1
|
|
|
<?php |
2
|
|
|
/** |
3
|
|
|
* @copyright Copyright (c) 2014 Carsten Brandt |
4
|
|
|
* @license https://github.com/cebe/markdown/blob/master/LICENSE |
5
|
|
|
* @link https://github.com/cebe/markdown#readme |
6
|
|
|
*/ |
7
|
|
|
|
8
|
|
|
namespace cebe\markdown\inline; |
9
|
|
|
|
10
|
|
|
// work around https://github.com/facebook/hhvm/issues/1120 |
11
|
1 |
|
defined('ENT_HTML401') || define('ENT_HTML401', 0); |
12
|
|
|
|
13
|
|
|
/** |
14
|
|
|
* Addes links and images as well as url markers. |
15
|
|
|
* |
16
|
|
|
* This trait conflicts with the HtmlTrait. If both are used together, |
17
|
|
|
* you have to define a resolution, by defining the HtmlTrait::parseInlineHtml |
18
|
|
|
* as private so it is not used directly: |
19
|
|
|
* |
20
|
|
|
* ```php |
21
|
|
|
* use block\HtmlTrait { |
22
|
|
|
* parseInlineHtml as private parseInlineHtml; |
23
|
|
|
* } |
24
|
|
|
* ``` |
25
|
|
|
* |
26
|
|
|
* If the method exists it is called internally by this trait. |
27
|
|
|
* |
28
|
|
|
* Also make sure to reset references on prepare(): |
29
|
|
|
* |
30
|
|
|
* ```php |
31
|
|
|
* protected function prepare() |
32
|
|
|
* { |
33
|
|
|
* // reset references |
34
|
|
|
* $this->references = []; |
35
|
|
|
* } |
36
|
|
|
* ``` |
37
|
|
|
*/ |
38
|
|
|
trait LinkTrait |
39
|
|
|
{ |
40
|
|
|
/** |
41
|
|
|
* @var array a list of defined references in this document. |
42
|
|
|
*/ |
43
|
|
|
protected $references = []; |
44
|
|
|
|
45
|
|
|
/** |
46
|
|
|
* Remove backslash from escaped characters |
47
|
|
|
* @param $text |
48
|
|
|
* @return string |
49
|
|
|
*/ |
50
|
55 |
|
protected function replaceEscape($text) |
51
|
|
|
{ |
52
|
55 |
|
$strtr = []; |
53
|
55 |
|
foreach($this->escapeCharacters as $char) { |
54
|
55 |
|
$strtr["\\$char"] = $char; |
55
|
|
|
} |
56
|
55 |
|
return strtr($text, $strtr); |
57
|
|
|
} |
58
|
|
|
|
59
|
|
|
/** |
60
|
|
|
* Parses a link indicated by `[`. |
61
|
|
|
* @marker [ |
62
|
|
|
*/ |
63
|
53 |
|
protected function parseLink($markdown) |
64
|
|
|
{ |
65
|
53 |
|
if (!in_array('parseLink', array_slice($this->context, 1)) && ($parts = $this->parseLinkOrImage($markdown)) !== false) { |
66
|
53 |
|
list($text, $url, $title, $offset, $key) = $parts; |
67
|
|
|
return [ |
68
|
|
|
[ |
69
|
53 |
|
'link', |
70
|
53 |
|
'text' => $this->parseInline($text), |
71
|
53 |
|
'url' => $url, |
72
|
53 |
|
'title' => $title, |
73
|
53 |
|
'refkey' => $key, |
74
|
53 |
|
'orig' => substr($markdown, 0, $offset), |
75
|
|
|
], |
76
|
53 |
|
$offset |
77
|
|
|
]; |
78
|
|
View Code Duplication |
} else { |
79
|
|
|
// remove all starting [ markers to avoid next one to be parsed as link |
80
|
11 |
|
$result = '['; |
81
|
11 |
|
$i = 1; |
82
|
11 |
|
while (isset($markdown[$i]) && $markdown[$i] === '[') { |
83
|
|
|
$result .= '['; |
84
|
|
|
$i++; |
