X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FNetwork%2FProbe.php;h=af2d1c9a16ef6d8254dbc8ca7b803e338097d324;hb=69e7c7fecac5d315930c52bfa9fb21954dcf8b5e;hp=5a70247afcea6427a68e21440ec4d8a51f5ea030;hpb=7052da745c17029971459fda2098bb3c917fab9f;p=friendica.git diff --git a/src/Network/Probe.php b/src/Network/Probe.php index 5a70247afc..af2d1c9a16 100644 --- a/src/Network/Probe.php +++ b/src/Network/Probe.php @@ -1,33 +1,38 @@ get_hostname(); $parts = parse_url($host); @@ -81,17 +88,17 @@ class Probe { } /** - * @brief Probes for XRD data + * @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 $host The host part of an url * - * @return array - * 'lrdd' => Link to LRDD endpoint - * 'lrdd-xml' => Link to LRDD endpoint in XML format - * 'lrdd-json' => Link to LRDD endpoint in JSON format + * @return array with template and type of the webfinger template for JSON or XML */ - private static function xrd($host) { - + private static function hostMeta($host) + { // Reset the static variable self::$baseurl = ''; @@ -102,37 +109,41 @@ class Probe { $redirects = 0; logger("Probing for ".$host, LOGGER_DEBUG); + $xrd = null; - $ret = z_fetch_url($ssl_url, false, $redirects, array('timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml')); - if (($ret['errno'] == CURLE_OPERATION_TIMEDOUT) && !self::ownHost($ssl_url)) { - logger("Probing timeout for ".$ssl_url, LOGGER_DEBUG); - return false; + $ret = Network::curl($ssl_url, false, $redirects, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']); + if ($ret['success']) { + $xml = $ret['body']; + $xrd = XML::parseString($xml, false); + $host_url = 'https://'.$host; } - $xml = $ret['body']; - - $xrd = parse_xml_string($xml, false); 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) { + $ret = Network::curl($url, false, $redirects, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']); + if (!empty($ret["errno"]) && ($ret['errno'] == CURLE_OPERATION_TIMEDOUT)) { logger("Probing timeout for ".$url, LOGGER_DEBUG); return false; } $xml = $ret['body']; - $xrd = parse_xml_string($xml, false); + $xrd = XML::parseString($xml, false); + $host_url = 'http://'.$host; } if (!is_object($xrd)) { logger("No xrd object found for ".$host, LOGGER_DEBUG); - return array(); + return []; } - $links = xml::element_to_array($xrd); + $links = XML::elementToArray($xrd); if (!isset($links["xrd"]["link"])) { logger("No xrd data found for ".$host, LOGGER_DEBUG); - return array(); + return []; } - $xrd_data = 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. + // $lrdd = array("application/jrd+json" => $host_url.'/.well-known/webfinger?resource={uri}'); foreach ($links["xrd"]["link"] as $value => $link) { if (!empty($link["@attributes"])) { @@ -143,16 +154,10 @@ class Probe { continue; } - if (($attributes["rel"] == "lrdd") - && ($attributes["type"] == "application/xrd+xml") - ) { - $xrd_data["lrdd-xml"] = $attributes["template"]; - } elseif (($attributes["rel"] == "lrdd") - && ($attributes["type"] == "application/json") - ) { - $xrd_data["lrdd-json"] = $attributes["template"]; - } elseif ($attributes["rel"] == "lrdd") { - $xrd_data["lrdd"] = $attributes["template"]; + if (($attributes["rel"] == "lrdd") && !empty($attributes["template"])) { + $type = (empty($attributes["type"]) ? '' : $attributes["type"]); + + $lrdd[$type] = $attributes["template"]; } } @@ -160,7 +165,7 @@ class Probe { logger("Probing successful for ".$host, LOGGER_DEBUG); - return $xrd_data; + return $lrdd; } /** @@ -178,13 +183,13 @@ class Probe { * amended 7/9/2011 to return an hcard which could save potentially loading * a lengthy content page to scrape dfrn attributes * - * @param string $webbie Address that should be probed + * @param string $webbie Address that should be probed * @param string $hcard_url Link to the hcard - is returned by reference * * @return string profile link */ - public static function webfingerDfrn($webbie, &$hcard_url) { - + public static function webfingerDfrn($webbie, &$hcard_url) + { $profile_link = ''; $links = self::lrdd($webbie); @@ -208,7 +213,7 @@ 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. * @@ -216,19 +221,19 @@ class Probe { * * @return array uri data */ - public static function lrdd($uri) { - - $lrdd = self::xrd($uri); + public static function lrdd($uri) + { + $lrdd = self::hostMeta($uri); $webfinger = null; if (is_bool($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"]; @@ -241,31 +246,27 @@ class Probe { $nick = array_pop($path_parts); do { - $lrdd = self::xrd($host); + $lrdd = self::hostMeta($host); $host .