3 * @copyright Copyright (C) 2010-2022, the Friendica project
5 * @license GNU AGPL version 3 or any later version
7 * This program is free software: you can redistribute it and/or modify
8 * it under the terms of the GNU Affero General Public License as
9 * published by the Free Software Foundation, either version 3 of the
10 * License, or (at your option) any later version.
12 * This program is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 * GNU Affero General Public License for more details.
17 * You should have received a copy of the GNU Affero General Public License
18 * along with this program. If not, see <https://www.gnu.org/licenses/>.
22 namespace Friendica\Util;
26 use Friendica\Content\OEmbed;
27 use Friendica\Core\Hook;
28 use Friendica\Core\Logger;
29 use Friendica\Database\Database;
30 use Friendica\Database\DBA;
32 use Friendica\Network\HTTPClient\Client\HttpClientAccept;
33 use Friendica\Network\HTTPException;
34 use Friendica\Network\HTTPClient\Client\HttpClientOptions;
37 * Get information about a given URL
39 * Class with methods for extracting certain content from an url
43 const DEFAULT_EXPIRATION_FAILURE = 'now + 1 day';
44 const DEFAULT_EXPIRATION_SUCCESS = 'now + 3 months';
47 * Maximum number of characters for the description
49 const MAX_DESC_COUNT = 250;
52 * Minimum number of characters for the description
54 const MIN_DESC_COUNT = 100;
57 * Fetch the content type of the given url
58 * @param string $url URL of the page
59 * @param string $accept content-type to accept
61 * @return array content type
63 public static function getContentType(string $url, string $accept = HttpClientAccept::DEFAULT, int $timeout = 0): array
65 if (!empty($timeout)) {
66 $options = [HttpClientOptions::TIMEOUT => $timeout];
71 $curlResult = DI::httpClient()->head($url, array_merge([HttpClientOptions::ACCEPT_CONTENT => $accept], $options));
73 // Workaround for systems that can't handle a HEAD request. Don't retry on timeouts.
74 if (!$curlResult->isSuccess() && ($curlResult->getReturnCode() >= 400) && !in_array($curlResult->getReturnCode(), [408, 504])) {
75 $curlResult = DI::httpClient()->get($url, $accept, array_merge([HttpClientOptions::CONTENT_LENGTH => 1000000], $options));
78 if (!$curlResult->isSuccess()) {
79 Logger::debug('Got HTTP Error', ['http error' => $curlResult->getReturnCode(), 'url' => $url]);
83 $contenttype = $curlResult->getHeader('Content-Type')[0] ?? '';
84 if (empty($contenttype)) {
85 return ['application', 'octet-stream'];
88 return explode('/', current(explode(';', $contenttype)));
92 * Search for chached embeddable data of an url otherwise fetch it
94 * @param string $url The url of the page which should be scraped
95 * @param bool $do_oembed The false option is used by the function fetch_oembed()
96 * to avoid endless loops
98 * @return array which contains needed data for embedding
99 * string 'url' => The url of the parsed page
100 * string 'type' => Content type
101 * string 'title' => (optional) The title of the content
102 * string 'text' => (optional) The description for the content
103 * string 'image' => (optional) A preview image of the content
104 * array 'images' => (optional) Array of preview pictures
105 * string 'keywords' => (optional) The tags which belong to the content
107 * @throws HTTPException\InternalServerErrorException
108 * @see ParseUrl::getSiteinfo() for more information about scraping
111 public static function getSiteinfoCached(string $url, bool $do_oembed = true): array
120 $urlHash = hash('sha256', $url);
122 $parsed_url = DBA::selectFirst('parsed_url', ['content'],
123 ['url_hash' => $urlHash, 'oembed' => $do_oembed]
125 if (!empty($parsed_url['content'])) {
126 $data = unserialize($parsed_url['content']);
130 $data = self::getSiteinfo($url, $do_oembed);
132 $expires = $data['expires'];
134 unset($data['expires']);
139 'url_hash' => $urlHash,
140 'oembed' => $do_oembed,
142 'content' => serialize($data),
143 'created' => DateTimeFormat::utcNow(),
144 'expires' => $expires,
146 Database::INSERT_UPDATE
153 * Parse a page for embeddable content information
155 * This method parses to url for meta data which can be used to embed
156 * the content. If available it prioritizes Open Graph meta tags.
157 * If this is not available it uses the twitter cards meta tags.
