private float freq; //prhase frequency in current doc as computed by phraseFreq().
internal PhraseScorer(Weight weight, TermPositions[] tps, int[] offsets, Similarity similarity, byte[] norms):base(similarity)
{
this.norms = norms;
this.weight = weight;
this.value_Renamed = weight.Value;
// convert tps to a list of phrase positions.
// note: phrase-position differs from term-position in that its position
// reflects the phrase offset: pp.pos = tp.pos - offset.
// this allows to easily identify a matching (exact) phrase
// when all PhrasePositions have exactly the same position.
for (int i = 0; i < tps.Length; i++)
{
PhrasePositions pp = new PhrasePositions(tps[i], offsets[i]);
if (last != null)
{
// add next to end of list
last.next = pp;
}
else
{
first = pp;
}
last = pp;
}
pq = new PhraseQueue(tps.Length); // construct empty pq
first.doc = - 1;
}
public TermSpans(TermPositions positions, Term term)
{
this.internalPositions = positions;
this.term = term;
internalDoc = - 1;
}
public TermSpans(TermPositions positions, Term term)
{
this.positions = positions;
this.term = term;
doc = - 1;
}
public virtual Scorer Scorer(IndexReader reader)
{
if (Enclosing_Instance.termArrays.Count == 0)
// optimize zero-term case
return null;
TermPositions[] tps = new TermPositions[Enclosing_Instance.termArrays.Count];
for (int i = 0; i < tps.Length; i++)
{
Term[] terms = (Term[]) Enclosing_Instance.termArrays[i];
TermPositions p;
if (terms.Length > 1)
p = new MultipleTermPositions(reader, terms);
else
p = reader.TermPositions(terms[0]);
if (p == null)
return null;
tps[i] = p;
}
if (Enclosing_Instance.slop == 0)
return new ExactPhraseScorer(this, tps, Enclosing_Instance.GetPositions(), similarity, reader.Norms(Enclosing_Instance.field));
else
return new SloppyPhraseScorer(this, tps, Enclosing_Instance.GetPositions(), similarity, Enclosing_Instance.slop, reader.Norms(Enclosing_Instance.field));
}
public virtual Scorer Scorer(IndexReader reader)
{
if (Enclosing_Instance.terms.Count == 0)
// optimize zero-term case
return null;
TermPositions[] tps = new TermPositions[Enclosing_Instance.terms.Count];
for (int i = 0; i < Enclosing_Instance.terms.Count; i++)
{
TermPositions p = reader.TermPositions((Term) Enclosing_Instance.terms[i]);
if (p == null)
return null;
tps[i] = p;
}
if (Enclosing_Instance.slop == 0)
// optimize exact case
return new ExactPhraseScorer(this, tps, Enclosing_Instance.GetPositions(), similarity, reader.Norms(Enclosing_Instance.field));
else
return new SloppyPhraseScorer(this, tps, Enclosing_Instance.GetPositions(), similarity, Enclosing_Instance.slop, reader.Norms(Enclosing_Instance.field));
}
private void InitBlock(Lucene.Net.Index.IndexReader reader, SpanTermQuery enclosingInstance)
{
this.reader = reader;
this.enclosingInstance = enclosingInstance;
positions = reader.TermPositions(Enclosing_Instance.term);
}
internal SloppyPhraseScorer(Weight weight, TermPositions[] tps, int[] positions, Similarity similarity, int slop, byte[] norms)
: base(weight, tps, positions, similarity, norms)
{
this.slop = slop;
}
public PayloadTermSpanScorer(PayloadTermWeight enclosingInstance, TermSpans spans, Weight weight, Similarity similarity, byte[] norms)
: base(spans, weight, similarity, norms)
{
InitBlock(enclosingInstance);
positions = spans.GetPositions();
}
public PayloadTermSpanScorer(PayloadTermWeight enclosingInstance, TermSpans spans, Weight weight, Similarity similarity, byte[] norms) : base(spans, weight, similarity, norms)
{
InitBlock(enclosingInstance);
positions = spans.GetPositions();
}
public virtual void TestPayloadsPos0()
{
for (int x = 0; x < 2; x++)
{
Directory dir = new MockRAMDirectory();
IndexWriter writer = new IndexWriter(dir, new TestPayloadAnalyzer(), true, IndexWriter.MaxFieldLength.LIMITED);
if (x == 1)
{
