1
|
|
|
<?php |
2
|
|
|
/** |
3
|
|
|
* FullTextSearch_ElasticSearch - Use Elasticsearch to index the content of your nextcloud |
4
|
|
|
* |
5
|
|
|
* This file is licensed under the Affero General Public License version 3 or |
6
|
|
|
* later. See the COPYING file. |
7
|
|
|
* |
8
|
|
|
* @author Maxence Lange <[email protected]> |
9
|
|
|
* @copyright 2018 |
10
|
|
|
* @license GNU AGPL version 3 or any later version |
11
|
|
|
* |
12
|
|
|
* This program is free software: you can redistribute it and/or modify |
13
|
|
|
* it under the terms of the GNU Affero General Public License as |
14
|
|
|
* published by the Free Software Foundation, either version 3 of the |
15
|
|
|
* License, or (at your option) any later version. |
16
|
|
|
* |
17
|
|
|
* This program is distributed in the hope that it will be useful, |
18
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
19
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
20
|
|
|
* GNU Affero General Public License for more details. |
21
|
|
|
* |
22
|
|
|
* You should have received a copy of the GNU Affero General Public License |
23
|
|
|
* along with this program. If not, see <http://www.gnu.org/licenses/>. |
24
|
|
|
* |
25
|
|
|
*/ |
26
|
|
|
|
27
|
|
|
namespace OCA\FullTextSearch_ElasticSearch\Platform; |
28
|
|
|
|
29
|
|
|
use Elasticsearch\Client; |
30
|
|
|
use Elasticsearch\ClientBuilder; |
31
|
|
|
use Elasticsearch\Common\Exceptions\BadRequest400Exception; |
32
|
|
|
use Exception; |
33
|
|
|
use OCA\FullTextSearch\IFullTextSearchPlatform; |
34
|
|
|
use OCA\FullTextSearch\IFullTextSearchProvider; |
35
|
|
|
use OCA\FullTextSearch\Model\DocumentAccess; |
36
|
|
|
use OCA\FullTextSearch\Model\Index; |
37
|
|
|
use OCA\FullTextSearch\Model\IndexDocument; |
38
|
|
|
use OCA\FullTextSearch\Model\Runner; |
39
|
|
|
use OCA\FullTextSearch\Model\SearchRequest; |
40
|
|
|
use OCA\FullTextSearch\Model\SearchResult; |
41
|
|
|
use OCA\FullTextSearch_ElasticSearch\AppInfo\Application; |
42
|
|
|
use OCA\FullTextSearch_ElasticSearch\Exceptions\AccessIsEmptyException; |
43
|
|
|
use OCA\FullTextSearch_ElasticSearch\Exceptions\ConfigurationException; |
44
|
|
|
use OCA\FullTextSearch_ElasticSearch\Service\ConfigService; |
45
|
|
|
use OCA\FullTextSearch_ElasticSearch\Service\IndexService; |
46
|
|
|
use OCA\FullTextSearch_ElasticSearch\Service\MiscService; |
47
|
|
|
use OCA\FullTextSearch_ElasticSearch\Service\SearchService; |
48
|
|
|
use OCP\AppFramework\QueryException; |
49
|
|
|
|
50
|
|
|
|
51
|
|
|
class ElasticSearchPlatform implements IFullTextSearchPlatform { |
52
|
|
|
|
53
|
|
|
/** @var ConfigService */ |
54
|
|
|
private $configService; |
55
|
|
|
|
56
|
|
|
/** @var IndexService */ |
57
|
|
|
private $indexService; |
58
|
|
|
|
59
|
|
|
/** @var SearchService */ |
60
|
|
|
private $searchService; |
61
|
|
|
|
62
|
|
|
/** @var MiscService */ |
63
|
|
|
private $miscService; |
64
|
|
|
|
65
|
|
|
/** @var Client */ |
66
|
|
|
private $client; |
67
|
|
|
|
68
|
|
|
/** @var Runner */ |
69
