Merge pull request #12615 from MrPetovan/task/12601-update-gserver-block

Drop UpdateGServer worker task if domain is blocked
2023-01-05 12:37:00 +01:00 · 2023-01-05 12:37:00 +01:00 · 2f28ec7af0
parent 0d2da55dfe 647ab1d04a
commit 2f28ec7af0
6 changed files with 179 additions and 16 deletions
--- a/src/Model/GServer.php
+++ b/src/Model/GServer.php
@ -44,7 +44,9 @@ use Friendica\Util\Network;
 use Friendica\Util\Strings;
 use Friendica\Util\XML;
 use Friendica\Network\HTTPException;
+use Friendica\Worker\UpdateGServer;
 use GuzzleHttp\Psr7\Uri;
+use Psr\Http\Message\UriInterface;

 /**
 * This class handles GServer related functions
@ -99,11 +101,11 @@ class GServer
 	 */
 	public static function add(string $url, bool $only_nodeinfo = false)
 	{
-		if (self::getID($url, false)) {
+		if (self::getID($url)) {
 			return;
 		}

-		Worker::add(Worker::PRIORITY_LOW, 'UpdateGServer', $url, $only_nodeinfo);
+		UpdateGServer::add(Worker::PRIORITY_LOW, $url, $only_nodeinfo);
 	}

 	/**
@ -191,8 +193,9 @@ class GServer
 			return false;
 		} else {
 			if (strtotime($gserver['next_contact']) < time()) {
-				Worker::add(Worker::PRIORITY_LOW, 'UpdateGServer', $gserver['url'], false);
+				UpdateGServer::add(Worker::PRIORITY_LOW, $gserver['url']);
 			}
+
 			return self::isDefunct($gserver);
 		}
 	}
@ -210,8 +213,9 @@ class GServer
 			return true;
 		} else {
 			if (strtotime($gserver['next_contact']) < time()) {
-				Worker::add(Worker::PRIORITY_LOW, 'UpdateGServer', $gserver['url'], false);
+				UpdateGServer::add(Worker::PRIORITY_LOW, $gserver['url']);
 			}
+
 			return !$gserver['failed'] && in_array($gserver['network'], Protocol::FEDERATED);
 		}
 	}
@ -252,7 +256,7 @@ class GServer
 		}

 		if (!empty($server) && (empty($gserver) || strtotime($gserver['next_contact']) < time())) {
-			Worker::add(Worker::PRIORITY_LOW, 'UpdateGServer', $server, false);
+			UpdateGServer::add(Worker::PRIORITY_LOW, $server);
 		}

 		return $reachable;
@ -375,7 +379,7 @@ class GServer
 			Logger::info('Reset failed status for server', ['url' => $gserver['url']]);

 			if (strtotime($gserver['next_contact']) < time()) {
-				Worker::add(Worker::PRIORITY_LOW, 'UpdateGServer', $gserver['url'], false);
+				UpdateGServer::add(Worker::PRIORITY_LOW, $gserver['url']);
 			}
 		}
 	}
@ -393,7 +397,7 @@ class GServer
 			Logger::info('Set failed status for server', ['url' => $gserver['url']]);

 			if (strtotime($gserver['next_contact']) < time()) {
-				Worker::add(Worker::PRIORITY_LOW, 'UpdateGServer', $gserver['url'], false);
+				UpdateGServer::add(Worker::PRIORITY_LOW, $gserver['url']);
 			}
 		}
 	}
@ -442,18 +446,41 @@ class GServer
 	 *
 	 * @return string cleaned URL
 	 * @throws Exception
+	 * @deprecated since 2023.03 Use cleanUri instead
 	 */
 	public static function cleanURL(string $dirtyUrl): string
 	{
 		try {
-			$url = str_replace('/index.php', '', trim($dirtyUrl, '/'));
-			return (string)(new Uri($url))->withUserInfo('')->withQuery('')->withFragment('');
+			return (string)self::cleanUri(new Uri($dirtyUrl));
 		} catch (\Throwable $e) {
-			Logger::warning('Invalid URL', ['dirtyUrl' => $dirtyUrl, 'url' => $url]);
+			Logger::warning('Invalid URL', ['dirtyUrl' => $dirtyUrl]);
 			return '';
 		}
 	}