= "/".array_shift($path_parts); } while (!$lrdd && (sizeof($path_parts) > 0)); } if (!$lrdd) { logger("No lrdd data found for ".$uri, LOGGER_DEBUG); - return array(); + return []; } - foreach ($lrdd as $key => $link) { + foreach ($lrdd as $type => $template) { if ($webfinger) { continue; } - if (!in_array($key, array("lrdd", "lrdd-xml", "lrdd-json"))) { - continue; - } - - $path = str_replace('{uri}', urlencode($uri), $link); - $webfinger = self::webfinger($path); + $path = str_replace('{uri}', urlencode($uri), $template); + $webfinger = self::webfinger($path, $type); if (!$webfinger && (strstr($uri, "@"))) { - $path = str_replace('{uri}', urlencode("acct:".$uri), $link); - $webfinger = self::webfinger($path); + $path = str_replace('{uri}', urlencode("acct:".$uri), $template); + $webfinger = self::webfinger($path, $type); } // Special treatment for Mastodon @@ -277,8 +278,8 @@ class Probe { $addr = $nick."@".$host; - $path = str_replace('{uri}', urlencode("acct:".$addr), $link); - $webfinger = self::webfinger($path); + $path = str_replace('{uri}', urlencode("acct:".$addr), $template); + $webfinger = self::webfinger($path, $type); } } @@ -287,17 +288,17 @@ class Probe { 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]]; } } @@ -307,23 +308,23 @@ class Probe { /** * @brief Fetch information (protocol endpoints and user information) about a given uri * - * @param string $uri Address that should be probed - * @param string $network Test for this specific network - * @param integer $uid User ID for the probe (only used for mails) - * @param boolean $cache Use cached values? + * @param string $uri Address that should be probed + * @param string $network Test for this specific network + * @param integer $uid User ID for the probe (only used for mails) + * @param boolean $cache Use cached values? * * @return array uri data */ - public static function uri($uri, $network = "", $uid = 0, $cache = true) { - + public static function uri($uri, $network = "", $uid = -1, $cache = true) + { if ($cache) { - $result = Cache::get("probe_url:".$network.":".$uri); + $result = Cache::get("Probe::uri:".$network.":".$uri); if (!is_null($result)) { return $result; } } - if ($uid == 0) { + if ($uid == -1) { $uid = local_user(); } @@ -333,10 +334,10 @@ class Probe { $data["url"] = $uri; } - if ($data["photo"] != "") { - $data["baseurl"] = matching_url(normalise_link($data["baseurl"]), normalise_link($data["photo"])); + if (x($data, "photo")) { + $data["baseurl"] = Network::getUrlMatch(normalise_link(defaults($data, "baseurl", "")), normalise_link($data["photo"])); } else { - $data["photo"] = App::get_baseurl().'/images/person-175.jpg'; + $data["photo"] = System::baseUrl().'/images/person-175.jpg'; } if (empty($data["name"])) { @@ -344,7 +345,7 @@ class Probe { $data["name"] = $data["nick"]; } - if ($data["name"] == "") { + if (!x($data, "name")) { $data["name"] = $data["url"]; } } @@ -357,25 +358,25 @@ class Probe { } } - if (self::$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($data['network'], array(NETWORK_PHANTOM, NETWORK_MAIL))) { - Cache::set("probe_url:".