AOEpeople /
crawler
Complex classes like CrawlerController often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes. You can also have a look at the cohesion graph to spot any un-connected, or weakly-connected components.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use CrawlerController, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 58 | class CrawlerController |
||
| 59 | { |
||
| 60 | const CLI_STATUS_NOTHING_PROCCESSED = 0; |
||
| 61 | const CLI_STATUS_REMAIN = 1; //queue not empty |
||
| 62 | const CLI_STATUS_PROCESSED = 2; //(some) queue items where processed |
||
| 63 | const CLI_STATUS_ABORTED = 4; //instance didn't finish |
||
| 64 | const CLI_STATUS_POLLABLE_PROCESSED = 8; |
||
| 65 | |||
| 66 | /** |
||
| 67 | * @var integer |
||
| 68 | */ |
||
| 69 | public $setID = 0; |
||
| 70 | |||
| 71 | /** |
||
| 72 | * @var string |
||
| 73 | */ |
||
| 74 | public $processID = ''; |
||
| 75 | |||
| 76 | /** |
||
| 77 | * One hour is max stalled time for the CLI |
||
| 78 | * If the process had the status "start" for 3600 seconds, it will be regarded stalled and a new process is started |
||
| 79 | * |
||
| 80 | * @var integer |
||
| 81 | */ |
||
| 82 | public $max_CLI_exec_time = 3600; |
||
| 83 | |||
| 84 | /** |
||
| 85 | * @var array |
||
| 86 | */ |
||
| 87 | public $duplicateTrack = []; |
||
| 88 | |||
| 89 | /** |
||
| 90 | * @var array |
||
| 91 | */ |
||
| 92 | public $downloadUrls = []; |
||
| 93 | |||
| 94 | /** |
||
| 95 | * @var array |
||
| 96 | */ |
||
| 97 | public $incomingProcInstructions = []; |
||
| 98 | |||
| 99 | /** |
||
| 100 | * @var array |
||
| 101 | */ |
||
| 102 | public $incomingConfigurationSelection = []; |
||
| 103 | |||
| 104 | /** |
||
| 105 | * @var bool |
||
| 106 | */ |
||
| 107 | public $registerQueueEntriesInternallyOnly = false; |
||
| 108 | |||
| 109 | /** |
||
| 110 | * @var array |
||
| 111 | */ |
||
| 112 | public $queueEntries = []; |
||
| 113 | |||
| 114 | /** |
||
| 115 | * @var array |
||
| 116 | */ |
||
| 117 | public $urlList = []; |
||
| 118 | |||
| 119 | /** |
||
| 120 | * @var boolean |
||
| 121 | */ |
||
| 122 | public $debugMode = false; |
||
| 123 | |||
| 124 | /** |
||
| 125 | * @var array |
||
| 126 | */ |
||
| 127 | public $extensionSettings = []; |
||
| 128 | |||
| 129 | /** |
||
| 130 | * Mount Point |
||
| 131 | * |
||
| 132 | * @var boolean |
||
| 133 | */ |
||
| 134 | public $MP = false; |
||
| 135 | |||
| 136 | /** |
||
| 137 | * @var string |
||
| 138 | */ |
||
| 139 | protected $processFilename; |
||
| 140 | |||
| 141 | /** |
||
| 142 | * Holds the internal access mode can be 'gui','cli' or 'cli_im' |
||
| 143 | * |
||
| 144 | * @var string |
||
| 145 | */ |
||
| 146 | protected $accessMode; |
||
| 147 | |||
| 148 | /** |
||
| 149 | * @var DatabaseConnection |
||
| 150 | */ |
||
| 151 | private $db; |
||
| 152 | |||
| 153 | /** |
||
| 154 | * @var BackendUserAuthentication |
||
| 155 | */ |
||
| 156 | private $backendUser; |
||
| 157 | |||
| 158 | /** |
||
| 159 | * @var integer |
||
| 160 | */ |
||
| 161 | private $scheduledTime = 0; |
||
| 162 | |||
| 163 | /** |
||
| 164 | * @var integer |
||
| 165 | */ |
||
| 166 | private $reqMinute = 0; |
||
| 167 | |||
| 168 | /** |
||
| 169 | * @var bool |
||
| 170 | */ |
||
| 171 | private $submitCrawlUrls = false; |
||
| 172 | |||
| 173 | /** |
||
| 174 | * @var bool |
||
| 175 | */ |
||
| 176 | private $downloadCrawlUrls = false; |
||
| 177 | |||
| 178 | /** |
||
| 179 | * @var QueueRepository |
||
| 180 | */ |
||
| 181 | protected $queueRepository; |
||
| 182 | |||
| 183 | /** |
||
| 184 | * Method to set the accessMode can be gui, cli or cli_im |
||
| 185 | * |
||
| 186 | * @return string |
||
| 187 | */ |
||
| 188 | 1 | public function getAccessMode() |
|
| 189 | { |
||
| 190 | 1 | return $this->accessMode; |
|
| 191 | } |
||
| 192 | |||
| 193 | /** |
||
| 194 | * @param string $accessMode |
||
| 195 | */ |
||
| 196 | 1 | public function setAccessMode($accessMode) |
|
| 197 | { |
||
| 198 | 1 | $this->accessMode = $accessMode; |
|
| 199 | 1 | } |
|
| 200 | |||
| 201 | /** |
||
| 202 | * Set disabled status to prevent processes from being processed |
||
| 203 | * |
||
| 204 | * @param bool $disabled (optional, defaults to true) |
||
| 205 | * @return void |
||
| 206 | */ |
||
| 207 | 3 | public function setDisabled($disabled = true) |
|
| 208 | { |
||
| 209 | 3 | if ($disabled) { |
|
| 210 | 2 | GeneralUtility::writeFile($this->processFilename, ''); |
|
| 211 | } else { |
||
| 212 | 1 | if (is_file($this->processFilename)) { |
|
| 213 | 1 | unlink($this->processFilename); |
|
| 214 | } |
||
| 215 | } |
||
| 216 | 3 | } |
|
| 217 | |||
| 218 | /** |
||
| 219 | * Get disable status |
||
| 220 | * |
||
| 221 | * @return bool true if disabled |
||
| 222 | */ |
||
| 223 | 3 | public function getDisabled() |
|
| 224 | { |
||
| 225 | 3 | if (is_file($this->processFilename)) { |
|
| 226 | 2 | return true; |
|
| 227 | } else { |
||
| 228 | 1 | return false; |
|
| 229 | } |
||
| 230 | } |
||
| 231 | |||
| 232 | /** |
||
| 233 | * @param string $filenameWithPath |
||
| 234 | * |
||
| 235 | * @return void |
||
| 236 | */ |
||
| 237 | 4 | public function setProcessFilename($filenameWithPath) |
|
| 238 | { |
||
| 239 | 4 | $this->processFilename = $filenameWithPath; |
|
| 240 | 4 | } |
|
| 241 | |||
| 242 | /** |
||
| 243 | * @return string |
||
| 244 | */ |
||
| 245 | 1 | public function getProcessFilename() |
|
| 246 | { |
||
| 247 | 1 | return $this->processFilename; |
|
| 248 | } |
||
| 249 | |||
| 250 | /************************************ |
||
| 251 | * |
||
| 252 | * Getting URLs based on Page TSconfig |
||
| 253 | * |
||
| 254 | ************************************/ |
||
| 255 | |||
| 256 | 28 | public function __construct() |
|
| 257 | { |
||
| 258 | 28 | $objectManager = GeneralUtility::makeInstance(ObjectManager::class); |
|
| 259 | 28 | $this->queueRepository = $objectManager->get(QueueRepository::class); |
|
| 260 | |||
| 261 | 28 | $this->db = $GLOBALS['TYPO3_DB']; |
|
| 262 | 28 | $this->backendUser = $GLOBALS['BE_USER']; |
|
| 263 | 28 | $this->processFilename = PATH_site . 'typo3temp/tx_crawler.proc'; |
|
| 264 | |||
| 265 | 28 | $settings = unserialize($GLOBALS['TYPO3_CONF_VARS']['EXT']['extConf']['crawler']); |
|
| 266 | 28 | $settings = is_array($settings) ? $settings : []; |
|
| 267 | |||
| 268 | // read ext_em_conf_template settings and set |
||
| 269 | 28 | $this->setExtensionSettings($settings); |
|
| 270 | |||
| 271 | // set defaults: |
||
| 272 | 28 | if (MathUtility::convertToPositiveInteger($this->extensionSettings['countInARun']) == 0) { |
|
| 273 | 21 | $this->extensionSettings['countInARun'] = 100; |
|
| 274 | } |
||
| 275 | |||
| 276 | 28 | $this->extensionSettings['processLimit'] = MathUtility::forceIntegerInRange($this->extensionSettings['processLimit'], 1, 99, 1); |
|
| 277 | 28 | } |
|
| 278 | |||
| 279 | /** |
||
| 280 | * Sets the extensions settings (unserialized pendant of $TYPO3_CONF_VARS['EXT']['extConf']['crawler']). |
||
| 281 | * |
||
| 282 | * @param array $extensionSettings |
||
| 283 | * @return void |
||
| 284 | */ |
||
| 285 | 37 | public function setExtensionSettings(array $extensionSettings) |
|
| 286 | { |
||
| 287 | 37 | $this->extensionSettings = $extensionSettings; |
|
| 288 | 37 | } |
|
| 289 | |||
| 290 | /** |
||
| 291 | * Check if the given page should be crawled |
||
| 292 | * |
||
| 293 | * @param array $pageRow |
||
| 294 | * @return false|string false if the page should be crawled (not excluded), true / skipMessage if it should be skipped |
||
| 295 | */ |
||
| 296 | 10 | public function checkIfPageShouldBeSkipped(array $pageRow) |
|
| 297 | { |
||
| 298 | 10 | $skipPage = false; |
|
| 299 | 10 | $skipMessage = 'Skipped'; // message will be overwritten later |
|
| 300 | |||
| 301 | // if page is hidden |
||
| 302 | 10 | if (!$this->extensionSettings['crawlHiddenPages']) { |
|
| 303 | 10 | if ($pageRow['hidden']) { |
|
| 304 | 1 | $skipPage = true; |
|
| 305 | 1 | $skipMessage = 'Because page is hidden'; |
|
| 306 | } |
||
| 307 | } |
||
| 308 | |||
| 309 | 10 | if (!$skipPage) { |
|
| 310 | 9 | if (GeneralUtility::inList('3,4', $pageRow['doktype']) || $pageRow['doktype'] >= 199) { |
|
| 311 | 3 | $skipPage = true; |
|
| 312 | 3 | $skipMessage = 'Because doktype is not allowed'; |
|
| 313 | } |
||
| 314 | } |
||
| 315 | |||
| 316 | 10 | if (!$skipPage) { |
|
| 317 | 6 | if (is_array($GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['crawler']['excludeDoktype'])) { |
|
| 318 | 2 | foreach ($GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['crawler']['excludeDoktype'] as $key => $doktypeList) { |
|
| 319 | 1 | if (GeneralUtility::inList($doktypeList, $pageRow['doktype'])) { |
|
| 320 | 1 | $skipPage = true; |
|
| 321 | 1 | $skipMessage = 'Doktype was excluded by "' . $key . '"'; |
|
| 322 | 1 | break; |
|
| 323 | } |
||
| 324 | } |
||
| 325 | } |
||
| 326 | } |
||
| 327 | |||
| 328 | 10 | if (!$skipPage) { |
|
| 329 | // veto hook |
||
| 330 | 5 | if (is_array($GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['crawler']['pageVeto'])) { |
|
| 331 | foreach ($GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['crawler']['pageVeto'] as $key => $func) { |
||
| 332 | $params = [ |
||
| 333 | 'pageRow' => $pageRow |
||
| 334 | ]; |
||
| 335 | // expects "false" if page is ok and "true" or a skipMessage if this page should _not_ be crawled |
||
| 336 | $veto = GeneralUtility::callUserFunction($func, $params, $this); |
||
| 337 | if ($veto !== false) { |
||
| 338 | $skipPage = true; |
||
| 339 | if (is_string($veto)) { |
||
| 340 | $skipMessage = $veto; |
||
| 341 | } else { |
||
| 342 | $skipMessage = 'Veto from hook "' . htmlspecialchars($key) . '"'; |
||
| 343 | } |
||
| 344 | // no need to execute other hooks if a previous one return a veto |
||
| 345 | break; |
||
| 346 | } |
||
| 347 | } |
||
| 348 | } |
||
| 349 | } |
||
| 350 | |||
| 351 | 10 | return $skipPage ? $skipMessage : false; |
|
| 352 | } |
||
| 353 | |||
| 354 | /** |
||
| 355 | * Wrapper method for getUrlsForPageId() |
||
| 356 | * It returns an array of configurations and no urls! |
||
| 357 | * |
||
| 358 | * @param array $pageRow Page record with at least dok-type and uid columns. |
||
| 359 | * @param string $skipMessage |
||
| 360 | * @return array |
||
| 361 | * @see getUrlsForPageId() |
||
| 362 | */ |
||
| 363 | 6 | public function getUrlsForPageRow(array $pageRow, &$skipMessage = '') |
|
| 364 | { |
||
| 365 | 6 | $message = $this->checkIfPageShouldBeSkipped($pageRow); |
|
| 366 | |||
| 367 | 6 | if ($message === false) { |
|
| 368 | 5 | $forceSsl = ($pageRow['url_scheme'] === 2) ? true : false; |
|
| 369 | 5 | $res = $this->getUrlsForPageId($pageRow['uid'], $forceSsl); |
|
| 370 | 5 | $skipMessage = ''; |
|
| 371 | } else { |
||
| 372 | 1 | $skipMessage = $message; |
|
| 373 | 1 | $res = []; |
|
| 374 | } |
||
| 375 | |||
| 376 | 6 | return $res; |
|
| 377 | } |
||
| 378 | |||
| 379 | /** |
||
| 380 | * This method is used to count if there are ANY unprocessed queue entries |
