1
0
mirror of https://git.FreeBSD.org/src.git synced 2024-12-16 10:20:30 +00:00
Commit Graph

393 Commits

Author SHA1 Message Date
Tim Kientzle
a3ac45275d Correct the cpio writers to not accept data for non-regular files.
In particular, the previous code led to archives that had
non-empty bodies following directory entries.  Not a fatal
problem, as bsdtar and GNU cpio are both happy to just skip
this bogus data, but it still shouldn't be there.

MFC after: 3 days
2007-10-12 04:11:31 +00:00
Tim Kientzle
04a832627b Correct the return values of the final zero-length block at EOF.
Return EOF immediately if an entry in a ZIP archive has no body.
In particular, the latter issue was causing bsdtar to emit spurious
warnings when extracting directory entries from ZIP archives.

MFC after: 3 days
2007-10-12 04:08:28 +00:00
Tim Kientzle
016e328d16 Fix the archive_write_data() function so it always returns
number of bytes written, even when used to write files to
disk.  Extend the test suite to verify the correct return
values for archive_write_data() and archive_write_data_block().

Thanks to: Bruce Mah, for stepping in promptly to back out the
   earlier broken version of this fix
Thanks to: Colin Percival, for pointing out the correct fix
MFC after: 5 days
Approved by: re (ksmith)
Pointy hat: \me
2007-09-21 04:52:43 +00:00
Tim Kientzle
d2d0f66a30 Fill in a missing 'e'
Thanks to: Kai Wang, for pointing this out
Approved by: re (bmah)
MFC after: 3 days
2007-09-19 16:37:45 +00:00
Bruce A. Mah
7b7b893394 Revert the last commit to libarchive. It introduced some regresssions,
most noticably the incorrect extraction of files by bsdtar.

This commit reverts:

	src/lib/libarchive/archive_write_disk.c 1.15
	src/lib/libarchive/test/test_write_disk.c 1.4

Approved by:	re (implicitly)
2007-09-18 20:20:37 +00:00
Tim Kientzle
8f3ba4ccf4 Correct the return value from archive_write_data()
(when used to restore files to disk) to match:
  * The documentation
  * The return values of this function when used
    to write files into an archive.

Approved by: re (bmah)
Pointy hat: \me
MFC after: 5 days
2007-09-18 04:20:21 +00:00
Tim Kientzle
68f0154dcf This commit updates libarchive to be compatible with
GNU tar 1.17's implementation of --posix --sparse,
at the cost of losing compatibility with GNU tar 1.16.
Fortunately, the 1.17 implementation actually makes sense,
so the libarchive code is now a bit more straightforward
than before.

Background:  GNU tar 1.16 defined a new way to store
sparse files in --posix archives.  Unfortunately,
the implementation incorrectly inserted several
blocks of null padding after each such entry.
As a result, non-GNU tar implementations saw the
archive as truncated after any sparse entry.
This was fixed in GNU tar 1.17 at the cost of
losing compatibility with GNU tar 1.16 for this
new format (which is not the default, so hopefully
rarely used).  Libarchive recently gained support
for reading the GNU tar 1.16 formats; this commit
updates it to read the GNU tar 1.17 variant instead.

Approved by: re (ksmith for libarchive portion)
Approved by: re (blanket for libarchive_test portion)
MFC after: 5 days
2007-08-18 21:53:25 +00:00
Tim Kientzle
9136384dc2 Make suid/sgid restore be "opportunistic" if
owner restore is not requested.  If you ask
for permissions to be restored but not owner,
you will now get no error if suid/sgid bits
cannot be set.  (It's a security hole to restore
suid/sgid bits if the owner/group aren't restored.)

This fixes an obscure problem where a simple
"tar -xf" with no other options will sometimes
fail gratuitously because of suid/sgid bits.
This is causing occasional problems for people
using bsdtar as a drop-in replacement for
"that other tar program." ;-)

Note: If you do ask for owner restore, then suid/sgid
restore failures still issue an error.  This
only suppresses the error in the case where an
suid/sgid bit restore fails because of an owner
mismatch and owner restore was not requested.

