source: rtems-graphics-toolkit/jpeg-8d/change.log @ 86b99f7

Last change on this file since 86b99f7 was 86b99f7, checked in by Alexandru-Sever Horin <alex.sever.h@…>, on Aug 1, 2012 at 10:40:32 PM

Added jpeg-8d version. Made modifications to compile for RTEMS, without man or binaries

  • Property mode set to 100644
File size: 13.8 KB
Line 
1CHANGE LOG for Independent JPEG Group's JPEG software
2
3
4Version 8d  15-Jan-2012
5-----------------------
6
7Add cjpeg -rgb option to create RGB JPEG files.
8Using this switch suppresses the conversion from RGB
9colorspace input to the default YCbCr JPEG colorspace.
10This feature allows true lossless JPEG coding of RGB color images.
11The recommended command for this purpose is currently
12cjpeg -rgb -block 1 -arithmetic.
13SmartScale capable decoder (introduced with IJG JPEG 8) required.
14Thank to Michael Koch for the initial suggestion.
15
16Add option to disable the region adjustment in the transupp crop code.
17Thank to Jeffrey Friedl for the suggestion.
18
19Thank to Richard Jones and Edd Dawson for various minor corrections.
20
21Thank to Akim Demaille for configure.ac cleanup.
22
23
24Version 8c  16-Jan-2011
25-----------------------
26
27Add option to compression library and cjpeg (-block N) to use
28different DCT block size.
29All N from 1 to 16 are possible.  Default is 8 (baseline format).
30Larger values produce higher compression,
31smaller values produce higher quality.
32SmartScale capable decoder (introduced with IJG JPEG 8) required.
33
34
35Version 8b  16-May-2010
36-----------------------
37
38Repair problem in new memory source manager with corrupt JPEG data.
39Thank to Ted Campbell and Samuel Chun for the report.
40
41Repair problem in Makefile.am test target.
42Thank to anonymous user for the report.
43
44Support MinGW installation with automatic configure.
45Thank to Volker Grabsch for the suggestion.
46
47
48Version 8a  28-Feb-2010
49-----------------------
50
51Writing tables-only datastreams via jpeg_write_tables works again.
52
53Support 32-bit BMPs (RGB image with Alpha channel) for read in cjpeg.
54Thank to Brett Blackham for the suggestion.
55
56Improve accuracy in floating point IDCT calculation.
57Thank to Robert Hooke for the hint.
58
59
60Version 8  10-Jan-2010
61----------------------
62
63jpegtran now supports the same -scale option as djpeg for "lossless" resize.
64An implementation of the JPEG SmartScale extension is required for this
65feature.  A (draft) specification of the JPEG SmartScale extension is
66available as a contributed document at ITU and ISO.  Revision 2 or later
67of the document is required (latest document version is Revision 3).
68The SmartScale extension will enable more features beside lossless resize
69in future implementations, as described in the document (new compression
70options).
71
72Add sanity check in BMP reader module to avoid cjpeg crash for empty input
73image (thank to Isaev Ildar of ISP RAS, Moscow, RU for reporting this error).
74
75Add data source and destination managers for read from and write to
76memory buffers.  New API functions jpeg_mem_src and jpeg_mem_dest.
77Thank to Roberto Boni from Italy for the suggestion.
78
79
80Version 7  27-Jun-2009
81----------------------
82
83New scaled DCTs implemented.
84djpeg now supports scalings N/8 with all N from 1 to 16.
85cjpeg now supports scalings 8/N with all N from 1 to 16.
86Scaled DCTs with size larger than 8 are now also used for resolving the
87common 2x2 chroma subsampling case without additional spatial resampling.
88Separate spatial resampling for those kind of files is now only necessary
89for N>8 scaling cases.
90Furthermore, separate scaled DCT functions are provided for direct resolving
91of the common asymmetric subsampling cases (2x1 and 1x2) without additional
92spatial resampling.
93
94cjpeg -quality option has been extended for support of separate quality
95settings for luminance and chrominance (or in general, for every provided
96quantization table slot).
97New API function jpeg_default_qtables() and q_scale_factor array in library.
98
99Added -nosmooth option to cjpeg, complementary to djpeg.
100New variable "do_fancy_downsampling" in library, complement to fancy
101upsampling.  Fancy upsampling now uses direct DCT scaling with sizes
102larger than 8.  The old method is not reversible and has been removed.
103
104Support arithmetic entropy encoding and decoding.
105Added files jaricom.c, jcarith.c, jdarith.c.
106
107Straighten the file structure:
108Removed files jidctred.c, jcphuff.c, jchuff.h, jdphuff.c, jdhuff.h.
109
110jpegtran has a new "lossless" cropping feature.
