We could not synchronize checks via GitHub's checks API since Scrutinizer's GitHub App is not installed for this repository.
1 | <?php |
||
2 | /** |
||
3 | * (c) Kitodo. Key to digital objects e.V. <[email protected]> |
||
4 | * |
||
5 | * This file is part of the Kitodo and TYPO3 projects. |
||
6 | * |
||
7 | * @license GNU General Public License version 3 or later. |
||
8 | * For the full copyright and license information, please read the |
||
9 | * LICENSE.txt file that was distributed with this source code. |
||
10 | */ |
||
11 | |||
12 | namespace Kitodo\Dlf\Controller; |
||
13 | |||
14 | use TYPO3\CMS\Core\Utility\GeneralUtility; |
||
15 | use Kitodo\Dlf\Common\Solr\Solr; |
||
16 | use Kitodo\Dlf\Domain\Model\Token; |
||
17 | use Kitodo\Dlf\Domain\Repository\CollectionRepository; |
||
18 | use Kitodo\Dlf\Domain\Repository\LibraryRepository; |
||
19 | use Kitodo\Dlf\Domain\Repository\TokenRepository; |
||
20 | |||
21 | /** |
||
22 | * Controller class for the plugin 'OAI-PMH Interface'. |
||
23 | * |
||
24 | * @package TYPO3 |
||
25 | * @subpackage dlf |
||
26 | * |
||
27 | * @access public |
||
28 | */ |
||
29 | class OaiPmhController extends AbstractController |
||
30 | { |
||
31 | /** |
||
32 | * @access protected |
||
33 | * @var TokenRepository |
||
34 | */ |
||
35 | protected $tokenRepository; |
||
36 | |||
37 | /** |
||
38 | * @access public |
||
39 | * |
||
40 | * @param TokenRepository $tokenRepository |
||
41 | * |
||
42 | * @return void |
||
43 | */ |
||
44 | public function injectTokenRepository(TokenRepository $tokenRepository) |
||
45 | { |
||
46 | $this->tokenRepository = $tokenRepository; |
||
47 | } |
||
48 | |||
49 | /** |
||
50 | * @access protected |
||
51 | * @var CollectionRepository |
||
52 | */ |
||
53 | protected $collectionRepository; |
||
54 | |||
55 | /** |
||
56 | * @access public |
||
57 | * |
||
58 | * @param CollectionRepository $collectionRepository |
||
59 | * |
||
60 | * @return void |
||
61 | */ |
||
62 | public function injectCollectionRepository(CollectionRepository $collectionRepository) |
||
63 | { |
||
64 | $this->collectionRepository = $collectionRepository; |
||
65 | } |
||
66 | |||
67 | /** |
||
68 | * @access protected |
||
69 | * @var LibraryRepository |
||
70 | */ |
||
71 | protected $libraryRepository; |
||
72 | |||
73 | /** |
||
74 | * @access public |
||
75 | * |
||
76 | * @param LibraryRepository $libraryRepository |
||
77 | * |
||
78 | * @return void |
||
79 | */ |
||
80 | public function injectLibraryRepository(LibraryRepository $libraryRepository) |
||
81 | { |
||
82 | $this->libraryRepository = $libraryRepository; |
||
83 | } |
||
84 | |||
85 | /** |
||
86 | * Initializes the current action |
||
87 | * |
||
88 | * @access public |
||
89 | * |
||
90 | * @return void |
||
91 | */ |
||
92 | public function initializeAction() |
||
93 | { |
||
94 | $this->request->setFormat('xml'); |
||
95 | } |
||
96 | |||
97 | /** |
||
98 | * @access protected |
||
99 | * @var string Did an error occur? |
||
100 | */ |
||
101 | protected $error; |
||
102 | |||
103 | /** |
||
104 | * @access protected |
||
105 | * @var array This holds the configuration for all supported metadata prefixes |
||
106 | */ |
||
107 | protected $formats = [ |
||
108 | 'oai_dc' => [ |
||
109 | 'schema' => 'http://www.openarchives.org/OAI/2.0/oai_dc.xsd', |
||
110 | 'namespace' => 'http://www.openarchives.org/OAI/2.0/oai_dc/', |
||
111 | 'requiredFields' => ['record_id'], |
||
112 | ], |
||
113 | 'epicur' => [ |
||
