> ... you can't implement Python generators as Lisp macros in any reasonable > way. You could do them in Scheme using call-with-current-continuation > but Lisp doesn't have that.
Well, okay, Scheme [same thing (to me), although I realize that they aren't, quite -- and CWCC is one place where they aren't!] But I don't follow why you can't come very close by appropriate macrification of closures. OTOH, this could be my lack of knowledge; it's possible that Python has somehow gone beyond what one can reasonably do in this way. But anyway, this wasn't the point of my post; rather, my point was that Python can't be extended at all (or at least not in the same way that Lisp can be), not that a given extension can or cannot be done in Lisp. -- http://mail.python.org/mailman/listinfo/python-list
