February 2006 - Pacman-dev - lists.archlinux.org

[pacman-dev] TODO update
by Aurelien Foret 05 Mar '06

05 Mar '06

Hello, I've closed items: [3] code synchro with pacman 2.9.8 [5a] db_write: database entries cleanup [11] _alpm prefix to private library functions On-going items: [7] (vmiklos) [8] (aurelien) the .lastupdate code was moved from the library to the frontend. FYI, I'm currently working on item [6] Here is the updated TODO: [2] possible db corruption because of pmsyncpkg_t struct: the spkg field is just a pointer to data from the cache, which is likely to change during sync_commit [5] database cleanup b) do not add name and version entries in "desc" file (already available in the directory name), and also for .PKGINFO file (in makepkg)? => on hold for now [6] more debug messages in libalpm some parts of the code still remain obscure (file conflicts handling for instance) even when running pacman with "--debug=-1" [7] remove exit() calls from libalpm the library should never call exit() but always return an error code to the frontend (for instance, drop MALLOC macro) [8] .lastupdate review Is the .lastupdate file really needed? Using mtime isn't enough? [9] review file conflict code See this thread: http://www.archlinux.org/pipermail/pacman-dev/2006-February/000202.html [10] pacman -Sw Outputs are wrong compared to pacman 2.9 Do not use log callbacks when the -w is used? or filter all outputs in trans.c depending on the downloadonly flag? => it should be done as the last step before a first release to avoid introducing too much changes to the CVS code at once. -- Aurelien

2 6

[pacman-dev] improving --ignore
by VMiklos 02 Mar '06

02 Mar '06

hi what about to the following improvement: old: $ sudo pacman -S misc-fonts --ignore corefonts :: group misc-fonts: corefonts fontconfig fontforge :: Install whole content? [Y/n] :: corefonts-1.3r4-4: local version is up to date. Upgrade anyway? [Y/n] :: fontforge-20060125-1: local version is up to date. Upgrade anyway? [Y/n] resolving dependencies... done. looking for inter-conflicts... done. Targets: corefonts-1.3r4-4 fontconfig-2.3.93-1 fontforge-20060125-1 Total Package Size: 6.8 MB Proceed with upgrade? [Y/n] n new: $ sudo pacman -S misc-fonts --ignore corefonts :: group misc-fonts: corefonts fontconfig fontforge :: Install whole content? [Y/n] :: About to install corefonts, but it is in IgnorePkg. Install anyway? [Y/n] :: corefonts-1.3r4-4: local version is up to date. Upgrade anyway? [Y/n] :: fontforge-20060125-1: local version is up to date. Upgrade anyway? [Y/n] resolving dependencies... done. looking for inter-conflicts... done. Targets: corefonts-1.3r4-4 fontconfig-2.3.93-1 fontforge-20060125-1 Total Package Size: 6.8 MB Proceed with upgrade? [Y/n] n comments? udv / greetings, VMiklos -- Developer of Frugalware Linux, to make things frugal - http://frugalware.org

2 2

[pacman-dev] release roadmap?
by VMiklos 01 Mar '06

01 Mar '06

hi all just wanted to suggest let's make a list of bugs that blocks us to release a public rc (and then the first stable) release of pacman3 we have several ideas, for example replacing the ugly chroot system() call, adding gettext support, etc, but these features were not in pacman2, and i think they're not "musthave" so what are the points from TODO that must be solved before a release? i think the only one is the lack of handling of package conflicts during packages replacement, and maybe the manpages of public funtions (via doxygen) udv / greetings, VMiklos -- Developer of Frugalware Linux, to make things frugal - http://frugalware.org

3 37

[pacman-dev] [PATCH] pacman: disable progressbar by default if the output is redirected
by VMiklos 01 Mar '06

01 Mar '06

hi http://darcs.frugalware.org/darcsweb/darcsweb.cgi?r=pacman;a=plain_commitdi…; idea from wget ;) udv / greetings, VMiklos -- Developer of Frugalware Linux, to make things frugal - http://frugalware.org

