- Nix 100%
i coincidentally ran into the same `-Wimplicit-function-declaration` issue while messing with ci thought it would like to have the fix too don't really know my way around nix yet but builds seemed fine when testing in a lix container Reviewed-on: https://git.gay/doll/pkgs/pulls/3 |
||
|---|---|---|
| modules | ||
| packages | ||
| support | ||
| .envrc | ||
| .gitignore | ||
| flake.lock | ||
| flake.nix | ||
| overlay.nix | ||
| README.md | ||
dollpkgs
more nixpkgs but for gay
mirrored between:
packages are built by https://hydra.dolly.sh
a binary cache is available: https://subby.blood.pet with public key
subby.blood.pet:1xqRlAxaTPJi/URlleX0ggGECwKzJbX+EVu6xY54wrc= (this will soon move)
contributing
this collection is for software and modules that do not exist in nixpkgs, and could, but the contribution model of nixpkgs (and it being on github) is not inviting to you.
pull requests are welcome on our git.gay mirror.
💕 don't be afraid, we don't bite. we want to help if its needed.
code that wasn't synthesized directly by the committer (such as LLM outputs) is strictly prohibited.
flake inputs MUST be used sparingly. if a consumer can't override a dependent package with an overlay, the package or module MUST be rearchitected to allow for this. we don't want to be in the zone of re-shipping flakes like crane.
all packages MUST be overridable with .override and .overrideAttrs.
all packages and modules SHOULD include checks and/or nixosTests. if not, please document why.
packages and modules MUST NOT be specific to a contributor's environment and machine. we want purity and portability. if this is hard, please ask for help on a draft pull request.
contributions SHOULD avoid breaking changes, and use the mechanisms built into nix and nixos for helping adjust when breaking changes must occur.