Home > Articles > Programming > Java

  • Print
  • + Share This
This chapter is from the book

Processing XML with XOM

One of the most important skills you can develop as a Java programmer is the ability to find suitable packages and classes that can be employed in your own projects. For obvious reasons, making use of a well-designed class library is much easier than developing one on your own.

Although Sun’s Java class library contains thousands of well-designed classes that cover a comprehensive range of development needs, the company isn’t the only supplier of packages that may prove useful to your efforts.

Dozens of Java packages are offered by other companies, groups, and individuals under a variety of commercial and open source licenses. Some of the most notable come from Apache Jakarta, a Java development project of the Apache Software Foundation that has produced the web application framework Struts, the Log4J logging class library, and many other popular libraries.

Another terrific open source Java class library is the XOM library, a tree-based package for XML processing that strives to be simple to learn, simple to use, and uncompromising in its adherence to well-formed XML.

The library was developed by the programmer and author Elliotte Rusty Harold based on his experience with Sun’s XML processing packages and other efforts to handle XML in Java.

The project was originally envisioned as a fork of JDOM, a popular tree-based model for representing an XML document. Harold has contributed code to that open source project and participated in its development.

Instead of forking the JDOM code, Harold decided to start from scratch and adopt some of its core design principles in XOM.

The library embodies the following principles:

  • XML documents are modeled as a tree with Java classes representing nodes on the tree such as elements, comments, processing instructions, and document type definitions. A programmer can add and remove nodes to manipulate the document in memory, a simple approach that can be implemented gracefully in Java.
  • All XML data produced by XOM is well-formed and has a well-formed namespace.
  • Each element of an XML document is represented as a class with constructor methods.
  • Object serialization is not supported. Instead, programmers are encouraged to use XML as the format for serialized data, enabling it to be readily exchanged with any software that reads XML regardless of the programming language in which it was developed.
  • The library relies on another XML parser to read XML documents and fill trees instead of doing this low-level work directly. XOM uses a SAX parser that must be downloaded and installed separately. Right now, the preferred parser is Apache Xerces 2.7.1.

XOM is available for download from the web address http://www.cafeconleche.org/XOM. The most current version at this writing is 1.1, which includes Xerces 2.7.1 in its distribution.

After you have downloaded XOM and added its packages to your system’s Classpath, you’re ready to begin using XOM.

The full installation instructions are available from the XOM and Xerces websites. The classes are distributed as JAR archive files—xom-1.1.jar, xercesImpl.jar, and xmlapis.jar. These files should be added to your system’s Classpath environment variable so that your Java programs can use XOM classes.

Creating an XML Document

The first application you will create, RssWriter, creates an XML document that contains the start of an RSS feed. The document is shown in Listing 19.2.

Listing 19.2. The Full Text of feed.rss

1: <?xml version="1.0"?>
2: <rss version="2.0">
3:   <channel>
4:     <title>Workbench</title>
5:     <link>http://www.cadenhead.org/workbench/</link>
6:   </channel>
7: </rss>

The base nu.xom package contains classes for a complete XML document (Document) and the nodes a document can contain (Attribute, Comment, DocType, Element, ProcessingInstruction, and Text).

The RssStarter application uses several of these classes. First, an Element object is created by specifying the element’s name as an argument:

Element rss = new Element("rss");

This statement creates an object for the root element of the document, rss. Element’s one-argument constructor can be used because the document does not employ a feature of XML called namespaces; if it did, a second argument would be necessary: the namespace uniform resource identifier (URI) of the element. The other classes in the XOM library support namespaces in a similar manner.

In the XML document in Listing 19.2, the rss element includes an attribute named version with the value “2.0”. An attribute can be created by specifying its name and value in consecutive arguments:

Attribute version = new Attribute("version", "2.0");

Attributes are added to an element by calling its addAttribute() method with the attribute as the only argument:


The text contained within an element is represented by the Text class, which is constructed by specifying the text as a String argument:

Text titleText = new Text("Workbench");

When composing an XML document, all of its elements end up inside a root element that is used to create a Document object—a Document constructor is called with the root element as an argument. In the RssStarter application, this element is called rss. Any Element object can be the root of a document:

Document doc = new Document(rss);

In XOM’s tree structure, the classes representing an XML document and its constituent parts are organized into a hierarchy below the generic superclass nu.xom.Node. This class has three subclasses in the same package: Attribute, LeafNode, and ParentNode.

