|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
declare(strict_types=1); |
|
4
|
|
|
|
|
5
|
|
|
namespace PHPHtmlParser\Dom; |
|
6
|
|
|
|
|
7
|
|
|
use PHPHtmlParser\Contracts\Dom\CleanerInterface; |
|
8
|
|
|
use PHPHtmlParser\Exceptions\LogicalException; |
|
9
|
|
|
use PHPHtmlParser\Options; |
|
10
|
|
|
|
|
11
|
|
|
class Cleaner implements CleanerInterface |
|
12
|
|
|
{ |
|
13
|
|
|
/** |
|
14
|
|
|
* Cleans the html of any none-html information. |
|
15
|
|
|
* |
|
16
|
|
|
* @throws LogicalException |
|
17
|
|
|
*/ |
|
18
|
297 |
|
public function clean(string $str, Options $options, string $defaultCharset): string |
|
19
|
|
|
{ |
|
20
|
297 |
|
if (!$options->isCleanupInput()) { |
|
21
|
|
|
// skip entire cleanup step |
|
22
|
15 |
|
return $str; |
|
23
|
|
|
} |
|
24
|
|
|
|
|
25
|
|
|
// check if the string is gziped |
|
26
|
282 |
|
$is_gzip = 0 === \mb_strpos($str, "\x1f" . "\x8b" . "\x08", 0, 'US-ASCII'); |
|
27
|
282 |
|
if ($is_gzip) { |
|
28
|
|
|
$str = \gzdecode($str); |
|
29
|
|
|
if ($str === false) { |
|
30
|
|
|
throw new LogicalException('gzdecode returned false. Error when trying to decode the string.'); |
|
31
|
|
|
} |
|
32
|
|
|
} |
|
33
|
|
|
|
|
34
|
|
|
// we must handle character encoding |
|
35
|
282 |
|
$str = $this->setUpRegexEncoding($str, $options, $defaultCharset); |
|
36
|
|
|
|
|
37
|
|
|
// remove white space before closing tags |
|
38
|
282 |
|
$str = \mb_eregi_replace("'\s+>", "'>", $str); |
|
39
|
282 |
|
if ($str === false) { |
|
40
|
|
|
throw new LogicalException('mb_eregi_replace returned false instead of a string. Error when attempting to clean single quotes.'); |
|
41
|
|
|
} |
|
42
|
282 |
|
$str = \mb_eregi_replace('"\s+>', '">', $str); |
|
43
|
282 |
|
if ($str === false) { |
|
44
|
|
|
throw new LogicalException('mb_eregi_replace returned false instead of a string. Error when attempting to clean double quotes.'); |
|
45
|
|
|
} |
|
46
|
|
|
|
|
47
|
|
|
// clean out the \n\r |
|
48
|
282 |
|
$replace = ' '; |
|
49
|
282 |
|
if ($options->isPreserveLineBreaks()) { |
|
50
|
3 |
|
$replace = ' '; |
|
51
|
|
|
} |
|
52
|
282 |
|
$str = \str_replace(["\r\n", "\r", "\n"], $replace, $str); |
|
53
|
282 |
|
if ($str === false) { |
|
54
|
|
|
throw new LogicalException('str_replace returned false instead of a string. Error when attempting to clean input string.'); |
|
55
|
|
|
} |
|
56
|
|
|
|
|
57
|
|
|
// strip the doctype |
|
58
|
282 |
|
$str = \mb_eregi_replace('<!doctype(.*?)>', '', $str); |
|
59
|
282 |
|
if ($str === false) { |
|
60
|
|
|
throw new LogicalException('mb_eregi_replace returned false instead of a string. Error when attempting to strip the doctype.'); |
|
61
|
|
|
} |
|
62
|
|
|
|
|
63
|
|
|
// strip out comments |
|
64
|
282 |
|
$str = \mb_eregi_replace('<!--(.*?)-->', '', $str); |
|
65
|
282 |
|
if ($str === false) { |
|
66
|
|
|
throw new LogicalException('mb_eregi_replace returned false instead of a string. Error when attempting to strip comments.'); |
|
67
|
|
|
} |
|
68
|
|
|
|
|
69
|
|
|
// strip out cdata |
|
70
|
282 |
|
