| 12 Jan 2025 |
 emily | nixpkgs-review is a live service game :) | 19:55:16 |
 Gaétan Lepage | I would be in favor of a backport too.
The changes are "breaking" in a way as we have made the tool fetch GHA results (instead of Ofborg's). But as emily said, the older version is now obsolete. | 19:56:55 |
 llakala | In reply to @niklaskorz:korz.dev
I guess there are not that many people reviewing packages while running stable anyhow I'd love to see a Stable team at some point, since this is a common issue and only gets more common later into the release cycle | 23:29:20 |
| llakala | * I'd love to see a Stable team at some point to help resolve this, since this is a common issue and only gets more common later into the release cycle | 23:29:30 |
 @sigmasquadron:matrix.org | Does NixOS care about lib/modules-load.d/ in systemd.packages`? | 23:43:19 |
| @sigmasquadron:matrix.org | * Does NixOS care about lib/modules-load.d/ in systemd.packages? | 23:43:28 |
 Perchun Pak | is it possible to remove an attribute in an overlay?
cleanAttrs = lib.flip removeAttrs [ ... ];
overlay = final: prev: ... # how to call cleanAttrs here?
| 23:44:46 |
| 13 Jan 2025 |
 pveierland | Has the public key of the nixos cache ever changed? Or is this an extremely unlikely event? | 01:57:37 |
 hexa | not to my knowledge | 01:58:40 |
| hexa | https://github.com/NixOS/rfcs/pull/149 | 01:59:45 |
| pveierland | What does the -1 part of the cache.nixos.org-1:<public_key_base64> name signify in the substituter config? | 02:02:23 |
| pveierland | It just seems like something added to have a unique key name | 02:14:02 |
 K900 | In reply to@pveierland:matrix.org
What does the -1 part of the cache.nixos.org-1:<public_key_base64> name signify in the substituter config? The key version, theoretically | 07:20:08 |
| pveierland | Version of the key itself, or the format of what is after the comma? | 07:21:14 |
| K900 | Of the key itself | 07:22:22 |
| K900 | It is not actually significant | 07:22:28 |
| K900 | It is purely a convention | 07:22:32 |
| K900 | Nix itself does not understand it | 07:22:37 |
 sss | hi all, is it possible to migrate from ~/.config/nixpkgs/overlays to flakes ? | 10:14:52 |
| K900 | This is a question for #Nix / NixOS | 10:18:27 |
| sss | understood | 10:56:02 |
| @sigmasquadron:matrix.org | Hey, I'm building a package with __structuredAttrs enabled, and I'm getting some bash variable expansion issues when setting paths in configureFlags. I have --with-initdir=$out/etc/init.d set, but for some reason, it's mapped to \$out/etc/init.d, and it ends up expanded to /ut/etc/init.d, and I don't know why. | 13:35:36 |
| @sigmasquadron:matrix.org | * Hey, I'm building a package with __structuredAttrs enabled, and I'm getting some bash variable expansion issues when setting paths in configureFlags. I have --with-initdir=$out/etc/init.d set, but for some reason, it's mapped to \$out/etc/init.d, and it ends up expanded to /ut/etc/init.d, and I don't know why. How do I ensure the arbitrarily-named outputs are passed correctly? | 13:36:05 |
| emily | ${placeholder "out"} | 13:36:39 |
| @sigmasquadron:matrix.org | Will it automagically figure out the right paths afterwards? | 13:37:35 |
 @paparodeo:matrix.org | not in nix-shell but in a builder it will. | 13:37:54 |
| @paparodeo:matrix.org | you might be able to get away with $(out) if the variable is getting expanded in a makefile. | 13:38:57 |
| @paparodeo:matrix.org | but placeholder seems nicer despite the problems with nix-shell. | 13:39:24 |
| @sigmasquadron:matrix.org | i think placeholder will be fine, since it's inside nixpkgs. | 13:39:34 |
 Ainz Ooal Gown | In reply to @atemu12:matrix.org
We had ofborg-like building a little while ago again; where did that go? That wasn't merged actually | 15:36:54 |
