Package org.apache.iotdb.db.mpp.plan.planner.plan.node.process

Class AggregationNode

java.lang.Object

org.apache.iotdb.db.mpp.plan.planner.plan.node.PlanNode

org.apache.iotdb.db.mpp.plan.planner.plan.node.process.ProcessNode

org.apache.iotdb.db.mpp.plan.planner.plan.node.process.MultiChildNode

org.apache.iotdb.db.mpp.plan.planner.plan.node.process.AggregationNode

All Implemented Interfaces:

org.apache.iotdb.consensus.common.request.IConsensusRequest

public class AggregationNode
extends MultiChildNode

This node is used to aggregate required series from multiple sources. The source data will be input as a TsBlock, it may be raw data or partial aggregation result. This node will output the final series aggregated result represented by TsBlock.

Field Summary

Fields

Modifier and Type	Field	Description
protected java.util.List<AggregationDescriptor>	aggregationDescriptorList
protected GroupByTimeParameter	groupByTimeParameter
protected Ordering	scanOrder

Fields inherited from class org.apache.iotdb.db.mpp.plan.planner.plan.node.process.MultiChildNode

children

Fields inherited from class org.apache.iotdb.db.mpp.plan.planner.plan.node.PlanNode

CHILD_COUNT_NO_LIMIT, NO_CHILD_ALLOWED, ONE_CHILD

Constructor Summary

Constructors

Constructor	Description
AggregationNode(PlanNodeId id, java.util.List<PlanNode> children, java.util.List<AggregationDescriptor> aggregationDescriptorList, GroupByTimeParameter groupByTimeParameter, Ordering scanOrder)
AggregationNode(PlanNodeId id, java.util.List<AggregationDescriptor> aggregationDescriptorList, GroupByTimeParameter groupByTimeParameter, Ordering scanOrder)

Method Summary

Modifier and Type	Method	Description
<R,C> R	accept(PlanVisitor<R,C> visitor, C context)
void	addChild(PlanNode child)
int	allowedChildCount()
PlanNode	clone()
static AggregationNode	deserialize(java.nio.ByteBuffer byteBuffer)
boolean	equals(java.lang.Object o)
java.util.List<AggregationDescriptor>	getAggregationDescriptorList()
java.util.List<PlanNode>	getChildren()
static java.util.List<AggregationDescriptor>	getDeduplicatedDescriptors(java.util.List<AggregationDescriptor> aggregationDescriptors)	If aggregation function COUNT and AVG for one time series appears at the same time, and outputs intermediate result, the output columns will be like | COUNT | COUNT | SUM |.
GroupByTimeParameter	getGroupByTimeParameter()
java.util.List<java.lang.String>	getOutputColumnNames()
Ordering	getScanOrder()
int	hashCode()
protected void	serializeAttributes(java.io.DataOutputStream stream)
protected void	serializeAttributes(java.nio.ByteBuffer byteBuffer)
void	setAggregationDescriptorList(java.util.List<AggregationDescriptor> aggregationDescriptorList)
java.lang.String	toString()

Methods inherited from class org.apache.iotdb.db.mpp.plan.planner.plan.node.process.MultiChildNode

setChildren

Methods inherited from class org.apache.iotdb.db.mpp.plan.planner.plan.node.PlanNode

cloneWithChildren, getPlanNodeId, serialize, serialize, serializeToByteBuffer, setPlanNodeId

Methods inherited from class java.lang.Object

finalize, getClass, notify, notifyAll, wait, wait, wait

Field Detail

aggregationDescriptorList

protected java.util.List<AggregationDescriptor> aggregationDescriptorList

groupByTimeParameter

@Nullable
protected GroupByTimeParameter groupByTimeParameter

scanOrder

protected Ordering scanOrder

Constructor Detail

AggregationNode

public AggregationNode(PlanNodeId id,
                       java.util.List<AggregationDescriptor> aggregationDescriptorList,
                       @Nullable
                       GroupByTimeParameter groupByTimeParameter,
                       Ordering scanOrder)

AggregationNode

public AggregationNode(PlanNodeId id,
                       java.util.List<PlanNode> children,
                       java.util.List<AggregationDescriptor> aggregationDescriptorList,
                       @Nullable
                       GroupByTimeParameter groupByTimeParameter,
                       Ordering scanOrder)

Method Detail

getAggregationDescriptorList

public java.util.List<AggregationDescriptor> getAggregationDescriptorList()

getGroupByTimeParameter

@Nullable
public GroupByTimeParameter getGroupByTimeParameter()

getScanOrder

public Ordering getScanOrder()

getChildren

public java.util.List<PlanNode> getChildren()

Specified by:

getChildren in class PlanNode

addChild

public void addChild(PlanNode child)

Specified by:

addChild in class PlanNode

allowedChildCount

public int allowedChildCount()

Specified by:

allowedChildCount in class PlanNode

clone

public PlanNode clone()

Specified by:

clone in class PlanNode

getOutputColumnNames

public java.util.List<java.lang.String> getOutputColumnNames()

Specified by:

getOutputColumnNames in class PlanNode

setAggregationDescriptorList

public void setAggregationDescriptorList(java.util.List<AggregationDescriptor> aggregationDescriptorList)

accept

public <R,C> R accept(PlanVisitor<R,C> visitor,
                            C context)

Overrides:

accept in class PlanNode

serializeAttributes

protected void serializeAttributes(java.nio.ByteBuffer byteBuffer)

Specified by:

serializeAttributes in class PlanNode

serializeAttributes

protected void serializeAttributes(java.io.DataOutputStream stream)
                            throws java.io.IOException

Specified by:

serializeAttributes in class PlanNode

Throws:

java.io.IOException

deserialize

public static AggregationNode deserialize(java.nio.ByteBuffer byteBuffer)

equals

public boolean equals(java.lang.Object o)

Overrides:

equals in class MultiChildNode

hashCode

public int hashCode()

Overrides:

hashCode in class MultiChildNode

getDeduplicatedDescriptors

public static java.util.List<AggregationDescriptor> getDeduplicatedDescriptors(java.util.List<AggregationDescriptor> aggregationDescriptors)

If aggregation function COUNT and AVG for one time series appears at the same time, and outputs intermediate result, the output columns will be like | COUNT | COUNT | SUM |. In this situation, one COUNT column is not needed. Therefore, when COUNT(or SUM) appears with AVG and outputs intermediate result(if they output final result, they will be all necessary), we need to REMOVE the COUNT aggregation, and only keep AVG function no matter their appearing order.

The related functions include AVG(COUNT AND SUM), FIRST_VALUE(FIRST_VALUE AND MIN_TIME), LAST_VALUE(LAST_VALUE AND MAX_TIME).

toString

public java.lang.String toString()

Overrides:

toString in class java.lang.Object