SEX.java

import java.io.IOException;
import java.io.StringReader;
import java.util.Iterator;
import java.util.Map;
import java.util.Stack;
import java.util.TreeMap;

import org.xml.sax.Attributes;
import org.xml.sax.InputSource;
import org.xml.sax.SAXException;
import org.xml.sax.XMLReader;
import org.xml.sax.helpers.DefaultHandler;
import org.xml.sax.helpers.XMLReaderFactory;

/**
 * Simple Expediant XML parser
 * 
 * @author Josh Rehman <[email protected]>
 * @date 10/13/2006
 */
public class SEX {

	/**
	 * Take a small chunk of XML and parse it into nested hashtables Do not
	 * support attributes, entities, or namespaces. Do not support any
	 * non-string types (validation is assumed to occur later)
	 */
	static String xml = 
			"<customer>\n"
				+ "\t<name>Josh</name>\n"
				+ "\t<email>[email protected]</email>\n"
				+ "\t<address>"
					+ "\t\t<street>123 Main Street</street>\n"
					+ "\t\t<city>Seal Beach</city>\n"
					+ "\t\t<state>CA</state>\n"
					+ "\t\t<region><country>USA</country></region>\n"
				+ "\t</address>\n"
			+ "</customer>";

	public static void main(String[] args) {
		Map result = parse(xml);
		printMap(result, 0);
	}
	
	static void printMap(Map map, int depth){
		Iterator it = map.keySet().iterator();
		while (it.hasNext()) {
			Object key = it.next();
			Object value = map.get(key);
			if (value instanceof String) {
				System.out.println(indent(depth) + key + ":" + value);
			} else if (value instanceof Map) {
				System.out.println(indent(depth) + key + ":");
				printMap((Map)value, depth + 1);
			}
		}
	}
	
	private static String indent(int depth){
		StringBuffer buf = new StringBuffer();
		buf.append(depth);
		for (int j = 0; j < depth; j++){buf.append("\t");}
		return buf.toString();
	}

	/**
	 * Parse an xml string into some nested Maps.
	 * 
	 * @param xml - a valid xml document in String form
	 * @return a nested map representing the document
	 * @see http://www.saxproject.org/quickstart.html
	 */
	static Map parse(String xml) {
		final Map result = new TreeMap();
		try {
			XMLReader xr = XMLReaderFactory.createXMLReader();
			/**
			 * Parse a very simple XML subset into nested Maps
			 * In particular we don't support attributes or mixed content (tags and text)
			 * All tag content is trim()ed.
			 * An interesting application of recursion
			 */
			xr.setContentHandler( new DefaultHandler(){
				String key, value; //needed especially for children.
				Stack stack = new Stack(); //stack of maps
				boolean isLeaf = false;
				public void startDocument() throws SAXException {
					stack.push(result);
				}
				// create a new entry in the current map, and push it on the stack
				public void startElement(String uri, String name, String qName, Attributes atts) throws SAXException {
					isLeaf = true;
					key = qName;
					Map newMap = new TreeMap();
				    Map parentMap = (Map)stack.peek();
					parentMap.put(key, newMap); 
					stack.push(newMap);
				}
				public void endElement(String uri, String name, String qName) throws SAXException {
					if (isLeaf) {
						// we need to pop then peek then put!
						// question for the class: why can't we pop twice?
						// it messes up the stack for succeeding siblings.
						Map newMap = (Map) stack.pop();
						Map parentMap = (Map) stack.peek();
						parentMap.put(key, value); // replaces newMap with string
					}
					isLeaf = false; //end elt methods called in sequence are not leaves
				}
				// just squirl this value away;
				public void characters(char[] ch, int start, int length) throws SAXException {
					value = new String(ch, start, length).trim();
				}
			});
			xr.parse(new InputSource(new StringReader(xml))); //do it
		} catch (SAXException e) {
			e.printStackTrace();
		} catch (IOException e) {
			e.printStackTrace();
		}
		
//		result.put("name", "Josh");
//		result.put("email", "[email protected]");
//		
//		Map address = new TreeMap();
//		address.put("street", "123 Main Street");
//		address.put("city", "Seal Beach");
//		address.put("state", "CA");
//		
//		result.put("address", address);
		return result;
	}
}