friendica/src/Model/GServer.php

1825 lines
55 KiB
PHP
Raw Normal View History

2019-10-03 11:20:36 +02:00
<?php
/**
2021-03-29 08:40:20 +02:00
* @copyright Copyright (C) 2010-2021, the Friendica project
2020-02-09 15:45:36 +01:00
*
* @license GNU AGPL version 3 or any later version
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as
* published by the Free Software Foundation, either version 3 of the
* License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with this program. If not, see <https://www.gnu.org/licenses/>.
*
2019-10-03 11:20:36 +02:00
*/
2020-02-09 15:45:36 +01:00
2019-10-03 11:20:36 +02:00
namespace Friendica\Model;
use DOMDocument;
use DOMXPath;
2021-01-09 20:18:22 +01:00
use Exception;
use Friendica\Core\Logger;
2019-10-03 11:20:36 +02:00
use Friendica\Core\Protocol;
use Friendica\Core\System;
2020-01-05 03:19:02 +01:00
use Friendica\Core\Worker;
2020-11-19 20:34:48 +01:00
use Friendica\Database\Database;
2019-10-03 11:20:36 +02:00
use Friendica\Database\DBA;
use Friendica\DI;
2019-10-03 11:20:36 +02:00
use Friendica\Module\Register;
use Friendica\Network\HTTPClientOptions;
use Friendica\Network\IHTTPResult;
use Friendica\Protocol\Relay;
2019-10-03 11:20:36 +02:00
use Friendica\Util\DateTimeFormat;
use Friendica\Util\Network;
2019-10-03 11:20:36 +02:00
use Friendica\Util\Strings;
use Friendica\Util\XML;
/**
2019-10-04 19:29:21 +02:00
* This class handles GServer related functions
2019-10-03 11:20:36 +02:00
*/
class GServer
{
2019-12-21 14:48:20 +01:00
// Directory types
const DT_NONE = 0;
const DT_POCO = 1;
const DT_MASTODON = 2;
// Methods to detect server types
// Non endpoint specific methods
const DETECT_MANUAL = 0;
const DETECT_HEADER = 1;
const DETECT_BODY = 2;
// Implementation specific endpoints
const DETECT_FRIENDIKA = 10;
const DETECT_FRIENDICA = 11;
const DETECT_STATUSNET = 12;
const DETECT_GNUSOCIAL = 13;
const DETECT_CONFIG_JSON = 14; // Statusnet, GNU Social, Older Hubzilla/Redmatrix
const DETECT_SITEINFO_JSON = 15; // Newer Hubzilla
const DETECT_MASTODON_API = 16;
const DETECT_STATUS_PHP = 17; // Nextcloud
const DETECT_V1_CONFIG = 18;
// Standardized endpoints
const DETECT_STATISTICS_JSON = 100;
const DETECT_NODEINFO_1 = 101;
const DETECT_NODEINFO_2 = 102;
2020-12-03 23:32:51 +01:00
/**
* Check for the existance of a server and adds it in the background if not existant
*
* @param string $url
* @param boolean $only_nodeinfo
* @return void
*/
public static function add(string $url, bool $only_nodeinfo = false)
{
if (self::getID($url, false)) {
return;
}
Worker::add(PRIORITY_LOW, 'UpdateGServer', $url, $only_nodeinfo);
}
/**
* Get the ID for the given server URL
*
* @param string $url
* @param boolean $no_check Don't check if the server hadn't been found
* @return int gserver id
*/
public static function getID(string $url, bool $no_check = false)
{
if (empty($url)) {
return null;
}
$url = self::cleanURL($url);
$gserver = DBA::selectFirst('gserver', ['id'], ['nurl' => Strings::normaliseLink($url)]);
if (DBA::isResult($gserver)) {
Logger::info('Got ID for URL', ['id' => $gserver['id'], 'url' => $url, 'callstack' => System::callstack(20)]);
return $gserver['id'];
}
if ($no_check || !self::check($url)) {
return null;
}
return self::getID($url, true);
}
/**
* Checks if the given server is reachable
*
* @param string $profile URL of the given profile
* @param string $server URL of the given server (If empty, taken from profile)
* @param string $network Network value that is used, when detection failed
* @param boolean $force Force an update.
*
* @return boolean 'true' if server seems vital
*/
public static function reachable(string $profile, string $server = '', string $network = '', bool $force = false)
{
if ($server == '') {
2020-08-01 18:15:18 +02:00
$contact = Contact::getByURL($profile, null, ['baseurl']);
if (!empty($contact['baseurl'])) {
$server = $contact['baseurl'];
}
}
if ($server == '') {
return true;
}
return self::check($server, $network, $force);
}
2020-12-03 16:47:50 +01:00
public static function getNextUpdateDate(bool $success, string $created = '', string $last_contact = '')
{
// On successful contact process check again next week
if ($success) {
return DateTimeFormat::utc('now +7 day');
}
$now = strtotime(DateTimeFormat::utcNow());
if ($created > $last_contact) {
$contact_time = strtotime($created);
} else {
$contact_time = strtotime($last_contact);
}
// If the last contact was less than 6 hours before then try again in 6 hours
if (($now - $contact_time) < (60 * 60 * 6)) {
return DateTimeFormat::utc('now +6 hour');
}
// If the last contact was less than 12 hours before then try again in 12 hours
if (($now - $contact_time) < (60 * 60 * 12)) {
return DateTimeFormat::utc('now +12 hour');
}
// If the last contact was less than 24 hours before then try tomorrow again
if (($now - $contact_time) < (60 * 60 * 24)) {
return DateTimeFormat::utc('now +1 day');
}
2020-12-03 16:47:50 +01:00
// If the last contact was less than a week before then try again in a week
if (($now - $contact_time) < (60 * 60 * 24 * 7)) {
return DateTimeFormat::utc('now +1 week');
}
// If the last contact was less than two weeks before then try again in two week
if (($now - $contact_time) < (60 * 60 * 24 * 14)) {
return DateTimeFormat::utc('now +2 week');
}
// If the last contact was less than a month before then try again in a month
if (($now - $contact_time) < (60 * 60 * 24 * 30)) {
return DateTimeFormat::utc('now +1 month');
}
// The system hadn't been successul contacted for more than a month, so try again in three months
return DateTimeFormat::utc('now +3 month');
}
2019-10-04 19:29:21 +02:00
/**
* Checks the state of the given server.
*
* @param string $server_url URL of the given server
* @param string $network Network value that is used, when detection failed
* @param boolean $force Force an update.
* @param boolean $only_nodeinfo Only use nodeinfo for server detection
2019-10-04 19:29:21 +02:00
*
2019-10-05 06:22:16 +02:00
* @return boolean 'true' if server seems vital
2019-10-04 19:29:21 +02:00
*/
public static function check(string $server_url, string $network = '', bool $force = false, bool $only_nodeinfo = false)
{
$server_url = self::cleanURL($server_url);
if ($server_url == '') {
return false;
}
$gserver = DBA::selectFirst('gserver', [], ['nurl' => Strings::normaliseLink($server_url)]);
if (DBA::isResult($gserver)) {
if ($gserver['created'] <= DBA::NULL_DATETIME) {
$fields = ['created' => DateTimeFormat::utcNow()];
$condition = ['nurl' => Strings::normaliseLink($server_url)];
DBA::update('gserver', $fields, $condition);
}
2020-12-03 23:32:51 +01:00
if (!$force && (strtotime($gserver['next_contact']) > time())) {
2019-10-04 07:42:54 +02:00
Logger::info('No update needed', ['server' => $server_url]);
2020-12-03 23:32:51 +01:00
return (!$gserver['failed']);
}
2020-12-03 23:32:51 +01:00
Logger::info('Server is outdated. Start discovery.', ['Server' => $server_url, 'Force' => $force]);
2019-10-04 07:42:54 +02:00
} else {
Logger::info('Server is unknown. Start discovery.', ['Server' => $server_url]);
}
return self::detect($server_url, $network, $only_nodeinfo);
}
/**
* Set failed server status
*
* @param string $url
*/
2020-12-03 16:47:50 +01:00
public static function setFailure(string $url)
{
2020-12-03 16:47:50 +01:00
$gserver = DBA::selectFirst('gserver', [], ['nurl' => Strings::normaliseLink($url)]);
if (DBA::isResult($gserver)) {
$next_update = self::getNextUpdateDate(false, $gserver['created'], $gserver['last_contact']);
DBA::update('gserver', ['failed' => true, 'last_failure' => DateTimeFormat::utcNow(),
'next_contact' => $next_update, 'detection-method' => null],
['nurl' => Strings::normaliseLink($url)]);
Logger::info('Set failed status for existing server', ['url' => $url]);
return;
}
DBA::insert('gserver', ['url' => $url, 'nurl' => Strings::normaliseLink($url),
'network' => Protocol::PHANTOM, 'created' => DateTimeFormat::utcNow(),
'failed' => true, 'last_failure' => DateTimeFormat::utcNow()]);
Logger::info('Set failed status for new server', ['url' => $url]);
}
2020-05-22 06:52:43 +02:00
/**
* Remove unwanted content from the given URL
*
* @param string $url
* @return string cleaned URL
*/
public static function cleanURL(string $url)
{
$url = trim($url, '/');
$url = str_replace('/index.php', '', $url);
$urlparts = parse_url($url);
unset($urlparts['user']);
unset($urlparts['pass']);
unset($urlparts['query']);
unset($urlparts['fragment']);
return Network::unparseURL($urlparts);
}
2020-05-22 06:52:43 +02:00
/**
* Return the base URL
*
* @param string $url
* @return string base URL
*/
private static function getBaseURL(string $url)
{
2020-05-22 06:52:43 +02:00
$urlparts = parse_url(self::cleanURL($url));
unset($urlparts['path']);
return Network::unparseURL($urlparts);
}
2019-10-03 11:20:36 +02:00
/**
2019-10-03 16:48:46 +02:00
* Detect server data (type, protocol, version number, ...)
