At 05:26 PM 28/04/2004 +1200, you wrote:

>Yes, UCS4 is the 4 byte unicode. As a string it would have lots of nulls
>in it, which would make working with standard code in C a little too
>interesting. So maybe we talk to it in utf-8? Although trying that using
>Database Workbench wasn't successful...

I guess not. Firebird has only UNICODE_FSS and it's a three-byte
set. Anyway, PJ will be online later...AFAIR, he made up an experimental
kit which can be downloaded somewhere...and I recall words like "plane"
floating around.

For Fb 1.5.1 (currently in pre-release testing) a character set NONE client
has been made capable of passing strings back and forth between client and
server without the transliteration errors. But my guess is you will still
be stumped trying to work with 4-byte characters, unless you decide to use
them output-only (i.e. no search/matching capability, no verification) and
store them as varchar(n*4) in OCTETS, i.e. "just a string of bytes".

/heLen