| *** milloni has quit IRC | 00:00 | |
| *** milloni has joined #buildstream | 00:00 | |
| *** bilelmoussaoui has joined #buildstream | 01:23 | |
| *** nimish has joined #buildstream | 02:20 | |
| *** nimish has quit IRC | 02:27 | |
| *** nimish has joined #buildstream | 02:29 | |
| *** nimish has quit IRC | 02:44 | |
| *** nimish has joined #buildstream | 02:45 | |
| *** xjuan has quit IRC | 03:41 | |
| *** nimish has quit IRC | 03:51 | |
| *** nimish has joined #buildstream | 06:33 | |
| *** bochecha has joined #buildstream | 07:10 | |
| *** bilelmoussaoui has quit IRC | 07:19 | |
| *** alatiera has joined #buildstream | 07:56 | |
| *** nimish has quit IRC | 08:21 | |
| *** nimish has joined #buildstream | 08:21 | |
| *** toscalix has joined #buildstream | 08:52 | |
| *** mohan43u has joined #buildstream | 08:52 | |
| *** sambishop has joined #buildstream | 09:11 | |
| *** nimish has quit IRC | 09:41 | |
| *** nimish has joined #buildstream | 09:42 | |
| *** rdale has joined #buildstream | 09:51 | |
| *** nimish has quit IRC | 09:57 | |
| *** nimish has joined #buildstream | 09:57 | |
| *** nimish has joined #buildstream | 09:57 | |
| *** raoul has joined #buildstream | 10:02 | |
| *** jonathanmaw has joined #buildstream | 10:06 | |
| *** nimish has quit IRC | 10:22 | |
| *** nimish has joined #buildstream | 10:23 | |
| gitlab-br-bot | valentindavid merged MR !1069 (valentindavid/git-reduced-history->master: buildstream/_gitsourcebase.py: Reduce git history for git describe.) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1069 | 10:33 |
|---|---|---|
| *** alatiera has quit IRC | 10:37 | |
| *** alatiera has joined #buildstream | 10:38 | |
| *** bilelmoussaoui has joined #buildstream | 10:55 | |
| *** nimish has quit IRC | 10:58 | |
| *** nimish has joined #buildstream | 10:58 | |
| *** cs-shadow has quit IRC | 11:04 | |
| jennis | jmac, I've made your suggested change to https://gitlab.com/BuildStream/buildstream/merge_requests/1072 | 11:06 |
| jennis | I assume we're happy to merge once pipeline passes? | 11:07 |
| *** rdale has quit IRC | 11:12 | |
| *** nimish has quit IRC | 11:23 | |
| *** nimish has joined #buildstream | 11:23 | |
| *** rdale has joined #buildstream | 11:25 | |
| gitlab-br-bot | jennis merged MR !1072 (jennis/doc_fixes_in_context->master: Small documentation/comment fixes in context.py) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1072 | 11:33 |
| *** cs-shadow has joined #buildstream | 11:35 | |
| jmac | jennis: Yes, go for it | 11:40 |
| jmac | I approved it when I reviewed it as it was a very minor nit-pick | 11:40 |
| jennis | yeah I noticed :) thanks jmac | 11:40 |
| *** nimish has quit IRC | 11:48 | |
| *** nimish has joined #buildstream | 11:49 | |
| *** nimish has quit IRC | 11:54 | |
| *** bilelmoussaoui has joined #buildstream | 11:54 | |
| *** nimish has joined #buildstream | 11:54 | |
| *** nimish has quit IRC | 12:14 | |
| *** nimish has joined #buildstream | 12:15 | |
| *** nimish has joined #buildstream | 12:15 | |
| *** abderrahim32 has quit IRC | 12:18 | |
| *** abderrahim32 has joined #buildstream | 12:18 | |
| *** nimish has quit IRC | 12:20 | |
| *** nimish has joined #buildstream | 12:21 | |
| *** slaf_ has joined #buildstream | 12:23 | |
| *** slaf has quit IRC | 12:25 | |
| *** slaf_ is now known as slaf | 12:25 | |
| *** nimish has quit IRC | 12:26 | |
| *** nimish has joined #buildstream | 12:26 | |
| *** bilelmoussaoui has quit IRC | 12:31 | |
| *** nimish has quit IRC | 12:41 | |
| *** nimish has joined #buildstream | 12:42 | |
| *** alatiera_ has joined #buildstream | 12:48 | |
| *** alatiera has quit IRC | 12:48 | |
| *** alatiera_ is now known as alatiera | 12:49 | |
| gitlab-br-bot | juergbi merged MR !1071 (raoul/cas-refactor->master: Cas refactor) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1071 | 12:56 |
| gitlab-br-bot | phildawson opened (was WIP) MR !1075 (phil/plugin-testing-api->master: Expose basic api for testing external plugins.) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1075 | 13:17 |
