public virtual void TestWriter()
{
try
{
TermVectorsWriter writer = new TermVectorsWriter(dir, seg, fieldInfos);
writer.OpenDocument();
Assert.IsTrue(writer.IsDocumentOpen() == true);
WriteField(writer, testFields[0]);
writer.CloseDocument();
writer.Close();
Assert.IsTrue(writer.IsDocumentOpen() == false);
//Check to see the files were created
Assert.IsTrue(dir.FileExists(seg + TermVectorsWriter.TVD_EXTENSION));
Assert.IsTrue(dir.FileExists(seg + TermVectorsWriter.TVX_EXTENSION));
//Now read it back in
TermVectorsReader reader = new TermVectorsReader(dir, seg, fieldInfos);
Assert.IsTrue(reader != null);
CheckTermVector(reader, 0, testFields[0]);
}
catch (System.IO.IOException e)
{
System.Console.Error.WriteLine(e.StackTrace);
Assert.IsTrue(false);
}
}
/// <summary> Merge the TermVectors from each of the segments into the new one.</summary>
/// <throws> IOException </throws>
private void MergeVectors()
{
TermVectorsWriter termVectorsWriter = new TermVectorsWriter(directory, segment, fieldInfos);
try
{
for (int r = 0; r < readers.Count; r++)
{
IndexReader reader = (IndexReader)readers[r];
int maxDoc = reader.MaxDoc();
for (int docNum = 0; docNum < maxDoc; docNum++)
{
// skip deleted docs
if (reader.IsDeleted(docNum))
{
continue;
}
termVectorsWriter.AddAllDocVectors(reader.GetTermFreqVectors(docNum));
if (checkAbort != null)
{
checkAbort.Work(300);
}
}
}
}
finally
{
termVectorsWriter.Close();
}
}
public virtual void TestMultipleDocuments()
{
try
{
TermVectorsWriter writer = new TermVectorsWriter(dir, seg, fieldInfos);
Assert.IsTrue(writer != null);
for (int i = 0; i < 10; i++)
{
WriteDocument(writer, testFields.Length);
}
writer.Close();
}
catch (System.IO.IOException e)
{
System.Console.Error.WriteLine(e.StackTrace);
Assert.IsTrue(false);
}
//Do some arbitrary tests
try
{
TermVectorsReader reader = new TermVectorsReader(dir, seg, fieldInfos);
for (int i = 0; i < 10; i++)
{
Assert.IsTrue(reader != null);
CheckTermVector(reader, 5, testFields[0]);
CheckTermVector(reader, 2, testFields[2]);
}
}
catch (System.IO.IOException e)
{
System.Console.Error.WriteLine(e.StackTrace);
Assert.IsTrue(false);
}
}
protected virtual void SetUp()
{
for (int i = 0; i < testFields.Length; i++)
{
fieldInfos.Add(testFields[i], true, true);
}
try
{
System.Array.Sort(testTerms);
for (int j = 0; j < 5; j++)
{
writer = new TermVectorsWriter(dir, seg, fieldInfos);
writer.OpenDocument();
for (int k = 0; k < testFields.Length; k++)
{
writer.OpenField(testFields[k]);
for (int i = 0; i < testTerms.Length; i++)
{
writer.AddTerm(testTerms[i], i);
}
writer.CloseField();
}
writer.CloseDocument();
writer.Close();
}
}
catch (System.IO.IOException e)
{
System.Console.Error.WriteLine(e.StackTrace);
Assert.IsTrue(false);
}
}
protected virtual void SetUp()
{
for (int i = 0; i < testFields.Length; i++)
{
fieldInfos.Add(testFields[i], true, true);
}
try
{
System.Array.Sort(testTerms);
for (int j = 0; j < 5; j++)
{
writer = new TermVectorsWriter(dir, seg, fieldInfos);
writer.OpenDocument();
for (int k = 0; k < testFields.Length; k++)
{
writer.OpenField(testFields[k]);
for (int i = 0; i < testTerms.Length; i++)
{
writer.AddTerm(testTerms[i], i);
}
writer.CloseField();
}
writer.CloseDocument();
writer.Close();
}
}
catch (System.IO.IOException e)
{
System.Console.Error.WriteLine(e.StackTrace);
Assert.IsTrue(false);
}
}
/// <summary> Merge the TermVectors from each of the segments into the new one.</summary>
/// <throws> IOException </throws>
private void MergeVectors()
{
TermVectorsWriter termVectorsWriter = new TermVectorsWriter(directory, segment, fieldInfos);
try
{
int idx = 0;
for (System.Collections.IEnumerator iter = readers.GetEnumerator(); iter.MoveNext();)
{
SegmentReader matchingSegmentReader = matchingSegmentReaders[idx++];
TermVectorsReader matchingVectorsReader = null;
if (matchingSegmentReader != null)
{
TermVectorsReader vectorsReader = matchingSegmentReader.GetTermVectorsReaderOrig();
// If the TV* files are an older format then they cannot read raw docs:
if (vectorsReader != null && vectorsReader.CanReadRawDocs())
{
matchingVectorsReader = vectorsReader;
}
}
IndexReader reader = (IndexReader)iter.Current;
if (reader.HasDeletions())
{
CopyVectorsWithDeletions(termVectorsWriter, matchingVectorsReader, reader);
}
else
{
CopyVectorsNoDeletions(termVectorsWriter, matchingVectorsReader, reader);
}
}
}
finally
{
termVectorsWriter.Close();
}
System.String fileName = segment + "." + IndexFileNames.VECTORS_INDEX_EXTENSION;
long tvxSize = directory.FileLength(fileName);
if (4 + ((long)mergedDocs) * 16 != tvxSize)
{
// 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("mergeVectors produced an invalid result: mergedDocs is " + mergedDocs + " but tvx size is " + tvxSize + " file=" + fileName + " file exists?=" + directory.FileExists(fileName) + "; now aborting this merge to prevent index corruption");
}
}
/// <summary> Merge the TermVectors from each of the segments into the new one.</summary>
/// <throws> IOException </throws>
private void MergeVectors()
{
TermVectorsWriter termVectorsWriter = new TermVectorsWriter(directory, segment, fieldInfos);
try
{
int idx = 0;
for (System.Collections.IEnumerator iter = readers.GetEnumerator(); iter.MoveNext(); )
{
SegmentReader matchingSegmentReader = matchingSegmentReaders[idx++];
TermVectorsReader matchingVectorsReader = null;
if (matchingSegmentReader != null)
{
TermVectorsReader vectorsReader = matchingSegmentReader.GetTermVectorsReaderOrig();
// If the TV* files are an older format then they cannot read raw docs:
if (vectorsReader != null && vectorsReader.CanReadRawDocs())
{
matchingVectorsReader = vectorsReader;
}
}
IndexReader reader = (IndexReader) iter.Current;
if (reader.HasDeletions())
{
CopyVectorsWithDeletions(termVectorsWriter, matchingVectorsReader, reader);
}
else
{
CopyVectorsNoDeletions(termVectorsWriter, matchingVectorsReader, reader);
}
}
}
finally
{
termVectorsWriter.Close();
}
System.String fileName = segment + "." + IndexFileNames.VECTORS_INDEX_EXTENSION;
long tvxSize = directory.FileLength(fileName);
if (4 + ((long) mergedDocs) * 16 != tvxSize)
// 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("mergeVectors produced an invalid result: mergedDocs is " + mergedDocs + " but tvx size is " + tvxSize + " file=" + fileName + " file exists?=" + directory.FileExists(fileName) + "; now aborting this merge to prevent index corruption");
}
public virtual void TestWriter()
{
try
{
TermVectorsWriter writer = new TermVectorsWriter(dir, seg, fieldInfos);
writer.OpenDocument();
Assert.IsTrue(writer.IsDocumentOpen() == true);
WriteField(writer, testFields[0]);
writer.CloseDocument();
writer.Close();
Assert.IsTrue(writer.IsDocumentOpen() == false);
//Check to see the files were created
Assert.IsTrue(dir.FileExists(seg + TermVectorsWriter.TVD_EXTENSION));
Assert.IsTrue(dir.FileExists(seg + TermVectorsWriter.TVX_EXTENSION));
//Now read it back in
TermVectorsReader reader = new TermVectorsReader(dir, seg, fieldInfos);
Assert.IsTrue(reader != null);
CheckTermVector(reader, 0, testFields[0]);
}
catch (System.IO.IOException e)
{
System.Console.Error.WriteLine(e.StackTrace);
Assert.IsTrue(false);
}
}
public virtual void TestMultipleDocuments()
