- Pacman-dev - lists.archlinux.org

[pacman-dev] lstat issue / fileconflict001 pactest
by Xavier 12 Nov '07

12 Nov '07

Following the discussion started in the comments of bug 7484, because I think it's more appropriate : http://bugs.archlinux.org/task/7484#comment20837 I recently discovered that a trailing / is very important when using lstat, for directory symlinks. In the following situation : symdir -> realdir/ lstat(symdir) stats the link itself, but lstat(symdir/) stats realdir. So lstat(symdir/) seems equivalent to stat(symdir). The archive extraction code in add.c that Dan rewrote didn't take this into consideration, so he added a new _alpm_lstat that removes the trailing / when there is one : http://projects.archlinux.org/git/?p=pacman.git;a=commit;h=b55abdce7aebb142… A little typo in _alpm_lstat (path instead of newpath) made this change not active yet. As Dan noticed, fixing this makes fileconflict001 pass. The only purpose of this mail is explaining why. So in this pactest, we have the following situation : dir/realdir/ dir/symdir -> realdir Then pkg1 installs dir/realdir/file , and pkg2 installs dir/symdir/file . Since in this case, symdir points to realdir, pkg1 and pkg2 have a fileconflict on "file". But if fileconflict001 pass after the bugfix mentioned above, it's not because pacman detects there is a package<->package conflict between dir/realdir/file and dir/symdir/file. Instead, it detects a filesystem <-> package conflict between dir/symdir/ from pkg2 (which is a real directory there), and the dir/symdir symlink on the filesystem. So that's plain wrong. The fileconflict code currently does a lstat(dir/symdir/), which doesn't stat the symdir symlink, but the real directory realdir. So it treated dir/symdir/ as a directory, and ignored it : 336 /* stat the file - if it exists, do some checks / 337 if(_alpm_lstat(path, &buf) != 0) { 338 continue; 339 } 340 if(S_ISDIR(buf.st_mode)) { 341 _alpm_log(PM_LOG_DEBUG, "%s is a directory, not a conflict\n", path); Now, with the new behavior of _alpm_lstat, the code does a lstat(dir/symdir), which stats the symlink. And the following code will treat this as a conflict. The case of fileconflict001 is probably another situation easier to detect during the transaction (coupled with a rollback system), rather than beforehand. Otherwise, maybe the file conflict code should keep using the normal lstat, instead of _alpm_lstat which removes the trailing /.

4 9

[pacman-dev] Generate doxygen man pages on "make dist"
by Aaron Griffin 12 Nov '07

12 Nov '07

Would anyone have a problem with generating doxygen man pages on the "make dist" target? The reason I ask, is that having a doxygen makedepend is breaking our little core/extra repo separation, and this is the cleanest fix. It also cleans up building pacman. If you ask me, I don't see a problem with pre-generating them Opinions?

2 2

[pacman-dev] TotalDownload option.
by Nathan Jones 12 Nov '07

12 Nov '07

Sample output: $ sudo pacman -S gcc gcc-fortran gcc-libs resolving dependencies... done. looking for inter-conflicts... done. Targets: gcc-libs-4.2.2-2 [1.06 MB] gcc-4.2.2-3 [8.92 MB] gcc-fortran-4.2.2-1 [2.91 MB] Total Package Size: 12.89 MB Total Download Size: 12.89 MB Proceed with installation? [Y/n] :: Retrieving packages from core... gcc-libs-4.2.2-2-i686 1081.1K 503.8K/s 00:00:02 [#--------------------] 8% gcc-4.2.2-3-i686 10.0M 302.1K/s 00:00:34 [################-----] 77% :: Retrieving packages from extra... gcc-fortran-4.2.2-1-... 12.9M 338.0K/s 00:00:39 [#####################] 100% checking package integrity... done. (3/3) checking for file conflicts [#####################] 100% (1/3) upgrading gcc-libs [#####################] 100% (2/3) upgrading gcc [#####################] 100% (3/3) upgrading gcc-fortran [#####################] 100%