158 * As fallback it uses standard html elements with meta informations
159 * like \<title\>Awesome Title\</title\> or
160 * \<meta name="description" content="An awesome description"\>
162 * @param string $url The url of the page which should be scraped
163 * @param bool $do_oembed The false option is used by the function fetch_oembed()
164 * to avoid endless loops
165 * @param int $count Internal counter to avoid endless loops
167 * @return array which contains needed data for embedding
168 * string 'url' => The url of the parsed page
169 * string 'type' => Content type (error, link, photo, image, audio, video)
170 * string 'title' => (optional) The title of the content
171 * string 'text' => (optional) The description for the content
172 * string 'image' => (optional) A preview image of the content
173 * array 'images' => (optional) Array of preview pictures
174 * string 'keywords' => (optional) The tags which belong to the content
176 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
177 * @todo https://developers.google.com/+/plugins/snippet/
179 * <meta itemprop="name" content="Awesome title">
180 * <meta itemprop="description" content="An awesome description">
181 * <meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
183 * <body itemscope itemtype="http://schema.org/Product">
184 * <h1 itemprop="name">Shiny Trinket</h1>
185 * <img itemprop="image" src="{image-url}" />
186 * <p itemprop="description">Shiny trinkets are shiny.</p>
190 public static function getSiteinfo(string $url, bool $do_oembed = true, int $count = 1): array
199 // Check if the URL does contain a scheme
200 $scheme = parse_url($url, PHP_URL_SCHEME);
203 $url = 'http://' . ltrim($url, '/');
206 $url = trim($url, "'\"");
208 $url = Network::stripTrackingQueryParams($url);
213 'expires' => DateTimeFormat::utc(self::DEFAULT_EXPIRATION_FAILURE),
217 Logger::warning('Endless loop detected', ['url' => $url]);
221 $type = self::getContentType($url);
222 Logger::info('Got content-type', ['content-type' => $type, 'url' => $url]);
223 if (!empty($type) && in_array($type[0], ['image', 'video', 'audio'])) {
224 $siteinfo['type'] = $type[0];
228 if ((count($type) >= 2) && (($type[0] != 'text') || ($type[1] != 'html'))) {
229 Logger::info('Unparseable content-type, quitting here, ', ['content-type' => $type, 'url' => $url]);
233 $curlResult = DI::httpClient()->get($url, HttpClientAccept::HTML, [HttpClientOptions::CONTENT_LENGTH => 1000000]);
234 if (!$curlResult->isSuccess() || empty($curlResult->getBody())) {
235 Logger::info('Empty body or error when fetching', ['url' => $url, 'success' => $curlResult->isSuccess(), 'code' => $curlResult->getReturnCode()]);
239 $siteinfo['expires'] = DateTimeFormat::utc(self::DEFAULT_EXPIRATION_SUCCESS);
241 if ($cacheControlHeader = $curlResult->getHeader('Cache-Control')[0] ?? '') {
242 if (preg_match('/max-age=([0-9]+)/i', $cacheControlHeader, $matches)) {
243 $maxAge = max(86400, (int)array_pop($matches));
244 $siteinfo['expires'] = DateTimeFormat::utc("now + $maxAge seconds");
248 $body = $curlResult->getBody();
251 $oembed_data = OEmbed::fetchURL($url, false, false);
253 if (!empty($oembed_data->type)) {
254 if (!in_array($oembed_data->type, ['error', 'rich', 'image', 'video', 'audio', ''])) {
255 $siteinfo['type'] = $oembed_data->type;
258 // See https://github.com/friendica/friendica/pull/5763#discussion_r217913178
259 if ($siteinfo['type'] != 'photo') {
260 if (!empty($oembed_data->title)) {
261 $siteinfo['title'] = trim($oembed_data->title);
263 if (!empty($oembed_data->description)) {
264 $siteinfo['text'] = trim($oembed_data->description);
266 if (!empty($oembed_data->author_name)) {
267 $siteinfo['author_name'] = trim($oembed_data->author_name);
269 if (!empty($oembed_data->author_url)) {
270 $siteinfo['author_url'] = trim($oembed_data->author_url);
272 if (!empty($oembed_data->provider_name)) {
273 $siteinfo['publisher_name'] = trim($oembed_data->provider_name);
275 if (!empty($oembed_data->provider_url)) {
276 $siteinfo['publisher_url'] = trim($oembed_data->provider_url);
278 if (!empty($oembed_data->thumbnail_url)) {
279 $siteinfo['image'] = $oembed_data->thumbnail_url;
286 // Look for a charset, first in headers
287 // Expected form: Content-Type: text/html; charset=ISO-8859-4
288 if (preg_match('/charset=([a-z0-9-_.\/]+)/i', $curlResult->getContentType(), $matches)) {
