| *** tristan has quit IRC | 00:33 | |
| *** jswagner has quit IRC | 01:24 | |
| *** tristan has joined #buildstream | 01:35 | |
| *** traveltissues has joined #buildstream | 06:14 | |
| *** traveltissues has quit IRC | 06:26 | |
| benschubert | jjardon: https://gitlab.com/BuildStream/buildstream/-/jobs/313985671 for the overnight :D I'll clean everything up and let you know which branch to clone :) | 06:56 |
|---|---|---|
| jjardon | OMG, amazing! :) | 06:58 |
| juergbi | \o/ | 06:58 |
| laurence | very cool | 07:00 |
| *** toscalix has joined #buildstream | 07:04 | |
| coldtom | \o/ | 07:04 |
| *** ChanServ sets mode: +o tristan | 07:37 | |
| tristan | woohoooooo ! | 07:37 |
| benschubert | jjardon: https://gitlab.com/freedesktop-sdk/freedesktop-sdk/commits/bschubert/freedesktop-sdk-19.08.3-bst2 would be branch, mind taking ownership of it and/or protecting it in gitlab? I'll make a PR to buildstream as soon as we decide on the exact name of the branch :) | 08:22 |
| *** traveltissues has joined #buildstream | 08:25 | |
| gitlab-br-bot | traveltissues approved MR !1623 (aevri/no_win32_fcntl->master: cli.py: no fcntl on Windows) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1623 | 08:30 |
| gitlab-br-bot | traveltissues approved MR !1627 (bschubert/fix-xdg-temppaths->master: setup.cfg: Fix xdg env vars to not affect the host) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1627 | 08:36 |
| gitlab-br-bot | traveltissues approved MR !1629 (bschubert/fix-windows-symlink->master: tests/frontend/buildcheckout.py: Create required symlinks on the fly) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1629 | 08:37 |
| benschubert | tristan: while you are here, mind having a look at https://gitlab.com/BuildStream/buildstream/merge_requests/1627? | 08:39 |
| benschubert | https://gitlab.com/BuildStream/bst-plugins-experimental/merge_requests/45 < for fixing the overnight tests | 08:40 |
| *** jonathanmaw has joined #buildstream | 08:41 | |
| tristan | benschubert, made some comments, not sure I understand how that patch changes anything | 08:55 |
| tristan | of course I prefer using the regular APIs for setting env vars, rather than inserting more custom voodoo into conftest.py | 08:55 |
| benschubert | tristan: the problem is that the current state is that you pollute the host system when calling "setup.py test" or "pytest", as 'D:...' only set the variable if it's not set. On any mahcine that has xdg, it would be set and you would hence pollute the host :) | 08:56 |
| benschubert | Doing it through conftest ensure that however you run the tests, we use either the 'BST_TEST_*' locations, which is an explicit action, or set it in the pytest tmp directory | 08:57 |
| tristan | benschubert, isnt there a syntax for unconditionally setting it in tox.ini ? | 08:57 |
| benschubert | in tox yes | 08:57 |
| benschubert | but neither setup.py nor pytest use tox | 08:57 |
| tristan | Ah so this is just to fix setup.py/setup.cfg, for running outside of tox ? | 08:58 |
| benschubert | correct | 08:58 |
| benschubert | the tox behavior doesn't change with this fix | 08:58 |
| tristan | And the problem is that if we set this in setup.cfg explicitly, then tox.ini cannot override it, because it will in turn invoke setup.py | 08:59 |
| benschubert | correct | 08:59 |
| tristan | It does look quite contorted | 09:00 |
| tristan | benschubert, and it seems to be missing explicit settings of D:BST_TEST_XDG_* in setup.cfg to replace the old thing doesnt it ? | 09:00 |
| benschubert | correct, and this is in order to allow users to override --basetemp, and thus have their temp files in the basetemp of their choosing | 09:01 |
| tristan | Hmmmm | 09:01 |
| tristan | And when they forget to add --basetemp, it will just use host tmp dir ? that would be bad | 09:02 |
| tristan | basetemp is in the domain of setup.cfg, not tox ? | 09:02 |
| benschubert | No, we set --basetemp in setup.cfg to be ./tmp | 09:02 |
| benschubert | yep it's pytest's option | 09:02 |
| tristan | I see | 09:02 |
| benschubert | but ./tmp doesn't make sense in some cases (when mounted in a docker container for example), and this change allows us to finally override it completely :) | 09:03 |
| tristan | That makes sense too | 09:04 |
| tristan | I've always assumed that you have write access to the checkout | 09:05 |
| benschubert | you do most of the time, but in my case, accessing something in the checkout means: docker -> ubuntu host -> samba -> windows host and back, which slows down my tests quite a bit :'D | 09:06 |
| benschubert | anyways, with the changes, the defaults won't change, and we would now respect the --basetemp override :) | 09:06 |
| benschubert | thanks! | 09:07 |
| tristan | no worries, I'm typing from a rather bureaucratic meeting | 09:09 |
| tristan | box ticking through issue checklists, silly stuff | 09:09 |
