On Mon, Feb 20, 2012 at 00:55, Joe Gilray <jgil...@gmail.com> wrote: > Hi Rodolfo, > > Thanks for the suggestions. Your re-factoring (no pun intended) of the > conds really cleaned things up. Much appreciated. > > About the unnecessary loops, good point, I simply changed to starting with > end set to 10 instead of 1000 and I think that helps a lot. I know that > there are other gross inefficiencies as well (for example, if > primes-from-to gets called to just add a few primes to the list it is very > slow as it goes through the whole list. A better implementation would > "know" that the values on storedlst are already checked and skip checking > them. >
Maybe you could generate the list of primes using... hmm.. Generators! http://docs.racket-lang.org/reference/Generators.html You could have something like (primes-up-to n) yield one prime at a time, up to n (the number to be factored). []'s Rodolfo Carvalho
____________________ Racket Users list: http://lists.racket-lang.org/users
