- Pacman-dev - lists.archlinux.org

[pacman-dev] [PATCH] pacman-key: fix broken logic around updating trust database
by Dan McGee 29 Aug '11

29 Aug '11

Any option that flips UPDATEDB=1 doesn't work right now due to what we thought was a good idea in commit cab1379a1ab14. Fix this by not including the update operation in the option count and special casing it where necessary. Also, bring back the helpful "Updating trust database" message. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- Allan broke it! :) But in all seriousness, this code is getting a bit hacky/dirty, so open to other suggestions if we can clean anything up. scripts/pacman-key.sh.in | 14 ++++++++++---- 1 files changed, 10 insertions(+), 4 deletions(-) diff --git a/scripts/pacman-key.sh.in b/scripts/pacman-key.sh.in index 1bac3d3..0b46b10 100644 --- a/scripts/pacman-key.sh.in +++ b/scripts/pacman-key.sh.in @@ -419,13 +419,16 @@ PACMAN_KEYRING_DIR=${PACMAN_KEYRING_DIR:-$(get_from "$CONFIG" "GPGDir" || echo " GPG_PACMAN=(gpg --homedir ${PACMAN_KEYRING_DIR} --no-permission-warning) # check only a single operation has been given +# don't include UPDATEDB in here as other opts can induce it numopt=$(( ADD + DELETE + EDITKEY + EXPORT + FINGER + IMPORT + IMPORT_TRUSTDB + - INIT + LISTKEYS + LISTSIGS + POPULATE + RECEIVE + UPDATEDB + VERIFY )) + INIT + LISTKEYS + LISTSIGS + POPULATE + RECEIVE + VERIFY )) case $numopt in 0) - error "$(gettext "no operation specified (use -h for help)")" - exit 1 + if (( ! UPDATEDB )); then + error "$(gettext "no operation specified (use -h for help)")" + exit 1 + fi ;; [!1]) error "$(gettext "Multiple operations specified")" @@ -450,6 +453,9 @@ esac (( RECEIVE )) && receive_keys (( VERIFY )) && "${GPG_PACMAN[@]}" --verify $SIGNATURE -(( UPDATEDB )) && "${GPG_PACMAN[@]}" --batch --check-trustdb +if (( UPDATEDB )); then + msg "$(gettext "Updating trust database...")" + "${GPG_PACMAN[@]}" --batch --check-trustdb +fi # vim: set ts=2 sw=2 noet: -- 1.7.6.1

1 0

[pacman-dev] [PATCH] Streamline alpm_splitdep() comparisons
by Dan McGee 29 Aug '11

29 Aug '11

This reduces from 5 to 3 the number of searches needed on the string looking for a comparison operator, since we can so a second quick comparison looking for '=' if we find '<' or '>'. It also makes every search doable with strchr() or memchr() rather than the slower strstr() method. In testing, only 10% of splitdep calls (~1600 / 16000) during an -Ss database load found a version comparison operator, so optimizing the not found path to be require less work makes sense. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/deps.c | 32 +++++++++++++++++++------------- 1 files changed, 19 insertions(+), 13 deletions(-) diff --git a/lib/libalpm/deps.c b/lib/libalpm/deps.c index e268157..639f14b 100644 --- a/lib/libalpm/deps.c +++ b/lib/libalpm/deps.c @@ -410,31 +410,37 @@ alpm_depend_t *_alpm_splitdep(const char *depstring) { alpm_depend_t *depend; const char *ptr, *version = NULL; + size_t deplen; if(depstring == NULL) { return NULL; } CALLOC(depend, 1, sizeof(alpm_depend_t), return NULL); + deplen = strlen(depstring); /* Find a version comparator if one exists. If it does, set the type and * increment the ptr accordingly so we can copy the right strings. */ - if((ptr = strstr(depstring, ">="))) { - depend->mod = ALPM_DEP_MOD_GE; - version = ptr + 2; - } else if((ptr = strstr(depstring, "<="))) { - depend->mod = ALPM_DEP_MOD_LE; - version = ptr + 2; - } else if((ptr = strstr(depstring, "="))) { + if((ptr = memchr(depstring, '<', deplen))) { + if(ptr[1] == '=') { + depend->mod = ALPM_DEP_MOD_LE; + version = ptr + 2; + } else { + depend->mod = ALPM_DEP_MOD_LT; + version = ptr + 1; + } + } else if((ptr = memchr(depstring, '>', deplen))) { + if(ptr[1] == '=') { + depend->mod = ALPM_DEP_MOD_GE; + version = ptr + 2; + } else { + depend->mod = ALPM_DEP_MOD_GT; + version = ptr + 1; + } + } else if((ptr = memchr(depstring, '=', deplen))) { /* Note: we must do =,<,> checks after <=, >= checks */ depend->mod = ALPM_DEP_MOD_EQ; version = ptr + 1; - } else if((ptr = strstr(depstring, "<"))) { - depend->mod = ALPM_DEP_MOD_LT; - version = ptr + 1; - } else if((ptr = strstr(depstring, ">"))) { - depend->mod = ALPM_DEP_MOD_GT; - version = ptr + 1; } else { /* no version specified, leave version and ptr NULL */ depend->mod = ALPM_DEP_MOD_ANY; -- 1.7.6.1

1 0

[pacman-dev] [PATCH] Run pacman test-suite with LC_ALL=C
by Allan McRae 29 Aug '11

29 Aug '11

Running the pacman test-suite in a non-English locale results in a few failures. Force the test-suite to run with LC_ALL=C. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- Not sure if this is the best approach or whether we should just have the three tests that require it to run under the C locale. Makefile.am | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/Makefile.am b/Makefile.am index 75760bb..fb7a4d3 100644 --- a/Makefile.am +++ b/Makefile.am @@ -22,7 +22,7 @@ dist_pkgdata_DATA = \ # run the pactest test suite and vercmp tests check-local: test/pacman test/util src/pacman src/util - $(PYTHON) $(top_srcdir)/test/pacman/pactest.py --debug=1 \ + LC_ALL=C $(PYTHON) $(top_srcdir)/test/pacman/pactest.py --debug=1 \ --test $(top_srcdir)/test/pacman/tests/*.py \ -p $(top_builddir)/src/pacman/pacman $(SH) $(top_srcdir)/test/util/pacsorttest.sh \ -- 1.7.6.1

3 2

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v4.0.0rc1-124-gf46db04
by dan＠archlinux.org 29 Aug '11