289 $charset = trim(trim(trim(array_pop($matches)), ';,'));
291 // Then in body that gets precedence
293 // - <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
294 // - <meta charset="utf-8">
295 // - <meta charset=utf-8>
296 // - <meta charSet="utf-8">
297 // We escape <style> and <script> tags since they can contain irrelevant charset information
298 // (see https://github.com/friendica/friendica/issues/9251#issuecomment-698636806)
299 Strings::performWithEscapedBlocks($body, '#<(?:style|script).*?</(?:style|script)>#ism', function ($body) use (&$charset) {
300 if (preg_match('/charset=["\']?([a-z0-9-_.\/]+)/i', $body, $matches)) {
301 $charset = trim(trim(trim(array_pop($matches)), ';,'));
306 $siteinfo['charset'] = $charset;
308 if ($charset && strtoupper($charset) != 'UTF-8') {
309 // See https://github.com/friendica/friendica/issues/5470#issuecomment-418351211
310 $charset = str_ireplace('latin-1', 'latin1', $charset);
312 Logger::info('detected charset', ['charset' => $charset]);
313 $body = iconv($charset, 'UTF-8//TRANSLIT', $body);
316 $body = mb_convert_encoding($body, 'HTML-ENTITIES', 'UTF-8');
322 $doc = new DOMDocument();
323 @$doc->loadHTML($body);
325 XML::deleteNode($doc, 'style');
326 XML::deleteNode($doc, 'option');
327 XML::deleteNode($doc, 'h1');
328 XML::deleteNode($doc, 'h2');
329 XML::deleteNode($doc, 'h3');
330 XML::deleteNode($doc, 'h4');
331 XML::deleteNode($doc, 'h5');
332 XML::deleteNode($doc, 'h6');
333 XML::deleteNode($doc, 'ol');
334 XML::deleteNode($doc, 'ul');
336 $xpath = new DOMXPath($doc);
338 $list = $xpath->query('//meta[@content]');
339 foreach ($list as $node) {
341 if ($node->attributes->length) {
342 foreach ($node->attributes as $attribute) {
343 $meta_tag[$attribute->name] = $attribute->value;
347 if (@$meta_tag['http-equiv'] == 'refresh') {
348 $path = $meta_tag['content'];
349 $pathinfo = explode(';', $path);
351 foreach ($pathinfo as $value) {
352 if (substr(strtolower($value), 0, 4) == 'url=') {
353 $content = substr($value, 4);
356 if ($content != '') {
357 $siteinfo = self::getSiteinfo($content, $do_oembed, ++$count);
363 $list = $xpath->query('//title');
364 if ($list->length > 0) {
365 $siteinfo['title'] = trim($list->item(0)->nodeValue);
368 $list = $xpath->query('//meta[@name]');
369 foreach ($list as $node) {
371 if ($node->attributes->length) {
372 foreach ($node->attributes as $attribute) {
373 $meta_tag[$attribute->name] = $attribute->value;
377 if (empty($meta_tag['content'])) {
381 $meta_tag['content'] = trim(html_entity_decode($meta_tag['content'], ENT_QUOTES, 'UTF-8'));
383 switch (strtolower($meta_tag['name'])) {
385 $siteinfo['title'] = trim($meta_tag['content']);
388 $siteinfo['text'] = trim($meta_tag['content']);
391 $siteinfo['image'] = $meta_tag['content'];
393 case 'twitter:image':
394 $siteinfo['image'] = $meta_tag['content'];
396 case 'twitter:image:src':
397 $siteinfo['image'] = $meta_tag['content'];
399 case 'twitter:description':
400 $siteinfo['text'] = trim($meta_tag['content']);
402 case 'twitter:title':
403 $siteinfo['title'] = trim($meta_tag['content']);
405 case 'twitter:player':
406 $siteinfo['player']['embed'] = trim($meta_tag['content']);
408 case 'twitter:player:stream':
409 $siteinfo['player']['stream'] = trim($meta_tag['content']);
411 case 'twitter:player:width':
412 $siteinfo['player']['width'] = intval($meta_tag['content']);
414 case 'twitter:player:height':
415 $siteinfo['player']['height'] = intval($meta_tag['content']);
418 $siteinfo['title'] = trim($meta_tag['content']);
420 case 'dc.description':
421 $siteinfo['text'] = trim($meta_tag['content']);
424 $siteinfo['publisher_name'] = trim($meta_tag['content']);
427 $keywords = explode(',', $meta_tag['content']);
429 case 'news_keywords':
430 $keywords = explode(',', $meta_tag['content']);
435 if (isset($keywords)) {
436 $siteinfo['keywords'] = [];
437 foreach ($keywords as $keyword) {
438 if (!in_array(trim($keyword), $siteinfo['keywords'])) {
439 $siteinfo['keywords'][] = trim($keyword);
444 $list = $xpath->query('//meta[@property]');
445 foreach ($list as $node) {
447 if ($node->attributes->length) {
448 foreach ($node->attributes as $attribute) {
449 $meta_tag[$attribute->name] = $attribute->value;
453 if (!empty($meta_tag['content'])) {
454 $meta_tag['content'] = trim(html_entity_decode($meta_tag['content'], ENT_QUOTES, 'UTF-8'));
