|
1
|
|
|
<?php |
|
2
|
|
|
/** |
|
3
|
|
|
* Markdown Extra - A text-to-HTML conversion tool for web writers |
|
4
|
|
|
* |
|
5
|
|
|
* @package php-markdown |
|
6
|
|
|
* @author Michel Fortin <[email protected]> |
|
7
|
|
|
* @copyright 2004-2019 Michel Fortin <https://michelf.com/projects/php-markdown/> |
|
8
|
|
|
* @copyright (Original Markdown) 2004-2006 John Gruber <https://daringfireball.net/projects/markdown/> |
|
9
|
|
|
*/ |
|
10
|
|
|
|
|
11
|
|
|
namespace Michelf; |
|
12
|
|
|
|
|
13
|
|
|
/** |
|
14
|
|
|
* Markdown Extra Parser Class |
|
15
|
|
|
*/ |
|
16
|
|
|
class MarkdownExtra extends \Michelf\Markdown { |
|
17
|
|
|
/** |
|
18
|
|
|
* Configuration variables |
|
19
|
|
|
*/ |
|
20
|
|
|
|
|
21
|
|
|
/** |
|
22
|
|
|
* Prefix for footnote ids. |
|
23
|
|
|
* @var string |
|
24
|
|
|
*/ |
|
25
|
|
|
public $fn_id_prefix = ""; |
|
26
|
|
|
|
|
27
|
|
|
/** |
|
28
|
|
|
* Optional title attribute for footnote links. |
|
29
|
|
|
* @var string |
|
30
|
|
|
*/ |
|
31
|
|
|
public $fn_link_title = ""; |
|
32
|
|
|
|
|
33
|
|
|
/** |
|
34
|
|
|
* Optional class attribute for footnote links and backlinks. |
|
35
|
|
|
* @var string |
|
36
|
|
|
*/ |
|
37
|
|
|
public $fn_link_class = "footnote-ref"; |
|
38
|
|
|
public $fn_backlink_class = "footnote-backref"; |
|
39
|
|
|
|
|
40
|
|
|
/** |
|
41
|
|
|
* Content to be displayed within footnote backlinks. The default is '↩'; |
|
42
|
|
|
* the U+FE0E on the end is a Unicode variant selector used to prevent iOS |
|
43
|
|
|
* from displaying the arrow character as an emoji. |
|
44
|
|
|
* Optionally use '^^' and '%%' to refer to the footnote number and |
|
45
|
|
|
* reference number respectively. {@see parseFootnotePlaceholders()} |
|
46
|
|
|
* @var string |
|
47
|
|
|
*/ |
|
48
|
|
|
public $fn_backlink_html = '↩︎'; |
|
49
|
|
|
|
|
50
|
|
|
/** |
|
51
|
|
|
* Optional title and aria-label attributes for footnote backlinks for |
|
52
|
|
|
* added accessibility (to ensure backlink uniqueness). |
|
53
|
|
|
* Use '^^' and '%%' to refer to the footnote number and reference number |
|
54
|
|
|
* respectively. {@see parseFootnotePlaceholders()} |
|
55
|
|
|
* @var string |
|
56
|
|
|
*/ |
|
57
|
|
|
public $fn_backlink_title = ""; |
|
58
|
|
|
public $fn_backlink_label = ""; |
|
59
|
|
|
|
|
60
|
|
|
/** |
|
61
|
|
|
* Class name for table cell alignment (%% replaced left/center/right) |
|
62
|
|
|
* For instance: 'go-%%' becomes 'go-left' or 'go-right' or 'go-center' |
|
63
|
|
|
* If empty, the align attribute is used instead of a class name. |
|
64
|
|
|
* @var string |
|
65
|
|
|
*/ |
|
66
|
|
|
public $table_align_class_tmpl = ''; |
|
67
|
|
|
|
|
68
|
|
|
/** |
|
69
|
|
|
* Optional class prefix for fenced code block. |
|
70
|
|
|
* @var string |
|
71
|
|
|
*/ |
|
72
|
|
|
public $code_class_prefix = ""; |
|
73
|
|
|
|
|
74
|
|
|
/** |
|
75
|
|
|
* Class attribute for code blocks goes on the `code` tag; |
|
76
|
|
|
* setting this to true will put attributes on the `pre` tag instead. |
|
77
|
|
|
* @var boolean |
|
78
|
|
|
*/ |
|
79
|
|
|
public $code_attr_on_pre = false; |
|
80
|
|
|
|
|
81
|
|
|
/** |
|
82
|
|
|
* Predefined abbreviations. |
|
83
|
|
|
* @var array |
|
84
|
|
|
*/ |
|
85
|
|
|
public $predef_abbr = array(); |
|
86
|
|
|
|
|
87
|
|
|
/** |
|
88
|
|
|
* Only convert atx-style headers if there's a space between the header and # |
|
89
|
|
|
* @var boolean |
|
90
|
|
|
*/ |
|
91
|
|
|
public $hashtag_protection = false; |
|
92
|
|
|
|
|
93
|
|
|
/** |
|
94
|
|
|
* Determines whether footnotes should be appended to the end of the document. |
|
95
|
|
|
* If true, footnote html can be retrieved from $this->footnotes_assembled. |
|
96
|
|
|
* @var boolean |
|
97
|
|
|
*/ |
|
98
|
|
|
public $omit_footnotes = false; |
|
99
|
|
|
|
|
100
|
|
|
|
|
101
|
|
|
/** |
|
102
|
|
|
* After parsing, the HTML for the list of footnotes appears here. |
|
103
|
|
|
* This is available only if $omit_footnotes == true. |
|
104
|
|
|
* |
|
105
|
|
|
* Note: when placing the content of `footnotes_assembled` on the page, |
|
106
|
|
|
* consider adding the attribute `role="doc-endnotes"` to the `div` or |
|
107
|
|
|
* `section` that will enclose the list of footnotes so they are |
|
108
|
|
|
* reachable to accessibility tools the same way they would be with the |
|
109
|
|
|
* default HTML output. |
|
110
|
|
|
* @var null|string |
|
111
|
|
|
*/ |
|
112
|
|
|
public $footnotes_assembled = null; |
|
113
|
|
|
|
|
114
|
|
|
/** |
|
115
|
|
|
* Parser implementation |
|
116
|
|
|
*/ |
|
117
|
|
|
|
|
118
|
|
|
/** |
|
119
|
|
|
* Constructor function. Initialize the parser object. |
|
120
|
|
|
* @return void |
|
|
|
|
|
|
121
|
|
|
*/ |
|
122
|
4 |
|
public function __construct() { |
|
123
|
|
|
// Add extra escapable characters before parent constructor |
|
124
|
|
|
// initialize the table. |
|
125
|
4 |
|
$this->escape_chars .= ':|'; |
|
126
|
|
|
|
|
127
|
|
|
// Insert extra document, block, and span transformations. |
|
128
|
|
|
// Parent constructor will do the sorting. |
|
129
|
4 |
|
$this->document_gamut += array( |
|
130
|
|
|
"doFencedCodeBlocks" => 5, |
|
131
|
|
|
"stripFootnotes" => 15, |
|
132
|
|
|
"stripAbbreviations" => 25, |
|
133
|
|
|
"appendFootnotes" => 50, |
|
134
|
|
|
); |
|
135
|
4 |
|
$this->block_gamut += array( |
|
136
|
|
|
"doFencedCodeBlocks" => 5, |
|
137
|
|
|
"doTables" => 15, |
|
138
|
|
|
"doDefLists" => 45, |
|
139
|
|
|
); |
|
140
|
4 |
|
$this->span_gamut += array( |
|
141
|
|
|
"doFootnotes" => 5, |
|
142
|
|
|
"doAbbreviations" => 70, |
|
143
|
|
|
); |
|
144
|
|
|
|
|
145
|
4 |
|
$this->enhanced_ordered_list = true; |
|
146
|
4 |
|
parent::__construct(); |
|
147
|
4 |
|
} |
|
148
|
|
|
|
|
149
|
|
|
|
|
150
|
|
|
/** |
|
151
|
|
|
* Extra variables used during extra transformations. |
|
152
|
|
|
* @var array |
|
153
|
|
|
*/ |
|
154
|
|
|
protected $footnotes = array(); |
|
155
|
|
|
protected $footnotes_ordered = array(); |
|
156
|
|
|
protected $footnotes_ref_count = array(); |
|
157
|
|
|
protected $footnotes_numbers = array(); |
|
158
|
|
|
protected $abbr_desciptions = array(); |
|
159
|
|
|
/** @var string */ |
|
160
|
|
|
protected $abbr_word_re = ''; |
|
161
|
|
|
|
|
162
|
|
|
/** |
|
163
|
|
|
* Give the current footnote number. |
|
164
|
|
|
* @var integer |
|
165
|
|
|
*/ |
|
166
|
|
|
protected $footnote_counter = 1; |
|
167
|
|
|
|
|
168
|
|
|
/** |
|
169
|
|
|
* Ref attribute for links |
|
170
|
|
|
* @var array |
|
171
|
|
|
*/ |
|
172
|
|
|
protected $ref_attr = array(); |
|
173
|
|
|
|
|
174
|
|
|
/** |
|
175
|
|
|
* Setting up Extra-specific variables. |
|
176
|
|
|
*/ |
|
177
|
62 |
|
protected function setup() { |
|
178
|
62 |
|
parent::setup(); |
|
179
|
|
|
|
|
180
|
62 |
|
$this->footnotes = array(); |
|
181
|
62 |
|
$this->footnotes_ordered = array(); |
|
182
|
62 |
|
$this->footnotes_ref_count = array(); |
|
183
|
62 |
|
$this->footnotes_numbers = array(); |
|
184
|
62 |
|
$this->abbr_desciptions = array(); |
|
185
|
62 |
|
$this->abbr_word_re = ''; |
|
186
|
62 |
|
$this->footnote_counter = 1; |
|
187
|
62 |
|
$this->footnotes_assembled = null; |
|
188
|
|
|
|
|
189
|
62 |
|
foreach ($this->predef_abbr as $abbr_word => $abbr_desc) { |
|
190
|
2 |
|
if ($this->abbr_word_re) |
|
191
|
1 |
|
$this->abbr_word_re .= '|'; |
|
192
|
2 |
|
$this->abbr_word_re .= preg_quote($abbr_word); |
|
193
|
2 |
|
$this->abbr_desciptions[$abbr_word] = trim($abbr_desc); |
|
194
|
|
|
} |
|
195
|
62 |
|
} |
|
196
|
|
|
|
|
197
|
|
|
/** |
|
198
|
|
|
* Clearing Extra-specific variables. |
|
199
|
|
|
*/ |
|
200
|
62 |
|
protected function teardown() { |
|
201
|
62 |
|
$this->footnotes = array(); |
|
202
|
62 |
|
$this->footnotes_ordered = array(); |
|
203
|
62 |
|
$this->footnotes_ref_count = array(); |
|
204
|
62 |
|
$this->footnotes_numbers = array(); |
|
205
|
62 |
|
$this->abbr_desciptions = array(); |
|
206
|
62 |
|
$this->abbr_word_re = ''; |
|
207
|
|
|
|
|
208
|
62 |
|
if ( ! $this->omit_footnotes ) |
|
209
|
62 |
|
$this->footnotes_assembled = null; |
|
210
|
|
|
|
|
211
|
62 |
|
parent::teardown(); |
|
212
|
62 |
|
} |
|
213
|
|
|
|
|
214
|
|
|
|
|
215
|
|
|
/** |
|
216
|
|
|
* Extra attribute parser |
|
217
|
|
|
*/ |
|
218
|
|
|
|
|
219
|
|
|
/** |
|
220
|
|
|
* Expression to use to catch attributes (includes the braces) |
|
221
|
|
|
* @var string |
|
222
|
|
|
*/ |
|
223
|
|
|
protected $id_class_attr_catch_re = '\{((?>[ ]*[#.a-z][-_:a-zA-Z0-9=]+){1,})[ ]*\}'; |
|
224
|
|
|
|
|
225
|
|
|
/** |
|
226
|
|
|
* Expression to use when parsing in a context when no capture is desired |
|
227
|
|
|
* @var string |
|
228
|
|
|
*/ |
|
229
|
|
|
protected $id_class_attr_nocatch_re = '\{(?>[ ]*[#.a-z][-_:a-zA-Z0-9=]+){1,}[ ]*\}'; |
|
230
|
|
|
|
|
231
|
|
|
/** |
|
232
|
|
|
* Parse attributes caught by the $this->id_class_attr_catch_re expression |
|
233
|
|
|
* and return the HTML-formatted list of attributes. |
|
234
|
|
|
* |
|
235
|
|
|
* Currently supported attributes are .class and #id. |
|
236
|
|
|
* |
|
237
|
|
|
* In addition, this method also supports supplying a default Id value, |
|
238
|
|
|
* which will be used to populate the id attribute in case it was not |
|
239
|
|
|
* overridden. |
|
240
|
|
|
* @param string $tag_name |
|
241
|
|
|
* @param string $attr |
|
242
|
|
|
* @param mixed $defaultIdValue |
|
243
|
|
|
* @param array $classes |
|
244
|
|
|
* @return string |
|
245
|
|
|
*/ |
|
246
|
29 |
|
protected function doExtraAttributes($tag_name, $attr, $defaultIdValue = null, $classes = array()) { |
|
|
|
|
|
|
247
|
29 |
|
if (empty($attr) && !$defaultIdValue && empty($classes)) { |
|
248
|
27 |
|
return ""; |
|
249
|
|
|
} |
|
250
|
|
|
|
|
251
|
|
|
// Split on components |
|
252
|
6 |
|
preg_match_all('/[#.a-z][-_:a-zA-Z0-9=]+/', $attr, $matches); |
|
253
|
6 |
|
$elements = $matches[0]; |
|
254
|
|
|
|
|
255
|
|
|