Approved by: re (bmah)
MFC after: 7 days
2007-08-12 17:35:05 +00:00
Tim Kientzle
f6d19aee04 Update the tests for reading the various GNU tar sparse formats.
In particular:
  * Include a second entry in all of the test archives (to catch errors
    with intermediate padding)
  * Test the GNU tar 1.17 version of "posix sparse format 1.0"
    instead of the GNU tar 1.16 version (the latter is no longer
    supported by GNU tar).

Right now, libarchive fails this test because I originally
implemented the GNU tar 1.16 version of "posix sparse format 1.0".
I'll fix libarchive shortly.

Approved by: re (blanket, libarchive testing)
2007-08-12 01:16:19 +00:00
Tim Kientzle
9e3fd30bac Two minor nits:
* Allow libarchive_test to compile on Interix again.
  * Track the test name (not just line number) when counting skipped tests.

Thanks to: Joerg Sonnenberger
Approved by: re (blanket; libarchive testing)
2007-07-31 05:03:27 +00:00
Tim Kientzle
8ea9716c0c Fix a strict aliasing warning from GCC 4.1.
Thanks to: Joerg Sonnenberger
Approved by: re (hrs)
MFC after: 3 days
2007-07-20 01:28:50 +00:00
Tim Kientzle
d3bb697513 archive_string_ensure() used to call exit(3) if it
couldn't allocate more memory for a string.  Change
this so it returns NULL in that case, and update
all of its callers to handle the error.  Some of
those callers can now return errors back to the
client instead of calling exit(3).

Approved by: re (bmah)
2007-07-15 19:13:59 +00:00
Tim Kientzle
75d0856ca5 Add archive_entry_copy_gname() and archive_entry_copy_uname()
functions.

Approved by: re (bmah)
MFC after: 3 days
2007-07-15 19:10:34 +00:00
Tim Kientzle
58bdc3275c Clarify one test.
Approved by: re (blanket, libarchive testing)
2007-07-15 17:16:42 +00:00
Tim Kientzle
a902b04074 Use an uncompressed test archive for gtar sparse format 1.0 format,
as that better exercises some internal read-combining logic than the
compressed archive.

Approved by: re (blanket, libarchive testing)
2007-07-14 17:54:14 +00:00
Tim Kientzle
2e4e46b540 Report each failed test once, but keep a count and report that count
if there was more than one.  In particular, this simplifies
test_tar_filenames.c, which has a tendency to be very noisy otherwise.

Approved by: re (blanket, libarchive testing)
2007-07-14 17:52:01 +00:00
Tim Kientzle
46dd1e6ee7 Restore the 'break' that was inadvertently removed in 1.57 of this file.
Without this, hardlinks get returned as symlinks.

Approved by: re (Ken Smith)
MFC after: 2 days
2007-07-14 05:53:51 +00:00
Tim Kientzle
f7f698bb22 Extend the basic tar reading test to exercise most types of
entries.  This doesn't cover everything yet, but it's a big improvement.

Approved by:  re (blanket, libarchive testing)
2007-07-14 05:35:17 +00:00
Tim Kientzle
99d4bb3999 Make the test for reading gtar sparse entries more robust;
it now verifies that the returned blocks have the correct data
at the correct file offsets, ignoring any null padding that
may exist.

Approved by: re (blanket, libarchive test suite)
2007-07-13 15:16:26 +00:00
Tim Kientzle
ceab112df6 New test suite test_read_pax_truncated probes libarchive
behavior with truncated or damaged pax archives.  This
tests most of the cases covered by the recent security advisory.

Approved by: re (blanket, libarchive test suite)
2007-07-13 15:14:35 +00:00
Tim Kientzle
8ed6656123 New file "read_open_memory.c" is a custom variant of
archive_read_open_memory.c that tries to test border
cases.  In particular, it copies over each returned block
so that formats or decompressors that read past the end
of a returned block will break.

Approved by: re (blanket, libarchive test suite)
2007-07-13 15:12:52 +00:00
Tim Kientzle
4b0489dd31 Fix running individual tests via "libarchive_test <number> <number> ..."
Approved by: re (blanket)
2007-07-13 15:09:07 +00:00
Colin Percival
612c3e7724 Correct multiple security issues in how libarchive handles corrupt
tar archives, including a potentially exploitable buffer overflow.

