private void WriteTerm(Term term)
{
int start = StringHelper.StringDifference(lastTerm.text, term.text);
int length = term.text.Length - start;
output.WriteVInt(start); // write shared prefix length
output.WriteVInt(length); // write delta length
output.WriteChars(term.text, start, length); // write delta chars
output.WriteVInt(fieldInfos.FieldNumber(term.field)); // write field num
lastTerm = term;
}
internal void Add(Term term, TermInfo ti)
{
int length = term.text.Length;
if (termTextBuffer.Length < length)
{
termTextBuffer = new char[(int)(length * 1.25)];
}
int i = 0;
System.Collections.Generic.IEnumerator <char> chars = term.text.GetEnumerator();
while (chars.MoveNext())
{
termTextBuffer[i++] = (char)chars.Current;
}
Add(fieldInfos.FieldNumber(term.field), termTextBuffer, 0, length, ti);
}
/// <summary> Add a complete document specified by all its term vectors. If document has no
/// term vectors, add value for tvx.
///
/// </summary>
/// <param name="vectors">
/// </param>
/// <throws> IOException </throws>
public void AddAllDocVectors(TermFreqVector[] vectors)
{
tvx.WriteLong(tvd.GetFilePointer());
tvx.WriteLong(tvf.GetFilePointer());
if (vectors != null)
{
int numFields = vectors.Length;
tvd.WriteVInt(numFields);
long[] fieldPointers = new long[numFields];
for (int i = 0; i < numFields; i++)
{
fieldPointers[i] = tvf.GetFilePointer();
int fieldNumber = fieldInfos.FieldNumber(vectors[i].GetField());
// 1st pass: write field numbers to tvd
tvd.WriteVInt(fieldNumber);
int numTerms = vectors[i].Size();
tvf.WriteVInt(numTerms);
TermPositionVector tpVector;
byte bits;
bool storePositions;
bool storeOffsets;
if (vectors[i] is TermPositionVector)
{
// May have positions & offsets
tpVector = (TermPositionVector)vectors[i];
storePositions = tpVector.Size() > 0 && tpVector.GetTermPositions(0) != null;
storeOffsets = tpVector.Size() > 0 && tpVector.GetOffsets(0) != null;
bits = (byte)((storePositions ? TermVectorsReader.STORE_POSITIONS_WITH_TERMVECTOR : (byte)0) + (storeOffsets ? TermVectorsReader.STORE_OFFSET_WITH_TERMVECTOR : (byte)0));
}
else
{
tpVector = null;
bits = 0;
storePositions = false;
storeOffsets = false;
}
tvf.WriteVInt(bits);
string[] terms = vectors[i].GetTerms();
int[] freqs = vectors[i].GetTermFrequencies();
int utf8Upto = 0;
utf8Results[1].length = 0;
for (int j = 0; j < numTerms; j++)
{
UnicodeUtil.UTF16toUTF8(terms[j], 0, terms[j].Length, utf8Results[utf8Upto]);
int start = StringHelper.bytesDifference(
utf8Results[1 - utf8Upto].result, utf8Results[1 - utf8Upto].length, utf8Results[utf8Upto].result, utf8Results[utf8Upto].length);
int length = utf8Results[utf8Upto].length - start;
tvf.WriteVInt(start); // write shared prefix length
tvf.WriteVInt(length); // write delta length
tvf.WriteBytes(utf8Results[utf8Upto].result, start, length); // write delta bytes
utf8Upto = 1 - utf8Upto;
int termFreq = freqs[j];
tvf.WriteVInt(termFreq);
if (storePositions)
{
int[] positions = tpVector.GetTermPositions(j);
if (positions == null)
{
throw new System.SystemException("Trying to write positions that are null!");
}
System.Diagnostics.Debug.Assert(positions.Length == termFreq);
// use delta encoding for positions
int lastPosition = 0;
for (int k = 0; k < positions.Length; k++)
{
int position = positions[k];
tvf.WriteVInt(position - lastPosition);
lastPosition = position;
}
}
if (storeOffsets)
{
TermVectorOffsetInfo[] offsets = tpVector.GetOffsets(j);
if (offsets == null)
{
throw new System.SystemException("Trying to write offsets that are null!");
}
System.Diagnostics.Debug.Assert(offsets.Length == termFreq);
// use delta encoding for offsets
int lastEndOffset = 0;
for (int k = 0; k < offsets.Length; k++)
{
int startOffset = offsets[k].GetStartOffset();
int endOffset = offsets[k].GetEndOffset();
tvf.WriteVInt(startOffset - lastEndOffset);
tvf.WriteVInt(endOffset - startOffset);
lastEndOffset = endOffset;
}
}
}
}
// 2nd pass: write field pointers to tvd