456 switch (strtolower($meta_tag['property'])) {
458 $siteinfo['image'] = $meta_tag['content'];
461 $siteinfo['image'] = $meta_tag['content'];
463 case 'og:image:secure_url':
464 $siteinfo['image'] = $meta_tag['content'];
467 $siteinfo['title'] = trim($meta_tag['content']);
469 case 'og:description':
470 $siteinfo['text'] = trim($meta_tag['content']);
473 $siteinfo['publisher_name'] = trim($meta_tag['content']);
476 $siteinfo['language'] = trim($meta_tag['content']);
479 $siteinfo['pagetype'] = trim($meta_tag['content']);
481 case 'twitter:description':
482 $siteinfo['text'] = trim($meta_tag['content']);
484 case 'twitter:title':
485 $siteinfo['title'] = trim($meta_tag['content']);
487 case 'twitter:image':
488 $siteinfo['image'] = $meta_tag['content'];
494 $list = $xpath->query("//script[@type='application/ld+json']");
495 foreach ($list as $node) {
496 if (!empty($node->nodeValue)) {
497 if ($jsonld = json_decode($node->nodeValue, true)) {
498 $siteinfo = self::parseParts($siteinfo, $jsonld);
503 if (!empty($siteinfo['player']['stream'])) {
504 // Only add player data to media arrays if there is no duplicate
505 $content_urls = array_merge(array_column($siteinfo['audio'] ?? [], 'content'), array_column($siteinfo['video'] ?? [], 'content'));
506 if (!in_array($siteinfo['player']['stream'], $content_urls)) {
507 $contenttype = self::getContentType($siteinfo['player']['stream']);
508 if (!empty($contenttype[0]) && in_array($contenttype[0], ['audio', 'video'])) {
509 $media = ['content' => $siteinfo['player']['stream']];
511 if (!empty($siteinfo['player']['embed'])) {
512 $media['embed'] = $siteinfo['player']['embed'];
515 $siteinfo[$contenttype[0]][] = $media;
520 if (!empty($siteinfo['image'])) {
521 $siteinfo['images'] = $siteinfo['images'] ?? [];
522 array_unshift($siteinfo['images'], ['url' => $siteinfo['image']]);
523 unset($siteinfo['image']);
526 $siteinfo = self::checkMedia($url, $siteinfo);
528 if (!empty($siteinfo['text']) && mb_strlen($siteinfo['text']) > self::MAX_DESC_COUNT) {
529 $siteinfo['text'] = mb_substr($siteinfo['text'], 0, self::MAX_DESC_COUNT) . '…';
530 $pos = mb_strrpos($siteinfo['text'], '.');
531 if ($pos > self::MIN_DESC_COUNT) {
532 $siteinfo['text'] = mb_substr($siteinfo['text'], 0, $pos + 1);
536 Logger::info('Siteinfo fetched', ['url' => $url, 'siteinfo' => $siteinfo]);
538 Hook::callAll('getsiteinfo', $siteinfo);
546 * Check the attached media elements.
547 * Fix existing data and add missing data.
549 * @param string $page_url
550 * @param array $siteinfo
553 private static function checkMedia(string $page_url, array $siteinfo) : array
555 if (!empty($siteinfo['images'])) {
556 array_walk($siteinfo['images'], function (&$image) use ($page_url) {
558 * According to the specifications someone could place a picture
559 * URL into the content field as well. But this doesn't seem to
560 * happen in the wild, so we don't cover it here.
562 if (!empty($image['url'])) {
563 $image['url'] = self::completeUrl($image['url'], $page_url);
564 $photodata = Images::getInfoFromURLCached($image['url']);
565 if (($photodata) && ($photodata[0] > 50) && ($photodata[1] > 50)) {
566 $image['src'] = $image['url'];
567 $image['width'] = $photodata[0];
568 $image['height'] = $photodata[1];
569 $image['contenttype'] = $photodata['mime'];
570 unset($image['url']);
580 $siteinfo['images'] = array_values(array_filter($siteinfo['images']));
583 foreach (['audio', 'video'] as $element) {
584 if (!empty($siteinfo[$element])) {
585 array_walk($siteinfo[$element], function (&$media) use ($page_url, &$siteinfo) {
590 foreach (['embed', 'content', 'url'] as $field) {
591 if (!empty($media[$field])) {
592 $media[$field] = self::completeUrl($media[$field], $page_url);
593 $type = self::getContentType($media[$field]);
594 if (($type[0] ?? '') == 'text') {
595 if ($field == 'embed') {
596 $embed = $media[$field];
598 $url = $media[$field];
600 } elseif (!empty($type[0])) {
601 $content = $media[$field];
602 $contenttype = implode('/', $type);
605 unset($media[$field]);
608 foreach (['image', 'preview'] as $field) {
609 if (!empty($media[$field])) {
610 $media[$field] = self::completeUrl($media[$field], $page_url);
615 $media['url'] = $url;
617 if (!empty($embed)) {
618 $media['embed'] = $embed;
619 if (empty($siteinfo['player']['embed'])) {
