Patrick Steinhardt <ps@xxxxxx> writes:
> Changes compared to v2:
>
> - I've simplified patch 1 to reset the preferred pack index instead
> of moving around some of the checks. This also causes us to print
> the warning for missing preferred packfiles again.
>
> - I've fixed the logic in patch 2 to find the preferred packfile to
> not return a packfile that would be rolled up in a geometric
> repack.
>
> - I've added an additional patch to split out preexisting tests for
> `--stdin-packs` into their own test file.
>
> - I've changed the unportable test added for geometric repacking
> with `-l` that used stat(1) to instead use our `test-tool chmtime`
> helper.
>
> - I've changed the logic that disables writing bitmaps in git-repack
> to cover more cases. It now always kicks in when doing a repack
> with `-l` that asks for bitmaps when connected to an alternate
> object directory.
>
> - In general, there's a bunch of small improvements left and right
> for the tests I'm adding.
>
> Thanks for all the feedback so far.
Loss of "stat" is very much appreciated, as the ones added
in the previous round were the only hits from
$ git grep '^[ ]*stat ' t/ ;# leading SP/HT plus "stat "
and I suspect among platforms that are not Windows, not Linux, and
not macOS, there may be ones that lack the tool.
With a few fix-ups to the test script I sent to the thread for the
review of [09/10] squashed in, this topic however seems to cause the
linux-TEST-vars job fail when queued at the tip of 'seen' (bbfd3bf)
[*1*]. When 'seen' is rewound by one merge to exclude this topic,
the CI runs OK (c35f3cd)[*2*]. I dug only far enough to identify
which topic among 40+ ones was causing the CI breakage. Perhaps the
code is operating correctly but the expectation in the test needs
updating? I dunno.
Thanks.
[References]
*1* https://github.com/git/git/actions/runs/4694670520/jobs/8323047888#step:5:313
*2* https://github.com/git/git/actions/runs/4694155668
