/**
* Licensed to the Apache Software Foundation (ASF) under one
* or more contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. The ASF licenses this file
* to you 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 at
*
* 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 org.apache.drill.exec.physical.impl.aggregate;
import javax.inject.Named;
import org.apache.drill.exec.exception.SchemaChangeException;
import org.apache.drill.exec.ops.FragmentContext;
import org.apache.drill.exec.physical.impl.aggregate.AggBatch.AggOutcome;
import org.apache.drill.exec.record.BatchSchema;
import org.apache.drill.exec.record.RecordBatch;
import org.apache.drill.exec.record.RecordBatch.IterOutcome;
import org.apache.drill.exec.record.VectorWrapper;
import org.apache.drill.exec.vector.allocator.VectorAllocator;
public abstract class AggTemplate implements Aggregator {
private static final org.slf4j.Logger logger = org.slf4j.LoggerFactory.getLogger(Aggregator.class);
private static final boolean EXTRA_DEBUG = false;
private static final String TOO_BIG_ERROR = "Couldn't add value to an empty batch. This likely means that a single value is too long for a varlen field.";
private IterOutcome lastOutcome = null;
private boolean first = true;
private boolean newSchema = false;
private int previousIndex = 0;
private int underlyingIndex = 0;
private int currentIndex;
private int addedRecordCount = 0;
private boolean pendingOutput = false;
private IterOutcome outcome;
private int outputCount = 0;
private RecordBatch incoming;
private BatchSchema schema;
private RecordBatch outgoing;
private VectorAllocator[] allocators;
private FragmentContext context;
private InternalBatch remainderBatch;
@Override
public void setup(FragmentContext context, RecordBatch incoming, RecordBatch outgoing, VectorAllocator[] allocators) throws SchemaChangeException {
this.allocators = allocators;
this.context = context;
this.incoming = incoming;
this.schema = incoming.getSchema();
this.allocators = allocators;
this.outgoing = outgoing;
setupInterior(incoming, outgoing);
this.currentIndex = this.getVectorIndex(underlyingIndex);
}
private void allocateOutgoing() {
for (VectorAllocator a : allocators) {
if(EXTRA_DEBUG) logger.debug("Allocating {} with {} records.", a, 20000);
a.alloc(20000);
}
}
@Override
public IterOutcome getOutcome() {
return outcome;
}
@Override
public int getOutputCount() {
return outputCount;
}
private AggOutcome tooBigFailure(){
context.fail(new Exception(TOO_BIG_ERROR));
this.outcome = IterOutcome.STOP;
return AggOutcome.CLEANUP_AND_RETURN;
}
@Override
public AggOutcome doWork() {
try{ // outside loop to ensure that first is set to false after the first run.
// if we're in the first state, allocate outgoing.
if(first){
allocateOutgoing();
}
// pick up a remainder batch if we have one.
if(remainderBatch != null){
if (!outputToBatch( previousIndex )) return tooBigFailure();
remainderBatch.clear();
remainderBatch = null;
return setOkAndReturn();
}
// setup for new output and pick any remainder.
if (pendingOutput) {
allocateOutgoing();
pendingOutput = false;
if (!outputToBatch( previousIndex)) return tooBigFailure();
}
if(newSchema){
return AggOutcome.UPDATE_AGGREGATOR;
}
if(lastOutcome != null){
outcome = lastOutcome;
return AggOutcome.CLEANUP_AND_RETURN;
}
outside: while(true){
// loop through existing records, adding as necessary.
for (; underlyingIndex < incoming.getRecordCount(); incIndex()) {
if(EXTRA_DEBUG) logger.debug("Doing loop with values underlying {}, current {}", underlyingIndex, currentIndex);
if (isSame( previousIndex, currentIndex )) {
if(EXTRA_DEBUG) logger.debug("Values were found the same, adding.");
addRecordInc(currentIndex);
} else {
if(EXTRA_DEBUG) logger.debug("Values were different, outputting previous batch.");
if (outputToBatch(previousIndex)) {
if(EXTRA_DEBUG) logger.debug("Output successful.");
addRecordInc(currentIndex);
} else {
if(EXTRA_DEBUG) logger.debug("Output failed.");
if(outputCount == 0) return tooBigFailure();
// mark the pending output but move forward for the next cycle.
