<?php
/**
- * @copyright Copyright (C) 2020, Friendica
+ * @copyright Copyright (C) 2010-2023, the Friendica project
*
* @license GNU AGPL version 3 or any later version
*
use DOMDocument;
use DOMXPath;
use Friendica\Content\OEmbed;
+use Friendica\Content\Text\HTML;
+use Friendica\Protocol\HTTP\MediaType;
use Friendica\Core\Hook;
use Friendica\Core\Logger;
use Friendica\Database\Database;
use Friendica\Database\DBA;
use Friendica\DI;
+use Friendica\Network\HTTPClient\Client\HttpClientAccept;
use Friendica\Network\HTTPException;
+use Friendica\Network\HTTPClient\Client\HttpClientOptions;
/**
* Get information about a given URL
/**
* Fetch the content type of the given url
- * @param string $url URL of the page
- * @return array content type
+ * @param string $url URL of the page
+ * @param string $accept content-type to accept
+ * @param int $timeout
+ * @return array content type
*/
- public static function getContentType(string $url)
+ public static function getContentType(string $url, string $accept = HttpClientAccept::DEFAULT, int $timeout = 0): array
{
- $curlResult = DI::httpRequest()->head($url);
+ if (!empty($timeout)) {
+ $options = [HttpClientOptions::TIMEOUT => $timeout];
+ } else {
+ $options = [];
+ }
+
+ $curlResult = DI::httpClient()->head($url, array_merge([HttpClientOptions::ACCEPT_CONTENT => $accept], $options));
+
+ // Workaround for systems that can't handle a HEAD request. Don't retry on timeouts.
+ if (!$curlResult->isSuccess() && ($curlResult->getReturnCode() >= 400) && !in_array($curlResult->getReturnCode(), [408, 504])) {
+ $curlResult = DI::httpClient()->get($url, $accept, array_merge([HttpClientOptions::CONTENT_LENGTH => 1000000], $options));
+ }
+
if (!$curlResult->isSuccess()) {
+ Logger::debug('Got HTTP Error', ['http error' => $curlResult->getReturnCode(), 'url' => $url]);
return [];
}
- $contenttype = $curlResult->getHeader('Content-Type');
+ $contenttype = $curlResult->getHeader('Content-Type')[0] ?? '';
if (empty($contenttype)) {
- return [];
+ return ['application', 'octet-stream'];
}
return explode('/', current(explode(';', $contenttype)));
}
/**
- * Search for chached embeddable data of an url otherwise fetch it
+ * Search for cached embeddable data of an url otherwise fetch it
*
* @param string $url The url of the page which should be scraped
* @param bool $do_oembed The false option is used by the function fetch_oembed()
* @see ParseUrl::getSiteinfo() for more information about scraping
* embeddable content
*/
- public static function getSiteinfoCached($url, $do_oembed = true): array
+ public static function getSiteinfoCached(string $url, bool $do_oembed = true): array
{
if (empty($url)) {
return [
* </body>
* @endverbatim
*/
- public static function getSiteinfo($url, $do_oembed = true, $count = 1)
+ public static function getSiteinfo(string $url, bool $do_oembed = true, int $count = 1): array
{
if (empty($url)) {
return [
];
if ($count > 10) {
- Logger::log('Endless loop detected for ' . $url, Logger::DEBUG);
+ Logger::warning('Endless loop detected', ['url' => $url]);
return $siteinfo;
}
return $siteinfo;
}
- $curlResult = DI::httpRequest()->get($url);
- if (!$curlResult->isSuccess()) {
+ $curlResult = DI::httpClient()->get($url, HttpClientAccept::HTML, [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) {
}
$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)) {
- $charset = trim(trim(trim(array_pop($matches)), ';,'));
- }
-
- // Then in body that gets precedence
- // Expected forms:
- // - <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
- // - <meta charset="utf-8">
- // - <meta charset=utf-8>
- // - <meta charSet="utf-8">
- // We escape <style> and <script> tags since they can contain irrelevant charset information
- // (see https://github.com/friendica/friendica/issues/9251#issuecomment-698636806)
- Strings::performWithEscapedBlocks($body, '#<(?:style|script).*?</(?:style|script)>#ism', function ($body) use (&$charset) {
- if (preg_match('/charset=["\']?([a-z0-9-_.\/]+)/i', $body, $matches)) {
- $charset = trim(trim(trim(array_pop($matches)), ';,'));
+ try {
+ // Look for a charset, first in headers
+ $mediaType = MediaType::fromContentType($curlResult->getContentType());
