- Pacman-dev - lists.archlinux.org

[pacman-dev] [PATCH] makepkg: handle pgp signatures with .sign extension
by Allan McRae 07 Nov '11

07 Nov '11

Detached sgnature files with extension .sign are accepted by gnupg. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- scripts/makepkg.sh.in | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index 3085bf5..5d048d2 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -334,7 +334,7 @@ in_array() { source_has_signatures(){ local file for file in "${source[@]}"; do - if [[ $file = *.@(sig|asc) ]]; then + if [[ $file = *.@(sig?(n)|asc) ]]; then return 0 fi done @@ -704,7 +704,7 @@ check_pgpsigs() { for file in "${source[@]}"; do file="$(get_filename "$file")" - if [[ ! $file = *.@(sig|asc) ]]; then + if [[ ! $file = *.@(sig?(n)|asc) ]]; then continue fi -- 1.7.7.2

1 0

[pacman-dev] Split debug support
by Martin Stolpe 04 Nov '11

04 Nov '11

Hello, I'm interested in the split debug support for makepkg. There seems to be some patches floating around: http://mailman.archlinux.org/pipermail/pacman- dev/2011-August/014203.html but they don't seem ready to be accepted. I have quite a few questions on my mind, hopefully somebody can answer them: -What are the requirements for implementing split packages? -Does anybody know how this is handled in other distros? -Why shouldn't the implementation use linux specific tools? -In reference to the previos question: Would it be possible to make this a linux only option? -Is makepkg only used in Archlinux and its dervatives? Regards Martin

1 0

[pacman-dev] [PATCH] Fix speed output edge case
by Allan McRae 03 Nov '11

03 Nov '11

Occasionally when the download rate showed 100.0 the output got messed up. This was caused by the rounding of a number between 99.95 and 100. Adjust the threshold to avoid this rounding issue. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- src/pacman/callback.c | 7 ++----- 1 files changed, 2 insertions(+), 5 deletions(-) diff --git a/src/pacman/callback.c b/src/pacman/callback.c index 5d7aed3..03d6266 100644 --- a/src/pacman/callback.c +++ b/src/pacman/callback.c @@ -698,11 +698,8 @@ void cb_dl_progress(const char *filename, off_t file_xfered, off_t file_total) printf(" %ls%-*s ", wcfname, padwid, ""); /* We will show 1.6M/s, 11.6M/s, but 116K/s and 1116K/s */ - if(rate_human < 9.95) { - printf("%6.1f %3s %1.2f%c/s ", - xfered_human, xfered_label, rate_human, rate_label[0]); - } else if(rate_human < 99.95) { - printf("%6.1f %3s %2.1f%c/s ", + if(rate_human < 99.95) { + printf("%6.1f %3s %4.1f%c/s ", xfered_human, xfered_label, rate_human, rate_label[0]); } else { printf("%6.1f %3s %4.f%c/s ", -- 1.7.7.2

2 2

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v4.0.0-69-g90477f1
by dan＠archlinux.org 01 Nov '11