2 1

[pacman-dev] [PATCH] remove corrupted packages if requested
by VMiklos 28 Feb '06

28 Feb '06

hi http://frugalware.org/~vmiklos/patches/libpacman-proposed/unlink_corrupt.di… udv / greetings, VMiklos -- Developer of Frugalware Linux, to make things frugal - http://frugalware.org

1 0

[pacman-dev] [PATCH] use sudo everywhere for makepkg -iS
by VMiklos 28 Feb '06

28 Feb '06

hi http://frugalware.org/~vmiklos/patches/libpacman-proposed/makepkg_install.d… if a package already built, and makepkg was called with -iS, then we should use sudo to install the package. if there were no -S and we're not root, then we should inform the user about we can't auto-install the packages - the same way as we do with the dependency auto-install. as the package install code was duplicated, i've moved it to a separate function udv / greetings, VMiklos -- Developer of Frugalware Linux, to make things frugal - http://frugalware.org

1 0

[pacman-dev] Away
by Aurelien Foret 24 Feb '06

24 Feb '06

I'll be away for the next 3 or 4 days (skiing). I hope the patch queue will not inflate too much in the meantime ;) -- Aurelien

1 0

[pacman-dev] [PATCH] whitespace fixes for libalpm
by VMiklos 24 Feb '06

24 Feb '06

hi http://frugalware.org/~vmiklos/patches/libpacman-proposed/alpm_db_cosmetics… somehow these tabs are replaced with spaces udv / greetings, VMiklos -- Developer of Frugalware Linux, to make things frugal - http://frugalware.org

3 5

[pacman-dev] test/sync999 - pacman segfault
by Aurelien Foret 23 Feb '06

23 Feb '06

Hi, Running pactest today showed that sync999 test case no longer works. Eventually, pactest return on investment does not seem null anymore :) It seems the flaw appeared with lib/libalpm/conflict.c:1.15. The same test with conflict.c:1.14 is ok. Here's the log from pactest: flow2: PM_OPT_DBPATH set to 'var/lib/pacman' flow2: PM_OPT_CACHEDIR set to 'var/cache/pacman/pkg' debug: opening database 'local' debug: opening database 'sync' :: Starting local database upgrade... flow1: checking for package replacements debug: loading package cache (infolevel=0x3) for repository 'sync' flow1: checking for package upgrades debug: loading package cache (infolevel=0x3) for repository 'local' debug: CACHE: fetching 'sync/pkg1' flow2: pkg1-1.0-1 elected for upgrade (1.0-1 => 1.0-2) debug: CACHE: fetching 'sync/pkg2' flow2: pkg2-1.0-1 elected for upgrade (1.0-1 => 1.0-2) resolving dependencies... flow1: resolving targets dependencies done. looking for inter-conflicts... flow1: looking for unresolvable dependencies debug: CACHE: fetching 'local/pkg1' debug: CACHE: fetching 'local/pkg2' flow1: looking for conflicts debug: targs vs db: found pkg2 as a conflict for pkg1 debug: targs vs targs: found pkg2 as a conflict for pkg1 debug: db vs targs: found pkg1 as a conflict for pkg2 flow2: package 'pkg1' is conflicting with 'pkg2' debug: CACHE: fetching 'local/pkg2' debug: package 'pkg1' provides its own conflict debug: resolving package 'pkg1' conflict flow2: electing 'pkg2' for removal flow2: removing 'pkg2' from target list flow2: package 'pkg2' is conflicting with 'pkg1' debug: CACHE: fetching 'local/pkg1' When handling conflict pkg1, pkg2 is removed from targets. Then, the library tries to handle pkg2 conflict... but find_pkginsync() returns NULL... I've committed a fix (sync.c:1.66) to skip conflict resolution in such a case. Thoughts? The test suite is ok with it ;) -- Aurelien

2 4

[pacman-dev] Sorry, test mail for me :)
by Christian Hamar 23 Feb '06

23 Feb '06

Hey. Just a test email, sorry from list members, i need to figure out why i got bounced on this list :( Regards -krix-

1 0