Johan Tibell <johan.tibell@gmail.com> writes:

> Typo? Bloom filters have O(1) lookup and tries O(m) lookup where m is the
> number of characters in the string.

Typically you need to examine the (whole) search string in order to
compute the hash function, so I think it is fair to consider them both
O(m).