01 Nov '11

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "The official pacman repository". The branch, master has been updated via 90477f156c18659eee6e8be75122169060309990 (commit) via 8b3717ef0d55719337b061b761979a5fd9b48fbc (commit) via 6df558177fac6bdcbd60de05038d6e6b7f224bd3 (commit) via ed3cd7573600bf8be2bda01ad75da8bae2f589fb (commit) via ba7a056d586f78954ac874a3bebc6325f7e6eeb1 (commit) via a4ce3edf95f148bc6eac16b6edd9bc281c7745af (commit) via d98ff04cc933253b7221fc170813f7d2365d3289 (commit) via 4c259d51f7e0d5fac0158f70b0ad3b0043930943 (commit) via f4875fab9bbe3011446032bebcbe232448a3e8d7 (commit) via 3343185473c79d556b29bda30c9c482c7fef5289 (commit) via 1052709921da50ec5203efaa313dbda75625a4e1 (commit) via 24881034f62cecb9e601fdfee9458f34f32b2507 (commit) via 8a9ce12a27970beb644952a83d27f54d1d7f751a (commit) via 5c1885b55eae870a10c91793d05a74d74a075e83 (commit) via 7a6b01d46c1da385dd34466e757beafa33e02afa (commit) via d5f0395dc13fb305e75a5e78de57e148d87f12d7 (commit) from a708e7d28e890c3e230a215629b7f5c58c57f58d (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- commit 90477f156c18659eee6e8be75122169060309990 Author: Dan McGee <dan(a)archlinux.org> Date: Thu Oct 27 17:23:12 2011 -0500 libalpm/util: don't use sprintf to convert from bin to hex This is a trivial operation that doesn't require calling a function over and over- just do some math and indexing into a character array. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 8b3717ef0d55719337b061b761979a5fd9b48fbc Author: Dan McGee <dan(a)archlinux.org> Date: Thu Oct 27 17:18:28 2011 -0500 Fix size reported in CALLOC allocation failure message Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 6df558177fac6bdcbd60de05038d6e6b7f224bd3 Author: Dan McGee <dan(a)archlinux.org> Date: Wed Oct 26 15:51:46 2011 -0500 Convert package and database archive reads to use file descriptors This gives us a bit more control and over the archive reading process, and a bit less is done behind the scenes. It also allows us to use fstat() in preference to stat(), which should avoid some potential race conditions. Some reorganization is necessary to move the stat calls after the open() calls. Error handling and cleanup in general is also improved, as we had several potential memory and file handle leaks before in some error paths. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit ed3cd7573600bf8be2bda01ad75da8bae2f589fb Author: Dan McGee <dan(a)archlinux.org> Date: Thu Oct 27 16:24:30 2011 -0500 libalpm/util: use low-level I/O for copyfile and checksum routines This removes an unnecessary level of buffering. We are not doing line-based I/O here, so we can read in blocks of 8K at a time directly from the file. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit ba7a056d586f78954ac874a3bebc6325f7e6eeb1 Author: Dan McGee <dan(a)archlinux.org> Date: Thu Oct 27 15:47:18 2011 -0500 Add OPEN() and CLOSE() util macros These wrap the normal open() and close() low-level I/O calls and ensure EINTR is handled correctly. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit a4ce3edf95f148bc6eac16b6edd9bc281c7745af Merge: d5f0395 d98ff04 Author: Dan McGee <dan(a)archlinux.org> Date: Tue Nov 1 10:26:45 2011 -0500 Merge branch 'maint' commit d5f0395dc13fb305e75a5e78de57e148d87f12d7 Author: Dan McGee <dan(a)archlinux.org> Date: Wed Oct 26 14:55:50 2011 -0500 libalpm/util: two stat() related cleanups First, use fstat() in preference to stat() since we already have an open file handle. This also removes the need to check for a symlink as that is not possible when a file is opened. Next, use archive_entry_mode() rather than archive_entry_stat() as we only use the mode portion of the stat struct and the call is much cheaper. Also delay it until it is necessary. Signed-off-by: Dan McGee <dan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: contrib/paccache.in | 3 + lib/libalpm/add.c | 17 ++-- lib/libalpm/be_package.c | 72 ++++++++----- lib/libalpm/be_sync.c | 56 ++++++---- lib/libalpm/conflict.c | 33 ++++--- lib/libalpm/dload.c | 16 +++- lib/libalpm/util.c | 199 +++++++++++++++++----------------- lib/libalpm/util.h | 21 ++++- src/pacman/conf.c | 5 +- src/util/Makefile.am | 2 +- test/pacman/tests/fileconflict007.py | 2 +- test/pacman/tests/symlink010.py | 24 ++++ test/pacman/tests/symlink011.py | 24 ++++ test/pacman/tests/symlink012.py | 24 ++++ 14 files changed, 316 insertions(+), 182 deletions(-) create mode 100644 test/pacman/tests/symlink010.py create mode 100644 test/pacman/tests/symlink011.py create mode 100644 test/pacman/tests/symlink012.py hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [GIT] The official pacman repository branch, maint, updated. v4.0.0-32-gd98ff04
by dan＠archlinux.org 01 Nov '11

