Mishaving character set conversion Java stored proc. vs JDBC application

We have an Oracle 8i database with internal character set UTF-8. Inside a Java stored procedure I can access my XML document through JDBC and I get the document in UTF-8 format without conversion. When I use nearly the same code inside a standalone JDBC client application Oracle tries to convert the document to ISO-8859-1 although the NLS_LANG parameter is set correctly to ".UTF8". When I run sqlplus in the same client environment Oracle make no character set conversion - just with JDBC. I also tried working with the thin and the oci8 JDBC driver - without any success. I also tried different JDK's (1.1.6 -> 1.3.0) - also without any success.

Does anyone know a solution for this problem ?

Kind regards,
Andreas Jung Received on Tue Jul 11 2000 - 06:21:02 CDT