|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
namespace Nabble\SemaltBlocker; |
|
4
|
|
|
|
|
5
|
|
|
/** |
|
6
|
|
|
* The most important class for this package. Basic usage: |
|
7
|
|
|
* ``` |
|
8
|
|
|
* Blocker::protect(); |
|
9
|
|
|
* ```. |
|
10
|
|
|
*/ |
|
11
|
|
|
class Blocker |
|
12
|
|
|
{ |
|
13
|
|
|
const SEPERATOR = ':'; |
|
14
|
|
|
|
|
15
|
|
|
public static $explanation = "Access to this website has been blocked because your referral is set to %s. <a href='%s'>Read why</a>"; |
|
16
|
|
|
|
|
17
|
|
|
private static $blocklist = './../../domains/blocked'; |
|
18
|
|
|
private static $reason = 'Not blocking, no reason given'; |
|
19
|
|
|
|
|
20
|
|
|
////////////////////////////////////////// |
|
21
|
|
|
// PUBLIC API // |
|
22
|
|
|
////////////////////////////////////////// |
|
23
|
|
|
|
|
24
|
|
|
/** |
|
25
|
|
|
* Block a page if referer is found on list of blocked domains. |
|
26
|
|
|
* |
|
27
|
|
|
* @param string $action If empty, send 403 response; if URL, redirect here; if non-empty string, print message |
|
28
|
|
|
*/ |
|
29
|
|
|
public static function protect($action = '') |
|
30
|
|
|
{ |
|
31
|
|
|
// Try to update the list |
|
32
|
|
|
if (!defined('SEMALT_UNIT_TESTING')) { |
|
33
|
|
|
Updater::update(); |
|
34
|
|
|
} |
|
35
|
|
|
|
|
36
|
|
|
// Simply stop here if referer is not on the list |
|
37
|
|
|
if (!self::isRefererOnBlocklist()) { |
|
38
|
|
|
return; |
|
39
|
|
|
} |
|
40
|
|
|
|
|
41
|
|
|
self::doBlock($action); |
|
42
|
|
|
|
|
43
|
|
|
// Stop execution altogether, bye bye bots |
|
44
|
|
|
if (!defined('SEMALT_UNIT_TESTING')) { |
|
45
|
|
|
exit; |
|
46
|
|
|
} |
|
47
|
|
|
} |
|
48
|
|
|
|
|
49
|
|
|
/** |
|
50
|
|
|
* @param bool $verbose Deprecated. Please use the explain() method instead. |
|
51
|
|
|
* |
|
52
|
|
|
* @return bool|string |
|
53
|
|
|
*/ |
|
54
|
|
|
public static function blocked($verbose = false) |
|
55
|
|
|
{ |
|
56
|
|
|
$blocked = self::isRefererOnBlocklist(); |
|
57
|
|
|
if ($verbose === true) { |
|
58
|
|
|
return self::$reason; |
|
59
|
|
|
} |
|
60
|
|
|
|
|
61
|
|
|
return $blocked; |
|
62
|
|
|
} |
|
63
|
|
|
|
|
64
|
|
|
/** |
|
65
|
|
|
* @return string |
|
66
|
|
|
*/ |
|
67
|
|
|
public static function explain() |
|
68
|
|
|
{ |
|
69
|
|
|
return self::$reason; |
|
70
|
|
|
} |
|
71
|
|
|
|
|
72
|
|
|
/** |
|
73
|
|
|
* Send a 403 Forbidden header. |
|
74
|
|
|
*/ |
|
75
|
|
|
public static function forbidden() |
|
76
|
|
|
{ |
|
77
|
|
|
$protocol = (isset($_SERVER['SERVER_PROTOCOL']) ? $_SERVER['SERVER_PROTOCOL'] : 'HTTP/1.0'); |
|
78
|
|
|
header($protocol . ' 403 Forbidden'); |
|
79
|
|
|
} |
|
80
|
|
|
|
|
81
|
|
|
/** |
|
82
|
|
|
* @return array |
|
83
|
|
|
*/ |
|
84
|
|
|
public static function getBlocklist() |
|
85
|
|
|
{ |
|
86
|
|
|
return self::parseBlocklist(self::getBlocklistContents()); |
|
87
|
|
|
} |
|
88
|
|
|
|
|
89
|
|
|
/** |
|
90
|
|
|
* @return string |
|
91
|
|
|
*/ |
|
92
|
|
|
public static function getBlocklistFilename() |
|
93
|
|
|
{ |
|
94
|
|
|
return __DIR__ . DIRECTORY_SEPARATOR . static::$blocklist; |
|
|
|
|
|
|
95
|
|
|
} |
|
96
|
|
|
|
|
97
|
|
|
////////////////////////////////////////// |
|
98
|
|
|
// PRIVATE FUNCTIONS // |
|
99
|
|
|
////////////////////////////////////////// |
|
100
|
|
|
|
|
101
|
|
|
/** |
|
102
|
|
|
* Responsible for sending action output. |
