public virtual void TestLazyFieldsAfterClose()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos, null);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
ISet <string> loadFieldNames = Support.Compatibility.SetFactory.CreateHashSet <string>();
loadFieldNames.Add(DocHelper.TEXT_FIELD_1_KEY);
loadFieldNames.Add(DocHelper.TEXT_FIELD_UTF1_KEY);
ISet <string> lazyFieldNames = Support.Compatibility.SetFactory.CreateHashSet <string>();
lazyFieldNames.Add(DocHelper.LARGE_LAZY_FIELD_KEY);
lazyFieldNames.Add(DocHelper.LAZY_FIELD_KEY);
lazyFieldNames.Add(DocHelper.LAZY_FIELD_BINARY_KEY);
lazyFieldNames.Add(DocHelper.TEXT_FIELD_UTF2_KEY);
SetBasedFieldSelector fieldSelector = new SetBasedFieldSelector(loadFieldNames, lazyFieldNames);
Document doc = reader.Doc(0, fieldSelector, null);
Assert.IsTrue(doc != null, "doc is null and it shouldn't be");
IFieldable field = doc.GetFieldable(DocHelper.LAZY_FIELD_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy, "field is not lazy and it should be");
reader.Dispose();
Assert.Throws <AlreadyClosedException>(() => { var value = field.StringValue(null); },
"did not hit AlreadyClosedException as expected");
}
private void Initialize(SegmentInfo si)
{
segment = si.name;
// Use compound file directory for some files, if it exists
Directory cfsDir = Directory();
if (Directory().FileExists(segment + ".cfs"))
{
cfsReader = new CompoundFileReader(Directory(), segment + ".cfs");
cfsDir = cfsReader;
}
// No compound file exists - use the multi-file format
fieldInfos = new FieldInfos(cfsDir, segment + ".fnm");
fieldsReader = new FieldsReader(cfsDir, segment, fieldInfos);
tis = new TermInfosReader(cfsDir, segment, fieldInfos);
// NOTE: the bitvector is stored using the regular directory, not cfs
if (HasDeletions(si))
deletedDocs = new BitVector(Directory(), segment + ".del");
// make sure that all index files have been read or are kept open
// so that if an index update removes them we'll still have them
freqStream = cfsDir.OpenInput(segment + ".frq");
proxStream = cfsDir.OpenInput(segment + ".prx");
OpenNorms(cfsDir);
if (fieldInfos.HasVectors())
{
// open term vector files only as needed
termVectorsReaderOrig = new TermVectorsReader(cfsDir, segment, fieldInfos);
}
}
public virtual void TestLoadFirst()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
LoadFirstFieldSelector fieldSelector = new LoadFirstFieldSelector();
Document doc = reader.Doc(0, fieldSelector);
Assert.IsTrue(doc != null, "doc is null and it shouldn't be");
int count = 0;
System.Collections.IList l = doc.GetFields();
for (System.Collections.IEnumerator iter = l.GetEnumerator(); iter.MoveNext();)
{
Field field = (Field)iter.Current;
Assert.IsTrue(field != null, "field is null and it shouldn't be");
System.String sv = field.StringValue();
Assert.IsTrue(sv != null, "sv is null and it shouldn't be");
count++;
}
Assert.IsTrue(count == 1, count + " does not equal: " + 1);
}
public override void CheckIntegrity()
{
EnsureOpen();
// stored fields
FieldsReader.CheckIntegrity();
// term vectors
TermVectorsReader termVectorsReader = TermVectorsReader;
if (termVectorsReader != null)
{
termVectorsReader.CheckIntegrity();
}
// terms/postings
if (core.fields != null)
{
core.fields.CheckIntegrity();
}
// norms
if (core.normsProducer != null)
{
core.normsProducer.CheckIntegrity();
}
// docvalues
if (dvProducers != null)
{
foreach (DocValuesProducer producer in dvProducers)
{
producer.CheckIntegrity();
}
}
}
public LazyField(FieldsReader enclosingInstance, System.String name, Field.Store store, Field.Index index, Field.TermVector termVector, int toRead, long pointer) : base(name, store, index, termVector)
{
InitBlock(enclosingInstance);
this.toRead = toRead;
this.pointer = pointer;
lazy = true;
}
public virtual void TestLazyFields()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos, null);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
ISet <string> loadFieldNames = Support.Compatibility.SetFactory.CreateHashSet <string>();
loadFieldNames.Add(DocHelper.TEXT_FIELD_1_KEY);
loadFieldNames.Add(DocHelper.TEXT_FIELD_UTF1_KEY);
ISet <string> lazyFieldNames = Support.Compatibility.SetFactory.CreateHashSet <string>();
//new String[]{DocHelper.LARGE_LAZY_FIELD_KEY, DocHelper.LAZY_FIELD_KEY, DocHelper.LAZY_FIELD_BINARY_KEY};
lazyFieldNames.Add(DocHelper.LARGE_LAZY_FIELD_KEY);
lazyFieldNames.Add(DocHelper.LAZY_FIELD_KEY);
lazyFieldNames.Add(DocHelper.LAZY_FIELD_BINARY_KEY);
lazyFieldNames.Add(DocHelper.TEXT_FIELD_UTF2_KEY);
SetBasedFieldSelector fieldSelector = new SetBasedFieldSelector(loadFieldNames, lazyFieldNames);
Document doc = reader.Doc(0, fieldSelector, null);
Assert.IsTrue(doc != null, "doc is null and it shouldn't be");
IFieldable field = doc.GetFieldable(DocHelper.LAZY_FIELD_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy, "field is not lazy and it should be");
System.String value_Renamed = field.StringValue(null);
Assert.IsTrue(value_Renamed != null, "value is null and it shouldn't be");
Assert.IsTrue(value_Renamed.Equals(DocHelper.LAZY_FIELD_TEXT) == true, value_Renamed + " is not equal to " + DocHelper.LAZY_FIELD_TEXT);
field = doc.GetFieldable(DocHelper.TEXT_FIELD_1_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy == false, "Field is lazy and it should not be");
field = doc.GetFieldable(DocHelper.TEXT_FIELD_UTF1_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy == false, "Field is lazy and it should not be");
Assert.IsTrue(field.StringValue(null).Equals(DocHelper.FIELD_UTF1_TEXT) == true, field.StringValue(null) + " is not equal to " + DocHelper.FIELD_UTF1_TEXT);
field = doc.GetFieldable(DocHelper.TEXT_FIELD_UTF2_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy == true, "Field is lazy and it should not be");
Assert.IsTrue(field.StringValue(null).Equals(DocHelper.FIELD_UTF2_TEXT) == true, field.StringValue(null) + " is not equal to " + DocHelper.FIELD_UTF2_TEXT);
field = doc.GetFieldable(DocHelper.LAZY_FIELD_BINARY_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.StringValue(null) == null, "stringValue isn't null for lazy binary field");
byte[] bytes = field.GetBinaryValue(null);
Assert.IsTrue(bytes != null, "bytes is null and it shouldn't be");
Assert.IsTrue(DocHelper.LAZY_FIELD_BINARY_BYTES.Length == bytes.Length, "");
for (int i = 0; i < bytes.Length; i++)
{
Assert.IsTrue(bytes[i] == DocHelper.LAZY_FIELD_BINARY_BYTES[i], "byte[" + i + "] is mismatched");
}
}
public LazyField(FieldsReader enclosingInstance, System.String name, Field.Store store, Field.Index index, Field.TermVector termVector, int toRead, long pointer, bool isBinary) : base(name, store, index, termVector)
{
InitBlock(enclosingInstance);
this.toRead = toRead;
this.pointer = pointer;
this.isBinary = isBinary;
if (isBinary)
{
binaryLength = toRead;
}
lazy = true;
}
public LazyField(FieldsReader enclosingInstance, System.String name, Field.Store store, int toRead, long pointer, bool isBinary, bool isCompressed) : base(name, store, Field.Index.NO, Field.TermVector.NO)
{
InitBlock(enclosingInstance);
this.toRead = toRead;
this.pointer = pointer;
this.internalIsBinary = isBinary;
if (isBinary)
{
internalBinaryLength = toRead;
}
lazy = true;
this.isCompressed = isCompressed;
}
public virtual void TestLoadSize()
{
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos);
Document doc;
doc = reader.Doc(0, new AnonymousClassFieldSelector(this));
Fieldable f1 = doc.GetFieldable(DocHelper.TEXT_FIELD_1_KEY);
Fieldable f3 = doc.GetFieldable(DocHelper.TEXT_FIELD_3_KEY);
Fieldable fb = doc.GetFieldable(DocHelper.LAZY_FIELD_BINARY_KEY);
Assert.IsTrue(f1.IsBinary());
Assert.IsTrue(!f3.IsBinary());
Assert.IsTrue(fb.IsBinary());
