The Document Object Model (DOM) is a tree-structured representation of the information in an XML document. It consists of several interfaces, the most important of which is the node . All are in the package org.w3c.dom , reflecting the influence of the World Wide Web Consortium (http://www.w3.org) in creating and promulgating the DOM. The DOM interfaces are shown in Table 21-1.
Table 21-1. DOM interfaces
|
Interface |
Function |
|---|---|
|
Document |
Top-level representation of an XML document |
|
Node |
Representation of any node in the XML tree |
|
Element |
An XML element |
|
Text |
A textual string |
You don’t have to implement these interfaces; the parser generates them. When you get to creating or modifying XML documents in Section 21.6, then you can create nodes. But even then there are implementing classes. Parsing an XML document with DOM is syntactically similar to processing a file with XSL, that is, you get a reference to a parser and call its methods with objects representing the input files. The difference is that the parser returns an XML DOM, a tree of objects in memory. Example 21-5 is code that simply parses an XML document.
Example 21-5. XParse.java
import java.io.*;
import org.w3c.dom.*;
import com.sun.xml.tree.*;
import org.xml.sax.SAXException;
import org.xml.sax.SAXParseException;
/** Parse an XML file using DOM.
*/
public class XParse {
/** Convert the file */
public static void parse(String fileName) {
try {
System.err.println("Parsing " + fileName + "...");
// Make the document a URL so relative DTD works.
String uri = "file:" + new File(fileName).getAbsolutePath( );
XmlDocument doc = XmlDocument.createXmlDocument(uri);
System.out.println("Parsed OK");
} catch (SAXParseException ex) {
System.err.println("+================================+");
System.err.println("| *Parse Error* |");
System.err.println("+================================+");
System.err.println("+ Line " + ex.getLineNumber ( )
+ ", uri " + ex.getSystemId ( ));
System.err.println(ex.getClass( ));
System.err.println(ex.getMessage( ));
System.err.println("+================================+");
} catch(SAXException ex) {
System.err.println("+================================+");
System.err.println("| *SAX XML Error* |");
System.err.println("+================================+");
System.err.println(ex.toString( ));
} catch (IOException ex) {
System.err.println("+================================+");
System.err.println("| *Input/Output Error* |");
System.err.println("+================================+");
System.err.println(ex.toString( ));
}
}
public static void main(String[] av) {
if (av.length == 0) {
System.err.println("Usage: XParse file");
return;
}
for (int i=0; i<av.length; i++) {
parse(av[i]);
}
}
}You then traverse the document. You can use the defined
TreeWalker
interface, or you can just use the
algorithm shown in Example 21-6.
Example 21-6. XTW.java (partial listing)
/* Process all the nodes, recursively. */
protected void doRecursive(Node p) {
if (p == null) {
return;
}
NodeList nodes = p.getChildNodes( );
int numElem = nodes.getLength( );
Debug.println("xml-tree", "Element has " + numElem + " children");
for (int i=0; i<numElem; i++) {
Node n = nodes.item(i);
if (n == null) {
continue;
}
doNode(n);
}
}A full code example using this approach is given in Section 21.7.