Hey Mike, no prefixes don't have counts, letters with prefixes have counts. Your example: Hello Jello Hello Jello
prefix length=8, prefix=llo Jell, letter=o, count=2 That's true for 'o' because 'o' follows both occurrences. Some more examples using your string. prefix length=1, prefix=l, letter=l, count=4 prefix length=1, prefix=l, letter=o, count=4 prefix length=1, prefix=H, letter=e, count=2 Thanks Julian, but I don't really follow how it helps with my problem since the outputs don't look to be what I'm asking for? I think one approach may be to solve the problem for a first k. Start with something like: (k+1) \< 'Hello Jello' | Hell | ello | llo | lo J | o Je | Jel | Jell | ello | First k characters are the prefix, last character is the letter. Yield <prefix>;<letter> For each boxed value, snub to get uniques, , and key to do counts. Then stick it together: <k>; <prefix>; <count> Do that for every k and concatenate. I think something like that might work? Emir ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
