[jdom-interest] Bug or Misunderstanding?

Robert J. Sanford, Jr. rsanford at trefs.com
Wed Jun 12 15:16:56 PDT 2002

I'm using the binary download version of B8 and have run into two minor
issues. I'm not sure if I'm configured incorrectly or if there is a bug
either in JDom or in the underlying parser (I have both the JAXP and Crimson
.jar files on my classpath so I assume that I'm using the JAXP API to the
Crimson parser).

The first problem is that if I turn validation on my test document fails the
validation even though XMLSpy 4.3 passes it. The exception being thrown is:
 org.jdom.JDOMException: Error on line 2 of document
   file:/f:/fantasy football/development/server/xml/fantasyFootball.xml:
   Element type "Project" is not declared.
 Caused by: org.xml.sax.SAXParseException: Element type "Project" is not
   at org.apache.crimson.parser.Parser2.error(Parser2.java:3160)
   at org.apache.crimson.parser.Parser2.maybeElement(Parser2.java:1322)
   at org.apache.crimson.parser.Parser2.parseInternal(Parser2.java:500)
   at org.apache.crimson.parser.Parser2.parse(Parser2.java:305)
   at org.apache.crimson.parser.XMLReaderImpl.parse(XMLReaderImpl.java:442)
   at org.jdom.input.SAXBuilder.build(SAXBuilder.java:354)

The complete XML document looks like:
   <?xml version="1.0" encoding="UTF-8"?>
   <Project xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
      <Object package="player" className="Player">

I know that this is an issue with validation because when I don't use
validation the entire file is parsed, Elements are created by my custom
JDOMFactory, and the XMLOutputter spews forth the file very nicely.

So why is XMLSpy saying the document is valid but JDOM is not? Should I be
using Xerces instead of Crimson or am I doing something completely wrong?

My second problem I think is caused by the validation failure. In my xml
schema I specify an attribute on the Object named "baseClass" which is
optional and has a default value of "com.wobbet.framework.BaseClass".
However, when I am iterating through the DOM tree and get the Object element
and call getAttributeValue("baseClass") I get a null return value. If what I
think is happening is happening then the failure to load/validate the schema
results in the parser not knowing that there is a default value for the
attribute. If that isn't the case then that means there is a bug somewhere
in the parsing, I'm guessing in Crimson since JDOM should just be using the
output from Crimson SAX to build the DOM and hopefully JDOM will recognize
the default values returned from there properly).

So is this one or more bugs, a configuration issue, a mis-conception on my
part or a combination of the three?

Many thanks and God's peace,


More information about the jdom-interest mailing list