+	/**
+	 * Remove unwanted content from the given URI
+	 *
+	 * @param UriInterface $dirtyUri
+	 *
+	 * @return UriInterface cleaned URI
+	 * @throws Exception
+	 */
+	public static function cleanUri(UriInterface $dirtyUri): string
+	{
+		return $dirtyUri
+			->withUserInfo('')
+			->withQuery('')
+			->withFragment('')
+			->withPath(
+				preg_replace(
+					'#(?:^|/)index\.php#',
+					'',
+					rtrim($dirtyUri->getPath(), '/')
+				)
+			);
+	}
+
 	/**
 	 * Detect server data (type, protocol, version number, ...)
 	 * The detected data is then updated or inserted in the gserver table.
--- a/src/Util/Network.php
+++ b/src/Util/Network.php
@ -29,6 +29,7 @@ use Friendica\Network\HTTPClient\Client\HttpClientAccept;
 use Friendica\Network\HTTPClient\Client\HttpClientOptions;
 use Friendica\Network\HTTPException\NotModifiedException;
 use GuzzleHttp\Psr7\Uri;
+use Psr\Http\Message\UriInterface;

 class Network
 {
@ -177,11 +178,28 @@ class Network
 	 * @param string $url The url to check the domain from
 	 *
 	 * @return boolean
+	 *
+	 * @deprecated since 2023.03 Use isUriBlocked instead
 	 */
 	public static function isUrlBlocked(string $url): bool
 	{
-		$host = @parse_url($url, PHP_URL_HOST);
-		if (!$host) {
+		try {
+			return self::isUriBlocked(new Uri($url));
+		} catch (\Throwable $e) {
+			Logger::warning('Invalid URL', ['url' => $url]);
+			return false;
+		}
+	}
+
+	/**
+	 * Checks if the provided URI domain is on the domain blocklist.
+	 *
+	 * @param UriInterface $uri
+	 * @return boolean
+	 */
+	public static function isUriBlocked(UriInterface $uri): bool
+	{
+		if (!$uri->getHost()) {
 			return false;
 		}

@ -191,7 +209,7 @@ class Network
 		}

 		foreach ($domain_blocklist as $domain_block) {
-			if (fnmatch(strtolower($domain_block['domain']), strtolower($host))) {
+			if (fnmatch(strtolower($domain_block['domain']), strtolower($uri->getHost()))) {
 				return true;
 			}
 		}
--- a/src/Worker/UpdateContact.php
+++ b/src/Worker/UpdateContact.php
@ -25,6 +25,7 @@ use Friendica\Core\Logger;
 use Friendica\Core\Worker;
 use Friendica\Model\Contact;
 use Friendica\Network\HTTPException\InternalServerErrorException;
+use Friendica\Util\Network;

 class UpdateContact
 {
@ -38,6 +39,11 @@ class UpdateContact
 	 */
 	public static function execute(int $contact_id)
 	{
+		// Silently dropping the task if the contact is blocked
+		if (Contact::isBlocked($contact_id)) {
+			return;
+		}
+
 		$success = Contact::updateFromProbe($contact_id);

 		Logger::info('Updated from probe', ['id' => $contact_id, 'success' => $success]);
@ -55,6 +61,11 @@ class UpdateContact
 			throw new \InvalidArgumentException('Invalid value provided for contact_id');
 		}

+		// Dropping the task if the contact is blocked
+		if (Contact::isBlocked($contact_id)) {
+			return 0;
+		}
+
 		return Worker::add($run_parameters, 'UpdateContact', $contact_id);
 	}
 }
--- a/src/Worker/UpdateGServer.php
+++ b/src/Worker/UpdateGServer.php
@ -22,9 +22,14 @@
 namespace Friendica\Worker;

 use Friendica\Core\Logger;
+use Friendica\Core\Worker;
 use Friendica\Database\DBA;
 use Friendica\Model\GServer;
+use Friendica\Network\HTTPException\InternalServerErrorException;
+use Friendica\Util\Network;
 use Friendica\Util\Strings;
+use GuzzleHttp\Psr7\Uri;
+use Psr\Http\Message\UriInterface;

 class UpdateGServer
 {
@ -34,8 +39,9 @@ class UpdateGServer
 	 * @param string  $server_url    Server URL
 	 * @param boolean $only_nodeinfo Only use nodeinfo for server detection
 	 * @return void
+	 * @throws \Exception
 	 */
-	public static function execute(string $server_url, bool $only_nodeinfo = false)
+	public static function execute(string $server_url, bool $only_nodeinfo)
 	{
 		if (empty($server_url)) {
 			return;
@ -47,6 +53,11 @@ class UpdateGServer
 			return;
 		}

