- Pacman-dev - lists.archlinux.org

[pacman-dev] [PATCH] conflict error return fix in sync.c
by Nagy Gabor 24 Jan '08

24 Jan '08

From 1225b562d9ce5eb96cc3a391d7a6eb5553ff7450 Mon Sep 17 00:00:00 2001 From: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> Date: Thu, 24 Jan 2008 22:52:58 +0100 Subject: [PATCH] conflict error return fix in sync.c *some memleak fixes *unify return lists (inner vs outer conflict) Note: this patch introduces some code duplication, which should be fixed, when fixing FS#7524 Signed-off-by: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> --- lib/libalpm/sync.c | 29 +++++++++++++---------------- 1 files changed, 13 insertions(+), 16 deletions(-) diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index 3e99795..603214d 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -485,9 +485,15 @@ int _alpm_sync_prepare(pmtrans_t *trans, pmdb_t *db_local, alpm_list_t *dbs_sync rsync = sync1; sync = sync2; } else { + _alpm_log(PM_LOG_ERROR, _("unresolvable package conflicts detected\n")); pm_errno = PM_ERR_CONFLICTING_DEPS; ret = -1; - *data = deps; + if(data) { + *data = deps; + } else { + alpm_list_free_inner(deps, (alpm_list_fn_free)_alpm_conflict_free); + alpm_list_free(deps); + } goto cleanup; } @@ -552,23 +558,14 @@ int _alpm_sync_prepare(pmtrans_t *trans, pmdb_t *db_local, alpm_list_t *dbs_sync sync->data = alpm_list_add(sync->data, local); } else { /* abort */ _alpm_log(PM_LOG_ERROR, _("unresolvable package conflicts detected\n")); - if(data) { - pmconflict_t *retconflict; - MALLOC(retconflict, sizeof(pmconflict_t), 0); - if(!retconflict) { - pm_errno = PM_ERR_MEMORY; - alpm_list_free_inner(*data, (alpm_list_fn_free)_alpm_conflict_free); - alpm_list_free(*data); - ret = -1; - goto cleanup; - } - *retconflict = *conflict; - *data = alpm_list_add(*data, retconflict); - } pm_errno = PM_ERR_CONFLICTING_DEPS; ret = -1; - alpm_list_free_inner(deps, (alpm_list_fn_free)_alpm_conflict_free); - alpm_list_free(deps); + if(data) { + *data = deps; + } else { + alpm_list_free_inner(deps, (alpm_list_fn_free)_alpm_conflict_free); + alpm_list_free(deps); + } goto cleanup; } } -- 1.5.3.7

1 0

[pacman-dev] Fwd: [arch-general] makepkg not stripping bin/libs
by Travis Willard 24 Jan '08

24 Jan '08

This feels more like pacman-dev material - forwarded there. ---------- Forwarded message ---------- From: Mark Constable <markc(a)renta.net> Date: Jan 24, 2008 6:58 AM Subject: [arch-general] makepkg not stripping bin/libs To: General Discusson about Arch Linux <arch-general(a)archlinux.org> makegpkg 3.1.1 line ~752 has this... find {,usr/{,local/},opt/}{bin,lib,sbin} -type f and I can't see how that will ever work, and it doesn't for me, however this does work (for me)... find -regex ".+/bin/.+" -o -regex ".+/lib/.+" -o -regex ".+/sbin/.+" -type f I recently rebuilt kdemod4 and it ended up being 1.9Gb whereas it should have been around >400Mb. --markc

3 3

[pacman-dev] Some small issues with backup relocation: upgrade044.py; needless .pacorig
by Nagy Gabor 24 Jan '08

24 Jan '08

Look at the following pactest file: ---upgrade044.py--- self.description = "Backup file relocation (package split)" lp = pmpkg("foo") lp.files = ["etc/foo.cfg*"] lp.backup = ["etc/foo.cfg"] self.addpkg2db("local", lp) p1 = pmpkg("foo", "1.0-2") self.addpkg(p1) p2 = pmpkg("libfoo", "1.0-2") p2.files = ["etc/foo.cfg**"] p2.backup = ["etc/foo.cfg"] self.addpkg(p2) self.filesystem = ["etc/foo.cfg"] self.args = "-U %s" % " ".join([p.filename() for p in p1, p2]) self.addrule("PKG_VERSION=foo|1.0-2") self.addrule("PKG_VERSION=libfoo|1.0-2") self.addrule("!FILE_PACSAVE=etc/foo.cfg") self.addrule("FILE_PACNEW=etc/foo.cfg") self.addrule("FILE_EXIST=etc/foo.cfg") ---------- This pactest fails, which means that we get different behaviour, if we had libfoo installed (see upgrade042.py and upgrade043.py). The difference is the immediate corollary of line 501 in add.c (libalpm). It's important to see, that if --force wasn't set we reach that line iff the file was backed up by the transaction, so .pacorig has no meaning in this case. So this .pacorig stuff has meaning with --force, but as I see backup handling with --force is not well defined :-P Bye, ngaba PS: and this .pacorig extension is quite misleading, it suggests "original config file" to me; using .pacsave is more suggestive here imho. ---------------------------------------------------- SZTE Egyetemi Könyvtár - http://www.bibl.u-szeged.hu This mail sent through IMP: http://horde.org/imp/

