Yeah I agree it is not the best solution, I kinda mentioned it might well be slow. But today is grep week so what can you do? As you point out it will slow in a geometric fashion. I have added an update that uses a hash solution - with grep of course :-)

Update

Just benchmarked the code on a 10,000 element data set, results not as bad as I feared - the iterative method is only 3 orders of magnitude slower. I think three orders of magnitude sounds better than 1000 times!

cheers

tachyon

s&&rsenoyhcatreve&&&s&n.+t&"$'$`$\"$\&"&ee&&y&srve&&d&&print