/*
     * checks two urls and determines if they are more than $min_similarity% similar. requires bash access, a decent OS, w3m and wdiff
     * @param $url first url
     * @param $other_url other url
     * @param $min_similarity an integer between 0 and 100, as in percentage
    */
    function urlEqualsUrl($url, $other_url, $min_similarity = 80) {

        $dir = "/tmp/equ_".substr(uniqid(),6);
        mkdir($dir);

        $cmd = "w3m -dump \"$url\" 2>/dev/null > $dir/1.html;
        w3m -dump \"$other_url\" 2>/dev/null > $dir/2.html;
        wdiff -nis $dir/1.html $dir/2.html | tail -2 | awk '{print $5}'";

        $percentages = explode(PHP_EOL, `$cmd`);

        $percentage = (substr($percentages[0], 0, -1) + substr($percentages[1], 0, -1)) / 2;

        return $percentage > $min_similarity;

    }