114 | 'schema' => 'http://www.persistent-identifier.de/xepicur/version1.0/xepicur.xsd', |
||
115 | 'namespace' => 'urn:nbn:de:1111-2004033116', |
||
116 | 'requiredFields' => ['purl', 'urn'], |
||
117 | ], |
||
118 | 'mets' => [ |
||
119 | 'schema' => 'http://www.loc.gov/standards/mets/version17/mets.v1-7.xsd', |
||
120 | 'namespace' => 'http://www.loc.gov/METS/', |
||
121 | 'requiredFields' => ['location'], |
||
122 | ] |
||
123 | ]; |
||
124 | |||
125 | /** |
||
126 | * @access protected |
||
127 | * @var array |
||
128 | */ |
||
129 | protected $parameters = []; |
||
130 | |||
131 | /** |
||
132 | * Delete expired resumption tokens |
||
133 | * |
||
134 | * @access protected |
||
135 | * |
||
136 | * @return void |
||
137 | */ |
||
138 | protected function deleteExpiredTokens() |
||
139 | { |
||
140 | // Delete expired resumption tokens. |
||
141 | $this->tokenRepository->deleteExpiredTokens($this->settings['expired']); |
||
142 | } |
||
143 | |||
144 | /** |
||
145 | * Load URL parameters |
||
146 | * |
||
147 | * @access protected |
||
148 | * |
||
149 | * @return void |
||
150 | */ |
||
151 | protected function getUrlParams() |
||
152 | { |
||
153 | $allowedParams = [ |
||
154 | 'verb', |
||
155 | 'identifier', |
||
156 | 'metadataPrefix', |
||
157 | 'from', |
||
158 | 'until', |
||
159 | 'set', |
||
160 | 'resumptionToken' |
||
161 | ]; |
||
162 | // Clear plugin variables. |
||
163 | $this->parameters = []; |
||
164 | // Set only allowed parameters. |
||
165 | foreach ($allowedParams as $param) { |
||
166 | if (GeneralUtility::_GP($param)) { |
||
167 | $this->parameters[$param] = GeneralUtility::_GP($param); |
||
168 | } |
||
169 | } |
||
170 | } |
||
171 | |||
172 | /** |
||
173 | * Get unqualified Dublin Core data. |
||
174 | * @see http://www.openarchives.org/OAI/openarchivesprotocol.html#dublincore |
||
175 | * |
||
176 | * @access private |
||
177 | * |
||
178 | * @param array $record The full record array |
||
179 | * |
||
180 | * @return array The mapped metadata array |
||
181 | */ |
||
182 | private function getDublinCoreData(array $record) |
||
183 | { |
||
184 | $metadata = []; |
||
185 | |||
186 | $metadata[] = ['dc:identifier' => $record['record_id']]; |
||
187 | |||
188 | $this->addDublinCoreData($metadata, 'dc:identifier', $record['purl']); |
||
189 | $this->addDublinCoreData($metadata, 'dc:identifier', $record['prod_id']); |
||
190 | $this->addDublinCoreData($metadata, 'dc:identifier', $record['urn']); |
||
191 | $this->addDublinCoreData($metadata, 'dc:title', $record['title']); |
||
192 | $this->addDublinCoreData($metadata, 'dc:creator', $record['author']); |
||
193 | $this->addDublinCoreData($metadata, 'dc:date', $record['year']); |
||
194 | $this->addDublinCoreData($metadata, 'dc:coverage', $record['place']); |
||
195 | |||
196 | $record[] = ['dc:format' => $record['application/mets+xml']]; |
||
197 | $record[] = ['dc:type' => $record['Text']]; |
||
198 | if (!empty($record['partof'])) { |
||
199 | $document = $this->documentRepository->findOneByPartof($metadata['partof']); |
||
200 | |||
201 | if ($document) { |
||
202 | $metadata[] = ['dc:relation' => $document->getRecordId()]; |
||
203 | } |
||
204 | } |
||
205 | $this->addDublinCoreData($metadata, 'dc:rights', $record['license']); |
||
206 | $this->addDublinCoreData($metadata, 'dc:rights', $record['terms']); |
||
207 | $this->addDublinCoreData($metadata, 'dc:rights', $record['restrictions']); |
||
208 | $this->addDublinCoreData($metadata, 'dc:rights', $record['out_of_print']); |
||
209 | $this->addDublinCoreData($metadata, 'dc:rights', $record['rights_info']); |
