- Pacman-dev - lists.archlinux.org

[pacman-dev] resolvedeps is negligent: resolvedeps.py
by Nagy Gabor 02 Dec '07

02 Dec '07

Hi! This is a not-to-commit pactest, I just refer to this old thread...: http://www.archlinux.org/pipermail/pacman-dev/2007-August/009072.html I don't know whether we should fix this or not... This is not a critical bug, just resolvedeps is a bit negligent, but checkdeps will report its negligence (however, pacman doesn't seem to be very informative...). By fixing this the code would be cleaner, more clever, but _notably_ slower, so I still don't know what to do... What do you think? ---resolvedeps.py--- self.description = "resolvedeps test" sp1 = pmpkg("pkg1") sp1.depends = [ "pkg2>=2.0-1" , "provision" ] self.addpkg2db("sync", sp1) sp2 = pmpkg("pkg2", "2.0-1") self.addpkg2db("sync", sp2) sp3 = pmpkg("provision") self.addpkg2db("sync", sp3) lp = pmpkg("pkg2", "1.0-1") lp.provides = [ "provision" ] self.addpkg2db("local", lp) self.args = "-S pkg1 pkg2" self.addrule("PACMAN_RETCODE=0") self.addrule("PKG_EXIST=pkg1") self.addrule("PKG_EXIST=pkg2") self.addrule("PKG_VERSION=pkg2|2.0-1") self.addrule("PKG_EXIST=provision") ---------------------------------------------------- SZTE Egyetemi Könyvtár - http://www.bibl.u-szeged.hu This mail sent through IMP: http://horde.org/imp/

2 15

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v3.0.0-632-g22c2043
by Dan McGee 01 Dec '07

01 Dec '07

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 22c2043160f30a9ed34fd2886edaf521029a9248 (commit) via 5f811dc0207ead061b89d3acc6466bb390fd242e (commit) via cff1e2a58ffbf878e03689b52e58d8a830b45aff (commit) via e95e346ac7e189d7facf34d1126b523ef3b87570 (commit) via 2ed6b482d2fce916466e44e37930f0b0c0d928bb (commit) via 5c21f0f15281fda6e0515e83c2e37989752bbfab (commit) from 72f40b3876263f7a8dcda1390026f43f599f8823 (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 22c2043160f30a9ed34fd2886edaf521029a9248 Author: Chantry Xavier <shiningxc(a)gmail.com> Date: Mon Nov 26 16:54:47 2007 +0100 Minor fixes in sync.c and deps.c. * memleak found by Nagy in checkdeps * an useless line found by Nagy in resolvedeps * data wasn't set to the missing dependencies in sync_prepare * use the MALLOC macro in resolvedeps Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 5f811dc0207ead061b89d3acc6466bb390fd242e Author: Aaron Griffin <aaronmgriffin(a)gmail.com> Date: Thu Nov 29 13:40:14 2007 -0600 Ensure 0644 permissions for install and changelog files Signed-off-by: Aaron Griffin <aaronmgriffin(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit cff1e2a58ffbf878e03689b52e58d8a830b45aff Author: Dan McGee <dan(a)archlinux.org> Date: Fri Nov 30 23:02:06 2007 -0600 Remove alpm_depcmp DEBUG output alpm_depcmp is simple enough that we shouldn't need constant debug logging, and the logging we do have in there negatively impacts our speed regardless of whether we actually are seeing it or not. If it ever needs debugging, it would be trivial to reimplement it by just temporarily reverting this commit. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit e95e346ac7e189d7facf34d1126b523ef3b87570 Author: Dan McGee <dan(a)archlinux.org> Date: Fri Nov 30 22:57:47 2007 -0600 Remove some null checks in _alpm_checkdeps Hopefully these are unnecessary, as they were probably added originally to cover up a problem instead of finding the root cause. I'd rather fail hard in these cases anyway. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 2ed6b482d2fce916466e44e37930f0b0c0d928bb Author: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> Date: Thu Nov 29 13:54:44 2007 +0100 alpm_checkdeps clean-up No bugfixes, just makes the code human-readable Signed-off-by: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 5c21f0f15281fda6e0515e83c2e37989752bbfab Author: Dan McGee <dan(a)archlinux.org> Date: Thu Nov 29 16:16:13 2007 -0600 Move requiredby computation before any display starts This should reduce the chances of people thinking pacman hung during the middle of something. Signed-off-by: Dan McGee <dan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: lib/libalpm/add.c | 2 + lib/libalpm/deps.c | 172 +++++++++++--------------------------------------- lib/libalpm/sync.c | 2 +- src/pacman/package.c | 8 ++- 4 files changed, 47 insertions(+), 137 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] November 76% OFF
by VIAGRA � Official Site 01 Dec '07

