beagled/Lucene.Net/Index/SegmentTermVector.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 namespace Lucene.Net.Index
  18 {
  19
  20
  21         public class SegmentTermVector : TermFreqVector
  22         {
  23                 private System.String field;
  24                 private System.String[] terms;
  25                 private int[] termFreqs;
  26
  27                 internal SegmentTermVector(System.String field, System.String[] terms, int[] termFreqs)
  28                 {
  29                         this.field = field;
  30                         this.terms = terms;
  31                         this.termFreqs = termFreqs;
  32                 }
  33
  34                 /// <summary> </summary>
  35                 /// <returns> The number of the Field this vector is associated with
  36                 /// </returns>
  37                 public virtual System.String GetField()
  38                 {
  39                         return field;
  40                 }
  41
  42                 public override System.String ToString()
  43                 {
  44                         System.Text.StringBuilder sb = new System.Text.StringBuilder();
  45                         sb.Append('{');
  46                         sb.Append(field).Append(": ");
  47             if (terms != null)
  48             {
  49                 for (int i = 0; i < terms.Length; i++)
  50                 {
  51                     if (i > 0)
  52                         sb.Append(", ");
  53                     sb.Append(terms[i]).Append('/').Append(termFreqs[i]);
  54                 }
  55             }
  56                         sb.Append('}');
  57
  58                         return sb.ToString();
  59                 }
  60
  61                 public virtual int Size()
  62                 {
  63                         return terms == null?0:terms.Length;
  64                 }
  65
  66                 public virtual System.String[] GetTerms()
  67                 {
  68                         return terms;
  69                 }
  70
  71                 public virtual int[] GetTermFrequencies()
  72                 {
  73                         return termFreqs;
  74                 }
  75
  76                 public virtual int IndexOf(System.String termText)
  77                 {
  78             if (terms == null)
  79                 return - 1;
  80             int res = System.Array.BinarySearch(terms, termText);
  81                         return res >= 0?res:- 1;
  82                 }
  83
  84                 public virtual int[] IndexesOf(System.String[] termNumbers, int start, int len)
  85                 {
  86                         // TODO: there must be a more efficient way of doing this.
  87                         //       At least, we could advance the lower bound of the terms array
  88                         //       as we find valid indexes. Also, it might be possible to leverage
  89                         //       this even more by starting in the middle of the termNumbers array
  90                         //       and thus dividing the terms array maybe in half with each found index.
  91                         int[] res = new int[len];
  92
  93                         for (int i = 0; i < len; i++)
  94                         {
  95                                 res[i] = IndexOf(termNumbers[start + i]);
  96                         }
  97                         return res;
  98                 }
  99         }
 100 }