|
1
|
|
|
<?php |
|
2
|
|
|
/** |
|
3
|
|
|
* SEOmatic plugin for Craft CMS |
|
4
|
|
|
* |
|
5
|
|
|
* A turnkey SEO implementation for Craft CMS that is comprehensive, powerful, |
|
6
|
|
|
* and flexible |
|
7
|
|
|
* |
|
8
|
|
|
* @link https://nystudio107.com |
|
9
|
|
|
* @copyright Copyright (c) 2017 nystudio107 |
|
10
|
|
|
*/ |
|
11
|
|
|
|
|
12
|
|
|
namespace nystudio107\seomatic\helpers; |
|
13
|
|
|
|
|
14
|
|
|
use ArrayAccess; |
|
15
|
|
|
use benf\neo\elements\Block as NeoBlock; |
|
16
|
|
|
use benf\neo\elements\db\BlockQuery as NeoBlockQuery; |
|
17
|
|
|
use craft\elements\db\MatrixBlockQuery; |
|
18
|
|
|
use craft\elements\db\TagQuery; |
|
19
|
|
|
use craft\elements\MatrixBlock; |
|
20
|
|
|
use craft\elements\Tag; |
|
21
|
|
|
use craft\helpers\HtmlPurifier; |
|
22
|
|
|
use craft\models\FieldLayout; |
|
23
|
|
|
use nystudio107\seomatic\helpers\Field as FieldHelper; |
|
24
|
|
|
use nystudio107\seomatic\Seomatic; |
|
25
|
|
|
use PhpScience\TextRank\TextRankFacade; |
|
26
|
|
|
use PhpScience\TextRank\Tool\StopWords\StopWordsAbstract; |
|
27
|
|
|
use Stringy\Stringy; |
|
28
|
|
|
use Traversable; |
|
29
|
|
|
use verbb\doxter\Doxter; |
|
30
|
|
|
use verbb\doxter\fields\data\DoxterData; |
|
31
|
|
|
use verbb\supertable\elements\db\SuperTableBlockQuery; |
|
32
|
|
|
use verbb\supertable\elements\SuperTableBlockElement as SuperTableBlock; |
|
33
|
|
|
use yii\base\InvalidConfigException; |
|
34
|
|
|
use function array_slice; |
|
35
|
|
|
use function function_exists; |
|
36
|
|
|
use function is_array; |
|
37
|
|
|
|
|
38
|
|
|
/** |
|
39
|
|
|
* @author nystudio107 |
|
40
|
|
|
* @package Seomatic |
|
41
|
|
|
* @since 3.0.0 |
|
42
|
|
|
*/ |
|
43
|
|
|
class Text |
|
44
|
|
|
{ |
|
45
|
|
|
// Constants |
|
46
|
|
|
// ========================================================================= |
|
47
|
|
|
|
|
48
|
|
|
public const LANGUAGE_MAP = [ |
|
49
|
|
|
'en' => 'English', |
|
50
|
|
|
'fr' => 'French', |
|
51
|
|
|
'de' => 'German', |
|
52
|
|
|
'it' => 'Italian', |
|
53
|
|
|
'no' => 'Norwegian', |
|
54
|
|
|
'es' => 'Spanish', |
|
55
|
|
|
]; |
|
56
|
|
|
|
|
57
|
|
|
// Public Static Methods |
|
58
|
|
|
// ========================================================================= |
|
59
|
|
|
|
|
60
|
|
|
/** |
|
61
|
|
|
* Truncates the string to a given length. If $substring is provided, and |
|
62
|
|
|
* truncating occurs, the string is further truncated so that the substring |
|
63
|
|
|
* may be appended without exceeding the desired length. |
|
64
|
|
|
* |
|
65
|
|
|
* @param string $string The string to truncate |
|
66
|
|
|
* @param int $length Desired length of the truncated string |
|
67
|
|
|
* @param string $substring The substring to append if it can fit |
|
68
|
|
|
* |
|
69
|
|
|
* @return string with the resulting $str after truncating |
