1
|
|
|
<?php |
2
|
|
|
|
3
|
|
|
namespace Mos\TextFilter; |
4
|
|
|
|
5
|
|
|
/** |
6
|
|
|
* Filter and format content. |
7
|
|
|
* |
8
|
|
|
*/ |
9
|
|
|
class CTextFilter |
10
|
|
|
{ |
11
|
|
|
/** |
12
|
|
|
* Supported filters. |
13
|
|
|
*/ |
14
|
|
|
private $filters = [ |
15
|
|
|
"jsonfrontmatter", |
16
|
|
|
"yamlfrontmatter", |
17
|
|
|
"bbcode", |
18
|
|
|
"clickable", |
19
|
|
|
"shortcode", |
20
|
|
|
"markdown", |
21
|
|
|
"nl2br", |
22
|
|
|
"purify", |
23
|
|
|
"titlefromh1", |
24
|
|
|
]; |
25
|
|
|
|
26
|
|
|
|
27
|
|
|
|
28
|
|
|
/** |
29
|
|
|
* Current document parsed. |
30
|
|
|
*/ |
31
|
|
|
private $current; |
32
|
|
|
|
33
|
|
|
|
34
|
|
|
|
35
|
|
|
/** |
36
|
|
|
* Call each filter. |
37
|
|
|
* |
38
|
|
|
* @deprecated deprecated since version 1.2 in favour of parse(). |
39
|
|
|
* |
40
|
|
|
* @param string $text the text to filter. |
41
|
|
|
* @param string|array $filters as comma separated list of filter, |
42
|
|
|
* or filters sent in as array. |
43
|
|
|
* |
44
|
|
|
* @return string the formatted text. |
45
|
|
|
*/ |
46
|
|
|
public function doFilter($text, $filters) |
47
|
|
|
{ |
48
|
|
|
// Define all valid filters with their callback function. |
49
|
|
|
$callbacks = [ |
50
|
|
|
'bbcode' => 'bbcode2html', |
51
|
|
|
'clickable' => 'makeClickable', |
52
|
|
|
'shortcode' => 'shortCode', |
53
|
|
|
'markdown' => 'markdown', |
54
|
|
|
'nl2br' => 'nl2br', |
55
|
|
|
'purify' => 'purify', |
56
|
|
|
]; |
57
|
|
|
|
58
|
|
|
// Make an array of the comma separated string $filters |
59
|
|
|
if (is_array($filters)) { |
60
|
|
|
$filter = $filters; |
61
|
|
|
} else { |
62
|
|
|
$filters = strtolower($filters); |
63
|
|
|
$filter = preg_replace('/\s/', '', explode(',', $filters)); |
64
|
|
|
} |
65
|
|
|
|
66
|
|
|
// For each filter, call its function with the $text as parameter. |
67
|
|
|
foreach ($filter as $key) { |
68
|
|
|
|
69
|
|
|
if (!isset($callbacks[$key])) { |
70
|
|
|
throw new Exception("The filter '$filters' is not a valid filter string due to '$key'."); |
71
|
|
|
} |
72
|
|
|
$text = call_user_func_array([$this, $callbacks[$key]], [$text]); |
73
|
|
|
} |
74
|
|
|
|
75
|
|
|
return $text; |
76
|
|
|
} |
77
|
|
|
|
78
|
|
|
|
79
|
|
|
|
80
|
|
|
/** |
81
|
|
|
* Return an array of all filters supported. |
82
|
|
|
* |
83
|
|
|
* @return array with strings of filters supported. |
84
|
|
|
*/ |
85
|
1 |
|
public function getFilters() |
86
|
|
|
{ |
87
|
1 |
|
return $this->filters; |
88
|
|
|
} |
89
|
|
|
|
90
|
|
|
|
91
|
|
|
|
92
|
|
|
/** |
93
|
|
|
* Check if filter is supported. |
94
|
|
|
* |
95
|
|
|
* @param string $filter to use. |
96
|
|
|
* |
97
|
|
|
* @throws mos/TextFilter/Exception when filter does not exists. |
98
|
|
|
* |
99
|
|
|
* @return boolean true if filter exists, false othwerwise. |
100
|
|
|
*/ |
101
|
2 |
|
public function hasFilter($filter) |
102
|
|
|
{ |
103
|
2 |
|
