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)
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::notice('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)), ';,'));
292 // Then in body that gets precedence
294 // - <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
295 // - <meta charset="utf-8">
296 // - <meta charset=utf-8>
297 // - <meta charSet="utf-8">
298 // We escape <style> and <script> tags since they can contain irrelevant charset information
299 // (see https://github.com/friendica/friendica/issues/9251#issuecomment-698636806)
300 Strings::performWithEscapedBlocks($body, '#<(?:style|script).*?</(?:style|script)>#ism', function ($body) use (&$charset) {
301 if (preg_match('/charset=["\']?([a-z0-9-_.\/]+)/i', $body, $matches)) {
302 $charset = trim(trim(trim(array_pop($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');
318 $doc = new DOMDocument();
319 @$doc->loadHTML($body);
321 XML::deleteNode($doc, 'style');
322 XML::deleteNode($doc, 'option');
323 XML::deleteNode($doc, 'h1');
324 XML::deleteNode($doc, 'h2');
325 XML::deleteNode($doc, 'h3');
326 XML::deleteNode($doc, 'h4');
327 XML::deleteNode($doc, 'h5');
328 XML::deleteNode($doc, 'h6');
329 XML::deleteNode($doc, 'ol');
330 XML::deleteNode($doc, 'ul');
332 $xpath = new DOMXPath($doc);
334 $list = $xpath->query('//meta[@content]');
335 foreach ($list as $node) {
337 if ($node->attributes->length) {
338 foreach ($node->attributes as $attribute) {
339 $meta_tag[$attribute->name] = $attribute->value;
343 if (@$meta_tag['http-equiv'] == 'refresh') {
344 $path = $meta_tag['content'];
345 $pathinfo = explode(';', $path);
347 foreach ($pathinfo as $value) {
348 if (substr(strtolower($value), 0, 4) == 'url=') {
349 $content = substr($value, 4);
352 if ($content != '') {
353 $siteinfo = self::getSiteinfo($content, $do_oembed, ++$count);
359 $list = $xpath->query('//title');
360 if ($list->length > 0) {
361 $siteinfo['title'] = trim($list->item(0)->nodeValue);
364 $list = $xpath->query('//meta[@name]');
365 foreach ($list as $node) {
367 if ($node->attributes->length) {
368 foreach ($node->attributes as $attribute) {
369 $meta_tag[$attribute->name] = $attribute->value;
373 if (empty($meta_tag['content'])) {
377 $meta_tag['content'] = trim(html_entity_decode($meta_tag['content'], ENT_QUOTES, 'UTF-8'));
379 switch (strtolower($meta_tag['name'])) {
381 $siteinfo['title'] = trim($meta_tag['content']);
384 $siteinfo['text'] = trim($meta_tag['content']);
387 $siteinfo['image'] = $meta_tag['content'];
389 case 'twitter:image':
390 $siteinfo['image'] = $meta_tag['content'];
392 case 'twitter:image:src':
393 $siteinfo['image'] = $meta_tag['content'];
395 case 'twitter:description':
396 $siteinfo['text'] = trim($meta_tag['content']);
398 case 'twitter:title':
399 $siteinfo['title'] = trim($meta_tag['content']);
401 case 'twitter:player':
402 $siteinfo['player']['embed'] = trim($meta_tag['content']);
404 case 'twitter:player:stream':
405 $siteinfo['player']['stream'] = trim($meta_tag['content']);
407 case 'twitter:player:width':
408 $siteinfo['player']['width'] = intval($meta_tag['content']);
410 case 'twitter:player:height':
411 $siteinfo['player']['height'] = intval($meta_tag['content']);
414 $siteinfo['title'] = trim($meta_tag['content']);
416 case 'dc.description':
417 $siteinfo['text'] = trim($meta_tag['content']);
420 $siteinfo['publisher_name'] = trim($meta_tag['content']);
423 $keywords = explode(',', $meta_tag['content']);
425 case 'news_keywords':
426 $keywords = explode(',', $meta_tag['content']);
