| *** narispo has quit IRC | 07:16 | |
| *** narispo has joined #buildstream | 07:16 | |
| *** samkirkham has joined #buildstream | 07:36 | |
| *** bochecha has joined #buildstream | 07:46 | |
| *** tristan has joined #buildstream | 08:01 | |
| *** rdale has joined #buildstream | 08:06 | |
| *** ChanServ sets mode: +o tristan | 08:22 | |
| *** toscalix has joined #buildstream | 08:23 | |
| *** tiagogomes has joined #buildstream | 08:50 | |
| *** flatmush_ is now known as flatmush | 08:50 | |
| *** traveltissues has joined #buildstream | 09:20 | |
| gitlab-br-bot | marge-bot123 closed issue #1107 (Add option to pass integration commands through filter elements) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1107 | 09:21 |
|---|---|---|
| gitlab-br-bot | marge-bot123 closed issue #1104 (Filter elements don't work on stacks) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1104 | 09:21 |
| gitlab-br-bot | marge-bot123 merged MR !1593 (coldtom/filter-element-improvements->master: plugins/elements/filter.py: Allow passing integration commands through, fail if dep is a stack element) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1593 | 09:21 |
| jjardon | persia: can you help getting the bst 1.4.1 update into Debian? We would really like it to be included in Ubuntu 19.10 which will be released soon | 09:24 |
| gitlab-br-bot | coldtom opened issue #1137 (Display whether an element builds artifacts when showing if it's buildable) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1137 | 09:28 |
| *** jonathanmaw has joined #buildstream | 09:29 | |
| *** benbrown has quit IRC | 09:54 | |
| *** benbrown has joined #buildstream | 09:54 | |
| *** bochecha has quit IRC | 10:17 | |
| *** phildawson_ has quit IRC | 10:45 | |
| *** phildawson_ has joined #buildstream | 10:45 | |
| *** ahmed89 has quit IRC | 10:56 | |
| * tlater[m] wonders if it's intentional that gitlab shows nothing under the "further information" heading here: https://gitlab.com/BuildStream/buildstream/blob/881257d4aa0ce77afbb57b00d9513334d0ee4b57/CONTRIBUTING.rst | 11:05 | |
| tlater[m] | I guess we'll need to create a manual toctree there, and somehow comment it out for sphinx? | 11:07 |
| *** ahmed89 has joined #buildstream | 11:37 | |
| *** phildawson_ has quit IRC | 11:38 | |
| *** bochecha has joined #buildstream | 12:09 | |
| gitlab-br-bot | tlater opened (was WIP) MR !1584 (jennis/frontend_documentation->master: Add UI section in CONTRIBUTING) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1584 | 12:12 |
| *** ahmed89 has quit IRC | 12:16 | |
| gitlab-br-bot | nanonyme opened issue #1138 (BuildStream 1.4 not handling relative git URL's properly in submodules) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1138 | 12:37 |
| *** phildawson_ has joined #buildstream | 12:41 | |
| gitlab-br-bot | tlater opened issue #1139 (Add UI/UX design principles to contributing documentation) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1139 | 12:52 |
| Kinnison | Is it not rendered as part of our documentation build? | 12:54 |
| tlater[m] | Kinnison: it is, but not when you open the document on gitlab as an innocent repository reader | 12:54 |
| tlater[m] | sphinx/gitlab use different link formats, basically. | 12:55 |
| * Kinnison shrugs, if you expect that kind of thing to work always, you're gonna be unlucky | 12:55 | |
| * Kinnison generally only expects on-git{lab,hub} rendering to work for README.md | 12:55 | |
| tlater[m] | Yep | 12:56 |
| tlater[m] | Since CONTRIBUTING.md is prominently displayed in our root directory on gitlab, I'd expect it to work there. | 12:56 |
| Kinnison | Oh, I wouldn't | 12:56 |
| Kinnison | :D | 12:56 |
| Kinnison | I literally only expect it for README because that gets auto-rendered | 12:57 |
| tlater[m] | Hrm | 12:57 |
| tristan | Errrm, more than that... we renamed it from the traditional HACKING filename *because* gitlab has a feature which specifically recognizes a CONTRIBUTING filename and links to it from the main page | 12:57 |
| tlater[m] | But there's a link at the top of the page and everything | 12:58 |
