public TermWeight(TermQuery enclosingInstance, Searcher searcher)
{
InitBlock(enclosingInstance);
this.similarity = Enclosing_Instance.GetSimilarity(searcher);
idfExp = similarity.IdfExplain(Enclosing_Instance.term, searcher);
idf = idfExp.GetIdf();
}
public Result Search (string term, int count, int start) {
try {
term = term.ToLower ();
Term htTerm = new Term ("hottext", term);
Query qq1 = new FuzzyQuery (htTerm);
Query qq2 = new TermQuery (htTerm);
qq2.SetBoost (10f);
Query qq3 = new PrefixQuery (htTerm);
qq3.SetBoost (10f);
DisjunctionMaxQuery q1 = new DisjunctionMaxQuery (0f);
q1.Add (qq1);
q1.Add (qq2);
q1.Add (qq3);
Query q2 = new TermQuery (new Term ("text", term));
q2.SetBoost (3f);
Query q3 = new TermQuery (new Term ("examples", term));
q3.SetBoost (3f);
DisjunctionMaxQuery q = new DisjunctionMaxQuery (0f);
q.Add (q1);
q.Add (q2);
q.Add (q3);
TopDocs top = SearchInternal (q, count, start);
Result r = new Result (term, searcher, top.ScoreDocs);
Results.Add (r);
return r;
} catch (IOException) {
Console.WriteLine ("No index in {0}", dir);
return null;
}
}
public Result FastSearch (string term, int number)
{
try {
term = term.ToLower ();
Query q1 = new TermQuery (new Term ("hottext", term));
Query q2 = new PrefixQuery (new Term ("hottext", term));
q2.SetBoost (0.5f);
DisjunctionMaxQuery q = new DisjunctionMaxQuery (0f);
q.Add (q1);
q.Add (q2);
TopDocs top = searcher.Search (q, number);
return new Result (term, searcher, top.ScoreDocs);
} catch (IOException) {
Console.WriteLine ("No index in {0}", dir);
return null;
}
}
public override Query Rewrite(IndexReader reader)
{
if (!termLongEnough)
{
// can only match if it's exact
return new TermQuery(term);
}
FilteredTermEnum enumerator = GetEnum(reader);
int maxClauseCount = BooleanQuery.GetMaxClauseCount();
ScoreTermQueue stQueue = new ScoreTermQueue(maxClauseCount);
ScoreTerm reusableST = null;
try
{
do
{
float score = 0.0f;
Term t = enumerator.Term();
if (t != null)
{
score = enumerator.Difference();
if (reusableST == null)
{
reusableST = new ScoreTerm(t, score);
}
else if (score >= reusableST.score)
{
// reusableST holds the last "rejected" entry, so, if
// this new score is not better than that, there's no
// need to try inserting it
reusableST.score = score;
reusableST.term = t;
}
else
{
continue;
}
reusableST = (ScoreTerm) stQueue.InsertWithOverflow(reusableST);
}
}
while (enumerator.Next());
}
finally
{
enumerator.Close();
}
BooleanQuery query = new BooleanQuery(true);
int size = stQueue.Size();
for (int i = 0; i < size; i++)
{
ScoreTerm st = (ScoreTerm) stQueue.Pop();
TermQuery tq = new TermQuery(st.term); // found a match
tq.SetBoost(GetBoost() * st.score); // set the boost
query.Add(tq, BooleanClause.Occur.SHOULD); // add to query
}
return query;
}
private void InitBlock(TermQuery enclosingInstance)
{
this.enclosingInstance = enclosingInstance;
}
public override Query Rewrite(IndexReader reader, MultiTermQuery query)
{
// Get the enum and start visiting terms. If we
// exhaust the enum before hitting either of the
// cutoffs, we use ConstantBooleanQueryRewrite; else,
// ConstantFilterRewrite:
System.Collections.ArrayList pendingTerms = new System.Collections.ArrayList();
int docCountCutoff = (int) ((docCountPercent / 100.0) * reader.MaxDoc());
int termCountLimit = System.Math.Min(BooleanQuery.GetMaxClauseCount(), termCountCutoff);
int docVisitCount = 0;
FilteredTermEnum enumerator = query.GetEnum(reader);
try
{
while (true)
{
Term t = enumerator.Term();
if (t != null)
{
pendingTerms.Add(t);
// Loading the TermInfo from the terms dict here
// should not be costly, because 1) the
// query/filter will load the TermInfo when it
// runs, and 2) the terms dict has a cache:
docVisitCount += reader.DocFreq(t);
}
if (pendingTerms.Count >= termCountLimit || docVisitCount >= docCountCutoff)
{
// Too many terms -- make a filter.
Query result = new ConstantScoreQuery(new MultiTermQueryWrapperFilter(query));
result.SetBoost(query.GetBoost());
return result;
}
else if (!enumerator.Next())
{
// Enumeration is done, and we hit a small
// enough number of terms & docs -- just make a
// BooleanQuery, now
System.Collections.IEnumerator it = pendingTerms.GetEnumerator();
BooleanQuery bq = new BooleanQuery(true);
while (it.MoveNext())
{
TermQuery tq = new TermQuery((Term) it.Current);
bq.Add(tq, BooleanClause.Occur.SHOULD);
}
// Strip scores
Query result = new ConstantScoreQuery(new QueryWrapperFilter(bq));
result.SetBoost(query.GetBoost());
query.IncTotalNumberOfTerms(pendingTerms.Count);
return result;
}
}
}
finally
{
enumerator.Close();
}
}
public override Query Rewrite(IndexReader reader, MultiTermQuery query)
{
FilteredTermEnum enumerator = query.GetEnum(reader);
BooleanQuery result = new BooleanQuery(true);
int count = 0;
try
{
do
{
Term t = enumerator.Term();
if (t != null)
{
TermQuery tq = new TermQuery(t); // found a match
tq.SetBoost(query.GetBoost() * enumerator.Difference()); // set the boost
result.Add(tq, BooleanClause.Occur.SHOULD); // add to query
count++;
}
}
while (enumerator.Next());
}
finally
{
enumerator.Close();
}
query.IncTotalNumberOfTerms(count);
return result;
}
public override Weight CreateWeight(Searcher searcher)
{
if (terms.Count == 1)
{
// optimize one-term case
Term term = (Term) terms[0];
Query termQuery = new TermQuery(term);
termQuery.SetBoost(GetBoost());
return termQuery.CreateWeight(searcher);
}
return new PhraseWeight(this, searcher);
}
private void InitBlock(TermQuery enclosingInstance)
{
this.enclosingInstance = enclosingInstance;
}
