1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
/* |
4
|
|
|
* This file is part of Crawler Detect - the web crawler detection library. |
5
|
|
|
* |
6
|
|
|
* (c) Mark Beech <[email protected]> |
7
|
|
|
* |
8
|
|
|
* This source file is subject to the MIT license that is bundled |
9
|
|
|
* with this source code in the file LICENSE. |
10
|
|
|
*/ |
11
|
|
|
|
12
|
|
|
use Jaybizzle\CrawlerDetect\CrawlerDetect; |
13
|
|
|
use Jaybizzle\CrawlerDetect\Fixtures\Crawlers; |
14
|
|
|
use PHPUnit\Framework\TestCase; |
15
|
|
|
|
16
|
|
|
final class UserAgentTest extends TestCase |
17
|
|
|
{ |
18
|
|
|
public $CrawlerDetect; |
19
|
|
|
|
20
|
|
|
/** @test */ |
21
|
|
View Code Duplication |
public function user_agents_are_bots() |
22
|
|
|
{ |
23
|
|
|
$this->CrawlerDetect = new CrawlerDetect(); |
24
|
|
|
$lines = file(__DIR__.'/crawlers.txt', FILE_IGNORE_NEW_LINES | FILE_SKIP_EMPTY_LINES); |
25
|
|
|
|
26
|
|
|
foreach ($lines as $line) { |
27
|
|
|
$test = $this->CrawlerDetect->isCrawler($line); |
28
|
|
|
$this->assertTrue($test, $line); |
29
|
|
|
} |
30
|
|
|
} |
31
|
|
|
|
32
|
|
|
/** @test */ |
33
|
|
View Code Duplication |
public function user_agents_are_devices() |
34
|
|
|
{ |
35
|
|
|
$this->CrawlerDetect = new CrawlerDetect(); |
36
|
|
|
$lines = file(__DIR__.'/devices.txt', FILE_IGNORE_NEW_LINES | FILE_SKIP_EMPTY_LINES); |
37
|
|
|
|
38
|
|
|
foreach ($lines as $line) { |
39
|
|
|
$test = $this->CrawlerDetect->isCrawler($line); |
40
|
|
|
$this->assertFalse($test, $line); |
41
|
|
|
} |
42
|
|
|
} |
43
|
|
|
|
44
|
|
|
/** @test */ |
45
|
|
View Code Duplication |
public function it_returns_correct_matched_bot_name() |
46
|
|
|
{ |
47
|
|
|
$this->CrawlerDetect = new CrawlerDetect(); |
48
|
|
|
$this->CrawlerDetect->isCrawler('Mozilla/5.0 (iPhone; CPU iPhone OS 7_1 like Mac OS X) AppleWebKit (KHTML, like Gecko) Mobile (compatible; Yahoo Ad monitoring; https://help.yahoo.com/kb/yahoo-ad-monitoring-SLN24857.html)'); |
49
|
|
|
|
50
|
|
|
$matches = $this->CrawlerDetect->getMatches(); |
51
|
|
|
|
52
|
|
|
$this->assertEquals($this->CrawlerDetect->getMatches(), 'monitoring', $matches); |
53
|
|
|
} |
54
|
|
|
|
55
|
|
|
/** @test */ |
56
|
|
View Code Duplication |
public function it_returns_full_matched_bot_name() |
57
|
|
|
{ |
58
|
|
|
$this->CrawlerDetect = new CrawlerDetect(); |
59
|
|
|
$this->CrawlerDetect->isCrawler('somenaughtybot'); |
60
|
|
|
|
61
|
|
|
$matches = $this->CrawlerDetect->getMatches(); |
62
|
|
|
|
63
|
|
|
$this->assertEquals($this->CrawlerDetect->getMatches(), 'somenaughtybot', $matches); |
64
|
|
|
} |
65
|
|
|
|
66
|
|
|
/** @test */ |
67
|
|
|
public function it_returns_null_when_no_bot_detected() |
68
|
|
|
{ |
69
|
|
|
