| Total Complexity | 44 |
| Total Lines | 356 |
| Duplicated Lines | 0 % |
| Changes | 0 | ||
Complex classes like Seeker 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 Seeker, and based on these observations, apply Extract Interface, too.
| 1 | <?php |
||
| 27 | final class Seeker extends Communicator implements SeekerInterface |
||
| 28 | { |
||
| 29 | use Timed; |
||
| 30 | |||
| 31 | private const INITIAL_PAGE = 1; |
||
| 32 | |||
| 33 | protected OptionSet $optionSet; |
||
| 34 | protected GoutteClient $client; |
||
| 35 | |||
| 36 | private array $targetDefinitions; |
||
| 37 | private LoggerInterface $logger; |
||
| 38 | private NodeProximityAssistant $nodeProximityAssistant; |
||
| 39 | private int $pageLimit; |
||
| 40 | private Scrapper $scrapper; |
||
| 41 | |||
| 42 | /** |
||
| 43 | * Create a new seeker. |
||
| 44 | * |
||
| 45 | * @throws Exception |
||
| 46 | */ |
||
| 47 | public function __construct( |
||
| 66 | ); |
||
| 67 | } |
||
| 68 | |||
| 69 | /** |
||
| 70 | * {@inheritdoc} |
||
| 71 | */ |
||
| 72 | public function seek( |
||
| 142 | } |
||
| 143 | |||
| 144 | private function crawlTarget(Target $target): void |
||
| 159 | } |
||
| 160 | |||
| 161 | private function searchAndScrape(Target $target, Crawler $crawler): void |
||
| 162 | { |
||
| 163 | if ($this->verbosity >= self::VERBOSITY_MEDIUM) { |
||
| 164 | $this->logger->info('Landing Document', [$crawler->html()]); |
||
| 165 | } |
||
| 166 | |||
| 167 | $searchFormConfig = $target->getSearch()[TargetKey::SEARCH_FORM]; |
||
| 168 | $searchFormSubmitButtonConfig = $searchFormConfig[TargetKey::SEARCH_FORM_SUBMIT_BUTTON] ?? []; |
||
| 169 | $formCrawler = $crawler->filter($searchFormConfig[TargetKey::SEARCH_FORM_SELECTOR]); |
||
| 170 | $submitButtonIdentifier = null; |
||
| 171 | $form = null; |
||
|
|
|||
| 172 | |||
| 173 | // check if submit button was defined |
||
| 174 | if (!empty($searchFormSubmitButtonConfig[TargetKey::SEARCH_FORM_SUBMIT_BUTTON_ID])) { |
||
| 175 | $submitButtonIdentifier = $searchFormSubmitButtonConfig[TargetKey::SEARCH_FORM_SUBMIT_BUTTON_ID]; |
||
| 176 | } elseif (!empty($searchFormSubmitButtonConfig[TargetKey::SEARCH_FORM_SUBMIT_BUTTON_TEXT])) { |
||
| 177 | $submitButtonIdentifier = $searchFormSubmitButtonConfig[TargetKey::SEARCH_FORM_SUBMIT_BUTTON_TEXT]; |
||
| 178 | } |
||
| 179 | |||
| 180 | try { |
||
| 181 | $form = empty($submitButtonIdentifier) |
||
| 182 | ? $formCrawler->form() |
||
| 183 | : $formCrawler->selectButton((string)$submitButtonIdentifier)->form(); |
||
| 184 | } catch (InvalidArgumentException $e) { |
||
| 185 | $this->logger->error($e); |
||
| 186 | } |
||
| 187 | |||
| 188 | if ($form instanceof Form) { |
||
| 189 | if ($this->verbosity >= self::VERBOSITY_MEDIUM) { |
||
| 190 | $this->logger->info('FORM', [$form]); |
||
| 191 | } |
||
| 192 | |||
| 193 | // Search each phrase/keyword one at a time. |
||
| 194 | foreach ($target->getSearch()[TargetKey::SEARCH_KEYWORDS] as $keyword) { |
||
| 195 | // set keyword |
||
| 196 | $form[$searchFormConfig[TargetKey::SEARCH_FORM_INPUT]] = $keyword; |
||
| 197 | |||
| 198 | $this->tell(FormattedMessage::get(FormattedMessage::SEARCH_KEYWORD, $keyword), self::COMM_DIRECTION_IN); |
||
| 199 | // submit search form |
||
| 200 | $resultCrawler = $this->client->submit($form); |
||
| 201 | |||
