|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
/** |
|
4
|
|
|
* (c) Kitodo. Key to digital objects e.V. <[email protected]> |
|
5
|
|
|
* |
|
6
|
|
|
* This file is part of the Kitodo and TYPO3 projects. |
|
7
|
|
|
* |
|
8
|
|
|
* @license GNU General Public License version 3 or later. |
|
9
|
|
|
* For the full copyright and license information, please read the |
|
10
|
|
|
* LICENSE.txt file that was distributed with this source code. |
|
11
|
|
|
*/ |
|
12
|
|
|
|
|
13
|
|
|
namespace Kitodo\Dlf\Middleware; |
|
14
|
|
|
|
|
15
|
|
|
use Kitodo\Dlf\Common\Helper; |
|
16
|
|
|
use Kitodo\Dlf\Common\Solr\Solr; |
|
17
|
|
|
use Kitodo\Dlf\Common\Solr\SearchResult\ResultDocument; |
|
18
|
|
|
use Psr\Http\Message\ResponseInterface; |
|
19
|
|
|
use Psr\Http\Message\ServerRequestInterface; |
|
20
|
|
|
use Psr\Http\Server\MiddlewareInterface; |
|
21
|
|
|
use Psr\Http\Server\RequestHandlerInterface; |
|
22
|
|
|
use TYPO3\CMS\Core\Http\Response; |
|
23
|
|
|
use TYPO3\CMS\Core\Site\SiteFinder; |
|
24
|
|
|
use TYPO3\CMS\Core\Utility\GeneralUtility; |
|
25
|
|
|
|
|
26
|
|
|
/** |
|
27
|
|
|
* Search in document Middleware for plugin 'Search' of the 'dlf' extension |
|
28
|
|
|
* |
|
29
|
|
|
* @author Alexander Bigga <[email protected]> |
|
30
|
|
|
* @author Beatrycze Volk <[email protected]> |
|
31
|
|
|
* @package TYPO3 |
|
32
|
|
|
* @subpackage dlf |
|
33
|
|
|
* @access public |
|
34
|
|
|
*/ |
|
35
|
|
|
class SearchInDocument implements MiddlewareInterface |
|
36
|
|
|
{ |
|
37
|
|
|
/** |
|
38
|
|
|
* This holds the solr instance |
|
39
|
|
|
* |
|
40
|
|
|
* @var \Kitodo\Dlf\Common\Solr\Solr |
|
41
|
|
|
* @access private |
|
42
|
|
|
*/ |
|
43
|
|
|
private $solr; |
|
44
|
|
|
|
|
45
|
|
|
/** |
|
46
|
|
|
* This holds the solr fields |
|
47
|
|
|
* |
|
48
|
|
|
* @var array |
|
49
|
|
|
* @access private |
|
50
|
|
|
*/ |
|
51
|
|
|
private $fields; |
|
52
|
|
|
|
|
53
|
|
|
/** |
|
54
|
|
|
* The process method of the middleware. |
|
55
|
|
|
* |
|
56
|
|
|
* @access public |
|
57
|
|
|
* |
|
58
|
|
|
* @param ServerRequestInterface $request |
|
59
|
|
|
* @param RequestHandlerInterface $handler |
|
60
|
|
|
* |
|
61
|
|
|
* @return ResponseInterface JSON response of documents |
|
62
|
|
|
*/ |
|
63
|
|
|
public function process(ServerRequestInterface $request, RequestHandlerInterface $handler): ResponseInterface |
|
64
|
|
|
{ |
|
65
|
|
|
$response = $handler->handle($request); |
|
66
|
|
|
// Get input parameters and decrypt core name. |
|
67
|
|
|
$parameters = $request->getParsedBody(); |
|
68
|
|
|
// Return if not this middleware |
|
69
|
|
|
if (!isset($parameters['middleware']) || ($parameters['middleware'] != 'dlf/search-in-document')) { |
|
70
|
|
|
return $response; |
|
71
|
|
|
} |
|
72
|
|
|
|
|
73
|
|
|
$encrypted = (string) $parameters['encrypted']; |
|
74
|
|
|
if (empty($encrypted)) { |
|
75
|
|
|
throw new \InvalidArgumentException('No valid parameter passed: ' . $parameters['middleware'] . ' ' . $parameters['encrypted'] . '!', 1580585079); |
|
76
|
|
|
} |
|
77
|
|
|
|
|
78
|
|
|
$output = [ |
|
79
|
|
|
'documents' => [], |
|
80
|
|
|
'numFound' => 0 |
|
81
|
|
|
]; |
|
82
|
|
|
|
|
83
|
|
|
$core = Helper::decrypt($encrypted); |
|
84
|
|
|
|
|
85
|
|
|
// Perform Solr query. |
|
86
|
|
|
$this->solr = Solr::getInstance($core); |
|
87
|
|
|
$this->fields = Solr::getFields(); |
|
88
|
|
|
|
|
89
|
|
|
if ($this->solr->ready) { |
|
90
|
|
|
$result = $this->executeSolrQuery($parameters); |
|
|
|
|
|
|
91
|
|
|
/** @scrutinizer ignore-call */ |
|
92
|
|
|
$output['numFound'] = $result->getNumFound(); |
|
93
|
|
|
$data = $result->getData(); |
|
94
|
|
|
$highlighting = $data['ocrHighlighting']; |
|
95
|
|
|
|
|
96
|
|
|
$siteFinder = GeneralUtility::makeInstance(SiteFinder::class); |
|
97
|
|
|
$site = $siteFinder->getSiteByPageId($parameters['pid']); |
|
98
|
|
|
|
|
99
|
|
|
foreach ($result as $record) { |
|
100
|
|
|
$resultDocument = new ResultDocument($record, $highlighting, $this->fields); |
|
101
|
|
|
|
|
102
|
|
|
$url = (string) $site->getRouter()->generateUri( |
|
103
|
|
