public virtual void TestNewestSegment_Mem()
{
Directory directory = NewDirectory();
IndexWriter writer = new IndexWriter(directory, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())));
Assert.IsNull(writer.NewestSegment());
writer.Dispose();
directory.Dispose();
}
public virtual void TestAddDocument()
{
Document testDoc = new Document();
DocHelper.SetupDoc(testDoc);
IndexWriter writer = new IndexWriter(Dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())));
writer.AddDocument(testDoc);
writer.Commit();
SegmentCommitInfo info = writer.NewestSegment();
writer.Dispose();
//After adding the document, we should be able to read it back in
SegmentReader reader = new SegmentReader(info, DirectoryReader.DEFAULT_TERMS_INDEX_DIVISOR, NewIOContext(Random()));
Assert.IsTrue(reader != null);
Document doc = reader.Document(0);
Assert.IsTrue(doc != null);
//System.out.println("Document: " + doc);
IndexableField[] fields = doc.GetFields("textField2");
Assert.IsTrue(fields != null && fields.Length == 1);
Assert.IsTrue(fields[0].StringValue.Equals(DocHelper.FIELD_2_TEXT));
Assert.IsTrue(fields[0].FieldType().StoreTermVectors);
fields = doc.GetFields("textField1");
Assert.IsTrue(fields != null && fields.Length == 1);
Assert.IsTrue(fields[0].StringValue.Equals(DocHelper.FIELD_1_TEXT));
Assert.IsFalse(fields[0].FieldType().StoreTermVectors);
fields = doc.GetFields("keyField");
Assert.IsTrue(fields != null && fields.Length == 1);
Assert.IsTrue(fields[0].StringValue.Equals(DocHelper.KEYWORD_TEXT));
fields = doc.GetFields(DocHelper.NO_NORMS_KEY);
Assert.IsTrue(fields != null && fields.Length == 1);
Assert.IsTrue(fields[0].StringValue.Equals(DocHelper.NO_NORMS_TEXT));
fields = doc.GetFields(DocHelper.TEXT_FIELD_3_KEY);
Assert.IsTrue(fields != null && fields.Length == 1);
Assert.IsTrue(fields[0].StringValue.Equals(DocHelper.FIELD_3_TEXT));
// test that the norms are not present in the segment if
// omitNorms is true
foreach (FieldInfo fi in reader.FieldInfos)
{
if (fi.Indexed)
{
Assert.IsTrue(fi.OmitsNorms() == (reader.GetNormValues(fi.Name) == null));
}
}
reader.Dispose();
}
private SegmentInfo IndexDoc(IndexWriter writer, System.String fileName)
{
System.IO.FileInfo file = new System.IO.FileInfo(System.IO.Path.Combine(workDir.FullName, fileName));
Document doc = FileDocument.Document(file);
writer.AddDocument(doc);
writer.Flush();
return writer.NewestSegment();
}
private SegmentInfo IndexDoc(IndexWriter writer, System.String fileName)
{
System.IO.DirectoryInfo file = new System.IO.DirectoryInfo(System.IO.Path.Combine(workDir.FullName, fileName));
Document doc = FileDocument.Document(file);
doc.Add(new Field("contents", new System.IO.StreamReader(file.FullName)));
writer.AddDocument(doc);
writer.Commit();
return writer.NewestSegment();
}
/// <summary>
/// Writes the document to the directory using the analyzer
/// and the similarity score; returns the SegmentInfo
/// describing the new segment
/// </summary>
public static SegmentCommitInfo WriteDoc(Random random, Directory dir, Analyzer analyzer, Similarity similarity, Document doc)
{
IndexWriter writer = new IndexWriter(dir, (new IndexWriterConfig(Util.LuceneTestCase.TEST_VERSION_CURRENT, analyzer)).SetSimilarity(similarity ?? IndexSearcher.DefaultSimilarity)); // LuceneTestCase.newIndexWriterConfig(random,
//writer.SetNoCFSRatio(0.0);
writer.AddDocument(doc);
writer.Commit();
SegmentCommitInfo info = writer.NewestSegment();
writer.Dispose();
return info;
}
public virtual void TestPreAnalyzedField()
{
IndexWriter writer = new IndexWriter(dir, new SimpleAnalyzer(), true, IndexWriter.MaxFieldLength.LIMITED);
