3 * @copyright Copyright (C) 2010-2023, 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\Network;
27 use Friendica\Core\Hook;
28 use Friendica\Core\Logger;
29 use Friendica\Core\Protocol;
30 use Friendica\Core\System;
31 use Friendica\Database\DBA;
33 use Friendica\Model\Contact;
34 use Friendica\Model\GServer;
35 use Friendica\Model\Profile;
36 use Friendica\Model\User;
37 use Friendica\Network\HTTPClient\Client\HttpClientAccept;
38 use Friendica\Network\HTTPClient\Client\HttpClientOptions;
39 use Friendica\Protocol\ActivityNamespace;
40 use Friendica\Protocol\ActivityPub;
41 use Friendica\Protocol\Diaspora;
42 use Friendica\Protocol\Email;
43 use Friendica\Protocol\Feed;
44 use Friendica\Protocol\Salmon;
45 use Friendica\Util\Crypto;
46 use Friendica\Util\DateTimeFormat;
47 use Friendica\Util\Network;
48 use Friendica\Util\Strings;
49 use Friendica\Util\XML;
50 use GuzzleHttp\Psr7\Uri;
53 * This class contain functions for probing URL
57 const HOST_META = '/.well-known/host-meta';
58 const WEBFINGER = '/.well-known/webfinger?resource={uri}';
61 * @var string Base URL
63 private static $baseurl;
66 * @var boolean Whether a timeout has occurred
68 private static $isTimeout;
71 * Checks if the provided network can be probed
73 * @param string $network
77 public static function isProbable(string $network): bool
79 return (in_array($network, array_merge(Protocol::FEDERATED, [Protocol::ZOT, Protocol::PHANTOM])));
83 * Remove stuff from an URI that doesn't belong there
85 * @param string $rawUri
86 * @return string Cleaned URI
88 public static function cleanURI(string $rawUri): string
90 // At first remove leading and trailing junk
91 $rawUri = trim($rawUri, "@#?: \t\n\r\0\x0B");
93 $rawUri = Network::convertToIdn($rawUri);
95 $uri = new Uri($rawUri);
96 if (!$uri->getScheme()) {
97 return $uri->__toString();
100 // Remove the URL fragment, since these shouldn't be part of any profile URL
101 $uri = $uri->withFragment('');
103 return $uri->__toString();
107 * Rearrange the array so that it always has the same order
109 * @param array $data Unordered data
110 * @return array Ordered data
112 private static function rearrangeData(array $data): array
114 $fields = ['name', 'given_name', 'family_name', 'nick', 'guid', 'url', 'addr', 'alias',
115 'photo', 'photo_medium', 'photo_small', 'header',
116 'account-type', 'community', 'keywords', 'location', 'about', 'xmpp', 'matrix',
117 'hide', 'batch', 'notify', 'poll', 'request', 'confirm', 'subscribe', 'poco',
118 'following', 'followers', 'inbox', 'outbox', 'sharedinbox',
119 'priority', 'network', 'pubkey', 'manually-approve', 'baseurl', 'gsid'];
121 $numeric_fields = ['gsid', 'hide', 'account-type', 'manually-approve'];
123 if (!empty($data['photo'])) {
124 $data['photo'] = Network::addBasePath($data['photo'], $data['url']);
126 if (!Network::isValidHttpUrl($data['photo'])) {
127 Logger::warning('Invalid URL for photo', ['url' => $data['url'], 'photo' => $data['photo']]);
128 unset($data['photo']);
133 foreach ($fields as $field) {
134 if (isset($data[$field])) {
135 if (in_array($field, $numeric_fields)) {
136 $newdata[$field] = (int)$data[$field];
138 $newdata[$field] = trim($data[$field]);
140 } elseif (!in_array($field, $numeric_fields)) {
141 $newdata[$field] = '';
143 $newdata[$field] = null;
147 $newdata['networks'] = [];
148 foreach ([Protocol::DIASPORA, Protocol::OSTATUS] as $network) {
149 if (!empty($data['networks'][$network])) {
150 $data['networks'][$network]['subscribe'] = $newdata['subscribe'] ?? '';
151 if (empty($data['networks'][$network]['baseurl'])) {
152 $data['networks'][$network]['baseurl'] = $newdata['baseurl'] ?? '';
154 $newdata['baseurl'] = $data['networks'][$network]['baseurl'];
156 if (!empty($newdata['baseurl'])) {
157 $newdata['gsid'] = $data['networks'][$network]['gsid'] = GServer::getID($newdata['baseurl']);
159 $newdata['gsid'] = $data['networks'][$network]['gsid'] = null;
162 $newdata['networks'][$network] = self::rearrangeData($data['networks'][$network]);
163 unset($newdata['networks'][$network]['networks']);
167 // We don't use the "priority" field anymore and replace it with a dummy.
168 $newdata['priority'] = 0;
174 * Check if the hostname belongs to the own server
176 * @param string $host The hostname that is to be checked
177 * @return bool Does the testes hostname belongs to the own server?
179 private static function ownHost(string $host): bool
181 $own_host = DI::baseUrl()->getHost();
183 $parts = parse_url($host);
185 if (!isset($parts['scheme'])) {
186 $parts = parse_url('http://' . $host);
189 if (!isset($parts['host'])) {
192 return $parts['host'] == $own_host;
196 * Probes for webfinger path via "host-meta"
198 * We have to check if the servers in the future still will offer this.
199 * It seems as if it was dropped from the standard.
201 * @param string $host The host part of an url
203 * @return array with template and type of the webfinger template for JSON or XML
204 * @throws HTTPException\InternalServerErrorException
206 private static function hostMeta(string $host): array
208 // Reset the static variable
211 // Handles the case when the hostname contains the scheme
212 if (!parse_url($host, PHP_URL_SCHEME)) {
213 $ssl_url = 'https://' . $host . self::HOST_META;
214 $url = 'http://' . $host . self::HOST_META;
216 $ssl_url = $host . self::HOST_META;
220 $xrd_timeout = DI::config()->get('system', 'xrd_timeout', 20);
222 Logger::info('Probing', ['host' => $host, 'ssl_url' => $ssl_url, 'url' => $url, 'callstack' => System::callstack(20)]);
225 $curlResult = DI::httpClient()->get($ssl_url, HttpClientAccept::XRD_XML, [HttpClientOptions::TIMEOUT => $xrd_timeout]);
226 $ssl_connection_error = ($curlResult->getErrorNumber() == CURLE_COULDNT_CONNECT) || ($curlResult->getReturnCode() == 0);
227 if ($curlResult->isSuccess()) {
228 $xml = $curlResult->getBody();
229 $xrd = XML::parseString($xml, true);
231 $host_url = 'https://' . $host;
235 } elseif ($curlResult->isTimeout()) {
236 Logger::info('Probing timeout', ['url' => $ssl_url]);
237 self::$isTimeout = true;
241 if (!is_object($xrd) && !empty($url)) {
242 $curlResult = DI::httpClient()->get($url, HttpClientAccept::XRD_XML, [HttpClientOptions::TIMEOUT => $xrd_timeout]);
243 $connection_error = ($curlResult->getErrorNumber() == CURLE_COULDNT_CONNECT) || ($curlResult->getReturnCode() == 0);
244 if ($curlResult->isTimeout()) {
245 Logger::info('Probing timeout', ['url' => $url]);
246 self::$isTimeout = true;
248 } elseif ($connection_error && $ssl_connection_error) {
249 self::$isTimeout = true;
253 $xml = $curlResult->getBody();
254 $xrd = XML::parseString($xml, true);
255 $host_url = 'http://'.$host;
257 if (!is_object($xrd)) {
258 Logger::info('No xrd object found', ['host' => $host]);
262 $links = XML::elementToArray($xrd);
263 if (!isset($links['xrd']['link'])) {
264 Logger::info('No xrd data found', ['host' => $host]);
270 foreach ($links['xrd']['link'] as $value => $link) {
271 if (!empty($link['@attributes'])) {
272 $attributes = $link['@attributes'];
273 } elseif ($value == '@attributes') {
279 if (!empty($attributes['rel']) && $attributes['rel'] == 'lrdd' && !empty($attributes['template'])) {
280 $type = (empty($attributes['type']) ? '' : $attributes['type']);
282 $lrdd[$type] = $attributes['template'];
286 if (Network::isUrlBlocked($host_url)) {
287 Logger::info('Domain is blocked', ['url' => $host]);
291 self::$baseurl = $host_url;
293 Logger::info('Probing successful', ['host' => $host]);
299 * Check an URI for LRDD data
301 * @param string $uri Address that should be probed
302 * @return array uri data
303 * @throws HTTPException\InternalServerErrorException
305 public static function lrdd(string $uri): array
307 $data = self::getWebfingerArray($uri);
311 $webfinger = $data['webfinger'];
313 if (empty($webfinger['links'])) {
314 Logger::info('No webfinger links found', ['uri' => $uri]);
320 foreach ($webfinger['links'] as $link) {
321 $data[] = ['@attributes' => $link];
324 if (!empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
325 foreach ($webfinger['aliases'] as $alias) {
339 * Fetch information (protocol endpoints and user information) about a given uri
341 * @param string $uri Address that should be probed
342 * @param string $network Test for this specific network
343 * @param integer $uid User ID for the probe (only used for mails)
344 * @param boolean $cache Use cached values?
