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);
}
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);
}
public /*internal*/ void AddDocument(System.String segment, Document doc)
{
// write field names
fieldInfos = new FieldInfos();
fieldInfos.Add(doc);
fieldInfos.Write(directory, segment + ".fnm");
// write field values
FieldsWriter fieldsWriter = new FieldsWriter(directory, segment, fieldInfos);
try
{
fieldsWriter.AddDocument(doc);
}
finally
{
fieldsWriter.Close();
}
// invert doc into postingTable
postingTable.Clear(); // clear postingTable
fieldLengths = new int[fieldInfos.Size()]; // init fieldLengths
fieldPositions = new int[fieldInfos.Size()]; // init fieldPositions
fieldOffsets = new int[fieldInfos.Size()]; // init fieldOffsets
fieldBoosts = new float[fieldInfos.Size()]; // init fieldBoosts
float boost = doc.GetBoost();
for (int i = 0; i < fieldBoosts.Length; i++)
{
fieldBoosts[i] = boost;
}
InvertDocument(doc);
// sort postingTable into an array
Posting[] postings = SortPostingTable();
/*
* for (int i = 0; i < postings.length; i++) {
* Posting posting = postings[i];
* System.out.print(posting.term);
* System.out.print(" freq=" + posting.freq);
* System.out.print(" pos=");
* System.out.print(posting.positions[0]);
* for (int j = 1; j < posting.freq; j++)
* System.out.print("," + posting.positions[j]);
* System.out.println("");
* }
*/
// write postings
WritePostings(postings, segment);
// write norms of indexed fields
WriteNorms(segment);
}
public void AddDocument(System.String segment, Document doc)
{
// write field names
fieldInfos = new FieldInfos();
fieldInfos.Add(doc);
fieldInfos.Write(directory, segment + ".fnm");
// write field values
FieldsWriter fieldsWriter = new FieldsWriter(directory, segment, fieldInfos);
try
{
fieldsWriter.AddDocument(doc);
}
finally
{
fieldsWriter.Close();
}
// invert doc into postingTable
postingTable.Clear(); // clear postingTable
fieldLengths = new int[fieldInfos.Size()]; // init fieldLengths
fieldPositions = new int[fieldInfos.Size()]; // init fieldPositions
fieldOffsets = new int[fieldInfos.Size()]; // init fieldOffsets
fieldBoosts = new float[fieldInfos.Size()]; // init fieldBoosts
float boost = doc.GetBoost();
for (int i = 0; i < fieldBoosts.Length; i++)
{
fieldBoosts[i] = boost;
}
InvertDocument(doc);
// sort postingTable into an array
Posting[] postings = SortPostingTable();
/*
for (int i = 0; i < postings.length; i++) {
Posting posting = postings[i];
System.out.print(posting.term);
System.out.print(" freq=" + posting.freq);
System.out.print(" pos=");
System.out.print(posting.positions[0]);
for (int j = 1; j < posting.freq; j++)
System.out.print("," + posting.positions[j]);
System.out.println("");
}
*/
// write postings
WritePostings(postings, segment);
// write norms of indexed fields
WriteNorms(segment);
}
/// <summary> </summary>
/// <returns> The number of documents in all of the readers
/// </returns>
/// <throws> IOException </throws>
private int MergeFields()
{
fieldInfos = new FieldInfos(); // merge field names
int docCount = 0;
for (int i = 0; i < readers.Count; i++)
{
IndexReader reader = (IndexReader)readers[i];
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION_OFFSET), true, true, true);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION), true, true, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_OFFSET), true, false, true);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR), true, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.INDEXED), false, false, false);
fieldInfos.Add(reader.GetFieldNames(IndexReader.FieldOption.UNINDEXED), false);
}
fieldInfos.Write(directory, segment + ".fnm");
FieldsWriter fieldsWriter = new FieldsWriter(directory, segment, fieldInfos);
// 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);
try
{
for (int i = 0; i < readers.Count; i++)
{
IndexReader reader = (IndexReader)readers[i];
int maxDoc = reader.MaxDoc();
for (int j = 0; j < maxDoc; j++)
{
if (!reader.IsDeleted(j))
{
// skip deleted docs
fieldsWriter.AddDocument(reader.Document(j, fieldSelectorMerge));
docCount++;
}
}
}
}
finally
{
fieldsWriter.Close();
}
return(docCount);
}
/// <summary> </summary>
/// <returns> The number of documents in all of the readers
/// </returns>
/// <throws> IOException </throws>
private int MergeFields()
{
fieldInfos = new FieldInfos(); // merge field names
int docCount = 0;
for (int i = 0; i < readers.Count; i++)
{
IndexReader reader = (IndexReader)readers[i];
