dll/3rdparty/libjpeg/change.log

   1 CHANGE LOG for Independent JPEG Group's JPEG software
   2
   3
   4 Version 9a  19-Jan-2014
   5 -----------------------
   6
   7 Add support for wide gamut color spaces (JFIF version 2).
   8 Improve clarity and accuracy in color conversion modules.
   9 Note: Requires rebuild of test images.
  10
  11 Extend the bit depth support to all values from 8 to 12
  12 (BITS_IN_JSAMPLE configuration option in jmorecfg.h).
  13 jpegtran now supports N bits sample data precision with all N from 8 to 12
  14 in a single instance.  Thank to Roland Fassauer for inspiration.
  15
  16 Try to resolve issues with new boolean type definition.
  17 Thank also to v4hn for suggestion.
  18
  19 Enable option to use default Huffman tables for lossless compression
  20 (for hardware solution), and in this case improve lossless RGB compression
  21 with reversible color transform.  Thank to Benny Alexandar for hint.
  22
  23 Extend the entropy decoding structure, so that extraneous bytes between
  24 compressed scan data and following marker can be reported correctly.
  25 Thank to Nigel Tao for hint.
  26
  27 Add jpegtran -wipe option and extension for -crop.
  28 Thank to Andrew Senior, David Clunie, and Josef Schmid for suggestion.
  29
  30
  31 Version 9  13-Jan-2013
  32 ----------------------
  33
  34 Add cjpeg -rgb1 option to create an RGB JPEG file, and insert
  35 a simple reversible color transform into the processing which
  36 significantly improves the compression.
  37 The recommended command for lossless coding of RGB images is now
  38 cjpeg -rgb1 -block 1 -arithmetic.
  39 As said, this option improves the compression significantly, but
  40 the files are not compatible with JPEG decoders prior to IJG v9
  41 due to the included color transform.
  42 The used color transform and marker signaling is compatible with
  43 other JPEG standards (e.g., JPEG-LS part 2).
  44
  45 Remove the automatic de-ANSI-fication support (Automake 1.12).
  46 Thank also to Nitin A Kamble for suggestion.
  47
  48 Add remark for jpeg_mem_dest() in jdatadst.c.
  49 Thank to Elie-Gregoire Khoury for the hint.
  50
  51 Support files with invalid component identifiers (created
  52 by Adobe PDF).  Thank to Robin Watts for the suggestion.
  53
  54 Adapt full buffer case in jcmainct.c for use with scaled DCT.
  55 Thank to Sergii Biloshytskyi for the suggestion.
  56
  57 Add type identifier for declaration of noreturn functions.
  58 Thank to Brett L. Moore for the suggestion.
  59
  60 Correct argument type in format string, avoid compiler warnings.
  61 Thank to Vincent Torri for hint.
  62
  63 Add missing #include directives in configuration checks, avoid
  64 configuration errors.  Thank to John Spencer for the hint.
  65
  66
  67 Version 8d  15-Jan-2012
  68 -----------------------
  69
  70 Add cjpeg -rgb option to create RGB JPEG files.
  71 Using this switch suppresses the conversion from RGB
  72 colorspace input to the default YCbCr JPEG colorspace.
  73 This feature allows true lossless JPEG coding of RGB color images.
  74 The recommended command for this purpose is currently
  75 cjpeg -rgb -block 1 -arithmetic.
  76 SmartScale capable decoder (introduced with IJG JPEG 8) required.
  77 Thank to Michael Koch for the initial suggestion.
  78
  79 Add option to disable the region adjustment in the transupp crop code.
  80 Thank to Jeffrey Friedl for the suggestion.
  81
  82 Thank to Richard Jones and Edd Dawson for various minor corrections.
  83
  84 Thank to Akim Demaille for configure.ac cleanup.
  85
  86
  87 Version 8c  16-Jan-2011
  88 -----------------------
  89
  90 Add option to compression library and cjpeg (-block N) to use
  91 different DCT block size.
  92 All N from 1 to 16 are possible.  Default is 8 (baseline format).
  93 Larger values produce higher compression,
  94 smaller values produce higher quality.