||
| 381 | * of a given page_id and the configuration which matches a given hash. |
||
| 382 | * If there if none, we can skip an inner detail check |
||
| 383 | * |
||
| 384 | * @param int $uid |
||
| 385 | * @param string $configurationHash |
||
| 386 | * @return boolean |
||
| 387 | */ |
||
| 388 | 7 | protected function noUnprocessedQueueEntriesForPageWithConfigurationHashExist($uid, $configurationHash) |
|
| 389 | { |
||
| 390 | 7 | $configurationHash = $this->db->fullQuoteStr($configurationHash, 'tx_crawler_queue'); |
|
| 391 | 7 | $res = $this->db->exec_SELECTquery('count(*) as anz', 'tx_crawler_queue', "page_id=" . intval($uid) . " AND configuration_hash=" . $configurationHash . " AND exec_time=0"); |
|
| 392 | 7 | $row = $this->db->sql_fetch_assoc($res); |
|
| 393 | |||
| 394 | 7 | return ($row['anz'] == 0); |
|
| 395 | } |
||
| 396 | |||
| 397 | /** |
||
| 398 | * Creates a list of URLs from input array (and submits them to queue if asked for) |
||
| 399 | * See Web > Info module script + "indexed_search"'s crawler hook-client using this! |
||
| 400 | * |
||
| 401 | * @param array Information about URLs from pageRow to crawl. |
||
| 402 | * @param array Page row |
||
| 403 | * @param integer Unix time to schedule indexing to, typically time() |
||
| 404 | * @param integer Number of requests per minute (creates the interleave between requests) |
||
| 405 | * @param boolean If set, submits the URLs to queue |
||
| 406 | * @param boolean If set (and submitcrawlUrls is false) will fill $downloadUrls with entries) |
||
| 407 | * @param array Array which is passed by reference and contains the an id per url to secure we will not crawl duplicates |
||
| 408 | * @param array Array which will be filled with URLS for download if flag is set. |
||
| 409 | * @param array Array of processing instructions |
||
| 410 | * @return string List of URLs (meant for display in backend module) |
||
| 411 | * |
||
| 412 | */ |
||
| 413 | 4 | public function urlListFromUrlArray( |
|
| 414 | array $vv, |
||
| 415 | array $pageRow, |
||
| 416 | $scheduledTime, |
||
| 417 | $reqMinute, |
||
| 418 | $submitCrawlUrls, |
||
| 419 | $downloadCrawlUrls, |
||
| 420 | array &$duplicateTrack, |
||
| 421 | array &$downloadUrls, |
||
| 422 | array $incomingProcInstructions |
||
| 423 | ) { |
||
| 424 | 4 | $urlList = ''; |
|
| 425 | // realurl support (thanks to Ingo Renner) |
||
| 426 | 4 | if (ExtensionManagementUtility::isLoaded('realurl') && $vv['subCfg']['realurl']) { |
|
| 427 | |||
| 428 | /** @var tx_realurl $urlObj */ |
||
| 429 | $urlObj = GeneralUtility::makeInstance('tx_realurl'); |
||
| 430 | |||
| 431 | if (!empty($vv['subCfg']['baseUrl'])) { |
||
| 432 | $urlParts = parse_url($vv['subCfg']['baseUrl']); |
||
| 433 | $host = strtolower($urlParts['host']); |
||
| 434 | $urlObj->host = $host; |
||
| 435 | |||
| 436 | // First pass, finding configuration OR pointer string: |
||
| 437 | $urlObj->extConf = isset($GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['realurl'][$urlObj->host]) ? $GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['realurl'][$urlObj->host] : $GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['realurl']['_DEFAULT']; |
||
| 438 | |||
| 439 | // If it turned out to be a string pointer, then look up the real config: |
||
| 440 | if (is_string($urlObj->extConf)) { |
||
| 441 | $urlObj->extConf = is_array($GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['realurl'][$urlObj->extConf]) ? $GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['realurl'][$urlObj->extConf] : $GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['realurl']['_DEFAULT']; |
||
| 442 | } |
||
| 443 | } |
||
| 444 | |||
| 445 | if (!$GLOBALS['TSFE']->sys_page) { |
||
| 446 | $GLOBALS['TSFE']->sys_page = GeneralUtility::makeInstance('TYPO3\CMS\Frontend\Page\PageRepository'); |
||
| 447 | } |
||
| 448 | |||
| 449 | if (!$GLOBALS['TSFE']->tmpl->rootLine[0]['uid']) { |
||
| 450 | $GLOBALS['TSFE']->tmpl->rootLine[0]['uid'] = $urlObj->extConf['pagePath']['rootpage_id']; |
||
| 451 | } |
||
| 452 | } |
||
| 453 | |||
| 454 | 4 | if (is_array($vv['URLs'])) { |
|
| 455 | 4 | $configurationHash = $this->getConfigurationHash($vv); |
|
| 456 | 4 | $skipInnerCheck = $this->noUnprocessedQueueEntriesForPageWithConfigurationHashExist($pageRow['uid'], $configurationHash); |
|
| 457 | |||
| 458 | 4 | foreach ($vv['URLs'] as $urlQuery) { |
|
| 459 | 4 | if ($this->drawURLs_PIfilter($vv['subCfg']['procInstrFilter'], $incomingProcInstructions)) { |
|
| 460 | |||
| 461 | // Calculate cHash: |
||
| 462 | 4 | if ($vv['subCfg']['cHash']) { |
|
| 463 | /* @var $cacheHash \TYPO3\CMS\Frontend\Page\CacheHashCalculator */ |
||
| 464 | $cacheHash = GeneralUtility::makeInstance('TYPO3\CMS\Frontend\Page\CacheHashCalculator'); |
||
| 465 | $urlQuery .= '&cHash=' . $cacheHash->generateForParameters($urlQuery); |
||
| 466 | } |
||
| 467 | |||
| 468 | // Create key by which to determine unique-ness: |
||
| 469 | 4 | $uKey = $urlQuery . '|' . $vv['subCfg']['userGroups'] . '|' . $vv['subCfg']['baseUrl'] . '|' . $vv['subCfg']['procInstrFilter']; |
|
| 470 | |||
| 471 | // realurl support (thanks to Ingo Renner) |
||
| 472 | 4 | $urlQuery = 'index.php' . $urlQuery; |
|
| 473 | 4 | if (ExtensionManagementUtility::isLoaded('realurl') && $vv['subCfg']['realurl']) { |
|
| 474 | $params = [ |
||
| 475 | 'LD' => [ |
||
| 476 | 'totalURL' => $urlQuery |
||
| 477 | ], |
||
| 478 | 'TCEmainHook' => true |
||
| 479 | ]; |
||
| 480 | $urlObj->encodeSpURL($params); |
||
|
|
|||
| 481 | $urlQuery = $params['LD']['totalURL']; |
||
| 482 | } |
||
| 483 | |||
| 484 | // Scheduled time: |
||
| 485 | 4 | $schTime = $scheduledTime + round(count($duplicateTrack) * (60 / $reqMinute)); |
|
| 486 | 4 | $schTime = floor($schTime / 60) * 60; |
|
| 487 | |||
| 488 | 4 | if (isset($duplicateTrack[$uKey])) { |
|
| 489 | |||
| 490 | //if the url key is registered just display it and do not resubmit is |
||
| 491 | $urlList = '<em><span class="typo3-dimmed">' . htmlspecialchars($urlQuery) . '</span></em><br/>'; |
||
| 492 | } else { |
||
| 493 | 4 | $urlList = '[' . date('d.m.y H:i', $schTime) . '] ' . htmlspecialchars($urlQuery); |
|
| 494 | 4 | $this->urlList[] = '[' . date('d.m.y H:i', $schTime) . '] ' . $urlQuery; |
|
| 495 | |||
| 496 | 4 | $theUrl = ($vv['subCfg']['baseUrl'] ? $vv['subCfg']['baseUrl'] : GeneralUtility::getIndpEnv('TYPO3_SITE_URL')) . $urlQuery; |
|
| 497 | |||
| 498 | // Submit for crawling! |
||
| 499 | 4 | if ($submitCrawlUrls) { |
|
| 500 | 4 | $added = $this->addUrl( |
|
| 501 | 4 | $pageRow['uid'], |
|
| 502 | 4 | $theUrl, |
|
| 503 | 4 | $vv['subCfg'], |
|
| 504 | 4 | $scheduledTime, |
|
| 505 | 4 | $configurationHash, |
|
| 506 | 4 | $skipInnerCheck |
|
| 507 | ); |
||
| 508 | 4 | if ($added === false) { |
|
| 509 | 4 | $urlList .= ' (Url already existed)'; |
|
| 510 | } |
||
| 511 | } elseif ($downloadCrawlUrls) { |
||
| 512 | $downloadUrls[$theUrl] = $theUrl; |
||
| 513 | } |
||
| 514 | |||
| 515 | 4 | $urlList .= '<br />'; |
|
| 516 | } |
||
| 517 | 4 | $duplicateTrack[$uKey] = true; |
|
| 518 | } |
||
| 519 | } |
||
| 520 | } else { |
||
| 521 | $urlList = 'ERROR - no URL generated'; |
||
| 522 | } |
||
| 523 | |||
| 524 | 4 | return $urlList; |
|
| 525 | } |
||
| 526 | |||
| 527 | /** |
||
| 528 | * Returns true if input processing instruction is among registered ones. |
||
| 529 | * |
||
| 530 | * @param string $piString PI to test |
||
| 531 | * @param array $incomingProcInstructions Processing instructions |
||
| 532 | * @return boolean |
||
| 533 | */ |
||
| 534 | 5 | public function drawURLs_PIfilter($piString, array $incomingProcInstructions) |
|
| 546 | |||
| 547 | 4 | public function getPageTSconfigForId($id) |
|
| 569 | |||
| 570 | /** |
||
| 571 | * This methods returns an array of configurations. |
||
| 572 | * And no urls! |
||
| 573 | * |
||
| 574 | * @param integer $id Page ID |
||
| 575 | * @param bool $forceSsl Use https |
||
| 576 | * @return array |
||
| 577 | * |
||
| 578 | * TODO: Should be switched back to protected - TNM 2018-11-16 |
||
| 579 | */ |
||
| 580 | 4 | public function getUrlsForPageId($id, $forceSsl = false) |
|
| 716 | |||
| 717 | /** |
||
| 718 | * Checks if a domain record exist and returns the base-url based on the record. If not the given baseUrl string is used. |
||
| 719 | * |
||
| 720 | * @param string $baseUrl |
||
| 721 | * @param integer $sysDomainUid |
||
| 722 | * @param bool $ssl |
||
| 723 | * @return string |
||
| 724 | */ |
||
| 725 | 4 | protected function getBaseUrlForConfigurationRecord($baseUrl, $sysDomainUid, $ssl = false) |
|
| 745 | |||
| 746 | public function getConfigurationsForBranch($rootid, $depth) |
||
| 791 | |||
| 792 | /** |
||
| 793 | * Check if a user has access to an item |
||
| 794 | * (e.g. get the group list of the current logged in user from $GLOBALS['TSFE']->gr_list) |
||
| 795 | * |
||
| 796 | * @see \TYPO3\CMS\Frontend\Page\PageRepository::getMultipleGroupsWhereClause() |
||
| 797 | * @param string $groupList Comma-separated list of (fe_)group UIDs from a user |
||
| 798 | * @param string $accessList Comma-separated list of (fe_)group UIDs of the item to access |
||
| 799 | * @return bool TRUE if at least one of the users group UIDs is in the access list or the access list is empty |
||
| 800 | */ |
||
| 801 | 3 | public function hasGroupAccess($groupList, $accessList) |
|
| 813 | |||
| 814 | /** |
||
| 815 | * Parse GET vars of input Query into array with key=>value pairs |
||
| 816 | * |
||
| 817 | * @param string $inputQuery Input query string |
||
| 818 | * @return array |
||
| 819 | */ |
||
| 820 | 7 | public function parseParams($inputQuery) |
|
| 835 | |||
| 836 | /** |
||
| 837 | * Will expand the parameters configuration to individual values. This follows a certain syntax of the value of each parameter. |
||
| 838 | * Syntax of values: |
||
| 839 | * - Basically: If the value is wrapped in [...] it will be expanded according to the following syntax, otherwise the value is taken literally |
||
| 840 | * - Configuration is splitted by "|" and the parts are processed individually and finally added together |
||
| 841 | * - For each configuration part: |
||
| 842 | * - "[int]-[int]" = Integer range, will be expanded to all values in between, values included, starting from low to high (max. 1000). Example "1-34" or "-40--30" |
||
| 843 | * - "_TABLE:[TCA table name];[_PID:[optional page id, default is current page]];[_ENABLELANG:1]" = Look up of table records from PID, filtering out deleted records. Example "_TABLE:tt_content; _PID:123" |
||
| 844 | * _ENABLELANG:1 picks only original records without their language overlays |
||
| 845 | * - Default: Literal value |
||
| 846 | * |
||
| 847 | * @param array $paramArray Array with key (GET var name) and values (value of GET var which is configuration for expansion) |
||
| 848 | * @param integer $pid Current page ID |
||
| 849 | * @return array |
||
| 850 | */ |
||
| 851 | 4 | public function expandParameters($paramArray, $pid) |
|
| 961 | |||
