[Python-Dev] Type/class (original) (raw)

Steven D. Majewski sdm7g@Virginia.EDU
Sat, 12 May 2001 16:53:26 -0400 (EDT)

On Sat, 12 May 2001, Guido van Rossum wrote:

Interesting. So is the rest of this thread, but since Python is not a prototype language and is unlikely to become one, I'd like to mention that Python 2.2 will likely allow you to choose either paradigm, on a per-class basis, using metaclasses.

As I said earlier: the only advantage would be if it could simplify things "under the hood" (compared to metaclasses) but could still provide the same Class semantics (with maybe a "proto" declaration sneaking it's nose in under the tent.) But I have no immediate idea on how to do that, and it sounds like you're pretty far along into an implementation already.

I'm finding metaclasses in Python useful for different things than they are in Smalltalk, and I expect that they will continue to play a less important role. But they are important because they control many "policy" aspects of Python classes/types: e.g. whether instances have a dict or a specific set of slots (maybe even typed slots), whether changes can be made to a class after it's been created, the semantics of multiple inheritance, and so on.

I guess my practical quesion, which I meant to ask before I got myself sidetracked into preaching prototypes is: How much of the existing plumbing (specifically the Don Beaudry hack) can I rely on in the future for the objective-C/python bridge ? With BOOST and Zope's extension classes relying on it, can I assume that it's being extended rather than replaced ? ( I guess I ought to take a look at the code! )

It's an interesting journey! The book I am reading about this: "Putting Metaclasses to Work" by Ira Forman and Scott Danforth. http://cseng.awl.com/book/0,3828,0201433052,00.html

Thanks for the reference. Talking about interesting journies:

Guido: did you ever imagine back at that first workshop at NIST that you and Python would be where you are today ?

-- Steve Majewski