|
70
|
|
|
*/ |
|
71
|
|
|
public static function truncate($string, $length, $substring = '…'): string |
|
72
|
|
|
{ |
|
73
|
|
|
$result = $string; |
|
74
|
|
|
|
|
75
|
|
|
if (!empty($string)) { |
|
76
|
|
|
$string = HtmlPurifier::process($string, ['HTML.Allowed' => '']); |
|
77
|
|
|
$string = html_entity_decode($string, ENT_NOQUOTES, 'UTF-8'); |
|
78
|
|
|
$result = (string)Stringy::create($string)->truncate($length, $substring); |
|
79
|
|
|
} |
|
80
|
|
|
|
|
81
|
|
|
return $result; |
|
82
|
|
|
} |
|
83
|
|
|
|
|
84
|
|
|
/** |
|
85
|
|
|
* Truncates the string to a given length, while ensuring that it does not |
|
86
|
|
|
* split words. If $substring is provided, and truncating occurs, the |
|
87
|
|
|
* string is further truncated so that the substring may be appended without |
|
88
|
|
|
* exceeding the desired length. |
|
89
|
|
|
* |
|
90
|
|
|
* @param string $string The string to truncate |
|
91
|
|
|
* @param int $length Desired length of the truncated string |
|
92
|
|
|
* @param string $substring The substring to append if it can fit |
|
93
|
|
|
* |
|
94
|
|
|
* @return string with the resulting $str after truncating |
|
95
|
|
|
*/ |
|
96
|
1 |
|
public static function truncateOnWord($string, $length, $substring = '…'): string |
|
97
|
|
|
{ |
|
98
|
1 |
|
$result = $string; |
|
99
|
|
|
|
|
100
|
1 |
|
if (!empty($string)) { |
|
101
|
1 |
|
$string = HtmlPurifier::process($string, ['HTML.Allowed' => '']); |
|
102
|
1 |
|
$string = html_entity_decode($string, ENT_NOQUOTES, 'UTF-8'); |
|
103
|
1 |
|
$result = (string)Stringy::create($string)->safeTruncate($length, $substring); |
|
104
|
|
|
} |
|
105
|
|
|
|
|
106
|
1 |
|
return $result; |
|
107
|
|
|
} |
|
108
|
|
|
|
|
109
|
|
|
/** |
|
110
|
|
|
* Extract plain old text from a field |
|
111
|
|
|
* |
|
112
|
|
|
* @param $field |
|
113
|
|
|
* |
|
114
|
|
|
* @return string |
|
115
|
|
|
*/ |
|
116
|
|
|
public static function extractTextFromField($field): string |
|
117
|
|
|
{ |
|
118
|
|
|
if (empty($field)) { |
|
119
|
|
|
return ''; |
|
120
|
|
|
} |
|
121
|
|
|
if ($field instanceof MatrixBlockQuery |
|
122
|
|
|
|| (self::isArrayLike($field) && $field[0] instanceof MatrixBlock)) { |
|
123
|
|
|
$result = self::extractTextFromMatrix($field); |
|
124
|
|
|
} elseif ($field instanceof NeoBlockQuery |
|
125
|
|
|
|| (self::isArrayLike($field) && $field[0] instanceof NeoBlock)) { |
|
126
|
|
|
$result = self::extractTextFromNeo($field); |
|
127
|
|
|
} elseif ($field instanceof SuperTableBlockQuery |
|
128
|
|
|
|| (self::isArrayLike($field) && $field[0] instanceof SuperTableBlock)) { |
|
129
|
|
|
$result = self::extractTextFromSuperTable($field); |
|
130
|
|
|
} elseif ($field instanceof TagQuery |
|
131
|
|
|
|| (self::isArrayLike($field) && $field[0] instanceof Tag)) { |
