From: Jeff King <peff@peff.net>
Date: Wed, 27 May 2015 06:50:31 GMT
Subject: Re: [RFC/WIP PATCH 06/11] remote.h: add get_remote_capabilities, request_capabilities
Message-ID: <20150527065030.GE885@peff.net>
In-Reply-To: <CAPig+cRfJKAQ8Q5PF1VfTAGA1njXAshC0RbnMv9cEp4bH_MN7A@mail.gmail.com>

On Tue, May 26, 2015 at 11:25:05PM -0400, Eric Sunshine wrote:

> > +               len = packet_read(in, &src_buf, &src_len,
> > +                                 packet_buffer, sizeof(packet_buffer),
> > +                                 PACKET_READ_GENTLE_ON_EOF |
> > +                                 PACKET_READ_CHOMP_NEWLINE);
> > +               if (len < 0)
> > +                       die_initial_contact(0);
> > +
> > +               if (!len)
> > +                       break;
> > +
> > +               if (len > 4 && skip_prefix(line, "ERR ", &arg))
> 
> The 'len > 4' check is needed because there's no guarantee that 'line'
> is NUL-terminated. Correct?

I think this was just blindly copied from get_remote_heads(). And I
think that code was being overly paranoid. Ever since f3a3214 (Make
send/receive-pack be closer to doing something interesting, 2005-06-29),
the pkt-line reader will add an extra NUL to the buffer to ease cases
like this.

-Peff