1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
declare(strict_types=1); |
4
|
|
|
|
5
|
|
|
/* |
6
|
|
|
* This file is part of the league/commonmark package. |
7
|
|
|
* |
8
|
|
|
* (c) Colin O'Dell <[email protected]> |
9
|
|
|
* |
10
|
|
|
* For the full copyright and license information, please view the LICENSE |
11
|
|
|
* file that was distributed with this source code. |
12
|
|
|
*/ |
13
|
|
|
|
14
|
|
|
namespace League\CommonMark\Extension\Autolink; |
15
|
|
|
|
16
|
|
|
use League\CommonMark\Extension\CommonMark\Node\Inline\Link; |
17
|
|
|
use League\CommonMark\Parser\Inline\InlineParserInterface; |
18
|
|
|
use League\CommonMark\Parser\Inline\InlineParserMatch; |
19
|
|
|
use League\CommonMark\Parser\InlineParserContext; |
20
|
|
|
|
21
|
|
|
final class UrlAutolinkParser implements InlineParserInterface |
22
|
|
|
{ |
23
|
|
|
private const ALLOWED_AFTER = [null, ' ', "\t", "\n", "\x0b", "\x0c", "\x0d", '*', '_', '~', '(']; |
24
|
|
|
|
25
|
|
|
// RegEx adapted from https://github.com/symfony/symfony/blob/4.2/src/Symfony/Component/Validator/Constraints/UrlValidator.php |
26
|
|
|
private const REGEX = '~ |
27
|
|
|
( |
28
|
|
|
# Must start with a supported scheme + auth, or "www" |
29
|
|
|
(?: |
30
|
|
|
(?:%s):// # protocol |
31
|
|
|
(?:([\.\pL\pN-]+:)?([\.\pL\pN-]+)@)? # basic auth |
32
|
|
|
|www\.) |
33
|
|
|
(?: |
34
|
|
|
(?:[\pL\pN\pS\-\.])+(?:\.?(?:[\pL\pN]|xn\-\-[\pL\pN-]+)+\.?) # a domain name |
35
|
|
|
| # or |
36
|
|
|
\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3} # an IP address |
37
|
|
|
| # or |
38
|
|
|
\[ |
39
|
|
|
(?:(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){6})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:::(?:(?:(?:[0-9a-f]{1,4})):){5})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:[0-9a-f]{1,4})))?::(?:(?:(?:[0-9a-f]{1,4})):){4})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,1}(?:(?:[0-9a-f]{1,4})))?::(?:(?:(?:[0-9a-f]{1,4})):){3})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,2}(?:(?:[0-9a-f]{1,4})))?::(?:(?:(?:[0-9a-f]{1,4})):){2})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,3}(?:(?:[0-9a-f]{1,4})))?::(?:(?:[0-9a-f]{1,4})):)(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,4}(?:(?:[0-9a-f]{1,4})))?::)(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,5}(?:(?:[0-9a-f]{1,4})))?::)(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,6}(?:(?:[0-9a-f]{1,4})))?::)))) |
40
|
|
|
\] # an IPv6 address |
41
|
|
|
) |
42
|
|
|
(?::[0-9]+)? # a port (optional) |
43
|
|
|
(?:/ (?:[\pL\pN\-._\~!$&\'()*+,;=:@]|%%[0-9A-Fa-f]{2})* )* # a path |
44
|
|
|
(?:\? (?:[\pL\pN\-._\~!$&\'()*+,;=:@/?]|%%[0-9A-Fa-f]{2})* )? # a query (optional) |
45
|
|
|
(?:\# (?:[\pL\pN\-._\~!$&\'()*+,;=:@/?]|%%[0-9A-Fa-f]{2})* )? # a fragment (optional) |
46
|
|
|
)~ixu'; |
47
|
|
|
|
48
|
|
|
/** |
49
|
|
|
* @var string[] |
50
|
|
|
* |
51
|
|
|
* @psalm-readonly |
52
|
|
|
*/ |
53
|
|
|
private $prefixes = ['www']; |
54
|
|
|
|
55
|
|
|
/** |
56
|
|
|
* @var string |
57
|
|
|
* |
58
|
|
|
* @psalm-readonly |
59
|
|
|
*/ |
60
|
|
|
private $finalRegex; |
61
|
|
|
|
62
|
|
|
/** |
63
|
|
|
* @param array<int, string> $allowedProtocols |
64
|
|
|
*/ |
65
|
