Lucene.Net/Analysis/RU/RussianLowerCaseFilter.cs

   1 using System;
   2 using Lucene.Net.Analysis;
   3
   4 namespace Lucene.Net.Analysis.Ru
   5 {
   6         /* ====================================================================
   7          * The Apache Software License, Version 1.1
   8          *
   9          * Copyright (c) 2001 The Apache Software Foundation.  All rights
  10          * reserved.
  11          *
  12          * Redistribution and use in source and binary forms, with or without
  13          * modification, are permitted provided that the following conditions
  14          * are met:
  15          *
  16          * 1. Redistributions of source code must retain the above copyright
  17          *    notice, this list of conditions and the following disclaimer.
  18          *
  19          * 2. Redistributions in binary form must reproduce the above copyright
  20          *    notice, this list of conditions and the following disclaimer in
  21          *    the documentation and/or other materials provided with the
  22          *    distribution.
  23          *
  24          * 3. The end-user documentation included with the redistribution,
  25          *    if any, must include the following acknowledgment:
  26          *       "This product includes software developed by the
  27          *        Apache Software Foundation (http://www.apache.org/)."
  28          *    Alternately, this acknowledgment may appear in the software itself,
  29          *    if and wherever such third-party acknowledgments normally appear.
  30          *
  31          * 4. The names "Apache" and "Apache Software Foundation" and
  32          *    "Apache Lucene" must not be used to endorse or promote products
  33          *    derived from this software without prior written permission. For
  34          *    written permission, please contact apache@apache.org.
  35          *
  36          * 5. Products derived from this software may not be called "Apache",
  37          *    "Apache Lucene", nor may "Apache" appear in their name, without
  38          *    prior written permission of the Apache Software Foundation.
  39          *
  40          * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
  41          * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
  42          * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
  43          * DISCLAIMED.  IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
  44          * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
  45          * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
  46          * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
  47          * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
  48          * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
  49          * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
  50          * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  51          * SUCH DAMAGE.
  52          * ====================================================================
  53          *
  54          * This software consists of voluntary contributions made by many
  55          * individuals on behalf of the Apache Software Foundation.  For more
  56          * information on the Apache Software Foundation, please see
  57          * <http://www.apache.org/>.
  58          */
  59
  60         /// <summary>
  61         /// Normalizes token text to lower case, analyzing given ("russian") charset.
  62         /// </summary>
  63         /// <author>Boris Okner, b.okner@rogers.com</author>
  64         /// <version>$Id: RussianLowerCaseFilter.cs,v 1.1.1.1 2004/04/29 22:53:51 trow Exp $</version>
  65         public sealed class RussianLowerCaseFilter : TokenFilter
  66         {
  67                 char[] charset;
  68
  69                 public RussianLowerCaseFilter(TokenStream _in, char[] charset) : base(_in)
  70                 {
  71                         this.charset = charset;
  72                 }
  73
  74                 public override Token Next()
  75                 {
  76                         Token t = input.Next();
  77
  78                         if (t == null)
  79                                 return null;
  80
  81                         String txt = t.TermText();
  82
  83                         char[] chArray = txt.ToCharArray();
  84                         for (int i = 0; i < chArray.Length; i++)
  85                         {
  86                                 chArray[i] = RussianCharsets.ToLowerCase(chArray[i], charset);
  87                         }
  88
  89                         String newTxt = new String(chArray);
  90                         // create new token
  91                         Token newToken = new Token(newTxt, t.StartOffset(), t.EndOffset());
  92
  93                         return newToken;
  94                 }
  95         }
  96 }