search/documents/resource_document.php

   1 <?php
   2 /**
   3 * Global Search Engine for Moodle
   4 * Michael Champanis (mchampan) [cynnical@gmail.com]
   5 * review 1.8+ : Valery Fremaux [valery.fremaux@club-internet.fr]
   6 * 2007/08/02
   7 *
   8 * document handling for all resources
   9 * This file contains the mapping between a resource and it's indexable counterpart,
  10 *
  11 * Functions for iterating and retrieving the necessary records are now also included
  12 * in this file, rather than mod/resource/lib.php
  13 **/
  14
  15 require_once("$CFG->dirroot/search/documents/document.php");
  16 require_once("$CFG->dirroot/mod/resource/lib.php");
  17
  18 /*
  19 * a class for representing searchable information
  20 *
  21 **/
  22 class ResourceSearchDocument extends SearchDocument {
  23     public function __construct(&$resource, $context_id) {
  24         // generic information; required
  25         $doc->docid     = $resource['trueid'];
  26         $doc->documenttype = SEARCH_TYPE_RESOURCE;
  27         $doc->itemtype     = $resource['type'];
  28         $doc->contextid    = $context_id;
  29
  30         $doc->title     = strip_tags($resource['name']);
  31         $doc->date      = $resource['timemodified'];
  32         $doc->author    = '';
  33         $doc->contents  = strip_tags($resource['summary']).' '.strip_tags($resource['alltext']);
  34         $doc->url       = resource_make_link($resource['id']);
  35
  36         // module specific information; optional
  37         $data = array();
  38
  39         // construct the parent class
  40         parent::__construct($doc, $data, $resource['course'], 0, 0, PATH_FOR_SEARCH_TYPE_RESOURCE);
  41     } //constructor
  42 } //ResourceSearchDocument
  43
  44 /**
  45 * constructs valid access links to information
  46 * @param resourceId the of the resource
  47 * @return a full featured link element as a string
  48 */
  49 function resource_make_link($resource_id) {
  50     global $CFG;
  51
  52     return $CFG->wwwroot.'/mod/resource/view.php?id='.$resource_id;
  53 } //resource_make_link
  54
  55 /**
  56 * part of standard API
  57 *
  58 */
  59 function resource_iterator() {
  60     //trick to leave search indexer functionality intact, but allow
  61     //this document to only use the below function to return info
  62     //to be searched
  63     return array(true);
  64   } //resource_iterator
  65
  66 /**
  67 * part of standard API
  68 * this function does not need a content iterator, returns all the info
  69 * itself;
  70 * @param notneeded to comply API, remember to fake the iterator array though
  71 * @return an array of searchable documents
  72 */
  73 function resource_get_content_for_index(&$notneeded) {
  74     global $CFG;
  75
  76     // starting with Moodle native resources
  77     $documents = array();
  78     $query = "
  79         SELECT
  80             id as trueid,
  81             r.*
  82         FROM
  83             {$CFG->prefix}resource as r
  84         WHERE
  85             alltext != '' AND
  86             alltext != ' ' AND
  87             alltext != '&nbsp;' AND
  88             type != 'file'
  89     ";
  90     $resources = get_records_sql($query);
  91
  92     foreach($resources as $aResource){
  93         $coursemodule = get_field('modules', 'id', 'name', 'resource');
  94         $cm = get_record('course_modules', 'course', $aResource->course, 'module', $coursemodule, 'instance', $aResource->id);
  95         $context = get_context_instance(CONTEXT_MODULE, $cm->id);
  96         $aResource->id = $cm->id;
  97         $documents[] = new ResourceSearchDocument(get_object_vars($aResource), $context->id);
  98         mtrace("finished $aResource->name");
  99     }
 100
 101     // special physical files handling
 102     /**
 103     * this sequence searches for a compatible physical stream handler for getting a text
 104     * equivalence for the content.
