Oracle JDBC unicode parameter corrupted (DB insert proc)

[dzenanz]

Calling a stored procedure from Java with ojdbc 1.4 corrupts (converts to ¿) unicode characters that are not contained in DB's current charset.
Calling this proc from C# works as expected (all unicode characters are properly stored).

String driver = "oracle.jdbc.OracleDriver";
Class.forName(driver);
DriverManager.registerDriver(new oracle.jdbc.OracleDriver());
conn = DriverManager.getConnection(url, username, password);
CallableStatement storedProc = conn.prepareCall("{call SMSC.p_smsu_campaign_insert(?,?,?,?,?,?)}");
 
String campaign="_šđčćž ò#±¦ţ!ę";
storedProc.setString("Campaign_name", campaign); //<----problem here!
//Campaign_name parameter in DB is nchar type (oracle's unicode type)
/*set other parameters*/
storedProc.execute();

Do you have any idea what is the cause?
Note: this is a simplification of my previous post

[dzenanz]

If anyone comes by the same problem again, here is the solution:

storedProc.setString("Campaign_name", campaign);
((OraclePreparedStatement) storedProc).setFormOfUse(1, OraclePreparedStatement.FORM_NCHAR); //<---this makes the difference!

[JosAH]

If anyone comes by the same problem again, here is the solution:

storedProc.setString("Campaign_name", campaign);
((OraclePreparedStatement) storedProc).setFormOfUse(1, OraclePreparedStatement.FORM_NCHAR); //<---this makes the difference!

Not that I know another solution to your problem but I find the solution you
showed here very ugly: it ties your code to the Oracle database forever. Better
isolate that trickery-dickery to a small utility class that can be changed when
you have to migrate to another database.

kind regards,

Jos