1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
declare(strict_types=1); |
4
|
|
|
|
5
|
|
|
namespace voku\helper; |
6
|
|
|
|
7
|
|
|
/** |
8
|
|
|
* Phonetic-Helper-Class |
9
|
|
|
* |
10
|
|
|
* @package voku\helper |
11
|
|
|
*/ |
12
|
|
|
final class StopWords |
13
|
|
|
{ |
14
|
|
|
/** |
15
|
|
|
* @var array |
16
|
|
|
*/ |
17
|
|
|
private static $availableLanguages = array( |
18
|
|
|
'ar', |
19
|
|
|
'bg', |
20
|
|
|
'ca', |
21
|
|
|
'cz', |
22
|
|
|
'da', |
23
|
|
|
'de', |
24
|
|
|
'el', |
25
|
|
|
'en', |
26
|
|
|
'eo', |
27
|
|
|
'es', |
28
|
|
|
'et', |
29
|
|
|
'fi', |
30
|
|
|
'fr', |
31
|
|
|
'hi', |
32
|
|
|
'hr', |
33
|
|
|
'hu', |
34
|
|
|
'id', |
35
|
|
|
'it', |
36
|
|
|
'ka', |
37
|
|
|
'lt', |
38
|
|
|
'lv', |
39
|
|
|
'nl', |
40
|
|
|
'no', |
41
|
|
|
'pl', |
42
|
|
|
'pt', |
43
|
|
|
'ro', |
44
|
|
|
'ru', |
45
|
|
|
'sk', |
46
|
|
|
'sv', |
47
|
|
|
'tr', |
48
|
|
|
'uk', |
49
|
|
|
'vi' |
50
|
|
|
); |
51
|
|
|
|
52
|
|
|
/** |
53
|
|
|
* @var array |
54
|
|
|
*/ |
55
|
|
|
private $stopWords = array(); |
56
|
|
|
|
57
|
|
|
/** |
58
|
|
|
* Load language-data from one language. |
59
|
|
|
* |
60
|
|
|
* @param string $language |
61
|
|
|
* |
62
|
|
|
* @throws StopWordsLanguageNotExists |
63
|
|
|
*/ |
64
|
2 |
View Code Duplication |
private function loadLanguageData(string $language = 'de') |
|
|
|
|
65
|
|
|
{ |
66
|
2 |
|
if (\in_array($language, self::$availableLanguages, true) === false) { |
67
|
|
|
throw new StopWordsLanguageNotExists('language not supported: ' . $language); |
68
|
|
|
} |
69
|
|
|
|
70
|
2 |
|
$this->stopWords[$language] = $this->getData($language); |
71
|
2 |
|
} |
72
|
|
|
|
73
|
|
|
/** |
74
|
|
|
* Get data from "/data/*.php". |
75
|
|
|
* |
76
|
|
|
* @param string $file |
77
|
|
|
* |
78
|
|
|
* @return array <p>Will return an empty array on error.</p> |
79
|
|
|
*/ |
80
|
2 |
|
private function getData(string $file): array |
81
|
|
|
{ |
82
|
2 |
|
static $RESULT_STOP_WORDS_CACHE = array(); |
83
|
|
|
|
84
|
2 |
|
if (isset($RESULT_STOP_WORDS_CACHE[$file])) { |
85
|
|
|
return $RESULT_STOP_WORDS_CACHE[$file]; |
86
|
|
|
} |
87
|
|
|
|
88
|
2 |
|
$file = __DIR__ . '/stopwords/' . $file . '.php'; |
89
|
2 |
|
if (file_exists($file)) { |
90
|
|
|
/** @noinspection PhpIncludeInspection */ |
91
|
2 |
|
$RESULT_STOP_WORDS_CACHE[$file] = require $file; |
92
|
|
|
} else { |
93
|
|
|
$RESULT_STOP_WORDS_CACHE[$file] = array(); |
94
|
|
|
} |
95
|
|
|
|
96
|
2 |
|
return $RESULT_STOP_WORDS_CACHE[$file]; |
97
|
|
|
} |
98
|
|
|
|
99
|
|
|
/** |
100
|
|
|
* Get the stop-words from one language. |
101
|
|
|
* |
102
|
|
|
* @param string $language |
103
|
|
|
* |
104
|
|
|
* @return array |
105
|
|
|
* |
106
|
|
|
* @throws StopWordsLanguageNotExists |
107
|
|
|
*/ |
108
|
2 |
View Code Duplication |
public function getStopWordsFromLanguage(string $language = 'de'): array |
|
|
|
|
109
|
|
|
{ |
110
|
2 |
|
if (\in_array($language, self::$availableLanguages, true) === false) { |
111
|
1 |
|
throw new StopWordsLanguageNotExists('language not supported: ' . $language); |
112
|
|
|
} |
113
|
|
|
|
114
|
1 |
|
if (!isset($this->stopWords[$language])) { |
115
|
1 |
|
$this->loadLanguageData($language); |
116
|
|
|
} |
117
|
|
|
|
118
|
1 |
|
return $this->stopWords[$language]; |
119
|
|
|
} |
120
|
|
|
|
121
|
1 |
|
private function loadLanguageDataAll() |
122
|
|
|
{ |
123
|
1 |
|
foreach (self::$availableLanguages as $language) { |
124
|
1 |
|
if (!isset($this->stopWords[$language])) { |
125
|
1 |
|
$this->loadLanguageData($language); |
126
|
|
|
} |
127
|
|
|
} |
128
|
1 |
|
} |
129
|
|
|
|
130
|
|
|
/** |
131
|
|
|
* Get all stop-words from all languages. |
132
|
|
|
* |
133
|
|
|
* @return array |
134
|
|
|
* |
135
|
|
|
* @throws StopWordsLanguageNotExists |
136
|
|
|
*/ |
137
|
1 |
|
public function getStopWordsAll(): array |
138
|
|
|
{ |
139
|
1 |
|
$this->loadLanguageDataAll(); |
140
|
|
|
|
141
|
1 |
|
return $this->stopWords; |
142
|
|
|
} |
143
|
|
|
} |
144
|
|
|
|
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.