| 1commit b69900ed0b2f914fc6c0a180dcb522dbe5b80ea7
2Author: Lasse Collin <lasse.collin@tukaani.org>
3Date: Sun Jun 30 18:02:27 2013 +0300
4
5 Man pages: Use similar syntax for synopsis as in xz.
6
7 The man pages of lzmainfo, xzmore, and xzdec had similar
8 constructs as the man page of xz had before the commit
9 eb6ca9854b8eb9fbf72497c1cf608d6b19d2d494. Eric S. Raymond
10 didn't mention these man pages in his bug report, but
11 it's nice to be consistent.
12
13commit cf4a1e1879d89be314ef3c064bd2656ea452f87e
14Author: Lasse Collin <lasse.collin@tukaani.org>
15Date: Sun Jun 30 15:55:09 2013 +0300
16
17 Update NEWS for 5.0.5.
18
19commit cb94bb6d1f34e1e93c2d634ea9c3b7dfb3981d05
20Author: Lasse Collin <lasse.collin@tukaani.org>
21Date: Sun Jun 30 15:54:38 2013 +0300
22
23 Bump version and soname for 5.0.5.
24
25commit b7dee202d5b041ccae028d0c5433b83cecbe9e5d
26Author: Lasse Collin <lasse.collin@tukaani.org>
27Date: Fri Jun 28 23:56:17 2013 +0300
28
29 xz: Fix return value type in io_write_buf().
30
31 It didn't affect the behavior of the code since -1
32 becomes true anyway.
33
34commit 265e7b44d804b47373f10b7da28350db7611cea6
35Author: Lasse Collin <lasse.collin@tukaani.org>
36Date: Fri Jun 28 18:46:13 2013 +0300
37
38 xz: Remove an outdated NetBSD-specific comment.
39
40 Nowadays errno == EFTYPE is documented in open(2).
41
42commit 78c2f8db902195468b8249c432252a6b281db836
43Author: Lasse Collin <lasse.collin@tukaani.org>
44Date: Fri Jun 28 18:09:47 2013 +0300
45
46 xz: Fix error detection of fcntl(fd, F_SETFL, flags) calls.
47
48 POSIX says that fcntl(fd, F_SETFL, flags) returns -1 on
49 error and "other than -1" on success. This is how it is
50 documented e.g. on OpenBSD too. On Linux, success with
51 F_SETFL is always 0 (at least accorinding to fcntl(2)
52 from man-pages 3.51).
53
54commit 91750dff8f2c654ff636f12a2acdffe5492374b3
55Author: Lasse Collin <lasse.collin@tukaani.org>
56Date: Fri Jun 28 17:36:47 2013 +0300
57
58 xz: Fix use of wrong variable in a fcntl() call.
59
60 Due to a wrong variable name, when writing a sparse file
61 to standard output, *all* file status flags were cleared
62 (to the extent the operating system allowed it) instead of
63 only clearing the O_APPEND flag. In practice this worked
64 fine in the common situations on GNU/Linux, but I didn't
65 check how it behaved elsewhere.
66
67 The original flags were still restored correctly. I still
68 changed the code to use a separate boolean variable to
69 indicate when the flags should be restored instead of
70 relying on a special value in stdout_flags.
71
72commit e11888a79a4a77a69afde60445880d44f63d01aa
73Author: Lasse Collin <lasse.collin@tukaani.org>
74Date: Wed Jun 26 13:30:57 2013 +0300
75
76 xz: Check the value of lzma_stream_flags.version in --list.
77
78 It is a no-op for now, but if an old xz version is used
79 together with a newer liblzma that supports something new,
80 then this check becomes important and will stop the old xz
81 from trying to parse files that it won't understand.
82
83commit f39ddd88f3222219ada88998cf30abfdd3e0e96c
84Author: Lasse Collin <lasse.collin@tukaani.org>
85Date: Wed Jun 26 12:17:00 2013 +0300
86
87 Build: Require Automake 1.12 and use serial-tests option.
88
89 It should actually still work with Automake 1.10 if
90 the serial-tests option is removed. Automake 1.13 started
91 using parallel tests by default and the option to get
92 the old behavior isn't supported before 1.12.
