UserAgentTest::user_agents_are_bots() - Code Metrics - Inspection of "Test against PHP 8 (#432)" - JayBizzle/Crawler-Detect - Measure and Improve Code Quality continuously with Scrutinizer

Completed

Push — master ( 60170b...6034e2 )

by Mark

created 2020-12-09 20:41 UTC

UserAgentTest::user_agents_are_bots() A

↳ Parent: UserAgentTest

Complexity

Conditions	2
Paths	2

Size

Total Lines

Duplication

Lines	10
Ratio	100 %

Importance

Changes

Metric	Value
cc	2
nc	2
nop	0
dl	10
loc	10
rs	9.9332
c	0
b	0
f	0

<?php

/*
 * This file is part of Crawler Detect - the web crawler detection library.
 *
 * (c) Mark Beech <[email protected]>
 *
 * This source file is subject to the MIT license that is bundled
 * with this source code in the file LICENSE.
 */

use Jaybizzle\CrawlerDetect\CrawlerDetect;
use Jaybizzle\CrawlerDetect\Fixtures\Crawlers;
use PHPUnit\Framework\TestCase;

final class UserAgentTest extends TestCase
{
    /** @test */
    public function user_agents_are_bots()
    {
        $this->CrawlerDetect = new CrawlerDetect();
class MyClass { }

$x = new MyClass();
$x->foo = true;
        $lines = file(__DIR__.'/crawlers.txt', FILE_IGNORE_NEW_LINES | FILE_SKIP_EMPTY_LINES);

        foreach ($lines as $line) {
            $test = $this->CrawlerDetect->isCrawler($line);
            $this->assertTrue($test, $line);
        }
    }

    /** @test */
    public function user_agents_are_devices()
    {
        $this->CrawlerDetect = new CrawlerDetect();
        $lines = file(__DIR__.'/devices.txt', FILE_IGNORE_NEW_LINES | FILE_SKIP_EMPTY_LINES);

        foreach ($lines as $line) {
            $test = $this->CrawlerDetect->isCrawler($line);
            $this->assertFalse($test, $line);
        }
    }

    /** @test */
    public function it_returns_correct_matched_bot_name()
    {
        $this->CrawlerDetect = new CrawlerDetect();
        $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)');

        $matches = $this->CrawlerDetect->getMatches();

        $this->assertEquals($this->CrawlerDetect->getMatches(), 'monitoring', $matches);
    }

    /** @test */
    public function it_returns_full_matched_bot_name()
    {
        $this->CrawlerDetect = new CrawlerDetect();
        $this->CrawlerDetect->isCrawler('somenaughtybot');

        $matches = $this->CrawlerDetect->getMatches();

        $this->assertEquals($this->CrawlerDetect->getMatches(), 'somenaughtybot', $matches);
    }

    /** @test */
    public function it_returns_null_when_no_bot_detected()
    {
        $this->CrawlerDetect = new CrawlerDetect();
        $this->CrawlerDetect->isCrawler('nothing to see here');

        $matches = $this->CrawlerDetect->getMatches();
$myVar = 'Value';
$higher = false;

if (rand(1, 6) > 3) {
    $higher = true;
} else {
    $higher = false;
}

        $this->assertNull($this->CrawlerDetect->getMatches());
    }

    /** @test */
    public function empty_user_agent()
    {
        $this->CrawlerDetect = new CrawlerDetect();
        $test = $this->CrawlerDetect->isCrawler('      ');

        $this->assertFalse($test);
    }

    /** @test */
    public function current_visitor()
    {
        $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}');

        $cd = new CrawlerDetect($headers);

        $this->assertTrue($cd->isCrawler());
    }

    /** @test */
    public function user_agent_passed_via_contructor()
    {
        $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)');

        $this->assertTrue($cd->isCrawler());
    }

    /** @test */
    public function http_from_header()
    {
        $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}');

        $cd = new CrawlerDetect($headers);

        $this->assertTrue($cd->isCrawler());
    }

    /** @test */
    public function matches_does_not_persit_across_multiple_calls()
    {
        $this->CrawlerDetect = new CrawlerDetect();
        $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)');
        $matches = $this->CrawlerDetect->getMatches();
        $this->assertEquals($this->CrawlerDetect->getMatches(), 'monitoring', $matches);

        $this->CrawlerDetect->isCrawler('This should not match');
        $matches = $this->CrawlerDetect->getMatches();
$myVar = 'Value';
$higher = false;

if (rand(1, 6) > 3) {
    $higher = true;
} else {
    $higher = false;
}
        $this->assertNull($this->CrawlerDetect->getMatches());
    }

