X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FNetwork%2FProbe.php;h=f4e3d1d43c55d433f875b840a781dac3171bceb4;hb=491fc5f78aadb787a57a0f0630779f457ff1bcb4;hp=539803b6e4d93acfc56ea918f5c2bc59f3ff75f4;hpb=a86ffd878dc3ca8638345781bda0f18071d439e5;p=friendica.git diff --git a/src/Network/Probe.php b/src/Network/Probe.php index 539803b6e4..f4e3d1d43c 100644 --- a/src/Network/Probe.php +++ b/src/Network/Probe.php @@ -9,23 +9,26 @@ namespace Friendica\Network; * @brief Functions for probing URL */ -use Friendica\App; -use Friendica\Core\System; +use DOMDocument; use Friendica\Core\Cache; use Friendica\Core\Config; -use Friendica\Database\DBM; +use Friendica\Core\Logger; +use Friendica\Core\Protocol; +use Friendica\Core\System; +use Friendica\Database\DBA; +use Friendica\Model\Contact; use Friendica\Model\Profile; use Friendica\Protocol\Email; use Friendica\Protocol\Feed; +use Friendica\Protocol\ActivityPub; use Friendica\Util\Crypto; +use Friendica\Util\DateTimeFormat; +use Friendica\Util\Network; +use Friendica\Util\Strings; use Friendica\Util\XML; - -use dba; -use DOMXPath; -use DOMDocument; +use DomXPath; require_once 'include/dba.php'; -require_once 'include/network.php'; /** * @brief This class contain functions for probing URL @@ -44,12 +47,12 @@ class Probe */ private static function rearrangeData($data) { - $fields = array("name", "nick", "guid", "url", "addr", "alias", + $fields = ["name", "nick", "guid", "url", "addr", "alias", "photo", "community", "keywords", "location", "about", "batch", "notify", "poll", "request", "confirm", "poco", - "priority", "network", "pubkey", "baseurl"); + "priority", "network", "pubkey", "baseurl"]; - $newdata = array(); + $newdata = []; foreach ($fields as $field) { if (isset($data[$field])) { $newdata[$field] = $data[$field]; @@ -73,7 +76,7 @@ class Probe */ private static function ownHost($host) { - $own_host = get_app()->get_hostname(); + $own_host = get_app()->getHostName(); $parts = parse_url($host); @@ -90,6 +93,9 @@ class Probe /** * @brief Probes for webfinger path via "host-meta" * + * We have to check if the servers in the future still will offer this. + * It seems as if it was dropped from the standard. + * * @param string $host The host part of an url * * @return array with template and type of the webfinger template for JSON or XML @@ -105,37 +111,38 @@ class Probe $xrd_timeout = Config::get('system', 'xrd_timeout', 20); $redirects = 0; - logger("Probing for ".$host, LOGGER_DEBUG); + Logger::log("Probing for ".$host, Logger::DEBUG); + $xrd = null; - $ret = z_fetch_url($ssl_url, false, $redirects, array('timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml')); - if ($ret['success']) { - $xml = $ret['body']; - $xrd = parse_xml_string($xml, false); + $curlResult = Network::curl($ssl_url, false, $redirects, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']); + if ($curlResult->isSuccess()) { + $xml = $curlResult->getBody(); + $xrd = XML::parseString($xml, false); $host_url = 'https://'.$host; } if (!is_object($xrd)) { - $ret = z_fetch_url($url, false, $redirects, array('timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml')); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { - logger("Probing timeout for ".$url, LOGGER_DEBUG); + $curlResult = Network::curl($url, false, $redirects, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']); + if ($curlResult->isTimeout()) { + Logger::log("Probing timeout for " . $url, Logger::DEBUG); return false; } - $xml = $ret['body']; - $xrd = parse_xml_string($xml, false); + $xml = $curlResult->getBody(); + $xrd = XML::parseString($xml, false); $host_url = 'http://'.$host; } if (!is_object($xrd)) { - logger("No xrd object found for ".$host, LOGGER_DEBUG); - return array(); + Logger::log("No xrd object found for ".$host, Logger::DEBUG); + return []; } $links = XML::elementToArray($xrd); if (!isset($links["xrd"]["link"])) { - logger("No xrd data found for ".$host, LOGGER_DEBUG); - return array(); + Logger::log("No xrd data found for ".$host, Logger::DEBUG); + return []; } - $lrdd = array(); + $lrdd = []; // The following webfinger path is defined in RFC 7033 https://tools.ietf.org/html/rfc7033 // Problem is that Hubzilla currently doesn't provide all data in the JSON webfinger // compared to the XML webfinger. So this is commented out by now. @@ -157,9 +164,9 @@ class Probe } } - self::$baseurl = "http://".$host; + self::$baseurl = $host_url; - logger("Probing successful for ".$host, LOGGER_DEBUG); + Logger::log("Probing successful for ".$host, Logger::DEBUG); return $lrdd; } @@ -189,7 +196,7 @@ class Probe $profile_link = ''; $links = self::lrdd($webbie); - logger('webfingerDfrn: '.$webbie.':'.print_r($links, true), LOGGER_DATA); + Logger::log('webfingerDfrn: '.$webbie.':'.print_r($links, true), Logger::DATA); if (count($links)) { foreach ($links as $link) { if ($link['@attributes']['rel'] === NAMESPACE_DFRN) { @@ -209,7 +216,7 @@ class Probe /** * @brief Check an URI for LRDD data * - * this is a replacement for the "lrdd" function in include/network.php. + * this is a replacement for the "lrdd" function. * It isn't used in this class and has some redundancies in the code. * When time comes we can check the existing calls for "lrdd" if we can rework them. * @@ -223,13 +230,13 @@ class Probe $webfinger = null; if (is_bool($lrdd)) { - return array(); + return []; } if (!