* The detected data is then updated or inserted in the gserver table.
2019-10-03 11:20:36 +02:00
*
* @param string $url URL of the given server
* @param string $network Network value that is used, when detection failed
* @param boolean $only_nodeinfo Only use nodeinfo for server detection
2019-10-03 11:20:36 +02:00
*
* @return boolean 'true' if server could be detected
*/
public static function detect(string $url, string $network = '', bool $only_nodeinfo = false)
2019-10-03 11:20:36 +02:00
{
2019-12-21 14:48:20 +01:00
Logger::info('Detect server type', ['server' => $url]);
$serverdata = ['detection-method' => self::DETECT_MANUAL];
2019-10-03 11:20:36 +02:00
2019-12-21 14:48:20 +01:00
$original_url = $url;
// Remove URL content that is not supposed to exist for a server url
$url = self::cleanURL($url);
// Get base URL
$baseurl = self::getBaseURL($url);
2019-12-21 14:48:20 +01:00
// If the URL missmatches, then we mark the old entry as failure
if ($url != $original_url) {
2020-12-03 16:47:50 +01:00
/// @todo What to do with "next_contact" here?
DBA::update('gserver', ['failed' => true, 'last_failure' => DateTimeFormat::utcNow()],
['nurl' => Strings::normaliseLink($original_url)]);
2019-12-21 14:48:20 +01:00
}
2019-10-03 11:20:36 +02:00
// When a nodeinfo is present, we don't need to dig further
$xrd_timeout = DI::config()->get('system', 'xrd_timeout');
$curlResult = DI::httpClient()->get($url . '/.well-known/nodeinfo', [HTTPClientOptions::TIMEOUT => $xrd_timeout]);
2019-10-03 11:20:36 +02:00
if ($curlResult->isTimeout()) {
self::setFailure($url);
2019-10-03 11:20:36 +02:00
return false;
}
// On a redirect follow the new host but mark the old one as failure
if ($curlResult->isSuccess() && (parse_url($url, PHP_URL_HOST) != parse_url($curlResult->getRedirectUrl(), PHP_URL_HOST))) {
$curlResult = DI::httpClient()->get($url, [HTTPClientOptions::TIMEOUT => $xrd_timeout]);
if (parse_url($url, PHP_URL_HOST) != parse_url($curlResult->getRedirectUrl(), PHP_URL_HOST)) {
Logger::info('Found redirect. Mark old entry as failure', ['old' => $url, 'new' => $curlResult->getRedirectUrl()]);
self::setFailure($url);
self::detect($curlResult->getRedirectUrl(), $network, $only_nodeinfo);
return false;
}
}
2019-10-03 11:20:36 +02:00
$nodeinfo = self::fetchNodeinfo($url, $curlResult);
if ($only_nodeinfo && empty($nodeinfo)) {
Logger::info('Invalid nodeinfo in nodeinfo-mode, server is marked as failure', ['url' => $url]);
self::setFailure($url);
return false;
}
2019-10-03 11:20:36 +02:00
2019-10-03 16:48:46 +02:00
// When nodeinfo isn't present, we use the older 'statistics.json' endpoint
2019-10-03 15:02:48 +02:00
if (empty($nodeinfo)) {
$nodeinfo = self::fetchStatistics($url);
}
2019-10-03 16:48:46 +02:00
// If that didn't work out well, we use some protocol specific endpoints
// For Friendica and Zot based networks we have to dive deeper to reveal more details
if (empty($nodeinfo['network']) || in_array($nodeinfo['network'], [Protocol::DFRN, Protocol::ZOT])) {
if (!empty($nodeinfo['detection-method'])) {
$serverdata['detection-method'] = $nodeinfo['detection-method'];
}
2019-10-03 11:20:36 +02:00
// Fetch the landing page, possibly it reveals some data
if (empty($nodeinfo['network'])) {
if ($baseurl == $url) {
$basedata = $serverdata;
} else {
$basedata = ['detection-method' => self::DETECT_MANUAL];
}
$curlResult = DI::httpClient()->get($baseurl, [HTTPClientOptions::TIMEOUT => $xrd_timeout]);
if ($curlResult->isSuccess()) {
if ((parse_url($baseurl, PHP_URL_HOST) != parse_url($curlResult->getRedirectUrl(), PHP_URL_HOST))) {
Logger::info('Found redirect. Mark old entry as failure', ['old' => $url, 'new' => $curlResult->getRedirectUrl()]);
self::setFailure($url);
self::detect($curlResult->getRedirectUrl(), $network, $only_nodeinfo);
return false;
}
$basedata = self::analyseRootHeader($curlResult, $basedata);
$basedata = self::analyseRootBody($curlResult, $basedata, $baseurl);
}
2020-01-12 13:50:00 +01:00
if (!$curlResult->isSuccess() || empty($curlResult->getBody()) || self::invalidBody($curlResult->getBody())) {
self::setFailure($url);
return false;
}
if ($baseurl == $url) {
$serverdata = $basedata;
} else {
// When the base path doesn't seem to contain a social network we try the complete path.
// Most detectable system have to be installed in the root directory.
// We checked the base to avoid false positives.
$curlResult = DI::httpClient()->get($url, [HTTPClientOptions::TIMEOUT => $xrd_timeout]);
if ($curlResult->isSuccess()) {
$urldata = self::analyseRootHeader($curlResult, $serverdata);
$urldata = self::analyseRootBody($curlResult, $urldata, $url);
$comparebase = $basedata;
unset($comparebase['info']);
unset($comparebase['site_name']);
$compareurl = $urldata;
unset($compareurl['info']);
unset($compareurl['site_name']);
// We assume that no one will install the identical system in the root and a subfolder
if (!empty(array_diff($comparebase, $compareurl))) {
$serverdata = $urldata;
}
}
}
}
if (empty($serverdata['network']) || ($serverdata['network'] == Protocol::ACTIVITYPUB)) {
$serverdata = self::detectMastodonAlikes($url, $serverdata);
2019-10-03 11:20:36 +02:00
}
// All following checks are done for systems that always have got a "host-meta" endpoint.
// With this check we don't have to waste time and ressources for dead systems.
// Also this hopefully prevents us from receiving abuse messages.
