1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
namespace Kitodo\Dlf\Tests\Functional\Api; |
4
|
|
|
|
5
|
|
|
use DateTime; |
6
|
|
|
use GuzzleHttp\Client as HttpClient; |
7
|
|
|
use Kitodo\Dlf\Common\Solr; |
8
|
|
|
use Kitodo\Dlf\Domain\Repository\SolrCoreRepository; |
9
|
|
|
use Kitodo\Dlf\Tests\Functional\FunctionalTestCase; |
10
|
|
|
use Phpoaipmh\Endpoint; |
11
|
|
|
use Phpoaipmh\Exception\OaipmhException; |
12
|
|
|
use SimpleXMLElement; |
13
|
|
|
use TYPO3\CMS\Extbase\Persistence\Generic\PersistenceManager; |
14
|
|
|
|
15
|
|
|
class OaiPmhTest extends FunctionalTestCase |
16
|
|
|
{ |
17
|
|
|
protected $disableJsonWrappedResponse = true; |
18
|
|
|
|
19
|
|
|
protected $coreExtensionsToLoad = [ |
20
|
|
|
'fluid', |
21
|
|
|
'fluid_styled_content', |
22
|
|
|
]; |
23
|
|
|
|
24
|
|
|
/** @var int */ |
25
|
|
|
protected $oaiPage = 20001; |
26
|
|
|
|
27
|
|
|
/** @var string */ |
28
|
|
|
protected $oaiUrl; |
29
|
|
|
|
30
|
|
|
/** @var int */ |
31
|
|
|
protected $oaiPageNoStoragePid = 20002; |
32
|
|
|
|
33
|
|
|
/** @var string */ |
34
|
|
|
protected $oaiUrlNoStoragePid; |
35
|
|
|
|
36
|
|
|
public function setUp(): void |
37
|
|
|
{ |
38
|
|
|
parent::setUp(); |
39
|
|
|
|
40
|
|
|
$this->oaiUrl = $this->baseUrl . '/index.php?id=' . $this->oaiPage; |
41
|
|
|
$this->oaiUrlNoStoragePid = $this->baseUrl . '/index.php?id=' . $this->oaiPageNoStoragePid; |
42
|
|
|
|
43
|
|
|
$this->importDataSet(__DIR__ . '/../../Fixtures/Common/documents_1.xml'); |
44
|
|
|
$this->importDataSet(__DIR__ . '/../../Fixtures/Common/metadata.xml'); |
45
|
|
|
$this->importDataSet(__DIR__ . '/../../Fixtures/Common/libraries.xml'); |
46
|
|
|
$this->importDataSet(__DIR__ . '/../../Fixtures/Common/pages.xml'); |
47
|
|
|
$this->importDataSet(__DIR__ . '/../../Fixtures/OaiPmh/pages.xml'); |
48
|
|
|
$this->importDataSet(__DIR__ . '/../../Fixtures/OaiPmh/solrcores.xml'); |
49
|
|
|
|
50
|
|
|
$this->persistenceManager = $this->objectManager->get(PersistenceManager::class); |
|
|
|
|
51
|
|
|
$this->solrCoreRepository = $this->initializeRepository(SolrCoreRepository::class, 20000); |
|
|
|
|
52
|
|
|
|
53
|
|
|
$this->setUpOaiSolr(); |
54
|
|
|
} |
55
|
|
|
|
56
|
|
|
protected function setUpOaiSolr() |
57
|
|
|
{ |
58
|
|
|
// Setup Solr only once for all tests in this suite |
59
|
|
|
static $solr = null; |
60
|
|
|
|
61
|
|
|
if ($solr === null) { |
62
|
|
|
$coreName = Solr::createCore(); |
63
|
|
|
$solr = Solr::getInstance($coreName); |
64
|
|
|
|
65
|
|
|
$this->importSolrDocuments($solr, __DIR__ . '/../../Fixtures/Common/documents_1.solr.json'); |
66
|
|
|
} |
67
|
|
|
|
68
|
|
|
$oaiCoreModel = $this->solrCoreRepository->findByUid(11001); |
69
|
|
|
$oaiCoreModel->setIndexName($solr->core); |
70
|
|
|
$this->solrCoreRepository->update($oaiCoreModel); |
71
|
|
|
$this->persistenceManager->persistAll(); |
72
|
|
|
} |
73
|
|
|
|
74
|
|
|
protected function importSolrDocuments(Solr $solr, string $path) |