|
132
|
|
|
$result = self::extractTextFromTags($field); |
|
133
|
|
|
} elseif ($field instanceof DoxterData) { |
|
134
|
|
|
$result = self::smartStripTags(Doxter::$plugin->getService()->parseMarkdown($field->getRaw())); |
|
135
|
|
|
} else { |
|
136
|
|
|
if (self::isArrayLike($field)) { |
|
137
|
|
|
$result = self::smartStripTags((string)$field[0]); |
|
138
|
|
|
} else { |
|
139
|
|
|
$result = self::smartStripTags((string)$field); |
|
140
|
|
|
} |
|
141
|
|
|
} |
|
142
|
|
|
|
|
143
|
|
|
//return $result; |
|
144
|
|
|
return self::sanitizeUserInput($result); |
|
145
|
|
|
} |
|
146
|
|
|
|
|
147
|
|
|
/** |
|
148
|
|
|
* Extract concatenated text from all of the tags in the $tagElement and |
|
149
|
|
|
* return as a comma-delimited string |
|
150
|
|
|
* |
|
151
|
|
|
* @param TagQuery|Tag[]|array $tags |
|
152
|
|
|
* |
|
153
|
|
|
* @return string |
|
154
|
|
|
*/ |
|
155
|
|
|
public static function extractTextFromTags($tags): string |
|
156
|
|
|
{ |
|
157
|
|
|
if (empty($tags)) { |
|
158
|
|
|
return ''; |
|
159
|
|
|
} |
|
160
|
|
|
$result = ''; |
|
161
|
|
|
// Iterate through all of the matrix blocks |
|
162
|
|
|
if ($tags instanceof TagQuery) { |
|
163
|
|
|
$tags = $tags->all(); |
|
164
|
|
|
} |
|
165
|
|
|
foreach ($tags as $tag) { |
|
166
|
|
|
$result .= $tag->title . ', '; |
|
167
|
|
|
} |
|
168
|
|
|
$result = rtrim($result, ', '); |
|
169
|
|
|
|
|
170
|
|
|
return $result; |
|
171
|
|
|
} |
|
172
|
|
|
|
|
173
|
|
|
/** |
|
174
|
|
|
* Extract text from all of the blocks in a matrix field, concatenating it |
|
175
|
|
|
* together. |
|
176
|
|
|
* |
|
177
|
|
|
* @param MatrixBlockQuery|MatrixBlock[]|array $blocks |
|
178
|
|
|
* @param string $fieldHandle |
|
179
|
|
|
* |
|
180
|
|
|
* @return string |
|
181
|
|
|
*/ |
|
182
|
|
|
public static function extractTextFromMatrix($blocks, $fieldHandle = ''): string |
|
183
|
|
|
{ |
|
184
|
|
|
if (empty($blocks)) { |
|
185
|
|
|
return ''; |
|
186
|
|
|
} |
|
187
|
|
|
$result = ''; |
|
188
|
|
|
// Iterate through all of the matrix blocks |
|
189
|
|
|
if ($blocks instanceof MatrixBlockQuery) { |
|
190
|
|
|
$blocks = $blocks->all(); |
|
191
|
|
|
} |
|
192
|
|
|
foreach ($blocks as $block) { |
|
193
|
|
|
try { |
|
194
|
|
|
$matrixBlockTypeModel = $block->getType(); |
|
195
|
|
|
} catch (InvalidConfigException $e) { |
|
196
|
|
|
$matrixBlockTypeModel = null; |
|
197
|
|
|
} |
|
198
|
|
|
// Find any text fields inside of the matrix block |
|
199
|
|
|
if ($matrixBlockTypeModel) { |
|
200
|
|
|
$fieldClasses = FieldHelper::FIELD_CLASSES[FieldHelper::TEXT_FIELD_CLASS_KEY]; |
|
201
|
|
|
$fields = $matrixBlockTypeModel->getCustomFields(); |
|
202
|
|
|
|
|
203
|
|
|
foreach ($fields as $field) { |
|
204
|
|
|
/** @var array $fieldClasses */ |
|
205
|
|
|
foreach ($fieldClasses as $fieldClassKey) { |
