3 * @copyright Copyright (C) 2020, Friendica
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;
26 use Friendica\Core\Hook;
27 use Friendica\Core\Logger;
28 use Friendica\Core\Protocol;
29 use Friendica\Core\System;
30 use Friendica\Database\DBA;
32 use Friendica\Model\Contact;
33 use Friendica\Model\GServer;
34 use Friendica\Model\Profile;
35 use Friendica\Model\User;
36 use Friendica\Protocol\ActivityNamespace;
37 use Friendica\Protocol\ActivityPub;
38 use Friendica\Protocol\Email;
39 use Friendica\Protocol\Feed;
40 use Friendica\Util\Crypto;
41 use Friendica\Util\DateTimeFormat;
42 use Friendica\Util\Network;
43 use Friendica\Util\Strings;
44 use Friendica\Util\XML;
47 * This class contain functions for probing URL
51 const WEBFINGER = '/.well-known/webfinger?resource={uri}';
53 private static $baseurl;
54 private static $istimeout;
57 * Remove stuff from an URI that doesn't belong there
60 * @return string Cleaned URI
62 public static function cleanURI(string $URI)
64 // At first remove leading and trailing junk
65 $URI = trim($URI, "@#?:/ \t\n\r\0\x0B");
67 $parts = parse_url($URI);
69 if (empty($parts['scheme'])) {
73 // Remove the URL fragment, since these shouldn't be part of any profile URL
74 unset($parts['fragment']);
76 $URI = Network::unparseURL($parts);
82 * Rearrange the array so that it always has the same order
84 * @param array $data Unordered data
86 * @return array Ordered data
88 private static function rearrangeData($data)
90 $fields = ["name", "nick", "guid", "url", "addr", "alias", "photo", "account-type",
91 "community", "keywords", "location", "about", "hide",
92 "batch", "notify", "poll", "request", "confirm", "subscribe", "poco",
93 "following", "followers", "inbox", "outbox", "sharedinbox",
94 "priority", "network", "pubkey", "manually-approve", "baseurl", "gsid"];
96 $numeric_fields = ["gsid", "hide", "account-type", "manually-approve"];
99 foreach ($fields as $field) {
100 if (isset($data[$field])) {
101 if (in_array($field, $numeric_fields)) {
102 $newdata[$field] = (int)$data[$field];
104 $newdata[$field] = $data[$field];
106 } elseif (!in_array($field, $numeric_fields)) {
107 $newdata[$field] = "";
109 $newdata[$field] = null;
113 // We don't use the "priority" field anymore and replace it with a dummy.
114 $newdata["priority"] = 0;
120 * Check if the hostname belongs to the own server
122 * @param string $host The hostname that is to be checked
124 * @return bool Does the testes hostname belongs to the own server?
126 private static function ownHost($host)
128 $own_host = DI::baseUrl()->getHostname();
130 $parts = parse_url($host);
132 if (!isset($parts['scheme'])) {
133 $parts = parse_url('http://'.$host);
136 if (!isset($parts['host'])) {
139 return $parts['host'] == $own_host;
143 * Probes for webfinger path via "host-meta"
145 * We have to check if the servers in the future still will offer this.
146 * It seems as if it was dropped from the standard.
148 * @param string $host The host part of an url
150 * @return array with template and type of the webfinger template for JSON or XML
151 * @throws HTTPException\InternalServerErrorException
153 private static function hostMeta($host)
155 // Reset the static variable
158 // Handles the case when the hostname contains the scheme
159 if (!parse_url($host, PHP_URL_SCHEME)) {
160 $ssl_url = "https://" . $host . "/.well-known/host-meta";
161 $url = "http://" . $host . "/.well-known/host-meta";
163 $ssl_url = $host . "/.well-known/host-meta";
167 $xrd_timeout = DI::config()->get('system', 'xrd_timeout', 20);
169 Logger::info('Probing', ['host' => $host, 'ssl_url' => $ssl_url, 'url' => $url, 'callstack' => System::callstack(20)]);
172 $curlResult = DI::httpRequest()->get($ssl_url, false, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']);
173 $ssl_connection_error = ($curlResult->getErrorNumber() == CURLE_COULDNT_CONNECT) || ($curlResult->getReturnCode() == 0);
174 if ($curlResult->isSuccess()) {
175 $xml = $curlResult->getBody();
176 $xrd = XML::parseString($xml, true);
178 $host_url = 'https://' . $host;
182 } elseif ($curlResult->isTimeout()) {
183 Logger::info('Probing timeout', ['url' => $ssl_url]);
184 self::$istimeout = true;
188 if (!is_object($xrd) && !empty($url)) {
189 $curlResult = DI::httpRequest()->get($url, false, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']);
190 $connection_error = ($curlResult->getErrorNumber() == CURLE_COULDNT_CONNECT) || ($curlResult->getReturnCode() == 0);
191 if ($curlResult->isTimeout()) {
192 Logger::info('Probing timeout', ['url' => $url]);
193 self::$istimeout = true;
195 } elseif ($connection_error && $ssl_connection_error) {
196 self::$istimeout = true;
200 $xml = $curlResult->getBody();
201 $xrd = XML::parseString($xml, true);
202 $host_url = 'http://'.$host;
204 if (!is_object($xrd)) {
205 Logger::info('No xrd object found', ['host' => $host]);
209 $links = XML::elementToArray($xrd);
210 if (!isset($links["xrd"]["link"])) {
211 Logger::info('No xrd data found', ['host' => $host]);
217 foreach ($links["xrd"]["link"] as $value => $link) {
218 if (!empty($link["@attributes"])) {
219 $attributes = $link["@attributes"];
220 } elseif ($value == "@attributes") {
226 if (!empty($attributes["rel"]) && $attributes["rel"] == "lrdd" && !empty($attributes["template"])) {
227 $type = (empty($attributes["type"]) ? '' : $attributes["type"]);
229 $lrdd[$type] = $attributes["template"];
233 self::$baseurl = $host_url;
235 Logger::info('Probing successful', ['host' => $host]);
241 * Perform Webfinger lookup and return DFRN data
243 * Given an email style address, perform webfinger lookup and
244 * return the resulting DFRN profile URL, or if no DFRN profile URL
245 * is located, returns an OStatus subscription template (prefixed
246 * with the string 'stat:' to identify it as on OStatus template).
247 * If this isn't an email style address just return $webbie.
248 * Return an empty string if email-style addresses but webfinger fails,
249 * or if the resultant personal XRD doesn't contain a supported
250 * subscription/friend-request attribute.
252 * amended 7/9/2011 to return an hcard which could save potentially loading
253 * a lengthy content page to scrape dfrn attributes
255 * @param string $webbie Address that should be probed
256 * @param string $hcard_url Link to the hcard - is returned by reference
258 * @return string profile link
259 * @throws HTTPException\InternalServerErrorException
261 public static function webfingerDfrn($webbie, &$hcard_url)
265 $links = self::lrdd($webbie);
266 Logger::debug('Result', ['url' => $webbie, 'links' => $links]);
267 if (!empty($links) && is_array($links)) {
268 foreach ($links as $link) {
269 if ($link['@attributes']['rel'] === ActivityNamespace::DFRN) {
270 $profile_link = $link['@attributes']['href'];
272 if (($link['@attributes']['rel'] === ActivityNamespace::OSTATUSSUB) && ($profile_link == "")) {
273 $profile_link = 'stat:'.$link['@attributes']['template'];
275 if ($link['@attributes']['rel'] === 'http://microformats.org/profile/hcard') {
276 $hcard_url = $link['@attributes']['href'];
280 return $profile_link;
284 * Check an URI for LRDD data
286 * @param string $uri Address that should be probed
288 * @return array uri data
289 * @throws HTTPException\InternalServerErrorException
291 public static function lrdd(string $uri)
293 $data = self::getWebfingerArray($uri);
297 $webfinger = $data['webfinger'];
299 if (empty($webfinger["links"])) {
300 Logger::info('No webfinger links found', ['uri' => $uri]);
306 foreach ($webfinger["links"] as $link) {
307 $data[] = ["@attributes" => $link];
310 if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) {
311 foreach ($webfinger["aliases"] as $alias) {
312 $data[] = ["@attributes" =>
322 * Fetch information (protocol endpoints and user information) about a given uri
324 * @param string $uri Address that should be probed
325 * @param string $network Test for this specific network
326 * @param integer $uid User ID for the probe (only used for mails)
327 * @param boolean $cache Use cached values?