Approved by:	re (kensmith, security blanket)
Reviewed by:	kientzle
Security:	FreeBSD-SA-07:05.libarchive
2007-07-12 15:00:28 +00:00
Tim Kientzle
ed3ba42250 Fix installworld: /usr/bin/printf isn't available then,
so use awk's printf for the formatting here instead.

Pointy hat: Yours Truly
Approved by: re
2007-07-07 16:55:42 +00:00
Tim Kientzle
dbb4eb7d9c Make test suite work with libarchive 1.3.1: Take advantage of
ARCHIVE_VERSION_STAMP to selectively disable tests that don't
apply to that version; new "skipping()" function reports skipped
tests; modify final summary to report component test failures and
skips.

Note:  I don't currently intend to MFC the test suite itself;
anyone interested should just checkout and use this version
of the test suite, which should work for any library version.

Approved by: re (Ken Smith, blanket)
2007-07-06 15:43:11 +00:00
Tim Kientzle
ab16ac785a New "version stamp" simplifies determining the exact version
of libarchive being used.  I've been taking advantage of this
with a recent round of updates to libarchive_test so that it
can test older and newer versions of the library.

Approved by: re (Ken Smith)
2007-07-06 15:36:38 +00:00
Tim Kientzle
3f6c3bcd84 Fix 'bsdtar -t' on tape drives. Libarchive uses the
skip() callback to skip over data when reading uncompressed
archives.  This gets invoked, for example, during tar -t
or tar -x with a filename argument.  The revised code
only calls [lf]seek() on regular files, instead of depending
on the kernel to return an error.

Thanks to: bde for explaining the implementation of lseek()
Thanks to: Daniel O'Connor for testing
Approved by: re (Ken Smith)
MFC after: 5 days
2007-06-26 03:06:48 +00:00
Tim Kientzle
734f944a06 Ouch. I partially screwed up the last commit by
enabling a test that's not ready yet. <sigh>

Pointy hat: /me
Approved by: re@
2007-06-22 05:49:13 +00:00
Tim Kientzle
17e60e6230 Support for writing the 'newc' cpio format, plus a minimal test harness
for the cpio formats.

Thanks to: Rudolf Marek
Approved by: re@
2007-06-22 05:47:00 +00:00
Tim Kientzle
7faa61e65d Track whether the current read stream supports seek(). For now, we
assume yes unless seek has previously failed, but I fear I'll have to
avoid seeks under other circumstances.  (For instance, tape drives on
FreeBSD seem to return garbage from lseek().)  Also, optimize away
zero-byte skips.
2007-06-18 00:36:54 +00:00
Tim Kientzle
d02e3c38ab Make 'ar' write test a tad more portable. 2007-06-13 03:40:02 +00:00
Tim Kientzle
0ddfde5d16 Read support for the new GNU tar sparse formats added in gtar 1.15 and
gtar 1.16.
2007-06-13 03:35:37 +00:00
Tim Kientzle
147e16377c Add some options to libarchive_test:
-k: like make -k, try to keep going after errors.
   -q: quiet
2007-06-13 03:30:46 +00:00
Tim Kientzle
e09eb83674 Fix a broken function declaration. 2007-06-13 03:29:15 +00:00
Stephane E. Potvin
04031e9ae2 Options spring cleanup:
- Add and document the KVM and KVM_SUPPORT options that
are needed for the ifmcstats(3) makefile
- Garbage collect unused variables
- Add missing inclusion of bsd.own.mk where needed

