public virtual void TestToString()
{
StopAnalyzer analyzer = new StopAnalyzer();
StopFilter.SetEnablePositionIncrementsDefault(true);
QueryParser qp = new QueryParser("field", analyzer);
qp.SetEnablePositionIncrements(true);
PhraseQuery q = (PhraseQuery)qp.Parse("\"this hi this is a test is\"");
Assert.AreEqual("field:\"? hi ? ? ? test\"", q.ToString());
q.Add(new Term("field", "hello"), 1);
Assert.AreEqual("field:\"? hi|hello ? ? ? test\"", q.ToString());
}
public virtual void TestToString()
{
StopAnalyzer analyzer = new StopAnalyzer();
StopFilter.SetEnablePositionIncrementsDefault(true);
QueryParser qp = new QueryParser("field", analyzer);
qp.SetEnablePositionIncrements(true);
PhraseQuery q = (PhraseQuery) qp.Parse("\"this hi this is a test is\"");
Assert.AreEqual("field:\"? hi ? ? ? test\"", q.ToString());
q.Add(new Term("field", "hello"), 1);
Assert.AreEqual("field:\"? hi|hello ? ? ? test\"", q.ToString());
}
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 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 virtual void TestPositionIncrement()
{
bool dflt = StopFilter.GetEnablePositionIncrementsDefault();
StopFilter.SetEnablePositionIncrementsDefault(true);
try
{
Lucene.Net.QueryParsers.QueryParser qp = new Lucene.Net.QueryParsers.QueryParser("a", new StopAnalyzer(new System.String[] { "the", "in", "are", "this" }));
qp.SetEnablePositionIncrements(true);
System.String qtxt = "\"the words in poisitions pos02578 are stopped in this phrasequery\"";
// 0 2 5 7 8
int[] expectedPositions = new int[]{1, 3, 4, 6, 9};
PhraseQuery pq = (PhraseQuery) qp.Parse(qtxt);
//System.out.println("Query text: "+qtxt);
//System.out.println("Result: "+pq);
Term[] t = pq.GetTerms();
int[] pos = pq.GetPositions();
for (int i = 0; i < t.Length; i++)
{
//System.out.println(i+". "+t[i]+" pos: "+pos[i]);
Assert.AreEqual(expectedPositions[i], pos[i], "term " + i + " = " + t[i] + " has wrong term-position!");
}
}
finally
{
StopFilter.SetEnablePositionIncrementsDefault(dflt);
}
}
