search/documents/physical_ppt.php

   1 <?php
   2 /**
   3 * Global Search Engine for Moodle
   4 *
   5 * @package search
   6 * @category core
   7 * @subpackage document_wrappers
   8 * @author Valery Fremaux [valery.fremaux@club-internet.fr] > 1.8
   9 * @date 2008/03/31
  10 * @license http://www.gnu.org/copyleft/gpl.html GNU Public License
  11 *
  12 * this is a format handler for getting text out of a proprietary binary format
  13 * so it can be indexed by Lucene search engine
  14 */
  15
  16 /*
  17 * first implementation is a trivial heuristic based on ppt character stream :
  18 * text sequence always starts with a 00 9F 0F 04 sequence followed by a 15 bytes
  19 * sequence
  20 * In this sequence is a A8 0F or A0 0F or AA 0F followed by a little-indian encoding of text buffer size
  21 * A8 0F denotes for ASCII text (local system monobyte encoding)
  22 * A0 0F denotes for UTF-16 encoding
  23 * AA 0F are non textual sequences
  24 * texts are either in ASCII or UTF-16
  25 * text ends on a new sequence start, or on a 00 00 NULL UTF-16 end of stream
  26 *
  27 * based on these following rules, here is a little empiric texte extractor for PPT
  28 */
  29
  30 /**
  31 * @param object $resource
  32 * @uses CFG, USER
  33 */
  34 function get_text_for_indexing_ppt(&$resource){
  35     global $CFG, $USER;
  36
  37     $indextext = null;
  38
  39     // SECURITY : do not allow non admin execute anything on system !!
  40     if (!isadmin($USER->id)) return;
  41
  42     $text = implode('', file("{$CFG->dataroot}/{$resource->course}/{$resource->reference}"));
  43
  44     $remains = $text;
  45     $fragments = array();
  46     while (preg_match('/\x00\x9F\x0F\x04.{9}(......)(.*)/s', $remains, $matches)){
  47         $unpacked = unpack("ncode/Llength", $matches[1]);
  48         $sequencecode = $unpacked['code'];
  49         $length = $unpacked['length'];
  50         // print "length : ".$length." ; segment type : ".sprintf("%x", $sequencecode)."<br/>";
  51         $followup = $matches[2];
  52         // local system encoding sequence
  53         if ($sequencecode == 0xA80F){
  54             $aFragment = substr($followup, 0, $length);
  55             $remains = substr($followup, $length);
  56             $fragments[] = $aFragment;
  57         }
  58         // denotes unicode encoded sequence
  59         elseif ($sequencecode == 0xA00F){
  60             $aFragment = substr($followup, 0, $length);
  61             // $aFragment = mb_convert_encoding($aFragment, 'UTF-16', 'UTF-8');
  62             $aFragment = preg_replace('/\xA0\x00\x19\x20/s', "'", $aFragment); // some quotes
  63             $aFragment = preg_replace('/\x00/s', "", $aFragment);
  64             $remains = substr($followup, $length);
  65             $fragments[] = $aFragment;
  66         }
  67         else{
  68             $remains = $followup;
  69         }
  70     }
  71     $indextext = implode(' ', $fragments);
  72     $indextext = preg_replace('/\x19\x20/', "'", $indextext); // some quotes
  73     $indextext = preg_replace('/\x09/', '', $indextext); // some extra chars
  74     $indextext = preg_replace('/\x0D/', "\n", $indextext); // some quotes
  75     $indextext = preg_replace('/\x0A/', "\n", $indextext); // some quotes
  76     $indextextprint = implode('<hr/>', $fragments);
  77
  78     // debug code
  79     // $logppt = fopen("C:/php5/logs/pptlog", "w");
  80     // fwrite($logppt, $indextext);
  81     // fclose($logppt);
  82
  83     if (!empty($CFG->block_search_limit_index_body)){
  84         $indextext = shorten($text, $CFG->block_search_limit_index_body);
  85     }
  86
  87     $indextext = mb_convert_encoding($indextext, 'UTF8', 'auto');
  88     return $indextext;
  89 }
  90 ?>