Merged Linkvalidator (tagged as 4.5RC1)
[Packages/TYPO3.CMS.git] / typo3 / sysext / linkvalidator / classes / class.tx_linkvalidator_processor.php
1 <?php
2 /***************************************************************
3 * Copyright notice
4 *
5 * (c) 2010 - 2011 Michael Miousse (michael.miousse@infoglobe.ca)
6 * All rights reserved
7 *
8 * This script is part of the TYPO3 project. The TYPO3 project is
9 * free software; you can redistribute it and/or modify
10 * it under the terms of the GNU General Public License as published by
11 * the Free Software Foundation; either version 2 of the License, or
12 * (at your option) any later version.
13 *
14 * The GNU General Public License can be found at
15 * http://www.gnu.org/copyleft/gpl.html.
16 *
17 * This script is distributed in the hope that it will be useful,
18 * but WITHOUT ANY WARRANTY; without even the implied warranty of
19 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
20 * GNU General Public License for more details.
21 *
22 * This copyright notice MUST APPEAR in all copies of the script!
23 ***************************************************************/
24
25 /**
26 * This class provides Processing plugin implementation.
27 *
28 * @author Michael Miousse <michael.miousse@infoglobe.ca>
29 * @author Jochen Rieger <j.rieger@connecta.ag>
30 * @package TYPO3
31 * @subpackage linkvalidator
32 */
33
34 $GLOBALS['LANG']->includeLLFile('EXT:linkvalidator/modfuncreport/locallang.xml');
35
36 class tx_linkvalidator_Processor {
37
38 /**
39 * Array of tables and fields to search for broken links.
40 *
41 * @var array
42 */
43 protected $searchFields = array();
44
45 /**
46 * List of comma seperated page uids (rootline downwards).
47 *
48 * @var string
49 */
50 protected $pidList = '';
51
52 /**
53 * Array of tables and the number of external links they contain.
54 *
55 * @var array
56 */
57 protected $linkCounts = array();
58
59 /**
60 * Array of tables and the number of broken external links they contain.
61 *
62 * @var array
63 */
64 protected $brokenLinkCounts = array();
65
66 /**
67 * Array of tables and records containing broken links.
68 *
69 * @var array
70 */
71 protected $recordsWithBrokenLinks = array();
72
73 /**
74 * Array for hooks for own checks.
75 *
76 * @var array
77 */
78 protected $hookObjectsArr = array();
79
80 /**
81 * Array with information about the current page.
82 *
83 * @var array
84 */
85 protected $extPageInTreeInfo = array();
86
87 /**
88 * Fill hookObjectsArr with different link types and possible XClasses.
89 */
90 public function __construct() {
91 // Hook to handle own checks
92 if (is_array($GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['linkvalidator']['checkLinks'])) {
93 foreach ($GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['linkvalidator']['checkLinks'] as $key => $classRef) {
94 $this->hookObjectsArr[$key] = &t3lib_div::getUserObj($classRef);
95 }
96 }
97 }
98
99 /**
100 * Init Function: Here all the needed configuration values are stored in class variables.
101 *
102 * @param array $searchField: list of fields in which to search for links
103 * @param string $pid: list of comma separated page uids in which to search for links
104 * @return void
105 */
106 public function init($searchField, $pid) {
107 $this->searchFields = $searchField;
108 $this->pidList = $pid;
109 }
110
111 /**
112 * Find all supported broken links and store them in tx_linkvalidator_link.