if (empty($serverdata['network']) && !self::validHostMeta($url)) {
self::setFailure($url);
2019-10-03 11:20:36 +02:00
return false;
}
if (empty($serverdata['network']) || in_array($serverdata['network'], [Protocol::DFRN, Protocol::ACTIVITYPUB])) {
2019-10-03 11:20:36 +02:00
$serverdata = self::detectFriendica($url, $serverdata);
}
2019-10-03 15:02:48 +02:00
// the 'siteinfo.json' is some specific endpoint of Hubzilla and Red
if (empty($serverdata['network']) || ($serverdata['network'] == Protocol::ZOT)) {
2019-10-03 11:20:36 +02:00
$serverdata = self::fetchSiteinfo($url, $serverdata);
}
2019-10-03 16:48:46 +02:00
// The 'siteinfo.json' doesn't seem to be present on older Hubzilla installations
2019-10-03 15:02:48 +02:00
if (empty($serverdata['network'])) {
$serverdata = self::detectHubzilla($url, $serverdata);
}
2020-12-22 09:24:56 +01:00
if (empty($serverdata['network']) || in_array($serverdata['detection-method'], [self::DETECT_MANUAL, self::DETECT_BODY])) {
$serverdata = self::detectPeertube($url, $serverdata);
}
2019-10-03 11:20:36 +02:00
if (empty($serverdata['network'])) {
$serverdata = self::detectNextcloud($url, $serverdata);
}
if (empty($serverdata['network'])) {
$serverdata = self::detectGNUSocial($url, $serverdata);
}
$serverdata = array_merge($nodeinfo, $serverdata);
2019-10-03 11:20:36 +02:00
} else {
$serverdata = $nodeinfo;
}
2019-12-21 14:48:20 +01:00
// Detect the directory type
$serverdata['directory-type'] = self::DT_NONE;
2019-10-03 17:30:07 +02:00
$serverdata = self::checkPoCo($url, $serverdata);
2019-12-21 14:48:20 +01:00
$serverdata = self::checkMastodonDirectory($url, $serverdata);
2019-10-03 17:30:07 +02:00
2019-10-03 11:20:36 +02:00
// We can't detect the network type. Possibly it is some system that we don't know yet
if (empty($serverdata['network'])) {
$serverdata['network'] = Protocol::PHANTOM;
}
2019-10-04 08:33:16 +02:00
// When we hadn't been able to detect the network type, we use the hint from the parameter
if (($serverdata['network'] == Protocol::PHANTOM) && !empty($network)) {
$serverdata['network'] = $network;
}
2019-10-03 11:20:36 +02:00
$serverdata['url'] = $url;
$serverdata['nurl'] = Strings::normaliseLink($url);
2019-10-03 17:30:07 +02:00
// We take the highest number that we do find
$registeredUsers = $serverdata['registered-users'] ?? 0;
// On an active server there has to be at least a single user
if (($serverdata['network'] != Protocol::PHANTOM) && ($registeredUsers == 0)) {
$registeredUsers = 1;
}
if ($serverdata['network'] == Protocol::PHANTOM) {
$serverdata['registered-users'] = max($registeredUsers, 1);
2019-10-04 08:33:16 +02:00
$serverdata = self::detectNetworkViaContacts($url, $serverdata);
}
2019-10-03 17:30:07 +02:00
2020-12-03 16:47:50 +01:00
$serverdata['next_contact'] = self::getNextUpdateDate(true);
2019-10-03 16:48:46 +02:00
$serverdata['last_contact'] = DateTimeFormat::utcNow();
$serverdata['failed'] = false;
2019-10-03 16:48:46 +02:00
2019-10-04 08:33:16 +02:00
$gserver = DBA::selectFirst('gserver', ['network'], ['nurl' => Strings::normaliseLink($url)]);
if (!DBA::isResult($gserver)) {
2019-10-03 11:20:36 +02:00
$serverdata['created'] = DateTimeFormat::utcNow();
2019-10-03 16:48:46 +02:00
$ret = DBA::insert('gserver', $serverdata);
$id = DBA::lastInsertId();
2019-10-03 11:20:36 +02:00
} else {
2019-10-05 06:22:16 +02:00
// Don't override the network with 'unknown' when there had been a valid entry before
2019-10-04 08:33:16 +02:00
if (($serverdata['network'] == Protocol::PHANTOM) && !empty($gserver['network'])) {
unset($serverdata['network']);
}
2019-10-03 16:48:46 +02:00
$ret = DBA::update('gserver', $serverdata, ['nurl' => $serverdata['nurl']]);
$gserver = DBA::selectFirst('gserver', ['id'], ['nurl' => $serverdata['nurl']]);
if (DBA::isResult($gserver)) {
$id = $gserver['id'];
}
}
if (!empty($serverdata['network']) && !empty($id) && ($serverdata['network'] != Protocol::PHANTOM)) {
$apcontacts = DBA::count('apcontact', ['gsid' => $id]);
$contacts = DBA::count('contact', ['uid' => 0, 'gsid' => $id]);
$max_users = max($apcontacts, $contacts, $registeredUsers, 1);
if ($max_users > $registeredUsers) {
Logger::info('Update registered users', ['id' => $id, 'url' => $serverdata['nurl'], 'registered-users' => $max_users]);
DBA::update('gserver', ['registered-users' => $max_users], ['id' => $id]);
}
2019-10-03 11:20:36 +02:00
}
2019-10-04 08:33:16 +02:00
if (!empty($serverdata['network']) && in_array($serverdata['network'], [Protocol::DFRN, Protocol::DIASPORA])) {
self::discoverRelay($url);
}
2019-10-03 16:48:46 +02:00
return $ret;
2019-10-03 11:20:36 +02:00
}
/**
2019-10-04 19:29:21 +02:00
* Fetch relay data from a given server url
*
* @param string $server_url address of the server
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
2020-05-17 15:59:05 +02:00
private static function discoverRelay(string $server_url)
{
Logger::info('Discover relay data', ['server' => $server_url]);
$curlResult = DI::httpClient()->get($server_url . '/.well-known/x-social-relay');
if (!$curlResult->isSuccess()) {
return;
}
$data = json_decode($curlResult->getBody(), true);
if (!is_array($data)) {
return;
}
// Sanitize incoming data, see https://github.com/friendica/friendica/issues/8565
$data['subscribe'] = (bool)$data['subscribe'] ?? false;
if (!$data['subscribe'] || empty($data['scope']) || !in_array(strtolower($data['scope']), ['all', 'tags'])) {
$data['scope'] = '';
$data['subscribe'] = false;
$data['tags'] = [];
}
$gserver = DBA::selectFirst('gserver', ['id', 'url', 'network', 'relay-subscribe', 'relay-scope'], ['nurl' => Strings::normaliseLink($server_url)]);
if (!DBA::isResult($gserver)) {
return;
}
if (($gserver['relay-subscribe'] != $data['subscribe']) || ($gserver['relay-scope'] != $data['scope'])) {
$fields = ['relay-subscribe' => $data['subscribe'], 'relay-scope' => $data['scope']];
DBA::update('gserver', $fields, ['id' => $gserver['id']]);
}
DBA::delete('gserver-tag', ['gserver-id' => $gserver['id']]);
if ($data['scope'] == 'tags') {
// Avoid duplicates
$tags = [];
foreach ($data['tags'] as $tag) {
$tag = mb_strtolower($tag);
if (strlen($tag) < 100) {
$tags[$tag] = $tag;
}
}
foreach ($tags as $tag) {
2020-11-19 20:34:48 +01:00
DBA::insert('gserver-tag', ['gserver-id' => $gserver['id'], 'tag' => $tag], Database::INSERT_IGNORE);
}
}
// Create or update the relay contact
$fields = [];
if (isset($data['protocols'])) {
if (isset($data['protocols']['diaspora'])) {
$fields['network'] = Protocol::DIASPORA;
if (isset($data['protocols']['diaspora']['receive'])) {
$fields['batch'] = $data['protocols']['diaspora']['receive'];
} elseif (is_string($data['protocols']['diaspora'])) {
$fields['batch'] = $data['protocols']['diaspora'];
}
}
if (isset($data['protocols']['dfrn'])) {
$fields['network'] = Protocol::DFRN;
if (isset($data['protocols']['dfrn']['receive'])) {
$fields['batch'] = $data['protocols']['dfrn']['receive'];
} elseif (is_string($data['protocols']['dfrn'])) {
$fields['batch'] = $data['protocols']['dfrn'];
}
}
if (isset($data['protocols']['activitypub'])) {
$fields['network'] = Protocol::ACTIVITYPUB;
if (!empty($data['protocols']['activitypub']['actor'])) {
$fields['url'] = $data['protocols']['activitypub']['actor'];
}
if (!empty($data['protocols']['activitypub']['receive'])) {
$fields['batch'] = $data['protocols']['activitypub']['receive'];
}
}
}
Logger::info('Discovery ended', ['server' => $server_url, 'data' => $fields]);
Relay::updateContact($gserver, $fields);
}
2019-10-04 19:29:21 +02:00
/**
* Fetch server data from '/statistics.json' on the given server
*
* @param string $url URL of the given server
*
* @return array server data
*/
private static function fetchStatistics(string $url)
2019-10-03 15:02:48 +02:00
{
$curlResult = DI::httpClient()->get($url . '/statistics.json');
2019-10-03 15:02:48 +02:00
if (!$curlResult->isSuccess()) {
return [];
}
$data = json_decode($curlResult->getBody(), true);
if (empty($data)) {
return [];
}
$serverdata = ['detection-method' => self::DETECT_STATISTICS_JSON];
2019-10-03 15:02:48 +02:00
if (!empty($data['version'])) {
$serverdata['version'] = $data['version'];
// Version numbers on statistics.json are presented with additional info, e.g.:
// 0.6.3.0-p1702cc1c, 0.6.99.0-p1b9ab160 or 3.4.3-2-1191.
