- Pacman-dev - lists.archlinux.org

[pacman-dev] [PATCH] trans_prepare: always sort trans->remove by deps
by Andrew Gregory 19 Dec '13

19 Dec '13

Packages can be removed during a sync transaction either directly or due to conflicts and need to be sorted. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> --- lib/libalpm/remove.c | 7 ------- lib/libalpm/sync.c | 5 ----- lib/libalpm/trans.c | 16 ++++++++++++++++ test/pacman/tests/TESTS | 1 + test/pacman/tests/sync046.py | 29 +++++++++++++++++++++++++++++ 5 files changed, 46 insertions(+), 12 deletions(-) create mode 100644 test/pacman/tests/sync046.py diff --git a/lib/libalpm/remove.c b/lib/libalpm/remove.c index 9417a61..6c50ab3 100644 --- a/lib/libalpm/remove.c +++ b/lib/libalpm/remove.c @@ -240,13 +240,6 @@ int _alpm_remove_prepare(alpm_handle_t *handle, alpm_list_t **data) } } - /* re-order w.r.t. dependencies */ - _alpm_log(handle, ALPM_LOG_DEBUG, "sorting by dependencies\n"); - lp = _alpm_sortbydeps(handle, trans->remove, NULL, 1); - /* free the old alltargs */ - alpm_list_free(trans->remove); - trans->remove = lp; - /* -Rcs == -Rc then -Rs */ if((trans->flags & ALPM_TRANS_FLAG_CASCADE) && (trans->flags & ALPM_TRANS_FLAG_RECURSE)) { diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index 7dbb8a0..ba71d89 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -625,11 +625,6 @@ int _alpm_sync_prepare(alpm_handle_t *handle, alpm_list_t **data) } goto cleanup; } - - /* re-order w.r.t. dependencies */ - alpm_list_t *add_orig = trans->add; - trans->add = _alpm_sortbydeps(handle, add_orig, trans->remove, 0); - alpm_list_free(add_orig); } for(i = trans->add; i; i = i->next) { /* update download size field */ diff --git a/lib/libalpm/trans.c b/lib/libalpm/trans.c index a795a1f..6ace823 100644 --- a/lib/libalpm/trans.c +++ b/lib/libalpm/trans.c @@ -39,6 +39,7 @@ #include "remove.h" #include "sync.h" #include "alpm.h" +#include "deps.h" /** \addtogroup alpm_trans Transaction Functions * @brief Functions to manipulate libalpm transactions @@ -134,6 +135,21 @@ int SYMEXPORT alpm_trans_prepare(alpm_handle_t *handle, alpm_list_t **data) } } + + if(!(trans->flags & ALPM_TRANS_FLAG_NODEPS)) { + _alpm_log(handle, ALPM_LOG_DEBUG, "sorting by dependencies\n"); + if(trans->add) { + alpm_list_t *add_orig = trans->add; + trans->add = _alpm_sortbydeps(handle, add_orig, trans->remove, 0); + alpm_list_free(add_orig); + } + if(trans->remove) { + alpm_list_t *rem_orig = trans->remove; + trans->remove = _alpm_sortbydeps(handle, rem_orig, NULL, 1); + alpm_list_free(rem_orig); + } + } + trans->state = STATE_PREPARED; return 0; diff --git a/test/pacman/tests/TESTS b/test/pacman/tests/TESTS index fc6a7e8..f52a1f7 100644 --- a/test/pacman/tests/TESTS +++ b/test/pacman/tests/TESTS @@ -158,6 +158,7 @@ TESTS += test/pacman/tests/sync042.py TESTS += test/pacman/tests/sync043.py TESTS += test/pacman/tests/sync044.py TESTS += test/pacman/tests/sync045.py +TESTS += test/pacman/tests/sync046.py TESTS += test/pacman/tests/sync050.py TESTS += test/pacman/tests/sync051.py TESTS += test/pacman/tests/sync052.py diff --git a/test/pacman/tests/sync046.py b/test/pacman/tests/sync046.py new file mode 100644 index 0000000..63e8cc3 --- /dev/null +++ b/test/pacman/tests/sync046.py @@ -0,0 +1,29 @@ +self.description = "Install a sync package with interdependent conflicts" + +sp = pmpkg("pkg1") +sp.conflicts = ["pkg2", "pkg3", "pkg4"] +self.addpkg2db("sync", sp); + +lp1 = pmpkg("pkg2") +lp1.depends = ["pkg3"] +lp1.install['pre_remove'] = "[ -f pkg3file ] && echo '' > pkg2ok" +self.addpkg2db("local", lp1); + +lp2 = pmpkg("pkg3") +lp2.files = ["pkg3file"] +self.addpkg2db("local", lp2); + +lp3 = pmpkg("pkg4") +lp3.depends = ["pkg3"] +lp3.install['pre_remove'] = "[ -f pkg3file ] && echo '' > pkg4ok" +self.addpkg2db("local", lp3); + +self.args = "-S %s --ask=4" % sp.name + +self.addrule("PACMAN_RETCODE=0") +self.addrule("PKG_EXIST=pkg1") +self.addrule("!PKG_EXIST=pkg2") +self.addrule("!PKG_EXIST=pkg3") +self.addrule("!PKG_EXIST=pkg4") +self.addrule("FILE_EXIST=pkg2ok") +self.addrule("FILE_EXIST=pkg4ok") -- 1.8.5.1

