/** Let's practice tokenizing a string.
 *  If you want to tokenize simply by spaces, then you can use the Scanner
 *    class and use next() repeatedly.
 *  But if you want to tokenize on a punctuation symbol, then you can
 *    use the String class' split() function.
 *  If you include \\s* on either side of the punctuation symbol when you call
 *    split, then we also trim spaces from the tokens.
 *  \\s* means 0 or more whitespace characters
 *  \\s+ means 1 or more whitespace characters
 *    For example, splitting on ",\\s+" would handle commas followed by
 *    1 or more spaces.  But it wouldn't separate:  4,5 
 *  If you simply want to tokenize based on all punctuation and space
 *    characters use the pre-defined sets {Punct} and {Space}.
 *    If you use both the Punct and Space sets without the \\s* on both sides,
 *    you get a lot of empty tokens.  The \\s* on both sides trims spaces.
 *  For detailed information on what delimiter-string to pass to split,
 *    see the online documentation on the "Pattern" class.
 */

public class Driver
{
  public static void main(String [] args)
  {
    String line = "dmchugh   ftp          davidm           Fri Oct 12 11:07 - down  (4+03:36)";
    String pattern = "\\s*[\\p{Punct}\\p{Space}]\\s*";
    showTokens(line, pattern);

    line = "4,5, 6,7";
    pattern = ",\\s+";
    showTokens(line, pattern);

    line = "moo-goo--gai-pan";
    pattern = "--";
    showTokens(line, pattern);
  }

  public static void showTokens(String input, String pattern)
  {
    // This is the most important step:  We need to call the split
    // function and put the result in an array of Strings.
    String [] token = input.split(pattern);

    // Print the tokens
    System.out.printf("\nHere are your tokens:\n");
    for (int i = 0; i < token.length; ++i)
      System.out.printf("Token #%d:  -----%s-----\n", i+1, token[i]);
  }
}