search/documents/physical_htm.php

   1 <?php
   2 /**
   3 * Global Search Engine for Moodle
   4 * add-on 1.8+ : Valery Fremaux [valery.fremaux@club-internet.fr]
   5 * 2007/08/02
   6 *
   7 * this is a format handler for getting text out of a proprietary binary format
   8 * so it can be indexed by Lucene search engine
   9 */
  10
  11 function get_text_for_indexing_htm(&$resource){
  12     global $CFG, $USER;
  13
  14     // SECURITY : do not allow non admin execute anything on system !!
  15     if (!isadmin($USER->id)) return;
  16
  17     // just get text
  18     $text = implode('', file("{$CFG->dataroot}/{$resource->course}/($resource->reference)"));
  19
  20     // extract keywords and other interesting meta information and put it back as real content for indexing
  21     if (preg_match('/(.*)<meta ([^>]*)>(.*)/is',$text, $matches)){
  22         $prefix = $matches[1];
  23         $meta_attributes = $matches[2];
  24         $suffix = $matches[3];
  25         if (preg_match('/name="(keywords|description)"/i', $attributes)){
  26             preg_match('/content="[^"]+"/i', $attributes, $matches);
  27             $text = $prefix.' '.$matches[1].' '.$suffix;
  28         }
  29     }
  30     // filter all html tags
  31     // $text = clean_text($text, FORMAT_PLAIN);
  32     // NOTE : this is done in ResourceSearchDocument __constructor
  33
  34     if (!empty($CFG->block_search_limit_index_body)){
  35         $text = shorten($text, $CFG->block_search_limit_index_body);
  36     }
  37     return $text;
  38 }
  39 ?>