private TermInfo termInfo = new TermInfo(); // minimize consing
/// <summary>Merge one term found in one or more segments. The array <code>smis</code>
/// contains segments that are positioned at the same term. <code>N</code>
/// is the number of cells in the array actually occupied.
///
/// </summary>
/// <param name="smis">array of segments
/// </param>
/// <param name="n">number of cells in the array actually occupied
/// </param>
private void MergeTermInfo(SegmentMergeInfo[] smis, int n)
{
long freqPointer = freqOutput.GetFilePointer();
long proxPointer = proxOutput.GetFilePointer();
int df = AppendPostings(smis, n); // append posting data
long skipPointer = WriteSkip();
if (df > 0)
{
// add an entry to the dictionary with pointers to prox and freq files
termInfo.Set(df, freqPointer, proxPointer, (int)(skipPointer - freqPointer));
termInfosWriter.Add(smis[0].term, termInfo);
}
}
/// <summary>Adds a new <Term, TermInfo> pair to the set.
/// Term must be lexicographically greater than all previous Terms added.
/// TermInfo pointers must be positive and greater than all previous.
/// </summary>
public /*internal*/ void Add(Term term, TermInfo ti)
{
if (!isIndex && term.CompareTo(lastTerm) <= 0)
{
throw new System.IO.IOException("term out of order");
}
if (ti.freqPointer < lastTi.freqPointer)
{
throw new System.IO.IOException("freqPointer out of order");
}
if (ti.proxPointer < lastTi.proxPointer)
{
throw new System.IO.IOException("proxPointer out of order");
}
if (!isIndex && size % indexInterval == 0)
{
other.Add(lastTerm, lastTi); // add an index term
}
WriteTerm(term); // write term
output.WriteVInt(ti.docFreq); // write doc freq
output.WriteVLong(ti.freqPointer - lastTi.freqPointer); // write pointers
output.WriteVLong(ti.proxPointer - lastTi.proxPointer);
if (ti.docFreq >= skipInterval)
{
output.WriteVInt(ti.skipOffset);
}
if (isIndex)
{
output.WriteVLong(other.output.GetFilePointer() - lastIndexPointer);
lastIndexPointer = other.output.GetFilePointer(); // write pointer
}
lastTi.Set(ti);
size++;
}
private void WritePostings(Posting[] postings, System.String segment)
{
OutputStream freq = null, prox = null;
TermInfosWriter tis = null;
TermVectorsWriter termVectorWriter = null;
try
{
//open files for inverse index storage
freq = directory.CreateFile(segment + ".frq");
prox = directory.CreateFile(segment + ".prx");
tis = new TermInfosWriter(directory, segment, fieldInfos);
TermInfo ti = new TermInfo();
System.String currentField = null;
for (int i = 0; i < postings.Length; i++)
{
Posting posting = postings[i];
// add an entry to the dictionary with pointers to prox and freq files
ti.Set(1, freq.GetFilePointer(), prox.GetFilePointer(), -1);
tis.Add(posting.term, ti);
// add an entry to the freq file
int postingFreq = posting.freq;
if (postingFreq == 1)
{
// optimize freq=1
freq.WriteVInt(1);
}
// set low bit of doc num.
