Mercurial > hgrepos > Python2 > PyMuPDF

comparison mupdf-source/thirdparty/libjpeg/change.log @ 2:b50eed0cc0ef upstream

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
ADD: MuPDF v1.26.7: the MuPDF source as downloaded by a default build of PyMuPDF 1.26.4. The directory name has changed: no version number in the expanded directory now.
author Franz Glasner <fzglas.hg@dom66.de>
date Mon, 15 Sep 2025 11:43:07 +0200
parents
children
comparison
equal deleted inserted replaced
1:1d09e1dec1d9 2:b50eed0cc0ef
1 CHANGE LOG for Independent JPEG Group's JPEG software
2
3
4 Version 9f 14-Jan-2024
5 -----------------------
6
7 Add build system for C++Builder/RAD Studio.
8
9 Add build system for Xcode (beside configure).
10
11 Add ARM64EC (Emulation Compatible) platform support in the
12 Visual Studio build.
13
14
15 Version 9e 16-Jan-2022
16 -----------------------
17
18 Include alternate user interface files for cjpeg/djpeg.
19
20 jcparam.c: change default chrominance DC quantization factor
21 for lossless support. Note: Requires rebuild of test images.
22
23 rdgif.c, cderror.h: add sanity check for GIF image dimensions.
24 Thank to Casper Sun for cjpeg potential vulnerability report.
25
26 Add ARM and ARM64 platform support in the Visual Studio build.
27
28
29 Version 9d 12-Jan-2020
30 -----------------------
31
32 Optimize the optimal Huffman code table generation to produce
33 slightly smaller files. Thank to John Korejwa for suggestion.
34 Note: Requires rebuild of testimgp.jpg.
35
36 Decoding Huffman: Use default tables if tables are not defined.
37 Thank to Simone Azzalin for report (Motion JPEG),
38 and to Martin Strunz for hint.
39
40 Add sanity check in optimal Huffman code table generation.
41 Thank to Adam Farley for suggestion.
42
43 rdtarga.c: use read_byte(), with EOF check, instead of getc()
44 in read_*_pixel().
45 Thank to Chijin Zhou for cjpeg potential vulnerability report.
46
47 jmemnobs.c: respect the max_memory_to_use setting in
48 jpeg_mem_available() computation. Thank to Sheng Shu and
49 Dongdong She for djpeg potential vulnerability report.
50
51 jdarith.c, jdhuff.c: avoid left shift of negative value
52 compiler warning in decode_mcu_AC_refine().
53 Thank to Indu Bhagat for suggestion.
54
55 Add x64 (64-bit) platform support, avoid compiler warnings.
56 Thank to Jonathan Potter, Feiyun Wang, and Sheng Shu for suggestion.
57
58 Adjust libjpeg version specification for pkg-config file.
59 Thank to Chen Chen for suggestion.
60
61 Restore GIF read and write support from libjpeg version 6a.
62 Thank to Wolfgang Werner (W.W.) Heinz for suggestion.
63
64 Improve consistency in raw (downsampled) image data processing mode.
65 Thank to Zhongyuan Zhou for hint.
66
67 Avoid out of bounds array read (AC derived table pointers)
68 in start pass in jdhuff.c. Thank to Peng Li for report.
69
70 Improve code sanity (jdhuff.c).
71 Thank to Reza Mirzazade farkhani for reports.
72
73 Add jpegtran -drop option; add options to the crop extension and wipe
74 to fill the extra area with content from the source image region,
75 instead of gray out.
76
77
78 Version 9c 14-Jan-2018
79 -----------------------
80
81 jpegtran: add an option to the -wipe switch to fill the region
82 with the average of adjacent blocks, instead of gray out.
83 Thank to Caitlyn Feddock and Maddie Ziegler for inspiration.
84
85 Make range extension bits adjustable (in jpegint.h).
86 Thank to Robin Watts for suggestion.
87
88 Provide macros for fflush() and ferror() in jinclude.h in order
89 to facilitate adaption by applications using an own FILE class.
