/lib/Controller/Validator/Basic.php

<?php

class Controller_Validator_Basic extends Controller_Validator_Abstract
{
    public function init()
    {
        parent::init();

        $this->alias = array_merge(
            $this->alias,
            array(
                'same' => 'eq',
                'different' => 'ne',
            )
        );
    }

    /* \section Value Test Rules: General */

    public function rule_required($a)
    {
        if ($a === '' || $a === false || $a === null) {
            return $this->fail('must not be empty');
        }
    }

    public function rule_len($a)
    {
        if (!is_string($a)) {
            return $this->fail('must be a string');
        }

        $this->prefix = 'length of ';

        return strlen($a);
    }

    /**
     * Requires a regex pattern as the
     * next rule in the chain.
     *
     * Please give your rule a custom error message.
     */
    public function rule_regex($a)
    {
        $opt = array();
        $rule = $this->pullRule();
        if ($rule[0] != '/') {
            $rule = '/^'.$rule.'*$/';
        }
        $opt['regexp'] = $rule;

        if (!filter_var($a, FILTER_VALIDATE_REGEXP, ['options' => $opt])) {
            return $this->fail('does not match the pattern');
        }
    }

    /* \section Value Test Rules: Value Lists */

    /**
     * Checks that value is in the list provided.
     * Syntax: |in|foo,bar,foobar.
     */
    public function rule_in($a)
    {
        $vals = $this->prep_in_vals($a);
        if (!in_array($a, $vals)) {
            return $this->fail('has an invalid value');
        }
    }

    /**
     * Checks that value is not in the list provided.
     * Syntax: |not_in|foo,bar,foobar.
     */
    public function rule_not_in($a)
    {
        $vals = $this->prep_in_vals($a);
        if (in_array($a, $vals)) {
            return $this->fail('has an invalid value');
        }
    }

    /* \section Value Comparison Rules: absolute value */

    /**
     * Inclusive range check.
     *
     * Overloaded: checks value for numbers,
     * string-length for other values.
     *
     * Next 2 rules must specify the min and max
     */
    public function rule_between($a)
    {
        $min = $this->pullRule(true);
        $max = $this->pullRule(true);

        if (is_numeric($a)) {
            if ($a < $min || $a > $max) {
                return $this->fail('must be between {{arg1}} and {{arg2}}', $min, $max);
            }
        } else {
            return $this->fail('Must be a numeric value');
        }
    }

    public function rule_gt($a)
    {
        $target = $this->pullRule(true);
        if ($a <= $target) {
            return $this->fail('must be greater than {{arg1}}', $target);
        }

        return $a;
    }

    public function rule_lt($a)
    {
        $target = $this->pullRule(true);
        if ($a >= $target) {
            return $this->fail('must be less than {{arg1}}', $target);
        }

        return $a;
    }

    public function rule_gte($a)
    {
        $target = $this->pullRule(true);
        if ($a < $target) {
            return $this->fail('must be at least {{arg1}}', $target);
        }

        return $a;
    }

    public function rule_lte($a)
    {
        $target = $this->pullRule(true);
        if ($a > $target) {
            return $this->fail('must be not greater than {{arg1}}', $target);
        }

        return $a;
    }

    public function rule_eq($a)
    {
        $target = $this->pullRule();
        if ($a !== $target) {
            return $this->fail('must be exactly {{arg1}}', $target);
        }

        return $a;
    }

    public function rule_eqf($a)
    {
        $target=$this->pullRule(true);
        if ($a !== $this->get($target)) {
            return $this->fail('must be same as {{arg1}}', $target);
        }
        return $a;
    }