$serverdata['version'] = preg_replace('=(.+)-(.{4,})=ism', '$1', $serverdata['version']);
}
if (!empty($data['name'])) {
$serverdata['site_name'] = $data['name'];
}
if (!empty($data['network'])) {
2020-01-12 22:07:40 +01:00
$serverdata['platform'] = strtolower($data['network']);
2019-10-03 15:02:48 +02:00
2020-01-12 22:07:40 +01:00
if ($serverdata['platform'] == 'diaspora') {
2019-10-03 15:02:48 +02:00
$serverdata['network'] = Protocol::DIASPORA;
2020-01-12 22:07:40 +01:00
} elseif ($serverdata['platform'] == 'friendica') {
2019-10-03 15:02:48 +02:00
$serverdata['network'] = Protocol::DFRN;
2019-10-03 16:48:46 +02:00
} elseif ($serverdata['platform'] == 'hubzilla') {
$serverdata['network'] = Protocol::ZOT;
2019-10-03 15:02:48 +02:00
} elseif ($serverdata['platform'] == 'redmatrix') {
$serverdata['network'] = Protocol::ZOT;
}
}
if (!empty($data['registrations_open'])) {
$serverdata['register_policy'] = Register::OPEN;
} else {
$serverdata['register_policy'] = Register::CLOSED;
}
return $serverdata;
}
2019-10-03 11:20:36 +02:00
/**
2019-10-04 19:29:21 +02:00
* Detect server type by using the nodeinfo data
2019-10-03 11:20:36 +02:00
*
* @param string $url address of the server
* @param IHTTPResult $httpResult
*
2019-10-03 11:20:36 +02:00
* @return array Server data
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
private static function fetchNodeinfo(string $url, IHTTPResult $httpResult)
2019-10-03 11:20:36 +02:00
{
if (!$httpResult->isSuccess()) {
return [];
}
$nodeinfo = json_decode($httpResult->getBody(), true);
2019-10-03 11:20:36 +02:00
2019-10-03 16:48:46 +02:00
if (!is_array($nodeinfo) || empty($nodeinfo['links'])) {
2019-10-03 11:20:36 +02:00
return [];
}
$nodeinfo1_url = '';
$nodeinfo2_url = '';
foreach ($nodeinfo['links'] as $link) {
if (!is_array($link) || empty($link['rel']) || empty($link['href'])) {
Logger::info('Invalid nodeinfo format', ['url' => $url]);
continue;
}
if ($link['rel'] == 'http://nodeinfo.diaspora.software/ns/schema/1.0') {
$nodeinfo1_url = $link['href'];
} elseif ($link['rel'] == 'http://nodeinfo.diaspora.software/ns/schema/2.0') {
$nodeinfo2_url = $link['href'];
}
}
if ($nodeinfo1_url . $nodeinfo2_url == '') {
return [];
}
$server = [];
// When the nodeinfo url isn't on the same host, then there is obviously something wrong
if (!empty($nodeinfo2_url) && (parse_url($url, PHP_URL_HOST) == parse_url($nodeinfo2_url, PHP_URL_HOST))) {
$server = self::parseNodeinfo2($nodeinfo2_url);
}
// When the nodeinfo url isn't on the same host, then there is obviously something wrong
if (empty($server) && !empty($nodeinfo1_url) && (parse_url($url, PHP_URL_HOST) == parse_url($nodeinfo1_url, PHP_URL_HOST))) {
$server = self::parseNodeinfo1($nodeinfo1_url);
}
return $server;
}
/**
2019-10-04 19:29:21 +02:00
* Parses Nodeinfo 1
2019-10-03 11:20:36 +02:00
*
* @param string $nodeinfo_url address of the nodeinfo path
* @return array Server data
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
2019-10-04 19:29:21 +02:00
private static function parseNodeinfo1(string $nodeinfo_url)
2019-10-03 11:20:36 +02:00
{
$curlResult = DI::httpClient()->get($nodeinfo_url);
2019-10-03 11:20:36 +02:00
if (!$curlResult->isSuccess()) {
2019-10-05 06:50:29 +02:00
return [];
2019-10-03 11:20:36 +02:00
}
$nodeinfo = json_decode($curlResult->getBody(), true);
if (!is_array($nodeinfo)) {
2019-10-05 06:50:29 +02:00
return [];
2019-10-03 11:20:36 +02:00
}
$server = ['detection-method' => self::DETECT_NODEINFO_1,
'register_policy' => Register::CLOSED];
2019-10-03 11:20:36 +02:00
if (!empty($nodeinfo['openRegistrations'])) {
$server['register_policy'] = Register::OPEN;
}
if (is_array($nodeinfo['software'])) {
2019-10-03 16:48:46 +02:00
if (!empty($nodeinfo['software']['name'])) {
2020-01-12 22:07:40 +01:00
$server['platform'] = strtolower($nodeinfo['software']['name']);
2019-10-03 11:20:36 +02:00
}
2019-10-03 16:48:46 +02:00
if (!empty($nodeinfo['software']['version'])) {
2019-10-03 11:20:36 +02:00
$server['version'] = $nodeinfo['software']['version'];
// Version numbers on Nodeinfo are presented with additional info, e.g.:
// 0.6.3.0-p1702cc1c, 0.6.99.0-p1b9ab160 or 3.4.3-2-1191.
$server['version'] = preg_replace('=(.+)-(.{4,})=ism', '$1', $server['version']);
}
}
2019-10-03 16:48:46 +02:00
if (!empty($nodeinfo['metadata']['nodeName'])) {
2019-10-03 11:20:36 +02:00
$server['site_name'] = $nodeinfo['metadata']['nodeName'];
}
if (!empty($nodeinfo['usage']['users']['total'])) {
$server['registered-users'] = max($nodeinfo['usage']['users']['total'], 1);
2019-10-03 11:20:36 +02:00
}
if (!empty($nodeinfo['protocols']['inbound']) && is_array($nodeinfo['protocols']['inbound'])) {
$protocols = [];
foreach ($nodeinfo['protocols']['inbound'] as $protocol) {
$protocols[$protocol] = true;
}
if (!empty($protocols['friendica'])) {
$server['network'] = Protocol::DFRN;
} elseif (!empty($protocols['activitypub'])) {
$server['network'] = Protocol::ACTIVITYPUB;
} elseif (!empty($protocols['diaspora'])) {
$server['network'] = Protocol::DIASPORA;
} elseif (!empty($protocols['ostatus'])) {
$server['network'] = Protocol::OSTATUS;
} elseif (!empty($protocols['gnusocial'])) {
$server['network'] = Protocol::OSTATUS;
2019-10-03 15:02:48 +02:00
} elseif (!empty($protocols['zot'])) {
$server['network'] = Protocol::ZOT;
2019-10-03 11:20:36 +02:00
}
}
2019-10-05 06:50:29 +02:00
if (empty($server)) {
return [];
2019-10-03 11:20:36 +02:00
}
return $server;
}
/**
2019-10-04 19:29:21 +02:00
* Parses Nodeinfo 2
2019-10-03 11:20:36 +02:00
*
* @see https://git.feneas.org/jaywink/nodeinfo2
2019-10-03 11:20:36 +02:00
* @param string $nodeinfo_url address of the nodeinfo path
* @return array Server data
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
2019-10-04 19:29:21 +02:00
private static function parseNodeinfo2(string $nodeinfo_url)
2019-10-03 11:20:36 +02:00
{
$curlResult = DI::httpClient()->get($nodeinfo_url);
2019-10-03 11:20:36 +02:00
if (!$curlResult->isSuccess()) {
2019-10-05 06:50:29 +02:00
return [];
2019-10-03 11:20:36 +02:00
}
$nodeinfo = json_decode($curlResult->getBody(), true);
if (!is_array($nodeinfo)) {
2019-10-05 06:50:29 +02:00
return [];
2019-10-03 11:20:36 +02:00
}
$server = ['detection-method' => self::DETECT_NODEINFO_2,
'register_policy' => Register::CLOSED];
2019-10-03 11:20:36 +02:00
if (!empty($nodeinfo['openRegistrations'])) {
$server['register_policy'] = Register::OPEN;
}
if (is_array($nodeinfo['software'])) {
2019-10-03 16:48:46 +02:00
if (!empty($nodeinfo['software']['name'])) {
2020-01-12 22:07:40 +01:00
$server['platform'] = strtolower($nodeinfo['software']['name']);
2019-10-03 11:20:36 +02:00
}
2019-10-03 16:48:46 +02:00
if (!empty($nodeinfo['software']['version'])) {
2019-10-03 11:20:36 +02:00
$server['version'] = $nodeinfo['software']['version'];
// Version numbers on Nodeinfo are presented with additional info, e.g.:
// 0.6.3.0-p1702cc1c, 0.6.99.0-p1b9ab160 or 3.4.3-2-1191.