{
try
{
TermVectorsWriter writer = new TermVectorsWriter(dir, seg, fieldInfos);
Assert.IsTrue(writer != null);
for (int i = 0; i < 10; i++)
{
WriteDocument(writer, testFields.Length);
}
writer.Close();
}
catch (System.IO.IOException e)
{
System.Console.Error.WriteLine(e.StackTrace);
Assert.IsTrue(false);
}
//Do some arbitrary tests
try
{
TermVectorsReader reader = new TermVectorsReader(dir, seg, fieldInfos);
for (int i = 0; i < 10; i++)
{
Assert.IsTrue(reader != null);
CheckTermVector(reader, 5, testFields[0]);
CheckTermVector(reader, 2, testFields[2]);
}
}
catch (System.IO.IOException e)
{
System.Console.Error.WriteLine(e.StackTrace);
Assert.IsTrue(false);
}
}
/// <summary> Merge the TermVectors from each of the segments into the new one.</summary>
/// <throws> IOException </throws>
private void MergeVectors()
{
TermVectorsWriter termVectorsWriter = new TermVectorsWriter(directory, segment, fieldInfos);
try
{
for (int r = 0; r < readers.Count; r++)
{
IndexReader reader = (IndexReader) readers[r];
int maxDoc = reader.MaxDoc();
for (int docNum = 0; docNum < maxDoc; docNum++)
{
// skip deleted docs
if (reader.IsDeleted(docNum))
continue;
termVectorsWriter.AddAllDocVectors(reader.GetTermFreqVectors(docNum));
}
}
}
finally
{
termVectorsWriter.Close();
}
}
/// <summary> Merge the TermVectors from each of the segments into the new one.</summary>
/// <throws> IOException </throws>
private void MergeVectors()
{
TermVectorsWriter termVectorsWriter =
new TermVectorsWriter(directory, segment, fieldInfos);
try
{
for (int r = 0; r < readers.Count; r++)
{
SegmentReader matchingSegmentReader = matchingSegmentReaders[r];
TermVectorsReader matchingVectorsReader;
bool hasMatchingReader;
if (matchingSegmentReader != null)
{
matchingVectorsReader = matchingSegmentReader.termVectorsReaderOrig;
// If the TV* files are an older format then they
// cannot read raw docs:
if (matchingVectorsReader != null && !matchingVectorsReader.CanReadRawDocs())
{
matchingVectorsReader = null;
hasMatchingReader = false;
}
else
hasMatchingReader = matchingVectorsReader != null;
}
else
{
hasMatchingReader = false;
matchingVectorsReader = null;
}
IndexReader reader = (IndexReader)readers[r];
bool hasDeletions = reader.HasDeletions();
int maxDoc = reader.MaxDoc();
for (int docNum = 0; docNum < maxDoc; )
{
// skip deleted docs
if (!hasDeletions || !reader.IsDeleted(docNum))
{
if (hasMatchingReader)
{
// We can optimize this case (doing a bulk
// byte copy) since the field numbers are
// identical
int start = docNum;
int numDocs = 0;
do
{
docNum++;
numDocs++;
if (docNum >= maxDoc)
break;
if (hasDeletions && matchingSegmentReader.IsDeleted(docNum))
{
docNum++;
break;
}
} while (numDocs < MAX_RAW_MERGE_DOCS);
matchingVectorsReader.RawDocs(rawDocLengths, rawDocLengths2, start, numDocs);
termVectorsWriter.AddRawDocuments(matchingVectorsReader, rawDocLengths, rawDocLengths2, numDocs);
if (checkAbort != null)
checkAbort.Work(300 * numDocs);
}
else
{
// NOTE: it's very important to first assign
// to vectors then pass it to
// termVectorsWriter.addAllDocVectors; see
// LUCENE-1282
TermFreqVector[] vectors = reader.GetTermFreqVectors(docNum);
termVectorsWriter.AddAllDocVectors(vectors);
docNum++;
if (checkAbort != null)
checkAbort.Work(300);
}
}
else
docNum++;
}
}
}
finally
{
termVectorsWriter.Close();
}
long tvxSize = directory.FileLength(segment + "." + IndexFileNames.VECTORS_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 + mergedDocs * 16 != tvxSize)