| phil | Hi all, anyone able to take a look at !1075 for me. It's my thoughts on what a basic API for use when testing none-core plugins might look like so would appreciate any feedback :) | 13:19 |
| cs-shadow | phil: I was planning to have a look, this seems really useful. In the meantime, it seems like the lint and docs jobs are failing so they might need a little love :) | 13:26 |
| *** abderrahim32 has quit IRC | 13:28 | |
| *** nimish has quit IRC | 13:32 | |
| *** nimish has joined #buildstream | 13:32 | |
| phil | cs-shadow, cheers. I'm looking into that now, but struggling to work out what's happening with the docs jobs. | 13:40 |
| *** tristan has joined #buildstream | 13:58 | |
| *** sambishop has quit IRC | 14:00 | |
| *** lachlan has joined #buildstream | 14:14 | |
| *** sambishop has joined #buildstream | 14:16 | |
| *** nimish has quit IRC | 14:17 | |
| *** nimish has joined #buildstream | 14:17 | |
| *** nimish has quit IRC | 14:27 | |
| *** nimish has joined #buildstream | 14:28 | |
| *** alatiera has quit IRC | 14:32 | |
| *** nimish has quit IRC | 14:33 | |
| *** nimish has joined #buildstream | 14:33 | |
| *** alatiera has joined #buildstream | 14:36 | |
| *** lachlan has quit IRC | 14:37 | |
| *** nimish has quit IRC | 14:38 | |
| *** nimish has joined #buildstream | 14:39 | |
| *** lachlan has joined #buildstream | 14:41 | |
| gitlab-br-bot | aevri opened MR !1076 (chadnan->master: bst-docker-import: fix Chandan's name) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1076 | 14:44 |
| *** nimish has quit IRC | 14:49 | |
| *** nimish has joined #buildstream | 14:49 | |
| *** alatiera has quit IRC | 14:56 | |
| *** sambishop has quit IRC | 14:56 | |
| *** sambishop has joined #buildstream | 14:58 | |
| gitlab-br-bot | aevri merged MR !1076 (chadnan->master: bst-docker-import: fix Chandan's name) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1076 | 15:13 |
| *** sambishop has quit IRC | 15:17 | |
| *** sambishop has joined #buildstream | 15:22 | |
| *** nimish has quit IRC | 15:29 | |
| *** nimish has joined #buildstream | 15:29 | |
| *** sambishop has quit IRC | 15:32 | |
| *** nimish has quit IRC | 15:35 | |
| *** nimish has joined #buildstream | 15:38 | |
| jennis | tristan, are you around? | 15:38 |
| *** tristan has quit IRC | 15:42 | |
| *** tristan has joined #buildstream | 15:44 | |
| *** sambishop has joined #buildstream | 15:46 | |
| *** ChanServ sets mode: +o tristan | 15:46 | |
| tristan | jennis, yep | 15:46 |
| tristan | jennis, whats up | 15:46 |
| tristan | jennis, Can I land the test refactor ? | 15:46 |
| jennis | ahh wonderful | 15:46 |
| jennis | no this is something unrelated :p | 15:47 |
| jennis | So basically I've been working on the bst artifact subcommands | 15:47 |
| jennis | and while for delete and checkout, we've been able to handle artifact refs relatively easily, for pull, I'd had to add a whole load of questionable logic in cli.py so that we can directly pull an artifact ref | 15:48 |
| *** nimish has quit IRC | 15:48 | |
| jennis | After discussing this with juergbi, we've agreed that this is not the proper way to do it | 15:48 |
| jennis | Here is what I tried doing for `bst artifact pull`: https://gitlab.com/BuildStream/buildstream/blob/6e12c3fe5d1067b97e1850355f4e69cc6905f8d4/buildstream/_frontend/cli.py#L992 | 15:49 |
| gitlab-br-bot | tristanvb opened MR !1077 (tristan/detoxing-integration-tests->master: Make integration tests parallelizable) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1077 | 15:49 |
| *** nimish has joined #buildstream | 15:49 | |
| jennis | https://gitlab.com/BuildStream/buildstream/blob/6e12c3fe5d1067b97e1850355f4e69cc6905f8d4/buildstream/_frontend/cli.py#L992 | 15:49 |
| *** nimish has joined #buildstream | 15:49 | |
| jennis | Sorry that was the wrong link | 15:49 |
| tristan | jennis, yeah that looks totally wrong, that logic should bleed out so high up | 15:50 |
| jennis | Now basically, when we're dealing with pulling elements, we pass the element to stream, this does the loading, and adds to the pull queue and so on | 15:50 |