+		// Silently dropping the worker task if the server domain is blocked
+		if (Network::isUrlBlocked($filtered)) {
+			return;
+		}
+
 		if (($filtered != $server_url) && DBA::exists('gserver', ['nurl' => Strings::normaliseLink($server_url)])) {
 			GServer::setFailureByUrl($server_url);
 			return;
@ -61,4 +72,23 @@ class UpdateGServer
 		$ret = GServer::check($filtered, '', true, $only_nodeinfo);
 		Logger::info('Updated gserver', ['url' => $filtered, 'result' => $ret]);
 	}
+
+	/**
+	 * @param array|int $run_parameters Priority constant or array of options described in Worker::add
+	 * @param string    $serverUrl
+	 * @param bool      $onlyNodeInfo   Only use NodeInfo for server detection
+	 * @return int
+	 * @throws InternalServerErrorException
+	 */
+	public static function add($run_parameters, string $serverUrl, bool $onlyNodeInfo = false): int
+	{
+		// Dropping the worker task if the server domain is blocked
+		if (Network::isUrlBlocked($serverUrl)) {
+			return 0;
+		}
+
+		// We have to convert the Uri back to string because worker parameters are saved in JSON format which
+		// doesn't allow for structured objects.
+		return Worker::add($run_parameters, 'UpdateGServer', $serverUrl, $onlyNodeInfo);
+	}
 }
--- a/src/Worker/UpdateGServers.php
+++ b/src/Worker/UpdateGServers.php
@ -27,6 +27,7 @@ use Friendica\Database\DBA;
 use Friendica\DI;
 use Friendica\Util\DateTimeFormat;
 use Friendica\Util\Strings;
+use GuzzleHttp\Psr7\Uri;

 class UpdateGServers
 {
@ -63,12 +64,12 @@ class UpdateGServers
 			// There are duplicated "url" but not "nurl". So we check both addresses instead of just overwriting them,
 			// since that would mean loosing data.
 			if (!empty($gserver['url'])) {
-				if (Worker::add(Worker::PRIORITY_LOW, 'UpdateGServer', $gserver['url'])) {
+				if (UpdateGServer::add(Worker::PRIORITY_LOW, $gserver['url'])) {
 					$count++;
 				}
 			}
 			if (!empty($gserver['nurl']) && ($gserver['nurl'] != Strings::normaliseLink($gserver['url']))) {
-				if (Worker::add(Worker::PRIORITY_LOW, 'UpdateGServer', $gserver['nurl'])) {
+				if (UpdateGServer::add(Worker::PRIORITY_LOW, $gserver['nurl'])) {
 					$count++;
 				}
 			}
--- a/tests/src/Model/GServerTest.php
+++ b/tests/src/Model/GServerTest.php
@ -0,0 +1,76 @@
+<?php
+/**
+ * @copyright Copyright (C) 2010-2023, 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/>.
+ *
+ */
+
+namespace Friendica\Test\src\Model;
+
+use Friendica\Model\GServer;
+use GuzzleHttp\Psr7\Uri;
+use Psr\Http\Message\UriInterface;
+
+class GServerTest extends \PHPUnit\Framework\TestCase
+{
+	public function dataCleanUri(): array
+	{
+		return [
+			'full-monty' => [
+				'expected' => new Uri('https://example.com/path'),
+				'dirtyUri' => new Uri('https://user:password@example.com/path?query=string#fragment'),
+			],
+			'index.php' => [
+				'expected' => new Uri('https://example.com'),
+				'dirtyUri' => new Uri('https://example.com/index.php'),
+			],
+			'index.php-2' => [
+				'expected' => new Uri('https://example.com/path/to/resource'),
+				'dirtyUri' => new Uri('https://example.com/index.php/path/to/resource'),
+			],
+			'index.php-path' => [
+				'expected' => new Uri('https://example.com/path/to'),
+				'dirtyUri' => new Uri('https://example.com/path/to/index.php'),
+			],
+			'index.php-path-2' => [
+				'expected' => new Uri('https://example.com/path/to/path/to/resource'),
+				'dirtyUri' => new Uri('https://example.com/path/to/index.php/path/to/resource'),
+			],
+			'index.php-slash' => [
+				'expected' => new Uri('https://example.com'),
+				'dirtyUri' => new Uri('https://example.com/index.php/'),
+			],
+			'index.php-slash-2' => [
+				'expected' => new Uri('https://example.com/path/to/resource'),
+				'dirtyUri' => new Uri('https://example.com/index.php/path/to/resource/'),
+			],
+		];
+	}
+
+	/**
+	 * @dataProvider dataCleanUri
+	 *
+	 * @param UriInterface $expected
+	 * @param UriInterface $dirtyUri
+	 * @return void
+	 * @throws \Exception
+	 */
+	public function testCleanUri(UriInterface $expected, UriInterface $dirtyUri)
+	{
+		$this->assertEquals($expected, GServer::cleanUri($dirtyUri));
+	}
+}