On Fri, Aug 02, 2002 at 08:55:21AM -0700, Sean O'Rourke wrote:
> I agree -- we should separate new from init at some level. I think the
> "new" opcode should still both allocate and initialize, while the
> normal sequence for getting a PMC should become
>
> x = pmc_new(...);
> x->vtable->init(interp, x);
>
> This would solve the continuation problem without breaking existing pasm
> code. It would involve tweaking a lot of existing C, but such is life.
> Any objections?
So in a loop, for a temporary, one should do this?
x = pmc_new(...);
while (...) {
x->vtable->init(interp, x);
...
}
and there's only one allocation made for the loop, but a fresh temporary
each time?
Nicholas Clark
--
Even better than the real thing: http://nms-cgi.sourceforge.net/
