|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
namespace Spatie\Robots; |
|
4
|
|
|
|
|
5
|
|
|
class RobotsTxt |
|
6
|
|
|
{ |
|
7
|
|
|
protected static $robotsCache = []; |
|
8
|
|
|
|
|
9
|
|
|
protected $disallowsPerUserAgent = []; |
|
10
|
|
|
|
|
11
|
|
|
public static function readFrom(string $source): self |
|
12
|
|
|
{ |
|
13
|
|
|
$content = @file_get_contents($source); |
|
14
|
|
|
|
|
15
|
|
|
return new self($content !== false ? $content : ''); |
|
16
|
|
|
} |
|
17
|
|
|
|
|
18
|
|
|
public function __construct(string $content) |
|
19
|
|
|
{ |
|
20
|
|
|
$this->disallowsPerUserAgent = $this->getDisallowsPerUserAgent($content); |
|
21
|
|
|
} |
|
22
|
|
|
|
|
23
|
|
|
public static function create(string $source): self |
|
24
|
|
|
{ |
|
25
|
|
|
if ( |
|
26
|
|
|
strpos($source, 'http') !== false |
|
27
|
|
|
&& strpos($source, 'robots.txt') !== false |
|
28
|
|
|
) { |
|
29
|
|
|
return self::readFrom($source); |
|
30
|
|
|
} |
|
31
|
|
|
|
|
32
|
|
|
return new self($source); |
|
33
|
|
|
} |
|
34
|
|
|
|
|
35
|
|
|
public function allows(string $url, ?string $userAgent = '*'): bool |
|
36
|
|
|
{ |
|
37
|
|
|
$requestUri = ''; |
|
38
|
|
|
|
|
39
|
|
|
$parts = parse_url($url); |
|
40
|
|
|
|
|
41
|
|
|
if ($parts !== false) { |
|
42
|
|
|
if (isset($parts['path'])) { |
|
43
|
|
|
$requestUri .= $parts['path']; |
|
44
|
|
|
} |
|
45
|
|
|
|
|
46
|
|
|
if (isset($parts['query'])) { |
|
47
|
|
|
$requestUri .= '?'.$parts['query']; |
|
48
|
|
|
} elseif ($this->hasEmptyQueryString($url)) { |
|
49
|
|
|
$requestUri .= '?'; |
|
50
|
|
|
} |
|
51
|
|
|
} |
|
52
|
|
|
|
|
53
|
|
|
$disallows = $this->disallowsPerUserAgent[$userAgent] ?? $this->disallowsPerUserAgent['*'] ?? []; |
|
54
|
|
|
|
|
55
|
|
|
return ! $this->pathIsDenied($requestUri, $disallows); |
|
56
|
|
|
} |
|
57
|
|
|
|
|
58
|
|
|
protected function pathIsDenied(string $requestUri, array $disallows): bool |
|
59
|
|
|
{ |
|
60
|
|
|
foreach ($disallows as $disallow) { |
|
61
|
|
|
if ($disallow === '') { |
|
62
|
|
|
continue; |
|
63
|
|
|
} |
|
64
|
|
|
|
|
65
|
|
|
$stopAtEndOfString = false; |
|
66
|
|
|
|
|
67
|
|
|
if ($disallow[-1] === '$') { |
|
68
|
|
|
// if the pattern ends with a dollar sign, the string must end there |
|
69
|
|
|
$disallow = substr($disallow, 0, -1); |
|
70
|
|
|
$stopAtEndOfString = true; |
|
71
|
|
|
} |
|
72
|
|
|
|
|
73
|
|
|
// convert to regexp |
|
74
|
|
|
$disallowRegexp = preg_quote($disallow, '/'); |
|
75
|
|
|
|
|
76
|
|
|
// the pattern must start at the beginning of the string... |
|
77
|
|
|
$disallowRegexp = '^'.$disallowRegexp; |
|
78
|
|
|
|
|
79
|
|
|
// ...and optionally stop at the end of the string |
|
80
|
|
|
if ($stopAtEndOfString) { |
|
81
|
|
|
$disallowRegexp .= '$'; |
|
82
|
|
|
} |
|
83
|
|
|
|
|
84
|
|
|
// replace (preg_quote'd) stars with an eager match |
|
85
|
|
|
$disallowRegexp = str_replace('\\*', '.*', $disallowRegexp); |
|
86
|
|
|
|
|
87
|
|
|
// enclose in delimiters |
|
88
|
|
|
$disallowRegexp = '/'.$disallowRegexp.'/'; |
|
89
|
|
|
|
|
90
|
|
|
if (preg_match($disallowRegexp, $requestUri) === 1) { |
|
91
|
|
|
return true; |
|
92
|
|
|
} |
|
93
|
|
|
} |
|
94
|
|
|
|
|
95
|
|
|
return false; |
