AJAX - The Complete Reference

Chapter 4: Data Formats 163

PART I

var items = xmlDoc.selectNodes("//bookmark");

for (var i=0;i<items.length;i++)

{

var title = items[i].selectNodes("title/text()")[0].text;

var url = items[i].selectNodes("url/text()")[0].text;

...snip...

/* print out the complete bookmark */

}

Again, we omit the gory output details and direct interested readers to view the complete

Internet Explorer–specific version at http://ajaxref.com/ch4/xmlxpathie.html.

Given that you have seen both of the major browsers support XPath, it is certainly

possible, as done previously, to create a wrapper to abstract the differing details away. In the

case of XPath and XSLT (to be discussed shortly), this is more than a bit of a chore, so we

opt to use one of the popular libraries online to do the task, in this case Sarissa (http://

sourceforge.net/projects/sarissa). Once the Sarissa library is included, we fetch the XML

tree, but then inform Sarissa that we will be using XSLT and XPath:

xmlDoc = xhr.responseXML;

xmlDoc.setProperty('SelectionNamespaces',

'xmlns:xsl="http://www.w3.org/1999/XSL/Transform"');

xmlDoc.setProperty('SelectionLanguage', 'XPath');

Now we run a selection to fetch the list of <bookmark> tags and once again iterate over

the list pulling out the values within each tag. You’ll note that the syntax Sarissa provides is

similar to the IE style, which is a bit cleaner than the Firefox approach.

var items = xmlDoc.selectNodes("//bookmark");

for (var i=0;i<items.length;i++)

{

var title = items[i].selectSingleNode("title").firstChild.nodeValue;

var url = items[i].selectSingleNode("url").firstChild.nodeValue;

... snip ...

/* print out the complete bookmark */

}

The Sarissa example, which works in a multitude of browsers, can be found at

http://ajaxref.com/ch4/xmlxpathsarissa.html. A version that uses the Google AJAXSLT

library (http://goog-ajaxslt.sourceforge.net/) is also supplied online at http://ajaxref.com/

ch4/xmlxpathgoogle.html, in case readers prefer this library.

Transforming Responses with XSLT

XSLT (eXtensible Style Language Transformations) is a powerful technology used to

convert one markup language to another (www.w3.org/TR/xslt). A common use of XSLT

is to create XHTML output templates from XML documents. We could spend a tremendous

amount of time explaining the syntax of XSLT, but instead we focus here on its application

within our simple example to illustrate the approach and pique reader’s interest in the

technology.