- Pacman-dev - lists.archlinux.org

[pacman-dev] pacman packaging
by Dan McGee 10 Jan '07

10 Jan '07

This was just another thought of mine. Pacman is currently distributed as one package on Archlinux (and I would assume on any distribution that is using it). I was thinking it should maybe be two packages, pacman and pacman-utils, or something similar. The two things can be split fairly cleanly- the first is used for installing packages and general maintenence, while the other is used for building packages, looking at the ABS tree, building a custom repo, etc. This also seems more important with a few other scripts being added recently (re-pacman, repo-add) pacman: pacman (and conf/man files), pacman.static, vercmp, rankmirrors, pacman-optimize pacman-utils: abs (and conf files), makepkg (and conf file), makeworld (TODO still wondering if this is used), gensync, resync, repo-add (TODO should this be renamed to addsync?), re-pacman In addition, pacman-utils could then pull in scripts such as srcpac and namcap that are likely installed by most developers anyway. This isn't so much development related, but I thought this was the best list to bring this up. -Dan

4 7

[pacman-dev] Install files and messages
by Dan McGee 10 Jan '07

10 Jan '07

I don't know what other peoples thoughts are, but I think the current install file system and scriptlets...seems like a drop in at some point that wasn't fully done right. Most packages currently use the file for one of two purposes that I can see: 1. display informational messages 2. perform operations that can't be done in the build phase. The install scriptlets were likely originally meant more for number 2, but has become more and more about providing number 1. Here are some relevant bugs and postings: metadata messages in PKGBUILD <http://bugs.archlinux.org/task/1571> display all post-install messages at once <http://bugs.archlinux.org/task/6143> global on-install system <http://bbs.archlinux.org/viewtopic.php?p=219152#219152> Here are my thoughts on the situation. As every package builder seems to have their own presentation format for displaying messages (e.g. echo ">>>...", echo " **...**"), it would be nice to have a way to unify it. Most (all?) messages are in one of two places- post_install or post_upgrade. I propose these messages should be moved out of these scriptlets, and moved into a variable, likely still located in the install file as this is packaged up with the package (unlike the PKGBUILD). These variables could be called install_msg and upgrade_msg, respectively. The scriptlets would still remain, but any output would be 'nonessential'. By changing where the messages stem from, it would make it easier for bug 6143 to be fixed, and would also make for cleaner output to pacman.log, now that all messages are being logged there. Finally, _alpm_runscriptlet seems pretty complicated...I'm wondering if it couldn't be cleaned up a bit by someone that is better with forks and popens. There is also some duplicate code. -Dan

4 5

[pacman-dev] makepkg fix (FS #5923)
by James Rosten 06 Jan '07

06 Jan '07

This is a fix for Flyspray bug #5923 http://bugs.archlinux.org/task/5923. Patch is below. ~ Jamie / yankees26

3 2

[pacman-dev] Long functions and other minor fixes
by James Rosten 03 Jan '07

03 Jan '07

This is my massacre of _alpm_load_pkg in package.c (just joking about the massacre). I moved the extremely large for loop into its own function (and made it a while loop). Feel free to change the name of _alpm_pkg_checker (first thing that came to mind). The patches for package.c and package.h are below. ~ Jamie / yankees26

3 4

[pacman-dev] My $0.02 on the Frugalware v.s. Arch bit...
by Scott Horowitz 02 Jan '07

02 Jan '07

I hate to wade into this, but let me just say... there's no doubt in my mind that this will never be resolved while the (inflammatory) sentiment in the pacman-g2 README exists. For starters, the whole point being made about the API not being stable, which seems to be points 1 and 2 in the README, are plain silly. No project should take a project's cvs development code and expect the API to remain stable; Frugalware knew what it was getting into when it grabbed this code to put into their repositories. There's a reason that pacman3 hasn't been released yet on Archlinux, and that there haven't been Archlinux frontends developed. Pacman3 development shouldn't be hindered because someone decided to release the code into the wild, and it's clearly going under some large refactoring. As for point 3, Aaron's supposed disinterest in contributor's patches - I think it has been quite clear from the mailing list that this isn't the case. The real justification behind this sentiment, which was brought up in the original email and I agree needs improvement, is that there could use more openness (such as in response to a WONTFIX patch) between all the parties. Then there's the stuff about how development has fallen behind schedule (what project hasn't?) and Aaron asked Frugalware to fork things (I'm pretty sure this was more tongue-in-cheeck about how Frugalware is doing things than anything). These seem pretty petty to me. In my opinion, two things need to happen, one from each side. On the Archlinux side, I think Aaron should make a better attempt to be more transparent in his responses, which would hopefully encourage debate. On the Frugalware side, I think the whining about API stability needs to be dropped and, rather than take the "fork" talk as a challenge, they should be asking what needs to be done to make the two paths more compatible. Happy Holidays, Scott

