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