| *** alatiera has quit IRC | 00:02 | |
| *** alatiera has joined #buildstream | 00:09 | |
| *** pro[m] has joined #buildstream | 00:45 | |
| *** Ox37b has joined #buildstream | 02:10 | |
| *** swick has quit IRC | 02:26 | |
| *** swick has joined #buildstream | 02:29 | |
| *** BubbatheVTOG has joined #buildstream | 03:13 | |
| *** alatiera has quit IRC | 03:32 | |
| *** nimish has quit IRC | 04:17 | |
| *** nimish has joined #buildstream | 04:35 | |
| *** notwa has joined #buildstream | 04:38 | |
| *** tristan has joined #buildstream | 04:58 | |
| *** wN has joined #buildstream | 05:30 | |
| *** nimish has quit IRC | 05:35 | |
| *** ChanServ sets mode: +o tristan | 05:40 | |
| *** tristan has quit IRC | 05:40 | |
| *** tristan has joined #buildstream | 05:40 | |
| tristan | cs-shadow, around ? | 05:41 |
|---|---|---|
| *** ChanServ sets mode: +o tristan | 05:41 | |
| *** nimish has joined #buildstream | 05:48 | |
| *** tristan has quit IRC | 06:15 | |
| *** mohan43u has quit IRC | 06:34 | |
| *** zelest has joined #buildstream | 07:11 | |
| *** mohan43u has joined #buildstream | 07:31 | |
| *** mohan43u has joined #buildstream | 07:47 | |
| *** bathtub_shark has joined #buildstream | 08:37 | |
| *** tristan has joined #buildstream | 09:02 | |
| *** ChanServ sets mode: +o tristan | 09:25 | |
| tristan | cs-shadow, sent an email, was hoping you or Gökçen might have that group photo of the last BeaverCon so I can include that in my blog post :) | 09:27 |
| tristan | Or, did anyone else get a copy of the group photo ? | 09:27 |
| *** Sembei has joined #buildstream | 09:36 | |
| *** jonathanmaw has joined #buildstream | 09:39 | |
| *** phildawson has joined #buildstream | 09:44 | |
| *** phildawson has quit IRC | 09:51 | |
| *** phildawson has joined #buildstream | 10:02 | |
| *** tpollard has joined #buildstream | 10:05 | |
| *** paulsherwood has joined #buildstream | 10:22 | |
| *** WSalmon has joined #buildstream | 10:28 | |
| *** valentind has joined #buildstream | 10:32 | |
| *** lachlan has joined #buildstream | 10:32 | |
| *** nimish has quit IRC | 10:35 | |
| cs-shadow | tristan: Hi, late start here :) Gokcen isn't online just yet but he says he'll send those around later today | 10:53 |
| cs-shadow | I assume that should be fine? | 10:54 |
| *** xxpor has joined #buildstream | 11:01 | |
| tristan | cs-shadow, Yay \o/ | 11:04 |
| *** SotK_ has joined #buildstream | 11:05 | |
| *** lachlan has quit IRC | 11:17 | |
| *** tjf has joined #buildstream | 11:19 | |
| *** kapil___ has joined #buildstream | 11:21 | |
| *** raoul has joined #buildstream | 11:29 | |
| *** toscalix has joined #buildstream | 11:30 | |
| *** lachlan has joined #buildstream | 11:47 | |
| *** toscalix has quit IRC | 12:02 | |
| gitlab-br-bot | marge-bot123 closed issue #916 (loosen dev dependencies) on buildstream https://gitlab.com/BuildStream/buildstream/issues/916 | 12:11 |
| gitlab-br-bot | marge-bot123 merged MR !1189 (tristan/optional-coverage->master: Make coverage optional) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1189 | 12:11 |
| *** Kristine has joined #buildstream | 12:23 | |
| *** lachlan has quit IRC | 12:48 | |
| *** lachlan has joined #buildstream | 12:51 | |
| *** wgwz has joined #buildstream | 12:55 | |
| *** kzantow_ has joined #buildstream | 12:55 | |
| gitlab-br-bot | tristanvb opened MR !1203 (tristan/optional-coverage-1.2->bst-1.2: Coverage optional in 1.2.x) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1203 | 13:03 |
| tristan | cs-shadow, | 13:03 |
| tristan | gah | 13:03 |
| tristan | cs-shadow, After merging !1189 I found out that there is still a problem with optionalizing coverage, will have to fix this separately | 13:04 |
| cs-shadow | tristan: what did i break now? :) | 13:04 |
| tristan | it was only working *because* the dev-requirements.txt seems not to have been updated well | 13:04 |
| tristan | I guess the makefile should rebuild all .txt from .in every time, not only if something changes | 13:04 |
| cs-shadow | tristan: Yes, I mentioned it in https://gitlab.com/BuildStream/buildstream/merge_requests/1189#note_145593067 | 13:06 |
| cs-shadow | we can either make all targets PHONY or recommend to use `make -B` | 13:06 |