| 202 | if ($this->verbosity >= self::VERBOSITY_MEDIUM) { |
||
| 203 | $this->logger->info( |
||
| 204 | FormattedMessage::get( |
||
| 205 | FormattedMessage::FIRST_RESULT_PAGE_FOR_KEYWORD_ON_TARGET, |
||
| 206 | $keyword, |
||
| 207 | $target->getName() |
||
| 208 | ), |
||
| 209 | [$resultCrawler === null ? '[NULL]' : $resultCrawler->html()] |
||
| 210 | ); |
||
| 211 | } |
||
| 212 | |||
| 213 | // crawl |
||
| 214 | $this->scrape($target, $resultCrawler); |
||
| 215 | } |
||
| 216 | } else { |
||
| 217 | $this->tell(FormattedMessage::get(FormattedMessage::TARGET_INVALID_SEARCH_FORM, $target->getName())); |
||
| 218 | } |
||
| 219 | } |
||
| 220 | |||
| 221 | private function scrape(Target $target, ?Crawler $crawler): void |
||
| 222 | { |
||
| 223 | if ($crawler === null) { |
||
| 224 | return; |
||
| 225 | } |
||
| 226 | |||
| 227 | $markup = $target->getMarkup(); |
||
| 228 | $titleLinkSelector = $markup[TargetKey::MARKUP_TITLE]; |
||
| 229 | $markupInside = $markup[TargetKey::special(TargetKey::MARKUP_INSIDE)] ?? []; |
||
| 230 | $markupHasInside = !empty($markupInside); |
||
| 231 | $markupHasItemWrapper = false; |
||
| 232 | |||
| 233 | // choose link as title link selector if it is not empty and it is not set to a special key |
||
| 234 | if (!(empty($markup[TargetKey::MARKUP_LINK]) || ConfigProvider::isSpecialKey( |
||
| 235 | $markup[TargetKey::MARKUP_LINK] |
||
| 236 | ))) { |
||
| 237 | $titleLinkSelector = $markup[TargetKey::MARKUP_LINK]; |
||
| 238 | } |
||
| 239 | |||
| 240 | // determine what item wrapper is |
||
| 241 | $markup[TargetKey::special(TargetKey::ITEM_WRAPPER)] = Scanner::firstNonEmpty( |
||
| 242 | $markup, |
||
| 243 | [ |
||
| 244 | TargetKey::special(TargetKey::ITEM_WRAPPER), |
||
| 245 | TargetKey::special(TargetKey::RESULT), |
||
| 246 | TargetKey::special(TargetKey::ITEM), |
||
| 247 | TargetKey::special(TargetKey::WRAPPER), |
||
| 248 | ] |
||
| 249 | ); |
||
| 250 | if (!empty($markup[TargetKey::special(TargetKey::ITEM_WRAPPER)])) { |
||
| 251 | $markupHasItemWrapper = true; |
||
| 252 | } |
||
| 253 | |||
| 254 | // Page by page we go... |
||
| 255 | $page = self::INITIAL_PAGE; |
||
| 256 | |||
| 257 | do { |
||
| 258 | $this->tell( |
||
| 259 | FormattedMessage::get(FormattedMessage::PROCESSING_PAGE_N, $page), |
||
| 260 | self::COMM_DIRECTION_NONE |
||
| 261 | ); |
||
| 262 | |||
| 263 | // scrape each by carving the insides |
||
| 264 | $items = $crawler->filter($titleLinkSelector); |
||
| 265 | |||
| 266 | if ($markupHasItemWrapper) { |
||
| 267 | $items = $crawler->filter($markup[TargetKey::special(TargetKey::ITEM_WRAPPER)]); |
||
| 268 | } |
||
| 269 | |||
| 270 | if (!$items->count()) { |
||
| 271 | $this->tell( |
||
| 272 | FormattedMessage::get(FormattedMessage::NO_ITEMS_FOUND_ON_PAGE_N, $page) |
||
| 273 | ); |
||
| 274 | } |
||
| 275 | |||
| 276 | $items->each( |
||
| 277 | function (Crawler $itemCrawler) use ( |
||
| 278 | $markupHasInside, |
||
| 279 | $markupInside, |
||
| 280 | $target, |
||
| 281 | $titleLinkSelector |
||
| 282 | ) { |
||
| 283 | $cursor = $target->getCursor(); |
||
| 284 | |||
| 285 | try { |
||
| 286 | $titleLinkCrawler = $markupHasInside ? $itemCrawler : $itemCrawler->filter($titleLinkSelector); |
||
| 287 | $titleLinkText = Scanner::cleanText($titleLinkCrawler->text()); |
||