01 Dec '07

1 0

[pacman-dev] more universal testdb -- bugs in our sync repos
by Nagy Gabor 30 Nov '07

30 Nov '07

Hi! Imho that would be helpful if testdb were more flexible. This may not be the proper ML (but I'm sure that reporting bugs is better than being silent), but I made some tests on sync repos, and I found the following inconsistencies in our sync dbs, mostly in community (I have only core, extra, community "enabled"): Unsatisfied dependencies: ------------------------- firefox-i18n -> firefox-i18n=2.0.0.9 dfm -> xfree86 fvwm-themes-devel -> fvwm-devel icomplete -> vim-devel perl-xml-feed -> perl-datetime-format-mail -> perl-datetime>=0.1705 perl-xml-rss -> perl-datetime-format-mail -> perl-datetime>=0.1705 perl-datetime-format-strptime -> perl-datetime>=0.1402 portaudio_cpp -> portaudio Strange dependency (cycle): --------------------------- archboot->archboot Some of them (like firefox-i18n) is not actual yet, but these things could be tested by a clever app (a modified testdb for internal use) to avoid these bugs. Bye, ngaba ---------------------------------------------------- SZTE Egyetemi Könyvtár - http://www.bibl.u-szeged.hu This mail sent through IMP: http://horde.org/imp/

2 2

[pacman-dev] [PATCH] New sync1005.py pactest
by Nagy Gabor 30 Nov '07

30 Nov '07

Hi! Patch attached. Note: I found an issue in resolvedeps, it simply pulls both pkg1s with this pactest (debug log), but the pactest doesn't fail [I don't know why]: ---sync1006.py--- self.description = "Conflicting package names in sync repos (2)" sp1 = pmpkg("pkg1") sp1.provides = [ "provision1" ] self.addpkg2db("sync1", sp1) sp2 = pmpkg("pkg1") sp2.provides = ["provision2"] self.addpkg2db("sync2", sp2) sp3 = pmpkg("pkg2") sp3.depends = ["provision1", "provision2"] self.addpkg2db("sync1", sp3) self.args = "-S pkg2" self.addrule("PACMAN_RETCODE=1") self.addrule("!PKG_EXIST=pkg1") self.addrule("!PKG_EXIST=pkg2") ---------------------------------------------------- SZTE Egyetemi Könyvtár - http://www.bibl.u-szeged.hu This mail sent through IMP: http://horde.org/imp/

2 2

[pacman-dev] Fast binary database
by Sivert Berg 29 Nov '07

29 Nov '07

Hello, I have been playing with the idea of speeding up pacman for a little while. This is my first try. It's a binary database that uses a ternary search tree to speedup file lookup and package lookup. The result can be found at http://owwsnap.com/sarcina-0.0.1.tar.gz. To compile it you need to run the normal ./configure && make, but you would probably want to edit src/sarcina.c and change the hardcoded paths. If you're running a new version of pacman (3.1 or something) you probably need to change libsarc/sarc.c:505: db_local = alpm_db_register("local"); into alpm_option_set_dbpath("/var/lib/pacman/"); db_local = alpm_db_register_local(); I see you have discussed the database question several times. I hope this code could give you some answers (is a binary database really faster etc). To run this program you first need to build the database using sarcina -u. Then you can: - find who owns a file (sarcina -o file) - list files that are outdated (sarcina -s) - "install" a package (sarcina -i package). This is pretty much just a proof-of-concept, so don't expect it to do much, but I think it might be a start. Sivert

