Re: [PATCH] t1500-rev-parse: rewrite each test to run in isolation

From: SZEDER Gábor <szeder@ira.uka.de>
Date: Apr 16, 2016, 10:23 UTC
Message-ID: <20160416122302.Horde.jjsXp-czGdXN7lDuc-Vg6pE@webmail.informatik.kit.edu>
In-Reply-To: <xmqqk2k2l0yx.fsf@gitster.mtv.corp.google.com>

Quoting Junio C Hamano <gitster@pobox.com>:

Show 7 quoted lines

> Applying this patch on top of sg/completion-updates topic makes the
> tests much more readable.  Given that sg/completion-updates topic is
> planned to be rerolled ($gmane/287839), I think it would be better
> to do this as a preparatory clean-up patch before it makes the tests
> uglier by doing "add --absolute-git-dir" patch in the middle.
>
> Gábor, what do you think?

Sure, go ahead. A standalone test cleanup patch could graduate to master faster than a 20+ patch series, so it shouldn't be held hostage.

The reroll of sg/completion-updates is almost ready. "Almost", because the refs completion speedup series (another 10+ patches...) is almost ready as well, but needs some final polishing and I want to trickle down some of the test updates into sg/completion-updates. Unfortunately, I don't expect to have time to finish it up in the next few days... but fortunately that'll leave you time to agree on how to rewrite t1500 and Michael to do the reroll ;)

Anyway, if Michael rerolls based on master and you have troubles you don't want to deal with with the conflicts between the reroll and my series, I would say just discard sg/completion-updates for the time being.

Gábor