    /* \section Value Test Rules: Numeric */

    /**
     * Checks for int or float.
     */
    public function rule_number($a)
    {
        if (!is_numeric($a)) {
            return $this->fail('must be a number');
        }
    }

    public function rule_decimal($a)
    {
        if (!preg_match('/^[0-9]+\.[0-9]+$/', $a)) {
            return $this->fail('must be a decimal number: eg 12.34');
        }
    }

    public function rule_phone($a)
    {
        if (strlen(preg_replace('/[^0-9]/', '', $a)) < 6) {
            return $this->fail('must contain valid phone number');
        }
    }

    /**
     * Checks for a specific number of
     * decimal places.
     *
     * Arg: int- number of places
     */
    public function rule_decimal_places($a)
    {
        $places = $this->pullRule();
        $pattern = sprintf('/^[0-9]+\.[0-9]{%s}$/', $places);
        if (!preg_match($pattern, $a)) {
            return $this->fail('Must have {{arg1}} decimal places', $places);
        }
    }

    public function rule_int($a)
    {
        if (!preg_match('/^[0-9]*$/', $a)) {
            return $this->fail('Must be an integer: eg 1234');
        }
    }

    /* \section Value Test Rules: String */

    /**
     * Test for A-Za-z.
     */
    public function rule_alpha($a)
    {
        $msg = 'must contain only letters';
        if (!preg_match('/^([A-Za-z])*$/', $a)) {
            return $this->fail($msg);
        }
    }

    /**
     * Test for A-Za-z0-9.
     */
    public function rule_alphanum($a)
    {
        $msg = 'must contain only digits and letters';
        if (!preg_match('/^([a-zA-Z0-9])*$/', $a)) {
            return $this->fail($msg);
        }
    }

    public function rule_type($a)
    {
        $this->prefix = 'type of ';

        return gettype($a);
    }

    public function rule_class($a)
    {
        if (!is_object($a)) {
            return $this->fail('is not an object');
        }
        $this->prefix = 'class of ';

        return get_class($a);
    }

    /* \section Value Test Rules: Boolean */

    /**
     * Validate for true|false|t|f|1|0|yes|no|y|n.
     *
     * Normalizes to lower case
     */
    public function rule_bool($a)
    {
        // We don't use PHP inbuilt test - a bit restrictive

        // Changes PHP true/false to 1, 0
        $a = strtolower($a);

        $vals = array('true', 'false', 't', 'f', 1, 0, 'yes', 'no', 'y', 'n');

        if (!in_array($a, $vals)) {
            return $this->fail('Must be a boolean value');
        }
    }

    /**
     * Validate for true|t|1|yes|y.
     *
     * Normalizes to lower case
     *
     * Useful for 'if' rules
     */
    public function rule_true($a)
    {
        // Changes PHP true to 1
        $a = strtolower($a);

        $vals = array('true', 't', 1, 'yes', 'y');

        if (!in_array($a, $vals)) {
            return $this->fail('Must be true');
        }
    }

    /**
     * Validate for false|f|0|no|n.
     *
     * Normalizes to lower case
     *
     * Useful for 'if' rules
     */
    public function rule_false($a)
    {
        // Changes PHP false to 0
        $a = strtolower($a);

        $vals = array('false', 'f', 0, 'no', 'n');

        if (!in_array($a, $vals)) {
            return $this->fail('Must be false');
        }
    }

    /* \section Value Test Rules: Date & Time */

    /**
     * Validate for ISO date in format YYYY-MM-DD.
     *
     * Also checks for valid month and day values
     */
    public function validate_iso_date($a)
    {
        $date = explode('-', $a);

        $msg = 'Must be date in format: YYYY-MMM-DD';

        if (count($date) != 3) {
            return $this->fail($msg);
        }

        if (strlen($date[0]) !== 4 || strlen($date[1]) !== 2 || strlen($date[2]) !== 2) {
            return $this->fail($msg);
        }

        if (!@checkdate($date[1], $date[2], $date[0])) {
            return $this->fail($msg);
        }
    }