To add a child to a parent node, call the parent’s appendChild() method with the node to add as the only argument. The following code creates three elements—a parent called domain and two of its children, name and dns:

Element channel = new Element("channel");
Element link = new Element("link");
Text linkText = new Text("http://www.cadenhead.org/workbench/");

The appendChild() method appends a new child below all other children of that parent. The preceding statements produce this XML fragment:


The appendChild() method also can be called with a String argument instead of a node. A Text object representing the string is created and added to the element:


After a tree has been created and filled with nodes, it can be displayed by calling the Document method toXML(), which returns the complete and well-formed XML document as a String.

Listing 19.3 shows the complete application.

Listing 19.3. The Full text of RssStarter.java

 1: import nu.xom.*;
 3: public class RssStarter {
 4:     public static void main(String[] arguments) {
 5:         // create an <rss> element to serve as the
 6:         Element rss = new Element("rss");
 8:         // add a version attribute to the element
 9:         Attribute version = new Attribute("version", "2.0");
10:         rss.addAttribute(version);
11:         // create a <channel> element and make it a child of <rss>
12:         Element channel = new Element("channel");
13:         rss.appendChild(channel);
14:         // create the channel's
15:         Element title = new Element("title");
16:         Text titleText = new Text("Workbench");
17:         title.appendChild(titleText);
18:         channel.appendChild(title);
19:         // create the channel's <link>
20:         Element link = new Element("link");
21:         Text linkText = new Text("http://www.cadenhead.org/workbench/");
22:         link.appendChild(linkText);
23:         channel.appendChild(link);
25:         // create a new document with <rss> as the root element
26:         Document doc = new Document(rss);
28:         // Display the XML document
29:         System.out.println(doc.toXML());
30:     }
31: }

The RssStarter application displays the XML document it creates on standard output. The following command runs the application and redirects its output to a file called feed.rss:

java RssStarter > feed.rss

XOM automatically precedes a document with an XML declaration.

The XML produced by this application contains no indentation; elements are stacked on the same line.

XOM only preserves significant whitespace when representing XML data—the spaces between elements in the RSS feed contained in Listing 19.2 are strictly for presentation purposes and are not produced automatically when XOM creates an XML document. A subsequent example demonstrates how to control indentation.

Modifying an XML Document

The next project, the DomainEditor application, makes several changes to the XML document that was just produced by the RssStarter application, feed.rss. The text enclosed by the link element is changed to “http://www.cadenhead.org/”, and a new item element is added:

  <title>Fuzzy Zoeller Sues Over Libelous Wikipedia Page</title>

Using the nu.xom package, XML documents can be loaded into a tree from several sources: a File, InputStream, Reader, or a URL (which is specified as a String instead of a java.net.URL object).

The Builder class represents a SAX parser that can load an XML document into a Document object. Constructor methods can be used to specify a particular parser or let XOM use the first available parser from this list: Xerces 2, Crimson, Piccolo, GNU Aelfred, Oracle, XP, Saxon Aelfred, or Dom4J Aelfred. If none of these is found, the parser specified by the system property org.xml.sax.driver is used. Constructors also determine whether the parser is validating or nonvalidating.

The Builder() and Builder(true) constructors both use the default parser—most likely a version of Xerces. The presence of the Boolean argument true in the second constructor configures the parser to be validating. It would be nonvalidating otherwise. A validating parser throws a nu.xom.ValidityException if the XML document doesn’t validate according to the rules of its document type definition.

The Builder object’s build() method loads an XML document from a source and returns a Document object:

Builder builder = new Builder();
File xmlFile = new File("feed.rss");
Document doc = builder.build(xmlFile);

These statements load an XML document from the file feed.rss barring one of two problems: A nu.xom.ParseException is thrown if the file does not contain well-formed XML, and a java.io.IOException is thrown if the input operation fails.