// Handle classes and IDs (only first ID taken into account) |
|
256
|
6 |
|
$attributes = array(); |
|
257
|
6 |
|
$id = false; |
|
258
|
6 |
|
foreach ($elements as $element) { |
|
259
|
4 |
|
if ($element[0] === '.') { |
|
260
|
4 |
|
$classes[] = substr($element, 1); |
|
261
|
4 |
|
} else if ($element[0] === '#') { |
|
262
|
4 |
|
if ($id === false) $id = substr($element, 1); |
|
263
|
1 |
|
} else if (strpos($element, '=') > 0) { |
|
264
|
1 |
|
$parts = explode('=', $element, 2); |
|
265
|
4 |
|
$attributes[] = $parts[0] . '="' . $parts[1] . '"'; |
|
266
|
|
|
} |
|
267
|
|
|
} |
|
268
|
|
|
|
|
269
|
6 |
|
if ($id === false || $id === '') { |
|
270
|
5 |
|
$id = $defaultIdValue; |
|
271
|
|
|
} |
|
272
|
|
|
|
|
273
|
|
|
// Compose attributes as string |
|
274
|
6 |
|
$attr_str = ""; |
|
275
|
6 |
|
if (!empty($id)) { |
|
276
|
4 |
|
$attr_str .= ' id="'.$this->encodeAttribute($id) .'"'; |
|
277
|
|
|
} |
|
278
|
6 |
|
if (!empty($classes)) { |
|
279
|
6 |
|
$attr_str .= ' class="'. implode(" ", $classes) . '"'; |
|
280
|
|
|
} |
|
281
|
6 |
|
if (!$this->no_markup && !empty($attributes)) { |
|
282
|
1 |
|
$attr_str .= ' '.implode(" ", $attributes); |
|
283
|
|
|
} |
|
284
|
6 |
|
return $attr_str; |
|
285
|
|
|
} |
|
286
|
|
|
|
|
287
|
|
|
/** |
|
288
|
|
|
* Strips link definitions from text, stores the URLs and titles in |
|
289
|
|
|
* hash references. |
|
290
|
|
|
* @param string $text |
|
291
|
|
|
* @return string |
|
292
|
|
|
*/ |
|
293
|
62 |
View Code Duplication |
protected function stripLinkDefinitions($text) { |
|
|
|
|
|
|
294
|
62 |
|
$less_than_tab = $this->tab_width - 1; |
|
295
|
|
|
|
|
296
|
|
|
// Link defs are in the form: ^[id]: url "optional title" |
|
297
|
62 |
|
$text = preg_replace_callback('{ |
|
298
|
62 |
|
^[ ]{0,'.$less_than_tab.'}\[(.+)\][ ]?: # id = $1 |
|
299
|
|
|
[ ]* |
|
300
|
|
|
\n? # maybe *one* newline |
|
301
|
|
|
[ ]* |
|
302
|
|
|
(?: |
|
303
|
|
|
<(.+?)> # url = $2 |
|
304
|
|
|
| |
|
305
|
|
|
(\S+?) # url = $3 |
|
306
|
|
|
) |
|
307
|
|
|
[ ]* |
|
308
|
|
|
\n? # maybe one newline |
|
309
|
|
|
[ ]* |
|
310
|
|
|
(?: |
|
311
|
|
|
(?<=\s) # lookbehind for whitespace |
|
312
|
|
|
["(] |
|
313
|
|
|
(.*?) # title = $4 |
|
314
|
|
|
[")] |
|
315
|
|
|
[ ]* |
|
316
|
|
|
)? # title is optional |
|
317
|
62 |
|
(?:[ ]* '.$this->id_class_attr_catch_re.' )? # $5 = extra id & class attr |
|
318
|
|
|
(?:\n+|\Z) |
|
319
|
|
|
}xm', |
|
320
|
62 |
|
array($this, '_stripLinkDefinitions_callback'), |
|
321
|
62 |
|
$text); |
|
322
|
62 |
|
return $text; |
|
323
|
|
|
} |
|
324
|
|
|
|
|
325
|
|
|
/** |
|
326
|
|
|
* Strip link definition callback |
|
327
|
|
|
* @param array $matches |
|
328
|
|
|
* @return string |
|
329
|
|
|
*/ |
|
330
|
11 |
|
protected function _stripLinkDefinitions_callback($matches) { |
|
331
|
11 |
|
$link_id = strtolower($matches[1]); |
|
332
|
11 |
|
$url = $matches[2] == '' ? $matches[3] : $matches[2]; |
|
333
|
11 |
|
$this->urls[$link_id] = $url; |
|
334
|
11 |
|
$this->titles[$link_id] =& $matches[4]; |
|
335
|
11 |
|
$this->ref_attr[$link_id] = $this->doExtraAttributes("", $dummy =& $matches[5]); |
|
336
|
11 |
|
return ''; // String that will replace the block |
|
337
|
|
|
} |
|
338
|
|
|
|
|
339
|
|
|
|
|
340
|
|
|
/** |
|
341
|
|
|
* HTML block parser |
|
342
|
|
|
*/ |
|
343
|
|
|
|
|
344
|
|
|
/** |
|
345
|
|
|
* Tags that are always treated as block tags |
|
346
|
|
|
* @var string |
|
347
|
|
|
*/ |
|
348
|
|
|
protected $block_tags_re = 'p|div|h[1-6]|blockquote|pre|table|dl|ol|ul|address|form|fieldset|iframe|hr|legend|article|section|nav|aside|hgroup|header|footer|figcaption|figure|details|summary'; |
|
349
|
|
|
|
|
350
|
|
|
/** |
|
351
|
|
|
* Tags treated as block tags only if the opening tag is alone on its line |
|
352
|
|
|
* @var string |
|
353
|
|
|
*/ |
|
354
|
|
|
protected $context_block_tags_re = 'script|noscript|style|ins|del|iframe|object|source|track|param|math|svg|canvas|audio|video'; |
|
355
|
|
|
|
|
356
|
|
|
/** |
|
357
|
|
|
* Tags where markdown="1" default to span mode: |
|
358
|
|
|
* @var string |
|
359
|
|
|
*/ |
|
360
|
|
|
protected $contain_span_tags_re = 'p|h[1-6]|li|dd|dt|td|th|legend|address'; |
|
361
|
|
|
|
|
362
|
|
|
/** |
|
363
|
|
|
* Tags which must not have their contents modified, no matter where |
|
364
|
|
|
* they appear |
|
365
|
|
|
* @var string |
|
366
|
|
|
*/ |
|
367
|
|
|
protected $clean_tags_re = 'script|style|math|svg'; |
|
368
|
|
|
|
|
369
|
|
|
/** |
|
370
|
|
|
* Tags that do not need to be closed. |
|
371
|
|
|
* @var string |
|
372
|
|
|
*/ |
|
373
|
|
|
protected $auto_close_tags_re = 'hr|img|param|source|track'; |
|
374
|
|
|
|
|
375
|
|
|
/** |
|
376
|
|
|
* Hashify HTML Blocks and "clean tags". |
|
377
|
|
|
* |
|
378
|
|
|
* We only want to do this for block-level HTML tags, such as headers, |
|
379
|
|
|
* lists, and tables. That's because we still want to wrap <p>s around |
|
380
|
|
|
* "paragraphs" that are wrapped in non-block-level tags, such as anchors, |
|
381
|
|
|
* phrase emphasis, and spans. The list of tags we're looking for is |
|
382
|
|
|
* hard-coded. |
|
383
|
|
|
* |
|
384
|
|
|
* This works by calling _HashHTMLBlocks_InMarkdown, which then calls |
|
385
|
|
|
* _HashHTMLBlocks_InHTML when it encounter block tags. When the markdown="1" |
|
386
|
|
|
* attribute is found within a tag, _HashHTMLBlocks_InHTML calls back |
|
387
|
|
|
* _HashHTMLBlocks_InMarkdown to handle the Markdown syntax within the tag. |
|
388
|
|
|
* These two functions are calling each other. It's recursive! |
|
389
|
|
|
* @param string $text |
|
390
|
|
|
* @return string |
|
391
|
|
|
*/ |
|
392
|
62 |
|
protected function hashHTMLBlocks($text) { |
|
393
|
62 |
|
if ($this->no_markup) { |
|
394
|
1 |
|
return $text; |
|
395
|
|
|
} |
|
396
|
|
|
|
|
397
|
|
|
// Call the HTML-in-Markdown hasher. |
|
398
|
61 |
|
list($text, ) = $this->_hashHTMLBlocks_inMarkdown($text); |
|
399
|
|
|
|
|
400
|
61 |
|
return $text; |
|
401
|
|
|
} |
|
402
|
|
|
|
|
403
|
|
|
/** |
|
404
|
|
|
* Parse markdown text, calling _HashHTMLBlocks_InHTML for block tags. |
|
405
|
|
|
* |
|
406
|
|
|
* * $indent is the number of space to be ignored when checking for code |
|
407
|
|
|
* blocks. This is important because if we don't take the indent into |
|
408
|
|
|
* account, something like this (which looks right) won't work as expected: |
|
409
|
|
|
* |
|
410
|
|
|
* <div> |
|
411
|
|
|
* <div markdown="1"> |
|
412
|
|
|
* Hello World. <-- Is this a Markdown code block or text? |
|
413
|
|
|
* </div> <-- Is this a Markdown code block or a real tag? |
|
414
|
|
|
* <div> |
|
415
|
|
|
* |
|
416
|
|
|
* If you don't like this, just don't indent the tag on which |
|
417
|
|
|
* you apply the markdown="1" attribute. |
|
418
|
|
|
* |
|
419
|
|
|
* * If $enclosing_tag_re is not empty, stops at the first unmatched closing |
|
420
|
|
|
* tag with that name. Nested tags supported. |
|
421
|
|
|
* |
|
422
|
|
|
* * If $span is true, text inside must treated as span. So any double |
|
423
|
|
|
* newline will be replaced by a single newline so that it does not create |
|
424
|
|
|
* paragraphs. |
|
425
|
|
|
* |
|
426
|
|
|
* Returns an array of that form: ( processed text , remaining text ) |
|
427
|
|
|
* |
|
428
|
|
|
* @param string $text |
|
429
|
|
|
* @param integer $indent |
|
430
|
|
|
* @param string $enclosing_tag_re |
|
431
|
|
|
* @param boolean $span |
|
432
|
|
|
* @return array |
|
433
|
|
|
*/ |
|
434
|
61 |
|
protected function _hashHTMLBlocks_inMarkdown($text, $indent = 0, |
|
435
|
|
|
$enclosing_tag_re = '', $span = false) |
|
436
|
|
|
{ |
|
437
|
|
|
|
|
438
|
61 |
|
if ($text === '') return array('', ''); |
|
439
|
|
|
|
|
440
|
|
|
// Regex to check for the presense of newlines around a block tag. |
|
441
|
61 |
|
$newline_before_re = '/(?:^\n?|\n\n)*$/'; |
|
442
|
|
|
$newline_after_re = |
|
443
|
61 |
|
'{ |
|
444
|
|
|
^ # Start of text following the tag. |
|
445
|
|
|
(?>[ ]*<!--.*?-->)? # Optional comment. |
|
446
|
|
|
[ ]*\n # Must be followed by newline. |
|
447
|
|
|
}xs'; |
|
448
|
|
|
|
|
449
|
|
|
// Regex to match any tag. |
|
450
|
|
|
$block_tag_re = |
|
451
|
|
|
'{ |
|
452
|
|
|
( # $2: Capture whole tag. |
|
453
|
|
|
</? # Any opening or closing tag. |
|
454
|
|
|
(?> # Tag name. |
|
455
|
61 |
|
' . $this->block_tags_re . ' | |
|
456
|
61 |
|
' . $this->context_block_tags_re . ' | |
|
457
|
61 |
|
' . $this->clean_tags_re . ' | |
|
458
|
61 |
|
(?!\s)'.$enclosing_tag_re . ' |
|
459
|
|
|
) |
|
460
|
|
|
(?: |
|
461
|
|
|
(?=[\s"\'/a-zA-Z0-9]) # Allowed characters after tag name. |
|
462
|
|
|
(?> |
|
463
|
|
|
".*?" | # Double quotes (can contain `>`) |
|
464
|
|
|
\'.*?\' | # Single quotes (can contain `>`) |
|
465
|
|
|
.+? # Anything but quotes and `>`. |
|
466
|
|
|
)*? |
|
467
|
|
|
)? |
|
468
|
|
|
> # End of tag. |
|
469
|
|
|
| |
|
470
|
|
|
<!-- .*? --> # HTML Comment |
|
471
|
|
|
| |
|
472
|
|
|
<\?.*?\?> | <%.*?%> # Processing instruction |
|
473
|
|
|
| |
|
474
|
|
|
<!\[CDATA\[.*?\]\]> # CData Block |
|
475
|
61 |
|
' . ( !$span ? ' # If not in span. |
|
476
|
|
|
| |
|
477
|
|
|
# Indented code block |
|
478
|
|
|
(?: ^[ ]*\n | ^ | \n[ ]*\n ) |
|
479
|
61 |
|
[ ]{' . ($indent + 4) . '}[^\n]* \n |
|
480
|
|
|
(?> |
|
481
|
61 |
|
(?: [ ]{' . ($indent + 4) . '}[^\n]* | [ ]* ) \n |
|
482
|
|
|
)* |
|
483
|
|
|
| |
|
484
|
|
|
# Fenced code block marker |
|
485
|
|
|
(?<= ^ | \n ) |
|
486
|
61 |
|
[ ]{0,' . ($indent + 3) . '}(?:~{3,}|`{3,}) |
|
487
|
|
|
[ ]* |
|
488
|
|
|
(?: \.?[-_:a-zA-Z0-9]+ )? # standalone class name |
|
489
|
|
|
[ ]* |
|
490
|
61 |
|
(?: ' . $this->id_class_attr_nocatch_re . ' )? # extra attributes |
|
491
|
|
|
[ ]* |
|
492
|
|
|
(?= \n ) |
|
493
|
61 |
|
' : '' ) . ' # End (if not is span). |
|
494
|
|
|
| |
|
495
|
|
|
# Code span marker |
|
496
|
|
|
# Note, this regex needs to go after backtick fenced |
|
497
|
|
|
