| *** hasebastian has joined #buildstream | 01:43 | |
| *** dftxbs3e has quit IRC | 03:45 | |
| *** jjardon has quit IRC | 03:45 | |
| *** pointswaves has quit IRC | 03:45 | |
| *** WSalmon has quit IRC | 03:45 | |
| *** jward has quit IRC | 03:45 | |
| *** robjh has quit IRC | 03:45 | |
| *** douglaswinship has quit IRC | 03:45 | |
| *** SamThursfield[m] has quit IRC | 03:45 | |
| *** abderrahim[m] has quit IRC | 03:45 | |
| *** doras has quit IRC | 03:45 | |
| *** cgm[m] has quit IRC | 03:45 | |
| *** krichter[m] has quit IRC | 03:45 | |
| *** awacheux[m] has quit IRC | 03:45 | |
| *** asingh_[m] has quit IRC | 03:45 | |
| *** theawless[m] has quit IRC | 03:45 | |
| *** walterve[m][m] has quit IRC | 03:45 | |
| *** tlater[m] has quit IRC | 03:45 | |
| *** ironfoot has quit IRC | 03:45 | |
| *** juergbi has quit IRC | 03:45 | |
| *** dftxbs3e_ has joined #buildstream | 03:46 | |
| *** jjardon has joined #buildstream | 03:46 | |
| *** doras has joined #buildstream | 03:46 | |
| *** pointswaves has joined #buildstream | 03:46 | |
| *** SamThursfield[m] has joined #buildstream | 03:46 | |
| *** jward has joined #buildstream | 03:46 | |
| *** abderrahim[m] has joined #buildstream | 03:46 | |
| *** douglaswinship has joined #buildstream | 03:46 | |
| *** WSalmon has joined #buildstream | 03:46 | |
| *** robjh has joined #buildstream | 03:46 | |
| *** krichter[m] has joined #buildstream | 03:46 | |
| *** cgm[m] has joined #buildstream | 03:46 | |
| *** awacheux[m] has joined #buildstream | 03:46 | |
| *** walterve[m][m] has joined #buildstream | 03:46 | |
| *** asingh_[m] has joined #buildstream | 03:46 | |
| *** tlater[m] has joined #buildstream | 03:46 | |
| *** theawless[m] has joined #buildstream | 03:46 | |
| *** ironfoot has joined #buildstream | 03:46 | |
| *** irc.acc.umu.se sets mode: +oo jjardon ironfoot | 03:46 | |
| *** Frazer61 has quit IRC | 05:02 | |
| *** juergbi has joined #buildstream | 07:11 | |
| *** benschubert has joined #buildstream | 07:24 | |
| WSalmon | https://gitlab.com/BuildStream/buildstream/-/merge_requests/1978 this is exciting juergbi sander :D cc cphang coldtom | 07:54 |
|---|---|---|
| *** tristan has quit IRC | 07:58 | |
| jjardon | douglaswinship: it worked! :) | 08:27 |
| jjardon | juergbi: when https://gitlab.com/BuildStream/buildstream/-/merge_requests/1978 get merged there will not be server implementation (yet) for the remote cache, right? | 08:28 |
| juergbi | jjardon: no external one, correct. the setup will be pretty much identical as it is right now. bst-artifact-server will support the new protocol | 08:30 |
| juergbi | also see my comment: https://gitlab.com/BuildStream/buildstream/-/merge_requests/1978#note_374769711 | 08:30 |
| jjardon | ok, thanks! | 08:32 |
| *** santi has joined #buildstream | 08:38 | |
| *** hasebastian has quit IRC | 08:40 | |
| *** hasebastian has joined #buildstream | 08:41 | |
| douglaswinship | jjardon: excellent! | 08:47 |
| douglaswinship | I think that thoroughly validates https://gitlab.com/BuildStream/bst-plugins-experimental/-/merge_requests/125 | 08:51 |
| douglaswinship | It's already been approved (thanks benschubert). Can anyone merge? | 08:52 |
| benschubert | Sure | 08:52 |
| benschubert | douglaswinship: set to merge on CI completion | 08:53 |
| *** tristan has joined #buildstream | 08:58 | |
| *** ChanServ sets mode: +o tristan | 08:58 | |
| coldtom | ooh nice douglaswinship | 08:59 |
| douglaswinship | any idea what's up with the runners? That CI pipeline's been scheduled for 15 hours, and not one job has started | 08:59 |
| benschubert | try cancel and restarting it? | 08:59 |
| douglaswinship | done (twice), and still none of the jobs have started | 09:06 |
| douglaswinship | (plus I expect I'm now further back in the queue) | 09:06 |
