- Pacman-dev - lists.archlinux.org

Re: [pacman-dev] [PATCH] Only query pacman upgrade when performing actual upgrade
by Xavier 09 Dec '07

09 Dec '07

On Sun, Dec 09, 2007 at 07:54:37PM +1000, Allan McRae wrote: > Attached is a patch that fixes FS#7147. The query about upgrading > pacman separately is not needed when the -p and -w flags are used. It > is really only a one line patch... > > I wasn't sure whether to use "alpm_trans_get_flags()" or "config->flags" > as this seems to vary across files. > > Allan > Ah yes, I see. I think you don't really have the choice before the trans is initialized (before alpm_trans_init call). You have to use config->flags. But in the part you edited, the trans is already initialized, so I guess we can use both there, without any differences. Since alpm_trans_get_flags is already used in that part, it's probably fine to use that. > From df921cc477d3c999bb8889b4c88cdb3867691c0c Mon Sep 17 00:00:00 2001 > From: Allan McRae <mcrae_allan(a)hotmail.com> > Date: Sun, 9 Dec 2007 19:44:30 +1000 > Subject: [PATCH] Only query pacman upgrade when performing actual upgrade > > Fixes FS#7147. Do not ask about upgrading pacman when -w and -p > flags are used. > > Signed-off-by: Allan McRae <mcrae_allan(a)hotmail.com> > --- > src/pacman/sync.c | 74 > +++++++++++++++++++++++++++------------------------- > 1 files changed, 38 insertions(+), 36 deletions(-) > The lines are broken again :( I believe the policy is to put the patch inline, but personally I would prefer a working attached patch, rather than a broken inline one. Besides, as vmiklos pointed out earlier, mutt shows them inline, even when they are attached. Why don't other mail clients do this :p

3 2

[pacman-dev] [PATCH] Updated Italian Translation fix
by Giovanni Scafora 09 Dec '07

09 Dec '07

Hi, here's an updated translation to fix 5 fuzzy translations. Thanks. -- Giovanni Scafora Arch Linux Trusted User (voidnull) http://www.archlinux.org linuxmania(a)gmail.com

1 0

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v3.0.6-664-gb6acb30
by Dan McGee 09 Dec '07

