nixos/systemd: Add package version tests to integration tests

[jpds]

Adds a package version test to the systemd integration tests.

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.

[LordGrimmauld]

i am a little confused, why is one an assert in and the other a succeed(" | grep") ? This feels more inconsistent than it could be.

[jpds]

@LordGrimmauld It might just be a preference thing but personally I prefer using assert for checking command outputs (which is also what the rest of the tests in this file do).

[LordGrimmauld]

@LordGrimmauld It might just be a preference thing but personally I prefer using assert for checking command outputs (which is also what the rest of the tests in this file do).

sure, but why not make both of them assert then, why test one with grep ?

[jpds]

Because grep is what's normally used for waiting and then checking log files.

[LordGrimmauld]

but you are grepping through stdout, not through a log file. I don't see the point still.

[jpds]

@LordGrimmauld This is pretty much the established pattern for using journalctl in nixos/tests/.

[LordGrimmauld]

Maybe, but in that case journalctl supports a --grep flag (and you'll probably want to add regex escapes, the version string will contain . which is a special thing in regex).

[LordGrimmauld]

Tested, the "happy path" works, but if the version is broken the test will stall.

[ElvishJerricco]

Wait I don't understand; why do we want this? It's not like there's any risk of a different version leaking in somehow, and this just makes it more annoying to bisect systemd bugs.

[LordGrimmauld]

There is scenarios in which this breaks. It just broke for me yesterday: The git merge between #442587 and #427968 will not report conflicts around the version field, yet it would fail to actually bump the version of systemd. However, because the hash would be bumped, in case of a cached source systemd would still build with the new source. And this test would catch that error, i believe/hope.
That said, we do have (lib.mesonOption "version-tag" finalAttrs.version), so we might be compiling in what we have anyways. Would need to do some more testing on this.