Should I fix a bug within nixpkgs if upstream won't?

Development
1

For the Zero Hydra Failure 23.11 initiative, I fixed the build and runtime errors of the package persepolis and adopted it as it was orphaned.
The software is abandoned upstream, but as it is relatively simple and relies on stable APIs, it works mostly well with up-to-date dependencies.

The only exception is its reliance on youtube-dl, which has been abandoned for well over two years now and broken for months. The alternative is yt-dlp, and up-to-date fork that works wonderfully. It would be very much possible to replace this dependency, and there is an issue on the upstream repo and even a PR to fix it, but as the repo is abandoned, it will not be merged, leaving this bug unfixed.

Now the question is; should I incorporate this fix in nixpkgs? I know we generally want to stay true to upstream and only fix incompatibilities with Nix, but I feel in this case, it might be reasonable to apply this small fix to make the package more useful for everyone.

4 Likes
2

sure, the mentioned PR is tiny and makes sure the software continues to work with online services which usually is more important.

5 Likes
3

ooi, why not fork upstream to make that change at this point?

2 Likes
4

Hmm, interesting idea. It’s a bit of a rabbit hole, but there are some issues I’m itching to fix, so it might be fun to dive in.

Though to be honest, I would rather spend my time contributing to the Nix ecosystem directly, that feels more impactful than maintaining one of many download managers.

3 Likes
5

i check the web site, it has a large team working on it…

and someone put a lot efforts into the repo and web site.

i wonder what happened to them .