|
|
|
private $runner; |
70
|
|
|
|
71
|
|
|
|
72
|
|
|
public function __construct( |
73
|
|
|
ConfigService $configService, IndexService $indexService, SearchService $searchService, |
74
|
|
|
MiscService $miscService |
75
|
|
|
) { |
76
|
|
|
$this->configService = $configService; |
77
|
|
|
$this->indexService = $indexService; |
78
|
|
|
$this->searchService = $searchService; |
79
|
|
|
$this->miscService = $miscService; |
80
|
|
|
} |
81
|
|
|
|
82
|
|
|
/** |
83
|
|
|
* return a unique Id of the platform. |
84
|
|
|
*/ |
85
|
|
|
public function getId() { |
86
|
|
|
return 'elastic_search'; |
87
|
|
|
} |
88
|
|
|
|
89
|
|
|
/** |
90
|
|
|
* return a unique Id of the platform. |
91
|
|
|
*/ |
92
|
|
|
public function getName() { |
93
|
|
|
return 'Elasticsearch'; |
94
|
|
|
} |
95
|
|
|
|
96
|
|
|
|
97
|
|
|
/** |
98
|
|
|
* @return string |
99
|
|
|
*/ |
100
|
|
|
public function getVersion() { |
101
|
|
|
return ''; |
102
|
|
|
} |
103
|
|
|
|
104
|
|
|
|
105
|
|
|
/** |
106
|
|
|
* @return array |
|
|
|
|
107
|
|
|
* @throws ConfigurationException |
108
|
|
|
*/ |
109
|
|
|
public function getConfiguration() { |
110
|
|
|
|
111
|
|
|
$result = []; |
112
|
|
|
$hosts = $this->configService->getElasticHost(); |
113
|
|
|
|
114
|
|
|
foreach ($hosts as $host) { |
115
|
|
|
$parsedHost = parse_url($host); |
116
|
|
|
$safeHost = $parsedHost['scheme'] . '://'; |
117
|
|
|
if (array_key_exists('user', $parsedHost)) { |
118
|
|
|
$safeHost .= $parsedHost['user'] . ':' . '********' . '@'; |
119
|
|
|
} |
120
|
|
|
$safeHost .= $parsedHost['host']; |
121
|
|
|
$safeHost .= ':' . $parsedHost['port']; |
122
|
|
|
|
123
|
|
|
$result[] = $safeHost; |
124
|
|
|
} |
125
|
|
|
|
126
|
|
|
return [ |
127
|
|
|
'elastic_host' => $result, |
128
|
|
|
'elastic_index' => $this->configService->getElasticIndex() |
129
|
|
|
]; |
130
|
|
|
} |
131
|
|
|
|
132
|
|
|
|
133
|
|
|
/** |
134
|
|
|
* @param Runner $runner |
135
|
|
|
*/ |
136
|
|
|
public function setRunner(Runner $runner) { |
137
|
|
|
$this->runner = $runner; |
138
|
|
|
} |
139
|
|
|
|
140
|
|
|
/** |
141
|
|
|
* @param $action |
142
|
|
|
* @param bool $force |
143
|
|
|
* |
144
|
|
|
* @throws Exception |
145
|
|
|
*/ |
146
|
|
|
private function updateRunnerAction($action, $force = false) { |
147
|
|
|
if ($this->runner === null) { |
148
|
|
|
return; |
149
|
|
|
} |
150
|
|
|
|
151
|
|
|
$this->runner->updateAction($action, $force); |
152
|
|
|
} |
153
|
|
|
|
154
|
|
|
|
155
|
|
|
/** |
156
|
|
|
* @param Index $index |
157
|
|
|
* @param string $message |
158
|
|
|
* @param string $exception |
159
|
|
|
* @param int $sev |
160
|
|
|
*/ |
161
|
|
|
private function updateNewIndexError($index, $message, $exception, $sev) { |
162
|
|
|
if ($this->runner === null) { |
163
|
|
|
return; |
164
|
|
|
} |
165
|
|
|
|
166
|
|
|
$this->runner->newIndexError($index, $message, $exception, $sev); |
167
|
|
|
} |
168
|
|
|
|
169
|
|
|
|
170
|
|
|
/** |
171
|
|
|
* @param Index $index |
172
|
|
|
* @param string $message |
