summaryrefslogtreecommitdiffstats
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* Not planning on porting check, export or unusedfedpatchesJesse Keating2010-02-061-16/+16
|
* Exit when we cannot find a build urlJesse Keating2010-02-051-0/+1
|
* Add some debugging outputJesse Keating2010-02-052-0/+4
|
* Fix up some loggingJesse Keating2010-02-051-4/+9
|
* Wire up chain-buildJesse Keating2010-02-051-4/+59
| | | | Some stuff stolen from /usr/bin/koji here, but it works.
* Handle chain buildsJesse Keating2010-02-051-3/+20
| | | | More stuff stolen from /usr/bin/koji here, but oh well.
* Add a function to find the latest remote commitJesse Keating2010-02-051-0/+18
| | | | | | | This is hardcoded to origin/master right now as chain-build is the only real consumer of this and chain-build only works on rawhide. Eventually this could probably take a branch name to get a commit for, or return all the heads or something.
* Handle the renamed moduleJesse Keating2010-02-051-37/+37
|
* Revert "Move fedpkg.py into a bin/ directory"Jesse Keating2010-02-051-0/+0
| | | | | | This reverts commit 6f09fff59e3d92fea106a79d3fba56c6f2be7115. Shouldn't be necessary to stash this in bin/ anymore.
* Rename the module 'pyfedpkg' to save headaches.Jesse Keating2010-02-051-0/+0
| | | | | Having the script and the module named the same was just causing too many weird python issues, so I gave up.
* Move fedpkg.py into a bin/ directoryJesse Keating2010-02-051-0/+0
| | | | | This helps eclipse not be confused by having fdpkg.py and fedpkg/__init__.py in the same directory. Silly eclipse.
* Handle scratch builds from srpmsJesse Keating2010-02-031-14/+86
| | | | | This drags in more of /usr/bin/koji but now handles passing an srpm into scratch-build and optionally tracking progress as it is uploaded.
* Handle scratch building with srpmsJesse Keating2010-02-031-56/+45
| | | | | | This cleans up from moving a chunk of the build function into it's own koji init function. It also adds a function to deal with uploading content to koji.
* Add a init_koji functionJesse Keating2010-02-031-0/+51
| | | | | I wanted to move the koji initialization out of the build command as we could use it in different functions.
* Wire up the unused_patches functionJesse Keating2010-02-021-2/+7
|
* Add an unused_patches functionJesse Keating2010-02-021-1/+22
|
* Wire up the build task the rest of the wayJesse Keating2010-02-011-2/+17
| | | | | This also makes use of the new cert library which needs some work still, so this code is somewhat fragile if you don't have proper certs.
* Add some functions watch koji tasksJesse Keating2010-02-011-0/+82
| | | | More code stolen from /usr/bin/koji
* Add a class to watch tasksJesse Keating2010-02-011-0/+94
| | | | | This is borrowed from /usr/bin/koji, so it can go away when koji grows a client library
* Finish the build functionJesse Keating2010-02-011-24/+75
| | | | | | | This borrows some from /usr/bin/koji but does it in a way that allows for clients to handle watching the task if they want or not. As /usr/bin/koji gets moved more toward a koji client library some of this code can go away.
* Use the new fedora_cert method to get a user nameJesse Keating2010-01-291-2/+8
|
* add extra macros to rpmbuild-md5 for newer rpm features in F-12Dennis Gilmore2010-01-291-1/+5
|
* clean up the old attempt at a fedora-cert libraryDennis Gilmore2010-01-281-104/+0
|
* make fedora-packager-setup use the new library so that you only need to run ↵Dennis Gilmore2010-01-281-5/+11
| | | | it once and it can get new certs
* make fedora-cert use the new libraryDennis Gilmore2010-01-281-94/+12
|
* make fedora_cert a library so we can import it in multiple placesDennis Gilmore2010-01-281-0/+104
|
* Wire up the build targetJesse Keating2010-01-071-2/+53
| | | | | This adds a few more functions necessary for sussing out secondary arch stuff, although see the comments as this may be changing
* Add a build functionJesse Keating2010-01-071-0/+56
| | | | | Currently this doesn't actually do anything, but the workings are there. Also make the git repo a object property.
* Define the target when creating the objectJesse Keating2010-01-071-0/+4
|
* Add a anon git url to be used by kojiJesse Keating2010-01-071-0/+1
|
* Change debug output a bitJesse Keating2010-01-061-1/+1
|
* Wire up mockbuildJesse Keating2010-01-061-2/+13
|
* Add a mockbuild functionJesse Keating2010-01-061-1/+39
|
* re-wire the clone commandJesse Keating2010-01-061-3/+3
|
* Finish up the clone functionJesse Keating2010-01-061-9/+23
| | | | | More use of the git module and it's lambda for running arbitrary git commands. There is no clone command within the git module.
* Wire up clogJesse Keating2010-01-061-2/+6
|
* Add a clog functionJesse Keating2010-01-061-0/+30
| | | | | | This could probably be done better, but I didn't want to just copy in the sed script, and my skills at taking sed and converting them into a python regex are not so hot. So if somebody cares, they can fix it.
* Wire up new.Jesse Keating2010-01-061-2/+5
|
* Fill in the 'new' targetJesse Keating2010-01-061-0/+15
| | | | | | | This is the first use of the 'git' library, and I'm not sure if I like it or not, but it does provide a wrapper to running various git tasks in something like a pythonic way. If nothing else it saves me from having to wrap everything myself.
* Handle return codes from fedpkgJesse Keating2010-01-051-7/+8
| | | | We now get a return code, rather than output
* Return is not a functionJesse Keating2010-01-051-1/+1
|
* Make sure we capture error output correctlyJesse Keating2010-01-051-19/+33
|
* Use logging to get output back to the callerJesse Keating2010-01-051-34/+40
| | | | | | Instead of returning output, just log and then return the returncode from the command being ran. This makes it a little nicer client side to deal with output.
* Use logging instead of prints or returning outputJesse Keating2010-01-051-7/+4
|
* Don't print, log instead.Jesse Keating2010-01-051-31/+31
|
* Use the logging system from fedpkgJesse Keating2010-01-051-3/+21
| | | | Adjust the options accordingly too
* Create a logging object that clients can useJesse Keating2010-01-051-0/+14
|
* Fix execution without --path being passedJesse Keating2010-01-052-2/+2
| | | | os.path.curdir() just returns "." and some things don't like that.
* wire up cleanJesse Keating2010-01-051-2/+14
|
* Add a clean functionJesse Keating2010-01-051-0/+31
|
generated by cgit (git 2.34.1) at 2024-05-12 11:22:27 +0000