90 Thank to Gerhard Huber for suggestion.
91
92 Add libjpeg pkg-config file. Thank to Mark Lavi, Vincent Torri,
93 Patrick McMunn, and Huw Davies for suggestion.
94
95 Add sanity checks in cjpeg image reader modules.
96 Thank to Bingchang, Liu for reports.
97
98
99 Version 9b 17-Jan-2016
100 -----------------------
101
102 Improvements and optimizations in DCT and color calculations.
103 Normalize range limit array composition and access pattern.
104 Thank to Sia Furler and Maddie Ziegler for inspiration.
105
106 Use merged upsample with scaled DCT sizes larger than 8.
107 Thank to Taylor Hatala for inspiration.
108
109 Check for excessive comment lengths in argument parsing in wrjpgcom.c.
110 Thank to Julian Cohen for hint.
111
112 Add makefile.b32 for use with Borland C++ 32-bit (bcc32).
113 Thank to Joe Slater for contribution.
114
115 Document 'f' specifier for jpegtran -crop specification.
116 Thank to Michele Martone for suggestion.
117
118 Use defined value from header instead of hardwired number in rdswitch.c.
119 Thank to Robert Sprowson for hint.
120
121
122 Version 9a 19-Jan-2014
123 -----------------------
124
125 Add support for wide gamut color spaces (JFIF version 2).
126 Improve clarity and accuracy in color conversion modules.
127 Note: Requires rebuild of test images.
128
129 Extend the bit depth support to all values from 8 to 12
130 (BITS_IN_JSAMPLE configuration option in jmorecfg.h).
131 jpegtran now supports N bits sample data precision with all N from 8 to 12
132 in a single instance. Thank to Roland Fassauer for inspiration.
133
134 Try to resolve issues with new boolean type definition.
135 Thank also to v4hn for suggestion.
136
137 Enable option to use default Huffman tables for lossless compression
138 (for hardware solution), and in this case improve lossless RGB compression
139 with reversible color transform. Thank to Benny Alexandar for hint.
140
141 Extend the entropy decoding structure, so that extraneous bytes between
142 compressed scan data and following marker can be reported correctly.
143 Thank to Nigel Tao for hint.
144
145 Add jpegtran -wipe option and extension for -crop.
146 Thank to Andrew Senior, David Clunie, and Josef Schmid for suggestion.
147
148
149 Version 9 13-Jan-2013
150 ----------------------
151
152 Add cjpeg -rgb1 option to create an RGB JPEG file, and insert
153 a simple reversible color transform into the processing which
154 significantly improves the compression.
155 The recommended command for lossless coding of RGB images is now
156 cjpeg -rgb1 -block 1 -arithmetic.
157 As said, this option improves the compression significantly, but
158 the files are not compatible with JPEG decoders prior to IJG v9
159 due to the included color transform.
160 The used color transform and marker signaling is compatible with
161 other JPEG standards (e.g., JPEG-LS part 2).
162
163 Remove the automatic de-ANSI-fication support (Automake 1.12).
164 Thank also to Nitin A Kamble for suggestion.
165
166 Add remark for jpeg_mem_dest() in jdatadst.c.
167 Thank to Elie-Gregoire Khoury for the hint.
168
169 Support files with invalid component identifiers (created
170 by Adobe PDF). Thank to Robin Watts for the suggestion.
171
172 Adapt full buffer case in jcmainct.c for use with scaled DCT.
173 Thank to Sergii Biloshytskyi for the suggestion.
174
175 Add type identifier for declaration of noreturn functions.
176 Thank to Brett L. Moore for the suggestion.
177
178 Correct argument type in format string, avoid compiler warnings.
179 Thank to Vincent Torri for hint.
180
181 Add missing #include directives in configuration checks, avoid
182 configuration errors. Thank to John Spencer for the hint.
183
184
185 Version 8d 15-Jan-2012
186 -----------------------
187
188 Add cjpeg -rgb option to create RGB JPEG files.
189 Using this switch suppresses the conversion from RGB
190 colorspace input to the default YCbCr JPEG colorspace.
191 This feature allows true lossless JPEG coding of RGB color images.
192 The recommended command for this purpose is currently
193 cjpeg -rgb -block 1 -arithmetic.
194 SmartScale capable decoder (introduced with IJG JPEG 8) required.
195 Thank to Michael Koch for the initial suggestion.
196
197 Add option to disable the region adjustment in the transupp crop code.
198 Thank to Jeffrey Friedl for the suggestion.
199
200 Thank to Richard Jones and Edd Dawson for various minor corrections.
201
202 Thank to Akim Demaille for configure.ac cleanup.
203
204
205 Version 8c 16-Jan-2011
206 -----------------------
207
208 Add option to compression library and cjpeg (-block N) to use
209 different DCT block size.