2 1

[pacman-dev] [PATCH] dload: allow curl to response to any auth challenge
by Dave Reisner 19 Dec '13

19 Dec '13

Previously, we only allowed the default of responding to basic auth challenges. Mirrors requiring authorization are far and away the edge case, but there's no sense in preventing access to them. Implements FS#38184. Signed-off-by: Dave Reisner <dreisner(a)archlinux.org> --- lib/libalpm/dload.c | 1 + 1 file changed, 1 insertion(+) diff --git a/lib/libalpm/dload.c b/lib/libalpm/dload.c index c74d2ad..d5906a3 100644 --- a/lib/libalpm/dload.c +++ b/lib/libalpm/dload.c @@ -301,6 +301,7 @@ static void curl_set_handle_opts(struct dload_payload *payload, curl_easy_setopt(curl, CURLOPT_NETRC, CURL_NETRC_OPTIONAL); curl_easy_setopt(curl, CURLOPT_SOCKOPTFUNCTION, dload_sockopt_cb); curl_easy_setopt(curl, CURLOPT_SOCKOPTDATA, (void *)handle); + curl_easy_setopt(curl, CURLOPT_HTTPAUTH, CURLAUTH_ANY); _alpm_log(handle, ALPM_LOG_DEBUG, "url: %s\n", payload->fileurl); -- 1.8.5.2

2 1

[pacman-dev] [PATCH 4/4] makepkg: git: update existing sources in srcdir without removing them first.
by SamLT 18 Dec '13

18 Dec '13

Hello, [sorry could not hit the reply button since I'm new to this list] > William Giokas 1007380 at gmail.com > Tue Dec 10 14:56:18 EST 2013 > > Previous message: [pacman-dev] [PATCH 4/4] makepkg: git: update existing sources in srcdir without removing them first. > Next message: [pacman-dev] [PATCH 4/4] makepkg: git: update existing sources in srcdir without removing them first. > Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] > > On Tue, Dec 10, 2013 at 09:39:51AM +0100, Lukas Jirkovsky wrote: > > On Mon, Dec 9, 2013 at 10:10 PM, Dave Reisner <d at falconindy.com> wrote: > > > This groundbreaking idea has been proposed and rejected several times > > > already. > > > > > > > Can you point me to where they were rejected? I have submitted the > > patches implementing this functionality in the past, but they were > > rejected because of technical problems, not because of the idea > > itself. This new patch series should fix all the issues that were > > present in my previous attempts. Also I remember that Allan set > > FS#35050 as due in 4.2.0 probably because I was working on it. > > For the tl;dr: > > Here's how we use git sources at the moment: > > - Create bare clone (no checked out files) > - Clone the bare clone to build dir > - Build > - Possible cleanup > > With a shallow clone the first two steps won't work at all. Look at the > manual page for git-clone(1): > > A shallow repository has a number of limitations (you cannot clone or > fetch from it, nor push from nor into it)... > > As you can see, we can't do the second step, and if we do the first step > with --depth=1, then it would essentially just give us a useless set of > files. I've also read the link you gave earlier about why shallow clones are bad: https://mailman.archlinux.org/pipermail/aur-general/2013-April/022938.html While I agree that the first step should *not* be a shallow clone, I fail to see why we clone the whole history for the build dir. AFAICS, the whole discussion in both of these threads is about the first step, I'm talking about the second step, ie making the sources to the build dir Uwe Koloska said "local clones are cheap", (https://mailman.archlinux.org/pipermail/aur-general/2013-April/022955.html ) while I agree it's fast it also copies a whole bunch of files which won't be used for the build (namely the .git directory). Although, this does not feel right(to me), I cannot find any tangible reason other that premature EOL of our beloved SSD *sick*. Anyway, why not just doing something like this for step 2: cd $git archive --format=tar --prefix=$pkgname/ . | tar xf - -C $srcdir (or even a shallow clone) I may be missing something, eg, I've never played with git submodules and don't know how well this approach would work with them. What do you think? Thanks for your time! sam > > Thanks, > -- > William Giokas | KaiSforza | http://kaictl.net/ > GnuPG Key: 0x73CD09CF > Fingerprint: F73F 50EF BBE2 9846 8306 E6B8 6902 06D8 73CD 09CF >