|
206
|
|
|
if ($field instanceof $fieldClassKey) { |
|
207
|
|
|
if ($field->handle === $fieldHandle || empty($fieldHandle)) { |
|
208
|
|
|
$result .= self::extractTextFromField($block[$field->handle]) . ' '; |
|
209
|
|
|
} |
|
210
|
|
|
} |
|
211
|
|
|
} |
|
212
|
|
|
} |
|
213
|
|
|
} |
|
214
|
|
|
} |
|
215
|
|
|
|
|
216
|
|
|
return $result; |
|
217
|
|
|
} |
|
218
|
|
|
|
|
219
|
|
|
/** |
|
220
|
|
|
* Extract text from all of the blocks in a Neo field, concatenating it |
|
221
|
|
|
* together. |
|
222
|
|
|
* |
|
223
|
|
|
* @param NeoBlockQuery|NeoBlock[]|array $blocks |
|
224
|
|
|
* @param string $fieldHandle |
|
225
|
|
|
* |
|
226
|
|
|
* @return string |
|
227
|
|
|
*/ |
|
228
|
|
|
public static function extractTextFromNeo($blocks, $fieldHandle = ''): string |
|
229
|
|
|
{ |
|
230
|
|
|
if (empty($blocks)) { |
|
231
|
|
|
return ''; |
|
232
|
|
|
} |
|
233
|
|
|
$result = ''; |
|
234
|
|
|
// Iterate through all of the matrix blocks |
|
235
|
|
|
if ($blocks instanceof NeoBlockQuery) { |
|
236
|
|
|
$blocks = $blocks->all(); |
|
237
|
|
|
} |
|
238
|
|
|
foreach ($blocks as $block) { |
|
239
|
|
|
$layout = $block->getFieldLayout(); |
|
240
|
|
|
// Find any text fields inside of the neo block |
|
241
|
|
|
if ($layout) { |
|
242
|
|
|
$fieldClasses = FieldHelper::FIELD_CLASSES[FieldHelper::TEXT_FIELD_CLASS_KEY]; |
|
243
|
|
|
$fieldElements = $layout->getCustomFieldElements(); |
|
244
|
|
|
foreach ($fieldElements as $fieldElement) { |
|
245
|
|
|
$field = $fieldElement->getField(); |
|
246
|
|
|
/** @var array $fieldClasses */ |
|
247
|
|
|
foreach ($fieldClasses as $fieldClassKey) { |
|
248
|
|
|
if ($field instanceof $fieldClassKey) { |
|
249
|
|
|
if ($field->handle === $fieldHandle || empty($fieldHandle)) { |
|
250
|
|
|
$result .= self::extractTextFromField($block[$field->handle]) . ' '; |
|
251
|
|
|
} |
|
252
|
|
|
} |
|
253
|
|
|
} |
|
254
|
|
|
} |
|
255
|
|
|
} |
|
256
|
|
|
} |
|
257
|
|
|
|
|
258
|
|
|
return $result; |
|
259
|
|
|
} |
|
260
|
|
|
|
|
261
|
|
|
/** |
|
262
|
|
|
* Extract text from all of the blocks in a matrix field, concatenating it |
|
263
|
|
|
* together. |
|
264
|
|
|
* |
|
265
|
|
|
* @param SuperTableBlockQuery|SuperTableBlock[]|array $blocks |
|
266
|
|
|
* @param string $fieldHandle |
|
267
|
|
|
* |
|
268
|
|
|
* @return string |
|
269
|
|
|
*/ |
|
270
|
|
|
public static function extractTextFromSuperTable($blocks, $fieldHandle = ''): string |
|
271
|
|
|
{ |
|
272
|
|
|
if (empty($blocks)) { |
|
273
|
|
|
return ''; |
|
274
|
|
|
} |
|
275
|
|
|
$result = ''; |
|
276
|
|
|
// Iterate through all of the supertable blocks |
|
277
|
|
|
if ($blocks instanceof SuperTableBlockQuery) { |
|
278
|
|
|
$blocks = $blocks->all(); |
|
279
|
|
|
} |
|
280
|
|
|
foreach ($blocks as $block) { |
|
281
|
|
|
try { |
