nystudio107 /
craft-seomatic
| Total Complexity | 75 |
| Total Lines | 441 |
| Duplicated Lines | 0 % |
| Coverage | 11.9% |
| Changes | 2 | ||
| Bugs | 1 | Features | 0 |
Complex classes like Text often do a lot of different things. To break such a class down, we need to identify a cohesive component within that class. A common approach to find such a component is to look for fields/methods that share the same prefixes, or suffixes.
Once you have determined the fields that belong together, you can apply the Extract Class refactoring. If the component makes sense as a sub-class, Extract Subclass is also a candidate, and is often faster.
While breaking up the class, it is a good idea to analyze how other classes use Text, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 41 | class Text |
||
| 42 | { |
||
| 43 | // Constants |
||
| 44 | // ========================================================================= |
||
| 45 | |||
| 46 | const LANGUAGE_MAP = [ |
||
| 47 | 'en' => 'English', |
||
| 48 | 'fr' => 'French', |
||
| 49 | 'de' => 'German', |
||
| 50 | 'it' => 'Italian', |
||
| 51 | 'no' => 'Norwegian', |
||
| 52 | 'es' => 'Spanish', |
||
| 53 | ]; |
||
| 54 | |||
| 55 | // Public Static Methods |
||
| 56 | // ========================================================================= |
||
| 57 | |||
| 58 | /** |
||
| 59 | * Truncates the string to a given length. If $substring is provided, and |
||
| 60 | * truncating occurs, the string is further truncated so that the substring |
||
| 61 | * may be appended without exceeding the desired length. |
||
| 62 | * |
||
| 63 | * @param string $string The string to truncate |
||
| 64 | * @param int $length Desired length of the truncated string |
||
| 65 | * @param string $substring The substring to append if it can fit |
||
| 66 | * |
||
| 67 | * @return string with the resulting $str after truncating |
||
| 68 | */ |
||
| 69 | public static function truncate($string, $length, $substring = '…'): string |
||
| 70 | { |
||
| 71 | $result = $string; |
||
| 72 | |||
| 73 | if (!empty($string)) { |
||
| 74 | $string = HtmlPurifier::process($string, ['HTML.Allowed' => '']); |
||
| 75 | $string = html_entity_decode($string, ENT_NOQUOTES, 'UTF-8'); |
||
| 76 | $result = (string)Stringy::create($string)->truncate($length, $substring); |
||
| 77 | } |
||
| 78 | |||
| 79 | return $result; |
||
| 80 | } |
||
| 81 | |||
| 82 | /** |
||
| 83 | * Truncates the string to a given length, while ensuring that it does not |
||
| 84 | * split words. If $substring is provided, and truncating occurs, the |
||
| 85 | * string is further truncated so that the substring may be appended without |
||
| 86 | * exceeding the desired length. |
||
| 87 | * |
||
| 88 | * @param string $string The string to truncate |
||
| 89 | * @param int $length Desired length of the truncated string |
||
| 90 | * @param string $substring The substring to append if it can fit |
||
| 91 | * |
||
| 92 | * @return string with the resulting $str after truncating |
||
| 93 | */ |
||
| 94 | 1 | public static function truncateOnWord($string, $length, $substring = '…'): string |
|
| 105 | } |
||
| 106 | |||
| 107 | /** |
||
| 108 | * Extract plain old text from a field |
||
