]> git.mxchange.org Git - friendica.git/blobdiff - src/Network/Probe.php
Revert "Replace IHTTPResult for CurlResult usages"
[friendica.git] / src / Network / Probe.php
index 41682572cd96c8dc16645ddfd92028179b453eb1..3fe035286f3bed729a11b947d06ac3eb551d063b 100644 (file)
@@ -23,7 +23,6 @@ namespace Friendica\Network;
 
 use DOMDocument;
 use DomXPath;
-use Friendica\Core\Cache\Duration;
 use Friendica\Core\Hook;
 use Friendica\Core\Logger;
 use Friendica\Core\Protocol;
@@ -33,11 +32,13 @@ use Friendica\DI;
 use Friendica\Model\Contact;
 use Friendica\Model\GServer;
 use Friendica\Model\Profile;
+use Friendica\Model\User;
 use Friendica\Protocol\ActivityNamespace;
 use Friendica\Protocol\ActivityPub;
 use Friendica\Protocol\Email;
 use Friendica\Protocol\Feed;
 use Friendica\Util\Crypto;
+use Friendica\Util\DateTimeFormat;
 use Friendica\Util\Network;
 use Friendica\Util\Strings;
 use Friendica\Util\XML;
@@ -90,17 +91,19 @@ class Probe
                                "community", "keywords", "location", "about", "hide",
                                "batch", "notify", "poll", "request", "confirm", "subscribe", "poco",
                                "following", "followers", "inbox", "outbox", "sharedinbox",
-                               "priority", "network", "pubkey", "baseurl", "gsid"];
+                               "priority", "network", "pubkey", "manually-approve", "baseurl", "gsid"];
+
+               $numeric_fields = ["gsid", "hide", "account-type", "manually-approve"];
 
                $newdata = [];
                foreach ($fields as $field) {
                        if (isset($data[$field])) {
-                               if (in_array($field, ["gsid", "hide", "account-type"])) {
+                               if (in_array($field, $numeric_fields)) {
                                        $newdata[$field] = (int)$data[$field];
                                } else {        
                                        $newdata[$field] = $data[$field];
                                }
-                       } elseif ($field != "gsid") {
+                       } elseif (!in_array($field, $numeric_fields)) {
                                $newdata[$field] = "";
                        } else {
                                $newdata[$field] = null;
@@ -166,7 +169,7 @@ class Probe
                Logger::info('Probing', ['host' => $host, 'ssl_url' => $ssl_url, 'url' => $url, 'callstack' => System::callstack(20)]);
                $xrd = null;
 
-               $curlResult = Network::curl($ssl_url, false, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']);
+               $curlResult = DI::httpRequest()->get($ssl_url, false, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']);
                $ssl_connection_error = ($curlResult->getErrorNumber() == CURLE_COULDNT_CONNECT) || ($curlResult->getReturnCode() == 0);
                if ($curlResult->isSuccess()) {
                        $xml = $curlResult->getBody();
@@ -177,16 +180,16 @@ class Probe
                                $host_url = $host;
                        }
                } elseif ($curlResult->isTimeout()) {
-                       Logger::info('Probing timeout', ['url' => $ssl_url], Logger::DEBUG);
+                       Logger::info('Probing timeout', ['url' => $ssl_url]);
                        self::$istimeout = true;
                        return [];
                }
 
                if (!is_object($xrd) && !empty($url)) {
-                       $curlResult = Network::curl($url, false, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']);
+                       $curlResult = DI::httpRequest()->get($url, false, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']);
                        $connection_error = ($curlResult->getErrorNumber() == CURLE_COULDNT_CONNECT) || ($curlResult->getReturnCode() == 0);
                        if ($curlResult->isTimeout()) {
-                               Logger::info('Probing timeout', ['url' => $url], Logger::DEBUG);
+                               Logger::info('Probing timeout', ['url' => $url]);
                                self::$istimeout = true;
                                return [];
                        } elseif ($connection_error && $ssl_connection_error) {
@@ -199,13 +202,13 @@ class Probe
                        $host_url = 'http://'.$host;
                }
                if (!is_object($xrd)) {
-                       Logger::log("No xrd object found for ".$host, Logger::DEBUG);
+                       Logger::info('No xrd object found', ['host' => $host]);
                        return [];
                }
 