| tristan | problem is that tests/testutils/artifactshare.py *imports* pytest_cov | 13:07 |
| tristan | cs-shadow, Nah well, we reviewed it together, I broke it | 13:08 |
| tristan | partly because the .txt files were not updated, so now I have the habit of updating requirements with `rm -f requirements/*.txt && make -C requirements/` | 13:08 |
| *** coldtom has joined #buildstream | 13:08 | |
| tristan | the former is better | 13:08 |
| tristan | nobody has to remember anything :) | 13:09 |
| tristan | I will take care of fixing the coverage optionality tomorrow | 13:09 |
| tristan | will have to optionalize the manual interaction with pytest_cov | 13:09 |
| tristan | and update the dev-requirement.txt | 13:09 |
| *** tristan has quit IRC | 13:15 | |
| *** coldtom has quit IRC | 13:21 | |
| *** tpollard has quit IRC | 13:21 | |
| *** pro[m] has quit IRC | 13:21 | |
| *** cgmcintyre[m] has quit IRC | 13:21 | |
| *** awacheux[m] has quit IRC | 13:21 | |
| *** m_22[m] has quit IRC | 13:21 | |
| *** ssssam[m] has quit IRC | 13:21 | |
| *** mattiasb has quit IRC | 13:21 | |
| *** skullone[m] has quit IRC | 13:21 | |
| *** inigomartinez has quit IRC | 13:21 | |
| *** rafaelff[m] has quit IRC | 13:21 | |
| *** waltervargas[m] has quit IRC | 13:21 | |
| *** mrmcq2u[m] has quit IRC | 13:21 | |
| *** Demos[m] has quit IRC | 13:21 | |
| *** nielsdg has quit IRC | 13:21 | |
| *** slaf has quit IRC | 13:21 | |
| *** coldtom has joined #buildstream | 13:30 | |
| *** slaf has joined #buildstream | 13:30 | |
| *** ChanServ sets mode: -o slaf | 13:31 | |
| *** ChanServ changes topic to "BuildStream 1.2.4 is out ! | https://gitlab.com/BuildStream/buildstream | Docs: https://docs.buildstream.build/ | IRC logs: https://irclogs.baserock.org/buildstream | Mailing List: https://mail.gnome.org/mailman/listinfo/buildstream-list | Roadmap: https://wiki.gnome.org/Projects/BuildStream/Roadmaps" | 13:31 | |
| *** tpollard has joined #buildstream | 13:31 | |
| *** Eickmeyer has joined #buildstream | 13:41 | |
| *** tristan has joined #buildstream | 13:41 | |
| *** alphis has joined #buildstream | 13:42 | |
| *** hiei has joined #buildstream | 13:46 | |
| tpollard | juergbi: around? | 14:01 |
| juergbi | yes | 14:08 |
| *** _Penguin has joined #buildstream | 14:25 | |
| *** abderrahim[m] has joined #buildstream | 14:47 | |
| *** theawless[m] has joined #buildstream | 14:52 | |
| *** samouy has joined #buildstream | 14:54 | |
| *** dbuch has joined #buildstream | 14:55 | |
| *** efdee has joined #buildstream | 14:56 | |
| tpollard | juergbi: just doing some of the refactoring for cached_artifact | 14:57 |
| tpollard | *cache | 14:57 |
| tpollard | here https://gitlab.com/BuildStream/buildstream/blob/master/buildstream/element.py#L1690 we set collectvdir to None | 14:57 |
| tpollard | then right at the end, we do a conditional check which we have already entered in that try except block | 14:58 |
| tpollard | and throw an unhandled elementerror | 14:58 |
| tpollard | is that expected, or should we be raising it when we originally set it to none? | 14:58 |
| *** toscalix has joined #buildstream | 14:58 | |
| tpollard | if we're unable to collect the contents, I would expect that we want to halt straight away, whereas it seems we still go about committing it | 14:59 |
| juergbi | tpollard: collectvdir is only for the 'files' part | 15:01 |
| juergbi | when we cache a failed build, that will be None but we still want to cache the artifact | 15:01 |
| juergbi | or at least it might be None, if the install-root doesn't exist | 15:02 |
| tpollard | Ok, but then the error goes unhandled right? | 15:03 |
| *** alatiera has joined #buildstream | 15:04 | |
| juergbi | tpollard: it's handled/raised by that check right at the end | 15:05 |
| * tpollard is confused | 15:06 | |
| tpollard | which is not out of the ordinary | 15:06 |
| *** ChanServ sets mode: +o tristan | 15:06 | |
| tristan | isn't install-root a marked directory, and as such guaranteed to exist anyway ? | 15:06 |
| *** awacheux[m] has joined #buildstream | 15:07 | |
| juergbi | normally, yes, but there could be corner cases where it doesn't exit | 15:07 |
| juergbi | *exist | 15:07 |
