksp->sortedSymbol[r].symbol = (BYTE)s; - } - rankStart[0] = 0; /* forget 0w symbols; this is beginning of weight(1) */ - } - - /* Build rankVal */ - { U32* const rankVal0 = wksp->rankVal[0]; - { int const rescale = (maxTableLog-tableLog) - 1; /* tableLog <= maxTableLog */ - U32 nextRankVal = 0; - U32 w; - for (w=1; wrankStats[w] << (w+rescale); - rankVal0[w] = curr; - } } - { U32 const minBits = tableLog+1 - maxW; - U32 consumed; - for (consumed = minBits; consumed < maxTableLog - minBits + 1; consumed++) { - U32* const rankValPtr = wksp->rankVal[consumed]; - U32 w; - for (w = 1; w < maxW+1; w++) { - rankValPtr[w] = rankVal0[w] >> consumed; - } } } } - - HUF_fillDTableX2(dt, maxTableLog, - wksp->sortedSymbol, - wksp->rankStart0, wksp->rankVal, maxW, - tableLog+1); - - dtd.tableLog = (BYTE)maxTableLog; - dtd.tableType = 1; - ZSTD_memcpy(DTable, &dtd, sizeof(dtd)); - return iSize; -} - - -FORCE_INLINE_TEMPLATE U32 -HUF_decodeSymbolX2(void* op, BIT_DStream_t* DStream, const HUF_DEltX2* dt, const U32 dtLog) -{ - size_t const val = BIT_lookBitsFast(DStream, dtLog); /* note : dtLog >= 1 */ - ZSTD_memcpy(op, &dt[val].sequence, 2); - BIT_skipBits(DStream, dt[val].nbBits); - return dt[val].length; -} - -FORCE_INLINE_TEMPLATE U32 -HUF_decodeLastSymbolX2(void* op, BIT_DStream_t* DStream, const HUF_DEltX2* dt, const U32 dtLog) -{ - size_t const val = BIT_lookBitsFast(DStream, dtLog); /* note : dtLog >= 1 */ - ZSTD_memcpy(op, &dt[val].sequence, 1); - if (dt[val].length==1) { - BIT_skipBits(DStream, dt[val].nbBits); - } else { - if (DStream->bitsConsumed < (sizeof(DStream->bitContainer)*8)) { - BIT_skipBits(DStream, dt[val].nbBits); - if (DStream->bitsConsumed > (sizeof(DStream->bitContainer)*8)) - /* ugly hack; works only because it's the last symbol. Note : can't easily extract nbBits from just this symbol */ - DStream->bitsConsumed = (sizeof(DStream->bitContainer)*8); - } - } - return 1; -} - -#define HUF_DECODE_SYMBOLX2_0(ptr, DStreamPtr) \ - ptr += HUF_decodeSymbolX2(ptr, DStreamPtr, dt, dtLog) - -#define HUF_DECODE_SYMBOLX2_1(ptr, DStreamPtr) \ - if (MEM_64bits() || (HUF_TABLELOG_MAX<=12)) \ - ptr += HUF_decodeSymbolX2(ptr, DStreamPtr, dt, dtLog) - -#define HUF_DECODE_SYMBOLX2_2(ptr, DStreamPtr) \ - if (MEM_64bits()) \ - ptr += HUF_decodeSymbolX2(ptr, DStreamPtr, dt, dtLog) - -HINT_INLINE size_t -HUF_decodeStreamX2(BYTE* p, BIT_DStream_t* bitDPtr, BYTE* const pEnd, - const HUF_DEltX2* const dt, const U32 dtLog) -{ - BYTE* const pStart = p; - - /* up to 8 symbols at a time */ - if ((size_t)(pEnd - p) >= sizeof(bitDPtr->bitContainer)) { - if (dtLog <= 11 && MEM_64bits()) { - /* up to 10 symbols at a time */ - while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p < pEnd-9)) { - HUF_DECODE_SYMBOLX2_0(p, bitDPtr); - HUF_DECODE_SYMBOLX2_0(p, bitDPtr); - HUF_DECODE_SYMBOLX2_0(p, bitDPtr); - HUF_DECODE_SYMBOLX2_0(p, bitDPtr); - HUF_DECODE_SYMBOLX2_0(p, bitDPtr); - } - } else { - /* up to 8 symbols at a time */ - while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p < pEnd-(sizeof(bitDPtr->bitContainer)-1))) { - HUF_DECODE_SYMBOLX2_2(p, bitDPtr); - HUF_DECODE_SYMBOLX2_1(p, bitDPtr); - HUF_DECODE_SYMBOLX2_2(p, bitDPtr); - HUF_DECODE_SYMBOLX2_0(p, bitDPtr); - } - } - } else { - BIT_reloadDStream(bitDPtr); - } - - /* closer to end : up to 2 symbols at a time */ - if ((size_t)(pEnd - p) >= 2) { - while ((BIT_reloadDStream(bitDPtr) == BIT_DStream_unfinished) & (p <= pEnd-2)) - HUF_DECODE_SYMBOLX2_0(p, bitDPtr); - - while (p <= pEnd-2) - HUF_DECODE_SYMBOLX2_0(p, bitDPtr); /* no need to reload : reached the end of DStream */ - } - - if (p < pEnd) - p += HUF_decodeLastSymbolX2(p, bitDPtr, dt, dtLog); - - return p-pStart; -} - -FORCE_INLINE_TEMPLATE size_t -HUF_decompress1X2_usingDTable_internal_body( - void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - const HUF_DTable* DTable) -{ - BIT_DStream_t bitD; - - /* Init */ - CHECK_F( BIT_initDStream(&bitD, cSrc, cSrcSize) ); - - /* decode */ - { BYTE* const ostart = (BYTE*) dst; - BYTE* const oend = ostart + dstSize; - const void* const dtPtr = DTable+1; /* force compiler to not use strict-aliasing */ - const HUF_DEltX2* const dt = (const HUF_DEltX2*)dtPtr; - DTableDesc const dtd = HUF_getDTableDesc(DTable); - HUF_decodeStreamX2(ostart, &bitD, oend, dt, dtd.tableLog); - } - - /* check */ - if (!BIT_endOfDStream(&bitD)) return ERROR(corruption_detected); - - /* decoded size */ - return dstSize; -} -FORCE_INLINE_TEMPLATE size_t -HUF_decompress4X2_usingDTable_internal_body( - void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - const HUF_DTable* DTable) -{ - if (cSrcSize < 10) return ERROR(corruption_detected); /* strict minimum : jump table + 1 byte per stream */ - - { const BYTE* const istart = (const BYTE*) cSrc; - BYTE* const ostart = (BYTE*) dst; - BYTE* const oend = ostart + dstSize; - BYTE* const olimit = oend - (sizeof(size_t)-1); - const void* const dtPtr = DTable+1; - const HUF_DEltX2* const dt = (const HUF_DEltX2*)dtPtr; - - /* Init */ - BIT_DStream_t bitD1; - BIT_DStream_t bitD2; - BIT_DStream_t bitD3; - BIT_DStream_t bitD4; - size_t const length1 = MEM_readLE16(istart); - size_t const length2 = MEM_readLE16(istart+2); - size_t const length3 = MEM_readLE16(istart+4); - size_t const length4 = cSrcSize - (length1 + length2 + length3 + 6); - const BYTE* const istart1 = istart + 6; /* jumpTable */ - const BYTE* const istart2 = istart1 + length1; - const BYTE* const istart3 = istart2 + length2; - const BYTE* const istart4 = istart3 + length3; - size_t const segmentSize = (dstSize+3) / 4; - BYTE* const opStart2 = ostart + segmentSize; - BYTE* const opStart3 = opStart2 + segmentSize; - BYTE* const opStart4 = opStart3 + segmentSize; - BYTE* op1 = ostart; - BYTE* op2 = opStart2; - BYTE* op3 = opStart3; - BYTE* op4 = opStart4; - U32 endSignal = 1; - DTableDesc const dtd = HUF_getDTableDesc(DTable); - U32 const dtLog = dtd.tableLog; - - if (length4 > cSrcSize) return ERROR(corruption_detected); /* overflow */ - if (opStart4 > oend) return ERROR(corruption_detected); /* overflow */ - CHECK_F( BIT_initDStream(&bitD1, istart1, length1) ); - CHECK_F( BIT_initDStream(&bitD2, istart2, length2) ); - CHECK_F( BIT_initDStream(&bitD3, istart3, length3) ); - CHECK_F( BIT_initDStream(&bitD4, istart4, length4) ); - - /* 16-32 symbols per loop (4-8 symbols per stream) */ - if ((size_t)(oend - op4) >= sizeof(size_t)) { - for ( ; (endSignal) & (op4 < olimit); ) { -#if defined(__clang__) && (defined(__x86_64__) || defined(__i386__)) - HUF_DECODE_SYMBOLX2_2(op1, &bitD1); - HUF_DECODE_SYMBOLX2_1(op1, &bitD1); - HUF_DECODE_SYMBOLX2_2(op1, &bitD1); - HUF_DECODE_SYMBOLX2_0(op1, &bitD1); - HUF_DECODE_SYMBOLX2_2(op2, &bitD2); - HUF_DECODE_SYMBOLX2_1(op2, &bitD2); - HUF_DECODE_SYMBOLX2_2(op2, &bitD2); - HUF_DECODE_SYMBOLX2_0(op2, &bitD2); - endSignal &= BIT_reloadDStreamFast(&bitD1) == BIT_DStream_unfinished; - endSignal &= BIT_reloadDStreamFast(&bitD2) == BIT_DStream_unfinished; - HUF_DECODE_SYMBOLX2_2(op3, &bitD3); - HUF_DECODE_SYMBOLX2_1(op3, &bitD3); - HUF_DECODE_SYMBOLX2_2(op3, &bitD3); - HUF_DECODE_SYMBOLX2_0(op3, &bitD3); - HUF_DECODE_SYMBOLX2_2(op4, &bitD4); - HUF_DECODE_SYMBOLX2_1(op4, &bitD4); - HUF_DECODE_SYMBOLX2_2(op4, &bitD4); - HUF_DECODE_SYMBOLX2_0(op4, &bitD4); - endSignal &= BIT_reloadDStreamFast(&bitD3) == BIT_DStream_unfinished; - endSignal &= BIT_reloadDStreamFast(&bitD4) == BIT_DStream_unfinished; -#else - HUF_DECODE_SYMBOLX2_2(op1, &bitD1); - HUF_DECODE_SYMBOLX2_2(op2, &bitD2); - HUF_DECODE_SYMBOLX2_2(op3, &bitD3); - HUF_DECODE_SYMBOLX2_2(op4, &bitD4); - HUF_DECODE_SYMBOLX2_1(op1, &bitD1); - HUF_DECODE_SYMBOLX2_1(op2, &bitD2); - HUF_DECODE_SYMBOLX2_1(op3, &bitD3); - HUF_DECODE_SYMBOLX2_1(op4, &bitD4); - HUF_DECODE_SYMBOLX2_2(op1, &bitD1); - HUF_DECODE_SYMBOLX2_2(op2, &bitD2); - HUF_DECODE_SYMBOLX2_2(op3, &bitD3); - HUF_DECODE_SYMBOLX2_2(op4, &bitD4); - HUF_DECODE_SYMBOLX2_0(op1, &bitD1); - HUF_DECODE_SYMBOLX2_0(op2, &bitD2); - HUF_DECODE_SYMBOLX2_0(op3, &bitD3); - HUF_DECODE_SYMBOLX2_0(op4, &bitD4); - endSignal = (U32)LIKELY((U32) - (BIT_reloadDStreamFast(&bitD1) == BIT_DStream_unfinished) - & (BIT_reloadDStreamFast(&bitD2) == BIT_DStream_unfinished) - & (BIT_reloadDStreamFast(&bitD3) == BIT_DStream_unfinished) - & (BIT_reloadDStreamFast(&bitD4) == BIT_DStream_unfinished)); -#endif - } - } - - /* check corruption */ - if (op1 > opStart2) return ERROR(corruption_detected); - if (op2 > opStart3) return ERROR(corruption_detected); - if (op3 > opStart4) return ERROR(corruption_detected); - /* note : op4 already verified within main loop */ - - /* finish bitStreams one by one */ - HUF_decodeStreamX2(op1, &bitD1, opStart2, dt, dtLog); - HUF_decodeStreamX2(op2, &bitD2, opStart3, dt, dtLog); - HUF_decodeStreamX2(op3, &bitD3, opStart4, dt, dtLog); - HUF_decodeStreamX2(op4, &bitD4, oend, dt, dtLog); - - /* check */ - { U32 const endCheck = BIT_endOfDStream(&bitD1) & BIT_endOfDStream(&bitD2) & BIT_endOfDStream(&bitD3) & BIT_endOfDStream(&bitD4); - if (!endCheck) return ERROR(corruption_detected); } - - /* decoded size */ - return dstSize; - } -} - -#if HUF_NEED_BMI2_FUNCTION -static BMI2_TARGET_ATTRIBUTE -size_t HUF_decompress4X2_usingDTable_internal_bmi2(void* dst, size_t dstSize, void const* cSrc, - size_t cSrcSize, HUF_DTable const* DTable) { - return HUF_decompress4X2_usingDTable_internal_body(dst, dstSize, cSrc, cSrcSize, DTable); -} -#endif - -#if HUF_NEED_DEFAULT_FUNCTION -static -size_t HUF_decompress4X2_usingDTable_internal_default(void* dst, size_t dstSize, void const* cSrc, - size_t cSrcSize, HUF_DTable const* DTable) { - return HUF_decompress4X2_usingDTable_internal_body(dst, dstSize, cSrc, cSrcSize, DTable); -} -#endif - -#if ZSTD_ENABLE_ASM_X86_64_BMI2 - -HUF_ASM_DECL void HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop(HUF_DecompressAsmArgs* args) ZSTDLIB_HIDDEN; - -static HUF_ASM_X86_64_BMI2_ATTRS size_t -HUF_decompress4X2_usingDTable_internal_bmi2_asm( - void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - const HUF_DTable* DTable) { - void const* dt = DTable + 1; - const BYTE* const iend = (const BYTE*)cSrc + 6; - BYTE* const oend = (BYTE*)dst + dstSize; - HUF_DecompressAsmArgs args; - { - size_t const ret = HUF_DecompressAsmArgs_init(&args, dst, dstSize, cSrc, cSrcSize, DTable); - FORWARD_IF_ERROR(ret, "Failed to init asm args"); - if (ret != 0) - return HUF_decompress4X2_usingDTable_internal_bmi2(dst, dstSize, cSrc, cSrcSize, DTable); - } - - assert(args.ip[0] >= args.ilimit); - HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop(&args); - - /* note : op4 already verified within main loop */ - assert(args.ip[0] >= iend); - assert(args.ip[1] >= iend); - assert(args.ip[2] >= iend); - assert(args.ip[3] >= iend); - assert(args.op[3] <= oend); - (void)iend; - - /* finish bitStreams one by one */ - { - size_t const segmentSize = (dstSize+3) / 4; - BYTE* segmentEnd = (BYTE*)dst; - int i; - for (i = 0; i < 4; ++i) { - BIT_DStream_t bit; - if (segmentSize <= (size_t)(oend - segmentEnd)) - segmentEnd += segmentSize; - else - segmentEnd = oend; - FORWARD_IF_ERROR(HUF_initRemainingDStream(&bit, &args, i, segmentEnd), "corruption"); - args.op[i] += HUF_decodeStreamX2(args.op[i], &bit, segmentEnd, (HUF_DEltX2 const*)dt, HUF_DECODER_FAST_TABLELOG); - if (args.op[i] != segmentEnd) - return ERROR(corruption_detected); - } - } - - /* decoded size */ - return dstSize; -} -#endif /* ZSTD_ENABLE_ASM_X86_64_BMI2 */ - -static size_t HUF_decompress4X2_usingDTable_internal(void* dst, size_t dstSize, void const* cSrc, - size_t cSrcSize, HUF_DTable const* DTable, int bmi2) -{ -#if DYNAMIC_BMI2 - if (bmi2) { -# if ZSTD_ENABLE_ASM_X86_64_BMI2 - return HUF_decompress4X2_usingDTable_internal_bmi2_asm(dst, dstSize, cSrc, cSrcSize, DTable); -# else - return HUF_decompress4X2_usingDTable_internal_bmi2(dst, dstSize, cSrc, cSrcSize, DTable); -# endif - } -#else - (void)bmi2; -#endif - -#if ZSTD_ENABLE_ASM_X86_64_BMI2 && defined(__BMI2__) - return HUF_decompress4X2_usingDTable_internal_bmi2_asm(dst, dstSize, cSrc, cSrcSize, DTable); -#else - return HUF_decompress4X2_usingDTable_internal_default(dst, dstSize, cSrc, cSrcSize, DTable); -#endif -} - -HUF_DGEN(HUF_decompress1X2_usingDTable_internal) - -size_t HUF_decompress1X2_usingDTable( - void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - const HUF_DTable* DTable) -{ - DTableDesc dtd = HUF_getDTableDesc(DTable); - if (dtd.tableType != 1) return ERROR(GENERIC); - return HUF_decompress1X2_usingDTable_internal(dst, dstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0); -} - -size_t HUF_decompress1X2_DCtx_wksp(HUF_DTable* DCtx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - void* workSpace, size_t wkspSize) -{ - const BYTE* ip = (const BYTE*) cSrc; - - size_t const hSize = HUF_readDTableX2_wksp(DCtx, cSrc, cSrcSize, - workSpace, wkspSize); - if (HUF_isError(hSize)) return hSize; - if (hSize >= cSrcSize) return ERROR(srcSize_wrong); - ip += hSize; cSrcSize -= hSize; - - return HUF_decompress1X2_usingDTable_internal(dst, dstSize, ip, cSrcSize, DCtx, /* bmi2 */ 0); -} - - -size_t HUF_decompress4X2_usingDTable( - void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - const HUF_DTable* DTable) -{ - DTableDesc dtd = HUF_getDTableDesc(DTable); - if (dtd.tableType != 1) return ERROR(GENERIC); - return HUF_decompress4X2_usingDTable_internal(dst, dstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0); -} - -static size_t HUF_decompress4X2_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - void* workSpace, size_t wkspSize, int bmi2) -{ - const BYTE* ip = (const BYTE*) cSrc; - - size_t hSize = HUF_readDTableX2_wksp(dctx, cSrc, cSrcSize, - workSpace, wkspSize); - if (HUF_isError(hSize)) return hSize; - if (hSize >= cSrcSize) return ERROR(srcSize_wrong); - ip += hSize; cSrcSize -= hSize; - - return HUF_decompress4X2_usingDTable_internal(dst, dstSize, ip, cSrcSize, dctx, bmi2); -} - -size_t HUF_decompress4X2_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - void* workSpace, size_t wkspSize) -{ - return HUF_decompress4X2_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, /* bmi2 */ 0); -} - - -#endif /* HUF_FORCE_DECOMPRESS_X1 */ - - -/* ***********************************/ -/* Universal decompression selectors */ -/* ***********************************/ - -size_t HUF_decompress1X_usingDTable(void* dst, size_t maxDstSize, - const void* cSrc, size_t cSrcSize, - const HUF_DTable* DTable) -{ - DTableDesc const dtd = HUF_getDTableDesc(DTable); -#if defined(HUF_FORCE_DECOMPRESS_X1) - (void)dtd; - assert(dtd.tableType == 0); - return HUF_decompress1X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0); -#elif defined(HUF_FORCE_DECOMPRESS_X2) - (void)dtd; - assert(dtd.tableType == 1); - return HUF_decompress1X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0); -#else - return dtd.tableType ? HUF_decompress1X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0) : - HUF_decompress1X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0); -#endif -} - -size_t HUF_decompress4X_usingDTable(void* dst, size_t maxDstSize, - const void* cSrc, size_t cSrcSize, - const HUF_DTable* DTable) -{ - DTableDesc const dtd = HUF_getDTableDesc(DTable); -#if defined(HUF_FORCE_DECOMPRESS_X1) - (void)dtd; - assert(dtd.tableType == 0); - return HUF_decompress4X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0); -#elif defined(HUF_FORCE_DECOMPRESS_X2) - (void)dtd; - assert(dtd.tableType == 1); - return HUF_decompress4X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0); -#else - return dtd.tableType ? HUF_decompress4X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0) : - HUF_decompress4X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, /* bmi2 */ 0); -#endif -} - - -#if !defined(HUF_FORCE_DECOMPRESS_X1) && !defined(HUF_FORCE_DECOMPRESS_X2) -typedef struct { U32 tableTime; U32 decode256Time; } algo_time_t; -static const algo_time_t algoTime[16 /* Quantization */][2 /* single, double */] = -{ - /* single, double, quad */ - {{0,0}, {1,1}}, /* Q==0 : impossible */ - {{0,0}, {1,1}}, /* Q==1 : impossible */ - {{ 150,216}, { 381,119}}, /* Q == 2 : 12-18% */ - {{ 170,205}, { 514,112}}, /* Q == 3 : 18-25% */ - {{ 177,199}, { 539,110}}, /* Q == 4 : 25-32% */ - {{ 197,194}, { 644,107}}, /* Q == 5 : 32-38% */ - {{ 221,192}, { 735,107}}, /* Q == 6 : 38-44% */ - {{ 256,189}, { 881,106}}, /* Q == 7 : 44-50% */ - {{ 359,188}, {1167,109}}, /* Q == 8 : 50-56% */ - {{ 582,187}, {1570,114}}, /* Q == 9 : 56-62% */ - {{ 688,187}, {1712,122}}, /* Q ==10 : 62-69% */ - {{ 825,186}, {1965,136}}, /* Q ==11 : 69-75% */ - {{ 976,185}, {2131,150}}, /* Q ==12 : 75-81% */ - {{1180,186}, {2070,175}}, /* Q ==13 : 81-87% */ - {{1377,185}, {1731,202}}, /* Q ==14 : 87-93% */ - {{1412,185}, {1695,202}}, /* Q ==15 : 93-99% */ -}; -#endif - -/** HUF_selectDecoder() : - * Tells which decoder is likely to decode faster, - * based on a set of pre-computed metrics. - * @return : 0==HUF_decompress4X1, 1==HUF_decompress4X2 . - * Assumption : 0 < dstSize <= 128 KB */ -U32 HUF_selectDecoder (size_t dstSize, size_t cSrcSize) -{ - assert(dstSize > 0); - assert(dstSize <= 128*1024); -#if defined(HUF_FORCE_DECOMPRESS_X1) - (void)dstSize; - (void)cSrcSize; - return 0; -#elif defined(HUF_FORCE_DECOMPRESS_X2) - (void)dstSize; - (void)cSrcSize; - return 1; -#else - /* decoder timing evaluation */ - { U32 const Q = (cSrcSize >= dstSize) ? 15 : (U32)(cSrcSize * 16 / dstSize); /* Q < 16 */ - U32 const D256 = (U32)(dstSize >> 8); - U32 const DTime0 = algoTime[Q][0].tableTime + (algoTime[Q][0].decode256Time * D256); - U32 DTime1 = algoTime[Q][1].tableTime + (algoTime[Q][1].decode256Time * D256); - DTime1 += DTime1 >> 5; /* small advantage to algorithm using less memory, to reduce cache eviction */ - return DTime1 < DTime0; - } -#endif -} - - -size_t HUF_decompress4X_hufOnly_wksp(HUF_DTable* dctx, void* dst, - size_t dstSize, const void* cSrc, - size_t cSrcSize, void* workSpace, - size_t wkspSize) -{ - /* validation checks */ - if (dstSize == 0) return ERROR(dstSize_tooSmall); - if (cSrcSize == 0) return ERROR(corruption_detected); - - { U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize); -#if defined(HUF_FORCE_DECOMPRESS_X1) - (void)algoNb; - assert(algoNb == 0); - return HUF_decompress4X1_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize); -#elif defined(HUF_FORCE_DECOMPRESS_X2) - (void)algoNb; - assert(algoNb == 1); - return HUF_decompress4X2_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize); -#else - return algoNb ? HUF_decompress4X2_DCtx_wksp(dctx, dst, dstSize, cSrc, - cSrcSize, workSpace, wkspSize): - HUF_decompress4X1_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize); -#endif - } -} - -size_t HUF_decompress1X_DCtx_wksp(HUF_DTable* dctx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize, - void* workSpace, size_t wkspSize) -{ - /* validation checks */ - if (dstSize == 0) return ERROR(dstSize_tooSmall); - if (cSrcSize > dstSize) return ERROR(corruption_detected); /* invalid */ - if (cSrcSize == dstSize) { ZSTD_memcpy(dst, cSrc, dstSize); return dstSize; } /* not compressed */ - if (cSrcSize == 1) { ZSTD_memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */ - - { U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize); -#if defined(HUF_FORCE_DECOMPRESS_X1) - (void)algoNb; - assert(algoNb == 0); - return HUF_decompress1X1_DCtx_wksp(dctx, dst, dstSize, cSrc, - cSrcSize, workSpace, wkspSize); -#elif defined(HUF_FORCE_DECOMPRESS_X2) - (void)algoNb; - assert(algoNb == 1); - return HUF_decompress1X2_DCtx_wksp(dctx, dst, dstSize, cSrc, - cSrcSize, workSpace, wkspSize); -#else - return algoNb ? HUF_decompress1X2_DCtx_wksp(dctx, dst, dstSize, cSrc, - cSrcSize, workSpace, wkspSize): - HUF_decompress1X1_DCtx_wksp(dctx, dst, dstSize, cSrc, - cSrcSize, workSpace, wkspSize); -#endif - } -} - - -size_t HUF_decompress1X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2) -{ - DTableDesc const dtd = HUF_getDTableDesc(DTable); -#if defined(HUF_FORCE_DECOMPRESS_X1) - (void)dtd; - assert(dtd.tableType == 0); - return HUF_decompress1X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2); -#elif defined(HUF_FORCE_DECOMPRESS_X2) - (void)dtd; - assert(dtd.tableType == 1); - return HUF_decompress1X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2); -#else - return dtd.tableType ? HUF_decompress1X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2) : - HUF_decompress1X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2); -#endif -} - -#ifndef HUF_FORCE_DECOMPRESS_X2 -size_t HUF_decompress1X1_DCtx_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2) -{ - const BYTE* ip = (const BYTE*) cSrc; - - size_t const hSize = HUF_readDTableX1_wksp_bmi2(dctx, cSrc, cSrcSize, workSpace, wkspSize, bmi2); - if (HUF_isError(hSize)) return hSize; - if (hSize >= cSrcSize) return ERROR(srcSize_wrong); - ip += hSize; cSrcSize -= hSize; - - return HUF_decompress1X1_usingDTable_internal(dst, dstSize, ip, cSrcSize, dctx, bmi2); -} -#endif - -size_t HUF_decompress4X_usingDTable_bmi2(void* dst, size_t maxDstSize, const void* cSrc, size_t cSrcSize, const HUF_DTable* DTable, int bmi2) -{ - DTableDesc const dtd = HUF_getDTableDesc(DTable); -#if defined(HUF_FORCE_DECOMPRESS_X1) - (void)dtd; - assert(dtd.tableType == 0); - return HUF_decompress4X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2); -#elif defined(HUF_FORCE_DECOMPRESS_X2) - (void)dtd; - assert(dtd.tableType == 1); - return HUF_decompress4X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2); -#else - return dtd.tableType ? HUF_decompress4X2_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2) : - HUF_decompress4X1_usingDTable_internal(dst, maxDstSize, cSrc, cSrcSize, DTable, bmi2); -#endif -} - -size_t HUF_decompress4X_hufOnly_wksp_bmi2(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize, void* workSpace, size_t wkspSize, int bmi2) -{ - /* validation checks */ - if (dstSize == 0) return ERROR(dstSize_tooSmall); - if (cSrcSize == 0) return ERROR(corruption_detected); - - { U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize); -#if defined(HUF_FORCE_DECOMPRESS_X1) - (void)algoNb; - assert(algoNb == 0); - return HUF_decompress4X1_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, bmi2); -#elif defined(HUF_FORCE_DECOMPRESS_X2) - (void)algoNb; - assert(algoNb == 1); - return HUF_decompress4X2_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, bmi2); -#else - return algoNb ? HUF_decompress4X2_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, bmi2) : - HUF_decompress4X1_DCtx_wksp_bmi2(dctx, dst, dstSize, cSrc, cSrcSize, workSpace, wkspSize, bmi2); -#endif - } -} - -#ifndef ZSTD_NO_UNUSED_FUNCTIONS -#ifndef HUF_FORCE_DECOMPRESS_X2 -size_t HUF_readDTableX1(HUF_DTable* DTable, const void* src, size_t srcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_readDTableX1_wksp(DTable, src, srcSize, - workSpace, sizeof(workSpace)); -} - -size_t HUF_decompress1X1_DCtx(HUF_DTable* DCtx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_decompress1X1_DCtx_wksp(DCtx, dst, dstSize, cSrc, cSrcSize, - workSpace, sizeof(workSpace)); -} - -size_t HUF_decompress1X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - HUF_CREATE_STATIC_DTABLEX1(DTable, HUF_TABLELOG_MAX); - return HUF_decompress1X1_DCtx (DTable, dst, dstSize, cSrc, cSrcSize); -} -#endif - -#ifndef HUF_FORCE_DECOMPRESS_X1 -size_t HUF_readDTableX2(HUF_DTable* DTable, const void* src, size_t srcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_readDTableX2_wksp(DTable, src, srcSize, - workSpace, sizeof(workSpace)); -} - -size_t HUF_decompress1X2_DCtx(HUF_DTable* DCtx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_decompress1X2_DCtx_wksp(DCtx, dst, dstSize, cSrc, cSrcSize, - workSpace, sizeof(workSpace)); -} - -size_t HUF_decompress1X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - HUF_CREATE_STATIC_DTABLEX2(DTable, HUF_TABLELOG_MAX); - return HUF_decompress1X2_DCtx(DTable, dst, dstSize, cSrc, cSrcSize); -} -#endif - -#ifndef HUF_FORCE_DECOMPRESS_X2 -size_t HUF_decompress4X1_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_decompress4X1_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, - workSpace, sizeof(workSpace)); -} -size_t HUF_decompress4X1 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - HUF_CREATE_STATIC_DTABLEX1(DTable, HUF_TABLELOG_MAX); - return HUF_decompress4X1_DCtx(DTable, dst, dstSize, cSrc, cSrcSize); -} -#endif - -#ifndef HUF_FORCE_DECOMPRESS_X1 -size_t HUF_decompress4X2_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_decompress4X2_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, - workSpace, sizeof(workSpace)); -} - -size_t HUF_decompress4X2 (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - HUF_CREATE_STATIC_DTABLEX2(DTable, HUF_TABLELOG_MAX); - return HUF_decompress4X2_DCtx(DTable, dst, dstSize, cSrc, cSrcSize); -} -#endif - -typedef size_t (*decompressionAlgo)(void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize); - -size_t HUF_decompress (void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ -#if !defined(HUF_FORCE_DECOMPRESS_X1) && !defined(HUF_FORCE_DECOMPRESS_X2) - static const decompressionAlgo decompress[2] = { HUF_decompress4X1, HUF_decompress4X2 }; -#endif - - /* validation checks */ - if (dstSize == 0) return ERROR(dstSize_tooSmall); - if (cSrcSize > dstSize) return ERROR(corruption_detected); /* invalid */ - if (cSrcSize == dstSize) { ZSTD_memcpy(dst, cSrc, dstSize); return dstSize; } /* not compressed */ - if (cSrcSize == 1) { ZSTD_memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */ - - { U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize); -#if defined(HUF_FORCE_DECOMPRESS_X1) - (void)algoNb; - assert(algoNb == 0); - return HUF_decompress4X1(dst, dstSize, cSrc, cSrcSize); -#elif defined(HUF_FORCE_DECOMPRESS_X2) - (void)algoNb; - assert(algoNb == 1); - return HUF_decompress4X2(dst, dstSize, cSrc, cSrcSize); -#else - return decompress[algoNb](dst, dstSize, cSrc, cSrcSize); -#endif - } -} - -size_t HUF_decompress4X_DCtx (HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - /* validation checks */ - if (dstSize == 0) return ERROR(dstSize_tooSmall); - if (cSrcSize > dstSize) return ERROR(corruption_detected); /* invalid */ - if (cSrcSize == dstSize) { ZSTD_memcpy(dst, cSrc, dstSize); return dstSize; } /* not compressed */ - if (cSrcSize == 1) { ZSTD_memset(dst, *(const BYTE*)cSrc, dstSize); return dstSize; } /* RLE */ - - { U32 const algoNb = HUF_selectDecoder(dstSize, cSrcSize); -#if defined(HUF_FORCE_DECOMPRESS_X1) - (void)algoNb; - assert(algoNb == 0); - return HUF_decompress4X1_DCtx(dctx, dst, dstSize, cSrc, cSrcSize); -#elif defined(HUF_FORCE_DECOMPRESS_X2) - (void)algoNb; - assert(algoNb == 1); - return HUF_decompress4X2_DCtx(dctx, dst, dstSize, cSrc, cSrcSize); -#else - return algoNb ? HUF_decompress4X2_DCtx(dctx, dst, dstSize, cSrc, cSrcSize) : - HUF_decompress4X1_DCtx(dctx, dst, dstSize, cSrc, cSrcSize) ; -#endif - } -} - -size_t HUF_decompress4X_hufOnly(HUF_DTable* dctx, void* dst, size_t dstSize, const void* cSrc, size_t cSrcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_decompress4X_hufOnly_wksp(dctx, dst, dstSize, cSrc, cSrcSize, - workSpace, sizeof(workSpace)); -} - -size_t HUF_decompress1X_DCtx(HUF_DTable* dctx, void* dst, size_t dstSize, - const void* cSrc, size_t cSrcSize) -{ - U32 workSpace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; - return HUF_decompress1X_DCtx_wksp(dctx, dst, dstSize, cSrc, cSrcSize, - workSpace, sizeof(workSpace)); -} -#endif diff --git a/dep/zstd/lib/decompress/huf_decompress_amd64.S b/dep/zstd/lib/decompress/huf_decompress_amd64.S deleted file mode 100644 index 49589cb61..000000000 --- a/dep/zstd/lib/decompress/huf_decompress_amd64.S +++ /dev/null @@ -1,585 +0,0 @@ -/* - * Copyright (c) Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#include "../common/portability_macros.h" - -/* Stack marking - * ref: https://wiki.gentoo.org/wiki/Hardened/GNU_stack_quickstart - */ -#if defined(__ELF__) && defined(__GNUC__) -.section .note.GNU-stack,"",%progbits -#endif - -#if ZSTD_ENABLE_ASM_X86_64_BMI2 - -/* Calling convention: - * - * %rdi contains the first argument: HUF_DecompressAsmArgs*. - * %rbp isn't maintained (no frame pointer). - * %rsp contains the stack pointer that grows down. - * No red-zone is assumed, only addresses >= %rsp are used. - * All register contents are preserved. - * - * TODO: Support Windows calling convention. - */ - -ZSTD_HIDE_ASM_FUNCTION(HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop) -ZSTD_HIDE_ASM_FUNCTION(HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop) -ZSTD_HIDE_ASM_FUNCTION(_HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop) -ZSTD_HIDE_ASM_FUNCTION(_HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop) -.global HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop -.global HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop -.global _HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop -.global _HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop -.text - -/* Sets up register mappings for clarity. - * op[], bits[], dtable & ip[0] each get their own register. - * ip[1,2,3] & olimit alias var[]. - * %rax is a scratch register. - */ - -#define op0 rsi -#define op1 rbx -#define op2 rcx -#define op3 rdi - -#define ip0 r8 -#define ip1 r9 -#define ip2 r10 -#define ip3 r11 - -#define bits0 rbp -#define bits1 rdx -#define bits2 r12 -#define bits3 r13 -#define dtable r14 -#define olimit r15 - -/* var[] aliases ip[1,2,3] & olimit - * ip[1,2,3] are saved every iteration. - * olimit is only used in compute_olimit. - */ -#define var0 r15 -#define var1 r9 -#define var2 r10 -#define var3 r11 - -/* 32-bit var registers */ -#define vard0 r15d -#define vard1 r9d -#define vard2 r10d -#define vard3 r11d - -/* Calls X(N) for each stream 0, 1, 2, 3. */ -#define FOR_EACH_STREAM(X) \ - X(0); \ - X(1); \ - X(2); \ - X(3) - -/* Calls X(N, idx) for each stream 0, 1, 2, 3. */ -#define FOR_EACH_STREAM_WITH_INDEX(X, idx) \ - X(0, idx); \ - X(1, idx); \ - X(2, idx); \ - X(3, idx) - -/* Define both _HUF_* & HUF_* symbols because MacOS - * C symbols are prefixed with '_' & Linux symbols aren't. - */ -_HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop: -HUF_decompress4X1_usingDTable_internal_bmi2_asm_loop: - /* Save all registers - even if they are callee saved for simplicity. */ - push %rax - push %rbx - push %rcx - push %rdx - push %rbp - push %rsi - push %rdi - push %r8 - push %r9 - push %r10 - push %r11 - push %r12 - push %r13 - push %r14 - push %r15 - - /* Read HUF_DecompressAsmArgs* args from %rax */ - movq %rdi, %rax - movq 0(%rax), %ip0 - movq 8(%rax), %ip1 - movq 16(%rax), %ip2 - movq 24(%rax), %ip3 - movq 32(%rax), %op0 - movq 40(%rax), %op1 - movq 48(%rax), %op2 - movq 56(%rax), %op3 - movq 64(%rax), %bits0 - movq 72(%rax), %bits1 - movq 80(%rax), %bits2 - movq 88(%rax), %bits3 - movq 96(%rax), %dtable - push %rax /* argument */ - push 104(%rax) /* ilimit */ - push 112(%rax) /* oend */ - push %olimit /* olimit space */ - - subq $24, %rsp - -.L_4X1_compute_olimit: - /* Computes how many iterations we can do safely - * %r15, %rax may be clobbered - * rbx, rdx must be saved - * op3 & ip0 mustn't be clobbered - */ - movq %rbx, 0(%rsp) - movq %rdx, 8(%rsp) - - movq 32(%rsp), %rax /* rax = oend */ - subq %op3, %rax /* rax = oend - op3 */ - - /* r15 = (oend - op3) / 5 */ - movabsq $-3689348814741910323, %rdx - mulq %rdx - movq %rdx, %r15 - shrq $2, %r15 - - movq %ip0, %rax /* rax = ip0 */ - movq 40(%rsp), %rdx /* rdx = ilimit */ - subq %rdx, %rax /* rax = ip0 - ilimit */ - movq %rax, %rbx /* rbx = ip0 - ilimit */ - - /* rdx = (ip0 - ilimit) / 7 */ - movabsq $2635249153387078803, %rdx - mulq %rdx - subq %rdx, %rbx - shrq %rbx - addq %rbx, %rdx - shrq $2, %rdx - - /* r15 = min(%rdx, %r15) */ - cmpq %rdx, %r15 - cmova %rdx, %r15 - - /* r15 = r15 * 5 */ - leaq (%r15, %r15, 4), %r15 - - /* olimit = op3 + r15 */ - addq %op3, %olimit - - movq 8(%rsp), %rdx - movq 0(%rsp), %rbx - - /* If (op3 + 20 > olimit) */ - movq %op3, %rax /* rax = op3 */ - addq $20, %rax /* rax = op3 + 20 */ - cmpq %rax, %olimit /* op3 + 20 > olimit */ - jb .L_4X1_exit - - /* If (ip1 < ip0) go to exit */ - cmpq %ip0, %ip1 - jb .L_4X1_exit - - /* If (ip2 < ip1) go to exit */ - cmpq %ip1, %ip2 - jb .L_4X1_exit - - /* If (ip3 < ip2) go to exit */ - cmpq %ip2, %ip3 - jb .L_4X1_exit - -/* Reads top 11 bits from bits[n] - * Loads dt[bits[n]] into var[n] - */ -#define GET_NEXT_DELT(n) \ - movq $53, %var##n; \ - shrxq %var##n, %bits##n, %var##n; \ - movzwl (%dtable,%var##n,2),%vard##n - -/* var[n] must contain the DTable entry computed with GET_NEXT_DELT - * Moves var[n] to %rax - * bits[n] <<= var[n] & 63 - * op[n][idx] = %rax >> 8 - * %ah is a way to access bits [8, 16) of %rax - */ -#define DECODE_FROM_DELT(n, idx) \ - movq %var##n, %rax; \ - shlxq %var##n, %bits##n, %bits##n; \ - movb %ah, idx(%op##n) - -/* Assumes GET_NEXT_DELT has been called. - * Calls DECODE_FROM_DELT then GET_NEXT_DELT - */ -#define DECODE_AND_GET_NEXT(n, idx) \ - DECODE_FROM_DELT(n, idx); \ - GET_NEXT_DELT(n) \ - -/* // ctz & nbBytes is stored in bits[n] - * // nbBits is stored in %rax - * ctz = CTZ[bits[n]] - * nbBits = ctz & 7 - * nbBytes = ctz >> 3 - * op[n] += 5 - * ip[n] -= nbBytes - * // Note: x86-64 is little-endian ==> no bswap - * bits[n] = MEM_readST(ip[n]) | 1 - * bits[n] <<= nbBits - */ -#define RELOAD_BITS(n) \ - bsfq %bits##n, %bits##n; \ - movq %bits##n, %rax; \ - andq $7, %rax; \ - shrq $3, %bits##n; \ - leaq 5(%op##n), %op##n; \ - subq %bits##n, %ip##n; \ - movq (%ip##n), %bits##n; \ - orq $1, %bits##n; \ - shlx %rax, %bits##n, %bits##n - - /* Store clobbered variables on the stack */ - movq %olimit, 24(%rsp) - movq %ip1, 0(%rsp) - movq %ip2, 8(%rsp) - movq %ip3, 16(%rsp) - - /* Call GET_NEXT_DELT for each stream */ - FOR_EACH_STREAM(GET_NEXT_DELT) - - .p2align 6 - -.L_4X1_loop_body: - /* Decode 5 symbols in each of the 4 streams (20 total) - * Must have called GET_NEXT_DELT for each stream - */ - FOR_EACH_STREAM_WITH_INDEX(DECODE_AND_GET_NEXT, 0) - FOR_EACH_STREAM_WITH_INDEX(DECODE_AND_GET_NEXT, 1) - FOR_EACH_STREAM_WITH_INDEX(DECODE_AND_GET_NEXT, 2) - FOR_EACH_STREAM_WITH_INDEX(DECODE_AND_GET_NEXT, 3) - FOR_EACH_STREAM_WITH_INDEX(DECODE_FROM_DELT, 4) - - /* Load ip[1,2,3] from stack (var[] aliases them) - * ip[] is needed for RELOAD_BITS - * Each will be stored back to the stack after RELOAD - */ - movq 0(%rsp), %ip1 - movq 8(%rsp), %ip2 - movq 16(%rsp), %ip3 - - /* Reload each stream & fetch the next table entry - * to prepare for the next iteration - */ - RELOAD_BITS(0) - GET_NEXT_DELT(0) - - RELOAD_BITS(1) - movq %ip1, 0(%rsp) - GET_NEXT_DELT(1) - - RELOAD_BITS(2) - movq %ip2, 8(%rsp) - GET_NEXT_DELT(2) - - RELOAD_BITS(3) - movq %ip3, 16(%rsp) - GET_NEXT_DELT(3) - - /* If op3 < olimit: continue the loop */ - cmp %op3, 24(%rsp) - ja .L_4X1_loop_body - - /* Reload ip[1,2,3] from stack */ - movq 0(%rsp), %ip1 - movq 8(%rsp), %ip2 - movq 16(%rsp), %ip3 - - /* Re-compute olimit */ - jmp .L_4X1_compute_olimit - -#undef GET_NEXT_DELT -#undef DECODE_FROM_DELT -#undef DECODE -#undef RELOAD_BITS -.L_4X1_exit: - addq $24, %rsp - - /* Restore stack (oend & olimit) */ - pop %rax /* olimit */ - pop %rax /* oend */ - pop %rax /* ilimit */ - pop %rax /* arg */ - - /* Save ip / op / bits */ - movq %ip0, 0(%rax) - movq %ip1, 8(%rax) - movq %ip2, 16(%rax) - movq %ip3, 24(%rax) - movq %op0, 32(%rax) - movq %op1, 40(%rax) - movq %op2, 48(%rax) - movq %op3, 56(%rax) - movq %bits0, 64(%rax) - movq %bits1, 72(%rax) - movq %bits2, 80(%rax) - movq %bits3, 88(%rax) - - /* Restore registers */ - pop %r15 - pop %r14 - pop %r13 - pop %r12 - pop %r11 - pop %r10 - pop %r9 - pop %r8 - pop %rdi - pop %rsi - pop %rbp - pop %rdx - pop %rcx - pop %rbx - pop %rax - ret - -_HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop: -HUF_decompress4X2_usingDTable_internal_bmi2_asm_loop: - /* Save all registers - even if they are callee saved for simplicity. */ - push %rax - push %rbx - push %rcx - push %rdx - push %rbp - push %rsi - push %rdi - push %r8 - push %r9 - push %r10 - push %r11 - push %r12 - push %r13 - push %r14 - push %r15 - - movq %rdi, %rax - movq 0(%rax), %ip0 - movq 8(%rax), %ip1 - movq 16(%rax), %ip2 - movq 24(%rax), %ip3 - movq 32(%rax), %op0 - movq 40(%rax), %op1 - movq 48(%rax), %op2 - movq 56(%rax), %op3 - movq 64(%rax), %bits0 - movq 72(%rax), %bits1 - movq 80(%rax), %bits2 - movq 88(%rax), %bits3 - movq 96(%rax), %dtable - push %rax /* argument */ - push %rax /* olimit */ - push 104(%rax) /* ilimit */ - - movq 112(%rax), %rax - push %rax /* oend3 */ - - movq %op3, %rax - push %rax /* oend2 */ - - movq %op2, %rax - push %rax /* oend1 */ - - movq %op1, %rax - push %rax /* oend0 */ - - /* Scratch space */ - subq $8, %rsp - -.L_4X2_compute_olimit: - /* Computes how many iterations we can do safely - * %r15, %rax may be clobbered - * rdx must be saved - * op[1,2,3,4] & ip0 mustn't be clobbered - */ - movq %rdx, 0(%rsp) - - /* We can consume up to 7 input bytes each iteration. */ - movq %ip0, %rax /* rax = ip0 */ - movq 40(%rsp), %rdx /* rdx = ilimit */ - subq %rdx, %rax /* rax = ip0 - ilimit */ - movq %rax, %r15 /* r15 = ip0 - ilimit */ - - /* rdx = rax / 7 */ - movabsq $2635249153387078803, %rdx - mulq %rdx - subq %rdx, %r15 - shrq %r15 - addq %r15, %rdx - shrq $2, %rdx - - /* r15 = (ip0 - ilimit) / 7 */ - movq %rdx, %r15 - - movabsq $-3689348814741910323, %rdx - movq 8(%rsp), %rax /* rax = oend0 */ - subq %op0, %rax /* rax = oend0 - op0 */ - mulq %rdx - shrq $3, %rdx /* rdx = rax / 10 */ - - /* r15 = min(%rdx, %r15) */ - cmpq %rdx, %r15 - cmova %rdx, %r15 - - movabsq $-3689348814741910323, %rdx - movq 16(%rsp), %rax /* rax = oend1 */ - subq %op1, %rax /* rax = oend1 - op1 */ - mulq %rdx - shrq $3, %rdx /* rdx = rax / 10 */ - - /* r15 = min(%rdx, %r15) */ - cmpq %rdx, %r15 - cmova %rdx, %r15 - - movabsq $-3689348814741910323, %rdx - movq 24(%rsp), %rax /* rax = oend2 */ - subq %op2, %rax /* rax = oend2 - op2 */ - mulq %rdx - shrq $3, %rdx /* rdx = rax / 10 */ - - /* r15 = min(%rdx, %r15) */ - cmpq %rdx, %r15 - cmova %rdx, %r15 - - movabsq $-3689348814741910323, %rdx - movq 32(%rsp), %rax /* rax = oend3 */ - subq %op3, %rax /* rax = oend3 - op3 */ - mulq %rdx - shrq $3, %rdx /* rdx = rax / 10 */ - - /* r15 = min(%rdx, %r15) */ - cmpq %rdx, %r15 - cmova %rdx, %r15 - - /* olimit = op3 + 5 * r15 */ - movq %r15, %rax - leaq (%op3, %rax, 4), %olimit - addq %rax, %olimit - - movq 0(%rsp), %rdx - - /* If (op3 + 10 > olimit) */ - movq %op3, %rax /* rax = op3 */ - addq $10, %rax /* rax = op3 + 10 */ - cmpq %rax, %olimit /* op3 + 10 > olimit */ - jb .L_4X2_exit - - /* If (ip1 < ip0) go to exit */ - cmpq %ip0, %ip1 - jb .L_4X2_exit - - /* If (ip2 < ip1) go to exit */ - cmpq %ip1, %ip2 - jb .L_4X2_exit - - /* If (ip3 < ip2) go to exit */ - cmpq %ip2, %ip3 - jb .L_4X2_exit - -#define DECODE(n, idx) \ - movq %bits##n, %rax; \ - shrq $53, %rax; \ - movzwl 0(%dtable,%rax,4),%r8d; \ - movzbl 2(%dtable,%rax,4),%r15d; \ - movzbl 3(%dtable,%rax,4),%eax; \ - movw %r8w, (%op##n); \ - shlxq %r15, %bits##n, %bits##n; \ - addq %rax, %op##n - -#define RELOAD_BITS(n) \ - bsfq %bits##n, %bits##n; \ - movq %bits##n, %rax; \ - shrq $3, %bits##n; \ - andq $7, %rax; \ - subq %bits##n, %ip##n; \ - movq (%ip##n), %bits##n; \ - orq $1, %bits##n; \ - shlxq %rax, %bits##n, %bits##n - - - movq %olimit, 48(%rsp) - - .p2align 6 - -.L_4X2_loop_body: - /* We clobber r8, so store it on the stack */ - movq %r8, 0(%rsp) - - /* Decode 5 symbols from each of the 4 streams (20 symbols total). */ - FOR_EACH_STREAM_WITH_INDEX(DECODE, 0) - FOR_EACH_STREAM_WITH_INDEX(DECODE, 1) - FOR_EACH_STREAM_WITH_INDEX(DECODE, 2) - FOR_EACH_STREAM_WITH_INDEX(DECODE, 3) - FOR_EACH_STREAM_WITH_INDEX(DECODE, 4) - - /* Reload r8 */ - movq 0(%rsp), %r8 - - FOR_EACH_STREAM(RELOAD_BITS) - - cmp %op3, 48(%rsp) - ja .L_4X2_loop_body - jmp .L_4X2_compute_olimit - -#undef DECODE -#undef RELOAD_BITS -.L_4X2_exit: - addq $8, %rsp - /* Restore stack (oend & olimit) */ - pop %rax /* oend0 */ - pop %rax /* oend1 */ - pop %rax /* oend2 */ - pop %rax /* oend3 */ - pop %rax /* ilimit */ - pop %rax /* olimit */ - pop %rax /* arg */ - - /* Save ip / op / bits */ - movq %ip0, 0(%rax) - movq %ip1, 8(%rax) - movq %ip2, 16(%rax) - movq %ip3, 24(%rax) - movq %op0, 32(%rax) - movq %op1, 40(%rax) - movq %op2, 48(%rax) - movq %op3, 56(%rax) - movq %bits0, 64(%rax) - movq %bits1, 72(%rax) - movq %bits2, 80(%rax) - movq %bits3, 88(%rax) - - /* Restore registers */ - pop %r15 - pop %r14 - pop %r13 - pop %r12 - pop %r11 - pop %r10 - pop %r9 - pop %r8 - pop %rdi - pop %rsi - pop %rbp - pop %rdx - pop %rcx - pop %rbx - pop %rax - ret - -#endif diff --git a/dep/zstd/lib/decompress/zstd_ddict.c b/dep/zstd/lib/decompress/zstd_ddict.c deleted file mode 100644 index ce335477b..000000000 --- a/dep/zstd/lib/decompress/zstd_ddict.c +++ /dev/null @@ -1,244 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -/* zstd_ddict.c : - * concentrates all logic that needs to know the internals of ZSTD_DDict object */ - -/*-******************************************************* -* Dependencies -*********************************************************/ -#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */ -#include "../common/cpu.h" /* bmi2 */ -#include "../common/mem.h" /* low level memory routines */ -#define FSE_STATIC_LINKING_ONLY -#include "../common/fse.h" -#define HUF_STATIC_LINKING_ONLY -#include "../common/huf.h" -#include "zstd_decompress_internal.h" -#include "zstd_ddict.h" - -#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1) -# include "../legacy/zstd_legacy.h" -#endif - - - -/*-******************************************************* -* Types -*********************************************************/ -struct ZSTD_DDict_s { - void* dictBuffer; - const void* dictContent; - size_t dictSize; - ZSTD_entropyDTables_t entropy; - U32 dictID; - U32 entropyPresent; - ZSTD_customMem cMem; -}; /* typedef'd to ZSTD_DDict within "zstd.h" */ - -const void* ZSTD_DDict_dictContent(const ZSTD_DDict* ddict) -{ - assert(ddict != NULL); - return ddict->dictContent; -} - -size_t ZSTD_DDict_dictSize(const ZSTD_DDict* ddict) -{ - assert(ddict != NULL); - return ddict->dictSize; -} - -void ZSTD_copyDDictParameters(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict) -{ - DEBUGLOG(4, "ZSTD_copyDDictParameters"); - assert(dctx != NULL); - assert(ddict != NULL); - dctx->dictID = ddict->dictID; - dctx->prefixStart = ddict->dictContent; - dctx->virtualStart = ddict->dictContent; - dctx->dictEnd = (const BYTE*)ddict->dictContent + ddict->dictSize; - dctx->previousDstEnd = dctx->dictEnd; -#ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION - dctx->dictContentBeginForFuzzing = dctx->prefixStart; - dctx->dictContentEndForFuzzing = dctx->previousDstEnd; -#endif - if (ddict->entropyPresent) { - dctx->litEntropy = 1; - dctx->fseEntropy = 1; - dctx->LLTptr = ddict->entropy.LLTable; - dctx->MLTptr = ddict->entropy.MLTable; - dctx->OFTptr = ddict->entropy.OFTable; - dctx->HUFptr = ddict->entropy.hufTable; - dctx->entropy.rep[0] = ddict->entropy.rep[0]; - dctx->entropy.rep[1] = ddict->entropy.rep[1]; - dctx->entropy.rep[2] = ddict->entropy.rep[2]; - } else { - dctx->litEntropy = 0; - dctx->fseEntropy = 0; - } -} - - -static size_t -ZSTD_loadEntropy_intoDDict(ZSTD_DDict* ddict, - ZSTD_dictContentType_e dictContentType) -{ - ddict->dictID = 0; - ddict->entropyPresent = 0; - if (dictContentType == ZSTD_dct_rawContent) return 0; - - if (ddict->dictSize < 8) { - if (dictContentType == ZSTD_dct_fullDict) - return ERROR(dictionary_corrupted); /* only accept specified dictionaries */ - return 0; /* pure content mode */ - } - { U32 const magic = MEM_readLE32(ddict->dictContent); - if (magic != ZSTD_MAGIC_DICTIONARY) { - if (dictContentType == ZSTD_dct_fullDict) - return ERROR(dictionary_corrupted); /* only accept specified dictionaries */ - return 0; /* pure content mode */ - } - } - ddict->dictID = MEM_readLE32((const char*)ddict->dictContent + ZSTD_FRAMEIDSIZE); - - /* load entropy tables */ - RETURN_ERROR_IF(ZSTD_isError(ZSTD_loadDEntropy( - &ddict->entropy, ddict->dictContent, ddict->dictSize)), - dictionary_corrupted, ""); - ddict->entropyPresent = 1; - return 0; -} - - -static size_t ZSTD_initDDict_internal(ZSTD_DDict* ddict, - const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType) -{ - if ((dictLoadMethod == ZSTD_dlm_byRef) || (!dict) || (!dictSize)) { - ddict->dictBuffer = NULL; - ddict->dictContent = dict; - if (!dict) dictSize = 0; - } else { - void* const internalBuffer = ZSTD_customMalloc(dictSize, ddict->cMem); - ddict->dictBuffer = internalBuffer; - ddict->dictContent = internalBuffer; - if (!internalBuffer) return ERROR(memory_allocation); - ZSTD_memcpy(internalBuffer, dict, dictSize); - } - ddict->dictSize = dictSize; - ddict->entropy.hufTable[0] = (HUF_DTable)((HufLog)*0x1000001); /* cover both little and big endian */ - - /* parse dictionary content */ - FORWARD_IF_ERROR( ZSTD_loadEntropy_intoDDict(ddict, dictContentType) , ""); - - return 0; -} - -ZSTD_DDict* ZSTD_createDDict_advanced(const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - ZSTD_customMem customMem) -{ - if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; - - { ZSTD_DDict* const ddict = (ZSTD_DDict*) ZSTD_customMalloc(sizeof(ZSTD_DDict), customMem); - if (ddict == NULL) return NULL; - ddict->cMem = customMem; - { size_t const initResult = ZSTD_initDDict_internal(ddict, - dict, dictSize, - dictLoadMethod, dictContentType); - if (ZSTD_isError(initResult)) { - ZSTD_freeDDict(ddict); - return NULL; - } } - return ddict; - } -} - -/*! ZSTD_createDDict() : -* Create a digested dictionary, to start decompression without startup delay. -* `dict` content is copied inside DDict. -* Consequently, `dict` can be released after `ZSTD_DDict` creation */ -ZSTD_DDict* ZSTD_createDDict(const void* dict, size_t dictSize) -{ - ZSTD_customMem const allocator = { NULL, NULL, NULL }; - return ZSTD_createDDict_advanced(dict, dictSize, ZSTD_dlm_byCopy, ZSTD_dct_auto, allocator); -} - -/*! ZSTD_createDDict_byReference() : - * Create a digested dictionary, to start decompression without startup delay. - * Dictionary content is simply referenced, it will be accessed during decompression. - * Warning : dictBuffer must outlive DDict (DDict must be freed before dictBuffer) */ -ZSTD_DDict* ZSTD_createDDict_byReference(const void* dictBuffer, size_t dictSize) -{ - ZSTD_customMem const allocator = { NULL, NULL, NULL }; - return ZSTD_createDDict_advanced(dictBuffer, dictSize, ZSTD_dlm_byRef, ZSTD_dct_auto, allocator); -} - - -const ZSTD_DDict* ZSTD_initStaticDDict( - void* sBuffer, size_t sBufferSize, - const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType) -{ - size_t const neededSpace = sizeof(ZSTD_DDict) - + (dictLoadMethod == ZSTD_dlm_byRef ? 