/** * Parses HTML document's meta tags, and fills document's metadata. * * @param HtmlDocument $document HTML document to scan for metadata. * @return void */ public function handle(HtmlDocument $document) { if (strlen($document->getTitle()) == 0) { $document->setTitle($this->getTitle($document)); } if (strlen($document->getDescription()) == 0) { $document->setDescription($document->getMetaContent('description')); } $this->imgElements = $document->extractElementAttributes('img'); if (strlen($document->getImage()) == 0) { $image = $this->getImage($document); if (strlen($image) > 0) { $document->setImage($image); } else { $imageCandidates = $this->getImageCandidates(); if (count($imageCandidates) === 1) { $document->setImage($imageCandidates[0]); } else { if (count($imageCandidates) > 1) { $document->setExtraField('IMAGES', $imageCandidates); } } } } }
/** * Parses HTML documents OpenGraph metadata * * @param HtmlDocument $document HTML document to be parsed. * @return void */ public function handle(HtmlDocument $document) { if (strlen($document->getTitle()) == 0) { $ogTitle = $document->getMetaContent('og:title'); if (strlen($ogTitle) > 0) { $document->setTitle($ogTitle); } } if (strlen($document->getDescription()) == 0) { $ogDescription = $document->getMetaContent('og:description'); if (strlen($ogDescription) > 0) { $document->setDescription($ogDescription); } } if (strlen($document->getImage()) == 0) { $ogImage = $document->getMetaContent('og:image:secure_url') ?: $document->getMetaContent('og:image'); if (strlen($ogImage) > 0) { $document->setImage($ogImage); } } if (!$document->getExtraField('SITE_NAME')) { $ogSiteName = $document->getMetaContent('og:site_name'); if (strlen($ogSiteName) > 0) { $document->setExtraField('SITE_NAME', $ogSiteName); } } /* Not really opengraph property :), but it's placed in opengraph parser to prevent executing full parser chain just to get favicon */ if (!$document->getExtraField('FAVICON')) { if ($favicon = $document->getLinkHref('icon')) { $document->setExtraField('FAVICON', $favicon); } } }