$lrdd) { $parts = @parse_url($uri); - if (!$parts) { - return array(); + if (!$parts || empty($parts["host"]) || empty($parts["path"])) { + return []; } $host = $parts["host"]; @@ -248,8 +255,8 @@ class Probe } if (!$lrdd) { - logger("No lrdd data found for ".$uri, LOGGER_DEBUG); - return array(); + Logger::log("No lrdd data found for ".$uri, Logger::DEBUG); + return []; } foreach ($lrdd as $type => $template) { @@ -280,21 +287,21 @@ class Probe } if (!is_array($webfinger["links"])) { - logger("No webfinger links found for ".$uri, LOGGER_DEBUG); + Logger::log("No webfinger links found for ".$uri, Logger::DEBUG); return false; } - $data = array(); + $data = []; foreach ($webfinger["links"] as $link) { - $data[] = array("@attributes" => $link); + $data[] = ["@attributes" => $link]; } if (is_array($webfinger["aliases"])) { foreach ($webfinger["aliases"] as $alias) { - $data[] = array("@attributes" => - array("rel" => "alias", - "href" => $alias)); + $data[] = ["@attributes" => + ["rel" => "alias", + "href" => $alias]]; } } @@ -311,10 +318,10 @@ class Probe * * @return array uri data */ - public static function uri($uri, $network = "", $uid = -1, $cache = true) + public static function uri($uri, $network = '', $uid = -1, $cache = true) { if ($cache) { - $result = Cache::get("Probe::uri:".$network.":".$uri); + $result = Cache::get('Probe::uri:' . $network . ':' . $uri); if (!is_null($result)) { return $result; } @@ -324,75 +331,93 @@ class Probe $uid = local_user(); } - $data = self::detect($uri, $network, $uid); + if ($network != Protocol::ACTIVITYPUB) { + $data = self::detect($uri, $network, $uid); + } else { + $data = null; + } + + $ap_profile = ActivityPub::probeProfile($uri); - if (!isset($data["url"])) { - $data["url"] = $uri; + if (!empty($ap_profile) && (defaults($data, 'network', '') != Protocol::DFRN)) { + $data = $ap_profile; } - if ($data["photo"] != "") { - $data["baseurl"] = matching_url(normalise_link($data["baseurl"]), normalise_link($data["photo"])); + if (!isset($data['url'])) { + $data['url'] = $uri; + } + + if (!empty($data['photo'])) { + $data['baseurl'] = Network::getUrlMatch(Strings::normaliseLink(defaults($data, 'baseurl', '')), Strings::normaliseLink($data['photo'])); } else { - $data["photo"] = System::baseUrl().'/images/person-175.jpg'; + $data['photo'] = System::baseUrl() . '/images/person-300.jpg'; } - if (empty($data["name"])) { - if (!empty($data["nick"])) { - $data["name"] = $data["nick"]; + if (empty($data['name'])) { + if (!empty($data['nick'])) { + $data['name'] = $data['nick']; } - if ($data["name"] == "") { - $data["name"] = $data["url"]; + if (empty($data['name'])) { + $data['name'] = $data['url']; } } - if (empty($data["nick"])) { - $data["nick"] = strtolower($data["name"]); + if (empty($data['nick'])) { + $data['nick'] = strtolower($data['name']); if (strpos($data['nick'], ' ')) { $data['nick'] = trim(substr($data['nick'], 0, strpos($data['nick'], ' '))); } } - if (self::$baseurl != "") { - $data["baseurl"] = self::$baseurl; + if (!empty(self::$baseurl)) { + $data['baseurl'] = self::$baseurl; } - if (!isset($data["network"])) { - $data["network"] = NETWORK_PHANTOM; + if (empty($data['network'])) { + $data['network'] = Protocol::PHANTOM; } $data = self::rearrangeData($data); // Only store into the cache if the value seems to be valid - if (!in_array($data['network'], array(NETWORK_PHANTOM, NETWORK_MAIL))) { - Cache::set("Probe::uri:".$network.":".$uri, $data, CACHE_DAY); + if (!in_array($data['network'], [Protocol::PHANTOM, Protocol::MAIL])) { + Cache::set('Probe::uri:' . $network . ':' . $uri, $data, Cache::DAY); /// @todo temporary fix - we need a real contact update function that updates only changing fields /// The biggest problem is the avatar picture that could have a reduced image size. /// It should only be updated if the existing picture isn't existing anymore. /// We only update the contact when it is no probing for a specific network. - if (($data['network'] != NETWORK_FEED) - && ($network == "") - && $data["name"] - && $data["nick"] - && $data["url"] - && $data["addr"] - && $data["poll"] + if (($data['network'] != Protocol::FEED) + && ($network == '') + && $data['name'] + && $data['nick'] + && $data['url'] + && $data['addr'] + && $data['poll'] ) { - $fields = array('name' => $data['name'], - 'nick' => $data['nick'], - 'url' => $data['url'], - 'addr' => $data['addr'], - 'photo' => $data['photo'], - 'keywords' => $data['keywords'], - 'location' => $data['location'], - 'about' => $data['about'], - 'notify' => $data['notify'], - 'network' => $data['network'], - 'server_url' => $data['baseurl']); - - $fieldnames = array(); + $fields = [ + 'name' => $data['name'], + 'nick' => $data['nick'], + 'url' => $data['url'], + 'addr' => $data['addr'], + 'photo' => $data['photo'], + 'keywords' => $data['keywords'], + 'location' => $data['location'], + 'about' => $data['about'], + 'notify' => $data['notify'], + 'network' => $data['network'], + 'server_url' => $data['baseurl'] + ]; + + // This doesn't cover the case when a community isn't a community anymore + if (!empty($data['community']) && $data['community']) { + $fields['community'] = $data['community']; + $fields['contact-type'] = Contact::ACCOUNT_TYPE_COMMUNITY; + } + + $fieldnames = []; foreach ($fields as $key => $val) { if (empty($val)) { @@ -402,32 +427,48 @@ class Probe } } - $fields['updated'] = DBM::date(); + $fields['updated'] = DateTimeFormat::utcNow(); - $condition = array('nurl' => normalise_link($data["url"])); + $condition = ['nurl' => Strings::normaliseLink($data['url'])]; - $old_fields = dba::select('gcontact', $fieldnames, $condition, array('limit' => 1)); + $old_fields = DBA::selectFirst('gcontact', $fieldnames, $condition); - dba::update('gcontact', $fields, $condition, $old_fields); + // When the gcontact doesn't exist, the value "true" will trigger an insert. + // In difference to the public contacts we want to have every contact + // in the world in our global contacts. + if (!