Ch23 xml processing_with_java

XML Processing
Chapter with Java

Topics in This Chapter

• Representing an entire XML document using the
Document Object Model (DOM) Level 2
• Using DOM to display the outline of an XML document in
a JTree
• Responding to individual XML parsing events with the
Simple API for XML Parsing (SAX) 2.0
• Printing the outline of an XML document using SAX
• Counting book orders using SAX
• Transforming XML using XSLT
• Invoking XSLT through custom JSP tags
• Hiding vendor-specific details with the Java API for XML
Processing (JAXP)

Chapter

23
X
ML is a “meta” markup language used to describe the structure of data.
XML has taken the computer industry by storm since its inception and is
now the markup language of choice for configuration files, data interchange,
B2B transactions, and Java 2 Enterprise architectures. XML is even being used to
represent calls to distributed objects through the Simple Object Access Protocol
(SOAP), an XML application.
XML has numerous advantages including being easy to read, easy to parse, exten-
sible, and widely adopted. In addition, you can define a grammar through a
Document Type Definition (DTD) to enforce application-specific syntax. However,
the greatest single advantage of XML is that the data can be easily processed by other
applications; XML data is not in a proprietary format. In essence, XML has done for
data what the Java language has done for programs:

Java = Portable Programs
XML = Portable Data
This chapter doesn’t focus on how to write XML but rather how to process XML
documents with Java. We show you how to use Java to process XML documents by
using the Document Object Model (DOM), the Simple API for XML (SAX), and the
Extensible Style sheet Language for Transformations (XSLT). If you are new to
XML, here are some good starting points for additional information:

1133

1134 Chapter 23 XML Processing with Java

XML 1.0 Specification
http://www.w3.org/TR/REC-xml

Sun Page on XML and Java
http://java.sun.com/xml/

WWW Consortium’s Home Page on XML
http://www.w3.org/XML/

Apache XML Project
http://xml.apache.org/

XML Resource Collection
http://xml.coverpages.org/

O’Reilly XML Resource Center
http://www.xml.com/

23.1 Parsing XML Documents with
DOM Level 2
The Document Object Model (DOM) represents an entire XML document in a
tree-like data structure that can be easily manipulated by a Java program. The advan-
tages of DOM are that it is relatively simple to use and you can modify the data struc-
ture in addition to extracting data from it. However, the disadvantage is that DOM
parses and stores the entire document, even if you only care about part of it. Section
23.3 (Parsing XML Documents with SAX 2.0) discusses an alternative approach
appropriate for cases when you are dealing with very large XML documents but care
about only small sections of them.

Installation and Setup
DOM is not a standard part of either Java 2 Standard Edition or the servlet and JSP
APIs. So, your first step is to download the appropriate classes and configure them
for use in your programs. Here is a summary of what is required.

1. Download a DOM-compliant parser. The parser provides the
Java classes that follow the DOM Level 2 API as specified by
the WWW Consortium. You can obtain a list of XML parsers in
Java at http://www.xml.com/pub/rg/Java_Parsers.

23.1 Parsing XML Documents with DOM Level 2 1135

We use the Apache Xerces-J parser in this book. See
http://xml.apache.org/xerces-j/. This parser also
comes with the complete DOM API in Javadoc format.
2. Download the Java API for XML Processing (JAXP). This API
provides a small layer on top of DOM that lets you plug in different
vendor’s parsers without making any changes to your basic code. See
http://java.sun.com/xml/.
3. Set your CLASSPATH to include the DOM classes. In the case of
Apache Xerces, you need to include xerces_install_dir
xerces.jar. For example, for desktop applications on Windows you
would do
set CLASSPATH=xerces_install_dirxerces.jar;%CLASSPATH%
If you wanted to use DOM from servlets and JSP, you would copy the
appropriate JAR file to the server’s lib directory (if supported),
unpack the JAR file (using jar -xvf) into the server’s classes
directory, or explicitly change the server’s CLASSPATH, usually by
modifying the server start-up script.
4. Set your CLASSPATH to include the JAXP classes. These classes
are in jaxp_install_dir/jaxp.jar. For example, on Unix/Linux and
the C shell, you would do
setenv CLASSPATH jaxp_install_dir/jaxp.jar:$CLASSPATH
For use from servlets and JSP, see the preceding step.
5. Bookmark the DOM Level 2 and JAXP APIs. The official
DOM specification can be found at http://www.w3.org/TR/
DOM-Level-2-Core/, but the API in Javadoc format that comes
with Apache Xerces is easier to read and also includes the JAXP and
SAX (see Section 23.3) APIs.
6. Print the JAXP specification for your reference. Download it
from http://java.sun.com/xml/jaxp-1_1-spec.pdf.

Parsing
With DOM processing, there are two high-level tasks: turning an XML document
into a DOM data structure and looking through that data structure for the data that
interests you. The following list summarizes the detailed steps needed to accomplish
these tasks.

1. Tell the system which parser you want to use. This can be done in
a number of ways: through the javax.xml.parsers.Document-
BuilderFactory system property, through jre_dir/lib/
jaxp.properties, through the J2EE Services API and the class spec-
ified in META-INF/services/javax.xml.parsers.Document-


BuilderFactory, or with a system-dependent default parser. The
system property is the easiest method. For example, the following
code permits users to specify the parser on the command line with the
-D option to java, and uses the Apache Xerces parser otherwise.
public static void main(String[] args) {
String jaxpPropertyName =
"javax.xml.parsers.DocumentBuilderFactory";
if (System.getProperty(jaxpPropertyName) == null) {
String apacheXercesPropertyValue =
"org.apache.xerces.jaxp.DocumentBuilderFactoryImpl";
System.setProperty(jaxpPropertyName,
apacheXercesPropertyValue);
}
...
}
2. Create a JAXP document builder. This is basically a wrapper
around a specific XML parser.
DocumentBuilderFactory builderFactory =
DocumentBuilderFactory.newInstance();
DocumentBuilder builder =
builderFactory.newDocumentBuilder();

Note that you can use the setNamespaceAware and
setValidating methods on the DocumentBuilderFactory to
make the parser namespace aware and validating, respectively.
3. Invoke the parser to create a Document representing an XML
document. You invoke the parser by calling the parse method of the
document builder, supplying an input stream, URI (represented as a
string), or org.xml.sax.InputSource. The Document class rep-
resents the parsed result in a tree structure.
Document document = builder.parse(someInputStream);
4. Normalize the tree. This means to combine textual nodes that were
on multiple lines and to eliminate empty textual nodes.
document.getDocumentElement().normalize();
5. Obtain the root node of the tree. This returns an Element, which
is a subclass of the more general Node class that represents an XML
element.
Element rootElement = document.getDocumentElement();
6. Examine various properties of the node. These properties include
the name of the element (getNodeName), the node type (getNode-
Type; compare the return value to predefined constants in the Node
class), the node value (getNodeValue; e.g., for text nodes the value
is the string between the element’s start and end tags), the attributes
used by the element’s start tag (getAttributes), and the child

23.2 DOM Example: Representing an XML Document as a JTree 1137

nodes (getChildNodes; i.e., the elements contained between the
current element’s start and end tags). You can recursively examine
each of the child nodes.
7. Modify properties of the nodes. Instead of just extracting data
from an XML document, you can modify the document by adding
child nodes (appendChild), removing child nodes (removeChild),
and changing the node’s value (setNodeValue). Unfortunately, how-
ever, DOM doesn’t provide a standard method of writing out a DOM
structure in textual format. So, you have to either do it yourself (print-
ing out a “<”, the node name, the attribute names and values with
equal signs between them and quotes around the values, a “>”, etc.) or
use one of the many existing packages that generate text from a DOM
element.

23.2 DOM Example: Representing
an XML Document as a JTree
Listing 23.1 shows a class that represents the basic structure of an XML document as
a JTree. Each element is represented as a node in the tree, with tree node being
either the element name or the element name followed by a list of the attributes in
parentheses. This class performs the following steps:

1. Parses and normalizes an XML document, then obtains the root ele-
ment. These steps are performed exactly as described in steps one
through five of the previous section.
2. Makes the root element into a JTree node. If the XML element has
attributes (as given by node.getAttributes), the tree node is rep-
resented by a string composed of the element name (getNodeName)
followed by the attributes and values in parentheses. If there are no
attributes (getLength applied to the result of node.get-
Attributes returns 0), then just the element name is used for the
tree node label.
3. Looks up the child elements of the current element using getChild-
Nodes, turns them into JTree nodes, and links those JTree nodes to
the parent tree node.
4. Recursively applies step 3 to each of the child elements.

