use the xmlstreamreader class

http://www.salesforce.com/us/developer/docs/apexcode/Content/apex_classes_XmlStream_reader.htm

here is a class that wraps it:

/* Copyright (c) 2008 salesforce.com, inc. All rights reserved. Redistribution and use in source and binary forms, with or without modification, are permitted provided that the following conditions are met: 1. Redistributions of source code must retain the above copyright notice, this list of conditions and the following disclaimer. 2. Redistributions in binary form must reproduce the above copyright notice, this list of conditions and the following disclaimer in the documentation and/or other materials provided with the distribution. 3. The name of the author may not be used to endorse or promote products derived from this software without specific prior written permission. THIS SOFTWARE IS PROVIDED BY THE AUTHOR "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. */ /* This class implements a small portion of the wc3 xml dom model. Generally useful for simple XML return objects. Note: stores namespaces in the attributes map for now. No node typing done at this time can parse into DOM trees the XML return objects from Google, Amazon and others. large parse trees will consume suprising amounts of memory */ public class XMLDom { // Constructor public XMLDom(string str) { parseFromString(str); } public XMLDom( ) { } public void parseFromString(string str) { XmlStreamReader reader = new XmlStreamReader(str); reader.setCoalescing(true); parseXmlReader (root , reader); } // debugging assistance public void dumpAll() { root.dumpAll(); } public void dumpList(Element[] l) { for(Element e:l) e.dump(); } // given a parent node and a stream reader, populates the tree below here (recursive) void parseXmlReader( Element parent, XmlStreamReader reader ) { try { while(reader.hasNext()) { if (reader.getEventType() == XmlTag.START_ELEMENT) { string nodeName = reader.getLocalName(); string pre = reader.getPrefix(); if ( pre != null && pre != '' ) { //system.debug( reader.getPrefix() ); nodeName = reader.getPrefix() + ':' + nodeName; } Element child = new Element( nodeName ); // add all attributes from this element for (integer i=0; i<reader.getAttributeCount(); i++) { child.attributes.put( reader.getAttributeLocalName(i), reader.getAttributeValueAt(i) ); } // add namespace info to each node/element ( for now storing on attributes map) for (integer j=0; j<reader.getNamespaceCount(); j++) { string prefix = 'xmlns'; if (reader.getnamespaceprefix(j)!=null) prefix = reader.getnamespaceprefix(j); child.attributes.put( prefix , reader.getnamespaceuriat(j) ); } parent.appendChild(child); // add the new element to current parent reader.next(); parseXmlReader(child, reader) ; // recurse } else if (reader.getEventType() == XmlTag.END_ELEMENT) { reader.next(); return; } else if (reader.getEventType() == XmlTag.CHARACTERS) { if ( ! reader.isWhitespace()) { parent.nodeValue += reader.getText(); // add text to current element } reader.next(); } else { reader.next(); } } } catch(Exception e) { // ParseError if we get a truncated response, allow it system.debug(e); } } // access nodes in the tree using these getters public List<Element> getElementsByTagName(string nam) { return root.getElementsByTagName(nam); } public Element getElementByTagName(string nam) { List<Element> r = root.getElementsByTagName(nam); if (r.size() == 0) return null; return r[0]; } // utility dom functions public Element ownerDocument() { return root; } // everything in the dom is found as childNodes under this root element public Element root = new Element('#document'); public integer debug =0; // dump out the element tree public String toXmlString() { return root.toXmlString(); } /* * Element class definition This following class implements a small portion of the wc3 xml dom model. Generally useful for simple XML return objects. for a properties and methods complete list see: http://www.w3schools.com/dom/dom_node.asp For simplicity, Nodes are the same as Elements in this class. Nodes have text directly in them, rather than a seperate text node child The following describes the implemented portion, some w3c properties are now methods. Property Description nodeName Returns the name of a node, depending on its type nodeValue Sets or returns the value of a node, depending on its type childNodes Returns a NodeList of child nodes for a node parentNode Returns the parent node of a node attributes Returns a NamedNodeMap of attributes for the element, also contains name space entries getElementByTagName() Returns list of elements matching tag name (document and