Re: What's cooking in git.git (Jun 2026, #02)

[Patrick Steinhardt <ps@pks.im>]

On Thu, Jun 04, 2026 at 02:21:22AM -0400, Jeff King wrote:
> On Thu, Jun 04, 2026 at 11:35:50AM +0900, Junio C Hamano wrote:
> 
> > * ps/t7527-fix-tap-output (2026-06-02) 4 commits
> >  - t: let prove fail when parsing invalid TAP output
> >  - t/lib-git-p4: silence output when killing p4d and its watchdog
> >  - t/test-lib: silence EBUSY errors on Windows during test cleanup
> >  - t7527: fix broken TAP output
> > 
> >  A recent regression in t7527 that broke TAP output has been fixed,
> >  some other test noise that also broke TAP output has been silenced,
> >  and 'prove' is now configured to fail on invalid TAP output to
> >  prevent future regressions.
> > 
> >  Expecting a (small and hopefully final) reroll.
> >  cf. <xmqqtsrlw09t.fsf@gitster.g>
> >  source: <20260603-pks-t7527-fix-tap-output-v2-0-cf3af5694e20@pks.im>
> 
> I guess this is the source of several CI failures on jch/seen that look
> like:
> 
>     Test Summary Report
>     -------------------
>     t7810-grep.sh                                    (Wstat: 0 Tests: 264 Failed: 0)
>     Parse errors: Unknown TAP token: "/bin/sh: warning: setlocale: LC_ALL: cannot change locale (en_US.UTF-8)"
> 
> This happens in the dockerized jobs whose images presumably have a very
> minimal locale setup. I don't know if this is a sign that the tests have
> never been doing quite what we expect on those platforms, or if it's
> simply noise that is now being caught.

Ah, thanks for flagging. I'll investigate.

Patrick