| benschubert | Is it even rendering for you? Gitlab gives up as the file is too big :/ | 12:58 |
| * tlater[m] needs to run, but thinks we should fix that | 12:58 | |
| tlater[m] | Worst case in favor of gitlab over sphinx | 12:58 |
| tlater[m] | benschubert: We made it a lot smaller | 12:58 |
| tlater[m] | The problem is it links to docs now, and those links only work in sphinx | 12:59 |
| benschubert | tlater[m]: oh :) Well, sphinx is still useful... | 12:59 |
| tristan | I think that if it's rendered in the documentation and if that's problematic, we can just make the documentation link to it | 12:59 |
| tristan | the contributing guide anyway is a single state of the union, where documentation is specific to a given release | 13:00 |
| persia | jjardon: What sort of help do you seek? | 13:10 |
| gitlab-br-bot | marge-bot123 merged MR !1584 (jennis/frontend_documentation->master: Add UI section in CONTRIBUTING) on buildstream https://gitlab.com/BuildStream/buildstream/merge_requests/1584 | 13:19 |
| jjardon | persia: someone with debian packaging knowledge to propose a MR to https://salsa.debian.org/gnome-team/buildstream | 13:20 |
| persia | jjardon: Maybe? Looking at that repo, there seem to be some conventions used to determine how to split the commits with which I'm not familiar, along with the construction of a repo that contains the code, which I've historically considered poor practice for packaging (so I suspect there are other ways in which my knowledge and the culture of that team may need reconcilliation). | 13:25 |
| *** tristan has quit IRC | 13:30 | |
| bochecha | jjardon: did you try asking the Debian-GNOME team? they have a channel on OFTC | 13:32 |
| jjardon | bochecha: no, thanks for the suggestion. I guess they are busy upgrading all the GNOME packages at the moment | 13:36 |
| bochecha | probably yeah | 13:36 |
| bochecha | also, since Ubuntu 19.10 releases soon, they might already have frozen the automatic imports from Debian | 13:36 |
| bochecha | which means if you want it in Ubuntu, then you should ask some Ubuntu dev to update it there | 13:36 |
| jjardon | :( | 13:36 |
| * bochecha asls | 13:36 | |
| jjardon | bochecha: jbicha is an Ubuntu dev, so hopefully he can do it | 13:37 |
| persia | Indeed. Feature Freeze for 19.10 was 22 August. Procedures may have changed, but I believe manual updates are permitted until 26 September with only minor administrative overhead. | 13:38 |
| persia | (my memory was that a developer had to inspect the change and assert it was primarily a bugfix upload and/or that any new features introduced were less risky than the bugs fixed) | 13:39 |
| bochecha | and buildstream is in universe anyway, so it shouldn't be too hard to get it updated before the release | 13:39 |
| persia | Best to ask jbicha :) | 13:47 |
| *** narispo has quit IRC | 13:48 | |
| persia | I'll still try to understand, and if more hands are needed, try to help, but I'm a bit out of touch for Ubuntu dev, and was never part of the GNOME team in Debian (who have done things slightly differently for a very long time) | 13:48 |
| *** narispo has joined #buildstream | 13:49 | |
| *** tiagogomes has quit IRC | 13:51 | |
| *** narispo has quit IRC | 13:53 | |
| *** narispo has joined #buildstream | 13:54 | |
| *** tristan has joined #buildstream | 13:56 | |
| *** narispo has quit IRC | 14:00 | |
| *** narispo has joined #buildstream | 14:00 | |
| coldtom | Can someone take a look at bst-plugins-experimental!34 please? | 14:07 |
| *** tpollard has quit IRC | 14:23 | |
| *** tpollard has joined #buildstream | 14:26 | |
| jjardon | tlater[m]: Hey, now that https://gitlab.com/BuildStream/buildstream/merge_requests/1540 is implemented; can we close https://gitlab.com/BuildStream/buildstream/issues/1050 or is there anything left to do? | 14:30 |
| tlater[m] | jjardon: I think so, it should work from the BuildStream side | 14:32 |
| tlater[m] | Though someone should look into whether it's actually possible from any of the other caches. | 14:33 |