| juergbi | e.g., I could imagine an element that uses read-write rootfs and then rmdirs install-root | 15:07 |
| juergbi | not extremely likely, of course | 15:07 |
| tristan | Seems like there is opportunity to make that more iron clad, and turn that error catching into an assertion instead ? | 15:08 |
| tpollard | so we'd cache the failed build, and then we'd error out? I can't see where that raised exception is excepted from a caller | 15:08 |
| tristan | tpollard, In general, every BstError is a user facing message | 15:10 |
| tristan | tpollard, which means either there is a system error (out of disk space ? network connectivity error ?), or the user did something wrong | 15:10 |
| tristan | tpollard, So there is no need to `except ElementError` almost ever, probably never | 15:10 |
| tristan | That is all handled by the highest level except BstError blocks | 15:11 |
| tristan | if an ArtifactError is raised and there is no need to add any additional context from the Element, then there is usually no need to handle it | 15:11 |
| tristan | (although that is a bit weird that the raised ArtifactError knows enough context to raise a fully contextual user facing error message) | 15:12 |
| tristan | But it can happen because we pass Elements to the ArtifactCache APIs for convenience reasons (even though it breaks vertical stacking order of modules) | 15:13 |
| tristan | Interestingly in the case of a missing install-root, the case seems a bit ambiguous, I would think that in almost any case that an install-root is missing, it is a programming error from the plugin python code or the core or something | 15:14 |
| tristan | but it is ambiguous, which is not so great | 15:14 |
| tristan | programming errors should be assertions, and it would be beneficial to have the API/rules reduce/remove such ambiguities | 15:15 |
| tristan | I think | 15:15 |
| *** mattiasb has joined #buildstream | 15:16 | |
| tpollard | Ok, I think I work with this for now then | 15:16 |
| juergbi | tpollard: fyi, tests/frontend/buildcheckout.py::test_install_to_build tests this code path | 15:20 |
| juergbi | (might also be unintentional, though, haven't looked into it) | 15:21 |
| juergbi | if anyone has some review cycles, !1190 should be ready | 15:22 |
| gitlab-br-bot | MR !1190: Improve import performance https://gitlab.com/BuildStream/buildstream/merge_requests/1190 | 15:22 |
| tpollard | I'll keep the exception at the end for now but raised from within element instead of artifact | 15:22 |
| juergbi | ok | 15:22 |
| tpollard | the tests did all pass though with me raising it before the artifact creation, so might be something to note | 15:22 |
| *** borbyu has joined #buildstream | 15:26 | |
| tristan | Hah, that could be a nice custom /leave message | 15:27 |
| *** Dynetrekk has joined #buildstream | 15:27 | |
| *** ssssam[m] has joined #buildstream | 15:37 | |
| *** alatiera has quit IRC | 15:37 | |
| *** alatiera has joined #buildstream | 15:38 | |
| *** m_22[m] has joined #buildstream | 15:38 | |
| *** cgmcintyre[m] has joined #buildstream | 15:45 | |
| *** dexter0 has joined #buildstream | 15:52 | |
| *** dexter0 has quit IRC | 15:52 | |
| *** Blastyr25 has joined #buildstream | 15:55 | |
| *** dpk has joined #buildstream | 16:13 | |
| *** SteveG has joined #buildstream | 16:13 | |
| *** SteveG has quit IRC | 16:14 | |
| *** zmo has joined #buildstream | 16:27 | |
| *** inigomartinez has joined #buildstream | 16:28 | |
| gitlab-br-bot | cs-shadow opened MR !1204 (chandan/always-regen-requirements->master: requirements/Makefile: Always rebuild requirements files) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1204 | 16:29 |
| cs-shadow | tristan: https://gitlab.com/BuildStream/buildstream/merge_requests/1204 for ensuring requirements files are always rebuilt | 16:29 |
| *** DHE has joined #buildstream | 16:30 | |
| gitlab-br-bot | jennis opened MR !1205 (jennis/compose_backwards->master: cleanup: Move yaml hanlding logic in _include.py to _yaml.py) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1205 | 16:31 |
| *** waltervargas[m] has joined #buildstream | 16:39 | |
| *** rafaelff[m] has joined #buildstream | 16:43 | |
| *** Lymkwi has joined #buildstream | 16:47 | |
