- Pacman-dev - lists.archlinux.org

[pacman-dev] dbs should be locked between _alpm_db_open and _alpm_db_close
by Nagy Gabor 27 Oct '07

27 Oct '07

Hi! The problem bellow doesn't appear with the current dummy front-end, but if you want to design a flexible alpm lib, you have to deal with it. The main problem, that the whole pkgcache concept _assumes_ that the db won't be modified by other processes. However, pacman locks db iff he modifies it, so he cannot trust his assumption. Let me explain. Imagine a nice pacman GUI which starts with alpm_initialize, opens the databases and let user do multiple operations. pacman_gui can run all the day, without unregistering the databases, alpm_release etc. etc. The main problem, that if pacman_gui (== libalpm) let an other pacman instance to modify the local db, he should always reload the pkgcache, because it may be outdated! So the following situation can happen: pacman_gui loaded the pkgcache to list the available local packages to the user, and now he is waiting. During this an other pacman instance removed lot of packages (he could lock anf unlock the db, since pacman_gui didn't lock anything yet) and exited. pacman_gui didn't detect anything from this. Now the user ask pacman_gui to install some packages: now he locks the dbs, but he trust its outdated pkgcache: for example its depchecks will use this outdated information... So _alpm_db_open should lock database, and _alpm_db_close should unlock it. The problem here, that now a non-root user cannot write /var/lib/pacman/local/db.lck, but we can solve this somehow. (Note, that I'm suggesting a lock-per-db method here.) Another (not-so-strict but slower) possible solution a .lastupdate-like method to detect if the database was modified or not after pkgcache load, and reload if needed. Bye, ngaba ---------------------------------------------------- SZTE Egyetemi Könyvtár - http://www.bibl.u-szeged.hu This mail sent through IMP: http://horde.org/imp/

2 2

[pacman-dev] testdb
by Nagy Gabor 27 Oct '07

27 Oct '07

Hi! First of all, many thanks to Xavier for this cool stuff. Some suggestions/feature requests: -you don't free() anything in testpkg.c; this is not a big problem, but rigorous c-programmers frees everything in even helloworld.c;-) -conflict checking also would be nice and easy to implement -a small speedup suggestion (however, speed is not important here): line 176: data = alpm_checkdeps(db, PM_TRANS_TYPE_ADD, alpm_db_getpkgcache(db)); here you could use a fake_empty_db as a first parameter: -O(n^2) operation -my biggest feature request: an option to fix the problems instead of just listing them (at least rewrite wrong requiredby fields, and a suggestion "pacman -S --asdeps [missingdeps]") I will share my results with you, this is a nice database-breaker-bug history: -testdb found many duplicated requiredby fields; I had to fix them with db-fix.sh + re-pacman + pacman -U (I have no net at home at the moment; see my biggest feature request ;-) -testdb found 4 other wrong requiredby fields (however, extra requiredby field doesn't cause an error now: see remove048.py): this is the category of the REMOVEUPGRADE transaction: requiredby004.py -and an other old bug (fixed): sysvinit needs awk, gawk provides awk; and sysvinit wasn't listed in requiredby field of gawk. Bye, ngaba ---------------------------------------------------- SZTE Egyetemi Könyvtár - http://www.bibl.u-szeged.hu This mail sent through IMP: http://horde.org/imp/

5 18

[pacman-dev] pacman doesn't realize about letter package versioning when doing upgrades
by Mateusz Jedrasik 27 Oct '07

27 Oct '07

Simply put: pacman -Syu <snip> warning: openssh: local (4.7p1-3) is newer than core (4.7p1-2) warning: openssl: local (0.9.8g-1) is newer than core (0.9.8f-1) <snip> :: openssh requires openssl>=0.9.8g :-) It simply does not realize that 0.9.8f-1 is newer than 0.9.8g-1. Cheers! //m. -- Mateusz Jędrasik <m.jedrasik(a)gmail.com> tel. +48(79)022-9393, +48(51)69-444-90 http://imachine.szklo.eu.org