  95 SmartScale capable decoder (introduced with IJG JPEG 8) required.
  96
  97
  98 Version 8b  16-May-2010
  99 -----------------------
 100
 101 Repair problem in new memory source manager with corrupt JPEG data.
 102 Thank to Ted Campbell and Samuel Chun for the report.
 103
 104 Repair problem in Makefile.am test target.
 105 Thank to anonymous user for the report.
 106
 107 Support MinGW installation with automatic configure.
 108 Thank to Volker Grabsch for the suggestion.
 109
 110
 111 Version 8a  28-Feb-2010
 112 -----------------------
 113
 114 Writing tables-only datastreams via jpeg_write_tables works again.
 115
 116 Support 32-bit BMPs (RGB image with Alpha channel) for read in cjpeg.
 117 Thank to Brett Blackham for the suggestion.
 118
 119 Improve accuracy in floating point IDCT calculation.
 120 Thank to Robert Hooke for the hint.
 121
 122
 123 Version 8  10-Jan-2010
 124 ----------------------
 125
 126 jpegtran now supports the same -scale option as djpeg for "lossless" resize.
 127 An implementation of the JPEG SmartScale extension is required for this
 128 feature.  A (draft) specification of the JPEG SmartScale extension is
 129 available as a contributed document at ITU and ISO.  Revision 2 or later
 130 of the document is required (latest document version is Revision 3).
 131 The SmartScale extension will enable more features beside lossless resize
 132 in future implementations, as described in the document (new compression
 133 options).
 134
 135 Add sanity check in BMP reader module to avoid cjpeg crash for empty input
 136 image (thank to Isaev Ildar of ISP RAS, Moscow, RU for reporting this error).
 137
 138 Add data source and destination managers for read from and write to
 139 memory buffers.  New API functions jpeg_mem_src and jpeg_mem_dest.
 140 Thank to Roberto Boni from Italy for the suggestion.
 141
 142
 143 Version 7  27-Jun-2009
 144 ----------------------
 145
 146 New scaled DCTs implemented.
 147 djpeg now supports scalings N/8 with all N from 1 to 16.
 148 cjpeg now supports scalings 8/N with all N from 1 to 16.
 149 Scaled DCTs with size larger than 8 are now also used for resolving the
 150 common 2x2 chroma subsampling case without additional spatial resampling.
 151 Separate spatial resampling for those kind of files is now only necessary
 152 for N>8 scaling cases.
 153 Furthermore, separate scaled DCT functions are provided for direct resolving
 154 of the common asymmetric subsampling cases (2x1 and 1x2) without additional
 155 spatial resampling.
 156
 157 cjpeg -quality option has been extended for support of separate quality
 158 settings for luminance and chrominance (or in general, for every provided
 159 quantization table slot).
 160 New API function jpeg_default_qtables() and q_scale_factor array in library.
 161
 162 Added -nosmooth option to cjpeg, complementary to djpeg.
 163 New variable "do_fancy_downsampling" in library, complement to fancy
 164 upsampling.  Fancy upsampling now uses direct DCT scaling with sizes
 165 larger than 8.  The old method is not reversible and has been removed.
 166
 167 Support arithmetic entropy encoding and decoding.
 168 Added files jaricom.c, jcarith.c, jdarith.c.
 169
 170 Straighten the file structure:
 171 Removed files jidctred.c, jcphuff.c, jchuff.h, jdphuff.c, jdhuff.h.
 172
 173 jpegtran has a new "lossless" cropping feature.
 174
 175 Implement -perfect option in jpegtran, new API function
 176 jtransform_perfect_transform() in transupp. (DP 204_perfect.dpatch)
 177
 178 Better error messages for jpegtran fopen failure.
 179 (DP 203_jpegtran_errmsg.dpatch)
 180
 181 Fix byte order issue with 16bit PPM/PGM files in rdppm.c/wrppm.c:
 182 according to Netpbm, the de facto standard implementation of the PNM formats,
 183 the most significant byte is first. (DP 203_rdppm.dpatch)
 184
 185 Add -raw option to rdjpgcom not to mangle the output.
 186 (DP 205_rdjpgcom_raw.dpatch)
 187
 188 Make rdjpgcom locale aware. (DP 201_rdjpgcom_locale.dpatch)
 189
 190 Add extern "C" to jpeglib.h.
 191 This avoids the need to put extern "C" { ... } around #include "jpeglib.h"
 192 in your C++ application.  Defining the symbol DONT_USE_EXTERN_C in the
 193 configuration prevents this. (DP 202_jpeglib.h_c++.dpatch)
 194
 195
 196 Version 6b  27-Mar-1998
 197 -----------------------
 198
 199 jpegtran has new features for lossless image transformations (rotation
 200 and flipping) as well as "lossless" reduction to grayscale.
