Re: [PATCH 3/3] t5312: prepare for reftable

From: Han-Wen Nienhuys <hanwen@google.com>
Date: Feb 3, 2022, 14:24 UTC
Message-ID: <CAFQ2z_OUqMx7WiTYHGrb5A0K1d_zNVTspM+6trw+u2rqRPjYwA@mail.gmail.com>
In-Reply-To: <220201.865ypy9te7.gmgdl@evledraar.gmail.com>

On Tue, Feb 1, 2022 at 10:19 PM Ævar Arnfjörð Bjarmason <avarab@gmail.com> wrote:

Show 12 quoted lines

> > -test_expect_success 'pack-refs does not drop broken refs during deletion' '
> > +test_expect_success REFFILES  'pack-refs does not drop broken refs during deletion' '
> >       git update-ref -d refs/heads/other &&
> >       git rev-parse refs/heads/main >actual &&
> >       test_cmp expect actual
>
> The setup for these is reffiles-specific, but it seems to me this is
> something we'd really like to test with reftable rather than skipping it
> entirely.
>
> I.e. the scenario described in the "we create..." comment in this file
> is something that might happen with reftable too, no?

That is tested in the 3 tests right above the ones I marked with REFFILES ('pack-refs does not silently delete broken loose ref'). The tests at the bottom check what happens if you have a missing SHA1 in a packed-refs file. The reftable backend does not have a packed-refs, so there is nothing to test.

-- 
Han-Wen Nienhuys - Google Munich
I work 80%. Don't expect answers from me on Fridays.
--

Google Germany GmbH, Erika-Mann-Strasse 33, 80636 Munich

Registergericht und -nummer: Hamburg, HRB 86891

Sitz der Gesellschaft: Hamburg

Geschäftsführer: Paul Manicle, Halimah DeLaine Prado