friendica/src/Network/Probe.php

2206 lines
61 KiB
PHP
Raw Normal View History

2016-07-03 20:27:16 +00:00
<?php
/**
2022-01-02 07:27:47 +00:00
* @copyright Copyright (C) 2010-2022, the Friendica project
*
* @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/>.
*
*/
2017-05-07 18:44:30 +00:00
namespace Friendica\Network;
2016-07-03 20:27:16 +00:00
use DOMDocument;
use DomXPath;
use Exception;
2020-06-06 18:54:04 +00:00
use Friendica\Core\Hook;
2018-10-29 21:20:46 +00:00
use Friendica\Core\Logger;
use Friendica\Core\Protocol;
use Friendica\Core\System;
use Friendica\Database\DBA;
use Friendica\DI;
use Friendica\Model\Contact;
2020-05-22 10:10:24 +00:00
use Friendica\Model\GServer;
2017-12-07 13:57:35 +00:00
use Friendica\Model\Profile;
2020-06-28 15:43:58 +00:00
use Friendica\Model\User;
use Friendica\Network\HTTPClient\Client\HttpClientAccept;
use Friendica\Network\HTTPClient\Client\HttpClientOptions;
2019-10-24 22:34:46 +00:00
use Friendica\Protocol\ActivityNamespace;
use Friendica\Protocol\ActivityPub;
use Friendica\Protocol\Email;
2017-12-13 07:02:52 +00:00
use Friendica\Protocol\Feed;
use Friendica\Util\Crypto;
2020-08-05 06:50:51 +00:00
use Friendica\Util\DateTimeFormat;
use Friendica\Util\Network;
use Friendica\Util\Strings;
use Friendica\Util\XML;
use GuzzleHttp\Psr7\Uri;
2017-05-11 15:53:04 +00:00
2016-07-08 20:55:39 +00:00
/**
2020-01-19 06:05:23 +00:00
* This class contain functions for probing URL
2016-07-08 20:55:39 +00:00
*/
class Probe
{
const HOST_META = '/.well-known/host-meta';
2020-06-14 13:37:28 +00:00
const WEBFINGER = '/.well-known/webfinger?resource={uri}';
/**
* @var string Base URL
*/
private static $baseurl;
/**
* @var boolean Whether a timeout has occured
*/
private static $isTimeout;
/**
* Checks if the provided network can be probed
*
* @param string $network
*
* @return boolean
*/
public static function isProbable(string $network): bool
{
return (in_array($network, array_merge(Protocol::FEDERATED, [Protocol::ZOT, Protocol::PHANTOM])));
}
2020-02-22 12:29:33 +00:00
/**
* Remove stuff from an URI that doesn't belong there
*
* @param string $rawUri
2020-02-22 12:29:33 +00:00
* @return string Cleaned URI
*/
public static function cleanURI(string $rawUri): string
2020-02-22 12:29:33 +00:00
{
// At first remove leading and trailing junk
2022-09-24 15:57:28 +00:00
$rawUri = trim($rawUri, "@#?: \t\n\r\0\x0B");
2020-02-22 12:29:33 +00:00
$rawUri = Network::convertToIdn($rawUri);
$uri = new Uri($rawUri);
if (!$uri->getScheme()) {
return $uri->__toString();
2020-02-22 12:29:33 +00:00
}
// Remove the URL fragment, since these shouldn't be part of any profile URL
$uri = $uri->withFragment('');
2020-02-22 12:29:33 +00:00
return $uri->__toString();
2020-02-22 12:29:33 +00:00
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Rearrange the array so that it always has the same order
2016-07-07 21:04:30 +00:00
*
* @param array $data Unordered data
* @return array Ordered data
*/
private static function rearrangeData(array $data): array
{
$fields = ['name', 'nick', 'guid', 'url', 'addr', 'alias', 'photo', 'header',
'account-type', 'community', 'keywords', 'location', 'about', 'xmpp', 'matrix',
'hide', 'batch', 'notify', 'poll', 'request', 'confirm', 'subscribe', 'poco',
'following', 'followers', 'inbox', 'outbox', 'sharedinbox',
'priority', 'network', 'pubkey', 'manually-approve', 'baseurl', 'gsid'];
2016-07-03 20:27:16 +00:00
$numeric_fields = ['gsid', 'hide', 'account-type', 'manually-approve'];
$newdata = [];
2017-05-08 15:06:57 +00:00
foreach ($fields as $field) {
if (isset($data[$field])) {
if (in_array($field, $numeric_fields)) {
$newdata[$field] = (int)$data[$field];
} else {
$newdata[$field] = $data[$field];
}
} elseif (!in_array($field, $numeric_fields)) {
$newdata[$field] = '';
2020-05-22 10:10:24 +00:00
} else {
$newdata[$field] = null;
2017-05-08 15:06:57 +00:00
}
}
2016-07-03 20:27:16 +00:00
// We don't use the "priority" field anymore and replace it with a dummy.
$newdata['priority'] = 0;
2016-07-03 20:27:16 +00:00
return $newdata;
}
/**
2020-01-19 06:05:23 +00:00
* Check if the hostname belongs to the own server
*
* @param string $host The hostname that is to be checked
* @return bool Does the testes hostname belongs to the own server?
*/
private static function ownHost(string $host): bool
{
$own_host = DI::baseUrl()->getHostname();
$parts = parse_url($host);
if (!isset($parts['scheme'])) {
$parts = parse_url('http://' . $host);
}
if (!isset($parts['host'])) {
return false;
}
return $parts['host'] == $own_host;
}
2016-07-03 20:27:16 +00:00
/**
* Probes for webfinger path via "host-meta"
2017-05-08 15:06:57 +00:00
*
2018-02-02 13:53:40 +00:00
* We have to check if the servers in the future still will offer this.
* It seems as if it was dropped from the standard.
*
* @param string $host The host part of an url
*
* @return array with template and type of the webfinger template for JSON or XML
2019-01-06 21:06:53 +00:00
* @throws HTTPException\InternalServerErrorException
2016-07-03 20:27:16 +00:00
*/
private static function hostMeta(string $host): array
{
2017-03-25 10:07:45 +00:00
// Reset the static variable
self::$baseurl = '';
// Handles the case when the hostname contains the scheme
if (!parse_url($host, PHP_URL_SCHEME)) {
$ssl_url = 'https://' . $host . self::HOST_META;
$url = 'http://' . $host . self::HOST_META;
} else {
$ssl_url = $host . self::HOST_META;
$url = '';
}
2016-07-03 20:27:16 +00:00
$xrd_timeout = DI::config()->get('system', 'xrd_timeout', 20);
2016-07-03 20:27:16 +00:00
Logger::info('Probing', ['host' => $host, 'ssl_url' => $ssl_url, 'url' => $url, 'callstack' => System::callstack(20)]);
$xrd = null;
2017-05-23 20:19:39 +00:00
$curlResult = DI::httpClient()->get($ssl_url, HttpClientAccept::XRD_XML, [HttpClientOptions::TIMEOUT => $xrd_timeout]);
$ssl_connection_error = ($curlResult->getErrorNumber() == CURLE_COULDNT_CONNECT) || ($curlResult->getReturnCode() == 0);
if ($curlResult->isSuccess()) {
$xml = $curlResult->getBody();
2020-04-27 14:35:50 +00:00
$xrd = XML::parseString($xml, true);
if (!empty($url)) {
$host_url = 'https://' . $host;
} else {
$host_url = $host;
}
} elseif ($curlResult->isTimeout()) {
2020-06-28 15:43:58 +00:00
Logger::info('Probing timeout', ['url' => $ssl_url]);
self::$isTimeout = true;
return [];
2017-02-01 21:35:01 +00:00
}
2016-07-03 20:27:16 +00:00
2020-03-08 19:07:36 +00:00
if (!is_object($xrd) && !empty($url)) {
$curlResult = DI::httpClient()->get($url, HttpClientAccept::XRD_XML, [HttpClientOptions::TIMEOUT => $xrd_timeout]);
$connection_error = ($curlResult->getErrorNumber() == CURLE_COULDNT_CONNECT) || ($curlResult->getReturnCode() == 0);
if ($curlResult->isTimeout()) {
2020-06-28 15:43:58 +00:00
Logger::info('Probing timeout', ['url' => $url]);
self::$isTimeout = true;
return [];
} elseif ($connection_error && $ssl_connection_error) {
self::$isTimeout = true;
return [];
}
$xml = $curlResult->getBody();
2020-04-27 14:35:50 +00:00
$xrd = XML::parseString($xml, true);
$host_url = 'http://'.$host;
2016-07-03 20:27:16 +00:00
}
2017-05-08 15:06:57 +00:00
if (!is_object($xrd)) {
2020-06-28 15:43:58 +00:00
Logger::info('No xrd object found', ['host' => $host]);
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$links = XML::elementToArray($xrd);
if (!isset($links['xrd']['link'])) {
2020-06-28 15:43:58 +00:00
Logger::info('No xrd data found', ['host' => $host]);
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
2020-06-14 13:37:28 +00:00
$lrdd = [];
2016-07-09 18:09:09 +00:00
foreach ($links['xrd']['link'] as $value => $link) {
if (!empty($link['@attributes'])) {
$attributes = $link['@attributes'];
} elseif ($value == '@attributes') {
2016-07-03 20:27:16 +00:00
$attributes = $link;
2017-05-08 15:06:57 +00:00
} else {
2016-07-03 20:27:16 +00:00
continue;
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
if (!empty($attributes['rel']) && $attributes['rel'] == 'lrdd' && !empty($attributes['template'])) {
$type = (empty($attributes['type']) ? '' : $attributes['type']);
$lrdd[$type] = $attributes['template'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
}
if (Network::isUrlBlocked($host_url)) {
Logger::info('Domain is blocked', ['url' => $host]);
return [];
}
2018-12-04 07:12:55 +00:00
self::$baseurl = $host_url;
2020-06-28 15:43:58 +00:00
Logger::info('Probing successful', ['host' => $host]);
2017-05-28 05:38:12 +00:00
return $lrdd;
2016-07-03 20:27:16 +00:00
}
2016-07-09 18:09:09 +00:00
/**
2020-01-19 06:05:23 +00:00
* Check an URI for LRDD data
2016-07-09 18:09:09 +00:00
*
* @param string $uri Address that should be probed
2016-07-09 18:09:09 +00:00
* @return array uri data
2019-01-06 21:06:53 +00:00
* @throws HTTPException\InternalServerErrorException
2016-07-09 18:09:09 +00:00
*/
public static function lrdd(string $uri): array
{
2020-06-14 13:37:28 +00:00
$data = self::getWebfingerArray($uri);
if (empty($data)) {
return [];
}
2020-06-14 13:37:28 +00:00
$webfinger = $data['webfinger'];
2016-07-09 18:09:09 +00:00
if (empty($webfinger['links'])) {
2020-06-28 15:43:58 +00:00
Logger::info('No webfinger links found', ['uri' => $uri]);
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-09 18:09:09 +00:00
$data = [];
2016-07-09 18:09:09 +00:00
foreach ($webfinger['links'] as $link) {
$data[] = ['@attributes' => $link];
2017-05-08 15:06:57 +00:00
}
2016-07-09 18:09:09 +00:00
if (!empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
foreach ($webfinger['aliases'] as $alias) {
$data[] = [
'@attributes' => [
'rel' => 'alias',
'href' => $alias,
]
];
2017-05-08 15:06:57 +00:00
}
}
2016-07-09 18:09:09 +00:00
return $data;
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Fetch information (protocol endpoints and user information) about a given uri
2016-07-07 21:04:30 +00:00
*
* @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?