1 1

[pacman-dev] [pacman-git] New release
by Dan McGee 24 Jan '08

24 Jan '08

I pushed a new release tonight. It is the current (88cbee) cut of the master branch. Stable enough for me to trust it here. :) Vastly reduced memory usage compared to v3.1.1 now- I posted a little writeup on my blog about how massif helped us examine this memory issue, as it wasn't a memleak problem at all. That entry is here: http://toofishes.net/blog/valgrind-330-and-new-massif/ As usual, if you encounter problems, reply here or open a bug report. I did enable a debug compile for this so you should be able to use gdb and have full symbols on a segfault backtrace or something. -Dan

1 0

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v3.1.1-27-g88cbee3
by Dan McGee 24 Jan '08

24 Jan '08

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 88cbee3c24768470cc0d4abe050e98b989807a67 (commit) via 69c6d59bb6efc1ae1bcf58d19d049d435dc974b2 (commit) via 8240da6cb3ff95ad480efe3e1876104024398fae (commit) via bd43a7f155cbd54d58fb45285e2eef1f51c88b00 (commit) via 22c900e7d57bfa7c9ed482f888ef200c0289ceec (commit) via 6b8f404a33b14ccb89f3adedf438f08b330a177a (commit) from 8ca6501ee1e3041f4fb186a33a9faf76f973efa2 (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 88cbee3c24768470cc0d4abe050e98b989807a67 Author: Chantry Xavier <shiningxc(a)gmail.com> Date: Mon Aug 13 00:17:32 2007 +0200 libalpm/sync.c : conflicts resolving cleanup. The original patch from Nagy tried to resolve target vs target conflicts, and so broke the following pactests : sync040, sync041 and sync990 Nagy's proposal to solve this situation was to choose the interactive way, ask the user how to deal with it: either remove pkg1 or remove pkg2 or stop here. So he left this as a TODO. But instead of trying to resolve these conflicts or asking the user, I tried to find a more conservative way, looking at what the current pactests expected: If between the two conflicting packages, one provides the other, pacman will keep that one and remove the other from the target list. That breaks sync893 and sync897. But Dan agreed these two looked weird and should be changed. This commit should close FS#8897, FS#8899 and FS#9024. Reference: http://www.archlinux.org/pipermail/pacman-dev/2007-October/009745.html http://www.archlinux.org/pipermail/pacman-dev/2007-December/010393.html Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 69c6d59bb6efc1ae1bcf58d19d049d435dc974b2 Author: Dan McGee <dan(a)archlinux.org> Date: Tue Jan 22 18:52:21 2008 -0600 Fix flip-flopped parameters to CALLOC macro The mixup causes a fail in the build using --enable-debug on x86_64 but not i686, so none of us caught this right away. Fix it. FS#9297. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 8240da6cb3ff95ad480efe3e1876104024398fae Author: Chantry Xavier <shiningxc(a)gmail.com> Date: Sat Jan 12 21:28:22 2008 +0100 libalpm/cache.c : don't duplicate packages in pkgcache. Edit _alpm_db_add_pkgincache to not duplicate packages, because this is not needed, is slower, and uses more memory. This made the max memory usage during base reinstall go from 10.4MB to 9.7MB. Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit bd43a7f155cbd54d58fb45285e2eef1f51c88b00 Author: Chantry Xavier <shiningxc(a)gmail.com> Date: Sat Dec 8 00:17:34 2007 +0100 testdb : add conflicts checking. Testdb will now scan the local database to check there are no conflicts. I'm not sure how it's possible to get this situation. At least pacman doesn't allow installing a package that conflicts with a local one, without removing the local one. But maybe in a very weird situation, or because of a bug, it could be possible. But anyway, this only requires 5 lines of code. Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 22c900e7d57bfa7c9ed482f888ef200c0289ceec Author: Chantry Xavier <shiningxc(a)gmail.com> Date: Sun Dec 9 14:49:34 2007 +0100 Add new public alpm_checkdbconflicts function. This function has a limited purpose, but might be interesting to do a sanity check from a frontend (eg testdb). Also removed the private _alpm_checkconflicts function to avoid confusion. This function was used only once in libalpm, in sync.c, and was just a single line anyway. Having to do it manually makes it explicit that we are looking for two kind of conflicts (targ vs targ and db vs targ). Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 6b8f404a33b14ccb89f3adedf438f08b330a177a Author: Chantry Xavier <shiningxc(a)gmail.com> Date: Wed Jan 23 20:25:44 2008 +0100 small memleak fix in sync.c. Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> ----------------------------------------------------------------------- Summary of changes: lib/libalpm/alpm.h | 2 + lib/libalpm/cache.c | 15 +-- lib/libalpm/conflict.c | 10 ++- lib/libalpm/conflict.h | 1 - lib/libalpm/deps.c | 4 +- lib/libalpm/sync.c | 261 +++++++++++++++++++-------------------------- lib/libalpm/trans.c | 3 +- pactest/tests/sync893.py | 4 +- pactest/tests/sync897.py | 6 +- src/util/testdb.c | 8 ++ 10 files changed, 139 insertions(+), 175 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] issues with cleaning a shared pacman cache with new -Sc behaviour
by Roman Kyrylych 22 Jan '08