| douglaswinship | Who manages these runners, and would poke them if they stopped working? | 09:06 |
| douglaswinship | jjardon's pipeline seems to be stuck too | 09:08 |
| juergbi | jjardon: the group runner bastion-debian10-c-2-4gib-ams3-01 is offline for 22 hours, do you know why? | 09:26 |
| juergbi | douglaswinship: that's the reason, afaik | 09:26 |
| juergbi | BuildStream core has another runner assigned (bastion-debiain-9), so it's not immediately blocked | 09:27 |
| jjardon | juergbi: no idea, let me check | 09:27 |
| jjardon | juergbi: mmm, seems one of the bastions is gone; I will setup a new one | 09:29 |
| juergbi | ta | 09:30 |
| douglaswinship | jjardon: thanks! | 09:30 |
| douglaswinship | I'd love to learn how all of this is managed, and participate, at some point. It's still all magic to me. | 09:31 |
| *** hasebastian has quit IRC | 09:32 | |
| jjardon | douglaswinship: happy to share with you; It's actually pretty simple. Let me setup then I will explain | 09:32 |
| douglaswinship | awesome! | 09:34 |
| WSalmon | standup in 5 min douglaswinship coldtom WSalmon | 09:39 |
| douglaswinship | How long will we continue these standups btw? until the end of the week? | 09:40 |
| douglaswinship | oh, sorry everyone. Those last 2 comments are in the wrong window | 09:43 |
| WSalmon | oh no | 09:43 |
| WSalmon | w/w | 09:43 |
| WSalmon | so sorry | 09:43 |
| *** tristan has quit IRC | 10:54 | |
| *** tristan has joined #buildstream | 11:35 | |
| *** ChanServ sets mode: +o tristan | 11:35 | |
| *** Frazer has joined #buildstream | 11:40 | |
| Frazer | hi, does anyone know a list of projects that use BuildStream, potentially outside of GitHub and not freedesktop-sdk? | 11:42 |
| coldtom | Frazer, gnome-build-meta, libreml, carbonOS (no idea how active this is, someone mentioned it on a freedesktop-sdk MR once), the webkit flatpak SDK, a bunch of stuff under the celduin namespace in gitlab all use bst | 11:52 |
| coldtom | probably a bunch of personal projects around too | 11:53 |
| *** hasebastian has joined #buildstream | 11:58 | |
| *** cphang has joined #buildstream | 12:18 | |
| Frazer | thanks coldtom, linguist need a large enough list of projects on github to count it and say its not enough yet, hopefully with these it will be | 12:34 |
| tristan | Take that Py_UNICODE ! | 13:15 |
| tristan | Bah ! | 13:15 |
| benschubert | tristan: ah? :D | 13:16 |
| tristan | benschubert, With lots of painful printing of octal values and fiddling, I've got a malloc + memcpy loop replacement for "".join() | 13:16 |
| tristan | Now who knows if it makes *any* difference at all performance wise | 13:17 |
| tristan | but it *should* not involve any copies or encoding/decoding, that said; I don't know for sure | 13:17 |
| tristan | It cost me dearly | 13:21 |
| tristan | My 'r' key | 13:21 |
| tristan | I will be missing that 'r' | 13:21 |
| * douglaswinship mourns the 'r' key | 13:27 | |
| * douglaswinship lays flowes on its gave. | 13:27 | |
| douglaswinship | jjardon: any progress on poking the runners for bst-plugins-experimental? Your pipeline and my pipeline are both still showing no progress at all. | 13:29 |
| tristan | ridiculously more expensive :'( | 13:42 |
| WSalmon | tristan, in bst-master can i not have the same named junction through a junction? when i try to link to the junctioned junction it just says that its the wrong type... | 13:43 |
| WSalmon | https://hastebin.com/jifepofiwo.md | 13:43 |
| tristan | WSalmon, that should work | 13:46 |
| jjardon | douglaswinship: busy day, sorry; I will take a look after work | 13:47 |
| tristan | WSalmon, this should demonstrate working links: https://gitlab.com/BuildStream/buildstream/-/tree/master/tests/format/link/simple-junctions/ | 13:54 |