 201
 202 jpegtran now copies comments by default; it has a -copy switch to enable
 203 copying all APPn blocks as well, or to suppress comments.  (Formerly it
 204 always suppressed comments and APPn blocks.)  jpegtran now also preserves
 205 JFIF version and resolution information.
 206
 207 New decompressor library feature: COM and APPn markers found in the input
 208 file can be saved in memory for later use by the application.  (Before,
 209 you had to code this up yourself with a custom marker processor.)
 210
 211 There is an unused field "void * client_data" now in compress and decompress
 212 parameter structs; this may be useful in some applications.
 213
 214 JFIF version number information is now saved by the decoder and accepted by
 215 the encoder.  jpegtran uses this to copy the source file's version number,
 216 to ensure "jpegtran -copy all" won't create bogus files that contain JFXX
 217 extensions but claim to be version 1.01.  Applications that generate their
 218 own JFXX extension markers also (finally) have a supported way to cause the
 219 encoder to emit JFIF version number 1.02.
 220
 221 djpeg's trace mode reports JFIF 1.02 thumbnail images as such, rather
 222 than as unknown APP0 markers.
 223
 224 In -verbose mode, djpeg and rdjpgcom will try to print the contents of
 225 APP12 markers as text.  Some digital cameras store useful text information
 226 in APP12 markers.
 227
 228 Handling of truncated data streams is more robust: blocks beyond the one in
 229 which the error occurs will be output as uniform gray, or left unchanged
 230 if decoding a progressive JPEG.  The appearance no longer depends on the
 231 Huffman tables being used.
 232
 233 Huffman tables are checked for validity much more carefully than before.
 234
 235 To avoid the Unisys LZW patent, djpeg's GIF output capability has been
 236 changed to produce "uncompressed GIFs", and cjpeg's GIF input capability
 237 has been removed altogether.  We're not happy about it either, but there
 238 seems to be no good alternative.
 239
 240 The configure script now supports building libjpeg as a shared library
 241 on many flavors of Unix (all the ones that GNU libtool knows how to
 242 build shared libraries for).  Use "./configure --enable-shared" to
 243 try this out.
 244
 245 New jconfig file and makefiles for Microsoft Visual C++ and Developer Studio.
 246 Also, a jconfig file and a build script for Metrowerks CodeWarrior
 247 on Apple Macintosh.  makefile.dj has been updated for DJGPP v2, and there
 248 are miscellaneous other minor improvements in the makefiles.
 249
 250 jmemmac.c now knows how to create temporary files following Mac System 7
 251 conventions.
 252
 253 djpeg's -map switch is now able to read raw-format PPM files reliably.
 254
 255 cjpeg -progressive -restart no longer generates any unnecessary DRI markers.
 256
 257 Multiple calls to jpeg_simple_progression for a single JPEG object
 258 no longer leak memory.
 259
 260
 261 Version 6a  7-Feb-96
 262 --------------------
 263
 264 Library initialization sequence modified to detect version mismatches
 265 and struct field packing mismatches between library and calling application.
 266 This change requires applications to be recompiled, but does not require
 267 any application source code change.
 268
 269 All routine declarations changed to the style "GLOBAL(type) name ...",
 270 that is, GLOBAL, LOCAL, METHODDEF, EXTERN are now macros taking the
 271 routine's return type as an argument.  This makes it possible to add
 272 Microsoft-style linkage keywords to all the routines by changing just
 273 these macros.  Note that any application code that was using these macros
 274 will have to be changed.
 275
 276 DCT coefficient quantization tables are now stored in normal array order
 277 rather than zigzag order.  Application code that calls jpeg_add_quant_table,
 278 or otherwise manipulates quantization tables directly, will need to be
 279 changed.  If you need to make such code work with either older or newer
 280 versions of the library, a test like "#if JPEG_LIB_VERSION >= 61" is
 281 recommended.
 282
 283 djpeg's trace capability now dumps DQT tables in natural order, not zigzag
 284 order.  This allows the trace output to be made into a "-qtables" file
 285 more easily.
 286
 287 New system-dependent memory manager module for use on Apple Macintosh.
 288
 289 Fix bug in cjpeg's -smooth option: last one or two scanlines would be
 290 duplicates of the prior line unless the image height mod 16 was 1 or 2.
 291
 292 Repair minor problems in VMS, BCC, MC6 makefiles.
 293
 294 New configure script based on latest GNU Autoconf.
 295
 296 Correct the list of include files needed by MetroWerks C for ccommand().
 297
 298 Numerous small documentation updates.
 299
 300
 301 Version 6  2-Aug-95
 302 -------------------
 303
 304 Progressive JPEG support: library can read and write full progressive JPEG
 305 files.  A "buffered image" mode supports incremental decoding for on-the-fly
 306 display of progressive images.  Simply recompiling an existing IJG-v5-based
 307 decoder with v6 should allow it to read progressive files, though of course
 308 without any special progressive display.
