python3Packages: deprecate 'format = "other/pyproject"'

[natsukium]

generated by following command
rg 'format = "other"' -l | xargs -n 1 sed -i 's/format = "other"/pyproject = false/g'

The following commits enabled assertions for format = "other/pyproject"
5d6d563
460c25c

Things done

• Built on platform(s)
  • x86_64-linux
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
• For non-Linux: Is sandboxing enabled in nix.conf? (See Nix manual)
  • sandbox = relaxed
  • sandbox = true
• Tested, as applicable:
  • NixOS test(s) (look inside nixos/tests)
  • and/or package tests
  • or, for functions and "core" functionality, tests in lib/tests or pkgs/test
  • made sure NixOS tests are linked to the relevant packages
• Tested compilation of all packages that depend on this change using nix-shell -p nixpkgs-review --run "nixpkgs-review rev HEAD". Note: all changes have to be committed, also see nixpkgs-review usage
• Tested basic functionality of all binary files (usually in ./result/bin/)
• 24.11 Release Notes (or backporting 23.11 and 24.05 Release notes)
  • (Package updates) Added a release notes entry if the change is major or breaking
  • (Module updates) Added a release notes entry if the change is significant
  • (Module addition) Added a release notes entry if adding a new NixOS module
• Fits CONTRIBUTING.md.

---

Add a 👍 reaction to pull requests you find important.

[Aleksanaa]

Might be better to keep lib.warn for a release, and change to assert/throw in the next.

[natsukium]

Since these changes are just replacements, and the release included is 25.05 in 6 months, I think we could throw an error.
We easily miss warnings, and I'm exhausted of pointing them out in reviews.

But if someone else also suggests lib.warn, I'll make the change.

[Aleksanaa]

Since these changes are just replacements, and the release included is 25.05 in 6 months, I think we could throw an error.

That will still be a show stopper for people using unstable channel. Such changes are not completely unacceptable, but we generally don't do this.

We easily miss warnings, and I'm exhausted of pointing them out in reviews.

Ofborg will turn it into an error so such new patterns won't be accepted into Nixpkgs.

[natsukium]

Ok, let's change it to warn. I found the evaluation errors around home-assistant, so I'll commit it along with this fix.
Thanks for your review.

[GaetanLepage]

I guess when converting format = "pyproject"; to pyproject = true, you often need to add setuptools to build-system no ?

[natsukium]

I guess when converting format = "pyproject"; to pyproject = true, you often need to add setuptools to build-system no ?

In many cases, simply change nativeBuildInputs to build-system. However, we cannot change them automatically. This is because nativeBuildInputs may contain other non-python modules, for example, rustc, pkg-config, and makeWrapper. And there are over 2,000 python modules that have nativeBuildInputs.

Also, to change format = setuptools, setuptools should be added to build-system, but it causes the build system to change. It means, dependency checks or something will become blockers, and we will also need a number of manual fixes.

[Aleksanaa]

You need tree-sitter and a Nix evaluator. Not impossible but will be more complex than a simple grep.

[natsukium]

You need tree-sitter and a Nix evaluator. Not impossible but will be more complex than a simple grep.

Good point. Yes, I have a fork of nixpkgs-lint.
nix-community/nixpkgs-lint@master...natsukium:nixpkgs-lint:python

[SuperSandro2000]

I was first confused by the commit titles and thought we abort evaluation instead of just logging a warning.

[natsukium]

I was first confused by the commit titles and thought we abort evaluation instead of just logging a warning.

Thanks, updated the commit messages.