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
|
|
|
use Psr\Http\Message\ResponseInterface; |
21
|
|
|
|
22
|
|
|
/** |
23
|
|
|
* Controller class for the plugin 'OAI-PMH Interface'. |
24
|
|
|
* |
25
|
|
|
* @package TYPO3 |
26
|
|
|
* @subpackage dlf |
27
|
|
|
* |
28
|
|
|
* @access public |
29
|
|
|
*/ |
30
|
|
|
class OaiPmhController extends AbstractController |
31
|
|
|
{ |
32
|
|
|
/** |
33
|
|
|
* @access protected |
34
|
|
|
* @var TokenRepository |
35
|
|
|
*/ |
36
|
|
|
protected $tokenRepository; |
37
|
|
|
|
38
|
|
|
/** |
39
|
|
|
* @access public |
40
|
|
|
* |
41
|
|
|
* @param TokenRepository $tokenRepository |
42
|
|
|
* |
43
|
|
|
* @return void |
44
|
|
|
*/ |
45
|
|
|
public function injectTokenRepository(TokenRepository $tokenRepository) |
46
|
|
|
{ |
47
|
|
|
$this->tokenRepository = $tokenRepository; |
48
|
|
|
} |
49
|
|
|
|
50
|
|
|
/** |
51
|
|
|
* @access protected |
52
|
|
|
* @var CollectionRepository |
53
|
|
|
*/ |
54
|
|
|
protected $collectionRepository; |
55
|
|
|
|
56
|
|
|
/** |
57
|
|
|
* @access public |
58
|
|
|
* |
59
|
|
|
* @param CollectionRepository $collectionRepository |
60
|
|
|
* |
61
|
|
|
* @return void |
62
|
|
|
*/ |
63
|
|
|
public function injectCollectionRepository(CollectionRepository $collectionRepository) |
64
|
|
|
{ |
65
|
|
|
$this->collectionRepository = $collectionRepository; |
66
|
|
|
} |
67
|
|
|
|
68
|
|
|
/** |
69
|
|
|
* @access protected |
70
|
|
|
* @var LibraryRepository |
71
|
|
|
*/ |
72
|
|
|
protected $libraryRepository; |
73
|
|
|
|
74
|
|
|
/** |
75
|
|
|
* @access public |
76
|
|
|
* |
77
|
|
|
* @param LibraryRepository $libraryRepository |
78
|
|
|
* |
79
|
|
|
* @return void |
80
|
|
|
*/ |
81
|
|
|
public function injectLibraryRepository(LibraryRepository $libraryRepository) |
82
|
|
|
{ |
83
|
|
|
$this->libraryRepository = $libraryRepository; |
84
|
|
|
} |
85
|
|
|
|
86
|
|
|
/** |
87
|
|
|
* Initializes the current action |
88
|
|
|
* |
89
|
|
|
* @access public |
90
|
|
|
* |
91
|
|
|
* @return void |
92
|
|
|
*/ |
93
|
|
|
public function initializeAction() |
94
|
|
|
{ |
95
|
|
|
$this->request = $this->request->withFormat("xml"); |
96
|
|
|
} |
97
|
|
|
|
98
|
|
|
/** |
99
|
|
|
* @access protected |
100
|
|
|
* @var string Did an error occur? |
101
|
|
|
*/ |
102
|
|
|
protected $error; |
103
|
|
|
|
104
|
|
|
/** |
105
|
|
|
* @access protected |
106
|
|
|
* @var array This holds the configuration for all supported metadata prefixes |
107
|
|
|
*/ |
108
|
|
|
protected $formats = [ |
109
|
|
|
'oai_dc' => [ |
110
|
|
|
'schema' => 'http://www.openarchives.org/OAI/2.0/oai_dc.xsd', |
111
|
|
|
'namespace' => 'http://www.openarchives.org/OAI/2.0/oai_dc/', |
112
|
|
|
'requiredFields' => ['record_id'], |
113
|
|
|
], |
114
|
|
|
'epicur' => [ |
115
|
|
|
'schema' => 'http://www.persistent-identifier.de/xepicur/version1.0/xepicur.xsd', |
116
|
|
|
'namespace' => 'urn:nbn:de:1111-2004033116', |
117
|
|
|
'requiredFields' => ['purl', 'urn'], |
118
|
|
|
], |
119
|
|
|
'mets' => [ |
120
|
|
|
'schema' => 'http://www.loc.gov/standards/mets/version17/mets.v1-7.xsd', |
121
|
|
|
'namespace' => 'http://www.loc.gov/METS/', |
122
|
|
|
'requiredFields' => ['location'], |
123
|
|
|
] |
124
|
|
|
]; |
125
|
|
|
|
126
|
|
|
/** |
127
|
|
|
* @access protected |
128
|
|
|
* @var array |
129
|
|
|
*/ |
130
|
|
|
protected $parameters = []; |
131
|
|
|
|
132
|
|
|
/** |
133
|
|
|
* Delete expired resumption tokens |
134
|
|
|
* |
135
|
|
|
* @access protected |
136
|
|
|
* |
137
|
|
|
* @return void |
138
|
|
|
*/ |
139
|
|
|
protected function deleteExpiredTokens() |
140
|
|
|
{ |
141
|
|
|
// Delete expired resumption tokens. |
142
|
|
|
$this->tokenRepository->deleteExpiredTokens($this->settings['expired']); |
143
|
|
|
} |
144
|
|
|
|
145
|
|
|
/** |
146
|
|
|
* Load URL parameters |
147
|
|
|
* |
148
|
|
|
* @access protected |
149