346 * @return array uri data
347 * @throws HTTPException\InternalServerErrorException
348 * @throws \ImagickException
350 public static function uri(string $uri, string $network = '', int $uid = -1): array
352 // Local profiles aren't probed via network
353 if (empty($network) && Contact::isLocal($uri)) {
354 $data = self::localProbe($uri);
361 $uid = DI::userSession()->getLocalUserId();
364 if (empty($network) || ($network == Protocol::ACTIVITYPUB)) {
365 $ap_profile = ActivityPub::probeProfile($uri);
370 self::$isTimeout = false;
372 if ($network != Protocol::ACTIVITYPUB) {
373 $data = self::detect($uri, $network, $uid, $ap_profile);
374 if (!is_array($data)) {
377 if (empty($data) || (!empty($ap_profile) && empty($network) && (($data['network'] ?? '') != Protocol::DFRN))) {
378 $networks = $data['networks'] ?? [];
379 unset($data['networks']);
380 if (!empty($data['network'])) {
381 $networks[$data['network']] = $data;
384 $data['networks'] = $networks;
385 } elseif (!empty($ap_profile)) {
386 $ap_profile['batch'] = '';
387 $data = array_merge($ap_profile, $data);
393 if (!isset($data['url'])) {
397 if (empty($data['photo'])) {
398 $data['photo'] = DI::baseUrl() . Contact::DEFAULT_AVATAR_PHOTO;
401 if (empty($data['name'])) {
402 if (!empty($data['nick'])) {
403 $data['name'] = $data['nick'];
406 if (empty($data['name'])) {
407 $data['name'] = $data['url'];
411 if (empty($data['nick'])) {
412 $data['nick'] = strtolower($data['name']);
414 if (strpos($data['nick'], ' ')) {
415 $data['nick'] = trim(substr($data['nick'], 0, strpos($data['nick'], ' ')));
419 if (!empty($data['baseurl']) && empty($data['gsid'])) {
420 $data['gsid'] = GServer::getID($data['baseurl']);
423 if (empty($data['network'])) {
424 $data['network'] = Protocol::PHANTOM;
427 // Ensure that local connections always are DFRN
428 if (($network == '') && ($data['network'] != Protocol::PHANTOM) && (self::ownHost($data['baseurl'] ?? '') || self::ownHost($data['url']))) {
429 $data['network'] = Protocol::DFRN;
432 if (!isset($data['hide']) && in_array($data['network'], Protocol::FEDERATED)) {
433 $data['hide'] = self::getHideStatus($data['url']);
436 return self::rearrangeData($data);
441 * Fetches the "hide" status from the profile
443 * @param string $url URL of the profile
444 * @return boolean "hide" status
446 private static function getHideStatus(string $url): bool
448 $curlResult = DI::httpClient()->get($url, HttpClientAccept::HTML, [HttpClientOptions::CONTENT_LENGTH => 1000000]);
449 if (!$curlResult->isSuccess()) {
453 // If it isn't a HTML file then exit
454 if (($curlResult->getContentType() != '') && !strstr(strtolower($curlResult->getContentType()), 'html')) {
458 $body = $curlResult->getBody();
463 $doc = new DOMDocument();
464 @$doc->loadHTML($body);
466 $xpath = new DOMXPath($doc);
468 $list = $xpath->query('//meta[@name]');
469 foreach ($list as $node) {
471 if ($node->attributes->length) {
472 foreach ($node->attributes as $attribute) {
473 $meta_tag[$attribute->name] = $attribute->value;
477 if (empty($meta_tag['content'])) {
481 $content = strtolower(trim($meta_tag['content']));
483 switch (strtolower(trim($meta_tag['name']))) {
484 case 'dfrn-global-visibility':
485 if ($content == 'false') {
490 if (strpos($content, 'noindex') !== false) {
501 * Fetch the "subscribe" and add it to the result
503 * @param array $result Result array
504 * @param array $webfinger Webfinger data
506 * @return array result Altered/unaltered result array
508 private static function getSubscribeLink(array $result, array $webfinger): array
510 if (empty($webfinger['links'])) {
514 foreach ($webfinger['links'] as $link) {
515 if (!empty($link['template']) && ($link['rel'] === ActivityNamespace::OSTATUSSUB)) {
516 $result['subscribe'] = $link['template'];
524 * Get webfinger data from a given URI
526 * @param string $uri URI
528 * @return array Webfinger data
529 * @throws HTTPException\InternalServerErrorException
531 public static function getWebfingerArray(string $uri): array
533 $parts = parse_url($uri);
535 if (!empty($parts['scheme']) && !empty($parts['host'])) {
536 $host = $parts['host'];
537 if (!empty($parts['port'])) {
538 $host .= ':' . $parts['port'];
541 $baseurl = $parts['scheme'] . '://' . $host;
546 $path_parts = explode('/', trim($parts['path'] ?? '', '/'));
547 if (!empty($path_parts)) {
548 $nick = ltrim(end($path_parts), '@');
549 $addr = $nick . '@' . $host;
552 $webfinger = self::getWebfinger($parts['scheme'] . '://' . $host . self::WEBFINGER, HttpClientAccept::JRD_JSON, $uri, $addr);
553 if (empty($webfinger)) {
554 $lrdd = self::hostMeta($host);
557 if (empty($webfinger) && empty($lrdd)) {
558 while (empty($lrdd) && empty($webfinger) && (sizeof($path_parts) > 1)) {
559 $host .= '/' . array_shift($path_parts);
560 $baseurl = $parts['scheme'] . '://' . $host;
563 $addr = $nick . '@' . $host;
566 $webfinger = self::getWebfinger($parts['scheme'] . '://' . $host . self::WEBFINGER, HttpClientAccept::JRD_JSON, $uri, $addr);
567 if (empty($webfinger)) {
568 $lrdd = self::hostMeta($host);
572 if (empty($lrdd) && empty($webfinger)) {
576 } elseif (strstr($uri, '@')) {
577 // Remove "acct:" from the URI
578 $uri = str_replace('acct:', '', $uri);
580 $host = substr($uri, strpos($uri, '@') + 1);
581 $nick = substr($uri, 0, strpos($uri, '@'));
584 $webfinger = self::getWebfinger('https://' . $host . self::WEBFINGER, HttpClientAccept::JRD_JSON, $uri, $addr);
585 if (self::$isTimeout) {
589 if (empty($webfinger)) {
590 $webfinger = self::getWebfinger('http://' . $host . self::WEBFINGER, HttpClientAccept::JRD_JSON, $uri, $addr);
591 if (self::$isTimeout) {
595 $baseurl = 'https://' . $host;
598 if (empty($webfinger)) {
599 $lrdd = self::hostMeta($host);
600 if (self::$isTimeout) {
603 $baseurl = self::$baseurl;
605 $baseurl = 'http://' . $host;
608 Logger::info('URI was not detectable', ['uri' => $uri]);
612 if (empty($webfinger)) {
613 foreach ($lrdd as $type => $template) {
618 $webfinger = self::getWebfinger($template, $type, $uri, $addr);
622 if (empty($webfinger)) {
626 if ($webfinger['detected'] == $addr) {
627 $webfinger['nick'] = $nick;
628 $webfinger['addr'] = $addr;
631 $webfinger['baseurl'] = $baseurl;
637 * Perform network request for webfinger data
639 * @param string $template
640 * @param string $type
642 * @param string $addr
644 * @return array webfinger results
646 private static function getWebfinger(string $template, string $type, string $uri, string $addr): array
648 if (Network::isUrlBlocked($template)) {
649 Logger::info('Domain is blocked', ['url' => $template]);
653 // First try the address because this is the primary purpose of webfinger
656 $path = str_replace('{uri}', urlencode('acct:' . $addr), $template);
657 $webfinger = self::webfinger($path, $type);
658 if (self::$isTimeout) {
664 if (empty($webfinger) && $uri != $addr) {
666 $path = str_replace('{uri}', urlencode($uri), $template);
667 $webfinger = self::webfinger($path, $type);
668 if (self::$isTimeout) {
673 if (empty($webfinger)) {
677 return ['webfinger' => $webfinger, 'detected' => $detected];
681 * Fetch information (protocol endpoints and user information) about a given uri
683 * This function is only called by the "uri" function that adds caching and rearranging of data.
