Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
|
|
include:
- Make new() comply with the documentation and return undef on
a read() failure.
- Re-adds bin/ptar, which disappeared from this distribution after
0.22 (sorry about that).
- Fix a file renaming bug that forgot to carry over path info
- Fix a bug where adding dirs on win32 gave 'permission denied'
- Add extra tests explicilty for Archive::Tar::File
- Move completely from FileHandle to IO::File
- Quell some annoying warnings about binmode on unopened filehandles
- Add tests for binary files included in a tarball
- The chown() code somehow didn't make it into the 1.05 release
- Patch _get_handle() to treat all IO::File handles as binary.
This should make win32 users happy
- A method called 'contains_file' that will tell you if a certain file
is already in the archive.
- Add a global variable $CHOWN that controls whether Archive::Tar
should attempt to chown() files or not when it can.
- NULL-byte padding was done also on files that had no real content,
like symlinks, thus ending up with a number of bytes not dividable
by 512.
- Always do a readlink on the full path, never just the file
- Make Archive::Tar write proper headers when dealing with symlinks
For this $Archive::Tar::FOLLOW_SYMLINKS is introduced
|
|
|
|
|
|
|
|
Suggested by der Mouse.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Bump PKGREVISION.
|
|
Changes since 3.10.20:
* arj_arcv.c, arj_file.c, arj_proc.c, arj_user.c, arjtypes.c,
environ.c, externs.c, externs.h, makefile:
Merged with the remaining part of TCO fixes to make up for 3.10.21
* defines.h, encode.c, environ.c, environ.h, fmemcmp.asm, misc.c,
rearj.c: Resync with TCO to close any outstanding bugs
* arj.c: Rudimentary Borland code caused the filenames in argv[] to
be mishandled under Win32
* register.c: REGISTER might fail due to _fput_* changes
|
|
|
|
Too many bug fixes and additions since last packaged version (0.11).
Please see the `Changes' file enclosed in the source distribution.
|
|
Fixes build on Solaris with native linker.
|
|
Fixes compilation on Solaris with SunPro compiler.
|
|
|
|
So you thought you had your files backed up onto that jaz cartridge -
until it came time to restore. Then you found out that you had bad
sectors and you've lost almost everything because gzip craps out 10%
of the way through your archive. The gzip Recovery Toolkit has a program
- gzrecover - that attempts to skip over bad data in a gzip archive and
to GNU tar that enables that program to skip over bad data and extract
whatever files might be there.
|
|
|
|
|
|
|
|
|
|
|
|
Salvage the `create parent directories and try again' path which is
removed by the previous commit.
|
|
|
|
|
|
|
|
gcc 2.95 does not understand.
DTRT with threading for platforms without native threads.
|
|
|
|
break out of the loop and avoid the infinite loop if attempts to
create, unlink and create all fail.
|
|
don't attempt to remove the current directory.
|
|
|
|
p7zip is a port of the Windows program 7za.exe.
7za.exe is a file archiver with the highest compression ratio.
|
|
|
|
|
|
|
|
properly handling broken archives.
|
|
|
|
tech-pkg@.
|
|
|
|
- Include time.h to use struct tm.
- Remove hardcoded /usr/include/varargs.h from makefile dependency.
|
|
bzip2 library is used instead of native one.
|
|
bypassing the check in configure (we can assume it's there).
Ew, working with non-up-to-date copies of directories...
|
|
deprecated (?) function. Even though, the module seems to build
correctly with the current version of curl in pkgsrc, so change the
check to look for an existent function.
This fixes the build of the curl module, thus solving the problem
exposed in latest kristerw@'s bulk build in NetBSD 2.0_BETA/i386.
Not bumping PKGREVISION because this was not packaged in the first
place.
|
|
configure uses curl_formparse(), which is not available in
curl-7.12.0. Since curl_formparse() is never used in libcompress and
curl is always available through buidlink, it is safe to skip the
test.
|
|
- Use lseek(2) instead of tell() on Linux.
- Use "${RM} -f" instead of "rm" in @unexec.
|