search/documents/physical_doc.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 /*
  12 * MS Word extractor
  13 */
  14
  15 function get_text_for_indexing_doc(&$resource){
  16     global $CFG, $USER;
  17
  18     // SECURITY : do not allow non admin execute anything on system !!
  19     if (!isadmin($USER->id)) return;
  20
  21     // just call pdftotext over stdout and capture the output
  22     if (!empty($CFG->block_search_word_to_text_cmd)){
  23         if (!file_exists("{$CFG->dirroot}/{$CFG->block_search_word_to_text_cmd}")){
  24             mtrace('Error with MSWord to text converter command : exectuable not found.');
  25         }
  26         else{
  27             $file = $CFG->dataroot.'/'.$resource->course.'/'.$resource->reference;
  28             $text_converter_cmd = "{$CFG->dirroot}/{$CFG->block_search_word_to_text_cmd} $file";
  29             if ($CFG->block_search_word_to_text_env){
  30                 putenv($CFG->block_search_word_to_text_env);
  31             }
  32             $result = shell_exec($text_converter_cmd);
  33             if ($result){
  34                 return mb_convert_encoding($result, 'UTF8', 'auto');
  35             }
  36             else{
  37                 mtrace('Error with MSWord to text converter command : execution failed.');
  38                 return '';
  39             }
  40         }
  41     }
  42     else {
  43         mtrace('Error with MSWord to text converter command : command not set up. Execute once search block configuration.');
  44         return '';
  45     }
  46 }
  47 ?>