else
{
freq.WriteVInt(0); // the document number
freq.WriteVInt(postingFreq); // frequency in doc
}
int lastPosition = 0; // write positions
int[] positions = posting.positions;
for (int j = 0; j < postingFreq; j++)
{
// use delta-encoding
int position = positions[j];
prox.WriteVInt(position - lastPosition);
lastPosition = position;
}
// check to see if we switched to a new Field
System.String termField = posting.term.Field();
if ((System.Object)currentField != (System.Object)termField)
{
// changing Field - see if there is something to save
currentField = termField;
FieldInfo fi = fieldInfos.FieldInfo(currentField);
if (fi.storeTermVector)
{
if (termVectorWriter == null)
{
termVectorWriter = new TermVectorsWriter(directory, segment, fieldInfos);
termVectorWriter.OpenDocument();
}
termVectorWriter.OpenField(currentField);
}
else if (termVectorWriter != null)
{
termVectorWriter.CloseField();
}
}
if (termVectorWriter != null && termVectorWriter.IsFieldOpen())
{
termVectorWriter.AddTerm(posting.term.Text(), postingFreq);
}
}
if (termVectorWriter != null)
{
termVectorWriter.CloseDocument();
}
}
finally
{
// make an effort to close all streams we can but remember and re-throw
// the first exception encountered in this process
System.IO.IOException keep = null;
if (freq != null)
{
try
{
freq.Close();
}
catch (System.IO.IOException e)
{
if (keep == null)
{
keep = e;
}
}
}
if (prox != null)
{
try
{
prox.Close();
}
catch (System.IO.IOException e)
{
if (keep == null)
{
keep = e;
}
}
}
if (tis != null)
{
try
{
tis.Close();
}
catch (System.IO.IOException e)
{
if (keep == null)
{
keep = e;
}
}
}
if (termVectorWriter != null)
{
try
{
termVectorWriter.Close();
}
catch (System.IO.IOException e)
{
if (keep == null)
{
keep = e;
}
}
}
if (keep != null)
{
throw new System.IO.IOException(keep.StackTrace);
}
}
}
private void WritePostings(Posting[] postings, System.String segment)
{
OutputStream freq = null, prox = null;
TermInfosWriter tis = null;
TermVectorsWriter termVectorWriter = null;
try
{
//open files for inverse index storage
freq = directory.CreateFile(segment + ".frq");
prox = directory.CreateFile(segment + ".prx");
tis = new TermInfosWriter(directory, segment, fieldInfos);
TermInfo ti = new TermInfo();
System.String currentField = null;
for (int i = 0; i < postings.Length; i++)
{
Posting posting = postings[i];
// add an entry to the dictionary with pointers to prox and freq files
ti.Set(1, freq.GetFilePointer(), prox.GetFilePointer(), - 1);
tis.Add(posting.term, ti);
// add an entry to the freq file
int postingFreq = posting.freq;
if (postingFreq == 1)
// optimize freq=1
freq.WriteVInt(1);
// set low bit of doc num.
else
{
freq.WriteVInt(0); // the document number
freq.WriteVInt(postingFreq); // frequency in doc
}
int lastPosition = 0; // write positions
int[] positions = posting.positions;
for (int j = 0; j < postingFreq; j++)
{
// use delta-encoding
int position = positions[j];
prox.WriteVInt(position - lastPosition);
lastPosition = position;
}
// check to see if we switched to a new Field
System.String termField = posting.term.Field();
if ((System.Object) currentField != (System.Object) termField)
{
// changing Field - see if there is something to save
currentField = termField;
FieldInfo fi = fieldInfos.FieldInfo(currentField);
if (fi.storeTermVector)
{
if (termVectorWriter == null)
{
termVectorWriter = new TermVectorsWriter(directory, segment, fieldInfos);
termVectorWriter.OpenDocument();
}
termVectorWriter.OpenField(currentField);
}
else if (termVectorWriter != null)
{
termVectorWriter.CloseField();
}
}
if (termVectorWriter != null && termVectorWriter.IsFieldOpen())
{
termVectorWriter.AddTerm(posting.term.Text(), postingFreq);
}
}
if (termVectorWriter != null)
termVectorWriter.CloseDocument();
}
finally
{
// make an effort to close all streams we can but remember and re-throw
// the first exception encountered in this process
System.IO.IOException keep = null;
if (freq != null)
try
{
freq.Close();
}
catch (System.IO.IOException e)
{
if (keep == null)
keep = e;
}
if (prox != null)
try
{
prox.Close();
}
catch (System.IO.IOException e)
{
if (keep == null)
keep = e;
}
if (tis != null)
try
{
tis.Close();
}
catch (System.IO.IOException e)
{
if (keep == null)
keep = e;
}
if (termVectorWriter != null)
try
{
termVectorWriter.Close();
}
catch (System.IO.IOException e)
{
if (keep == null)
keep = e;
}
if (keep != null)
{
throw new System.IO.IOException(keep.StackTrace);
}
}
}