|
$parameters['pid'], |
|
104
|
|
|
[ |
|
105
|
|
|
'tx_dlf[id]' => !empty($resultDocument->getUid()) ? $resultDocument->getUid() : $parameters['uid'], |
|
106
|
|
|
'tx_dlf[page]' => $resultDocument->getPage(), |
|
107
|
|
|
'tx_dlf[highlight_word]' => $parameters['q'] |
|
108
|
|
|
] |
|
109
|
|
|
); |
|
110
|
|
|
|
|
111
|
|
|
$document = [ |
|
112
|
|
|
'id' => $resultDocument->getId(), |
|
113
|
|
|
'uid' => !empty($resultDocument->getUid()) ? $resultDocument->getUid() : $parameters['uid'], |
|
114
|
|
|
'page' => $resultDocument->getPage(), |
|
115
|
|
|
'snippet' => $resultDocument->getSnippets(), |
|
116
|
|
|
'highlight' => $resultDocument->getHighlightsIds(), |
|
117
|
|
|
'url' => $url |
|
118
|
|
|
]; |
|
119
|
|
|
$output['documents'][] = $document; |
|
120
|
|
|
} |
|
121
|
|
|
} |
|
122
|
|
|
|
|
123
|
|
|
// Create response object. |
|
124
|
|
|
/** @var Response $response */ |
|
125
|
|
|
$response = GeneralUtility::makeInstance(Response::class); |
|
126
|
|
|
$response->getBody()->write(json_encode($output)); |
|
127
|
|
|
return $response; |
|
128
|
|
|
} |
|
129
|
|
|
|
|
130
|
|
|
/** |
|
131
|
|
|
* Execute SOLR query. |
|
132
|
|
|
* |
|
133
|
|
|
* @access private |
|
134
|
|
|
* |
|
135
|
|
|
* @param array $parameters array of query parameters |
|
136
|
|
|
* |
|
137
|
|
|
* @return \Solarium\Core\Query\Result\ResultInterface result |
|
138
|
|
|
*/ |
|
139
|
|
|
private function executeSolrQuery($parameters) |
|
140
|
|
|
{ |
|
141
|
|
|
$query = $this->solr->service->createSelect(); |
|
142
|
|
|
$query->setFields([$this->fields['id'], $this->fields['uid'], $this->fields['page']]); |
|
143
|
|
|
$query->setQuery($this->getQuery($parameters)); |
|
144
|
|
|
$query->setStart(intval($parameters['start']))->setRows(20); |
|
145
|
|
|
$query->addSort($this->fields['page'], $query::SORT_ASC); |
|
146
|
|
|
$query->getHighlighting(); |
|
147
|
|
|
$solrRequest = $this->solr->service->createRequest($query); |
|
148
|
|
|
|
|
149
|
|
|
// it is necessary to add the custom parameters to the request |
|
150
|
|
|
// because query object doesn't allow custom parameters |
|
151
|
|
|
|
|
152
|
|
|
// field for which highlighting is going to be performed, |
|
153
|
|
|
// is required if you want to have OCR highlighting |
|
154
|
|
|
$solrRequest->addParam('hl.ocr.fl', $this->fields['fulltext']); |
|
155
|
|
|
// return the coordinates of highlighted search as absolute coordinates |
|
156
|
|
|
$solrRequest->addParam('hl.ocr.absoluteHighlights', 'on'); |
|
157
|
|
|
// max amount of snippets for a single page |
|
158
|
|
|
$solrRequest->addParam('hl.snippets', 40); |
|
159
|
|
|
// we store the fulltext on page level and can disable this option |
|
160
|
|
|
$solrRequest->addParam('hl.ocr.trackPages', 'off'); |
|
161
|
|
|
|
|
162
|
|
|
$response = $this->solr->service->executeRequest($solrRequest); |
|
163
|
|
|
return $this->solr->service->createResult($query, $response); |
|
164
|
|
|
} |
|
165
|
|
|
|
|
166
|
|
|
/** |
|
167
|
|
|
* Build SOLR query for given fields and parameters. |
|
168
|
|
|
* |
|
169
|
|
|
* @access private |
|
170
|
|
|
* |
|
171
|
|
|
* @param array|object $parameters parsed from request body |
|
172
|
|
|
* |
|
173
|
|
|
* @return string SOLR query |
|
174
|
|
|
*/ |
|
175
|
|
|
private function getQuery($parameters) |
|
176
|
|
|
{ |
|
177
|
|
|
return $this->fields['fulltext'] . ':(' . Solr::escapeQuery((string) $parameters['q']) . ') AND ' . $this->fields['uid'] . ':' . $this->getUid($parameters['uid']); |
|
178
|
|
|
} |
|
179
|
|
|
|
|
180
|
|
|
/** |
|
181
|
|
|
* Check if uid is number, if yes convert it to int, |
|
182
|
|
|
* otherwise leave uid not changed. |
|
183
|
|
|
* |
|
184
|
|
|
* @access private |
|
185
|
|
|
* |
|
186
|
|
|
* @param string $uid of the document |
|
187
|
|
|
* |
|
188
|
|
|
* @return int|string uid of the document |
|
189
|
|
|
*/ |
|
190
|
|
|
private function getUid($uid) |
|
191
|
|
|
{ |
|
192
|
|
|
return is_numeric($uid) ? intval($uid) : $uid; |
|
193
|
|
|
} |
|
194
|
|
|
} |
|
195
|
|
|
|
kitodo /
kitodo-presentation
Loading history...