 309
 310 New "jpegtran" application performs lossless transcoding between different
 311 JPEG formats; primarily, it can be used to convert baseline to progressive
 312 JPEG and vice versa.  In support of jpegtran, the library now allows lossless
 313 reading and writing of JPEG files as DCT coefficient arrays.  This ability
 314 may be of use in other applications.
 315
 316 Notes for programmers:
 317 * We changed jpeg_start_decompress() to be able to suspend; this makes all
 318 decoding modes available to suspending-input applications.  However,
 319 existing applications that use suspending input will need to be changed
 320 to check the return value from jpeg_start_decompress().  You don't need to
 321 do anything if you don't use a suspending data source.
 322 * We changed the interface to the virtual array routines: access_virt_array
 323 routines now take a count of the number of rows to access this time.  The
 324 last parameter to request_virt_array routines is now interpreted as the
 325 maximum number of rows that may be accessed at once, but not necessarily
 326 the height of every access.
 327
 328
 329 Version 5b  15-Mar-95
 330 ---------------------
 331
 332 Correct bugs with grayscale images having v_samp_factor > 1.
 333
 334 jpeg_write_raw_data() now supports output suspension.
 335
 336 Correct bugs in "configure" script for case of compiling in
 337 a directory other than the one containing the source files.
 338
 339 Repair bug in jquant1.c: sometimes didn't use as many colors as it could.
 340
 341 Borland C makefile and jconfig file work under either MS-DOS or OS/2.
 342
 343 Miscellaneous improvements to documentation.
 344
 345
 346 Version 5a  7-Dec-94
 347 --------------------
 348
 349 Changed color conversion roundoff behavior so that grayscale values are
 350 represented exactly.  (This causes test image files to change.)
 351
 352 Make ordered dither use 16x16 instead of 4x4 pattern for a small quality
 353 improvement.
 354
 355 New configure script based on latest GNU Autoconf.
 356 Fix configure script to handle CFLAGS correctly.
 357 Rename *.auto files to *.cfg, so that configure script still works if
 358 file names have been truncated for DOS.
 359
 360 Fix bug in rdbmp.c: didn't allow for extra data between header and image.
 361
 362 Modify rdppm.c/wrppm.c to handle 2-byte raw PPM/PGM formats for 12-bit data.
 363
 364 Fix several bugs in rdrle.c.
 365
 366 NEED_SHORT_EXTERNAL_NAMES option was broken.
 367
 368 Revise jerror.h/jerror.c for more flexibility in message table.
 369
 370 Repair oversight in jmemname.c NO_MKTEMP case: file could be there
 371 but unreadable.
 372
 373
 374 Version 5  24-Sep-94
 375 --------------------
 376
 377 Version 5 represents a nearly complete redesign and rewrite of the IJG
 378 software.  Major user-visible changes include:
 379   * Automatic configuration simplifies installation for most Unix systems.
 380   * A range of speed vs. image quality tradeoffs are supported.
 381     This includes resizing of an image during decompression: scaling down
 382     by a factor of 1/2, 1/4, or 1/8 is handled very efficiently.
 383   * New programs rdjpgcom and wrjpgcom allow insertion and extraction
 384     of text comments in a JPEG file.
 385
 386 The application programmer's interface to the library has changed completely.
 387 Notable improvements include:
 388   * We have eliminated the use of callback routines for handling the
 389     uncompressed image data.  The application now sees the library as a
 390     set of routines that it calls to read or write image data on a
 391     scanline-by-scanline basis.
 392   * The application image data is represented in a conventional interleaved-
 393     pixel format, rather than as a separate array for each color channel.
 394     This can save a copying step in many programs.
 395   * The handling of compressed data has been cleaned up: the application can
 396     supply routines to source or sink the compressed data.  It is possible to
 397     suspend processing on source/sink buffer overrun, although this is not
 398     supported in all operating modes.
 399   * All static state has been eliminated from the library, so that multiple
 400     instances of compression or decompression can be active concurrently.
 401   * JPEG abbreviated datastream formats are supported, ie, quantization and
 402     Huffman tables can be stored separately from the image data.
 403   * And not only that, but the documentation of the library has improved
 404     considerably!
 405
 406
 407 The last widely used release before the version 5 rewrite was version 4A of
 408 18-Feb-93.  Change logs before that point have been discarded, since they
 409 are not of much interest after the rewrite.