package projet;
 
import java.io.IOException;
import java.io.InputStreamReader;
import java.io.Reader;
import java.net.URL;
import java.net.URLConnection;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.StringTokenizer;
import javax.swing.text.BadLocationException;
import javax.swing.text.EditorKit;
import javax.swing.text.SimpleAttributeSet;
import javax.swing.text.html.HTML;
import javax.swing.text.html.HTMLDocument;
import javax.swing.text.html.HTMLEditorKit;
 
public class robot {
@SuppressWarnings("unchecked" )
HashMap<String, ArrayList> Index ;
@SuppressWarnings("unchecked" )
public robot () {
Index = new HashMap<String, ArrayList> ();
}
@SuppressWarnings("unchecked" )
public HashMap<String, ArrayList> Link(String adresse) throws IOException, BadLocationException{
//Charger la page
URL url = new URL(adresse);
URLConnection uconnection = url.openConnection();
Reader rd = new InputStreamReader(uconnection.getInputStream());
//lire le document HTML
EditorKit kit = new HTMLEditorKit();
HTMLDocument doc = (HTMLDocument) kit.createDefaultDocument();
doc.putProperty("IgnoreCharsetDirective", new Boolean(true));
kit.read(rd, doc, 0);
//Parcourir la balise lien
HTMLDocument.Iterator it = doc.getIterator(HTML.Tag.A);
while (it.isValid()) {
SimpleAttributeSet s = (SimpleAttributeSet) it.getAttributes();
String link = (String) s.getAttribute(HTML.Attribute.HREF);
if (link != null && link.startsWith("http" ) && link.endsWith(".html" )) {
String Nom=(String) s.getAttribute(HTML.Attribute.TITLE);
//découpage des mots du titre du lien corespondant
if(Nom!=null){
StringTokenizer tmp = new StringTokenizer (Nom," " );
ArrayList<String> tmp2 = new ArrayList<String> () ;
 
while (tmp.hasMoreTokens()){
String tmp3=tmp.nextToken();
if(tmp3.length()>2)
tmp2.add(tmp3);
 
}
Index.put(link, tmp2);
}
}
it.next();
}
return Index;
}
 
}

1. String findLink(String myTag) {
2.         for (Map.Entry<String, ArrayList> entry : Index.entrySet()) {
3.              if(entry.getValue().contrains(myTag)) {
4.                    return entry.getKey();
5.              } 
6.        }
7. }