|
|
|
* |
150
|
|
|
* @return void |
151
|
|
|
*/ |
152
|
|
|
protected function getUrlParams() |
153
|
|
|
{ |
154
|
|
|
$allowedParams = [ |
155
|
|
|
'verb', |
156
|
|
|
'identifier', |
157
|
|
|
'metadataPrefix', |
158
|
|
|
'from', |
159
|
|
|
'until', |
160
|
|
|
'set', |
161
|
|
|
'resumptionToken' |
162
|
|
|
]; |
163
|
|
|
// Clear plugin variables. |
164
|
|
|
$this->parameters = []; |
165
|
|
|
// Set only allowed parameters. |
166
|
|
|
foreach ($allowedParams as $param) { |
167
|
|
|
// replace with $this->request->getQueryParams() when dropping support for Typo3 v11, see Deprecation-100596 |
168
|
|
|
if (GeneralUtility::_GP($param)) { |
169
|
|
|
$this->parameters[$param] = GeneralUtility::_GP($param); |
170
|
|
|
} |
171
|
|
|
} |
172
|
|
|
} |
173
|
|
|
|
174
|
|
|
/** |
175
|
|
|
* Get unqualified Dublin Core data. |
176
|
|
|
* @see http://www.openarchives.org/OAI/openarchivesprotocol.html#dublincore |
177
|
|
|
* |
178
|
|
|
* @access private |
179
|
|
|
* |
180
|
|
|
* @param array $record The full record array |
181
|
|
|
* |
182
|
|
|
* @return array The mapped metadata array |
183
|
|
|
*/ |
184
|
|
|
private function getDublinCoreData(array $record) |
185
|
|
|
{ |
186
|
|
|
$metadata = []; |
187
|
|
|
|
188
|
|
|
$metadata[] = ['dc:identifier' => $record['record_id']]; |
189
|
|
|
|
190
|
|
|
$this->addDublinCoreData($metadata, 'dc:identifier', $record['purl']); |
191
|
|
|
$this->addDublinCoreData($metadata, 'dc:identifier', $record['prod_id']); |
192
|
|
|
$this->addDublinCoreData($metadata, 'dc:identifier', $record['urn']); |
193
|
|
|
$this->addDublinCoreData($metadata, 'dc:title', $record['title']); |
194
|
|
|
$this->addDublinCoreData($metadata, 'dc:creator', $record['author']); |
195
|
|
|
$this->addDublinCoreData($metadata, 'dc:date', $record['year']); |
196
|
|
|
$this->addDublinCoreData($metadata, 'dc:coverage', $record['place']); |
197
|
|
|
|
198
|
|
|
$record[] = ['dc:format' => $record['application/mets+xml']]; |
199
|
|
|
$record[] = ['dc:type' => $record['Text']]; |
200
|
|
|
if (!empty($record['partof'])) { |
201
|
|
|
$document = $this->documentRepository->findOneByPartof($metadata['partof']); |
|
|
|
|
202
|
|
|
|
203
|
|
|
if ($document) { |
204
|
|
|
$metadata[] = ['dc:relation' => $document->getRecordId()]; |
|
|
|
|
205
|
|
|
} |
206
|
|
|
} |
207
|
|
|
$this->addDublinCoreData($metadata, 'dc:rights', $record['license']); |
|
|
|
|
208
|
|
|
$this->addDublinCoreData($metadata, 'dc:rights', $record['terms']); |
209
|
|
|
$this->addDublinCoreData($metadata, 'dc:rights', $record['restrictions']); |
210
|
|
|
$this->addDublinCoreData($metadata, 'dc:rights', $record['out_of_print']); |
211
|
|
|
$this->addDublinCoreData($metadata, 'dc:rights', $record['rights_info']); |
212
|
|
|
|
213
|
|
|
return $metadata; |
214
|
|
|
} |
215
|
|
|
|
216
|
|
|
/** |
217
|
|
|
* Add Dublin Core data. |
218
|
|
|
* |
219
|
|
|
* @access private |
220
|
|
|
* |
221
|
|
|
* @param array $metadata The mapped metadata array passed as reference |
222
|
|
|
* @param string $key The key to which record value should be assigned |
223
|
|
|
* @param string $value The key from record array |
224
|
|
|
* |
225
|
|
|
* @return void |
226
|
|
|
*/ |
227
|
|
|
private function addDublinCoreData(&$metadata, $key, $value) |
228
|
|
|
{ |
229
|
|
|
if (!empty($value)) { |
230
|
|
|
$metadata[] = [$key => $value]; |
231
|
|
|
} |
232
|
|
|
} |
233
|
|
|
|
234
|
|
|
/** |
235
|
|
|
* Get METS data. |
236
|
|
|
* @see http://www.loc.gov/standards/mets/docs/mets.v1-7.html |
237
|
|
|
* |
238
|
|
|
* @access protected |
239
|
|
|
* |
240
|
|
|
* @param array $record The full record array |
241
|
|
|
* |
242
|
|
|
* @return string The fetched METS XML |
243
|
|
|
*/ |
244
|
|
|
protected function getMetsData(array $record) |
245
|
|
|
{ |
246
|
|
|
$mets = null; |
247
|
|
|
// Load METS file. |
248
|
|
|
