IRC logs for #buildstream for Thursday, 2019-10-31

« Wednesday, 2019-10-30 Index Friday, 2019-11-01 »
*** narispo has quit IRC01:11
*** narispo has joined #buildstream01:12
*** ikerperez has joined #buildstream07:58
*** dtf has joined #buildstream09:04
*** phildawson_ has joined #buildstream09:13
*** dtf has joined #buildstream09:15
*** rdale has joined #buildstream09:24
*** tiagogomes has joined #buildstream09:41
*** santi has joined #buildstream09:54
*** phil has joined #buildstream10:01
*** phildawson_ has quit IRC10:02
*** jonathanmaw has joined #buildstream10:12
gitlab-br-botmarge-bot123 merged MR !1676 (aevri/enable_spawn_ci_5a->master: spawn tests: enable most non-integration) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/167610:25
*** lachlan has joined #buildstream10:30
*** lachlan has quit IRC10:31
*** lachlan has joined #buildstream10:31
tpollardIf I'm adding a new dependency in an MR, should I run the Make updater which will add in other dependency updates, or just add my dependency manually to the req list?10:42
tlater[m]tpollard: Remove the irrelevant dependency updates from the diff, but still use the updater, I'd say10:50
* tlater[m] thinks we haven't cared too much about churning that list, though10:51
tpollard'tests/frontend/init.py::test_bad_project_name[Micheal Jackson]'10:59
tpollardbravo10:59
*** lachlan has quit IRC11:00
tlater[m]Hah!11:00
*** lachlan has joined #buildstream11:01
*** dtf has quit IRC11:11
*** phoenix has joined #buildstream11:23
*** phildawson_ has joined #buildstream11:24
*** phil has quit IRC11:25
*** phoenix has quit IRC11:31
jonathanmawaevri: as someone interested in this particular issue and with experience working with .gitlab-ci.yml, would you like to review https://gitlab.com/BuildStream/buildstream/merge_requests/1678 ?11:33
*** lachlan has quit IRC11:56
*** phil has joined #buildstream11:59
*** phildawson_ has quit IRC12:00
coldtomis there a way to disable all attempts at pulling using bst master?12:02
tpollardcoldtom: there's a few non official ways, but build has a --remote option12:04
tpollardwhich you can give a dummy value12:04
tpollardI think setting global fetchers to 0 also does the trick12:05
coldtomtyvm12:05
tlater[m]coldtom: For reference, an actual proper way to do that was proposed and we'd like to get it in before 2.012:07
*** phil has quit IRC12:09
*** phildawson_ has joined #buildstream12:11
*** lachlan has joined #buildstream12:31
gitlab-br-bottlater opened (was WIP) MR !1660 (tlater/annihilate_update_state->master: Remove update_state) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/166012:57
*** lachlan has quit IRC13:07
*** santi has quit IRC13:12
*** santi has joined #buildstream13:45
*** santi has quit IRC14:00
*** santi has joined #buildstream14:12
*** phil has joined #buildstream14:22
*** phildawson_ has quit IRC14:22
*** lachlan has joined #buildstream14:30
*** dtf has joined #buildstream14:48
*** phildawson_ has joined #buildstream14:54
*** phil has quit IRC14:55
*** santi has quit IRC15:43
*** bochecha has joined #buildstream15:46
benschubertjuergbi: I was looking at the source caching thing, with the speed up about not having to check if we have it in the source cache. Should we add a new Consistency.IN_SOURCE_CACHE  that is lower than CACHED and only look for this when we need just this?15:55
benschubertor just 'if consistency.RESOLVED && ${in_source_cache} ?15:56
gitlab-br-botjonathanmaw opened (was WIP) MR !1678 (jonathan/win32_ci->master: Add native windows CI configuration) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/167815:56
juergbibenschubert: I don't think it makes sense to add this as separate value to that enum as individual source plugins can't answer this15:56
benschubertfair15:56
gitlab-br-botmarge-bot123 closed issue #1163 (Add CI for native windows support) on buildstream https://gitlab.com/BuildStream/buildstream/issues/116315:57
gitlab-br-botmarge-bot123 merged MR !1678 (jonathan/win32_ci->master: Add native windows CI configuration) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/167815:57
