new PMCs

[Dan Sugalski]

Okay, so as I'm working on the very, very late python bytecode 
translation documentation I'm coming across stuff we need to 
consider, and this is the first of them: The generic new PMC. (No, 
I've not yet finished the doc, nor have I dug through most of the 
past few week's mail. First things first, though it may all come in a 
big dump later)

So, anyway, ponder the new PMC. A common thing, lots of code needs to 
create new PMCs, more for us than other interpreters, as we don't do 
nearly so much implicit PMC creation. (Arguably we ought to, but 
that's a fight for a different time)

What I'm finding is that there's a *lot* of:
   new Px, Undef
in the code I'm seeing, both what I'm creating at work as part of the 
compiler and what I'm working on for the python code generator. (And 
yeah, I know I'm running into Leo's work here. We'll merge the stuff 
in a bit) This is not necessarily a problem, but... it happens so 
often, it seems worth making:

   new Px
a synonym for
   new Px, Undef
and I think I'd like to do just that. It'd shrink the bytecode a fair 
amount, and this is going to be damned common, so I think it'd be for 
the best. (Having generic new PMCs allocated by C code be Undef PMCs 
also makes some sense here)
--
				Dan

--------------------------------------it's like this-------------------
Dan Sugalski                          even samurai
[EMAIL PROTECTED]                         have teddy bears and even
                                      teddy bears get drunk