 105     *
 106     */
 107     if (@$CFG->block_search_enable_file_indexing){
 108         $query = "
 109             SELECT
 110                r.id as trueid,
 111                cm.id as id,
 112                r.course as course,
 113                r.name as name,
 114                r.summary as summary,
 115                r.alltext as alltext,
 116                r.reference as reference,
 117                r.type as type,
 118                r.timemodified as timemodified
 119             FROM
 120                 {$CFG->prefix}resource as r,
 121                 {$CFG->prefix}course_modules as cm,
 122                 {$CFG->prefix}modules as m
 123             WHERE
 124                r.type = 'file' AND
 125                cm.instance = r.id AND
 126                cm.course = r.course AND
 127                cm.module = m.id AND
 128                m.name = 'resource'
 129         ";
 130         $resources = get_records_sql($query);
 131
 132         // invokes external content extractor if exists.
 133         foreach($resources as $aResource){
 134             // fetches a physical indexable document and adds it to documents passed by ref
 135             $coursemodule = get_field('modules', 'id', 'name', 'resource');
 136             $cm = get_record('course_modules', 'id', $aResource->id);
 137             $context = get_context_instance(CONTEXT_MODULE, $cm->id);
 138             resource_get_physical_file($aResource, $context->id, false, $documents);
 139         }
 140     }
 141     return $documents;
 142 } //resource_get_content_for_index
 143
 144 /**
 145 * get text from a physical file
 146 * @param resource a resource for which to fetch some representative text
 147 * @param getsingle if true, returns a single search document, elsewhere return the array
 148 * given as documents increased by one
 149 * @param documents the array of documents, by ref, where to add the new document.
 150 * @return a search document when unique or false.
 151 */
 152 function resource_get_physical_file(&$resource, $context_id, $getsingle, &$documents = null){
 153     global $CFG;
 154
 155     // cannot index empty references
 156     if (empty($resource->reference)) return false;
 157
 158     // cannot index remote resources
 159     if (resource_is_url($resource->reference)){
 160         mtrace("Cannot index remote URLs.");
 161         return false;
 162     }
 163
 164     $fileparts = pathinfo($resource->reference);
 165     // cannot index unknown or masked types
 166     if (empty($fileparts['extension'])) {
 167         return false;
 168     }
 169
 170     // cannot index non existent file
 171     $file = "{$CFG->dataroot}/{$resource->course}/{$resource->reference}";
 172     if (!file_exists($file)){
 173         mtrace("Missing resource file $file : will not be indexed.");
 174         return false;
 175     }
 176
 177     $ext = strtolower($fileparts['extension']);
 178
 179     // cannot index unallowed or unhandled types
 180     if (!preg_match("/\b$ext\b/i", $CFG->block_search_filetypes)) {
 181         mtrace($fileparts['extension'] . ' is not an allowed extension for indexing');
 182         return false;
 183     }
 184     if (file_exists($CFG->dirroot.'/search/documents/physical_'.$ext.'.php')){
 185         include_once($CFG->dirroot.'/search/documents/physical_'.$ext.'.php');
 186         $function_name = 'get_text_for_indexing_'.$ext;
 187         $resource->alltext = $function_name($resource);
 188         if (!empty($resource->alltext)){
 189             if ($getsingle){
 190                 return new ResourceSearchDocument(get_object_vars($resource));
 191             }
 192             else{
 193                 $documents[] = new ResourceSearchDocument(get_object_vars($resource), $context_id);
 194             }
 195             mtrace("finished file $resource->name as {$resource->reference}");
 196         }
 197     }
 198     else{
 199         mtrace("fulltext handler not found for $ext type");
 200     }
 201     return false;
 202 }
 203
 204 /**
 205 * part of standard API.