Document doc = new Document();
doc.Add(new Field("preanalyzed", new AnonymousClassTokenStream(this), TermVector.NO));
writer.AddDocument(doc);
writer.Flush();
SegmentInfo info = writer.NewestSegment();
writer.Close();
SegmentReader reader = SegmentReader.Get(info);
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());
}
private SegmentCommitInfo IndexDoc(IndexWriter writer, string fileName)
{
FileInfo file = new FileInfo(Path.Combine(WorkDir.FullName, fileName));
Document doc = new Document();
StreamReader @is = new StreamReader(file.FullName);
doc.Add(new TextField("contents", @is));
writer.AddDocument(doc);
writer.Commit();
@is.Dispose();
return writer.NewestSegment();
}
public virtual void TestPositionIncrementGap()
{
Analyzer analyzer = new AnonymousClassAnalyzer(this);
IndexWriter writer = new IndexWriter(dir, analyzer, true, IndexWriter.MaxFieldLength.LIMITED);
Document doc = new Document();
doc.Add(new Field("repeated", "repeated one", Field.Store.YES, Field.Index.ANALYZED));
doc.Add(new Field("repeated", "repeated two", Field.Store.YES, Field.Index.ANALYZED));
writer.AddDocument(doc);
writer.Flush();
SegmentInfo info = writer.NewestSegment();
writer.Close();
SegmentReader reader = SegmentReader.Get(info);
TermPositions termPositions = reader.TermPositions(new Term("repeated", "repeated"));
Assert.IsTrue(termPositions.Next());
int freq = termPositions.Freq();
Assert.AreEqual(2, freq);
Assert.AreEqual(0, termPositions.NextPosition());
Assert.AreEqual(502, termPositions.NextPosition());
}
public override void SetUp()
{
base.SetUp();
/*
* for (int i = 0; i < testFields.Length; i++) {
* fieldInfos.Add(testFields[i], true, true, testFieldsStorePos[i], testFieldsStoreOff[i]);
* }
*/
Array.Sort(TestTerms);
int tokenUpto = 0;
for (int i = 0; i < TestTerms.Length; i++)
{
Positions[i] = new int[TERM_FREQ];
// first position must be 0
for (int j = 0; j < TERM_FREQ; j++)
{
// positions are always sorted in increasing order
Positions[i][j] = (int)(j * 10 + new Random(1).NextDouble() * 10);
TestToken token = Tokens[tokenUpto++] = new TestToken(this);
token.Text = TestTerms[i];
token.Pos = Positions[i][j];
token.StartOffset = j * 10;
token.EndOffset = j * 10 + TestTerms[i].Length;
}
}
Array.Sort(Tokens);
Dir = NewDirectory();
IndexWriter writer = new IndexWriter(Dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MyAnalyzer(this)).SetMaxBufferedDocs(-1).SetMergePolicy(NewLogMergePolicy(false, 10)).SetUseCompoundFile(false));
Document doc = new Document();
for (int i = 0; i < TestFields.Length; i++)
{
FieldType customType = new FieldType(TextField.TYPE_NOT_STORED);
if (TestFieldsStorePos[i] && TestFieldsStoreOff[i])
{
customType.StoreTermVectors = true;
customType.StoreTermVectorPositions = true;
customType.StoreTermVectorOffsets = true;
}
else if (TestFieldsStorePos[i] && !TestFieldsStoreOff[i])
{
customType.StoreTermVectors = true;
customType.StoreTermVectorPositions = true;
}
else if (!TestFieldsStorePos[i] && TestFieldsStoreOff[i])
{
customType.StoreTermVectors = true;
customType.StoreTermVectorOffsets = true;
}
else
{
customType.StoreTermVectors = true;
}
doc.Add(new Field(TestFields[i], "", customType));
}
//Create 5 documents for testing, they all have the same
//terms
for (int j = 0; j < 5; j++)
{
writer.AddDocument(doc);
}
writer.Commit();
Seg = writer.NewestSegment();
writer.Dispose();
FieldInfos = SegmentReader.ReadFieldInfos(Seg);
}
public void TestNewestSegment_Renamed()
{
RAMDirectory directory = new RAMDirectory();
IndexWriter writer = new IndexWriter(directory, new SimpleAnalyzer(), IndexWriter.MaxFieldLength.LIMITED);