2016-07-07 21:04:30 +00:00
*
* @return array uri data
2019-01-06 21:06:53 +00:00
* @throws HTTPException\InternalServerErrorException
* @throws \ImagickException
2016-07-07 21:04:30 +00:00
*/
public static function uri(string $uri, string $network = '', int $uid = -1): array
{
// Local profiles aren't probed via network
2021-07-19 04:15:57 +00:00
if (empty($network) && Contact::isLocal($uri)) {
$data = self::localProbe($uri);
if (!empty($data)) {
return $data;
}
}
2017-09-23 22:53:44 +00:00
if ($uid == -1) {
$uid = DI::userSession()->getLocalUserId();
2017-05-08 15:06:57 +00:00
}
2020-06-25 20:29:21 +00:00
if (empty($network) || ($network == Protocol::ACTIVITYPUB)) {
$ap_profile = ActivityPub::probeProfile($uri);
2020-06-25 20:29:21 +00:00
} else {
$ap_profile = [];
}
self::$isTimeout = false;
2020-06-04 21:55:14 +00:00
if ($network != Protocol::ACTIVITYPUB) {
2020-06-25 20:29:21 +00:00
$data = self::detect($uri, $network, $uid, $ap_profile);
2020-06-14 13:37:28 +00:00
if (!is_array($data)) {
$data = [];
}
if (empty($data) || (!empty($ap_profile) && empty($network) && (($data['network'] ?? '') != Protocol::DFRN))) {
$data = $ap_profile;
2019-07-04 19:31:42 +00:00
} elseif (!empty($ap_profile)) {
$ap_profile['batch'] = '';
2019-07-04 19:31:42 +00:00
$data = array_merge($ap_profile, $data);
}
2020-06-25 20:29:21 +00:00
} else {
$data = $ap_profile;
}
2016-07-03 20:27:16 +00:00
2018-11-10 13:24:53 +00:00
if (!isset($data['url'])) {
$data['url'] = $uri;
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
2020-05-24 20:40:00 +00:00
if (empty($data['photo'])) {
$data['photo'] = DI::baseUrl() . Contact::DEFAULT_AVATAR_PHOTO;
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
2018-11-10 13:24:53 +00:00
if (empty($data['name'])) {
if (!empty($data['nick'])) {
$data['name'] = $data['nick'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
if (empty($data['name'])) {
2018-11-10 13:24:53 +00:00
$data['name'] = $data['url'];
2017-05-08 15:06:57 +00:00
}
2016-07-10 07:27:40 +00:00
}
2018-11-10 13:24:53 +00:00
if (empty($data['nick'])) {
$data['nick'] = strtolower($data['name']);
2016-07-03 20:27:16 +00:00
2017-05-08 15:06:57 +00:00
if (strpos($data['nick'], ' ')) {
$data['nick'] = trim(substr($data['nick'], 0, strpos($data['nick'], ' ')));
2017-05-08 15:06:57 +00:00
}
}
2020-05-22 10:10:24 +00:00
if (!empty($data['baseurl']) && empty($data['gsid'])) {
$data['gsid'] = GServer::getID($data['baseurl']);
}
2018-11-10 13:24:53 +00:00
if (empty($data['network'])) {
$data['network'] = Protocol::PHANTOM;
}
2016-07-03 20:27:16 +00:00
// Ensure that local connections always are DFRN
if (($network == '') && ($data['network'] != Protocol::PHANTOM) && (self::ownHost($data['baseurl'] ?? '') || self::ownHost($data['url']))) {
$data['network'] = Protocol::DFRN;
}
if (!isset($data['hide']) && in_array($data['network'], Protocol::FEDERATED)) {
$data['hide'] = self::getHideStatus($data['url']);
}
2020-08-06 18:53:45 +00:00
return self::rearrangeData($data);
2016-07-03 20:27:16 +00:00
}
/**
* Fetches the "hide" status from the profile
*
* @param string $url URL of the profile
* @return boolean "hide" status
*/
private static function getHideStatus(string $url): bool
{
$curlResult = DI::httpClient()->get($url, HttpClientAccept::HTML, [HttpClientOptions::CONTENT_LENGTH => 1000000]);
if (!$curlResult->isSuccess()) {
return false;
}
// If it isn't a HTML file then exit
if (($curlResult->getContentType() != '') && !strstr(strtolower($curlResult->getContentType()), 'html')) {
return false;
}
$body = $curlResult->getBody();
2021-04-10 05:46:19 +00:00
if (empty($body)) {
return false;
}
$doc = new DOMDocument();
@$doc->loadHTML($body);
$xpath = new DOMXPath($doc);
$list = $xpath->query('//meta[@name]');
foreach ($list as $node) {
$meta_tag = [];
if ($node->attributes->length) {
foreach ($node->attributes as $attribute) {
$meta_tag[$attribute->name] = $attribute->value;
}
}
if (empty($meta_tag['content'])) {
continue;
}
$content = strtolower(trim($meta_tag['content']));
switch (strtolower(trim($meta_tag['name']))) {
case 'dfrn-global-visibility':
if ($content == 'false') {
return true;
}
break;
case 'robots':
if (strpos($content, 'noindex') !== false) {
return true;
}
break;
}
}
return false;
}
/**
* Fetch the "subscribe" and add it to the result
*
* @param array $result Result array
* @param array $webfinger Webfinger data
*
* @return array result Altered/unaltered result array
*/
private static function getSubscribeLink(array $result, array $webfinger): array
{
if (empty($webfinger['links'])) {
return $result;
}
foreach ($webfinger['links'] as $link) {
if (!empty($link['template']) && ($link['rel'] === ActivityNamespace::OSTATUSSUB)) {
$result['subscribe'] = $link['template'];
}
}
return $result;
}
2020-06-14 13:37:28 +00:00
/**
* Get webfinger data from a given URI
*
* @param string $uri URI
*
* @return array Webfinger data
2021-12-02 13:02:26 +00:00
* @throws HTTPException\InternalServerErrorException
2020-06-14 13:37:28 +00:00
*/
2021-12-02 13:02:26 +00:00
private static function getWebfingerArray(string $uri): array
2020-06-14 13:37:28 +00:00
{
$parts = parse_url($uri);
if (!empty($parts['scheme']) && !empty($parts['host'])) {
$host = $parts['host'];
if (!empty($parts['port'])) {
2021-12-02 13:02:26 +00:00
$host .= ':' . $parts['port'];
2020-06-14 13:37:28 +00:00
}
$baseurl = $parts['scheme'] . '://' . $host;
$nick = '';
$addr = '';
2021-12-02 13:02:26 +00:00
$path_parts = explode('/', trim($parts['path'] ?? '', '/'));
2020-06-14 13:37:28 +00:00
if (!empty($path_parts)) {
$nick = ltrim(end($path_parts), '@');
$addr = $nick . '@' . $host;
2020-06-14 13:37:28 +00:00
}
$webfinger = self::getWebfinger($parts['scheme'] . '://' . $host . self::WEBFINGER, HttpClientAccept::JRD_JSON, $uri, $addr);
2020-06-14 13:37:28 +00:00
if (empty($webfinger)) {
$lrdd = self::hostMeta($host);
}
if (empty($webfinger) && empty($lrdd)) {
while (empty($lrdd) && empty($webfinger) && (sizeof($path_parts) > 1)) {
2021-12-02 13:02:26 +00:00
$host .= '/' . array_shift($path_parts);
2020-06-14 13:37:28 +00:00
$baseurl = $parts['scheme'] . '://' . $host;
if (!empty($nick)) {
2021-12-02 13:02:26 +00:00
$addr = $nick . '@' . $host;
2020-06-14 13:37:28 +00:00
}
$webfinger = self::getWebfinger($parts['scheme'] . '://' . $host . self::WEBFINGER, HttpClientAccept::JRD_JSON, $uri, $addr);
2020-06-14 13:37:28 +00:00
if (empty($webfinger)) {
$lrdd = self::hostMeta($host);
}
}
if (empty($lrdd) && empty($webfinger)) {
return [];
}
}
} elseif (strstr($uri, '@')) {
// Remove "acct:" from the URI
$uri = str_replace('acct:', '', $uri);
$host = substr($uri, strpos($uri, '@') + 1);
$nick = substr($uri, 0, strpos($uri, '@'));
$addr = $uri;
$webfinger = self::getWebfinger('https://' . $host . self::WEBFINGER, HttpClientAccept::JRD_JSON, $uri, $addr);
if (self::$isTimeout) {
2020-06-14 13:37:28 +00:00
return [];
}
if (empty($webfinger)) {
$webfinger = self::getWebfinger('http://' . $host . self::WEBFINGER, HttpClientAccept::JRD_JSON, $uri, $addr);
if (self::$isTimeout) {
2020-06-14 13:37:28 +00:00
return [];
}
} else {
$baseurl = 'https://' . $host;
}
if (empty($webfinger)) {
$lrdd = self::hostMeta($host);
if (self::$isTimeout) {
2020-06-14 13:37:28 +00:00
return [];
}
$baseurl = self::$baseurl;
} else {
$baseurl = 'http://' . $host;
}
} else {
Logger::info('URI was not detectable', ['uri' => $uri]);
return [];
}
if (empty($webfinger)) {
foreach ($lrdd as $type => $template) {
if ($webfinger) {
continue;
}
$webfinger = self::getWebfinger($template, $type, $uri, $addr);
}
}
if (empty($webfinger)) {
return [];
}
if ($webfinger['detected'] == $addr) {
$webfinger['nick'] = $nick;
$webfinger['addr'] = $addr;
}
$webfinger['baseurl'] = $baseurl;
return $webfinger;
}
/**
* Perform network request for webfinger data
*
* @param string $template
* @param string $type
* @param string $uri
* @param string $addr
*
2020-06-14 13:37:28 +00:00
* @return array webfinger results
*/
private static function getWebfinger(string $template, string $type, string $uri, string $addr): array
2020-06-14 13:37:28 +00:00
{
if (Network::isUrlBlocked($template)) {
Logger::info('Domain is blocked', ['url' => $template]);
return [];
}
2020-06-14 13:37:28 +00:00
// First try the address because this is the primary purpose of webfinger
if (!empty($addr)) {
$detected = $addr;
$path = str_replace('{uri}', urlencode('acct:' . $addr), $template);
2020-06-14 13:37:28 +00:00
$webfinger = self::webfinger($path, $type);
if (self::$isTimeout) {
2020-06-14 13:37:28 +00:00
return [];
}
}
// Then try the URI
if (empty($webfinger) && $uri != $addr) {
$detected = $uri;
$path = str_replace('{uri}', urlencode($uri), $template);
$webfinger = self::webfinger($path, $type);
if (self::$isTimeout) {
2020-06-14 13:37:28 +00:00
return [];
}
}
if (empty($webfinger)) {
return [];
}
return ['webfinger' => $webfinger, 'detected' => $detected];
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Fetch information (protocol endpoints and user information) about a given uri
2016-07-10 16:44:48 +00:00
*
* This function is only called by the "uri" function that adds caching and rearranging of data.