| 962 | /** |
||
| 963 | * Compiling URLs from parameter array (output of expandParameters()) |
||
| 964 | * The number of URLs will be the multiplication of the number of parameter values for each key |
||
| 965 | * |
||
| 966 | * @param array $paramArray Output of expandParameters(): Array with keys (GET var names) and for each an array of values |
||
| 967 | * @param array $urls URLs accumulated in this array (for recursion) |
||
| 968 | * @return array |
||
| 969 | */ |
||
| 970 | 7 | public function compileUrls($paramArray, $urls = []) |
|
| 995 | |||
| 996 | /************************************ |
||
| 997 | * |
||
| 998 | * Crawler log |
||
| 999 | * |
||
| 1000 | ************************************/ |
||
| 1001 | |||
| 1002 | /** |
||
| 1003 | * Return array of records from crawler queue for input page ID |
||
| 1004 | * |
||
| 1005 | * @param integer $id Page ID for which to look up log entries. |
||
| 1006 | * @param string$filter Filter: "all" => all entries, "pending" => all that is not yet run, "finished" => all complete ones |
||
| 1007 | * @param boolean $doFlush If TRUE, then entries selected at DELETED(!) instead of selected! |
||
| 1008 | * @param boolean $doFullFlush |
||
| 1009 | * @param integer $itemsPerPage Limit the amount of entries per page default is 10 |
||
| 1010 | * @return array |
||
| 1011 | */ |
||
| 1012 | 4 | public function getLogEntriesForPageId($id, $filter = '', $doFlush = false, $doFullFlush = false, $itemsPerPage = 10) |
|
| 1041 | |||
| 1042 | /** |
||
| 1043 | * Return array of records from crawler queue for input set ID |
||
| 1044 | * |
||
| 1045 | * @param integer $set_id Set ID for which to look up log entries. |
||
| 1046 | * @param string $filter Filter: "all" => all entries, "pending" => all that is not yet run, "finished" => all complete ones |
||
| 1047 | * @param boolean $doFlush If TRUE, then entries selected at DELETED(!) instead of selected! |
||
| 1048 | * @param integer $itemsPerPage Limit the amount of entires per page default is 10 |
||
| 1049 | * @return array |
||
| 1050 | */ |
||
| 1051 | 6 | public function getLogEntriesForSetId($set_id, $filter = '', $doFlush = false, $doFullFlush = false, $itemsPerPage = 10) |
|
| 1080 | |||
| 1081 | /** |
||
| 1082 | * Removes queue entries |
||
| 1083 | * |
||
| 1084 | * @param string $where SQL related filter for the entries which should be removed |
||
| 1085 | * @return void |
||
| 1086 | */ |
||
| 1087 | 10 | protected function flushQueue($where = '') |
|
| 1120 | |||
| 1121 | /** |
||
| 1122 | * Adding call back entries to log (called from hooks typically, see indexed search class "class.crawler.php" |
||
| 1123 | * |
||
| 1124 | * @param integer $setId Set ID |
||
| 1125 | * @param array $params Parameters to pass to call back function |
||
| 1126 | * @param string $callBack Call back object reference, eg. 'EXT:indexed_search/class.crawler.php:&tx_indexedsearch_crawler' |
||
| 1127 | * @param integer $page_id Page ID to attach it to |
||
| 1128 | * @param integer $schedule Time at which to activate |
||
| 1129 | * @return void |
||
| 1130 | */ |
||
| 1131 | public function addQueueEntry_callBack($setId, $params, $callBack, $page_id = 0, $schedule = 0) |
||
| 1150 | |||
| 1151 | /************************************ |
||
| 1152 | * |
||
| 1153 | * URL setting |
||
| 1154 | * |
||
| 1155 | ************************************/ |
||
| 1156 | |||
| 1157 | /** |
||
| 1158 | * Setting a URL for crawling: |
||
| 1159 | * |
||
| 1160 | * @param integer $id Page ID |
||
| 1161 | * @param string $url Complete URL |
||
| 1162 | * @param array $subCfg Sub configuration array (from TS config) |
||
| 1163 | * @param integer $tstamp Scheduled-time |
||
| 1164 | * @param string $configurationHash (optional) configuration hash |
||
| 1165 | * @param bool $skipInnerDuplicationCheck (optional) skip inner duplication check |
||
| 1166 | * @return bool |
||
| 1167 | */ |
||
| 1168 | 4 | public function addUrl( |
|
| 1169 | $id, |
||
| 1170 | $url, |
||
| 1171 | array $subCfg, |
||
| 1172 | $tstamp, |
||
| 1173 | $configurationHash = '', |
||
| 1174 | $skipInnerDuplicationCheck = false |
||
| 1175 | ) { |
||
| 1176 | 4 | $urlAdded = false; |
|
| 1177 | 4 | $rows = []; |
|
| 1178 | |||
| 1179 | // Creating parameters: |
||
| 1180 | $parameters = [ |
||
| 1181 | 4 | 'url' => $url |
|
| 1182 | ]; |
||
| 1183 | |||
| 1184 | // fe user group simulation: |
||
| 1185 | 4 | $uGs = implode(',', array_unique(GeneralUtility::intExplode(',', $subCfg['userGroups'], true))); |
|
| 1186 | 4 | if ($uGs) { |
|
| 1187 | $parameters['feUserGroupList'] = $uGs; |
||
| 1188 | } |
||
| 1189 | |||
| 1190 | // Setting processing instructions |
||
| 1191 | 4 | $parameters['procInstructions'] = GeneralUtility::trimExplode(',', $subCfg['procInstrFilter']); |
|
| 1192 | 4 | if (is_array($subCfg['procInstrParams.'])) { |
|
| 1193 | 4 | $parameters['procInstrParams'] = $subCfg['procInstrParams.']; |
|
| 1194 | } |
||
| 1195 | |||
| 1196 | // Possible TypoScript Template Parents |
||
| 1197 | 4 | $parameters['rootTemplatePid'] = $subCfg['rootTemplatePid']; |
|
| 1198 | |||
| 1199 | // Compile value array: |
||
| 1200 | 4 | $parameters_serialized = serialize($parameters); |
|
| 1201 | $fieldArray = [ |
||
| 1202 | 4 | 'page_id' => intval($id), |
|
| 1203 | 4 | 'parameters' => $parameters_serialized, |
|
| 1204 | 4 | 'parameters_hash' => GeneralUtility::shortMD5($parameters_serialized), |
|
| 1205 | 4 | 'configuration_hash' => $configurationHash, |
|
| 1206 | 4 | 'scheduled' => $tstamp, |
|
| 1207 | 4 | 'exec_time' => 0, |
|
| 1208 | 4 | 'set_id' => intval($this->setID), |
|
| 1209 | 4 | 'result_data' => '', |
|
| 1210 | 4 | 'configuration' => $subCfg['key'], |
|
| 1211 | ]; |
||
| 1212 | |||
| 1213 | 4 | if ($this->registerQueueEntriesInternallyOnly) { |
|
| 1214 | //the entries will only be registered and not stored to the database |
||
| 1215 | $this->queueEntries[] = $fieldArray; |
||
| 1216 | } else { |
||
| 1217 | 4 | if (!$skipInnerDuplicationCheck) { |
|
| 1218 | // check if there is already an equal entry |
||
| 1219 | 4 | $rows = $this->getDuplicateRowsIfExist($tstamp, $fieldArray); |
|
| 1220 | } |
||
| 1221 | |||
| 1222 | 4 | if (count($rows) == 0) { |
|
| 1223 | 4 | $this->db->exec_INSERTquery('tx_crawler_queue', $fieldArray); |
|
| 1224 | 4 | $uid = $this->db->sql_insert_id(); |
|
| 1225 | 4 | $rows[] = $uid; |
|
| 1226 | 4 | $urlAdded = true; |
|
| 1227 | |||
| 1228 | // The event dispatcher is deprecated since crawler v6.4.0, will be removed in crawler v7.0.0. |
||
| 1229 | // Please use the Signal instead. |
||
| 1230 | 4 | EventDispatcher::getInstance()->post('urlAddedToQueue', $this->setID, ['uid' => $uid, 'fieldArray' => $fieldArray]); |
|
| 1231 | |||
| 1232 | 4 | $signalPayload = ['uid' => $uid, 'fieldArray' => $fieldArray]; |
|
| 1233 | 4 | SignalSlotUtility::emitSignal( |
|
| 1234 | 4 | __CLASS__, |
|
| 1235 | 4 | SignalSlotUtility::SIGNAL_URL_ADDED_TO_QUEUE, |
|
| 1236 | 4 | $signalPayload |
|
| 1237 | ); |
||
| 1238 | |||
| 1239 | } else { |
||
| 1240 | // The event dispatcher is deprecated since crawler v6.4.0, will be removed in crawler v7.0.0. |
||
| 1241 | // Please use the Signal instead. |
||
| 1242 | 2 | EventDispatcher::getInstance()->post('duplicateUrlInQueue', $this->setID, ['rows' => $rows, 'fieldArray' => $fieldArray]); |
|
| 1243 | |||
| 1244 | 2 | $signalPayload = ['rows' => $rows, 'fieldArray' => $fieldArray]; |
|
| 1245 | 2 | SignalSlotUtility::emitSignal( |
|
| 1246 | 2 | __CLASS__, |
|
| 1247 | 2 | SignalSlotUtility::SIGNAL_DUPLICATE_URL_IN_QUEUE, |
|
| 1248 | 2 | $signalPayload |
|
| 1249 | ); |
||
| 1250 | } |
||
| 1251 | } |
||
| 1252 | |||
| 1253 | 4 | return $urlAdded; |
|
| 1254 | } |
||
| 1255 | |||
| 1256 | /** |
||
| 1257 | * This method determines duplicates for a queue entry with the same parameters and this timestamp. |
||
| 1258 | * If the timestamp is in the past, it will check if there is any unprocessed queue entry in the past. |
||
| 1259 | * If the timestamp is in the future it will check, if the queued entry has exactly the same timestamp |
||
| 1260 | * |
||
| 1261 | * @param int $tstamp |
||
| 1262 | * @param array $fieldArray |
||
| 1263 | * |
||
| 1264 | * @return array |
||
| 1265 | */ |
||
| 1266 | 4 | protected function getDuplicateRowsIfExist($tstamp, $fieldArray) |
|
| 1267 | { |
||
| 1268 | 4 | $rows = []; |
|
| 1269 | |||
| 1270 | 4 | $currentTime = $this->getCurrentTime(); |
|
| 1271 | |||
| 1272 | //if this entry is scheduled with "now" |
||
| 1273 | 4 | if ($tstamp <= $currentTime) { |
|
| 1274 | 1 | if ($this->extensionSettings['enableTimeslot']) { |
|
| 1275 | 1 | $timeBegin = $currentTime - 100; |
|
| 1276 | 1 | $timeEnd = $currentTime + 100; |
|
| 1277 | 1 | $where = ' ((scheduled BETWEEN ' . $timeBegin . ' AND ' . $timeEnd . ' ) OR scheduled <= ' . $currentTime . ') '; |
|
| 1278 | } else { |
||
| 1279 | 1 | $where = 'scheduled <= ' . $currentTime; |
|
| 1280 | } |
||
| 1281 | 3 | } elseif ($tstamp > $currentTime) { |
|
| 1282 | //entry with a timestamp in the future need to have the same schedule time |
||
| 1283 | 3 | $where = 'scheduled = ' . $tstamp ; |
|
| 1284 | } |
||
| 1285 | |||
| 1286 | 4 | if (!empty($where)) { |
|
| 1287 | 4 | $result = $this->db->exec_SELECTgetRows( |
|
| 1288 | 4 | 'qid', |
|
| 1289 | 4 | 'tx_crawler_queue', |
|
| 1290 | $where . |
||
| 1291 | 4 | ' AND NOT exec_time' . |
|
| 1292 | 4 | ' AND NOT process_id ' . |
|
| 1293 | 4 | ' AND page_id=' . intval($fieldArray['page_id']) . |
|
| 1294 | 4 | ' AND parameters_hash = ' . $this->db->fullQuoteStr($fieldArray['parameters_hash'], 'tx_crawler_queue') |
|
| 1295 | ); |
||
| 1296 | |||
| 1297 | 4 | if (is_array($result)) { |
|
| 1298 | 4 | foreach ($result as $value) { |
|
| 1299 | 2 | $rows[] = $value['qid']; |
|
| 1300 | } |
||
| 1301 | } |
||
| 1302 | } |
||
| 1303 | |||
| 1304 | 4 | return $rows; |
|
| 1305 | } |
||
| 1306 | |||
| 1307 | /** |
||
| 1308 | * Returns the current system time |
||
| 1309 | * |
||
| 1310 | * @return int |
||
| 1311 | */ |
||
| 1312 | public function getCurrentTime() |
||
| 1316 | |||
| 1317 | /************************************ |
||
| 1318 | * |
||
| 1319 | * URL reading |
||
| 1320 | * |
||
| 1321 | ************************************/ |
||
| 1322 | |||
| 1323 | /** |
||
| 1324 | * Read URL for single queue entry |
||
| 1325 | * |
||
| 1326 | * @param integer $queueId |
||
| 1327 | * @param boolean $force If set, will process even if exec_time has been set! |
||
| 1328 | * @return integer |
||
| 1329 | */ |
||
| 1330 | public function readUrl($queueId, $force = false) |
||
| 1331 | { |
||
| 1332 | $ret = 0; |
||
| 1333 | if ($this->debugMode) { |
||
| 1334 | GeneralUtility::devlog('crawler-readurl start ' . microtime(true), __FUNCTION__); |
||
| 1335 | } |
||
| 1336 | // Get entry: |
||
| 1337 | list($queueRec) = $this->db->exec_SELECTgetRows( |
||
| 1338 | '*', |
||
| 1339 | 'tx_crawler_queue', |
||
| 1340 | 'qid=' . intval($queueId) . ($force ? '' : ' AND exec_time=0 AND process_scheduled > 0') |