429 Logger::debug('Unsupported meta-tag found', ['meta-tag' => $meta_tag]);
433 if (isset($keywords)) {
434 $siteinfo['keywords'] = [];
435 foreach ($keywords as $keyword) {
436 if (!in_array(trim($keyword), $siteinfo['keywords'])) {
437 $siteinfo['keywords'][] = trim($keyword);
442 $list = $xpath->query('//meta[@property]');
443 foreach ($list as $node) {
445 if ($node->attributes->length) {
446 foreach ($node->attributes as $attribute) {
447 $meta_tag[$attribute->name] = $attribute->value;
451 if (!empty($meta_tag['content'])) {
452 $meta_tag['content'] = trim(html_entity_decode($meta_tag['content'], ENT_QUOTES, 'UTF-8'));
454 switch (strtolower($meta_tag['property'])) {
456 $siteinfo['image'] = $meta_tag['content'];
459 $siteinfo['image'] = $meta_tag['content'];
461 case 'og:image:secure_url':
462 $siteinfo['image'] = $meta_tag['content'];
465 $siteinfo['title'] = trim($meta_tag['content']);
467 case 'og:description':
468 $siteinfo['text'] = trim($meta_tag['content']);
471 $siteinfo['publisher_name'] = trim($meta_tag['content']);
474 $siteinfo['language'] = trim($meta_tag['content']);
477 $siteinfo['pagetype'] = trim($meta_tag['content']);
479 case 'twitter:description':
480 $siteinfo['text'] = trim($meta_tag['content']);
482 case 'twitter:title':
483 $siteinfo['title'] = trim($meta_tag['content']);
485 case 'twitter:image':
486 $siteinfo['image'] = $meta_tag['content'];
492 $list = $xpath->query("//script[@type='application/ld+json']");
493 foreach ($list as $node) {
494 if (!empty($node->nodeValue)) {
495 if ($jsonld = json_decode($node->nodeValue, true)) {
496 $siteinfo = self::parseParts($siteinfo, $jsonld);
501 if (!empty($siteinfo['player']['stream'])) {
502 // Only add player data to media arrays if there is no duplicate
503 $content_urls = array_merge(array_column($siteinfo['audio'] ?? [], 'content'), array_column($siteinfo['video'] ?? [], 'content'));
504 if (!in_array($siteinfo['player']['stream'], $content_urls)) {
505 $contenttype = self::getContentType($siteinfo['player']['stream']);
506 if (!empty($contenttype[0]) && in_array($contenttype[0], ['audio', 'video'])) {
507 $media = ['content' => $siteinfo['player']['stream']];
509 if (!empty($siteinfo['player']['embed'])) {
510 $media['embed'] = $siteinfo['player']['embed'];
513 $siteinfo[$contenttype[0]][] = $media;
518 if (!empty($siteinfo['image'])) {
519 $siteinfo['images'] = $siteinfo['images'] ?? [];
520 array_unshift($siteinfo['images'], ['url' => $siteinfo['image']]);
521 unset($siteinfo['image']);
524 $siteinfo = self::checkMedia($url, $siteinfo);
526 if (!empty($siteinfo['text']) && mb_strlen($siteinfo['text']) > self::MAX_DESC_COUNT) {
527 $siteinfo['text'] = mb_substr($siteinfo['text'], 0, self::MAX_DESC_COUNT) . '…';
528 $pos = mb_strrpos($siteinfo['text'], '.');
529 if ($pos > self::MIN_DESC_COUNT) {
530 $siteinfo['text'] = mb_substr($siteinfo['text'], 0, $pos + 1);
534 Logger::info('Siteinfo fetched', ['url' => $url, 'siteinfo' => $siteinfo]);
536 Hook::callAll('getsiteinfo', $siteinfo);
544 * Check the attached media elements.
545 * Fix existing data and add missing data.
547 * @param string $page_url
548 * @param array $siteinfo
551 private static function checkMedia(string $page_url, array $siteinfo) : array
553 if (!empty($siteinfo['images'])) {
554 array_walk($siteinfo['images'], function (&$image) use ($page_url) {
556 * According to the specifications someone could place a picture
557 * URL into the content field as well. But this doesn't seem to
558 * happen in the wild, so we don't cover it here.
