X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=src%2FUtil%2FNetwork.php;h=30c4798a5a985dc1db09d4545a04f73c7009d9d5;hb=180b81c6e74bdc8d677f447fe796fb10783c96dd;hp=2cad22acf83791bd7e2d8dad7a480f10175bdaa1;hpb=8db0e090d7afc3fbfeaf7a587140c80c98233484;p=friendica.git

diff --git a/src/Util/Network.php b/src/Util/Network.php
index 2cad22acf8..30c4798a5a 100644
--- a/src/Util/Network.php
+++ b/src/Util/Network.php
@@ -1,6 +1,6 @@
 <?php
 /**
- * @copyright Copyright (C) 2010-2022, the Friendica project
+ * @copyright Copyright (C) 2010-2023, the Friendica project
  *
  * @license GNU AGPL version 3 or any later version
  *
@@ -25,8 +25,11 @@ use Friendica\Core\Hook;
 use Friendica\Core\Logger;
 use Friendica\DI;
 use Friendica\Model\Contact;
+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
 {
@@ -48,6 +51,7 @@ class Network
 	 * and check DNS to see if it's real (or check if is a valid IP address)
 	 *
 	 * @param string $url The URL to be validated
+	 *
 	 * @return string|boolean The actual working URL, false else
 	 * @throws \Friendica\Network\HTTPException\InternalServerErrorException
 	 */
@@ -66,14 +70,31 @@ class Network
 			$url = 'http://' . $url;
 		}
 
-		/// @TODO Really suppress function outcomes? Why not find them + debug them?
-		$h = @parse_url($url);
+		$xrd_timeout = DI::config()->get('system', 'xrd_timeout');
+		$host = parse_url($url, PHP_URL_HOST);
 
-		if (!empty($h['host']) && (@dns_get_record($h['host'], DNS_A + DNS_CNAME) || filter_var($h['host'], FILTER_VALIDATE_IP))) {
-			return $url;
+		if (empty($host) || !(filter_var($host, FILTER_VALIDATE_IP) || @dns_get_record($host . '.', DNS_A + DNS_AAAA))) {
+			return false;
 		}
 
-		return false;
+		if (in_array(parse_url($url, PHP_URL_SCHEME), ['https', 'http'])) {
+			$options = [HttpClientOptions::VERIFY => true, HttpClientOptions::TIMEOUT => $xrd_timeout];
+			$curlResult = DI::httpClient()->head($url, $options);
+
+			// Workaround for systems that can't handle a HEAD request. Don't retry on timeouts.
+			if (!$curlResult->isSuccess() && ($curlResult->getReturnCode() >= 400) && !in_array($curlResult->getReturnCode(), [408, 504])) {
+				$curlResult = DI::httpClient()->get($url, HttpClientAccept::DEFAULT, $options);
+			}
+
+			if (!$curlResult->isSuccess()) {
+				Logger::notice('Url not reachable', ['host' => $host, 'url' => $url]);
+				return false;
+			} elseif ($curlResult->isRedirectUrl()) {
+				$url = $curlResult->getRedirectUrl();
+			}
+		}
+
+		return $url;
 	}
 
 	/**
@@ -95,7 +116,7 @@ class Network
 		$h = substr($addr, strpos($addr, '@') + 1);
 
 		// Concerning the @ see here: https://stackoverflow.com/questions/36280957/dns-get-record-a-temporary-server-error-occurred
-		if ($h && (@dns_get_record($h, DNS_A + DNS_MX) || filter_var($h, FILTER_VALIDATE_IP))) {
+		if ($h && (@dns_get_record($h, DNS_A + DNS_AAAA + DNS_MX) || filter_var($h, FILTER_VALIDATE_IP))) {
 			return true;
 		}
 		if ($h && @dns_get_record($h, DNS_CNAME + DNS_MX)) {
@@ -157,11 +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;
 		}
 
