Peder Bergebakken Sundt
Peder Bergebakken Sundt
As a editor LSP i think both are fine, in nixpkgs CI i think neither should be enforced.
> I don't think there is a case where a = a makes sense? > i feel it lessens readability, especially in the middle of other attribute assignments that don't...
also a neat editor suggestion :+1:
It's become a bit random when it happens, but i believe a toot might be to blame, as today it only crashed after i scrolled a few posts, reproducibly so.
Is there a reason to prefer cargoLock over cargoHash? Cache reuse?
There is also the fact that with `cargoLock` each dep becomes a derivation, making for better cache reuse. But on the whole I'd say its speed and cache benefits are...
> > This causes the testtray binary to complain about missing assets. > > This is an issue with the testtray test upstream, I've opened [libsdl-org/SDL#13252](https://github.com/libsdl-org/SDL/pull/13252) to address this. should...
builds when i cherry pick this and the ibus fix on master
please view https://github.com/NixOS/ofborg/issues/689#issuecomment-2295412796
> Sources could be retrieved from the PyPI tarball otherwise, but would lose the ability to run the tests (this is in context of distributing and testing in nixpkgs)