| 22 Nov 2024 |
 K900 | I really, REALLY want gtk2 gone at this point | 08:33:45 |
 eliasp | looking at the Sentry reports at https://crash-reports.kde.org/ I can't find any reports with os.name = NixOS - there should be at least some, since I had a few crashes recently and wanted to see, whether there are other reports of those as well or whether they're NixOS specific...
does anyone have some insight how kcrash handles all this, so I could dig into this and find out, why no reports are sent? | 09:02:00 |
| eliasp | ...and it would be also interesting to know, whether any of the kcrash tooling has a way to "simulate" crash to test all this? | 09:07:49 |
| K900 | We don't have enough debug info to submit those currently | 09:09:29 |
| K900 | The next staging cycle should get us there | 09:09:37 |
| eliasp | ah, because my backtraces looked quite complete, so I assumed they'd be good enough to be sent…
nonetheless, then the UI is misleading and doesn't tell me "Your crash data are bad - I'm not gonna send anything to KDE" :) | 09:10:50 |
| eliasp | 
Download Screenshot_20241122_101122.png | 09:12:09 |
| eliasp | and I just found a few quite insightful blog posts by Harald:
- https://planet.kde.org/harald-sitter-2024-08-20-plasma-crash-course-kcrash/
- https://planet.kde.org/harald-sitter-2024-08-27-plasma-crash-course-coredumpd/
- https://planet.kde.org/harald-sitter-2024-09-04-plasma-crash-course-drkonqi/
| 09:13:07 |
|  Morgan (@numinit) joined the room. | 17:51:31 |
| 24 Nov 2024 |
|  @ixlun:matrix.org changed their display name from Matthew Leach to Matthew L. | 00:43:18 |
| K900 | @emily can you take a look at what we're actually breaking with qtwebengine5 | 08:14:53 |
 emily | like if we tried to remove it now? | 08:15:09 |
| emily | or if we don't get it fixed on Darwin for 24.11? | 08:15:21 |
| emily | the hacky fix for Darwin is pretty simple so we can probably just go ahead with that for now | 08:15:32 |
| emily | but it does seem like it's better to get rid of it sooner rather than later for 25.05 | 08:15:43 |
| K900 | Like if we remove and/or knownVulnerabilities it | 08:16:03 |
| emily | I can try and just delete it and see what nixpkgs-review thinks | 08:16:09 |
| K900 | It will drag half of Plasma 5 to heck, that's for sure | 08:16:12 |
| emily | tomorrow probably | 08:16:14 |
| emily | maybe you should drop Plasma 5 firts :P | 08:16:22 |
| K900 | But the good thing about that is we also want to get rid of Plasma 5 | 08:16:24 |
| emily | * maybe you should drop Plasma 5 first :P | 08:16:25 |
| K900 | Yes | 08:16:26 |
| emily | I wish you could easily grep for Qt/KDE package names and reliably narrow it down to specific versions | 08:16:46 |
| K900 | I wish we didn't have two of them to begin with | 08:17:23 |
| emily | scrolling through rg -g '!pkgs/kde' qtwebengine and guessing at what's using 5/6, it really doesn't look too bad | 08:19:19 |
| emily | okay lol Calamares doesn't seem good | 08:20:05 |
| emily | and RStudio seems painful | 08:20:13 |
| emily | looks like Calamares has Qt 6 support of some kind? | 08:20:34 |
| K900 | I'm not sure how mature that is but I can ask adg | 08:21:11 |