685 * @param string $uri Address that should be probed
686 * @param string $network Test for this specific network
687 * @param integer $uid User ID for the probe (only used for mails)
688 * @param array $ap_profile Previously probed AP profile
689 * @return array URI data
690 * @throws HTTPException\InternalServerErrorException
692 private static function detect(string $uri, string $network, int $uid, array $ap_profile): array
696 'network' => $network,
701 Hook::callAll('probe_detect', $hookData);
703 if (isset($hookData['result'])) {
704 return is_array($hookData['result']) ? $hookData['result'] : [];
707 $parts = parse_url($uri);
708 if (empty($parts['scheme']) && empty($parts['host']) && (empty($parts['path']) || strpos($parts['path'], '@') === false)) {
709 Logger::info('URI was not detectable', ['uri' => $uri]);
713 // If the URI starts with "mailto:" then jump directly to the mail detection
714 if (strpos($uri, 'mailto:') !== false) {
715 $uri = str_replace('mailto:', '', $uri);
716 return self::mail($uri, $uid);
719 if ($network == Protocol::MAIL) {
720 return self::mail($uri, $uid);
723 Logger::info('Probing start', ['uri' => $uri]);
725 if (!empty($ap_profile['addr']) && ($ap_profile['addr'] != $uri)) {
726 $data = self::getWebfingerArray($ap_profile['addr']);
730 $data = self::getWebfingerArray($uri);
734 if (!empty($parts['scheme'])) {
735 return self::feed($uri);
736 } elseif (!empty($uid)) {
737 return self::mail($uri, $uid);
743 $webfinger = $data['webfinger'];
744 $nick = $data['nick'] ?? '';
745 $addr = $data['addr'] ?? '';
746 $baseurl = $data['baseurl'] ?? '';
750 if (in_array($network, ['', Protocol::DFRN])) {
751 $result = self::dfrn($webfinger);
753 if ((!$result && ($network == '')) || ($network == Protocol::DIASPORA)) {
754 $result = self::diaspora($webfinger);
756 $result['networks'][Protocol::DIASPORA] = self::diaspora($webfinger);
758 if ((!$result && ($network == '')) || ($network == Protocol::OSTATUS)) {
759 $result = self::ostatus($webfinger);
761 $result['networks'][Protocol::OSTATUS] = self::ostatus($webfinger);
763 if (in_array($network, ['', Protocol::ZOT])) {
764 $result = self::zot($webfinger, $result, $baseurl);
766 if ((!$result && ($network == '')) || ($network == Protocol::PUMPIO)) {
767 $result = self::pumpio($webfinger, $addr, $baseurl);
769 if (empty($result['network']) && empty($ap_profile['network']) || ($network == Protocol::FEED)) {
770 $result = self::feed($uri);
772 // We overwrite the detected nick with our try if the previous routines hadn't detected it.
773 // Additionally, it is overwritten when the nickname doesn't make sense (contains spaces).
774 if ((empty($result['nick']) || (strstr($result['nick'], ' '))) && ($nick != '')) {
775 $result['nick'] = $nick;
778 if (empty($result['addr']) && ($addr != '')) {
779 $result['addr'] = $addr;
783 $result = self::getSubscribeLink($result, $webfinger);
785 if (empty($result['network'])) {
786 $result['network'] = Protocol::PHANTOM;
789 if (empty($result['baseurl']) && !empty($baseurl)) {
790 $result['baseurl'] = $baseurl;
793 if (empty($result['url'])) {
794 $result['url'] = $uri;
797 Logger::info('Probing done', ['uri' => $uri, 'network' => $result['network']]);
803 * Check for Zot contact
805 * @param array $webfinger Webfinger data
806 * @param array $data previously probed data
807 * @param string $baseUrl Base URL
809 * @return array Zot data
810 * @throws HTTPException\InternalServerErrorException
812 private static function zot(array $webfinger, array $data, string $baseurl): array
814 if (!empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
815 foreach ($webfinger['aliases'] as $alias) {
816 if (substr($alias, 0, 5) == 'acct:') {
817 $data['addr'] = substr($alias, 5);
822 if (!empty($webfinger['subject']) && (substr($webfinger['subject'], 0, 5) == 'acct:')) {
823 $data['addr'] = substr($webfinger['subject'], 5);
827 foreach ($webfinger['links'] as $link) {
828 if (($link['rel'] == 'http://purl.org/zot/protocol') && !empty($link['href'])) {
829 $zot_url = $link['href'];
833 if (empty($zot_url) && !empty($data['addr']) && !empty($baseurl)) {
834 $condition = ['nurl' => Strings::normaliseLink($baseurl), 'platform' => ['hubzilla']];
835 if (!DBA::exists('gserver', $condition)) {
838 $zot_url = $baseurl . '/.well-known/zot-info?address=' . $data['addr'];
841 if (empty($zot_url)) {
845 $data = self::pollZot($zot_url, $data);
847 if (!empty($data['url']) && !empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
848 foreach ($webfinger['aliases'] as $alias) {
849 if (!strstr($alias, '@') && Strings::normaliseLink($alias) != Strings::normaliseLink($data['url'])) {
850 $data['alias'] = $alias;
858 public static function pollZot(string $url, array $data): array
860 $curlResult = DI::httpClient()->get($url, HttpClientAccept::JSON);
861 if ($curlResult->isTimeout()) {
864 $content = $curlResult->getBody();
869 $json = json_decode($content, true);
870 if (!is_array($json)) {
874 if (empty($data['network'])) {
875 if (!empty($json['protocols']) && in_array('zot', $json['protocols'])) {
876 $data['network'] = Protocol::ZOT;
877 } elseif (!isset($json['protocols'])) {
878 $data['network'] = Protocol::ZOT;
882 if (!empty($json['guid']) && empty($data['guid'])) {
883 $data['guid'] = $json['guid'];
885 if (!empty($json['key']) && empty($data['pubkey'])) {
886 $data['pubkey'] = $json['key'];
888 if (!empty($json['name'])) {
889 $data['name'] = $json['name'];
891 if (!empty($json['photo'])) {
892 $data['photo'] = $json['photo'];
893 if (!empty($json['photo_updated'])) {
894 $data['photo'] .= '?rev=' . urlencode($json['photo_updated']);
897 if (!empty($json['address'])) {
898 $data['addr'] = $json['address'];
900 if (!empty($json['url'])) {
901 $data['url'] = $json['url'];
903 if (!empty($json['connections_url'])) {
904 $data['poco'] = $json['connections_url'];
906 if (isset($json['searchable'])) {
907 $data['hide'] = !$json['searchable'];
909 if (!empty($json['public_forum'])) {
910 $data['community'] = $json['public_forum'];
911 $data['account-type'] = User::PAGE_FLAGS_COMMUNITY;
914 if (!empty($json['profile'])) {
915 $profile = $json['profile'];
916 if (!empty($profile['description'])) {
917 $data['about'] = $profile['description'];
919 if (!empty($profile['keywords'])) {
920 $keywords = implode(', ', $profile['keywords']);
921 if (!empty($keywords)) {
922 $data['keywords'] = $keywords;
927 if (!empty($profile['region'])) {
928 $loc['region'] = $profile['region'];
930 if (!empty($profile['country'])) {
931 $loc['country-name'] = $profile['country'];
933 $location = Profile::formatLocation($loc);
934 if (!empty($location)) {
935 $data['location'] = $location;
943 * Perform a webfinger request.
945 * For details see RFC 7033: <https://tools.ietf.org/html/rfc7033>
947 * @param string $url Address that should be probed
948 * @param string $type type
950 * @return array webfinger data
951 * @throws HTTPException\InternalServerErrorException
953 public static function webfinger(string $url, string $type): array
955 $xrd_timeout = DI::config()->get('system', 'xrd_timeout', 20);
957 $curlResult = DI::httpClient()->get($url, $type, [HttpClientOptions::TIMEOUT => $xrd_timeout]);
958 if ($curlResult->isTimeout()) {
959 self::$isTimeout = true;
962 $data = $curlResult->getBody();
964 $webfinger = json_decode($data, true);
965 if (!empty($webfinger)) {
966 if (!isset($webfinger['links'])) {
967 Logger::info('No json webfinger links', ['url' => $url]);
973 // If it is not JSON, maybe it is XML
974 $xrd = XML::parseString($data, true);
975 if (!is_object($xrd)) {
976 Logger::info('No webfinger data retrievable', ['url' => $url]);
980 $xrd_arr = XML::elementToArray($xrd);
981 if (!isset($xrd_arr['xrd']['link'])) {
982 Logger::info('No XML webfinger links', ['url' => $url]);
988 if (!empty($xrd_arr['xrd']['subject'])) {
989 $webfinger['subject'] = $xrd_arr['xrd']['subject'];
992 if (!empty($xrd_arr['xrd']['alias'])) {
993 $webfinger['aliases'] = $xrd_arr['xrd']['alias'];
996 $webfinger['links'] = [];
998 foreach ($xrd_arr['xrd']['link'] as $value => $data) {
999 if (!empty($data['@attributes'])) {
1000 $attributes = $data['@attributes'];
1001 } elseif ($value == '@attributes') {
1002 $attributes = $data;
1007 $webfinger['links'][] = $attributes;
1013 * Poll the Friendica specific noscrape page.
1015 * "noscrape" is a faster alternative to fetch the data from the hcard.
1016 * This functionality was originally created for the directory.
1018 * @param string $noscrape_url Link to the noscrape page
1019 * @param array $data The already fetched data
1021 * @return array noscrape data
1022 * @throws HTTPException\InternalServerErrorException
1024 private static function pollNoscrape(string $noscrape_url, array $data): array
1026 $curlResult = DI::httpClient()->get($noscrape_url, HttpClientAccept::JSON);
1027 if ($curlResult->isTimeout()) {
1028 self::$isTimeout = true;
1031 $content = $curlResult->getBody();
1033 Logger::info('Empty body', ['url' => $noscrape_url]);
1037 $json = json_decode($content, true);
1038 if (!is_array($json)) {
1039 Logger::info('No json data', ['url' => $noscrape_url]);
1043 if (!empty($json['fn'])) {
1044 $data['name'] = $json['fn'];
1047 if (!empty($json['addr'])) {
1048 $data['addr'] = $json['addr'];
1051 if (!empty($json['nick'])) {
1052 $data['nick'] = $json['nick'];
1055 if (!empty($json['guid'])) {
1056 $data['guid'] = $json['guid'];
1059 if (!empty($json['comm'])) {
1060 $data['community'] = $json['comm'];
1063 if (!empty($json['tags'])) {
1064 $keywords = implode(', ', $json['tags']);
1065 if ($keywords != '') {
1066 $data['keywords'] = $keywords;
1070 $location = Profile::formatLocation($json);
1072 $data['location'] = $location;
1075 if (!empty($json['about'])) {
1076 $data['about'] = $json['about'];
1079 if (!empty($json['xmpp'])) {
1080 $data['xmpp'] = $json['xmpp'];
1083 if (!empty($json['matrix'])) {
1084 $data['matrix'] = $json['matrix'];
1087 if (!empty($json['key'])) {
1088 $data['pubkey'] = $json['key'];
1091 if (!empty($json['photo'])) {
1092 $data['photo'] = $json['photo'];
1095 if (!empty($json['dfrn-request'])) {
1096 $data['request'] = $json['dfrn-request'];
1099 if (!empty($json['dfrn-confirm'])) {
1100 $data['confirm'] = $json['dfrn-confirm'];
1103 if (!empty($json['dfrn-notify'])) {
1104 $data['notify'] = $json['dfrn-notify'];
1107 if (!empty($json['dfrn-poll'])) {
1108 $data['poll'] = $json['dfrn-poll'];
1111 if (isset($json['hide'])) {
1112 $data['hide'] = (bool)$json['hide'];
1114 $data['hide'] = false;
1121 * Check for valid DFRN data
1123 * @param array $data DFRN data
1125 * @return int Number of errors
1127 public static function validDfrn(array $data): int
1130 if (!isset($data['key'])) {
1133 if (!isset($data['dfrn-request'])) {
1136 if (!isset($data['dfrn-confirm'])) {
1139 if (!isset($data['dfrn-notify'])) {
1142 if (!isset($data['dfrn-poll'])) {
1149 * Fetch data from a DFRN profile page and via "noscrape"
1151 * @param string $profile_link Link to the profile page
1152 * @return array profile data
1153 * @throws HTTPException\InternalServerErrorException
1154 * @throws \ImagickException
1156 public static function profile(string $profile_link): array
1160 Logger::info('Check profile', ['link' => $profile_link]);
1162 // Fetch data via noscrape - this is faster
1163 $noscrape_url = str_replace(['/hcard/', '/profile/'], '/noscrape/', $profile_link);
1164 $data = self::pollNoscrape($noscrape_url, $data);
1166 if (!isset($data['notify'])
1167 || !isset($data['confirm'])
1168 || !isset($data['request'])
1169 || !isset($data['poll'])
1170 || !isset($data['name'])
1171 || !isset($data['photo'])
1173 $data = self::pollHcard($profile_link, $data, true);
1178 if (empty($data['addr']) || empty($data['nick'])) {
1179 $probe_data = self::uri($profile_link);
1180 $data['addr'] = ($data['addr'] ?? '') ?: $probe_data['addr'];
1181 $data['nick'] = ($data['nick'] ?? '') ?: $probe_data['nick'];
1184 $prof_data['addr'] = $data['addr'];
1185 $prof_data['nick'] = $data['nick'];
1186 $prof_data['dfrn-request'] = $data['request'] ?? null;
1187 $prof_data['dfrn-confirm'] = $data['confirm'] ?? null;
1188 $prof_data['dfrn-notify'] = $data['notify'] ?? null;
1189 $prof_data['dfrn-poll'] = $data['poll'] ?? null;
1190 $prof_data['photo'] = $data['photo'] ?? null;
1191 $prof_data['fn'] = $data['name'] ?? null;
1192 $prof_data['key'] = $data['pubkey'] ?? null;
1194 Logger::debug('Result', ['link' => $profile_link, 'data' => $prof_data]);
1200 * Check for DFRN contact
1202 * @param array $webfinger Webfinger data
1203 * @return array DFRN data
1204 * @throws HTTPException\InternalServerErrorException
1206 private static function dfrn(array $webfinger): array
1210 // The array is reversed to take into account the order of preference for same-rel links
1211 // See: https://tools.ietf.org/html/rfc7033#section-4.4.4
1212 foreach (array_reverse($webfinger['links']) as $link) {
1213 if (($link['rel'] == ActivityNamespace::DFRN) && !empty($link['href'])) {
1214 $data['network'] = Protocol::DFRN;
1215 } elseif (($link['rel'] == ActivityNamespace::FEED) && !empty($link['href'])) {
1216 $data['poll'] = $link['href'];
1217 } elseif (($link['rel'] == 'http://webfinger.net/rel/profile-page') && (($link['type'] ?? '') == 'text/html') && !empty($link['href'])) {
1218 $data['url'] = $link['href'];
1219 } elseif (($link['rel'] == 'http://microformats.org/profile/hcard') && !empty($link['href'])) {
1220 $hcard_url = $link['href'];
1221 } elseif (($link['rel'] == ActivityNamespace::POCO) && !empty($link['href'])) {
1222 $data['poco'] = $link['href'];
1223 } elseif (($link['rel'] == 'http://webfinger.net/rel/avatar') && !empty($link['href'])) {
1224 $data['photo'] = $link['href'];
1225 } elseif (($link['rel'] == 'http://joindiaspora.com/seed_location') && !empty($link['href'])) {
1226 $data['baseurl'] = trim($link['href'], '/');
1227 } elseif (($link['rel'] == 'http://joindiaspora.com/guid') && !empty($link['href'])) {
1228 $data['guid'] = $link['href'];
1229 } elseif (($link['rel'] == 'diaspora-public-key') && !empty($link['href'])) {
1230 $data['pubkey'] = base64_decode($link['href']);
1232 if (strstr($data['pubkey'], 'RSA ')) {
1233 $data['pubkey'] = Crypto::rsaToPem($data['pubkey']);
1238 if (!empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
1239 foreach ($webfinger['aliases'] as $alias) {
1240 if (empty($data['url']) && !strstr($alias, '@')) {
1241 $data['url'] = $alias;
1242 } elseif (!strstr($alias, '@') && Strings::normaliseLink($alias) != Strings::normaliseLink($data['url'])) {
1243 $data['alias'] = $alias;
1244 } elseif (substr($alias, 0, 5) == 'acct:') {