111
112Implement -perfect option in jpegtran, new API function
113jtransform_perfect_transform() in transupp. (DP 204_perfect.dpatch)
114
115Better error messages for jpegtran fopen failure.
116(DP 203_jpegtran_errmsg.dpatch)
117
118Fix byte order issue with 16bit PPM/PGM files in rdppm.c/wrppm.c:
119according to Netpbm, the de facto standard implementation of the PNM formats,
120the most significant byte is first. (DP 203_rdppm.dpatch)
121
122Add -raw option to rdjpgcom not to mangle the output.
123(DP 205_rdjpgcom_raw.dpatch)
124
125Make rdjpgcom locale aware. (DP 201_rdjpgcom_locale.dpatch)
126
127Add extern "C" to jpeglib.h.
128This avoids the need to put extern "C" { ... } around #include "jpeglib.h"
129in your C++ application.  Defining the symbol DONT_USE_EXTERN_C in the
130configuration prevents this. (DP 202_jpeglib.h_c++.dpatch)
131
132
133Version 6b  27-Mar-1998
134-----------------------
135
136jpegtran has new features for lossless image transformations (rotation
137and flipping) as well as "lossless" reduction to grayscale.
138
139jpegtran now copies comments by default; it has a -copy switch to enable
140copying all APPn blocks as well, or to suppress comments.  (Formerly it
141always suppressed comments and APPn blocks.)  jpegtran now also preserves
142JFIF version and resolution information.
143
144New decompressor library feature: COM and APPn markers found in the input
145file can be saved in memory for later use by the application.  (Before,
146you had to code this up yourself with a custom marker processor.)
147
148There is an unused field "void * client_data" now in compress and decompress
149parameter structs; this may be useful in some applications.
150
151JFIF version number information is now saved by the decoder and accepted by
152the encoder.  jpegtran uses this to copy the source file's version number,
153to ensure "jpegtran -copy all" won't create bogus files that contain JFXX
154extensions but claim to be version 1.01.  Applications that generate their
155own JFXX extension markers also (finally) have a supported way to cause the
156encoder to emit JFIF version number 1.02.
157
158djpeg's trace mode reports JFIF 1.02 thumbnail images as such, rather
159than as unknown APP0 markers.
160
161In -verbose mode, djpeg and rdjpgcom will try to print the contents of
162APP12 markers as text.  Some digital cameras store useful text information
163in APP12 markers.
164
165Handling of truncated data streams is more robust: blocks beyond the one in
166which the error occurs will be output as uniform gray, or left unchanged
167if decoding a progressive JPEG.  The appearance no longer depends on the
168Huffman tables being used.
169
170Huffman tables are checked for validity much more carefully than before.
171
172To avoid the Unisys LZW patent, djpeg's GIF output capability has been
173changed to produce "uncompressed GIFs", and cjpeg's GIF input capability
174has been removed altogether.  We're not happy about it either, but there
175seems to be no good alternative.
176
177The configure script now supports building libjpeg as a shared library
178on many flavors of Unix (all the ones that GNU libtool knows how to
179build shared libraries for).  Use "./configure --enable-shared" to
180try this out.
181
182New jconfig file and makefiles for Microsoft Visual C++ and Developer Studio.
183Also, a jconfig file and a build script for Metrowerks CodeWarrior
184on Apple Macintosh.  makefile.dj has been updated for DJGPP v2, and there
185are miscellaneous other minor improvements in the makefiles.
186
187jmemmac.c now knows how to create temporary files following Mac System 7
188conventions.
189
190djpeg's -map switch is now able to read raw-format PPM files reliably.
191
192cjpeg -progressive -restart no longer generates any unnecessary DRI markers.
193
194Multiple calls to jpeg_simple_progression for a single JPEG object
195no longer leak memory.
196
197
198Version 6a  7-Feb-96
199--------------------
200
201Library initialization sequence modified to detect version mismatches
202and struct field packing mismatches between library and calling application.
203This change requires applications to be recompiled, but does not require
204any application source code change.
205
206All routine declarations changed to the style "GLOBAL(type) name ...",
207that is, GLOBAL, LOCAL, METHODDEF, EXTERN are now macros taking the
208routine's return type as an argument.  This makes it possible to add
209Microsoft-style linkage keywords to all the routines by changing just
210these macros.  Note that any application code that was using these macros
211will have to be changed.
212
213DCT coefficient quantization tables are now stored in normal array order
214rather than zigzag order.  Application code that calls jpeg_add_quant_table,
215or otherwise manipulates quantization tables directly, will need to be
216changed.  If you need to make such code work with either older or newer
217versions of the library, a test like "#if JPEG_LIB_VERSION >= 61" is
218recommended.
219
220djpeg's trace capability now dumps DQT tables in natural order, not zigzag
221order.  This allows the trace output to be made into a "-qtables" file
222more easily.
223
224New system-dependent memory manager module for use on Apple Macintosh.