Assert.IsNull(writer.NewestSegment());
}
public virtual void TestTokenReuse()
{
Analyzer analyzer = new AnalyzerAnonymousInnerClassHelper2(this);
IndexWriter writer = new IndexWriter(Dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, analyzer));
Document doc = new Document();
doc.Add(NewTextField("f1", "a 5 a a", Field.Store.YES));
writer.AddDocument(doc);
writer.Commit();
SegmentCommitInfo info = writer.NewestSegment();
writer.Dispose();
SegmentReader reader = new SegmentReader(info, DirectoryReader.DEFAULT_TERMS_INDEX_DIVISOR, NewIOContext(Random()));
DocsAndPositionsEnum termPositions = MultiFields.GetTermPositionsEnum(reader, reader.LiveDocs, "f1", new BytesRef("a"));
Assert.IsTrue(termPositions.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
int freq = termPositions.Freq();
Assert.AreEqual(3, freq);
Assert.AreEqual(0, termPositions.NextPosition());
Assert.IsNotNull(termPositions.Payload);
Assert.AreEqual(6, termPositions.NextPosition());
Assert.IsNull(termPositions.Payload);
Assert.AreEqual(7, termPositions.NextPosition());
Assert.IsNull(termPositions.Payload);
reader.Dispose();
}
public virtual void TestPreAnalyzedField()
{
IndexWriter writer = new IndexWriter(Dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random())));
Document doc = new Document();
doc.Add(new TextField("preanalyzed", new TokenStreamAnonymousInnerClassHelper(this)));
writer.AddDocument(doc);
writer.Commit();
SegmentCommitInfo info = writer.NewestSegment();
writer.Dispose();
SegmentReader reader = new SegmentReader(info, DirectoryReader.DEFAULT_TERMS_INDEX_DIVISOR, NewIOContext(Random()));
DocsAndPositionsEnum termPositions = reader.TermPositionsEnum(new Term("preanalyzed", "term1"));
Assert.IsTrue(termPositions.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(1, termPositions.Freq());
Assert.AreEqual(0, termPositions.NextPosition());
termPositions = reader.TermPositionsEnum(new Term("preanalyzed", "term2"));
Assert.IsTrue(termPositions.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(2, termPositions.Freq());
Assert.AreEqual(1, termPositions.NextPosition());
Assert.AreEqual(3, termPositions.NextPosition());
termPositions = reader.TermPositionsEnum(new Term("preanalyzed", "term3"));
Assert.IsTrue(termPositions.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
Assert.AreEqual(1, termPositions.Freq());
Assert.AreEqual(2, termPositions.NextPosition());
reader.Dispose();
}
public virtual void TestPositionIncrementGap()
{
Analyzer analyzer = new AnalyzerAnonymousInnerClassHelper(this);
IndexWriter writer = new IndexWriter(Dir, NewIndexWriterConfig(TEST_VERSION_CURRENT, analyzer));
Document doc = new Document();
doc.Add(NewTextField("repeated", "repeated one", Field.Store.YES));
doc.Add(NewTextField("repeated", "repeated two", Field.Store.YES));
writer.AddDocument(doc);
writer.Commit();
SegmentCommitInfo info = writer.NewestSegment();
writer.Dispose();
SegmentReader reader = new SegmentReader(info, DirectoryReader.DEFAULT_TERMS_INDEX_DIVISOR, NewIOContext(Random()));
DocsAndPositionsEnum termPositions = MultiFields.GetTermPositionsEnum(reader, MultiFields.GetLiveDocs(reader), "repeated", new BytesRef("repeated"));
Assert.IsTrue(termPositions.NextDoc() != DocIdSetIterator.NO_MORE_DOCS);
int freq = termPositions.Freq();
Assert.AreEqual(2, freq);
Assert.AreEqual(0, termPositions.NextPosition());
Assert.AreEqual(502, termPositions.NextPosition());
reader.Dispose();
}
public override void SetUp()
{
base.SetUp();
/*
for (int i = 0; i < testFields.length; i++) {
fieldInfos.add(testFields[i], true, true, testFieldsStorePos[i], testFieldsStoreOff[i]);