7 11

[pacman-dev] pacman-optimize updates
by James Rosten 31 Dec '06

31 Dec '06

I came up with a patch for FS #5009 (the error when /var/lib/pacman is a loopback device or other mounted partition. I have not been able to test it yet, as I am on my mac (so sending this with Mail, so let us hope it doesn't mess this up :P), but in a few hours I'll be on my Arch computer and I'll test it then and obviously report back. Also, fixed the step number in the comments ('cause I am that nit pickish), because there were two step 1's. ~ Jamie / yankees26

1 0

[pacman-dev] Long functions and other minor fixes
by James Rosten 30 Dec '06

30 Dec '06

I've refactored _alpm_checkconflicts, but there appears to be something slightly off. With a normal cvs build of pacman3, trying to install sonata-svn with normal sonata produces this: loading package data... done.checking dependencies... :: sonata-svn conflicts with sonata. Remove sonata? [Y/n] y error: failed to prepare transaction (conflicting dependencies) :: sonata-svn: conflicts with sonata But with my patch: loading package data... done.checking dependencies... done. checking for file conflicts (1/1) [###############################################################] 100% error: failed to prepare transaction (conflicting files) sonata-svn: /usr/bin/sonata exists in filesystemsonata-svn: /usr/lib/python2.4/site-packages/mmkeys.so exists in filesystemsonata-svn: /usr/lib/python2.4/site-packages/mpdclient3.py exists in filesystemsonata-svn: /usr/lib/python2.4/site-packages/mpdclient3.pyc exists in filesystemsonata-svn: /usr/lib/python2.4/site-packages/sonata.py exists in filesystemsonata-svn: /usr/lib/python2.4/site-packages/sonata.pyc exists in filesystemsonata-svn: /usr/share/applications/sonata.desktop exists in filesystemsonata-svn: /usr/share/locale/de/LC_MESSAGES/sonata.mo exists in filesystemsonata-svn: /usr/share/locale/fr/LC_MESSAGES/sonata.mo exists in filesystemsonata-svn: /usr/share/locale/pl/LC_MESSAGES/sonata.mo exists in filesystemsonata-svn: /usr/share/locale/ru/LC_MESSAGES/sonata.mo exists in filesystemsonata-svn: /usr/share/pixmaps/sonata.png exists in filesystemsonata-svn: /usr/share/pixmaps/sonatacd.png exists in filesystemsonata-svn: /usr/share/pixmaps/sonatacd_large.png exists in filesystemsonata-svn: /usr/share/sonata/CHANGELOG exists in filesystemsonata-svn: /usr/share/sonata/README exists in filesystemsonata-svn: /usr/share/sonata/TODO exists in filesystemsonata-svn: /usr/share/sonata/TRANSLATORS exists in filesystem errors occurred, no packages were upgraded. As you can see it flipped out, but even with the normal pacman3 it still fails to install. Both were executed with: sudo ./pacman.static -A /var/abs/local/sonata-svn/sonata-svn-246-1.pkg.tar.gz Using -U instead of -A yields the same results. So below is the patch for you guys to look over and see what might be screwing this up (cause I have no idea). ~ Jamie / yankees26

1 0

[pacman-dev] Long functions and other minor fixes
by James Rosten 30 Dec '06

30 Dec '06

After trying to build pacman3 so I could test something for stonecrest, I found some errors in my patch and Dan's (very small, you forgot to put the label error before return(-1);) while compiling. So below is the better patch than the other patch (which should be ignored). It patches package.c, package.h, and sync.c of libalpm. ~ Jamie / yankees26

