Its implementation was quite broken:
* add_loadtarget() might have silently filtered out some targets when
replacing an older version.
* This was used in sync.c to determine whether a target is implicit or not,
which is incorrect behavior. Before this patch we silently removed user
confirmed replacements; now we always warn on a replacement.
* remove001.py behavior was quite odd in adding same target 5 times to the
target list, we can change this behavior to be a failure.
Signed-off-by: Nagy Gabor <ngaba@bibl.u-szeged.hu>
[Xav: changed remove001 pactest accordingly]
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
[Dan: rewrote commit message]
Signed-off-by: Dan McGee <dan@archlinux.org>
This hasn't been updated in forever, and a simple python program could
regenerate it anyway. Or even grep.
Signed-off-by: Dan McGee <dan@archlinux.org>
This is in anticipation of removing the -A/--add option from the pacman
frontend. I've went through each of the pacman pactests that used the -A
operation and decided whether they were worth keeping, whether there was
already an upgrade test doing the same thing, or whether it should be moved
over.
The GIT rename log should make several of the moves obvious, but for those
that were deleted:
add001: handled by upgrade004
add002: worthless
add004: worthless
add010: handled by upgrade011
add050: handled by upgrade010
Signed-off-by: Dan McGee <dan@archlinux.org>
This is the first step of fixing FS#9547. This should not break any existing
code that may rely on this function behaving the way it did, and should be
good for inclusion in a maint release.
In addition, update pactest so it fills the FILENAME field in the DB entries
it creates so we can move forward with a real fix to this issue.
Signed-off-by: Dan McGee <dan@archlinux.org>
If the end of the pB list was reached before the end of pA, we failed to
read any remaining files from the pA list. Add an additional loop to ensure
all entries of pA are added to the return list regardless of whether we have
reached the end of pB.
This new loop also eliminates the now-unnecessary check for a null pB, as we
need to ensure we are excluding directories in the resulting output anyway.
Signed-off-by: Dan McGee <dan@archlinux.org>
This patch fixes upgrade040.py and upgrade041.py (041 now fails!):
* the old pactests didn't check the existence of the relocated file
* upgrade041.py was broken due to a typo (missing comma)
New upgrade046.py pactest was added (derived from the fixed upgrade041.py).
This fails because the file relocation check is _hacked_ to conflict.c, and
_alpm_db_find_fileconflicts is not called in case of --force.
Signed-off-by: Nagy Gabor <ngaba@bibl.u-szeged.hu>
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
Add a bunch of guards around function calls like open() and stat() to ensure
we are not going to get ourselves a python error. This made implementing and
testing the new upgrade045 pactest much easier, as its whole purpose was to
create a dead symlink and debug a segfault of pacman (which caused no DB
entries to be written) to support the previously checked in fix for FS#9235
(commit 0c2206f542). Both of these cases are
now non-fatal in pactest.
Signed-off-by: Dan McGee <dan@archlinux.org>
With --unneeded option 'pacman -R' doesn't stop in case of dependency error;
it removes the needed-dependency targets from the target-list instead. See
also: http://archlinux.org/pipermail/pacman-dev/2007-October/009653.html .
The patch also adds a new causingpkg field to pmdepmissing_t which indicates
the to-be-removed package which would cause a dependency break. This is
needed, because miss->depend.name may be a provision. miss->causingpkg will
be useful in -R dependency error messages too.
[Xavier: renamed inducer to causingpkg, removed the _alpm_pkgname_pkg_cmp
helper function as requested by Aaron. This might be added by a further
commit. Other small cleanups, updated manpage and bash completion.]
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
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.htmlhttp://www.archlinux.org/pipermail/pacman-dev/2007-December/010393.html
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
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@archlinux.org>
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@bibl.u-szeged.hu>
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
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@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
Change the 'provname provver' format to 'provname=provver'.
In .PKGINFO, the provisions are copied from the PKGBUILD without quotes. So
the provision version was actually handled as a different provision...
See FS#9171.
Dan: Unfortunately we have to change our original specification for
versioned provisions with this patch, but it ends up being the simpler and
cleaner solution in the long run, and if there is any time to change it the
time is now before many packages have been built. Keeping the ' ' based
format would have required us to do special parsing in repo-add, as well as
being susceptible to users not using quotes in their provides array.
Hopefully this will resolve the issues we had with our initial plan. Sorry
for the confusion.
Acked-by: Nagy Gabor <ngaba@bibl.u-szeged.hu>
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
Also add 3 new pactests to cover both the -U and -S operations for versioned
provisions.
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
This pactest demonstrates that we should copy the reason between the
to-be-replaced and replaced packages
Signed-off-by: Nagy Gabor <ngaba@bibl.u-szeged.hu>
Signed-off-by: Dan McGee <dan@archlinux.org>
This reverts commit e28973169d.
