X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;ds=sidebyside;f=src%2FNetwork%2FProbe.php;h=5872ae587b53fe84efd4248699a98a2d9cf5c064;hb=e27915a819397f391b5c50b5bf15e2475eb084bf;hp=b547c430564e9977e69e2f13e88faa8869bab271;hpb=fc9865176e4358c12a41370768632f0a8db3e655;p=friendica.git

diff --git a/src/Network/Probe.php b/src/Network/Probe.php
index b547c43056..5872ae587b 100644
--- a/src/Network/Probe.php
+++ b/src/Network/Probe.php
@@ -26,9 +26,11 @@ use DomXPath;
 use Friendica\Core\Cache\Duration;
 use Friendica\Core\Logger;
 use Friendica\Core\Protocol;
+use Friendica\Core\System;
 use Friendica\Database\DBA;
 use Friendica\DI;
 use Friendica\Model\Contact;
+use Friendica\Model\GServer;
 use Friendica\Model\Profile;
 use Friendica\Protocol\ActivityNamespace;
 use Friendica\Protocol\ActivityPub;
@@ -47,6 +49,31 @@ class Probe
 	private static $baseurl;
 	private static $istimeout;
 
+	/**
+	 * Remove stuff from an URI that doesn't belong there
+	 *
+	 * @param string $URI
+	 * @return string Cleaned URI
+	 */
+	public static function cleanURI(string $URI)
+	{
+		// At first remove leading and trailing junk
+		$URI = trim($URI, "@#?:/ \t\n\r\0\x0B");
+
+		$parts = parse_url($URI);
+
+		if (empty($parts['scheme'])) {
+			return $URI;
+		}
+
+		// Remove the URL fragment, since these shouldn't be part of any profile URL
+		unset($parts['fragment']);
+
+		$URI = Network::unparseURL($parts);
+
+		return $URI;
+	}
+
 	/**
 	 * Rearrange the array so that it always has the same order
 	 *
@@ -60,14 +87,16 @@ class Probe
 				"community", "keywords", "location", "about", "hide",
 				"batch", "notify", "poll", "request", "confirm", "poco",
 				"following", "followers", "inbox", "outbox", "sharedinbox",
-				"priority", "network", "pubkey", "baseurl"];
+				"priority", "network", "pubkey", "baseurl", "gsid"];
 
 		$newdata = [];
 		foreach ($fields as $field) {
 			if (isset($data[$field])) {
 				$newdata[$field] = $data[$field];
-			} else {
+			} elseif ($field != "gsid") {
 				$newdata[$field] = "";
+			} else {
+				$newdata[$field] = null;
 			}
 		}
 
@@ -116,27 +145,37 @@ class Probe
 		// Reset the static variable
 		self::$baseurl = '';
 
-		$ssl_url = "https://".$host."/.well-known/host-meta";
-		$url = "http://".$host."/.well-known/host-meta";
+		// Handles the case when the hostname contains the scheme
+		if (!parse_url($host, PHP_URL_SCHEME)) {
+			$ssl_url = "https://" . $host . "/.well-known/host-meta";
+			$url = "http://" . $host . "/.well-known/host-meta";
+		} else {
+			$ssl_url = $host . "/.well-known/host-meta";
+			$url = '';
+		}
 
 		$xrd_timeout = DI::config()->get('system', 'xrd_timeout', 20);
 
-		Logger::log("Probing for ".$host, Logger::DEBUG);
+		Logger::info('Probing', ['host' => $host, 'ssl_url' => $ssl_url, 'url' => $url, 'callstack' => System::callstack(20)]);
 		$xrd = null;
 
 		$curlResult = Network::curl($ssl_url, false, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']);
 		$ssl_connection_error = ($curlResult->getErrorNumber() == CURLE_COULDNT_CONNECT) || ($curlResult->getReturnCode() == 0);
 		if ($curlResult->isSuccess()) {
 			$xml = $curlResult->getBody();
-			$xrd = XML::parseString($xml, false);
-			$host_url = 'https://'.$host;
+			$xrd = XML::parseString($xml, true);
+			if (!empty($url)) {
+				$host_url = 'https://' . $host;
+			} else {
+				$host_url = $host;
+			}
 		} elseif ($curlResult->isTimeout()) {
 			Logger::info('Probing timeout', ['url' => $ssl_url], Logger::DEBUG);
 			self::$istimeout = true;
 			return false;
 		}
 
-		if (!is_object($xrd)) {
+		if (!is_object($xrd) && !empty($url)) {
 			$curlResult = Network::curl($url, false, ['timeout' => $xrd_timeout, 'accept_content' => 'application/xrd+xml']);
 			$connection_error = ($curlResult->getErrorNumber() == CURLE_COULDNT_CONNECT) || ($curlResult->getReturnCode() == 0);
 			if ($curlResult->isTimeout()) {
@@ -149,7 +188,7 @@ class Probe
 			}
 
 			$xml = $curlResult->getBody();
-			$xrd = XML::parseString($xml, false);
+			$xrd = XML::parseString($xml, true);
 			$host_url = 'http://'.$host;
 		}
 		if (!is_object($xrd)) {
@@ -256,10 +295,6 @@ class Probe
 	/**
 	 * Check an URI for LRDD data
 	 *
-	 * this is a replacement for the "lrdd" function.
-	 * It isn't used in this class and has some redundancies in the code.
-	 * When time comes we can check the existing calls for "lrdd" if we can rework them.
-	 *
 	 * @param string $uri Address that should be probed
 	 *
 	 * @return array uri data
@@ -280,7 +315,7 @@ class Probe
 				return [];
 			}
 
-			$host = $parts["host"];
+			$host = $parts['scheme'] . '://' . $parts["host"];
 			if (!empty($parts["port"])) {
 				$host .= ':'.$parts["port"];
 			}
@@ -338,7 +373,7 @@ class Probe
 			$data[] = ["@attributes" => $link];
 		}
 
-		if (is_array($webfinger["aliases"])) {
+		if (!empty($webfinger["aliases"]) && is_array($webfinger["aliases"])) {
 			foreach ($webfinger["aliases"] as $alias) {
 				$data[] = ["@attributes" =>
 							["rel" => "alias",
@@ -429,6 +464,10 @@ class Probe
 			$data['baseurl'] = self::$baseurl;
 		}
 
+		if (!empty($data['baseurl']) && empty($data['gsid'])) {
+			$data['gsid'] = GServer::getID($data['baseurl']);
+		}
+
 		if (empty($data['network'])) {
 			$data['network'] = Protocol::PHANTOM;
 		}
@@ -904,7 +943,7 @@ class Probe
 		}
 
 		// If it is not JSON, maybe it is XML
-		$xrd = XML::parseString($data, false);
+		$xrd = XML::parseString($data, true);
 		if (!is_object($xrd)) {
 			Logger::log("No webfinger data retrievable for ".$url, Logger::DEBUG);
 			return false;