    /**
     * Validate for ISO time.
     *
     * Requires a complete hour:minute:second time with
     * the optional ':' separators.
     *
     * Checks for hh:mm[[:ss][.**..] where * = microseconds
     * Also checks for valid # of hours, mins, secs
     */
    public function rule_iso_time($a)
    {
        $pattern = "/^([0-9]{2}):([0-9]{2})(?::([0-9]{2})(?:(?:\.[0-9]{1,}))?)?$/";
        $msg = 'Must be a valid ISO time';

        if (preg_match($pattern, $a, $matches)) {
            if ($matches[1] > 24) {
                return $this->fail($msg);
            }

            if ($matches[2] > 59) {
                return $this->fail($msg);
            }

            if (isset($matches[3]) && $matches[3] > 59) {
                return $this->fail($msg);
            }
        } else {
            return $this->fail($msg);
        }
    }

    /**
     * Validate ISO datetime in the format:.
     *
     * YYYY-MM-DD hh:mm:ss with optional microseconds
     */
    public function rule_iso_datetime($a)
    {
        $parts = explode(' ', $a);
        $msg = 'Must be a valid ISO datetime';

        if (count($parts) != 2) {
            return $this->fail($msg);
        }

        try {
            $this->rule_iso_date($parts[0]);
        } catch (Exception $e) {
            return $this->fail($msg);
        }
        try {
            $this->rule_iso_time($parts[1]);
        } catch (Exception $e) {
            return $this->fail($msg);
        }
    }

    /**
     * Checks any PHP datetime format:
     * http://www.php.net/manual/en/datetime.formats.date.php.
     */
    public function rule_before($a)
    {
        $time = $this->pullRule();
        if (strtotime($a) >= strtotime($time)) {
            return $this->fail('Must be before {{arg1}}', $time);
        }
    }

    /**
     * Checks any PHP datetime format:
     * http://www.php.net/manual/en/datetime.formats.date.php.
     */
    public function rule_after($a)
    {
        $time = $this->pullRule();
        if (strtotime($a) <= strtotime($time)) {
            return $this->fail('Must be after {{arg1}}', $time);
        }
    }

    /* \section Value Test Rules: Postal, Email & Credit Card */

    public function rule_email($a)
    {
        if (!filter_var($a, FILTER_VALIDATE_EMAIL)) {
            return $this->fail('Must be a valid email address');
        }
    }

    /**
     * Checks for a 5 digit or extended US zip.
     */
    public function rule_zip($a)
    {
        if (!preg_match('/^\d{5}(-\d{4})?$/', $a)) {
            return $this->fail('Must be a valid ZIP code');
        }
    }

    /**
     * Validate for credit card number.
     *
     * Uses the Luhn Mod 10 check
     */
    public function rule_credit_card($a)
    {
        // Card formats keep changing and there is too high a risk
        // of false negatives if we get clever. So we just check it
        // with the Luhn Mod 10 formula

        // Calculate the Luhn check number

        $msg = 'Not a valid card number';

        $sum = 0;
        $alt = false;

        for ($i = strlen($a) - 1; $i >= 0; --$i) {
            $n = substr($a, $i, 1);
            if ($alt) {
                //square n
                $n *= 2;
                if ($n > 9) {
                    //calculate remainder
                    $n = ($n % 10) + 1;
                }
            }
            $sum += $n;
            $alt = !$alt;
        }

        // If $sum divides exactly by 10 it's valid

        if (!($sum % 10 == 0)) {
            return $this->fail($msg);
        } else {
            // Luhn check seems to return true for any string of 0s

            $stripped = str_replace('0', '', $a);

            if (strlen($stripped) == 0) {
                return $this->fail($msg);
            }
        }
    }

    /**
     * Validate a card "expires end" date.
     */
    public function rule_card_to_date($a)
    {
        $msg = 'Not a valid date';
        if (!$this->card_date_parser($a, 'to')) {
            return $this->fail($msg);
        }
    }

