| 27 Aug 2025 |
|  @nullcube:matrix.org left the room. | 12:14:29 |
| 29 Aug 2025 |
 K900 | So does anyone want to do more deleting around https://github.com/NixOS/nixpkgs/pull/435812 | 10:18:08 |
 Grimmauld (any/all) | oh hey K900 more eval errors wrt fcitx! | 14:02:15 |
| Grimmauld (any/all) | oh wait just unfortunate timing of unstable-small advance, sorry | 14:03:08 |
| Grimmauld (any/all) | * ~~oh hey K900 more eval errors wrt fcitx!~~ | 14:04:18 |
| Grimmauld (any/all) | i guess i go straight to master then | 14:04:35 |
| K900 | Huh | 14:05:08 |
| Grimmauld (any/all) | wait what the hell is going on | 14:05:51 |
| Grimmauld (any/all) | i did nix flake update, but it did not give me up-to-date nixpkgs | 14:06:06 |
| Grimmauld (any/all) | OH GOD github timeouted me | 14:06:16 |
| Grimmauld (any/all) | yeah no disregard, not your fault | 14:06:31 |
| Grimmauld (any/all) | huh no that wasn't it... Seems i have some flake that pins nixpkgs instead of doign a follows=, this is pain | 14:07:56 |
| Grimmauld (any/all) | oh no, its homemanager that is exploding | 14:14:27 |
| Grimmauld (any/all) | ugh | 14:14:28 |
| Grimmauld (any/all) | HM was a mistake. | 14:15:25 |
 aidalgol | KMyMoney is broken because alkimia has an alias that throws with a message about it being removed.
error: The libsForQt5.alkimia package and the corresponding top-level alkimia alias have been removed, as KDE Gear 5 and Plasma 5 have reached end of life.
But there's an alkimia in top-level/all-packages.nix that uses kdePackages.callPackage, and the alkimia nix source appears to be building against Qt6. So is this alias incorrectly shadowing this definition, or do I need to reference it differently?
| 22:13:51 |
 emily | aliases will shadow all-packages | 22:35:56 |
| emily | where is the alias defined? | 22:36:01 |
| aidalgol | https://github.com/NixOS/nixpkgs/blob/425e45b2a10122f752d44b3fd2e7bc778daf1ff3/pkgs/top-level/aliases.nix#L56 | 23:01:22 |
 Sandro 🐧 | Found one breakage on one of my systems due to the Plasma 5 removals https://github.com/NixOS/nixpkgs/pull/438418 | 23:03:33 |
| emily | wow, I'm surprised this hasn't caught more stuff | 23:09:07 |
| emily | can you try making the throws filtered on !(builtins.hasAttr k super)? | 23:09:53 |
| emily | and also move plasma5Throws into the mapAliases call perhaps? | 23:10:12 |
| emily | which would have caught this | 23:10:14 |
| aidalgol | It looks like it should already be getting filtered. mapAliases is invoked on that giant attrset, which is merged with plasma5Throws, and mapAliases already has that checkInPkgs filter. So I think we should figure out why that's not catching it before moving things around. | 23:17:57 |
| emily | no | 23:45:02 |
| emily | mapAliases { … } // … is (mapAliases { … }) // … | 23:45:16 |
| emily | and the check only causes errors, doesn't filter | 23:45:21 |
| emily | so the fix is mapAliases ({ … } // …) except that might cause a horrible reindentation so maybe just add mapAliases to plasma5Throws, but more importantly it needs a filter. or just dropping the problematic packages from the list I guess | 23:45:49 |
| 30 Aug 2025 |
| Sandro 🐧 | Has anyone a good idea how to do https://github.com/NixOS/nixpkgs/issues/409986#issuecomment-3217982330 by default so that dolphin works better for everyone?
| 01:47:21 |