# code blocks but it should also be kept outside of the |
|
498
|
|
|
# "if not in span" condition adding backticks to the parser |
|
499
|
|
|
`+ |
|
500
|
|
|
) |
|
501
|
|
|
}xs'; |
|
502
|
|
|
|
|
503
|
|
|
|
|
504
|
61 |
|
$depth = 0; // Current depth inside the tag tree. |
|
505
|
61 |
|
$parsed = ""; // Parsed text that will be returned. |
|
506
|
|
|
|
|
507
|
|
|
// Loop through every tag until we find the closing tag of the parent |
|
508
|
|
|
// or loop until reaching the end of text if no parent tag specified. |
|
509
|
|
|
do { |
|
510
|
|
|
// Split the text using the first $tag_match pattern found. |
|
511
|
|
|
// Text before pattern will be first in the array, text after |
|
512
|
|
|
// pattern will be at the end, and between will be any catches made |
|
513
|
|
|
// by the pattern. |
|
514
|
61 |
|
$parts = preg_split($block_tag_re, $text, 2, |
|
515
|
61 |
|
PREG_SPLIT_DELIM_CAPTURE); |
|
516
|
|
|
|
|
517
|
|
|
// If in Markdown span mode, add a empty-string span-level hash |
|
518
|
|
|
// after each newline to prevent triggering any block element. |
|
519
|
61 |
|
if ($span) { |
|
520
|
1 |
|
$void = $this->hashPart("", ':'); |
|
521
|
1 |
|
$newline = "\n$void"; |
|
522
|
1 |
|
$parts[0] = $void . str_replace("\n", $newline, $parts[0]) . $void; |
|
523
|
|
|
} |
|
524
|
|
|
|
|
525
|
61 |
|
$parsed .= $parts[0]; // Text before current tag. |
|
526
|
|
|
|
|
527
|
|
|
// If end of $text has been reached. Stop loop. |
|
528
|
61 |
|
if (count($parts) < 3) { |
|
529
|
61 |
|
$text = ""; |
|
530
|
61 |
|
break; |
|
531
|
|
|
} |
|
532
|
|
|
|
|
533
|
39 |
|
$tag = $parts[1]; // Tag to handle. |
|
534
|
39 |
|
$text = $parts[2]; // Remaining text after current tag. |
|
535
|
|
|
|
|
536
|
|
|
// Check for: Fenced code block marker. |
|
537
|
|
|
// Note: need to recheck the whole tag to disambiguate backtick |
|
538
|
|
|
// fences from code spans |
|
539
|
39 |
|
if (preg_match('{^\n?([ ]{0,' . ($indent + 3) . '})(~{3,}|`{3,})[ ]*(?:\.?[-_:a-zA-Z0-9]+)?[ ]*(?:' . $this->id_class_attr_nocatch_re . ')?[ ]*\n?$}', $tag, $capture)) { |
|
540
|
|
|
// Fenced code block marker: find matching end marker. |
|
541
|
4 |
|
$fence_indent = strlen($capture[1]); // use captured indent in re |
|
542
|
4 |
|
$fence_re = $capture[2]; // use captured fence in re |
|
543
|
4 |
View Code Duplication |
if (preg_match('{^(?>.*\n)*?[ ]{' . ($fence_indent) . '}' . $fence_re . '[ ]*(?:\n|$)}', $text, |
|
|
|
|
|
|
544
|
4 |
|
$matches)) |
|
545
|
|
|
{ |
|
546
|
|
|
// End marker found: pass text unchanged until marker. |
|
547
|
4 |
|
$parsed .= $tag . $matches[0]; |
|
548
|
4 |
|
$text = substr($text, strlen($matches[0])); |
|
549
|
|
|
} |
|
550
|
|
|
else { |
|
551
|
|
|
// No end marker: just skip it. |
|
552
|
4 |
|
$parsed .= $tag; |
|
553
|
|
|
} |
|
554
|
|
|
} |
|
555
|
|
|
// Check for: Indented code block. |
|
556
|
39 |
|
else if ($tag[0] === "\n" || $tag[0] === " ") { |
|
557
|
|
|
// Indented code block: pass it unchanged, will be handled |
|
558
|
|
|
// later. |
|
559
|
23 |
|
$parsed .= $tag; |
|
560
|
|
|
} |
|
561
|
|
|
// Check for: Code span marker |
|
562
|
|
|
// Note: need to check this after backtick fenced code blocks |
|
563
|
27 |
|
else if ($tag[0] === "`") { |
|
564
|
|
|
// Find corresponding end marker. |
|
565
|
13 |
|
$tag_re = preg_quote($tag); |
|
566
|
13 |
View Code Duplication |
if (preg_match('{^(?>.+?|\n(?!\n))*?(?<!`)' . $tag_re . '(?!`)}', |
|
|
|
|
|
|
567
|
13 |
|
$text, $matches)) |
|
568
|
|
|
{ |
|
569
|
|
|
// End marker found: pass text unchanged until marker. |
|
570
|
12 |
|
$parsed .= $tag . $matches[0]; |
|
571
|
12 |
|
$text = substr($text, strlen($matches[0])); |
|
572
|
|
|
} |
|
573
|
|
|
else { |
|
574
|
|
|
// Unmatched marker: just skip it. |
|
575
|
13 |
|
$parsed .= $tag; |
|
576
|
|
|
} |
|
577
|
|
|
} |
|
578
|
|
|
// Check for: Opening Block level tag or |
|
579
|
|
|
// Opening Context Block tag (like ins and del) |
|
580
|
|
|
// used as a block tag (tag is alone on it's line). |
|
581
|
24 |
|
else if (preg_match('{^<(?:' . $this->block_tags_re . ')\b}', $tag) || |
|
582
|
19 |
|
( preg_match('{^<(?:' . $this->context_block_tags_re . ')\b}', $tag) && |
|
583
|
19 |
|
preg_match($newline_before_re, $parsed) && |
|
584
|
24 |
|
preg_match($newline_after_re, $text) ) |
|
585
|
|
|
) |
|
586
|
|
|
{ |
|
587
|
|
|
// Need to parse tag and following text using the HTML parser. |
|
588
|
|
|
list($block_text, $text) = |
|
589
|
10 |
|
$this->_hashHTMLBlocks_inHTML($tag . $text, "hashBlock", true); |
|
590
|
|
|
|
|
591
|
|
|
// Make sure it stays outside of any paragraph by adding newlines. |
|
592
|
10 |
|
$parsed .= "\n\n$block_text\n\n"; |
|
593
|
|
|
} |
|
594
|
|
|
// Check for: Clean tag (like script, math) |
|
595
|
|
|
// HTML Comments, processing instructions. |
|
596
|
19 |
|
else if (preg_match('{^<(?:' . $this->clean_tags_re . ')\b}', $tag) || |
|
597
|
19 |
|
$tag[1] === '!' || $tag[1] === '?') |
|
598
|
|
|
{ |
|
599
|
|
|
// Need to parse tag and following text using the HTML parser. |
|
600
|
|
|
// (don't check for markdown attribute) |
|
601
|
|
|
list($block_text, $text) = |
|
602
|
3 |
|
$this->_hashHTMLBlocks_inHTML($tag . $text, "hashClean", false); |
|
603
|
|
|
|
|
604
|
3 |
|
$parsed .= $block_text; |
|
605
|
|
|
} |
|
606
|
|
|
// Check for: Tag with same name as enclosing tag. |
|
607
|
16 |
|
else if ($enclosing_tag_re !== '' && |
|
608
|
|
|
// Same name as enclosing tag. |
|
609
|
16 |
|
preg_match('{^</?(?:' . $enclosing_tag_re . ')\b}', $tag)) |
|
610
|
|
|
{ |
|
611
|
|
|
// Increase/decrease nested tag count. |
|
612
|
3 |
View Code Duplication |
if ($tag[1] === '/') { |
|
|
|
|
|
|
613
|
3 |
|
$depth--; |
|
614
|
|
|
} else if ($tag[strlen($tag)-2] !== '/') { |
|
615
|
|
|
$depth++; |
|
616
|
|
|
} |
|
617
|
|
|
|
|
618
|
3 |
|
if ($depth < 0) { |
|
619
|
|
|
// Going out of parent element. Clean up and break so we |
|
620
|
|
|
// return to the calling function. |
|
621
|
3 |
|
$text = $tag . $text; |
|
622
|
3 |
|
break; |
|
623
|
|
|
} |
|
624
|
|
|
|
|
625
|
|
|
$parsed .= $tag; |
|
626
|
|
|
} |
|
627
|
|
|
else { |
|
628
|
13 |
|
$parsed .= $tag; |
|
629
|
|
|
} |
|
630
|
39 |
|
} while ($depth >= 0); |
|
631
|
|
|
|
|
632
|
61 |
|
return array($parsed, $text); |
|
633
|
|
|
} |
|
634
|
|
|
|
|
635
|
|
|
/** |
|
636
|
|
|
* Parse HTML, calling _HashHTMLBlocks_InMarkdown for block tags. |
|
637
|
|
|
* |
|
638
|
|
|
* * Calls $hash_method to convert any blocks. |
|
639
|
|
|
* * Stops when the first opening tag closes. |
|
640
|
|
|
* * $md_attr indicate if the use of the `markdown="1"` attribute is allowed. |
|
641
|
|
|
* (it is not inside clean tags) |
|
642
|
|
|
* |
|
643
|
|
|
* Returns an array of that form: ( processed text , remaining text ) |
|
644
|
|
|
* @param string $text |
|
645
|
|
|
* @param string $hash_method |
|
646
|
|
|
* @param bool $md_attr Handle `markdown="1"` attribute |
|
647
|
|
|
* @return array |
|
648
|
|
|
*/ |
|
649
|
12 |
|
protected function _hashHTMLBlocks_inHTML($text, $hash_method, $md_attr) { |
|
650
|
12 |
|
if ($text === '') return array('', ''); |
|
651
|
|
|
|
|
652
|
|
|
// Regex to match `markdown` attribute inside of a tag. |
|
653
|
12 |
|
$markdown_attr_re = ' |
|
654
|
|
|
{ |
|
655
|
|
|
\s* # Eat whitespace before the `markdown` attribute |
|
656
|
|
|
markdown |
|
657
|
|
|
\s*=\s* |
|
658
|
|
|
(?> |
|
659
|
|
|
(["\']) # $1: quote delimiter |
|
660
|
|
|
(.*?) # $2: attribute value |
|
661
|
|
|
\1 # matching delimiter |
|
662
|
|
|
| |
|
663
|
|
|
([^\s>]*) # $3: unquoted attribute value |
|
664
|
|
|
) |
|
665
|
|
|
() # $4: make $3 always defined (avoid warnings) |
|
666
|
|
|
}xs'; |
|
667
|
|
|
|
|
668
|
|
|
// Regex to match any tag. |
|
669
|
12 |
|
$tag_re = '{ |
|
670
|
|
|
( # $2: Capture whole tag. |
|
671
|
|
|
</? # Any opening or closing tag. |
|
672
|
|
|
[\w:$]+ # Tag name. |
|
673
|
|
|
(?: |
|
674
|
|
|
(?=[\s"\'/a-zA-Z0-9]) # Allowed characters after tag name. |
|
675
|
|
|
(?> |
|
676
|
|
|
".*?" | # Double quotes (can contain `>`) |
|
677
|
|
|
\'.*?\' | # Single quotes (can contain `>`) |
|
678
|
|
|
.+? # Anything but quotes and `>`. |
|
679
|
|
|
)*? |
|
680
|
|
|
)? |
|
681
|
|
|
> # End of tag. |
|
682
|
|
|
| |
|
683
|
|
|
<!-- .*? --> # HTML Comment |
|
684
|
|
|
| |
|
685
|
|
|
<\?.*?\?> | <%.*?%> # Processing instruction |
|
686
|
|
|
| |
|
687
|
|
|
<!\[CDATA\[.*?\]\]> # CData Block |
|
688
|
|
|
) |
|
689
|
|
|
}xs'; |
|
690
|
|
|
|
|
691
|
12 |
|
$original_text = $text; // Save original text in case of faliure. |
|
692
|
|
|
|
|
693
|
12 |
|
$depth = 0; // Current depth inside the tag tree. |
|
694
|
12 |
|
$block_text = ""; // Temporary text holder for current text. |
|
695
|
12 |
|
$parsed = ""; // Parsed text that will be returned. |
|
696
|
12 |
|
$base_tag_name_re = ''; |
|
697
|
|
|
|
|
698
|
|
|
// Get the name of the starting tag. |
|
699
|
|
|
// (This pattern makes $base_tag_name_re safe without quoting.) |
|
700
|
12 |
|
if (preg_match('/^<([\w:$]*)\b/', $text, $matches)) |
|
701
|
10 |
|
$base_tag_name_re = $matches[1]; |
|
702
|
|
|
|
|
703
|
|
|
// Loop through every tag until we find the corresponding closing tag. |
|
704
|
|
|
do { |
|
705
|
|
|
// Split the text using the first $tag_match pattern found. |
|
706
|
|
|
// Text before pattern will be first in the array, text after |
|
707
|
|
|
// pattern will be at the end, and between will be any catches made |
|
708
|
|
|
// by the pattern. |
|
709
|
12 |
|
$parts = preg_split($tag_re, $text, 2, PREG_SPLIT_DELIM_CAPTURE); |
|
710
|
|
|
|
|
711
|
12 |
|
if (count($parts) < 3) { |
|
712
|
|
|
// End of $text reached with unbalenced tag(s). |
|
713
|
|
|
// In that case, we return original text unchanged and pass the |
|
714
|
|
|
// first character as filtered to prevent an infinite loop in the |
|
715
|
|
|
// parent function. |
|
716
|