113 *
114 * @param array $checkOptions: list of hook object to activate
115 * @param boolean $considerHidden: defines whether to look into hidden fields or not
116 * @return void
117 */
118 public function getLinkStatistics($checkOptions = array(), $considerHidden = FALSE) {
119 $results = array();
120 $GLOBALS['TYPO3_DB']->exec_DELETEquery('tx_linkvalidator_link', 'record_pid in (' . $this->pidList . ') or ( record_uid IN (' . $this->pidList . ') and table_name like \'pages\')');
121
122 // let's traverse all configured tables
123 foreach ($this->searchFields as $table => $fields) {
124 if($table == 'pages'){
125 $where = 'deleted = 0 AND uid IN (' . $this->pidList . ')';
126 }
127 else{
128 $where = 'deleted = 0 AND pid IN (' . $this->pidList . ')';
129 }
130 if (!$considerHidden) {
131 $where .= t3lib_BEfunc::BEenableFields($table);
132 }
133 // if table is not configured, we assume the ext is not installed and therefore no need to check it
134 if (!is_array($GLOBALS['TCA'][$table])) continue;
135
136 // re-init selectFields for table
137 $selectFields = 'uid, pid';
138 $selectFields .= ', ' . $GLOBALS['TCA'][$table]['ctrl']['label'] . ', ' . implode(', ', $fields);
139
140 // TODO: only select rows that have content in at least one of the relevant fields (via OR)
141 $res = $GLOBALS['TYPO3_DB']->exec_SELECTquery($selectFields, $table, $where);
142 // Get record rows of table
143 while ($row = $GLOBALS['TYPO3_DB']->sql_fetch_assoc($res)) {
144 // Analyse each record
145 $this->analyzeRecord($results, $table, $fields, $row);
146 }
147 }
148
149 foreach ($this->hookObjectsArr as $key => $hookObj) {
150 if ((is_array($results[$key])) && empty($checkOptions) || (is_array($results[$key]) && $checkOptions[$key])) {
151 // check'em!
152 foreach ($results[$key] as $entryKey => $entryValue) {
153 $table = $entryValue['table'];
154 $record = array();
155 $record['headline'] = $entryValue['row'][$GLOBALS['TCA'][$table]['ctrl']['label']];
156 $record['record_pid'] = $entryValue['row']['pid'];
157 $record['record_uid'] = $entryValue['uid'];
158 $record['table_name'] = $table;
159 $record['link_title'] = $entryValue['link_title'];
160 $record['field'] = $entryValue['field'];
161 $record['last_check'] = time();
162
163 $this->recordReference = $entryValue['substr']['recordRef'];
164
165 $this->pageWithAnchor = $entryValue['pageAndAnchor'];
166
167 if (!empty($this->pageWithAnchor)) {
168 // page with anchor, e.g. 18#1580
169 $url = $this->pageWithAnchor;
170 } else {
171 $url = $entryValue['substr']['tokenValue'];
172 }
173
174 $this->linkCounts[$table]++;
175 $checkURL = $hookObj->checkLink($url, $entryValue, $this);
176 // broken link found!
177 if (!$checkURL) {
178 $response = array();
179 $response['valid'] = FALSE;
180 $response['errorParams'] = $hookObj->getErrorParams();
181 $this->brokenLinkCounts[$table]++;
182 $record['link_type'] = $key;
183 $record['url'] = $url;
184 $record['url_response'] = serialize($response);
185 $GLOBALS['TYPO3_DB']->exec_INSERTquery('tx_linkvalidator_link', $record);
186 } elseif (t3lib_div::_GP('showalllinks')) {
187 $response = array();
188 $response['valid'] = TRUE;
189 $this->brokenLinkCounts[$table]++;
190 $record['url'] = $url;
191 $record['link_type'] = $key;
192 $record['url_response'] = serialize($response);
193 $GLOBALS['TYPO3_DB']->exec_INSERTquery('tx_linkvalidator_link', $record);
194 }
195 }
196 }
197 }
198 }
199
200
201 /**
202 * Find all supported broken links for a specific record.
203 *
204 * @param array $results: array of broken links
205 * @param string $table: table name of the record
206 * @param array $fields: array of fields to analyze
207 * @param array $record: record to analyse
208 * @return void
209 */
210 public function analyzeRecord(&$results, $table, $fields, $record) {
211
212 // array to store urls from relevant field contents
213 $urls = array();
214
215 $referencedRecordType = '';
216 // last-parsed link element was a page.
