hyracks/hyracks-storage-am-btree/src/main/java/edu/uci/ics/hyracks/storage/am/btree/api/IPrefixSlotManager.java

/*
 * Copyright 2009-2010 by The Regents of the University of California
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * you may obtain a copy of the License from
 * 
 *     http://www.apache.org/licenses/LICENSE-2.0
 * 
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

package edu.uci.ics.hyracks.storage.am.btree.api;

import edu.uci.ics.hyracks.dataflow.common.data.accessors.ITupleReference;
import edu.uci.ics.hyracks.storage.am.common.api.ISlotManager;
import edu.uci.ics.hyracks.storage.am.common.api.ITreeIndexTupleReference;
import edu.uci.ics.hyracks.storage.am.common.ophelpers.FindTupleMode;
import edu.uci.ics.hyracks.storage.am.common.ophelpers.FindTupleNoExactMatchPolicy;
import edu.uci.ics.hyracks.storage.am.common.ophelpers.MultiComparator;

/**
 *  A slot consists of two fields. The first field is 1 byte and it indicates the slot number of 
 *  a prefix tuple that is called the first field prefixSlotOff. The second field is 3 bytes and 
 *  it points to the start offset of a tuple that is called the second field tupleOff.
 *  
 *  We distinguish between two slot types:
 *      1) prefix slots that point to prefix tuples (a frame is assumed to have a field numPrefixTuples)
 *      2) tuple slots that point to data tuples (a frame is assumed to have a field numTuples)
 *      
 *  A tuple slot contains a tuple pointer and a pointer to a prefix slot (prefix slot number).
 *  
 *  INSERT procedure:
 *      - A tuple insertion may use an existing prefix tuple
 *      - A tuple insertion may never create a new prefix tuple
 *  
 *  Modifying the prefix slots would be extremely expensive because potentially all tuples slots 
 *  would have to change their prefix slot pointers. All prefixes are recomputed during a reorg 
 *  or compaction.
 */
public interface IPrefixSlotManager extends ISlotManager {
    // TODO: Clean up interface after extending ISlotManager.

    public int decodeFirstSlotField(int slot);

    public int decodeSecondSlotField(int slot);

    public int encodeSlotFields(int firstField, int secondField);

    public int findSlot(ITupleReference searchKey, ITreeIndexTupleReference frameTuple,
            ITreeIndexTupleReference framePrefixTuple, MultiComparator multiCmp, FindTupleMode mode,
            FindTupleNoExactMatchPolicy matchPolicy);

    public int insertSlot(int slot, int tupleOff);

    /** 
     * @return the prefix slot number or FieldPrefixSlotManager.TUPLE_UNCOMPRESSED if none found
     */
    public int findPrefix(ITupleReference tuple, ITreeIndexTupleReference framePrefixTuple);

    public int getTupleSlotStartOff();

    public int getTupleSlotEndOff();

    public int getPrefixSlotStartOff();

    public int getPrefixSlotEndOff();

    public int getTupleSlotOff(int tupleIndex);

    public int getPrefixSlotOff(int tupleIndex);

    public int getSlotSize();

    public void setSlot(int offset, int value);

    // functions for testing
    public void setPrefixSlot(int tupleIndex, int slot);

    public void setMultiComparator(MultiComparator cmp);
}