1
|
|
|
<?php |
2
|
|
|
namespace Redaxscript\Html; |
3
|
|
|
|
4
|
|
|
use DOMDocument; |
5
|
|
|
use DOMNode; |
6
|
|
|
use Redaxscript\Model; |
7
|
|
|
use function in_array; |
8
|
|
|
use function is_object; |
9
|
|
|
use function is_string; |
10
|
|
|
use function mb_convert_encoding; |
11
|
|
|
use function range; |
12
|
|
|
use function strlen; |
13
|
|
|
use function trim; |
14
|
|
|
|
15
|
|
|
/** |
16
|
|
|
* parent class to purify html against xss |
17
|
|
|
* |
18
|
|
|
* @since 3.0.0 |
19
|
|
|
* |
20
|
|
|
* @package Redaxscript |
21
|
|
|
* @category Html |
22
|
|
|
* @author Henry Ruhs |
23
|
|
|
*/ |
24
|
|
|
|
25
|
|
|
class Purifier |
26
|
|
|
{ |
27
|
|
|
/** |
28
|
|
|
* array of allowed tags |
29
|
|
|
* |
30
|
|
|
* @var array |
31
|
|
|
*/ |
32
|
|
|
|
33
|
|
|
protected $_allowedTags = |
34
|
|
|
[ |
35
|
|
|
'#text', |
36
|
|
|
'br', |
37
|
|
|
'caption', |
38
|
|
|
'div', |
39
|
|
|
'dd', |
40
|
|
|
'dl', |
41
|
|
|
'dt', |
42
|
|
|
'em', |
43
|
|
|
'h1', |
44
|
|
|
'h2', |
45
|
|
|
'h3', |
46
|
|
|
'h4', |
47
|
|
|
'h5', |
48
|
|
|
'h6', |
49
|
|
|
'li', |
50
|
|
|
'p', |
51
|
|
|
'pre', |
52
|
|
|
'ol', |
53
|
|
|
'span', |
54
|
|
|
'strike', |
55
|
|
|
'strong', |
56
|
|
|
'sub', |
57
|
|
|
'sup', |
58
|
|
|
'table', |
59
|
|
|
'tbody', |
60
|
|
|
'tfoot', |
61
|
|
|
'thead', |
62
|
|
|
'td', |
63
|
|
|
'th', |
64
|
|
|
'tr', |
65
|
|
|
'strong', |
66
|
|
|
'u', |
67
|
|
|
'ul', |
68
|
|
|
'wbr' |
69
|
|
|
]; |
70
|
|
|
|
71
|
|
|
/** |
72
|
|
|
* purify the html |
73
|
|
|
* |
74
|
|
|
* @since 3.0.0 |
75
|
|
|
* |
76
|
|
|
* @param string $html html to be purified |
77
|
|
|
* @param bool $filter optional filter |
78
|
|
|
* |
79
|
|
|
* @return string|null |
80
|
|
|
*/ |
81
|
|
|
|
82
|
11 |
|
public function purify(string $html = null, bool $filter = true) : ?string |
83
|
|
|
{ |
84
|
11 |
|
$settingModel = new Model\Setting(); |
85
|
11 |
|
$charset = $settingModel->get('charset'); |
86
|
11 |
|
$html = mb_convert_encoding($html, 'html-entities', $charset); |
87
|
|
|
|
88
|
|
|
/* filter html */ |
89
|
|
|
|
90
|
11 |
|
if ($filter && strlen($html)) |
91
|
|
|
{ |
92
|
11 |
|
$html = $this->_process($html); |
93
|
|
|
} |
94
|
11 |
|
return $html; |
95
|
|
|
} |
96
|
|
|
|
97
|
|
|
/** |
98
|
|
|
* process the html |
99
|
|
|
* |
100
|
|
|
* @since 3.0.0 |
101
|
|
|
* |
102
|
|
|
* @param string|object $html html to be processed |
103
|
|
|
* |
104
|
|
|
* @return string|null |
105
|
|
|
*/ |
106
|
|
|
|
107
|
11 |
|
protected function _process($html = null) : ?string |
108
|
|
|
{ |
109
|
11 |
|
if (is_object($html)) |
110
|
|
|
{ |
111
|
|
|
/* process children */ |
112
|
|
|
|
113
|
11 |
|
if ($html->hasChildNodes()) |
114
|
|
|
{ |
115
|
5 |
|
$range = range($html->childNodes->length - 1, 0); |
116
|
5 |
|
foreach ($range as $i) |