AssertSizeEquals(2 * DocHelper.FIELD_1_TEXT.Length, f1.BinaryValue());
Assert.AreEqual(DocHelper.FIELD_3_TEXT, f3.StringValue());
AssertSizeEquals(DocHelper.LAZY_FIELD_BINARY_BYTES.Length, fb.BinaryValue());
reader.Close();
}
public virtual void Test()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
Document doc = reader.Doc(0, null);
Assert.IsTrue(doc != null);
Assert.IsTrue(doc.GetField(DocHelper.TEXT_FIELD_1_KEY) != null);
Fieldable field = doc.GetField(DocHelper.TEXT_FIELD_2_KEY);
Assert.IsTrue(field != null);
Assert.IsTrue(field.IsTermVectorStored() == true);
Assert.IsTrue(field.IsStoreOffsetWithTermVector() == true);
Assert.IsTrue(field.IsStorePositionWithTermVector() == true);
Assert.IsTrue(field.GetOmitNorms() == false);
Assert.IsTrue(field.GetOmitTf() == false);
field = doc.GetField(DocHelper.TEXT_FIELD_3_KEY);
Assert.IsTrue(field != null);
Assert.IsTrue(field.IsTermVectorStored() == false);
Assert.IsTrue(field.IsStoreOffsetWithTermVector() == false);
Assert.IsTrue(field.IsStorePositionWithTermVector() == false);
Assert.IsTrue(field.GetOmitNorms() == true);
Assert.IsTrue(field.GetOmitTf() == false);
field = doc.GetField(DocHelper.NO_TF_KEY);
Assert.IsTrue(field != null);
Assert.IsTrue(field.IsTermVectorStored() == false);
Assert.IsTrue(field.IsStoreOffsetWithTermVector() == false);
Assert.IsTrue(field.IsStorePositionWithTermVector() == false);
Assert.IsTrue(field.GetOmitNorms() == false);
Assert.IsTrue(field.GetOmitTf() == true);
reader.Close();
}
public virtual void Test()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
try
{
FieldsReader reader = new FieldsReader(dir, "test", fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
Document doc = reader.Doc(0);
Assert.IsTrue(doc != null);
Assert.IsTrue(doc.GetField("textField1") != null);
Field field = doc.GetField("textField2");
Assert.IsTrue(field != null);
Assert.IsTrue(field.IsTermVectorStored() == true);
reader.Close();
}
catch (System.IO.IOException e)
{
System.Console.Error.WriteLine(e.StackTrace);
Assert.IsTrue(false);
}
}
public virtual void Test()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
try
{
FieldsReader reader = new FieldsReader(dir, "test", fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
Document doc = reader.Doc(0);
Assert.IsTrue(doc != null);
Assert.IsTrue(doc.GetField("textField1") != null);
Field field = doc.GetField("textField2");
Assert.IsTrue(field != null);
Assert.IsTrue(field.IsTermVectorStored() == true);
reader.Close();
}
catch (System.IO.IOException e)
{
System.Console.Error.WriteLine(e.StackTrace);
Assert.IsTrue(false);
}
}
public virtual void TestLazyFieldsAfterClose()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
System.Collections.Hashtable loadFieldNames = new System.Collections.Hashtable();
SupportClass.CollectionsHelper.AddIfNotContains(loadFieldNames, DocHelper.TEXT_FIELD_1_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(loadFieldNames, DocHelper.TEXT_FIELD_UTF1_KEY);
System.Collections.Hashtable lazyFieldNames = new System.Collections.Hashtable();
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.LARGE_LAZY_FIELD_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.LAZY_FIELD_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.LAZY_FIELD_BINARY_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.TEXT_FIELD_UTF2_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.COMPRESSED_TEXT_FIELD_2_KEY);
SetBasedFieldSelector fieldSelector = new SetBasedFieldSelector(loadFieldNames, lazyFieldNames);
Document doc = reader.Doc(0, fieldSelector);
Assert.IsTrue(doc != null, "doc is null and it shouldn't be");
Fieldable field = doc.GetFieldable(DocHelper.LAZY_FIELD_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy(), "field is not lazy and it should be");
reader.Close();
try
{
field.StringValue();
Assert.Fail("did not hit AlreadyClosedException as expected");
}
catch (AlreadyClosedException e)
{
// expected
}
}
internal void OpenDocStores(SegmentInfo si)
{
lock (this)
{
System.Diagnostics.Debug.Assert(si.name.Equals(segment));
if (fieldsReaderOrig == null)
{
Directory storeDir;
if (si.GetDocStoreOffset() != - 1)
{
if (si.GetDocStoreIsCompoundFile())
{
System.Diagnostics.Debug.Assert(storeCFSReader == null);
storeCFSReader = new CompoundFileReader(dir, si.GetDocStoreSegment() + "." + IndexFileNames.COMPOUND_FILE_STORE_EXTENSION, readBufferSize);
storeDir = storeCFSReader;
System.Diagnostics.Debug.Assert(storeDir != null);
}
else
{
storeDir = dir;
System.Diagnostics.Debug.Assert(storeDir != null);
}
}
else if (si.GetUseCompoundFile())
{
// In some cases, we were originally opened when CFS
// was not used, but then we are asked to open doc
// stores after the segment has switched to CFS
if (cfsReader == null)
{
cfsReader = new CompoundFileReader(dir, segment + "." + IndexFileNames.COMPOUND_FILE_EXTENSION, readBufferSize);
}
storeDir = cfsReader;
System.Diagnostics.Debug.Assert(storeDir != null);
}
else
{
storeDir = dir;
System.Diagnostics.Debug.Assert(storeDir != null);
}
System.String storesSegment;
if (si.GetDocStoreOffset() != - 1)
{
storesSegment = si.GetDocStoreSegment();
}
else
{
storesSegment = segment;
}
fieldsReaderOrig = new FieldsReader(storeDir, storesSegment, fieldInfos, readBufferSize, si.GetDocStoreOffset(), si.docCount);
// Verify two sources of "maxDoc" agree:
if (si.GetDocStoreOffset() == - 1 && fieldsReaderOrig.Size() != si.docCount)
{
throw new CorruptIndexException("doc counts differ for segment " + segment + ": fieldsReader shows " + fieldsReaderOrig.Size() + " but segmentInfo shows " + si.docCount);
}
if (fieldInfos.HasVectors())
{
// open term vector files only as needed
termVectorsReaderOrig = new TermVectorsReader(storeDir, storesSegment, fieldInfos, readBufferSize, si.GetDocStoreOffset(), si.docCount);
}
}
}
}
public virtual void TestLazyFields()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
System.Collections.Hashtable loadFieldNames = new System.Collections.Hashtable();
SupportClass.CollectionsHelper.AddIfNotContains(loadFieldNames, DocHelper.TEXT_FIELD_1_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(loadFieldNames, DocHelper.TEXT_FIELD_UTF1_KEY);
System.Collections.Hashtable lazyFieldNames = new System.Collections.Hashtable();
//new String[]{DocHelper.LARGE_LAZY_FIELD_KEY, DocHelper.LAZY_FIELD_KEY, DocHelper.LAZY_FIELD_BINARY_KEY};
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.LARGE_LAZY_FIELD_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.LAZY_FIELD_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.LAZY_FIELD_BINARY_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.TEXT_FIELD_UTF2_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.COMPRESSED_TEXT_FIELD_2_KEY);
SetBasedFieldSelector fieldSelector = new SetBasedFieldSelector(loadFieldNames, lazyFieldNames);
Document doc = reader.Doc(0, fieldSelector);
Assert.IsTrue(doc != null, "doc is null and it shouldn't be");
Fieldable field = doc.GetFieldable(DocHelper.LAZY_FIELD_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy(), "field is not lazy and it should be");
System.String value_Renamed = field.StringValue();
Assert.IsTrue(value_Renamed != null, "value is null and it shouldn't be");
Assert.IsTrue(value_Renamed.Equals(DocHelper.LAZY_FIELD_TEXT) == true, value_Renamed + " is not equal to " + DocHelper.LAZY_FIELD_TEXT);
field = doc.GetFieldable(DocHelper.COMPRESSED_TEXT_FIELD_2_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy(), "field is not lazy and it should be");
Assert.IsTrue(field.BinaryValue() == null, "binaryValue isn't null for lazy string field");
value_Renamed = field.StringValue();
Assert.IsTrue(value_Renamed != null, "value is null and it shouldn't be");