4 3

[pacman-dev] Bug with rxvt-unicode version check
by bardo 28 Nov '07

28 Nov '07

Hi all. I'm experiencing a strange problem: rxvt-unicode should be upgraded from 8.4-1 to 8.5a-1, but when it comes the time of checking if the package has been cached, 8.4-1 is marked as a match, and gets installed instead of the new version. I can reproduce this whenever I want, since pacman tries to upgrade rxvt-unicode with every -Su. The relevant part of pacman -Syu --debug is attached, I can upload the whole log if needed (~750KB). Corrado

5 10

[pacman-dev] [PATCH] Optimize string comparison on 1 character compare
by Dan McGee 27 Nov '07

27 Nov '07

No need to do strcmp when we can check for direct equality, but is it worth the optimization? Signed-off-by: Dan McGee <dan(a)archlinux.org> --- lib/libalpm/be_files.c | 4 ++-- lib/libalpm/util.c | 2 +- src/pacman/sync.c | 4 ++-- src/pacman/util.c | 2 +- src/util/testdb.c | 2 +- 5 files changed, 7 insertions(+), 7 deletions(-) diff --git a/lib/libalpm/be_files.c b/lib/libalpm/be_files.c index 078136e..74932d2 100644 --- a/lib/libalpm/be_files.c +++ b/lib/libalpm/be_files.c @@ -170,7 +170,7 @@ pmpkg_t *_alpm_db_scan(pmdb_t *db, const char *target) /* search for a specific package (by name only) */ rewinddir(db->handle); while(!found && (ent = readdir(db->handle)) != NULL) { - if(!strcmp(ent->d_name, ".") || !strcmp(ent->d_name, "..")) { + if(*(ent->d_name) == '.' || !strcmp(ent->d_name, "..")) { continue; } /* stat the entry, make sure it's a directory */ @@ -201,7 +201,7 @@ pmpkg_t *_alpm_db_scan(pmdb_t *db, const char *target) if(ent == NULL) { return(NULL); } - if(!strcmp(ent->d_name, ".") || !strcmp(ent->d_name, "..")) { + if(*(ent->d_name) == '.' || !strcmp(ent->d_name, "..")) { isdir = 0; continue; } diff --git a/lib/libalpm/util.c b/lib/libalpm/util.c index 2a42948..3ce1f04 100644 --- a/lib/libalpm/util.c +++ b/lib/libalpm/util.c @@ -463,7 +463,7 @@ int _alpm_rmrf(const char *path) for(dp = readdir(dirp); dp != NULL; dp = readdir(dirp)) { if(dp->d_ino) { sprintf(name, "%s/%s", path, dp->d_name); - if(strcmp(dp->d_name, "..") && strcmp(dp->d_name, ".")) { + if(strcmp(dp->d_name, "..") && *(dp->d_name) != '.') { errflag += _alpm_rmrf(name); } } diff --git a/src/pacman/sync.c b/src/pacman/sync.c index 58e616e..e9c16f5 100644 --- a/src/pacman/sync.c +++ b/src/pacman/sync.c @@ -61,7 +61,7 @@ static int sync_cleandb(const char *dbpath) { int found = 0; char *dname = ent->d_name; - if(!strcmp(dname, ".") || !strcmp(dname, "..")) { + if(*dname == '.' || !strcmp(dname, "..")) { continue; } /* skip the local and sync directories */ @@ -145,7 +145,7 @@ static int sync_cleancache(int level) char path[PATH_MAX]; pmpkg_t *localpkg = NULL, *dbpkg = NULL; - if(!strcmp(ent->d_name, ".") || !strcmp(ent->d_name, "..")) { + if(*(ent->d_name) == '.' || !strcmp(ent->d_name, "..")) { continue; } /* build the full filepath */ diff --git a/src/pacman/util.c b/src/pacman/util.c index a925be3..5025b3f 100644 --- a/src/pacman/util.c +++ b/src/pacman/util.c @@ -154,7 +154,7 @@ int rmrf(const char *path) if(dp->d_ino) { char name[PATH_MAX]; sprintf(name, "%s/%s", path, dp->d_name); - if(strcmp(dp->d_name, "..") && strcmp(dp->d_name, ".")) { + if(strcmp(dp->d_name, "..") && *(dp->d_name) != '.') { errflag += rmrf(name); } } diff --git a/src/util/testdb.c b/src/util/testdb.c index 31b4ff0..c3f86d8 100644 --- a/src/util/testdb.c +++ b/src/util/testdb.c @@ -75,7 +75,7 @@ static int db_test(char *dbpath) } while ((ent = readdir(dir)) != NULL) { - if(!strcmp(ent->d_name, ".") || !strcmp(ent->d_name, "..")) { + if(*(ent->d_name) == '.' || !strcmp(ent->d_name, "..")) { continue; } /* check for desc, depends, and files */ -- 1.5.3.6