| tristan | I dont think it even makes sense for the cli.py to every know what the artifact cache API is | 15:51 |
| jennis | yes, and the reason I did this is because I don't see how we can cleanly handle artifact refs | 15:51 |
| tristan | cli.py/app.py should be only talking to stream | 15:51 |
| jennis | Ok, thanks for that clarification | 15:51 |
| jennis | But even if I moved this logic to stream, we're not really handling artifact refs until we hit CASCache | 15:51 |
| tristan | stream is the main interface of the bst core, and everything in _frontend/ should go through it... | 15:52 |
| jennis | ok thanks | 15:52 |
| tristan | initialization is currently messy (i.e. a frontend needs to load the Context() and Project() itself)... anyway, you get the idea | 15:52 |
| jennis | But now it's a question of, how do I integrate artifact ref handling into the pipeline | 15:52 |
| tristan | What is the pipeline ? | 15:53 |
| tristan | You mean the loaded element graph ? | 15:53 |
| jennis | into the scheduler and queues, because these all work with ElementJob | 15:53 |
| tristan | I don't believe that you do, ok so basically you are pointing out that "pull" is different in the sense that we need to construct an element in order to do something ? | 15:54 |
| jennis | So I'm extending pull to be able to take an artifact ref as well as an element name here | 15:55 |
| tristan | jennis, I think the cli probably determines whether a given target is an element or a ref first, and then uses a different Stream() pathway for pulling refs | 15:55 |
| jennis | ok, so we need to create another Stream method to handle the pulling of refs? | 15:55 |
| tristan | Well, I don't see how you are going to do it the same way | 15:55 |
| tristan | A ref can be pulled directly | 15:56 |
| tristan | Loading an element basically produces that ref, which can then be pulled directly | 15:56 |
| jennis | But then there is the loading of the project and gathering the user conf and project conf to get the remotes and then the initialising of the remotes as CASCache objects and then pulling the ref | 15:57 |
| tristan | When pulling element targets, you really need to load the project first of all, which you dont need for a ref | 15:57 |
| jennis | which is all the messy logic I've currently got in cli.py | 15:57 |
| tristan | And you also need to resolve the refs on the fly when you deal with elements | 15:57 |
| jennis | Well, you do need to load the project to get the config, and thus, the urls of the remotes? | 15:57 |
| tristan | or that is unclear actually | 15:58 |
| tristan | perhaps only at build time there is a question of which artifact is preferred depending on strict mode etc | 15:58 |
| juergbi | the current logic in the MR also allows specifying a mix of elements and artifact refs. shall we just disallow that? or should stream support this mix in a single session? | 15:58 |
| tristan | jennis, the Project perhaps, but not the full data model | 15:58 |
| jennis | ok sure | 15:59 |
| jennis | So basically, we're good to write a Stream method which figures out the remotes and pulls the refs directly? | 15:59 |
| tristan | juergbi, Can we determine what it is by the format of the string ? | 15:59 |
| juergbi | yes | 15:59 |
| tristan | jennis, I would say yes | 16:00 |
| jennis | ok that's good to hear, thanks tristan | 16:00 |
| juergbi | this should still go via scheduler, though, right? | 16:00 |
| * jennis got the impression that it didn't | 16:00 | |
| juergbi | or are we saying we skip the regular UI if refs are specified? | 16:01 |
| juergbi | seems a bit odd from the user perspective | 16:01 |
| juergbi | although I could understand it from the implementation perspective | 16:01 |
| jennis | juergbi, do you think the 'odd from the user perspective' is still odd if we restrict to only refs or only elements? | 16:03 |
| juergbi | would definitely be better than the mixed mode right now in the MR | 16:03 |
| juergbi | it would still not be ideal, imo | 16:03 |
| jennis | Yeah, but still odd in the sense that if you use the command with an element, and then use it again with a ref, what you see as a user looks completely different? | 16:04 |
