Oracle FAQ Your Portal to the Oracle Knowledge Grid
HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US
 

Home -> Community -> Usenet -> c.d.o.server -> Re: db character set

Re: db character set

From: Laurenz Albe <invite_at_spam.to.invalid>
Date: 25 Jan 2006 08:52:25 GMT
Message-ID: <1138179144.1031@proxy.dienste.wien.at>


Frank van Bortel <frank.van.bortel_at_gmail.com> wrote:

>> Is it possible to use a different character set (e.g. UTF8 instead of
>> ASCII) for a single schema or is it a global setting for the db instance?
>> 
>> If the latter, is it possible to change the character set for an
>> existing db instance? How?

>
> No. Yes. As per documentation on http://tahiti.oracle.com.
> But only if the set you want to migrate *to* is a superset of
> what you currently use (so UTF->ASCII is out of the question).
>
> And do not forget about the National Character set (UTF by default),
> as addition to the Database Character set

I would like to elaborate on these correct statements.

You cannot have a different character set for part of the database, but from Oracle 9 on you can have a different character set for a text column.

There is a 'national character set' that you chose at database creation time. It is a Unicode character set, and it's best to take AL16UTF16.

You can use NCHAR, NVARCHAR and NCLOB columns, which are in the chosen national character set instead of the database character set.

Maybe that can solve your problem.

As for changing the character set, it is impossible in all but very few cases. Basically, if your database is not US7ASCII, you're out of luck.

There's a chapter 'Character Set Migration' in the Database Globalization Support Guide that might be of interest for you.

Yours,
Laurenz Albe Received on Wed Jan 25 2006 - 02:52:25 CST

Original text of this message

HOME | ASK QUESTION | ADD INFO | SEARCH | E-MAIL US