writer.SetAllowMinus1Position();
}
Document doc = new Document();
System.IO.MemoryStream ms = new System.IO.MemoryStream();
System.IO.StreamWriter sw = new System.IO.StreamWriter(ms);
sw.Write("a a b c d e a f g h i j a b k k");
// flush to stream & reset it's position so it can be read
sw.Flush();
ms.Position = 0;
doc.Add(new Field("content", new System.IO.StreamReader(ms)));
writer.AddDocument(doc);
IndexReader r = writer.GetReader();
TermPositions tp = r.TermPositions(new Term("content", "a"));
int count = 0;
Assert.IsTrue(tp.Next());
// "a" occurs 4 times
Assert.AreEqual(4, tp.Freq());
int expected;
if (x == 1)
{
expected = System.Int32.MaxValue;
}
else
{
expected = 0;
}
Assert.AreEqual(expected, tp.NextPosition());
if (x == 1)
{
continue;
}
Assert.AreEqual(1, tp.NextPosition());
Assert.AreEqual(3, tp.NextPosition());
Assert.AreEqual(6, tp.NextPosition());
// only one doc has "a"
Assert.IsFalse(tp.Next());
IndexSearcher is_Renamed = new IndexSearcher(r);
SpanTermQuery stq1 = new SpanTermQuery(new Term("content", "a"));
SpanTermQuery stq2 = new SpanTermQuery(new Term("content", "k"));
SpanQuery[] sqs = new SpanQuery[] { stq1, stq2 };
SpanNearQuery snq = new SpanNearQuery(sqs, 30, false);
count = 0;
bool sawZero = false;
//System.out.println("\ngetPayloadSpans test");
Lucene.Net.Search.Spans.Spans pspans = snq.GetSpans(is_Renamed.GetIndexReader());
while (pspans.Next())
{
//System.out.println(pspans.doc() + " - " + pspans.start() + " - "+ pspans.end());
System.Collections.Generic.ICollection <byte[]> payloads = pspans.GetPayload();
sawZero |= pspans.Start() == 0;
for (System.Collections.IEnumerator it = payloads.GetEnumerator(); it.MoveNext();)
{
count++;
System.Object generatedAux2 = it.Current;
//System.out.println(new String((byte[]) it.next()));
}
}
Assert.AreEqual(5, count);
Assert.IsTrue(sawZero);
//System.out.println("\ngetSpans test");
Lucene.Net.Search.Spans.Spans spans = snq.GetSpans(is_Renamed.GetIndexReader());
count = 0;
sawZero = false;
while (spans.Next())
{
count++;
sawZero |= spans.Start() == 0;
//System.out.println(spans.doc() + " - " + spans.start() + " - " + spans.end());
}
Assert.AreEqual(4, count);
Assert.IsTrue(sawZero);
//System.out.println("\nPayloadSpanUtil test");
sawZero = false;
PayloadSpanUtil psu = new PayloadSpanUtil(is_Renamed.GetIndexReader());
System.Collections.Generic.ICollection <byte[]> pls = psu.GetPayloadsForQuery(snq);
count = pls.Count;
for (System.Collections.IEnumerator it = pls.GetEnumerator(); it.MoveNext();)
{
System.String s = new System.String(System.Text.UTF8Encoding.UTF8.GetChars((byte[])it.Current));
//System.out.println(s);
sawZero |= s.Equals("pos: 0");
}
Assert.AreEqual(5, count);
Assert.IsTrue(sawZero);
writer.Close();
is_Renamed.GetIndexReader().Close();
dir.Close();
}
}
public virtual void TestSetPosition()
{
Analyzer analyzer = new AnonymousClassAnalyzer(this);
Directory store = new MockRAMDirectory();
IndexWriter writer = new IndexWriter(store, analyzer, true, IndexWriter.MaxFieldLength.LIMITED);
Document d = new Document();
d.Add(new Field("field", "bogus", Field.Store.YES, Field.Index.ANALYZED));
writer.AddDocument(d);
writer.Optimize();
writer.Close();
IndexSearcher searcher = new IndexSearcher(store);
TermPositions pos = searcher.GetIndexReader().TermPositions(new Term("field", "1"));
pos.Next();
// first token should be at position 0
Assert.AreEqual(0, pos.NextPosition());
pos = searcher.GetIndexReader().TermPositions(new Term("field", "2"));
pos.Next();
// second token should be at position 2
Assert.AreEqual(2, pos.NextPosition());
PhraseQuery q;
ScoreDoc[] hits;
q = new PhraseQuery();
q.Add(new Term("field", "1"));
q.Add(new Term("field", "2"));
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(0, hits.Length);
// same as previous, just specify positions explicitely.
