| *** phoenix has quit IRC | 03:15 | |
| *** tristan has quit IRC | 05:41 | |
| *** traveltissues has quit IRC | 05:43 | |
| *** traveltissues has joined #buildstream | 05:43 | |
| gitlab-br-bot | tristanvb merged MR !1851 (valentindavid/include-options-from-junction->master: Process options in includes files with the options of their junction) on buildstream https://gitlab.com/BuildStream/buildstream/-/merge_requests/1851 | 06:02 |
|---|---|---|
| *** tristan has joined #buildstream | 06:02 | |
| *** ChanServ sets mode: +o tristan | 06:03 | |
| *** narispo has joined #buildstream | 06:31 | |
| *** narispo has joined #buildstream | 06:31 | |
| douglaswinship | juergbi: thanks for your help yesterday! | 06:55 |
| douglaswinship | I also found out that bst shell has a '--mount' option, that lets you do the same thing on a one-off basis, without having to edit project.conf | 06:56 |
| *** benschubert has joined #buildstream | 07:07 | |
| * tristan finds it mind boggling and kind of cool what fdsdk / g-b-m have been doing with filter elements | 07:08 | |
| tristan | See this one for instance: https://gitlab.gnome.org/GNOME/gnome-build-meta/-/blob/master/elements/sdk/gst-plugins-good.bst | 07:09 |
| tristan | We're not filtering here, but filter elements quite accidentally give you the opportunity to rewrite runtime dependencies | 07:09 |
| tristan | In this case, we've effectively axed the fdsdk builds of gdk-pixbuf, gst-plugins-base, GTK+3 and libsoup, and replaced them with elements built in gnome-build-meta | 07:10 |
| tristan | At this point depending on the filter element will result in having a gst-plugins-good which is runtime linked against a different set of the same libraries it was built against | 07:11 |
| juergbi | ah, interesting, haven't seen this either. it might be somewhat error prone, though. besides the possible dependency incompatibility, g-b-m might also forget updating the dependency list when the one in fdsdk gets modified | 07:13 |
| tristan | I think in this context it's fairly safe, given the ABI guarantees of fdsdk | 07:15 |
| tristan | But yes there is opportunity for that | 07:15 |
| juergbi | it guarantees it won't enable a plugin? | 07:16 |
| juergbi | I don't remember the details otoh | 07:16 |
| tristan | In stable, it must in some way do that yes | 07:16 |
| tristan | in master it cannot, but in stable because of the nature of runtimes/extensions/apps, the runtimes must retain backwards and forwards compatibility | 07:16 |
| juergbi | valentind: I think I've finally tracked down the FileNotFoundError (of a blob) on push after build. it can happen when using independent artifact and CAS servers and the artifact proto is on the artifact server but at least one blob of the artifact is missing from the CAS server | 07:17 |
| juergbi | I'll try to write a test case for this and fix it as part of !1843 | 07:17 |
| gitlab-br-bot | MR !1843: Fix handling of missing blobs in `ArtifactCache.pull()` https://gitlab.com/BuildStream/buildstream/-/merge_requests/1843 | 07:18 |
| juergbi | the bug is one of the two hard things in computer science: cache invalidation | 07:18 |
| douglaswinship | juergbi: well at least you don't have to name it. :) | 07:38 |
| *** rdale has joined #buildstream | 07:43 | |
| *** santi has joined #buildstream | 08:35 | |
| *** tpollard has joined #buildstream | 08:37 | |
| *** lachlan has joined #buildstream | 09:10 | |
| *** phoenix has joined #buildstream | 09:39 | |
| *** phoenix has quit IRC | 09:53 | |
| *** phoenix has joined #buildstream | 09:58 | |
| *** lachlan has quit IRC | 10:01 | |
| *** lachlan has joined #buildstream | 10:11 | |
| gitlab-br-bot | willsalmon opened MR !1866 (willsalmon/more_as_str->master: Adding as_str too Mapping nodes) on buildstream https://gitlab.com/BuildStream/buildstream/-/merge_requests/1866 | 10:14 |
| WSalmon | i feel like benschubert would have most context for ^ | 10:17 |
| *** lachlan has quit IRC | 10:18 | |
| *** lachlan has joined #buildstream | 10:19 | |
| *** tristan has quit IRC | 10:21 | |
| *** tristan has joined #buildstream | 10:34 | |
| *** ChanServ sets mode: +o tristan | 10:34 | |
| *** lachlan has quit IRC | 11:07 | |
| *** lachlan has joined #buildstream | 11:21 | |
| *** lachlan has quit IRC | 11:39 | |
| *** lachlan has joined #buildstream | 11:47 | |
| *** lachlan has quit IRC | 11:54 | |
| *** cs-shadow has joined #buildstream | 11:58 | |