benschubertthere is interestingly nowhere in the code base where we have = Consistency.RESOLVED15:57
juergbiwe return RESOLVED15:57
juergbiin source plugins15:57
benschubertoh!15:57
benschubertConsistency.RESOLVED == has_ref right? or is there some corner cases? (except local/workspace)15:59
*** santi has joined #buildstream16:00
juergbibenschubert: that's pretty much it16:00
benschubertOk, so do we actually need Consistency.RESOLVED?16:00
juergbi(and Consistency.CACHED implies this as well)16:00
benschubertor can we just do get_ref() ?16:01
juergbiyes but that should always be quick to determine16:01
benschubertAFAIK it's not since it's mixed with getting the CACHED consistency16:01
juergbiyes, I meant, for the source plugin it should always be quick to determine whether it's >= RESOLVED16:02
juergbichecking get_ref() is not None instead could be a possibility16:02
benschubertI guess there is two ways here:16:02
benschubert1) get rid of RESOLVED, and use get_ref() wherever we need it to be resolved16:02
benschubert2) split get_consistency() in two for get_resolved() and get_cached()16:02
juergbiright, 1) does seem fairly sensible16:03
juergbiiirc, it will require fixes in some plugins but should be simple enough16:03
juergbiand they would already be not complying to the API per documentation, afaict16:03
benschubertOk! However, then do we actually need CACHED? or could we 'just' have is_cached() ?16:03
juergbithen we could indeed just have cached but that would definitely be a breaking change16:04
benschubertok! I'll play with it, see what I can do and if it makes sense to do this change will move it to the ML :)16:04
juergbiprobably makes sense to go that way but maybe in two steps16:04
juergbisounds good16:04
benschubertthanks a lot!16:04
benschubertdo you see any other reason than for opening a workspace that a source should be in state 'CACHED' ?16:06
*** lachlan has quit IRC16:06
*** lachlan has joined #buildstream16:09
juergbibenschubert: can't think of a reason right now16:14
benschubertAwesome, thanks16:14
juergbi(the distinction is also relevant in the fetch job if it's not in the source cache yet, of course)16:15
benschubertright :)16:19
*** tiagogomes has quit IRC16:25
*** bochecha has quit IRC16:27
*** lachlan has quit IRC16:29
*** tiagogomes has joined #buildstream16:32
*** lachlan has joined #buildstream16:52
*** lachlan has quit IRC16:53
*** jonathanmaw has quit IRC17:02
*** tiagogomes has quit IRC17:11
*** lachlan has joined #buildstream17:22
*** narispo has quit IRC17:37
*** santi has quit IRC17:40
benschubertjuergbi: https://gitlab.com/BuildStream/buildstream/blob/master/src/buildstream/_gitsourcebase.py#L473 :'D17:44
benschubertI'll send a patch now for that17:44
juergbiyep, that's what I was referring to, ta17:45
gitlab-br-botBenjaminSchubert opened MR !1680 (bschubert/fix-git-ref->master: _gitsourcebase.py: Return 'None' when the ref is unset) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/168017:47
benschubertjuergbi: should I also check about the tags?17:47
benschubert(I don't think so, based on get_consistency)17:48
*** lachlan has quit IRC17:48
juergbiI agree17:48
juergbimirror.ref should suffice17:48
benschubertwe'll see if the pipeline passes :)17:50
benschuberthttps://gitlab.com/BuildStream/buildstream/merge_requests/1680/ making the git plugin respect the source API :D18:23
*** slaf_ has joined #buildstream18:28
*** slaf_ has joined #buildstream18:28
*** slaf_ has joined #buildstream18:28
*** slaf_ has joined #buildstream18:28
*** slaf_ has joined #buildstream18:29
*** slaf_ has joined #buildstream18:29
*** slaf has quit IRC18:29
*** slaf_ is now known as slaf18:29
gitlab-br-botcs-shadow approved MR !1677 (traveltissues/1184->master: buildgrid-compose.yml: reduce log level to WARN) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/167719:08
*** phoenix has joined #buildstream20:45
*** dtf has quit IRC20:51
*** rdale has quit IRC22:42
« Wednesday, 2019-10-30 Index Friday, 2019-11-01 »

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!