0 : dictSize); - ZSTD_DDict* const ddict = (ZSTD_DDict*)sBuffer; - assert(sBuffer != NULL); - assert(dict != NULL); - if ((size_t)sBuffer & 7) return NULL; /* 8-aligned */ - if (sBufferSize < neededSpace) return NULL; - if (dictLoadMethod == ZSTD_dlm_byCopy) { - ZSTD_memcpy(ddict+1, dict, dictSize); /* local copy */ - dict = ddict+1; - } - if (ZSTD_isError( ZSTD_initDDict_internal(ddict, - dict, dictSize, - ZSTD_dlm_byRef, dictContentType) )) - return NULL; - return ddict; -} - - -size_t ZSTD_freeDDict(ZSTD_DDict* ddict) -{ - if (ddict==NULL) return 0; /* support free on NULL */ - { ZSTD_customMem const cMem = ddict->cMem; - ZSTD_customFree(ddict->dictBuffer, cMem); - ZSTD_customFree(ddict, cMem); - return 0; - } -} - -/*! ZSTD_estimateDDictSize() : - * Estimate amount of memory that will be needed to create a dictionary for decompression. - * Note : dictionary created by reference using ZSTD_dlm_byRef are smaller */ -size_t ZSTD_estimateDDictSize(size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod) -{ - return sizeof(ZSTD_DDict) + (dictLoadMethod == ZSTD_dlm_byRef ? 0 : dictSize); -} - -size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict) -{ - if (ddict==NULL) return 0; /* support sizeof on NULL */ - return sizeof(*ddict) + (ddict->dictBuffer ? ddict->dictSize : 0) ; -} - -/*! ZSTD_getDictID_fromDDict() : - * Provides the dictID of the dictionary loaded into `ddict`. - * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty. - * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */ -unsigned ZSTD_getDictID_fromDDict(const ZSTD_DDict* ddict) -{ - if (ddict==NULL) return 0; - return ZSTD_getDictID_fromDict(ddict->dictContent, ddict->dictSize); -} diff --git a/dep/zstd/lib/decompress/zstd_ddict.h b/dep/zstd/lib/decompress/zstd_ddict.h deleted file mode 100644 index bd03268b5..000000000 --- a/dep/zstd/lib/decompress/zstd_ddict.h +++ /dev/null @@ -1,44 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -#ifndef ZSTD_DDICT_H -#define ZSTD_DDICT_H - -/*-******************************************************* - * Dependencies - *********************************************************/ -#include "../common/zstd_deps.h" /* size_t */ -#include "../zstd.h" /* ZSTD_DDict, and several public functions */ - - -/*-******************************************************* - * Interface - *********************************************************/ - -/* note: several prototypes are already published in `zstd.h` : - * ZSTD_createDDict() - * ZSTD_createDDict_byReference() - * ZSTD_createDDict_advanced() - * ZSTD_freeDDict() - * ZSTD_initStaticDDict() - * ZSTD_sizeof_DDict() - * ZSTD_estimateDDictSize() - * ZSTD_getDictID_fromDict() - */ - -const void* ZSTD_DDict_dictContent(const ZSTD_DDict* ddict); -size_t ZSTD_DDict_dictSize(const ZSTD_DDict* ddict); - -void ZSTD_copyDDictParameters(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict); - - - -#endif /* ZSTD_DDICT_H */ diff --git a/dep/zstd/lib/decompress/zstd_decompress.c b/dep/zstd/lib/decompress/zstd_decompress.c deleted file mode 100644 index 0031e98cf..000000000 --- a/dep/zstd/lib/decompress/zstd_decompress.c +++ /dev/null @@ -1,2230 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -/* *************************************************************** -* Tuning parameters -*****************************************************************/ -/*! - * HEAPMODE : - * Select how default decompression function ZSTD_decompress() allocates its context, - * on stack (0), or into heap (1, default; requires malloc()). - * Note that functions with explicit context such as ZSTD_decompressDCtx() are unaffected. - */ -#ifndef ZSTD_HEAPMODE -# define ZSTD_HEAPMODE 1 -#endif - -/*! -* LEGACY_SUPPORT : -* if set to 1+, ZSTD_decompress() can decode older formats (v0.1+) -*/ -#ifndef ZSTD_LEGACY_SUPPORT -# define ZSTD_LEGACY_SUPPORT 0 -#endif - -/*! - * MAXWINDOWSIZE_DEFAULT : - * maximum window size accepted by DStream __by default__. - * Frames requiring more memory will be rejected. - * It's possible to set a different limit using ZSTD_DCtx_setMaxWindowSize(). - */ -#ifndef ZSTD_MAXWINDOWSIZE_DEFAULT -# define ZSTD_MAXWINDOWSIZE_DEFAULT (((U32)1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT) + 1) -#endif - -/*! - * NO_FORWARD_PROGRESS_MAX : - * maximum allowed nb of calls to ZSTD_decompressStream() - * without any forward progress - * (defined as: no byte read from input, and no byte flushed to output) - * before triggering an error. - */ -#ifndef ZSTD_NO_FORWARD_PROGRESS_MAX -# define ZSTD_NO_FORWARD_PROGRESS_MAX 16 -#endif - - -/*-******************************************************* -* Dependencies -*********************************************************/ -#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */ -#include "../common/mem.h" /* low level memory routines */ -#define FSE_STATIC_LINKING_ONLY -#include "../common/fse.h" -#define HUF_STATIC_LINKING_ONLY -#include "../common/huf.h" -#include "../common/xxhash.h" /* XXH64_reset, XXH64_update, XXH64_digest, XXH64 */ -#include "../common/zstd_internal.h" /* blockProperties_t */ -#include "zstd_decompress_internal.h" /* ZSTD_DCtx */ -#include "zstd_ddict.h" /* ZSTD_DDictDictContent */ -#include "zstd_decompress_block.h" /* ZSTD_decompressBlock_internal */ - -#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1) -# include "../legacy/zstd_legacy.h" -#endif - - - -/************************************* - * Multiple DDicts Hashset internals * - *************************************/ - -#define DDICT_HASHSET_MAX_LOAD_FACTOR_COUNT_MULT 4 -#define DDICT_HASHSET_MAX_LOAD_FACTOR_SIZE_MULT 3 /* These two constants represent SIZE_MULT/COUNT_MULT load factor without using a float. - * Currently, that means a 0.75 load factor. - * So, if count * COUNT_MULT / size * SIZE_MULT != 0, then we've exceeded - * the load factor of the ddict hash set. - */ - -#define DDICT_HASHSET_TABLE_BASE_SIZE 64 -#define DDICT_HASHSET_RESIZE_FACTOR 2 - -/* Hash function to determine starting position of dict insertion within the table - * Returns an index between [0, hashSet->ddictPtrTableSize] - */ -static size_t ZSTD_DDictHashSet_getIndex(const ZSTD_DDictHashSet* hashSet, U32 dictID) { - const U64 hash = XXH64(&dictID, sizeof(U32), 0); - /* DDict ptr table size is a multiple of 2, use size - 1 as mask to get index within [0, hashSet->ddictPtrTableSize) */ - return hash & (hashSet->ddictPtrTableSize - 1); -} - -/* Adds DDict to a hashset without resizing it. - * If inserting a DDict with a dictID that already exists in the set, replaces the one in the set. - * Returns 0 if successful, or a zstd error code if something went wrong. - */ -static size_t ZSTD_DDictHashSet_emplaceDDict(ZSTD_DDictHashSet* hashSet, const ZSTD_DDict* ddict) { - const U32 dictID = ZSTD_getDictID_fromDDict(ddict); - size_t idx = ZSTD_DDictHashSet_getIndex(hashSet, dictID); - const size_t idxRangeMask = hashSet->ddictPtrTableSize - 1; - RETURN_ERROR_IF(hashSet->ddictPtrCount == hashSet->ddictPtrTableSize, GENERIC, "Hash set is full!"); - DEBUGLOG(4, "Hashed index: for dictID: %u is %zu", dictID, idx); - while (hashSet->ddictPtrTable[idx] != NULL) { - /* Replace existing ddict if inserting ddict with same dictID */ - if (ZSTD_getDictID_fromDDict(hashSet->ddictPtrTable[idx]) == dictID) { - DEBUGLOG(4, "DictID already exists, replacing rather than adding"); - hashSet->ddictPtrTable[idx] = ddict; - return 0; - } - idx &= idxRangeMask; - idx++; - } - DEBUGLOG(4, "Final idx after probing for dictID %u is: %zu", dictID, idx); - hashSet->ddictPtrTable[idx] = ddict; - hashSet->ddictPtrCount++; - return 0; -} - -/* Expands hash table by factor of DDICT_HASHSET_RESIZE_FACTOR and - * rehashes all values, allocates new table, frees old table. - * Returns 0 on success, otherwise a zstd error code. - */ -static size_t ZSTD_DDictHashSet_expand(ZSTD_DDictHashSet* hashSet, ZSTD_customMem customMem) { - size_t newTableSize = hashSet->ddictPtrTableSize * DDICT_HASHSET_RESIZE_FACTOR; - const ZSTD_DDict** newTable = (const ZSTD_DDict**)ZSTD_customCalloc(sizeof(ZSTD_DDict*) * newTableSize, customMem); - const ZSTD_DDict** oldTable = hashSet->ddictPtrTable; - size_t oldTableSize = hashSet->ddictPtrTableSize; - size_t i; - - DEBUGLOG(4, "Expanding DDict hash table! Old size: %zu new size: %zu", oldTableSize, newTableSize); - RETURN_ERROR_IF(!newTable, memory_allocation, "Expanded hashset allocation failed!"); - hashSet->ddictPtrTable = newTable; - hashSet->ddictPtrTableSize = newTableSize; - hashSet->ddictPtrCount = 0; - for (i = 0; i < oldTableSize; ++i) { - if (oldTable[i] != NULL) { - FORWARD_IF_ERROR(ZSTD_DDictHashSet_emplaceDDict(hashSet, oldTable[i]), ""); - } - } - ZSTD_customFree((void*)oldTable, customMem); - DEBUGLOG(4, "Finished re-hash"); - return 0; -} - -/* Fetches a DDict with the given dictID - * Returns the ZSTD_DDict* with the requested dictID. If it doesn't exist, then returns NULL. - */ -static const ZSTD_DDict* ZSTD_DDictHashSet_getDDict(ZSTD_DDictHashSet* hashSet, U32 dictID) { - size_t idx = ZSTD_DDictHashSet_getIndex(hashSet, dictID); - const size_t idxRangeMask = hashSet->ddictPtrTableSize - 1; - DEBUGLOG(4, "Hashed index: for dictID: %u is %zu", dictID, idx); - for (;;) { - size_t currDictID = ZSTD_getDictID_fromDDict(hashSet->ddictPtrTable[idx]); - if (currDictID == dictID || currDictID == 0) { - /* currDictID == 0 implies a NULL ddict entry */ - break; - } else { - idx &= idxRangeMask; /* Goes to start of table when we reach the end */ - idx++; - } - } - DEBUGLOG(4, "Final idx after probing for dictID %u is: %zu", dictID, idx); - return hashSet->ddictPtrTable[idx]; -} - -/* Allocates space for and returns a ddict hash set - * The hash set's ZSTD_DDict* table has all values automatically set to NULL to begin with. - * Returns NULL if allocation failed. - */ -static ZSTD_DDictHashSet* ZSTD_createDDictHashSet(ZSTD_customMem customMem) { - ZSTD_DDictHashSet* ret = (ZSTD_DDictHashSet*)ZSTD_customMalloc(sizeof(ZSTD_DDictHashSet), customMem); - DEBUGLOG(4, "Allocating new hash set"); - if (!ret) - return NULL; - ret->ddictPtrTable = (const ZSTD_DDict**)ZSTD_customCalloc(DDICT_HASHSET_TABLE_BASE_SIZE * sizeof(ZSTD_DDict*), customMem); - if (!ret->ddictPtrTable) { - ZSTD_customFree(ret, customMem); - return NULL; - } - ret->ddictPtrTableSize = DDICT_HASHSET_TABLE_BASE_SIZE; - ret->ddictPtrCount = 0; - return ret; -} - -/* Frees the table of ZSTD_DDict* within a hashset, then frees the hashset itself. - * Note: The ZSTD_DDict* within the table are NOT freed. - */ -static void ZSTD_freeDDictHashSet(ZSTD_DDictHashSet* hashSet, ZSTD_customMem customMem) { - DEBUGLOG(4, "Freeing ddict hash set"); - if (hashSet && hashSet->ddictPtrTable) { - ZSTD_customFree((void*)hashSet->ddictPtrTable, customMem); - } - if (hashSet) { - ZSTD_customFree(hashSet, customMem); - } -} - -/* Public function: Adds a DDict into the ZSTD_DDictHashSet, possibly triggering a resize of the hash set. - * Returns 0 on success, or a ZSTD error. - */ -static size_t ZSTD_DDictHashSet_addDDict(ZSTD_DDictHashSet* hashSet, const ZSTD_DDict* ddict, ZSTD_customMem customMem) { - DEBUGLOG(4, "Adding dict ID: %u to hashset with - Count: %zu Tablesize: %zu", ZSTD_getDictID_fromDDict(ddict), hashSet->ddictPtrCount, hashSet->ddictPtrTableSize); - if (hashSet->ddictPtrCount * DDICT_HASHSET_MAX_LOAD_FACTOR_COUNT_MULT / hashSet->ddictPtrTableSize * DDICT_HASHSET_MAX_LOAD_FACTOR_SIZE_MULT != 0) { - FORWARD_IF_ERROR(ZSTD_DDictHashSet_expand(hashSet, customMem), ""); - } - FORWARD_IF_ERROR(ZSTD_DDictHashSet_emplaceDDict(hashSet, ddict), ""); - return 0; -} - -/*-************************************************************* -* Context management -***************************************************************/ -size_t ZSTD_sizeof_DCtx (const ZSTD_DCtx* dctx) -{ - if (dctx==NULL) return 0; /* support sizeof NULL */ - return sizeof(*dctx) - + ZSTD_sizeof_DDict(dctx->ddictLocal) - + dctx->inBuffSize + dctx->outBuffSize; -} - -size_t ZSTD_estimateDCtxSize(void) { return sizeof(ZSTD_DCtx); } - - -static size_t ZSTD_startingInputLength(ZSTD_format_e format) -{ - size_t const startingInputLength = ZSTD_FRAMEHEADERSIZE_PREFIX(format); - /* only supports formats ZSTD_f_zstd1 and ZSTD_f_zstd1_magicless */ - assert( (format == ZSTD_f_zstd1) || (format == ZSTD_f_zstd1_magicless) ); - return startingInputLength; -} - -static void ZSTD_DCtx_resetParameters(ZSTD_DCtx* dctx) -{ - assert(dctx->streamStage == zdss_init); - dctx->format = ZSTD_f_zstd1; - dctx->maxWindowSize = ZSTD_MAXWINDOWSIZE_DEFAULT; - dctx->outBufferMode = ZSTD_bm_buffered; - dctx->forceIgnoreChecksum = ZSTD_d_validateChecksum; - dctx->refMultipleDDicts = ZSTD_rmd_refSingleDDict; -} - -static void ZSTD_initDCtx_internal(ZSTD_DCtx* dctx) -{ - dctx->staticSize = 0; - dctx->ddict = NULL; - dctx->ddictLocal = NULL; - dctx->dictEnd = NULL; - dctx->ddictIsCold = 0; - dctx->dictUses = ZSTD_dont_use; - dctx->inBuff = NULL; - dctx->inBuffSize = 0; - dctx->outBuffSize = 0; - dctx->streamStage = zdss_init; -#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1) - dctx->legacyContext = NULL; - dctx->previousLegacyVersion = 0; -#endif - dctx->noForwardProgress = 0; - dctx->oversizedDuration = 0; -#if DYNAMIC_BMI2 - dctx->bmi2 = ZSTD_cpuSupportsBmi2(); -#endif - dctx->ddictSet = NULL; - ZSTD_DCtx_resetParameters(dctx); -#ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION - dctx->dictContentEndForFuzzing = NULL; -#endif -} - -ZSTD_DCtx* ZSTD_initStaticDCtx(void *workspace, size_t workspaceSize) -{ - ZSTD_DCtx* const dctx = (ZSTD_DCtx*) workspace; - - if ((size_t)workspace & 7) return NULL; /* 8-aligned */ - if (workspaceSize < sizeof(ZSTD_DCtx)) return NULL; /* minimum size */ - - ZSTD_initDCtx_internal(dctx); - dctx->staticSize = workspaceSize; - dctx->inBuff = (char*)(dctx+1); - return dctx; -} - -static ZSTD_DCtx* ZSTD_createDCtx_internal(ZSTD_customMem customMem) { - if ((!customMem.customAlloc) ^ (!customMem.customFree)) return NULL; - - { ZSTD_DCtx* const dctx = (ZSTD_DCtx*)ZSTD_customMalloc(sizeof(*dctx), customMem); - if (!dctx) return NULL; - dctx->customMem = customMem; - ZSTD_initDCtx_internal(dctx); - return dctx; - } -} - -ZSTD_DCtx* ZSTD_createDCtx_advanced(ZSTD_customMem customMem) -{ - return ZSTD_createDCtx_internal(customMem); -} - -ZSTD_DCtx* ZSTD_createDCtx(void) -{ - DEBUGLOG(3, "ZSTD_createDCtx"); - return ZSTD_createDCtx_internal(ZSTD_defaultCMem); -} - -static void ZSTD_clearDict(ZSTD_DCtx* dctx) -{ - ZSTD_freeDDict(dctx->ddictLocal); - dctx->ddictLocal = NULL; - dctx->ddict = NULL; - dctx->dictUses = ZSTD_dont_use; -} - -size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx) -{ - if (dctx==NULL) return 0; /* support free on NULL */ - RETURN_ERROR_IF(dctx->staticSize, memory_allocation, "not compatible with static DCtx"); - { ZSTD_customMem const cMem = dctx->customMem; - ZSTD_clearDict(dctx); - ZSTD_customFree(dctx->inBuff, cMem); - dctx->inBuff = NULL; -#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1) - if (dctx->legacyContext) - ZSTD_freeLegacyStreamContext(dctx->legacyContext, dctx->previousLegacyVersion); -#endif - if (dctx->ddictSet) { - ZSTD_freeDDictHashSet(dctx->ddictSet, cMem); - dctx->ddictSet = NULL; - } - ZSTD_customFree(dctx, cMem); - return 0; - } -} - -/* no longer useful */ -void ZSTD_copyDCtx(ZSTD_DCtx* dstDCtx, const ZSTD_DCtx* srcDCtx) -{ - size_t const toCopy = (size_t)((char*)(&dstDCtx->inBuff) - (char*)dstDCtx); - ZSTD_memcpy(dstDCtx, srcDCtx, toCopy); /* no need to copy workspace */ -} - -/* Given a dctx with a digested frame params, re-selects the correct ZSTD_DDict based on - * the requested dict ID from the frame. If there exists a reference to the correct ZSTD_DDict, then - * accordingly sets the ddict to be used to decompress the frame. - * - * If no DDict is found, then no action is taken, and the ZSTD_DCtx::ddict remains as-is. - * - * ZSTD_d_refMultipleDDicts must be enabled for this function to be called. - */ -static void ZSTD_DCtx_selectFrameDDict(ZSTD_DCtx* dctx) { - assert(dctx->refMultipleDDicts && dctx->ddictSet); - DEBUGLOG(4, "Adjusting DDict based on requested dict ID from frame"); - if (dctx->ddict) { - const ZSTD_DDict* frameDDict = ZSTD_DDictHashSet_getDDict(dctx->ddictSet, dctx->fParams.dictID); - if (frameDDict) { - DEBUGLOG(4, "DDict found!"); - ZSTD_clearDict(dctx); - dctx->dictID = dctx->fParams.dictID; - dctx->ddict = frameDDict; - dctx->dictUses = ZSTD_use_indefinitely; - } - } -} - - -/*-************************************************************* - * Frame header decoding - ***************************************************************/ - -/*! ZSTD_isFrame() : - * Tells if the content of `buffer` starts with a valid Frame Identifier. - * Note : Frame Identifier is 4 bytes. If `size < 4`, @return will always be 0. - * Note 2 : Legacy Frame Identifiers are considered valid only if Legacy Support is enabled. - * Note 3 : Skippable Frame Identifiers are considered valid. */ -unsigned ZSTD_isFrame(const void* buffer, size_t size) -{ - if (size < ZSTD_FRAMEIDSIZE) return 0; - { U32 const magic = MEM_readLE32(buffer); - if (magic == ZSTD_MAGICNUMBER) return 1; - if ((magic & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) return 1; - } -#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1) - if (ZSTD_isLegacy(buffer, size)) return 1; -#endif - return 0; -} - -/*! ZSTD_isSkippableFrame() : - * Tells if the content of `buffer` starts with a valid Frame Identifier for a skippable frame. - * Note : Frame Identifier is 4 bytes. If `size < 4`, @return will always be 0. - */ -unsigned ZSTD_isSkippableFrame(const void* buffer, size_t size) -{ - if (size < ZSTD_FRAMEIDSIZE) return 0; - { U32 const magic = MEM_readLE32(buffer); - if ((magic & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) return 1; - } - return 0; -} - -/** ZSTD_frameHeaderSize_internal() : - * srcSize must be large enough to reach header size fields. - * note : only works for formats ZSTD_f_zstd1 and ZSTD_f_zstd1_magicless. - * @return : size of the Frame Header - * or an error code, which can be tested with ZSTD_isError() */ -static size_t ZSTD_frameHeaderSize_internal(const void* src, size_t srcSize, ZSTD_format_e format) -{ - size_t const minInputSize = ZSTD_startingInputLength(format); - RETURN_ERROR_IF(srcSize < minInputSize, srcSize_wrong, ""); - - { BYTE const fhd = ((const BYTE*)src)[minInputSize-1]; - U32 const dictID= fhd & 3; - U32 const singleSegment = (fhd >> 5) & 1; - U32 const fcsId = fhd >> 6; - return minInputSize + !singleSegment - + ZSTD_did_fieldSize[dictID] + ZSTD_fcs_fieldSize[fcsId] - + (singleSegment && !fcsId); - } -} - -/** ZSTD_frameHeaderSize() : - * srcSize must be >= ZSTD_frameHeaderSize_prefix. - * @return : size of the Frame Header, - * or an error code (if srcSize is too small) */ -size_t ZSTD_frameHeaderSize(const void* src, size_t srcSize) -{ - return ZSTD_frameHeaderSize_internal(src, srcSize, ZSTD_f_zstd1); -} - - -/** ZSTD_getFrameHeader_advanced() : - * decode Frame Header, or require larger `srcSize`. - * note : only works for formats ZSTD_f_zstd1 and ZSTD_f_zstd1_magicless - * @return : 0, `zfhPtr` is correctly filled, - * >0, `srcSize` is too small, value is wanted `srcSize` amount, - * or an error code, which can be tested using ZSTD_isError() */ -size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize, ZSTD_format_e format) -{ - const BYTE* ip = (const BYTE*)src; - size_t const minInputSize = ZSTD_startingInputLength(format); - - ZSTD_memset(zfhPtr, 0, sizeof(*zfhPtr)); /* not strictly necessary, but static analyzer do not understand that zfhPtr is only going to be read only if return value is zero, since they are 2 different signals */ - if (srcSize < minInputSize) return minInputSize; - RETURN_ERROR_IF(src==NULL, GENERIC, "invalid parameter"); - - if ( (format != ZSTD_f_zstd1_magicless) - && (MEM_readLE32(src) != ZSTD_MAGICNUMBER) ) { - if ((MEM_readLE32(src) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { - /* skippable frame */ - if (srcSize < ZSTD_SKIPPABLEHEADERSIZE) - return ZSTD_SKIPPABLEHEADERSIZE; /* magic number + frame length */ - ZSTD_memset(zfhPtr, 0, sizeof(*zfhPtr)); - zfhPtr->frameContentSize = MEM_readLE32((const char *)src + ZSTD_FRAMEIDSIZE); - zfhPtr->frameType = ZSTD_skippableFrame; - return 0; - } - RETURN_ERROR(prefix_unknown, ""); - } - - /* ensure there is enough `srcSize` to fully read/decode frame header */ - { size_t const fhsize = ZSTD_frameHeaderSize_internal(src, srcSize, format); - if (srcSize < fhsize) return fhsize; - zfhPtr->headerSize = (U32)fhsize; - } - - { BYTE const fhdByte = ip[minInputSize-1]; - size_t pos = minInputSize; - U32 const dictIDSizeCode = fhdByte&3; - U32 const checksumFlag = (fhdByte>>2)&1; - U32 const singleSegment = (fhdByte>>5)&1; - U32 const fcsID = fhdByte>>6; - U64 windowSize = 0; - U32 dictID = 0; - U64 frameContentSize = ZSTD_CONTENTSIZE_UNKNOWN; - RETURN_ERROR_IF((fhdByte & 0x08) != 0, frameParameter_unsupported, - "reserved bits, must be zero"); - - if (!singleSegment) { - BYTE const wlByte = ip[pos++]; - U32 const windowLog = (wlByte >> 3) + ZSTD_WINDOWLOG_ABSOLUTEMIN; - RETURN_ERROR_IF(windowLog > ZSTD_WINDOWLOG_MAX, frameParameter_windowTooLarge, ""); - windowSize = (1ULL << windowLog); - windowSize += (windowSize >> 3) * (wlByte&7); - } - switch(dictIDSizeCode) - { - default: - assert(0); /* impossible */ - ZSTD_FALLTHROUGH; - case 0 : break; - case 1 : dictID = ip[pos]; pos++; break; - case 2 : dictID = MEM_readLE16(ip+pos); pos+=2; break; - case 3 : dictID = MEM_readLE32(ip+pos); pos+=4; break; - } - switch(fcsID) - { - default: - assert(0); /* impossible */ - ZSTD_FALLTHROUGH; - case 0 : if (singleSegment) frameContentSize = ip[pos]; break; - case 1 : frameContentSize = MEM_readLE16(ip+pos)+256; break; - case 2 : frameContentSize = MEM_readLE32(ip+pos); break; - case 3 : frameContentSize = MEM_readLE64(ip+pos); break; - } - if (singleSegment) windowSize = frameContentSize; - - zfhPtr->frameType = ZSTD_frame; - zfhPtr->frameContentSize = frameContentSize; - zfhPtr->windowSize = windowSize; - zfhPtr->blockSizeMax = (unsigned) MIN(windowSize, ZSTD_BLOCKSIZE_MAX); - zfhPtr->dictID = dictID; - zfhPtr->checksumFlag = checksumFlag; - } - return 0; -} - -/** ZSTD_getFrameHeader() : - * decode Frame Header, or require larger `srcSize`. - * note : this function does not consume input, it only reads it. - * @return : 0, `zfhPtr` is correctly filled, - * >0, `srcSize` is too small, value is wanted `srcSize` amount, - * or an error code, which can be tested using ZSTD_isError() */ -size_t ZSTD_getFrameHeader(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize) -{ - return ZSTD_getFrameHeader_advanced(zfhPtr, src, srcSize, ZSTD_f_zstd1); -} - -/** ZSTD_getFrameContentSize() : - * compatible with legacy mode - * @return : decompressed size of the single frame pointed to be `src` if known, otherwise - * - ZSTD_CONTENTSIZE_UNKNOWN if the size cannot be determined - * - ZSTD_CONTENTSIZE_ERROR if an error occurred (e.g. invalid magic number, srcSize too small) */ -unsigned long long ZSTD_getFrameContentSize(const void *src, size_t srcSize) -{ -#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1) - if (ZSTD_isLegacy(src, srcSize)) { - unsigned long long const ret = ZSTD_getDecompressedSize_legacy(src, srcSize); - return ret == 0 ? ZSTD_CONTENTSIZE_UNKNOWN : ret; - } -#endif - { ZSTD_frameHeader zfh; - if (ZSTD_getFrameHeader(&zfh, src, srcSize) != 0) - return ZSTD_CONTENTSIZE_ERROR; - if (zfh.frameType == ZSTD_skippableFrame) { - return 0; - } else { - return zfh.frameContentSize; - } } -} - -static size_t readSkippableFrameSize(void const* src, size_t srcSize) -{ - size_t const skippableHeaderSize = ZSTD_SKIPPABLEHEADERSIZE; - U32 sizeU32; - - RETURN_ERROR_IF(srcSize < ZSTD_SKIPPABLEHEADERSIZE, srcSize_wrong, ""); - - sizeU32 = MEM_readLE32((BYTE const*)src + ZSTD_FRAMEIDSIZE); - RETURN_ERROR_IF((U32)(sizeU32 + ZSTD_SKIPPABLEHEADERSIZE) < sizeU32, - frameParameter_unsupported, ""); - { - size_t const skippableSize = skippableHeaderSize + sizeU32; - RETURN_ERROR_IF(skippableSize > srcSize, srcSize_wrong, ""); - return skippableSize; - } -} - -/*! ZSTD_readSkippableFrame() : - * Retrieves a zstd skippable frame containing data given by src, and writes it to dst buffer. - * - * The parameter magicVariant will receive the magicVariant that was supplied when the frame was written, - * i.e. magicNumber - ZSTD_MAGIC_SKIPPABLE_START. This can be NULL if the caller is not interested - * in the magicVariant. - * - * Returns an error if destination buffer is not large enough, or if the frame is not skippable. - * - * @return : number of bytes written or a ZSTD error. - */ -ZSTDLIB_API size_t ZSTD_readSkippableFrame(void* dst, size_t dstCapacity, unsigned* magicVariant, - const void* src, size_t srcSize) -{ - U32 const magicNumber = MEM_readLE32(src); - size_t skippableFrameSize = readSkippableFrameSize(src, srcSize); - size_t skippableContentSize = skippableFrameSize - ZSTD_SKIPPABLEHEADERSIZE; - - /* check input validity */ - RETURN_ERROR_IF(!ZSTD_isSkippableFrame(src, srcSize), frameParameter_unsupported, ""); - RETURN_ERROR_IF(skippableFrameSize < ZSTD_SKIPPABLEHEADERSIZE || skippableFrameSize > srcSize, srcSize_wrong, ""); - RETURN_ERROR_IF(skippableContentSize > dstCapacity, dstSize_tooSmall, ""); - - /* deliver payload */ - if (skippableContentSize > 0 && dst != NULL) - ZSTD_memcpy(dst, (const BYTE *)src + ZSTD_SKIPPABLEHEADERSIZE, skippableContentSize); - if (magicVariant != NULL) - *magicVariant = magicNumber - ZSTD_MAGIC_SKIPPABLE_START; - return skippableContentSize; -} - -/** ZSTD_findDecompressedSize() : - * compatible with legacy mode - * `srcSize` must be the exact length of some number of ZSTD compressed and/or - * skippable frames - * @return : decompressed size of the frames contained */ -unsigned long long ZSTD_findDecompressedSize(const void* src, size_t srcSize) -{ - unsigned long long totalDstSize = 0; - - while (srcSize >= ZSTD_startingInputLength(ZSTD_f_zstd1)) { - U32 const magicNumber = MEM_readLE32(src); - - if ((magicNumber & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { - size_t const skippableSize = readSkippableFrameSize(src, srcSize); - if (ZSTD_isError(skippableSize)) { - return ZSTD_CONTENTSIZE_ERROR; - } - assert(skippableSize <= srcSize); - - src = (const BYTE *)src + skippableSize; - srcSize -= skippableSize; - continue; - } - - { unsigned long long const ret = ZSTD_getFrameContentSize(src, srcSize); - if (ret >= ZSTD_CONTENTSIZE_ERROR) return ret; - - /* check for overflow */ - if (totalDstSize + ret < totalDstSize) return ZSTD_CONTENTSIZE_ERROR; - totalDstSize += ret; - } - { size_t const frameSrcSize = ZSTD_findFrameCompressedSize(src, srcSize); - if (ZSTD_isError(frameSrcSize)) { - return ZSTD_CONTENTSIZE_ERROR; - } - - src = (const BYTE *)src + frameSrcSize; - srcSize -= frameSrcSize; - } - } /* while (srcSize >= ZSTD_frameHeaderSize_prefix) */ - - if (srcSize) return ZSTD_CONTENTSIZE_ERROR; - - return totalDstSize; -} - -/** ZSTD_getDecompressedSize() : - * compatible with legacy mode - * @return : decompressed size if known, 0 otherwise - note : 0 can mean any of the following : - - frame content is empty - - decompressed size field is not present in frame header - - frame header unknown / not supported - - frame header not complete (`srcSize` too small) */ -unsigned long long ZSTD_getDecompressedSize(const void* src, size_t srcSize) -{ - unsigned long long const ret = ZSTD_getFrameContentSize(src, srcSize); - ZSTD_STATIC_ASSERT(ZSTD_CONTENTSIZE_ERROR < ZSTD_CONTENTSIZE_UNKNOWN); - return (ret >= ZSTD_CONTENTSIZE_ERROR) ? 0 : ret; -} - - -/** ZSTD_decodeFrameHeader() : - * `headerSize` must be the size provided by ZSTD_frameHeaderSize(). - * If multiple DDict references are enabled, also will choose the correct DDict to use. - * @return : 0 if success, or an error code, which can be tested using ZSTD_isError() */ -static size_t ZSTD_decodeFrameHeader(ZSTD_DCtx* dctx, const void* src, size_t headerSize) -{ - size_t const result = ZSTD_getFrameHeader_advanced(&(dctx->fParams), src, headerSize, dctx->format); - if (ZSTD_isError(result)) return result; /* invalid header */ - RETURN_ERROR_IF(result>0, srcSize_wrong, "headerSize too small"); - - /* Reference DDict requested by frame if dctx references multiple ddicts */ - if (dctx->refMultipleDDicts == ZSTD_rmd_refMultipleDDicts && dctx->ddictSet) { - ZSTD_DCtx_selectFrameDDict(dctx); - } - -#ifndef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION - /* Skip the dictID check in fuzzing mode, because it makes the search - * harder. - */ - RETURN_ERROR_IF(dctx->fParams.dictID && (dctx->dictID != dctx->fParams.dictID), - dictionary_wrong, ""); -#endif - dctx->validateChecksum = (dctx->fParams.checksumFlag && !dctx->forceIgnoreChecksum) ? 1 : 0; - if (dctx->validateChecksum) XXH64_reset(&dctx->xxhState, 0); - dctx->processedCSize += headerSize; - return 0; -} - -static ZSTD_frameSizeInfo ZSTD_errorFrameSizeInfo(size_t ret) -{ - ZSTD_frameSizeInfo frameSizeInfo; - frameSizeInfo.compressedSize = ret; - frameSizeInfo.decompressedBound = ZSTD_CONTENTSIZE_ERROR; - return frameSizeInfo; -} - -static ZSTD_frameSizeInfo ZSTD_findFrameSizeInfo(const void* src, size_t srcSize) -{ - ZSTD_frameSizeInfo frameSizeInfo; - ZSTD_memset(&frameSizeInfo, 0, sizeof(ZSTD_frameSizeInfo)); - -#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1) - if (ZSTD_isLegacy(src, srcSize)) - return ZSTD_findFrameSizeInfoLegacy(src, srcSize); -#endif - - if ((srcSize >= ZSTD_SKIPPABLEHEADERSIZE) - && (MEM_readLE32(src) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { - frameSizeInfo.compressedSize = readSkippableFrameSize(src, srcSize); - assert(ZSTD_isError(frameSizeInfo.compressedSize) || - frameSizeInfo.compressedSize <= srcSize); - return frameSizeInfo; - } else { - const BYTE* ip = (const BYTE*)src; - const BYTE* const ipstart = ip; - size_t remainingSize = srcSize; - size_t nbBlocks = 0; - ZSTD_frameHeader zfh; - - /* Extract Frame Header */ - { size_t const ret = ZSTD_getFrameHeader(&zfh, src, srcSize); - if (ZSTD_isError(ret)) - return ZSTD_errorFrameSizeInfo(ret); - if (ret > 0) - return ZSTD_errorFrameSizeInfo(ERROR(srcSize_wrong)); - } - - ip += zfh.headerSize; - remainingSize -= zfh.headerSize; - - /* Iterate over each block */ - while (1) { - blockProperties_t blockProperties; - size_t const cBlockSize = ZSTD_getcBlockSize(ip, remainingSize, &blockProperties); - if (ZSTD_isError(cBlockSize)) - return ZSTD_errorFrameSizeInfo(cBlockSize); - - if (ZSTD_blockHeaderSize + cBlockSize > remainingSize) - return ZSTD_errorFrameSizeInfo(ERROR(srcSize_wrong)); - - ip += ZSTD_blockHeaderSize + cBlockSize; - remainingSize -= ZSTD_blockHeaderSize + cBlockSize; - nbBlocks++; - - if (blockProperties.lastBlock) break; - } - - /* Final frame content checksum */ - if (zfh.checksumFlag) { - if (remainingSize < 4) - return ZSTD_errorFrameSizeInfo(ERROR(srcSize_wrong)); - ip += 4; - } - - frameSizeInfo.compressedSize = (size_t)(ip - ipstart); - frameSizeInfo.decompressedBound = (zfh.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN) - ? zfh.frameContentSize - : nbBlocks * zfh.blockSizeMax; - return frameSizeInfo; - } -} - -/** ZSTD_findFrameCompressedSize() : - * compatible with legacy mode - * `src` must point to the start of a ZSTD frame, ZSTD legacy frame, or skippable frame - * `srcSize` must be at least as large as the frame contained - * @return : the compressed size of the frame starting at `src` */ -size_t ZSTD_findFrameCompressedSize(const void *src, size_t srcSize) -{ - ZSTD_frameSizeInfo const frameSizeInfo = ZSTD_findFrameSizeInfo(src, srcSize); - return frameSizeInfo.compressedSize; -} - -/** ZSTD_decompressBound() : - * compatible with legacy mode - * `src` must point to the start of a ZSTD frame or a skippeable frame - * `srcSize` must be at least as large as the frame contained - * @return : the maximum decompressed size of the compressed source - */ -unsigned long long ZSTD_decompressBound(const void* src, size_t srcSize) -{ - unsigned long long bound = 0; - /* Iterate over each frame */ - while (srcSize > 0) { - ZSTD_frameSizeInfo const frameSizeInfo = ZSTD_findFrameSizeInfo(src, srcSize); - size_t const compressedSize = frameSizeInfo.compressedSize; - unsigned long long const decompressedBound = frameSizeInfo.decompressedBound; - if (ZSTD_isError(compressedSize) || decompressedBound == ZSTD_CONTENTSIZE_ERROR) - return ZSTD_CONTENTSIZE_ERROR; - assert(srcSize >= compressedSize); - src = (const BYTE*)src + compressedSize; - srcSize -= compressedSize; - bound += decompressedBound; - } - return bound; -} - - -/*-************************************************************* - * Frame decoding - ***************************************************************/ - -/** ZSTD_insertBlock() : - * insert `src` block into `dctx` history. Useful to track uncompressed blocks. */ -size_t ZSTD_insertBlock(ZSTD_DCtx* dctx, const void* blockStart, size_t blockSize) -{ - DEBUGLOG(5, "ZSTD_insertBlock: %u bytes", (unsigned)blockSize); - ZSTD_checkContinuity(dctx, blockStart, blockSize); - dctx->previousDstEnd = (const char*)blockStart + blockSize; - return blockSize; -} - - -static size_t ZSTD_copyRawBlock(void* dst, size_t dstCapacity, - const void* src, size_t srcSize) -{ - DEBUGLOG(5, "ZSTD_copyRawBlock"); - RETURN_ERROR_IF(srcSize > dstCapacity, dstSize_tooSmall, ""); - if (dst == NULL) { - if (srcSize == 0) return 0; - RETURN_ERROR(dstBuffer_null, ""); - } - ZSTD_memcpy(dst, src, srcSize); - return srcSize; -} - -static size_t ZSTD_setRleBlock(void* dst, size_t dstCapacity, - BYTE b, - size_t regenSize) -{ - RETURN_ERROR_IF(regenSize > dstCapacity, dstSize_tooSmall, ""); - if (dst == NULL) { - if (regenSize == 0) return 0; - RETURN_ERROR(dstBuffer_null, ""); - } - ZSTD_memset(dst, b, regenSize); - return regenSize; -} - -static void ZSTD_DCtx_trace_end(ZSTD_DCtx const* dctx, U64 uncompressedSize, U64 compressedSize, unsigned streaming) -{ -#if ZSTD_TRACE - if (dctx->traceCtx && ZSTD_trace_decompress_end != NULL) { - ZSTD_Trace trace; - ZSTD_memset(&trace, 0, sizeof(trace)); - trace.version = ZSTD_VERSION_NUMBER; - trace.streaming = streaming; - if (dctx->ddict) { - trace.dictionaryID = ZSTD_getDictID_fromDDict(dctx->ddict); - trace.dictionarySize = ZSTD_DDict_dictSize(dctx->ddict); - trace.dictionaryIsCold = dctx->ddictIsCold; - } - trace.uncompressedSize = (size_t)uncompressedSize; - trace.compressedSize = (size_t)compressedSize; - trace.dctx = dctx; - ZSTD_trace_decompress_end(dctx->traceCtx, &trace); - } -#else - (void)dctx; - (void)uncompressedSize; - (void)compressedSize; - (void)streaming; -#endif -} - - -/*! ZSTD_decompressFrame() : - * @dctx must be properly initialized - * will update *srcPtr and *srcSizePtr, - * to make *srcPtr progress by one frame. */ -static size_t ZSTD_decompressFrame(ZSTD_DCtx* dctx, - void* dst, size_t dstCapacity, - const void** srcPtr, size_t *srcSizePtr) -{ - const BYTE* const istart = (const BYTE*)(*srcPtr); - const BYTE* ip = istart; - BYTE* const ostart = (BYTE*)dst; - BYTE* const oend = dstCapacity != 0 ? ostart + dstCapacity : ostart; - BYTE* op = ostart; - size_t remainingSrcSize = *srcSizePtr; - - DEBUGLOG(4, "ZSTD_decompressFrame (srcSize:%i)", (int)*srcSizePtr); - - /* check */ - RETURN_ERROR_IF( - remainingSrcSize < ZSTD_FRAMEHEADERSIZE_MIN(dctx->format)+ZSTD_blockHeaderSize, - srcSize_wrong, ""); - - /* Frame Header */ - { size_t const frameHeaderSize = ZSTD_frameHeaderSize_internal( - ip, ZSTD_FRAMEHEADERSIZE_PREFIX(dctx->format), dctx->format); - if (ZSTD_isError(frameHeaderSize)) return frameHeaderSize; - RETURN_ERROR_IF(remainingSrcSize < frameHeaderSize+ZSTD_blockHeaderSize, - srcSize_wrong, ""); - FORWARD_IF_ERROR( ZSTD_decodeFrameHeader(dctx, ip, frameHeaderSize) , ""); - ip += frameHeaderSize; remainingSrcSize -= frameHeaderSize; - } - - /* Loop on each block */ - while (1) { - size_t decodedSize; - blockProperties_t blockProperties; - size_t const cBlockSize = ZSTD_getcBlockSize(ip, remainingSrcSize, &blockProperties); - if (ZSTD_isError(cBlockSize)) return cBlockSize; - - ip += ZSTD_blockHeaderSize; - remainingSrcSize -= ZSTD_blockHeaderSize; - RETURN_ERROR_IF(cBlockSize > remainingSrcSize, srcSize_wrong, ""); - - switch(blockProperties.blockType) - { - case bt_compressed: - decodedSize = ZSTD_decompressBlock_internal(dctx, op, (size_t)(oend-op), ip, cBlockSize, /* frame */ 1, not_streaming); - break; - case bt_raw : - decodedSize = ZSTD_copyRawBlock(op, (size_t)(oend-op), ip, cBlockSize); - break; - case bt_rle : - decodedSize = ZSTD_setRleBlock(op, (size_t)(oend-op), *ip, blockProperties.origSize); - break; - case bt_reserved : - default: - RETURN_ERROR(corruption_detected, "invalid block type"); - } - - if (ZSTD_isError(decodedSize)) return decodedSize; - if (dctx->validateChecksum) - XXH64_update(&dctx->xxhState, op, decodedSize); - if (decodedSize != 0) - op += decodedSize; - assert(ip != NULL); - ip += cBlockSize; - remainingSrcSize -= cBlockSize; - if (blockProperties.lastBlock) break; - } - - if (dctx->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN) { - RETURN_ERROR_IF((U64)(op-ostart) != dctx->fParams.frameContentSize, - corruption_detected, ""); - } - if (dctx->fParams.checksumFlag) { /* Frame content checksum verification */ - RETURN_ERROR_IF(remainingSrcSize<4, checksum_wrong, ""); - if (!dctx->forceIgnoreChecksum) { - U32 const checkCalc = (U32)XXH64_digest(&dctx->xxhState); - U32 checkRead; - checkRead = MEM_readLE32(ip); - RETURN_ERROR_IF(checkRead != checkCalc, checksum_wrong, ""); - } - ip += 4; - remainingSrcSize -= 4; - } - ZSTD_DCtx_trace_end(dctx, (U64)(op-ostart), (U64)(ip-istart), /* streaming */ 0); - /* Allow caller to get size read */ - *srcPtr = ip; - *srcSizePtr = remainingSrcSize; - return (size_t)(op-ostart); -} - -static size_t ZSTD_decompressMultiFrame(ZSTD_DCtx* dctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const void* dict, size_t dictSize, - const ZSTD_DDict* ddict) -{ - void* const dststart = dst; - int moreThan1Frame = 0; - - DEBUGLOG(5, "ZSTD_decompressMultiFrame"); - assert(dict==NULL || ddict==NULL); /* either dict or ddict set, not both */ - - if (ddict) { - dict = ZSTD_DDict_dictContent(ddict); - dictSize = ZSTD_DDict_dictSize(ddict); - } - - while (srcSize >= ZSTD_startingInputLength(dctx->format)) { - -#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT >= 1) - if (ZSTD_isLegacy(src, srcSize)) { - size_t decodedSize; - size_t const frameSize = ZSTD_findFrameCompressedSizeLegacy(src, srcSize); - if (ZSTD_isError(frameSize)) return frameSize; - RETURN_ERROR_IF(dctx->staticSize, memory_allocation, - "legacy support is not compatible with static dctx"); - - decodedSize = ZSTD_decompressLegacy(dst, dstCapacity, src, frameSize, dict, dictSize); - if (ZSTD_isError(decodedSize)) return decodedSize; - - assert(decodedSize <= dstCapacity); - dst = (BYTE*)dst + decodedSize; - dstCapacity -= decodedSize; - - src = (const BYTE*)src + frameSize; - srcSize -= frameSize; - - continue; - } -#endif - - { U32 const magicNumber = MEM_readLE32(src); - DEBUGLOG(4, "reading magic number %08X (expecting %08X)", - (unsigned)magicNumber, ZSTD_MAGICNUMBER); - if ((magicNumber & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { - size_t const skippableSize = readSkippableFrameSize(src, srcSize); - FORWARD_IF_ERROR(skippableSize, "readSkippableFrameSize failed"); - assert(skippableSize <= srcSize); - - src = (const BYTE *)src + skippableSize; - srcSize -= skippableSize; - continue; - } } - - if (ddict) { - /* we were called from ZSTD_decompress_usingDDict */ - FORWARD_IF_ERROR(ZSTD_decompressBegin_usingDDict(dctx, ddict), ""); - } else { - /* this will initialize correctly with no dict if dict == NULL, so - * use this in all cases but ddict */ - FORWARD_IF_ERROR(ZSTD_decompressBegin_usingDict(dctx, dict, dictSize), ""); - } - ZSTD_checkContinuity(dctx, dst, dstCapacity); - - { const size_t res = ZSTD_decompressFrame(dctx, dst, dstCapacity, - &src, &srcSize); - RETURN_ERROR_IF( - (ZSTD_getErrorCode(res) == ZSTD_error_prefix_unknown) - && (moreThan1Frame==1), - srcSize_wrong, - "At least one frame successfully completed, " - "but following bytes are garbage: " - "it's more likely to be a srcSize error, " - "specifying more input bytes than size of frame(s). " - "Note: one could be unlucky, it might be a corruption error instead, " - "happening right at the place where we expect zstd magic bytes. " - "But this is _much_ less likely than a srcSize field error."); - if (ZSTD_isError(res)) return res; - assert(res <= dstCapacity); - if (res != 0) - dst = (BYTE*)dst + res; - dstCapacity -= res; - } - moreThan1Frame = 1; - } /* while (srcSize >= ZSTD_frameHeaderSize_prefix) */ - - RETURN_ERROR_IF(srcSize, srcSize_wrong, "input not entirely consumed"); - - return (size_t)((BYTE*)dst - (BYTE*)dststart); -} - -size_t ZSTD_decompress_usingDict(ZSTD_DCtx* dctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const void* dict, size_t dictSize) -{ - return ZSTD_decompressMultiFrame(dctx, dst, dstCapacity, src, srcSize, dict, dictSize, NULL); -} - - -static ZSTD_DDict const* ZSTD_getDDict(ZSTD_DCtx* dctx) -{ - switch (dctx->dictUses) { - default: - assert(0 /* Impossible */); - ZSTD_FALLTHROUGH; - case ZSTD_dont_use: - ZSTD_clearDict(dctx); - return NULL; - case ZSTD_use_indefinitely: - return dctx->ddict; - case ZSTD_use_once: - dctx->dictUses = ZSTD_dont_use; - return dctx->ddict; - } -} - -size_t ZSTD_decompressDCtx(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize) -{ - return ZSTD_decompress_usingDDict(dctx, dst, dstCapacity, src, srcSize, ZSTD_getDDict(dctx)); -} - - -size_t ZSTD_decompress(void* dst, size_t dstCapacity, const void* src, size_t srcSize) -{ -#if defined(ZSTD_HEAPMODE) && (ZSTD_HEAPMODE>=1) - size_t regenSize; - ZSTD_DCtx* const dctx = ZSTD_createDCtx_internal(ZSTD_defaultCMem); - RETURN_ERROR_IF(dctx==NULL, memory_allocation, "NULL pointer!"); - regenSize = ZSTD_decompressDCtx(dctx, dst, dstCapacity, src, srcSize); - ZSTD_freeDCtx(dctx); - return regenSize; -#else /* stack mode */ - ZSTD_DCtx dctx; - ZSTD_initDCtx_internal(&dctx); - return ZSTD_decompressDCtx(&dctx, dst, dstCapacity, src, srcSize); -#endif -} - - -/*-************************************** -* Advanced Streaming Decompression API -* Bufferless and synchronous -****************************************/ -size_t ZSTD_nextSrcSizeToDecompress(ZSTD_DCtx* dctx) { return dctx->expected; } - -/** - * Similar to ZSTD_nextSrcSizeToDecompress(), but when when a block input can be streamed, - * we allow taking a partial block as the input. Currently only raw uncompressed blocks can - * be streamed. - * - * For blocks that can be streamed, this allows us to reduce the latency until we produce - * output, and avoid copying the input. - * - * @param inputSize - The total amount of input that the caller currently has. - */ -static size_t ZSTD_nextSrcSizeToDecompressWithInputSize(ZSTD_DCtx* dctx, size_t inputSize) { - if (!