+ if (isset($mediaType->parameters['charset'])) {
+ $charset = $mediaType->parameters['charset'];
}
- });
+ } catch(\InvalidArgumentException $e) {}
$siteinfo['charset'] = $charset;
// 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);
}
$body = mb_convert_encoding($body, 'HTML-ENTITIES', 'UTF-8');
+ if (empty($body)) {
+ return $siteinfo;
+ }
+
$doc = new DOMDocument();
@$doc->loadHTML($body);
+ $siteinfo['charset'] = HTML::extractCharset($doc) ?? $siteinfo['charset'];
+
XML::deleteNode($doc, 'style');
XML::deleteNode($doc, 'option');
XML::deleteNode($doc, 'h1');
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;
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;
$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 ($jsonld = json_decode($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'], '.');
Hook::callAll('getsiteinfo', $siteinfo);
+ ksort($siteinfo);
+
return $siteinfo;
}
/**
- * Parse the Json-Ld parts
+ * 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 (($photodata) && ($photodata[0] > 50) && ($photodata[1] > 50)) {
+ $image['src'] = $image['url'];
+ $image['width'] = $photodata[0];
+ $image['height'] = $photodata[1];
+ $image['contenttype'] = $photodata['mime'];
+ $image['blurhash'] = $photodata['blurhash'] ?? null;
+ unset($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 $string): array
+ {
+ $arr_tags = str_getcsv($string);
+ if (count($arr_tags)) {
+ // add the # sign to every tag
+ array_walk($arr_tags, [self::class, 'arrAddHashes']);
+
+ return $arr_tags;
+ }
+ return [];
+ }
+
+ /**
+ * Add a hasht sign to a string
+ *
+ * This method is used as callback function
+ *
+ * @param string $tag The pure tag name
+ * @param int $k Counter for internal use
+ *
+ * @return void
+ */
+ private static function arrAddHashes(string &$tag, int $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(string $url, string $scheme): string
+ {
+ $urlarr = parse_url($url);
+
+ // If the url does already 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
*
- * @param array $siteinfo
- * @param array $jsonld
* @return array siteinfo
*/
- private static function parseParts(array $siteinfo, array $jsonld)
+ private static function parseParts(array $siteinfo, array $jsonld): array
{
if (!empty($jsonld['@graph']) && is_array($jsonld['@graph'])) {
foreach ($jsonld['@graph'] as $part) {
- $siteinfo = self::parseJsonLd($siteinfo, $part);
+ if (!empty($part) && is_array($part)) {
+ $siteinfo = self::parseParts($siteinfo, $part);
+ }
}
} elseif (!empty($jsonld['@type'])) {
$siteinfo = self::parseJsonLd($siteinfo, $jsonld);
}
if ($numeric_keys) {
foreach ($jsonld as $part) {
- $siteinfo = self::parseParts($siteinfo, $part);
- }
+ if (!empty($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
+ *
* @return array siteinfo
*/
- private static function parseJsonLd(array $siteinfo, array $jsonld)
+ private static function parseJsonLd(array $siteinfo, array $jsonld): array
{
$type = JsonLD::fetchElement($jsonld, '@type');
if (empty($type)) {
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, $jsonld);
case 'Organization':
- case 'NewsMediaOrganization':
+ case 'Airline':
+ case 'Consortium':
+ case 'Corporation':
+ case 'EducationalOrganization':
+ case 'FundingScheme':
+ case 'GovernmentOrganization':
+ case 'LibrarySystem':
case 'LocalBusiness':
- return self::parseJsonLdWebOrganization($siteinfo, $jsonld);
+ case 'MedicalOrganization':
+ case 'NGO':
+ case 'NewsMediaOrganization':
+ case 'Project':
+ case 'SportsOrganization':
+ case 'WorkersUnion':
+ return self::parseJsonLdWebOrganization($siteinfo, $jsonld);
case 'Person':
+ case 'Patient':
+ case 'PerformingGroup':
+ case 'DanceGroup';
+ case 'MusicGroup':
+ case 'TheaterGroup':
return self::parseJsonLdWebPerson($siteinfo, $jsonld);
- case 'Audio':
case 'AudioObject':
- return self::parseJsonLdAudio($siteinfo, $jsonld);
-
+ case 'Audio':
+ return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'audio');
case 'VideoObject':
+ return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'video');
case 'ImageObject':
-
- case 'WPHeader': // Temp
- case 'WPSideBar': // Temp
- case 'WPFooter': // Temp
-
- case 'LiveBlogPosting':
- case 'SocialMediaPosting':
- case 'BreadcrumbList':