93
94 At least for now, parallel tests don't improve anything
95 in XZ Utils but they hide the progress output from
96 test_compress.sh.
97
98commit cb84e278027a90e9827a6f4d3bb0b4d4744a2fbb
99Author: Lasse Collin <lasse.collin@tukaani.org>
100Date: Sun Jun 23 17:36:47 2013 +0300
101
102 xz: Validate Uncompressed Size from Block Header in list.c.
103
104 This affects only "xz -lvv". Normal decompression with xz
105 already detected if Block Header and Index had mismatched
106 Uncompressed Size fields. So this just makes "xz -lvv"
107 show such files as corrupt instead of showing the
108 Uncompressed Size from Index.
109
110commit f01780fce454c7489f7dcbf806299b50da5f51b7
111Author: Lasse Collin <lasse.collin@tukaani.org>
112Date: Wed Jun 26 10:58:58 2013 +0300
113
114 Update THANKS.
115
116commit d98ede7d700b892e32d9c2f46563b6ebc566786d
117Author: Lasse Collin <lasse.collin@tukaani.org>
118Date: Fri Jun 21 22:04:45 2013 +0300
119
120 xz: Make the man page more friendly to doclifter.
121
122 Thanks to Eric S. Raymond.
123
124commit 19b447b64b3f520cd5b11429000b092f7c76709b
125Author: Lasse Collin <lasse.collin@tukaani.org>
126Date: Fri Jun 21 21:54:59 2013 +0300
127
128 xz: A couple of man page fixes.
129
130 Now the interaction of presets and custom filter chains
131 is described correctly. Earlier it contradicted itself.
132
133 Thanks to DevHC who reported these issues on IRC to me
134 on 2012-12-14.
135
136commit 45edf2966fc9a4d2eae8f84b2fa027fb4fa1df8b
137Author: Lasse Collin <lasse.collin@tukaani.org>
138Date: Fri Jun 21 21:50:26 2013 +0300
139
140 xz: Fix interaction between preset and custom filter chains.
141
142 There was somewhat illogical behavior when --extreme was
143 specified and mixed with custom filter chains.
144
145 Before this commit, "xz -9 --lzma2 -e" was equivalent
146 to "xz --lzma2". After it is equivalent to "xz -6e"
147 (all earlier preset options get forgotten when a custom
148 filter chain is specified and the default preset is 6
149 to which -e is applied). I find this less illogical.
150
151 This also affects the meaning of "xz -9e --lzma2 -7".
152 Earlier it was equivalent to "xz -7e" (the -e specified
153 before a custom filter chain wasn't forgotten). Now it
154 is "xz -7". Note that "xz -7e" still is the same as "xz -e7".
155
156 Hopefully very few cared about this in the first place,
157 so pretty much no one should even notice this change.
158
159 Thanks to Conley Moorhous.
160
161commit b065984e5a9272eb50bc0c6d3731e6199c0ae8a8
162Author: Lasse Collin <lasse.collin@tukaani.org>
163Date: Fri Apr 8 17:53:05 2011 +0300
164
165 xz: Change size_t to uint32_t in a few places.
166
167commit 32be621f52f2e1686db88baa7b01dc1ae338f426
168Author: Lasse Collin <lasse.collin@tukaani.org>
169Date: Sat Apr 27 22:07:46 2013 +0300
170
171 Build: Use -Wvla with GCC if supported.
172
173 Variable-length arrays are mandatory in C99 but optional in C11.
174 The code doesn't currently use any VLAs and it shouldn't in the
175 future either to stay compatible with C11 without requiring any
176 optional C11 features.
177
178commit efb07cfba65e9e05984c02cd796c1b0338ce04dc
179Author: Lasse Collin <lasse.collin@tukaani.org>
180Date: Mon Apr 15 19:29:09 2013 +0300
181
182 xzdec: Improve the --help message.
