From: Marcel Werk <burntime@woltlab.com>
Date: Wed, 9 Jul 2014 10:36:32 +0000 (+0200)
Subject: Improved detection of utf-8 encoded strings
X-Git-Tag: 2.1.0_Alpha_1~601
X-Git-Url: https://git.stricted.de/?a=commitdiff_plain;h=a92ec792d48b2a9c7290ef517c1054932f9e47dd;p=GitHub%2FWoltLab%2FWCF.git

Improved detection of utf-8 encoded strings
---

diff --git a/wcfsetup/install/files/lib/system/WCF.class.php b/wcfsetup/install/files/lib/system/WCF.class.php
index e40feb8647..1afab98fce 100644
--- a/wcfsetup/install/files/lib/system/WCF.class.php
+++ b/wcfsetup/install/files/lib/system/WCF.class.php
@@ -746,7 +746,7 @@ class WCF {
 		}
 		
 		$path = str_replace('/index.php', '', str_replace($scriptName, '', $_SERVER['REQUEST_URI']));
-		if (!StringUtil::isASCII($path) && !StringUtil::isUTF8($path)) {
+		if (!StringUtil::isUTF8($path)) {
 			$path = StringUtil::convertEncoding('ISO-8859-1', 'UTF-8', $path);
 		}
 		$path = FileUtil::removeLeadingSlash($path);
diff --git a/wcfsetup/install/files/lib/system/bbcode/KeywordHighlighter.class.php b/wcfsetup/install/files/lib/system/bbcode/KeywordHighlighter.class.php
index 7b62166727..691e65d6d0 100644
--- a/wcfsetup/install/files/lib/system/bbcode/KeywordHighlighter.class.php
+++ b/wcfsetup/install/files/lib/system/bbcode/KeywordHighlighter.class.php
@@ -95,7 +95,7 @@ class KeywordHighlighter extends SingletonFactory {
 	 */
 	protected function parseKeywords($keywordString) {
 		// convert encoding if necessary
-		if (!StringUtil::isASCII($keywordString) && !StringUtil::isUTF8($keywordString)) {
+		if (!StringUtil::isUTF8($keywordString)) {
 			$keywordString = StringUtil::convertEncoding('ISO-8859-1', 'UTF-8', $keywordString);
 		}
 		
diff --git a/wcfsetup/install/files/lib/util/StringUtil.class.php b/wcfsetup/install/files/lib/util/StringUtil.class.php
index a51a41fa88..f1f4a6061c 100644
--- a/wcfsetup/install/files/lib/util/StringUtil.class.php
+++ b/wcfsetup/install/files/lib/util/StringUtil.class.php
@@ -524,7 +524,7 @@ final class StringUtil {
 	 * @return	boolean
 	 */
 	public static function isUTF8($string) {
-		/*return preg_match('/^(
+		return preg_match('/^(
 				[\x09\x0A\x0D\x20-\x7E]*		# ASCII
 			|	[\xC2-\xDF][\x80-\xBF]			# non-overlong 2-byte
 			|	\xE0[\xA0-\xBF][\x80-\xBF]		# excluding overlongs
@@ -534,16 +534,6 @@ final class StringUtil {
 			|	[\xF1-\xF3][\x80-\xBF]{3}		# planes 4-15
 			|	\xF4[\x80-\x8F][\x80-\xBF]{2}		# plane 16
 			)*$/x', $string);
-		*/
-		return preg_match('/(
-				[\xC2-\xDF][\x80-\xBF]			# non-overlong 2-byte
-			|	\xE0[\xA0-\xBF][\x80-\xBF]		# excluding overlongs
-			|	[\xE1-\xEC\xEE\xEF][\x80-\xBF]{2}	# straight 3-byte
-			|	\xED[\x80-\x9F][\x80-\xBF]		# excluding surrogates
-			|	\xF0[\x90-\xBF][\x80-\xBF]{2}		# planes 1-3
-			|	[\xF1-\xF3][\x80-\xBF]{3}		# planes 4-15
-			|	\xF4[\x80-\x8F][\x80-\xBF]{2}		# plane 16
-			)/x', $string);
 	}
 	
 	/**
diff --git a/wcfsetup/install/files/lib/util/UserUtil.class.php b/wcfsetup/install/files/lib/util/UserUtil.class.php
index 6553ea887d..b4aaa225a8 100644
--- a/wcfsetup/install/files/lib/util/UserUtil.class.php
+++ b/wcfsetup/install/files/lib/util/UserUtil.class.php
@@ -107,7 +107,7 @@ final class UserUtil {
 	public static function getUserAgent() {
 		if (isset($_SERVER['HTTP_USER_AGENT'])) {
 			$userAgent = $_SERVER['HTTP_USER_AGENT'];
-			if (!StringUtil::isASCII($userAgent) && !StringUtil::isUTF8($userAgent)) {
+			if (!StringUtil::isUTF8($userAgent)) {
 				$userAgent = StringUtil::convertEncoding('ISO-8859-1', 'UTF-8', $userAgent);
 			}
 			
@@ -263,7 +263,7 @@ final class UserUtil {
 		}
 		
 		// fix encoding
-		if (!StringUtil::isASCII($REQUEST_URI) && !StringUtil::isUTF8($REQUEST_URI)) {
+		if (!StringUtil::isUTF8($REQUEST_URI)) {
 			$REQUEST_URI = StringUtil::convertEncoding('ISO-8859-1', 'UTF-8', $REQUEST_URI);
 		}