| 30 Jul 2025 |
 maralorn | A big change typed on my stupid phone. | 12:24:23 |
 dana | big change - the default ghc upgrade from 429212? | 12:25:03 |
| maralorn | Regarding staging-next I think that branch gets created fresh at some point the PR that you are looking for will be staging-next -> master. | 12:26:42 |
| maralorn | * | 12:27:22 |
 Alyssa Ross | It never gets recreated — staging just gets merged into it when it's time for a new cycle. | 12:33:19 |
| dana | In reply to @qyliss:fairydust.space
It never gets recreated — staging just gets merged into it when it's time for a new cycle. How can I find commits/PRs that correspond to this? | 12:35:33 |
|  Marc Jakobi joined the room. | 12:37:24 |
| Alyssa Ross | git log --merges --grep="staging.*staging-next" origin/staging-next would probably be reasonably accurate | 12:37:33 |
| Alyssa Ross | Next staging-next won't be for at least a week, because we just started staging-next-25.05 | 12:38:11 |
| dana | You mean this one, right? https://github.com/NixOS/nixpkgs/pull/429274
And no more commits will be added to it, only fixes? | 12:41:14 |
| Alyssa Ross | I think you're confused — you asked how to find merges of staging into staging-next. That's a PR to merge staging-next into master. New commits can still be added to it, as long as they don't cause mass rebuilds on Hydra. | 12:48:40 |
| dana | Ah, I see. I guess I thought staging-next is created alongside a PR to release/master and only fixes are merged into it. Thanks! | 12:52:40 |
| Alyssa Ross | Staging is merged into staging-next, and then a staging-next to master PR is opened. New commits can still be added to staging-next after that as long as they don't cause rebuilds, or, rarely, if the rebuilds are judged to be worth it. | 12:54:52 |
|  @dmjio:matrix.org joined the room. | 18:06:36 |
| 31 Jul 2025 |
|  cartocor joined the room. | 08:41:04 |
|  Sammy (It/Its) left the room. | 09:47:03 |
 sterni (he/him) | First proper eval of Stackage LTS 24 / ghc 9.10.2 https://hydra.nixos.org/eval/1817278?compare=1817170 | 11:21:36 |
 Pierre Thierry | I'm kephas on Github | 13:51:39 |
| 1 Aug 2025 |
|  _low_ changed their display name from low to _low_. | 21:06:04 |
| 2 Aug 2025 |
 magic_rb | is there a way to get a more minimal haskell bin out of nixpkgs? basic things end up being 3GB in closure size because they pull in GHC through primitive. I've yet to try static builds but that takes ages so isnt really scalable | 10:37:25 |
| maralorn | @magic_rb:matrix.redalder.org: Have a look at justStaticExecutables in the nixpkgs manual. | 11:17:00 |
| magic_rb | thanks | 11:18:12 |
| maralorn | Even nicer but minimally more complicated ist enableSeparateBinOutput. | 11:19:16 |
| magic_rb | oh my so much better | 11:21:32 |
| magic_rb | 3GB to 64MB | 11:21:36 |
| magic_rb | thanks a lot | 11:21:37 |
| magic_rb | nice, enableSeparateBinOutput also just works | 11:22:47 |
| magic_rb | this makes haskell acceptable for the usecase i wanted. (reimplementing environment.etc and i dont feel like doing it with a shell script, rather gonna write a haskell binary) | 11:23:20 |
 emily | In reply to @magic_rb:matrix.redalder.org
this makes haskell acceptable for the usecase i wanted. (reimplementing environment.etc and i dont feel like doing it with a shell script, rather gonna write a haskell binary) you know about system.etc.overlay right? | 11:25:11 |
| magic_rb | this is not for nixos | 11:25:18 |
