| Total Complexity | 55 |
| Total Lines | 508 |
| Duplicated Lines | 0 % |
| Changes | 5 | ||
| Bugs | 0 | Features | 0 |
Complex classes like ConvertHelper_URLFinder 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 ConvertHelper_URLFinder, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 25 | class ConvertHelper_URLFinder implements Interface_Optionable |
||
| 26 | { |
||
| 27 | public const ERROR_INVALID_DETECTOR_CLASS = 87901; |
||
| 28 | |||
| 29 | use Traits_Optionable; |
||
| 30 | |||
| 31 | /** |
||
| 32 | * @var array<string,URLInfo> |
||
| 33 | */ |
||
| 34 | private $urls = array(); |
||
| 35 | |||
| 36 | /** |
||
| 37 | * @var array<string,URLInfo> |
||
| 38 | */ |
||
| 39 | private $emails = array(); |
||
| 40 | |||
| 41 | /** |
||
| 42 | * @var array<string,bool> |
||
| 43 | */ |
||
| 44 | private $enabledDetectorClasses = array( |
||
| 45 | ConvertHelper_URLFinder_Detector_Tel::class => true, |
||
| 46 | ConvertHelper_URLFinder_Detector_HTMLAttributes::class => false, |
||
| 47 | ConvertHelper_URLFinder_Detector_IPV4::class => true |
||
| 48 | ); |
||
| 49 | |||
| 50 | /** |
||
| 51 | * @var string[] |
||
| 52 | */ |
||
| 53 | private $matches = array(); |
||
| 54 | |||
| 55 | /** |
||
| 56 | * @var string[] |
||
| 57 | */ |
||
| 58 | private $boundaries = array( |
||
| 59 | "\n", |
||
| 60 | "\t", |
||
| 61 | "\r", |
||
| 62 | '"', |
||
| 63 | "'", |
||
| 64 | '|', |
||
| 65 | ',', |
||
| 66 | ';', |
||
| 67 | '<', |
||
| 68 | '>' |
||
| 69 | ); |
||
| 70 | |||
| 71 | /** |
||
| 72 | * @var ConvertHelper_URLFinder_Detector[] |
||
| 73 | */ |
||
| 74 | private $detectors = array(); |
||
| 75 | |||
| 76 | /** |
||
| 77 | * @var string |
||
| 78 | */ |
||
| 79 | private $subject; |
||
| 80 | |||
| 81 | /** |
||
| 82 | * @var bool |
||
| 83 | */ |
||
| 84 | private $parsed = false; |
||
| 85 | |||
| 86 | public function __construct(string $subject) |
||
| 87 | { |
||
| 88 | $this->subject = $subject; |
||
| 89 | } |
||
| 90 | |||
| 91 | public function getDefaultOptions() : array |
||
| 92 | { |
||
| 93 | return array( |
||
| 94 | 'include-emails' => false, |
||
| 95 | 'omit-mailto' => false, |
||
| 96 | 'sorting' => false, |
||
| 97 | 'normalize' => false |
||
| 98 | ); |
||
| 99 | } |
||
| 100 | |||
| 101 | /** |
||
| 102 | * Whether all URLs should be normalized (parameters ordered alphabetically, |
||
| 103 | * whitespace removal). This ensures that URL duplicates are detected even |
||
| 104 | * if they have a different order of parameters. |
||
| 105 | * |
||
| 106 | * @param bool $enabled |
||
| 107 | * @return $this |
||
| 108 | */ |
||
| 109 | public function enableNormalizing(bool $enabled=true) : ConvertHelper_URLFinder |
||
| 110 | { |
||
| 111 | $this->setOption('normalize', $enabled); |
||
| 112 | return $this; |
||
| 113 | } |
||
| 114 | |||
| 115 | /** |
||
| 116 | * Whether to enable sorting the URLs alphabetically (disabled by default). |
||
| 117 | * |
||
| 118 | * @param bool $enabled |
||
| 119 | * @return $this |
||
| 120 | */ |
||
| 121 | public function enableSorting(bool $enabled=true) : ConvertHelper_URLFinder |
||
| 122 | { |
||
| 123 | $this->setOption('sorting', $enabled); |
||
| 124 | return $this; |
||
| 125 | } |
||
| 126 | |||
| 127 | /** |
||
| 128 | * Whether to include email addresses in the search. |
||
| 129 | * This is only relevant when using the getURLs() |
||
| 130 | * method. |
||
| 131 | * |
||
| 132 | * @param bool $include |
||
| 133 | * @return ConvertHelper_URLFinder |
||
| 134 | */ |
||
| 135 | public function includeEmails(bool $include=true) : ConvertHelper_URLFinder |
||
| 136 | { |
||
| 137 | $this->setOption('include-emails', $include); |
||
| 138 | return $this; |
||
| 139 | } |
||
| 140 | |||
| 141 | /** |
||
| 142 | * Whether to omit the mailto: that is automatically added to all email addresses. |