620 $siteinfo['player']['embed'] = $embed;
623 if (!empty($content)) {
624 $media['src'] = $content;
626 if (!empty($contenttype)) {
627 $media['contenttype'] = $contenttype;
629 if (empty($url) && empty($content) && empty($embed)) {
635 $siteinfo[$element] = array_values(array_filter($siteinfo[$element]));
637 if (empty($siteinfo[$element])) {
638 unset($siteinfo[$element]);
645 * Convert tags from CSV to an array
647 * @param string $string Tags
649 * @return array with formatted Hashtags
651 public static function convertTagsToArray(string $string): array
653 $arr_tags = str_getcsv($string);
654 if (count($arr_tags)) {
655 // add the # sign to every tag
656 array_walk($arr_tags, ['self', 'arrAddHashes']);
664 * Add a hasht sign to a string
666 * This method is used as callback function
668 * @param string $tag The pure tag name
669 * @param int $k Counter for internal use
673 private static function arrAddHashes(string &$tag, int $k)
679 * Add a scheme to an url
681 * The src attribute of some html elements (e.g. images)
682 * can miss the scheme so we need to add the correct
685 * @param string $url The url which possibly does have
686 * a missing scheme (a link to an image)
687 * @param string $scheme The url with a correct scheme
688 * (e.g. the url from the webpage which does contain the image)
690 * @return string The url with a scheme
692 private static function completeUrl(string $url, string $scheme): string
694 $urlarr = parse_url($url);
696 // If the url does allready have an scheme
697 // we can stop the process here
698 if (isset($urlarr['scheme'])) {
702 $schemearr = parse_url($scheme);
704 $complete = $schemearr['scheme'] . '://' . $schemearr['host'];
706 if (!empty($schemearr['port'])) {
707 $complete .= ':' . $schemearr['port'];
710 if (!empty($urlarr['path'])) {
711 if (strpos($urlarr['path'], '/') !== 0) {
715 $complete .= $urlarr['path'];
718 if (!empty($urlarr['query'])) {
719 $complete .= '?' . $urlarr['query'];
722 if (!empty($urlarr['fragment'])) {
723 $complete .= '#' . $urlarr['fragment'];
730 * Parse the Json-Ld parts of a web page
732 * @param array $siteinfo
733 * @param array $jsonld
735 * @return array siteinfo
737 private static function parseParts(array $siteinfo, array $jsonld): array
739 if (!empty($jsonld['@graph']) && is_array($jsonld['@graph'])) {
740 foreach ($jsonld['@graph'] as $part) {
741 if (!empty($part) && is_array($part)) {
742 $siteinfo = self::parseParts($siteinfo, $part);
745 } elseif (!empty($jsonld['@type'])) {
746 $siteinfo = self::parseJsonLd($siteinfo, $jsonld);
747 } elseif (!empty($jsonld)) {
748 $keys = array_keys($jsonld);
749 $numeric_keys = true;
750 foreach ($keys as $key) {
752 $numeric_keys = false;
756 foreach ($jsonld as $part) {
757 if (!empty($part) && is_array($part)) {
758 $siteinfo = self::parseParts($siteinfo, $part);
764 array_walk_recursive($siteinfo, function (&$element) {
765 if (is_string($element)) {
766 $element = trim(strip_tags(html_entity_decode($element, ENT_COMPAT, 'UTF-8')));
774 * Improve the siteinfo with information from the provided JSON-LD information
775 * @see https://jsonld.com/
776 * @see https://schema.org/
778 * @param array $siteinfo
779 * @param array $jsonld
781 * @return array siteinfo
783 private static function parseJsonLd(array $siteinfo, array $jsonld): array
785 $type = JsonLD::fetchElement($jsonld, '@type');
787 Logger::info('Empty type', ['url' => $siteinfo['url']]);
791 // Silently ignore some types that aren't processed
792 if (in_array($type, ['SiteNavigationElement', 'JobPosting', 'CreativeWork', 'MusicAlbum',
793 'WPHeader', 'WPSideBar', 'WPFooter', 'LegalService', 'MusicRecording',
794 'ItemList', 'BreadcrumbList', 'Blog', 'Dataset', 'Product'])) {
800 case 'AdvertiserContentArticle':
803 case 'SatiricalArticle':
804 case 'ScholarlyArticle':
805 case 'SocialMediaPosting':
807 case 'ReportageNewsArticle':
808 case 'SocialMediaPosting':
810 case 'LiveBlogPosting':
811 case 'DiscussionForumPosting':
812 return self::parseJsonLdArticle($siteinfo, $jsonld);
816 case 'CollectionPage':
820 case 'MedicalWebPage':
823 case 'RealEstateListing':
824 case 'SearchResultsPage':
830 return self::parseJsonLdWebPage($siteinfo, $jsonld);
832 return self::parseJsonLdWebSite($siteinfo, $jsonld);
837 case 'EducationalOrganization':
838 case 'FundingScheme':
