beagled/Lucene.Net/Analysis/DE/GermanStemFilter.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.DE
  21 {
  22
  23         /// <summary> A filter that stems German words. It supports a table of words that should
  24         /// not be stemmed at all. The stemmer used can be changed at runtime after the
  25         /// filter object is created (as long as it is a GermanStemmer).
  26         ///
  27         /// </summary>
  28         /// <author>     Gerhard Schwarz
  29         /// </author>
  30         /// <version>    $Id: GermanStemFilter.cs,v 1.2 2005/01/17 19:54:27 joeshaw Exp $
  31         /// </version>
  32         public sealed class GermanStemFilter : TokenFilter
  33         {
  34                 /// <summary> The actual token in the input stream.</summary>
  35                 private Token token = null;
  36                 private GermanStemmer stemmer = null;
  37                 private System.Collections.Hashtable exclusionSet = null;
  38
  39                 public GermanStemFilter(TokenStream in_Renamed) : base(in_Renamed)
  40                 {
  41                         stemmer = new GermanStemmer();
  42                 }
  43
  44                 /// <summary> Builds a GermanStemFilter that uses an exclusiontable.</summary>
  45                 /// <deprecated> Use {@link #GermanStemFilter(Lucene.Net.Analysis.TokenStream, java.util.Set)} instead.
  46                 /// </deprecated>
  47                 public GermanStemFilter(TokenStream in_Renamed, System.Collections.Hashtable exclusiontable):this(in_Renamed)
  48                 {
  49                         exclusionSet = new System.Collections.Hashtable(new System.Collections.Hashtable(exclusiontable));
  50                 }
  51
  52                 /// <returns>  Returns the next token in the stream, or null at EOS
  53                 /// </returns>
  54                 public override Token Next()
  55                 {
  56                         if ((token = input.Next()) == null)
  57                         {
  58                                 return null;
  59                         }
  60                         // Check the exclusiontable
  61                         else if (exclusionSet != null && exclusionSet.Contains(token.TermText()))
  62                         {
  63                                 return token;
  64                         }
  65                         else
  66                         {
  67                                 System.String s = stemmer.Stem(token.TermText());
  68                                 // If not stemmed, dont waste the time creating a new token
  69                                 if (!s.Equals(token.TermText()))
  70                                 {
  71                                         return new Token(s, token.StartOffset(), token.EndOffset(), token.Type());
  72                                 }
  73                                 return token;
  74                         }
  75                 }
  76
  77                 /// <summary> Set a alternative/custom GermanStemmer for this filter.</summary>
  78                 public void  SetStemmer(GermanStemmer stemmer)
  79                 {
  80                         if (stemmer != null)
  81                         {
  82                                 this.stemmer = stemmer;
  83                         }
  84                 }
  85
  86                 /// <summary> Set an alternative exclusion list for this filter.</summary>
  87                 /// <deprecated> Use {@link #SetExclusionSet(java.util.Set)} instead.
  88                 /// </deprecated>
  89                 public void  SetExclusionTable(System.Collections.Hashtable exclusiontable)
  90                 {
  91                         exclusionSet = new System.Collections.Hashtable(new System.Collections.Hashtable(exclusiontable));
  92                 }
  93
  94                 /// <summary> Set an alternative exclusion list for this filter.</summary>
  95                 public void  SetExclusionSet(System.Collections.Hashtable exclusionSet)
  96                 {
  97                         this.exclusionSet = exclusionSet;
  98                 }
  99         }
 100 }