Assert.IsTrue(value_Renamed.Equals(DocHelper.FIELD_2_COMPRESSED_TEXT) == true, value_Renamed + " is not equal to " + DocHelper.FIELD_2_COMPRESSED_TEXT);
field = doc.GetFieldable(DocHelper.TEXT_FIELD_1_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy() == false, "Field is lazy and it should not be");
field = doc.GetFieldable(DocHelper.TEXT_FIELD_UTF1_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy() == false, "Field is lazy and it should not be");
Assert.IsTrue(field.StringValue().Equals(DocHelper.FIELD_UTF1_TEXT) == true, field.StringValue() + " is not equal to " + DocHelper.FIELD_UTF1_TEXT);
field = doc.GetFieldable(DocHelper.TEXT_FIELD_UTF2_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy() == true, "Field is lazy and it should not be");
Assert.IsTrue(field.StringValue().Equals(DocHelper.FIELD_UTF2_TEXT) == true, field.StringValue() + " is not equal to " + DocHelper.FIELD_UTF2_TEXT);
field = doc.GetFieldable(DocHelper.LAZY_FIELD_BINARY_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.StringValue() == null, "stringValue isn't null for lazy binary field");
byte[] bytes = field.BinaryValue();
Assert.IsTrue(bytes != null, "bytes is null and it shouldn't be");
Assert.IsTrue(DocHelper.LAZY_FIELD_BINARY_BYTES.Length == bytes.Length, "");
for (int i = 0; i < bytes.Length; i++)
{
Assert.IsTrue(bytes[i] == DocHelper.LAZY_FIELD_BINARY_BYTES[i], "byte[" + i + "] is mismatched");
}
}
public virtual void Test()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
Document doc = reader.Doc(0, null);
Assert.IsTrue(doc != null);
Assert.IsTrue(doc.GetField(DocHelper.TEXT_FIELD_1_KEY) != null);
IFieldable field = doc.GetField(DocHelper.TEXT_FIELD_2_KEY);
Assert.IsTrue(field != null);
Assert.IsTrue(field.IsTermVectorStored == true);
Assert.IsTrue(field.IsStoreOffsetWithTermVector == true);
Assert.IsTrue(field.IsStorePositionWithTermVector == true);
Assert.IsTrue(field.OmitNorms == false);
Assert.IsTrue(field.OmitTermFreqAndPositions == false);
field = doc.GetField(DocHelper.TEXT_FIELD_3_KEY);
Assert.IsTrue(field != null);
Assert.IsTrue(field.IsTermVectorStored == false);
Assert.IsTrue(field.IsStoreOffsetWithTermVector == false);
Assert.IsTrue(field.IsStorePositionWithTermVector == false);
Assert.IsTrue(field.OmitNorms == true);
Assert.IsTrue(field.OmitTermFreqAndPositions == false);
field = doc.GetField(DocHelper.NO_TF_KEY);
Assert.IsTrue(field != null);
Assert.IsTrue(field.IsTermVectorStored == false);
Assert.IsTrue(field.IsStoreOffsetWithTermVector == false);
Assert.IsTrue(field.IsStorePositionWithTermVector == false);
Assert.IsTrue(field.OmitNorms == false);
Assert.IsTrue(field.OmitTermFreqAndPositions == true);
reader.Dispose();
}
private void InitBlock(FieldsReader enclosingInstance)
{
this.enclosingInstance = enclosingInstance;
}
private int CopyFieldsWithDeletions(FieldSelector fieldSelectorMerge, FieldsWriter fieldsWriter, IndexReader reader, FieldsReader matchingFieldsReader)
{
int docCount = 0;
int maxDoc = reader.MaxDoc();
if (matchingFieldsReader != null)
{
// We can bulk-copy because the fieldInfos are "congruent"
for (int j = 0; j < maxDoc; )
{
if (reader.IsDeleted(j))
{
// skip deleted docs
++j;
continue;
}
// We can optimize this case (doing a bulk byte copy) since the field
// numbers are identical
int start = j, numDocs = 0;
do
{
j++;
numDocs++;
if (j >= maxDoc)
break;
if (reader.IsDeleted(j))
{
j++;
break;
}
}
while (numDocs < MAX_RAW_MERGE_DOCS);
IndexInput stream = matchingFieldsReader.RawDocs(rawDocLengths, start, numDocs);
fieldsWriter.AddRawDocuments(stream, rawDocLengths, numDocs);
docCount += numDocs;
checkAbort.Work(300 * numDocs);
}
}
else
{
for (int j = 0; j < maxDoc; j++)
{
if (reader.IsDeleted(j))
{
// skip deleted docs
continue;
}
// NOTE: it's very important to first assign to doc then pass it to
// termVectorsWriter.addAllDocVectors; see LUCENE-1282
Document doc = reader.Document(j, fieldSelectorMerge);
fieldsWriter.AddDocument(doc);
docCount++;
checkAbort.Work(300);
}
}
return docCount;
}
public LazyField(FieldsReader enclosingInstance, System.String name, Field.Store store, int toRead, long pointer, bool isBinary):base(name, store, Field.Index.NO, Field.TermVector.NO)
{
InitBlock(enclosingInstance);
this.toRead = toRead;
this.pointer = pointer;
this.isBinary = isBinary;
if (isBinary)
binaryLength = toRead;
lazy = true;
}
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.Open(file);
Assert.IsTrue(tmpDir != null);
IndexWriter writer = new IndexWriter(tmpDir, new WhitespaceAnalyzer(), true, IndexWriter.MaxFieldLength.LIMITED);
writer.SetUseCompoundFile(false);
writer.AddDocument(testDoc);
writer.Close();
Assert.IsTrue(fieldInfos != null);
FieldsReader reader;
long lazyTime = 0;
long regularTime = 0;
int length = 50;
System.Collections.Hashtable lazyFieldNames = new System.Collections.Hashtable();
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, 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, TEST_SEGMENT_NAME, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
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 = (DateTime.Now.Ticks / TimeSpan.TicksPerMillisecond);
//On my machine this was always 0ms.
value_Renamed = field.StringValue();
finish = (DateTime.Now.Ticks / TimeSpan.TicksPerMillisecond);
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, TEST_SEGMENT_NAME, fieldInfos);
doc = reader.Doc(0, fieldSelector);
field = doc.GetFieldable(DocHelper.LARGE_LAZY_FIELD_KEY);
Assert.IsTrue(field.IsLazy() == true, "field is not lazy");
start = (DateTime.Now.Ticks / TimeSpan.TicksPerMillisecond);
//On my machine this took around 50 - 70ms
value_Renamed = field.StringValue();
finish = (DateTime.Now.Ticks / TimeSpan.TicksPerMillisecond);
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");
}
private void Initialize(SegmentInfo si)
{
segment = si.name;
this.si = si;
bool success = false;
try
{
// Use compound file directory for some files, if it exists
Directory cfsDir = Directory();
if (si.GetUseCompoundFile())
{
cfsReader = new CompoundFileReader(Directory(), segment + ".cfs");
cfsDir = cfsReader;
}
// No compound file exists - use the multi-file format
fieldInfos = new FieldInfos(cfsDir, segment + ".fnm");
fieldsReader = new FieldsReader(cfsDir, segment, fieldInfos);
// Verify two sources of "maxDoc" agree:
if (fieldsReader.Size() != si.docCount)
{
throw new System.SystemException("doc counts differ for segment " + si.name + ": fieldsReader shows " + fieldsReader.Size() + " but segmentInfo shows " + si.docCount);
}
tis = new TermInfosReader(cfsDir, segment, fieldInfos);
// NOTE: the bitvector is stored using the regular directory, not cfs
if (HasDeletions(si))
{
deletedDocs = new BitVector(Directory(), si.GetDelFileName());
// Verify # deletes does not exceed maxDoc for this segment:
if (deletedDocs.Count() > MaxDoc())
{
throw new System.SystemException("number of deletes (" + deletedDocs.Count() + ") exceeds max doc (" + MaxDoc() + ") for segment " + si.name);
}
}
// make sure that all index files have been read or are kept open
// so that if an index update removes them we'll still have them
freqStream = cfsDir.OpenInput(segment + ".frq");
proxStream = cfsDir.OpenInput(segment + ".prx");
OpenNorms(cfsDir);
if (fieldInfos.HasVectors())
{
// open term vector files only as needed
termVectorsReaderOrig = new TermVectorsReader(cfsDir, segment, fieldInfos);
}
success = true;
}
finally
{
// With lock-less commits, it's entirely possible (and
// fine) to hit a FileNotFound exception above. In
// this case, we want to explicitly close any subset
// of things that were opened so that we don't have to
// wait for a GC to do so.
