> Hmm, I understand what you're saying, but how ion earth do I create
> a function that reorders the result based on all the different
> characters ^ . * etc that could cause this?
Write a function that strips out the characters you want to ignore
(returning the rest of the given string) and then create an index on
that function.
(Please note: that's an answer to the question that you asked above,
not necessarily a recommendation that you choose this particular
collation strategy).
-- Korry