Listing 23.2 shows a class that creates the JTree just described and places it into
a JFrame. Both the parser and the XML document can be specified by the user. The
parser is specified when the user invokes the program with
java -Djavax.xml.parsers.DocumentBuilderFactory=xxx XMLFrame


If no parser is specified, the Apache Xerces parser is used. The XML document
can be supplied on the command line, but if it is not given, a JFileChooser is used
to interactively select the file of interest. The file extensions shown by the
JFileChooser are limited to xml and tld (JSP tag library descriptors) through use
of the ExtensionFileFilter class of Listing 23.3.
Figure 23–1 shows the initial file chooser used to select the perennials.xml
file (Listing 23.4; see Listing 23.5 for the DTD). Figures 23–2 and 23–3 show the
result in its unexpanded and partially expanded forms, respectively.
Note that because the XML file specifies a DTD, Xerces-J will attempt to parse
the DTD even though no validation of the document is performed. If
perennials.dtd is not available on-line, then you can place the DTD in a dtds
subdirectory (below the directory containing XMLTree) and change the DOCTYPE in
perennials.xml to
<!DOCTYPE perennials SYSTEM "dtds/perennials.dtd">

Figure 23–1 JFileChooser that uses ExtensionFileFilter (Listing 23.3) to
interactively select an XML file.

Figure 23–2 JTree representation of root node and top-level child elements of
perennials.xml (Listing 23.4).


Figure 23–3 JTree representation of perennials.xml with several nodes
expanded.

Listing 23.1 XMLTree.java
import java.awt.*;
import javax.swing.*;
import javax.swing.tree.*;
import java.io.*;
import org.w3c.dom.*;
import javax.xml.parsers.*;

/** Given a filename or a name and an input stream,
* this class generates a JTree representing the
* XML structure contained in the file or stream.
* Parses with DOM then copies the tree structure
* (minus text and comment nodes).
*/

public class XMLTree extends JTree {
public XMLTree(String filename) throws IOException {
this(filename, new FileInputStream(new File(filename)));
}

public XMLTree(String filename, InputStream in) {
super(makeRootNode(in));
}

(continued)


Listing 23.1 XMLTree.java (continued)
// This method needs to be static so that it can be called
// from the call to the parent constructor (super), which
// occurs before the object is really built.

private static DefaultMutableTreeNode
makeRootNode(InputStream in) {
try {
// Use JAXP's DocumentBuilderFactory so that there
// is no code here that is dependent on a particular
// DOM parser. Use the system property
// javax.xml.parsers.DocumentBuilderFactory (set either
// from Java code or by using the -D option to "java").
// or jre_dir/lib/jaxp.properties to specify this.
DocumentBuilderFactory builderFactory =
DocumentBuilderFactory.newInstance();
DocumentBuilder builder =
builderFactory.newDocumentBuilder();
// Standard DOM code from hereon. The "parse"
// method invokes the parser and returns a fully parsed
// Document object. We'll then recursively descend the
// tree and copy non-text nodes into JTree nodes.
Document document = builder.parse(in);
document.getDocumentElement().normalize();
Element rootElement = document.getDocumentElement();
DefaultMutableTreeNode rootTreeNode =
buildTree(rootElement);
return(rootTreeNode);
} catch(Exception e) {
String errorMessage =
"Error making root node: " + e;
System.err.println(errorMessage);
e.printStackTrace();
return(new DefaultMutableTreeNode(errorMessage));
}
}

private static DefaultMutableTreeNode
buildTree(Element rootElement) {
// Make a JTree node for the root, then make JTree
// nodes for each child and add them to the root node.
// The addChildren method is recursive.
DefaultMutableTreeNode rootTreeNode =
new DefaultMutableTreeNode(treeNodeLabel(rootElement));
addChildren(rootTreeNode, rootElement);
return(rootTreeNode);
}

(continued)


private static void addChildren
(DefaultMutableTreeNode parentTreeNode,
Node parentXMLElement) {
// Recursive method that finds all the child elements
// and adds them to the parent node. We have two types
// of nodes here: the ones corresponding to the actual
// XML structure and the entries of the graphical JTree.
// The convention is that nodes corresponding to the
// graphical JTree will have the word "tree" in the
// variable name. Thus, "childElement" is the child XML
// element whereas "childTreeNode" is the JTree element.
// This method just copies the non-text and non-comment
// nodes from the XML structure to the JTree structure.

NodeList childElements =
parentXMLElement.getChildNodes();
for(int i=0; i<childElements.getLength(); i++) {
Node childElement = childElements.item(i);
if (!(childElement instanceof Text ||
childElement instanceof Comment)) {
DefaultMutableTreeNode childTreeNode =
new DefaultMutableTreeNode
(treeNodeLabel(childElement));
parentTreeNode.add(childTreeNode);
addChildren(childTreeNode, childElement);
}
}
}

// If the XML element has no attributes, the JTree node
// will just have the name of the XML element. If the
// XML element has attributes, the names and values of the
// attributes will be listed in parens after the XML
// element name. For example:
// XML Element: <blah>
// JTree Node: blah
// XML Element: <blah foo="bar" baz="quux">
// JTree Node: blah (foo=bar, baz=quux)

private static String treeNodeLabel(Node childElement) {
NamedNodeMap elementAttributes =
childElement.getAttributes();
String treeNodeLabel = childElement.getNodeName();

(continued)


if (elementAttributes != null &&
elementAttributes.getLength() > 0) {
treeNodeLabel = treeNodeLabel + " (";
int numAttributes = elementAttributes.getLength();
for(int i=0; i<numAttributes; i++) {
Node attribute = elementAttributes.item(i);
if (i > 0) {
treeNodeLabel = treeNodeLabel + ", ";
}
treeNodeLabel =
treeNodeLabel + attribute.getNodeName() +
"=" + attribute.getNodeValue();
}
treeNodeLabel = treeNodeLabel + ")";
}
return(treeNodeLabel);
}
}

Listing 23.2 XMLFrame.java
import java.awt.*;
import java.io.*;

/** Invokes an XML parser on an XML document and displays
* the document in a JTree. Both the parser and the
* document can be specified by the user. The parser
* is specified by invoking the program with
* java -Djavax.xml.parsers.DocumentBuilderFactory=xxx XMLFrame
* If no parser is specified, the Apache Xerces parser is used.
* The XML document can be supplied on the command
* line, but if it is not given, a JFileChooser is used
* to interactively select the file of interest.
*/

public class XMLFrame extends JFrame {
"javax.xml.parsers.DocumentBuilderFactory";

(continued)


Listing 23.2 XMLFrame.java (continued)
// Pass the parser factory in on the command line with
// -D to override the use of the Apache parser.
"org.apache.xerces.jaxp.DocumentBuilderFactoryImpl";
}
String filename;
if (args.length > 0) {
filename = args[0];
} else {
String[] extensions = { "xml", "tld" };
WindowUtilities.setNativeLookAndFeel();
filename = ExtensionFileFilter.getFileName(".",
"XML Files",
extensions);
if (filename == null) {
filename = "test.xml";
}
}
new XMLFrame(filename);
}

public XMLFrame(String filename) {
try {
JTree tree = new XMLTree(filename);
JFrame frame = new JFrame(filename);
frame.addWindowListener(new ExitListener());
Container content = frame.getContentPane();
content.add(new JScrollPane(tree));
frame.pack();
frame.setVisible(true);
} catch(IOException ioe) {
System.out.println("Error creating tree: " + ioe);
}
}
}


Listing 23.3 ExtensionFileFilter.java
import java.io.File;
import java.util.*;
import javax.swing.filechooser.FileFilter;

/** A FileFilter that lets you specify which file extensions
* will be displayed. Also includes a static getFileName
* method that users can call to pop up a JFileChooser for
* a set of file extensions.
* <P>
* Adapted from Sun SwingSet demo.
*/

public class ExtensionFileFilter extends FileFilter {
public static final int LOAD = 0;
public static final int SAVE = 1;
private String description;
private boolean allowDirectories;
private Hashtable extensionsTable = new Hashtable();
private boolean allowAll = false;

public ExtensionFileFilter(boolean allowDirectories) {
this.allowDirectories = allowDirectories;
}

public ExtensionFileFilter() {
this(true);
}

public static String getFileName(String initialDirectory,
String description,
String extension) {
String[] extensions = new String[]{ extension };
return(getFileName(initialDirectory, description,
extensions, LOAD));
}

String description,
String extension,
int mode) {
String[] extensions = new String[]{ extension };
extensions, mode));
}

(continued)


