com.google.cloud.dataflow.sdk.runners

Class PipelineRunner<ResultT extends PipelineResult>

java.lang.Object

com.google.cloud.dataflow.sdk.runners.PipelineRunner<ResultT>

Type Parameters:

ResultT - the type of the result of run(com.google.cloud.dataflow.sdk.Pipeline).

Direct Known Subclasses:

BlockingDataflowPipelineRunner, DataflowPipelineRunner, DirectPipelineRunner, TestDataflowPipelineRunner

public abstract class PipelineRunner<ResultT extends PipelineResult>
extends Object

A PipelineRunner can execute, translate, or otherwise process a Pipeline.

Constructor Summary

Constructors

Constructor and Description
PipelineRunner()

Method Summary

Modifier and Type	Method and Description
<OutputT extends POutput,InputT extends PInput> OutputT	apply(PTransform<InputT,OutputT> transform, InputT input) Applies a transform to the given input, returning the output.
static PipelineRunner<? extends PipelineResult>	fromOptions(PipelineOptions options) Constructs a runner from the provided options.
abstract ResultT	run(Pipeline pipeline) Processes the given Pipeline, returning the results.

Methods inherited from class java.lang.Object

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

Constructor Detail

PipelineRunner

public PipelineRunner()

Method Detail

fromOptions

public static PipelineRunner<? extends PipelineResult> fromOptions(PipelineOptions options)

Constructs a runner from the provided options.

Returns:

The newly created runner.

run

public abstract ResultT run(Pipeline pipeline)

Processes the given Pipeline, returning the results.

apply

public <OutputT extends POutput,InputT extends PInput> OutputT apply(PTransform<InputT,OutputT> transform,
                                                                     InputT input)

Applies a transform to the given input, returning the output.

The default implementation calls PTransform.apply(input), but can be overridden to customize behavior for a particular runner.