Lucene.Net/Analysis/DE/WordlistLoader.cs

   1 using System;
   2 using System.IO;
   3 using System.Collections;
   4
   5 namespace Lucene.Net.Analysis.De
   6 {
   7         /* ====================================================================
   8          * The Apache Software License, Version 1.1
   9          *
  10          * Copyright (c) 2001 The Apache Software Foundation.  All rights
  11          * reserved.
  12          *
  13          * Redistribution and use in source and binary forms, with or without
  14          * modification, are permitted provided that the following conditions
  15          * are met:
  16          *
  17          * 1. Redistributions of source code must retain the above copyright
  18          *    notice, this list of conditions and the following disclaimer.
  19          *
  20          * 2. Redistributions in binary form must reproduce the above copyright
  21          *    notice, this list of conditions and the following disclaimer in
  22          *    the documentation and/or other materials provided with the
  23          *    distribution.
  24          *
  25          * 3. The end-user documentation included with the redistribution,
  26          *    if any, must include the following acknowledgment:
  27          *       "This product includes software developed by the
  28          *        Apache Software Foundation (http://www.apache.org/)."
  29          *    Alternately, this acknowledgment may appear in the software itself,
  30          *    if and wherever such third-party acknowledgments normally appear.
  31          *
  32          * 4. The names "Apache" and "Apache Software Foundation" and
  33          *    "Apache Lucene" must not be used to endorse or promote products
  34          *    derived from this software without prior written permission. For
  35          *    written permission, please contact apache@apache.org.
  36          *
  37          * 5. Products derived from this software may not be called "Apache",
  38          *    "Apache Lucene", nor may "Apache" appear in their name, without
  39          *    prior written permission of the Apache Software Foundation.
  40          *
  41          * THIS SOFTWARE IS PROVIDED ``AS IS'' AND ANY EXPRESSED OR IMPLIED
  42          * WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
  43          * OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
  44          * DISCLAIMED.  IN NO EVENT SHALL THE APACHE SOFTWARE FOUNDATION OR
  45          * ITS CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
  46          * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
  47          * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
  48          * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
  49          * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
  50          * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
  51          * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
  52          * SUCH DAMAGE.
  53          * ====================================================================
  54          *
  55          * This software consists of voluntary contributions made by many
  56          * individuals on behalf of the Apache Software Foundation.  For more
  57          * information on the Apache Software Foundation, please see
  58          * <http://www.apache.org/>.
  59          */
  60
  61         /// <summary>
  62         /// Loads a text file and adds every line as an entry to a Hashtable. Every line
  63         /// should contain only one word. If the file is not found or on any error, an
  64         /// empty table is returned.
  65         /// </summary>
  66         /// <author>Gerhard Schwarz</author>
  67         /// <version>$Id: WordlistLoader.cs,v 1.1.1.1 2004/04/29 22:53:51 trow Exp $</version>
  68         public class WordlistLoader
  69         {
  70                 /// <summary>
  71                 /// </summary>
  72                 /// <param name="path">Path to the wordlist</param>
  73                 /// <param name="wordfile">Name of the wordlist</param>
  74                 /// <returns></returns>
  75                 public static Hashtable GetWordtable( String path, String wordfile )
  76                 {
  77                         if ( path == null || wordfile == null )
  78                         {
  79                                 return new Hashtable();
  80                         }
  81                         return GetWordtable(new FileInfo(path + "\\" + wordfile));
  82                 }
  83
  84                 /// <summary>
  85                 /// </summary>
  86                 /// <param name="wordfile">Complete path to the wordlist</param>
  87                 /// <returns></returns>
  88                 public static Hashtable GetWordtable( String wordfile )
  89                 {
  90                         if ( wordfile == null )
  91                         {
  92                                 return new Hashtable();
  93                         }
  94                         return GetWordtable( new FileInfo( wordfile ) );
  95                 }
  96
  97                 /// <summary>
  98                 ///
  99                 /// </summary>
 100                 /// <param name="wordfile">File containing the wordlist</param>
 101                 /// <returns></returns>
 102                 public static Hashtable GetWordtable( FileInfo wordfile )
 103                 {
 104                         if ( wordfile == null )
 105                         {
 106                                 return new Hashtable();
 107                         }
 108                         Hashtable result = null;
 109                         try
 110                         {
 111                                 StreamReader lnr = new StreamReader(wordfile.FullName);
 112                                 String word = null;
 113                                 String[] stopwords = new String[100];
 114                                 int wordcount = 0;
 115                                 while ( ( word = lnr.ReadLine() ) != null )
 116                                 {
 117                                         wordcount++;
 118                                         if ( wordcount == stopwords.Length )
 119                                         {
 120                                                 String[] tmp = new String[stopwords.Length + 50];
 121                                                 Array.Copy( stopwords, 0, tmp, 0, wordcount );
 122                                                 stopwords = tmp;
 123                                         }
 124                                         stopwords[wordcount-1] = word;
 125                                 }
 126                                 result = MakeWordTable( stopwords, wordcount );
 127                         }
 128                                 // On error, use an empty table
 129                         catch (IOException)
 130                         {
 131                                 result = new Hashtable();
 132                         }
 133                         return result;
 134                 }
 135
 136                 /// <summary>
 137                 /// Builds the wordlist table.
 138                 /// </summary>
 139                 /// <param name="words">Word that where read</param>
 140                 /// <param name="length">Amount of words that where read into <tt>words</tt></param>
 141                 /// <returns></returns>
 142                 private static Hashtable MakeWordTable( String[] words, int length )
 143                 {
 144                         Hashtable table = new Hashtable( length );
 145                         for ( int i = 0; i < length; i++ )
 146                         {
 147                                 table.Add(words[i], words[i]);
 148                         }
 149                         return table;
 150                 }
 151         }
 152 }