Storing multilingual characters in Oracle

From: Sandeep Dubey <dubey.sandeep@xxxxxxxxx>
To: oracle-l@xxxxxxxxxxxxx
Date: Fri, 29 Apr 2005 09:31:32 -0400

Our database currently storing english characters now needs to store
lanaguages like chinese, korean and Japanese. NLS_CHARACTERSET is
currently WE8ISO8859P1. I need to store multilingual characters in
approximately 20% of the tables. There are three options:

1. Alter the columns to NCHAR. =20
2. Alter database charachter set AL32UTF8
3. Create a new database with AL32UTF8 charcter set and do export import.

Are there any known issues of using Nchar on Oracle 9.2.0.4? I found
on a metalink note recommending NOT to use nchar and rather change
character set. Anyone has any bad experience with Nchar/Nvarchar2?

Is  WE8ISO8859P1 a strict subset of AL32UTF8?=20

Or Option 3 is the best way to go. I am trying to avoid this option
because of outage time I have to take for export and import.

Please share your experience.

Regards,

Sandeep Dubey
--
http://www.freelists.org/webpage/oracle-l

Follow-Ups:

Re: Storing multilingual characters in Oracle
From: Ana Choto

References:
- Quick compare of PL/SQL records
  - From: Ranko Mosic

Storing multilingual characters in Oracle

Other related posts: