]> git.ipfire.org Git - thirdparty/libarchive.git/log
thirdparty/libarchive.git
8 years agoRemove unneeded return. 798/head
Mariusz Zaborski [Fri, 7 Oct 2016 19:37:55 +0000 (21:37 +0200)] 
Remove unneeded return.

8 years agoStyle.
Mariusz Zaborski [Fri, 7 Oct 2016 19:34:24 +0000 (21:34 +0200)] 
Style.

8 years agoProperly free and close when exiting write_hierarchy() on failed
Martin Matuska [Tue, 4 Oct 2016 10:09:28 +0000 (12:09 +0200)] 
Properly free and close when exiting write_hierarchy() on failed
archive_read_next_header2().

Fixes #794

8 years agoMerge pull request #791 from hartzell/bug/disk-variable-masked
Tim Kientzle [Mon, 3 Oct 2016 00:38:07 +0000 (17:38 -0700)] 
Merge pull request #791 from hartzell/bug/disk-variable-masked

Fix scoping error with "disk" variable

8 years agoFix scoping error with "disk" variable 791/head
George Hartzell [Thu, 29 Sep 2016 21:30:44 +0000 (14:30 -0700)] 
Fix scoping error with "disk" variable

The variable named 'disk' inside the loop was hiding the variable named
'disk' at the outer level.  The outer variable was never used, get rid
of it.

Move the call to archive_read_disk_set_standard_lookup(disk) inside the
loop so that it Does What It Should.

8 years agoMerge pull request #790 from Tarsnap/uninitialized-subst-result
Tim Kientzle [Thu, 29 Sep 2016 04:09:29 +0000 (21:09 -0700)] 
Merge pull request #790 from Tarsnap/uninitialized-subst-result

Initialize subst rule->result pointer to NULL

8 years agoInitialize subst rule->result pointer to NULL 790/head
Graham Percival [Wed, 28 Sep 2016 18:02:34 +0000 (11:02 -0700)] 
Initialize subst rule->result pointer to NULL

This is not directly useful in libarchive at the present time, but
it might be in the future.  If a project which uses this code adds
an atexit(3) handler to clean up memory, and somebody runs:
    bsdtar somestring -t
then "omestring" will be interpreted as an (invalid) substitution
rule and call lafe_errc().  However, if the atexit() handler then
calls cleanup_substition(), it will try to free(rule->result)
which was not initialized.

Sponsored by: Tarsnap Backup Inc.

8 years agoMerge pull request #787 from praiskup/fix-umask
Tim Kientzle [Wed, 28 Sep 2016 13:30:51 +0000 (06:30 -0700)] 
Merge pull request #787 from praiskup/fix-umask

tests: fix testsuite for umask issues

8 years agoMerge pull request #788 from Tarsnap/check-test-mode
Tim Kientzle [Wed, 28 Sep 2016 13:29:41 +0000 (06:29 -0700)] 
Merge pull request #788 from Tarsnap/check-test-mode

Check test mode

8 years agoAdd chmod() after make_{dir,file} in test suites 788/head
Graham Percival [Mon, 26 Sep 2016 19:50:29 +0000 (12:50 -0700)] 
Add chmod() after make_{dir,file} in test suites

This is the same change that
    38b3f516df865e2dffd0a1bbc9feb923ab2efc38
made to
    libarchive/test/main.c
, but now applied to:
    cat/test/main.c
    cpio/test/main.c
    tar/test/main.c

Sponsored by: Tarsnap Backup Inc.

8 years agoAdd and use assertion_file_mode() in other tests
Graham Percival [Mon, 26 Sep 2016 19:20:25 +0000 (12:20 -0700)] 
Add and use assertion_file_mode() in other tests

This adds the assertion_file_mode() function from
    libarchive/test/main.c
and applies it to
    cat/test/main.c
    cpio/test/main.c
    tar/test/main.c

Sponsored by: Tarsnap Backup Inc.

8 years agotests: fix testsuite for umask issues 787/head
Pavel Raiskup [Mon, 26 Sep 2016 10:56:52 +0000 (12:56 +0200)] 
tests: fix testsuite for umask issues

Reproducible with umask 0077.

