more problems with 10-fix_potential_overflows.dpatch

[Henrique de Moraes Holschuh]

On Wed, 21 Jul 2010, Patrick Goetz wrote:
> The patch listed below switches from using strcpy to strlcpy, which
> is questionable due to the lack of standardization of strlcpy.  But

That is an argument you should never make.  If a system doesn't have
strlcpy, it is uninteresting crap, and it will have to provide one through
gnulib or something.

Debian systems *always* have strlcpy, anyway.

> moreover, does it even make sense to take the sizeof of a UNIX
> socket?

What you're talking about?  It is taking the size of an array inside a
structure, so as not to write past the end of that array.

> @@ -1246,7 +1246,7 @@
> 
>    /* can't have this and a file for input */
>    sunsock.sun_family = AF_UNIX;
> -  strcpy(sunsock.sun_path, output_socket);
> +  strlcpy(sunsock.sun_path, output_socket, sizeof(sunsock.sun_path));
>    unlink(output_socket);

-- 
  "One disk to rule them all, One disk to find them. One disk to bring
  them all and in the darkness grind them. In the Land of Redmond
  where the shadows lie." -- The Silicon Valley Tarot
  Henrique Holschuh