search/documents/physical_ppt.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 * first implementation is a trivial heuristic based on ppt character stream :
  13 * text sequence always starts with a 00 9F 0F 04 sequence followed by a 15 bytes
  14 * sequence
  15 * In this sequence is a A8 0F or A0 0F or AA 0F followed by a little-indian encoding of text buffer size
  16 * A8 0F denotes for ASCII text (local system monobyte encoding)
  17 * A0 0F denotes for UTF-16 encoding
  18 * AA 0F are non textual sequences
  19 * texts are either in ASCII or UTF-16
  20 * text ends on a new sequence start, or on a 00 00 NULL UTF-16 end of stream
  21 *
  22 * based on these following rules, here is a little empiric texte extractor for PPT
  23 */
  24
  25 function get_text_for_indexing_ppt(&$resource){
  26     global $CFG, $USER;
  27
  28     $indextext = null;
  29
  30     // SECURITY : do not allow non admin execute anything on system !!
  31     if (!isadmin($USER->id)) return;
  32
  33     $text = implode('', file("{$CFG->dataroot}/{$resource->course}/{$resource->reference}"));
  34
  35     $remains = $text;
  36     $fragments = array();
  37     while (preg_match('/\x00\x9F\x0F\x04.{9}(......)(.*)/s', $remains, $matches)){
  38         $unpacked = unpack("ncode/Llength", $matches[1]);
  39         $sequencecode = $unpacked['code'];
  40         $length = $unpacked['length'];
  41         // print "length : ".$length." ; segment type : ".sprintf("%x", $sequencecode)."<br/>";
  42         $followup = $matches[2];
  43         // local system encoding sequence
  44         if ($sequencecode == 0xA80F){
  45             $aFragment = substr($followup, 0, $length);
  46             $remains = substr($followup, $length);
  47             $fragments[] = $aFragment;
  48         }
  49         // denotes unicode encoded sequence
  50         elseif ($sequencecode == 0xA00F){
  51             $aFragment = substr($followup, 0, $length);
  52             // $aFragment = mb_convert_encoding($aFragment, 'UTF-16', 'UTF-8');
  53             $aFragment = preg_replace('/\xA0\x00\x19\x20/s', "'", $aFragment); // some quotes
  54             $aFragment = preg_replace('/\x00/s', "", $aFragment);
  55             $remains = substr($followup, $length);
  56             $fragments[] = $aFragment;
  57         }
  58         else{
  59             $remains = $followup;
  60         }
  61     }
  62     $indextext = implode(' ', $fragments);
  63     $indextext = preg_replace('/\x19\x20/', "'", $indextext); // some quotes
  64     $indextext = preg_replace('/\x09/', '', $indextext); // some extra chars
  65     $indextext = preg_replace('/\x0D/', "\n", $indextext); // some quotes
  66     $indextext = preg_replace('/\x0A/', "\n", $indextext); // some quotes
  67     $indextextprint = implode('<hr/>', $fragments);
  68
  69     $logppt = fopen("C:/php5/logs/pptlog", "w");
  70     fwrite($logppt, $indextext);
  71     fclose($logppt);
  72
  73     if (!empty($CFG->block_search_limit_index_body)){
  74         $indextext = shorten($text, $CFG->block_search_limit_index_body);
  75     }
  76
  77     $indextext = mb_convert_encoding($indextext, 'UTF8', 'auto');
  78     return $indextext;
  79 }
  80 ?>