                $links = XML::elementToArray($xrd);
                if (!isset($links["xrd"]["link"])) {
-                       Logger::log("No xrd data found for ".$host, Logger::DEBUG);
+                       Logger::info('No xrd data found', ['host' => $host]);
                        return [];
                }
 
@@ -229,7 +232,7 @@ class Probe
 
                self::$baseurl = $host_url;
 
-               Logger::log("Probing successful for ".$host, Logger::DEBUG);
+               Logger::info('Probing successful', ['host' => $host]);
 
                return $lrdd;
        }
@@ -260,7 +263,7 @@ class Probe
                $profile_link = '';
 
                $links = self::lrdd($webbie);
-               Logger::log('webfingerDfrn: '.$webbie.':'.print_r($links, true), Logger::DATA);
+               Logger::debug('Result', ['url' => $webbie, 'links' => $links]);
                if (!empty($links) && is_array($links)) {
                        foreach ($links as $link) {
                                if ($link['@attributes']['rel'] === ActivityNamespace::DFRN) {
@@ -294,7 +297,7 @@ class Probe
                $webfinger = $data['webfinger'];
 
                if (empty($webfinger["links"])) {
-                       Logger::log("No webfinger links found for ".$uri, Logger::DEBUG);
+                       Logger::info('No webfinger links found', ['uri' => $uri]);
                        return [];
                }
 
@@ -327,13 +330,13 @@ class Probe
         * @throws HTTPException\InternalServerErrorException
         * @throws \ImagickException
         */
-       public static function uri($uri, $network = '', $uid = -1, $cache = true)
+       public static function uri($uri, $network = '', $uid = -1)
        {
-               $cachekey = 'Probe::uri:' . $network . ':' . $uri;
-               if ($cache) {
-                       $result = DI::cache()->get($cachekey);
-                       if (!is_null($result)) {
-                               return $result;
+               // Local profiles aren't probed via network
+               if (empty($network) && strpos($uri, DI::baseUrl()->getHostname())) {
+                       $data = self::localProbe($uri);
+                       if (!empty($data)) {
+                               return $data;
                        }
                }
 
@@ -342,7 +345,7 @@ class Probe
                }
 
                if (empty($network) || ($network == Protocol::ACTIVITYPUB)) {
-                       $ap_profile = ActivityPub::probeProfile($uri, !$cache);
+                       $ap_profile = ActivityPub::probeProfile($uri);
                } else {
                        $ap_profile = [];
                }
@@ -369,7 +372,7 @@ class Probe
                }
 
                if (empty($data['photo'])) {
-                       $data['photo'] = DI::baseUrl() . '/images/person-300.jpg';
+                       $data['photo'] = DI::baseUrl() . Contact::DEFAULT_AVATAR_PHOTO;
                }
 
                if (empty($data['name'])) {
@@ -407,14 +410,7 @@ class Probe
                        $data['hide'] = self::getHideStatus($data['url']);
                }
 
-               $data = self::rearrangeData($data);
-
-               // Only store into the cache if the value seems to be valid
-               if (!in_array($data['network'], [Protocol::PHANTOM, Protocol::MAIL])) {
-                       DI::cache()->set($cachekey, $data, Duration::DAY);
-               }
-
-               return $data;
+               return self::rearrangeData($data);
        }
 
 
@@ -427,7 +423,7 @@ class Probe
         */
        private static function getHideStatus($url)
        {
-               $curlResult = Network::curl($url);
+               $curlResult = DI::httpRequest()->get($url);
                if (!$curlResult->isSuccess()) {
                        return false;
                }
@@ -718,7 +714,14 @@ class Probe
 
                Logger::info('Probing start', ['uri' => $uri]);
 