75
|
|
|
{ |
76
|
|
|
$jsonDocuments = json_decode(file_get_contents($path), true); |
77
|
|
|
|
78
|
|
|
$updateQuery = $solr->service->createUpdate(); |
79
|
|
|
$documents = array_map(function ($jsonDoc) use ($updateQuery) { |
80
|
|
|
$document = $updateQuery->createDocument(); |
81
|
|
|
foreach ($jsonDoc as $key => $value) { |
82
|
|
|
$document->setField($key, $value); |
|
|
|
|
83
|
|
|
} |
84
|
|
|
return $document; |
85
|
|
|
}, $jsonDocuments); |
86
|
|
|
$updateQuery->addDocuments($documents); |
87
|
|
|
$updateQuery->addCommit(); |
88
|
|
|
$solr->service->update($updateQuery); |
89
|
|
|
} |
90
|
|
|
|
91
|
|
|
/** |
92
|
|
|
* @test |
93
|
|
|
*/ |
94
|
|
|
public function correctlyRespondsOnBadVerb() |
95
|
|
|
{ |
96
|
|
|
$client = new HttpClient(); |
97
|
|
|
$response = $client->get($this->baseUrl, [ |
98
|
|
|
'query' => [ |
99
|
|
|
'id' => $this->oaiPage, |
100
|
|
|
'verb' => 'nastyVerb', |
101
|
|
|
], |
102
|
|
|
]); |
103
|
|
|
$xml = new SimpleXMLElement((string) $response->getBody()); |
104
|
|
|
|
105
|
|
|
$this->assertEquals('badVerb', (string) $xml->error['code']); |
106
|
|
|
|
107
|
|
|
// The base URL may be different from the one used that we actually used, |
108
|
|
|
// but it shouldn't contain the verb argument |
109
|
|
|
$this->assertStringNotContainsString('nastyVerb', (string) $xml->request); |
110
|
|
|
|
111
|
|
|
// For bad verbs, the <request> element must not contain any attributes |
112
|
|
|
// - http://www.openarchives.org/OAI/openarchivesprotocol.html#XMLResponse |
113
|
|
|
// - http://www.openarchives.org/OAI/openarchivesprotocol.html#ErrorConditions |
114
|
|
|
$this->assertEmpty($xml->request->attributes()); |
115
|
|
|
} |
116
|
|
|
|
117
|
|
|
/** |
118
|
|
|
* @test |
119
|
|
|
*/ |
120
|
|
|
public function canIdentify() |
121
|
|
|
{ |
122
|
|
|
$oai = Endpoint::build($this->oaiUrl); |
123
|
|
|
$identity = $oai->identify(); |
124
|
|
|
|
125
|
|
|
$this->assertEquals('Identify', (string) $identity->request['verb']); |
126
|
|
|
$this->assertEquals('Default Library - OAI Repository', (string) $identity->Identify->repositoryName); |
127
|
|
|
$this->assertUtcDateString((string) $identity->Identify->earliestDatestamp); |
128
|
|
|
$this->assertEquals('[email protected]', (string) $identity->Identify->adminEmail); |
129
|
|
|
} |
130
|
|
|
|
131
|
|
|
/** |
132
|
|
|
* @test |
133
|
|
|
*/ |
134
|
|
|
public function identifyGivesFallbackDatestampWhenNoDocuments() |
135
|
|
|
{ |
136
|
|
|
$oai = Endpoint::build($this->oaiUrlNoStoragePid); |
137
|
|
|
$identity = $oai->identify(); |
138
|
|
|
|
139
|
|
|
$this->assertUtcDateString((string) $identity->Identify->earliestDatestamp); |
140
|
|
|
} |
141
|
|
|
|
142
|
|
|
/** |
143
|
|
|
* @test |
144
|
|
|
*/ |
145
|
|
|
public function canListMetadataFormats() |
146
|
|
|
{ |
147
|
|
|
$oai = Endpoint::build($this->oaiUrl); |
148
|
|
|
$formats = $oai->listMetadataFormats(); |
149
|
|
|
|
150
|