839 case 'GovernmentOrganization':
840 case 'LibrarySystem':
841 case 'LocalBusiness':
842 case 'MedicalOrganization':
844 case 'NewsMediaOrganization':
846 case 'SportsOrganization':
848 return self::parseJsonLdWebOrganization($siteinfo, $jsonld);
851 case 'PerformingGroup':
855 return self::parseJsonLdWebPerson($siteinfo, $jsonld);
858 return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'audio');
860 return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'video');
862 return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'images');
864 Logger::info('Unknown type', ['type' => $type, 'url' => $siteinfo['url']]);
870 * Fetch author and publisher data
872 * @param array $siteinfo
873 * @param array $jsonld
875 * @return array siteinfo
877 private static function parseJsonLdAuthor(array $siteinfo, array $jsonld): array
881 if (!empty($jsonld['publisher']) && is_array($jsonld['publisher'])) {
882 $content = JsonLD::fetchElement($jsonld, 'publisher', 'name');
883 if (!empty($content) && is_string($content)) {
884 $jsonldinfo['publisher_name'] = trim($content);
887 $content = JsonLD::fetchElement($jsonld, 'publisher', 'url');
888 if (!empty($content) && is_string($content)) {
889 $jsonldinfo['publisher_url'] = trim($content);
892 $brand = JsonLD::fetchElement($jsonld, 'publisher', 'brand', '@type', 'Organization');
893 if (!empty($brand) && is_array($brand)) {
894 $content = JsonLD::fetchElement($brand, 'name');
895 if (!empty($content) && is_string($content)) {
896 $jsonldinfo['publisher_name'] = trim($content);
899 $content = JsonLD::fetchElement($brand, 'url');
900 if (!empty($content) && is_string($content)) {
901 $jsonldinfo['publisher_url'] = trim($content);
904 $content = JsonLD::fetchElement($brand, 'logo', 'url');
905 if (!empty($content) && is_string($content)) {
906 $jsonldinfo['publisher_img'] = trim($content);
910 $logo = JsonLD::fetchElement($jsonld, 'publisher', 'logo');
911 if (!empty($logo) && is_array($logo)) {
912 $content = JsonLD::fetchElement($logo, 'url');
913 if (!empty($content) && is_string($content)) {
914 $jsonldinfo['publisher_img'] = trim($content);
917 } elseif (!empty($jsonld['publisher']) && is_string($jsonld['publisher'])) {
918 $jsonldinfo['publisher_name'] = trim($jsonld['publisher']);
921 if (!empty($jsonld['author']) && is_array($jsonld['author'])) {
922 $content = JsonLD::fetchElement($jsonld, 'author', 'name');
923 if (!empty($content) && is_string($content)) {
924 $jsonldinfo['author_name'] = trim($content);
927 $content = JsonLD::fetchElement($jsonld, 'author', 'sameAs');
928 if (!empty($content) && is_string($content)) {
929 $jsonldinfo['author_url'] = trim($content);
932 $content = JsonLD::fetchElement($jsonld, 'author', 'url');
933 if (!empty($content) && is_string($content)) {
934 $jsonldinfo['author_url'] = trim($content);
937 $logo = JsonLD::fetchElement($jsonld, 'author', 'logo');
938 if (!empty($logo) && is_array($logo)) {
939 $content = JsonLD::fetchElement($logo, 'url');
940 if (!empty($content) && is_string($content)) {
941 $jsonldinfo['author_img'] = trim($content);
944 } elseif (!empty($jsonld['author']) && is_string($jsonld['author'])) {
945 $jsonldinfo['author_name'] = trim($jsonld['author']);
948 Logger::info('Fetched Author information', ['fetched' => $jsonldinfo]);
950 return array_merge($siteinfo, $jsonldinfo);
954 * Fetch data from the provided JSON-LD Article type
955 * @see https://schema.org/Article
957 * @param array $siteinfo
958 * @param array $jsonld
960 * @return array siteinfo
962 private static function parseJsonLdArticle(array $siteinfo, array $jsonld): array
966 $content = JsonLD::fetchElement($jsonld, 'headline');
967 if (!empty($content) && is_string($content)) {
968 $jsonldinfo['title'] = trim($content);
971 $content = JsonLD::fetchElement($jsonld, 'alternativeHeadline');
972 if (!empty($content) && is_string($content) && (($jsonldinfo['title'] ?? '') != trim($content))) {
973 $jsonldinfo['alternative_title'] = trim($content);
976 $content = JsonLD::fetchElement($jsonld, 'description');
977 if (!empty($content) && is_string($content)) {
978 $jsonldinfo['text'] = trim($content);
981 $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
982 if (!empty($content)) {
983 $jsonldinfo['image'] = trim($content);
986 $content = JsonLD::fetchElement($jsonld, 'image', 'url', '@type', 'ImageObject');
