| 21 Sep 2023 |
 @ronnypfannschmidt:matrix.org | In reply to @crtified:crtified.me
The first one is probably evaluated in a different flake and thus "breaks" the expectation of being pure, while the second one is impure on its own as you rarely have a pure shell env (and the concrete result of nixpkgs depends on your local flake registry) as per https://nixos.org/manual/nix/stable/language/builtins.html#builtins-getFlake - unless impure is allowed the flake needs to be locked
nix shell with a flake auto-updates a lock if necessary | 10:52:13 |
| @ronnypfannschmidt:matrix.org | however nix-shell will also put the flake into the nix store before evaluating | 10:52:55 |
|  bembo set a profile picture. | 16:01:57 |
| 26 Sep 2023 |
 @penguincoder:matrix.wolfie.pw | I'm trying to use flake inputs as a way to track several other non-flake git repos. The idea is that after CI passes on the other repos, I will update the lock for that particular input. I then want to use my flake to build packages for these other, non-flake, inputs. I can add the input pretty easily, and flake-parts provides me with inputs' as an attribute to my modules. This much works great, but when I go to pass the input in mkPoetryApp { projectDir = inputs'.my-python-app; }, it fails with cannot coerce a set into a string.
which is okay, I guess, it's an attrset. That makes sense... but none of the attrs that I know exist in the input schema seem to work. I can't get lib.debug.traceVal to ever print anything out, either. What can I give mkPoetryApp to figure out where the path-in-the-store-on-my-disk is for projectDir or is there a better way to manage the non-flake inputs? What I don't want to do is have to manually change the rev using fetchGit each time CI changes.
| 19:02:22 |
| 21 Sep 2023 |
|  dedmunwalk joined the room. | 23:06:10 |
| 22 Sep 2023 |
|  bl1nk changed their profile picture. | 11:21:25 |
| bl1nk changed their profile picture. | 11:23:10 |
|  Oro (any/all) changed their profile picture. | 20:45:43 |
| 23 Sep 2023 |
|  Snuupy joined the room. | 10:17:54 |
|  @0xmrtt:projectsegfau.lt removed their profile picture. | 12:37:12 |
| @0xmrtt:projectsegfau.lt removed their display name 0xMRTT. | 12:44:20 |
| @0xmrtt:projectsegfau.lt left the room. | 12:52:08 |
|  heyoni joined the room. | 20:19:03 |
| 24 Sep 2023 |
 @aidalgol:matrix.org | If I have a function call several times with the same argument values, does it have to compute it for every call, or just once? | 08:51:35 |
| @aidalgol:matrix.org | * If I have a function call several times in a flake.nix with the same argument values, does it have to compute it for every call, or just once? | 08:51:50 |
|  mib 🥐 joined the room. | 12:21:19 |
| 25 Sep 2023 |
|  bertof joined the room. | 10:43:46 |
|  eva joined the room. | 15:45:15 |
|  Las left the room. | 19:44:14 |
| 26 Sep 2023 |
| Oro (any/all) changed their profile picture. | 05:51:42 |
| 27 Sep 2023 |
|  amnicolist joined the room. | 02:13:56 |
| amnicolist | Hi flakes, I'm new to nixos/flakes and this chat. I have a question about flake-parts and overlays. I hope its okay to ask here.
I'm trying to get the [rust-overlay](https://github.com/oxalica/rust-overlay
| 02:17:58 |
| amnicolist | * Hi flakes, I'm new to nixos/flakes and this chat. I have a question about flake-parts and overlays. I hope its okay to ask here.
I'm trying to get the [rust-overlay](https://github.com/oxalica/rust-overlay) working with an existing repo that I forked. However, I can't figure out how to get access to the rust-bin set once I follow the https://flake.parts/overlays documentation
| 02:19:19 |
| amnicolist | * Hi flakes, I'm new to nixos/flakes and this chat. I have a question about flake-parts and overlays. I hope its okay to ask here.
I'm trying to get the rust-overlay working with an existing repo that I forked. However, I can't figure out how to get access to the rust-bin set once I follow the https://flake.parts/overlays documentation
| 02:19:32 |
 Charles | Would you be amenable to an alternative that doesn't involve overlays | 02:43:36 |
| mib 🥐 changed their display name from mib to mib 🥐. | 05:53:08 |
|  belvaleth joined the room. | 13:25:07 |
 @antifuchs:asf.computer | amnicolist: you're following the "consuming an overlay" section? | 17:47:28 |
| @antifuchs:asf.computer | that will define the "pkgs" attribute available in the perSystem closure's arguments as the overlaid version, but you then also have to go and use that pkgs attribute to define the system / home-manager configuration... that's typically what trips me up | 17:49:43 |
| @antifuchs:asf.computer | but anyway, how are you planning to use the overlay? (: | 17:50:28 |
