Peter Sinnott <[EMAIL PROTECTED]> wrote:
> I had a quick look and things looked ok. The parents are stored in a
> hash but they are retrieved sorted by their value which increments as
> parents are added.
> What was different between the 2 machines ( perl versions/os/whatever )?
It's ok on:
$ un
On Fri, Mar 25, 2005 at 11:31:58AM +0100, Leopold Toetsch wrote:
> Matt Diephouse <[EMAIL PROTECTED]> wrote:
> > Leopold Toetsch <[EMAIL PROTECTED]> wrote:
> >> Seems to be another ordering problem, where a hash is used instead of an
> >> array with inherited vtable methods. pylong.dump has already
Matt Diephouse <[EMAIL PROTECTED]> wrote:
> Leopold Toetsch <[EMAIL PROTECTED]> wrote:
>> Seems to be another ordering problem, where a hash is used instead of an
>> array with inherited vtable methods. pylong.dump has already the wrong
>> entry:
>>
>> 'destroy' => 'default',
>>
>> in the suoe
Leopold Toetsch <[EMAIL PROTECTED]> wrote:
> Seems to be another ordering problem, where a hash is used instead of an
> array with inherited vtable methods. pylong.dump has already the wrong
> entry:
>
> 'destroy' => 'default',
>
> in the suoer hash.
I don't think it's an ordering problem (
Investigating some recent dynclass/py*.t I found another problem in the
PMC compiler (classes/pmc2c2.pl, lib/Parrot/Pmc2c.pm).
On one computer I have in dynclasses/pylong.c temp_base_vtable
Parrot_default_destroy,
on the second I got:
Parrot_BigInt_destroy,
The latter is cor
