|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
namespace PhpOffice\PhpSpreadsheet\Reader\Security; |
|
4
|
|
|
|
|
5
|
|
|
use PhpOffice\PhpSpreadsheet\Reader\Exception; |
|
6
|
|
|
|
|
7
|
|
|
class XmlScanner |
|
8
|
|
|
{ |
|
9
|
|
|
/** |
|
10
|
|
|
* Identifies whether the thread-safe libxmlDisableEntityLoader() function is available. |
|
11
|
|
|
* |
|
12
|
|
|
* @var bool |
|
13
|
|
|
*/ |
|
14
|
|
|
private $libxmlDisableEntityLoader = false; |
|
15
|
|
|
|
|
16
|
|
|
private $pattern; |
|
17
|
|
|
|
|
18
|
96 |
|
public function __construct($pattern = '<!DOCTYPE') |
|
19
|
|
|
{ |
|
20
|
96 |
|
$this->pattern = $pattern; |
|
21
|
96 |
|
$this->libxmlDisableEntityLoader = $this->identifyLibxmlDisableEntityLoaderAvailability(); |
|
22
|
|
|
|
|
23
|
96 |
|
if ($this->libxmlDisableEntityLoader) { |
|
24
|
96 |
|
libxml_disable_entity_loader(true); |
|
25
|
|
|
} |
|
26
|
96 |
|
} |
|
27
|
|
|
|
|
28
|
96 |
|
private function identifyLibxmlDisableEntityLoaderAvailability() |
|
29
|
|
|
{ |
|
30
|
96 |
|
if (PHP_MAJOR_VERSION == 7) { |
|
31
|
96 |
|
switch (PHP_MINOR_VERSION) { |
|
32
|
96 |
|
case 2: |
|
33
|
96 |
|
return PHP_RELEASE_VERSION >= 1; |
|
34
|
|
|
case 1: |
|
35
|
|
|
return PHP_RELEASE_VERSION >= 13; |
|
36
|
|
|
case 0: |
|
37
|
|
|
return PHP_RELEASE_VERSION >= 27; |
|
38
|
|
|
} |
|
39
|
|
|
|
|
40
|
|
|
return true; |
|
41
|
|
|
} |
|
42
|
|
|
|
|
43
|
|
|
return false; |
|
44
|
|
|
} |
|
45
|
|
|
|
|
46
|
|
|
/** |
|
47
|
|
|
* Scan the XML for use of <!ENTITY to prevent XXE/XEE attacks. |
|
48
|
|
|
* |
|
49
|
|
|
* @param mixed $xml |
|
50
|
|
|
* |
|
51
|
|
|
* @throws Exception |
|
52
|
|
|
* |
|
53
|
|
|
* @return string |
|
54
|
|
|
*/ |
|
55
|
77 |
|
public function scan($xml) |
|
56
|
|
|
{ |
|
57
|
77 |
|
$pattern = '/encoding="(.*?)"/'; |
|
58
|
77 |
|
$result = preg_match($pattern, $xml, $matches); |
|
59
|
77 |
|
$charset = $result ? $matches[1] : 'UTF-8'; |
|
60
|
|
|
|
|
61
|
77 |
|
if ($charset !== 'UTF-8') { |
|
62
|
1 |
|
$xml = mb_convert_encoding($xml, 'UTF-8', $charset); |
|
63
|
|
|
} |
|
64
|
|
|
|
|
65
|
|
|
// Don't rely purely on libxml_disable_entity_loader() |
|
66
|
77 |
|
$pattern = '/\\0?' . implode('\\0?', str_split($this->pattern)) . '\\0?/'; |
|
67
|
77 |
|
if (preg_match($pattern, $xml)) { |
|
68
|
5 |
|
throw new Exception('Detected use of ENTITY in XML, spreadsheet file load() aborted to prevent XXE/XEE attacks'); |
|
69
|
|
|
} |
|
70
|
|
|
|
|
71
|
72 |
|
return $xml; |
|
72
|
|
|
} |
|
73
|
|
|
|
|
74
|
|
|
/** |
|
75
|
|
|
* Scan theXML for use of <!ENTITY to prevent XXE/XEE attacks. |
|
76
|
|
|
* |
|
77
|
|
|
* @param string $filestream |
|
78
|
|
|
* |
|
79
|
|
|
* @throws Exception |
|
80
|
|
|
* |
|
81
|
|
|
* @return string |
|
82
|
|
|
*/ |
|
83
|
20 |
|
public function scanFile($filestream) |
|
84
|
|
|
{ |
|
85
|
20 |
|
return $this->scan(file_get_contents($filestream)); |
|
86
|
|
|
} |
|
87
|
|
|
} |
|
88
|
|
|
|
PHPOffice /
PhpSpreadsheet
