Re: [PATCH v4 4/4] sha1_name: minimize OID comparisons during disambiguation

From: Jeff King <peff@peff.net>
Date: Oct 10, 2017, 12:36 UTC
Message-ID: <20171010123634.3hdqxpo6mbl6jhbx@sigill.intra.peff.net>
In-Reply-To: <338aab6a-1181-d740-1bf2-2ac86749a6b2@gmail.com>

On Tue, Oct 10, 2017 at 08:16:27AM -0400, Derrick Stolee wrote:

Show 20 quoted lines

> > > +	mad->init_len = 0;
> > > +	if (!match) {
> > > +		nth_packed_object_oid(&oid, p, first);
> > > +		extend_abbrev_len(&oid, mad);
> > If we have zero objects in the pack, what would nth_packed_object_oid()
> > be returning here?
> > 
> > So I actually think we do want an early return, not just when
> > open_packed_index() fails, but also when p->num_objects is zero.
> > 
> > -Peff
> 
> Sorry about this. I caught this while I was writing my cover letter and
> amended my last commit to include the following:
> 
>     if (open_pack_index(p))
>         return;
> 
> After I amended the commit, I forgot to 'format-patch' again. I can send a
> diff between the commits after review has calmed.

OK, I think that makes more sense. But note the p->num_objects thing I mentioned. If I do:

  git pack-objects .git/objects/pack/pack </dev/null

then I have a pack with zero objects, which I think we'd similarly want to return early from. I.e., I think we need:

  if (p->num_objects)
	return;

Technically that also covers open_pack_index() failure, too, but that's a subtlety I don't think we should rely on.

-Peff