- case 'ItemList':
- case 'LegalService':
- case 'MusicGroup':
- case 'Blog':
- case 'BlogPosting':
- case 'Dataset':
- case 'CollectionPage':
- case 'ImageGallery':
- // 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
+ *
* @return array siteinfo
*/
- private static function parseJsonLdAuthor(array $siteinfo, array $jsonld)
+ private static function parseJsonLdAuthor(array $siteinfo, array $jsonld): array
{
$jsonldinfo = [];
$jsonldinfo['publisher_name'] = trim($content);
}
- $content = JsonLD::fetchElement($jsonld, 'publisher', 'sameAs');
- if (!empty($content) && is_string($content)) {
- $jsonldinfo['publisher_url'] = trim($content);
- }
-
$content = JsonLD::fetchElement($jsonld, 'publisher', 'url');
if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_url'] = trim($content);
$brand = JsonLD::fetchElement($jsonld, 'publisher', 'brand', '@type', 'Organization');
if (!empty($brand) && is_array($brand)) {
- $content = JsonLD::fetchElement($brand, 'name', '@type', '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($content) && is_string($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
+ *
* @return array siteinfo
*/
- private static function parseJsonLdArticle(array $siteinfo, array $jsonld)
+ private static function parseJsonLdArticle(array $siteinfo, array $jsonld): array
{
$jsonldinfo = [];
$siteinfo['keywords'][] = trim($keyword);
}
}
- } else {
+ } 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);
Logger::info('Fetched article information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
}
/**
- * 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
+ *
* @return array siteinfo
*/
- private static function parseJsonLdWebPage(array $siteinfo, array $jsonld)
+ private static function parseJsonLdWebPage(array $siteinfo, array $jsonld): array
{
$jsonldinfo = [];
}
$content = JsonLD::fetchElement($jsonld, 'description');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['text'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'image');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['image'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['image'] = trim($content);
}
$jsonldinfo = self::parseJsonLdAuthor($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
+ *
* @return array siteinfo
*/
- private static function parseJsonLdWebSite(array $siteinfo, array $jsonld)
+ private static function parseJsonLdWebSite(array $siteinfo, array $jsonld): array
{
$jsonldinfo = [];
$content = JsonLD::fetchElement($jsonld, 'name');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_name'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'description');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_description'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'url');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_url'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['image'] = trim($content);
}
}
/**
- * 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
+ *
* @return array siteinfo
*/
- private static function parseJsonLdWebOrganization(array $siteinfo, array $jsonld)
+ private static function parseJsonLdWebOrganization(array $siteinfo, array $jsonld): array
{
$jsonldinfo = [];
$content = JsonLD::fetchElement($jsonld, 'name');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_name'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'description');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_description'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'url');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_url'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'logo', 'url', '@type', 'ImageObject');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['publisher_img'] = trim($content);
+ } elseif (!empty($content) && is_array($content)) {
+ $jsonldinfo['publisher_img'] = trim($content[0]);
+ }
+
+ $content = JsonLD::fetchElement($jsonld, 'brand', 'name', '@type', 'Organization');
+ if (!empty($content) && is_string($content)) {
+ $jsonldinfo['publisher_name'] = trim($content);
+ }
+
+ $content = JsonLD::fetchElement($jsonld, 'brand', 'url', '@type', 'Organization');
+ if (!empty($content) && is_string($content)) {
+ $jsonldinfo['publisher_url'] = trim($content);
}