329 * @return array uri data
330 * @throws HTTPException\InternalServerErrorException
331 * @throws \ImagickException
333 public static function uri($uri, $network = '', $uid = -1)
335 // Local profiles aren't probed via network
336 if (empty($network) && strpos($uri, DI::baseUrl()->getHostname())) {
337 $data = self::localProbe($uri);
347 if (empty($network) || ($network == Protocol::ACTIVITYPUB)) {
348 $ap_profile = ActivityPub::probeProfile($uri);
353 self::$istimeout = false;
355 if ($network != Protocol::ACTIVITYPUB) {
356 $data = self::detect($uri, $network, $uid, $ap_profile);
357 if (!is_array($data)) {
360 if (empty($data) || (!empty($ap_profile) && empty($network) && (($data['network'] ?? '') != Protocol::DFRN))) {
362 } elseif (!empty($ap_profile)) {
363 $ap_profile['batch'] = '';
364 $data = array_merge($ap_profile, $data);
370 if (!isset($data['url'])) {
374 if (empty($data['photo'])) {
375 $data['photo'] = DI::baseUrl() . Contact::DEFAULT_AVATAR_PHOTO;
378 if (empty($data['name'])) {
379 if (!empty($data['nick'])) {
380 $data['name'] = $data['nick'];
383 if (empty($data['name'])) {
384 $data['name'] = $data['url'];
388 if (empty($data['nick'])) {
389 $data['nick'] = strtolower($data['name']);
391 if (strpos($data['nick'], ' ')) {
392 $data['nick'] = trim(substr($data['nick'], 0, strpos($data['nick'], ' ')));
396 if (!empty($data['baseurl']) && empty($data['gsid'])) {
397 $data['gsid'] = GServer::getID($data['baseurl']);
400 if (empty($data['network'])) {
401 $data['network'] = Protocol::PHANTOM;
404 // Ensure that local connections always are DFRN
405 if (($network == '') && ($data['network'] != Protocol::PHANTOM) && (self::ownHost($data['baseurl'] ?? '') || self::ownHost($data['url']))) {
406 $data['network'] = Protocol::DFRN;
409 if (!isset($data['hide']) && in_array($data['network'], Protocol::FEDERATED)) {
410 $data['hide'] = self::getHideStatus($data['url']);
413 return self::rearrangeData($data);
418 * Fetches the "hide" status from the profile
420 * @param string $url URL of the profile
422 * @return boolean "hide" status
424 private static function getHideStatus($url)
426 $curlResult = DI::httpRequest()->get($url, false, ['content_length' => 1000000]);
427 if (!$curlResult->isSuccess()) {
431 // If it isn't a HTML file then exit
432 if (!in_array('html', $curlResult->getContentType())) {
436 $body = $curlResult->getBody();
438 $doc = new DOMDocument();
439 @$doc->loadHTML($body);
441 $xpath = new DOMXPath($doc);
443 $list = $xpath->query('//meta[@name]');
444 foreach ($list as $node) {
446 if ($node->attributes->length) {
447 foreach ($node->attributes as $attribute) {
448 $meta_tag[$attribute->name] = $attribute->value;
452 if (empty($meta_tag['content'])) {
456 $content = strtolower(trim($meta_tag['content']));
458 switch (strtolower(trim($meta_tag['name']))) {
459 case 'dfrn-global-visibility':
460 if ($content == 'false') {
465 if (strpos($content, 'noindex') !== false) {
476 * Fetch the "subscribe" and add it to the result
478 * @param array $result
479 * @param array $webfinger
480 * @return array result
482 private static function getSubscribeLink(array $result, array $webfinger)
484 if (empty($webfinger['links'])) {
488 foreach ($webfinger['links'] as $link) {
489 if (!empty($link['template']) && ($link['rel'] === ActivityNamespace::OSTATUSSUB)) {
490 $result['subscribe'] = $link['template'];
498 * Get webfinger data from a given URI
501 * @return array Webfinger array
503 private static function getWebfingerArray(string $uri)
505 $parts = parse_url($uri);
507 if (!empty($parts['scheme']) && !empty($parts['host'])) {
508 $host = $parts['host'];
509 if (!empty($parts['port'])) {
510 $host .= ':'.$parts['port'];
513 $baseurl = $parts['scheme'] . '://' . $host;
518 $path_parts = explode("/", trim($parts['path'] ?? '', "/"));
519 if (!empty($path_parts)) {
520 $nick = ltrim(end($path_parts), '@');
521 // When the last part of the URI is numeric then it is most likely an ID and not a nick name
522 if (!is_numeric($nick)) {
523 $addr = $nick."@".$host;
529 $webfinger = self::getWebfinger($parts['scheme'] . '://' . $host . self::WEBFINGER, 'application/jrd+json', $uri, $addr);
530 if (empty($webfinger)) {
531 $lrdd = self::hostMeta($host);
534 if (empty($webfinger) && empty($lrdd)) {
535 while (empty($lrdd) && empty($webfinger) && (sizeof($path_parts) > 1)) {
536 $host .= "/".array_shift($path_parts);
537 $baseurl = $parts['scheme'] . '://' . $host;
540 $addr = $nick."@".$host;
543 $webfinger = self::getWebfinger($parts['scheme'] . '://' . $host . self::WEBFINGER, 'application/jrd+json', $uri, $addr);
544 if (empty($webfinger)) {
545 $lrdd = self::hostMeta($host);
549 if (empty($lrdd) && empty($webfinger)) {
553 } elseif (strstr($uri, '@')) {
554 // Remove "acct:" from the URI
555 $uri = str_replace('acct:', '', $uri);
557 $host = substr($uri, strpos($uri, '@') + 1);
558 $nick = substr($uri, 0, strpos($uri, '@'));
561 $webfinger = self::getWebfinger('https://' . $host . self::WEBFINGER, 'application/jrd+json', $uri, $addr);
562 if (self::$istimeout) {
566 if (empty($webfinger)) {
567 $webfinger = self::getWebfinger('http://' . $host . self::WEBFINGER, 'application/jrd+json', $uri, $addr);
568 if (self::$istimeout) {
572 $baseurl = 'https://' . $host;
575 if (empty($webfinger)) {
576 $lrdd = self::hostMeta($host);
577 if (self::$istimeout) {
580 $baseurl = self::$baseurl;
582 $baseurl = 'http://' . $host;
585 Logger::info('URI was not detectable', ['uri' => $uri]);
589 if (empty($webfinger)) {
590 foreach ($lrdd as $type => $template) {
595 $webfinger = self::getWebfinger($template, $type, $uri, $addr);
599 if (empty($webfinger)) {
603 if ($webfinger['detected'] == $addr) {
604 $webfinger['nick'] = $nick;
605 $webfinger['addr'] = $addr;
608 $webfinger['baseurl'] = $baseurl;
614 * Perform network request for webfinger data
616 * @param string $template
617 * @param string $type
619 * @param string $addr
620 * @return array webfinger results
622 private static function getWebfinger(string $template, string $type, string $uri, string $addr)
624 // First try the address because this is the primary purpose of webfinger
627 $path = str_replace('{uri}', urlencode("acct:" . $addr), $template);
628 $webfinger = self::webfinger($path, $type);
629 if (self::$istimeout) {
635 if (empty($webfinger) && $uri != $addr) {
637 $path = str_replace('{uri}', urlencode($uri), $template);
638 $webfinger = self::webfinger($path, $type);
639 if (self::$istimeout) {
644 if (empty($webfinger)) {
648 return ['webfinger' => $webfinger, 'detected' => $detected];
652 * Fetch information (protocol endpoints and user information) about a given uri
654 * This function is only called by the "uri" function that adds caching and rearranging of data.