||
| 1341 | ); |
||
| 1342 | |||
| 1343 | if (!is_array($queueRec)) { |
||
| 1344 | return; |
||
| 1345 | } |
||
| 1346 | |||
| 1347 | $parameters = unserialize($queueRec['parameters']); |
||
| 1348 | if ($parameters['rootTemplatePid']) { |
||
| 1349 | $this->initTSFE((int)$parameters['rootTemplatePid']); |
||
| 1350 | } else { |
||
| 1351 | GeneralUtility::sysLog( |
||
| 1352 | 'Page with (' . $queueRec['page_id'] . ') could not be crawled, please check your crawler configuration. Perhaps no Root Template Pid is set', |
||
| 1353 | 'crawler', |
||
| 1354 | GeneralUtility::SYSLOG_SEVERITY_WARNING |
||
| 1355 | ); |
||
| 1356 | } |
||
| 1357 | |||
| 1358 | $signalPayload = [$queueId, $queueRec]; |
||
| 1359 | SignalSlotUtility::emitSignal( |
||
| 1360 | __CLASS__, |
||
| 1361 | SignalSlotUtility::SIGNAL_QUEUEITEM_PREPROCESS, |
||
| 1362 | $signalPayload |
||
| 1363 | ); |
||
| 1364 | |||
| 1365 | // Set exec_time to lock record: |
||
| 1366 | $field_array = ['exec_time' => $this->getCurrentTime()]; |
||
| 1367 | |||
| 1368 | if (isset($this->processID)) { |
||
| 1369 | //if mulitprocessing is used we need to store the id of the process which has handled this entry |
||
| 1370 | $field_array['process_id_completed'] = $this->processID; |
||
| 1371 | } |
||
| 1372 | $this->db->exec_UPDATEquery('tx_crawler_queue', 'qid=' . intval($queueId), $field_array); |
||
| 1373 | |||
| 1374 | $result = $this->readUrl_exec($queueRec); |
||
| 1375 | $resultData = unserialize($result['content']); |
||
| 1376 | |||
| 1377 | //atm there's no need to point to specific pollable extensions |
||
| 1378 | if (is_array($GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['crawler']['pollSuccess'])) { |
||
| 1379 | foreach ($GLOBALS['TYPO3_CONF_VARS']['EXTCONF']['crawler']['pollSuccess'] as $pollable) { |
||
| 1380 | // only check the success value if the instruction is runnig |
||
| 1381 | // it is important to name the pollSuccess key same as the procInstructions key |
||
| 1382 | if (is_array($resultData['parameters']['procInstructions']) && in_array( |
||
| 1383 | $pollable, |
||
| 1384 | $resultData['parameters']['procInstructions'] |
||
| 1385 | ) |
||
| 1386 | ) { |
||
| 1387 | if (!empty($resultData['success'][$pollable]) && $resultData['success'][$pollable]) { |
||
| 1388 | $ret |= self::CLI_STATUS_POLLABLE_PROCESSED; |
||
| 1389 | } |
||
| 1390 | } |
||
| 1391 | } |
||
| 1392 | } |
||
| 1393 | |||
| 1394 | // Set result in log which also denotes the end of the processing of this entry. |
||
| 1395 | $field_array = ['result_data' => serialize($result)]; |
||
| 1396 | |||
| 1397 | $signalPayload = [$queueId, $field_array]; |
||
| 1398 | SignalSlotUtility::emitSignal( |
||
| 1399 | __CLASS__, |
||
| 1400 | SignalSlotUtility::SIGNAL_QUEUEITEM_POSTPROCESS, |
||
| 1401 | $signalPayload |
||
| 1402 | ); |
||
| 1403 | |||
| 1404 | $this->db->exec_UPDATEquery('tx_crawler_queue', 'qid=' . intval($queueId), $field_array); |
||
| 1405 | |||
| 1406 | if ($this->debugMode) { |
||
| 1407 | GeneralUtility::devlog('crawler-readurl stop ' . microtime(true), __FUNCTION__); |
||
| 1408 | } |
||
| 1409 | |||
| 1410 | return $ret; |
||
| 1411 | } |
||
| 1412 | |||
| 1413 | /** |
||
| 1414 | * Read URL for not-yet-inserted log-entry |
||
| 1415 | * |
||
| 1416 | * @param array $field_array Queue field array, |
||
| 1417 | * |
||
| 1418 | * @return string |
||
| 1419 | */ |
||
| 1420 | public function readUrlFromArray($field_array) |
||
| 1421 | { |
||
| 1422 | |||
| 1423 | // Set exec_time to lock record: |
||
| 1424 | $field_array['exec_time'] = $this->getCurrentTime(); |
||
| 1425 | $this->db->exec_INSERTquery('tx_crawler_queue', $field_array); |
||
| 1426 | $queueId = $field_array['qid'] = $this->db->sql_insert_id(); |
||
| 1427 | |||
| 1428 | $result = $this->readUrl_exec($field_array); |
||
| 1429 | |||
| 1430 | // Set result in log which also denotes the end of the processing of this entry. |
||
| 1431 | $field_array = ['result_data' => serialize($result)]; |
||
| 1432 | |||
| 1433 | $signalPayload = [$queueId, $field_array]; |
||
| 1434 | SignalSlotUtility::emitSignal( |
||
| 1435 | __CLASS__, |
||
| 1436 | SignalSlotUtility::SIGNAL_QUEUEITEM_POSTPROCESS, |
||
| 1437 | $signalPayload |
||
| 1438 | ); |
||
| 1439 | |||
| 1440 | $this->db->exec_UPDATEquery('tx_crawler_queue', 'qid=' . intval($queueId), $field_array); |
||
| 1441 | |||
| 1442 | return $result; |
||
| 1443 | } |
||
| 1444 | |||
| 1445 | /** |
||
| 1446 | * Read URL for a queue record |
||
| 1447 | * |
||
| 1448 | * @param array $queueRec Queue record |
||
| 1449 | * @return string |
||
| 1450 | */ |
||
| 1451 | public function readUrl_exec($queueRec) |
||
| 1452 | { |
||
| 1453 | // Decode parameters: |
||
| 1454 | $parameters = unserialize($queueRec['parameters']); |
||
| 1455 | $result = 'ERROR'; |
||
| 1456 | if (is_array($parameters)) { |
||
| 1457 | if ($parameters['_CALLBACKOBJ']) { // Calling object: |
||
| 1458 | $objRef = $parameters['_CALLBACKOBJ']; |
||
| 1459 | $callBackObj = &GeneralUtility::getUserObj($objRef); |
||
| 1460 | if (is_object($callBackObj)) { |
||
| 1461 | unset($parameters['_CALLBACKOBJ']); |
||
| 1462 | $result = ['content' => serialize($callBackObj->crawler_execute($parameters, $this))]; |
||
| 1463 | } else { |
||
| 1464 | $result = ['content' => 'No object: ' . $objRef]; |
||
| 1465 | } |
||
| 1466 | } else { // Regular FE request: |
||
| 1467 | |||
| 1468 | // Prepare: |
||
| 1469 | $crawlerId = $queueRec['qid'] . ':' . md5($queueRec['qid'] . '|' . $queueRec['set_id'] . '|' . $GLOBALS['TYPO3_CONF_VARS']['SYS']['encryptionKey']); |
||
| 1470 | |||
| 1471 | // Get result: |
||
| 1472 | $result = $this->requestUrl($parameters['url'], $crawlerId); |
||
| 1473 | |||
| 1474 | // The event dispatcher is deprecated since crawler v6.4.0, will be removed in crawler v7.0.0. |
||
| 1475 | // Please use the Signal instead. |
||
| 1476 | EventDispatcher::getInstance()->post('urlCrawled', $queueRec['set_id'], ['url' => $parameters['url'], 'result' => $result]); |
||
| 1477 | |||
| 1478 | $signalPayload = ['url' => $parameters['url'], 'result' => $result]; |
||
| 1479 | SignalSlotUtility::emitSignal( |
||
| 1480 | __CLASS__, |
||
| 1481 | SignalSlotUtility::SIGNAL_URL_CRAWLED, |
||
| 1482 | $signalPayload |
||
| 1483 | ); |
||
| 1484 | } |
||
| 1485 | } |
||
| 1486 | |||
| 1487 | return $result; |
||
| 1488 | } |
||
| 1489 | |||
| 1490 | /** |
||
| 1491 | * Gets the content of a URL. |
||
| 1492 | * |
||
| 1493 | * @param string $originalUrl URL to read |
||
| 1494 | * @param string $crawlerId Crawler ID string (qid + hash to verify) |
||
| 1495 | * @param integer $timeout Timeout time |
||
| 1496 | * @param integer $recursion Recursion limiter for 302 redirects |
||
| 1497 | * @return array |
||
| 1498 | */ |
||
| 1499 | 2 | public function requestUrl($originalUrl, $crawlerId, $timeout = 2, $recursion = 10) |
|
| 1592 | |||
| 1593 | /** |
||
| 1594 | * Gets the base path of the website frontend. |
||
| 1595 | * (e.g. if you call http://mydomain.com/cms/index.php in |
||
| 1596 | * the browser the base path is "/cms/") |
||
| 1597 | * |
||
| 1598 | * @return string Base path of the website frontend |
||
| 1599 | */ |
||
| 1600 | protected function getFrontendBasePath() |
||
| 1601 | { |
||
| 1602 | $frontendBasePath = '/'; |
||
| 1603 | |||
| 1604 | // Get the path from the extension settings: |
||
| 1605 | if (isset($this->extensionSettings['frontendBasePath']) && $this->extensionSettings['frontendBasePath']) { |
||
| 1606 | $frontendBasePath = $this->extensionSettings['frontendBasePath']; |
||
| 1607 | // If empty, try to use config.absRefPrefix: |
||
| 1608 | } elseif (isset($GLOBALS['TSFE']->absRefPrefix) && !empty($GLOBALS['TSFE']->absRefPrefix)) { |
||
| 1609 | $frontendBasePath = $GLOBALS['TSFE']->absRefPrefix; |
||
| 1610 | // If not in CLI mode the base path can be determined from $_SERVER environment: |
||
| 1611 | } elseif (!defined('TYPO3_REQUESTTYPE_CLI') || !TYPO3_REQUESTTYPE_CLI) { |
||
| 1612 | $frontendBasePath = GeneralUtility::getIndpEnv('TYPO3_SITE_PATH'); |
||
| 1613 | } |
||
| 1614 | |||
| 1615 | // Base path must be '/<pathSegements>/': |
||
| 1616 | if ($frontendBasePath != '/') { |
||
| 1617 | $frontendBasePath = '/' . ltrim($frontendBasePath, '/'); |
||
| 1618 | $frontendBasePath = rtrim($frontendBasePath, '/') . '/'; |
||
| 1619 | } |
||
| 1620 | |||
| 1621 | return $frontendBasePath; |
||
| 1622 | } |
||
| 1623 | |||
| 1624 | /** |
||
| 1625 | * Executes a shell command and returns the outputted result. |
||
| 1626 | * |
||
| 1627 | * @param string $command Shell command to be executed |
||
| 1628 | * @return string Outputted result of the command execution |
||
| 1629 | */ |
||
| 1630 | protected function executeShellCommand($command) |
||
| 1631 | { |
||
| 1632 | $result = shell_exec($command); |
||
| 1633 | return $result; |
||
| 1634 | } |
||
| 1635 | |||
| 1636 | /** |
||
| 1637 | * Reads HTTP response from the given stream. |
||
| 1638 | * |
||
| 1639 | * @param resource $streamPointer Pointer to connection stream. |
||
| 1640 | * @return array Associative array with the following items: |
||
| 1641 | * headers <array> Response headers sent by server. |
||
| 1642 | * content <array> Content, with each line as an array item. |
||
| 1643 | */ |
||
| 1644 | 1 | protected function getHttpResponseFromStream($streamPointer) |
|
| 1645 | { |
||
| 1646 | 1 | $response = ['headers' => [], 'content' => []]; |
|
| 1647 | |||
| 1648 | 1 | if (is_resource($streamPointer)) { |
|
| 1649 | // read headers |
||
| 1650 | 1 | while ($line = fgets($streamPointer, '2048')) { |
|
| 1651 | 1 | $line = trim($line); |
|
| 1652 | 1 | if ($line !== '') { |
|
| 1653 | 1 | $response['headers'][] = $line; |
|
| 1654 | } else { |
||
| 1655 | 1 | break; |
|
| 1656 | } |
||
| 1657 | } |
||
| 1658 | |||
| 1659 | // read content |
||
| 1660 | 1 | while ($line = fgets($streamPointer, '2048')) { |
|
| 1661 | 1 | $response['content'][] = $line; |
|
| 1662 | } |
||
| 1663 | } |
||
| 1664 | |||
| 1665 | 1 | return $response; |
|
| 1666 | } |
||
| 1667 | |||
| 1668 | /** |
||
| 1669 | * @param message |
||
| 1670 | */ |
||
| 1671 | 2 | protected function log($message) |
|
| 1672 | { |
||
| 1673 | 2 | if (!empty($this->extensionSettings['logFileName'])) { |
|
| 1674 | $fileResult = @file_put_contents($this->extensionSettings['logFileName'], date('Ymd His') . ' ' . $message . PHP_EOL, FILE_APPEND); |
||
| 1675 | if (!$fileResult) { |
||
| 1676 | GeneralUtility::devLog('File "' . $this->extensionSettings['logFileName'] . '" could not be written, please check file permissions.', 'crawler', LogLevel::INFO); |
||
| 1677 | } |
||
| 1678 | } |
||
| 1679 | 2 | } |
|
| 1680 | |||
| 1681 | /** |
||
| 1682 | * Builds HTTP request headers. |
||
| 1683 | * |
||
| 1684 | * @param array $url |
||
| 1685 | * @param string $crawlerId |
||
| 1686 | * |
||
| 1687 | * @return array |
||
| 1688 | */ |
||
| 1689 | 6 | protected function buildRequestHeaderArray(array $url, $crawlerId) |
|
| 1690 | { |
||
| 1691 | 6 | $reqHeaders = []; |
