public virtual void TestFilterWorks()
{
Directory dir = new RAMDirectory();
IndexWriter writer = new IndexWriter(dir, new SimpleAnalyzer(), true, IndexWriter.MaxFieldLength.LIMITED);
for (int i = 0; i < 500; i++)
{
Document document = new Document();
document.Add(new Field("field", English.IntToEnglish(i) + " equals " + English.IntToEnglish(i), Field.Store.NO, Field.Index.ANALYZED));
writer.AddDocument(document);
}
writer.Close();
IndexReader reader = IndexReader.Open(dir, true);
SpanTermQuery query = new SpanTermQuery(new Term("field", English.IntToEnglish(10).Trim()));
SpanQueryFilter filter = new SpanQueryFilter(query);
SpanFilterResult result = filter.BitSpans(reader);
DocIdSet docIdSet = result.DocIdSet;
Assert.IsTrue(docIdSet != null, "docIdSet is null and it shouldn't be");
AssertContainsDocId("docIdSet doesn't contain docId 10", docIdSet, 10);
var spans = result.Positions;
Assert.IsTrue(spans != null, "spans is null and it shouldn't be");
int size = GetDocIdSetSize(docIdSet);
Assert.IsTrue(spans.Count == size, "spans Size: " + spans.Count + " is not: " + size);
for (System.Collections.IEnumerator iterator = spans.GetEnumerator(); iterator.MoveNext(); )
{
SpanFilterResult.PositionInfo info = (SpanFilterResult.PositionInfo) iterator.Current;
Assert.IsTrue(info != null, "info is null and it shouldn't be");
//The doc should indicate the bit is on
AssertContainsDocId("docIdSet doesn't contain docId " + info.Doc, docIdSet, info.Doc);
//There should be two positions in each
Assert.IsTrue(info.Positions.Count == 2, "info.getPositions() Size: " + info.Positions.Count + " is not: " + 2);
}
reader.Close();
}
public virtual void TestFilterWorks()
{
Directory dir = new RAMDirectory();
IndexWriter writer = new IndexWriter(dir, new SimpleAnalyzer(), true);
for (int i = 0; i < 500; i++)
{
Document document = new Document();
document.Add(new Field("field", English.IntToEnglish(i) + " equals " + English.IntToEnglish(i), Field.Store.NO, Field.Index.TOKENIZED));
writer.AddDocument(document);
}
writer.Close();
IndexReader reader = IndexReader.Open(dir);
SpanTermQuery query = new SpanTermQuery(new Term("field", English.IntToEnglish(10).Trim()));
SpanQueryFilter filter = new SpanQueryFilter(query);
SpanFilterResult result = filter.BitSpans(reader);
System.Collections.BitArray bits = result.GetBits();
Assert.IsTrue(bits != null, "bits is null and it shouldn't be");
Assert.IsTrue(bits.Get(10), "tenth bit is not on");
System.Collections.IList spans = result.GetPositions();
Assert.IsTrue(spans != null, "spans is null and it shouldn't be");
int cardinality = 0;
for (int i = 0; i < bits.Count; i++)
{
if (bits.Get(i)) cardinality++;
}
Assert.IsTrue(spans.Count == cardinality, "spans Size: " + spans.Count + " is not: " + cardinality);
for (System.Collections.IEnumerator iterator = spans.GetEnumerator(); iterator.MoveNext(); )
{
SpanFilterResult.PositionInfo info = (SpanFilterResult.PositionInfo) iterator.Current;
Assert.IsTrue(info != null, "info is null and it shouldn't be");
//The doc should indicate the bit is on
Assert.IsTrue(bits.Get(info.GetDoc()), "Bit is not on and it should be");
//There should be two positions in each
Assert.IsTrue(info.GetPositions().Count == 2, "info.getPositions() Size: " + info.GetPositions().Count + " is not: " + 2);
}
reader.Close();
}
public virtual void TestFilterWorks()
{
Directory dir = new RAMDirectory();
IndexWriter writer = new IndexWriter(dir, new SimpleAnalyzer(), true, IndexWriter.MaxFieldLength.LIMITED, null);
for (int i = 0; i < 500; i++)
{
Document document = new Document();
document.Add(new Field("field", English.IntToEnglish(i) + " equals " + English.IntToEnglish(i), Field.Store.NO, Field.Index.ANALYZED));
