summaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2010-11-23sourceballs: avoid failing twicePierre Schmitz1-0/+7
2010-11-23sourceballs: fix outputPierre Schmitz1-8/+3
2010-11-23sourceballs: be less verbosePierre Schmitz1-6/+21
2010-11-23Rewrite sourceballs to increase performance and reliabilityPierre Schmitz4-257/+101
* Decrease file stats as much as possible * Create a list of all packages and meta data only once * Create a list of available source packages only once * Create a list of expected packages only once * Combine all three scripts into one to share data and code * Use as much information from the db files as possible and avoid using svn * Avoid attempting to create the same source package twice Logic works as follows: 1) create a list of all packages 2) Check for each package if we need a src package and create one 3) During this process create a list of all src packages that should be there 4) Diff both lists for the cleanup
2010-11-22Attempt to fix integrity-checkPierre Schmitz1-5/+6
2010-11-22check_archlinux: fix for python2 usagePierre Schmitz3-2/+2
2010-11-22sourceballs: check if source package was really createdPierre Schmitz1-2/+10
2010-11-22Prefer any packagesPierre Schmitz1-3/+3
2010-11-22sourceballs-cleanup: respect .force and .skip filesPierre Schmitz1-2/+8
2010-11-22Add wrappers for repo-add and repo-removePierre Schmitz3-3/+82
2010-11-22sourceballs: add for split packagesPierre Schmitz1-0/+21
2010-11-22Set CARCH as needed by some PKGBUILDsPierre Schmitz1-0/+3
PKGBUILDs expect CARCH to be set. This is needed when sourcing them directly.
2010-11-21texlive sources are already on ftp.archlinux.orgPierre Schmitz1-0/+20
2010-11-21sourceballs: Of course sort does only operate on linesPierre Schmitz1-1/+1
2010-11-21Blacklist more duplicate data packagesPierre Schmitz1-0/+5
2010-11-21sourceballs.skip: ufoai-data has the same sources as ufoaiPierre Schmitz1-0/+1
2010-11-21sourceballs.skip: tremulous-data has the same sources as tremulousPierre Schmitz1-0/+1
2010-11-21sourceballs.skip: torcs-data has the same sources as torcsPierre Schmitz1-0/+1
2010-11-21sourceballs.skip: nexuiz-data has the same sources as nexuizPierre Schmitz1-0/+1
2010-11-21sourceballs: Support any packagesPierre Schmitz2-3/+23
2010-11-21sourceballs: Don't write any logPierre Schmitz6-44/+15
This is consistent with the other dbscripts. The output will be send to the mailinglist.
2010-11-21sourceballs: respect the dryrun settingPierre Schmitz1-4/+6
2010-11-21sourceballs: check if PKGBUILD existsPierre Schmitz2-4/+10
2010-11-21sourceballs: fix error handlingPierre Schmitz2-9/+32
2010-11-21When on sigurd the svn repo is actually called svn-packagesPierre Schmitz1-1/+1
2010-11-21sourceballs: add LGPL2.1Pierre Schmitz1-1/+1
2010-11-20sourceballs: Lock the repo while reading the db filePierre Schmitz1-3/+10
2010-11-20sourceballs: skip in a clean way if source package should not be createdPierre Schmitz1-1/+1
2010-11-20Simplify sourceballs creationPierre Schmitz6-56/+32
* Read package lists directly from DB file * Make SVNREPO configurable
2010-11-20Fix sourceballs cron jobPierre Schmitz8-29/+84
* add unit test for sourceballs and cleanup * introduce SRCPOOL and LOGDIR variables in config
2010-10-12Dont try to index empty reposPierre Schmitz1-2/+2
2010-10-11Fix missing output redirectionDan McGee1-1/+1
Signed-off-by: Dan McGee <dan@archlinux.org>
2010-09-17Move output redirectionDan McGee1-3/+3
Signed-off-by: Dan McGee <dan@archlinux.org>
2010-09-13Fix typosPierre Schmitz2-2/+2
2010-09-12Merge branch 'master' of gerolde.archlinux.org:/srv/projects/git/dbscriptsPierre Schmitz1-1/+1
2010-09-12_grep_pkginfo: use a more specific formatPierre Schmitz1-6/+6
2010-09-12Add multilib-testing to web update scriptDan McGee1-1/+1
Signed-off-by: Dan McGee <dan@archlinux.org>
2010-09-12check_pkgsvn: Don't assume the same PKGBUILD for all architecturesPierre Schmitz1-5/+5
2010-09-12check_pkgsvn: check if pkgname is in sync with svnPierre Schmitz1-12/+16
In addition to pkgver and pkgrel it is checked if pkgname is defined in the PKGBUILD.
2010-09-11Fix check_pkgreposPierre Schmitz1-6/+6
Don't use globing as this will match a package of the same version but another arch. We can be more specific here as there are no longer packages without $arch extension.
2010-09-11Use local config instead of guessing by hostnamePierre Schmitz9-63/+48
Using the hostname to decide which repos to use is not releiable and hard to test. Instead use config.local to configure these. config files for sigurd and gerolde were added which can be copied or symlinked to config.local on the specific host.
2010-09-11Fix check_repo_permissionPierre Schmitz1-6/+4
2010-09-11Remove obsolete cron jobPierre Schmitz1-41/+0
2010-09-11adding multilib-testing repositoryPierre Schmitz1-1/+1
2010-09-05Check if package exists in any other repository on update20100906Pierre Schmitz3-3/+71
This also checks if the sam package exists within the old package layout (without package pool)
2010-09-05Fix updating of same package into different repositories at aoncePierre Schmitz2-4/+24
See FS#20745
2010-09-02Set correct group after touching the db filePierre Schmitz4-0/+19
* When writing the db file ensure that it has write permission of the group which owns the parent directory. * This should make the adjust-permissions cron job obsolete.
2010-09-02Simplify case statementDan McGee1-18/+5
Don't duplicate the user:group pair all over the place; put all of the associated repos and directories in the same case statement for a given set of credentials. Signed-off-by: Dan McGee <dan@archlinux.org> Signed-off-by: Pierre Schmitz <pierre@archlinux.de>
2010-09-01Simplify repo configurationPierre Schmitz7-14/+36
* Repositories can now be defined in the config file for each host * added community-staging, gnome-unstable and kde-unstable * Exception is the adjust-permission cron-job; but we might want to use acls in future anyway Signed-off-by: Pierre Schmitz <pierre@archlinux.de>
2010-08-25update-web-db: flip the arch/repo loopsDan McGee1-2/+2
This will really make sure [core] packages flow to the top of the list, regardless of architecture. They should anyway, but anything that was only updated in the first architecture updated could get artificially pushed down before. Signed-off-by: Dan McGee <dan@archlinux.org>