Re: unicode character set

It depends on the characters (languages) you have to store.

If you need to support English only, AL32UTF8 is the best choice, since each char will use exactly 1 byte.

For most Western European languages, AL32UTF8 as well, for example in Italian we have

1) 95% of chars are the same as English, so they take 1 byte
2) 4.99% are accented chars, they take 2 bytes
3) 0.01% is EURO currency symbol, it takes 3 bytes

Do read the Globalization Support Guide, especially the chapter "Supporting Multilingual Databases with Unicode".

Don't use the legacy UTF8, it has been replaced by AL32UT8.

HTH
Alberto

On 5/4/07, genegurevich_at_discoverfinancial.com <genegurevich_at_discoverfinancial.com> wrote:
>
> Everybody:
>
> How do I pick a unicode character set? I need to build a new database and
> the vendor's requirement is a unicode char set. There are several unicode
> character sets. How do I decide on which one to use? Are there any
> rules/criteria?
>
> thank you
>
> Gene Gurevich
>
>
> --
> http://www.freelists.org/webpage/oracle-l
>
>
>

-- 
Alberto Dell'Era
"dulce bellum inexpertis"
--
http://www.freelists.org/webpage/oracle-l