|
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/6.3/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-]|%%[0-9A-Fa-f]{2})+:)?((?:[\_\.\pL\pN-]|%%[0-9A-Fa-f]{2})+)@)? # basic auth |
|
32
|
|
|
|www\.) |
|
33
|
|
|
(?: |
|
34
|
|
|
(?: |
|
35
|
|
|
(?:xn--[a-z0-9-]++\.)*+xn--[a-z0-9-]++ # a domain name using punycode |
|
36
|
|
|
| |
|
37
|
|
|
(?:[\pL\pN\pS\pM\-\_]++\.)+[\pL\pN\pM]++ # a multi-level domain name |
|
38
|
|
|
| |
|
39
|
|
|
[a-z0-9\-\_]++ # a single-level domain name |
|
40
|
|
|
)\.? |
|
41
|
|
|
| # or |
|
42
|
|
|
\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3} # an IP address |
|
43
|
|
|
| # or |
|
44
|
|
|
\[ |
|
45
|
|
|
(?:(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){6})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:::(?:(?:(?:[0-9a-f]{1,4})):){5})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:[0-9a-f]{1,4})))?::(?:(?:(?:[0-9a-f]{1,4})):){4})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,1}(?:(?:[0-9a-f]{1,4})))?::(?:(?:(?:[0-9a-f]{1,4})):){3})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,2}(?:(?:[0-9a-f]{1,4})))?::(?:(?:(?:[0-9a-f]{1,4})):){2})(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,3}(?:(?:[0-9a-f]{1,4})))?::(?:(?:[0-9a-f]{1,4})):)(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,4}(?:(?:[0-9a-f]{1,4})))?::)(?:(?:(?:(?:(?:[0-9a-f]{1,4})):(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9]))\.){3}(?:(?:25[0-5]|(?:[1-9]|1[0-9]|2[0-4])?[0-9])))))))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,5}(?:(?:[0-9a-f]{1,4})))?::)(?:(?:[0-9a-f]{1,4})))|(?:(?:(?:(?:(?:(?:[0-9a-f]{1,4})):){0,6}(?:(?:[0-9a-f]{1,4})))?::)))) |
|
46
|
|
|
\] # an IPv6 address |
|
47
|
|
|
) |
|
48
|
|
|
(?::[0-9]+)? # a port (optional) |
|
49
|
|
|
(?:/ (?:[\pL\pN\-._\~!$&\'()*+,;=:@]|%%[0-9A-Fa-f]{2})* )* # a path |
|
50
|
|
|
(?:\? (?:[\pL\pN\-._\~!$&\'\[\]()*+,;=:@/?]|%%[0-9A-Fa-f]{2})* )? # a query (optional) |
|
51
|
|
|
(?:\# (?:[\pL\pN\-._\~!$&\'()*+,;=:@/?]|%%[0-9A-Fa-f]{2})* )? # a fragment (optional) |
|
52
|
|
|
)~ixu'; |
|
53
|
|
|
|
|
54
|
|
|
/** |
|
55
|
|
|
* @var string[] |
|
56
|
|
|
* |
|
57
|
|
|
* @psalm-readonly |
|
58
|
|
|
*/ |
|
59
|
|
|
private array $prefixes = ['www.']; |
|
60
|
|
|
|
|
61
|
|
|
/** |
|
62
|
|
|
* @psalm-var non-empty-string |
|
63
|
|
|
* |
|
64
|
|
|
* @psalm-readonly |
|
65
|
|
|
*/ |
|
66
|
|
|
private string $finalRegex; |
|
67
|
|
|
|
|
68
|
|
|
private string $defaultProtocol; |
|
69
|
|
|
|
|
70
|
|
|
/** |
|
71
|
|
|
* @param array<int, string> $allowedProtocols |
|
72
|
|
|
*/ |
|
73
|
190 |
|
public function __construct(array $allowedProtocols = ['http', 'https', 'ftp'], string $defaultProtocol = 'http') |
|
74
|
|
|
{ |
|
75
|
|
|
/** |
|
76
|
|
|
* @psalm-suppress PropertyTypeCoercion |
|
77
|
|
|
*/ |
|
78
|
190 |
|
$this->finalRegex = \sprintf(self::REGEX, \implode('|', $allowedProtocols)); |
|
79
|
|
|
|
|
80
|
190 |
|
foreach ($allowedProtocols as $protocol) { |
|
81
|
190 |
|
$this->prefixes[] = $protocol . '://'; |
|
82
|
|
|
} |
|
83
|
|
|
|
|
84
|
190 |
|
$this->defaultProtocol = $defaultProtocol; |