987 if (!empty($content) && is_string($content)) {
988 $jsonldinfo['image'] = trim($content);
991 if (!empty($jsonld['keywords']) && !is_array($jsonld['keywords'])) {
992 $content = JsonLD::fetchElement($jsonld, 'keywords');
993 if (!empty($content)) {
994 $siteinfo['keywords'] = [];
995 $keywords = explode(',', $content);
996 foreach ($keywords as $keyword) {
997 $siteinfo['keywords'][] = trim($keyword);
1000 } elseif (!empty($jsonld['keywords'])) {
1001 $content = JsonLD::fetchElementArray($jsonld, 'keywords');
1002 if (!empty($content) && is_array($content)) {
1003 $jsonldinfo['keywords'] = $content;
1007 $content = JsonLD::fetchElement($jsonld, 'datePublished');
1008 if (!empty($content) && is_string($content)) {
1009 $jsonldinfo['published'] = DateTimeFormat::utc($content);
1012 $content = JsonLD::fetchElement($jsonld, 'dateModified');
1013 if (!empty($content) && is_string($content)) {
1014 $jsonldinfo['modified'] = DateTimeFormat::utc($content);
1017 $jsonldinfo = self::parseJsonLdAuthor($jsonldinfo, $jsonld);
1019 Logger::info('Fetched article information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
1021 return array_merge($siteinfo, $jsonldinfo);
1025 * Fetch data from the provided JSON-LD WebPage type
1026 * @see https://schema.org/WebPage
1028 * @param array $siteinfo
1029 * @param array $jsonld
1031 * @return array siteinfo
1033 private static function parseJsonLdWebPage(array $siteinfo, array $jsonld): array
1037 $content = JsonLD::fetchElement($jsonld, 'name');
1038 if (!empty($content)) {
1039 $jsonldinfo['title'] = trim($content);
1042 $content = JsonLD::fetchElement($jsonld, 'description');
1043 if (!empty($content) && is_string($content)) {
1044 $jsonldinfo['text'] = trim($content);
1047 $content = JsonLD::fetchElement($jsonld, 'image');
1048 if (!empty($content) && is_string($content)) {
1049 $jsonldinfo['image'] = trim($content);
1052 $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
1053 if (!empty($content) && is_string($content)) {
1054 $jsonldinfo['image'] = trim($content);
1057 $jsonldinfo = self::parseJsonLdAuthor($jsonldinfo, $jsonld);
1059 Logger::info('Fetched WebPage information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
1061 return array_merge($siteinfo, $jsonldinfo);
1065 * Fetch data from the provided JSON-LD WebSite type
1066 * @see https://schema.org/WebSite
1068 * @param array $siteinfo
1069 * @param array $jsonld
1071 * @return array siteinfo
1073 private static function parseJsonLdWebSite(array $siteinfo, array $jsonld): array
1077 $content = JsonLD::fetchElement($jsonld, 'name');
1078 if (!empty($content) && is_string($content)) {
1079 $jsonldinfo['publisher_name'] = trim($content);
1082 $content = JsonLD::fetchElement($jsonld, 'description');
1083 if (!empty($content) && is_string($content)) {
1084 $jsonldinfo['publisher_description'] = trim($content);
1087 $content = JsonLD::fetchElement($jsonld, 'url');
1088 if (!empty($content) && is_string($content)) {
1089 $jsonldinfo['publisher_url'] = trim($content);
1092 $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
1093 if (!empty($content) && is_string($content)) {
1094 $jsonldinfo['image'] = trim($content);
1097 $jsonldinfo = self::parseJsonLdAuthor($jsonldinfo, $jsonld);
1099 Logger::info('Fetched WebSite information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
1100 return array_merge($siteinfo, $jsonldinfo);
1104 * Fetch data from the provided JSON-LD Organization type
1105 * @see https://schema.org/Organization
1107 * @param array $siteinfo
1108 * @param array $jsonld
1110 * @return array siteinfo
1112 private static function parseJsonLdWebOrganization(array $siteinfo, array $jsonld): array
1116 $content = JsonLD::fetchElement($jsonld, 'name');
1117 if (!empty($content) && is_string($content)) {
1118 $jsonldinfo['publisher_name'] = trim($content);
1121 $content = JsonLD::fetchElement($jsonld, 'description');
1122 if (!empty($content) && is_string($content)) {
1123 $jsonldinfo['publisher_description'] = trim($content);
1126 $content = JsonLD::fetchElement($jsonld, 'url');
1127 if (!empty($content) && is_string($content)) {
1128 $jsonldinfo['publisher_url'] = trim($content);
1131 $content = JsonLD::fetchElement($jsonld, 'logo', 'url', '@type', 'ImageObject');
1132 if (!empty($content) && is_string($content)) {
