/// <summary>
/// Creates an <see cref="T:IDictionary{string, WeightedSpanTerm}"/> from the given <see cref="Query"/> and <see cref="Analysis.TokenStream"/>. Uses a supplied
/// <see cref="IndexReader"/> to properly Weight terms (for gradient highlighting).
/// </summary>
/// <param name="query"><see cref="Query"/> that caused hit</param>
/// <param name="tokenStream"><see cref="Analysis.TokenStream"/> of text to be highlighted</param>
/// <param name="fieldName">restricts Term's used based on field name</param>
/// <param name="reader">to use for scoring</param>
/// <returns>Map of <see cref="WeightedSpanTerm"/>s with quasi tf/idf scores</returns>
/// <exception cref="IOException">If there is a low-level I/O error</exception>
public virtual IDictionary <string, WeightedSpanTerm> GetWeightedSpanTermsWithScores(
Query query, TokenStream tokenStream, string fieldName, IndexReader reader)
{
this.fieldName = fieldName?.Intern();
this.tokenStream = tokenStream;
IDictionary <string, WeightedSpanTerm> terms = new PositionCheckingMap <string>();
Extract(query, terms);
int totalNumDocs = reader.MaxDoc;
var weightedTerms = terms.Keys;
try
{
foreach (var wt in weightedTerms)
{
terms.TryGetValue(wt, out WeightedSpanTerm weightedSpanTerm);
int docFreq = reader.DocFreq(new Term(fieldName, weightedSpanTerm.Term));
// IDF algorithm taken from DefaultSimilarity class
float idf = (float)(Math.Log((float)totalNumDocs / (double)(docFreq + 1)) + 1.0);
weightedSpanTerm.Weight *= idf;
}
}
finally
{
IOUtils.Dispose(internalReader);
}
return(terms);
}
/// <summary>
/// Creates an <see cref="T:IDictionary{string, WeightedSpanTerm}"/> from the given <see cref="Query"/> and <see cref="Analysis.TokenStream"/>.
/// </summary>
/// <param name="query"><see cref="Query"/> that caused hit</param>
/// <param name="tokenStream"><see cref="Analysis.TokenStream"/> of text to be highlighted</param>
/// <param name="fieldName">restricts Term's used based on field name</param>
/// <returns>Map containing <see cref="WeightedSpanTerm"/>s</returns>
/// <exception cref="IOException">If there is a low-level I/O error</exception>
public virtual IDictionary <string, WeightedSpanTerm> GetWeightedSpanTerms(Query query, TokenStream tokenStream,
string fieldName)
{
if (fieldName != null)
{
this.fieldName = fieldName.Intern();
}
else
{
this.fieldName = null;
}
IDictionary <string, WeightedSpanTerm> terms = new PositionCheckingMap <string>();
this.tokenStream = tokenStream;
try
{
Extract(query, terms);
}
finally
{
IOUtils.Dispose(internalReader);
}
return(terms);
}
/// <summary>
/// Creates a Map of <c>WeightedSpanTerms</c> from the given <c>Query</c> and <c>TokenStream</c>.
/// </summary>
/// <param name="query">query that caused hit</param>
/// <param name="tokenStream">tokenStream of text to be highlighted</param>
/// <param name="fieldName">restricts Term's used based on field name</param>
/// <returns>Map containing WeightedSpanTerms</returns>
public IDictionary <String, WeightedSpanTerm> GetWeightedSpanTerms(Query query, TokenStream tokenStream,
String fieldName)
{
if (fieldName != null)
{
this.fieldName = StringHelper.Intern(fieldName);
}
else
{
this.fieldName = null;
}
IDictionary <String, WeightedSpanTerm> terms = new PositionCheckingMap <String>();
this.tokenStream = tokenStream;
try
{
Extract(query, terms);
}
finally
{
CloseReaders();
}
return(terms);
}
/// <summary>
/// Creates a Map of <c>WeightedSpanTerms</c> from the given <c>Query</c> and <c>TokenStream</c>. Uses a supplied
/// <c>IndexReader</c> to properly Weight terms (for gradient highlighting).
/// </summary>
/// <param name="query">Query that caused hit</param>
/// <param name="tokenStream">Tokenstream of text to be highlighted</param>
/// <param name="fieldName">restricts Term's used based on field name</param>
/// <param name="reader">to use for scoring</param>
/// <returns>Map of WeightedSpanTerms with quasi tf/idf scores</returns>
public IDictionary <String, WeightedSpanTerm> GetWeightedSpanTermsWithScores(Query query, TokenStream tokenStream,
String fieldName, IndexReader reader)
{
if (fieldName != null)
{
this.fieldName = StringHelper.Intern(fieldName);
}
else
{
this.fieldName = null;
}
this.tokenStream = tokenStream;
IDictionary <String, WeightedSpanTerm> terms = new PositionCheckingMap <String>();
Extract(query, terms);
int totalNumDocs = reader.NumDocs();
var weightedTerms = terms.Keys;
try
{
foreach (var wt in weightedTerms)
{
WeightedSpanTerm weightedSpanTerm = terms[wt];
int docFreq = reader.DocFreq(new Term(fieldName, weightedSpanTerm.Term));
// docFreq counts deletes
if (totalNumDocs < docFreq)
{
docFreq = totalNumDocs;
}
// IDF algorithm taken from DefaultSimilarity class
float idf = (float)(Math.Log((float)totalNumDocs / (double)(docFreq + 1)) + 1.0);
weightedSpanTerm.Weight *= idf;
}
}
finally
{
CloseReaders();
}
return(terms);
}