| juergbi | however, we could say that we do it like that for now and consider moving this to the scheduler/pipeline later | 16:04 |
| jennis | That's good | 16:04 |
| juergbi | up to tristan ;) | 16:04 |
| jennis | Another thing I wanted to address is the '--remote' option | 16:04 |
| jennis | Can this only be a remote that exists in the project.conf(s) or user.conf? | 16:05 |
| jennis | Or should we just be able to specify any old url? (I just don't know how we deal with certificates and keys when this is the case) especially when it comes to pushing | 16:05 |
| juergbi | I thought I saw a discussion about this somewhere | 16:06 |
| juergbi | for pushing I think we can indeed not support CLI URLs due to typically requiring authentication | 16:06 |
| juergbi | this might be something to discuss at the gathering | 16:07 |
| juergbi | I proposed a topic with regards to artifact server configuration in projects and junctions | 16:07 |
| juergbi | would probably fit in there | 16:07 |
| tristan | sorry stepped out | 16:12 |
| tristan | had a thought, reading backlog... | 16:12 |
| tristan | I had the same thought as juergbi | 16:12 |
| tristan | that maybe these would bypass the scheduler | 16:12 |
| tristan | thats very un-nice | 16:13 |
| tristan | I was thinking it would be a dummy element per ref or something like that | 16:13 |
| tristan | juergbi, jennis, it would be nice too if we didnt run the scheduler *twice* just because refs and elements were intermingled on the command line | 16:14 |
| *** nimish has quit IRC | 16:14 | |
| *** nimish has joined #buildstream | 16:14 | |
| *** sambishop has quit IRC | 16:15 | |
| juergbi | yes, I think we should either not allow intermingling or combine them | 16:15 |
| tristan | so we'd have a.) the cli.py passes the opaque targets to Stream.pull()... b.) Stream.pull() sorts the elements from the refs, and loads a build graph of those, and resolves `--deps` for only those... c.) Stream.pull() adds dummy elements for any ref which was specified and appends those to the list of elements... d.) Stream.pull() invokes the scheduler with a PullQueue() | 16:15 |
| tristan | With that, we should also document explicitly that recursive `--deps` options are not supported on refs, each specified ref will be pulled with a `--deps none` criteria | 16:16 |
| tristan | That is, whether we allow intermingling or not, we need to specify that for refs | 16:17 |
| tristan | it might not be true at push time, and it may not be true forever for pull (funky dynamic scheduling ?) | 16:17 |
| jennis | Oh, they're not options, that's a proposed order of things | 16:19 |
| jennis | tristan, could you expand on the dummy element? I'm not really sure I'm following this bit | 16:21 |
| jennis | Is this just *something* that's going to invoke the loading of the project and initialising of the caches? | 16:22 |
| tristan | jennis, We can create an internal dummy element which has no dependencies and it's name is derived from the artifact name (I think we can derive the original element name from the artifact name) | 16:24 |
| tristan | jennis, that element needs to hardwire itself to produce the right key | 16:24 |
| tristan | This might be useful in other commands which involve handling refs and dont need project loading | 16:25 |
| jennis | ok, currently that's a big grey area for me, but I'll dive into it and see where I get | 16:25 |
| tristan | jennis, Anyway... once you have a dummyelement which you can instantiate like DummyElement(project_name, element_name, cache_key) | 16:25 |
| tristan | which produces the correct artifact name | 16:26 |
| jennis | And parsing the ref should be easy enough | 16:26 |
| tristan | You can just pop those into a PullQueue along with any other elements | 16:26 |
| jennis | ok, thanks tristan, juergbi | 16:27 |
| tristan | jennis, It can be tough to do it through all the existing cache key mechanics... it might make more sense to move the ArtifactCache.get_artifact_fullname() -> Element.get_artifact_name() | 16:28 |