Listing 23.3 ExtensionFileFilter.java (continued)
String description,
String[] extensions) {
extensions, LOAD));
}

/** Pops up a JFileChooser that lists files with the
* specified extensions. If the mode is SAVE, then the
* dialog will have a Save button; otherwise, the dialog
* will have an Open button. Returns a String corresponding
* to the file's pathname, or null if Cancel was selected.
*/

String description,
String[] extensions,
int mode) {
ExtensionFileFilter filter = new ExtensionFileFilter();
filter.setDescription(description);
for(int i=0; i<extensions.length; i++) {
String extension = extensions[i];
filter.addExtension(extension, true);
}
JFileChooser chooser =
new JFileChooser(initialDirectory);
chooser.setFileFilter(filter);
int selectVal = (mode==SAVE) ? chooser.showSaveDialog(null)
: chooser.showOpenDialog(null);
if (selectVal == JFileChooser.APPROVE_OPTION) {
String path = chooser.getSelectedFile().getAbsolutePath();
return(path);
} else {
JOptionPane.showMessageDialog(null, "No file selected.");
return(null);
}
}

public void addExtension(String extension,
boolean caseInsensitive) {
if (caseInsensitive) {
extension = extension.toLowerCase();
}

(continued)


Listing 23.3 ExtensionFileFilter.java (continued)
if (!extensionsTable.containsKey(extension)) {
extensionsTable.put(extension,
new Boolean(caseInsensitive));
if (extension.equals("*") ||
extension.equals("*.*") ||
extension.equals(".*")) {
allowAll = true;
}
}
}

public boolean accept(File file) {
if (file.isDirectory()) {
return(allowDirectories);
}
if (allowAll) {
return(true);
}
String name = file.getName();
int dotIndex = name.lastIndexOf('.');
if ((dotIndex == -1) || (dotIndex == name.length() - 1)) {
return(false);
}
String extension = name.substring(dotIndex + 1);
if (extensionsTable.containsKey(extension)) {
return(true);
}
Enumeration keys = extensionsTable.keys();
while(keys.hasMoreElements()) {
String possibleExtension = (String)keys.nextElement();
Boolean caseFlag =
(Boolean)extensionsTable.get(possibleExtension);
if ((caseFlag != null) &&
(caseFlag.equals(Boolean.FALSE)) &&
(possibleExtension.equalsIgnoreCase(extension))) {
return(true);
}
}
return(false);
}

public void setDescription(String description) {
this.description = description;
}
public String getDescription() {
return(description);
}
}


Listing 23.4 perennials.xml
<?xml version="1.0" ?>
<!DOCTYPE perennials SYSTEM
"http://archive.corewebprogramming.com/dtds/perennials.dtd">
<perennials>
<daylily status="in-stock">
<cultivar>Luxury Lace</cultivar>
<award>
<name>Stout Medal</name>
<year>1965</year>
</award>
<award>
<name note="small-flowered">Annie T. Giles</name>
<year>1965</year>
</award>
<award>
<name>Lenington All-American</name>
<year>1970</year>
</award>
<bloom code="M">Midseason</bloom>
<cost discount="3" currency="US">11.75</cost>
</daylily>
<cultivar>Green Flutter</cultivar>
<award>
<year>1976</year>
</award>
<award>
<name note="small-flowered">Annie T. Giles</name>
<year>1970</year>
</award>
<bloom code="M">Midseason</bloom>
<cost discount="3+" currency="US">7.50</cost>
</daylily>
<daylily status="sold-out">
<cultivar>My Belle</cultivar>
<award>
<year>1984</year>
</award>
<bloom code="E">Early</bloom>
<cost currency="US">12.00</cost>
</daylily>

(continued)


Listing 23.4 perennials.xml (continued)
<cultivar>Stella De Oro</cultivar>
<award>
<year>1985</year>
</award>
<award>
<name note="miniature">Donn Fishcer Memorial Cup</name>
<year>1979</year>
</award>
<bloom code="E-L">Early to Late</bloom>
<cost discount="10+" currency="US">5.00</cost>
</daylily>
<daylily status="limited">
<cultivar>Brocaded Gown</cultivar>
<award>
<year>1989</year>
</award>
<bloom code="E">Early</bloom>
<cost currency="US" discount="3+">14.50</cost>
</daylily>
</perennials>

Listing 23.5 perennials.dtd
<?xml version="1.0" encoding="ISO-8859-1" ?>

<!ELEMENT perennials (daylily)*>

<!ELEMENT daylily (cultivar, award*, bloom, cost)+>
<!ATTLIST daylily
status (in-stock | limited | sold-out) #REQUIRED>

<!ELEMENT cultivar (#PCDATA)>

<!ELEMENT award (name, year)>

<!ELEMENT name (#PCDATA)>
<!ATTLIST name
note CDATA #IMPLIED>

(continued)

23.3 Parsing XML Documents with SAX 2.0 1149

Listing 23.5 perennials.dtd (continued)
<!ELEMENT year (#PCDATA)>

<!ELEMENT bloom (#PCDATA)>
<!ATTLIST bloom
code (E | EM | M | ML | L | E-L) #REQUIRED>

<!ELEMENT cost (#PCDATA)>
<!ATTLIST cost
discount CDATA #IMPLIED>
<!ATTLIST cost
currency (US | UK | CAN) "US">

23.3 Parsing XML Documents
with SAX 2.0
DOM processing is relatively straightforward since the DOM classes do all the “real”
parsing—you just have to look through the parsed result for the data you want. How-
ever, DOM can be quite wasteful if you only care about a small part of the document.
For example, suppose that you want to extract the first word from an XML document
representing an entire dictionary. DOM would require you to parse and store the
entire XML document (which could be huge in this case). With SAX, you need only
store the parts you care about and can stop parsing whenever you want. On the other
hand, SAX is a bit more work. The idea is that the system tells you when certain pars-
ing events such as finding a start tag (<language rating="good">), an end tag
(</language>), or a tag body (e.g., Java between the aforementioned start and
end tags). You have to decide what to do when these events occur. Are you a JSP pro-
grammer? Does this process sound familiar? It should—SAX processing is very simi-
lar to the way you go about defining custom JSP tag libraries (Section 20.7).

SAX is not a standard part of either Java 2 Standard Edition or the servlet and JSP
for use in your programs. Here is a summary of what is required.

1. Download a SAX-compliant parser. The parser provides the Java
classes that follow the SAX 2 API as specified by the WWW Consortium.
You can obtain a list of XML parsers in Java at http://www.xml.com/
pub/rg/Java_Parsers. We use the Apache Xerces-J parser in this
book. See http://xml.apache.org/xerces-j/. This parser
comes with the complete SAX API in Javadoc format.


2. Download the Java API for XML Processing (JAXP). This API
provides a small layer on top of SAX that lets you plug in different ven-
dor’s parsers without making any changes to your basic code. See
http://java.sun.com/xml/.
3. Set your CLASSPATH to include the SAX classes. In the case of
Apache Xerces, you need to include xerces_install_dir
xerces.jar. For example, on Windows you would do
set CLASSPATH=xerces_install_dirxerces.jar;%CLASSPATH%
If you wanted to use DOM from servlets and JSP, you would copy the
appropriate JAR file to the server’s lib directory (if supported),
unpack the JAR file (using jar -xvf) into the server’s classes
directory, or explicitly change the server’s CLASSPATH, usually by
modifying the server startup script.
4. Set your CLASSPATH to include the JAXP classes. These classes
are in jaxp_install_dir/jaxp.jar. For example, on Unix/Linux
and the C shell, you would do
setenv CLASSPATH jaxp_install_dir/jaxp.jar:$CLASSPATH
For use from servlets and JSP, see the preceding step.
5. Bookmark the SAX 2 and JAXP APIs. You can browse the
official API at http://www.megginson.com/SAX/Java/
javadoc/, but the API that comes with Apache Xerces is easier to
use because it is on your local system and is integrated with the
DOM and JAXP APIs. More information on SAX can be found
at http://www.megginson.com/SAX/.

Parsing
With SAX processing, there are two high-level tasks: creating a content handler and
invoking the parser with the designated content handler. The following list summa-
rizes the detailed steps needed to accomplish these tasks.

