* Search for chached embeddable data of an url otherwise fetch it
*
* @param string $url The url of the page which should be scraped
- * @param bool $no_guessing If true the parse doens't search for
- * preview pictures
* @param bool $do_oembed The false option is used by the function fetch_oembed()
* to avoid endless loops
*
* string 'type' => Content type
* string 'title' => (optional) The title of the content
* string 'text' => (optional) The description for the content
- * string 'image' => (optional) A preview image of the content (only available if $no_geuessing = false)
+ * string 'image' => (optional) A preview image of the content
* array 'images' => (optional) Array of preview pictures
* string 'keywords' => (optional) The tags which belong to the content
*
* @see ParseUrl::getSiteinfo() for more information about scraping
* embeddable content
*/
- public static function getSiteinfoCached($url, $no_guessing = false, $do_oembed = true): array
+ public static function getSiteinfoCached($url, $do_oembed = true): array
{
if (empty($url)) {
return [
$urlHash = hash('sha256', $url);
$parsed_url = DBA::selectFirst('parsed_url', ['content'],
- ['url_hash' => $urlHash, 'guessing' => !$no_guessing, 'oembed' => $do_oembed]
+ ['url_hash' => $urlHash, 'oembed' => $do_oembed]
);
if (!empty($parsed_url['content'])) {
$data = unserialize($parsed_url['content']);
return $data;
}
- $data = self::getSiteinfo($url, $no_guessing, $do_oembed);
+ $data = self::getSiteinfo($url, $do_oembed);
$expires = $data['expires'];
'parsed_url',
[
'url_hash' => $urlHash,
- 'guessing' => !$no_guessing,
'oembed' => $do_oembed,
'url' => $url,
'content' => serialize($data),
* \<meta name="description" content="An awesome description"\>
*
* @param string $url The url of the page which should be scraped
- * @param bool $no_guessing If true the parse doens't search for
- * preview pictures
* @param bool $do_oembed The false option is used by the function fetch_oembed()
* to avoid endless loops
* @param int $count Internal counter to avoid endless loops
* string 'type' => Content type (error, link, photo, image, audio, video)
* string 'title' => (optional) The title of the content
* string 'text' => (optional) The description for the content
- * string 'image' => (optional) A preview image of the content (only available if $no_guessing = false)
+ * string 'image' => (optional) A preview image of the content
* array 'images' => (optional) Array of preview pictures
* string 'keywords' => (optional) The tags which belong to the content
*
* </body>
* @endverbatim
*/
- public static function getSiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1)
+ public static function getSiteinfo($url, $do_oembed = true, $count = 1)
{
if (empty($url)) {
return [
$body = $curlResult->getBody();
if ($do_oembed) {
- $oembed_data = OEmbed::fetchURL($url);
+ $oembed_data = OEmbed::fetchURL($url, false, false);
if (!empty($oembed_data->type)) {
if (!in_array($oembed_data->type, ['error', 'rich', 'image', 'video', 'audio', ''])) {
// See https://github.com/friendica/friendica/pull/5763#discussion_r217913178
if ($siteinfo['type'] != 'photo') {
- if (isset($oembed_data->title)) {
+ if (!empty($oembed_data->title)) {
$siteinfo['title'] = trim($oembed_data->title);
}
- if (isset($oembed_data->description)) {
+ if (!empty($oembed_data->description)) {
$siteinfo['text'] = trim($oembed_data->description);
}
- if (isset($oembed_data->thumbnail_url)) {
+ if (!empty($oembed_data->author_name)) {
+ $siteinfo['author_name'] = trim($oembed_data->author_name);
+ }
+ if (!empty($oembed_data->author_url)) {
+ $siteinfo['author_url'] = trim($oembed_data->author_url);
+ }
+ if (!empty($oembed_data->provider_name)) {
+ $siteinfo['publisher_name'] = trim($oembed_data->provider_name);
+ }
+ if (!empty($oembed_data->provider_url)) {
+ $siteinfo['publisher_url'] = trim($oembed_data->provider_url);
+ }
+ if (!empty($oembed_data->thumbnail_url)) {
$siteinfo['image'] = $oembed_data->thumbnail_url;
}
}
}
}
if ($content != '') {
- $siteinfo = self::getSiteinfo($content, $no_guessing, $do_oembed, ++$count);
+ $siteinfo = self::getSiteinfo($content, $do_oembed, ++$count);
return $siteinfo;
}
}
if (!empty($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);
- }
+ $siteinfo = self::parseParts($siteinfo, $jsonld);
}
}
}
}
if (!empty($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'], '.');
return $siteinfo;
}
+ /**
+ * Check the attached media elements.
+ * Fix existing data and add missing data.
+ *
+ * @param string $page_url
+ * @param array $siteinfo
+ * @return void
+ */
+ private static function checkMedia(string $page_url, array $siteinfo)
+ {
+ if (!empty($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.
+ $image['url'] = self::completeUrl($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 = [];
+ }
+ });
+
+ $siteinfo['images'] = array_values(array_filter($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'] = $embed;
+ if (!empty($media['main'])) {
+ $siteinfo['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) {
+ $siteinfo = self::parseParts($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) {
+ $siteinfo = self::parseParts($siteinfo, $part);
+ }
+ }
+ }
+
+ 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
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
$jsonldinfo = [];
if (!empty($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', '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)) {
- $content = JsonLD::fetchElement($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, 'sameAs');
+ if (!empty($content) && is_string($content)) {
+ $jsonldinfo['publisher_url'] = trim($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
}
$content = JsonLD::fetchElement($jsonld, 'alternativeHeadline');
- if (!empty($content) && is_string($content)) {
+ if (!empty($content) && is_string($content) && (($jsonldinfo['title'] ?? '') != trim($content))) {
$jsonldinfo['alternative_title'] = trim($content);
}
$jsonldinfo['text'] = trim($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);
+ }
+ }
+ } else {
+ $content = JsonLD::fetchElementArray($jsonld, 'keywords');
+ if (!empty($content) && is_array($content)) {
+ $jsonldinfo['keywords'] = $content;
+ }
+ }
$jsonldinfo = self::parseJsonLdAuthor($jsonldinfo, $jsonld);
}
/**
- * 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
$jsonldinfo['text'] = trim($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);
$jsonldinfo = self::parseJsonLdAuthor($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
}
/**
- * 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
$jsonldinfo['publisher_name'] = trim($content);
}
+ $content = JsonLD::fetchElement($jsonld, 'description');
+ if (!empty($content)) {
+ $jsonldinfo['publisher_description'] = trim($content);
+ }
+
+ $content = JsonLD::fetchElement($jsonld, 'sameAs');
+ if (!empty($content) && is_string($content)) {
+ $jsonldinfo['publisher_url'] = trim($content);
+ }
+
$content = JsonLD::fetchElement($jsonld, 'url');
if (!empty($content)) {
$jsonldinfo['publisher_url'] = trim($content);
$jsonldinfo['publisher_img'] = trim($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
$jsonldinfo['author_description'] = trim($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);
}
/**
- * 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;
}
}
projects / friendica.git / blobdiff