Resolves: #782

8 years ago"bunzip2 -V" does not exit with 0, use "bunzip2 -h" instead
Martin Matuska [Sun, 25 Sep 2016 00:03:23 +0000 (02:03 +0200)] 
"bunzip2 -V" does not exit with 0, use "bunzip2 -h" instead

8 years agoAdd READ_DATA dir permission to @owner to make test cleanup work for non-root
Martin Matuska [Sat, 24 Sep 2016 23:30:34 +0000 (01:30 +0200)] 
Add READ_DATA dir permission to @owner to make test cleanup work for non-root

8 years agoMerge pull request #774 from Tarsnap/check-test-mode
Tim Kientzle [Sat, 24 Sep 2016 21:07:33 +0000 (14:07 -0700)] 
Merge pull request #774 from Tarsnap/check-test-mode

chmod and check the mode after creating a file or dir

8 years agoIssue 774, 782: chmod() after creating a file or dir 774/head
Graham Percival [Fri, 23 Sep 2016 17:56:21 +0000 (10:56 -0700)] 
Issue 774, 782: chmod() after creating a file or dir

If the user's system has a default umask, then
    mkdir(pathname, mode);
will report "success" even if the created dir does not match the specified
mode.  Presumably that is desired in the general case, but when it comes to
testing libarchive this can generate false errors in `make check`.

chmod() is not affected by umask, so we call that after creating the file
or directory.

Sponsored by: Tarsnap Backup Inc.

8 years agoIssue 761: Heap overflow reading corrupted 7Zip files
Tim Kientzle [Mon, 19 Sep 2016 01:14:58 +0000 (18:14 -0700)] 
Issue 761:  Heap overflow reading corrupted 7Zip files

The sample file that demonstrated this had multiple 'EmptyStream'
attributes.  The first one ended up being used to calculate
certain statistics, then was overwritten by the second which
was incompatible with those statistics.

The fix here is to reject any header with multiple EmptyStream
attributes.  While here, also reject headers with multiple
EmptyFile, AntiFile, Name, or Attributes markers.

8 years agoMerge branch 'master' of github.com:libarchive/libarchive
Tim Kientzle [Mon, 19 Sep 2016 00:31:24 +0000 (17:31 -0700)] 
Merge branch 'master' of github.com:libarchive/libarchive

8 years agoIssue 747 (and others?): Avoid OOB read when parsing multiple long lines
Tim Kientzle [Mon, 19 Sep 2016 00:27:47 +0000 (17:27 -0700)] 
Issue 747 (and others?):  Avoid OOB read when parsing multiple long lines

The mtree bidder needs to look several lines ahead
in the input.  It does this by extending the read-ahead
and parsing subsequent lines from the same growing buffer.
A bookkeeping error when extending the read-ahead would
sometimes lead it to significantly over-count the
size of the line being read.

8 years agoMerge pull request #781 from praiskup/build-fix-undeclared-var
Tim Kientzle [Mon, 12 Sep 2016 13:21:48 +0000 (06:21 -0700)] 
Merge pull request #781 from praiskup/build-fix-undeclared-var

ftbfs: declare 'r' variable for non NFSv4 ACLs too

8 years agoftbfs: declare 'r' variable for non NFSv4 ACLs too 781/head
Pavel Raiskup [Mon, 12 Sep 2016 11:58:53 +0000 (13:58 +0200)] 
ftbfs: declare 'r' variable for non NFSv4 ACLs too

8 years agoMerge branch 'master' of github.com:libarchive/libarchive
Tim Kientzle [Mon, 12 Sep 2016 01:38:32 +0000 (18:38 -0700)] 
Merge branch 'master' of github.com:libarchive/libarchive

8 years agoUse O_NOFOLLOW to avoid traversing symlinks when restoring data-carrying hardlinks.
Tim Kientzle [Sun, 11 Sep 2016 21:33:30 +0000 (14:33 -0700)] 
Use O_NOFOLLOW to avoid traversing symlinks when restoring data-carrying hardlinks.

TODO: Consider using this flag in other situations as well.

Thanks to Pavel Raiskup.

8 years agoReject non-regular sparse files
Tim Kientzle [Sun, 11 Sep 2016 21:20:31 +0000 (14:20 -0700)] 
Reject non-regular sparse files

Thanks to Pavel Raiskup

8 years agoRemove dead comments
Tim Kientzle [Sun, 11 Sep 2016 21:13:52 +0000 (14:13 -0700)] 
Remove dead comments

8 years agoFixes for Issue #745 and Issue #746 from Doran Moppert.
Tim Kientzle [Sun, 11 Sep 2016 20:21:57 +0000 (13:21 -0700)] 
Fixes for Issue #745 and Issue #746 from Doran Moppert.