3 5

[pacman-dev] small bug in libalpm/remove.c
by Nagy Gabor 27 Oct '07

27 Oct '07

Hi! Look at line 122-123: _alpm_log(PM_LOG_ERROR, _("could not find %s in database -- skipping\n"), miss->depend.name); Probably you want to say miss->target here. You can treat this mail as a patch, since it is very easy to fix it (now I'm working on removing pmdepend_t and I don't want to create a new branch;-) Bye, ngaba

4 9

[pacman-dev] [PATCH] Dependency error/log messages fix
by Nagy Gabor 26 Oct '07

26 Oct '07

From 4ee4c412b7ce0a688a81b1f856ef6462c8ee269d Mon Sep 17 00:00:00 2001 From: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> Date: Sat, 27 Oct 2007 01:14:35 +0200 Subject: [PATCH] Dependency error/log messages fix The old code used only the depend.name in messages, which might have not been informative. The new code uses the whole dependency string in %DEPENDS% format. --- lib/libalpm/deps.c | 18 +++++++++++------- lib/libalpm/sync.c | 4 +++- src/util/testdb.c | 4 +++- 3 files changed, 17 insertions(+), 9 deletions(-) diff --git a/lib/libalpm/deps.c b/lib/libalpm/deps.c index 7f4fb0b..4beaf70 100644 --- a/lib/libalpm/deps.c +++ b/lib/libalpm/deps.c @@ -306,7 +306,7 @@ alpm_list_t *_alpm_checkdeps(pmdb_t *db, pmtranstype_t op, if(alpm_depcmp(pkg, depend)) { _alpm_log(PM_LOG_DEBUG, "checkdeps: dependency '%s' has moved from '%s' to '%s'\n", - depend->name, alpm_pkg_get_name(oldpkg), alpm_pkg_get_name(pkg)); + k->data, alpm_pkg_get_name(oldpkg), alpm_pkg_get_name(pkg)); satisfied = 1; break; } @@ -322,7 +322,7 @@ alpm_list_t *_alpm_checkdeps(pmdb_t *db, pmtranstype_t op, /* we ignore packages that will be updated because we know * that the updated ones don't satisfy depend */ _alpm_log(PM_LOG_DEBUG, "checkdeps: dependency '%s' satisfied by installed package '%s'\n", - depend->name, alpm_pkg_get_name(pkg)); + k->data, alpm_pkg_get_name(pkg)); satisfied = 1; break; } @@ -380,7 +380,7 @@ alpm_list_t *_alpm_checkdeps(pmdb_t *db, pmtranstype_t op, /* else if still not found... */ if(!found) { _alpm_log(PM_LOG_DEBUG, "missing dependency '%s' for package '%s'\n", - depend->name, alpm_pkg_get_name(tp)); + j->data, alpm_pkg_get_name(tp)); miss = _alpm_depmiss_new(alpm_pkg_get_name(tp), PM_DEP_TYPE_DEPEND, depend->mod, depend->name, depend->version); if(!_alpm_depmiss_isin(miss, baddeps)) { @@ -426,7 +426,7 @@ alpm_list_t *_alpm_checkdeps(pmdb_t *db, pmtranstype_t op, pmpkg_t *pkg = l->data; if(alpm_depcmp(pkg, depend) && !_alpm_pkg_find(alpm_pkg_get_name(pkg), packages)) { _alpm_log(PM_LOG_DEBUG, "checkdeps: dependency '%s' satisfied by installed package '%s'\n", - depend->name, alpm_pkg_get_name(pkg)); + k->data, alpm_pkg_get_name(pkg)); satisfied = 1; break; } @@ -686,8 +686,10 @@ int _alpm_resolvedeps(pmdb_t *local, alpm_list_t *dbs_sync, pmpkg_t *syncpkg, for(j = *list; j && !found; j = j->next) { pmpkg_t *sp = j->data; if(alpm_depcmp(sp, missdep)) { + char *missdepstring = alpm_dep_get_string(missdep); _alpm_log(PM_LOG_DEBUG, "%s satisfies dependency %s -- skipping\n", - alpm_pkg_get_name(sp), missdep->name); + alpm_pkg_get_name(sp), missdepstring); + free(missdepstring); found = 1; } } @@ -713,8 +715,10 @@ int _alpm_resolvedeps(pmdb_t *local, alpm_list_t *dbs_sync, pmpkg_t *syncpkg, } if(!found) { - _alpm_log(PM_LOG_ERROR, _("cannot resolve dependencies for \"%s\" (\"%s\" is not in the package set)\n"), - miss->target, missdep->name); + char *missdepstring = alpm_dep_get_string(missdep); + _alpm_log(PM_LOG_ERROR, _("cannot resolve \"%s\" dependency of \"%s\"\n"), + missdepstring, miss->target); + free(missdepstring); if(data) { if((miss = malloc(sizeof(pmdepmissing_t))) == NULL) { _alpm_log(PM_LOG_ERROR, _("malloc failure: could not allocate %d bytes\n"), sizeof(pmdepmissing_t)); diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index 3e12ffb..f4ed6dd 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -674,8 +674,10 @@ int _alpm_sync_prepare(pmtrans_t *trans, pmdb_t *db_local, alpm_list_t *dbs_sync pmsyncpkg_t *sp = l->data; pmpkg_t *sppkg = sp->pkg; if(alpm_depcmp(sppkg, &(miss->depend))) { + char *missdepstring = alpm_dep_get_string(&(miss->depend)); _alpm_log(PM_LOG_DEBUG, "sync: dependency '%s' satisfied by package '%s'\n", - miss->depend.name, alpm_pkg_get_name(sppkg)); + missdepstring, alpm_pkg_get_name(sppkg)); + free(missdepstring); satisfied = 1; } } diff --git a/src/util/testdb.c b/src/util/testdb.c index 1331f2c..a06456a 100644 --- a/src/util/testdb.c +++ b/src/util/testdb.c @@ -178,8 +178,10 @@ int main(int argc, char **argv) for(i = data; i; i = alpm_list_next(i)) { pmdepmissing_t *miss = alpm_list_getdata(i); pmdepend_t *dep = alpm_miss_get_dep(miss); + char *depstring = alpm_dep_get_string(dep); printf("missing dependency for %s : %s\n", alpm_miss_get_target(miss), - alpm_dep_get_name(dep)); + depstring); + free(depstring); } /* check requiredby */ -- 1.5.3.4

2 1

[pacman-dev] alpm_list_remove is buggy
by Nagy Gabor 26 Oct '07

26 Oct '07

Hi! That function seems totally odd. /* TODO I modified this to remove ALL matching items. Do we need a remove_first? */ <- this is partly true: *data contains only the last found "i->data", and the function doesn't return with an alpm_list. MEMLEAK!! Bye, ngaba PS: I don't create a patch, because I dunno want do you want: remove the matching _list_ or remove the first matching _item_'s data. My vote: 2nd. (Because then I don't have to resubmit my -Ru patch ;-)

2 2

[pacman-dev] Pacman maint branch (was: [GIT] The official pacman repository branch, maint now at v3.0.0-54-g09253b4)
by Dan McGee 26 Oct '07

26 Oct '07

On 10/25/07, Dan McGee <dan(a)archlinux.org> wrote: > Hello, > > This is an automated email from the git hooks/update script, it was > generated because a ref change was pushed to the repository. > > Updating branch, maint, > as a new branch > to 09253b48d60589d42d9486dd361eaaf30e364e59 (commit) So, as you can maybe tell from the above commit, I decided to see if I could import our CVS history post-3.0.0 into GIT. Long story short, I figured it out and did just that. Down with CVS! (or to Aaron- we really don't need the pacman CVS repo anymore as all that history has been imported into GIT.) If we do need another 3.0.X release, we can do it from the maint branch. Once we make a 3.1 release, the maint branch will probably move to the 3.1.1 tag and be based off of there, so be aware of that. Until then, it will be stuck to the 3.0.X series. -Dan

3 2

[pacman-dev] [GIT] The official pacman repository branch, maint now at v3.0.0-54-g09253b4
by Dan McGee 26 Oct '07