|
103
|
|
|
* |
|
104
|
|
|
* @param string $action |
|
105
|
|
|
*/ |
|
106
|
|
|
private static function doBlock($action = '') |
|
107
|
|
|
{ |
|
108
|
|
|
// Clear buffered output |
|
109
|
|
|
if (!defined('SEMALT_UNIT_TESTING')) { |
|
110
|
|
|
self::cls(); |
|
111
|
|
|
} |
|
112
|
|
|
|
|
113
|
|
|
// Take user defined action |
|
114
|
|
|
self::blockAction($action); |
|
115
|
|
|
|
|
116
|
|
|
// If a human comes by, don't just serve a blank page |
|
117
|
|
|
echo sprintf(self::$explanation, self::getHttpReferer(), 'https://www.google.com/#q=' . urlencode(preg_replace('/https?:\/\//', '', self::getHttpReferer()) . ' referral spam')); |
|
118
|
|
|
} |
|
119
|
|
|
|
|
120
|
|
|
/** |
|
121
|
|
|
* Execute desired action. |
|
122
|
|
|
* |
|
123
|
|
|
* @param string $action |
|
124
|
|
|
*/ |
|
125
|
|
|
private static function blockAction($action = '') |
|
126
|
|
|
{ |
|
127
|
|
|
// Redirect or 403 |
|
128
|
|
|
if (filter_var($action, FILTER_VALIDATE_URL)) { |
|
129
|
|
|
self::redirect($action); |
|
130
|
|
|
} else { |
|
131
|
|
|
self::forbidden(); |
|
132
|
|
|
if (!empty($action)) { |
|
133
|
|
|
echo $action . '<br/>'; |
|
134
|
|
|
} // tell them something nice |
|
135
|
|
|
} |
|
136
|
|
|
} |
|
137
|
|
|
|
|
138
|
|
|
/** |
|
139
|
|
|
* Clear output buffer. |
|
140
|
|
|
*/ |
|
141
|
|
|
private static function cls() |
|
142
|
|
|
{ |
|
143
|
|
|
while (ob_get_level()) { |
|
144
|
|
|
ob_end_clean(); |
|
145
|
|
|
} |
|
146
|
|
|
} |
|
147
|
|
|
|
|
148
|
|
|
/** |
|
149
|
|
|
* Redirect to a url by sending the appropriate header. |
|
150
|
|
|
* |
|
151
|
|
|
* @param string $url |
|
152
|
|
|
*/ |
|
153
|
|
|
private static function redirect($url) |
|
154
|
|
|
{ |
|
155
|
|
|
header('Location: ' . $url); |
|
156
|
|
|
} |
|
157
|
|
|
|
|
158
|
|
|
/** |
|
159
|
|
|
* The public use of this function is undocumented. |
|
160
|
|
|
* |
|
161
|
|
|
* @return bool |
|
162
|
|
|
*/ |
|
163
|
|
|
public static function isRefererOnBlocklist() |
|
164
|
|
|
{ |
|
165
|
|
|
$referer = self::getHttpReferer(); |
|
166
|
|
|
if ($referer === null) { |
|
167
|
|
|
self::$reason = 'Not blocking because referer header is not set or empty'; |
|
168
|
|
|
|
|
169
|
|
|
return false; |
|
170
|
|
|
} |
|
171
|
|
|
|
|
172
|
|
|
return self::isUrlOnBlocklist($referer, 'referer'); |
|
173
|
|
|
} |
|
174
|
|
|
|
|
175
|
|
|
/** |
|
176
|
|
|
* The public use of this function is undocumented. |
|
177
|
|
|
* |
|
178
|
|
|
* @param string $url |
|
179
|
|
|
* @param string $entity |
|
180
|
|
|
* |
|
181
|
|
|
* @return bool |
|
182
|
|
|
*/ |
|
183
|
|
|
public static function isUrlOnBlocklist($url, $entity = 'url') |
|
184
|
|
|
{ |
|
185
|
|
|
$rootDomain = Domainparser::getRootDomain($url); |
|
186
|
|
|
if ($rootDomain === false) { |
|
187
|
|
|
self::$reason = "Not blocking because we couldn't parse root domain"; |
|
188
|
|
|
|
|
189
|
|
|
return false; |
|
190
|
|
|
} |
|
191
|
|
|
|
|
192
|
|
|
$blocklist = self::getConcatenateBlocklist(); |
|
193
|
|
View Code Duplication |
if (substr_count($blocklist, self::SEPERATOR . $rootDomain . self::SEPERATOR)) { |
|
|
|
|
|
|
194
|
|
|
self::$reason = 'Blocking because ' . $entity . ' root domain (' . $rootDomain . ') is found on blocklist'; |
|
195
|
|
|
|
|
196
|
|
|
return true; |
|
197
|
|
|
} |
|
198
|
|
|
|
|
199
|
|
|
