Re: [PATCH v2 01/27] t1400: Fix name and expected result of one test

From: Michael Haggerty <mhagger@alum.mit.edu>
Date: Mar 31, 2014, 21:49 UTC
Message-ID: <5339E2FC.8080403@alum.mit.edu>
In-Reply-To: <xmqqppl23vjl.fsf@gitster.dls.corp.google.com>

On 03/31/2014 11:30 PM, Junio C Hamano wrote:

Show 25 quoted lines

> Michael Haggerty <mhagger@alum.mit.edu> writes:
> 
>> The test
>>
>>     stdin -z create ref fails with zero new value
>>
>> actually passes an empty new value, not a zero new value.  So rename
>> the test s/zero/empty/, and change the expected error from
>>
>>     fatal: create $c given zero new value
>>
>> to
>>
>>     fatal: create $c missing <newvalue>
> 
> I have a feeling that "zero new value" might have been done by a
> non-native (like me) to say "no new value"; "missing newvalue"
> sounds like a good phrasing to use.
> 
>> Of course, this makes the test fail now, so mark it
>> test_expect_failure.  The failure will be fixed later in this patch
>> series.
> 
> That sounds somewhat strange.  Why not just give a single-liner to
> update-ref.c instead?

This is because there really is a difference between the two errors, and "git update-ref" tries to emit distinct error messages for them:

* "zero new value" means that the new value was 0{40}
* "missing <newvalue>" means that the new value was absent

The problem is that it is not distinguishing between these two cases correctly, and fixing *that* is more than a one-liner.

Michael

-- 
Michael Haggerty
mhagger@alum.mit.edu
http://softwareswirl.blogspot.com/