+function poco_reachable($profile, $server = "", $network = "", $force = false) {
+
+ if ($server == "")
+ $server = poco_detect_server($profile);
+
+ if ($server == "")
+ return true;
+
+ return poco_check_server($server, $network, $force);
+}
+
+function poco_detect_server($profile) {
+
+ // Try to detect the server path based upon some known standard paths
+ $server_url = "";
+
+ if ($server_url == "") {
+ $friendica = preg_replace("=(https?://)(.*)/profile/(.*)=ism", "$1$2", $profile);
+ if ($friendica != $profile) {
+ $server_url = $friendica;
+ $network = NETWORK_DFRN;
+ }
+ }
+
+ if ($server_url == "") {
+ $diaspora = preg_replace("=(https?://)(.*)/u/(.*)=ism", "$1$2", $profile);
+ if ($diaspora != $profile) {
+ $server_url = $diaspora;
+ $network = NETWORK_DIASPORA;
+ }
+ }
+
+ if ($server_url == "") {
+ $red = preg_replace("=(https?://)(.*)/channel/(.*)=ism", "$1$2", $profile);
+ if ($red != $profile) {
+ $server_url = $red;
+ $network = NETWORK_DIASPORA;
+ }
+ }
+
+ return $server_url;
+}
+
+function poco_alternate_ostatus_url($url) {
+ return(preg_match("=https?://.+/user/\d+=ism", $url, $matches));
+}
+
+function poco_last_updated($profile, $force = false) {
+
+ $gcontacts = q("SELECT * FROM `gcontact` WHERE `nurl` = '%s'",
+ dbesc(normalise_link($profile)));
+
+ if ($gcontacts[0]["created"] == "0000-00-00 00:00:00")
+ q("UPDATE `gcontact` SET `created` = '%s' WHERE `nurl` = '%s'",
+ dbesc(datetime_convert()), dbesc(normalise_link($profile)));
+
+ if ($gcontacts[0]["server_url"] != "")
+ $server_url = $gcontacts[0]["server_url"];
+ else
+ $server_url = poco_detect_server($profile);
+
+ if ($server_url != "") {
+ if (!poco_check_server($server_url, $gcontacts[0]["network"], $force)) {
+
+ if ($force)
+ q("UPDATE `gcontact` SET `last_failure` = '%s' WHERE `nurl` = '%s'",
+ dbesc(datetime_convert()), dbesc(normalise_link($profile)));
+
+ return false;
+ }
+
+ q("UPDATE `gcontact` SET `server_url` = '%s' WHERE `nurl` = '%s'",
+ dbesc($server_url), dbesc(normalise_link($profile)));
+ }
+
+ if (in_array($gcontacts[0]["network"], array("", NETWORK_FEED))) {
+ $server = q("SELECT `network` FROM `gserver` WHERE `nurl` = '%s' AND `network` != ''",
+ dbesc(normalise_link($server_url)));
+
+ if ($server)
+ q("UPDATE `gcontact` SET `network` = '%s' WHERE `nurl` = '%s'",
+ dbesc($server[0]["network"]), dbesc(normalise_link($profile)));
+ else
+ return;
+ }
+
+ // noscrape is really fast so we don't cache the call.