-               $data = self::getWebfingerArray($uri);
+               if (!empty($ap_profile['addr']) && ($ap_profile['addr'] != $uri)) {
+                       $data = self::getWebfingerArray($ap_profile['addr']);
+               }
+
+               if (empty($data)) {
+                       $data = self::getWebfingerArray($uri);
+               }
+
                if (empty($data)) {
                        if (!empty($parts['scheme'])) {
                                return self::feed($uri);
@@ -841,7 +844,7 @@ class Probe
 
        public static function pollZot($url, $data)
        {
-               $curlResult = Network::curl($url);
+               $curlResult = DI::httpRequest()->get($url);
                if ($curlResult->isTimeout()) {
                        return $data;
                }
@@ -892,7 +895,7 @@ class Probe
                }
                if (!empty($json['public_forum'])) {
                        $data['community'] = $json['public_forum'];
-                       $data['account-type'] = Contact::PAGE_COMMUNITY;
+                       $data['account-type'] = User::PAGE_FLAGS_COMMUNITY;
                }
 
                if (!empty($json['profile'])) {
@@ -938,7 +941,7 @@ class Probe
        {
                $xrd_timeout = DI::config()->get('system', 'xrd_timeout', 20);
 
-               $curlResult = Network::curl($url, false, ['timeout' => $xrd_timeout, 'accept_content' => $type]);
+               $curlResult = DI::httpRequest()->get($url, false, ['timeout' => $xrd_timeout, 'accept_content' => $type]);
                if ($curlResult->isTimeout()) {
                        self::$istimeout = true;
                        return [];
@@ -948,7 +951,7 @@ class Probe
                $webfinger = json_decode($data, true);
                if (!empty($webfinger)) {
                        if (!isset($webfinger["links"])) {
-                               Logger::log("No json webfinger links for ".$url, Logger::DEBUG);
+                               Logger::info('No json webfinger links', ['url' => $url]);
                                return [];
                        }
                        return $webfinger;
@@ -957,13 +960,13 @@ class Probe
                // If it is not JSON, maybe it is XML
                $xrd = XML::parseString($data, true);
                if (!is_object($xrd)) {
-                       Logger::log("No webfinger data retrievable for ".$url, Logger::DEBUG);
+                       Logger::info('No webfinger data retrievable', ['url' => $url]);
                        return [];
                }
 
                $xrd_arr = XML::elementToArray($xrd);
                if (!isset($xrd_arr["xrd"]["link"])) {
-                       Logger::log("No XML webfinger links for ".$url, Logger::DEBUG);
+                       Logger::info('No XML webfinger links', ['url' => $url]);
                        return [];
                }
 
@@ -1007,20 +1010,20 @@ class Probe
         */
        private static function pollNoscrape($noscrape_url, $data)
        {
-               $curlResult = Network::curl($noscrape_url);
+               $curlResult = DI::httpRequest()->get($noscrape_url);
                if ($curlResult->isTimeout()) {
                        self::$istimeout = true;
                        return [];
                }
                $content = $curlResult->getBody();
                if (!$content) {
-                       Logger::log("Empty body for ".$noscrape_url, Logger::DEBUG);
+                       Logger::info('Empty body', ['url' => $noscrape_url]);
                        return [];
                }
 
                $json = json_decode($content, true);
                if (!is_array($json)) {
-                       Logger::log("No json data for ".$noscrape_url, Logger::DEBUG);
+                       Logger::info('No json data', ['url' => $noscrape_url]);
                        return [];
                }
 
@@ -1134,7 +1137,7 @@ class Probe
        {
                $data = [];
 
-               Logger::log("Check profile ".$profile_link, Logger::DEBUG);
+               Logger::info('Check profile', ['link' => $profile_link]);
 
                // Fetch data via noscrape - this is faster
                $noscrape_url = str_replace(["/hcard/", "/profile/"], "/noscrape/", $profile_link);
@@ -1168,7 +1171,7 @@ class Probe
                $prof_data["fn"]           = $data['name']    ?? null;
                $prof_data["key"]          = $data['pubkey']  ?? null;
 