q = new PhraseQuery();
q.Add(new Term("field", "1"), 0);
q.Add(new Term("field", "2"), 1);
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(0, hits.Length);
// specifying correct positions should find the phrase.
q = new PhraseQuery();
q.Add(new Term("field", "1"), 0);
q.Add(new Term("field", "2"), 2);
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(1, hits.Length);
q = new PhraseQuery();
q.Add(new Term("field", "2"));
q.Add(new Term("field", "3"));
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(1, hits.Length);
q = new PhraseQuery();
q.Add(new Term("field", "3"));
q.Add(new Term("field", "4"));
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(0, hits.Length);
// phrase query would find it when correct positions are specified.
q = new PhraseQuery();
q.Add(new Term("field", "3"), 0);
q.Add(new Term("field", "4"), 0);
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(1, hits.Length);
// phrase query should fail for non existing searched term
// even if there exist another searched terms in the same searched position.
q = new PhraseQuery();
q.Add(new Term("field", "3"), 0);
q.Add(new Term("field", "9"), 0);
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(0, hits.Length);
// multi-phrase query should succed for non existing searched term
// because there exist another searched terms in the same searched position.
MultiPhraseQuery mq = new MultiPhraseQuery();
mq.Add(new Term[] { new Term("field", "3"), new Term("field", "9") }, 0);
hits = searcher.Search(mq, null, 1000).scoreDocs;
Assert.AreEqual(1, hits.Length);
q = new PhraseQuery();
q.Add(new Term("field", "2"));
q.Add(new Term("field", "4"));
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(1, hits.Length);
q = new PhraseQuery();
q.Add(new Term("field", "3"));
q.Add(new Term("field", "5"));
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(1, hits.Length);
q = new PhraseQuery();
q.Add(new Term("field", "4"));
q.Add(new Term("field", "5"));
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(1, hits.Length);
q = new PhraseQuery();
q.Add(new Term("field", "2"));
q.Add(new Term("field", "5"));
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(0, hits.Length);
// should not find "1 2" because there is a gap of 1 in the index
QueryParser qp = new QueryParser("field", new StopWhitespaceAnalyzer(false));
q = (PhraseQuery)qp.Parse("\"1 2\"");
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(0, hits.Length);
// omitted stop word cannot help because stop filter swallows the increments.
q = (PhraseQuery)qp.Parse("\"1 stop 2\"");
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(0, hits.Length);
// query parser alone won't help, because stop filter swallows the increments.
qp.SetEnablePositionIncrements(true);
q = (PhraseQuery)qp.Parse("\"1 stop 2\"");
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(0, hits.Length);
// stop filter alone won't help, because query parser swallows the increments.
qp.SetEnablePositionIncrements(false);
q = (PhraseQuery)qp.Parse("\"1 stop 2\"");
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(0, hits.Length);
// when both qp qnd stopFilter propagate increments, we should find the doc.
qp = new QueryParser("field", new StopWhitespaceAnalyzer(true));
qp.SetEnablePositionIncrements(true);
q = (PhraseQuery)qp.Parse("\"1 stop 2\"");
hits = searcher.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(1, hits.Length);
}
public TermSpans(TermPositions positions, Term term)
{
this.positions = positions;
this.term = term;
doc = -1;
}
public TermSpans(TermPositions positions, Term term)
{
this.internalPositions = positions;
this.term = term;
internalDoc = -1;
}