3 3

[pacman-dev] [GIT] The official pacman repository branch, master, updated. v4.1.2-219-g31fcdec
by allan＠archlinux.org 15 Dec '13

15 Dec '13

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 31fcdec619a5d738221b6629ec5e48205b4e04fe (commit) via e8ec7e54e5c293fd4e3e36b8aeb4f7d60a44a466 (commit) via 841c60f2b396c3b59c7d673dc39dacc3ccbf5658 (commit) via 76a05e94c19922ee27a9c3b751ec3836c13a1bab (commit) via 916c7085d829e6fec4f31b7ea3dbbdf571f9cd81 (commit) via b847d1d8e7a032afa51c7ccce1b253ec451bb6b7 (commit) via 9749fb033c8794e8ac4b931678659aa2b79ae1b1 (commit) via ae2c9bba5a0de58454814c5fa438cf1db52707ae (commit) via af2c0a15371058ba93d6bb423b3fe90e45380d7a (commit) via 20127e732c66897506449c38baff9ec1a893032c (commit) via 759c48cd34235a0650f8189223b812dad53780cc (commit) via 1340ed792501c3a2c65b83ff90b232e6e01f5537 (commit) from a15fd48016014249a13db7a437b5e4f5937f6158 (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 31fcdec619a5d738221b6629ec5e48205b4e04fe Author: Allan McRae <allan(a)archlinux.org> Date: Fri Dec 13 20:53:12 2013 +1000 Fix pactest README The --test option no longer exists. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit e8ec7e54e5c293fd4e3e36b8aeb4f7d60a44a466 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Wed Dec 11 11:58:27 2013 -0500 remove useless continue Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 841c60f2b396c3b59c7d673dc39dacc3ccbf5658 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Sat Nov 30 20:45:02 2013 -0500 db.c: require unique database names Allowing multiple databases with the same name causes conflicts as they both point to the same database file but may use different servers, usages, or siglevels. Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 76a05e94c19922ee27a9c3b751ec3836c13a1bab Author: Sébastien Luttringer <seblu(a)seblu.net> Date: Tue Dec 10 01:59:24 2013 +0100 makepkg: fix consistency of downloaded sources permissions Running 'makepkg -g' or 'makepkg' to download source files results in different permissions on the files if the user has a non-default umask. Put the umask definition at the "beginning" of the makepkg script to ensure all files generated by makepkg have a 0022 umask. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 916c7085d829e6fec4f31b7ea3dbbdf571f9cd81 Author: Andrew Gregory <andrew.gregory.8(a)gmail.com> Date: Mon Nov 25 11:16:47 2013 -0500 alpm_sync_sysupgrade: skip packages being removed Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit b847d1d8e7a032afa51c7ccce1b253ec451bb6b7 Author: Allan McRae <allan(a)archlinux.org> Date: Thu Nov 21 12:55:39 2013 +1000 Color output from pm_vasprintf This makes sure warning and error messages from _alpm_log are colored. Fixes FS#35160. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 9749fb033c8794e8ac4b931678659aa2b79ae1b1 Author: Jason St. John <jstjohn(a)purdue.edu> Date: Thu Nov 14 23:06:24 2013 -0500 Fix misspellings and some grammar issues in output of some pacman contrib scripts Affected files: -- contrib/bacman.sh.in -- contrib/paccache.sh.in -- contrib/pacdiff.sh.in -- contrib/rankmirrors.sh.in Signed-off-by: Jason St. John <jstjohn(a)purdue.edu> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit ae2c9bba5a0de58454814c5fa438cf1db52707ae Author: Vasiliy Korchagin <vasiliy.korchagin(a)gmail.com> Date: Mon Nov 18 13:01:38 2013 +1000 pacdiff: do not remove pacfile after viewing Automatically removing the file after viewing is not good if on viewing you decide to deal with it later. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit af2c0a15371058ba93d6bb423b3fe90e45380d7a Author: Jason St. John <jstjohn(a)purdue.edu> Date: Fri Nov 15 23:56:05 2013 -0500 pacscripts: Improve error message when sudo binary is not found Signed-off-by: Jason St. John <jstjohn(a)purdue.edu> Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 20127e732c66897506449c38baff9ec1a893032c Author: Wolfgang Bumiller <wry.git(a)bumiller.com> Date: Fri Nov 15 20:54:27 2013 +1000 Close file descirptor before fork FreeBSD chroot(2) fails with EPERM when a directory file descriptor is open with either `kern.chroot_allow_open_directories` being 0, or when the process already is inside a chroot. This is exposed in alpm_run_chroot that uses opendir() to open a file descriptor to the current directory before doing the forking and chrooting. Since the file descriptor is not used in the forked process, we close it. Fixes FS#36161. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 759c48cd34235a0650f8189223b812dad53780cc Author: Allan McRae <allan(a)archlinux.org> Date: Fri Nov 15 12:08:35 2013 +1000 .gitignore: Add intl directory. Signed-off-by: Allan McRae <allan(a)archlinux.org> commit 1340ed792501c3a2c65b83ff90b232e6e01f5537 Author: Allan McRae <allan(a)archlinux.org> Date: Fri Nov 15 11:59:18 2013 +1000 paccache: clarify help output The help output referred to variables in the script rather than what they were labelled in the parameter argument. Signed-off-by: Allan McRae <allan(a)archlinux.org> ----------------------------------------------------------------------- Summary of changes: .gitignore | 1 + contrib/bacman.sh.in | 4 ++-- contrib/paccache.sh.in | 10 +++++----- contrib/pacdiff.sh.in | 13 +++++++------ contrib/pacscripts.sh.in | 4 ++-- contrib/rankmirrors.sh.in | 41 +++++++++++++++++++++++++++++------------ doc/pacman.conf.5.txt | 9 +++++---- lib/libalpm/db.c | 13 +++++++++++++ lib/libalpm/sync.c | 6 +++++- lib/libalpm/util.c | 1 + scripts/makepkg.sh.in | 6 +++--- src/pacman/util.c | 6 ++++-- test/pacman/README | 5 ++--- 13 files changed, 79 insertions(+), 40 deletions(-) hooks/post-receive -- The official pacman repository

