/// <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="terms">the terms in the phrase
/// </param>
/// <param name="searcher">the document collection being searched
/// </param>
/// <returns> an IDFExplain object that includes both an idf
/// score factor for the phrase and an explanation
/// for each term.
/// </returns>
/// <throws> IOException </throws>
public virtual IDFExplanation idfExplain(IList <Lucene.Net.Index.Term> terms, Searcher searcher)
{
if (SupportedMethods.overridesCollectionIDF)
{
float idf = Idf(terms, searcher);
return(new IDFExplanation
{
GetIdf = () => idf,
Explain = () => "Inexplicable"
});
}
int max = searcher.MaxDoc();
float idf2 = 0.0f;
System.Text.StringBuilder exp = new System.Text.StringBuilder();
foreach (Term term in terms)
{
int df = searcher.DocFreq(term);
idf2 += Idf(df, max);
exp.Append(" ");
exp.Append(term.Text());
exp.Append("=");
exp.Append(df);
}
float fIdf = idf2;
return(new IDFExplanation
{
GetIdf = () => fIdf,
Explain = () => exp.ToString()
});
}
// count # deletions, return -1 if unknown.
private int CountDeletions(Searcher s)
{
int cnt = -1;
if (s is IndexSearcher)
{
cnt = s.MaxDoc() - ((IndexSearcher)s).GetIndexReader().NumDocs();
}
return(cnt);
}
/// <summary> Computes a score factor for a simple term and returns an explanation
/// for that score factor.
///
/// <p/>
/// The default implementation uses:
///
/// <pre>
/// idf(searcher.docFreq(term), searcher.maxDoc());
/// </pre>
///
/// Note that {@link Searcher#MaxDoc()} is used instead of
/// {@link Lucene.Net.Index.IndexReader#NumDocs()} because it is
/// proportional to {@link Searcher#DocFreq(Term)} , i.e., when one is
/// inaccurate, so is the other, and in the same direction.
///
/// </summary>
/// <param name="term">the term in question
/// </param>
/// <param name="searcher">the document collection being searched
/// </param>
/// <returns> an IDFExplain object that includes both an idf score factor
/// and an explanation for the term.
/// </returns>
/// <throws> IOException </throws>
public virtual IDFExplanation IdfExplain(Term term, Searcher searcher)
{
if (SupportedMethods.overridesTermIDF)
{
float idf = Idf(term, searcher);
return(new AnonymousClassIDFExplanation(idf, this));
}
int df = searcher.DocFreq(term);
int max = searcher.MaxDoc();
float idf2 = Idf(df, max);
return(new AnonymousClassIDFExplanation1(df, max, idf2, this));
}
/// <summary> Computes a score factor for a simple term and returns an explanation
/// for that score factor.
///
/// <p/>
/// The default implementation uses:
///
/// <pre>
/// idf(searcher.docFreq(term), searcher.maxDoc());
/// </pre>
///
/// Info: <see cref="Searcher.MaxDoc()"/> is used instead of
/// <see cref="Lucene.Net.Index.IndexReader.NumDocs()"/> because it is
/// proportional to <see cref="Searcher.DocFreq(Term)"/> , i.e., when one is
/// inaccurate, so is the other, and in the same direction.
///
/// </summary>
/// <param name="term">the term in question
/// </param>
/// <param name="searcher">the document collection being searched
/// </param>
/// <returns> an IDFExplain object that includes both an idf score factor
/// and an explanation for the term.