||
210 | |||
211 | return $metadata; |
||
212 | } |
||
213 | |||
214 | /** |
||
215 | * Add Dublin Core data. |
||
216 | * |
||
217 | * @access private |
||
218 | * |
||
219 | * @param array $metadata The mapped metadata array passed as reference |
||
220 | * @param string $key The key to which record value should be assigned |
||
221 | * @param string $value The key from record array |
||
222 | * |
||
223 | * @return void |
||
224 | */ |
||
225 | private function addDublinCoreData(&$metadata, $key, $value) |
||
226 | { |
||
227 | if (!empty($value)) { |
||
228 | $metadata[] = [$key => $value]; |
||
229 | } |
||
230 | } |
||
231 | |||
232 | /** |
||
233 | * Get METS data. |
||
234 | * @see http://www.loc.gov/standards/mets/docs/mets.v1-7.html |
||
235 | * |
||
236 | * @access protected |
||
237 | * |
||
238 | * @param array $record The full record array |
||
239 | * |
||
240 | * @return string The fetched METS XML |
||
241 | */ |
||
242 | protected function getMetsData(array $record) |
||
243 | { |
||
244 | $mets = null; |
||
245 | // Load METS file. |
||
246 | $xml = new \DOMDocument(); |
||
247 | if ($xml->load($record['location'])) { |
||
248 | // Get root element. |
||
249 | $root = $xml->getElementsByTagNameNS($this->formats['mets']['namespace'], 'mets'); |
||
250 | if ($root->item(0) instanceof \DOMNode) { |
||
251 | $mets = $xml->saveXML($root->item(0)); |
||
252 | } else { |
||
253 | $this->logger->error('No METS part found in document with location "' . $record['location'] . '"'); |
||
254 | } |
||
255 | } else { |
||
256 | $this->logger->error('Could not load XML file from "' . $record['location'] . '"'); |
||
257 | } |
||
258 | return $mets; |
||
259 | } |
||
260 | |||
261 | /** |
||
262 | * The main method of the plugin |
||
263 | * |
||
264 | * @access public |
||
265 | * |
||
266 | * @return void |
||
267 | */ |
||
268 | public function mainAction() |
||
269 | { |
||
270 | // Get allowed GET and POST variables. |
||
271 | $this->getUrlParams(); |
||
272 | |||
273 | // Delete expired resumption tokens. |
||
274 | $this->deleteExpiredTokens(); |
||
275 | |||
276 | switch ($this->parameters['verb']) { |
||
277 | case 'GetRecord': |
||
278 | $this->verbGetRecord(); |
||
279 | break; |
||
280 | case 'Identify': |
||
281 | $this->verbIdentify(); |
||
282 | break; |
||
283 | case 'ListIdentifiers': |
||
284 | $this->verbListIdentifiers(); |
||
285 | break; |
||
286 | case 'ListMetadataFormats': |
||
287 | $this->verbListMetadataFormats(); |
||
288 | break; |
||
289 | case 'ListRecords': |
||
290 | $this->verbListRecords(); |
||
291 | break; |
||
292 | case 'ListSets': |
||
293 | $this->verbListSets(); |
||
294 | break; |
||
295 | default: |
||
296 | $this->error = 'badVerb'; |
||
297 | break; |
||
298 | } |
||
299 | |||
300 | $this->view->assign('parameters', $this->parameters); |
||
301 | $this->view->assign('error', $this->error); |
||
302 | |||
303 | return; |
||
304 | } |
||
305 | |||
306 | /** |
||
307 | * Continue with resumption token |
||
308 | * |
||
309 | * @access protected |
||
310 | * |
||
311 | * @return array|null list of uids |
||
312 | */ |
||
313 | protected function resume(): ?array |
||
314 | { |
||
315 | $token = $this->tokenRepository->findOneByToken($this->parameters['resumptionToken']); |
||
316 | |||
317 | if ($token) { |
||
318 | $options = $token->getOptions(); |
||
319 | } |
||
320 | if (is_array($options)) { |
||
0 ignored issues
–
show
Comprehensibility
Best Practice
introduced
by
Loading history...