$str = \mb_eregi_replace("<!\[CDATA\[(.*?)\]\]>", '', $str); |
|
71
|
282 |
|
if ($str === false) { |
|
72
|
|
|
throw new LogicalException('mb_eregi_replace returned false instead of a string. Error when attempting to strip out cdata.'); |
|
73
|
|
|
} |
|
74
|
|
|
|
|
75
|
|
|
// strip out <script> tags |
|
76
|
282 |
|
if ($options->isRemoveScripts()) { |
|
77
|
279 |
|
$str = \mb_eregi_replace("<\s*script[^>]*[^/]>(.*?)<\s*/\s*script\s*>", '', $str); |
|
78
|
279 |
|
if ($str === false) { |
|
79
|
|
|
throw new LogicalException('mb_eregi_replace returned false instead of a string. Error when attempting to remove scripts 1.'); |
|
80
|
|
|
} |
|
81
|
279 |
|
$str = \mb_eregi_replace("<\s*script\s*>(.*?)<\s*/\s*script\s*>", '', $str); |
|
82
|
279 |
|
if ($str === false) { |
|
83
|
|
|
throw new LogicalException('mb_eregi_replace returned false instead of a string. Error when attempting to remove scripts 2.'); |
|
84
|
|
|
} |
|
85
|
|
|
} |
|
86
|
|
|
|
|
87
|
|
|
// strip out <style> tags |
|
88
|
282 |
|
if ($options->isRemoveStyles()) { |
|
89
|
279 |
|
$str = \mb_eregi_replace("<\s*style[^>]*[^/]>(.*?)<\s*/\s*style\s*>", '', $str); |
|
90
|
279 |
|
if ($str === false) { |
|
91
|
|
|
throw new LogicalException('mb_eregi_replace returned false instead of a string. Error when attempting to strip out style tags 1.'); |
|
92
|
|
|
} |
|
93
|
279 |
|
$str = \mb_eregi_replace("<\s*style\s*>(.*?)<\s*/\s*style\s*>", '', $str); |
|
94
|
279 |
|
if ($str === false) { |
|
95
|
|
|
throw new LogicalException('mb_eregi_replace returned false instead of a string. Error when attempting to strip out style tags 2.'); |
|
96
|
|
|
} |
|
97
|
|
|
} |
|
98
|
|
|
|
|
99
|
|
|
// strip smarty scripts |
|
100
|
282 |
|
if ($options->isRemoveSmartyScripts()) { |
|
101
|
279 |
|
$str = \mb_eregi_replace("(\{\w)(.*?)(\})", '', $str); |
|
102
|
279 |
|
if ($str === false) { |
|
103
|
|
|
throw new LogicalException('mb_eregi_replace returned false instead of a string. Error when attempting to remove smarty scripts.'); |
|
104
|
|
|
} |
|
105
|
|
|
} |
|
106
|
|
|
|
|
107
|
282 |
|
return $str; |
|
108
|
|
|
} |
|
109
|
|
|
|
|
110
|
|
|
/** |
|
111
|
|
|
* Sets up the mb_regex_encoding and converts the text to that encoding. |
|
112
|
|
|
* |
|
113
|
|
|
* @throws LogicalException |
|
114
|
|
|
*/ |
|
115
|
282 |
|
private function setUpRegexEncoding(string $str, Options $options, string $defaultCharset): string |
|
116
|
|
|
{ |
|
117
|
282 |
|
$encoding = $defaultCharset; |
|
118
|
282 |
|
$enforceEncoding = $options->getEnforceEncoding(); |
|
119
|
282 |
|
if ($enforceEncoding !== null) { |
|
120
|
|
|
// they want to enforce the given encoding |
|
121
|
|
|
$encoding = $enforceEncoding; |
|
122
|
|
|
} |
|
123
|
|
|
|
|
124
|
282 |
|
if (!\mb_regex_encoding($encoding)) { |
|
125
|
|
|
throw new LogicalException('Character encoding was not able to be changed to ' . $encoding . '.'); |
|
126
|
|
|
} |
|
127
|
|
|
|
|
128
|
282 |
|
return \mb_convert_encoding($str, $encoding); |
|
129
|
|
|
} |
|
130
|
|
|
} |
|
131
|
|
|
|
paquettg /
php-html-parser