/// </returns>
/// <throws> IOException </throws>
public virtual IDFExplanation IdfExplain(Term term, Searcher searcher)
{
if (SupportedMethods.overridesTermIDF)
{
float idf = Idf(term, searcher);
return(new IDFExplanation
{
GetIdf = () => idf,
Explain = () => "Inexplicable"
});
}
int df = searcher.DocFreq(term);
int max = searcher.MaxDoc();
float idf2 = Idf(df, max);
return(new IDFExplanation
{
GetIdf = () => idf2,
Explain = () => "idf(docFreq=" + df + ", maxDocs=" + max + ")"
});
}
public static float EXPLAIN_SCORE_TOLERANCE_DELTA = 0.00025f; // {{See: LUCENENET-288}} Intentional diversion from Java Lucene per above comment
/// <summary> Tests that all documents up to maxDoc which are *not* in the
/// expected result set, have an explanation which indicates no match
/// (ie: Explanation value of 0.0f)
/// </summary>
public static void CheckNoMatchExplanations(Query q, System.String defaultFieldName, Searcher searcher, int[] results)
{
System.String d = q.ToString(defaultFieldName);
System.Collections.Hashtable ignore = new System.Collections.Hashtable();
for (int i = 0; i < results.Length; i++)
{
SupportClass.CollectionsHelper.AddIfNotContains(ignore, (System.Int32) results[i]);
}
int maxDoc = searcher.MaxDoc();
for (int doc = 0; doc < maxDoc; doc++)
{
if (ignore.Contains((System.Int32) doc))
continue;
Explanation exp = searcher.Explain(q, doc);
Assert.IsNotNull(exp, "Explanation of [[" + d + "]] for #" + doc + " is null");
Assert.AreEqual(0.0f, exp.GetValue(), 0.0f, "Explanation of [[" + d + "]] for #" + doc + " doesn't indicate non-match: " + exp.ToString());
}
}
public static float EXPLAIN_SCORE_TOLERANCE_DELTA = 0.00025f; // {{See: LUCENENET-288}} Intentional diversion from Java Lucene per above comment
/// <summary> Tests that all documents up to maxDoc which are *not* in the
/// expected result set, have an explanation which indicates no match
/// (ie: Explanation value of 0.0f)
/// </summary>
public static void CheckNoMatchExplanations(Query q, System.String defaultFieldName, Searcher searcher, int[] results)
{
System.String d = q.ToString(defaultFieldName);
System.Collections.Hashtable ignore = new System.Collections.Hashtable();
for (int i = 0; i < results.Length; i++)
{
SupportClass.CollectionsHelper.AddIfNotContains(ignore, (System.Int32)results[i]);
}
int maxDoc = searcher.MaxDoc();
for (int doc = 0; doc < maxDoc; doc++)
{
if (ignore.Contains((System.Int32)doc))
{
continue;
}
Explanation exp = searcher.Explain(q, doc);
Assert.IsNotNull(exp, "Explanation of [[" + d + "]] for #" + doc + " is null");
Assert.AreEqual(0.0f, exp.GetValue(), 0.0f, "Explanation of [[" + d + "]] for #" + doc + " doesn't indicate non-match: " + exp.ToString());
}
}
/// <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="terms">the terms in the phrase
/// </param>
/// <param name="searcher">the document collection being searched
/// </param>
/// <returns> an IDFExplain object that includes both an idf
/// score factor for the phrase and an explanation
/// for each term.
/// </returns>
/// <throws> IOException </throws>
public virtual IDFExplanation idfExplain(System.Collections.ICollection terms, Searcher searcher)
{
if (SupportedMethods.overridesCollectionIDF)
{
float idf = Idf(terms, searcher);
return(new AnonymousClassIDFExplanation2(idf, this));
}
int max = searcher.MaxDoc();
float idf2 = 0.0f;
System.Text.StringBuilder exp = new System.Text.StringBuilder();
foreach (Term term in terms)
{
int df = searcher.DocFreq(term);
idf2 += Idf(df, max);
exp.Append(" ");
exp.Append(term.Text());
exp.Append("=");
exp.Append(df);
}
float fIdf = idf2;
return(new AnonymousClassIDFExplanation3(fIdf, exp, this));
}
/// <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="terms">the terms in the phrase
/// </param>
/// <param name="searcher">the document collection being searched
/// </param>
/// <returns> an IDFExplain object that includes both an idf
/// score factor for the phrase and an explanation
/// for each term.
/// </returns>
/// <throws> IOException </throws>
public virtual IDFExplanation idfExplain(System.Collections.ICollection terms, Searcher searcher)
{
if (SupportedMethods.overridesCollectionIDF)
{
float idf = Idf(terms, searcher);
return new AnonymousClassIDFExplanation2(idf, this);
}
int max = searcher.MaxDoc();
float idf2 = 0.0f;
System.Text.StringBuilder exp = new System.Text.StringBuilder();
foreach (Term term in terms)
{
int df = searcher.DocFreq(term);
idf2 += Idf(df, max);
exp.Append(" ");
exp.Append(term.Text());
exp.Append("=");
exp.Append(df);
}
float fIdf = idf2;
return new AnonymousClassIDFExplanation3(fIdf, exp, this);
}
/// <summary> Computes a score factor for a simple term and returns an explanation
/// for that score factor.
///
/// <p/>
/// The default implementation uses:
///
/// <pre>
/// idf(searcher.docFreq(term), searcher.maxDoc());
/// </pre>
///
/// Note that {@link Searcher#MaxDoc()} is used instead of
/// {@link Lucene.Net.Index.IndexReader#NumDocs()} because it is
/// proportional to {@link Searcher#DocFreq(Term)} , i.e., when one is
/// inaccurate, so is the other, and in the same direction.
