Revert "freebsd.yml: fix warning"

[xtqqczze]

Use jq instead of jaq in freebsd workflows.

As of 2026-09-23, the deprecation for oniguruma was removed. See FreshPorts:

devel/oniguruma: Remove DEPRECATED line because there are many consumers

and consumers are not getting off oniguruma any time soon.

Reverts #14488.
Conflicts with #14947.

[oech3]

This should NOT close #14947 .
The script is can be used on the system missing jq, not only on the CI. It is also serious for reproducibility of the result.

[github-actions]

GNU testsuite comparison:

Congrats! The gnu test tests/id/setgid is no longer failing!

[xtqqczze]

The script is can be used on the system missing jq, not only on the CI. It is also serious for reproducibility of the result.

Sorry, I may be misunderstanding the issue here. The script installs jq using pkg install, so could you please clarify what you mean by the script being used on systems without jq?

[oech3]

What?! show-utils itself does not (and should not) install jq.

[xtqqczze]

I mean the script that was changed in .github/workflows/freebsd.yml.

[oech3]

#14947 you tried to close is PR about the script show-utils, not *.yml .

[oech3]

Do we need to revert this just for famous command name? Reverting reduces 2 lines, but it is also done at #14947

[xtqqczze]

CICD / Build (macos-latest, x86_64-apple-darwin, feat_os_unix, true, PGO) (pull_request) failure tracked by #14888.

[xtqqczze]

Do we need to revert this just for famous command name?

I'd rather use the same command everywhere for consistency. For example, jq is available on all GitHub Actions runners and on every version of macOS. I don't see a strong reason to use jaq solely on FreeBSD.

[oech3]

OK. Would you merge this by yourself? Then I rebase my PR.

[xtqqczze]

OK. Would you merge this by yourself? Then I rebase my PR.

OK, could you please mark #14947 as draft for now? Alternatively, if you prefer, you can rebase on this PR (to include 91874f2) and I’ll close it.

[oech3]

marked as draft