01 Nov '11

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "The official pacman repository". The branch, maint has been updated via d98ff04cc933253b7221fc170813f7d2365d3289 (commit) via 4c259d51f7e0d5fac0158f70b0ad3b0043930943 (commit) via f4875fab9bbe3011446032bebcbe232448a3e8d7 (commit) from 3343185473c79d556b29bda30c9c482c7fef5289 (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- commit d98ff04cc933253b7221fc170813f7d2365d3289 Author: Dave Reisner <d(a)falconindy.com> Date: Sun Oct 30 10:14:37 2011 -0400 src/util: link vercmp against .lo, not the .o This seems to fix FS#26652. Signed-off-by: Dave Reisner <dreisner(a)archlinux.org> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 4c259d51f7e0d5fac0158f70b0ad3b0043930943 Author: Dave Reisner <dreisner(a)archlinux.org> Date: Sat Oct 22 13:16:39 2011 -0400 dload: remove redundant conditional Replacing the strdup when after the first NULL check assures that we get continue with payload->remote_name defined. Signed-off-by: Dave Reisner <dreisner(a)archlinux.org> commit f4875fab9bbe3011446032bebcbe232448a3e8d7 Author: Dave Reisner <dreisner(a)archlinux.org> Date: Mon Oct 17 09:47:06 2011 -0400 dload: chmod tempfiles to respect umask Dan: fix mask calculation, add it to the success/fail block instead. Signed-off-by: Dave Reisner <dreisner(a)archlinux.org> Signed-off-by: Dan McGee <dan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: lib/libalpm/dload.c | 18 +++++++++++++++--- src/util/Makefile.am | 2 +- 2 files changed, 16 insertions(+), 4 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH] src/util: link vercmp against .lo, not the .o
by Dave Reisner 01 Nov '11

01 Nov '11

This seems to fix FS#26652. Signed-off-by: Dave Reisner <dreisner(a)archlinux.org> --- I'm no libtool expert, but this seems to fix the build with --disable-static and the dependencies still look correct for vercmp. src/util/Makefile.am | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/util/Makefile.am b/src/util/Makefile.am index 11308fb..e4af56c 100644 --- a/src/util/Makefile.am +++ b/src/util/Makefile.am @@ -31,6 +31,6 @@ testpkg_SOURCES = testpkg.c testpkg_LDADD = $(top_builddir)/lib/libalpm/.libs/libalpm.la vercmp_SOURCES = vercmp.c -vercmp_LDADD = $(top_builddir)/lib/libalpm/version.o +vercmp_LDADD = $(top_builddir)/lib/libalpm/version.lo # vim:set ts=2 sw=2 noet: -- 1.7.7.1

2 1

[pacman-dev] [PATCH] libalpm/util: don't use sprintf to convert from bin to hex
by Dan McGee 27 Oct '11

27 Oct '11

This is a trivial operation that doesn't require calling a function over and over- just do some math and indexing into a character array. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/util.c | 24 ++++++++++++++---------- 1 files changed, 14 insertions(+), 10 deletions(-) diff --git a/lib/libalpm/util.c b/lib/libalpm/util.c index 5056b98..57869d3 100644 --- a/lib/libalpm/util.c +++ b/lib/libalpm/util.c @@ -816,6 +816,8 @@ static int sha2_file(const char *path, unsigned char output[32], int is224) } #endif +static const char *hex_digits = "0123456789abcdef"; + /** Get the md5 sum of file. * @param filename name of the file * @return the checksum on success, NULL on error @@ -829,8 +831,7 @@ char SYMEXPORT *alpm_compute_md5sum(const char *filename) ASSERT(filename != NULL, return NULL); - /* allocate 32 chars plus 1 for null */ - CALLOC(md5sum, 33, sizeof(char), return NULL); + MALLOC(md5sum, (size_t)33, return NULL); /* defined above for OpenSSL, otherwise defined in md5.h */ ret = md5_file(filename, output); @@ -841,10 +842,12 @@ char SYMEXPORT *alpm_compute_md5sum(const char *filename) /* Convert the result to something readable */ for (i = 0; i < 16; i++) { - /* sprintf is acceptable here because we know our output */ - sprintf(md5sum +(i * 2), "%02x", output[i]); + int pos = i * 2; + /* high 4 bits are first digit, low 4 are second */ + md5sum[pos] = hex_digits[output[i] >> 4]; + md5sum[pos + 1] = hex_digits[output[i] & 0x0f]; } - + md5sum[32] = '\0'; return md5sum; } @@ -861,8 +864,7 @@ char SYMEXPORT *alpm_compute_sha256sum(const char *filename) ASSERT(filename != NULL, return NULL); - /* allocate 64 chars plus 1 for null */ - CALLOC(sha256sum, 65, sizeof(char), return NULL); + MALLOC(sha256sum, (size_t)65, return NULL); /* defined above for OpenSSL, otherwise defined in sha2.h */ ret = sha2_file(filename, output, 0); @@ -873,10 +875,12 @@ char SYMEXPORT *alpm_compute_sha256sum(const char *filename) /* Convert the result to something readable */ for (i = 0; i < 32; i++) { - /* sprintf is acceptable here because we know our output */ - sprintf(sha256sum +(i * 2), "%02x", output[i]); + int pos = i * 2; + /* high 4 bits are first digit, low 4 are second */ + sha256sum[pos] = hex_digits[output[i] >> 4]; + sha256sum[pos + 1] = hex_digits[output[i] & 0x0f]; } - + sha256sum[64] = '\0'; return sha256sum; } -- 1.7.7