|
282
|
|
|
$superTableBlockTypeModel = $block->getType(); |
|
283
|
|
|
} catch (InvalidConfigException $e) { |
|
284
|
|
|
$superTableBlockTypeModel = null; |
|
285
|
|
|
} |
|
286
|
|
|
// Find any text fields inside of the matrix block |
|
287
|
|
|
if ($superTableBlockTypeModel) { |
|
288
|
|
|
$fieldClasses = FieldHelper::FIELD_CLASSES[FieldHelper::TEXT_FIELD_CLASS_KEY]; |
|
289
|
|
|
/** @var ?FieldLayout $layout */ |
|
290
|
|
|
// The SuperTableBlockType class lacks @mixin FieldLayoutBehavior in its annotations |
|
291
|
|
|
/** @phpstan-ignore-next-line */ |
|
292
|
|
|
$layout = $superTableBlockTypeModel->getFieldLayout(); |
|
293
|
|
|
$fieldElements = $layout->getCustomFieldElements(); |
|
294
|
|
|
foreach ($fieldElements as $fieldElement) { |
|
295
|
|
|
$field = $fieldElement->getField(); |
|
296
|
|
|
/** @var array $fieldClasses */ |
|
297
|
|
|
foreach ($fieldClasses as $fieldClassKey) { |
|
298
|
|
|
if ($field instanceof $fieldClassKey) { |
|
299
|
|
|
if ($field->handle === $fieldHandle || empty($fieldHandle)) { |
|
300
|
|
|
$result .= self::extractTextFromField($block[$field->handle]) . ' '; |
|
301
|
|
|
} |
|
302
|
|
|
} |
|
303
|
|
|
} |
|
304
|
|
|
} |
|
305
|
|
|
} |
|
306
|
|
|
} |
|
307
|
|
|
|
|
308
|
|
|
return $result; |
|
309
|
|
|
} |
|
310
|
|
|
|
|
311
|
|
|
/** |
|
312
|
|
|
* Return the most important keywords extracted from the text as a comma- |
|
313
|
|
|
* delimited string |
|
314
|
|
|
* |
|
315
|
|
|
* @param string $text |
|
316
|
|
|
* @param int $limit |
|
317
|
|
|
* @param bool $useStopWords |
|
318
|
|
|
* |
|
319
|
|
|
* @return string |
|
320
|
|
|
*/ |
|
321
|
|
|
public static function extractKeywords($text, $limit = 15, $useStopWords = true): string |
|
322
|
|
|
{ |
|
323
|
|
|
if (empty($text)) { |
|
324
|
|
|
return ''; |
|
325
|
|
|
} |
|
326
|
|
|
$api = new TextRankFacade(); |
|
327
|
|
|
// Set the stop words that should be ignored |
|
328
|
|
|
if ($useStopWords) { |
|
329
|
|
|
$language = strtolower(substr(Seomatic::$language, 0, 2)); |
|
330
|
|
|
$stopWords = self::stopWordsForLanguage($language); |
|
331
|
|
|
if ($stopWords !== null) { |
|
332
|
|
|
$api->setStopWords($stopWords); |
|
333
|
|
|
} |
|
334
|
|
|
} |
|
335
|
|
|
// Array of the most important keywords: |
|
336
|
|
|
$keywords = $api->getOnlyKeyWords(self::cleanupText($text)); |
|
337
|
|
|
|
|
338
|
|
|
// If it's empty, just return the text |
|
339
|
|
|
if (empty($keywords)) { |
|
340
|
|
|
return $text; |
|
341
|
|
|
} |
|
342
|
|
|
|
|
343
|
|
|
$result = implode(', ', array_slice(array_keys($keywords), 0, $limit)); |
|
344
|
|
|
|
|
345
|
|
|
return self::sanitizeUserInput($result); |
|
346
|
|
|
} |
|
347
|
|
|
|
|
348
|
|
|
/** |
|
349
|
|
|
* Extract a summary consisting of the 3 most important sentences from the |
|
350
|
|
|