1. Tell the system which parser you want to use. This can be done in
a number of ways: through the javax.xml.parsers.SAX-
ParserFactory system property, through jre_dir/lib/
jaxp.properties, through the J2EE Services API and the class
specified in META-INF/services/javax.xml.parsers.SAX-
ParserFactory, or with a system-dependent default parser. The
system property is the easiest method. For example, the following
code permits users to specify the parser on the command line with the
-D option to java, and uses the Apache Xerces parser otherwise.
"javax.xml.parsers.SAXParserFactory";

23.3 Parsing XML Documents with SAX 2.0 1151

"org.apache.xerces.jaxp.SAXParserFactoryImpl";
}
...
}
2. Create a parser instance. First make an instance of a parser factory,
then use that to create a parser object.
SAXParserFactory factory = SAXParserFactory.newInstance();
SAXParser parser = factory.newSAXParser();
Note that you can use the setNamespaceAware and set-
Validating methods on the SAXParserFactory to make the
parser namespace aware and validating, respectively.
3. Create a content handler to respond to parsing events. This
handler is typically a subclass of DefaultHandler. You override any
or all of the following placeholders
• startDocument, endDocument
Use these methods to respond to the start and end of the document;
they take no arguments.
• startElement, endElement
Use these methods to respond to the start and end tags of an ele-
ment. The startElement method takes four arguments: the
namespace URI (a String; empty if no namespace), the
namespace or prefix (a String; empty if no namespace), the fully
qualified element name (a String; i.e., "prefix:mainName" if
there is a namespace; "mainName" otherwise), and an
Attributes object representing the attributes of the start tag. The
endElement method takes the same arguments except for the
attributes (since end tags are not permitted attributes).
• characters, ignoreableWhitespace
Use these methods to respond to the tag body. They take three
arguments: a char array, a start index, and an end index. A common
approach is to turn the relevant part of the character array into a
String by passing all three arguments to the String constructor.
Non-whitespace data is always reported to the characters
method. Whitespace is always reported to the ignoreable-
Whitespace method a parser is run in validating mode, but can be
reported to either method otherwise.
4. Invoke the parser with the designated content handler. You
invoke the parser by calling the parse method, supplying an input
stream, URI (represented as a string), or org.xml.sax.Input-
Source along with the content handler.
parser.parse(filename, handler);
The content handler does the rest.


23.4 SAX Example 1: Printing the
Outline of an XML Document

Listing 23.7 shows a content handler that responds to three parts of an XML docu-
ment: start tags, end tags, and tag bodies. It overrides the startElement, end-
Element, and characters methods to accomplish this. The handler simply prints
out the start element, end element, and first word of tag body, with two spaces of
indentation for each nesting level. To accomplish this task, the content handler over-
rides the following three methods:

• startElement
This method prints a message indicating that it found the start tag for
the element name. Any attributes associated with the element are
listed in parentheses. The method also puts spaces in front of the
printout, as specified by the indentation variable (initially 0).
Finally, it adds 2 to this variable.
• endElement
This method subtracts 2 from the indentation variable and then
prints a message indicating that it found the end tag for the element.
• characters
This method prints the first word of the tag body, leaving the
indentation level unchanged.

Listing 23.8 shows a program that lets the user specify a SAX-compliant parser and
an XML file, then invokes the parser with the outline-printing content handler just
described (and shown in Listing 23.7). Figure 23–4 shows the initial result, and Listing
23.6 shows the top part of the output when orders.xml (Listing 23.9) is selected.

Figure 23–4 Interactively selecting the orders.xml file.

23.4 SAX Example 1: Printing the Outline of an XML Document 1153

Listing 23.6 Partial output of SAXPrinter applied to orders.xml
Start tag: orders
Start tag: order
Start tag: count
37
End tag: count
Start tag: price
49.99
End tag: price
Start tag: book
Start tag: isbn
0130897930
End tag: isbn
Start tag: title
Core...
End tag: title
Start tag: authors
Start tag: author
Marty...
End tag: author
Start tag: author
Larry...
End tag: author
End tag: authors
End tag: book
End tag: order
Start tag: order
Start tag: count
1
End tag: count
Start tag: price
9.95
End tag: price
Start tag: yacht
Start tag: manufacturer
Luxury...
End tag: manufacturer
Start tag: model
M-1
End tag: model
Start tag: standardFeatures (oars=plastic, lifeVests=none)
false
End tag: standardFeatures
End tag: yacht
End tag: order
... (Rest of results omitted)
End tag: orders


Listing 23.7 PrintHandler.java
import org.xml.sax.*;
import org.xml.sax.helpers.*;
import java.util.StringTokenizer;

/** A SAX handler that prints out the start tags, end tags,
* and first word of tag body. Indents two spaces
* for each nesting level.
*/

public class PrintHandler extends DefaultHandler {
private int indentation = 0;

/** When you see a start tag, print it out and then
* increase indentation by two spaces. If the
* element has attributes, place them in parens
* after the element name.
*/

public void startElement(String namespaceUri,
String localName,
String qualifiedName,
Attributes attributes)
throws SAXException {
indent(indentation);
System.out.print("Start tag: " + qualifiedName);
int numAttributes = attributes.getLength();
// For <someTag> just print out "someTag". But for
// <someTag att1="Val1" att2="Val2">, print out
// "someTag (att1=Val1, att2=Val2).
if (numAttributes > 0) {
System.out.print(" (");
for(int i=0; i<numAttributes; i++) {
if (i>0) {
System.out.print(", ");
}
System.out.print(attributes.getQName(i) + "=" +
attributes.getValue(i));
}
System.out.print(")");
}
System.out.println();
indentation = indentation + 2;
}

(continued)


Listing 23.7 PrintHandler.java (continued)
/** When you see the end tag, print it out and decrease
* indentation level by 2.
*/

public void endElement(String namespaceUri,
String localName,
String qualifiedName)
indentation = indentation - 2;
System.out.println("End tag: " + qualifiedName);
}

/** Print out the first word of each tag body. */

public void characters(char[] chars,
int startIndex,
int endIndex) {
String data = new String(chars, startIndex, endIndex);
// Whitespace makes up default StringTokenizer delimeters
StringTokenizer tok = new StringTokenizer(data);
if (tok.hasMoreTokens()) {
System.out.print(tok.nextToken());
if (tok.hasMoreTokens()) {
System.out.println("...");
} else {
System.out.println();
}
}
}

private void indent(int indentation) {
for(int i=0; i<indentation; i++) {
System.out.print(" ");
}
}
}


Listing 23.8 SAXPrinter.java

/** A program that uses SAX to print out the start tags,
* end tags, and first word of tag body of an XML file.
*/

public class SAXPrinter {
}
String filename;
filename = args[0];
} else {
String[] extensions = { "xml", "tld" };
"XML Files",
extensions);
filename = "test.xml";
}
}
printOutline(filename);
System.exit(0);
}

public static void printOutline(String filename) {
DefaultHandler handler = new PrintHandler();
try {
"Error parsing " + filename + ": " + e;
}
}
}


Listing 23.9 orders.xml
<?xml version="1.0" ?>
<orders>
<order>
<count>37</count>
<price>49.99</price>
<book>
<isbn>0130897930</isbn>
<title>Core Web Programming Second Edition</title>
<authors>
<author>Marty Hall</author>
<author>Larry Brown</author>
</authors>
</book>
</order>
<order>
<count>1</count>
<price>9.95</price>
<yacht>
<manufacturer>Luxury Yachts, Inc.</manufacturer>
<model>M-1</model>
<standardFeatures oars="plastic"
lifeVests="none">
false
</standardFeatures>
</yacht>
</order>
<order>
<count>3</count>
<book>
<isbn>B000059Z4H</isbn>
<title>Harry Potter and the Order of the Phoenix</title>
<authors>
<author>J.K. Rowling</author>
</authors>
</book>
</order>
<order>
<count>2</count>
<price>10452689.01</price>
<yacht>
<manufacturer>We B Boats, Inc.</manufacturer>
<model>236-A</model>
<standardFeatures bowlingAlley="double"
tennisCourt="grass">

(continued)


Listing 23.9 orders.xml (continued)
true
</standardFeatures>
</yacht>
</order>
<order>
<count>13</count>
<book>
<isbn>0130897930</isbn>
<title>Core Web Programming Second Edition</title>
<authors>
<author>Marty Hall</author>
<author>Larry Brown</author>
</authors>
</book>
</order>
</orders>

23.5 SAX Example 2: Counting
Book Orders
One of the advantages of SAX over DOM is that SAX does not require you to process
and store the entire document; you can quickly skip over the parts that do not inter-
est you. The following example looks for sections of an XML file that look like this:
<orders>
...
<count>23</count>
<book>
<isbn>0130897930</isbn>
...
</book>
...
</orders>
The idea is that the program will count up how many copies of Core Web
Programming Second Edition (you did recognize that ISBN number, right?) are con-
tained in a set of orders. Thus, it can skip over most elements. Since SAX, unlike
DOM, does not store anything automatically, we need to take care of storing the
pieces of data that are of interest. The one difficulty in that regard is that the isbn
element comes after the count element. So, we need to record every count tempo-
rarily but only add the temporary value to the running total when the ISBN number
matches. To accomplish this task, the content handler (Listing 23.10) overrides the
following four methods:

23.5 SAX Example 2: Counting Book Orders 1159

• startElement
This method checks whether the name of the element is either count
or isbn. If so, it sets a flag that tells the characters method to be
on the lookout.
• endElement
This method checks whether the name of the element is either count
or isbn. If so, it turns off the flag that the characters method
watches.
• characters
If the count flag is set, this method parses the tag body as an int and
records the result in a temporary location. If the isbn flag is set, the
method reads the tag body and compares it to the ISBN number of
the second edition of Core Web Programming. If this comparison
results in a match, then the temporary count is added to the running
count.
• endDocument
This method prints out the running count. If the number of copies is
less than 250 (a real slacker!), it urges the user to buy more copies in
the future.

