Java Code Examples for org.apache.hadoop.hbase.KeyValue#FAMILY_LENGTH_SIZE

/**
 * Estimate based on keyvalue's serialization format in the RPC layer. Note that there is an extra
 * SIZEOF_INT added to the size here that indicates the actual length of the cell for cases where
 * cell's are serialized in a contiguous format (For eg in RPCs).
 * @return Estimate of the <code>cell</code> size in bytes plus an extra SIZEOF_INT indicating the
 *         actual cell length.
 */
public static int estimatedSerializedSizeOf(final Cell cell) {
  if (cell instanceof ExtendedCell) {
    return ((ExtendedCell) cell).getSerializedSize(true) + Bytes.SIZEOF_INT;
  }

  return getSumOfCellElementLengths(cell) +
      // Use the KeyValue's infrastructure size presuming that another implementation would have
      // same basic cost.
      KeyValue.ROW_LENGTH_SIZE + KeyValue.FAMILY_LENGTH_SIZE +
      // Serialization is probably preceded by a length (it is in the KeyValueCodec at least).
      Bytes.SIZEOF_INT;
}
 

@Override
public int length() {
  // For ROW_COL blooms we use bytes
  // <RK length> (2 bytes) , <RK>, 0 (one byte CF length), <CQ>, <TS> (8 btes), <TYPE> ( 1 byte)
  return KeyValue.ROW_LENGTH_SIZE + this.t.getRowLength() + KeyValue.FAMILY_LENGTH_SIZE
      + this.t.getQualifierLength() + KeyValue.TIMESTAMP_TYPE_SIZE;
}
 

private int compressSingleKeyValue(DataOutputStream out, Cell cell, Cell prevCell)
    throws IOException {
  int flag = 0; // Do not use more bits that can fit into a byte
  int kLength = KeyValueUtil.keyLength(cell);
  int vLength = cell.getValueLength();

  long timestamp;
  long diffTimestamp = 0;
  int diffTimestampFitsInBytes = 0;
  int timestampFitsInBytes;
  int commonPrefix = 0;

  if (prevCell == null) {
    timestamp = cell.getTimestamp();
    if (timestamp < 0) {
      flag |= FLAG_TIMESTAMP_SIGN;
      timestamp = -timestamp;
    }
    timestampFitsInBytes = ByteBufferUtils.longFitsIn(timestamp);
    flag |= (timestampFitsInBytes - 1) << SHIFT_TIMESTAMP_LENGTH;
    // put column family
    byte familyLength = cell.getFamilyLength();
    out.write(familyLength);
    PrivateCellUtil.writeFamily(out, cell, familyLength);
  } else {
    // Finding common prefix
    int preKeyLength = KeyValueUtil.keyLength(prevCell);
    commonPrefix = PrivateCellUtil.findCommonPrefixInFlatKey(cell, prevCell, true, false);
    if (kLength == preKeyLength) {
      flag |= FLAG_SAME_KEY_LENGTH;
    }
    if (vLength == prevCell.getValueLength()) {
      flag |= FLAG_SAME_VALUE_LENGTH;
    }
    if (cell.getTypeByte() == prevCell.getTypeByte()) {
      flag |= FLAG_SAME_TYPE;
    }
    // don't compress timestamp and type using prefix encode timestamp
    timestamp = cell.getTimestamp();
    diffTimestamp = prevCell.getTimestamp() - timestamp;
    boolean negativeTimestamp = timestamp < 0;
    if (negativeTimestamp) {
      timestamp = -timestamp;
    }
    timestampFitsInBytes = ByteBufferUtils.longFitsIn(timestamp);
    boolean minusDiffTimestamp = diffTimestamp < 0;
    if (minusDiffTimestamp) {
      diffTimestamp = -diffTimestamp;
    }
    diffTimestampFitsInBytes = ByteBufferUtils.longFitsIn(diffTimestamp);
    if (diffTimestampFitsInBytes < timestampFitsInBytes) {
      flag |= (diffTimestampFitsInBytes - 1) << SHIFT_TIMESTAMP_LENGTH;
      flag |= FLAG_TIMESTAMP_IS_DIFF;
      if (minusDiffTimestamp) {
        flag |= FLAG_TIMESTAMP_SIGN;
      }
    } else {
      flag |= (timestampFitsInBytes - 1) << SHIFT_TIMESTAMP_LENGTH;
      if (negativeTimestamp) {
        flag |= FLAG_TIMESTAMP_SIGN;
      }
    }
  }
  out.write(flag);
  if ((flag & FLAG_SAME_KEY_LENGTH) == 0) {
    ByteBufferUtils.putCompressedInt(out, kLength);
  }
  if ((flag & FLAG_SAME_VALUE_LENGTH) == 0) {
    ByteBufferUtils.putCompressedInt(out, vLength);
  }
  ByteBufferUtils.putCompressedInt(out, commonPrefix);
  short rLen = cell.getRowLength();
  if (commonPrefix < rLen + KeyValue.ROW_LENGTH_SIZE) {
    // Previous and current rows are different. Copy the differing part of
    // the row, skip the column family, and copy the qualifier.
    PrivateCellUtil.writeRowKeyExcludingCommon(cell, rLen, commonPrefix, out);
    PrivateCellUtil.writeQualifier(out, cell, cell.getQualifierLength());
  } else {
    // The common part includes the whole row. As the column family is the
    // same across the whole file, it will automatically be included in the
    // common prefix, so we need not special-case it here.
    // What we write here is the non common part of the qualifier
    int commonQualPrefix = commonPrefix - (rLen + KeyValue.ROW_LENGTH_SIZE)
        - (cell.getFamilyLength() + KeyValue.FAMILY_LENGTH_SIZE);
    PrivateCellUtil.writeQualifierSkippingBytes(out, cell, cell.getQualifierLength(),
      commonQualPrefix);
  }
  if ((flag & FLAG_TIMESTAMP_IS_DIFF) == 0) {
    ByteBufferUtils.putLong(out, timestamp, timestampFitsInBytes);
  } else {
    ByteBufferUtils.putLong(out, diffTimestamp, diffTimestampFitsInBytes);
  }