|
|
return array($original_text[0], substr($original_text, 1)); |
|
717
|
|
|
} |
|
718
|
|
|
|
|
719
|
12 |
|
$block_text .= $parts[0]; // Text before current tag. |
|
720
|
12 |
|
$tag = $parts[1]; // Tag to handle. |
|
721
|
12 |
|
$text = $parts[2]; // Remaining text after current tag. |
|
722
|
|
|
|
|
723
|
|
|
// Check for: Auto-close tag (like <hr/>) |
|
724
|
|
|
// Comments and Processing Instructions. |
|
725
|
12 |
|
if (preg_match('{^</?(?:' . $this->auto_close_tags_re . ')\b}', $tag) || |
|
726
|
12 |
|
$tag[1] === '!' || $tag[1] === '?') |
|
727
|
|
|
{ |
|
728
|
|
|
// Just add the tag to the block as if it was text. |
|
729
|
4 |
|
$block_text .= $tag; |
|
730
|
|
|
} |
|
731
|
|
|
else { |
|
732
|
|
|
// Increase/decrease nested tag count. Only do so if |
|
733
|
|
|
// the tag's name match base tag's. |
|
734
|
10 |
|
if (preg_match('{^</?' . $base_tag_name_re . '\b}', $tag)) { |
|
735
|
10 |
View Code Duplication |
if ($tag[1] === '/') { |
|
|
|
|
|
|
736
|
10 |
|
$depth--; |
|
737
|
10 |
|
} else if ($tag[strlen($tag)-2] !== '/') { |
|
738
|
10 |
|
$depth++; |
|
739
|
|
|
} |
|
740
|
|
|
} |
|
741
|
|
|
|
|
742
|
|
|
// Check for `markdown="1"` attribute and handle it. |
|
743
|
10 |
|
if ($md_attr && |
|
744
|
10 |
|
preg_match($markdown_attr_re, $tag, $attr_m) && |
|
745
|
10 |
|
preg_match('/^1|block|span$/', $attr_m[2] . $attr_m[3])) |
|
746
|
|
|
{ |
|
747
|
|
|
// Remove `markdown` attribute from opening tag. |
|
748
|
3 |
|
$tag = preg_replace($markdown_attr_re, '', $tag); |
|
749
|
|
|
|
|
750
|
|
|
// Check if text inside this tag must be parsed in span mode. |
|
751
|
3 |
|
$mode = $attr_m[2] . $attr_m[3]; |
|
752
|
3 |
|
$span_mode = $mode === 'span' || ($mode !== 'block' && |
|
753
|
3 |
|
preg_match('{^<(?:' . $this->contain_span_tags_re . ')\b}', $tag)); |
|
754
|
|
|
|
|
755
|
|
|
// Calculate indent before tag. |
|
756
|
3 |
|
if (preg_match('/(?:^|\n)( *?)(?! ).*?$/', $block_text, $matches)) { |
|
757
|
3 |
|
$strlen = $this->utf8_strlen; |
|
758
|
3 |
|
$indent = $strlen($matches[1], 'UTF-8'); |
|
759
|
|
|
} else { |
|
760
|
|
|
$indent = 0; |
|
761
|
|
|
} |
|
762
|
|
|
|
|
763
|
|
|
// End preceding block with this tag. |
|
764
|
3 |
|
$block_text .= $tag; |
|
765
|
3 |
|
$parsed .= $this->$hash_method($block_text); |
|
766
|
|
|
|
|
767
|
|
|
// Get enclosing tag name for the ParseMarkdown function. |
|
768
|
|
|
// (This pattern makes $tag_name_re safe without quoting.) |
|
769
|
3 |
|
preg_match('/^<([\w:$]*)\b/', $tag, $matches); |
|
770
|
3 |
|
$tag_name_re = $matches[1]; |
|
771
|
|
|
|
|
772
|
|
|
// Parse the content using the HTML-in-Markdown parser. |
|
773
|
|
|
list ($block_text, $text) |
|
774
|
3 |
|
= $this->_hashHTMLBlocks_inMarkdown($text, $indent, |
|
775
|
3 |
|
$tag_name_re, $span_mode); |
|
776
|
|
|
|
|
777
|
|
|
// Outdent markdown text. |
|
778
|
3 |
|
if ($indent > 0) { |
|
779
|
1 |
|
$block_text = preg_replace("/^[ ]{1,$indent}/m", "", |
|
780
|
1 |
|
$block_text); |
|
781
|
|
|
} |
|
782
|
|
|
|
|
783
|
|
|
// Append tag content to parsed text. |
|
784
|
3 |
|
if (!$span_mode) { |
|
785
|
3 |
|
$parsed .= "\n\n$block_text\n\n"; |
|
786
|
|
|
} else { |
|
787
|
1 |
|
$parsed .= (string) $block_text; |
|
788
|
|
|
} |
|
789
|
|
|
|
|
790
|
|
|
// Start over with a new block. |
|
791
|
3 |
|
$block_text = ""; |
|
792
|
|
|
} |
|
793
|
10 |
|
else $block_text .= $tag; |
|
794
|
|
|
} |
|
795
|
|
|
|
|
796
|
12 |
|
} while ($depth > 0); |
|
797
|
|
|
|
|
798
|
|
|
// Hash last block text that wasn't processed inside the loop. |
|
799
|
12 |
|
$parsed .= $this->$hash_method($block_text); |
|
800
|
|
|
|
|
801
|
12 |
|
return array($parsed, $text); |
|
802
|
|
|
} |
|
803
|
|
|
|
|
804
|
|
|
/** |
|
805
|
|
|
* Called whenever a tag must be hashed when a function inserts a "clean" tag |
|
806
|
|
|
* in $text, it passes through this function and is automaticaly escaped, |
|
807
|
|
|
* blocking invalid nested overlap. |
|
808
|
|
|
* @param string $text |
|
809
|
|
|
* @return string |
|
810
|
|
|
*/ |
|
811
|
3 |
|
protected function hashClean($text) { |
|
812
|
3 |
|
return $this->hashPart($text, 'C'); |
|
813
|
|
|
} |
|
814
|
|
|
|
|
815
|
|
|
/** |
|
816
|
|
|
* Turn Markdown link shortcuts into XHTML <a> tags. |
|
817
|
|
|
* @param string $text |
|
818
|
|
|
* @return string |
|
819
|
|
|
*/ |
|
820
|
62 |
View Code Duplication |
protected function doAnchors($text) { |
|
|
|
|
|
|
821
|
62 |
|
if ($this->in_anchor) { |
|
822
|
15 |
|
return $text; |
|
823
|
|
|
} |
|
824
|
62 |
|
$this->in_anchor = true; |
|
825
|
|
|
|
|
826
|
|
|
// First, handle reference-style links: [link text] [id] |
|
827
|
62 |
|
$text = preg_replace_callback('{ |
|
828
|
|
|
( # wrap whole match in $1 |
|
829
|
|
|
\[ |
|
830
|
62 |
|
(' . $this->nested_brackets_re . ') # link text = $2 |
|
831
|
|
|
\] |
|
832
|
|
|
|
|
833
|
|
|
[ ]? # one optional space |
|
834
|
|
|
(?:\n[ ]*)? # one optional newline followed by spaces |
|
835
|
|
|
|
|
836
|
|
|
\[ |
|
837
|
|
|
(.*?) # id = $3 |
|
838
|
|
|
\] |
|
839
|
|
|
) |
|
840
|
|
|
}xs', |
|
841
|
62 |
|
array($this, '_doAnchors_reference_callback'), $text); |
|
842
|
|
|
|
|
843
|
|
|
// Next, inline-style links: [link text](url "optional title") |
|
844
|
62 |
|
$text = preg_replace_callback('{ |
|
845
|
|
|
( # wrap whole match in $1 |
|
846
|
|
|
\[ |
|
847
|
62 |
|
(' . $this->nested_brackets_re . ') # link text = $2 |
|
848
|
|
|
\] |
|
849
|
|
|
\( # literal paren |
|
850
|
|
|
[ \n]* |
|
851
|
|
|
(?: |
|
852
|
|
|
<(.+?)> # href = $3 |
|
853
|
|
|
| |
|
854
|
62 |
|
(' . $this->nested_url_parenthesis_re . ') # href = $4 |
|
855
|
|
|
) |
|
856
|
|
|
[ \n]* |
|
857
|
|
|
( # $5 |
|
858
|
|
|
([\'"]) # quote char = $6 |
|
859
|
|
|
(.*?) # Title = $7 |
|
860
|
|
|
\6 # matching quote |
|
861
|
|
|
[ \n]* # ignore any spaces/tabs between closing quote and ) |
|
862
|
|
|
)? # title is optional |
|
863
|
|
|
\) |
|
864
|
62 |
|
(?:[ ]? ' . $this->id_class_attr_catch_re . ' )? # $8 = id/class attributes |
|
865
|
|
|
) |
|
866
|
|
|
}xs', |
|
867
|
62 |
|
array($this, '_doAnchors_inline_callback'), $text); |
|
868
|
|
|
|
|
869
|
|
|
// Last, handle reference-style shortcuts: [link text] |
|
870
|
|
|
// These must come last in case you've also got [link text][1] |
|
871
|
|
|
// or [link text](/foo) |
|
872
|
62 |
|
$text = preg_replace_callback('{ |
|
873
|
|
|
( # wrap whole match in $1 |
|
874
|
|
|
\[ |
|
875
|
|
|
([^\[\]]+) # link text = $2; can\'t contain [ or ] |
|
876
|
|
|
\] |
|
877
|
|
|
) |
|
878
|
|
|
}xs', |
|
879
|
62 |
|
array($this, '_doAnchors_reference_callback'), $text); |
|
880
|
|
|
|
|
881
|
62 |
|
$this->in_anchor = false; |
|
882
|
62 |
|
return $text; |
|
883
|
|
|
} |
|
884
|
|
|
|
|
885
|
|
|
/** |
|
886
|
|
|
* Callback for reference anchors |
|
887
|
|
|
* @param array $matches |
|
888
|
|
|
* @return string |
|
889
|
|
|
*/ |
|
890
|
11 |
|
protected function _doAnchors_reference_callback($matches) { |
|
891
|
11 |
|
$whole_match = $matches[1]; |
|
892
|
11 |
|
$link_text = $matches[2]; |
|
893
|
11 |
|
$link_id =& $matches[3]; |
|
894
|
|
|
|
|
895
|
11 |
|
if ($link_id == "") { |
|
896
|
|
|
// for shortcut links like [this][] or [this]. |
|
897
|
7 |
|
$link_id = $link_text; |
|
898
|
|
|
} |
|
899
|
|
|
|
|
900
|
|
|
// lower-case and turn embedded newlines into spaces |
|
901
|
11 |
|
$link_id = strtolower($link_id); |
|
902
|
11 |
|
$link_id = preg_replace('{[ ]?\n}', ' ', $link_id); |
|
903
|
|
|
|
|
904
|
11 |
|
if (isset($this->urls[$link_id])) { |
|
905
|
10 |
|
$url = $this->urls[$link_id]; |
|
906
|
10 |
|
$url = $this->encodeURLAttribute($url); |
|
907
|
|
|
|
|
908
|
10 |
|
$result = "<a href=\"$url\""; |
|
909
|
10 |
View Code Duplication |
if ( isset( $this->titles[$link_id] ) ) { |
|
|
|
|
|
|
910
|
7 |
|
$title = $this->titles[$link_id]; |
|
911
|
7 |
|
$title = $this->encodeAttribute($title); |
|
912
|
7 |
|
$result .= " title=\"$title\""; |
|
913
|
|
|
} |
|
914
|
10 |
|
if (isset($this->ref_attr[$link_id])) |
|
915
|
10 |
|
$result .= $this->ref_attr[$link_id]; |
|
916
|
|
|
|
|
917
|
10 |
|
$link_text = $this->runSpanGamut($link_text); |
|
918
|
10 |
|
$result .= ">$link_text</a>"; |
|
919
|
10 |
|
$result = $this->hashPart($result); |
|
920
|
|
|
} |
|
921
|
|
|
else { |
|
922
|
3 |
|
$result = $whole_match; |
|
923
|
|
|
} |
|
924
|
11 |
|
return $result; |
|
925
|
|
|
} |
|
926
|
|
|
|
|
927
|
|
|
/** |
|
928
|
|
|
* Callback for inline anchors |
|
929
|
|
|
* @param array $matches |
|
930
|
|
|
* @return string |
|
931
|
|
|
*/ |
|
932
|
12 |
|
protected function _doAnchors_inline_callback($matches) { |
|
933
|
12 |
|
$link_text = $this->runSpanGamut($matches[2]); |
|
934
|
12 |
|
$url = $matches[3] === '' ? $matches[4] : $matches[3]; |
|
935
|
12 |
|
$title_quote =& $matches[6]; |
|
936
|
12 |
|
$title =& $matches[7]; |
|
937
|
12 |
|
$attr = $this->doExtraAttributes("a", $dummy =& $matches[8]); |
|
938
|
|
|
|
|
939
|
|
|
// if the URL was of the form <s p a c e s> it got caught by the HTML |
|
940
|
|
|
// tag parser and hashed. Need to reverse the process before using the URL. |
|
941
|
12 |
|
$unhashed = $this->unhash($url); |
|
942
|
12 |
|
if ($unhashed !== $url) |
|
943
|
2 |
|
$url = preg_replace('/^<(.*)>$/', '\1', $unhashed); |
|
944
|
|
|
|
|
945
|
12 |
|
$url = $this->encodeURLAttribute($url); |
|
946
|
|
|
|
|
947
|
12 |
|
$result = "<a href=\"$url\""; |
|
948
|
12 |
|
if (isset($title) && $title_quote) { |
|
949
|
6 |
|
$title = $this->encodeAttribute($title); |
|
950
|
6 |
|
$result .= " title=\"$title\""; |
|
951
|
|
|
} |
|
952
|
12 |
|
$result .= $attr; |
|
953
|
|
|
|
|
954
|
12 |
|
$link_text = $this->runSpanGamut($link_text); |
|
955
|
12 |
|
$result .= ">$link_text</a>"; |
|
956
|
|
|
|
|
957
|
12 |
|
return $this->hashPart($result); |
|
958
|
|
|
} |
|
959
|
|
|
|
|
960
|
|
|
/** |
|
961
|
|
|
