Lasse Collin [Wed, 7 Jan 2015 17:08:06 +0000 (19:08 +0200)]
xz: Don't fail if stdin doesn't support O_NONBLOCK.
It's a problem at least on OpenBSD which doesn't support
O_NONBLOCK on e.g. /dev/null. I'm not surprised if it's
a problem on other OSes too since this behavior is allowed
in POSIX-1.2008.
The code relying on this behavior was committed in June 2013
and included in 5.1.3alpha released on 2013-10-26. Clearly
the development releases only get limited testing.
Lasse Collin [Tue, 6 Jan 2015 18:30:15 +0000 (20:30 +0200)]
Tests: Don't hide unexpected error messages in test_files.sh.
Hiding them makes no sense since normally there's no error
when testing the "good" files. With "bad" files errors are
expected and then it makes sense to keep the messages hidden.
Lasse Collin [Tue, 30 Dec 2014 09:17:16 +0000 (11:17 +0200)]
Update Solaris notes in INSTALL.
Mention the possible "make check" failure on Solaris in the
Solaris-specific section of INSTALL. It was already in
section 4.5 but it is better mention it in the OS-specific
section too.
Lasse Collin [Sun, 21 Dec 2014 16:11:17 +0000 (18:11 +0200)]
Bump version and soname for 5.2.0.
I know that soname != app version, but I skip AGE=1
in -version-info to make the soname match the liblzma
version anyway. It doesn't hurt anything as long as
it doesn't conflict with library versioning rules.
Lasse Collin [Sat, 20 Dec 2014 18:41:48 +0000 (20:41 +0200)]
Windows: Update the build script and README-Windows.txt.
The 32-bit build is now for i686 or newer because the
prebuilt MinGW-w64 toolchains include i686 code in the
executables even if one uses -march=i486.
The build script builds 32-bit SSE2 enabled version too.
Run-time detection of SSE2 support would be nice (on any OS)
but it's not implemented in XZ Utils yet.
Adrien Nader [Wed, 3 Dec 2014 19:00:53 +0000 (20:00 +0100)]
po/fr: remove fuzzy marker for error messages that will be kept in English.
The following is a copy of a comment inside fr.po:
Note from translator on "file status flags".
The following entry is kept un-translated on purpose. It is difficult to
translate and should only happen in exceptional circumstances which means
that translating would:
- lose some of the meaning
- make it more difficult to look up in search engines; it might happen one
in
a million times, if we dilute the error message in 20 languages, it will be
almost impossible to find an explanation and support for the error.
Lasse Collin [Tue, 25 Nov 2014 10:32:05 +0000 (12:32 +0200)]
liblzma: Verify the filter chain in threaded encoder initialization.
This way an invalid filter chain is detected at the Stream
encoder initialization instead of delaying it to the first
call to lzma_code() which triggers the initialization of
the actual filter encoder(s).
Lasse Collin [Mon, 10 Nov 2014 16:45:01 +0000 (18:45 +0200)]
xzdiff: Create a temporary directory to hold a temporary file.
This avoids the possibility of "File name too long" when
creating a temp file when the input file name is very long.
This also means that other users on the system can no longer
see the input file names in /tmp (or whatever $TMPDIR is)
since the temporary directory will have a generic name. This
usually doesn't matter since on many systems one can see
the arguments given to all processes anyway.
The number X chars to mktemp where increased from 6 to 10.
Note that with some shells temp files or dirs won't be used at all.
Lasse Collin [Thu, 9 Oct 2014 15:42:14 +0000 (18:42 +0300)]
xzgrep: Avoid passing both -q and -l to grep.
The behavior of grep -ql varies:
- GNU grep behaves like grep -q.
- OpenBSD grep behaves like grep -l.
POSIX doesn't make it 100 % clear what behavior is expected.
Anyway, using both -q and -l at the same time makes no sense
so both options simply should never be used at the same time.
Clang and nowadays also GCC accept any -Wfoobar option
but then may give a warning that an unknown warning option
was specified. To avoid adding unsupported warning options,
the options are now tested with -Werror.
Lasse Collin [Tue, 5 Aug 2014 19:03:30 +0000 (22:03 +0300)]
liblzma: Add support for lzma_block.ignore_check.
Note that this slightly changes how lzma_block_header_decode()
has been documented. Earlier it said that the .version is set
to the lowest required value, but now it says that the .version
field is kept unchanged if possible. In practice this doesn't
affect any old code, because before this commit the only
possible .version was 0.
Lasse Collin [Sun, 3 Aug 2014 18:32:25 +0000 (21:32 +0300)]
liblzma: SHA-256: Optimize the Maj macro slightly.
The Maj macro is used where multiple things are added
together, so making Maj a sum of two expressions allows
some extra freedom for the compiler to schedule the
instructions.
I learned this trick from
<http://www.hackersdelight.org/corres.txt>.
Lasse Collin [Sun, 3 Aug 2014 17:33:38 +0000 (20:33 +0300)]
liblzma: SHA-256: Unroll a little more.
This way a branch isn't needed for each operation
to choose between blk0 and blk2, and still the code
doesn't grow as much as it would with full unrolling.
Lasse Collin [Sun, 25 May 2014 18:45:56 +0000 (21:45 +0300)]
liblzma: Use lzma_alloc_zero() in LZ encoder initialization.
This avoids a memzero() call for a newly-allocated memory,
which can be expensive when encoding small streams with
an over-sized dictionary.
To avoid using lzma_alloc_zero() for memory that doesn't
need to be zeroed, lzma_mf.son is now allocated separately,
which requires handling it separately in normalize() too.
Thanks to Vincenzo Innocente for reporting the problem.
Lasse Collin [Fri, 25 Apr 2014 14:53:42 +0000 (17:53 +0300)]
Build: Fix the combination of --disable-xzdec --enable-lzmadec.
In this case "make install" could fail if the man page directory
didn't already exist at the destination. If it did exist, a
dangling symlink was created there. Now the link is omitted
instead. This isn't the best fix but it's better than the old
behavior.