writer.AddDocument(document, null);
}
writer.Close();
IndexReader reader = IndexReader.Open(dir, true, null);
SpanTermQuery query = new SpanTermQuery(new Term("field", English.IntToEnglish(10).Trim()));
SpanQueryFilter filter = new SpanQueryFilter(query);
SpanFilterResult result = filter.BitSpans(reader, null);
DocIdSet docIdSet = result.DocIdSet;
Assert.IsTrue(docIdSet != null, "docIdSet is null and it shouldn't be");
AssertContainsDocId("docIdSet doesn't contain docId 10", docIdSet, 10);
var spans = result.Positions;
Assert.IsTrue(spans != null, "spans is null and it shouldn't be");
int size = GetDocIdSetSize(docIdSet);
Assert.IsTrue(spans.Count == size, "spans Size: " + spans.Count + " is not: " + size);
for (System.Collections.IEnumerator iterator = spans.GetEnumerator(); iterator.MoveNext();)
{
SpanFilterResult.PositionInfo info = (SpanFilterResult.PositionInfo)iterator.Current;
Assert.IsTrue(info != null, "info is null and it shouldn't be");
//The doc should indicate the bit is on
AssertContainsDocId("docIdSet doesn't contain docId " + info.Doc, docIdSet, info.Doc);
//There should be two positions in each
Assert.IsTrue(info.Positions.Count == 2, "info.getPositions() Size: " + info.Positions.Count + " is not: " + 2);
}
reader.Close();
}
public void TestEnforceDeletions()
{
Directory dir = new MockRAMDirectory();
IndexWriter writer = new IndexWriter(dir, new WhitespaceAnalyzer(), IndexWriter.MaxFieldLength.UNLIMITED);
IndexReader reader = writer.GetReader();
IndexSearcher searcher = new IndexSearcher(reader);
// add a doc, refresh the reader, and check that its there
Document doc = new Document();
doc.Add(new Field("id", "1", Field.Store.YES, Field.Index.NOT_ANALYZED));
writer.AddDocument(doc);
reader = RefreshReader(reader);
searcher = new IndexSearcher(reader);
TopDocs docs = searcher.Search(new MatchAllDocsQuery(), 1);
Assert.AreEqual(1, docs.TotalHits, "Should find a hit...");
SpanFilter startFilter = new SpanQueryFilter(new SpanTermQuery(new Term("id", "1")));
// ignore deletions
CachingSpanFilter filter = new CachingSpanFilter(startFilter, CachingWrapperFilter.DeletesMode.IGNORE);
docs = searcher.Search(new MatchAllDocsQuery(), filter, 1);
Assert.AreEqual(1, docs.TotalHits, "[query + filter] Should find a hit...");
ConstantScoreQuery constantScore = new ConstantScoreQuery(filter);
docs = searcher.Search(constantScore, 1);
Assert.AreEqual(1, docs.TotalHits, "[just filter] Should find a hit...");
// now delete the doc, refresh the reader, and see that it's not there
writer.DeleteDocuments(new Term("id", "1"));
reader = RefreshReader(reader);
searcher = new IndexSearcher(reader);
docs = searcher.Search(new MatchAllDocsQuery(), filter, 1);
Assert.AreEqual(0, docs.TotalHits, "[query + filter] Should *not* find a hit...");
docs = searcher.Search(constantScore, 1);
Assert.AreEqual(1, docs.TotalHits, "[just filter] Should find a hit...");
// force cache to regenerate:
filter = new CachingSpanFilter(startFilter, CachingWrapperFilter.DeletesMode.RECACHE);
writer.AddDocument(doc);
reader = RefreshReader(reader);
searcher = new IndexSearcher(reader);
docs = searcher.Search(new MatchAllDocsQuery(), filter, 1);
Assert.AreEqual(1, docs.TotalHits, "[query + filter] Should find a hit...");
constantScore = new ConstantScoreQuery(filter);
docs = searcher.Search(constantScore, 1);
Assert.AreEqual(1, docs.TotalHits, "[just filter] Should find a hit...");
// make sure we get a cache hit when we reopen readers
// that had no new deletions
IndexReader newReader = RefreshReader(reader);
Assert.IsTrue(reader != newReader);
reader = newReader;
searcher = new IndexSearcher(reader);