1245 $data['addr'] = substr($alias, 5);
1250 if (!empty($webfinger['subject']) && (substr($webfinger['subject'], 0, 5) == 'acct:')) {
1251 $data['addr'] = substr($webfinger['subject'], 5);
1254 if (!isset($data['network']) || ($hcard_url == '')) {
1258 // Fetch data via noscrape - this is faster
1259 $noscrape_url = str_replace('/hcard/', '/noscrape/', $hcard_url);
1260 $data = self::pollNoscrape($noscrape_url, $data);
1262 if (isset($data['notify'])
1263 && isset($data['confirm'])
1264 && isset($data['request'])
1265 && isset($data['poll'])
1266 && isset($data['name'])
1267 && isset($data['photo'])
1272 $data = self::pollHcard($hcard_url, $data, true);
1278 * Poll the hcard page (Diaspora and Friendica specific)
1280 * @param string $hcard_url Link to the hcard page
1281 * @param array $data The already fetched data
1282 * @param boolean $dfrn Poll DFRN specific data
1283 * @return array hcard data
1284 * @throws HTTPException\InternalServerErrorException
1286 private static function pollHcard(string $hcard_url, array $data, bool $dfrn = false): array
1288 $curlResult = DI::httpClient()->get($hcard_url, HttpClientAccept::HTML);
1289 if ($curlResult->isTimeout()) {
1290 self::$isTimeout = true;
1293 $content = $curlResult->getBody();
1294 if (empty($content)) {
1298 $doc = new DOMDocument();
1299 if (!@$doc->loadHTML($content)) {
1303 $xpath = new DomXPath($doc);
1305 $vcards = $xpath->query("//div[contains(concat(' ', @class, ' '), ' vcard ')]");
1306 if (!is_object($vcards)) {
1310 if (!isset($data['baseurl'])) {
1311 $data['baseurl'] = '';
1314 if ($vcards->length > 0) {
1315 $vcard = $vcards->item(0);
1317 // We have to discard the guid from the hcard in favour of the guid from lrdd
1318 // Reason: Hubzilla doesn't use the value "uid" in the hcard like Diaspora does.
1319 $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' uid ')]", $vcard); // */
1320 if (($search->length > 0) && empty($data['guid'])) {
1321 $data['guid'] = $search->item(0)->nodeValue;
1324 $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' nickname ')]", $vcard); // */
1325 if ($search->length > 0) {
1326 $data['nick'] = $search->item(0)->nodeValue;
1329 $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' fn ')]", $vcard); // */
1330 if ($search->length > 0) {
1331 $data['name'] = $search->item(0)->nodeValue;
1334 $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' given_name ')]", $vcard); // */
1335 if ($search->length > 0) {
1336 $data["given_name"] = $search->item(0)->nodeValue;
1339 $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' family_name ')]", $vcard); // */
1340 if ($search->length > 0) {
1341 $data["family_name"] = $search->item(0)->nodeValue;
1344 $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' searchable ')]", $vcard); // */
1345 if ($search->length > 0) {
1346 $data['hide'] = (strtolower($search->item(0)->nodeValue) != 'true');
1349 $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' key ')]", $vcard); // */
1350 if ($search->length > 0) {
1351 $data['pubkey'] = $search->item(0)->nodeValue;
1352 if (strstr($data['pubkey'], 'RSA ')) {
1353 $data['pubkey'] = Crypto::rsaToPem($data['pubkey']);
1357 $search = $xpath->query("//*[@id='pod_location']", $vcard); // */
1358 if ($search->length > 0) {
1359 $data['baseurl'] = trim($search->item(0)->nodeValue, '/');
1364 if (!empty($vcard)) {
1365 $photos = $xpath->query("//*[contains(concat(' ', @class, ' '), ' photo ') or contains(concat(' ', @class, ' '), ' avatar ')]", $vcard); // */
1366 foreach ($photos as $photo) {
1368 foreach ($photo->attributes as $attribute) {
1369 $attr[$attribute->name] = trim($attribute->value);
1372 if (isset($attr['src']) && isset($attr['width'])) {
1373 $avatars[$attr['width']] = self::fixAvatar($attr['src'], $data['baseurl']);
1376 // We don't have a width. So we just take everything that we got.
1377 // This is a Hubzilla workaround which doesn't send a width.
1378 if (!$avatars && !empty($attr['src'])) {
1379 $avatars[] = self::fixAvatar($attr['src'], $data['baseurl']);
1386 $data['photo'] = array_pop($avatars);
1388 $data['photo_medium'] = array_pop($avatars);
1392 $data['photo_small'] = array_pop($avatars);
1397 // Poll DFRN specific data
1398 $search = $xpath->query("//link[contains(concat(' ', @rel), ' dfrn-')]");
1399 if ($search->length > 0) {
1400 foreach ($search as $link) {
1401 //$data['request'] = $search->item(0)->nodeValue;
1403 foreach ($link->attributes as $attribute) {
1404 $attr[$attribute->name] = trim($attribute->value);
1407 $data[substr($attr['rel'], 5)] = $attr['href'];
1411 // Older Friendica versions had used the "uid" field differently than newer versions
1412 if (!empty($data['nick']) && !empty($data['guid']) && ($data['nick'] == $data['guid'])) {
1413 unset($data['guid']);
1421 * Check for Diaspora contact
1423 * @param array $webfinger Webfinger data
1425 * @return array Diaspora data
1426 * @throws HTTPException\InternalServerErrorException
1428 private static function diaspora(array $webfinger): array
1433 // The array is reversed to take into account the order of preference for same-rel links
1434 // See: https://tools.ietf.org/html/rfc7033#section-4.4.4
1435 foreach (array_reverse($webfinger['links']) as $link) {
1436 if (($link['rel'] == 'http://microformats.org/profile/hcard') && !empty($link['href'])) {
1437 $hcard_url = $link['href'];
1438 } elseif (($link['rel'] == 'http://joindiaspora.com/seed_location') && !empty($link['href'])) {
1439 $data['baseurl'] = trim($link['href'], '/');
1440 } elseif (($link['rel'] == 'http://joindiaspora.com/guid') && !empty($link['href'])) {
1441 $data['guid'] = $link['href'];
1442 } elseif (($link['rel'] == 'http://webfinger.net/rel/profile-page') && (($link['type'] ?? '') == 'text/html') && !empty($link['href'])) {
1443 $data['url'] = $link['href'];
1444 } elseif (($link['rel'] == 'http://webfinger.net/rel/profile-page') && empty($link['type']) && !empty($link['href'])) {
1445 $profile_url = $link['href'];
1446 } elseif (($link['rel'] == ActivityNamespace::FEED) && !empty($link['href'])) {
1447 $data['poll'] = $link['href'];
1448 } elseif (($link['rel'] == ActivityNamespace::POCO) && !empty($link['href'])) {
1449 $data['poco'] = $link['href'];
1450 } elseif (($link['rel'] == 'salmon') && !empty($link['href'])) {
1451 $data['notify'] = $link['href'];
1452 } elseif (($link['rel'] == 'diaspora-public-key') && !empty($link['href'])) {
1453 $data['pubkey'] = base64_decode($link['href']);
1455 if (strstr($data['pubkey'], 'RSA ')) {
1456 $data['pubkey'] = Crypto::rsaToPem($data['pubkey']);
1461 if (empty($data['url']) && !empty($profile_url)) {
1462 $data['url'] = $profile_url;
1465 if (empty($data['url']) || empty($hcard_url)) {
1469 if (!empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
1470 foreach ($webfinger['aliases'] as $alias) {
1471 if (Strings::normaliseLink($alias) != Strings::normaliseLink($data['url']) && ! strstr($alias, '@')) {
1472 $data['alias'] = $alias;
1473 } elseif (substr($alias, 0, 5) == 'acct:') {
1474 $data['addr'] = substr($alias, 5);
1479 if (!empty($webfinger['subject']) && (substr($webfinger['subject'], 0, 5) == 'acct:')) {
1480 $data['addr'] = substr($webfinger['subject'], 5);
1483 // Fetch further information from the hcard
1484 $data = self::pollHcard($hcard_url, $data);
1490 if (!empty($data['url'])
1491 && !empty($data['guid'])
1492 && !empty($data['baseurl'])
1493 && !empty($data['pubkey'])
1494 && !empty($hcard_url)
1496 $data['network'] = Protocol::DIASPORA;
1497 $data['manually-approve'] = false;
1499 // The Diaspora handle must always be lowercase
1500 if (!empty($data['addr'])) {
1501 $data['addr'] = strtolower($data['addr']);
1504 // We have to overwrite the detected value for "notify" since Hubzilla doesn't send it
1505 $data['notify'] = $data['baseurl'] . '/receive/users/' . $data['guid'];
1506 $data['batch'] = $data['baseurl'] . '/receive/public';
1515 * Check for OStatus contact
