7 * Fix regressions introduced in 5.0.6:
9 - Fix building with non-GNU make.
11 - Fix invalid Libs.private value in liblzma.pc which broke
12 static linking against liblzma if the linker flags were
13 taken from pkg-config.
18 * xzgrep now exits with status 0 if at least one file matched.
20 * A few minor portability and build system fixes
25 * lzmadec and liblzma's lzma_alone_decoder(): Support decompressing
26 .lzma files that have less common settings in the headers
27 (dictionary size other than 2^n or 2^n + 2^(n-1), or uncompressed
28 size greater than 256 GiB). The limitations existed to avoid false
29 positives when detecting .lzma files. The lc + lp <= 4 limitation
30 still remains since liblzma's LZMA decoder has that limitation.
32 NOTE: xz's .lzma support or liblzma's lzma_auto_decoder() are NOT
33 affected by this change. They still consider uncommon .lzma headers
34 as not being in the .lzma format. Changing this would give way too
39 - Interaction of preset and custom filter chain options was
40 made less illogical. This affects only certain less typical
41 uses cases so few people are expected to notice this change.
43 Now when a custom filter chain option (e.g. --lzma2) is
44 specified, all preset options (-0 ... -9, -e) earlier are on
45 the command line are completely forgotten. Similarly, when
46 a preset option is specified, all custom filter chain options
47 earlier on the command line are completely forgotten.
49 Example 1: "xz -9 --lzma2=preset=5 -e" is equivalent to "xz -e"
50 which is equivalent to "xz -6e". Earlier -e didn't put xz back
51 into preset mode and thus the example command was equivalent
52 to "xz --lzma2=preset=5".
54 Example 2: "xz -9e --lzma2=preset=5 -7" is equivalent to
55 "xz -7". Earlier a custom filter chain option didn't make
56 xz forget the -e option so the example was equivalent to
59 - Fixes and improvements to error handling.
61 - Various fixes to the man page.
63 * xzless: Fixed to work with "less" versions 448 and later.
65 * xzgrep: Made -h an alias for --no-filename.
67 * Include the previously missing debug/translation.bash which can
68 be useful for translators.
70 * Include a build script for Mac OS X. This has been in the Git
71 repository since 2010 but due to a mistake in Makefile.am the
72 script hasn't been included in a release tarball before.
79 - Fix lzma_index_init(). It could crash if memory allocation
82 - Fix the possibility of an incorrect LZMA_BUF_ERROR when a BCJ
83 filter is used and the application only provides exactly as
84 much output space as is the uncompressed size of the file.
86 - Fix a bug in doc/examples_old/xz_pipe_decompress.c. It didn't
87 check if the last call to lzma_code() really returned
88 LZMA_STREAM_END, which made the program think that truncated
91 - New example programs in doc/examples (old programs are now in
92 doc/examples_old). These have more comments and more detailed
95 * Fix "xz -lvv foo.xz". It could crash on some corrupted files.
97 * Fix output of "xz --robot -lv" and "xz --robot -lvv" which
98 incorrectly printed the filename also in the "foo (x/x)" format.
100 * Fix exit status of "xzdiff foo.xz bar.xz".
102 * Fix exit status of "xzgrep foo binary_file".
104 * Fix portability to EBCDIC systems.
106 * Fix a configure issue on AIX with the XL C compiler. See INSTALL
109 * Update French, German, Italian, and Polish translations.
116 - A memory leak was fixed.
118 - lzma_stream_buffer_encode() no longer creates an empty .xz
119 Block if encoding an empty buffer. Such an empty Block with
120 LZMA2 data would trigger a bug in 5.0.1 and older (see the
121 first bullet point in 5.0.2 notes). When releasing 5.0.2,
122 I thought that no encoder creates this kind of files but
125 - Validate function arguments better in a few functions. Most
126 importantly, specifying an unsupported integrity check to
127 lzma_stream_buffer_encode() no longer creates a corrupt .xz
128 file. Probably no application tries to do that, so this
129 shouldn't be a big problem in practice.
131 - Document that lzma_block_buffer_encode(),
132 lzma_easy_buffer_encode(), lzma_stream_encoder(), and
133 lzma_stream_buffer_encode() may return LZMA_UNSUPPORTED_CHECK.
135 - The return values of the _memusage() functions are now
138 * Fix command name detection in xzgrep. xzegrep and xzfgrep now
139 correctly use egrep and fgrep instead of grep.
141 * French translation was added.
146 * LZMA2 decompressor now correctly accepts LZMA2 streams with no
147 uncompressed data. Previously it considered them corrupt. The
148 bug can affect applications that use raw LZMA2 streams. It is
149 very unlikely to affect .xz files because no compressor creates
150 .xz files with empty LZMA2 streams. (Empty .xz files are a
151 different thing than empty LZMA2 streams.)
153 * "xz --suffix=.foo filename.foo" now refuses to compress the
154 file due to it already having the suffix .foo. It was already
155 documented on the man page, but the code lacked the test.
157 * "xzgrep -l foo bar.xz" works now.
159 * Polish translation was added.
164 * xz --force now (de)compresses files that have setuid, setgid,
165 or sticky bit set and files that have multiple hard links.
166 The man page had it documented this way already, but the code
169 * gzip and bzip2 support in xzdiff was fixed.
173 * Minor fix to Czech translation
178 Only the most important changes compared to 4.999.9beta are listed
179 here. One change is especially important:
181 * The memory usage limit is now disabled by default. Some scripts
182 written before this change may have used --memory=max on xz command
183 line or in XZ_OPT. THESE USES OF --memory=max SHOULD BE REMOVED
184 NOW, because they interfere with user's ability to set the memory
185 usage limit himself. If user-specified limit causes problems to
186 your script, blame the user.
188 Other significant changes:
190 * Added support for XZ_DEFAULTS environment variable. This variable
191 allows users to set default options for xz, e.g. default memory
192 usage limit or default compression level. Scripts that use xz
193 must never set or unset XZ_DEFAULTS. Scripts should use XZ_OPT
194 instead if they need a way to pass options to xz via an
195 environment variable.
197 * The compression settings associated with the preset levels
198 -0 ... -9 have been changed. --extreme was changed a little too.
199 It is now less likely to make compression worse, but with some
200 files the new --extreme may compress slightly worse than the old
203 * If a preset level (-0 ... -9) is specified after a custom filter
204 chain options have been used (e.g. --lzma2), the custom filter
205 chain will be forgotten. Earlier the preset options were
206 completely ignored after custom filter chain options had been
209 * xz will create sparse files when decompressing if the uncompressed
210 data contains long sequences of binary zeros. This is done even
211 when writing to standard output that is connected to a regular
212 file and certain additional conditions are met to make it safe.
214 * Support for "xz --list" was added. Combine with --verbose or
215 --verbose --verbose (-vv) for detailed output.
217 * I had hoped that liblzma API would have been stable after
218 4.999.9beta, but there have been a couple of changes in the
219 advanced features, which don't affect most applications:
221 - Index handling code was revised. If you were using the old
222 API, you will get a compiler error (so it's easy to notice).
224 - A subtle but important change was made to the Block handling
225 API. lzma_block.version has to be initialized even for
226 lzma_block_header_decode(). Code that doesn't do it will work
227 for now, but might break in the future, which makes this API
230 * The major soname has been bumped to 5.0.0. liblzma API and ABI
231 are now stable, so the need to recompile programs linking against
232 liblzma shouldn't arise soon.