From: Jakub Narebski <jnareb@gmail.com>
Date: Mon, 03 Sep 2007 09:52:51 GMT
Subject: Re: [PATCH] gitweb: Fix file name quoting in simplified out patch detection for --cc output
Message-ID: <200709031152.51373.jnareb@gmail.com>
In-Reply-To: <7vabs4oaxh.fsf@gitster.siamese.dyndns.org>

Junio C Hamano wrote:
> Jakub Narebski <jnareb@gmail.com> writes:
> 
> > diff --git a/gitweb/gitweb.perl b/gitweb/gitweb.perl
> > index 8c1e02c..5d424e5 100755
> > --- a/gitweb/gitweb.perl
> > +++ b/gitweb/gitweb.perl
> > @@ -3140,8 +3140,8 @@ sub git_patchset_body {
> >  			# compact combined diff output can have some patches skipped
> >  			# find which patch (using pathname of result) we are at now
> >  			my $to_name;
> > -			if ($patch_line =~ m!^diff --cc "?(.*)"?$!) {
> > -				$to_name = $1;
> > +			if ($patch_line =~ m!^diff --cc ("?.*"?)$!) {
> > +				$to_name = unquote($1);
> >  			}
> 
> This patch obviously makes sense, but I wonder why you use the
> form m!foobar! here, instead of far simpler to read /foobar/
> form?

Consistency (kind of). For the ordinary diff output I use m!regexp!
instead of simpler /regexp/ to avoid 'leaning toothpicks' syndrome:

  m!^diff --git "?a/.*"? "?b/.*"?!

But as it is now not near, then I guess /regexp/ would be better...

-- 
Jakub Narebski
Poland