@@ -171,7 +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;
 			}
 		}
@@ -240,6 +278,7 @@ class Network
 	 *
 	 * @param string $domain
 	 * @param array  $domain_list
+	 *
 	 * @return boolean
 	 */
 	public static function isDomainAllowed(string $domain, array $domain_list): bool
@@ -278,6 +317,7 @@ class Network
 	 * Remove Google Analytics and other tracking platforms params from URL
 	 *
 	 * @param string $url Any user-submitted URL that may contain tracking params
+	 *
 	 * @return string The same URL stripped of tracking parameters
 	 */
 	public static function stripTrackingQueryParams(string $url): string
@@ -331,10 +371,12 @@ class Network
 	 *
 	 * @param string $url
 	 * @param string $basepath
+	 *
 	 * @return string url
 	 */
-	public static function addBasePath(string $url, string $basepath)
+	public static function addBasePath(string $url, string $basepath): string
 	{
+		$url = trim($url);
 		if (!empty(parse_url($url, PHP_URL_SCHEME)) || empty(parse_url($basepath, PHP_URL_SCHEME)) || empty($url) || empty(parse_url($url))) {
 			return $url;
 		}
@@ -353,6 +395,7 @@ class Network
 	 *
 	 * @param string $url1
 	 * @param string $url2
+	 *
 	 * @return string The matching part or empty string on error
 	 */
 	public static function getUrlMatch(string $url1, string $url2): string
@@ -440,6 +483,7 @@ class Network
 	 * Glue url parts together
 	 *
 	 * @param array $parsed URL parts
+	 *
 	 * @return string|null The glued URL or null on error
 	 * @deprecated since version 2021.12, use GuzzleHttp\Psr7\Uri::fromParts($parts) instead
 	 */
@@ -456,21 +500,22 @@ class Network
 		$scheme    = $get('scheme');
 		$query     = $get('query');
 		$fragment  = $get('fragment');
-		$authority = ($userinfo !== null ? $userinfo . '@' : '') . 
+		$authority = ($userinfo !== null ? $userinfo . '@' : '') .
 						$get('host') .
 						($port ? ":$port" : '');
 
-		return	(strlen($scheme) ? $scheme . ':' : '') .
-			(strlen($authority) ? '//' . $authority : '') .
+		return	(!empty($scheme) ? $scheme . ':' : '') .
+			(!empty($authority) ? '//' . $authority : '') .
 			$get('path') .
-			(strlen($query) ? '?' . $query : '') .
-			(strlen($fragment) ? '#' . $fragment : '');
+			(!empty($query) ? '?' . $query : '') .
+			(!empty($fragment) ? '#' . $fragment : '');
 	}
 
 	/**
 	 * Convert an URI to an IDN compatible URI
 	 *
 	 * @param string $uri
+	 *
 	 * @return string
 	 */
 	public static function convertToIdn(string $uri): string
@@ -478,7 +523,7 @@ class Network
 		$parts = parse_url($uri);
 		if (!empty($parts['scheme']) && !empty($parts['host'])) {
 			$parts['host'] = idn_to_ascii($parts['host']);
-			$uri = Uri::fromParts($parts);
+			$uri = (string)Uri::fromParts($parts);
 		} else {
 			$parts = explode('@', $uri);
 			if (count($parts) == 2) {
@@ -495,6 +540,7 @@ class Network
 	 * Switch the scheme of an url between http and https
 	 *
 	 * @param string $url
+	 *
 	 * @return string Switched URL
 	 */
 	public static function switchScheme(string $url): string
@@ -518,6 +564,7 @@ class Network
 	 *
 	 * @param string $path
 	 * @param array  $additionalParams Associative array of parameters
+	 *
 	 * @return string
 	 */
 	public static function appendQueryParam(string $path, array $additionalParams): string
@@ -544,6 +591,7 @@ class Network
 	 *
 	 * @param string $etag          The page etag
 	 * @param string $last_modified The page last modification UTC date
+	 *
 	 * @return void
 	 * @throws \Exception
 	 */
@@ -582,6 +630,7 @@ class Network
 	 * Check if the given URL is a local link
 	 *
 	 * @param string $url
+	 *
 	 * @return bool
 	 */
 	public static function isLocalLink(string $url): bool