|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
/** |
|
4
|
|
|
* @package s9e\TextFormatter |
|
5
|
|
|
* @copyright Copyright (c) 2010-2017 The s9e Authors |
|
6
|
|
|
* @license http://www.opensource.org/licenses/mit-license.php The MIT License |
|
7
|
|
|
*/ |
|
8
|
|
|
namespace s9e\TextFormatter\Plugins\Censor; |
|
9
|
|
|
|
|
10
|
|
|
class Helper |
|
11
|
|
|
{ |
|
12
|
|
|
/** |
|
13
|
|
|
* @var string Regexp matching whitelisted words |
|
14
|
|
|
*/ |
|
15
|
|
|
public $allowed; |
|
16
|
|
|
|
|
17
|
|
|
/** |
|
18
|
|
|
* @var string Name of attribute used for the replacement |
|
19
|
|
|
*/ |
|
20
|
|
|
public $attrName = 'with'; |
|
21
|
|
|
|
|
22
|
|
|
/** |
|
23
|
|
|
* @var string Default string used to replace censored words |
|
24
|
|
|
*/ |
|
25
|
|
|
public $defaultReplacement = '****'; |
|
26
|
|
|
|
|
27
|
|
|
/** |
|
28
|
|
|
* @var string Regexp matching blacklisted words in plain text |
|
29
|
|
|
*/ |
|
30
|
|
|
public $regexp = '/(?!)/'; |
|
31
|
|
|
|
|
32
|
|
|
/** |
|
33
|
|
|
* @var string Regexp matching blacklisted words in HTML |
|
34
|
|
|
*/ |
|
35
|
|
|
public $regexpHtml = '/(?!)/'; |
|
36
|
|
|
|
|
37
|
|
|
/** |
|
38
|
|
|
* @var array Array of [regexp => replacement] |
|
39
|
|
|
*/ |
|
40
|
|
|
public $replacements = []; |
|
41
|
|
|
|
|
42
|
|
|
/** |
|
43
|
|
|
* @var string Name of the tag used to mark censored words |
|
44
|
|
|
*/ |
|
45
|
|
|
public $tagName = 'CENSOR'; |
|
46
|
|
|
|
|
47
|
|
|
/** |
|
48
|
|
|
* Constructor |
|
49
|
|
|
* |
|
50
|
|
|
* @param array $config Helper's config |
|
51
|
|
|
*/ |
|
52
|
20 |
|
public function __construct(array $config) |
|
53
|
|
|
{ |
|
54
|
20 |
|
foreach ($config as $k => $v) |
|
55
|
|
|
{ |
|
56
|
20 |
|
$this->$k = $v; |
|
57
|
20 |
|
} |
|
58
|
20 |
|
} |
|
59
|
|
|
|
|
60
|
|
|
/** |
|
61
|
|
|
* Censor text nodes inside of HTML code |
|
62
|
|
|
* |
|
63
|
|
|
* NOTE: will only recognize attributes that are enclosed in double quotes |
|
64
|
|
|
* |
|
65
|
|
|
* @param string $html Original HTML |
|
66
|
|
|
* @param bool $censorAttributes Whether to censor the content of attributes |
|
67
|
|
|
* @return string Censored HTML |
|
68
|
|
|
*/ |
|
69
|
14 |
|
public function censorHtml($html, $censorAttributes = false) |
|
70
|
|
|
{ |
|
71
|
14 |
|
$attributesExpr = ''; |
|
72
|
|
|
if ($censorAttributes) |
|
73
|
14 |
|
{ |
|
74
|
2 |
|
$attributesExpr = '|[^<">]*+(?=<|$|"(?> [-\\w]+="[^"]*+")*+\\/?>)'; |
|
75
|
2 |
|
} |
|
76
|
|
|
|
|
77
|
|
|
// Modify the original regexp so that it only matches text nodes and optionally attribute |
|
78
|
|
|
// values |
|
79
|
14 |
|
$delim = $this->regexpHtml[0]; |
|
80
|
14 |
|
$pos = strrpos($this->regexpHtml, $delim); |
|
81
|
|
|
$regexp = $delim |
|
82
|
|
|
. '(?<!&#)(?<!&)' |
|
83
|
14 |
|
. substr($this->regexpHtml, 1, $pos - 1) |
|
84
|
14 |
|
. '(?=[^<>]*+(?=<|$)' . $attributesExpr . ')' |
|
85
|
14 |
|
. substr($this->regexpHtml, $pos); |
|
86
|
|
|
|
|
87
|
14 |
|
return preg_replace_callback( |