210 All N from 1 to 16 are possible. Default is 8 (baseline format).
211 Larger values produce higher compression,
212 smaller values produce higher quality.
213 SmartScale capable decoder (introduced with IJG JPEG 8) required.
214
215
216 Version 8b 16-May-2010
217 -----------------------
218
219 Repair problem in new memory source manager with corrupt JPEG data.
220 Thank to Ted Campbell and Samuel Chun for the report.
221
222 Repair problem in Makefile.am test target.
223 Thank to anonymous user for the report.
224
225 Support MinGW installation with automatic configure.
226 Thank to Volker Grabsch for the suggestion.
227
228
229 Version 8a 28-Feb-2010
230 -----------------------
231
232 Writing tables-only datastreams via jpeg_write_tables works again.
233
234 Support 32-bit BMPs (RGB image with Alpha channel) for read in cjpeg.
235 Thank to Brett Blackham for the suggestion.
236
237 Improve accuracy in floating point IDCT calculation.
238 Thank to Robert Hooke for the hint.
239
240
241 Version 8 10-Jan-2010
242 ----------------------
243
244 jpegtran now supports the same -scale option as djpeg for "lossless" resize.
245 An implementation of the JPEG SmartScale extension is required for this
246 feature. A (draft) specification of the JPEG SmartScale extension is
247 available as a contributed document at ITU and ISO. Revision 2 or later
248 of the document is required (latest document version is Revision 3).
249 The SmartScale extension will enable more features beside lossless resize
250 in future implementations, as described in the document (new compression
251 options).
252
253 Add sanity check in BMP reader module to avoid cjpeg crash for empty input
254 image (thank to Isaev Ildar of ISP RAS, Moscow, RU for reporting this error).
255
256 Add data source and destination managers for read from and write to
257 memory buffers. New API functions jpeg_mem_src and jpeg_mem_dest.
258 Thank to Roberto Boni from Italy for the suggestion.
259
260
261 Version 7 27-Jun-2009
262 ----------------------
263
264 New scaled DCTs implemented.
265 djpeg now supports scalings N/8 with all N from 1 to 16.
266 cjpeg now supports scalings 8/N with all N from 1 to 16.
267 Scaled DCTs with size larger than 8 are now also used for resolving the
268 common 2x2 chroma subsampling case without additional spatial resampling.
269 Separate spatial resampling for those kind of files is now only necessary
270 for N>8 scaling cases.
271 Furthermore, separate scaled DCT functions are provided for direct resolving
272 of the common asymmetric subsampling cases (2x1 and 1x2) without additional
273 spatial resampling.
274
275 cjpeg -quality option has been extended for support of separate quality
276 settings for luminance and chrominance (or in general, for every provided
277 quantization table slot).
278 New API function jpeg_default_qtables() and q_scale_factor array in library.
279
280 Added -nosmooth option to cjpeg, complementary to djpeg.
281 New variable "do_fancy_downsampling" in library, complement to fancy
282 upsampling. Fancy upsampling now uses direct DCT scaling with sizes
283 larger than 8. The old method is not reversible and has been removed.
284
285 Support arithmetic entropy encoding and decoding.
286 Added files jaricom.c, jcarith.c, jdarith.c.
287
288 Straighten the file structure:
289 Removed files jidctred.c, jcphuff.c, jchuff.h, jdphuff.c, jdhuff.h.
290
291 jpegtran has a new "lossless" cropping feature.
292
293 Implement -perfect option in jpegtran, new API function
294 jtransform_perfect_transform() in transupp. (DP 204_perfect.dpatch)
295
296 Better error messages for jpegtran fopen failure.
297 (DP 203_jpegtran_errmsg.dpatch)
298
299 Fix byte order issue with 16bit PPM/PGM files in rdppm.c/wrppm.c:
300 according to Netpbm, the de facto standard implementation of the PNM formats,
301 the most significant byte is first. (DP 203_rdppm.dpatch)
302
303 Add -raw option to rdjpgcom not to mangle the output.
304 (DP 205_rdjpgcom_raw.dpatch)
305
306 Make rdjpgcom locale aware. (DP 201_rdjpgcom_locale.dpatch)
307
308 Add extern "C" to jpeglib.h.
309 This avoids the need to put extern "C" { ... } around #include "jpeglib.h"
310 in your C++ application. Defining the symbol DONT_USE_EXTERN_C in the
311 configuration prevents this. (DP 202_jpeglib.h_c++.dpatch)
312
313
314 Version 6b 27-Mar-1998
315 -----------------------
316
317 jpegtran has new features for lossless image transformations (rotation
318 and flipping) as well as "lossless" reduction to grayscale.
