Tuesday, 2020-12-29

*** rfolco has quit IRC00:02
*** rfolco has joined #zuul00:03
*** tosky has quit IRC00:39
*** zenkuro has quit IRC00:45
*** holser has quit IRC01:02
*** holser has joined #zuul01:04
*** holser has quit IRC01:43
*** rfolco has quit IRC01:51
*** bhavikdbavishi has joined #zuul03:55
*** ykarel has joined #zuul04:21
*** bhavikdbavishi has quit IRC05:25
*** saneax has joined #zuul05:29
*** evrardjp has quit IRC05:33
*** evrardjp has joined #zuul05:33
*** saneax has quit IRC06:15
*** saneax has joined #zuul06:58
*** bhavikdbavishi has joined #zuul07:27
*** bhavikdbavishi1 has joined #zuul09:50
*** bhavikdbavishi has quit IRC09:52
*** bhavikdbavishi1 is now known as bhavikdbavishi09:52
*** hamalq has joined #zuul10:47
*** hamalq has quit IRC10:52
*** hamalq has joined #zuul10:55
*** hamalq has quit IRC11:00
*** bhavikdbavishi has quit IRC11:40
*** zenkuro has joined #zuul11:45
*** rfolco has joined #zuul11:57
*** bhavikdbavishi has joined #zuul12:01
*** tosky has joined #zuul12:05
*** bhavikdbavishi1 has joined #zuul12:50
*** bhavikdbavishi has quit IRC12:52
*** bhavikdbavishi1 is now known as bhavikdbavishi12:52
*** bhavikdbavishi has quit IRC13:24
*** holser has joined #zuul14:14
*** zenkuro has quit IRC15:07
*** zenkuro has joined #zuul15:07
*** sshnaidm|rover is now known as sshnaidm|afk15:34
*** noonedeadpunk has quit IRC15:48
*** holser has quit IRC15:48
*** noonedeadpunk has joined #zuul15:50
*** saneax has quit IRC16:30
*** bhavikdbavishi has joined #zuul16:35
*** noonedeadpunk has quit IRC16:40
*** noonedeadpunk_ has joined #zuul16:48
*** hamalq has joined #zuul17:01
*** ykarel has quit IRC17:24
*** hamalq has quit IRC18:04
*** bhavikdbavishi has quit IRC18:10
*** hamalq has joined #zuul18:16
*** hamalq has quit IRC18:20
*** hamalq has joined #zuul18:31
*** hamalq has quit IRC18:36
*** saneax has joined #zuul19:33
*** saneax has quit IRC19:39
*** masterpe has joined #zuul20:20
masterpeI'm using zuul with the gitlab driver, but currently it is only cloning the default branch but what is the best why to change to use the source_branch. I was not able to find the answer in the docu or the driver code?20:20
clarkbmasterpe: zuul should prepare all branches for a repo and then you can check them out in the jobs20:58
clarkbmasterpe: do you mean that a gitlab PR is made against branch foo but the driver is defaulting to using master/main in the job?20:59
clarkbbecause the default checkout on the job side should be the PR itself (then you can switch from there)20:59
masterpeYes20:59
clarkbhrm the default checkout should be for the change/PR merged against whatever its target is. Then on the test node that checked out state has the name of the target21:01
clarkbthere isn't a thing to force that beacuse it should just do that anyway21:01
clarkbreally quickly looking at the driver source it seems to set the event.branch values to the gitlab event's target_branch21:02
*** holser has joined #zuul21:02
clarkbwhich is how it would do what I describe21:02
masterpeThat is the target branch but how do I checkout the source branch?21:05
clarkbmasterpe: you mean the originating side of the PR? eg the branch on my fork?21:05
masterpeYes21:06
clarkbI don't think you can because zuul is only looking at the target side (the events originate there and zuul is not likely to be tied to your forks)21:06
clarkbwhat you can do is look at the delta between master and refs/origin/master since refs/origin/master is the pristine state and master is the pristine state + the PR21:07
clarkbthis might be a sitaution where zuul shows its original integration with gerrit too much (gerrit doesn't do forks so there really isn't a source "side" to consider). It might be useful to describe the use case here to either determine if a functional workaround is doable or if zuul needs to be extended somehow21:09
clarkbmasterpe: double checking the event ingestion code I don't think it records taht info from the event at all21:14
clarkbthe "change_url" is recorded though so possible the job could look it up via that url?21:14
*** holser has quit IRC21:16
*** holser has joined #zuul21:17
masterpeI only do a branch in that branch I change the code. After that I do a Merge request to the default branch.21:18
masterpeIn the webhook from Gitlab I also see the source_branch variable21:21
clarkbright I think zuul is interested in the merge request (because that reflects a potential future state). If you want to test branches before they get merge requested you may be able to use the gl_push trigger along with a ref regex to filter out branches to ignore?21:22
clarkbmasterpe: yes, I expect the event emitted includes the info but zuul's event ingestion doesn't seem to do anything with it that I can see21:23
clarkbI think the piece of info I'm missing is how is that source_branch information and content useful to jobs running under zuul. Zuul is buitl around examining proposed future states and testing them as if they had merged. Describing how the source side before constructing the proposed state is useful would probably help me21:26
*** aprice has quit IRC21:28
*** aprice has joined #zuul21:30
clarkbI have to pop out now but will check back in later to see if there is more info21:34
masterpeThanks for your help for now.21:37
masterpeI was thinking in using {{ zuul.projects['git.example.com/org/project'].checkout }} -> https://zuul-ci.org/docs/zuul/reference/jobs.html#var-zuul.projects.checkout to change the branch in a pre playbook.21:38
masterpeBut this still gives me the default branch and not the branch where I'm working in.21:41
*** rfolco has quit IRC21:43
*** rfolco has joined #zuul22:11
*** rfolco has quit IRC22:12
*** rfolco has joined #zuul22:13
*** rfolco has quit IRC22:18
*** holser has quit IRC23:09
*** holser has joined #zuul23:35
*** rfolco has joined #zuul23:44

Generated by irclog2html.py 2.17.2 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!