+ if (($gcontacts[0]["server_url"] != "") AND ($gcontacts[0]["nick"] != "")) {
+
+ // Use noscrape if possible
+ $server = q("SELECT `noscrape` FROM `gserver` WHERE `nurl` = '%s' AND `noscrape` != ''", dbesc(normalise_link($gcontacts[0]["server_url"])));
+
+ if ($server) {
+ $noscraperet = z_fetch_url($server[0]["noscrape"]."/".$gcontacts[0]["nick"]);
+
+ if ($noscraperet["success"] AND ($noscraperet["body"] != "")) {
+
+ $noscrape = json_decode($noscraperet["body"], true);
+
+ if (($noscrape["fn"] != "") AND ($noscrape["fn"] != $gcontacts[0]["name"]))
+ q("UPDATE `gcontact` SET `name` = '%s' WHERE `nurl` = '%s'",
+ dbesc($noscrape["fn"]), dbesc(normalise_link($profile)));
+
+ if (($noscrape["photo"] != "") AND ($noscrape["photo"] != $gcontacts[0]["photo"]))
+ q("UPDATE `gcontact` SET `photo` = '%s' WHERE `nurl` = '%s'",
+ dbesc($noscrape["photo"]), dbesc(normalise_link($profile)));
+
+ if (($noscrape["updated"] != "") AND ($noscrape["updated"] != $gcontacts[0]["updated"]))
+ q("UPDATE `gcontact` SET `updated` = '%s' WHERE `nurl` = '%s'",
+ dbesc($noscrape["updated"]), dbesc(normalise_link($profile)));
+
+ if (($noscrape["gender"] != "") AND ($noscrape["gender"] != $gcontacts[0]["gender"]))
+ q("UPDATE `gcontact` SET `gender` = '%s' WHERE `nurl` = '%s'",
+ dbesc($noscrape["gender"]), dbesc(normalise_link($profile)));
+
+ if (($noscrape["pdesc"] != "") AND ($noscrape["pdesc"] != $gcontacts[0]["about"]))
+ q("UPDATE `gcontact` SET `about` = '%s' WHERE `nurl` = '%s'",
+ dbesc($noscrape["pdesc"]), dbesc(normalise_link($profile)));
+
+ if (($noscrape["about"] != "") AND ($noscrape["about"] != $gcontacts[0]["about"]))
+ q("UPDATE `gcontact` SET `about` = '%s' WHERE `nurl` = '%s'",
+ dbesc($noscrape["about"]), dbesc(normalise_link($profile)));
+
+ if (isset($noscrape["comm"]) AND ($noscrape["comm"] != $gcontacts[0]["community"]))
+ q("UPDATE `gcontact` SET `community` = %d WHERE `nurl` = '%s'",
+ intval($noscrape["comm"]), dbesc(normalise_link($profile)));
+
+ if (isset($noscrape["tags"]))
+ $keywords = implode(" ", $noscrape["tags"]);
+ else
+ $keywords = "";
+
+ if (($keywords != "") AND ($keywords != $gcontacts[0]["keywords"]))
+ q("UPDATE `gcontact` SET `keywords` = '%s' WHERE `nurl` = '%s'",
+ dbesc($keywords), dbesc(normalise_link($profile)));
+
+ $location = $noscrape["locality"];
+
+ if ($noscrape["region"] != "") {
+ if ($location != "")
+ $location .= ", ";
+
+ $location .= $noscrape["region"];
+ }
+
+ if ($noscrape["country-name"] != "") {
+ if ($location != "")
+ $location .= ", ";
+
+ $location .= $noscrape["country-name"];
+ }
+
+ if (($location != "") AND ($location != $gcontacts[0]["location"]))
+ q("UPDATE `gcontact` SET `location` = '%s' WHERE `nurl` = '%s'",
+ dbesc($location), dbesc(normalise_link($profile)));
+
+ // If we got data from noscrape then mark the contact as reachable
+ if (is_array($noscrape) AND count($noscrape))
+ q("UPDATE `gcontact` SET `last_contact` = '%s' WHERE `nurl` = '%s'",
+ dbesc(datetime_convert()), dbesc(normalise_link($profile)));
+
+ return $noscrape["updated"];
+ }
+ }
+ }
+
+ // If we only can poll the feed, then we only do this once a while
+ if (!$force AND !poco_do_update($gcontacts[0]["created"], $gcontacts[0]["updated"], $gcontacts[0]["last_failure"], $gcontacts[0]["last_contact"]))
+ return $gcontacts[0]["updated"];
+
+ $data = probe_url($profile);
+
+ // Is the profile link the alternate OStatus link notation? (http://domain.tld/user/4711)