$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) + if (($data['network'] != Protocol::FEED) && ($network == "") && $data["name"] && $data["nick"] @@ -383,19 +384,96 @@ class Probe { && $data["addr"] && $data["poll"] ) { - q("UPDATE `contact` SET `name` = '%s', `nick` = '%s', `url` = '%s', `addr` = '%s', - `notify` = '%s', `poll` = '%s', `alias` = '%s', `success_update` = '%s' - WHERE `nurl` = '%s' AND NOT `self` AND `uid` = 0", - dbesc($data["name"]), - dbesc($data["nick"]), - dbesc($data["url"]), - dbesc($data["addr"]), - dbesc($data["notify"]), - dbesc($data["poll"]), - dbesc($data["alias"]), - dbesc(datetime_convert()), - dbesc(normalise_link($data['url'])) - ); + $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)) { + unset($fields[$key]); + } else { + $fieldnames[] = $key; + } + } + + $fields['updated'] = DateTimeFormat::utcNow(); + + $condition = ['nurl' => normalise_link($data["url"])]; + + $old_fields = DBA::selectFirst('gcontact', $fieldnames, $condition); + + // When the gcontact doesn't exist, the value "true" will trigger an insert. + // In difference to the public contacts we want to have every contact + // in the world in our global contacts. + if (!$old_fields) { + $old_fields = true; + + // These values have to be set only on insert + $fields['photo'] = $data['photo']; + $fields['created'] = DateTimeFormat::utcNow(); + } + + 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)) { + unset($fields[$key]); + } else { + $fieldnames[] = $key; + } + } + + $condition = ['nurl' => normalise_link($data["url"]), 'self' => false, 'uid' => 0]; + + // "$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); + + $fields['name-date'] = DateTimeFormat::utcNow(); + $fields['uri-date'] = DateTimeFormat::utcNow(); + $fields['success_update'] = DateTimeFormat::utcNow(); + + DBA::update('contact', $fields, $condition, $old_fields); } } @@ -409,7 +487,8 @@ class Probe { * * @return string switched URL */ - private static function switchScheme($url) { + private static function switchScheme($url) + { $parts = parse_url($url); if (!isset($parts['scheme'])) { @@ -428,12 +507,14 @@ class Probe { /** * @brief Checks if a profile url should be OStatus but only provides partial information * - * @param array $webfinger Webfinger data - * @param string $lrdd Path template for webfinger request + * @param array $webfinger Webfinger data + * @param string $lrdd Path template for webfinger request + * @param string $type type * * @return array fixed webfinger data */ - private static function fixOstatus($webfinger, $lrdd) { + private static function fixOStatus($webfinger, $lrdd, $type) + { if (empty($webfinger['links']) || empty($webfinger['subject'])) { return $webfinger; } @@ -456,7 +537,7 @@ class Probe { $url = self::switchScheme($webfinger['subject']); $path = str_replace('{uri}', urlencode($url), $lrdd); - $webfinger2 = self::webfinger($path); + $webfinger2 = self::webfinger($path, $type); // Is the new webfinger detectable as OStatus? if (self::ostatus($webfinger2, true)) { @@ -471,13 +552,14 @@ class Probe { * * This function is only called by the "uri" function that adds caching and rearranging of data. * - * @param string $uri Address that should be probed - * @param string $network Test for this specific network - * @param integer $uid User ID for the probe (only used for mails) + * @param string $uri Address that should be probed + * @param string $network Test for this specific network + * @param integer $uid User ID for the probe (only used for mails) * * @return array uri data */ - private static function detect($uri, $network, $uid) { + private static function detect($uri, $network, $uid) + { $parts = parse_url($uri); if (!empty($parts["scheme"]) && !empty($parts["host"]) && !empty($parts["path"])) { @@ -487,19 +569,19 @@ class Probe { } if ($host == 'twitter.com') { - return array("network" => NETWORK_TWITTER); + return ["network" => Protocol::TWITTER]; } - $lrdd = self::xrd($host); + $lrdd = self::hostMeta($host); if (is_bool($lrdd)) { - return array(); + return []; } $path_parts = explode("/", trim($parts["path"], "/")); while (!$lrdd && (sizeof($path_parts) > 1)) { $host .= "/".array_shift($path_parts); - $lrdd = self::xrd($host); + $lrdd = self::hostMeta($host); } if (!$lrdd) { logger('No XRD data was found for '.$uri, LOGGER_DEBUG); @@ -511,7 +593,6 @@ class Probe { $nick = ltrim($nick, '@'); $addr = $nick."@".