Re: [PATCH 2/3] Teach 'git remote' that the config var branch.*.rebase can be 'interactive'

From: Johannes Schindelin <johannes.schindelin@gmx.de>
Date: Jan 13, 2016, 12:03 UTC
Message-ID: <alpine.DEB.2.20.1601131302340.2964@virtualbox>
In-Reply-To: <vpq7fjdyfvu.fsf@anie.imag.fr>

Hi Matthieu,

On Wed, 13 Jan 2016, Matthieu Moy wrote:

Show 20 quoted lines

> Johannes Schindelin <johannes.schindelin@gmx.de> writes:
> 
> > diff --git a/builtin/remote.c b/builtin/remote.c
> > index 6694cf2..0af8300 100644
> > --- a/builtin/remote.c
> > +++ b/builtin/remote.c
> >  static struct string_list branch_list;
> > @@ -312,6 +312,8 @@ static int config_read_branches(const char *key, const char *value, void *cb)
> >  				info->rebase = v;
> >  			else if (!strcmp(value, "preserve"))
> >  				info->rebase = 1;
> > +			else if (!strcmp(value, "interactive"))
> > +				info->rebase = INTERACTIVE_REBASE;
> 
> What happens if one has branch.*.rebase=interactive, and wants to make
> an exception? Does
> 
>   git pull --rebase=true
> 
> cancel the 'interactive' part?

It is the same situation as before (just substitute a branch.*.rebase=preserve setting): yes, the config is parsed first, then the command line, so the command line wins.

> I guess it is, but if so I think it should be tested and documented.

Is this really necessary, given that the behavior has not changed from before?

Ciao, Dscho