$server['version'] = preg_replace('=(.+)-(.{4,})=ism', '$1', $server['version']);
}
}
2019-10-03 16:48:46 +02:00
if (!empty($nodeinfo['metadata']['nodeName'])) {
2019-10-03 11:20:36 +02:00
$server['site_name'] = $nodeinfo['metadata']['nodeName'];
}
if (!empty($nodeinfo['usage']['users']['total'])) {
$server['registered-users'] = max($nodeinfo['usage']['users']['total'], 1);
2019-10-03 11:20:36 +02:00
}
if (!empty($nodeinfo['protocols'])) {
$protocols = [];
foreach ($nodeinfo['protocols'] as $protocol) {
if (is_string($protocol)) {
$protocols[$protocol] = true;
}
2019-10-03 11:20:36 +02:00
}
2019-12-21 14:48:20 +01:00
if (!empty($protocols['dfrn'])) {
2019-10-03 11:20:36 +02:00
$server['network'] = Protocol::DFRN;
} elseif (!empty($protocols['activitypub'])) {
$server['network'] = Protocol::ACTIVITYPUB;
} elseif (!empty($protocols['diaspora'])) {
$server['network'] = Protocol::DIASPORA;
} elseif (!empty($protocols['ostatus'])) {
$server['network'] = Protocol::OSTATUS;
} elseif (!empty($protocols['gnusocial'])) {
$server['network'] = Protocol::OSTATUS;
2019-10-03 15:02:48 +02:00
} elseif (!empty($protocols['zot'])) {
$server['network'] = Protocol::ZOT;
2019-10-03 11:20:36 +02:00
}
}
if (empty($server)) {
2019-10-05 06:50:29 +02:00
return [];
2019-10-03 11:20:36 +02:00
}
return $server;
}
2019-10-04 19:29:21 +02:00
/**
2019-10-05 06:22:16 +02:00
* Fetch server information from a 'siteinfo.json' file on the given server
2019-10-04 19:29:21 +02:00
*
* @param string $url URL of the given server
* @param array $serverdata array with server data
*
* @return array server data
*/
private static function fetchSiteinfo(string $url, array $serverdata)
2019-10-03 11:20:36 +02:00
{
$curlResult = DI::httpClient()->get($url . '/siteinfo.json');
2019-10-03 11:20:36 +02:00
if (!$curlResult->isSuccess()) {
return $serverdata;
}
$data = json_decode($curlResult->getBody(), true);
if (empty($data)) {
return $serverdata;
}
if (in_array($serverdata['detection-method'], [self::DETECT_HEADER, self::DETECT_BODY, self::DETECT_MANUAL])) {
$serverdata['detection-method'] = self::DETECT_SITEINFO_JSON;
}
2019-10-03 16:48:46 +02:00
if (!empty($data['url'])) {
2020-01-12 22:07:40 +01:00
$serverdata['platform'] = strtolower($data['platform']);
2019-10-03 11:20:36 +02:00
$serverdata['version'] = $data['version'];
}
if (!empty($data['plugins'])) {
if (in_array('pubcrawl', $data['plugins'])) {
$serverdata['network'] = Protocol::ACTIVITYPUB;
} elseif (in_array('diaspora', $data['plugins'])) {
$serverdata['network'] = Protocol::DIASPORA;
} elseif (in_array('gnusoc', $data['plugins'])) {
$serverdata['network'] = Protocol::OSTATUS;
} else {
$serverdata['network'] = Protocol::ZOT;
}
}
if (!empty($data['site_name'])) {
$serverdata['site_name'] = $data['site_name'];
}
if (!empty($data['channels_total'])) {
$serverdata['registered-users'] = max($data['channels_total'], 1);
2019-10-03 11:20:36 +02:00
}
if (!empty($data['register_policy'])) {
switch ($data['register_policy']) {
2019-10-03 15:02:48 +02:00
case 'REGISTER_OPEN':
2019-10-03 11:20:36 +02:00
$serverdata['register_policy'] = Register::OPEN;
break;
2019-10-03 15:02:48 +02:00
case 'REGISTER_APPROVE':
2019-10-03 11:20:36 +02:00
$serverdata['register_policy'] = Register::APPROVE;
break;
2019-10-03 15:02:48 +02:00
case 'REGISTER_CLOSED':
2019-10-03 11:20:36 +02:00
default:
$serverdata['register_policy'] = Register::CLOSED;
break;
}
}
return $serverdata;
}
2019-10-04 19:29:21 +02:00
/**
* Checks if the server contains a valid host meta file
*
* @param string $url URL of the given server
*
2019-10-05 06:22:16 +02:00
* @return boolean 'true' if the server seems to be vital
2019-10-04 19:29:21 +02:00
*/
private static function validHostMeta(string $url)
{
$xrd_timeout = DI::config()->get('system', 'xrd_timeout');
$curlResult = DI::httpClient()->get($url . '/.well-known/host-meta', [HTTPClientOptions::TIMEOUT => $xrd_timeout]);
if (!$curlResult->isSuccess()) {
return false;
}
2020-04-27 16:35:50 +02:00
$xrd = XML::parseString($curlResult->getBody());
if (!is_object($xrd)) {
return false;
}
$elements = XML::elementToArray($xrd);
if (empty($elements) || empty($elements['xrd']) || empty($elements['xrd']['link'])) {
return false;
}
$valid = false;
foreach ($elements['xrd']['link'] as $link) {
// When there is more than a single "link" element, the array looks slightly different
if (!empty($link['@attributes'])) {
$link = $link['@attributes'];
}
if (empty($link['rel']) || empty($link['template'])) {
continue;
}
if ($link['rel'] == 'lrdd') {
// When the webfinger host is the same like the system host, it should be ok.
$valid = (parse_url($url, PHP_URL_HOST) == parse_url($link['template'], PHP_URL_HOST));
}
}
return $valid;
}
2019-10-04 19:29:21 +02:00
/**
* Detect the network of the given server via their known contacts
*
* @param string $url URL of the given server
* @param array $serverdata array with server data
*
* @return array server data
*/
private static function detectNetworkViaContacts(string $url, array $serverdata)
2019-10-04 08:33:16 +02:00
{
$contacts = [];
2019-10-04 08:33:16 +02:00
$apcontacts = DBA::select('apcontact', ['url'], ['baseurl' => [$url, $serverdata['nurl']]]);
2020-04-27 16:35:50 +02:00
while ($apcontact = DBA::fetch($apcontacts)) {
$contacts[Strings::normaliseLink($apcontact['url'])] = $apcontact['url'];
2019-10-04 08:33:16 +02:00
}
DBA::close($apcontacts);
$pcontacts = DBA::select('contact', ['url', 'nurl'], ['uid' => 0, 'baseurl' => [$url, $serverdata['nurl']]]);
2020-04-27 16:35:50 +02:00
while ($pcontact = DBA::fetch($pcontacts)) {
2019-10-04 08:33:16 +02:00
$contacts[$pcontact['nurl']] = $pcontact['url'];
}
DBA::close($pcontacts);
if (empty($contacts)) {
return $serverdata;
}
foreach ($contacts as $contact) {
$probed = Contact::getByURL($contact);
if (!empty($probed) && in_array($probed['network'], Protocol::FEDERATED)) {
2019-10-04 08:33:16 +02:00
$serverdata['network'] = $probed['network'];
break;
}
}
$serverdata['registered-users'] = max($serverdata['registered-users'], count($contacts), 1);
2019-10-04 08:33:16 +02:00
return $serverdata;
}
2019-10-04 19:29:21 +02:00
/**
2019-10-05 06:22:16 +02:00
* Checks if the given server does have a '/poco' endpoint.
* This is used for the 'PortableContact' functionality,
* which is used by both Friendica and Hubzilla.
2019-10-04 19:29:21 +02:00
*
* @param string $url URL of the given server
* @param array $serverdata array with server data
*
* @return array server data
*/
private static function checkPoCo(string $url, array $serverdata)
2019-10-03 17:30:07 +02:00
{
2019-12-21 14:48:20 +01:00
$serverdata['poco'] = '';
$curlResult = DI::httpClient()->get($url . '/poco');
2019-10-03 17:30:07 +02:00
if (!$curlResult->isSuccess()) {
return $serverdata;
}
$data = json_decode($curlResult->getBody(), true);
if (empty($data)) {
return $serverdata;
}
if (!empty($data['totalResults'])) {
$registeredUsers = $serverdata['registered-users'] ?? 0;
$serverdata['registered-users'] = max($data['totalResults'], $registeredUsers, 1);
2019-12-21 14:48:20 +01:00
$serverdata['directory-type'] = self::DT_POCO;
2019-10-03 17:30:07 +02:00
$serverdata['poco'] = $url . '/poco';
2019-12-21 14:48:20 +01:00
}
return $serverdata;
}
/**
* Checks if the given server does have a Mastodon style directory endpoint.