return in_array($filter, $this->filters); |
104
|
|
|
} |
105
|
|
|
|
106
|
|
|
|
107
|
|
|
|
108
|
|
|
/** |
109
|
|
|
* Add array items to frontmatter. |
110
|
|
|
* |
111
|
|
|
* @param array|null $matter key value array with items to add |
112
|
|
|
* or null if empty. |
113
|
|
|
* |
114
|
|
|
* @return $this |
115
|
|
|
*/ |
116
|
3 |
|
private function addToFrontmatter($matter) |
117
|
1 |
|
{ |
118
|
3 |
|
if (empty($matter)) { |
119
|
2 |
|
return $this; |
120
|
|
|
} |
121
|
|
|
|
122
|
2 |
|
if (is_null($this->current->frontmatter)) { |
123
|
2 |
|
$this->current->frontmatter = []; |
124
|
2 |
|
} |
125
|
|
|
|
126
|
2 |
|
$this->current->frontmatter = array_merge($this->current->frontmatter, $matter); |
127
|
2 |
|
return $this; |
128
|
|
|
} |
129
|
|
|
|
130
|
|
|
|
131
|
|
|
|
132
|
|
|
/** |
133
|
|
|
* Call a specific filter and store its details. |
134
|
|
|
* |
135
|
|
|
* @param string $filter to use. |
136
|
|
|
* |
137
|
|
|
* @throws mos/TextFilter/Exception when filter does not exists. |
138
|
|
|
* |
139
|
|
|
* @return string the formatted text. |
140
|
|
|
*/ |
141
|
4 |
|
private function parseFactory($filter) |
142
|
|
|
{ |
143
|
|
|
// Define single tasks filter with a callback. |
144
|
|
|
$callbacks = [ |
145
|
4 |
|
"bbcode" => "bbcode2html", |
146
|
4 |
|
"clickable" => "makeClickable", |
147
|
4 |
|
"shortcode" => "shortCode", |
148
|
4 |
|
"markdown" => "markdown", |
149
|
4 |
|
"nl2br" => "nl2br", |
150
|
4 |
|
"purify" => "purify", |
151
|
4 |
|
]; |
152
|
|
|
|
153
|
|
|
// Do the specific filter |
154
|
4 |
|
$text = $this->current->text; |
155
|
1 |
|
switch ($filter) { |
156
|
4 |
|
case "jsonfrontmatter": |
157
|
3 |
|
$res = $this->jsonFrontMatter($text); |
158
|
4 |
|
$this->current->text = $res["text"]; |
159
|
3 |
|
$this->addToFrontmatter($res["frontmatter"]); |
160
|
3 |
|
break; |
161
|
|
|
|
162
|
2 |
|
case "yamlfrontmatter": |
163
|
|
|
$res = $this->yamlFrontMatter($text); |
164
|
|
|
$this->current->text = $res["text"]; |
165
|
|
|
$this->addToFrontmatter($res["frontmatter"]); |
166
|
|
|
break; |
167
|
|
|
|
168
|
2 |
|
case "titlefromh1": |
169
|
1 |
|
$title = $this->getTitleFromFirstH1($text); |
170
|
1 |
|
$this->current->text = $text; |
171
|
1 |
|
if (!isset($this->current->frontmatter["title"])) { |
172
|
1 |
|
$this->addToFrontmatter(["title" => $title]); |
173
|
1 |
|
} |
174
|
2 |
|
break; |
175
|
|
|
|
176
|
2 |
|
case "bbcode": |
177
|
2 |
|
case "clickable": |
178
|
2 |
|
case "shortcode": |
179
|
2 |
|
case "markdown": |
180
|
2 |
|
case "nl2br": |
181
|
2 |
|
case "purify": |
182
|
2 |
|
$this->current->text = call_user_func_array( |
183
|
2 |
|
[$this, $callbacks[$filter]], |
184
|
2 |
|
[$text] |
185
|
2 |
|
); |
186
|
2 |
|
break; |
187
|
|
|
|
188
|
|
|
default: |
189
|
|
|
throw new Exception("The filter '$filter' is not a valid filter string."); |
190
|
|
|
} |
191
|
4 |
|
} |
192
|
|
|
|
193
|
|
|
|
194
|
|
|
|
