How do I get subset of a Java XML org.w3c.dom.Document ?

Question

I have a XML org.w3c.dom.Document object.

It looks sorta like this:

<A>
  <B>
     <C/>
     <D/>
     <E/>
     <F/>
   </B>
   <G>
     <H/>
     <H/>
     <J/>
   </G>
 </A>

How can I convert the Document object so that it strips off the root node and returns another Document object subset (selected by name) that looks like this:

<G>
   <H/>
   <H/>
   <J/>
</G>

I am hoping for something like this:

...
Document doc = db.parse(file);
Document subdoc = doc.getDocumentSubsetByName("G"); //imaginary method name
NodeList nodeList = subdoc.getElementsByTagName("H");

But I am having trouble finding such a thing.

---

The answer turns out to be something like this:

...
Document doc = db.parse();
doc.getDocumentElement.normalize();
NodeList a = doc.getElementsByTagName("A");
Element AsubNode = null;
if (a.item(0) != null) {
   AsubNode = (Element) a.item(0);
   nodeList = AsubNode.getElementsByTagName("G");
...

Answer 1

You can simply use getElementsByTagName("G") to get the G elements, then pick one of them and call getElementsByTagName("H") on that.

Answer 2

Of course, you could always use XPath to do the same thing:

import javax.xml.xpath.XPath;
import javax.xml.xpath.XPathConstants;
import javax.xml.xpath.XPathFactory;
import org.w3c.dom.NodeList;

final XPath xpath = XPathFactory.newInstance().newXPath();
final NodeList list = (NodeList) xpath.evaluate("/A/G/H", 
    doc.getDocumentElement(), XPathConstants.NODESET);

This begins to pay off when the path to your elements begins to become more complex (requiring attribute predicates, etc..)