The open opcode

[Moritz Lenz]

Jonathan Leto wrote:
> Would making the open opcode a dynop create a noticeable decrease in
> performance? That is the only reason I can see that we would not want
> to go that route.

As I mentioned on #parrot, it's not clear to me why 'open' needs to be
an opcode (or a dynop, for that matter) at all. It sounds like the thing
you'd find as a method call in a (core) library, really.

So deprecating 'open' as an opcode (and replacing it by an appropriate
library function, if that doesn't exist already) seems to solve that
problem :-)

Cheers,
Moritz