8 years agoFix the test cases for Issue #745 and Issue #746
Tim Kientzle [Sun, 11 Sep 2016 20:19:05 +0000 (13:19 -0700)] 
Fix the test cases for Issue #745 and Issue #746

Thanks to Doran Moppert for pointing out the inconsistencies here.

8 years agoMerge pull request #775 from mmatuska/glibc_2_24_readdir_r
Tim Kientzle [Sat, 10 Sep 2016 17:05:55 +0000 (10:05 -0700)] 
Merge pull request #775 from mmatuska/glibc_2_24_readdir_r

Glibc 2.24 deprecates readdir_r()

8 years agoMerge pull request #779 from nacho/fix-769
Tim Kientzle [Fri, 9 Sep 2016 06:51:39 +0000 (23:51 -0700)] 
Merge pull request #779 from nacho/fix-769

test_write_format_gnutar_filenames: use AE_IFLNK instead of S_IFLNK

8 years agotest_write_format_gnutar_filenames: use AE_IFLNK instead of S_IFLNK 779/head
Ignacio Casal Quinteiro [Thu, 8 Sep 2016 12:53:18 +0000 (14:53 +0200)] 
test_write_format_gnutar_filenames: use AE_IFLNK instead of S_IFLNK

Since S_IFLNK will fail on windows.

https://github.com/libarchive/libarchive/issues/769

8 years agoMerge pull request #778 from mmatuska/acl_error_handling
Tim Kientzle [Mon, 5 Sep 2016 17:35:24 +0000 (10:35 -0700)] 
Merge pull request #778 from mmatuska/acl_error_handling

Handle all possible errors from ACL functions

8 years agoACL error handling 778/head
Martin Matuska [Mon, 15 Aug 2016 22:32:42 +0000 (00:32 +0200)] 
ACL error handling

8 years agoFill in checks and support for acl_get_fd_np
Tim Kientzle [Sun, 4 Sep 2016 23:35:07 +0000 (16:35 -0700)] 
Fill in checks and support for acl_get_fd_np

8 years agoMerge branch 'fix_acls_inside_subdirs' of https://github.com/mmatuska/libarchive...
Tim Kientzle [Sun, 4 Sep 2016 23:15:28 +0000 (16:15 -0700)] 
Merge branch 'fix_acls_inside_subdirs' of https://github.com/mmatuska/libarchive into mmatuska-fix_acls_inside_subdirs

8 years agoTest for PR#777: verify ACL reading during directory walks
Tim Kientzle [Sun, 4 Sep 2016 23:08:26 +0000 (16:08 -0700)] 
Test for PR#777: verify ACL reading during directory walks

This verifies the issue mentioned in Pull Request #777:
When using archive_read_disk to read a directory tree,
libarchive would attempt to read ACLs with the wrong
directory.

8 years agoFix FreeBSD build: prefer struct xvfsconf to struct vfsconf
Tim Kientzle [Sun, 4 Sep 2016 23:07:27 +0000 (16:07 -0700)] 
Fix FreeBSD build:  prefer struct xvfsconf to struct vfsconf

8 years agoMultiple bugfixes for setup_acls() 777/head
Martin Matuska [Mon, 15 Aug 2016 21:54:37 +0000 (23:54 +0200)] 
Multiple bugfixes for setup_acls()

8 years agoConvert README to markdown
Tim Kientzle [Sun, 4 Sep 2016 16:36:39 +0000 (09:36 -0700)] 
Convert README to markdown

8 years agoConvert README to markdown
Tim Kientzle [Sun, 4 Sep 2016 16:28:47 +0000 (09:28 -0700)] 
Convert README to markdown

8 years agoConvert README to markdown
Tim Kientzle [Sat, 3 Sep 2016 18:33:50 +0000 (11:33 -0700)] 
Convert README to markdown

8 years agoConvert README to markdown
Tim Kientzle [Sat, 3 Sep 2016 18:26:20 +0000 (11:26 -0700)] 
Convert README to markdown

8 years agoConvert README to markdown
Tim Kientzle [Sat, 3 Sep 2016 18:10:15 +0000 (11:10 -0700)] 
Convert README to markdown

8 years agoDo not use deprecated readdir_r() on glibc 2.24 and later 775/head
Martin Matuska [Sun, 28 Aug 2016 01:04:17 +0000 (03:04 +0200)] 
Do not use deprecated readdir_r() on glibc 2.24 and later

8 years agoIssue #731: Reject tar entries >= INT64_MAX
Tim Kientzle [Mon, 22 Aug 2016 00:31:49 +0000 (17:31 -0700)] 
Issue #731:  Reject tar entries >= INT64_MAX

Note that the code that parses file sizes returns INT64_MAX on
overflow.

