9 - Fix lzma_index_init(). It could crash if memory allocation
12 - Fix the possibility of an incorrect LZMA_BUF_ERROR when a BCJ
13 filter is used and the application only provides exactly as
14 much output space as is the uncompressed size of the file.
16 - Fix a bug in doc/examples_old/xz_pipe_decompress.c. It didn't
17 check if the last call to lzma_code() really returned
18 LZMA_STREAM_END, which made the program think that truncated
21 - New example programs in doc/examples (old programs are now in
22 doc/examples_old). These have more comments and more detailed
25 * Fix "xz -lvv foo.xz". It could crash on some corrupted files.
27 * Fix output of "xz --robot -lv" and "xz --robot -lvv" which
28 incorrectly printed the filename also in the "foo (x/x)" format.
30 * Fix exit status of "xzdiff foo.xz bar.xz".
32 * Fix exit status of "xzgrep foo binary_file".
34 * Fix portability to EBCDIC systems.
36 * Fix a configure issue on AIX with the XL C compiler. See INSTALL
39 * Update French, German, Italian, and Polish translations.
46 - A memory leak was fixed.
48 - lzma_stream_buffer_encode() no longer creates an empty .xz
49 Block if encoding an empty buffer. Such an empty Block with
50 LZMA2 data would trigger a bug in 5.0.1 and older (see the
51 first bullet point in 5.0.2 notes). When releasing 5.0.2,
52 I thought that no encoder creates this kind of files but
55 - Validate function arguments better in a few functions. Most
56 importantly, specifying an unsupported integrity check to
57 lzma_stream_buffer_encode() no longer creates a corrupt .xz
58 file. Probably no application tries to do that, so this
59 shouldn't be a big problem in practice.
61 - Document that lzma_block_buffer_encode(),
62 lzma_easy_buffer_encode(), lzma_stream_encoder(), and
63 lzma_stream_buffer_encode() may return LZMA_UNSUPPORTED_CHECK.
65 - The return values of the _memusage() functions are now
68 * Fix command name detection in xzgrep. xzegrep and xzfgrep now
69 correctly use egrep and fgrep instead of grep.
71 * French translation was added.
76 * LZMA2 decompressor now correctly accepts LZMA2 streams with no
77 uncompressed data. Previously it considered them corrupt. The
78 bug can affect applications that use raw LZMA2 streams. It is
79 very unlikely to affect .xz files because no compressor creates
80 .xz files with empty LZMA2 streams. (Empty .xz files are a
81 different thing than empty LZMA2 streams.)
83 * "xz --suffix=.foo filename.foo" now refuses to compress the
84 file due to it already having the suffix .foo. It was already
85 documented on the man page, but the code lacked the test.
87 * "xzgrep -l foo bar.xz" works now.
89 * Polish translation was added.
94 * xz --force now (de)compresses files that have setuid, setgid,
95 or sticky bit set and files that have multiple hard links.
96 The man page had it documented this way already, but the code
99 * gzip and bzip2 support in xzdiff was fixed.
103 * Minor fix to Czech translation
108 Only the most important changes compared to 4.999.9beta are listed
109 here. One change is especially important:
111 * The memory usage limit is now disabled by default. Some scripts
112 written before this change may have used --memory=max on xz command
113 line or in XZ_OPT. THESE USES OF --memory=max SHOULD BE REMOVED
114 NOW, because they interfere with user's ability to set the memory
115 usage limit himself. If user-specified limit causes problems to
116 your script, blame the user.
118 Other significant changes:
120 * Added support for XZ_DEFAULTS environment variable. This variable
121 allows users to set default options for xz, e.g. default memory
122 usage limit or default compression level. Scripts that use xz
123 must never set or unset XZ_DEFAULTS. Scripts should use XZ_OPT
124 instead if they need a way to pass options to xz via an
125 environment variable.
127 * The compression settings associated with the preset levels
128 -0 ... -9 have been changed. --extreme was changed a little too.
129 It is now less likely to make compression worse, but with some
130 files the new --extreme may compress slightly worse than the old
133 * If a preset level (-0 ... -9) is specified after a custom filter
134 chain options have been used (e.g. --lzma2), the custom filter
135 chain will be forgotten. Earlier the preset options were
136 completely ignored after custom filter chain options had been
139 * xz will create sparse files when decompressing if the uncompressed
140 data contains long sequences of binary zeros. This is done even
141 when writing to standard output that is connected to a regular
142 file and certain additional conditions are met to make it safe.
144 * Support for "xz --list" was added. Combine with --verbose or
145 --verbose --verbose (-vv) for detailed output.
147 * I had hoped that liblzma API would have been stable after
148 4.999.9beta, but there have been a couple of changes in the
149 advanced features, which don't affect most applications:
151 - Index handling code was revised. If you were using the old
152 API, you will get a compiler error (so it's easy to notice).
154 - A subtle but important change was made to the Block handling
155 API. lzma_block.version has to be initialized even for
156 lzma_block_header_decode(). Code that doesn't do it will work
157 for now, but might break in the future, which makes this API
160 * The major soname has been bumped to 5.0.0. liblzma API and ABI
161 are now stable, so the need to recompile programs linking against
162 liblzma shouldn't arise soon.