22 Jan '08

Hi! I've found an issue that appeared after the change in -Sc behaviour. It now removes "uninstalled" packages from cache which is good for majority of users, but is bad for shared pacman cache users. With a pacman cache shared between machines it is impossible to correctly clean it with pacman if a set of installed packages on them is different. I remember there was at least one script posted here or on forums that removes old version of packages from a specified directory (it was provided as a helpful tool for managing local repositories AFAIR), but I cannot find it. Could someone who have this (or similar) script please share it? It needs to be posted to our wiki page about a shared cache, and maybe Phil's rucksack script needs to be modified as well. -- Roman Kyrylych (Роман Кирилич)

6 6

[pacman-dev] [GIT] The official pacman repository branch, maint, updated. v3.1.1-4-gbba6265
by Dan McGee 22 Jan '08

22 Jan '08

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, maint has been updated via bba62655fe06989e89576e66a97607da88aa9916 (commit) via 73ee64d49f8703e2b732ae03771a63ef56208601 (commit) from ba7687f58eab73627ba9325ad117ef2fbbeebb12 (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 ----------------------------------------------------------------- ----------------------------------------------------------------------- Summary of changes: scripts/makepkg.sh.in | 2 +- src/pacman/add.c | 2 +- src/pacman/callback.c | 2 +- 3 files changed, 3 insertions(+), 3 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v3.1.1-21-g8ca6501
by Dan McGee 22 Jan '08

