Need help to upload Excel Chinese data into Oracle

Question

Hello
I am very new at internationalization. We are writing an application that
reads Chinese data from a Microsoft Excel file and writes it out into Oracle
8i. This is what we have done:
1. Use JDBC-ODBC driver to read the Excel data directly from .xls.
2. Write data into oracle database.
When we open Excel, we can view the Chinese characters. When we use SQL
PLus, we see only ????. When we write another servlet to retrieve the data,
it is displayed as ????. The few questions I have are:
1. How can I verify that the Chinese data has been accurately inserted into
the database?
2. Do I need to set anything special (environment variables, config files
etc) in BEA Weblogic 5.1 and Oracle 8i?
The environment we have are:
1. BEA Weblogic SP9 running on WIn NT.
2. Oracle 8i running on Solaris.
Thanks

Answer 1

Make sure that the characters are correct in the app server after you get
them out of Oracle and before you put them into the web page. This will help
you determine if the problem is in the JDBC access to Oracle or the encoding
into the web page. Make sure that you verify that the hex values of each
character from the db test case are what you expect.
Peace,
Cameron Purdy
Tangosol Inc.
<< Tangosol Server: How Weblogic applications are customized >>
<< Download now from http://www.tangosol.com/download.jsp >>
"Bernard Ong" <[email protected]> wrote in message
news:[email protected]...
Hi Kev
1. "It doesn't work" means that I still see ??? in the database. WHen I
retrieve the data, it is still displayed as ??? in a browser.
2. Codeset to use to display should be UTF-8.
3. Charset in database server is UTF8. I did a "select * from
database_parameters" to get this value.
4. I copied the Chinese characters from Excel and pasted on Notepad. Icould
see the Chinese characters with the Chinese emulator turned on. WHen Icopy
??? from SQL*Plus and paste it into notepad, I see ???.
Presume:
1. Yes, running SQL*PLUS on BEA machine.
2. BEA machine running on WIndows 200 professional. We downloaded aChinese
emulator. WIndows 2000 is the English version.
Cheers
"Kevin Lu" <[email protected]> wrote in message
news:[email protected]...
Hi Bernard,
Please first clarify:-
1) "It doesn't work" means... cannot display the chinese character as
you
desired?
2) Which codeset that you want it to display, Chinese traditional big5?UTF-8,
i would think it's still in Java/JDBC codeset.
3) What's the charset in the database server? Get help from DBA if you
not
sure
how to obtain this information.
4) Try paste the result(select via SQL*Plus) onto Win Notepad, what's
the
outcome?
Presume:
1) you're running SQL*Plus on your BEA machine?
2) BEA machine is running on Chinese O/S or hv NJStar installed andrunning?
Let me know if i could be of further help.
-Kev
"Bernard Ong" <[email protected]> wrote:
Thanks Kev.
We had changed the registry in the BEA machine to
AMERICAN_AMERICA.UTF8.
THis was done for the Oracle registry in the BEA machine.
Unfortunately,
it
still doesn't work.
Cheers,
Bernard
"Kevin Lu" <[email protected]> wrote in message
news:[email protected]...
Hi Bernard,
I take about the same approach as you while resolving codeset
problem.
So,
first
i make sure SQL*Plus could display my desired result before proceedto
application(be
it Swing, servlet, JSP console etc..). You may try set the NLS_LANGon
your client
machine(same machine as where you run SQL*Plus) to the SAME as
database
server
character set, for instance America_American.US7ASCII. You should
able
to
see
the Chinese characters.
Hope this help with your first question. I can't help for the restanyway.
good luck!
-Kev
"Bernard Ong" <[email protected]> wrote:
Hello
I am very new at internationalization. We are writing an
application
that
reads Chinese data from a Microsoft Excel file and writes it outinto
Oracle
8i. This is what we have done:
1. Use JDBC-ODBC driver to read the Excel data directly from .xls.
2. Write data into oracle database.
When we open Excel, we can view the Chinese characters. When we useSQL
PLus, we see only ????. When we write another servlet to retrievethe
data,
it is displayed as ????. The few questions I have are:
1. How can I verify that the Chinese data has been accurately
inserted
into
the database?
2. Do I need to set anything special (environment variables, configfiles
etc) in BEA Weblogic 5.1 and Oracle 8i?
The environment we have are:
1. BEA Weblogic SP9 running on WIn NT.
2. Oracle 8i running on Solaris.
Thanks