29 Aug '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 f46db04f98cb63b48208026865943e42e7ece2f4 (commit) via 8973875a1fc52ec35c255afd34c9cd7d5c285caa (commit) via b221af660d101301606e263cef73b358937516a6 (commit) via 040083b97fab61c8afc16a1c49a8384d097c272a (commit) via 3a458783a2617b7b22a756f95b5f66f3da79146f (commit) via 12a6c77fdd0c465761f4f9518eff0eeeda668d20 (commit) via 7ceeebf1505dba655b43e095f5392367a3a0f9b8 (commit) via e1b9f7b3005c4d6db9cc84e95a42d4beba6c7e24 (commit) via 29dede2eb76ab5a139d4e8236be1037a7a86b6e5 (commit) via cab1379a1ab14e29414c5fdf6252d1f5ea7263fb (commit) via 710e83999bbf8ae614b9310744c5306a7cbc858b (commit) via d1240f67eab6dfa7e3344cd84dd8849b38ce337c (commit) from 2cfcc874b9332ad207398b9e20dc8880d93e8ae4 (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 f46db04f98cb63b48208026865943e42e7ece2f4 Merge: 8973875 12a6c77 Author: Dan McGee <dan(a)archlinux.org> Date: Sun Aug 28 23:51:05 2011 -0500 Merge branch 'allan/pacman-key' commit 8973875a1fc52ec35c255afd34c9cd7d5c285caa Author: Dan McGee <dan(a)archlinux.org> Date: Sun Aug 28 23:05:20 2011 -0500 _alpm_splitdep(): don't pass bogus length value to strndup If we fell through to the ALPM_DEP_MOD_ANY case, ptr would be NULL, and we would pass (0 - <str>), which is a rather large negative number or bogus positive number, depending on signed/unsigned. Just use strdup in the case where we don't have a ptr available. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit b221af660d101301606e263cef73b358937516a6 Author: Dan McGee <dan(a)archlinux.org> Date: Sun Aug 28 22:20:41 2011 -0500 Database read optimizations Hard to believe there was still more room to improve on this, but I found an easily correctable oversight tonight. Our databases (both sync and local) contain many blank lines, and we were not moving onto the next line right away in these cases; instead we would proceed through our strcmp() conditional checks as normal. Some local numbers follow to show the effects of this patch: Sync `-Ss foobarbaz`: 71,709 blank lines skipped early ~1,505,889 strcmp() calls avoided (21 per line) ~15% speed improvement (.210 --> .179 sec) Local `-Qs foobarbaz`: 6,823 blank lines skipped early 115,991 strcmp() calls avoided (17 per line) ~6% speed improvement (.080 -> .071 sec) Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 040083b97fab61c8afc16a1c49a8384d097c272a Author: Dan McGee <dan(a)archlinux.org> Date: Sun Aug 28 23:41:17 2011 -0500 Allow access to package origin data Add new alpm_pkg_get_origin() method, use it in the front end now that the enum constants are publicly available. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 3a458783a2617b7b22a756f95b5f66f3da79146f Author: Diogo Sousa <diogogsousa(a)gmail.com> Date: Mon Aug 29 02:51:32 2011 +0100 Removed multiple definitions of pkgfrom_t libalpm now exports type alpm_pkgfrom_t in alpm.h, which may be used by frontends. Pacman now uses alpm_pkgfrom_t instead of replicating that type (pkg_from as was in src/pacman/package.h) Updated API change in README. Signed-off-by: Diogo Sousa <diogogsousa(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 12a6c77fdd0c465761f4f9518eff0eeeda668d20 Author: Dan McGee <dan(a)archlinux.org> Date: Thu Aug 25 12:59:27 2011 -0500 pacman-key: have --init add more options to default gpg.conf This adds a add_gpg_conf_option() helper function which tries to be intelligent and only add not found options, and those which have not been explicitly commented out. The new options added are 'no-greeting', 'no-permission-warning', and a default 'keyserver'. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 7ceeebf1505dba655b43e095f5392367a3a0f9b8 Author: Dan McGee <dan(a)archlinux.org> Date: Thu Aug 25 12:59:26 2011 -0500 pacman-key: refine permission and locking checks * secring.gpg can be 600, readable by root user only * ensure grep for lock-never option in check_keyring doesn't catch comments Signed-off-by: Dan McGee <dan(a)archlinux.org> commit e1b9f7b3005c4d6db9cc84e95a42d4beba6c7e24 Author: Allan McRae <allan(a)archlinux.org> Date: Tue Aug 23 16:10:06 2011 +1000 pacman-key: rework and document holding keys in keyring The HoldKey option was undocumented and was not suited for pacman.conf. Instead use the file "/etc/pacman.d/gnupg/heldkeys" to contain a list of keys not to be removed from the pacman keyring with the --populate option. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 29dede2eb76ab5a139d4e8236be1037a7a86b6e5 Author: Allan McRae <allan(a)archlinux.org> Date: Tue Aug 23 15:46:46 2011 +1000 pacman-key: Improve documentation for --populate Signed-off-by: Allan McRae <allan(a)archlinux.org> commit cab1379a1ab14e29414c5fdf6252d1f5ea7263fb Author: Allan McRae <allan(a)archlinux.org> Date: Tue Aug 23 15:12:08 2011 +1000 pacman-key: update trust database for relevant operations After most operations that touch the keyring, it is a good idea to always run a check on the trustdb as this prevents gpg complaining on later operations. Inspiration-from: Dave Reisner <dreisner(a)archlinux.org> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 710e83999bbf8ae614b9310744c5306a7cbc858b Author: Allan McRae <allan(a)archlinux.org> Date: Tue Aug 23 15:00:19 2011 +1000 pacman-key: import everything then revoke on --populate The optimization of only importing keys that were not to be later revoked was a not smart enough. For example, if a key was in both a repos keyring and its revoke list, alternate runs of pacman-key --populate would add then remove the key from the pacman keyring. This problem is made worse when considering the possibility of multiple keyrings being imported. Instead, import all keys followed by the revoking of all keys. This may result in a key being added then revoked, but that is not much of an issue given that is a very fast operation. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit d1240f67eab6dfa7e3344cd84dd8849b38ce337c Author: Allan McRae <allan(a)archlinux.org> Date: Mon Aug 15 22:51:58 2011 +1000 pacman-key: rework importing distro/repo provided keyrings The current --reload option, apart from being non-clear in its naming, is very limited in that only one keyring can be provided. A distribution may want to provide multiple keyrings for various subsets of its organisation or custom repo providers may also want to provide a keyring. This patch adds a --populate option that reads keyrings from (by default) /usr/share/pacman/keyrings. A keyring is named foo.gpg, with optional foo-revoked file providing a list of revoked key ids. These files are required to be signed (detached) by a key trusted by pacman-key, in practice probably by the key that signed the package providing these files. The --populate flag either updates the pacman keyring using all keyrings in the directory or individual keyrings can be specified. Signed-off-by: Allan McRae <allan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: README | 4 +- doc/pacman-key.8.txt | 23 +++++- lib/libalpm/alpm.h | 11 +++ lib/libalpm/be_local.c | 8 ++- lib/libalpm/be_sync.c | 5 +- lib/libalpm/deps.c | 6 +- lib/libalpm/package.c | 7 ++ lib/libalpm/package.h | 6 -- scripts/Makefile.am | 1 + scripts/pacman-key.sh.in | 223 ++++++++++++++++++++++++---------------------- src/pacman/package.c | 7 +- src/pacman/package.h | 9 +-- src/pacman/query.c | 4 +- src/pacman/sync.c | 4 +- 14 files changed, 183 insertions(+), 135 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH] Standardize output when db.lck is present
by pangyanhan＠gmail.com 29 Aug '11

29 Aug '11

From: Pang Yan Han <pangyanhan(a)gmail.com> When db.lck file is present, pacman -Syy and other Sync operations involving transactions show: :: Synchronizing package databases... error: failed to update core (unable to lock database) error: failed to update extra (unable to lock database) error: failed to update community (unable to lock database) error: failed to update multilib (unable to lock database) error: failed to synchronize any databases Whereas pacman -U <pkg> shows: error: failed to init transaction (unable to lock database) if you're sure a package manager is not already running, you can remove /var/lib/pacman/db.lck which is much more meaningful. Standardize the output to that of pacman -U when db.lck is present. Reported-by: Allan McRae <allan(a)archlinux.org> Signed-off-by: Pang Yan Han <pangyanhan(a)gmail.com> --- src/pacman/sync.c | 9 +++++++++ 1 files changed, 9 insertions(+), 0 deletions(-) diff --git a/src/pacman/sync.c b/src/pacman/sync.c index ea32a26..37f6ecb 100644 --- a/src/pacman/sync.c +++ b/src/pacman/sync.c @@ -288,6 +288,15 @@ static int sync_synctree(int level, alpm_list_t *syncs) ret = alpm_db_update((level < 2 ? 0 : 1), db); if(ret < 0) { + enum _alpm_errno_t err = alpm_errno(config->handle); + pm_fprintf(stderr, ALPM_LOG_ERROR, _("failed to init transaction (%s)\n"), + alpm_strerror(err)); + if(err == ALPM_ERR_HANDLE_LOCK) { + fprintf(stderr, _(" if you're sure a package manager is not already\n" + " running, you can remove %s\n"), + alpm_option_get_lockfile(config->handle)); + return 0; + } pm_fprintf(stderr, ALPM_LOG_ERROR, _("failed to update %s (%s)\n"), alpm_db_get_name(db), alpm_strerror(alpm_errno(config->handle))); } else if(ret == 1) { -- 1.7.6.1.385.gb7fcd0

2 1

[pacman-dev] [PATCH] Allow access to package origin data
by Dan McGee 29 Aug '11

29 Aug '11

Add new alpm_pkg_get_origin() method, use it in the front end now that the enum constants are publicly available. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- README | 2 +- lib/libalpm/alpm.h | 5 +++++ lib/libalpm/package.c | 7 +++++++ src/pacman/package.c | 7 +++---- src/pacman/package.h | 2 +- src/pacman/query.c | 4 ++-- src/pacman/sync.c | 4 ++-- 7 files changed, 21 insertions(+), 10 deletions(-) diff --git a/README b/README index fbd40d2..c14fd3c 100644 --- a/README +++ b/README @@ -429,7 +429,7 @@ API CHANGES BETWEEN 3.5 AND 4.0 alpm_option_get_default_siglevel(), alpm_option_set_default_siglevel(), alpm_option_get_gpgdir(), alpm_option_set_gpgdir(), alpm_db_get_siglevel(), alpm_sigresult_cleanup(), alpm_db_check_pgp_signature(), alpm_pkg_check_pgp_signature(), - alpm_pkg_get_sha256sum(), alpm_pkg_get_base64_sig() + alpm_pkg_get_origin(), alpm_pkg_get_sha256sum(), alpm_pkg_get_base64_sig() - list functions: alpm_list_to_array(), alpm_list_previous() - structs: diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index 2f62565..42e6038 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -598,6 +598,11 @@ const char *alpm_pkg_get_name(alpm_pkg_t *pkg); */ const char *alpm_pkg_get_version(alpm_pkg_t *pkg); +/** Returns the origin of the package. + * @return a #alpm_pkgfrom_t constant, -1 on error + */ +alpm_pkgfrom_t alpm_pkg_get_origin(alpm_pkg_t *pkg); + /** Returns the package description. * @param pkg a pointer to package * @return a reference to an internal string diff --git a/lib/libalpm/package.c b/lib/libalpm/package.c index 3d73a43..487b56d 100644 --- a/lib/libalpm/package.c +++ b/lib/libalpm/package.c @@ -185,6 +185,13 @@ const char SYMEXPORT *alpm_pkg_get_version(alpm_pkg_t *pkg) return pkg->version; } +alpm_pkgfrom_t SYMEXPORT alpm_pkg_get_origin(alpm_pkg_t *pkg) +{ + ASSERT(pkg != NULL, return -1); + pkg->handle->pm_errno = 0; + return pkg->origin; +} + const char SYMEXPORT *alpm_pkg_get_desc(alpm_pkg_t *pkg) { ASSERT(pkg != NULL, return NULL); diff --git a/src/pacman/package.c b/src/pacman/package.c index f7065d5..5654944 100644 --- a/src/pacman/package.c +++ b/src/pacman/package.c @@ -61,7 +61,7 @@ static void deplist_display(const char *title, * @param from the type of package we are dealing with * @param extra should we show extra information */ -void dump_pkg_full(alpm_pkg_t *pkg, alpm_pkgfrom_t from, int extra) +void dump_pkg_full(alpm_pkg_t *pkg, int extra) { const char *reason; time_t bdate, idate; @@ -69,10 +69,9 @@ void dump_pkg_full(alpm_pkg_t *pkg, alpm_pkgfrom_t from, int extra) const char *label; double size; alpm_list_t *requiredby = NULL; + alpm_pkgfrom_t from; - if(pkg == NULL) { - return; - } + from = alpm_pkg_get_origin(pkg); /* set variables here, do all output below */ bdate = alpm_pkg_get_builddate(pkg); diff --git a/src/pacman/package.h b/src/pacman/package.h index 4bbc2ba..890b4fe 100644 --- a/src/pacman/package.h +++ b/src/pacman/package.h @@ -22,7 +22,7 @@ #include <alpm.h> -void dump_pkg_full(alpm_pkg_t *pkg, alpm_pkgfrom_t from, int extra); +void dump_pkg_full(alpm_pkg_t *pkg, int extra); void dump_pkg_backups(alpm_pkg_t *pkg); void dump_pkg_files(alpm_pkg_t *pkg, int quiet); diff --git a/src/pacman/query.c b/src/pacman/query.c index ec98c99..5f2b37e 100644 --- a/src/pacman/query.c +++ b/src/pacman/query.c @@ -460,9 +460,9 @@ static int display(alpm_pkg_t *pkg) if(config->op_q_info) { if(config->op_q_isfile) { - dump_pkg_full(pkg, PKG_FROM_FILE, 0); + dump_pkg_full(pkg, 0); } else { - dump_pkg_full(pkg, PKG_FROM_LOCALDB, config->op_q_info > 1); + dump_pkg_full(pkg, config->op_q_info > 1); } } if(config->op_q_list) { diff --git a/src/pacman/sync.c b/src/pacman/sync.c index a7fe5be..e97da37 100644 --- a/src/pacman/sync.c +++ b/src/pacman/sync.c @@ -469,7 +469,7 @@ static int sync_info(alpm_list_t *syncs, alpm_list_t *targets) alpm_pkg_t *pkg = alpm_list_getdata(k); if(strcmp(alpm_pkg_get_name(pkg), pkgstr) == 0) { - dump_pkg_full(pkg, PKG_FROM_SYNCDB, config->op_s_info > 1); + dump_pkg_full(pkg, config->op_s_info > 1); foundpkg = 1; break; } @@ -494,7 +494,7 @@ static int sync_info(alpm_list_t *syncs, alpm_list_t *targets) for(j = alpm_db_get_pkgcache(db); j; j = alpm_list_next(j)) { alpm_pkg_t *pkg = alpm_list_getdata(j); - dump_pkg_full(pkg, PKG_FROM_SYNCDB, config->op_s_info > 1); + dump_pkg_full(pkg, config->op_s_info > 1); } } } -- 1.7.6

