X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FUtil%2FParseUrl.php;h=04afc927ba6cec4229c8167a6bcafe66e4c317d9;hb=720a43461d67ab229de0aecfc5008f22cc4c1c54;hp=e57be2412f66838082b015def920caa40eca07a0;hpb=08771d96c2c5ea27e004991ffd923f9a1f326e87;p=friendica.git diff --git a/src/Util/ParseUrl.php b/src/Util/ParseUrl.php index e57be2412f..04afc927ba 100644 --- a/src/Util/ParseUrl.php +++ b/src/Util/ParseUrl.php @@ -1,6 +1,6 @@ head($url); + $curlResult = DI::httpClient()->head($url); if (!$curlResult->isSuccess()) { return []; } - $contenttype = $curlResult->getHeader('Content-Type'); + $contenttype = $curlResult->getHeader('Content-Type')[0] ?? ''; if (empty($contenttype)) { return []; } @@ -197,7 +198,7 @@ class ParseUrl ]; if ($count > 10) { - Logger::log('Endless loop detected for ' . $url, Logger::DEBUG); + Logger::notice('Endless loop detected', ['url' => $url]); return $siteinfo; } @@ -213,26 +214,21 @@ class ParseUrl return $siteinfo; } - $curlResult = DI::httpRequest()->get($url); - if (!$curlResult->isSuccess()) { + $curlResult = DI::httpClient()->get($url, [HttpClientOptions::CONTENT_LENGTH => 1000000]); + if (!$curlResult->isSuccess() || empty($curlResult->getBody())) { + Logger::info('Empty body or error when fetching', ['url' => $url, 'success' => $curlResult->isSuccess(), 'code' => $curlResult->getReturnCode()]); return $siteinfo; } $siteinfo['expires'] = DateTimeFormat::utc(self::DEFAULT_EXPIRATION_SUCCESS); - // If the file is too large then exit - if (($curlResult->getInfo()['download_content_length'] ?? 0) > 1000000) { - return $siteinfo; - } - - if ($cacheControlHeader = $curlResult->getHeader('Cache-Control')) { + if ($cacheControlHeader = $curlResult->getHeader('Cache-Control')[0] ?? '') { if (preg_match('/max-age=([0-9]+)/i', $cacheControlHeader, $matches)) { $maxAge = max(86400, (int)array_pop($matches)); $siteinfo['expires'] = DateTimeFormat::utc("now + $maxAge seconds"); } } - $header = $curlResult->getHeader(); $body = $curlResult->getBody(); if ($do_oembed) { @@ -273,7 +269,7 @@ class ParseUrl $charset = ''; // Look for a charset, first in headers // Expected form: Content-Type: text/html; charset=ISO-8859-4 - if (preg_match('/charset=([a-z0-9-_.\/]+)/i', $header, $matches)) { + if (preg_match('/charset=([a-z0-9-_.\/]+)/i', $curlResult->getContentType(), $matches)) { $charset = trim(trim(trim(array_pop($matches)), ';,')); } @@ -297,7 +293,7 @@ class ParseUrl // See https://github.com/friendica/friendica/issues/5470#issuecomment-418351211 $charset = str_ireplace('latin-1', 'latin1', $charset); - Logger::log('detected charset ' . $charset, Logger::DEBUG); + Logger::info('detected charset', ['charset' => $charset]); $body = iconv($charset, 'UTF-8//TRANSLIT', $body); } @@ -380,18 +376,24 @@ class ParseUrl case 'twitter:image:src': $siteinfo['image'] = $meta_tag['content']; break; - case 'twitter:card': - // Detect photo pages - if ($meta_tag['content'] == 'summary_large_image') { - $siteinfo['type'] = 'photo'; - } - break; case 'twitter:description': $siteinfo['text'] = trim($meta_tag['content']); break; case 'twitter:title': $siteinfo['title'] = trim($meta_tag['content']); break; + case 'twitter:player': + $siteinfo['player']['embed'] = trim($meta_tag['content']); + break; + case 'twitter:player:stream': + $siteinfo['player']['stream'] = trim($meta_tag['content']); + break; + case 'twitter:player:width': + $siteinfo['player']['width'] = intval($meta_tag['content']); + break; + case 'twitter:player:height': + $siteinfo['player']['height'] = intval($meta_tag['content']); + break; case 'dc.title': $siteinfo['title'] = trim($meta_tag['content']); break; @@ -450,6 +452,12 @@ class ParseUrl case 'og:site_name': $siteinfo['publisher_name'] = trim($meta_tag['content']); break; + case 'og:locale': + $siteinfo['language'] = trim($meta_tag['content']); + break; + case 'og:type': + $siteinfo['pagetype'] = trim($meta_tag['content']); + break; case 