183
184 The options are now ordered in the same order as in xz's help
185 message.
186
187 Descriptions were added to the options that are ignored.
188 I left them in parenthesis even if it looks a bit weird
189 because I find it easier to spot the ignored vs. non-ignored
190 options from the list that way.
191
192commit e3c8be13699e2813f5e2879d8187444b46d82d89
193Author: Lasse Collin <lasse.collin@tukaani.org>
194Date: Fri Apr 5 19:25:40 2013 +0300
195
196 Update THANKS.
197
198commit ad8282efe483612f6b5544f9a0d2e4914fb2532a
199Author: Jeff Bastian <jbastian@redhat.com>
200Date: Wed Apr 3 13:59:17 2013 +0200
201
202 xzgrep: make the '-h' option to be --no-filename equivalent
203
204 * src/scripts/xzgrep.in: Accept the '-h' option in argument parsing.
205
206commit 9271a3eb0e022b23e8712154be851d0afe4c02e4
207Author: Lasse Collin <lasse.collin@tukaani.org>
208Date: Fri Apr 5 19:34:09 2013 +0300
209
210 liblzma: Be less picky in lzma_alone_decoder().
211
212 To avoid false positives when detecting .lzma files,
213 rare values in dictionary size and uncompressed size fields
214 were rejected. They will still be rejected if .lzma files
215 are decoded with lzma_auto_decoder(), but when using
216 lzma_alone_decoder() directly, such files will now be accepted.
217 Hopefully this is an OK compromise.
218
219 This doesn't affect xz because xz still has its own file
220 format detection code. This does affect lzmadec though.
221 So after this commit lzmadec will accept files that xz or
222 xz-emulating-lzma doesn't.
223
224 NOTE: lzma_alone_decoder() still won't decode all .lzma files
225 because liblzma's LZMA decoder doesn't support lc + lp > 4.
226
227 Reported here:
228 http://sourceforge.net/projects/lzmautils/forums/forum/708858/topic/7068827
229
230 Conflicts:
231 src/liblzma/common/alone_decoder.c
232 src/liblzma/common/alone_decoder.h
233
234commit 211b931cee58626c1d2e021810cb108cb5cbc10f
235Author: Lasse Collin <lasse.collin@tukaani.org>
236Date: Tue Mar 5 19:14:50 2013 +0200
237
238 Avoid unneeded use of awk in xzless.
239
240 Use "read" instead of "awk" in xzless to get the version
241 number of "less". The need for awk was introduced in
242 the commit db5c1817fabf7cbb9e4087b1576eb26f0747338e.
243
244 Thanks to Ariel P for the patch.
245
246commit 9f62fd9605eade23b62b07a235d1f02156f7a5c6
247Author: Jonathan Nieder <jrnieder@gmail.com>
248Date: Mon Nov 19 00:10:10 2012 -0800
249
250 xzless: Make "less -V" parsing more robust
251
252 In v4.999.9beta~30 (xzless: Support compressed standard input,
253 2009-08-09), xzless learned to parse ���less -V��� output to figure out
254 whether less is new enough to handle $LESSOPEN settings starting
255 with ���|-���. That worked well for a while, but the version string from
256 ���less��� versions 448 (June, 2012) is misparsed, producing a warning:
257
258 $ xzless /tmp/test.xz; echo $?
259 /usr/bin/xzless: line 49: test: 456 (GNU regular expressions): \
260 integer expression expected
261 0
262
263 More precisely, modern ���less��� lists the regexp implementation along
264 with its version number, and xzless passes the entire version number
265 with attached parenthetical phrase as a number to "test $a -gt $b",
266 producing the above confusing message.
267
268 $ less-444 -V | head -1
269 less 444
270 $ less -V | head -1
271 less 456 (no regular expressions)
272
273 So relax the pattern matched --- instead of expecting "less <number>",
274 look for a line of the form "less <number>[ (extra parenthetical)]".