1 0

[pacman-dev] [PATCH] Removed multiple definitions of pkgfrom_t
by Diogo Sousa 29 Aug '11

29 Aug '11

libalpm now exports type alpm_pkgfrom_t in alpm.h, which may be used by frontends. Pacman now uses alpm_pkgfrom_t instead of replicating that type (pkg_from as was in src/pacman/package.h) Updated API change in README. Signed-off-by: Diogo Sousa <diogogsousa(a)gmail.com> --- README | 2 +- lib/libalpm/alpm.h | 6 ++++++ lib/libalpm/package.h | 6 ------ src/pacman/package.c | 2 +- src/pacman/package.h | 9 +-------- 5 files changed, 9 insertions(+), 16 deletions(-) diff --git a/README b/README index d5071a9..fbd40d2 100644 --- a/README +++ b/README @@ -435,7 +435,7 @@ API CHANGES BETWEEN 3.5 AND 4.0 - structs: alpm_backup_t, alpm_file_t, alpm_filelist_t - enums: - alpm_siglevel_t, alpm_sigstatus_t, alpm_sigvalidity_t + alpm_siglevel_t, alpm_sigstatus_t, alpm_sigvalidity_t, alpm_pkgfrom_t - error codes: ALPM_ERR_DB_INVALID, ALPM_ERR_DB_INVALID_SIG, ALPM_ERR_GPGME, ALPM_ERR_PKG_INVALID_CHECKSUM, ALPM_ERR_PKG_INVALID_SIG, ALPM_ERR_SIG_INVALID, diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index c94cdf7..2f62565 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -60,6 +60,12 @@ typedef enum _alpm_pkgreason_t { ALPM_PKG_REASON_DEPEND = 1 } alpm_pkgreason_t; +typedef enum _alpm_pkgfrom_t { + PKG_FROM_FILE = 1, + PKG_FROM_LOCALDB, + PKG_FROM_SYNCDB +} alpm_pkgfrom_t; + /** Types of version constraints in dependency specs. */ typedef enum _alpm_depmod_t { /** No version constraint */ diff --git a/lib/libalpm/package.h b/lib/libalpm/package.h index 707405d..5962b1d 100644 --- a/lib/libalpm/package.h +++ b/lib/libalpm/package.h @@ -34,12 +34,6 @@ #include "db.h" #include "signing.h" -typedef enum _alpm_pkgfrom_t { - PKG_FROM_FILE = 1, - PKG_FROM_LOCALDB, - PKG_FROM_SYNCDB -} alpm_pkgfrom_t; - /** Package operations struct. This struct contains function pointers to * all methods used to access data in a package to allow for things such * as lazy package intialization (such as used by the file backend). Each diff --git a/src/pacman/package.c b/src/pacman/package.c index a8c15eb..f7065d5 100644 --- a/src/pacman/package.c +++ b/src/pacman/package.c @@ -61,7 +61,7 @@ static void deplist_display(const char *title, * @param from the type of package we are dealing with * @param extra should we show extra information */ -void dump_pkg_full(alpm_pkg_t *pkg, enum pkg_from from, int extra) +void dump_pkg_full(alpm_pkg_t *pkg, alpm_pkgfrom_t from, int extra) { const char *reason; time_t bdate, idate; diff --git a/src/pacman/package.h b/src/pacman/package.h index 6f71d20..4bbc2ba 100644 --- a/src/pacman/package.h +++ b/src/pacman/package.h @@ -22,14 +22,7 @@ #include <alpm.h> -/* TODO it would be nice if we didn't duplicate a backend type */ -enum pkg_from { - PKG_FROM_FILE = 1, - PKG_FROM_LOCALDB, - PKG_FROM_SYNCDB -}; - -void dump_pkg_full(alpm_pkg_t *pkg, enum pkg_from from, int extra); +void dump_pkg_full(alpm_pkg_t *pkg, alpm_pkgfrom_t from, int extra); void dump_pkg_backups(alpm_pkg_t *pkg); void dump_pkg_files(alpm_pkg_t *pkg, int quiet); -- 1.7.6.1

2 1

[pacman-dev] [PATCH] _alpm_splitdep(): don't pass bogus length value to strndup
by Dan McGee 29 Aug '11

29 Aug '11

If we fell through to the ALPM_DEP_MOD_ANY case, ptr would be NULL, and we would pass (0 - <str>), which is a rather large negative number or bogus positive number, depending on signed/unsigned. Just use strdup in the case where we don't have a ptr available. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/deps.c | 6 +++++- 1 files changed, 5 insertions(+), 1 deletions(-) diff --git a/lib/libalpm/deps.c b/lib/libalpm/deps.c index 47b7637..e268157 100644 --- a/lib/libalpm/deps.c +++ b/lib/libalpm/deps.c @@ -441,7 +441,11 @@ alpm_depend_t *_alpm_splitdep(const char *depstring) } /* copy the right parts to the right places */ - STRNDUP(depend->name, depstring, ptr - depstring, return NULL); + if(ptr) { + STRNDUP(depend->name, depstring, ptr - depstring, return NULL); + } else { + STRDUP(depend->name, depstring, return NULL); + } depend->name_hash = _alpm_hash_sdbm(depend->name); if(version) { STRDUP(depend->version, version, return NULL); -- 1.7.6

1 0

[pacman-dev] [PATCH] Database read optimizations
by Dan McGee 29 Aug '11

29 Aug '11

Hard to believe there was still more room to improve on this, but I found an easily correctable oversight tonight. Our databases (both sync and local) contain many blank lines, and we were not moving onto the next line right away in these cases; instead we would proceed through our strcmp() conditional checks as normal. Some local numbers follow to show the effects of this patch: Sync `-Ss foobarbaz`: 71,709 blank lines skipped early ~1,505,889 strcmp() calls avoided (21 per line) ~15% speed improvement (.210 --> .179 sec) Local `-Qs foobarbaz`: 6,823 blank lines skipped early 115,991 strcmp() calls avoided (17 per line) ~6% speed improvement (.080 -> .071 sec) Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/be_local.c | 8 +++++++- lib/libalpm/be_sync.c | 5 ++++- 2 files changed, 11 insertions(+), 2 deletions(-) diff --git a/lib/libalpm/be_local.c b/lib/libalpm/be_local.c index 1a46dfd..be02bb5 100644 --- a/lib/libalpm/be_local.c +++ b/lib/libalpm/be_local.c @@ -580,7 +580,13 @@ static int local_db_read(alpm_pkg_t *info, alpm_dbinfrq_t inforeq) goto error; } while(!feof(fp)) { - READ_NEXT(); + if(fgets(line, sizeof(line), fp) == NULL && !feof(fp)) { + goto error; + } + if(_alpm_strip_newline(line) == 0) { + /* length of stripped line was zero */ + continue; + } if(strcmp(line, "%NAME%") == 0) { READ_NEXT(); if(strcmp(line, info->name) != 0) { diff --git a/lib/libalpm/be_sync.c b/lib/libalpm/be_sync.c index ac99e05..12d5b7f 100644 --- a/lib/libalpm/be_sync.c +++ b/lib/libalpm/be_sync.c @@ -526,7 +526,10 @@ static int sync_db_read(alpm_db_t *db, struct archive *archive, int ret; while((ret = _alpm_archive_fgets(archive, &buf)) == ARCHIVE_OK) { char *line = buf.line; - _alpm_strip_newline(line); + if(_alpm_strip_newline(line) == 0) { + /* length of stripped line was zero */ + continue; + } if(strcmp(line, "%NAME%") == 0) { READ_NEXT(); -- 1.7.6

1 0

[pacman-dev] [PATCH] Removed multiple definitions of pkgfrom_t
by Diogo Sousa 29 Aug '11