'twitter:description': $siteinfo['text'] = trim($meta_tag['content']); break; @@ -466,38 +474,37 @@ class ParseUrl $list = $xpath->query("//script[@type='application/ld+json']"); foreach ($list as $node) { if (!empty($node->nodeValue)) { - $nodevalue = html_entity_decode($node->nodeValue, ENT_COMPAT, 'UTF-8'); - if ($jsonld = json_decode($nodevalue, true)) { - if (!empty($jsonld['@graph']) && is_array($jsonld['@graph'])) { - foreach ($jsonld['@graph'] as $part) { - $siteinfo = self::parseJsonLd($siteinfo, $part); - } - } else { - $siteinfo = self::parseJsonLd($siteinfo, $jsonld); - } + if ($jsonld = json_decode($node->nodeValue, true)) { + $siteinfo = self::parseParts($siteinfo, $jsonld); } } } - // Prevent to have a photo type without an image - if ((empty($siteinfo['image']) || !empty($siteinfo['text'])) && ($siteinfo['type'] == 'photo')) { - $siteinfo['type'] = 'link'; + if (!empty($siteinfo['player']['stream'])) { + // Only add player data to media arrays if there is no duplicate + $content_urls = array_merge(array_column($siteinfo['audio'] ?? [], 'content'), array_column($siteinfo['video'] ?? [], 'content')); + if (!in_array($siteinfo['player']['stream'], $content_urls)) { + $contenttype = self::getContentType($siteinfo['player']['stream']); + if (!empty($contenttype[0]) && in_array($contenttype[0], ['audio', 'video'])) { + $media = ['content' => $siteinfo['player']['stream']]; + + if (!empty($siteinfo['player']['embed'])) { + $media['embed'] = $siteinfo['player']['embed']; + } + + $siteinfo[$contenttype[0]][] = $media; + } + } } if (!empty($siteinfo['image'])) { - $src = self::completeUrl($siteinfo['image'], $url); - + $siteinfo['images'] = $siteinfo['images'] ?? []; + array_unshift($siteinfo['images'], ['url' => $siteinfo['image']]); unset($siteinfo['image']); - - $photodata = Images::getInfoFromURLCached($src); - - if (($photodata) && ($photodata[0] > 10) && ($photodata[1] > 10)) { - $siteinfo['images'][] = ['src' => $src, - 'width' => $photodata[0], - 'height' => $photodata[1]]; - } } + $siteinfo = self::checkMedia($url, $siteinfo); + if (!empty($siteinfo['text']) && mb_strlen($siteinfo['text']) > self::MAX_DESC_COUNT) { $siteinfo['text'] = mb_substr($siteinfo['text'], 0, self::MAX_DESC_COUNT) . '…'; $pos = mb_strrpos($siteinfo['text'], '.'); @@ -510,12 +517,236 @@ class ParseUrl Hook::callAll('getsiteinfo', $siteinfo); + ksort($siteinfo); + + return $siteinfo; + } + + /** + * Check the attached media elements. + * Fix existing data and add missing data. + * + * @param string $page_url + * @param array $siteinfo + * @return array + */ + private static function checkMedia(string $page_url, array $siteinfo) : array + { + if (!empty($siteinfo['images'])) { + array_walk($siteinfo['images'], function (&$image) use ($page_url) { + // According to the specifications someone could place a picture url into the content field as well. + // But this doesn't seem to happen in the wild, so we don't cover it here. + if (!empty($image['url'])) { + $image['url'] = self::completeUrl($image['url'], $page_url); + $photodata = Images::getInfoFromURLCached($image['url']); + if (!empty($photodata) && ($photodata[0] > 50) && ($photodata[1] > 50)) { + $image['src'] = $image['url']; + $image['width'] = $photodata[0]; + $image['height'] = $photodata[1]; + $image['contenttype'] = $photodata['mime']; + unset($image['url']); + ksort($image); + } else { + $image = []; + } + } else { + $image = []; + } + }); + + $siteinfo['images'] = array_values(array_filter($siteinfo['images'])); + } + + foreach (['audio', 'video'] as $element) { + if (!empty($siteinfo[$element])) { + array_walk($siteinfo[$element], function (&$media) use ($page_url, &$siteinfo) { + $url = ''; + $embed = ''; + $content = ''; + $contenttype = ''; + foreach (['embed', 'content', 'url'] as $field) { + if (!empty($media[$field])) { + $media[$field] = self::completeUrl($media[$field], $page_url); + $type = self::getContentType($media[$field]); + if (($type[0] ?? '') == 'text') { + if ($field == 'embed') { + $embed = $media[$field]; + } else { + $url = $media[$field]; + } + } elseif (!empty($type[0])) { + $content = $media[$field]; + $contenttype = implode('/', $type); + } + } + unset($media[$field]); + } + + foreach (['image', 'preview'] as $field) { + if (!empty($media[$field])) { + $media[$field] = self::completeUrl($media[$field], $page_url); + } + } + + if (!empty($url)) { + $media['url'] = $url; + } + if (!empty($embed)) { + $media['embed'] = $embed; + if (empty($siteinfo['player']['embed'])) { + $siteinfo['player']['embed'] = $embed; + } + } + if (!empty($content)) { + $media['src'] = $content; + } + if (!empty($contenttype)) { + $media['contenttype'] = $contenttype; + } + if (empty($url) && empty($content) && empty($embed)) { + $media = []; + } + ksort($media); + }); + + $siteinfo[$element] = array_values(array_filter($siteinfo[$element])); + } + if (empty($siteinfo[$element])) { + unset($siteinfo[$element]); + } + } + return $siteinfo; + } + + /** + * Convert tags from CSV to an array + * + * @param string $string Tags + * @return array with formatted Hashtags + */ + public static function convertTagsToArray($string) + { + $arr_tags = str_getcsv($string); + if (count($arr_tags)) { + // add the # sign to every tag + array_walk($arr_tags, ["self", "arrAddHashes"]); + + return $arr_tags; + } + } + + /** + * Add a hasht sign to a string + * + * This method is used as callback function + * + * @param string $tag The pure tag name + * @param int $k Counter for internal use + * @return void + */ + private static function arrAddHashes(&$tag, $k) + { + $tag = "#" . $tag; + } + + /** + * Add a scheme to an url + * + * The src attribute of some html elements (e.g. images) + * can miss the scheme so we need to add the correct + * scheme + * + * @param string $url The url which possibly does have + * a missing scheme (a link to an image) + * @param string $scheme The url with a correct scheme + * (e.g. the url from the webpage which does contain the image) + * + * @return string The url with a scheme + */ + private static function completeUrl($url, $scheme) + { + $urlarr = parse_url($url); + + // If the url does allready have an scheme + // we can stop the process here + if (isset($urlarr["scheme"])) { + return($url); + } + + $schemearr = parse_url($scheme); + + $complete = $schemearr["scheme"]."://".$schemearr["host"]; + + if (!empty($schemearr["port"])) { + $complete .= ":".$schemearr["port"]; + } + + if (!empty($urlarr["path"])) { + if (strpos($urlarr["path"], "/") !== 0) { + $complete .= "/"; + } + + $complete .= $urlarr["path"]; + } + + if (!empty($urlarr["query"])) { + $complete .= "?".$urlarr["query"]; + } + + if (!empty($urlarr["fragment"])) { + $complete .= "#".$urlarr["fragment"]; + } + + return($complete); + } + + /** + * Parse the Json-Ld parts of a web page + * + * @param array $siteinfo + * @param array $jsonld + * @return array siteinfo + */ + private static function parseParts(array $siteinfo, array $jsonld) + { + if (!empty($jsonld['@graph']) && is_array($jsonld['@graph'])) { + foreach ($jsonld['@graph'] as $part) { + if (!empty($part) && is_array($part)) { + $siteinfo = self::parseParts($siteinfo, $part); + } + } + } elseif (!empty($jsonld['@type'])) { + $siteinfo = self::parseJsonLd($siteinfo, $jsonld); + } elseif (!empty($jsonld)) { + $keys = array_keys($jsonld); + $numeric_keys = true; + foreach ($keys as $key) { + if (!is_int($key)) { + $numeric_keys = false; + } + } + if ($numeric_keys) { + foreach ($jsonld as $part) { + if (!empty($part) && is_array($part)) { + $siteinfo = self::parseParts($siteinfo, $part); + } + } + } + } + + array_walk_recursive($siteinfo, function (&$element) { + if (is_string($element)) { + $element = trim(strip_tags(html_entity_decode($element, ENT_COMPAT, 'UTF-8'))); + } + }); + return $siteinfo; } /** * Improve the siteinfo with