This code might fit better in the frontend than in the backend finally.
Ref: http://www.archlinux.org/pipermail/pacman-dev/2007-November/010150.html
I also changed it for fixing FS#8763 :
if there is exactly one provider, pacman will pull it and print a warning.
if there are several providers, pacman will list them and fail. It's up to
the user to pick one. Add sync501 pactest to reflect that.
The previous fileconflict check (package vs filesystem) skipped the conflict
when the file on the filesystem was a directory or a symlink to a directory,
no matter what the file in the package was.
Now, the conflict will only be skipped if the file in the package is a
directory (so compatible with a dir or a dir symlink on the filesystem).
So in the case of 8156 (new fileconflict003 pactest for this case), instead
of silently ignoring the extraction of the test symlink, pacman will now
fail because of a file conflict between the test symlink in the pkg2 package
and the test directory on the filesystem.
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
Trying to make a symlink at the top level previously made pactest fail.
For example : "test -> test2/" as a file in a package.
The path to the test symlink was empty in this case, but the python code
still tried to chdir in "", which failed.
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
Originally noticed in FS#9024, but was fixed in previous changes anyway.
However, it doesn't hurt to still check it.
Also add a pactest from Chantry Xavier for the original problem to ensure
we can't reproduce it.
Signed-off-by: Dan McGee <dan@archlinux.org>
Thanks to the proactive backup handling, we don't need to add the moving
file to the skip_add list.
The backup handling will make sure nothing gets overwritten.
Ref: http://www.archlinux.org/pipermail/pacman-dev/2007-December/010610.html
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
This was the case of the bash packaging error where a file was removed from
the package but not the backup array.
I just added a sanity check so that only the files from the backup array
that are also in the filelist are used.
I had to edit upgrade026 pactest slightly : it required the file to be
copied to .pacsave instead of moved. But just moving it should be enough, as
we agreed on the ML :
http://www.archlinux.org/pipermail/pacman-dev/2007-December/010440.html
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
This adds a pactest for the relocation of a config file between two packages
(case of etc/profile moving from bash to filesystem).
While running this pactest, I found out that chk_filedifference didn't work
correctly with an empty list as second argument. So that's fixed now.
Ref: http://www.archlinux.org/pipermail/pacman-dev/2007-December/010610.html
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
This is something pacman can do on its own straight from the archive, and we
will reduce the chance of problems occurring becuase of inproper FILELIST
generation as we have had in the past with special characters in filenames.
Once we remove it from makepkg. we can remove any usage of it from all of
our other tools, including pacman, pactest, and contrib/ utilities.
Note that removing it from pacman uncovered a few other bugs anyway, so this
was probably a good move.
Signed-off-by: Dan McGee <dan@archlinux.org>
This is the bash case when the /etc/profile file was removed by error from
the package, but stayed in the backup array.
Ref:
http://www.archlinux.org/pipermail/arch-dev-public/2007-December/003556.html
Also fixed a little typo in add.c, but it's disabled code.
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
query002 and sync1100 had PACMAN_OUTPUT rules that looked at the
build/install date (localized).
Instead of looking at the month name, it will now check the year, which
should be safer.
I also had to add another pactest (query005) for keeping the same coverage.
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
You can use foo<2.0 and foo>2.0 as depend
add046.py and add047.py pactests were added to check this
Signed-off-by: Nagy Gabor <ngaba@bibl.u-szeged.hu>
Signed-off-by: Dan McGee <dan@archlinux.org>
Due to commit da1222de2e30aabcae9d17bbfa10bbf0672338af, we can now use
fakechroot to completely run scriptlet pactests
Use "which" functionality so as not to REQUIRE it for all users
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
Check for existence of the logfile first, and output an error if not found
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
Update the GPL boilerplate to direct people to the GNU website for a copy of
the license, as well as bump all of Judd's copyrights to 2007.
Signed-off-by: Dan McGee <dan@archlinux.org>
I started playing around with gcov today and it showed a few places in the
code that we don't test at all. This is the start of ensuring that we
execute most of the code in our codebase.
Signed-off-by: Dan McGee <dan@archlinux.org>
This pactest checks what happens if a package exists in two sync repos.
Signed-off-by: Nagy Gabor <ngaba@bibl.u-szeged.hu>
[Dan: added sync1006, same test with different versions]
Signed-off-by: Dan McGee <dan@archlinux.org>
checkdeps and resolvedeps now take both a remove list and an install list as
arguments, allowing dependencies to be calculated correctly.
This broke the sync990 pactest, but this pactest used dependencies and
provides in an unusual way, so it has been changed.
Dan: the sync990 pactest was just plain wrong. It didn't satisfy the
dependencies correctly, so should never have succeeded.