09 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 b6acb30aae5504736fd23f97402389baabbddb1b (commit) via 4437033d2a822d41820c67ab98d95db97daa7fd8 (commit) via 2ef78355fc65a38bd64e430c8f19e2ae9ab7648f (commit) via 97fe3d3f5adecfbf9d6bf201161d743361107e32 (commit) via f0664fbd93f0002bc568b24de2206e31187fda85 (commit) via c49b2a00d628c2ecb86b3908c1901ab3edee0b1a (commit) via 43eacf2852a15828a4becff74efad04f85305a92 (commit) via 145c8eb52d7c782748bdc1e6a087c932243903bc (commit) from 9615d82343148301884bfc79d87e2f408aad64bd (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 b6acb30aae5504736fd23f97402389baabbddb1b Author: Chantry Xavier <shiningxc(a)gmail.com> Date: Sat Dec 8 09:58:49 2007 +0100 src/pacman/: use the FREELIST macro when possible. Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 4437033d2a822d41820c67ab98d95db97daa7fd8 Author: Chantry Xavier <shiningxc(a)gmail.com> Date: Sat Dec 8 00:01:32 2007 +0100 testdb.c : add vim modeline according to HACKING, and reindent the file. Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 2ef78355fc65a38bd64e430c8f19e2ae9ab7648f Author: Chantry Xavier <shiningxc(a)gmail.com> Date: Sun Dec 9 02:04:28 2007 +0100 _alpm_unpack: return 1 when the file is not found. If _alpm_unpack has a specific file to extract (not NULL), but doesn't find it, it'll now return 1, for indicating the failure. Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 97fe3d3f5adecfbf9d6bf201161d743361107e32 Author: Chantry Xavier <shiningxc(a)gmail.com> Date: Fri Dec 7 09:45:20 2007 -0600 libalpm/server.c : fix segfault when downloading failed. The alpm_list_free(complete) needs to be done OUTSIDE the loop walking through the server list. Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit f0664fbd93f0002bc568b24de2206e31187fda85 Author: Dan McGee <dan(a)archlinux.org> Date: Wed Dec 5 20:34:38 2007 -0600 Update pacman manpage with description of --clean option Signed-off-by: Dan McGee <dan(a)archlinux.org> commit c49b2a00d628c2ecb86b3908c1901ab3edee0b1a Author: Chantry Xavier <shiningxc(a)gmail.com> Date: Wed Dec 5 23:47:54 2007 +0100 pacman/sync.c : improve the sync db cleanup feature. This feature (introduced by b118ce55bd01c7ebd42b5b6d4a0f34aa925701d8 as a part of -Sc) could actually be helpful in the 3.0 -> 3.1 transition, because all sync dbs will be left in /var/lib/pacman/, while the updated ones will go to /var/lib/pacman/sync/. So it'll now clean everything in /var/lib/pacman/, and only the unused databases in /var/lib/pacman/sync/ (with the exception of local/ and sync/ in both cases). Note: This feature is undocumented. I wonder if moving it to another option, something like -S --dbclean, wouldn't help for documenting it. Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 43eacf2852a15828a4becff74efad04f85305a92 Author: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> Date: Wed Dec 5 12:02:41 2007 +0100 Remove duplicated PM_TRANS_CONV_INSTALL_IGNOREPKG message Now alpm_checkdeps first search for literals, then search for no-literal satisfiers Signed-off-by: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> [Dan: fix spelling of INTALL, fix line wrapping] Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 145c8eb52d7c782748bdc1e6a087c932243903bc Author: Nathan Jones <nathanj(a)insightbb.com> Date: Wed Dec 5 10:45:23 2007 -0500 Update zsh completion. Update zsh completion file to version 0.5-2 from AUR and add the new options in pacman 3.1. Signed-off-by: Nathan Jones <nathanj(a)insightbb.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: contrib/zsh_completion | 85 ++++++++++++++++----- doc/pacman.8.txt | 19 +++-- lib/libalpm/deps.c | 4 +- lib/libalpm/server.c | 2 +- lib/libalpm/util.c | 3 +- src/pacman/add.c | 3 +- src/pacman/callback.c | 4 +- src/pacman/remove.c | 3 +- src/pacman/sync.c | 27 ++++--- src/util/testdb.c | 198 ++++++++++++++++++++++++------------------------ 10 files changed, 200 insertions(+), 148 deletions(-) hooks/post-receive -- The official pacman repository

