Re: [PATCH v2] refspec: make @ a synonym of HEAD

From: Junio C Hamano <gitster@pobox.com>
Date: Nov 25, 2020, 01:53 UTC
Message-ID: <xmqq360ygq2g.fsf@gitster.c.googlers.com>
In-Reply-To: <CAMP44s2ChcCjhjksS0s5BoYznqLAoXyFvSkP4GxSCh_ALusOtQ@mail.gmail.com>

Felipe Contreras <felipe.contreras@gmail.com> writes:

Show 19 quoted lines

> On Tue, Nov 24, 2020 at 6:37 PM Junio C Hamano <gitster@pobox.com> wrote:
>> Felipe Contreras <felipe.contreras@gmail.com> writes:
>
>> > +test_expect_success 'push with @' '
>> > +
>> > +     mk_test testrepo heads/master &&
>> > +     git checkout master &&
>> > +     git push testrepo @ &&
>> > +     check_push_result testrepo $the_commit heads/master
>> > +
>> > +'
>>
>> This is OK, but shouldn't this be placed before the tests with
>> various configuration?  Something along the lines of the attached,
>> but with the body of the loop properly reindented, would also give
>> us a better test coverage at the same time.
>
> I don't see much value in those tests, since I don't see how if one
> passes another one would fail. But I guess it cannot hurt.

That can only be said based on the knowledge of the implementation detail of the code immediately after this patch gets applied. Any future change to the code for whatever reason (e.g. refactoring) can make the current assumption invalid.

As the proposed log message says,

    Since commit 9ba89f484e git learned how to push to a remote branch using
    the source @, for example:

      git push origin @:$dst

    However, if the right-hand side is missing, the push fails:

      git push origin @

we care about both of these forms working, not just the singleton form, so it is not just "not hurt", but is actively a good thing, to protect both forms from future breakage. After all, that is why we have tests.

Thanks.