Optimize HttpUtility.JavaScriptStringEncode by using SearchValues (#102917)

* Optimize HttpUtility.JavaScriptStringEncode by using SearchValues for invalid JavaScript characters.

* remove duplicated test

* more suggestions

* EncodeCore and AppendSpanFormattable

* Remove check for null

* Use StackallocThreshold const

Author: TrayanZapryanov

src/libraries/System.Web.HttpUtility/src/System.Web.HttpUtility.csproj

@@ -15,6 +15,10 @@
     <Compile Include="System\Web\Util\Utf16StringValidator.cs" />
     <Compile Include="$(CommonPath)System\HexConverter.cs"
              Link="Common\System\HexConverter.cs" />
+    <Compile Include="$(CommonPath)System\Text\ValueStringBuilder.cs"
+             Link="Common\System\Text\ValueStringBuilder.cs" />
+    <Compile Include="$(CommonPath)System\Text\ValueStringBuilder.AppendSpanFormattable.cs"
+             Link="Common\System\Text\ValueStringBuilder.AppendSpanFormattable.cs" />
   </ItemGroup>
 
   <ItemGroup>

src/libraries/System.Web.HttpUtility/src/System/Web/HttpUtility.cs

@@ -237,12 +237,8 @@ public static NameValueCollection ParseQueryString(string query, Encoding encodi
         [return: NotNullIfNotNull(nameof(bytes))]
         public static byte[]? UrlDecodeToBytes(byte[]? bytes, int offset, int count) => HttpEncoder.UrlDecode(bytes, offset, count);
 
-        public static string JavaScriptStringEncode(string? value) => HttpEncoder.JavaScriptStringEncode(value);
+        public static string JavaScriptStringEncode(string? value) => HttpEncoder.JavaScriptStringEncode(value, false);
 
-        public static string JavaScriptStringEncode(string? value, bool addDoubleQuotes)
-        {
-            string encoded = HttpEncoder.JavaScriptStringEncode(value);
-            return addDoubleQuotes ? "\"" + encoded + "\"" : encoded;
-        }
+        public static string JavaScriptStringEncode(string? value, bool addDoubleQuotes) => HttpEncoder.JavaScriptStringEncode(value, addDoubleQuotes);
     }
 }

src/libraries/System.Web.HttpUtility/src/System/Web/Util/HttpEncoder.cs

@@ -20,22 +20,13 @@ internal static class HttpEncoder
         private static readonly SearchValues<byte> s_urlSafeBytes = SearchValues.Create(
             "!()*-.0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ_abcdefghijklmnopqrstuvwxyz"u8);
 
-        private static void AppendCharAsUnicodeJavaScript(StringBuilder builder, char c)
-        {
-            builder.Append($"\\u{(int)c:x4}");
-        }
-
-        private static bool CharRequiresJavaScriptEncoding(char c) =>
-            c < 0x20 // control chars always have to be encoded
-                || c == '\"' // chars which must be encoded per JSON spec
-                || c == '\\'
-                || c == '\'' // HTML-sensitive chars encoded for safety
-                || c == '<'
-                || c == '>'
-                || (c == '&')
-                || c == '\u0085' // newline chars (see Unicode 6.2, Table 5-1 [http://www.unicode.org/versions/Unicode6.2.0/ch05.pdf]) have to be encoded
-                || c == '\u2028'
-                || c == '\u2029';
+        private static readonly SearchValues<char> s_invalidJavaScriptChars = SearchValues.Create(
+            // Any Control, < 32 (' ')
+            "\u0000\u0001\u0002\u0003\u0004\u0005\u0006\u0007\u0008\u0009\u000A\u000B\u000C\u000D\u000E\u000F\u0010\u0011\u0012\u0013\u0014\u0015\u0016\u0017\u0018\u0019\u001A\u001B\u001C\u001D\u001E\u001F" +
+            // Chars which must be encoded per JSON spec / HTML-sensitive chars encoded for safety
+            "\"&'<>\\" +
+            // newline chars (see Unicode 6.2, Table 5-1 [http://www.unicode.org/versions/Unicode6.2.0/ch05.pdf]) have to be encoded
+            "\u0085\u2028\u2029");
 