117
|
|
|
{ |
118
|
5 |
|
$this->_process($html->childNodes->item($i)); |
119
|
|
|
} |
120
|
|
|
} |
121
|
|
|
|
122
|
|
|
/* strip tags and attributes */ |
123
|
|
|
|
124
|
11 |
|
$this->_stripTags($html); |
125
|
11 |
|
$this->_stripAttributes($html); |
126
|
|
|
} |
127
|
11 |
|
else if (is_string($html)) |
128
|
|
|
{ |
129
|
11 |
|
$doc = $this->_createDocument($html); |
130
|
11 |
|
$this->_process($doc->documentElement); |
131
|
11 |
|
return trim($doc->saveHTML()); |
132
|
|
|
} |
133
|
11 |
|
return null; |
134
|
|
|
} |
135
|
|
|
|
136
|
|
|
/** |
137
|
|
|
* strip the tags |
138
|
|
|
* |
139
|
|
|
* @since 3.0.0 |
140
|
|
|
* |
141
|
|
|
* @param DOMNode $node document node |
142
|
|
|
*/ |
143
|
|
|
|
144
|
11 |
|
protected function _stripTags(DOMNode $node = null) : void |
145
|
|
|
{ |
146
|
11 |
|
if (!in_array($node->nodeName, $this->_allowedTags)) |
147
|
|
|
{ |
148
|
10 |
|
$fragment = $this->_createFragment($node); |
149
|
10 |
|
$node->parentNode->replaceChild($fragment, $node); |
150
|
|
|
} |
151
|
11 |
|
} |
152
|
|
|
|
153
|
|
|
/** |
154
|
|
|
* strip the attributes |
155
|
|
|
* |
156
|
|
|
* @since 3.0.0 |
157
|
|
|
* |
158
|
|
|
* @param DOMNode $node document node |
159
|
|
|
*/ |
160
|
|
|
|
161
|
11 |
|
protected function _stripAttributes(DOMNode $node = null) : void |
162
|
|
|
{ |
163
|
11 |
|
while ($node->hasAttributes()) |
|
|
|
|
164
|
|
|
{ |
165
|
10 |
|
$node->removeAttributeNode($node->attributes->item(0)); |
166
|
|
|
} |
167
|
11 |
|
} |
168
|
|
|
|
169
|
|
|
/** |
170
|
|
|
* create the document |
171
|
|
|
* |
172
|
|
|
* @since 2.4.0 |
173
|
|
|
* |
174
|
|
|
* @param string $html html to be loaded |
175
|
|
|
* |
176
|
|
|
* @return DOMDocument |
177
|
|
|
*/ |
178
|
|
|
|
179
|
11 |
|
protected function _createDocument(string $html = null) : DOMDocument |
180
|
|
|
{ |
181
|
11 |
|
$doc = new DOMDocument(); |
182
|
11 |
|
$doc->loadHTML($html, LIBXML_HTML_NODEFDTD | LIBXML_HTML_NOIMPLIED); |
183
|
11 |
|
return $doc; |
184
|
|
|
} |
185
|
|
|
|
186
|
|
|
/** |
187
|
|
|
* create the fragment |
188
|
|
|
* |
189
|
|
|
* @since 3.0.0 |
190
|
|
|
* |
191
|
|
|
* @param DOMNode $node document node |
192
|
|
|
* |
193
|
|
|
* @return DOMNode |
194
|
|
|
*/ |
195
|
|
|
|
196
|
10 |
|
protected function _createFragment(DOMNode $node = null) : DOMNode |
197
|
|
|
{ |
198
|
10 |
|
$fragment = $node->ownerDocument->createDocumentFragment(); |
199
|
|
|
|
200
|
|
|
/* process nodes */ |
201
|
|
|
|
202
|
10 |
|
while ($node->childNodes->length) |
203
|
|
|
{ |
204
|
1 |
|
$fragment->appendChild($node->firstChild); |
205
|
|
|
} |
206
|
10 |
|
return $fragment; |
207
|
|
|
} |
208
|
|
|
} |
209
|
|
|
|
If a variable is not always an object, we recommend to add an additional type check to ensure your method call is safe: