| 8 Dec 2025 |
 dramforever | i don't think so | 17:05:11 |
| dramforever | i can try to mess with sandbox-build-dir though | 17:08:35 |
| dramforever | maybe it's some cgroup and auto-allocate-uids stuff? | 17:18:34 |
 Mic92 | no. | 17:29:44 |
| Mic92 | no to what? | 17:37:20 |
| Mic92 | Probably should be merged before the release: https://github.com/NixOS/nix/pull/14739 | 20:16:00 |
| Mic92 | https://github.com/NixOS/nix/pull/14743 and now this one as well | 22:27:09 |
| 9 Dec 2025 |
 niksnut | Release notes PR, please add anything relevant that's missing: https://github.com/NixOS/nix/pull/14752 | 15:57:43 |
 teto | I cant edit but for #8871 (teto) it's a PR that makes nix flake show keep going even when an attribute in "packages" is not a derivation. ie., it lists all attributes and names the ones that are not derivations (it previously didn't so you have to check which one was not a derivation) | 16:31:35 |
 John Ericson | Sergei Zimmerman (xokdvium): what exactly did you find with 2.4 and nulls? | 16:56:00 |
| John Ericson | my guess that it never parsed the nulls, but also back then it wasn't making a data structure | 16:56:20 |
| John Ericson | it was just "shotgun parsing" in LocalDerivationGoal | 16:56:30 |
| John Ericson | was there an error? | 16:56:36 |
 Sergei Zimmerman (xokdvium) | In reply to @Ericson2314:matrix.org
my guess that it never parsed the nulls, but also back then it wasn't making a data structure It also failed to build the derivation with a message like expected a list | 16:56:53 |
| John Ericson | huh ok | 16:57:08 |
| Sergei Zimmerman (xokdvium) | derivation-show also | 16:57:16 |
| John Ericson | so Nixpkgs has just been wrong this whole time? | 16:57:17 |
| John Ericson | nix derivation show made an error? | 16:57:47 |
| Sergei Zimmerman (xokdvium) | In reply to @Ericson2314:matrix.org
so Nixpkgs has just been wrong this whole time? I guess so. I can double check since a was doing that in a rush this morning, but I tested 2.4’s show-derivation and building too. | 16:58:14 |
| John Ericson | I would not expect 2.4 to make an error about this since the parsing is in the builder | 16:58:50 |
| John Ericson | but maybe you just meant you checked what the derivation looked like? | 16:59:04 |
| John Ericson | (in case it was a nix language -> parsing issue?) | 16:59:24 |
 Taeer Bar-Yam | It seems like nixpkgs has been "wrong", but for some reason that I still can't figure out, it does work what they're doing (in the case of __structuredAttrs = false;) | 17:08:03 |
| John Ericson | that null becomes the empty string? | 17:08:55 |
| John Ericson | so structured attrs true was always broken, but not structured attrs false? | 17:09:17 |
| Taeer Bar-Yam | No. null becomes empty string in my own example that i cooked up, not using inputDerivation. In that case it fails to build.
inputDerivation somehow gets rid of allowedReferences, despite setting it to null, and it succeeds to build
| 17:09:54 |
| Taeer Bar-Yam | I think so | 17:10:12 |
| John Ericson | OK so I was fixing the wrong branch :) | 17:10:35 |
| Taeer Bar-Yam | honestly both branches seem broken | 17:11:19 |
| John Ericson | code branch | 17:11:23 |