| 109 | * |
||
| 110 | * @param $field |
||
| 111 | * |
||
| 112 | * @return string |
||
| 113 | */ |
||
| 114 | public static function extractTextFromField($field): string |
||
| 115 | { |
||
| 116 | if (empty($field)) { |
||
| 117 | return ''; |
||
| 118 | } |
||
| 119 | if ($field instanceof MatrixBlockQuery |
||
| 120 | || (is_array($field) && $field[0] instanceof MatrixBlock)) { |
||
| 121 | $result = self::extractTextFromMatrix($field); |
||
| 122 | } elseif ($field instanceof NeoBlockQuery |
||
| 123 | || (is_array($field) && $field[0] instanceof NeoBlock)) { |
||
| 124 | $result = self::extractTextFromNeo($field); |
||
| 125 | } elseif ($field instanceof SuperTableBlockQuery |
||
| 126 | || (is_array($field) && $field[0] instanceof SuperTableBlock)) { |
||
| 127 | $result = self::extractTextFromSuperTable($field); |
||
| 128 | } elseif ($field instanceof TagQuery |
||
| 129 | || (is_array($field) && $field[0] instanceof Tag)) { |
||
| 130 | $result = self::extractTextFromTags($field); |
||
| 131 | } elseif ($field instanceof DoxterData) { |
||
| 132 | $result = self::smartStripTags(Doxter::$plugin->getService()->parseMarkdown($field->getRaw())); |
||
| 133 | } else { |
||
| 134 | if (is_array($field)) { |
||
| 135 | $result = self::smartStripTags((string)$field[0]); |
||
| 136 | } else { |
||
| 137 | $result = self::smartStripTags((string)$field); |
||
| 138 | } |
||
| 139 | } |
||
| 140 | |||
| 141 | //return $result; |
||
| 142 | return self::sanitizeUserInput($result); |
||
| 143 | } |
||
| 144 | |||
| 145 | /** |
||
| 146 | * Extract concatenated text from all of the tags in the $tagElement and |
||
| 147 | * return as a comma-delimited string |
||
| 148 | * |
||
| 149 | * @param TagQuery|Tag[]|array $tags |
||
| 150 | * |
||
| 151 | * @return string |
||
| 152 | */ |
||
| 153 | public static function extractTextFromTags($tags): string |
||
| 169 | } |
||
| 170 | |||
| 171 | /** |
||
| 172 | * Extract text from all of the blocks in a matrix field, concatenating it |
||
| 173 | * together. |
||
| 174 | * |
||
| 175 | * @param MatrixBlockQuery|MatrixBlock[]|array $blocks |
||
| 176 | * @param string $fieldHandle |
||
| 177 | * |
||
| 178 | * @return string |
||
| 179 | */ |
||
| 180 | public static function extractTextFromMatrix($blocks, $fieldHandle = ''): string |
||
| 181 | { |
||
| 182 | if (empty($blocks)) { |
||
| 183 | return ''; |
||
| 184 | } |
||
| 185 | $result = ''; |
||
| 186 | // Iterate through all of the matrix blocks |
||
| 187 | if ($blocks instanceof MatrixBlockQuery) { |
||
| 188 | $blocks = $blocks->all(); |
||
| 189 | } |
||
| 190 | foreach ($blocks as $block) { |
||
| 191 | try { |
||
| 192 | $matrixBlockTypeModel = $block->getType(); |
||
| 193 | } catch (InvalidConfigException $e) { |
||
| 194 | $matrixBlockTypeModel = null; |
||
| 195 | } |
||
| 196 | // Find any text fields inside of the matrix block |
||
| 197 | if ($matrixBlockTypeModel) { |
||
| 198 | $fieldClasses = FieldHelper::FIELD_CLASSES[FieldHelper::TEXT_FIELD_CLASS_KEY]; |
||
| 199 | $fields = $matrixBlockTypeModel->getFields(); |
||
| 200 | |||
| 201 | foreach ($fields as $field) { |
||
| 202 | /** @var array $fieldClasses */ |
||
| 203 | foreach ($fieldClasses as $fieldClassKey) { |