(dctx->stage == ZSTDds_decompressBlock || dctx->stage == ZSTDds_decompressLastBlock)) - return dctx->expected; - if (dctx->bType != bt_raw) - return dctx->expected; - return BOUNDED(1, inputSize, dctx->expected); -} - -ZSTD_nextInputType_e ZSTD_nextInputType(ZSTD_DCtx* dctx) { - switch(dctx->stage) - { - default: /* should not happen */ - assert(0); - ZSTD_FALLTHROUGH; - case ZSTDds_getFrameHeaderSize: - ZSTD_FALLTHROUGH; - case ZSTDds_decodeFrameHeader: - return ZSTDnit_frameHeader; - case ZSTDds_decodeBlockHeader: - return ZSTDnit_blockHeader; - case ZSTDds_decompressBlock: - return ZSTDnit_block; - case ZSTDds_decompressLastBlock: - return ZSTDnit_lastBlock; - case ZSTDds_checkChecksum: - return ZSTDnit_checksum; - case ZSTDds_decodeSkippableHeader: - ZSTD_FALLTHROUGH; - case ZSTDds_skipFrame: - return ZSTDnit_skippableFrame; - } -} - -static int ZSTD_isSkipFrame(ZSTD_DCtx* dctx) { return dctx->stage == ZSTDds_skipFrame; } - -/** ZSTD_decompressContinue() : - * srcSize : must be the exact nb of bytes expected (see ZSTD_nextSrcSizeToDecompress()) - * @return : nb of bytes generated into `dst` (necessarily <= `dstCapacity) - * or an error code, which can be tested using ZSTD_isError() */ -size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize) -{ - DEBUGLOG(5, "ZSTD_decompressContinue (srcSize:%u)", (unsigned)srcSize); - /* Sanity check */ - RETURN_ERROR_IF(srcSize != ZSTD_nextSrcSizeToDecompressWithInputSize(dctx, srcSize), srcSize_wrong, "not allowed"); - ZSTD_checkContinuity(dctx, dst, dstCapacity); - - dctx->processedCSize += srcSize; - - switch (dctx->stage) - { - case ZSTDds_getFrameHeaderSize : - assert(src != NULL); - if (dctx->format == ZSTD_f_zstd1) { /* allows header */ - assert(srcSize >= ZSTD_FRAMEIDSIZE); /* to read skippable magic number */ - if ((MEM_readLE32(src) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { /* skippable frame */ - ZSTD_memcpy(dctx->headerBuffer, src, srcSize); - dctx->expected = ZSTD_SKIPPABLEHEADERSIZE - srcSize; /* remaining to load to get full skippable frame header */ - dctx->stage = ZSTDds_decodeSkippableHeader; - return 0; - } } - dctx->headerSize = ZSTD_frameHeaderSize_internal(src, srcSize, dctx->format); - if (ZSTD_isError(dctx->headerSize)) return dctx->headerSize; - ZSTD_memcpy(dctx->headerBuffer, src, srcSize); - dctx->expected = dctx->headerSize - srcSize; - dctx->stage = ZSTDds_decodeFrameHeader; - return 0; - - case ZSTDds_decodeFrameHeader: - assert(src != NULL); - ZSTD_memcpy(dctx->headerBuffer + (dctx->headerSize - srcSize), src, srcSize); - FORWARD_IF_ERROR(ZSTD_decodeFrameHeader(dctx, dctx->headerBuffer, dctx->headerSize), ""); - dctx->expected = ZSTD_blockHeaderSize; - dctx->stage = ZSTDds_decodeBlockHeader; - return 0; - - case ZSTDds_decodeBlockHeader: - { blockProperties_t bp; - size_t const cBlockSize = ZSTD_getcBlockSize(src, ZSTD_blockHeaderSize, &bp); - if (ZSTD_isError(cBlockSize)) return cBlockSize; - RETURN_ERROR_IF(cBlockSize > dctx->fParams.blockSizeMax, corruption_detected, "Block Size Exceeds Maximum"); - dctx->expected = cBlockSize; - dctx->bType = bp.blockType; - dctx->rleSize = bp.origSize; - if (cBlockSize) { - dctx->stage = bp.lastBlock ? ZSTDds_decompressLastBlock : ZSTDds_decompressBlock; - return 0; - } - /* empty block */ - if (bp.lastBlock) { - if (dctx->fParams.checksumFlag) { - dctx->expected = 4; - dctx->stage = ZSTDds_checkChecksum; - } else { - dctx->expected = 0; /* end of frame */ - dctx->stage = ZSTDds_getFrameHeaderSize; - } - } else { - dctx->expected = ZSTD_blockHeaderSize; /* jump to next header */ - dctx->stage = ZSTDds_decodeBlockHeader; - } - return 0; - } - - case ZSTDds_decompressLastBlock: - case ZSTDds_decompressBlock: - DEBUGLOG(5, "ZSTD_decompressContinue: case ZSTDds_decompressBlock"); - { size_t rSize; - switch(dctx->bType) - { - case bt_compressed: - DEBUGLOG(5, "ZSTD_decompressContinue: case bt_compressed"); - rSize = ZSTD_decompressBlock_internal(dctx, dst, dstCapacity, src, srcSize, /* frame */ 1, is_streaming); - dctx->expected = 0; /* Streaming not supported */ - break; - case bt_raw : - assert(srcSize <= dctx->expected); - rSize = ZSTD_copyRawBlock(dst, dstCapacity, src, srcSize); - FORWARD_IF_ERROR(rSize, "ZSTD_copyRawBlock failed"); - assert(rSize == srcSize); - dctx->expected -= rSize; - break; - case bt_rle : - rSize = ZSTD_setRleBlock(dst, dstCapacity, *(const BYTE*)src, dctx->rleSize); - dctx->expected = 0; /* Streaming not supported */ - break; - case bt_reserved : /* should never happen */ - default: - RETURN_ERROR(corruption_detected, "invalid block type"); - } - FORWARD_IF_ERROR(rSize, ""); - RETURN_ERROR_IF(rSize > dctx->fParams.blockSizeMax, corruption_detected, "Decompressed Block Size Exceeds Maximum"); - DEBUGLOG(5, "ZSTD_decompressContinue: decoded size from block : %u", (unsigned)rSize); - dctx->decodedSize += rSize; - if (dctx->validateChecksum) XXH64_update(&dctx->xxhState, dst, rSize); - dctx->previousDstEnd = (char*)dst + rSize; - - /* Stay on the same stage until we are finished streaming the block. */ - if (dctx->expected > 0) { - return rSize; - } - - if (dctx->stage == ZSTDds_decompressLastBlock) { /* end of frame */ - DEBUGLOG(4, "ZSTD_decompressContinue: decoded size from frame : %u", (unsigned)dctx->decodedSize); - RETURN_ERROR_IF( - dctx->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN - && dctx->decodedSize != dctx->fParams.frameContentSize, - corruption_detected, ""); - if (dctx->fParams.checksumFlag) { /* another round for frame checksum */ - dctx->expected = 4; - dctx->stage = ZSTDds_checkChecksum; - } else { - ZSTD_DCtx_trace_end(dctx, dctx->decodedSize, dctx->processedCSize, /* streaming */ 1); - dctx->expected = 0; /* ends here */ - dctx->stage = ZSTDds_getFrameHeaderSize; - } - } else { - dctx->stage = ZSTDds_decodeBlockHeader; - dctx->expected = ZSTD_blockHeaderSize; - } - return rSize; - } - - case ZSTDds_checkChecksum: - assert(srcSize == 4); /* guaranteed by dctx->expected */ - { - if (dctx->validateChecksum) { - U32 const h32 = (U32)XXH64_digest(&dctx->xxhState); - U32 const check32 = MEM_readLE32(src); - DEBUGLOG(4, "ZSTD_decompressContinue: checksum : calculated %08X :: %08X read", (unsigned)h32, (unsigned)check32); - RETURN_ERROR_IF(check32 != h32, checksum_wrong, ""); - } - ZSTD_DCtx_trace_end(dctx, dctx->decodedSize, dctx->processedCSize, /* streaming */ 1); - dctx->expected = 0; - dctx->stage = ZSTDds_getFrameHeaderSize; - return 0; - } - - case ZSTDds_decodeSkippableHeader: - assert(src != NULL); - assert(srcSize <= ZSTD_SKIPPABLEHEADERSIZE); - ZSTD_memcpy(dctx->headerBuffer + (ZSTD_SKIPPABLEHEADERSIZE - srcSize), src, srcSize); /* complete skippable header */ - dctx->expected = MEM_readLE32(dctx->headerBuffer + ZSTD_FRAMEIDSIZE); /* note : dctx->expected can grow seriously large, beyond local buffer size */ - dctx->stage = ZSTDds_skipFrame; - return 0; - - case ZSTDds_skipFrame: - dctx->expected = 0; - dctx->stage = ZSTDds_getFrameHeaderSize; - return 0; - - default: - assert(0); /* impossible */ - RETURN_ERROR(GENERIC, "impossible to reach"); /* some compiler require default to do something */ - } -} - - -static size_t ZSTD_refDictContent(ZSTD_DCtx* dctx, const void* dict, size_t dictSize) -{ - dctx->dictEnd = dctx->previousDstEnd; - dctx->virtualStart = (const char*)dict - ((const char*)(dctx->previousDstEnd) - (const char*)(dctx->prefixStart)); - dctx->prefixStart = dict; - dctx->previousDstEnd = (const char*)dict + dictSize; -#ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION - dctx->dictContentBeginForFuzzing = dctx->prefixStart; - dctx->dictContentEndForFuzzing = dctx->previousDstEnd; -#endif - return 0; -} - -/*! ZSTD_loadDEntropy() : - * dict : must point at beginning of a valid zstd dictionary. - * @return : size of entropy tables read */ -size_t -ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy, - const void* const dict, size_t const dictSize) -{ - const BYTE* dictPtr = (const BYTE*)dict; - const BYTE* const dictEnd = dictPtr + dictSize; - - RETURN_ERROR_IF(dictSize <= 8, dictionary_corrupted, "dict is too small"); - assert(MEM_readLE32(dict) == ZSTD_MAGIC_DICTIONARY); /* dict must be valid */ - dictPtr += 8; /* skip header = magic + dictID */ - - ZSTD_STATIC_ASSERT(offsetof(ZSTD_entropyDTables_t, OFTable) == offsetof(ZSTD_entropyDTables_t, LLTable) + sizeof(entropy->LLTable)); - ZSTD_STATIC_ASSERT(offsetof(ZSTD_entropyDTables_t, MLTable) == offsetof(ZSTD_entropyDTables_t, OFTable) + sizeof(entropy->OFTable)); - ZSTD_STATIC_ASSERT(sizeof(entropy->LLTable) + sizeof(entropy->OFTable) + sizeof(entropy->MLTable) >= HUF_DECOMPRESS_WORKSPACE_SIZE); - { void* const workspace = &entropy->LLTable; /* use fse tables as temporary workspace; implies fse tables are grouped together */ - size_t const workspaceSize = sizeof(entropy->LLTable) + sizeof(entropy->OFTable) + sizeof(entropy->MLTable); -#ifdef HUF_FORCE_DECOMPRESS_X1 - /* in minimal huffman, we always use X1 variants */ - size_t const hSize = HUF_readDTableX1_wksp(entropy->hufTable, - dictPtr, dictEnd - dictPtr, - workspace, workspaceSize); -#else - size_t const hSize = HUF_readDTableX2_wksp(entropy->hufTable, - dictPtr, (size_t)(dictEnd - dictPtr), - workspace, workspaceSize); -#endif - RETURN_ERROR_IF(HUF_isError(hSize), dictionary_corrupted, ""); - dictPtr += hSize; - } - - { short offcodeNCount[MaxOff+1]; - unsigned offcodeMaxValue = MaxOff, offcodeLog; - size_t const offcodeHeaderSize = FSE_readNCount(offcodeNCount, &offcodeMaxValue, &offcodeLog, dictPtr, (size_t)(dictEnd-dictPtr)); - RETURN_ERROR_IF(FSE_isError(offcodeHeaderSize), dictionary_corrupted, ""); - RETURN_ERROR_IF(offcodeMaxValue > MaxOff, dictionary_corrupted, ""); - RETURN_ERROR_IF(offcodeLog > OffFSELog, dictionary_corrupted, ""); - ZSTD_buildFSETable( entropy->OFTable, - offcodeNCount, offcodeMaxValue, - OF_base, OF_bits, - offcodeLog, - entropy->workspace, sizeof(entropy->workspace), - /* bmi2 */0); - dictPtr += offcodeHeaderSize; - } - - { short matchlengthNCount[MaxML+1]; - unsigned matchlengthMaxValue = MaxML, matchlengthLog; - size_t const matchlengthHeaderSize = FSE_readNCount(matchlengthNCount, &matchlengthMaxValue, &matchlengthLog, dictPtr, (size_t)(dictEnd-dictPtr)); - RETURN_ERROR_IF(FSE_isError(matchlengthHeaderSize), dictionary_corrupted, ""); - RETURN_ERROR_IF(matchlengthMaxValue > MaxML, dictionary_corrupted, ""); - RETURN_ERROR_IF(matchlengthLog > MLFSELog, dictionary_corrupted, ""); - ZSTD_buildFSETable( entropy->MLTable, - matchlengthNCount, matchlengthMaxValue, - ML_base, ML_bits, - matchlengthLog, - entropy->workspace, sizeof(entropy->workspace), - /* bmi2 */ 0); - dictPtr += matchlengthHeaderSize; - } - - { short litlengthNCount[MaxLL+1]; - unsigned litlengthMaxValue = MaxLL, litlengthLog; - size_t const litlengthHeaderSize = FSE_readNCount(litlengthNCount, &litlengthMaxValue, &litlengthLog, dictPtr, (size_t)(dictEnd-dictPtr)); - RETURN_ERROR_IF(FSE_isError(litlengthHeaderSize), dictionary_corrupted, ""); - RETURN_ERROR_IF(litlengthMaxValue > MaxLL, dictionary_corrupted, ""); - RETURN_ERROR_IF(litlengthLog > LLFSELog, dictionary_corrupted, ""); - ZSTD_buildFSETable( entropy->LLTable, - litlengthNCount, litlengthMaxValue, - LL_base, LL_bits, - litlengthLog, - entropy->workspace, sizeof(entropy->workspace), - /* bmi2 */ 0); - dictPtr += litlengthHeaderSize; - } - - RETURN_ERROR_IF(dictPtr+12 > dictEnd, dictionary_corrupted, ""); - { int i; - size_t const dictContentSize = (size_t)(dictEnd - (dictPtr+12)); - for (i=0; i<3; i++) { - U32 const rep = MEM_readLE32(dictPtr); dictPtr += 4; - RETURN_ERROR_IF(rep==0 || rep > dictContentSize, - dictionary_corrupted, ""); - entropy->rep[i] = rep; - } } - - return (size_t)(dictPtr - (const BYTE*)dict); -} - -static size_t ZSTD_decompress_insertDictionary(ZSTD_DCtx* dctx, const void* dict, size_t dictSize) -{ - if (dictSize < 8) return ZSTD_refDictContent(dctx, dict, dictSize); - { U32 const magic = MEM_readLE32(dict); - if (magic != ZSTD_MAGIC_DICTIONARY) { - return ZSTD_refDictContent(dctx, dict, dictSize); /* pure content mode */ - } } - dctx->dictID = MEM_readLE32((const char*)dict + ZSTD_FRAMEIDSIZE); - - /* load entropy tables */ - { size_t const eSize = ZSTD_loadDEntropy(&dctx->entropy, dict, dictSize); - RETURN_ERROR_IF(ZSTD_isError(eSize), dictionary_corrupted, ""); - dict = (const char*)dict + eSize; - dictSize -= eSize; - } - dctx->litEntropy = dctx->fseEntropy = 1; - - /* reference dictionary content */ - return ZSTD_refDictContent(dctx, dict, dictSize); -} - -size_t ZSTD_decompressBegin(ZSTD_DCtx* dctx) -{ - assert(dctx != NULL); -#if ZSTD_TRACE - dctx->traceCtx = (ZSTD_trace_decompress_begin != NULL) ? ZSTD_trace_decompress_begin(dctx) : 0; -#endif - dctx->expected = ZSTD_startingInputLength(dctx->format); /* dctx->format must be properly set */ - dctx->stage = ZSTDds_getFrameHeaderSize; - dctx->processedCSize = 0; - dctx->decodedSize = 0; - dctx->previousDstEnd = NULL; - dctx->prefixStart = NULL; - dctx->virtualStart = NULL; - dctx->dictEnd = NULL; - dctx->entropy.hufTable[0] = (HUF_DTable)((HufLog)*0x1000001); /* cover both little and big endian */ - dctx->litEntropy = dctx->fseEntropy = 0; - dctx->dictID = 0; - dctx->bType = bt_reserved; - ZSTD_STATIC_ASSERT(sizeof(dctx->entropy.rep) == sizeof(repStartValue)); - ZSTD_memcpy(dctx->entropy.rep, repStartValue, sizeof(repStartValue)); /* initial repcodes */ - dctx->LLTptr = dctx->entropy.LLTable; - dctx->MLTptr = dctx->entropy.MLTable; - dctx->OFTptr = dctx->entropy.OFTable; - dctx->HUFptr = dctx->entropy.hufTable; - return 0; -} - -size_t ZSTD_decompressBegin_usingDict(ZSTD_DCtx* dctx, const void* dict, size_t dictSize) -{ - FORWARD_IF_ERROR( ZSTD_decompressBegin(dctx) , ""); - if (dict && dictSize) - RETURN_ERROR_IF( - ZSTD_isError(ZSTD_decompress_insertDictionary(dctx, dict, dictSize)), - dictionary_corrupted, ""); - return 0; -} - - -/* ====== ZSTD_DDict ====== */ - -size_t ZSTD_decompressBegin_usingDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict) -{ - DEBUGLOG(4, "ZSTD_decompressBegin_usingDDict"); - assert(dctx != NULL); - if (ddict) { - const char* const dictStart = (const char*)ZSTD_DDict_dictContent(ddict); - size_t const dictSize = ZSTD_DDict_dictSize(ddict); - const void* const dictEnd = dictStart + dictSize; - dctx->ddictIsCold = (dctx->dictEnd != dictEnd); - DEBUGLOG(4, "DDict is %s", - dctx->ddictIsCold ? "~cold~" : "hot!"); - } - FORWARD_IF_ERROR( ZSTD_decompressBegin(dctx) , ""); - if (ddict) { /* NULL ddict is equivalent to no dictionary */ - ZSTD_copyDDictParameters(dctx, ddict); - } - return 0; -} - -/*! ZSTD_getDictID_fromDict() : - * Provides the dictID stored within dictionary. - * if @return == 0, the dictionary is not conformant with Zstandard specification. - * It can still be loaded, but as a content-only dictionary. */ -unsigned ZSTD_getDictID_fromDict(const void* dict, size_t dictSize) -{ - if (dictSize < 8) return 0; - if (MEM_readLE32(dict) != ZSTD_MAGIC_DICTIONARY) return 0; - return MEM_readLE32((const char*)dict + ZSTD_FRAMEIDSIZE); -} - -/*! ZSTD_getDictID_fromFrame() : - * Provides the dictID required to decompress frame stored within `src`. - * If @return == 0, the dictID could not be decoded. - * This could for one of the following reasons : - * - The frame does not require a dictionary (most common case). - * - The frame was built with dictID intentionally removed. - * Needed dictionary is a hidden information. - * Note : this use case also happens when using a non-conformant dictionary. - * - `srcSize` is too small, and as a result, frame header could not be decoded. - * Note : possible if `srcSize < ZSTD_FRAMEHEADERSIZE_MAX`. - * - This is not a Zstandard frame. - * When identifying the exact failure cause, it's possible to use - * ZSTD_getFrameHeader(), which will provide a more precise error code. */ -unsigned ZSTD_getDictID_fromFrame(const void* src, size_t srcSize) -{ - ZSTD_frameHeader zfp = { 0, 0, 0, ZSTD_frame, 0, 0, 0 }; - size_t const hError = ZSTD_getFrameHeader(&zfp, src, srcSize); - if (ZSTD_isError(hError)) return 0; - return zfp.dictID; -} - - -/*! ZSTD_decompress_usingDDict() : -* Decompression using a pre-digested Dictionary -* Use dictionary without significant overhead. */ -size_t ZSTD_decompress_usingDDict(ZSTD_DCtx* dctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const ZSTD_DDict* ddict) -{ - /* pass content and size in case legacy frames are encountered */ - return ZSTD_decompressMultiFrame(dctx, dst, dstCapacity, src, srcSize, - NULL, 0, - ddict); -} - - -/*===================================== -* Streaming decompression -*====================================*/ - -ZSTD_DStream* ZSTD_createDStream(void) -{ - DEBUGLOG(3, "ZSTD_createDStream"); - return ZSTD_createDCtx_internal(ZSTD_defaultCMem); -} - -ZSTD_DStream* ZSTD_initStaticDStream(void *workspace, size_t workspaceSize) -{ - return ZSTD_initStaticDCtx(workspace, workspaceSize); -} - -ZSTD_DStream* ZSTD_createDStream_advanced(ZSTD_customMem customMem) -{ - return ZSTD_createDCtx_internal(customMem); -} - -size_t ZSTD_freeDStream(ZSTD_DStream* zds) -{ - return ZSTD_freeDCtx(zds); -} - - -/* *** Initialization *** */ - -size_t ZSTD_DStreamInSize(void) { return ZSTD_BLOCKSIZE_MAX + ZSTD_blockHeaderSize; } -size_t ZSTD_DStreamOutSize(void) { return ZSTD_BLOCKSIZE_MAX; } - -size_t ZSTD_DCtx_loadDictionary_advanced(ZSTD_DCtx* dctx, - const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType) -{ - RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong, ""); - ZSTD_clearDict(dctx); - if (dict && dictSize != 0) { - dctx->ddictLocal = ZSTD_createDDict_advanced(dict, dictSize, dictLoadMethod, dictContentType, dctx->customMem); - RETURN_ERROR_IF(dctx->ddictLocal == NULL, memory_allocation, "NULL pointer!"); - dctx->ddict = dctx->ddictLocal; - dctx->dictUses = ZSTD_use_indefinitely; - } - return 0; -} - -size_t ZSTD_DCtx_loadDictionary_byReference(ZSTD_DCtx* dctx, const void* dict, size_t dictSize) -{ - return ZSTD_DCtx_loadDictionary_advanced(dctx, dict, dictSize, ZSTD_dlm_byRef, ZSTD_dct_auto); -} - -size_t ZSTD_DCtx_loadDictionary(ZSTD_DCtx* dctx, const void* dict, size_t dictSize) -{ - return ZSTD_DCtx_loadDictionary_advanced(dctx, dict, dictSize, ZSTD_dlm_byCopy, ZSTD_dct_auto); -} - -size_t ZSTD_DCtx_refPrefix_advanced(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType) -{ - FORWARD_IF_ERROR(ZSTD_DCtx_loadDictionary_advanced(dctx, prefix, prefixSize, ZSTD_dlm_byRef, dictContentType), ""); - dctx->dictUses = ZSTD_use_once; - return 0; -} - -size_t ZSTD_DCtx_refPrefix(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize) -{ - return ZSTD_DCtx_refPrefix_advanced(dctx, prefix, prefixSize, ZSTD_dct_rawContent); -} - - -/* ZSTD_initDStream_usingDict() : - * return : expected size, aka ZSTD_startingInputLength(). - * this function cannot fail */ -size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize) -{ - DEBUGLOG(4, "ZSTD_initDStream_usingDict"); - FORWARD_IF_ERROR( ZSTD_DCtx_reset(zds, ZSTD_reset_session_only) , ""); - FORWARD_IF_ERROR( ZSTD_DCtx_loadDictionary(zds, dict, dictSize) , ""); - return ZSTD_startingInputLength(zds->format); -} - -/* note : this variant can't fail */ -size_t ZSTD_initDStream(ZSTD_DStream* zds) -{ - DEBUGLOG(4, "ZSTD_initDStream"); - return ZSTD_initDStream_usingDDict(zds, NULL); -} - -/* ZSTD_initDStream_usingDDict() : - * ddict will just be referenced, and must outlive decompression session - * this function cannot fail */ -size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* dctx, const ZSTD_DDict* ddict) -{ - FORWARD_IF_ERROR( ZSTD_DCtx_reset(dctx, ZSTD_reset_session_only) , ""); - FORWARD_IF_ERROR( ZSTD_DCtx_refDDict(dctx, ddict) , ""); - return ZSTD_startingInputLength(dctx->format); -} - -/* ZSTD_resetDStream() : - * return : expected size, aka ZSTD_startingInputLength(). - * this function cannot fail */ -size_t ZSTD_resetDStream(ZSTD_DStream* dctx) -{ - FORWARD_IF_ERROR(ZSTD_DCtx_reset(dctx, ZSTD_reset_session_only), ""); - return ZSTD_startingInputLength(dctx->format); -} - - -size_t ZSTD_DCtx_refDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict) -{ - RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong, ""); - ZSTD_clearDict(dctx); - if (ddict) { - dctx->ddict = ddict; - dctx->dictUses = ZSTD_use_indefinitely; - if (dctx->refMultipleDDicts == ZSTD_rmd_refMultipleDDicts) { - if (dctx->ddictSet == NULL) { - dctx->ddictSet = ZSTD_createDDictHashSet(dctx->customMem); - if (!dctx->ddictSet) { - RETURN_ERROR(memory_allocation, "Failed to allocate memory for hash set!"); - } - } - assert(!dctx->staticSize); /* Impossible: ddictSet cannot have been allocated if static dctx */ - FORWARD_IF_ERROR(ZSTD_DDictHashSet_addDDict(dctx->ddictSet, ddict, dctx->customMem), ""); - } - } - return 0; -} - -/* ZSTD_DCtx_setMaxWindowSize() : - * note : no direct equivalence in ZSTD_DCtx_setParameter, - * since this version sets windowSize, and the other sets windowLog */ -size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize) -{ - ZSTD_bounds const bounds = ZSTD_dParam_getBounds(ZSTD_d_windowLogMax); - size_t const min = (size_t)1 << bounds.lowerBound; - size_t const max = (size_t)1 << bounds.upperBound; - RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong, ""); - RETURN_ERROR_IF(maxWindowSize < min, parameter_outOfBound, ""); - RETURN_ERROR_IF(maxWindowSize > max, parameter_outOfBound, ""); - dctx->maxWindowSize = maxWindowSize; - return 0; -} - -size_t ZSTD_DCtx_setFormat(ZSTD_DCtx* dctx, ZSTD_format_e format) -{ - return ZSTD_DCtx_setParameter(dctx, ZSTD_d_format, (int)format); -} - -ZSTD_bounds ZSTD_dParam_getBounds(ZSTD_dParameter dParam) -{ - ZSTD_bounds bounds = { 0, 0, 0 }; - switch(dParam) { - case ZSTD_d_windowLogMax: - bounds.lowerBound = ZSTD_WINDOWLOG_ABSOLUTEMIN; - bounds.upperBound = ZSTD_WINDOWLOG_MAX; - return bounds; - case ZSTD_d_format: - bounds.lowerBound = (int)ZSTD_f_zstd1; - bounds.upperBound = (int)ZSTD_f_zstd1_magicless; - ZSTD_STATIC_ASSERT(ZSTD_f_zstd1 < ZSTD_f_zstd1_magicless); - return bounds; - case ZSTD_d_stableOutBuffer: - bounds.lowerBound = (int)ZSTD_bm_buffered; - bounds.upperBound = (int)ZSTD_bm_stable; - return bounds; - case ZSTD_d_forceIgnoreChecksum: - bounds.lowerBound = (int)ZSTD_d_validateChecksum; - bounds.upperBound = (int)ZSTD_d_ignoreChecksum; - return bounds; - case ZSTD_d_refMultipleDDicts: - bounds.lowerBound = (int)ZSTD_rmd_refSingleDDict; - bounds.upperBound = (int)ZSTD_rmd_refMultipleDDicts; - return bounds; - default:; - } - bounds.error = ERROR(parameter_unsupported); - return bounds; -} - -/* ZSTD_dParam_withinBounds: - * @return 1 if value is within dParam bounds, - * 0 otherwise */ -static int ZSTD_dParam_withinBounds(ZSTD_dParameter dParam, int value) -{ - ZSTD_bounds const bounds = ZSTD_dParam_getBounds(dParam); - if (ZSTD_isError(bounds.error)) return 0; - if (value < bounds.lowerBound) return 0; - if (value > bounds.upperBound) return 0; - return 1; -} - -#define CHECK_DBOUNDS(p,v) { \ - RETURN_ERROR_IF(!ZSTD_dParam_withinBounds(p, v), parameter_outOfBound, ""); \ -} - -size_t ZSTD_DCtx_getParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int* value) -{ - switch (param) { - case ZSTD_d_windowLogMax: - *value = (int)ZSTD_highbit32((U32)dctx->maxWindowSize); - return 0; - case ZSTD_d_format: - *value = (int)dctx->format; - return 0; - case ZSTD_d_stableOutBuffer: - *value = (int)dctx->outBufferMode; - return 0; - case ZSTD_d_forceIgnoreChecksum: - *value = (int)dctx->forceIgnoreChecksum; - return 0; - case ZSTD_d_refMultipleDDicts: - *value = (int)dctx->refMultipleDDicts; - return 0; - default:; - } - RETURN_ERROR(parameter_unsupported, ""); -} - -size_t ZSTD_DCtx_setParameter(ZSTD_DCtx* dctx, ZSTD_dParameter dParam, int value) -{ - RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong, ""); - switch(dParam) { - case ZSTD_d_windowLogMax: - if (value == 0) value = ZSTD_WINDOWLOG_LIMIT_DEFAULT; - CHECK_DBOUNDS(ZSTD_d_windowLogMax, value); - dctx->maxWindowSize = ((size_t)1) << value; - return 0; - case ZSTD_d_format: - CHECK_DBOUNDS(ZSTD_d_format, value); - dctx->format = (ZSTD_format_e)value; - return 0; - case ZSTD_d_stableOutBuffer: - CHECK_DBOUNDS(ZSTD_d_stableOutBuffer, value); - dctx->outBufferMode = (ZSTD_bufferMode_e)value; - return 0; - case ZSTD_d_forceIgnoreChecksum: - CHECK_DBOUNDS(ZSTD_d_forceIgnoreChecksum, value); - dctx->forceIgnoreChecksum = (ZSTD_forceIgnoreChecksum_e)value; - return 0; - case ZSTD_d_refMultipleDDicts: - CHECK_DBOUNDS(ZSTD_d_refMultipleDDicts, value); - if (dctx->staticSize != 0) { - RETURN_ERROR(parameter_unsupported, "Static dctx does not support multiple DDicts!"); - } - dctx->refMultipleDDicts = (ZSTD_refMultipleDDicts_e)value; - return 0; - default:; - } - RETURN_ERROR(parameter_unsupported, ""); -} - -size_t ZSTD_DCtx_reset(ZSTD_DCtx* dctx, ZSTD_ResetDirective reset) -{ - if ( (reset == ZSTD_reset_session_only) - || (reset == ZSTD_reset_session_and_parameters) ) { - dctx->streamStage = zdss_init; - dctx->noForwardProgress = 0; - } - if ( (reset == ZSTD_reset_parameters) - || (reset == ZSTD_reset_session_and_parameters) ) { - RETURN_ERROR_IF(dctx->streamStage != zdss_init, stage_wrong, ""); - ZSTD_clearDict(dctx); - ZSTD_DCtx_resetParameters(dctx); - } - return 0; -} - - -size_t ZSTD_sizeof_DStream(const ZSTD_DStream* dctx) -{ - return ZSTD_sizeof_DCtx(dctx); -} - -size_t ZSTD_decodingBufferSize_min(unsigned long long windowSize, unsigned long long frameContentSize) -{ - size_t const blockSize = (size_t) MIN(windowSize, ZSTD_BLOCKSIZE_MAX); - /* space is needed to store the litbuffer after the output of a given block without stomping the extDict of a previous run, as well as to cover both windows against wildcopy*/ - unsigned long long const neededRBSize = windowSize + blockSize + ZSTD_BLOCKSIZE_MAX + (WILDCOPY_OVERLENGTH * 2); - unsigned long long const neededSize = MIN(frameContentSize, neededRBSize); - size_t const minRBSize = (size_t) neededSize; - RETURN_ERROR_IF((unsigned long long)minRBSize != neededSize, - frameParameter_windowTooLarge, ""); - return minRBSize; -} - -size_t ZSTD_estimateDStreamSize(size_t windowSize) -{ - size_t const blockSize = MIN(windowSize, ZSTD_BLOCKSIZE_MAX); - size_t const inBuffSize = blockSize; /* no block can be larger */ - size_t const outBuffSize = ZSTD_decodingBufferSize_min(windowSize, ZSTD_CONTENTSIZE_UNKNOWN); - return ZSTD_estimateDCtxSize() + inBuffSize + outBuffSize; -} - -size_t ZSTD_estimateDStreamSize_fromFrame(const void* src, size_t srcSize) -{ - U32 const windowSizeMax = 1U << ZSTD_WINDOWLOG_MAX; /* note : should be user-selectable, but requires an additional parameter (or a dctx) */ - ZSTD_frameHeader zfh; - size_t const err = ZSTD_getFrameHeader(&zfh, src, srcSize); - if (ZSTD_isError(err)) return err; - RETURN_ERROR_IF(err>0, srcSize_wrong, ""); - RETURN_ERROR_IF(zfh.windowSize > windowSizeMax, - frameParameter_windowTooLarge, ""); - return ZSTD_estimateDStreamSize((size_t)zfh.windowSize); -} - - -/* ***** Decompression ***** */ - -static int ZSTD_DCtx_isOverflow(ZSTD_DStream* zds, size_t const neededInBuffSize, size_t const neededOutBuffSize) -{ - return (zds->inBuffSize + zds->outBuffSize) >= (neededInBuffSize + neededOutBuffSize) * ZSTD_WORKSPACETOOLARGE_FACTOR; -} - -static void ZSTD_DCtx_updateOversizedDuration(ZSTD_DStream* zds, size_t const neededInBuffSize, size_t const neededOutBuffSize) -{ - if (ZSTD_DCtx_isOverflow(zds, neededInBuffSize, neededOutBuffSize)) - zds->oversizedDuration++; - else - zds->oversizedDuration = 0; -} - -static int ZSTD_DCtx_isOversizedTooLong(ZSTD_DStream* zds) -{ - return zds->oversizedDuration >= ZSTD_WORKSPACETOOLARGE_MAXDURATION; -} - -/* Checks that the output buffer hasn't changed if ZSTD_obm_stable is used. */ -static size_t ZSTD_checkOutBuffer(ZSTD_DStream const* zds, ZSTD_outBuffer const* output) -{ - ZSTD_outBuffer const expect = zds->expectedOutBuffer; - /* No requirement when ZSTD_obm_stable is not enabled. */ - if (zds->outBufferMode != ZSTD_bm_stable) - return 0; - /* Any buffer is allowed in zdss_init, this must be the same for every other call until - * the context is reset. - */ - if (zds->streamStage == zdss_init) - return 0; - /* The buffer must match our expectation exactly. */ - if (expect.dst == output->dst && expect.pos == output->pos && expect.size == output->size) - return 0; - RETURN_ERROR(dstBuffer_wrong, "ZSTD_d_stableOutBuffer enabled but output differs!"); -} - -/* Calls ZSTD_decompressContinue() with the right parameters for ZSTD_decompressStream() - * and updates the stage and the output buffer state. This call is extracted so it can be - * used both when reading directly from the ZSTD_inBuffer, and in buffered input mode. - * NOTE: You must break after calling this function since the streamStage is modified. - */ -static size_t ZSTD_decompressContinueStream( - ZSTD_DStream* zds, char** op, char* oend, - void const* src, size_t srcSize) { - int const isSkipFrame = ZSTD_isSkipFrame(zds); - if (zds->outBufferMode == ZSTD_bm_buffered) { - size_t const dstSize = isSkipFrame ? 0 : zds->outBuffSize - zds->outStart; - size_t const decodedSize = ZSTD_decompressContinue(zds, - zds->outBuff + zds->outStart, dstSize, src, srcSize); - FORWARD_IF_ERROR(decodedSize, ""); - if (!decodedSize && !isSkipFrame) { - zds->streamStage = zdss_read; - } else { - zds->outEnd = zds->outStart + decodedSize; - zds->streamStage = zdss_flush; - } - } else { - /* Write directly into the output buffer */ - size_t const dstSize = isSkipFrame ? 0 : (size_t)(oend - *op); - size_t const decodedSize = ZSTD_decompressContinue(zds, *op, dstSize, src, srcSize); - FORWARD_IF_ERROR(decodedSize, ""); - *op += decodedSize; - /* Flushing is not needed. */ - zds->streamStage = zdss_read; - assert(*op <= oend); - assert(zds->outBufferMode == ZSTD_bm_stable); - } - return 0; -} - -size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inBuffer* input) -{ - const char* const src = (const char*)input->src; - const char* const istart = input->pos != 0 ? src + input->pos : src; - const char* const iend = input->size != 0 ? src + input->size : src; - const char* ip = istart; - char* const dst = (char*)output->dst; - char* const ostart = output->pos != 0 ? dst + output->pos : dst; - char* const oend = output->size != 0 ? dst + output->size : dst; - char* op = ostart; - U32 someMoreWork = 1; - - DEBUGLOG(5, "ZSTD_decompressStream"); - RETURN_ERROR_IF( - input->pos > input->size, - srcSize_wrong, - "forbidden. in: pos: %u vs size: %u", - (U32)input->pos, (U32)input->size); - RETURN_ERROR_IF( - output->pos > output->size, - dstSize_tooSmall, - "forbidden. out: pos: %u vs size: %u", - (U32)output->pos, (U32)output->size); - DEBUGLOG(5, "input size : %u", (U32)(input->size - input->pos)); - FORWARD_IF_ERROR(ZSTD_checkOutBuffer(zds, output), ""); - - while (someMoreWork) { - switch(zds->streamStage) - { - case zdss_init : - DEBUGLOG(5, "stage zdss_init => transparent reset "); - zds->streamStage = zdss_loadHeader; - zds->lhSize = zds->inPos = zds->outStart = zds->outEnd = 0; -#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1) - zds->legacyVersion = 0; -#endif - zds->hostageByte = 0; - zds->expectedOutBuffer = *output; - ZSTD_FALLTHROUGH; - - case zdss_loadHeader : - DEBUGLOG(5, "stage zdss_loadHeader (srcSize : %u)", (U32)(iend - ip)); -#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1) - if (zds->legacyVersion) { - RETURN_ERROR_IF(zds->staticSize, memory_allocation, - "legacy support is incompatible with static dctx"); - { size_t const hint = ZSTD_decompressLegacyStream(zds->legacyContext, zds->legacyVersion, output, input); - if (hint==0) zds->streamStage = zdss_init; - return hint; - } } -#endif - { size_t const hSize = ZSTD_getFrameHeader_advanced(&zds->fParams, zds->headerBuffer, zds->lhSize, zds->format); - if (zds->refMultipleDDicts && zds->ddictSet) { - ZSTD_DCtx_selectFrameDDict(zds); - } - DEBUGLOG(5, "header size : %u", (U32)hSize); - if (ZSTD_isError(hSize)) { -#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1) - U32 const legacyVersion = ZSTD_isLegacy(istart, iend-istart); - if (legacyVersion) { - ZSTD_DDict const* const ddict = ZSTD_getDDict(zds); - const void* const dict = ddict ? ZSTD_DDict_dictContent(ddict) : NULL; - size_t const dictSize = ddict ? ZSTD_DDict_dictSize(ddict) : 0; - DEBUGLOG(5, "ZSTD_decompressStream: detected legacy version v0.%u", legacyVersion); - RETURN_ERROR_IF(zds->staticSize, memory_allocation, - "legacy support is incompatible with static dctx"); - FORWARD_IF_ERROR(ZSTD_initLegacyStream(&zds->legacyContext, - zds->previousLegacyVersion, legacyVersion, - dict, dictSize), ""); - zds->legacyVersion = zds->previousLegacyVersion = legacyVersion; - { size_t const hint = ZSTD_decompressLegacyStream(zds->legacyContext, legacyVersion, output, input); - if (hint==0) zds->streamStage = zdss_init; /* or stay in stage zdss_loadHeader */ - return hint; - } } -#endif - return hSize; /* error */ - } - if (hSize != 0) { /* need more input */ - size_t const toLoad = hSize - zds->lhSize; /* if hSize!=0, hSize > zds->lhSize */ - size_t const remainingInput = (size_t)(iend-ip); - assert(iend >= ip); - if (toLoad > remainingInput) { /* not enough input to load full header */ - if (remainingInput > 0) { - ZSTD_memcpy(zds->headerBuffer + zds->lhSize, ip, remainingInput); - zds->lhSize += remainingInput; - } - input->pos = input->size; - return (MAX((size_t)ZSTD_FRAMEHEADERSIZE_MIN(zds->format), hSize) - zds->lhSize) + ZSTD_blockHeaderSize; /* remaining header bytes + next block header */ - } - assert(ip != NULL); - ZSTD_memcpy(zds->headerBuffer + zds->lhSize, ip, toLoad); zds->lhSize = hSize; ip += toLoad; - break; - } } - - /* check for single-pass mode opportunity */ - if (zds->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN - && zds->fParams.frameType != ZSTD_skippableFrame - && (U64)(size_t)(oend-op) >= zds->fParams.frameContentSize) { - size_t const cSize = ZSTD_findFrameCompressedSize(istart, (size_t)(iend-istart)); - if (cSize <= (size_t)(iend-istart)) { - /* shortcut : using single-pass mode */ - size_t const decompressedSize = ZSTD_decompress_usingDDict(zds, op, (size_t)(oend-op), istart, cSize, ZSTD_getDDict(zds)); - if (ZSTD_isError(decompressedSize)) return decompressedSize; - DEBUGLOG(4, "shortcut to single-pass ZSTD_decompress_usingDDict()") - ip = istart + cSize; - op += decompressedSize; - zds->expected = 0; - zds->streamStage = zdss_init; - someMoreWork = 0; - break; - } } - - /* Check output buffer is large enough for ZSTD_odm_stable. */ - if (zds->outBufferMode == ZSTD_bm_stable - && zds->fParams.frameType != ZSTD_skippableFrame - && zds->fParams.frameContentSize != ZSTD_CONTENTSIZE_UNKNOWN - && (U64)(size_t)(oend-op) < zds->fParams.frameContentSize) { - RETURN_ERROR(dstSize_tooSmall, "ZSTD_obm_stable passed but ZSTD_outBuffer is too small"); - } - - /* Consume header (see ZSTDds_decodeFrameHeader) */ - DEBUGLOG(4, "Consume header"); - FORWARD_IF_ERROR(ZSTD_decompressBegin_usingDDict(zds, ZSTD_getDDict(zds)), ""); - - if ((MEM_readLE32(zds->headerBuffer) & ZSTD_MAGIC_SKIPPABLE_MASK) == ZSTD_MAGIC_SKIPPABLE_START) { /* skippable frame */ - zds->expected = MEM_readLE32(zds->headerBuffer + ZSTD_FRAMEIDSIZE); - zds->stage = ZSTDds_skipFrame; - } else { - FORWARD_IF_ERROR(ZSTD_decodeFrameHeader(zds, zds->headerBuffer, zds->lhSize), ""); - zds->expected = ZSTD_blockHeaderSize; - zds->stage = ZSTDds_decodeBlockHeader; - } - - /* control buffer memory usage */ - DEBUGLOG(4, "Control max memory usage (%u KB <= max %u KB)", - (U32)(zds->fParams.windowSize >>10), - (U32)(zds->maxWindowSize >> 10) ); - zds->fParams.windowSize = MAX(zds->fParams.windowSize, 1U << ZSTD_WINDOWLOG_ABSOLUTEMIN); - RETURN_ERROR_IF(zds->fParams.windowSize > zds->maxWindowSize, - frameParameter_windowTooLarge, ""); - - /* Adapt buffer sizes to frame header instructions */ - { size_t const neededInBuffSize = MAX(zds->fParams.blockSizeMax, 4 /* frame checksum */); - size_t const neededOutBuffSize = zds->outBufferMode == ZSTD_bm_buffered - ? ZSTD_decodingBufferSize_min(zds->fParams.windowSize, zds->fParams.frameContentSize) - : 0; - - ZSTD_DCtx_updateOversizedDuration(zds, neededInBuffSize, neededOutBuffSize); - - { int const tooSmall = (zds->inBuffSize < neededInBuffSize) || (zds->outBuffSize < neededOutBuffSize); - int const tooLarge = ZSTD_DCtx_isOversizedTooLong(zds); - - if (tooSmall || tooLarge) { - size_t const bufferSize = neededInBuffSize + neededOutBuffSize; - DEBUGLOG(4, "inBuff : from %u to %u", - (U32)zds->inBuffSize, (U32)neededInBuffSize); - DEBUGLOG(4, "outBuff : from %u to %u", - (U32)zds->outBuffSize, (U32)neededOutBuffSize); - if (zds->staticSize) { /* static DCtx */ - DEBUGLOG(4, "staticSize : %u", (U32)zds->staticSize); - assert(zds->staticSize >= sizeof(ZSTD_DCtx)); /* controlled at init */ - RETURN_ERROR_IF( - bufferSize > zds->staticSize - sizeof(ZSTD_DCtx), - memory_allocation, ""); - } else { - ZSTD_customFree(zds->inBuff, zds->customMem); - zds->inBuffSize = 0; - zds->outBuffSize = 0; - zds->inBuff = (char*)ZSTD_customMalloc(bufferSize, zds->customMem); - RETURN_ERROR_IF(zds->inBuff == NULL, memory_allocation, ""); - } - zds->inBuffSize = neededInBuffSize; - zds->outBuff = zds->inBuff + zds->inBuffSize; - zds->outBuffSize = neededOutBuffSize; - } } } - zds->streamStage = zdss_read; - ZSTD_FALLTHROUGH; - - case zdss_read: - DEBUGLOG(5, "stage zdss_read"); - { size_t const neededInSize = ZSTD_nextSrcSizeToDecompressWithInputSize(zds, (size_t)(iend - ip)); - DEBUGLOG(5, "neededInSize = %u", (U32)neededInSize); - if (neededInSize==0) { /* end of frame */ - zds->streamStage = zdss_init; - someMoreWork = 0; - break; - } - if ((size_t)(iend-ip) >= neededInSize) { /* decode directly from src */ - FORWARD_IF_ERROR(ZSTD_decompressContinueStream(zds, &op, oend, ip, neededInSize), ""); - ip += neededInSize; - /* Function modifies the stage so we must break */ - break; - } } - if (ip==iend) { someMoreWork = 0; break; } /* no more input */ - zds->streamStage = zdss_load; - ZSTD_FALLTHROUGH; - - case zdss_load: - { size_t const neededInSize = ZSTD_nextSrcSizeToDecompress(zds); - size_t const toLoad = neededInSize - zds->inPos; - int const isSkipFrame = ZSTD_isSkipFrame(zds); - size_t loadedSize; - /* At this point we shouldn't be decompressing a block that we can stream. */ - assert(neededInSize == ZSTD_nextSrcSizeToDecompressWithInputSize(zds, iend - ip)); - if (isSkipFrame) { - loadedSize = MIN(toLoad, (size_t)(iend-ip)); - } else { - RETURN_ERROR_IF(toLoad > zds->inBuffSize - zds->inPos, - corruption_detected, - "should never happen"); - loadedSize = ZSTD_limitCopy(zds->inBuff + zds->inPos, toLoad, ip, (size_t)(iend-ip)); - } - ip += loadedSize; - zds->inPos += loadedSize; - if (loadedSize < toLoad) { someMoreWork = 0; break; } /* not enough input, wait for more */ - - /* decode loaded input */ - zds->inPos = 0; /* input is consumed */ - FORWARD_IF_ERROR(ZSTD_decompressContinueStream(zds, &op, oend, zds->inBuff, neededInSize), ""); - /* Function modifies the stage so we must break */ - break; - } - case zdss_flush: - { size_t const toFlushSize = zds->outEnd - zds->outStart; - size_t const flushedSize = ZSTD_limitCopy(op, (size_t)(oend-op), zds->outBuff + zds->outStart, toFlushSize); - op += flushedSize; - zds->outStart += flushedSize; - if (flushedSize == toFlushSize) { /* flush completed */ - zds->streamStage = zdss_read; - if ( (zds->outBuffSize < zds->fParams.frameContentSize) - && (zds->outStart + zds->fParams.blockSizeMax > zds->outBuffSize) ) { - DEBUGLOG(5, "restart filling outBuff from beginning (left:%i, needed:%u)", - (int)(zds->outBuffSize - zds->outStart), - (U32)zds->fParams.blockSizeMax); - zds->outStart = zds->outEnd = 0; - } - break; - } } - /* cannot complete flush */ - someMoreWork = 0; - break; - - default: - assert(0); /* impossible */ - RETURN_ERROR(GENERIC, "impossible to reach"); /* some compiler require default to do something */ - } } - - /* result */ - input->pos = (size_t)(ip - (const char*)(input->src)); - output->pos = (size_t)(op - (char*)(output->dst)); - - /* Update the expected output buffer for ZSTD_obm_stable. */ - zds->expectedOutBuffer = *output; - - if ((ip==istart) && (op==ostart)) { /* no forward progress */ - zds->noForwardProgress ++; - if (zds->noForwardProgress >= ZSTD_NO_FORWARD_PROGRESS_MAX) { - RETURN_ERROR_IF(op==oend, dstSize_tooSmall, ""); - RETURN_ERROR_IF(ip==iend, srcSize_wrong, ""); - assert(0); - } - } else { - zds->noForwardProgress = 0; - } - { size_t nextSrcSizeHint = ZSTD_nextSrcSizeToDecompress(zds); - if (!nextSrcSizeHint) { /* frame fully decoded */ - if (zds->outEnd == zds->outStart) { /* output fully flushed */ - if (zds->hostageByte) { - if (input->pos >= input->size) { - /* can't release hostage (not present) */ - zds->streamStage = zdss_read; - return 1; - } - input->pos++; /* release hostage */ - } /* zds->hostageByte */ - return 0; - } /* zds->outEnd == zds->outStart */ - if (!zds->hostageByte) { /* output not fully flushed; keep last byte as hostage; will be released when all output is flushed */ - input->pos--; /* note : pos > 0, otherwise, impossible to finish reading last block */ - zds->hostageByte=1; - } - return 1; - } /* nextSrcSizeHint==0 */ - nextSrcSizeHint += ZSTD_blockHeaderSize * (ZSTD_nextInputType(zds) == ZSTDnit_block); /* preload header of next block */ - assert(zds->inPos <= nextSrcSizeHint); - nextSrcSizeHint -= zds->inPos; /* part already loaded*/ - return nextSrcSizeHint; - } -} - -size_t ZSTD_decompressStream_simpleArgs ( - ZSTD_DCtx* dctx, - void* dst, size_t dstCapacity, size_t* dstPos, - const void* src, size_t srcSize, size_t* srcPos) -{ - ZSTD_outBuffer output = { dst, dstCapacity, *dstPos }; - ZSTD_inBuffer input = { src, srcSize, *srcPos }; - /* ZSTD_compress_generic() will check validity of dstPos and srcPos */ - size_t const cErr = ZSTD_decompressStream(dctx, &output, &input); - *dstPos = output.pos; - *srcPos = input.pos; - return cErr; -} diff --git a/dep/zstd/lib/decompress/zstd_decompress_block.c b/dep/zstd/lib/decompress/zstd_decompress_block.c deleted file mode 100644 index 2e44d30d2..000000000 --- a/dep/zstd/lib/decompress/zstd_decompress_block.c +++ /dev/null @@ -1,2072 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -/* zstd_decompress_block : - * this module takes care of decompressing _compressed_ block */ - -/*-******************************************************* -* Dependencies -*********************************************************/ -#include "../common/zstd_deps.h" /* ZSTD_memcpy, ZSTD_memmove, ZSTD_memset */ -#include "../common/compiler.h" /* prefetch */ -#include "../common/cpu.h" /* bmi2 */ -#include "../common/mem.h" /* low level memory routines */ -#define FSE_STATIC_LINKING_ONLY -#include "../common/fse.h" -#define HUF_STATIC_LINKING_ONLY -#include "../common/huf.h" -#include "../common/zstd_internal.h" -#include "zstd_decompress_internal.h" /* ZSTD_DCtx */ -#include "zstd_ddict.h" /* ZSTD_DDictDictContent */ -#include "zstd_decompress_block.h" - -/*_******************************************************* -* Macros -**********************************************************/ - -/* These two optional macros force the use one way or another of the two - * ZSTD_decompressSequences implementations. You can't force in both directions - * at the same time. - */ -#if defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT) && \ - defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG) -#error "Cannot force the use of the short and the long ZSTD_decompressSequences variants!" -#endif - - -/*_******************************************************* -* Memory operations -**********************************************************/ -static void ZSTD_copy4(void* dst, const void* src) { ZSTD_memcpy(dst, src, 4); } - - -/*-************************************************************* - * Block decoding - ***************************************************************/ - -/*! ZSTD_getcBlockSize() : - * Provides the size of compressed block from block header `src` */ -size_t ZSTD_getcBlockSize(const void* src, size_t srcSize, - blockProperties_t* bpPtr) -{ - RETURN_ERROR_IF(srcSize < ZSTD_blockHeaderSize, srcSize_wrong, ""); - - { U32 const cBlockHeader = MEM_readLE24(src); - U32 const cSize = cBlockHeader >> 3; - bpPtr->lastBlock = cBlockHeader & 1; - bpPtr->blockType = (blockType_e)((cBlockHeader >> 1) & 3); - bpPtr->origSize = cSize; /* only useful for RLE */ - if (bpPtr->blockType == bt_rle) return 1; - RETURN_ERROR_IF(bpPtr->blockType == bt_reserved, corruption_detected, ""); - return cSize; - } -} - -/* Allocate buffer for literals, either overlapping current dst, or split between dst and litExtraBuffer, or stored entirely within litExtraBuffer */ -static void ZSTD_allocateLiteralsBuffer(ZSTD_DCtx* dctx, void* const dst, const size_t dstCapacity, const size_t litSize, - const streaming_operation streaming, const size_t expectedWriteSize, const unsigned splitImmediately) -{ - if (streaming == not_streaming && dstCapacity > ZSTD_BLOCKSIZE_MAX + WILDCOPY_OVERLENGTH + litSize + WILDCOPY_OVERLENGTH) - { - /* room for litbuffer to fit without read faulting */ - dctx->litBuffer = (BYTE*)dst + ZSTD_BLOCKSIZE_MAX + WILDCOPY_OVERLENGTH; - dctx->litBufferEnd = dctx->litBuffer + litSize; - dctx->litBufferLocation = ZSTD_in_dst; - } - else if (litSize > ZSTD_LITBUFFEREXTRASIZE) - { - /* won't fit in litExtraBuffer, so it will be split between end of dst and extra buffer */ - if (splitImmediately) { - /* won't fit in litExtraBuffer, so it will be split between end of dst and extra buffer */ - dctx->litBuffer = (BYTE*)dst + expectedWriteSize - litSize + ZSTD_LITBUFFEREXTRASIZE - WILDCOPY_OVERLENGTH; - dctx->litBufferEnd = dctx->litBuffer + litSize - ZSTD_LITBUFFEREXTRASIZE; - } - else { - /* initially this will be stored entirely in dst during huffman decoding, it will partially shifted to litExtraBuffer after */ - dctx->litBuffer = (BYTE*)dst + expectedWriteSize - litSize; - dctx->litBufferEnd = (BYTE*)dst + expectedWriteSize; - } - dctx->litBufferLocation = ZSTD_split; - } - else - { - /* fits entirely within litExtraBuffer, so no split is necessary */ - dctx->litBuffer = dctx->litExtraBuffer; - dctx->litBufferEnd = dctx->litBuffer + litSize; - dctx->litBufferLocation = ZSTD_not_in_dst; - } -} - -/* Hidden declaration for fullbench */ -size_t ZSTD_decodeLiteralsBlock(ZSTD_DCtx* dctx, - const void* src, size_t srcSize, - void* dst, size_t dstCapacity, const streaming_operation streaming); -/*! ZSTD_decodeLiteralsBlock() : - * Where it is possible to do so without being stomped by the output during decompression, the literals block will be stored - * in the dstBuffer. If there is room to do so, it will be stored in full in the excess dst space after where the current - * block will be output. Otherwise it will be stored at the end of the current dst blockspace, with a small portion being - * stored in dctx->litExtraBuffer to help keep it "ahead" of the current output write. - * - * @return : nb of bytes read from src (< srcSize ) - * note : symbol not declared but exposed for fullbench */ -size_t ZSTD_decodeLiteralsBlock(ZSTD_DCtx* dctx, - const void* src, size_t srcSize, /* note : srcSize < BLOCKSIZE */ - void* dst, size_t dstCapacity, const streaming_operation streaming) -{ - DEBUGLOG(5, "ZSTD_decodeLiteralsBlock"); - RETURN_ERROR_IF(srcSize < MIN_CBLOCK_SIZE, corruption_detected, ""); - - { const BYTE* const istart = (const BYTE*) src; - symbolEncodingType_e const litEncType = (symbolEncodingType_e)(istart[0] & 3); - - switch(litEncType) - { - case set_repeat: - DEBUGLOG(5, "set_repeat flag : re-using stats from previous compressed literals block"); - RETURN_ERROR_IF(dctx->litEntropy==0, dictionary_corrupted, ""); - ZSTD_FALLTHROUGH; - - case set_compressed: - RETURN_ERROR_IF(srcSize < 5, corruption_detected, "srcSize >= MIN_CBLOCK_SIZE == 3; here we need up to 5 for case 3"); - { size_t lhSize, litSize, litCSize; - U32 singleStream=0; - U32 const lhlCode = (istart[0] >> 2) & 3; - U32 const lhc = MEM_readLE32(istart); - size_t hufSuccess; - size_t expectedWriteSize = MIN(ZSTD_BLOCKSIZE_MAX, dstCapacity); - switch(lhlCode) - { - case 0: case 1: default: /* note : default is impossible, since lhlCode into [0..3] */ - /* 2 - 2 - 10 - 10 */ - singleStream = !lhlCode; - lhSize = 3; - litSize = (lhc >> 4) & 0x3FF; - litCSize = (lhc >> 14) & 0x3FF; - break; - case 2: - /* 2 - 2 - 14 - 14 */ - lhSize = 4; - litSize = (lhc >> 4) & 0x3FFF; - litCSize = lhc >> 18; - break; - case 3: - /* 2 - 2 - 18 - 18 */ - lhSize = 5; - litSize = (lhc >> 4) & 0x3FFFF; - litCSize = (lhc >> 22) + ((size_t)istart[4] << 10); - break; - } - RETURN_ERROR_IF(litSize > 0 && dst == NULL, dstSize_tooSmall, "NULL not handled"); - RETURN_ERROR_IF(litSize > ZSTD_BLOCKSIZE_MAX, corruption_detected, ""); - RETURN_ERROR_IF(litCSize + lhSize > srcSize, corruption_detected, ""); - RETURN_ERROR_IF(expectedWriteSize < litSize , dstSize_tooSmall, ""); - ZSTD_allocateLiteralsBuffer(dctx, dst, dstCapacity, litSize, streaming, expectedWriteSize, 0); - - /* prefetch huffman table if cold */ - if (dctx->ddictIsCold && (litSize > 768 /* heuristic */)) { - PREFETCH_AREA(dctx->HUFptr, sizeof(dctx->entropy.hufTable)); - } - - if (litEncType==set_repeat) { - if (singleStream) { - hufSuccess = HUF_decompress1X_usingDTable_bmi2( - dctx->litBuffer, litSize, istart+lhSize, litCSize, - dctx->HUFptr, ZSTD_DCtx_get_bmi2(dctx)); - } else { - hufSuccess = HUF_decompress4X_usingDTable_bmi2( - dctx->litBuffer, litSize, istart+lhSize, litCSize, - dctx->HUFptr, ZSTD_DCtx_get_bmi2(dctx)); - } - } else { - if (singleStream) { -#if defined(HUF_FORCE_DECOMPRESS_X2) - hufSuccess = HUF_decompress1X_DCtx_wksp( - dctx->entropy.hufTable, dctx->litBuffer, litSize, - istart+lhSize, litCSize, dctx->workspace, - sizeof(dctx->workspace)); -#else - hufSuccess = HUF_decompress1X1_DCtx_wksp_bmi2( - dctx->entropy.hufTable, dctx->litBuffer, litSize, - istart+lhSize, litCSize, dctx->workspace, - sizeof(dctx->workspace), ZSTD_DCtx_get_bmi2(dctx)); -#endif - } else { - hufSuccess = HUF_decompress4X_hufOnly_wksp_bmi2( - dctx->entropy.hufTable, dctx->litBuffer, litSize, - istart+lhSize, litCSize, dctx->workspace, - sizeof(dctx->workspace), ZSTD_DCtx_get_bmi2(dctx)); - } - } - if (dctx->litBufferLocation == ZSTD_split) - { - ZSTD_memcpy(dctx->litExtraBuffer, dctx->litBufferEnd - ZSTD_LITBUFFEREXTRASIZE, ZSTD_LITBUFFEREXTRASIZE); - ZSTD_memmove(dctx->litBuffer + ZSTD_LITBUFFEREXTRASIZE - WILDCOPY_OVERLENGTH, dctx->litBuffer, litSize - ZSTD_LITBUFFEREXTRASIZE); - dctx->litBuffer += ZSTD_LITBUFFEREXTRASIZE - WILDCOPY_OVERLENGTH; - dctx->litBufferEnd -= WILDCOPY_OVERLENGTH; - } - - RETURN_ERROR_IF(HUF_isError(hufSuccess), corruption_detected, ""); - - dctx->litPtr = dctx->litBuffer; - dctx->litSize = litSize; - dctx->litEntropy = 1; - if (litEncType==set_compressed) dctx->HUFptr = dctx->entropy.hufTable; - return litCSize + lhSize; - } - - case set_basic: - { size_t litSize, lhSize; - U32 const lhlCode = ((istart[0]) >> 2) & 3; - size_t expectedWriteSize = MIN(ZSTD_BLOCKSIZE_MAX, dstCapacity); - switch(lhlCode) - { - case 0: case 2: default: /* note : default is impossible, since lhlCode into [0..3] */ - lhSize = 1; - litSize = istart[0] >> 3; - break; - case 1: - lhSize = 2; - litSize = MEM_readLE16(istart) >> 4; - break; - case 3: - lhSize = 3; - litSize = MEM_readLE24(istart) >> 4; - break; - } - - RETURN_ERROR_IF(litSize > 0 && dst == NULL, dstSize_tooSmall, "NULL not handled"); - RETURN_ERROR_IF(expectedWriteSize < litSize, dstSize_tooSmall, ""); - ZSTD_allocateLiteralsBuffer(dctx, dst, dstCapacity, litSize, streaming, expectedWriteSize, 1); - if (lhSize+litSize+WILDCOPY_OVERLENGTH > srcSize) { /* risk reading beyond src buffer with wildcopy */ - RETURN_ERROR_IF(litSize+lhSize > srcSize, corruption_detected, ""); - if (dctx->litBufferLocation == ZSTD_split) - { - ZSTD_memcpy(dctx->litBuffer, istart + lhSize, litSize - ZSTD_LITBUFFEREXTRASIZE); - ZSTD_memcpy(dctx->litExtraBuffer, istart + lhSize + litSize - ZSTD_LITBUFFEREXTRASIZE, ZSTD_LITBUFFEREXTRASIZE); - } - else - { - ZSTD_memcpy(dctx->litBuffer, istart + lhSize, litSize); - } - dctx->litPtr = dctx->litBuffer; - dctx->litSize = litSize; - return lhSize+litSize; - } - /* direct reference into compressed stream */ - dctx->litPtr = istart+lhSize; - dctx->litSize = litSize; - dctx->litBufferEnd = dctx->litPtr + litSize; - dctx->litBufferLocation = ZSTD_not_in_dst; - return lhSize+litSize; - } - - case set_rle: - { U32 const lhlCode = ((istart[0]) >> 2) & 3; - size_t litSize, lhSize; - size_t expectedWriteSize = MIN(ZSTD_BLOCKSIZE_MAX, dstCapacity); - switch(lhlCode) - { - case 0: case 2: default: /* note : default is impossible, since lhlCode into [0..3] */ - lhSize = 1; - litSize = istart[0] >> 3; - break; - case 1: - lhSize = 2; - litSize = MEM_readLE16(istart) >> 4; - break; - case 3: - lhSize = 3; - litSize = MEM_readLE24(istart) >> 4; - RETURN_ERROR_IF(srcSize<4, corruption_detected, "srcSize >= MIN_CBLOCK_SIZE == 3; here we need lhSize+1 = 4"); - break; - } - RETURN_ERROR_IF(litSize > 0 && dst == NULL, dstSize_tooSmall, "NULL not handled"); - RETURN_ERROR_IF(litSize > ZSTD_BLOCKSIZE_MAX, corruption_detected, ""); - RETURN_ERROR_IF(expectedWriteSize < litSize, dstSize_tooSmall, ""); - ZSTD_allocateLiteralsBuffer(dctx, dst, dstCapacity, litSize, streaming, expectedWriteSize, 1); - if (dctx->litBufferLocation == ZSTD_split) - { - ZSTD_memset(dctx->litBuffer, istart[lhSize], litSize - ZSTD_LITBUFFEREXTRASIZE); - ZSTD_memset(dctx->litExtraBuffer, istart[lhSize], ZSTD_LITBUFFEREXTRASIZE); - } - else - { - ZSTD_memset(dctx->litBuffer, istart[lhSize], litSize); - } - dctx->litPtr = dctx->litBuffer; - dctx->litSize = litSize; - return lhSize+1; - } - default: - RETURN_ERROR(corruption_detected, "impossible"); - } - } -} - -/* Default FSE distribution tables. - * These are pre-calculated FSE decoding tables using default distributions as defined in specification : - * https://github.com/facebook/zstd/blob/release/doc/zstd_compression_format.md#default-distributions - * They were generated programmatically with following method : - * - start from default distributions, present in /lib/common/zstd_internal.h - * - generate tables normally, using ZSTD_buildFSETable() - * - printout the content of tables - * - pretify output, report below, test with fuzzer to ensure it's correct */ - -/* Default FSE distribution table for Literal Lengths */ -static const ZSTD_seqSymbol LL_defaultDTable[(1<tableLog = 0; - DTableH->fastMode = 0; - - cell->nbBits = 0; - cell->nextState = 0; - assert(nbAddBits < 255); - cell->nbAdditionalBits = nbAddBits; - cell->baseValue = baseValue; -} - - -/* ZSTD_buildFSETable() : - * generate FSE decoding table for one symbol (ll, ml or off) - * cannot fail if input is valid => - * all inputs are presumed validated at this stage */ -FORCE_INLINE_TEMPLATE -void ZSTD_buildFSETable_body(ZSTD_seqSymbol* dt, - const short* normalizedCounter, unsigned maxSymbolValue, - const U32* baseValue, const U8* nbAdditionalBits, - unsigned tableLog, void* wksp, size_t wkspSize) -{ - ZSTD_seqSymbol* const tableDecode = dt+1; - U32 const maxSV1 = maxSymbolValue + 1; - U32 const tableSize = 1 << tableLog; - - U16* symbolNext = (U16*)wksp; - BYTE* spread = (BYTE*)(symbolNext + MaxSeq + 1); - U32 highThreshold = tableSize - 1; - - - /* Sanity Checks */ - assert(maxSymbolValue <= MaxSeq); - assert(tableLog <= MaxFSELog); - assert(wkspSize >= ZSTD_BUILD_FSE_TABLE_WKSP_SIZE); - (void)wkspSize; - /* Init, lay down lowprob symbols */ - { ZSTD_seqSymbol_header DTableH; - DTableH.tableLog = tableLog; - DTableH.fastMode = 1; - { S16 const largeLimit= (S16)(1 << (tableLog-1)); - U32 s; - for (s=0; s= largeLimit) DTableH.fastMode=0; - assert(normalizedCounter[s]>=0); - symbolNext[s] = (U16)normalizedCounter[s]; - } } } - ZSTD_memcpy(dt, &DTableH, sizeof(DTableH)); - } - - /* Spread symbols */ - assert(tableSize <= 512); - /* Specialized symbol spreading for the case when there are - * no low probability (-1 count) symbols. When compressing - * small blocks we avoid low probability symbols to hit this - * case, since header decoding speed matters more. - */ - if (highThreshold == tableSize - 1) { - size_t const tableMask = tableSize-1; - size_t const step = FSE_TABLESTEP(tableSize); - /* First lay down the symbols in order. - * We use a uint64_t to lay down 8 bytes at a time. This reduces branch - * misses since small blocks generally have small table logs, so nearly - * all symbols have counts <= 8. We ensure we have 8 bytes at the end of - * our buffer to handle the over-write. - */ - { - U64 const add = 0x0101010101010101ull; - size_t pos = 0; - U64 sv = 0; - U32 s; - for (s=0; s highThreshold) position = (position + step) & tableMask; /* lowprob area */ - } } - assert(position == 0); /* position must reach all cells once, otherwise normalizedCounter is incorrect */ - } - - /* Build Decoding table */ - { - U32 u; - for (u=0; u max, corruption_detected, ""); - { U32 const symbol = *(const BYTE*)src; - U32 const baseline = baseValue[symbol]; - U8 const nbBits = nbAdditionalBits[symbol]; - ZSTD_buildSeqTable_rle(DTableSpace, baseline, nbBits); - } - *DTablePtr = DTableSpace; - return 1; - case set_basic : - *DTablePtr = defaultTable; - return 0; - case set_repeat: - RETURN_ERROR_IF(!flagRepeatTable, corruption_detected, ""); - /* prefetch FSE table if used */ - if (ddictIsCold && (nbSeq > 24 /* heuristic */)) { - const void* const pStart = *DTablePtr; - size_t const pSize = sizeof(ZSTD_seqSymbol) * (SEQSYMBOL_TABLE_SIZE(maxLog)); - PREFETCH_AREA(pStart, pSize); - } - return 0; - case set_compressed : - { unsigned tableLog; - S16 norm[MaxSeq+1]; - size_t const headerSize = FSE_readNCount(norm, &max, &tableLog, src, srcSize); - RETURN_ERROR_IF(FSE_isError(headerSize), corruption_detected, ""); - RETURN_ERROR_IF(tableLog > maxLog, corruption_detected, ""); - ZSTD_buildFSETable(DTableSpace, norm, max, baseValue, nbAdditionalBits, tableLog, wksp, wkspSize, bmi2); - *DTablePtr = DTableSpace; - return headerSize; - } - default : - assert(0); - RETURN_ERROR(GENERIC, "impossible"); - } -} - -size_t ZSTD_decodeSeqHeaders(ZSTD_DCtx* dctx, int* nbSeqPtr, - const void* src, size_t srcSize) -{ - const BYTE* const istart = (const BYTE*)src; - const BYTE* const iend = istart + srcSize; - const BYTE* ip = istart; - int nbSeq; - DEBUGLOG(5, "ZSTD_decodeSeqHeaders"); - - /* check */ - RETURN_ERROR_IF(srcSize < MIN_SEQUENCES_SIZE, srcSize_wrong, ""); - - /* SeqHead */ - nbSeq = *ip++; - if (!nbSeq) { - *nbSeqPtr=0; - RETURN_ERROR_IF(srcSize != 1, srcSize_wrong, ""); - return 1; - } - if (nbSeq > 0x7F) { - if (nbSeq == 0xFF) { - RETURN_ERROR_IF(ip+2 > iend, srcSize_wrong, ""); - nbSeq = MEM_readLE16(ip) + LONGNBSEQ; - ip+=2; - } else { - RETURN_ERROR_IF(ip >= iend, srcSize_wrong, ""); - nbSeq = ((nbSeq-0x80)<<8) + *ip++; - } - } - *nbSeqPtr = nbSeq; - - /* FSE table descriptors */ - RETURN_ERROR_IF(ip+1 > iend, srcSize_wrong, ""); /* minimum possible size: 1 byte for symbol encoding types */ - { symbolEncodingType_e const LLtype = (symbolEncodingType_e)(*ip >> 6); - symbolEncodingType_e const OFtype = (symbolEncodingType_e)((*ip >> 4) & 3); - symbolEncodingType_e const MLtype = (symbolEncodingType_e)((*ip >> 2) & 3); - ip++; - - /* Build DTables */ - { size_t const llhSize = ZSTD_buildSeqTable(dctx->entropy.LLTable, &dctx->LLTptr, - LLtype, MaxLL, LLFSELog, - ip, iend-ip, - LL_base, LL_bits, - LL_defaultDTable, dctx->fseEntropy, - dctx->ddictIsCold, nbSeq, - dctx->workspace, sizeof(dctx->workspace), - ZSTD_DCtx_get_bmi2(dctx)); - RETURN_ERROR_IF(ZSTD_isError(llhSize), corruption_detected, "ZSTD_buildSeqTable failed"); - ip += llhSize; - } - - { size_t const ofhSize = ZSTD_buildSeqTable(dctx->entropy.OFTable, &dctx->OFTptr, - OFtype, MaxOff, OffFSELog, - ip, iend-ip, - OF_base, OF_bits, - OF_defaultDTable, dctx->fseEntropy, - dctx->ddictIsCold, nbSeq, - dctx->workspace, sizeof(dctx->workspace), - ZSTD_DCtx_get_bmi2(dctx)); - RETURN_ERROR_IF(ZSTD_isError(ofhSize), corruption_detected, "ZSTD_buildSeqTable failed"); - ip += ofhSize; - } - - { size_t const mlhSize = ZSTD_buildSeqTable(dctx->entropy.MLTable, &dctx->MLTptr, - MLtype, MaxML, MLFSELog, - ip, iend-ip, - ML_base, ML_bits, - ML_defaultDTable, dctx->fseEntropy, - dctx->ddictIsCold, nbSeq, - dctx->workspace, sizeof(dctx->workspace), - ZSTD_DCtx_get_bmi2(dctx)); - RETURN_ERROR_IF(ZSTD_isError(mlhSize), corruption_detected, "ZSTD_buildSeqTable failed"); - ip += mlhSize; - } - } - - return ip-istart; -} - - -typedef struct { - size_t litLength; - size_t matchLength; - size_t offset; -} seq_t; - -typedef struct { - size_t state; - const ZSTD_seqSymbol* table; -} ZSTD_fseState; - -typedef struct { - BIT_DStream_t DStream; - ZSTD_fseState stateLL; - ZSTD_fseState stateOffb; - ZSTD_fseState stateML; - size_t prevOffset[ZSTD_REP_NUM]; -} seqState_t; - -/*! ZSTD_overlapCopy8() : - * Copies 8 bytes from ip to op and updates op and ip where ip <= op. - * If the offset is < 8 then the offset is spread to at least 8 bytes. - * - * Precondition: *ip <= *op - * Postcondition: *op - *op >= 8 - */ -HINT_INLINE void ZSTD_overlapCopy8(BYTE** op, BYTE const** ip, size_t offset) { - assert(*ip <= *op); - if (offset < 8) { - /* close range match, overlap */ - static const U32 dec32table[] = { 0, 1, 2, 1, 4, 4, 4, 4 }; /* added */ - static const int dec64table[] = { 8, 8, 8, 7, 8, 9,10,11 }; /* subtracted */ - int const sub2 = dec64table[offset]; - (*op)[0] = (*ip)[0]; - (*op)[1] = (*ip)[1]; - (*op)[2] = (*ip)[2]; - (*op)[3] = (*ip)[3]; - *ip += dec32table[offset]; - ZSTD_copy4(*op+4, *ip); - *ip -= sub2; - } else { - ZSTD_copy8(*op, *ip); - } - *ip += 8; - *op += 8; - assert(*op - *ip >= 8); -} - -/*! ZSTD_safecopy() : - * Specialized version of memcpy() that is allowed to READ up to WILDCOPY_OVERLENGTH past the input buffer - * and write up to 16 bytes past oend_w (op >= oend_w is allowed). - * This function is only called in the uncommon case where the sequence is near the end of the block. It - * should be fast for a single long sequence, but can be slow for several short sequences. - * - * @param ovtype controls the overlap detection - * - ZSTD_no_overlap: The source and destination are guaranteed to be at least WILDCOPY_VECLEN bytes apart. - * - ZSTD_overlap_src_before_dst: The src and dst may overlap and may be any distance apart. - * The src buffer must be before the dst buffer. - */ -static void ZSTD_safecopy(BYTE* op, const BYTE* const oend_w, BYTE const* ip, ptrdiff_t length, ZSTD_overlap_e ovtype) { - ptrdiff_t const diff = op - ip; - BYTE* const oend = op + length; - - assert((ovtype == ZSTD_no_overlap && (diff <= -8 || diff >= 8 || op >= oend_w)) || - (ovtype == ZSTD_overlap_src_before_dst && diff >= 0)); - - if (length < 8) { - /* Handle short lengths. */ - while (op < oend) *op++ = *ip++; - return; - } - if (ovtype == ZSTD_overlap_src_before_dst) { - /* Copy 8 bytes and ensure the offset >= 8 when there can be overlap. */ - assert(length >= 8); - ZSTD_overlapCopy8(&op, &ip, diff); - length -= 8; - assert(op - ip >= 8); - assert(op <= oend); - } - - if (oend <= oend_w) { - /* No risk of overwrite. */ - ZSTD_wildcopy(op, ip, length, ovtype); - return; - } - if (op <= oend_w) { - /* Wildcopy until we get close to the end. */ - assert(oend > oend_w); - ZSTD_wildcopy(op, ip, oend_w - op, ovtype); - ip += oend_w - op; - op += oend_w - op; - } - /* Handle the leftovers. */ - while (op < oend) *op++ = *ip++; -} - -/* ZSTD_safecopyDstBeforeSrc(): - * This version allows overlap with dst before src, or handles the non-overlap case with dst after src - * Kept separate from more common ZSTD_safecopy case to avoid performance impact to the safecopy common case */ -static void ZSTD_safecopyDstBeforeSrc(BYTE* op, BYTE const* ip, ptrdiff_t length) { - ptrdiff_t const diff = op - ip; - BYTE* const oend = op + length; - - if (length < 8 || diff > -8) { - /* Handle short lengths, close overlaps, and dst not before src. */ - while (op < oend) *op++ = *ip++; - return; - } - - if (op <= oend - WILDCOPY_OVERLENGTH && diff < -WILDCOPY_VECLEN) { - ZSTD_wildcopy(op, ip, oend - WILDCOPY_OVERLENGTH - op, ZSTD_no_overlap); - ip += oend - WILDCOPY_OVERLENGTH - op; - op += oend - WILDCOPY_OVERLENGTH - op; - } - - /* Handle the leftovers. */ - while (op < oend) *op++ = *ip++; -} - -/* ZSTD_execSequenceEnd(): - * This version handles cases that are near the end of the output buffer. It requires - * more careful checks to make sure there is no overflow. By separating out these hard - * and unlikely cases, we can speed up the common cases. - * - * NOTE: This function needs to be fast for a single long sequence, but doesn't need - * to be optimized for many small sequences, since those fall into ZSTD_execSequence(). - */ -FORCE_NOINLINE -size_t ZSTD_execSequenceEnd(BYTE* op, - BYTE* const oend, seq_t sequence, - const BYTE** litPtr, const BYTE* const litLimit, - const BYTE* const prefixStart, const BYTE* const virtualStart, const BYTE* const dictEnd) -{ - BYTE* const oLitEnd = op + sequence.litLength; - size_t const sequenceLength = sequence.litLength + sequence.matchLength; - const BYTE* const iLitEnd = *litPtr + sequence.litLength; - const BYTE* match = oLitEnd - sequence.offset; - BYTE* const oend_w = oend - WILDCOPY_OVERLENGTH; - - /* bounds checks : careful of address space overflow in 32-bit mode */ - RETURN_ERROR_IF(sequenceLength > (size_t)(oend - op), dstSize_tooSmall, "last match must fit within dstBuffer"); - RETURN_ERROR_IF(sequence.litLength > (size_t)(litLimit - *litPtr), corruption_detected, "try to read beyond literal buffer"); - assert(op < op + sequenceLength); - assert(oLitEnd < op + sequenceLength); - - /* copy literals */ - ZSTD_safecopy(op, oend_w, *litPtr, sequence.litLength, ZSTD_no_overlap); - op = oLitEnd; - *litPtr = iLitEnd; - - /* copy Match */ - if (sequence.offset > (size_t)(oLitEnd - prefixStart)) { - /* offset beyond prefix */ - RETURN_ERROR_IF(sequence.offset > (size_t)(oLitEnd - virtualStart), corruption_detected, ""); - match = dictEnd - (prefixStart - match); - if (match + sequence.matchLength <= dictEnd) { - ZSTD_memmove(oLitEnd, match, sequence.matchLength); - return sequenceLength; - } - /* span extDict & currentPrefixSegment */ - { size_t const length1 = dictEnd - match; - ZSTD_memmove(oLitEnd, match, length1); - op = oLitEnd + length1; - sequence.matchLength -= length1; - match = prefixStart; - } - } - ZSTD_safecopy(op, oend_w, match, sequence.matchLength, ZSTD_overlap_src_before_dst); - return sequenceLength; -} - -/* ZSTD_execSequenceEndSplitLitBuffer(): - * This version is intended to be used during instances where the litBuffer is still split. It is kept separate to avoid performance impact for the good case. - */ -FORCE_NOINLINE -size_t ZSTD_execSequenceEndSplitLitBuffer(BYTE* op, - BYTE* const oend, const BYTE* const oend_w, seq_t sequence, - const BYTE** litPtr, const BYTE* const litLimit, - const BYTE* const prefixStart, const BYTE* const virtualStart, const BYTE* const dictEnd) -{ - BYTE* const oLitEnd = op + sequence.litLength; - size_t const sequenceLength = sequence.litLength + sequence.matchLength; - const BYTE* const iLitEnd = *litPtr + sequence.litLength; - const BYTE* match = oLitEnd - sequence.offset; - - - /* bounds checks : careful of address space overflow in 32-bit mode */ - RETURN_ERROR_IF(sequenceLength > (size_t)(oend - op), dstSize_tooSmall, "last match must fit within dstBuffer"); - RETURN_ERROR_IF(sequence.litLength > (size_t)(litLimit - *litPtr), corruption_detected, "try to read beyond literal buffer"); - assert(op < op + sequenceLength); - assert(oLitEnd < op + sequenceLength); - - /* copy literals */ - RETURN_ERROR_IF(op > *litPtr && op < *litPtr + sequence.litLength, dstSize_tooSmall, "output should not catch up to and overwrite literal buffer"); - ZSTD_safecopyDstBeforeSrc(op, *litPtr, sequence.litLength); - op = oLitEnd; - *litPtr = iLitEnd; - - /* copy Match */ - if (sequence.offset > (size_t)(oLitEnd - prefixStart)) { - /* offset beyond prefix */ - RETURN_ERROR_IF(sequence.offset > (size_t)(oLitEnd - virtualStart), corruption_detected, ""); - match = dictEnd - (prefixStart - match); - if (match + sequence.matchLength <= dictEnd) { - ZSTD_memmove(oLitEnd, match, sequence.matchLength); - return sequenceLength; - } - /* span extDict & currentPrefixSegment */ - { size_t const length1 = dictEnd - match; - ZSTD_memmove(oLitEnd, match, length1); - op = oLitEnd + length1; - sequence.matchLength -= length1; - match = prefixStart; - } - } - ZSTD_safecopy(op, oend_w, match, sequence.matchLength, ZSTD_overlap_src_before_dst); - return sequenceLength; -} - -HINT_INLINE -size_t ZSTD_execSequence(BYTE* op, - BYTE* const oend, seq_t sequence, - const BYTE** litPtr, const BYTE* const litLimit, - const BYTE* const prefixStart, const BYTE* const virtualStart, const BYTE* const dictEnd) -{ - BYTE* const oLitEnd = op + sequence.litLength; - size_t const sequenceLength = sequence.litLength + sequence.matchLength; - BYTE* const oMatchEnd = op + sequenceLength; /* risk : address space overflow (32-bits) */ - BYTE* const oend_w = oend - WILDCOPY_OVERLENGTH; /* risk : address space underflow on oend=NULL */ - const BYTE* const iLitEnd = *litPtr + sequence.litLength; - const BYTE* match = oLitEnd - sequence.offset; - - assert(op != NULL /* Precondition */); - assert(oend_w < oend /* No underflow */); - /* Handle edge cases in a slow path: - * - Read beyond end of literals - * - Match end is within WILDCOPY_OVERLIMIT of oend - * - 32-bit mode and the match length overflows - */ - if (UNLIKELY( - iLitEnd > litLimit || - oMatchEnd > oend_w || - (MEM_32bits() && (size_t)(oend - op) < sequenceLength + WILDCOPY_OVERLENGTH))) - return ZSTD_execSequenceEnd(op, oend, sequence, litPtr, litLimit, prefixStart, virtualStart, dictEnd); - - /* Assumptions (everything else goes into ZSTD_execSequenceEnd()) */ - assert(op <= oLitEnd /* No overflow */); - assert(oLitEnd < oMatchEnd /* Non-zero match & no overflow */); - assert(oMatchEnd <= oend /* No underflow */); - assert(iLitEnd <= litLimit /* Literal length is in bounds */); - assert(oLitEnd <= oend_w /* Can wildcopy literals */); - assert(oMatchEnd <= oend_w /* Can wildcopy matches */); - - /* Copy Literals: - * Split out litLength <= 16 since it is nearly always true. +1.6% on gcc-9. - * We likely don't need the full 32-byte wildcopy. - */ - assert(WILDCOPY_OVERLENGTH >= 16); - ZSTD_copy16(op, (*litPtr)); - if (UNLIKELY(sequence.litLength > 16)) { - ZSTD_wildcopy(op + 16, (*litPtr) + 16, sequence.litLength - 16, ZSTD_no_overlap); - } - op = oLitEnd; - *litPtr = iLitEnd; /* update for next sequence */ - - /* Copy Match */ - if (sequence.offset > (size_t)(oLitEnd - prefixStart)) { - /* offset beyond prefix -> go into extDict */ - RETURN_ERROR_IF(UNLIKELY(sequence.offset > (size_t)(oLitEnd - virtualStart)), corruption_detected, ""); - match = dictEnd + (match - prefixStart); - if (match + sequence.matchLength <= dictEnd) { - ZSTD_memmove(oLitEnd, match, sequence.matchLength); - return sequenceLength; - } - /* span extDict & currentPrefixSegment */ - { size_t const length1 = dictEnd - match; - ZSTD_memmove(oLitEnd, match, length1); - op = oLitEnd + length1; - sequence.matchLength -= length1; - match = prefixStart; - } - } - /* Match within prefix of 1 or more bytes */ - assert(op <= oMatchEnd); - assert(oMatchEnd <= oend_w); - assert(match >= prefixStart); - assert(sequence.matchLength >= 1); - - /* Nearly all offsets are >= WILDCOPY_VECLEN bytes, which means we can use wildcopy - * without overlap checking. - */ - if (LIKELY(sequence.offset >= WILDCOPY_VECLEN)) { - /* We bet on a full wildcopy for matches, since we expect matches to be - * longer than literals (in general). In silesia, ~10% of matches are longer - * than 16 bytes. - */ - ZSTD_wildcopy(op, match, (ptrdiff_t)sequence.matchLength, ZSTD_no_overlap); - return sequenceLength; - } - assert(sequence.offset < WILDCOPY_VECLEN); - - /* Copy 8 bytes and spread the offset to be >= 8. */ - ZSTD_overlapCopy8(&op, &match, sequence.offset); - - /* If the match length is > 8 bytes, then continue with the wildcopy. */ - if (sequence.matchLength > 8) { - assert(op < oMatchEnd); - ZSTD_wildcopy(op, match, (ptrdiff_t)sequence.matchLength - 8, ZSTD_overlap_src_before_dst); - } - return sequenceLength; -} - -HINT_INLINE -size_t ZSTD_execSequenceSplitLitBuffer(BYTE* op, - BYTE* const oend, const BYTE* const oend_w, seq_t sequence, - const BYTE** litPtr, const BYTE* const litLimit, - const BYTE* const prefixStart, const BYTE* const virtualStart, const BYTE* const dictEnd) -{ - BYTE* const oLitEnd = op + sequence.litLength; - size_t const sequenceLength = sequence.litLength + sequence.matchLength; - BYTE* const oMatchEnd = op + sequenceLength; /* risk : address space overflow (32-bits) */ - const BYTE* const iLitEnd = *litPtr + sequence.litLength; - const BYTE* match = oLitEnd - sequence.offset; - - assert(op != NULL /* Precondition */); - assert(oend_w < oend /* No underflow */); - /* Handle edge cases in a slow path: - * - Read beyond end of literals - * - Match end is within WILDCOPY_OVERLIMIT of oend - * - 32-bit mode and the match length overflows - */ - if (UNLIKELY( - iLitEnd > litLimit || - oMatchEnd > oend_w || - (MEM_32bits() && (size_t)(oend - op) < sequenceLength + WILDCOPY_OVERLENGTH))) - return ZSTD_execSequenceEndSplitLitBuffer(op, oend, oend_w, sequence, litPtr, litLimit, prefixStart, virtualStart, dictEnd); - - /* Assumptions (everything else goes into ZSTD_execSequenceEnd()) */ - assert(op <= oLitEnd /* No overflow */); - assert(oLitEnd < oMatchEnd /* Non-zero match & no overflow */); - assert(oMatchEnd <= oend /* No underflow */); - assert(iLitEnd <= litLimit /* Literal length is in bounds */); - assert(oLitEnd <= oend_w /* Can wildcopy literals */); - assert(oMatchEnd <= oend_w /* Can wildcopy matches */); - - /* Copy Literals: - * Split out litLength <= 16 since it is nearly always true. +1.6% on gcc-9. - * We likely don't need the full 32-byte wildcopy. - */ - assert(WILDCOPY_OVERLENGTH >= 16); - ZSTD_copy16(op, (*litPtr)); - if (UNLIKELY(sequence.litLength > 16)) { - ZSTD_wildcopy(op+16, (*litPtr)+16, sequence.litLength-16, ZSTD_no_overlap); - } - op = oLitEnd; - *litPtr = iLitEnd; /* update for next sequence */ - - /* Copy Match */ - if (sequence.offset > (size_t)(oLitEnd - prefixStart)) { - /* offset beyond prefix -> go into extDict */ - RETURN_ERROR_IF(UNLIKELY(sequence.offset > (size_t)(oLitEnd - virtualStart)), corruption_detected, ""); - match = dictEnd + (match - prefixStart); - if (match + sequence.matchLength <= dictEnd) { - ZSTD_memmove(oLitEnd, match, sequence.matchLength); - return sequenceLength; - } - /* span extDict & currentPrefixSegment */ - { size_t const length1 = dictEnd - match; - ZSTD_memmove(oLitEnd, match, length1); - op = oLitEnd + length1; - sequence.matchLength -= length1; - match = prefixStart; - } } - /* Match within prefix of 1 or more bytes */ - assert(op <= oMatchEnd); - assert(oMatchEnd <= oend_w); - assert(match >= prefixStart); - assert(sequence.matchLength >= 1); - - /* Nearly all offsets are >= WILDCOPY_VECLEN bytes, which means we can use wildcopy - * without overlap checking. - */ - if (LIKELY(sequence.offset >= WILDCOPY_VECLEN)) { - /* We bet on a full wildcopy for matches, since we expect matches to be - * longer than literals (in general). In silesia, ~10% of matches are longer - * than 16 bytes. - */ - ZSTD_wildcopy(op, match, (ptrdiff_t)sequence.matchLength, ZSTD_no_overlap); - return sequenceLength; - } - assert(sequence.offset < WILDCOPY_VECLEN); - - /* Copy 8 bytes and spread the offset to be >= 8. */ - ZSTD_overlapCopy8(&op, &match, sequence.offset); - - /* If the match length is > 8 bytes, then continue with the wildcopy. */ - if (sequence.matchLength > 8) { - assert(op < oMatchEnd); - ZSTD_wildcopy(op, match, (ptrdiff_t)sequence.matchLength-8, ZSTD_overlap_src_before_dst); - } - return sequenceLength; -} - - -static void -ZSTD_initFseState(ZSTD_fseState* DStatePtr, BIT_DStream_t* bitD, const ZSTD_seqSymbol* dt) -{ - const void* ptr = dt; - const ZSTD_seqSymbol_header* const DTableH = (const ZSTD_seqSymbol_header*)ptr; - DStatePtr->state = BIT_readBits(bitD, DTableH->tableLog); - DEBUGLOG(6, "ZSTD_initFseState : val=%u using %u bits", - (U32)DStatePtr->state, DTableH->tableLog); - BIT_reloadDStream(bitD); - DStatePtr->table = dt + 1; -} - -FORCE_INLINE_TEMPLATE void -ZSTD_updateFseStateWithDInfo(ZSTD_fseState* DStatePtr, BIT_DStream_t* bitD, U16 nextState, U32 nbBits) -{ - size_t const lowBits = BIT_readBits(bitD, nbBits); - DStatePtr->state = nextState + lowBits; -} - -/* We need to add at most (ZSTD_WINDOWLOG_MAX_32 - 1) bits to read the maximum - * offset bits. But we can only read at most (STREAM_ACCUMULATOR_MIN_32 - 1) - * bits before reloading. This value is the maximum number of bytes we read - * after reloading when we are decoding long offsets. - */ -#define LONG_OFFSETS_MAX_EXTRA_BITS_32 \ - (ZSTD_WINDOWLOG_MAX_32 > STREAM_ACCUMULATOR_MIN_32 \ - ? ZSTD_WINDOWLOG_MAX_32 - STREAM_ACCUMULATOR_MIN_32 \ - : 0) - -typedef enum { ZSTD_lo_isRegularOffset, ZSTD_lo_isLongOffset=1 } ZSTD_longOffset_e; - -FORCE_INLINE_TEMPLATE seq_t -ZSTD_decodeSequence(seqState_t* seqState, const ZSTD_longOffset_e longOffsets) -{ - seq_t seq; - const ZSTD_seqSymbol* const llDInfo = seqState->stateLL.table + seqState->stateLL.state; - const ZSTD_seqSymbol* const mlDInfo = seqState->stateML.table + seqState->stateML.state; - const ZSTD_seqSymbol* const ofDInfo = seqState->stateOffb.table + seqState->stateOffb.state; - seq.matchLength = mlDInfo->baseValue; - seq.litLength = llDInfo->baseValue; - { U32 const ofBase = ofDInfo->baseValue; - BYTE const llBits = llDInfo->nbAdditionalBits; - BYTE const mlBits = mlDInfo->nbAdditionalBits; - BYTE const ofBits = ofDInfo->nbAdditionalBits; - BYTE const totalBits = llBits+mlBits+ofBits; - - U16 const llNext = llDInfo->nextState; - U16 const mlNext = mlDInfo->nextState; - U16 const ofNext = ofDInfo->nextState; - U32 const llnbBits = llDInfo->nbBits; - U32 const mlnbBits = mlDInfo->nbBits; - U32 const ofnbBits = ofDInfo->nbBits; - /* - * As gcc has better branch and block analyzers, sometimes it is only - * valuable to mark likelyness for clang, it gives around 3-4% of - * performance. - */ - - /* sequence */ - { size_t offset; - #if defined(__clang__) - if (LIKELY(ofBits > 1)) { - #else - if (ofBits > 1) { - #endif - ZSTD_STATIC_ASSERT(ZSTD_lo_isLongOffset == 1); - ZSTD_STATIC_ASSERT(LONG_OFFSETS_MAX_EXTRA_BITS_32 == 5); - assert(ofBits <= MaxOff); - if (MEM_32bits() && longOffsets && (ofBits >= STREAM_ACCUMULATOR_MIN_32)) { - U32 const extraBits = ofBits - MIN(ofBits, 32 - seqState->DStream.bitsConsumed); - offset = ofBase + (BIT_readBitsFast(&seqState->DStream, ofBits - extraBits) << extraBits); - BIT_reloadDStream(&seqState->DStream); - if (extraBits) offset += BIT_readBitsFast(&seqState->DStream, extraBits); - assert(extraBits <= LONG_OFFSETS_MAX_EXTRA_BITS_32); /* to avoid another reload */ - } else { - offset = ofBase + BIT_readBitsFast(&seqState->DStream, ofBits/*>0*/); /* <= (ZSTD_WINDOWLOG_MAX-1) bits */ - if (MEM_32bits()) BIT_reloadDStream(&seqState->DStream); - } - seqState->prevOffset[2] = seqState->prevOffset[1]; - seqState->prevOffset[1] = seqState->prevOffset[0]; - seqState->prevOffset[0] = offset; - } else { - U32 const ll0 = (llDInfo->baseValue == 0); - if (LIKELY((ofBits == 0))) { - offset = seqState->prevOffset[ll0]; - seqState->prevOffset[1] = seqState->prevOffset[!ll0]; - seqState->prevOffset[0] = offset; - } else { - offset = ofBase + ll0 + BIT_readBitsFast(&seqState->DStream, 1); - { size_t temp = (offset==3) ? seqState->prevOffset[0] - 1 : seqState->prevOffset[offset]; - temp += !temp; /* 0 is not valid; input is corrupted; force offset to 1 */ - if (offset != 1) seqState->prevOffset[2] = seqState->prevOffset[1]; - seqState->prevOffset[1] = seqState->prevOffset[0]; - seqState->prevOffset[0] = offset = temp; - } } } - seq.offset = offset; - } - - #if defined(__clang__) - if (UNLIKELY(mlBits > 0)) - #else - if (mlBits > 0) - #endif - seq.matchLength += BIT_readBitsFast(&seqState->DStream, mlBits/*>0*/); - - if (MEM_32bits() && (mlBits+llBits >= STREAM_ACCUMULATOR_MIN_32-LONG_OFFSETS_MAX_EXTRA_BITS_32)) - BIT_reloadDStream(&seqState->DStream); - if (MEM_64bits() && UNLIKELY(totalBits >= STREAM_ACCUMULATOR_MIN_64-(LLFSELog+MLFSELog+OffFSELog))) - BIT_reloadDStream(&seqState->DStream); - /* Ensure there are enough bits to read the rest of data in 64-bit mode. */ - ZSTD_STATIC_ASSERT(16+LLFSELog+MLFSELog+OffFSELog < STREAM_ACCUMULATOR_MIN_64); - - #if defined(__clang__) - if (UNLIKELY(llBits > 0)) - #else - if (llBits > 0) - #endif - seq.litLength += BIT_readBitsFast(&seqState->DStream, llBits/*>0*/); - - if (MEM_32bits()) - BIT_reloadDStream(&seqState->DStream); - - DEBUGLOG(6, "seq: litL=%u, matchL=%u, offset=%u", - (U32)seq.litLength, (U32)seq.matchLength, (U32)seq.offset); - - ZSTD_updateFseStateWithDInfo(&seqState->stateLL, &seqState->DStream, llNext, llnbBits); /* <= 9 bits */ - ZSTD_updateFseStateWithDInfo(&seqState->stateML, &seqState->DStream, mlNext, mlnbBits); /* <= 9 bits */ - if (MEM_32bits()) BIT_reloadDStream(&seqState->DStream); /* <= 18 bits */ - ZSTD_updateFseStateWithDInfo(&seqState->stateOffb, &seqState->DStream, ofNext, ofnbBits); /* <= 8 bits */ - } - - return seq; -} - -#ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION -MEM_STATIC int ZSTD_dictionaryIsActive(ZSTD_DCtx const* dctx, BYTE const* prefixStart, BYTE const* oLitEnd) -{ - size_t const windowSize = dctx->fParams.windowSize; - /* No dictionary used. */ - if (dctx->dictContentEndForFuzzing == NULL) return 0; - /* Dictionary is our prefix. */ - if (prefixStart == dctx->dictContentBeginForFuzzing) return 1; - /* Dictionary is not our ext-dict. */ - if (dctx->dictEnd != dctx->dictContentEndForFuzzing) return 0; - /* Dictionary is not within our window size. */ - if ((size_t)(oLitEnd - prefixStart) >= windowSize) return 0; - /* Dictionary is active. */ - return 1; -} - -MEM_STATIC void ZSTD_assertValidSequence( - ZSTD_DCtx const* dctx, - BYTE const* op, BYTE const* oend, - seq_t const seq, - BYTE const* prefixStart, BYTE const* virtualStart) -{ -#if DEBUGLEVEL >= 1 - size_t const windowSize = dctx->fParams.windowSize; - size_t const sequenceSize = seq.litLength + seq.matchLength; - BYTE const* const oLitEnd = op + seq.litLength; - DEBUGLOG(6, "Checking sequence: litL=%u matchL=%u offset=%u", - (U32)seq.litLength, (U32)seq.matchLength, (U32)seq.offset); - assert(op <= oend); - assert((size_t)(oend - op) >= sequenceSize); - assert(sequenceSize <= ZSTD_BLOCKSIZE_MAX); - if (ZSTD_dictionaryIsActive(dctx, prefixStart, oLitEnd)) { - size_t const dictSize = (size_t)((char const*)dctx->dictContentEndForFuzzing - (char const*)dctx->dictContentBeginForFuzzing); - /* Offset must be within the dictionary. */ - assert(seq.offset <= (size_t)(oLitEnd - virtualStart)); - assert(seq.offset <= windowSize + dictSize); - } else { - /* Offset must be within our window. */ - assert(seq.offset <= windowSize); - } -#else - (void)dctx, (void)op, (void)oend, (void)seq, (void)prefixStart, (void)virtualStart; -#endif -} -#endif - -#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG - - -FORCE_INLINE_TEMPLATE size_t -DONT_VECTORIZE -ZSTD_decompressSequences_bodySplitLitBuffer( ZSTD_DCtx* dctx, - void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame) -{ - const BYTE* ip = (const BYTE*)seqStart; - const BYTE* const iend = ip + seqSize; - BYTE* const ostart = (BYTE*)dst; - BYTE* const oend = ostart + maxDstSize; - BYTE* op = ostart; - const BYTE* litPtr = dctx->litPtr; - const BYTE* litBufferEnd = dctx->litBufferEnd; - const BYTE* const prefixStart = (const BYTE*) (dctx->prefixStart); - const BYTE* const vBase = (const BYTE*) (dctx->virtualStart); - const BYTE* const dictEnd = (const BYTE*) (dctx->dictEnd); - DEBUGLOG(5, "ZSTD_decompressSequences_bodySplitLitBuffer"); - (void)frame; - - /* Regen sequences */ - if (nbSeq) { - seqState_t seqState; - dctx->fseEntropy = 1; - { U32 i; for (i=0; ientropy.rep[i]; } - RETURN_ERROR_IF( - ERR_isError(BIT_initDStream(&seqState.DStream, ip, iend-ip)), - corruption_detected, ""); - ZSTD_initFseState(&seqState.stateLL, &seqState.DStream, dctx->LLTptr); - ZSTD_initFseState(&seqState.stateOffb, &seqState.DStream, dctx->OFTptr); - ZSTD_initFseState(&seqState.stateML, &seqState.DStream, dctx->MLTptr); - assert(dst != NULL); - - ZSTD_STATIC_ASSERT( - BIT_DStream_unfinished < BIT_DStream_completed && - BIT_DStream_endOfBuffer < BIT_DStream_completed && - BIT_DStream_completed < BIT_DStream_overflow); - - /* decompress without overrunning litPtr begins */ - { - seq_t sequence = ZSTD_decodeSequence(&seqState, isLongOffset); - /* Align the decompression loop to 32 + 16 bytes. - * - * zstd compiled with gcc-9 on an Intel i9-9900k shows 10% decompression - * speed swings based on the alignment of the decompression loop. This - * performance swing is caused by parts of the decompression loop falling - * out of the DSB. The entire decompression loop should fit in the DSB, - * when it can't we get much worse performance. You can measure if you've - * hit the good case or the bad case with this perf command for some - * compressed file test.zst: - * - * perf stat -e cycles -e instructions -e idq.all_dsb_cycles_any_uops \ - * -e idq.all_mite_cycles_any_uops -- ./zstd -tq test.zst - * - * If you see most cycles served out of the MITE you've hit the bad case. - * If you see most cycles served out of the DSB you've hit the good case. - * If it is pretty even then you may be in an okay case. - * - * This issue has been reproduced on the following CPUs: - * - Kabylake: Macbook Pro (15-inch, 2019) 2.4 GHz Intel Core i9 - * Use Instruments->Counters to get DSB/MITE cycles. - * I never got performance swings, but I was able to - * go from the good case of mostly DSB to half of the - * cycles served from MITE. - * - Coffeelake: Intel i9-9900k - * - Coffeelake: Intel i7-9700k - * - * I haven't been able to reproduce the instability or DSB misses on any - * of the following CPUS: - * - Haswell - * - Broadwell: Intel(R) Xeon(R) CPU E5-2680 v4 @ 2.40GH - * - Skylake - * - * Alignment is done for each of the three major decompression loops: - * - ZSTD_decompressSequences_bodySplitLitBuffer - presplit section of the literal buffer - * - ZSTD_decompressSequences_bodySplitLitBuffer - postsplit section of the literal buffer - * - ZSTD_decompressSequences_body - * Alignment choices are made to minimize large swings on bad cases and influence on performance - * from changes external to this code, rather than to overoptimize on the current commit. - * - * If you are seeing performance stability this script can help test. - * It tests on 4 commits in zstd where I saw performance change. - * - * https://gist.github.com/terrelln/9889fc06a423fd5ca6e99351564473f4 - */ -#if defined(__GNUC__) && defined(__x86_64__) - __asm__(".p2align 6"); -# if __GNUC__ >= 7 - /* good for gcc-7, gcc-9, and gcc-11 */ - __asm__("nop"); - __asm__(".p2align 5"); - __asm__("nop"); - __asm__(".p2align 4"); -# if __GNUC__ == 8 || __GNUC__ == 10 - /* good for gcc-8 and gcc-10 */ - __asm__("nop"); - __asm__(".p2align 3"); -# endif -# endif -#endif - - /* Handle the initial state where litBuffer is currently split between dst and litExtraBuffer */ - for (; litPtr + sequence.litLength <= dctx->litBufferEnd; ) { - size_t const oneSeqSize = ZSTD_execSequenceSplitLitBuffer(op, oend, litPtr + sequence.litLength - WILDCOPY_OVERLENGTH, sequence, &litPtr, litBufferEnd, prefixStart, vBase, dictEnd); -#if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) && defined(FUZZING_ASSERT_VALID_SEQUENCE) - assert(!ZSTD_isError(oneSeqSize)); - if (frame) ZSTD_assertValidSequence(dctx, op, oend, sequence, prefixStart, vBase); -#endif - if (UNLIKELY(ZSTD_isError(oneSeqSize))) - return oneSeqSize; - DEBUGLOG(6, "regenerated sequence size : %u", (U32)oneSeqSize); - op += oneSeqSize; - if (UNLIKELY(!--nbSeq)) - break; - BIT_reloadDStream(&(seqState.DStream)); - sequence = ZSTD_decodeSequence(&seqState, isLongOffset); - } - - /* If there are more sequences, they will need to read literals from litExtraBuffer; copy over the remainder from dst and update litPtr and litEnd */ - if (nbSeq > 0) { - const size_t leftoverLit = dctx->litBufferEnd - litPtr; - if (leftoverLit) - { - RETURN_ERROR_IF(leftoverLit > (size_t)(oend - op), dstSize_tooSmall, "remaining lit must fit within dstBuffer"); - ZSTD_safecopyDstBeforeSrc(op, litPtr, leftoverLit); - sequence.litLength -= leftoverLit; - op += leftoverLit; - } - litPtr = dctx->litExtraBuffer; - litBufferEnd = dctx->litExtraBuffer + ZSTD_LITBUFFEREXTRASIZE; - dctx->litBufferLocation = ZSTD_not_in_dst; - { - size_t const oneSeqSize = ZSTD_execSequence(op, oend, sequence, &litPtr, litBufferEnd, prefixStart, vBase, dictEnd); -#if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) && defined(FUZZING_ASSERT_VALID_SEQUENCE) - assert(!ZSTD_isError(oneSeqSize)); - if (frame) ZSTD_assertValidSequence(dctx, op, oend, sequence, prefixStart, vBase); -#endif - if (UNLIKELY(ZSTD_isError(oneSeqSize))) - return oneSeqSize; - DEBUGLOG(6, "regenerated sequence size : %u", (U32)oneSeqSize); - op += oneSeqSize; - if (--nbSeq) - BIT_reloadDStream(&(seqState.DStream)); - } - } - } - - if (nbSeq > 0) /* there is remaining lit from extra buffer */ - { - -#if defined(__GNUC__) && defined(__x86_64__) - __asm__(".p2align 6"); - __asm__("nop"); -# if __GNUC__ != 7 - /* worse for gcc-7 better for gcc-8, gcc-9, and gcc-10 and clang */ - __asm__(".p2align 4"); - __asm__("nop"); - __asm__(".p2align 3"); -# elif __GNUC__ >= 11 - __asm__(".p2align 3"); -# else - __asm__(".p2align 5"); - __asm__("nop"); - __asm__(".p2align 3"); -# endif -#endif - - for (; ; ) { - seq_t const sequence = ZSTD_decodeSequence(&seqState, isLongOffset); - size_t const oneSeqSize = ZSTD_execSequence(op, oend, sequence, &litPtr, litBufferEnd, prefixStart, vBase, dictEnd); -#if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) && defined(FUZZING_ASSERT_VALID_SEQUENCE) - assert(!ZSTD_isError(oneSeqSize)); - if (frame) ZSTD_assertValidSequence(dctx, op, oend, sequence, prefixStart, vBase); -#endif - if (UNLIKELY(ZSTD_isError(oneSeqSize))) - return oneSeqSize; - DEBUGLOG(6, "regenerated sequence size : %u", (U32)oneSeqSize); - op += oneSeqSize; - if (UNLIKELY(!--nbSeq)) - break; - BIT_reloadDStream(&(seqState.DStream)); - } - } - - /* check if reached exact end */ - DEBUGLOG(5, "ZSTD_decompressSequences_bodySplitLitBuffer: after decode loop, remaining nbSeq : %i", nbSeq); - RETURN_ERROR_IF(nbSeq, corruption_detected, ""); - RETURN_ERROR_IF(BIT_reloadDStream(&seqState.DStream) < BIT_DStream_completed, corruption_detected, ""); - /* save reps for next block */ - { U32 i; for (i=0; ientropy.rep[i] = (U32)(seqState.prevOffset[i]); } - } - - /* last literal segment */ - if (dctx->litBufferLocation == ZSTD_split) /* split hasn't been reached yet, first get dst then copy litExtraBuffer */ - { - size_t const lastLLSize = litBufferEnd - litPtr; - RETURN_ERROR_IF(lastLLSize > (size_t)(oend - op), dstSize_tooSmall, ""); - if (op != NULL) { - ZSTD_memmove(op, litPtr, lastLLSize); - op += lastLLSize; - } - litPtr = dctx->litExtraBuffer; - litBufferEnd = dctx->litExtraBuffer + ZSTD_LITBUFFEREXTRASIZE; - dctx->litBufferLocation = ZSTD_not_in_dst; - } - { size_t const lastLLSize = litBufferEnd - litPtr; - RETURN_ERROR_IF(lastLLSize > (size_t)(oend-op), dstSize_tooSmall, ""); - if (op != NULL) { - ZSTD_memcpy(op, litPtr, lastLLSize); - op += lastLLSize; - } - } - - return op-ostart; -} - -FORCE_INLINE_TEMPLATE size_t -DONT_VECTORIZE -ZSTD_decompressSequences_body(ZSTD_DCtx* dctx, - void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame) -{ - const BYTE* ip = (const BYTE*)seqStart; - const BYTE* const iend = ip + seqSize; - BYTE* const ostart = (BYTE*)dst; - BYTE* const oend = dctx->litBufferLocation == ZSTD_not_in_dst ? ostart + maxDstSize : dctx->litBuffer; - BYTE* op = ostart; - const BYTE* litPtr = dctx->litPtr; - const BYTE* const litEnd = litPtr + dctx->litSize; - const BYTE* const prefixStart = (const BYTE*)(dctx->prefixStart); - const BYTE* const vBase = (const BYTE*)(dctx->virtualStart); - const BYTE* const dictEnd = (const BYTE*)(dctx->dictEnd); - DEBUGLOG(5, "ZSTD_decompressSequences_body"); - (void)frame; - - /* Regen sequences */ - if (nbSeq) { - seqState_t seqState; - dctx->fseEntropy = 1; - { U32 i; for (i = 0; i < ZSTD_REP_NUM; i++) seqState.prevOffset[i] = dctx->entropy.rep[i]; } - RETURN_ERROR_IF( - ERR_isError(BIT_initDStream(&seqState.DStream, ip, iend - ip)), - corruption_detected, ""); - ZSTD_initFseState(&seqState.stateLL, &seqState.DStream, dctx->LLTptr); - ZSTD_initFseState(&seqState.stateOffb, &seqState.DStream, dctx->OFTptr); - ZSTD_initFseState(&seqState.stateML, &seqState.DStream, dctx->MLTptr); - assert(dst != NULL); - - ZSTD_STATIC_ASSERT( - BIT_DStream_unfinished < BIT_DStream_completed && - BIT_DStream_endOfBuffer < BIT_DStream_completed && - BIT_DStream_completed < BIT_DStream_overflow); - -#if defined(__GNUC__) && defined(__x86_64__) - __asm__(".p2align 6"); - __asm__("nop"); -# if __GNUC__ >= 7 - __asm__(".p2align 5"); - __asm__("nop"); - __asm__(".p2align 3"); -# else - __asm__(".p2align 4"); - __asm__("nop"); - __asm__(".p2align 3"); -# endif -#endif - - for ( ; ; ) { - seq_t const sequence = ZSTD_decodeSequence(&seqState, isLongOffset); - size_t const oneSeqSize = ZSTD_execSequence(op, oend, sequence, &litPtr, litEnd, prefixStart, vBase, dictEnd); -#if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) && defined(FUZZING_ASSERT_VALID_SEQUENCE) - assert(!ZSTD_isError(oneSeqSize)); - if (frame) ZSTD_assertValidSequence(dctx, op, oend, sequence, prefixStart, vBase); -#endif - if (UNLIKELY(ZSTD_isError(oneSeqSize))) - return oneSeqSize; - DEBUGLOG(6, "regenerated sequence size : %u", (U32)oneSeqSize); - op += oneSeqSize; - if (UNLIKELY(!