*
* @param string $url URL of the given server
* @param array $serverdata array with server data
*
* @return array server data
*/
public static function checkMastodonDirectory(string $url, array $serverdata)
{
$curlResult = DI::httpClient()->get($url . '/api/v1/directory?limit=1');
2019-12-21 14:48:20 +01:00
if (!$curlResult->isSuccess()) {
return $serverdata;
}
$data = json_decode($curlResult->getBody(), true);
if (empty($data)) {
return $serverdata;
}
if (count($data) == 1) {
$serverdata['directory-type'] = self::DT_MASTODON;
2019-10-03 17:30:07 +02:00
}
return $serverdata;
}
/**
* Detects Peertube via their known endpoint
*
* @param string $url URL of the given server
* @param array $serverdata array with server data
*
* @return array server data
*/
private static function detectPeertube(string $url, array $serverdata)
{
$curlResult = DI::httpClient()->get($url . '/api/v1/config');
if (!$curlResult->isSuccess() || ($curlResult->getBody() == '')) {
return $serverdata;
}
$data = json_decode($curlResult->getBody(), true);
if (empty($data)) {
return $serverdata;
}
if (!empty($data['instance']) && !empty($data['serverVersion'])) {
$serverdata['platform'] = 'peertube';
$serverdata['version'] = $data['serverVersion'];
$serverdata['network'] = Protocol::ACTIVITYPUB;
if (!empty($data['instance']['name'])) {
$serverdata['site_name'] = $data['instance']['name'];
}
if (!empty($data['instance']['shortDescription'])) {
$serverdata['info'] = $data['instance']['shortDescription'];
}
if (!empty($data['signup'])) {
if (!empty($data['signup']['allowed'])) {
$serverdata['register_policy'] = Register::OPEN;
}
}
if (in_array($serverdata['detection-method'], [self::DETECT_HEADER, self::DETECT_BODY, self::DETECT_MANUAL])) {
$serverdata['detection-method'] = self::DETECT_V1_CONFIG;
}
}
return $serverdata;
}
2019-10-04 19:29:21 +02:00
/**
* Detects the version number of a given server when it was a NextCloud installation
*
* @param string $url URL of the given server
* @param array $serverdata array with server data
*
* @return array server data
*/
private static function detectNextcloud(string $url, array $serverdata)
2019-10-03 11:20:36 +02:00
{
$curlResult = DI::httpClient()->get($url . '/status.php');
2019-10-03 11:20:36 +02:00
if (!$curlResult->isSuccess() || ($curlResult->getBody() == '')) {
return $serverdata;
}
$data = json_decode($curlResult->getBody(), true);
if (empty($data)) {
return $serverdata;
}
if (!empty($data['version'])) {
$serverdata['platform'] = 'nextcloud';
$serverdata['version'] = $data['version'];
$serverdata['network'] = Protocol::ACTIVITYPUB;
if (in_array($serverdata['detection-method'], [self::DETECT_HEADER, self::DETECT_BODY, self::DETECT_MANUAL])) {
$serverdata['detection-method'] = self::DETECT_STATUS_PHP;
}
2019-10-03 11:20:36 +02:00
}
return $serverdata;
}
2019-10-04 19:29:21 +02:00
/**
* Detects data from a given server url if it was a mastodon alike system
*
* @param string $url URL of the given server
* @param array $serverdata array with server data
*
* @return array server data
*/
private static function detectMastodonAlikes(string $url, array $serverdata)
2019-10-03 11:20:36 +02:00
{
$curlResult = DI::httpClient()->get($url . '/api/v1/instance');
2019-10-03 11:20:36 +02:00
if (!$curlResult->isSuccess() || ($curlResult->getBody() == '')) {
return $serverdata;
}
$data = json_decode($curlResult->getBody(), true);
if (empty($data)) {
return $serverdata;
}
if (in_array($serverdata['detection-method'], [self::DETECT_HEADER, self::DETECT_BODY, self::DETECT_MANUAL])) {
$serverdata['detection-method'] = self::DETECT_MASTODON_API;
}
2019-10-03 15:02:48 +02:00
if (!empty($data['version'])) {
2019-10-03 11:20:36 +02:00
$serverdata['platform'] = 'mastodon';
$serverdata['version'] = $data['version'] ?? '';
2019-10-03 11:20:36 +02:00
$serverdata['network'] = Protocol::ACTIVITYPUB;
}
2019-10-03 15:02:48 +02:00
if (!empty($data['title'])) {
$serverdata['site_name'] = $data['title'];
}
2020-01-13 06:57:05 +01:00
if (!empty($data['title']) && empty($serverdata['platform']) && empty($serverdata['network'])) {
$serverdata['platform'] = 'mastodon';
$serverdata['network'] = Protocol::ACTIVITYPUB;
}
2019-10-03 15:02:48 +02:00
if (!empty($data['description'])) {
$serverdata['info'] = trim($data['description']);
}
2019-10-03 11:20:36 +02:00
if (!empty($data['stats']['user_count'])) {
$serverdata['registered-users'] = max($data['stats']['user_count'], 1);
2019-10-03 11:20:36 +02:00
}
if (!empty($serverdata['version']) && preg_match('/.*?\(compatible;\s(.*)\s(.*)\)/ism', $serverdata['version'], $matches)) {
2020-01-12 22:07:40 +01:00
$serverdata['platform'] = strtolower($matches[1]);
$serverdata['version'] = $matches[2];
}
2020-01-13 06:57:05 +01:00
if (!empty($serverdata['version']) && strstr(strtolower($serverdata['version']), 'pleroma')) {
$serverdata['platform'] = 'pleroma';
$serverdata['version'] = trim(str_ireplace('pleroma', '', $serverdata['version']));
}
if (!empty($serverdata['platform']) && strstr($serverdata['platform'], 'pleroma')) {
$serverdata['version'] = trim(str_ireplace('pleroma', '', $serverdata['platform']));
2019-10-03 11:20:36 +02:00
$serverdata['platform'] = 'pleroma';
}
return $serverdata;
}
2019-10-04 19:29:21 +02:00
/**
* Detects data from typical Hubzilla endpoints
*
* @param string $url URL of the given server
* @param array $serverdata array with server data
*
* @return array server data
*/
private static function detectHubzilla(string $url, array $serverdata)
2019-10-03 15:02:48 +02:00
{
$curlResult = DI::httpClient()->get($url . '/api/statusnet/config.json');
2019-10-03 15:02:48 +02:00
if (!$curlResult->isSuccess() || ($curlResult->getBody() == '')) {
return $serverdata;
}
$data = json_decode($curlResult->getBody(), true);
if (empty($data) || empty($data['site'])) {
2019-10-03 15:02:48 +02:00
return $serverdata;
}
if (!empty($data['site']['name'])) {
$serverdata['site_name'] = $data['site']['name'];
}
if (!empty($data['site']['platform'])) {
2020-01-12 22:07:40 +01:00
$serverdata['platform'] = strtolower($data['site']['platform']['PLATFORM_NAME']);
2019-10-03 15:02:48 +02:00
$serverdata['version'] = $data['site']['platform']['STD_VERSION'];
$serverdata['network'] = Protocol::ZOT;
}
2019-10-03 16:48:46 +02:00
if (!empty($data['site']['hubzilla'])) {
2020-01-12 22:07:40 +01:00
$serverdata['platform'] = strtolower($data['site']['hubzilla']['PLATFORM_NAME']);
2019-10-03 15:02:48 +02:00
$serverdata['version'] = $data['site']['hubzilla']['RED_VERSION'];
$serverdata['network'] = Protocol::ZOT;
}
2019-10-03 16:48:46 +02:00
if (!empty($data['site']['redmatrix'])) {
if (!empty($data['site']['redmatrix']['PLATFORM_NAME'])) {
2020-01-12 22:07:40 +01:00
$serverdata['platform'] = strtolower($data['site']['redmatrix']['PLATFORM_NAME']);
2019-10-03 16:48:46 +02:00
} elseif (!empty($data['site']['redmatrix']['RED_PLATFORM'])) {
2020-01-12 22:07:40 +01:00
$serverdata['platform'] = strtolower($data['site']['redmatrix']['RED_PLATFORM']);
2019-10-03 15:02:48 +02:00
}
$serverdata['version'] = $data['site']['redmatrix']['RED_VERSION'];
$serverdata['network'] = Protocol::ZOT;
}
$private = false;
$inviteonly = false;
$closed = false;
if (!empty($data['site']['closed'])) {
$closed = self::toBoolean($data['site']['closed']);
}
if (!empty($data['site']['private'])) {
$private = self::toBoolean($data['site']['private']);
}
if (!empty($data['site']['inviteonly'])) {
$inviteonly = self::toBoolean($data['site']['inviteonly']);
}
if (!$closed && !$private and $inviteonly) {
$serverdata['register_policy'] = Register::APPROVE;
2019-10-03 15:02:48 +02:00
} elseif (!$closed && !$private) {
$serverdata['register_policy'] = Register::OPEN;
2019-10-03 15:02:48 +02:00
} else {
$serverdata['register_policy'] = Register::CLOSED;
}
if (!empty($serverdata['network']) && in_array($serverdata['detection-method'],
[self::DETECT_HEADER, self::DETECT_BODY, self::DETECT_MANUAL])) {
$serverdata['detection-method'] = self::DETECT_CONFIG_JSON;
2019-10-03 15:02:48 +02:00
}
return $serverdata;
}
2019-10-04 19:29:21 +02:00
/**
* Converts input value to a boolean value
*
* @param string|integer $val
*
* @return boolean
*/
2019-10-03 15:02:48 +02:00
private static function toBoolean($val)
{
if (($val == 'true') || ($val == 1)) {
return true;
} elseif (($val == 'false') || ($val == 0)) {
return false;
}
return $val;
}
2019-10-04 19:29:21 +02:00
/**
* Detect if the URL belongs to a GNU Social server
*
* @param string $url URL of the given server
* @param array $serverdata array with server data
*
* @return array server data
