python3Packages.limits: fix input tarball non-reproducibility#171832
python3Packages.limits: fix input tarball non-reproducibility#171832vcunat merged 1 commit intoNixOS:staging-nextfrom
Conversation
|
cc @fabaff |
Can we please tell upstream that's not okay? |
|
It could be that I messed things up with the hash. |
I'm still investigating, but ofborg was fine with that hash according to the logs, and it doesn't match any other version I can find still published.
I'll file an issue when I've confirmed it's on their side. I suggest not blocking on this so staging-next can get greener faster? :) |
|
Successful fetch on ofborg: Failed fetch on Hydra: The archive size seems to have changed. |
|
See #84312 |
|
Interesting, thank you! I didn't know about this quirk. |
|
This is how I solved it for another package but I don't think it will work in this case: |
Context: NixOS#84312 documents that using .gitattributes on GitHub results in non-reproducible release tarballs. versioneer uses this to provide version numbers for Python libraries. Hack around that issue by deleting the file containing substitutions in the fixed hash input derivation, and recreate the required info from the derivation's version info at patchPhase time.
|
@dotlambda thanks for the reference, and there are actually a few other Python packages using versioneer that have similar workarounds. PTAL. |
|
@ofborg build python39Packages.limits python310Packages.limits |
Description of changes
Fixes regression from last bump (22f5d32), it looks like the
release tarball was changed upstream.
Tested:
nix-build . -A home-assistantnow passes on staging-next.Things done
sandbox = trueset innix.conf? (See Nix manual)nix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage./result/bin/)nixos/doc/manual/md-to-db.shto update generated release notes