Java 类org.apache.lucene.util.PagedBytes.PagedBytesDataOutput 实例源码

项目:lams    文件:TermInfosReaderIndex.java   
/**
 * Loads the segment information at segment load time.
 * 
 * @param indexEnum
 *          the term enum.
 * @param indexDivisor
 *          the index divisor.
 * @param tiiFileLength
 *          the size of the tii file, used to approximate the size of the
 *          buffer.
 * @param totalIndexInterval
 *          the total index interval.
 */
TermInfosReaderIndex(SegmentTermEnum indexEnum, int indexDivisor, long tiiFileLength, int totalIndexInterval) throws IOException {
  this.totalIndexInterval = totalIndexInterval;
  indexSize = 1 + ((int) indexEnum.size - 1) / indexDivisor;
  skipInterval = indexEnum.skipInterval;
  // this is only an inital size, it will be GCed once the build is complete
  long initialSize = (long) (tiiFileLength * 1.5) / indexDivisor;
  PagedBytes dataPagedBytes = new PagedBytes(estimatePageBits(initialSize));
  PagedBytesDataOutput dataOutput = dataPagedBytes.getDataOutput();

  final int bitEstimate = 1+MathUtil.log(tiiFileLength, 2);
  GrowableWriter indexToTerms = new GrowableWriter(bitEstimate, indexSize, PackedInts.DEFAULT);

  String currentField = null;
  List<String> fieldStrs = new ArrayList<>();
  int fieldCounter = -1;
  for (int i = 0; indexEnum.next(); i++) {
    Term term = indexEnum.term();
    if (currentField == null || !currentField.equals(term.field())) {
      currentField = term.field();
      fieldStrs.add(currentField);
      fieldCounter++;
    }
    TermInfo termInfo = indexEnum.termInfo();
    indexToTerms.set(i, dataOutput.getPosition());
    dataOutput.writeVInt(fieldCounter);
    dataOutput.writeString(term.text());
    dataOutput.writeVInt(termInfo.docFreq);
    if (termInfo.docFreq >= skipInterval) {
      dataOutput.writeVInt(termInfo.skipOffset);
    }
    dataOutput.writeVLong(termInfo.freqPointer);
    dataOutput.writeVLong(termInfo.proxPointer);
    dataOutput.writeVLong(indexEnum.indexPointer);
    for (int j = 1; j < indexDivisor; j++) {
      if (!indexEnum.next()) {
        break;
      }
    }
  }

  fields = new Term[fieldStrs.size()];
  for (int i = 0; i < fields.length; i++) {
    fields[i] = new Term(fieldStrs.get(i));
  }

  dataPagedBytes.freeze(true);
  dataInput = dataPagedBytes.getDataInput();
  indexToDataOffset = indexToTerms.getMutable();

  long ramBytesUsed = RamUsageEstimator.shallowSizeOf(fields);
  ramBytesUsed += RamUsageEstimator.shallowSizeOf(dataInput);
  ramBytesUsed += fields.length * RamUsageEstimator.shallowSizeOfInstance(Term.class);
  ramBytesUsed += dataPagedBytes.ramBytesUsed();
  ramBytesUsed += indexToDataOffset.ramBytesUsed();
  this.ramBytesUsed = ramBytesUsed;
}
项目:search    文件:TermInfosReaderIndex.java   
/**
 * Loads the segment information at segment load time.
 * 
 * @param indexEnum
 *          the term enum.
 * @param indexDivisor
 *          the index divisor.
 * @param tiiFileLength
 *          the size of the tii file, used to approximate the size of the
 *          buffer.
 * @param totalIndexInterval
 *          the total index interval.
 */
TermInfosReaderIndex(SegmentTermEnum indexEnum, int indexDivisor, long tiiFileLength, int totalIndexInterval) throws IOException {
  this.totalIndexInterval = totalIndexInterval;
  indexSize = 1 + ((int) indexEnum.size - 1) / indexDivisor;
  skipInterval = indexEnum.skipInterval;
  // this is only an inital size, it will be GCed once the build is complete
  long initialSize = (long) (tiiFileLength * 1.5) / indexDivisor;
  PagedBytes dataPagedBytes = new PagedBytes(estimatePageBits(initialSize));
  PagedBytesDataOutput dataOutput = dataPagedBytes.getDataOutput();

  final int bitEstimate = 1+MathUtil.log(tiiFileLength, 2);
  GrowableWriter indexToTerms = new GrowableWriter(bitEstimate, indexSize, PackedInts.DEFAULT);

