Is there a holdup on reviewing Nix PRs?

I’ve had a trivial PR against Nix open for 2 months now, with no activity. I’ve mentioned it twice already in PRs ready for review and still nothing. The PR is very tiny, most of the +20 is just tests.

Given that, is there something preventing small Nix PRs from being reviewed right now?

https://github.com/NixOS/nix/pull/4281

It’s pretty common for some PRs to get overlooked. Just ping me if you want some movement. Unfortunately I don’t have as much time as I used to.

Nixpkgs now is going through ~2700 PRs a month, that’s about 90 a day. https://github.com/NixOS/nixpkgs/pulse/monthly

Would also like to thank @Sandro for doing 30-120 reviews a day for the past month.

EDIT:
This was about NixOS/nix not NixOS/nixpkgs… oops! :slight_smile:

2 Likes

Oof! Is Nixpkgs and Nix using the same resources for code reviews? I’d imagine the skillset is slightly different.

Oh sorry, I’m so used to people posting about nixpkgs PRs.

Yes, NixOS/nix largely has eelco as a single bus factor.

Not at all. NixOS/nix is largely a C++ code base with a fair amount of complexity. NixOS/nixpkgs we can leverage a lot of community tools, and expertise is usually package dependent. So we can lean users of applications to help with PR reviews.

The governance of Nix{,pkgs} is that Eelco is a “laissez faire” (hands-off) BDFL who mostly works on nix (the package manager). Nixpkgs is largely curated, moderated, and governed by volunteers in an unstructured way (which I call technical chaos).

10 Likes