-               Logger::log("Result for profile ".$profile_link.": ".print_r($prof_data, true), Logger::DEBUG);
+               Logger::debug('Result', ['link' => $profile_link, 'data' => $prof_data]);
 
                return $prof_data;
        }
@@ -1265,7 +1268,7 @@ class Probe
         */
        private static function pollHcard($hcard_url, $data, $dfrn = false)
        {
-               $curlResult = Network::curl($hcard_url);
+               $curlResult = DI::httpRequest()->get($hcard_url);
                if ($curlResult->isTimeout()) {
                        self::$istimeout = true;
                        return [];
@@ -1453,6 +1456,7 @@ class Probe
                        && !empty($hcard_url)
                ) {
                        $data["network"] = Protocol::DIASPORA;
+                       $data["manually-approve"] = false;
 
                        // The Diaspora handle must always be lowercase
                        if (!empty($data["addr"])) {
@@ -1519,7 +1523,7 @@ class Probe
                                                        $pubkey = substr($pubkey, 5);
                                                }
                                        } elseif (Strings::normaliseLink($pubkey) == 'http://') {
-                                               $curlResult = Network::curl($pubkey);
+                                               $curlResult = DI::httpRequest()->get($pubkey);
                                                if ($curlResult->isTimeout()) {
                                                        self::$istimeout = true;
                                                        return $short ? false : [];
@@ -1543,6 +1547,7 @@ class Probe
                        && isset($data["url"])
                ) {
                        $data["network"] = Protocol::OSTATUS;
+                       $data["manually-approve"] = false;
                } else {
                        return $short ? false : [];
                }
@@ -1552,7 +1557,7 @@ class Probe
                }
 
                // Fetch all additional data from the feed
-               $curlResult = Network::curl($data["poll"]);
+               $curlResult = DI::httpRequest()->get($data["poll"]);
                if ($curlResult->isTimeout()) {
                        self::$istimeout = true;
                        return [];
@@ -1604,7 +1609,7 @@ class Probe
         */
        private static function pumpioProfileData($profile_link)
        {
-               $curlResult = Network::curl($profile_link);
+               $curlResult = DI::httpRequest()->get($profile_link);
                if (!$curlResult->isSuccess()) {
                        return [];
                }
@@ -1784,6 +1789,9 @@ class Probe
                $base = $xpath->evaluate('string(/html/head/base/@href)') ?: $base;
 
                $baseParts = parse_url($base);
+               if (empty($baseParts['host'])) {
+                       return $href;
+               }
 
                // Naked domain case (scheme://basehost)
                $path = $baseParts['path'] ?? '/';
@@ -1835,7 +1843,7 @@ class Probe
         */
        private static function feed($url, $probe = true)
        {
-               $curlResult = Network::curl($url);
+               $curlResult = DI::httpRequest()->get($url);
                if ($curlResult->isTimeout()) {
                        self::$istimeout = true;
                        return [];
@@ -1919,7 +1927,7 @@ class Probe
                }
 
                $msgs = Email::poll($mbox, $uri);
-               Logger::log('searching '.$uri.', '.count($msgs).' messages found.', Logger::DEBUG);
+               Logger::info('Messages found', ['uri' => $uri, 'count' => count($msgs)]);
 
                if (!count($msgs)) {
                        return [];
@@ -2002,8 +2010,220 @@ class Probe
 
                $fixed = $scheme.$host.$port.$path.$query.$fragment;
 
