| 19 Sep 2025 |
 emily | fwiw this breaks .override interfaces (but I agree it's better overall) | 02:15:46 |
| emily | pkgs/games/maptool/default.nix → pkgs/games/maptool/package.nix doesn't look right | 02:16:07 |
 Tomodachi94 (they/them) | Oops, definitely not intended | 02:18:51 |
| Tomodachi94 (they/them) | Fixed in latest push | 02:21:19 |
| Tomodachi94 (they/them) | https://github.com/NixOS/nixpkgs/pull/443954#pullrequestreview-3242588635
Can't seem to find any consensus about the pattern of pname versus pname<version>, does anyone know anything? | 02:29:29 |
| emily | your proposed ? thing doesn't work | 02:57:31 |
| emily | callPackage will override it with pkgs.jdk | 02:57:36 |
| emily | you have to name it something else | 02:57:43 |
| emily | fwiw I don't think dependency overrides should count as public interface anyway but | 02:58:09 |
| Tomodachi94 (they/them) | Would be a good topic for an RFC | 03:35:19 |
| emily | ideally, you could define the interface that's exposed and the internal references separately | 03:39:30 |
| emily | package sets / multi-version packages and some wonky details about overrides are basically the only things stopping everything from moving to by-name I think | 03:40:55 |
 msgilligan | The Gradle 9.1.0 PR was merged! You can now build Java 25 apps (using either zulu25 or graalvm-ce) and Gradle using Nixpkgs master. (I assume Maven works fine because it is less sensitive to Java version bumps than Gradle, but I haven't tested it) | 14:30:22 |
| msgilligan | I'm watching the Temurin chimney and I think I see white smoke. | 14:58:07 |
| msgilligan | But we have a bunch of Temurin updates to the existing versions that should be made, I think. I ran the update python script and generated this PR: https://github.com/NixOS/nixpkgs/pull/443891 | 14:59:04 |
| msgilligan | Should that be split into multiple PRs (one for each version bump maybe) and targeted to staging? | 15:00:01 |
| msgilligan | * But we have a bunch of Temurin updates to the existing versions that should be made, I think. I ran the update python script and generated this (massive-rebuild-triggering) PR: https://github.com/NixOS/nixpkgs/pull/443891 | 15:00:47 |
| msgilligan | https://adoptium.net/temurin/releases/?version=25&os=any&arch=any | 15:04:04 |
| msgilligan | The website is updated but nothing is actually there except alpine x64 | 15:04:54 |
| msgilligan | Here's a status page:
https://github.com/adoptium/temurin/issues/96 | 15:07:34 |
| emily | one PR to staging is fine | 15:13:44 |
| emily | we're about to start a cycle in the next couple days | 15:13:47 |
| emily | any other mass rebuild Java PRs to land? | 15:13:59 |
| msgilligan | What is a "cycle"? | 15:14:36 |
 Alyssa Ross | https://github.com/NixOS/nixpkgs/blob/master/CONTRIBUTING.md#staging | 15:15:23 |
| emily | https://github.com/NixOS/nixpkgs/blob/master/CONTRIBUTING.md#staging | 15:15:27 |
| emily | ha | 15:15:27 |
| msgilligan | Read the fine manual 🤣 | 15:15:49 |
| emily | staging-next-25.05 is about to merge to release-25.05, so we will soon have staging-next cut from staging and start to build | 15:15:53 |
| msgilligan | OK PR 443891 is rebased to staging. (I rebased and force-pushed before switching the target branch with the GitHub PR UI and that confused some of the tests. But I force-pushed a second time and that seems to have reset everything) | 15:46:31 |
