docs/include/lib/html_resource_parser.inc.php

   1 <?php
   2 /************************************************************************/
   3 /* AContent                                                             */
   4 /************************************************************************/
   5 /* Copyright (c) 2010                                                   */
   6 /* Inclusive Design Institute                                           */
   7 /*                                                                      */
   8 /* This program is free software. You can redistribute it and/or        */
   9 /* modify it under the terms of the GNU General Public License          */
  10 /* as published by the Free Software Foundation.                        */
  11 /************************************************************************/
  12
  13 function get_html_resources($text) {
  14         global $_course_id;
  15
  16         $resources = array();
  17
  18         $handler = new XML_HTMLSax_Handler();
  19
  20         $parser = new XML_HTMLSax();
  21         $parser->set_object($handler);
  22         $parser->set_element_handler('openHandler','closeHandler');
  23
  24         $parser->parse($text);
  25
  26         foreach ($handler->resources as $resource) {
  27                 $url_parts = @parse_url($resource);
  28
  29                 if (isset($url_parts['scheme'])) {
  30                         // we don't want full urls
  31                         continue;
  32                 }
  33
  34                 if ((substr($resource, 0, 1) == '/')) {
  35                         // we don't want absolute urls
  36                         continue;
  37                 }
  38
  39                 // make sure this resource exists in this course's content directory:
  40                 $resource_server_path = realpath(TR_CONTENT_DIR . $_course_id. '/' . $resource);
  41                 if (file_exists($resource_server_path) && is_file($resource_server_path)) {
  42                         $resources[$resource] = $resource_server_path;
  43                 }
  44         }
  45
  46         return $resources;
  47 }
  48
  49 /*
  50         the following resources are to be identified:
  51         even if some of these can't be images, they can still be files in the content dir.
  52         theoretically the only urls we wouldn't deal with would be for a <!DOCTYPE and <form>
  53
  54         img             => src
  55         a               => href                         // ignore if href doesn't exist (ie. <a name>)
  56         object  => data | classid       // probably only want data
  57         applet  => classid | archive                    // whatever these two are should double check to see if it's a valid file (not a dir)
  58         link    => href
  59         script  => src
  60         form    => action
  61         input   => src
  62         iframe  => src
  63 */
  64 class XML_HTMLSax_Handler {
  65         var $elements = array(  'img'    => 'src',
  66                                                         'a'              => 'href',
  67                                                         'object' =>  array('data',    'classid'),
  68                                                         'applet' =>  array('classid', 'archive'),
  69                                                         'link'   => 'href',
  70                                                         'script' => 'src',
  71                                                         'form'   => 'action',
  72                                                         'input'  => 'src',
  73                                                         'iframe' => 'src',
  74                                                         'embed'  => 'src',
  75                                                         'param'  => 'value');
  76         var $resources = array();
  77
  78     function XML_HTMLSax_Handler() {
  79                 $this->resources = array();
  80         }
  81
  82     function openHandler(& $parser,$name,$attrs) {
  83                 $name = strtolower($name);
  84                 $attrs = array_change_key_case($attrs, CASE_LOWER);
  85
  86                 /* check if this attribute specifies the files in different ways: (ie. java) */
  87                 if (is_array($this->elements[$name])) {
  88                         $items = $this->elements[$name];
  89
  90                         foreach ($items as $item) {
  91                                 if ($attrs[$item] != '') {
  92                                         /* some attributes allow a listing of files to include seperated by commas (ie. applet->archive). */
  93                                         if (strpos($attrs[$item], ',') !== false) {
  94                                                 $files = explode(',', $attrs[$item]);
  95                                                 foreach ($files as $file) {
  96                                                         $this->resources[] = trim($file);
  97                                                 }
  98                                         } else {
  99                                                 $this->resources[] = $attrs[$item];
 100                                         }
 101                                 }
 102                         }
 103                 } else if (isset($this->elements[$name]) && ($attrs[$this->elements[$name]] != '')) {
 104                         /* we know exactly which attribute contains the reference to the file. */
 105                         $this->resources[] = $attrs[$this->elements[$name]];
 106                 }
 107     }
 108     function closeHandler(& $parser,$name) { }
 109 }
 110 ?>