  if ((flag & FLAG_SAME_TYPE) == 0) {
    out.write(cell.getTypeByte());
  }
  PrivateCellUtil.writeValue(out, cell, vLength);
  return kLength + vLength + KeyValue.KEYVALUE_INFRASTRUCTURE_SIZE;
}
 

private int compressSingleKeyValue(DataOutputStream out, Cell cell, Cell prevCell)
    throws IOException {
  int flag = 0; // Do not use more bits than will fit into a byte
  int kLength = KeyValueUtil.keyLength(cell);
  int vLength = cell.getValueLength();

  if (prevCell == null) {
    // copy the key, there is no common prefix with none
    out.write(flag);
    ByteBufferUtils.putCompressedInt(out, kLength);
    ByteBufferUtils.putCompressedInt(out, vLength);
    ByteBufferUtils.putCompressedInt(out, 0);
    PrivateCellUtil.writeFlatKey(cell, (DataOutput)out);
    // Write the value part
    PrivateCellUtil.writeValue(out, cell, cell.getValueLength());
  } else {
    int preKeyLength = KeyValueUtil.keyLength(prevCell);
    int preValLength = prevCell.getValueLength();
    // find a common prefix and skip it
    int commonPrefix = PrivateCellUtil.findCommonPrefixInFlatKey(cell, prevCell, true, false);

    if (kLength == preKeyLength) {
      flag |= FLAG_SAME_KEY_LENGTH;
    }
    if (vLength == prevCell.getValueLength()) {
      flag |= FLAG_SAME_VALUE_LENGTH;
    }
    if (cell.getTypeByte() == prevCell.getTypeByte()) {
      flag |= FLAG_SAME_TYPE;
    }

    byte[] curTsBuf = Bytes.toBytes(cell.getTimestamp());
    int commonTimestampPrefix = findCommonTimestampPrefix(curTsBuf,
        Bytes.toBytes(prevCell.getTimestamp()));

    flag |= commonTimestampPrefix << SHIFT_TIMESTAMP_LENGTH;

