| *** narispo has quit IRC | 00:27 | |
| *** narispo has joined #buildstream | 00:27 | |
| *** narispo has quit IRC | 00:31 | |
| *** narispo has joined #buildstream | 00:34 | |
| *** narispo has quit IRC | 01:10 | |
| *** narispo has joined #buildstream | 01:10 | |
| *** phoenix has joined #buildstream | 05:54 | |
| *** narispo has quit IRC | 05:55 | |
| *** narispo has joined #buildstream | 05:55 | |
| *** phoenix has quit IRC | 06:17 | |
| *** tristan has quit IRC | 06:37 | |
| *** juergbi has quit IRC | 06:51 | |
| *** juergbi has joined #buildstream | 06:58 | |
| *** juergbi has quit IRC | 07:27 | |
| *** juergbi has joined #buildstream | 07:31 | |
| *** traveltissues has joined #buildstream | 07:37 | |
| *** tpollard has joined #buildstream | 09:08 | |
| *** santi has joined #buildstream | 09:29 | |
| *** santi has quit IRC | 09:31 | |
| *** santi has joined #buildstream | 09:33 | |
| benschubert | Our aarch64 tests seem to more or less always be failing: https://gitlab.com/BuildStream/buildstream/-/jobs/369697841 is someone still interested in those tests and can fix the runners? | 09:46 |
|---|---|---|
| gitlab-br-bot | BenjaminSchubert approved MR !1706 (chandan/interactive-tests->master: Add tests for interactive BuildStream operations) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1706 | 09:54 |
| *** jonathanmaw has joined #buildstream | 09:54 | |
| gitlab-br-bot | BenjaminSchubert approved MR !1725 (1211-stop-using-non-posix-features-in-small-test-suite-tests-to-avoid-fake-test-failures->master: Resolve "Stop using non-POSIX features in small test suite tests to avoid fake test failures") on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1725 | 10:01 |
| benschubert | coldtom: any plan to work on !1607 ? Or would you mind me taking it over? | 10:10 |
| gitlab-br-bot | MR !1607: WIP: Run standardised source tests from external plugins https://gitlab.com/BuildStream/buildstream/merge_requests/1607 | 10:10 |
| coldtom | benschubert: i don't think i'll have time any time soon, feel free to get it over the line | 10:11 |
| benschubert | Sure, thanks! | 10:12 |
| *** phildawson has joined #buildstream | 10:14 | |
| *** tme5 has joined #buildstream | 10:26 | |
| *** lachlan has joined #buildstream | 10:31 | |
| gitlab-br-bot | juergbi opened MR !1748 (juerg/runcli-unused-parameter->master: testing/runcli.py: Remove unused configure parameter from run() methods) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1748 | 10:47 |
| gitlab-br-bot | BenjaminSchubert approved MR !1748 (juerg/runcli-unused-parameter->master: testing/runcli.py: Remove unused configure parameter from run() methods) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1748 | 10:49 |
| gitlab-br-bot | juergbi opened MR !1749 (juerg/cas-directory-reset->master: _sandboxreapi.py: Reset CasBasedDirectory instead of recreating it) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1749 | 10:50 |
| *** lachlan has quit IRC | 10:56 | |
| *** cs-shadow has joined #buildstream | 11:24 | |
| *** rdale has joined #buildstream | 11:31 | |
| *** lachlan has joined #buildstream | 11:37 | |
| *** lachlan has quit IRC | 11:41 | |
| gitlab-br-bot | juergbi merged MR !1748 (juerg/runcli-unused-parameter->master: testing/runcli.py: Remove unused configure parameter from run() methods) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1748 | 11:46 |
| *** narispo has quit IRC | 11:51 | |
| juergbi | benschubert: are you happy with my reply? https://gitlab.com/BuildStream/buildstream/merge_requests/1749#note_255105970 | 11:52 |
| *** narispo has joined #buildstream | 11:52 | |
| benschubert | yep thanks! | 11:52 |
| cs-shadow | quick q - is marge healthy these days, or should I merge things manually? | 12:28 |
| *** lachlan has joined #buildstream | 12:30 | |
| *** lachlan has quit IRC | 12:36 | |
| tme5 | i've run into a bit of a catch-22 with my work on the git source plugin. there's not much documentation on it, but Source.get_source_fetchers is supposed to iterate over all sources we will be fetching from | 12:38 |
| tme5 | however, I can't make that list complete without actually doing the fetching.. which seems like it might be problematic | 12:38 |
| *** toscalix has joined #buildstream | 12:44 | |
| *** narispo has quit IRC | 12:52 | |
| *** narispo has joined #buildstream | 12:53 | |
| tme5 | do I understand that correctly? is there any harm in not implementing get_source_fetchers? | 12:54 |
| tme5 | it seems to be only used in source.py and in one function | 12:55 |
| coldtom | tme5: the docs here suggest it's optional https://docs.buildstream.build/1.4.1/buildstream.source.html#abstract-methods | 12:56 |
| coldtom | ...unless you're using SourceFetchers, like the git plugin does | 12:57 |