8 20

[pacman-dev] documenting the base-devel split
by Jonathan 12 Nov '07

12 Nov '07

I haven't been able to find any arch documentation (besides the mailing lists) on the splitting of base into base and base-devel groups. The main need is for people to understand the need to install the base-devel group before building packages. the makepkg man page seems like a good place for this. Jonathan 'wide-eye'

7 13

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v3.0.0-531-gd0d5848
by Aaron Griffin 11 Nov '07

11 Nov '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 d0d58489ff8b4458719e4bceb6a5d7290c99588a (commit) via 2ee90ddae23dd86c68223c0d6c49f0b92d62429d (commit) from bdab234d977dd2e9417a39f5191e495d5c460ee7 (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 d0d58489ff8b4458719e4bceb6a5d7290c99588a Author: Aaron Griffin <aaronmgriffin(a)gmail.com> Date: Tue Nov 6 22:50:21 2007 -0600 Add STRDUP macro to mirror MALLOC/CALLOC Signed-off-by: Aaron Griffin <aaronmgriffin(a)gmail.com> commit 2ee90ddae23dd86c68223c0d6c49f0b92d62429d Author: Aaron Griffin <aaronmgriffin(a)gmail.com> Date: Tue Nov 6 00:55:45 2007 -0600 Maintain list tail pointers in the head node List head nodes contain null 'prev' pointer, which we can (ab)use to maintain a back reference to the tail pointer of the list. While list additions are not _significantly_ improved, they are still sped up. Original $ time pacman -Qo /usr/bin/wtpt /usr/bin/wtpt is owned by lcms 1.17-2 real 0m3.623s user 0m1.883s sys 0m1.473s New $ time pacman -Qo /usr/bin/wtpt /usr/bin/wtpt is owned by lcms 1.17-2 real 0m2.006s user 0m0.263s sys 0m1.627s Signed-off-by: Aaron Griffin <aaronmgriffin(a)gmail.com> ----------------------------------------------------------------------- Summary of changes: lib/libalpm/alpm_list.c | 59 +++++++++++++++++++++++++++++++++++------------ lib/libalpm/remove.c | 3 +- lib/libalpm/util.h | 2 + 3 files changed, 48 insertions(+), 16 deletions(-) hooks/post-receive -- The official pacman repository

5 12

[pacman-dev] [PATCH] Incorrect usage of alpm_db_whatprovides in sync.c
by Nagy Gabor 10 Nov '07

10 Nov '07

From 82e113f07a8d70e8cc0a1cbd2b1743981494d333 Mon Sep 17 00:00:00 2001 From: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> Date: Sat, 10 Nov 2007 18:11:40 +0100 Subject: [PATCH] Incorrect usage of alpm_db_whatprovides in sync.c The old code thought that alpm_db_whatprovides returns with a list of strings (package names). Signed-off-by: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> --- lib/libalpm/sync.c | 8 ++++---- 1 files changed, 4 insertions(+), 4 deletions(-) diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index 7791e74..e696b58 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -280,9 +280,9 @@ int _alpm_sync_addtarget(pmtrans_t *trans, pmdb_t *db_local, alpm_list_t *dbs_sy if(!p) { RET_ERR(PM_ERR_PKG_NOT_FOUND, -1); } + spkg = (pmpkg_t *) p->data; _alpm_log(PM_LOG_DEBUG, "found '%s' as a provision for '%s'\n", - (char *)p->data, targ); - spkg = _alpm_db_get_pkgfromcache(db, p->data); + alpm_pkg_get_name(spkg), targ); alpm_list_free(p); } } @@ -304,9 +304,9 @@ int _alpm_sync_addtarget(pmtrans_t *trans, pmdb_t *db_local, alpm_list_t *dbs_sy pmdb_t *db = j->data; alpm_list_t *p = _alpm_db_whatprovides(db, targ); if(p) { + spkg = (pmpkg_t *) p->data; _alpm_log(PM_LOG_DEBUG, "found '%s' as a provision for '%s' in db '%s'\n", - (char *)p->data, targ, db->treename); - spkg = _alpm_db_get_pkgfromcache(db, p->data); + alpm_pkg_get_name(spkg), targ, db->treename); alpm_list_free(p); } } -- 1.5.3.4

