eigen_3_4_0: init at 3.4.0, migrate eigen{,2,_3_4_0} to by-name #389101
Conversation
github-actions
bot
added
10.rebuild-darwin: 1
pbsds
changed the title
There was a problem hiding this comment.
Is this a verbatim copy of by-name/ei/eigen/include-dir.patch?
There was a problem hiding this comment.
Well then we might as well reference it by ../... but also I'd prefer the "more generic expression" approach
There was a problem hiding this comment.
I don't want to resort to ../.. reach around, as i find them to be brittle.
IMO the maintenance burden is far larger when two packages have to go though a common generic.nix as they diverge more and more. I trust the tarball compression to outdo any compression we do by hand trying to re-using code.
There was a problem hiding this comment.
IMO the maintenance burden is far larger when two packages have to go though a common generic.nix
Yes, because that requires making .override and .overrideAttrs actually work, which could be considered one of the main value propositions of Nixpkgs
There was a problem hiding this comment.
My thinking was that fetchpatch {} could just go behind optionals (versionAtMost finalAttrs.version "3.4.0") (where versionAtMost = flip versionAtLeast, unless I'm confusing myself) in the eigen3 expression and then we'd just callPackage twice or use overrideAttrs in all-packages.nix (or when defining eigenVersions if we were doing that)
There was a problem hiding this comment.
to satisfy nixpkgs-vet we still have to add a file at pkgs/by-name/ei/eigen_3_4_0/package.nix rather than all-packages.nix, unless you want to expose eigen_3_4_0 in eigen.passthru?
There was a problem hiding this comment.
to satisfy nixpkgs-vet
Is this about the by-name check? Can't we just ignore it since we're adding it there for a reason and not because we haven't heard of by-name?
There was a problem hiding this comment.
I prefer not giving myself the license to circumvent codified guidelines and CI checks.
wegank
added
12.approvals: 1
|
I just rebased on top of #397641 which also benefits from |
pbsds
force-pushed
the
migrate-eigen-1741728350
branch
from
b42e966 to
c9bcec0
Compare
ofborg
bot
removed
the
2.status: merge conflict
|
5 participants
#364362 (comment) wants eigenVersions, but thats a bit overkill. Hence this, which follows package-naming
all should be in cache
Things done
nix.conf? (See Nix manual)sandbox = relaxedsandbox = truenix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage./result/bin/)Add a 👍 reaction to pull requests you find important.