I've kinda decided to go with 7 bytes of attribute information for
each Unicode character (as opposed to 5). This allows for more
flexibility in seamlessly upgrading to new Unicode specs, and makes my
life as a coder, much easier (and the code will run faster too).
This inflates the database size from 327680 to 458752 bytes.
Anyone have a problem with $classpath _requiring_ a 458752 byte file
for java.lang.Character to work? :)
(and do keep in mind that the entire file is never read into memory.
this is really just a disk space issue.)
--
Paul Fisher * [EMAIL PROTECTED]
