[rules-users] question about encoding in rules

Greetings, I am having some issues with encoding in my rules. I am using a drl like this: package pt.xpto.mydroolstest import pt.xpto.model.SimpleFact import pt.xpto.model.SimpleResult import java.util.ArrayList import java.util.Collection global Collection results rule 'first-rule' when SimpleFact(description=="someDescription", value < "500") and SimpleFact(description=="anotherDescription", value=="1"); then SimpleResult result = new SimpleResult(); ArrayList aList= new ArrayList(); result.setDescription("Some String with portuguese chars like ç and à"); aList.add("Another String with portuguese chars like é and ã."); result.setValue(aList); results.add(result); end My problem is that when i retrieve the results from firing the rules, the encoding of the strings with the portuguese chars is all messed up. From the tests i did, this happens when the rules are compiled by drools. If i make the following modification to the rules result.setDescription(new String("Some String with portuguese chars like ç and à".getBytes(),"utf-8")); and load it like this, then the strings are retrieved ok. RuleBase ruleBase = RuleBaseLoader.getInstance().loadFromReader(new InputStreamReader(DroolsTest.class.getResourceAsStream( "Rules.drl" ))); Yet the same modification does not work if the rule is defined in the BRMS and loaded by the rules agent. I noticed that in the portuguese example shown in http://labs.jboss.com/drools/ no portuguese special chars are used, is this a limitation of the rules engine or am I just using wrong? Can you point me in the right direction? best regards, João Mota