summaryrefslogtreecommitdiffstats
path: root/src/pyfedpkg/__init__.py
Commit message (Collapse)AuthorAgeFilesLines
* Add "fedpkg clone --initial-merge" argumentinitial-merge-command/2Hans Ulrich Niedermann2010-08-101-2/+5
|
* Add "fedpkg initial-merge" commandHans Ulrich Niedermann2010-08-091-0/+2
| | | | | | | | | | | | | | | | | | | | | | | usage: fedpkg.py initial-merge [-h] [-n] [repo-path [repo-path ...]] Performs a 'git merge' of all git branches with the same content (i.e. with the same package spec files, patch files, etc.), regardless of their history. This is useful after Fedora's dist-cvs to dist-git migration, as often different branches have different histories but the same content on the filesystem. After these initial merges of identical trees, future merges between the branches will be a lot easier: Easier to follow in the dependency graph, and easier to perform without conflicts. positional arguments: repo-path Path to a repo to initial-merge optional arguments: -h, --help show this help message and exit -n, --dry-run Whether to run without actually merging
* Remove function call results as default param valuesHans Ulrich Niedermann2010-08-081-7/+22
| | | | | | | | | The function call is evaluated once when the module code is read. However, the os.getcwd() function might give different results at run time than at module reading time. Replaced those default param values with None, and let the functions call os.getcwd() if the param happens to be None.
* Fix rawhide branch detection (define .distshort)Hans Ulrich Niedermann2010-08-051-0/+1
| | | | | | Fix rawhide branch detection (define .distshort) Also define PackageModule.distshort for rawhide branches.
* Enable branch switching without any rpm contentJesse Keating2010-08-021-51/+60
| | | | | Without rpm content, we can't init the packagemodule, so don't try to do that when switching branches.
* Don't clog the first line of a changelogJesse Keating2010-08-021-4/+3
| | | | It's redundant.
* Also log on chain buildingJesse Keating2010-08-021-0/+3
|
* Add a nvr property to the class and use itJesse Keating2010-08-021-0/+2
| | | | | We could probably use this in more places, but this is a good start. Ticket #30
* set the loacl rpms macros to match whats set in olpc-release,Dennis Gilmore2010-08-011-2/+5
| | | | fedora-release and redhat-release. fixes a bug on local rpm creation
* Remove unused importJesse Keating2010-07-311-1/+0
|
* Fix a hashtype bugJesse Keating2010-07-311-1/+1
|
* Fix chainbuild to work for rawhide at leastJesse Keating2010-07-311-2/+4
| | | | Need more logic here to work with other chainable targets like epel.
* Make fedpkg clone -b f?? package workJesse Keating2010-07-291-0/+6
| | | | Also works for el? and olpc?
* Try to use kitchen when available.Jesse Keating2010-07-291-1/+4
| | | | This should make things work on el4/5
* Move 'sources' to not be part of the classJesse Keating2010-07-291-51/+61
| | | | | This might work around a weird issue where the koji builders can't read the git config strings in the chroot
* Fix checking for unpushed changes on a branchJesse Keating2010-07-281-2/+4
|
* Always target dist-rawhide when building from masterJesse Keating2010-07-271-1/+1
|
* Allow anonymous checkouts with fedpkgJesse Keating2010-07-271-1/+5
| | | | Comes from https://fedorahosted.org/fedora-packager/ticket/25
* Fix cloning with branches for our new branch schemeJesse Keating2010-07-271-3/+4
|
* Remove the koji.stg hackJesse Keating2010-07-271-4/+0
|
* Update to production URLSJesse Keating2010-07-271-4/+4
|
* Impliment pushing. Thanks Jochen@herr-schmitt.deJesse Keating2010-07-261-0/+9
|
* Rework how we deal with branchesJesse Keating2010-07-231-22/+51
| | | | | | Create a new list_branches function to discover the local and remote branches. Allow creating a new branch if the requested local branch doesn't already exist.
