From: Jeff King <peff@peff.net>
Date: Sun, 18 Nov 2007 03:12:16 GMT
Subject: Re: [PATCH 1/3] send-pack: track errors for each ref
Message-ID: <20071118031215.GB4560@sigill.intra.peff.net>
In-Reply-To: <7v6400z6tk.fsf@gitster.siamese.dyndns.org>

On Sat, Nov 17, 2007 at 05:21:59PM -0800, Junio C Hamano wrote:

> > Sorry, I don't see the style nit you're mentioning here.
> 
> I think Daniel is referring to "Put 'else' on the same line as
> the brace that closes the corresponding 'if' opened", iow:
> 
> 	if (...) {
>         	...
> 	} else {
> 		...
> 	}

Ah. We are really inconsistent about that (I count about 333 non-cuddled
(but possible to cuddle) versus 499 cuddled). Should we put something in
the CodingGuidelines? I am also happy to roll this change into a style
patch.

For reference, I counted with:

# cuddled:
grep '} *else' *.c | wc -l

# possible but non-cuddled
# when we find an 'else', if the previous line had a '}'
# on it, then it is a hit. There are a few false positives
# when the "} else" is from an outer block scope
# And no, this sed invocation probably isn't totally portable. :)
sed -n '/else/{x; /}/{x;p}}; h' *.c | wc -l
# or to check the accuracy
sed -n '/else/{x; /}/{p;x;p}}; h' *.c

-Peff