On Wed, Oct 22, 2003 at 08:17:06AM -0700, Bob Goolsby (bogoolsb) wrote: > Actually, that is an argument for running two sets of tests, one vanilla, > the other Tainted. > > And that raises the question "Who's bug is it?" if something passes the > test package under normal conditions, but fails under -T. Are we seeing a > real problem in the Package, or another -T weirdity?
If a module hits a taint bug in the core, its Perl's bug but its the module's problem. Unless the author is going to make the decision to not support taint mode, they've got to be aware of the problem and attempt a work around. Same issue with any core bug I'm afraid. Fortunately, there's an easy way to tell the difference between a proper taint failure and a failure due to a taint bug. If the taint test dies with an "Insecure dependency" you've probably got a normal taint failure. If it fails in any other way differently then a normal run you've got a bug in taint. -- Michael G Schwern [EMAIL PROTECTED] http://www.pobox.com/~schwern/ AY! The ground beef, she is burning my groin! http://sluggy.com/d/990105.html