173
|
|
|
* @param string $status |
174
|
|
|
* @param int $type |
175
|
|
|
*/ |
176
|
|
|
private function updateNewIndexResult($index, $message, $status, $type) { |
177
|
|
|
if ($this->runner === null) { |
178
|
|
|
return; |
179
|
|
|
} |
180
|
|
|
|
181
|
|
|
$this->runner->newIndexResult($index, $message, $status, $type); |
182
|
|
|
} |
183
|
|
|
|
184
|
|
|
|
185
|
|
|
/** |
186
|
|
|
* Called when loading the platform. |
187
|
|
|
* |
188
|
|
|
* Loading some container and connect to ElasticSearch. |
189
|
|
|
* |
190
|
|
|
* @throws ConfigurationException |
191
|
|
|
* @throws QueryException |
192
|
|
|
* @throws Exception |
193
|
|
|
*/ |
194
|
|
|
public function loadPlatform() { |
195
|
|
|
$app = new Application(); |
|
|
|
|
196
|
|
|
|
197
|
|
|
try { |
198
|
|
|
$this->connectToElastic($this->configService->getElasticHost()); |
199
|
|
|
} catch (ConfigurationException $e) { |
200
|
|
|
throw $e; |
201
|
|
|
} |
202
|
|
|
} |
203
|
|
|
|
204
|
|
|
|
205
|
|
|
/** |
206
|
|
|
* not used yet. |
207
|
|
|
* |
208
|
|
|
* @return bool |
209
|
|
|
*/ |
210
|
|
|
public function testPlatform() { |
211
|
|
|
return $this->client->ping(); |
212
|
|
|
} |
213
|
|
|
|
214
|
|
|
|
215
|
|
|
/** |
216
|
|
|
* called before any index |
217
|
|
|
* |
218
|
|
|
* We create a general index. |
219
|
|
|
* |
220
|
|
|
* @throws ConfigurationException |
221
|
|
|
* @throws BadRequest400Exception |
222
|
|
|
*/ |
223
|
|
|
public function initializeIndex() { |
224
|
|
|
$this->indexService->initializeIndex($this->client); |
225
|
|
|
} |
226
|
|
|
|
227
|
|
|
|
228
|
|
|
/** |
229
|
|
|
* resetIndex(); |
230
|
|
|
* |
231
|
|
|
* Called when admin wants to remove an index specific to a $provider. |
232
|
|
|
* $provider can be null, meaning a reset of the whole index. |
233
|
|
|
* |
234
|
|
|
* @param string $providerId |
235
|
|
|
* |
236
|
|
|
* @throws ConfigurationException |
237
|
|
|
*/ |
238
|
|
|
public function resetIndex($providerId) { |
239
|
|
|
if ($providerId === 'all') { |
240
|
|
|
$this->indexService->resetIndexAll($this->client); |
241
|
|
|
} else { |
242
|
|
|
$this->indexService->resetIndex($this->client, $providerId); |
243
|
|
|
} |
244
|
|
|
} |
245
|
|
|
|
246
|
|
|
|
247
|
|
|
/** |
248
|
|
|
* @deprecated |
249
|
|
|
* |
250
|
|
|
* @param IFullTextSearchProvider $provider |
251
|
|
|
* @param $documents |
252
|
|
|
*/ |
253
|
|
|
public function indexDocuments(IFullTextSearchProvider $provider, $documents) { |
254
|
|
|
|
255
|
|
|
} |
256
|
|
|
|
257
|
|
|
|
258
|
|
|
/** |
259
|
|
|
* @param IFullTextSearchProvider $provider |
260
|
|
|
* @param IndexDocument $document |
261
|
|
|
* |
262
|
|
|
* @return Index |
263
|
|
|
*/ |
264
|
|
|
public function indexDocument(IFullTextSearchProvider $provider, IndexDocument $document) { |
265
|
|
|
|
266
|
|
|
$document->initHash(); |
267
|
|
|
|
268
|
|
|
try { |
269
|
|
|
$result = $this->indexService->indexDocument($this->client, $provider, $document); |
270
|
|
|
|
271
|
|
|