$xml = new \DOMDocument(); |
249
|
|
|
if ($xml->load($record['location'])) { |
250
|
|
|
// Get root element. |
251
|
|
|
$root = $xml->getElementsByTagNameNS($this->formats['mets']['namespace'], 'mets'); |
252
|
|
|
if ($root->item(0) instanceof \DOMNode) { |
253
|
|
|
$mets = $xml->saveXML($root->item(0)); |
254
|
|
|
} else { |
255
|
|
|
$this->logger->error('No METS part found in document with location "' . $record['location'] . '"'); |
256
|
|
|
} |
257
|
|
|
} else { |
258
|
|
|
$this->logger->error('Could not load XML file from "' . $record['location'] . '"'); |
259
|
|
|
} |
260
|
|
|
return $mets; |
261
|
|
|
} |
262
|
|
|
|
263
|
|
|
/** |
264
|
|
|
* The main method of the plugin |
265
|
|
|
* |
266
|
|
|
* @access public |
267
|
|
|
* |
268
|
|
|
* @return ResponseInterface |
269
|
|
|
*/ |
270
|
|
|
public function mainAction(): ResponseInterface |
271
|
|
|
{ |
272
|
|
|
// Get allowed GET and POST variables. |
273
|
|
|
$this->getUrlParams(); |
274
|
|
|
|
275
|
|
|
// Delete expired resumption tokens. |
276
|
|
|
$this->deleteExpiredTokens(); |
277
|
|
|
|
278
|
|
|
switch ($this->parameters['verb'] ?? null) { |
279
|
|
|
case 'GetRecord': |
280
|
|
|
$this->verbGetRecord(); |
281
|
|
|
break; |
282
|
|
|
case 'Identify': |
283
|
|
|
$this->verbIdentify(); |
284
|
|
|
break; |
285
|
|
|
case 'ListIdentifiers': |
286
|
|
|
$this->verbListIdentifiers(); |
287
|
|
|
break; |
288
|
|
|
case 'ListMetadataFormats': |
289
|
|
|
$this->verbListMetadataFormats(); |
290
|
|
|
break; |
291
|
|
|
case 'ListRecords': |
292
|
|
|
$this->verbListRecords(); |
293
|
|
|
break; |
294
|
|
|
case 'ListSets': |
295
|
|
|
$this->verbListSets(); |
296
|
|
|
break; |
297
|
|
|
default: |
298
|
|
|
$this->error = 'badVerb'; |
299
|
|
|
break; |
300
|
|
|
} |
301
|
|
|
|
302
|
|
|
$this->view->assign('parameters', $this->parameters); |
303
|
|
|
$this->view->assign('error', $this->error); |
304
|
|
|
|
305
|
|
|
return $this->htmlResponse(); |
306
|
|
|
} |
307
|
|
|
|
308
|
|
|
/** |
309
|
|
|
* Continue with resumption token |
310
|
|
|
* |
311
|
|
|
* @access protected |
312
|
|
|
* |
313
|
|
|
* @return array|null list of uids |
314
|
|
|
*/ |
315
|
|
|
protected function resume(): ?array |
316
|
|
|
{ |
317
|
|
|
$token = $this->tokenRepository->findOneByToken($this->parameters['resumptionToken']); |
|
|
|
|
318
|
|
|
|
319
|
|
|
if ($token) { |
320
|
|
|
$options = $token->getOptions(); |
|
|
|
|
321
|
|
|
} |
322
|
|
|
if (is_array($options)) { |
|
|
|
|
323
|
|
|
return $options; |
324
|
|
|
} else { |
325
|
|
|
// No resumption token found or resumption token expired. |
326
|
|
|
$this->error = 'badResumptionToken'; |
327
|
|
|
return null; |
328
|
|
|
} |
329
|
|
|
} |
330
|
|
|
|
331
|
|
|
/** |
332
|
|
|
* Process verb "GetRecord" |
333
|
|
|
* |
334
|
|
|
* @access protected |
335
|
|
|
* |
336
|
|
|
* @return void |
337
|
|
|
*/ |
338
|
|
|
protected function verbGetRecord() |
339
|
|
|
{ |
340
|
|
|
if (count($this->parameters) !== 3 || empty($this->parameters['metadataPrefix']) || empty($this->parameters['identifier'])) { |
341
|
|
|
$this->error = 'badArgument'; |
342
|
|
|
return; |
343
|
|
|
} |
344
|
|
|
|
345
|
|
|
if (!array_key_exists($this->parameters['metadataPrefix'], $this->formats)) { |
346
|
|
|
$this->error = 'cannotDisseminateFormat'; |
347
|
|
|
return; |
348
|
|
|
} |
349
|
|
|
|
350
|
|
|
$document = $this->documentRepository->getOaiRecord($this->settings, $this->parameters); |
351
|
|
|
|
352
|
|
|
if (!$document['uid']) { |
353
|
|
|
$this->error = 'idDoesNotExist'; |
354
|
|
|
return; |
355
|
|
|
} |
356
|
|
|
|
357
|
|
|
// Check for required fields. |
358
|
|
|
foreach ($this->formats[$this->parameters['metadataPrefix']]['requiredFields'] as $required) { |
359
|
|
|
if (empty($document[$required])) { |
360
|
|
|
$this->error = 'cannotDisseminateFormat'; |
361
|
|
|
return; |
362
|
|
|
} |
363
|
|
|
} |