| tristan | Maybe there's a bug somewhere... | 13:54 |
| * tristan getting kicked out of closing coffee shop | 13:55 | |
| WSalmon | https://gitlab.com/BuildStream/buildstream/-/blob/master/tests/format/link/simple-junctions/elements/subsubproject-link.bst right but these are diffrent names, in real projects you keep refuring to the base project by the same project, eg my suppor_project and final_project both want to call FD freedesktop-sdk.bst | 13:57 |
| *** tristan has quit IRC | 13:58 | |
| *** tristan has joined #buildstream | 14:12 | |
| *** ChanServ sets mode: +o tristan | 14:12 | |
| tristan | WSalmon, in your paste, bst-boardsupport.bst is a local junction to a project which has a junction element called freedesktop-sdk.bst correct ? | 14:22 |
| tristan | and `bst show bst-boardsupport.bst:freedesktop-sdk.bst` shows that as a valid junction element ? | 14:22 |
| tristan | https://hastebin.com/jifepofiwo.md lacks a bit of context :-S | 14:23 |
| WSalmon | tristan, that is all correct | 14:28 |
| WSalmon | tristan, https://gitlab.com/celduin/bsps/bst-boardsupport/-/commits/willsalmon/multi_vim_bst2_plus in that branch if you cd sample/rpi4-features | 14:32 |
| tristan | WSalmon, I will investigate, looking at that it all looks like it should work :-S | 14:41 |
| *** dylan-m has joined #buildstream | 17:42 | |
| dylan-m | Hey, has anyone played with running unit tests for an element inside buildstream? I ended up making this and it works, using the include directive to copy the regular element definition but adding `make check` as an install step: https://github.com/endlessm/endless-sdk-flatpak/blob/master/elements/tests/sdk/eos-metrics.bst. So our CI system does `bst build tests/sdk/eos-metrics.bst` and the like. But it isn't perfect, particularly | 17:42 |
| dylan-m | because if I have a workspace open for an element, I need to open a _separate_ workspace (pointing at the same code) to test the thing. I'm curious what other folks are doing here :) | 17:42 |
| philn | i kind-of abuse integration-commands, not to run unittests, but to check my recipe is minimally usable, so if it ships a command, i run it there for instance | 17:48 |
| philn | not sure this was the initial purpose :) | 17:48 |
| abderrahim[m] | dylan-m: there is a plan to implement something like that, but it will be for bst 2.x | 17:49 |
| dylan-m | abderrahim: Ooh, okay. So would that be like a test environment and commands for an element? | 17:52 |
| dylan-m | I'm definitely super happy about running our tests in buildstream - it was a bit of churn getting an x server and dbus in the sandbox, but it feels really nice having reproducible tests :) | 17:53 |
| abderrahim[m] | Not really well defined yet | 17:54 |
| * abderrahim[m] looks for the issue | 17:54 | |
| dylan-m | abderrahim: Ah, must be these two: | 17:58 |
| dylan-m | https://gitlab.com/BuildStream/buildstream/-/issues/776 | 17:58 |
| dylan-m | https://gitlab.com/BuildStream/buildstream/-/issues/777 | 17:58 |
| * dylan-m subscribes to them :) | 17:58 | |
| *** hasebastian has quit IRC | 18:09 | |
| *** santi has quit IRC | 18:16 | |
| cphang | dylan-m the libreML project has done this in a rudimentary fashion. See https://gitlab.com/libreml/libreml/-/blob/master/elements/python-modules/pandas.bst | 18:46 |
| *** Trevinho has quit IRC | 22:42 | |
| *** benschubert has quit IRC | 22:43 | |
| *** benschubert has joined #buildstream | 22:43 | |
| *** Trevinho has joined #buildstream | 22:43 | |
| *** Trevinho has quit IRC | 22:46 | |
| *** benschubert has quit IRC | 22:48 | |
| *** lchlan has quit IRC | 22:50 | |
| *** jjardon has quit IRC | 22:50 | |
Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!