Excerpts from Tommy Pettersson's message of Thu Jan 17 12:18:19 +0100 2008:
> On Thu, Jan 17, 2008 at 09:26:46AM +0100, Nicolas Pouillard wrote:
> > Excerpts from Tommy Pettersson's message of Wed Jan 16 21:18:15 +0100 2008:
> > > On Wed, Jan 16, 2008 at 07:36:30PM +0000, Nicolas Pouillard wrote:
> > > > Wed Jan 16 20:34:26 CET 2008 [EMAIL PROTECTED]
> > > > * Add --{allow,dont-allow,mark}-conflicts to darcs pull.
> > > >
> > > > This patch also merge the --external option to pull_conflicts_options
> > > > like with apply.
> > >
> > > Hmmm... I don't think merging --external-merger to the
> > > multi-choice option is a good idea. A common case is to have
> > > 'ALL external-merger my_favorite_merger" in the .darcs/defaults
> > > file, and expect 'darcs apply --mark-conflicts' to use
> > > my_favorite_merger.
> >
> > So, does makes sense to extract the external-merger option from
> > the
> > multi-choice option for both pull and apply?
>
> Absolutely! In fact, I thought this was the case for Apply,
> until I just looked at the code and saw it treats
> --external-merge (without 'r' at the end) as a multi-choice of
> the other conflict options.
Yes, that's why I've done the same thing to pull.
> Maybe I was confusing it with the
> --diff-command option, which is how --external-merger (with 'r'
> at the end) should work, imho.
Hum, what is the external-merge*r* thing you talk about, I only know about
external-merge.
> But to have Pull equal Apply in semantics of external-merge
> (whatever they are), as is now done with your patch, is of
> course the right thing.
Yes, but having --external-merge out of this option group could ease the case
you've talk about.
So I vote for having --external-merge out the multi-choice group for both
pull and apply.
--
Nicolas Pouillard aka Ertai
_______________________________________________
darcs-devel mailing list
darcs-devel@darcs.net
http://lists.osuosl.org/mailman/listinfo/darcs-devel