656 * @param string $uri Address that should be probed
657 * @param string $network Test for this specific network
658 * @param integer $uid User ID for the probe (only used for mails)
659 * @param array $ap_profile Previously probed AP profile
661 * @return array uri data
662 * @throws HTTPException\InternalServerErrorException
664 private static function detect(string $uri, string $network, int $uid, array $ap_profile)
668 'network' => $network,
673 Hook::callAll('probe_detect', $hookData);
675 if ($hookData['result']) {
676 if (!is_array($hookData['result'])) {
679 return $hookData['result'];
683 $parts = parse_url($uri);
685 if (!empty($parts['scheme']) && !empty($parts['host'])) {
686 if (in_array($parts['host'], ['twitter.com', 'mobile.twitter.com'])) {
687 return self::twitter($uri);
689 } elseif (strstr($uri, '@')) {
690 // If the URI starts with "mailto:" then jump directly to the mail detection
691 if (strpos($uri, 'mailto:') !== false) {
692 $uri = str_replace('mailto:', '', $uri);
693 return self::mail($uri, $uid);
696 if ($network == Protocol::MAIL) {
697 return self::mail($uri, $uid);
700 if (Strings::endsWith($uri, '@twitter.com')
701 || Strings::endsWith($uri, '@mobile.twitter.com')
703 return self::twitter($uri);
706 Logger::info('URI was not detectable', ['uri' => $uri]);
710 Logger::info('Probing start', ['uri' => $uri]);
712 if (!empty($ap_profile['addr']) && ($ap_profile['addr'] != $uri)) {
713 $data = self::getWebfingerArray($ap_profile['addr']);
717 $data = self::getWebfingerArray($uri);
721 if (!empty($parts['scheme'])) {
722 return self::feed($uri);
723 } elseif (!empty($uid)) {
724 return self::mail($uri, $uid);
730 $webfinger = $data['webfinger'];
731 $nick = $data['nick'] ?? '';
732 $addr = $data['addr'] ?? '';
733 $baseurl = $data['baseurl'] ?? '';
737 if (in_array($network, ["", Protocol::DFRN])) {
738 $result = self::dfrn($webfinger);
740 if ((!$result && ($network == "")) || ($network == Protocol::DIASPORA)) {
741 $result = self::diaspora($webfinger);
743 if ((!$result && ($network == "")) || ($network == Protocol::OSTATUS)) {
744 $result = self::ostatus($webfinger);
746 if (in_array($network, ['', Protocol::ZOT])) {
747 $result = self::zot($webfinger, $result, $baseurl);
749 if ((!$result && ($network == "")) || ($network == Protocol::PUMPIO)) {
750 $result = self::pumpio($webfinger, $addr);
752 if (empty($result['network']) && empty($ap_profile['network']) || ($network == Protocol::FEED)) {
753 $result = self::feed($uri);
755 // We overwrite the detected nick with our try if the previois routines hadn't detected it.
756 // Additionally it is overwritten when the nickname doesn't make sense (contains spaces).
757 if ((empty($result["nick"]) || (strstr($result["nick"], " "))) && ($nick != "")) {
758 $result["nick"] = $nick;
761 if (empty($result["addr"]) && ($addr != "")) {
762 $result["addr"] = $addr;
766 $result = self::getSubscribeLink($result, $webfinger);
768 if (empty($result["network"])) {
769 $result["network"] = Protocol::PHANTOM;
772 if (empty($result['baseurl']) && !empty($baseurl)) {
773 $result['baseurl'] = $baseurl;
776 if (empty($result["url"])) {
777 $result["url"] = $uri;
780 Logger::info('Probing done', ['uri' => $uri, 'network' => $result["network"]]);
786 * Check for Zot contact
788 * @param array $webfinger Webfinger data
789 * @param array $data previously probed data
791 * @return array Zot data
792 * @throws HTTPException\InternalServerErrorException
794 private static function zot($webfinger, $data, $baseurl)
796 if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) {
797 foreach ($webfinger["aliases"] as $alias) {
798 if (substr($alias, 0, 5) == 'acct:') {
799 $data["addr"] = substr($alias, 5);
804 if (!empty($webfinger["subject"]) && (substr($webfinger["subject"], 0, 5) == "acct:")) {
805 $data["addr"] = substr($webfinger["subject"], 5);
809 foreach ($webfinger['links'] as $link) {
810 if (($link['rel'] == 'http://purl.org/zot/protocol') && !empty($link['href'])) {
811 $zot_url = $link['href'];
815 if (empty($zot_url) && !empty($data['addr']) && !empty($baseurl)) {
816 $condition = ['nurl' => Strings::normaliseLink($baseurl), 'platform' => ['hubzilla']];
817 if (!DBA::exists('gserver', $condition)) {
820 $zot_url = $baseurl . '/.well-known/zot-info?address=' . $data['addr'];
823 if (empty($zot_url)) {
827 $data = self::pollZot($zot_url, $data);
829 if (!empty($data['url']) && !empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
830 foreach ($webfinger['aliases'] as $alias) {
831 if (!strstr($alias, '@') && Strings::normaliseLink($alias) != Strings::normaliseLink($data['url'])) {
832 $data['alias'] = $alias;
840 public static function pollZot($url, $data)
842 $curlResult = DI::httpRequest()->get($url);
843 if ($curlResult->isTimeout()) {
846 $content = $curlResult->getBody();
851 $json = json_decode($content, true);
852 if (!is_array($json)) {
856 if (empty($data['network'])) {
857 if (!empty($json['protocols']) && in_array('zot', $json['protocols'])) {
858 $data['network'] = Protocol::ZOT;
859 } elseif (!isset($json['protocols'])) {
860 $data['network'] = Protocol::ZOT;
864 if (!empty($json['guid']) && empty($data['guid'])) {
865 $data['guid'] = $json['guid'];
867 if (!empty($json['key']) && empty($data['pubkey'])) {
868 $data['pubkey'] = $json['key'];
870 if (!empty($json['name'])) {
871 $data['name'] = $json['name'];
873 if (!empty($json['photo'])) {
874 $data['photo'] = $json['photo'];
875 if (!empty($json['photo_updated'])) {
876 $data['photo'] .= '?rev=' . urlencode($json['photo_updated']);
879 if (!empty($json['address'])) {
880 $data['addr'] = $json['address'];
882 if (!empty($json['url'])) {
883 $data['url'] = $json['url'];
885 if (!empty($json['connections_url'])) {
886 $data['poco'] = $json['connections_url'];
888 if (isset($json['searchable'])) {
889 $data['hide'] = !$json['searchable'];
891 if (!empty($json['public_forum'])) {
892 $data['community'] = $json['public_forum'];
893 $data['account-type'] = User::PAGE_FLAGS_COMMUNITY;
896 if (!empty($json['profile'])) {
897 $profile = $json['profile'];
898 if (!empty($profile['description'])) {
899 $data['about'] = $profile['description'];
901 if (!empty($profile['keywords'])) {
902 $keywords = implode(', ', $profile['keywords']);
903 if (!empty($keywords)) {
904 $data['keywords'] = $keywords;
909 if (!empty($profile['region'])) {
910 $loc['region'] = $profile['region'];
912 if (!empty($profile['country'])) {
913 $loc['country-name'] = $profile['country'];
915 $location = Profile::formatLocation($loc);
916 if (!empty($location)) {
917 $data['location'] = $location;
925 * Perform a webfinger request.