The CountBooks class (Listing 23.11) invokes a user-specifiable parser on an
XML file with CountHandler as the parser’s content handler. Figure 23–5 shows
the initial result and Figure 23–6 shows the final result, after orders.xml (Listing
23.9) is used as input.

Figure 23–5 Interactively selecting the orders.xml file.

Figure 23–6 Result of running CountBooks on orders.xml.


Listing 23.10 CountHandler.java
import java.util.StringTokenizer;

/** A SAX parser handler that keeps track of the number
* of copies of Core Web Programming ordered. Entries
* that look like this will be recorded:
* <XMP>
* ...
* <count>23</count>
* <book>
* <isbn>0130897930</isbn>
* ...
* </book>
* </XMP>
* All other entries will be ignored -- different books,
* orders for yachts, things that are not even orders, etc.
*/

public class CountHandler extends DefaultHandler {
private boolean collectCount = false;
private boolean collectISBN = false;
private int currentCount = 0;
private int totalCount = 0;

/** If you start the "count" or "isbn" elements,
* set a flag so that the characters method can check
* the value of the tag body.
*/

public void startElement(String namespaceUri,
String localName,
String qualifiedName,
Attributes attributes)
if (qualifiedName.equals("count")) {
collectCount = true;
currentCount = 0;
} else if (qualifiedName.equals("isbn")) {
collectISBN = true;
}
}

(continued)


Listing 23.10 CountHandler.java (continued)
/** If you end the "count" or "isbn" elements,
* set a flag so that the characters method will no
* longer check the value of the tag body.
*/

public void endElement(String namespaceUri,
String localName,
String qualifiedName)
if (qualifiedName.equals("count")) {
collectCount = false;
} else if (qualifiedName.equals("isbn")) {
collectISBN = false;
}
}

/** Since the "count" entry comes before the "book"
* entry (which contains "isbn"), we have to temporarily
* record all counts we see. Later, if we find a
* matching "isbn" entry, we will record that temporary
* count.
*/

public void characters(char[] chars,
int startIndex,
int endIndex) {
if (collectCount || collectISBN) {
String dataString =
new String(chars, startIndex, endIndex).trim();
if (collectCount) {
try {
currentCount = Integer.parseInt(dataString);
} catch(NumberFormatException nfe) {
System.err.println("Ignoring malformed count: " +
dataString);
}
} else if (collectISBN) {
if (dataString.equals("0130897930")) {
totalCount = totalCount + currentCount;
}
}
}
}

(continued)


Listing 23.10 CountHandler.java (continued)
/** Report the total number of copies ordered.
* Gently chide underachievers.
*/

public void endDocument() throws SAXException {
String message =
"You ordered " + totalCount + " copies of n" +
"Core Web Programming Second Edition.n";
if (totalCount < 250) {
message = message + "Please order more next time!";
} else {
message = message + "Thanks for your order.";
}
JOptionPane.showMessageDialog(null, message);
}
}

Listing 23.11 CountBooks.java

/** A program using SAX to keep track of the number
* of copies of Core Web Programming ordered. Entries
* that look like this will be recorded:<XMP>
* ...
* <count>23</count>
* <book>
* <isbn>0130897930</isbn>
* ...
* </book>
*
* </XMP>All other entries will be ignored -- different books,
* orders for yachts, things that are not even orders, etc.
*/

public class CountBooks {

(continued)


Listing 23.11 CountBooks.java (continued)
}
String filename;
filename = args[0];
} else {
String[] extensions = { "xml" };
"XML Files",
extensions);
filename = "orders.xml";
}
}
countBooks(filename);
System.exit(0);
}

private static void countBooks(String filename) {
DefaultHandler handler = new CountHandler();
try {
"Error parsing " + filename + ": " + e;
}
}
}


23.6 Transforming XML with XSLT
XSLT is a language for transforming XML documents into HTML, XML, or other
types of documents. When performing a transformation, an XSLT engine converts
the XML document according to formatting rules and XPath addresses specified in
an XML style sheet (XSL). The XPath information identifies the different parts of
the XML document for processing, and the style sheet information identifies the lay-
out of the output.
The benefit of XSLT is that you can define multiple style sheets for transforming a
single XML document. For example, a database could return a query in an XML for-
mat, and depending on the client protocol, HTTP or WAP, a servlet could use differ-
ent style sheets to convert the data into HTML or WML, respectively. As another
example of an XSLT application, consider an e-commerce business order; the order
could be sent to the supplier in an XML format and then processed by the recipient
with XSLT, using different XSL documents to convert the original order into sepa-
rate billing and shipping documents.
Specifications for XSLT, XSL, and XPath technologies are maintained by the
WWW Consortium. These specifications are located at:

XSLT 1.0
http://www.w3.org/TR/xslt.html

XSL 1.0
http://www.w3.org/TR/xsl/

XPath 1.0
http://www.w3.org/TR/xpath.html

Upcoming specifications are summarized at http://www.w3.org/Style/XSL/.
In addition, an excellent XSLT resource site, sponsored by GoXML, is located at
http://www.xslt.com/.

XSLT is not a standard part of either Java 2 Standard Edition or the servlet and JSP
for use in your programs. Here is a summary of what is required:

1. Download an XSLT-compliant transformer. The transformer
provides the Java classes that follow the XSLT 1.0 specification as
specified by the WWW Consortium. You can obtain a list of XSLT
parsers at http://www.w3.org/Style/XSL/ or

23.6 Transforming XML with XSLT 1165

http://www.xslt.com/xslt_tools_engines.htm. We use
the Apache Xalan-J transformer in this book. See
http://xml.apache.org/xalan-j/.
2. Set your CLASSPATH to include the DOM and SAX classes. XSLT
builds upon DOM and SAX for handling the document processing. In
the case of Apache Xalan-J, you need to include xerces.jar in the
CLASSPATH. See Section 23.1 (Parsing XML Documents with DOM
Level 2) and Section 23.3 (Parsing XML Documents with SAX 2.0) for
configuration of Apache Xerces-J. Note that xerces.jar is included
in the Xalan-J installation directory.
3. Set your CLASSPATH to include the XSLT classes. With Xalan,
these classes are in xalan_install_dirxalan.jar. For exam-
ple, for desktop application on Windows, you would do
set CLASSPATH=xalan_install_dirxalan.jar;
%CLASSPATH%
On Unix/Linux and the C shell, you would do
setenv CLASSPATH xalan_install_dir/xalan.jar:
$CLASSPATH
If you wanted to use XSLT from servlets and JSP, you would copy the
appropriate DOM, SAX, and XSLT JAR files to the server’s lib direc-
tory (if supported), unpack the JAR files (using jar -xvf) into the
server’s classes directory, or explicitly change the server’s CLASSPATH,
usually by modifying the server’s startup script.
4. Bookmark the XSL 1.0 and XPath 1.0 specifications. The official
documentation for these two specifications can be found at
http://www.w3.org/Style/XSL/.
5. Bookmark the XSLT specification. The official XSLT specification
can be found at http://www.w3.org/TR/xslt.html. The XSLT
specification is implemented in Apache Xalan through the Transfor-
mation API for XML (TrAX). The complete TrAX API comes with
Xalan-J in Javadoc format and is also available on-line at
http://xml.apache.org/xalan-j/apidocs/.

Translating
With XSLT processing, there are two high-level tasks, establishing an XSL template
from which to build a transformer and invoking the transformer on the XML docu-
ment. The following list summarizes the detailed steps needed to accomplish these
tasks.