$host; - } elseif (strstr($uri, '@')) { // If the URI starts with "mailto:" then jump directly to the mail detection if (strpos($uri, 'mailto:') !== false) { @@ -519,7 +600,7 @@ 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 @@ -529,12 +610,12 @@ class Probe { $nick = substr($uri, 0, strpos($uri, '@')); if (strpos($uri, '@twitter.com')) { - return array("network" => NETWORK_TWITTER); + return ["network" => Protocol::TWITTER]; } - $lrdd = self::xrd($host); + $lrdd = self::hostMeta($host); if (is_bool($lrdd)) { - return array(); + return []; } if (!$lrdd) { @@ -542,7 +623,6 @@ class Probe { return self::mail($uri, $uid); } $addr = $uri; - } else { logger("Uri ".$uri." was not detectable", LOGGER_DEBUG); return false; @@ -552,19 +632,17 @@ class Probe { /// @todo Do we need the prefix "acct:" or "acct://"? - foreach ($lrdd as $key => $link) { + foreach ($lrdd as $type => $template) { if ($webfinger) { continue; } - if (!in_array($key, array("lrdd", "lrdd-xml", "lrdd-json"))) { - continue; - } + // At first try it with the given uri - $path = str_replace('{uri}', urlencode($uri), $link); - $webfinger = self::webfinger($path); + $path = str_replace('{uri}', urlencode($uri), $template); + $webfinger = self::webfinger($path, $type); // Fix possible problems with GNU Social probing to wrong scheme - $webfinger = self::fixOstatus($webfinger, $link); + $webfinger = self::fixOStatus($webfinger, $template, $type); // We cannot be sure that the detected address was correct, so we don't use the values if ($webfinger && ($uri != $addr)) { @@ -574,16 +652,17 @@ class Probe { // Try webfinger with the address (user@domain.tld) if (!$webfinger) { - $path = str_replace('{uri}', urlencode($addr), $link); - $webfinger = self::webfinger($path); + $path = str_replace('{uri}', urlencode($addr), $template); + $webfinger = self::webfinger($path, $type); } // Mastodon needs to have it with "acct:" if (!$webfinger) { - $path = str_replace('{uri}', urlencode("acct:".$addr), $link); - $webfinger = self::webfinger($path); + $path = str_replace('{uri}', urlencode("acct:".$addr), $template); + $webfinger = self::webfinger($path, $type); } } + if (!$webfinger) { return self::feed($uri); } @@ -592,19 +671,19 @@ class Probe { logger("Probing ".$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. @@ -618,6 +697,14 @@ class Probe { } } + if (empty($result["network"])) { + $result["network"] = Protocol::PHANTOM; + } + + if (empty($result["url"])) { + $result["url"] = $uri; + } + logger($uri." is ".$result["network"], LOGGER_DEBUG); if (empty($result["baseurl"])) { @@ -626,7 +713,6 @@ class Probe { $result["baseurl"] = substr($result["url"], 0, $pos).$host; } } - return $result; } @@ -635,42 +721,45 @@ class Probe { * * For details see RFC 7033: * - * @param string $url Address that should be probed + * @param string $url Address that should be probed + * @param string $type type * * @return array webfinger data */ - private static function webfinger($url) { - + private static function webfinger($url, $type) + { $xrd_timeout = Config::get('system', 'xrd_timeout', 20); $redirects = 0; - $ret = z_fetch_url($url, false, $redirects, array('timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml')); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { + $ret = Network::curl($url, false, $redirects, ['timeout' => $xrd_timeout, 'accept_content' => $type]); + if (!empty($ret["errno"]) && ($ret['errno'] == CURLE_OPERATION_TIMEDOUT)) { return false; } $data = $ret['body']; - $xrd = parse_xml_string($data, false); - - if (!is_object($xrd)) { - // If it is not XML, maybe it is JSON - $webfinger = json_decode($data, true); - + $webfinger = json_decode($data, true); + if (is_array($webfinger)) { if (!isset($webfinger["links"])) { logger("No json webfinger links for ".$url, LOGGER_DEBUG); return false; } - return $webfinger; } - $xrd_arr = xml::element_to_array($xrd); + // If it is not JSON, maybe it is XML + $xrd = XML::parseString($data, false); + if (!is_object($xrd)) { + logger("No webfinger data retrievable for ".