|
|||
321 | return $options; |
||
322 | } else { |
||
323 | // No resumption token found or resumption token expired. |
||
324 | $this->error = 'badResumptionToken'; |
||
325 | return null; |
||
326 | } |
||
327 | } |
||
328 | |||
329 | /** |
||
330 | * Process verb "GetRecord" |
||
331 | * |
||
332 | * @access protected |
||
333 | * |
||
334 | * @return void |
||
335 | */ |
||
336 | protected function verbGetRecord() |
||
337 | { |
||
338 | if (count($this->parameters) !== 3 || empty($this->parameters['metadataPrefix']) || empty($this->parameters['identifier'])) { |
||
339 | $this->error = 'badArgument'; |
||
340 | return; |
||
341 | } |
||
342 | |||
343 | if (!array_key_exists($this->parameters['metadataPrefix'], $this->formats)) { |
||
344 | $this->error = 'cannotDisseminateFormat'; |
||
345 | return; |
||
346 | } |
||
347 | |||
348 | $document = $this->documentRepository->getOaiRecord($this->settings, $this->parameters); |
||
349 | |||
350 | if (!$document['uid']) { |
||
351 | $this->error = 'idDoesNotExist'; |
||
352 | return; |
||
353 | } |
||
354 | |||
355 | // Check for required fields. |
||
356 | foreach ($this->formats[$this->parameters['metadataPrefix']]['requiredFields'] as $required) { |
||
357 | if (empty($document[$required])) { |
||
358 | $this->error = 'cannotDisseminateFormat'; |
||
359 | return; |
||
360 | } |
||
361 | } |
||
362 | |||
363 | // we need the collections as array later |
||
364 | $document['collections'] = explode(' ', $document['collections']); |
||
365 | |||
366 | // Add metadata |
||
367 | switch ($this->parameters['metadataPrefix']) { |
||
368 | case 'oai_dc': |
||
369 | $document['metadata'] = $this->getDublinCoreData($document); |
||
370 | break; |
||
371 | case 'epicur': |
||
372 | $document['metadata'] = $document; |
||
373 | break; |
||
374 | case 'mets': |
||
375 | $document['metadata'] = $this->getMetsData($document); |
||
376 | break; |
||
377 | } |
||
378 | |||
379 | $this->view->assign('record', $document); |
||
380 | } |
||
381 | |||
382 | /** |
||
383 | * Process verb "Identify" |
||
384 | * |
||
385 | * @access protected |
||
386 | * |
||
387 | * @return void |
||
388 | */ |
||
389 | protected function verbIdentify() |
||
390 | { |
||
391 | $library = $this->libraryRepository->findByUid($this->settings['library']); |
||
392 | |||
393 | $oaiIdentifyInfo = []; |
||
394 | |||
395 | if (!$oaiIdentifyInfo) { |
||
396 | $this->logger->notice('Incomplete plugin configuration'); |
||
397 | } |
||
398 | |||
399 | $oaiIdentifyInfo['oai_label'] = $library ? $library->getOaiLabel() : ''; |
||
400 | // Use default values for an installation with incomplete plugin configuration. |
||
401 | if (empty($oaiIdentifyInfo['oai_label'])) { |
||
402 | $oaiIdentifyInfo['oai_label'] = 'Kitodo.Presentation OAI-PMH Interface (default configuration)'; |
||
403 | $this->logger->notice('Incomplete plugin configuration (oai_label is missing)'); |
||
404 | } |
||
405 | |||
406 | $oaiIdentifyInfo['contact'] = $library ? $library->getContact() : ''; |
||
407 | if (empty($oaiIdentifyInfo['contact'])) { |
||
408 | $oaiIdentifyInfo['contact'] = '[email protected]'; |
||
409 | $this->logger->notice('Incomplete plugin configuration (contact is missing)'); |
||
410 | } |
||
411 | |||
412 | $document = $this->documentRepository->findOldestDocument(); |
||
413 | |||
414 | if ($document) { |
||
415 | $oaiIdentifyInfo['earliestDatestamp'] = gmdate('Y-m-d\TH:i:s\Z', $document->getTstamp()->getTimestamp()); |
||
416 | } else { |
||
417 | // Provide a fallback timestamp if no document is found |
||
418 | $oaiIdentifyInfo['earliestDatestamp'] = '0000-00-00T00:00:00Z'; |
||
419 | |||
420 | // access storagePid from TypoScript |
||
421 | $pageSettings = $this->configurationManager->getConfiguration($this->configurationManager::CONFIGURATION_TYPE_FULL_TYPOSCRIPT); |
||
422 | $storagePid = $pageSettings["plugin."]["tx_dlf."]["persistence."]["storagePid"]; |
||
423 | if ($storagePid > 0) { |
||
424 | $this->logger->notice('No records found with PID ' . $storagePid); |