}
*/
System.Array.Sort(testTerms);
int tokenUpto = 0;
for (int i = 0; i < testTerms.Length; i++)
{
positions[i] = new int[TERM_FREQ];
offsets[i] = new TermVectorOffsetInfo[TERM_FREQ];
// first position must be 0
for (int j = 0; j < TERM_FREQ; j++)
{
// positions are always sorted in increasing order
positions[i][j] = (int) (j * 10 + (new System.Random().NextDouble()) * 10);
// offsets are always sorted in increasing order
offsets[i][j] = new TermVectorOffsetInfo(j * 10, j * 10 + testTerms[i].Length);
TestToken token = tokens[tokenUpto++] = new TestToken(this);
token.text = testTerms[i];
token.pos = positions[i][j];
token.startOffset = offsets[i][j].GetStartOffset();
token.endOffset = offsets[i][j].GetEndOffset();
}
}
System.Array.Sort(tokens);
IndexWriter writer = new IndexWriter(dir, new MyAnalyzer(this), true, IndexWriter.MaxFieldLength.LIMITED);
writer.SetUseCompoundFile(false);
Document doc = new Document();
for (int i = 0; i < testFields.Length; i++)
{
Field.TermVector tv;
if (testFieldsStorePos[i] && testFieldsStoreOff[i])
tv = Field.TermVector.WITH_POSITIONS_OFFSETS;
else if (testFieldsStorePos[i] && !testFieldsStoreOff[i])
tv = Field.TermVector.WITH_POSITIONS;
else if (!testFieldsStorePos[i] && testFieldsStoreOff[i])
tv = Field.TermVector.WITH_OFFSETS;
else
tv = Field.TermVector.YES;
doc.Add(new Field(testFields[i], "", Field.Store.NO, Field.Index.ANALYZED, tv));
}
//Create 5 documents for testing, they all have the same
//terms
for (int j = 0; j < 5; j++)
writer.AddDocument(doc);
writer.Flush();
seg = writer.NewestSegment().name;
writer.Close();
fieldInfos = new FieldInfos(dir, seg + "." + IndexFileNames.FIELD_INFOS_EXTENSION);
}
public virtual void TestLazyPerformance()
{
System.String tmpIODir = SupportClass.AppSettings.Get("tempDir", "");
System.String userName = System.Environment.UserName;
System.String path = tmpIODir + System.IO.Path.DirectorySeparatorChar.ToString() + "lazyDir" + userName;
System.IO.FileInfo file = new System.IO.FileInfo(path);
_TestUtil.RmDir(file);
FSDirectory tmpDir = FSDirectory.GetDirectory(file);
Assert.IsTrue(tmpDir != null);
IndexWriter writer = new IndexWriter(tmpDir, new WhitespaceAnalyzer(), true);
writer.SetUseCompoundFile(false);
writer.AddDocument(testDoc);
writer.Close();
segmentName = writer.NewestSegment().name;
Assert.IsTrue(fieldInfos != null);
FieldsReader reader;
long lazyTime = 0;
long regularTime = 0;
int length = 50;
System.Collections.Hashtable lazyFieldNames = new System.Collections.Hashtable();
lazyFieldNames.Add(DocHelper.LARGE_LAZY_FIELD_KEY, DocHelper.LARGE_LAZY_FIELD_KEY);
SetBasedFieldSelector fieldSelector = new SetBasedFieldSelector(new System.Collections.Hashtable(), lazyFieldNames);
for (int i = 0; i < length; i++)
{
reader = new FieldsReader(tmpDir, segmentName, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
Lucene.Net.Documents.Document doc;
doc = reader.Doc(0, null); //Load all of them
Assert.IsTrue(doc != null, "doc is null and it shouldn't be");
Fieldable field = doc.GetFieldable(DocHelper.LARGE_LAZY_FIELD_KEY);
Assert.IsTrue(field.IsLazy() == false, "field is lazy");
System.String value_Renamed;
long start;
long finish;
start = System.DateTime.Now.Millisecond;
//On my machine this was always 0ms.
value_Renamed = field.StringValue();
finish = System.DateTime.Now.Millisecond;
Assert.IsTrue(value_Renamed != null, "value is null and it shouldn't be");
Assert.IsTrue(field != null, "field is null and it shouldn't be");
regularTime += (finish - start);
reader.Close();
reader = null;
doc = null;
//Hmmm, are we still in cache???
