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