$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); return false; } - $webfinger = array(); + $webfinger = []; if (!empty($xrd_arr["xrd"]["subject"])) { $webfinger["subject"] = $xrd_arr["xrd"]["subject"]; @@ -680,7 +769,7 @@ 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"])) { @@ -703,13 +792,14 @@ class Probe { * This functionality was originally created for the directory. * * @param string $noscrape_url Link to the noscrape page - * @param array $data The already fetched data + * @param array $data The already fetched data * * @return array noscrape data */ - private static function pollNoscrape($noscrape_url, $data) { - $ret = z_fetch_url($noscrape_url); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { + private static function pollNoscrape($noscrape_url, $data) + { + $ret = Network::curl($noscrape_url); + if (!empty($ret["errno"]) && ($ret['errno'] == CURLE_OPERATION_TIMEDOUT)) { return false; } $content = $ret['body']; @@ -736,6 +826,10 @@ class Probe { $data["nick"] = $json["nick"]; } + if (!empty($json["guid"])) { + $data["guid"] = $json["guid"]; + } + if (!empty($json["comm"])) { $data["community"] = $json["comm"]; } @@ -747,7 +841,7 @@ class Probe { } } - $location = formatted_location($json); + $location = Profile::formatLocation($json); if ($location) { $data["location"] = $location; } @@ -790,7 +884,8 @@ class Probe { * * @return int Number of errors */ - public static function validDfrn($data) { + public static function validDfrn($data) + { $errors = 0; if (!isset($data['key'])) { $errors ++; @@ -817,35 +912,40 @@ class Probe { * * @return array profile data */ - public static function profile($profile_link) { - - $data = array(); + public static function profile($profile_link) + { + $data = []; logger("Check profile ".$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"] = $data["pubkey"]; @@ -862,13 +962,13 @@ class Probe { * * @return array DFRN data */ - private static function dfrn($webfinger) { - + private static function dfrn($webfinger) + { $hcard_url = ""; - $data = array(); + $data = []; foreach ($webfinger["links"] as $link) { if (($link["rel"] == NAMESPACE_DFRN) && ($link["href"] != "")) { - $data["network"] = NETWORK_DFRN; + $data["network"] = Protocol::DFRN; } elseif (($link["rel"] == NAMESPACE_FEED) && ($link["href"] != "")) { $data["poll"] = $link["href"]; } elseif (($link["rel"] == "http://webfinger.net/rel/profile-page") && ($link["type"] == "text/html") && ($link["href"] != "")) { @@ -888,19 +988,27 @@ class Probe { //if (strstr($data["pubkey"], 'RSA ') || ($link["type"] == "RSA")) if (strstr($data["pubkey"], 'RSA ')) { - $data["pubkey"] = rsatopem($data["pubkey"]); + $data["pubkey"] = Crypto::rsaToPem($data["pubkey"]); } } } - if (is_array($webfinger["aliases"])) { + if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) { foreach ($webfinger["aliases"] as $alias) { - if (substr($alias, 0, 5) == 'acct:') { + if (empty($data["url"]) && !strstr($alias, "@")) { + $data["url"] = $alias; + } elseif (!strstr($alias, "@") && normalise_link($alias) != normalise_link($data["url"])) { + $data["alias"] = $alias; + } elseif (substr($alias, 0, 5) == 'acct:') { $data["addr"] = substr($alias, 5); } } } + if (!empty($webfinger["subject"]) && (substr($webfinger["subject"], 0, 5) == "acct:")) { + $data["addr"] = substr($webfinger["subject"], 5); + } + if (!isset($data["network"]) || ($hcard_url == "")) { return false; } @@ -927,15 +1035,16 @@ class Probe { /** * @brief Poll the hcard page (Diaspora and Friendica specific) * - * @param string $hcard_url Link to the hcard page - * @param array $data The already fetched data - * @param boolean $dfrn Poll DFRN specific data + * @param string $hcard_url Link to the hcard page + * @param array $data The already fetched data + * @param boolean $dfrn Poll DFRN specific data * * @return array hcard data */ - private static function pollHcard($hcard_url, $data, $dfrn = false) { - $ret = z_fetch_url($hcard_url); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { + private static function pollHcard($hcard_url, $data, $dfrn = false) + { + $ret = Network::curl($hcard_url); + if (!empty($ret["errno"]) && ($ret['errno'] == CURLE_OPERATION_TIMEDOUT)) { return false; } $content = $ret['body']; @@ -955,13 +1064,17 @@ class Probe { return false; } + if (!isset($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"] = $search->item(0)->nodeValue; } @@ -984,7 +1097,7 @@ class Probe { if ($search->length > 0) { $data["pubkey"] = $search->item(0)->nodeValue; if (strstr($data["pubkey"], 'RSA ')) { - $data["pubkey"] = rsatopem($data["pubkey"]); + $data["pubkey"] = Crypto::rsaToPem($data["pubkey"]); } } @@ -994,22 +1107,24 @@ 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[] = $attr["src"]; + } } } @@ -1024,7 +1139,7 @@ class Probe { if ($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($attribute->value); } @@ -1034,7 +1149,7 @@ class Probe { } // Older Friendica versions had used the "uid" field differently than newer versions - if ($data["nick"] == $data["guid"]) { + if (!empty($data["nick"]) && !empty($data["guid"]) && ($data["nick"] == $data["guid"])) { unset($data["guid"]); } } @@ -1050,10 +1165,10 @@ class Probe { * * @return array Diaspora data */ - private static function diaspora($webfinger) { - + private static function diaspora($webfinger) + { $hcard_url = ""; - $data = array(); + $data = []; foreach ($webfinger["links"] as $link) { if (($link["rel"] == "http://microformats.org/profile/hcard") && ($link["href"] != "")) { $hcard_url = $link["href"]; @@ -1074,7 +1189,7 @@ class Probe { //if (strstr($data["pubkey"], 'RSA ') || ($link["type"] == "RSA")) if (strstr($data["pubkey"], 'RSA ')) { - $data["pubkey"] = rsatopem($data["pubkey"]); + $data["pubkey"] = Crypto::rsaToPem($data["pubkey"]); } } } @@ -1083,14 +1198,20 @@ 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, "@")) { $data["alias"] = $alias; + } elseif (substr($alias, 0, 5) == 'acct:') { + $data["addr"] = substr($alias, 5); } } } + if (!empty($webfinger["subject"]) && (substr($webfinger["subject"], 0, 5) == 'acct:')) { + $data["addr"] = substr($webfinger["subject"], 5); + } + // Fetch further information from the hcard $data = self::pollHcard($hcard_url, $data); @@ -1104,10 +1225,12 @@ class Probe { && isset($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"]; @@ -1123,57 +1246,64 @@ class Probe { * @brief Check for OStatus contact * * @param array $webfinger Webfinger data - * @param bool $short Short detection mode + * @param bool $short Short detection mode * * @return array|bool OStatus data or "false" on error or "true" on short mode */ - private static function ostatus($webfinger, $short = false) { - $data = array(); - if (is_array($webfinger["aliases"])) { + private static function ostatus($webfinger, $short = false) + { + $data = []; + + if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) { foreach ($webfinger["aliases"] as $alias) { - if (strstr($alias, "@")) { + if (strstr($alias, "@") && !strstr(normalise_link($alias), "http://")) { $data["addr"] = str_replace('acct:', '', $alias); } } } - if (is_string($webfinger["subject"]) && strstr($webfinger["subject"], "@")) { + if (!empty($webfinger["subject"]) && strstr($webfinger["subject"], "@") + && !strstr(normalise_link($webfinger["subject"]), "http://") + ) { $data["addr"] = str_replace('acct:', '', $webfinger["subject"]); } - $pubkey = ""; - foreach ($webfinger["links"] as $link) { - if (($link["rel"] == "http://webfinger.net/rel/profile-page") - && ($link["type"] == "text/html") - && ($link["href"] != "") - ) { - $data["url"] = $link["href"]; - } elseif (($link["rel"] == "salmon") && ($link["href"] != "")) { - $data["notify"] = $link["href"]; - } elseif (($link["rel"] == NAMESPACE_FEED) && ($link["href"] != "")) { - $data["poll"] = $link["href"]; - } elseif (($link["rel"] == "magic-public-key") && ($link["href"] != "")) { - $pubkey = $link["href"]; - if (substr($pubkey, 0, 5) === 'data:') { - if (strstr($pubkey, ',')) { - $pubkey = substr($pubkey, strpos($pubkey, ',') + 1); - } else { - $pubkey = substr($pubkey, 5); - } - } elseif (normalise_link($pubkey) == 