85
|
|
|
} |
86
|
11 |
|
return [['text', $result], $i]; |
87
|
|
|
} |
88
|
|
|
} |
89
|
|
|
|
90
|
|
|
/** |
91
|
|
|
* Parses an image indicated by `![`. |
92
|
|
|
* @marker ![ |
93
|
|
|
*/ |
94
|
10 |
|
protected function parseImage($markdown) |
95
|
|
|
{ |
96
|
10 |
|
if (($parts = $this->parseLinkOrImage(substr($markdown, 1))) !== false) { |
97
|
10 |
|
list($text, $url, $title, $offset, $key) = $parts; |
98
|
|
|
|
99
|
|
|
return [ |
100
|
|
|
[ |
101
|
10 |
|
'image', |
102
|
10 |
|
'text' => $text, |
103
|
10 |
|
'url' => $url, |
104
|
10 |
|
'title' => $title, |
105
|
10 |
|
'refkey' => $key, |
106
|
10 |
|
'orig' => substr($markdown, 0, $offset + 1), |
107
|
|
|
], |
108
|
10 |
|
$offset + 1 |
109
|
|
|
]; |
110
|
|
View Code Duplication |
} else { |
111
|
|
|
// remove all starting [ markers to avoid next one to be parsed as link |
112
|
3 |
|
$result = '!'; |
113
|
3 |
|
$i = 1; |
114
|
3 |
|
while (isset($markdown[$i]) && $markdown[$i] === '[') { |
115
|
3 |
|
$result .= '['; |
116
|
3 |
|
$i++; |
117
|
|
|
} |
118
|
3 |
|
return [['text', $result], $i]; |
119
|
|
|
} |
120
|
|
|
} |
121
|
|
|
|
122
|
53 |
|
protected function parseLinkOrImage($markdown) |
123
|
|
|
{ |
124
|
53 |
|
if (strpos($markdown, ']') !== false && preg_match('/\[((?>[^\]\[]+|(?R))*)\]/', $markdown, $textMatches)) { // TODO improve bracket regex |
125
|
53 |
|
$text = $textMatches[1]; |
126
|
53 |
|
$offset = strlen($textMatches[0]); |
127
|
53 |
|
$markdown = substr($markdown, $offset); |
128
|
|
|
|
129
|
|
|
$pattern = <<<REGEXP |
130
|
53 |
|
/(?(R) # in case of recursion match parentheses |
131
|
|
|
\(((?>[^\s()]+)|(?R))*\) |
132
|
|
|
| # else match a link with title |
133
|
|
|
^\(\s*(((?>[^\s()]+)|(?R))*)(\s+"(.*?)")?\s*\) |
134
|
|
|
)/x |
135
|
|
|
REGEXP; |
136
|
53 |
|
if (preg_match($pattern, $markdown, $refMatches)) { |
137
|
|
|
// inline link |
138
|
|
|
return [ |
139
|
36 |
|
$text, |
140
|
36 |
|
isset($refMatches[2]) ? $this->replaceEscape($refMatches[2]) : '', // url |
141
|
36 |
|
empty($refMatches[5]) ? null: $refMatches[5], // title |
142
|
36 |
|
$offset + strlen($refMatches[0]), // offset |
143
|
|
|
null, // reference key |
144
|
|
|
]; |
145
|
43 |
|
} elseif (preg_match('/^([ \n]?\[(.*?)\])?/s', $markdown, $refMatches)) { |
146
|
|
|
// reference style link |
147
|
43 |
|
if (empty($refMatches[2])) { |
148
|
36 |
|
$key = strtolower($text); |
149
|
|
|
} else { |
150
|
17 |
|
$key = strtolower($refMatches[2]); |
151
|
|
|
} |
152
|
|
|
return [ |
153
|
43 |
|
$text, |
154
|
|
|
null, // url |
155
|
|
|
null, // title |
156
|
43 |
|
$offset + strlen($refMatches[0]), // offset |
157
|
43 |
|
$key, |
158
|
|
|
]; |
159
|
|
|
} |
160
|
|
|
} |
161
|
3 |
|
return false; |
162
|
|
|
} |
163
|
|
|
|
164
|
|
|
/** |
165
|
|
|
* Parses inline HTML. |
166
|
|
|