1 0

[pacman-dev] Hypotetic pacman's db structure (SQL)
by JJDaNiMoTh 09 Nov '07

09 Nov '07

Hello to all. I'm writing an example databe structure for pacman db ( may be useful when someone decide to use SQLite as backend, and not simple text file). [1] Tell me your impression, if something can be made, revisioned or deleted. Views isn't complete, because I need some data stored in the db to test it; and, of course, I'm writing a simple script that parse all packages information stored under /var/lib/pacman to this db schemas ( if anyone wants help, it's welcome :D ) Thanks [1] http://rafb.net/p/XxK62l68.html (Version 0.01) -- JJDaNiMoTh - ArchLinux Trusted User

5 7

Re: [pacman-dev] Some pacman and delta test results
by Dale Ogilvie 09 Nov '07

09 Nov '07

Why wget-xdelta.sh as Xfercommand? I thought I'd try out the Xfercommand "corner-case" of pacman 3.1, as coincidentally I had wget-xdelta.sh in my current pacman.conf. Seems to work ok, unless that segfault I saw was Xfercommand related. Now, I've removed it to take advantage of that delta chain Nathan implemented. Dale

1 0

[pacman-dev] requiredby handling apparently just broke
by Xavier 09 Nov '07

09 Nov '07

Here are my last upgrades: [2007-11-07 22:50] upgraded pacman-git (20071105-1 -> 20071107-1) [2007-11-08 00:22] upgraded glibc (2.7-4 -> 2.7-5) [2007-11-08 08:53] upgraded deltatest (1.0.1-1 -> 1.0.1-2) [2007-11-08 08:53] upgraded libsamplerate (0.1.2-3 -> 0.1.2-4) [2007-11-08 14:57] upgraded deltatest (1.0.1-2 -> 1.0.2rc3-1) [2007-11-09 08:36] upgraded fakeroot (1.7.1-3 -> 1.8.4-1) [2007-11-09 08:36] upgraded mutt (1.5.16-1 -> 1.5.17-1) [2007-11-09 08:36] upgraded vi (7.1.135-1 -> 7.1.154-1) [2007-11-09 08:36] upgraded vim (7.1.135-1 -> 7.1.154-1) [2007-11-09 08:36] upgraded xpdf (3.02_pl1-1 -> 3.02_pl2-1) [2007-11-09 08:42] upgraded git (1.5.3.5-2 -> 1.5.3.5-2) [2007-11-09 18:56] upgraded glibc (2.7-5 -> 2.7-6) [2007-11-09 18:56] upgraded glproto (1.4.8-1 -> 1.4.9-1) [2007-11-09 18:57] upgraded kernel26 (2.6.23.1-6 -> 2.6.23.1-7) [2007-11-09 18:57] upgraded xf86-video-intel (2.1.1-2 -> 2.1.1-3) And now, when I run testdb : > testdb missing requiredby for bash : glibc missing requiredby for module-init-tools : kernel26 missing requiredby for perl : vim I didn't investigate this issue yet, but this would indicate an issue in _alpm_trans_update_depends which has worked great for a while now, and wasn't edited recently. Though, alpm_list_add / alpm_list_remove were changed recently, and Nagy noticed a few bugs in them (didn't look at it either, so not sure if it's really related), but I think the problem might come from there.

3 4

[pacman-dev] Gettext support in makeworld
by Giovanni Scafora 09 Nov '07

09 Nov '07

Hi guys, I attached makeworld.sh.in with gettext support. Thanks. -- Giovanni Scafora Arch Linux Trusted User (voidnull) http://www.archlinux.org linuxmania(a)gmail.com

2 3