||
| 204 | if ($field instanceof $fieldClassKey) { |
||
| 205 | if ($field->handle === $fieldHandle || empty($fieldHandle)) { |
||
| 206 | $result .= self::extractTextFromField($block[$field->handle]) . ' '; |
||
| 207 | } |
||
| 208 | } |
||
| 209 | } |
||
| 210 | } |
||
| 211 | } |
||
| 212 | } |
||
| 213 | |||
| 214 | return $result; |
||
| 215 | } |
||
| 216 | |||
| 217 | /** |
||
| 218 | * Extract text from all of the blocks in a Neo field, concatenating it |
||
| 219 | * together. |
||
| 220 | * |
||
| 221 | * @param NeoBlockQuery|NeoBlock[]|array $blocks |
||
| 222 | * @param string $fieldHandle |
||
| 223 | * |
||
| 224 | * @return string |
||
| 225 | */ |
||
| 226 | public static function extractTextFromNeo($blocks, $fieldHandle = ''): string |
||
| 227 | { |
||
| 228 | if (empty($blocks)) { |
||
| 229 | return ''; |
||
| 230 | } |
||
| 231 | $result = ''; |
||
| 232 | // Iterate through all of the matrix blocks |
||
| 233 | if ($blocks instanceof NeoBlockQuery) { |
||
| 234 | $blocks = $blocks->all(); |
||
| 235 | } |
||
| 236 | foreach ($blocks as $block) { |
||
| 237 | $layout = $block->getFieldLayout(); |
||
| 238 | // Find any text fields inside of the matrix block |
||
| 239 | if ($layout) { |
||
| 240 | $fieldClasses = FieldHelper::FIELD_CLASSES[FieldHelper::TEXT_FIELD_CLASS_KEY]; |
||
| 241 | $fields = $layout->getFields(); |
||
| 242 | |||
| 243 | foreach ($fields as $field) { |
||
| 244 | /** @var array $fieldClasses */ |
||
| 245 | foreach ($fieldClasses as $fieldClassKey) { |
||
| 246 | if ($field instanceof $fieldClassKey) { |
||
| 247 | if ($field->handle === $fieldHandle || empty($fieldHandle)) { |
||
| 248 | $result .= self::extractTextFromField($block[$field->handle]) . ' '; |
||
| 249 | } |
||
| 250 | } |
||
| 251 | } |
||
| 252 | } |
||
| 253 | } |
||
| 254 | } |
||
| 255 | |||
| 256 | return $result; |
||
| 257 | } |
||
| 258 | |||
| 259 | /** |
||
| 260 | * Extract text from all of the blocks in a matrix field, concatenating it |
||
| 261 | * together. |
||
| 262 | * |
||
| 263 | * @param SuperTableBlockQuery|SuperTableBlock[]|array $blocks |
||
| 264 | * @param string $fieldHandle |
||
| 265 | * |
||
| 266 | * @return string |
||
| 267 | */ |
||
| 268 | public static function extractTextFromSuperTable($blocks, $fieldHandle = ''): string |
||
| 269 | { |
||
| 270 | if (empty($blocks)) { |
||
| 271 | return ''; |
||
| 272 | } |
||
| 273 | $result = ''; |
||
| 274 | // Iterate through all of the matrix blocks |
||
| 275 | if ($blocks instanceof SuperTableBlockQuery) { |
||
| 276 | $blocks = $blocks->all(); |
||
| 277 | } |
||
| 278 | foreach ($blocks as $block) { |
||
| 279 | try { |
||
| 280 | /** @var SuperTableBlockTypeModel $superTableBlockTypeModel */ |
||
| 281 | $superTableBlockTypeModel = $block->getType(); |
||
| 282 | } catch (InvalidConfigException $e) { |
||
| 283 | $superTableBlockTypeModel = null; |
||
| 284 | } |
||
| 285 | // Find any text fields inside of the matrix block |
||
| 286 | if ($superTableBlockTypeModel) { |
||
| 287 | $fieldClasses = FieldHelper::FIELD_CLASSES[FieldHelper::TEXT_FIELD_CLASS_KEY]; |
||
| 288 | // The SuperTableBlockTypeModel class lacks @mixin FieldLayoutBehavior in its annotations |
||