| *** Exterminador has joined #buildstream | 16:48 | |
| tpollard | when running against master 'Invalid requirement: '!nocover: -rrequirements/cov-requirements.txt' | 16:51 |
| tpollard | It looks like a path. File '!nocover: -rrequirements/cov-requirements.txt' does not exist.' | 16:51 |
| tpollard | the file is there locally, but if I use --recreate or py36-nocover I still get the same error, any tips? | 16:51 |
| tpollard | anyone else having issues running the tests against master? | 16:56 |
| jmac | I'll try now | 16:56 |
| tpollard | seems fine in CI | 16:57 |
| tpollard | cheers jmac | 16:58 |
| tpollard | (I've also done a fresh clone of the repo) | 16:58 |
| jmac | I'm getting a load of errors because I don't have pytest_cov installed | 16:59 |
| jmac | Once I manually installed that, it appears fine | 17:00 |
| *** jonathanmaw has quit IRC | 17:02 | |
| juergbi | tpollard: is it possible you have an old version of tox installed? | 17:03 |
| *** toscalix has quit IRC | 17:03 | |
| tpollard | doesn't for me (also tried a reinstall of buildtsream via pip, and manualy install of pytest_cov via pip) | 17:03 |
| tpollard | I thought recreate would handle this but it doesn't want to play ball | 17:04 |
| jmac | Oh, I wasn't using tox, I'll try that next | 17:04 |
| tpollard | https://paste.gnome.org/pc4spbpqr is the error | 17:05 |
| *** bparker has joined #buildstream | 17:06 | |
| jmac | tpollard: Yes, I'm getting that exact error on master now I'm using tox | 17:06 |
| tpollard | thanks for checking jmac! | 17:07 |
| juergbi | tpollard: seems to work here. tox --version? | 17:08 |
| tpollard | 2.5.0 | 17:08 |
| *** pro[m] has joined #buildstream | 17:09 | |
| juergbi | I think that's too old | 17:09 |
| juergbi | WSL had this issue, iirc | 17:09 |
| juergbi | I have 3.6.1 here, which works | 17:09 |
| *** frangdlt has joined #buildstream | 17:12 | |
| tpollard | juergbi: seems I have 3.7 under pip, but my path is picking up 2.5 from apt I guess | 17:15 |
| tpollard | (Ubuntu 18.04) | 17:15 |
| *** nielsdg has joined #buildstream | 17:15 | |
| *** phildawson has quit IRC | 17:15 | |
| *** alexteves has joined #buildstream | 17:16 | |
| *** kapil___ has quit IRC | 17:17 | |
| *** phildawson has joined #buildstream | 17:17 | |
| *** nimish has joined #buildstream | 17:18 | |
| *** Demos[m] has joined #buildstream | 17:25 | |
| * tpollard removes his apt installed tox | 17:28 | |
| *** mrmcq2u[m] has joined #buildstream | 17:36 | |
| *** skullone[m] has joined #buildstream | 17:44 | |
| *** alatiera has quit IRC | 17:48 | |
| *** alatiera has joined #buildstream | 17:49 | |
| *** alatiera has quit IRC | 17:50 | |
| *** alatiera has joined #buildstream | 17:50 | |
| gitlab-br-bot | aevri approved MR !1204 (chandan/always-regen-requirements->master: requirements/Makefile: Always rebuild requirements files) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1204 | 18:19 |
| *** alatiera_ has joined #buildstream | 18:22 | |
| *** phildawson has quit IRC | 18:26 | |
| *** phildawson has joined #buildstream | 18:26 | |
| *** alatiera has quit IRC | 18:36 | |
| *** raoul_ has joined #buildstream | 18:46 | |
| *** raoul has quit IRC | 18:47 | |
| *** raoul_ has quit IRC | 18:49 | |
| *** alatiera has joined #buildstream | 18:51 | |
| *** raoul has joined #buildstream | 18:59 | |
| *** raoul has quit IRC | 19:05 | |
| *** raoul has joined #buildstream | 19:05 | |
| *** tristan has quit IRC | 19:26 | |
| *** cs-shadow has quit IRC | 19:37 | |
| *** raoul has quit IRC | 19:49 | |
| *** raoul has joined #buildstream | 19:49 | |
| *** phildawson has quit IRC | 19:50 | |
| *** swick has quit IRC | 19:53 | |
| *** raoul has quit IRC | 20:12 | |
| *** cs-shadow has joined #buildstream | 20:25 | |
| gitlab-br-bot | marge-bot123 merged MR !1204 (chandan/always-regen-requirements->master: requirements/Makefile: Always rebuild requirements files) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1204 | 20:36 |
| *** lachlan has quit IRC | 21:05 | |
| *** swick has joined #buildstream | 21:12 | |
| *** nimish has quit IRC | 21:12 | |
| *** alatiera has quit IRC | 22:59 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!