$this->CrawlerDetect = new CrawlerDetect(); |
70
|
|
|
$this->CrawlerDetect->isCrawler('nothing to see here'); |
71
|
|
|
|
72
|
|
|
$this->assertNull($this->CrawlerDetect->getMatches()); |
73
|
|
|
} |
74
|
|
|
|
75
|
|
|
/** @test */ |
76
|
|
|
public function empty_user_agent() |
77
|
|
|
{ |
78
|
|
|
$this->CrawlerDetect = new CrawlerDetect(); |
79
|
|
|
$test = $this->CrawlerDetect->isCrawler(' '); |
80
|
|
|
|
81
|
|
|
$this->assertFalse($test); |
82
|
|
|
} |
83
|
|
|
|
84
|
|
|
/** @test */ |
85
|
|
|
public function current_visitor() |
86
|
|
|
{ |
87
|
|
|
$headers = (array) json_decode('{"DOCUMENT_ROOT":"\/home\/test\/public_html","GATEWAY_INTERFACE":"CGI\/1.1","HTTP_ACCEPT":"*\/*","HTTP_ACCEPT_ENCODING":"gzip, deflate","HTTP_CACHE_CONTROL":"no-cache","HTTP_CONNECTION":"Keep-Alive","HTTP_FROM":"bingbot(at)microsoft.com","HTTP_HOST":"www.test.com","HTTP_PRAGMA":"no-cache","HTTP_USER_AGENT":"Mozilla\/5.0 (compatible; bingbot\/2.0; +http:\/\/www.bing.com\/bingbot.htm)","PATH":"\/bin:\/usr\/bin","QUERY_STRING":"order=closingDate","REDIRECT_STATUS":"200","REMOTE_ADDR":"127.0.0.1","REMOTE_PORT":"3360","REQUEST_METHOD":"GET","REQUEST_URI":"\/?test=testing","SCRIPT_FILENAME":"\/home\/test\/public_html\/index.php","SCRIPT_NAME":"\/index.php","SERVER_ADDR":"127.0.0.1","SERVER_ADMIN":"[email protected]","SERVER_NAME":"www.test.com","SERVER_PORT":"80","SERVER_PROTOCOL":"HTTP\/1.1","SERVER_SIGNATURE":"","SERVER_SOFTWARE":"Apache","UNIQUE_ID":"Vx6MENRxerBUSDEQgFLAAAAAS","PHP_SELF":"\/index.php","REQUEST_TIME_FLOAT":1461619728.0705,"REQUEST_TIME":1461619728}'); |
88
|
|
|
|
89
|
|
|
$cd = new CrawlerDetect($headers); |
90
|
|
|
|
91
|
|
|
$this->assertTrue($cd->isCrawler()); |
92
|
|
|
} |
93
|
|
|
|
94
|
|
|
/** @test */ |
95
|
|
|
public function user_agent_passed_via_contructor() |
96
|
|
|
{ |
97
|
|
|
$cd = new CrawlerDetect(null, 'Mozilla/5.0 (iPhone; CPU iPhone OS 7_1 like Mac OS X) AppleWebKit (KHTML, like Gecko) Mobile (compatible; Yahoo Ad monitoring; https://help.yahoo.com/kb/yahoo-ad-monitoring-SLN24857.html)'); |
98
|
|
|
|
99
|
|
|
$this->assertTrue($cd->isCrawler()); |
100
|
|
|
} |
101
|
|
|
|
102
|
|
|
/** @test */ |
103
|
|
|
public function http_from_header() |
104
|
|
|
{ |
105
|
|
|
$headers = (array) json_decode('{"DOCUMENT_ROOT":"\/home\/test\/public_html","GATEWAY_INTERFACE":"CGI\/1.1","HTTP_ACCEPT":"*\/*","HTTP_ACCEPT_ENCODING":"gzip, deflate","HTTP_CACHE_CONTROL":"no-cache","HTTP_CONNECTION":"Keep-Alive","HTTP_FROM":"googlebot(at)googlebot.com","HTTP_HOST":"www.test.com","HTTP_PRAGMA":"no-cache","HTTP_USER_AGENT":"Mozilla\/5.0 (Macintosh; Intel Mac OS X 10_8_4) AppleWebKit\/537.36 (KHTML, like Gecko) Chrome\/28.0.1500.71 