 206 * returns a single resource search document based on a resource_entry id
 207 * @param id the id of the accessible document
 208 * @return a searchable object or null if failure
 209 */
 210 function resource_single_document($id, $itemtype) {
 211     global $CFG;
 212
 213     // rewriting with legacy moodle databse API
 214     $query = "
 215         SELECT
 216            r.id as trueid,
 217            cm.id as id,
 218            r.course as course,
 219            r.name as name,
 220            r.summary as summary,
 221            r.alltext as alltext,
 222            r.reference as reference,
 223            r.type as type,
 224            r.timemodified as timemodified
 225         FROM
 226             {$CFG->prefix}resource as r,
 227             {$CFG->prefix}course_modules as cm,
 228             {$CFG->prefix}modules as m
 229         WHERE
 230             cm.instance = r.id AND
 231             cm.course = r.course AND
 232             cm.module = m.id AND
 233             m.name = 'resource' AND
 234             ((r.type != 'file' AND
 235             r.alltext != '' AND
 236             r.alltext != ' ' AND
 237             r.alltext != '&nbsp;') OR
 238             r.type = 'file') AND
 239             r.id = '{$id}'
 240     ";
 241     $resource = get_record_sql($query);
 242
 243     if ($resource){
 244         $coursemodule = get_field('modules', 'id', 'name', 'resource');
 245         $cm = get_record('course_modules', 'id', $resource->id);
 246         $context = get_context_instance(CONTEXT_MODULE, $cm->id);
 247         if ($resource->type == 'file' && @$CFG->block_search_enable_file_indexing){
 248             $document = resource_get_physical_file($resource, true, $context->id);
 249             if (!$document) mtrace("Warning : this document {$resource->name} will not be indexed");
 250             return $document;
 251         }
 252         else{
 253             return new ResourceSearchDocument(get_object_vars($resource), $context->id);
 254         }
 255     }
 256     mtrace("null resource");
 257     return null;
 258 } //resource_single_document
 259
 260 /**
 261 * dummy delete function that aggregates id with itemtype.
 262 * this was here for a reason, but I can't remember it at the moment.
 263 *
 264 */
 265 function resource_delete($info, $itemtype) {
 266     $object->id = $info;
 267     $object->itemtype = $itemtype;
 268     return $object;
 269 } //resource_delete
 270
 271 /**
 272 * returns the var names needed to build a sql query for addition/deletions
 273 *
 274 */
 275 function resource_db_names() {
 276     //[primary id], [table name], [time created field name], [time modified field name], [additional where conditions for sql]
 277     return array(array('id', 'resource', 'timemodified', 'timemodified', '*', " (alltext != '' AND alltext != ' ' AND alltext != '&nbsp;' AND TYPE != 'file') OR TYPE = 'file' "));
 278 } //resource_db_names
 279
 280 /**
 281 * this function handles the access policy to contents indexed as searchable documents. If this
 282 * function does not exist, the search engine assumes access is allowed.
 283 * @param path the access path to the module script code
 284 * @param itemtype the information subclassing (usefull for complex modules, defaults to 'standard')
 285 * @param this_id the item id within the information class denoted by itemtype. In resources, this id
 286 * points to the resource record and not to the module that shows it.
 287 * @param user the user record denoting the user who searches
 288 * @param group_id the current group used by the user when searching
 289 * @return true if access is allowed, false elsewhere
 290 */
 291 function resource_check_text_access($path, $itemtype, $this_id, $user, $group_id, $context_id){
 292     global $CFG;
 293
 294     include_once("{$CFG->dirroot}/{$path}/lib.php");
 295
 296     $r = get_record('resource', 'id', $this_id);
 297     $module_context = get_record('context', 'id', $context_id);
 298     $cm = get_record('course_modules', 'id', $module_context->instance);
 299
 300     //check if found course module is visible
 301     if (!$cm->visible and !has_capability('moodle/course:viewhiddenactivities', $module_context)){
 302         return false;
 303     }
 304
 305     return true;
 306 } //resource_check_text_access
 307 ?>