Re: [PATCH 2/2] expand --pretty=format color options

From: Johannes Schindelin <johannes.schindelin@gmx.de>
Date: Jan 20, 2009, 14:58 UTC
Message-ID: <alpine.DEB.1.00.0901201553200.5159@intel-tinevez-2-302>
In-Reply-To: <20090120142301.GC10688@sigill.intra.peff.net>

Hi,

On Tue, 20 Jan 2009, Jeff King wrote:

Show 19 quoted lines

> On Tue, Jan 20, 2009 at 11:36:08AM +0100, Johannes Schindelin wrote:
> 
> > > > Of course. But the problem is that rev-list is _already_ contaminated 
> > > > by --pretty=format:%Cred. Or do you mean, you really want rev-list to 
> > > > unconditionally output color in such a case?
> > > 
> > > No, rev-list is not contaminated with UI color options.  %Cred _always_ 
> > > outputs the color, even when the user turned off color explicitely, 
> > > using --no-color.
> > 
> > BTW I would find it very logical for rev-list not to output any color at 
> > all when %C(yellow) is specified, as your code respects the diff UI 
> > options, which are implicitly turned off for rev-list (as rev-list is no 
> > UI), just like the coloring of "commit <name>" is implicitly turned off 
> > for rev-list.
> 
> Now I'm confused. Should color in --pretty=format always be on, or
> should it respect color settings? You seem to be advocating both sides
> in the two paragraphs.

No, I am just very bad at relating my thoughts.

What I tried to say is "plumbing does not, and should not, change behavior depending on diff.color".

With %Cred, I was just lazy, and did not make a check if diff.color is true, which I regret now.

But then, its behavior still does not depend on diff.color when using plumbing.

It does not even depend on it when using porcelain :-)

Show 5 quoted lines

> The behavior I would propose it along the lines of:
> 
>  - plumbing _always_ has color off
> 
>  - porcelain respects color.* config, --color, etc

Right, that'd be the sane behavior, even for %Cred.

Ciao, Dscho