[plt-scheme] Is there a set data-type?

Does something like this sounds interesting and generalizable for infinite
sets?
Mixing hashes on finite sets for speed and recurrence for infinite sets in
the default case (trivial example):

(define h (make-hash))

(for-each (λ(n)(hash-set! h n #t))
         '(0 2 4 6 8)) ; could be a lot larger of course

(define (odd? n)
 (if (> n 0)
      (hash-ref h n
               ; default case:
                (λ()(odd? (- n 10))))
     #f))


> (even? 9876)
#t
> (even? 9877)
#f

Or maybe that's just a well-known fact...

Laurent

On Fri, Feb 5, 2010 at 15:42, Laurent <laurent.orseau at gmail.com> wrote:
> On Fri, Feb 5, 2010 at 15:41, Laurent <laurent.orseau at gmail.com> wrote:
>> On Fri, Feb 5, 2010 at 15:40, Laurent <laurent.orseau at gmail.com> wrote:
>>> On Fri, Feb 5, 2010 at 15:12, Matthias Felleisen <matthias at ccs.neu.edu>
wrote:
>>>>
>>>> On Feb 5, 2010, at 8:29 AM, Todd O'Bryan wrote:
>>>>
>>>>> What do people use for a set data-type? If there isn't one built-in,
>>>>> we're going to use a hash-table with #t as all the values.
>>>>
>>>> Just a tease: how would you use hashtable to represent infinite sets?
>>>
>>> From the top of my head, though that does probably not fit to the
>>> definition of an infinite set:
>>>
>>> (define towards-infinity-and-beyond (make-hash))
>>> (hash-set! towards-infinity-and-beyond 'zero '())
>>> (hash-set! 'me towards-infinity-and-beyond)
>>
>> oops, please read on the last line:
>> (hash-set! towards-infinity-and-beyond) 'me towards-infinity-and-beyond)
>
> rahhh, well you got the idea... (and I'm going to get some sleep)
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.racket-lang.org/users/archive/attachments/20100205/c59384da/attachment.html>