if (!success)
{
DoClose();
}
}
}
public virtual void TestLazyFields()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
System.Collections.Hashtable loadFieldNames = new System.Collections.Hashtable();
SupportClass.CollectionsHelper.AddIfNotContains(loadFieldNames, DocHelper.TEXT_FIELD_1_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(loadFieldNames, DocHelper.TEXT_FIELD_UTF1_KEY);
System.Collections.Hashtable lazyFieldNames = new System.Collections.Hashtable();
//new String[]{DocHelper.LARGE_LAZY_FIELD_KEY, DocHelper.LAZY_FIELD_KEY, DocHelper.LAZY_FIELD_BINARY_KEY};
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.LARGE_LAZY_FIELD_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.LAZY_FIELD_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.LAZY_FIELD_BINARY_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.TEXT_FIELD_UTF2_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.COMPRESSED_TEXT_FIELD_2_KEY);
SetBasedFieldSelector fieldSelector = new SetBasedFieldSelector(loadFieldNames, lazyFieldNames);
Document doc = reader.Doc(0, fieldSelector);
Assert.IsTrue(doc != null, "doc is null and it shouldn't be");
Fieldable field = doc.GetFieldable(DocHelper.LAZY_FIELD_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy(), "field is not lazy and it should be");
System.String value_Renamed = field.StringValue();
Assert.IsTrue(value_Renamed != null, "value is null and it shouldn't be");
Assert.IsTrue(value_Renamed.Equals(DocHelper.LAZY_FIELD_TEXT) == true, value_Renamed + " is not equal to " + DocHelper.LAZY_FIELD_TEXT);
field = doc.GetFieldable(DocHelper.COMPRESSED_TEXT_FIELD_2_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy(), "field is not lazy and it should be");
Assert.IsTrue(field.BinaryValue() == null, "binaryValue isn't null for lazy string field");
value_Renamed = field.StringValue();
Assert.IsTrue(value_Renamed != null, "value is null and it shouldn't be");
Assert.IsTrue(value_Renamed.Equals(DocHelper.FIELD_2_COMPRESSED_TEXT) == true, value_Renamed + " is not equal to " + DocHelper.FIELD_2_COMPRESSED_TEXT);
field = doc.GetFieldable(DocHelper.TEXT_FIELD_1_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy() == false, "Field is lazy and it should not be");
field = doc.GetFieldable(DocHelper.TEXT_FIELD_UTF1_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy() == false, "Field is lazy and it should not be");
Assert.IsTrue(field.StringValue().Equals(DocHelper.FIELD_UTF1_TEXT) == true, field.StringValue() + " is not equal to " + DocHelper.FIELD_UTF1_TEXT);
field = doc.GetFieldable(DocHelper.TEXT_FIELD_UTF2_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy() == true, "Field is lazy and it should not be");
Assert.IsTrue(field.StringValue().Equals(DocHelper.FIELD_UTF2_TEXT) == true, field.StringValue() + " is not equal to " + DocHelper.FIELD_UTF2_TEXT);
field = doc.GetFieldable(DocHelper.LAZY_FIELD_BINARY_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.StringValue() == null, "stringValue isn't null for lazy binary field");
byte[] bytes = field.BinaryValue();
Assert.IsTrue(bytes != null, "bytes is null and it shouldn't be");
Assert.IsTrue(DocHelper.LAZY_FIELD_BINARY_BYTES.Length == bytes.Length, "");
for (int i = 0; i < bytes.Length; i++)
{
Assert.IsTrue(bytes[i] == DocHelper.LAZY_FIELD_BINARY_BYTES[i], "byte[" + i + "] is mismatched");
}
}
private void Initialize(SegmentInfo si)
{
segment = si.name;
// Use compound file directory for some files, if it exists
Directory cfsDir = Directory();
if (Directory().FileExists(segment + ".cfs"))
{
cfsReader = new CompoundFileReader(Directory(), segment + ".cfs");
cfsDir = cfsReader;
}
// No compound file exists - use the multi-file format
fieldInfos = new FieldInfos(cfsDir, segment + ".fnm");
fieldsReader = new FieldsReader(cfsDir, segment, fieldInfos);
tis = new TermInfosReader(cfsDir, segment, fieldInfos);
// NOTE: the bitvector is stored using the regular directory, not cfs
if (HasDeletions(si))
deletedDocs = new BitVector(Directory(), segment + ".del");
// make sure that all index files have been read or are kept open
// so that if an index update removes them we'll still have them
freqStream = cfsDir.OpenInput(segment + ".frq");
proxStream = cfsDir.OpenInput(segment + ".prx");
OpenNorms(cfsDir);
if (fieldInfos.HasVectors())
{
// open term vector files only as needed
termVectorsReaderOrig = new TermVectorsReader(cfsDir, segment, fieldInfos);
}
}
public LazyField(FieldsReader enclosingInstance, System.String name, Field.Store store, Field.Index index, Field.TermVector termVector, int toRead, long pointer):base(name, store, index, termVector)
{
InitBlock(enclosingInstance);
this.toRead = toRead;
this.pointer = pointer;
lazy = true;
}
private void assertCompressedFields29(Directory dir, bool shouldStillBeCompressed)
{
int count = 0;
int TEXT_PLAIN_LENGTH = TEXT_TO_COMPRESS.Length * 2;
// FieldSelectorResult.SIZE returns 2*number_of_chars for String fields:
int BINARY_PLAIN_LENGTH = BINARY_TO_COMPRESS.Length;
IndexReader reader = IndexReader.Open(dir, true);
try
{
// look into sub readers and check if raw merge is on/off
var readers = new System.Collections.Generic.List <IndexReader>();
ReaderUtil.GatherSubReaders(readers, reader);
foreach (IndexReader ir in readers)
{
FieldsReader fr = ((SegmentReader)ir).GetFieldsReader();
Assert.IsTrue(shouldStillBeCompressed != fr.CanReadRawDocs(),
"for a 2.9 index, FieldsReader.canReadRawDocs() must be false and other way round for a trunk index");
}
// test that decompression works correctly
for (int i = 0; i < reader.MaxDoc; i++)
{
if (!reader.IsDeleted(i))
{
Document d = reader.Document(i);
if (d.Get("content3") != null)
{
continue;
}
count++;
IFieldable compressed = d.GetFieldable("compressed");
if (int.Parse(d.Get("id")) % 2 == 0)
{
Assert.IsFalse(compressed.IsBinary);
Assert.AreEqual(TEXT_TO_COMPRESS, compressed.StringValue,
"incorrectly decompressed string");
}
else
{
Assert.IsTrue(compressed.IsBinary);
Assert.IsTrue(BINARY_TO_COMPRESS.SequenceEqual(compressed.GetBinaryValue()),
"incorrectly decompressed binary");
}
}
}
//check if field was decompressed after optimize
for (int i = 0; i < reader.MaxDoc; i++)
{
if (!reader.IsDeleted(i))
{
Document d = reader.Document(i, new AnonymousFieldSelector());
if (d.Get("content3") != null)
{
continue;
}
count++;
// read the size from the binary value using BinaryReader (this prevents us from doing the shift ops ourselves):
// ugh, Java uses Big-Endian streams, so we need to do it manually.