1517 * @param array $webfinger Webfinger data
1518 * @param bool $short Short detection mode
1520 * @return array|bool OStatus data or "false" on error or "true" on short mode
1521 * @throws HTTPException\InternalServerErrorException
1523 private static function ostatus(array $webfinger, bool $short = false)
1527 if (!empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
1528 foreach ($webfinger['aliases'] as $alias) {
1529 if (strstr($alias, '@') && !strstr(Strings::normaliseLink($alias), 'http://')) {
1530 $data['addr'] = str_replace('acct:', '', $alias);
1535 if (!empty($webfinger['subject']) && strstr($webfinger['subject'], '@')
1536 && !strstr(Strings::normaliseLink($webfinger['subject']), 'http://')
1538 $data['addr'] = str_replace('acct:', '', $webfinger['subject']);
1541 if (!empty($webfinger['links'])) {
1542 // The array is reversed to take into account the order of preference for same-rel links
1543 // See: https://tools.ietf.org/html/rfc7033#section-4.4.4
1544 foreach (array_reverse($webfinger['links']) as $link) {
1545 if (($link['rel'] == 'http://webfinger.net/rel/profile-page')
1546 && (($link['type'] ?? '') == 'text/html')
1547 && ($link['href'] != '')
1549 $data['url'] = $data['alias'] = $link['href'];
1550 } elseif (($link['rel'] == 'salmon') && !empty($link['href'])) {
1551 $data['notify'] = $link['href'];
1552 } elseif (($link['rel'] == ActivityNamespace::FEED) && !empty($link['href'])) {
1553 $data['poll'] = $link['href'];
1554 } elseif (($link['rel'] == 'magic-public-key') && !empty($link['href'])) {
1555 $pubkey = $link['href'];
1557 if (substr($pubkey, 0, 5) === 'data:') {
1558 if (strstr($pubkey, ',')) {
1559 $pubkey = substr($pubkey, strpos($pubkey, ',') + 1);
1561 $pubkey = substr($pubkey, 5);
1563 } elseif (Strings::normaliseLink($pubkey) == 'http://') {
1564 $curlResult = DI::httpClient()->get($pubkey, HttpClientAccept::MAGIC_KEY);
1565 if ($curlResult->isTimeout()) {
1566 self::$isTimeout = true;
1567 return $short ? false : [];
1569 Logger::debug('Fetched public key', ['Content-Type' => $curlResult->getHeader('Content-Type'), 'url' => $pubkey]);
1570 $pubkey = $curlResult->getBody();
1574 $data['pubkey'] = Salmon::magicKeyToPem($pubkey);
1575 } catch (\Throwable $e) {
1582 if (isset($data['notify']) && isset($data['pubkey'])
1583 && isset($data['poll'])
1584 && isset($data['url'])
1586 $data['network'] = Protocol::OSTATUS;
1587 $data['manually-approve'] = false;
1589 return $short ? false : [];
1596 // Fetch all additional data from the feed
1597 $curlResult = DI::httpClient()->get($data['poll'], HttpClientAccept::FEED_XML);
1598 if ($curlResult->isTimeout()) {
1599 self::$isTimeout = true;
1602 $feed = $curlResult->getBody();
1603 $feed_data = Feed::import($feed);
1608 if (!empty($feed_data['header']['author-name'])) {
1609 $data['name'] = $feed_data['header']['author-name'];
1611 if (!empty($feed_data['header']['author-nick'])) {
1612 $data['nick'] = $feed_data['header']['author-nick'];
1614 if (!empty($feed_data['header']['author-avatar'])) {
1615 $data['photo'] = self::fixAvatar($feed_data['header']['author-avatar'], $data['url']);
1617 if (!empty($feed_data['header']['author-id'])) {
1618 $data['alias'] = $feed_data['header']['author-id'];
1620 if (!empty($feed_data['header']['author-location'])) {
1621 $data['location'] = $feed_data['header']['author-location'];
1623 if (!empty($feed_data['header']['author-about'])) {
1624 $data['about'] = $feed_data['header']['author-about'];
1626 // OStatus has serious issues when the url doesn't fit (ssl vs. non ssl)
1627 // So we take the value that we just fetched, although the other one worked as well
1628 if (!empty($feed_data['header']['author-link'])) {
1629 $data['url'] = $feed_data['header']['author-link'];
1632 if ($data['url'] == $data['alias']) {
1633 $data['alias'] = '';
1636 /// @todo Fetch location and "about" from the feed as well
1641 * Fetch data from a pump.io profile page
1643 * @param string $profile_link Link to the profile page
1645 * @return array Profile data
1647 private static function pumpioProfileData(string $profile_link, string $baseurl): array
1649 $curlResult = DI::httpClient()->get($profile_link, HttpClientAccept::HTML);
1650 if (!$curlResult->isSuccess() || empty($curlResult->getBody())) {
1654 $doc = new DOMDocument();
1655 if (!@$doc->loadHTML($curlResult->getBody())) {
1659 $xpath = new DomXPath($doc);
1662 $data['name'] = $xpath->query("//span[contains(@class, 'p-name')]")->item(0)->nodeValue;
1664 if ($data['name'] == '') {
1665 // This is ugly - but pump.io doesn't seem to know a better way for it
1666 $data['name'] = trim($xpath->query("//h1[@class='media-header']")->item(0)->nodeValue);
1667 $pos = strpos($data['name'], chr(10));
1669 $data['name'] = trim(substr($data['name'], 0, $pos));
1673 $data['location'] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'p-locality')]");
1675 if ($data['location'] == '') {
1676 $data['location'] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'location')]");
1679 $data['about'] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'p-note')]");
1681 if ($data['about'] == '') {
1682 $data['about'] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'summary')]");
1685 $avatar = $xpath->query("//img[contains(@class, 'u-photo')]")->item(0);
1687 $avatar = $xpath->query("//img[@class='img-rounded media-object']")->item(0);
1690 foreach ($avatar->attributes as $attribute) {
1691 if (($attribute->name == 'src') && !empty($attribute->value)) {
1692 $data['photo'] = Network::addBasePath($attribute->value, $baseurl);
1701 * Check for pump.io contact
1703 * @param array $webfinger Webfinger data
1704 * @param string $addr
1706 * @return array pump.io data
1708 private static function pumpio(array $webfinger, string $addr, string $baseurl): array
1711 // The array is reversed to take into account the order of preference for same-rel links
1712 // See: https://tools.ietf.org/html/rfc7033#section-4.4.4
1713 foreach (array_reverse($webfinger['links']) as $link) {
1714 if (($link['rel'] == 'http://webfinger.net/rel/profile-page')
1715 && (($link['type'] ?? '') == 'text/html')
1716 && ($link['href'] != '')
1718 $data['url'] = $link['href'];
1719 } elseif (($link['rel'] == 'activity-inbox') && ($link['href'] != '')) {
1720 $data['notify'] = $link['href'];
1721 } elseif (($link['rel'] == 'activity-outbox') && ($link['href'] != '')) {
1722 $data['poll'] = $link['href'];
1723 } elseif (($link['rel'] == 'dialback') && ($link['href'] != '')) {
1724 $data['dialback'] = $link['href'];
1727 if (isset($data['poll']) && isset($data['notify'])
1728 && isset($data['dialback'])
1729 && isset($data['url'])
1731 // by now we use these fields only for the network type detection
1732 // So we unset all data that isn't used at the moment
1733 unset($data['dialback']);
1735 $data['network'] = Protocol::PUMPIO;
1740 $profile_data = self::pumpioProfileData($data['url'], $baseurl);
1742 if (!$profile_data) {
1746 $data = array_merge($data, $profile_data);
1748 if (($addr != '') && ($data['name'] != '')) {
1749 $name = trim(str_replace($addr, '', $data['name']));
1751 $data['name'] = $name;
1759 * Checks HTML page for RSS feed link
1761 * @param string $url Page link
1762 * @param string $body Page body string
1764 * @return string|false Feed link or false if body was invalid HTML document
1766 public static function getFeedLink(string $url, string $body)
1772 $doc = new DOMDocument();
1773 if (!@$doc->loadHTML($body)) {
1777 $xpath = new DOMXPath($doc);
1779 $feedUrl = $xpath->evaluate('string(/html/head/link[@type="application/rss+xml" and @rel="alternate"]/@href)');
1780 $feedUrl = $feedUrl ?: $xpath->evaluate('string(/html/head/link[@type="application/atom+xml" and @rel="alternate"]/@href)');
1782 $feedUrl = $feedUrl ? self::ensureAbsoluteLinkFromHTMLDoc($feedUrl, $url, $xpath) : '';
1788 * Return an absolute URL in the context of a HTML document retrieved from the provided URL.
