|
1
|
|
|
<?php |
|
2
|
|
|
|
|
3
|
|
|
declare(strict_types=1); |
|
4
|
|
|
|
|
5
|
|
|
/** |
|
6
|
|
|
* Saito - The Threaded Web Forum |
|
7
|
|
|
* |
|
8
|
|
|
* @copyright Copyright (c) the Saito Project Developers |
|
9
|
|
|
* @link https://github.com/Schlaefer/Saito |
|
10
|
|
|
* @license http://opensource.org/licenses/MIT |
|
11
|
|
|
*/ |
|
12
|
|
|
|
|
13
|
|
|
namespace Plugin\BbcodeParser\src\Lib\jBBCode\Visitors; |
|
14
|
|
|
|
|
15
|
|
|
use Plugin\BbcodeParser\src\Lib\Helper\UrlParserTrait; |
|
16
|
|
|
|
|
17
|
|
|
/** |
|
18
|
|
|
* Handles all implicit linking in a text (autolink URLs, tags, ...) |
|
19
|
|
|
*/ |
|
20
|
|
|
class JbbCodeAutolinkVisitor extends JbbCodeTextVisitor |
|
21
|
|
|
{ |
|
22
|
|
|
use UrlParserTrait; |
|
23
|
|
|
|
|
24
|
|
|
protected $_disallowedTags = ['code']; |
|
25
|
|
|
|
|
26
|
|
|
/** |
|
27
|
|
|
* {@inheritDoc} |
|
28
|
|
|
*/ |
|
29
|
|
|
protected function _processTextNode($string, $node) |
|
30
|
|
|
{ |
|
31
|
|
|
// don't auto-link in url tags; problem is that 'urlWithAttributes' definition |
|
32
|
|
|
// reuses 'url' tag with ParseContent = true |
|
33
|
|
|
if ($node->getParent()->getTagName() === 'url') { |
|
34
|
|
|
return $string; |
|
35
|
|
|
} |
|
36
|
|
|
$string = $this->hashLink($string); |
|
37
|
|
|
$string = $this->atUserLink($string); |
|
38
|
|
|
|
|
39
|
|
|
return $this->autolink($string); |
|
40
|
|
|
} |
|
41
|
|
|
|
|
42
|
|
|
/** |
|
43
|
|
|
* Links @<username> to the user's profile. |
|
44
|
|
|
* |
|
45
|
|
|
* @param string $string The Text to be parsed. |
|
46
|
|
|
* @return string The text with usernames linked. |
|
47
|
|
|
*/ |
|
48
|
|
|
protected function atUserLink(string $string): string |
|
49
|
|
|
{ |
|
50
|
|
|
$tags = []; |
|
51
|
|
|
|
|
52
|
|
|
/* |
|
53
|
|
|
* - '\pP' all unicode punctuation marks |
|
54
|
|
|
* - '<' if nl2br has taken place whatchout for <br /> linebreaks |
|
55
|
|
|
*/ |
|
56
|
|
|
$hasTags = preg_match_all('/(\s|^)@([^\s\pP<]+)/m', $string, $tags); |
|
57
|
|
|
if (!$hasTags) { |
|
58
|
|
|
return $string; |
|
59
|
|
|
} |
|
60
|
|
|
|
|
61
|
|
|
// would be cleaner to pass userlist by value, but for performance reasons |
|
62
|
|
|
// we only query the db if we actually have any @ tags |
|
63
|
|
|
$users = $this->_sOptions->get('UserList')->get(); |
|
64
|
|
|
sort($users); |
|
65
|
|
|
$names = []; |
|
66
|
|
|
if (empty($tags[2]) === false) { |
|
67
|
|
|
$tags = $tags[2]; |
|
68
|
|
|
foreach ($tags as $tag) { |
|
69
|
|
|
if (in_array($tag, $users)) { |
|
70
|
|
|
$names[$tag] = 1; |
|
71
|
|
|
} else { |
|
72
|
|
|
$continue = 0; |
|
73
|
|
|
foreach ($users as $user) { |
|
74
|
|
|
if (mb_strpos($user, $tag) === 0) { |
|
75
|
|
|
$names[$user] = 1; |
|
76
|
|
|
$continue = true; |
|
77
|
|
|
} |
|
78
|
|
|
if ($continue === false) { |
|
79
|
|
|
break; |
|
80
|
|
|
} elseif ($continue !== 0) { |
|
81
|
|
|
$continue = false; |
|
82
|
|
|
} |
|
83
|
|
|
} |
|
84
|
|
|
} |
|
85
|
|
|
} |
|
86
|
|
|
} |
|
87
|
|
|
krsort($names); |
|
88
|
|
|
$baseUrl = $this->_sOptions->get('webroot') . $this->_sOptions->get('atBaseUrl'); |
|
89
|
|
|
foreach ($names as $name => $v) { |
|
90
|
|
|
$title = urlencode($name); |
|
91
|
|
|
$link = $this->_url( |
|
92
|
|
|
$baseUrl . $title, |
|
93
|
|
|
"@$name", |
|
94
|
|
|
false |
|
95
|
|
|
); |
|
96
|
|
|
$string = str_replace("@$name", $link, $string); |
|
97
|
|
|
} |
|
98
|
|
|
|
|
99
|
|
|
return $string; |
|
100
|
|
|
} |
|
101
|
|
|
|
|
102
|
|
|
/** |
|
103
|
|
|
* Autolinks URLs not surrounded by explicit URL-tags for user-convenience. |
|
104
|
|
|
* |
|
105
|
|
|
* @param string $string The text to be parsed for URLs. |
|
106
|
|
|
* @return string The text with URLs linked. |
|
