The Homu model looks quite interesting. If homu would trigger a hydra build and only push to master after the build succeeds, I believe that it would simplify a lot of things nicely.
Master would already have binary caches from the builds so we could use it as the unstable channel. This is quite big because as a small-time contributor I never really know when my contributions are going to hit unstable right now. And using master means that I might have to recompile quite a lot of things myself. Master would always be stable in term of syntax so people can rebase on top of it to build their pull-requests. Right now I have to chose between either using master with maybe broken things and no binary caches, or use the nixos-unstable branch (that is also in another repo) and not know if my changes are going to merge on master upfront. Even staging could be removed if builds are sped-up enough. On Tue, 24 Nov 2015 at 13:55 Matthias Beyer <m...@beyermatthias.de> wrote: > It seems someone added this to the nixpkgs! Awesome! > > On 22-11-2015 17:09:43, Matthias Beyer wrote: > > Hi, > > > > I wanted to ask whether it would be a good idea to use montion-bot[0] > for our > > nixpkgs repository? > > > > Another idea would be to use homu[1]. Please argue. > > > > [0]: https://github.com/facebook/mention-bot > > [1]: homu.io > > -- > Mit freundlichen Grüßen, > Kind regards, > Matthias Beyer > > Proudly sent with mutt. > Happily signed with gnupg. > _______________________________________________ > nix-dev mailing list > nix-dev@lists.science.uu.nl > http://lists.science.uu.nl/mailman/listinfo/nix-dev >
_______________________________________________ nix-dev mailing list nix-dev@lists.science.uu.nl http://lists.science.uu.nl/mailman/listinfo/nix-dev