1 0

[pacman-dev] Long functions and other minor fixes
by Dan McGee 29 Dec '06

29 Dec '06

Some more functions that are worth taking a look at on the libalpm side: add.c: _alpm_add_commit be_files.c: _alpm_db_read _alpm_db_write conflict.c: _alpm_checkconflicts _alpm_db_find_conflicts package.c: _alpm_pkg_load (getting a bit long) remove.c: _alpm_remove_commit server.c: _alpm_downloadfiles_forreal sync.c: _alpm_sync_sysupgrade (getting there) _alpm_sync_prepare _alpm_sync_commit And some minor things I found in the code- cleaned up an extern enum, changed 'md5' variable to 'hash' as it can be either md5 or sha1, resorted db.h for clarity, and fixed some English grammar issues. Here is a patch: Index: lib/libalpm/alpm.h =================================================================== RCS file: /home/cvs-pacman/pacman-lib/lib/libalpm/alpm.h,v retrieving revision 1.64 diff -u -u -r1.64 alpm.h --- lib/libalpm/alpm.h 22 Nov 2006 09:03:41 -0000 1.64 +++ lib/libalpm/alpm.h 22 Dec 2006 18:34:01 -0000 @@ -399,7 +399,7 @@ /* * Errors */ -extern enum __pmerrno_t { +enum __pmerrno_t { PM_ERR_MEMORY = 1, PM_ERR_SYSTEM, PM_ERR_BADPERMS, @@ -469,7 +469,9 @@ /* Downloading */ PM_ERR_CONNECT_FAILED, PM_ERR_FORK_FAILED -} pm_errno; +}; + +extern enum __pmerrno_t pm_errno; char *alpm_strerror(int err); Index: lib/libalpm/backup.c =================================================================== RCS file: /home/cvs-pacman/pacman-lib/lib/libalpm/backup.c,v retrieving revision 1.7 diff -u -u -r1.7 backup.c --- lib/libalpm/backup.c 31 Oct 2006 06:39:59 -0000 1.7 +++ lib/libalpm/backup.c 22 Dec 2006 18:34:01 -0000 @@ -55,9 +55,9 @@ ptr++; /* now str points to the filename and ptr points to the md5 or sha1 hash */ if(!strcmp(file, str)) { - char *md5 = strdup(ptr); + char *hash = strdup(ptr); FREE(str); - return(md5); + return(hash); } FREE(str); } Index: lib/libalpm/db.h =================================================================== RCS file: /home/cvs-pacman/pacman-lib/lib/libalpm/db.h,v retrieving revision 1.20 diff -u -u -r1.20 db.h --- lib/libalpm/db.h 20 Nov 2006 09:10:24 -0000 1.20 +++ lib/libalpm/db.h 22 Dec 2006 18:34:01 -0000 @@ -44,10 +44,14 @@ pmlist_t *servers; }; +/* db.c, database general calls */ pmdb_t *_alpm_db_new(char *root, char *dbpath, char *treename); void _alpm_db_free(void *data); int _alpm_db_cmp(const void *db1, const void *db2); pmlist_t *_alpm_db_search(pmdb_t *db, pmlist_t *needles); +pmdb_t *_alpm_db_register(char *treename, alpm_cb_db_register callback); + +/* be.c, backend specific */ int _alpm_db_install(pmdb_t *db, const char *dbfile); int _alpm_db_open(pmdb_t *db); void _alpm_db_close(pmdb_t *db); @@ -58,7 +62,6 @@ int _alpm_db_remove(pmdb_t *db, pmpkg_t *info); int _alpm_db_getlastupdate(pmdb_t *db, char *ts); int _alpm_db_setlastupdate(pmdb_t *db, char *ts); -pmdb_t *_alpm_db_register(char *treename, alpm_cb_db_register callback); #endif /* _ALPM_DB_H */ Index: lib/libalpm/sync.c =================================================================== RCS file: /home/cvs-pacman/pacman-lib/lib/libalpm/sync.c,v retrieving revision 1.85 diff -u -u -r1.85 sync.c --- lib/libalpm/sync.c 1 Dec 2006 09:32:30 -0000 1.85 +++ lib/libalpm/sync.c 22 Dec 2006 18:34:03 -0000 @@ -382,7 +382,7 @@ { pmlist_t *deps = NULL; pmlist_t *list = NULL; /* list allowing checkdeps usage with data from trans->packages */ - pmlist_t *trail = NULL; /* breadcrum list to avoid running into circles */ + pmlist_t *trail = NULL; /* breadcrumb list to avoid running in circles */ pmlist_t *asked = NULL; pmlist_t *i, *j, *k, *l; int ret = 0;

2 4

[pacman-dev] pacman-optimize updates
by Dan McGee 28 Dec '06

28 Dec '06

pacman-optimize updates. Bug #5009 is out there, I'll think about it and try to make a patch hopefully without screwing up my own database. -Dan Index: pacman-optimize =================================================================== RCS file: /home/cvs-pacman/pacman-lib/scripts/pacman-optimize,v retrieving revision 1.3 diff -u -r1.3 pacman-optimize --- pacman-optimize 21 Dec 2006 01:53:41 -0000 1.3 +++ pacman-optimize 22 Dec 2006 19:37:41 -0000 @@ -20,7 +20,7 @@ # USA. # -myver='2.9.8' +myver='3.0.0' usage() { echo "pacman-optimize $myver" @@ -59,10 +59,6 @@ dbroot=$1 fi -if [ "`id -u`" != 0 ]; then - die "You must be root to optimize the database" -fi - # make sure pacman isn't running if [ -f /tmp/pacman.lck ]; then die "Pacman lockfile was found. Cannot run while pacman is running." @@ -72,6 +68,11 @@ die "$dbroot does not exist or is not a directory" fi +#if [ "$EUID" != 0 ]; then +if [ ! -w "$dbroot" ]; then + die "You must have correct permissions to optimize the database" +fi + # don't let pacman run while we do this touch /tmp/pacman.lck

2 2