pendingOutput = true;
previousIndex = currentIndex;
incIndex();
return setOkAndReturn();
}
}
previousIndex = currentIndex;
}
InternalBatch previous = null;
try{
while(true){
previous = new InternalBatch(incoming);
IterOutcome out = incoming.next();
if(EXTRA_DEBUG) logger.debug("Received IterOutcome of {}", out);
switch(out){
case NONE:
lastOutcome = out;
if(addedRecordCount > 0){
if( !outputToBatchPrev( previous, previousIndex, outputCount) ) remainderBatch = previous;
if(EXTRA_DEBUG) logger.debug("Received no more batches, returning.");
return setOkAndReturn();
}else{
outcome = out;
return AggOutcome.CLEANUP_AND_RETURN;
}
case NOT_YET:
this.outcome = out;
return AggOutcome.RETURN_OUTCOME;
case OK_NEW_SCHEMA:
if(EXTRA_DEBUG) logger.debug("Received new schema. Batch has {} records.", incoming.getRecordCount());
if(addedRecordCount > 0){
if( !outputToBatchPrev( previous, previousIndex, outputCount) ) remainderBatch = previous;
if(EXTRA_DEBUG) logger.debug("Wrote out end of previous batch, returning.");
newSchema = true;
return setOkAndReturn();
}
cleanup();
return AggOutcome.UPDATE_AGGREGATOR;
case OK:
resetIndex();
if(incoming.getRecordCount() == 0){
continue;
}else{
if(isSamePrev(previousIndex , previous, currentIndex)){
if(EXTRA_DEBUG) logger.debug("New value was same as last value of previous batch, adding.");
addRecordInc(currentIndex);
previousIndex = currentIndex;
incIndex();
if(EXTRA_DEBUG) logger.debug("Continuing outside");
continue outside;
}else{ // not the same
if(EXTRA_DEBUG) logger.debug("This is not the same as the previous, add record and continue outside.");
previousIndex = currentIndex;
if(addedRecordCount > 0){
if( !outputToBatchPrev( previous, previousIndex, outputCount) ){
remainderBatch = previous;
return setOkAndReturn();
}
continue outside;
}
}
}
case STOP:
default:
lastOutcome = out;
outcome = out;
return AggOutcome.CLEANUP_AND_RETURN;
}
}
}finally{
// make sure to clear previous if we haven't saved it.
if(remainderBatch == null && previous != null){
previous.clear();
}
}
}
}finally{
if(first) first = !first;
}
}
private final void incIndex(){
underlyingIndex++;
if(underlyingIndex >= incoming.getRecordCount()){
currentIndex = Integer.MAX_VALUE;
return;
}
currentIndex = getVectorIndex(underlyingIndex);
}
private final void resetIndex(){
underlyingIndex = -1;
incIndex();
}
private final AggOutcome setOkAndReturn(){
if(first){
this.outcome = IterOutcome.OK_NEW_SCHEMA;
}else{
this.outcome = IterOutcome.OK;
}
for(VectorWrapper<?> v : outgoing){
v.getValueVector().getMutator().setValueCount(outputCount);
}
return AggOutcome.RETURN_OUTCOME;
}
private final boolean outputToBatch(int inIndex){
boolean success = outputRecordKeys(inIndex, outputCount) //
&& outputRecordValues(outputCount) //
&& resetValues();
if(success){
if(EXTRA_DEBUG) logger.debug("Outputting values to {}", outputCount);
outputCount++;
addedRecordCount = 0;
}
return success;
}
private final boolean outputToBatchPrev(InternalBatch b1, int inIndex, int outIndex){
boolean success = outputRecordKeysPrev(b1, inIndex, outIndex) //
&& outputRecordValues(outIndex) //
&& resetValues();
if(success){
outputCount++;
addedRecordCount = 0;
}
return success;
}
private void addRecordInc(int index){
addRecord(index);
this.addedRecordCount++;
}
@Override
public void cleanup(){
if(remainderBatch != null) remainderBatch.clear();
}
public abstract void setupInterior(@Named("incoming") RecordBatch incoming, @Named("outgoing") RecordBatch outgoing) throws SchemaChangeException;
public abstract boolean isSame(@Named("index1") int index1, @Named("index2") int index2);
public abstract boolean isSamePrev(@Named("b1Index") int b1Index, @Named("b1") InternalBatch b1, @Named("b2Index") int b2Index);
public abstract void addRecord(@Named("index") int index);
public abstract boolean outputRecordKeys(@Named("inIndex") int inIndex, @Named("outIndex") int outIndex);
public abstract boolean outputRecordKeysPrev(@Named("previous") InternalBatch previous, @Named("previousIndex") int previousIndex, @Named("outIndex") int outIndex);
public abstract boolean outputRecordValues(@Named("outIndex") int outIndex);
public abstract int getVectorIndex(@Named("recordIndex") int recordIndex);
public abstract boolean resetValues();
}