$siteinfo['type'] = 'link';
}
- if (!empty($siteinfo['image']) && empty($siteinfo['images'])) {
- $src = self::completeUrl($siteinfo['image'], $url);
-
+ if (!empty($siteinfo['image'])) {
+ $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'], '.');
}
/**
- * 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 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
*
- * @param array $siteinfo
- * @param array $jsonld
+ * 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) {
- $siteinfo = self::parseJsonLd($siteinfo, $part);
+ $siteinfo = self::parseParts($siteinfo, $part);
}
} elseif (!empty($jsonld['@type'])) {
$siteinfo = self::parseJsonLd($siteinfo, $jsonld);
/**
* Improve the siteinfo with information from the provided JSON-LD information
* @see https://jsonld.com/
+ * @see https://schema.org/
*
* @param array $siteinfo
* @param array $jsonld
}
// Silently ignore some types that aren't processed
- if (in_array($type, ['SiteNavigationElement', 'JobPosting', 'CreativeWork',
- 'WPHeader', 'WPSideBar', 'WPFooter', 'LegalService',
+ if (in_array($type, ['SiteNavigationElement', 'JobPosting', 'CreativeWork', 'MusicAlbum',
+ 'WPHeader', 'WPSideBar', 'WPFooter', 'LegalService', 'MusicRecording',
'ItemList', 'BreadcrumbList', 'Blog', 'Dataset', 'Product'])) {
return $siteinfo;
}
return self::parseJsonLdWebOrganization($siteinfo, $jsonld);
case 'Person':
case 'Patient':
- case 'PerformingGroup':
+ case 'PerformingGroup':
case 'DanceGroup';
case 'MusicGroup':
case 'TheaterGroup':
}
/**
- * 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
$brand = JsonLD::fetchElement($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', '@type', 'brand');
+
+ $content = JsonLD::fetchElement($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($content) && is_string($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
}
/**
- * 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
$jsonldinfo = self::parseJsonLdAuthor($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
}
/**
- * 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
}
/**
- * 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
}
/**
- * Improve the siteinfo with information from the provided JSON-LD MediaObject
+ * Fetch data from the provided JSON-LD MediaObject type
* @see https://schema.org/MediaObject
*
* @param array $siteinfo
$content = JsonLD::fetchElement($jsonld, 'url');
if (!empty($content)) {
- $media['src'] = trim($content);
+ $media['url'] = trim($content);
+ }
+
+ $content = JsonLD::fetchElement($jsonld, 'mainEntityOfPage');
+ if (!empty($content)) {
+ $media['main'] = Strings::compareLink($content, $siteinfo['url']);
}
$content = JsonLD::fetchElement($jsonld, 'description');
$media['width'] = trim($content);
}
- $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
- if (!empty($content)) {
- $media['preview'] = trim($content);
- }
-
$content = JsonLD::fetchElement($jsonld, 'image');
if (!empty($content)) {
$media['image'] = trim($content);
}
- Logger::info('Fetched Media information', ['url' => $siteinfo['url'], 'fetched' => $media]);
- $siteinfo[$name][] = $media;
- 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 .= "/";
+ $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
+ if (!empty($content) && (($media['image'] ?? '') != trim($content))) {
+ if (!empty($media['image'])) {
+ $media['preview'] = trim($content);
+ } else {
+ $media['image'] = trim($content);
}
-
- $complete .= $urlarr["path"];
}
- if (!empty($urlarr["query"])) {
- $complete .= "?".$urlarr["query"];
- }
-
- if (!empty($urlarr["fragment"])) {
- $complete .= "#".$urlarr["fragment"];
- }
-
- return($complete);
+ Logger::info('Fetched Media information', ['url' => $siteinfo['url'], 'fetched' => $media]);
+ $siteinfo[$name][] = $media;
+ return $siteinfo;
}
}
projects / friendica.git / blobdiff