364
|
|
|
|
365
|
|
|
// we need the collections as array later |
366
|
|
|
$document['collections'] = explode(' ', $document['collections']); |
367
|
|
|
|
368
|
|
|
// Add metadata |
369
|
|
|
switch ($this->parameters['metadataPrefix']) { |
370
|
|
|
case 'oai_dc': |
371
|
|
|
$document['metadata'] = $this->getDublinCoreData($document); |
372
|
|
|
break; |
373
|
|
|
case 'epicur': |
374
|
|
|
$document['metadata'] = $document; |
375
|
|
|
break; |
376
|
|
|
case 'mets': |
377
|
|
|
$document['metadata'] = $this->getMetsData($document); |
378
|
|
|
break; |
379
|
|
|
} |
380
|
|
|
|
381
|
|
|
$this->view->assign('record', $document); |
382
|
|
|
} |
383
|
|
|
|
384
|
|
|
/** |
385
|
|
|
* Process verb "Identify" |
386
|
|
|
* |
387
|
|
|
* @access protected |
388
|
|
|
* |
389
|
|
|
* @return void |
390
|
|
|
*/ |
391
|
|
|
protected function verbIdentify() |
392
|
|
|
{ |
393
|
|
|
$library = $this->libraryRepository->findByUid($this->settings['library']); |
394
|
|
|
|
395
|
|
|
$oaiIdentifyInfo = []; |
396
|
|
|
|
397
|
|
|
if (!$oaiIdentifyInfo) { |
|
|
|
|
398
|
|
|
$this->logger->notice('Incomplete plugin configuration'); |
399
|
|
|
} |
400
|
|
|
|
401
|
|
|
$oaiIdentifyInfo['oai_label'] = $library ? $library->getOaiLabel() : ''; |
402
|
|
|
// Use default values for an installation with incomplete plugin configuration. |
403
|
|
|
if (empty($oaiIdentifyInfo['oai_label'])) { |
404
|
|
|
$oaiIdentifyInfo['oai_label'] = 'Kitodo.Presentation OAI-PMH Interface (default configuration)'; |
405
|
|
|
$this->logger->notice('Incomplete plugin configuration (oai_label is missing)'); |
406
|
|
|
} |
407
|
|
|
|
408
|
|
|
$oaiIdentifyInfo['contact'] = $library ? $library->getContact() : ''; |
409
|
|
|
if (empty($oaiIdentifyInfo['contact'])) { |
410
|
|
|
$oaiIdentifyInfo['contact'] = '[email protected]'; |
411
|
|
|
$this->logger->notice('Incomplete plugin configuration (contact is missing)'); |
412
|
|
|
} |
413
|
|
|
|
414
|
|
|
$document = $this->documentRepository->findOldestDocument(); |
415
|
|
|
|
416
|
|
|
if ($document) { |
417
|
|
|
$oaiIdentifyInfo['earliestDatestamp'] = gmdate('Y-m-d\TH:i:s\Z', $document->getTstamp()->getTimestamp()); |
418
|
|
|
} else { |
419
|
|
|
// Provide a fallback timestamp if no document is found |
420
|
|
|
$oaiIdentifyInfo['earliestDatestamp'] = '0000-00-00T00:00:00Z'; |
421
|
|
|
|
422
|
|
|
// access storagePid from TypoScript |
423
|
|
|
$pageSettings = $this->configurationManager->getConfiguration($this->configurationManager::CONFIGURATION_TYPE_FULL_TYPOSCRIPT); |
424
|
|
|
$storagePid = $pageSettings["plugin."]["tx_dlf."]["persistence."]["storagePid"]; |
425
|
|
|
if ($storagePid > 0) { |
426
|
|
|
$this->logger->notice('No records found with PID ' . $storagePid); |
427
|
|
|
} else { |
428
|
|
|
$this->logger->notice('No records found'); |
429
|
|
|
} |
430
|
|
|
} |
431
|
|
|
$this->view->assign('oaiIdentifyInfo', $oaiIdentifyInfo); |
432
|
|
|
} |
433
|
|
|
|
434
|
|
|
/** |
435
|
|
|
* Process verb "ListIdentifiers" |
436
|
|
|
* |
437
|
|
|
* @access protected |
438
|
|
|
* |
439
|
|
|
* @return void |
440
|
|
|
*/ |
441
|
|
|
protected function verbListIdentifiers() |
442
|
|
|
{ |
443
|
|
|
// If we have a resumption token we can continue our work |
444
|
|
|
if (!empty($this->parameters['resumptionToken'])) { |
445
|
|
|
// "resumptionToken" is an exclusive argument. |
446
|
|
|
if (count($this->parameters) > 2) { |
447
|
|
|
$this->error = 'badArgument'; |
448
|
|
|
return; |
449
|
|
|
} else { |
450
|
|
|
// return next chunk of documents |
451
|
|
|
$resultSet = $this->resume(); |
452
|
|
|
if (is_array($resultSet)) { |
453
|
|
|
$listIdentifiers = $this->generateOutputForDocumentList($resultSet); |
454
|
|
|
$this->view->assign('listIdentifiers', $listIdentifiers); |
455
|
|
|
} |
456
|
|
|
return; |
457
|
|
|
} |
458
|
|
|
} |
459
|
|
|