$index = $this->indexService->parseIndexResult($document->getIndex(), $result); |
|
|
|
|
272
|
|
|
|
273
|
|
|
$this->updateNewIndexResult( |
274
|
|
|
$document->getIndex(), json_encode($result), 'ok', |
275
|
|
|
Runner::RESULT_TYPE_SUCCESS |
276
|
|
|
); |
277
|
|
|
|
278
|
|
|
return $index; |
279
|
|
|
} catch (Exception $e) { |
280
|
|
|
$this->updateNewIndexResult( |
281
|
|
|
$document->getIndex(), '', 'issue while indexing, testing with empty content', |
282
|
|
|
Runner::RESULT_TYPE_WARNING |
283
|
|
|
); |
284
|
|
|
|
285
|
|
|
$this->manageIndexErrorException($document, $e); |
286
|
|
|
} |
287
|
|
|
|
288
|
|
|
try { |
289
|
|
|
$result = $this->indexDocumentError($provider, $document, $e); |
290
|
|
|
$index = $this->indexService->parseIndexResult($document->getIndex(), $result); |
|
|
|
|
291
|
|
|
|
292
|
|
|
$this->updateNewIndexResult( |
293
|
|
|
$document->getIndex(), json_encode($result), 'ok', |
294
|
|
|
Runner::RESULT_TYPE_WARNING |
295
|
|
|
); |
296
|
|
|
|
297
|
|
|
return $index; |
298
|
|
|
} catch (Exception $e) { |
299
|
|
|
$this->updateNewIndexResult( |
300
|
|
|
$document->getIndex(), '', 'fail', |
301
|
|
|
Runner::RESULT_TYPE_FAIL |
302
|
|
|
); |
303
|
|
|
$this->manageIndexErrorException($document, $e); |
304
|
|
|
} |
305
|
|
|
|
306
|
|
|
return $document->getIndex(); |
307
|
|
|
} |
308
|
|
|
|
309
|
|
|
|
310
|
|
|
/** |
311
|
|
|
* @param IFullTextSearchProvider $provider |
312
|
|
|
* @param IndexDocument $document |
313
|
|
|
* @param Exception $e |
314
|
|
|
* |
315
|
|
|
* @return array |
|
|
|
|
316
|
|
|
* @throws AccessIsEmptyException |
317
|
|
|
* @throws ConfigurationException |
318
|
|
|
* @throws \Exception |
319
|
|
|
*/ |
320
|
|
|
private function indexDocumentError( |
321
|
|
|
IFullTextSearchProvider $provider, IndexDocument $document, Exception $e |
322
|
|
|
) { |
323
|
|
|
|
324
|
|
|
$this->updateRunnerAction('indexDocumentWithoutContent', true); |
325
|
|
|
|
326
|
|
|
$document->setContent(''); |
327
|
|
|
// $index = $document->getIndex(); |
328
|
|
|
// $index->unsetStatus(Index::INDEX_CONTENT); |
329
|
|
|
|
330
|
|
|
$result = $this->indexService->indexDocument($this->client, $provider, $document); |
331
|
|
|
|
332
|
|
|
return $result; |
333
|
|
|
} |
334
|
|
|
|
335
|
|
|
|
336
|
|
|
/** |
337
|
|
|
* @param IndexDocument $document |
338
|
|
|
* @param Exception $e |
339
|
|
|
*/ |
340
|
|
|
private function manageIndexErrorException(IndexDocument $document, Exception $e) { |
341
|
|
|
|
342
|
|
|
$message = $this->parseIndexErrorException($e); |
343
|
|
|
$document->getIndex() |
344
|
|
|
->addError($message, get_class($e), Index::ERROR_SEV_3); |
345
|
|
|
$this->updateNewIndexError( |
346
|
|
|
$document->getIndex(), $message, get_class($e), Index::ERROR_SEV_3 |
347
|
|
|
); |
348
|
|
|
} |
349
|
|
|
|
350
|
|
|
|
351
|
|
|
/** |
352
|
|
|
* @param Exception $e |
353
|
|
|
* |
354
|
|
|
* @return string |
355
|
|
|
*/ |
356
|
|
|
private function parseIndexErrorException(Exception $e) { |