// // 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("mergeVectors produced an invalid result: mergedDocs is " + mergedDocs + " but tvx size is " + tvxSize + "; now aborting this merge to prevent index corruption");
}
/// <summary> Merge the TermVectors from each of the segments into the new one.</summary>
/// <throws> IOException </throws>
private void MergeVectors()
{
TermVectorsWriter termVectorsWriter =
new TermVectorsWriter(directory, segment, fieldInfos);
try
{
for (int r = 0; r < readers.Count; r++)
{
SegmentReader matchingSegmentReader = matchingSegmentReaders[r];
TermVectorsReader matchingVectorsReader;
bool hasMatchingReader;
if (matchingSegmentReader != null)
{
matchingVectorsReader = matchingSegmentReader.termVectorsReaderOrig;
// If the TV* files are an older format then they
// cannot read raw docs:
if (matchingVectorsReader != null && !matchingVectorsReader.CanReadRawDocs())
{
matchingVectorsReader = null;
hasMatchingReader = false;
}
else
{
hasMatchingReader = matchingVectorsReader != null;
}
}
else
{
hasMatchingReader = false;
matchingVectorsReader = null;
}
IndexReader reader = (IndexReader)readers[r];
bool hasDeletions = reader.HasDeletions();
int maxDoc = reader.MaxDoc();
for (int docNum = 0; docNum < maxDoc;)
{
// skip deleted docs
if (!hasDeletions || !reader.IsDeleted(docNum))
{
if (hasMatchingReader)
{
// We can optimize this case (doing a bulk
// byte copy) since the field numbers are
// identical
int start = docNum;
int numDocs = 0;
do
{
docNum++;
numDocs++;
if (docNum >= maxDoc)
{
break;
}
if (hasDeletions && matchingSegmentReader.IsDeleted(docNum))
{
docNum++;
break;
}
} while (numDocs < MAX_RAW_MERGE_DOCS);
matchingVectorsReader.RawDocs(rawDocLengths, rawDocLengths2, start, numDocs);
termVectorsWriter.AddRawDocuments(matchingVectorsReader, rawDocLengths, rawDocLengths2, numDocs);
if (checkAbort != null)
{
checkAbort.Work(300 * numDocs);
}
}
else
{
// NOTE: it's very important to first assign
// to vectors then pass it to
// termVectorsWriter.addAllDocVectors; see
// LUCENE-1282
TermFreqVector[] vectors = reader.GetTermFreqVectors(docNum);
termVectorsWriter.AddAllDocVectors(vectors);
docNum++;
if (checkAbort != null)
{
checkAbort.Work(300);
}
}
}
else
{
docNum++;
}
}
}
}
finally
{
termVectorsWriter.Close();
}
long tvxSize = directory.FileLength(segment + "." + IndexFileNames.VECTORS_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 + mergedDocs * 16 != tvxSize)
// // 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("mergeVectors produced an invalid result: mergedDocs is " + mergedDocs + " but tvx size is " + tvxSize + "; now aborting this merge to prevent index corruption");
}
/// <summary> Merge the TermVectors from each of the segments into the new one.</summary>
/// <throws> IOException </throws>
private void MergeVectors()
{
TermVectorsWriter termVectorsWriter = new TermVectorsWriter(directory, segment, fieldInfos);
try
{
for (int r = 0; r < readers.Count; r++)
{
IndexReader reader = (IndexReader) readers[r];
int maxDoc = reader.MaxDoc();
for (int docNum = 0; docNum < maxDoc; docNum++)
{
// skip deleted docs
if (reader.IsDeleted(docNum))
continue;
termVectorsWriter.AddAllDocVectors(reader.GetTermFreqVectors(docNum));
if (checkAbort != null)
checkAbort.Work(300);
}
}
}
finally
{
termVectorsWriter.Close();
}
System.Diagnostics.Debug.Assert(4 + mergedDocs * 8 == directory.FileLength(segment + "." + IndexFileNames.VECTORS_INDEX_EXTENSION),
"after MergeVectors: tvx size mismatch: " + mergedDocs + " docs vs " +
directory.FileLength(segment + "." + IndexFileNames.VECTORS_INDEX_EXTENSION) +
" length in bytes of " + segment + "." + IndexFileNames.VECTORS_INDEX_EXTENSION);
}