Improve string encoding times.

Cache a reference to Encoding.UTF8 - the property access is (rather surprisingly) significant. Additionally, when we detect that the string is all ASCII (due to the computed length in bytes being the length in characters), we can perform the encoding very efficiently ourselves.
jskeet · Jun 11, 2015 · 35e4dbd · 35e4dbd
1 parent 954e720
commit 35e4dbd
Show file tree

Hide file tree

Showing 2 changed files with 17 additions and 5 deletions.
diff --git a/csharp/src/ProtocolBuffers/CodedOutputStream.ComputeSize.cs b/csharp/src/ProtocolBuffers/CodedOutputStream.ComputeSize.cs
@@ -135,7 +135,7 @@ public static int ComputeBoolSize(int fieldNumber, bool value)
         /// </summary>
         public static int ComputeStringSize(int fieldNumber, String value)
         {
-            int byteArraySize = Encoding.UTF8.GetByteCount(value);
+            int byteArraySize = UTF8.GetByteCount(value);
             return ComputeTagSize(fieldNumber) +
                    ComputeRawVarint32Size((uint) byteArraySize) +
                    byteArraySize;
@@ -323,7 +323,7 @@ public static int ComputeBoolSizeNoTag(bool value)
         /// </summary>
         public static int ComputeStringSizeNoTag(String value)
         {
-            int byteArraySize = Encoding.UTF8.GetByteCount(value);
+            int byteArraySize = UTF8.GetByteCount(value);
             return ComputeRawVarint32Size((uint) byteArraySize) +
                    byteArraySize;
         }

diff --git a/csharp/src/ProtocolBuffers/CodedOutputStream.cs b/csharp/src/ProtocolBuffers/CodedOutputStream.cs
@@ -58,6 +58,8 @@ namespace Google.Protobuf
     /// </remarks>
     public sealed partial class CodedOutputStream : ICodedOutputStream
     {
+        private static readonly Encoding UTF8 = Encoding.UTF8;
+
         /// <summary>
         /// The buffer size used by CreateInstance(Stream).
         /// </summary>
@@ -294,16 +296,26 @@ public void WriteString(int fieldNumber, string fieldName, string value)
             WriteTag(fieldNumber, WireFormat.WireType.LengthDelimited);
             // Optimise the case where we have enough space to write
             // the string directly to the buffer, which should be common.
-            int length = Encoding.UTF8.GetByteCount(value);
+            int length = UTF8.GetByteCount(value);
             WriteRawVarint32((uint) length);
             if (limit - position >= length)
             {
-                Encoding.UTF8.GetBytes(value, 0, value.Length, buffer, position);
+                if (length == value.Length) // Must be all ASCII...
+                {
+                                for (int i = 0; i < length; i++)
+                                {
+                        buffer[position + i] = (byte)value[i];
+                    }
+                }
+                else
+                {
+                    UTF8.GetBytes(value, 0, value.Length, buffer, position);
+                }
                 position += length;
             }
             else
             {
-                byte[] bytes = Encoding.UTF8.GetBytes(value);
+                byte[] bytes = UTF8.GetBytes(value);
                 WriteRawBytes(bytes);
             }
         }