| 23 Aug 2023 |
| Michael joined the room. | 03:14:30 |
| @rover:aguiarvieira.pt joined the room. | 07:30:51 |
| @brian:bmcgee.ie left the room. | 09:57:34 |
| @aloisw:kde.org joined the room. | 12:13:56 |
| @aloisw:kde.org | 12:14:08 |
| 25 Aug 2023 |
| @lenny:flipdot.org changed their display name from Lenny. 7913 to Lenny.. | 12:05:03 |
| 27 Aug 2023 |
@antifuchs:asf.computer | so I have a github job that regularly runs nix flake update and files a PR with the version bump. Is there a nice way to visualize the things that change, especially around what packages got updated? Ideally as a follow-up github action? (: | 16:21:04 |
Ilan Joselevich (Kranzes) | nix flake update --commit-lock-file | 16:22:10 |
Ilan Joselevich (Kranzes) | This commits the file after an update with a commit description/message indicating which things got updated | 16:22:28 |
@antifuchs:asf.computer | oooooh, TIL! | 16:23:16 |
CRTified | But that only shows what changed for the inputs, i.e. the git rev | 16:24:12 |
Ilan Joselevich (Kranzes) | yup, if you want something for actual derivations you could use nix store diff-closures | 16:24:35 |
@antifuchs:asf.computer | Hm hm hm. I think yeah, the closure diff would be preferable. Guess I’d have to identify the thing I would like to diff; but it would probably work. Hmmmmm. | 17:17:23 |
| Moritz joined the room. | 17:21:26 |
@antifuchs:asf.computer | the alternative would be to not open a PR and push to the default branch directly, it's about the same amount of scrutiny that these updates get (: | 18:18:32 |
| lord_fomo joined the room. | 21:22:29 |
| catch22 joined the room. | 22:08:15 |
| 30 Aug 2023 |
mewp | what is the proper method to have a nixos module use a package defined in the same flake? is there a current system available somewhere so that I could use packages.${currentSystem}. something? | 10:50:22 |
mewp | oh, Imanaged to find pkgs.system, nevermind | 10:56:09 |
mewp | * oh, I managed to find `pkgs.system`, nevermind | 10:56:17 |
@petrichor:envs.net | mewp: see also https://github.com/hercules-ci/flake-parts which might help with maintainability in the long run | 11:00:14 |
| @adham-omran:matrix.org left the room. | 16:00:08 |
| Chinchilla Optional joined the room. | 22:08:00 |
Chinchilla Optional | How can I specify a branch for a local (git+file:) flake? | 22:11:05 |
Chinchilla Optional | Oh, looks like `?ref=<branchname>' | 22:13:30 |
| 31 Aug 2023 |
@arianvp:matrix.org | how is nixConfig option supposed to work? | 09:25:44 |
@arianvp:matrix.org | I tried to add the nixConfig.bash-prompt option but my prompt doesnt change in nix-develop | 09:25:55 |
| industrialrobot changed their display name from IndustrialRobot (she/her) to IndustrialRobot. | 10:50:49 |
@antifuchs:asf.computer | I don't think all the settings are piped through? this seems like it would be better to set in the dev shell definition of the flake | 11:52:32 |
@antifuchs:asf.computer | that said, I haven't gotten the binary cache (the example given in the docs) to work with nixConfig either | 11:52:52 |