  String currentField = null;
  List<String> fieldStrs = new ArrayList<>();
  int fieldCounter = -1;
  for (int i = 0; indexEnum.next(); i++) {
    Term term = indexEnum.term();
    if (currentField == null || !currentField.equals(term.field())) {
      currentField = term.field();
      fieldStrs.add(currentField);
      fieldCounter++;
    }
    TermInfo termInfo = indexEnum.termInfo();
    indexToTerms.set(i, dataOutput.getPosition());
    dataOutput.writeVInt(fieldCounter);
    dataOutput.writeString(term.text());
    dataOutput.writeVInt(termInfo.docFreq);
    if (termInfo.docFreq >= skipInterval) {
      dataOutput.writeVInt(termInfo.skipOffset);
    }
    dataOutput.writeVLong(termInfo.freqPointer);
    dataOutput.writeVLong(termInfo.proxPointer);
    dataOutput.writeVLong(indexEnum.indexPointer);
    for (int j = 1; j < indexDivisor; j++) {
      if (!indexEnum.next()) {
        break;
      }
    }
  }

  fields = new Term[fieldStrs.size()];
  for (int i = 0; i < fields.length; i++) {
    fields[i] = new Term(fieldStrs.get(i));
  }

  dataPagedBytes.freeze(true);
  dataInput = dataPagedBytes.getDataInput();
  indexToDataOffset = indexToTerms.getMutable();

  long ramBytesUsed = RamUsageEstimator.shallowSizeOf(fields);
  ramBytesUsed += RamUsageEstimator.shallowSizeOf(dataInput);
  ramBytesUsed += fields.length * RamUsageEstimator.shallowSizeOfInstance(Term.class);
  ramBytesUsed += dataPagedBytes.ramBytesUsed();
  ramBytesUsed += indexToDataOffset.ramBytesUsed();
  this.ramBytesUsed = ramBytesUsed;
}
项目:NYBC    文件:TermInfosReaderIndex.java   
/**
 * Loads the segment information at segment load time.
 * 
 * @param indexEnum
 *          the term enum.
 * @param indexDivisor
 *          the index divisor.
 * @param tiiFileLength
 *          the size of the tii file, used to approximate the size of the
 *          buffer.
 * @param totalIndexInterval
 *          the total index interval.
 */
TermInfosReaderIndex(SegmentTermEnum indexEnum, int indexDivisor, long tiiFileLength, int totalIndexInterval) throws IOException {
  this.totalIndexInterval = totalIndexInterval;
  indexSize = 1 + ((int) indexEnum.size - 1) / indexDivisor;
  skipInterval = indexEnum.skipInterval;
  // this is only an inital size, it will be GCed once the build is complete
  long initialSize = (long) (tiiFileLength * 1.5) / indexDivisor;
  PagedBytes dataPagedBytes = new PagedBytes(estimatePageBits(initialSize));
  PagedBytesDataOutput dataOutput = dataPagedBytes.getDataOutput();

  final int bitEstimate = 1+MathUtil.log(tiiFileLength, 2);
  GrowableWriter indexToTerms = new GrowableWriter(bitEstimate, indexSize, PackedInts.DEFAULT);

  String currentField = null;
  List<String> fieldStrs = new ArrayList<String>();
  int fieldCounter = -1;
  for (int i = 0; indexEnum.next(); i++) {
    Term term = indexEnum.term();
    if (currentField == null || !currentField.equals(term.field())) {
      currentField = term.field();
      fieldStrs.add(currentField);
      fieldCounter++;
    }
    TermInfo termInfo = indexEnum.termInfo();
    indexToTerms.set(i, dataOutput.getPosition());
    dataOutput.writeVInt(fieldCounter);
    dataOutput.writeString(term.text());
    dataOutput.writeVInt(termInfo.docFreq);
    if (termInfo.docFreq >= skipInterval) {
      dataOutput.writeVInt(termInfo.skipOffset);
    }
    dataOutput.writeVLong(termInfo.freqPointer);
    dataOutput.writeVLong(termInfo.proxPointer);
    dataOutput.writeVLong(indexEnum.indexPointer);
    for (int j = 1; j < indexDivisor; j++) {
      if (!indexEnum.next()) {
        break;
      }
    }
  }

