|
|
Boost : |
From: David Abrahams (dave_at_[hidden])
Date: 2004-09-28 06:58:00
"Victor A. Wagner Jr." <vawjr_at_[hidden]> writes:
> nope, nor does:
> if( (unsigned)(*endname + 1) > 256 || !isspace(*endname) && *endname
> != '\\' && *endname != '/' )
>
> the problem just gets postponed and eventually someone tries to access
> 0xfdfdfdfd
>
> however, changing the malloc back several lines to be one larger, then
> putting a '\0' at the end _does_ seem to fix it.
>
> Checked into CVS.
Thanks!
-- Dave Abrahams Boost Consulting http://www.boost-consulting.com
Boost list run by bdawes at acm.org, gregod at cs.rpi.edu, cpdaniel at pacbell.net, john at johnmaddock.co.uk