| * coldtom should read better | 12:57 | |
| tme5 | I had read the docs, it says "This only *needs* to be implemented for sources that need to download from multiple URLs while fetching" | 12:57 |
| tme5 | but i don't know all those URLs before fetching | 12:58 |
| *** lachlan has joined #buildstream | 12:58 | |
| *** lachlan has quit IRC | 12:59 | |
| gitlab-br-bot | juergbi merged MR !1749 (juerg/cas-directory-reset->master: _sandboxreapi.py: Reset CasBasedDirectory instead of recreating it) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1749 | 13:00 |
| *** santi has quit IRC | 13:14 | |
| *** narispo has quit IRC | 13:23 | |
| *** narispo has joined #buildstream | 13:24 | |
| *** toscalix has quit IRC | 13:32 | |
| *** toscalix has joined #buildstream | 13:33 | |
| benschubert | tme5: Why wouldn't you know? The sources should have been tracked at fetch time, so you'll have a ref, and know all the submodules that are present right? | 13:48 |
| *** santi has joined #buildstream | 13:50 | |
| coldtom | benschubert: i think tme5 is trying to recursively submodule | 13:56 |
| coldtom | some projects may rely on their submodules' submodules | 13:57 |
| benschubert | coldtom: you _should_ still have all urls at fetch time no? Your source is meant to be fully resolved once you fetch | 13:57 |
| benschubert | So we might have a problem that the git source base doesn't handle recursive submodules in the metadata, which we could fix | 13:57 |
| coldtom | ah, i think we have crossed wires on source fetchers/sources perhaps | 13:58 |
| coldtom | in which case, yes, everything needed _should_ be there | 13:59 |
| *** lachlan has joined #buildstream | 14:07 | |
| tme5 | benschubert, my understanding based on the name is that SourceFetchers are only useful for fetching sources | 14:13 |
| tme5 | so it stood to reason that i should assume my source is not fetched when thinking about them | 14:13 |
| *** lachlan has quit IRC | 14:14 | |
| tme5 | is this not correct? | 14:14 |
| benschubert | That seems a correct assumption | 14:15 |
| cs-shadow | there's a difference between not having sources fetched and not knowing what to fetch. source fetchers can assume that sources aren't fetched locally, but they should still know that what they need to fetch without fetching it | 14:18 |
| tme5 | but my issue is that I can't know completely what I need to fetch without fetching anything | 14:18 |
| tme5 | because i'm adding recursion to submodule checkout | 14:19 |
| *** lachlan has joined #buildstream | 14:20 | |
| cs-shadow | I understand that you need to clone one repo to find out about its submodules and so on, but all of this should happen at track time. Once we've finished tracking we should know exactly what needs to be fetched on subsequent runs | 14:20 |
| tme5 | i'm not sure i understand | 14:21 |
| tme5 | what do you mean by 'track' ? I thought 'tracking' was automatically updating the refs in an element file. Isn't that an optional stage? | 14:22 |
| benschubert | tme5: it's optional in the sense that you don't _need_ to do it if your element already has a ref (or you can set the ref manually) | 14:23 |
| cs-shadow | yeah, automatically updating the refs is optional, having the refs is not optional at all | 14:23 |
| cs-shadow | so, what I'm saying is that the ref should contain enough information to be able to fetch everything that it needs to fetch at that revision | 14:24 |
| tme5 | in the singular, parent repo? that's not correct as far as i know | 14:24 |
| cs-shadow | not sure if I understand - what's not correct? | 14:25 |
| tme5 | "the ref should contain enough information to be able to fetch everything that needs to fetch at that revision [of the git repo]" | 14:25 |
| tme5 | afaik a git repo only stores its immediate submodules, it doesn't know anything about its subsubmodules, etc. | 14:27 |
| cs-shadow | ah, I see what you mean. I meant "should" as in what should happen ideally. our current handling of submodules certainly needs work | 14:28 |
| tme5 | but i'm not sure the Source class really has the vocabulary to deal with what git needs to do | 14:30 |
| *** lachlan has quit IRC | 14:30 | |
| cs-shadow | what do you think we need extra? | 14:32 |
| cs-shadow | To me recursive submodules sounds very similar to the pip source, where you can list some packages as direct dependencies, but don't know all transitive deps beforehand. At track time, we figure out the full set of dependencies and store them as the ref for the source. | 14:32 |
| cs-shadow | Can't we do the same in this case? i.e. fetch everything recursively at track time, and then record the version information the refer? | 14:32 |
| cs-shadow | *and then record the version information in the ref field | 14:33 |