|
|
$formatMap = []; |
151
|
|
|
foreach ($formats as $format) { |
152
|
|
|
$formatMap[(string) $format->metadataPrefix] = $format; |
153
|
|
|
} |
154
|
|
|
|
155
|
|
|
$this->assertEquals('http://www.loc.gov/METS/', (string) $formatMap['mets']->metadataNamespace); |
156
|
|
|
} |
157
|
|
|
|
158
|
|
|
/** |
159
|
|
|
* @test |
160
|
|
|
*/ |
161
|
|
|
public function canListRecords() |
162
|
|
|
{ |
163
|
|
|
$oai = Endpoint::build($this->oaiUrl); |
164
|
|
|
$result = $oai->listRecords('mets'); |
165
|
|
|
|
166
|
|
|
$record = $result->current(); |
167
|
|
|
$metsRoot = $record->metadata->children('http://www.loc.gov/METS/')[0]; |
168
|
|
|
$this->assertNotNull($metsRoot); |
169
|
|
|
$this->assertEquals('mets', $metsRoot->getName()); |
170
|
|
|
} |
171
|
|
|
|
172
|
|
|
/** |
173
|
|
|
* @test |
174
|
|
|
*/ |
175
|
|
|
public function noRecordsUntil1900() |
176
|
|
|
{ |
177
|
|
|
$this->expectException(OaipmhException::class); |
178
|
|
|
$this->expectExceptionMessage('empty list'); |
179
|
|
|
|
180
|
|
|
$oai = Endpoint::build($this->oaiUrl); |
181
|
|
|
$result = $oai->listRecords('mets', null, (new DateTime())->setDate(1900, 1, 1)); |
182
|
|
|
|
183
|
|
|
$result->current(); |
184
|
|
|
} |
185
|
|
|
|
186
|
|
|
/** |
187
|
|
|
* @test |
188
|
|
|
*/ |
189
|
|
|
public function canUseResumptionToken() |
190
|
|
|
{ |
191
|
|
|
// NOTE: cursor and expirationDate are optional by the specification, |
192
|
|
|
// but we include them in our implementation |
193
|
|
|
|
194
|
|
|
$client = new HttpClient(); |
195
|
|
|
|
196
|
|
|
// The general handling of resumption tokens should be the same for these verbs |
197
|
|
|
foreach (['ListIdentifiers', 'ListRecords'] as $verb) { |
198
|
|
|
// Check that we get a proper resumption token when starting a list |
199
|
|
|
$response = $client->get($this->baseUrl, [ |
200
|
|
|
'query' => [ |
201
|
|
|
'id' => $this->oaiPage, |
202
|
|
|
'verb' => $verb, |
203
|
|
|
'metadataPrefix' => 'mets', |
204
|
|
|
], |
205
|
|
|
]); |
206
|
|
|
$xml = new SimpleXMLElement((string) $response->getBody()); |
207
|
|
|
|
208
|
|
|
$resumptionToken = $xml->$verb->resumptionToken; |
209
|
|
|
$this->assertEquals('0', (string) $resumptionToken['cursor']); |
210
|
|
|
$this->assertInFuture((string) $resumptionToken['expirationDate']); |
211
|
|
|
$this->assertNotEmpty((string) $resumptionToken); |
212
|
|
|
|
213
|
|
|
// Store list size to check that it remains constant (and check its sanity) |
214
|
|
|
$completeListSize = (int) $resumptionToken['completeListSize']; |
215
|
|
|
$this->assertGreaterThan(2, $completeListSize); // we have more than two documents in document set |
216
|
|
|
|
217
|
|
|
// Check that we can resume and get a proper cursor value |
218
|
|
|
$cursor = 1; |
219
|
|
|
do { |
220
|
|
|
$response = $client->get($this->baseUrl, [ |
221
|
|
|
'query' => [ |
222
|
|
|
'id' => $this->oaiPage, |
223
|
|
|
'verb' => $verb, |
224
|
|
|