long lastFieldPointer = fieldPointers[0];
for (int i = 1; i < numFields; i++)
{
long fieldPointer = fieldPointers[i];
tvd.WriteVLong(fieldPointer - lastFieldPointer);
lastFieldPointer = fieldPointer;
}
}
else
{
tvd.WriteVInt(0);
}
}
internal void Add(Term term, TermInfo ti)
{
UnicodeUtil.UTF16toUTF8(term.Text, 0, term.Text.Length, utf8Result);
Add(fieldInfos.FieldNumber(term.Field), utf8Result.result, utf8Result.length, ti);
}
internal void AddDocument(Document doc)
{
indexStream.WriteLong(fieldsStream.GetFilePointer());
int storedCount = 0;
System.Collections.IEnumerator fieldIterator = doc.GetFields().GetEnumerator();
while (fieldIterator.MoveNext())
{
Fieldable field = (Fieldable)fieldIterator.Current;
if (field.IsStored())
{
storedCount++;
}
}
fieldsStream.WriteVInt(storedCount);
fieldIterator = doc.GetFields().GetEnumerator();
while (fieldIterator.MoveNext())
{
Fieldable field = (Fieldable)fieldIterator.Current;
// if the field as an instanceof FieldsReader.FieldForMerge, we're in merge mode
// and field.binaryValue() already returns the compressed value for a field
// with isCompressed()==true, so we disable compression in that case
bool disableCompression = (field is FieldsReader.FieldForMerge);
if (field.IsStored())
{
fieldsStream.WriteVInt(fieldInfos.FieldNumber(field.Name()));
byte bits = 0;
if (field.IsTokenized())
{
bits |= FieldsWriter.FIELD_IS_TOKENIZED;
}
if (field.IsBinary())
{
bits |= FieldsWriter.FIELD_IS_BINARY;
}
if (field.IsCompressed())
{
bits |= FieldsWriter.FIELD_IS_COMPRESSED;
}
fieldsStream.WriteByte(bits);
if (field.IsCompressed())
{
// compression is enabled for the current field
byte[] data = null;
if (disableCompression)
{
// optimized case for merging, the data
// is already compressed
data = field.BinaryValue();
}
else
{
// check if it is a binary field
if (field.IsBinary())
{
data = Compress(field.BinaryValue());
}
else
{
data = Compress(System.Text.Encoding.GetEncoding("UTF-8").GetBytes(field.StringValue()));
}
}
int len = data.Length;
fieldsStream.WriteVInt(len);
fieldsStream.WriteBytes(data, len);
}
else
{
// compression is disabled for the current field
if (field.IsBinary())
{
byte[] data = field.BinaryValue();
int len = data.Length;
fieldsStream.WriteVInt(len);
fieldsStream.WriteBytes(data, len);
}
else
{
fieldsStream.WriteString(field.StringValue());
}
}
}
}
}
internal void AddDocument(Document doc)
{
indexStream.WriteLong(fieldsStream.GetFilePointer());
int storedCount = 0;
foreach (Field field in doc.Fields())
{
if (field.IsStored())
{
storedCount++;
}
}
fieldsStream.WriteVInt(storedCount);
foreach (Field field in doc.Fields())
{
if (field.IsStored())
{
fieldsStream.WriteVInt(fieldInfos.FieldNumber(field.Name()));
byte bits = 0;
if (field.IsTokenized())
{
bits |= FieldsWriter.FIELD_IS_TOKENIZED;
}
if (field.IsBinary())
{
bits |= FieldsWriter.FIELD_IS_BINARY;
}
if (field.IsCompressed())
{
bits |= FieldsWriter.FIELD_IS_COMPRESSED;
}
fieldsStream.WriteByte(bits);
if (field.IsCompressed())
{
// compression is enabled for the current field
byte[] data = null;
// check if it is a binary field
if (field.IsBinary())
{
data = Compress(field.BinaryValue());
}
else
{
data = Compress(System.Text.Encoding.GetEncoding("UTF-8").GetBytes(field.StringValue()));
}
int len = data.Length;
fieldsStream.WriteVInt(len);
fieldsStream.WriteBytes(data, len);
}
else
{
// compression is disabled for the current field
if (field.IsBinary())
{
byte[] data = field.BinaryValue();
int len = data.Length;
fieldsStream.WriteVInt(len);
fieldsStream.WriteBytes(data, len);
}
else
{
fieldsStream.WriteString(field.StringValue());
}
}
}
}
}
/// <summary> Retrieve the term vector for the given document and field</summary>
/// <param name="docNum">The document number to retrieve the vector for
/// </param>
/// <param name="field">The field within the document to retrieve
/// </param>
/// <returns> The TermFreqVector for the document and field or null if there is no termVector for this field.