-               Logger::log('Base: '.$base.' - Avatar: '.$avatar.' - Fixed: '.$fixed, Logger::DATA);
+               Logger::debug('Avatar fixed', ['base' => $base, 'avatar' => $avatar, 'fixed' => $fixed]);
 
                return $fixed;
        }
+
+       /**
+        * Fetch the last date that the contact had posted something (publically)
+        *
+        * @param string $data  probing result
+        * @return string last activity
+        */
+       public static function getLastUpdate(array $data)
+       {
+               $uid = User::getIdForURL($data['url']);
+               if (!empty($uid)) {
+                       $contact = Contact::selectFirst(['url', 'last-item'], ['self' => true, 'uid' => $uid]);
+                       if (!empty($contact['last-item'])) {
+                               return $contact['last-item'];
+                       }
+               }
+
+               if ($lastUpdate = self::updateFromNoScrape($data)) {
+                       return $lastUpdate;
+               }
+
+               if (!empty($data['outbox'])) {
+                       return self::updateFromOutbox($data['outbox'], $data);
+               } elseif (!empty($data['poll']) && ($data['network'] == Protocol::ACTIVITYPUB)) {
+                       return self::updateFromOutbox($data['poll'], $data);
+               } elseif (!empty($data['poll'])) {
+                       return self::updateFromFeed($data);
+               }
+
+               return '';
+       }
+
+       /**
+        * Fetch the last activity date from the "noscrape" endpoint
+        *
+        * @param array $data Probing result
+        * @return string last activity
+        *
+        * @return bool 'true' if update was successful or the server was unreachable
+        */
+       private static function updateFromNoScrape(array $data)
+       {
+               if (empty($data['baseurl'])) {
+                       return '';
+               }
+
+               // Check the 'noscrape' endpoint when it is a Friendica server
+               $gserver = DBA::selectFirst('gserver', ['noscrape'], ["`nurl` = ? AND `noscrape` != ''",
+                       Strings::normaliseLink($data['baseurl'])]);
+               if (!DBA::isResult($gserver)) {
+                       return '';
+               }
+
+               $curlResult = DI::httpRequest()->get($gserver['noscrape'] . '/' . $data['nick']);
+
+               if ($curlResult->isSuccess() && !empty($curlResult->getBody())) {
+                       $noscrape = json_decode($curlResult->getBody(), true);
+                       if (!empty($noscrape) && !empty($noscrape['updated'])) {
+                               return DateTimeFormat::utc($noscrape['updated'], DateTimeFormat::MYSQL);
+                       }
+               }
+
+               return '';
+       }
+
+       /**
+        * Fetch the last activity date from an ActivityPub Outbox
+        *
+        * @param string $feed
+        * @param array  $data Probing result
+        * @return string last activity
+        * @throws \Friendica\Network\HTTPException\InternalServerErrorException
+        */
+       private static function updateFromOutbox(string $feed, array $data)
+       {
+               $outbox = ActivityPub::fetchContent($feed);
+               if (empty($outbox)) {
+                       return '';
+               }
+
+               if (!empty($outbox['orderedItems'])) {
+                       $items = $outbox['orderedItems'];
+               } elseif (!empty($outbox['first']['orderedItems'])) {
+                       $items = $outbox['first']['orderedItems'];
+               } elseif (!empty($outbox['first']['href']) && ($outbox['first']['href'] != $feed)) {
+                       return self::updateFromOutbox($outbox['first']['href'], $data);
+               } elseif (!empty($outbox['first'])) {
+                       if (is_string($outbox['first']) && ($outbox['first'] != $feed)) {
+                               return self::updateFromOutbox($outbox['first'], $data);
+                       } else {
+                               Logger::warning('Unexpected data', ['outbox' => $outbox]);
+                       }
+                       return '';
+               } else {
+                       $items = [];
+               }
+
+               $last_updated = '';
+               foreach ($items as $activity) {
+                       if (!empty($activity['published'])) {
+                               $published =  DateTimeFormat::utc($activity['published']);
+                       } elseif (!empty($activity['object']['published'])) {
+                               $published =  DateTimeFormat::utc($activity['object']['published']);
+                       } else {
+                               continue;