29 Aug '11

libalpm now exports type alpm_pkgfrom_t in alpm.h, which may be used by frontends. Pacman now uses alpm_pkgfrom_t instead of replicating that type (pkg_from as was in src/pacman/package.h) Signed-off-by: Diogo Sousa <diogogsousa(a)gmail.com> --- lib/libalpm/alpm.h | 6 ++++++ lib/libalpm/package.h | 6 ------ src/pacman/package.c | 2 +- src/pacman/package.h | 9 +-------- 4 files changed, 8 insertions(+), 15 deletions(-) diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index fc8f0bc..b1e9a88 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -197,6 +197,12 @@ typedef struct _alpm_file_t { mode_t mode; } alpm_file_t; +typedef enum _alpm_pkgfrom_t { + PKG_FROM_FILE = 1, + PKG_FROM_LOCALDB, + PKG_FROM_SYNCDB +} alpm_pkgfrom_t; + /** Package filelist container */ typedef struct _alpm_filelist_t { size_t count; diff --git a/lib/libalpm/package.h b/lib/libalpm/package.h index 25d1b1a..a75f6ae 100644 --- a/lib/libalpm/package.h +++ b/lib/libalpm/package.h @@ -34,12 +34,6 @@ #include "db.h" #include "signing.h" -typedef enum _alpm_pkgfrom_t { - PKG_FROM_FILE = 1, - PKG_FROM_LOCALDB, - PKG_FROM_SYNCDB -} alpm_pkgfrom_t; - /** Package operations struct. This struct contains function pointers to * all methods used to access data in a package to allow for things such * as lazy package intialization (such as used by the file backend). Each diff --git a/src/pacman/package.c b/src/pacman/package.c index 80c6bf2..ef483ef 100644 --- a/src/pacman/package.c +++ b/src/pacman/package.c @@ -61,7 +61,7 @@ static void deplist_display(const char *title, * @param from the type of package we are dealing with * @param extra should we show extra information */ -void dump_pkg_full(alpm_pkg_t *pkg, enum pkg_from from, int extra) +void dump_pkg_full(alpm_pkg_t *pkg, alpm_pkgfrom_t from, int extra) { const char *reason; time_t bdate, idate; diff --git a/src/pacman/package.h b/src/pacman/package.h index 6f71d20..4bbc2ba 100644 --- a/src/pacman/package.h +++ b/src/pacman/package.h @@ -22,14 +22,7 @@ #include <alpm.h> -/* TODO it would be nice if we didn't duplicate a backend type */ -enum pkg_from { - PKG_FROM_FILE = 1, - PKG_FROM_LOCALDB, - PKG_FROM_SYNCDB -}; - -void dump_pkg_full(alpm_pkg_t *pkg, enum pkg_from from, int extra); +void dump_pkg_full(alpm_pkg_t *pkg, alpm_pkgfrom_t from, int extra); void dump_pkg_backups(alpm_pkg_t *pkg); void dump_pkg_files(alpm_pkg_t *pkg, int quiet); -- 1.7.6.1

3 3

[pacman-dev] [PATCH 1/3] lib/libalpm/signing.c: Fix memory leak in decode_signature() in case of error.
by Diogo Sousa 29 Aug '11

29 Aug '11

Signed-off-by: Diogo Sousa <diogogsousa(a)gmail.com> --- lib/libalpm/signing.c | 1 + 1 files changed, 1 insertions(+), 0 deletions(-) diff --git a/lib/libalpm/signing.c b/lib/libalpm/signing.c index 8db4a7b..67f7adf 100644 --- a/lib/libalpm/signing.c +++ b/lib/libalpm/signing.c @@ -175,6 +175,7 @@ static int decode_signature(const char *base64_data, size_t destlen = len * 3 / 4; MALLOC(*data, destlen, goto error); if(base64_decode(*data, &destlen, usline, len)) { + free(*data); goto error; } *data_len = destlen; -- 1.7.6.1

2 4

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v4.0.0rc1-112-g2cfcc87
by dan＠archlinux.org 29 Aug '11

29 Aug '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 2cfcc874b9332ad207398b9e20dc8880d93e8ae4 (commit) via dc3336c27728fc16d2f9e68cb818648e7ca88467 (commit) via c7e4005e5ced965db49d73067fe40c516b6c34df (commit) via de43d00db071a04653cff592607647bb9c01d025 (commit) via 12387ca4e5c16994bacad52c142ef94868ba545a (commit) via 0dad2f6e6223e2aa00c43b59f49bf83082a0d6bb (commit) via a12a4ea3960c98817aa60df5a8879d118a16e2a7 (commit) via 455ca55f4efecf48bf86904e4a67a4b124cf1838 (commit) via a2002b8f690e12a7e8c75a3a9184570d27709c71 (commit) via 87fb8f5d5795be9b2bcb025eac26c94d7bb1e7b3 (commit) via 84d6de806bd18f4a6c0940de5ee60904f650539c (commit) via 5a07b75b225cbc731fbfb915a9c9edf0c300ccc4 (commit) via be72e100159e6cd50ac64e3639f8e4f98218ccae (commit) via c5982a3eb5287afdec6c9b859f81795c773da8fe (commit) from e1dce078b247142f7db80bf2d338e431df34691e (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 2cfcc874b9332ad207398b9e20dc8880d93e8ae4 Author: Dan McGee <dan(a)archlinux.org> Date: Thu Aug 25 18:29:00 2011 -0500 Better error handling out of package load method There are many other ways to fail a package load other than "file not found". We should also use the correct error code in this case. Clean it up a bit in the various callers. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit dc3336c27728fc16d2f9e68cb818648e7ca88467 Author: Dan McGee <dan(a)archlinux.org> Date: Thu Aug 25 16:40:39 2011 -0500 Refactor some args out of pkg_load_internal Just pass the entire sync package in if we have it; that way we can do any necessary operations involving it rather than have a parameter list growing endlessly. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit c7e4005e5ced965db49d73067fe40c516b6c34df Author: Dan McGee <dan(a)archlinux.org> Date: Thu Aug 25 10:29:02 2011 -0500 Add more info to debug key display Signed-off-by: Dan McGee <dan(a)archlinux.org> commit de43d00db071a04653cff592607647bb9c01d025 Author: Dan McGee <dan(a)archlinux.org> Date: Wed Aug 24 13:24:42 2011 -0500 Refactor signature result return format I was trying to take a shortcut and not introduce a wrapper struct for the signature results, so packed it all into alpm_sigresult_t in the first iteration. However, this is painful when one wants to add new fields or only return information regarding a single signature. Refactor the type into a few components which are exposed to the end user, and will allow a lot more future flexibility. This also exposes more information regarding the key to the frontend than was previously available. The "private" void *data pointer is used by the library to store the actual key object returned by gpgme; it is typed this way so the frontend has no expectations of what is there, and so we don't have any hard gpgme requirement in our public API. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 12387ca4e5c16994bacad52c142ef94868ba545a Author: Diogo Sousa <diogogsousa(a)gmail.com> Date: Sat Aug 27 22:03:45 2011 +0100 lib/libalpm/signing.c: Fix memory leak in decode_signature() in case of error. Signed-off-by: Diogo Sousa <diogogsousa(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 0dad2f6e6223e2aa00c43b59f49bf83082a0d6bb Author: Lukas Fleischer <archlinux(a)cryptocrack.de> Date: Fri Aug 26 13:10:38 2011 +0200 lib/libalpm/util.c: Fix two memory leaks Free "md5sum" if md5_file() fails in alpm_compute_md5sum(). Free "sha256sum" if sha2_file() fails in alpm_compute_sha256sum(). Signed-off-by: Lukas Fleischer <archlinux(a)cryptocrack.de> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit a12a4ea3960c98817aa60df5a8879d118a16e2a7 Author: Dan McGee <dan(a)archlinux.org> Date: Sun Aug 28 19:46:09 2011 -0500 Check local database status flag in db_write sanity check block Do all the checks at once; this also avoids the 'return' call after we have allocated memory for "pkgpath" as well as tweaked the umask. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 455ca55f4efecf48bf86904e4a67a4b124cf1838 Author: Lukas Fleischer <archlinux(a)cryptocrack.de> Date: Fri Aug 26 13:10:37 2011 +0200 be_sync.c: Fix memory leak in alpm_db_update() Free "syncpath" and restore umask if we fail to grab a lock. Signed-off-by: Lukas Fleischer <archlinux(a)cryptocrack.de> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit a2002b8f690e12a7e8c75a3a9184570d27709c71 Author: Lukas Fleischer <archlinux(a)cryptocrack.de> Date: Fri Aug 26 13:10:35 2011 +0200 pacman.c: Remove redundant strdup() in parsearg_global() config_set_arch() already calls strdup(). Remove strdup() from the config_set_arch() invocation to avoid a memory leak. Signed-off-by: Lukas Fleischer <archlinux(a)cryptocrack.de> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 87fb8f5d5795be9b2bcb025eac26c94d7bb1e7b3 Author: Dan McGee <dan(a)archlinux.org> Date: Thu Aug 25 18:37:23 2011 -0500 Make sync error message smarter on unfound targets We had two issues here. One is a file with an absolute path passed to -S results in a cryptic error message due to the database name being '\0'. The second is not realizing you should be doing -U instead of -S. Fix both of these to transform this: $ sudo pacman -S /tmp/binutils-2.21.1-2-i686.pkg.tar.xz error: database not found: to this: $ sudo pacman -S /tmp/binutils-2.21.1-2-i686.pkg.tar.xz error: target not found: /tmp/binutils-2.21.1-2-i686.pkg.tar.xz warning: '/tmp/binutils-2.21.1-2-i686.pkg.tar.xz' is a file, did you mean -U/--upgrade instead of -S/--sync? Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 84d6de806bd18f4a6c0940de5ee60904f650539c Author: Dan McGee <dan(a)archlinux.org> Date: Sat Aug 27 08:45:39 2011 -0500 Fix possible mismatched type with several curl arguments After commit 2e7d0023150664, we use off_t rather than long variables. Use the _LARGE variants of the methods to indicate we are passing off_t sized variables, and cast using (curl_off_t) accordingly. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 5a07b75b225cbc731fbfb915a9c9edf0c300ccc4 Author: Dan McGee <dan(a)archlinux.org> Date: Thu Aug 25 17:41:47 2011 -0500 Always process validity value returned by gpgme Don't force 'never'; you should be checking both the status and validity anyway. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit be72e100159e6cd50ac64e3639f8e4f98218ccae Author: Dan McGee <dan(a)archlinux.org> Date: Thu Aug 25 17:41:27 2011 -0500 Fix small memory leak in sig check code Signed-off-by: Dan McGee <dan(a)archlinux.org> commit c5982a3eb5287afdec6c9b859f81795c773da8fe Author: Dan McGee <dan(a)archlinux.org> Date: Thu Aug 25 17:14:19 2011 -0500 strtrim: don't move empty string There were many cases where the string coming in was a blank line, e.g. "\n\0", length 1. The trim routine starts by trimming leading spaces, thus trimming everything. We would then proceed to do a memmove of the NULL byte, which is completely worthless as we can just assign it instead. Signed-off-by: Dan McGee <dan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: lib/libalpm/alpm.h | 28 +++++++--- lib/libalpm/be_local.c | 6 +-- lib/libalpm/be_package.c | 68 ++++++++++++----------- lib/libalpm/be_sync.c | 2 + lib/libalpm/dload.c | 6 ++- lib/libalpm/package.h | 3 +- lib/libalpm/signing.c | 133 ++++++++++++++++++++++----------------------- lib/libalpm/signing.h | 2 +- lib/libalpm/sync.c | 3 +- lib/libalpm/util.c | 9 +++- src/pacman/package.c | 8 ++-- src/pacman/pacman.c | 2 +- src/pacman/query.c | 14 ++++- src/pacman/sync.c | 9 ++- src/pacman/util.c | 22 +++++--- src/pacman/util.h | 2 +- src/util/pactree.c | 7 ++- src/util/testpkg.c | 3 + valgrind.supp | 1 - 19 files changed, 186 insertions(+), 142 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH 1/4] pacman.c: Remove redundant strdup() in parsearg_global()
by Lukas Fleischer 29 Aug '11