* Turn Markdown image shortcuts into <img> tags. |
|
962
|
|
|
* @param string $text |
|
963
|
|
|
* @return string |
|
964
|
|
|
*/ |
|
965
|
62 |
View Code Duplication |
protected function doImages($text) { |
|
|
|
|
|
|
966
|
|
|
// First, handle reference-style labeled images: ![alt text][id] |
|
967
|
62 |
|
$text = preg_replace_callback('{ |
|
968
|
|
|
( # wrap whole match in $1 |
|
969
|
|
|
!\[ |
|
970
|
62 |
|
(' . $this->nested_brackets_re . ') # alt text = $2 |
|
971
|
|
|
\] |
|
972
|
|
|
|
|
973
|
|
|
[ ]? # one optional space |
|
974
|
|
|
(?:\n[ ]*)? # one optional newline followed by spaces |
|
975
|
|
|
|
|
976
|
|
|
\[ |
|
977
|
|
|
(.*?) # id = $3 |
|
978
|
|
|
\] |
|
979
|
|
|
|
|
980
|
|
|
) |
|
981
|
|
|
}xs', |
|
982
|
62 |
|
array($this, '_doImages_reference_callback'), $text); |
|
983
|
|
|
|
|
984
|
|
|
// Next, handle inline images:  |
|
985
|
|
|
// Don't forget: encode * and _ |
|
986
|
62 |
|
$text = preg_replace_callback('{ |
|
987
|
|
|
( # wrap whole match in $1 |
|
988
|
|
|
!\[ |
|
989
|
62 |
|
(' . $this->nested_brackets_re . ') # alt text = $2 |
|
990
|
|
|
\] |
|
991
|
|
|
\s? # One optional whitespace character |
|
992
|
|
|
\( # literal paren |
|
993
|
|
|
[ \n]* |
|
994
|
|
|
(?: |
|
995
|
|
|
<(\S*)> # src url = $3 |
|
996
|
|
|
| |
|
997
|
62 |
|
(' . $this->nested_url_parenthesis_re . ') # src url = $4 |
|
998
|
|
|
) |
|
999
|
|
|
[ \n]* |
|
1000
|
|
|
( # $5 |
|
1001
|
|
|
([\'"]) # quote char = $6 |
|
1002
|
|
|
(.*?) # title = $7 |
|
1003
|
|
|
\6 # matching quote |
|
1004
|
|
|
[ \n]* |
|
1005
|
|
|
)? # title is optional |
|
1006
|
|
|
\) |
|
1007
|
62 |
|
(?:[ ]? ' . $this->id_class_attr_catch_re . ' )? # $8 = id/class attributes |
|
1008
|
|
|
) |
|
1009
|
|
|
}xs', |
|
1010
|
62 |
|
array($this, '_doImages_inline_callback'), $text); |
|
1011
|
|
|
|
|
1012
|
62 |
|
return $text; |
|
1013
|
|
|
} |
|
1014
|
|
|
|
|
1015
|
|
|
/** |
|
1016
|
|
|
* Callback for referenced images |
|
1017
|
|
|
* @param array $matches |
|
1018
|
|
|
* @return string |
|
1019
|
|
|
*/ |
|
1020
|
3 |
|
protected function _doImages_reference_callback($matches) { |
|
1021
|
3 |
|
$whole_match = $matches[1]; |
|
1022
|
3 |
|
$alt_text = $matches[2]; |
|
1023
|
3 |
|
$link_id = strtolower($matches[3]); |
|
1024
|
|
|
|
|
1025
|
3 |
|
if ($link_id === "") { |
|
1026
|
|
|
$link_id = strtolower($alt_text); // for shortcut links like ![this][]. |
|
1027
|
|
|
} |
|
1028
|
|
|
|
|
1029
|
3 |
|
$alt_text = $this->encodeAttribute($alt_text); |
|
1030
|
3 |
|
if (isset($this->urls[$link_id])) { |
|
1031
|
3 |
|
$url = $this->encodeURLAttribute($this->urls[$link_id]); |
|
1032
|
3 |
|
$result = "<img src=\"$url\" alt=\"$alt_text\""; |
|
1033
|
3 |
View Code Duplication |
if(file_exists($url)) |
|
|
|
|
|
|
1034
|
|
|
{ |
|
1035
|
|
|
list($width, $height, $type, $attr) = getimagesize($url); |
|
|
|
|
|
|
1036
|
|
|
if(isset($width)) $result .= " width=\"$width\""; |
|
1037
|
|
|
if(isset($height)) $result .= " height=\"$height\""; |
|
1038
|
|
|
if(isset($width) && isset($height)) $result .= " loading=\"lazy\""; |
|
1039
|
|
|
} |
|
1040
|
3 |
View Code Duplication |
if (isset($this->titles[$link_id])) { |
|
|
|
|
|
|
1041
|
3 |
|
$title = $this->titles[$link_id]; |
|
1042
|
3 |
|
$title = $this->encodeAttribute($title); |
|
1043
|
3 |
|
$result .= " title=\"$title\""; |
|
1044
|
|
|
} |
|
1045
|
3 |
|
if (isset($this->ref_attr[$link_id])) { |
|
1046
|
3 |
|
$result .= $this->ref_attr[$link_id]; |
|
1047
|
|
|
} |
|
1048
|
3 |
|
$result .= $this->empty_element_suffix; |
|
1049
|
3 |
|
$result = $this->hashPart($result); |
|
1050
|
|
|
} |
|
1051
|
|
|
else { |
|
1052
|
|
|
// If there's no such link ID, leave intact: |
|
1053
|
|
|
$result = $whole_match; |
|
1054
|
|
|
} |
|
1055
|
|
|
|
|
1056
|
3 |
|
return $result; |
|
1057
|
|
|
} |
|
1058
|
|
|
|
|
1059
|
|
|
/** |
|
1060
|
|
|
* Callback for inline images |
|
1061
|
|
|
* @param array $matches |
|
1062
|
|
|
* @return string |
|
1063
|
|
|
*/ |
|
1064
|
3 |
|
protected function _doImages_inline_callback($matches) { |
|
1065
|
3 |
|
$alt_text = $matches[2]; |
|
1066
|
3 |
|
$url = $matches[3] === '' ? $matches[4] : $matches[3]; |
|
1067
|
3 |
|
$title_quote =& $matches[6]; |
|
1068
|
3 |
|
$title =& $matches[7]; |
|
1069
|
3 |
|
$attr = $this->doExtraAttributes("img", $dummy =& $matches[8]); |
|
1070
|
|
|
|
|
1071
|
3 |
|
$alt_text = $this->encodeAttribute($alt_text); |
|
1072
|
3 |
|
$url = $this->encodeURLAttribute($url); |
|
1073
|
3 |
|
$result = "<img src=\"$url\" alt=\"$alt_text\""; |
|
1074
|
3 |
View Code Duplication |
if(file_exists($url)) |
|
|
|
|
|
|
1075
|
|
|
{ |
|
1076
|
|
|
list($width, $height, $type, $attr) = getimagesize($url); |
|
|
|
|
|
|
1077
|
|
|
if(isset($width)) $result .= " width=\"$width\""; |
|
1078
|
|
|
if(isset($height)) $result .= " height=\"$height\""; |
|
1079
|
|
|
if(isset($width) && isset($height)) $result .= " loading=\"lazy\""; |
|
1080
|
|
|
} |
|
1081
|
3 |
|
if (isset($title) && $title_quote) { |
|
1082
|
2 |
|
$title = $this->encodeAttribute($title); |
|
1083
|
2 |
|
$result .= " title=\"$title\""; // $title already quoted |
|
1084
|
|
|
} |
|
1085
|
3 |
|
$result .= $attr; |
|
1086
|
3 |
|
$result .= $this->empty_element_suffix; |
|
1087
|
|
|
|
|
1088
|
3 |
|
return $this->hashPart($result); |
|
1089
|
|
|
} |
|
1090
|
|
|
|
|
1091
|
|
|
/** |
|
1092
|
|
|
* Process markdown headers. Redefined to add ID and class attribute support. |
|
1093
|
|
|
* @param string $text |
|
1094
|
|
|
* @return string |
|
1095
|
|
|
*/ |
|
1096
|
62 |
|
protected function doHeaders($text) { |
|
1097
|
|
|
// Setext-style headers: |
|
1098
|
|
|
// Header 1 {#header1} |
|
1099
|
|
|
// ======== |
|
1100
|
|
|
// |
|
1101
|
|
|
// Header 2 {#header2 .class1 .class2} |
|
1102
|
|
|
// -------- |
|
1103
|
|
|
// |
|
1104
|
62 |
|
$text = preg_replace_callback( |
|
1105
|
|
|
'{ |
|
1106
|
|
|
(^.+?) # $1: Header text |
|
1107
|
62 |
|
(?:[ ]+ ' . $this->id_class_attr_catch_re . ' )? # $3 = id/class attributes |
|
1108
|
|
|
[ ]*\n(=+|-+)[ ]*\n+ # $3: Header footer |
|
1109
|
|
|
}mx', |
|
1110
|
62 |
|
array($this, '_doHeaders_callback_setext'), $text); |
|
1111
|
|
|
|
|
1112
|
|
|
// atx-style headers: |
|
1113
|
|
|
// # Header 1 {#header1} |
|
1114
|
|
|
// ## Header 2 {#header2} |
|
1115
|
|
|
// ## Header 2 with closing hashes ## {#header3.class1.class2} |
|
1116
|
|
|
// ... |
|
1117
|
|
|
// ###### Header 6 {.class2} |
|
1118
|
|
|
// |
|
1119
|
62 |
|
$text = preg_replace_callback('{ |
|
1120
|
|
|
^(\#{1,6}) # $1 = string of #\'s |
|
1121
|
62 |
|
[ ]'.($this->hashtag_protection ? '+' : '*').' |
|
1122
|
|
|
(.+?) # $2 = Header text |
|
1123
|
|
|
[ ]* |
|
1124
|
|
|
\#* # optional closing #\'s (not counted) |
|
1125
|
62 |
|
(?:[ ]+ ' . $this->id_class_attr_catch_re . ' )? # $3 = id/class attributes |
|
1126
|
|
|
[ ]* |
|
1127
|
|
|
\n+ |
|
1128
|
|
|
}xm', |
|
1129
|
62 |
|
array($this, '_doHeaders_callback_atx'), $text); |
|
1130
|
|
|
|
|
1131
|
62 |
|
return $text; |
|
1132
|
|
|
} |
|
1133
|
|
|
|
|
1134
|
|
|
/** |
|
1135
|
|
|
* Callback for setext headers |
|
1136
|
|
|
* @param array $matches |
|
1137
|
|
|
* @return string |
|
1138
|
|
|
*/ |
|
1139
|
6 |
|
protected function _doHeaders_callback_setext($matches) { |
|
1140
|
6 |
View Code Duplication |
if ($matches[3] === '-' && preg_match('{^- }', $matches[1])) { |
|
|
|
|
|
|
1141
|
1 |
|
return $matches[0]; |
|
1142
|
|
|
} |
|
1143
|
|
|
|
|
1144
|
5 |
|
$level = $matches[3][0] === '=' ? 1 : 2; |
|
1145
|
|
|
|
|
1146
|
5 |
|
$defaultId = is_callable($this->header_id_func) ? call_user_func($this->header_id_func, $matches[1]) : null; |
|
1147
|
|
|
|
|
1148
|
5 |
|
$attr = $this->doExtraAttributes("h$level", $dummy =& $matches[2], $defaultId); |
|
1149
|
5 |
|
$block = "<h$level$attr>" . $this->runSpanGamut($matches[1]) . "</h$level>"; |
|
1150
|
5 |
|
return "\n" . $this->hashBlock($block) . "\n\n"; |
|
1151
|
|
|
} |
|
1152
|
|
|
|
|
1153
|
|
|
/** |
|
1154
|
|
|
* Callback for atx headers |
|
1155
|
|
|
* @param array $matches |
|
1156
|
|
|
* @return string |
|
1157
|
|
|
*/ |
|
1158
|
12 |
|
protected function _doHeaders_callback_atx($matches) { |
|
1159
|
12 |
|
$level = strlen($matches[1]); |
|
1160
|
|
|
|
|
1161
|
12 |
|
$defaultId = is_callable($this->header_id_func) ? call_user_func($this->header_id_func, $matches[2]) : null; |
|
1162
|
12 |
|
$attr = $this->doExtraAttributes("h$level", $dummy =& $matches[3], $defaultId); |
|
1163
|
12 |
|
$block = "<h$level$attr>" . $this->runSpanGamut($matches[2]) . "</h$level>"; |
|
1164
|
12 |
|
return "\n" . $this->hashBlock($block) . "\n\n"; |
|
1165
|
|
|
} |
|
1166
|
|
|
|
|
1167
|
|
|
/** |
|
1168
|
|
|
* Form HTML tables. |
|
1169
|
|
|
* @param string $text |
|
1170
|
|
|
* @return string |
|
1171
|
|
|
*/ |
|
1172
|
62 |
|
protected function doTables($text) { |
|
1173
|
62 |
|
$less_than_tab = $this->tab_width - 1; |
|
1174
|
|
|
// Find tables with leading pipe. |
|
1175
|
|
|
// |
|
1176
|
|
|
// | Header 1 | Header 2 |
|
1177
|
|
|
// | -------- | -------- |
|
1178
|
|
|
// | Cell 1 | Cell 2 |
|
1179
|
|
|
// | Cell 3 | Cell 4 |
|
1180
|
62 |
|
$text = preg_replace_callback(' |
|
1181
|
|
|
{ |
|
1182
|
|
|
^ # Start of a line |
|
1183
|
62 |
|
[ ]{0,' . $less_than_tab . '} # Allowed whitespace. |
|
1184
|
|
|
[|] # Optional leading pipe (present) |
|
1185
|
|
|
(.+) \n # $1: Header row (at least one pipe) |
|
1186
|
|
|
|
|
1187
|
62 |
|
[ ]{0,' . $less_than_tab . '} # Allowed whitespace. |
|
1188
|
|
|
[|] ([ ]*[-:]+[-| :]*) \n # $2: Header underline |
|
1189
|
|
|
|
|
1190
|
|
|
( # $3: Cells |
|
1191
|
|
|
(?> |
|
1192
|
|
|
[ ]* # Allowed whitespace. |
|
1193
|
|
|
[|] .* \n # Row content. |
|
1194
|
|
|
)* |
|
1195
|
|
|
) |
|
1196
|
|
|
(?=\n|\Z) # Stop at final double newline. |
|
1197
|
|
|
}xm', |
|
1198
|
62 |
|
array($this, '_doTable_leadingPipe_callback'), $text); |