|
| 1692 | 6 | $reqHeaders[] = 'GET ' . $url['path'] . ($url['query'] ? '?' . $url['query'] : '') . ' HTTP/1.0'; |
|
| 1693 | 6 | $reqHeaders[] = 'Host: ' . $url['host']; |
|
| 1694 | 6 | if (stristr($url['query'], 'ADMCMD_previewWS')) { |
|
| 1695 | 2 | $reqHeaders[] = 'Cookie: $Version="1"; be_typo_user="1"; $Path=/'; |
|
| 1696 | } |
||
| 1697 | 6 | $reqHeaders[] = 'Connection: close'; |
|
| 1698 | 6 | if ($url['user'] != '') { |
|
| 1699 | 2 | $reqHeaders[] = 'Authorization: Basic ' . base64_encode($url['user'] . ':' . $url['pass']); |
|
| 1700 | } |
||
| 1701 | 6 | $reqHeaders[] = 'X-T3crawler: ' . $crawlerId; |
|
| 1702 | 6 | $reqHeaders[] = 'User-Agent: TYPO3 crawler'; |
|
| 1703 | 6 | return $reqHeaders; |
|
| 1704 | } |
||
| 1705 | |||
| 1706 | /** |
||
| 1707 | * Check if the submitted HTTP-Header contains a redirect location and built new crawler-url |
||
| 1708 | * |
||
| 1709 | * @param array $headers HTTP Header |
||
| 1710 | * @param string $user HTTP Auth. User |
||
| 1711 | * @param string $pass HTTP Auth. Password |
||
| 1712 | * @return bool|string |
||
| 1713 | */ |
||
| 1714 | 12 | protected function getRequestUrlFrom302Header($headers, $user = '', $pass = '') |
|
| 1715 | { |
||
| 1716 | 12 | $header = []; |
|
| 1717 | 12 | if (!is_array($headers)) { |
|
| 1718 | 1 | return false; |
|
| 1719 | } |
||
| 1720 | 11 | if (!(stristr($headers[0], '301 Moved') || stristr($headers[0], '302 Found') || stristr($headers[0], '302 Moved'))) { |
|
| 1721 | 2 | return false; |
|
| 1722 | } |
||
| 1723 | |||
| 1724 | 9 | foreach ($headers as $hl) { |
|
| 1725 | 9 | $tmp = explode(": ", $hl); |
|
| 1726 | 9 | $header[trim($tmp[0])] = trim($tmp[1]); |
|
| 1727 | 9 | if (trim($tmp[0]) == 'Location') { |
|
| 1728 | 9 | break; |
|
| 1729 | } |
||
| 1730 | } |
||
| 1731 | 9 | if (!array_key_exists('Location', $header)) { |
|
| 1732 | 3 | return false; |
|
| 1733 | } |
||
| 1734 | |||
| 1735 | 6 | if ($user != '') { |
|
| 1736 | 3 | if (!($tmp = parse_url($header['Location']))) { |
|
| 1737 | 1 | return false; |
|
| 1738 | } |
||
| 1739 | 2 | $newUrl = $tmp['scheme'] . '://' . $user . ':' . $pass . '@' . $tmp['host'] . $tmp['path']; |
|
| 1740 | 2 | if ($tmp['query'] != '') { |
|
| 1741 | 2 | $newUrl .= '?' . $tmp['query']; |
|
| 1742 | } |
||
| 1743 | } else { |
||
| 1744 | 3 | $newUrl = $header['Location']; |
|
| 1745 | } |
||
| 1746 | 5 | return $newUrl; |
|
| 1747 | } |
||
| 1748 | |||
| 1749 | /************************** |
||
| 1750 | * |
||
| 1751 | * tslib_fe hooks: |
||
| 1752 | * |
||
| 1753 | **************************/ |
||
| 1754 | |||
| 1755 | /** |
||
| 1756 | * Initialization hook (called after database connection) |
||
| 1757 | * Takes the "HTTP_X_T3CRAWLER" header and looks up queue record and verifies if the session comes from the system (by comparing hashes) |
||
| 1758 | * |
||
| 1759 | * @param array $params Parameters from frontend |
||
| 1760 | * @param object $ref TSFE object (reference under PHP5) |
||
| 1761 | * @return void |
||
| 1762 | * |
||
| 1763 | * FIXME: Look like this is not used, in commit 9910d3f40cce15f4e9b7bcd0488bf21f31d53ebc it's added as public, |
||
| 1764 | * FIXME: I think this can be removed. (TNM) |
||
| 1765 | */ |
||
| 1766 | public function fe_init(&$params, $ref) |
||
| 1767 | { |
||
| 1768 | // Authenticate crawler request: |
||
| 1769 | if (isset($_SERVER['HTTP_X_T3CRAWLER'])) { |
||
| 1770 | list($queueId, $hash) = explode(':', $_SERVER['HTTP_X_T3CRAWLER']); |
||
| 1771 | list($queueRec) = $this->db->exec_SELECTgetSingleRow('*', 'tx_crawler_queue', 'qid=' . intval($queueId)); |
||
| 1772 | |||
| 1773 | // If a crawler record was found and hash was matching, set it up: |
||
| 1774 | if (is_array($queueRec) && $hash === md5($queueRec['qid'] . '|' . $queueRec['set_id'] . '|' . $GLOBALS['TYPO3_CONF_VARS']['SYS']['encryptionKey'])) { |
||
| 1775 | $params['pObj']->applicationData['tx_crawler']['running'] = true; |
||
| 1776 | $params['pObj']->applicationData['tx_crawler']['parameters'] = unserialize($queueRec['parameters']); |
||
| 1777 | $params['pObj']->applicationData['tx_crawler']['log'] = []; |
||
| 1778 | } else { |
||
| 1779 | die('No crawler entry found!'); |
||
| 1780 | } |
||
| 1781 | } |
||
| 1782 | } |
||
| 1783 | |||
| 1784 | /***************************** |
||
| 1785 | * |
||
| 1786 | * Compiling URLs to crawl - tools |
||
| 1787 | * |
||
| 1788 | *****************************/ |
||
| 1789 | |||
| 1790 | /** |
||
| 1791 | * @param integer $id Root page id to start from. |
||
| 1792 | * @param integer $depth Depth of tree, 0=only id-page, 1= on sublevel, 99 = infinite |
||
| 1793 | * @param integer $scheduledTime Unix Time when the URL is timed to be visited when put in queue |
||
| 1794 | * @param integer $reqMinute Number of requests per minute (creates the interleave between requests) |
||
| 1795 | * @param boolean $submitCrawlUrls If set, submits the URLs to queue in database (real crawling) |
||
| 1796 | * @param boolean $downloadCrawlUrls If set (and submitcrawlUrls is false) will fill $downloadUrls with entries) |
||
| 1797 | * @param array $incomingProcInstructions Array of processing instructions |
||
| 1798 | * @param array $configurationSelection Array of configuration keys |
||
| 1799 | * @return string |
||
| 1800 | */ |
||
| 1801 | public function getPageTreeAndUrls( |
||
| 1802 | $id, |
||
| 1803 | $depth, |
||
| 1804 | $scheduledTime, |
||
| 1805 | $reqMinute, |
||
| 1806 | $submitCrawlUrls, |
||
| 1807 | $downloadCrawlUrls, |
||
| 1808 | array $incomingProcInstructions, |
||
| 1809 | array $configurationSelection |
||
| 1810 | ) { |
||
| 1811 | global $BACK_PATH; |
||
| 1812 | global $LANG; |
||
| 1813 | if (!is_object($LANG)) { |
||
| 1814 | $LANG = GeneralUtility::makeInstance(LanguageService::class); |
||
| 1815 | $LANG->init(0); |
||
| 1816 | } |
||
| 1817 | $this->scheduledTime = $scheduledTime; |
||
| 1818 | $this->reqMinute = $reqMinute; |
||
| 1819 | $this->submitCrawlUrls = $submitCrawlUrls; |
||
| 1820 | $this->downloadCrawlUrls = $downloadCrawlUrls; |
||
| 1821 | $this->incomingProcInstructions = $incomingProcInstructions; |
||
| 1822 | $this->incomingConfigurationSelection = $configurationSelection; |
||
| 1823 | |||
| 1824 | $this->duplicateTrack = []; |
||
| 1825 | $this->downloadUrls = []; |
||
| 1826 | |||
| 1827 | // Drawing tree: |
||
| 1828 | /* @var PageTreeView $tree */ |
||
| 1829 | $tree = GeneralUtility::makeInstance(PageTreeView::class); |
||
| 1830 | $perms_clause = $GLOBALS['BE_USER']->getPagePermsClause(1); |
||
| 1831 | $tree->init('AND ' . $perms_clause); |
||
| 1832 | |||
| 1833 | $pageInfo = BackendUtility::readPageAccess($id, $perms_clause); |
||
| 1834 | if (is_array($pageInfo)) { |
||
| 1835 | // Set root row: |
||
| 1836 | $tree->tree[] = [ |
||
| 1837 | 'row' => $pageInfo, |
||
| 1838 | 'HTML' => IconUtility::getIconForRecord('pages', $pageInfo) |
||
| 1839 | ]; |
||
| 1840 | } |
||
| 1841 | |||
| 1842 | // Get branch beneath: |
||
| 1843 | if ($depth) { |
||
| 1844 | $tree->getTree($id, $depth, ''); |
||
| 1845 | } |
||
| 1846 | |||
| 1847 | // Traverse page tree: |
||
| 1848 | $code = ''; |
||
| 1849 | |||
| 1850 | foreach ($tree->tree as $data) { |
||
| 1851 | $this->MP = false; |
||
| 1852 | |||
| 1853 | // recognize mount points |
||
| 1854 | if ($data['row']['doktype'] == 7) { |
||
| 1855 | $mountpage = $this->db->exec_SELECTgetRows('*', 'pages', 'uid = ' . $data['row']['uid']); |
||
| 1856 | |||
| 1857 | // fetch mounted pages |
||
| 1858 | $this->MP = $mountpage[0]['mount_pid'] . '-' . $data['row']['uid']; |
||
| 1859 | |||
| 1860 | $mountTree = GeneralUtility::makeInstance(PageTreeView::class); |
||
| 1861 | $mountTree->init('AND ' . $perms_clause); |
||
| 1862 | $mountTree->getTree($mountpage[0]['mount_pid'], $depth, ''); |
||
| 1863 | |||
| 1864 | foreach ($mountTree->tree as $mountData) { |
||
| 1865 | $code .= $this->drawURLs_addRowsForPage( |
||
| 1866 | $mountData['row'], |
||
| 1867 | $mountData['HTML'] . BackendUtility::getRecordTitle('pages', $mountData['row'], true) |
||
| 1868 | ); |
||
| 1869 | } |
||
| 1870 | |||
| 1871 | // replace page when mount_pid_ol is enabled |
||
| 1872 | if ($mountpage[0]['mount_pid_ol']) { |
||
| 1873 | $data['row']['uid'] = $mountpage[0]['mount_pid']; |
||
| 1874 | } else { |
||
| 1875 | // if the mount_pid_ol is not set the MP must not be used for the mountpoint page |
||
| 1876 | $this->MP = false; |
||
| 1877 | } |
||
| 1878 | } |
||
| 1879 | |||
| 1880 | $code .= $this->drawURLs_addRowsForPage( |
||
| 1881 | $data['row'], |
||
| 1882 | $data['HTML'] . BackendUtility::getRecordTitle('pages', $data['row'], true) |
||
| 1883 | ); |
||
| 1884 | } |
||
| 1885 | |||
| 1886 | return $code; |
||
| 1887 | } |
||
| 1888 | |||
| 1889 | /** |
||
| 1890 | * Expands exclude string |
||
| 1891 | * |
||
| 1892 | * @param string $excludeString Exclude string |
||
| 1893 | * @return array |
||
| 1894 | */ |
||
| 1895 | 1 | public function expandExcludeString($excludeString) |
|
| 1896 | { |
||
| 1897 | // internal static caches; |
||
| 1898 | 1 | static $expandedExcludeStringCache; |
|
| 1899 | 1 | static $treeCache; |
|
| 1900 | |||
| 1901 | 1 | if (empty($expandedExcludeStringCache[$excludeString])) { |
|
| 1902 | 1 | $pidList = []; |
|
| 1903 | |||
| 1904 | 1 | if (!empty($excludeString)) { |
|
| 1905 | /** @var PageTreeView $tree */ |
||
| 1906 | $tree = GeneralUtility::makeInstance(PageTreeView::class); |
||
| 1907 | $tree->init('AND ' . $this->backendUser->getPagePermsClause(1)); |
||
| 1908 | |||
| 1909 | $excludeParts = GeneralUtility::trimExplode(',', $excludeString); |
||
| 1910 | |||
| 1911 | foreach ($excludeParts as $excludePart) { |
||
| 1912 | list($pid, $depth) = GeneralUtility::trimExplode('+', $excludePart); |
||
| 1913 | |||
| 1914 | // default is "page only" = "depth=0" |
||
| 1915 | if (empty($depth)) { |
||
| 1916 | $depth = (stristr($excludePart, '+')) ? 99 : 0; |
||
| 1917 | } |
||
| 1918 | |||
| 1919 | $pidList[] = $pid; |
||
| 1920 | |||
| 1921 | if ($depth > 0) { |
||
| 1922 | if (empty($treeCache[$pid][$depth])) { |
||
| 1923 | $tree->reset(); |
||
| 1924 | $tree->getTree($pid, $depth); |
||
| 1925 | $treeCache[$pid][$depth] = $tree->tree; |
||
| 1926 | } |
||
| 1927 | |||
| 1928 | foreach ($treeCache[$pid][$depth] as $data) { |
||
| 1929 | $pidList[] = $data['row']['uid']; |
||
| 1930 | } |
||
| 1931 | } |
||
| 1932 | } |
||
| 1933 | } |
||
| 1934 | |||
| 1935 | 1 | $expandedExcludeStringCache[$excludeString] = array_unique($pidList); |
|
| 1936 | } |
||
| 1937 | |||
| 1938 | 1 | return $expandedExcludeStringCache[$excludeString]; |
|
| 1939 | } |
||
| 1940 | |||
| 1941 | /** |
||
| 1942 | * Create the rows for display of the page tree |
||
| 1943 | * For each page a number of rows are shown displaying GET variable configuration |
||
| 1944 | * |
||
| 1945 | * @param array Page row |
||
| 1946 | * @param string Page icon and title for row |
||
| 1947 | * @return string HTML <tr> content (one or more) |
||
| 1948 | */ |
||