| 289 | /** @phpstan-ignore-next-line */ |
||
| 290 | $fields = $superTableBlockTypeModel->getFields(); |
||
| 291 | |||
| 292 | foreach ($fields as $field) { |
||
| 293 | /** @var array $fieldClasses */ |
||
| 294 | foreach ($fieldClasses as $fieldClassKey) { |
||
| 295 | if ($field instanceof $fieldClassKey) { |
||
| 296 | if ($field->handle === $fieldHandle || empty($fieldHandle)) { |
||
| 297 | $result .= self::extractTextFromField($block[$field->handle]) . ' '; |
||
| 298 | } |
||
| 299 | } |
||
| 300 | } |
||
| 301 | } |
||
| 302 | } |
||
| 303 | } |
||
| 304 | |||
| 305 | return $result; |
||
| 306 | } |
||
| 307 | |||
| 308 | /** |
||
| 309 | * Return the most important keywords extracted from the text as a comma- |
||
| 310 | * delimited string |
||
| 311 | * |
||
| 312 | * @param string $text |
||
| 313 | * @param int $limit |
||
| 314 | * @param bool $useStopWords |
||
| 315 | * |
||
| 316 | * @return string |
||
| 317 | */ |
||
| 318 | public static function extractKeywords($text, $limit = 15, $useStopWords = true): string |
||
| 319 | { |
||
| 320 | if (empty($text)) { |
||
| 321 | return ''; |
||
| 322 | } |
||
| 323 | $api = new TextRankFacade(); |
||
| 324 | // Set the stop words that should be ignored |
||
| 325 | if ($useStopWords) { |
||
| 326 | $language = strtolower(substr(Seomatic::$language, 0, 2)); |
||
| 327 | $stopWords = self::stopWordsForLanguage($language); |
||
| 328 | if ($stopWords !== null) { |
||
| 329 | $api->setStopWords($stopWords); |
||
| 330 | } |
||
| 331 | } |
||
| 332 | // Array of the most important keywords: |
||
| 333 | $keywords = $api->getOnlyKeyWords(self::cleanupText($text)); |
||
| 334 | |||
| 335 | // If it's empty, just return the text |
||
| 336 | if (empty($keywords)) { |
||
| 337 | return $text; |
||
| 338 | } |
||
| 339 | |||
| 340 | $result = implode(', ', array_slice(array_keys($keywords), 0, $limit)); |
||
| 341 | |||
| 342 | return self::sanitizeUserInput($result); |
||
| 343 | } |
||
| 344 | |||
| 345 | /** |
||
| 346 | * Extract a summary consisting of the 3 most important sentences from the |
||
| 347 | * text |
||
| 348 | * |
||
| 349 | * @param string $text |
||
| 350 | * @param bool $useStopWords |
||
| 351 | * |
||
| 352 | * @return string |
||
| 353 | */ |
||
| 354 | public static function extractSummary($text, $useStopWords = true): string |
||
| 355 | { |
||
| 356 | if (empty($text)) { |
||
| 357 | return ''; |
||
| 358 | } |
||
| 359 | $api = new TextRankFacade(); |
||
| 360 | // Set the stop words that should be ignored |
||
| 361 | if ($useStopWords) { |
||
| 362 | $language = strtolower(substr(Seomatic::$language, 0, 2)); |
||
| 363 | $stopWords = self::stopWordsForLanguage($language); |
||
| 364 | if ($stopWords !== null) { |
||
| 365 | $api->setStopWords($stopWords); |
||
| 366 | } |
||
| 367 | } |
||
| 368 | // Array of the most important keywords: |
||
| 369 | $sentences = $api->getHighlights(self::cleanupText($text)); |
||
| 370 | |||
| 371 | // If it's empty, just return the text |
||
| 372 | if (empty($sentences)) { |
||
| 373 | return $text; |
||
| 374 | } |
||
| 375 | |||
| 376 | $result = implode(' ', $sentences); |
||
| 377 | |||
| 378 | return self::sanitizeUserInput($result); |
||
| 379 | } |
||
| 380 | |||
