1
|
|
|
<?php |
2
|
|
|
/** |
3
|
|
|
* FullTextSearch - Full text search framework for 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\Service; |
28
|
|
|
|
29
|
|
|
use Exception; |
30
|
|
|
use OCA\FullTextSearch\Db\IndexesRequest; |
31
|
|
|
use OCA\FullTextSearch\Exceptions\DatabaseException; |
32
|
|
|
use OCA\FullTextSearch\Exceptions\IndexDoesNotExistException; |
33
|
|
|
use OCA\FullTextSearch\Exceptions\InterruptException; |
34
|
|
|
use OCA\FullTextSearch\Exceptions\NoResultException; |
35
|
|
|
use OCA\FullTextSearch\Exceptions\TickDoesNotExistException; |
36
|
|
|
use OCA\FullTextSearch\IFullTextSearchPlatform; |
37
|
|
|
use OCA\FullTextSearch\IFullTextSearchProvider; |
38
|
|
|
use OCA\FullTextSearch\Model\ExtendedIndex; |
39
|
|
|
use OCA\FullTextSearch\Model\Index; |
40
|
|
|
use OCA\FullTextSearch\Model\IndexDocument; |
41
|
|
|
use OCA\FullTextSearch\Model\ProviderIndexes; |
42
|
|
|
use OCA\FullTextSearch\Model\Runner; |
43
|
|
|
|
44
|
|
|
class IndexService { |
45
|
|
|
|
46
|
|
|
/** @var IndexesRequest */ |
47
|
|
|
private $indexesRequest; |
48
|
|
|
|
49
|
|
|
/** @var ConfigService */ |
50
|
|
|
private $configService; |
51
|
|
|
|
52
|
|
|
/** @var ProviderService */ |
53
|
|
|
private $providerService; |
54
|
|
|
|
55
|
|
|
/** @var PlatformService */ |
56
|
|
|
private $platformService; |
57
|
|
|
|
58
|
|
|
/** @var MiscService */ |
59
|
|
|
private $miscService; |
60
|
|
|
|
61
|
|
|
|
62
|
|
|
/** @var Runner */ |
63
|
|
|
private $runner = null; |
64
|
|
|
|
65
|
|
|
/** |
66
|
|
|
* IndexService constructor. |
67
|
|
|
* |
68
|
|
|
* @param IndexesRequest $indexesRequest |
69
|
|
|
* @param ConfigService $configService |
70
|
|
|
* @param ProviderService $providerService |
71
|
|
|
* @param PlatformService $platformService |
72
|
|
|
* @param MiscService $miscService |
73
|
|
|
*/ |
74
|
|
View Code Duplication |
public function __construct( |
|
|
|
|
75
|
|
|
IndexesRequest $indexesRequest, ConfigService $configService, ProviderService $providerService, |
76
|
|
|
PlatformService $platformService, MiscService $miscService |
77
|
|
|
) { |
78
|
|
|
$this->indexesRequest = $indexesRequest; |
79
|
|
|
$this->configService = $configService; |
80
|
|
|
$this->providerService = $providerService; |
81
|
|
|
$this->platformService = $platformService; |
82
|
|
|
$this->miscService = $miscService; |
83
|
|
|
} |
84
|
|
|
|
85
|
|
|
|
86
|
|
|
/** |
87
|
|
|
* @param Runner $runner |
88
|
|
|
*/ |
89
|
|
|
public function setRunner(Runner $runner) { |
90
|
|
|
$this->runner = $runner; |
91
|
|
|
} |
92
|
|
|
|
93
|
|
|
|
94
|
|
|
/** |
95
|
|
|
* @param $action |
96
|
|
|
* |
97
|
|
|
* @throws InterruptException |
98
|
|
|
* @throws TickDoesNotExistException |
99
|
|
|
*/ |
100
|
|
|
private function updateRunner($action) { |
101
|
|
|
if ($this->runner === null) { |
102
|
|
|
return; |
103
|
|
|
} |
104
|
|
|
|
105
|
|
|
$this->runner->update($action); |
106
|
|
|
} |
107
|
|
|
|
108
|
|
|
|
109
|
|
|
/** |
110
|
|
|
* @param IFullTextSearchPlatform $platform |
111
|
|
|
* @param IFullTextSearchProvider $provider |
112
|
|
|
* @param string $userId |
113
|
|
|
* |
114
|
|
|
* @throws Exception |
115
|
|
|
*/ |
116
|
|
|
public function indexProviderContentFromUser( |
117
|
|
|
IFullTextSearchPlatform $platform, IFullTextSearchProvider $provider, $userId |
