JAXP & DOM with Namespaces

Question

A quick question: when an XML document using a namespace is parsed into a DOM tree, is Element.getNodeName() supposed to return the tag name of the element with or without the namespace prefix? Also, does Element.getElementsByTagName() follow the same standard?
When I use the Oracle XML Parser through JAXP, Element.getNodeName() returns the namespace prefix along with the name, but Element.getElementsByTagName(String name) requires that I leave off the namespace prefix. When I use the default parser included with Java 1.4.1, the behavior is different. Does anyone know what the correct behavior is?

Answer 1

The XML DOM core specification can be found here: http://www.w3.org/TR/2002/WD-DOM-Level-3-Core-20021022/core.htm
From that, it looks like Nodes have separate fields for namespace and name, so I would lean towards there having to be separate accessors as well, so getNodeName() should not return the namespace. That should be handled by getNamespaceURI(). Although, that returns the URI, not the alias, but I don't think an alias is used in any other method anyways.
Also, there is a getElementsByTagNameNS which allows searching for tags within a given namespace, so I would say that getElementsByTagName behaves as it should.