System.GC.Collect();
reader = new FieldsReader(tmpDir, segmentName, fieldInfos);
doc = reader.Doc(0, fieldSelector);
field = doc.GetFieldable(DocHelper.LARGE_LAZY_FIELD_KEY);
Assert.IsTrue(field.IsLazy() == true, "field is not lazy");
start = System.DateTime.Now.Millisecond;
//On my machine this took around 50 - 70ms
value_Renamed = field.StringValue();
finish = System.DateTime.Now.Millisecond;
Assert.IsTrue(value_Renamed != null, "value is null and it shouldn't be");
lazyTime += (finish - start);
reader.Close();
}
System.Console.Out.WriteLine("Average Non-lazy time (should be very close to zero): " + regularTime / length + " ms for " + length + " reads");
System.Console.Out.WriteLine("Average Lazy Time (should be greater than zero): " + lazyTime / length + " ms for " + length + " reads");
}
public virtual void TestAddDocument()
{
Document testDoc = new Document();
DocHelper.SetupDoc(testDoc);
Analyzer analyzer = new WhitespaceAnalyzer();
IndexWriter writer = new IndexWriter(dir, analyzer, true, IndexWriter.MaxFieldLength.LIMITED);
writer.AddDocument(testDoc);
writer.Flush();
SegmentInfo info = writer.NewestSegment();
writer.Close();
//After adding the document, we should be able to read it back in
SegmentReader reader = SegmentReader.Get(info);
Assert.IsTrue(reader != null);
Document doc = reader.Document(0);
Assert.IsTrue(doc != null);
//System.out.println("Document: " + doc);
Fieldable[] fields = doc.GetFields("textField2");
Assert.IsTrue(fields != null && fields.Length == 1);
Assert.IsTrue(fields[0].StringValue().Equals(DocHelper.FIELD_2_TEXT));
Assert.IsTrue(fields[0].IsTermVectorStored());
fields = doc.GetFields("textField1");
Assert.IsTrue(fields != null && fields.Length == 1);
Assert.IsTrue(fields[0].StringValue().Equals(DocHelper.FIELD_1_TEXT));
Assert.IsFalse(fields[0].IsTermVectorStored());
fields = doc.GetFields("keyField");
Assert.IsTrue(fields != null && fields.Length == 1);
Assert.IsTrue(fields[0].StringValue().Equals(DocHelper.KEYWORD_TEXT));
fields = doc.GetFields(DocHelper.NO_NORMS_KEY);
Assert.IsTrue(fields != null && fields.Length == 1);
Assert.IsTrue(fields[0].StringValue().Equals(DocHelper.NO_NORMS_TEXT));
fields = doc.GetFields(DocHelper.TEXT_FIELD_3_KEY);
Assert.IsTrue(fields != null && fields.Length == 1);
Assert.IsTrue(fields[0].StringValue().Equals(DocHelper.FIELD_3_TEXT));
// test that the norms are not present in the segment if
// omitNorms is true
for (int i = 0; i < reader.core_ForNUnit.fieldInfos_ForNUnit.Size(); i++)
{
FieldInfo fi = reader.core_ForNUnit.fieldInfos_ForNUnit.FieldInfo(i);
if (fi.isIndexed_ForNUnit)
{
Assert.IsTrue(fi.omitNorms_ForNUnit == !reader.HasNorms(fi.name_ForNUnit));
}
}
}
public override void SetUp()
{
base.SetUp();
fieldInfos = new FieldInfos();
DocHelper.SetupDoc(testDoc);
fieldInfos.Add(testDoc);
IndexWriter writer = new IndexWriter(dir, new WhitespaceAnalyzer(), true);
writer.SetUseCompoundFile(false);
writer.AddDocument(testDoc);
writer.Close();
segmentName = writer.NewestSegment().name;
}
public virtual void TestTokenReuse()
{
Analyzer analyzer = new AnonymousClassAnalyzer1(this);
IndexWriter writer = new IndexWriter(dir, analyzer, true, IndexWriter.MaxFieldLength.LIMITED);
Document doc = new Document();
doc.Add(new Field("f1", "a 5 a a", Field.Store.YES, Field.Index.ANALYZED));
writer.AddDocument(doc);
writer.Flush();
SegmentInfo info = writer.NewestSegment();
writer.Close();
SegmentReader reader = SegmentReader.Get(info);
TermPositions termPositions = reader.TermPositions(new Term("f1", "a"));
Assert.IsTrue(termPositions.Next());
int freq = termPositions.Freq();
Assert.AreEqual(3, freq);
Assert.AreEqual(0, termPositions.NextPosition());
Assert.AreEqual(true, termPositions.IsPayloadAvailable());
Assert.AreEqual(6, termPositions.NextPosition());
Assert.AreEqual(false, termPositions.IsPayloadAvailable());