Elements are retrieved from the tree by calling a method of their parent node.

A Document object’s getRootElement() method returns the root element of the document:

Element root = doc.getRootElement();

In the XML document feed.rss, the root element is domains.

Elements with names can be retrieved by calling their parent node’s getFirstChildElement() method with the name as a String argument:

Element channel = root.getFirstChildElement("channel");

This statement retrieves the channel element contained in the rss element (or null if that element could not be found). Like other examples, this is simplified by the lack of a namespace in the document; there are also methods where a name and namespace are arguments.

When several elements within a parent have the same name, the parent node’s getChildElements() method can be used instead:

Elements children = channel.getChildElements()

The getChildElements() method returns an Elements object containing each of the elements. This object is a read-only list and does not change automatically if the parent node’s contents change after getChildElements() is called.

Elements has a size() method containing an integer count of the elements it holds. This can be used in a loop to cycle through each element in turn beginning with the one at position 0. There’s a get() method to retrieve each element; call it with the integer position of the element to be retrieved:

for (int i = 0; i < children.size(); i++) {    Element link = children.get(i);

This for loop cycles through each child element of the channel element.

Elements without names can be retrieved by calling their parent node’s getChild() method with one argument: an integer indicating the element’s position within the parent node:

Text linkText = (Text) link.getChild(0);

This statement creates the Text object for the text “http://www.cadenhead.org/workbench/” found within the link element. Text elements always will be at position 0 within their enclosing parent.

To work with this text as a string, call the Text object’s getValue() method, as in this statement:

if (linkText.getValue().equals("http://www.cadenhead.org/workbench/"))
    // ...

The DomainEditor application only modifies a link element enclosing the text “http://www.cadenhead.org/workbench/”. The application makes the following changes: The text of the link element is deleted, the new text “http://www.cadenhead.org/” is added in its place, and then a new item element is added.

A parent node has two removeChild() methods to delete a child node from the document. Calling the method with an integer deletes the child at that position:

Element channel = domain.getFirstChildElement("channel");
Element link = dns.getFirstChildElement("link");

These statements would delete the Text object contained within the channel’s first link element.

Calling the removeChild() method with a node as an argument deletes that particular node. Extending the previous example, the link element could be deleted with this statement:


Listing 19.4 shows the source code of the DomainEditor application.

Listing 19.4. The Full Text of DomainEditor.java

 1: import java.io.*;
 2: import nu.xom.*;
 4: public class DomainEditor {
 5:     public static void main(String[] arguments) throws IOException {
 6:         try {
 7:             // create a tree from the XML document feed.rss
 8:             Builder builder = new Builder();
 9:             File xmlFile = new File("feed.rss");
10:             Document doc = builder.build(xmlFile);
12:             // get the root element <rss>
13:             Element root = doc.getRootElement();
15:             // get its <channel> element
16:             Element channel = root.getFirstChildElement("channel");
18:             // get its <link> elements
19:             Elements children = channel.getChildElements();
20:             for (int i = 0; i < children.size(); i++) {
22:                 // get a <link> element
23:                 Element link = children.get(i);
25:                 // get its text
26:                 Text linkText = (Text) link.getChild(0);
28:                 // update any link matching a URL
29:                 if (linkText.getValue().equals(
30:                     "http://www.cadenhead.org/workbench/")) {
32:                     // update the link's text
33:                     link.removeChild(0);
34:                     link.appendChild("http://www.cadenhead.org/");
35:                 }
36:             }
38:             // create new elements and attributes to add
39:             Element item = new Element("item");
40:             Element itemTitle = new Element("title");
42:             // add them to the <channel> element
43:             itemTitle.appendChild(
44:                 "Fuzzy Zoeller Sues Over Libelous Wikipedia Page"
45:             );
46:             item.appendChild(itemTitle);
47:             channel.appendChild(item);
49:             // display the XML document
50:             System.out.println(doc.toXML());
51:         } catch (ParsingException pe) {
52:             System.out.println("Error parsing document: " + pe.getMessage());
53:             pe.printStackTrace();
54:             System.exit(-1);
55:         }
56:     }
57: }

The DomainEditor application displays the modified XML document to standard output, so it can be run with the following command to produce a file named feeds2.rss:

java DomainEditor > feed2.rss

Formatting an XML Document

As described earlier, XOM does not retain insignificant whitespace when representing XML documents. This is in keeping with one of XOM’s design goals—to disregard anything that has no syntactic significance in XML. (Another example of this is how text is treated identically whether created using character entities, CDATA sections, or regular characters.)