$old_fields) { + $old_fields = true; - $fields = array('name' => $data['name'], - 'nick' => $data['nick'], - 'url' => $data['url'], - 'addr' => $data['addr'], - 'alias' => $data['alias'], - 'keywords' => $data['keywords'], - 'location' => $data['location'], - 'about' => $data['about'], - 'batch' => $data['batch'], - 'notify' => $data['notify'], - 'poll' => $data['poll'], - 'request' => $data['request'], - 'confirm' => $data['confirm'], - 'poco' => $data['poco'], - 'network' => $data['network'], - 'success_update' => DBM::date()); + // These values have to be set only on insert + $fields['photo'] = $data['photo']; + $fields['created'] = DateTimeFormat::utcNow(); + } - $fieldnames = array(); + DBA::update('gcontact', $fields, $condition, $old_fields); + + $fields = [ + 'name' => $data['name'], + 'nick' => $data['nick'], + 'url' => $data['url'], + 'addr' => $data['addr'], + 'alias' => $data['alias'], + 'keywords' => $data['keywords'], + 'location' => $data['location'], + 'about' => $data['about'], + 'batch' => $data['batch'], + 'notify' => $data['notify'], + 'poll' => $data['poll'], + 'request' => $data['request'], + 'confirm' => $data['confirm'], + 'poco' => $data['poco'], + 'network' => $data['network'], + 'pubkey' => $data['pubkey'], + 'priority' => $data['priority'], + 'writable' => true, + 'rel' => Contact::SHARING + ]; + + $fieldnames = []; foreach ($fields as $key => $val) { if (empty($val)) { @@ -437,11 +478,19 @@ class Probe } } - $condition = array('nurl' => normalise_link($data["url"]), 'self' => false, 'uid' => 0); + $condition = ['nurl' => Strings::normaliseLink($data['url']), 'self' => false, 'uid' => 0]; - $old_fields = dba::select('contact', $fieldnames, $condition, array('limit' => 1)); + // "$old_fields" will return a "false" when the contact doesn't exist. + // This won't trigger an insert. This is intended, since we only need + // public contacts for everyone we store items from. + // We don't need to store every contact on the planet. + $old_fields = DBA::selectFirst('contact', $fieldnames, $condition); - dba::update('contact', $fields, $condition, $old_fields); + $fields['name-date'] = DateTimeFormat::utcNow(); + $fields['uri-date'] = DateTimeFormat::utcNow(); + $fields['success_update'] = DateTimeFormat::utcNow(); + + DBA::update('contact', $fields, $condition, $old_fields); } } @@ -530,29 +579,29 @@ class Probe { $parts = parse_url($uri); - if (!empty($parts["scheme"]) && !empty($parts["host"]) && !empty($parts["path"])) { + if (!empty($parts["scheme"]) && !empty($parts["host"])) { $host = $parts["host"]; if (!empty($parts["port"])) { $host .= ':'.$parts["port"]; } if ($host == 'twitter.com') { - return array("network" => NETWORK_TWITTER); + return ["network" => Protocol::TWITTER]; } $lrdd = self::hostMeta($host); if (is_bool($lrdd)) { - return array(); + return []; } - $path_parts = explode("/", trim($parts["path"], "/")); + $path_parts = explode("/", trim(defaults($parts, 'path', ''), "/")); while (!$lrdd && (sizeof($path_parts) > 1)) { $host .= "/".array_shift($path_parts); $lrdd = self::hostMeta($host); } if (!$lrdd) { - logger('No XRD data was found for '.$uri, LOGGER_DEBUG); + Logger::log('No XRD data was found for '.$uri, Logger::DEBUG); return self::feed($uri); } $nick = array_pop($path_parts); @@ -568,7 +617,7 @@ class Probe return self::mail($uri, $uid); } - if ($network == NETWORK_MAIL) { + if ($network == Protocol::MAIL) { return self::mail($uri, $uid); } // Remove "acct:" from the URI @@ -578,21 +627,21 @@ class Probe $nick = substr($uri, 0, strpos($uri, '@')); if (strpos($uri, '@twitter.com')) { - return array("network" => NETWORK_TWITTER); + return ["network" => Protocol::TWITTER]; } $lrdd = self::hostMeta($host); if (is_bool($lrdd)) { - return array(); + return []; } if (!$lrdd) { - logger('No XRD data was found for '.$uri, LOGGER_DEBUG); + Logger::log('No XRD data was found for '.$uri, Logger::DEBUG); return self::mail($uri, $uid); } $addr = $uri; } else { - logger("Uri ".$uri." was not detectable", LOGGER_DEBUG); + Logger::log("Uri ".$uri." was not detectable", Logger::DEBUG); return false; } @@ -637,21 +686,21 @@ class Probe $result = false; - logger("Probing ".$uri, LOGGER_DEBUG); + Logger::log("Probing ".$uri, Logger::DEBUG); - if (in_array($network, array("", NETWORK_DFRN))) { + if (in_array($network, ["", Protocol::DFRN])) { $result = self::dfrn($webfinger); } - if ((!$result && ($network == "")) || ($network == NETWORK_DIASPORA)) { + if ((!$result && ($network == "")) || ($network == Protocol::DIASPORA)) { $result = self::diaspora($webfinger); } - if ((!$result && ($network == "")) || ($network == NETWORK_OSTATUS)) { + if ((!