On 2006-03-23, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote: > The solution that would have the most utility would be one where the > elements are generated one-by-one, loop-like, so that they can be used > in the body of a loop, and to avoid the fact that even with exclusion > the cardinality of the target set EX^n could be in the millions even > with a full list of wc's, that is, a list containing at least one wc of > every length in 2..(n-1). I don't know enough Lisp, Haskell or > Qi/Prolog to know if the solutions so far can be modified to do this. > The Python program is too slow for large sets.
In Haskell you can get this essentially for free, due to its laziness. -- Aaron Denney -><- -- http://mail.python.org/mailman/listinfo/python-list