Re: heads-up: git-index-pack in "next" is broken

From: Nicolas Pitre <nico@cam.org>
Date: Oct 18, 2006, 13:13 UTC
Message-ID: <Pine.LNX.4.64.0610180902580.1971@xanadu.home>
In-Reply-To: <Pine.LNX.4.63.0610181159050.14200@wbgn013.biozentrum.uni-wuerzburg.de>

On Wed, 18 Oct 2006, Johannes Schindelin wrote:

Show 13 quoted lines

> Hi,
> 
> On Tue, 17 Oct 2006, Junio C Hamano wrote:
> 
> > +/*
> > + * Even if sizeof(union delta_base) == 24 on 64-bit archs, we really want
> > + * to memcmp() only the first 20 bytes.
> > + */
> > +#define UNION_BASE_SZ	20
> 
> Excuse me for joining the game, but why don't you just use the 
> recently introduced hashcmp() for that purpose? AFAIU you do exactly that, 
> you compare hashes.

Yes, and that is what I did originally.

But that could lead to false assumptions (and this thread already proved this code has its share of false assumption leads already). The thing is that the memory chunk that is being compared is not always the same kind of hash as usually used with hashcmp(). Throughout the code hashcmp() is always used with a 20-byte sha1 digest. In this case it can be either a 20-byte sha1 digest, or a long offset value. And by using hashcmp() I would be afraid someone else could assume the hash is always a sha1 digest which it is not.

Nicolas