8 years agoIssue #744 (part of Issue #743): Enforce sandbox with very long pathnames
Tim Kientzle [Mon, 22 Aug 2016 00:11:45 +0000 (17:11 -0700)] 
Issue #744 (part of Issue #743): Enforce sandbox with very long pathnames

Because check_symlinks is handled separately from the deep-directory
support, very long pathnames cause problems.  Previously, the code
ignored most failures to lstat() a path component.  In particular,
this led to check_symlinks always passing for very long paths, which
in turn provides a way to evade the symlink checks in the sandboxing
code.

We now fail on unrecognized lstat() failures, which plugs this
hole at the cost of disabling deep directory support when the
user requests sandboxing.

TODO:  This probably cannot be completely fixed without
entirely reimplementing the deep directory support to
integrate the symlink checks.  I want to reimplement the
deep directory hanlding someday anyway; openat() and
related system calls now provide a much cleaner way to
handle deep directories than the chdir approach used by this
code.

8 years agoIssue #748: Zip decompression failure with highly-compressed data
Tim Kientzle [Mon, 22 Aug 2016 00:01:54 +0000 (17:01 -0700)] 
Issue #748: Zip decompression failure with highly-compressed data

Previously, we stopped driving the decompressor as soon as
we hit end of input, but in some cases, the decompressor
has internal state that can continue generating output
even when there is no more input.  So we now feed zero-length
blocks into the decompressor until the decompressor tells us
it is finished.

8 years agoIssue #767: Buffer overflow printing a filename
Tim Kientzle [Sun, 21 Aug 2016 17:51:43 +0000 (10:51 -0700)] 
Issue #767:  Buffer overflow printing a filename

The safe_fprintf function attempts to ensure clean output for an
arbitrary sequence of bytes by doing a trial conversion of the
multibyte characters to wide characters -- if the resulting wide
character is printable then we pass through the corresponding bytes
unaltered, otherwise, we convert them to C-style ASCII escapes.

The stack trace in Issue #767 suggest that the 20-byte buffer
was getting overflowed trying to format a non-printable multibyte
character.  This should only happen if there is a valid multibyte
character of more than 5 bytes that was unprintable.  (Each byte
would get expanded to a four-charcter octal-style escape of the form
"\123" resulting in >20 characters for the >5 byte multibyte character.)

I've not been able to reproduce this, but have expanded the conversion
buffer to 128 bytes on the belief that no multibyte character set
has a single character of more than 32 bytes.

8 years agoIssue #770: Be more careful about extra_length
Tim Kientzle [Sun, 21 Aug 2016 16:25:00 +0000 (09:25 -0700)] 
Issue #770:  Be more careful about extra_length

Hanno Böck pointed out that the loop here computes
extra_length - 4 without first checking for possible underflow.

In addition to fixing this, I also added a bunch of error
checks so Zip parsing will fail if any Zip extra field
is malformed.  Among other things, this uncovered an
old bug that would skip a trailing extra field with
zero-sized data.

Note that we still simply ignore well-formed
extra fields that we don't understand.

8 years agoMerge pull request #768 from hannob/master
Tim Kientzle [Sun, 21 Aug 2016 16:22:43 +0000 (09:22 -0700)] 
Merge pull request #768 from hannob/master

Remove unused variable to avoid compiler warning.

8 years agoconditionally declare r on ACL_TYPE_NFS4 768/head
Hanno [Thu, 18 Aug 2016 03:51:59 +0000 (05:51 +0200)] 
conditionally declare r on ACL_TYPE_NFS4

8 years agoRemove unused variable to avoid compiler warning.
Hanno [Wed, 17 Aug 2016 08:31:26 +0000 (10:31 +0200)] 
Remove unused variable to avoid compiler warning.