3 3

[pacman-dev] [PATCH] Unify dump_pkg_full in pacman [-Si, -Qip, -Qi and -Qii]
by Nagy Gabor 27 Nov '07

27 Nov '07

Patch attached. Notes: 1. 'Compressed size' is ugly [too long, no space before :], improvements welcome 2. The following commit is misleading in case of sync packages (isize == 0 means isize is not filled in, not isize == csize): http://projects.archlinux.org/git/? p=pacman.git;a=commit;h=5e12d3dec99e7a506683cf625fa4344f57df0b77 Bye ---------------------------------------------------- SZTE Egyetemi Könyvtár - http://www.bibl.u-szeged.hu This mail sent through IMP: http://horde.org/imp/

2 2

[pacman-dev] [PATCH] Fix several memleaks, mostly related to errors handling.
by Chantry Xavier 26 Nov '07

26 Nov '07

* The frontend calls alpm_trans_prepare(&data), and in case of errors, receive the missing dependencies / conflicts / etc in the data pointer. It apparently needs to free this structure totally with : alpm_list_free_inner(data, free) alpm_list_free(data) So I added alpm_list_free_inner(data, free) in pacman/{sync.c,remove.c,add,c} * in _alpm_sync_prepare, the deps and asked lists were not freed in case of errors (unresolvable conflicts). Besides the code for handling this case was duplicated. * in _alpm_remove_commit, free was used instead of alpm_list_free for newfiles. * newline fix in pacman/sync.c Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> --- lib/libalpm/remove.c | 2 +- lib/libalpm/sync.c | 42 ++++++++++++++++++------------------------ src/pacman/add.c | 1 + src/pacman/remove.c | 1 + src/pacman/sync.c | 3 ++- 5 files changed, 23 insertions(+), 26 deletions(-) diff --git a/lib/libalpm/remove.c b/lib/libalpm/remove.c index 592cf39..e1f19ec 100644 --- a/lib/libalpm/remove.c +++ b/lib/libalpm/remove.c @@ -315,7 +315,7 @@ int _alpm_remove_commit(pmtrans_t *trans, pmdb_t *db) (pkg_count - alpm_list_count(targ) + 1)); position++; } - free(newfiles); + alpm_list_free(newfiles); } /* set progress to 100% after we finish unlinking files */ diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index f03a78b..b49bbfb 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -496,12 +496,13 @@ int _alpm_sync_prepare(pmtrans_t *trans, pmdb_t *db_local, alpm_list_t *dbs_sync if(deps) { int errorout = 0; alpm_list_t *asked = NULL; + pmconflict_t *conflict = NULL; for(i = deps; i && !errorout; i = i->next) { - pmconflict_t *conflict = i->data; pmsyncpkg_t *sync; pmpkg_t *found = NULL; + conflict = i->data; _alpm_log(PM_LOG_DEBUG, "package '%s' conflicts with '%s'\n", conflict->package1, conflict->package2); /* check if the conflicting package is about to be removed/replaced. @@ -614,42 +615,35 @@ int _alpm_sync_prepare(pmtrans_t *trans, pmdb_t *db_local, alpm_list_t *dbs_sync /* abort */ _alpm_log(PM_LOG_ERROR, _("unresolvable package