| 288 | $linkCrawler = $titleLinkCrawler->selectLink($titleLinkText); |
||
| 289 | |||
| 290 | // link crawler is empty in this case the link may be a parent element |
||
| 291 | // so we must find closest link: |
||
| 292 | if (!count($linkCrawler)) { |
||
| 293 | $linkCrawler = $this->nodeProximityAssistant->closest('a[href]', $titleLinkCrawler); |
||
| 294 | } |
||
| 295 | |||
| 296 | // simply get link from crawler |
||
| 297 | $link = $linkCrawler->link(); |
||
| 298 | |||
| 299 | $this->tell($titleLinkText, self::COMM_DIRECTION_FLAT); |
||
| 300 | } catch (InvalidArgumentException $e) { |
||
| 301 | $this->tell( |
||
| 302 | FormattedMessage::get( |
||
| 303 | ($markupHasInside |
||
| 304 | ? FormattedMessage::UNABLE_TO_RETRIEVE_SCRAP_FOR_X |
||
| 305 | : FormattedMessage::NO_TITLE_LINK_FOUND_FOR_X), |
||
| 306 | $cursor + 1 |
||
| 307 | ) |
||
| 308 | ); |
||
| 309 | $this->logger->error($e); |
||
| 310 | |||
| 311 | // escape 'each' block |
||
| 312 | return; |
||
| 313 | } |
||
| 314 | |||
| 315 | $target->incrementCursor(); |
||
| 316 | |||
| 317 | if ($markupHasInside) { |
||
| 318 | // grab handle on detail |
||
| 319 | $itemCrawler = $this->client->click($link); |
||
| 320 | $markupHasFocus = !empty($markupInside[TargetKey::special(TargetKey::MARKUP_FOCUS)]); |
||
| 321 | |||
| 322 | // focus detail crawler on section if specified |
||
| 323 | if ($markupHasFocus) { |
||
| 324 | $itemCrawler = $itemCrawler->filter( |
||
| 325 | $markupInside[TargetKey::special(TargetKey::MARKUP_FOCUS)] |
||
| 326 | ); |
||
| 327 | } |
||
| 328 | } |
||
| 329 | |||
| 330 | // collect the scrap... |
||
| 331 | $this->scrapper->collect( |
||
| 332 | new TitleLink($titleLinkText, $link->getUri()), |
||
| 333 | $target, |
||
| 334 | $itemCrawler |
||
| 335 | ); |
||
| 336 | } |
||
| 337 | ); |
||
| 338 | |||
| 339 | if ($page < $this->pageLimit && $target->hasPager()) { |
||
| 340 | // Look for next page. |
||
| 341 | // An InvalidArgumentException may be thrown if a 'next' link does not exist. |
||
| 342 | try { |
||
| 343 | // Select pager |
||
| 344 | $pager = $crawler->filter($target->getPager()[TargetKey::PAGER_SELECTOR]); |
||
| 345 | // Grab pager/next link |
||
| 346 | $nextLink = $pager->link(); |
||
| 347 | // Click it! |
||
| 348 | $crawler = $this->client->click($nextLink); |
||
| 349 | } catch (InvalidArgumentException $e) { |
||
| 350 | // Next link doesn't exist |
||
| 351 | $crawler = null; |
||
| 352 | |||
| 353 | if ($this->verbosity >= self::VERBOSITY_HIGH) { |
||
| 354 | $errorMessage = FormattedMessage::get( |
||
| 355 | FormattedMessage::TARGET_PAGER_NEXT_NOT_FOUND, |
||
| 356 | $target->getName(), |
||
| 357 | $target->getPager()[TargetKey::PAGER_SELECTOR], |
||
| 358 | $target->getPager()[TargetKey::PAGER_TEXT] |
||
| 359 | ); |
||
| 360 | |||
| 361 | $this->tell($errorMessage); |
||
| 362 | $this->logger->error($errorMessage); |
||
| 363 | } |
||
| 364 | } |
||
| 365 | |||
| 366 | ++$page; |
||
| 367 | } else { |
||
| 368 | $crawler = null; |
||
| 369 | } |
||
| 370 | |||
| 371 | // back-off |
||
| 372 | sleep($this->optionSet->getBackOff()); |
||
| 373 | } while ($crawler !== null); // unless Crawler died... |
||
| 374 | } |
||
| 375 | |||
| 376 | private function getTargetBuilder(): TargetBuilder |
||
| 383 | ) |
||
| 384 | ) |
||
| 385 | ); |
||
| 388 |