234 |
|
public function __construct(array $allowedProtocols = ['http', 'https', 'ftp']) |
66
|
|
|
{ |
67
|
234 |
|
$this->finalRegex = \sprintf(self::REGEX, \implode('|', $allowedProtocols)); |
68
|
|
|
|
69
|
234 |
|
foreach ($allowedProtocols as $protocol) { |
70
|
234 |
|
$this->prefixes[] = $protocol . '://'; |
71
|
|
|
} |
72
|
234 |
|
} |
73
|
|
|
|
74
|
234 |
|
public function getMatchDefinition(): InlineParserMatch |
75
|
|
|
{ |
76
|
234 |
|
return InlineParserMatch::oneOf(...$this->prefixes); |
77
|
|
|
} |
78
|
|
|
|
79
|
99 |
|
public function parse(string $match, InlineParserContext $inlineContext): bool |
80
|
|
|
{ |
81
|
99 |
|
$cursor = $inlineContext->getCursor(); |
82
|
|
|
|
83
|
|
|
// Autolinks can only come at the beginning of a line, after whitespace, or certain delimiting characters |
84
|
99 |
|
$previousChar = $cursor->peek(-1); |
85
|
99 |
|
if (! \in_array($previousChar, self::ALLOWED_AFTER, true)) { |
86
|
|
|
return false; |
87
|
|
|
} |
88
|
|
|
|
89
|
|
|
// Check if we have a valid URL |
90
|
99 |
|
if (! \preg_match($this->finalRegex, $cursor->getRemainder(), $matches)) { |
91
|
3 |
|
return false; |
92
|
|
|
} |
93
|
|
|
|
94
|
96 |
|
$url = $matches[0]; |
95
|
|
|
|
96
|
|
|
// Does the URL end with punctuation that should be stripped? |
97
|
96 |
|
if (\preg_match('/(.+)([?!.,:*_~]+)$/', $url, $matches)) { |
98
|
|
|
// Add the punctuation later |
99
|
24 |
|
$url = $matches[1]; |
100
|
|
|
} |
101
|
|
|
|
102
|
|
|
// Does the URL end with something that looks like an entity reference? |
103
|
96 |
|
if (\preg_match('/(.+)(&[A-Za-z0-9]+;)$/', $url, $matches)) { |
104
|
6 |
|
$url = $matches[1]; |
105
|
|
|
} |
106
|
|
|
|
107
|
|
|
// Does the URL need unmatched parens chopped off? |
108
|
96 |
|
if (\substr($url, -1) === ')' && ($diff = self::diffParens($url)) > 0) { |
109
|
15 |
|
$url = \substr($url, 0, -$diff); |
110
|
|
|
} |
111
|
|
|
|
112
|
96 |
|
$cursor->advanceBy(\mb_strlen($url)); |
113
|
|
|
|
114
|
|
|
// Auto-prefix 'http://' onto 'www' URLs |
115
|
96 |
|
if (\substr($url, 0, 4) === 'www.') { |
116
|
54 |
|
$inlineContext->getContainer()->appendChild(new Link('http://' . $url, $url)); |
117
|
|
|
|
118
|
54 |
|
return true; |
119
|
|
|
} |
120
|
|
|
|
121
|
42 |
|
$inlineContext->getContainer()->appendChild(new Link($url, $url)); |
122
|
|
|
|
123
|
42 |
|
return true; |
124
|
|
|
} |
125
|
|
|
|
126
|
|
|
/** |
127
|
|
|
* @psalm-pure |
128
|
|
|
*/ |
129
|
24 |
|
private static function diffParens(string $content): int |
130
|
|
|
{ |
131
|
|
|
// Scan the entire autolink for the total number of parentheses. |
132
|
|
|
// If there is a greater number of closing parentheses than opening ones, |
133
|
|
|
// we don’t consider ANY of the last characters as part of the autolink, |
134
|
|
|
// in order to facilitate including an autolink inside a parenthesis. |
135
|
24 |
|
\preg_match_all('/[()]/', $content, $matches); |
136
|
|
|
|
137
|
24 |
|
$charCount = ['(' => 0, ')' => 0]; |
138
|
24 |
|
foreach ($matches[0] as $char) { |
139
|
24 |
|
$charCount[$char]++; |
140
|
|
|
} |
141
|
|
|
|
142
|
24 |
|
return $charCount[')'] - $charCount['(']; |
143
|
|
|
} |
144
|
|
|
} |
145
|
|
|
|