'resumptionToken' => (string) $resumptionToken, |
225
|
|
|
], |
226
|
|
|
]); |
227
|
|
|
$xml = new SimpleXMLElement((string) $response->getBody()); |
228
|
|
|
|
229
|
|
|
$resumptionToken = $xml->$verb->resumptionToken; |
230
|
|
|
$tokenStr = (string) $resumptionToken; |
231
|
|
|
|
232
|
|
|
$this->assertEquals($cursor, (string) $resumptionToken['cursor']); // settings.limit = 1 |
233
|
|
|
$this->assertEquals($completeListSize, (string) $resumptionToken['completeListSize']); |
234
|
|
|
|
235
|
|
|
// The last resumptionToken is empty and doesn't have expirationDate |
236
|
|
|
$isLastBatch = $cursor + 1 >= $completeListSize; |
237
|
|
|
$this->assertEquals($isLastBatch, empty((string) $resumptionToken['expirationDate'])); |
238
|
|
|
$this->assertEquals($isLastBatch, empty($tokenStr)); |
239
|
|
|
|
240
|
|
|
$cursor++; |
241
|
|
|
} while ($tokenStr); |
242
|
|
|
} |
243
|
|
|
} |
244
|
|
|
|
245
|
|
|
/** |
246
|
|
|
* @test |
247
|
|
|
*/ |
248
|
|
|
public function noResumptionTokenForCompleteList() |
249
|
|
|
{ |
250
|
|
|
$client = new HttpClient(); |
251
|
|
|
|
252
|
|
|
foreach (['ListIdentifiers', 'ListRecords'] as $verb) { |
253
|
|
|
$response = $client->get($this->baseUrl, [ |
254
|
|
|
'query' => [ |
255
|
|
|
'id' => $this->oaiPage, |
256
|
|
|
'verb' => $verb, |
257
|
|
|
'metadataPrefix' => 'mets', |
258
|
|
|
'set' => 'collection-with-single-document', |
259
|
|
|
], |
260
|
|
|
]); |
261
|
|
|
$xml = new SimpleXMLElement((string) $response->getBody()); |
262
|
|
|
|
263
|
|
|
$this->assertEquals(1, count($xml->$verb->children())); |
264
|
|
|
$this->assertEmpty($xml->$verb->resumptionToken); |
265
|
|
|
} |
266
|
|
|
} |
267
|
|
|
|
268
|
|
|
/** |
269
|
|
|
* @test |
270
|
|
|
*/ |
271
|
|
|
public function canListAndResumeIdentifiers() |
272
|
|
|
{ |
273
|
|
|
$oai = Endpoint::build($this->oaiUrl); |
274
|
|
|
$result = $oai->listIdentifiers('mets'); |
275
|
|
|
|
276
|
|
|
$record = $result->current(); |
277
|
|
|
$this->assertEquals('oai:de:slub-dresden:db:id-476251419', $record->identifier); |
278
|
|
|
$this->assertEquals(['collection-with-single-document', 'music'], (array) $record->setSpec); |
279
|
|
|
|
280
|
|
|
// This should use a resumption token because settings.limit is 1 |
281
|
|
|
$record = $result->next(); |
282
|
|
|
$this->assertEquals('oai:de:slub-dresden:db:id-476248086', $record->identifier); |
283
|
|
|
} |
284
|
|
|
|
285
|
|
|
protected function parseUtc(string $dateTime) |
286
|
|
|
{ |
287
|
|
|
return DateTime::createFromFormat('Y-m-d\TH:i:s\Z', $dateTime); |
288
|
|
|
} |
289
|
|
|
|
290
|
|
|
protected function assertUtcDateString(string $dateTime) |
291
|
|
|
{ |
292
|
|
|
$this->assertInstanceOf(DateTime::class, $this->parseUtc($dateTime)); |
293
|
|
|
} |
294
|
|
|
|
295
|
|
|
protected function assertInFuture(string $dateTime) |
296
|
|
|
{ |
297
|
|
|
$this->assertGreaterThan(new DateTime(), $this->parseUtc($dateTime)); |
298
|
|
|
} |
299
|
|
|
} |
300
|
|
|
|