$result && ($network == "")) || ($network == Protocol::OSTATUS)) { $result = self::ostatus($webfinger); } - if ((!$result && ($network == "")) || ($network == NETWORK_PUMPIO)) { - $result = self::pumpio($webfinger); + if ((!$result && ($network == "")) || ($network == Protocol::PUMPIO)) { + $result = self::pumpio($webfinger, $addr); } - if ((!$result && ($network == "")) || ($network == NETWORK_FEED)) { + if ((!$result && ($network == "")) || ($network == Protocol::FEED)) { $result = self::feed($uri); } else { // We overwrite the detected nick with our try if the previois routines hadn't detected it. @@ -665,7 +714,15 @@ class Probe } } - logger($uri." is ".$result["network"], LOGGER_DEBUG); + if (empty($result["network"])) { + $result["network"] = Protocol::PHANTOM; + } + + if (empty($result["url"])) { + $result["url"] = $uri; + } + + Logger::log($uri." is ".$result["network"], Logger::DEBUG); if (empty($result["baseurl"])) { $pos = strpos($result["url"], $host); @@ -673,7 +730,6 @@ class Probe $result["baseurl"] = substr($result["url"], 0, $pos).$host; } } - return $result; } @@ -692,35 +748,35 @@ class Probe $xrd_timeout = Config::get('system', 'xrd_timeout', 20); $redirects = 0; - $ret = z_fetch_url($url, false, $redirects, array('timeout' => $xrd_timeout, 'accept_content' => $type)); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { + $curlResult = Network::curl($url, false, $redirects, ['timeout' => $xrd_timeout, 'accept_content' => $type]); + if ($curlResult->isTimeout()) { return false; } - $data = $ret['body']; + $data = $curlResult->getBody(); $webfinger = json_decode($data, true); if (is_array($webfinger)) { if (!isset($webfinger["links"])) { - logger("No json webfinger links for ".$url, LOGGER_DEBUG); + Logger::log("No json webfinger links for ".$url, Logger::DEBUG); return false; } return $webfinger; } // If it is not JSON, maybe it is XML - $xrd = parse_xml_string($data, false); + $xrd = XML::parseString($data, false); if (!is_object($xrd)) { - logger("No webfinger data retrievable for ".$url, LOGGER_DEBUG); + Logger::log("No webfinger data retrievable for ".$url, Logger::DEBUG); return false; } $xrd_arr = XML::elementToArray($xrd); if (!isset($xrd_arr["xrd"]["link"])) { - logger("No XML webfinger links for ".$url, LOGGER_DEBUG); + Logger::log("No XML webfinger links for ".$url, Logger::DEBUG); return false; } - $webfinger = array(); + $webfinger = []; if (!empty($xrd_arr["xrd"]["subject"])) { $webfinger["subject"] = $xrd_arr["xrd"]["subject"]; @@ -730,7 +786,7 @@ class Probe $webfinger["aliases"] = $xrd_arr["xrd"]["alias"]; } - $webfinger["links"] = array(); + $webfinger["links"] = []; foreach ($xrd_arr["xrd"]["link"] as $value => $data) { if (!empty($data["@attributes"])) { @@ -759,19 +815,19 @@ class Probe */ private static function pollNoscrape($noscrape_url, $data) { - $ret = z_fetch_url($noscrape_url); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { + $curlResult = Network::curl($noscrape_url); + if ($curlResult->isTimeout()) { return false; } - $content = $ret['body']; + $content = $curlResult->getBody(); if (!$content) { - logger("Empty body for ".$noscrape_url, LOGGER_DEBUG); + Logger::log("Empty body for ".$noscrape_url, Logger::DEBUG); return false; } $json = json_decode($content, true); if (!is_array($json)) { - logger("No json data for ".$noscrape_url, LOGGER_DEBUG); + Logger::log("No json data for ".$noscrape_url, Logger::DEBUG); return false; } @@ -875,38 +931,43 @@ class Probe */ public static function profile($profile_link) { - $data = array(); + $data = []; - logger("Check profile ".$profile_link, LOGGER_DEBUG); + Logger::log("Check profile ".$profile_link, Logger::DEBUG); // Fetch data via noscrape - this is faster - $noscrape_url = str_replace(array("/hcard/", "/profile/"), "/noscrape/", $profile_link); + $noscrape_url = str_replace(["/hcard/", "/profile/"], "/noscrape/", $profile_link); $data = self::pollNoscrape($noscrape_url, $data); if (!isset($data["notify"]) || !isset($data["confirm"]) || !isset($data["request"]) || !isset($data["poll"]) - || !isset($data["poco"]) || !isset($data["name"]) || !isset($data["photo"]) ) { $data = self::pollHcard($profile_link, $data, true); } - $prof_data = array(); + $prof_data = []; + + if (empty($data["addr"]) || empty($data["nick"])) { + $probe_data = self::uri($profile_link); + $data["addr"] = defaults($data, "addr", $probe_data["addr"]); + $data["nick"] = defaults($data, "nick", $probe_data["nick"]); + } + $prof_data["addr"] = $data["addr"]; $prof_data["nick"] = $data["nick"]; - $prof_data["dfrn-request"] = $data["request"]; - $prof_data["dfrn-confirm"] = $data["confirm"]; - $prof_data["dfrn-notify"] = $data["notify"]; - $prof_data["dfrn-poll"] = $data["poll"]; - $prof_data["dfrn-poco"] = $data["poco"]; - $prof_data["photo"] = $data["photo"]; - $prof_data["fn"] = $data["name"]; - $prof_data["key"] = $data["pubkey"]; + $prof_data["dfrn-request"] = defaults($data, 'request', null); + $prof_data["dfrn-confirm"] = defaults($data, 'confirm', null); + $prof_data["dfrn-notify"] = defaults($data, 'notify' , null); + $prof_data["dfrn-poll"] = defaults($data, 'poll' , null); + $prof_data["photo"] = defaults($data, 'photo' , null); + $prof_data["fn"] = defaults($data, 'name' , null); + $prof_data["key"] = defaults($data, 'pubkey' , null); - logger("Result for profile ".