Age | Commit message (Collapse) | Author |
|
This patch sources a PKGBUILD in a subshells instead of directly.
This way we don't polute our scope or overwrite our vars which might lead to unexpected behavior.
|
|
db-functions now sets an individual $WORKDIR and implements trap functinos that
remove locks on exit or error. There are new functions to lock and unlock the running script.
misc-scripts/ftpdir-cleanup was renamed to ftpdir-cleanup-repo as the cron-job had the same name.
Script names have to be unique when using db-functions.
|
|
|
|
|
|
There is no need to have a variable for things like PKGBUILD
that are very unlikely to ever change.
|
|
devtools take care of this anyway
|
|
Only inform of errors and processes that might take longer.
|
|
|
|
|
|
|
|
The package pool dir depends on the host; so we solve this similar to
get_repos_for_host()
Note: There is no "os" subdir for the package pool.
|
|
This is a relative var, descendent from FTP_BASE. We do this because
we also need a relative path to the package pool dir
|
|
This should save lots of transfer time with the mirrors :)
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
|
|
Signed-off-by: Pierre Schmitz <pierre@archlinux.de>
|
|
|
|
With this patch packages with different compressions are accepted. It is
ensured that one cannot have the same package with different compression
extensions.
The new functions getpkgfile{,s} are used to sanitize globed filenames.
Signed-off-by: Pierre Schmitz <pierre@archlinux.de>
|
|
Add 'any' to the items looped over in the ${ARCHES[@]} loop
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Dan McGee <dan@archlinux.org>
|
|
We can't really do the basedir magic from db-functions as it is just being
sourced and is not the currently executing script. Although a bit
repetitive, it is a lot safer to just include the config file everywhere.
Noticed this when trying to run the latest available scripts on sigurd.
Signed-off-by: Dan McGee <dan@archlinux.org>
|
|
Removed the need for makepkg.conf
* Killed off CARCH
* Added PKGEXT to the config file
* Remove source_makepkg function
* Source config file where makepkg.conf was needed
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
DBEXT contains the full extension for the db files (.db.tar.gz)
and is recorded in the config file
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Dan McGee <dan@archlinux.org>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
[Aaron: keep WORKDIR creation outside the loop, as its
removed at the end]
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Because we lock/unlock in a loop, simply remove the cleanup trap
at the end of the process, and manually clean up the work dir
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
* also added error msg for repo_lock in db-functions
[Aaron: keep quotes around $WORKDIR]
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
[Aaron: Don't remove quotes around $WORKDIR]
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
* When calling cleanup at the end of db-update, the function
repo_unlock is called, but the repo-locks have already been unlocked!
So we get a spurious error message. This commit replaces "cleanup" by
"rm -rf $WORKDIR".
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Original-work-by: François Charette <francois@archlinux.org>
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
repo-add handles this on its own
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Also fixes the bug in the previous commit, in which the
arch-independent packages would be deleted from the staging
directory after only one iteration.
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
This could be useful if we move community to a separate
SVN repo
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
ARCHES is defined in config. db-update should now loop
over all arches on each run and update whatever is in the
staging dir.
Remove the db-*64 scripts due to this
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Conflicts:
db-update
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
This prevents some errors where packages where removed
from the staging dir when they shouldn't have been.
Original-idea-by: Pierre Schmitz <pierre@archlinux.org>
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|
|
This assumes the -q flag patch is accepted for pacman 8)
Signed-off-by: Aaron Griffin <aaronmgriffin@gmail.com>
|