29 Aug '11

config_set_arch() already calls strdup(). Remove strdup() from the config_set_arch() invocation to avoid a memory leak. Signed-off-by: Lukas Fleischer <archlinux(a)cryptocrack.de> --- src/pacman/pacman.c | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/src/pacman/pacman.c b/src/pacman/pacman.c index 71413bd..d285a05 100644 --- a/src/pacman/pacman.c +++ b/src/pacman/pacman.c @@ -385,7 +385,7 @@ static int parsearg_global(int opt) switch(opt) { case OP_ARCH: check_optarg(); - config_set_arch(strdup(optarg)); + config_set_arch(optarg); break; case OP_ASK: check_optarg(); -- 1.7.6.1

2 7

[pacman-dev] Inconsistent output when db.lck present
by Allan McRae 28 Aug '11

28 Aug '11

For example: allan@mugen ~ > sudo touch /var/lib/pacman/db.lck Password: allan@mugen ~ > pacman -Syu :: Synchronizing package databases... error: failed to update allanbrokeit (unable to lock database) error: failed to update kernel64 (unable to lock database) error: failed to update testing (unable to lock database) error: failed to update core (unable to lock database) error: failed to update extra (unable to lock database) error: failed to update community-testing (unable to lock database) error: failed to update community (unable to lock database) error: failed to synchronize any databases allan@mugen ~ > pacman -U /var/abs/local/pacman-git/pacman-git-20110828-1-i686.pkg.tar.xz error: failed to init transaction (unable to lock database) if you're sure a package manager is not already running, you can remove /var/lib/pacman/db.lck Allan

3 2

[pacman-dev] [PATCH] Fix possible mismatched type with several curl arguments
by Dan McGee 27 Aug '11

27 Aug '11

After commit 2e7d0023150664, we use off_t rather than long variables. Use the _LARGE variants of the methods to indicate we are passing off_t sized variables, and cast using (curl_off_t) accordingly. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/dload.c | 6 ++++-- 1 files changed, 4 insertions(+), 2 deletions(-) diff --git a/lib/libalpm/dload.c b/lib/libalpm/dload.c index 6ae6917..8cdb075 100644 --- a/lib/libalpm/dload.c +++ b/lib/libalpm/dload.c @@ -205,7 +205,8 @@ static void curl_set_handle_opts(struct dload_payload *payload, curl_easy_setopt(handle->curl, CURLOPT_WRITEHEADER, (void *)payload); if(payload->max_size) { - curl_easy_setopt(handle->curl, CURLOPT_MAXFILESIZE, payload->max_size); + curl_easy_setopt(handle->curl, CURLOPT_MAXFILESIZE_LARGE, + (curl_off_t)payload->max_size); } if(useragent != NULL) { @@ -220,7 +221,8 @@ static void curl_set_handle_opts(struct dload_payload *payload, } else if(stat(payload->tempfile_name, &st) == 0 && payload->allow_resume) { /* a previous partial download exists, resume from end of file. */ payload->tempfile_openmode = "ab"; - curl_easy_setopt(handle->curl, CURLOPT_RESUME_FROM, (long)st.st_size); + curl_easy_setopt(handle->curl, CURLOPT_RESUME_FROM_LARGE, + (curl_off_t)st.st_size); _alpm_log(handle, ALPM_LOG_DEBUG, "tempfile found, attempting continuation\n"); payload->initial_size = st.st_size; } -- 1.7.6

2 1

[pacman-dev] [PATCH 1/5] Remove trans is NULL check in QUESTION/EVENT/PROGRESS macros
by Dan McGee 26 Aug '11

26 Aug '11

trans cannot (and better not) be NULL at any point when these are being called. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/trans.h | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/lib/libalpm/trans.h b/lib/libalpm/trans.h index 9a60e71..6d5c08d 100644 --- a/lib/libalpm/trans.h +++ b/lib/libalpm/trans.h @@ -49,19 +49,19 @@ struct __alpm_trans_t { #define EVENT(t, e, d1, d2) \ do { \ - if((t) && (t)->cb_event) { \ + if((t)->cb_event) { \ (t)->cb_event(e, d1, d2); \ } \ } while(0) #define QUESTION(t, q, d1, d2, d3, r) \ do { \ - if((t) && (t)->cb_conv) { \ + if((t)->cb_conv) { \ (t)->cb_conv(q, d1, d2, d3, r); \ } \ } while(0) #define PROGRESS(t, e, p, per, h, r) \ do { \ - if((t) && (t)->cb_progress) { \ + if((t)->cb_progress) { \ (t)->cb_progress(e, p, per, h, r); \ } \ } while(0) -- 1.7.6

4 7

[pacman-dev] [PATCH 1/2] pacman-key: refine permission and locking checks
by Dan McGee 26 Aug '11

26 Aug '11

* secring.gpg can be 600, readable by root user only * ensure grep for lock-never option in check_keyring doesn't catch comments Signed-off-by: Dan McGee <dan(a)archlinux.org> --- scripts/pacman-key.sh.in | 6 +++--- 1 files changed, 3 insertions(+), 3 deletions(-) diff --git a/scripts/pacman-key.sh.in b/scripts/pacman-key.sh.in index 819ec69..5b4320d 100644 --- a/scripts/pacman-key.sh.in +++ b/scripts/pacman-key.sh.in @@ -108,7 +108,8 @@ initialize() { [[ -f ${PACMAN_KEYRING_DIR}/pubring.gpg ]] || touch ${PACMAN_KEYRING_DIR}/pubring.gpg [[ -f ${PACMAN_KEYRING_DIR}/secring.gpg ]] || touch ${PACMAN_KEYRING_DIR}/secring.gpg [[ -f ${PACMAN_KEYRING_DIR}/trustdb.gpg ]] || "${GPG_PACMAN[@]}" --update-trustdb - chmod 644 ${PACMAN_KEYRING_DIR}/{{pub,sec}ring,trustdb}.gpg + chmod 644 ${PACMAN_KEYRING_DIR}/{pubring,trustdb}.gpg + chmod 600 ${PACMAN_KEYRING_DIR}/secring.gpg # gpg.conf [[ -f ${PACMAN_KEYRING_DIR}/gpg.conf ]] || touch ${PACMAN_KEYRING_DIR}/gpg.conf @@ -120,7 +121,6 @@ initialize() { check_keyring() { if [[ ! -r ${PACMAN_KEYRING_DIR}/pubring.gpg || \ - ! -r ${PACMAN_KEYRING_DIR}/secring.gpg || \ ! -r ${PACMAN_KEYRING_DIR}/trustdb.gpg ]]; then error "$(gettext "You do not have sufficient permissions to read the %s keyring...")" "pacman" msg "$(gettext "Use '%s' to correct the keyring permissions.")" "pacman-key --init" @@ -128,7 +128,7 @@ check_keyring() { fi if (( (EXPORT || FINGER || LIST || VERIFY) && EUID != 0 )); then - if ! grep -w -q "lock-never" ${PACMAN_KEYRING_DIR}/gpg.conf &>/dev/null; then + if ! grep -q "^[[:space:]]*lock-never[[:space:]]*$" ${PACMAN_KEYRING_DIR}/gpg.conf &>/dev/null; then error "$(gettext "You do not have sufficient permissions to run this command...")" msg "$(gettext "Use '%s' to correct the keyring permissions.")" "pacman-key --init" exit 1 -- 1.7.6.1

2 3

[pacman-dev] [PATCH 2/5] pacman-key: import everything then revoke on --populate
by Allan McRae 26 Aug '11

26 Aug '11

The optimization of only importing keys that were not to be later revoked was a not smart enough. For example, if a key was in both a repos keyring and its revoke list, alternate runs of pacman-key --populate would add then remove the key from the pacman keyring. This problem is made worse when considering the possibility of multiple keyrings being imported. Instead, import all keys followed by the revoking of all keys. This may result in a key being added then revoked, but that is not much of an issue given that is a very fast operation. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- scripts/pacman-key.sh.in | 23 ++++++++++------------- 1 files changed, 10 insertions(+), 13 deletions(-) diff --git a/scripts/pacman-key.sh.in b/scripts/pacman-key.sh.in index 8580100..3e31abb 100644 --- a/scripts/pacman-key.sh.in +++ b/scripts/pacman-key.sh.in @@ -199,7 +199,16 @@ populate_keyring() { local key local key_id - # Read the key ids to an array. The conversion from whatever is inside the file + # Add keys from requested keyrings + for keyring in ${KEYRINGIDS[@]}; do + msg "$(gettext "Appending keys from %s.gpg...")" "$keyring" + local add_keys="$("${GPG_NOKEYRING[@]}" --keyring "${KEYRING_IMPORT_DIR}/${keyring}.gpg" --with-colons --list-keys | grep ^pub | cut -d: -f5)" + for key_id in ${add_keys}; do + "${GPG_NOKEYRING[@]}" --keyring "${KEYRING_IMPORT_DIR}/${keyring}.gpg" --export "${key_id}" | "${GPG_PACMAN[@]}" --import + done + done + + # Read the revoked key IDs to an array. The conversion from whatever is inside the file # to key ids is important, because key ids are the only guarantee of identification # for the keys. local -A removed_ids @@ -236,18 +245,6 @@ populate_keyring() { done fi - # Add keys from requested keyrings - for keyring in ${KEYRINGIDS[@]}; do - msg "$(gettext "Appending keys from %s.gpg...")" "$keyring" - local add_keys="$("${GPG_NOKEYRING[@]}" --keyring "${KEYRING_IMPORT_DIR}/${keyring}.gpg" --with-colons --list-keys | grep ^pub | cut -d: -f5)" - for key_id in ${add_keys}; do - # There is no point in adding a key that will be deleted right after - if [[ -z "${removed_ids[$key_id]}" ]]; then - "${GPG_NOKEYRING[@]}" --keyring "${KEYRING_IMPORT_DIR}/${keyring}.gpg" --export "${key_id}" | "${GPG_PACMAN[@]}" --import - fi - done - done - # Remove the keys not marked to keep if (( ${#removed_ids[@]} > 0 )); then msg "$(gettext "Removing revoked keys from keyring...")" -- 1.7.6

3 6

[pacman-dev] [PATCH] Make sync error message smarter on unfound targets
by Dan McGee 25 Aug '11

25 Aug '11

We had two issues here. One is a file with an absolute path passed to -S results in a cryptic error message due to the database name being '\0'. The second is not realizing you should be doing -U instead of -S. Fix both of these to transform this: $ sudo pacman -S /tmp/binutils-2.21.1-2-i686.pkg.tar.xz error: database not found: to this: $ sudo pacman -S /tmp/binutils-2.21.1-2-i686.pkg.tar.xz error: target not found: /tmp/binutils-2.21.1-2-i686.pkg.tar.xz warning: '/tmp/binutils-2.21.1-2-i686.pkg.tar.xz' is a file, did you mean -U/--upgrade instead of -S/--sync? Signed-off-by: Dan McGee <dan(a)archlinux.org> --- src/pacman/sync.c | 9 ++++++--- 1 files changed, 6 insertions(+), 3 deletions(-) diff --git a/src/pacman/sync.c b/src/pacman/sync.c index ea32a26..a7fe5be 100644 --- a/src/pacman/sync.c +++ b/src/pacman/sync.c @@ -672,8 +672,6 @@ static int process_targname(alpm_list_t *dblist, const char *targname) /* #FS#23342 - skip ignored packages when user says no */ if(alpm_errno(config->handle) == ALPM_ERR_PKG_IGNORED) { pm_printf(ALPM_LOG_WARNING, _("skipping target: %s\n"), targname); - /* TODO how to do this, we shouldn't be fucking with it from the frontend */ - /* pm_errno = 0; */ return 0; } @@ -693,7 +691,7 @@ static int process_target(const char *target) int ret = 0; alpm_list_t *dblist = NULL; - if(targname) { + if(targname && targname != targstring) { alpm_db_t *db = NULL; *targname = '\0'; @@ -716,6 +714,11 @@ static int process_target(const char *target) } cleanup: free(targstring); + if(ret && access(target, R_OK) == 0) { + pm_fprintf(stderr, ALPM_LOG_WARNING, + _("'%s' is a file, did you mean %s instead of %s?"), + target, "-U/--upgrade", "-S/--sync"); + } return ret; } -- 1.7.6

1 0

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v4.0.0rc1-98-ge1dce07
by dan＠archlinux.org 25 Aug '11

25 Aug '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 e1dce078b247142f7db80bf2d338e431df34691e (commit) via 6aa9fdce5ac59f4137567c17043fdd9e73b03518 (commit) via 0ee3ce70a88a42387d19f0d58901fda21b0161b7 (commit) via 73fcf17041cf00e0b6c3082bd354b8eaab412dab (commit) via 30cad47fb91dec0748551cb7d8607871b95785e5 (commit) via f7a3c4c8df95ad99e05dfd29c61d6877b13b09da (commit) via 2e7d002315066400ccb1e62dacb3500fa3aa6015 (commit) via f0357e415c10c10e691c1ed881b1c1ebe32e3084 (commit) via 5c48ca3239f56dfdae1d82ce03140bc310a6213e (commit) from 268d0bbcbed29b684af7478467bab4bdc2785636 (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 e1dce078b247142f7db80bf2d338e431df34691e Author: Dan McGee <dan(a)archlinux.org> Date: Wed Aug 24 23:19:06 2011 -0500 Remove argument from check_pgp_helper This one wasn't all that necessary as we only used it in one place in the function, which can be checked easily enough at the call site. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 6aa9fdce5ac59f4137567c17043fdd9e73b03518 Author: Dan McGee <dan(a)archlinux.org> Date: Thu Aug 25 00:07:03 2011 -0500 Remove trans is NULL check in QUESTION/EVENT/PROGRESS macros trans cannot (and better not) be NULL at any point when these are being called. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 0ee3ce70a88a42387d19f0d58901fda21b0161b7 Author: Dan McGee <dan(a)archlinux.org> Date: Wed Aug 24 22:13:26 2011 -0500 Remove short/long label distinction We only used short labels in one place, and the short label is always the first character of the long label anyway. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 73fcf17041cf00e0b6c3082bd354b8eaab412dab Author: Dan McGee <dan(a)archlinux.org> Date: Wed Aug 24 21:55:56 2011 -0500 Slight refresh of the download progress callback This cleans up some of the mess we have here. * switch to long units for the download size * omit the .0 decimal part from the download rate * omit the almost always zero HH: from estimated time if eta_h == 0 * Display --:-- if eta_h > 99; formatting was screwed up before The net result of this is we usually have 1 more character to use for filename display. Before: extra 500.9K 1242.4K/s 00:00:00 [######################] 100% community-testing 947.0B 28.2M/s 00:00:00 [######################] 100% multilib 26.5K 405.1K/s 00:00:00 [######################] 100% community 450.6K 1238.3K/s 00:00:00 [######################] 100% After: extra 500.9 KiB 1118K/s 00:00 [######################] 100% community-testing 947.0 B 23M/s 00:00 [######################] 100% multilib 26.5 KiB 255K/s 00:00 [######################] 100% community 450.6 KiB 1211K/s 00:00 [######################] 100% Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 30cad47fb91dec0748551cb7d8607871b95785e5 Author: Dan McGee <dan(a)archlinux.org> Date: Wed Aug 24 21:52:32 2011 -0500 Add a few more sizes to humanize_sizes() Because why the hell not? Exbibyte, zebibyte, and yobibyte are going in, even though nothing bigger than the 2^60 exbibyte can be represented using an off_t variable anyway. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit f7a3c4c8df95ad99e05dfd29c61d6877b13b09da Author: Dan McGee <dan(a)archlinux.org> Date: Wed Aug 24 19:58:14 2011 -0500 Finish large file download attack prevention This handles the no Content-Length header problem as stated in the comments of FS#23413. We add a quick check to the callback that will force an abort if the downloaded data exceeds the payload size, and then check for this error in the post-download cleanup code. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 2e7d002315066400ccb1e62dacb3500fa3aa6015 Author: Dan McGee <dan(a)archlinux.org> Date: Wed Aug 24 19:48:03 2011 -0500 Use off_t rather than double where possible Beautiful of libcurl to use floating point types for what are never fractional values. We can do better, and we usually want these values in their integer form anyway. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit f0357e415c10c10e691c1ed881b1c1ebe32e3084 Author: Dan McGee <dan(a)archlinux.org> Date: Thu Aug 25 16:04:09 2011 -0500 Add new 'lt' and 'zh_TW' translations from transifex Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 5c48ca3239f56dfdae1d82ce03140bc310a6213e Author: Dan McGee <dan(a)archlinux.org> Date: Thu Aug 25 15:53:25 2011 -0500 Update existing translations from Transifex Signed-off-by: Dan McGee <dan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: lib/libalpm/be_package.c | 3 +- lib/libalpm/be_sync.c | 3 +- lib/libalpm/dload.c | 39 +- lib/libalpm/dload.h | 4 +- lib/libalpm/po/LINGUAS | 2 + lib/libalpm/po/{sr(a)latin.po => lt.po} | 297 ++++++------ lib/libalpm/po/sv.po | 11 +- lib/libalpm/po/uk.po | 186 ++++---- lib/libalpm/po/{zh_CN.po => zh_TW.po} | 290 ++++++------ lib/libalpm/signing.c | 7 +- lib/libalpm/signing.h | 3 +- lib/libalpm/trans.h | 6 +- scripts/po/LINGUAS | 2 + scripts/po/ca.po | 10 +- scripts/po/cs.po | 10 +- scripts/po/da.po | 10 +- scripts/po/de.po | 91 +++- scripts/po/el.po | 10 +- scripts/po/en_GB.po | 10 +- scripts/po/es.po | 10 +- scripts/po/fi.po | 10 +- scripts/po/fr.po | 10 +- scripts/po/hu.po | 10 +- scripts/po/it.po | 10 +- scripts/po/kk.po | 10 +- scripts/po/{it.po => lt.po} | 728 ++++++++++++++------------- scripts/po/nb.po | 10 +- scripts/po/pl.po | 10 +- scripts/po/pt.po | 10 +- scripts/po/pt_BR.po | 10 +- scripts/po/ro.po | 10 +- scripts/po/ru.po | 10 +- scripts/po/sk.po | 10 +- scripts/po/sr.po | 10 +- scripts/po/sr(a)latin.po | 10 +- scripts/po/sv.po | 25 +- scripts/po/tr.po | 10 +- scripts/po/uk.po | 10 +- scripts/po/zh_CN.po | 10 +- scripts/po/{pacman-scripts.pot => zh_TW.po} | 14 +- src/pacman/callback.c | 32 +- src/pacman/package.c | 4 +- src/pacman/po/LINGUAS | 2 + src/pacman/po/ca.po | 36 +- src/pacman/po/cs.po | 36 +- src/pacman/po/da.po | 40 +- src/pacman/po/de.po | 43 +- src/pacman/po/el.po | 42 +- src/pacman/po/en_GB.po | 42 +- src/pacman/po/es.po | 36 +- src/pacman/po/fi.po | 36 +- src/pacman/po/fr.po | 42 +- src/pacman/po/hu.po | 36 +- src/pacman/po/it.po | 42 +- src/pacman/po/kk.po | 36 +- src/pacman/po/{pacman.pot => lt.po} | 547 +++++++++++--------- src/pacman/po/nb.po | 36 +- src/pacman/po/pl.po | 36 +- src/pacman/po/pt.po | 36 +- src/pacman/po/pt_BR.po | 36 +- src/pacman/po/ro.po | 42 +- src/pacman/po/ru.po | 36 +- src/pacman/po/sk.po | 42 +- src/pacman/po/sr.po | 36 +- src/pacman/po/sr(a)latin.po | 36 +- src/pacman/po/sv.po | 166 ++++--- src/pacman/po/tr.po | 36 +- src/pacman/po/uk.po | 351 +++++++------ src/pacman/po/zh_CN.po | 36 +- src/pacman/po/{zh_CN.po => zh_TW.po} | 530 ++++++++++---------- src/pacman/util.c | 22 +- src/pacman/util.h | 2 +- 72 files changed, 2466 insertions(+), 2006 deletions(-) copy lib/libalpm/po/{sr(a)latin.po => lt.po} (54%) copy lib/libalpm/po/{zh_CN.po => zh_TW.po} (57%) copy scripts/po/{it.po => lt.po} (50%) copy scripts/po/{pacman-scripts.pot => zh_TW.po} (98%) copy src/pacman/po/{pacman.pot => lt.po} (54%) copy src/pacman/po/{zh_CN.po => zh_TW.po} (58%) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH] Remove short/long label distinction
by Dan McGee 25 Aug '11