'http://') { - $ret = z_fetch_url($pubkey); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { - return false; + $pubkey = ""; + if (is_array($webfinger["links"])) { + foreach ($webfinger["links"] as $link) { + if (($link["rel"] == "http://webfinger.net/rel/profile-page") + && ($link["type"] == "text/html") + && ($link["href"] != "") + ) { + $data["url"] = $link["href"]; + } elseif (($link["rel"] == "salmon") && ($link["href"] != "")) { + $data["notify"] = $link["href"]; + } elseif (($link["rel"] == NAMESPACE_FEED) && ($link["href"] != "")) { + $data["poll"] = $link["href"]; + } elseif (($link["rel"] == "magic-public-key") && ($link["href"] != "")) { + $pubkey = $link["href"]; + + if (substr($pubkey, 0, 5) === 'data:') { + if (strstr($pubkey, ',')) { + $pubkey = substr($pubkey, strpos($pubkey, ',') + 1); + } else { + $pubkey = substr($pubkey, 5); + } + } elseif (normalise_link($pubkey) == 'http://') { + $ret = Network::curl($pubkey); + if (!empty($ret["errno"]) && ($ret['errno'] == CURLE_OPERATION_TIMEDOUT)) { + return false; + } + $pubkey = $ret['body']; } - $pubkey = $ret['body']; - } - $key = explode(".", $pubkey); + $key = explode(".", $pubkey); - if (sizeof($key) >= 3) { - $m = base64url_decode($key[1]); - $e = base64url_decode($key[2]); - $data["pubkey"] = metopem($m, $e); + if (sizeof($key) >= 3) { + $m = base64url_decode($key[1]); + $e = base64url_decode($key[2]); + $data["pubkey"] = Crypto::meToPem($m, $e); + } } } } @@ -1182,7 +1312,7 @@ class Probe { && isset($data["poll"]) && isset($data["url"]) ) { - $data["network"] = NETWORK_OSTATUS; + $data["network"] = Protocol::OSTATUS; } else { return false; } @@ -1192,38 +1322,48 @@ class Probe { } // Fetch all additional data from the feed - $ret = z_fetch_url($data["poll"]); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { + $ret = Network::curl($data["poll"]); + if (!empty($ret["errno"]) && ($ret['errno'] == CURLE_OPERATION_TIMEDOUT)) { return false; } $feed = $ret['body']; - $feed_data = feed_import($feed, $dummy1, $dummy2, $dummy3, true); + $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"] = $feed_data["header"]["author-link"]; } + + if (($data['poll'] == $data['url']) && ($data["alias"] != '')) { + $data['url'] = $data["alias"]; + $data["alias"] = ''; + } + /// @todo Fetch location and "about" from the feed as well return $data; } @@ -1235,8 +1375,8 @@ class Probe { * * @return array profile data */ - private static function pumpioProfileData($profile_link) { - + private static function pumpioProfileData($profile_link) + { $doc = new DOMDocument(); if (!@$doc->loadHTMLFile($profile_link)) { return false; @@ -1244,16 +1384,35 @@ 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')]"); + + if ($data["location"] == '') { + $data["location"] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'location')]"); + } + + $data["about"] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'p-note')]"); - // 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["about"] == '') { + $data["about"] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'summary')]"); } - $avatar = $xpath->query("//img[@class='img-rounded media-object']")->item(0); + $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") { @@ -1262,9 +1421,6 @@ class Probe { } } - $data["location"] = $xpath->query("//p[@class='location']")->item(0)->nodeValue; - $data["about"] = $xpath->query("//p[@class='summary']")->item(0)->nodeValue; - return $data; } @@ -1275,9 +1431,9 @@ class Probe { * * @return array pump.io data */ - private static function pumpio($webfinger) { - - $data = array(); + private static function pumpio($webfinger, $addr) + { + $data = []; foreach ($webfinger["links"] as $link) { if (($link["rel"] == "http://webfinger.net/rel/profile-page") && ($link["type"] == "text/html") @@ -1300,7 +1456,7 @@ class Probe { // So we unset all data that isn't used at the moment unset($data["dialback"]); - $data["network"] = NETWORK_PUMPIO; + $data["network"] = Protocol::PUMPIO; } else { return false; } @@ -1313,6 +1469,13 @@ 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 $data; } @@ -1323,7 +1486,8 @@ class Probe { * * @return string feed link */ - private static function getFeedLink($url) { + private static function getFeedLink($url) + { $doc = new DOMDocument(); if (!