Logger::info('Fetched Organization information', ['url' => $siteinfo['url'], 'fetched' => $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
+ *
* @return array siteinfo
*/
- private static function parseJsonLdWebPerson(array $siteinfo, array $jsonld)
+ private static function parseJsonLdWebPerson(array $siteinfo, array $jsonld): array
{
$jsonldinfo = [];
$content = JsonLD::fetchElement($jsonld, 'name');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['author_name'] = trim($content);
}
$content = JsonLD::fetchElement($jsonld, 'description');
- if (!empty($content)) {
+ if (!empty($content) && is_string($content)) {
$jsonldinfo['author_description'] = trim($content);
}
+ $content = JsonLD::fetchElement($jsonld, 'sameAs');
+ if (!empty($content) && is_string($content)) {
+ $jsonldinfo['author_url'] = trim($content);
+ }
+
$content = JsonLD::fetchElement($jsonld, 'url');
- if (!empty($content)) {
+ if (!empty($content) && is_string($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);
}
}
/**
- * Improve the siteinfo with information from the provided JSON-LD Audio information
+ * Fetch data from the provided JSON-LD MediaObject type
+ * @see https://schema.org/MediaObject
*
* @param array $siteinfo
* @param array $jsonld
+ *
* @return array siteinfo
*/
- private static function parseJsonLdAudio(array $siteinfo, array $jsonld)
+ private static function parseJsonLdMediaObject(array $siteinfo, array $jsonld, string $name): array
{
- $jsonldinfo = [];
+ $media = [];
- $content = JsonLD::fetchElement($jsonld, 'description');
- if (!empty($content)) {
- $jsonldinfo['audio_description'] = trim($content);
+ $content = JsonLD::fetchElement($jsonld, 'caption');
+ if (!empty($content) && is_string($content)) {
+ $media['caption'] = trim($content);
}
- $content = JsonLD::fetchElement($jsonld, 'name');
- if (!empty($content)) {
- $jsonldinfo['audio_description'] = trim($content);
+ $content = JsonLD::fetchElement($jsonld, 'url');
+ if (!empty($content) && is_string($content)) {
+ $media['url'] = trim($content);
}
- $content = JsonLD::fetchElement($jsonld, 'contentUrl');
- if (!empty($content)) {
- $jsonldinfo['audio_url'] = trim($content);
+ $content = JsonLD::fetchElement($jsonld, 'mainEntityOfPage');
+ if (!empty($content) && is_string($content)) {
+ $media['main'] = Strings::compareLink($content, $siteinfo['url']);
}
- $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
- if (!empty($content)) {
- $jsonldinfo['audio_img'] = trim($content);
+ $content = JsonLD::fetchElement($jsonld, 'description');
+ if (!empty($content) && is_string($content)) {
+ $media['description'] = trim($content);
}
- Logger::info('Fetched Audio information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
- return array_merge($siteinfo, $jsonldinfo);
- }
-
- /**
- * 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;
+ $content = JsonLD::fetchElement($jsonld, 'name');
+ if (!empty($content) && (($media['description'] ?? '') != trim($content))) {
+ $media['name'] = trim($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;
- }
- /**
- * 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);
+ $content = JsonLD::fetchElement($jsonld, 'contentUrl');
+ if (!empty($content) && is_string($content)) {
+ $media['content'] = trim($content);
}
- $schemearr = parse_url($scheme);
-
- $complete = $schemearr["scheme"]."://".$schemearr["host"];
-
- if (!empty($schemearr["port"])) {
- $complete .= ":".$schemearr["port"];
+ $content = JsonLD::fetchElement($jsonld, 'embedUrl');
+ if (!empty($content) && is_string($content)) {
+ $media['embed'] = trim($content);
}
- if (!empty($urlarr["path"])) {
- if (strpos($urlarr["path"], "/") !== 0) {
- $complete .= "/";
- }
+ $content = JsonLD::fetchElement($jsonld, 'height');
+ if (!empty($content) && is_string($content)) {
+ $media['height'] = trim($content);
+ }
- $complete .= $urlarr["path"];
+ $content = JsonLD::fetchElement($jsonld, 'width');
+ if (!empty($content) && is_string($content)) {
+ $media['width'] = trim($content);
}
- if (!empty($urlarr["query"])) {
- $complete .= "?".$urlarr["query"];
+ $content = JsonLD::fetchElement($jsonld, 'image');
+ if (!empty($content) && is_string($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;
}
}