From: Jeff King <peff@peff.net>
Date: Thu, 27 Apr 2023 16:57:09 GMT
Subject: Re: [PATCH v3 3/4] parse_commit(): handle broken whitespace-only timestamp
Message-ID: <20230427165709.GC1976451@coredump.intra.peff.net>
In-Reply-To: <xmqqleidl2ch.fsf@gitster.g>

On Thu, Apr 27, 2023 at 09:25:02AM -0700, Junio C Hamano wrote:

> Jeff King <peff@peff.net> writes:
> 
> > In practice this can't cause us to walk off the end of an array, because
> > we always add an extra NUL byte to the end of objects we load from disk
> > (as a defense against exactly this kind of bug). However, you can see
> > the behavior in action when "committer" is the final header (which it
> > usually is, unless there's an encoding ...
> 
> ... or it is a signed commit or a commit that merges a signed tag.
> 
> There is no need for us to be exhaustive here, but I just wondered
> which one of these three commit object headers is more common.  I
> guess the reason "encoding" came to your mind first is because it is
> the oldest among the three.

Mostly the others did not occur to me at all. :)

I expect that "gpgsig" lines are probably the most common these days,
but that may be my biased view (I guess in the kernel workflow it is
probably signed tags).

-Peff