--nbSeq)) - break; - BIT_reloadDStream(&(seqState.DStream)); - } - - /* check if reached exact end */ - DEBUGLOG(5, "ZSTD_decompressSequences_body: after decode loop, remaining nbSeq : %i", nbSeq); - RETURN_ERROR_IF(nbSeq, corruption_detected, ""); - RETURN_ERROR_IF(BIT_reloadDStream(&seqState.DStream) < BIT_DStream_completed, corruption_detected, ""); - /* save reps for next block */ - { U32 i; for (i=0; ientropy.rep[i] = (U32)(seqState.prevOffset[i]); } - } - - /* last literal segment */ - { size_t const lastLLSize = litEnd - litPtr; - RETURN_ERROR_IF(lastLLSize > (size_t)(oend-op), dstSize_tooSmall, ""); - if (op != NULL) { - ZSTD_memcpy(op, litPtr, lastLLSize); - op += lastLLSize; - } - } - - return op-ostart; -} - -static size_t -ZSTD_decompressSequences_default(ZSTD_DCtx* dctx, - void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame) -{ - return ZSTD_decompressSequences_body(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset, frame); -} - -static size_t -ZSTD_decompressSequencesSplitLitBuffer_default(ZSTD_DCtx* dctx, - void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame) -{ - return ZSTD_decompressSequences_bodySplitLitBuffer(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset, frame); -} -#endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG */ - -#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT - -FORCE_INLINE_TEMPLATE size_t -ZSTD_prefetchMatch(size_t prefetchPos, seq_t const sequence, - const BYTE* const prefixStart, const BYTE* const dictEnd) -{ - prefetchPos += sequence.litLength; - { const BYTE* const matchBase = (sequence.offset > prefetchPos) ? dictEnd : prefixStart; - const BYTE* const match = matchBase + prefetchPos - sequence.offset; /* note : this operation can overflow when seq.offset is really too large, which can only happen when input is corrupted. - * No consequence though : memory address is only used for prefetching, not for dereferencing */ - PREFETCH_L1(match); PREFETCH_L1(match+CACHELINE_SIZE); /* note : it's safe to invoke PREFETCH() on any memory address, including invalid ones */ - } - return prefetchPos + sequence.matchLength; -} - -/* This decoding function employs prefetching - * to reduce latency impact of cache misses. - * It's generally employed when block contains a significant portion of long-distance matches - * or when coupled with a "cold" dictionary */ -FORCE_INLINE_TEMPLATE size_t -ZSTD_decompressSequencesLong_body( - ZSTD_DCtx* dctx, - void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame) -{ - const BYTE* ip = (const BYTE*)seqStart; - const BYTE* const iend = ip + seqSize; - BYTE* const ostart = (BYTE*)dst; - BYTE* const oend = dctx->litBufferLocation == ZSTD_in_dst ? dctx->litBuffer : ostart + maxDstSize; - BYTE* op = ostart; - const BYTE* litPtr = dctx->litPtr; - const BYTE* litBufferEnd = dctx->litBufferEnd; - const BYTE* const prefixStart = (const BYTE*) (dctx->prefixStart); - const BYTE* const dictStart = (const BYTE*) (dctx->virtualStart); - const BYTE* const dictEnd = (const BYTE*) (dctx->dictEnd); - (void)frame; - - /* Regen sequences */ - if (nbSeq) { -#define STORED_SEQS 8 -#define STORED_SEQS_MASK (STORED_SEQS-1) -#define ADVANCED_SEQS STORED_SEQS - seq_t sequences[STORED_SEQS]; - int const seqAdvance = MIN(nbSeq, ADVANCED_SEQS); - seqState_t seqState; - int seqNb; - size_t prefetchPos = (size_t)(op-prefixStart); /* track position relative to prefixStart */ - - dctx->fseEntropy = 1; - { int i; for (i=0; ientropy.rep[i]; } - assert(dst != NULL); - assert(iend >= ip); - RETURN_ERROR_IF( - ERR_isError(BIT_initDStream(&seqState.DStream, ip, iend-ip)), - corruption_detected, ""); - ZSTD_initFseState(&seqState.stateLL, &seqState.DStream, dctx->LLTptr); - ZSTD_initFseState(&seqState.stateOffb, &seqState.DStream, dctx->OFTptr); - ZSTD_initFseState(&seqState.stateML, &seqState.DStream, dctx->MLTptr); - - /* prepare in advance */ - for (seqNb=0; (BIT_reloadDStream(&seqState.DStream) <= BIT_DStream_completed) && (seqNblitBufferLocation == ZSTD_split && litPtr + sequences[(seqNb - ADVANCED_SEQS) & STORED_SEQS_MASK].litLength > dctx->litBufferEnd) - { - /* lit buffer is reaching split point, empty out the first buffer and transition to litExtraBuffer */ - const size_t leftoverLit = dctx->litBufferEnd - litPtr; - if (leftoverLit) - { - RETURN_ERROR_IF(leftoverLit > (size_t)(oend - op), dstSize_tooSmall, "remaining lit must fit within dstBuffer"); - ZSTD_safecopyDstBeforeSrc(op, litPtr, leftoverLit); - sequences[(seqNb - ADVANCED_SEQS) & STORED_SEQS_MASK].litLength -= leftoverLit; - op += leftoverLit; - } - litPtr = dctx->litExtraBuffer; - litBufferEnd = dctx->litExtraBuffer + ZSTD_LITBUFFEREXTRASIZE; - dctx->litBufferLocation = ZSTD_not_in_dst; - oneSeqSize = ZSTD_execSequence(op, oend, sequences[(seqNb - ADVANCED_SEQS) & STORED_SEQS_MASK], &litPtr, litBufferEnd, prefixStart, dictStart, dictEnd); -#if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) && defined(FUZZING_ASSERT_VALID_SEQUENCE) - assert(!ZSTD_isError(oneSeqSize)); - if (frame) ZSTD_assertValidSequence(dctx, op, oend, sequences[(seqNb - ADVANCED_SEQS) & STORED_SEQS_MASK], prefixStart, dictStart); -#endif - if (ZSTD_isError(oneSeqSize)) return oneSeqSize; - - prefetchPos = ZSTD_prefetchMatch(prefetchPos, sequence, prefixStart, dictEnd); - sequences[seqNb & STORED_SEQS_MASK] = sequence; - op += oneSeqSize; - } - else - { - /* lit buffer is either wholly contained in first or second split, or not split at all*/ - oneSeqSize = dctx->litBufferLocation == ZSTD_split ? - ZSTD_execSequenceSplitLitBuffer(op, oend, litPtr + sequences[(seqNb - ADVANCED_SEQS) & STORED_SEQS_MASK].litLength - WILDCOPY_OVERLENGTH, sequences[(seqNb - ADVANCED_SEQS) & STORED_SEQS_MASK], &litPtr, litBufferEnd, prefixStart, dictStart, dictEnd) : - ZSTD_execSequence(op, oend, sequences[(seqNb - ADVANCED_SEQS) & STORED_SEQS_MASK], &litPtr, litBufferEnd, prefixStart, dictStart, dictEnd); -#if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) && defined(FUZZING_ASSERT_VALID_SEQUENCE) - assert(!ZSTD_isError(oneSeqSize)); - if (frame) ZSTD_assertValidSequence(dctx, op, oend, sequences[(seqNb - ADVANCED_SEQS) & STORED_SEQS_MASK], prefixStart, dictStart); -#endif - if (ZSTD_isError(oneSeqSize)) return oneSeqSize; - - prefetchPos = ZSTD_prefetchMatch(prefetchPos, sequence, prefixStart, dictEnd); - sequences[seqNb & STORED_SEQS_MASK] = sequence; - op += oneSeqSize; - } - } - RETURN_ERROR_IF(seqNblitBufferLocation == ZSTD_split && litPtr + sequence->litLength > dctx->litBufferEnd) - { - const size_t leftoverLit = dctx->litBufferEnd - litPtr; - if (leftoverLit) - { - RETURN_ERROR_IF(leftoverLit > (size_t)(oend - op), dstSize_tooSmall, "remaining lit must fit within dstBuffer"); - ZSTD_safecopyDstBeforeSrc(op, litPtr, leftoverLit); - sequence->litLength -= leftoverLit; - op += leftoverLit; - } - litPtr = dctx->litExtraBuffer; - litBufferEnd = dctx->litExtraBuffer + ZSTD_LITBUFFEREXTRASIZE; - dctx->litBufferLocation = ZSTD_not_in_dst; - { - size_t const oneSeqSize = ZSTD_execSequence(op, oend, *sequence, &litPtr, litBufferEnd, prefixStart, dictStart, dictEnd); -#if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) && defined(FUZZING_ASSERT_VALID_SEQUENCE) - assert(!ZSTD_isError(oneSeqSize)); - if (frame) ZSTD_assertValidSequence(dctx, op, oend, sequences[seqNb&STORED_SEQS_MASK], prefixStart, dictStart); -#endif - if (ZSTD_isError(oneSeqSize)) return oneSeqSize; - op += oneSeqSize; - } - } - else - { - size_t const oneSeqSize = dctx->litBufferLocation == ZSTD_split ? - ZSTD_execSequenceSplitLitBuffer(op, oend, litPtr + sequence->litLength - WILDCOPY_OVERLENGTH, *sequence, &litPtr, litBufferEnd, prefixStart, dictStart, dictEnd) : - ZSTD_execSequence(op, oend, *sequence, &litPtr, litBufferEnd, prefixStart, dictStart, dictEnd); -#if defined(FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION) && defined(FUZZING_ASSERT_VALID_SEQUENCE) - assert(!ZSTD_isError(oneSeqSize)); - if (frame) ZSTD_assertValidSequence(dctx, op, oend, sequences[seqNb&STORED_SEQS_MASK], prefixStart, dictStart); -#endif - if (ZSTD_isError(oneSeqSize)) return oneSeqSize; - op += oneSeqSize; - } - } - - /* save reps for next block */ - { U32 i; for (i=0; ientropy.rep[i] = (U32)(seqState.prevOffset[i]); } - } - - /* last literal segment */ - if (dctx->litBufferLocation == ZSTD_split) /* first deplete literal buffer in dst, then copy litExtraBuffer */ - { - size_t const lastLLSize = litBufferEnd - litPtr; - RETURN_ERROR_IF(lastLLSize > (size_t)(oend - op), dstSize_tooSmall, ""); - if (op != NULL) { - ZSTD_memmove(op, litPtr, lastLLSize); - op += lastLLSize; - } - litPtr = dctx->litExtraBuffer; - litBufferEnd = dctx->litExtraBuffer + ZSTD_LITBUFFEREXTRASIZE; - } - { size_t const lastLLSize = litBufferEnd - litPtr; - RETURN_ERROR_IF(lastLLSize > (size_t)(oend-op), dstSize_tooSmall, ""); - if (op != NULL) { - ZSTD_memmove(op, litPtr, lastLLSize); - op += lastLLSize; - } - } - - return op-ostart; -} - -static size_t -ZSTD_decompressSequencesLong_default(ZSTD_DCtx* dctx, - void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame) -{ - return ZSTD_decompressSequencesLong_body(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset, frame); -} -#endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT */ - - - -#if DYNAMIC_BMI2 - -#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG -static BMI2_TARGET_ATTRIBUTE size_t -DONT_VECTORIZE -ZSTD_decompressSequences_bmi2(ZSTD_DCtx* dctx, - void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame) -{ - return ZSTD_decompressSequences_body(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset, frame); -} -static BMI2_TARGET_ATTRIBUTE size_t -DONT_VECTORIZE -ZSTD_decompressSequencesSplitLitBuffer_bmi2(ZSTD_DCtx* dctx, - void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame) -{ - return ZSTD_decompressSequences_bodySplitLitBuffer(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset, frame); -} -#endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG */ - -#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT -static BMI2_TARGET_ATTRIBUTE size_t -ZSTD_decompressSequencesLong_bmi2(ZSTD_DCtx* dctx, - void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame) -{ - return ZSTD_decompressSequencesLong_body(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset, frame); -} -#endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT */ - -#endif /* DYNAMIC_BMI2 */ - -typedef size_t (*ZSTD_decompressSequences_t)( - ZSTD_DCtx* dctx, - void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame); - -#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG -static size_t -ZSTD_decompressSequences(ZSTD_DCtx* dctx, void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame) -{ - DEBUGLOG(5, "ZSTD_decompressSequences"); -#if DYNAMIC_BMI2 - if (ZSTD_DCtx_get_bmi2(dctx)) { - return ZSTD_decompressSequences_bmi2(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset, frame); - } -#endif - return ZSTD_decompressSequences_default(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset, frame); -} -static size_t -ZSTD_decompressSequencesSplitLitBuffer(ZSTD_DCtx* dctx, void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame) -{ - DEBUGLOG(5, "ZSTD_decompressSequencesSplitLitBuffer"); -#if DYNAMIC_BMI2 - if (ZSTD_DCtx_get_bmi2(dctx)) { - return ZSTD_decompressSequencesSplitLitBuffer_bmi2(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset, frame); - } -#endif - return ZSTD_decompressSequencesSplitLitBuffer_default(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset, frame); -} -#endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG */ - - -#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT -/* ZSTD_decompressSequencesLong() : - * decompression function triggered when a minimum share of offsets is considered "long", - * aka out of cache. - * note : "long" definition seems overloaded here, sometimes meaning "wider than bitstream register", and sometimes meaning "farther than memory cache distance". - * This function will try to mitigate main memory latency through the use of prefetching */ -static size_t -ZSTD_decompressSequencesLong(ZSTD_DCtx* dctx, - void* dst, size_t maxDstSize, - const void* seqStart, size_t seqSize, int nbSeq, - const ZSTD_longOffset_e isLongOffset, - const int frame) -{ - DEBUGLOG(5, "ZSTD_decompressSequencesLong"); -#if DYNAMIC_BMI2 - if (ZSTD_DCtx_get_bmi2(dctx)) { - return ZSTD_decompressSequencesLong_bmi2(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset, frame); - } -#endif - return ZSTD_decompressSequencesLong_default(dctx, dst, maxDstSize, seqStart, seqSize, nbSeq, isLongOffset, frame); -} -#endif /* ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT */ - - - -#if !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT) && \ - !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG) -/* ZSTD_getLongOffsetsShare() : - * condition : offTable must be valid - * @return : "share" of long offsets (arbitrarily defined as > (1<<23)) - * compared to maximum possible of (1< 22) total += 1; - } - - assert(tableLog <= OffFSELog); - total <<= (OffFSELog - tableLog); /* scale to OffFSELog */ - - return total; -} -#endif - -size_t -ZSTD_decompressBlock_internal(ZSTD_DCtx* dctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, const int frame, const streaming_operation streaming) -{ /* blockType == blockCompressed */ - const BYTE* ip = (const BYTE*)src; - /* isLongOffset must be true if there are long offsets. - * Offsets are long if they are larger than 2^STREAM_ACCUMULATOR_MIN. - * We don't expect that to be the case in 64-bit mode. - * In block mode, window size is not known, so we have to be conservative. - * (note: but it could be evaluated from current-lowLimit) - */ - ZSTD_longOffset_e const isLongOffset = (ZSTD_longOffset_e)(MEM_32bits() && (!frame || (dctx->fParams.windowSize > (1ULL << STREAM_ACCUMULATOR_MIN)))); - DEBUGLOG(5, "ZSTD_decompressBlock_internal (size : %u)", (U32)srcSize); - - RETURN_ERROR_IF(srcSize >= ZSTD_BLOCKSIZE_MAX, srcSize_wrong, ""); - - /* Decode literals section */ - { size_t const litCSize = ZSTD_decodeLiteralsBlock(dctx, src, srcSize, dst, dstCapacity, streaming); - DEBUGLOG(5, "ZSTD_decodeLiteralsBlock : %u", (U32)litCSize); - if (ZSTD_isError(litCSize)) return litCSize; - ip += litCSize; - srcSize -= litCSize; - } - - /* Build Decoding Tables */ - { - /* These macros control at build-time which decompressor implementation - * we use. If neither is defined, we do some inspection and dispatch at - * runtime. - */ -#if !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT) && \ - !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG) - int usePrefetchDecoder = dctx->ddictIsCold; -#endif - int nbSeq; - size_t const seqHSize = ZSTD_decodeSeqHeaders(dctx, &nbSeq, ip, srcSize); - if (ZSTD_isError(seqHSize)) return seqHSize; - ip += seqHSize; - srcSize -= seqHSize; - - RETURN_ERROR_IF(dst == NULL && nbSeq > 0, dstSize_tooSmall, "NULL not handled"); - -#if !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT) && \ - !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG) - if ( !usePrefetchDecoder - && (!frame || (dctx->fParams.windowSize > (1<<24))) - && (nbSeq>ADVANCED_SEQS) ) { /* could probably use a larger nbSeq limit */ - U32 const shareLongOffsets = ZSTD_getLongOffsetsShare(dctx->OFTptr); - U32 const minShare = MEM_64bits() ? 7 : 20; /* heuristic values, correspond to 2.73% and 7.81% */ - usePrefetchDecoder = (shareLongOffsets >= minShare); - } -#endif - - dctx->ddictIsCold = 0; - -#if !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT) && \ - !defined(ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG) - if (usePrefetchDecoder) -#endif -#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_SHORT - return ZSTD_decompressSequencesLong(dctx, dst, dstCapacity, ip, srcSize, nbSeq, isLongOffset, frame); -#endif - -#ifndef ZSTD_FORCE_DECOMPRESS_SEQUENCES_LONG - /* else */ - if (dctx->litBufferLocation == ZSTD_split) - return ZSTD_decompressSequencesSplitLitBuffer(dctx, dst, dstCapacity, ip, srcSize, nbSeq, isLongOffset, frame); - else - return ZSTD_decompressSequences(dctx, dst, dstCapacity, ip, srcSize, nbSeq, isLongOffset, frame); -#endif - } -} - - -void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst, size_t dstSize) -{ - if (dst != dctx->previousDstEnd && dstSize > 0) { /* not contiguous */ - dctx->dictEnd = dctx->previousDstEnd; - dctx->virtualStart = (const char*)dst - ((const char*)(dctx->previousDstEnd) - (const char*)(dctx->prefixStart)); - dctx->prefixStart = dst; - dctx->previousDstEnd = dst; - } -} - - -size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize) -{ - size_t dSize; - ZSTD_checkContinuity(dctx, dst, dstCapacity); - dSize = ZSTD_decompressBlock_internal(dctx, dst, dstCapacity, src, srcSize, /* frame */ 0, not_streaming); - dctx->previousDstEnd = (char*)dst + dSize; - return dSize; -} diff --git a/dep/zstd/lib/decompress/zstd_decompress_block.h b/dep/zstd/lib/decompress/zstd_decompress_block.h deleted file mode 100644 index c61a9d0c4..000000000 --- a/dep/zstd/lib/decompress/zstd_decompress_block.h +++ /dev/null @@ -1,68 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -#ifndef ZSTD_DEC_BLOCK_H -#define ZSTD_DEC_BLOCK_H - -/*-******************************************************* - * Dependencies - *********************************************************/ -#include "../common/zstd_deps.h" /* size_t */ -#include "../zstd.h" /* DCtx, and some public functions */ -#include "../common/zstd_internal.h" /* blockProperties_t, and some public functions */ -#include "zstd_decompress_internal.h" /* ZSTD_seqSymbol */ - - -/* === Prototypes === */ - -/* note: prototypes already published within `zstd.h` : - * ZSTD_decompressBlock() - */ - -/* note: prototypes already published within `zstd_internal.h` : - * ZSTD_getcBlockSize() - * ZSTD_decodeSeqHeaders() - */ - - - /* Streaming state is used to inform allocation of the literal buffer */ -typedef enum { - not_streaming = 0, - is_streaming = 1 -} streaming_operation; - -/* ZSTD_decompressBlock_internal() : - * decompress block, starting at `src`, - * into destination buffer `dst`. - * @return : decompressed block size, - * or an error code (which can be tested using ZSTD_isError()) - */ -size_t ZSTD_decompressBlock_internal(ZSTD_DCtx* dctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, const int frame, const streaming_operation streaming); - -/* ZSTD_buildFSETable() : - * generate FSE decoding table for one symbol (ll, ml or off) - * this function must be called with valid parameters only - * (dt is large enough, normalizedCounter distribution total is a power of 2, max is within range, etc.) - * in which case it cannot fail. - * The workspace must be 4-byte aligned and at least ZSTD_BUILD_FSE_TABLE_WKSP_SIZE bytes, which is - * defined in zstd_decompress_internal.h. - * Internal use only. - */ -void ZSTD_buildFSETable(ZSTD_seqSymbol* dt, - const short* normalizedCounter, unsigned maxSymbolValue, - const U32* baseValue, const U8* nbAdditionalBits, - unsigned tableLog, void* wksp, size_t wkspSize, - int bmi2); - - -#endif /* ZSTD_DEC_BLOCK_H */ diff --git a/dep/zstd/lib/decompress/zstd_decompress_internal.h b/dep/zstd/lib/decompress/zstd_decompress_internal.h deleted file mode 100644 index 2b5a53850..000000000 --- a/dep/zstd/lib/decompress/zstd_decompress_internal.h +++ /dev/null @@ -1,236 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - - -/* zstd_decompress_internal: - * objects and definitions shared within lib/decompress modules */ - - #ifndef ZSTD_DECOMPRESS_INTERNAL_H - #define ZSTD_DECOMPRESS_INTERNAL_H - - -/*-******************************************************* - * Dependencies - *********************************************************/ -#include "../common/mem.h" /* BYTE, U16, U32 */ -#include "../common/zstd_internal.h" /* constants : MaxLL, MaxML, MaxOff, LLFSELog, etc. */ - - - -/*-******************************************************* - * Constants - *********************************************************/ -static UNUSED_ATTR const U32 LL_base[MaxLL+1] = { - 0, 1, 2, 3, 4, 5, 6, 7, - 8, 9, 10, 11, 12, 13, 14, 15, - 16, 18, 20, 22, 24, 28, 32, 40, - 48, 64, 0x80, 0x100, 0x200, 0x400, 0x800, 0x1000, - 0x2000, 0x4000, 0x8000, 0x10000 }; - -static UNUSED_ATTR const U32 OF_base[MaxOff+1] = { - 0, 1, 1, 5, 0xD, 0x1D, 0x3D, 0x7D, - 0xFD, 0x1FD, 0x3FD, 0x7FD, 0xFFD, 0x1FFD, 0x3FFD, 0x7FFD, - 0xFFFD, 0x1FFFD, 0x3FFFD, 0x7FFFD, 0xFFFFD, 0x1FFFFD, 0x3FFFFD, 0x7FFFFD, - 0xFFFFFD, 0x1FFFFFD, 0x3FFFFFD, 0x7FFFFFD, 0xFFFFFFD, 0x1FFFFFFD, 0x3FFFFFFD, 0x7FFFFFFD }; - -static UNUSED_ATTR const U8 OF_bits[MaxOff+1] = { - 0, 1, 2, 3, 4, 5, 6, 7, - 8, 9, 10, 11, 12, 13, 14, 15, - 16, 17, 18, 19, 20, 21, 22, 23, - 24, 25, 26, 27, 28, 29, 30, 31 }; - -static UNUSED_ATTR const U32 ML_base[MaxML+1] = { - 3, 4, 5, 6, 7, 8, 9, 10, - 11, 12, 13, 14, 15, 16, 17, 18, - 19, 20, 21, 22, 23, 24, 25, 26, - 27, 28, 29, 30, 31, 32, 33, 34, - 35, 37, 39, 41, 43, 47, 51, 59, - 67, 83, 99, 0x83, 0x103, 0x203, 0x403, 0x803, - 0x1003, 0x2003, 0x4003, 0x8003, 0x10003 }; - - -/*-******************************************************* - * Decompression types - *********************************************************/ - typedef struct { - U32 fastMode; - U32 tableLog; - } ZSTD_seqSymbol_header; - - typedef struct { - U16 nextState; - BYTE nbAdditionalBits; - BYTE nbBits; - U32 baseValue; - } ZSTD_seqSymbol; - - #define SEQSYMBOL_TABLE_SIZE(log) (1 + (1 << (log))) - -#define ZSTD_BUILD_FSE_TABLE_WKSP_SIZE (sizeof(S16) * (MaxSeq + 1) + (1u << MaxFSELog) + sizeof(U64)) -#define ZSTD_BUILD_FSE_TABLE_WKSP_SIZE_U32 ((ZSTD_BUILD_FSE_TABLE_WKSP_SIZE + sizeof(U32) - 1) / sizeof(U32)) - -typedef struct { - ZSTD_seqSymbol LLTable[SEQSYMBOL_TABLE_SIZE(LLFSELog)]; /* Note : Space reserved for FSE Tables */ - ZSTD_seqSymbol OFTable[SEQSYMBOL_TABLE_SIZE(OffFSELog)]; /* is also used as temporary workspace while building hufTable during DDict creation */ - ZSTD_seqSymbol MLTable[SEQSYMBOL_TABLE_SIZE(MLFSELog)]; /* and therefore must be at least HUF_DECOMPRESS_WORKSPACE_SIZE large */ - HUF_DTable hufTable[HUF_DTABLE_SIZE(HufLog)]; /* can accommodate HUF_decompress4X */ - U32 rep[ZSTD_REP_NUM]; - U32 workspace[ZSTD_BUILD_FSE_TABLE_WKSP_SIZE_U32]; -} ZSTD_entropyDTables_t; - -typedef enum { ZSTDds_getFrameHeaderSize, ZSTDds_decodeFrameHeader, - ZSTDds_decodeBlockHeader, ZSTDds_decompressBlock, - ZSTDds_decompressLastBlock, ZSTDds_checkChecksum, - ZSTDds_decodeSkippableHeader, ZSTDds_skipFrame } ZSTD_dStage; - -typedef enum { zdss_init=0, zdss_loadHeader, - zdss_read, zdss_load, zdss_flush } ZSTD_dStreamStage; - -typedef enum { - ZSTD_use_indefinitely = -1, /* Use the dictionary indefinitely */ - ZSTD_dont_use = 0, /* Do not use the dictionary (if one exists free it) */ - ZSTD_use_once = 1 /* Use the dictionary once and set to ZSTD_dont_use */ -} ZSTD_dictUses_e; - -/* Hashset for storing references to multiple ZSTD_DDict within ZSTD_DCtx */ -typedef struct { - const ZSTD_DDict** ddictPtrTable; - size_t ddictPtrTableSize; - size_t ddictPtrCount; -} ZSTD_DDictHashSet; - -#ifndef ZSTD_DECODER_INTERNAL_BUFFER -# define ZSTD_DECODER_INTERNAL_BUFFER (1 << 16) -#endif - -#define ZSTD_LBMIN 64 -#define ZSTD_LBMAX (128 << 10) - -/* extra buffer, compensates when dst is not large enough to store litBuffer */ -#define ZSTD_LITBUFFEREXTRASIZE BOUNDED(ZSTD_LBMIN, ZSTD_DECODER_INTERNAL_BUFFER, ZSTD_LBMAX) - -typedef enum { - ZSTD_not_in_dst = 0, /* Stored entirely within litExtraBuffer */ - ZSTD_in_dst = 1, /* Stored entirely within dst (in memory after current output write) */ - ZSTD_split = 2 /* Split between litExtraBuffer and dst */ -} ZSTD_litLocation_e; - -struct ZSTD_DCtx_s -{ - const ZSTD_seqSymbol* LLTptr; - const ZSTD_seqSymbol* MLTptr; - const ZSTD_seqSymbol* OFTptr; - const HUF_DTable* HUFptr; - ZSTD_entropyDTables_t entropy; - U32 workspace[HUF_DECOMPRESS_WORKSPACE_SIZE_U32]; /* space needed when building huffman tables */ - const void* previousDstEnd; /* detect continuity */ - const void* prefixStart; /* start of current segment */ - const void* virtualStart; /* virtual start of previous segment if it was just before current one */ - const void* dictEnd; /* end of previous segment */ - size_t expected; - ZSTD_frameHeader fParams; - U64 processedCSize; - U64 decodedSize; - blockType_e bType; /* used in ZSTD_decompressContinue(), store blockType between block header decoding and block decompression stages */ - ZSTD_dStage stage; - U32 litEntropy; - U32 fseEntropy; - XXH64_state_t xxhState; - size_t headerSize; - ZSTD_format_e format; - ZSTD_forceIgnoreChecksum_e forceIgnoreChecksum; /* User specified: if == 1, will ignore checksums in compressed frame. Default == 0 */ - U32 validateChecksum; /* if == 1, will validate checksum. Is == 1 if (fParams.checksumFlag == 1) and (forceIgnoreChecksum == 0). */ - const BYTE* litPtr; - ZSTD_customMem customMem; - size_t litSize; - size_t rleSize; - size_t staticSize; -#if DYNAMIC_BMI2 != 0 - int bmi2; /* == 1 if the CPU supports BMI2 and 0 otherwise. CPU support is determined dynamically once per context lifetime. */ -#endif - - /* dictionary */ - ZSTD_DDict* ddictLocal; - const ZSTD_DDict* ddict; /* set by ZSTD_initDStream_usingDDict(), or ZSTD_DCtx_refDDict() */ - U32 dictID; - int ddictIsCold; /* if == 1 : dictionary is "new" for working context, and presumed "cold" (not in cpu cache) */ - ZSTD_dictUses_e dictUses; - ZSTD_DDictHashSet* ddictSet; /* Hash set for multiple ddicts */ - ZSTD_refMultipleDDicts_e refMultipleDDicts; /* User specified: if == 1, will allow references to multiple DDicts. Default == 0 (disabled) */ - - /* streaming */ - ZSTD_dStreamStage streamStage; - char* inBuff; - size_t inBuffSize; - size_t inPos; - size_t maxWindowSize; - char* outBuff; - size_t outBuffSize; - size_t outStart; - size_t outEnd; - size_t lhSize; -#if defined(ZSTD_LEGACY_SUPPORT) && (ZSTD_LEGACY_SUPPORT>=1) - void* legacyContext; - U32 previousLegacyVersion; - U32 legacyVersion; -#endif - U32 hostageByte; - int noForwardProgress; - ZSTD_bufferMode_e outBufferMode; - ZSTD_outBuffer expectedOutBuffer; - - /* workspace */ - BYTE* litBuffer; - const BYTE* litBufferEnd; - ZSTD_litLocation_e litBufferLocation; - BYTE litExtraBuffer[ZSTD_LITBUFFEREXTRASIZE + WILDCOPY_OVERLENGTH]; /* literal buffer can be split between storage within dst and within this scratch buffer */ - BYTE headerBuffer[ZSTD_FRAMEHEADERSIZE_MAX]; - - size_t oversizedDuration; - -#ifdef FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION - void const* dictContentBeginForFuzzing; - void const* dictContentEndForFuzzing; -#endif - - /* Tracing */ -#if ZSTD_TRACE - ZSTD_TraceCtx traceCtx; -#endif -}; /* typedef'd to ZSTD_DCtx within "zstd.h" */ - -MEM_STATIC int ZSTD_DCtx_get_bmi2(const struct ZSTD_DCtx_s *dctx) { -#if DYNAMIC_BMI2 != 0 - return dctx->bmi2; -#else - (void)dctx; - return 0; -#endif -} - -/*-******************************************************* - * Shared internal functions - *********************************************************/ - -/*! ZSTD_loadDEntropy() : - * dict : must point at beginning of a valid zstd dictionary. - * @return : size of dictionary header (size of magic number + dict ID + entropy tables) */ -size_t ZSTD_loadDEntropy(ZSTD_entropyDTables_t* entropy, - const void* const dict, size_t const dictSize); - -/*! ZSTD_checkContinuity() : - * check if next `dst` follows previous position, where decompression ended. - * If yes, do nothing (continue on current segment). - * If not, classify previous segment as "external dictionary", and start a new segment. - * This function cannot fail. */ -void ZSTD_checkContinuity(ZSTD_DCtx* dctx, const void* dst, size_t dstSize); - - -#endif /* ZSTD_DECOMPRESS_INTERNAL_H */ diff --git a/dep/zstd/lib/zdict.h b/dep/zstd/lib/zdict.h deleted file mode 100644 index f1e139a40..000000000 --- a/dep/zstd/lib/zdict.h +++ /dev/null @@ -1,452 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef DICTBUILDER_H_001 -#define DICTBUILDER_H_001 - -#if defined (__cplusplus) -extern "C" { -#endif - - -/*====== Dependencies ======*/ -#include /* size_t */ - - -/* ===== ZDICTLIB_API : control library symbols visibility ===== */ -#ifndef ZDICTLIB_VISIBILITY -# if defined(__GNUC__) && (__GNUC__ >= 4) -# define ZDICTLIB_VISIBILITY __attribute__ ((visibility ("default"))) -# else -# define ZDICTLIB_VISIBILITY -# endif -#endif -#if defined(ZSTD_DLL_EXPORT) && (ZSTD_DLL_EXPORT==1) -# define ZDICTLIB_API __declspec(dllexport) ZDICTLIB_VISIBILITY -#elif defined(ZSTD_DLL_IMPORT) && (ZSTD_DLL_IMPORT==1) -# define ZDICTLIB_API __declspec(dllimport) ZDICTLIB_VISIBILITY /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/ -#else -# define ZDICTLIB_API ZDICTLIB_VISIBILITY -#endif - -/******************************************************************************* - * Zstd dictionary builder - * - * FAQ - * === - * Why should I use a dictionary? - * ------------------------------ - * - * Zstd can use dictionaries to improve compression ratio of small data. - * Traditionally small files don't compress well because there is very little - * repetition in a single sample, since it is small. But, if you are compressing - * many similar files, like a bunch of JSON records that share the same - * structure, you can train a dictionary on ahead of time on some samples of - * these files. Then, zstd can use the dictionary to find repetitions that are - * present across samples. This can vastly improve compression ratio. - * - * When is a dictionary useful? - * ---------------------------- - * - * Dictionaries are useful when compressing many small files that are similar. - * The larger a file is, the less benefit a dictionary will have. Generally, - * we don't expect dictionary compression to be effective past 100KB. And the - * smaller a file is, the more we would expect the dictionary to help. - * - * How do I use a dictionary? - * -------------------------- - * - * Simply pass the dictionary to the zstd compressor with - * `ZSTD_CCtx_loadDictionary()`. The same dictionary must then be passed to - * the decompressor, using `ZSTD_DCtx_loadDictionary()`. There are other - * more advanced functions that allow selecting some options, see zstd.h for - * complete documentation. - * - * What is a zstd dictionary? - * -------------------------- - * - * A zstd dictionary has two pieces: Its header, and its content. The header - * contains a magic number, the dictionary ID, and entropy tables. These - * entropy tables allow zstd to save on header costs in the compressed file, - * which really matters for small data. The content is just bytes, which are - * repeated content that is common across many samples. - * - * What is a raw content dictionary? - * --------------------------------- - * - * A raw content dictionary is just bytes. It doesn't have a zstd dictionary - * header, a dictionary ID, or entropy tables. Any buffer is a valid raw - * content dictionary. - * - * How do I train a dictionary? - * ---------------------------- - * - * Gather samples from your use case. These samples should be similar to each - * other. If you have several use cases, you could try to train one dictionary - * per use case. - * - * Pass those samples to `ZDICT_trainFromBuffer()` and that will train your - * dictionary. There are a few advanced versions of this function, but this - * is a great starting point. If you want to further tune your dictionary - * you could try `ZDICT_optimizeTrainFromBuffer_cover()`. If that is too slow - * you can try `ZDICT_optimizeTrainFromBuffer_fastCover()`. - * - * If the dictionary training function fails, that is likely because you - * either passed too few samples, or a dictionary would not be effective - * for your data. Look at the messages that the dictionary trainer printed, - * if it doesn't say too few samples, then a dictionary would not be effective. - * - * How large should my dictionary be? - * ---------------------------------- - * - * A reasonable dictionary size, the `dictBufferCapacity`, is about 100KB. - * The zstd CLI defaults to a 110KB dictionary. You likely don't need a - * dictionary larger than that. But, most use cases can get away with a - * smaller dictionary. The advanced dictionary builders can automatically - * shrink the dictionary for you, and select a the smallest size that - * doesn't hurt compression ratio too much. See the `shrinkDict` parameter. - * A smaller dictionary can save memory, and potentially speed up - * compression. - * - * How many samples should I provide to the dictionary builder? - * ------------------------------------------------------------ - * - * We generally recommend passing ~100x the size of the dictionary - * in samples. A few thousand should suffice. Having too few samples - * can hurt the dictionaries effectiveness. Having more samples will - * only improve the dictionaries effectiveness. But having too many - * samples can slow down the dictionary builder. - * - * How do I determine if a dictionary will be effective? - * ----------------------------------------------------- - * - * Simply train a dictionary and try it out. You can use zstd's built in - * benchmarking tool to test the dictionary effectiveness. - * - * # Benchmark levels 1-3 without a dictionary - * zstd -b1e3 -r /path/to/my/files - * # Benchmark levels 1-3 with a dictionary - * zstd -b1e3 -r /path/to/my/files -D /path/to/my/dictionary - * - * When should I retrain a dictionary? - * ----------------------------------- - * - * You should retrain a dictionary when its effectiveness drops. Dictionary - * effectiveness drops as the data you are compressing changes. Generally, we do - * expect dictionaries to "decay" over time, as your data changes, but the rate - * at which they decay depends on your use case. Internally, we regularly - * retrain dictionaries, and if the new dictionary performs significantly - * better than the old dictionary, we will ship the new dictionary. - * - * I have a raw content dictionary, how do I turn it into a zstd dictionary? - * ------------------------------------------------------------------------- - * - * If you have a raw content dictionary, e.g. by manually constructing it, or - * using a third-party dictionary builder, you can turn it into a zstd - * dictionary by using `ZDICT_finalizeDictionary()`. You'll also have to - * provide some samples of the data. It will add the zstd header to the - * raw content, which contains a dictionary ID and entropy tables, which - * will improve compression ratio, and allow zstd to write the dictionary ID - * into the frame, if you so choose. - * - * Do I have to use zstd's dictionary builder? - * ------------------------------------------- - * - * No! You can construct dictionary content however you please, it is just - * bytes. It will always be valid as a raw content dictionary. If you want - * a zstd dictionary, which can improve compression ratio, use - * `ZDICT_finalizeDictionary()`. - * - * What is the attack surface of a zstd dictionary? - * ------------------------------------------------ - * - * Zstd is heavily fuzz tested, including loading fuzzed dictionaries, so - * zstd should never crash, or access out-of-bounds memory no matter what - * the dictionary is. However, if an attacker can control the dictionary - * during decompression, they can cause zstd to generate arbitrary bytes, - * just like if they controlled the compressed data. - * - ******************************************************************************/ - - -/*! ZDICT_trainFromBuffer(): - * Train a dictionary from an array of samples. - * Redirect towards ZDICT_optimizeTrainFromBuffer_fastCover() single-threaded, with d=8, steps=4, - * f=20, and accel=1. - * Samples must be stored concatenated in a single flat buffer `samplesBuffer`, - * supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order. - * The resulting dictionary will be saved into `dictBuffer`. - * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`) - * or an error code, which can be tested with ZDICT_isError(). - * Note: Dictionary training will fail if there are not enough samples to construct a - * dictionary, or if most of the samples are too small (< 8 bytes being the lower limit). - * If dictionary training fails, you should use zstd without a dictionary, as the dictionary - * would've been ineffective anyways. If you believe your samples would benefit from a dictionary - * please open an issue with details, and we can look into it. - * Note: ZDICT_trainFromBuffer()'s memory usage is about 6 MB. - * Tips: In general, a reasonable dictionary has a size of ~ 100 KB. - * It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`. - * In general, it's recommended to provide a few thousands samples, though this can vary a lot. - * It's recommended that total size of all samples be about ~x100 times the target size of dictionary. - */ -ZDICTLIB_API size_t ZDICT_trainFromBuffer(void* dictBuffer, size_t dictBufferCapacity, - const void* samplesBuffer, - const size_t* samplesSizes, unsigned nbSamples); - -typedef struct { - int compressionLevel; /*< optimize for a specific zstd compression level; 0 means default */ - unsigned notificationLevel; /*< Write log to stderr; 0 = none (default); 1 = errors; 2 = progression; 3 = details; 4 = debug; */ - unsigned dictID; /*< force dictID value; 0 means auto mode (32-bits random value) - * NOTE: The zstd format reserves some dictionary IDs for future use. - * You may use them in private settings, but be warned that they - * may be used by zstd in a public dictionary registry in the future. - * These dictionary IDs are: - * - low range : <= 32767 - * - high range : >= (2^31) - */ -} ZDICT_params_t; - -/*! ZDICT_finalizeDictionary(): - * Given a custom content as a basis for dictionary, and a set of samples, - * finalize dictionary by adding headers and statistics according to the zstd - * dictionary format. - * - * Samples must be stored concatenated in a flat buffer `samplesBuffer`, - * supplied with an array of sizes `samplesSizes`, providing the size of each - * sample in order. The samples are used to construct the statistics, so they - * should be representative of what you will compress with this dictionary. - * - * The compression level can be set in `parameters`. You should pass the - * compression level you expect to use in production. The statistics for each - * compression level differ, so tuning the dictionary for the compression level - * can help quite a bit. - * - * You can set an explicit dictionary ID in `parameters`, or allow us to pick - * a random dictionary ID for you, but we can't guarantee no collisions. - * - * The dstDictBuffer and the dictContent may overlap, and the content will be - * appended to the end of the header. If the header + the content doesn't fit in - * maxDictSize the beginning of the content is truncated to make room, since it - * is presumed that the most profitable content is at the end of the dictionary, - * since that is the cheapest to reference. - * - * `maxDictSize` must be >= max(dictContentSize, ZSTD_DICTSIZE_MIN). - * - * @return: size of dictionary stored into `dstDictBuffer` (<= `maxDictSize`), - * or an error code, which can be tested by ZDICT_isError(). - * Note: ZDICT_finalizeDictionary() will push notifications into stderr if - * instructed to, using notificationLevel>0. - * NOTE: This function currently may fail in several edge cases including: - * * Not enough samples - * * Samples are uncompressible - * * Samples are all exactly the same - */ -ZDICTLIB_API size_t ZDICT_finalizeDictionary(void* dstDictBuffer, size_t maxDictSize, - const void* dictContent, size_t dictContentSize, - const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples, - ZDICT_params_t parameters); - - -/*====== Helper functions ======*/ -ZDICTLIB_API unsigned ZDICT_getDictID(const void* dictBuffer, size_t dictSize); /**< extracts dictID; @return zero if error (not a valid dictionary) */ -ZDICTLIB_API size_t ZDICT_getDictHeaderSize(const void* dictBuffer, size_t dictSize); /* returns dict header size; returns a ZSTD error code on failure */ -ZDICTLIB_API unsigned ZDICT_isError(size_t errorCode); -ZDICTLIB_API const char* ZDICT_getErrorName(size_t errorCode); - - - -#ifdef ZDICT_STATIC_LINKING_ONLY - -/* ==================================================================================== - * The definitions in this section are considered experimental. - * They should never be used with a dynamic library, as they may change in the future. - * They are provided for advanced usages. - * Use them only in association with static linking. - * ==================================================================================== */ - -#define ZDICT_DICTSIZE_MIN 256 -/* Deprecated: Remove in v1.6.0 */ -#define ZDICT_CONTENTSIZE_MIN 128 - -/*! ZDICT_cover_params_t: - * k and d are the only required parameters. - * For others, value 0 means default. - */ -typedef struct { - unsigned k; /* Segment size : constraint: 0 < k : Reasonable range [16, 2048+] */ - unsigned d; /* dmer size : constraint: 0 < d <= k : Reasonable range [6, 16] */ - unsigned steps; /* Number of steps : Only used for optimization : 0 means default (40) : Higher means more parameters checked */ - unsigned nbThreads; /* Number of threads : constraint: 0 < nbThreads : 1 means single-threaded : Only used for optimization : Ignored if ZSTD_MULTITHREAD is not defined */ - double splitPoint; /* Percentage of samples used for training: Only used for optimization : the first nbSamples * splitPoint samples will be used to training, the last nbSamples * (1 - splitPoint) samples will be used for testing, 0 means default (1.0), 1.0 when all samples are used for both training and testing */ - unsigned shrinkDict; /* Train dictionaries to shrink in size starting from the minimum size and selects the smallest dictionary that is shrinkDictMaxRegression% worse than the largest dictionary. 0 means no shrinking and 1 means shrinking */ - unsigned shrinkDictMaxRegression; /* Sets shrinkDictMaxRegression so that a smaller dictionary can be at worse shrinkDictMaxRegression% worse than the max dict size dictionary. */ - ZDICT_params_t zParams; -} ZDICT_cover_params_t; - -typedef struct { - unsigned k; /* Segment size : constraint: 0 < k : Reasonable range [16, 2048+] */ - unsigned d; /* dmer size : constraint: 0 < d <= k : Reasonable range [6, 16] */ - unsigned f; /* log of size of frequency array : constraint: 0 < f <= 31 : 1 means default(20)*/ - unsigned steps; /* Number of steps : Only used for optimization : 0 means default (40) : Higher means more parameters checked */ - unsigned nbThreads; /* Number of threads : constraint: 0 < nbThreads : 1 means single-threaded : Only used for optimization : Ignored if ZSTD_MULTITHREAD is not defined */ - double splitPoint; /* Percentage of samples used for training: Only used for optimization : the first nbSamples * splitPoint samples will be used to training, the last nbSamples * (1 - splitPoint) samples will be used for testing, 0 means default (0.75), 1.0 when all samples are used for both training and testing */ - unsigned accel; /* Acceleration level: constraint: 0 < accel <= 10, higher means faster and less accurate, 0 means default(1) */ - unsigned shrinkDict; /* Train dictionaries to shrink in size starting from the minimum size and selects the smallest dictionary that is shrinkDictMaxRegression% worse than the largest dictionary. 0 means no shrinking and 1 means shrinking */ - unsigned shrinkDictMaxRegression; /* Sets shrinkDictMaxRegression so that a smaller dictionary can be at worse shrinkDictMaxRegression% worse than the max dict size dictionary. */ - - ZDICT_params_t zParams; -} ZDICT_fastCover_params_t; - -/*! ZDICT_trainFromBuffer_cover(): - * Train a dictionary from an array of samples using the COVER algorithm. - * Samples must be stored concatenated in a single flat buffer `samplesBuffer`, - * supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order. - * The resulting dictionary will be saved into `dictBuffer`. - * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`) - * or an error code, which can be tested with ZDICT_isError(). - * See ZDICT_trainFromBuffer() for details on failure modes. - * Note: ZDICT_trainFromBuffer_cover() requires about 9 bytes of memory for each input byte. - * Tips: In general, a reasonable dictionary has a size of ~ 100 KB. - * It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`. - * In general, it's recommended to provide a few thousands samples, though this can vary a lot. - * It's recommended that total size of all samples be about ~x100 times the target size of dictionary. - */ -ZDICTLIB_API size_t ZDICT_trainFromBuffer_cover( - void *dictBuffer, size_t dictBufferCapacity, - const void *samplesBuffer, const size_t *samplesSizes, unsigned nbSamples, - ZDICT_cover_params_t parameters); - -/*! ZDICT_optimizeTrainFromBuffer_cover(): - * The same requirements as above hold for all the parameters except `parameters`. - * This function tries many parameter combinations and picks the best parameters. - * `*parameters` is filled with the best parameters found, - * dictionary constructed with those parameters is stored in `dictBuffer`. - * - * All of the parameters d, k, steps are optional. - * If d is non-zero then we don't check multiple values of d, otherwise we check d = {6, 8}. - * if steps is zero it defaults to its default value. - * If k is non-zero then we don't check multiple values of k, otherwise we check steps values in [50, 2000]. - * - * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`) - * or an error code, which can be tested with ZDICT_isError(). - * On success `*parameters` contains the parameters selected. - * See ZDICT_trainFromBuffer() for details on failure modes. - * Note: ZDICT_optimizeTrainFromBuffer_cover() requires about 8 bytes of memory for each input byte and additionally another 5 bytes of memory for each byte of memory for each thread. - */ -ZDICTLIB_API size_t ZDICT_optimizeTrainFromBuffer_cover( - void* dictBuffer, size_t dictBufferCapacity, - const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples, - ZDICT_cover_params_t* parameters); - -/*! ZDICT_trainFromBuffer_fastCover(): - * Train a dictionary from an array of samples using a modified version of COVER algorithm. - * Samples must be stored concatenated in a single flat buffer `samplesBuffer`, - * supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order. - * d and k are required. - * All other parameters are optional, will use default values if not provided - * The resulting dictionary will be saved into `dictBuffer`. - * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`) - * or an error code, which can be tested with ZDICT_isError(). - * See ZDICT_trainFromBuffer() for details on failure modes. - * Note: ZDICT_trainFromBuffer_fastCover() requires 6 * 2^f bytes of memory. - * Tips: In general, a reasonable dictionary has a size of ~ 100 KB. - * It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`. - * In general, it's recommended to provide a few thousands samples, though this can vary a lot. - * It's recommended that total size of all samples be about ~x100 times the target size of dictionary. - */ -ZDICTLIB_API size_t ZDICT_trainFromBuffer_fastCover(void *dictBuffer, - size_t dictBufferCapacity, const void *samplesBuffer, - const size_t *samplesSizes, unsigned nbSamples, - ZDICT_fastCover_params_t parameters); - -/*! ZDICT_optimizeTrainFromBuffer_fastCover(): - * The same requirements as above hold for all the parameters except `parameters`. - * This function tries many parameter combinations (specifically, k and d combinations) - * and picks the best parameters. `*parameters` is filled with the best parameters found, - * dictionary constructed with those parameters is stored in `dictBuffer`. - * All of the parameters d, k, steps, f, and accel are optional. - * If d is non-zero then we don't check multiple values of d, otherwise we check d = {6, 8}. - * if steps is zero it defaults to its default value. - * If k is non-zero then we don't check multiple values of k, otherwise we check steps values in [50, 2000]. - * If f is zero, default value of 20 is used. - * If accel is zero, default value of 1 is used. - * - * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`) - * or an error code, which can be tested with ZDICT_isError(). - * On success `*parameters` contains the parameters selected. - * See ZDICT_trainFromBuffer() for details on failure modes. - * Note: ZDICT_optimizeTrainFromBuffer_fastCover() requires about 6 * 2^f bytes of memory for each thread. - */ -ZDICTLIB_API size_t ZDICT_optimizeTrainFromBuffer_fastCover(void* dictBuffer, - size_t dictBufferCapacity, const void* samplesBuffer, - const size_t* samplesSizes, unsigned nbSamples, - ZDICT_fastCover_params_t* parameters); - -typedef struct { - unsigned selectivityLevel; /* 0 means default; larger => select more => larger dictionary */ - ZDICT_params_t zParams; -} ZDICT_legacy_params_t; - -/*! ZDICT_trainFromBuffer_legacy(): - * Train a dictionary from an array of samples. - * Samples must be stored concatenated in a single flat buffer `samplesBuffer`, - * supplied with an array of sizes `samplesSizes`, providing the size of each sample, in order. - * The resulting dictionary will be saved into `dictBuffer`. - * `parameters` is optional and can be provided with values set to 0 to mean "default". - * @return: size of dictionary stored into `dictBuffer` (<= `dictBufferCapacity`) - * or an error code, which can be tested with ZDICT_isError(). - * See ZDICT_trainFromBuffer() for details on failure modes. - * Tips: In general, a reasonable dictionary has a size of ~ 100 KB. - * It's possible to select smaller or larger size, just by specifying `dictBufferCapacity`. - * In general, it's recommended to provide a few thousands samples, though this can vary a lot. - * It's recommended that total size of all samples be about ~x100 times the target size of dictionary. - * Note: ZDICT_trainFromBuffer_legacy() will send notifications into stderr if instructed to, using notificationLevel>0. - */ -ZDICTLIB_API size_t ZDICT_trainFromBuffer_legacy( - void* dictBuffer, size_t dictBufferCapacity, - const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples, - ZDICT_legacy_params_t parameters); - - -/* Deprecation warnings */ -/* It is generally possible to disable deprecation warnings from compiler, - for example with -Wno-deprecated-declarations for gcc - or _CRT_SECURE_NO_WARNINGS in Visual. - Otherwise, it's also possible to manually define ZDICT_DISABLE_DEPRECATE_WARNINGS */ -#ifdef ZDICT_DISABLE_DEPRECATE_WARNINGS -# define ZDICT_DEPRECATED(message) ZDICTLIB_API /* disable deprecation warnings */ -#else -# define ZDICT_GCC_VERSION (__GNUC__ * 100 + __GNUC_MINOR__) -# if defined (__cplusplus) && (__cplusplus >= 201402) /* C++14 or greater */ -# define ZDICT_DEPRECATED(message) [[deprecated(message)]] ZDICTLIB_API -# elif defined(__clang__) || (ZDICT_GCC_VERSION >= 405) -# define ZDICT_DEPRECATED(message) ZDICTLIB_API __attribute__((deprecated(message))) -# elif (ZDICT_GCC_VERSION >= 301) -# define ZDICT_DEPRECATED(message) ZDICTLIB_API __attribute__((deprecated)) -# elif defined(_MSC_VER) -# define ZDICT_DEPRECATED(message) ZDICTLIB_API __declspec(deprecated(message)) -# else -# pragma message("WARNING: You need to implement ZDICT_DEPRECATED for this compiler") -# define ZDICT_DEPRECATED(message) ZDICTLIB_API -# endif -#endif /* ZDICT_DISABLE_DEPRECATE_WARNINGS */ - -ZDICT_DEPRECATED("use ZDICT_finalizeDictionary() instead") -size_t ZDICT_addEntropyTablesFromBuffer(void* dictBuffer, size_t dictContentSize, size_t dictBufferCapacity, - const void* samplesBuffer, const size_t* samplesSizes, unsigned nbSamples); - - -#endif /* ZDICT_STATIC_LINKING_ONLY */ - -#if defined (__cplusplus) -} -#endif - -#endif /* DICTBUILDER_H_001 */ diff --git a/dep/zstd/lib/zstd.h b/dep/zstd/lib/zstd.h deleted file mode 100644 index a88ae7bf8..000000000 --- a/dep/zstd/lib/zstd.h +++ /dev/null @@ -1,2575 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ -#if defined (__cplusplus) -extern "C" { -#endif - -#ifndef ZSTD_H_235446 -#define ZSTD_H_235446 - -/* ====== Dependency ======*/ -#include /* INT_MAX */ -#include /* size_t */ - - -/* ===== ZSTDLIB_API : control library symbols visibility ===== */ -#ifndef ZSTDLIB_VISIBLE -# if