||
425 | } else { |
||
426 | $this->logger->notice('No records found'); |
||
427 | } |
||
428 | } |
||
429 | $this->view->assign('oaiIdentifyInfo', $oaiIdentifyInfo); |
||
430 | } |
||
431 | |||
432 | /** |
||
433 | * Process verb "ListIdentifiers" |
||
434 | * |
||
435 | * @access protected |
||
436 | * |
||
437 | * @return void |
||
438 | */ |
||
439 | protected function verbListIdentifiers() |
||
440 | { |
||
441 | // If we have a resumption token we can continue our work |
||
442 | if (!empty($this->parameters['resumptionToken'])) { |
||
443 | // "resumptionToken" is an exclusive argument. |
||
444 | if (count($this->parameters) > 2) { |
||
445 | $this->error = 'badArgument'; |
||
446 | return; |
||
447 | } else { |
||
448 | // return next chunk of documents |
||
449 | $resultSet = $this->resume(); |
||
450 | if (is_array($resultSet)) { |
||
451 | $listIdentifiers = $this->generateOutputForDocumentList($resultSet); |
||
452 | $this->view->assign('listIdentifiers', $listIdentifiers); |
||
453 | } |
||
454 | return; |
||
455 | } |
||
456 | } |
||
457 | // "metadataPrefix" is required and "identifier" is not allowed. |
||
458 | if (empty($this->parameters['metadataPrefix']) || !empty($this->parameters['identifier'])) { |
||
459 | $this->error = 'badArgument'; |
||
460 | return; |
||
461 | } |
||
462 | if (!in_array($this->parameters['metadataPrefix'], array_keys($this->formats))) { |
||
463 | $this->error = 'cannotDisseminateFormat'; |
||
464 | return; |
||
465 | } |
||
466 | try { |
||
467 | $documentSet = $this->fetchDocumentSet(); |
||
468 | } catch (\Exception $exception) { |
||
469 | $this->error = 'idDoesNotExist'; |
||
470 | return; |
||
471 | } |
||
472 | // create new and empty document list |
||
473 | $resultSet = []; |
||
474 | if (is_array($documentSet)) { |
||
475 | $resultSet['elements'] = $documentSet; |
||
476 | $resultSet['metadata'] = [ |
||
477 | 'cursor' => 0, |
||
478 | 'completeListSize' => count($documentSet), |
||
479 | 'metadataPrefix' => $this->parameters['metadataPrefix'], |
||
480 | ]; |
||
481 | } |
||
482 | |||
483 | $listIdentifiers = $this->generateOutputForDocumentList($resultSet); |
||
484 | $this->view->assign('listIdentifiers', $listIdentifiers); |
||
485 | } |
||
486 | |||
487 | /** |
||
488 | * Process verb "ListMetadataFormats" |
||
489 | * |
||
490 | * @access protected |
||
491 | * |
||
492 | * @return void |
||
493 | */ |
||
494 | protected function verbListMetadataFormats() |
||
495 | { |
||
496 | $resArray = []; |
||
497 | // check for the optional "identifier" parameter |
||
498 | if (isset($this->parameters['identifier'])) { |
||
499 | $resArray = $this->documentRepository->findOneByRecordId($this->parameters['identifier']); |
||
500 | } |
||
501 | |||
502 | $resultSet = []; |
||
503 | foreach ($this->formats as $prefix => $details) { |
||
504 | if (!empty($resArray)) { |
||
505 | // check, if all required fields are available for a given identifier |
||
506 | foreach ($details['requiredFields'] as $required) { |
||
507 | $methodName = 'get' . GeneralUtility::underscoredToUpperCamelCase($required); |
||
508 | if (empty($resArray->$methodName())) { |
||
509 | // Skip metadata formats whose requirements are not met. |
||
510 | continue 2; |
||
511 | } |
||
512 | } |
||
513 | } |
||
514 | $details['prefix'] = $prefix; |
||
515 | $resultSet[] = $details; |
||
516 | } |
||
517 | $this->view->assign('metadataFormats', $resultSet); |
||
518 | } |
||
519 | |||
520 | /** |
||
521 | * Process verb "ListRecords" |
||
522 | * |
||
523 | * @access protected |
||
524 | * |
||
525 | * @return void |
||
526 | */ |
||
527 | protected function verbListRecords() |
||
528 | { |
||
529 | // Check for invalid arguments. |
||
530 | if (!empty($this->parameters['resumptionToken'])) { |
||
531 | // "resumptionToken" is an exclusive argument. |
||
532 | if (count($this->parameters) > 2) { |
||
533 | $this->error = 'badArgument'; |
||
534 | return; |
||
535 | } else { |
||