107
|
|
|
*/ |
|
108
|
|
|
protected function autolink(string $string): string |
|
109
|
|
|
{ |
|
110
|
|
|
$replace = function (array $matches): string { |
|
111
|
|
|
/// don't link locally |
|
112
|
|
|
if (strpos($matches['element'], 'file://') !== false) { |
|
113
|
|
|
return $matches['element']; |
|
114
|
|
|
} |
|
115
|
|
|
|
|
116
|
|
|
/// exclude punctuation at end of sentence from URLs |
|
117
|
|
|
$ignoredEndChars = implode('|', [',', '\?', ',', '\.', '\)', '!']); |
|
118
|
|
|
preg_match( |
|
119
|
|
|
'/(?P<element>.*?)(?P<suffix>' . $ignoredEndChars . ')?$/', |
|
120
|
|
|
$matches['element'], |
|
121
|
|
|
$m |
|
122
|
|
|
); |
|
123
|
|
|
|
|
124
|
|
|
/// exclude ignored end chars if paired in URL foo.com/bar_(baz) |
|
125
|
|
|
if (!empty($m['suffix'])) { |
|
126
|
|
|
$ignoredIfNotPaired = [ |
|
127
|
|
|
['open' => '(', 'close' => ')'], |
|
128
|
|
|
]; |
|
129
|
|
|
foreach ($ignoredIfNotPaired as $pair) { |
|
130
|
|
|
$isUnpaired = substr_count($m['element'], $pair['open']) > substr_count($m['element'], $pair['close']); |
|
131
|
|
|
if ($isUnpaired) { |
|
132
|
|
|
$m['element'] .= $m['suffix']; |
|
133
|
|
|
unset($m['suffix']); |
|
134
|
|
|
} |
|
135
|
|
|
} |
|
136
|
|
|
} |
|
137
|
|
|
|
|
138
|
|
|
/// keep ['element'] and ['suffix'] and include ['prefix']; (array) for phpstan |
|
139
|
|
|
$matches = (array)($m + $matches); |
|
140
|
|
|
|
|
141
|
|
|
if (strpos($matches['element'], '://') === false) { |
|
142
|
|
|
$matches['element'] = 'http://' . $matches['element']; |
|
143
|
|
|
} |
|
144
|
|
|
$matches += [ |
|
145
|
|
|
'prefix' => '', |
|
146
|
|
|
'suffix' => '', |
|
147
|
|
|
]; |
|
148
|
|
|
|
|
149
|
|
|
$url = $this->_url( |
|
150
|
|
|
$matches['element'], |
|
151
|
|
|
$matches['element'], |
|
152
|
|
|
false, |
|
153
|
|
|
true |
|
154
|
|
|
); |
|
155
|
|
|
|
|
156
|
|
|
return $matches['prefix'] . $url . $matches['suffix']; |
|
157
|
|
|
}; |
|
158
|
|
|
|
|
159
|
|
|
//# autolink http://urls |
|
160
|
|
|
$string = preg_replace_callback( |
|
161
|
|
|
"#(?<=^|[\n (])(?P<element>[\w]+?://.*?[^ \"\n\r\t<]*)#is", |
|
162
|
|
|
$replace, |
|
163
|
|
|
$string |
|
164
|
|
|
); |
|
165
|
|
|
|
|
166
|
|
|
//# autolink without http://, i.e. www.foo.bar/baz |
|
167
|
|
|
$string = preg_replace_callback( |
|
168
|
|
|
"#(?P<prefix>^|[\n (])(?P<element>(www|ftp)\.[\w\-]+\.[\w\-.\~]+(?:/[^ \"\t\n\r<]*)?)#is", |
|
169
|
|
|
$replace, |
|
170
|
|
|
$string |
|
171
|
|
|
); |
|
172
|
|
|
|
|
173
|
|
|
//# autolink email |
|
174
|
|
|
$string = preg_replace_callback( |
|
175
|
|
|
"#(?<=^|[\n ])(?P<content>([a-z0-9&\-_.]+?)@([\w\-]+\.([\w\-\.]+\.)*[\w]+))#i", |
|
176
|
|
|
function ($matches) { |
|
177
|
|
|
return $this->_email($matches['content']); |
|
178
|
|
|
}, |
|
179
|
|
|
$string |
|
180
|
|
|
); |
|
181
|
|
|
|
|
182
|
|
|
return $string; |
|
183
|
|
|
} |
|
184
|
|
|
|
|
185
|
|
|
/** |
|
186
|
|
|
* Links #<posting-ID> to that posting. |
|
187
|
|
|
* |
|
188
|
|
|
* @param string $string Text to be parsed for #<id>. |
|
189
|
|
|
* @return string Text containing hash-links. |
|
190
|
|
|
*/ |
|
191
|
|
|
protected function hashLink(string $string): string |
|
192
|
|
|
{ |
|
193
|
|
|
$baseUrl = $this->_sOptions->get('webroot') . $this->_sOptions->get('hashBaseUrl'); |
|
194
|
|
|
$string = preg_replace_callback( |
|
195
|
|
|
'/(?<=\s|^|]|\()(?<tag>#)(?<element>\d+)(?!\w)/', |
|
196
|
|
|
function (array $m) use ($baseUrl): string { |
|
197
|
|
|
$hash = $m['element']; |
|
198
|
|
|
|
|
199
|
|
|
return $this->_url($baseUrl . $hash, '#' . $hash); |
|
200
|
|
|
}, |
|
201
|
|
|
$string |
|
202
|
|
|
); |
|
203
|
|
|
|
|
204
|
|
|
return $string; |
|
205
|
|
|
} |
|
206
|
|
|
} |
|
207
|
|
|
|
Schlaefer /
Saito