// "metadataPrefix" is required and "identifier" is not allowed. |
460
|
|
|
if (empty($this->parameters['metadataPrefix']) || !empty($this->parameters['identifier'])) { |
461
|
|
|
$this->error = 'badArgument'; |
462
|
|
|
return; |
463
|
|
|
} |
464
|
|
|
if (!in_array($this->parameters['metadataPrefix'], array_keys($this->formats))) { |
465
|
|
|
$this->error = 'cannotDisseminateFormat'; |
466
|
|
|
return; |
467
|
|
|
} |
468
|
|
|
try { |
469
|
|
|
$documentSet = $this->fetchDocumentSet(); |
470
|
|
|
} catch (\Exception $exception) { |
471
|
|
|
$this->error = 'idDoesNotExist'; |
472
|
|
|
return; |
473
|
|
|
} |
474
|
|
|
// create new and empty document list |
475
|
|
|
$resultSet = []; |
476
|
|
|
if (is_array($documentSet)) { |
|
|
|
|
477
|
|
|
$resultSet['elements'] = $documentSet; |
478
|
|
|
$resultSet['metadata'] = [ |
479
|
|
|
'cursor' => 0, |
480
|
|
|
'completeListSize' => count($documentSet), |
481
|
|
|
'metadataPrefix' => $this->parameters['metadataPrefix'], |
482
|
|
|
]; |
483
|
|
|
} |
484
|
|
|
|
485
|
|
|
$listIdentifiers = $this->generateOutputForDocumentList($resultSet); |
486
|
|
|
$this->view->assign('listIdentifiers', $listIdentifiers); |
487
|
|
|
} |
488
|
|
|
|
489
|
|
|
/** |
490
|
|
|
* Process verb "ListMetadataFormats" |
491
|
|
|
* |
492
|
|
|
* @access protected |
493
|
|
|
* |
494
|
|
|
* @return void |
495
|
|
|
*/ |
496
|
|
|
protected function verbListMetadataFormats() |
497
|
|
|
{ |
498
|
|
|
$resArray = []; |
499
|
|
|
// check for the optional "identifier" parameter |
500
|
|
|
if (isset($this->parameters['identifier'])) { |
501
|
|
|
$resArray = $this->documentRepository->findOneByRecordId($this->parameters['identifier']); |
|
|
|
|
502
|
|
|
} |
503
|
|
|
|
504
|
|
|
$resultSet = []; |
505
|
|
|
foreach ($this->formats as $prefix => $details) { |
506
|
|
|
if (!empty($resArray)) { |
507
|
|
|
// check, if all required fields are available for a given identifier |
508
|
|
|
foreach ($details['requiredFields'] as $required) { |
509
|
|
|
$methodName = 'get' . GeneralUtility::underscoredToUpperCamelCase($required); |
510
|
|
|
if (empty($resArray->$methodName())) { |
511
|
|
|
// Skip metadata formats whose requirements are not met. |
512
|
|
|
continue 2; |
513
|
|
|
} |
514
|
|
|
} |
515
|
|
|
} |
516
|
|
|
$details['prefix'] = $prefix; |
517
|
|
|
$resultSet[] = $details; |
518
|
|
|
} |
519
|
|
|
$this->view->assign('metadataFormats', $resultSet); |
520
|
|
|
} |
521
|
|
|
|
522
|
|
|
/** |
523
|
|
|
* Process verb "ListRecords" |
524
|
|
|
* |
525
|
|
|
* @access protected |
526
|
|
|
* |
527
|
|
|
* @return void |
528
|
|
|
*/ |
529
|
|
|
protected function verbListRecords() |
530
|
|
|
{ |
531
|
|
|
// Check for invalid arguments. |
532
|
|
|
if (!empty($this->parameters['resumptionToken'])) { |
533
|
|
|
// "resumptionToken" is an exclusive argument. |
534
|
|
|
if (count($this->parameters) > 2) { |
535
|
|
|
$this->error = 'badArgument'; |
536
|
|
|
return; |
537
|
|
|
} else { |
538
|
|
|
// return next chunk of documents |
539
|
|
|
$resultSet = $this->resume(); |
540
|
|
|
if (is_array($resultSet)) { |
541
|
|
|
$listRecords = $this->generateOutputForDocumentList($resultSet); |
542
|
|
|
$this->parameters['metadataPrefix'] = $resultSet['metadata']['metadataPrefix']; |
543
|
|
|
$this->view->assign('listRecords', $listRecords); |
544
|
|
|
} |
545
|
|
|
return; |
546
|
|
|
} |
547
|
|
|
} |
548
|
|
|
if (empty($this->parameters['metadataPrefix']) || !empty($this->parameters['identifier'])) { |
549
|
|
|
// "metadataPrefix" is required and "identifier" is not allowed. |
550
|
|
|
$this->error = 'badArgument'; |
551
|
|
|
return; |
552
|
|
|
} |
553
|
|
|
// Check "metadataPrefix" for valid value. |
554
|
|
|
if (!in_array($this->parameters['metadataPrefix'], array_keys($this->formats))) { |
555
|
|
|
$this->error = 'cannotDisseminateFormat'; |
556
|
|
|
return; |
557
|
|
|
} |
558
|
|
|
