Using ODBC data source connection special language characters are replaced by a question mark (?).
This document describes how to resolve this.
Special language characters replaced by ?.
When using ANSI ODBC driver manager, the characters received from the driver manager are bounded to the the ANSI client locale, ie OS locale.
For example case, the the character ő ( letter O with double accute accent) is not in CP1252 code page on Windows. The character failed to translate to the codepage 1252.
ODBC Data Source
Resolving the problem
In order to receive the characters in Unicode, customers are advised to use Unicode driver manager. To enable Unicode driver manager, in the connection string for a Netezza datasource, add the keyword @UNICODE as below
(dblogicname "GOSALES" "DSN=DataSource;UID=udmsqc2;PWD=password@UNICODE" NZ )
For the generic ODBC datasource in the portal, there is a check box "Unicode ODBC". So we don't need manually to add the keyword @UNOCODE.