[JSoup][1] is pretty nice and getting popular. Here's how you could just parse the whole table:


 URL url = new URL("http://www.nseindia.com/content/equities/niftysparks.htm");
 Document doc = Jsoup.parse(url, 3000);
 
 Element table = doc.select("table[title=Avgångar:]").first();
 
 Iterator<Element> it = table.select("td").iterator();
 
 //we know the third td element is where we wanna start so we call .next twice
 it.next();
 it.next();
 while(it.hasNext()){
 // do what ever you want with the td element here

 //iterate three times to get to the next td you want. checking after the first
 // one to make sure
 // we're not at the end of the table.
 it.next();
 if(!it.hasNext()){ 
 break;
 }
 it.next();
 it.next();
 }


 [1]: http://jsoup.org/