195
|
|
|
/** |
196
|
|
|
* Call each filter and return array with details of the formatted content. |
197
|
|
|
* |
198
|
|
|
* @param string $text the text to filter. |
199
|
|
|
* @param array $filter array of filters to use. |
200
|
|
|
* |
201
|
|
|
* @throws mos/TextFilter/Exception when filterd does not exists. |
202
|
|
|
* |
203
|
|
|
* @return array with the formatted text and additional details. |
204
|
|
|
*/ |
205
|
4 |
|
public function parse($text, $filter) |
206
|
|
|
{ |
207
|
4 |
|
$this->current = new \stdClass(); |
208
|
4 |
|
$this->current->frontmatter = null; |
209
|
4 |
|
$this->current->text = $text; |
210
|
|
|
|
211
|
4 |
|
foreach ($filter as $key) { |
212
|
4 |
|
$this->parseFactory($key); |
213
|
4 |
|
} |
214
|
|
|
|
215
|
4 |
|
return $this->current; |
216
|
|
|
} |
217
|
|
|
|
218
|
|
|
|
219
|
|
|
|
220
|
|
|
/** |
221
|
|
|
* Extract front matter from text. |
222
|
|
|
* |
223
|
|
|
* @param string $text the text to be parsed. |
224
|
|
|
* @param string $startToken the start token. |
225
|
|
|
* @param string $stopToken the stop token. |
226
|
|
|
* |
227
|
|
|
* @return array with the formatted text and the front matter. |
228
|
|
|
*/ |
229
|
3 |
|
private function extractFrontMatter($text, $startToken, $stopToken) |
230
|
|
|
{ |
231
|
3 |
|
$tokenLength = strlen($startToken); |
232
|
|
|
|
233
|
3 |
|
$start = strpos($text, $startToken); |
234
|
|
|
// Is a valid start? |
235
|
3 |
|
if ($start !== false && $start !== 0) { |
236
|
|
|
if ($text[$start - 1] !== "\n") { |
237
|
|
|
$start = false; |
238
|
|
|
} |
239
|
|
|
} |
240
|
|
|
|
241
|
3 |
|
$frontmatter = null; |
242
|
3 |
|
if ($start !== false) { |
243
|
3 |
|
$stop = strpos($text, $stopToken, $tokenLength - 1); |
244
|
|
|
|
245
|
3 |
|
if ($stop !== false && $text[$stop - 1] === "\n") { |
246
|
2 |
|
$length = $stop - ($start + $tokenLength); |
247
|
|
|
|
248
|
2 |
|
$frontmatter = substr($text, $start + $tokenLength, $length); |
249
|
2 |
|
$textStart = substr($text, 0, $start); |
250
|
2 |
|
$text = $textStart . substr($text, $stop + $tokenLength); |
251
|
2 |
|
} |
252
|
3 |
|
} |
253
|
|
|
|
254
|
3 |
|
return [$text, $frontmatter]; |
255
|
|
|
} |
256
|
|
|
|
257
|
|
|
|
258
|
|
|
|
259
|
|
|
/** |
260
|
|
|
* Extract JSON front matter from text. |
261
|
|
|
* |
262
|
|
|
* @param string $text the text to be parsed. |
263
|
|
|
* |
264
|
|
|
* @return array with the formatted text and the front matter. |
265
|
|
|
*/ |
266
|
3 |
|
public function jsonFrontMatter($text) |
267
|
|
|
{ |
268
|
3 |
|
list($text, $frontmatter) = $this->extractFrontMatter($text, "{{{\n", "}}}\n"); |
269
|
|
|
|
270
|
3 |
|
if (!empty($frontmatter)) { |
271
|
2 |
|
$frontmatter = json_decode($frontmatter, true); |
272
|
|
|
|
273
|
2 |
|
if (is_null($frontmatter)) { |
274
|
|
|
throw new Exception("Failed parsing JSON frontmatter."); |
275
|
|
|
} |
276
|
2 |
|
} |
277
|
|
|
|
278
|
|
|