||
| 143 | * |
||
| 144 | * @param bool $omit |
||
| 145 | * @return ConvertHelper_URLFinder |
||
| 146 | */ |
||
| 147 | public function omitMailto(bool $omit=true) : ConvertHelper_URLFinder |
||
| 151 | } |
||
| 152 | |||
| 153 | /** |
||
| 154 | * Splits the string by a list of word boundaries, so that all relevant |
||
| 155 | * words are separated into individual lines. Each line is then checked |
||
| 156 | * to keep only strings that are more or less likely to contain a domain name. |
||
| 157 | * |
||
| 158 | * @param string $subject |
||
| 159 | * @return string[] |
||
| 160 | */ |
||
| 161 | private function splitSubject(string $subject) : array |
||
| 162 | { |
||
| 163 | $subject = str_replace($this->boundaries, ' ', $subject); |
||
| 164 | $lines = ConvertHelper::explodeTrim(' ', $subject); |
||
| 165 | |||
| 166 | unset($subject); |
||
| 167 | |||
| 168 | $keep = array(); |
||
| 169 | |||
| 170 | foreach ($lines as $line) |
||
| 171 | { |
||
| 172 | $line = $this->analyzeLine($line); |
||
| 173 | |||
| 174 | if($line !== null) { |
||
| 175 | $keep[] = $line; |
||
| 176 | } |
||
| 177 | } |
||
| 178 | |||
| 179 | return array_unique($keep); |
||
| 180 | } |
||
| 181 | |||
| 182 | /** |
||
| 183 | * Analyzes a single line to see if it is likely to contain a domain name. |
||
| 184 | * |
||
| 185 | * @param string $line |
||
| 186 | * @return string|null |
||
| 187 | */ |
||
| 188 | private function analyzeLine(string $line) : ?string |
||
| 189 | { |
||
| 190 | // Strip punctuation from the beginning and end, |
||
| 191 | // to exclude the end of phrases, e.g. "domain.com." |
||
| 192 | $line = trim($line, '.'); |
||
| 193 | |||
| 194 | // Handle detecting an URI scheme |
||
| 195 | if(strstr($line, ':') !== false) |
||
| 196 | { |
||
| 197 | $scheme = URISchemes::detectScheme($line); |
||
| 198 | |||
| 199 | if ($scheme !== null) |
||
| 200 | { |
||
| 201 | return $line; |
||
| 202 | } |
||
| 203 | } |
||
| 204 | |||
| 205 | // From here on out, the only things we can still |
||
| 206 | // detect are email addresses and domain names. |
||
| 207 | |||
| 208 | // No dot? Then it's certainly not a domain name. |
||
| 209 | if(strstr($line, '.') === false) { |
||
| 210 | return null; |
||
| 211 | } |
||
| 212 | |||
| 213 | return $line; |
||
| 214 | } |
||
| 215 | |||
| 216 | /** |
||
| 217 | * Filters the subject string before trying to detect regular HTTP/HTTPS |
||
| 218 | * URLs as well as email addresses that are domain-based. |
||
| 219 | * |
||
| 220 | * @param string $subject |
||
| 221 | * @return string |
||
| 222 | */ |
||
| 223 | private function filterSubjectBefore(string $subject) : string |
||
| 224 | { |
||
| 225 | $subject = stripslashes($subject); |
||
| 226 | |||
| 227 | foreach($this->detectors as $detector) |
||
| 228 | { |
||
| 229 | // Avoid processing the string if it is not needed. |
||
| 230 | if($detector->getRunPosition() !== ConvertHelper_URLFinder_Detector::RUN_BEFORE || !$detector->isValidFor($subject)) { |
||
| 231 | continue; |
||
| 232 | } |
||
| 233 | |||
| 234 | $subject = $detector->processString($subject); |
||
| 235 | |||
| 236 | $this->matches = array_merge($this->matches, $detector->getMatches()); |
||
| 237 | } |
||
| 238 | |||
| 239 | return $subject; |
||
| 240 | } |
||
| 241 | |||
| 242 | /** |
||
| 243 | * @param string $className |
||
| 244 | * @return ConvertHelper_URLFinder_Detector |
||
| 245 | * @throws ConvertHelper_Exception |
||
| 246 | */ |
||
| 247 | private function createDetector(string $className) : ConvertHelper_URLFinder_Detector |
||
| 248 | { |
||
| 249 | $detector = new $className(); |
||
| 250 | |||
| 251 | if($detector instanceof ConvertHelper_URLFinder_Detector) |
||
| 252 | { |
||
| 253 | return $detector; |
||
| 254 | } |
||
| 255 | |||
| 256 | throw new ConvertHelper_Exception( |