319
320 jpegtran now copies comments by default; it has a -copy switch to enable
321 copying all APPn blocks as well, or to suppress comments. (Formerly it
322 always suppressed comments and APPn blocks.) jpegtran now also preserves
323 JFIF version and resolution information.
324
325 New decompressor library feature: COM and APPn markers found in the input
326 file can be saved in memory for later use by the application. (Before,
327 you had to code this up yourself with a custom marker processor.)
328
329 There is an unused field "void * client_data" now in compress and decompress
330 parameter structs; this may be useful in some applications.
331
332 JFIF version number information is now saved by the decoder and accepted by
333 the encoder. jpegtran uses this to copy the source file's version number,
334 to ensure "jpegtran -copy all" won't create bogus files that contain JFXX
335 extensions but claim to be version 1.01. Applications that generate their
336 own JFXX extension markers also (finally) have a supported way to cause the
337 encoder to emit JFIF version number 1.02.
338
339 djpeg's trace mode reports JFIF 1.02 thumbnail images as such, rather
340 than as unknown APP0 markers.
341
342 In -verbose mode, djpeg and rdjpgcom will try to print the contents of
343 APP12 markers as text. Some digital cameras store useful text information
344 in APP12 markers.
345
346 Handling of truncated data streams is more robust: blocks beyond the one in
347 which the error occurs will be output as uniform gray, or left unchanged
348 if decoding a progressive JPEG. The appearance no longer depends on the
349 Huffman tables being used.
350
351 Huffman tables are checked for validity much more carefully than before.
352
353 To avoid the Unisys LZW patent, djpeg's GIF output capability has been
354 changed to produce "uncompressed GIFs", and cjpeg's GIF input capability
355 has been removed altogether. We're not happy about it either, but there
356 seems to be no good alternative.
357
358 The configure script now supports building libjpeg as a shared library
359 on many flavors of Unix (all the ones that GNU libtool knows how to
360 build shared libraries for). Use "./configure --enable-shared" to
361 try this out.
362
363 New jconfig file and makefiles for Microsoft Visual C++ and Developer Studio.
364 Also, a jconfig file and a build script for Metrowerks CodeWarrior
365 on Apple Macintosh. makefile.dj has been updated for DJGPP v2, and there
366 are miscellaneous other minor improvements in the makefiles.
367
368 jmemmac.c now knows how to create temporary files following Mac System 7
369 conventions.
370
371 djpeg's -map switch is now able to read raw-format PPM files reliably.
372
373 cjpeg -progressive -restart no longer generates any unnecessary DRI markers.
374
375 Multiple calls to jpeg_simple_progression for a single JPEG object
376 no longer leak memory.
377
378
379 Version 6a 7-Feb-96
380 --------------------
381
382 Library initialization sequence modified to detect version mismatches
383 and struct field packing mismatches between library and calling application.
384 This change requires applications to be recompiled, but does not require
385 any application source code change.
386
387 All routine declarations changed to the style "GLOBAL(type) name ...",
388 that is, GLOBAL, LOCAL, METHODDEF, EXTERN are now macros taking the
389 routine's return type as an argument. This makes it possible to add
390 Microsoft-style linkage keywords to all the routines by changing just
391 these macros. Note that any application code that was using these macros
392 will have to be changed.
393
394 DCT coefficient quantization tables are now stored in normal array order
395 rather than zigzag order. Application code that calls jpeg_add_quant_table,
396 or otherwise manipulates quantization tables directly, will need to be
397 changed. If you need to make such code work with either older or newer
398 versions of the library, a test like "#if JPEG_LIB_VERSION >= 61" is
399 recommended.
400
401 djpeg's trace capability now dumps DQT tables in natural order, not zigzag
402 order. This allows the trace output to be made into a "-qtables" file
403 more easily.
404
405 New system-dependent memory manager module for use on Apple Macintosh.
406
407 Fix bug in cjpeg's -smooth option: last one or two scanlines would be
408 duplicates of the prior line unless the image height mod 16 was 1 or 2.