Assert.AreEqual(7, termPositions.NextPosition());
Assert.AreEqual(false, termPositions.IsPayloadAvailable());
}
/// <summary> Writes the document to the directory using the analyzer
/// and the similarity score; returns the SegmentInfo
/// describing the new segment
/// </summary>
/// <param name="dir">
/// </param>
/// <param name="analyzer">
/// </param>
/// <param name="similarity">
/// </param>
/// <param name="doc">
/// </param>
/// <throws> IOException </throws>
public static SegmentInfo WriteDoc(Directory dir, Analyzer analyzer, Similarity similarity, Document doc)
{
IndexWriter writer = new IndexWriter(dir, analyzer, IndexWriter.MaxFieldLength.LIMITED);
writer.SetSimilarity(similarity);
//writer.setUseCompoundFile(false);
writer.AddDocument(doc);
writer.Flush();
SegmentInfo info = writer.NewestSegment();
writer.Close();
return info;
}
public virtual void TestLiveChangeToCFS()
{
Directory dir = NewDirectory();
IndexWriterConfig iwc = new IndexWriterConfig(TEST_VERSION_CURRENT, new MockAnalyzer(Random()));
iwc.SetMergePolicy(NewLogMergePolicy(true));
// Start false:
iwc.SetUseCompoundFile(false);
iwc.MergePolicy.NoCFSRatio = 0.0d;
IndexWriter w = new IndexWriter(dir, iwc);
// Change to true:
w.Config.SetUseCompoundFile(true);
Document doc = new Document();
doc.Add(NewStringField("field", "foo", Store.NO));
w.AddDocument(doc);
w.Commit();
Assert.IsTrue(w.NewestSegment().Info.UseCompoundFile, "Expected CFS after commit");
doc.Add(NewStringField("field", "foo", Store.NO));
w.AddDocument(doc);
w.Commit();
w.ForceMerge(1);
w.Commit();
// no compound files after merge
Assert.IsFalse(w.NewestSegment().Info.UseCompoundFile, "Expected Non-CFS after merge");
MergePolicy lmp = w.Config.MergePolicy;
lmp.NoCFSRatio = 1.0;
lmp.MaxCFSSegmentSizeMB = double.PositiveInfinity;
w.AddDocument(doc);
w.ForceMerge(1);
w.Commit();
Assert.IsTrue(w.NewestSegment().Info.UseCompoundFile, "Expected CFS after merge");
w.Dispose();
dir.Dispose();
}
public override void SetUp()
{
base.SetUp();
/*
* for (int i = 0; i < testFields.length; i++) {
* fieldInfos.add(testFields[i], true, true, testFieldsStorePos[i], testFieldsStoreOff[i]);
* }
*/
System.Array.Sort(testTerms);
int tokenUpto = 0;
for (int i = 0; i < testTerms.Length; i++)
{
positions[i] = new int[TERM_FREQ];
offsets[i] = new TermVectorOffsetInfo[TERM_FREQ];
// first position must be 0
for (int j = 0; j < TERM_FREQ; j++)
{
// positions are always sorted in increasing order
positions[i][j] = (int)(j * 10 + (new System.Random().NextDouble()) * 10);
// offsets are always sorted in increasing order
offsets[i][j] = new TermVectorOffsetInfo(j * 10, j * 10 + testTerms[i].Length);
TestToken token = tokens[tokenUpto++] = new TestToken(this);
token.text = testTerms[i];
token.pos = positions[i][j];
token.startOffset = offsets[i][j].StartOffset;
token.endOffset = offsets[i][j].EndOffset;
}
}
System.Array.Sort(tokens);
IndexWriter writer = new IndexWriter(dir, new MyAnalyzer(this), true, IndexWriter.MaxFieldLength.LIMITED, null);
writer.UseCompoundFile = false;
Document doc = new Document();
for (int i = 0; i < testFields.Length; i++)
{
Field.TermVector tv;
if (testFieldsStorePos[i] && testFieldsStoreOff[i])
{
tv = Field.TermVector.WITH_POSITIONS_OFFSETS;
}
else if (testFieldsStorePos[i] && !testFieldsStoreOff[i])
{
tv = Field.TermVector.WITH_POSITIONS;
}
else if (!testFieldsStorePos[i] && testFieldsStoreOff[i])
{
tv = Field.TermVector.WITH_OFFSETS;
}
else
{
tv = Field.TermVector.YES;
}
doc.Add(new Field(testFields[i], "", Field.Store.NO, Field.Index.ANALYZED, tv));
}
//Create 5 documents for testing, they all have the same
//terms
for (int j = 0; j < 5; j++)
{
writer.AddDocument(doc, null);
}
writer.Commit(null);
seg = writer.NewestSegment().name;
writer.Close();
fieldInfos = new FieldInfos(dir, seg + "." + IndexFileNames.FIELD_INFOS_EXTENSION, null);
}