Java and regular expressions

[kosmos890]

I want to get the favicon of a web page using jsoup.

 Document doc = null;
 String str="http://www.java-forums.org";

        try {
            doc = Jsoup.connect(str).get();

        } catch (IOException ex) {
            Exceptions.printStackTrace(ex);
        }
   
   String urlToIcon=doc.select("link[rel~=(?i)(shortcut)?(\\\\s)+icon]").attr("abs:href"));

The problem is the regular expression (probably whitespaces \\\\s).
But I test the regex with this code and the regex works.

String attributeValues="icon shortcut icon ICON SHORTCUT ICON";
        
   String regex="(?i)(shortcut)?(\\\\s)+icon";
   String[ ] split = attributesValues.split(regex);
        
   for(String s:split){
       System.out.println(s);
    }

I test the regex here and the regex works also.

[KevinWorkman]

So, what's the problem?

[kosmos890]

From an html document, I want to extract tags like
<link rel="shortcut icon" href="/favicon.ico">
<link rel="icon" href="/favicon.ico">
<link rel="ICON" href="/favicon.ico">
but with my regex I can't.
I am beginner and I can't understand why my regex doesn't work.

If I simplify my regex to (?i)icon then I lose tags like
<link rel="shortcut icon" href="/favicon.ico">

If I change my regex to (?i).*icon then I get tags like
<link rel="apple-touch-icon" href="icon.png">
I don't want these tags.

[DarrylBurke]

You want to parse HTML, use a HTML parser. Not regex.

db