| tristan | jennis, and make sure everything asks Element.get_artifact_name() instead... then have DummyElement just override that | 16:28 |
| gitlab-br-bot | tristanvb merged MR !1077 (tristan/detoxing-integration-tests->master: Make integration tests parallelizable) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1077 | 16:32 |
| tristan | jennis, How about the tests refactor... can we land that ? | 16:33 |
| tristan | https://gitlab.com/BuildStream/buildstream/merge_requests/1062 | 16:34 |
| tristan | gitlab-br-bot, tell me about !1062 | 16:34 |
| gitlab-br-bot | MR !1062: General refactor in tests directory https://gitlab.com/BuildStream/buildstream/merge_requests/1062 | 16:34 |
| tristan | :) | 16:34 |
| cs-shadow | tristan: hi, would you be open to toxifying the 1.2 branch? (The reason I ask is that there are chances that the dependencies will drift over time on that branch, especially since we don't work on that often) | 16:34 |
| tristan | cs-shadow, I definitely think we should port all of the tox related things there *if* we're not going to do a stable in a while | 16:35 |
| tristan | that discussion is going to happen at the gathering but I wouldnt object to being prepared in advance | 16:35 |
| cs-shadow | tristan: ok cool, I'll have a go at it. Just wanted to make sure that you don't have any reservations against it | 16:36 |
| jennis | oh nice, I'll bear that in mind. You can land the tests refactor, it won't take me long to change checkout -> artifact checkout | 16:36 |
| tristan | juergbi, you said 1.9.x but I think 1.90.x rings a bell, and leaves a nicely paranoid namespace just incase | 16:36 |
| juergbi | tristan: I meant 1.90.x | 16:37 |
| tristan | jennis, thanks :) | 16:37 |
| tristan | juergbi, ah, yeah indeed | 16:38 |
| tristan | Are there things planned for monday in London ? | 16:44 |
| tristan | laurence, ? | 16:44 |
| tristan | or it's really just from tuesday | 16:44 |
| *** nimish has quit IRC | 16:44 | |
| *** nimish has joined #buildstream | 16:45 | |
| Kinnison | Those of us from mcr don't get to London until late on Monday so I doubt there're plans | 16:45 |
| Kinnison | At least plans which would involve us | 16:46 |
| Kinnison | :-D | 16:46 |
| Kinnison | I don't think I get to my hotel until gone 9pm :( | 16:46 |
| tristan | I see | 16:48 |
| tristan | And on thursday it's off to brussels | 16:48 |
| Kinnison | I thought that was Friday, but I'm going back to mcr anyway :-D | 16:50 |
| * Kinnison thought tue/wed was buildstream gathering, and thu was this "build tools" hack day | 16:50 | |
| cs-shadow | Kinnison: right, 29-30 is more buildstream focused, similar to our previous events. And 31st will have a more diverse audience. This should be similar to the hackathon that we did in NY after BazelCon, in case that rings a bell in people's minds | 16:52 |
| Kinnison | cool | 16:52 |
| * Kinnison is looking forward to it all | 16:52 | |
| Kinnison | Though monday night will suck | 16:52 |
| laurence | tristan, nothing on monday | 16:53 |
| tristan | Gotcha :) | 16:57 |
| *** tristan has quit IRC | 17:01 | |
| *** raoul has quit IRC | 17:15 | |
| *** nimish has quit IRC | 17:15 | |
| *** nimish has joined #buildstream | 17:15 | |
| *** nimish has quit IRC | 17:20 | |
| *** nimish has joined #buildstream | 17:21 | |
| *** nimish has quit IRC | 17:36 | |
| *** nimish has joined #buildstream | 17:37 | |
| jennis | Doesn't look like the CASCache refactor took into account `bst-artifact-server` | 17:44 |
| gitlab-br-bot | aevri opened MR !1078 (aevri/shell_separator_hint->master: cli.py: add a hint about '--' to 'bst shell' help) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1078 | 17:46 |
| gitlab-br-bot | jennis opened issue #867 (`bst-artifact-server` no longer works) on buildstream https://gitlab.com/BuildStream/buildstream/issues/867 | 17:49 |
| *** nimish has quit IRC | 17:52 | |
| *** nimish has joined #buildstream | 17:52 | |
| *** lachlan has quit IRC | 17:52 | |
| jennis | Ahh, fixed it | 17:56 |
| jennis | I think | 17:56 |
| jennis | wow, how did I do that | 17:56 |
| *** nimish has quit IRC | 17:57 | |