560 if (!empty($image['url'])) {
561 $image['url'] = self::completeUrl($image['url'], $page_url);
562 $photodata = Images::getInfoFromURLCached($image['url']);
563 if (($photodata) && ($photodata[0] > 50) && ($photodata[1] > 50)) {
564 $image['src'] = $image['url'];
565 $image['width'] = $photodata[0];
566 $image['height'] = $photodata[1];
567 $image['contenttype'] = $photodata['mime'];
568 unset($image['url']);
578 $siteinfo['images'] = array_values(array_filter($siteinfo['images']));
581 foreach (['audio', 'video'] as $element) {
582 if (!empty($siteinfo[$element])) {
583 array_walk($siteinfo[$element], function (&$media) use ($page_url, &$siteinfo) {
588 foreach (['embed', 'content', 'url'] as $field) {
589 if (!empty($media[$field])) {
590 $media[$field] = self::completeUrl($media[$field], $page_url);
591 $type = self::getContentType($media[$field]);
592 if (($type[0] ?? '') == 'text') {
593 if ($field == 'embed') {
594 $embed = $media[$field];
596 $url = $media[$field];
598 } elseif (!empty($type[0])) {
599 $content = $media[$field];
600 $contenttype = implode('/', $type);
603 unset($media[$field]);
606 foreach (['image', 'preview'] as $field) {
607 if (!empty($media[$field])) {
608 $media[$field] = self::completeUrl($media[$field], $page_url);
613 $media['url'] = $url;
615 if (!empty($embed)) {
616 $media['embed'] = $embed;
617 if (empty($siteinfo['player']['embed'])) {
618 $siteinfo['player']['embed'] = $embed;
621 if (!empty($content)) {
622 $media['src'] = $content;
624 if (!empty($contenttype)) {
625 $media['contenttype'] = $contenttype;
627 if (empty($url) && empty($content) && empty($embed)) {
633 $siteinfo[$element] = array_values(array_filter($siteinfo[$element]));
635 if (empty($siteinfo[$element])) {
636 unset($siteinfo[$element]);
643 * Convert tags from CSV to an array
645 * @param string $string Tags
646 * @return array with formatted Hashtags
648 public static function convertTagsToArray(string $string): array
650 $arr_tags = str_getcsv($string);
651 if (count($arr_tags)) {
652 // add the # sign to every tag
653 array_walk($arr_tags, ['self', 'arrAddHashes']);
661 * Add a hasht sign to a string
663 * This method is used as callback function
665 * @param string $tag The pure tag name
666 * @param int $k Counter for internal use
669 private static function arrAddHashes(string &$tag, int $k)
675 * Add a scheme to an url
677 * The src attribute of some html elements (e.g. images)
678 * can miss the scheme so we need to add the correct
681 * @param string $url The url which possibly does have
682 * a missing scheme (a link to an image)
683 * @param string $scheme The url with a correct scheme
684 * (e.g. the url from the webpage which does contain the image)
686 * @return string The url with a scheme
688 private static function completeUrl(string $url, string $scheme): string
690 $urlarr = parse_url($url);
692 // If the url does allready have an scheme
693 // we can stop the process here
694 if (isset($urlarr['scheme'])) {
698 $schemearr = parse_url($scheme);
700 $complete = $schemearr['scheme'] . '://' . $schemearr['host'];
702 if (!empty($schemearr['port'])) {
703 $complete .= ':' . $schemearr['port'];
706 if (!empty($urlarr['path'])) {
707 if (strpos($urlarr['path'], '/') !== 0) {
711 $complete .= $urlarr['path'];
714 if (!empty($urlarr['query'])) {
715 $complete .= '?' . $urlarr['query'];
718 if (!empty($urlarr['fragment'])) {
719 $complete .= '#' . $urlarr['fragment'];
726 * Parse the Json-Ld parts of a web page
728 * @param array $siteinfo
729 * @param array $jsonld
730 * @return array siteinfo