* Imports go at the topJesse Keating2010-07-231-1/+2
|
* When SSL connection fails, print it on screen.Thomas Spura2010-07-231-2/+6
| | | | | | | E.g. when the handshake fails, because your certificate is too old, fedpkg will gracefully fail and not completely crash. Signed-off-by: Thomas Spura <tomspur@fedoraproject.org>
* Change how we determine branchesJesse Keating2010-07-221-17/+42
| | | | | | | | | | This requires that remote branches follow a naming scheme of <release>/master and <release>/topic where release is like "f13" or "epel6". If we don't find a f<something, epel<something>, or olpc<something>, we assume that it is either master, or a branch of master for rawhide, and treat it like rawhide. To determine rawhide target we look at all the f branches, find the latest one and add 1 to the number. Seems to work and will keep working until Fedora 99.
* Use log not printJesse Keating2010-07-151-1/+1
|
* Added new command 'switch-branch' for working with branches.Léon Keijser2010-07-151-0/+22
|
* Print out the user eqiv koji commandJesse Keating2010-07-141-0/+13
| | | | | Even though we don't use koji directly, it's helpful to users if we show them the command they would have to use to duplicate what we are doing.
* Log koji url and wrap the session init in a tryJesse Keating2010-07-141-1/+5
|
* Fix ver rel output when subpackages are involved.Jesse Keating2010-07-141-4/+12
| | | | The idea here came from Xavier Lamien. I just modified it slightly.
* Fix cloning with --branchesJesse Keating2010-07-131-2/+2
|
* Handle importing an srpm into an empty repoJesse Keating2010-07-081-3/+2
| | | | | | In the case of a fresh module created upstream, we don't have a spec file to use for information, so we have to make a git repo outside of the module.
* Don't write duplicate lines to sourcesJesse Keating2010-07-061-1/+4
| | | | | This can happen if somebody uses the "upload" target on the same file more than once.
* Remove incorrect commentJesse Keating2010-07-061-3/+0
|
* Rename source uploading functionJesse Keating2010-07-061-4/+14
| | | | | | This way we can reuse it for new-sources as well as just plain upload. The difference is new-sources replaces current content, upload just adds additional content.
* Use curl directly for file uploads.Jesse Keating2010-07-061-1/+12
| | | | | This is the easy way out until we have a progress bar support via the pycurl method.
* Pass the whole file path to the upload functionJesse Keating2010-07-061-6/+6
| | | | | | The otherside will reduce it down to basename when it stores the file. Frequently I upload source from a different directory (say an upstream clone and source build)
* Implement new-sources.Devan Goodwin2010-07-061-5/+182
| | | | | Introduces Lookaside object for interacting with the cache, as well as a GitIgnore object for managing the project's .gitignore file.
* Create a (simple) diff commandJesse Keating2010-07-011-0/+29
|
* Add filtering so we only check out branches we want with clone --branchesCasey Dahlin2010-06-251-2/+4
|
* Point branches to the correct upstream for clone --branchesCasey Dahlin2010-06-251-5/+21
| | | | Also get rid of fedpkg.git and improve some documentation
* Implement fedpkg.py clone --branchesCasey Dahlin2010-06-251-9/+39
|
* Add a commit functionJesse Keating2010-06-241-0/+35
| | | | | This should work well whether you're on a real tty or not, but I didn't test it for the "not a tty" aspect.
* Make use of our new function to run commands.Jesse Keating2010-06-231-78/+28
| | | | Hurray for code removal!
* Add some logging to our process calls.Jesse Keating2010-06-231-0/+5
|
* Add a function to run commandsJesse Keating2010-06-231-0/+56
| | | | | This saves some code duplication and allows us to actually stream output as it happens, when on a real tty. Woo!
* Move import_srpm out of the classJesse Keating2010-06-231-81/+79
| | | | | This is so that we can use it before we have a full fledged module to deal with, like creating something local to play with.
* Better handle imports when not all the files pre-exist.Jesse Keating2010-06-231-5/+15
| | | | | This will be useful when we start importing into fresh blank repos, but that still needs more work.
* Handle our common files not existing yet.Jesse Keating2010-06-181-2/+3
|