* text |
|
351
|
|
|
* |
|
352
|
|
|
* @param string $text |
|
353
|
|
|
* @param bool $useStopWords |
|
354
|
|
|
* |
|
355
|
|
|
* @return string |
|
356
|
|
|
*/ |
|
357
|
|
|
public static function extractSummary($text, $useStopWords = true): string |
|
358
|
|
|
{ |
|
359
|
|
|
if (empty($text)) { |
|
360
|
|
|
return ''; |
|
361
|
|
|
} |
|
362
|
|
|
$api = new TextRankFacade(); |
|
363
|
|
|
// Set the stop words that should be ignored |
|
364
|
|
|
if ($useStopWords) { |
|
365
|
|
|
$language = strtolower(substr(Seomatic::$language, 0, 2)); |
|
366
|
|
|
$stopWords = self::stopWordsForLanguage($language); |
|
367
|
|
|
if ($stopWords !== null) { |
|
368
|
|
|
$api->setStopWords($stopWords); |
|
369
|
|
|
} |
|
370
|
|
|
} |
|
371
|
|
|
// Array of the most important keywords: |
|
372
|
|
|
$sentences = $api->getHighlights(self::cleanupText($text)); |
|
373
|
|
|
|
|
374
|
|
|
// If it's empty, just return the text |
|
375
|
|
|
if (empty($sentences)) { |
|
376
|
|
|
return $text; |
|
377
|
|
|
} |
|
378
|
|
|
|
|
379
|
|
|
$result = implode(' ', $sentences); |
|
380
|
|
|
|
|
381
|
|
|
return self::sanitizeUserInput($result); |
|
382
|
|
|
} |
|
383
|
|
|
|
|
384
|
|
|
|
|
385
|
|
|
/** |
|
386
|
|
|
* Sanitize user input by decoding any HTML Entities, URL decoding the text, |
|
387
|
|
|
* then removing any newlines, stripping tags, stripping Twig tags, and changing |
|
388
|
|
|
* single {}'s into ()'s |
|
389
|
|
|
* |
|
390
|
|
|
* @param $str |
|
391
|
|
|
* @return string |
|
392
|
|
|
*/ |
|
393
|
2 |
|
public static function sanitizeUserInput($str): string |
|
394
|
|
|
{ |
|
395
|
|
|
// Do some general cleanup |
|
396
|
2 |
|
$str = html_entity_decode($str, ENT_NOQUOTES, 'UTF-8'); |
|
397
|
2 |
|
$str = rawurldecode($str); |
|
398
|
|
|
// Remove any linebreaks |
|
399
|
2 |
|
$str = (string)preg_replace("/\r|\n/", "", $str); |
|
400
|
2 |
|
$str = HtmlPurifier::process($str, ['HTML.Allowed' => '']); |
|
401
|
2 |
|
$str = html_entity_decode($str, ENT_NOQUOTES, 'UTF-8'); |
|
402
|
|
|
// Remove any embedded Twig code |
|
403
|
2 |
|
$str = preg_replace('/{{.*?}}/', '', $str); |
|
404
|
2 |
|
$str = preg_replace('/{%.*?%}/', '', $str); |
|
405
|
|
|
// Change single brackets to parenthesis |
|
406
|
2 |
|
$str = preg_replace('/{/', '(', $str); |
|
407
|
2 |
|
$str = preg_replace('/}/', ')', $str); |
|
408
|
2 |
|
if (empty($str)) { |
|
409
|
2 |
|
$str = ''; |
|
410
|
|
|
} |
|
411
|
|
|
|
|
412
|
2 |
|
return $str; |
|
413
|
|
|
} |
|
414
|
|
|
|
|
415
|
|
|
/** |
|
416
|
|
|
* Strip HTML tags, but replace them with a space rather than just eliminating them |
|
417
|
|
|
* |
|
418
|
|
|
* @param $str |
|
419
|
|
|
* @return string |
|
420
|
|
|
*/ |
|
421
|
|
|
public static function smartStripTags($str) |
|
422
|
|
|
{ |
|
423
|
|
|
$str = str_replace('<', ' <', $str); |
