1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
namespace Onoi\Tesa\Tests; |
4
|
|
|
|
5
|
|
|
use Onoi\Tesa\StopwordAnalyzer; |
6
|
|
|
|
7
|
|
|
/** |
8
|
|
|
* @covers \Onoi\Tesa\StopwordAnalyzer |
9
|
|
|
* @group onoi-tesa |
10
|
|
|
* |
11
|
|
|
* @license GNU GPL v2+ |
12
|
|
|
* @since 0.1 |
13
|
|
|
* |
14
|
|
|
* @author mwjames |
15
|
|
|
*/ |
16
|
|
|
class StopwordAnalyzerTest extends \PHPUnit_Framework_TestCase { |
17
|
|
|
|
18
|
|
|
public function testCanConstruct() { |
19
|
|
|
|
20
|
|
|
$this->assertInstanceOf( |
21
|
|
|
'\Onoi\Tesa\StopwordAnalyzer', |
22
|
|
|
new StopwordAnalyzer() |
23
|
|
|
); |
24
|
|
|
} |
25
|
|
|
|
26
|
|
|
public function testCustomStopwordList() { |
27
|
|
|
|
28
|
|
|
$instance = new StopwordAnalyzer(); |
29
|
|
|
$instance->setCustomStopwordList( array( 'zoo' => array( 'Foo' ) ) ); |
30
|
|
|
|
31
|
|
|
$this->assertTrue( |
32
|
|
|
$instance->isStopWord( 'Foo' ) |
33
|
|
|
); |
34
|
|
|
} |
35
|
|
|
|
36
|
|
|
/** |
37
|
|
|
* @dataProvider defaultListStopWordsProvider |
38
|
|
|
*/ |
39
|
|
|
public function testUncachedIsWordCheckForDefaultList( $word, $expected ) { |
40
|
|
|
|
41
|
|
|
$instance = new StopwordAnalyzer(); |
42
|
|
|
$instance->loadListBy( StopwordAnalyzer::DEFAULT_STOPWORDLIST ); |
43
|
|
|
|
44
|
|
|
$this->assertEquals( |
45
|
|
|
$expected, |
46
|
|
|
$instance->isStopWord( $word ) |
47
|
|
|
); |
48
|
|
|
} |
49
|
|
|
|
50
|
|
|
/** |
51
|
|
|
* @dataProvider listByLanguageStopWordsProvider |
52
|
|
|
*/ |
53
|
|
|
public function testListByLanguage( $languageCode, $word, $expected ) { |
54
|
|
|
|
55
|
|
|
$instance = new StopwordAnalyzer(); |
56
|
|
|
$instance->loadListByLanguage( $languageCode ); |
57
|
|
|
|
58
|
|
|
$this->assertEquals( |
59
|
|
|
$expected, |
60
|
|
|
$instance->isStopWord( $word ) |
61
|
|
|
); |
62
|
|
|
} |
63
|
|
|
|
64
|
|
|
public function testTryToLoadUnlistedLanguage() { |
65
|
|
|
|
66
|
|
|
$instance = new StopwordAnalyzer(); |
67
|
|
|
$instance->loadListByLanguage( 'foo' ); |
68
|
|
|
|
69
|
|
|
$this->assertFalse( |
70
|
|
|
$instance->isStopWord( 'Bar' ) |
71
|
|
|
); |
72
|
|
|
} |
73
|
|
|
|
74
|
|
|
public function testToLoadFromCache() { |
75
|
|
|
|
76
|
|
|
$cache = $this->getMockBuilder( '\Onoi\Cache\Cache' ) |
77
|
|
|
->disableOriginalConstructor() |
78
|
|
|
->setMethods( array( 'contains', 'fetch' ) ) |
79
|
|
|
->getMockForAbstractClass(); |
80
|
|
|
|
81
|
|
|
$cache->expects( $this->once() ) |
82
|
|
|
->method( 'contains' ) |
83
|
|
|
->with( $this->stringContains( 'onoi:tesa:stopword' ) ) |
84
|
|
|
->will( $this->returnValue( true ) ); |
85
|
|
|
|
86
|
|
|
$cache->expects( $this->once() ) |
87
|
|
|
->method( 'fetch' ); |
88
|
|
|
|
89
|
|
|
$instance = new StopwordAnalyzer( $cache ); |
90
|
|
|
$instance->loadListByLanguage( 'foo' ); |
91
|
|
|
} |
92
|
|
|
|
93
|
|
|
public function defaultListStopWordsProvider() { |
94
|
|
|
|
95
|
|
|
$provider[] = array( |
|
|
|
|
96
|
|
|
'Foo', |
97
|
|
|
false |
98
|
|
|
); |
99
|
|
|
|
100
|
|
|
// en-list |
101
|
|
|
$provider[] = array( |
102
|
|
|
'about', |
103
|
|
|
true |
104
|
|
|
); |
105
|
|
|
|
106
|
|
|
// de-list |
107
|
|
|
$provider[] = array( |
108
|
|
|
'über', |
109
|
|
|
true |
110
|
|
|
); |
111
|
|
|
|
112
|
|
|
// fr-list |
113
|
|
|
$provider[] = array( |
114
|
|
|
'avoir', |
115
|
|
|
true |
116
|
|
|
); |
117
|
|
|
|
118
|
|
|
// es-list |
119
|
|
|
$provider[] = array( |
120
|
|
|
'aquellos', |
121
|
|
|
true |
122
|
|
|
); |
123
|
|
|
|
124
|
|
|
// ja-list (not default) |
125
|
|
|
$provider[] = array( |
126
|
|
|
'かつて', |
127
|
|
|
false |
128
|
|
|
); |
129
|
|
|
|
130
|
|
|
return $provider; |
131
|
|
|
} |
132
|
|
|
|
133
|
|
|
public function listByLanguageStopWordsProvider() { |
134
|
|
|
|
135
|
|
|
// ja-list |
136
|
|
|
$provider[] = array( |
|
|
|
|
137
|
|
|
'ja', |
138
|
|
|
'かつて', |
139
|
|
|
true |
140
|
|
|
); |
141
|
|
|
|
142
|
|
|
return $provider; |
143
|
|
|
} |
144
|
|
|
|
145
|
|
|
} |
146
|
|
|
|
Adding an explicit array definition is generally preferable to implicit array definition as it guarantees a stable state of the code.
Let’s take a look at an example:
As you can see in this example, the array
$myArray
is initialized the first time when the foreach loop is entered. You can also see that the value of thebar
key is only written conditionally; thus, its value might result from a previous iteration.This might or might not be intended. To make your intention clear, your code more readible and to avoid accidental bugs, we recommend to add an explicit initialization $myArray = array() either outside or inside the foreach loop.