conflicts detected\n")); errorout = 1; - if(data) { - if((conflict = malloc(sizeof(pmconflict_t))) == NULL) { - _alpm_log(PM_LOG_ERROR, _("malloc failure: could not allocate %zd bytes\n"), sizeof(pmconflict_t)); - FREELIST(*data); - pm_errno = PM_ERR_MEMORY; - ret = -1; - goto cleanup; - } - *conflict = *(pmconflict_t *)i->data; - *data = alpm_list_add(*data, conflict); - } } } } else { _alpm_log(PM_LOG_ERROR, _("unresolvable package conflicts detected\n")); errorout = 1; - if(data) { - if((conflict = malloc(sizeof(pmconflict_t))) == NULL) { - _alpm_log(PM_LOG_ERROR, _("malloc failure: could not allocate %zd bytes\n"), sizeof(pmconflict_t)); - FREELIST(*data); - pm_errno = PM_ERR_MEMORY; - ret = -1; - goto cleanup; - } - *conflict = *(pmconflict_t *)i->data; - *data = alpm_list_add(*data, conflict); - } } } if(errorout) { + /* The last conflict was unresolvable, so we duplicate it and add it to *data */ pm_errno = PM_ERR_CONFLICTING_DEPS; + if(data) { + pmconflict_t *lastconflict = conflict; + if((conflict = malloc(sizeof(pmconflict_t))) == NULL) { + _alpm_log(PM_LOG_ERROR, _("malloc failure: could not allocate %zd bytes\n"), + sizeof(pmconflict_t)); + FREELIST(*data); + pm_errno = PM_ERR_MEMORY; + } else { + *conflict = *lastconflict; + *data = alpm_list_add(*data, conflict); + } + } + FREELIST(asked); + FREELIST(deps); ret = -1; goto cleanup; } - FREELIST(deps); FREELIST(asked); + FREELIST(deps); } EVENT(trans, PM_TRANS_EVT_INTERCONFLICTS_DONE, NULL, NULL); } diff --git a/src/pacman/add.c b/src/pacman/add.c index 7d18749..e04707f 100644 --- a/src/pacman/add.c +++ b/src/pacman/add.c @@ -175,6 +175,7 @@ int pacman_add(alpm_list_t *targets) break; } add_cleanup(); + alpm_list_free_inner(data, free); alpm_list_free(data); return(1); } diff --git a/src/pacman/remove.c b/src/pacman/remove.c index e1e4f04..1028d9e 100644 --- a/src/pacman/remove.c +++ b/src/pacman/remove.c @@ -133,6 +133,7 @@ int pacman_remove(alpm_list_t *targets) depstring); free(depstring); } + alpm_list_free_inner(data, free); alpm_list_free(data); break; default: diff --git a/src/pacman/sync.c b/src/pacman/sync.c index 41d18a9..889c682 100644 --- a/src/pacman/sync.c +++ b/src/pacman/sync.c @@ -621,7 +621,7 @@ int sync_trans(alpm_list_t *targets, int sync_only) case PM_ERR_CONFLICTING_DEPS: for(i = data; i; i = alpm_list_next(i)) { pmconflict_t *conflict = alpm_list_getdata(i); - printf(_(":: %s: conflicts with %s"), + printf(_(":: %s: conflicts with %s\n"), alpm_conflict_get_package1(conflict), alpm_conflict_get_package2(conflict)); } break; @@ -706,6 +706,7 @@ int sync_trans(alpm_list_t *targets, int sync_only) /* Step 4: release transaction resources */ cleanup: if(data) { + alpm_list_free_inner(data, free); alpm_list_free(data); } if(alpm_trans_release() == -1) { -- 1.5.3.6

5 9