927 * For details see RFC 7033: <https://tools.ietf.org/html/rfc7033>
929 * @param string $url Address that should be probed
930 * @param string $type type
932 * @return array webfinger data
933 * @throws HTTPException\InternalServerErrorException
935 public static function webfinger($url, $type)
937 $xrd_timeout = DI::config()->get('system', 'xrd_timeout', 20);
939 $curlResult = DI::httpRequest()->get($url, false, ['timeout' => $xrd_timeout, 'accept_content' => $type]);
940 if ($curlResult->isTimeout()) {
941 self::$istimeout = true;
944 $data = $curlResult->getBody();
946 $webfinger = json_decode($data, true);
947 if (!empty($webfinger)) {
948 if (!isset($webfinger["links"])) {
949 Logger::info('No json webfinger links', ['url' => $url]);
955 // If it is not JSON, maybe it is XML
956 $xrd = XML::parseString($data, true);
957 if (!is_object($xrd)) {
958 Logger::info('No webfinger data retrievable', ['url' => $url]);
962 $xrd_arr = XML::elementToArray($xrd);
963 if (!isset($xrd_arr["xrd"]["link"])) {
964 Logger::info('No XML webfinger links', ['url' => $url]);
970 if (!empty($xrd_arr["xrd"]["subject"])) {
971 $webfinger["subject"] = $xrd_arr["xrd"]["subject"];
974 if (!empty($xrd_arr["xrd"]["alias"])) {
975 $webfinger["aliases"] = $xrd_arr["xrd"]["alias"];
978 $webfinger["links"] = [];
980 foreach ($xrd_arr["xrd"]["link"] as $value => $data) {
981 if (!empty($data["@attributes"])) {
982 $attributes = $data["@attributes"];
983 } elseif ($value == "@attributes") {
989 $webfinger["links"][] = $attributes;
995 * Poll the Friendica specific noscrape page.
997 * "noscrape" is a faster alternative to fetch the data from the hcard.
998 * This functionality was originally created for the directory.
1000 * @param string $noscrape_url Link to the noscrape page
1001 * @param array $data The already fetched data
1003 * @return array noscrape data
1004 * @throws HTTPException\InternalServerErrorException
1006 private static function pollNoscrape($noscrape_url, $data)
1008 $curlResult = DI::httpRequest()->get($noscrape_url);
1009 if ($curlResult->isTimeout()) {
1010 self::$istimeout = true;
1013 $content = $curlResult->getBody();
1015 Logger::info('Empty body', ['url' => $noscrape_url]);
1019 $json = json_decode($content, true);
1020 if (!is_array($json)) {
1021 Logger::info('No json data', ['url' => $noscrape_url]);
1025 if (!empty($json["fn"])) {
1026 $data["name"] = $json["fn"];
1029 if (!empty($json["addr"])) {
1030 $data["addr"] = $json["addr"];
1033 if (!empty($json["nick"])) {
1034 $data["nick"] = $json["nick"];
1037 if (!empty($json["guid"])) {
1038 $data["guid"] = $json["guid"];
1041 if (!empty($json["comm"])) {
1042 $data["community"] = $json["comm"];
1045 if (!empty($json["tags"])) {
1046 $keywords = implode(", ", $json["tags"]);
1047 if ($keywords != "") {
1048 $data["keywords"] = $keywords;
1052 $location = Profile::formatLocation($json);
1054 $data["location"] = $location;
1057 if (!empty($json["about"])) {
1058 $data["about"] = $json["about"];
1061 if (!empty($json["key"])) {
1062 $data["pubkey"] = $json["key"];
1065 if (!empty($json["photo"])) {
1066 $data["photo"] = $json["photo"];
1069 if (!empty($json["dfrn-request"])) {
1070 $data["request"] = $json["dfrn-request"];
1073 if (!empty($json["dfrn-confirm"])) {
1074 $data["confirm"] = $json["dfrn-confirm"];
1077 if (!empty($json["dfrn-notify"])) {
1078 $data["notify"] = $json["dfrn-notify"];
1081 if (!empty($json["dfrn-poll"])) {
1082 $data["poll"] = $json["dfrn-poll"];
1085 if (isset($json["hide"])) {
1086 $data["hide"] = (bool)$json["hide"];
1088 $data["hide"] = false;
1095 * Check for valid DFRN data
1097 * @param array $data DFRN data
1099 * @return int Number of errors
1101 public static function validDfrn($data)
1104 if (!isset($data['key'])) {
1107 if (!isset($data['dfrn-request'])) {
1110 if (!isset($data['dfrn-confirm'])) {
1113 if (!isset($data['dfrn-notify'])) {
1116 if (!isset($data['dfrn-poll'])) {
1123 * Fetch data from a DFRN profile page and via "noscrape"
1125 * @param string $profile_link Link to the profile page
1127 * @return array profile data
1128 * @throws HTTPException\InternalServerErrorException
1129 * @throws \ImagickException
1131 public static function profile($profile_link)
1135 Logger::info('Check profile', ['link' => $profile_link]);
1137 // Fetch data via noscrape - this is faster
1138 $noscrape_url = str_replace(["/hcard/", "/profile/"], "/noscrape/", $profile_link);
1139 $data = self::pollNoscrape($noscrape_url, $data);
1141 if (!isset($data["notify"])
1142 || !isset($data["confirm"])
1143 || !isset($data["request"])
1144 || !isset($data["poll"])
1145 || !isset($data["name"])
1146 || !isset($data["photo"])
1148 $data = self::pollHcard($profile_link, $data, true);
1153 if (empty($data["addr"]) || empty($data["nick"])) {
1154 $probe_data = self::uri($profile_link);
1155 $data["addr"] = ($data["addr"] ?? '') ?: $probe_data["addr"];
1156 $data["nick"] = ($data["nick"] ?? '') ?: $probe_data["nick"];
1159 $prof_data["addr"] = $data["addr"];
1160 $prof_data["nick"] = $data["nick"];
1161 $prof_data["dfrn-request"] = $data['request'] ?? null;
1162 $prof_data["dfrn-confirm"] = $data['confirm'] ?? null;
1163 $prof_data["dfrn-notify"] = $data['notify'] ?? null;
1164 $prof_data["dfrn-poll"] = $data['poll'] ?? null;
1165 $prof_data["photo"] = $data['photo'] ?? null;
1166 $prof_data["fn"] = $data['name'] ?? null;
1167 $prof_data["key"] = $data['pubkey'] ?? null;
1169 Logger::debug('Result', ['link' => $profile_link, 'data' => $prof_data]);
1175 * Check for DFRN contact
1177 * @param array $webfinger Webfinger data
1179 * @return array DFRN data
1180 * @throws HTTPException\InternalServerErrorException
1182 private static function dfrn($webfinger)
1186 // The array is reversed to take into account the order of preference for same-rel links
1187 // See: https://tools.ietf.org/html/rfc7033#section-4.4.4
1188 foreach (array_reverse($webfinger["links"]) as $link) {
1189 if (($link["rel"] == ActivityNamespace::DFRN) && !empty($link["href"])) {
1190 $data["network"] = Protocol::DFRN;
1191 } elseif (($link["rel"] == ActivityNamespace::FEED) && !empty($link["href"])) {
1192 $data["poll"] = $link["href"];
1193 } elseif (($link["rel"] == "http://webfinger.net/rel/profile-page") && (($link["type"] ?? "") == "text/html") && !empty($link["href"])) {
1194 $data["url"] = $link["href"];
1195 } elseif (($link["rel"] == "http://microformats.org/profile/hcard") && !empty($link["href"])) {
1196 $hcard_url = $link["href"];
1197 } elseif (($link["rel"] == ActivityNamespace::POCO) && !empty($link["href"])) {
1198 $data["poco"] = $link["href"];
1199 } elseif (($link["rel"] == "http://webfinger.net/rel/avatar") && !empty($link["href"])) {
1200 $data["photo"] = $link["href"];
1201 } elseif (($link["rel"] == "http://joindiaspora.com/seed_location") && !empty($link["href"])) {
1202 $data["baseurl"] = trim($link["href"], '/');
1203 } elseif (($link["rel"] == "http://joindiaspora.com/guid") && !empty($link["href"])) {
1204 $data["guid"] = $link["href"];
1205 } elseif (($link["rel"] == "diaspora-public-key") && !empty($link["href"])) {
1206 $data["pubkey"] = base64_decode($link["href"]);
1208 //if (strstr($data["pubkey"], 'RSA ') || ($link["type"] == "RSA"))
1209 if (strstr($data["pubkey"], 'RSA ')) {
1210 $data["pubkey"] = Crypto::rsaToPem($data["pubkey"]);
1215 if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) {
1216 foreach ($webfinger["aliases"] as $alias) {
1217 if (empty($data["url"]) && !strstr($alias, "@")) {
1218 $data["url"] = $alias;
1219 } elseif (!strstr($alias, "@") && Strings::normaliseLink($alias) != Strings::normaliseLink($data["url"])) {
1220 $data["alias"] = $alias;
