We have trouble with the Euro character when saving and locating it from the Oracle 10g while using Oracle 10.2..3 JDBC driver. The issue only happens throughout a JUnit test running under Linux. The Euro figures came back from database after saving are total messed up. Oracle continues to be set up to make use of character set "WE8MSWIN1252". Can it be that Linux cannot use this character set?
The correct answer is possible code page 1252 is Microsoft's own extension of ISO-8859-1 (a.k.a. Latin-1), and it is rarely utilized on Linux, which has a tendency to make use of the latter. The euro symbol isn't a part of Latin-1, that might be why it cannot be shown correctly on Linux. Are you able to not use UTF-8? It'll work exactly the same way on sides.
It isn't a Linux factor. It is a known Oracle bug within the retrieve code from the jdbc driver. There is a patch available but you will need use of Oracle Metalink to download it. Another alternative recommended there's to make use of the OCI driver rather than the thin one. That might be a choice for you personally.
This bug, which been around within the 10.2..3 driver, is bound within the 10.2..4 jdbc driver.