Re: Database character set conversion.

From: Martin Berger <martin.a.berger_at_gmail.com>
Date: Wed, 12 Aug 2009 19:28:33 +0200
Message-Id: <E7C4DE10-2A9B-428D-AEA6-E67AD4E0B03C_at_gmail.com>

Fred,
as mentioned by others, csscan is a good start! all char, varchar2, nvarchar2 etc columns can be more or less easily switched to ... varchar2(<size> CHAR) {instead of varchar2(<size> BYTE) which it will be probably in your case) by going through *_tab_cols
CLOBs are a little bit more fun:
please check Note:257772.1 - CLOBs and NCLOBs character set storage in Oracle Release 8i, 9i and 10g , as you will probably double the size of your CLOBs, or other way round, if CLOBs are stored in row right now, they probably fall out of row just because of your cenversion. (with all its aspects on storage size and performance) hth
  Martin

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

Am 12.08.2009 um 09:58 schrieb Fred Tilly:



> All,

>

> We have two applications that use one database, and the database is  

> configured with character set WE8ISO8859P1. One of the applications  

> needs to handle multiple languages so the vendor wants us to convert  

> ot a 16 bit character set. However the other application in the  

> database has not been tested against a 16bit database, and there are  

> a lot of tables with fields defined as varchar2(4000) and clobs.

> Has anyone actually done such a conversion and is there anything we  

> should be looking out for.

> P.S. Please do not suggest two databases as management have already  

> said no.

> Thanks

> Fred.

> --

> http://www.freelists.org/webpage/oracle-l

>

>


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

Received on Wed Aug 12 2009 - 12:28:33 CDT