From: Junio C Hamano <gitster@pobox.com>
Date: Thu, 13 Jul 2017 19:30:33 GMT
Subject: Re: [PATCH] git-p4: parse marshal output "p4 -G" in p4 changes
Message-ID: <xmqqa848w2w6.fsf@gitster.mtv.corp.google.com>
In-Reply-To: <CAKYtbVaxR0sdL_k=vy-aT5wEzvCTzDcM6Q-i0hO6jLMzjEUwmA@mail.gmail.com>

Miguel Torroja <miguel.torroja@gmail.com> writes:

> I've just sent in reply to your previous e-mail three different patches.
>
> * The first patch is just to show some broken tests,
> * Second patch is to fix the original issue I had (the one that
> initiated this thread)
> * Third patch is the one that filters out "info" messages in p4CmdList
> (this time default is reversed and set to False, what is the original
> behaviour). The two test cases that are cured with this change have to
> set explicitely skip_info=True.

The approach looks reasonable.  By having tests that expect failure
upfront, the series clearly shows how the code changes in later
steps make things better.

Thanks.  Will replace.