private void CheckInvariants(IndexWriter writer)
{
_TestUtil.SyncConcurrentMerges(writer);
int maxBufferedDocs = writer.GetMaxBufferedDocs();
int mergeFactor = writer.GetMergeFactor();
int maxMergeDocs = writer.GetMaxMergeDocs();
int ramSegmentCount = writer.GetNumBufferedDocuments();
Assert.IsTrue(ramSegmentCount < maxBufferedDocs);
int lowerBound = -1;
int upperBound = maxBufferedDocs;
int numSegments = 0;
int segmentCount = writer.GetSegmentCount();
for (int i = segmentCount - 1; i >= 0; i--)
{
int docCount = writer.GetDocCount(i);
Assert.IsTrue(docCount > lowerBound);
if (docCount <= upperBound)
{
numSegments++;
}
else
{
if (upperBound * mergeFactor <= maxMergeDocs)
{
Assert.IsTrue(numSegments < mergeFactor);
}
do
{
lowerBound = upperBound;
upperBound *= mergeFactor;
}while (docCount > upperBound);
numSegments = 1;
}
}
if (upperBound * mergeFactor <= maxMergeDocs)
{
Assert.IsTrue(numSegments < mergeFactor);
}
System.String[] files = writer.GetDirectory().ListAll();
int segmentCfsCount = 0;
for (int i = 0; i < files.Length; i++)
{
if (files[i].EndsWith(".cfs"))
{
segmentCfsCount++;
}
}
Assert.AreEqual(segmentCount, segmentCfsCount);
}
/// <throws> IOException </throws>
/// <seealso cref="IndexModifier#SetMergeFactor(int)">
/// </seealso>
public virtual int GetMergeFactor()
{
lock (directory)
{
AssureOpen();
CreateIndexWriter();
return(indexWriter.GetMergeFactor());
}
}
private void CheckInvariants(IndexWriter writer)
{
writer.WaitForMerges();
int maxBufferedDocs = writer.GetMaxBufferedDocs();
int mergeFactor = writer.GetMergeFactor();
int maxMergeDocs = writer.GetMaxMergeDocs();
int ramSegmentCount = writer.GetNumBufferedDocuments();
Assert.IsTrue(ramSegmentCount < maxBufferedDocs);
int lowerBound = -1;
int upperBound = maxBufferedDocs;
int numSegments = 0;
int segmentCount = writer.GetSegmentCount();
for (int i = segmentCount - 1; i >= 0; i--)
{
int docCount = writer.GetDocCount(i);
Assert.IsTrue(docCount > lowerBound);
if (docCount <= upperBound)
{
numSegments++;
}
else
{
if (upperBound * mergeFactor <= maxMergeDocs)
{
Assert.IsTrue(numSegments < mergeFactor);
}
do
{
lowerBound = upperBound;
upperBound *= mergeFactor;
}while (docCount > upperBound);
numSegments = 1;
}
}
if (upperBound * mergeFactor <= maxMergeDocs)
{
Assert.IsTrue(numSegments < mergeFactor);
}
}
private void CheckInvariants(IndexWriter writer)
{
_TestUtil.SyncConcurrentMerges(writer);
int maxBufferedDocs = writer.GetMaxBufferedDocs();
int mergeFactor = writer.GetMergeFactor();
int maxMergeDocs = writer.GetMaxMergeDocs();
int ramSegmentCount = writer.GetNumBufferedDocuments();
Assert.IsTrue(ramSegmentCount < maxBufferedDocs);
int lowerBound = - 1;
int upperBound = maxBufferedDocs;
int numSegments = 0;
int segmentCount = writer.GetSegmentCount();
for (int i = segmentCount - 1; i >= 0; i--)
{
int docCount = writer.GetDocCount(i);
Assert.IsTrue(docCount > lowerBound);
if (docCount <= upperBound)
{
numSegments++;
}
else
{
if (upperBound * mergeFactor <= maxMergeDocs)
{
Assert.IsTrue(numSegments < mergeFactor);
}
do
{
lowerBound = upperBound;
upperBound *= mergeFactor;
}
while (docCount > upperBound);
numSegments = 1;
}
}
if (upperBound * mergeFactor <= maxMergeDocs)
{
Assert.IsTrue(numSegments < mergeFactor);
}
System.String[] files = writer.GetDirectory().ListAll();
int segmentCfsCount = 0;
for (int i = 0; i < files.Length; i++)
{
if (files[i].EndsWith(".cfs"))
{
segmentCfsCount++;
}
}
Assert.AreEqual(segmentCount, segmentCfsCount);
}