| 381 | |||
| 382 | /** |
||
| 383 | * Sanitize user input by decoding any HTML Entities, URL decoding the text, |
||
| 384 | * then removing any newlines, stripping tags, stripping Twig tags, and changing |
||
| 385 | * single {}'s into ()'s |
||
| 386 | * |
||
| 387 | * @param $str |
||
| 388 | * @return string |
||
| 389 | */ |
||
| 390 | 2 | public static function sanitizeUserInput($str): string |
|
| 391 | { |
||
| 392 | // Do some general cleanup |
||
| 393 | 2 | $str = html_entity_decode($str, ENT_NOQUOTES, 'UTF-8'); |
|
| 394 | 2 | $str = rawurldecode($str); |
|
| 395 | // Remove any linebreaks |
||
| 396 | 2 | $str = (string)preg_replace("/\r|\n/", "", $str); |
|
| 397 | 2 | $str = HtmlPurifier::process($str, ['HTML.Allowed' => '']); |
|
| 398 | 2 | $str = html_entity_decode($str, ENT_NOQUOTES, 'UTF-8'); |
|
| 399 | // Remove any embedded Twig code |
||
| 400 | 2 | $str = preg_replace('/{{.*?}}/', '', $str); |
|
| 401 | 2 | $str = preg_replace('/{%.*?%}/', '', $str); |
|
| 402 | // Change single brackets to parenthesis |
||
| 403 | 2 | $str = preg_replace('/{/', '(', $str); |
|
| 404 | 2 | $str = preg_replace('/}/', ')', $str); |
|
| 405 | 2 | if (empty($str)) { |
|
| 406 | 2 | $str = ''; |
|
| 407 | } |
||
| 408 | |||
| 409 | 2 | return $str; |
|
| 410 | } |
||
| 411 | |||
| 412 | /** |
||
| 413 | * Strip HTML tags, but replace them with a space rather than just eliminating them |
||
| 414 | * |
||
| 415 | * @param $str |
||
| 416 | * @return string |
||
| 417 | */ |
||
| 418 | public static function smartStripTags($str) |
||
| 426 | } |
||
| 427 | |||
| 428 | /** |
||
| 429 | * Clean up the passed in text by converting it to UTF-8, stripping tags, |
||
| 430 | * removing whitespace, and decoding HTML entities |
||
| 431 | * |
||
| 432 | * @param string $text |
||
| 433 | * |
||
| 434 | * @return string |
||
| 435 | */ |
||
| 436 | public static function cleanupText($text): string |
||
| 437 | { |
||
| 438 | if (empty($text)) { |
||
| 439 | return ''; |
||
| 440 | } |
||
| 441 | // Convert to UTF-8 |
||
| 442 | if (function_exists('iconv')) { |
||
| 443 | $text = iconv(mb_detect_encoding($text, mb_detect_order(), true), 'UTF-8//IGNORE', $text); |
||
|
Bug
introduced
by
|
|||
| 444 | } else { |
||
| 445 | ini_set('mbstring.substitute_character', 'none'); |
||
| 446 | $text = mb_convert_encoding($text, 'UTF-8', 'UTF-8'); |
||
| 447 | } |
||
| 448 | // Strip HTML tags |
||
| 449 | $text = HtmlPurifier::process($text, ['HTML.Allowed' => '']); |
||
| 450 | $text = html_entity_decode($text, ENT_NOQUOTES, 'UTF-8'); |
||
| 451 | // Remove excess whitespace |
||
| 452 | $text = preg_replace('/\s{2,}/u', ' ', $text); |
||
| 453 | // Decode any HTML entities |
||
| 454 | $text = html_entity_decode($text); |
||
| 455 | |||
| 456 | return $text; |
||
| 457 | } |
||
| 458 | |||
| 459 | // Protected Static Methods |
||
| 460 | // ========================================================================= |
||
| 461 | |||
| 462 | /** |
||
| 463 | * @param string $language |
||
| 464 | * |
||
| 465 | * @return null|StopWordsAbstract |
||
| 466 | */ |
||
| 467 | protected static function stopWordsForLanguage(string $language) |
||
| 482 | } |
||
| 483 | } |
||
| 484 |
Loading history...