Today’s next project is the DomainWriter application, which adds a comment to the beginning of the XML document feeds2.rss and serializes it with indented lines, producing the version shown in Listing 19.5.

Listing 19.5. The Full Text of feeds2.rss

 1: <?xml version="1.0"?>
 2: <rss version="2.0">
 3:   <channel>
 4:     <title>Workbench</title>
 5:     <link>http://www.cadenhead.org/</link>
 6:     <item>
 7:       <title>Fuzzy Zoeller Sues Over Libelous Wikipedia Page</title>
 8:     </item>
 9:   </channel>
10: </rss>

The Serializer class in nu.xom offers control over how an XML document is formatted when it is displayed or stored serially. Indentation, character encoding, line breaks, and other formatting are established by objects of this class.

A Serializer object can be created by specifying an output stream and character encoding as arguments to the constructor:

File inFile = new File(arguments[0]);
FileOutputStream fos = new FileOutputStream("new_" +
Serializer output = new Serializer(fos, "ISO-8859-1");

These statements serialize a file using the ISO-8859-1 character encoding. The file is given a name based on a command-line argument.

Serializer supports 20 encodings, including ISO-10646-UCS-2, ISO-8859-1 through ISO-8859-10, ISO-8859-13 through ISO-8859-16, UTF-8, and UTF-16. There’s also a Serializer() constructor that takes only an output stream as an argument; this uses the UTF-8 encoding by default.

Indentation is set by calling the serializer’s setIndentation() method with an integer argument specifying the number of spaces:


An entire XML document is written to the serializer destination by calling the serializer’s write() method with the document as an argument:


The DomainWriter application inserts a comment atop the XML document instead of appending it at the end of a parent node’s children. This requires another method of the parent node, insertChild(), which is called with two arguments—the element to add and the integer position of the insertion:

Builder builder = new Builder();
Document doc = builder.build(arguments[0]);
Comment timestamp = new Comment("File created " +
    new java.util.Date());
doc.insertChild(timestamp, 0);

The comment is placed at position 0 atop the document, moving the domains tag down one line but remaining below the XML declaration.

Listing 19.6 contains the source code of the application.

Listing 19.6. The Full Text of DomainWriter.java

 1: import java.io.*;
 2: import nu.xom.*;
 4: public class DomainWriter {
 5:     public static void main(String[] arguments) throws IOException {
 6:         try {
 7:             // Create a tree from an XML document
 8:             // specified as a command-line argument
 9:             Builder builder = new Builder();
10:             Document doc = builder.build(arguments[0]);
12:             // Create a comment with the current time and date
13:              Comment timestamp = new Comment("File created "
14:                  + new java.util.Date());
16:             // Add the comment above everything else in the
17:             // document
18:             doc.insertChild(timestamp, 0);
20:             // Create a file output stream to a new file
21:             File inFile = new File(arguments[0]);
22:             FileOutputStream fos = new FileOutputStream("new_" +
24:             // Using a serializer with indention set to 2 spaces,
25:             // write the XML document to the file
26:             Serializer output = new Serializer(fos, "ISO-8859-1");
27:             output.setIndent(2);
28:             output.write(doc);
29:         } catch (ParsingException pe) {
30:             System.out.println("Error parsing document: " + pe.getMessage());
31:             pe.printStackTrace();
32:             System.exit(-1);
33:         }
34:     }
35: }

The DomainWriter application takes an XML filename as a command-line argument when run:

java DomainWriter feeds2.rss

This command produces a file called new_feeds2.rss that contains an indented copy of the XML document with a time stamp inserted as a comment.

