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

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