try { |
559
|
|
|
$documentSet = $this->fetchDocumentSet(); |
560
|
|
|
} catch (\Exception $exception) { |
561
|
|
|
$this->error = 'idDoesNotExist'; |
562
|
|
|
return; |
563
|
|
|
} |
564
|
|
|
$resultSet = []; |
565
|
|
|
if (count($documentSet) > 0) { |
566
|
|
|
$resultSet['elements'] = $documentSet; |
567
|
|
|
$resultSet['metadata'] = [ |
568
|
|
|
'cursor' => 0, |
569
|
|
|
'completeListSize' => count($documentSet), |
570
|
|
|
'metadataPrefix' => $this->parameters['metadataPrefix'], |
571
|
|
|
]; |
572
|
|
|
} |
573
|
|
|
|
574
|
|
|
$resultSet = $this->generateOutputForDocumentList($resultSet); |
575
|
|
|
$this->view->assign('listRecords', $resultSet); |
576
|
|
|
} |
577
|
|
|
|
578
|
|
|
/** |
579
|
|
|
* Process verb "ListSets" |
580
|
|
|
* |
581
|
|
|
* @access protected |
582
|
|
|
* |
583
|
|
|
* @return void |
584
|
|
|
*/ |
585
|
|
|
protected function verbListSets() |
586
|
|
|
{ |
587
|
|
|
// It is required to set a oai_name inside the collection record to be shown in oai-pmh plugin. |
588
|
|
|
$this->settings['hideEmptyOaiNames'] = true; |
589
|
|
|
|
590
|
|
|
$oaiSets = $this->collectionRepository->findCollectionsBySettings($this->settings); |
591
|
|
|
|
592
|
|
|
$this->view->assign('oaiSets', $oaiSets); |
593
|
|
|
} |
594
|
|
|
|
595
|
|
|
/** |
596
|
|
|
* Fetch records |
597
|
|
|
* |
598
|
|
|
* @access protected |
599
|
|
|
* |
600
|
|
|
* @return array matching records or empty array if there were some errors |
601
|
|
|
*/ |
602
|
|
|
protected function fetchDocumentSet(): array |
603
|
|
|
{ |
604
|
|
|
$documentSet = []; |
605
|
|
|
$solrQuery = ''; |
606
|
|
|
// Check "set" for valid value. |
607
|
|
|
if (!empty($this->parameters['set'])) { |
608
|
|
|
// For SOLR we need the index_name of the collection, |
609
|
|
|
// For DB Query we need the UID of the collection |
610
|
|
|
|
611
|
|
|
$result = $this->collectionRepository->getIndexNameForSolr($this->settings, $this->parameters['set']); |
612
|
|
|
$resArray = $result->fetchAssociative(); |
613
|
|
|
if ($resArray) { |
614
|
|
|
if ($resArray['index_query'] != "") { |
615
|
|
|
$solrQuery .= '(' . $resArray['index_query'] . ')'; |
616
|
|
|
} else { |
617
|
|
|
$solrQuery .= 'collection:' . '"' . $resArray['index_name'] . '"'; |
618
|
|
|
} |
619
|
|
|
} else { |
620
|
|
|
$this->error = 'noSetHierarchy'; |
621
|
|
|
return $documentSet; |
622
|
|
|
} |
623
|
|
|
} else { |
624
|
|
|
// If no set is specified we have to query for all collections |
625
|
|
|
$solrQuery .= 'collection:* NOT collection:""'; |
626
|
|
|
} |
627
|
|
|
// Check for required fields. |
628
|
|
|
foreach ($this->formats[$this->parameters['metadataPrefix']]['requiredFields'] as $required) { |
629
|
|
|
$solrQuery .= ' NOT ' . $required . ':""'; |
630
|
|
|
} |
631
|
|
|
// toplevel="true" is always required |
632
|
|
|
$solrQuery .= ' AND toplevel:true'; |
633
|
|
|
|
634
|
|
|
$from = $this->getFrom(); |
635
|
|
|
$until = $this->getUntil($from); |
636
|
|
|
|
637
|
|
|
$this->checkGranularity(); |
638
|
|
|
|
639
|
|
|
if ($this->error === 'badArgument') { |
640
|
|
|
return $documentSet; |
641
|
|
|
} |
642
|
|
|
|
643
|
|
|
$solrQuery .= ' AND timestamp:[' . $from . ' TO ' . $until . ']'; |
644
|
|
|
|
645
|
|
|
$solr = Solr::getInstance($this->settings['solrcore']); |
646
|
|
|
if (!$solr->ready) { |
647
|
|
|
$this->logger->error('Apache Solr not available'); |
648
|
|
|
return $documentSet; |
649
|
|
|
} |
650
|
|
|
if ($this->settings['solr_limit'] > 0) { |
651
|
|
|
$solr->limit = $this->settings['solr_limit']; |
652
|
|
|
} |
653
|
|
|
// We only care about the UID in the results and want them sorted |
654
|
|
|
$parameters = [ |
655
|
|
|
"fields" => "uid", |
656
|
|
|
"sort" => [ |
657
|
|
|
"uid" => "asc" |
658
|
|
|
] |
659
|
|
|
]; |
660
|
|
|
$parameters['query'] = $solrQuery; |
661
|
|
|
$result = $solr->searchRaw($parameters); |
662
|