2016-07-07 21:04:30 +00:00
*
2020-06-25 20:29:21 +00:00
* @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 array $ap_profile Previously probed AP profile
* @return array URI data
2019-01-06 21:06:53 +00:00
* @throws HTTPException\InternalServerErrorException
2016-07-07 21:04:30 +00:00
*/
private static function detect(string $uri, string $network, int $uid, array $ap_profile): array
{
2020-06-06 18:54:04 +00:00
$hookData = [
'uri' => $uri,
'network' => $network,
'uid' => $uid,
'result' => null,
2020-06-06 18:54:04 +00:00
];
Hook::callAll('probe_detect', $hookData);
if (isset($hookData['result'])) {
return is_array($hookData['result']) ? $hookData['result'] : [];
2020-06-14 13:37:28 +00:00
}
2020-06-14 13:37:28 +00:00
$parts = parse_url($uri);
if (empty($parts['scheme']) && empty($parts['host']) && !strstr($parts['path'], '@')) {
2020-06-14 13:37:28 +00:00
Logger::info('URI was not detectable', ['uri' => $uri]);
return [];
2016-07-03 20:27:16 +00:00
}
// If the URI starts with "mailto:" then jump directly to the mail detection
if (strpos($uri, 'mailto:') !== false) {
$uri = str_replace('mailto:', '', $uri);
return self::mail($uri, $uid);
}
if ($network == Protocol::MAIL) {
return self::mail($uri, $uid);
}
2020-06-14 13:37:28 +00:00
Logger::info('Probing start', ['uri' => $uri]);
2020-06-10 19:19:10 +00:00
if (!empty($ap_profile['addr']) && ($ap_profile['addr'] != $uri)) {
$data = self::getWebfingerArray($ap_profile['addr']);
}
if (empty($data)) {
$data = self::getWebfingerArray($uri);
}
2020-06-14 13:37:28 +00:00
if (empty($data)) {
if (!empty($parts['scheme'])) {
return self::feed($uri);
} elseif (!empty($uid)) {
return self::mail($uri, $uid);
} else {
return [];
2020-06-10 19:19:10 +00:00
}
2016-07-03 20:27:16 +00:00
}
2020-06-14 13:37:28 +00:00
$webfinger = $data['webfinger'];
$nick = $data['nick'] ?? '';
$addr = $data['addr'] ?? '';
$baseurl = $data['baseurl'] ?? '';
2016-07-03 20:27:16 +00:00
$result = [];
2016-07-03 20:27:16 +00:00
if (in_array($network, ['', Protocol::DFRN])) {
2016-07-03 20:27:16 +00:00
$result = self::dfrn($webfinger);
2017-05-08 15:06:57 +00:00
}
if ((!$result && ($network == '')) || ($network == Protocol::DIASPORA)) {
2019-12-23 14:26:06 +00:00
$result = self::diaspora($webfinger);
2017-05-08 15:06:57 +00:00
}
if ((!$result && ($network == '')) || ($network == Protocol::OSTATUS)) {
2019-12-23 14:30:48 +00:00
$result = self::ostatus($webfinger);
2017-05-08 15:06:57 +00:00
}
2019-12-23 15:07:49 +00:00
if (in_array($network, ['', Protocol::ZOT])) {
2020-06-14 13:37:28 +00:00
$result = self::zot($webfinger, $result, $baseurl);
2019-12-23 15:07:49 +00:00
}
if ((!$result && ($network == '')) || ($network == Protocol::PUMPIO)) {
2019-12-23 14:26:06 +00:00
$result = self::pumpio($webfinger, $addr);
2017-05-08 15:06:57 +00:00
}
2020-06-25 20:29:21 +00:00
if (empty($result['network']) && empty($ap_profile['network']) || ($network == Protocol::FEED)) {
2019-12-23 14:30:48 +00:00
$result = self::feed($uri);
2017-05-08 15:06:57 +00:00
} else {
2016-07-03 20:27:16 +00:00
// We overwrite the detected nick with our try if the previois routines hadn't detected it.
// Additionally it is overwritten when the nickname doesn't make sense (contains spaces).
if ((empty($result['nick']) || (strstr($result['nick'], ' '))) && ($nick != '')) {
$result['nick'] = $nick;
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
if (empty($result['addr']) && ($addr != '')) {
$result['addr'] = $addr;
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
}
$result = self::getSubscribeLink($result, $webfinger);
if (empty($result['network'])) {
$result['network'] = Protocol::PHANTOM;
}
2020-06-14 13:37:28 +00:00
if (empty($result['baseurl']) && !empty($baseurl)) {
$result['baseurl'] = $baseurl;
}
if (empty($result['url'])) {
$result['url'] = $uri;
}
Logger::info('Probing done', ['uri' => $uri, 'network' => $result['network']]);
2016-07-03 22:49:38 +00:00
2016-07-03 20:27:16 +00:00
return $result;
}
2019-12-22 17:20:11 +00:00
/**
* Check for Zot contact
*
* @param array $webfinger Webfinger data
* @param array $data previously probed data
* @param string $baseUrl Base URL
*
2019-12-22 17:20:11 +00:00
* @return array Zot data
* @throws HTTPException\InternalServerErrorException
*/
private static function zot(array $webfinger, array $data, string $baseurl): array
2019-12-22 08:13:12 +00:00
{
if (!empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
foreach ($webfinger['aliases'] as $alias) {
2019-12-23 14:26:06 +00:00
if (substr($alias, 0, 5) == 'acct:') {
$data['addr'] = substr($alias, 5);
2019-12-23 14:26:06 +00:00
}
}
}
if (!empty($webfinger['subject']) && (substr($webfinger['subject'], 0, 5) == 'acct:')) {
$data['addr'] = substr($webfinger['subject'], 5);
2019-12-23 14:26:06 +00:00
}
2019-12-22 08:13:12 +00:00
$zot_url = '';
foreach ($webfinger['links'] as $link) {
2019-12-22 17:20:11 +00:00
if (($link['rel'] == 'http://purl.org/zot/protocol') && !empty($link['href'])) {
2019-12-22 08:13:12 +00:00
$zot_url = $link['href'];
}
}
2020-06-14 13:37:28 +00:00
if (empty($zot_url) && !empty($data['addr']) && !empty($baseurl)) {
$condition = ['nurl' => Strings::normaliseLink($baseurl), 'platform' => ['hubzilla']];
2019-12-23 14:26:06 +00:00
if (!DBA::exists('gserver', $condition)) {
2019-12-23 15:07:49 +00:00
return $data;
2019-12-23 14:26:06 +00:00
}
2020-06-14 13:37:28 +00:00
$zot_url = $baseurl . '/.well-known/zot-info?address=' . $data['addr'];
2019-12-22 08:13:12 +00:00
}
2019-12-24 10:19:14 +00:00
if (empty($zot_url)) {
return $data;
}
$data = self::pollZot($zot_url, $data);
if (!empty($data['url']) && !empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
foreach ($webfinger['aliases'] as $alias) {
if (!strstr($alias, '@') && Strings::normaliseLink($alias) != Strings::normaliseLink($data['url'])) {
$data['alias'] = $alias;
}
}
2019-12-22 08:13:12 +00:00
}
return $data;
}
2022-06-23 08:18:38 +00:00
public static function pollZot(string $url, array $data): array
2019-12-22 08:13:12 +00:00
{
$curlResult = DI::httpClient()->get($url, HttpClientAccept::JSON);
2019-12-22 08:13:12 +00:00
if ($curlResult->isTimeout()) {
return $data;
}
$content = $curlResult->getBody();
if (!$content) {
return $data;
}
$json = json_decode($content, true);
if (!is_array($json)) {
return $data;
}
2019-12-23 21:10:54 +00:00
2019-12-23 14:26:06 +00:00
if (empty($data['network'])) {
if (!empty($json['protocols']) && in_array('zot', $json['protocols'])) {
$data['network'] = Protocol::ZOT;
} elseif (!isset($json['protocols'])) {
$data['network'] = Protocol::ZOT;
}
2019-12-22 08:13:12 +00:00
}
2019-12-23 14:26:06 +00:00
if (!empty($json['guid']) && empty($data['guid'])) {
2019-12-22 08:13:12 +00:00
$data['guid'] = $json['guid'];
}
2019-12-23 14:26:06 +00:00
if (!empty($json['key']) && empty($data['pubkey'])) {
2019-12-22 08:13:12 +00:00
$data['pubkey'] = $json['key'];
}
if (!empty($json['name'])) {
$data['name'] = $json['name'];
}
2019-12-23 14:26:06 +00:00
if (!empty($json['photo'])) {
2019-12-22 17:20:11 +00:00
$data['photo'] = $json['photo'];
2019-12-23 14:26:06 +00:00
if (!empty($json['photo_updated'])) {
$data['photo'] .= '?rev=' . urlencode($json['photo_updated']);
}
2019-12-22 17:20:11 +00:00
}
2019-12-22 08:13:12 +00:00
if (!empty($json['address'])) {
$data['addr'] = $json['address'];
}
if (!empty($json['url'])) {
$data['url'] = $json['url'];
}
if (!empty($json['connections_url'])) {
$data['poco'] = $json['connections_url'];
}
if (isset($json['searchable'])) {
$data['hide'] = !$json['searchable'];
}
if (!empty($json['public_forum'])) {
$data['community'] = $json['public_forum'];
2020-06-28 15:43:58 +00:00
$data['account-type'] = User::PAGE_FLAGS_COMMUNITY;
2019-12-22 08:13:12 +00:00
}
if (!empty($json['profile'])) {
$profile = $json['profile'];
if (!empty($profile['description'])) {
$data['about'] = $profile['description'];
}
if (!empty($profile['keywords'])) {
$keywords = implode(', ', $profile['keywords']);
if (!empty($keywords)) {
$data['keywords'] = $keywords;
}
}
$loc = [];
if (!empty($profile['region'])) {
$loc['region'] = $profile['region'];
}
if (!empty($profile['country'])) {
$loc['country-name'] = $profile['country'];
}
$location = Profile::formatLocation($loc);
if (!empty($location)) {
$data['location'] = $location;
}
}
return $data;
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Perform a webfinger request.