/// <summary> Tests the IndexReader.getFieldNames implementation</summary>
/// <throws> Exception on error </throws>
public virtual void TestGetFieldNames()
{
RAMDirectory d = new MockRAMDirectory();
// set up writer
IndexWriter writer = new IndexWriter(d, new StandardAnalyzer(), true);
AddDocumentWithFields(writer);
writer.Close();
// set up reader
IndexReader reader = IndexReader.Open(d);
System.Collections.ICollection fieldNames = reader.GetFieldNames(IndexReader.FieldOption.ALL);
Assert.IsTrue(CollectionContains(fieldNames, "keyword"));
Assert.IsTrue(CollectionContains(fieldNames, "text"));
Assert.IsTrue(CollectionContains(fieldNames, "unindexed"));
Assert.IsTrue(CollectionContains(fieldNames, "unstored"));
reader.Close();
// add more documents
writer = new IndexWriter(d, new StandardAnalyzer(), false);
// want to get some more segments here
for (int i = 0; i < 5 * writer.GetMergeFactor(); i++)
{
AddDocumentWithFields(writer);
}
// new fields are in some different segments (we hope)
for (int i = 0; i < 5 * writer.GetMergeFactor(); i++)
{
AddDocumentWithDifferentFields(writer);
}
// new termvector fields
for (int i = 0; i < 5 * writer.GetMergeFactor(); i++)
{
AddDocumentWithTermVectorFields(writer);
}
writer.Close();
// verify fields again
reader = IndexReader.Open(d);
fieldNames = reader.GetFieldNames(IndexReader.FieldOption.ALL);
Assert.AreEqual(13, fieldNames.Count); // the following fields
Assert.IsTrue(CollectionContains(fieldNames, "keyword"));
Assert.IsTrue(CollectionContains(fieldNames, "text"));
Assert.IsTrue(CollectionContains(fieldNames, "unindexed"));
Assert.IsTrue(CollectionContains(fieldNames, "unstored"));
Assert.IsTrue(CollectionContains(fieldNames, "keyword2"));
Assert.IsTrue(CollectionContains(fieldNames, "text2"));
Assert.IsTrue(CollectionContains(fieldNames, "unindexed2"));
Assert.IsTrue(CollectionContains(fieldNames, "unstored2"));
Assert.IsTrue(CollectionContains(fieldNames, "tvnot"));
Assert.IsTrue(CollectionContains(fieldNames, "termvector"));
Assert.IsTrue(CollectionContains(fieldNames, "tvposition"));
Assert.IsTrue(CollectionContains(fieldNames, "tvoffset"));
Assert.IsTrue(CollectionContains(fieldNames, "tvpositionoffset"));
// verify that only indexed fields were returned
fieldNames = reader.GetFieldNames(IndexReader.FieldOption.INDEXED);
Assert.AreEqual(11, fieldNames.Count); // 6 original + the 5 termvector fields
Assert.IsTrue(CollectionContains(fieldNames, "keyword"));
Assert.IsTrue(CollectionContains(fieldNames, "text"));
Assert.IsTrue(CollectionContains(fieldNames, "unstored"));
Assert.IsTrue(CollectionContains(fieldNames, "keyword2"));
Assert.IsTrue(CollectionContains(fieldNames, "text2"));
Assert.IsTrue(CollectionContains(fieldNames, "unstored2"));
Assert.IsTrue(CollectionContains(fieldNames, "tvnot"));
Assert.IsTrue(CollectionContains(fieldNames, "termvector"));
Assert.IsTrue(CollectionContains(fieldNames, "tvposition"));
Assert.IsTrue(CollectionContains(fieldNames, "tvoffset"));
Assert.IsTrue(CollectionContains(fieldNames, "tvpositionoffset"));
// verify that only unindexed fields were returned
fieldNames = reader.GetFieldNames(IndexReader.FieldOption.UNINDEXED);
Assert.AreEqual(2, fieldNames.Count); // the following fields
Assert.IsTrue(CollectionContains(fieldNames, "unindexed"));
Assert.IsTrue(CollectionContains(fieldNames, "unindexed2"));
// verify index term vector fields
fieldNames = reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR);
Assert.AreEqual(1, fieldNames.Count); // 1 field has term vector only