1133 $jsonldinfo['publisher_img'] = trim($content);
1134 } elseif (!empty($content) && is_array($content)) {
1135 $jsonldinfo['publisher_img'] = trim($content[0]);
1138 $content = JsonLD::fetchElement($jsonld, 'brand', 'name', '@type', 'Organization');
1139 if (!empty($content) && is_string($content)) {
1140 $jsonldinfo['publisher_name'] = trim($content);
1143 $content = JsonLD::fetchElement($jsonld, 'brand', 'url', '@type', 'Organization');
1144 if (!empty($content) && is_string($content)) {
1145 $jsonldinfo['publisher_url'] = trim($content);
1148 Logger::info('Fetched Organization information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
1149 return array_merge($siteinfo, $jsonldinfo);
1153 * Fetch data from the provided JSON-LD Person type
1154 * @see https://schema.org/Person
1156 * @param array $siteinfo
1157 * @param array $jsonld
1159 * @return array siteinfo
1161 private static function parseJsonLdWebPerson(array $siteinfo, array $jsonld): array
1165 $content = JsonLD::fetchElement($jsonld, 'name');
1166 if (!empty($content) && is_string($content)) {
1167 $jsonldinfo['author_name'] = trim($content);
1170 $content = JsonLD::fetchElement($jsonld, 'description');
1171 if (!empty($content) && is_string($content)) {
1172 $jsonldinfo['author_description'] = trim($content);
1175 $content = JsonLD::fetchElement($jsonld, 'sameAs');
1176 if (!empty($content) && is_string($content)) {
1177 $jsonldinfo['author_url'] = trim($content);
1180 $content = JsonLD::fetchElement($jsonld, 'url');
1181 if (!empty($content) && is_string($content)) {
1182 $jsonldinfo['author_url'] = trim($content);
1185 $content = JsonLD::fetchElement($jsonld, 'image', 'url', '@type', 'ImageObject');
1186 if (!empty($content) && !is_string($content)) {
1187 Logger::notice('Unexpected return value for the author image', ['content' => $content]);
1190 if (!empty($content) && is_string($content)) {
1191 $jsonldinfo['author_img'] = trim($content);
1194 Logger::info('Fetched Person information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
1195 return array_merge($siteinfo, $jsonldinfo);
1199 * Fetch data from the provided JSON-LD MediaObject type
1200 * @see https://schema.org/MediaObject
1202 * @param array $siteinfo
1203 * @param array $jsonld
1205 * @return array siteinfo
1207 private static function parseJsonLdMediaObject(array $siteinfo, array $jsonld, string $name): array
1211 $content = JsonLD::fetchElement($jsonld, 'caption');
1212 if (!empty($content) && is_string($content)) {
1213 $media['caption'] = trim($content);
1216 $content = JsonLD::fetchElement($jsonld, 'url');
1217 if (!empty($content) && is_string($content)) {
1218 $media['url'] = trim($content);
1221 $content = JsonLD::fetchElement($jsonld, 'mainEntityOfPage');
1222 if (!empty($content) && is_string($content)) {
1223 $media['main'] = Strings::compareLink($content, $siteinfo['url']);
1226 $content = JsonLD::fetchElement($jsonld, 'description');
1227 if (!empty($content) && is_string($content)) {
1228 $media['description'] = trim($content);
1231 $content = JsonLD::fetchElement($jsonld, 'name');
1232 if (!empty($content) && (($media['description'] ?? '') != trim($content))) {
1233 $media['name'] = trim($content);
1236 $content = JsonLD::fetchElement($jsonld, 'contentUrl');
1237 if (!empty($content) && is_string($content)) {
1238 $media['content'] = trim($content);
1241 $content = JsonLD::fetchElement($jsonld, 'embedUrl');
1242 if (!empty($content) && is_string($content)) {
1243 $media['embed'] = trim($content);
1246 $content = JsonLD::fetchElement($jsonld, 'height');
1247 if (!empty($content) && is_string($content)) {
1248 $media['height'] = trim($content);
1251 $content = JsonLD::fetchElement($jsonld, 'width');
1252 if (!empty($content) && is_string($content)) {
1253 $media['width'] = trim($content);
1256 $content = JsonLD::fetchElement($jsonld, 'image');
1257 if (!empty($content) && is_string($content)) {
1258 $media['image'] = trim($content);
1261 $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
1262 if (!empty($content) && (($media['image'] ?? '') != trim($content))) {
1263 if (!empty($media['image'])) {
1264 $media['preview'] = trim($content);
1266 $media['image'] = trim($content);
1270 Logger::info('Fetched Media information', ['url' => $siteinfo['url'], 'fetched' => $media]);
1271 $siteinfo[$name][] = $media;