2016-07-11 17:48:37 +00:00
*
* For details see RFC 7033: <https://tools.ietf.org/html/rfc7033>
2016-07-07 21:04:30 +00:00
*
* @param string $url Address that should be probed
* @param string $type type
*
2016-07-07 21:04:30 +00:00
* @return array webfinger data
2019-01-06 21:06:53 +00:00
* @throws HTTPException\InternalServerErrorException
2016-07-07 21:04:30 +00:00
*/
public static function webfinger(string $url, string $type): array
{
$xrd_timeout = DI::config()->get('system', 'xrd_timeout', 20);
2016-07-03 20:27:16 +00:00
$curlResult = DI::httpClient()->get($url, $type, [HttpClientOptions::TIMEOUT => $xrd_timeout]);
if ($curlResult->isTimeout()) {
self::$isTimeout = true;
return [];
2017-02-01 21:35:01 +00:00
}
$data = $curlResult->getBody();
2017-02-01 21:35:01 +00:00
$webfinger = json_decode($data, true);
if (!empty($webfinger)) {
if (!isset($webfinger['links'])) {
2020-06-28 15:43:58 +00:00
Logger::info('No json webfinger links', ['url' => $url]);
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
return $webfinger;
}
// If it is not JSON, maybe it is XML
2020-04-27 14:35:50 +00:00
$xrd = XML::parseString($data, true);
if (!is_object($xrd)) {
2020-06-28 15:43:58 +00:00
Logger::info('No webfinger data retrievable', ['url' => $url]);
return [];
}
$xrd_arr = XML::elementToArray($xrd);
if (!isset($xrd_arr['xrd']['link'])) {
2020-06-28 15:43:58 +00:00
Logger::info('No XML webfinger links', ['url' => $url]);
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$webfinger = [];
2016-07-03 20:27:16 +00:00
if (!empty($xrd_arr['xrd']['subject'])) {
$webfinger['subject'] = $xrd_arr['xrd']['subject'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
if (!empty($xrd_arr['xrd']['alias'])) {
$webfinger['aliases'] = $xrd_arr['xrd']['alias'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$webfinger['links'] = [];
2016-07-03 20:27:16 +00:00
foreach ($xrd_arr['xrd']['link'] as $value => $data) {
if (!empty($data['@attributes'])) {
$attributes = $data['@attributes'];
} elseif ($value == '@attributes') {
2016-07-03 20:27:16 +00:00
$attributes = $data;
2017-05-08 15:06:57 +00:00
} else {
2016-07-03 20:27:16 +00:00
continue;
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$webfinger['links'][] = $attributes;
2016-07-03 20:27:16 +00:00
}
return $webfinger;
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Poll the Friendica specific noscrape page.
2016-07-07 21:04:30 +00:00
*
2016-07-11 17:48:37 +00:00
* "noscrape" is a faster alternative to fetch the data from the hcard.
* This functionality was originally created for the directory.
2016-07-10 16:44:48 +00:00
*
* @param string $noscrape_url Link to the noscrape page
* @param array $data The already fetched data
*
2016-07-07 21:04:30 +00:00
* @return array noscrape data
2019-01-06 21:06:53 +00:00
* @throws HTTPException\InternalServerErrorException
2016-07-07 21:04:30 +00:00
*/
private static function pollNoscrape(string $noscrape_url, array $data): array
{
$curlResult = DI::httpClient()->get($noscrape_url, HttpClientAccept::JSON);
if ($curlResult->isTimeout()) {
self::$isTimeout = true;
return $data;
2017-02-01 21:35:01 +00:00
}
$content = $curlResult->getBody();
2017-02-01 21:35:01 +00:00
if (!$content) {
2020-06-28 15:43:58 +00:00
Logger::info('Empty body', ['url' => $noscrape_url]);
return $data;
2017-02-01 21:35:01 +00:00
}
2016-07-03 21:11:21 +00:00
$json = json_decode($content, true);
2017-05-08 15:06:57 +00:00
if (!is_array($json)) {
2020-06-28 15:43:58 +00:00
Logger::info('No json data', ['url' => $noscrape_url]);
return $data;
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (!empty($json['fn'])) {
$data['name'] = $json['fn'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (!empty($json['addr'])) {
$data['addr'] = $json['addr'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (!empty($json['nick'])) {
$data['nick'] = $json['nick'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (!empty($json['guid'])) {
$data['guid'] = $json['guid'];
2017-10-28 11:48:29 +00:00
}
if (!empty($json['comm'])) {
$data['community'] = $json['comm'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (!empty($json['tags'])) {
$keywords = implode(', ', $json['tags']);
if ($keywords != '') {
$data['keywords'] = $keywords;
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
}
$location = Profile::formatLocation($json);
2017-05-08 15:06:57 +00:00
if ($location) {
$data['location'] = $location;
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (!empty($json['about'])) {
$data['about'] = $json['about'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (!empty($json['xmpp'])) {
$data['xmpp'] = $json['xmpp'];
}
if (!empty($json['matrix'])) {
$data['matrix'] = $json['matrix'];
}
if (!empty($json['key'])) {
$data['pubkey'] = $json['key'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (!empty($json['photo'])) {
$data['photo'] = $json['photo'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (!empty($json['dfrn-request'])) {
$data['request'] = $json['dfrn-request'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (!empty($json['dfrn-confirm'])) {
$data['confirm'] = $json['dfrn-confirm'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (!empty($json['dfrn-notify'])) {
$data['notify'] = $json['dfrn-notify'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (!empty($json['dfrn-poll'])) {
$data['poll'] = $json['dfrn-poll'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
if (isset($json['hide'])) {
$data['hide'] = (bool)$json['hide'];
} else {
$data['hide'] = false;
}
2016-07-03 21:11:21 +00:00
return $data;
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Check for valid DFRN data
2016-07-07 21:04:30 +00:00
*
* @param array $data DFRN data
*
* @return int Number of errors
*/
public static function validDfrn(array $data): int
{
$errors = 0;
2017-05-08 15:06:57 +00:00
if (!isset($data['key'])) {
$errors ++;
2017-05-08 15:06:57 +00:00
}
if (!isset($data['dfrn-request'])) {
$errors ++;
2017-05-08 15:06:57 +00:00
}
if (!isset($data['dfrn-confirm'])) {
$errors ++;
2017-05-08 15:06:57 +00:00
}
if (!isset($data['dfrn-notify'])) {
$errors ++;
2017-05-08 15:06:57 +00:00
}
if (!isset($data['dfrn-poll'])) {
$errors ++;
2017-05-08 15:06:57 +00:00
}
return $errors;
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Fetch data from a DFRN profile page and via "noscrape"
2016-07-07 21:04:30 +00:00
*
* @param string $profile_link Link to the profile page
2016-07-07 21:04:30 +00:00
* @return array profile data
2019-01-06 21:06:53 +00:00
* @throws HTTPException\InternalServerErrorException
* @throws \ImagickException
2016-07-07 21:04:30 +00:00
*/
public static function profile(string $profile_link): array
{
$data = [];
2020-06-28 15:43:58 +00:00
Logger::info('Check profile', ['link' => $profile_link]);
// Fetch data via noscrape - this is faster
$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['name'])
|| !isset($data['photo'])
2017-05-08 15:36:04 +00:00
) {
$data = self::pollHcard($profile_link, $data, true);
2017-05-08 15:06:57 +00:00
}
$prof_data = [];
if (empty($data['addr']) || empty($data['nick'])) {
$probe_data = self::uri($profile_link);
$data['addr'] = ($data['addr'] ?? '') ?: $probe_data['addr'];
$data['nick'] = ($data['nick'] ?? '') ?: $probe_data['nick'];
}
$prof_data['addr'] = $data['addr'];
$prof_data['nick'] = $data['nick'];
$prof_data['dfrn-request'] = $data['request'] ?? null;
$prof_data['dfrn-confirm'] = $data['confirm'] ?? null;
$prof_data['dfrn-notify'] = $data['notify'] ?? null;
$prof_data['dfrn-poll'] = $data['poll'] ?? null;
$prof_data['photo'] = $data['photo'] ?? null;
$prof_data['fn'] = $data['name'] ?? null;
$prof_data['key'] = $data['pubkey'] ?? null;
2020-06-28 15:43:58 +00:00
Logger::debug('Result', ['link' => $profile_link, 'data' => $prof_data]);
return $prof_data;
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Check for DFRN contact
2016-07-07 21:04:30 +00:00
*
* @param array $webfinger Webfinger data
* @return array DFRN data
2019-01-06 21:06:53 +00:00
* @throws HTTPException\InternalServerErrorException
2016-07-07 21:04:30 +00:00
*/
private static function dfrn(array $webfinger): array
{
$hcard_url = '';
$data = [];
// The array is reversed to take into account the order of preference for same-rel links
// See: https://tools.ietf.org/html/rfc7033#section-4.4.4
foreach (array_reverse($webfinger['links']) as $link) {
if (($link['rel'] == ActivityNamespace::DFRN) && !empty($link['href'])) {
$data['network'] = Protocol::DFRN;
} elseif (($link['rel'] == ActivityNamespace::FEED) && !empty($link['href'])) {
$data['poll'] = $link['href'];
} elseif (($link['rel'] == 'http://webfinger.net/rel/profile-page') && (($link['type'] ?? '') == 'text/html') && !empty($link['href'])) {
$data['url'] = $link['href'];
} elseif (($link['rel'] == 'http://microformats.org/profile/hcard') && !empty($link['href'])) {
$hcard_url = $link['href'];
} elseif (($link['rel'] == ActivityNamespace::POCO) && !empty($link['href'])) {
$data['poco'] = $link['href'];
} elseif (($link['rel'] == 'http://webfinger.net/rel/avatar') && !empty($link['href'])) {
$data['photo'] = $link['href'];
} elseif (($link['rel'] == 'http://joindiaspora.com/seed_location') && !empty($link['href'])) {
$data['baseurl'] = trim($link['href'], '/');