1 0

Re: [pacman-dev] [PATCH] Improve changelog handling
by Xavier 09 Dec '07

09 Dec '07

On Sat, Dec 08, 2007 at 02:47:23AM +1000, Allan McRae wrote: > From 47affca982ec2d449bd6d96846172052dbdd3da6 Mon Sep 17 00:00:00 2001 > From: Allan McRae <mcrae_allan(a)hotmail.com> > Date: Sat, 8 Dec 2007 02:35:00 +1000 > Subject: [PATCH] Improve changelog handling > > Allows dumping of changelog for both installed packages and from package > files > (See FS#7321). Moves querying of changelog file in alpm backend. > > Signed-off-by: Allan McRae <mcrae_allan(a)hotmail.com> > --- > lib/libalpm/alpm.h | 1 + > lib/libalpm/package.c | 71 > +++++++++++++++++++++++++++++++++++++++++++++++++ > src/pacman/package.c | 33 ++++++++++------------ > src/pacman/package.h | 2 +- > src/pacman/query.c | 9 +----- > 5 files changed, 89 insertions(+), 27 deletions(-) > > diff --git a/lib/libalpm/alpm.h b/lib/libalpm/alpm.h > index 1e18ad9..7f6c9eb 100644 > --- a/lib/libalpm/alpm.h > +++ b/lib/libalpm/alpm.h > @@ -216,6 +216,7 @@ alpm_list_t *alpm_pkg_get_deltas(pmpkg_t *pkg); > alpm_list_t *alpm_pkg_get_replaces(pmpkg_t *pkg); > alpm_list_t *alpm_pkg_get_files(pmpkg_t *pkg); > alpm_list_t *alpm_pkg_get_backup(pmpkg_t *pkg); > +alpm_list_t *alpm_pkg_get_changelog(pmpkg_t *pkg); > unsigned short alpm_pkg_has_scriptlet(pmpkg_t *pkg); > > unsigned long alpm_pkg_download_size(pmpkg_t *newpkg, pmdb_t *db_local); > diff --git a/lib/libalpm/package.c b/lib/libalpm/package.c > index 172456d..bbdb95d 100644 > --- a/lib/libalpm/package.c > +++ b/lib/libalpm/package.c > @@ -491,6 +491,77 @@ alpm_list_t SYMEXPORT *alpm_pkg_get_backup(pmpkg_t > *pkg) > return pkg->backup; > } > > +alpm_list_t SYMEXPORT *alpm_pkg_get_changelog(pmpkg_t *pkg) > +{ > + ALPM_LOG_FUNC; > + > + /* Sanity checks */ > + ASSERT(handle != NULL, return(NULL)); > + ASSERT(pkg != NULL, return(NULL)); > + > + int ret = ARCHIVE_OK; > + alpm_list_t *cl = NULL; > + char clfile[PATH_MAX]; > + char *tmpfile = NULL; > + struct archive *archive; > + struct archive_entry *entry; > + int fd = -1; > + char line[PATH_MAX]; > + > + if(pkg->origin == PKG_FROM_CACHE) { > + snprintf(clfile, PATH_MAX, "%s/%s/%s-%s/changelog", > + alpm_option_get_dbpath(), > + alpm_db_get_name(alpm_db_register_local()), I believe you should use handle->db_local here rather than alpm_db_register_local(); > + alpm_pkg_get_name(pkg), > + alpm_pkg_get_version(pkg)); > + } else if(pkg->origin == PKG_FROM_FILE) { > + const char *pkgfile = pkg->origin_data.file; > + > + if((archive = archive_read_new()) == NULL) { > + RET_ERR(PM_ERR_LIBARCHIVE_ERROR, NULL); > + } > + > + archive_read_support_compression_all(archive); > + archive_read_support_format_all(archive); > + > + if (archive_read_open_filename(archive, pkgfile, > + ARCHIVE_DEFAULT_BYTES_PER_BLOCK) != ARCHIVE_OK) { > + RET_ERR(PM_ERR_PKG_OPEN, NULL); > + } > + > + while((ret = archive_read_next_header(archive, &entry)) == > ARCHIVE_OK) { > + const char *entry_name = archive_entry_pathname(entry); > + > + if(strcmp(entry_name, ".CHANGELOG") == 0) { > + tmpfile = strdup("/tmp/alpm_XXXXXX"); > + fd = mkstemp(tmpfile); > + archive_read_data_into_fd(archive, fd); > + break; > + } > + } > + snprintf(clfile, PATH_MAX, "%s", tmpfile); > + } > + You probably got this part from _alpm_pkg_load, I guess it's alright (except you probably want to add an archive_read_finish(archive); there). But you could also use _alpm_unpack, like in _alpm_runscriptlet. Btw, I couldn't get your patch to apply cleanly, because of many line breaks. So I started rewriting it from the start, but in the same time, I wanted to experiment reading directly from the archive, which resulted in the patch I submitted somewhere else in this thread. I don't know if it's better or not, but if it is, all the credits still go to you, for looking at the problem in the first place, and providing a good patch to work with. So thanks again.

