From 45ff7cec805755088dee3dc17f158e89f79bfeb5 Mon Sep 17 00:00:00 2001
From: UbitUmarov
Date: Wed, 28 Oct 2015 21:27:56 +0000
Subject:  fix cut points of UTF-8 strings

---
 OpenSim/Framework/Util.cs | 38 ++++++++++++++++++++++++++++----------
 1 file changed, 28 insertions(+), 10 deletions(-)

(limited to 'OpenSim/Framework')

diff --git a/OpenSim/Framework/Util.cs b/OpenSim/Framework/Util.cs
index d1cdaa6..3e6d8ef 100644
--- a/OpenSim/Framework/Util.cs
+++ b/OpenSim/Framework/Util.cs
@@ -1945,17 +1945,26 @@ namespace OpenSim.Framework
         /// <returns></returns>
         public static byte[] StringToBytes256(string str)
         {
-            if (String.IsNullOrEmpty(str)) { return Utils.EmptyBytes; }
-            if (str.Length > 254) str = str.Remove(254);
-            if (!str.EndsWith("\0")) { str += "\0"; }
+            if (String.IsNullOrEmpty(str))
+                return Utils.EmptyBytes;
+
+            if (!str.EndsWith("\0"))
+                str += "\0";
             
             // Because this is UTF-8 encoding and not ASCII, it's possible we
             // might have gotten an oversized array even after the string trim
             byte[] data = UTF8.GetBytes(str);
+
             if (data.Length > 256)
             {
-                Array.Resize<byte>(ref data, 256);
-                data[255] = 0;
+                int cut = 255;
+                if((data[cut] & 0x80 ) != 0 )
+                    {
+                    while(cut > 0 && (data[cut] & 0xc0) != 0xc0)
+                        cut--;
+                    }
+                Array.Resize<byte>(ref data, cut + 1);
+                data[cut] = 0;
             }
 
             return data;
@@ -1987,17 +1996,26 @@ namespace OpenSim.Framework
         /// <returns></returns>
         public static byte[] StringToBytes1024(string str)
         {
-            if (String.IsNullOrEmpty(str)) { return Utils.EmptyBytes; }
-            if (str.Length > 1023) str = str.Remove(1023);
-            if (!str.EndsWith("\0")) { str += "\0"; }
+            if (String.IsNullOrEmpty(str))
+                return Utils.EmptyBytes;
+
+            if (!str.EndsWith("\0"))
+                 str += "\0";
 
             // Because this is UTF-8 encoding and not ASCII, it's possible we
             // might have gotten an oversized array even after the string trim
             byte[] data = UTF8.GetBytes(str);
+
             if (data.Length > 1024)
             {
-                Array.Resize<byte>(ref data, 1024);
-                data[1023] = 0;
+                int cut = 1023;
+                if((data[cut] & 0x80 ) != 0 )
+                    {
+                    while(cut > 0 && (data[cut] & 0xc0) != 0xc0)
+                        cut--;
+                    }
+                Array.Resize<byte>(ref data, cut + 1);
+                data[cut] = 0;
             }
 
             return data;
-- 
cgit v1.1