DSXchange

I am trying to transform and convert a set of data from MS Access to Oracle9i.

The data contains chinese characters. Seen from one of the posting here that I should use set NLS map to BIG5 or GB2312, tried both but did not work for me. I was not able to view the source data. All the chinese characters came out as "??" when I tried to view data from the ODBC stage connecting to the Access table.

However, I was able to view the chinese characters(Simplified Chinese) from MS Access program correctly.

Can anyone advise? Any help rendered is much appreciated. [:)]

1. What is your code page setting for your Oracle?
2. Have you tried to extract the chinese chars from access then write them to a sequential file? Can you read the chinese character this way?

Regards,
Dick Fong

When I do a direct extraction from Access to a sequential file, the chinese characters appears as "??" in the text file as well. Any further advise?

As for question 1, what should be the correct code page setting? Because my DBA is not around today and I am only familiar with basics of DataStage, I am not sure where to get this information on settings.

There are other, non-DataStage factors that can affect successful acquisition of character sets. For example, check that your code page setting is correct, your LANG environment variable is correct, and any database-specific settings are correct.
Within DataStage, the maps serve to convert between the external character set encoding and the Unicode encoding used internally by DataStage. However, a different map may be required for communication between your DataStage server and client (the map name ends in "-CS" for client/server). This may affect how successfully you can view the data using DataStage's data browser (the View Data button).
Basically, you need to use the same map (code page) as you are using for Access, and the "-CS" equivalent for your DataStage clients.

Ray Wurlod
Education and Consulting Services
ABN 57 092 448 518