} elseif (($link['rel'] == 'http://joindiaspora.com/guid') && !empty($link['href'])) {
$data['guid'] = $link['href'];
} elseif (($link['rel'] == 'diaspora-public-key') && !empty($link['href'])) {
$data['pubkey'] = base64_decode($link['href']);
if (strstr($data['pubkey'], 'RSA ')) {
$data['pubkey'] = Crypto::rsaToPem($data['pubkey']);
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
}
}
if (!empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
foreach ($webfinger['aliases'] as $alias) {
if (empty($data['url']) && !strstr($alias, '@')) {
$data['url'] = $alias;
} elseif (!strstr($alias, '@') && Strings::normaliseLink($alias) != Strings::normaliseLink($data['url'])) {
$data['alias'] = $alias;
2017-10-28 11:48:29 +00:00
} elseif (substr($alias, 0, 5) == 'acct:') {
$data['addr'] = substr($alias, 5);
}
2017-07-10 22:35:32 +00:00
}
}
if (!empty($webfinger['subject']) && (substr($webfinger['subject'], 0, 5) == 'acct:')) {
$data['addr'] = substr($webfinger['subject'], 5);
}
if (!isset($data['network']) || ($hcard_url == '')) {
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
2016-07-03 21:11:21 +00:00
// Fetch data via noscrape - this is faster
$noscrape_url = str_replace('/hcard/', '/noscrape/', $hcard_url);
$data = self::pollNoscrape($noscrape_url, $data);
2016-07-03 21:11:21 +00:00
if (isset($data['notify'])
&& isset($data['confirm'])
&& isset($data['request'])
&& isset($data['poll'])
&& isset($data['name'])
&& isset($data['photo'])
2017-05-08 15:36:04 +00:00
) {
2016-07-03 21:11:21 +00:00
return $data;
2017-05-08 15:06:57 +00:00
}
2016-07-03 21:11:21 +00:00
$data = self::pollHcard($hcard_url, $data, true);
2016-07-03 20:27:16 +00:00
return $data;
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Poll the hcard page (Diaspora and Friendica specific)
2016-07-07 21:04:30 +00:00
*
* @param string $hcard_url Link to the hcard page
* @param array $data The already fetched data
* @param boolean $dfrn Poll DFRN specific data
2016-07-07 21:04:30 +00:00
* @return array hcard data
2019-01-06 21:06:53 +00:00
* @throws HTTPException\InternalServerErrorException
2016-07-07 21:04:30 +00:00
*/
private static function pollHcard(string $hcard_url, array $data, bool $dfrn = false): array
{
$curlResult = DI::httpClient()->get($hcard_url, HttpClientAccept::HTML);
if ($curlResult->isTimeout()) {
self::$isTimeout = true;
return [];
2017-02-01 21:35:01 +00:00
}
$content = $curlResult->getBody();
2021-04-10 05:46:19 +00:00
if (empty($content)) {
return [];
2017-02-01 21:35:01 +00:00
}
2017-05-11 15:53:04 +00:00
$doc = new DOMDocument();
2017-05-08 15:06:57 +00:00
if (!@$doc->loadHTML($content)) {
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
2017-05-11 15:53:04 +00:00
$xpath = new DomXPath($doc);
2016-07-03 20:27:16 +00:00
$vcards = $xpath->query("//div[contains(concat(' ', @class, ' '), ' vcard ')]");
2017-05-08 15:06:57 +00:00
if (!is_object($vcards)) {
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
if (!isset($data['baseurl'])) {
$data['baseurl'] = '';
}
if ($vcards->length > 0) {
$vcard = $vcards->item(0);
2016-07-03 20:27:16 +00:00
// 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) && empty($data['guid'])) {
$data['guid'] = $search->item(0)->nodeValue;
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' nickname ')]", $vcard); // */
2017-05-08 15:06:57 +00:00
if ($search->length > 0) {
$data['nick'] = $search->item(0)->nodeValue;
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' fn ')]", $vcard); // */
2017-05-08 15:06:57 +00:00
if ($search->length > 0) {
$data['name'] = $search->item(0)->nodeValue;
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' searchable ')]", $vcard); // */
2017-05-08 15:06:57 +00:00
if ($search->length > 0) {
$data['searchable'] = $search->item(0)->nodeValue;
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$search = $xpath->query("//*[contains(concat(' ', @class, ' '), ' key ')]", $vcard); // */
if ($search->length > 0) {
$data['pubkey'] = $search->item(0)->nodeValue;
if (strstr($data['pubkey'], 'RSA ')) {
$data['pubkey'] = Crypto::rsaToPem($data['pubkey']);
2017-05-08 15:06:57 +00:00
}
}
2016-07-03 20:27:16 +00:00
$search = $xpath->query("//*[@id='pod_location']", $vcard); // */
2017-05-08 15:06:57 +00:00
if ($search->length > 0) {
$data['baseurl'] = trim($search->item(0)->nodeValue, '/');
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
}
$avatar = [];
if (!empty($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($attribute->value);
}
2016-07-03 20:27:16 +00:00
if (isset($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[] = $attr['src'];
}
}
2016-07-03 20:27:16 +00:00
}
if (sizeof($avatar)) {
ksort($avatar);
$data['photo'] = self::fixAvatar(array_pop($avatar), $data['baseurl']);
2016-07-03 20:27:16 +00:00
}
if ($dfrn) {
// Poll DFRN specific data
$search = $xpath->query("//link[contains(concat(' ', @rel), ' dfrn-')]");
if ($search->length > 0) {
2017-05-08 15:06:57 +00:00
foreach ($search as $link) {
//$data['request'] = $search->item(0)->nodeValue;
$attr = [];
2017-05-08 15:06:57 +00:00
foreach ($link->attributes as $attribute) {
2016-07-03 20:27:16 +00:00
$attr[$attribute->name] = trim($attribute->value);
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$data[substr($attr['rel'], 5)] = $attr['href'];
2016-07-03 20:27:16 +00:00
}
}
// Older Friendica versions had used the "uid" field differently than newer versions
if (!empty($data['nick']) && !empty($data['guid']) && ($data['nick'] == $data['guid'])) {
unset($data['guid']);
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
}
return $data;
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Check for Diaspora contact
2016-07-07 21:04:30 +00:00
*
* @param array $webfinger Webfinger data
*
2016-07-07 21:04:30 +00:00
* @return array Diaspora data
2019-01-06 21:06:53 +00:00
* @throws HTTPException\InternalServerErrorException
2016-07-07 21:04:30 +00:00
*/
private static function diaspora(array $webfinger): array
{
$hcard_url = '';
2019-12-23 14:26:06 +00:00
$data = [];
2019-12-22 08:13:12 +00:00
// The array is reversed to take into account the order of preference for same-rel links
// See: https://tools.ietf.org/html/rfc7033#section-4.4.4
foreach (array_reverse($webfinger['links']) as $link) {
if (($link['rel'] == 'http://microformats.org/profile/hcard') && !empty($link['href'])) {
$hcard_url = $link['href'];
} elseif (($link['rel'] == 'http://joindiaspora.com/seed_location') && !empty($link['href'])) {
$data['baseurl'] = trim($link['href'], '/');
} elseif (($link['rel'] == 'http://joindiaspora.com/guid') && !empty($link['href'])) {
$data['guid'] = $link['href'];
} elseif (($link['rel'] == 'http://webfinger.net/rel/profile-page') && (($link['type'] ?? '') == 'text/html') && !empty($link['href'])) {
$data['url'] = $link['href'];
} elseif (($link['rel'] == 'http://webfinger.net/rel/profile-page') && empty($link['type']) && !empty($link['href'])) {
$profile_url = $link['href'];
} elseif (($link['rel'] == ActivityNamespace::FEED) && !empty($link['href'])) {
$data['poll'] = $link['href'];
} elseif (($link['rel'] == ActivityNamespace::POCO) && !empty($link['href'])) {
$data['poco'] = $link['href'];
} elseif (($link['rel'] == 'salmon') && !empty($link['href'])) {
$data['notify'] = $link['href'];
} elseif (($link['rel'] == 'diaspora-public-key') && !empty($link['href'])) {
$data['pubkey'] = base64_decode($link['href']);
if (strstr($data['pubkey'], 'RSA ')) {
$data['pubkey'] = Crypto::rsaToPem($data['pubkey']);
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
}
}
if (empty($data['url']) && !empty($profile_url)) {
$data['url'] = $profile_url;
}
if (empty($data['url']) || empty($hcard_url)) {
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
if (!empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
foreach ($webfinger['aliases'] as $alias) {
if (Strings::normaliseLink($alias) != Strings::normaliseLink($data['url']) && ! strstr($alias, '@')) {
$data['alias'] = $alias;
} elseif (substr($alias, 0, 5) == 'acct:') {
$data['addr'] = substr($alias, 5);
2017-05-08 15:06:57 +00:00
}
}
}
2016-07-03 20:27:16 +00:00
if (!empty($webfinger['subject']) && (substr($webfinger['subject'], 0, 5) == 'acct:')) {
$data['addr'] = substr($webfinger['subject'], 5);
2017-10-28 11:48:29 +00:00
}
2016-07-03 20:27:16 +00:00
// Fetch further information from the hcard
$data = self::pollHcard($hcard_url, $data);
2016-07-03 20:27:16 +00:00
2019-12-23 14:30:48 +00:00
if (!$data) {
return [];
2019-12-23 14:30:48 +00:00
}
if (!empty($data['url'])
&& !empty($data['guid'])
&& !empty($data['baseurl'])
&& !empty($data['pubkey'])
2019-12-22 08:13:12 +00:00
&& !empty($hcard_url)
2017-05-08 15:36:04 +00:00
) {
$data['network'] = Protocol::DIASPORA;
$data['manually-approve'] = false;
2016-07-03 20:27:16 +00:00
// The Diaspora handle must always be lowercase
if (!empty($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'];
$data['batch'] = $data['baseurl'] . '/receive/public';
2019-12-23 14:30:48 +00:00
} else {
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
return $data;
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Check for OStatus contact
2016-07-07 21:04:30 +00:00
*
* @param array $webfinger Webfinger data
* @param bool $short Short detection mode