|
424
|
|
|
$str = HtmlPurifier::process($str, ['HTML.Allowed' => '']); |
|
425
|
|
|
$str = html_entity_decode($str, ENT_NOQUOTES, 'UTF-8'); |
|
426
|
|
|
$str = str_replace(' ', ' ', $str); |
|
427
|
|
|
|
|
428
|
|
|
return $str; |
|
429
|
|
|
} |
|
430
|
|
|
|
|
431
|
|
|
/** |
|
432
|
|
|
* Clean up the passed in text by converting it to UTF-8, stripping tags, |
|
433
|
|
|
* removing whitespace, and decoding HTML entities |
|
434
|
|
|
* |
|
435
|
|
|
* @param string $text |
|
436
|
|
|
* |
|
437
|
|
|
* @return string |
|
438
|
|
|
*/ |
|
439
|
|
|
public static function cleanupText($text): string |
|
440
|
|
|
{ |
|
441
|
|
|
if (empty($text)) { |
|
442
|
|
|
return ''; |
|
443
|
|
|
} |
|
444
|
|
|
// Convert to UTF-8 |
|
445
|
|
|
if (function_exists('iconv')) { |
|
446
|
|
|
$text = iconv(mb_detect_encoding($text, mb_detect_order(), true), 'UTF-8//IGNORE', $text); |
|
|
|
|
|
|
447
|
|
|
} else { |
|
448
|
|
|
ini_set('mbstring.substitute_character', 'none'); |
|
449
|
|
|
$text = mb_convert_encoding($text, 'UTF-8', 'UTF-8'); |
|
450
|
|
|
} |
|
451
|
|
|
// Strip HTML tags |
|
452
|
|
|
$text = HtmlPurifier::process($text, ['HTML.Allowed' => '']); |
|
|
|
|
|
|
453
|
|
|
$text = html_entity_decode($text, ENT_NOQUOTES, 'UTF-8'); |
|
454
|
|
|
// Remove excess whitespace |
|
455
|
|
|
$text = preg_replace('/\s{2,}/u', ' ', $text); |
|
456
|
|
|
// Decode any HTML entities |
|
457
|
|
|
$text = html_entity_decode($text); |
|
458
|
|
|
|
|
459
|
|
|
return $text; |
|
460
|
|
|
} |
|
461
|
|
|
|
|
462
|
|
|
/** |
|
463
|
|
|
* Is $var an array or array-like object? |
|
464
|
|
|
* |
|
465
|
|
|
* @param $var |
|
466
|
|
|
* @return bool |
|
467
|
|
|
*/ |
|
468
|
|
|
public static function isArrayLike($var): bool |
|
469
|
|
|
{ |
|
470
|
|
|
return is_array($var) || ($var instanceof ArrayAccess && $var instanceof Traversable); |
|
471
|
|
|
} |
|
472
|
|
|
|
|
473
|
|
|
// Protected Static Methods |
|
474
|
|
|
// ========================================================================= |
|
475
|
|
|
|
|
476
|
|
|
/** |
|
477
|
|
|
* @param string $language |
|
478
|
|
|
* |
|
479
|
|
|
* @return null|StopWordsAbstract |
|
480
|
|
|
*/ |
|
481
|
|
|
protected static function stopWordsForLanguage(string $language) |
|
482
|
|
|
{ |
|
483
|
|
|
$stopWords = null; |
|
484
|
|
|
if (!empty(self::LANGUAGE_MAP[$language])) { |
|
485
|
|
|
$language = self::LANGUAGE_MAP[$language]; |
|
486
|
|
|
} else { |
|
487
|
|
|
$language = 'English'; |
|
488
|
|
|
} |
|
489
|
|
|
|
|
490
|
|
|
$className = 'PhpScience\\TextRank\\Tool\\StopWords\\' . ucfirst($language); |
|
491
|
|
|
if (class_exists($className)) { |
|
492
|
|
|
$stopWords = new $className(); |
|
493
|
|
|
} |
|
494
|
|
|
|
|
495
|
|
|
return $stopWords; |
|
496
|
|
|
} |
|
497
|
|
|
} |
|
498
|
|
|
|