536 | // return next chunk of documents |
||
537 | $resultSet = $this->resume(); |
||
538 | if (is_array($resultSet)) { |
||
539 | $listRecords = $this->generateOutputForDocumentList($resultSet); |
||
540 | $this->parameters['metadataPrefix'] = $resultSet['metadata']['metadataPrefix']; |
||
541 | $this->view->assign('listRecords', $listRecords); |
||
542 | } |
||
543 | return; |
||
544 | } |
||
545 | } |
||
546 | if (empty($this->parameters['metadataPrefix']) || !empty($this->parameters['identifier'])) { |
||
547 | // "metadataPrefix" is required and "identifier" is not allowed. |
||
548 | $this->error = 'badArgument'; |
||
549 | return; |
||
550 | } |
||
551 | // Check "metadataPrefix" for valid value. |
||
552 | if (!in_array($this->parameters['metadataPrefix'], array_keys($this->formats))) { |
||
553 | $this->error = 'cannotDisseminateFormat'; |
||
554 | return; |
||
555 | } |
||
556 | try { |
||
557 | $documentSet = $this->fetchDocumentSet(); |
||
558 | } catch (\Exception $exception) { |
||
559 | $this->error = 'idDoesNotExist'; |
||
560 | return; |
||
561 | } |
||
562 | $resultSet = []; |
||
563 | if (count($documentSet) > 0) { |
||
564 | $resultSet['elements'] = $documentSet; |
||
565 | $resultSet['metadata'] = [ |
||
566 | 'cursor' => 0, |
||
567 | 'completeListSize' => count($documentSet), |
||
568 | 'metadataPrefix' => $this->parameters['metadataPrefix'], |
||
569 | ]; |
||
570 | } |
||
571 | |||
572 | $resultSet = $this->generateOutputForDocumentList($resultSet); |
||
573 | $this->view->assign('listRecords', $resultSet); |
||
574 | } |
||
575 | |||
576 | /** |
||
577 | * Process verb "ListSets" |
||
578 | * |
||
579 | * @access protected |
||
580 | * |
||
581 | * @return void |
||
582 | */ |
||
583 | protected function verbListSets() |
||
584 | { |
||
585 | // It is required to set a oai_name inside the collection record to be shown in oai-pmh plugin. |
||
586 | $this->settings['hideEmptyOaiNames'] = true; |
||
587 | |||
588 | $oaiSets = $this->collectionRepository->findCollectionsBySettings($this->settings); |
||
589 | |||
590 | $this->view->assign('oaiSets', $oaiSets); |
||
591 | } |
||
592 | |||
593 | /** |
||
594 | * Fetch records |
||
595 | * |
||
596 | * @access protected |
||
597 | * |
||
598 | * @return array matching records or empty array if there were some errors |
||
599 | */ |
||
600 | protected function fetchDocumentSet(): array |
||
601 | { |
||
602 | $documentSet = []; |
||
603 | $solrQuery = ''; |
||
604 | // Check "set" for valid value. |
||
605 | if (!empty($this->parameters['set'])) { |
||
606 | // For SOLR we need the index_name of the collection, |
||
607 | // For DB Query we need the UID of the collection |
||
608 | |||
609 | $result = $this->collectionRepository->getIndexNameForSolr($this->settings, $this->parameters['set']); |
||
610 | $resArray = $result->fetchAssociative(); |
||
611 | if ($resArray) { |
||
612 | if ($resArray['index_query'] != "") { |
||
613 | $solrQuery .= '(' . $resArray['index_query'] . ')'; |
||
614 | } else { |
||
615 | $solrQuery .= 'collection:' . '"' . $resArray['index_name'] . '"'; |
||
616 | } |
||
617 | } else { |
||
618 | $this->error = 'noSetHierarchy'; |
||
619 | return $documentSet; |
||
620 | } |
||
621 | } else { |
||
622 | // If no set is specified we have to query for all collections |
||
623 | $solrQuery .= 'collection:* NOT collection:""'; |
||
624 | } |
||
625 | // Check for required fields. |
||
626 | foreach ($this->formats[$this->parameters['metadataPrefix']]['requiredFields'] as $required) { |
||
627 | $solrQuery .= ' NOT ' . $required . ':""'; |
||
628 | } |
||
629 | // toplevel="true" is always required |
||
630 | $solrQuery .= ' AND toplevel:true'; |
||
631 | |||
632 | $from = $this->getFrom(); |
||
633 | $until = $this->getUntil($from); |
||
634 | |||
635 | $this->checkGranularity(); |
||
636 | |||
637 | if ($this->error === 'badArgument') { |
||
638 | return $documentSet; |
||
639 | } |
||
640 | |||
641 | $solrQuery .= ' AND timestamp:[' . $from . ' TO ' . $until . ']'; |