+ // Then check the other link and delete this one
+ if (($data["network"] == NETWORK_OSTATUS) AND poco_alternate_ostatus_url($profile) AND
+ (normalise_link($profile) == normalise_link($data["alias"])) AND
+ (normalise_link($profile) != normalise_link($data["url"]))) {
+
+ // Delete the old entry
+ q("DELETE FROM `gcontact` WHERE `nurl` = '%s'", dbesc(normalise_link($profile)));
+ q("DELETE FROM `glink` WHERE `gcid` = %d", intval($gcontacts[0]["id"]));
+
+ poco_check($data["url"], $data["name"], $data["network"], $data["photo"], $gcontacts[0]["about"], $gcontacts[0]["location"],
+ $gcontacts[0]["gender"], $gcontacts[0]["keywords"], $data["addr"], $gcontacts[0]["updated"], $gcontacts[0]["generation"]);
+
+ poco_last_updated($data["url"], $force);
+
+ return false;
+ }
+
+ if (($data["poll"] == "") OR (in_array($data["network"], array(NETWORK_FEED, NETWORK_PHANTOM)))) {
+ q("UPDATE `gcontact` SET `last_failure` = '%s' WHERE `nurl` = '%s'",
+ dbesc(datetime_convert()), dbesc(normalise_link($profile)));
+ return false;
+ }
+
+ if (($data["name"] != "") AND ($data["name"] != $gcontacts[0]["name"]))
+ q("UPDATE `gcontact` SET `name` = '%s' WHERE `nurl` = '%s'",
+ dbesc($data["name"]), dbesc(normalise_link($profile)));
+
+ if (($data["nick"] != "") AND ($data["nick"] != $gcontacts[0]["nick"]))
+ q("UPDATE `gcontact` SET `nick` = '%s' WHERE `nurl` = '%s'",
+ dbesc($data["nick"]), dbesc(normalise_link($profile)));
+
+ if (($data["addr"] != "") AND ($data["addr"] != $gcontacts[0]["connect"]))
+ q("UPDATE `gcontact` SET `connect` = '%s' WHERE `nurl` = '%s'",
+ dbesc($data["addr"]), dbesc(normalise_link($profile)));
+
+ if (($data["photo"] != "") AND ($data["photo"] != $gcontacts[0]["photo"]))
+ q("UPDATE `gcontact` SET `photo` = '%s' WHERE `nurl` = '%s'",
+ dbesc($data["photo"]), dbesc(normalise_link($profile)));
+
+ if (($data["baseurl"] != "") AND ($data["baseurl"] != $gcontacts[0]["server_url"]))
+ q("UPDATE `gcontact` SET `server_url` = '%s' WHERE `nurl` = '%s'",
+ dbesc($data["baseurl"]), dbesc(normalise_link($profile)));
+
+ $feedret = z_fetch_url($data["poll"]);
+
+ if (!$feedret["success"]) {
+ q("UPDATE `gcontact` SET `last_failure` = '%s' WHERE `nurl` = '%s'",
+ dbesc(datetime_convert()), dbesc(normalise_link($profile)));
+ return false;
+ }
+
+ $doc = new DOMDocument();
+ @$doc->loadXML($feedret["body"]);
+
+ $xpath = new DomXPath($doc);
+ $xpath->registerNamespace('atom', "http://www.w3.org/2005/Atom");
+
+ $entries = $xpath->query('/atom:feed/atom:entry');
+
+ $last_updated = "";
+
+ foreach ($entries AS $entry) {
+ $published = $xpath->query('atom:published/text()', $entry)->item(0)->nodeValue;
+ $updated = $xpath->query('atom:updated/text()', $entry)->item(0)->nodeValue;
+
+ if ($last_updated < $published)
+ $last_updated = $published;
+
+ if ($last_updated < $updated)
+ $last_updated = $updated;
+ }
+
+ // Maybe there aren't any entries. Then check if it is a valid feed
+ if ($last_updated == "")
+ if ($xpath->query('/atom:feed')->length > 0)
+ $last_updated = "0000-00-00 00:00:00";
+
+ q("UPDATE `gcontact` SET `updated` = '%s', `last_contact` = '%s' WHERE `nurl` = '%s'",
+ dbesc($last_updated), dbesc(datetime_convert()), dbesc(normalise_link($profile)));
+
+ if (($gcontacts[0]["generation"] == 0))
+ q("UPDATE `gcontact` SET `generation` = 9 WHERE `nurl` = '%s'",
+ dbesc(normalise_link($profile)));
+
+ return($last_updated);
+}
+
+function poco_do_update($created, $updated, $last_failure, $last_contact) {
