From: David Kastrup <dak@gnu.org>
Date: Tue, 04 Feb 2014 21:48:57 GMT
Subject: Re: [PATCH] blame.c: prepare_lines should not call xrealloc for every line
Message-ID: <87zjm6v99y.fsf@fencepost.gnu.org>
In-Reply-To: <xmqqr47i7dt4.fsf@gitster.dls.corp.google.com>

Junio C Hamano <gitster@pobox.com> writes:

> David Kastrup <dak@gnu.org> writes:
>
>> Ok, I now wrote
>>
>> 	for (p = buf;; num++, p++) {
>> 		p = memchr(p, '\n', end - p);
>> 		if (!p)
>> 			break;
>> 	}
>
> Looks still wrong (perhaps this is a taste issue).
>
> 	num++ is not "loop control", but the real action of this
> 	loop to count lines.  It is better left inside.

Ok.

> 	p++ is "loop control", and belongs to the third part of
> 	for(;;).

No, it isn't.  The "real" loop control is the p = memchr line.  p++ only
skips over the newline.

> 	Isn't the normal continuation condition "p < end"?

memchr returns NULL when not finding anything any more.

> so something like
>
> 	for (p = buf; p < end; p++) {
>         	p = find the end of this line
>                 if (!p)
>                 	break;
> 		num++;
> 	}
>
> perhaps?

Would crash on incomplete last line.

-- 
David Kastrup