docs/include/lib/resources_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 if (!defined('TR_INCLUDE_PATH')) { exit; }
  14
  15 global $db;
  16
  17 define('TR_INCLUDE_PATH', '../include/');
  18 include_once(TR_INCLUDE_PATH.'../home/classes/ContentUtility.class.php');
  19
  20 $body_text      = htmlspecialchars($stripslashes($_POST['body_text']));
  21 $body           = html_entity_decode($body_text);
  22
  23 require(TR_INCLUDE_PATH.'classes/XML/XML_HTMLSax/XML_HTMLSax.php');     /* for XML_HTMLSax */
  24 require(TR_INCLUDE_PATH.'../home/imscc/include/ims_template.inc.php');                          /* for ims templates + print_organizations() */
  25
  26 /*
  27 the following resources are to be identified:
  28 even if some of these can't be images, they can still be files in the content dir.
  29 theoretically the only urls we wouldn't deal with would be for a <!DOCTYPE and <form>
  30
  31 img             => src
  32 a               => href                         // ignore if href doesn't exist (ie. <a name>)
  33 object  => data | classid       // probably only want data
  34 applet  => classid | archive                    // whatever these two are should double check to see if it's a valid file (not a dir)
  35 script  => src
  36 input   => src
  37 iframe  => src
  38 */
  39
  40 class MyHandler {
  41         function MyHandler(){}
  42         function openHandler(& $parser,$name,$attrs) {
  43                 global $my_files;
  44
  45                 $name = strtolower($name);
  46                 $attrs = array_change_key_case($attrs, CASE_LOWER);
  47
  48                 $elements = array(      'img'           => 'src',
  49                                                         'a'                     => 'href',
  50                                                         'object'        => array('data', 'classid'),
  51                                                         'applet'        => array('classid', 'archive'),
  52                                                         'script'        => 'src',
  53                                                         'input'         => 'src',
  54                                                         'iframe'        => 'src',
  55                                                         'embed'         => 'src',
  56                                                         );
  57
  58                 /* check if this attribute specifies the files in different ways: (ie. java) */
  59                 if (is_array($elements[$name])) {
  60                         $items = $elements[$name];
  61
  62                         foreach ($items as $item) {
  63                                 if ($attrs[$item] != '') {
  64
  65                                         /* some attributes allow a listing of files to include seperated by commas (ie. applet->archive). */
  66                                         if (strpos($attrs[$item], ',') !== false) {
  67                                                 $files = explode(',', $attrs[$item]);
  68                                                 foreach ($files as $file) {
  69                                                         $my_files[] = trim($file);
  70                                                 }
  71                                         } else {
  72                                                 $my_files[] = $attrs[$item];
  73                                         }
  74                                 }
  75                         }
  76                 } else if (isset($elements[$name]) && ($attrs[$elements[$name]] != '')) {
  77                         /* we know exactly which attribute contains the reference to the file. */
  78                         $my_files[] = $attrs[$elements[$name]];
  79                 }
  80         }
  81         function closeHandler(& $parser,$name) { }
  82         }
  83
  84 /* get all the content */
  85 $handler=new MyHandler();
  86 $parser = new XML_HTMLSax();
  87 $parser->set_object($handler);
  88 $parser->set_element_handler('openHandler','closeHandler');
  89
  90 /* generate the resources and save the HTML files */
  91
  92 global $parser, $my_files;
  93 global $course_id;
  94
  95 /* add the resource dependancies */
  96 $my_files               = array();
  97 $content_files  = "\n";
  98
  99 //in order to control if some [media] is in the body_text
 100 //$body = ContentUtility::embedMedia($body_t);
 101
 102 $parser->parse($body);
 103
 104 // find all [media] resources
 105 preg_match_all("/\[media[0-9a-z\|]*\](.*)\[\/media\]/i",$body,$media_matches);
 106 $my_files = array_merge($media_matches[1], $my_files);
 107
 108 // add by Cindy Li.
 109 // This resolves the problem introduced by [media] tag: when [media] is
 110 // parsed into <object>, same resource appears a few times in <object> with different
 111 // format to cater for different browsers or players. This way creates problem that different
 112 // formats in <object> are all parsed and considered as different resource. array_unique()
 113 // call solves this problem. But, it introduces the new problem that when a same resource
 114 // appears at different places in the content and users do want to have them with different
 115 // alternatives. With this solution, this same resource only shows up once at "adapt content"
 116 // and only can have one alternative associated with. Table and scripts need to be re-designed
 117 // to solve this problem, for example, include line number in table.
 118 $my_files = array_unique($my_files);
 119
 120 /* handle @import */
 121 $import_files   = get_import_files($body);
 122
 123 if (count($import_files) > 0) $my_files = array_merge($my_files, $import_files);
 124
 125 $i=0;
 126
 127 foreach ($my_files as $file) {
 128         /* filter out full urls */
 129         $url_parts = @parse_url($file);
 130 //      if (isset($url_parts['scheme']) && substr($file, 0, strlen(TR_BASE_HREF)) != TR_BASE_HREF) {
 131 //              continue;
 132 //      }
 133
 134         /* file should be relative to content. let's double check */
 135         if ((substr($file, 0, 1) == '/')) {
 136                 continue;
 137         }
 138
 139     // The URL of the movie from youtube.com has been converted above in embed_media().
 140         // For example:  http://www.youtube.com/watch?v=a0ryB0m0MiM is converted to
 141         // http://www.youtube.com/v/a0ryB0m0MiM to make it playable. This creates the problem
 142         // that the parsed-out url (http://www.youtube.com/v/a0ryB0m0MiM) does not match with
 143         // the URL saved in content table (http://www.youtube.com/watch?v=a0ryB0m0MiM).
 144         // The code below is to convert the URL back to original.
 145         $file = preg_replace("/(http:\/\/[a-z0-9\.]*)?youtube.com\/v\/(.*)/",
 146                              "\\1youtube.com/watch?v=\\2", $file);
 147
 148         $resources[$i] = $file;
 149         $i++;
 150 }
 151 ?>