public SpanWeight(SpanQuery query, IndexSearcher searcher)
{
this.m_similarity = searcher.Similarity;
this.m_query = query;
m_termContexts = new Dictionary <Term, TermContext>();
ISet <Term> terms = new JCG.SortedSet <Term>();
query.ExtractTerms(terms);
IndexReaderContext context = searcher.TopReaderContext;
TermStatistics[] termStats = new TermStatistics[terms.Count];
int i = 0;
foreach (Term term in terms)
{
TermContext state = TermContext.Build(context, term);
termStats[i] = searcher.TermStatistics(term, state);
m_termContexts[term] = state;
i++;
}
string field = query.Field;
if (field != null)
{
m_stats = m_similarity.ComputeWeight(query.Boost, searcher.CollectionStatistics(query.Field), termStats);
}
}
/// <summary>
/// Computes a score factor for a simple term and returns an explanation
/// for that score factor.
///
/// <para/>
/// The default implementation uses:
///
/// <code>
/// Idf(docFreq, searcher.MaxDoc);
/// </code>
///
/// Note that <see cref="CollectionStatistics.MaxDoc"/> is used instead of
/// <see cref="Lucene.Net.Index.IndexReader.NumDocs"/> because also
/// <see cref="TermStatistics.DocFreq"/> is used, and when the latter
/// is inaccurate, so is <see cref="CollectionStatistics.MaxDoc"/>, and in the same direction.
/// In addition, <see cref="CollectionStatistics.MaxDoc"/> is more efficient to compute
/// </summary>
/// <param name="collectionStats"> Collection-level statistics </param>
/// <param name="termStats"> Term-level statistics for the term </param>
/// <returns> An Explain object that includes both an idf score factor
/// and an explanation for the term. </returns>
public virtual Explanation IdfExplain(CollectionStatistics collectionStats, TermStatistics termStats)
{
long df = termStats.DocFreq;
long max = collectionStats.MaxDoc;
float idf = Idf(df, max);
return(new Explanation(idf, "idf(docFreq=" + df + ", maxDocs=" + max + ")"));
}
public PhraseWeight(PhraseQuery outerInstance, IndexSearcher searcher)
{
this.outerInstance = outerInstance;
this.similarity = searcher.Similarity;
IndexReaderContext context = searcher.TopReaderContext;
states = new TermContext[outerInstance.terms.Count];
TermStatistics[] termStats = new TermStatistics[outerInstance.terms.Count];
for (int i = 0; i < outerInstance.terms.Count; i++)
{
Term term = outerInstance.terms[i];
states[i] = TermContext.Build(context, term);
termStats[i] = searcher.TermStatistics(term, states[i]);
}
stats = similarity.ComputeWeight(outerInstance.Boost, searcher.CollectionStatistics(outerInstance.field), termStats);
}
/// <summary>
/// Fills all member fields defined in {@code BasicStats} in {@code stats}.
/// Subclasses can override this method to fill additional stats.
/// </summary>
protected internal virtual void FillBasicStats(BasicStats stats, CollectionStatistics collectionStats, TermStatistics termStats)
{
// #positions(field) must be >= #positions(term)
Debug.Assert(collectionStats.SumTotalTermFreq() == -1 || collectionStats.SumTotalTermFreq() >= termStats.TotalTermFreq());
long numberOfDocuments = collectionStats.MaxDoc;
long docFreq = termStats.DocFreq();
long totalTermFreq = termStats.TotalTermFreq();
// codec does not supply totalTermFreq: substitute docFreq
if (totalTermFreq == -1)
{
totalTermFreq = docFreq;
}
long numberOfFieldTokens;
float avgFieldLength;
long sumTotalTermFreq = collectionStats.SumTotalTermFreq();
if (sumTotalTermFreq <= 0)
{
// field does not exist;
// We have to provide something if codec doesnt supply these measures,
// or if someone omitted frequencies for the field... negative values cause
// NaN/Inf for some scorers.
numberOfFieldTokens = docFreq;
avgFieldLength = 1;
}
else
{
numberOfFieldTokens = sumTotalTermFreq;
avgFieldLength = (float)numberOfFieldTokens / numberOfDocuments;
}
// TODO: add sumDocFreq for field (numberOfFieldPostings)
stats.NumberOfDocuments = numberOfDocuments;
stats.NumberOfFieldTokens = numberOfFieldTokens;
stats.AvgFieldLength = avgFieldLength;
stats.DocFreq = docFreq;
stats.TotalTermFreq = totalTermFreq;
}
// idf used for phrase queries
public override Explanation IdfExplain(CollectionStatistics collectionStats, TermStatistics[] termStats)
{
return new Explanation(1.0f, "Inexplicable");
}
/// <summary>
/// Computes the collection probability of the current term in addition to the
/// usual statistics.
