| 26 Apr 2021 |
 _hc | #fosshost:matrix.org | 12:55:40 |
 proletarius101 | In reply to @eighthave:matrix.org
its not a pure static site, Apache automatically handles localizations Cdn would be a much better solution | 12:55:49 |
| _hc | we're approved, they just need to finish setting up | 12:56:04 |
| proletarius101 | In reply to @eighthave:matrix.org
we're approved, they just need to finish setting up I'll keep an eye on it then | 12:56:42 |
 @rdfg77:kde.org | In reply to @proletarius101:matrix.org
linsui not sure if you are familiar with nekohasegai, but I can't find a way to contact the upstram for this new app https://gitlab.com/fdroid/rfp/-/issues/1719 I only know that they are the developer of NekoX. I thought you can ping them in the issue directly. | 13:22:17 |
| _hc | izzy: I fixed issuebot, it posts again | 14:29:21 |
 izzy | Great, thanks a lot! What was causing it btw? | 14:53:17 |
| _hc | I messed up the issuebot.get_modules() function when I moved it to be tested. the path where it was running the glob was off by one | 15:04:15 |
| _hc | so it returned no modules | 15:04:23 |
| _hc | my test case set it up properly, but not in real world | 15:04:39 |
| _hc | so I added a test case for just that | 15:04:47 |
| izzy | <so it returned no modules> Ah, that explains, yes! Why must reality always mess with our nice theories #D Thanks for making it work again! I'm going on to hammer the open MRs then (1 down, 127 to go =:-0) | 17:43:50 |
|  @freenode_foobar48:matrix.org joined the room. | 20:05:04 |
| 27 Apr 2021 |
|  jochensp1 joined the room. | 01:09:11 |
| jochensp1 left the room. | 01:11:49 |
 Passiing | Was there a talk on why there are so many glosseries? | 08:41:57 |
| _hc | stencil: you mean in Weblate? | 08:56:36 |
 jochensp | _hc: could we keep build logs of disabled builds on the repo? Currently they are removed here: https://gitlab.com/fdroid/fdroidserver/-/blob/master/fdroidserver/update.py#L440 | 11:12:07 |
| jochensp | maybe only remove them if the build entry is removed? | 11:12:38 |
| jochensp | context: we usually disable failed builds so they don't get retried and take up build time but then the log is gone and the app author can't see the problem | 11:14:25 |
| jochensp | maybe we could do that in update.scan_repo_files? I thought about reworking that already to output the build queue, would be great to talk to you about that before I start digging | 11:19:39 |
| jochensp | (queue would rather be update.main() but same area of code) | 11:27:32 |
| jochensp | ..also, maybe it makes more sense to have this all in buildbot | 11:27:57 |
| _hc | I agree that's something we need to improve on, but I don't have a clear idea of the issues and possible pitfalls. maybe that's something we could discuss at the meeting? I know uniq has thought about that area of the processs more than I | 11:48:50 |
| _hc | he implemented the build log support, for example | 11:48:59 |
| _hc | that would change the conception of how the repo files are managed | 11:49:31 |
| _hc | since the beginning, disabled: removed all the files for that entry | 11:49:52 |
| _hc | AFAIK | 11:50:01 |
| _hc | its also been used as a way to remove everything related to a release | 11:50:40 |
| _hc | honestly, that seems like something better solved by moving to buildbot, it would give us that for free | 11:51:14 |