8 years agoMerge pull request #755 from mmatuska/fix_acl_get_flagset_np
Tim Kientzle [Sat, 13 Aug 2016 16:38:16 +0000 (09:38 -0700)] 
Merge pull request #755 from mmatuska/fix_acl_get_flagset_np

fix use of acl_get_flagset_np() on FreeBSD

8 years agoCorrect the usage of PATH_MAX as reported in Issue #744.
Tim Kientzle [Wed, 10 Aug 2016 01:35:38 +0000 (21:35 -0400)] 
Correct the usage of PATH_MAX as reported in Issue #744.

8 years agoMerge branch 'master' of github.com:libarchive/libarchive
Tim Kientzle [Wed, 10 Aug 2016 01:32:00 +0000 (21:32 -0400)] 
Merge branch 'master' of github.com:libarchive/libarchive

8 years agoTest cases for Github Issue #744, #745, and #746.
Tim Kientzle [Wed, 10 Aug 2016 01:31:36 +0000 (21:31 -0400)] 
Test cases for Github Issue #744, #745, and #746.

8 years agoMerge pull request #757 from Tarsnap/compile-test-zip-high
Tim Kientzle [Fri, 5 Aug 2016 19:57:05 +0000 (12:57 -0700)] 
Merge pull request #757 from Tarsnap/compile-test-zip-high

Add test_..._zip_high_compression.c to automake

9 years agoAdd test_..._zip_high_compression.c to automake 757/head
Graham Percival [Fri, 5 Aug 2016 17:55:46 +0000 (10:55 -0700)] 
Add test_..._zip_high_compression.c to automake

Sponsored by: Tarsnap Backup Inc.

9 years agofix use of acl_get_flagset_np() on FreeBSD 755/head
Martin Matuska [Thu, 4 Aug 2016 13:36:03 +0000 (15:36 +0200)] 
fix use of acl_get_flagset_np() on FreeBSD

9 years agoMerge pull request #751 from aahancoc/patch-1
Tim Kientzle [Wed, 27 Jul 2016 01:58:19 +0000 (18:58 -0700)] 
Merge pull request #751 from aahancoc/patch-1

Fixed build error in test_read_format_zip_high_compression.c

9 years agoFixed build error in test_read_format_zip_high_compression.c 751/head
aahancoc [Tue, 26 Jul 2016 20:03:54 +0000 (16:03 -0400)] 
Fixed build error in test_read_format_zip_high_compression.c

C89 (what this library is compiled in) requires for loop iterators to be initialized outside the for loop.

9 years agoIssue 748: Regression test
Tim Kientzle [Sun, 24 Jul 2016 05:11:46 +0000 (22:11 -0700)] 
Issue 748: Regression test

9 years agoMerge pull request #750 from cpatulea/master
Tim Kientzle [Sun, 24 Jul 2016 00:49:22 +0000 (17:49 -0700)] 
Merge pull request #750 from cpatulea/master

ustar: fix out of bounds read on empty string ("") filename.

9 years agoustar: fix out of bounds read on empty string ("") filename. 750/head
Catalin Patulea [Fri, 22 Jul 2016 15:27:31 +0000 (11:27 -0400)] 
ustar: fix out of bounds read on empty string ("") filename.

9 years agoIssue 720: Allow dashes in third-party version numbers
Tim Kientzle [Thu, 21 Jul 2016 03:27:45 +0000 (20:27 -0700)] 
Issue 720: Allow dashes in third-party version numbers

9 years agoIssue 695: Check return value of fwrite/fclose
Tim Kientzle [Thu, 21 Jul 2016 03:22:25 +0000 (20:22 -0700)] 
Issue 695:  Check return value of fwrite/fclose

9 years agoMerge pull request #739 from ppentchev/roam-cppcheck
Tim Kientzle [Sun, 17 Jul 2016 20:05:01 +0000 (13:05 -0700)] 
Merge pull request #739 from ppentchev/roam-cppcheck

Fix two minor complaints by cppcheck.

9 years agoMerge branch 'master' of github.com:libarchive/libarchive
Tim Kientzle [Sun, 17 Jul 2016 15:29:23 +0000 (08:29 -0700)] 
Merge branch 'master' of github.com:libarchive/libarchive

9 years agoFix check for GCC >= 4.8
Tim Kientzle [Sun, 17 Jul 2016 15:28:51 +0000 (08:28 -0700)] 
Fix check for GCC >= 4.8

9 years agoMerge pull request #737 from ppentchev/roam-typo-candidate
Tim Kientzle [Sun, 17 Jul 2016 14:36:34 +0000 (07:36 -0700)] 
Merge pull request #737 from ppentchev/roam-typo-candidate

Fix a typographical error in a struct member: candiate -> candidate

9 years agoMerge pull request #736 from ppentchev/roam-typos
Tim Kientzle [Sun, 17 Jul 2016 14:35:45 +0000 (07:35 -0700)] 
Merge pull request #736 from ppentchev/roam-typos

Fix a couple of typographical errors.

9 years agoFix two minor complaints by cppcheck. 739/head
Peter Pentchev [Thu, 7 Jul 2016 15:22:04 +0000 (18:22 +0300)] 
Fix two minor complaints by cppcheck.

9 years agoFix a typo in a struct member: candiate -> candidate. 737/head
Peter Pentchev [Thu, 7 Jul 2016 08:12:08 +0000 (11:12 +0300)] 
Fix a typo in a struct member: candiate -> candidate.

9 years agoFix a couple of typographical errors. 736/head
Peter Pentchev [Thu, 7 Jul 2016 08:05:31 +0000 (11:05 +0300)] 
Fix a couple of typographical errors.

9 years agoMerge LZMA MP detection from configure.
Joerg Sonnenberger [Sat, 2 Jul 2016 08:44:03 +0000 (10:44 +0200)] 
Merge LZMA MP detection from configure.

9 years agoMerge struct vfsconf detection from configure for DragonFly.
Joerg Sonnenberger [Sat, 2 Jul 2016 08:34:06 +0000 (10:34 +0200)] 
Merge struct vfsconf detection from configure for DragonFly.

9 years agoMerge pull request #733 from tpgxyz/master
Tim Kientzle [Fri, 1 Jul 2016 02:17:03 +0000 (19:17 -0700)] 
Merge pull request #733 from tpgxyz/master

Support GNU tar --blocking-factor and GNU cpio --dereference

9 years agosort it 733/head
tpg [Thu, 30 Jun 2016 19:12:13 +0000 (21:12 +0200)] 
sort it

9 years agoadd GNU compatible blocking-factor alias, add cpio long alias for dereference
tpg [Wed, 29 Jun 2016 19:04:08 +0000 (21:04 +0200)] 
add GNU compatible blocking-factor alias, add cpio long alias for dereference

9 years agoMerge pull request #727 from jcfr/use-cmake-builtin-module-CheckStructHasMember
Tim Kientzle [Tue, 28 Jun 2016 05:11:35 +0000 (22:11 -0700)] 
Merge pull request #727 from jcfr/use-cmake-builtin-module-CheckStructHasMember

Use CMake module "CheckStructHasMember" and remove custom "CheckStructMember"

9 years agoFreeBSD bug 204157: If filesystem does not support birthtime, then make all birthtim...
Tim Kientzle [Fri, 24 Jun 2016 05:16:03 +0000 (22:16 -0700)] 
FreeBSD bug 204157:  If filesystem does not support birthtime, then make all birthtime comparisons succeed.

This is a bit of a hack:  We should also have tests that verify
different birthtimes.  When/if that happens, we'll need to add
a real system probe to determine whether birthtime is actually
supported on this particular filesystem and skip tests accordingly.

9 years agoIssue 686: Keep retrying until buffer is big enough
Tim Kientzle [Fri, 24 Jun 2016 05:07:19 +0000 (22:07 -0700)] 
Issue 686: Keep retrying until buffer is big enough

@duy-github pointed out that commit 9134c3a fixed
one place where this mistake occurred, but left three
other places.

9 years agoFor getvfsbyname on DragonFly, struct vfsconf has to be used and not
Joerg Sonnenberger [Thu, 23 Jun 2016 13:16:16 +0000 (15:16 +0200)] 
For getvfsbyname on DragonFly, struct vfsconf has to be used and not
struct xvfsconf as on FreeBSD.

9 years agoAvoid use of C99 for-scope declarations to fix issue #729.
Joerg Sonnenberger [Wed, 22 Jun 2016 21:09:27 +0000 (23:09 +0200)] 
Avoid use of C99 for-scope declarations to fix issue #729.

9 years agoIgnore the MT encoder in XZ 5.2 prerelease versions.
Joerg Sonnenberger [Wed, 22 Jun 2016 21:03:43 +0000 (23:03 +0200)] 
Ignore the MT encoder in XZ 5.2 prerelease versions.