| *** jonathanmaw has quit IRC | 18:00 | |
| jennis | Quick review for anyone that's up for it: https://gitlab.com/BuildStream/buildstream/merge_requests/1079 | 18:00 |
| jmac | Approved | 18:01 |
| jmac | (Take it out of WIP, though!) | 18:01 |
| gitlab-br-bot | jennis opened (was WIP) MR !1079 (jennis/fix_bst_artifact_server->master: Ensure that we can execute `bst-artifact-server`) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1079 | 18:02 |
| jennis | Done | 18:02 |
| jennis | Set to merge | 18:02 |
| jmac | Thanks! | 18:02 |
| jennis | Perhaps we should have a test to ensure this command returns a help message or something | 18:02 |
| *** lachlan has joined #buildstream | 18:16 | |
| *** lachlan has quit IRC | 18:29 | |
| gitlab-br-bot | jennis closed issue #867 (`bst-artifact-server` no longer works) on buildstream https://gitlab.com/BuildStream/buildstream/issues/867 | 18:33 |
| gitlab-br-bot | jennis merged MR !1079 (jennis/fix_bst_artifact_server->master: Ensure that we can execute `bst-artifact-server`) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1079 | 18:33 |
| *** toscalix has quit IRC | 18:53 | |
| *** lachlan has joined #buildstream | 19:12 | |
| *** oknf[m] has left #buildstream | 19:13 | |
| *** tristan has joined #buildstream | 19:25 | |
| *** lachlan has quit IRC | 19:37 | |
| *** xjuan has joined #buildstream | 19:43 | |
| *** lachlan has joined #buildstream | 20:05 | |
| gitlab-br-bot | tristanvb closed issue #712 (BuildStream spends a long time pulling/looking to pull before doing anything useful) on buildstream https://gitlab.com/BuildStream/buildstream/issues/712 | 20:10 |
| gitlab-br-bot | tristanvb merged MR !1067 (tristan/element-processing-order->master: Scheduler refactor, fix processing order) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1067 | 20:10 |
| gitlab-br-bot | tristanvb opened MR !1080 (tristan/cleanup-staging-tempdir->master: element.py: Cleanup temporary staging directories on termination) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1080 | 20:18 |
| gitlab-br-bot | tristanvb opened MR !1081 (tristan/error-message-regression->master: sandbox/sandbox.py: Display failed commands in the detail string) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1081 | 20:19 |
| *** lachlan has quit IRC | 20:49 | |
| gitlab-br-bot | tristanvb merged MR !1080 (tristan/cleanup-staging-tempdir->master: element.py: Cleanup temporary staging directories on termination) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1080 | 21:17 |
| *** tristan has quit IRC | 21:27 | |
| *** tristan has joined #buildstream | 21:46 | |
| *** bochecha has quit IRC | 22:05 | |
| *** bochecha has joined #buildstream | 22:06 | |
| *** tristan has quit IRC | 22:39 | |
| *** johnward has quit IRC | 22:40 | |
| *** benbrown has quit IRC | 22:41 | |
| *** paulsherwood has quit IRC | 22:41 | |
| *** paulsherwood has joined #buildstream | 22:41 | |
| *** benbrown has joined #buildstream | 22:41 | |
| *** johnward has joined #buildstream | 22:41 | |
| *** WSalmon has quit IRC | 22:42 | |
| *** mablanch has quit IRC | 22:42 | |
| *** WSalmon has joined #buildstream | 22:42 | |
| *** mablanch has joined #buildstream | 22:43 | |
| *** tristan has joined #buildstream | 22:59 | |
| *** bochecha has quit IRC | 23:01 | |
| valentind | I really do not understand test tests/integration/workspace.py::test_workspace_update_dependency_failed | 23:03 |
| valentind | I think it has passed until now by chance only. | 23:04 |
| valentind | Are all files from a workspace supposed to be older than staged dependencies? | 23:05 |
| valentind | Ah the file is not supposed to change. | 23:07 |
| valentind | It is. Just mtimes are supposed to be updated. | 23:19 |
| *** ChanServ sets mode: +o tristan | 23:44 | |
| tristan | Looks like the bzr plugin doesnt handle parallelism that well, I'm seeing the "atomic" moves not being very atomic | 23:45 |
| *** tristan has quit IRC | 23:49 | |
| *** xjuan has quit IRC | 23:49 | |
| *** tristan has joined #buildstream | 23:57 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!