Cache GB2312 encoding lookup, #1076

apache · Jan 11, 2025 · c07cc87 · c07cc87
1 parent 66f3e3c
commit c07cc87
Show file tree

Hide file tree

Showing 5 changed files with 16 additions and 8 deletions.
diff --git a/src/Lucene.Net.Analysis.Common/Analysis/Hunspell/Dictionary.cs b/src/Lucene.Net.Analysis.Common/Analysis/Hunspell/Dictionary.cs
@@ -746,7 +746,7 @@ private static Encoding GetSystemEncoding(string encoding) // LUCENENET: CA1822:
             }
             if ("ISO8859-14".Equals(encoding, StringComparison.OrdinalIgnoreCase))
             {
-                return new ISO8859_14Encoding();
+                return ISO8859_14Encoding.Default;
             }
             // .NET doesn't recognize the encoding without a dash between ISO and the number
             // https://msdn.microsoft.com/en-us/library/system.text.encodinginfo.getencoding(v=vs.110).aspx

diff --git a/src/Lucene.Net.Analysis.Common/Analysis/Hunspell/ISO8859_14Decoder.cs b/src/Lucene.Net.Analysis.Common/Analysis/Hunspell/ISO8859_14Decoder.cs
@@ -28,6 +28,11 @@ namespace Lucene.Net.Analysis.Hunspell
     [ExceptionToClassNameConvention]
     internal sealed class ISO8859_14Encoding : Encoding
     {
+        /// <summary>
+        /// The default singleton instance of the <see cref="ISO8859_14Encoding"/> class.
+        /// </summary>
+        public static new ISO8859_14Encoding Default { get; } = new ISO8859_14Encoding();
+
         private static readonly Decoder decoder = new ISO8859_14Decoder();
         public override Decoder GetDecoder()
         {
@@ -119,4 +124,4 @@ public override int GetChars(byte[] bytesIn, int byteIndex, int byteCount, char[
             return writeCount;
         }
     }
-}
+}
diff --git a/src/Lucene.Net.Analysis.SmartCn/Hhmm/AbstractDictionary.cs b/src/Lucene.Net.Analysis.SmartCn/Hhmm/AbstractDictionary.cs
@@ -32,14 +32,17 @@ namespace Lucene.Net.Analysis.Cn.Smart.Hhmm
     /// </summary>
     internal abstract class AbstractDictionary
     {
+        // LUCENENET specific: cached GB2312 encoding to avoid repeated calls to Encoding.GetEncoding("GB2312")
+        protected static readonly Encoding gb2312Encoding = Encoding.GetEncoding("GB2312");
+
         /// <summary>
         /// First Chinese Character in GB2312 (15 * 94)
         /// Characters in GB2312 are arranged in a grid of 94 * 94, 0-14 are unassigned or punctuation.
         /// </summary>
         public const int GB2312_FIRST_CHAR = 1410;
 
         /// <summary>
-        /// Last Chinese Character in GB2312 (87 * 94). 
+        /// Last Chinese Character in GB2312 (87 * 94).
         /// Characters in GB2312 are arranged in a grid of 94 * 94, 88-94 are unassigned.
         /// </summary>
         public const int GB2312_CHAR_NUM = 87 * 94;
@@ -98,7 +101,7 @@ public virtual string GetCCByGB2312Id(int ccid)
             try
             {
                 //String cchar = new String(buffer, "GB2312");
-                string cchar = Encoding.GetEncoding("GB2312").GetString(buffer);
+                string cchar = gb2312Encoding.GetString(buffer); // LUCENENET specific: use cached encoding instance
                 return cchar;
             }
             catch (Exception e) when (e.IsUnsupportedEncodingException()) // Encoding is not supported by the platform
@@ -117,15 +120,15 @@ public virtual short GetGB2312Id(char ch)
             try
             {
                 //byte[] buffer = Character.ToString(ch).getBytes("GB2312");
-                byte[] buffer = Encoding.GetEncoding("GB2312").GetBytes(ch.ToString());
+                byte[] buffer = gb2312Encoding.GetBytes(ch.ToString()); // LUCENENET specific: use cached encoding instance
                 //byte[] buffer = Encoding.GetEncoding("hz-gb-2312").GetBytes(ch.ToString());
                 if (buffer.Length != 2)
                 {
                     // Should be a two-byte character
                     return -1;
                 }
                 int b0 = (buffer[0] & 0x0FF) - 161; // Code starts from A1, therefore subtract 0xA1=161
-                int b1 = (buffer[1] & 0x0FF) - 161; // There is no Chinese char for the first and last symbol. 
+                int b1 = (buffer[1] & 0x0FF) - 161; // There is no Chinese char for the first and last symbol.
                                                     // Therefore, each code page only has 16*6-2=94 characters.
                 return (short)(b0 * 94 + b1);
             }

diff --git a/src/Lucene.Net.Analysis.SmartCn/Hhmm/BigramDictionary.cs b/src/Lucene.Net.Analysis.SmartCn/Hhmm/BigramDictionary.cs
@@ -302,7 +302,7 @@ public virtual void LoadFromFile(string dctFilePath)
                         byte[] lchBuffer = new byte[length];
                         dctFile.Read(lchBuffer, 0, lchBuffer.Length);
                         //tmpword = new String(lchBuffer, "GB2312");
-                        tmpword = Encoding.GetEncoding("GB2312").GetString(lchBuffer);
+                        tmpword = gb2312Encoding.GetString(lchBuffer); // LUCENENET specific: use cached encoding instance from base class
                         //tmpword = Encoding.GetEncoding("hz-gb-2312").GetString(lchBuffer);
                         if (i != 3755 + GB2312_FIRST_CHAR)
                         {

diff --git a/src/Lucene.Net.Analysis.SmartCn/Hhmm/WordDictionary.cs b/src/Lucene.Net.Analysis.SmartCn/Hhmm/WordDictionary.cs
@@ -395,7 +395,7 @@ private int LoadMainDataFromFile(string dctFilePath)
                         {
                             byte[] lchBuffer = new byte[length];
                             dctFile.Read(lchBuffer, 0, lchBuffer.Length);
-                            tmpword = Encoding.GetEncoding("GB2312").GetString(lchBuffer);
+                            tmpword = gb2312Encoding.GetString(lchBuffer); // LUCENENET specific: use cached encoding instance from base class
                             wordItem_charArrayTable[i][j] = tmpword.ToCharArray();
                         }
                         else