1 0

[pacman-dev] Test suite failures using fakechroot-2.17.1
by Allan McRae 14 Dec '13

14 Dec '13

Using fakechroot-2.17.1 from the Arch [testing] repo, I get 4 pactest failures. All appear to be due to install scriptlets failing. The ldconfig tests work so the _alpm_run_chroot commands work. I think it has something to do with symlinks and the shell... The guilty upstream commit is 52502dbbb - "It is safe to use relative paths which won't escape from fake chroot". Can someone help figuring out whether this is a pacman testsuite issue or a fakechroot issue? Allan

2 3

[pacman-dev] More flexibility on *-vcs building
by 郑文辉(Techlive Zheng) 13 Dec '13

13 Dec '13

Now, we can specify 'fragment' parameter to the source url of a VCS package to build a specific version, I wonder if it is possible to make this parameter also accept environment variables, and let `makepkg` handles a depends array elememnt in PKGBUILD like `*-vcs=SPECVIED-vesion` automatically. This way, we can build the a static version of a package but use the source fetched from git.

2 1

[pacman-dev] [PATCH] Fix consistency of downloaded sources permissions
by Sébastien Luttringer 13 Dec '13

13 Dec '13

Calls to makepg -g and makepkg to download a source files, result in different permissions on the file if user umask != 0022. Run makepkg with -g option, will download source files before makepkg set umask to 0022. Downloaded files permissions will depend on user umask. Run bare makepkg, will call download source routines after umask was set to 0022. Downloaded files permissions will be group and world readable. A collateral damage: When a user who has a restricted umask (like 077), update a PKGBUKLD with updpkgsums (which call makepkg -g), and call a devtools scripts (extra-i686-build) to build the package, he will get a 'Permission denied', because the builder (another makepkg call) will be run as nobody user. Another side effect, when several users share a SRCDEST directory, they cannot access to files generated by another user with restricted umask. Altough, this can be workarounded by default ACL in the SRCDEST directory. The oldest commit with this umask is the first git commit (d04baab) with no revelent informations provided about the purpose of using an umask. The last commit moving the umask was 171808. The precious commit message refered to bug FS#9242 and FS#9362 which recommend to put umask at the top in makepkg. This patch put the umask definition at the "beginning" of the makepkg script. This let us rely that all files generated by makepkg will be with 0022 umask. Signed-off-by: Sébastien Luttringer <seblu(a)seblu.net> --- scripts/makepkg.sh.in | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/scripts/makepkg.sh.in b/scripts/makepkg.sh.in index 28e8e7a..81354df 100644 --- a/scripts/makepkg.sh.in +++ b/scripts/makepkg.sh.in @@ -2544,6 +2544,9 @@ There is NO WARRANTY, to the extent permitted by law.\n")" # PROGRAM START +# ensure we have a sane umask set +umask 0022 + # determine whether we have gettext; make it a no-op if we do not if ! type -p gettext >/dev/null; then gettext() { @@ -2979,9 +2982,6 @@ else fi fi -# ensure we have a sane umask set -umask 0022 - # get back to our src directory so we can begin with sources mkdir -p "$srcdir" chmod a-s "$srcdir" -- 1.8.4.2

4 5

[pacman-dev] [PATCH] Fix pactest README
by Allan McRae 13 Dec '13

13 Dec '13

The --tests option no longer exists. Signed-off-by: Allan McRae <allan(a)archlinux.org> --- test/pacman/README | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-) diff --git a/test/pacman/README b/test/pacman/README index 958ff28..c7aeb10 100644 --- a/test/pacman/README +++ b/test/pacman/README @@ -52,13 +52,12 @@ Upon completion, it checks that: Usage ===== -pactest will run the suite of tests defined by the "--test" parameter. +pactest will run the suite of test files passed as arguments. Example: - ./pactest.py --test tests/*.py + ./pactest.py tests/*.py This example will run all tests from the "tests" directory. -Note: several "--test" options can be passed to pactest. Use the "help" option to get the full list of parameters: ./pactest.py --help -- 1.8.5.1

1 0

[pacman-dev] Bug in libalpm: sizeof(off_t)
by Jeremy Heiner 11 Dec '13

11 Dec '13

An app developer can sometimes choose how big they want off_t to be (by #defining _FILE_OFFSET_BITS). The default for i686 is 32 bits (a bit odd, that). Pacman's configure attempts to choose 64 bits, and succeeds in doing so for the Arch builds (but doesn't force that choice). Bad things happen when the app developer chooses differently than configure did, but not until runtime. For example, the app's .o files will think sizeof(alpm_file_t) == 12, whereas libalpm thinks it == 16. The compile and link phases don't catch this. One possible solution might be to "mv alpm.h alpm.h.in", and add something like: #if sizeof(off_t) != @???@ #error ??? #endif Not sure yet what to put in those "???" spots, but the basic idea is to prevent the compiler from generating any .o files that disagree with the ABI of /usr/lib/libalpm. Hope that all made sense, and that I'm not duplicating - I searched the mailing list and the bug database and found zero hits. Anyone see a better solution? Jeremy

4 22

[pacman-dev] [PATCH] remove useless continue
by Andrew Gregory 11 Dec '13

11 Dec '13

Signed-off-by: Andrew Gregory <andrew.gregory.8(a)gmail.com> --- lib/libalpm/sync.c | 1 - 1 file changed, 1 deletion(-) diff --git a/lib/libalpm/sync.c b/lib/libalpm/sync.c index 551f926..7dbb8a0 100644 --- a/lib/libalpm/sync.c +++ b/lib/libalpm/sync.c @@ -536,7 +536,6 @@ int _alpm_sync_prepare(alpm_handle_t *handle, alpm_list_t **data) trans->add = alpm_list_remove(trans->add, rsync, _alpm_pkg_cmp, NULL); /* rsync is not a transaction target anymore */ trans->unresolvable = alpm_list_add(trans->unresolvable, rsync); - continue; } alpm_list_free_inner(deps, (alpm_list_fn_free)_alpm_conflict_free); -- 1.8.5.1

1 0