*
* @return array|bool OStatus data or "false" on error or "true" on short mode
2019-01-06 21:06:53 +00:00
* @throws HTTPException\InternalServerErrorException
2016-07-07 21:04:30 +00:00
*/
private static function ostatus(array $webfinger, bool $short = false)
{
2019-12-23 14:26:06 +00:00
$data = [];
if (!empty($webfinger['aliases']) && is_array($webfinger['aliases'])) {
foreach ($webfinger['aliases'] as $alias) {
if (strstr($alias, '@') && !strstr(Strings::normaliseLink($alias), 'http://')) {
$data['addr'] = str_replace('acct:', '', $alias);
}
}
}
if (!empty($webfinger['subject']) && strstr($webfinger['subject'], '@')
&& !strstr(Strings::normaliseLink($webfinger['subject']), 'http://')
) {
$data['addr'] = str_replace('acct:', '', $webfinger['subject']);
}
2017-10-28 11:48:29 +00:00
if (!empty($webfinger['links'])) {
// The array is reversed to take into account the order of preference for same-rel links
// See: https://tools.ietf.org/html/rfc7033#section-4.4.4
foreach (array_reverse($webfinger['links']) as $link) {
if (($link['rel'] == 'http://webfinger.net/rel/profile-page')
&& (($link['type'] ?? '') == 'text/html')
&& ($link['href'] != '')
2017-08-29 04:51:02 +00:00
) {
$data['url'] = $data['alias'] = $link['href'];
} elseif (($link['rel'] == 'salmon') && !empty($link['href'])) {
$data['notify'] = $link['href'];
} elseif (($link['rel'] == ActivityNamespace::FEED) && !empty($link['href'])) {
$data['poll'] = $link['href'];
} elseif (($link['rel'] == 'magic-public-key') && !empty($link['href'])) {
$pubkey = $link['href'];
2017-08-29 04:51:02 +00:00
if (substr($pubkey, 0, 5) === 'data:') {
if (strstr($pubkey, ',')) {
$pubkey = substr($pubkey, strpos($pubkey, ',') + 1);
} else {
$pubkey = substr($pubkey, 5);
}
} elseif (Strings::normaliseLink($pubkey) == 'http://') {
$curlResult = DI::httpClient()->get($pubkey, HttpClientAccept::MAGIC_KEY);
if ($curlResult->isTimeout()) {
self::$isTimeout = true;
return $short ? false : [];
2017-08-29 04:51:02 +00:00
}
2022-03-29 17:26:31 +00:00
Logger::debug('Fetched public key', ['Content-Type' => $curlResult->getHeader('Content-Type'), 'url' => $pubkey]);
$pubkey = $curlResult->getBody();
}
2016-07-03 20:27:16 +00:00
$key = explode('.', $pubkey);
2016-07-03 20:27:16 +00:00
2017-08-29 04:51:02 +00:00
if (sizeof($key) >= 3) {
$m = Strings::base64UrlDecode($key[1]);
$e = Strings::base64UrlDecode($key[2]);
$data['pubkey'] = Crypto::meToPem($m, $e);
2017-08-29 04:51:02 +00:00
}
2016-07-03 20:27:16 +00:00
}
}
}
if (isset($data['notify']) && isset($data['pubkey'])
&& isset($data['poll'])
&& isset($data['url'])
2017-05-08 15:36:04 +00:00
) {
$data['network'] = Protocol::OSTATUS;
$data['manually-approve'] = false;
2019-12-22 08:13:12 +00:00
} else {
return $short ? false : [];
}
if ($short) {
return true;
}
2016-07-03 20:27:16 +00:00
// Fetch all additional data from the feed
$curlResult = DI::httpClient()->get($data['poll'], HttpClientAccept::FEED_XML);
2018-10-11 05:45:04 +00:00
if ($curlResult->isTimeout()) {
self::$isTimeout = true;
return [];
}
2018-10-11 05:45:04 +00:00
$feed = $curlResult->getBody();
$feed_data = Feed::import($feed);
if (!$feed_data) {
return [];
}
2017-08-21 20:21:04 +00:00
if (!empty($feed_data['header']['author-name'])) {
$data['name'] = $feed_data['header']['author-name'];
}
if (!empty($feed_data['header']['author-nick'])) {
$data['nick'] = $feed_data['header']['author-nick'];
}
if (!empty($feed_data['header']['author-avatar'])) {
$data['photo'] = self::fixAvatar($feed_data['header']['author-avatar'], $data['url']);
}
if (!empty($feed_data['header']['author-id'])) {
$data['alias'] = $feed_data['header']['author-id'];
}
if (!empty($feed_data['header']['author-location'])) {
$data['location'] = $feed_data['header']['author-location'];
}
if (!empty($feed_data['header']['author-about'])) {
$data['about'] = $feed_data['header']['author-about'];
}
2016-07-03 20:27:16 +00:00
// OStatus has serious issues when the the url doesn't fit (ssl vs. non ssl)
// So we take the value that we just fetched, although the other one worked as well
if (!empty($feed_data['header']['author-link'])) {
$data['url'] = $feed_data['header']['author-link'];
}
2017-08-21 20:21:04 +00:00
if ($data['url'] == $data['alias']) {
$data['alias'] = '';
2017-08-21 20:21:04 +00:00
}
2016-07-03 20:27:16 +00:00
/// @todo Fetch location and "about" from the feed as well
return $data;
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Fetch data from a pump.io profile page
2016-07-07 21:04:30 +00:00
*
* @param string $profile_link Link to the profile page
*
* @return array Profile data
2016-07-07 21:04:30 +00:00
*/
private static function pumpioProfileData(string $profile_link): array
{
$curlResult = DI::httpClient()->get($profile_link, HttpClientAccept::HTML);
2021-04-10 05:46:19 +00:00
if (!$curlResult->isSuccess() || empty($curlResult->getBody())) {
return [];
}
2017-05-11 15:53:04 +00:00
$doc = new DOMDocument();
if (!@$doc->loadHTML($curlResult->getBody())) {
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
2017-05-11 15:53:04 +00:00
$xpath = new DomXPath($doc);
2016-07-03 20:27:16 +00:00
$data = [];
$data['name'] = $xpath->query("//span[contains(@class, 'p-name')]")->item(0)->nodeValue;
2016-07-03 20:27:16 +00:00
if ($data['name'] == '') {
2018-02-13 05:54:09 +00:00
// 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));
2018-02-13 05:54:09 +00:00
if ($pos) {
$data['name'] = trim(substr($data['name'], 0, $pos));
2018-02-13 05:54:09 +00:00
}
}
$data['location'] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'p-locality')]");
2018-02-13 05:54:09 +00:00
if ($data['location'] == '') {
$data['location'] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'location')]");
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$data['about'] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'p-note')]");
2018-02-13 05:54:09 +00:00
if ($data['about'] == '') {
$data['about'] = XML::getFirstNodeValue($xpath, "//p[contains(@class, 'summary')]");
2018-02-13 05:54:09 +00:00
}
$avatar = $xpath->query("//img[contains(@class, 'u-photo')]")->item(0);
if (!$avatar) {
$avatar = $xpath->query("//img[@class='img-rounded media-object']")->item(0);
}
2017-05-08 15:06:57 +00:00
if ($avatar) {
foreach ($avatar->attributes as $attribute) {
if ($attribute->name == 'src') {
$data['photo'] = trim($attribute->value);
2017-05-08 15:06:57 +00:00
}
}
}
2016-07-03 20:27:16 +00:00
return $data;
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Check for pump.io contact
2016-07-07 21:04:30 +00:00
*
2019-12-22 17:20:11 +00:00
* @param array $webfinger Webfinger data
* @param string $addr
*
2016-07-07 21:04:30 +00:00
* @return array pump.io data
*/
private static function pumpio(array $webfinger, string $addr): array
{
2019-12-23 14:26:06 +00:00
$data = [];
// The array is reversed to take into account the order of preference for same-rel links
// See: https://tools.ietf.org/html/rfc7033#section-4.4.4
foreach (array_reverse($webfinger['links']) as $link) {
if (($link['rel'] == 'http://webfinger.net/rel/profile-page')
&& (($link['type'] ?? '') == 'text/html')
&& ($link['href'] != '')
2017-05-08 15:36:04 +00:00
) {
$data['url'] = $link['href'];
} elseif (($link['rel'] == 'activity-inbox') && ($link['href'] != '')) {
$data['notify'] = $link['href'];
} elseif (($link['rel'] == 'activity-outbox') && ($link['href'] != '')) {
$data['poll'] = $link['href'];
} elseif (($link['rel'] == 'dialback') && ($link['href'] != '')) {
$data['dialback'] = $link['href'];
}
}
if (isset($data['poll']) && isset($data['notify'])
&& isset($data['dialback'])
&& isset($data['url'])
2017-05-08 15:36:04 +00:00
) {
2016-07-03 20:27:16 +00:00
// by now we use these fields only for the network type detection
// So we unset all data that isn't used at the moment
unset($data['dialback']);
2016-07-03 20:27:16 +00:00
$data['network'] = Protocol::PUMPIO;
2017-05-08 15:06:57 +00:00
} else {
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$profile_data = self::pumpioProfileData($data['url']);
2016-07-03 20:27:16 +00:00
2017-05-08 15:06:57 +00:00
if (!$profile_data) {
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$data = array_merge($data, $profile_data);
2018-02-13 07:10:05 +00:00
if (($addr != '') && ($data['name'] != '')) {
$name = trim(str_replace($addr, '', $data['name']));
if ($name != '') {
$data['name'] = $name;
}
}
2016-07-03 20:27:16 +00:00
return $data;
}
2016-07-07 21:04:30 +00:00
/**
* Checks HTML page for RSS feed link
2016-07-07 21:04:30 +00:00
*
* @param string $url Page link
* @param string $body Page body string
*
* @return string|false Feed link or false if body was invalid HTML document
2016-07-07 21:04:30 +00:00
*/
public static function getFeedLink(string $url, string $body)
{
if (empty($body)) {
return '';
}
$doc = new DOMDocument();
if (!@$doc->loadHTML($body)) {
return false;
2017-05-08 15:06:57 +00:00
}
$xpath = new DOMXPath($doc);
$feedUrl = $xpath->evaluate('string(/html/head/link[@type="application/rss+xml" and @rel="alternate"]/@href)');
$feedUrl = $feedUrl ? self::ensureAbsoluteLinkFromHTMLDoc($feedUrl, $url, $xpath) : '';
return $feedUrl;
}
/**
* Return an absolute URL in the context of a HTML document retrieved from the provided URL.