||
642 | |||
643 | $solr = Solr::getInstance($this->settings['solrcore']); |
||
644 | if (!$solr->ready) { |
||
645 | $this->logger->error('Apache Solr not available'); |
||
646 | return $documentSet; |
||
647 | } |
||
648 | if ((int) $this->settings['solr_limit'] > 0) { |
||
649 | $solr->limit = (int) $this->settings['solr_limit']; |
||
650 | } |
||
651 | // We only care about the UID in the results and want them sorted |
||
652 | $parameters = [ |
||
653 | "fields" => "uid", |
||
654 | "sort" => [ |
||
655 | "uid" => "asc" |
||
656 | ] |
||
657 | ]; |
||
658 | $parameters['query'] = $solrQuery; |
||
659 | $result = $solr->searchRaw($parameters); |
||
660 | if (empty($result)) { |
||
661 | $this->error = 'noRecordsMatch'; |
||
662 | return $documentSet; |
||
663 | } |
||
664 | foreach ($result as $doc) { |
||
665 | $documentSet[] = $doc->uid; |
||
666 | } |
||
667 | return $documentSet; |
||
668 | } |
||
669 | |||
670 | /** |
||
671 | * Get 'from' query parameter. |
||
672 | * |
||
673 | * @access private |
||
674 | * |
||
675 | * @return string |
||
676 | */ |
||
677 | private function getFrom(): string |
||
678 | { |
||
679 | $from = "*"; |
||
680 | // Check "from" for valid value. |
||
681 | if (!empty($this->parameters['from'])) { |
||
682 | // Is valid format? |
||
683 | $date = $this->getDate('from'); |
||
684 | if (is_array($date)) { |
||
685 | $from = $this->getDateFromTimestamp($date, '.000Z'); |
||
686 | } else { |
||
687 | $this->error = 'badArgument'; |
||
688 | } |
||
689 | } |
||
690 | return $from; |
||
691 | } |
||
692 | |||
693 | /** |
||
694 | * Get 'until' query parameter. |
||
695 | * |
||
696 | * @access private |
||
697 | * |
||
698 | * @param string $from start date |
||
699 | * |
||
700 | * @return string |
||
701 | */ |
||
702 | private function getUntil(string $from): string |
||
703 | { |
||
704 | $until = "*"; |
||
705 | // Check "until" for valid value. |
||
706 | if (!empty($this->parameters['until'])) { |
||
707 | // Is valid format? |
||
708 | $date = $this->getDate('until'); |
||
709 | if (is_array($date)) { |
||
710 | $until = $this->getDateFromTimestamp($date, '.999Z'); |
||
711 | if ($from != "*" && $from > $until) { |
||
712 | $this->error = 'badArgument'; |
||
713 | } |
||
714 | } else { |
||
715 | $this->error = 'badArgument'; |
||
716 | } |
||
717 | } |
||
718 | return $until; |
||
719 | } |
||
720 | |||
721 | /** |
||
722 | * Get date from parameter string. |
||
723 | * |
||
724 | * @access private |
||
725 | * |
||
726 | * @param string $dateType |
||
727 | * |
||
728 | * @return array|false |
||
729 | */ |
||
730 | private function getDate(string $dateType) |
||
731 | { |
||
732 | return strptime($this->parameters[$dateType], '%Y-%m-%dT%H:%M:%SZ') ?: strptime($this->parameters[$dateType], '%Y-%m-%d'); |
||
733 | } |
||
734 | |||
735 | /** |
||
736 | * Get date from timestamp. |
||
737 | * |
||
738 | * @access private |
||
739 | * |
||
740 | * @param array $date |
||
741 | * @param string $end |
||
742 | * |
||
743 | * @return string |
||
744 | */ |
||
745 | private function getDateFromTimestamp(array $date, string $end): string |
||
746 | { |
||
747 | $timestamp = gmmktime( |
||
748 | $date['tm_hour'], |
||
749 | $date['tm_min'], |
||
750 | $date['tm_sec'], |
||
751 | $date['tm_mon'] + 1, |
||
752 | $date['tm_mday'], |
||
753 | $date['tm_year'] + 1900 |
||
754 | ); |
||
755 | return date("Y-m-d", $timestamp) . 'T' . date("H:i:s", $timestamp) . $end; |
||
756 | } |
||
757 | |||
758 | /** |
||
759 | * Check "from" and "until" for same granularity. |
||
760 | * |
||
761 | * @access private |
||
762 | * |
||
763 | * @return void |
||
764 | */ |
||
765 | private function checkGranularity(): void |
||
766 | { |
||
767 | if ( |
||
768 | !empty($this->parameters['from']) |
||
769 | && !empty($this->parameters['until']) |
||
770 | ) { |
||
771 | if (strlen($this->parameters['from']) != strlen($this->parameters['until'])) { |
||
772 | $this->error = 'badArgument'; |
||
773 | } |
||
774 | } |