357
|
|
|
|
358
|
|
|
$arr = json_decode($e->getMessage(), true); |
359
|
|
|
if (!is_array($arr)) { |
360
|
|
|
return $e->getMessage(); |
361
|
|
|
} |
362
|
|
|
|
363
|
|
|
if (array_key_exists('reason', $arr['error']['root_cause'][0])) { |
364
|
|
|
return $arr['error']['root_cause'][0]['reason']; |
365
|
|
|
} |
366
|
|
|
|
367
|
|
|
return $e->getMessage(); |
368
|
|
|
} |
369
|
|
|
|
370
|
|
|
|
371
|
|
|
/** |
372
|
|
|
* {@inheritdoc} |
373
|
|
|
* @throws ConfigurationException |
374
|
|
|
*/ |
375
|
|
|
public function deleteIndexes($indexes) { |
376
|
|
|
try { |
377
|
|
|
$this->indexService->deleteIndexes($this->client, $indexes); |
378
|
|
|
} catch (ConfigurationException $e) { |
379
|
|
|
throw $e; |
380
|
|
|
} |
381
|
|
|
} |
382
|
|
|
|
383
|
|
|
|
384
|
|
|
/** |
385
|
|
|
* {@inheritdoc} |
386
|
|
|
* @throws ConfigurationException |
387
|
|
|
* @throws Exception |
388
|
|
|
*/ |
389
|
|
|
public function searchDocuments( |
390
|
|
|
IFullTextSearchProvider $provider, DocumentAccess $access, SearchRequest $request |
391
|
|
|
) { |
392
|
|
|
return null; |
393
|
|
|
// return $this->searchService->searchDocuments($this->client, $provider, $access, $request); |
394
|
|
|
} |
395
|
|
|
|
396
|
|
|
|
397
|
|
|
/** |
398
|
|
|
* {@inheritdoc} |
399
|
|
|
* @throws Exception |
400
|
|
|
*/ |
401
|
|
|
public function searchRequest(SearchResult $result, DocumentAccess $access) { |
402
|
|
|
$this->searchService->searchRequest($this->client, $result, $access); |
403
|
|
|
} |
404
|
|
|
|
405
|
|
|
|
406
|
|
|
/** |
407
|
|
|
* @param string $providerId |
408
|
|
|
* @param string $documentId |
409
|
|
|
* |
410
|
|
|
* @return IndexDocument |
411
|
|
|
* @throws ConfigurationException |
412
|
|
|
*/ |
413
|
|
|
public function getDocument($providerId, $documentId) { |
414
|
|
|
return $this->searchService->getDocument($this->client, $providerId, $documentId); |
415
|
|
|
} |
416
|
|
|
|
417
|
|
|
|
418
|
|
|
/** |
419
|
|
|
* @param array $hosts |
420
|
|
|
* |
421
|
|
|
* @throws Exception |
422
|
|
|
*/ |
423
|
|
|
private function connectToElastic($hosts) { |
424
|
|
|
|
425
|
|
|
try { |
426
|
|
|
$hosts = array_map([MiscService::class, 'noEndSlash'], $hosts); |
427
|
|
|
$this->client = ClientBuilder::create() |
428
|
|
|
->setHosts($hosts) |
429
|
|
|
->setRetries(3) |
430
|
|
|
->build(); |
431
|
|
|
|
432
|
|
|
// } |
433
|
|
|
// catch (CouldNotConnectToHost $e) { |
434
|
|
|
// $this 'CouldNotConnectToHost'; |
435
|
|
|
// $previous = $e->getPrevious(); |
436
|
|
|
// if ($previous instanceof MaxRetriesException) { |
437
|
|
|
// echo "Max retries!"; |
438
|
|
|
// } |
439
|
|
|
} catch (Exception $e) { |
440
|
|
|
throw $e; |
441
|
|
|
// echo ' ElasticSearchPlatform::load() Exception --- ' . $e->getMessage() . "\n"; |
442
|
|
|
} |
443
|
|
|
} |
444
|
|
|
|
445
|
|
|
|
446
|
|
|
} |
447
|
|
|
|
This check looks for the generic type
array
as a return type and suggests a more specific type. This type is inferred from the actual code.