225
226Fix bug in cjpeg's -smooth option: last one or two scanlines would be
227duplicates of the prior line unless the image height mod 16 was 1 or 2.
228
229Repair minor problems in VMS, BCC, MC6 makefiles.
230
231New configure script based on latest GNU Autoconf.
232
233Correct the list of include files needed by MetroWerks C for ccommand().
234
235Numerous small documentation updates.
236
237
238Version 6  2-Aug-95
239-------------------
240
241Progressive JPEG support: library can read and write full progressive JPEG
242files.  A "buffered image" mode supports incremental decoding for on-the-fly
243display of progressive images.  Simply recompiling an existing IJG-v5-based
244decoder with v6 should allow it to read progressive files, though of course
245without any special progressive display.
246
247New "jpegtran" application performs lossless transcoding between different
248JPEG formats; primarily, it can be used to convert baseline to progressive
249JPEG and vice versa.  In support of jpegtran, the library now allows lossless
250reading and writing of JPEG files as DCT coefficient arrays.  This ability
251may be of use in other applications.
252
253Notes for programmers:
254* We changed jpeg_start_decompress() to be able to suspend; this makes all
255decoding modes available to suspending-input applications.  However,
256existing applications that use suspending input will need to be changed
257to check the return value from jpeg_start_decompress().  You don't need to
258do anything if you don't use a suspending data source.
259* We changed the interface to the virtual array routines: access_virt_array
260routines now take a count of the number of rows to access this time.  The
261last parameter to request_virt_array routines is now interpreted as the
262maximum number of rows that may be accessed at once, but not necessarily
263the height of every access.
264
265
266Version 5b  15-Mar-95
267---------------------
268
269Correct bugs with grayscale images having v_samp_factor > 1.
270
271jpeg_write_raw_data() now supports output suspension.
272
273Correct bugs in "configure" script for case of compiling in
274a directory other than the one containing the source files.
275
276Repair bug in jquant1.c: sometimes didn't use as many colors as it could.
277
278Borland C makefile and jconfig file work under either MS-DOS or OS/2.
279
280Miscellaneous improvements to documentation.
281
282
283Version 5a  7-Dec-94
284--------------------
285
286Changed color conversion roundoff behavior so that grayscale values are
287represented exactly.  (This causes test image files to change.)
288
289Make ordered dither use 16x16 instead of 4x4 pattern for a small quality
290improvement.
291
292New configure script based on latest GNU Autoconf.
293Fix configure script to handle CFLAGS correctly.
294Rename *.auto files to *.cfg, so that configure script still works if
295file names have been truncated for DOS.
296
297Fix bug in rdbmp.c: didn't allow for extra data between header and image.
298
299Modify rdppm.c/wrppm.c to handle 2-byte raw PPM/PGM formats for 12-bit data.
300
301Fix several bugs in rdrle.c.
302
303NEED_SHORT_EXTERNAL_NAMES option was broken.
304
305Revise jerror.h/jerror.c for more flexibility in message table.
306
307Repair oversight in jmemname.c NO_MKTEMP case: file could be there
308but unreadable.
309
310
311Version 5  24-Sep-94
312--------------------
313
314Version 5 represents a nearly complete redesign and rewrite of the IJG
315software.  Major user-visible changes include:
316  * Automatic configuration simplifies installation for most Unix systems.
317  * A range of speed vs. image quality tradeoffs are supported.
318    This includes resizing of an image during decompression: scaling down
319    by a factor of 1/2, 1/4, or 1/8 is handled very efficiently.
320  * New programs rdjpgcom and wrjpgcom allow insertion and extraction
321    of text comments in a JPEG file.
322
323The application programmer's interface to the library has changed completely.
324Notable improvements include:
325  * We have eliminated the use of callback routines for handling the
326    uncompressed image data.  The application now sees the library as a
327    set of routines that it calls to read or write image data on a
328    scanline-by-scanline basis.
329  * The application image data is represented in a conventional interleaved-
330    pixel format, rather than as a separate array for each color channel.
331    This can save a copying step in many programs.
332  * The handling of compressed data has been cleaned up: the application can
333    supply routines to source or sink the compressed data.  It is possible to
334    suspend processing on source/sink buffer overrun, although this is not
335    supported in all operating modes.
336  * All static state has been eliminated from the library, so that multiple
337    instances of compression or decompression can be active concurrently.
338  * JPEG abbreviated datastream formats are supported, ie, quantization and
339    Huffman tables can be stored separately from the image data.
340  * And not only that, but the documentation of the library has improved
341    considerably!
342
343
344The last widely used release before the version 5 rewrite was version 4A of
34518-Feb-93.  Change logs before that point have been discarded, since they
346are not of much interest after the rewrite.
Note: See TracBrowser for help on using the repository browser.