|
1199
|
|
|
|
|
1200
|
|
|
// Find tables without leading pipe. |
|
1201
|
|
|
// |
|
1202
|
|
|
// Header 1 | Header 2 |
|
1203
|
|
|
// -------- | -------- |
|
1204
|
|
|
// Cell 1 | Cell 2 |
|
1205
|
|
|
// Cell 3 | Cell 4 |
|
1206
|
62 |
|
$text = preg_replace_callback(' |
|
1207
|
|
|
{ |
|
1208
|
|
|
^ # Start of a line |
|
1209
|
62 |
|
[ ]{0,' . $less_than_tab . '} # Allowed whitespace. |
|
1210
|
|
|
(\S.*[|].*) \n # $1: Header row (at least one pipe) |
|
1211
|
|
|
|
|
1212
|
62 |
|
[ ]{0,' . $less_than_tab . '} # Allowed whitespace. |
|
1213
|
|
|
([-:]+[ ]*[|][-| :]*) \n # $2: Header underline |
|
1214
|
|
|
|
|
1215
|
|
|
( # $3: Cells |
|
1216
|
|
|
(?> |
|
1217
|
|
|
.* [|] .* \n # Row content |
|
1218
|
|
|
)* |
|
1219
|
|
|
) |
|
1220
|
|
|
(?=\n|\Z) # Stop at final double newline. |
|
1221
|
|
|
}xm', |
|
1222
|
62 |
|
array($this, '_DoTable_callback'), $text); |
|
1223
|
|
|
|
|
1224
|
62 |
|
return $text; |
|
1225
|
|
|
} |
|
1226
|
|
|
|
|
1227
|
|
|
/** |
|
1228
|
|
|
* Callback for removing the leading pipe for each row |
|
1229
|
|
|
* @param array $matches |
|
1230
|
|
|
* @return string |
|
1231
|
|
|
*/ |
|
1232
|
1 |
|
protected function _doTable_leadingPipe_callback($matches) { |
|
1233
|
1 |
|
$head = $matches[1]; |
|
1234
|
1 |
|
$underline = $matches[2]; |
|
1235
|
1 |
|
$content = $matches[3]; |
|
1236
|
|
|
|
|
1237
|
1 |
|
$content = preg_replace('/^ *[|]/m', '', $content); |
|
1238
|
|
|
|
|
1239
|
1 |
|
return $this->_doTable_callback(array($matches[0], $head, $underline, $content)); |
|
1240
|
|
|
} |
|
1241
|
|
|
|
|
1242
|
|
|
/** |
|
1243
|
|
|
* Make the align attribute in a table |
|
1244
|
|
|
* @param string $alignname |
|
1245
|
|
|
* @return string |
|
1246
|
|
|
*/ |
|
1247
|
1 |
|
protected function _doTable_makeAlignAttr($alignname) { |
|
1248
|
1 |
|
if (empty($this->table_align_class_tmpl)) { |
|
1249
|
1 |
|
return " align=\"$alignname\""; |
|
1250
|
|
|
} |
|
1251
|
|
|
|
|
1252
|
|
|
$classname = str_replace('%%', $alignname, $this->table_align_class_tmpl); |
|
1253
|
|
|
return " class=\"$classname\""; |
|
1254
|
|
|
} |
|
1255
|
|
|
|
|
1256
|
|
|
/** |
|
1257
|
|
|
* Calback for processing tables |
|
1258
|
|
|
* @param array $matches |
|
1259
|
|
|
* @return string |
|
1260
|
|
|
*/ |
|
1261
|
1 |
|
protected function _doTable_callback($matches) { |
|
1262
|
1 |
|
$head = $matches[1]; |
|
1263
|
1 |
|
$underline = $matches[2]; |
|
1264
|
1 |
|
$content = $matches[3]; |
|
1265
|
|
|
|
|
1266
|
|
|
// Remove any tailing pipes for each line. |
|
1267
|
1 |
|
$head = preg_replace('/[|] *$/m', '', $head); |
|
1268
|
1 |
|
$underline = preg_replace('/[|] *$/m', '', $underline); |
|
1269
|
1 |
|
$content = preg_replace('/[|] *$/m', '', $content); |
|
1270
|
|
|
|
|
1271
|
|
|
// Reading alignement from header underline. |
|
1272
|
1 |
|
$separators = preg_split('/ *[|] */', $underline); |
|
1273
|
1 |
|
foreach ($separators as $n => $s) { |
|
1274
|
1 |
|
if (preg_match('/^ *-+: *$/', $s)) |
|
1275
|
1 |
|
$attr[$n] = $this->_doTable_makeAlignAttr('right'); |
|
|
|
|
|
|
1276
|
1 |
|
else if (preg_match('/^ *:-+: *$/', $s)) |
|
1277
|
1 |
|
$attr[$n] = $this->_doTable_makeAlignAttr('center'); |
|
|
|
|
|
|
1278
|
1 |
|
else if (preg_match('/^ *:-+ *$/', $s)) |
|
1279
|
1 |
|
$attr[$n] = $this->_doTable_makeAlignAttr('left'); |
|
1280
|
|
|
else |
|
1281
|
1 |
|
$attr[$n] = ''; |
|
1282
|
|
|
} |
|
1283
|
|
|
|
|
1284
|
|
|
// Parsing span elements, including code spans, character escapes, |
|
1285
|
|
|
// and inline HTML tags, so that pipes inside those gets ignored. |
|
1286
|
1 |
|
$head = $this->parseSpan($head); |
|
1287
|
1 |
|
$headers = preg_split('/ *[|] */', $head); |
|
1288
|
1 |
|
$col_count = count($headers); |
|
1289
|
1 |
|
$attr = array_pad($attr, $col_count, ''); |
|
1290
|
|
|
|
|
1291
|
|
|
// Write column headers. |
|
1292
|
1 |
|
$text = "<table>\n"; |
|
1293
|
1 |
|
$text .= "<thead>\n"; |
|
1294
|
1 |
|
$text .= "<tr>\n"; |
|
1295
|
1 |
|
foreach ($headers as $n => $header) { |
|
1296
|
1 |
|
$text .= " <th$attr[$n]>" . $this->runSpanGamut(trim($header)) . "</th>\n"; |
|
1297
|
|
|
} |
|
1298
|
1 |
|
$text .= "</tr>\n"; |
|
1299
|
1 |
|
$text .= "</thead>\n"; |
|
1300
|
|
|
|
|
1301
|
|
|
// Split content by row. |
|
1302
|
1 |
|
$rows = explode("\n", trim($content, "\n")); |
|
1303
|
|
|
|
|
1304
|
1 |
|
$text .= "<tbody>\n"; |
|
1305
|
1 |
|
foreach ($rows as $row) { |
|
1306
|
|
|
// Parsing span elements, including code spans, character escapes, |
|
1307
|
|
|
// and inline HTML tags, so that pipes inside those gets ignored. |
|
1308
|
1 |
|
$row = $this->parseSpan($row); |
|
1309
|
|
|
|
|
1310
|
|
|
// Split row by cell. |
|
1311
|
1 |
|
$row_cells = preg_split('/ *[|] */', $row, $col_count); |
|
1312
|
1 |
|
$row_cells = array_pad($row_cells, $col_count, ''); |
|
1313
|
|
|
|
|
1314
|
1 |
|
$text .= "<tr>\n"; |
|
1315
|
1 |
|
foreach ($row_cells as $n => $cell) { |
|
1316
|
1 |
|
$text .= " <td$attr[$n]>" . $this->runSpanGamut(trim($cell)) . "</td>\n"; |
|
1317
|
|
|
} |
|
1318
|
1 |
|
$text .= "</tr>\n"; |
|
1319
|
|
|
} |
|
1320
|
1 |
|
$text .= "</tbody>\n"; |
|
1321
|
1 |
|
$text .= "</table>"; |
|
1322
|
|
|
|
|
1323
|
1 |
|
return $this->hashBlock($text) . "\n"; |
|
1324
|
|
|
} |
|
1325
|
|
|
|
|
1326
|
|
|
/** |
|
1327
|
|
|
* Form HTML definition lists. |
|
1328
|
|
|
* @param string $text |
|
1329
|
|
|
* @return string |
|
1330
|
|
|
*/ |
|
1331
|
62 |
|
protected function doDefLists($text) { |
|
1332
|
62 |
|
$less_than_tab = $this->tab_width - 1; |
|
1333
|
|
|
|
|
1334
|
|
|
// Re-usable pattern to match any entire dl list: |
|
1335
|
|
|
$whole_list_re = '(?> |
|
1336
|
|
|
( # $1 = whole list |
|
1337
|
|
|
( # $2 |
|
1338
|
62 |
|
[ ]{0,' . $less_than_tab . '} |
|
1339
|
|
|
((?>.*\S.*\n)+) # $3 = defined term |
|
1340
|
|
|
\n? |
|
1341
|
62 |
|
[ ]{0,' . $less_than_tab . '}:[ ]+ # colon starting definition |
|
1342
|
|
|
) |
|
1343
|
|
|
(?s:.+?) |
|
1344
|
|
|
( # $4 |
|
1345
|
|
|
\z |
|
1346
|
|
|
| |
|
1347
|
|
|
\n{2,} |
|
1348
|
|
|
(?=\S) |
|
1349
|
|
|
(?! # Negative lookahead for another term |
|
1350
|
62 |
|
[ ]{0,' . $less_than_tab . '} |
|
1351
|
|
|
(?: \S.*\n )+? # defined term |
|
1352
|
|
|
\n? |
|
1353
|
62 |
|
[ ]{0,' . $less_than_tab . '}:[ ]+ # colon starting definition |
|
1354
|
|
|
) |
|
1355
|
|
|
(?! # Negative lookahead for another definition |
|
1356
|
62 |
|
[ ]{0,' . $less_than_tab . '}:[ ]+ # colon starting definition |
|
1357
|
|
|
) |
|
1358
|
|
|
) |
|
1359
|
|
|
) |
|
1360
|
|
|
)'; // mx |
|
1361
|
|
|
|
|
1362
|
62 |
|
$text = preg_replace_callback('{ |
|
1363
|
|
|
(?>\A\n?|(?<=\n\n)) |
|
1364
|
62 |
|
' . $whole_list_re . ' |
|
1365
|
|
|
}mx', |
|
1366
|
62 |
|
array($this, '_doDefLists_callback'), $text); |
|
1367
|
|
|
|
|
1368
|
62 |
|
return $text; |
|
1369
|
|
|
} |
|
1370
|
|
|
|
|
1371
|
|
|
/** |
|
1372
|
|
|
* Callback for processing definition lists |
|
1373
|
|
|
* @param array $matches |
|
1374
|
|
|
* @return string |
|
1375
|
|
|
*/ |
|
1376
|
1 |
|
protected function _doDefLists_callback($matches) { |
|
1377
|
|
|
// Re-usable patterns to match list item bullets and number markers: |
|
1378
|
1 |
|
$list = $matches[1]; |
|
1379
|
|
|
|
|
1380
|
|
|
// Turn double returns into triple returns, so that we can make a |
|
1381
|
|
|
// paragraph for the last item in a list, if necessary: |
|
1382
|
1 |
|
$result = trim($this->processDefListItems($list)); |
|
1383
|
1 |
|
$result = "<dl>\n" . $result . "\n</dl>"; |
|
1384
|
1 |
|
return $this->hashBlock($result) . "\n\n"; |
|
1385
|
|
|
} |
|
1386
|
|
|
|
|
1387
|
|
|
/** |
|
1388
|
|
|
* Process the contents of a single definition list, splitting it |
|
1389
|
|
|
* into individual term and definition list items. |
|
1390
|
|
|
* @param string $list_str |
|
1391
|
|
|
* @return string |
|
1392
|
|
|
*/ |
|
1393
|
1 |
|
protected function processDefListItems($list_str) { |
|
1394
|
|
|
|
|
1395
|
1 |
|
$less_than_tab = $this->tab_width - 1; |
|
1396
|
|
|
|
|
1397
|
|
|
// Trim trailing blank lines: |
|
1398
|
1 |
|
$list_str = preg_replace("/\n{2,}\\z/", "\n", $list_str); |
|
1399
|
|
|
|
|
1400
|
|
|
// Process definition terms. |
|
1401
|
1 |
|
$list_str = preg_replace_callback('{ |
|
1402
|
|
|
(?>\A\n?|\n\n+) # leading line |
|
1403
|
|
|
( # definition terms = $1 |
|
1404
|
1 |
|
[ ]{0,' . $less_than_tab . '} # leading whitespace |
|
1405
|
|
|
(?!\:[ ]|[ ]) # negative lookahead for a definition |
|
1406
|
|
|
# mark (colon) or more whitespace. |
|
1407
|
|
|
(?> \S.* \n)+? # actual term (not whitespace). |
|
1408
|
|
|
) |
|
1409
|
|
|
(?=\n?[ ]{0,3}:[ ]) # lookahead for following line feed |
|
1410
|
|
|
# with a definition mark. |
|
1411
|
|
|
}xm', |
|
1412
|
1 |
|
array($this, '_processDefListItems_callback_dt'), $list_str); |
|
1413
|
|
|
|
|
1414
|
|
|
// Process actual definitions. |
|
1415
|
1 |
|
$list_str = preg_replace_callback('{ |
|
1416
|
|
|
\n(\n+)? # leading line = $1 |
|
1417
|
|
|
( # marker space = $2 |
|
1418
|
1 |
|
[ ]{0,' . $less_than_tab . '} # whitespace before colon |
|
1419
|
|
|
\:[ ]+ # definition mark (colon) |
|
1420
|
|
|
) |
|
1421
|
|
|
((?s:.+?)) # definition text = $3 |
|
1422
|
|
|
(?= \n+ # stop at next definition mark, |
|
1423
|
|
|
(?: # next term or end of text |
|
1424
|
1 |
|
[ ]{0,' . $less_than_tab . '} \:[ ] | |
|
1425
|
|
|
<dt> | \z |
|
1426
|
|
|
) |
|
1427
|
|
|
) |
|
1428
|
|
|
}xm', |
|
1429
|
1 |
|
array($this, '_processDefListItems_callback_dd'), $list_str); |
|
1430
|
|
|
|
|
1431
|
1 |
|
return $list_str; |
|
1432
|
|
|
} |
|
1433
|
|
|
|
|
1434
|
|
|
/** |
|
1435
|
|
|
* Callback for <dt> elements in definition lists |
|
1436
|
|
|
* @param array $matches |
|
1437
|
|
|
* @return string |
|
1438
|
|
|
*/ |
|
1439
|
1 |
|
protected function _processDefListItems_callback_dt($matches) { |
|
1440
|
1 |
|
