Lucene.Net/Search/PhraseScorer.cs

   1 using Lucene.Net.Util;
   2 using Lucene.Net.Index;
   3
   4 namespace Lucene.Net.Search
   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         abstract class PhraseScorer : Scorer
  61         {
  62                 private Weight weight;
  63                 protected byte[] norms;
  64                 protected float value;
  65
  66                 protected PhraseQueue pq;
  67                 protected PhrasePositions first, last;
  68
  69                 private float freq;
  70
  71                 internal PhraseScorer(Weight weight, TermPositions[] tps, Similarity similarity,
  72                         byte[] norms) : base(similarity)
  73                 {
  74                         this.norms = norms;
  75                         this.weight = weight;
  76                         this.value = weight.GetValue();
  77
  78                         // use PQ to build a sorted list of PhrasePositions
  79                         pq = new PhraseQueue(tps.Length);
  80                         for (int i = 0; i < tps.Length; i++)
  81                                 pq.Put(new PhrasePositions(tps[i], i));
  82                         PqToList();
  83                 }
  84
  85                 public override void Score(HitCollector results, int end)
  86                 {
  87                         Similarity similarity = GetSimilarity();
  88                         while (last.doc < end)
  89                         {                         // find doc w/ all the terms
  90                                 while (first.doc < last.doc)
  91                                 {                 // scan forward in first
  92                                         do
  93                                         {
  94                                                 first.Next();
  95                                         } while (first.doc < last.doc);
  96                                         FirstToLast();
  97                                         if (last.doc >= end)
  98                                                 return;
  99                                 }
 100
 101                                 // found doc with all terms
 102                                 freq = PhraseFreq();                        // check for phrase
 103
 104                                 if (freq > 0.0)
 105                                 {
 106                                         float score = similarity.Tf(freq)*value;  // compute score
 107                                         score *= Similarity.DecodeNorm(norms[first.doc]); // normalize
 108                                         results.Collect(first.doc, score);        // add to results
 109                                 }
 110                                 last.Next();                              // resume scanning
 111                         }
 112                 }
 113
 114                 protected abstract float PhraseFreq() ;
 115
 116                 protected void PqToList()
 117                 {
 118                         last = first = null;
 119                         while (pq.Top() != null)
 120                         {
 121                                 PhrasePositions pp = (PhrasePositions)pq.Pop();
 122                                 if (last != null)
 123                                 {                         // add next to end of list
 124                                         last.next = pp;
 125                                 }
 126                                 else
 127                                         first = pp;
 128                                 last = pp;
 129                                 pp.next = null;
 130                         }
 131                 }
 132
 133                 protected void FirstToLast()
 134                 {
 135                         last.next = first;                        // move first to end of list
 136                         last = first;
 137                         first = first.next;
 138                         last.next = null;
 139                 }
 140
 141                 class PhraseScorerHitCollector : HitCollector
 142                 {
 143                         public override void Collect(int d, float score) {}
 144                 }
 145
 146                 public override Explanation Explain(int doc)
 147                 {
 148                         Explanation tfExplanation = new Explanation();
 149
 150                         Score(new PhraseScorerHitCollector(), doc+1);
 151
 152                         float phraseFreq = (first.doc == doc) ? freq : 0.0f;
 153                         tfExplanation.SetValue(GetSimilarity().Tf(phraseFreq));
 154                         tfExplanation.SetDescription("tf(phraseFreq=" + phraseFreq + ")");
 155
 156                         return tfExplanation;
 157                 }
 158         }
 159 }