409
410 Repair minor problems in VMS, BCC, MC6 makefiles.
411
412 New configure script based on latest GNU Autoconf.
413
414 Correct the list of include files needed by MetroWerks C for ccommand().
415
416 Numerous small documentation updates.
417
418
419 Version 6 2-Aug-95
420 -------------------
421
422 Progressive JPEG support: library can read and write full progressive JPEG
423 files. A "buffered image" mode supports incremental decoding for on-the-fly
424 display of progressive images. Simply recompiling an existing IJG-v5-based
425 decoder with v6 should allow it to read progressive files, though of course
426 without any special progressive display.
427
428 New "jpegtran" application performs lossless transcoding between different
429 JPEG formats; primarily, it can be used to convert baseline to progressive
430 JPEG and vice versa. In support of jpegtran, the library now allows lossless
431 reading and writing of JPEG files as DCT coefficient arrays. This ability
432 may be of use in other applications.
433
434 Notes for programmers:
435 * We changed jpeg_start_decompress() to be able to suspend; this makes all
436 decoding modes available to suspending-input applications. However,
437 existing applications that use suspending input will need to be changed
438 to check the return value from jpeg_start_decompress(). You don't need to
439 do anything if you don't use a suspending data source.
440 * We changed the interface to the virtual array routines: access_virt_array
441 routines now take a count of the number of rows to access this time. The
442 last parameter to request_virt_array routines is now interpreted as the
443 maximum number of rows that may be accessed at once, but not necessarily
444 the height of every access.
445
446
447 Version 5b 15-Mar-95
448 ---------------------
449
450 Correct bugs with grayscale images having v_samp_factor > 1.
451
452 jpeg_write_raw_data() now supports output suspension.
453
454 Correct bugs in "configure" script for case of compiling in
455 a directory other than the one containing the source files.
456
457 Repair bug in jquant1.c: sometimes didn't use as many colors as it could.
458
459 Borland C makefile and jconfig file work under either MS-DOS or OS/2.
460
461 Miscellaneous improvements to documentation.
462
463
464 Version 5a 7-Dec-94
465 --------------------
466
467 Changed color conversion roundoff behavior so that grayscale values are
468 represented exactly. (This causes test image files to change.)
469
470 Make ordered dither use 16x16 instead of 4x4 pattern for a small quality
471 improvement.
472
473 New configure script based on latest GNU Autoconf.
474 Fix configure script to handle CFLAGS correctly.
475 Rename *.auto files to *.cfg, so that configure script still works if
476 file names have been truncated for DOS.
477
478 Fix bug in rdbmp.c: didn't allow for extra data between header and image.
479
480 Modify rdppm.c/wrppm.c to handle 2-byte raw PPM/PGM formats for 12-bit data.
481
482 Fix several bugs in rdrle.c.
483
484 NEED_SHORT_EXTERNAL_NAMES option was broken.
485
486 Revise jerror.h/jerror.c for more flexibility in message table.
487
488 Repair oversight in jmemname.c NO_MKTEMP case: file could be there
489 but unreadable.
490
491
492 Version 5 24-Sep-94
493 --------------------
494
495 Version 5 represents a nearly complete redesign and rewrite of the IJG
496 software. Major user-visible changes include:
497 * Automatic configuration simplifies installation for most Unix systems.
498 * A range of speed vs. image quality tradeoffs are supported.
499 This includes resizing of an image during decompression: scaling down
500 by a factor of 1/2, 1/4, or 1/8 is handled very efficiently.
501 * New programs rdjpgcom and wrjpgcom allow insertion and extraction
502 of text comments in a JPEG file.
503
504 The application programmer's interface to the library has changed completely.
505 Notable improvements include:
506 * We have eliminated the use of callback routines for handling the
507 uncompressed image data. The application now sees the library as a
508 set of routines that it calls to read or write image data on a
509 scanline-by-scanline basis.
510 * The application image data is represented in a conventional interleaved-
511 pixel format, rather than as a separate array for each color channel.
512 This can save a copying step in many programs.
513 * The handling of compressed data has been cleaned up: the application can
514 supply routines to source or sink the compressed data. It is possible to
515 suspend processing on source/sink buffer overrun, although this is not
516 supported in all operating modes.
517 * All static state has been eliminated from the library, so that multiple
518 instances of compression or decompression can be active concurrently.
519 * JPEG abbreviated datastream formats are supported, ie, quantization and
520 Huffman tables can be stored separately from the image data.
521 * And not only that, but the documentation of the library has improved
522 considerably!
523
524
525 The last widely used release before the version 5 rewrite was version 4A of
526 18-Feb-93. Change logs before that point have been discarded, since they
527 are not of much interest after the rewrite.