information from the provided JSON-LD information * @see https://jsonld.com/ + * @see https://schema.org/ * * @param array $siteinfo * @param array $jsonld @@ -524,35 +755,91 @@ class ParseUrl private static function parseJsonLd(array $siteinfo, array $jsonld) { $type = JsonLD::fetchElement($jsonld, '@type'); + if (empty($type)) { + Logger::info('Empty type', ['url' => $siteinfo['url']]); + return $siteinfo; + } + + // Silently ignore some types that aren't processed + if (in_array($type, ['SiteNavigationElement', 'JobPosting', 'CreativeWork', 'MusicAlbum', + 'WPHeader', 'WPSideBar', 'WPFooter', 'LegalService', 'MusicRecording', + 'ItemList', 'BreadcrumbList', 'Blog', 'Dataset', 'Product'])) { + return $siteinfo; + } switch ($type) { case 'Article': + case 'AdvertiserContentArticle': case 'NewsArticle': + case 'Report': + case 'SatiricalArticle': + case 'ScholarlyArticle': + case 'SocialMediaPosting': + case 'TechArticle': + case 'ReportageNewsArticle': + case 'SocialMediaPosting': + case 'BlogPosting': + case 'LiveBlogPosting': + case 'DiscussionForumPosting': return self::parseJsonLdArticle($siteinfo, $jsonld); case 'WebPage': + case 'AboutPage': + case 'CheckoutPage': + case 'CollectionPage': + case 'ContactPage': + case 'FAQPage': + case 'ItemPage': + case 'MedicalWebPage': + case 'ProfilePage': + case 'QAPage': + case 'RealEstateListing': + case 'SearchResultsPage': + case 'MediaGallery': + case 'ImageGallery': + case 'VideoGallery': + case 'RadioEpisode': + case 'Event': return self::parseJsonLdWebPage($siteinfo, $jsonld); case 'WebSite': return self::parseJsonLdWebSite($siteinfo, $jsonld); case 'Organization': + case 'Airline': + case 'Consortium': + case 'Corporation': + case 'EducationalOrganization': + case 'FundingScheme': + case 'GovernmentOrganization': + case 'LibrarySystem': + case 'LocalBusiness': + case 'MedicalOrganization': + case 'NGO': + case 'NewsMediaOrganization': + case 'Project': + case 'SportsOrganization': + case 'WorkersUnion': return self::parseJsonLdWebOrganization($siteinfo, $jsonld); case 'Person': + case 'Patient': + case 'PerformingGroup': + case 'DanceGroup'; + case 'MusicGroup': + case 'TheaterGroup': return self::parseJsonLdWebPerson($siteinfo, $jsonld); - case 'BreadcrumbList': - case 'Audio': /// @todo Can contain direct media links to audio - can be interesting in the future + case 'AudioObject': + case 'Audio': + return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'audio'); case 'VideoObject': + return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'video'); case 'ImageObject': - case 'LiveBlogPosting': - case 'SocialMediaPosting': - // quit silently - return $siteinfo; + return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'images'); default: - Logger::info('Unsupported or unknown type', ['type' => $type, 'url' => $siteinfo['url']]); + Logger::info('Unknown type', ['type' => $type, 'url' => $siteinfo['url']]); return $siteinfo; } } /** - * Improve the siteinfo with information from the provided JSON-LD information concerning authors and publishers + * Fetch author and publisher data * * @param array $siteinfo * @param array $jsonld @@ -563,54 +850,80 @@ class ParseUrl $jsonldinfo = []; if (!empty($jsonld['publisher']) && is_array($jsonld['publisher'])) { - $content = JsonLD::fetchElement($jsonld, 'publisher', 'name', '@type', 'Organization'); + $content = JsonLD::fetchElement($jsonld, 'publisher', 'name'); if (!empty($content) && is_string($content)) { $jsonldinfo['publisher_name'] = trim($content); } - $content = JsonLD::fetchElement($jsonld, 'publisher', 'url', '@type', 'Organization'); + $content = JsonLD::fetchElement($jsonld, 'publisher', 'url'); if (!empty($content) && is_string($content)) { $jsonldinfo['publisher_url'] = trim($content); } $brand = JsonLD::fetchElement($jsonld, 'publisher', 'brand', '@type', 'Organization'); - if (!empty($brand)) { - $content = JsonLD::fetchElement($brand, 'name', '@type', 