732 private static function parseParts(array $siteinfo, array $jsonld): array
734 if (!empty($jsonld['@graph']) && is_array($jsonld['@graph'])) {
735 foreach ($jsonld['@graph'] as $part) {
736 if (!empty($part) && is_array($part)) {
737 $siteinfo = self::parseParts($siteinfo, $part);
740 } elseif (!empty($jsonld['@type'])) {
741 $siteinfo = self::parseJsonLd($siteinfo, $jsonld);
742 } elseif (!empty($jsonld)) {
743 $keys = array_keys($jsonld);
744 $numeric_keys = true;
745 foreach ($keys as $key) {
747 $numeric_keys = false;
751 foreach ($jsonld as $part) {
752 if (!empty($part) && is_array($part)) {
753 $siteinfo = self::parseParts($siteinfo, $part);
759 array_walk_recursive($siteinfo, function (&$element) {
760 if (is_string($element)) {
761 $element = trim(strip_tags(html_entity_decode($element, ENT_COMPAT, 'UTF-8')));
769 * Improve the siteinfo with information from the provided JSON-LD information
770 * @see https://jsonld.com/
771 * @see https://schema.org/
773 * @param array $siteinfo
774 * @param array $jsonld
775 * @return array siteinfo
777 private static function parseJsonLd(array $siteinfo, array $jsonld): array
779 $type = JsonLD::fetchElement($jsonld, '@type');
781 Logger::info('Empty type', ['url' => $siteinfo['url']]);
785 // Silently ignore some types that aren't processed
786 if (in_array($type, ['SiteNavigationElement', 'JobPosting', 'CreativeWork', 'MusicAlbum',
787 'WPHeader', 'WPSideBar', 'WPFooter', 'LegalService', 'MusicRecording',
788 'ItemList', 'BreadcrumbList', 'Blog', 'Dataset', 'Product'])) {
794 case 'AdvertiserContentArticle':
797 case 'SatiricalArticle':
798 case 'ScholarlyArticle':
799 case 'SocialMediaPosting':
801 case 'ReportageNewsArticle':
802 case 'SocialMediaPosting':
804 case 'LiveBlogPosting':
805 case 'DiscussionForumPosting':
806 return self::parseJsonLdArticle($siteinfo, $jsonld);
810 case 'CollectionPage':
814 case 'MedicalWebPage':
817 case 'RealEstateListing':
818 case 'SearchResultsPage':
824 return self::parseJsonLdWebPage($siteinfo, $jsonld);
826 return self::parseJsonLdWebSite($siteinfo, $jsonld);
831 case 'EducationalOrganization':
832 case 'FundingScheme':
833 case 'GovernmentOrganization':
834 case 'LibrarySystem':
835 case 'LocalBusiness':
836 case 'MedicalOrganization':
838 case 'NewsMediaOrganization':
840 case 'SportsOrganization':
842 return self::parseJsonLdWebOrganization($siteinfo, $jsonld);
845 case 'PerformingGroup':
849 return self::parseJsonLdWebPerson($siteinfo, $jsonld);
852 return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'audio');
854 return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'video');
856 return self::parseJsonLdMediaObject($siteinfo, $jsonld, 'images');
858 Logger::info('Unknown type', ['type' => $type, 'url' => $siteinfo['url']]);
864 * Fetch author and publisher data
866 * @param array $siteinfo
867 * @param array $jsonld
868 * @return array siteinfo
870 private static function parseJsonLdAuthor(array $siteinfo, array $jsonld): array
874 if (!empty($jsonld['publisher']) && is_array($jsonld['publisher'])) {
875 $content = JsonLD::fetchElement($jsonld, 'publisher', 'name');
876 if (!empty($content) && is_string($content)) {
877 $jsonldinfo['publisher_name'] = trim($content);
880 $content = JsonLD::fetchElement($jsonld, 'publisher', 'url');
881 if (!empty($content) && is_string($content)) {
882 $jsonldinfo['publisher_url'] = trim($content);
885 $brand = JsonLD::fetchElement($jsonld, 'publisher', 'brand', '@type', 'Organization');
886 if (!empty($brand) && is_array($brand)) {
887 $content = JsonLD::fetchElement($brand, 'name');
888 if (!empty($content) && is_string($content)) {
