mkDerivation: make allowedReference/allowedRequisites nullable when __structuredAttrs is true

[ShamrockLee]

stdenv.mkDerivation { allowedRequisites = null; } should work whether __structuredAttrs is false or true.

Things done

• Built on platform:
  • x86_64-linux
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
• Tested, as applicable:
  • NixOS tests in nixos/tests.
  • Package tests at passthru.tests.
  • Tests in lib/tests or pkgs/test for functions and "core" functionality.
• Ran nixpkgs-review on this PR. See nixpkgs-review usage.
• Tested basic functionality of all binary files, usually in ./result/bin/.
• Nixpkgs Release Notes
  • Package update: when the change is major or breaking.
• NixOS Release Notes
  • Module addition: when adding a new NixOS module.
  • Module update: when the change is significant.
• Fits CONTRIBUTING.md, pkgs/README.md, maintainers/README.md and other READMEs.

---

Add a 👍 reaction to pull requests you find important.

[ShamrockLee]

nixpkgs-review result

Generated using nixpkgs-review-gha

Command: nixpkgs-review pr 462141
Commit: b803d3076034aef35a40b3ba2bc0f2458e639840 (subsequent changes)
Merge: d380d3cad68a2e82687871ca55bd475076d52b14

Logs: https://github.com/ShamrockLee/nixpkgs-review-gha/actions/runs/19401209750

---

x86_64-linux

✅ 1 package built:

• tests.overriding

---

aarch64-linux

✅ 1 package built:

• tests.overriding

---

x86_64-darwin (sandbox = true)

✅ 1 package built:

• tests.overriding

---

aarch64-darwin (sandbox = true)

✅ 1 package built:

• tests.overriding

[wolfgangwalther]

Why would I set these to null instead of [] to disable?

[ShamrockLee]

Why would I set these to null instead of [] to disable?

These two variables are the whitelists. allowedReferences = [ ]; forbids this derivation from depending on any other packages.

[wolfgangwalther]

I'm trying to imagine when setting allowedRe... = null is useful - and I think it only makes sense when overriding attrs, right?

But consider this:

mkDerivation {
  __structuredAttrs = true;
  outputChecks.allowedRequisites = [ foo ];
}

When overriding this with allowedRequisites = null;, it will not have the intended effect - foo will still be in there.

When overriding attrs of a structuredAttrs-enabled derivation, then you just need to override outputChecks.... and not the non-structured-attrs variant.

We only have this compatibility hack to be able to create derivations that don't specify __structuredAttrs at all, i.e. rely on the default, and will then still work when flipping the default later.

I think we don't need to support overrideAttrs for this thing - it's discouraged from using inside Nixpkgs anyway.

[ShamrockLee]

If there is a package with { allowedRequisites = [ ... ]; }, and I want to override it so to be { __structuredAttrs = true; }; it would then be helpful to cancel the top-level allowedRequisites with null.