118
|
|
|
) { |
119
|
|
|
$this->updateRunner('generateIndex' . $provider->getName()); |
120
|
|
|
$documents = $provider->generateIndexableDocuments($userId); |
121
|
|
|
|
122
|
|
|
//$maxSize = sizeof($documents); |
|
|
|
|
123
|
|
|
|
124
|
|
|
$toIndex = $this->updateDocumentsWithCurrIndex($provider, $documents); |
125
|
|
|
$this->indexChunks($platform, $provider, $toIndex); |
126
|
|
|
} |
127
|
|
|
|
128
|
|
|
|
129
|
|
|
/** |
130
|
|
|
* @param IFullTextSearchProvider $provider |
131
|
|
|
* @param IndexDocument[] $documents |
132
|
|
|
* |
133
|
|
|
* @return IndexDocument[] |
134
|
|
|
* @throws InterruptException |
135
|
|
|
* @throws TickDoesNotExistException |
136
|
|
|
*/ |
137
|
|
|
private function updateDocumentsWithCurrIndex(IFullTextSearchProvider $provider, array $documents) { |
|
|
|
|
138
|
|
|
|
139
|
|
|
$currIndex = $this->getProviderIndexFromProvider($provider); |
140
|
|
|
$result = []; |
141
|
|
|
foreach ($documents as $document) { |
142
|
|
|
$this->updateRunner('compareWithCurrentIndex'); |
143
|
|
|
|
144
|
|
|
$index = $currIndex->getIndex($document->getId()); |
145
|
|
|
if ($index === null) { |
146
|
|
|
$index = new Index($document->getProviderId(), $document->getId()); |
147
|
|
|
$index->setStatus(Index::INDEX_FULL); |
148
|
|
|
$index->setLastIndex(); |
149
|
|
|
} |
150
|
|
|
|
151
|
|
|
$document->setIndex($index); |
152
|
|
|
if (!$this->isDocumentUpToDate($provider, $document)) { |
153
|
|
|
$result[] = $document; |
154
|
|
|
} |
155
|
|
|
|
156
|
|
|
} |
157
|
|
|
|
158
|
|
|
return $result; |
159
|
|
|
} |
160
|
|
|
|
161
|
|
|
|
162
|
|
|
/** |
163
|
|
|
* @param IFullTextSearchProvider $provider |
164
|
|
|
* @param IndexDocument $document |
165
|
|
|
* |
166
|
|
|
* @return bool |
167
|
|
|
*/ |
168
|
|
|
private function isDocumentUpToDate(IFullTextSearchProvider $provider, IndexDocument $document) { |
169
|
|
|
$index = $document->getIndex(); |
170
|
|
|
|
171
|
|
|
if (!$index->isStatus(Index::INDEX_OK)) { |
172
|
|
|
return false; |
173
|
|
|
} |
174
|
|
|
|
175
|
|
|
if ($index->isStatus(Index::INDEX_META) || $index->isStatus(Index::INDEX_CONTENT)) { |
176
|
|
|
return false; |
177
|
|
|
} |
178
|
|
|
|
179
|
|
|
return $provider->isDocumentUpToDate($document); |
180
|
|
|
} |
181
|
|
|
|
182
|
|
|
|
183
|
|
|
/** |
184
|
|
|
* @param IFullTextSearchProvider $provider |
185
|
|
|
* |
186
|
|
|
* @return ProviderIndexes |
187
|
|
|
*/ |
188
|
|
|
private function getProviderIndexFromProvider(IFullTextSearchProvider $provider) { |
189
|
|
|
$indexes = $this->indexesRequest->getIndexesFromProvider($provider); |
190
|
|
|
|
191
|
|
|
return new ProviderIndexes($indexes); |
192
|
|
|
} |
193
|
|
|
|
194
|
|
|
|
195
|
|
|
/** |
196
|
|
|
* @param IFullTextSearchPlatform $platform |
197
|
|
|
* @param IFullTextSearchProvider $provider |
198
|
|
|
* @param IndexDocument[] $documents |
199
|
|
|
* |
200
|
|
|
* @throws Exception |
201
|
|
|
*/ |
202
|
|
|
private function indexChunks( |
203
|
|
|
IFullTextSearchPlatform $platform, IFullTextSearchProvider $provider, $documents |
204
|
|
|
) { |
205
|
|
|
$chunkSize = $this->configService->getAppValue(ConfigService::CHUNK_INDEX); |
206
|
|
|
|
207
|
|
|
$max = sizeof($documents); |
208
|
|
|
for ($i = 0; $i < $max; $i++) { |