||
775 | } |
||
776 | |||
777 | /** |
||
778 | * Fetch more information for document list |
||
779 | * |
||
780 | * @access protected |
||
781 | * |
||
782 | * @param array $documentListSet |
||
783 | * |
||
784 | * @return array of enriched records |
||
785 | */ |
||
786 | protected function generateOutputForDocumentList(array $documentListSet) |
||
787 | { |
||
788 | $documentsToProcess = array_splice($documentListSet['elements'], 0, (int) $this->settings['limit']); |
||
789 | if (empty($documentsToProcess)) { |
||
790 | $this->error = 'noRecordsMatch'; |
||
791 | return []; |
||
792 | } |
||
793 | $verb = $this->parameters['verb']; |
||
794 | |||
795 | $documents = $this->documentRepository->getOaiDocumentList($documentsToProcess); |
||
796 | |||
797 | $records = []; |
||
798 | while ($resArray = $documents->fetchAssociative()) { |
||
799 | // we need the collections as array later |
||
800 | $resArray['collections'] = explode(' ', $resArray['collections']); |
||
801 | |||
802 | if ($verb === 'ListRecords') { |
||
803 | // Add metadata node. |
||
804 | $metadataPrefix = $this->parameters['metadataPrefix']; |
||
805 | if (!$metadataPrefix) { |
||
806 | // If we resume an action the metadataPrefix is stored with the documentSet |
||
807 | $metadataPrefix = $documentListSet['metadata']['metadataPrefix']; |
||
808 | } |
||
809 | switch ($metadataPrefix) { |
||
810 | case 'oai_dc': |
||
811 | $resArray['metadata'] = $this->getDublinCoreData($resArray); |
||
812 | break; |
||
813 | case 'epicur': |
||
814 | $resArray['metadata'] = $resArray; |
||
815 | break; |
||
816 | case 'mets': |
||
817 | $resArray['metadata'] = $this->getMetsData($resArray); |
||
818 | break; |
||
819 | } |
||
820 | } |
||
821 | |||
822 | $records[] = $resArray; |
||
823 | } |
||
824 | |||
825 | $this->generateResumptionTokenForDocumentListSet($documentListSet, count($documentsToProcess)); |
||
826 | |||
827 | return $records; |
||
828 | } |
||
829 | |||
830 | /** |
||
831 | * Generate resumption token |
||
832 | * |
||
833 | * @access protected |
||
834 | * |
||
835 | * @param array $documentListSet |
||
836 | * @param int $numShownDocuments |
||
837 | * |
||
838 | * @return void |
||
839 | */ |
||
840 | protected function generateResumptionTokenForDocumentListSet(array $documentListSet, int $numShownDocuments) |
||
841 | { |
||
842 | // The cursor specifies how many elements have already been returned in previous requests |
||
843 | // See http://www.openarchives.org/OAI/openarchivesprotocol.html#FlowControl |
||
844 | $currentCursor = $documentListSet['metadata']['cursor']; |
||
845 | |||
846 | if (count($documentListSet['elements']) !== 0) { |
||
847 | $resumptionToken = uniqid('', false); |
||
848 | |||
849 | $documentListSet['metadata']['cursor'] += $numShownDocuments; |
||
850 | |||
851 | // create new token |
||
852 | $newToken = GeneralUtility::makeInstance(Token::class); |
||
853 | $newToken->setToken($resumptionToken); |
||
854 | $newToken->setOptions($documentListSet); |
||
855 | |||
856 | // add to tokenRepository |
||
857 | $this->tokenRepository->add($newToken); |
||
858 | } else { |
||
859 | // Result set complete. We don't need a token. |
||
860 | $resumptionToken = ''; |
||
861 | } |
||
862 | |||
863 | $resumptionTokenInfo = []; |
||
864 | $resumptionTokenInfo['token'] = $resumptionToken; |
||
865 | $resumptionTokenInfo['cursor'] = $currentCursor; |
||
866 | $resumptionTokenInfo['completeListSize'] = $documentListSet['metadata']['completeListSize']; |
||
867 | $expireDateTime = new \DateTime(); |
||
868 | $expireDateTime->add(new \DateInterval('PT' . $this->settings['expired'] . 'S')); |
||
869 | $resumptionTokenInfo['expired'] = $expireDateTime; |
||
870 | |||
871 | $omitResumptionToken = $currentCursor === 0 && $numShownDocuments >= $documentListSet['metadata']['completeListSize']; |
||
872 | if (!$omitResumptionToken) { |
||
873 | $this->view->assign('resumptionToken', $resumptionTokenInfo); |
||
874 | } |
||
875 | } |
||
876 | } |
||
877 |