26 Oct '07

Hello, This is an automated email from the git hooks/update script, it was generated because a ref change was pushed to the repository. Updating branch, maint, as a new branch to 09253b48d60589d42d9486dd361eaaf30e364e59 (commit) - Log ----------------------------------------------------------------- ----------------------------------------------------------------------- hooks/update --- Git Source Code Management System hooks/update refs/heads/maint \ 0000000000000000000000000000000000000000 \ 09253b48d60589d42d9486dd361eaaf30e364e59

1 0

[pacman-dev] [GIT] The official pacman repository branch, master now at v3.0.0-482-g0cff7c6
by Dan McGee 25 Oct '07

25 Oct '07

Hello, This is an automated email from the git hooks/update script, it was generated because a ref change was pushed to the repository. Updating branch, master, via 0cff7c6bdf5461c3286edc940aabb88d63345381 (commit) via 89ac8aa9c45486aa4f4b9599bb094f1d54ff1b66 (commit) via 581769b72d882fdc05bdbdc588db97187329e5b5 (commit) via aa545f6798e526dfd23779b488a86296d1f52d12 (commit) via 616668fb71db5045ae727586031ce32fd18e92cc (commit) via bd15c23e1690d0c85632d7950cd79d17c65a4e17 (commit) via 8b1fb61df26649e1dec626ef9120fc787a831fcc (commit) via 0c4f7d821c1cd4ce559476595e2108f75b3dde7c (commit) via 46eaa5c54200bfda402412e2f5b7df0eddc0230a (commit) from f5b77eb9894ea92a08fac586ddf03e9ba70c3de8 (commit) - Log ----------------------------------------------------------------- commit 0cff7c6bdf5461c3286edc940aabb88d63345381 Author: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> Date: Thu Oct 25 02:31:28 2007 +0200 Add alpm_dep_get_string method Public alpm_dep_get_string function is introduced, which converts a pmdepend_t structure to printable string in %DEPENDS% format. This function is now used in pacman to print dependency error messages. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 89ac8aa9c45486aa4f4b9599bb094f1d54ff1b66 Author: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> Date: Wed Oct 24 22:58:34 2007 +0200 _alpm_depmiss_isin fix The old code used memcmp, which is not good for comparing strings: "pkgname"'\0''\0' should be equal to "pkgname"'\0''a' for example. The new code uses strcmp. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 581769b72d882fdc05bdbdc588db97187329e5b5 Author: Dan McGee <dan(a)archlinux.org> Date: Wed Oct 24 00:37:50 2007 -0500 Add strndup implemention for environments that are missing it Signed-off-by: Dan McGee <dan(a)archlinux.org> commit aa545f6798e526dfd23779b488a86296d1f52d12 Author: Dan McGee <dan(a)archlinux.org> Date: Tue Oct 23 23:27:03 2007 -0500 One more added include to help compilation elsewhere Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 616668fb71db5045ae727586031ce32fd18e92cc Author: Dan McGee <dan(a)archlinux.org> Date: Tue Oct 23 19:24:39 2007 -0500 Replace UK mirror with a new updated one Signed-off-by: Dan McGee <dan(a)archlinux.org> commit bd15c23e1690d0c85632d7950cd79d17c65a4e17 Author: Dan McGee <dan(a)archlinux.org> Date: Tue Oct 23 00:35:42 2007 -0500 Print and typecast time_t correctly The recommended C99 way to print the value of a time_t is to cast it to uintmax_t or intmax_t. Do this to ensure compatability with all platforms. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 8b1fb61df26649e1dec626ef9120fc787a831fcc Author: Dan McGee <dan(a)archlinux.org> Date: Mon Oct 22 23:52:55 2007 -0500 Ensure all localization stuff is correctly guarded Anything dealing with libintl and localization should be correctly guarded inside an ENABLE_NLS block on both the pacman and libalpm sides. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 0c4f7d821c1cd4ce559476595e2108f75b3dde7c Author: Dan McGee <dan(a)archlinux.org> Date: Mon Oct 22 21:33:47 2007 -0700 Fix broken or missing includes Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 46eaa5c54200bfda402412e2f5b7df0eddc0230a Author: Dan McGee <dan(a)archlinux.org> Date: Mon Oct 22 01:43:34 2007 -0700 Fix invalid static scoping of strverscmp Signed-off-by: Dan McGee <dan(a)archlinux.org> ----------------------------------------------------------------------- Diffstat: etc/pacman.d/mirrorlist.in | 2 +- lib/libalpm/alpm.c | 3 +++ lib/libalpm/alpm.h | 1 + lib/libalpm/be_files.c | 7 +++---- lib/libalpm/deps.c | 41 +++++++++++++++++++++++++++++++++++++++-- lib/libalpm/server.c | 5 +++-- lib/libalpm/util.c | 2 +- lib/libalpm/util.h | 16 +++++++++------- src/pacman/add.c | 18 +++--------------- src/pacman/pacman.c | 5 ++++- src/pacman/remove.c | 4 +++- src/pacman/sync.c | 19 ++++--------------- src/pacman/util.c | 23 +++++++++++++++++++++++ src/pacman/util.h | 14 +++++++++----- src/util/testdb.c | 1 + 15 files changed, 107 insertions(+), 54 deletions(-) hooks/update --- Git Source Code Management System hooks/update refs/heads/master \ f5b77eb9894ea92a08fac586ddf03e9ba70c3de8 \ 0cff7c6bdf5461c3286edc940aabb88d63345381