| benschubert | have fun with that! | 09:09 |
| tristan | hah | 09:09 |
| *** lachlan has joined #buildstream | 09:11 | |
| gitlab-br-bot | aevri opened (was WIP) MR !1624 (aevri/platform_win32->master: win32: _platform/win32: add support for win32) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1624 | 09:34 |
| gitlab-br-bot | marge-bot123 merged MR !1623 (aevri/no_win32_fcntl->master: cli.py: no fcntl on Windows) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1623 | 09:38 |
| *** traveltissues has quit IRC | 09:50 | |
| *** lachlan has quit IRC | 09:54 | |
| *** tristan has quit IRC | 10:02 | |
| gitlab-br-bot | BenjaminSchubert approved MR !1624 (aevri/platform_win32->master: win32: _platform/win32: add support for win32) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1624 | 10:04 |
| *** lachlan has joined #buildstream | 10:06 | |
| benschubert | Hey, could someone make a new release for https://gitlab.com/BuildStream/bst-plugins-experimental ? | 10:23 |
| *** traveltissues has joined #buildstream | 10:33 | |
| *** lachlan has quit IRC | 10:35 | |
| gitlab-br-bot | BenjaminSchubert opened MR !1630 (bschubert/fix-empty-buildtree->master: _artifactcache.py: Don't push artifact blobs when no files are present) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1630 | 10:40 |
| gitlab-br-bot | marge-bot123 merged MR !1627 (bschubert/fix-xdg-temppaths->master: setup.cfg: Fix xdg env vars to not affect the host) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1627 | 10:49 |
| *** lachlan has joined #buildstream | 10:50 | |
| benschubert | tlater[m], traveltissues is https://gitlab.com/BuildStream/buildstream/merge_requests/1620#note_227362273 fine with you ? | 10:51 |
| traveltissues | lgtm | 10:51 |
| tlater[m] | Also fine by me benschubert :) | 10:53 |
| benschubert | awesome, to marge then, thanks! | 10:53 |
| * tlater[m] wonders what caused the variance | 10:53 | |
| tlater[m] | Probably some update process or somesuch - I assume you ran it under WSL? | 10:53 |
| benschubert | it was on docker on windows | 10:54 |
| benschubert | and it wasn't my benchmark machine | 10:54 |
| benschubert | so many things running :) | 10:54 |
| tlater[m] | Yeah, that kind of variance seems very reasonable then. | 10:55 |
| traveltissues | benschubert, are you ok with !1628 | 10:55 |
| gitlab-br-bot | MR !1628: Defer workspace source commits https://gitlab.com/BuildStream/buildstream/merge_requests/1628 | 10:55 |
| gitlab-br-bot | BenjaminSchubert approved MR !1628 (traveltissues/1159->master: Defer workspace source commits) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1628 | 10:56 |
| gitlab-br-bot | tlater opened (was WIP) MR !1608 (tlater/progress-tests->master: Improve assertions around element loading progress reporting) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1608 | 10:56 |
| benschubert | traveltissues: yup thanks! | 10:56 |
| traveltissues | thanks | 10:57 |
| gitlab-br-bot | BenjaminSchubert approved MR !1608 (tlater/progress-tests->master: Improve assertions around element loading progress reporting) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1608 | 10:59 |
| tlater[m] | Ha, ta benschubert - I'll wait for another pair of eyes, I don't want to sign off an MR that says "bug free" | 10:59 |
| benschubert | I had one nit though ;) | 11:00 |
| *** traveltissues has left #buildstream | 11:01 | |
| tlater[m] | Ah, it's a good nit too | 11:01 |
| *** traveltissues has joined #buildstream | 11:01 | |
| *** lachlan has quit IRC | 11:03 | |
| *** toscalix has quit IRC | 11:04 | |
| *** phildawson_ has joined #buildstream | 11:04 | |
| *** phil has quit IRC | 11:05 | |
| gitlab-br-bot | tmewett opened issue #1162 (git and git_tag sources don't support recursive submodules) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1162 | 11:17 |
| *** narispo has quit IRC | 11:18 | |
| *** traveltissues has quit IRC | 11:26 | |
| *** traveltissues has joined #buildstream | 11:26 | |
| *** lachlan has joined #buildstream | 11:54 | |
| *** phildawson_ has quit IRC | 12:05 | |
| *** lachlan has quit IRC | 12:18 | |
| gitlab-br-bot | jonathanmaw approved MR !1608 (tlater/progress-tests->master: Improve assertions around element loading progress reporting) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1608 | 12:32 |
| benschubert | traveltissues: about your xfail comment, I wonder whether we should not instead use https://docs.pytest.org/en/latest/reference.html#confval-xfail_strict? | 12:56 |
| *** traveltissues has quit IRC | 13:09 | |
| *** traveltissues has joined #buildstream | 13:09 | |
| traveltissues | benschubert, i didn't know about that, yes that's probably sensible | 13:10 |
| traveltissues | but at the same time adding the strict keyword doesn't hurt | 13:11 |
| gitlab-br-bot | jonathanmaw opened issue #1163 (Add CI for native windows support) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1163 | 13:11 |