1790 * Loosely based on RFC 1808
1792 * @see https://tools.ietf.org/html/rfc1808
1794 * @param string $href The potential relative href found in the HTML document
1795 * @param string $base The HTML document URL
1796 * @param DOMXPath $xpath The HTML document XPath
1798 * @return string Absolute URL
1800 private static function ensureAbsoluteLinkFromHTMLDoc(string $href, string $base, DOMXPath $xpath): string
1802 if (filter_var($href, FILTER_VALIDATE_URL)) {
1806 $base = $xpath->evaluate('string(/html/head/base/@href)') ?: $base;
1808 $baseParts = parse_url($base);
1809 if (empty($baseParts['host'])) {
1813 // Naked domain case (scheme://basehost)
1814 $path = $baseParts['path'] ?? '/';
1816 // Remove the filename part of the path if it exists (/base/path/file)
1817 $path = implode('/', array_slice(explode('/', $path), 0, -1));
1819 $hrefParts = parse_url($href);
1821 if (!empty($hrefParts['path'])) {
1822 // Root path case (/path) including relative scheme case (//host/path)
1823 if ($hrefParts['path'] && $hrefParts['path'][0] == '/') {
1824 $path = $hrefParts['path'];
1826 $path = $path . '/' . $hrefParts['path'];
1828 // Resolve arbitrary relative path
1829 // Lifted from https://www.php.net/manual/en/function.realpath.php#84012
1830 $parts = array_filter(explode('/', $path), 'strlen');
1832 foreach ($parts as $part) {
1833 if ('.' == $part) continue;
1834 if ('..' == $part) {
1835 array_pop($absolutes);
1837 $absolutes[] = $part;
1841 $path = '/' . implode('/', $absolutes);
1845 // Relative scheme case (//host/path)
1846 $baseParts['host'] = $hrefParts['host'] ?? $baseParts['host'];
1847 $baseParts['path'] = $path;
1848 unset($baseParts['query']);
1849 unset($baseParts['fragment']);
1851 return Network::unparseURL($baseParts);
1855 * Check for feed contact
1857 * @param string $url Profile link
1858 * @param boolean $probe Do a probe if the page contains a feed link
1860 * @return array feed data
1861 * @throws HTTPException\InternalServerErrorException
1863 private static function feed(string $url, bool $probe = true): array
1866 $curlResult = DI::httpClient()->get($url, HttpClientAccept::FEED_XML);
1867 } catch(\Throwable $e) {
1868 DI::logger()->info('Error requesting feed URL', ['url' => $url, 'exception' => $e]);
1872 if ($curlResult->isTimeout()) {
1873 self::$isTimeout = true;
1877 $feed = $curlResult->getBody();
1878 $feed_data = Feed::import($feed);
1885 $feed_url = self::getFeedLink($url, $feed);
1891 return self::feed($feed_url, false);
1894 if (!empty($feed_data['header']['author-name'])) {
1895 $data['name'] = $feed_data['header']['author-name'];
1898 if (!empty($feed_data['header']['author-nick'])) {
1899 $data['nick'] = $feed_data['header']['author-nick'];
1902 if (!empty($feed_data['header']['author-avatar'])) {
1903 $data['photo'] = $feed_data['header']['author-avatar'];
1906 if (!empty($feed_data['header']['author-id'])) {
1907 $data['alias'] = $feed_data['header']['author-id'];
1910 $data['url'] = $url;
1911 $data['poll'] = $url;
1913 $data['network'] = Protocol::FEED;
1919 * Check for mail contact
1921 * @param string $uri Profile link
1922 * @param integer $uid User ID
1924 * @return array mail data
1925 * @throws \Exception
1927 private static function mail(string $uri, int $uid): array
1929 if (!Network::isEmailDomainValid($uri)) {
1937 $user = DBA::selectFirst('user', ['prvkey'], ['uid' => $uid]);
1939 $condition = ["`uid` = ? AND `server` != ''", $uid];
1940 $fields = ['pass', 'user', 'server', 'port', 'ssltype', 'mailbox'];
1941 $mailacct = DBA::selectFirst('mailacct', $fields, $condition);
1943 if (!DBA::isResult($user) || !DBA::isResult($mailacct)) {
1947 $mailbox = Email::constructMailboxName($mailacct);
1949 openssl_private_decrypt(hex2bin($mailacct['pass']), $password, $user['prvkey']);
1950 $mbox = Email::connect($mailbox, $mailacct['user'], $password);
1955 $msgs = Email::poll($mbox, $uri);
1956 Logger::info('Messages found', ['uri' => $uri, 'count' => count($msgs)]);
1958 if (!count($msgs)) {
1962 $phost = substr($uri, strpos($uri, '@') + 1);
1966 'network' => Protocol::MAIL,
1967 'name' => substr($uri, 0, strpos($uri, '@')),
1968 'photo' => Network::lookupAvatarByEmail($uri),
1969 'url' => 'mailto:' . $uri,
1970 'notify' => 'smtp ' . Strings::getRandomHex(),
1971 'poll' => 'email ' . Strings::getRandomHex(),
1974 $data['nick'] = $data['name'];
1976 $x = Email::messageMeta($mbox, $msgs[0]);
1978 if (stristr($x[0]->from, $uri)) {
1979 $adr = imap_rfc822_parse_adrlist($x[0]->from, '');
1980 } elseif (stristr($x[0]->to, $uri)) {
1981 $adr = imap_rfc822_parse_adrlist($x[0]->to, '');
1985 foreach ($adr as $feadr) {
1986 if ((strcasecmp($feadr->mailbox, $data['name']) == 0)
1987 &&(strcasecmp($feadr->host, $phost) == 0)
1988 && (strlen($feadr->personal))
1990 $personal = imap_mime_header_decode($feadr->personal);
1992 foreach ($personal as $perspart) {
1993 if ($perspart->charset != 'default') {
1994 $data['name'] .= iconv($perspart->charset, 'UTF-8//IGNORE', $perspart->text);
1996 $data['name'] .= $perspart->text;
2003 if (!empty($mbox)) {
2011 * Mix two paths together to possibly fix missing parts
2013 * @param string $avatar Path to the avatar
2014 * @param string $base Another path that is hopefully complete
2016 * @return string fixed avatar path
2017 * @throws \Exception
2019 public static function fixAvatar(string $avatar, string $base): string
2021 $base_parts = parse_url($base);
2023 // Remove all parts that could create a problem
2024 unset($base_parts['path']);
2025 unset($base_parts['query']);
2026 unset($base_parts['fragment']);
2028 $avatar_parts = parse_url($avatar);
2031 $parts = array_merge($base_parts, $avatar_parts);
2033 // And put them together again
2034 $scheme = isset($parts['scheme']) ? $parts['scheme'] . '://' : '';
2035 $host = isset($parts['host']) ? $parts['host'] : '';
2036 $port = isset($parts['port']) ? ':' . $parts['port'] : '';
2037 $path = isset($parts['path']) ? $parts['path'] : '';
2038 $query = isset($parts['query']) ? '?' . $parts['query'] : '';
2039 $fragment = isset($parts['fragment']) ? '#' . $parts['fragment'] : '';
2041 $fixed = $scheme.$host.$port.$path.$query.$fragment;
2043 Logger::debug('Avatar fixed', ['base' => $base, 'avatar' => $avatar, 'fixed' => $fixed]);
2049 * Fetch the last date that the contact had posted something (publically)
2051 * @param array $data probing result
2053 * @return string last activity
2055 public static function getLastUpdate(array $data): string
2057 $uid = User::getIdForURL($data['url']);
2059 $contact = Contact::selectFirst(['url', 'last-item'], ['self' => true, 'uid' => $uid]);
2060 if (!empty($contact['last-item'])) {
2061 return $contact['last-item'];
2065 if ($lastUpdate = self::updateFromNoScrape($data)) {
2069 if (!empty($data['outbox'])) {
2070 return self::updateFromOutbox($data['outbox'], $data);
