private void testEnum(int lower, int upper) { NumericRangeQuery <int> q = NumericRangeQuery.NewIntRange("field4", 4, lower, upper, true, true); FilteredTermEnum termEnum = q.GetEnum(searcher.IndexReader, null); try { int count = 0; do { Term t = termEnum.Term; if (t != null) { int val = NumericUtils.PrefixCodedToInt(t.Text); Assert.True(val >= lower && val <= upper, "value not in bounds"); count++; } else { break; } } while (termEnum.Next(null)); Assert.False(termEnum.Next(null)); Console.WriteLine("TermEnum on 'field4' for range [" + lower + "," + upper + "] contained " + count + " terms."); } finally { termEnum.Close(); } }
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 Query Rewrite(IndexReader reader) { 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); }
public override Query Rewrite(IndexReader reader) { FilteredTermEnum enumerator = GetEnum(reader); int maxClauseCount = BooleanQuery.GetMaxClauseCount(); ScoreTermQueue stQueue = new ScoreTermQueue(maxClauseCount); try { do { float minScore = 0.0f; float score = 0.0f; Term t = enumerator.Term(); if (t != null) { score = enumerator.Difference(); // terms come in alphabetical order, therefore if queue is full and score // not bigger than minScore, we can skip if (stQueue.Size() < maxClauseCount || score > minScore) { stQueue.Insert(new ScoreTerm(t, score)); minScore = ((ScoreTerm)stQueue.Top()).score; // maintain minScore } } }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); }
public override Query Rewrite(IndexReader reader) { FilteredTermEnum enumerator = GetEnum(reader); BooleanQuery query = new BooleanQuery(true); try { do { Term t = enumerator.Term(); if (t != null) { TermQuery tq = new TermQuery(t); // found a match tq.SetBoost(GetBoost() * enumerator.Difference()); // set the boost query.Add(tq, BooleanClause.Occur.SHOULD); // add to query } }while (enumerator.Next()); } finally { enumerator.Close(); } return(query); }
public override Query Rewrite(IndexReader reader, IState state) { if (!termLongEnough) { // can only match if it's exact return(new TermQuery(Term)); } int maxSize = BooleanQuery.MaxClauseCount; // TODO: Java uses a PriorityQueue. Using Linq, we can emulate it, // however it's considerable slower than the java counterpart. // this should be a temporary thing, fixed before release SortedList <ScoreTerm, ScoreTerm> stQueue = new SortedList <ScoreTerm, ScoreTerm>(); FilteredTermEnum enumerator = GetEnum(reader, state); try { ScoreTerm st = new ScoreTerm(); do { Term t = enumerator.Term; if (t == null) { break; } float score = enumerator.Difference(); //ignore uncompetetive hits if (stQueue.Count >= maxSize && score <= stQueue.Keys.First().score) { continue; } // add new entry in PQ st.term = t; st.score = score; stQueue.Add(st, st); // possibly drop entries from queue if (stQueue.Count > maxSize) { st = stQueue.Keys.First(); stQueue.Remove(st); } else { st = new ScoreTerm(); } }while (enumerator.Next(state)); } finally { enumerator.Close(); } BooleanQuery query = new BooleanQuery(true); foreach (ScoreTerm st in stQueue.Keys) { TermQuery tq = new TermQuery(st.term); // found a match tq.Boost = Boost * st.score; // set the boost query.Add(tq, Occur.SHOULD); // add to query } return(query); }