Package org.apache.druid.query.aggregation

Interface PostAggregator

All Superinterfaces:

Cacheable

All Known Implementing Classes:

ArithmeticPostAggregator, ConstantPostAggregator, DoubleGreatestPostAggregator, DoubleLeastPostAggregator, ExpressionPostAggregator, FieldAccessPostAggregator, FinalizingFieldAccessPostAggregator, HyperUniqueFinalizingPostAggregator, JavaScriptPostAggregator, LongGreatestPostAggregator, LongLeastPostAggregator

public interface PostAggregator
extends Cacheable

Functionally similar to an Aggregator. See the Aggregator interface for more comments.

Method Summary

Modifier and Type	Method	Description
Object	compute(Map<String,Object> combinedAggregators)
PostAggregator	decorate(Map<String,AggregatorFactory> aggregators)	Allows returning an enriched post aggregator, built from contextual information available from the given map of AggregatorFactory keyed by their names.
Comparator	getComparator()
Set<String>	getDependentFields()
String	getName()
default ValueType	getType()	Deprecated.
default ColumnType	getType(ColumnInspector signature)	Return the output type of a row processed with this post aggregator.

Methods inherited from interface org.apache.druid.java.util.common.Cacheable

getCacheKey

Method Details

getDependentFields

Set<String> getDependentFields()

getComparator

Comparator getComparator()

compute

@Nullable
Object compute(Map<String,Object> combinedAggregators)

getName

@Nullable
String getName()

getType

@Nullable
default ColumnType getType(ColumnInspector signature)

Return the output type of a row processed with this post aggregator. Refer to the ColumnType javadocs for details on the implications of choosing a type.

Parameters:

signature -

getType

@Deprecated
default ValueType getType()

Deprecated.

This method is deprecated and will be removed soon. Use getType(ColumnInspector) instead. Do not call this method, it will likely produce incorrect results, it exists for backwards compatibility.

decorate

PostAggregator decorate(Map<String,AggregatorFactory> aggregators)

Allows returning an enriched post aggregator, built from contextual information available from the given map of AggregatorFactory keyed by their names. Callers must call this method before calling compute(java.util.Map<java.lang.String, java.lang.Object>) or getComparator(). This is typically done in the constructor of queries which support post aggregators, via Queries.prepareAggregations(java.util.List<org.apache.druid.query.aggregation.AggregatorFactory>, java.util.List<org.apache.druid.query.aggregation.PostAggregator>).