| 8 Jan 2025 |
 raboof | that's fair. it's fairly mundane code though, and only executed when doing updates | 16:05:06 |
| raboof | probably mainly outside. my first approach was indeed to package it outside of nixpkgs, but then I ran into the fact that there's quite some infrastructure inside of nixpkgs that is not easy to re-use outside (without resorting to copy-pasting), which is how I ended up doing it inside instead. | 16:09:55 |
| raboof | having the binary package already in nixpkgs might also be convenient when we start packaging the source-built jdk24, as it looks like we currently bootstrap source-building of Java X with the corresponding binary Java X | 16:10:15 |
| 11 Jan 2025 |
|  j4m3s joined the room. | 01:19:10 |
| 15 Jan 2025 |
|  SomeoneSerge (utc+3) changed their display name from SomeoneSerge (utc+3) to SomeoneSerge. | 19:02:27 |
| 16 Jan 2025 |
| j4m3s set their display name to j4m3s. | 10:27:58 |
| 22 Jan 2025 |
|  samasaur joined the room. | 19:54:23 |
| samasaur | I'm not sure whether this is exactly the right place to put it, but I recently opened a PR attempting to standardize outputs of the JDKs in nixpkgs on darwin. It makes them produce a "JDK bundle" at $out/Library/Java/JavaVirtualMachines/${name}-${version}.jdk (mirroring the install location for JDKs on macOS), and provides a bundle attribute that points to this path. Having this consistency will make it a lot easier to install JDKs properly on darwin. I'd love to hear people's thoughts! | 19:57:48 |
| 31 Jan 2025 |
| SomeoneSerge (utc+3) changed their display name from SomeoneSerge to SomeoneSerge (Bruxelles). | 19:34:48 |
| raboof changed their display name from raboof to raboof@FOSDEM. | 23:54:20 |
| 2 Feb 2025 |
 Tomodachi94 (they/them) | In reply to@samasaur:matrix.org
I'm not sure whether this is exactly the right place to put it, but I recently opened a PR attempting to standardize outputs of the JDKs in nixpkgs on darwin. It makes them produce a "JDK bundle" at $out/Library/Java/JavaVirtualMachines/${name}-${version}.jdk (mirroring the install location for JDKs on macOS), and provides a bundle attribute that points to this path. Having this consistency will make it a lot easier to install JDKs properly on darwin. I'd love to hear people's thoughts! This sounds like an awesome idea, I'll take a look | 03:18:09 |
| Tomodachi94 (they/them) | I currently don't have access to a NixOS machine though, so I'm limited to just reviewing diffs (this is why I've been so inactive this month) | 03:18:35 |
| Tomodachi94 (they/them) | This is totally up @emily's alley too (darwin and Java) | 03:32:54 |
| samasaur | yeah i was actually talking with emily about it when i did the work originally! they get credit for the idea of putting JDK bundles at the path that mirrors the install location | 03:44:43 |
| samasaur | i'll request their review too | 03:48:49 |
| samasaur | and I just force-pushed | 03:49:19 |
| samasaur | As best I can tell, the PR doesn't introduce any new breakages | 03:49:39 |
| Tomodachi94 (they/them) | I've been meaning to mark a bunch of those Java packages as broken since I did some Ant refactors way back in November/early December, could be a fun project for anyone wanting to do Java stuff | 04:49:46 |
| Tomodachi94 (they/them) | (and potentially make the build reports of some of my PRs more appealing, which is totally unrelated /s 🙃) | 04:50:38 |
|  gmodena joined the room. | 10:54:57 |
| samasaur | funnily enough i actually realized while doing this that I barely use Java anymore | 18:41:55 |
| samasaur | I still keep four JDK versions installed on my mac though | 18:42:12 |
| samasaur | and now I'm committed to getting this queue of changes in | 18:42:38 |
| raboof changed their display name from raboof@FOSDEM to raboof. | 22:18:45 |
| 3 Feb 2025 |
|  Lorenz Leutgeb changed their display name from Lorenz Leutgeb 📞6343 to Lorenz Leutgeb. | 08:53:22 |
| SomeoneSerge (utc+3) changed their display name from SomeoneSerge (Bruxelles) to SomeoneSerge (Gand St. Pieters). | 13:40:23 |
| 4 Feb 2025 |
| samasaur | I'm reaching the limit of my Nix-fu on this PR. I ran a nixpkgs-review run, which gave me a list of broken packages. Then I fed them into hydra-check to eliminate those that were already broken even though they weren't marked broken. But I'm left with a very long list of packages that allegedly were broken by my PR, even though when I test packages from that list locally they don't build before or after the PR | 06:33:31 |
| samasaur | is there some strategy I'm missing to track down which packages, if any, were actually broken by this PR? | 06:33:53 |
 FliegendeWurst | In reply to @samasaur:matrix.org
I'm reaching the limit of my Nix-fu on this PR. I ran a nixpkgs-review run, which gave me a list of broken packages. Then I fed them into hydra-check to eliminate those that were already broken even though they weren't marked broken. But I'm left with a very long list of packages that allegedly were broken by my PR, even though when I test packages from that list locally they don't build before or after the PR By "before your PR", do you mean current master? Or the commit you branched off from master (potentially much older)? | 06:37:26 |
| samasaur | well FWIW I did just rebase two days ago so the difference isn't going to be all that much | 06:42:34 |