    /**
     * Validate a card "valid from" date.
     */
    public function rule_card_from_date($a)
    {
        $msg = 'Not a valid date';
        if (!$this->card_date_parser($a, 'from')) {
            return $this->fail($msg);
        }
    }

    /* \section Value Conversion Rules: General */

    /**
     * Strips out the characters matched by the
     * pattern in the argument.
     *
     * The pattern should include the pattern delimiter, eg:
     *
     * /my_pattern/
     */
    public function to_strip_regex($a)
    {
        $pattern = $this->pullRule();

        return preg_replace($pattern, '', $a);
    }

    /* \section Value Conversion Rules: Numeric */

    public function rule_to_int($a)
    {
        return (int) $a = preg_replace('/[^0-9]/', '', $a);
    }

    public function rule_to_number($a)
    {
        return (int) $a = preg_replace('/[^0-9\.]/', '', $a);
    }

    public function rule_to_float($a)
    {
        return (int) $a = preg_replace('/[^0-9\.]/', '', $a);
    }

    public function rule_to_digits_and_single_spaces($a)
    {
        $a = preg_replace("/[^\d ]/", '', $a);

        return $this->rule_strip_extra_space($a);
    }

    /* \section Value Conversion Rules: String */

    public function rule_to_trim($a)
    {
        return trim($a);
    }

    public function rule_to_ltrim($a)
    {
        return ltrim($a);
    }

    public function rule_to_rtrim($a)
    {
        return rtrim($a);
    }

    /**
     * Strip out all white space.
     */
    public function rule_to_strip_space($a)
    {
        return preg_replace("/\s/", '', $a);
    }

    /**
     * Reduce sequential whitespaces to a single space.
     */
    public function rule_to_strip_extra_space($a)
    {
        return $this->strip_excess_whitespace($a);
    }

    /**
     * Strip to A-Za-z0-9.
     */
    public function rule_to_alpha($a)
    {
        return preg_replace('/[^a-zA-Z]/', '', $a);
    }

    /**
     * Test for unicode letter characters.
     *
     * Should work even is PCRE compiled
     * without "--enable-unicode-properties".
     */
    public function rule_to_alpha_unicode($a)
    {
        return preg_replace('/(*UTF8)[^\p{L}]/u', '', $a);
    }

    public function rule_to_alpha_num($a)
    {
        return preg_replace('/[^a-zA-Z0-9]/', '', $a);
    }

    public function rule_to_lower($a)
    {
        return  strtolower($a);
    }

    public function rule_to_upper($a)
    {
        return $this->mb_str_to_upper($a);
    }

    public function rule_to_upper_words($a)
    {
        return $this->mb_str_to_upper_words($a);
    }

    /* \section Value Conversion Rules: Alphanumeric */

    /**
     * Strip to unicode letter characters and 0-9.
     *
     * Should work even is PCRE compiled
     * without "--enable-unicode-properties".
     */
    public function rule_to_alpha_num_unicode($a)
    {
        return preg_replace('/(*UTF8)[^\p{L}0-9]/u', '', $a);
    }

    public function rule_to_alpha_num_dash($a)
    {
        return preg_replace('/[^a-zA-Z0-9_-]/', '', $a);
    }

    /**
     * Truncates, and adds '...' to
     * end of string.
     *
     * Requires parameter 'length'
     */
    public function rule_to_truncate($a)
    {
        $len = $this->pullRule();

        return $this->mb_truncate($a, $len);
    }

    /**
     * Requires parameters: length,
     * custom string to end of string.
     */
    public function rule_to_truncate_custom($a)
    {
        $len = $this->pullRule();
        $append = $this->pullRule();

        return $this->mb_truncate($a, $len, $append);
    }

    /**
     * Strip to unicode letter characters and 0-9, -, _.
     *
     * Requires PCRE compiled with  "--enable-unicode-properties".
     * Most distros these days will offer this
     */
    public function rule_to_alpha_num_dash_unicode($a)
    {
        return preg_replace('/(*UTF8)[^\p{L}0-9_-]/u', '', $a);
    }