'brand'); + if (!empty($brand) && is_array($brand)) { + $content = JsonLD::fetchElement($brand, 'name'); if (!empty($content) && is_string($content)) { $jsonldinfo['publisher_name'] = trim($content); } + + $content = JsonLD::fetchElement($brand, 'url'); + if (!empty($content) && is_string($content)) { + $jsonldinfo['publisher_url'] = trim($content); + } + + $content = JsonLD::fetchElement($brand, 'logo', 'url'); + if (!empty($content) && is_string($content)) { + $jsonldinfo['publisher_img'] = trim($content); + } } + + $logo = JsonLD::fetchElement($jsonld, 'publisher', 'logo'); + if (!empty($logo) && is_array($logo)) { + $content = JsonLD::fetchElement($logo, 'url'); + if (!empty($content) && is_string($content)) { + $jsonldinfo['publisher_img'] = trim($content); + } + } + } elseif (!empty($jsonld['publisher']) && is_string($jsonld['publisher'])) { + $jsonldinfo['publisher_name'] = trim($jsonld['publisher']); } if (!empty($jsonld['author']) && is_array($jsonld['author'])) { - $content = JsonLD::fetchElement($jsonld, 'author', 'name', '@type', 'Organization'); + $content = JsonLD::fetchElement($jsonld, 'author', 'name'); if (!empty($content) && is_string($content)) { - $jsonldinfo['publisher_name'] = trim($content); + $jsonldinfo['author_name'] = trim($content); } - $content = JsonLD::fetchElement($jsonld, 'author', 'url', '@type', 'Organization'); + $content = JsonLD::fetchElement($jsonld, 'author', 'sameAs'); if (!empty($content) && is_string($content)) { - $jsonldinfo['publisher_url'] = trim($content); + $jsonldinfo['author_url'] = trim($content); } - $content = JsonLD::fetchElement($jsonld, 'author', 'name', '@type', 'Person'); + $content = JsonLD::fetchElement($jsonld, 'author', 'url'); if (!empty($content) && is_string($content)) { - $jsonldinfo['author_name'] = trim($content); + $jsonldinfo['author_url'] = trim($content); } - $content = JsonLD::fetchElement($jsonld, 'author', 'url', '@type', 'Person'); - if (!empty($content) && is_string($content)) { - $jsonldinfo['author_url'] = trim($content); + $logo = JsonLD::fetchElement($jsonld, 'author', 'logo'); + if (!empty($logo) && is_array($logo)) { + $content = JsonLD::fetchElement($logo, 'url'); + if (!empty($content) && is_string($content)) { + $jsonldinfo['author_img'] = trim($content); + } } + } elseif (!empty($jsonld['author']) && is_string($jsonld['author'])) { + $jsonldinfo['author_name'] = trim($jsonld['author']); } - Logger::info('Fetched author information', ['fetched' => $jsonldinfo]); + Logger::info('Fetched Author information', ['fetched' => $jsonldinfo]); return array_merge($siteinfo, $jsonldinfo); } /** - * Improve the siteinfo with information from the provided JSON-LD Article information + * Fetch data from the provided JSON-LD Article type + * @see https://schema.org/Article * * @param array $siteinfo * @param array $jsonld @@ -626,7 +939,7 @@ class ParseUrl } $content = JsonLD::fetchElement($jsonld, 'alternativeHeadline'); - if (!empty($content) && is_string($content)) { + if (!empty($content) && is_string($content) && (($jsonldinfo['title'] ?? '') != trim($content))) { $jsonldinfo['alternative_title'] = trim($content); } @@ -635,16 +948,41 @@ class ParseUrl $jsonldinfo['text'] = trim($content); } + $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl'); + if (!empty($content)) { + $jsonldinfo['image'] = trim($content); + } + $content = JsonLD::fetchElement($jsonld, 'image', 'url', '@type', 'ImageObject'); if (!empty($content) && is_string($content)) { $jsonldinfo['image'] = trim($content); } -/// @todo Check for the correct separator, also check for dpublicates before adding -// $content = JsonLD::fetchElement($jsonld, 'keywords'); -// if (!empty($content) && is_string($content)) { -// $jsonldinfo['keywords'] = trim($content); -// } + if (!empty($jsonld['keywords']) && !is_array($jsonld['keywords'])) { + $content = JsonLD::fetchElement($jsonld, 'keywords'); + if (!empty($content)) { + $siteinfo['keywords'] = []; + $keywords = explode(',', $content); + foreach ($keywords as $keyword) { + $siteinfo['keywords'][] = trim($keyword); + } + } + } elseif (!empty($jsonld['keywords'])) { + $content = JsonLD::fetchElementArray($jsonld, 'keywords'); + if (!empty($content) && is_array($content)) { + $jsonldinfo['keywords'] = $content; + } + } + + $content = JsonLD::fetchElement($jsonld, 'datePublished'); + if (!empty($content) && is_string($content)) { + $jsonldinfo['published'] = DateTimeFormat::utc($content); + } + + $content = JsonLD::fetchElement($jsonld, 'dateModified'); + if (!empty($content) && is_string($content)) { + $jsonldinfo['modified'] = DateTimeFormat::utc($content); + } $jsonldinfo = self::parseJsonLdAuthor($jsonldinfo, $jsonld); @@ -654,7 +992,8 @@ class ParseUrl } /** - * Improve the siteinfo with information from the provided JSON-LD WebPage information + * Fetch data from the provided JSON-LD WebPage type + * @see https://schema.org/WebPage * * @param array $siteinfo * @param array $jsonld @@ -674,6 +1013,11 @@ class ParseUrl $jsonldinfo['text'] = trim($content); } + $content = JsonLD::fetchElement($jsonld, 'image'); + if (!empty($content)) { + $jsonldinfo['image'] = trim($content); + } + $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl'); if (!empty($content)) { $jsonldinfo['image'] = trim($content); @@ -681,13 +1025,14 @@ class ParseUrl $jsonldinfo = self::parseJsonLdAuthor($jsonldinfo, $jsonld); - Logger::info('Fetched webpage information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]); + Logger::info('Fetched WebPage information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]); return array_merge($siteinfo, $jsonldinfo); } /** - * Improve the siteinfo with information from the provided JSON-LD WebSite information + * Fetch data from the provided JSON-LD WebSite type + * @see https://schema.org/WebSite * * @param array $siteinfo * @param array $jsonld @@ -724,7 +1069,8 @@ class ParseUrl } /** - * Improve the siteinfo with information from the provided JSON-LD Organization information + * Fetch data from the provided JSON-LD Organization type + * @see https://schema.org/Organization * * @param array $siteinfo * @param array $jsonld @@ -739,6 +1085,11 @@ class ParseUrl $jsonldinfo['publisher_name'] = trim($content); } + $content = JsonLD::fetchElement($jsonld, 'description'); + if (!empty($content)) { + $jsonldinfo['publisher_description'] = trim($content); + } + $content = JsonLD::fetchElement($jsonld, 'url'); if (!empty($content)) { $jsonldinfo['publisher_url'] = trim($content); @@ -749,12 +1100,23 @@ class ParseUrl $jsonldinfo['publisher_img'] = trim($content); } + $content = JsonLD::fetchElement($jsonld, 'brand', 'name', '@type', 'Organization'); + if (!empty($content)) { + $jsonldinfo['publisher_name'] = trim($content); + } + + $content = JsonLD::fetchElement($jsonld, 'brand', 'url', '@type', 'Organization'); + if (!empty($content)) { + $jsonldinfo['publisher_url'] = trim($content); + } + Logger::info('Fetched Organization information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]); return array_merge($siteinfo, $jsonldinfo); } /** - * Improve the siteinfo with information from the provided JSON-LD Person information + * Fetch data from the provided JSON-LD Person type + * @see https://schema.org/Person * * @param array $siteinfo * @param array $jsonld @@ -774,13 +1136,22 @@ class ParseUrl $jsonldinfo['author_description'] = trim($content); } + $content = JsonLD::fetchElement($jsonld, 'sameAs'); + if (!empty($content) && is_string($content)) { + $jsonldinfo['author_url'] = trim($content); + } + $content = JsonLD::fetchElement($jsonld, 'url'); if (!empty($content)) { $jsonldinfo['author_url'] = trim($content); } $content = JsonLD::fetchElement($jsonld, 'image', 'url', '@type', 'ImageObject'); - if (!empty($content)) { + if (!empty($content) && !is_string($content)) { + Logger::notice('Unexpected return