|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
namespace PhpOffice\PhpSpreadsheet\Reader\Security; |
|
4
|
|
|
|
|
5
|
|
|
use PhpOffice\PhpSpreadsheet\Reader; |
|
6
|
|
|
|
|
7
|
|
|
class XmlScanner |
|
8
|
|
|
{ |
|
9
|
|
|
private string $pattern; |
|
10
|
|
|
|
|
11
|
|
|
/** @var ?callable */ |
|
12
|
|
|
private $callback; |
|
13
|
|
|
|
|
14
|
1296 |
|
public function __construct(string $pattern = '<!DOCTYPE') |
|
15
|
|
|
{ |
|
16
|
1296 |
|
$this->pattern = $pattern; |
|
17
|
|
|
} |
|
18
|
|
|
|
|
19
|
1295 |
|
public static function getInstance(Reader\IReader $reader): self |
|
20
|
|
|
{ |
|
21
|
1295 |
|
$pattern = ($reader instanceof Reader\Html) ? '<!ENTITY' : '<!DOCTYPE'; |
|
22
|
|
|
|
|
23
|
1295 |
|
return new self($pattern); |
|
24
|
|
|
} |
|
25
|
|
|
|
|
26
|
1 |
|
public function setAdditionalCallback(callable $callback): void |
|
27
|
|
|
{ |
|
28
|
1 |
|
$this->callback = $callback; |
|
29
|
|
|
} |
|
30
|
|
|
|
|
31
|
2 |
|
private static function forceString(mixed $arg): string |
|
32
|
|
|
{ |
|
33
|
2 |
|
return is_string($arg) ? $arg : ''; |
|
34
|
|
|
} |
|
35
|
|
|
|
|
36
|
1248 |
|
private function toUtf8(string $xml): string |
|
37
|
|
|
{ |
|
38
|
1248 |
|
$charset = $this->findCharSet($xml); |
|
39
|
1248 |
|
if ($charset !== 'UTF-8') { |
|
40
|
1248 |
|
$xml = self::forceString(mb_convert_encoding($xml, 'UTF-8', $charset)); |
|
41
|
|
|
|
|
42
|
1248 |
|
$charset = $this->findCharSet($xml); |
|
43
|
2 |
|
if ($charset !== 'UTF-8') { |
|
44
|
|
|
throw new Reader\Exception('Suspicious Double-encoded XML, spreadsheet file load() aborted to prevent XXE/XEE attacks'); |
|
45
|
2 |
|
} |
|
46
|
2 |
|
} |
|
47
|
2 |
|
|
|
48
|
2 |
|
return $xml; |
|
49
|
|
|
} |
|
50
|
|
|
|
|
51
|
|
|
private function findCharSet(string $xml): string |
|
52
|
1246 |
|
{ |
|
53
|
|
|
$patterns = [ |
|
54
|
|
|
'/encoding="([^"]*]?)"/', |
|
55
|
|
|
"/encoding='([^']*?)'/", |
|
56
|
|
|
]; |
|
57
|
|
|
|
|
58
|
|
|
foreach ($patterns as $pattern) { |
|
59
|
|
|
if (preg_match($pattern, $xml, $matches)) { |
|
60
|
1248 |
|
return strtoupper($matches[1]); |
|
61
|
|
|
} |
|
62
|
1248 |
|
} |
|
63
|
|
|
|
|
64
|
1248 |
|
return 'UTF-8'; |
|
65
|
|
|
} |
|
66
|
|
|
|
|
67
|
1246 |
|
/** |
|
68
|
|
|
* Scan the XML for use of <!ENTITY to prevent XXE/XEE attacks. |
|
69
|
1246 |
|
* |
|
70
|
5 |
|
* @param false|string $xml |
|
71
|
|
|
*/ |
|
72
|
|
|
public function scan($xml): string |
|
73
|
1241 |
|
{ |
|
74
|
1 |
|
$xml = "$xml"; |
|
75
|
|
|
|
|
76
|
|
|
$xml = $this->toUtf8($xml); |
|
77
|
1241 |
|
|
|
78
|
|
|
// Don't rely purely on libxml_disable_entity_loader() |
|
79
|
|
|
$pattern = '/\\0?' . implode('\\0?', str_split($this->pattern)) . '\\0?/'; |
|
80
|
|
|
|
|
81
|
|
|
if (preg_match($pattern, $xml)) { |
|
82
|
|
|
throw new Reader\Exception('Detected use of ENTITY in XML, spreadsheet file load() aborted to prevent XXE/XEE attacks'); |
|
83
|
470 |
|
} |
|
84
|
|
|
|
|
85
|
470 |
|
if ($this->callback !== null) { |
|
86
|
|
|
$xml = call_user_func($this->callback, $xml); |
|
87
|
|
|
} |
|
88
|
|
|
|
|
89
|
|
|
return $xml; |
|
90
|
|
|
} |
|
91
|
|
|
|
|
92
|
|
|
/** |
|
93
|
|
|
* Scan theXML for use of <!ENTITY to prevent XXE/XEE attacks. |
|
94
|
|
|
*/ |
|
95
|
|
|
public function scanFile(string $filestream): string |
|
96
|
|
|
{ |
|
97
|
|
|
return $this->scan(file_get_contents($filestream)); |
|
98
|
|
|
} |
|
99
|
|
|
} |
|
100
|
|
|
|
PHPOffice /
PhpSpreadsheet