defined(__GNUC__) && (__GNUC__ >= 4) && !defined(__MINGW32__) -# define ZSTDLIB_VISIBLE __attribute__ ((visibility ("default"))) -# define ZSTDLIB_HIDDEN __attribute__ ((visibility ("hidden"))) -# else -# define ZSTDLIB_VISIBLE -# define ZSTDLIB_HIDDEN -# endif -#endif -#if defined(ZSTD_DLL_EXPORT) && (ZSTD_DLL_EXPORT==1) -# define ZSTDLIB_API __declspec(dllexport) ZSTDLIB_VISIBLE -#elif defined(ZSTD_DLL_IMPORT) && (ZSTD_DLL_IMPORT==1) -# define ZSTDLIB_API __declspec(dllimport) ZSTDLIB_VISIBLE /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/ -#else -# define ZSTDLIB_API ZSTDLIB_VISIBLE -#endif - - -/******************************************************************************* - Introduction - - zstd, short for Zstandard, is a fast lossless compression algorithm, targeting - real-time compression scenarios at zlib-level and better compression ratios. - The zstd compression library provides in-memory compression and decompression - functions. - - The library supports regular compression levels from 1 up to ZSTD_maxCLevel(), - which is currently 22. Levels >= 20, labeled `--ultra`, should be used with - caution, as they require more memory. The library also offers negative - compression levels, which extend the range of speed vs. ratio preferences. - The lower the level, the faster the speed (at the cost of compression). - - Compression can be done in: - - a single step (described as Simple API) - - a single step, reusing a context (described as Explicit context) - - unbounded multiple steps (described as Streaming compression) - - The compression ratio achievable on small data can be highly improved using - a dictionary. Dictionary compression can be performed in: - - a single step (described as Simple dictionary API) - - a single step, reusing a dictionary (described as Bulk-processing - dictionary API) - - Advanced experimental functions can be accessed using - `#define ZSTD_STATIC_LINKING_ONLY` before including zstd.h. - - Advanced experimental APIs should never be used with a dynamically-linked - library. They are not "stable"; their definitions or signatures may change in - the future. Only static linking is allowed. -*******************************************************************************/ - -/*------ Version ------*/ -#define ZSTD_VERSION_MAJOR 1 -#define ZSTD_VERSION_MINOR 5 -#define ZSTD_VERSION_RELEASE 2 -#define ZSTD_VERSION_NUMBER (ZSTD_VERSION_MAJOR *100*100 + ZSTD_VERSION_MINOR *100 + ZSTD_VERSION_RELEASE) - -/*! ZSTD_versionNumber() : - * Return runtime library version, the value is (MAJOR*100*100 + MINOR*100 + RELEASE). */ -ZSTDLIB_API unsigned ZSTD_versionNumber(void); - -#define ZSTD_LIB_VERSION ZSTD_VERSION_MAJOR.ZSTD_VERSION_MINOR.ZSTD_VERSION_RELEASE -#define ZSTD_QUOTE(str) #str -#define ZSTD_EXPAND_AND_QUOTE(str) ZSTD_QUOTE(str) -#define ZSTD_VERSION_STRING ZSTD_EXPAND_AND_QUOTE(ZSTD_LIB_VERSION) - -/*! ZSTD_versionString() : - * Return runtime library version, like "1.4.5". Requires v1.3.0+. */ -ZSTDLIB_API const char* ZSTD_versionString(void); - -/* ************************************* - * Default constant - ***************************************/ -#ifndef ZSTD_CLEVEL_DEFAULT -# define ZSTD_CLEVEL_DEFAULT 3 -#endif - -/* ************************************* - * Constants - ***************************************/ - -/* All magic numbers are supposed read/written to/from files/memory using little-endian convention */ -#define ZSTD_MAGICNUMBER 0xFD2FB528 /* valid since v0.8.0 */ -#define ZSTD_MAGIC_DICTIONARY 0xEC30A437 /* valid since v0.7.0 */ -#define ZSTD_MAGIC_SKIPPABLE_START 0x184D2A50 /* all 16 values, from 0x184D2A50 to 0x184D2A5F, signal the beginning of a skippable frame */ -#define ZSTD_MAGIC_SKIPPABLE_MASK 0xFFFFFFF0 - -#define ZSTD_BLOCKSIZELOG_MAX 17 -#define ZSTD_BLOCKSIZE_MAX (1<= `ZSTD_compressBound(srcSize)`. - * @return : compressed size written into `dst` (<= `dstCapacity), - * or an error code if it fails (which can be tested using ZSTD_isError()). */ -ZSTDLIB_API size_t ZSTD_compress( void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - int compressionLevel); - -/*! ZSTD_decompress() : - * `compressedSize` : must be the _exact_ size of some number of compressed and/or skippable frames. - * `dstCapacity` is an upper bound of originalSize to regenerate. - * If user cannot imply a maximum upper bound, it's better to use streaming mode to decompress data. - * @return : the number of bytes decompressed into `dst` (<= `dstCapacity`), - * or an errorCode if it fails (which can be tested using ZSTD_isError()). */ -ZSTDLIB_API size_t ZSTD_decompress( void* dst, size_t dstCapacity, - const void* src, size_t compressedSize); - -/*! ZSTD_getFrameContentSize() : requires v1.3.0+ - * `src` should point to the start of a ZSTD encoded frame. - * `srcSize` must be at least as large as the frame header. - * hint : any size >= `ZSTD_frameHeaderSize_max` is large enough. - * @return : - decompressed size of `src` frame content, if known - * - ZSTD_CONTENTSIZE_UNKNOWN if the size cannot be determined - * - ZSTD_CONTENTSIZE_ERROR if an error occurred (e.g. invalid magic number, srcSize too small) - * note 1 : a 0 return value means the frame is valid but "empty". - * note 2 : decompressed size is an optional field, it may not be present, typically in streaming mode. - * When `return==ZSTD_CONTENTSIZE_UNKNOWN`, data to decompress could be any size. - * In which case, it's necessary to use streaming mode to decompress data. - * Optionally, application can rely on some implicit limit, - * as ZSTD_decompress() only needs an upper bound of decompressed size. - * (For example, data could be necessarily cut into blocks <= 16 KB). - * note 3 : decompressed size is always present when compression is completed using single-pass functions, - * such as ZSTD_compress(), ZSTD_compressCCtx() ZSTD_compress_usingDict() or ZSTD_compress_usingCDict(). - * note 4 : decompressed size can be very large (64-bits value), - * potentially larger than what local system can handle as a single memory segment. - * In which case, it's necessary to use streaming mode to decompress data. - * note 5 : If source is untrusted, decompressed size could be wrong or intentionally modified. - * Always ensure return value fits within application's authorized limits. - * Each application can set its own limits. - * note 6 : This function replaces ZSTD_getDecompressedSize() */ -#define ZSTD_CONTENTSIZE_UNKNOWN (0ULL - 1) -#define ZSTD_CONTENTSIZE_ERROR (0ULL - 2) -ZSTDLIB_API unsigned long long ZSTD_getFrameContentSize(const void *src, size_t srcSize); - -/*! ZSTD_getDecompressedSize() : - * NOTE: This function is now obsolete, in favor of ZSTD_getFrameContentSize(). - * Both functions work the same way, but ZSTD_getDecompressedSize() blends - * "empty", "unknown" and "error" results to the same return value (0), - * while ZSTD_getFrameContentSize() gives them separate return values. - * @return : decompressed size of `src` frame content _if known and not empty_, 0 otherwise. */ -ZSTDLIB_API unsigned long long ZSTD_getDecompressedSize(const void* src, size_t srcSize); - -/*! ZSTD_findFrameCompressedSize() : Requires v1.4.0+ - * `src` should point to the start of a ZSTD frame or skippable frame. - * `srcSize` must be >= first frame size - * @return : the compressed size of the first frame starting at `src`, - * suitable to pass as `srcSize` to `ZSTD_decompress` or similar, - * or an error code if input is invalid */ -ZSTDLIB_API size_t ZSTD_findFrameCompressedSize(const void* src, size_t srcSize); - - -/*====== Helper functions ======*/ -#define ZSTD_COMPRESSBOUND(srcSize) ((srcSize) + ((srcSize)>>8) + (((srcSize) < (128<<10)) ? (((128<<10) - (srcSize)) >> 11) /* margin, from 64 to 0 */ : 0)) /* this formula ensures that bound(A) + bound(B) <= bound(A+B) as long as A and B >= 128 KB */ -ZSTDLIB_API size_t ZSTD_compressBound(size_t srcSize); /*!< maximum compressed size in worst case single-pass scenario */ -ZSTDLIB_API unsigned ZSTD_isError(size_t code); /*!< tells if a `size_t` function result is an error code */ -ZSTDLIB_API const char* ZSTD_getErrorName(size_t code); /*!< provides readable string from an error code */ -ZSTDLIB_API int ZSTD_minCLevel(void); /*!< minimum negative compression level allowed, requires v1.4.0+ */ -ZSTDLIB_API int ZSTD_maxCLevel(void); /*!< maximum compression level available */ -ZSTDLIB_API int ZSTD_defaultCLevel(void); /*!< default compression level, specified by ZSTD_CLEVEL_DEFAULT, requires v1.5.0+ */ - - -/*************************************** -* Explicit context -***************************************/ -/*= Compression context - * When compressing many times, - * it is recommended to allocate a context just once, - * and re-use it for each successive compression operation. - * This will make workload friendlier for system's memory. - * Note : re-using context is just a speed / resource optimization. - * It doesn't change the compression ratio, which remains identical. - * Note 2 : In multi-threaded environments, - * use one different context per thread for parallel execution. - */ -typedef struct ZSTD_CCtx_s ZSTD_CCtx; -ZSTDLIB_API ZSTD_CCtx* ZSTD_createCCtx(void); -ZSTDLIB_API size_t ZSTD_freeCCtx(ZSTD_CCtx* cctx); /* accept NULL pointer */ - -/*! ZSTD_compressCCtx() : - * Same as ZSTD_compress(), using an explicit ZSTD_CCtx. - * Important : in order to behave similarly to `ZSTD_compress()`, - * this function compresses at requested compression level, - * __ignoring any other parameter__ . - * If any advanced parameter was set using the advanced API, - * they will all be reset. Only `compressionLevel` remains. - */ -ZSTDLIB_API size_t ZSTD_compressCCtx(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - int compressionLevel); - -/*= Decompression context - * When decompressing many times, - * it is recommended to allocate a context only once, - * and re-use it for each successive compression operation. - * This will make workload friendlier for system's memory. - * Use one context per thread for parallel execution. */ -typedef struct ZSTD_DCtx_s ZSTD_DCtx; -ZSTDLIB_API ZSTD_DCtx* ZSTD_createDCtx(void); -ZSTDLIB_API size_t ZSTD_freeDCtx(ZSTD_DCtx* dctx); /* accept NULL pointer */ - -/*! ZSTD_decompressDCtx() : - * Same as ZSTD_decompress(), - * requires an allocated ZSTD_DCtx. - * Compatible with sticky parameters. - */ -ZSTDLIB_API size_t ZSTD_decompressDCtx(ZSTD_DCtx* dctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize); - - -/********************************************* -* Advanced compression API (Requires v1.4.0+) -**********************************************/ - -/* API design : - * Parameters are pushed one by one into an existing context, - * using ZSTD_CCtx_set*() functions. - * Pushed parameters are sticky : they are valid for next compressed frame, and any subsequent frame. - * "sticky" parameters are applicable to `ZSTD_compress2()` and `ZSTD_compressStream*()` ! - * __They do not apply to "simple" one-shot variants such as ZSTD_compressCCtx()__ . - * - * It's possible to reset all parameters to "default" using ZSTD_CCtx_reset(). - * - * This API supersedes all other "advanced" API entry points in the experimental section. - * In the future, we expect to remove from experimental API entry points which are redundant with this API. - */ - - -/* Compression strategies, listed from fastest to strongest */ -typedef enum { ZSTD_fast=1, - ZSTD_dfast=2, - ZSTD_greedy=3, - ZSTD_lazy=4, - ZSTD_lazy2=5, - ZSTD_btlazy2=6, - ZSTD_btopt=7, - ZSTD_btultra=8, - ZSTD_btultra2=9 - /* note : new strategies _might_ be added in the future. - Only the order (from fast to strong) is guaranteed */ -} ZSTD_strategy; - -typedef enum { - - /* compression parameters - * Note: When compressing with a ZSTD_CDict these parameters are superseded - * by the parameters used to construct the ZSTD_CDict. - * See ZSTD_CCtx_refCDict() for more info (superseded-by-cdict). */ - ZSTD_c_compressionLevel=100, /* Set compression parameters according to pre-defined cLevel table. - * Note that exact compression parameters are dynamically determined, - * depending on both compression level and srcSize (when known). - * Default level is ZSTD_CLEVEL_DEFAULT==3. - * Special: value 0 means default, which is controlled by ZSTD_CLEVEL_DEFAULT. - * Note 1 : it's possible to pass a negative compression level. - * Note 2 : setting a level does not automatically set all other compression parameters - * to default. Setting this will however eventually dynamically impact the compression - * parameters which have not been manually set. The manually set - * ones will 'stick'. */ - /* Advanced compression parameters : - * It's possible to pin down compression parameters to some specific values. - * In which case, these values are no longer dynamically selected by the compressor */ - ZSTD_c_windowLog=101, /* Maximum allowed back-reference distance, expressed as power of 2. - * This will set a memory budget for streaming decompression, - * with larger values requiring more memory - * and typically compressing more. - * Must be clamped between ZSTD_WINDOWLOG_MIN and ZSTD_WINDOWLOG_MAX. - * Special: value 0 means "use default windowLog". - * Note: Using a windowLog greater than ZSTD_WINDOWLOG_LIMIT_DEFAULT - * requires explicitly allowing such size at streaming decompression stage. */ - ZSTD_c_hashLog=102, /* Size of the initial probe table, as a power of 2. - * Resulting memory usage is (1 << (hashLog+2)). - * Must be clamped between ZSTD_HASHLOG_MIN and ZSTD_HASHLOG_MAX. - * Larger tables improve compression ratio of strategies <= dFast, - * and improve speed of strategies > dFast. - * Special: value 0 means "use default hashLog". */ - ZSTD_c_chainLog=103, /* Size of the multi-probe search table, as a power of 2. - * Resulting memory usage is (1 << (chainLog+2)). - * Must be clamped between ZSTD_CHAINLOG_MIN and ZSTD_CHAINLOG_MAX. - * Larger tables result in better and slower compression. - * This parameter is useless for "fast" strategy. - * It's still useful when using "dfast" strategy, - * in which case it defines a secondary probe table. - * Special: value 0 means "use default chainLog". */ - ZSTD_c_searchLog=104, /* Number of search attempts, as a power of 2. - * More attempts result in better and slower compression. - * This parameter is useless for "fast" and "dFast" strategies. - * Special: value 0 means "use default searchLog". */ - ZSTD_c_minMatch=105, /* Minimum size of searched matches. - * Note that Zstandard can still find matches of smaller size, - * it just tweaks its search algorithm to look for this size and larger. - * Larger values increase compression and decompression speed, but decrease ratio. - * Must be clamped between ZSTD_MINMATCH_MIN and ZSTD_MINMATCH_MAX. - * Note that currently, for all strategies < btopt, effective minimum is 4. - * , for all strategies > fast, effective maximum is 6. - * Special: value 0 means "use default minMatchLength". */ - ZSTD_c_targetLength=106, /* Impact of this field depends on strategy. - * For strategies btopt, btultra & btultra2: - * Length of Match considered "good enough" to stop search. - * Larger values make compression stronger, and slower. - * For strategy fast: - * Distance between match sampling. - * Larger values make compression faster, and weaker. - * Special: value 0 means "use default targetLength". */ - ZSTD_c_strategy=107, /* See ZSTD_strategy enum definition. - * The higher the value of selected strategy, the more complex it is, - * resulting in stronger and slower compression. - * Special: value 0 means "use default strategy". */ - /* LDM mode parameters */ - ZSTD_c_enableLongDistanceMatching=160, /* Enable long distance matching. - * This parameter is designed to improve compression ratio - * for large inputs, by finding large matches at long distance. - * It increases memory usage and window size. - * Note: enabling this parameter increases default ZSTD_c_windowLog to 128 MB - * except when expressly set to a different value. - * Note: will be enabled by default if ZSTD_c_windowLog >= 128 MB and - * compression strategy >= ZSTD_btopt (== compression level 16+) */ - ZSTD_c_ldmHashLog=161, /* Size of the table for long distance matching, as a power of 2. - * Larger values increase memory usage and compression ratio, - * but decrease compression speed. - * Must be clamped between ZSTD_HASHLOG_MIN and ZSTD_HASHLOG_MAX - * default: windowlog - 7. - * Special: value 0 means "automatically determine hashlog". */ - ZSTD_c_ldmMinMatch=162, /* Minimum match size for long distance matcher. - * Larger/too small values usually decrease compression ratio. - * Must be clamped between ZSTD_LDM_MINMATCH_MIN and ZSTD_LDM_MINMATCH_MAX. - * Special: value 0 means "use default value" (default: 64). */ - ZSTD_c_ldmBucketSizeLog=163, /* Log size of each bucket in the LDM hash table for collision resolution. - * Larger values improve collision resolution but decrease compression speed. - * The maximum value is ZSTD_LDM_BUCKETSIZELOG_MAX. - * Special: value 0 means "use default value" (default: 3). */ - ZSTD_c_ldmHashRateLog=164, /* Frequency of inserting/looking up entries into the LDM hash table. - * Must be clamped between 0 and (ZSTD_WINDOWLOG_MAX - ZSTD_HASHLOG_MIN). - * Default is MAX(0, (windowLog - ldmHashLog)), optimizing hash table usage. - * Larger values improve compression speed. - * Deviating far from default value will likely result in a compression ratio decrease. - * Special: value 0 means "automatically determine hashRateLog". */ - - /* frame parameters */ - ZSTD_c_contentSizeFlag=200, /* Content size will be written into frame header _whenever known_ (default:1) - * Content size must be known at the beginning of compression. - * This is automatically the case when using ZSTD_compress2(), - * For streaming scenarios, content size must be provided with ZSTD_CCtx_setPledgedSrcSize() */ - ZSTD_c_checksumFlag=201, /* A 32-bits checksum of content is written at end of frame (default:0) */ - ZSTD_c_dictIDFlag=202, /* When applicable, dictionary's ID is written into frame header (default:1) */ - - /* multi-threading parameters */ - /* These parameters are only active if multi-threading is enabled (compiled with build macro ZSTD_MULTITHREAD). - * Otherwise, trying to set any other value than default (0) will be a no-op and return an error. - * In a situation where it's unknown if the linked library supports multi-threading or not, - * setting ZSTD_c_nbWorkers to any value >= 1 and consulting the return value provides a quick way to check this property. - */ - ZSTD_c_nbWorkers=400, /* Select how many threads will be spawned to compress in parallel. - * When nbWorkers >= 1, triggers asynchronous mode when invoking ZSTD_compressStream*() : - * ZSTD_compressStream*() consumes input and flush output if possible, but immediately gives back control to caller, - * while compression is performed in parallel, within worker thread(s). - * (note : a strong exception to this rule is when first invocation of ZSTD_compressStream2() sets ZSTD_e_end : - * in which case, ZSTD_compressStream2() delegates to ZSTD_compress2(), which is always a blocking call). - * More workers improve speed, but also increase memory usage. - * Default value is `0`, aka "single-threaded mode" : no worker is spawned, - * compression is performed inside Caller's thread, and all invocations are blocking */ - ZSTD_c_jobSize=401, /* Size of a compression job. This value is enforced only when nbWorkers >= 1. - * Each compression job is completed in parallel, so this value can indirectly impact the nb of active threads. - * 0 means default, which is dynamically determined based on compression parameters. - * Job size must be a minimum of overlap size, or ZSTDMT_JOBSIZE_MIN (= 512 KB), whichever is largest. - * The minimum size is automatically and transparently enforced. */ - ZSTD_c_overlapLog=402, /* Control the overlap size, as a fraction of window size. - * The overlap size is an amount of data reloaded from previous job at the beginning of a new job. - * It helps preserve compression ratio, while each job is compressed in parallel. - * This value is enforced only when nbWorkers >= 1. - * Larger values increase compression ratio, but decrease speed. - * Possible values range from 0 to 9 : - * - 0 means "default" : value will be determined by the library, depending on strategy - * - 1 means "no overlap" - * - 9 means "full overlap", using a full window size. - * Each intermediate rank increases/decreases load size by a factor 2 : - * 9: full window; 8: w/2; 7: w/4; 6: w/8; 5:w/16; 4: w/32; 3:w/64; 2:w/128; 1:no overlap; 0:default - * default value varies between 6 and 9, depending on strategy */ - - /* note : additional experimental parameters are also available - * within the experimental section of the API. - * At the time of this writing, they include : - * ZSTD_c_rsyncable - * ZSTD_c_format - * ZSTD_c_forceMaxWindow - * ZSTD_c_forceAttachDict - * ZSTD_c_literalCompressionMode - * ZSTD_c_targetCBlockSize - * ZSTD_c_srcSizeHint - * ZSTD_c_enableDedicatedDictSearch - * ZSTD_c_stableInBuffer - * ZSTD_c_stableOutBuffer - * ZSTD_c_blockDelimiters - * ZSTD_c_validateSequences - * ZSTD_c_useBlockSplitter - * ZSTD_c_useRowMatchFinder - * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them. - * note : never ever use experimentalParam? names directly; - * also, the enums values themselves are unstable and can still change. - */ - ZSTD_c_experimentalParam1=500, - ZSTD_c_experimentalParam2=10, - ZSTD_c_experimentalParam3=1000, - ZSTD_c_experimentalParam4=1001, - ZSTD_c_experimentalParam5=1002, - ZSTD_c_experimentalParam6=1003, - ZSTD_c_experimentalParam7=1004, - ZSTD_c_experimentalParam8=1005, - ZSTD_c_experimentalParam9=1006, - ZSTD_c_experimentalParam10=1007, - ZSTD_c_experimentalParam11=1008, - ZSTD_c_experimentalParam12=1009, - ZSTD_c_experimentalParam13=1010, - ZSTD_c_experimentalParam14=1011, - ZSTD_c_experimentalParam15=1012 -} ZSTD_cParameter; - -typedef struct { - size_t error; - int lowerBound; - int upperBound; -} ZSTD_bounds; - -/*! ZSTD_cParam_getBounds() : - * All parameters must belong to an interval with lower and upper bounds, - * otherwise they will either trigger an error or be automatically clamped. - * @return : a structure, ZSTD_bounds, which contains - * - an error status field, which must be tested using ZSTD_isError() - * - lower and upper bounds, both inclusive - */ -ZSTDLIB_API ZSTD_bounds ZSTD_cParam_getBounds(ZSTD_cParameter cParam); - -/*! ZSTD_CCtx_setParameter() : - * Set one compression parameter, selected by enum ZSTD_cParameter. - * All parameters have valid bounds. Bounds can be queried using ZSTD_cParam_getBounds(). - * Providing a value beyond bound will either clamp it, or trigger an error (depending on parameter). - * Setting a parameter is generally only possible during frame initialization (before starting compression). - * Exception : when using multi-threading mode (nbWorkers >= 1), - * the following parameters can be updated _during_ compression (within same frame): - * => compressionLevel, hashLog, chainLog, searchLog, minMatch, targetLength and strategy. - * new parameters will be active for next job only (after a flush()). - * @return : an error code (which can be tested using ZSTD_isError()). - */ -ZSTDLIB_API size_t ZSTD_CCtx_setParameter(ZSTD_CCtx* cctx, ZSTD_cParameter param, int value); - -/*! ZSTD_CCtx_setPledgedSrcSize() : - * Total input data size to be compressed as a single frame. - * Value will be written in frame header, unless if explicitly forbidden using ZSTD_c_contentSizeFlag. - * This value will also be controlled at end of frame, and trigger an error if not respected. - * @result : 0, or an error code (which can be tested with ZSTD_isError()). - * Note 1 : pledgedSrcSize==0 actually means zero, aka an empty frame. - * In order to mean "unknown content size", pass constant ZSTD_CONTENTSIZE_UNKNOWN. - * ZSTD_CONTENTSIZE_UNKNOWN is default value for any new frame. - * Note 2 : pledgedSrcSize is only valid once, for the next frame. - * It's discarded at the end of the frame, and replaced by ZSTD_CONTENTSIZE_UNKNOWN. - * Note 3 : Whenever all input data is provided and consumed in a single round, - * for example with ZSTD_compress2(), - * or invoking immediately ZSTD_compressStream2(,,,ZSTD_e_end), - * this value is automatically overridden by srcSize instead. - */ -ZSTDLIB_API size_t ZSTD_CCtx_setPledgedSrcSize(ZSTD_CCtx* cctx, unsigned long long pledgedSrcSize); - -typedef enum { - ZSTD_reset_session_only = 1, - ZSTD_reset_parameters = 2, - ZSTD_reset_session_and_parameters = 3 -} ZSTD_ResetDirective; - -/*! ZSTD_CCtx_reset() : - * There are 2 different things that can be reset, independently or jointly : - * - The session : will stop compressing current frame, and make CCtx ready to start a new one. - * Useful after an error, or to interrupt any ongoing compression. - * Any internal data not yet flushed is cancelled. - * Compression parameters and dictionary remain unchanged. - * They will be used to compress next frame. - * Resetting session never fails. - * - The parameters : changes all parameters back to "default". - * This removes any reference to any dictionary too. - * Parameters can only be changed between 2 sessions (i.e. no compression is currently ongoing) - * otherwise the reset fails, and function returns an error value (which can be tested using ZSTD_isError()) - * - Both : similar to resetting the session, followed by resetting parameters. - */ -ZSTDLIB_API size_t ZSTD_CCtx_reset(ZSTD_CCtx* cctx, ZSTD_ResetDirective reset); - -/*! ZSTD_compress2() : - * Behave the same as ZSTD_compressCCtx(), but compression parameters are set using the advanced API. - * ZSTD_compress2() always starts a new frame. - * Should cctx hold data from a previously unfinished frame, everything about it is forgotten. - * - Compression parameters are pushed into CCtx before starting compression, using ZSTD_CCtx_set*() - * - The function is always blocking, returns when compression is completed. - * Hint : compression runs faster if `dstCapacity` >= `ZSTD_compressBound(srcSize)`. - * @return : compressed size written into `dst` (<= `dstCapacity), - * or an error code if it fails (which can be tested using ZSTD_isError()). - */ -ZSTDLIB_API size_t ZSTD_compress2( ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize); - - -/*********************************************** -* Advanced decompression API (Requires v1.4.0+) -************************************************/ - -/* The advanced API pushes parameters one by one into an existing DCtx context. - * Parameters are sticky, and remain valid for all following frames - * using the same DCtx context. - * It's possible to reset parameters to default values using ZSTD_DCtx_reset(). - * Note : This API is compatible with existing ZSTD_decompressDCtx() and ZSTD_decompressStream(). - * Therefore, no new decompression function is necessary. - */ - -typedef enum { - - ZSTD_d_windowLogMax=100, /* Select a size limit (in power of 2) beyond which - * the streaming API will refuse to allocate memory buffer - * in order to protect the host from unreasonable memory requirements. - * This parameter is only useful in streaming mode, since no internal buffer is allocated in single-pass mode. - * By default, a decompression context accepts window sizes <= (1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT). - * Special: value 0 means "use default maximum windowLog". */ - - /* note : additional experimental parameters are also available - * within the experimental section of the API. - * At the time of this writing, they include : - * ZSTD_d_format - * ZSTD_d_stableOutBuffer - * ZSTD_d_forceIgnoreChecksum - * ZSTD_d_refMultipleDDicts - * Because they are not stable, it's necessary to define ZSTD_STATIC_LINKING_ONLY to access them. - * note : never ever use experimentalParam? names directly - */ - ZSTD_d_experimentalParam1=1000, - ZSTD_d_experimentalParam2=1001, - ZSTD_d_experimentalParam3=1002, - ZSTD_d_experimentalParam4=1003 - -} ZSTD_dParameter; - -/*! ZSTD_dParam_getBounds() : - * All parameters must belong to an interval with lower and upper bounds, - * otherwise they will either trigger an error or be automatically clamped. - * @return : a structure, ZSTD_bounds, which contains - * - an error status field, which must be tested using ZSTD_isError() - * - both lower and upper bounds, inclusive - */ -ZSTDLIB_API ZSTD_bounds ZSTD_dParam_getBounds(ZSTD_dParameter dParam); - -/*! ZSTD_DCtx_setParameter() : - * Set one compression parameter, selected by enum ZSTD_dParameter. - * All parameters have valid bounds. Bounds can be queried using ZSTD_dParam_getBounds(). - * Providing a value beyond bound will either clamp it, or trigger an error (depending on parameter). - * Setting a parameter is only possible during frame initialization (before starting decompression). - * @return : 0, or an error code (which can be tested using ZSTD_isError()). - */ -ZSTDLIB_API size_t ZSTD_DCtx_setParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int value); - -/*! ZSTD_DCtx_reset() : - * Return a DCtx to clean state. - * Session and parameters can be reset jointly or separately. - * Parameters can only be reset when no active frame is being decompressed. - * @return : 0, or an error code, which can be tested with ZSTD_isError() - */ -ZSTDLIB_API size_t ZSTD_DCtx_reset(ZSTD_DCtx* dctx, ZSTD_ResetDirective reset); - - -/**************************** -* Streaming -****************************/ - -typedef struct ZSTD_inBuffer_s { - const void* src; /**< start of input buffer */ - size_t size; /**< size of input buffer */ - size_t pos; /**< position where reading stopped. Will be updated. Necessarily 0 <= pos <= size */ -} ZSTD_inBuffer; - -typedef struct ZSTD_outBuffer_s { - void* dst; /**< start of output buffer */ - size_t size; /**< size of output buffer */ - size_t pos; /**< position where writing stopped. Will be updated. Necessarily 0 <= pos <= size */ -} ZSTD_outBuffer; - - - -/*-*********************************************************************** -* Streaming compression - HowTo -* -* A ZSTD_CStream object is required to track streaming operation. -* Use ZSTD_createCStream() and ZSTD_freeCStream() to create/release resources. -* ZSTD_CStream objects can be reused multiple times on consecutive compression operations. -* It is recommended to re-use ZSTD_CStream since it will play nicer with system's memory, by re-using already allocated memory. -* -* For parallel execution, use one separate ZSTD_CStream per thread. -* -* note : since v1.3.0, ZSTD_CStream and ZSTD_CCtx are the same thing. -* -* Parameters are sticky : when starting a new compression on the same context, -* it will re-use the same sticky parameters as previous compression session. -* When in doubt, it's recommended to fully initialize the context before usage. -* Use ZSTD_CCtx_reset() to reset the context and ZSTD_CCtx_setParameter(), -* ZSTD_CCtx_setPledgedSrcSize(), or ZSTD_CCtx_loadDictionary() and friends to -* set more specific parameters, the pledged source size, or load a dictionary. -* -* Use ZSTD_compressStream2() with ZSTD_e_continue as many times as necessary to -* consume input stream. The function will automatically update both `pos` -* fields within `input` and `output`. -* Note that the function may not consume the entire input, for example, because -* the output buffer is already full, in which case `input.pos < input.size`. -* The caller must check if input has been entirely consumed. -* If not, the caller must make some room to receive more compressed data, -* and then present again remaining input data. -* note: ZSTD_e_continue is guaranteed to make some forward progress when called, -* but doesn't guarantee maximal forward progress. This is especially relevant -* when compressing with multiple threads. The call won't block if it can -* consume some input, but if it can't it will wait for some, but not all, -* output to be flushed. -* @return : provides a minimum amount of data remaining to be flushed from internal buffers -* or an error code, which can be tested using ZSTD_isError(). -* -* At any moment, it's possible to flush whatever data might remain stuck within internal buffer, -* using ZSTD_compressStream2() with ZSTD_e_flush. `output->pos` will be updated. -* Note that, if `output->size` is too small, a single invocation with ZSTD_e_flush might not be enough (return code > 0). -* In which case, make some room to receive more compressed data, and call again ZSTD_compressStream2() with ZSTD_e_flush. -* You must continue calling ZSTD_compressStream2() with ZSTD_e_flush until it returns 0, at which point you can change the -* operation. -* note: ZSTD_e_flush will flush as much output as possible, meaning when compressing with multiple threads, it will -* block until the flush is complete or the output buffer is full. -* @return : 0 if internal buffers are entirely flushed, -* >0 if some data still present within internal buffer (the value is minimal estimation of remaining size), -* or an error code, which can be tested using ZSTD_isError(). -* -* Calling ZSTD_compressStream2() with ZSTD_e_end instructs to finish a frame. -* It will perform a flush and write frame epilogue. -* The epilogue is required for decoders to consider a frame completed. -* flush operation is the same, and follows same rules as calling ZSTD_compressStream2() with ZSTD_e_flush. -* You must continue calling ZSTD_compressStream2() with ZSTD_e_end until it returns 0, at which point you are free to -* start a new frame. -* note: ZSTD_e_end will flush as much output as possible, meaning when compressing with multiple threads, it will -* block until the flush is complete or the output buffer is full. -* @return : 0 if frame fully completed and fully flushed, -* >0 if some data still present within internal buffer (the value is minimal estimation of remaining size), -* or an error code, which can be tested using ZSTD_isError(). -* -* *******************************************************************/ - -typedef ZSTD_CCtx ZSTD_CStream; /**< CCtx and CStream are now effectively same object (>= v1.3.0) */ - /* Continue to distinguish them for compatibility with older versions <= v1.2.0 */ -/*===== ZSTD_CStream management functions =====*/ -ZSTDLIB_API ZSTD_CStream* ZSTD_createCStream(void); -ZSTDLIB_API size_t ZSTD_freeCStream(ZSTD_CStream* zcs); /* accept NULL pointer */ - -/*===== Streaming compression functions =====*/ -typedef enum { - ZSTD_e_continue=0, /* collect more data, encoder decides when to output compressed result, for optimal compression ratio */ - ZSTD_e_flush=1, /* flush any data provided so far, - * it creates (at least) one new block, that can be decoded immediately on reception; - * frame will continue: any future data can still reference previously compressed data, improving compression. - * note : multithreaded compression will block to flush as much output as possible. */ - ZSTD_e_end=2 /* flush any remaining data _and_ close current frame. - * note that frame is only closed after compressed data is fully flushed (return value == 0). - * After that point, any additional data starts a new frame. - * note : each frame is independent (does not reference any content from previous frame). - : note : multithreaded compression will block to flush as much output as possible. */ -} ZSTD_EndDirective; - -/*! ZSTD_compressStream2() : Requires v1.4.0+ - * Behaves about the same as ZSTD_compressStream, with additional control on end directive. - * - Compression parameters are pushed into CCtx before starting compression, using ZSTD_CCtx_set*() - * - Compression parameters cannot be changed once compression is started (save a list of exceptions in multi-threading mode) - * - output->pos must be <= dstCapacity, input->pos must be <= srcSize - * - output->pos and input->pos will be updated. They are guaranteed to remain below their respective limit. - * - endOp must be a valid directive - * - When nbWorkers==0 (default), function is blocking : it completes its job before returning to caller. - * - When nbWorkers>=1, function is non-blocking : it copies a portion of input, distributes jobs to internal worker threads, flush to output whatever is available, - * and then immediately returns, just indicating that there is some data remaining to be flushed. - * The function nonetheless guarantees forward progress : it will return only after it reads or write at least 1+ byte. - * - Exception : if the first call requests a ZSTD_e_end directive and provides enough dstCapacity, the function delegates to ZSTD_compress2() which is always blocking. - * - @return provides a minimum amount of data remaining to be flushed from internal buffers - * or an error code, which can be tested using ZSTD_isError(). - * if @return != 0, flush is not fully completed, there is still some data left within internal buffers. - * This is useful for ZSTD_e_flush, since in this case more flushes are necessary to empty all buffers. - * For ZSTD_e_end, @return == 0 when internal buffers are fully flushed and frame is completed. - * - after a ZSTD_e_end directive, if internal buffer is not fully flushed (@return != 0), - * only ZSTD_e_end or ZSTD_e_flush operations are allowed. - * Before starting a new compression job, or changing compression parameters, - * it is required to fully flush internal buffers. - */ -ZSTDLIB_API size_t ZSTD_compressStream2( ZSTD_CCtx* cctx, - ZSTD_outBuffer* output, - ZSTD_inBuffer* input, - ZSTD_EndDirective endOp); - - -/* These buffer sizes are softly recommended. - * They are not required : ZSTD_compressStream*() happily accepts any buffer size, for both input and output. - * Respecting the recommended size just makes it a bit easier for ZSTD_compressStream*(), - * reducing the amount of memory shuffling and buffering, resulting in minor performance savings. - * - * However, note that these recommendations are from the perspective of a C caller program. - * If the streaming interface is invoked from some other language, - * especially managed ones such as Java or Go, through a foreign function interface such as jni or cgo, - * a major performance rule is to reduce crossing such interface to an absolute minimum. - * It's not rare that performance ends being spent more into the interface, rather than compression itself. - * In which cases, prefer using large buffers, as large as practical, - * for both input and output, to reduce the nb of roundtrips. - */ -ZSTDLIB_API size_t ZSTD_CStreamInSize(void); /**< recommended size for input buffer */ -ZSTDLIB_API size_t ZSTD_CStreamOutSize(void); /**< recommended size for output buffer. Guarantee to successfully flush at least one complete compressed block. */ - - -/* ***************************************************************************** - * This following is a legacy streaming API, available since v1.0+ . - * It can be replaced by ZSTD_CCtx_reset() and ZSTD_compressStream2(). - * It is redundant, but remains fully supported. - * Streaming in combination with advanced parameters and dictionary compression - * can only be used through the new API. - ******************************************************************************/ - -/*! - * Equivalent to: - * - * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); - * ZSTD_CCtx_refCDict(zcs, NULL); // clear the dictionary (if any) - * ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel); - */ -ZSTDLIB_API size_t ZSTD_initCStream(ZSTD_CStream* zcs, int compressionLevel); -/*! - * Alternative for ZSTD_compressStream2(zcs, output, input, ZSTD_e_continue). - * NOTE: The return value is different. ZSTD_compressStream() returns a hint for - * the next read size (if non-zero and not an error). ZSTD_compressStream2() - * returns the minimum nb of bytes left to flush (if non-zero and not an error). - */ -ZSTDLIB_API size_t ZSTD_compressStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output, ZSTD_inBuffer* input); -/*! Equivalent to ZSTD_compressStream2(zcs, output, &emptyInput, ZSTD_e_flush). */ -ZSTDLIB_API size_t ZSTD_flushStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output); -/*! Equivalent to ZSTD_compressStream2(zcs, output, &emptyInput, ZSTD_e_end). */ -ZSTDLIB_API size_t ZSTD_endStream(ZSTD_CStream* zcs, ZSTD_outBuffer* output); - - -/*-*************************************************************************** -* Streaming decompression - HowTo -* -* A ZSTD_DStream object is required to track streaming operations. -* Use ZSTD_createDStream() and ZSTD_freeDStream() to create/release resources. -* ZSTD_DStream objects can be re-used multiple times. -* -* Use ZSTD_initDStream() to start a new decompression operation. -* @return : recommended first input size -* Alternatively, use advanced API to set specific properties. -* -* Use ZSTD_decompressStream() repetitively to consume your input. -* The function will update both `pos` fields. -* If `input.pos < input.size`, some input has not been consumed. -* It's up to the caller to present again remaining data. -* The function tries to flush all data decoded immediately, respecting output buffer size. -* If `output.pos < output.size`, decoder has flushed everything it could. -* But if `output.pos == output.size`, there might be some data left within internal buffers., -* In which case, call ZSTD_decompressStream() again to flush whatever remains in the buffer. -* Note : with no additional input provided, amount of data flushed is necessarily <= ZSTD_BLOCKSIZE_MAX. -* @return : 0 when a frame is completely decoded and fully flushed, -* or an error code, which can be tested using ZSTD_isError(), -* or any other value > 0, which means there is still some decoding or flushing to do to complete current frame : -* the return value is a suggested next input size (just a hint for better latency) -* that will never request more than the remaining frame size. -* *******************************************************************************/ - -typedef ZSTD_DCtx ZSTD_DStream; /**< DCtx and DStream are now effectively same object (>= v1.3.0) */ - /* For compatibility with versions <= v1.2.0, prefer differentiating them. */ -/*===== ZSTD_DStream management functions =====*/ -ZSTDLIB_API ZSTD_DStream* ZSTD_createDStream(void); -ZSTDLIB_API size_t ZSTD_freeDStream(ZSTD_DStream* zds); /* accept NULL pointer */ - -/*===== Streaming decompression functions =====*/ - -/* This function is redundant with the advanced API and equivalent to: - * - * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only); - * ZSTD_DCtx_refDDict(zds, NULL); - */ -ZSTDLIB_API size_t ZSTD_initDStream(ZSTD_DStream* zds); - -ZSTDLIB_API size_t ZSTD_decompressStream(ZSTD_DStream* zds, ZSTD_outBuffer* output, ZSTD_inBuffer* input); - -ZSTDLIB_API size_t ZSTD_DStreamInSize(void); /*!< recommended size for input buffer */ -ZSTDLIB_API size_t ZSTD_DStreamOutSize(void); /*!< recommended size for output buffer. Guarantee to successfully flush at least one complete block in all circumstances. */ - - -/************************** -* Simple dictionary API -***************************/ -/*! ZSTD_compress_usingDict() : - * Compression at an explicit compression level using a Dictionary. - * A dictionary can be any arbitrary data segment (also called a prefix), - * or a buffer with specified information (see zdict.h). - * Note : This function loads the dictionary, resulting in significant startup delay. - * It's intended for a dictionary used only once. - * Note 2 : When `dict == NULL || dictSize < 8` no dictionary is used. */ -ZSTDLIB_API size_t ZSTD_compress_usingDict(ZSTD_CCtx* ctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const void* dict,size_t dictSize, - int compressionLevel); - -/*! ZSTD_decompress_usingDict() : - * Decompression using a known Dictionary. - * Dictionary must be identical to the one used during compression. - * Note : This function loads the dictionary, resulting in significant startup delay. - * It's intended for a dictionary used only once. - * Note : When `dict == NULL || dictSize < 8` no dictionary is used. */ -ZSTDLIB_API size_t ZSTD_decompress_usingDict(ZSTD_DCtx* dctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const void* dict,size_t dictSize); - - -/*********************************** - * Bulk processing dictionary API - **********************************/ -typedef struct ZSTD_CDict_s ZSTD_CDict; - -/*! ZSTD_createCDict() : - * When compressing multiple messages or blocks using the same dictionary, - * it's recommended to digest the dictionary only once, since it's a costly operation. - * ZSTD_createCDict() will create a state from digesting a dictionary. - * The resulting state can be used for future compression operations with very limited startup cost. - * ZSTD_CDict can be created once and shared by multiple threads concurrently, since its usage is read-only. - * @dictBuffer can be released after ZSTD_CDict creation, because its content is copied within CDict. - * Note 1 : Consider experimental function `ZSTD_createCDict_byReference()` if you prefer to not duplicate @dictBuffer content. - * Note 2 : A ZSTD_CDict can be created from an empty @dictBuffer, - * in which case the only thing that it transports is the @compressionLevel. - * This can be useful in a pipeline featuring ZSTD_compress_usingCDict() exclusively, - * expecting a ZSTD_CDict parameter with any data, including those without a known dictionary. */ -ZSTDLIB_API ZSTD_CDict* ZSTD_createCDict(const void* dictBuffer, size_t dictSize, - int compressionLevel); - -/*! ZSTD_freeCDict() : - * Function frees memory allocated by ZSTD_createCDict(). - * If a NULL pointer is passed, no operation is performed. */ -ZSTDLIB_API size_t ZSTD_freeCDict(ZSTD_CDict* CDict); - -/*! ZSTD_compress_usingCDict() : - * Compression using a digested Dictionary. - * Recommended when same dictionary is used multiple times. - * Note : compression level is _decided at dictionary creation time_, - * and frame parameters are hardcoded (dictID=yes, contentSize=yes, checksum=no) */ -ZSTDLIB_API size_t ZSTD_compress_usingCDict(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const ZSTD_CDict* cdict); - - -typedef struct ZSTD_DDict_s ZSTD_DDict; - -/*! ZSTD_createDDict() : - * Create a digested dictionary, ready to start decompression operation without startup delay. - * dictBuffer can be released after DDict creation, as its content is copied inside DDict. */ -ZSTDLIB_API ZSTD_DDict* ZSTD_createDDict(const void* dictBuffer, size_t dictSize); - -/*! ZSTD_freeDDict() : - * Function frees memory allocated with ZSTD_createDDict() - * If a NULL pointer is passed, no operation is performed. */ -ZSTDLIB_API size_t ZSTD_freeDDict(ZSTD_DDict* ddict); - -/*! ZSTD_decompress_usingDDict() : - * Decompression using a digested Dictionary. - * Recommended when same dictionary is used multiple times. */ -ZSTDLIB_API size_t ZSTD_decompress_usingDDict(ZSTD_DCtx* dctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const ZSTD_DDict* ddict); - - -/******************************** - * Dictionary helper functions - *******************************/ - -/*! ZSTD_getDictID_fromDict() : Requires v1.4.0+ - * Provides the dictID stored within dictionary. - * if @return == 0, the dictionary is not conformant with Zstandard specification. - * It can still be loaded, but as a content-only dictionary. */ -ZSTDLIB_API unsigned ZSTD_getDictID_fromDict(const void* dict, size_t dictSize); - -/*! ZSTD_getDictID_fromCDict() : Requires v1.5.0+ - * Provides the dictID of the dictionary loaded into `cdict`. - * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty. - * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */ -ZSTDLIB_API unsigned ZSTD_getDictID_fromCDict(const ZSTD_CDict* cdict); - -/*! ZSTD_getDictID_fromDDict() : Requires v1.4.0+ - * Provides the dictID of the dictionary loaded into `ddict`. - * If @return == 0, the dictionary is not conformant to Zstandard specification, or empty. - * Non-conformant dictionaries can still be loaded, but as content-only dictionaries. */ -ZSTDLIB_API unsigned ZSTD_getDictID_fromDDict(const ZSTD_DDict* ddict); - -/*! ZSTD_getDictID_fromFrame() : Requires v1.4.0+ - * Provides the dictID required to decompressed the frame stored within `src`. - * If @return == 0, the dictID could not be decoded. - * This could for one of the following reasons : - * - The frame does not require a dictionary to be decoded (most common case). - * - The frame was built with dictID intentionally removed. Whatever dictionary is necessary is a hidden information. - * Note : this use case also happens when using a non-conformant dictionary. - * - `srcSize` is too small, and as a result, the frame header could not be decoded (only possible if `srcSize < ZSTD_FRAMEHEADERSIZE_MAX`). - * - This is not a Zstandard frame. - * When identifying the exact failure cause, it's possible to use ZSTD_getFrameHeader(), which will provide a more precise error code. */ -ZSTDLIB_API unsigned ZSTD_getDictID_fromFrame(const void* src, size_t srcSize); - - -/******************************************************************************* - * Advanced dictionary and prefix API (Requires v1.4.0+) - * - * This API allows dictionaries to be used with ZSTD_compress2(), - * ZSTD_compressStream2(), and ZSTD_decompressDCtx(). Dictionaries are sticky, and - * only reset with the context is reset with ZSTD_reset_parameters or - * ZSTD_reset_session_and_parameters. Prefixes are single-use. - ******************************************************************************/ - - -/*! ZSTD_CCtx_loadDictionary() : Requires v1.4.0+ - * Create an internal CDict from `dict` buffer. - * Decompression will have to use same dictionary. - * @result : 0, or an error code (which can be tested with ZSTD_isError()). - * Special: Loading a NULL (or 0-size) dictionary invalidates previous dictionary, - * meaning "return to no-dictionary mode". - * Note 1 : Dictionary is sticky, it will be used for all future compressed frames. - * To return to "no-dictionary" situation, load a NULL dictionary (or reset parameters). - * Note 2 : Loading a dictionary involves building tables. - * It's also a CPU consuming operation, with non-negligible impact on latency. - * Tables are dependent on compression parameters, and for this reason, - * compression parameters can no longer be changed after loading a dictionary. - * Note 3 :`dict` content will be copied internally. - * Use experimental ZSTD_CCtx_loadDictionary_byReference() to reference content instead. - * In such a case, dictionary buffer must outlive its users. - * Note 4 : Use ZSTD_CCtx_loadDictionary_advanced() - * to precisely select how dictionary content must be interpreted. */ -ZSTDLIB_API size_t ZSTD_CCtx_loadDictionary(ZSTD_CCtx* cctx, const void* dict, size_t dictSize); - -/*! ZSTD_CCtx_refCDict() : Requires v1.4.0+ - * Reference a prepared dictionary, to be used for all next compressed frames. - * Note that compression parameters are enforced from within CDict, - * and supersede any compression parameter previously set within CCtx. - * The parameters ignored are labelled as "superseded-by-cdict" in the ZSTD_cParameter enum docs. - * The ignored parameters will be used again if the CCtx is returned to no-dictionary mode. - * The dictionary will remain valid for future compressed frames using same CCtx. - * @result : 0, or an error code (which can be tested with ZSTD_isError()). - * Special : Referencing a NULL CDict means "return to no-dictionary mode". - * Note 1 : Currently, only one dictionary can be managed. - * Referencing a new dictionary effectively "discards" any previous one. - * Note 2 : CDict is just referenced, its lifetime must outlive its usage within CCtx. */ -ZSTDLIB_API size_t ZSTD_CCtx_refCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict); - -/*! ZSTD_CCtx_refPrefix() : Requires v1.4.0+ - * Reference a prefix (single-usage dictionary) for next compressed frame. - * A prefix is **only used once**. Tables are discarded at end of frame (ZSTD_e_end). - * Decompression will need same prefix to properly regenerate data. - * Compressing with a prefix is similar in outcome as performing a diff and compressing it, - * but performs much faster, especially during decompression (compression speed is tunable with compression level). - * @result : 0, or an error code (which can be tested with ZSTD_isError()). - * Special: Adding any prefix (including NULL) invalidates any previous prefix or dictionary - * Note 1 : Prefix buffer is referenced. It **must** outlive compression. - * Its content must remain unmodified during compression. - * Note 2 : If the intention is to diff some large src data blob with some prior version of itself, - * ensure that the window size is large enough to contain the entire source. - * See ZSTD_c_windowLog. - * Note 3 : Referencing a prefix involves building tables, which are dependent on compression parameters. - * It's a CPU consuming operation, with non-negligible impact on latency. - * If there is a need to use the same prefix multiple times, consider loadDictionary instead. - * Note 4 : By default, the prefix is interpreted as raw content (ZSTD_dct_rawContent). - * Use experimental ZSTD_CCtx_refPrefix_advanced() to alter dictionary interpretation. */ -ZSTDLIB_API size_t ZSTD_CCtx_refPrefix(ZSTD_CCtx* cctx, - const void* prefix, size_t prefixSize); - -/*! ZSTD_DCtx_loadDictionary() : Requires v1.4.0+ - * Create an internal DDict from dict buffer, - * to be used to decompress next frames. - * The dictionary remains valid for all future frames, until explicitly invalidated. - * @result : 0, or an error code (which can be tested with ZSTD_isError()). - * Special : Adding a NULL (or 0-size) dictionary invalidates any previous dictionary, - * meaning "return to no-dictionary mode". - * Note 1 : Loading a dictionary involves building tables, - * which has a non-negligible impact on CPU usage and latency. - * It's recommended to "load once, use many times", to amortize the cost - * Note 2 :`dict` content will be copied internally, so `dict` can be released after loading. - * Use ZSTD_DCtx_loadDictionary_byReference() to reference dictionary content instead. - * Note 3 : Use ZSTD_DCtx_loadDictionary_advanced() to take control of - * how dictionary content is loaded and interpreted. - */ -ZSTDLIB_API size_t ZSTD_DCtx_loadDictionary(ZSTD_DCtx* dctx, const void* dict, size_t dictSize); - -/*! ZSTD_DCtx_refDDict() : Requires v1.4.0+ - * Reference a prepared dictionary, to be used to decompress next frames. - * The dictionary remains active for decompression of future frames using same DCtx. - * - * If called with ZSTD_d_refMultipleDDicts enabled, repeated calls of this function - * will store the DDict references in a table, and the DDict used for decompression - * will be determined at decompression time, as per the dict ID in the frame. - * The memory for the table is allocated on the first call to refDDict, and can be - * freed with ZSTD_freeDCtx(). - * - * @result : 0, or an error code (which can be tested with ZSTD_isError()). - * Note 1 : Currently, only one dictionary can be managed. - * Referencing a new dictionary effectively "discards" any previous one. - * Special: referencing a NULL DDict means "return to no-dictionary mode". - * Note 2 : DDict is just referenced, its lifetime must outlive its usage from DCtx. - */ -ZSTDLIB_API size_t ZSTD_DCtx_refDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict); - -/*! ZSTD_DCtx_refPrefix() : Requires v1.4.0+ - * Reference a prefix (single-usage dictionary) to decompress next frame. - * This is the reverse operation of ZSTD_CCtx_refPrefix(), - * and must use the same prefix as the one used during compression. - * Prefix is **only used once**. Reference is discarded at end of frame. - * End of frame is reached when ZSTD_decompressStream() returns 0. - * @result : 0, or an error code (which can be tested with ZSTD_isError()). - * Note 1 : Adding any prefix (including NULL) invalidates any previously set prefix or dictionary - * Note 2 : Prefix buffer is referenced. It **must** outlive decompression. - * Prefix buffer must remain unmodified up to the end of frame, - * reached when ZSTD_decompressStream() returns 0. - * Note 3 : By default, the prefix is treated as raw content (ZSTD_dct_rawContent). - * Use ZSTD_CCtx_refPrefix_advanced() to alter dictMode (Experimental section) - * Note 4 : Referencing a raw content prefix has almost no cpu nor memory cost. - * A full dictionary is more costly, as it requires building tables. - */ -ZSTDLIB_API size_t ZSTD_DCtx_refPrefix(ZSTD_DCtx* dctx, - const void* prefix, size_t prefixSize); - -/* === Memory management === */ - -/*! ZSTD_sizeof_*() : Requires v1.4.0+ - * These functions give the _current_ memory usage of selected object. - * Note that object memory usage can evolve (increase or decrease) over time. */ -ZSTDLIB_API size_t ZSTD_sizeof_CCtx(const ZSTD_CCtx* cctx); -ZSTDLIB_API size_t ZSTD_sizeof_DCtx(const ZSTD_DCtx* dctx); -ZSTDLIB_API size_t ZSTD_sizeof_CStream(const ZSTD_CStream* zcs); -ZSTDLIB_API size_t ZSTD_sizeof_DStream(const ZSTD_DStream* zds); -ZSTDLIB_API size_t ZSTD_sizeof_CDict(const ZSTD_CDict* cdict); -ZSTDLIB_API size_t ZSTD_sizeof_DDict(const ZSTD_DDict* ddict); - -#endif /* ZSTD_H_235446 */ - - -/* ************************************************************************************** - * ADVANCED AND EXPERIMENTAL FUNCTIONS - **************************************************************************************** - * The definitions in the following section are considered experimental. - * They are provided for advanced scenarios. - * They should never be used with a dynamic library, as prototypes may change in the future. - * Use them only in association with static linking. - * ***************************************************************************************/ - -#if defined(ZSTD_STATIC_LINKING_ONLY) && !defined(ZSTD_H_ZSTD_STATIC_LINKING_ONLY) -#define ZSTD_H_ZSTD_STATIC_LINKING_ONLY - -/* This can be overridden externally to hide static symbols. */ -#ifndef ZSTDLIB_STATIC_API -# if defined(ZSTD_DLL_EXPORT) && (ZSTD_DLL_EXPORT==1) -# define ZSTDLIB_STATIC_API __declspec(dllexport) ZSTDLIB_VISIBLE -# elif defined(ZSTD_DLL_IMPORT) && (ZSTD_DLL_IMPORT==1) -# define ZSTDLIB_STATIC_API __declspec(dllimport) ZSTDLIB_VISIBLE -# else -# define ZSTDLIB_STATIC_API ZSTDLIB_VISIBLE -# endif -#endif - -/* Deprecation warnings : - * Should these warnings be a problem, it is generally possible to disable them, - * typically with -Wno-deprecated-declarations for gcc or _CRT_SECURE_NO_WARNINGS in Visual. - * Otherwise, it's also possible to define ZSTD_DISABLE_DEPRECATE_WARNINGS. - */ -#ifdef ZSTD_DISABLE_DEPRECATE_WARNINGS -# define ZSTD_DEPRECATED(message) ZSTDLIB_STATIC_API /* disable deprecation warnings */ -#else -# if defined (__cplusplus) && (__cplusplus >= 201402) /* C++14 or greater */ -# define ZSTD_DEPRECATED(message) [[deprecated(message)]] ZSTDLIB_STATIC_API -# elif (defined(GNUC) && (GNUC > 4 || (GNUC == 4 && GNUC_MINOR >= 5))) || defined(__clang__) -# define ZSTD_DEPRECATED(message) ZSTDLIB_STATIC_API __attribute__((deprecated(message))) -# elif defined(__GNUC__) && (__GNUC__ >= 3) -# define ZSTD_DEPRECATED(message) ZSTDLIB_STATIC_API __attribute__((deprecated)) -# elif defined(_MSC_VER) -# define ZSTD_DEPRECATED(message) ZSTDLIB_STATIC_API __declspec(deprecated(message)) -# else -# pragma message("WARNING: You need to implement ZSTD_DEPRECATED for this compiler") -# define ZSTD_DEPRECATED(message) ZSTDLIB_STATIC_API -# endif -#endif /* ZSTD_DISABLE_DEPRECATE_WARNINGS */ - -/**************************************************************************************** - * experimental API (static linking only) - **************************************************************************************** - * The following symbols and constants - * are not planned to join "stable API" status in the near future. - * They can still change in future versions. - * Some of them are planned to remain in the static_only section indefinitely. - * Some of them might be removed in the future (especially when redundant with existing stable functions) - * ***************************************************************************************/ - -#define ZSTD_FRAMEHEADERSIZE_PREFIX(format) ((format) == ZSTD_f_zstd1 ? 5 : 1) /* minimum input size required to query frame header size */ -#define ZSTD_FRAMEHEADERSIZE_MIN(format) ((format) == ZSTD_f_zstd1 ? 6 : 2) -#define ZSTD_FRAMEHEADERSIZE_MAX 18 /* can be useful for static allocation */ -#define ZSTD_SKIPPABLEHEADERSIZE 8 - -/* compression parameter bounds */ -#define ZSTD_WINDOWLOG_MAX_32 30 -#define ZSTD_WINDOWLOG_MAX_64 31 -#define ZSTD_WINDOWLOG_MAX ((int)(sizeof(size_t) == 4 ? ZSTD_WINDOWLOG_MAX_32 : ZSTD_WINDOWLOG_MAX_64)) -#define ZSTD_WINDOWLOG_MIN 10 -#define ZSTD_HASHLOG_MAX ((ZSTD_WINDOWLOG_MAX < 30) ? ZSTD_WINDOWLOG_MAX : 30) -#define ZSTD_HASHLOG_MIN 6 -#define ZSTD_CHAINLOG_MAX_32 29 -#define ZSTD_CHAINLOG_MAX_64 30 -#define ZSTD_CHAINLOG_MAX ((int)(sizeof(size_t) == 4 ? ZSTD_CHAINLOG_MAX_32 : ZSTD_CHAINLOG_MAX_64)) -#define ZSTD_CHAINLOG_MIN ZSTD_HASHLOG_MIN -#define ZSTD_SEARCHLOG_MAX (ZSTD_WINDOWLOG_MAX-1) -#define ZSTD_SEARCHLOG_MIN 1 -#define ZSTD_MINMATCH_MAX 7 /* only for ZSTD_fast, other strategies are limited to 6 */ -#define ZSTD_MINMATCH_MIN 3 /* only for ZSTD_btopt+, faster strategies are limited to 4 */ -#define ZSTD_TARGETLENGTH_MAX ZSTD_BLOCKSIZE_MAX -#define ZSTD_TARGETLENGTH_MIN 0 /* note : comparing this constant to an unsigned results in a tautological test */ -#define ZSTD_STRATEGY_MIN ZSTD_fast -#define ZSTD_STRATEGY_MAX ZSTD_btultra2 - - -#define ZSTD_OVERLAPLOG_MIN 0 -#define ZSTD_OVERLAPLOG_MAX 9 - -#define ZSTD_WINDOWLOG_LIMIT_DEFAULT 27 /* by default, the streaming decoder will refuse any frame - * requiring larger than (1< 0: - * If litLength != 0: - * rep == 1 --> offset == repeat_offset_1 - * rep == 2 --> offset == repeat_offset_2 - * rep == 3 --> offset == repeat_offset_3 - * If litLength == 0: - * rep == 1 --> offset == repeat_offset_2 - * rep == 2 --> offset == repeat_offset_3 - * rep == 3 --> offset == repeat_offset_1 - 1 - * - * Note: This field is optional. ZSTD_generateSequences() will calculate the value of - * 'rep', but repeat offsets do not necessarily need to be calculated from an external - * sequence provider's perspective. For example, ZSTD_compressSequences() does not - * use this 'rep' field at all (as of now). - */ -} ZSTD_Sequence; - -typedef struct { - unsigned windowLog; /**< largest match distance : larger == more compression, more memory needed during decompression */ - unsigned chainLog; /**< fully searched segment : larger == more compression, slower, more memory (useless for fast) */ - unsigned hashLog; /**< dispatch table : larger == faster, more memory */ - unsigned searchLog; /**< nb of searches : larger == more compression, slower */ - unsigned minMatch; /**< match length searched : larger == faster decompression, sometimes less compression */ - unsigned targetLength; /**< acceptable match size for optimal parser (only) : larger == more compression, slower */ - ZSTD_strategy strategy; /**< see ZSTD_strategy definition above */ -} ZSTD_compressionParameters; - -typedef struct { - int contentSizeFlag; /**< 1: content size will be in frame header (when known) */ - int checksumFlag; /**< 1: generate a 32-bits checksum using XXH64 algorithm at end of frame, for error detection */ - int noDictIDFlag; /**< 1: no dictID will be saved into frame header (dictID is only useful for dictionary compression) */ -} ZSTD_frameParameters; - -typedef struct { - ZSTD_compressionParameters cParams; - ZSTD_frameParameters fParams; -} ZSTD_parameters; - -typedef enum { - ZSTD_dct_auto = 0, /* dictionary is "full" when starting with ZSTD_MAGIC_DICTIONARY, otherwise it is "rawContent" */ - ZSTD_dct_rawContent = 1, /* ensures dictionary is always loaded as rawContent, even if it starts with ZSTD_MAGIC_DICTIONARY */ - ZSTD_dct_fullDict = 2 /* refuses to load a dictionary if it does not respect Zstandard's specification, starting with ZSTD_MAGIC_DICTIONARY */ -} ZSTD_dictContentType_e; - -typedef enum { - ZSTD_dlm_byCopy = 0, /**< Copy dictionary content internally */ - ZSTD_dlm_byRef = 1 /**< Reference dictionary content -- the dictionary buffer must outlive its users. */ -} ZSTD_dictLoadMethod_e; - -typedef enum { - ZSTD_f_zstd1 = 0, /* zstd frame format, specified in zstd_compression_format.md (default) */ - ZSTD_f_zstd1_magicless = 1 /* Variant of zstd frame format, without initial 4-bytes magic number. - * Useful to save 4 bytes per generated frame. - * Decoder cannot recognise automatically this format, requiring this instruction. */ -} ZSTD_format_e; - -typedef enum { - /* Note: this enum controls ZSTD_d_forceIgnoreChecksum */ - ZSTD_d_validateChecksum = 0, - ZSTD_d_ignoreChecksum = 1 -} ZSTD_forceIgnoreChecksum_e; - -typedef enum { - /* Note: this enum controls ZSTD_d_refMultipleDDicts */ - ZSTD_rmd_refSingleDDict = 0, - ZSTD_rmd_refMultipleDDicts = 1 -} ZSTD_refMultipleDDicts_e; - -typedef enum { - /* Note: this enum and the behavior it controls are effectively internal - * implementation details of the compressor. They are expected to continue - * to evolve and should be considered only in the context of extremely - * advanced performance tuning. - * - * Zstd currently supports the use of a CDict in three ways: - * - * - The contents of the CDict can be copied into the working context. This - * means that the compression can search both the dictionary and input - * while operating on a single set of internal tables. This makes - * the compression faster per-byte of input. However, the initial copy of - * the CDict's tables incurs a fixed cost at the beginning of the - * compression. For small compressions (< 8 KB), that copy can dominate - * the cost of the compression. - * - * - The CDict's tables can be used in-place. In this model, compression is - * slower per input byte, because the compressor has to search two sets of - * tables. However, this model incurs no start-up cost (as long as the - * working context's tables can be reused). For small inputs, this can be - * faster than copying the CDict's tables. - * - * - The CDict's tables are not used at all, and instead we use the working - * context alone to reload the dictionary and use params based on the source - * size. See ZSTD_compress_insertDictionary() and ZSTD_compress_usingDict(). - * This method is effective when the dictionary sizes are very small relative - * to the input size, and the input size is fairly large to begin with. - * - * Zstd has a simple internal heuristic that selects which strategy to use - * at the beginning of a compression. However, if experimentation shows that - * Zstd is making poor choices, it is possible to override that choice with - * this enum. - */ - ZSTD_dictDefaultAttach = 0, /* Use the default heuristic. */ - ZSTD_dictForceAttach = 1, /* Never copy the dictionary. */ - ZSTD_dictForceCopy = 2, /* Always copy the dictionary. */ - ZSTD_dictForceLoad = 3 /* Always reload the dictionary */ -} ZSTD_dictAttachPref_e; - -typedef enum { - ZSTD_lcm_auto = 0, /**< Automatically determine the compression mode based on the compression level. - * Negative compression levels will be uncompressed, and positive compression - * levels will be compressed. */ - ZSTD_lcm_huffman = 1, /**< Always attempt Huffman compression. Uncompressed literals will still be - * emitted if Huffman compression is not profitable. */ - ZSTD_lcm_uncompressed = 2 /**< Always emit uncompressed literals. */ -} ZSTD_literalCompressionMode_e; - -typedef enum { - /* Note: This enum controls features which are conditionally beneficial. Zstd typically will make a final - * decision on whether or not to enable the feature (ZSTD_ps_auto), but setting the switch to ZSTD_ps_enable - * or ZSTD_ps_disable allow for a force enable/disable the feature. - */ - ZSTD_ps_auto = 0, /* Let the library automatically determine whether the feature shall be enabled */ - ZSTD_ps_enable = 1, /* Force-enable the feature */ - ZSTD_ps_disable = 2 /* Do not use the feature */ -} ZSTD_paramSwitch_e; - -/*************************************** -* Frame size functions -***************************************/ - -/*! ZSTD_findDecompressedSize() : - * `src` should point to the start of a series of ZSTD encoded and/or skippable frames - * `srcSize` must be the _exact_ size of this series - * (i.e. there should be a frame boundary at `src + srcSize`) - * @return : - decompressed size of all data in all successive frames - * - if the decompressed size cannot be determined: ZSTD_CONTENTSIZE_UNKNOWN - * - if an error occurred: ZSTD_CONTENTSIZE_ERROR - * - * note 1 : decompressed size is an optional field, that may not be present, especially in streaming mode. - * When `return==ZSTD_CONTENTSIZE_UNKNOWN`, data to decompress could be any size. - * In which case, it's necessary to use streaming mode to decompress data. - * note 2 : decompressed size is always present when compression is done with ZSTD_compress() - * note 3 : decompressed size can be very large (64-bits value), - * potentially larger than what local system can handle as a single memory segment. - * In which case, it's necessary to use streaming mode to decompress data. - * note 4 : If source is untrusted, decompressed size could be wrong or intentionally modified. - * Always ensure result fits within application's authorized limits. - * Each application can set its own limits. - * note 5 : ZSTD_findDecompressedSize handles multiple frames, and so it must traverse the input to - * read each contained frame header. This is fast as most of the data is skipped, - * however it does mean that all frame data must be present and valid. */ -ZSTDLIB_STATIC_API unsigned long long ZSTD_findDecompressedSize(const void* src, size_t srcSize); - -/*! ZSTD_decompressBound() : - * `src` should point to the start of a series of ZSTD encoded and/or skippable frames - * `srcSize` must be the _exact_ size of this series - * (i.e. there should be a frame boundary at `src + srcSize`) - * @return : - upper-bound for the decompressed size of all data in all successive frames - * - if an error occurred: ZSTD_CONTENTSIZE_ERROR - * - * note 1 : an error can occur if `src` contains an invalid or incorrectly formatted frame. - * note 2 : the upper-bound is exact when the decompressed size field is available in every ZSTD encoded frame of `src`. - * in this case, `ZSTD_findDecompressedSize` and `ZSTD_decompressBound` return the same value. - * note 3 : when the decompressed size field isn't available, the upper-bound for that frame is calculated by: - * upper-bound = # blocks * min(128 KB, Window_Size) - */ -ZSTDLIB_STATIC_API unsigned long long ZSTD_decompressBound(const void* src, size_t srcSize); - -/*! ZSTD_frameHeaderSize() : - * srcSize must be >= ZSTD_FRAMEHEADERSIZE_PREFIX. - * @return : size of the Frame Header, - * or an error code (if srcSize is too small) */ -ZSTDLIB_STATIC_API size_t ZSTD_frameHeaderSize(const void* src, size_t srcSize); - -typedef enum { - ZSTD_sf_noBlockDelimiters = 0, /* Representation of ZSTD_Sequence has no block delimiters, sequences only */ - ZSTD_sf_explicitBlockDelimiters = 1 /* Representation of ZSTD_Sequence contains explicit block delimiters */ -} ZSTD_sequenceFormat_e; - -/*! ZSTD_generateSequences() : - * Generate sequences using ZSTD_compress2, given a source buffer. - * - * Each block will end with a dummy sequence - * with offset == 0, matchLength == 0, and litLength == length of last literals. - * litLength may be == 0, and if so, then the sequence of (of: 0 ml: 0 ll: 0) - * simply acts as a block delimiter. - * - * zc can be used to insert custom compression params. - * This function invokes ZSTD_compress2 - * - * The output of this function can be fed into ZSTD_compressSequences() with CCtx - * setting of ZSTD_c_blockDelimiters as ZSTD_sf_explicitBlockDelimiters - * @return : number of sequences generated - */ - -ZSTDLIB_STATIC_API size_t ZSTD_generateSequences(ZSTD_CCtx* zc, ZSTD_Sequence* outSeqs, - size_t outSeqsSize, const void* src, size_t srcSize); - -/*! ZSTD_mergeBlockDelimiters() : - * Given an array of ZSTD_Sequence, remove all sequences that represent block delimiters/last literals - * by merging them into into the literals of the next sequence. - * - * As such, the final generated result has no explicit representation of block boundaries, - * and the final last literals segment is not represented in the sequences. - * - * The output of this function can be fed into ZSTD_compressSequences() with CCtx - * setting of ZSTD_c_blockDelimiters as ZSTD_sf_noBlockDelimiters - * @return : number of sequences left after merging - */ -ZSTDLIB_STATIC_API size_t ZSTD_mergeBlockDelimiters(ZSTD_Sequence* sequences, size_t seqsSize); - -/*! ZSTD_compressSequences() : - * Compress an array of ZSTD_Sequence, generated from the original source buffer, into dst. - * If a dictionary is included, then the cctx should reference the dict. (see: ZSTD_CCtx_refCDict(), ZSTD_CCtx_loadDictionary(), etc.) - * The entire source is compressed into a single frame. - * - * The compression behavior changes based on cctx params. In particular: - * If ZSTD_c_blockDelimiters == ZSTD_sf_noBlockDelimiters, the array of ZSTD_Sequence is expected to contain - * no block delimiters (defined in ZSTD_Sequence). Block boundaries are roughly determined based on - * the block size derived from the cctx, and sequences may be split. This is the default setting. - * - * If ZSTD_c_blockDelimiters == ZSTD_sf_explicitBlockDelimiters, the array of ZSTD_Sequence is expected to contain - * block delimiters (defined in ZSTD_Sequence). Behavior is undefined if no block delimiters are provided. - * - * If ZSTD_c_validateSequences == 0, this function will blindly accept the sequences provided. Invalid sequences cause undefined - * behavior. If ZSTD_c_validateSequences == 1, then if sequence is invalid (see doc/zstd_compression_format.md for - * specifics regarding offset/matchlength requirements) then the function will bail out and return an error. - * - * In addition to the two adjustable experimental params, there are other important cctx params. - * - ZSTD_c_minMatch MUST be set as less than or equal to the smallest match generated by the match finder. It has a minimum value of ZSTD_MINMATCH_MIN. - * - ZSTD_c_compressionLevel accordingly adjusts the strength of the entropy coder, as it would in typical compression. - * - ZSTD_c_windowLog affects offset validation: this function will return an error at higher debug levels if a provided offset - * is larger than what the spec allows for a given window log and dictionary (if present). See: doc/zstd_compression_format.md - * - * Note: Repcodes are, as of now, always re-calculated within this function, so ZSTD_Sequence::rep is unused. - * Note 2: Once we integrate ability to ingest repcodes, the explicit block delims mode must respect those repcodes exactly, - * and cannot emit an RLE block that disagrees with the repcode history - * @return : final compressed size or a ZSTD error. - */ -ZSTDLIB_STATIC_API size_t ZSTD_compressSequences(ZSTD_CCtx* const cctx, void* dst, size_t dstSize, - const ZSTD_Sequence* inSeqs, size_t inSeqsSize, - const void* src, size_t srcSize); - - -/*! ZSTD_writeSkippableFrame() : - * Generates a zstd skippable frame containing data given by src, and writes it to dst buffer. - * - * Skippable frames begin with a a 4-byte magic number. There are 16 possible choices of magic number, - * ranging from ZSTD_MAGIC_SKIPPABLE_START to ZSTD_MAGIC_SKIPPABLE_START+15. - * As such, the parameter magicVariant controls the exact skippable frame magic number variant used, so - * the magic number used will be ZSTD_MAGIC_SKIPPABLE_START + magicVariant. - * - * Returns an error if destination buffer is not large enough, if the source size is not representable - * with a 4-byte unsigned int, or if the parameter magicVariant is greater than 15 (and therefore invalid). - * - * @return : number of bytes written or a ZSTD error. - */ -ZSTDLIB_STATIC_API size_t ZSTD_writeSkippableFrame(void* dst, size_t dstCapacity, - const void* src, size_t srcSize, unsigned magicVariant); - -/*! ZSTD_readSkippableFrame() : - * Retrieves a zstd skippable frame containing data given by src, and writes it to dst buffer. - * - * The parameter magicVariant will receive the magicVariant that was supplied when the frame was written, - * i.e. magicNumber - ZSTD_MAGIC_SKIPPABLE_START. This can be NULL if the caller is not interested - * in the magicVariant. - * - * Returns an error if destination buffer is not large enough, or if the frame is not skippable. - * - * @return : number of bytes written or a ZSTD error. - */ -ZSTDLIB_API size_t ZSTD_readSkippableFrame(void* dst, size_t dstCapacity, unsigned* magicVariant, - const void* src, size_t srcSize); - -/*! ZSTD_isSkippableFrame() : - * Tells if the content of `buffer` starts with a valid Frame Identifier for a skippable frame. - */ -ZSTDLIB_API unsigned ZSTD_isSkippableFrame(const void* buffer, size_t size); - - - -/*************************************** -* Memory management -***************************************/ - -/*! ZSTD_estimate*() : - * These functions make it possible to estimate memory usage - * of a future {D,C}Ctx, before its creation. - * - * ZSTD_estimateCCtxSize() will provide a memory budget large enough - * for any compression level up to selected one. - * Note : Unlike ZSTD_estimateCStreamSize*(), this estimate - * does not include space for a window buffer. - * Therefore, the estimation is only guaranteed for single-shot compressions, not streaming. - * The estimate will assume the input may be arbitrarily large, - * which is the worst case. - * - * When srcSize can be bound by a known and rather "small" value, - * this fact can be used to provide a tighter estimation - * because the CCtx compression context will need less memory. - * This tighter estimation can be provided by more advanced functions - * ZSTD_estimateCCtxSize_usingCParams(), which can be used in tandem with ZSTD_getCParams(), - * and ZSTD_estimateCCtxSize_usingCCtxParams(), which can be used in tandem with ZSTD_CCtxParams_setParameter(). - * Both can be used to estimate memory using custom compression parameters and arbitrary srcSize limits. - * - * Note 2 : only single-threaded compression is supported. - * ZSTD_estimateCCtxSize_usingCCtxParams() will return an error code if ZSTD_c_nbWorkers is >= 1. - */ -ZSTDLIB_STATIC_API size_t ZSTD_estimateCCtxSize(int compressionLevel); -ZSTDLIB_STATIC_API size_t ZSTD_estimateCCtxSize_usingCParams(ZSTD_compressionParameters cParams); -ZSTDLIB_STATIC_API size_t ZSTD_estimateCCtxSize_usingCCtxParams(const ZSTD_CCtx_params* params); -ZSTDLIB_STATIC_API size_t ZSTD_estimateDCtxSize(void); - -/*! ZSTD_estimateCStreamSize() : - * ZSTD_estimateCStreamSize() will provide a budget large enough for any compression level up to selected one. - * It will also consider src size to be arbitrarily "large", which is worst case. - * If srcSize is known to always be small, ZSTD_estimateCStreamSize_usingCParams() can provide a tighter estimation. - * ZSTD_estimateCStreamSize_usingCParams() can be used in tandem with ZSTD_getCParams() to create cParams from compressionLevel. - * ZSTD_estimateCStreamSize_usingCCtxParams() can be used in tandem with ZSTD_CCtxParams_setParameter(). Only single-threaded compression is supported. This function will return an error code if ZSTD_c_nbWorkers is >= 1. - * Note : CStream size estimation is only correct for single-threaded compression. - * ZSTD_DStream memory budget depends on window Size. - * This information can be passed manually, using ZSTD_estimateDStreamSize, - * or deducted from a valid frame Header, using ZSTD_estimateDStreamSize_fromFrame(); - * Note : if streaming is init with function ZSTD_init?Stream_usingDict(), - * an internal ?Dict will be created, which additional size is not estimated here. - * In this case, get total size by adding ZSTD_estimate?DictSize */ -ZSTDLIB_STATIC_API size_t ZSTD_estimateCStreamSize(int compressionLevel); -ZSTDLIB_STATIC_API size_t ZSTD_estimateCStreamSize_usingCParams(ZSTD_compressionParameters cParams); -ZSTDLIB_STATIC_API size_t ZSTD_estimateCStreamSize_usingCCtxParams(const ZSTD_CCtx_params* params); -ZSTDLIB_STATIC_API size_t ZSTD_estimateDStreamSize(size_t windowSize); -ZSTDLIB_STATIC_API size_t ZSTD_estimateDStreamSize_fromFrame(const void* src, size_t srcSize); - -/*! ZSTD_estimate?DictSize() : - * ZSTD_estimateCDictSize() will bet that src size is relatively "small", and content is copied, like ZSTD_createCDict(). - * ZSTD_estimateCDictSize_advanced() makes it possible to control compression parameters precisely, like ZSTD_createCDict_advanced(). - * Note : dictionaries created by reference (`ZSTD_dlm_byRef`) are logically smaller. - */ -ZSTDLIB_STATIC_API size_t ZSTD_estimateCDictSize(size_t dictSize, int compressionLevel); -ZSTDLIB_STATIC_API size_t ZSTD_estimateCDictSize_advanced(size_t dictSize, ZSTD_compressionParameters cParams, ZSTD_dictLoadMethod_e dictLoadMethod); -ZSTDLIB_STATIC_API size_t ZSTD_estimateDDictSize(size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod); - -/*! ZSTD_initStatic*() : - * Initialize an object using a pre-allocated fixed-size buffer. - * workspace: The memory area to emplace the object into. - * Provided pointer *must be 8-bytes aligned*. - * Buffer must outlive object. - * workspaceSize: Use ZSTD_estimate*Size() to determine - * how large workspace must be to support target scenario. - * @return : pointer to object (same address as workspace, just different type), - * or NULL if error (size too small, incorrect alignment, etc.) - * Note : zstd will never resize nor malloc() when using a static buffer. - * If the object requires more memory than available, - * zstd will just error out (typically ZSTD_error_memory_allocation). - * Note 2 : there is no corresponding "free" function. - * Since workspace is allocated externally, it must be freed externally too. - * Note 3 : cParams : use ZSTD_getCParams() to convert a compression level - * into its associated cParams. - * Limitation 1 : currently not compatible with internal dictionary creation, triggered by - * ZSTD_CCtx_loadDictionary(), ZSTD_initCStream_usingDict() or ZSTD_initDStream_usingDict(). - * Limitation 2 : static cctx currently not compatible with multi-threading. - * Limitation 3 : static dctx is incompatible with legacy support. - */ -ZSTDLIB_STATIC_API ZSTD_CCtx* ZSTD_initStaticCCtx(void* workspace, size_t workspaceSize); -ZSTDLIB_STATIC_API ZSTD_CStream* ZSTD_initStaticCStream(void* workspace, size_t workspaceSize); /**< same as ZSTD_initStaticCCtx() */ - -ZSTDLIB_STATIC_API ZSTD_DCtx* ZSTD_initStaticDCtx(void* workspace, size_t workspaceSize); -ZSTDLIB_STATIC_API ZSTD_DStream* ZSTD_initStaticDStream(void* workspace, size_t workspaceSize); /**< same as ZSTD_initStaticDCtx() */ - -ZSTDLIB_STATIC_API const ZSTD_CDict* ZSTD_initStaticCDict( - void* workspace, size_t workspaceSize, - const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - ZSTD_compressionParameters cParams); - -ZSTDLIB_STATIC_API const ZSTD_DDict* ZSTD_initStaticDDict( - void* workspace, size_t workspaceSize, - const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType); - - -/*! Custom memory allocation : - * These prototypes make it possible to pass your own allocation/free functions. - * ZSTD_customMem is provided at creation time, using ZSTD_create*_advanced() variants listed below. - * All allocation/free operations will be completed using these custom variants instead of regular ones. - */ -typedef void* (*ZSTD_allocFunction) (void* opaque, size_t size); -typedef void (*ZSTD_freeFunction) (void* opaque, void* address); -typedef struct { ZSTD_allocFunction customAlloc; ZSTD_freeFunction customFree; void* opaque; } ZSTD_customMem; -static -#ifdef __GNUC__ -__attribute__((__unused__)) -#endif -ZSTD_customMem const ZSTD_defaultCMem = { NULL, NULL, NULL }; /**< this constant defers to stdlib's functions */ - -ZSTDLIB_STATIC_API ZSTD_CCtx* ZSTD_createCCtx_advanced(ZSTD_customMem customMem); -ZSTDLIB_STATIC_API ZSTD_CStream* ZSTD_createCStream_advanced(ZSTD_customMem customMem); -ZSTDLIB_STATIC_API ZSTD_DCtx* ZSTD_createDCtx_advanced(ZSTD_customMem customMem); -ZSTDLIB_STATIC_API ZSTD_DStream* ZSTD_createDStream_advanced(ZSTD_customMem customMem); - -ZSTDLIB_STATIC_API ZSTD_CDict* ZSTD_createCDict_advanced(const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - ZSTD_compressionParameters cParams, - ZSTD_customMem customMem); - -/*! Thread pool : - * These prototypes make it possible to share a thread pool among multiple compression contexts. - * This can limit resources for applications with multiple threads where each one uses - * a threaded compression mode (via ZSTD_c_nbWorkers parameter). - * ZSTD_createThreadPool creates a new thread pool with a given number of threads. - * Note that the lifetime of such pool must exist while being used. - * ZSTD_CCtx_refThreadPool assigns a thread pool to a context (use NULL argument value - * to use an internal thread pool). - * ZSTD_freeThreadPool frees a thread pool, accepts NULL pointer. - */ -typedef struct POOL_ctx_s ZSTD_threadPool; -ZSTDLIB_STATIC_API ZSTD_threadPool* ZSTD_createThreadPool(size_t numThreads); -ZSTDLIB_STATIC_API void ZSTD_freeThreadPool (ZSTD_threadPool* pool); /* accept NULL pointer */ -ZSTDLIB_STATIC_API size_t ZSTD_CCtx_refThreadPool(ZSTD_CCtx* cctx, ZSTD_threadPool* pool); - - -/* - * This API is temporary and is expected to change or disappear in the future! - */ -ZSTDLIB_STATIC_API ZSTD_CDict* ZSTD_createCDict_advanced2( - const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - const ZSTD_CCtx_params* cctxParams, - ZSTD_customMem customMem); - -ZSTDLIB_STATIC_API ZSTD_DDict* ZSTD_createDDict_advanced( - const void* dict, size_t dictSize, - ZSTD_dictLoadMethod_e dictLoadMethod, - ZSTD_dictContentType_e dictContentType, - ZSTD_customMem customMem); - - -/*************************************** -* Advanced compression functions -***************************************/ - -/*! ZSTD_createCDict_byReference() : - * Create a digested dictionary for compression - * Dictionary content is just referenced, not duplicated. - * As a consequence, `dictBuffer` **must** outlive CDict, - * and its content must remain unmodified throughout the lifetime of CDict. - * note: equivalent to ZSTD_createCDict_advanced(), with dictLoadMethod==ZSTD_dlm_byRef */ -ZSTDLIB_STATIC_API ZSTD_CDict* ZSTD_createCDict_byReference(const void* dictBuffer, size_t dictSize, int compressionLevel); - -/*! ZSTD_getCParams() : - * @return ZSTD_compressionParameters structure for a selected compression level and estimated srcSize. - * `estimatedSrcSize` value is optional, select 0 if not known */ -ZSTDLIB_STATIC_API ZSTD_compressionParameters ZSTD_getCParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize); - -/*! ZSTD_getParams() : - * same as ZSTD_getCParams(), but @return a full `ZSTD_parameters` object instead of sub-component `ZSTD_compressionParameters`. - * All fields of `ZSTD_frameParameters` are set to default : contentSize=1, checksum=0, noDictID=0 */ -ZSTDLIB_STATIC_API ZSTD_parameters ZSTD_getParams(int compressionLevel, unsigned long long estimatedSrcSize, size_t dictSize); - -/*! ZSTD_checkCParams() : - * Ensure param values remain within authorized range. - * @return 0 on success, or an error code (can be checked with ZSTD_isError()) */ -ZSTDLIB_STATIC_API size_t ZSTD_checkCParams(ZSTD_compressionParameters params); - -/*! ZSTD_adjustCParams() : - * optimize params for a given `srcSize` and `dictSize`. - * `srcSize` can be unknown, in which case use ZSTD_CONTENTSIZE_UNKNOWN. - * `dictSize` must be `0` when there is no dictionary. - * cPar can be invalid : all parameters will be clamped within valid range in the @return struct. - * This function never fails (wide contract) */ -ZSTDLIB_STATIC_API ZSTD_compressionParameters ZSTD_adjustCParams(ZSTD_compressionParameters cPar, unsigned long long srcSize, size_t dictSize); - -/*! ZSTD_compress_advanced() : - * Note : this function is now DEPRECATED. - * It can be replaced by ZSTD_compress2(), in combination with ZSTD_CCtx_setParameter() and other parameter setters. - * This prototype will generate compilation warnings. */ -ZSTD_DEPRECATED("use ZSTD_compress2") -size_t ZSTD_compress_advanced(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const void* dict,size_t dictSize, - ZSTD_parameters params); - -/*! ZSTD_compress_usingCDict_advanced() : - * Note : this function is now DEPRECATED. - * It can be replaced by ZSTD_compress2(), in combination with ZSTD_CCtx_loadDictionary() and other parameter setters. - * This prototype will generate compilation warnings. */ -ZSTD_DEPRECATED("use ZSTD_compress2 with ZSTD_CCtx_loadDictionary") -size_t ZSTD_compress_usingCDict_advanced(ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, - const void* src, size_t srcSize, - const ZSTD_CDict* cdict, - ZSTD_frameParameters fParams); - - -/*! ZSTD_CCtx_loadDictionary_byReference() : - * Same as ZSTD_CCtx_loadDictionary(), but dictionary content is referenced, instead of being copied into CCtx. - * It saves some memory, but also requires that `dict` outlives its usage within `cctx` */ -ZSTDLIB_STATIC_API size_t ZSTD_CCtx_loadDictionary_byReference(ZSTD_CCtx* cctx, const void* dict, size_t dictSize); - -/*! ZSTD_CCtx_loadDictionary_advanced() : - * Same as ZSTD_CCtx_loadDictionary(), but gives finer control over - * how to load the dictionary (by copy ? by reference ?) - * and how to interpret it (automatic ? force raw mode ? full mode only ?) */ -ZSTDLIB_STATIC_API size_t ZSTD_CCtx_loadDictionary_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType); - -/*! ZSTD_CCtx_refPrefix_advanced() : - * Same as ZSTD_CCtx_refPrefix(), but gives finer control over - * how to interpret prefix content (automatic ? force raw mode (default) ? full mode only ?) */ -ZSTDLIB_STATIC_API size_t ZSTD_CCtx_refPrefix_advanced(ZSTD_CCtx* cctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType); - -/* === experimental parameters === */ -/* these parameters can be used with ZSTD_setParameter() - * they are not guaranteed to remain supported in the future */ - - /* Enables rsyncable mode, - * which makes compressed files more rsync friendly - * by adding periodic synchronization points to the compressed data. - * The target average block size is ZSTD_c_jobSize / 2. - * It's possible to modify the job size to increase or decrease - * the granularity of the synchronization point. - * Once the jobSize is smaller than the window size, - * it will result in compression ratio degradation. - * NOTE 1: rsyncable mode only works when multithreading is enabled. - * NOTE 2: rsyncable performs poorly in combination with long range mode, - * since it will decrease the effectiveness of synchronization points, - * though mileage may vary. - * NOTE 3: Rsyncable mode limits maximum compression speed to ~400 MB/s. - * If the selected compression level is already running significantly slower, - * the overall speed won't be significantly impacted. - */ - #define ZSTD_c_rsyncable ZSTD_c_experimentalParam1 - -/* Select a compression format. - * The value must be of type ZSTD_format_e. - * See ZSTD_format_e enum definition for details */ -#define ZSTD_c_format ZSTD_c_experimentalParam2 - -/* Force back-reference distances to remain < windowSize, - * even when referencing into Dictionary content (default:0) */ -#define ZSTD_c_forceMaxWindow ZSTD_c_experimentalParam3 - -/* Controls whether the contents of a CDict - * are used in place, or copied into the working context. - * Accepts values from the ZSTD_dictAttachPref_e enum. - * See the comments on that enum for an explanation of the feature. */ -#define ZSTD_c_forceAttachDict ZSTD_c_experimentalParam4 - -/* Controlled with ZSTD_paramSwitch_e enum. - * Default is ZSTD_ps_auto. - * Set to ZSTD_ps_disable to never compress literals. - * Set to ZSTD_ps_enable to always compress literals. (Note: uncompressed literals - * may still be emitted if huffman is not beneficial to use.) - * - * By default, in ZSTD_ps_auto, the library will decide at runtime whether to use - * literals compression based on the compression parameters - specifically, - * negative compression levels do not use literal compression. - */ -#define ZSTD_c_literalCompressionMode ZSTD_c_experimentalParam5 - -/* Tries to fit compressed block size to be around targetCBlockSize. - * No target when targetCBlockSize == 0. - * There is no guarantee on compressed block size (default:0) */ -#define ZSTD_c_targetCBlockSize ZSTD_c_experimentalParam6 - -/* User's best guess of source size. - * Hint is not valid when srcSizeHint == 0. - * There is no guarantee that hint is close to actual source size, - * but compression ratio may regress significantly if guess considerably underestimates */ -#define ZSTD_c_srcSizeHint ZSTD_c_experimentalParam7 - -/* Controls whether the new and experimental "dedicated dictionary search - * structure" can be used. This feature is still rough around the edges, be - * prepared for surprising behavior! - * - * How to use it: - * - * When using a CDict, whether to use this feature or not is controlled at - * CDict creation, and it must be set in a CCtxParams set passed into that - * construction (via ZSTD_createCDict_advanced2()). A compression will then - * use the feature or not based on how the CDict was constructed; the value of - * this param, set in the CCtx, will have no effect. - * - * However, when a dictionary buffer is passed into a CCtx, such as via - * ZSTD_CCtx_loadDictionary(), this param can be set on the CCtx to control - * whether the CDict that is created internally can use the feature or not. - * - * What it does: - * - * Normally, the internal data structures of the CDict are analogous to what - * would be stored in a CCtx after compressing the contents of a dictionary. - * To an approximation, a compression using a dictionary can then use those - * data structures to simply continue what is effectively a streaming - * compression where the simulated compression of the dictionary