209
|
|
|
|
210
|
|
|
$this->updateRunner('indexChunk'); |
211
|
|
|
try { |
212
|
|
|
$chunk = array_splice($documents, 0, $chunkSize); |
213
|
|
|
$this->indexChunk($platform, $provider, $chunk); |
214
|
|
|
|
215
|
|
|
/** @var IndexDocument $doc */ |
216
|
|
|
foreach ($chunk as $doc) { |
217
|
|
|
$doc->__destruct(); // because. |
218
|
|
|
} |
219
|
|
|
} catch (NoResultException $e) { |
220
|
|
|
return; |
221
|
|
|
} catch (Exception $e) { |
222
|
|
|
throw $e; |
223
|
|
|
} |
224
|
|
|
} |
225
|
|
|
} |
226
|
|
|
|
227
|
|
|
|
228
|
|
|
/** |
229
|
|
|
* @param IFullTextSearchPlatform $platform |
230
|
|
|
* @param IFullTextSearchProvider $provider |
231
|
|
|
* @param IndexDocument[] $chunk |
232
|
|
|
* |
233
|
|
|
* @throws NoResultException |
234
|
|
|
* @throws DatabaseException |
235
|
|
|
*/ |
236
|
|
|
private function indexChunk( |
237
|
|
|
IFullTextSearchPlatform $platform, IFullTextSearchProvider $provider, $chunk |
238
|
|
|
) { |
239
|
|
|
if (sizeof($chunk) === 0) { |
240
|
|
|
throw new NoResultException(); |
241
|
|
|
} |
242
|
|
|
|
243
|
|
|
$documents = $provider->fillIndexDocuments($chunk); |
244
|
|
|
$toIndex = $this->filterDocumentsToIndex($documents); |
245
|
|
|
$indexes = $platform->indexDocuments($provider, $toIndex); |
246
|
|
|
$this->updateIndexes($indexes); |
247
|
|
|
} |
248
|
|
|
|
249
|
|
|
|
250
|
|
|
/** |
251
|
|
|
* @param IndexDocument[] $documents |
252
|
|
|
* |
253
|
|
|
* @return array |
254
|
|
|
*/ |
255
|
|
|
private function filterDocumentsToIndex($documents) { |
256
|
|
|
$toIndex = []; |
257
|
|
|
foreach ($documents as $document) { |
258
|
|
|
// TODO - rework the index/not_index |
259
|
|
|
// echo '==st==' . $document->getIndex()->getStatus() . "\n"; |
260
|
|
|
if (!$document->getIndex() |
261
|
|
|
->isStatus(Index::INDEX_IGNORE)) { |
262
|
|
|
$toIndex[] = $document; |
263
|
|
|
} |
264
|
|
|
} |
265
|
|
|
|
266
|
|
|
return $toIndex; |
267
|
|
|
} |
268
|
|
|
|
269
|
|
|
|
270
|
|
|
/** |
271
|
|
|
* @param IFullTextSearchPlatform $platform |
272
|
|
|
* @param IFullTextSearchProvider $provider |
273
|
|
|
* @param Index $index |
274
|
|
|
* |
275
|
|
|
* @internal param int|string $documentId |
276
|
|
|
* @throws Exception |
277
|
|
|
*/ |
278
|
|
|
public function updateDocument( |
279
|
|
|
IFullTextSearchPlatform $platform, IFullTextSearchProvider $provider, Index $index |
280
|
|
|
) { |
281
|
|
|
$document = $provider->updateDocument($index); |
282
|
|
|
if ($document === null) { |
283
|
|
|
$this->indexesRequest->deleteIndex($index); |
284
|
|
|
|
285
|
|
|
return; |
286
|
|
|
} |
287
|
|
|
|
288
|
|
|
$index = $platform->indexDocument($provider, $document); |
289
|
|
|
$this->updateIndex($index); |
290
|
|
|
} |
291
|
|
|
|
292
|
|
|
|
293
|
|
|
/** |
294
|
|
|
* @param Index[] $indexes |
295
|
|
|
* |
296
|
|
|
* @throws DatabaseException |
297
|
|
|
*/ |
298
|
|
|
public function updateIndexes($indexes) { |
299
|
|
|
|
300
|
|
|
try { |
301
|
|
|
foreach ($indexes as $index) { |
302
|
|
|
$this->updateIndex($index); |
303
|
|
|
} |
304
|
|
|
} catch (Exception $e) { |
305
|
|
|
throw new DatabaseException($e->getMessage()); |
306
|
|
|
} |
307
|
|
|
} |
308
|
|
|
|
309
|
|
|
|
310
|
|
|
/** |
311
|
|
|
* @param Index $index |
312
|
|
|
* |
313
|
|
|
* @throws Exception |
314