Evaluating XOM

These three sample applications cover the core features of the main XOM package and are representative of its straightforward approach to XML processing.

There also are smaller nu.xom.canonical, nu.xom.converters, nu.xom.xinclude, and nu.xom.xslt packages to support XInclude, XSLT, canonical XML serialization, and conversions between the XOM model for XML and the one used by DOM and SAX.

Listing 19.7 contains an application that works with XML from a dynamic source: RSS feeds of recently updated web content from the producer of the feed. The RssFilter application searches the feed for specified text in headlines, producing a new XML document that contains only the matching items and shorter indentation. It also modifies the feed’s title and adds an RSS 0.91 document type declaration if one is needed in an RSS 0.91 format feed.

Listing 19.7. The Full Text of RssFilter.java

 1: import nu.xom.*;
 3: public class RssFilter {
 4:     public static void main(String[] arguments) {
 6:         if (arguments.length < 2) {
 7:             System.out.println("Usage: java RssFilter rssFile searchTerm");
 8:             System.exit(-1);
 9:         }
11:         // Save the RSS location and search term
12:         String rssFile = arguments[0];
13:         String searchTerm = arguments[1];
15:         try {
16:             // Fill a tree with an RSS file's XML data
17:             // The file can be local or something on the
18:             // Web accessible via a URL.
19:             Builder bob = new Builder();
20:             Document doc = bob.build(rssFile);
22:             // Get the file's root element (<rss>)
23:             Element rss = doc.getRootElement();
25:             // Get the element's
26:             Attribute rssVersion = rss.getAttribute("version");
27:             String version = rssVersion.getValue();
29:             // Add the DTD for RSS 0.91 feeds, if needed
30:             if ( (version.equals("0.91")) & (doc.getDocType() == null) ) {
31:                 DocType rssDtd = new DocType("rss",
32:                     "http://my.netscape.com/publish/formats/rss-0.91.dtd");
33:                 doc.insertChild(rssDtd, 0);
34:              }
36:             // Get the first (and only) <channel> element
37:             Element channel = rss.getFirstChildElement("channel");
39:             // Get its <title> element
40:             Element title = channel.getFirstChildElement("title");
41:             Text titleText = (Text)title.getChild(0);
43:             // Change the title to reflect the search term
44:             titleText.setValue(titleText.getValue() + ": Search for " +
45:                 searchTerm + " articles");
47:             // Get all of the <item> elements and loop through them
48:             Elements items = channel.getChildElements("item");
49:             for (int i = 0; i < items.size(); i++) {
50:                 // Get an <item> element
51:                 Element item = items.get(i);
53:                 // Look for a <title> element inside it
54:                 Element itemTitle = item.getFirstChildElement("title");
56:                 // If found, look for its contents
57:                 if (itemTitle != null) {
58:                     Text itemTitleText = (Text) itemTitle.getChild(0);
60:                     // If the search text is not found in the item,
61:                     // delete it from the tree
62:                     if (itemTitleText.toString().indexOf(searchTerm) == -1)
63:                         channel.removeChild(item);
64:                 }
65:             } 
67:             // Display the results with a serializer
68:             Serializer output = new Serializer(System.out);
69:             output.setIndent(2);
70:             output.write(doc);
71:         } catch (Exception exc) {
72:             System.out.println("Error: " + exc.getMessage());
73:             exc.printStackTrace();
74:         }
75:     }
76: }

One feed that can be used to test the application is the one from the Toronto Star newspaper. The following command searches it for items with titles that mention the word “snow”:

java RssFilter http://www.thestar.com/rss/000-082-672?searchMode=Lineup snow

Comments in the application’s source code describe its functionality.

XOM’s design is strongly informed by one overriding principle: enforced simplicity.

On the website for the class library, Harold states that XOM “should help inexperienced developers do the right thing and keep them from doing the wrong thing. The learning curve needs to be really shallow, and that includes not relying on best practices that are known in the community but are not obvious at first glance.”

The new class library is useful for Java programmers whose Java programs require a steady diet of XML.

  • + Share This
  • 🔖 Save To Your Account