Safari\/537.36","PATH":"\/bin:\/usr\/bin","QUERY_STRING":"order=closingDate","REDIRECT_STATUS":"200","REMOTE_ADDR":"127.0.0.1","REMOTE_PORT":"3360","REQUEST_METHOD":"GET","REQUEST_URI":"\/?test=testing","SCRIPT_FILENAME":"\/home\/test\/public_html\/index.php","SCRIPT_NAME":"\/index.php","SERVER_ADDR":"127.0.0.1","SERVER_ADMIN":"[email protected]","SERVER_NAME":"www.test.com","SERVER_PORT":"80","SERVER_PROTOCOL":"HTTP\/1.1","SERVER_SIGNATURE":"","SERVER_SOFTWARE":"Apache","UNIQUE_ID":"Vx6MENRxerBUSDEQgFLAAAAAS","PHP_SELF":"\/index.php","REQUEST_TIME_FLOAT":1461619728.0705,"REQUEST_TIME":1461619728}'); |
106
|
|
|
|
107
|
|
|
$cd = new CrawlerDetect($headers); |
108
|
|
|
|
109
|
|
|
$this->assertTrue($cd->isCrawler()); |
110
|
|
|
} |
111
|
|
|
|
112
|
|
|
/** @test */ |
113
|
|
|
public function matches_does_not_persit_across_multiple_calls() |
114
|
|
|
{ |
115
|
|
|
$this->CrawlerDetect = new CrawlerDetect(); |
116
|
|
|
$this->CrawlerDetect->isCrawler('Mozilla/5.0 (iPhone; CPU iPhone OS 7_1 like Mac OS X) AppleWebKit (KHTML, like Gecko) Mobile (compatible; Yahoo Ad monitoring; https://help.yahoo.com/kb/yahoo-ad-monitoring-SLN24857.html)'); |
117
|
|
|
$matches = $this->CrawlerDetect->getMatches(); |
118
|
|
|
$this->assertEquals($this->CrawlerDetect->getMatches(), 'monitoring', $matches); |
119
|
|
|
|
120
|
|
|
$this->CrawlerDetect->isCrawler('This should not match'); |
121
|
|
|
$matches = $this->CrawlerDetect->getMatches(); |
|
|
|
|
122
|
|
|
$this->assertNull($this->CrawlerDetect->getMatches()); |
123
|
|
|
} |
124
|
|
|
|
125
|
|
|
/** @test */ |
126
|
|
|
public function the_regex_patterns_are_unique() |
127
|
|
|
{ |
128
|
|
|
$crawlers = new Crawlers(); |
129
|
|
|
|
130
|
|
|
$this->assertEquals(count($crawlers->getAll()), count(array_unique($crawlers->getAll()))); |
131
|
|
|
} |
132
|
|
|
|
133
|
|
|
/** @test */ |
134
|
|
|
public function there_are_no_regex_collisions() |
135
|
|
|
{ |
136
|
|
|
$crawlers = new Crawlers(); |
137
|
|
|
|
138
|
|
|
foreach ($crawlers->getAll() as $key1 => $regex) { |
139
|
|
|
foreach ($crawlers->getAll() as $key2 => $compare) { |
140
|
|
|
// Dont check this regex against itself |
141
|
|
|
if ($key1 != $key2) { |
142
|
|
|
preg_match('/'.$regex.'/i', stripslashes($compare), $matches); |
143
|
|
|
|
144
|
|
|
$this->assertEmpty($matches, $regex.' collided with '.$compare); |
145
|
|
|
} |
146
|
|
|
} |
147
|
|
|
} |
148
|
|
|
} |
149
|
|
|
} |
150
|
|
|
|
This check looks for variable assignements that are either overwritten by other assignments or where the variable is not used subsequently.
Both the
$myVar
assignment in line 1 and the$higher
assignment in line 2 are dead. The first because$myVar
is never used and the second because$higher
is always overwritten for every possible time line.