25 Aug '11

We only used short labels in one place, and the short label is always the first character of the long label anyway. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- src/pacman/callback.c | 8 ++++---- src/pacman/package.c | 4 ++-- src/pacman/util.c | 22 +++++++++------------- src/pacman/util.h | 2 +- 4 files changed, 16 insertions(+), 20 deletions(-) diff --git a/src/pacman/callback.c b/src/pacman/callback.c index 63f7b55..04b3a52 100644 --- a/src/pacman/callback.c +++ b/src/pacman/callback.c @@ -663,12 +663,12 @@ void cb_dl_progress(const char *filename, off_t file_xfered, off_t file_total) } - rate_human = humanize_size((off_t)rate, '\0', 0, &rate_label); - xfered_human = humanize_size(xfered, '\0', 1, &xfered_label); + rate_human = humanize_size((off_t)rate, '\0', &rate_label); + xfered_human = humanize_size(xfered, '\0', &xfered_label); printf(" %ls%-*s ", wcfname, padwid, ""); - printf("%6.1f %3s %4.f%s/s ", - xfered_human, xfered_label, rate_human, rate_label); + printf("%6.1f %3s %4.f%c/s ", + xfered_human, xfered_label, rate_human, rate_label[0]); if(eta_h == 0) { printf("%02u:%02u", eta_m, eta_s); } else if(eta_h < 100) { diff --git a/src/pacman/package.c b/src/pacman/package.c index 47b6623..a8c15eb 100644 --- a/src/pacman/package.c +++ b/src/pacman/package.c @@ -120,14 +120,14 @@ void dump_pkg_full(alpm_pkg_t *pkg, enum pkg_from from, int extra) deplist_display(_("Conflicts With :"), alpm_pkg_get_conflicts(pkg)); deplist_display(_("Replaces :"), alpm_pkg_get_replaces(pkg)); - size = humanize_size(alpm_pkg_get_size(pkg), 'K', 1, &label); + size = humanize_size(alpm_pkg_get_size(pkg), 'K', &label); if(from == PKG_FROM_SYNCDB) { printf(_("Download Size : %6.2f %s\n"), size, label); } else if(from == PKG_FROM_FILE) { printf(_("Compressed Size: %6.2f %s\n"), size, label); } - size = humanize_size(alpm_pkg_get_isize(pkg), 'K', 1, &label); + size = humanize_size(alpm_pkg_get_isize(pkg), 'K', &label); printf(_("Installed Size : %6.2f %s\n"), size, label); string_display(_("Packager :"), alpm_pkg_get_packager(pkg)); diff --git a/src/pacman/util.c b/src/pacman/util.c index 995b275..f2afbb3 100644 --- a/src/pacman/util.c +++ b/src/pacman/util.c @@ -776,7 +776,7 @@ static alpm_list_t *create_verbose_row(alpm_pkg_t *pkg, int install) ret = alpm_list_add(ret, str); /* and size */ - size = humanize_size(alpm_pkg_get_size(pkg), 'M', 1, &label); + size = humanize_size(alpm_pkg_get_size(pkg), 'M', &label); pm_asprintf(&str, "%.2f %s", size, label); ret = alpm_list_add(ret, str); @@ -839,19 +839,19 @@ void display_targets(const alpm_list_t *pkgs, int install) printf("\n"); if(install) { - size = humanize_size(dlsize, 'M', 1, &label); + size = humanize_size(dlsize, 'M', &label); printf(_("Total Download Size: %.2f %s\n"), size, label); if(!(config->flags & ALPM_TRANS_FLAG_DOWNLOADONLY)) { - size = humanize_size(isize, 'M', 1, &label); + size = humanize_size(isize, 'M', &label); printf(_("Total Installed Size: %.2f %s\n"), size, label); /* only show this net value if different from raw installed size */ if(rsize > 0) { - size = humanize_size(isize - rsize, 'M', 1, &label); + size = humanize_size(isize - rsize, 'M', &label); printf(_("Net Upgrade Size: %.2f %s\n"), size, label); } } } else { - size = humanize_size(isize, 'M', 1, &label); + size = humanize_size(isize, 'M', &label); printf(_("Total Removed Size: %.2f %s\n"), size, label); } @@ -914,21 +914,17 @@ static char *pkg_get_location(alpm_pkg_t *pkg) * * @return the size in the appropriate unit */ -double humanize_size(off_t bytes, const char target_unit, int long_labels, - const char **label) +double humanize_size(off_t bytes, const char target_unit, const char **label) { - static const char *shortlabels[] = {"B", "K", "M", "G", - "T", "P", "E", "Z", "Y"}; - static const char *longlabels[] = {"B", "KiB", "MiB", "GiB", + static const char *labels[] = {"B", "KiB", "MiB", "GiB", "TiB", "PiB", "EiB", "ZiB", "YiB"}; - static const int unitcount = sizeof(shortlabels) / sizeof(shortlabels[0]); + static const int unitcount = sizeof(labels) / sizeof(labels[0]); - const char **labels = long_labels ? longlabels : shortlabels; double val = (double)bytes; int index; for(index = 0; index < unitcount - 1; index++) { - if(target_unit != '\0' && shortlabels[index][0] == target_unit) { + if(target_unit != '\0' && labels[index][0] == target_unit) { break; } else if(target_unit == '\0' && val <= 2048.0 && val >= -2048.0) { break; diff --git a/src/pacman/util.h b/src/pacman/util.h index 2411e65..6669847 100644 --- a/src/pacman/util.h +++ b/src/pacman/util.h @@ -53,7 +53,7 @@ char *strtrim(char *str); char *strreplace(const char *str, const char *needle, const char *replace); alpm_list_t *strsplit(const char *str, const char splitchar); void string_display(const char *title, const char *string); -double humanize_size(off_t bytes, const char target_unit, int long_labels, const char **label); +double humanize_size(off_t bytes, const char target_unit, const char **label); int table_display(const char *title, const alpm_list_t *header, const alpm_list_t *rows); void list_display(const char *title, const alpm_list_t *list); void list_display_linebreak(const char *title, const alpm_list_t *list); -- 1.7.6

1 0

[pacman-dev] [PATCH 1/2] Add a few more sizes to humanize_sizes()
by Dan McGee 25 Aug '11

25 Aug '11

Because why the hell not? Exbibyte, zebibyte, and yobibyte are going in, even though nothing bigger than the 2^60 exbibyte can be represented using an off_t variable anyway. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- src/pacman/util.c | 6 ++++-- 1 files changed, 4 insertions(+), 2 deletions(-) diff --git a/src/pacman/util.c b/src/pacman/util.c index 585f402..995b275 100644 --- a/src/pacman/util.c +++ b/src/pacman/util.c @@ -917,8 +917,10 @@ static char *pkg_get_location(alpm_pkg_t *pkg) double humanize_size(off_t bytes, const char target_unit, int long_labels, const char **label) { - static const char *shortlabels[] = {"B", "K", "M", "G", "T", "P"}; - static const char *longlabels[] = {"B", "KiB", "MiB", "GiB", "TiB", "PiB"}; + static const char *shortlabels[] = {"B", "K", "M", "G", + "T", "P", "E", "Z", "Y"}; + static const char *longlabels[] = {"B", "KiB", "MiB", "GiB", + "TiB", "PiB", "EiB", "ZiB", "YiB"}; static const int unitcount = sizeof(shortlabels) / sizeof(shortlabels[0]); const char **labels = long_labels ? longlabels : shortlabels; -- 1.7.6

