I use Zend Lucene to index swedish texts. The problem is that lucene tokenizes words at swedish chars åäö. For example the word "världens" becomes two words "v" and "ldens" in the index.
Is there a way to add characters that zend lucene should accept and not tokenize at?