gettext, texinfo: Remove spurious xz

[rhelmot]

Description of changes

part of #296581

If xz is overridden with a derivation that does not have an explicit bin output, these will fail. These changed instances are the ones impacted by the FreeBSD stdenv providing a bootstrap-files xz for early boot.

diffoscope indicates that the builds with and without xz are identical, so just remove.

Things done

• Built on platform(s)
  • x86_64-linux
  • aarch64-linux
  • x86_64-darwin
  • aarch64-darwin
  • x86_64-freebsd
• 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.

[SuperSandro2000]

Isn't this automatically chosen here?

[wolfgangwalther]

Similar to the other xz.bin this seems to have been introduced as part of #7701 in 2014:
4dccb22#diff-a05908b6eefb0d3630d5d06934fbfb4f0294382eeefef4438d9facb44a5bc291

[SuperSandro2000]

strictDeps is already set and --with-xz seems to be not properly checked and I think right now it is inneffective. We would probably need to move xz to buildInputs to properly work but it doesn't seem to do anything, at least nothing is linked against it.

Suggested change

(lib.getBin xz)

[SuperSandro2000]

This seems wrong to me especially for cross compilation

[wolfgangwalther]

Is your comment about xz.bin in the first place or about the change to lib.getBin xz?

[wolfgangwalther]

note: xz.bin was introduced in 2014 in 3ec413c#diff-bea856dc5528c15510a6ac9eafd6765a6608ff351b3f77b75c2eb1cd7d68eb14 as part of #7701

[Ericson2314]

I think @SuperSandro2000 means this should be a nativeBuildInput

[SuperSandro2000]

Is your comment about xz.bin in the first place

yes

[SuperSandro2000]

We don't need that here anymore. At the time the tarball was in the xz format but now we have gz and strictDeps is already set.

I've also built the package with and without xz and diffoscope didn't find changes except the store path change.

[Ericson2314]

Please put the description in the commit messages (goes for many of the other PRs too).

I think it's OK to just leave the texinfo one as-is with the suspicious use of buildInputs as that's an orthogonal problem, but if you are feeling motivated you could add a strictDeps = true and see if nativeBuildInputs is better (simulating cross).

[SuperSandro2000]

Suggested change

	xz
	xz.bin
	(lib.getBin xz)

I've run diffoscope on a gettext build with and without xz and booth are equal. I think we can just drop xz entirely from here.

and we can also drop the "--with-xz" arg in line 31.

[rhelmot]

I have changed the commits as per feedback, removing both uses!