import org.w3c.dom.Attr;
import org.w3c.dom.Document;
import org.w3c.dom.NamedNodeMap;
import org.w3c.dom.Node;
import org.w3c.dom.NodeList;
import org.w3c.tidy.Tidy;

import java.io.*;
import java.net.*;

class TidyExample {
  public static void main(String[] args) throws IOException {
    if(args.length!=1) {
      System.out.println("This program requires a URL as parameter.");
      return;
    }
    
    // The following lines creates an instance of jtidy and prevents it 
    // from writing most of its error and warning messages. The messages that 
    // cannot be prevented are redirected to the file "parselog"
    Tidy t=new Tidy();    
    t.setMakeClean(true);
    t.setQuiet(true);
    t.setOnlyErrors(true);
    t.setShowWarnings(false);  
    t.setErrout(new PrintWriter(new FileWriter("parselog")));     

    // Download and parse:
    URL u=new URL(args[0]);
    URLConnection uc=u.openConnection();
    Document d=t.parseDOM(uc.getInputStream(), null);
    
    // Get the root node of the tree:
    Node root=d;
    
    // Get the body node in an ugly way:
    Node html=root.getLastChild();
    Node body=html.getLastChild();

    // Print the contents of the body node:
    NodeList children = body.getChildNodes();
    if(children==null) return;
    
    int len = children.getLength();
    for(int i=0; i<len; i++ ) {
      // Get this child:
      Node node=children.item(i);
      
      // And print it:
      int type = node.getNodeType();
      switch(type) {
        case Node.TEXT_NODE:
          System.out.println("Text:"+node.getNodeValue());
          break;
        case Node.ELEMENT_NODE:
          System.out.println("Node:"+node.getNodeName());
          NamedNodeMap attrs = node.getAttributes();
          for(int j=0; j<attrs.getLength(); j++) {
            System.out.println("  Attr name:"+attrs.item(j).getNodeName());
            System.out.println("  Attr value:"+attrs.item(j).getNodeValue());
          }	 
          break;
      }    
    }
  }
}