1 1

[pacman-dev] [PATCH 1/2] Use long rather than double where possible
by Dan McGee 25 Aug '11

25 Aug '11

Beautiful of libcurl to use floating point types for what are never fractional values. We can do better, and we usually want these values in their integer form anyway. Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/dload.c | 16 ++++++++-------- lib/libalpm/dload.h | 2 +- 2 files changed, 9 insertions(+), 9 deletions(-) diff --git a/lib/libalpm/dload.c b/lib/libalpm/dload.c index 8387ae6..bd5ccce 100644 --- a/lib/libalpm/dload.c +++ b/lib/libalpm/dload.c @@ -43,7 +43,7 @@ #include "handle.h" #ifdef HAVE_LIBCURL -static double prevprogress; /* last download amount */ +static long prevprogress; /* last download amount */ static const char *get_filename(const char *url) { @@ -76,7 +76,7 @@ static int curl_progress(void *file, double dltotal, double dlnow, double UNUSED ultotal, double UNUSED ulnow) { struct dload_payload *payload = (struct dload_payload *)file; - double current_size, total_size; + long current_size, total_size; /* SIGINT sent, abort by alerting curl */ if(dload_interrupted) { @@ -88,20 +88,20 @@ static int curl_progress(void *file, double dltotal, double dlnow, return 0; } - current_size = payload->initial_size + dlnow; - total_size = payload->initial_size + dltotal; + current_size = payload->initial_size + (long)dlnow; + total_size = payload->initial_size + (long)dltotal; - if(DOUBLE_EQ(dltotal, 0) || DOUBLE_EQ(prevprogress, total_size)) { + if(DOUBLE_EQ(dltotal, 0.0) || prevprogress == total_size) { return 0; } /* initialize the progress bar here to avoid displaying it when * a repo is up to date and nothing gets downloaded */ - if(DOUBLE_EQ(prevprogress, 0)) { + if(prevprogress == 0) { payload->handle->dlcb(payload->remote_name, 0, (long)dltotal); } - payload->handle->dlcb(payload->remote_name, (long)current_size, (long)total_size); + payload->handle->dlcb(payload->remote_name, current_size, total_size); prevprogress = current_size; @@ -216,7 +216,7 @@ static void curl_set_handle_opts(struct dload_payload *payload, payload->tempfile_openmode = "ab"; curl_easy_setopt(handle->curl, CURLOPT_RESUME_FROM, (long)st.st_size); _alpm_log(handle, ALPM_LOG_DEBUG, "tempfile found, attempting continuation\n"); - payload->initial_size = (double)st.st_size; + payload->initial_size = st.st_size; } } diff --git a/lib/libalpm/dload.h b/lib/libalpm/dload.h index 13034f7..f973f2d 100644 --- a/lib/libalpm/dload.h +++ b/lib/libalpm/dload.h @@ -33,7 +33,7 @@ struct dload_payload { char *destfile_name; char *content_disp_name; char *fileurl; - double initial_size; + long initial_size; long max_size; int force; int allow_resume; -- 1.7.6

1 1

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v4.0.0rc1-89-g268d0bb
by dan＠archlinux.org 25 Aug '11

25 Aug '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 268d0bbcbed29b684af7478467bab4bdc2785636 (commit) via 30d978a966c1d2a619acc49f204e4da1e0010c83 (commit) via cc03d6366a12cd1926e46265c63fba98f41faaae (commit) via a8a4395098b4ddced5204b386e476565e68508b5 (commit) via 6731d0a9407e00f755d237e889c2e3f5db9cd276 (commit) via dddd6a46a01167f4cfdae1133da34a6d245c9bde (commit) from 762cbf574b66ccbe6d4931e426b6d9e95b191bd5 (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 268d0bbcbed29b684af7478467bab4bdc2785636 Author: Diogo Sousa <diogogsousa(a)gmail.com> Date: Wed Aug 24 17:01:11 2011 +0100 Improved alpm_list_mmerge() performance (fixed coding style) Improved alpm_list_mmerge() performance by removing an extra pass to obtain the tail node. This was actually suggested by a TODO comment. Signed-off-by: Diogo Sousa <diogogsousa(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 30d978a966c1d2a619acc49f204e4da1e0010c83 Author: Dan McGee <dan(a)archlinux.org> Date: Tue Aug 23 01:49:29 2011 -0500 vercmp: ensure 2.0a and 2.0.a do not compare equal We had this interesting set of facts conundrum, according to vercmp return values: 2.0a < 2.0 2.0 < 2.0.a 2.0a == 2.0.a This introduces a code change that ensures '2.0a < 2.0.a' as would be expected by the first two comparisons. Unfortunately this stays us a bit further from upstream RPM code, but those are the breaks (in RPM, the versions involving 'a' do in fact compare the same, but they are both greater than the bare '2.0'). Signed-off-by: Dan McGee <dan(a)archlinux.org> commit cc03d6366a12cd1926e46265c63fba98f41faaae Author: Dan McGee <dan(a)archlinux.org> Date: Tue Aug 23 01:47:45 2011 -0500 Style-match rpmvercmp code with upstream Not sure how or why some of this differed, but it is easy enough to set it back to how it was so it is easier to diff. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit a8a4395098b4ddced5204b386e476565e68508b5 Author: Dan McGee <dan(a)archlinux.org> Date: Tue Aug 23 01:05:02 2011 -0500 Retrieve default SigLevel value from backend after initial setting This ensures we grab and use the library default once we have processed the global SigLevel setting, but before processing the repo-specific settings. This means the following two configs will now evaluate the same, as the backend currently defaults to 'Optional': Config 1: [options] # nothing here [repo] SigLevel = TrustAll Config 2: [options] SigLevel = Optional [repo] SigLevel = TrustAll Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 6731d0a9407e00f755d237e889c2e3f5db9cd276 Author: Dave Reisner <dreisner(a)archlinux.org> Date: Mon Aug 22 19:07:35 2011 -0400 sync: halt file discovery if repo has no servers This avoids error spam when no servers are configured for a repo and a sync operation is performed: Proceed with installation? [Y/n] y :: Retrieving packages from testing... warning: failed to retrieve some files from testing warning: failed to retrieve some files from testing warning: failed to retrieve some files from testing warning: failed to retrieve some files from testing warning: failed to retrieve some files from testing warning: failed to retrieve some files from testing warning: failed to retrieve some files from testing warning: failed to retrieve some files from testing warning: failed to retrieve some files from testing warning: failed to retrieve some files from testing Signed-off-by: Dave Reisner <dreisner(a)archlinux.org> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit dddd6a46a01167f4cfdae1133da34a6d245c9bde Author: Jakob Gruber <jakob.gruber(a)gmail.com> Date: Sat Aug 20 09:28:07 2011 +0200 Fix formatting in pacman-key manpage Signed-off-by: Jakob Gruber <jakob.gruber(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: doc/pacman-key.8.txt | 4 ++-- doc/pacman.8.txt | 2 +- doc/vercmp.8.txt | 2 +- lib/libalpm/alpm_list.c | 25 +++++++++++++++---------- lib/libalpm/sync.c | 8 ++++++++ lib/libalpm/version.c | 27 ++++++++++++++++----------- src/pacman/conf.c | 3 +++ test/util/vercmptest.sh | 6 ++++++ 8 files changed, 52 insertions(+), 25 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [PATCH] Improved alpm_list_mmerge() performance (fixed coding style)
by Diogo Sousa 24 Aug '11

24 Aug '11

Improved alpm_list_mmerge() performance by removing an extra pass to obtain the tail node. This was actually suggested by a TODO comment. I hope the coding style is ok this time :) Thanks Signed-off-by: Diogo Sousa <diogogsousa(a)gmail.com> --- lib/libalpm/alpm_list.c | 25 +++++++++++++++---------- 1 files changed, 15 insertions(+), 10 deletions(-) diff --git a/lib/libalpm/alpm_list.c b/lib/libalpm/alpm_list.c index 83b9824..0ab9356 100644 --- a/lib/libalpm/alpm_list.c +++ b/lib/libalpm/alpm_list.c @@ -206,12 +206,18 @@ alpm_list_t SYMEXPORT *alpm_list_join(alpm_list_t *first, alpm_list_t *second) */ alpm_list_t SYMEXPORT *alpm_list_mmerge(alpm_list_t *left, alpm_list_t *right, alpm_list_fn_cmp fn) { - alpm_list_t *newlist, *lp; + alpm_list_t *newlist, *lp, *tail_ptr, *left_tail_ptr, *right_tail_ptr; - if(left == NULL) + if(left == NULL) { return right; - if(right == NULL) + } + if(right == NULL) { return left; + } + + /* Save tail node pointers for future use */ + left_tail_ptr = left->prev; + right_tail_ptr = right->prev; if(fn(left->data, right->data) <= 0) { newlist = left; @@ -242,19 +248,18 @@ alpm_list_t SYMEXPORT *alpm_list_mmerge(alpm_list_t *left, alpm_list_t *right, a if(left != NULL) { lp->next = left; left->prev = lp; + tail_ptr = left_tail_ptr; } else if(right != NULL) { lp->next = right; right->prev = lp; + tail_ptr = right_tail_ptr; } - - /* Find our tail pointer - * TODO maintain this in the algorithm itself */ - lp = newlist; - while(lp && lp->next) { - lp = lp->next; + else { + tail_ptr = lp; } - newlist->prev = lp; + + newlist->prev = tail_ptr; return newlist; } -- 1.7.6

2 1