1 1

[pacman-dev] [PATCH] Add test directory for testing private libalpm functions.
by Nathan Jones 07 Dec '07

07 Dec '07

I found this useful when testing the shortest_delta_path code. Recently, I ran valgrind on the test which helped to discover the memory leak. Not sure if you guys will want to apply this, but here it is. Signed-off-by: Nathan Jones <nathanj(a)insightbb.com> --- Makefile.am | 2 +- configure.ac | 1 + lib/libalpm/test/.gitignore | 3 + lib/libalpm/test/Makefile.am | 47 +++++++++++++++++ lib/libalpm/test/deltatest.c | 112 ++++++++++++++++++++++++++++++++++++++++++ 5 files changed, 164 insertions(+), 1 deletions(-) create mode 100644 lib/libalpm/test/.gitignore create mode 100644 lib/libalpm/test/Makefile.am create mode 100644 lib/libalpm/test/deltatest.c diff --git a/Makefile.am b/Makefile.am index e07321b..c68786a 100644 --- a/Makefile.am +++ b/Makefile.am @@ -1,4 +1,4 @@ -SUBDIRS = lib/libalpm src/util src/pacman scripts doc etc po pactest contrib +SUBDIRS = lib/libalpm lib/libalpm/test src/util src/pacman scripts doc etc po pactest contrib # Some files automatically included, so they aren't specified below: # AUTHORS, COPYING, NEWS, README diff --git a/configure.ac b/configure.ac index 9a5f3b3..f9f27a7 100644 --- a/configure.ac +++ b/configure.ac @@ -309,6 +309,7 @@ AC_DEFINE_UNQUOTED([DBEXT], "$DBEXT", [The file extension used by pacman databas AC_CONFIG_FILES([ lib/libalpm/Makefile lib/libalpm/po/Makefile.in +lib/libalpm/test/Makefile src/pacman/Makefile src/util/Makefile scripts/Makefile diff --git a/lib/libalpm/test/.gitignore b/lib/libalpm/test/.gitignore new file mode 100644 index 0000000..64ebea7 --- /dev/null +++ b/lib/libalpm/test/.gitignore @@ -0,0 +1,3 @@ +deltatest +.deps +.libs diff --git a/lib/libalpm/test/Makefile.am b/lib/libalpm/test/Makefile.am new file mode 100644 index 0000000..b033018 --- /dev/null +++ b/lib/libalpm/test/Makefile.am @@ -0,0 +1,47 @@ +AUTOMAKE_OPTIONS = gnu + +bin_PROGRAMS = deltatest + +DEFS = -DLOCALEDIR=\"@localedir@\" @DEFS@ + +AM_CFLAGS = -pedantic -D_GNU_SOURCE -I.. -I../../.. + +if ENABLE_VISIBILITY_CC +AM_CFLAGS += -fvisibility=internal +endif +if ENABLE_GNU89_INLINE_CC +AM_CFLAGS += -fgnu89-inline +endif + +deltatest_SOURCES = deltatest.c +deltatest_LDADD = ../.libs/libalpm.a + +libalpm_sources = \ + ../add.h ../add.c \ + ../alpm.h ../alpm.c \ + ../alpm_list.h ../alpm_list.c \ + ../backup.h ../backup.c \ + ../be_files.c ../\ + ../cache.h ../cache.c \ + ../conflict.h ../conflict.c \ + ../db.h ../db.c \ + ../delta.h ../delta.c \ + ../deps.h ../deps.c \ + ../error.h ../error.c \ + ../group.h ../group.c \ + ../handle.h ../handle.c \ + ../log.h ../log.c \ + ../md5.h ../md5.c \ + ../package.h ../package.c \ + ../remove.h ../remove.c \ + ../server.h ../server.c \ + ../sync.h ../sync.c \ + ../trans.h ../trans.c \ + ../util.h ../util.c + +all: ../.libs/libalpm.a + +../.libs/libalpm.a: $(libalpm_sources) + cd .. && make + +# vim:set ts=2 sw=2 noet: diff --git a/lib/libalpm/test/deltatest.c b/lib/libalpm/test/deltatest.c new file mode 100644 index 0000000..e2c3c2e --- /dev/null +++ b/lib/libalpm/test/deltatest.c @@ -0,0 +1,112 @@ +/* + * deltatest.c + * + * Copyright (c) 2007 by Judd Vinet <jvinet(a)zeroflux.org> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation; either version 2 of the License, or + * (at your option) any later version. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, + * USA. + */ +#include "string.h" +#include "assert.h" + +#include "alpm.h" +#include "alpm_list.h" +#include "delta.h" + +int main() +{ + alpm_list_t *deltas = NULL, *path = NULL; + pmdelta_t d1, d2, d3, d4, d5; + + strcpy(d1.from, "1.0-1"); + strcpy(d1.to, "1.1-1"); + d1.size = 10000; + + strcpy(d2.from, "1.1-1"); + strcpy(d2.to, "1.2-1"); + d2.size = 12000; + + strcpy(d3.from, "1.2-1"); + strcpy(d3.to, "1.2-2"); + d3.size = 8000; + + strcpy(d4.from, "1.2-2"); + strcpy(d4.to, "1.3-1"); + d4.size = 15000; + + strcpy(d5.from, "1.0-1"); + strcpy(d5.to, "1.3-1"); + d5.size = 20000; + + deltas = alpm_list_add(deltas, &d1); + deltas = alpm_list_add(deltas, &d2); + deltas = alpm_list_add(deltas, &d3); + deltas = alpm_list_add(deltas, &d4); + deltas = alpm_list_add(deltas, &d5); + + /* test standard path */ + path = _alpm_shortest_delta_path(deltas, "1.0-1", "1.1-1"); + assert(alpm_list_count(path) == 1); + assert(_alpm_delta_path_size(path) == 10000); + alpm_list_free(path); + + path = _alpm_shortest_delta_path(deltas, "1.0-1", "1.2-1"); + assert(alpm_list_count(path) == 2); + assert(_alpm_delta_path_size(path) == 22000); + alpm_list_free(path); + + path = _alpm_shortest_delta_path(deltas, "1.0-1", "1.2-2"); + assert(alpm_list_count(path) == 3); + assert(_alpm_delta_path_size(path) == 30000); + alpm_list_free(path); + + path = _alpm_shortest_delta_path(deltas, "1.0-1", "1.3-1"); + assert(alpm_list_count(path) == 1); + assert(_alpm_delta_path_size(path) == 20000); + alpm_list_free(path); + + path = _alpm_shortest_delta_path(deltas, "1.2-2", "1.3-1"); + assert(alpm_list_count(path) == 1); + assert(_alpm_delta_path_size(path) == 15000); + alpm_list_free(path); + + path = _alpm_shortest_delta_path(deltas, "1.2-1", "1.3-1"); + assert(alpm_list_count(path) == 2); + assert(_alpm_delta_path_size(path) == 23000); + alpm_list_free(path); + + /* test no path */ + path = _alpm_shortest_delta_path(deltas, "0.9-1", "1.3-1"); + assert(alpm_list_count(path) == 0); + alpm_list_free(path); + + path = _alpm_shortest_delta_path(deltas, "1.0-1", "1.4-1"); + assert(alpm_list_count(path) == 0); + alpm_list_free(path); + + path = _alpm_shortest_delta_path(deltas, "0.9-1", "1.4-1"); + assert(alpm_list_count(path) == 0); + alpm_list_free(path); + + /* test reverse direction */ + path = _alpm_shortest_delta_path(deltas, "1.1-1", "1.0-1"); + assert(alpm_list_count(path) == 0); + alpm_list_free(path); + + alpm_list_free(deltas); + + return 0; +} + -- 1.5.3.7

