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
  17 using System;
  18
  19 namespace Lucene.Net.Index
  20 {
  21
  22
  23         public class SegmentTermVector : TermFreqVector
  24         {
  25                 private System.String field;
  26                 private System.String[] terms;
  27                 private int[] termFreqs;
  28
  29                 internal SegmentTermVector(System.String field, System.String[] terms, int[] termFreqs)
  30                 {
  31                         this.field = field;
  32                         this.terms = terms;
  33                         this.termFreqs = termFreqs;
  34                 }
  35
  36                 /// <summary> </summary>
  37                 /// <returns> The number of the field this vector is associated with
  38                 /// </returns>
  39                 public virtual System.String GetField()
  40                 {
  41                         return field;
  42                 }
  43
  44                 public override System.String ToString()
  45                 {
  46                         System.Text.StringBuilder sb = new System.Text.StringBuilder();
  47                         sb.Append('{');
  48                         sb.Append(field).Append(": ");
  49                         if (terms != null)
  50                         {
  51                                 for (int i = 0; i < terms.Length; i++)
  52                                 {
  53                                         if (i > 0)
  54                                                 sb.Append(", ");
  55                                         sb.Append(terms[i]).Append('/').Append(termFreqs[i]);
  56                                 }
  57                         }
  58                         sb.Append('}');
  59
  60                         return sb.ToString();
  61                 }
  62
  63                 public virtual int Size()
  64                 {
  65                         return terms == null?0:terms.Length;
  66                 }
  67
  68                 public virtual System.String[] GetTerms()
  69                 {
  70                         return terms;
  71                 }
  72
  73                 public virtual int[] GetTermFrequencies()
  74                 {
  75                         return termFreqs;
  76                 }
  77
  78                 public virtual int IndexOf(System.String termText)
  79                 {
  80                         if (terms == null)
  81                                 return - 1;
  82                         int res = System.Array.BinarySearch(terms, termText);
  83                         return res >= 0?res:- 1;
  84                 }
  85
  86                 public virtual int[] IndexesOf(System.String[] termNumbers, int start, int len)
  87                 {
  88                         // TODO: there must be a more efficient way of doing this.
  89                         //       At least, we could advance the lower bound of the terms array
  90                         //       as we find valid indexes. Also, it might be possible to leverage
  91                         //       this even more by starting in the middle of the termNumbers array
  92                         //       and thus dividing the terms array maybe in half with each found index.
  93                         int[] res = new int[len];
  94
  95                         for (int i = 0; i < len; i++)
  96                         {
  97                                 res[i] = IndexOf(termNumbers[start + i]);
  98                         }
  99                         return res;
 100                 }
 101         }
 102 }