| 1949 | public function drawURLs_addRowsForPage(array $pageRow, $pageTitleAndIcon) |
||
| 1950 | { |
||
| 1951 | $skipMessage = ''; |
||
| 1952 | |||
| 1953 | // Get list of configurations |
||
| 1954 | $configurations = $this->getUrlsForPageRow($pageRow, $skipMessage); |
||
| 1955 | |||
| 1956 | if (count($this->incomingConfigurationSelection) > 0) { |
||
| 1957 | // remove configuration that does not match the current selection |
||
| 1958 | foreach ($configurations as $confKey => $confArray) { |
||
| 1959 | if (!in_array($confKey, $this->incomingConfigurationSelection)) { |
||
| 1960 | unset($configurations[$confKey]); |
||
| 1961 | } |
||
| 1962 | } |
||
| 1963 | } |
||
| 1964 | |||
| 1965 | // Traverse parameter combinations: |
||
| 1966 | $c = 0; |
||
| 1967 | $content = ''; |
||
| 1968 | if (count($configurations)) { |
||
| 1969 | foreach ($configurations as $confKey => $confArray) { |
||
| 1970 | |||
| 1971 | // Title column: |
||
| 1972 | if (!$c) { |
||
| 1973 | $titleClm = '<td rowspan="' . count($configurations) . '">' . $pageTitleAndIcon . '</td>'; |
||
| 1974 | } else { |
||
| 1975 | $titleClm = ''; |
||
| 1976 | } |
||
| 1977 | |||
| 1978 | if (!in_array($pageRow['uid'], $this->expandExcludeString($confArray['subCfg']['exclude']))) { |
||
| 1979 | |||
| 1980 | // URL list: |
||
| 1981 | $urlList = $this->urlListFromUrlArray( |
||
| 1982 | $confArray, |
||
| 1983 | $pageRow, |
||
| 1984 | $this->scheduledTime, |
||
| 1985 | $this->reqMinute, |
||
| 1986 | $this->submitCrawlUrls, |
||
| 1987 | $this->downloadCrawlUrls, |
||
| 1988 | $this->duplicateTrack, |
||
| 1989 | $this->downloadUrls, |
||
| 1990 | $this->incomingProcInstructions // if empty the urls won't be filtered by processing instructions |
||
| 1991 | ); |
||
| 1992 | |||
| 1993 | // Expanded parameters: |
||
| 1994 | $paramExpanded = ''; |
||
| 1995 | $calcAccu = []; |
||
| 1996 | $calcRes = 1; |
||
| 1997 | foreach ($confArray['paramExpanded'] as $gVar => $gVal) { |
||
| 1998 | $paramExpanded .= ' |
||
| 1999 | <tr> |
||
| 2000 | <td class="bgColor4-20">' . htmlspecialchars('&' . $gVar . '=') . '<br/>' . |
||
| 2001 | '(' . count($gVal) . ')' . |
||
| 2002 | '</td> |
||
| 2003 | <td class="bgColor4" nowrap="nowrap">' . nl2br(htmlspecialchars(implode(chr(10), $gVal))) . '</td> |
||
| 2004 | </tr> |
||
| 2005 | '; |
||
| 2006 | $calcRes *= count($gVal); |
||
| 2007 | $calcAccu[] = count($gVal); |
||
| 2008 | } |
||
| 2009 | $paramExpanded = '<table class="lrPadding c-list param-expanded">' . $paramExpanded . '</table>'; |
||
| 2010 | $paramExpanded .= 'Comb: ' . implode('*', $calcAccu) . '=' . $calcRes; |
||
| 2011 | |||
| 2012 | // Options |
||
| 2013 | $optionValues = ''; |
||
| 2014 | if ($confArray['subCfg']['userGroups']) { |
||
| 2015 | $optionValues .= 'User Groups: ' . $confArray['subCfg']['userGroups'] . '<br/>'; |
||
| 2016 | } |
||
| 2017 | if ($confArray['subCfg']['baseUrl']) { |
||
| 2018 | $optionValues .= 'Base Url: ' . $confArray['subCfg']['baseUrl'] . '<br/>'; |
||
| 2019 | } |
||
| 2020 | if ($confArray['subCfg']['procInstrFilter']) { |
||
| 2021 | $optionValues .= 'ProcInstr: ' . $confArray['subCfg']['procInstrFilter'] . '<br/>'; |
||
| 2022 | } |
||
| 2023 | |||
| 2024 | // Compile row: |
||
| 2025 | $content .= ' |
||
| 2026 | <tr class="bgColor' . ($c % 2 ? '-20' : '-10') . '"> |
||
| 2027 | ' . $titleClm . ' |
||
| 2028 | <td>' . htmlspecialchars($confKey) . '</td> |
||
| 2029 | <td>' . nl2br(htmlspecialchars(rawurldecode(trim(str_replace('&', chr(10) . '&', GeneralUtility::implodeArrayForUrl('', $confArray['paramParsed'])))))) . '</td> |
||
| 2030 | <td>' . $paramExpanded . '</td> |
||
| 2031 | <td nowrap="nowrap">' . $urlList . '</td> |
||
| 2032 | <td nowrap="nowrap">' . $optionValues . '</td> |
||
| 2033 | <td nowrap="nowrap">' . DebugUtility::viewArray($confArray['subCfg']['procInstrParams.']) . '</td> |
||
| 2034 | </tr>'; |
||
| 2035 | } else { |
||
| 2036 | $content .= '<tr class="bgColor' . ($c % 2 ? '-20' : '-10') . '"> |
||
| 2037 | ' . $titleClm . ' |
||
| 2038 | <td>' . htmlspecialchars($confKey) . '</td> |
||
| 2039 | <td colspan="5"><em>No entries</em> (Page is excluded in this configuration)</td> |
||
| 2040 | </tr>'; |
||
| 2041 | } |
||
| 2042 | |||
| 2043 | $c++; |
||
| 2044 | } |
||
| 2045 | } else { |
||
| 2046 | $message = !empty($skipMessage) ? ' (' . $skipMessage . ')' : ''; |
||
| 2047 | |||
| 2048 | // Compile row: |
||
| 2049 | $content .= ' |
||
| 2050 | <tr class="bgColor-20" style="border-bottom: 1px solid black;"> |
||
| 2051 | <td>' . $pageTitleAndIcon . '</td> |
||
| 2052 | <td colspan="6"><em>No entries</em>' . $message . '</td> |
||
| 2053 | </tr>'; |
||
| 2054 | } |
||
| 2055 | |||
| 2056 | return $content; |
||
| 2057 | } |
||
| 2058 | |||
| 2059 | /***************************** |
||
| 2060 | * |
||
| 2061 | * CLI functions |
||
| 2062 | * |
||
| 2063 | *****************************/ |
||
| 2064 | |||
| 2065 | /** |
||
| 2066 | * Main function for running from Command Line PHP script (cron job) |
||
| 2067 | * See ext/crawler/cli/crawler_cli.phpsh for details |
||
| 2068 | * |
||
| 2069 | * @return int number of remaining items or false if error |
||
| 2070 | */ |
||
| 2071 | public function CLI_main() |
||
| 2072 | { |
||
| 2073 | $this->setAccessMode('cli'); |
||
| 2074 | $result = self::CLI_STATUS_NOTHING_PROCCESSED; |
||
| 2075 | $cliObj = GeneralUtility::makeInstance(CrawlerCommandLineController::class); |
||
| 2076 | |||
| 2077 | if (isset($cliObj->cli_args['-h']) || isset($cliObj->cli_args['--help'])) { |
||
| 2078 | $cliObj->cli_validateArgs(); |
||
| 2079 | $cliObj->cli_help(); |
||
| 2080 | exit; |
||
| 2081 | } |
||
| 2082 | |||
| 2083 | if (!$this->getDisabled() && $this->CLI_checkAndAcquireNewProcess($this->CLI_buildProcessId())) { |
||
| 2084 | $countInARun = $cliObj->cli_argValue('--countInARun') ? intval($cliObj->cli_argValue('--countInARun')) : $this->extensionSettings['countInARun']; |
||
| 2085 | // Seconds |
||
| 2086 | $sleepAfterFinish = $cliObj->cli_argValue('--sleepAfterFinish') ? intval($cliObj->cli_argValue('--sleepAfterFinish')) : $this->extensionSettings['sleepAfterFinish']; |
||
| 2087 | // Milliseconds |
||
| 2088 | $sleepTime = $cliObj->cli_argValue('--sleepTime') ? intval($cliObj->cli_argValue('--sleepTime')) : $this->extensionSettings['sleepTime']; |
||
| 2089 | |||
| 2090 | try { |
||
| 2091 | // Run process: |
||
| 2092 | $result = $this->CLI_run($countInARun, $sleepTime, $sleepAfterFinish); |
||
| 2093 | } catch (\Exception $e) { |
||
| 2094 | $this->CLI_debug(get_class($e) . ': ' . $e->getMessage()); |
||
| 2095 | $result = self::CLI_STATUS_ABORTED; |
||
| 2096 | } |
||
| 2097 | |||
| 2098 | // Cleanup |
||
| 2099 | $this->db->exec_DELETEquery('tx_crawler_process', 'assigned_items_count = 0'); |
||
| 2100 | |||
| 2101 | //TODO can't we do that in a clean way? |
||
| 2102 | $releaseStatus = $this->CLI_releaseProcesses($this->CLI_buildProcessId()); |
||
| 2103 | |||
| 2104 | $this->CLI_debug("Unprocessed Items remaining:" . $this->queueRepository->countUnprocessedItems() . " (" . $this->CLI_buildProcessId() . ")"); |
||
| 2105 | $result |= ($this->queueRepository->countUnprocessedItems() > 0 ? self::CLI_STATUS_REMAIN : self::CLI_STATUS_NOTHING_PROCCESSED); |
||
| 2106 | } else { |
||
| 2107 | $result |= self::CLI_STATUS_ABORTED; |
||
| 2108 | } |
||
| 2109 | |||
| 2110 | return $result; |
||
| 2111 | } |
||
| 2112 | |||
| 2113 | /** |
||
| 2114 | * Function executed by crawler_im.php cli script. |
||
| 2115 | * |
||
| 2116 | * @return void |
||
| 2117 | */ |
||
| 2118 | public function CLI_main_im() |
||
| 2119 | { |
||
| 2120 | $this->setAccessMode('cli_im'); |
||
| 2121 | |||
| 2122 | $cliObj = GeneralUtility::makeInstance(QueueCommandLineController::class); |
||
| 2123 | |||
| 2124 | // Force user to admin state and set workspace to "Live": |
||
| 2125 | $this->backendUser->user['admin'] = 1; |
||
| 2126 | $this->backendUser->setWorkspace(0); |
||
| 2127 | |||
| 2128 | // Print help |
||
| 2129 | if (!isset($cliObj->cli_args['_DEFAULT'][1])) { |
||
| 2130 | $cliObj->cli_validateArgs(); |
||
| 2131 | $cliObj->cli_help(); |
||
| 2132 | exit; |
||
| 2133 | } |
||
| 2134 | |||
| 2135 | $cliObj->cli_validateArgs(); |
||
| 2136 | |||
| 2137 | if ($cliObj->cli_argValue('-o') === 'exec') { |
||
| 2138 | $this->registerQueueEntriesInternallyOnly = true; |
||
| 2139 | } |
||
| 2140 | |||
| 2141 | if (isset($cliObj->cli_args['_DEFAULT'][2])) { |
||
| 2142 | // Crawler is called over TYPO3 BE |
||
| 2143 | $pageId = MathUtility::forceIntegerInRange($cliObj->cli_args['_DEFAULT'][2], 0); |
||
| 2144 | } else { |
||
| 2145 | // Crawler is called over cli |
||
| 2146 | $pageId = MathUtility::forceIntegerInRange($cliObj->cli_args['_DEFAULT'][1], 0); |
||
| 2147 | } |
||
| 2148 | |||
| 2149 | $configurationKeys = $this->getConfigurationKeys($cliObj); |
||
| 2150 | |||
| 2151 | if (!is_array($configurationKeys)) { |
||
| 2152 | $configurations = $this->getUrlsForPageId($pageId); |
||
| 2153 | if (is_array($configurations)) { |
||
| 2154 | $configurationKeys = array_keys($configurations); |
||
| 2155 | } else { |
||
| 2156 | $configurationKeys = []; |
||
| 2157 | } |
||
| 2158 | } |
||
| 2159 | |||
| 2160 | if ($cliObj->cli_argValue('-o') === 'queue' || $cliObj->cli_argValue('-o') === 'exec') { |
||
| 2161 | $reason = new Reason(); |
||
| 2162 | $reason->setReason(Reason::REASON_GUI_SUBMIT); |
||
| 2163 | $reason->setDetailText('The cli script of the crawler added to the queue'); |
||
| 2164 | |||
| 2165 | // The event dispatcher is deprecated since crawler v6.4.0, will be removed in crawler v7.0.0. |
||
| 2166 | // Please use the Signal instead. |
||
| 2167 | EventDispatcher::getInstance()->post( |
||
| 2168 | 'invokeQueueChange', |
||
| 2169 | $this->setID, |
||
| 2170 | ['reason' => $reason] |
||
| 2171 | ); |
||
| 2172 | |||
| 2173 | $signalPayload = ['reason' => $reason]; |
||
| 2174 | SignalSlotUtility::emitSignal( |
||
| 2175 | __CLASS__, |
||
| 2176 | SignalSlotUtility::SIGNAL_INVOKE_QUEUE_CHANGE, |
||
| 2177 | $signalPayload |
||
| 2178 | ); |
||
| 2179 | |||
| 2180 | } |
||
| 2181 | |||
| 2182 | if ($this->extensionSettings['cleanUpOldQueueEntries']) { |
||
| 2183 | $this->cleanUpOldQueueEntries(); |
||
| 2184 | } |
||
| 2185 | |||
| 2186 | $this->setID = (int) GeneralUtility::md5int(microtime()); |
||
| 2187 | $this->getPageTreeAndUrls( |
||
| 2188 | $pageId, |
||
| 2189 | MathUtility::forceIntegerInRange($cliObj->cli_argValue('-d'), 0, 99), |
||
| 2190 | $this->getCurrentTime(), |
||
| 2191 | MathUtility::forceIntegerInRange($cliObj->cli_isArg('-n') ? $cliObj->cli_argValue('-n') : 30, 1, 1000), |
||
| 2192 | $cliObj->cli_argValue('-o') === 'queue' || $cliObj->cli_argValue('-o') === 'exec', |
||
| 2193 | $cliObj->cli_argValue('-o') === 'url', |
||
| 2194 | GeneralUtility::trimExplode(',', $cliObj->cli_argValue('-proc'), true), |
||
| 2195 | $configurationKeys |
||
| 2196 | ); |
||
| 2197 | |||
| 2198 | if ($cliObj->cli_argValue('-o') === 'url') { |
||
| 2199 | $cliObj->cli_echo(implode(chr(10), $this->downloadUrls) . chr(10), true); |