9 years agoOlder Solaris at least has no _PATH_TTY, so provide a fallback.
Joerg Sonnenberger [Wed, 22 Jun 2016 13:00:34 +0000 (15:00 +0200)] 
Older Solaris at least has no _PATH_TTY, so provide a fallback.

9 years agoUse CMake module "CheckStructHasMember" and remove custom "CheckStructMember" 727/head
Jean-Christophe Fillion-Robin [Wed, 22 Jun 2016 06:40:54 +0000 (02:40 -0400)] 
Use CMake module "CheckStructHasMember" and remove custom "CheckStructMember"

9 years agoDummy out copy_acls if sys/acl.h is missing or ACL support is disabled.
Joerg Sonnenberger [Tue, 21 Jun 2016 16:29:07 +0000 (18:29 +0200)] 
Dummy out copy_acls if sys/acl.h is missing or ACL support is disabled.

9 years agoMerge pull request #726 from mmatuska/cpio_version
Tim Kientzle [Tue, 21 Jun 2016 15:01:34 +0000 (08:01 -0700)] 
Merge pull request #726 from mmatuska/cpio_version

Fix test for cpio version dash fix

9 years agoSIGRTMAX doesn't exist on all systems, so compute the largest used
Joerg Sonnenberger [Tue, 21 Jun 2016 13:21:20 +0000 (15:21 +0200)] 
SIGRTMAX doesn't exist on all systems, so compute the largest used
signal number. Don't bother with dynamically allocating the array, just
use a static array instead. Fix ctype use.

9 years agoIn the past, OpenSSL was not needed by the main libarchive when the hash
Joerg Sonnenberger [Tue, 21 Jun 2016 13:12:20 +0000 (15:12 +0200)] 
In the past, OpenSSL was not needed by the main libarchive when the hash
functions were provided by libc, e.g. on NetBSD. With the introduction
of the de-/encryption support, this is no longer the case. As such,
explicitly check for the necessary EVP routines as well to make sure
that -lcrypto is added as necessary.

9 years agoMany systems have bswap16 in system headers, so undef it first to avoid
Joerg Sonnenberger [Tue, 21 Jun 2016 13:03:08 +0000 (15:03 +0200)] 
Many systems have bswap16 in system headers, so undef it first to avoid
compiler warnings.

9 years agoFix ctype use. Avoid empty loop bodies.
Joerg Sonnenberger [Tue, 21 Jun 2016 12:45:37 +0000 (14:45 +0200)] 
Fix ctype use. Avoid empty loop bodies.

9 years agoFix argument range of ctype functions.
Joerg Sonnenberger [Tue, 21 Jun 2016 12:34:02 +0000 (14:34 +0200)] 
Fix argument range of ctype functions.

9 years agoSupport maintainer mode
Joerg Sonnenberger [Tue, 21 Jun 2016 12:23:36 +0000 (14:23 +0200)] 
Support maintainer mode

The time stamp checks for configure and Makefile.in are sometimes problematic, so allow disabling them.

9 years agoFix test for cpio version dash fix 726/head
Martin Matuska [Tue, 21 Jun 2016 06:30:54 +0000 (08:30 +0200)] 
Fix test for cpio version dash fix

9 years agoMerge pull request #725 from mmatuska/cpio_version
Tim Kientzle [Tue, 21 Jun 2016 02:54:56 +0000 (19:54 -0700)] 
Merge pull request #725 from mmatuska/cpio_version

Report cpio version string with one dash only

9 years agoReport cpio version string with one dash only 725/head
Martin Matuska [Mon, 20 Jun 2016 23:55:50 +0000 (01:55 +0200)] 
Report cpio version string with one dash only

9 years agoLibarchive 3.2.1 v3.2.1
Tim Kientzle [Mon, 20 Jun 2016 03:30:48 +0000 (20:30 -0700)] 
Libarchive 3.2.1

9 years agoAdd missing reference file to distribution
Tim Kientzle [Mon, 20 Jun 2016 02:56:02 +0000 (19:56 -0700)] 
Add missing reference file to distribution

9 years agoFix array overflow in test suite
Tim Kientzle [Sun, 19 Jun 2016 23:11:24 +0000 (16:11 -0700)] 
Fix array overflow in test suite