byte[] encodedSize = d.GetFieldable("compressed").GetBinaryValue().Take(4).Reverse().ToArray();
int actualSize = BitConverter.ToInt32(encodedSize, 0);
int compressedSize = int.Parse(d.Get("compressedSize"));
bool binary = int.Parse(d.Get("id")) % 2 > 0;
int shouldSize = shouldStillBeCompressed
? compressedSize
: (binary ? BINARY_PLAIN_LENGTH : TEXT_PLAIN_LENGTH);
Assert.AreEqual(shouldSize, actualSize, "size incorrect");
if (!shouldStillBeCompressed)
{
Assert.IsFalse(compressedSize == actualSize,
"uncompressed field should have another size than recorded in index");
}
}
}
Assert.AreEqual(34 * 2, count, "correct number of tests");
}
finally
{
reader.Dispose();
}
}
public virtual void TestLazyFieldsAfterClose()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
ISet<string> loadFieldNames = Support.Compatibility.SetFactory.GetSet<string>();
loadFieldNames.Add(DocHelper.TEXT_FIELD_1_KEY);
loadFieldNames.Add(DocHelper.TEXT_FIELD_UTF1_KEY);
ISet<string> lazyFieldNames = Support.Compatibility.SetFactory.GetSet<string>();
lazyFieldNames.Add(DocHelper.LARGE_LAZY_FIELD_KEY);
lazyFieldNames.Add(DocHelper.LAZY_FIELD_KEY);
lazyFieldNames.Add(DocHelper.LAZY_FIELD_BINARY_KEY);
lazyFieldNames.Add(DocHelper.TEXT_FIELD_UTF2_KEY);
SetBasedFieldSelector fieldSelector = new SetBasedFieldSelector(loadFieldNames, lazyFieldNames);
Document doc = reader.Doc(0, fieldSelector);
Assert.IsTrue(doc != null, "doc is null and it shouldn't be");
IFieldable field = doc.GetFieldable(DocHelper.LAZY_FIELD_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy, "field is not lazy and it should be");
reader.Dispose();
Assert.Throws<AlreadyClosedException>(() => { var value = field.StringValue; },
"did not hit AlreadyClosedException as expected");
}
public virtual void TestLazyPerformance()
{
System.String tmpIODir = AppSettings.Get("tempDir", "");
System.String userName = System.Environment.UserName;
System.String path = tmpIODir + System.IO.Path.DirectorySeparatorChar.ToString() + "lazyDir" + userName;
System.IO.DirectoryInfo file = new System.IO.DirectoryInfo(path);
_TestUtil.RmDir(file);
FSDirectory tmpDir = FSDirectory.Open(file);
Assert.IsTrue(tmpDir != null);
IndexWriter writer = new IndexWriter(tmpDir, new WhitespaceAnalyzer(), true, IndexWriter.MaxFieldLength.LIMITED);
writer.UseCompoundFile = false;
writer.AddDocument(testDoc);
writer.Close();
Assert.IsTrue(fieldInfos != null);
FieldsReader reader;
long lazyTime = 0;
long regularTime = 0;
int length = 50;
ISet<string> lazyFieldNames = Support.Compatibility.SetFactory.GetSet<string>();
lazyFieldNames.Add(DocHelper.LARGE_LAZY_FIELD_KEY);
SetBasedFieldSelector fieldSelector = new SetBasedFieldSelector(Support.Compatibility.SetFactory.GetSet<string>(), lazyFieldNames);
for (int i = 0; i < length; i++)
{
reader = new FieldsReader(tmpDir, TEST_SEGMENT_NAME, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
Document doc;
doc = reader.Doc(0, null); //Load all of them
Assert.IsTrue(doc != null, "doc is null and it shouldn't be");
IFieldable 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 = (DateTime.Now.Ticks / TimeSpan.TicksPerMillisecond);
//On my machine this was always 0ms.
value_Renamed = field.StringValue;
finish = (DateTime.Now.Ticks / TimeSpan.TicksPerMillisecond);
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.Dispose();
reader = null;
doc = null;
//Hmmm, are we still in cache???
System.GC.Collect();
reader = new FieldsReader(tmpDir, TEST_SEGMENT_NAME, fieldInfos);
doc = reader.Doc(0, fieldSelector);
field = doc.GetFieldable(DocHelper.LARGE_LAZY_FIELD_KEY);
Assert.IsTrue(field.IsLazy == true, "field is not lazy");
start = (DateTime.Now.Ticks / TimeSpan.TicksPerMillisecond);
//On my machine this took around 50 - 70ms
value_Renamed = field.StringValue;
finish = (DateTime.Now.Ticks / TimeSpan.TicksPerMillisecond);
Assert.IsTrue(value_Renamed != null, "value is null and it shouldn't be");
lazyTime += (finish - start);
reader.Dispose();
}
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 Test()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, segmentName, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
Lucene.Net.Documents.Document doc = reader.Doc(0, null);
Assert.IsTrue(doc != null);
Assert.IsTrue(doc.GetField(DocHelper.TEXT_FIELD_1_KEY) != null);
Field field = doc.GetField(DocHelper.TEXT_FIELD_2_KEY);
Assert.IsTrue(field != null);
Assert.IsTrue(field.IsTermVectorStored() == true);
Assert.IsTrue(field.IsStoreOffsetWithTermVector() == true);
Assert.IsTrue(field.IsStorePositionWithTermVector() == true);
Assert.IsTrue(field.GetOmitNorms() == false);
field = doc.GetField(DocHelper.TEXT_FIELD_3_KEY);
Assert.IsTrue(field != null);
Assert.IsTrue(field.IsTermVectorStored() == false);
Assert.IsTrue(field.IsStoreOffsetWithTermVector() == false);
Assert.IsTrue(field.IsStorePositionWithTermVector() == false);
Assert.IsTrue(field.GetOmitNorms() == true);
reader.Close();
}
public LazyField(FieldsReader enclosingInstance, System.String name, Field.Store store, Field.Index index, Field.TermVector termVector, int toRead, long pointer, bool isBinary, bool isCompressed)
: base(name, store, index, termVector)
{
InitBlock(enclosingInstance);
this.toRead = toRead;
this.pointer = pointer;
this.internalIsBinary = isBinary;
if (isBinary)
internalBinaryLength = toRead;
lazy = true;
this.isCompressed = isCompressed;
}
private int CopyFieldsNoDeletions(FieldSelector fieldSelectorMerge, FieldsWriter fieldsWriter, IndexReader reader, FieldsReader matchingFieldsReader)
{
int maxDoc = reader.MaxDoc();
int docCount = 0;
if (matchingFieldsReader != null)
{
// We can bulk-copy because the fieldInfos are "congruent"
while (docCount < maxDoc)
{
int len = System.Math.Min(MAX_RAW_MERGE_DOCS, maxDoc - docCount);
IndexInput stream = matchingFieldsReader.RawDocs(rawDocLengths, docCount, len);
fieldsWriter.AddRawDocuments(stream, rawDocLengths, len);
docCount += len;
checkAbort.Work(300 * len);
}
}
else
{
for (; docCount < maxDoc; docCount++)
{
// NOTE: it's very important to first assign to doc then pass it to
// termVectorsWriter.addAllDocVectors; see LUCENE-1282
Document doc = reader.Document(docCount, fieldSelectorMerge);
fieldsWriter.AddDocument(doc);
checkAbort.Work(300);
}
}
return docCount;
}
/// <summary> </summary>
/// <returns> The number of documents in all of the readers
/// </returns>
/// <throws> CorruptIndexException if the index is corrupt </throws>
/// <throws> IOException if there is a low-level IO error </throws>
private int MergeFields()
{
if (!mergeDocStores)
{
// When we are not merging by doc stores, that means
// all segments were written as part of a single
// autoCommit=false IndexWriter session, so their field
// name -> number mapping are the same. So, we start
// with the fieldInfos of the last segment in this
// case, to keep that numbering.
