... what's with the [ 0 ] at the end?

If you liked that, take another look at kcott's  ( ... )[0][0] approach. That said, it's still not an O(n) solution and therefore not, in general, and all things being equal, to be preferred IMHO.