library/HTMLPurifier/AttrDef/Lang.php

   1 <?php
   2
   3 /**
   4  * Validates the HTML attribute lang, effectively a language code.
   5  * @note Built according to RFC 3066, which obsoleted RFC 1766
   6  */
   7 class HTMLPurifier_AttrDef_Lang extends HTMLPurifier_AttrDef
   8 {
   9
  10     public function validate($string, $config, $context) {
  11
  12         $string = trim($string);
  13         if (!$string) return false;
  14
  15         $subtags = explode('-', $string);
  16         $num_subtags = count($subtags);
  17
  18         if ($num_subtags == 0) return false; // sanity check
  19
  20         // process primary subtag : $subtags[0]
  21         $length = strlen($subtags[0]);
  22         switch ($length) {
  23             case 0:
  24                 return false;
  25             case 1:
  26                 if (! ($subtags[0] == 'x' || $subtags[0] == 'i') ) {
  27                     return false;
  28                 }
  29                 break;
  30             case 2:
  31             case 3:
  32                 if (! ctype_alpha($subtags[0]) ) {
  33                     return false;
  34                 } elseif (! ctype_lower($subtags[0]) ) {
  35                     $subtags[0] = strtolower($subtags[0]);
  36                 }
  37                 break;
  38             default:
  39                 return false;
  40         }
  41
  42         $new_string = $subtags[0];
  43         if ($num_subtags == 1) return $new_string;
  44
  45         // process second subtag : $subtags[1]
  46         $length = strlen($subtags[1]);
  47         if ($length == 0 || ($length == 1 && $subtags[1] != 'x') || $length > 8 || !ctype_alnum($subtags[1])) {
  48             return $new_string;
  49         }
  50         if (!ctype_lower($subtags[1])) $subtags[1] = strtolower($subtags[1]);
  51
  52         $new_string .= '-' . $subtags[1];
  53         if ($num_subtags == 2) return $new_string;
  54
  55         // process all other subtags, index 2 and up
  56         for ($i = 2; $i < $num_subtags; $i++) {
  57             $length = strlen($subtags[$i]);
  58             if ($length == 0 || $length > 8 || !ctype_alnum($subtags[$i])) {
  59                 return $new_string;
  60             }
  61             if (!ctype_lower($subtags[$i])) {
  62                 $subtags[$i] = strtolower($subtags[$i]);
  63             }
  64             $new_string .= '-' . $subtags[$i];
  65         }
  66
  67         return $new_string;
  68
  69     }
  70
  71 }
  72
  73 // vim: et sw=4 sts=4