217 $wasPage = TRUE;
218
219 // flag whether row contains a broken link in some field or not
220 $rowContainsBrokenLink = FALSE;
221
222 // put together content of all relevant fields
223 $haystack = '';
224 $htmlParser = t3lib_div::makeInstance('t3lib_parsehtml');
225
226 $idRecord = $record['uid'];
227
228 // get all references
229 foreach ($fields as $field) {
230 $haystack .= $record[$field] . ' --- ';
231 $conf = $GLOBALS['TCA'][$table]['columns'][$field]['config'];
232
233 $valueField = $record[$field];
234
235 // Check if a TCA configured field has softreferences defined (see TYPO3 Core API document)
236 if ($conf['softref'] && strlen($valueField)) {
237 // Explode the list of softreferences/parameters
238 $softRefs = t3lib_BEfunc::explodeSoftRefParserList($conf['softref']);
239 // Traverse soft references
240 foreach ($softRefs as $spKey => $spParams) {
241 // create / get object
242 $softRefObj = &t3lib_BEfunc::softRefParserObj($spKey);
243
244 // If there was an object returned...:
245 if (is_object($softRefObj)) {
246
247 // Do processing
248 $resultArray = $softRefObj->findRef($table, $field, $idRecord, $valueField, $spKey, $spParams);
249 if (!empty($resultArray['elements'])) {
250
251 if ($spKey == 'typolink_tag') {
252 $linkTags = $htmlParser->splitIntoBlock('link', $resultArray['content']);
253
254 for ($i = 1; $i < count($linkTags); $i += 2) {
255 $referencedRecordType = '';
256 foreach($resultArray['elements'] as $element) {
257 $type = '';
258 $r = $element['subst'];
259
260 if (!empty($r['tokenID'])) {
261 if (substr_count($linkTags[$i], $r['tokenID'])) {
262 // Type of referenced record
263 if (strpos($r['recordRef'], 'pages') !== FALSE) {
264 $currentR = $r;
265 // contains number of the page
266 $referencedRecordType = $r['tokenValue'];
267 $wasPage = TRUE;
268 }
269 // append number of content element to the page saved in the last loop
270 elseif ((strpos($r['recordRef'], 'tt_content') !== FALSE) && ($wasPage === TRUE)) {
271 $referencedRecordType = $referencedRecordType . '#c' . $r['tokenValue'];
272 $wasPage = FALSE;
273 } else {
274 $currentR = $r;
275 }
276 $title = strip_tags($linkTags[$i]);
277 }
278 }
279 }
280 foreach ($this->hookObjectsArr as $keyArr => $hookObj) {
281 $type = $hookObj->fetchType($currentR, $type, $keyArr);
282 }
283
284 $results[$type][$table . ':' . $field . ':' . $idRecord . ':' . $currentR["tokenID"]]["substr"] = $currentR;
285 $results[$type][$table . ':' . $field . ':' . $idRecord . ':' . $currentR["tokenID"]]["row"] = $record;
286 $results[$type][$table . ':' . $field . ':' . $idRecord . ':' . $currentR["tokenID"]]["table"] = $table;
287 $results[$type][$table . ':' . $field . ':' . $idRecord . ':' . $currentR["tokenID"]]["field"] = $field;
288 $results[$type][$table . ':' . $field . ':' . $idRecord . ':' . $currentR["tokenID"]]["uid"] = $idRecord;
289 $results[$type][$table . ':' . $field . ':' . $idRecord . ':' . $currentR["tokenID"]]["link_title"] = $title;
290 $results[$type][$table . ':' . $field . ':' . $idRecord . ':' . $currentR["tokenID"]]["pageAndAnchor"] = $referencedRecordType;
291
292 }
293 } else {
294
295 foreach ($resultArray['elements'] as $element) {
296 $r = $element['subst'];
297 $title = '';
298 $type = '';
299
300 if (!empty($r)) {
301 // Parse string for special TYPO3 <link> tag:
302
303 foreach ($this->hookObjectsArr as $keyArr => $hookObj) {
304 $type = $hookObj->fetchType($r, $type, $keyArr);
305 }
306 $results[$type][$table . ':' . $field . ':' . $idRecord . ':' . $r["tokenID"]]["substr"] = $r;
307 $results[$type][$table . ':' . $field . ':' . $idRecord . ':' . $r["tokenID"]]["row"] = $record;
308 $results[$type][$table . ':' . $field . ':' . $idRecord . ':' . $r["tokenID"]]["table"] = $table;
309 $results[$type][$table . ':' . $field . ':' . $idRecord . ':' . $r["tokenID"]]["field"] = $field;
310 $results[$type][$table . ':' . $field . ':' . $idRecord . ':' . $r["tokenID"]]["uid"] = $idRecord;
311
312 }
313 }
314 }
315 }
316 }
317 }
318 }
319 }
320 }
321
322 /**
323 * Fill a markerarray with the number of links found in a list of pages.