*
* Loosely based on RFC 1808
*
* @see https://tools.ietf.org/html/rfc1808
*
* @param string $href The potential relative href found in the HTML document
* @param string $base The HTML document URL
* @param DOMXPath $xpath The HTML document XPath
*
* @return string Absolute URL
*/
private static function ensureAbsoluteLinkFromHTMLDoc(string $href, string $base, DOMXPath $xpath): string
{
if (filter_var($href, FILTER_VALIDATE_URL)) {
return $href;
2017-05-08 15:06:57 +00:00
}
$base = $xpath->evaluate('string(/html/head/base/@href)') ?: $base;
$baseParts = parse_url($base);
2020-07-25 08:07:22 +00:00
if (empty($baseParts['host'])) {
return $href;
}
// Naked domain case (scheme://basehost)
$path = $baseParts['path'] ?? '/';
// Remove the filename part of the path if it exists (/base/path/file)
$path = implode('/', array_slice(explode('/', $path), 0, -1));
$hrefParts = parse_url($href);
2020-12-31 07:54:56 +00:00
if (!empty($hrefParts['path'])) {
// Root path case (/path) including relative scheme case (//host/path)
if ($hrefParts['path'] && $hrefParts['path'][0] == '/') {
$path = $hrefParts['path'];
} else {
$path = $path . '/' . $hrefParts['path'];
// Resolve arbitrary relative path
// Lifted from https://www.php.net/manual/en/function.realpath.php#84012
$parts = array_filter(explode('/', $path), 'strlen');
$absolutes = [];
2020-12-31 07:54:56 +00:00
foreach ($parts as $part) {
if ('.' == $part) continue;
if ('..' == $part) {
array_pop($absolutes);
} else {
$absolutes[] = $part;
}
}
2020-12-31 07:54:56 +00:00
$path = '/' . implode('/', $absolutes);
}
}
// Relative scheme case (//host/path)
$baseParts['host'] = $hrefParts['host'] ?? $baseParts['host'];
$baseParts['path'] = $path;
unset($baseParts['query']);
unset($baseParts['fragment']);
return Network::unparseURL($baseParts);
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Check for feed contact
2016-07-07 21:04:30 +00:00
*
* @param string $url Profile link
2016-07-07 21:04:30 +00:00
* @param boolean $probe Do a probe if the page contains a feed link
*
2016-07-07 21:04:30 +00:00
* @return array feed data
2019-01-06 21:06:53 +00:00
* @throws HTTPException\InternalServerErrorException
2016-07-07 21:04:30 +00:00
*/
private static function feed(string $url, bool $probe = true): array
{
$curlResult = DI::httpClient()->get($url, HttpClientAccept::FEED_XML);
if ($curlResult->isTimeout()) {
self::$isTimeout = true;
return [];
2017-02-01 21:35:01 +00:00
}
$feed = $curlResult->getBody();
$feed_data = Feed::import($feed);
2016-07-03 20:27:16 +00:00
if (!$feed_data) {
2017-05-08 15:06:57 +00:00
if (!$probe) {
return [];
2017-05-08 15:06:57 +00:00
}
$feed_url = self::getFeedLink($url, $feed);
2017-05-08 15:06:57 +00:00
if (!$feed_url) {
return [];
2017-05-08 15:06:57 +00:00
}
return self::feed($feed_url, false);
}
2016-07-03 20:27:16 +00:00
if (!empty($feed_data['header']['author-name'])) {
$data['name'] = $feed_data['header']['author-name'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
if (!empty($feed_data['header']['author-nick'])) {
$data['nick'] = $feed_data['header']['author-nick'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
if (!empty($feed_data['header']['author-avatar'])) {
$data['photo'] = $feed_data['header']['author-avatar'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
if (!empty($feed_data['header']['author-id'])) {
$data['alias'] = $feed_data['header']['author-id'];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$data['url'] = $url;
$data['poll'] = $url;
2016-07-03 20:27:16 +00:00
$data['network'] = Protocol::FEED;
2016-07-03 20:27:16 +00:00
return $data;
}
2016-07-07 21:04:30 +00:00
/**
2020-01-19 06:05:23 +00:00
* Check for mail contact
2016-07-07 21:04:30 +00:00
*
* @param string $uri Profile link
2016-07-07 21:04:30 +00:00
* @param integer $uid User ID
*
2016-07-07 21:04:30 +00:00
* @return array mail data
2019-01-06 21:06:53 +00:00
* @throws \Exception
2016-07-07 21:04:30 +00:00
*/
private static function mail(string $uri, int $uid): array
{
if (!Network::isEmailDomainValid($uri)) {
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
if ($uid == 0) {
return [];
}
$user = DBA::selectFirst('user', ['prvkey'], ['uid' => $uid]);
2016-07-03 20:27:16 +00:00
$condition = ["`uid` = ? AND `server` != ''", $uid];
$fields = ['pass', 'user', 'server', 'port', 'ssltype', 'mailbox'];
$mailacct = DBA::selectFirst('mailacct', $fields, $condition);
2016-07-03 20:27:16 +00:00
if (!DBA::isResult($user) || !DBA::isResult($mailacct)) {
return [];
}
$mailbox = Email::constructMailboxName($mailacct);
$password = '';
openssl_private_decrypt(hex2bin($mailacct['pass']), $password, $user['prvkey']);
$mbox = Email::connect($mailbox, $mailacct['user'], $password);
if (!$mbox) {
return [];
2017-05-08 15:06:57 +00:00
}
2016-07-03 20:27:16 +00:00
$msgs = Email::poll($mbox, $uri);
2020-06-28 15:43:58 +00:00
Logger::info('Messages found', ['uri' => $uri, 'count' => count($msgs)]);
if (!count($msgs)) {
return [];
}
2017-05-08 16:28:30 +00:00
$phost = substr($uri, strpos($uri, '@') + 1);
2016-07-03 20:27:16 +00:00
$data = [
'addr' => $uri,
'network' => Protocol::MAIL,
'name' => substr($uri, 0, strpos($uri, '@')),
'photo' => Network::lookupAvatarByEmail($uri),
'url' => 'mailto:' . $uri,
'notify' => 'smtp ' . Strings::getRandomHex(),
'poll' => 'email ' . Strings::getRandomHex(),
];
$data['nick'] = $data['name'];
2016-07-03 20:27:16 +00:00
$x = Email::messageMeta($mbox, $msgs[0]);
2017-05-08 15:06:57 +00:00
if (stristr($x[0]->from, $uri)) {
2016-07-03 20:27:16 +00:00
$adr = imap_rfc822_parse_adrlist($x[0]->from, '');
2017-05-08 15:06:57 +00:00
} elseif (stristr($x[0]->to, $uri)) {
2016-07-03 20:27:16 +00:00
$adr = imap_rfc822_parse_adrlist($x[0]->to, '');
2017-05-08 15:06:57 +00:00
}
2017-05-08 15:06:57 +00:00
if (isset($adr)) {
foreach ($adr as $feadr) {
if ((strcasecmp($feadr->mailbox, $data['name']) == 0)
2016-07-03 20:27:16 +00:00
&&(strcasecmp($feadr->host, $phost) == 0)
2017-05-08 15:36:04 +00:00
&& (strlen($feadr->personal))
) {
2016-07-03 20:27:16 +00:00
$personal = imap_mime_header_decode($feadr->personal);
$data['name'] = '';
2017-05-08 15:06:57 +00:00
foreach ($personal as $perspart) {
if ($perspart->charset != 'default') {
$data['name'] .= iconv($perspart->charset, 'UTF-8//IGNORE', $perspart->text);
2017-05-08 15:06:57 +00:00
} else {
$data['name'] .= $perspart->text;
2017-05-08 15:06:57 +00:00
}
}
2016-07-03 20:27:16 +00:00
}
}
}
2017-09-23 22:53:44 +00:00
if (!empty($mbox)) {
imap_close($mbox);
}
2016-07-03 20:27:16 +00:00
return $data;
}
/**
2020-01-19 06:05:23 +00:00
* 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
*
* @return string fixed avatar path
2019-01-06 21:06:53 +00:00
* @throws \Exception
*/
public static function fixAvatar(string $avatar, string $base): string
{
$base_parts = parse_url($base);
// Remove all parts that could create a problem
unset($base_parts['path']);
unset($base_parts['query']);
unset($base_parts['fragment']);
$avatar_parts = parse_url($avatar);
// Now we mix them
$parts = array_merge($base_parts, $avatar_parts);
// And put them together again
2017-05-08 15:36:04 +00:00
$scheme = isset($parts['scheme']) ? $parts['scheme'] . '://' : '';
$host = isset($parts['host']) ? $parts['host'] : '';
$port = isset($parts['port']) ? ':' . $parts['port'] : '';
$path = isset($parts['path']) ? $parts['path'] : '';
$query = isset($parts['query']) ? '?' . $parts['query'] : '';