||
| 2200 | } elseif ($cliObj->cli_argValue('-o') === 'exec') { |
||
| 2201 | $cliObj->cli_echo("Executing " . count($this->urlList) . " requests right away:\n\n"); |
||
| 2202 | $cliObj->cli_echo(implode(chr(10), $this->urlList) . chr(10)); |
||
| 2203 | $cliObj->cli_echo("\nProcessing:\n"); |
||
| 2204 | |||
| 2205 | foreach ($this->queueEntries as $queueRec) { |
||
| 2206 | $p = unserialize($queueRec['parameters']); |
||
| 2207 | $cliObj->cli_echo($p['url'] . ' (' . implode(',', $p['procInstructions']) . ') => '); |
||
| 2208 | |||
| 2209 | $result = $this->readUrlFromArray($queueRec); |
||
| 2210 | |||
| 2211 | $requestResult = unserialize($result['content']); |
||
| 2212 | if (is_array($requestResult)) { |
||
| 2213 | $resLog = is_array($requestResult['log']) ? chr(10) . chr(9) . chr(9) . implode(chr(10) . chr(9) . chr(9), $requestResult['log']) : ''; |
||
| 2214 | $cliObj->cli_echo('OK: ' . $resLog . chr(10)); |
||
| 2215 | } else { |
||
| 2216 | $cliObj->cli_echo('Error checking Crawler Result: ' . substr(preg_replace('/\s+/', ' ', strip_tags($result['content'])), 0, 30000) . '...' . chr(10)); |
||
| 2217 | } |
||
| 2218 | } |
||
| 2219 | } elseif ($cliObj->cli_argValue('-o') === 'queue') { |
||
| 2220 | $cliObj->cli_echo("Putting " . count($this->urlList) . " entries in queue:\n\n"); |
||
| 2221 | $cliObj->cli_echo(implode(chr(10), $this->urlList) . chr(10)); |
||
| 2222 | } else { |
||
| 2223 | $cliObj->cli_echo(count($this->urlList) . " entries found for processing. (Use -o to decide action):\n\n", true); |
||
| 2224 | $cliObj->cli_echo(implode(chr(10), $this->urlList) . chr(10), true); |
||
| 2225 | } |
||
| 2226 | } |
||
| 2227 | |||
| 2228 | /** |
||
| 2229 | * Function executed by crawler_im.php cli script. |
||
| 2230 | * |
||
| 2231 | * @return bool |
||
| 2232 | */ |
||
| 2233 | public function CLI_main_flush() |
||
| 2234 | { |
||
| 2235 | $this->setAccessMode('cli_flush'); |
||
| 2236 | $cliObj = GeneralUtility::makeInstance(FlushCommandLineController::class); |
||
| 2237 | |||
| 2238 | // Force user to admin state and set workspace to "Live": |
||
| 2239 | $this->backendUser->user['admin'] = 1; |
||
| 2240 | $this->backendUser->setWorkspace(0); |
||
| 2241 | |||
| 2242 | // Print help |
||
| 2243 | if (!isset($cliObj->cli_args['_DEFAULT'][1])) { |
||
| 2244 | $cliObj->cli_validateArgs(); |
||
| 2245 | $cliObj->cli_help(); |
||
| 2246 | exit; |
||
| 2247 | } |
||
| 2248 | |||
| 2249 | $cliObj->cli_validateArgs(); |
||
| 2250 | $pageId = MathUtility::forceIntegerInRange($cliObj->cli_args['_DEFAULT'][1], 0); |
||
| 2251 | $fullFlush = ($pageId == 0); |
||
| 2252 | |||
| 2253 | $mode = $cliObj->cli_argValue('-o'); |
||
| 2254 | |||
| 2255 | switch ($mode) { |
||
| 2256 | case 'all': |
||
| 2257 | $result = $this->getLogEntriesForPageId($pageId, '', true, $fullFlush); |
||
| 2258 | break; |
||
| 2259 | case 'finished': |
||
| 2260 | case 'pending': |
||
| 2261 | $result = $this->getLogEntriesForPageId($pageId, $mode, true, $fullFlush); |
||
| 2262 | break; |
||
| 2263 | default: |
||
| 2264 | $cliObj->cli_validateArgs(); |
||
| 2265 | $cliObj->cli_help(); |
||
| 2266 | $result = false; |
||
| 2267 | } |
||
| 2268 | |||
| 2269 | return $result !== false; |
||
| 2270 | } |
||
| 2271 | |||
| 2272 | /** |
||
| 2273 | * Obtains configuration keys from the CLI arguments |
||
| 2274 | * |
||
| 2275 | * @param QueueCommandLineController $cliObj |
||
| 2276 | * @return array |
||
| 2277 | * |
||
| 2278 | * @deprecated since crawler v6.3.0, will be removed in crawler v7.0.0. |
||
| 2279 | */ |
||
| 2280 | protected function getConfigurationKeys(QueueCommandLineController $cliObj) |
||
| 2281 | { |
||
| 2282 | $parameter = trim($cliObj->cli_argValue('-conf')); |
||
| 2283 | return ($parameter != '' ? GeneralUtility::trimExplode(',', $parameter) : []); |
||
| 2284 | } |
||
| 2285 | |||
| 2286 | /** |
||
| 2287 | * Running the functionality of the CLI (crawling URLs from queue) |
||
| 2288 | * |
||
| 2289 | * @param int $countInARun |
||
| 2290 | * @param int $sleepTime |
||
| 2291 | * @param int $sleepAfterFinish |
||
| 2292 | * @return string |
||
| 2293 | */ |
||
| 2294 | public function CLI_run($countInARun, $sleepTime, $sleepAfterFinish) |
||
| 2295 | { |
||
| 2296 | $result = 0; |
||
| 2297 | $counter = 0; |
||
| 2298 | |||
| 2299 | // First, run hooks: |
||
| 2300 | $this->CLI_runHooks(); |
||
| 2301 | |||
| 2302 | // Clean up the queue |
||
| 2303 | if (intval($this->extensionSettings['purgeQueueDays']) > 0) { |
||
| 2304 | $purgeDate = $this->getCurrentTime() - 24 * 60 * 60 * intval($this->extensionSettings['purgeQueueDays']); |
||
| 2305 | $del = $this->db->exec_DELETEquery( |
||
| 2306 | 'tx_crawler_queue', |
||
| 2307 | 'exec_time!=0 AND exec_time<' . $purgeDate |
||
| 2308 | ); |
||
| 2309 | if (false == $del) { |
||
| 2310 | GeneralUtility::devLog('Records could not be deleted.', 'crawler', LogLevel::INFO); |
||
| 2311 | } |
||
| 2312 | } |
||
| 2313 | |||
| 2314 | // Select entries: |
||
| 2315 | //TODO Shouldn't this reside within the transaction? |
||
| 2316 | $rows = $this->db->exec_SELECTgetRows( |
||
| 2317 | 'qid,scheduled', |
||
| 2318 | 'tx_crawler_queue', |
||
| 2319 | 'exec_time=0 |
||
| 2320 | AND process_scheduled= 0 |
||
| 2321 | AND scheduled<=' . $this->getCurrentTime(), |
||
| 2322 | '', |
||
| 2323 | 'scheduled, qid', |
||
| 2324 | intval($countInARun) |
||
| 2325 | ); |
||
| 2326 | |||
| 2327 | if (count($rows) > 0) { |
||
| 2328 | $quidList = []; |
||
| 2329 | |||
| 2330 | foreach ($rows as $r) { |
||
| 2331 | $quidList[] = $r['qid']; |
||
| 2332 | } |
||
| 2333 | |||
| 2334 | $processId = $this->CLI_buildProcessId(); |
||
| 2335 | |||
| 2336 | //reserve queue entries for process |
||
| 2337 | $this->db->sql_query('BEGIN'); |
||
| 2338 | //TODO make sure we're not taking assigned queue-entires |
||
| 2339 | $this->db->exec_UPDATEquery( |
||
| 2340 | 'tx_crawler_queue', |
||
| 2341 | 'qid IN (' . implode(',', $quidList) . ')', |
||
| 2342 | [ |
||
| 2343 | 'process_scheduled' => intval($this->getCurrentTime()), |
||
| 2344 | 'process_id' => $processId |
||
| 2345 | ] |
||
| 2346 | ); |
||
| 2347 | |||
| 2348 | //save the number of assigned queue entrys to determine who many have been processed later |
||
| 2349 | $numberOfAffectedRows = $this->db->sql_affected_rows(); |
||
| 2350 | $this->db->exec_UPDATEquery( |
||
| 2351 | 'tx_crawler_process', |
||
| 2352 | "process_id = '" . $processId . "'", |
||
| 2353 | [ |
||
| 2354 | 'assigned_items_count' => intval($numberOfAffectedRows) |
||
| 2355 | ] |
||
| 2356 | ); |
||
| 2357 | |||
| 2358 | if ($numberOfAffectedRows == count($quidList)) { |
||
| 2359 | $this->db->sql_query('COMMIT'); |
||
| 2360 | } else { |
||
| 2361 | $this->db->sql_query('ROLLBACK'); |
||
| 2362 | $this->CLI_debug("Nothing processed due to multi-process collision (" . $this->CLI_buildProcessId() . ")"); |
||
| 2363 | return ($result | self::CLI_STATUS_ABORTED); |
||
| 2364 | } |
||
| 2365 | |||
| 2366 | foreach ($rows as $r) { |
||
| 2367 | $result |= $this->readUrl($r['qid']); |
||
| 2368 | |||
| 2369 | $counter++; |
||
| 2370 | usleep(intval($sleepTime)); // Just to relax the system |
||
| 2371 | |||
| 2372 | // if during the start and the current read url the cli has been disable we need to return from the function |
||
| 2373 | // mark the process NOT as ended. |
||
| 2374 | if ($this->getDisabled()) { |
||
| 2375 | return ($result | self::CLI_STATUS_ABORTED); |
||
| 2376 | } |
||
| 2377 | |||
| 2378 | if (!$this->CLI_checkIfProcessIsActive($this->CLI_buildProcessId())) { |
||
| 2379 | $this->CLI_debug("conflict / timeout (" . $this->CLI_buildProcessId() . ")"); |
||
| 2380 | |||
| 2381 | //TODO might need an additional returncode |
||
| 2382 | $result |= self::CLI_STATUS_ABORTED; |
||
| 2383 | break; //possible timeout |
||
| 2384 | } |
||
| 2385 | } |
||
| 2386 | |||
| 2387 | sleep(intval($sleepAfterFinish)); |
||
| 2388 | |||
| 2389 | $msg = 'Rows: ' . $counter; |
||
| 2390 | $this->CLI_debug($msg . " (" . $this->CLI_buildProcessId() . ")"); |
||
| 2391 | } else { |
||
| 2392 | $this->CLI_debug("Nothing within queue which needs to be processed (" . $this->CLI_buildProcessId() . ")"); |
||
| 2393 | } |
||
| 2394 | |||
| 2395 | if ($counter > 0) { |
||
| 2396 | $result |= self::CLI_STATUS_PROCESSED; |
||
| 2397 | } |
||
| 2398 | |||
| 2399 | return $result; |
||
| 2400 | } |
||
| 2401 | |||
| 2402 | /** |
||
| 2403 | * Activate hooks |
||
| 2404 | * |
||
| 2405 | * @return void |
||
| 2406 | */ |
||
| 2407 | public function CLI_runHooks() |
||
| 2408 | { |
||
| 2409 | global $TYPO3_CONF_VARS; |
||
| 2410 | if (is_array($TYPO3_CONF_VARS['EXTCONF']['crawler']['cli_hooks'])) { |
||
| 2411 | foreach ($TYPO3_CONF_VARS['EXTCONF']['crawler']['cli_hooks'] as $objRef) { |
||
| 2412 | $hookObj = &GeneralUtility::getUserObj($objRef); |
||
| 2413 | if (is_object($hookObj)) { |
||
| 2414 | $hookObj->crawler_init($this); |
||
| 2415 | } |
||
| 2416 | } |
||
| 2417 | } |
||
| 2418 | } |
||
| 2419 | |||
| 2420 | /** |
||
| 2421 | * Try to acquire a new process with the given id |
||
| 2422 | * also performs some auto-cleanup for orphan processes |
||
| 2423 | * @todo preemption might not be the most elegant way to clean up |
||
| 2424 | * |
||
| 2425 | * @param string $id identification string for the process |
||
| 2426 | * @return boolean |
||
| 2427 | */ |
||
| 2428 | public function CLI_checkAndAcquireNewProcess($id) |
||
| 2429 | { |
||
| 2430 | $ret = true; |
||
| 2431 | |||
| 2432 | $systemProcessId = getmypid(); |
||
| 2433 | if ($systemProcessId < 1) { |
||
| 2434 | return false; |
||
| 2435 | } |
||
| 2436 | |||
| 2437 | $processCount = 0; |
||
| 2438 | $orphanProcesses = []; |
||
| 2439 | |||
| 2440 | $this->db->sql_query('BEGIN'); |
||
| 2441 | |||
| 2442 | $res = $this->db->exec_SELECTquery( |
||
| 2443 | 'process_id,ttl', |
||
| 2444 | 'tx_crawler_process', |
||
| 2445 | 'active=1 AND deleted=0' |
||
| 2446 | ); |
||
| 2447 | |||
| 2448 | $currentTime = $this->getCurrentTime(); |
||
| 2449 | |||
| 2450 | while ($row = $this->db->sql_fetch_assoc($res)) { |
||
| 2451 | if ($row['ttl'] < $currentTime) { |
||
| 2452 | $orphanProcesses[] = $row['process_id']; |
||
| 2453 | } else { |
||
| 2454 | $processCount++; |
||
| 2455 | } |
||
| 2456 | } |
||
| 2457 | |||
| 2458 | // if there are less than allowed active processes then add a new one |
||
| 2459 | if ($processCount < intval($this->extensionSettings['processLimit'])) { |
||
| 2460 | $this->CLI_debug("add process " . $this->CLI_buildProcessId() . " (" . ($processCount + 1) . "/" . intval($this->extensionSettings['processLimit']) . ")"); |
||
| 2461 | |||
| 2462 | // create new process record |
||
| 2463 | $this->db->exec_INSERTquery( |
||
| 2464 | 'tx_crawler_process', |
||
| 2465 | [ |
||
| 2466 | 'process_id' => $id, |
||
| 2467 | 'active' => '1', |
||
| 2468 | 'ttl' => ($currentTime + intval($this->extensionSettings['processMaxRunTime'])), |
||
| 2469 | 'system_process_id' => $systemProcessId |
||