+ $now = strtotime(datetime_convert());
+
+ if ($updated > $last_contact)
+ $contact_time = strtotime($updated);
+ else
+ $contact_time = strtotime($last_contact);
+
+ $failure_time = strtotime($last_failure);
+ $created_time = strtotime($created);
+
+ // If there is no "created" time then use the current time
+ if ($created_time <= 0)
+ $created_time = $now;
+
+ // If the last contact was less than 24 hours then don't update
+ if (($now - $contact_time) < (60 * 60 * 24))
+ return false;
+
+ // If the last failure was less than 24 hours then don't update
+ if (($now - $failure_time) < (60 * 60 * 24))
+ return false;
+
+ // If the last contact was less than a week ago and the last failure is older than a week then don't update
+ //if ((($now - $contact_time) < (60 * 60 * 24 * 7)) AND ($contact_time > $failure_time))
+ // return false;
+
+ // If the last contact time was more than a week ago and the contact was created more than a week ago, then only try once a week
+ if ((($now - $contact_time) > (60 * 60 * 24 * 7)) AND (($now - $created_time) > (60 * 60 * 24 * 7)) AND (($now - $failure_time) < (60 * 60 * 24 * 7)))
+ return false;
+
+ // If the last contact time was more than a month ago and the contact was created more than a month ago, then only try once a month
+ if ((($now - $contact_time) > (60 * 60 * 24 * 30)) AND (($now - $created_time) > (60 * 60 * 24 * 30)) AND (($now - $failure_time) < (60 * 60 * 24 * 30)))
+ return false;
+
+ return true;
+}
+
+function poco_to_boolean($val) {
+ if (($val == "true") OR ($val == 1))
+ return(true);
+ if (($val == "false") OR ($val == 0))
+ return(false);
+
+ return ($val);
+}
+
+function poco_check_server($server_url, $network = "", $force = false) {
+
+ if ($server_url == "")
+ return false;
+
+ $servers = q("SELECT * FROM `gserver` WHERE `nurl` = '%s'", dbesc(normalise_link($server_url)));
+ if ($servers) {
+
+ if ($servers[0]["created"] == "0000-00-00 00:00:00")
+ q("UPDATE `gserver` SET `created` = '%s' WHERE `nurl` = '%s'",
+ dbesc(datetime_convert()), dbesc(normalise_link($server_url)));
+
+ $poco = $servers[0]["poco"];
+ $noscrape = $servers[0]["noscrape"];
+
+ if ($network == "")
+ $network = $servers[0]["network"];
+
+ $last_contact = $servers[0]["last_contact"];
+ $last_failure = $servers[0]["last_failure"];
+ $version = $servers[0]["version"];
+ $platform = $servers[0]["platform"];
+ $site_name = $servers[0]["site_name"];
+ $info = $servers[0]["info"];
+ $register_policy = $servers[0]["register_policy"];
+
+ if (!$force AND !poco_do_update($servers[0]["created"], "", $last_failure, $last_contact)) {
+ logger("Use cached data for server ".$server_url, LOGGER_DEBUG);
+ return ($last_contact >= $last_failure);
+ }
+ } else {
+ $poco = "";
+ $noscrape = "";
+ $version = "";
+ $platform = "";
+ $site_name = "";
+ $info = "";
+ $register_policy = -1;
+
+ $last_contact = "0000-00-00 00:00:00";
+ $last_failure = "0000-00-00 00:00:00";
+ }
+ logger("Server ".$server_url." is outdated or unknown. Start discovery. Force: ".$force." Created: ".$servers[0]["created"]." Failure: ".$last_failure." Contact: ".$last_contact, LOGGER_DEBUG);
+
+ $failure = false;
+ $orig_last_failure = $last_failure;
+
+ // Check if the page is accessible via SSL.
+ $server_url = str_replace("http://", "https://", $server_url);
+ $serverret = z_fetch_url($server_url."/.well-known/host-meta");
+
+ // Maybe the page is unencrypted only?
