public virtual void TestCaching()
{
Directory dir = new RAMDirectory();
IndexWriter writer = new IndexWriter(dir, new SimpleAnalyzer(), IndexWriter.MaxFieldLength.LIMITED);
Document doc = new Document();
TokenStream stream = new AnonymousClassTokenStream(this);
stream = new CachingTokenFilter(stream);
doc.Add(new Field("preanalyzed", stream, TermVector.NO));
// 1) we consume all tokens twice before we add the doc to the index
checkTokens(stream);
stream.Reset();
checkTokens(stream);
// 2) now add the document to the index and verify if all tokens are indexed
// don't reset the stream here, the DocumentWriter should do that implicitly
writer.AddDocument(doc);
writer.Close();
IndexReader reader = IndexReader.Open(dir);
TermPositions termPositions = reader.TermPositions(new Term("preanalyzed", "term1"));
Assert.IsTrue(termPositions.Next());
Assert.AreEqual(1, termPositions.Freq());
Assert.AreEqual(0, termPositions.NextPosition());
termPositions.Seek(new Term("preanalyzed", "term2"));
Assert.IsTrue(termPositions.Next());
Assert.AreEqual(2, termPositions.Freq());
Assert.AreEqual(1, termPositions.NextPosition());
Assert.AreEqual(3, termPositions.NextPosition());
termPositions.Seek(new Term("preanalyzed", "term3"));
Assert.IsTrue(termPositions.Next());
Assert.AreEqual(1, termPositions.Freq());
Assert.AreEqual(2, termPositions.NextPosition());
reader.Close();
// 3) reset stream and consume tokens again
stream.Reset();
checkTokens(stream);
}
public virtual void TestCaching()
{
Directory dir = new RAMDirectory();
IndexWriter writer = new IndexWriter(dir, new SimpleAnalyzer());
Document doc = new Document();
TokenStream stream = new AnonymousClassTokenStream(this);
stream = new CachingTokenFilter(stream);
doc.Add(new Field("preanalyzed", stream, TermVector.NO));
// 1) we consume all tokens twice before we add the doc to the index
CheckTokens(stream);
stream.Reset();
CheckTokens(stream);
// 2) now add the document to the index and verify if all tokens are indexed
// don't reset the stream here, the DocumentWriter should do that implicitly
writer.AddDocument(doc);
writer.Close();
IndexReader reader = IndexReader.Open(dir);
TermPositions termPositions = reader.TermPositions(new Term("preanalyzed", "term1"));
Assert.IsTrue(termPositions.Next());
Assert.AreEqual(1, termPositions.Freq());
Assert.AreEqual(0, termPositions.NextPosition());
termPositions.Seek(new Term("preanalyzed", "term2"));
Assert.IsTrue(termPositions.Next());
Assert.AreEqual(2, termPositions.Freq());
Assert.AreEqual(1, termPositions.NextPosition());
Assert.AreEqual(3, termPositions.NextPosition());
termPositions.Seek(new Term("preanalyzed", "term3"));
Assert.IsTrue(termPositions.Next());
Assert.AreEqual(1, termPositions.Freq());
Assert.AreEqual(2, termPositions.NextPosition());
reader.Close();
// 3) reset stream and consume tokens again
stream.Reset();
CheckTokens(stream);
}
public virtual void TestNegativePositions()
{
TokenStream tokens = new AnonymousClassTokenStream(this);
MockRAMDirectory dir = new MockRAMDirectory();
IndexWriter w = new IndexWriter(dir, new WhitespaceAnalyzer(), true, IndexWriter.MaxFieldLength.UNLIMITED);
Document doc = new Document();
doc.Add(new Field("field", tokens));
w.AddDocument(doc);
w.Commit();
IndexSearcher s = new IndexSearcher(dir);
PhraseQuery pq = new PhraseQuery();
pq.Add(new Term("field", "a"));
pq.Add(new Term("field", "b"));
pq.Add(new Term("field", "c"));
ScoreDoc[] hits = s.Search(pq, null, 1000).scoreDocs;
Assert.AreEqual(1, hits.Length);
Query q = new SpanTermQuery(new Term("field", "a"));
hits = s.Search(q, null, 1000).scoreDocs;
Assert.AreEqual(1, hits.Length);
TermPositions tps = s.GetIndexReader().TermPositions(new Term("field", "a"));
Assert.IsTrue(tps.Next());
Assert.AreEqual(1, tps.Freq());
Assert.AreEqual(0, tps.NextPosition());
w.Close();
Assert.IsTrue(_TestUtil.CheckIndex(dir));
s.Close();
dir.Close();
}