///
/// </summary>
/// <param name="term">the term in question
/// </param>
/// <param name="searcher">the document collection being searched
/// </param>
/// <returns> an IDFExplain object that includes both an idf score factor
/// and an explanation for the term.
/// </returns>
/// <throws> IOException </throws>
public virtual IDFExplanation IdfExplain(Term term, Searcher searcher)
{
if (SupportedMethods.overridesTermIDF)
{
float idf = Idf(term, searcher);
return new AnonymousClassIDFExplanation(idf, this);
}
int df = searcher.DocFreq(term);
int max = searcher.MaxDoc();
float idf2 = Idf(df, max);
return new AnonymousClassIDFExplanation1(df, max, idf2, this);
}
public virtual float Idf(Term term, Searcher searcher)
{
return Idf(searcher.DocFreq(term), searcher.MaxDoc());
}
/// <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="terms">the terms in the phrase
/// </param>
/// <param name="searcher">the document collection being searched
/// </param>
/// <returns> an IDFExplain object that includes both an idf
/// score factor for the phrase and an explanation
/// for each term.
/// </returns>
/// <throws> IOException </throws>
public virtual IDFExplanation idfExplain(IList<Lucene.Net.Index.Term> terms, Searcher searcher)
{
if (SupportedMethods.overridesCollectionIDF)
{
float idf = Idf(terms, searcher);
return new IDFExplanation
{
GetIdf = () => idf,
Explain = () => "Inexplicable"
};
}
int max = searcher.MaxDoc();
float idf2 = 0.0f;
System.Text.StringBuilder exp = new System.Text.StringBuilder();
foreach (Term term in terms)
{
int df = searcher.DocFreq(term);
idf2 += Idf(df, max);
exp.Append(" ");
exp.Append(term.Text());
exp.Append("=");
exp.Append(df);
}
float fIdf = idf2;
return new IDFExplanation
{
GetIdf = () => fIdf,
Explain = () => exp.ToString()
};
}
/// <summary> Computes a score factor for a simple term and returns an explanation
/// for that score factor.
///
/// <p/>
/// The default implementation uses:
///
/// <pre>
/// idf(searcher.docFreq(term), searcher.maxDoc());
/// </pre>
///
/// Info: <see cref="Searcher.MaxDoc()"/> is used instead of
/// <see cref="Lucene.Net.Index.IndexReader.NumDocs()"/> because it is
/// proportional to <see cref="Searcher.DocFreq(Term)"/> , i.e., when one is
/// inaccurate, so is the other, and in the same direction.
///
/// </summary>
/// <param name="term">the term in question
/// </param>
/// <param name="searcher">the document collection being searched
/// </param>
/// <returns> an IDFExplain object that includes both an idf score factor
/// and an explanation for the term.
/// </returns>
/// <throws> IOException </throws>
public virtual IDFExplanation IdfExplain(Term term, Searcher searcher)
{
if (SupportedMethods.overridesTermIDF)
{
float idf = Idf(term, searcher);
return new IDFExplanation
{ GetIdf = () => idf,
Explain = () => "Inexplicable"
};
}
int df = searcher.DocFreq(term);
int max = searcher.MaxDoc();
float idf2 = Idf(df, max);
return new IDFExplanation
{
GetIdf = () => idf2,
Explain = () => "idf(docFreq=" + df + ", maxDocs=" + max + ")"
};
}
/// <summary>Computes a score factor for a simple term.
///
/// <p>The default implementation is:<pre>
/// return idf(searcher.docFreq(term), searcher.maxDoc());
/// </pre>
///
/// Note that {@link Searcher#MaxDoc()} is used instead of
/// {@link IndexReader#NumDocs()} because it is proportional to
/// {@link Searcher#DocFreq(Term)} , i.e., when one is inaccurate,
/// so is the other, and in the same direction.
///
/// </summary>
/// <param name="term">the term in question
/// </param>
/// <param name="searcher">the document collection being searched
/// </param>
/// <returns> a score factor for the term
/// </returns>
public virtual float Idf(Term term, Searcher searcher)
{
return(Ldf(searcher.DocFreq(term), searcher.MaxDoc()));
}
// count # deletions, return -1 if unknown.
private int CountDeletions(Searcher s)
{
int cnt = - 1;
if (s is IndexSearcher)
{
cnt = s.MaxDoc() - ((IndexSearcher) s).GetIndexReader().NumDocs();
}
return cnt;
}