+ $xmlobj = @simplexml_load_string($serverret["body"],'SimpleXMLElement',0, "http://docs.oasis-open.org/ns/xri/xrd-1.0");
+ if (!$serverret["success"] OR ($serverret["body"] == "") OR (@sizeof($xmlobj) == 0) OR !is_object($xmlobj)) {
+ $server_url = str_replace("https://", "http://", $server_url);
+ $serverret = z_fetch_url($server_url."/.well-known/host-meta");
+
+ $xmlobj = @simplexml_load_string($serverret["body"],'SimpleXMLElement',0, "http://docs.oasis-open.org/ns/xri/xrd-1.0");
+ }
+
+ if (!$serverret["success"] OR ($serverret["body"] == "") OR (sizeof($xmlobj) == 0) OR !is_object($xmlobj)) {
+ // Workaround for bad configured servers (known nginx problem)
+ if ($serverret["debug"]["http_code"] != "403") {
+ $last_failure = datetime_convert();
+ $failure = true;
+ }
+ } elseif ($network == NETWORK_DIASPORA)
+ $last_contact = datetime_convert();
+
+ if (!$failure) {
+ // Test for Diaspora
+ $serverret = z_fetch_url($server_url);
+
+ $lines = explode("\n",$serverret["header"]);
+ if(count($lines))
+ foreach($lines as $line) {
+ $line = trim($line);
+ if(stristr($line,'X-Diaspora-Version:')) {
+ $platform = "Diaspora";
+ $version = trim(str_replace("X-Diaspora-Version:", "", $line));
+ $version = trim(str_replace("x-diaspora-version:", "", $version));
+ $network = NETWORK_DIASPORA;
+ }
+ }
+ }
+
+ if (!$failure) {
+ // Test for Statusnet
+ // Will also return data for Friendica and GNU Social - but it will be overwritten later
+ // The "not implemented" is a special treatment for really, really old Friendica versions
+ $serverret = z_fetch_url($server_url."/api/statusnet/version.json");
+ if ($serverret["success"] AND ($serverret["body"] != '{"error":"not implemented"}') AND ($serverret["body"] != '') AND (strlen($serverret["body"]) < 250)) {
+ $platform = "StatusNet";
+ $version = trim($serverret["body"], '"');
+ $network = NETWORK_OSTATUS;
+ }
+
+ // Test for GNU Social
+ $serverret = z_fetch_url($server_url."/api/gnusocial/version.json");
+ if ($serverret["success"] AND ($serverret["body"] != '{"error":"not implemented"}') AND ($serverret["body"] != '') AND (strlen($serverret["body"]) < 250)) {
+ $platform = "GNU Social";
+ $version = trim($serverret["body"], '"');
+ $network = NETWORK_OSTATUS;
+ }
+
+ $serverret = z_fetch_url($server_url."/api/statusnet/config.json");
+ if ($serverret["success"]) {
+ $data = json_decode($serverret["body"]);
+
+ if (isset($data->site->server)) {
+ $last_contact = datetime_convert();
+
+ if (isset($data->site->hubzilla)) {
+ $platform = $data->site->hubzilla->PLATFORM_NAME;
+ $version = $data->site->hubzilla->RED_VERSION;
+ $network = NETWORK_DIASPORA;
+ }
+ if (isset($data->site->redmatrix)) {
+ if (isset($data->site->redmatrix->PLATFORM_NAME))
+ $platform = $data->site->redmatrix->PLATFORM_NAME;
+ elseif (isset($data->site->redmatrix->RED_PLATFORM))
+ $platform = $data->site->redmatrix->RED_PLATFORM;
+
+ $version = $data->site->redmatrix->RED_VERSION;
+ $network = NETWORK_DIASPORA;
+ }
+ if (isset($data->site->friendica)) {
+ $platform = $data->site->friendica->FRIENDICA_PLATFORM;
+ $version = $data->site->friendica->FRIENDICA_VERSION;
+ $network = NETWORK_DFRN;
+ }
+
+ $site_name = $data->site->name;
+
+ $data->site->closed = poco_to_boolean($data->site->closed);
+ $data->site->private = poco_to_boolean($data->site->private);
+ $data->site->inviteonly = poco_to_boolean($data->site->inviteonly);
+
+ if (!$data->site->closed AND !$data->site->private and $data->site->inviteonly)