SegmentReader sr = (SegmentReader)readers[readers.Count - 1];
fieldInfos = (FieldInfos)sr.fieldInfos.Clone();
}
else
{
fieldInfos = new FieldInfos(); // merge field names
}
for (int i = 0; i < readers.Count; i++)
{
IndexReader reader = (IndexReader)readers[i];
if (reader is SegmentReader)
{
SegmentReader segmentReader = (SegmentReader)reader;
for (int j = 0; j < segmentReader.GetFieldInfos().Size(); j++)
{
FieldInfo fi = segmentReader.GetFieldInfos().FieldInfo(j);
fieldInfos.Add(fi.name, fi.isIndexed, fi.storeTermVector, fi.storePositionWithTermVector, fi.storeOffsetWithTermVector, !reader.HasNorms(fi.name), fi.storePayloads, fi.omitTf);
}
}
else
{
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION_OFFSET), true, true, true, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION), true, true, false, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_OFFSET), true, false, true, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR), true, false, false, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.OMIT_TF), false, false, false, false, true);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.STORES_PAYLOADS), false, false, false, true, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.INDEXED), false, false, false, false, false);
fieldInfos.Add(reader.GetFieldNames(IndexReader.FieldOption.UNINDEXED), false);
}
}
fieldInfos.Write(directory, segment + ".fnm");
int docCount = 0;
SetMatchingSegmentReaders();
if (mergeDocStores)
{
// for merging we don't want to compress/uncompress the data, so to tell the FieldsReader that we're
// in merge mode, we use this FieldSelector
FieldSelector fieldSelectorMerge = new AnonymousClassFieldSelector(this);
// merge field values
FieldsWriter fieldsWriter = new FieldsWriter(directory, segment, fieldInfos);
try
{
for (int i = 0; i < readers.Count; i++)
{
IndexReader reader = (IndexReader)readers[i];
SegmentReader matchingSegmentReader = matchingSegmentReaders[i];
FieldsReader matchingFieldsReader;
bool hasMatchingReader;
if (matchingSegmentReader != null)
{
FieldsReader fieldsReader = matchingSegmentReader.GetFieldsReader();
if (fieldsReader != null && !fieldsReader.CanReadRawDocs())
{
matchingFieldsReader = null;
hasMatchingReader = false;
}
else
{
matchingFieldsReader = fieldsReader;
hasMatchingReader = true;
}
}
else
{
hasMatchingReader = false;
matchingFieldsReader = null;
}
int maxDoc = reader.MaxDoc();
bool hasDeletions = reader.HasDeletions();
for (int j = 0; j < maxDoc;)
{
if (!hasDeletions || !reader.IsDeleted(j))
{ // skip deleted docs
if (hasMatchingReader)
{
// We can optimize this case (doing a bulk
// byte copy) since the field numbers are
// identical
int start = j;
int numDocs = 0;
do
{
j++;
numDocs++;
if (j >= maxDoc)
{
break;
}
if (hasDeletions && matchingSegmentReader.IsDeleted(j))
{
j++;
break;
}
} while (numDocs < MAX_RAW_MERGE_DOCS);
IndexInput stream = matchingFieldsReader.RawDocs(rawDocLengths, start, numDocs);
fieldsWriter.AddRawDocuments(stream, rawDocLengths, numDocs);
docCount += numDocs;
if (checkAbort != null)
{
checkAbort.Work(300 * numDocs);
}
}
else
{
// NOTE: it's very important to first assign
// to doc then pass it to
// termVectorsWriter.addAllDocVectors; see
// LUCENE-1282
Document doc = reader.Document(j, fieldSelectorMerge);
fieldsWriter.AddDocument(doc);
j++;
docCount++;
if (checkAbort != null)
{
checkAbort.Work(300);
}
}
}
else
{
j++;
}
}
}
}
finally
{
fieldsWriter.Close();
}
long fdxFileLength = directory.FileLength(segment + "." + IndexFileNames.FIELDS_INDEX_EXTENSION);
// {{dougsale-2.4.0}
// this shouldn't be a problem for us - if it is,
// then it's not a JRE bug...
//if (4+docCount*8 != fdxFileLength)
// // This is most likely a bug in Sun JRE 1.6.0_04/_05;
// // we detect that the bug has struck, here, and
// // throw an exception to prevent the corruption from
// // entering the index. See LUCENE-1282 for
// // details.
// throw new RuntimeException("mergeFields produced an invalid result: docCount is " + docCount + " but fdx file size is " + fdxFileLength + "; now aborting this merge to prevent index corruption");
}
else
{
// If we are skipping the doc stores, that means there
// are no deletions in any of these segments, so we
// just sum numDocs() of each segment to get total docCount
for (int i = 0; i < readers.Count; i++)
{
docCount += ((IndexReader)readers[i]).NumDocs();
}
}
return(docCount);
}
public virtual void TestLoadSize()
{
FieldsReader reader = new FieldsReader(dir, segmentName, fieldInfos);
Lucene.Net.Documents.Document doc;
doc = reader.Doc(0, new AnonymousClassFieldSelector(this));
Fieldable f1 = doc.GetFieldable(DocHelper.TEXT_FIELD_1_KEY);
Fieldable f3 = doc.GetFieldable(DocHelper.TEXT_FIELD_3_KEY);
Fieldable fb = doc.GetFieldable(DocHelper.LAZY_FIELD_BINARY_KEY);
Assert.IsTrue(f1.IsBinary());
Assert.IsTrue(!f3.IsBinary());
Assert.IsTrue(fb.IsBinary());
AssertSizeEquals(2 * DocHelper.FIELD_1_TEXT.Length, f1.BinaryValue());
Assert.AreEqual(DocHelper.FIELD_3_TEXT, f3.StringValue());
AssertSizeEquals(DocHelper.LAZY_FIELD_BINARY_BYTES.Length, fb.BinaryValue());
reader.Close();
}
private int CopyFieldsWithDeletions(FieldsWriter fieldsWriter, IndexReader reader, FieldsReader matchingFieldsReader)
{
int docCount = 0;
int maxDoc = reader.MaxDoc;
if (matchingFieldsReader != null)
{
// We can bulk-copy because the fieldInfos are "congruent"
for (int j = 0; j < maxDoc;)
{
if (reader.IsDeleted(j))
{
// skip deleted docs
++j;
continue;
}
// We can optimize this case (doing a bulk byte copy) since the field
// numbers are identical
int start = j, numDocs = 0;
do
{
j++;
numDocs++;
if (j >= maxDoc)
{
break;
}
if (reader.IsDeleted(j))
{
j++;
break;
}
}while (numDocs < MAX_RAW_MERGE_DOCS);
IndexInput stream = matchingFieldsReader.RawDocs(rawDocLengths, start, numDocs);
fieldsWriter.AddRawDocuments(stream, rawDocLengths, numDocs);
docCount += numDocs;
checkAbort.Work(300 * numDocs);
}
}
else
{
for (int j = 0; j < maxDoc; j++)
{
if (reader.IsDeleted(j))
{
// skip deleted docs
continue;
}
// NOTE: it's very important to first assign to doc then pass it to
// termVectorsWriter.addAllDocVectors; see LUCENE-1282
Document doc = reader.Document(j);
fieldsWriter.AddDocument(doc);
docCount++;
checkAbort.Work(300);
}
}
return(docCount);
}
private void Initialize(SegmentInfo si)
{
segment = si.name;
this.si = si;
bool success = false;
try
{
// Use compound file directory for some files, if it exists
Directory cfsDir = Directory();
if (si.GetUseCompoundFile())
{
cfsReader = new CompoundFileReader(Directory(), segment + ".cfs");
cfsDir = cfsReader;
}
// No compound file exists - use the multi-file format
fieldInfos = new FieldInfos(cfsDir, segment + ".fnm");
fieldsReader = new FieldsReader(cfsDir, segment, fieldInfos);
// Verify two sources of "maxDoc" agree:
if (fieldsReader.Size() != si.docCount)
{
throw new System.SystemException("doc counts differ for segment " + si.name + ": fieldsReader shows " + fieldsReader.Size() + " but segmentInfo shows " + si.docCount);
}
tis = new TermInfosReader(cfsDir, segment, fieldInfos);
// NOTE: the bitvector is stored using the regular directory, not cfs
if (HasDeletions(si))
{
deletedDocs = new BitVector(Directory(), si.GetDelFileName());
// Verify # deletes does not exceed maxDoc for this segment:
if (deletedDocs.Count() > MaxDoc())
{
throw new System.SystemException("number of deletes (" + deletedDocs.Count() + ") exceeds max doc (" + MaxDoc() + ") for segment " + si.name);
}
}
// make sure that all index files have been read or are kept open
// so that if an index update removes them we'll still have them
freqStream = cfsDir.OpenInput(segment + ".frq");
proxStream = cfsDir.OpenInput(segment + ".prx");
OpenNorms(cfsDir);
if (fieldInfos.HasVectors())
{
// open term vector files only as needed
termVectorsReaderOrig = new TermVectorsReader(cfsDir, segment, fieldInfos);
}
success = true;
}
finally
{
// With lock-less commits, it's entirely possible (and
// fine) to hit a FileNotFound exception above. In
// this case, we want to explicitly close any subset
// of things that were opened so that we don't have to
// wait for a GC to do so.
