search/indexer.php

   1 <?php
   2   /* The indexer logic -
   3    * Look through each installed module's search document class file (/search/documents)
   4    * for necessary search functions, and if they're present add the content to the index.
   5    * Repeat this for blocks.
   6    *
   7    * Because the iterator/retrieval functions are now stored in /search/documents/mod_document.php,
   8    * /mod/mod/lib.php doesn't have to be modified - and thus the search module becomes quite
   9    * self-sufficient. URL's are now stored in the index, stopping us from needing to require
  10    * the class files to generate a results page.
  11    *
  12    * Along with the index data, each document's summary gets stored in the database
  13    * and synchronised to the index (flat file) via the primary key ('id') which is mapped
  14    * to the 'db_id' field in the index
  15    * */
  16
  17   //this'll take some time, set up the environment
  18   @set_time_limit(0);
  19   @ob_implicit_flush(true);
  20   @ob_end_flush();
  21
  22   require_once('../config.php');
  23   require_once("$CFG->dirroot/search/lib.php");
  24
  25   //only administrators can index the moodle installation, because access to all pages is required
  26   require_login();
  27
  28   if (empty($CFG->enableglobalsearch)) {
  29     error('Global searching is not enabled.');
  30   }
  31
  32   if (!isadmin()) {
  33     error("You need to be an admin user to use this page.", "$CFG->wwwroot/login/index.php");
  34   } //if
  35
  36   //confirmation flag to prevent accidental reindexing (indexersplash.php is the correct entry point)
  37   $sure = strtolower(optional_param('areyousure', '', PARAM_ALPHA));
  38
  39   if ($sure != 'yes') {
  40     mtrace("<pre>Sorry, you need to confirm indexing via <a href='indexersplash.php'>indexersplash.php</a>"
  41           .". (<a href='index.php'>Back to query page</a>).</pre>");
  42
  43     exit(0);
  44   } //if
  45
  46   //check for php5 (lib.php)
  47   if (!search_check_php5()) {
  48     $phpversion = phpversion();
  49     mtrace("Sorry, global search requires PHP 5.0.0 or later (currently using version $phpversion)");
  50     exit(0);
  51   } //if
  52
  53   //php5 found, continue including php5-only files
  54   //require_once("$CFG->dirroot/search/Zend/Search/Lucene.php");
  55   require_once("$CFG->dirroot/search/indexlib.php");
  56
  57   mtrace('<pre>Server Time: '.date('r',time())."\n");
  58
  59   if ($CFG->search_indexer_busy == '1') {
  60     //means indexing was not finished previously
  61     mtrace("Warning: Indexing was not successfully completed last time, restarting.\n");
  62   } //if
  63
  64   //turn on busy flag
  65   set_config('search_indexer_busy', '1');
  66
  67   //paths
  68   $index_path = SEARCH_INDEX_PATH;
  69   $index_db_file = "$CFG->dirroot/search/db/$CFG->dbtype.sql";
  70   $dbcontrol = new IndexDBControl();
  71
  72   //setup directory in data root
  73   if (!file_exists($index_path)) {
  74     mtrace("Data directory ($index_path) does not exist, attempting to create.");
  75     if (!mkdir($index_path)) {
  76       search_pexit("Error creating data directory at: $index_path. Please correct.");
  77     } else {
  78       mtrace("Directory successfully created.");
  79     } //else
  80   } else {
  81     mtrace("Using $index_path as data directory.");
  82   } //else
  83
  84   $index = new Zend_Search_Lucene($index_path, true);
  85
  86   if (!$dbcontrol->checkDB()) {
  87     search_pexit("Database error. Please check settings/files.");
  88   } //if
  89
  90   //begin timer
  91   search_stopwatch();
  92   mtrace("Starting activity modules\n");
  93
  94   //the presence of the required search functions -
  95   // * mod_iterator
  96   // * mod_get_content_for_index
  97   //are the sole basis for including a module in the index at the moment.
  98
  99   if ($mods = get_records_select('modules' /*'index this module?' where statement*/)) {
 100     //add virtual modules onto the back of the array
 101     $mods = array_merge($mods, search_get_additional_modules());
 102
 103     foreach ($mods as $mod) {
 104       $class_file = $CFG->dirroot.'/search/documents/'.$mod->name.'_document.php';
 105
 106       if (file_exists($class_file)) {
 107         include_once($class_file);
 108
 109         //build function names
 110         $iter_function = $mod->name.'_iterator';
 111         $index_function = $mod->name.'_get_content_for_index';
 112
 113         $counter = 0;
 114         $doc = new stdClass;
 115
 116         if (function_exists($index_function) && function_exists($iter_function)) {
 117           mtrace("Processing module function $index_function ...");
 118
 119           foreach ($iter_function() as $i) {
 120             $documents = $index_function($i);
 121
 122             //begin transaction
 123
 124             foreach($documents as $document) {
 125               $counter++;
 126
 127               //object to insert into db
 128               $dbid = $dbcontrol->addDocument($document);
 129
 130               //synchronise db with index
 131               $document->addField(Zend_Search_Lucene_Field::Keyword('dbid', $dbid));
 132
 133               //add document to index
 134               $index->addDocument($document);
 135
 136               //commit every x new documents, and print a status message
 137               if (($counter%2000) == 0) {
 138                 $index->commit();
 139                 mtrace(".. $counter");
 140               } //if
 141             } //foreach
 142
 143             //end transaction
 144
 145           } //foreach
 146
 147           //commit left over documents, and finish up
 148           $index->commit();
 149
 150           mtrace("-- $counter documents indexed");
 151           mtrace("done.\n");
 152         } //if
 153       } //if
 154     } //foreach
 155   } //if
 156
 157   //finished modules
 158   mtrace('Finished activity modules');
 159   search_stopwatch();
 160
 161   //now blocks...
 162   //
 163
 164   mtrace(".<br/><a href='index.php'>Back to query page</a>.");
 165   mtrace('</pre>');
 166
 167   //finished, turn busy flag off
 168   set_config("search_indexer_busy", "0");
 169
 170   //mark the time we last updated
 171   set_config("search_indexer_run_date", time());
 172
 173   //and the index size
 174   set_config("search_index_size", (int)$index->count());
 175
 176 ?>