RE: Unicode character set

>
> Test it. End evaluate sensitivity of endiannes.
> My memory reminds me that UCS2 strings may have
> introducer string to indicate endianness.
> This issue needs to be analysed. Anyway, adjusting
> endiannes is a matter of adding a couple ifdefs.
>
> Try to enable it in RDB$CHARACTER_SETS or ask David
> what is the purpose of mentioned charset.
>
> In fact, I'm not very interested directly in this issues now (but
> I was interested in WIN1251 charset and fixed a couple tweaks
> in it). It is not so hard, I think you may fix it yourself in
> any case.
>
> > Peter
>
> Nickolay Samofatov
>

UNICODE character set actually sort-of worked as a general
character set -- it was supposed to be part of the original
Interbase 4.0 release.

It was very late in the release cycle that I realized
the translate-endian-format problems in the remote
communication protocol couldn't be solved without
rev'ing the protocol ID, and the politics of the situation
didn't allow that (at the time).

And, after that, Internationalization never made it to
the top of the priority list for ib 5.x or 6.x - so it
never got fixed.

UNICODE worked fine when the local and remote machine
implemented the same endian format, but due to the
remote problems we pulled it from 4.0 at near to
the last minute.

Dave