| 2470 | ] |
||
| 2471 | ); |
||
| 2472 | } else { |
||
| 2473 | $this->CLI_debug("Processlimit reached (" . ($processCount) . "/" . intval($this->extensionSettings['processLimit']) . ")"); |
||
| 2474 | $ret = false; |
||
| 2475 | } |
||
| 2476 | |||
| 2477 | $this->CLI_releaseProcesses($orphanProcesses, true); // maybe this should be somehow included into the current lock |
||
| 2478 | $this->CLI_deleteProcessesMarkedDeleted(); |
||
| 2479 | |||
| 2480 | $this->db->sql_query('COMMIT'); |
||
| 2481 | |||
| 2482 | return $ret; |
||
| 2483 | } |
||
| 2484 | |||
| 2485 | /** |
||
| 2486 | * Release a process and the required resources |
||
| 2487 | * |
||
| 2488 | * @param mixed $releaseIds string with a single process-id or array with multiple process-ids |
||
| 2489 | * @param boolean $withinLock show whether the DB-actions are included within an existing lock |
||
| 2490 | * @return boolean |
||
| 2491 | */ |
||
| 2492 | public function CLI_releaseProcesses($releaseIds, $withinLock = false) |
||
| 2493 | { |
||
| 2494 | if (!is_array($releaseIds)) { |
||
| 2495 | $releaseIds = [$releaseIds]; |
||
| 2496 | } |
||
| 2497 | |||
| 2498 | if (!count($releaseIds) > 0) { |
||
| 2499 | return false; //nothing to release |
||
| 2500 | } |
||
| 2501 | |||
| 2502 | if (!$withinLock) { |
||
| 2503 | $this->db->sql_query('BEGIN'); |
||
| 2504 | } |
||
| 2505 | |||
| 2506 | // some kind of 2nd chance algo - this way you need at least 2 processes to have a real cleanup |
||
| 2507 | // this ensures that a single process can't mess up the entire process table |
||
| 2508 | |||
| 2509 | // mark all processes as deleted which have no "waiting" queue-entires and which are not active |
||
| 2510 | $this->db->exec_UPDATEquery( |
||
| 2511 | 'tx_crawler_queue', |
||
| 2512 | 'process_id IN (SELECT process_id FROM tx_crawler_process WHERE active=0 AND deleted=0)', |
||
| 2513 | [ |
||
| 2514 | 'process_scheduled' => 0, |
||
| 2515 | 'process_id' => '' |
||
| 2516 | ] |
||
| 2517 | ); |
||
| 2518 | $this->db->exec_UPDATEquery( |
||
| 2519 | 'tx_crawler_process', |
||
| 2520 | 'active=0 AND deleted=0 |
||
| 2521 | AND NOT EXISTS ( |
||
| 2522 | SELECT * FROM tx_crawler_queue |
||
| 2523 | WHERE tx_crawler_queue.process_id = tx_crawler_process.process_id |
||
| 2524 | AND tx_crawler_queue.exec_time = 0 |
||
| 2525 | )', |
||
| 2526 | [ |
||
| 2527 | 'deleted' => '1', |
||
| 2528 | 'system_process_id' => 0 |
||
| 2529 | ] |
||
| 2530 | ); |
||
| 2531 | // mark all requested processes as non-active |
||
| 2532 | $this->db->exec_UPDATEquery( |
||
| 2533 | 'tx_crawler_process', |
||
| 2534 | 'process_id IN (\'' . implode('\',\'', $releaseIds) . '\') AND deleted=0', |
||
| 2535 | [ |
||
| 2536 | 'active' => '0' |
||
| 2537 | ] |
||
| 2538 | ); |
||
| 2539 | $this->db->exec_UPDATEquery( |
||
| 2540 | 'tx_crawler_queue', |
||
| 2541 | 'exec_time=0 AND process_id IN ("' . implode('","', $releaseIds) . '")', |
||
| 2542 | [ |
||
| 2543 | 'process_scheduled' => 0, |
||
| 2544 | 'process_id' => '' |
||
| 2545 | ] |
||
| 2546 | ); |
||
| 2547 | |||
| 2548 | if (!$withinLock) { |
||
| 2549 | $this->db->sql_query('COMMIT'); |
||
| 2550 | } |
||
| 2551 | |||
| 2552 | return true; |
||
| 2553 | } |
||
| 2554 | |||
| 2555 | /** |
||
| 2556 | * Delete processes marked as deleted |
||
| 2557 | * |
||
| 2558 | * @return void |
||
| 2559 | */ |
||
| 2560 | 1 | public function CLI_deleteProcessesMarkedDeleted() |
|
| 2561 | { |
||
| 2562 | 1 | $this->db->exec_DELETEquery('tx_crawler_process', 'deleted = 1'); |
|
| 2563 | 1 | } |
|
| 2564 | |||
| 2565 | /** |
||
| 2566 | * Check if there are still resources left for the process with the given id |
||
| 2567 | * Used to determine timeouts and to ensure a proper cleanup if there's a timeout |
||
| 2568 | * |
||
| 2569 | * @param string identification string for the process |
||
| 2570 | * @return boolean determines if the process is still active / has resources |
||
| 2571 | * |
||
| 2572 | * FIXME: Please remove Transaction, not needed as only a select query. |
||
| 2573 | */ |
||
| 2574 | public function CLI_checkIfProcessIsActive($pid) |
||
| 2575 | { |
||
| 2576 | $ret = false; |
||
| 2577 | $this->db->sql_query('BEGIN'); |
||
| 2578 | $res = $this->db->exec_SELECTquery( |
||
| 2579 | 'process_id,active,ttl', |
||
| 2580 | 'tx_crawler_process', |
||
| 2581 | 'process_id = \'' . $pid . '\' AND deleted=0', |
||
| 2582 | '', |
||
| 2583 | 'ttl', |
||
| 2584 | '0,1' |
||
| 2585 | ); |
||
| 2586 | if ($row = $this->db->sql_fetch_assoc($res)) { |
||
| 2587 | $ret = intVal($row['active']) == 1; |
||
| 2588 | } |
||
| 2589 | $this->db->sql_query('COMMIT'); |
||
| 2590 | |||
| 2591 | return $ret; |
||
| 2592 | } |
||
| 2593 | |||
| 2594 | /** |
||
| 2595 | * Create a unique Id for the current process |
||
| 2596 | * |
||
| 2597 | * @return string the ID |
||
| 2598 | */ |
||
| 2599 | 2 | public function CLI_buildProcessId() |
|
| 2600 | { |
||
| 2601 | 2 | if (!$this->processID) { |
|
| 2602 | 1 | $this->processID = GeneralUtility::shortMD5($this->microtime(true)); |
|
| 2603 | } |
||
| 2604 | 2 | return $this->processID; |
|
| 2605 | } |
||
| 2606 | |||
| 2607 | /** |
||
| 2608 | * @param bool $get_as_float |
||
| 2609 | * |
||
| 2610 | * @return mixed |
||
| 2611 | */ |
||
| 2612 | protected function microtime($get_as_float = false) |
||
| 2613 | { |
||
| 2614 | return microtime($get_as_float); |
||
| 2615 | } |
||
| 2616 | |||
| 2617 | /** |
||
| 2618 | * Prints a message to the stdout (only if debug-mode is enabled) |
||
| 2619 | * |
||
| 2620 | * @param string $msg the message |
||
| 2621 | */ |
||
| 2622 | public function CLI_debug($msg) |
||
| 2623 | { |
||
| 2624 | if (intval($this->extensionSettings['processDebug'])) { |
||
| 2625 | echo $msg . "\n"; |
||
| 2626 | flush(); |
||
| 2627 | } |
||
| 2628 | } |
||
| 2629 | |||
| 2630 | /** |
||
| 2631 | * Get URL content by making direct request to TYPO3. |
||
| 2632 | * |
||
| 2633 | * @param string $url Page URL |
||
| 2634 | * @param int $crawlerId Crawler-ID |
||
| 2635 | * @return array |
||
| 2636 | */ |
||
| 2637 | 2 | protected function sendDirectRequest($url, $crawlerId) |
|
| 2638 | { |
||
| 2639 | 2 | $parsedUrl = parse_url($url); |
|
| 2640 | 2 | if (!is_array($parsedUrl)) { |
|
| 2641 | return []; |
||
| 2642 | } |
||
| 2643 | |||
| 2644 | 2 | $requestHeaders = $this->buildRequestHeaderArray($parsedUrl, $crawlerId); |
|
| 2645 | |||
| 2646 | 2 | $cmd = escapeshellcmd($this->extensionSettings['phpPath']); |
|
| 2647 | 2 | $cmd .= ' '; |
|
| 2648 | 2 | $cmd .= escapeshellarg(ExtensionManagementUtility::extPath('crawler') . 'cli/bootstrap.php'); |
|
| 2649 | 2 | $cmd .= ' '; |
|
| 2650 | 2 | $cmd .= escapeshellarg($this->getFrontendBasePath()); |
|
| 2651 | 2 | $cmd .= ' '; |
|
| 2652 | 2 | $cmd .= escapeshellarg($url); |
|
| 2653 | 2 | $cmd .= ' '; |
|
| 2654 | 2 | $cmd .= escapeshellarg(base64_encode(serialize($requestHeaders))); |
|
| 2655 | |||
| 2656 | 2 | $startTime = microtime(true); |
|
| 2657 | 2 | $content = $this->executeShellCommand($cmd); |
|
| 2658 | 2 | $this->log($url . ' ' . (microtime(true) - $startTime)); |
|
| 2659 | |||
| 2660 | $result = [ |
||
| 2661 | 2 | 'request' => implode("\r\n", $requestHeaders) . "\r\n\r\n", |
|
| 2662 | 2 | 'headers' => '', |
|
| 2663 | 2 | 'content' => $content |
|
| 2664 | ]; |
||
| 2665 | |||
| 2666 | 2 | return $result; |
|
| 2667 | } |
||
| 2668 | |||
| 2669 | /** |
||
| 2670 | * Cleans up entries that stayed for too long in the queue. These are: |
||
| 2671 | * - processed entries that are over 1.5 days in age |
||
| 2672 | * - scheduled entries that are over 7 days old |
||
| 2673 | * |
||
| 2674 | * @return void |
||
| 2675 | * |
||
| 2676 | * TODO: Should be switched back to protected - TNM 2018-11-16 |
||
| 2677 | */ |
||
| 2678 | public function cleanUpOldQueueEntries() |
||
| 2679 | { |
||
| 2680 | $processedAgeInSeconds = $this->extensionSettings['cleanUpProcessedAge'] * 86400; // 24*60*60 Seconds in 24 hours |
||
| 2681 | $scheduledAgeInSeconds = $this->extensionSettings['cleanUpScheduledAge'] * 86400; |
||
| 2682 | |||
| 2683 | $now = time(); |
||
| 2684 | $condition = '(exec_time<>0 AND exec_time<' . ($now - $processedAgeInSeconds) . ') OR scheduled<=' . ($now - $scheduledAgeInSeconds); |
||
| 2685 | $this->flushQueue($condition); |
||
| 2686 | } |
||
| 2687 | |||
| 2688 | /** |
||
| 2689 | * Initializes a TypoScript Frontend necessary for using TypoScript and TypoLink functions |
||
| 2690 | * |
||
| 2691 | * @param int $id |
||
| 2692 | * @param int $typeNum |
||
| 2693 | * |
||
| 2694 | * @throws \TYPO3\CMS\Core\Error\Http\ServiceUnavailableException |
||
| 2695 | * |
||
| 2696 | * @return void |
||
| 2697 | */ |
||
| 2698 | protected function initTSFE($id = 1, $typeNum = 0) |
||
| 2699 | { |
||
| 2700 | EidUtility::initTCA(); |
||
| 2701 | $timeTracker = GeneralUtility::makeInstance(TimeTracker::class); |
||
| 2702 | $timeTracker->start(); |
||
| 2703 | |||
| 2704 | $GLOBALS['TSFE'] = GeneralUtility::makeInstance(TypoScriptFrontendController::class, $GLOBALS['TYPO3_CONF_VARS'], $id, $typeNum); |
||
| 2705 | $GLOBALS['TSFE']->sys_page = GeneralUtility::makeInstance(PageRepository::class); |
||
| 2706 | $GLOBALS['TSFE']->sys_page->init(true); |
||
| 2707 | $GLOBALS['TSFE']->connectToDB(); |
||
| 2708 | $GLOBALS['TSFE']->initFEuser(); |
||
| 2709 | $GLOBALS['TSFE']->determineId(); |
||
| 2710 | $GLOBALS['TSFE']->initTemplate(); |
||
| 2711 | $GLOBALS['TSFE']->rootLine = $GLOBALS['TSFE']->sys_page->getRootLine($id, ''); |
||
| 2712 | $GLOBALS['TSFE']->getConfigArray(); |
||
| 2713 | PageGenerator::pagegenInit(); |
||
| 2714 | } |
||
| 2715 | |||
| 2716 | /** |
||
| 2717 | * Returns a md5 hash generated from a serialized configuration array. |
||
| 2718 | * |
||
| 2719 | * @param array $configuration |
||
| 2720 | * |
||
| 2721 | * @return string |
||
| 2722 | */ |
||
| 2723 | 9 | protected function getConfigurationHash(array $configuration) { |
|
| 2724 | 9 | unset($configuration['paramExpanded']); |
|
| 2725 | 9 | unset($configuration['URLs']); |
|
| 2726 | 9 | return md5(serialize($configuration)); |
|
| 2727 | } |
||
| 2728 | |||
| 2729 | /** |
||
| 2730 | * Check whether the Crawling Protocol should be http or https |
||
| 2731 | * |
||
| 2732 | * @param $crawlerConfiguration |
||
| 2733 | * @param $pageConfiguration |
||
| 2734 | * |
||
| 2735 | * @return bool |
||
| 2736 | */ |
||
| 2737 | 6 | protected function isCrawlingProtocolHttps($crawlerConfiguration, $pageConfiguration) { |
|
| 2749 | } |
||
| 2750 |
Loading history...
If you define a variable conditionally, it can happen that it is not defined for all execution paths.
Let’s take a look at an example:
In the above example, the variable $x is defined if you pass “foo” or “bar” as argument for $a. However, since the switch statement has no default case statement, if you pass any other value, the variable $x would be undefined.
Available Fixes
Check for existence of the variable explicitly:
Define a default value for the variable:
Add a value for the missing path: