| 6 Dec 2021 |
balsoft | You might be able to do something using filterSource et al (e.g. clearSourceWith from nixpkgs) | 18:07:57 |
balsoft | But I'm not sure, last time I tried that it didn't work | 18:08:04 |
flyx | well I'd have assumed that filterSource just generates another intermediate derivation depending on self | 18:09:03 |
flyx | and meta-question: why is this room not part of the NixOS space when I join it? | 18:09:37 |
flyx | In reply to @balsoft:balsoft.ru But I'm not sure, last time I tried that it didn't work so I did test this, doesn't work here either. guess to make it work, we'd need to have some way of specifying the filter right in the nix flake definition, so that self is already filtered | 18:31:44 |
tomberek | In reply to @flyx:flyx.org hi folks! how can I prevent every edit of flake.nix to cause everything depending on self being rebuilt on the next nix build? flake.nix is checked into git so it is part of self, but obviously not part of the sources being built, so an edit should only affect those derivations whose definitions were changed by the edit this keeps coming up, this would be good to clarify/fix/describe | 19:46:02 |
| 7 Dec 2021 |
brendanzab | Hey, I have a flake that uses github:oxalica/rust-overlay - in the rust overlay you can select different versions of the rust toolchain (eg. stable.latest, stable."1.56.0", beta.latest) to build with. I was hoping to be able to test with a range of different versions, either locally on CI. Is there a way to ‘parametrise’ flakes to allow for something like this? | 00:11:06 |
balsoft | You can parametrise flake outputs | 00:11:59 |
brendanzab | ah cool - do you know of any examples of this? | 00:21:13 |
brendanzab | * ah cool - do you know of any examples of this that I can look at? | 00:21:27 |
brendanzab | (might have missed it in the flake docs) | 00:21:46 |
balsoft | In reply to @brendanzab:matrix.org ah cool - do you know of any examples of this that I can look at? Uh, I don't think I have anything public really | 00:25:06 |
balsoft | It's really late ATM | 00:25:19 |
brendanzab | would I just add something like rustChannel ? { toolchain = "stable", version = "latest" } to the function returned from the outputs? | 00:25:20 |
brendanzab | np | 00:25:25 |
balsoft | In reply to @brendanzab:matrix.org would I just add something like rustChannel ? { toolchain = "stable", version = "latest" } to the function returned from the outputs? No, you should make more outputs :) | 00:25:36 |
balsoft | lib.cartesianProductOfSets can be useful | 00:25:52 |
brendanzab | ahh so another layer of nesting from flake-utils.lib.eachDefaultSystem? | 00:26:05 |
balsoft | Yep | 00:26:14 |
balsoft | In checks I think | 00:26:16 |
brendanzab | cool | 00:27:14 |
brendanzab | thanks! | 00:27:17 |
brendanzab | I'll give it a go - still a massive beginner at nix 😅 | 00:27:46 |
| 8 Dec 2021 |
| Robert Hensing (roberth) changed their profile picture. | 17:40:49 |
| 9 Dec 2021 |
| bmabsout joined the room. | 19:46:43 |
bmabsout | Hello! I'm trying to run a function from flake so that the output is in a nix shell, I ended up doing things manually like this nix shell --impure --expr "(builtins.getFlake \"nixpkgs/nixos-21.11\").legacyPackages.x86_64-linux.python39.withPackages (p: [p.numpy])" is there a better way to do this? | 19:48:28 |
bmabsout | Is there a way to do something similar in meaning to nix shell nixpkgs/nixos-21.11#python39.withPackages --something "(p: [p.numpy])", I might be getting a few concepts wrong here | 19:51:16 |
bmabsout | * Is there a way to do something similar in meaning to nix shell nixpkgs/nixos-21.11#python39.withPackages --something "(p: [p.numpy])"? I might be getting a few concepts wrong here | 19:51:32 |
bmabsout | * Hello! I'm trying to run a function from a flake so that the output is in a nix shell, I ended up doing things manually like this nix shell --impure --expr "(builtins.getFlake \"nixpkgs/nixos-21.11\").legacyPackages.x86_64-linux.python39.withPackages (p: [p.numpy])" is there a better way to do this? | 20:31:56 |
| 10 Dec 2021 |
| linj joined the room. | 00:51:32 |