Approved by: kan (mentor)
Reviewed by: ru
2007-06-13 02:08:04 +00:00
Tim Kientzle
8b63161182 Don't lose leading '/' for pathnames exactly 101 bytes long.
Also, update the test harness to exercise this case.
2007-06-11 05:17:30 +00:00
Tim Kientzle
b48b40f1f8 libarchive 2.2.3
* "compression_program" support uses an external program
  * Portability: no longer uses "struct stat" as a primary
    data interchange structure internally
  * Part of the above: refactor archive_entry to separate
    out copy_stat() and stat() functions
  * More complete tests for archive_entry
  * Finish archive_entry_clone()
  * Isolate major()/minor()/makedev() in archive_entry; remove
    these from everywhere else.
  * Bug fix: properly handle decompression look-ahead at end-of-data
  * Bug fixes to 'ar' support
  * Fix memory leak in ZIP reader
  * Portability: better timegm() emulation in iso9660 reader
  * New write_disk flags to suppress auto dir creation and not
    overwrite newer files (for future cpio front-end)
  * Simplify trailing-'/' fixup when writing tar and pax
  * Test enhancements:  fix various compiler warnings, improve
    portability, add lots of new tests.
  * Documentation: document new functions, first draft of
    libarchive_internals.3

MFC after: 14 days
Thanks to: Joerg Sonnenberger (compression_program)
Thanks to: Kai Wang (ar)
Thanks to: Colin Percival (many small fixes)
Thanks to: Many others who sent me various patches and problem reports.
2007-05-29 01:00:21 +00:00
Colin Percival
a30eadfc15 Replace "sizeof(struct bucket[cache_size])" with
"cache_size * sizeof(struct bucket)".  The former is valid in C99 but can
confuse earlier compilers, while the latter is a standard idiom which all
C compilers understand.

Approved by:	kientzle
2007-05-22 15:37:58 +00:00
Colin Percival
3662c7b8ad Don't test for NULL when it is both unnecessary (the pointer is checked
against NULL when it is first allocated) and pointless (we've already
dereferenced the pointer several times).

Found by:	Coverity Prevent(tm)
CID:		3204
2007-05-21 04:45:24 +00:00
Colin Percival
51bbfbd15c Remove pointless code: Don't assign a value to a variable when we're
going to overwrite it with a new value a few lines later.

Visual inspection of the surrounding code indicates that the code does
what it's supposed to do; i.e., the pointless code wasn't supposed to
be doing something other than what it was doing.

CID:		3323
Found by:	Coverity Prevent(tm)
2007-05-21 04:22:38 +00:00
Colin Percival
a7e2f06359 Don't forget to free a string allocated by strdup where necessary.
Found by:	Coverity Prevent
Coverity ID:	CID 3132
2007-05-19 05:09:09 +00:00
Colin Percival
9fbf190fc5 s/@VERSION@/@ARCHIVE_VERSION@/
This is a no-op as far as FreeBSD is concerned, but makes libarchive
more autoconf-friendly.

Approved by:	kientzle
2007-05-02 05:29:55 +00:00
Tim Kientzle
f1a870981d Fix a memory leak in the uname/gname lookup cache.
Thanks to: VMiklos
2007-04-20 15:32:13 +00:00
Colin Percival
6fc0ea8474 In libarchive: Downgrade ARCHIVE_FATAL and ARCHIVE_FAILED errors which
occur on the write side of extracting a file to ARCHIVE_WARN errors
when returning them from archive_read_extract.

In bsdtar: Use the return code from archive_read_data_into_fd and
archive_read_extract to determine whether we should continue trying to
extract an archive after one of the entries fails.

This commit makes extracting a truncated tarball complain once about
the archive being truncated, instead of complaining twice (once when
trying to extract an entry, and once when trying to seek to the next
entry).

Discussed with:	kientzle
2007-04-16 04:04:50 +00:00
Tim Kientzle
2a5449a7ca Properly cleanup the UID/GID lookup data. 2007-04-15 04:43:12 +00:00
Tim Kientzle
1eb296a82c Don't free a NULL pointer. 2007-04-15 04:42:52 +00:00
Tim Kientzle
0c3e4592a6 Clean up a lot of memory leaks in the libarchive test harness. 2007-04-15 04:30:02 +00:00
Tim Kientzle
edddd60899 Thanks to Colin for catching my mixup. The original problem wasn't
suppressing the second error, it was failing to address the first.
2007-04-15 01:01:20 +00:00
Tim Kientzle
f912fb118f Consolidate numeric limit macros in one place; include them
only on platforms that need them.  FreeBSD doesn't.
2007-04-15 00:53:38 +00:00
Tim Kientzle
b723fe5b61 Pass through error message if any operation fails, not just the last one. 2007-04-14 22:49:37 +00:00