A simple scanner that uses regular expressions to parse strings

scanner.php

<?php

class Scanner
{
    private $_scans;
    private $_idx;

    public function __construct(array $scans)
    {
        $this->_scans = array();
        foreach ($scans as $regex => $callback) {
            $regex = '/(^' . $regex . ')/';
            $this->_scans[$regex] = $callback;
        }
    }

    public function scan($str)
    {
        $idx = 0;
        $cont = true;
        do {
            $cont = false;
            foreach ($this->_scans as $regex => $callback) {
                $matches = array();
                if (preg_match($regex, substr($str, $idx), $matches)) {
                    $idx += strlen($matches[0]);
                    if (is_callable($callback)) {
                        call_user_func($callback, $this, $matches[0]);
                    }
                    $cont = true;
                    break;
                }
            }
        } while ($cont);

        // Was this successfully parsed?
        return ($idx === strlen($str));
    }
}

// if __name__ == "__main__":
if (!debug_backtrace()) {

    class Tokenizer
    {
        public $tokens;

        public function __construct()
        {
            $this->tokens = array();
        }

        public function identifier(Scanner $scanner, $token)
        {
            $this->tokens[] = "(IDENT, $token)";
        }

        public function operator(Scanner $scanner, $token)
        {
            $this->tokens[] = "(OPERATOR, $token)";
        }

        public function digit(Scanner $scanner, $token)
        {
            $this->tokens[] = "(DIGIT, $token)";
        }

        public function endStmnt(Scanner $scanner, $token)
        {
            $this->tokens[] = 'END_STATEMENT';
        }
    }

    $tokenizer = new Tokenizer;

    $scanner = new Scanner(
        array(
            '[a-zA-Z_]\w*' => array($tokenizer, 'identifier'),
            '[+\-*=]' => array($tokenizer, 'operator'),
            '[0-9]+(\.[0-9]+)?' => array($tokenizer, 'digit'),
            ';' => array($tokenizer, 'endStmnt'),
            '\s+' => null
        )
    );
    $res = $scanner->scan('foo = 5.5 * 30; bar = bar - 60;');

    if ($res) { echo "TRUE\n"; } else { echo "FALSE\n"; }
    echo join("\n", $tokenizer->tokens);
}