Re: Unexpected behavior in (d)iff while interactively working through conflicts--I think diff3 would be nicer

On Dec 14, 2007 4:25 AM, Alan Barrett <apb@cequrux.com> wrote:
> On Thu, 13 Dec 2007, Ben Collins-Sussman wrote:
> > > +<<<<<<< .mine
> > > +moona
> > > +=======
> > > +bar
> > > +>>>>>>> .r2
> > >
> > > What I got was diff2 output, while what I expected was diff3 output.
> > What exactly would diff3 output look like?
>
> I think he means this:
>
> <<<<<<< .mine
> moona
> ||||||| .r1
> foo
> =======
> bar
> >>>>>>> .r2
>
> Both the existing format and the format that I think is wanted are
> produced by the traditional diff3 program. The existing format looks
> like the output from "diff3 -m -E .mine .r1 .r2", while the desired
> format is more like the output from "diff3 -m -A .mine .r1 .r2".

I agree with you about what the desired format is, but I think you're
mischaracterizing the current output.

The current output, the first time you run (d)iff (ie, assuming you
haven't edited the merged file yourself) is essentially diff2(r1,
diff3(mine, r1, r2)).

--dave

-- 
David Glasser | glasser_at_davidglasser.net | http://www.davidglasser.net/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org