left off. - * Which is to say, the search structures in the CDict are normally the same - * format as in the CCtx. - * - * It is possible to do better, since the CDict is not like a CCtx: the search - * structures are written once during CDict creation, and then are only read - * after that, while the search structures in the CCtx are both read and - * written as the compression goes along. This means we can choose a search - * structure for the dictionary that is read-optimized. - * - * This feature enables the use of that different structure. - * - * Note that some of the members of the ZSTD_compressionParameters struct have - * different semantics and constraints in the dedicated search structure. It is - * highly recommended that you simply set a compression level in the CCtxParams - * you pass into the CDict creation call, and avoid messing with the cParams - * directly. - * - * Effects: - * - * This will only have any effect when the selected ZSTD_strategy - * implementation supports this feature. Currently, that's limited to - * ZSTD_greedy, ZSTD_lazy, and ZSTD_lazy2. - * - * Note that this means that the CDict tables can no longer be copied into the - * CCtx, so the dict attachment mode ZSTD_dictForceCopy will no longer be - * usable. The dictionary can only be attached or reloaded. - * - * In general, you should expect compression to be faster--sometimes very much - * so--and CDict creation to be slightly slower. Eventually, we will probably - * make this mode the default. - */ -#define ZSTD_c_enableDedicatedDictSearch ZSTD_c_experimentalParam8 - -/* ZSTD_c_stableInBuffer - * Experimental parameter. - * Default is 0 == disabled. Set to 1 to enable. - * - * Tells the compressor that the ZSTD_inBuffer will ALWAYS be the same - * between calls, except for the modifications that zstd makes to pos (the - * caller must not modify pos). This is checked by the compressor, and - * compression will fail if it ever changes. This means the only flush - * mode that makes sense is ZSTD_e_end, so zstd will error if ZSTD_e_end - * is not used. The data in the ZSTD_inBuffer in the range [src, src + pos) - * MUST not be modified during compression or you will get data corruption. - * - * When this flag is enabled zstd won't allocate an input window buffer, - * because the user guarantees it can reference the ZSTD_inBuffer until - * the frame is complete. But, it will still allocate an output buffer - * large enough to fit a block (see ZSTD_c_stableOutBuffer). This will also - * avoid the memcpy() from the input buffer to the input window buffer. - * - * NOTE: ZSTD_compressStream2() will error if ZSTD_e_end is not used. - * That means this flag cannot be used with ZSTD_compressStream(). - * - * NOTE: So long as the ZSTD_inBuffer always points to valid memory, using - * this flag is ALWAYS memory safe, and will never access out-of-bounds - * memory. However, compression WILL fail if you violate the preconditions. - * - * WARNING: The data in the ZSTD_inBuffer in the range [dst, dst + pos) MUST - * not be modified during compression or you will get data corruption. This - * is because zstd needs to reference data in the ZSTD_inBuffer to find - * matches. Normally zstd maintains its own window buffer for this purpose, - * but passing this flag tells zstd to use the user provided buffer. - */ -#define ZSTD_c_stableInBuffer ZSTD_c_experimentalParam9 - -/* ZSTD_c_stableOutBuffer - * Experimental parameter. - * Default is 0 == disabled. Set to 1 to enable. - * - * Tells he compressor that the ZSTD_outBuffer will not be resized between - * calls. Specifically: (out.size - out.pos) will never grow. This gives the - * compressor the freedom to say: If the compressed data doesn't fit in the - * output buffer then return ZSTD_error_dstSizeTooSmall. This allows us to - * always decompress directly into the output buffer, instead of decompressing - * into an internal buffer and copying to the output buffer. - * - * When this flag is enabled zstd won't allocate an output buffer, because - * it can write directly to the ZSTD_outBuffer. It will still allocate the - * input window buffer (see ZSTD_c_stableInBuffer). - * - * Zstd will check that (out.size - out.pos) never grows and return an error - * if it does. While not strictly necessary, this should prevent surprises. - */ -#define ZSTD_c_stableOutBuffer ZSTD_c_experimentalParam10 - -/* ZSTD_c_blockDelimiters - * Default is 0 == ZSTD_sf_noBlockDelimiters. - * - * For use with sequence compression API: ZSTD_compressSequences(). - * - * Designates whether or not the given array of ZSTD_Sequence contains block delimiters - * and last literals, which are defined as sequences with offset == 0 and matchLength == 0. - * See the definition of ZSTD_Sequence for more specifics. - */ -#define ZSTD_c_blockDelimiters ZSTD_c_experimentalParam11 - -/* ZSTD_c_validateSequences - * Default is 0 == disabled. Set to 1 to enable sequence validation. - * - * For use with sequence compression API: ZSTD_compressSequences(). - * Designates whether or not we validate sequences provided to ZSTD_compressSequences() - * during function execution. - * - * Without validation, providing a sequence that does not conform to the zstd spec will cause - * undefined behavior, and may produce a corrupted block. - * - * With validation enabled, a if sequence is invalid (see doc/zstd_compression_format.md for - * specifics regarding offset/matchlength requirements) then the function will bail out and - * return an error. - * - */ -#define ZSTD_c_validateSequences ZSTD_c_experimentalParam12 - -/* ZSTD_c_useBlockSplitter - * Controlled with ZSTD_paramSwitch_e enum. - * Default is ZSTD_ps_auto. - * Set to ZSTD_ps_disable to never use block splitter. - * Set to ZSTD_ps_enable to always use block splitter. - * - * By default, in ZSTD_ps_auto, the library will decide at runtime whether to use - * block splitting based on the compression parameters. - */ -#define ZSTD_c_useBlockSplitter ZSTD_c_experimentalParam13 - -/* ZSTD_c_useRowMatchFinder - * Controlled with ZSTD_paramSwitch_e enum. - * Default is ZSTD_ps_auto. - * Set to ZSTD_ps_disable to never use row-based matchfinder. - * Set to ZSTD_ps_enable to force usage of row-based matchfinder. - * - * By default, in ZSTD_ps_auto, the library will decide at runtime whether to use - * the row-based matchfinder based on support for SIMD instructions and the window log. - * Note that this only pertains to compression strategies: greedy, lazy, and lazy2 - */ -#define ZSTD_c_useRowMatchFinder ZSTD_c_experimentalParam14 - -/* ZSTD_c_deterministicRefPrefix - * Default is 0 == disabled. Set to 1 to enable. - * - * Zstd produces different results for prefix compression when the prefix is - * directly adjacent to the data about to be compressed vs. when it isn't. - * This is because zstd detects that the two buffers are contiguous and it can - * use a more efficient match finding algorithm. However, this produces different - * results than when the two buffers are non-contiguous. This flag forces zstd - * to always load the prefix in non-contiguous mode, even if it happens to be - * adjacent to the data, to guarantee determinism. - * - * If you really care about determinism when using a dictionary or prefix, - * like when doing delta compression, you should select this option. It comes - * at a speed penalty of about ~2.5% if the dictionary and data happened to be - * contiguous, and is free if they weren't contiguous. We don't expect that - * intentionally making the dictionary and data contiguous will be worth the - * cost to memcpy() the data. - */ -#define ZSTD_c_deterministicRefPrefix ZSTD_c_experimentalParam15 - -/*! ZSTD_CCtx_getParameter() : - * Get the requested compression parameter value, selected by enum ZSTD_cParameter, - * and store it into int* value. - * @return : 0, or an error code (which can be tested with ZSTD_isError()). - */ -ZSTDLIB_STATIC_API size_t ZSTD_CCtx_getParameter(const ZSTD_CCtx* cctx, ZSTD_cParameter param, int* value); - - -/*! ZSTD_CCtx_params : - * Quick howto : - * - ZSTD_createCCtxParams() : Create a ZSTD_CCtx_params structure - * - ZSTD_CCtxParams_setParameter() : Push parameters one by one into - * an existing ZSTD_CCtx_params structure. - * This is similar to - * ZSTD_CCtx_setParameter(). - * - ZSTD_CCtx_setParametersUsingCCtxParams() : Apply parameters to - * an existing CCtx. - * These parameters will be applied to - * all subsequent frames. - * - ZSTD_compressStream2() : Do compression using the CCtx. - * - ZSTD_freeCCtxParams() : Free the memory, accept NULL pointer. - * - * This can be used with ZSTD_estimateCCtxSize_advanced_usingCCtxParams() - * for static allocation of CCtx for single-threaded compression. - */ -ZSTDLIB_STATIC_API ZSTD_CCtx_params* ZSTD_createCCtxParams(void); -ZSTDLIB_STATIC_API size_t ZSTD_freeCCtxParams(ZSTD_CCtx_params* params); /* accept NULL pointer */ - -/*! ZSTD_CCtxParams_reset() : - * Reset params to default values. - */ -ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_reset(ZSTD_CCtx_params* params); - -/*! ZSTD_CCtxParams_init() : - * Initializes the compression parameters of cctxParams according to - * compression level. All other parameters are reset to their default values. - */ -ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_init(ZSTD_CCtx_params* cctxParams, int compressionLevel); - -/*! ZSTD_CCtxParams_init_advanced() : - * Initializes the compression and frame parameters of cctxParams according to - * params. All other parameters are reset to their default values. - */ -ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_init_advanced(ZSTD_CCtx_params* cctxParams, ZSTD_parameters params); - -/*! ZSTD_CCtxParams_setParameter() : Requires v1.4.0+ - * Similar to ZSTD_CCtx_setParameter. - * Set one compression parameter, selected by enum ZSTD_cParameter. - * Parameters must be applied to a ZSTD_CCtx using - * ZSTD_CCtx_setParametersUsingCCtxParams(). - * @result : a code representing success or failure (which can be tested with - * ZSTD_isError()). - */ -ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_setParameter(ZSTD_CCtx_params* params, ZSTD_cParameter param, int value); - -/*! ZSTD_CCtxParams_getParameter() : - * Similar to ZSTD_CCtx_getParameter. - * Get the requested value of one compression parameter, selected by enum ZSTD_cParameter. - * @result : 0, or an error code (which can be tested with ZSTD_isError()). - */ -ZSTDLIB_STATIC_API size_t ZSTD_CCtxParams_getParameter(const ZSTD_CCtx_params* params, ZSTD_cParameter param, int* value); - -/*! ZSTD_CCtx_setParametersUsingCCtxParams() : - * Apply a set of ZSTD_CCtx_params to the compression context. - * This can be done even after compression is started, - * if nbWorkers==0, this will have no impact until a new compression is started. - * if nbWorkers>=1, new parameters will be picked up at next job, - * with a few restrictions (windowLog, pledgedSrcSize, nbWorkers, jobSize, and overlapLog are not updated). - */ -ZSTDLIB_STATIC_API size_t ZSTD_CCtx_setParametersUsingCCtxParams( - ZSTD_CCtx* cctx, const ZSTD_CCtx_params* params); - -/*! ZSTD_compressStream2_simpleArgs() : - * Same as ZSTD_compressStream2(), - * but using only integral types as arguments. - * This variant might be helpful for binders from dynamic languages - * which have troubles handling structures containing memory pointers. - */ -ZSTDLIB_STATIC_API size_t ZSTD_compressStream2_simpleArgs ( - ZSTD_CCtx* cctx, - void* dst, size_t dstCapacity, size_t* dstPos, - const void* src, size_t srcSize, size_t* srcPos, - ZSTD_EndDirective endOp); - - -/*************************************** -* Advanced decompression functions -***************************************/ - -/*! ZSTD_isFrame() : - * Tells if the content of `buffer` starts with a valid Frame Identifier. - * Note : Frame Identifier is 4 bytes. If `size < 4`, @return will always be 0. - * Note 2 : Legacy Frame Identifiers are considered valid only if Legacy Support is enabled. - * Note 3 : Skippable Frame Identifiers are considered valid. */ -ZSTDLIB_STATIC_API unsigned ZSTD_isFrame(const void* buffer, size_t size); - -/*! ZSTD_createDDict_byReference() : - * Create a digested dictionary, ready to start decompression operation without startup delay. - * Dictionary content is referenced, and therefore stays in dictBuffer. - * It is important that dictBuffer outlives DDict, - * it must remain read accessible throughout the lifetime of DDict */ -ZSTDLIB_STATIC_API ZSTD_DDict* ZSTD_createDDict_byReference(const void* dictBuffer, size_t dictSize); - -/*! ZSTD_DCtx_loadDictionary_byReference() : - * Same as ZSTD_DCtx_loadDictionary(), - * but references `dict` content instead of copying it into `dctx`. - * This saves memory if `dict` remains around., - * However, it's imperative that `dict` remains accessible (and unmodified) while being used, so it must outlive decompression. */ -ZSTDLIB_STATIC_API size_t ZSTD_DCtx_loadDictionary_byReference(ZSTD_DCtx* dctx, const void* dict, size_t dictSize); - -/*! ZSTD_DCtx_loadDictionary_advanced() : - * Same as ZSTD_DCtx_loadDictionary(), - * but gives direct control over - * how to load the dictionary (by copy ? by reference ?) - * and how to interpret it (automatic ? force raw mode ? full mode only ?). */ -ZSTDLIB_STATIC_API size_t ZSTD_DCtx_loadDictionary_advanced(ZSTD_DCtx* dctx, const void* dict, size_t dictSize, ZSTD_dictLoadMethod_e dictLoadMethod, ZSTD_dictContentType_e dictContentType); - -/*! ZSTD_DCtx_refPrefix_advanced() : - * Same as ZSTD_DCtx_refPrefix(), but gives finer control over - * how to interpret prefix content (automatic ? force raw mode (default) ? full mode only ?) */ -ZSTDLIB_STATIC_API size_t ZSTD_DCtx_refPrefix_advanced(ZSTD_DCtx* dctx, const void* prefix, size_t prefixSize, ZSTD_dictContentType_e dictContentType); - -/*! ZSTD_DCtx_setMaxWindowSize() : - * Refuses allocating internal buffers for frames requiring a window size larger than provided limit. - * This protects a decoder context from reserving too much memory for itself (potential attack scenario). - * This parameter is only useful in streaming mode, since no internal buffer is allocated in single-pass mode. - * By default, a decompression context accepts all window sizes <= (1 << ZSTD_WINDOWLOG_LIMIT_DEFAULT) - * @return : 0, or an error code (which can be tested using ZSTD_isError()). - */ -ZSTDLIB_STATIC_API size_t ZSTD_DCtx_setMaxWindowSize(ZSTD_DCtx* dctx, size_t maxWindowSize); - -/*! ZSTD_DCtx_getParameter() : - * Get the requested decompression parameter value, selected by enum ZSTD_dParameter, - * and store it into int* value. - * @return : 0, or an error code (which can be tested with ZSTD_isError()). - */ -ZSTDLIB_STATIC_API size_t ZSTD_DCtx_getParameter(ZSTD_DCtx* dctx, ZSTD_dParameter param, int* value); - -/* ZSTD_d_format - * experimental parameter, - * allowing selection between ZSTD_format_e input compression formats - */ -#define ZSTD_d_format ZSTD_d_experimentalParam1 -/* ZSTD_d_stableOutBuffer - * Experimental parameter. - * Default is 0 == disabled. Set to 1 to enable. - * - * Tells the decompressor that the ZSTD_outBuffer will ALWAYS be the same - * between calls, except for the modifications that zstd makes to pos (the - * caller must not modify pos). This is checked by the decompressor, and - * decompression will fail if it ever changes. Therefore the ZSTD_outBuffer - * MUST be large enough to fit the entire decompressed frame. This will be - * checked when the frame content size is known. The data in the ZSTD_outBuffer - * in the range [dst, dst + pos) MUST not be modified during decompression - * or you will get data corruption. - * - * When this flags is enabled zstd won't allocate an output buffer, because - * it can write directly to the ZSTD_outBuffer, but it will still allocate - * an input buffer large enough to fit any compressed block. This will also - * avoid the memcpy() from the internal output buffer to the ZSTD_outBuffer. - * If you need to avoid the input buffer allocation use the buffer-less - * streaming API. - * - * NOTE: So long as the ZSTD_outBuffer always points to valid memory, using - * this flag is ALWAYS memory safe, and will never access out-of-bounds - * memory. However, decompression WILL fail if you violate the preconditions. - * - * WARNING: The data in the ZSTD_outBuffer in the range [dst, dst + pos) MUST - * not be modified during decompression or you will get data corruption. This - * is because zstd needs to reference data in the ZSTD_outBuffer to regenerate - * matches. Normally zstd maintains its own buffer for this purpose, but passing - * this flag tells zstd to use the user provided buffer. - */ -#define ZSTD_d_stableOutBuffer ZSTD_d_experimentalParam2 - -/* ZSTD_d_forceIgnoreChecksum - * Experimental parameter. - * Default is 0 == disabled. Set to 1 to enable - * - * Tells the decompressor to skip checksum validation during decompression, regardless - * of whether checksumming was specified during compression. This offers some - * slight performance benefits, and may be useful for debugging. - * Param has values of type ZSTD_forceIgnoreChecksum_e - */ -#define ZSTD_d_forceIgnoreChecksum ZSTD_d_experimentalParam3 - -/* ZSTD_d_refMultipleDDicts - * Experimental parameter. - * Default is 0 == disabled. Set to 1 to enable - * - * If enabled and dctx is allocated on the heap, then additional memory will be allocated - * to store references to multiple ZSTD_DDict. That is, multiple calls of ZSTD_refDDict() - * using a given ZSTD_DCtx, rather than overwriting the previous DDict reference, will instead - * store all references. At decompression time, the appropriate dictID is selected - * from the set of DDicts based on the dictID in the frame. - * - * Usage is simply calling ZSTD_refDDict() on multiple dict buffers. - * - * Param has values of byte ZSTD_refMultipleDDicts_e - * - * WARNING: Enabling this parameter and calling ZSTD_DCtx_refDDict(), will trigger memory - * allocation for the hash table. ZSTD_freeDCtx() also frees this memory. - * Memory is allocated as per ZSTD_DCtx::customMem. - * - * Although this function allocates memory for the table, the user is still responsible for - * memory management of the underlying ZSTD_DDict* themselves. - */ -#define ZSTD_d_refMultipleDDicts ZSTD_d_experimentalParam4 - - -/*! ZSTD_DCtx_setFormat() : - * This function is REDUNDANT. Prefer ZSTD_DCtx_setParameter(). - * Instruct the decoder context about what kind of data to decode next. - * This instruction is mandatory to decode data without a fully-formed header, - * such ZSTD_f_zstd1_magicless for example. - * @return : 0, or an error code (which can be tested using ZSTD_isError()). */ -ZSTD_DEPRECATED("use ZSTD_DCtx_setParameter() instead") -size_t ZSTD_DCtx_setFormat(ZSTD_DCtx* dctx, ZSTD_format_e format); - -/*! ZSTD_decompressStream_simpleArgs() : - * Same as ZSTD_decompressStream(), - * but using only integral types as arguments. - * This can be helpful for binders from dynamic languages - * which have troubles handling structures containing memory pointers. - */ -ZSTDLIB_STATIC_API size_t ZSTD_decompressStream_simpleArgs ( - ZSTD_DCtx* dctx, - void* dst, size_t dstCapacity, size_t* dstPos, - const void* src, size_t srcSize, size_t* srcPos); - - -/******************************************************************** -* Advanced streaming functions -* Warning : most of these functions are now redundant with the Advanced API. -* Once Advanced API reaches "stable" status, -* redundant functions will be deprecated, and then at some point removed. -********************************************************************/ - -/*===== Advanced Streaming compression functions =====*/ - -/*! ZSTD_initCStream_srcSize() : - * This function is DEPRECATED, and equivalent to: - * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); - * ZSTD_CCtx_refCDict(zcs, NULL); // clear the dictionary (if any) - * ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel); - * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize); - * - * pledgedSrcSize must be correct. If it is not known at init time, use - * ZSTD_CONTENTSIZE_UNKNOWN. Note that, for compatibility with older programs, - * "0" also disables frame content size field. It may be enabled in the future. - * This prototype will generate compilation warnings. - */ -ZSTD_DEPRECATED("use ZSTD_CCtx_reset, see zstd.h for detailed instructions") -size_t ZSTD_initCStream_srcSize(ZSTD_CStream* zcs, - int compressionLevel, - unsigned long long pledgedSrcSize); - -/*! ZSTD_initCStream_usingDict() : - * This function is DEPRECATED, and is equivalent to: - * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); - * ZSTD_CCtx_setParameter(zcs, ZSTD_c_compressionLevel, compressionLevel); - * ZSTD_CCtx_loadDictionary(zcs, dict, dictSize); - * - * Creates of an internal CDict (incompatible with static CCtx), except if - * dict == NULL or dictSize < 8, in which case no dict is used. - * Note: dict is loaded with ZSTD_dct_auto (treated as a full zstd dictionary if - * it begins with ZSTD_MAGIC_DICTIONARY, else as raw content) and ZSTD_dlm_byCopy. - * This prototype will generate compilation warnings. - */ -ZSTD_DEPRECATED("use ZSTD_CCtx_reset, see zstd.h for detailed instructions") -size_t ZSTD_initCStream_usingDict(ZSTD_CStream* zcs, - const void* dict, size_t dictSize, - int compressionLevel); - -/*! ZSTD_initCStream_advanced() : - * This function is DEPRECATED, and is approximately equivalent to: - * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); - * // Pseudocode: Set each zstd parameter and leave the rest as-is. - * for ((param, value) : params) { - * ZSTD_CCtx_setParameter(zcs, param, value); - * } - * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize); - * ZSTD_CCtx_loadDictionary(zcs, dict, dictSize); - * - * dict is loaded with ZSTD_dct_auto and ZSTD_dlm_byCopy. - * pledgedSrcSize must be correct. - * If srcSize is not known at init time, use value ZSTD_CONTENTSIZE_UNKNOWN. - * This prototype will generate compilation warnings. - */ -ZSTD_DEPRECATED("use ZSTD_CCtx_reset, see zstd.h for detailed instructions") -size_t ZSTD_initCStream_advanced(ZSTD_CStream* zcs, - const void* dict, size_t dictSize, - ZSTD_parameters params, - unsigned long long pledgedSrcSize); - -/*! ZSTD_initCStream_usingCDict() : - * This function is DEPRECATED, and equivalent to: - * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); - * ZSTD_CCtx_refCDict(zcs, cdict); - * - * note : cdict will just be referenced, and must outlive compression session - * This prototype will generate compilation warnings. - */ -ZSTD_DEPRECATED("use ZSTD_CCtx_reset and ZSTD_CCtx_refCDict, see zstd.h for detailed instructions") -size_t ZSTD_initCStream_usingCDict(ZSTD_CStream* zcs, const ZSTD_CDict* cdict); - -/*! ZSTD_initCStream_usingCDict_advanced() : - * This function is DEPRECATED, and is approximately equivalent to: - * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); - * // Pseudocode: Set each zstd frame parameter and leave the rest as-is. - * for ((fParam, value) : fParams) { - * ZSTD_CCtx_setParameter(zcs, fParam, value); - * } - * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize); - * ZSTD_CCtx_refCDict(zcs, cdict); - * - * same as ZSTD_initCStream_usingCDict(), with control over frame parameters. - * pledgedSrcSize must be correct. If srcSize is not known at init time, use - * value ZSTD_CONTENTSIZE_UNKNOWN. - * This prototype will generate compilation warnings. - */ -ZSTD_DEPRECATED("use ZSTD_CCtx_reset and ZSTD_CCtx_refCDict, see zstd.h for detailed instructions") -size_t ZSTD_initCStream_usingCDict_advanced(ZSTD_CStream* zcs, - const ZSTD_CDict* cdict, - ZSTD_frameParameters fParams, - unsigned long long pledgedSrcSize); - -/*! ZSTD_resetCStream() : - * This function is DEPRECATED, and is equivalent to: - * ZSTD_CCtx_reset(zcs, ZSTD_reset_session_only); - * ZSTD_CCtx_setPledgedSrcSize(zcs, pledgedSrcSize); - * Note: ZSTD_resetCStream() interprets pledgedSrcSize == 0 as ZSTD_CONTENTSIZE_UNKNOWN, but - * ZSTD_CCtx_setPledgedSrcSize() does not do the same, so ZSTD_CONTENTSIZE_UNKNOWN must be - * explicitly specified. - * - * start a new frame, using same parameters from previous frame. - * This is typically useful to skip dictionary loading stage, since it will re-use it in-place. - * Note that zcs must be init at least once before using ZSTD_resetCStream(). - * If pledgedSrcSize is not known at reset time, use macro ZSTD_CONTENTSIZE_UNKNOWN. - * If pledgedSrcSize > 0, its value must be correct, as it will be written in header, and controlled at the end. - * For the time being, pledgedSrcSize==0 is interpreted as "srcSize unknown" for compatibility with older programs, - * but it will change to mean "empty" in future version, so use macro ZSTD_CONTENTSIZE_UNKNOWN instead. - * @return : 0, or an error code (which can be tested using ZSTD_isError()) - * This prototype will generate compilation warnings. - */ -ZSTD_DEPRECATED("use ZSTD_CCtx_reset, see zstd.h for detailed instructions") -size_t ZSTD_resetCStream(ZSTD_CStream* zcs, unsigned long long pledgedSrcSize); - - -typedef struct { - unsigned long long ingested; /* nb input bytes read and buffered */ - unsigned long long consumed; /* nb input bytes actually compressed */ - unsigned long long produced; /* nb of compressed bytes generated and buffered */ - unsigned long long flushed; /* nb of compressed bytes flushed : not provided; can be tracked from caller side */ - unsigned currentJobID; /* MT only : latest started job nb */ - unsigned nbActiveWorkers; /* MT only : nb of workers actively compressing at probe time */ -} ZSTD_frameProgression; - -/* ZSTD_getFrameProgression() : - * tells how much data has been ingested (read from input) - * consumed (input actually compressed) and produced (output) for current frame. - * Note : (ingested - consumed) is amount of input data buffered internally, not yet compressed. - * Aggregates progression inside active worker threads. - */ -ZSTDLIB_STATIC_API ZSTD_frameProgression ZSTD_getFrameProgression(const ZSTD_CCtx* cctx); - -/*! ZSTD_toFlushNow() : - * Tell how many bytes are ready to be flushed immediately. - * Useful for multithreading scenarios (nbWorkers >= 1). - * Probe the oldest active job, defined as oldest job not yet entirely flushed, - * and check its output buffer. - * @return : amount of data stored in oldest job and ready to be flushed immediately. - * if @return == 0, it means either : - * + there is no active job (could be checked with ZSTD_frameProgression()), or - * + oldest job is still actively compressing data, - * but everything it has produced has also been flushed so far, - * therefore flush speed is limited by production speed of oldest job - * irrespective of the speed of concurrent (and newer) jobs. - */ -ZSTDLIB_STATIC_API size_t ZSTD_toFlushNow(ZSTD_CCtx* cctx); - - -/*===== Advanced Streaming decompression functions =====*/ - -/*! - * This function is deprecated, and is equivalent to: - * - * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only); - * ZSTD_DCtx_loadDictionary(zds, dict, dictSize); - * - * note: no dictionary will be used if dict == NULL or dictSize < 8 - * Note : this prototype will be marked as deprecated and generate compilation warnings on reaching v1.5.x - */ -ZSTDLIB_STATIC_API size_t ZSTD_initDStream_usingDict(ZSTD_DStream* zds, const void* dict, size_t dictSize); - -/*! - * This function is deprecated, and is equivalent to: - * - * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only); - * ZSTD_DCtx_refDDict(zds, ddict); - * - * note : ddict is referenced, it must outlive decompression session - * Note : this prototype will be marked as deprecated and generate compilation warnings on reaching v1.5.x - */ -ZSTDLIB_STATIC_API size_t ZSTD_initDStream_usingDDict(ZSTD_DStream* zds, const ZSTD_DDict* ddict); - -/*! - * This function is deprecated, and is equivalent to: - * - * ZSTD_DCtx_reset(zds, ZSTD_reset_session_only); - * - * re-use decompression parameters from previous init; saves dictionary loading - * Note : this prototype will be marked as deprecated and generate compilation warnings on reaching v1.5.x - */ -ZSTDLIB_STATIC_API size_t ZSTD_resetDStream(ZSTD_DStream* zds); - - -/********************************************************************* -* Buffer-less and synchronous inner streaming functions -* -* This is an advanced API, giving full control over buffer management, for users which need direct control over memory. -* But it's also a complex one, with several restrictions, documented below. -* Prefer normal streaming API for an easier experience. -********************************************************************* */ - -/** - Buffer-less streaming compression (synchronous mode) - - A ZSTD_CCtx object is required to track streaming operations. - Use ZSTD_createCCtx() / ZSTD_freeCCtx() to manage resource. - ZSTD_CCtx object can be re-used multiple times within successive compression operations. - - Start by initializing a context. - Use ZSTD_compressBegin(), or ZSTD_compressBegin_usingDict() for dictionary compression. - It's also possible to duplicate a reference context which has already been initialized, using ZSTD_copyCCtx() - - Then, consume your input using ZSTD_compressContinue(). - There are some important considerations to keep in mind when using this advanced function : - - ZSTD_compressContinue() has no internal buffer. It uses externally provided buffers only. - - Interface is synchronous : input is consumed entirely and produces 1+ compressed blocks. - - Caller must ensure there is enough space in `dst` to store compressed data under worst case scenario. - Worst case evaluation is provided by ZSTD_compressBound(). - ZSTD_compressContinue() doesn't guarantee recover after a failed compression. - - ZSTD_compressContinue() presumes prior input ***is still accessible and unmodified*** (up to maximum distance size, see WindowLog). - It remembers all previous contiguous blocks, plus one separated memory segment (which can itself consists of multiple contiguous blocks) - - ZSTD_compressContinue() detects that prior input has been overwritten when `src` buffer overlaps. - In which case, it will "discard" the relevant memory section from its history. - - Finish a frame with ZSTD_compressEnd(), which will write the last block(s) and optional checksum. - It's possible to use srcSize==0, in which case, it will write a final empty block to end the frame. - Without last block mark, frames are considered unfinished (hence corrupted) by compliant decoders. - - `ZSTD_CCtx` object can be re-used (ZSTD_compressBegin()) to compress again. -*/ - -/*===== Buffer-less streaming compression functions =====*/ -ZSTDLIB_STATIC_API size_t ZSTD_compressBegin(ZSTD_CCtx* cctx, int compressionLevel); -ZSTDLIB_STATIC_API size_t ZSTD_compressBegin_usingDict(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, int compressionLevel); -ZSTDLIB_STATIC_API size_t ZSTD_compressBegin_usingCDict(ZSTD_CCtx* cctx, const ZSTD_CDict* cdict); /**< note: fails if cdict==NULL */ -ZSTDLIB_STATIC_API size_t ZSTD_copyCCtx(ZSTD_CCtx* cctx, const ZSTD_CCtx* preparedCCtx, unsigned long long pledgedSrcSize); /**< note: if pledgedSrcSize is not known, use ZSTD_CONTENTSIZE_UNKNOWN */ - -ZSTDLIB_STATIC_API size_t ZSTD_compressContinue(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); -ZSTDLIB_STATIC_API size_t ZSTD_compressEnd(ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); - -/* The ZSTD_compressBegin_advanced() and ZSTD_compressBegin_usingCDict_advanced() are now DEPRECATED and will generate a compiler warning */ -ZSTD_DEPRECATED("use advanced API to access custom parameters") -size_t ZSTD_compressBegin_advanced(ZSTD_CCtx* cctx, const void* dict, size_t dictSize, ZSTD_parameters params, unsigned long long pledgedSrcSize); /**< pledgedSrcSize : If srcSize is not known at init time, use ZSTD_CONTENTSIZE_UNKNOWN */ -ZSTD_DEPRECATED("use advanced API to access custom parameters") -size_t ZSTD_compressBegin_usingCDict_advanced(ZSTD_CCtx* const cctx, const ZSTD_CDict* const cdict, ZSTD_frameParameters const fParams, unsigned long long const pledgedSrcSize); /* compression parameters are already set within cdict. pledgedSrcSize must be correct. If srcSize is not known, use macro ZSTD_CONTENTSIZE_UNKNOWN */ -/** - Buffer-less streaming decompression (synchronous mode) - - A ZSTD_DCtx object is required to track streaming operations. - Use ZSTD_createDCtx() / ZSTD_freeDCtx() to manage it. - A ZSTD_DCtx object can be re-used multiple times. - - First typical operation is to retrieve frame parameters, using ZSTD_getFrameHeader(). - Frame header is extracted from the beginning of compressed frame, so providing only the frame's beginning is enough. - Data fragment must be large enough to ensure successful decoding. - `ZSTD_frameHeaderSize_max` bytes is guaranteed to always be large enough. - @result : 0 : successful decoding, the `ZSTD_frameHeader` structure is correctly filled. - >0 : `srcSize` is too small, please provide at least @result bytes on next attempt. - errorCode, which can be tested using ZSTD_isError(). - - It fills a ZSTD_frameHeader structure with important information to correctly decode the frame, - such as the dictionary ID, content size, or maximum back-reference distance (`windowSize`). - Note that these values could be wrong, either because of data corruption, or because a 3rd party deliberately spoofs false information. - As a consequence, check that values remain within valid application range. - For example, do not allocate memory blindly, check that `windowSize` is within expectation. - Each application can set its own limits, depending on local restrictions. - For extended interoperability, it is recommended to support `windowSize` of at least 8 MB. - - ZSTD_decompressContinue() needs previous data blocks during decompression, up to `windowSize` bytes. - ZSTD_decompressContinue() is very sensitive to contiguity, - if 2 blocks don't follow each other, make sure that either the compressor breaks contiguity at the same place, - or that previous contiguous segment is large enough to properly handle maximum back-reference distance. - There are multiple ways to guarantee this condition. - - The most memory efficient way is to use a round buffer of sufficient size. - Sufficient size is determined by invoking ZSTD_decodingBufferSize_min(), - which can @return an error code if required value is too large for current system (in 32-bits mode). - In a round buffer methodology, ZSTD_decompressContinue() decompresses each block next to previous one, - up to the moment there is not enough room left in the buffer to guarantee decoding another full block, - which maximum size is provided in `ZSTD_frameHeader` structure, field `blockSizeMax`. - At which point, decoding can resume from the beginning of the buffer. - Note that already decoded data stored in the buffer should be flushed before being overwritten. - - There are alternatives possible, for example using two or more buffers of size `windowSize` each, though they consume more memory. - - Finally, if you control the compression process, you can also ignore all buffer size rules, - as long as the encoder and decoder progress in "lock-step", - aka use exactly the same buffer sizes, break contiguity at the same place, etc. - - Once buffers are setup, start decompression, with ZSTD_decompressBegin(). - If decompression requires a dictionary, use ZSTD_decompressBegin_usingDict() or ZSTD_decompressBegin_usingDDict(). - - Then use ZSTD_nextSrcSizeToDecompress() and ZSTD_decompressContinue() alternatively. - ZSTD_nextSrcSizeToDecompress() tells how many bytes to provide as 'srcSize' to ZSTD_decompressContinue(). - ZSTD_decompressContinue() requires this _exact_ amount of bytes, or it will fail. - - @result of ZSTD_decompressContinue() is the number of bytes regenerated within 'dst' (necessarily <= dstCapacity). - It can be zero : it just means ZSTD_decompressContinue() has decoded some metadata item. - It can also be an error code, which can be tested with ZSTD_isError(). - - A frame is fully decoded when ZSTD_nextSrcSizeToDecompress() returns zero. - Context can then be reset to start a new decompression. - - Note : it's possible to know if next input to present is a header or a block, using ZSTD_nextInputType(). - This information is not required to properly decode a frame. - - == Special case : skippable frames == - - Skippable frames allow integration of user-defined data into a flow of concatenated frames. - Skippable frames will be ignored (skipped) by decompressor. - The format of skippable frames is as follows : - a) Skippable frame ID - 4 Bytes, Little endian format, any value from 0x184D2A50 to 0x184D2A5F - b) Frame Size - 4 Bytes, Little endian format, unsigned 32-bits - c) Frame Content - any content (User Data) of length equal to Frame Size - For skippable frames ZSTD_getFrameHeader() returns zfhPtr->frameType==ZSTD_skippableFrame. - For skippable frames ZSTD_decompressContinue() always returns 0 : it only skips the content. -*/ - -/*===== Buffer-less streaming decompression functions =====*/ -typedef enum { ZSTD_frame, ZSTD_skippableFrame } ZSTD_frameType_e; -typedef struct { - unsigned long long frameContentSize; /* if == ZSTD_CONTENTSIZE_UNKNOWN, it means this field is not available. 0 means "empty" */ - unsigned long long windowSize; /* can be very large, up to <= frameContentSize */ - unsigned blockSizeMax; - ZSTD_frameType_e frameType; /* if == ZSTD_skippableFrame, frameContentSize is the size of skippable content */ - unsigned headerSize; - unsigned dictID; - unsigned checksumFlag; -} ZSTD_frameHeader; - -/*! ZSTD_getFrameHeader() : - * decode Frame Header, or requires larger `srcSize`. - * @return : 0, `zfhPtr` is correctly filled, - * >0, `srcSize` is too small, value is wanted `srcSize` amount, - * or an error code, which can be tested using ZSTD_isError() */ -ZSTDLIB_STATIC_API size_t ZSTD_getFrameHeader(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize); /**< doesn't consume input */ -/*! ZSTD_getFrameHeader_advanced() : - * same as ZSTD_getFrameHeader(), - * with added capability to select a format (like ZSTD_f_zstd1_magicless) */ -ZSTDLIB_STATIC_API size_t ZSTD_getFrameHeader_advanced(ZSTD_frameHeader* zfhPtr, const void* src, size_t srcSize, ZSTD_format_e format); -ZSTDLIB_STATIC_API size_t ZSTD_decodingBufferSize_min(unsigned long long windowSize, unsigned long long frameContentSize); /**< when frame content size is not known, pass in frameContentSize == ZSTD_CONTENTSIZE_UNKNOWN */ - -ZSTDLIB_STATIC_API size_t ZSTD_decompressBegin(ZSTD_DCtx* dctx); -ZSTDLIB_STATIC_API size_t ZSTD_decompressBegin_usingDict(ZSTD_DCtx* dctx, const void* dict, size_t dictSize); -ZSTDLIB_STATIC_API size_t ZSTD_decompressBegin_usingDDict(ZSTD_DCtx* dctx, const ZSTD_DDict* ddict); - -ZSTDLIB_STATIC_API size_t ZSTD_nextSrcSizeToDecompress(ZSTD_DCtx* dctx); -ZSTDLIB_STATIC_API size_t ZSTD_decompressContinue(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); - -/* misc */ -ZSTDLIB_STATIC_API void ZSTD_copyDCtx(ZSTD_DCtx* dctx, const ZSTD_DCtx* preparedDCtx); -typedef enum { ZSTDnit_frameHeader, ZSTDnit_blockHeader, ZSTDnit_block, ZSTDnit_lastBlock, ZSTDnit_checksum, ZSTDnit_skippableFrame } ZSTD_nextInputType_e; -ZSTDLIB_STATIC_API ZSTD_nextInputType_e ZSTD_nextInputType(ZSTD_DCtx* dctx); - - - - -/* ============================ */ -/** Block level API */ -/* ============================ */ - -/*! - Block functions produce and decode raw zstd blocks, without frame metadata. - Frame metadata cost is typically ~12 bytes, which can be non-negligible for very small blocks (< 100 bytes). - But users will have to take in charge needed metadata to regenerate data, such as compressed and content sizes. - - A few rules to respect : - - Compressing and decompressing require a context structure - + Use ZSTD_createCCtx() and ZSTD_createDCtx() - - It is necessary to init context before starting - + compression : any ZSTD_compressBegin*() variant, including with dictionary - + decompression : any ZSTD_decompressBegin*() variant, including with dictionary - + copyCCtx() and copyDCtx() can be used too - - Block size is limited, it must be <= ZSTD_getBlockSize() <= ZSTD_BLOCKSIZE_MAX == 128 KB - + If input is larger than a block size, it's necessary to split input data into multiple blocks - + For inputs larger than a single block, consider using regular ZSTD_compress() instead. - Frame metadata is not that costly, and quickly becomes negligible as source size grows larger than a block. - - When a block is considered not compressible enough, ZSTD_compressBlock() result will be 0 (zero) ! - ===> In which case, nothing is produced into `dst` ! - + User __must__ test for such outcome and deal directly with uncompressed data - + A block cannot be declared incompressible if ZSTD_compressBlock() return value was != 0. - Doing so would mess up with statistics history, leading to potential data corruption. - + ZSTD_decompressBlock() _doesn't accept uncompressed data as input_ !! - + In case of multiple successive blocks, should some of them be uncompressed, - decoder must be informed of their existence in order to follow proper history. - Use ZSTD_insertBlock() for such a case. -*/ - -/*===== Raw zstd block functions =====*/ -ZSTDLIB_STATIC_API size_t ZSTD_getBlockSize (const ZSTD_CCtx* cctx); -ZSTDLIB_STATIC_API size_t ZSTD_compressBlock (ZSTD_CCtx* cctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); -ZSTDLIB_STATIC_API size_t ZSTD_decompressBlock(ZSTD_DCtx* dctx, void* dst, size_t dstCapacity, const void* src, size_t srcSize); -ZSTDLIB_STATIC_API size_t ZSTD_insertBlock (ZSTD_DCtx* dctx, const void* blockStart, size_t blockSize); /**< insert uncompressed block into `dctx` history. Useful for multi-blocks decompression. */ - - -#endif /* ZSTD_H_ZSTD_STATIC_LINKING_ONLY */ - -#if defined (__cplusplus) -} -#endif diff --git a/dep/zstd/lib/zstd_errors.h b/dep/zstd/lib/zstd_errors.h deleted file mode 100644 index fa3686b77..000000000 --- a/dep/zstd/lib/zstd_errors.h +++ /dev/null @@ -1,95 +0,0 @@ -/* - * Copyright (c) Yann Collet, Facebook, Inc. - * All rights reserved. - * - * This source code is licensed under both the BSD-style license (found in the - * LICENSE file in the root directory of this source tree) and the GPLv2 (found - * in the COPYING file in the root directory of this source tree). - * You may select, at your option, one of the above-listed licenses. - */ - -#ifndef ZSTD_ERRORS_H_398273423 -#define ZSTD_ERRORS_H_398273423 - -#if defined (__cplusplus) -extern "C" { -#endif - -/*===== dependency =====*/ -#include /* size_t */ - - -/* ===== ZSTDERRORLIB_API : control library symbols visibility ===== */ -#ifndef ZSTDERRORLIB_VISIBILITY -# if defined(__GNUC__) && (__GNUC__ >= 4) -# define ZSTDERRORLIB_VISIBILITY __attribute__ ((visibility ("default"))) -# else -# define ZSTDERRORLIB_VISIBILITY -# endif -#endif -#if defined(ZSTD_DLL_EXPORT) && (ZSTD_DLL_EXPORT==1) -# define ZSTDERRORLIB_API __declspec(dllexport) ZSTDERRORLIB_VISIBILITY -#elif defined(ZSTD_DLL_IMPORT) && (ZSTD_DLL_IMPORT==1) -# define ZSTDERRORLIB_API __declspec(dllimport) ZSTDERRORLIB_VISIBILITY /* It isn't required but allows to generate better code, saving a function pointer load from the IAT and an indirect jump.*/ -#else -# define ZSTDERRORLIB_API ZSTDERRORLIB_VISIBILITY -#endif - -/*-********************************************* - * Error codes list - *-********************************************* - * Error codes _values_ are pinned down since v1.3.1 only. - * Therefore, don't rely on values if you may link to any version < v1.3.1. - * - * Only values < 100 are considered stable. - * - * note 1 : this API shall be used with static linking only. - * dynamic linking is not yet officially supported. - * note 2 : Prefer relying on the enum than on its value whenever possible - * This is the only supported way to use the error list < v1.3.1 - * note 3 : ZSTD_isError() is always correct, whatever the library version. - **********************************************/ -typedef enum { - ZSTD_error_no_error = 0, - ZSTD_error_GENERIC = 1, - ZSTD_error_prefix_unknown = 10, - ZSTD_error_version_unsupported = 12, - ZSTD_error_frameParameter_unsupported = 14, - ZSTD_error_frameParameter_windowTooLarge = 16, - ZSTD_error_corruption_detected = 20, - ZSTD_error_checksum_wrong = 22, - ZSTD_error_dictionary_corrupted = 30, - ZSTD_error_dictionary_wrong = 32, - ZSTD_error_dictionaryCreation_failed = 34, - ZSTD_error_parameter_unsupported = 40, - ZSTD_error_parameter_outOfBound = 42, - ZSTD_error_tableLog_tooLarge = 44, - ZSTD_error_maxSymbolValue_tooLarge = 46, - ZSTD_error_maxSymbolValue_tooSmall = 48, - ZSTD_error_stage_wrong = 60, - ZSTD_error_init_missing = 62, - ZSTD_error_memory_allocation = 64, - ZSTD_error_workSpace_tooSmall= 66, - ZSTD_error_dstSize_tooSmall = 70, - ZSTD_error_srcSize_wrong = 72, - ZSTD_error_dstBuffer_null = 74, - /* following error codes are __NOT STABLE__, they can be removed or changed in future versions */ - ZSTD_error_frameIndex_tooLarge = 100, - ZSTD_error_seekableIO = 102, - ZSTD_error_dstBuffer_wrong = 104, - ZSTD_error_srcBuffer_wrong = 105, - ZSTD_error_maxCode = 120 /* never EVER use this value directly, it can change in future versions! Use ZSTD_isError() instead */ -} ZSTD_ErrorCode; - -/*! ZSTD_getErrorCode() : - convert a `size_t` function result into a `ZSTD_ErrorCode` enum type, - which can be used to compare with enum list published above */ -ZSTDERRORLIB_API ZSTD_ErrorCode ZSTD_getErrorCode(size_t functionResult); -ZSTDERRORLIB_API const char* ZSTD_getErrorString(ZSTD_ErrorCode code); /**< Same as ZSTD_getErrorName, but using a `ZSTD_ErrorCode` enum argument */ - - -#if defined (__cplusplus) -} -#endif - -#endif /* ZSTD_ERRORS_H_398273423 */ diff --git a/dep/zstd/zstd.vcxproj b/dep/zstd/zstd.vcxproj deleted file mode 100644 index c785949fd..000000000 --- a/dep/zstd/zstd.vcxproj +++ /dev/null @@ -1,76 +0,0 @@ - - - - - {73EE0C55-6FFE-44E7-9C12-BAA52434A797} - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - TurnOffAllWarnings - $(ProjectDir)include;$(SolutionDir)dep\zlib\include;%(AdditionalIncludeDirectories) - - - - \ No newline at end of file diff --git a/dep/zstd/zstd.vcxproj.filters b/dep/zstd/zstd.vcxproj.filters deleted file mode 100644 index 349f8109a..000000000 --- a/dep/zstd/zstd.vcxproj.filters +++ /dev/null @@ -1,189 +0,0 @@ - - - - - {085cb7cf-b82e-4fbc-93f1-28fd067a43be} - - - {23a28848-3ccc-47e4-a375-cb97761304c3} - - - {90d9824f-f178-4f36-9b47-9b2471598977} - - - - - common - - - common - - - common - - - common - - - common - - - common - - - common - - - common - - - common - - - common - - - common - - - common - - - common - - - common - - - common - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - decompress - - - decompress - - - decompress - - - - - common - - - common - - - common - - - common - - - common - - - common - - - common - - - common - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - compress - - - decompress - - - decompress - - - decompress - - - decompress - - - \ No newline at end of file