1 0

[pacman-dev] [GIT] The official pacman repository branch, maint, updated. v4.0.0-29-g3343185
by dan＠archlinux.org 27 Oct '11

27 Oct '11

This is an automated email from the git hooks/post-receive script. It was generated because a ref change was pushed to the repository containing the project "The official pacman repository". The branch, maint has been updated via 3343185473c79d556b29bda30c9c482c7fef5289 (commit) via 1052709921da50ec5203efaa313dbda75625a4e1 (commit) via 24881034f62cecb9e601fdfee9458f34f32b2507 (commit) via 8a9ce12a27970beb644952a83d27f54d1d7f751a (commit) via 5c1885b55eae870a10c91793d05a74d74a075e83 (commit) via 7a6b01d46c1da385dd34466e757beafa33e02afa (commit) from 2da59e1aa91fa4160ec1c8ec84d94d8141b2a832 (commit) Those revisions listed above that are new to this repository have not appeared on any other notification email; so we list those revisions in full, below. - Log ----------------------------------------------------------------- commit 3343185473c79d556b29bda30c9c482c7fef5289 Author: Dan McGee <dan(a)archlinux.org> Date: Thu Oct 27 14:59:24 2011 -0500 Introduce ALPM_BUFFER_SIZE constant This takes the place of three previously used constants: ARCHIVE_DEFAULT_BYTES_PER_BLOCK, BUFFER_SIZE, and CPBUFSIZE. In libarchive 3.0, the first constant will be no more, so we can ensure we are forward-compatible by removing our usage of it now. The rest are unified for consistency. By default, we will use the value of BUFSIZ provided by <stdio.h>, which is 8192 on Linux. If that is undefined, a default value is provided. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 1052709921da50ec5203efaa313dbda75625a4e1 Author: Dave Reisner <dreisner(a)archlinux.org> Date: Mon Oct 24 10:29:15 2011 -0400 paccache: add vim modeline Signed-off-by: Dave Reisner <dreisner(a)archlinux.org> commit 24881034f62cecb9e601fdfee9458f34f32b2507 Author: Dave Reisner <dreisner(a)archlinux.org> Date: Mon Oct 24 10:28:35 2011 -0400 paccache: ensure seen/seenarch vars are set Doesn't do a whole lot of good to compare against values that are never set. Fixes bug where -vvv output wasn't grouping packages together properly. Signed-off-by: Dave Reisner <dreisner(a)archlinux.org> commit 8a9ce12a27970beb644952a83d27f54d1d7f751a Author: Dan McGee <dan(a)archlinux.org> Date: Wed Oct 26 17:32:46 2011 -0500 Fix issues with replacing unowned symlinks There aretwo seperate issues in the same block of file conflict checking code here: 1) If realpath errored, such as when a symlink was broken, we would call 'continue' rather than simply exit this particular method of resolution. This was likely just a copy-paste mistake as the previous resolving steps all use loops where continue makes sense. Refactor the check so we only proceed if realpath is successful, and continue with the rest of the checks either way. 2) The real problem this code was trying to solve was canonicalizing path component (e.g., directory) symlinks. The final