/// </summary>
protected internal override void FillBasicStats(BasicStats stats, CollectionStatistics collectionStats, TermStatistics termStats)
{
base.FillBasicStats(stats, collectionStats, termStats);
LMStats lmStats = (LMStats)stats;
lmStats.CollectionProbability = collectionModel.ComputeProbability(stats);
}
/// <summary>
/// Computes a score factor for a phrase.
///
/// <p>
/// The default implementation sums the idf factor for
/// each term in the phrase.
/// </summary>
/// <param name="collectionStats"> collection-level statistics </param>
/// <param name="termStats"> term-level statistics for the terms in the phrase </param>
/// <returns> an Explain object that includes both an idf
/// score factor for the phrase and an explanation
/// for each term. </returns>
public virtual Explanation IdfExplain(CollectionStatistics collectionStats, TermStatistics[] termStats)
{
long max = collectionStats.MaxDoc();
float idf = 0.0f;
Explanation exp = new Explanation();
exp.Description = "idf(), sum of:";
foreach (TermStatistics stat in termStats)
{
long df = stat.DocFreq();
float termIdf = Idf(df, max);
exp.AddDetail(new Explanation(termIdf, "idf(docFreq=" + df + ", maxDocs=" + max + ")"));
idf += termIdf;
}
exp.Value = idf;
return exp;
}
/// <summary>
/// Computes a score factor for a simple term and returns an explanation
/// for that score factor.
///
/// <p>
/// The default implementation uses:
///
/// <pre class="prettyprint">
/// idf(docFreq, searcher.maxDoc());
/// </pre>
///
/// Note that <seealso cref="CollectionStatistics#maxDoc()"/> is used instead of
/// <seealso cref="Lucene.Net.Index.IndexReader#numDocs() IndexReader#numDocs()"/> because also
/// <seealso cref="TermStatistics#docFreq()"/> is used, and when the latter
/// is inaccurate, so is <seealso cref="CollectionStatistics#maxDoc()"/>, and in the same direction.
/// In addition, <seealso cref="CollectionStatistics#maxDoc()"/> is more efficient to compute
/// </summary>
/// <param name="collectionStats"> collection-level statistics </param>
/// <param name="termStats"> term-level statistics for the term </param>
/// <returns> an Explain object that includes both an idf score factor
/// and an explanation for the term. </returns>
public virtual Explanation IdfExplain(CollectionStatistics collectionStats, TermStatistics termStats)
{
long df = termStats.DocFreq();
long max = collectionStats.MaxDoc();
float idf = Idf(df, max);
return new Explanation(idf, "idf(docFreq=" + df + ", maxDocs=" + max + ")");
}
/// <summary>
/// Computes the collection probability of the current term in addition to the
/// usual statistics.
/// </summary>
protected internal override void FillBasicStats(BasicStats stats, CollectionStatistics collectionStats, TermStatistics termStats)
{
base.FillBasicStats(stats, collectionStats, termStats);
LMStats lmStats = (LMStats)stats;
lmStats.CollectionProbability = m_collectionModel.ComputeProbability(stats);
}
/// <summary>
/// Fills all member fields defined in {@code BasicStats} in {@code stats}.
/// Subclasses can override this method to fill additional stats.
/// </summary>
protected internal virtual void FillBasicStats(BasicStats stats, CollectionStatistics collectionStats, TermStatistics termStats)
{
// #positions(field) must be >= #positions(term)
Debug.Assert(collectionStats.SumTotalTermFreq() == -1 || collectionStats.SumTotalTermFreq() >= termStats.TotalTermFreq());
long numberOfDocuments = collectionStats.MaxDoc;
long docFreq = termStats.DocFreq();
long totalTermFreq = termStats.TotalTermFreq();
// codec does not supply totalTermFreq: substitute docFreq
if (totalTermFreq == -1)
{
totalTermFreq = docFreq;
}
long numberOfFieldTokens;
float avgFieldLength;
long sumTotalTermFreq = collectionStats.SumTotalTermFreq();
if (sumTotalTermFreq <= 0)
{
// field does not exist;
// We have to provide something if codec doesnt supply these measures,
// or if someone omitted frequencies for the field... negative values cause
// NaN/Inf for some scorers.
numberOfFieldTokens = docFreq;
avgFieldLength = 1;
}
else
{
numberOfFieldTokens = sumTotalTermFreq;
avgFieldLength = (float)numberOfFieldTokens / numberOfDocuments;
}
// TODO: add sumDocFreq for field (numberOfFieldPostings)
stats.NumberOfDocuments = numberOfDocuments;
stats.NumberOfFieldTokens = numberOfFieldTokens;
stats.AvgFieldLength = avgFieldLength;
stats.DocFreq = docFreq;
stats.TotalTermFreq = totalTermFreq;
}