|
1
|
|
|
<?php |
|
2
|
|
|
namespace vipnytt\RobotsTxtParser\Core; |
|
3
|
|
|
|
|
4
|
|
|
use DateTimeZone; |
|
5
|
|
|
use vipnytt\RobotsTxtParser\Exceptions\ParserException; |
|
6
|
|
|
|
|
7
|
|
|
/** |
|
8
|
|
|
* Trait Toolbox |
|
9
|
|
|
* |
|
10
|
|
|
* @package vipnytt\RobotsTxtParser\Core |
|
11
|
|
|
*/ |
|
12
|
|
|
trait Toolbox |
|
13
|
|
|
{ |
|
14
|
|
|
/** |
|
15
|
|
|
* Check basic rule |
|
16
|
|
|
* |
|
17
|
|
|
* @param string $path |
|
18
|
|
|
* @param array $paths |
|
19
|
|
|
* @return bool |
|
20
|
|
|
*/ |
|
21
|
|
|
protected function checkPath($path, array $paths) |
|
22
|
|
|
{ |
|
23
|
|
|
foreach ($paths as $rule) { |
|
24
|
|
|
$escape = ['?' => '\?', '.' => '\.', '*' => '.*']; |
|
25
|
|
|
foreach ($escape as $search => $replace) { |
|
26
|
|
|
$rule = str_replace($search, $replace, $rule); |
|
27
|
|
|
} |
|
28
|
|
|
/** |
|
29
|
|
|
* Warning: preg_match need to be replaced |
|
30
|
|
|
* |
|
31
|
|
|
* Bug report |
|
32
|
|
|
* @link https://github.com/t1gor/Robots.txt-Core-Class/issues/62 |
|
33
|
|
|
* |
|
34
|
|
|
* An robots.txt parser, where a bug-fix is planned |
|
35
|
|
|
* @link https://github.com/diggin/Diggin_RobotRules |
|
36
|
|
|
* |
|
37
|
|
|
* The solution? |
|
38
|
|
|
* PHP PEG (parsing expression grammar) |
|
39
|
|
|
* @link https://github.com/hafriedlander/php-peg |
|
40
|
|
|
*/ |
|
41
|
|
|
try { |
|
42
|
|
|
if (!preg_match('#' . $rule . '#', $path)) { |
|
43
|
|
|
// Rule does not match |
|
44
|
|
|
continue; |
|
45
|
|
|
} else if (mb_stripos($rule, '$') === false) { |
|
46
|
|
|
// No special parsing required |
|
47
|
|
|
return true; |
|
48
|
|
|
} else if (($wildcardPos = mb_strrpos($rule, '*')) !== false) { |
|
49
|
|
|
// Rule contains both an end anchor ($) and wildcard (*) |
|
50
|
|
|
$afterWildcard = mb_substr($rule, $wildcardPos + 1, mb_strlen($rule) - $wildcardPos - 2); |
|
51
|
|
|
if ($afterWildcard == mb_substr($path, -mb_strlen($afterWildcard))) { |
|
52
|
|
|
return true; |
|
53
|
|
|
} |
|
54
|
|
|
} else if (mb_substr($rule, 0, -1) == $path) { |
|
55
|
|
|
// Rule does contains an end anchor |
|
56
|
|
|
return true; |
|
57
|
|
|
} |
|
58
|
|
|
} catch (\Exception $e) { |
|
59
|
|
|
// An preg_match bug has occurred |
|
60
|
|
|
} |
|
61
|
|
|
} |
|
62
|
|
|
return false; |
|
63
|
|
|
} |
|
64
|
|
|
|
|
65
|
|
|
/** |
|
66
|
|
|
* Generate directive/rule pair |
|
67
|
|
|
* |
|
68
|
|
|
* @param string $line |
|
69
|
|
|
* @param array $whiteList |
|
70
|
|
|
* @return array|false |
|
71
|
|
|
*/ |
|
72
|
|
|
protected function generateRulePair($line, array $whiteList) |
|
73
|
|
|
{ |
|
74
|
|
|
$whiteList = array_map('mb_strtolower', $whiteList); |
|
75
|
|
|
// Split by directive and rule |
|
76
|
|
|
$pair = array_map('trim', mb_split(':', $line, 2)); |
|
77
|
|
|
// Check if the line contains a rule |
|
78
|
|
|
if ( |
|
79
|
|
|
empty($pair[1]) || |
|
80
|
|
|
empty($pair[0]) || |
|
81
|
|
|
!in_array(($pair[0] = mb_strtolower($pair[0])), $whiteList) |
|
82
|
|
|
) { |
|
83
|
|
|