  fields = new Term[fieldStrs.size()];
  for (int i = 0; i < fields.length; i++) {
    fields[i] = new Term(fieldStrs.get(i));
  }

  dataPagedBytes.freeze(true);
  dataInput = dataPagedBytes.getDataInput();
  indexToDataOffset = indexToTerms.getMutable();
}
项目:read-open-source-code    文件:TermInfosReaderIndex.java   
/**
 * Loads the segment information at segment load time.
 * 
 * @param indexEnum
 *          the term enum.
 * @param indexDivisor
 *          the index divisor.
 * @param tiiFileLength
 *          the size of the tii file, used to approximate the size of the
 *          buffer.
 * @param totalIndexInterval
 *          the total index interval.
 */
TermInfosReaderIndex(SegmentTermEnum indexEnum, int indexDivisor, long tiiFileLength, int totalIndexInterval) throws IOException {
  this.totalIndexInterval = totalIndexInterval;
  indexSize = 1 + ((int) indexEnum.size - 1) / indexDivisor;
  skipInterval = indexEnum.skipInterval;
  // this is only an inital size, it will be GCed once the build is complete
  long initialSize = (long) (tiiFileLength * 1.5) / indexDivisor;
  PagedBytes dataPagedBytes = new PagedBytes(estimatePageBits(initialSize));
  PagedBytesDataOutput dataOutput = dataPagedBytes.getDataOutput();

  final int bitEstimate = 1+MathUtil.log(tiiFileLength, 2);
  GrowableWriter indexToTerms = new GrowableWriter(bitEstimate, indexSize, PackedInts.DEFAULT);

  String currentField = null;
  List<String> fieldStrs = new ArrayList<String>();
  int fieldCounter = -1;
  for (int i = 0; indexEnum.next(); i++) {
    Term term = indexEnum.term();
    if (currentField == null || !currentField.equals(term.field())) {
      currentField = term.field();
      fieldStrs.add(currentField);
      fieldCounter++;
    }
    TermInfo termInfo = indexEnum.termInfo();
    indexToTerms.set(i, dataOutput.getPosition());
    dataOutput.writeVInt(fieldCounter);
    dataOutput.writeString(term.text());
    dataOutput.writeVInt(termInfo.docFreq);
    if (termInfo.docFreq >= skipInterval) {
      dataOutput.writeVInt(termInfo.skipOffset);
    }
    dataOutput.writeVLong(termInfo.freqPointer);
    dataOutput.writeVLong(termInfo.proxPointer);
    dataOutput.writeVLong(indexEnum.indexPointer);
    for (int j = 1; j < indexDivisor; j++) {
      if (!indexEnum.next()) {
        break;
      }
    }
  }

  fields = new Term[fieldStrs.size()];
  for (int i = 0; i < fields.length; i++) {
    fields[i] = new Term(fieldStrs.get(i));
  }

  dataPagedBytes.freeze(true);
  dataInput = dataPagedBytes.getDataInput();
  indexToDataOffset = indexToTerms.getMutable();

  ramBytesUsed = fields.length * (RamUsageEstimator.NUM_BYTES_OBJECT_REF + RamUsageEstimator.shallowSizeOfInstance(Term.class))
      + dataPagedBytes.ramBytesUsed() + indexToDataOffset.ramBytesUsed();
}
项目:read-open-source-code    文件:TermInfosReaderIndex.java   
/**
 * Loads the segment information at segment load time.
 * 
 * @param indexEnum
 *          the term enum.
 * @param indexDivisor
 *          the index divisor.
 * @param tiiFileLength
 *          the size of the tii file, used to approximate the size of the
 *          buffer.
 * @param totalIndexInterval
 *          the total index interval.
 */
TermInfosReaderIndex(SegmentTermEnum indexEnum, int indexDivisor, long tiiFileLength, int totalIndexInterval) throws IOException {
  this.totalIndexInterval = totalIndexInterval;
  indexSize = 1 + ((int) indexEnum.size - 1) / indexDivisor;
  skipInterval = indexEnum.skipInterval;
  // this is only an inital size, it will be GCed once the build is complete
  long initialSize = (long) (tiiFileLength * 1.5) / indexDivisor;
  PagedBytes dataPagedBytes = new PagedBytes(estimatePageBits(initialSize));
  PagedBytesDataOutput dataOutput = dataPagedBytes.getDataOutput();