889 $jsonldinfo['publisher_name'] = trim($content);
892 $content = JsonLD::fetchElement($brand, 'url');
893 if (!empty($content) && is_string($content)) {
894 $jsonldinfo['publisher_url'] = trim($content);
897 $content = JsonLD::fetchElement($brand, 'logo', 'url');
898 if (!empty($content) && is_string($content)) {
899 $jsonldinfo['publisher_img'] = trim($content);
903 $logo = JsonLD::fetchElement($jsonld, 'publisher', 'logo');
904 if (!empty($logo) && is_array($logo)) {
905 $content = JsonLD::fetchElement($logo, 'url');
906 if (!empty($content) && is_string($content)) {
907 $jsonldinfo['publisher_img'] = trim($content);
910 } elseif (!empty($jsonld['publisher']) && is_string($jsonld['publisher'])) {
911 $jsonldinfo['publisher_name'] = trim($jsonld['publisher']);
914 if (!empty($jsonld['author']) && is_array($jsonld['author'])) {
915 $content = JsonLD::fetchElement($jsonld, 'author', 'name');
916 if (!empty($content) && is_string($content)) {
917 $jsonldinfo['author_name'] = trim($content);
920 $content = JsonLD::fetchElement($jsonld, 'author', 'sameAs');
921 if (!empty($content) && is_string($content)) {
922 $jsonldinfo['author_url'] = trim($content);
925 $content = JsonLD::fetchElement($jsonld, 'author', 'url');
926 if (!empty($content) && is_string($content)) {
927 $jsonldinfo['author_url'] = trim($content);
930 $logo = JsonLD::fetchElement($jsonld, 'author', 'logo');
931 if (!empty($logo) && is_array($logo)) {
932 $content = JsonLD::fetchElement($logo, 'url');
933 if (!empty($content) && is_string($content)) {
934 $jsonldinfo['author_img'] = trim($content);
937 } elseif (!empty($jsonld['author']) && is_string($jsonld['author'])) {
938 $jsonldinfo['author_name'] = trim($jsonld['author']);
941 Logger::info('Fetched Author information', ['fetched' => $jsonldinfo]);
943 return array_merge($siteinfo, $jsonldinfo);
947 * Fetch data from the provided JSON-LD Article type
948 * @see https://schema.org/Article
950 * @param array $siteinfo
951 * @param array $jsonld
952 * @return array siteinfo
954 private static function parseJsonLdArticle(array $siteinfo, array $jsonld): array
958 $content = JsonLD::fetchElement($jsonld, 'headline');
959 if (!empty($content) && is_string($content)) {
960 $jsonldinfo['title'] = trim($content);
963 $content = JsonLD::fetchElement($jsonld, 'alternativeHeadline');
964 if (!empty($content) && is_string($content) && (($jsonldinfo['title'] ?? '') != trim($content))) {
965 $jsonldinfo['alternative_title'] = trim($content);
968 $content = JsonLD::fetchElement($jsonld, 'description');
969 if (!empty($content) && is_string($content)) {
970 $jsonldinfo['text'] = trim($content);
973 $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
974 if (!empty($content)) {
975 $jsonldinfo['image'] = trim($content);
978 $content = JsonLD::fetchElement($jsonld, 'image', 'url', '@type', 'ImageObject');
979 if (!empty($content) && is_string($content)) {
980 $jsonldinfo['image'] = trim($content);
983 if (!empty($jsonld['keywords']) && !is_array($jsonld['keywords'])) {
984 $content = JsonLD::fetchElement($jsonld, 'keywords');
985 if (!empty($content)) {
986 $siteinfo['keywords'] = [];
987 $keywords = explode(',', $content);
988 foreach ($keywords as $keyword) {
989 $siteinfo['keywords'][] = trim($keyword);
992 } elseif (!empty($jsonld['keywords'])) {
993 $content = JsonLD::fetchElementArray($jsonld, 'keywords');
994 if (!empty($content) && is_array($content)) {
995 $jsonldinfo['keywords'] = $content;
999 $content = JsonLD::fetchElement($jsonld, 'datePublished');
1000 if (!empty($content) && is_string($content)) {
1001 $jsonldinfo['published'] = DateTimeFormat::utc($content);
1004 $content = JsonLD::fetchElement($jsonld, 'dateModified');
1005 if (!empty($content) && is_string($content)) {
1006 $jsonldinfo['modified'] = DateTimeFormat::utc($content);
1009 $jsonldinfo = self::parseJsonLdAuthor($jsonldinfo, $jsonld);
1011 Logger::info('Fetched article information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
1013 return array_merge($siteinfo, $jsonldinfo);
1017 * Fetch data from the provided JSON-LD WebPage type
1018 * @see https://schema.org/WebPage
1020 * @param array $siteinfo
1021 * @param array $jsonld
1022 * @return array siteinfo
1024 private static function parseJsonLdWebPage(array $siteinfo, array $jsonld): array
1028 $content = JsonLD::fetchElement($jsonld, 'name');
1029 if (!empty($content)) {
1030 $jsonldinfo['title'] = trim($content);
1033 $content = JsonLD::fetchElement($jsonld, 'description');
1034 if (!empty($content)) {
1035 $jsonldinfo['text'] = trim($content);
1038 $content = JsonLD::fetchElement($jsonld, 'image');
1039 if (!empty($content)) {
1040 $jsonldinfo['image'] = trim($content);
1043 $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
1044 if (!empty($content)) {
1045 $jsonldinfo['image'] = trim($content);
1048 $jsonldinfo = self::parseJsonLdAuthor($jsonldinfo, $jsonld);
1050 Logger::info('Fetched WebPage information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
1052 return array_merge($siteinfo, $jsonldinfo);
1056 * Fetch data from the provided JSON-LD WebSite type
1057 * @see https://schema.org/WebSite
1059 * @param array $siteinfo
1060 * @param array $jsonld
1061 * @return array siteinfo
1063 private static function parseJsonLdWebSite(array $siteinfo, array $jsonld): array
1067 $content = JsonLD::fetchElement($jsonld, 'name');
1068 if (!empty($content)) {
1069 $jsonldinfo['publisher_name'] = trim($content);
1072 $content = JsonLD::fetchElement($jsonld, 'description');
1073 if (!empty($content)) {
1074 $jsonldinfo['publisher_description'] = trim($content);
1077 $content = JsonLD::fetchElement($jsonld, 'url');
1078 if (!empty($content)) {
1079 $jsonldinfo['publisher_url'] = trim($content);
1082 $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
1083 if (!empty($content)) {
1084 $jsonldinfo['image'] = trim($content);
1087 $jsonldinfo = self::parseJsonLdAuthor($jsonldinfo, $jsonld);
1089 Logger::info('Fetched WebSite information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
1090 return array_merge($siteinfo, $jsonldinfo);
1094 * Fetch data from the provided JSON-LD Organization type
1095 * @see https://schema.org/Organization
1097 * @param array $siteinfo
1098 * @param array $jsonld
1099 * @return array siteinfo
1101 private static function parseJsonLdWebOrganization(array $siteinfo, array $jsonld): array
1105 $content = JsonLD::fetchElement($jsonld, 'name');
1106 if (!empty($content)) {
1107 $jsonldinfo['publisher_name'] = trim($content);
1110 $content = JsonLD::fetchElement($jsonld, 'description');
1111 if (!empty($content)) {
1112 $jsonldinfo['publisher_description'] = trim($content);
1115 $content = JsonLD::fetchElement($jsonld, 'url');
1116 if (!empty($content)) {
1117 $jsonldinfo['publisher_url'] = trim($content);
1120 $content = JsonLD::fetchElement($jsonld, 'logo', 'url', '@type', 'ImageObject');
1121 if (!empty($content)) {
1122 $jsonldinfo['publisher_img'] = trim($content);
1125 $content = JsonLD::fetchElement($jsonld, 'brand', 'name', '@type', 'Organization');
1126 if (!empty($content)) {
1127 $jsonldinfo['publisher_name'] = trim($content);
1130 $content = JsonLD::fetchElement($jsonld, 'brand', 'url', '@type', 'Organization');