return [ |
279
|
3 |
|
"text" => $text, |
280
|
|
|
"frontmatter" => $frontmatter |
281
|
3 |
|
]; |
282
|
|
|
} |
283
|
|
|
|
284
|
|
|
|
285
|
|
|
|
286
|
|
|
/** |
287
|
|
|
* Extract YAML front matter from text. |
288
|
|
|
* |
289
|
|
|
* @param string $text the text to be parsed. |
290
|
|
|
* |
291
|
|
|
* @return array with the formatted text and the front matter. |
292
|
|
|
*/ |
293
|
|
|
public function yamlFrontMatter($text) |
294
|
|
|
{ |
295
|
|
|
$needle = "---\n"; |
296
|
|
|
list($text, $frontmatter) = $this->extractFrontMatter($text, $needle, $needle); |
297
|
|
|
|
298
|
|
|
if (function_exists("yaml_parse") && !empty($frontmatter)) { |
299
|
|
|
$frontmatter = yaml_parse($needle . $frontmatter); |
300
|
|
|
|
301
|
|
|
if ($frontmatter === false) { |
302
|
|
|
throw new Exception("Failed parsing YAML frontmatter."); |
303
|
|
|
} |
304
|
|
|
} |
305
|
|
|
|
306
|
|
|
return [ |
307
|
|
|
"text" => $text, |
308
|
|
|
"frontmatter" => $frontmatter |
309
|
|
|
]; |
310
|
|
|
} |
311
|
|
|
|
312
|
|
|
|
313
|
|
|
|
314
|
|
|
/** |
315
|
|
|
* Get the title from the first H1. |
316
|
|
|
* |
317
|
|
|
* @param string $text the text to be parsed. |
318
|
|
|
* |
319
|
|
|
* @return string|null with the title, if its found. |
320
|
|
|
*/ |
321
|
1 |
|
public function getTitleFromFirstH1($text) |
322
|
|
|
{ |
323
|
1 |
|
$matches = []; |
324
|
1 |
|
$title = null; |
325
|
|
|
|
326
|
1 |
|
if (preg_match("#<h1.*?>(.*)</h1>#", $text, $matches)) { |
327
|
1 |
|
$title = strip_tags($matches[1]); |
328
|
1 |
|
} |
329
|
|
|
|
330
|
1 |
|
return $title; |
331
|
|
|
} |
332
|
|
|
|
333
|
|
|
|
334
|
|
|
|
335
|
|
|
/** |
336
|
|
|
* Helper, BBCode formatting converting to HTML. |
337
|
|
|
* |
338
|
|
|
* @param string $text The text to be converted. |
339
|
|
|
* |
340
|
|
|
* @return string the formatted text. |
341
|
|
|
* |
342
|
|
|
* @link http://dbwebb.se/coachen/reguljara-uttryck-i-php-ger-bbcode-formattering |
343
|
|
|
*/ |
344
|
3 |
|
public function bbcode2html($text) |
345
|
|
|
{ |
346
|
|
|
$search = [ |
347
|
3 |
|
'/\[b\](.*?)\[\/b\]/is', |
348
|
3 |
|
'/\[i\](.*?)\[\/i\]/is', |
349
|
3 |
|
'/\[u\](.*?)\[\/u\]/is', |
350
|
3 |
|
'/\[img\](https?.*?)\[\/img\]/is', |
351
|
3 |
|
'/\[url\](https?.*?)\[\/url\]/is', |
352
|
|
|
'/\[url=(https?.*?)\](.*?)\[\/url\]/is' |
353
|
3 |
|
]; |
354
|
|
|
|
355
|
|
|
$replace = [ |
356
|
3 |
|
'<strong>$1</strong>', |
357
|
3 |
|
'<em>$1</em>', |
358
|
3 |
|
'<u>$1</u>', |
359
|
3 |
|
'<img src="$1" />', |
360
|
3 |
|
'<a href="$1">$1</a>', |
361
|
|
|
'<a href="$1">$2</a>' |
362
|
3 |
|
]; |
363
|
|
|
|
364
|
3 |
|
return preg_replace($search, $replace, $text); |
365
|
|
|
} |
366
|
|
|
|
367
|
|
|
|
368
|
|
|
|
369
|
|
|
/** |
370
|
|
|
* Make clickable links from URLs in text. |
371
|
|
|
* |
372
|
|
|
* @param string $text the text that should be formatted. |
373
|
|
|
* |
374
|
|
|
* @return string with formatted anchors. |