    /** @test */
    public function the_regex_patterns_are_unique()
    {
        $crawlers = new Crawlers();

        $this->assertEquals(count($crawlers->getAll()), count(array_unique($crawlers->getAll())));
    }

    /** @test */
    public function there_are_no_regex_collisions()
    {
        $crawlers = new Crawlers();

        foreach ($crawlers->getAll() as $key1 => $regex) {
            foreach ($crawlers->getAll() as $key2 => $compare) {
                // Dont check this regex against itself
                if ($key1 != $key2) {
                    preg_match('/'.$regex.'/i', stripslashes($compare), $matches);

                    $this->assertEmpty($matches, $regex.' collided with '.$compare);
                }
            }
        }
    }
}


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		/** @test */
19	View Code Duplication	public function user_agents_are_bots()
20		{
21		$this->CrawlerDetect = new CrawlerDetect();
		0 ignored issues – show Bug introduced 2020-12-09 20:01 UTC by Report Bug Copy Issue Report The property `CrawlerDetect` does not exist. Did you maybe forget to declare it? In PHP it is possible to write to properties without declaring them. For example, the following is perfectly valid PHP code: class MyClass { } $x = new MyClass(); $x->foo = true; Generally, it is a good practice to explictly declare properties to avoid accidental typos and provide IDE auto-completion: class MyClass { public $foo; } $x = new MyClass(); $x->foo = true; Loading history...
22		$lines = file(__DIR__.'/crawlers.txt', FILE_IGNORE_NEW_LINES \| FILE_SKIP_EMPTY_LINES);
23
24		foreach ($lines as $line) {
25		$test = $this->CrawlerDetect->isCrawler($line);
26		$this->assertTrue($test, $line);
27		}
28		}
29
30		/** @test */
31	View Code Duplication	public function user_agents_are_devices()
32		{
33		$this->CrawlerDetect = new CrawlerDetect();
34		$lines = file(__DIR__.'/devices.txt', FILE_IGNORE_NEW_LINES \| FILE_SKIP_EMPTY_LINES);
35
36		foreach ($lines as $line) {
37		$test = $this->CrawlerDetect->isCrawler($line);
38		$this->assertFalse($test, $line);
39		}
40		}
41
42		/** @test */
43	View Code Duplication	public function it_returns_correct_matched_bot_name()
44		{
45		$this->CrawlerDetect = new CrawlerDetect();
46		$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)');
47
48		$matches = $this->CrawlerDetect->getMatches();
49
50		$this->assertEquals($this->CrawlerDetect->getMatches(), 'monitoring', $matches);
51		}
52
53		/** @test */
54	View Code Duplication	public function it_returns_full_matched_bot_name()
55		{
56		$this->CrawlerDetect = new CrawlerDetect();
57		$this->CrawlerDetect->isCrawler('somenaughtybot');
58
59		$matches = $this->CrawlerDetect->getMatches();
60
61		$this->assertEquals($this->CrawlerDetect->getMatches(), 'somenaughtybot', $matches);
62		}
63
64		/** @test */
65	View Code Duplication	public function it_returns_null_when_no_bot_detected()
66		{
67		$this->CrawlerDetect = new CrawlerDetect();
68		$this->CrawlerDetect->isCrawler('nothing to see here');
69
70		$matches = $this->CrawlerDetect->getMatches();
		0 ignored issues – show Unused Code introduced 2020-12-09 20:10 UTC by Report Bug Copy Issue Report `$matches` is not used, you could remove the assignment. This check looks for variable assignements that are either overwritten by other assignments or where the variable is not used subsequently. $myVar = 'Value'; $higher = false; if (rand(1, 6) > 3) { $higher = true; } else { $higher = false; } Both the `$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. Loading history...
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 = $this->CrawlerDetect->getMatches();
		0 ignored issues – show Unused Code introduced 2020-08-25 19:32 UTC by Report Bug Copy Issue Report `$matches` is not used, you could remove the assignment. This check looks for variable assignements that are either overwritten by other assignments or where the variable is not used subsequently. $myVar = 'Value'; $higher = false; if (rand(1, 6) > 3) { $higher = true; } else { $higher = false; } Both the `$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. Loading history...
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

JayBizzle / Crawler-Detect

Push — master ( 60170b...6034e2 )

UserAgentTest::user_agents_are_bots() A

Complexity

Size

Duplication

Importance

Duplication Side-by-Side

Filter issues like