3 2

Re: [pacman-dev] [PATCH] Improve changelog handling
by Xavier 07 Dec '07

07 Dec '07

On Sat, Dec 08, 2007 at 02:47:23AM +1000, Allan McRae wrote: > diff --git a/src/pacman/package.c b/src/pacman/package.c > index ac3f820..00c3535 100644 > --- a/src/pacman/package.c > +++ b/src/pacman/package.c > @@ -232,29 +232,26 @@ void dump_pkg_files(pmpkg_t *pkg) > fflush(stdout); > } > > -/* Display the changelog of an installed package > +/* Display the changelog of a package > */ > -void dump_pkg_changelog(char *clfile, const char *pkgname) > +void dump_pkg_changelog(pmpkg_t *pkg) > { > - FILE* fp = NULL; > - char line[PATH_MAX+1]; > - > - if((fp = fopen(clfile, "r")) == NULL) > - { > - fprintf(stderr, _("error: no changelog available for '%s'.\n"), > pkgname); > + alpm_list_t *changelog = NULL; > + alpm_list_t *i; > + char line[PATH_MAX]; > + > + changelog = alpm_pkg_get_changelog(pkg); > + if(changelog == NULL) { > + fprintf(stderr, _("error: no changelog available for '%s'.\n"), > alpm_pkg_get_name(pkg)); > return; > - } > - else > - { > - while(!feof(fp)) > - { > - fgets(line, (int)PATH_MAX, fp); > - printf("%s", line); > - line[0] = '\0'; > + } else { > + for(i = changelog; i; i = alpm_list_next(i)) { > + snprintf(line, PATH_MAX-1, "%s", (char*)alpm_list_getdata(i)); > + fprintf(stdout, "%s\n", line); > } > - fclose(fp); > - return; > } > + > + FREELIST(changelog); > } > Dan, you said earlier the FREELIST macros shouldn't be used in the frontend. I don't know why I said "ah, ok", because I actually don't understand this, and Nagy didn't either. There are several other places in the frontend where it could be used instead of alpm_list_free_inner(list, free); alpm_list_free(list);

2 3

[pacman-dev] [PATCH] Fix memory leak in shortest_delta_path.
by Nathan Jones 07 Dec '07

07 Dec '07

It is possible for the if statement to never succeed, causing path to never be freed. It is also possible for the if statement to succeed more than once per loop, which could have caused a segfault. Signed-off-by: Nathan Jones <nathanj(a)insightbb.com> --- lib/libalpm/delta.c | 6 ++++-- 1 files changed, 4 insertions(+), 2 deletions(-) diff --git a/lib/libalpm/delta.c b/lib/libalpm/delta.c index 8daac0c..b59dd52 100644 --- a/lib/libalpm/delta.c +++ b/lib/libalpm/delta.c @@ -170,15 +170,15 @@ static alpm_list_t *shortest_delta_path(alpm_list_t *deltas, /* If this vertex has already been visited in the path, go to the * next vertex. */ - if(alpm_list_find_ptr(path, v)) + if(alpm_list_find_ptr(path, v)) { continue; + } /* Once we find a vertex that starts at the 'from' version, * recursively find the shortest path using the 'to' version of this * current vertex as the 'from' version in the function call. */ if(strcmp(v->from, from) == 0) { alpm_list_t *newpath = alpm_list_copy(path); - alpm_list_free(path); newpath = alpm_list_add(newpath, v); newpath = shortest_delta_path(deltas, v->to, to, newpath); @@ -197,6 +197,8 @@ static alpm_list_t *shortest_delta_path(alpm_list_t *deltas, } } + alpm_list_free(path); + return(shortest); } -- 1.5.3.7