|
85
|
|
|
} |
|
86
|
|
|
|
|
87
|
190 |
|
public function getMatchDefinition(): InlineParserMatch |
|
88
|
|
|
{ |
|
89
|
190 |
|
return InlineParserMatch::oneOf(...$this->prefixes); |
|
90
|
|
|
} |
|
91
|
|
|
|
|
92
|
86 |
|
public function parse(InlineParserContext $inlineContext): bool |
|
93
|
|
|
{ |
|
94
|
86 |
|
$cursor = $inlineContext->getCursor(); |
|
95
|
|
|
|
|
96
|
|
|
// Autolinks can only come at the beginning of a line, after whitespace, or certain delimiting characters |
|
97
|
86 |
|
$previousChar = $cursor->peek(-1); |
|
98
|
86 |
|
if (! \in_array($previousChar, self::ALLOWED_AFTER, true)) { |
|
99
|
2 |
|
return false; |
|
100
|
|
|
} |
|
101
|
|
|
|
|
102
|
|
|
// Check if we have a valid URL |
|
103
|
84 |
|
if (! \preg_match($this->finalRegex, $cursor->getRemainder(), $matches)) { |
|
104
|
|
|
return false; |
|
105
|
|
|
} |
|
106
|
|
|
|
|
107
|
84 |
|
$url = $matches[0]; |
|
108
|
|
|
|
|
109
|
|
|
// Does the URL end with punctuation that should be stripped? |
|
110
|
84 |
|
if (\preg_match('/(.+?)([?!.,:*_~]+)$/', $url, $matches)) { |
|
111
|
|
|
// Add the punctuation later |
|
112
|
18 |
|
$url = $matches[1]; |
|
113
|
|
|
} |
|
114
|
|
|
|
|
115
|
|
|
// Does the URL end with something that looks like an entity reference? |
|
116
|
84 |
|
if (\preg_match('/(.+)(&[A-Za-z0-9]+;)$/', $url, $matches)) { |
|
117
|
4 |
|
$url = $matches[1]; |
|
118
|
|
|
} |
|
119
|
|
|
|
|
120
|
|
|
// Does the URL need unmatched parens chopped off? |
|
121
|
84 |
|
if (\substr($url, -1) === ')' && ($diff = self::diffParens($url)) > 0) { |
|
122
|
10 |
|
$url = \substr($url, 0, -$diff); |
|
123
|
|
|
} |
|
124
|
|
|
|
|
125
|
84 |
|
$cursor->advanceBy(\mb_strlen($url, 'UTF-8')); |
|
126
|
|
|
|
|
127
|
|
|
// Auto-prefix 'http(s)://' onto 'www' URLs |
|
128
|
84 |
|
if (\substr($url, 0, 4) === 'www.') { |
|
129
|
46 |
|
$inlineContext->getContainer()->appendChild(new Link($this->defaultProtocol . '://' . $url, $url)); |
|
130
|
|
|
|
|
131
|
46 |
|
return true; |
|
132
|
|
|
} |
|
133
|
|
|
|
|
134
|
38 |
|
$inlineContext->getContainer()->appendChild(new Link($url, $url)); |
|
135
|
|
|
|
|
136
|
38 |
|
return true; |
|
137
|
|
|
} |
|
138
|
|
|
|
|
139
|
|
|
/** |
|
140
|
|
|
* @psalm-pure |
|
141
|
|
|
*/ |
|
142
|
16 |
|
private static function diffParens(string $content): int |
|
143
|
|
|
{ |
|
144
|
|
|
// Scan the entire autolink for the total number of parentheses. |
|
145
|
|
|
// If there is a greater number of closing parentheses than opening ones, |
|
146
|
|
|
// we don’t consider ANY of the last characters as part of the autolink, |
|
147
|
|
|
// in order to facilitate including an autolink inside a parenthesis. |
|
148
|
16 |
|
\preg_match_all('/[()]/', $content, $matches); |
|
149
|
|
|
|
|
150
|
16 |
|
$charCount = ['(' => 0, ')' => 0]; |
|
151
|
16 |
|
foreach ($matches[0] as $char) { |
|
152
|
16 |
|
$charCount[$char]++; |
|
153
|
|
|
} |
|
154
|
|
|
|
|
155
|
16 |
|
return $charCount[')'] - $charCount['(']; |
|
156
|
|
|
} |
|
157
|
|
|
} |
|
158
|
|
|
|
thephpleague /
commonmark