1221 } elseif (substr($alias, 0, 5) == 'acct:') {
1222 $data["addr"] = substr($alias, 5);
1227 if (!empty($webfinger["subject"]) && (substr($webfinger["subject"], 0, 5) == "acct:")) {
1228 $data["addr"] = substr($webfinger["subject"], 5);
1231 if (!isset($data["network"]) || ($hcard_url == "")) {
1235 // Fetch data via noscrape - this is faster
1236 $noscrape_url = str_replace("/hcard/", "/noscrape/", $hcard_url);
1237 $data = self::pollNoscrape($noscrape_url, $data);
1239 if (isset($data["notify"])
1240 && isset($data["confirm"])
1241 && isset($data["request"])
1242 && isset($data["poll"])
1243 && isset($data["name"])
1244 && isset($data["photo"])
1249 $data = self::pollHcard($hcard_url, $data, true);
1255 * Poll the hcard page (Diaspora and Friendica specific)
1257 * @param string $hcard_url Link to the hcard page
1258 * @param array $data The already fetched data
1259 * @param boolean $dfrn Poll DFRN specific data
1261 * @return array hcard data
1262 * @throws HTTPException\InternalServerErrorException
1264 private static function pollHcard($hcard_url, $data, $dfrn = false)
1266 $curlResult = DI::httpRequest()->get($hcard_url);
1267 if ($curlResult->isTimeout()) {
1268 self::$istimeout = true;
1271 $content = $curlResult->getBody();
1276 $doc = new DOMDocument();
1277 if (!@$doc->loadHTML($content)) {
1281 $xpath = new DomXPath($doc);
1283 $vcards = $xpath->query("//div[contains(concat(' ', @class, ' '), ' vcard ')]");
1284 if (!is_object($vcards)) {
1288 if (!isset($data["baseurl"])) {
1289 $data["baseurl"] = "";
1292 if ($vcards->length > 0) {
1293 $vcard = $vcards->item(0);
1295 // We have to discard the guid from the hcard in favour of the guid from lrdd
1296 // Reason: Hubzilla doesn't use the value "uid" in the hcard like Diaspora does.
1297 $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' uid ')]", $vcard); // */
1298 if (($search->length > 0) && empty($data["guid"])) {
1299 $data["guid"] = $search->item(0)->nodeValue;
1302 $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' nickname ')]", $vcard); // */
1303 if ($search->length > 0) {
1304 $data["nick"] = $search->item(0)->nodeValue;
1307 $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' fn ')]", $vcard); // */
1308 if ($search->length > 0) {
1309 $data["name"] = $search->item(0)->nodeValue;
1312 $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' searchable ')]", $vcard); // */
1313 if ($search->length > 0) {
1314 $data["searchable"] = $search->item(0)->nodeValue;
1317 $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' key ')]", $vcard); // */
1318 if ($search->length > 0) {
1319 $data["pubkey"] = $search->item(0)->nodeValue;
1320 if (strstr($data["pubkey"], 'RSA ')) {
1321 $data["pubkey"] = Crypto::rsaToPem($data["pubkey"]);
1325 $search = $xpath->query("//*[@id='pod_location']", $vcard); // */
1326 if ($search->length > 0) {
1327 $data["baseurl"] = trim($search->item(0)->nodeValue, "/");
1332 if (!empty($vcard)) {
1333 $photos = $xpath->query("//*[contains(concat(' ', @class, ' '), ' photo ') or contains(concat(' ', @class, ' '), ' avatar ')]", $vcard); // */
1334 foreach ($photos as $photo) {
1336 foreach ($photo->attributes as $attribute) {
1337 $attr[$attribute->name] = trim($attribute->value);
1340 if (isset($attr["src"]) && isset($attr["width"])) {
1341 $avatar[$attr["width"]] = $attr["src"];
1344 // We don't have a width. So we just take everything that we got.
1345 // This is a Hubzilla workaround which doesn't send a width.
1346 if ((sizeof($avatar) == 0) && !empty($attr["src"])) {
1347 $avatar[] = $attr["src"];
1352 if (sizeof($avatar)) {
1354 $data["photo"] = self::fixAvatar(array_pop($avatar), $data["baseurl"]);
1358 // Poll DFRN specific data
1359 $search = $xpath->query("//link[contains(concat(' ', @rel), ' dfrn-')]");
1360 if ($search->length > 0) {
1361 foreach ($search as $link) {
1362 //$data["request"] = $search->item(0)->nodeValue;
1364 foreach ($link->attributes as $attribute) {
1365 $attr[$attribute->name] = trim($attribute->value);
1368 $data[substr($attr["rel"], 5)] = $attr["href"];
1372 // Older Friendica versions had used the "uid" field differently than newer versions
1373 if (!empty($data["nick"]) && !empty($data["guid"]) && ($data["nick"] == $data["guid"])) {
1374 unset($data["guid"]);
1383 * Check for Diaspora contact
1385 * @param array $webfinger Webfinger data
1387 * @return array Diaspora data
1388 * @throws HTTPException\InternalServerErrorException
1390 private static function diaspora($webfinger)
1395 // The array is reversed to take into account the order of preference for same-rel links
1396 // See: https://tools.ietf.org/html/rfc7033#section-4.4.4
1397 foreach (array_reverse($webfinger["links"]) as $link) {
1398 if (($link["rel"] == "http://microformats.org/profile/hcard") && !empty($link["href"])) {
1399 $hcard_url = $link["href"];
1400 } elseif (($link["rel"] == "http://joindiaspora.com/seed_location") && !empty($link["href"])) {
1401 $data["baseurl"] = trim($link["href"], '/');
1402 } elseif (($link["rel"] == "http://joindiaspora.com/guid") && !empty($link["href"])) {
1403 $data["guid"] = $link["href"];
1404 } elseif (($link["rel"] == "http://webfinger.net/rel/profile-page") && (($link["type"] ?? "") == "text/html") && !empty($link["href"])) {
1405 $data["url"] = $link["href"];
1406 } elseif (($link["rel"] == ActivityNamespace::FEED) && !empty($link["href"])) {
1407 $data["poll"] = $link["href"];
1408 } elseif (($link["rel"] == ActivityNamespace::POCO) && !empty($link["href"])) {
1409 $data["poco"] = $link["href"];
1410 } elseif (($link["rel"] == "salmon") && !empty($link["href"])) {
1411 $data["notify"] = $link["href"];
1412 } elseif (($link["rel"] == "diaspora-public-key") && !empty($link["href"])) {
1413 $data["pubkey"] = base64_decode($link["href"]);
1415 //if (strstr($data["pubkey"], 'RSA ') || ($link["type"] == "RSA"))
1416 if (strstr($data["pubkey"], 'RSA ')) {
1417 $data["pubkey"] = Crypto::rsaToPem($data["pubkey"]);
1422 if (empty($data["url"]) || empty($hcard_url)) {
1426 if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) {
1427 foreach ($webfinger["aliases"] as $alias) {
1428 if (Strings::normaliseLink($alias) != Strings::normaliseLink($data["url"]) && ! strstr($alias, "@")) {
1429 $data["alias"] = $alias;
1430 } elseif (substr($alias, 0, 5) == 'acct:') {
1431 $data["addr"] = substr($alias, 5);
1436 if (!empty($webfinger["subject"]) && (substr($webfinger["subject"], 0, 5) == 'acct:')) {
1437 $data["addr"] = substr($webfinger["subject"], 5);
1440 // Fetch further information from the hcard
1441 $data = self::pollHcard($hcard_url, $data);
1447 if (!empty($data["url"])
1448 && !empty($data["guid"])
1449 && !empty($data["baseurl"])
1450 && !empty($data["pubkey"])
1451 && !empty($hcard_url)
1453 $data["network"] = Protocol::DIASPORA;
1454 $data["manually-approve"] = false;
1456 // The Diaspora handle must always be lowercase
1457 if (!empty($data["addr"])) {
1458 $data["addr"] = strtolower($data["addr"]);
1461 // We have to overwrite the detected value for "notify" since Hubzilla doesn't send it
1462 $data["notify"] = $data["baseurl"] . "/receive/users/" . $data["guid"];
1463 $data["batch"] = $data["baseurl"] . "/receive/public";
1472 * Check for OStatus contact
1474 * @param array $webfinger Webfinger data
1475 * @param bool $short Short detection mode
1477 * @return array|bool OStatus data or "false" on error or "true" on short mode
1478 * @throws HTTPException\InternalServerErrorException
1480 private static function ostatus($webfinger, $short = false)
1484 if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) {
1485 foreach ($webfinger["aliases"] as $alias) {
1486 if (strstr($alias, "@") && !strstr(Strings::normaliseLink($alias), "http://")) {
