Le mercredi 24 avril 2002, à 06:48 , Edwards, Waverly a écrit :

> <<<
> if you know the values of SIMM and BOBO you can put these are
> longs also.
>
> in this way you should cut out string transformations and just
> work on longs no?
>>>>
>
> Yep, I definitely wanted to cut out the string comparisons but
> the strings were 1-4 characters in length.  If they were all
> four characters in length I could do just that but the
> variability of the string lengths throws a monkey wrench into
> the gears.  Atleast I thought so until some ideas were proposed
> on how to deal with the string lengths.

well, as tedd has gone to vegas, i'll step in... if your comparaisons
[that is the things on the other side of your case statements]
are known in advance, perhaps a tree, or a hash table?

:-j