324 *
325 * @param string $curPage: comma separated list of page uids
326 * @return array markerarray with the number of links found
327 */
328 public function getLinkCounts($curPage) {
329 $markerArray = array();
330 if (($res = $GLOBALS['TYPO3_DB']->exec_SELECTquery(
331 'count(uid) as nbBrokenLinks,link_type',
332 'tx_linkvalidator_link',
333 'record_pid in (' . $this->pidList . ')',
334 'link_type'
335 ))) {
336 while (($row = $GLOBALS['TYPO3_DB']->sql_fetch_assoc($res))) {
337 $markerArray[$row['link_type']] = $row['nbBrokenLinks'];
338 $markerArray['brokenlinkCount'] += $row['nbBrokenLinks'];
339 }
340 }
341 return $markerArray;
342 }
343
344 /**
345 * Calls t3lib_tsfeBeUserAuth::extGetTreeList.
346 * Although this duplicates the function t3lib_tsfeBeUserAuth::extGetTreeList
347 * this is necessary to create the object that is used recursively by the original function.
348 *
349 * Generates a list of page uids from $id. List does not include $id itself.
350 * The only pages excluded from the list are deleted pages.
351 *
352 * level in the tree to start collecting uids. Zero means
353 * 'start right away', 1 = 'next level and out'
354 *
355 * @param integer Start page id
356 * @param integer Depth to traverse down the page tree.
357 * @param integer $begin is an optional integer that determines at which
358 * @param string Perms clause
359 * @return string Returns the list with a comma in the end (if any pages selected!)
360 */
361 public function extGetTreeList($id, $depth, $begin = 0, $permsClause) {
362 $depth = intval($depth);
363 $begin = intval($begin);
364 $id = intval($id);
365 $theList = '';
366
367 if ($depth > 0) {
368 $res = $GLOBALS['TYPO3_DB']->exec_SELECTquery(
369 'uid,title',
370 'pages',
371 'pid=' . $id . ' AND deleted=0 AND ' . $permsClause
372 );
373 while ($row = $GLOBALS['TYPO3_DB']->sql_fetch_assoc($res)) {
374 if ($begin <= 0) {
375 $theList .= $row['uid'] . ',';
376 $this->extPageInTreeInfo[] = array($row['uid'], htmlspecialchars($row['title'], $depth));
377 }
378 if ($depth > 1) {
379 $theList .= $this->extGetTreeList($row['uid'], $depth - 1, $begin - 1, $permsClause);
380 }
381 }
382 }
383 return $theList;
384 }
385
386
387 }
388
389 if (defined('TYPO3_MODE') && isset($GLOBALS['TYPO3_CONF_VARS'][TYPO3_MODE]['XCLASS']['ext/linkvalidator/classes/class.tx_linkvalidator_processor.php'])) {
390 include_once($GLOBALS['TYPO3_CONF_VARS'][TYPO3_MODE]['XCLASS']['ext/linkvalidator/classes/class.tx_linkvalidator_processor.php']);
391 }
392 ?>