int missCount = filter.missCount;
docs = searcher.Search(constantScore, 1);
Assert.AreEqual(1, docs.TotalHits, "[just filter] Should find a hit...");
Assert.AreEqual(missCount, filter.missCount);
// now delete the doc, refresh the reader, and see that it's not there
writer.DeleteDocuments(new Term("id", "1"));
reader = RefreshReader(reader);
searcher = new IndexSearcher(reader);
docs = searcher.Search(new MatchAllDocsQuery(), filter, 1);
Assert.AreEqual(0, docs.TotalHits, "[query + filter] Should *not* find a hit...");
docs = searcher.Search(constantScore, 1);
Assert.AreEqual(0, docs.TotalHits, "[just filter] Should *not* find a hit...");
}
public void TestEnforceDeletions()
{
Directory dir = new MockRAMDirectory();
IndexWriter writer = new IndexWriter(dir, new WhitespaceAnalyzer(), IndexWriter.MaxFieldLength.UNLIMITED);
IndexReader reader = writer.GetReader();
IndexSearcher searcher = new IndexSearcher(reader);
// add a doc, refresh the reader, and check that its there
Document doc = new Document();
doc.Add(new Field("id", "1", Field.Store.YES, Field.Index.NOT_ANALYZED));
writer.AddDocument(doc);
reader = RefreshReader(reader);
searcher = new IndexSearcher(reader);
TopDocs docs = searcher.Search(new MatchAllDocsQuery(), 1);
Assert.AreEqual(1, docs.TotalHits, "Should find a hit...");
SpanFilter startFilter = new SpanQueryFilter(new SpanTermQuery(new Term("id", "1")));
// ignore deletions
CachingSpanFilter filter = new CachingSpanFilter(startFilter, CachingWrapperFilter.DeletesMode.IGNORE);
docs = searcher.Search(new MatchAllDocsQuery(), filter, 1);
Assert.AreEqual(1, docs.TotalHits, "[query + filter] Should find a hit...");
ConstantScoreQuery constantScore = new ConstantScoreQuery(filter);
docs = searcher.Search(constantScore, 1);
Assert.AreEqual(1, docs.TotalHits, "[just filter] Should find a hit...");
// now delete the doc, refresh the reader, and see that it's not there
writer.DeleteDocuments(new Term("id", "1"));
reader = RefreshReader(reader);
searcher = new IndexSearcher(reader);
docs = searcher.Search(new MatchAllDocsQuery(), filter, 1);
Assert.AreEqual(0, docs.TotalHits, "[query + filter] Should *not* find a hit...");
docs = searcher.Search(constantScore, 1);
Assert.AreEqual(1, docs.TotalHits, "[just filter] Should find a hit...");
// force cache to regenerate:
filter = new CachingSpanFilter(startFilter, CachingWrapperFilter.DeletesMode.RECACHE);
writer.AddDocument(doc);
reader = RefreshReader(reader);
searcher = new IndexSearcher(reader);
docs = searcher.Search(new MatchAllDocsQuery(), filter, 1);
Assert.AreEqual(1, docs.TotalHits, "[query + filter] Should find a hit...");
constantScore = new ConstantScoreQuery(filter);
docs = searcher.Search(constantScore, 1);
Assert.AreEqual(1, docs.TotalHits, "[just filter] Should find a hit...");
// make sure we get a cache hit when we reopen readers
// that had no new deletions
IndexReader newReader = RefreshReader(reader);
Assert.IsTrue(reader != newReader);
reader = newReader;
searcher = new IndexSearcher(reader);
int missCount = filter.missCount;
docs = searcher.Search(constantScore, 1);
Assert.AreEqual(1, docs.TotalHits, "[just filter] Should find a hit...");
Assert.AreEqual(missCount, filter.missCount);
// now delete the doc, refresh the reader, and see that it's not there
writer.DeleteDocuments(new Term("id", "1"));
reader = RefreshReader(reader);
searcher = new IndexSearcher(reader);
docs = searcher.Search(new MatchAllDocsQuery(), filter, 1);
Assert.AreEqual(0, docs.TotalHits, "[query + filter] Should *not* find a hit...");
docs = searcher.Search(constantScore, 1);
Assert.AreEqual(0, docs.TotalHits, "[just filter] Should *not* find a hit...");
}