+ $register_policy = REGISTER_APPROVE;
+ elseif (!$data->site->closed AND !$data->site->private)
+ $register_policy = REGISTER_OPEN;
+ else
+ $register_policy = REGISTER_CLOSED;
+ }
+ }
+ }
+
+ // Query statistics.json. Optional package for Diaspora, Friendica and Redmatrix
+ if (!$failure) {
+ $serverret = z_fetch_url($server_url."/statistics.json");
+ if ($serverret["success"]) {
+ $data = json_decode($serverret["body"]);
+ if ($version == "")
+ $version = $data->version;
+
+ $site_name = $data->name;
+
+ if (isset($data->network) AND ($platform == ""))
+ $platform = $data->network;
+
+ if ($platform == "Diaspora")
+ $network = NETWORK_DIASPORA;
+
+ if ($data->registrations_open)
+ $register_policy = REGISTER_OPEN;
+ else
+ $register_policy = REGISTER_CLOSED;
+
+ if (isset($data->version))
+ $last_contact = datetime_convert();
+ }
+ }
+
+ // Check for noscrape
+ // Friendica servers could be detected as OStatus servers
+ if (!$failure AND in_array($network, array(NETWORK_DFRN, NETWORK_OSTATUS))) {
+ $serverret = z_fetch_url($server_url."/friendica/json");
+
+ if (!$serverret["success"])
+ $serverret = z_fetch_url($server_url."/friendika/json");
+
+ if ($serverret["success"]) {
+ $data = json_decode($serverret["body"]);
+
+ if (isset($data->version)) {
+ $last_contact = datetime_convert();
+ $network = NETWORK_DFRN;
+
+ $noscrape = $data->no_scrape_url;
+ $version = $data->version;
+ $site_name = $data->site_name;
+ $info = $data->info;
+ $register_policy_str = $data->register_policy;
+ $platform = $data->platform;
+
+ switch ($register_policy_str) {
+ case "REGISTER_CLOSED":
+ $register_policy = REGISTER_CLOSED;
+ break;
+ case "REGISTER_APPROVE":
+ $register_policy = REGISTER_APPROVE;
+ break;
+ case "REGISTER_OPEN":
+ $register_policy = REGISTER_OPEN;
+ break;
+ }
+ }
+ }
+ }
+
+ // Look for poco
+ if (!$failure) {
+ $serverret = z_fetch_url($server_url."/poco");
+ if ($serverret["success"]) {
+ $data = json_decode($serverret["body"]);
+ if (isset($data->totalResults)) {
+ $poco = $server_url."/poco";
+ $last_contact = datetime_convert();
+ }
+ }
+ }
+
+ // Check again if the server exists
+ $servers = q("SELECT `nurl` FROM `gserver` WHERE `nurl` = '%s'", dbesc(normalise_link($server_url)));
+
+ if ($servers)
+ q("UPDATE `gserver` SET `url` = '%s', `version` = '%s', `site_name` = '%s', `info` = '%s', `register_policy` = %d, `poco` = '%s', `noscrape` = '%s',
+ `network` = '%s', `platform` = '%s', `last_contact` = '%s', `last_failure` = '%s' WHERE `nurl` = '%s'",
+ dbesc($server_url),
+ dbesc($version),
+ dbesc($site_name),
+ dbesc($info),
+ intval($register_policy),
+ dbesc($poco),
+ dbesc($noscrape),
+ dbesc($network),
+ dbesc($platform),
+ dbesc($last_contact),
+ dbesc($last_failure),
+ dbesc(normalise_link($server_url))
+ );
+ else
+ q("INSERT INTO `gserver` (`url`, `nurl`, `version`, `site_name`, `info`, `register_policy`, `poco`, `noscrape`, `network`, `platform`, `created`, `last_contact`, `last_failure`)
+ VALUES ('%s', '%s', '%s', '%s', '%s', %d, '%s', '%s', '%s', '%s', '%s', '%s', '%s')",
+ dbesc($server_url),
+ dbesc(normalise_link($server_url)),
+ dbesc($version),
+ dbesc($site_name),
+ dbesc($info),
+ intval($register_policy),
+ dbesc($poco),
+ dbesc($noscrape),
+ dbesc($network),
+ dbesc($platform),
+ dbesc(datetime_convert()),
+ dbesc($last_contact),
+ dbesc($last_failure),
+ dbesc(datetime_convert())
+ );
+
+ logger("End discovery for server ".$server_url, LOGGER_DEBUG);
+
+ return !$failure;
+}
+