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\Eid; |
14
|
|
|
|
15
|
|
|
use Kitodo\Dlf\Common\Helper; |
16
|
|
|
use Kitodo\Dlf\Common\Solr; |
17
|
|
|
use Kitodo\Dlf\Common\SolrSearchResult\ResultDocument; |
18
|
|
|
use Psr\Http\Message\ResponseInterface; |
19
|
|
|
use Psr\Http\Message\ServerRequestInterface; |
20
|
|
|
use TYPO3\CMS\Core\Http\Response; |
21
|
|
|
use TYPO3\CMS\Core\Utility\GeneralUtility; |
22
|
|
|
|
23
|
|
|
/** |
24
|
|
|
* eID search in document for plugin 'Search' of the 'dlf' extension |
25
|
|
|
* |
26
|
|
|
* @author Alexander Bigga <[email protected]> |
27
|
|
|
* @package TYPO3 |
28
|
|
|
* @subpackage dlf |
29
|
|
|
* @access public |
30
|
|
|
*/ |
31
|
|
|
class SearchInDocument |
32
|
|
|
{ |
33
|
|
|
/** |
34
|
|
|
* The main method of the eID script |
35
|
|
|
* |
36
|
|
|
* @param ServerRequestInterface $request |
37
|
|
|
* @return ResponseInterface JSON response of search suggestions |
38
|
|
|
*/ |
39
|
|
|
public function main(ServerRequestInterface $request) |
40
|
|
|
{ |
41
|
|
|
$output = [ |
42
|
|
|
'documents' => [], |
43
|
|
|
'numFound' => 0 |
44
|
|
|
]; |
45
|
|
|
// Get input parameters and decrypt core name. |
46
|
|
|
$parameters = $request->getParsedBody(); |
47
|
|
|
if ($parameters === null) { |
48
|
|
|
throw new \InvalidArgumentException('No parameters passed!', 1632322297); |
49
|
|
|
} |
50
|
|
|
$encrypted = (string) $parameters['encrypted']; |
51
|
|
|
$start = intval($parameters['start']); |
52
|
|
|
if (empty($encrypted)) { |
53
|
|
|
throw new \InvalidArgumentException('No valid parameter passed!', 1580585079); |
54
|
|
|
} |
55
|
|
|
|
56
|
|
|
$core = Helper::decrypt($encrypted); |
57
|
|
|
|
58
|
|
|
// Perform Solr query. |
59
|
|
|
$solr = Solr::getInstance($core); |
60
|
|
|
$fields = Solr::getFields(); |
61
|
|
|
|
62
|
|
|
if ($solr->ready) { |
63
|
|
|
$query = $solr->service->createSelect(); |
64
|
|
|
$query->setFields([$fields['id'], $fields['uid'], $fields['page']]); |
65
|
|
|
$query->setQuery($this->getQuery($fields, $parameters)); |
66
|
|
|
$query->setStart($start)->setRows(20); |
67
|
|
|
$query->addSort($fields['page'], $query::SORT_ASC); |
68
|
|
|
$query->getHighlighting(); |
69
|
|
|
$solrRequest = $solr->service->createRequest($query); |
70
|
|
|
|
71
|
|
|
// it is necessary to add the custom parameters to the request |
72
|
|
|
// because query object doesn't allow custom parameters |
73
|
|
|
|
74
|
|
|
// field for which highlighting is going to be performed, |
75
|
|
|
// is required if you want to have OCR highlighting |
76
|
|
|
$solrRequest->addParam('hl.ocr.fl', $fields['fulltext']); |
77
|
|
|
// return the coordinates of highlighted search as absolute coordinates |
78
|
|
|
$solrRequest->addParam('hl.ocr.absoluteHighlights', 'on'); |
79
|
|
|
// max amount of snippets for a single page |
80
|
|
|
$solrRequest->addParam('hl.snippets', 40); |
81
|
|
|
// we store the fulltext on page level and can disable this option |
82
|
|
|
$solrRequest->addParam('hl.ocr.trackPages', 'off'); |
83
|
|
|
|
84
|
|
|
$response = $solr->service->executeRequest($solrRequest); |
85
|
|
|
$result = $solr->service->createResult($query, $response); |
86
|
|
|
/** @scrutinizer ignore-call */ |
87
|
|
|
$output['numFound'] = $result->getNumFound(); |
88
|
|
|
$data = $result->getData(); |
89
|
|
|
$highlighting = $data['ocrHighlighting']; |
90
|
|
|
|
91
|
|
|
foreach ($result as $record) { |
92
|
|
|
$resultDocument = new ResultDocument($record, $highlighting, $fields); |
93
|
|
|
|
94
|
|
|
$document = [ |
95
|
|
|
'id' => $resultDocument->getId(), |
96
|
|
|
'uid' => !empty($resultDocument->getUid()) ? $resultDocument->getUid() : $parameters['uid'], |
97
|
|
|
'page' => $resultDocument->getPage(), |
98
|
|
|
'snippet' => $resultDocument->getSnippets(), |
99
|
|
|
'highlight' => $resultDocument->getHighlightsIds() |
100
|
|
|
]; |
101
|
|
|
$output['documents'][] = $document; |
102
|
|
|
} |
103
|
|
|
} |
104
|
|
|
// Create response object. |
105
|
|
|
/** @var Response $response */ |
106
|
|
|
$response = GeneralUtility::makeInstance(Response::class); |
107
|
|
|
$response->getBody()->write(json_encode($output)); |
108
|
|
|
return $response; |
109
|
|
|
} |
110
|
|
|
|
111
|
|
|
/** |
112
|
|
|
* Build SOLR query for given fields and parameters. |
113
|
|
|
* |
114
|
|
|
* @access private |
115
|
|
|
* |
116
|
|
|
* @param array $fields array of SOLR index fields |
117
|
|
|
* @param array|object $parameters parsed from request body |
118
|
|
|
* |
119
|
|
|
* @return string SOLR query |
120
|
|
|
*/ |
121
|
|
|
private function getQuery($fields, $parameters) |
122
|
|
|
{ |
123
|
|
|
return $fields['fulltext'] . ':(' . Solr::escapeQuery((string) $parameters['q']) . ') AND ' . $fields['uid'] . ':' . $this->getUid($parameters['uid']); |
124
|
|
|
} |
125
|
|
|
|
126
|
|
|
/** |
127
|
|
|
* Check if uid is number, if yes convert it to int, |
128
|
|
|
* otherwise leave uid not changed. |
129
|
|
|
* |
130
|
|
|
* @access private |
131
|
|
|
* |
132
|
|
|
* @param string $uid of the document |
133
|
|
|
* |
134
|
|
|
* @return int|string uid of the document |
135
|
|
|
*/ |
136
|
|
|
private function getUid($uid) |
137
|
|
|
{ |
138
|
|
|
return is_numeric($uid) ? intval($uid) : $uid; |
139
|
|
|
} |
140
|
|
|
} |
141
|
|
|
|