Re: How Oracle handling multilingual character strings

From: Bastorff <nospam_at_localhost.zz>
Date: Sat, 23 Sep 2000 21:27:17 +0200
Message-ID: <vv7z5.13036$Fl2.112722@nntpserver.swip.net>

"Eric Leung" <eleungks_at_netvigator.com> wrote in message news:39C8FED0.B318457E_at_netvigator.com...
> We are also having multilingual problem, we are using Oracle 8.1.6. And
>
> the
> Oracle is set up with UTF-8 as the charater set and national character
> set.
>

UTF-8 is an encoding method, not a character set/repertoaire. It could imply a (Unicode) character set though, as defined by Oracle manuals.

> Does anyone know how many bytes does a Chinese (Simplified or
> Traditional)
> character take?
>

With UTF-8, how many octets used depends on the character code point, it could be one octet or four, or even more. I guess it depends on how you define the characters.

There are plenty of "unicode" resources out there... A good tutorial to start with http://www.hut.fi/u/jkorpela/chars.html

> We are inserting and updating the database through JDBC.
> Do we need to process the character string before pass it to the
> database?
>

I don't know the program design issues here, but it would seem strange if this wasn't "hidden" by Oracle?

Regards,
/Fad

> Eric Leung
>
>
Received on Sat Sep 23 2000 - 14:27:17 CDT