component, if not a directory, should not be handled at all in this loop. Add a !S_ISLNK() condition to the loop so we only call this for real files. There are few other small cleanups to the debug messages that I made while debugging this problem- we don't need to keep printing the file name, and ensure every block that sets resolved_conflict to true prints a debug message so we know how it was resolved. This fixes the expected failures from symlink010.py and symlink011.py, while still ensuring the fix for fileconflict007.py works. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 5c1885b55eae870a10c91793d05a74d74a075e83 Author: Dan McGee <dan(a)archlinux.org> Date: Wed Oct 26 16:43:36 2011 -0500 Add some unowned symlink replacement tests These should all prevent installation, and yet two of the three tests currently fail. Not good. The best way to see what is going on here is to diff the three new tests side by side- there is only a small difference between the three tests, and that is in the destination of the symlink in question that should never be overwritten. symlink010.py: myprogsuffix -> myprog symlink011.py: myprogsuffix -> broken symlink012.py: myprogsuffix -> otherprog Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 7a6b01d46c1da385dd34466e757beafa33e02afa Author: Dan McGee <dan(a)archlinux.org> Date: Wed Oct 26 15:44:55 2011 -0500 Don't realloc a 0-length files array when loading packages There is some pecular behavior going on here when a package is loaded that has no files, as is very common in our test suite. When we enter the realloc/sort code, a package without files will call the following: files = realloc(NULL, 0); One would assume this is a no-op, returning a NULL pointer, but that is not the case and valgrind later reports we are leaking memory. Fix the whole thing by skipping the reallocation and sort steps if the pointer is NULL, as we have nothing to do. Note that the package still gets marked as 'files loaded', becuase although there were none, we tried and were successful. Signed-off-by: Dan McGee <dan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: contrib/paccache.in | 3 +++ lib/libalpm/add.c | 11 +++++++---- lib/libalpm/be_package.c | 22 ++++++++++++---------- lib/libalpm/be_sync.c | 2 +- lib/libalpm/conflict.c | 33 +++++++++++++++++++-------------- lib/libalpm/util.c | 20 ++++++++------------ lib/libalpm/util.h | 7 +++++++ test/pacman/tests/fileconflict007.py | 2 +- test/pacman/tests/symlink010.py | 24 ++++++++++++++++++++++++ test/pacman/tests/symlink011.py | 24 ++++++++++++++++++++++++ test/pacman/tests/symlink012.py | 24 ++++++++++++++++++++++++ 11 files changed, 130 insertions(+), 42 deletions(-) create mode 100644 test/pacman/tests/symlink010.py create mode 100644 test/pacman/tests/symlink011.py create mode 100644 test/pacman/tests/symlink012.py hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH 1/3] libalpm/util: two stat() related cleanups
by Dan McGee 26 Oct '11

