beagled/Lucene.Net/Analysis/RU/RussianStemFilter.cs

   1 /*
   2  * Copyright 2004 The Apache Software Foundation
   3  *
   4  * Licensed under the Apache License, Version 2.0 (the "License");
   5  * you may not use this file except in compliance with the License.
   6  * You may obtain a copy of the License at
   7  *
   8  * http://www.apache.org/licenses/LICENSE-2.0
   9  *
  10  * Unless required by applicable law or agreed to in writing, software
  11  * distributed under the License is distributed on an "AS IS" BASIS,
  12  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  13  * See the License for the specific language governing permissions and
  14  * limitations under the License.
  15  */
  16 using System;
  17 using Token = Lucene.Net.Analysis.Token;
  18 using TokenFilter = Lucene.Net.Analysis.TokenFilter;
  19 using TokenStream = Lucene.Net.Analysis.TokenStream;
  20 namespace Lucene.Net.Analysis.RU
  21 {
  22
  23         /// <summary> A filter that stems Russian words. The implementation was inspired by GermanStemFilter.
  24         /// The input should be filtered by RussianLowerCaseFilter before passing it to RussianStemFilter ,
  25         /// because RussianStemFilter only works  with lowercase part of any "russian" charset.
  26         ///
  27         /// </summary>
  28         /// <author>     Boris Okner, b.okner@rogers.com
  29         /// </author>
  30         /// <version>    $Id: RussianStemFilter.cs,v 1.2 2005/01/17 19:54:28 joeshaw Exp $
  31         /// </version>
  32         public sealed class RussianStemFilter : TokenFilter
  33         {
  34                 /// <summary> The actual token in the input stream.</summary>
  35                 private Token token = null;
  36                 private RussianStemmer stemmer = null;
  37
  38                 public RussianStemFilter(TokenStream in_Renamed, char[] charset):base(in_Renamed)
  39                 {
  40                         stemmer = new RussianStemmer(charset);
  41                 }
  42
  43                 /// <returns>  Returns the next token in the stream, or null at EOS
  44                 /// </returns>
  45                 public override Token Next()
  46                 {
  47                         if ((token = input.Next()) == null)
  48                         {
  49                                 return null;
  50                         }
  51                         else
  52                         {
  53                                 System.String s = stemmer.Stem(token.TermText());
  54                                 if (!s.Equals(token.TermText()))
  55                                 {
  56                                         return new Token(s, token.StartOffset(), token.EndOffset(), token.Type());
  57                                 }
  58                                 return token;
  59                         }
  60                 }
  61
  62                 /// <summary> Set a alternative/custom RussianStemmer for this filter.</summary>
  63                 public void  SetStemmer(RussianStemmer stemmer)
  64                 {
  65                         if (stemmer != null)
  66                         {
  67                                 this.stemmer = stemmer;
  68                         }
  69                 }
  70         }
  71 }