lib/htmlpurifier/HTMLPurifier/URIParser.php

   1 <?php
   2
   3 require_once 'HTMLPurifier/URI.php';
   4
   5 /**
   6  * Parses a URI into the components and fragment identifier as specified
   7  * by RFC 3986.
   8  */
   9 class HTMLPurifier_URIParser
  10 {
  11
  12     /**
  13      * Instance of HTMLPurifier_PercentEncoder to do normalization with.
  14      */
  15     var $percentEncoder;
  16
  17     function HTMLPurifier_URIParser() {
  18         $this->percentEncoder = new HTMLPurifier_PercentEncoder();
  19     }
  20
  21     /**
  22      * Parses a URI.
  23      * @param $uri string URI to parse
  24      * @return HTMLPurifier_URI representation of URI. This representation has
  25      *         not been validated yet and may not conform to RFC.
  26      */
  27     function parse($uri) {
  28
  29         $uri = $this->percentEncoder->normalize($uri);
  30
  31         // Regexp is as per Appendix B.
  32         // Note that ["<>] are an addition to the RFC's recommended
  33         // characters, because they represent external delimeters.
  34         $r_URI = '!'.
  35             '(([^:/?#"<>]+):)?'. // 2. Scheme
  36             '(//([^/?#"<>]*))?'. // 4. Authority
  37             '([^?#"<>]*)'.       // 5. Path
  38             '(\?([^#"<>]*))?'.   // 7. Query
  39             '(#([^"<>]*))?'.     // 8. Fragment
  40             '!';
  41
  42         $matches = array();
  43         $result = preg_match($r_URI, $uri, $matches);
  44
  45         if (!$result) return false; // *really* invalid URI
  46
  47         // seperate out parts
  48         $scheme     = !empty($matches[1]) ? $matches[2] : null;
  49         $authority  = !empty($matches[3]) ? $matches[4] : null;
  50         $path       = $matches[5]; // always present, can be empty
  51         $query      = !empty($matches[6]) ? $matches[7] : null;
  52         $fragment   = !empty($matches[8]) ? $matches[9] : null;
  53
  54         // further parse authority
  55         if ($authority !== null) {
  56             $r_authority = "/^((.+?)@)?(\[[^\]]+\]|[^:]*)(:(\d*))?/";
  57             $matches = array();
  58             preg_match($r_authority, $authority, $matches);
  59             $userinfo   = !empty($matches[1]) ? $matches[2] : null;
  60             $host       = !empty($matches[3]) ? $matches[3] : '';
  61             $port       = !empty($matches[4]) ? (int) $matches[5] : null;
  62         } else {
  63             $port = $host = $userinfo = null;
  64         }
  65
  66         return new HTMLPurifier_URI(
  67             $scheme, $userinfo, $host, $port, $path, $query, $fragment);
  68     }
  69
  70 }
  71