reactos/dll/3rdparty/libjpeg/change.log

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