275 While at it, improve the behavior when no matching line is found ---
276 instead of producing a cryptic message, we can fall back on a LESSPIPE
277 setting that is supported by all versions of ���less���.
278
279 The implementation uses "awk" for simplicity. Hopefully that���s
280 portable enough.
281
282 Reported-by: J��rg-Volker Peetz <jvpeetz@web.de>
283 Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
284
285commit 1d05980f5b5c2c94d833001daccacce4a466876e
286Author: Lasse Collin <lasse.collin@tukaani.org>
287Date: Wed Oct 3 15:54:24 2012 +0300
288
289 xz: Fix the note about --rsyncable on the man page.
290
291commit fb68497333598688d309a92838d91fd560f7e9f0
292Author: Lasse Collin <lasse.collin@tukaani.org>
293Date: Fri Sep 28 20:11:09 2012 +0300
294
295 xz: Improve handling of failed realloc in xrealloc.
296
297 Thanks to Jim Meyering.
298
299commit 75013db6d4d63c195bd8b8d45729b4be0665a812
300Author: Lasse Collin <lasse.collin@tukaani.org>
301Date: Sat Dec 15 20:01:02 2012 +0200
302
303 A few typo fixes to comments and the xz man page.
304
305 Thanks to Jim Meyering.
306
307commit e44b21839b1dcbac5097be39b87dd2ddb6e114fd
308Author: Lasse Collin <lasse.collin@tukaani.org>
309Date: Thu Aug 2 17:13:30 2012 +0300
310
311 Build: Bump gettext version requirement to 0.18.
312
313 Otherwise too old version of m4/lib-link.m4 gets included
314 when autoreconf -fi is run.
315
316commit fd3dbb23ca7e75a7a888d7e897c381dc06308307
317Author: Lasse Collin <lasse.collin@tukaani.org>
318Date: Thu Jul 5 07:36:28 2012 +0300
319
320 Tests: Remove tests/test_block.c that had gotten committed accidentally.
321
322commit 05a735d279d74af437c31f25f69aded4713c1a3d
323Author: Lasse Collin <lasse.collin@tukaani.org>
324Date: Thu Jul 5 07:33:35 2012 +0300
325
326 Build: Include macosx/build.sh in the distribution.
327
328 It has been in the Git repository since 2010 but probably
329 few people have seen it since it hasn't been included in
330 the release tarballs. :-(
331
332commit 4e6d62793b5e7b87edcc93c7ded072c1ecd94173
333Author: Lasse Collin <lasse.collin@tukaani.org>
334Date: Thu Jul 5 07:24:45 2012 +0300
335
336 Docs: Fix the name LZMA Utils -> XZ Utils in debug/README.
337
338commit dd95b5e7614baf1f07a1316b5106bd616a9efa79
339Author: Lasse Collin <lasse.collin@tukaani.org>
340Date: Thu Jul 5 07:23:17 2012 +0300
341
342 Include debug/translation.bash in the distribution.
343
344 Also fix the script name mentioned in README.
345
|
1commit 20778053a07eb90c159c1377ca8dc05a90fd530b
2Author: Lasse Collin <lasse.collin@tukaani.org>
3Date: Fri Jun 22 14:36:16 2012 +0300
4
5 xz: Update man page date to match the latest update.
6
7commit 2cefa84af676da37d7e9c466d55d46c67ab00c22
8Author: Lasse Collin <lasse.collin@tukaani.org>
--- 6412 unchanged lines hidden --- | 346commit 20778053a07eb90c159c1377ca8dc05a90fd530b
347Author: Lasse Collin <lasse.collin@tukaani.org>
348Date: Fri Jun 22 14:36:16 2012 +0300
349
350 xz: Update man page date to match the latest update.
351
352commit 2cefa84af676da37d7e9c466d55d46c67ab00c22
353Author: Lasse Collin <lasse.collin@tukaani.org>
--- 6412 unchanged lines hidden --- |