|
|
|
*/ |
315
|
|
|
private function updateIndex(Index $index) { |
316
|
|
|
|
317
|
|
|
if ($index->isStatus(Index::INDEX_REMOVE)) { |
318
|
|
|
|
319
|
|
|
if ($index->isStatus(Index::INDEX_OK)) { |
320
|
|
|
$this->indexesRequest->deleteIndex($index); |
321
|
|
|
|
322
|
|
|
return; |
323
|
|
|
} |
324
|
|
|
$index->setStatus(Index::INDEX_FAILED); |
325
|
|
|
$this->indexesRequest->update($index); |
326
|
|
|
|
327
|
|
|
return; |
328
|
|
|
} |
329
|
|
|
|
330
|
|
|
if ($index->isStatus(Index::INDEX_DONE)) { |
331
|
|
|
$index->setStatus(Index::INDEX_OK, true); |
332
|
|
|
} |
333
|
|
|
|
334
|
|
|
if (!$this->indexesRequest->update($index)) { |
335
|
|
|
$this->indexesRequest->create($index); |
336
|
|
|
} |
337
|
|
|
} |
338
|
|
|
|
339
|
|
|
|
340
|
|
|
/** |
341
|
|
|
* @param string $providerId |
342
|
|
|
* @param string|int $documentId |
343
|
|
|
* @param int $status |
344
|
|
|
* @param bool $reset |
345
|
|
|
* |
346
|
|
|
* @throws DatabaseException |
347
|
|
|
*/ |
348
|
|
|
public function updateIndexStatus($providerId, $documentId, $status, $reset = false) { |
|
|
|
|
349
|
|
|
try { |
350
|
|
|
$curr = $this->getIndex($providerId, $documentId); |
351
|
|
|
} catch (IndexDoesNotExistException $e) { |
352
|
|
|
$curr = new Index($providerId, $documentId); |
353
|
|
|
$curr->setStatus(Index::INDEX_FULL); |
354
|
|
|
} |
355
|
|
|
|
356
|
|
|
$curr->setStatus($status); |
357
|
|
|
$this->updateIndexes([$curr]); |
358
|
|
|
} |
359
|
|
|
|
360
|
|
|
|
361
|
|
|
/** |
362
|
|
|
* @param string $providerId |
363
|
|
|
* @param string|int $documentId |
364
|
|
|
* |
365
|
|
|
* @return ExtendedIndex |
366
|
|
|
* @throws IndexDoesNotExistException |
367
|
|
|
*/ |
368
|
|
|
public function getIndex($providerId, $documentId) { |
369
|
|
|
return $this->indexesRequest->getIndex($providerId, $documentId); |
370
|
|
|
} |
371
|
|
|
|
372
|
|
|
|
373
|
|
|
/** |
374
|
|
|
* @return Index[] |
375
|
|
|
*/ |
376
|
|
|
public function getQueuedIndexes() { |
377
|
|
|
return $this->indexesRequest->getQueuedIndexes(); |
378
|
|
|
} |
379
|
|
|
|
380
|
|
|
|
381
|
|
|
/** |
382
|
|
|
* @param string $providerId |
383
|
|
|
* |
384
|
|
|
* @throws Exception |
385
|
|
|
*/ |
386
|
|
|
public function resetIndex($providerId = '') { |
387
|
|
|
$platform = $this->platformService->getPlatform(); |
388
|
|
|
|
389
|
|
|
if ($providerId === '') { |
390
|
|
|
$platform->removeIndex(null); |
391
|
|
|
$this->providerService->setProvidersAsNotIndexed(); |
392
|
|
|
$this->indexesRequest->reset(); |
393
|
|
|
|
394
|
|
|
return; |
395
|
|
|
} else { |
396
|
|
|
$providers = [$this->providerService->getProvider($providerId)]; |
397
|
|
|
} |
398
|
|
|
|
399
|
|
|
foreach ($providers AS $provider) { |
400
|
|
|
$platform->removeIndex($provider); |
401
|
|
|
$this->providerService->setProviderAsIndexed($provider, false); |
402
|
|
|
$this->indexesRequest->deleteFromProviderId($provider->getId()); |
403
|
|
|
} |
404
|
|
|
} |
405
|
|
|
|
406
|
|
|
|
407
|
|
|
} |
Duplicated code is one of the most pungent code smells. If you need to duplicate the same code in three or more different places, we strongly encourage you to look into extracting the code into a single class or operation.
You can also find more detailed suggestions in the “Code” section of your repository.