/// </returns>
/// <throws> IOException if there is an error reading the term vector files </throws>
public /*internal*/ virtual TermFreqVector Get(int docNum, System.String field)
{
// Check if no term vectors are available for this segment at all
int fieldNumber = fieldInfos.FieldNumber(field);
TermFreqVector result = null;
if (tvx != null)
{
//We need to account for the FORMAT_SIZE at when seeking in the tvx
//We don't need to do this in other seeks because we already have the
// file pointer
//that was written in another file
tvx.Seek((docNum * 8L) + TermVectorsWriter.FORMAT_SIZE);
//System.out.println("TVX Pointer: " + tvx.getFilePointer());
long position = tvx.ReadLong();
tvd.Seek(position);
int fieldCount = tvd.ReadVInt();
//System.out.println("Num Fields: " + fieldCount);
// There are only a few fields per document. We opt for a full scan
// rather then requiring that they be ordered. We need to read through
// all of the fields anyway to get to the tvf pointers.
int number = 0;
int found = -1;
for (int i = 0; i < fieldCount; i++)
{
if (tvdFormat == TermVectorsWriter.FORMAT_VERSION)
{
number = tvd.ReadVInt();
}
else
{
number += tvd.ReadVInt();
}
if (number == fieldNumber)
{
found = i;
}
}
// This field, although valid in the segment, was not found in this
// document
if (found != -1)
{
// Compute position in the tvf file
position = 0;
for (int i = 0; i <= found; i++)
{
position += tvd.ReadVLong();
}
result = ReadTermVector(field, position);
}
else
{
//System.out.println("Field not found");
}
}
else
{
//System.out.println("No tvx file");
}
return(result);
}
public virtual void Get(int docNum, System.String field, TermVectorMapper mapper)
{
if (tvx != null)
{
int fieldNumber = fieldInfos.FieldNumber(field);
//We need to account for the FORMAT_SIZE at when seeking in the tvx
//We don't need to do this in other seeks because we already have the
// file pointer
//that was written in another file
SeekTvx(docNum);
//System.out.println("TVX Pointer: " + tvx.getFilePointer());
long tvdPosition = tvx.ReadLong();
tvd.Seek(tvdPosition);
int fieldCount = tvd.ReadVInt();
//System.out.println("Num Fields: " + fieldCount);
// There are only a few fields per document. We opt for a full scan
// rather then requiring that they be ordered. We need to read through
// all of the fields anyway to get to the tvf pointers.
int number = 0;
int found = -1;
for (int i = 0; i < fieldCount; i++)
{
if (format >= FORMAT_VERSION)
{
number = tvd.ReadVInt();
}
else
{
number += tvd.ReadVInt();
}
if (number == fieldNumber)
{
found = i;
}
}
// This field, although valid in the segment, was not found in this
// document
if (found != -1)
{
// Compute position in the tvf file
long position;
if (format >= FORMAT_VERSION2)
{
position = tvx.ReadLong();
}
else
{
position = tvd.ReadVLong();
}
for (int i = 1; i <= found; i++)
{
position += tvd.ReadVLong();
}
mapper.SetDocumentNumber(docNum);
ReadTermVector(field, position, mapper);
}
else
{
//System.out.println("Fieldable not found");
}
}
else
{
//System.out.println("No tvx file");
}
}
// Tokenizes the fields of a document into Postings.
private void InvertDocument(Document doc)
{
foreach (Field field in doc.Fields())
{
System.String fieldName = field.Name();
int fieldNumber = fieldInfos.FieldNumber(fieldName);
int length = fieldLengths[fieldNumber]; // length of field
int position = fieldPositions[fieldNumber]; // position in field
if (length > 0)
{
position += analyzer.GetPositionIncrementGap(fieldName);
}
int offset = fieldOffsets[fieldNumber]; // offset field
if (field.IsIndexed())
{
if (!field.IsTokenized())
{
// un-tokenized field
System.String stringValue = field.StringValue();
if (field.IsStoreOffsetWithTermVector())
{
AddPosition(fieldName, stringValue, position++, new TermVectorOffsetInfo(offset, offset + stringValue.Length));
}
else
{
AddPosition(fieldName, stringValue, position++, null);
}
offset += stringValue.Length;
length++;
}
else
{
System.IO.TextReader reader; // find or make Reader
if (field.ReaderValue() != null)
{
reader = field.ReaderValue();
}
else if (field.StringValue() != null)
{
reader = new System.IO.StringReader(field.StringValue());
}
else
{
throw new System.ArgumentException("field must have either String or Reader value");
}
// Tokenize field and add to postingTable
TokenStream stream = analyzer.TokenStream(fieldName, reader);
try
{
Token lastToken = null;
for (Token t = stream.Next(); t != null; t = stream.Next())
{
position += (t.GetPositionIncrement() - 1);
if (field.IsStoreOffsetWithTermVector())
{
AddPosition(fieldName, t.TermText(), position++, new TermVectorOffsetInfo(offset + t.StartOffset(), offset + t.EndOffset()));
}
else
{
AddPosition(fieldName, t.TermText(), position++, null);
}
lastToken = t;
if (++length > maxFieldLength)
{
if (infoStream != null)
{
infoStream.WriteLine("maxFieldLength " + maxFieldLength + " reached, ignoring following tokens");
}
break;
}
}
if (lastToken != null)
{
offset += lastToken.EndOffset() + 1;
}
}
finally
{
stream.Close();
}
}
fieldLengths[fieldNumber] = length; // save field length
fieldPositions[fieldNumber] = position; // save field position
fieldBoosts[fieldNumber] *= field.GetBoost();
fieldOffsets[fieldNumber] = offset;
}
}
}
