From: Johannes Schindelin <johannes.schindelin@gmx.de>
Date: Wed, 22 Jun 2016 16:14:43 GMT
Subject: Re: [PATCH 5/5] format-patch: avoid freopen()
Message-ID: <alpine.DEB.2.20.1606221811300.10382@virtualbox>
In-Reply-To: <xmqqeg7pns0a.fsf@gitster.mtv.corp.google.com>

Hi Junio,

On Wed, 22 Jun 2016, Junio C Hamano wrote:

> Johannes Schindelin <Johannes.Schindelin@gmx.de> writes:
> 
> > But there's a rub... If you specify --color *explicitly*, use_color is set
> > to GIT_COLOR_ALWAYS and the file indeed contains ANSI sequences (i.e. my
> > analysis above left out the command-line part).
> 
> Heh, the command-line is the _ONLY_ thing I raised, as we knew
> ui.color is not an issue in this codepath, in $gmane/297757.
> 
> Going back to that and reading again, I suggested to check with
> GIT_COLOR_AUTO (i.e. if it is left to "auto", disable it) because I
> think the former is a much more future-proof way (imagine that we
> may add --color=<some new setting> in the future) than checking with
> GIT_COLOR_ALWAYS (i.e. if it is not explicitly set to "always",
> disable it).

Well, if I change `rev.diffopt.use_color != GIT_COLOR_ALWAYS` to
`rev.diffopt.use_color == GIT_COLOR_AUTO`, then the files will contain
ugly ANSI color sequences if I run `git format-patch -o . -3`.

The reason is, as I tried to explain, that the use_color field is *not*
initialized to GIT_COLOR_AUTO (which is equivalent to 2), but to -1.

So the difference between your version and mine is that mine works ;-)

Ciao,
Dscho