+                       }
+
+                       if ($last_updated < $published) {
+                               $last_updated = $published;
+                       }
+               }
+
+               if (!empty($last_updated)) {
+                       return $last_updated;
+               }
+
+               return '';
+       }
+
+       /**
+        * Fetch the last activity date from an XML feed
+        *
+        * @param array $data Probing result
+        * @return string last activity
+        */
+       private static function updateFromFeed(array $data)
+       {
+               // Search for the newest entry in the feed
+               $curlResult = DI::httpRequest()->get($data['poll']);
+               if (!$curlResult->isSuccess()) {
+                       return '';
+               }
+
+               $doc = new DOMDocument();
+               @$doc->loadXML($curlResult->getBody());
+
+               $xpath = new DOMXPath($doc);
+               $xpath->registerNamespace('atom', 'http://www.w3.org/2005/Atom');
+
+               $entries = $xpath->query('/atom:feed/atom:entry');
+
+               $last_updated = '';
+
+               foreach ($entries as $entry) {
+                       $published_item = $xpath->query('atom:published/text()', $entry)->item(0);
+                       $updated_item   = $xpath->query('atom:updated/text()'  , $entry)->item(0);
+                       $published      = !empty($published_item->nodeValue) ? DateTimeFormat::utc($published_item->nodeValue) : null;
+                       $updated        = !empty($updated_item->nodeValue) ? DateTimeFormat::utc($updated_item->nodeValue) : null;
+
+                       if (empty($published) || empty($updated)) {
+                               Logger::notice('Invalid entry for XPath.', ['entry' => $entry, 'url' => $data['url']]);
+                               continue;
+                       }
+
+                       if ($last_updated < $published) {
+                               $last_updated = $published;
+                       }
+
+                       if ($last_updated < $updated) {
+                               $last_updated = $updated;
+                       }
+               }
+
+               if (!empty($last_updated)) {
+                       return $last_updated;
+               }
+
+               return '';
+       }
+
+       /**
+        * Probe data from local profiles without network traffic
+        *
+        * @param string $url
+        * @return array probed data
+        */
+       private static function localProbe(string $url)
+       {
+               $uid = User::getIdForURL($url);
+               if (empty($uid)) {
+                       return [];
+               }
+
+               $profile = User::getOwnerDataById($uid);
+               if (empty($profile)) {
+                       return [];
+               }
+
+               $approfile = ActivityPub\Transmitter::getProfile($uid);
+               if (empty($approfile)) {
+                       return [];
+               }
+
+               if (empty($profile['gsid'])) {
+                       $profile['gsid'] = GServer::getID($approfile['generator']['url']);
+               }
+
+               $data = ['name' => $profile['name'], 'nick' => $profile['nick'], 'guid' => $approfile['diaspora:guid'] ?? '',
+                       'url' => $profile['url'], 'addr' => $profile['addr'], 'alias' => $profile['alias'],
+                       'photo' => $profile['photo'], 'account-type' => $profile['contact-type'],
+                       'community' => ($profile['contact-type'] == User::ACCOUNT_TYPE_COMMUNITY),
+                       'keywords' => $profile['keywords'], 'location' => $profile['location'], 'about' => $profile['about'], 
+                       'hide' => !$profile['net-publish'], 'batch' => '', 'notify' => $profile['notify'],
+                       'poll' => $profile['poll'], 'request' => $profile['request'], 'confirm' => $profile['confirm'],
+                       'subscribe' => $approfile['generator']['url'] . '/follow?url={uri}', 'poco' => $profile['poco'], 
+                       'following' => $approfile['following'], 'followers' => $approfile['followers'],
+                       'inbox' => $approfile['inbox'], 'outbox' => $approfile['outbox'],
+                       'sharedinbox' => $approfile['endpoints']['sharedInbox'], 'network' => Protocol::DFRN, 
+                       'pubkey' => $profile['upubkey'], 'baseurl' => $approfile['generator']['url'], 'gsid' => $profile['gsid'],
+                       'manually-approve' => in_array($profile['page-flags'], [User::PAGE_FLAGS_NORMAL, User::PAGE_FLAGS_PRVGROUP])];
+               return self::rearrangeData($data);              
+       }
 }