| *** phoenix has quit IRC | 12:30 | |
| *** tristan has quit IRC | 12:59 | |
| *** santi has quit IRC | 13:07 | |
| *** santi has joined #buildstream | 13:08 | |
| *** lachlan has joined #buildstream | 13:13 | |
| *** lachlan has quit IRC | 13:38 | |
| *** lachlan has joined #buildstream | 13:40 | |
| *** lachlan has quit IRC | 13:53 | |
| *** lachlan has joined #buildstream | 14:18 | |
| jjardon | juergbi: yes, It's very fragile, but we could not find other solution | 14:23 |
| jjardon | any idea if we could do something more robust with buildstream and junctions? | 14:24 |
| jjardon | gnome-build-meta is not only building flatpaks so what tristan said about stability is not 100% true | 14:24 |
| benschubert | jjardon: the aim would be to replace some dependencies of an element in a junction by others in the top level project right? | 14:30 |
| *** tristan has joined #buildstream | 14:31 | |
| benschubert | it would require gbm to have the same structure, but including the element in yours? The name of the dependencies would be resolved in your context. You would not be able to remove any though | 14:33 |
| benschubert | unless you create empty elements for those ones | 14:33 |
| gitlab-br-bot | BenjaminSchubert approved MR !1866 (willsalmon/more_as_str->master: Adding as_str too Mapping nodes) on buildstream https://gitlab.com/BuildStream/buildstream/-/merge_requests/1866 | 14:37 |
| gitlab-br-bot | BenjaminSchubert unapproved MR !1866 (willsalmon/more_as_str->master: Adding as_str too Mapping nodes) on buildstream https://gitlab.com/BuildStream/buildstream/-/merge_requests/1866 | 14:37 |
| tpollard | jjardon: what are you trying to achieve, I presume I'm missing some backscroll here | 14:41 |
| gitlab-br-bot | BenjaminSchubert approved MR !1866 (willsalmon/more_as_str->master: Adding as_str too Mapping nodes) on buildstream https://gitlab.com/BuildStream/buildstream/-/merge_requests/1866 | 14:44 |
| coldtom | tpollard, changing the runtime-depends of an element from a junction iiuc, e.g. https://gitlab.gnome.org/GNOME/gnome-build-meta/-/blob/master/elements/sdk/gst-plugins-good.bst | 14:45 |
| *** lachlan has quit IRC | 14:47 | |
| *** lachlan has joined #buildstream | 14:47 | |
| tpollard | 'patch or overlay' I assume? | 14:51 |
| jjardon | runtime and build time dependencies | 14:51 |
| tpollard | or are wet talking about a junction going up the stack to change a top level element? | 14:51 |
| tpollard | s/wet/we | 14:51 |
| jjardon | benschubert: for example; gstreamer depends on GTK+, but GTK+ in fdsdk is different from the GTK+ in gnome-build-meta | 14:52 |
| jjardon | tpollard: ^ | 14:52 |
| *** lachlan has quit IRC | 14:52 | |
| tpollard | yep, gotcha | 14:53 |
| WSalmon | benschubert, you happy for me to merge that MR, i havnt merged anything here in ages so cant rember the eticat | 14:53 |
| WSalmon | ? | 14:53 |
| benschubert | WSalmon: go ahead (assign to Marge that is :D) | 14:53 |
| WSalmon | Is the CI completly broken today? | 14:55 |
| *** tpollard has quit IRC | 14:57 | |
| *** tpollard has joined #buildstream | 14:57 | |
| gitlab-br-bot | valentindavid opened MR !1867 (valentindavid/bst-1/retry-cas-calls->bst-1: Handle grpc errors of type UNAVAILABLE and ABORTED) on buildstream https://gitlab.com/BuildStream/buildstream/-/merge_requests/1867 | 14:58 |
| *** lachlan has joined #buildstream | 14:59 | |
| *** lachlan has quit IRC | 15:02 | |
| *** lachlan has joined #buildstream | 15:02 | |
| *** lachlan has quit IRC | 15:14 | |
| *** santi has quit IRC | 15:53 | |
| *** santi has joined #buildstream | 15:54 | |
| *** phoenix has joined #buildstream | 16:34 | |
| *** phoenix has quit IRC | 16:36 | |
| *** santi has quit IRC | 16:36 | |
| *** santi has joined #buildstream | 16:37 | |
| *** phoenix has joined #buildstream | 16:39 | |
| *** santi has quit IRC | 17:34 | |
| *** toscalix has joined #buildstream | 17:58 | |
| *** tpollard has quit IRC | 18:19 | |
| *** tpollard has joined #buildstream | 18:29 | |
| *** tpollard has quit IRC | 18:34 | |
| *** benschubert has quit IRC | 19:11 | |
| *** tpollard has joined #buildstream | 19:15 | |
| *** tpollard has quit IRC | 19:45 | |
| *** santi has joined #buildstream | 19:55 | |
| *** toscalix has quit IRC | 20:23 | |
| *** cs-shadow has quit IRC | 20:51 | |
| *** santi has quit IRC | 21:41 | |
| *** rdale has quit IRC | 21:43 | |
| *** phoenix has quit IRC | 23:15 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!