if (!success)
{
DoClose();
}
}
}
private int CopyFieldsNoDeletions(FieldsWriter fieldsWriter, IndexReader reader, FieldsReader matchingFieldsReader)
{
int maxDoc = reader.MaxDoc;
int docCount = 0;
if (matchingFieldsReader != null)
{
// We can bulk-copy because the fieldInfos are "congruent"
while (docCount < maxDoc)
{
int len = System.Math.Min(MAX_RAW_MERGE_DOCS, maxDoc - docCount);
IndexInput stream = matchingFieldsReader.RawDocs(rawDocLengths, docCount, len);
fieldsWriter.AddRawDocuments(stream, rawDocLengths, len);
docCount += len;
checkAbort.Work(300 * len);
}
}
else
{
for (; docCount < maxDoc; docCount++)
{
// NOTE: it's very important to first assign to doc then pass it to
// termVectorsWriter.addAllDocVectors; see LUCENE-1282
Document doc = reader.Document(docCount);
fieldsWriter.AddDocument(doc);
checkAbort.Work(300);
}
}
return(docCount);
}
public virtual void TestLazyFields()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
ISet<string> loadFieldNames = Support.Compatibility.SetFactory.GetSet<string>();
loadFieldNames.Add(DocHelper.TEXT_FIELD_1_KEY);
loadFieldNames.Add(DocHelper.TEXT_FIELD_UTF1_KEY);
ISet<string> lazyFieldNames = Support.Compatibility.SetFactory.GetSet<string>();
//new String[]{DocHelper.LARGE_LAZY_FIELD_KEY, DocHelper.LAZY_FIELD_KEY, DocHelper.LAZY_FIELD_BINARY_KEY};
lazyFieldNames.Add(DocHelper.LARGE_LAZY_FIELD_KEY);
lazyFieldNames.Add(DocHelper.LAZY_FIELD_KEY);
lazyFieldNames.Add(DocHelper.LAZY_FIELD_BINARY_KEY);
lazyFieldNames.Add(DocHelper.TEXT_FIELD_UTF2_KEY);
SetBasedFieldSelector fieldSelector = new SetBasedFieldSelector(loadFieldNames, lazyFieldNames);
Document doc = reader.Doc(0, fieldSelector);
Assert.IsTrue(doc != null, "doc is null and it shouldn't be");
IFieldable field = doc.GetFieldable(DocHelper.LAZY_FIELD_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy, "field is not lazy and it should be");
System.String value_Renamed = field.StringValue;
Assert.IsTrue(value_Renamed != null, "value is null and it shouldn't be");
Assert.IsTrue(value_Renamed.Equals(DocHelper.LAZY_FIELD_TEXT) == true, value_Renamed + " is not equal to " + DocHelper.LAZY_FIELD_TEXT);
field = doc.GetFieldable(DocHelper.TEXT_FIELD_1_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy == false, "Field is lazy and it should not be");
field = doc.GetFieldable(DocHelper.TEXT_FIELD_UTF1_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy == false, "Field is lazy and it should not be");
Assert.IsTrue(field.StringValue.Equals(DocHelper.FIELD_UTF1_TEXT) == true, field.StringValue + " is not equal to " + DocHelper.FIELD_UTF1_TEXT);
field = doc.GetFieldable(DocHelper.TEXT_FIELD_UTF2_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy == true, "Field is lazy and it should not be");
Assert.IsTrue(field.StringValue.Equals(DocHelper.FIELD_UTF2_TEXT) == true, field.StringValue + " is not equal to " + DocHelper.FIELD_UTF2_TEXT);
field = doc.GetFieldable(DocHelper.LAZY_FIELD_BINARY_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.StringValue == null, "stringValue isn't null for lazy binary field");
byte[] bytes = field.GetBinaryValue();
Assert.IsTrue(bytes != null, "bytes is null and it shouldn't be");
Assert.IsTrue(DocHelper.LAZY_FIELD_BINARY_BYTES.Length == bytes.Length, "");
for (int i = 0; i < bytes.Length; i++)
{
Assert.IsTrue(bytes[i] == DocHelper.LAZY_FIELD_BINARY_BYTES[i], "byte[" + i + "] is mismatched");
}
}
public virtual void TestLazyFieldsAfterClose()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
System.Collections.Hashtable loadFieldNames = new System.Collections.Hashtable();
SupportClass.CollectionsHelper.AddIfNotContains(loadFieldNames, DocHelper.TEXT_FIELD_1_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(loadFieldNames, DocHelper.TEXT_FIELD_UTF1_KEY);
System.Collections.Hashtable lazyFieldNames = new System.Collections.Hashtable();
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.LARGE_LAZY_FIELD_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.LAZY_FIELD_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.LAZY_FIELD_BINARY_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.TEXT_FIELD_UTF2_KEY);
SupportClass.CollectionsHelper.AddIfNotContains(lazyFieldNames, DocHelper.COMPRESSED_TEXT_FIELD_2_KEY);
SetBasedFieldSelector fieldSelector = new SetBasedFieldSelector(loadFieldNames, lazyFieldNames);
Document doc = reader.Doc(0, fieldSelector);
Assert.IsTrue(doc != null, "doc is null and it shouldn't be");
Fieldable field = doc.GetFieldable(DocHelper.LAZY_FIELD_KEY);
Assert.IsTrue(field != null, "field is null and it shouldn't be");
Assert.IsTrue(field.IsLazy(), "field is not lazy and it should be");
reader.Close();
try
{
field.StringValue();
Assert.Fail("did not hit AlreadyClosedException as expected");
}
catch (AlreadyClosedException e)
{
// expected
}
}
public virtual void TestLoadFirst()
{
Assert.IsTrue(dir != null);
Assert.IsTrue(fieldInfos != null);
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos);
Assert.IsTrue(reader != null);
Assert.IsTrue(reader.Size() == 1);
LoadFirstFieldSelector fieldSelector = new LoadFirstFieldSelector();
Document doc = reader.Doc(0, fieldSelector);
Assert.IsTrue(doc != null, "doc is null and it shouldn't be");
int count = 0;
var l = doc.GetFields();
for (System.Collections.IEnumerator iter = l.GetEnumerator(); iter.MoveNext(); )
{
Field field = (Field) iter.Current;
Assert.IsTrue(field != null, "field is null and it shouldn't be");
System.String sv = field.StringValue;
Assert.IsTrue(sv != null, "sv is null and it shouldn't be");
count++;
}
Assert.IsTrue(count == 1, count + " does not equal: " + 1);
}
private void Initialize(SegmentInfo si, int readBufferSize, bool doOpenStores)
{
segment = si.name;
this.si = si;
this.readBufferSize = readBufferSize;
bool success = false;
try
{
// Use compound file directory for some files, if it exists
Directory cfsDir = Directory();
if (si.GetUseCompoundFile())
{
cfsReader = new CompoundFileReader(Directory(), segment + "." + IndexFileNames.COMPOUND_FILE_EXTENSION, readBufferSize);
cfsDir = cfsReader;
}
Directory storeDir;
if (doOpenStores)
{
if (si.GetDocStoreOffset() != -1)
{
if (si.GetDocStoreIsCompoundFile())
{
storeCFSReader = new CompoundFileReader(Directory(), si.GetDocStoreSegment() + "." + IndexFileNames.COMPOUND_FILE_STORE_EXTENSION, readBufferSize);
storeDir = storeCFSReader;
}
else
{
storeDir = Directory();
}
}
else
{
storeDir = cfsDir;
}
}
else
{
storeDir = null;
}
fieldInfos = new FieldInfos(cfsDir, segment + ".fnm");
bool anyProx = false;
int numFields = fieldInfos.Size();
for (int i = 0; !anyProx && i < numFields; i++)
{
if (!fieldInfos.FieldInfo(i).omitTf)
{
anyProx = true;
}
}
System.String fieldsSegment;
if (si.GetDocStoreOffset() != -1)
{
fieldsSegment = si.GetDocStoreSegment();
}
else
{
fieldsSegment = segment;
}
if (doOpenStores)
{
fieldsReader = new FieldsReader(storeDir, fieldsSegment, fieldInfos, readBufferSize, si.GetDocStoreOffset(), si.docCount);
// Verify two sources of "maxDoc" agree:
if (si.GetDocStoreOffset() == -1 && fieldsReader.Size() != si.docCount)
{
throw new CorruptIndexException("doc counts differ for segment " + si.name + ": fieldsReader shows " + fieldsReader.Size() + " but segmentInfo shows " + si.docCount);
}
}
tis = new TermInfosReader(cfsDir, segment, fieldInfos, readBufferSize);
LoadDeletedDocs();
// make sure that all index files have been read or are kept open
// so that if an index update removes them we'll still have them
freqStream = cfsDir.OpenInput(segment + ".frq", readBufferSize);
if (anyProx)
{
proxStream = cfsDir.OpenInput(segment + ".prx", readBufferSize);
}
OpenNorms(cfsDir, readBufferSize);
if (doOpenStores && fieldInfos.HasVectors())
{
// open term vector files only as needed
System.String vectorsSegment;
if (si.GetDocStoreOffset() != -1)
{
vectorsSegment = si.GetDocStoreSegment();
}
else
{
vectorsSegment = segment;
}
termVectorsReaderOrig = new TermVectorsReader(storeDir, vectorsSegment, fieldInfos, readBufferSize, si.GetDocStoreOffset(), si.docCount);
}
success = true;
}
finally
{
// With lock-less commits, it's entirely possible (and
// fine) to hit a FileNotFound exception above. In
// this case, we want to explicitly close any subset
// of things that were opened so that we don't have to
// wait for a GC to do so.