    /* \section Value Conversion Rules: Date & Time */

    public function rule_to_iso_date($a)
    {
        $a = preg_replace("/[^T0-9\/\-\(\): ]/", '', $a);

        return $this->rule_iso_date($a);
    }

    /* \section Value Conversion Rules: Data Sanitization */

    public function rule_to_strip_tags($a)
    {
        return strip_tags($a);
    }

    public function rule_to_quote_meta($a)
    {
        return quotemeta($a);
    }

    public function rule_to_add_slashes($a)
    {
        return addslashes($a);
    }

    public function rule_to_strip_slashes($a)
    {
        return stripslashes($a);
    }

    /**
     * Strip out attack characters from names & addresses
     * and other strings where they have no place.
     *
     * Strips: * ^ <> ? ! () | / \ [] + = % ; ~ `
     */
    public function rule_to_strip_nasties($a)
    {
        return preg_replace("|[\*\^<?>!\"\(\)\|\\\\/\[\]\+=#%;~`]|", '', $a);
    }

    /* \section Value Conversion Rules: Phone, Name, Address */

    /**
     * Normalizes, then validates.
     */
    public function rule_to_zip($a)
    {
        // Change 12345 1234 to 12345-1234
        $a = preg_replace('/[^0-9-]/', '', $a);
        $this->rule_zip($a);
    }

    // *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-
    // PERSONAL NAMES (European style)
    // *-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-

    /**
     * Useful for cleaning up input where you don't want to present
     * an error to the user - eg a checkout where ease of use is
     * more important than accuracy. Can save a lot of re-keying.
     *
     * Some libraries don't clean up names if already in mixed case.
     * Experience shows this isn't very useful, as many users
     * will type mAry, JOSepH etc.
     *
     * Can only be a best guess - but much better than nothing:
     * has been in production for years without any negative
     * customer feedback.
     *
     * Set $is_capitalise_prefix if you want prefixes in upper:
     * Von Trapp vs von Trapp. You would do this to format
     * a last name for use in salutations:
     *
     * Dear Mr Von Trapp
     *
     * Can handle full names, 1st only, middle only, last only.
     *
     * Cleans up extra whitespace.
     */
    public function rule_to_name($name, $is_capitalise_prefix = false)
    {
        /*
        A name can have up to 5 components, space delimited:

        Worst case:
            salutation | forenames    | prefix(es) | main name   | suffix
            Ms         | Jo-Sue Ellen | de la      | Mer-Savarin | III

        Rules for forenames

        1) Capitalise 1st char and after a hyphen.

        Rules for special case prefixes: von, de etc

        1) Set capitalisation at runtime.

            There seem to be no fixed rules, but
            lower case is commonly used as part of a
            whole name:

            John von Trapp

            While it is normally capitalised as part of
            a salutation:

            Dear Mr Von Trapp

            By default we store in the lower case form.
            Set the param $is_capitalise_prefix to true
            to capitalise.

        2) In default mode, St is capitalised,
            other prefixes lower cased. We retain user's
            choice of punctuation with St./St

        Rules for main name:

        1) Capitalise after a hyphen in the main name:

            Smythington-Fenwick

        2) Capitalise after Mc at start of name -
            this is pretty much a universal rule

            MCDONALD => McDonald

        3) Unless user has capitalised after the M,
            do NOT capitalise after Mac at start of name:

            - Many Scottish Mac names
                are not capitalised: Macaulay, Macdonald

            - Many non-Scottish names start with Mac:
                eg Macon - we want to avoid "MacOn";

            - The Cpan name modules and some style manuals
                force MacDonald, but this seems to
                create more problems than it solves.

            macrae => Macrae

        4) Capitalise after O'

            o'grady => O'Grady
        */

        // If name string is empty, bail out

        if (empty($name)) {
            return '';
        }

        // Setup special case prefix lookup list.
        // These are prefixes that are not capitalised.

