io.cdap.cdap.etl.api.batch

Class BatchAggregator<GROUP_KEY,GROUP_VALUE,OUT>

java.lang.Object

io.cdap.cdap.etl.api.batch.BatchConfigurable<BatchAggregatorContext>

io.cdap.cdap.etl.api.batch.BatchAggregator<GROUP_KEY,GROUP_VALUE,OUT>

Type Parameters:

GROUP_KEY - group key type. Must be a supported type

GROUP_VALUE - group value type. Must be a supported type

OUT - output object type

All Implemented Interfaces:

Aggregator<GROUP_KEY,GROUP_VALUE,OUT>, Destroyable, PipelineConfigurable, StageLifecycle<BatchRuntimeContext>, SubmitterLifecycle<BatchAggregatorContext>

@Beta
public abstract class BatchAggregator<GROUP_KEY,GROUP_VALUE,OUT>
extends BatchConfigurable<BatchAggregatorContext>
implements Aggregator<GROUP_KEY,GROUP_VALUE,OUT>, PipelineConfigurable, StageLifecycle<BatchRuntimeContext>

An Aggregator used in batch programs. As it is used in batch programs, a BatchAggregator must be parameterized with supported group key and value classes. Group keys and values can be a byte[], Boolean, Integer, Long, Float, Double, String, or StructuredRecord. If the group key is not one of those types and is being used in mapreduce, it must implement Hadoop's org.apache.hadoop.io.WritableComparable interface. If the group value is not one of those types and is being used in mapreduce, it must implement Hadoop's org.apache.hadoop.io.Writable interface. If the aggregator is being used in spark, both the group key and value must implement the Serializable interface.

Field Summary

Fields

Modifier and Type	Field and Description
static String	PLUGIN_TYPE

Constructor Summary

Constructors

Constructor and Description
BatchAggregator()

Method Summary

Modifier and Type	Method and Description
void	configurePipeline(PipelineConfigurer pipelineConfigurer) Configure the pipeline.
void	destroy() Destroy the Batch Aggregator.
void	initialize(BatchRuntimeContext context) Initialize the Batch Aggregator.
void	prepareRun(BatchAggregatorContext context) Prepare a pipeline run.

Methods inherited from class io.cdap.cdap.etl.api.batch.BatchConfigurable

onRunFinish

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface io.cdap.cdap.etl.api.Aggregator

aggregate, groupBy

Field Detail

PLUGIN_TYPE

public static final String PLUGIN_TYPE

See Also:

Constant Field Values

Constructor Detail

BatchAggregator

public BatchAggregator()

Method Detail

configurePipeline

public void configurePipeline(PipelineConfigurer pipelineConfigurer)

Configure the pipeline. This is run once when the pipeline is being published. This is where you perform any static logic, like creating required datasets, performing schema validation, setting output schema, and things of that nature.

Specified by:

configurePipeline in interface PipelineConfigurable

Overrides:

configurePipeline in class BatchConfigurable<BatchAggregatorContext>

Parameters:

pipelineConfigurer - the configurer used to add required datasets and streams

Throws:

ValidationException - if the given config is invalid

prepareRun

public void prepareRun(BatchAggregatorContext context)
                throws Exception

Prepare a pipeline run. This is run every time before a pipeline runs in order to help set up the run. This is where you would set things like the number of partitions to use when grouping, and setting the group key and value classes if they are not known at compile time.

Specified by:

prepareRun in interface SubmitterLifecycle<BatchAggregatorContext>

Specified by:

prepareRun in class BatchConfigurable<BatchAggregatorContext>

Parameters:

context - batch execution context

Throws:

Exception

initialize

public void initialize(BatchRuntimeContext context)
                throws Exception

Initialize the Batch Aggregator. Executed inside the Batch Run. This method is guaranteed to be invoked before any calls to Aggregator.groupBy(Object, Emitter) and Aggregator.aggregate(Object, Iterator, Emitter) are made.

Specified by:

initialize in interface StageLifecycle<BatchRuntimeContext>

Parameters:

context - BatchRuntimeContext

Throws:

Exception - if there is any error during initialization

destroy

public void destroy()

Destroy the Batch Aggregator. Executed at the end of the Batch Run.

Specified by:

destroy in interface Destroyable