| gitlab-br-bot | jjardon closed issue #1050 (Not possible to use generic servers implementing the remote-apis as remote cache) on buildstream https://gitlab.com/BuildStream/buildstream/issues/1050 | 14:33 |
| tlater[m] | Heh, thanks for making the executive decision there ;) | 14:34 |
| jjardon | tlater[m]: cool! thanks. I will close and open separate issues if we found any problem | 14:34 |
| jjardon | tlater[m]: I guess this was only tested with BuildGrid ? | 14:34 |
| tlater[m] | jjardon: Yup | 14:40 |
| jjardon | ok, thanks! | 14:40 |
| juergbi | jjardon: I've noticed one leftover piece that might be an issue in master but should be fixed in !1601 | 14:41 |
| gitlab-br-bot | MR !1601: WIP: Do not directly communicate with CAS server https://gitlab.com/BuildStream/buildstream/merge_requests/1601 | 14:41 |
| jjardon | juergbi: thanks; I guess this makes this feature to depend on casd? (ie, very difficult to backport to bst-1.x)? | 14:43 |
| juergbi | jjardon: no, not really, I just stumbled upon it while doing casd-related work | 14:44 |
| juergbi | 1540 might require a bit of an effort to backport in general but the fix in 1601 should be simple enough. https://gitlab.com/BuildStream/buildstream/merge_requests/1601/diffs?commit_id=09b0b8152b6bbfb14e821b883f60bcfc2b60e0e8 | 14:45 |
| jjardon | ok, thanks juergbi | 14:46 |
| juergbi | could also move this commit to a separate MR if you need it urgently in master | 14:46 |
| jjardon | juergbi: no worries, not urgent | 15:09 |
| *** tiagogomes has joined #buildstream | 15:55 | |
| *** bochecha has quit IRC | 16:34 | |
| *** toscalix has quit IRC | 16:43 | |
| * tlater[m] wonders why junction meta elements are loaded with deps when they are never allowed to specify deps | 16:46 | |
| tlater[m] | If that's just an oversight things may become a lot simpler... | 16:46 |
| juergbi | tlater[m]: that sounds like an oversight | 16:54 |
| tlater[m] | juergbi: Might be something like needing to load deps so that we know later down the line and can tell the user that that's not allowed | 16:56 |
| tlater[m] | Haven't experimented enough to tell yet | 16:56 |
| tlater[m] | Ah, yes, indeed | 16:57 |
| tlater[m] | Test to assert that failure works fails the moment I switch methods | 16:57 |
| tlater[m] | Grmbl | 16:58 |
| tlater[m] | That means we actually do need to load the deps and that in turn means I should probably show a task counter for this | 16:59 |
| * tlater[m] wonders if it would make sense to short-circuit that specifically for junctions | 17:00 | |
| tlater[m] | Or maybe add some plugin API to say "this plugin is not allowed to have dependencies" | 17:00 |
| juergbi | tlater[m]: we already have BST_FORBID_BDEPENDS and BST_FORBID_RDEPENDS which we both set for junctions | 17:03 |
| tlater[m] | I suppose those are just queried a lot later than they could be | 17:03 |
| tlater[m] | Presumably because we don't know what the element's `kind` is until after we've loaded it | 17:04 |
| juergbi | ah, yes, that's after meta, of course | 17:04 |
| tlater[m] | I suppose I'll want to take a look at how feasible it is to get at that information before meta | 17:07 |
| tlater[m] | Or well, before we load the meta element | 17:07 |
| *** cs-shadow has joined #buildstream | 17:10 | |
| *** thinkl33t has joined #buildstream | 17:29 | |
| *** phoenix has joined #buildstream | 17:55 | |
| *** traveltissues has quit IRC | 18:24 | |
| *** phoenix has quit IRC | 18:55 | |
| benschubert | tlater[m]: we should still probably have a counter, loading a junction takes... minuites at least for any thing reasonably sized | 19:30 |
| juergbi | benschubert: fetching subproject should already go via scheduler, though | 19:34 |
| benschubert | juergbi: ah then I misunderstood tlater[m]'s comments | 19:35 |
| *** rdale has quit IRC | 19:39 | |
| *** thinkl33t has quit IRC | 19:48 | |
| *** jonathanmaw has quit IRC | 19:58 | |
| tlater[m] | benschubert: this is specifically for loading metaelements for a junction's metaelement dependencies, which should be 0, or we should error out | 20:11 |
| *** narispo has quit IRC | 23:06 | |
| *** narispo has joined #buildstream | 23:06 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!