1. Tell the system which parser you want to use for transforma-
tions. This can be done in a number of ways: through the
javax.xml.transform.TransformFactory system property,


through the jre_dir/lib/jaxp.properties, through the J2EE
Services API and the class specified in the META-INF/services/
javax.xml.transform.TransformFactory, or with a sys-
tem-dependent default processor. As XSLT depends on DOM and
SAX, you can tell the system which DOM and SAX parser to use for
processing the document. See Section 23.1 and Section 23.3 for infor-
mation on configuring DOM and SAX parsers. By default, Apache
Xalan-J uses the Apache Xerces-J DOM and SAX parsers.
2. Establish a factory in which to create transformers. Before
processing an XML document, you first need to establish a
TransformerFactory. The factory allows you to create different
transformers for different style sheet templates.
TransformerFactory factory =
TransformerFactory.newInstance();
3. Generate a transformer for a particular style sheet template.
For each style sheet you can generate a separate transformer to apply
to multiple XML documents.
Source xsl = new StreamSource(xslStream);
Templates template = factory.newTemplates(xsl);
Transformer transformer = template.newTransformer();
Typically, the XSL source is a StreamSource object. You can easily
convert an XSL document to a StreamSource through a File,
Reader, InputStream, or URI (represented as a string) reference
to the document.
4. Invoke the transformer to process the source document. You
invoke the transformation by calling the transform method, supply-
ing the XML source and a Result object to receive the transformed
document.
Source xml = new StreamSource(xmlStream);
Result result = new StreamResult(outputStream);
tranformer.transform(xml, result);
Similar to the XSL source, the XML source is typically a Stream-
Source constructed from a File, Reader, InputStream, or URI.
The transformed StreamResult can be a File, Writer, Output-
Stream or URI.

Listing 23.12 presents a class for preforming XSLT transformations of documents.
The XML and XSL source documents can be either Readers or Files, and the
resulting transformed document can be a Writer or File. The advantage of han-
dling the documents as Readers and Writers is that they can remain in memory
and can easily be processed as strings by a StringReader or CharArrayReader
for the source documents and a StringWriter or CharArrayWriter for the
result document. For example, a servlet could receive a database query as an XML

23.6 Transforming XML with XSLT 1167

character stream, process the input using XSLT, and deliver the result as an HTML
document to a browser client. At no time do the XML document and transformed
result need to reside on disk.

Listing 23.12 XslTransformer.java
package cwp;

import javax.xml.transform.*;
import javax.xml.transform.stream.*;
import java.io.*;
import java.util.*;

/** Creates an XSLT transformer for processing an XML document.
* A new transformer, along with a style template are created
* for each document transformation. The XSLT, DOM, and
* SAX processors are based on system default parameters.
*/

public class XslTransformer {
private TransformerFactory factory;

public XslTransformer() {
factory = TransformerFactory.newInstance();
}

/** Transform an XML and XSL document as <code>Reader</code>s,
* placing the resulting transformed document in a
* <code>Writer</code>. Convenient for handling an XML
* document as a String (<code>StringReader</code>) residing
* in memory, not on disk. The output document could easily be
* handled as a String (<code>StringWriter</code>) or as a
* <code>JSPWriter</code> in a JavaServer page.
*/

public void process(Reader xmlFile, Reader xslFile,
Writer output)
throws TransformerException {
process(new StreamSource(xmlFile),
new StreamSource(xslFile),
new StreamResult(output));
}

/** Transform an XML and XSL document as <code>File</code>s,
* placing the resulting transformed document in a
* <code>Writer</code>. The output document could easily
* be handled as a String (<code>StringWriter</code)> or as
* a <code>JSPWriter</code> in a JavaServer page.
*/

(continued)


Listing 23.12 XslTransformer.java (continued)
public void process(File xmlFile, File xslFile,
Writer output)
new StreamResult(output));
}

/** Transform an XML <code>File</code> based on an XSL
* <code>File</code>, placing the resulting transformed
* document in a <code>OutputStream</code>. Convenient for
* handling the result as a <code>FileOutputStream</code> or
* <code>ByteArrayOutputStream</code>.
*/

public void process(File xmlFile, File xslFile,
OutputStream out)
new StreamResult(out));
}

/** Transform an XML source using XSLT based on a new template
* for the source XSL document. The resulting transformed
* document is placed in the passed in <code>Result</code>
* object.
*/

public void process(Source xml, Source xsl, Result result)
try {
Templates template = factory.newTemplates(xsl);
Transformer transformer = template.newTransformer();
transformer.transform(xml, result);
} catch(TransformerConfigurationException tce) {
throw new TransformerException(
tce.getMessageAndLocation());
} catch (TransformerException te) {
throw new TransformerException(
te.getMessageAndLocation());
}
}
}

23.7 XSLT Example 1: XSLT Document Editor 1169

23.7 XSLT Example 1: XSLT
Document Editor

Listing 23.13 shows a simple Swing document editor that presents three tabbed
panes: one for an XML document, one for an XSL style sheet, and one for a resulting
XSLT-transformed document. Both the XML and XSL document panes are editable,
so after you load the XML and XSL files from disk you can edit the documents
directly. Each tabbed pane contains a scrollable DocumentPane that inherits from a
JEditorPane (see Listing 23.14). The XML and XSL panes are treated as plain
text, and the XSLT pane is treated as HTML. If an XML file and XSL file are loaded,
selecting the XSLT tab will invoke an XslTransformer (Listing 23.12) to process
the XML file by using the style sheet, and present the results as HTML in the XSLT
document pane.

Listing 23.13 XsltExample.java
import java.awt.*;
import java.awt.event.*;
import javax.swing.event.*;
import java.io.*;
import cwp.XslTransformer;

/** A document editor to process XML and XSL text using
* XSLT and presenting the results as HTML. Three tabbed panes
* are presented: an editable text pane for the XML document,
* an editable text pane for the XSL style sheet, and a non-
* editable HTML pane for the HTML result. If an XML and XSL
* file are loaded, then selecting the XSLT tab will perform
* the transformation and present the results. If there is
* a problem processing the XML or XSL document, then a
* message box is popped up describing the problem.
*/

public class XsltExample extends JFrame
implements ChangeListener {
private static final int XML = 0;
private static final int XSL = 1;
private static final int XSLT = 2;
private static final String DEFAULT_TITLE = "XSLT Example";

(continued)


Listing 23.13 XsltExample.java (continued)
private static final String[] tabTitles =
{ "XML", "XSL", "XSLT" };
private static final String[] extensions =
{ "xml", "xsl", "html" };
private Action openAction, saveAction, exitAction;
private JTabbedPane tabbedPane;
private DocumentPane[] documents;
private XslTransformer transformer;

public XsltExample() {
super(DEFAULT_TITLE);
transformer = new XslTransformer();
Container content = getContentPane();
content.setBackground(SystemColor.control);

// Set up menus
JMenuBar menubar = new JMenuBar();
openAction = new OpenAction();
saveAction = new SaveAction();
exitAction = new ExitAction();
JMenu fileMenu = new JMenu("File");
fileMenu.add(openAction);
fileMenu.add(saveAction);
fileMenu.add(exitAction);
menubar.add(fileMenu);
setJMenuBar(menubar);

// Set up tabbed panes
tabbedPane = new JTabbedPane();
documents = new DocumentPane[3];
for(int i=0; i<3; i++) {
documents[i] = new DocumentPane();
JPanel panel = new JPanel();
JScrollPane scrollPane = new JScrollPane(documents[i]);
panel.add(scrollPane);
tabbedPane.add(tabTitles[i], scrollPane);
}
documents[XSLT].setContentType(DocumentPane.HTML);
// JEditorPane has a bug, whereas the setText method does
// not properly recognize an HTML document that has a META
// element containing a CONTENT-TYPE, unless the EditorKit
// is first created through setPage. Xalan automatically
// adds a META CONTENT-TYPE to the document.