Assert.IsTrue(CollectionContains(fieldNames, "termvector"));
fieldNames = reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION);
Assert.AreEqual(1, fieldNames.Count); // 4 fields are indexed with term vectors
Assert.IsTrue(CollectionContains(fieldNames, "tvposition"));
fieldNames = reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_OFFSET);
Assert.AreEqual(1, fieldNames.Count); // 4 fields are indexed with term vectors
Assert.IsTrue(CollectionContains(fieldNames, "tvoffset"));
fieldNames = reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION_OFFSET);
Assert.AreEqual(1, fieldNames.Count); // 4 fields are indexed with term vectors
Assert.IsTrue(CollectionContains(fieldNames, "tvpositionoffset"));
reader.Close();
d.Close();
}
public virtual void TestTermVectors()
{
RAMDirectory d = new MockRAMDirectory();
// set up writer
IndexWriter writer = new IndexWriter(d, new StandardAnalyzer(), true);
// want to get some more segments here
// new termvector fields
for (int i = 0; i < 5 * writer.GetMergeFactor(); i++)
{
Document doc = new Document();
doc.Add(new Field("tvnot", "one two two three three three", Field.Store.YES, Field.Index.TOKENIZED, Field.TermVector.NO));
doc.Add(new Field("termvector", "one two two three three three", Field.Store.YES, Field.Index.TOKENIZED, Field.TermVector.YES));
doc.Add(new Field("tvoffset", "one two two three three three", Field.Store.YES, Field.Index.TOKENIZED, Field.TermVector.WITH_OFFSETS));
doc.Add(new Field("tvposition", "one two two three three three", Field.Store.YES, Field.Index.TOKENIZED, Field.TermVector.WITH_POSITIONS));
doc.Add(new Field("tvpositionoffset", "one two two three three three", Field.Store.YES, Field.Index.TOKENIZED, Field.TermVector.WITH_POSITIONS_OFFSETS));
writer.AddDocument(doc);
}
writer.Close();
IndexReader reader = IndexReader.Open(d);
FieldSortedTermVectorMapper mapper = new FieldSortedTermVectorMapper(new TermVectorEntryFreqSortedComparator());
reader.GetTermFreqVector(0, mapper);
System.Collections.IDictionary map = mapper.GetFieldToTerms();
Assert.IsTrue(map != null, "map is null and it shouldn't be");
Assert.IsTrue(map.Count == 4, "map Size: " + map.Count + " is not: " + 4);
System.Collections.IDictionary set_Renamed = (System.Collections.IDictionary) map["termvector"];
for (System.Collections.IEnumerator iterator = set_Renamed.Keys.GetEnumerator(); iterator.MoveNext(); )
{
TermVectorEntry entry = (TermVectorEntry) iterator.Current;
Assert.IsTrue(entry != null, "entry is null and it shouldn't be");
System.Console.Out.WriteLine("Entry: " + entry);
}
}
private void CheckInvariants(IndexWriter writer)
{
writer.WaitForMerges();
int maxBufferedDocs = writer.GetMaxBufferedDocs();
int mergeFactor = writer.GetMergeFactor();
int maxMergeDocs = writer.GetMaxMergeDocs();
int ramSegmentCount = writer.GetNumBufferedDocuments();
Assert.IsTrue(ramSegmentCount < maxBufferedDocs);
int lowerBound = - 1;
int upperBound = maxBufferedDocs;
int numSegments = 0;
int segmentCount = writer.GetSegmentCount();
for (int i = segmentCount - 1; i >= 0; i--)
{
int docCount = writer.GetDocCount(i);
Assert.IsTrue(docCount > lowerBound);
if (docCount <= upperBound)
{
numSegments++;
}
else
{
if (upperBound * mergeFactor <= maxMergeDocs)
{
Assert.IsTrue(numSegments < mergeFactor);
}
do
{
lowerBound = upperBound;
upperBound *= mergeFactor;
}
while (docCount > upperBound);
numSegments = 1;
}
}
if (upperBound * mergeFactor <= maxMergeDocs)
{
Assert.IsTrue(numSegments < mergeFactor);
}
}