|
|
if (empty($result)) { |
663
|
|
|
$this->error = 'noRecordsMatch'; |
664
|
|
|
return $documentSet; |
665
|
|
|
} |
666
|
|
|
foreach ($result as $doc) { |
667
|
|
|
$documentSet[] = $doc->uid; |
668
|
|
|
} |
669
|
|
|
return $documentSet; |
670
|
|
|
} |
671
|
|
|
|
672
|
|
|
/** |
673
|
|
|
* Get 'from' query parameter. |
674
|
|
|
* |
675
|
|
|
* @access private |
676
|
|
|
* |
677
|
|
|
* @return string |
678
|
|
|
*/ |
679
|
|
|
private function getFrom(): string |
680
|
|
|
{ |
681
|
|
|
$from = "*"; |
682
|
|
|
// Check "from" for valid value. |
683
|
|
|
if (!empty($this->parameters['from'])) { |
684
|
|
|
// Is valid format? |
685
|
|
|
$date = $this->getDate('from'); |
686
|
|
|
if (is_array($date)) { |
|
|
|
|
687
|
|
|
$from = $this->getDateFromTimestamp($date, '.000Z'); |
688
|
|
|
} else { |
689
|
|
|
$this->error = 'badArgument'; |
690
|
|
|
} |
691
|
|
|
} |
692
|
|
|
return $from; |
693
|
|
|
} |
694
|
|
|
|
695
|
|
|
/** |
696
|
|
|
* Get 'until' query parameter. |
697
|
|
|
* |
698
|
|
|
* @access private |
699
|
|
|
* |
700
|
|
|
* @param string $from start date |
701
|
|
|
* |
702
|
|
|
* @return string |
703
|
|
|
*/ |
704
|
|
|
private function getUntil(string $from): string |
705
|
|
|
{ |
706
|
|
|
$until = "*"; |
707
|
|
|
// Check "until" for valid value. |
708
|
|
|
if (!empty($this->parameters['until'])) { |
709
|
|
|
// Is valid format? |
710
|
|
|
$date = $this->getDate('until'); |
711
|
|
|
if (is_array($date)) { |
|
|
|
|
712
|
|
|
$until = $this->getDateFromTimestamp($date, '.999Z'); |
713
|
|
|
if ($from != "*" && $from > $until) { |
714
|
|
|
$this->error = 'badArgument'; |
715
|
|
|
} |
716
|
|
|
} else { |
717
|
|
|
$this->error = 'badArgument'; |
718
|
|
|
} |
719
|
|
|
} |
720
|
|
|
return $until; |
721
|
|
|
} |
722
|
|
|
|
723
|
|
|
/** |
724
|
|
|
* Get date from parameter string. |
725
|
|
|
* |
726
|
|
|
* @access private |
727
|
|
|
* |
728
|
|
|
* @param string $dateType |
729
|
|
|
* |
730
|
|
|
* @return array|false |
731
|
|
|
*/ |
732
|
|
|
private function getDate(string $dateType) |
733
|
|
|
{ |
734
|
|
|
return strptime($this->parameters[$dateType], '%Y-%m-%dT%H:%M:%SZ') ?: strptime($this->parameters[$dateType], '%Y-%m-%d'); |
735
|
|
|
} |
736
|
|
|
|
737
|
|
|
/** |
738
|
|
|
* Get date from timestamp. |
739
|
|
|
* |
740
|
|
|
* @access private |
741
|
|
|
* |
742
|
|
|
* @param array $date |
743
|
|
|
* @param string $end |
744
|
|
|
* |
745
|
|
|
* @return string |
746
|
|
|
*/ |
747
|
|
|
private function getDateFromTimestamp(array $date, string $end): string |
748
|
|
|
{ |
749
|
|
|
$timestamp = gmmktime( |
750
|
|
|
$date['tm_hour'], |
751
|
|
|
$date['tm_min'], |
752
|
|
|
$date['tm_sec'], |
753
|
|
|
$date['tm_mon'] + 1, |
754
|
|
|
$date['tm_mday'], |
755
|
|
|
$date['tm_year'] + 1900 |
756
|
|
|
); |
757
|
|
|
return date("Y-m-d", $timestamp) . 'T' . date("H:i:s", $timestamp) . $end; |
758
|
|
|
} |
759
|
|
|
|
760
|
|
|
/** |
761
|
|
|
* Check "from" and "until" for same granularity. |
762
|
|
|
* |
763
|
|
|
* @access private |
764
|
|
|
* |
765
|
|
|
* @return void |
766
|
|
|
*/ |
767
|
|
|
private function checkGranularity(): void |
768
|
|
|
{ |
769
|
|
|
if ( |
770
|
|
|
!empty($this->parameters['from']) |
771
|
|
|
&& !empty($this->parameters['until']) |
772
|
|
|
) { |
773
|
|
|
if (strlen($this->parameters['from']) != strlen($this->parameters['until'])) { |
774
|
|
|
$this->error = 'badArgument'; |
775
|
|
|
} |
776
|
|
|
} |
777
|
|
|
} |
778
|
|
|
|
779
|
|
|
/** |
780
|
|
|
* Fetch more information for document list |
781
|
|
|
* |
782
|
|
|
* @access protected |
783
|
|
|
* |
784
|
|
|
* @param array $documentListSet |
785
|
|
|
* |
786
|
|
|
* @return array of enriched records |
787
|
|
|
*/ |
788
|
|
|
protected function generateOutputForDocumentList(array $documentListSet) |
789
|
|
|
{ |
790
|
|
|
// check whether any result elements are available |