element) firstChild() Returns the first child of a node removeChild() Removes a child node appendChild() Adds a new child node to the end of the list of children of a node getAttribute() Returns the value of an attribute hasChildNodes() Returns whether the element has any child nodes isEqualNode() Checks if two nodes are equal textContent() returns the textual content of a node cloneNode() Clones a node hasAttributes() Returns whether the element has any attributes isSameNode() Checks if two nodes are the same node ownerDocument() Returns the root element (document object) for a node *** NOT Implemented at this time *** lastChild() Returns the last child of a node nodeType Returns the type of a node , all nodes are the same type currently baseURI Returns the absolute base URI of a node localName Returns the local part of the name of a node namespaceURI Returns the namespace URI of a node nextSibling Returns the node immediately following a node insertBefore() Inserts a new child node before an existing child node replaceChild() Replaces a child node */ public class Element { // Element(Element p, string n) { parentNode = p; nodeName = n; } public Element(string n) { nodeName = n; } public Element() { } public string getAttribute(string name) { return attributes.get(name); } public void appendChild(Element e) { e.ParentNode = this; this.childNodes.add(e); } public void removeChild(Element e) { Element p = e.parentNode; List<Element> kids = new List<Element> {}; for( Element ee: e.parentNode.childNodes) { if (ee != e) kids.add(ee); } p.childNodes = kids; } // traverse below this node, returning all matching nodes by name public List<Element> getElementsByTagName(string nam) { List<Element> ret = new List<Element>{}; if (nam == this.nodeName) ret.add(this); for (Element c: this.childNodes) { ret.addAll( c.getElementsByTagName(nam) ); // decend tree } return ret; } // like above, but just returns the first one that matches public Element getElementByTagName(string nam) { List<Element> r = getElementsByTagName(nam); if (r.size() == 0) return null; return r[0]; } // first one that matches, just return the nodeValue public string getValue(string nam) { Element e = getElementByTagName(nam); return (e==null?null:e.nodeValue); } // some debugging help public void dump() { dump('');} public void dump(string pre) { // just current node system.debug( pre + ' ' +this.nodeName + '->' + this.nodeValue + ' ' + this.attributes ); } public void dumpAll() { dumpAll(''); } public void dumpAll(string pre) { // node and all children system.debug( pre + this.nodeName + '->' + this.nodeValue + ' ' + this.attributes ); for (Element c: this.childNodes) { c.dumpAll(pre+' '); } } public string toXmlString() { string ret = '<' + this.nodeName + ' '; for ( string a : attributes.keySet() ) { ret += a + '=\'' + attributes.get(a) + '\' ' ; } ret += '>'; if (nodeValue == '' ) ret += '\n'; for (Element c: this.childNodes) { ret += c.toXmlString() ;//+ '\n'; } if (nodeValue != '' ) ret += nodeValue; //else ret += '\n'; return ret + '</' + this.nodeName + '>\n'; } /* * experimental path based patern matching, sort of like xpath, * but simpler, just matches a path() string with the pattern supplied */ // * /bookstore/book/.* // /.*book/.* // /.*/book$ public List<Element> getElementsByPath(string path) { List<Element> ret = new List<Element>{}; // system.debug( path + ' ' + this.path()); if ( Pattern.matches(path, this.path()) ) ret.add(this); for (Element c: this.childNodes) ret.addAll( c.getElementsByPath(path) ); return ret; } public string path() { Element t = this; string ret = t.nodeName; while (t.parentNode != null && t.parentNode.nodeName != '#document') { t = t.parentNode; ret = t.nodeName + '/'+ret; } return '/'+ret; } // utility methods public Element firstChild() { if ( this.childNodes.size() == 0 ) return null; return this.childNodes[0]; } public string textContent() { return this.nodeValue; } public boolean hasChildNodes() { return childNodes.size()>0; } public boolean isEqualNode(Element comp) { return this.nodeName == comp.nodeName; } public Element cloneNode() { return this.clone(); } public boolean hasAttributes() { return ! attributes.isEmpty(); } public boolean isSameNode(Element comp) { return this === comp; } public Element ownerDocument() { Element ret = this; while( ret.parentNode != null) { ret = ret.parentNode; } return ret; } // properties public Element parentNode = null; // only root has a null parent public string nodeName = ''; public string nodeValue = ''; public List<Element> childNodes = new List<Element>(); public map<String,String> attributes = new map<String,String>(); } }