$profile_link.": ".print_r($prof_data, true), LOGGER_DEBUG); + Logger::log("Result for profile ".$profile_link.": ".print_r($prof_data, true), Logger::DEBUG); return $prof_data; } @@ -921,25 +982,27 @@ class Probe private static function dfrn($webfinger) { $hcard_url = ""; - $data = array(); - foreach ($webfinger["links"] as $link) { - if (($link["rel"] == NAMESPACE_DFRN) && ($link["href"] != "")) { - $data["network"] = NETWORK_DFRN; - } elseif (($link["rel"] == NAMESPACE_FEED) && ($link["href"] != "")) { + $data = []; + // The array is reversed to take into account the order of preference for same-rel links + // See: https://tools.ietf.org/html/rfc7033#section-4.4.4 + foreach (array_reverse($webfinger["links"]) as $link) { + if (($link["rel"] == NAMESPACE_DFRN) && !empty($link["href"])) { + $data["network"] = Protocol::DFRN; + } elseif (($link["rel"] == NAMESPACE_FEED) && !empty($link["href"])) { $data["poll"] = $link["href"]; - } elseif (($link["rel"] == "http://webfinger.net/rel/profile-page") && ($link["type"] == "text/html") && ($link["href"] != "")) { + } elseif (($link["rel"] == "http://webfinger.net/rel/profile-page") && (defaults($link, "type", "") == "text/html") && !empty($link["href"])) { $data["url"] = $link["href"]; - } elseif (($link["rel"] == "http://microformats.org/profile/hcard") && ($link["href"] != "")) { + } elseif (($link["rel"] == "http://microformats.org/profile/hcard") && !empty($link["href"])) { $hcard_url = $link["href"]; - } elseif (($link["rel"] == NAMESPACE_POCO) && ($link["href"] != "")) { + } elseif (($link["rel"] == NAMESPACE_POCO) && !empty($link["href"])) { $data["poco"] = $link["href"]; - } elseif (($link["rel"] == "http://webfinger.net/rel/avatar") && ($link["href"] != "")) { + } elseif (($link["rel"] == "http://webfinger.net/rel/avatar") && !empty($link["href"])) { $data["photo"] = $link["href"]; - } elseif (($link["rel"] == "http://joindiaspora.com/seed_location") && ($link["href"] != "")) { + } elseif (($link["rel"] == "http://joindiaspora.com/seed_location") && !empty($link["href"])) { $data["baseurl"] = trim($link["href"], '/'); - } elseif (($link["rel"] == "http://joindiaspora.com/guid") && ($link["href"] != "")) { + } elseif (($link["rel"] == "http://joindiaspora.com/guid") && !empty($link["href"])) { $data["guid"] = $link["href"]; - } elseif (($link["rel"] == "diaspora-public-key") && ($link["href"] != "")) { + } elseif (($link["rel"] == "diaspora-public-key") && !empty($link["href"])) { $data["pubkey"] = base64_decode($link["href"]); //if (strstr($data["pubkey"], 'RSA ') || ($link["type"] == "RSA")) @@ -949,9 +1012,11 @@ class Probe } } - if (is_array($webfinger["aliases"])) { + if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) { foreach ($webfinger["aliases"] as $alias) { - if (normalise_link($alias) != normalise_link($data["url"]) && ! strstr($alias, "@")) { + if (empty($data["url"]) && !strstr($alias, "@")) { + $data["url"] = $alias; + } elseif (!strstr($alias, "@") && Strings::normaliseLink($alias) != Strings::normaliseLink($data["url"])) { $data["alias"] = $alias; } elseif (substr($alias, 0, 5) == 'acct:') { $data["addr"] = substr($alias, 5); @@ -959,6 +1024,10 @@ class Probe } } + if (!empty($webfinger["subject"]) && (substr($webfinger["subject"], 0, 5) == "acct:")) { + $data["addr"] = substr($webfinger["subject"], 5); + } + if (!isset($data["network"]) || ($hcard_url == "")) { return false; } @@ -993,11 +1062,11 @@ class Probe */ private static function pollHcard($hcard_url, $data, $dfrn = false) { - $ret = z_fetch_url($hcard_url); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { + $curlResult = Network::curl($hcard_url); + if ($curlResult->isTimeout()) { return false; } - $content = $ret['body']; + $content = $curlResult->getBody(); if (!$content) { return false; } @@ -1014,13 +1083,17 @@ class Probe return false; } + if (!isset($data["baseurl"])) { + $data["baseurl"] = ""; + } + if ($vcards->length > 0) { $vcard = $vcards->item(0); // We have to discard the guid from the hcard in favour of the guid from lrdd // Reason: Hubzilla doesn't use the value "uid" in the hcard like Diaspora does. $search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' uid ')]", $vcard); // */ - if (($search->length > 0) && ($data["guid"] == "")) { + if (($search->length > 0) && empty($data["guid"])) { $data["guid"] = $search->item(0)->nodeValue; } @@ -1053,22 +1126,24 @@ class Probe } } - $avatar = array(); - $photos = $xpath->query("//*[contains(concat(' ', @class, ' '), ' photo ') or contains(concat(' ', @class, ' '), ' avatar ')]", $vcard); // */ - foreach ($photos as $photo) { - $attr = array(); - foreach ($photo->attributes as $attribute) { - $attr[$attribute->name] = trim($attribute->value); - } + $avatar = []; + if (!empty($vcard)) { + $photos = $xpath->query("//*[contains(concat(' ', @class, ' '), ' photo ') or contains(concat(' ', @class, ' '), ' avatar ')]", $vcard); // */ + foreach ($photos as $photo) { + $attr = []; + foreach ($photo->attributes as $attribute) { + $attr[$attribute->name] = trim($attribute->value); + } - if (isset($attr["src"]) && isset($attr["width"])) { - $avatar[$attr["width"]] = $attr["src"]; - } + if (isset($attr["src"]) && isset($attr["width"])) { + $avatar[$attr["width"]] = $attr["src"]; + } - // We don't have a width. So we just take everything