|
1
|
|
|
<?php |
|
2
|
|
|
/** |
|
3
|
|
|
* User: jensk |
|
4
|
|
|
* Date: 1-3-2017 |
|
5
|
|
|
* Time: 10:22 |
|
6
|
|
|
*/ |
|
7
|
|
|
|
|
8
|
|
|
namespace library\search\indexer; |
|
9
|
|
|
|
|
10
|
|
|
|
|
11
|
|
|
use library\search\DocumentTokenizer; |
|
12
|
|
|
use library\search\Indexer; |
|
13
|
|
|
use library\storage\JsonStorage; |
|
14
|
|
|
|
|
15
|
|
|
class TermCount |
|
16
|
|
|
{ |
|
17
|
|
|
protected $dbHandle; |
|
18
|
|
|
protected $documents; |
|
19
|
|
|
protected $filters; |
|
20
|
|
|
protected $storage; |
|
21
|
|
|
|
|
22
|
|
|
/** |
|
23
|
|
|
* TermCount constructor. |
|
24
|
|
|
* |
|
25
|
|
|
* @param resource $dbHandle |
|
26
|
|
|
* @param array $documents |
|
27
|
|
|
* @param array $filters |
|
28
|
|
|
* @param JsonStorage $jsonStorage |
|
29
|
|
|
*/ |
|
30
|
|
|
public function __construct($dbHandle, $documents, $filters, $jsonStorage) |
|
31
|
|
|
{ |
|
32
|
|
|
$this->dbHandle = $dbHandle; |
|
33
|
|
|
$this->documents = $documents; |
|
34
|
|
|
$this->filters = $filters; |
|
35
|
|
|
$this->storage = $jsonStorage; |
|
36
|
|
|
} |
|
37
|
|
|
|
|
38
|
|
|
public function execute() |
|
39
|
|
|
{ |
|
40
|
|
|
foreach ($this->documents as $document) { |
|
41
|
|
|
$tokenizer = new DocumentTokenizer($document, $this->storage); |
|
42
|
|
|
$tokens = $tokenizer->getTokens(); |
|
43
|
|
|
$documentTermCount = $this->applyFilters($tokens); |
|
44
|
|
|
$this->storeDocumentTermCount($document, $documentTermCount); |
|
45
|
|
|
} |
|
46
|
|
|
} |
|
47
|
|
|
|
|
48
|
|
View Code Duplication |
protected function applyFilters($tokens) |
|
|
|
|
|
|
49
|
|
|
{ |
|
50
|
|
|
foreach ($this->filters as $filterName) { |
|
51
|
|
|
$filterClassName = '\library\search\filters\\' . $filterName; |
|
52
|
|
|
$filter = new $filterClassName($tokens); |
|
53
|
|
|
$tokens = $filter->getFilterResults(); |
|
54
|
|
|
} |
|
55
|
|
|
return $tokens; |
|
56
|
|
|
} |
|
57
|
|
|
|
|
58
|
|
|
protected function storeDocumentTermCount($document, $documentTermCount) |
|
59
|
|
|
{ |
|
60
|
|
|
$db = $this->dbHandle; |
|
61
|
|
|
$sqlStart = ' |
|
62
|
|
|
INSERT INTO `term_count` (`documentPath`, `term`, `count`, `field`) |
|
63
|
|
|
VALUES '; |
|
64
|
|
|
$sql = $sqlStart; |
|
65
|
|
|
$values = array(); |
|
66
|
|
|
$quotedDocumentPath = $db->quote($document->path); |
|
|
|
|
|
|
67
|
|
|
$i = 0; |
|
68
|
|
|
foreach ($documentTermCount as $field => $countArray) { |
|
69
|
|
|
$quotedField = $db->quote($field); |
|
|
|
|
|
|
70
|
|
|
foreach ($countArray as $term => $count) { |
|
71
|
|
|
$values[] = $quotedDocumentPath . ', ' . $db->quote($term) . ', ' . $db->quote($count) . ', ' . $quotedField; |
|
|
|
|
|
|
72
|
|
|
$i += 1; |
|
73
|
|
|
if ($i >= Indexer::SQLITE_MAX_COMPOUND_SELECT) { |
|
74
|
|
|
$this->executeStoreDocumentTermCount($values, $sql, $db); |
|
75
|
|
|
$values = array(); |
|
76
|
|
|
$sql = $sqlStart; |
|
77
|
|
|
$i = 0; |
|
78
|
|
|
} |
|
79
|
|
|
} |
|
80
|
|
|
} |
|
81
|
|
|
if (count($values) != 0) { |
|
82
|
|
|
$this->executeStoreDocumentTermCount($values, $sql, $db); |
|
83
|
|
|
} |
|
84
|
|
|
} |
|
85
|
|
|
|
|
86
|
|
|
/** |
|
87
|
|
|
* @param $values |
|
88
|
|
|
* @param $sql |
|
89
|
|
|
* @param $db |
|
90
|
|
|
* |
|
91
|
|
|
* @throws \Exception |
|
92
|
|
|
*/ |
|
93
|
|
|
protected function executeStoreDocumentTermCount($values, $sql, $db) |
|
94
|
|
|
{ |
|
95
|
|
|
$sql .= '(' . implode('),' . PHP_EOL . '(', $values) . ');'; |
|
96
|
|
|
|
|
97
|
|
|
$stmt = $db->prepare($sql); |
|
98
|
|
View Code Duplication |
if ($stmt === false || !$stmt->execute()) { |
|
|
|
|
|
|
99
|
|
|
$errorInfo = $db->errorInfo(); |
|
100
|
|
|
$errorMsg = $errorInfo[2]; |
|
101
|
|
|
throw new \Exception('SQLite Exception: ' . $errorMsg . ' in SQL: <br /><pre>' . $sql . '</pre>'); |
|
102
|
|
|
} |
|
103
|
|
|
} |
|
104
|
|
|
} |
jenskooij /
cloudcontrol
Loading history...
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.