$fragment = isset($parts['fragment']) ? '#' . $parts['fragment'] : '';
$fixed = $scheme.$host.$port.$path.$query.$fragment;
2020-06-28 15:43:58 +00:00
Logger::debug('Avatar fixed', ['base' => $base, 'avatar' => $avatar, 'fixed' => $fixed]);
return $fixed;
}
2020-08-05 06:50:51 +00:00
/**
* Fetch the last date that the contact had posted something (publically)
*
* @param array $data probing result
*
2020-08-05 06:50:51 +00:00
* @return string last activity
*/
public static function getLastUpdate(array $data): string
2020-08-05 06:50:51 +00:00
{
$uid = User::getIdForURL($data['url']);
if (!empty($uid)) {
$contact = Contact::selectFirst(['url', 'last-item'], ['self' => true, 'uid' => $uid]);
if (!empty($contact['last-item'])) {
return $contact['last-item'];
}
}
2020-08-05 06:50:51 +00:00
if ($lastUpdate = self::updateFromNoScrape($data)) {
return $lastUpdate;
}
if (!empty($data['outbox'])) {
return self::updateFromOutbox($data['outbox'], $data);
} elseif (!empty($data['poll']) && ($data['network'] == Protocol::ACTIVITYPUB)) {
return self::updateFromOutbox($data['poll'], $data);
} elseif (!empty($data['poll'])) {
return self::updateFromFeed($data);
}
return '';
}
/**
* Fetch the last activity date from the "noscrape" endpoint
*
* @param array $data Probing result
*
* @return string last activity or true if update was successful or the server was unreachable
2020-08-05 06:50:51 +00:00
*/
private static function updateFromNoScrape(array $data): string
2020-08-05 06:50:51 +00:00
{
2020-08-05 07:51:15 +00:00
if (empty($data['baseurl'])) {
return '';
}
2020-08-05 06:50:51 +00:00
// Check the 'noscrape' endpoint when it is a Friendica server
$gserver = DBA::selectFirst('gserver', ['noscrape'], ["`nurl` = ? AND `noscrape` != ''",
2020-08-05 07:51:15 +00:00
Strings::normaliseLink($data['baseurl'])]);
2020-08-05 06:50:51 +00:00
if (!DBA::isResult($gserver)) {
return '';
}
$curlResult = DI::httpClient()->get($gserver['noscrape'] . '/' . $data['nick'], HttpClientAccept::JSON);
2020-08-05 06:50:51 +00:00
if ($curlResult->isSuccess() && !empty($curlResult->getBody())) {
$noscrape = json_decode($curlResult->getBody(), true);
if (!empty($noscrape) && !empty($noscrape['updated'])) {
return DateTimeFormat::utc($noscrape['updated'], DateTimeFormat::MYSQL);
}
}
return '';
}
/**
* Fetch the last activity date from an ActivityPub Outbox
*
* @param string $feed
* @param array $data Probing result
*
2020-08-05 06:50:51 +00:00
* @return string last activity
* @throws \Friendica\Network\HTTPException\InternalServerErrorException
*/
private static function updateFromOutbox(string $feed, array $data): string
2020-08-05 06:50:51 +00:00
{
$outbox = ActivityPub::fetchContent($feed);
if (empty($outbox)) {
return '';
}
if (!empty($outbox['orderedItems'])) {
$items = $outbox['orderedItems'];
} elseif (!empty($outbox['first']['orderedItems'])) {
$items = $outbox['first']['orderedItems'];
} elseif (!empty($outbox['first']['href']) && ($outbox['first']['href'] != $feed)) {
return self::updateFromOutbox($outbox['first']['href'], $data);
} elseif (!empty($outbox['first'])) {
if (is_string($outbox['first']) && ($outbox['first'] != $feed)) {
return self::updateFromOutbox($outbox['first'], $data);
} else {
Logger::warning('Unexpected data', ['outbox' => $outbox]);
}
return '';
} else {
$items = [];
}
$last_updated = '';
foreach ($items as $activity) {
if (!empty($activity['published'])) {
$published = DateTimeFormat::utc($activity['published']);
} elseif (!empty($activity['object']['published'])) {
$published = DateTimeFormat::utc($activity['object']['published']);
} else {
continue;
}
if ($last_updated < $published) {
$last_updated = $published;
}
}
if (!empty($last_updated)) {
return $last_updated;
}
return '';
}
/**
* Fetch the last activity date from an XML feed
*
* @param array $data Probing result
* @return string last activity
*/
private static function updateFromFeed(array $data): string
2020-08-05 06:50:51 +00:00
{
// Search for the newest entry in the feed
$curlResult = DI::httpClient()->get($data['poll'], HttpClientAccept::ATOM_XML);
if (!$curlResult->isSuccess() || !$curlResult->getBody()) {
2020-08-05 06:50:51 +00:00
return '';
}
$doc = new DOMDocument();
@$doc->loadXML($curlResult->getBody());
$xpath = new DOMXPath($doc);
$xpath->registerNamespace('atom', 'http://www.w3.org/2005/Atom');
$entries = $xpath->query('/atom:feed/atom:entry');
$last_updated = '';
foreach ($entries as $entry) {
$published_item = $xpath->query('atom:published/text()', $entry)->item(0);
$updated_item = $xpath->query('atom:updated/text()' , $entry)->item(0);
$published = !empty($published_item->nodeValue) ? DateTimeFormat::utc($published_item->nodeValue) : null;
$updated = !empty($updated_item->nodeValue) ? DateTimeFormat::utc($updated_item->nodeValue) : null;
if (empty($published) || empty($updated)) {
Logger::notice('Invalid entry for XPath.', ['entry' => $entry, 'url' => $data['url']]);
continue;
}
if ($last_updated < $published) {
$last_updated = $published;
}
if ($last_updated < $updated) {
$last_updated = $updated;
}
}
if (!empty($last_updated)) {
return $last_updated;
}
return '';
}
2020-08-18 19:45:01 +00:00
/**
* Probe data from local profiles without network traffic
*
* @param string $url
*
2020-08-18 19:45:01 +00:00
* @return array probed data
* @throws HTTPException\InternalServerErrorException
* @throws HTTPException\NotFoundException
2020-08-18 19:45:01 +00:00
*/
private static function localProbe(string $url): array
{
try {
$uid = User::getIdForURL($url);
if (!$uid) {
throw new HTTPException\NotFoundException('User not found.');
}
$owner = User::getOwnerDataById($uid);
$approfile = ActivityPub\Transmitter::getProfile($uid);
if (empty($owner['gsid'])) {
$owner['gsid'] = GServer::getID($approfile['generator']['url']);
}
$data = [
'name' => $owner['name'], 'nick' => $owner['nick'], 'guid' => $approfile['diaspora:guid'] ?? '',
'url' => $owner['url'], 'addr' => $owner['addr'], 'alias' => $owner['alias'],
'photo' => User::getAvatarUrl($owner),
'header' => $owner['header'] ? Contact::getHeaderUrlForId($owner['id'], $owner['updated']) : '',
'account-type' => $owner['contact-type'], 'community' => ($owner['contact-type'] == User::ACCOUNT_TYPE_COMMUNITY),
'keywords' => $owner['keywords'], 'location' => $owner['location'], 'about' => $owner['about'],
'xmpp' => $owner['xmpp'], 'matrix' => $owner['matrix'],
'hide' => !$owner['net-publish'], 'batch' => '', 'notify' => $owner['notify'],
'poll' => $owner['poll'], 'request' => $owner['request'], 'confirm' => $owner['confirm'],
'subscribe' => $approfile['generator']['url'] . '/follow?url={uri}', 'poco' => $owner['poco'],
'following' => $approfile['following'], 'followers' => $approfile['followers'],
'inbox' => $approfile['inbox'], 'outbox' => $approfile['outbox'],
'sharedinbox' => $approfile['endpoints']['sharedInbox'], 'network' => Protocol::DFRN,
'pubkey' => $owner['upubkey'], 'baseurl' => $approfile['generator']['url'], 'gsid' => $owner['gsid'],
'manually-approve' => in_array($owner['page-flags'], [User::PAGE_FLAGS_NORMAL, User::PAGE_FLAGS_PRVGROUP])
];
} catch (Exception $e) {
// Default values for non existing targets
$data = [
'name' => $url, 'nick' => $url, 'url' => $url, 'network' => Protocol::PHANTOM,
'photo' => DI::baseUrl() . Contact::DEFAULT_AVATAR_PHOTO
];
}
return self::rearrangeData($data);
}
2016-07-03 20:27:16 +00:00
}