2071 } elseif (!empty($data['poll']) && ($data['network'] == Protocol::ACTIVITYPUB)) {
2072 return self::updateFromOutbox($data['poll'], $data);
2073 } elseif (!empty($data['poll'])) {
2074 return self::updateFromFeed($data);
2081 * Fetch the last activity date from the "noscrape" endpoint
2083 * @param array $data Probing result
2085 * @return string last activity or true if update was successful or the server was unreachable
2087 private static function updateFromNoScrape(array $data): string
2089 if (empty($data['baseurl'])) {
2093 // Check the 'noscrape' endpoint when it is a Friendica server
2094 $gserver = DBA::selectFirst('gserver', ['noscrape'], ["`nurl` = ? AND `noscrape` != ''",
2095 Strings::normaliseLink($data['baseurl'])]);
2096 if (!DBA::isResult($gserver)) {
2100 $curlResult = DI::httpClient()->get($gserver['noscrape'] . '/' . $data['nick'], HttpClientAccept::JSON);
2102 if ($curlResult->isSuccess() && !empty($curlResult->getBody())) {
2103 $noscrape = json_decode($curlResult->getBody(), true);
2104 if (!empty($noscrape) && !empty($noscrape['updated'])) {
2105 return DateTimeFormat::utc($noscrape['updated'], DateTimeFormat::MYSQL);
2113 * Fetch the last activity date from an ActivityPub Outbox
2115 * @param string $feed
2116 * @param array $data Probing result
2118 * @return string last activity
2119 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2121 private static function updateFromOutbox(string $feed, array $data): string
2123 $outbox = ActivityPub::fetchContent($feed);
2124 if (empty($outbox)) {
2128 if (!empty($outbox['orderedItems'])) {
2129 $items = $outbox['orderedItems'];
2130 } elseif (!empty($outbox['first']['orderedItems'])) {
2131 $items = $outbox['first']['orderedItems'];
2132 } elseif (!empty($outbox['first']['href']) && ($outbox['first']['href'] != $feed)) {
2133 return self::updateFromOutbox($outbox['first']['href'], $data);
2134 } elseif (!empty($outbox['first'])) {
2135 if (is_string($outbox['first']) && ($outbox['first'] != $feed)) {
2136 return self::updateFromOutbox($outbox['first'], $data);
2138 Logger::warning('Unexpected data', ['outbox' => $outbox]);
2146 foreach ($items as $activity) {
2147 if (!empty($activity['published'])) {
2148 $published = DateTimeFormat::utc($activity['published']);
2149 } elseif (!empty($activity['object']['published'])) {
2150 $published = DateTimeFormat::utc($activity['object']['published']);
2155 if ($last_updated < $published) {
2156 $last_updated = $published;
2160 if (!empty($last_updated)) {
2161 return $last_updated;
2168 * Fetch the last activity date from an XML feed
2170 * @param array $data Probing result
2171 * @return string last activity
2173 private static function updateFromFeed(array $data): string
2175 // Search for the newest entry in the feed
2176 $curlResult = DI::httpClient()->get($data['poll'], HttpClientAccept::ATOM_XML);
2177 if (!$curlResult->isSuccess() || !$curlResult->getBody()) {
2181 $doc = new DOMDocument();
2182 @$doc->loadXML($curlResult->getBody());
2184 $xpath = new DOMXPath($doc);
2185 $xpath->registerNamespace('atom', 'http://www.w3.org/2005/Atom');
2187 $entries = $xpath->query('/atom:feed/atom:entry');
2191 foreach ($entries as $entry) {
2192 $published_item = $xpath->query('atom:published/text()', $entry)->item(0);
2193 $updated_item = $xpath->query('atom:updated/text()' , $entry)->item(0);
2194 $published = !empty($published_item->nodeValue) ? DateTimeFormat::utc($published_item->nodeValue) : null;
2195 $updated = !empty($updated_item->nodeValue) ? DateTimeFormat::utc($updated_item->nodeValue) : null;
2197 if (empty($published) || empty($updated)) {
2198 Logger::notice('Invalid entry for XPath.', ['entry' => $entry, 'url' => $data['url']]);
2202 if ($last_updated < $published) {
2203 $last_updated = $published;
2206 if ($last_updated < $updated) {
2207 $last_updated = $updated;
2211 if (!empty($last_updated)) {
2212 return $last_updated;
2219 * Probe data from local profiles without network traffic
2221 * @param string $url
2223 * @return array probed data
2224 * @throws HTTPException\InternalServerErrorException
2225 * @throws HTTPException\NotFoundException
2227 private static function localProbe(string $url): array
2230 $uid = User::getIdForURL($url);
2232 throw new HTTPException\NotFoundException('User not found.');
2235 $owner = User::getOwnerDataById($uid);
2236 $approfile = ActivityPub\Transmitter::getProfile($uid);
2238 $split_name = Diaspora::splitName($owner['name']);
2240 if (empty($owner['gsid'])) {
2241 $owner['gsid'] = GServer::getID($approfile['generator']['url']);
2245 'name' => $owner['name'], 'nick' => $owner['nick'], 'guid' => $approfile['diaspora:guid'] ?? '',
2246 'url' => $owner['url'], 'addr' => $owner['addr'], 'alias' => $owner['alias'],
2247 'photo' => User::getAvatarUrl($owner),
2248 'header' => $owner['header'] ? Contact::getHeaderUrlForId($owner['id'], $owner['updated']) : '',
2249 'account-type' => $owner['contact-type'], 'community' => ($owner['contact-type'] == User::ACCOUNT_TYPE_COMMUNITY),
2250 'keywords' => $owner['keywords'], 'location' => $owner['location'], 'about' => $owner['about'],
2251 'xmpp' => $owner['xmpp'], 'matrix' => $owner['matrix'],
2252 'hide' => !$owner['net-publish'], 'batch' => '', 'notify' => $owner['notify'],
2253 'poll' => $owner['poll'], 'request' => $owner['request'], 'confirm' => $owner['confirm'],
2254 'subscribe' => $approfile['generator']['url'] . '/contact/follow?url={uri}', 'poco' => $owner['poco'],
2255 'following' => $approfile['following'], 'followers' => $approfile['followers'],
2256 'inbox' => $approfile['inbox'], 'outbox' => $approfile['outbox'],
2257 'sharedinbox' => $approfile['endpoints']['sharedInbox'], 'network' => Protocol::DFRN,
2258 'pubkey' => $owner['upubkey'], 'baseurl' => $approfile['generator']['url'], 'gsid' => $owner['gsid'],
2259 'manually-approve' => in_array($owner['page-flags'], [User::PAGE_FLAGS_NORMAL, User::PAGE_FLAGS_PRVGROUP]),
2261 Protocol::DIASPORA => [
2262 'name' => $owner['name'],
2263 'given_name' => $split_name['first'],
2264 'family_name' => $split_name['last'],
2265 'nick' => $owner['nick'],
2266 'guid' => $approfile['diaspora:guid'],
2267 'url' => $owner['url'],
2268 'addr' => $owner['addr'],
2269 'alias' => $owner['alias'],
2270 'photo' => $owner['photo'],
2271 'photo_medium' => $owner['thumb'],
2272 'photo_small' => $owner['micro'],
2273 'batch' => $approfile['generator']['url'] . '/receive/public',
2274 'notify' => $owner['notify'],
2275 'poll' => $owner['poll'],
2276 'poco' => $owner['poco'],
2277 'network' => Protocol::DIASPORA,
2278 'pubkey' => $owner['upubkey'],
2282 } catch (Exception $e) {
2283 // Default values for nonexistent targets
2285 'name' => $url, 'nick' => $url, 'url' => $url, 'network' => Protocol::PHANTOM,
2286 'photo' => DI::baseUrl() . Contact::DEFAULT_AVATAR_PHOTO
2290 return self::rearrangeData($data);