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION_OFFSET), true, true, true);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION), true, true, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_OFFSET), true, false, true);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR), true, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.INDEXED), false, false, false);
fieldInfos.Add(reader.GetFieldNames(IndexReader.FieldOption.UNINDEXED), false);
}
fieldInfos.Write(directory, segment + ".fnm");
FieldsWriter fieldsWriter = new FieldsWriter(directory, segment, fieldInfos);
try
{
for (int i = 0; i < readers.Count; i++)
{
IndexReader reader = (IndexReader)readers[i];
int maxDoc = reader.MaxDoc();
for (int j = 0; j < maxDoc; j++)
{
if (!reader.IsDeleted(j))
{
// skip deleted docs
fieldsWriter.AddDocument(reader.Document(j));
docCount++;
}
}
}
}
finally
{
fieldsWriter.Close();
}
return(docCount);
}
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;
}
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;
}
/// <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);
}
}
else
{
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION_OFFSET), true, true, true, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION), true, true, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_OFFSET), true, false, true, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR), true, false, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.STORES_PAYLOADS), false, false, false, true);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.INDEXED), false, false, false, false);
fieldInfos.Add(reader.GetFieldNames(IndexReader.FieldOption.UNINDEXED), false);
}
}
fieldInfos.Write(directory, segment + ".fnm");
int docCount = 0;
if (mergeDocStores)
{
// If the i'th reader is a SegmentReader and has
// identical fieldName -> number mapping, then this
// array will be non-null at position i:
SegmentReader[] matchingSegmentReaders = new SegmentReader[readers.Count];
// If this reader is a SegmentReader, and all of its
// field name -> number mappings match the "merged"
// FieldInfos, then we can do a bulk copy of the
// stored fields:
for (int i = 0; i < readers.Count; i++)
{
IndexReader reader = (IndexReader)readers[i];
if (reader is SegmentReader)
{
SegmentReader segmentReader = (SegmentReader)reader;
bool same = true;
FieldInfos segmentFieldInfos = segmentReader.GetFieldInfos();
for (int j = 0; same && j < segmentFieldInfos.Size(); j++)
{
same = fieldInfos.FieldName(j).Equals(segmentFieldInfos.FieldName(j));
}
if (same)
{
matchingSegmentReaders[i] = segmentReader;
}
}
}
// Used for bulk-reading raw bytes for stored fields
int[] rawDocLengths = new int[MAX_RAW_MERGE_DOCS];
// 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;
if (matchingSegmentReader != null)
{
matchingFieldsReader = matchingSegmentReader.GetFieldsReader();
}
else
{
matchingFieldsReader = null;
}
int maxDoc = reader.MaxDoc();
for (int j = 0; j < maxDoc;)
{
if (!reader.IsDeleted(j))
{
// skip deleted docs
if (matchingSegmentReader != null)
{
// 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++;
}while (j < maxDoc && !matchingSegmentReader.IsDeleted(j) && 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
{
fieldsWriter.AddDocument(reader.Document(j, fieldSelectorMerge));
j++;
docCount++;
if (checkAbort != null)
{
checkAbort.Work(300);
}
}
}
else
{
j++;
}
}
}
}
finally
{
fieldsWriter.Close();
}
}
// 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
{
for (int i = 0; i < readers.Count; i++)
{
docCount += ((IndexReader)readers[i]).NumDocs();
}
}
return(docCount);
}
/// <summary> </summary>
/// <returns> The number of documents in all of the readers
/// </returns>
/// <throws> IOException </throws>
private int MergeFields()
{
fieldInfos = new FieldInfos(); // merge field names
int docCount = 0;
for (int i = 0; i < readers.Count; i++)
{
IndexReader reader = (IndexReader) readers[i];
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION_OFFSET), true, true, true);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION), true, true, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_OFFSET), true, false, true);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR), true, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.INDEXED), false, false, false);
fieldInfos.Add(reader.GetFieldNames(IndexReader.FieldOption.UNINDEXED), false);
}
fieldInfos.Write(directory, segment + ".fnm");
FieldsWriter fieldsWriter = new FieldsWriter(directory, segment, fieldInfos);
try
{
for (int i = 0; i < readers.Count; i++)
{
IndexReader reader = (IndexReader) readers[i];
int maxDoc = reader.MaxDoc();