(continued)


documents[XSLT].loadFile("XSLT-Instructions.html");
documents[XSLT].setEditable(false);
tabbedPane.addChangeListener(this);
content.add(tabbedPane, BorderLayout.CENTER);

setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE);
setSize(450, 350);
setVisible(true);
}

/** Checks to see which tabbed pane was selected by the
* user. If the XML and XSL panes hold a document, then
* selecting the XSLT tab will perform the transformation.
*/

public void stateChanged(ChangeEvent event) {
int index = tabbedPane.getSelectedIndex();
switch (index) {
case XSLT: if (documents[XML].isLoaded() &&
documents[XSL].isLoaded()) {
doTransform();
}
case XML:
case XSL: updateMenuAndTitle(index);
break;
default:
}
}

/** Retrieve the documents in the XML and XSL pages
* as text (String), pipe into a StringReader, and
* perform the XSLT transformation. If an exception
* occurs, present the problem in a message dialog.
*/

private void doTransform() {
StringWriter strWriter = new StringWriter();
try {
Reader xmlInput =
new StringReader(documents[XML].getText());
Reader xslInput =
new StringReader(documents[XSL].getText());
transformer = new XslTransformer();
transformer.process(xmlInput, xslInput, strWriter);

(continued)


} catch(TransformerException te) {
JOptionPane.showMessageDialog(this,
"Error: " + te.getMessage());
}
documents[XSLT].setText(strWriter.toString());
}

/** Update the title of the application to present
* the name of the file loaded into the selected
* tabbed pane. Also, update the menu options (Save,
* Load) based on which tab is selected.
*/

private void updateMenuAndTitle(int index) {
if ((index > -1) && (index < documents.length)) {
saveAction.setEnabled(documents[index].isLoaded());
openAction.setEnabled(documents[index].isEditable());
String title = DEFAULT_TITLE;
String filename = documents[index].getFilename();
if (filename.length() > 0) {
title += " - [" + filename + "]";
}
setTitle(title);
}
}

/** Open a file dialog to either load a new file to or save
* the existing file in the present document pane.
*/

private void updateDocument(int mode) {
int index = tabbedPane.getSelectedIndex();
String description = tabTitles[index] + " Files";
String filename = ExtensionFileFilter.getFileName(".",
description,
extensions[index],
mode);
if (filename != null) {
if (mode==ExtensionFileFilter.SAVE) {
documents[index].saveFile(filename);
} else {
documents[index].loadFile(filename);
}
updateMenuAndTitle(index);
}
}

(continued)


new XsltExample();
}

// Open menu action to load a new file into a
// document when selected.
class OpenAction extends AbstractAction {
public OpenAction() {
super("Open ...");
}
public void actionPerformed(ActionEvent event) {
updateDocument(ExtensionFileFilter.LOAD);
}
}

// Save menu action to save the document in the
// selected pane to a file.
class SaveAction extends AbstractAction {
public SaveAction() {
super("Save");
setEnabled(false);
}
updateDocument(ExtensionFileFilter.SAVE);
}
}

// Exit menu action to close the application.
class ExitAction extends AbstractAction {
public ExitAction() {
super("Exit");
}
System.exit(0);
}
}
}


Listing 23.14 DocumentPane.java
import java.awt.*;
import java.awt.event.*;
import java.io.*;
import java.net.*;

/** A JEditorPane with support for loading and saving the
* document. The document should be one of two
* types: "text/plain" (default) or "text/html".
*/

public class DocumentPane extends JEditorPane {
public static final String TEXT = "text/plain";
public static final String HTML = "text/html";

private boolean loaded = false;
private String filename = "";

/** Set the current page displayed in the editor pane,
* replacing the existing document.
*/

public void setPage(URL url) {
loaded = false;
try {
super.setPage(url);
File file = new File(getPage().toString());
setFilename(file.getName());
loaded = true;
} catch (IOException ioe) {
System.err.println("Unable to set page: " + url);
}
}

/** Set the text in the document page, replace the exiting
* document.
*/

public void setText(String text) {
super.setText(text);
setFilename("");
loaded = true;
}

(continued)


Listing 23.14 DocumentPane.java (continued)
/** Load a file into the editor pane.
*
* Note that the setPage method of JEditorPane checks the
* URL of the currently loaded page against the URL of the
* new page to laod. If the two URLs are the same, then
* the page is <b>not</b> reloaded.
*/

public void loadFile(String filename) {
try {
File file = new File(filename);
setPage(file.toURL());
} catch (IOException mue) {
System.err.println("Unable to load file: " + filename);
}
}

public void saveFile(String filename) {
try {
File file = new File(filename);
FileWriter writer = new FileWriter(file);
writer.write(getText());
writer.close();
setFilename(file.getName());
} catch (IOException ioe) {
System.err.println("Unable to save file: " + filename);
}
}

/** Return the name of the file loaded into the editor pane. */

public String getFilename() {
return(filename);
}

/** Set the filename of the document. */

public void setFilename(String filename) {
this.filename = filename;
}

/** Return true if a document is loaded into the editor
* page, either through <code>setPage</code> or
* <code>setText</code>.
*/

public boolean isLoaded() {
return(loaded);
}
}


The results for the XsltExample are shown in Figure 23–7 through Figure
23–9. Specifically, the result for the XML document pane with the loaded file,
perennials.xml (Listing 23.4), is shown in Figure 23–7. The XSL document pane
with the loaded file, perennials.xsl (Listing 23.15), is shown in Figure 23–8.
Finally, the XSLT transformation of the XML document is presented in Figure 23–8.
For this example, all daylilies awarded a Stout Medal are selected from the XML file
and listed in an HTML TABLE. For each daylily matching the criteria, the year of
hybridization, cultivar name, bloom season, and cost are presented in the table.
Note that for Apache Xalan-J to perform the XSLT transformation, the DTD,
perennials.dtd, must be accessible on-line from http://archive.coreweb-
programming.com/dtds/. If you would like to test this example locally, place the
file, perennials.dtd, in a dtds subdirectory below the XML file, and change the
DOCTYPE statement from
<!DOCTYPE perennials SYSTEM
"http://archive.corewebprogramming.com/dtds/perennials.dtd">
to

Note that in the XSL file, if you include a doctype-public attribute for the
xsl:output element, then Xalan will include a DOCTYPE statement for the first
line of the output document.

Core Approach

Include a doctype-public attribute in the xsl:output element to
produce a DOCTYPE statement in the transformed output.

Listing 23.15 perennials.xsl
<?xml version="1.0"?>
<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="html"
doctype-public ="-//W3C//DTD HTML 4.0 Transitional//EN"/>

<xsl:template match="/">
<HTML>
<HEAD>

(continued)

Listing 23.15 perennials.xsl (continued)
<TITLE>Daylilies</TITLE>
</HEAD>
<BODY>
<TABLE CELLPADDING="3">
<CAPTION>Stout Medal Award</CAPTION>
<TR>
<TH>Year</TH>
<TH>Cultivar</TH>
<TH>Bloom Season</TH>
<TH>Cost</TH>
</TR>

<xsl:apply-templates
select="/perennials/daylily[award/name='Stout Medal']"/>
<TR>
<TD COLSPAN="4" ALIGN="CENTER">
E-early M-midseason L-late</TD>
</TR>
</TABLE>
</BODY>
</HTML>
</xsl:template>

<xsl:template match="daylily">
<TR>
<TD><xsl:value-of select="award/year"/></TD>
<TD><xsl:value-of select="cultivar"/></TD>

<TD ALIGN="CENTER"><xsl:value-of select="bloom/@code"/></TD>
<TD ALIGN="RIGHT"><xsl:value-of select="cost"/></TD>
</TR>
</xsl:template>

</xsl:stylesheet>


Figure 23–7 Presentation of XML tabbed pane in XsltExample with
perennials.xml (Listing 23.4) loaded.

Figure 23–8 Presentation of XSL tabbed pane in XsltExample with
perennials.xsl (Listing 23.15) loaded.

23.8 XSLT Example 2: Custom JSP Tag 1179

Figure 23–9 Result of XSLT transformation of perennials.xml (Listing 23.4) and
perennials.xsl (Listing 23.15).