// Line does not contain any supported directive |
|
84
|
|
|
return false; |
|
85
|
|
|
} |
|
86
|
|
|
return [ |
|
87
|
|
|
'directive' => $pair[0], |
|
88
|
|
|
'value' => $pair[1], |
|
89
|
|
|
]; |
|
90
|
|
|
} |
|
91
|
|
|
|
|
92
|
|
|
/** |
|
93
|
|
|
* Validate directive |
|
94
|
|
|
* |
|
95
|
|
|
* @param string $directive |
|
96
|
|
|
* @param array $directives |
|
97
|
|
|
* @return string |
|
98
|
|
|
* @throws ParserException |
|
99
|
|
|
*/ |
|
100
|
|
|
protected function validateDirective($directive, array $directives) |
|
101
|
|
|
{ |
|
102
|
|
|
if (!in_array($directive, $directives, true)) { |
|
103
|
|
|
throw new ParserException('Directive is not allowed here'); |
|
104
|
|
|
} |
|
105
|
|
|
return mb_strtolower($directive); |
|
106
|
|
|
} |
|
107
|
|
|
|
|
108
|
|
|
/** |
|
109
|
|
|
* Client rate as specified in the `Robot exclusion standard` version 2.0 draft |
|
110
|
|
|
* rate = numDocuments / timeUnit |
|
111
|
|
|
* @link http://www.conman.org/people/spc/robots2.html#format.directives.request-rate |
|
112
|
|
|
* |
|
113
|
|
|
* @param $string |
|
114
|
|
|
* @return int|float|false |
|
115
|
|
|
*/ |
|
116
|
|
|
protected function draftParseRate($string) |
|
117
|
|
|
{ |
|
118
|
|
|
$parts = array_map('trim', mb_split('/', $string)); |
|
119
|
|
|
if (count($parts) != 2) { |
|
120
|
|
|
return false; |
|
121
|
|
|
} |
|
122
|
|
|
$multiplier = 1; |
|
123
|
|
|
switch (mb_substr(mb_strtolower(preg_replace('/[^A-Za-z]/', '', $parts[1])), 0, 1)) { |
|
124
|
|
|
case 'm': |
|
125
|
|
|
$multiplier = 60; |
|
126
|
|
|
break; |
|
127
|
|
|
case 'h': |
|
128
|
|
|
$multiplier = 3600; |
|
129
|
|
|
break; |
|
130
|
|
|
case 'd': |
|
131
|
|
|
$multiplier = 86400; |
|
132
|
|
|
break; |
|
133
|
|
|
} |
|
134
|
|
|
$num = floatval(preg_replace('/[^0-9]/', '', $parts[0])); |
|
135
|
|
|
$sec = floatval(preg_replace('/[^0-9.]/', '', $parts[1])) * $multiplier; |
|
136
|
|
|
$rate = $sec / $num; |
|
137
|
|
|
return $rate > 0 ? $rate : false; |
|
138
|
|
|
} |
|
139
|
|
|
|
|
140
|
|
|
/** |
|
141
|
|
|
* Client timestamp range as specified in the `Robot exclusion standard` version 2.0 draft |
|
142
|
|
|
* @link http://www.conman.org/people/spc/robots2.html#format.directives.visit-time |
|
143
|
|
|
* |
|
144
|
|
|
* @param $string |
|
145
|
|
|
* @return array|bool |
|
146
|
|
|
*/ |
|
147
|
|
|
protected function draftParseTime($string) |
|
148
|
|
|
{ |
|
149
|
|
|
$array = preg_replace('/[^0-9]/', '', mb_split('-', $string)); |
|
150
|
|
|
if ( |
|
151
|
|
|
count($array) != 2 || |
|
152
|
|
|
($fromTime = date_create_from_format('Hi', $array[0], new DateTimeZone('UTC'))) === false || |
|
153
|
|
|
($toTime = date_create_from_format('Hi', $array[1], new DateTimeZone('UTC'))) === false |
|
154
|
|
|
) { |
|
155
|
|
|
return false; |
|
156
|
|
|
} |
|
157
|
|
|
return [ |
|
158
|
|
|
'from' => date_format($fromTime, 'Hi'), |
|
159
|
|
|
'to' => date_format($toTime, 'Hi'), |
|
160
|
|
|
]; |
|
161
|
|
|
} |
|
162
|
|
|
} |
|
163
|
|
|
|
VIPnytt /
RobotsTxtParser
