/// <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++)
                {
                    Monodoc.Lucene.Net.Index.IndexReader reader = (Monodoc.Lucene.Net.Index.IndexReader)readers[r];
                    int maxDoc = reader.MaxDoc();
                    for (int docNum = 0; docNum < maxDoc; docNum++)
                    {
                        // skip deleted docs
                        if (reader.IsDeleted(docNum))
                        {
                            continue;
                        }
                        termVectorsWriter.OpenDocument();

                        // get all term vectors
                        TermFreqVector[] sourceTermVector = reader.GetTermFreqVectors(docNum);

                        if (sourceTermVector != null)
                        {
                            for (int f = 0; f < sourceTermVector.Length; f++)
                            {
                                // translate Field numbers
                                TermFreqVector termVector = sourceTermVector[f];
                                termVectorsWriter.OpenField(termVector.GetField());
                                System.String[] terms = termVector.GetTerms();
                                int[]           freqs = termVector.GetTermFrequencies();

                                for (int t = 0; t < terms.Length; t++)
                                {
                                    termVectorsWriter.AddTerm(terms[t], freqs[t]);
                                }
                            }
                            termVectorsWriter.CloseDocument();
                        }
                    }
                }
            }
            finally
            {
                termVectorsWriter.Close();
            }
        }
Example #2
0
        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);
                }
            }
        }
Example #3
0
		/// <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++)
				{
					Monodoc.Lucene.Net.Index.IndexReader reader = (Monodoc.Lucene.Net.Index.IndexReader) readers[r];
					int maxDoc = reader.MaxDoc();
					for (int docNum = 0; docNum < maxDoc; docNum++)
					{
						// skip deleted docs
						if (reader.IsDeleted(docNum))
						{
							continue;
						}
						termVectorsWriter.OpenDocument();
						
						// get all term vectors
						TermFreqVector[] sourceTermVector = reader.GetTermFreqVectors(docNum);
						
						if (sourceTermVector != null)
						{
							for (int f = 0; f < sourceTermVector.Length; f++)
							{
								// translate Field numbers
								TermFreqVector termVector = sourceTermVector[f];
								termVectorsWriter.OpenField(termVector.GetField());
								System.String[] terms = termVector.GetTerms();
								int[] freqs = termVector.GetTermFrequencies();
								
								for (int t = 0; t < terms.Length; t++)
								{
									termVectorsWriter.AddTerm(terms[t], freqs[t]);
								}
							}
							termVectorsWriter.CloseDocument();
						}
					}
				}
			}
			finally
			{
				termVectorsWriter.Close();
			}
		}
Example #4
0
		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);
				}
			}
		}