        // Prefixes which are capitalised such as "St"
        // can be omitted.

        // We omit prefixes that are also common names,
        // such as "Della", "Di" and "Ben"

        $prefixes = array(
            'ap' => array('upper' => 'Ap', 'lower' => 'ap'),
            'da' => array('upper' => 'Da', 'lower' => 'da'),
            'de' => array('upper' => 'De', 'lower' => 'de'),
            'del' => array('upper' => 'Del', 'lower' => 'del'),
            'der' => array('upper' => 'Der', 'lower' => 'der'),
            'du' => array('upper' => 'Du', 'lower' => 'du'),
            'la' => array('upper' => 'La', 'lower' => 'la'),
            'le' => array('upper' => 'Le', 'lower' => 'le'),
            'lo' => array('upper' => 'Lo', 'lower' => 'lo'),
            'van' => array('upper' => 'Van', 'lower' => 'van'),
            'von' => array('upper' => 'Von', 'lower' => 'von'),
           );

        // Set up suffix lookup list

        // We preserve user's preferred punctuation: Sr./Sr

        $suffixes = array(
                        'i' => 'I',
                        'ii' => 'II',
                        'iii' => 'III',
                        'iv' => 'IV',
                        'v' => 'V',
                        'vi' => 'VI',
                        'vii' => 'VII',
                        'viii' => 'VIII',
                        'ix' => 'IX',
                        'x' => 'X',
                        'jr.' => 'Jr.',
                        'jr' => 'Jr',
                        'jnr.' => 'Jnr.',
                        'jnr' => 'Jnr',
                        'sr.' => 'Sr.',
                        'sr' => 'Sr',
                        'snr.' => 'Snr.',
                        'snr' => 'Snr',
                        '1st' => '1st',
                        '2nd' => '2nd',
                        '3rd' => '3rd',
                        '4th' => '4th',
                        '5th' => '5th',
                        '6th' => '6th',
                        '7th' => '7th',
                        '8th' => '8th',
                        '9th' => '9th',
                        '10th' => '10th',
                        '1st.' => '1st.',
                        '2nd.' => '2nd.',
                        '3rd.' => '3rd.',
                        '4th.' => '4th.',
                        '5th.' => '5th.',
                        '6th.' => '6th.',
                        '7th.' => '7th.',
                        '8th.' => '8th.',
                        '9th.' => '9th.',
                        '10th.' => '10th.',
                       );

        // Clean out extra whitespace

        $name = $this->strip_excess_whitespace(trim($name));

        // Try to parse into forenames, main name, suffix

        $parts = explode(' ', $name);

        if (count($parts) == 1) {
            // Must be the main name

            $name_main = array_pop($parts);
            $name_fname = false;
            $name_suffix = false;
        } else {
            // We have more than one part to parse

            // Is the last part a suffix?
            // We assume name can have only one suffix

            $part = array_pop($parts);
            $normalised_part = strtolower($part);

            if (array_key_exists($normalised_part, $suffixes)) {
                // Last part is a suffix

                $name_main = array_pop($parts);
                $name_suffix = $suffixes[$normalised_part];
            } else {
                // Last part is the main name

                $name_main = $part;
                $name_suffix = false;
            }
        }

        // Anything left is a salutation, initial or forname

        if (count($parts) > 0) {
            $name_fnames = $parts;
        } else {
            $name_fnames = false;
        }

        // We build the name from first to last:

        $new_name = array();

        // Set case for the forenames

        if ($name_fnames) {
            foreach ($name_fnames as $fname) {
                $parts = array();
                $fname = strtolower($fname);

                // Do hypenated parts separately

                $exploded_fname = explode('-', $fname);

                foreach ($exploded_fname as $part) {
                    // If it is one of our special case prefixes
                    // we use the appropriate value
                    // Else, we capitalise

                    if (array_key_exists($part, $prefixes)) {
                        if ($is_capitalise_prefix !== false) {
                            $parts[] = $prefixes[$part]['upper'];
                        } else {
                            $parts[] = $prefixes[$part]['lower'];
                        }
                    } else {
                        // It is a normal forename, salutation or initial
                        // We capitalise it.

                        $parts[] = ucfirst($part);
                    }
                }

                $new_name[] = implode('-', $parts);
            }
        }

        // Set case for the main name

        $name_main_original = $name_main;
        $name_main = strtolower($name_main);

        // Do hypenated parts separately

        $exploded_main_original = explode('-', $name_main_original);
        $exploded_main = explode('-', $name_main);

        $parts = array();

        foreach ($exploded_main as $key => $part) {
            $part_original = $exploded_main_original[$key];

            if (substr($part, 0, 2) == 'mc') {
                // Do "Mc"

                // Uppercase the 3rd character

                $a = substr($part, 2);
                $parts[] = 'Mc'.ucfirst($a);
            } elseif (substr($part, 0, 3) == 'mac') {
                // Do "Mac"

                // Lowercase the 3rd character
                // unless user has submitted
                // a correct looking name

                if (preg_match('|^Mac[A-Z][a-z]*$|', $part_original)) {
                    $parts[] = $part_original;
                } else {
                    $parts[] = ucfirst($part);
                }
            } elseif (substr($part, 0, 2) == "o'") {
                // Do O'
                // Uppercase the 3rd character

                $a = substr($part, 2);
                $parts[] = "O'".ucwords($a);
            } else {
                // It is a plain-jane name

                $parts[] = ucfirst($part);
            }
        }

        $new_name[] = implode('-', $parts);

        if ($name_suffix !== false) {
            $new_name[] = $name_suffix;
        }

        // Assemble the new name

        $output = implode(' ', $new_name);

        return $output;
    }

    /* \section Helper Functions */

    /**
     * Reduce sequential whitespaces to a single space.
     */
    protected function strip_excess_whitespace($a)
    {
        return preg_replace('/\s\s+/', ' ', $a);
    }

    /**
     * Helper for validating card to and from dates.
     */
    protected function card_date_parser($a, $type)
    {
        // Strip out any slash

        $date = str_replace('/', '', $a);

        // Check that we have 4 digits

        if (!preg_match('|^[0-9]{4}$|', $date)) {
            return false;
        }

        $month = substr($date, 0, 2);
        $year = substr($date, 2, 2);

        // Check month is logical

        if ($month > 12) {
            return false;
        }

        $parts = array(date('Y'), date('m'), 1);
        $now_datetime = new DateTime(implode('-', $parts));

        $parts = array('20'.$year, $month, '1');
        $card_datetime = new DateTime(implode('-', $parts));

        $interval = $now_datetime->diff($card_datetime);
        $days = $interval->format('%R%a days');

        if ($type == 'from') {
            // Check from date is older or equal to current month

            if ($days <= 0 && $days > -3650) {
                return true;
            } else {
                return false;
            }
        } elseif ($type == 'to') {
            // Check to date is newer or equal to current month

            if ($days >= 0 && $days < 3650) {
                return true;
            } else {
                return false;
            }
        } else {
            $msg = "Bad date type '$type' in card-date validation";
            throw new Error($msg);
        }
    }

    /**
     * Explode and trim a comma-delmited list
     * for the in and not_in rules.
     */
    protected function prep_in_vals($a)
    {
        $vals = $this->pullRule();
        if (is_array($vals)) {
            return $vals;
        }
        $vals = explode(',', $vals);
        array_walk($vals, function ($val) {
            return trim($val);
        });
       // create_function('&$val', '$val = trim($val);'));
        return $vals;
    }
}