Re: [PATCH v15 7/7] t/t7507: tests for broken behavior of status

From: Junio C Hamano <gitster@pobox.com>
Date: May 3, 2016, 15:47 UTC
Message-ID: <xmqq4mafp2hg.fsf@gitster.mtv.corp.google.com>
In-Reply-To: <CAPig+cR7pPHZv_z3G+BsLPqP7WYSVUb_7c2qmM+0y-TFeWjaSg@mail.gmail.com>

Eric Sunshine <sunshine@sunshineco.com> writes:

Show 16 quoted lines

>>>> One previous commit (title: commit: add a commit.verbose config
>>>> variable) introduced a new config variable named commit.verbose,
>>>> so care should be taken that it would not affect the behavior of
>>>> status.
>>>>
>>>> Another previous commit (title: "parse-options.c: make OPTION_COUNTUP
>>>> respect "unspecified" values") changes the initial value of verbose
>>>> from 0 to -1. This can cause git-status to display a verbose output even
>>>> when it isn't supposed to.
>>>> ...
>
> This patch's commit message is perhaps a bit too detailed about what
> could have gone wrong in earlier patches in this series; indeed, it
> misled Junio into thinking that patches in this series did break
> behavior, when in fact, it was instead previous rounds of this series
> which were buggy.

Indeed. Please forget everything I said about expect-failure, if the top two paragraphs are describing breakages that this series does *NOT* introduce. I was misled by them--and others will, too. These two paragraphs do not belong to the log message.

Thanks for clarifying.