Character set conversion to UTF8 from WE8ISO8859P1 to UTF8 
Author Message
 Character set conversion to UTF8 from WE8ISO8859P1 to UTF8

Hi

We r trying to change database character set from WE8ISO8859P1 to UTF8
on windows NT having oracle 8.0.4.
We have tried all ways as mentioned in oracle docs as well as
Metalink.But getting the error that UTF8 is not the superset.

Please can u confirm that

1.Weather is it possible to change from WE8ISO8859P1 to UTF8 using
Alter database command. (Server side). We are getting error that
target is not superset.

2. If not then is it necessary to create db in US7ASCII to later chang
to UTF8.
3. If yes then does it have any relation with Windos NT or Unix OS
supported character sets.

Thanks in advance

Shailesh Mishra



Fri, 25 Jun 2004 14:09:59 GMT
 Character set conversion to UTF8 from WE8ISO8859P1 to UTF8

Quote:

>Hi

>We r trying to change database character set from WE8ISO8859P1 to UTF8
>on windows NT having oracle 8.0.4.
>We have tried all ways as mentioned in oracle docs as well as
>Metalink.But getting the error that UTF8 is not the superset.

>Please can u confirm that

>1.Weather is it possible to change from WE8ISO8859P1 to UTF8 using
>Alter database command. (Server side). We are getting error that
>target is not superset.

>2. If not then is it necessary to create db in US7ASCII to later chang
>to UTF8.
>3. If yes then does it have any relation with Windos NT or Unix OS
>supported character sets.

>Thanks in advance

>Shailesh Mishra

If you want to change the charecter set you must bild an new database. There is
no other way to solve this problem.
--
__________________________________________________________
News suchen, lesen, schreiben mit http://newsgroups.web.de


Fri, 25 Jun 2004 18:53:57 GMT
 Character set conversion to UTF8 from WE8ISO8859P1 to UTF8

want to change the charecter set you must bild an new database. There is

Quote:
> no other way to solve this problem.
> --
> __________________________________________________________
> News suchen, lesen, schreiben mit http://newsgroups.web.de

Not ( totally ) true. I have at least four 8i databases which were
previously US7ASCII which I have just changed to UTF8.

However .... there were some characters in the database that were beyond
ASCII ( like an accented E ) that somehow got inserted into the database via
JDBC. So an extra step to ALTER DATABASE blah ... I have to search for
characters that were beyond ASCII and convert them using Java to the proper
UTF8 encoding.

As to the original post's question, the docs only show a sample using
US7ASCII, and it says:

"The target character set is a strict superset if and only if each and every
codepoint in the source character set is available in the target character
set, with the same corresponding codepoint value."

From my experience, latin characters in iso8859-1 would therefore be encoded
differently in UTF-8. For example, an accented E is 1 octet in iso-8859-1
but it would be 2 octets in UTF-8 ... which I think is what Oracle is
complaining about ... not that it was trying to convert characters when you
do ALTER DATABASE. It is probably complaining because it KNOWS that some of
your data will be garbage if you do proceed to convert them to UTF-8.



Sat, 26 Jun 2004 13:57:11 GMT
 
 [ 3 post ] 

 Relevant Pages 

1. character set conversion: AL24UTFFSS to UTF8

2. Characters beyond ASCII in database ( US7ASCII and UTF8 character set )

3. Solution: UCS2 / UTF8 / Oracle / Fail to convert between UTF8 and UCS2

4. Converting from WE8ISO8859P1 to UTF8

5. JA16SJIS stored in WE8ISO8859P1 needed to be converted and stored in UTF8

6. exp data from WE8ISO8859P1 and imp in UTF8

7. database character sets utf8

8. Export & Import with UTF8 Character Set

9. utf8 - ascii conversion

10. Database conversion from US7ASCII to UTF8

11. utf8 <-> sjis conversion problem

12. utf8 ,sqlplus, unix, sqlldr and character translation


 
Powered by phpBB® Forum Software