|
88
|
14 |
|
$regexp, |
|
89
|
|
|
function ($m) |
|
90
|
|
|
{ |
|
91
|
13 |
|
return htmlspecialchars($this->getReplacement($m[0]), ENT_QUOTES); |
|
92
|
14 |
|
}, |
|
93
|
|
|
$html |
|
94
|
14 |
|
); |
|
95
|
|
|
} |
|
96
|
|
|
|
|
97
|
|
|
/** |
|
98
|
|
|
* Censor given plain text |
|
99
|
|
|
* |
|
100
|
|
|
* @param string $text Original text |
|
101
|
|
|
* @return string Censored text |
|
102
|
|
|
*/ |
|
103
|
3 |
|
public function censorText($text) |
|
104
|
|
|
{ |
|
105
|
3 |
|
return preg_replace_callback( |
|
106
|
3 |
|
$this->regexp, |
|
107
|
3 |
|
function ($m) |
|
108
|
|
|
{ |
|
109
|
3 |
|
return $this->getReplacement($m[0]); |
|
110
|
3 |
|
}, |
|
111
|
|
|
$text |
|
112
|
3 |
|
); |
|
113
|
|
|
} |
|
114
|
|
|
|
|
115
|
|
|
/** |
|
116
|
|
|
* Test whether given word is censored |
|
117
|
|
|
* |
|
118
|
|
|
* @param string $word |
|
119
|
|
|
* @return bool |
|
120
|
|
|
*/ |
|
121
|
3 |
|
public function isCensored($word) |
|
122
|
|
|
{ |
|
123
|
3 |
|
return (preg_match($this->regexp, $word) && !$this->isAllowed($word)); |
|
124
|
|
|
} |
|
125
|
|
|
|
|
126
|
|
|
/** |
|
127
|
|
|
* Build and return the censor tag that matches given word |
|
128
|
|
|
* |
|
129
|
|
|
* @param string $word Word to censor |
|
130
|
|
|
* @return string Censor tag, complete with its replacement attribute |
|
131
|
|
|
*/ |
|
132
|
|
|
protected function buildTag($word) |
|
133
|
|
|
{ |
|
134
|
|
|
$startTag = '<' . $this->tagName; |
|
135
|
|
|
$replacement = $this->getReplacement($word); |
|
136
|
|
|
if ($replacement !== $this->defaultReplacement) |
|
137
|
|
|
{ |
|
138
|
|
|
$startTag .= ' ' . $this->attrName . '="' . htmlspecialchars($replacement, ENT_COMPAT) . '"'; |
|
139
|
|
|
} |
|
140
|
|
|
|
|
141
|
|
|
return $startTag . '>' . $word . '</' . $this->tagName . '>'; |
|
142
|
|
|
} |
|
143
|
|
|
|
|
144
|
|
|
/** |
|
145
|
|
|
* Get the replacement for given word |
|
146
|
|
|
* |
|
147
|
|
|
* @param string $word Original word |
|
148
|
|
|
* @return string Replacement if the word is censored, or the original word otherwise |
|
149
|
|
|
*/ |
|
150
|
16 |
|
protected function getReplacement($word) |
|
151
|
|
|
{ |
|
152
|
16 |
|
if ($this->isAllowed($word)) |
|
153
|
16 |
|
{ |
|
154
|
2 |
|
return $word; |
|
155
|
|
|
} |
|
156
|
|
|
|
|
157
|
16 |
|
foreach ($this->replacements as list($regexp, $replacement)) |
|
158
|
|
|
{ |
|
159
|
5 |
|
if (preg_match($regexp, $word)) |
|
160
|
5 |
|
{ |
|
161
|
5 |
|
return $replacement; |
|
162
|
|
|
} |
|
163
|
11 |
|
} |
|
164
|
|
|
|
|
165
|
11 |
|
return $this->defaultReplacement; |
|
166
|
|
|
} |
|
167
|
|
|
|
|
168
|
|
|
/** |
|
169
|
|
|
* Test whether given word is allowed (whitelisted) |
|
170
|
|
|
* |
|
171
|
|
|
* @param string $word |
|
172
|
|
|
* @return bool |
|
173
|
|
|
*/ |
|
174
|
18 |
|
protected function isAllowed($word) |
|
175
|
|
|
{ |
|
176
|
18 |
|
return (isset($this->allowed) && preg_match($this->allowed, $word)); |
|
177
|
|
|
} |
|
178
|
|
|
} |
s9e /
TextFormatter
