gistfile1.txt

<?php

function tokenize($message) {
    preg_match_all("#[@\#][\w]+|http://[^\s]+|[:;][^\s]*[\(\)D\]\[\{\}pP]|[\w'\-]+#", $message, $matches);
    
    return $matches;
}

$messages = array(
    "@leemcveigh Yup, they're not pretty.",
    "@RihannaMaQueen in november ;D caaant waaaaait , my idol , your just amazing & so inspirational &lt;3",
    "convinced my mum to drive me to derby so I don't have to spend the night on my own there :') PHEW!",
    "@littlefoxrocks Whose digs, Em??? HA Ha",
    "#shoreditchtriangle which is a Nice place to go for lunch ? :s",
    "@festivalbicycle I loved the Fesitval of the Bicycle yesterday: great atmosphere. My son now wants to join Cycling Club Hackney: result!!"
);
foreach ($messages as $message) {
    print $message . "\n";
    print_r(tokenize($message));
}