for (int j = 0; j < maxDoc; j++)
if (!reader.IsDeleted(j))
{
// skip deleted docs
fieldsWriter.AddDocument(reader.Document(j));
docCount++;
}
}
}
finally
{
fieldsWriter.Close();
}
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;
}
/// <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);
}
/// <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);
}
}
else
{
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION_OFFSET), true, true, true, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION), true, true, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_OFFSET), true, false, true, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR), true, false, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.STORES_PAYLOADS), false, false, false, true);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.INDEXED), false, false, false, false);
fieldInfos.Add(reader.GetFieldNames(IndexReader.FieldOption.UNINDEXED), false);
}
}
fieldInfos.Write(directory, segment + ".fnm");
int docCount = 0;
if (mergeDocStores)
{
// If the i'th reader is a SegmentReader and has
// identical fieldName -> number mapping, then this
// array will be non-null at position i:
SegmentReader[] matchingSegmentReaders = new SegmentReader[readers.Count];
// If this reader is a SegmentReader, and all of its
// field name -> number mappings match the "merged"
// FieldInfos, then we can do a bulk copy of the
// stored fields:
for (int i = 0; i < readers.Count; i++)
{
IndexReader reader = (IndexReader) readers[i];
if (reader is SegmentReader)
{
SegmentReader segmentReader = (SegmentReader) reader;
bool same = true;
FieldInfos segmentFieldInfos = segmentReader.GetFieldInfos();
for (int j = 0; same && j < segmentFieldInfos.Size(); j++)
same = fieldInfos.FieldName(j).Equals(segmentFieldInfos.FieldName(j));
if (same)
{
matchingSegmentReaders[i] = segmentReader;
}
}
}
// Used for bulk-reading raw bytes for stored fields
int[] rawDocLengths = new int[MAX_RAW_MERGE_DOCS];
// 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;
if (matchingSegmentReader != null)
matchingFieldsReader = matchingSegmentReader.GetFieldsReader();
else
matchingFieldsReader = null;
int maxDoc = reader.MaxDoc();
for (int j = 0; j < maxDoc; )
{
if (!reader.IsDeleted(j))
{
// skip deleted docs
if (matchingSegmentReader != null)
{
// 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++;
}
while (j < maxDoc && !matchingSegmentReader.IsDeleted(j) && 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
{
fieldsWriter.AddDocument(reader.Document(j, fieldSelectorMerge));
j++;
docCount++;
if (checkAbort != null)
checkAbort.Work(300);
}
}
else
j++;
}
}
}
finally
{
fieldsWriter.Close();
}
System.Diagnostics.Debug.Assert(docCount*8 == directory.FileLength(segment + "." + IndexFileNames.FIELDS_INDEX_EXTENSION),
"after MergeFields: fdx size mismatch: " + docCount + " docs vs " +
directory.FileLength(segment + "." + IndexFileNames.FIELDS_INDEX_EXTENSION) +
" length in bytes of " + segment + "." + IndexFileNames.FIELDS_INDEX_EXTENSION);
}
// 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
for (int i = 0; i < readers.Count; i++)
docCount += ((IndexReader) readers[i]).NumDocs();
return docCount;
}
/// <summary> </summary>
/// <returns> The number of documents in all of the readers
/// </returns>
/// <throws> IOException </throws>
private int MergeFields()
{
fieldInfos = new FieldInfos(); // merge field names
int docCount = 0;
for (int i = 0; i < readers.Count; i++)
{
IndexReader reader = (IndexReader) readers[i];
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION_OFFSET), true, true, true);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_POSITION), true, true, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR_WITH_OFFSET), true, false, true);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.TERMVECTOR), true, false, false);
AddIndexed(reader, fieldInfos, reader.GetFieldNames(IndexReader.FieldOption.INDEXED), false, false, false);
fieldInfos.Add(reader.GetFieldNames(IndexReader.FieldOption.UNINDEXED), false);
}
fieldInfos.Write(directory, segment + ".fnm");
FieldsWriter fieldsWriter = new FieldsWriter(directory, segment, fieldInfos);
// 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);
try
{
for (int i = 0; i < readers.Count; i++)
{
IndexReader reader = (IndexReader) readers[i];
int maxDoc = reader.MaxDoc();
for (int j = 0; j < maxDoc; j++)
if (!reader.IsDeleted(j))
{
// skip deleted docs
fieldsWriter.AddDocument(reader.Document(j, fieldSelectorMerge));
docCount++;
}
}
}
finally
{
fieldsWriter.Close();
}
return docCount;
}