| *** traveltissues has quit IRC | 13:14 | |
| *** traveltissues has joined #buildstream | 13:14 | |
| *** lachlan has joined #buildstream | 13:26 | |
| *** lachlan has quit IRC | 13:28 | |
| benschubert | traveltissues: right | 13:29 |
| *** lachlan has joined #buildstream | 13:31 | |
| *** rdale has quit IRC | 13:32 | |
| *** phildawson_ has joined #buildstream | 13:33 | |
| *** rdale has joined #buildstream | 13:34 | |
| *** traveltissues has quit IRC | 13:57 | |
| *** traveltissues has joined #buildstream | 13:58 | |
| tlater[m] | Ergh, looks like new gobject-introspection breaks buildstream 1.4's ostree plugin | 13:59 |
| tlater[m] | Yup, API for this changed: https://lazka.github.io/pgi-docs/#OSTree-1.0/classes/Repo.html#OSTree.Repo.remote_gpg_import | 14:07 |
| tlater[m] | BuildStream invokes `repo.remote_gpg_import(remote, stream, None, 0, None)` | 14:07 |
| tlater[m] | But whatever that 0 was has disappeared from the API | 14:07 |
| * tlater[m] wonders how we should handle this | 14:08 | |
| tlater[m] | Clearly the upstream API changed without a version change | 14:08 |
| tlater[m] | So we can't really detect it | 14:08 |
| tlater[m] | Weird try/catch bullshit? | 14:09 |
| * tlater[m] is annoyed at gobject introspection :| | 14:09 | |
| coldtom | tlater[m | 14:11 |
| coldtom | isn't that from ostree | 14:11 |
| coldtom | changing their API, not gobject-introspection | 14:12 |
| tlater[m] | Yeah | 14:13 |
| tlater[m] | It's ostree's gobject introspection API stuff | 14:13 |
| tlater[m] | Oh, I suppose | 14:13 |
| * tlater[m] wonders if there's a way to walk back through their API | 14:14 | |
| tlater[m] | To see what version we're running, and what we should be expecting | 14:14 |
| *** traveltissues has quit IRC | 14:26 | |
| *** traveltissues has joined #buildstream | 14:26 | |
| tlater[m] | coldtom: Looks like whatever that was was never in ostree's actual API | 14:27 |
| tlater[m] | There have only been 4 changes to that function in the history of the project, and none of them removed an argument | 14:27 |
| gitlab-br-bot | tlater opened issue #1164 (OSTree plugin fails with: `TypeError: OSTree.Repo.remote_gpg_import() takes exactly 5 arguments (6 given)`) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1164 | 14:41 |
| traveltissues | is marge ok? it doesn't seem to be collecting jobs | 14:59 |
| * tlater[m] swears he didn't break her | 15:55 | |
| *** traveltissues has quit IRC | 16:11 | |
| benschubert | ^ seems marge got stuck on a stuck pipeline :( | 16:26 |
| benschubert | I just restarted it | 16:26 |
| gitlab-br-bot | marge-bot123 merged MR !1628 (traveltissues/1159->master: Defer workspace source commits) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1628 | 16:33 |
| gitlab-br-bot | marge-bot123 closed issue #1159 (Follow-up from "element.py: remove call to _source_cached()" -- test_workspace_visible changed to XFAIL) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1159 | 16:33 |
| *** lachlan has quit IRC | 16:56 | |
| *** jonathanmaw has quit IRC | 17:01 | |
| *** lachlan has joined #buildstream | 17:12 | |
| *** lachlan has quit IRC | 17:14 | |
| gitlab-br-bot | marge-bot123 merged MR !1629 (bschubert/fix-windows-symlink->master: tests/frontend/buildcheckout.py: Create required symlinks on the fly) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1629 | 17:15 |
| gitlab-br-bot | marge-bot123 closed issue #1157 (Handle unexpected buildbox-casd termination in asyncio event loop) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1157 | 18:12 |
| gitlab-br-bot | marge-bot123 merged MR !1620 (bschubert/casd-listen-failures->master: Listen for casd failures and abort the run when they happen) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1620 | 18:12 |
| gitlab-br-bot | marge-bot123 closed issue #1145 (When building an artifact with no build-tree generated, an exception is thrown) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1145 | 18:53 |
| gitlab-br-bot | marge-bot123 merged MR !1630 (bschubert/fix-empty-buildtree->master: _artifactcache.py: Don't push artifact blobs when no files are present) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1630 | 18:53 |
| gitlab-br-bot | BenjaminSchubert opened MR !1634 (bschubert/fuse-permissions->master: _fuse/mount.py: Monitor the fuse process while waiting for the mount) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1634 | 19:11 |
| *** tristan has joined #buildstream | 20:24 | |
| jjardon | benschubert: branch name is freedesktop-sdk-19.08.3-bst2 | 21:50 |
| gitlab-br-bot | jjardon opened issue #1165 (Arch support should not be hardcoded in the tool) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1165 | 22:22 |
| *** narispo has joined #buildstream | 22:57 | |
| *** tristan has quit IRC | 23:06 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!