From: Junio C Hamano <gitster@pobox.com>
Date: Fri, 28 Feb 2020 20:39:50 GMT
Subject: Re: [PATCH 1/1] midx.c: fix an integer overflow
Message-ID: <xmqqzhd2cup5.fsf@gitster-ct.c.googlers.com>
In-Reply-To: <20200228185525.GB1408759@coredump.intra.peff.net>

Jeff King <peff@peff.net> writes:

>> -	for (i = 0; i < m->num_objects - 1; i++) {
>> +	for (i = 0; i + 1 < m->num_objects; i++) {
>>  		struct object_id oid1, oid2;
>>  
>>  		nth_midxed_object_oid(&oid1, m, i);
> [...]           nth_midxed_object_oid(&oid2, m, i + 1);
>
> Perhaps it would be simpler as:
>
>   for (i = 1; i < m->num_objects; i++) {
>           ...
> 	  nth_midxed_object_oid(&oid1, m, i - 1);
> 	  nth_midxed_object_oid(&oid2, m, i);
> 	  ...
>   }

"Count up while i+1 is smaller than..." looked extremely unnatural and
it was hard to grok, at least to me.  This

	for (i = 0; i < m->num_objects - 1; i++) {

might have been more palatable, but yours is much better.

> Though I almost wonder if we should be catching "m->num_objects == 0"
> early and declaring the midx to be bogus (it's not _technically_ wrong,
> but I'd have to suspect a bug in anything that generated a 0-object midx
> file).

That, too ;-)