William D Clinger wrote:
> Per Bothner wrote:
>  > Random accesses to a position in a string that has not
>  > been previously accessed is not in itself useful.
>
> Untrue.  The Boyer-Moore algorithm for fast string
> searching uses random accesses to positions that have
> not been previously accessed [1].

Yes, but I think you can implement this for UTF-8 or UTF-16 strings
using offsets to the underlying bytes or shorts. I don't think that you
need character offsets.

Regards,

Alan
--
Dr Alan Watson
Centro de Radioastronomía y Astrofísica
Universidad Astronómico Nacional de México