22 Jan '08

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 8ca6501ee1e3041f4fb186a33a9faf76f973efa2 (commit) via 9247ddbe8aaebf00bf9b93a753b73aa09b5de014 (commit) via 36264a3ab9ec83ffe505dade32115a31e858122e (commit) via cbcf542ad22c479f3e0be9698e13c52687953ec9 (commit) via 2a7101c049dd1874da09d7d120f8855c61e55809 (commit) via b2914bf0af388f369865859292b1c7342e785303 (commit) via 927af790ee3ff1495acd2c6b33378a7ab20e0c67 (commit) via bba62655fe06989e89576e66a97607da88aa9916 (commit) via 73ee64d49f8703e2b732ae03771a63ef56208601 (commit) via ba7687f58eab73627ba9325ad117ef2fbbeebb12 (commit) via 003adb7646c6e7d00b6cac26586aa3678ade660b (commit) via 3b464dc89774ea83d4fe4c837b751d4ecc637e43 (commit) via cee314fc5e93e20392072c0182781b59d3fdbb5e (commit) from f6785dcb89e380f0f3062b4b7aa485e1cd3c5830 (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 8ca6501ee1e3041f4fb186a33a9faf76f973efa2 Author: Dan McGee <dan(a)archlinux.org> Date: Mon Jan 21 19:47:03 2008 -0600 pactest: remove check for root permissions message Pacman isn't designed to fail when under fakeroot anymore, so kill this check off which fails anyway if the logfile is missing (such as when the --valgrind flag is used). Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 9247ddbe8aaebf00bf9b93a753b73aa09b5de014 Author: Dan McGee <dan(a)archlinux.org> Date: Mon Jan 14 00:59:11 2008 -0600 conflict.c: move debug message out of inner loop We have a debug message in the target vs. target file conflict check, and this is a bit rediculous when it comes to watching output from something like smoke001.py. Instead, put the output outside this inner loop so we only see it at most once per target, which is much more reasonable. Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 36264a3ab9ec83ffe505dade32115a31e858122e Author: Dan McGee <dan(a)archlinux.org> Date: Mon Jan 14 00:33:13 2008 -0600 Fix memleak found by add004.py Signed-off-by: Dan McGee <dan(a)archlinux.org> commit cbcf542ad22c479f3e0be9698e13c52687953ec9 Merge: 2a7101c049dd1874da09d7d120f8855c61e55809 bba62655fe06989e89576e66a97607da88aa9916 Author: Dan McGee <dan(a)archlinux.org> Date: Mon Jan 21 19:42:08 2008 -0600 Merge branch 'maint' commit 2a7101c049dd1874da09d7d120f8855c61e55809 Author: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> Date: Sun Jan 13 22:15:10 2008 +0100 New --asexplicit option This is the symmetric of --asdeps, install packages explicitly. Documentation and completion files were updated accordingly. Added sync301.py and upgrade032.py pactest files to test this. I also made a little modification in ALLDEPS handling too. Signed-off-by: Nagy Gabor <ngaba(a)bibl.u-szeged.hu> Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> commit b2914bf0af388f369865859292b1c7342e785303 Author: Chantry Xavier <shiningxc(a)gmail.com> Date: Tue Jan 22 01:28:05 2008 +0100 Move the deptest code from frontend to backend. The deptest code (pacman -T) used by makepkg was mostly in the frontend. There were 2 drawbacks: 1) the public splitdep function returns a pmdepend_t struct, but the _alpm_dep_free function for freeing it is private. So there was a memleak. 2) there is a helper in the backend (satisfycmp in deps.c) which makes this function much easier. So this adds a new public alpm_deptest in libalpm/deps.c, which cleans pacman_deptest in pacman/deptest.c a lot. Besides, alpm_splitdep was made private, because the frontend no longer requires it, and _alpm_dep_free is also private. Finally the deptest001 pactest was extended. Signed-off-by: Chantry Xavier <shiningxc(a)gmail.com> Signed-off-by: Dan McGee <dan(a)archlinux.org> commit 927af790ee3ff1495acd2c6b33378a7ab20e0c67 Author: Dan McGee <dan(a)archlinux.org> Date: Sat Jan 12 01:27:02 2008 -0600 Convert deltas to dynamic allocation Another elimination of a static length structure in libalpm. Should result in a little more memory saved during execution of packages with lots of deltas attached. Signed-off-by: Dan McGee <dan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: TRANSLATORS | 17 ++++++--- configure.ac | 9 +++-- contrib/bash_completion | 2 + contrib/zsh_completion | 1 + doc/pacman.8.txt | 7 +++- lib/libalpm/add.c | 37 ++++++++----------- lib/libalpm/alpm.h | 5 ++- lib/libalpm/be_files.c | 5 ++- lib/libalpm/conflict.c | 6 ++-- lib/libalpm/delta.c | 23 ++++++++---- lib/libalpm/delta.h | 18 ++++----- lib/libalpm/deps.c | 27 +++++++++++++- lib/libalpm/deps.h | 4 ++- lib/libalpm/package.c | 5 ++- pactest/pmtest.py | 5 --- pactest/tests/deptest001.py | 24 +++++------- pactest/tests/{sync030.py => sync031.py} | 10 +++--- pactest/tests/{upgrade031.py => upgrade032.py} | 10 +++--- po/de.po | 4 +- scripts/makepkg.sh.in | 8 ++++- src/pacman/add.c | 2 +- src/pacman/callback.c | 2 +- src/pacman/deptest.c | 46 ++++------------------- src/pacman/pacman.c | 7 ++++ src/pacman/util.c | 4 +- 25 files changed, 155 insertions(+), 133 deletions(-) copy pactest/tests/{sync030.py => sync031.py} (57%) copy pactest/tests/{upgrade031.py => upgrade032.py} (54%) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] pacman -S on already installed package doesn't ask for reinstall
by Roman Kyrylych 21 Jan '08

21 Jan '08

Hi! # pacman -S ttf-ms-fonts warning: ttf-ms-fonts-2.0-1 is up to date -- reinstalling It was asking user for confirmation before which I think was logical, because if user just forgot he/she has this package installed - this would allow to avoid creating of unnecessary .pacnew files. Either I missed this change or forgot about it. Could someone enlighten me why this was done? -- Roman Kyrylych (Роман Кирилич)

5 6

[pacman-dev] bad order of messages in output
by Vojtěch Gondžala 20 Jan '08

20 Jan '08

Hi, I type this command: `pacman -R pacman` and output look very crazy: # pacman -R pacman :: pacman is designated as a HoldPkg. Remove anyway? [Y/n] loading package data... n failed. error: failed to add target 'pacman' (not confirmed)

1 0