if (!success)
{
DoClose();
}
}
}
public virtual void TestLoadSize()
{
FieldsReader reader = new FieldsReader(dir, TEST_SEGMENT_NAME, fieldInfos);
Document doc;
doc = reader.Doc(0, new AnonymousClassFieldSelector(this));
IFieldable f1 = doc.GetFieldable(DocHelper.TEXT_FIELD_1_KEY);
IFieldable f3 = doc.GetFieldable(DocHelper.TEXT_FIELD_3_KEY);
IFieldable fb = doc.GetFieldable(DocHelper.LAZY_FIELD_BINARY_KEY);
Assert.IsTrue(f1.IsBinary);
Assert.IsTrue(!f3.IsBinary);
Assert.IsTrue(fb.IsBinary);
AssertSizeEquals(2 * DocHelper.FIELD_1_TEXT.Length, f1.GetBinaryValue());
Assert.AreEqual(DocHelper.FIELD_3_TEXT, f3.StringValue);
AssertSizeEquals(DocHelper.LAZY_FIELD_BINARY_BYTES.Length, fb.GetBinaryValue());
reader.Dispose();
}
/// <summary> </summary>
/// <returns> The number of documents in all of the readers
/// </returns>
/// <throws> CorruptIndexException if the index is corrupt </throws>
/// <throws> IOException if there is a low-level IO error </throws>
private int MergeFields()
{
if (!mergeDocStores)
{
// When we are not merging by doc stores, their field
// name -> number mapping are the same. So, we start
// with the fieldInfos of the last segment in this
// case, to keep that numbering.
SegmentReader sr = (SegmentReader)readers[readers.Count - 1];
fieldInfos = (FieldInfos)sr.core.fieldInfos.Clone();
}
else
{
fieldInfos = new FieldInfos(); // merge field names
}
foreach (IndexReader reader in readers)
{
if (reader is SegmentReader)
{
SegmentReader segmentReader = (SegmentReader)reader;
FieldInfos readerFieldInfos = segmentReader.FieldInfos();
int numReaderFieldInfos = readerFieldInfos.Size();
for (int j = 0; j < numReaderFieldInfos; j++)
{
FieldInfo fi = readerFieldInfos.FieldInfo(j);
fieldInfos.Add(fi.name, fi.isIndexed, fi.storeTermVector, fi.storePositionWithTermVector, fi.storeOffsetWithTermVector, !reader.HasNorms(fi.name), fi.storePayloads, fi.omitTermFreqAndPositions);
}
}
else
{
AddIndexed(reader, fieldInfos, reader.GetFieldNames(FieldOption.TERMVECTOR_WITH_POSITION_OFFSET), true, true, true, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(FieldOption.TERMVECTOR_WITH_POSITION), true, true, false, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(FieldOption.TERMVECTOR_WITH_OFFSET), true, false, true, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(FieldOption.TERMVECTOR), true, false, false, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(FieldOption.OMIT_TERM_FREQ_AND_POSITIONS), false, false, false, false, true);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(FieldOption.STORES_PAYLOADS), false, false, false, true, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(FieldOption.INDEXED), false, false, false, false, false);
fieldInfos.Add(reader.GetFieldNames(FieldOption.UNINDEXED), false);
}
}
fieldInfos.Write(directory, segment + ".fnm");
int docCount = 0;
SetMatchingSegmentReaders();
if (mergeDocStores)
{
// merge field values
FieldsWriter fieldsWriter = new FieldsWriter(directory, segment, fieldInfos);
try
{
int idx = 0;
foreach (IndexReader reader in readers)
{
SegmentReader matchingSegmentReader = matchingSegmentReaders[idx++];
FieldsReader matchingFieldsReader = null;
if (matchingSegmentReader != null)
{
FieldsReader fieldsReader = matchingSegmentReader.GetFieldsReader();
if (fieldsReader != null && fieldsReader.CanReadRawDocs())
{
matchingFieldsReader = fieldsReader;
}
}
if (reader.HasDeletions)
{
docCount += CopyFieldsWithDeletions(fieldsWriter, reader, matchingFieldsReader);
}
else
{
docCount += CopyFieldsNoDeletions(fieldsWriter, reader, matchingFieldsReader);
}
}
}
finally
{
fieldsWriter.Dispose();
}
System.String fileName = segment + "." + IndexFileNames.FIELDS_INDEX_EXTENSION;
long fdxFileLength = directory.FileLength(fileName);
if (4 + ((long)docCount) * 8 != fdxFileLength)
{
// This is most likely a bug in Sun JRE 1.6.0_04/_05;
// we detect that the bug has struck, here, and
// throw an exception to prevent the corruption from
// entering the index. See LUCENE-1282 for
// details.
throw new System.SystemException("mergeFields produced an invalid result: docCount is " + docCount + " but fdx file size is " + fdxFileLength + " file=" + fileName + " file exists?=" + directory.FileExists(fileName) + "; now aborting this merge to prevent index corruption");
}
}
// If we are skipping the doc stores, that means there
// are no deletions in any of these segments, so we
// just sum numDocs() of each segment to get total docCount
else
{
foreach (IndexReader reader in readers)
{
docCount += reader.NumDocs();
}
}
return(docCount);
}
private void InitBlock(FieldsReader enclosingInstance)
{
this.Enclosing_Instance = enclosingInstance;
}
private void Initialize(SegmentInfo si, int readBufferSize, bool doOpenStores)
{
segment = si.name;
this.si = si;
this.readBufferSize = readBufferSize;
bool success = false;
try
{
// Use compound file directory for some files, if it exists
Directory cfsDir = Directory();
if (si.GetUseCompoundFile())
{
cfsReader = new CompoundFileReader(Directory(), segment + "." + IndexFileNames.COMPOUND_FILE_EXTENSION, readBufferSize);
cfsDir = cfsReader;
}
Directory storeDir;
if (doOpenStores)
{
if (si.GetDocStoreOffset() != - 1)
{
if (si.GetDocStoreIsCompoundFile())
{
storeCFSReader = new CompoundFileReader(Directory(), si.GetDocStoreSegment() + "." + IndexFileNames.COMPOUND_FILE_STORE_EXTENSION, readBufferSize);
storeDir = storeCFSReader;
}
else
{
storeDir = Directory();
}
}
else
{
storeDir = cfsDir;
}
}
else
storeDir = null;
fieldInfos = new FieldInfos(cfsDir, segment + ".fnm");
bool anyProx = false;
int numFields = fieldInfos.Size();
for (int i = 0; !anyProx && i < numFields; i++)
if (!fieldInfos.FieldInfo(i).omitTf)
anyProx = true;
System.String fieldsSegment;
if (si.GetDocStoreOffset() != - 1)
fieldsSegment = si.GetDocStoreSegment();
else
fieldsSegment = segment;
if (doOpenStores)
{
fieldsReader = new FieldsReader(storeDir, fieldsSegment, fieldInfos, readBufferSize, si.GetDocStoreOffset(), si.docCount);
// Verify two sources of "maxDoc" agree:
if (si.GetDocStoreOffset() == - 1 && fieldsReader.Size() != si.docCount)
{
throw new CorruptIndexException("doc counts differ for segment " + si.name + ": fieldsReader shows " + fieldsReader.Size() + " but segmentInfo shows " + si.docCount);
}
}
tis = new TermInfosReader(cfsDir, segment, fieldInfos, readBufferSize);
LoadDeletedDocs();
// make sure that all index files have been read or are kept open
// so that if an index update removes them we'll still have them
freqStream = cfsDir.OpenInput(segment + ".frq", readBufferSize);
if (anyProx)
proxStream = cfsDir.OpenInput(segment + ".prx", readBufferSize);
OpenNorms(cfsDir, readBufferSize);
if (doOpenStores && fieldInfos.HasVectors())
{
// open term vector files only as needed
System.String vectorsSegment;
if (si.GetDocStoreOffset() != - 1)
vectorsSegment = si.GetDocStoreSegment();
else
vectorsSegment = segment;
termVectorsReaderOrig = new TermVectorsReader(storeDir, vectorsSegment, fieldInfos, readBufferSize, si.GetDocStoreOffset(), si.docCount);
}
success = true;
}
finally
{
// With lock-less commits, it's entirely possible (and
// fine) to hit a FileNotFound exception above. In
// this case, we want to explicitly close any subset
// of things that were opened so that we don't have to
// wait for a GC to do so.
if (!success)
{
DoClose();
}
}
}
