|
1
|
|
|
<?php |
|
2
|
|
|
/** |
|
3
|
|
|
* Bookmarks_FullTextSearch - Indexing bookmarks |
|
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\Bookmarks_FullTextSearch\Service; |
|
28
|
|
|
|
|
29
|
|
|
|
|
30
|
|
|
use OCA\Bookmarks\Controller\Lib\Bookmarks; |
|
31
|
|
|
use OCA\Bookmarks_FullTextSearch\Exceptions\WebpageIsNotIndexableException; |
|
32
|
|
|
use OCA\Bookmarks_FullTextSearch\Model\BookmarksDocument; |
|
33
|
|
|
use OCA\FullTextSearch\Model\DocumentAccess; |
|
34
|
|
|
use OCA\FullTextSearch\Model\Index; |
|
35
|
|
|
use OCA\FullTextSearch\Model\IndexDocument; |
|
36
|
|
|
use OCA\FullTextSearch\Model\Runner; |
|
37
|
|
|
use OCP\AppFramework\IAppContainer; |
|
38
|
|
|
use OCP\AppFramework\QueryException; |
|
39
|
|
|
|
|
40
|
|
|
class BookmarksService { |
|
41
|
|
|
|
|
42
|
|
|
const DOCUMENT_TYPE = 'bookmarks'; |
|
43
|
|
|
|
|
44
|
|
|
|
|
45
|
|
|
/** @var ConfigService */ |
|
46
|
|
|
private $configService; |
|
47
|
|
|
|
|
48
|
|
|
/** @var MiscService */ |
|
49
|
|
|
private $miscService; |
|
50
|
|
|
|
|
51
|
|
|
/** @var Bookmarks */ |
|
52
|
|
|
private $bookmarksClass; |
|
53
|
|
|
|
|
54
|
|
|
|
|
55
|
|
|
/** |
|
56
|
|
|
* BookmarksService constructor. |
|
57
|
|
|
* |
|
58
|
|
|
* @param IAppContainer $container |
|
59
|
|
|
* @param ConfigService $configService |
|
60
|
|
|
* @param MiscService $miscService |
|
61
|
|
|
*/ |
|
62
|
|
View Code Duplication |
public function __construct( |
|
|
|
|
|
|
63
|
|
|
IAppContainer $container, ConfigService $configService, MiscService $miscService |
|
64
|
|
|
) { |
|
65
|
|
|
$this->configService = $configService; |
|
66
|
|
|
$this->miscService = $miscService; |
|
67
|
|
|
|
|
68
|
|
|
try { |
|
69
|
|
|
$this->bookmarksClass = $container->query(Bookmarks::class); |
|
70
|
|
|
} catch (QueryException $e) { |
|
71
|
|
|
/** we do nothing */ |
|
72
|
|
|
} |
|
73
|
|
|
} |
|
74
|
|
|
|
|
75
|
|
|
|
|
76
|
|
|
/** |
|
77
|
|
|
* @param Runner $runner |
|
78
|
|
|
* @param string $userId |
|
79
|
|
|
* |
|
80
|
|
|
* @return BookmarksDocument[] |
|
81
|
|
|
*/ |
|
82
|
|
|
public function getBookmarksFromUser(Runner $runner, $userId) { |
|
83
|
|
|
|
|
84
|
|
|
$bookmarks = $this->bookmarksClass->findBookmarks($userId, 0, 'id', [], false, -1); |
|
85
|
|
|
|
|
86
|
|
|
$documents = []; |
|
87
|
|
|
foreach ($bookmarks as $bookmark) { |
|
88
|
|
|
$document = $this->generateBookmarksDocumentFromBookmark($bookmark, $userId); |
|
89
|
|
|
|
|
90
|
|
|
$documents[] = $document; |
|
91
|
|
|
} |
|
92
|
|
|
|
|
93
|
|
|
return $documents; |
|
94
|
|
|
} |
|
95
|
|
|
|
|
96
|
|
|
|
|
97
|
|
|
/** |
|
98
|
|
|
* @param BookmarksDocument $document |
|
99
|
|
|
* |
|
100
|
|
|
* @throws WebpageIsNotIndexableException |
|
101
|
|
|
*/ |
|
102
|
|
|
public function updateDocumentFromBookmarksDocument(BookmarksDocument $document) { |
|
103
|
|
|
$userId = $document->getAccess() |
|
104
|
|
|
->getOwnerId(); |
|
105
|
|
|
|
|
106
|
|
|
$bookmark = $this->bookmarksClass->findUniqueBookmark($document->getId(), $userId); |
|
107
|
|
|
$document->addPart('description', $bookmark['description']); |
|
108
|
|
|
|
|
109
|
|
|
$html = $this->getWebpageFromUrl($document->getSource()); |
|
110
|
|
|
$document->setContent(base64_encode($html), IndexDocument::ENCODED_BASE64); |
|
111
|
|
|
} |
|
112
|
|
|
|
|
113
|
|
|
|
|
114
|
|
|
/** |
|
115
|
|
|
* @param $url |
|
116
|
|
|
* |
|
117
|
|
|
* @return mixed |
|
118
|
|
|
* @throws WebpageIsNotIndexableException |
|
119
|
|
|
*/ |
|
120
|
|
|
private function getWebpageFromUrl($url) { |
|
121
|
|
|
$curl = curl_init($url); |