||
| 257 | 'Not a valid detector class.', |
||
| 258 | sprintf( |
||
| 259 | 'The class [%s] is not an instance of [%s].', |
||
| 260 | $className, |
||
| 261 | ConvertHelper_URLFinder_Detector::class |
||
| 262 | ), |
||
| 263 | self::ERROR_INVALID_DETECTOR_CLASS |
||
| 264 | ); |
||
| 265 | } |
||
| 266 | |||
| 267 | /** |
||
| 268 | * Fetches all URLs that can be found in the subject string. |
||
| 269 | * |
||
| 270 | * @return string[] |
||
| 271 | */ |
||
| 272 | public function getURLs() : array |
||
| 273 | { |
||
| 274 | $this->parse(); |
||
| 275 | |||
| 276 | $result = $this->getItemsAsString($this->urls); |
||
| 277 | |||
| 278 | if($this->getBoolOption('include-emails')) |
||
| 279 | { |
||
| 280 | $result = array_merge($result, $this->getEmails()); |
||
| 281 | } |
||
| 282 | |||
| 283 | if($this->getBoolOption('sorting')) |
||
| 284 | { |
||
| 285 | usort($result, function(string $a, string $b) { |
||
| 286 | return strnatcasecmp($a, $b); |
||
| 287 | }); |
||
| 288 | } |
||
| 289 | |||
| 290 | return $result; |
||
| 291 | } |
||
| 292 | |||
| 293 | /** |
||
| 294 | * @param array<string,URLInfo> $collection |
||
| 295 | * @return string[] |
||
| 296 | */ |
||
| 297 | private function getItemsAsString(array $collection) : array |
||
| 298 | { |
||
| 299 | $normalize = $this->getBoolOption('normalize'); |
||
| 300 | |||
| 301 | $result = array(); |
||
| 302 | |||
| 303 | foreach($collection as $url => $info) { |
||
| 304 | if($normalize) { |
||
| 305 | $url = $info->getNormalized(); |
||
| 306 | } |
||
| 307 | |||
| 308 | if(!in_array($url, $result)) { |
||
| 309 | $result[] = $url; |
||
| 310 | } |
||
| 311 | } |
||
| 312 | |||
| 313 | return $result; |
||
| 314 | } |
||
| 315 | |||
| 316 | /** |
||
| 317 | * Instantiates the selected detector classes, which are |
||
| 318 | * used to detect specific elements in the target string |
||
| 319 | * (beyond regular URLs and Email addresses). |
||
| 320 | * |
||
| 321 | * @throws ConvertHelper_Exception |
||
| 322 | */ |
||
| 323 | private function initDetectors() : void |
||
| 324 | { |
||
| 325 | foreach($this->enabledDetectorClasses as $className => $enabled) |
||
| 326 | { |
||
| 327 | if($enabled) { |
||
| 328 | $this->detectors[] = $this->createDetector($className); |
||
| 329 | } |
||
| 330 | } |
||
| 331 | } |
||
| 332 | |||
| 333 | /** |
||
| 334 | * Parses the specified string to detect all URLs and Email addresses. |
||
| 335 | * For accurate results, this does not use a regex, but splits the |
||
| 336 | * string into a list of strings that are likely to be either an URL |
||
| 337 | * or Email address. Each of these is then checked for a valid scheme |
||
| 338 | * or domain name extension. |
||
| 339 | */ |
||
| 340 | private function parse() : void |
||
| 341 | { |
||
| 342 | if($this->parsed) { |
||
| 343 | return; |
||
| 344 | } |
||
| 345 | |||
| 346 | $this->parsed = true; |
||
| 347 | |||
| 348 | $this->initDetectors(); |
||
| 349 | $this->detectMatches($this->subject); |
||
| 350 | |||
| 351 | unset($this->subject); |
||
| 352 | |||
| 353 | foreach($this->matches as $match) |
||
| 354 | { |
||
| 355 | $info = parseURL($match); |
||
| 356 | |||
| 357 | if($info->isEmail()) |
||
| 358 | { |
||
| 359 | $this->emails[$this->filterEmailAddress($match)] = $info; |
||
| 360 | continue; |
||
| 361 | } |
||
| 362 | |||
| 363 | $this->urls[$match] = $info; |
||
| 364 | } |
||
| 365 | } |
||
| 366 | |||
| 367 | /** |
||
| 368 | * Enables the search for relative URLs in HTML attributes. |
||
| 369 | * |
||
| 370 | * @param bool $enable |
||
| 371 | * @return $this |
||
| 372 | */ |
||
| 373 | public function enableHTMLAttributes(bool $enable=true) : ConvertHelper_URLFinder |
||
| 374 | { |
||
| 375 | $this->enabledDetectorClasses[ConvertHelper_URLFinder_Detector_HTMLAttributes::class] = $enable; |
||
| 376 | return $this; |
||
| 377 | } |
||
| 378 | |||
| 379 | /** |
||
| 380 | * Ensures that the email address has the `mailto:` scheme prepended, |
||
| 381 | * and lowercases it to avoid case mixups. |
||
| 382 | * |
||
| 383 | * @param string $email |
||
| 384 | * @return string |
||
| 385 | */ |
||
| 386 | private function filterEmailAddress(string $email) : string |
||
| 387 | { |
||
| 388 | if(stristr($email, 'mailto:') === false) { |
||
| 389 | $email = 'mailto:'.$email; |
||
| 390 | } |
||
| 391 | |||
| 392 | return strtolower($email); |
||
| 393 | } |
||
| 394 | |||
| 395 | /** |
||
| 396 | * Detects all URL and Email matches in the specified string. |
||
| 397 | * |
||
| 398 | * @param string $subject |
||
| 399 | */ |
||
| 400 | private function detectMatches(string $subject) : void |
||
| 401 | { |
||
| 402 | $subject = $this->filterSubjectBefore($subject); |
||
| 403 | |||
| 404 | $lines = $this->splitSubject($subject); |
||
| 405 | $domains = new ConvertHelper_URLFinder_DomainExtensions(); |
||
| 406 | |||
| 407 | foreach ($lines as $line) |
||
| 408 | { |
||
| 409 | $scheme = URISchemes::detectScheme($line); |
||
| 410 | if($scheme !== null) { |
||
| 411 | $this->matches[] = $line; |
||
| 412 | continue; |
||
| 413 | } |
||
| 414 | |||
| 415 | $extension = $this->detectDomainExtension($line); |
||
| 416 | |||
| 417 | if($domains->nameExists($extension)) { |
||
| 418 | $this->matches[] = $line; |
||
| 419 | } |
||
| 420 | } |
||
| 421 | |||
| 422 | $this->filterSubjectAfter($subject); |
||
| 423 | } |
||
| 424 | |||
| 425 | private function filterSubjectAfter(string $subject) : void |
||
| 426 | { |
||
| 427 | // Sort the matches from longest to shortest, to avoid |
||
| 428 | // replacing parts of URLs. |
||
| 429 | $remove = $this->matches; |
||
| 430 | usort($remove, function(string $a, string $b) { |
||
| 431 | return strlen($b) - strlen($a); |
||
| 432 | }); |
||
| 433 | |||
| 434 | $subject = str_replace($remove, ' ', $subject); |
||
| 435 | |||
| 436 | foreach($this->detectors as $detector) |
||
| 437 | { |
||
| 438 | if($detector->getRunPosition() !== ConvertHelper_URLFinder_Detector::RUN_AFTER || !$detector->isValidFor($subject)) { |
||
| 439 | continue; |
||
| 440 | } |
||
| 441 | |||
| 442 | $subject = $detector->processString($subject); |
||
| 443 | |||
| 444 | $this->matches = array_merge($this->matches, $detector->getMatches()); |
||
| 445 | } |
||
| 446 | } |
||
| 447 | |||
| 448 | /** |
||
| 449 | * Attempts to extract a valid domain name extension from |
||
| 450 | * the specified URL. |
||
| 451 | * |
||
| 452 | * @param string $url |
||
| 453 | * @return string |
||
| 454 | * @see ConvertHelper_URLFinder_DomainExtensions |
||
| 455 | */ |
||
| 456 | private function detectDomainExtension(string $url) : string |
||
| 457 | { |
||
| 458 | $boundaries = array('/', '?'); |
||
| 459 | |||
| 460 | // Remove the path or query parts to access the domain extension only |
||
| 461 | foreach($boundaries as $boundary) { |
||
| 462 | if(strstr($url, $boundary)) { |
||
| 463 | $parts = explode($boundary, $url); |
||
| 464 | $url = array_shift($parts); |
||
| 465 | break; |
||
| 466 | } |
||
| 467 | } |
||
| 468 | |||
| 469 | $parts = explode('.', $url); |
||
| 470 | |||
| 471 | return array_pop($parts); |
||
| 472 | } |
||
| 473 | |||
| 474 | /** |
||
| 475 | * Retrieves all email addresses from the subject string. |
||
| 476 | * |
||
| 477 | * @return string[] |
||
| 478 | * |
||
| 479 | * @see omitMailto() |
||
| 480 | */ |
||
| 481 | public function getEmails() : array |
||
| 504 | } |
||
| 505 | |||
| 506 | /** |
||
| 507 | * Retrieves all URLs as URLInfo instances. |
||
| 508 | * |
||
| 509 | * @return URLInfo[] |
||
| 510 | */ |
||
| 511 | public function getInfos() : array |
||
| 533 | } |
||
| 534 | } |
||
| 535 |