1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
/* |
4
|
|
|
* This file is part of the league/commonmark package. |
5
|
|
|
* |
6
|
|
|
* (c) Colin O'Dell <[email protected]> |
7
|
|
|
* |
8
|
|
|
* Original code based on the CommonMark JS reference parser (https://bitly.com/commonmark-js) |
9
|
|
|
* - (c) John MacFarlane |
10
|
|
|
* |
11
|
|
|
* For the full copyright and license information, please view the LICENSE |
12
|
|
|
* file that was distributed with this source code. |
13
|
|
|
*/ |
14
|
|
|
|
15
|
|
|
namespace League\CommonMark\Util; |
16
|
|
|
|
17
|
|
|
use League\CommonMark\Cursor; |
18
|
|
|
|
19
|
|
|
final class LinkParserHelper |
20
|
|
|
{ |
21
|
|
|
/** |
22
|
|
|
* Attempt to parse link destination |
23
|
|
|
* |
24
|
|
|
* @param Cursor $cursor |
25
|
|
|
* |
26
|
|
|
* @return null|string The string, or null if no match |
27
|
|
|
*/ |
28
|
399 |
|
public static function parseLinkDestination(Cursor $cursor): ?string |
29
|
|
|
{ |
30
|
399 |
|
if ($res = $cursor->match(RegexHelper::REGEX_LINK_DESTINATION_BRACES)) { |
31
|
|
|
// Chop off surrounding <..>: |
32
|
18 |
|
return UrlEncoder::unescapeAndEncode( |
33
|
18 |
|
RegexHelper::unescape(\substr($res, 1, -1)) |
34
|
|
|
); |
35
|
|
|
} |
36
|
|
|
|
37
|
381 |
|
if ($cursor->getCharacter() === '<') { |
38
|
6 |
|
return null; |
39
|
|
|
} |
40
|
|
|
|
41
|
375 |
|
$oldState = $cursor->saveState(); |
42
|
375 |
|
$openParens = 0; |
43
|
375 |
|
while (($c = $cursor->getCharacter()) !== null) { |
44
|
372 |
|
if ($c === '\\' && RegexHelper::isEscapable($cursor->peek())) { |
45
|
18 |
|
$cursor->advanceBy(2); |
46
|
372 |
|
} elseif ($c === '(') { |
47
|
18 |
|
$cursor->advance(); |
48
|
18 |
|
$openParens++; |
49
|
372 |
|
} elseif ($c === ')') { |
50
|
120 |
|
if ($openParens < 1) { |
51
|
117 |
|
break; |
52
|
|
|
} |
53
|
|
|
|
54
|
18 |
|
$cursor->advance(); |
55
|
18 |
|
$openParens--; |
56
|
369 |
|
} elseif (\preg_match(RegexHelper::REGEX_WHITESPACE_CHAR, $c)) { |
57
|
171 |
|
break; |
58
|
|
|
} else { |
59
|
369 |
|
$cursor->advance(); |
60
|
|
|
} |
61
|
|
|
} |
62
|
|
|
|
63
|
375 |
|
$newPos = $cursor->getPosition(); |
64
|
375 |
|
$cursor->restoreState($oldState); |
65
|
|
|
|
66
|
375 |
|
$cursor->advanceBy($newPos - $cursor->getPosition()); |
67
|
|
|
|
68
|
375 |
|
$res = $cursor->getPreviousText(); |
69
|
|
|
|
70
|
375 |
|
return UrlEncoder::unescapeAndEncode( |
71
|
375 |
|
RegexHelper::unescape($res) |
72
|
|
|
); |
73
|
|
|
} |
74
|
|
|
|
75
|
|
|
/** |
76
|
|
|
* @param Cursor $cursor |
77
|
|
|
* |
78
|
|
|
* @return int |
79
|
|
|
*/ |
80
|
405 |
|
public static function parseLinkLabel(Cursor $cursor): int |
81
|
|
|
{ |
82
|
405 |
|
$match = $cursor->match('/^\[(?:[^\\\\\[\]]|\\\\.){0,1000}\]/'); |
83
|
405 |
|
if ($match === null) { |
84
|
210 |
|
return 0; |
85
|
|
|
} |
86
|
|
|
|
87
|
378 |
|
$length = \mb_strlen($match, 'utf-8'); |
88
|
|
|
|
89
|
378 |
|
if ($length > 1001) { |
90
|
|
|
return 0; |
91
|
|
|
} |
92
|
|
|
|
93
|
378 |
|
return $length; |
94
|
|
|
} |
95
|
|
|
|
96
|
|
|
/** |
97
|
|
|
* Attempt to parse link title (sans quotes) |
98
|
|
|
* |
99
|
|
|
* @param Cursor $cursor |
100
|
|
|
* |
101
|
|
|
* @return null|string The string, or null if no match |
102
|
|
|
*/ |
103
|
174 |
|
public static function parseLinkTitle(Cursor $cursor): ?string |
104
|
|
|
{ |
105
|
174 |
|
if ($title = $cursor->match('/' . RegexHelper::PARTIAL_LINK_TITLE . '/')) { |
106
|
|
|
// Chop off quotes from title and unescape |
107
|
138 |
|
return RegexHelper::unescape(substr($title, 1, -1)); |
108
|
|
|
} |
109
|
|
|
|
110
|
36 |
|
return null; |
111
|
|
|
} |
112
|
|
|
} |
113
|
|
|
|