* @marker < |
167
|
|
|
*/ |
168
|
30 |
|
protected function parseLt($text) |
169
|
|
|
{ |
170
|
30 |
|
if (strpos($text, '>') !== false) { |
171
|
26 |
|
if (!in_array('parseLink', $this->context)) { // do not allow links in links |
172
|
26 |
View Code Duplication |
if (preg_match('/^<([^\s]*?@[^\s]*?\.\w+?)>/', $text, $matches)) { |
173
|
|
|
// email address |
174
|
|
|
return [ |
175
|
3 |
|
['email', $this->replaceEscape($matches[1])], |
176
|
3 |
|
strlen($matches[0]) |
177
|
|
|
]; |
178
|
26 |
|
} elseif (preg_match('/^<([a-z]{3,}:\/\/[^\s]+?)>/', $text, $matches)) { |
179
|
|
|
// URL |
180
|
|
|
return [ |
181
|
10 |
|
['url', $this->replaceEscape($matches[1])], |
182
|
10 |
|
strlen($matches[0]) |
183
|
|
|
]; |
184
|
|
|
} |
185
|
|
|
} |
186
|
|
|
// try inline HTML if it was neither a URL nor email if HtmlTrait is included. |
187
|
16 |
|
if (method_exists($this, 'parseInlineHtml')) { |
188
|
16 |
|
return $this->parseInlineHtml($text); |
|
|
|
|
189
|
|
|
} |
190
|
|
|
} |
191
|
11 |
|
return [['text', '<'], 1]; |
192
|
|
|
} |
193
|
|
|
|
194
|
3 |
|
protected function renderEmail($block) |
195
|
|
|
{ |
196
|
3 |
|
$email = htmlspecialchars($block[1], ENT_NOQUOTES | ENT_SUBSTITUTE, 'UTF-8'); |
197
|
3 |
|
return "<a href=\"mailto:$email\">$email</a>"; |
198
|
|
|
} |
199
|
|
|
|
200
|
10 |
View Code Duplication |
protected function renderUrl($block) |
|
|
|
|
201
|
|
|
{ |
202
|
10 |
|
$url = htmlspecialchars($block[1], ENT_COMPAT | ENT_HTML401, 'UTF-8'); |
203
|
10 |
|
$decodedUrl = urldecode($block[1]); |
204
|
10 |
|
$secureUrlText = preg_match('//u', $decodedUrl) ? $decodedUrl : $block[1]; |
205
|
10 |
|
$text = htmlspecialchars($secureUrlText, ENT_NOQUOTES | ENT_SUBSTITUTE, 'UTF-8'); |
206
|
10 |
|
return "<a href=\"$url\">$text</a>"; |
207
|
|
|
} |
208
|
|
|
|
209
|
43 |
|
protected function lookupReference($key) |
210
|
|
|
{ |
211
|
43 |
|
$normalizedKey = preg_replace('/\s+/', ' ', $key); |
212
|
43 |
|
if (isset($this->references[$key]) || isset($this->references[$key = $normalizedKey])) { |
213
|
29 |
|
return $this->references[$key]; |
214
|
|
|
} |
215
|
23 |
|
return false; |
216
|
|
|
} |
217
|
|
|
|
218
|
35 |
|
protected function renderLink($block) |
219
|
|
|
{ |
220
|
35 |
View Code Duplication |
if (isset($block['refkey'])) { |
221
|
28 |
|
if (($ref = $this->lookupReference($block['refkey'])) !== false) { |
222
|
19 |
|
$block = array_merge($block, $ref); |
223
|
|
|
} else { |
224
|
15 |
|
if (strncmp($block['orig'], '[', 1) === 0) { |
225
|
15 |
|
return '[' . $this->renderAbsy($this->parseInline(substr($block['orig'], 1))); |
226
|
|
|
} |
227
|
|
|
return $block['orig']; |
228
|
|
|
} |
229
|
|
|
} |
230
|
34 |
|
return '<a href="' . htmlspecialchars($block['url'], ENT_COMPAT | ENT_HTML401, 'UTF-8') . '"' |
231
|
34 |
|
