SAX Parser includes ignorable whitespaces in the character() method

I am writing an app which need to access all text content in XML. According to the ContentHandler API, this could be accomplished by using a validating parser and the characters() method.

But with the Xerces parser, the characters() method could contain ignorable whitespaces (XML formatting whitespaces). I have no way to tell if the whitespace is ignorable whitespace or is part of the XML content.

Has anybody else run into the problem? I tested with both Xerces 2.9.1 and Xerces 2.11. They behave the same way.

Joe Zhu