William A. Rowe, Jr. wrote:
> From: "Jim Jagielski" <[EMAIL PROTECTED]>
> Sent: Tuesday, February 20, 2001 2:48 PM
>
> >
> > yes, part of the feedback is whether this is an ignorable error...
> > I'm not sure it is...
>
> I'll second either Greg or you deciding to fix, dump, bump and roll. I looked
> at this problem early today, and my first gut instinct was an ap_pstrdup before
> we muck with it. It clearly could be lethal, but I can't see a specific example
> to hold up and say "We Broke This ." Perhaps noone but the mod_perl or mod_python
> folks could be bit, but that doesn't matter - if someone will be bit - let's just
> chuck it.
>
I think the decision has been made already (got to get that Xterm
installed in my gray matter one of these days... it's dangerous
to be offline for anything resembling minutes nowadays :) :) ).
I couldn't find a specific example either, nor did any testing
appear to tickle it, but also wasn't comfy with this and it
did seem like a bug just waiting to be hit us. Thanks
to Greg for making sure that others were looking at this. The
other feedback was about tmpnam and the UnixWare 7/SCO warnings
(the ap_getpass one I'm sure is innoculous, the mod_include
ones look weird). I'd like for us to settle on these as well.
--
===========================================================================
Jim Jagielski [|] [EMAIL PROTECTED] [|] http://www.jaguNET.com/
"Casanova will have many weapons; To beat him you will
have to have more than forks and flatulence."