@$doc->loadHTMLFile($url)) { @@ -1345,7 +1509,7 @@ class Probe { $feed_url = ""; foreach ($feeds as $feed) { - $attr = array(); + $attr = []; foreach ($feed->attributes as $attribute) { $attr[$attribute->name] = trim($attribute->value); } @@ -1361,18 +1525,20 @@ class Probe { /** * @brief Check for feed contact * - * @param string $url Profile link + * @param string $url Profile link * @param boolean $probe Do a probe if the page contains a feed link * * @return array feed data */ - private static function feed($url, $probe = true) { - $ret = z_fetch_url($url); - if ($ret['errno'] == CURLE_OPERATION_TIMEDOUT) { + private static function feed($url, $probe = true) + { + $ret = Network::curl($url); + if (!empty($ret["errno"]) && ($ret['errno'] == CURLE_OPERATION_TIMEDOUT)) { return false; } $feed = $ret['body']; - $feed_data = feed_import($feed, $dummy1, $dummy2, $dummy3, true); + $dummy1 = $dummy2 = $dummy3 = null; + $feed_data = Feed::import($feed, $dummy1, $dummy2, $dummy3, true); if (!$feed_data) { if (!$probe) { @@ -1388,32 +1554,32 @@ class Probe { return self::feed($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; } @@ -1421,32 +1587,40 @@ class Probe { /** * @brief Check for mail contact * - * @param string $uri Profile link + * @param string $uri Profile link * @param integer $uid User ID * * @return array mail data */ - private static function mail($uri, $uid) { + private static function mail($uri, $uid) + { + if (!Network::isEmailDomainValid($uri)) { + return false; + } - if (!validate_email($uri)) { + if ($uid == 0) { return false; } - $x = q("SELECT `prvkey` FROM `user` WHERE `uid` = %d LIMIT 1", intval($uid)); + $user = DBA::selectFirst('user', ['prvkey'], ['uid' => $uid]); - $r = q("SELECT * FROM `mailacct` WHERE `uid` = %d AND `server` != '' LIMIT 1", intval($uid)); + $condition = ["`uid` = ? AND `server` != ''", $uid]; + $fields = ['pass', 'user', 'server', 'port', 'ssltype', 'mailbox']; + $mailacct = DBA::selectFirst('mailacct', $fields, $condition); - if (dbm::is_result($x) && dbm::is_result($r)) { - $mailbox = construct_mailbox_name($r[0]); - $password = ''; - openssl_private_decrypt(hex2bin($r[0]['pass']), $password, $x[0]['prvkey']); - $mbox = email_connect($mailbox, $r[0]['user'], $password); - if (!mbox) { - return false; - } + if (!DBA::isResult($user) || !DBA::isResult($mailacct)) { + return false; } - $msgs = email_poll($mbox, $uri); + $mailbox = Email::constructMailboxName($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, $uri); logger('searching '.$uri.', '.count($msgs).' messages found.', LOGGER_DEBUG); if (!count($msgs)) { @@ -1455,17 +1629,17 @@ class Probe { $phost = substr($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); - $data["url"] = 'http://'.$phost."/".$data["nick"]; + $data["photo"] = Network::lookupAvatarByEmail($uri); + $data["url"] = 'mailto:'.$uri; $data["notify"] = 'smtp '.random_string(); $data["poll"] = 'email '.random_string(); - $x = email_msg_meta($mbox, $msgs[0]); + $x = Email::messageMeta($mbox, $msgs[0]); if (stristr($x[0]->from, $uri)) { $adr = imap_rfc822_parse_adrlist($x[0]->from, ''); } elseif (stristr($x[0]->to, $uri)) { @@ -1491,8 +1665,9 @@ class Probe { } } } - imap_close($mbox); - + if (!empty($mbox)) { + imap_close($mbox); + } return $data; } @@ -1500,11 +1675,12 @@ class Probe { * @brief Mix two paths together to possibly fix missing parts * * @param string $avatar Path to the avatar - * @param string $base Another path that is hopefully complete + * @param string $base Another path that is hopefully complete * * @return string fixed avatar path */ - public static function fixAvatar($avatar, $base) { + public static function fixAvatar($avatar, $base) + { $base_parts = parse_url($base); // Remove all parts that could create a problem