From: Junio C Hamano <gitster@pobox.com>
Date: Thu, 16 Aug 2012 23:40:25 GMT
Subject: Re: [RFC/PATCH] t3300-*.sh: Fix a TAP parse error
Message-ID: <7vk3wyxv6e.fsf@alter.siamese.dyndns.org>
In-Reply-To: <20120725205120.GD4732@burratino>

Jonathan Nieder <jrnieder@gmail.com> writes:

> Ramsay Jones wrote:
>> Junio C Hamano wrote:
>
>>> Observing that all well-written test scripts we have begin with this
>>> boilerplate line:
>>>
>>>	test_expect_success setup '
>>>
>>> I wouldn't mind introducing a new helper function test_setup that
>>> behaves like test_expect_success but is meant to be used in the
>>> first "set-up" phase of the tests in a test script.
>
> Neat.  This could be used for later set-up tests, too, perhaps with a
> long-term goal of making non set-up tests independent of each other
> (reorderable and skippable).
>
> [...]
>> [1] For example, what should/will happen if someone uses test_must_fail,
>> test_might_fail, etc., within the test_fixture script? Should they simply
>> be banned within a text_fixture?
>
> Why wouldn't they act just like they do in test_expect_success blocks?
>
> FWIW I find Junio's test_setup name more self-explanatory.  What
> mnemonic should I be using to remember the _fixture name?

I see that I was distracted by the "where does the fixture come
from" and did not follow through.

I think what it does makes sense (I haven't checked all the
redirections, though).  Do we want to resurrect the topic?  It needs
a paragraph in the proposed commit log and t/README to explain the
motivation and the usage.

Thanks.