Robert Hensing
Robert Hensing
The current expressions seems to be kind of geared towards flakes, but also not really using flakes as intended. Initializing `pkgs` is currently a mess in the whole ecosystem. I...
I've changed the user to a have a lower uid and an underscore prefix, as hinted by the previously referenced discussion https://github.com/NixOS/nix/pull/4532. Is there anything else I should change?
@LnL7 I've implemented the change. Do you think this can be merged?
I've added `meta.maintainers` with the recommendation to use `maintainers.alice or "alice"` instead of the usual Nixpkgs way. This wasn't strictly necessary for building nix-darwin configurations, but allows to build on...
Is there anything I can do?
> Absolut paths are probably _only_ sensible as a development helper. and the relative paths like subflakes can be locked by locking their parent flake instead.
> but it's better not to copy the tree for a development shell The `hhvm` package "looks" like it requires the copy, but the expression should be lazy enough to...
`pathType` on a store path is generally not something you'll want to do, because it blocks the evaluator, causing the evaluation heap to exist for longer than necessary (wasting memory...
Right, it used to be that you could just avoid putting your project sources in the store (except for derivation sources, by necessity). I hope #6530 makes that work a...
I've started a draf spec for the protocol, calling it ACES: Auto-Completion for Executables and Shells. It's close to what this PR implements, but uses `--aces-*` instead of `--bash-*` to...