1487 $data["addr"] = str_replace('acct:', '', $alias);
1492 if (!empty($webfinger["subject"]) && strstr($webfinger["subject"], "@")
1493 && !strstr(Strings::normaliseLink($webfinger["subject"]), "http://")
1495 $data["addr"] = str_replace('acct:', '', $webfinger["subject"]);
1498 if (!empty($webfinger["links"])) {
1499 // The array is reversed to take into account the order of preference for same-rel links
1500 // See: https://tools.ietf.org/html/rfc7033#section-4.4.4
1501 foreach (array_reverse($webfinger["links"]) as $link) {
1502 if (($link["rel"] == "http://webfinger.net/rel/profile-page")
1503 && (($link["type"] ?? "") == "text/html")
1504 && ($link["href"] != "")
1506 $data["url"] = $data["alias"] = $link["href"];
1507 } elseif (($link["rel"] == "salmon") && !empty($link["href"])) {
1508 $data["notify"] = $link["href"];
1509 } elseif (($link["rel"] == ActivityNamespace::FEED) && !empty($link["href"])) {
1510 $data["poll"] = $link["href"];
1511 } elseif (($link["rel"] == "magic-public-key") && !empty($link["href"])) {
1512 $pubkey = $link["href"];
1514 if (substr($pubkey, 0, 5) === 'data:') {
1515 if (strstr($pubkey, ',')) {
1516 $pubkey = substr($pubkey, strpos($pubkey, ',') + 1);
1518 $pubkey = substr($pubkey, 5);
1520 } elseif (Strings::normaliseLink($pubkey) == 'http://') {
1521 $curlResult = DI::httpRequest()->get($pubkey);
1522 if ($curlResult->isTimeout()) {
1523 self::$istimeout = true;
1524 return $short ? false : [];
1526 $pubkey = $curlResult->getBody();
1529 $key = explode(".", $pubkey);
1531 if (sizeof($key) >= 3) {
1532 $m = Strings::base64UrlDecode($key[1]);
1533 $e = Strings::base64UrlDecode($key[2]);
1534 $data["pubkey"] = Crypto::meToPem($m, $e);
1540 if (isset($data["notify"]) && isset($data["pubkey"])
1541 && isset($data["poll"])
1542 && isset($data["url"])
1544 $data["network"] = Protocol::OSTATUS;
1545 $data["manually-approve"] = false;
1547 return $short ? false : [];
1554 // Fetch all additional data from the feed
1555 $curlResult = DI::httpRequest()->get($data["poll"]);
1556 if ($curlResult->isTimeout()) {
1557 self::$istimeout = true;
1560 $feed = $curlResult->getBody();
1561 $feed_data = Feed::import($feed);
1566 if (!empty($feed_data["header"]["author-name"])) {
1567 $data["name"] = $feed_data["header"]["author-name"];
1569 if (!empty($feed_data["header"]["author-nick"])) {
1570 $data["nick"] = $feed_data["header"]["author-nick"];
1572 if (!empty($feed_data["header"]["author-avatar"])) {
1573 $data["photo"] = self::fixAvatar($feed_data["header"]["author-avatar"], $data["url"]);
1575 if (!empty($feed_data["header"]["author-id"])) {
1576 $data["alias"] = $feed_data["header"]["author-id"];
1578 if (!empty($feed_data["header"]["author-location"])) {
1579 $data["location"] = $feed_data["header"]["author-location"];
1581 if (!empty($feed_data["header"]["author-about"])) {
1582 $data["about"] = $feed_data["header"]["author-about"];
1584 // OStatus has serious issues when the the url doesn't fit (ssl vs. non ssl)
1585 // So we take the value that we just fetched, although the other one worked as well
1586 if (!empty($feed_data["header"]["author-link"])) {
1587 $data["url"] = $feed_data["header"]["author-link"];
1590 if ($data["url"] == $data["alias"]) {
1591 $data["alias"] = '';
1594 /// @todo Fetch location and "about" from the feed as well
1599 * Fetch data from a pump.io profile page
1601 * @param string $profile_link Link to the profile page
1603 * @return array profile data
1605 private static function pumpioProfileData($profile_link)
1607 $curlResult = DI::httpRequest()->get($profile_link);
1608 if (!$curlResult->isSuccess()) {
1612 $doc = new DOMDocument();
1613 if (!@$doc->loadHTML($curlResult->getBody())) {
1617 $xpath = new DomXPath($doc);
1621 $data["name"] = $xpath->query("//span[contains(@class, 'p-name')]")->item(0)->nodeValue;
1623 if ($data["name"] == '') {
1624 // This is ugly - but pump.io doesn't seem to know a better way for it
1625 $data["name"] = trim($xpath->query("//h1[@class='media-header']")->item(0)->nodeValue);
1626 $pos = strpos($data["name"], chr(10));
1628 $data["name"] = trim(substr($data["name"], 0, $pos));
1632 $data["location"] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'p-locality')]");
1634 if ($data["location"] == '') {
1635 $data["location"] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'location')]");
1638 $data["about"] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'p-note')]");
1640 if ($data["about"] == '') {
1641 $data["about"] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'summary')]");
1644 $avatar = $xpath->query("//img[contains(@class, 'u-photo')]")->item(0);
1646 $avatar = $xpath->query("//img[@class='img-rounded media-object']")->item(0);
1649 foreach ($avatar->attributes as $attribute) {
1650 if ($attribute->name == "src") {
1651 $data["photo"] = trim($attribute->value);
1660 * Check for pump.io contact
1662 * @param array $webfinger Webfinger data
1663 * @param string $addr
1664 * @return array pump.io data
1666 private static function pumpio($webfinger, $addr)
1669 // The array is reversed to take into account the order of preference for same-rel links
1670 // See: https://tools.ietf.org/html/rfc7033#section-4.4.4
1671 foreach (array_reverse($webfinger["links"]) as $link) {
1672 if (($link["rel"] == "http://webfinger.net/rel/profile-page")
1673 && (($link["type"] ?? "") == "text/html")
1674 && ($link["href"] != "")
1676 $data["url"] = $link["href"];
1677 } elseif (($link["rel"] == "activity-inbox") && ($link["href"] != "")) {
1678 $data["notify"] = $link["href"];
1679 } elseif (($link["rel"] == "activity-outbox") && ($link["href"] != "")) {
1680 $data["poll"] = $link["href"];
1681 } elseif (($link["rel"] == "dialback") && ($link["href"] != "")) {
1682 $data["dialback"] = $link["href"];
1685 if (isset($data["poll"]) && isset($data["notify"])
1686 && isset($data["dialback"])
1687 && isset($data["url"])
1689 // by now we use these fields only for the network type detection
1690 // So we unset all data that isn't used at the moment
1691 unset($data["dialback"]);
1693 $data["network"] = Protocol::PUMPIO;
1698 $profile_data = self::pumpioProfileData($data["url"]);
1700 if (!$profile_data) {
1704 $data = array_merge($data, $profile_data);
1706 if (($addr != '') && ($data['name'] != '')) {
1707 $name = trim(str_replace($addr, '', $data['name']));
1709 $data['name'] = $name;
1717 * Check for twitter contact
1719 * @param string $uri
1721 * @return array twitter data
1723 private static function twitter($uri)
1725 if (preg_match('=([^@]+)@(?:mobile\.)?twitter\.com$=i', $uri, $matches)) {
1726 $nick = $matches[1];
1727 } elseif (preg_match('=^https?://(?:mobile\.)?twitter\.com/(.+)=i', $uri, $matches)) {
1728 $nick = $matches[1];
1734 $data['url'] = 'https://twitter.com/' . $nick;
1735 $data['addr'] = $nick . '@twitter.com';
1736 $data['nick'] = $data['name'] = $nick;
1737 $data['network'] = Protocol::TWITTER;
1738 $data['baseurl'] = 'https://twitter.com';
1744 * Checks HTML page for RSS feed link
1746 * @param string $url Page link
1747 * @param string $body Page body string
1748 * @return string|false Feed link or false if body was invalid HTML document
1750 public static function getFeedLink(string $url, string $body)
1752 $doc = new DOMDocument();
1753 if (!@$doc->loadHTML($body)) {
1757 $xpath = new DOMXPath($doc);
1759 $feedUrl = $xpath->evaluate('string(/html/head/link[@type="application/rss+xml" and @rel="alternate"]/@href)');
1761 $feedUrl = $feedUrl ? self::ensureAbsoluteLinkFromHTMLDoc($feedUrl, $url, $xpath) : '';
1767 * Return an absolute URL in the context of a HTML document retrieved from the provided URL.
