Should I be using JDBC getNString() instead of getString()?

Question

We are building a Java application backed by an Oracle database that we access using JDBC (drivers ojdbc6.jar and orai18n.jar). The database schema stores text columns primarily using the NVARCHAR2 data type.

The JDBC documentation for the JDBC ResultSet says that getNString() is particularly intended for use with the NCHAR, NVARCHAR etc. data types, but at the moment we are only using getString().

This seems to work fine, so I am wondering why I should use getNString() rather than getString(). Is getString() going to start failing if non-ASCII characters are entered, or is the Oracle JDBC driver indifferent as to which method I should use?

EDIT: Seems that it may be database-dependent: SQL Server doesn't seem to mind which you use, depending on the connection parameters. Does anyone have any specific information on Oracle?

Answer 1

I have done a test on our application and it seems that getNString() is unnecessary with Java 6, JDBC 6, Oracle JDBC 6 drivers and Oracle 11.1.0.6.0. The test string I used was "Δ, Й, ק, ‎ م, ๗, あ, 叶, 葉, and 말", copied from http://en.wikipedia.org/wiki/Unicode.

Most of our data access is done via stored procedures. Java was able to set and retrieve the above test string correctly via setObject() and getString() (not setString() for abstraction reasons), collecting data from the interface and writing it back to the interface as expected.

Hence getString() works ok for Unicode data with Oracle 11g (like SQL Server as in the above link) so we will continue to use this rather than getNString().