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
[Dan: some variable renaming, clarification in commit message]
Signed-off-by: Dan McGee <dan@archlinux.org>
This operation made sense in the days before sync DBs existed, but it no
longer has the same usefulness it once did.
Signed-off-by: Dan McGee <dan@archlinux.org>
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
Signed-off-by: Nagy Gabor <ngaba@bibl.u-szeged.hu>
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
This patch introduces versioned provisions in "provision 1.0-1" format.
_alpm_db_whatprovides was modified accordingly (added sync500.py),
alpm_depcmp was modified accordingly (add043.py passes now; added add044.py
and add045.py).
Notes:
alpm_db_search now uses the whole versioned %PROVIDES% string in its search.
debug logging was simplified in alpm_depcmp.
Signed-off-by: Nagy Gabor <ngaba@bibl.u-szeged.hu>
[Xavier: fixed a few typos, duplicate const strings with strdup before
modifying them, put some debugging back in alpm_depcmp, minor code cleanups
(var/function renaming), added a note in PKGBUILD man page.]
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
[Dan: made strcmp checks clearer, added a comment]
Signed-off-by: Dan McGee <dan@archlinux.org>
Remove any checks dealing with requiredby from pactest (but not actually
from the pactests themselves). Of course, we should probably find a new way
to check requiredby values of packages since there is no guarantee our code
is working perfectly.
Signed-off-by: Dan McGee <dan@archlinux.org>
The pactest demonstrates what happens if fileconflict was found after the
removal part of a sync transaction.
Signed-off-by: Nagy Gabor <ngaba@bibl.u-szeged.hu>
Signed-off-by: Dan McGee <dan@archlinux.org>
Currently alpm_depcmp uses pkg->version as a version number for provisions,
which is odd. The failure of the pactest demonstrates this.
Signed-off-by: Nagy Gabor <ngaba@bibl.u-szeged.hu>
Signed-off-by: Dan McGee <dan@archlinux.org>
Also tell pactest to reset IgnoreGroup like it does for IgnorePkg.
Signed-off-by: Nathan Jones <nathanj@insightbb.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
One currently should succeed (006), and 005 fails.
requiredby005.py is originally from Nagy Gabor <ngaba@petra.hos.u-szeged.hu>.
Signed-off-by: Dan McGee <dan@archlinux.org>
Actually use python commands to generate the filelist instead of os.system
calls that don't necessarily work everywhere. Noticed when running "make
check" on FreeBSD where the tar program is actually bsdtar.
Signed-off-by: Dan McGee <dan@archlinux.org>
As seen with the recent upgrade of pacman and the removal of the
pacman.d/current mirrorlist, files that were formerly in the backup array
get deleted upon their removal, which could be dangerous. Instead, we should
use the combined backup array of the old and new package. This fix should
address this issue in a relatively straightforward way.
In addition, old files should be moved to pacsave locations as expected.
Signed-off-by: Dan McGee <dan@archlinux.org>
That is the problem mentioned by Nagy there (with suggestions for fixing it) :
http://www.archlinux.org/pipermail/pacman-dev/2007-August/009082.html
If a dependency conflicts with a local package and has to replace it,
the PM_SYNC_TYPE_DEPEND information is lost, and the resulting install
reason is wrong (the package is marked as explictly installed).
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
remove048 is the case mentioned there (fails in 3.0 but works in 3.1) :
http://www.archlinux.org/pipermail/pacman-dev/2007-September/009294.html
It's the same as remove046 with -R instead of -Rc.
sync060 is a case reported this morning on IRC :
a pacman -Su wanted to replace gensplashutils by gensplash,
but pacman said gensplashutils was required by initscripts-gensplash,
while initscripts-gensplash was not even installed.
This is also fixed in the current 3.1 code though.
upgrade02{4,5} are the backup handling problem I described there :
http://www.archlinux.org/pipermail/pacman-dev/2007-September/009376.html
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
The sync package pkg3 depends on the sync package pkg2, but pkg1 replaces pkg2.
So pkg2 is added to the remove list.
When checkdeps checks if pkg2 can be removed, it looks at the packages that require pkg2
(with the RequiredBy field of pkg2), but this doesn't contain pkg3 at this point.
RequiredBy fields are only updated after the packages are installed.
Proposed solution : checkdeps should know both the remove and upgrade list to decide properly.
Reference:
http://www.archlinux.org/pipermail/pacman-dev/2007-July/008972.html
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
This patch cleans up and fix the _alpm_trans_update_depends function
and ensure that all requiredby fields are filled in case of multiple satisfiers
(previously, the handling of mutliple satisfiers in that function was inconsistent).
This makes a special case handling of requiredby in commit_single_pkg() obsolete,
and so allows cleaning that code as well.