value for the author image', ['content' => $content]); + } + + if (!empty($content) && is_string($content)) { $jsonldinfo['author_img'] = trim($content); } @@ -789,84 +1160,78 @@ class ParseUrl } /** - * Convert tags from CSV to an array + * Fetch data from the provided JSON-LD MediaObject type + * @see https://schema.org/MediaObject * - * @param string $string Tags - * @return array with formatted Hashtags + * @param array $siteinfo + * @param array $jsonld + * @return array siteinfo */ - public static function convertTagsToArray($string) + private static function parseJsonLdMediaObject(array $siteinfo, array $jsonld, string $name) { - $arr_tags = str_getcsv($string); - if (count($arr_tags)) { - // add the # sign to every tag - array_walk($arr_tags, ["self", "arrAddHashes"]); + $media = []; - return $arr_tags; + $content = JsonLD::fetchElement($jsonld, 'caption'); + if (!empty($content)) { + $media['caption'] = trim($content); } - } - /** - * Add a hasht sign to a string - * - * This method is used as callback function - * - * @param string $tag The pure tag name - * @param int $k Counter for internal use - * @return void - */ - private static function arrAddHashes(&$tag, $k) - { - $tag = "#" . $tag; - } + $content = JsonLD::fetchElement($jsonld, 'url'); + if (!empty($content)) { + $media['url'] = trim($content); + } - /** - * Add a scheme to an url - * - * The src attribute of some html elements (e.g. images) - * can miss the scheme so we need to add the correct - * scheme - * - * @param string $url The url which possibly does have - * a missing scheme (a link to an image) - * @param string $scheme The url with a correct scheme - * (e.g. the url from the webpage which does contain the image) - * - * @return string The url with a scheme - */ - private static function completeUrl($url, $scheme) - { - $urlarr = parse_url($url); + $content = JsonLD::fetchElement($jsonld, 'mainEntityOfPage'); + if (!empty($content)) { + $media['main'] = Strings::compareLink($content, $siteinfo['url']); + } - // If the url does allready have an scheme - // we can stop the process here - if (isset($urlarr["scheme"])) { - return($url); + $content = JsonLD::fetchElement($jsonld, 'description'); + if (!empty($content)) { + $media['description'] = trim($content); } - $schemearr = parse_url($scheme); + $content = JsonLD::fetchElement($jsonld, 'name'); + if (!empty($content) && (($media['description'] ?? '') != trim($content))) { + $media['name'] = trim($content); + } - $complete = $schemearr["scheme"]."://".$schemearr["host"]; + $content = JsonLD::fetchElement($jsonld, 'contentUrl'); + if (!empty($content)) { + $media['content'] = trim($content); + } - if (!empty($schemearr["port"])) { - $complete .= ":".$schemearr["port"]; + $content = JsonLD::fetchElement($jsonld, 'embedUrl'); + if (!empty($content)) { + $media['embed'] = trim($content); } - if (!empty($urlarr["path"])) { - if (strpos($urlarr["path"], "/") !== 0) { - $complete .= "/"; - } + $content = JsonLD::fetchElement($jsonld, 'height'); + if (!empty($content)) { + $media['height'] = trim($content); + } - $complete .= $urlarr["path"]; + $content = JsonLD::fetchElement($jsonld, 'width'); + if (!empty($content)) { + $media['width'] = trim($content); } - if (!empty($urlarr["query"])) { - $complete .= "?".$urlarr["query"]; + $content = JsonLD::fetchElement($jsonld, 'image'); + if (!empty($content)) { + $media['image'] = trim($content); } - if (!empty($urlarr["fragment"])) { - $complete .= "#".$urlarr["fragment"]; + $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl'); + if (!empty($content) && (($media['image'] ?? '') != trim($content))) { + if (!empty($media['image'])) { + $media['preview'] = trim($content); + } else { + $media['image'] = trim($content); + } } - return($complete); + Logger::info('Fetched Media information', ['url' => $siteinfo['url'], 'fetched' => $media]); + $siteinfo[$name][] = $media; + return $siteinfo; } }