|
96
|
|
|
} |
|
97
|
|
|
|
|
98
|
|
|
/** |
|
99
|
|
|
* Checks for an empty query string. |
|
100
|
|
|
* |
|
101
|
|
|
* This works around the fact that parse_url() will not set the 'query' key when the query string is empty. |
|
102
|
|
|
* See: https://bugs.php.net/bug.php?id=78385 |
|
103
|
|
|
*/ |
|
104
|
|
|
protected function hasEmptyQueryString(string $url) : bool |
|
105
|
|
|
{ |
|
106
|
|
|
if ($url === '') { |
|
107
|
|
|
return false; |
|
108
|
|
|
} |
|
109
|
|
|
|
|
110
|
|
|
if ($url[-1] === '?') { // ends with ? |
|
111
|
|
|
return true; |
|
112
|
|
|
} |
|
113
|
|
|
|
|
114
|
|
|
if (strpos($url, '?#') !== false) { // empty query string, followed by a fragment |
|
115
|
|
|
return true; |
|
116
|
|
|
} |
|
117
|
|
|
|
|
118
|
|
|
return false; |
|
119
|
|
|
} |
|
120
|
|
|
|
|
121
|
|
|
protected function getDisallowsPerUserAgent(string $content): array |
|
122
|
|
|
{ |
|
123
|
|
|
$lines = explode(PHP_EOL, $content); |
|
124
|
|
|
|
|
125
|
|
|
$lines = array_filter($lines); |
|
126
|
|
|
|
|
127
|
|
|
$disallowsPerUserAgent = []; |
|
128
|
|
|
|
|
129
|
|
|
$currentUserAgent = null; |
|
130
|
|
|
|
|
131
|
|
|
foreach ($lines as $line) { |
|
132
|
|
|
if ($this->isCommentLine($line)) { |
|
133
|
|
|
continue; |
|
134
|
|
|
} |
|
135
|
|
|
|
|
136
|
|
|
if ($this->isUserAgentLine($line)) { |
|
137
|
|
|
$disallowsPerUserAgent[$this->parseUserAgent($line)] = []; |
|
138
|
|
|
|
|
139
|
|
|
$currentUserAgent = &$disallowsPerUserAgent[$this->parseUserAgent($line)]; |
|
140
|
|
|
|
|
141
|
|
|
continue; |
|
142
|
|
|
} |
|
143
|
|
|
|
|
144
|
|
|
if ($currentUserAgent === null) { |
|
145
|
|
|
continue; |
|
146
|
|
|
} |
|
147
|
|
|
|
|
148
|
|
|
$disallowUrl = $this->parseDisallow($line); |
|
149
|
|
|
|
|
150
|
|
|
$currentUserAgent[$disallowUrl] = $disallowUrl; |
|
151
|
|
|
} |
|
152
|
|
|
|
|
153
|
|
|
return $disallowsPerUserAgent; |
|
154
|
|
|
} |
|
155
|
|
|
|
|
156
|
|
|
protected function isCommentLine(string $line): bool |
|
157
|
|
|
{ |
|
158
|
|
|
return strpos(trim($line), '#') === 0; |
|
159
|
|
|
} |
|
160
|
|
|
|
|
161
|
|
|
protected function isUserAgentLine(string $line): bool |
|
162
|
|
|
{ |
|
163
|
|
|
return strpos(trim(strtolower($line)), 'user-agent') === 0; |
|
164
|
|
|
} |
|
165
|
|
|
|
|
166
|
|
|
protected function parseUserAgent(string $line): string |
|
167
|
|
|
{ |
|
168
|
|
|
return trim(str_replace('user-agent', '', strtolower(trim($line))), ': '); |
|
169
|
|
|
} |
|
170
|
|
|
|
|
171
|
|
|
protected function parseDisallow(string $line): string |
|
172
|
|
|
{ |
|
173
|
|
|
return trim(substr_replace(strtolower(trim($line)), '', 0, 8), ': '); |
|
174
|
|
|
} |
|
175
|
|
|
|
|
176
|
|
|
/** |
|
177
|
|
|
* @deprecated |
|
178
|
|
|
*/ |
|
179
|
|
|
protected function concernsDirectory(string $path): bool |
|
180
|
|
|
{ |
|
181
|
|
|
return substr($path, strlen($path) - 1, 1) === '/'; |
|
182
|
|
|
} |
|
183
|
|
|
|
|
184
|
|
|
/** |
|
185
|
|
|
* @deprecated |
|
186
|
|
|
*/ |
|
187
|
|
|
protected function isUrlInDirectory(string $url, string $path): bool |
|
188
|
|
|
{ |
|
189
|
|
|
return strpos($url, $path) === 0; |
|
190
|
|
|
} |
|
191
|
|
|
} |
|
192
|
|
|
|
spatie /
robots-txt