    // Check if current and previous values are the same. Compare value
    // length first as an optimization.
    if (vLength == preValLength
        && PrivateCellUtil.matchingValue(cell, prevCell, vLength, preValLength)) {
      flag |= FLAG_SAME_VALUE;
    }

    out.write(flag);
    if ((flag & FLAG_SAME_KEY_LENGTH) == 0) {
      ByteBufferUtils.putCompressedInt(out, kLength);
    }
    if ((flag & FLAG_SAME_VALUE_LENGTH) == 0) {
      ByteBufferUtils.putCompressedInt(out, vLength);
    }
    ByteBufferUtils.putCompressedInt(out, commonPrefix);
    short rLen = cell.getRowLength();
    if (commonPrefix < rLen + KeyValue.ROW_LENGTH_SIZE) {
      // Previous and current rows are different. Copy the differing part of
      // the row, skip the column family, and copy the qualifier.
      PrivateCellUtil.writeRowKeyExcludingCommon(cell, rLen, commonPrefix, out);
      PrivateCellUtil.writeQualifier(out, cell, cell.getQualifierLength());
    } else {
      // The common part includes the whole row. As the column family is the
      // same across the whole file, it will automatically be included in the
      // common prefix, so we need not special-case it here.
      // What we write here is the non common part of the qualifier
      int commonQualPrefix = commonPrefix - (rLen + KeyValue.ROW_LENGTH_SIZE)
          - (cell.getFamilyLength() + KeyValue.FAMILY_LENGTH_SIZE);
      PrivateCellUtil.writeQualifierSkippingBytes(out, cell, cell.getQualifierLength(),
        commonQualPrefix);
    }
    // Write non common ts part
    out.write(curTsBuf, commonTimestampPrefix, KeyValue.TIMESTAMP_SIZE - commonTimestampPrefix);

    // Write the type if it is not the same as before.
    if ((flag & FLAG_SAME_TYPE) == 0) {
      out.write(cell.getTypeByte());
    }

    // Write the value if it is not the same as before.
    if ((flag & FLAG_SAME_VALUE) == 0) {
      PrivateCellUtil.writeValue(out, cell, vLength);
    }
  }
  return kLength + vLength + KeyValue.KEYVALUE_INFRASTRUCTURE_SIZE;
}
 

private void writeKeyExcludingCommon(Cell cell, int commonPrefix, DataOutputStream out)
    throws IOException {
  short rLen = cell.getRowLength();
  if (commonPrefix < rLen + KeyValue.ROW_LENGTH_SIZE) {
    // Previous and current rows are different. Need to write the differing part followed by
    // cf,q,ts and type
    PrivateCellUtil.writeRowKeyExcludingCommon(cell, rLen, commonPrefix, out);
    byte fLen = cell.getFamilyLength();
    out.writeByte(fLen);
    PrivateCellUtil.writeFamily(out, cell, fLen);
    PrivateCellUtil.writeQualifier(out, cell, cell.getQualifierLength());
    out.writeLong(cell.getTimestamp());
    out.writeByte(cell.getTypeByte());
  } else {
    // The full row key part is common. CF part will be common for sure as we deal with Cells in
    // same family. Just need write the differing part in q, ts and type
    commonPrefix = commonPrefix - (rLen + KeyValue.ROW_LENGTH_SIZE)
        - (cell.getFamilyLength() + KeyValue.FAMILY_LENGTH_SIZE);
    int qLen = cell.getQualifierLength();
    int commonQualPrefix = Math.min(commonPrefix, qLen);
    int qualPartLenToWrite = qLen - commonQualPrefix;
    if (qualPartLenToWrite > 0) {
      PrivateCellUtil.writeQualifierSkippingBytes(out, cell, qLen, commonQualPrefix);
    }
    commonPrefix -= commonQualPrefix;
    // Common part in TS also?
    if (commonPrefix > 0) {
      int commonTimestampPrefix = Math.min(commonPrefix, KeyValue.TIMESTAMP_SIZE);
      if (commonTimestampPrefix < KeyValue.TIMESTAMP_SIZE) {
        byte[] curTsBuf = Bytes.toBytes(cell.getTimestamp());
        out.write(curTsBuf, commonTimestampPrefix, KeyValue.TIMESTAMP_SIZE
            - commonTimestampPrefix);
      }
      commonPrefix -= commonTimestampPrefix;
      if (commonPrefix == 0) {
        out.writeByte(cell.getTypeByte());
      }
    } else {
      out.writeLong(cell.getTimestamp());
      out.writeByte(cell.getTypeByte());
    }
  }
}
 

@Override
public int getLength() {
  return KEYVALUE_INFRASTRUCTURE_SIZE + KeyValue.ROW_LENGTH_SIZE + row.getLength()
      + KeyValue.FAMILY_LENGTH_SIZE + family.getLength() + qualifier.getLength()
      + KeyValue.TIMESTAMP_SIZE + KeyValue.TYPE_SIZE + value.getLength();
}