Sorry if this double posts. I just wanna say, as a new member, I'm impressed by the speed, volume, and quality of responses in this community. Not being simply told to RTFM or such is a really great change from many other places.
> Or use structure properties to turn them into sequences (and then use for and > friends). That would work for the TO-LIST abstraction or for using a sequence as an intermediate representation for collection types. But, what about something like the Speakable behaviorial type? > I imagine that one could use impersonators for vectors and _mutable_ lists. Racket's a pretty big language. I think everytime I try and do something, I end up in a new unexplored corner of the language. I'll check impersonators out, though according to Sam's comment they won't necessarily work. > You could do that, or you could handle them specially in your generic > function. That's probably the most straight forward way, but it means dealing with built-in types upfront rather than allowing for later extensibility (unless I missed something).
____________________ Racket Users list: http://lists.racket-lang.org/users