26 Oct '11

First, use fstat() in preference to stat() since we already have an open file handle. This also removes the need to check for a symlink as that is not possible when a file is opened. Next, use archive_entry_mode() rather than archive_entry_stat() as we only use the mode portion of the stat struct and the call is much cheaper. Also delay it until it is necessary. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/util.c | 27 +++++++++++++-------------- 1 files changed, 13 insertions(+), 14 deletions(-) diff --git a/lib/libalpm/util.c b/lib/libalpm/util.c index 0d499ad..af9cf42 100644 --- a/lib/libalpm/util.c +++ b/lib/libalpm/util.c @@ -160,12 +160,10 @@ int _alpm_copyfile(const char *src, const char *dest) } } - /* chmod dest to permissions of src, as long as it is not a symlink */ + /* chmod dest to permissions of src */ struct stat statbuf; - if(!stat(src, &statbuf)) { - if(! S_ISLNK(statbuf.st_mode)) { - fchmod(fileno(out), statbuf.st_mode); - } + if(!fstat(fileno(in), &statbuf)) { + fchmod(fileno(out), statbuf.st_mode); } else { /* stat was unsuccessful */ ret = 1; @@ -313,18 +311,11 @@ int _alpm_unpack(alpm_handle_t *handle, const char *archive, const char *prefix, } while(archive_read_next_header(_archive, &entry) == ARCHIVE_OK) { - const struct stat *st; - const char *entryname; /* the name of the file in the archive */ + const char *entryname; + mode_t mode; - st = archive_entry_stat(entry); entryname = archive_entry_pathname(entry); - if(S_ISREG(st->st_mode)) { - archive_entry_set_perm(entry, 0644); - } else if(S_ISDIR(st->st_mode)) { - archive_entry_set_perm(entry, 0755); - } - /* If specific files were requested, skip entries that don't match. */ if(list) { char *entry_prefix = strdup(entryname); @@ -345,6 +336,13 @@ int _alpm_unpack(alpm_handle_t *handle, const char *archive, const char *prefix, } } + mode = archive_entry_mode(entry); + if(S_ISREG(mode)) { + archive_entry_set_perm(entry, 0644); + } else if(S_ISDIR(mode)) { + archive_entry_set_perm(entry, 0755); + } + /* Extract the archive entry. */ int readret = archive_read_extract(_archive, entry, 0); if(readret == ARCHIVE_WARN) { @@ -555,6 +553,7 @@ int _alpm_run_chroot(alpm_handle_t *handle, const char *path, char *const argv[] } umask(0022); execv(path, argv); + /* execv only returns if there was an error */ fprintf(stderr, _("call to execv failed (%s)\n"), strerror(errno)); exit(1); } else { -- 1.7.7

3 4

[pacman-dev] [PATCH 1/3] Don't realloc a 0-length files array when loading packages
by Dan McGee 26 Oct '11

26 Oct '11

There is some pecular behavior going on here when a package is loaded that has no files, as is very common in our test suite. When we enter the realloc/sort code, a package without files will call the following: files = realloc(NULL, 0); One would assume this is a no-op, returning a NULL pointer, but that is not the case and valgrind later reports we are leaking memory. Fix the whole thing by skipping the reallocation and sort steps if the pointer is NULL, as we have nothing to do. Note that the package still gets marked as 'files loaded', becuase although there were none, we tried and were successful. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- For maint; so valgrind messages can stay cleaner and it is a pretty low-risk, obvious fix. lib/libalpm/be_package.c | 18 ++++++++++-------- 1 files changed, 10 insertions(+), 8 deletions(-) diff --git a/lib/libalpm/be_package.c b/lib/libalpm/be_package.c index c20c703..c8e08e2 100644 --- a/lib/libalpm/be_package.c +++ b/lib/libalpm/be_package.c @@ -482,17 +482,19 @@ alpm_pkg_t *_alpm_pkg_load_internal(alpm_handle_t *handle, newpkg->origin_data.file = strdup(pkgfile); newpkg->ops = get_file_pkg_ops(); newpkg->handle = handle; + newpkg->infolevel = INFRQ_BASE | INFRQ_DESC | INFRQ_SCRIPTLET; if(full) { - /* attempt to hand back any memory we don't need */ - files = realloc(files, sizeof(alpm_file_t) * files_count); - /* "checking for conflicts" requires a sorted list, ensure that here */ - _alpm_log(handle, ALPM_LOG_DEBUG, "sorting package filelist for %s\n", pkgfile); - newpkg->files.files = files_msort(files, files_count); + if(files) { + /* attempt to hand back any memory we don't need */ + files = realloc(files, sizeof(alpm_file_t) * files_count); + /* "checking for conflicts" requires a sorted list, ensure that here */ + _alpm_log(handle, ALPM_LOG_DEBUG, + "sorting package filelist for %s\n", pkgfile); + newpkg->files.files = files_msort(files, files_count); + } newpkg->files.count = files_count; - newpkg->infolevel = INFRQ_BASE | INFRQ_DESC | INFRQ_FILES | INFRQ_SCRIPTLET; - } else { - newpkg->infolevel = INFRQ_BASE | INFRQ_DESC | INFRQ_SCRIPTLET; + newpkg->infolevel |= INFRQ_FILES; } return newpkg; -- 1.7.7

1 2