|
122
|
|
|
curl_setopt($curl, CURLOPT_CONNECTTIMEOUT, 10); |
|
123
|
|
|
curl_setopt($curl, CURLOPT_TIMEOUT, 10); |
|
124
|
|
|
|
|
125
|
|
|
curl_setopt($curl, CURLOPT_RETURNTRANSFER, true); |
|
126
|
|
|
|
|
127
|
|
|
curl_setopt($curl, CURLOPT_SSL_VERIFYPEER, false); |
|
128
|
|
|
curl_setopt($curl, CURLOPT_SSL_VERIFYHOST, false); |
|
129
|
|
|
|
|
130
|
|
|
$html = curl_exec($curl); |
|
131
|
|
|
if (curl_error($curl)) { |
|
132
|
|
|
throw new WebpageIsNotIndexableException('Webpage is not reachable - ' . $url); |
|
133
|
|
|
} |
|
134
|
|
|
|
|
135
|
|
|
curl_close($curl); |
|
136
|
|
|
|
|
137
|
|
|
return $html; |
|
138
|
|
|
} |
|
139
|
|
|
|
|
140
|
|
|
|
|
141
|
|
|
/** |
|
142
|
|
|
* @param IndexDocument $document |
|
143
|
|
|
* |
|
144
|
|
|
* @return bool |
|
145
|
|
|
*/ |
|
146
|
|
|
public function isDocumentUpToDate($document) { |
|
147
|
|
|
$index = $document->getIndex(); |
|
148
|
|
|
|
|
149
|
|
|
if ($index->getStatus() !== Index::INDEX_OK) { |
|
150
|
|
|
return false; |
|
151
|
|
|
} |
|
152
|
|
|
|
|
153
|
|
|
$s = $this->configService->getAppValue(ConfigService::BOOKMARKS_TTL) * 3600 * 24; |
|
154
|
|
|
|
|
155
|
|
|
return ($index->getLastIndex() > time() - $s); |
|
156
|
|
|
} |
|
157
|
|
|
|
|
158
|
|
|
|
|
159
|
|
|
/** |
|
160
|
|
|
* @param $bookmark |
|
161
|
|
|
* @param $userId |
|
162
|
|
|
* |
|
163
|
|
|
* @return BookmarksDocument |
|
164
|
|
|
*/ |
|
165
|
|
|
private function generateBookmarksDocumentFromBookmark($bookmark, $userId) { |
|
166
|
|
|
$document = new BookmarksDocument($bookmark['id']); |
|
167
|
|
|
|
|
168
|
|
|
$document->setAccess(new DocumentAccess($userId)) |
|
169
|
|
|
->setModifiedTime($bookmark['lastmodified']) |
|
170
|
|
|
->setSource($bookmark['url']) |
|
171
|
|
|
->setTitle($bookmark['title']) |
|
172
|
|
|
->setTags($bookmark['tags']); |
|
173
|
|
|
|
|
174
|
|
|
return $document; |
|
175
|
|
|
} |
|
176
|
|
|
|
|
177
|
|
|
|
|
178
|
|
|
/** |
|
179
|
|
|
* @param Index $index |
|
180
|
|
|
* |
|
181
|
|
|
* @return null|BookmarksDocument |
|
182
|
|
|
*/ |
|
183
|
|
|
public function updateDocument(Index $index) { |
|
184
|
|
|
try { |
|
185
|
|
|
$document = $this->generateDocumentFromIndex($index); |
|
186
|
|
|
|
|
187
|
|
|
return $document; |
|
188
|
|
|
} catch (WebpageIsNotIndexableException $e) { |
|
189
|
|
|
return null; |
|
190
|
|
|
} |
|
191
|
|
|
} |
|
192
|
|
|
|
|
193
|
|
|
|
|
194
|
|
|
/** |
|
195
|
|
|
* @param Index $index |
|
196
|
|
|
* |
|
197
|
|
|
* @return BookmarksDocument |
|
198
|
|
|
* @throws WebpageIsNotIndexableException |
|
199
|
|
|
*/ |
|
200
|
|
|
private function generateDocumentFromIndex(Index $index) { |
|
201
|
|
|
|
|
202
|
|
|
$bookmark = |
|
203
|
|
|
$this->bookmarksClass->findUniqueBookmark( |
|
204
|
|
|
$index->getDocumentId(), $index->getOwnerId() |
|
205
|
|
|
); |
|
206
|
|
|
|
|
207
|
|
|
if (sizeof($bookmark) === 0) { |
|
208
|
|
|
$index->setStatus(Index::INDEX_REMOVE); |
|
209
|
|
|
$document = new BookmarksDocument($index->getDocumentId()); |
|
210
|
|
|
$document->setIndex($index); |
|
211
|
|
|
|
|
212
|
|
|
return $document; |
|
213
|
|
|
} |
|
214
|
|
|
|
|
215
|
|
|
$document = $this->generateBookmarksDocumentFromBookmark($bookmark, $index->getOwnerId()); |
|
216
|
|
|
$document->setIndex($index); |
|
217
|
|
|
|
|
218
|
|
|
$this->updateDocumentFromBookmarksDocument($document); |
|
219
|
|
|
|
|
220
|
|
|
return $document; |
|
221
|
|
|
} |
|
222
|
|
|
|
|
223
|
|
|
|
|
224
|
|
|
} |
nextcloud /
bookmarks_fulltextsearch
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.