23.8 XSLT Example 2:
Custom JSP Tag

In this example, a custom JSP tag is used along with XSLT to output a listing of Stout
Medal daylilies in an HTML table, formatted specifically for the client’s browser
type. Netscape 4.7 and earlier do not support the HTML 4.0 elements: THEAD,
TBODY, and TFOOT. However, these three elements are supported by Internet
Explorer 4.x and later (see Section 2.4, “Tables,” for details on these elements). Thus,
in this example, two separate style sheets are used to process the XML file,
perennials.xml (Listing 23.4), and depending on which client browser accesses
the JavaServer Page, Daylilies.jsp (Listing 23.18), the correct style sheet is
applied. The first XSL document, perennials-ie.xsl (Listing 23.16) formats the
daylilies in a table suitable for Internet Explorer by using THEAD, TBODY, and TFOOT
elements, and the second XSL document, perennials-ns.xsl (Listing 23.17)
formats the daylilies in a basic table suitable for Netscape.
The Tag Library Descriptor (TLD) for the custom JSP tag, xsltransform,
used in Daylilies.jsp, is presented in xsltransform.tld (Listing 23.19). The
tag class for this custom tag is cwp.tags.XslTransformTag. Three attributes are


defined for the tag: xml, the source XML file (required), xslie, the XSL style sheet
targeting Internet Explorer, and xslns (required), the XSL style sheet targeting
Netscape. The xslns style sheet is required because this is the default style sheet
applied if the client browser is not Internet Explorer. For additional information on
custom tags, see Section 20.7 (Defining Custom JSP Tags).
The tag class, XslTransformTag, is shown in Listing 23.20. The doStartTag
method builds the File objects for the XML and XSL document, where the XSL
document applied for the style sheet is determined by the User-Agent header in
the HTTP request. After the source files are determined, the XSLT transform is per-
formed with XslTransformer (Listing 23.12) and the result is sent to the Jsp-
Writer.
This example requires numerous files that must be located in the proper directo-
ries on the server to run correctly. In Figure 23–10, we illustrate where to place the
files on a Tomcat server. If the DTD, perennials.dtd, is not accessible on-line
from http://www.corewebprogramming.com/dtds/, then place the DTD file
in a dtds subdirectory as illustrated and modify the DOCTYPE to

The result for Internet Explorer 5.0 on Windows 2000 is shown in Figure 23–11 and
the result for Netscape 4.7 on Windows 98 is shown in Figure 23–12.

TOMCAT_HOME/webapps/ROOT/

cwp/Daylilies.jsp

cwp-tags/xsltransform.tld

WEB-INF/classes/cwp/perennials.xml
perennials-ie.xml
perennials-ns.xsl
XslTransformer.class

dtds/perennials.dtd (if local)

tags/XslTransformTag.class

Figure 23–10 Location of files for custom JSP tag example on Tomcat.

Listing 23.16 perennials-ie.xsl


<xsl:stylesheet
xmlns:xsl="http://www.w3.org/1999/XSL/Transform"
version="1.0">
<xsl:output method="html"/>
<TABLE CELLPADDING="3" RULES="GROUPS" ALIGN="CENTER">
<COLGROUP>
<COL ALIGN="CENTER"/>
<COL ALIGN="LEFT"/>
<COL ALIGN="CENTER"/>
</COLGROUP>
<COLGROUP ALIGN="RIGHT"/>
<THEAD>
<TR>
<TH>Year</TH>
<TH>Cultivar</TH>
<TH>Cost</TH>
</TR>
</THEAD>
<TBODY>

</TBODY>
<TFOOT>
<TR>
<TD COLSPAN="4">E-early M-midseason L-late</TD>
</TR>
</TFOOT>
</TABLE>
</xsl:template>
<TR>
<TD><xsl:value-of select="bloom/@code"/></TD>
<TD><xsl:value-of select="cost"/></TD>
</TR>
</xsl:template>
</xsl:stylesheet>

Listing 23.17 perennials-ns.xsl


<xsl:stylesheet version="1.0"
xmlns:xsl="http://www.w3.org/1999/XSL/Transform">
<xsl:output method="html"/>
<TABLE CELLPADDING="3" BORDER="1" ALIGN="CENTER">
<TR>
<TH>Year</TH>
<TH>Cultivar</TH>
<TH>Cost</TH>
</TR>

<TR>
<TD COLSPAN="4" ALIGN="CENTER">
E-early M-midseason L-late</TD>
</TR>
</TABLE>
</xsl:template>
<TR>
<TD ALIGN="CENTER"><xsl:value-of select="bloom/@code"/></TD>
<TD ALIGN="RIGHT"><xsl:value-of select="cost"/></TD>
</TR>
</xsl:template>
</xsl:stylesheet>


Listing 23.18 Daylilies.jsp
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<html>
<head>
<title>Daylilies</title>
</head>
<body>
<%@ taglib uri="cwp-tags/xsltransform.tld" prefix="cwp" %>

<H1 ALIGN="CENTER">Katie's Favorite Daylilies
<p>
<cwp:xsltransform xml='perennials.xml'
xslie='perennials-ie.xsl'
xslns='perennials-ns.xsl'
/>

</body>
</html>

Listing 23.19 xsltransform.tld
<?xml version="1.0" encoding="ISO-8859-1" ?>
<!DOCTYPE taglib
PUBLIC "-//Sun Microsystems, Inc.//DTD JSP Tag Library 1.1//EN"
"http://java.sun.com/j2ee/dtds/web-jsptaglibrary_1_1.dtd">

<taglib>
<tlibversion>1.0</tlibversion>
<jspversion>1.1</jspversion>
<shortname>cwp</shortname>
<urn></urn>
<info>
A tag library from Core Web Programming,
http://www.corewebprogramming.com/.
</info>

<tag>
<name>xsltransform</name>
<tagclass>cwp.tags.XslTransformTag</tagclass>
<info>Applies xslt transform based on browser type.</info>
<attribute>

(continued)


Listing 23.19 xsltransform.tld (continued)
<name>xml</name>
<required>yes</required>
</attribute>
<attribute>
<name>xslie</name>
<required>false</required>
</attribute>
<attribute>
<name>xslns</name>
<required>true</required>
</attribute>
</tag>
</taglib>

Listing 23.20 XslTransformTag.java
package cwp.tags;

import java.io.*;
import javax.servlet.*;
import javax.servlet.jsp.*;
import javax.servlet.http.*;
import javax.servlet.jsp.tagext.*;
import cwp.XslTransformer;

/** A tag that translates an XML document to HTML using XSLT.
* Depending on the client browser type, either an XSL style
* targeting Internet Explorer or Netscape (default) is
* applied.
*/

public class XslTransformTag extends TagSupport {
private static final String FS =
System.getProperty("file.separator");
private static final int IE = 1;
private static final int NS = 2;
private String xml, xslie, xslns;
public void setXml(String xml) {
this.xml = xml;
}

(continued)


Listing 23.20 XslTransformTag.java (continued)
public String getXml() {
return(xml);
}

public void setXslie(String xslie) {
this.xslie = xslie;
}

public String getXslie() {
return(xslie);
}

public void setXslns(String xslns) {
this.xslns = xslns;
}

public String getXslns() {
return(xslns);
}

public int doStartTag() throws JspException {
// Determine the path to XML and XSL source files.
// The path of SERVLET_HOME/WEB-INF/classes/cwp/ is
// assumed for the location of the source files.
String FS = System.getProperty("file.separator");
ServletContext context = pageContext.getServletContext();
String path = context.getRealPath(FS) + "WEB-INF" + FS +
"classes" + FS + "cwp" + FS;

HttpServletRequest request =
(HttpServletRequest)pageContext.getRequest();

// Use either IE or NS style sheet depending on
// browser type.
File xslFile = null;
if ((browserType(request) == IE) && (getXslie() != null)) {
xslFile = new File(path + getXslie());
} else {
xslFile = new File(path + getXslns());
}
File xmlFile = new File(path + getXml());
try {
JspWriter out = pageContext.getOut();
XslTransformer transformer = new XslTransformer();
transformer.process(xmlFile, xslFile, out);
}

(continued)


Listing 23.20 XslTransformTag.java (continued)
catch(TransformerException tx) {
context.log("XslTransformTag: " + tx.getMessage());
}
return(SKIP_BODY);
}

// Determine the browser type based on the User-Agent
// HTTP request header.
private int browserType(HttpServletRequest request) {
int type = NS;
String userAgent = request.getHeader("User-Agent");
if ((userAgent != null) &&
(userAgent.indexOf("IE") >=0)) {
type = IE;
}
return(type);
}
}

Figure 23–11 Transformation of perennials.xml through a custom JSP tag for
Internet Explorer 5.0 on Windows 2000.

23.9 Summary 1187

Figure 23–12 Transformation of perennials.xml through a custom JSP tag on
Netscape 4.7 on Windows 98.

23.9 Summary
Wow! This wraps up the section on server-side programming. Now you know how to
process XML documents by using DOM, SAX, and XSLT. You also know how to
write servlets and JavaServer Pages, perform HTTP tunneling, and communicate
with databases by using JDBC. No doubt you’ve already started to put these technol-
ogies to use in new and exciting Web applications. No doubt your boss is impressed
(or are you the boss now?).
In the next section we move back to the client side and discuss JavaScript, an
interpreted language that runs in the browser. JavaScript can be applied in a variety
of ways to make Web pages more flexible and dynamic, but one of the major applica-
tions is to check the format of HTML form data before it is submitted to the server.

Ch23 xml processing_with_java

More Related Content

What's hot (20)

Similar to Ch23 xml processing_with_java (20)

Recently uploaded (20)

Ch23 xml processing_with_java