. (empty($block['title']) ? '' : ' title="' . htmlspecialchars($block['title'], ENT_COMPAT | ENT_HTML401 | ENT_SUBSTITUTE, 'UTF-8') . '"') |
232
|
34 |
|
. '>' . $this->renderAbsy($block['text']) . '</a>'; |
233
|
|
|
} |
234
|
|
|
|
235
|
6 |
|
protected function renderImage($block) |
236
|
|
|
{ |
237
|
6 |
View Code Duplication |
if (isset($block['refkey'])) { |
238
|
4 |
|
if (($ref = $this->lookupReference($block['refkey'])) !== false) { |
239
|
|
|
$block = array_merge($block, $ref); |
240
|
|
|
} else { |
241
|
4 |
|
if (strncmp($block['orig'], '![', 2) === 0) { |
242
|
4 |
|
return '![' . $this->renderAbsy($this->parseInline(substr($block['orig'], 2))); |
243
|
|
|
} |
244
|
|
|
return $block['orig']; |
245
|
|
|
} |
246
|
|
|
} |
247
|
4 |
|
return '<img src="' . htmlspecialchars($block['url'], ENT_COMPAT | ENT_HTML401, 'UTF-8') . '"' |
248
|
4 |
|
. ' alt="' . htmlspecialchars($block['text'], ENT_COMPAT | ENT_HTML401 | ENT_SUBSTITUTE, 'UTF-8') . '"' |
249
|
4 |
|
. (empty($block['title']) ? '' : ' title="' . htmlspecialchars($block['title'], ENT_COMPAT | ENT_HTML401 | ENT_SUBSTITUTE, 'UTF-8') . '"') |
250
|
4 |
|
. ($this->html5 ? '>' : ' />'); |
251
|
|
|
} |
252
|
|
|
|
253
|
|
|
// references |
254
|
|
|
|
255
|
141 |
|
protected function identifyReference($line) |
256
|
|
|
{ |
257
|
141 |
|
return ($line[0] === ' ' || $line[0] === '[') && preg_match('/^ {0,3}\[(.+?)\]:\s*([^\s]+?)(?:\s+[\'"](.+?)[\'"])?\s*$/', $line); |
258
|
|
|
} |
259
|
|
|
|
260
|
|
|
/** |
261
|
|
|
* Consume link references |
262
|
|
|
*/ |
263
|
19 |
|
protected function consumeReference($lines, $current) |
264
|
|
|
{ |
265
|
19 |
|
while (isset($lines[$current]) && preg_match('/^ {0,3}\[(.+?)\]:\s*(.+?)(?:\s+[\(\'"](.+?)[\)\'"])?\s*$/', $lines[$current], $matches)) { |
266
|
19 |
|
$label = strtolower($matches[1]); |
267
|
|
|
|
268
|
19 |
|
$this->references[$label] = [ |
269
|
19 |
|
'url' => $this->replaceEscape($matches[2]), |
270
|
|
|
]; |
271
|
19 |
View Code Duplication |
if (isset($matches[3])) { |
272
|
9 |
|
$this->references[$label]['title'] = $matches[3]; |
273
|
|
|
} else { |
274
|
|
|
// title may be on the next line |
275
|
16 |
|
if (isset($lines[$current + 1]) && preg_match('/^\s+[\(\'"](.+?)[\)\'"]\s*$/', $lines[$current + 1], $matches)) { |
276
|
2 |
|
$this->references[$label]['title'] = $matches[1]; |
277
|
2 |
|
$current++; |
278
|
|
|
} |
279
|
|
|
} |
280
|
19 |
|
$current++; |
281
|
|
|
} |
282
|
19 |
|
return [false, --$current]; |
283
|
|
|
} |
284
|
|
|
|
285
|
|
|
abstract protected function parseInline($text); |
286
|
|
|
abstract protected function renderAbsy($blocks); |
287
|
|
|
} |
288
|
|
|
|
This check marks calls to methods that do not seem to exist on an object.
This is most likely the result of a method being renamed without all references to it being renamed likewise.