         [return: NotNullIfNotNull(nameof(value))]
         internal static string? HtmlAttributeEncode(string? value)
@@ -137,79 +128,71 @@ private static int IndexOfHtmlAttributeEncodingChars(string s) =>
 
         private static bool IsNonAsciiByte(byte b) => b >= 0x7F || b < 0x20;
 
-        internal static string JavaScriptStringEncode(string? value)
+        internal static string JavaScriptStringEncode(string? value, bool addDoubleQuotes)
         {
-            if (string.IsNullOrEmpty(value))
+            int i = value.AsSpan().IndexOfAny(s_invalidJavaScriptChars);
+            if (i < 0)
             {
-                return string.Empty;
+                return addDoubleQuotes ? $"\"{value}\"" : value ?? string.Empty;
             }
 
-            StringBuilder? b = null;
-            int startIndex = 0;
-            int count = 0;
-            for (int i = 0; i < value.Length; i++)
-            {
-                char c = value[i];
+            return EncodeCore(value, i, addDoubleQuotes);
 
-                // Append the unhandled characters (that do not require special treament)
-                // to the string builder when special characters are detected.
-                if (CharRequiresJavaScriptEncoding(c))
+            static string EncodeCore(ReadOnlySpan<char> value, int i, bool addDoubleQuotes)
+            {
+                var vsb = new ValueStringBuilder(stackalloc char[StackallocThreshold]);
+                if (addDoubleQuotes)
                 {
-                    b ??= new StringBuilder(value.Length + 5);
-
-                    if (count > 0)
-                    {
-                        b.Append(value, startIndex, count);
-                    }
-
-                    startIndex = i + 1;
-                    count = 0;
+                    vsb.Append('"');
+                }
 
+                ReadOnlySpan<char> chars = value;
+                do
+                {
+                    vsb.Append(chars.Slice(0, i));
+                    char c = chars[i];
+                    chars = chars.Slice(i + 1);
                     switch (c)
                     {
                         case '\r':
-                            b.Append("\\r");
+                            vsb.Append("\\r");
                             break;
                         case '\t':
-                            b.Append("\\t");
+                            vsb.Append("\\t");
                             break;
                         case '\"':
-                            b.Append("\\\"");
+                            vsb.Append("\\\"");
                             break;
                         case '\\':
-                            b.Append("\\\\");
+                            vsb.Append("\\\\");
                             break;
                         case '\n':
-                            b.Append("\\n");
+                            vsb.Append("\\n");
                             break;
                         case '\b':
-                            b.Append("\\b");
+                            vsb.Append("\\b");
                             break;
                         case '\f':
-                            b.Append("\\f");
+                            vsb.Append("\\f");
                             break;
                         default:
-                            AppendCharAsUnicodeJavaScript(b, c);
+                            vsb.Append("\\u");
+                            vsb.AppendSpanFormattable((int)c, "x4");
                             break;
                     }
-                }
-                else
+
+                    i = chars.IndexOfAny(s_invalidJavaScriptChars);
+                } while (i >= 0);
+
+                vsb.Append(chars);
+
+                if (addDoubleQuotes)
                 {
-                    count++;
+                    vsb.Append('"');
                 }
-            }
 
-            if (b == null)
-            {
-                return value;
-            }
-
-            if (count > 0)
-            {
-                b.Append(value, startIndex, count);
+                return vsb.ToString();
             }
-
-            return b.ToString();
         }
 
         [return: NotNullIfNotNull(nameof(bytes))]

src/libraries/System.Web.HttpUtility/tests/HttpUtility/HttpUtilityTest.cs

@@ -310,6 +310,7 @@ public static IEnumerable<object[]> JavaScriptStringEncodeData
                 yield return new object[] { "", "" };
                 yield return new object[] {"No escaping needed.", "No escaping needed."};
                 yield return new object[] {"The \t and \n will need to be escaped.", "The \\t and \\n will need to be escaped."};
+                yield return new object[] {"The \t and \n will need to be escaped.>", "The \\t and \\n will need to be escaped.\\u003e" };
                 for (char c = char.MinValue; c < TestMaxChar; c++)
                 {
                     if (c >= 0 && c <= 7 || c == 11 || c >= 14 && c <= 31 || c == 38 || c == 39 || c == 60 || c == 62 || c == 133 || c == 8232 || c == 8233)