Also fixed upgrade056 pactest because :
1) the requiredby fields were wrong, and this wouldn't happen with the fixed _alpm_trans_update_depends().
2) this is a very unusual case anyway (and handling all corner cases combined to a broken database
seems nearly impossible to achieve).
References :
http://www.archlinux.org/pipermail/pacman-dev/2007-July/008919.htmlhttp://www.archlinux.org/pipermail/pacman-dev/2007-July/008920.html
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
A side effect of the previous commit ( ea9a756eea )
is that it's now possible to use versioned conflicts.
Add two new conflict pactests for showing it.
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
.gitignore works recursively, so we don't need Makefile and Makefile.in
in all of the subdirectory .gitignore files.
Signed-off-by: Dan McGee <dan@archlinux.org>
This test installs three packages with a circular dependency,
to check everything still goes fine in that case.
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
This replaces the former -D operation that was undocumented and rather
hacky. It can be used with add, upgrade, or sync transactions and will affect
all packages installed. Should close FS #7193.
Also tell makepkg to use this new flag.
Signed-off-by: Dan McGee <dan@archlinux.org>
sync022 was added here:
39b6549655
This pactest reverses the installed package to see if it is correctly
picked, in order to test some further changes to this depcheck code that
currently makes sync022 fail.
Signed-off-by: Dan McGee <dan@archlinux.org>
This indicates a problem in the current removedeps function
(no version checking) for which Nagy proposed a fix here :
http://www.archlinux.org/pipermail/pacman-dev/2007-June/008539.html
But as he mentioned there, a better fix might be to use the checkdeps
function in the removedeps one.
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
pactest was sleeping for 3 seconds, which is unnecessary. Reduce it to 1.5
seconds and clean up the code a bit anyway.
Signed-off-by: Dan McGee <dan@archlinux.org>
Add the ability to check the permissions and type of a file within the
framework of pactest. Two new rules can be used:
self.addrule("FILE_TYPE=bin/foo|file")
self.addrule("FILE_MODE=bin/bar|644")
TODO: add the ability to add different types of files (eg links) via the test
package building framework, and add the ability to change the modes on files.
Signed-off-by: Dan McGee <dan@archlinux.org>
I previously introduced some patches to make just about every path in
pacman/libalpm configurable; doing this with the lockfile seemed a bit too
far and we really should just place the lockfile where it belongs- with the
DB that needs locking.
More details in this thread:
http://archlinux.org/pipermail/pacman-dev/2007-June/008499.html
Signed-off-by: Dan McGee <dan@archlinux.org>
As described in the bug report, installing taglib-rcc (sync900 pactest)
works fine. But installing taglib back (sync901 pactest) doesn't.
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
Same as remove040, with cascaded remove instead of simple remove.
Signed-off-by: Chantry Xavier <shiningxc@gmail.com>
Signed-off-by: Dan McGee <dan@archlinux.org>
* Add vim modeline to Makefile.am and configure.ac
* Fix white space in Makefile.am and configure.ac
* Add contrib/wget-xdelta.sh to EXTRA_DIST in Makefile.am
Signed-off-by: Andrew Fyfe <andrew@neptune-one.net>
Signed-off-by: Dan McGee <dan@archlinux.org>
Most of these new tests deal with provisions upon sync or upgrade being
changed by packages, and pacman not being smart enough to deal with
it yet.
Signed-off-by: Dan McGee <dan@archlinux.org>
This change allows us to use all autoconf specified paths, most notably
$(localstatedir). It is quite a change and touches a lot of files, as
all references to the DB and cache were done with the ROOTDIR as a prefix.
* add --lock command-line option to pacman to specify the location of the
lockfile (this can now be specified at configure time by setting the
$localstatedir path).
* Rip quite a few settings out of configure.ac as they are now picked by
setting the paths during configure or make.
* Fix bug with /tmp fallback for sync downloads not working correctly
(related to root location, now the system tmp dir is used).
* Simplified the parameters to some libalpm functions, and added get/set
for the new lockfile option.
* Renamed several of the DEFS to names without the PM_ prefix.
Signed-off-by: Dan McGee <dan@archlinux.org>
These should test various aspects of the DB when dealing with requiredby
entries. At least one fails without Nagy's alpm_depcmp patch.
Signed-off-by: Dan McGee <dan@archlinux.org>
* Move all .cvsignore files to .gitignore for switch in VCS. In addition,
delete ones that were unnecessary because they only contained Makefile
and Makefile.am.
Signed-off-by: Dan McGee <dan@archlinux.org>
reordering and adding ones that were forgotten (noticed when trying to
compile after reordering).
* Updated the HACKING file to include information on #include usage.
* print -> vprint in "making dir" function in pactest.