*/
private static function detectGNUSocial(string $url, array $serverdata)
2019-10-03 11:20:36 +02:00
{
// Test for GNU Social
$curlResult = DI::httpClient()->get($url . '/api/gnusocial/version.json');
2019-10-03 11:20:36 +02:00
if ($curlResult->isSuccess() && ($curlResult->getBody() != '{"error":"not implemented"}') &&
($curlResult->getBody() != '') && (strlen($curlResult->getBody()) < 30)) {
$serverdata['platform'] = 'gnusocial';
2019-10-03 11:20:36 +02:00
// Remove junk that some GNU Social servers return
$serverdata['version'] = str_replace(chr(239) . chr(187) . chr(191), '', $curlResult->getBody());
2020-01-12 13:50:00 +01:00
$serverdata['version'] = str_replace(["\r", "\n", "\t"], '', $serverdata['version']);
2019-10-03 11:20:36 +02:00
$serverdata['version'] = trim($serverdata['version'], '"');
$serverdata['network'] = Protocol::OSTATUS;
if (in_array($serverdata['detection-method'], [self::DETECT_HEADER, self::DETECT_BODY, self::DETECT_MANUAL])) {
$serverdata['detection-method'] = self::DETECT_GNUSOCIAL;
}
return $serverdata;
2019-10-03 11:20:36 +02:00
}
// Test for Statusnet
$curlResult = DI::httpClient()->get($url . '/api/statusnet/version.json');
2019-10-03 11:20:36 +02:00
if ($curlResult->isSuccess() && ($curlResult->getBody() != '{"error":"not implemented"}') &&
($curlResult->getBody() != '') && (strlen($curlResult->getBody()) < 30)) {
2020-01-12 13:50:00 +01:00
2019-10-03 11:20:36 +02:00
// Remove junk that some GNU Social servers return
$serverdata['version'] = str_replace(chr(239).chr(187).chr(191), '', $curlResult->getBody());
2020-01-12 13:50:00 +01:00
$serverdata['version'] = str_replace(["\r", "\n", "\t"], '', $serverdata['version']);
2019-10-03 11:20:36 +02:00
$serverdata['version'] = trim($serverdata['version'], '"');
2020-01-12 13:50:00 +01:00
if (!empty($serverdata['version']) && strtolower(substr($serverdata['version'], 0, 7)) == 'pleroma') {
$serverdata['platform'] = 'pleroma';
$serverdata['version'] = trim(str_ireplace('pleroma', '', $serverdata['version']));
$serverdata['network'] = Protocol::ACTIVITYPUB;
} else {
$serverdata['platform'] = 'statusnet';
$serverdata['network'] = Protocol::OSTATUS;
}
if (in_array($serverdata['detection-method'], [self::DETECT_HEADER, self::DETECT_BODY, self::DETECT_MANUAL])) {
$serverdata['detection-method'] = self::DETECT_STATUSNET;
}
2019-10-03 11:20:36 +02:00
}
return $serverdata;
}
2019-10-04 19:29:21 +02:00
/**
* Detect if the URL belongs to a Friendica server
*
* @param string $url URL of the given server
* @param array $serverdata array with server data
*
* @return array server data
*/
private static function detectFriendica(string $url, array $serverdata)
2019-10-03 11:20:36 +02:00
{
$curlResult = DI::httpClient()->get($url . '/friendica/json');
2019-10-03 11:20:36 +02:00
if (!$curlResult->isSuccess()) {
$curlResult = DI::httpClient()->get($url . '/friendika/json');
$friendika = true;
$platform = 'Friendika';
} else {
$friendika = false;
$platform = 'Friendica';
2019-10-03 11:20:36 +02:00
}
if (!$curlResult->isSuccess()) {
return $serverdata;
}
$data = json_decode($curlResult->getBody(), true);
if (empty($data) || empty($data['version'])) {
return $serverdata;
}
if (in_array($serverdata['detection-method'], [self::DETECT_HEADER, self::DETECT_BODY, self::DETECT_MANUAL])) {
$serverdata['detection-method'] = $friendika ? self::DETECT_FRIENDIKA : self::DETECT_FRIENDICA;
}
2019-10-03 11:20:36 +02:00
$serverdata['network'] = Protocol::DFRN;
$serverdata['version'] = $data['version'];
if (!empty($data['no_scrape_url'])) {
$serverdata['noscrape'] = $data['no_scrape_url'];
}
if (!empty($data['site_name'])) {
$serverdata['site_name'] = $data['site_name'];
}
if (!empty($data['info'])) {
$serverdata['info'] = trim($data['info']);
}
$register_policy = ($data['register_policy'] ?? '') ?: 'REGISTER_CLOSED';
2019-10-03 11:20:36 +02:00
switch ($register_policy) {
case 'REGISTER_OPEN':
$serverdata['register_policy'] = Register::OPEN;
break;
case 'REGISTER_APPROVE':
$serverdata['register_policy'] = Register::APPROVE;
break;
case 'REGISTER_CLOSED':
case 'REGISTER_INVITATION':
$serverdata['register_policy'] = Register::CLOSED;
break;
default:
Logger::info('Register policy is invalid', ['policy' => $register_policy, 'server' => $url]);
$serverdata['register_policy'] = Register::CLOSED;
break;
}
$serverdata['platform'] = strtolower($data['platform'] ?? $platform);
2019-10-03 11:20:36 +02:00
return $serverdata;
}
2019-10-04 19:29:21 +02:00
/**
* Analyses the landing page of a given server for hints about type and system of that server
*
* @param object $curlResult result of curl execution
* @param array $serverdata array with server data
* @param string $url Server URL
2019-10-04 19:29:21 +02:00
*
* @return array server data
*/
private static function analyseRootBody($curlResult, array $serverdata, string $url)
2019-10-03 11:20:36 +02:00
{
2021-04-10 07:46:19 +02:00
if (empty($curlResult->getBody())) {
return $serverdata;
}
2019-10-03 11:20:36 +02:00
$doc = new DOMDocument();
@$doc->loadHTML($curlResult->getBody());
$xpath = new DOMXPath($doc);
$title = trim(XML::getFirstNodeValue($xpath, '//head/title/text()'));
if (!empty($title)) {
$serverdata['site_name'] = $title;
}
$list = $xpath->query('//meta[@name]');
foreach ($list as $node) {
$attr = [];
if ($node->attributes->length) {
foreach ($node->attributes as $attribute) {
2019-11-19 11:02:35 +01:00
$value = trim($attribute->value);
if (empty($value)) {
2019-10-03 11:20:36 +02:00
continue;
}
2019-11-19 11:02:35 +01:00
$attr[$attribute->name] = $value;
2019-10-03 11:20:36 +02:00
}
if (empty($attr['name']) || empty($attr['content'])) {
continue;
}
2019-10-03 11:20:36 +02:00
}
2019-10-03 23:39:48 +02:00
2019-10-03 11:20:36 +02:00
if ($attr['name'] == 'description') {
$serverdata['info'] = $attr['content'];
}
if (in_array($attr['name'], ['application-name', 'al:android:app_name', 'al:ios:app_name',
'twitter:app:name:googleplay', 'twitter:app:name:iphone', 'twitter:app:name:ipad'])) {
2020-01-12 22:07:40 +01:00
$serverdata['platform'] = strtolower($attr['content']);
2019-10-03 11:20:36 +02:00
if (in_array($attr['content'], ['Misskey', 'Write.as'])) {
$serverdata['network'] = Protocol::ACTIVITYPUB;
}
}
2020-01-13 06:57:05 +01:00
if (($attr['name'] == 'generator') && (empty($serverdata['platform']) || (substr(strtolower($attr['content']), 0, 9) == 'wordpress'))) {
2020-01-12 22:07:40 +01:00
$serverdata['platform'] = strtolower($attr['content']);
2019-10-03 11:20:36 +02:00
$version_part = explode(' ', $attr['content']);
2019-10-03 16:48:46 +02:00
if (count($version_part) == 2) {
2019-10-03 11:20:36 +02:00
if (in_array($version_part[0], ['WordPress'])) {
$serverdata['platform'] = 'wordpress';
2019-10-03 11:20:36 +02:00
$serverdata['version'] = $version_part[1];
// We still do need a reliable test if some AP plugin is activated
if (DBA::exists('apcontact', ['baseurl' => $url])) {
$serverdata['network'] = Protocol::ACTIVITYPUB;
} else {
$serverdata['network'] = Protocol::FEED;
}
if ($serverdata['detection-method'] == self::DETECT_MANUAL) {
$serverdata['detection-method'] = self::DETECT_BODY;
}
2019-10-03 11:20:36 +02:00
}
if (in_array($version_part[0], ['Friendika', 'Friendica'])) {
2020-01-12 22:07:40 +01:00
$serverdata['platform'] = strtolower($version_part[0]);
2019-10-03 11:20:36 +02:00
$serverdata['version'] = $version_part[1];
$serverdata['network'] = Protocol::DFRN;
}
}
}
}
$list = $xpath->query('//meta[@property]');
foreach ($list as $node) {
$attr = [];
if ($node->attributes->length) {
foreach ($node->attributes as $attribute) {
2019-11-19 11:02:35 +01:00
$value = trim($attribute->value);
if (empty($value)) {
2019-10-03 11:20:36 +02:00
continue;
}
2019-11-19 11:02:35 +01:00
$attr[$attribute->name] = $value;
2019-10-03 11:20:36 +02:00
}
if (empty($attr['property']) || empty($attr['content'])) {
continue;
}
2019-10-03 11:20:36 +02:00
}
if ($attr['property'] == 'og:site_name') {
$serverdata['site_name'] = $attr['content'];
}
if ($attr['property'] == 'og:description') {
$serverdata['info'] = $attr['content'];
}
if ($attr['property'] == 'og:platform') {
2020-01-12 22:07:40 +01:00
$serverdata['platform'] = strtolower($attr['content']);
if (in_array($attr['content'], ['PeerTube'])) {
$serverdata['network'] = Protocol::ACTIVITYPUB;
}
2019-10-03 11:20:36 +02:00
}
if ($attr['property'] == 'generator') {
2020-01-12 22:07:40 +01:00
$serverdata['platform'] = strtolower($attr['content']);
if (in_array($attr['content'], ['hubzilla'])) {
// We later check which compatible protocol modules are loaded.
