I'm using JAXP 1.1's DocumentBuilder and Xerces 2.2 to load an XML string into a DOM. A simple example string is: <page>&lt;</page>.

When I examine the DOM structure in my debugger, I see an Element node with a single child Text node that contains the resolved entity reference text - in this case the single character '<'.  What I want to see is an Element node with an EntityReference node child and subtree.

If I substitute the Crimson parser for Xerces, I can toggle between these two modes using the docBuilderFactory.setExpandEntityReferences(boolean) method, but this method appears to have no effect on Xerces.

What am I doing wrong?

- Luke -

Reply via email to