| 9 Oct 2025 |
 ivy | is there a way to define arbitrary variables in a .cabal file that are maintained in the outputted 2nix | 07:16:05 |
 chreekat | Well I know there are extension fields supported in .cabal files , but I don't know if cabal2nix knows or cares about them. (Extension fields are only documented in one place to my knowledge: https://cabal.readthedocs.io/en/stable/file-format-changelog.html#cabal-version-1-1-4) | 10:51:38 |
 sterni | In reply to @ivy:faggot.sh
is there a way to define arbitrary variables in a .cabal file that are maintained in the outputted 2nix no. (what for?) | 12:33:02 |
| ivy | okay ill add some and then add to cabal 2nix | 12:35:33 |
| ivy | i want to track the verson of the karabiner driver in kmonad upstream | 12:36:07 |
| ivy | ill patch cabal2nix and do a pr for that for extension fields (i am very experienced with haskell) | 12:36:43 |
| ivy | should these x- fields be put into passthru or to the derivation itself | 12:41:22 |
| ivy | i’m feeling passthru therefore i should write a passthru field to derivation adt | 12:42:05 |
| ivy | idk what do yall think | 12:42:18 |
| ivy | but also could be cool to pass through to extra attributes | 12:45:48 |
| sterni | well does kmonad even set these x-fields? would Hackage preserve them? | 13:05:49 |
| sterni | in principle seems like a feasible idea, passthru seems about right though x-revision should continue to be treated specially probably | 13:06:51 |
