Re: [PATCH] git-p4: Add failing test case for name-rev vs symbolic-ref

From: Luke Diamand <luke@diamand.org>
Date: Mar 27, 2017, 06:57 UTC
Message-ID: <CAE5ih78jhR3GwMBdPyi5oOHByC2fVxefExStc9pCNMTu-1Os7w@mail.gmail.com>
In-Reply-To: <xmqqefxj4ofm.fsf@gitster.mtv.corp.google.com>

On 27 March 2017 at 00:18, Junio C Hamano <gitster@pobox.com> wrote:

Show 23 quoted lines

> Luke Diamand <luke@diamand.org> writes:
>
>> As per the discussion about use of "git name-rev" vs "git symbolic-ref" in
>> git-p4 here:
>>
>> http://marc.info/?l=git&m=148979063421355
>>
>> git-p4 could get confused about the branch it is on and affects
>> the git-p4.allowSubmit <branchname> option. This adds a failing
>> test case for the problem.
>>
>> Luke Diamand (1):
>>   git-p4: add failing test for name-rev rather than symbolic-ref
>>
>>  t/t9807-git-p4-submit.sh | 16 ++++++++++++++++
>>  1 file changed, 16 insertions(+)
>
> Ahh, thanks for tying loose ends.  I completely forgot about that
> topic.
>
> If we queue this and then update the function in git-p4.py that
> (mis)uses name-rev to learn the current branch to use symbolic-ref
> instead, can we flip the "expect_failure" to "expect_success"?

Yes. The test passes with your change.

>
> IOW, can we have a follow up to this patch that fixes a known
> breakage the patch documents ;-)?

Yes.

Regards! Luke