1 0

[pacman-dev] [BUG] segfault when the download fails
by Xavier 07 Dec '07

07 Dec '07

I believe my mirror gave me a way to reproduce easily the issue Nagy described there : http://bugs.archlinux.org/task/8350#comment20223 However, it doesn't look at all like the output in the original bug report, where the progress bar is at 64%. So I wasn't sure where to report this information, so I'm reporting it here for now before investigating further. Here is the --debug output: Targets: hunspell-1.2.1-2 openoffice-fr-2.3.1-1 Total Download Size: 15.08 MB [00:27:01] debug: using cachedir: /var/cache/pacman/pkg/ :: Retrieving packages from extra... error: Internal pacman error: Segmentation fault. Please submit a full bug report with --debug if appropriate. [00:27:01] debug: using 'hunspell-1.2.1-2-i686.pkg.tar.gz' for download progress [00:27:01] debug: connected to mir1.archlinuxfr.org successfully [00:27:02] debug: using 'openoffice-fr-2.3.1-1-i686.pkg.tar.gz' for download progress [00:27:02] error: failed retrieving file 'openoffice-fr-2.3.1-1-i686.pkg.tar.gz' from mir1.archlinuxfr.org : Not Found [00:27:02] debug: using 'hunspell-1.2.1-2-i686.pkg.tar.gz' for download progress [00:27:02] error: segmentation fault So pacman first downloads hunspell-1.2.1-2-i686.pkg.tar.gz and succeeds, then it tries to download openoffice-fr-2.3.1-1-i686.pkg.tar.gz and fails. So after that, it tries using the second server in the list. It starts with hunspell-1.2.1-2-i686.pkg.tar.gz again, and then bam, segfault. While I'm finishing this mail, the bug is already gone, since that package is now available on my first mirror. But when it wasn't available, the issue was 100% reproducible, I made it happen 10 times in a row.

3 9

[pacman-dev] Directory permission checking in makepkg
by Allan McRae 06 Dec '07

06 Dec '07

Hi all, I was looking at bug FS#5908 [1] which asks for directory permissions to be checked somewhere in order to stop screwing your system when installing a poorly made package (as directory permissions get overwritten when extracting packages). This at first glance seems about a two line addition to the tidy_install function of makepkg. However, can anybody come up with a case where package directories shouldn't have 755 permissions? If so, I guess this would require either the addition of some option which would stop the directory permissions getting fixed (dirperm?) or just checking/fixing the permissions of a predetermined set of directories. If needed, which is preferable. Cheers, Allan [1] http://bugs.archlinux.org/task/5908

4 5

[pacman-dev] backup handling ugly mess
by Xavier 06 Dec '07

06 Dec '07

Following the recent bash breakage, Pierre was wondering why the /etc/profile just stayed there on the filesystem : http://www.archlinux.org/pipermail/arch-dev-public/2007-December/003556.html I didn't understand this at all. As stonecrest noted earlier, when the "current" mirrorlist was removed from pacman package, that file totally disappeared during the upgrade, even though it was in the old backup array : http://www.archlinux.org/pipermail/pacman-dev/2007-September/009376.html This caused the following commit : http://projects.archlinux.org/git/?p=pacman.git;a=commitdiff;h=843d368ef60a… The first part of this commit is supposed to temporarily add all old backup entries (besides the new) to the NoUpgrade array. Because of a bug, it didn't have any effect. The second part was an idea of me to handle the pacsave during an removeupgrade just like we already did for a remove, but I finally think this is a really stupid / misleading / confusing idea. This code shouldn't even be reached during a removeupgrade if the first part was fixed anyway. So that was just a little sum up, now description of the bug: alpm_pkg_get_backup(newpkg) : newpkg is the one that will be installed, so this only return a list of elements like "file" alpm_pkg_get_backup(oldpkg) : oldpkg is a locally installed package, so this return a list of elements like "file<tab>md5sum" Since the md5sum wasn't removed there, adding these files had 0 effect. Now, this can easily be fixed. But then all files in either old or new backup array that were in oldpkg but not in newpkg will just stay there on the filesystem. They won't be renamed to .pacsave . Why the bash case was different to the pacman one is that, even if /etc/profile was removed from the newpkg, it was still in the backup array of the newpkg (which is the only one pacman looks at). So the file wasn't removed and was just left on the filesystem. Even though the bash case is a packaging error, it's a real case that happened, so it might be worth adding it to the pactests.

3 7