On Fri, Jan 28, 2000 at 08:33:15PM +0100, [EMAIL PROTECTED] wrote:
> I'm using assert() like statements in my code for self checking. Twice
> today, such a forced program termination reported more than I expected:
>
> - snip --
>
> Attempt to free unreferenced scalar during global destruction.
> Use of uninitialized value during global destruction.
> Explicit blessing to '' (assuming package main) during global
> destruction.
>
> - snip --
>
> As I understand perldiag, this seems to be caused by some C code. I do
> not know if this is Event code. Maybe it is (because the script is a
> server based on Event and I never wrote C extensions of Perl yet), so I
> post it here. Does anyone know how to localize where such a message
> comes from?
Yes, I think there are some bug in the cleanup phase of Event. Part of
the problem is that perl's global destruction phase is somewhat
unpredictable. This is a known bug in perl. I'm not sure if Event can
work around this bug, so I haven't tried very hard to fix it. Is it
really important? I figured it was only important if you were embedding
perl in another application.
--
"Does `competition' have an abstract purpose?"
via, but not speaking for Deutsche Bank