1769 * Loosely based on RFC 1808
1771 * @see https://tools.ietf.org/html/rfc1808
1773 * @param string $href The potential relative href found in the HTML document
1774 * @param string $base The HTML document URL
1775 * @param DOMXPath $xpath The HTML document XPath
1778 private static function ensureAbsoluteLinkFromHTMLDoc(string $href, string $base, DOMXPath $xpath)
1780 if (filter_var($href, FILTER_VALIDATE_URL)) {
1784 $base = $xpath->evaluate('string(/html/head/base/@href)') ?: $base;
1786 $baseParts = parse_url($base);
1787 if (empty($baseParts['host'])) {
1791 // Naked domain case (scheme://basehost)
1792 $path = $baseParts['path'] ?? '/';
1794 // Remove the filename part of the path if it exists (/base/path/file)
1795 $path = implode('/', array_slice(explode('/', $path), 0, -1));
1797 $hrefParts = parse_url($href);
1799 // Root path case (/path) including relative scheme case (//host/path)
1800 if ($hrefParts['path'] && $hrefParts['path'][0] == '/') {
1801 $path = $hrefParts['path'];
1803 $path = $path . '/' . $hrefParts['path'];
1805 // Resolve arbitrary relative path
1806 // Lifted from https://www.php.net/manual/en/function.realpath.php#84012
1807 $parts = array_filter(explode('/', $path), 'strlen');
1808 $absolutes = array();
1809 foreach ($parts as $part) {
1810 if ('.' == $part) continue;
1811 if ('..' == $part) {
1812 array_pop($absolutes);
1814 $absolutes[] = $part;
1818 $path = '/' . implode('/', $absolutes);
1821 // Relative scheme case (//host/path)
1822 $baseParts['host'] = $hrefParts['host'] ?? $baseParts['host'];
1823 $baseParts['path'] = $path;
1824 unset($baseParts['query']);
1825 unset($baseParts['fragment']);
1827 return Network::unparseURL($baseParts);
1831 * Check for feed contact
1833 * @param string $url Profile link
1834 * @param boolean $probe Do a probe if the page contains a feed link
1836 * @return array feed data
1837 * @throws HTTPException\InternalServerErrorException
1839 private static function feed($url, $probe = true)
1841 $curlResult = DI::httpRequest()->get($url);
1842 if ($curlResult->isTimeout()) {
1843 self::$istimeout = true;
1846 $feed = $curlResult->getBody();
1847 $feed_data = Feed::import($feed);
1854 $feed_url = self::getFeedLink($url, $feed);
1860 return self::feed($feed_url, false);
1863 if (!empty($feed_data["header"]["author-name"])) {
1864 $data["name"] = $feed_data["header"]["author-name"];
1867 if (!empty($feed_data["header"]["author-nick"])) {
1868 $data["nick"] = $feed_data["header"]["author-nick"];
1871 if (!empty($feed_data["header"]["author-avatar"])) {
1872 $data["photo"] = $feed_data["header"]["author-avatar"];
1875 if (!empty($feed_data["header"]["author-id"])) {
1876 $data["alias"] = $feed_data["header"]["author-id"];
1879 $data["url"] = $url;
1880 $data["poll"] = $url;
1882 $data["network"] = Protocol::FEED;
1888 * Check for mail contact
1890 * @param string $uri Profile link
1891 * @param integer $uid User ID
1893 * @return array mail data
1894 * @throws \Exception
1896 private static function mail($uri, $uid)
1898 if (!Network::isEmailDomainValid($uri)) {
1906 $user = DBA::selectFirst('user', ['prvkey'], ['uid' => $uid]);
1908 $condition = ["`uid` = ? AND `server` != ''", $uid];
1909 $fields = ['pass', 'user', 'server', 'port', 'ssltype', 'mailbox'];
1910 $mailacct = DBA::selectFirst('mailacct', $fields, $condition);
1912 if (!DBA::isResult($user) || !DBA::isResult($mailacct)) {
1916 $mailbox = Email::constructMailboxName($mailacct);
1918 openssl_private_decrypt(hex2bin($mailacct['pass']), $password, $user['prvkey']);
1919 $mbox = Email::connect($mailbox, $mailacct['user'], $password);
1924 $msgs = Email::poll($mbox, $uri);
1925 Logger::info('Messages found', ['uri' => $uri, 'count' => count($msgs)]);
1927 if (!count($msgs)) {
1931 $phost = substr($uri, strpos($uri, '@') + 1);
1934 $data["addr"] = $uri;
1935 $data["network"] = Protocol::MAIL;
1936 $data["name"] = substr($uri, 0, strpos($uri, '@'));
1937 $data["nick"] = $data["name"];
1938 $data["photo"] = Network::lookupAvatarByEmail($uri);
1939 $data["url"] = 'mailto:'.$uri;
1940 $data["notify"] = 'smtp ' . Strings::getRandomHex();
1941 $data["poll"] = 'email ' . Strings::getRandomHex();
1943 $x = Email::messageMeta($mbox, $msgs[0]);
1944 if (stristr($x[0]->from, $uri)) {
1945 $adr = imap_rfc822_parse_adrlist($x[0]->from, '');
1946 } elseif (stristr($x[0]->to, $uri)) {
1947 $adr = imap_rfc822_parse_adrlist($x[0]->to, '');
1950 foreach ($adr as $feadr) {
1951 if ((strcasecmp($feadr->mailbox, $data["name"]) == 0)
1952 &&(strcasecmp($feadr->host, $phost) == 0)
1953 && (strlen($feadr->personal))
1955 $personal = imap_mime_header_decode($feadr->personal);
1957 foreach ($personal as $perspart) {
1958 if ($perspart->charset != "default") {
1959 $data["name"] .= iconv($perspart->charset, 'UTF-8//IGNORE', $perspart->text);
1961 $data["name"] .= $perspart->text;
1965 $data["name"] = Strings::escapeTags($data["name"]);
1969 if (!empty($mbox)) {
1976 * Mix two paths together to possibly fix missing parts
1978 * @param string $avatar Path to the avatar
1979 * @param string $base Another path that is hopefully complete
1981 * @return string fixed avatar path
1982 * @throws \Exception
1984 public static function fixAvatar($avatar, $base)
1986 $base_parts = parse_url($base);
1988 // Remove all parts that could create a problem
1989 unset($base_parts['path']);
1990 unset($base_parts['query']);
1991 unset($base_parts['fragment']);
1993 $avatar_parts = parse_url($avatar);
1996 $parts = array_merge($base_parts, $avatar_parts);
1998 // And put them together again
1999 $scheme = isset($parts['scheme']) ? $parts['scheme'] . '://' : '';
2000 $host = isset($parts['host']) ? $parts['host'] : '';
2001 $port = isset($parts['port']) ? ':' . $parts['port'] : '';
2002 $path = isset($parts['path']) ? $parts['path'] : '';
2003 $query = isset($parts['query']) ? '?' . $parts['query'] : '';
2004 $fragment = isset($parts['fragment']) ? '#' . $parts['fragment'] : '';
2006 $fixed = $scheme.$host.$port.$path.$query.$fragment;
2008 Logger::debug('Avatar fixed', ['base' => $base, 'avatar' => $avatar, 'fixed' => $fixed]);