  final int bitEstimate = 1+MathUtil.log(tiiFileLength, 2);
  GrowableWriter indexToTerms = new GrowableWriter(bitEstimate, indexSize, PackedInts.DEFAULT);

  String currentField = null;
  List<String> fieldStrs = new ArrayList<String>();
  int fieldCounter = -1;
  for (int i = 0; indexEnum.next(); i++) {
    Term term = indexEnum.term();
    if (currentField == null || !currentField.equals(term.field())) {
      currentField = term.field();
      fieldStrs.add(currentField);
      fieldCounter++;
    }
    TermInfo termInfo = indexEnum.termInfo();
    indexToTerms.set(i, dataOutput.getPosition());
    dataOutput.writeVInt(fieldCounter);
    dataOutput.writeString(term.text());
    dataOutput.writeVInt(termInfo.docFreq);
    if (termInfo.docFreq >= skipInterval) {
      dataOutput.writeVInt(termInfo.skipOffset);
    }
    dataOutput.writeVLong(termInfo.freqPointer);
    dataOutput.writeVLong(termInfo.proxPointer);
    dataOutput.writeVLong(indexEnum.indexPointer);
    for (int j = 1; j < indexDivisor; j++) {
      if (!indexEnum.next()) {
        break;
      }
    }
  }

  fields = new Term[fieldStrs.size()];
  for (int i = 0; i < fields.length; i++) {
    fields[i] = new Term(fieldStrs.get(i));
  }

  dataPagedBytes.freeze(true);
  dataInput = dataPagedBytes.getDataInput();
  indexToDataOffset = indexToTerms.getMutable();

  ramBytesUsed = fields.length * (RamUsageEstimator.NUM_BYTES_OBJECT_REF + RamUsageEstimator.shallowSizeOfInstance(Term.class))
      + dataPagedBytes.ramBytesUsed() + indexToDataOffset.ramBytesUsed();
}
项目:read-open-source-code    文件:TermInfosReaderIndex.java   
/**
 * Loads the segment information at segment load time.
 * 
 * @param indexEnum
 *          the term enum.
 * @param indexDivisor
 *          the index divisor.
 * @param tiiFileLength
 *          the size of the tii file, used to approximate the size of the
 *          buffer.
 * @param totalIndexInterval
 *          the total index interval.
 */
TermInfosReaderIndex(SegmentTermEnum indexEnum, int indexDivisor, long tiiFileLength, int totalIndexInterval) throws IOException {
  this.totalIndexInterval = totalIndexInterval;
  indexSize = 1 + ((int) indexEnum.size - 1) / indexDivisor;
  skipInterval = indexEnum.skipInterval;
  // this is only an inital size, it will be GCed once the build is complete
  long initialSize = (long) (tiiFileLength * 1.5) / indexDivisor;
  PagedBytes dataPagedBytes = new PagedBytes(estimatePageBits(initialSize));
  PagedBytesDataOutput dataOutput = dataPagedBytes.getDataOutput();

  final int bitEstimate = 1+MathUtil.log(tiiFileLength, 2);
  GrowableWriter indexToTerms = new GrowableWriter(bitEstimate, indexSize, PackedInts.DEFAULT);