$serverdata['network'] = Protocol::ZOT;
}
2019-10-03 11:20:36 +02:00
}
}
if (!empty($serverdata['network']) && ($serverdata['detection-method'] == self::DETECT_MANUAL)) {
$serverdata['detection-method'] = self::DETECT_BODY;
}
2019-10-03 11:20:36 +02:00
return $serverdata;
}
2019-10-04 19:29:21 +02:00
/**
* Analyses the header data of a given server for hints about type and system of that server
*
* @param object $curlResult result of curl execution
* @param array $serverdata array with server data
*
* @return array server data
*/
private static function analyseRootHeader($curlResult, array $serverdata)
2019-10-03 11:20:36 +02:00
{
if ($curlResult->getHeader('server') == 'Mastodon') {
$serverdata['platform'] = 'mastodon';
$serverdata['network'] = Protocol::ACTIVITYPUB;
2019-10-03 11:20:36 +02:00
} elseif ($curlResult->inHeader('x-diaspora-version')) {
$serverdata['platform'] = 'diaspora';
$serverdata['network'] = Protocol::DIASPORA;
$serverdata['version'] = $curlResult->getHeader('x-diaspora-version')[0] ?? '';
2019-10-03 11:20:36 +02:00
} elseif ($curlResult->inHeader('x-friendica-version')) {
$serverdata['platform'] = 'friendica';
$serverdata['network'] = Protocol::DFRN;
$serverdata['version'] = $curlResult->getHeader('x-friendica-version')[0] ?? '';
} else {
return $serverdata;
}
if ($serverdata['detection-method'] == self::DETECT_MANUAL) {
$serverdata['detection-method'] = self::DETECT_HEADER;
2019-10-03 11:20:36 +02:00
}
2019-10-03 11:20:36 +02:00
return $serverdata;
}
2019-12-21 07:39:22 +01:00
2020-01-12 13:50:00 +01:00
/**
* Test if the body contains valid content
*
* @param string $body
* @return boolean
*/
private static function invalidBody(string $body)
{
// Currently we only test for a HTML element.
// Possibly we enhance this in the future.
return !strpos($body, '>');
}
/**
* Update GServer entries
*/
public static function discover()
{
// Update the server list
self::discoverFederation();
$no_of_queries = 5;
$requery_days = intval(DI::config()->get('system', 'poco_requery_days'));
if ($requery_days == 0) {
$requery_days = 7;
}
$last_update = date('c', time() - (60 * 60 * 24 * $requery_days));
$gservers = DBA::select('gserver', ['id', 'url', 'nurl', 'network', 'poco', 'directory-type'],
["NOT `failed` AND `directory-type` != ? AND `last_poco_query` < ?", GServer::DT_NONE, $last_update],
['order' => ['RAND()']]);
while ($gserver = DBA::fetch($gservers)) {
2020-08-01 07:55:27 +02:00
Logger::info('Update peer list', ['server' => $gserver['url'], 'id' => $gserver['id']]);
Worker::add(PRIORITY_LOW, 'UpdateServerPeers', $gserver['url']);
Logger::info('Update directory', ['server' => $gserver['url'], 'id' => $gserver['id']]);
Worker::add(PRIORITY_LOW, 'UpdateServerDirectory', $gserver);
$fields = ['last_poco_query' => DateTimeFormat::utcNow()];
DBA::update('gserver', $fields, ['nurl' => $gserver['nurl']]);
if (--$no_of_queries == 0) {
break;
}
}
DBA::close($gservers);
}
/**
* Discover federated servers
*/
private static function discoverFederation()
{
$last = DI::config()->get('poco', 'last_federation_discovery');
if ($last) {
$next = $last + (24 * 60 * 60);
if ($next > time()) {
return;
}
}
// Discover federated servers
$protocols = ['activitypub', 'diaspora', 'dfrn', 'ostatus'];
foreach ($protocols as $protocol) {
$query = '{nodes(protocol:"' . $protocol . '"){host}}';
$curlResult = DI::httpClient()->fetch('https://the-federation.info/graphql?query=' . urlencode($query));
if (!empty($curlResult)) {
$data = json_decode($curlResult, true);
if (!empty($data['data']['nodes'])) {
foreach ($data['data']['nodes'] as $server) {
// Using "only_nodeinfo" since servers that are listed on that page should always have it.
2020-12-03 23:32:51 +01:00
self::add('https://' . $server['host'], true);
}
}
}
}
// Disvover Mastodon servers
$accesstoken = DI::config()->get('system', 'instances_social_key');
if (!empty($accesstoken)) {
$api = 'https://instances.social/api/1.0/instances/list?count=0';
$curlResult = DI::httpClient()->get($api, [HTTPClientOptions::HEADERS => ['Authorization' => ['Bearer ' . $accesstoken]]]);
if ($curlResult->isSuccess()) {
$servers = json_decode($curlResult->getBody(), true);
foreach ($servers['instances'] as $server) {
$url = (is_null($server['https_score']) ? 'http' : 'https') . '://' . $server['name'];
2020-12-03 23:32:51 +01:00
self::add($url);
}
}
}
DI::config()->set('poco', 'last_federation_discovery', time());
}
2021-01-09 20:18:22 +01:00
/**
* Set the protocol for the given server
*
* @param int $gsid Server id
* @param int $protocol Protocol id
* @return void
* @throws Exception
2021-01-09 20:18:22 +01:00
*/
public static function setProtocol(int $gsid, int $protocol)
2021-01-09 20:18:22 +01:00
{
if (empty($gsid)) {
return;
}
$gserver = DBA::selectFirst('gserver', ['protocol', 'url'], ['id' => $gsid]);
if (!DBA::isResult($gserver)) {
return;
}
$old = $gserver['protocol'];
if (!is_null($old)) {
/*
The priority for the protocols is:
1. ActivityPub
2. DFRN via Diaspora
3. Legacy DFRN
4. Diaspora
5. OStatus
*/
// We don't need to change it when nothing is to be changed
if ($old == $protocol) {
return;
}
// We don't want to mark a server as OStatus when it had been marked with any other protocol before
if ($protocol == Post\DeliveryData::OSTATUS) {
return;
}
// If the server is marked as ActivityPub then we won't change it to anything different
if ($old == Post\DeliveryData::ACTIVITYPUB) {
return;
}
// Don't change it to anything lower than DFRN if the new one wasn't ActivityPub
if (($old == Post\DeliveryData::DFRN) && ($protocol != Post\DeliveryData::ACTIVITYPUB)) {
return;
}
// Don't change it to Diaspora when it is a legacy DFRN server
if (($old == Post\DeliveryData::LEGACY_DFRN) && ($protocol == Post\DeliveryData::DIASPORA)) {
return;
}
}
Logger::info('Protocol for server', ['protocol' => $protocol, 'old' => $old, 'id' => $gsid, 'url' => $gserver['url'], 'callstack' => System::callstack(20)]);
2021-01-09 20:18:22 +01:00
DBA::update('gserver', ['protocol' => $protocol], ['id' => $gsid]);
}
/**
* Fetch the protocol of the given server
*
* @param int $gsid Server id
* @return int
* @throws Exception
*/
public static function getProtocol(int $gsid)
{
if (empty($gsid)) {
return null;
}
$gserver = DBA::selectFirst('gserver', ['protocol'], ['id' => $gsid]);
if (DBA::isResult($gserver)) {
return $gserver['protocol'];
}
return null;
}
2019-10-03 11:20:36 +02:00
}