1 0

[pacman-dev] [PATCH] alpm_dep_get_string
by Nagy Gabor 25 Oct '07

25 Oct '07

Hi! If you look into libalpm code, you will realize that usually only depend->name is used in log/error messages; which can be confusing in some context (sometimes, alpm assumes that the missing dependency is a canonical package name). See my note here, for example: http://www.archlinux.org/pipermail/pacman-dev/2007-October/009671.html So I started to work on this. Most cases dependency is "created" with alpm_splitdep from a string; in these cases we can use that string in error messages. However, in few cases dependency "comes from" pmdepmissing_t, so we have to convert pmdepend_t to string. As a first step, I coded this converter-function; which is going to be used in resolvedeps only (in libalpm). ---------------------------- From 074e80216543c631f51079b4c27efbf8dcefff16 Mon Sep 17 00:00:00 2001 From: Nagy Gabor <ngaba@@bibl..u-szeged..hu> Date: Thu, 25 Oct 2007 02:16:48 +0200 Subject: [PATCH] alpm_dep_get_string Public alpm_dep_get_string function is introduced, which converts a pmdepend_t structure to printable string in %DEPENDS% format. This function is now used in pacman to print dependency error messages. --- lib/libalpm/alpm.h | 1 + lib/libalpm/deps.c | 34 ++++++++++++++++++++++++++++++++++ src/pacman/add.c | 18 +++--------------- src/pacman/remove.c | 4 +++- src/pacman/sync.c | 18 +++--------------- 5 files changed, 44 insertions(+), 31 deletions(-) diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h index 880bbeb..2000db4 100644 --- a/lib/libalpm/alpm.h +++ b/lib/libalpm/alpm.h @@ -379,6 +379,7 @@ pmdepend_t *alpm_miss_get_dep(pmdepmissing_t *miss); pmdepmod_t alpm_dep_get_mod(pmdepend_t *dep); const char *alpm_dep_get_name(pmdepend_t *dep); const char *alpm_dep_get_version(pmdepend_t *dep); +char *alpm_dep_get_string(pmdepend_t *dep); /* * File conflicts diff --git a/lib/libalpm/deps.c b/lib/libalpm/deps.c index b7e49be..55d3ee7 100644 --- a/lib/libalpm/deps.c +++ b/lib/libalpm/deps.c @@ -835,4 +835,38 @@ const char SYMEXPORT *alpm_dep_get_version(pmdepend_t *dep) return dep->version; } +/* the return-string must be freed! */ +char SYMEXPORT *alpm_dep_get_string(pmdepend_t *dep) +{ + ALPM_LOG_FUNC; + + /* Sanity checks */ + ASSERT(handle != NULL, return(NULL)); + ASSERT(dep != NULL, return(NULL)); + + char *ptr; + char *depstring = malloc(sizeof(pmdepend_t)); + if(depstring == NULL) { + _alpm_log(PM_LOG_ERROR, _("malloc failure: could not allocate %d bytes\n"), sizeof(pmdepend_t)); + return NULL; + } + + strcpy(depstring, dep->name); + ptr = depstring + strlen(depstring); + switch(dep->mod) { + case