Sigh, forgot to reply all. Bradley
----- Forwarded message from Bradley Baetz <[EMAIL PROTECTED]> ----- Date: Wed, 12 Feb 2003 07:22:17 +1100 From: Bradley Baetz <[EMAIL PROTECTED]> To: Tim Bunce <[EMAIL PROTECTED]> Subject: Re: Bradley Baetz contact info In-Reply-To: <[EMAIL PROTECTED]> User-Agent: Mutt/1.4i On Tue, Feb 11, 2003 at 02:38:06PM +0000, Tim Bunce wrote: > Goggle says [EMAIL PROTECTED] > Google doesn't know that I've graduated :) > http://www.rosat.mpe-garching.mpg.de/mailing-lists/dbi/2002-12/msg00013.html > > Tim. I had a report (via bugzilla mail lists) of this failing under 5.6.0 on OSX, but I haven't had time to test yet. I'm wondering if this is just the tests themselves which have the problem - the file-based taint stuff was a bit fragile, and its possible that the different taint behaviour may cause this. 5.005 was more fragile = | $foo = 1 if $tainted_var; | sometimes taints in 5.005, but not 5.6.1, for example. I'll see if I can find someone who has 5.005_03 for me to test on. If someone could test by trying DBI->prepare(substr("$0$^X",0,0)) - that should fail taint tests if TaintIn is set, regardless of perl version. Bradley ----- End forwarded message -----