$terms = explode("\n", trim($matches[1])); |
|
1441
|
1 |
|
$text = ''; |
|
1442
|
1 |
|
foreach ($terms as $term) { |
|
1443
|
1 |
|
$term = $this->runSpanGamut(trim($term)); |
|
1444
|
1 |
|
$text .= "\n<dt>" . $term . "</dt>"; |
|
1445
|
|
|
} |
|
1446
|
1 |
|
return $text . "\n"; |
|
1447
|
|
|
} |
|
1448
|
|
|
|
|
1449
|
|
|
/** |
|
1450
|
|
|
* Callback for <dd> elements in definition lists |
|
1451
|
|
|
* @param array $matches |
|
1452
|
|
|
* @return string |
|
1453
|
|
|
*/ |
|
1454
|
1 |
|
protected function _processDefListItems_callback_dd($matches) { |
|
1455
|
1 |
|
$leading_line = $matches[1]; |
|
1456
|
1 |
|
$marker_space = $matches[2]; |
|
1457
|
1 |
|
$def = $matches[3]; |
|
1458
|
|
|
|
|
1459
|
1 |
|
if ($leading_line || preg_match('/\n{2,}/', $def)) { |
|
1460
|
|
|
// Replace marker with the appropriate whitespace indentation |
|
1461
|
1 |
|
$def = str_repeat(' ', strlen($marker_space)) . $def; |
|
1462
|
1 |
|
$def = $this->runBlockGamut($this->outdent($def . "\n\n")); |
|
1463
|
1 |
|
$def = "\n". $def ."\n"; |
|
1464
|
|
|
} |
|
1465
|
|
|
else { |
|
1466
|
1 |
|
$def = rtrim($def); |
|
1467
|
1 |
|
$def = $this->runSpanGamut($this->outdent($def)); |
|
1468
|
|
|
} |
|
1469
|
|
|
|
|
1470
|
1 |
|
return "\n<dd>" . $def . "</dd>\n"; |
|
1471
|
|
|
} |
|
1472
|
|
|
|
|
1473
|
|
|
/** |
|
1474
|
|
|
* Adding the fenced code block syntax to regular Markdown: |
|
1475
|
|
|
* |
|
1476
|
|
|
* ~~~ |
|
1477
|
|
|
* Code block |
|
1478
|
|
|
* ~~~ |
|
1479
|
|
|
* |
|
1480
|
|
|
* @param string $text |
|
1481
|
|
|
* @return string |
|
1482
|
|
|
*/ |
|
1483
|
62 |
|
protected function doFencedCodeBlocks($text) { |
|
1484
|
|
|
|
|
1485
|
62 |
|
$text = preg_replace_callback('{ |
|
1486
|
|
|
(?:\n|\A) |
|
1487
|
|
|
# 1: Opening marker |
|
1488
|
|
|
( |
|
1489
|
|
|
(?:~{3,}|`{3,}) # 3 or more tildes/backticks. |
|
1490
|
|
|
) |
|
1491
|
|
|
[ ]* |
|
1492
|
|
|
(?: |
|
1493
|
|
|
\.?([-_:a-zA-Z0-9]+) # 2: standalone class name |
|
1494
|
|
|
)? |
|
1495
|
|
|
[ ]* |
|
1496
|
|
|
(?: |
|
1497
|
62 |
|
' . $this->id_class_attr_catch_re . ' # 3: Extra attributes |
|
1498
|
|
|
)? |
|
1499
|
|
|
[ ]* \n # Whitespace and newline following marker. |
|
1500
|
|
|
|
|
1501
|
|
|
# 4: Content |
|
1502
|
|
|
( |
|
1503
|
|
|
(?> |
|
1504
|
|
|
(?!\1 [ ]* \n) # Not a closing marker. |
|
1505
|
|
|
.*\n+ |
|
1506
|
|
|
)+ |
|
1507
|
|
|
) |
|
1508
|
|
|
|
|
1509
|
|
|
# Closing marker. |
|
1510
|
|
|
\1 [ ]* (?= \n ) |
|
1511
|
|
|
}xm', |
|
1512
|
62 |
|
array($this, '_doFencedCodeBlocks_callback'), $text); |
|
1513
|
|
|
|
|
1514
|
62 |
|
return $text; |
|
1515
|
|
|
} |
|
1516
|
|
|
|
|
1517
|
|
|
/** |
|
1518
|
|
|
* Callback to process fenced code blocks |
|
1519
|
|
|
* @param array $matches |
|
1520
|
|
|
* @return string |
|
1521
|
|
|
*/ |
|
1522
|
4 |
|
protected function _doFencedCodeBlocks_callback($matches) { |
|
1523
|
4 |
|
$classname =& $matches[2]; |
|
1524
|
4 |
|
$attrs =& $matches[3]; |
|
1525
|
4 |
|
$codeblock = $matches[4]; |
|
1526
|
|
|
|
|
1527
|
4 |
|
if ($this->code_block_content_func) { |
|
1528
|
|
|
$codeblock = call_user_func($this->code_block_content_func, $codeblock, $classname); |
|
1529
|
|
|
} else { |
|
1530
|
4 |
|
$codeblock = htmlspecialchars($codeblock, ENT_NOQUOTES); |
|
1531
|
|
|
} |
|
1532
|
|
|
|
|
1533
|
4 |
|
$codeblock = preg_replace_callback('/^\n+/', |
|
1534
|
4 |
|
array($this, '_doFencedCodeBlocks_newlines'), $codeblock); |
|
1535
|
|
|
|
|
1536
|
4 |
|
$classes = array(); |
|
1537
|
4 |
|
if ($classname !== "") { |
|
1538
|
4 |
|
if ($classname[0] === '.') { |
|
1539
|
|
|
$classname = substr($classname, 1); |
|
1540
|
|
|
} |
|
1541
|
4 |
|
$classes[] = $this->code_class_prefix . $classname; |
|
1542
|
|
|
} |
|
1543
|
4 |
|
$attr_str = $this->doExtraAttributes($this->code_attr_on_pre ? "pre" : "code", $attrs, null, $classes); |
|
1544
|
4 |
|
$pre_attr_str = $this->code_attr_on_pre ? $attr_str : ''; |
|
1545
|
4 |
|
$code_attr_str = $this->code_attr_on_pre ? '' : $attr_str; |
|
1546
|
4 |
|
$codeblock = "<pre$pre_attr_str><code$code_attr_str>$codeblock</code></pre>"; |
|
1547
|
|
|
|
|
1548
|
4 |
|
return "\n\n".$this->hashBlock($codeblock)."\n\n"; |
|
1549
|
|
|
} |
|
1550
|
|
|
|
|
1551
|
|
|
/** |
|
1552
|
|
|
* Replace new lines in fenced code blocks |
|
1553
|
|
|
* @param array $matches |
|
1554
|
|
|
* @return string |
|
1555
|
|
|
*/ |
|
1556
|
2 |
|
protected function _doFencedCodeBlocks_newlines($matches) { |
|
1557
|
2 |
|
return str_repeat("<br$this->empty_element_suffix", |
|
1558
|
2 |
|
strlen($matches[0])); |
|
1559
|
|
|
} |
|
1560
|
|
|
|
|
1561
|
|
|
/** |
|
1562
|
|
|
* Redefining emphasis markers so that emphasis by underscore does not |
|
1563
|
|
|
* work in the middle of a word. |
|
1564
|
|
|
* @var array |
|
1565
|
|
|
*/ |
|
1566
|
|
|
protected $em_relist = array( |
|
1567
|
|
|
'' => '(?:(?<!\*)\*(?!\*)|(?<![a-zA-Z0-9_])_(?!_))(?![\.,:;]?\s)', |
|
1568
|
|
|
'*' => '(?<![\s*])\*(?!\*)', |
|
1569
|
|
|
'_' => '(?<![\s_])_(?![a-zA-Z0-9_])', |
|
1570
|
|
|
); |
|
1571
|
|
|
protected $strong_relist = array( |
|
1572
|
|
|
'' => '(?:(?<!\*)\*\*(?!\*)|(?<![a-zA-Z0-9_])__(?!_))(?![\.,:;]?\s)', |
|
1573
|
|
|
'**' => '(?<![\s*])\*\*(?!\*)', |
|
1574
|
|
|
'__' => '(?<![\s_])__(?![a-zA-Z0-9_])', |
|
1575
|
|
|
); |
|
1576
|
|
|
protected $em_strong_relist = array( |
|
1577
|
|
|
'' => '(?:(?<!\*)\*\*\*(?!\*)|(?<![a-zA-Z0-9_])___(?!_))(?![\.,:;]?\s)', |
|
1578
|
|
|
'***' => '(?<![\s*])\*\*\*(?!\*)', |
|
1579
|
|
|
'___' => '(?<![\s_])___(?![a-zA-Z0-9_])', |
|
1580
|
|
|
); |
|
1581
|
|
|
|
|
1582
|
|
|
/** |
|
1583
|
|
|
* Parse text into paragraphs |
|
1584
|
|
|
* @param string $text String to process in paragraphs |
|
1585
|
|
|
* @param boolean $wrap_in_p Whether paragraphs should be wrapped in <p> tags |
|
1586
|
|
|
* @return string HTML output |
|
1587
|
|
|
*/ |
|
1588
|
62 |
|
protected function formParagraphs($text, $wrap_in_p = true) { |
|
1589
|
|
|
// Strip leading and trailing lines: |
|
1590
|
62 |
|
$text = preg_replace('/\A\n+|\n+\z/', '', $text); |
|
1591
|
|
|
|
|
1592
|
62 |
|
$grafs = preg_split('/\n{2,}/', $text, -1, PREG_SPLIT_NO_EMPTY); |
|
1593
|
|
|
|
|
1594
|
|
|
// Wrap <p> tags and unhashify HTML blocks |
|
1595
|
62 |
|
foreach ($grafs as $key => $value) { |
|
1596
|
62 |
|
$value = trim($this->runSpanGamut($value)); |
|
1597
|
|
|
|
|
1598
|
|
|
// Check if this should be enclosed in a paragraph. |
|
1599
|
|
|
// Clean tag hashes & block tag hashes are left alone. |
|
1600
|
62 |
|
$is_p = $wrap_in_p && !preg_match('/^B\x1A[0-9]+B|^C\x1A[0-9]+C$/', $value); |
|
1601
|
|
|
|
|
1602
|
62 |
|
if ($is_p) { |
|
1603
|
58 |
|
$value = "<p>$value</p>"; |
|
1604
|
|
|
} |
|
1605
|
62 |
|
$grafs[$key] = $value; |
|
1606
|
|
|
} |
|
1607
|
|
|
|
|
1608
|
|
|
// Join grafs in one text, then unhash HTML tags. |
|
1609
|
62 |
|
$text = implode("\n\n", $grafs); |
|
1610
|
|
|
|
|
1611
|
|
|
// Finish by removing any tag hashes still present in $text. |
|
1612
|
62 |
|
$text = $this->unhash($text); |
|
1613
|
|
|
|
|
1614
|
62 |
|
return $text; |
|
1615
|
|
|
} |
|
1616
|
|
|
|
|
1617
|
|
|
|
|
1618
|
|
|
/** |
|
1619
|
|
|
* Footnotes - Strips link definitions from text, stores the URLs and |
|
1620
|
|
|
* titles in hash references. |
|
1621
|
|
|
* @param string $text |
|
1622
|
|
|
* @return string |
|
1623
|
|
|
*/ |
|
1624
|
62 |
View Code Duplication |
protected function stripFootnotes($text) { |
|
|
|
|
|
|
1625
|
62 |
|
$less_than_tab = $this->tab_width - 1; |
|
1626
|
|
|
|
|
1627
|
|
|
// Link defs are in the form: [^id]: url "optional title" |
|
1628
|
62 |
|
$text = preg_replace_callback('{ |
|
1629
|
62 |
|
^[ ]{0,' . $less_than_tab . '}\[\^(.+?)\][ ]?: # note_id = $1 |
|
1630
|
|
|
[ ]* |
|
1631
|
|
|
\n? # maybe *one* newline |
|
1632
|
|
|
( # text = $2 (no blank lines allowed) |
|
1633
|
|
|
(?: |
|
1634
|
|
|
.+ # actual text |
|
1635
|
|
|
| |
|
1636
|
|
|
\n # newlines but |
|
1637
|
|
|
(?!\[.+?\][ ]?:\s)# negative lookahead for footnote or link definition marker. |
|
1638
|
|
|
(?!\n+[ ]{0,3}\S)# ensure line is not blank and followed |
|
1639
|
|
|
# by non-indented content |
|
1640
|
|
|
)* |
|
1641
|
|
|
) |
|
1642
|
|
|
}xm', |
|
1643
|
62 |
|
array($this, '_stripFootnotes_callback'), |
|
1644
|
62 |
|
$text); |
|
1645
|
62 |
|
return $text; |
|
1646
|
|
|
} |
|
1647
|
|
|
|
|
1648
|
|
|
/** |
|
1649
|
|
|
* Callback for stripping footnotes |
|
1650
|
|
|
* @param array $matches |
|
1651
|
|
|
* @return string |
|
1652
|
|
|
*/ |
|
1653
|
1 |
|
protected function _stripFootnotes_callback($matches) { |
|
1654
|
1 |
|
$note_id = $this->fn_id_prefix . $matches[1]; |
|
1655
|
1 |
|
$this->footnotes[$note_id] = $this->outdent($matches[2]); |
|
1656
|
1 |
|
return ''; // String that will replace the block |
|
1657
|
|
|
} |
|
1658
|
|
|
|
|
1659
|
|
|
/** |
|
1660
|
|
|
* Replace footnote references in $text [^id] with a special text-token |
|
1661
|
|
|
* which will be replaced by the actual footnote marker in appendFootnotes. |
|
1662
|
|
|
* @param string $text |
|
1663
|
|
|
* @return string |
|
1664
|
|
|
*/ |
|
1665
|
62 |
|
protected function doFootnotes($text) { |
|
1666
|
62 |
|
if (!$this->in_anchor) { |
|
1667
|
62 |
|
$text = preg_replace('{\[\^(.+?)\]}', "F\x1Afn:\\1\x1A:", $text); |
|
1668
|
|
|
} |
|
1669
|
62 |
|
return $text; |
|
1670
|
|
|
} |
|
1671
|
|
|
|
|
1672
|
|
|
/** |
|
1673
|
|
|
* Append footnote list to text |
|
1674
|
|
|
* @param string $text |
|
1675
|
|
|
* @return string |
|
1676
|
|
|
*/ |
|
1677
|
62 |
|
protected function appendFootnotes($text) { |
|
1678
|
62 |
|
$text = preg_replace_callback('{F\x1Afn:(.*?)\x1A:}', |
|
1679
|
62 |
|
array($this, '_appendFootnotes_callback'), $text); |
|
1680
|
|
|
|
|
1681
|
62 |
|
if ( ! empty( $this->footnotes_ordered ) ) { |
|
1682
|
1 |
|
$this->_doFootnotes(); |
|
1683
|
1 |
|
if ( ! $this->omit_footnotes ) { |
