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 * @contributor Tatsuva Shirai 20090530
  10 * @date 2008/03/31
  11 * @license http://www.gnu.org/copyleft/gpl.html GNU Public License
  12 *
  13 * this is a format handler for getting text out of a proprietary binary format
  14 * so it can be indexed by Lucene search engine
  15 */
  16
  17 /*
  18 * first implementation is a trivial heuristic based on ppt character stream :
  19 * text sequence always starts with a 00 9F 0F 04 sequence followed by a 15 bytes
  20 * sequence
  21 * In this sequence is a A8 0F or A0 0F or AA 0F followed by a little-indian encoding of text buffer size
  22 * A8 0F denotes for ASCII text (local system monobyte encoding)
  23 * A0 0F denotes for UTF-16 encoding
  24 * AA 0F are non textual sequences
  25 * texts are either in ASCII or UTF-16
  26 * text ends on a new sequence start, or on a 00 00 NULL UTF-16 end of stream
  27 *
  28 * based on these following rules, here is a little empiric texte extractor for PPT
  29 */
  30
  31 /**
  32 * @param object $resource
  33 * @uses $CFG
  34 */
  35 function get_text_for_indexing_ppt(&$resource, $directfile = ''){
  36     global $CFG;
  37
  38     $indextext = null;
  39
  40     // SECURITY : do not allow non admin execute anything on system !!
  41     if (!has_capability('moodle/site:doanything', get_context_instance(CONTEXT_SYSTEM))) return;
  42
  43     if ($directfile == ''){
  44         $text = implode('', file("{$CFG->dataroot}/{$resource->course}/{$resource->reference}"));
  45     } else {
  46         $text = implode('', file("{$CFG->dataroot}/{$directfile}"));
  47     }
  48
  49     $remains = $text;
  50     $fragments = array();
  51     while (preg_match('/\x00\x9F\x0F\x04.{9}(......)(.*)/s', $remains, $matches)){
  52         $unpacked = unpack("ncode/Llength", $matches[1]);
  53         $sequencecode = $unpacked['code'];
  54         $length = $unpacked['length'];
  55         // print "length : ".$length." ; segment type : ".sprintf("%x", $sequencecode)."<br/>";
  56         $followup = $matches[2];
  57         // local system encoding sequence
  58         if ($sequencecode == 0xA80F){
  59             $aFragment = substr($followup, 0, $length);
  60             $remains = substr($followup, $length);
  61             $fragments[] = $aFragment;
  62         }
  63         // denotes unicode encoded sequence
  64         elseif ($sequencecode == 0xA00F){
  65             $aFragment = substr($followup, 0, $length);
  66             // $aFragment = mb_convert_encoding($aFragment, 'UTF-16', 'UTF-8');
  67             $aFragment = preg_replace('/\xA0\x00\x19\x20/s', "'", $aFragment); // some quotes
  68             $aFragment = preg_replace('/\x00/s', "", $aFragment);
  69             $remains = substr($followup, $length);
  70             $fragments[] = $aFragment;
  71         }
  72         else{
  73             $remains = $followup;
  74         }
  75     }
  76     $indextext = implode(' ', $fragments);
  77     $indextext = preg_replace('/\x19\x20/', "'", $indextext); // some quotes
  78     $indextext = preg_replace('/\x09/', '', $indextext); // some extra chars
  79     $indextext = preg_replace('/\x0D/', "\n", $indextext); // some quotes
  80     $indextext = preg_replace('/\x0A/', "\n", $indextext); // some quotes
  81     $indextextprint = implode('<hr/>', $fragments);
  82
  83     // debug code
  84     // $logppt = fopen("C:/php5/logs/pptlog", "w");
  85     // fwrite($logppt, $indextext);
  86     // fclose($logppt);
  87
  88     if (!empty($CFG->block_search_limit_index_body)){
  89         $indextext = shorten_text($text, $CFG->block_search_limit_index_body);
  90     }
  91
  92     $indextext = mb_convert_encoding($indextext, 'UTF-8', 'auto'); // Shirai 20090530 - MDL19342
  93     return $indextext;
  94 }
  95 ?>