| tme5 | i suppose we could, that would change how the source works in this case quite a lot | 14:39 |
| tme5 | to be honest, though, looking at __do_fetch it looks like get_source_fetchers is just a convenience method in the absence of an actual fetch method on the source | 14:42 |
| tme5 | so if I implement fetch myself I don't have to worry about this problem | 14:42 |
| tme5 | i think, anyway.. | 14:43 |
| *** lachlan has joined #buildstream | 14:54 | |
| *** lachlan has quit IRC | 14:59 | |
| tme5 | all green on the git test suite! | 15:08 |
| *** narispo has quit IRC | 15:19 | |
| *** narispo has joined #buildstream | 15:20 | |
| *** narispo has quit IRC | 15:32 | |
| *** lachlan has joined #buildstream | 15:32 | |
| *** narispo has joined #buildstream | 15:33 | |
| *** lachlan has quit IRC | 15:36 | |
| gitlab-br-bot | cs-shadow merged MR !1706 (chandan/interactive-tests->master: Add tests for interactive BuildStream operations) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1706 | 15:42 |
| benschubert | o/ | 15:42 |
| tpollard | nice | 15:42 |
| cs-shadow | next, I need to add some docs on this topic :) | 15:42 |
| *** lachlan has joined #buildstream | 15:51 | |
| *** lachlan has quit IRC | 15:54 | |
| *** phildawson_ has joined #buildstream | 16:10 | |
| *** phildawson has quit IRC | 16:10 | |
| *** toscalix has quit IRC | 16:20 | |
| *** toscalix has joined #buildstream | 16:21 | |
| *** toscalix has quit IRC | 16:24 | |
| *** santi has quit IRC | 16:29 | |
| gitlab-br-bot | juergbi opened MR !1751 (juerg/buildbox-run-userchroot->master: Add CI job to test buildbox-run-userchroot) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1751 | 16:31 |
| gitlab-br-bot | juergbi opened MR !1752 (juerg/buildbox-run-bubblewrap->master: Add CI job to test buildbox-run-bubblewrap and buildbox-fuse) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1752 | 16:39 |
| *** toscalix has joined #buildstream | 16:42 | |
| *** traveltissues has quit IRC | 16:44 | |
| *** toscalix has quit IRC | 16:45 | |
| *** toscalix_ has joined #buildstream | 16:45 | |
| *** toscalix_ has quit IRC | 16:45 | |
| *** toscalix has joined #buildstream | 16:46 | |
| gitlab-br-bot | juergbi opened (was WIP) MR !1738 (juerg/buildbox-run->master: Add buildbox-run sandboxing backend) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1738 | 16:47 |
| gitlab-br-bot | cs-shadow opened MR !1754 (chandan/fix-manifest->master: Minor packaging-related fixups) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1754 | 16:49 |
| gitlab-br-bot | BenjaminSchubert approved MR !1754 (chandan/fix-manifest->master: Minor packaging-related fixups) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1754 | 16:50 |
| gitlab-br-bot | tpollard opened (was WIP) MR !1745 (tpollard/messagerate->master: Make message & status rendering be tick driven by default) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1745 | 17:04 |
| *** santi has joined #buildstream | 17:12 | |
| *** tpollard has quit IRC | 17:13 | |
| *** lachlan has joined #buildstream | 17:21 | |
| *** lachlan has quit IRC | 17:24 | |
| *** toscalix has quit IRC | 17:27 | |
| *** slaf has quit IRC | 17:29 | |
| *** slaf has joined #buildstream | 17:36 | |
| *** slaf has joined #buildstream | 17:36 | |
| *** slaf has joined #buildstream | 17:36 | |
| *** slaf has joined #buildstream | 17:37 | |
| *** slaf has joined #buildstream | 17:37 | |
| *** slaf has joined #buildstream | 17:37 | |
| *** phildawson_ has quit IRC | 17:37 | |
| *** slaf has joined #buildstream | 17:38 | |
| *** slaf has joined #buildstream | 17:38 | |
| *** slaf has joined #buildstream | 17:38 | |
| *** slaf has joined #buildstream | 17:38 | |
| *** slaf has joined #buildstream | 17:39 | |
| *** slaf has joined #buildstream | 17:39 | |
| *** slaf has joined #buildstream | 17:39 | |
| *** santi has quit IRC | 17:46 | |
| *** tme5 has quit IRC | 17:47 | |
| gitlab-br-bot | cs-shadow merged MR !1754 (chandan/fix-manifest->master: Minor packaging-related fixups) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1754 | 17:52 |
| *** lachlan has joined #buildstream | 17:54 | |
| *** narispo has quit IRC | 17:56 | |
| *** narispo has joined #buildstream | 17:56 | |
| *** lachlan has quit IRC | 17:58 | |
| *** tiagogomes has quit IRC | 18:24 | |
| *** jonathanmaw has quit IRC | 18:39 | |
| *** narispo has quit IRC | 19:13 | |
| *** narispo has joined #buildstream | 21:25 | |
| *** cs-shadow has quit IRC | 21:39 | |
| *** traveltissues has joined #buildstream | 21:43 | |
| *** narispo has quit IRC | 22:22 | |
| *** narispo has joined #buildstream | 22:25 | |
| *** narispo has quit IRC | 22:32 | |
| *** traveltissues has quit IRC | 22:34 | |
| *** narispo has joined #buildstream | 22:35 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!