PM_DEP_MOD_ANY: + break; + case PM_DEP_MOD_EQ: + sprintf(ptr, "=%s", dep->version); + break; + case PM_DEP_MOD_GE: + sprintf(ptr, ">=%s", dep->version); + break; + case PM_DEP_MOD_LE: + sprintf(ptr, "<=%s", dep->version); + break; + } + + return(depstring); +} /* vim: set ts=2 sw=2 noet: */ diff --git a/src/pacman/add.c b/src/pacman/add.c index 0b59a23..f883a6b 100644 --- a/src/pacman/add.c +++ b/src/pacman/add.c @@ -137,26 +137,14 @@ int pacman_add(alpm_list_t *targets) for(i = data; i; i = alpm_list_next(i)) { pmdepmissing_t *miss = alpm_list_getdata(i); pmdepend_t *dep = alpm_miss_get_dep(miss); + char *depstring = alpm_dep_get_string(dep); /* TODO indicate if the error was a virtual package or not: * :: %s: requires %s, provided by %s */ printf(_(":: %s: requires %s\n"), alpm_miss_get_target(miss), - alpm_dep_get_name(dep)); - switch(alpm_dep_get_mod(dep)) { - case PM_DEP_MOD_ANY: - break; - case PM_DEP_MOD_EQ: - printf("=%s", alpm_dep_get_version(dep)); - break; - case PM_DEP_MOD_GE: - printf(">=%s", alpm_dep_get_version(dep)); - break; - case PM_DEP_MOD_LE: - printf("<=%s", alpm_dep_get_version(dep)); - break; - } - printf("\n"); + depstring); + free(depstring); } break; case PM_ERR_CONFLICTING_DEPS: diff --git a/src/pacman/remove.c b/src/pacman/remove.c index 860bf49..dce479e 100644 --- a/src/pacman/remove.c +++ b/src/pacman/remove.c @@ -129,8 +129,10 @@ int pacman_remove(alpm_list_t *targets) for(i = data; i; i = alpm_list_next(i)) { pmdepmissing_t *miss = alpm_list_getdata(i); pmdepend_t *dep = alpm_miss_get_dep(miss); + char *depstring = alpm_dep_get_string(dep); printf(_(":: %s: requires %s\n"), alpm_miss_get_target(miss), - alpm_dep_get_name(dep)); + depstring); + free(depstring); } alpm_list_free(data); break; diff --git a/src/pacman/sync.c b/src/pacman/sync.c index bf6eed1..2da7f55 100644 --- a/src/pacman/sync.c +++ b/src/pacman/sync.c @@ -558,22 +558,10 @@ int sync_trans(alpm_list_t *targets, int sync_only) for(i = data; i; i = alpm_list_next(i)) { pmdepmissing_t *miss = alpm_list_getdata(i); pmdepend_t *dep = alpm_miss_get_dep(miss); + char *depstring = alpm_dep_get_string(dep); printf(_(":: %s: requires %s\n"), alpm_miss_get_target(miss), - alpm_dep_get_name(dep)); - switch(alpm_dep_get_mod(dep)) { - case PM_DEP_MOD_ANY: - break; - case PM_DEP_MOD_EQ: - printf("=%s", alpm_dep_get_version(dep)); - break; - case PM_DEP_MOD_GE: - printf(">=%s", alpm_dep_get_version(dep)); - break; - case PM_DEP_MOD_LE: - printf("<=%s", alpm_dep_get_version(dep)); - break; - } - printf("\n"); + depstring); + free(depstring); } break; case PM_ERR_CONFLICTING_DEPS: -- 1.5.3.4

3 2