2014 * Fetch the last date that the contact had posted something (publically)
2016 * @param string $data probing result
2017 * @return string last activity
2019 public static function getLastUpdate(array $data)
2021 $uid = User::getIdForURL($data['url']);
2023 $contact = Contact::selectFirst(['url', 'last-item'], ['self' => true, 'uid' => $uid]);
2024 if (!empty($contact['last-item'])) {
2025 return $contact['last-item'];
2029 if ($lastUpdate = self::updateFromNoScrape($data)) {
2033 if (!empty($data['outbox'])) {
2034 return self::updateFromOutbox($data['outbox'], $data);
2035 } elseif (!empty($data['poll']) && ($data['network'] == Protocol::ACTIVITYPUB)) {
2036 return self::updateFromOutbox($data['poll'], $data);
2037 } elseif (!empty($data['poll'])) {
2038 return self::updateFromFeed($data);
2045 * Fetch the last activity date from the "noscrape" endpoint
2047 * @param array $data Probing result
2048 * @return string last activity
2050 * @return bool 'true' if update was successful or the server was unreachable
2052 private static function updateFromNoScrape(array $data)
2054 if (empty($data['baseurl'])) {
2058 // Check the 'noscrape' endpoint when it is a Friendica server
2059 $gserver = DBA::selectFirst('gserver', ['noscrape'], ["`nurl` = ? AND `noscrape` != ''",
2060 Strings::normaliseLink($data['baseurl'])]);
2061 if (!DBA::isResult($gserver)) {
2065 $curlResult = DI::httpRequest()->get($gserver['noscrape'] . '/' . $data['nick']);
2067 if ($curlResult->isSuccess() && !empty($curlResult->getBody())) {
2068 $noscrape = json_decode($curlResult->getBody(), true);
2069 if (!empty($noscrape) && !empty($noscrape['updated'])) {
2070 return DateTimeFormat::utc($noscrape['updated'], DateTimeFormat::MYSQL);
2078 * Fetch the last activity date from an ActivityPub Outbox
2080 * @param string $feed
2081 * @param array $data Probing result
2082 * @return string last activity
2083 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
2085 private static function updateFromOutbox(string $feed, array $data)
2087 $outbox = ActivityPub::fetchContent($feed);
2088 if (empty($outbox)) {
2092 if (!empty($outbox['orderedItems'])) {
2093 $items = $outbox['orderedItems'];
2094 } elseif (!empty($outbox['first']['orderedItems'])) {
2095 $items = $outbox['first']['orderedItems'];
2096 } elseif (!empty($outbox['first']['href']) && ($outbox['first']['href'] != $feed)) {
2097 return self::updateFromOutbox($outbox['first']['href'], $data);
2098 } elseif (!empty($outbox['first'])) {
2099 if (is_string($outbox['first']) && ($outbox['first'] != $feed)) {
2100 return self::updateFromOutbox($outbox['first'], $data);
2102 Logger::warning('Unexpected data', ['outbox' => $outbox]);
2110 foreach ($items as $activity) {
2111 if (!empty($activity['published'])) {
2112 $published = DateTimeFormat::utc($activity['published']);
2113 } elseif (!empty($activity['object']['published'])) {
2114 $published = DateTimeFormat::utc($activity['object']['published']);
2119 if ($last_updated < $published) {
2120 $last_updated = $published;
2124 if (!empty($last_updated)) {
2125 return $last_updated;
2132 * Fetch the last activity date from an XML feed
2134 * @param array $data Probing result
2135 * @return string last activity
2137 private static function updateFromFeed(array $data)
2139 // Search for the newest entry in the feed
2140 $curlResult = DI::httpRequest()->get($data['poll']);
2141 if (!$curlResult->isSuccess()) {
2145 $doc = new DOMDocument();
2146 @$doc->loadXML($curlResult->getBody());
2148 $xpath = new DOMXPath($doc);
2149 $xpath->registerNamespace('atom', 'http://www.w3.org/2005/Atom');
2151 $entries = $xpath->query('/atom:feed/atom:entry');
2155 foreach ($entries as $entry) {
2156 $published_item = $xpath->query('atom:published/text()', $entry)->item(0);
2157 $updated_item = $xpath->query('atom:updated/text()' , $entry)->item(0);
2158 $published = !empty($published_item->nodeValue) ? DateTimeFormat::utc($published_item->nodeValue) : null;
2159 $updated = !empty($updated_item->nodeValue) ? DateTimeFormat::utc($updated_item->nodeValue) : null;
2161 if (empty($published) || empty($updated)) {
2162 Logger::notice('Invalid entry for XPath.', ['entry' => $entry, 'url' => $data['url']]);
2166 if ($last_updated < $published) {
2167 $last_updated = $published;
2170 if ($last_updated < $updated) {
2171 $last_updated = $updated;
2175 if (!empty($last_updated)) {
2176 return $last_updated;
2183 * Probe data from local profiles without network traffic
2185 * @param string $url
2186 * @return array probed data
2188 private static function localProbe(string $url)
2190 $uid = User::getIdForURL($url);
2195 $profile = User::getOwnerDataById($uid);
2196 if (empty($profile)) {
2200 $approfile = ActivityPub\Transmitter::getProfile($uid);
2201 if (empty($approfile)) {
2205 if (empty($profile['gsid'])) {
2206 $profile['gsid'] = GServer::getID($approfile['generator']['url']);
2209 $data = ['name' => $profile['name'], 'nick' => $profile['nick'], 'guid' => $approfile['diaspora:guid'] ?? '',
2210 'url' => $profile['url'], 'addr' => $profile['addr'], 'alias' => $profile['alias'],
2211 'photo' => $profile['photo'], 'account-type' => $profile['contact-type'],
2212 'community' => ($profile['contact-type'] == User::ACCOUNT_TYPE_COMMUNITY),
2213 'keywords' => $profile['keywords'], 'location' => $profile['location'], 'about' => $profile['about'],
2214 'hide' => !$profile['net-publish'], 'batch' => '', 'notify' => $profile['notify'],
2215 'poll' => $profile['poll'], 'request' => $profile['request'], 'confirm' => $profile['confirm'],
2216 'subscribe' => $approfile['generator']['url'] . '/follow?url={uri}', 'poco' => $profile['poco'],
2217 'following' => $approfile['following'], 'followers' => $approfile['followers'],
2218 'inbox' => $approfile['inbox'], 'outbox' => $approfile['outbox'],
2219 'sharedinbox' => $approfile['endpoints']['sharedInbox'], 'network' => Protocol::DFRN,
2220 'pubkey' => $profile['upubkey'], 'baseurl' => $approfile['generator']['url'], 'gsid' => $profile['gsid'],
2221 'manually-approve' => in_array($profile['page-flags'], [User::PAGE_FLAGS_NORMAL, User::PAGE_FLAGS_PRVGROUP])];
2222 return self::rearrangeData($data);