From: Alex Riesen <raa.lkml@gmail.com>
Date: Wed, 14 Mar 2007 17:15:32 GMT
Subject: Re: [PATCH] Allow git-diff exit with codes similar to diff(1)
Message-ID: <81b0412b0703141015m1f7cb68dkf89fe13a968a30d0@mail.gmail.com>
In-Reply-To: <7vodmvah5e.fsf@assigned-by-dhcp.cox.net>

On 3/14/07, Junio C Hamano <junkio@cox.net> wrote:
> >> > -     return run_diff_files_cmd(&rev, argc, argv);
> >> > +     result = run_diff_files_cmd(&rev, argc, argv);
> >> > +     return rev.diffopt.diff_exit_code ? rev.diffopt.exit_code: result;
> >> >  }
> >>
> >> Yuck.  Let's call the former "exit_with_status" (meaning, the
> >> caller instructed us to do that) and the latter "has_changes".
> >
> > I like "exit_with_status". But has_changes looks confusing
> > good near return value of run_diff_files_cmd, which "has"
> > nothing. Or do you mean to highlight this "difference"?
>
> Maybe 'found_changes' would be a better name, then?
>
>         if (exit_with_status)
>                 return !!found_changes;
>         else
>                 return result;

Nah... has_changes has almost the same meaning,
and I'm weary regarding making it a changing counter
(which one can assume it to be when seeing !!found_changes).
And I already sent the patch off.