$hostname = Domainparser::getHostname($url); |
|
200
|
|
View Code Duplication |
if (substr_count($blocklist, self::SEPERATOR . $hostname . self::SEPERATOR)) { |
|
|
|
|
|
|
201
|
|
|
self::$reason = 'Blocking because ' . $entity . ' hostname (' . $hostname . ') is found on blocklist'; |
|
202
|
|
|
|
|
203
|
|
|
return true; |
|
204
|
|
|
} |
|
205
|
|
|
|
|
206
|
|
|
$path = Domainparser::getPath($url); |
|
207
|
|
|
if (trim($path, '/')) { |
|
208
|
|
View Code Duplication |
if (substr_count($blocklist, self::SEPERATOR . $rootDomain . $path . self::SEPERATOR)) { |
|
|
|
|
|
|
209
|
|
|
self::$reason = 'Blocking because ' . $entity . ' root domain/path (' . $rootDomain . $path . ') is found on blocklist'; |
|
210
|
|
|
|
|
211
|
|
|
return true; |
|
212
|
|
|
} |
|
213
|
|
View Code Duplication |
if (substr_count($blocklist, self::SEPERATOR . $hostname . $path . self::SEPERATOR)) { |
|
|
|
|
|
|
214
|
|
|
self::$reason = 'Blocking because ' . $entity . ' hostname/path (' . $hostname . $path . ') is found on blocklist'; |
|
215
|
|
|
|
|
216
|
|
|
return true; |
|
217
|
|
|
} |
|
218
|
|
|
} |
|
219
|
|
|
|
|
220
|
|
|
self::$reason = 'Not blocking because ' . $entity . ' (' . $url . ') is not matched against blocklist'; |
|
221
|
|
|
|
|
222
|
|
|
return false; |
|
223
|
|
|
} |
|
224
|
|
|
|
|
225
|
|
|
/** |
|
226
|
|
|
* Returns HTTP Referer if it is available and not empty, null otherwise. |
|
227
|
|
|
* |
|
228
|
|
|
* @return string|null |
|
229
|
|
|
*/ |
|
230
|
|
|
private static function getHttpReferer() |
|
231
|
|
|
{ |
|
232
|
|
|
if (isset($_SERVER['HTTP_REFERER']) && !empty($_SERVER['HTTP_REFERER'])) { |
|
233
|
|
|
return $_SERVER['HTTP_REFERER']; |
|
234
|
|
|
} |
|
235
|
|
|
} |
|
236
|
|
|
|
|
237
|
|
|
/** |
|
238
|
|
|
* @return string |
|
239
|
|
|
*/ |
|
240
|
|
|
private static function getBlocklistContents() |
|
241
|
|
|
{ |
|
242
|
|
|
$blocklistContent = file_get_contents(self::getBlocklistFilename()); |
|
243
|
|
|
|
|
244
|
|
|
return $blocklistContent; |
|
245
|
|
|
} |
|
246
|
|
|
|
|
247
|
|
|
/** |
|
248
|
|
|
* @return string |
|
249
|
|
|
*/ |
|
250
|
|
|
private static function getConcatenateBlocklist() |
|
251
|
|
|
{ |
|
252
|
|
|
return self::concatenateBlocklist(self::getBlocklistContents()); |
|
253
|
|
|
} |
|
254
|
|
|
|
|
255
|
|
|
/** |
|
256
|
|
|
* @param string $blocklistContent |
|
257
|
|
|
* |
|
258
|
|
|
* @return array |
|
259
|
|
|
*/ |
|
260
|
|
|
private static function parseBlocklist($blocklistContent) |
|
261
|
|
|
{ |
|
262
|
|
|
return array_map('trim', array_filter(explode(PHP_EOL, strtolower($blocklistContent)))); |
|
263
|
|
|
} |
|
264
|
|
|
|
|
265
|
|
|
/** |
|
266
|
|
|
* @param string $blocklistContent |
|
267
|
|
|
* |
|
268
|
|
|
* @return string |
|
269
|
|
|
*/ |
|
270
|
|
|
private static function concatenateBlocklist($blocklistContent) |
|
271
|
|
|
{ |
|
272
|
|
|
return self::SEPERATOR . str_replace(PHP_EOL, self::SEPERATOR, strtolower($blocklistContent)) . self::SEPERATOR; |
|
273
|
|
|
} |
|
274
|
|
|
} |
|
275
|
|
|
|
Let’s assume you have a class which uses late-static binding:
The code above will run fine in your PHP runtime. However, if you now create a sub-class and call the
getSomeVariable()on that sub-class, you will receive a runtime error:In the case above, it makes sense to update
SomeClassto useselfinstead: