Question

I’m working on a set of validation classes and am currently building plugins for applying various validation rules. I’ve built the following class for validating a UK postcode:

class PostcodeUk extends abstr\Prop implements iface\Prop
{
    const 

        /**
         * Defines the regular expression against which to test postal code
         * 
         * @see http://en.wikipedia.org/wiki/Postcodes_in_the_United_Kingdom#Validation UK postal code validation rules on Wikipedia 
         */
        PATTERN = '/^(GIR 0AA)|(((A[BL]|B[ABDHLNRSTX]?|C[ABFHMORTVW]|D[ADEGHLNTY]|E[HNX]?|F[KY]|G[LUY]?|H[ADGPRSUX]|I[GMPV]|JE|K[ATWY]|L[ADELNSU]?|M[EKL]?|N[EGNPRW]?|O[LX]|P[AEHLOR]|R[GHM]|S[AEGKLMNOPRSTY]?|T[ADFNQRSW]|UB|W[ADFNRSV]|YO|ZE)[1-9]?[0-9]|((E|N|NW|SE|SW|W)1|EC[1-4]|WC[12])[A-HJKMNPR-Y]|(SW|W)([2-9]|[1-9][0-9])|EC[1-9][0-9]) [0-9][ABD-HJLNP-UW-Z]{2})$/';

    /**
     *
     * @return bool True if valid
     * @throws \InvalidArgumentException 
     */
    public function isValid ()
    {
        $valid  = false;
        $data   = $this -> getData ();

        switch (gettype ($data))
        {
            case 'NULL'     :
                $valid  = true;
            break;
            case 'string'   :
                $valid  = preg_match (static::PATTERN, $data) > 0;
            break;
            default         :
                throw new \InvalidArgumentException (__CLASS__ . ': This property cannot be applied to data of type ' . gettype ($data));
            break;
        }

        return ($valid);
    }
}

The regex defined in PostcodeUk::PATTERN was derived from the on given in Wikipedia’s article on UK postcodes. However, the regex as given detects valid postcode strings contained within bigger blocks of text. I want it to exactly match a valid post code only, excluding preceding and following characters. So (SW1A 0AA) should be passed as valid, but (foobarSW1A 0AA) should not.

I added the anchors to the regex (^ at the start and $ at the end) to try and force it to only accept a string that consists of only a postcode as valid. However, the class still passes postcodes with padding and/or non-postcode strings wrapping it.

What am I doing wrong? I thought adding the anchors would be enough to get the behaviour I wanted.

Answer 1

Add the anchors as:

^(?:regex)$

^foo|bar$ is not the same as ^(?:foo|bar)$.

You should also use \z instead of $. $ allows an optional line break at the end of string, while \z is a strict end of string match.