791
|
|
|
if (empty($documentListSet) || empty($documentListSet['elements'])) { |
792
|
|
|
$this->error = 'noRecordsMatch'; |
793
|
|
|
return []; |
794
|
|
|
} |
795
|
|
|
// consume result elements from list to implement pagination logic of resumptionToken |
796
|
|
|
$documentsToProcess = array_splice($documentListSet['elements'], 0, $this->settings['limit']); |
797
|
|
|
$verb = $this->parameters['verb']; |
798
|
|
|
|
799
|
|
|
$documents = $this->documentRepository->getOaiDocumentList($documentsToProcess); |
800
|
|
|
|
801
|
|
|
$records = []; |
802
|
|
|
while ($resArray = $documents->fetchAssociative()) { |
803
|
|
|
// we need the collections as array later |
804
|
|
|
$resArray['collections'] = explode(' ', $resArray['collections']); |
805
|
|
|
|
806
|
|
|
if ($verb === 'ListRecords') { |
807
|
|
|
// Add metadata node. |
808
|
|
|
$metadataPrefix = $this->parameters['metadataPrefix']; |
809
|
|
|
if (!$metadataPrefix) { |
810
|
|
|
// If we resume an action the metadataPrefix is stored with the documentSet |
811
|
|
|
$metadataPrefix = $documentListSet['metadata']['metadataPrefix']; |
812
|
|
|
} |
813
|
|
|
switch ($metadataPrefix) { |
814
|
|
|
case 'oai_dc': |
815
|
|
|
$resArray['metadata'] = $this->getDublinCoreData($resArray); |
816
|
|
|
break; |
817
|
|
|
case 'epicur': |
818
|
|
|
$resArray['metadata'] = $resArray; |
819
|
|
|
break; |
820
|
|
|
case 'mets': |
821
|
|
|
$resArray['metadata'] = $this->getMetsData($resArray); |
822
|
|
|
break; |
823
|
|
|
} |
824
|
|
|
} |
825
|
|
|
|
826
|
|
|
$records[] = $resArray; |
827
|
|
|
} |
828
|
|
|
|
829
|
|
|
$this->generateResumptionTokenForDocumentListSet($documentListSet, count($documentsToProcess)); |
830
|
|
|
|
831
|
|
|
return $records; |
832
|
|
|
} |
833
|
|
|
|
834
|
|
|
/** |
835
|
|
|
* Generate resumption token |
836
|
|
|
* |
837
|
|
|
* @access protected |
838
|
|
|
* |
839
|
|
|
* @param array $documentListSet |
840
|
|
|
* @param int $numShownDocuments |
841
|
|
|
* |
842
|
|
|
* @return void |
843
|
|
|
*/ |
844
|
|
|
protected function generateResumptionTokenForDocumentListSet(array $documentListSet, int $numShownDocuments) |
845
|
|
|
{ |
846
|
|
|
// The cursor specifies how many elements have already been returned in previous requests |
847
|
|
|
// See http://www.openarchives.org/OAI/openarchivesprotocol.html#FlowControl |
848
|
|
|
$currentCursor = $documentListSet['metadata']['cursor']; |
849
|
|
|
|
850
|
|
|
if (count($documentListSet['elements']) !== 0) { |
851
|
|
|
$resumptionToken = uniqid('', false); |
852
|
|
|
|
853
|
|
|
$documentListSet['metadata']['cursor'] += $numShownDocuments; |
854
|
|
|
|
855
|
|
|
// create new token |
856
|
|
|
$newToken = GeneralUtility::makeInstance(Token::class); |
857
|
|
|
$newToken->setToken($resumptionToken); |
858
|
|
|
$newToken->setOptions($documentListSet); |
859
|
|
|
|
860
|
|
|
// add to tokenRepository |
861
|
|
|
$this->tokenRepository->add($newToken); |
862
|
|
|
} else { |
863
|
|
|
// Result set complete. We don't need a token. |
864
|
|
|
$resumptionToken = ''; |
865
|
|
|
} |
866
|
|
|
|
867
|
|
|
$resumptionTokenInfo = []; |
868
|
|
|
$resumptionTokenInfo['token'] = $resumptionToken; |
869
|
|
|
$resumptionTokenInfo['cursor'] = $currentCursor; |
870
|
|
|
$resumptionTokenInfo['completeListSize'] = $documentListSet['metadata']['completeListSize']; |
871
|
|
|
$expireDateTime = new \DateTime(); |
872
|
|
|
$expireDateTime->add(new \DateInterval('PT' . $this->settings['expired'] . 'S')); |
873
|
|
|
$resumptionTokenInfo['expired'] = $expireDateTime; |
874
|
|
|
|
875
|
|
|
$omitResumptionToken = $currentCursor === 0 && $numShownDocuments >= $documentListSet['metadata']['completeListSize']; |
876
|
|
|
if (!$omitResumptionToken) { |
877
|
|
|
$this->view->assign('resumptionToken', $resumptionTokenInfo); |
878
|
|
|
} |
879
|
|
|
} |
880
|
|
|
} |
881
|
|
|
|