that we got. - // This is a Hubzilla workaround which doesn't send a width. - if ((sizeof($avatar) == 0) && !empty($attr["src"])) { - $avatar[] = $attr["src"]; + // We don't have a width. So we just take everything that we got. + // This is a Hubzilla workaround which doesn't send a width. + if ((sizeof($avatar) == 0) && !empty($attr["src"])) { + $avatar[] = $attr["src"]; + } } } @@ -1083,7 +1158,7 @@ class Probe if ($search->length > 0) { foreach ($search as $link) { //$data["request"] = $search->item(0)->nodeValue; - $attr = array(); + $attr = []; foreach ($link->attributes as $attribute) { $attr[$attribute->name] = trim($attribute->value); } @@ -1093,7 +1168,7 @@ class Probe } // Older Friendica versions had used the "uid" field differently than newer versions - if ($data["nick"] == $data["guid"]) { + if (!empty($data["nick"]) && !empty($data["guid"]) && ($data["nick"] == $data["guid"])) { unset($data["guid"]); } } @@ -1112,23 +1187,25 @@ class Probe private static function diaspora($webfinger) { $hcard_url = ""; - $data = array(); - foreach ($webfinger["links"] as $link) { - if (($link["rel"] == "http://microformats.org/profile/hcard") && ($link["href"] != "")) { + $data = []; + // The array is reversed to take into account the order of preference for same-rel links + // See: https://tools.ietf.org/html/rfc7033#section-4.4.4 + foreach (array_reverse($webfinger["links"]) as $link) { + if (($link["rel"] == "http://microformats.org/profile/hcard") && !empty($link["href"])) { $hcard_url = $link["href"]; - } elseif (($link["rel"] == "http://joindiaspora.com/seed_location") && ($link["href"] != "")) { + } elseif (($link["rel"] == "http://joindiaspora.com/seed_location") && !empty($link["href"])) { $data["baseurl"] = trim($link["href"], '/'); - } elseif (($link["rel"] == "http://joindiaspora.com/guid") && ($link["href"] != "")) { + } elseif (($link["rel"] == "http://joindiaspora.com/guid") && !empty($link["href"])) { $data["guid"] = $link["href"]; - } elseif (($link["rel"] == "http://webfinger.net/rel/profile-page") && ($link["type"] == "text/html") && ($link["href"] != "")) { + } elseif (($link["rel"] == "http://webfinger.net/rel/profile-page") && (defaults($link, "type", "") == "text/html") && !empty($link["href"])) { $data["url"] = $link["href"]; - } elseif (($link["rel"] == NAMESPACE_FEED) && ($link["href"] != "")) { + } elseif (($link["rel"] == NAMESPACE_FEED) && !empty($link["href"])) { $data["poll"] = $link["href"]; - } elseif (($link["rel"] == NAMESPACE_POCO) && ($link["href"] != "")) { + } elseif (($link["rel"] == NAMESPACE_POCO) && !empty($link["href"])) { $data["poco"] = $link["href"]; - } elseif (($link["rel"] == "salmon") && ($link["href"] != "")) { + } elseif (($link["rel"] == "salmon") && !empty($link["href"])) { $data["notify"] = $link["href"]; - } elseif (($link["rel"] == "diaspora-public-key") && ($link["href"] != "")) { + } elseif (($link["rel"] == "diaspora-public-key") && !empty($link["href"])) { $data["pubkey"] = base64_decode($link["href"]); //if (strstr($data["pubkey"], 'RSA ') || ($link["type"] == "RSA")) @@ -1142,9 +1219,9 @@ class Probe return false; } - if (is_array($webfinger["aliases"])) { + if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) { foreach ($webfinger["aliases"] as $alias) { - if (normalise_link($alias) != normalise_link($data["url"]) && ! strstr($alias, "@")) { + if (Strings::normaliseLink($alias) != Strings::normaliseLink($data["url"]) && ! strstr($alias, "@")) { $data["alias"] = $alias; } elseif (substr($alias, 0, 5) == 'acct:') { $data["addr"] = substr($alias, 5); @@ -1169,10 +1246,12 @@ class Probe && isset($data["pubkey"]) && ($hcard_url != "") ) { - $data["network"] = NETWORK_DIASPORA; + $data["network"] = Protocol::DIASPORA; // The Diaspora handle must always be lowercase - $data["addr"] = strtolower($data["addr"]); + if (!empty($data["addr"])) { + $data["addr"] = strtolower($data["addr"]); + } // We have to overwrite the detected value for "notify" since Hubzilla doesn't send it $data["notify"] = $data["baseurl"] . "/receive/users/" . $data["guid"]; @@ -1194,35 +1273,37 @@ class Probe */ private static function ostatus($webfinger, $short = false) { - $data = array(); + $data = []; - if (is_array($webfinger["aliases"])) { + if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) { foreach ($webfinger["aliases"] as $alias) { - if (strstr($alias, "@") && !strstr(normalise_link($alias), "http://")) { + if (strstr($alias, "@") && !strstr(Strings::normaliseLink($alias), "http://")) { $data["addr"] = str_replace('acct:', '', $alias); } } } - if (is_string($webfinger["subject"]) && strstr($webfinger["subject"], "@") - && !strstr(normalise_link($webfinger["subject"]), "http://") + if (!empty($webfinger["subject"]) && strstr($webfinger["subject"], "@") + && !strstr(Strings::normaliseLink($webfinger["subject"]), "http://") ) { $data["addr"] = str_replace('acct:', '', $webfinger["subject"]); } $pubkey = ""; if (is_array($webfinger["links"])) { - foreach ($webfinger["links"] as $link) { + // The array is reversed to take into account the order of preference for same-rel links + // See: https://tools.ietf.org/html/rfc7033#section-4.4.4 + foreach (array_reverse($webfinger["links"]) as $link) { if (($link["rel"] == "http://webfinger.net/rel/profile-page") - && ($link["type"] == "text/html") + && (defaults($link, "type", "") == "text/html") && ($link["href"] != "") ) { $data["url"] = $link["href"]; - } elseif (($link["rel"] == "salmon") && ($link["href"] != "")) { + } elseif (($link["rel"] == "salmon") && !empty($link["href"])) { $data["notify"] = $link["href"]; - } elseif (($link["rel"] == NAMESPACE_FEED) && ($link["href"] != "")) { + } elseif (($link["rel"] == NAMESPACE_FEED) && !empty($link["href"])) { $data["poll"] = $link["href"]; - } elseif (($link["rel"] == "magic-public-key") && ($link["href"] != "")) { + } elseif (($link["rel"] == "magic-public-key") && !empty($link["href"])) { $pubkey = $link["href"]; if (substr($pubkey, 0, 5) === 'data:') { @@ -1231,19 +1312,19 @@ class Probe } else { $pubkey = substr($pubkey, 5); } - } elseif (normalise_link($pubkey) == 'http://') { - $ret = z_fetch_url($pubkey); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { + } elseif (Strings::normaliseLink($pubkey) == 'http://') { + $curlResult = Network::curl($pubkey); + if ($curlResult->isTimeout()) { return false; } - $pubkey = $ret['body']; + $pubkey = $curlResult['body']; } $key = explode(".", $pubkey); if (sizeof($key) >= 3) { - $m = base64url_decode($key[1]); - $e = base64url_decode($key[2]); + $m = Strings::base64UrlDecode($key[1]); + $e = Strings::base64UrlDecode($key[2]); $data["pubkey"] = Crypto::meToPem($m, $e); } } @@ -1254,7 +1335,7 @@ class Probe && isset($data["poll"]) && isset($data["url"]) ) { - $data["network"] = NETWORK_OSTATUS; + $data["network"] = Protocol::OSTATUS; } else { return false; } @@ -1264,37 +1345,40 @@ class Probe } // Fetch all additional data from the feed - $ret = z_fetch_url($data["poll"]); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { + $curlResult = Network::curl($data["poll"]); + if ($curlResult->isTimeout()) { return false; } - $feed = $ret['body']; + $feed = $curlResult->getBody(); + $dummy1 = null; + $dummy2 = null; + $dummy2 = null; $feed_data = Feed::import($feed, $dummy1, $dummy2, $dummy3, true); if (!$feed_data) { return false; } - if ($feed_data["header"]["author-name"] != "") { + if (!empty($feed_data["header"]["author-name"])) { $data["name"] = $feed_data["header"]["author-name"]; } - if ($feed_data["header"]["author-nick"] != "") { + if (!empty($feed_data["header"]["author-nick"])) { $data["nick"] = $feed_data["header"]["author-nick"]; } - if ($feed_data["header"]["author-avatar"] != "") { + if (!empty($feed_data["header"]["author-avatar"])) { $data["photo"] = self::fixAvatar($feed_data["header"]["author-avatar"], $data["url"]); } - if ($feed_data["header"]["author-id"] != "") { + if (!empty($feed_data["header"]["author-id"])) { $data["alias"] = $feed_data["header"]["author-id"]; } - if ($feed_data["header"]["author-location"] != "") { + if (!empty($feed_data["header"]["author-location"])) { $data["location"] = $feed_data["header"]["author-location"]; } - if ($feed_data["header"]["author-about"] != "") { + if (!empty($feed_data["header"]["author-about"])) { $data["about"] = $feed_data["header"]["author-about"]; } // OStatus has serious issues when the the url doesn't fit (ssl vs. non ssl) // So we take the value that we just fetched, although the other one worked as well - if ($feed_data["header"]["author-link"] != "") { + if (!empty($feed_data["header"]["author-link"])) { $data["url"] = $feed_data["header"]["author-link"]; } @@ -1323,16 +1407,35 @@ class Probe $xpath = new DomXPath($doc); - $data = array(); + $data = []; + + $data["name"] = $xpath->query("//span[contains(@class, 'p-name')]")->item(0)->nodeValue; + + if ($data["name"] == '') { + // This is ugly - but pump.io doesn't seem to know a better way for it + $data["name"] = trim($xpath->query("//h1[@class='media-header']")->item(0)->nodeValue); + $pos = strpos($data["name"], chr(10)); + if ($pos) { + $data["name"] = trim(substr($data["name"], 0, $pos)); + } + } + + $data["location"] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'p-locality')]"); - // This is ugly - but pump.io doesn't seem to know a better way for it - $data["name"] = trim($xpath->query("//h1[@class='media-header']")->item(0)->nodeValue); - $pos = strpos($data["name"], chr(10)); - if ($pos) { - $data["name"] = trim(substr($data["name"], 0, $pos)); + if ($data["location"] == '') { + $data["location"] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'location')]"); } - $avatar = $xpath->query("//img[@class='img-rounded media-object']")->item(0); + $data["about"] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'p-note')]"); + + if ($data["about"] == '') { + $data["about"] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'summary')]"); + } + + $avatar = $xpath->query("//img[contains(@class, 'u-photo')]")->item(0); + if (!