gistfile1.java

package bagcheck.util;

import java.util.Stack;
import java.util.regex.Matcher;
import java.util.regex.Pattern;

/**
 * Clean up Text
 * <p/>
 * User: sam
 * Date: Aug 5, 2010
 * Time: 9:25:02 AM
 */
public class Text {

  private static Pattern matchTags = Pattern.compile("<(/?[^><]+?)>|<([^><]+?)");
  private static Pattern allowedTags = Pattern.compile("/?(br|p|a|b|i|ol|li|ul|blockquote)");
  private static Pattern allowedATag = Pattern.compile("a\\s+href=\"([^\"]+?)\"");
  private static Pattern allowedHref = Pattern.compile("https?://[a-zA-Z0-9-_./]+");

  public static String strip(String input) {
    Stack<String> tags = new Stack<String>();
    input = input.replace("&", "&amp;");
    StringBuffer sb = new StringBuffer();
    // Match what we allow
    Matcher m1 = matchTags.matcher(input);
    while (m1.find()) {
      String tag = m1.group(1);
      if (tag == null) {
        m1.appendReplacement(sb, "&lt;" + m1.group(2));
      } else if (allowedTags.matcher(tag).matches()) {
        if (tag.startsWith("/")) {
          if (tags.size() > 0 && tags.pop().equals(tag.substring(1))) {
          } else {
            m1.appendReplacement(sb, "");
            continue;
          }
        } else {
          if (!tag.equals("br")) tags.push(tag);
        }
        m1.appendReplacement(sb, "<" + tag + ">");
      } else {
        Matcher m2 = allowedATag.matcher(tag);
        if (m2.matches() && allowedHref.matcher(m2.group(1)).matches()) {
          m1.appendReplacement(sb, "<" + tag + " rel=\"nofollow\">");
          tags.add("a");
        } else {
          m1.appendReplacement(sb, "");
        }
      }
    }
    m1.appendTail(sb);
    return sb.toString().replace("\n", "<br>");
  }

  public static void main(String[] args) {
    System.out.println(strip("<a href=\"http://lessthan10.com\">1 < 10</a>"));
  }
}