[prev in list] [next in list] [prev in thread] [next in thread] 

List:       git
Subject:    Re: [PATCH 1/5] t3400: demonstrate failure with format.useAutoBase
From:       Junio C Hamano <gitster () pobox ! com>
Date:       2019-11-30 17:25:10
Message-ID: xmqqimn12s5l.fsf () gitster-ct ! c ! googlers ! com
[Download RAW message or body]

Eric Sunshine <sunshine@sunshineco.com> writes:

> On Tue, Nov 26, 2019 at 9:09 PM Denton Liu <liu.denton@gmail.com> wrote:
>> Ever since bb52995f3e (format-patch: introduce format.useAutoBase
>> configuration, 2016-04-26), `git rebase` has been broken when
>> `format.useAutoBase = true`. Demonstrate that failure here.
>
> What specifically does "broken" mean here?
>
>> Reported-by: Christian Biesinger <cbiesinger@google.com>
>> Signed-off-by: Denton Liu <liu.denton@gmail.com>
>> ---
>> diff --git a/t/t3400-rebase.sh b/t/t3400-rebase.sh
>> @@ -159,6 +159,12 @@ test_expect_success 'fail when upstream arg is missing and not configured' '
>> +test_expect_failure 'rebase works with format.useAutoBase' '
>> +       test_config format.useAutoBase true &&
>> +       git checkout topic &&
>> +       git rebase master
>> +'
>
> Having read both the commit message and the test itself, I'm not wiser
> about what is actually "broken" or what this is demonstrating.

True.

The tests must be crystal clear what kind of brokenness it is
demonstrating, not just "this test is expected to fail", especially
when the "expect failure in one step, fix and flip expectation in a
separate step" pattern is used; otherwise it becomes doubly puzzling.

Thanks.
[prev in list] [next in list] [prev in thread] [next in thread]