375
|
|
|
* |
376
|
|
|
* @link http://dbwebb.se/coachen/lat-php-funktion-make-clickable-automatiskt-skapa-klickbara-lankar |
377
|
|
|
*/ |
378
|
1 |
|
public function makeClickable($text) |
379
|
|
|
{ |
380
|
1 |
|
return preg_replace_callback( |
381
|
1 |
|
'#\b(?<![href|src]=[\'"])https?://[^\s()<>]+(?:\([\w\d]+\)|([^[:punct:]\s]|/))#', |
382
|
|
|
function ($matches) { |
383
|
1 |
|
return "<a href='{$matches[0]}'>{$matches[0]}</a>"; |
384
|
1 |
|
}, |
385
|
|
|
$text |
386
|
1 |
|
); |
387
|
|
|
} |
388
|
|
|
|
389
|
|
|
|
390
|
|
|
|
391
|
|
|
/** |
392
|
|
|
* Format text according to HTML Purifier. |
393
|
|
|
* |
394
|
|
|
* @param string $text that should be formatted. |
395
|
|
|
* |
396
|
|
|
* @return string as the formatted html-text. |
397
|
|
|
*/ |
398
|
1 |
|
public function purify($text) |
399
|
|
|
{ |
400
|
1 |
|
$config = \HTMLPurifier_Config::createDefault(); |
401
|
1 |
|
$config->set("Cache.DefinitionImpl", null); |
402
|
|
|
//$config->set('Cache.SerializerPath', '/home/user/absolute/path'); |
403
|
|
|
|
404
|
1 |
|
$purifier = new \HTMLPurifier($config); |
405
|
|
|
|
406
|
1 |
|
return $purifier->purify($text); |
407
|
|
|
} |
408
|
|
|
|
409
|
|
|
|
410
|
|
|
|
411
|
|
|
/** |
412
|
|
|
* Format text according to Markdown syntax. |
413
|
|
|
* |
414
|
|
|
* @param string $text the text that should be formatted. |
415
|
|
|
* |
416
|
|
|
* @return string as the formatted html-text. |
417
|
|
|
*/ |
418
|
7 |
|
public function markdown($text) |
419
|
|
|
{ |
420
|
7 |
|
return \Michelf\MarkdownExtra::defaultTransform($text); |
421
|
|
|
} |
422
|
|
|
|
423
|
|
|
|
424
|
|
|
|
425
|
|
|
/** |
426
|
|
|
* For convenience access to nl2br |
427
|
|
|
* |
428
|
|
|
* @param string $text text to be converted. |
429
|
|
|
* |
430
|
|
|
* @return string the formatted text. |
431
|
|
|
*/ |
432
|
1 |
|
public function nl2br($text) |
433
|
|
|
{ |
434
|
1 |
|
return nl2br($text); |
435
|
|
|
} |
436
|
|
|
|
437
|
|
|
|
438
|
|
|
|
439
|
|
|
/** |
440
|
|
|
* Shortcode to to quicker format text as HTML. |
441
|
|
|
* |
442
|
|
|
* @param string $text text to be converted. |
443
|
|
|
* |
444
|
|
|
* @return string the formatted text. |
445
|
|
|
*/ |
446
|
1 |
|
public function shortCode($text) |
447
|
|
|
{ |
448
|
|
|
$patterns = [ |
449
|
1 |
|
'/\[(FIGURE)[\s+](.+)\]/', |
450
|
1 |
|
]; |
451
|
|
|
|
452
|
1 |
|
return preg_replace_callback( |
453
|
1 |
|
$patterns, |
454
|
1 |
|
function ($matches) { |
455
|
1 |
|
switch ($matches[1]) { |
456
|
|
|
|
457
|
1 |
|
case 'FIGURE': |
458
|
1 |
|
return self::ShortCodeFigure($matches[2]); |
459
|
|
|
break; |
|
|
|
|
460
|
|
|
|
461
|
|
|
default: |
462
|
|
|
return "{$matches[1]} is unknown shortcode."; |
463
|
|
|
} |
464
|
1 |
|
}, |
465
|
|
|
$text |
466
|
1 |
|
); |
467
|
|
|
} |
468
|
|
|
|
469
|
|
|
|
470
|
|
|
|
471
|
|
|
/** |
472
|
|
|
