Unicode is supported by the Pentaho BI Platform. Follow these guidelines for successful translations.
These files will be read as ISO-8859-1 encoded files. To store Unicode characters in these files they must be encoded using Java, \uNNNN, or XML, NNNN;, encoding and stored as ASCII files. You can use a text editor such as SC Unipad (http://www.unipad.org) to edit the Pentaho .properties files and save them in the correct format.
These files must be stored with all Unicode characters encoded using XML, NNNN;, encoding.
Unicode characters stored in the XSLs must be stored using XML, NNNN;, encoding. Unicode characters in messages stored in org/Pentaho/ui/messages.properties that are used in XSLs should be accessed using msg:getXslString and output escaping should be disabled.
Unicode characters in report definitions for JFreeReport should be stored using XML, NNNN;, encoding. In addition the report configuration section should specify that Identity-H encoding and embedded fonts should be used for PDF output.
You need to use fonts that support Unicode characters such as 'Ariel Unicode MS'