|
1684
|
1 |
|
$text .= "\n\n"; |
|
1685
|
1 |
|
$text .= "<div class=\"footnotes\" role=\"doc-endnotes\">\n"; |
|
1686
|
1 |
|
$text .= "<hr" . $this->empty_element_suffix . "\n"; |
|
1687
|
1 |
|
$text .= $this->footnotes_assembled; |
|
1688
|
1 |
|
$text .= "</div>"; |
|
1689
|
|
|
} |
|
1690
|
|
|
} |
|
1691
|
62 |
|
return $text; |
|
1692
|
|
|
} |
|
1693
|
|
|
|
|
1694
|
|
|
|
|
1695
|
|
|
/** |
|
1696
|
|
|
* Generates the HTML for footnotes. Called by appendFootnotes, even if |
|
1697
|
|
|
* footnotes are not being appended. |
|
1698
|
|
|
* @return void |
|
1699
|
|
|
*/ |
|
1700
|
1 |
|
protected function _doFootnotes() { |
|
1701
|
1 |
|
$attr = array(); |
|
1702
|
1 |
|
if ($this->fn_backlink_class !== "") { |
|
1703
|
1 |
|
$class = $this->fn_backlink_class; |
|
1704
|
1 |
|
$class = $this->encodeAttribute($class); |
|
1705
|
1 |
|
$attr['class'] = " class=\"$class\""; |
|
1706
|
|
|
} |
|
1707
|
1 |
|
$attr['role'] = " role=\"doc-backlink\""; |
|
1708
|
1 |
|
$num = 0; |
|
1709
|
|
|
|
|
1710
|
1 |
|
$text = "<ol>\n\n"; |
|
1711
|
1 |
|
while (!empty($this->footnotes_ordered)) { |
|
1712
|
1 |
|
$footnote = reset($this->footnotes_ordered); |
|
1713
|
1 |
|
$note_id = key($this->footnotes_ordered); |
|
1714
|
1 |
|
unset($this->footnotes_ordered[$note_id]); |
|
1715
|
1 |
|
$ref_count = $this->footnotes_ref_count[$note_id]; |
|
1716
|
1 |
|
unset($this->footnotes_ref_count[$note_id]); |
|
1717
|
1 |
|
unset($this->footnotes[$note_id]); |
|
1718
|
|
|
|
|
1719
|
1 |
|
$footnote .= "\n"; // Need to append newline before parsing. |
|
1720
|
1 |
|
$footnote = $this->runBlockGamut("$footnote\n"); |
|
1721
|
1 |
|
$footnote = preg_replace_callback('{F\x1Afn:(.*?)\x1A:}', |
|
1722
|
1 |
|
array($this, '_appendFootnotes_callback'), $footnote); |
|
1723
|
|
|
|
|
1724
|
1 |
|
$num++; |
|
1725
|
1 |
|
$note_id = $this->encodeAttribute($note_id); |
|
1726
|
|
|
|
|
1727
|
|
|
// Prepare backlink, multiple backlinks if multiple references |
|
1728
|
|
|
// Do not create empty backlinks if the html is blank |
|
1729
|
1 |
|
$backlink = ""; |
|
1730
|
1 |
|
if (!empty($this->fn_backlink_html)) { |
|
1731
|
1 |
|
for ($ref_num = 1; $ref_num <= $ref_count; ++$ref_num) { |
|
1732
|
1 |
|
if (!empty($this->fn_backlink_title)) { |
|
1733
|
|
|
$attr['title'] = ' title="' . $this->encodeAttribute($this->fn_backlink_title) . '"'; |
|
1734
|
|
|
} |
|
1735
|
1 |
|
if (!empty($this->fn_backlink_label)) { |
|
1736
|
|
|
$attr['label'] = ' aria-label="' . $this->encodeAttribute($this->fn_backlink_label) . '"'; |
|
1737
|
|
|
} |
|
1738
|
1 |
|
$parsed_attr = $this->parseFootnotePlaceholders( |
|
1739
|
1 |
|
implode('', $attr), |
|
1740
|
1 |
|
$num, |
|
1741
|
1 |
|
$ref_num |
|
1742
|
|
|
); |
|
1743
|
1 |
|
$backlink_text = $this->parseFootnotePlaceholders( |
|
1744
|
1 |
|
$this->fn_backlink_html, |
|
1745
|
1 |
|
$num, |
|
1746
|
1 |
|
$ref_num |
|
1747
|
|
|
); |
|
1748
|
1 |
|
$ref_count_mark = $ref_num > 1 ? $ref_num : ''; |
|
1749
|
1 |
|
$backlink .= " <a href=\"#fnref$ref_count_mark:$note_id\"$parsed_attr>$backlink_text</a>"; |
|
1750
|
|
|
} |
|
1751
|
1 |
|
$backlink = trim($backlink); |
|
1752
|
|
|
} |
|
1753
|
|
|
|
|
1754
|
|
|
// Add backlink to last paragraph; create new paragraph if needed. |
|
1755
|
1 |
|
if (!empty($backlink)) { |
|
1756
|
1 |
|
if (preg_match('{</p>$}', $footnote)) { |
|
1757
|
1 |
|
$footnote = substr($footnote, 0, -4) . " $backlink</p>"; |
|
1758
|
|
|
} else { |
|
1759
|
1 |
|
$footnote .= "\n\n<p>$backlink</p>"; |
|
1760
|
|
|
} |
|
1761
|
|
|
} |
|
1762
|
|
|
|
|
1763
|
1 |
|
$text .= "<li id=\"fn:$note_id\" role=\"doc-endnote\">\n"; |
|
1764
|
1 |
|
$text .= $footnote . "\n"; |
|
1765
|
1 |
|
$text .= "</li>\n\n"; |
|
1766
|
|
|
} |
|
1767
|
1 |
|
$text .= "</ol>\n"; |
|
1768
|
|
|
|
|
1769
|
1 |
|
$this->footnotes_assembled = $text; |
|
1770
|
1 |
|
} |
|
1771
|
|
|
|
|
1772
|
|
|
/** |
|
1773
|
|
|
* Callback for appending footnotes |
|
1774
|
|
|
* @param array $matches |
|
1775
|
|
|
* @return string |
|
1776
|
|
|
*/ |
|
1777
|
1 |
|
protected function _appendFootnotes_callback($matches) { |
|
1778
|
1 |
|
$node_id = $this->fn_id_prefix . $matches[1]; |
|
1779
|
|
|
|
|
1780
|
|
|
// Create footnote marker only if it has a corresponding footnote *and* |
|
1781
|
|
|
// the footnote hasn't been used by another marker. |
|
1782
|
1 |
|
if (isset($this->footnotes[$node_id])) { |
|
1783
|
1 |
|
$num =& $this->footnotes_numbers[$node_id]; |
|
1784
|
1 |
|
if (!isset($num)) { |
|
1785
|
|
|
// Transfer footnote content to the ordered list and give it its |
|
1786
|
|
|
// number |
|
1787
|
1 |
|
$this->footnotes_ordered[$node_id] = $this->footnotes[$node_id]; |
|
1788
|
1 |
|
$this->footnotes_ref_count[$node_id] = 1; |
|
1789
|
1 |
|
$num = $this->footnote_counter++; |
|
1790
|
1 |
|
$ref_count_mark = ''; |
|
1791
|
|
|
} else { |
|
1792
|
1 |
|
$ref_count_mark = $this->footnotes_ref_count[$node_id] += 1; |
|
1793
|
|
|
} |
|
1794
|
|
|
|
|
1795
|
1 |
|
$attr = ""; |
|
1796
|
1 |
|
if ($this->fn_link_class !== "") { |
|
1797
|
1 |
|
$class = $this->fn_link_class; |
|
1798
|
1 |
|
$class = $this->encodeAttribute($class); |
|
1799
|
1 |
|
$attr .= " class=\"$class\""; |
|
1800
|
|
|
} |
|
1801
|
1 |
|
if ($this->fn_link_title !== "") { |
|
1802
|
|
|
$title = $this->fn_link_title; |
|
1803
|
|
|
$title = $this->encodeAttribute($title); |
|
1804
|
|
|
$attr .= " title=\"$title\""; |
|
1805
|
|
|
} |
|
1806
|
1 |
|
$attr .= " role=\"doc-noteref\""; |
|
1807
|
|
|
|
|
1808
|
1 |
|
$attr = str_replace("%%", $num, $attr); |
|
1809
|
1 |
|
$node_id = $this->encodeAttribute($node_id); |
|
1810
|
|
|
|
|
1811
|
|
|
return |
|
1812
|
1 |
|
"<sup id=\"fnref$ref_count_mark:$node_id\">". |
|
1813
|
1 |
|
"<a href=\"#fn:$node_id\"$attr>$num</a>". |
|
1814
|
1 |
|
"</sup>"; |
|
1815
|
|
|
} |
|
1816
|
|
|
|
|
1817
|
1 |
|
return "[^" . $matches[1] . "]"; |
|
1818
|
|
|
} |
|
1819
|
|
|
|
|
1820
|
|
|
/** |
|
1821
|
|
|
* Build footnote label by evaluating any placeholders. |
|
1822
|
|
|
* - ^^ footnote number |
|
1823
|
|
|
* - %% footnote reference number (Nth reference to footnote number) |
|
1824
|
|
|
* @param string $label |
|
1825
|
|
|
* @param int $footnote_number |
|
1826
|
|
|
* @param int $reference_number |
|
1827
|
|
|
* @return string |
|
1828
|
|
|
*/ |
|
1829
|
1 |
|
protected function parseFootnotePlaceholders($label, $footnote_number, $reference_number) { |
|
1830
|
1 |
|
return str_replace( |
|
1831
|
1 |
|
array('^^', '%%'), |
|
1832
|
1 |
|
array($footnote_number, $reference_number), |
|
1833
|
1 |
|
$label |
|
1834
|
|
|
); |
|
1835
|
|
|
} |
|
1836
|
|
|
|
|
1837
|
|
|
|
|
1838
|
|
|
/** |
|
1839
|
|
|
* Abbreviations - strips abbreviations from text, stores titles in hash |
|
1840
|
|
|
* references. |
|
1841
|
|
|
* @param string $text |
|
1842
|
|
|
* @return string |
|
1843
|
|
|
*/ |
|
1844
|
62 |
View Code Duplication |
protected function stripAbbreviations($text) { |
|
|
|
|
|
|
1845
|
62 |
|
$less_than_tab = $this->tab_width - 1; |
|
1846
|
|
|
|
|
1847
|
|
|
// Link defs are in the form: [id]*: url "optional title" |
|
1848
|
62 |
|
$text = preg_replace_callback('{ |
|
1849
|
62 |
|
^[ ]{0,' . $less_than_tab . '}\*\[(.+?)\][ ]?: # abbr_id = $1 |
|
1850
|
|
|
(.*) # text = $2 (no blank lines allowed) |
|
1851
|
|
|
}xm', |
|
1852
|
62 |
|
array($this, '_stripAbbreviations_callback'), |
|
1853
|
62 |
|
$text); |
|
1854
|
62 |
|
return $text; |
|
1855
|
|
|
} |
|
1856
|
|
|
|
|
1857
|
|
|
/** |
|
1858
|
|
|
* Callback for stripping abbreviations |
|
1859
|
|
|
* @param array $matches |
|
1860
|
|
|
* @return string |
|
1861
|
|
|
*/ |
|
1862
|
1 |
|
protected function _stripAbbreviations_callback($matches) { |
|
1863
|
1 |
|
$abbr_word = $matches[1]; |
|
1864
|
1 |
|
$abbr_desc = $matches[2]; |
|
1865
|
1 |
|
if ($this->abbr_word_re) { |
|
1866
|
1 |
|
$this->abbr_word_re .= '|'; |
|
1867
|
|
|
} |
|
1868
|
1 |
|
$this->abbr_word_re .= preg_quote($abbr_word); |
|
1869
|
1 |
|
$this->abbr_desciptions[$abbr_word] = trim($abbr_desc); |
|
1870
|
1 |
|
return ''; // String that will replace the block |
|
1871
|
|
|
} |
|
1872
|
|
|
|
|
1873
|
|
|
/** |
|
1874
|
|
|
* Find defined abbreviations in text and wrap them in <abbr> elements. |
|
1875
|
|
|
* @param string $text |
|
1876
|
|
|
* @return string |
|
1877
|
|
|
*/ |
|
1878
|
62 |
|
protected function doAbbreviations($text) { |
|
1879
|
62 |
|
if ($this->abbr_word_re) { |
|
1880
|
|
|
// cannot use the /x modifier because abbr_word_re may |
|
1881
|
|
|
// contain significant spaces: |
|
1882
|
3 |
|
$text = preg_replace_callback('{' . |
|
1883
|
|
|
'(?<![\w\x1A])' . |
|
1884
|
3 |
|
'(?:' . $this->abbr_word_re . ')' . |
|
1885
|
3 |
|
'(?![\w\x1A])' . |
|
1886
|
3 |
|
'}', |
|
1887
|
3 |
|
array($this, '_doAbbreviations_callback'), $text); |
|
1888
|
|
|
} |
|
1889
|
62 |
|
return $text; |
|
1890
|
|
|
} |
|
1891
|
|
|
|
|
1892
|
|
|
/** |
|
1893
|
|
|
* Callback for processing abbreviations |
|
1894
|
|
|
* @param array $matches |
|
1895
|
|
|
* @return string |
|
1896
|
|
|
*/ |
|
1897
|
3 |
|
protected function _doAbbreviations_callback($matches) { |
|
1898
|
3 |
|
$abbr = $matches[0]; |
|
1899
|
3 |
|
if (isset($this->abbr_desciptions[$abbr])) { |
|
1900
|
3 |
|
$desc = $this->abbr_desciptions[$abbr]; |
|
1901
|
3 |
|
if (empty($desc)) { |
|
1902
|
|
|
return $this->hashPart("<abbr>$abbr</abbr>"); |
|
1903
|
|
|
} |
|
1904
|
3 |
|
$desc = $this->encodeAttribute($desc); |
|
1905
|
3 |
|
return $this->hashPart("<abbr title=\"$desc\">$abbr</abbr>"); |
|
1906
|
|
|
} |
|
1907
|
|
|
return $matches[0]; |
|
1908
|
|
|
} |
|
1909
|
|
|
} |
|
1910
|
|
|
|
michelf /
php-markdown
Loading history...
Adding a
@returnannotation to a constructor is not recommended, since a constructor does not have a meaningful return value.Please refer to the PHP core documentation on constructors.