* Init shortcode handling by preparing the option list to an array, for those using arguments. |
473
|
|
|
* |
474
|
|
|
* @param string $options for the shortcode. |
475
|
|
|
* |
476
|
|
|
* @return array with all the options. |
477
|
|
|
*/ |
478
|
1 |
|
public static function shortCodeInit($options) |
479
|
|
|
{ |
480
|
1 |
|
preg_match_all('/[a-zA-Z0-9]+="[^"]+"|\S+/', $options, $matches); |
481
|
|
|
|
482
|
1 |
|
$res = array(); |
483
|
1 |
|
foreach ($matches[0] as $match) { |
484
|
1 |
|
$pos = strpos($match, '='); |
485
|
1 |
|
if ($pos === false) { |
486
|
|
|
$res[$match] = true; |
487
|
|
|
} else { |
488
|
1 |
|
$key = substr($match, 0, $pos); |
489
|
1 |
|
$val = trim(substr($match, $pos+1), '"'); |
490
|
1 |
|
$res[$key] = $val; |
491
|
|
|
} |
492
|
1 |
|
} |
493
|
|
|
|
494
|
1 |
|
return $res; |
495
|
|
|
} |
496
|
|
|
|
497
|
|
|
|
498
|
|
|
|
499
|
|
|
/** |
500
|
|
|
* Shortcode for <figure>. |
501
|
|
|
* |
502
|
|
|
* Usage example: [FIGURE src="img/home/me.jpg" caption="Me" alt="Bild på mig" nolink="nolink"] |
503
|
|
|
* |
504
|
|
|
* @param string $options for the shortcode. |
505
|
|
|
* |
506
|
|
|
* @return array with all the options. |
507
|
|
|
*/ |
508
|
1 |
|
public static function shortCodeFigure($options) |
509
|
|
|
{ |
510
|
|
|
// Merge incoming options with default and expose as variables |
511
|
1 |
|
$options= array_merge( |
512
|
|
|
[ |
513
|
1 |
|
'id' => null, |
514
|
1 |
|
'class' => null, |
515
|
1 |
|
'src' => null, |
516
|
1 |
|
'title' => null, |
517
|
1 |
|
'alt' => null, |
518
|
1 |
|
'caption' => null, |
519
|
1 |
|
'href' => null, |
520
|
1 |
|
'nolink' => false, |
521
|
1 |
|
], |
522
|
1 |
|
self::ShortCodeInit($options) |
523
|
1 |
|
); |
524
|
1 |
|
extract($options, EXTR_SKIP); |
525
|
|
|
|
526
|
1 |
|
$id = $id ? " id='$id'" : null; |
527
|
1 |
|
$class = $class ? " class='figure $class'" : " class='figure'"; |
528
|
1 |
|
$title = $title ? " title='$title'" : null; |
529
|
|
|
|
530
|
1 |
|
if (!$alt && $caption) { |
531
|
1 |
|
$alt = $caption; |
532
|
1 |
|
} |
533
|
|
|
|
534
|
1 |
|
if (!$href) { |
535
|
1 |
|
$pos = strpos($src, '?'); |
536
|
1 |
|
$href = $pos ? substr($src, 0, $pos) : $src; |
537
|
1 |
|
} |
538
|
|
|
|
539
|
1 |
|
$start = null; |
540
|
1 |
|
$end = null; |
541
|
1 |
|
if (!$nolink) { |
542
|
1 |
|
$start = "<a href='{$href}'>"; |
543
|
1 |
|
$end = "</a>"; |
544
|
1 |
|
} |
545
|
|
|
|
546
|
|
|
$html = <<<EOD |
547
|
1 |
|
<figure{$id}{$class}> |
548
|
1 |
|
{$start}<img src='{$src}' alt='{$alt}'{$title}/>{$end} |
549
|
1 |
|
<figcaption markdown=1>{$caption}</figcaption> |
550
|
1 |
|
</figure> |
551
|
1 |
|
EOD; |
552
|
|
|
|
553
|
1 |
|
return $html; |
554
|
|
|
} |
555
|
|
|
} |
556
|
|
|
|
The break statement is not necessary if it is preceded for example by a return statement:
If you would like to keep this construct to be consistent with other case statements, you can safely mark this issue as a false-positive.