1131 if (!empty($content)) {
1132 $jsonldinfo['publisher_url'] = trim($content);
1135 Logger::info('Fetched Organization information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
1136 return array_merge($siteinfo, $jsonldinfo);
1140 * Fetch data from the provided JSON-LD Person type
1141 * @see https://schema.org/Person
1143 * @param array $siteinfo
1144 * @param array $jsonld
1145 * @return array siteinfo
1147 private static function parseJsonLdWebPerson(array $siteinfo, array $jsonld): array
1151 $content = JsonLD::fetchElement($jsonld, 'name');
1152 if (!empty($content)) {
1153 $jsonldinfo['author_name'] = trim($content);
1156 $content = JsonLD::fetchElement($jsonld, 'description');
1157 if (!empty($content)) {
1158 $jsonldinfo['author_description'] = trim($content);
1161 $content = JsonLD::fetchElement($jsonld, 'sameAs');
1162 if (!empty($content) && is_string($content)) {
1163 $jsonldinfo['author_url'] = trim($content);
1166 $content = JsonLD::fetchElement($jsonld, 'url');
1167 if (!empty($content)) {
1168 $jsonldinfo['author_url'] = trim($content);
1171 $content = JsonLD::fetchElement($jsonld, 'image', 'url', '@type', 'ImageObject');
1172 if (!empty($content) && !is_string($content)) {
1173 Logger::notice('Unexpected return value for the author image', ['content' => $content]);
1176 if (!empty($content) && is_string($content)) {
1177 $jsonldinfo['author_img'] = trim($content);
1180 Logger::info('Fetched Person information', ['url' => $siteinfo['url'], 'fetched' => $jsonldinfo]);
1181 return array_merge($siteinfo, $jsonldinfo);
1185 * Fetch data from the provided JSON-LD MediaObject type
1186 * @see https://schema.org/MediaObject
1188 * @param array $siteinfo
1189 * @param array $jsonld
1190 * @return array siteinfo
1192 private static function parseJsonLdMediaObject(array $siteinfo, array $jsonld, string $name): array
1196 $content = JsonLD::fetchElement($jsonld, 'caption');
1197 if (!empty($content)) {
1198 $media['caption'] = trim($content);
1201 $content = JsonLD::fetchElement($jsonld, 'url');
1202 if (!empty($content)) {
1203 $media['url'] = trim($content);
1206 $content = JsonLD::fetchElement($jsonld, 'mainEntityOfPage');
1207 if (!empty($content)) {
1208 $media['main'] = Strings::compareLink($content, $siteinfo['url']);
1211 $content = JsonLD::fetchElement($jsonld, 'description');
1212 if (!empty($content)) {
1213 $media['description'] = trim($content);
1216 $content = JsonLD::fetchElement($jsonld, 'name');
1217 if (!empty($content) && (($media['description'] ?? '') != trim($content))) {
1218 $media['name'] = trim($content);
1221 $content = JsonLD::fetchElement($jsonld, 'contentUrl');
1222 if (!empty($content)) {
1223 $media['content'] = trim($content);
1226 $content = JsonLD::fetchElement($jsonld, 'embedUrl');
1227 if (!empty($content)) {
1228 $media['embed'] = trim($content);
1231 $content = JsonLD::fetchElement($jsonld, 'height');
1232 if (!empty($content)) {
1233 $media['height'] = trim($content);
1236 $content = JsonLD::fetchElement($jsonld, 'width');
1237 if (!empty($content)) {
1238 $media['width'] = trim($content);
1241 $content = JsonLD::fetchElement($jsonld, 'image');
1242 if (!empty($content)) {
1243 $media['image'] = trim($content);
1246 $content = JsonLD::fetchElement($jsonld, 'thumbnailUrl');
1247 if (!empty($content) && (($media['image'] ?? '') != trim($content))) {
1248 if (!empty($media['image'])) {
1249 $media['preview'] = trim($content);
1251 $media['image'] = trim($content);
1255 Logger::info('Fetched Media information', ['url' => $siteinfo['url'], 'fetched' => $media]);
1256 $siteinfo[$name][] = $media;