  String currentField = null;
  List<String> fieldStrs = new ArrayList<>();
  int fieldCounter = -1;
  for (int i = 0; indexEnum.next(); i++) {
    Term term = indexEnum.term();
    if (currentField == null || !currentField.equals(term.field())) {
      currentField = term.field();
      fieldStrs.add(currentField);
      fieldCounter++;
    }
    TermInfo termInfo = indexEnum.termInfo();
    indexToTerms.set(i, dataOutput.getPosition());
    dataOutput.writeVInt(fieldCounter);
    dataOutput.writeString(term.text());
    dataOutput.writeVInt(termInfo.docFreq);
    if (termInfo.docFreq >= skipInterval) {
      dataOutput.writeVInt(termInfo.skipOffset);
    }
    dataOutput.writeVLong(termInfo.freqPointer);
    dataOutput.writeVLong(termInfo.proxPointer);
    dataOutput.writeVLong(indexEnum.indexPointer);
    for (int j = 1; j < indexDivisor; j++) {
      if (!indexEnum.next()) {
        break;
      }
    }
  }

  fields = new Term[fieldStrs.size()];
  for (int i = 0; i < fields.length; i++) {
    fields[i] = new Term(fieldStrs.get(i));
  }

  dataPagedBytes.freeze(true);
  dataInput = dataPagedBytes.getDataInput();
  indexToDataOffset = indexToTerms.getMutable();

  long ramBytesUsed = RamUsageEstimator.shallowSizeOf(fields);
  ramBytesUsed += RamUsageEstimator.shallowSizeOf(dataInput);
  ramBytesUsed += fields.length * RamUsageEstimator.shallowSizeOfInstance(Term.class);
  ramBytesUsed += dataPagedBytes.ramBytesUsed();
  ramBytesUsed += indexToDataOffset.ramBytesUsed();
  this.ramBytesUsed = ramBytesUsed;
}
项目:Maskana-Gestor-de-Conocimiento    文件:TermInfosReaderIndex.java   
/**
 * Loads the segment information at segment load time.
 * 
 * @param indexEnum
 *          the term enum.
 * @param indexDivisor
 *          the index divisor.
 * @param tiiFileLength
 *          the size of the tii file, used to approximate the size of the
 *          buffer.
 * @param totalIndexInterval
 *          the total index interval.
 */
TermInfosReaderIndex(SegmentTermEnum indexEnum, int indexDivisor, long tiiFileLength, int totalIndexInterval) throws IOException {
  this.totalIndexInterval = totalIndexInterval;
  indexSize = 1 + ((int) indexEnum.size - 1) / indexDivisor;
  skipInterval = indexEnum.skipInterval;
  // this is only an inital size, it will be GCed once the build is complete
  long initialSize = (long) (tiiFileLength * 1.5) / indexDivisor;
  PagedBytes dataPagedBytes = new PagedBytes(estimatePageBits(initialSize));
  PagedBytesDataOutput dataOutput = dataPagedBytes.getDataOutput();

  final int bitEstimate = 1+MathUtil.log(tiiFileLength, 2);
  GrowableWriter indexToTerms = new GrowableWriter(bitEstimate, indexSize, PackedInts.DEFAULT);

  String currentField = null;
  List<String> fieldStrs = new ArrayList<String>();
  int fieldCounter = -1;
  for (int i = 0; indexEnum.next(); i++) {
    Term term = indexEnum.term();
    if (currentField == null || !currentField.equals(term.field())) {
      currentField = term.field();
      fieldStrs.add(currentField);
      fieldCounter++;
    }
    TermInfo termInfo = indexEnum.termInfo();
    indexToTerms.set(i, dataOutput.getPosition());
    dataOutput.writeVInt(fieldCounter);
    dataOutput.writeString(term.text());
    dataOutput.writeVInt(termInfo.docFreq);
    if (termInfo.docFreq >= skipInterval) {
      dataOutput.writeVInt(termInfo.skipOffset);
    }
    dataOutput.writeVLong(termInfo.freqPointer);
    dataOutput.writeVLong(termInfo.proxPointer);
    dataOutput.writeVLong(indexEnum.indexPointer);
    for (int j = 1; j < indexDivisor; j++) {
      if (!indexEnum.next()) {
        break;
      }
    }
  }

  fields = new Term[fieldStrs.size()];
  for (int i = 0; i < fields.length; i++) {
    fields[i] = new Term(fieldStrs.get(i));
  }

  dataPagedBytes.freeze(true);
  dataInput = dataPagedBytes.getDataInput();
  indexToDataOffset = indexToTerms.getMutable();
}