org.apache.druid.query.aggregation

Interface Aggregator

All Superinterfaces:

AutoCloseable, Closeable

All Known Implementing Classes:

CardinalityAggregator, CountAggregator, DelegatingAggregator, DoubleAnyAggregator, DoubleFirstAggregator, DoubleLastAggregator, DoubleMaxAggregator, DoubleMeanAggregator, DoubleMinAggregator, DoubleSumAggregator, ExpressionLambdaAggregator, FilteredAggregator, FloatAnyAggregator, FloatFirstAggregator, FloatLastAggregator, FloatMaxAggregator, FloatMinAggregator, FloatSumAggregator, HistogramAggregator, HyperUniquesAggregator, JavaScriptAggregator, LongAnyAggregator, LongConstantAggregator, LongFirstAggregator, LongLastAggregator, LongMaxAggregator, LongMinAggregator, LongSumAggregator, NoopAggregator, NullableNumericAggregator, NumericAnyAggregator, NumericFirstAggregator, NumericLastAggregator, StringAnyAggregator, StringColumnDoubleAggregatorWrapper, StringColumnFloatAggregatorWrapper, StringColumnLongAggregatorWrapper, StringFirstAggregator, StringLastAggregator, SuppressedAggregatorFactory.SuppressedAggregator

public interface Aggregator
extends Closeable

An Aggregator is an object that can aggregate metrics. Its aggregation-related methods (namely, aggregate() and get()) do not take any arguments as the assumption is that the Aggregator was given something in its constructor that it can use to get at the next bit of data. Thus, an Aggregator can be thought of as a closure over some other thing that is stateful and changes between calls to aggregate(). This is currently (as of this documentation) implemented through the use of ColumnValueSelector objects. During ingestion, OnheapIncrementalIndex uses instances of this class from multiple threads. In particular, ingestion threads call aggregate() simultaneously with query threads calling the various "get" methods. Instances of this class must be thread-safe.

Method Summary

Modifier and Type	Method and Description
void	aggregate() Performs aggregation.
default long	aggregateWithSize() Performs aggregation and returns the increase in required on-heap memory caused by this aggregation step.
void	close()
Object	get()
default double	getDouble() The default implementation casts getFloat() to double.
float	getFloat()
long	getLong()
default boolean	isNull() returns true if aggregator's output type is primitive long/double/float and aggregated value is null, but when aggregated output type is Object, this method always returns false, and users are advised to check nullability for the object returned by get() method.

Method Detail

aggregate

void aggregate()

Performs aggregation.

aggregateWithSize

default long aggregateWithSize()

Performs aggregation and returns the increase in required on-heap memory caused by this aggregation step.

The default implementation of this method calls aggregate() and returns 0.

If overridden, this method must include the JVM object overheads in the size estimation and must ensure not to underestimate required memory as that might lead to OOM errors.

Returns:

Increase in required on-heap memory caused by this aggregation step.

get

@Nullable
Object get()

getFloat

float getFloat()

getLong

long getLong()

getDouble

default double getDouble()

The default implementation casts getFloat() to double. This default method is added to enable smooth backward compatibility, please re-implement it if your aggregators work with numeric double columns.

isNull

default boolean isNull()

returns true if aggregator's output type is primitive long/double/float and aggregated value is null, but when aggregated output type is Object, this method always returns false, and users are advised to check nullability for the object returned by get() method. The default implementation always return false to enable smooth backward compatibility, re-implement if your aggregator is nullable.

close

void close()

Specified by:

close in interface AutoCloseable

Specified by:

close in interface Closeable