$avatar) { + $avatar = $xpath->query("//img[@class='img-rounded media-object']")->item(0); + } if ($avatar) { foreach ($avatar->attributes as $attribute) { if ($attribute->name == "src") { @@ -1341,9 +1444,6 @@ class Probe } } - $data["location"] = $xpath->query("//p[@class='location']")->item(0)->nodeValue; - $data["about"] = $xpath->query("//p[@class='summary']")->item(0)->nodeValue; - return $data; } @@ -1354,12 +1454,14 @@ class Probe * * @return array pump.io data */ - private static function pumpio($webfinger) + private static function pumpio($webfinger, $addr) { - $data = array(); - foreach ($webfinger["links"] as $link) { + $data = []; + // The array is reversed to take into account the order of preference for same-rel links + // See: https://tools.ietf.org/html/rfc7033#section-4.4.4 + foreach (array_reverse($webfinger["links"]) as $link) { if (($link["rel"] == "http://webfinger.net/rel/profile-page") - && ($link["type"] == "text/html") + && (defaults($link, "type", "") == "text/html") && ($link["href"] != "") ) { $data["url"] = $link["href"]; @@ -1379,7 +1481,7 @@ class Probe // So we unset all data that isn't used at the moment unset($data["dialback"]); - $data["network"] = NETWORK_PUMPIO; + $data["network"] = Protocol::PUMPIO; } else { return false; } @@ -1392,6 +1494,13 @@ class Probe $data = array_merge($data, $profile_data); + if (($addr != '') && ($data['name'] != '')) { + $name = trim(str_replace($addr, '', $data['name'])); + if ($name != '') { + $data['name'] = $name; + } + } + return $data; } @@ -1425,7 +1534,7 @@ class Probe $feed_url = ""; foreach ($feeds as $feed) { - $attr = array(); + $attr = []; foreach ($feed->attributes as $attribute) { $attr[$attribute->name] = trim($attribute->value); } @@ -1448,11 +1557,12 @@ class Probe */ private static function feed($url, $probe = true) { - $ret = z_fetch_url($url); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { + $curlResult = Network::curl($url); + if ($curlResult->isTimeout()) { return false; } - $feed = $ret['body']; + $feed = $curlResult->getBody(); + $dummy1 = $dummy2 = $dummy3 = null; $feed_data = Feed::import($feed, $dummy1, $dummy2, $dummy3, true); if (!$feed_data) { @@ -1469,32 +1579,32 @@ class Probe return self::feed($feed_url, false); } - if ($feed_data["header"]["author-name"] != "") { + if (!empty($feed_data["header"]["author-name"])) { $data["name"] = $feed_data["header"]["author-name"]; } - if ($feed_data["header"]["author-nick"] != "") { + if (!empty($feed_data["header"]["author-nick"])) { $data["nick"] = $feed_data["header"]["author-nick"]; } - if ($feed_data["header"]["author-avatar"] != "") { + if (!empty($feed_data["header"]["author-avatar"])) { $data["photo"] = $feed_data["header"]["author-avatar"]; } - if ($feed_data["header"]["author-id"] != "") { + if (!empty($feed_data["header"]["author-id"])) { $data["alias"] = $feed_data["header"]["author-id"]; } $data["url"] = $url; $data["poll"] = $url; - if ($feed_data["header"]["author-link"] != "") { + if (!empty($feed_data["header"]["author-link"])) { $data["baseurl"] = $feed_data["header"]["author-link"]; } else { $data["baseurl"] = $data["url"]; } - $data["network"] = NETWORK_FEED; + $data["network"] = Protocol::FEED; return $data; } @@ -1509,44 +1619,50 @@ class Probe */ private static function mail($uri, $uid) { - if (!validate_email($uri)) { + if (!Network::isEmailDomainValid($uri)) { return false; } - if ($uid != 0) { - $x = q("SELECT `prvkey` FROM `user` WHERE `uid` = %d LIMIT 1", intval($uid)); + if ($uid == 0) { + return false; + } - $r = q("SELECT * FROM `mailacct` WHERE `uid` = %d AND `server` != '' LIMIT 1", intval($uid)); + $user = DBA::selectFirst('user', ['prvkey'], ['uid' => $uid]); - if (DBM::is_result($x) && DBM::is_result($r)) { - $mailbox = Email::constructMailboxName($r[0]); - $password = ''; - openssl_private_decrypt(hex2bin($r[0]['pass']), $password, $x[0]['prvkey']); - $mbox = Email::connect($mailbox, $r[0]['user'], $password); - if (!mbox) { - return false; - } - } + $condition = ["`uid` = ? AND `server` != ''", $uid]; + $fields = ['pass', 'user', 'server', 'port', 'ssltype', 'mailbox']; + $mailacct = DBA::selectFirst('mailacct', $fields, $condition); - $msgs = Email::poll($mbox, $uri); - logger('searching '.$uri.', '.count($msgs).' messages found.', LOGGER_DEBUG); + if (!DBA::isResult($user) || !DBA::isResult($mailacct)) { + return false; + } - if (!count($msgs)) { - return false; - } + $mailbox = Email::constructMailboxName($mailacct); + $password = ''; + openssl_private_decrypt(hex2bin($mailacct['pass']), $password, $user['prvkey']); + $mbox = Email::connect($mailbox, $mailacct['user'], $password); + if (!$mbox) { + return false; + } + + $msgs = Email::poll($mbox, $uri); + Logger::log('searching '.$uri.', '.count($msgs).' messages found.', Logger::DEBUG); + + if (!count($msgs)) { + return false; } $phost = substr($uri, strpos($uri, '@') + 1); - $data = array(); + $data = []; $data["addr"] = $uri; - $data["network"] = NETWORK_MAIL; + $data["network"] = Protocol::MAIL; $data["name"] = substr($uri, 0, strpos($uri, '@')); $data["nick"] = $data["name"]; - $data["photo"] = avatar_img($uri); + $data["photo"] = Network::lookupAvatarByEmail($uri); $data["url"] = 'mailto:'.$uri; - $data["notify"] = 'smtp '.random_string(); - $data["poll"] = 'email '.random_string(); + $data["notify"] = 'smtp ' . Strings::getRandomHex(); + $data["poll"] = 'email ' . Strings::getRandomHex(); $x = Email::messageMeta($mbox, $msgs[0]); if (stristr($x[0]->from, $uri)) { @@ -1570,14 +1686,13 @@ class Probe } } - $data["name"] = notags($data["name"]); + $data["name"] = Strings::escapeTags($data["name"]); } } } if (!empty($mbox)) { imap_close($mbox); } - return $data; } @@ -1613,7 +1728,7 @@ class Probe $fixed = $scheme.$host.$port.$path.$query.$fragment; - logger('Base: '.$base.' - Avatar: '.$avatar.' - Fixed: '.$fixed, LOGGER_DATA); + Logger::log('Base: '.$base.' - Avatar: '.$avatar.' - Fixed: '.$fixed, Logger::DATA); return $fixed; }