org.apache.crunch.impl.mr

Class MRPipeline

java.lang.Object

org.apache.crunch.impl.mr.MRPipeline

All Implemented Interfaces:

Pipeline

public class MRPipeline
extends Object
implements Pipeline

Pipeline implementation that is executed within Hadoop MapReduce.

Constructor Summary

Constructors

Constructor and Description
MRPipeline(Class<?> jarClass) Instantiate with a default Configuration and name.
MRPipeline(Class<?> jarClass, org.apache.hadoop.conf.Configuration conf) Instantiate with a custom configuration and default naming.
MRPipeline(Class<?> jarClass, String name) Instantiate with a custom pipeline name.
MRPipeline(Class<?> jarClass, String name, org.apache.hadoop.conf.Configuration conf) Instantiate with a custom name and configuration.

Method Summary

Methods

Modifier and Type	Method and Description
<T> SourceTarget<T>	createIntermediateOutput(PType<T> ptype)
org.apache.hadoop.fs.Path	createTempPath()
PipelineResult	done() Run any remaining jobs required to generate outputs and then clean up any intermediate data files that were created in this run or previous calls to run.
void	enableDebug() Turn on debug logging for jobs that are run from this pipeline.
org.apache.hadoop.conf.Configuration	getConfiguration() Returns the Configuration instance associated with this pipeline.
<T> ReadableSource<T>	getMaterializeSourceTarget(PCollection<T> pcollection) Retrieve a ReadableSourceTarget that provides access to the contents of a PCollection.
String	getName() Returns the name of this pipeline.
int	getNextAnonymousStageId()
<T> Iterable<T>	materialize(PCollection<T> pcollection) Create the given PCollection and read the data it contains into the returned Collection instance for client use.
org.apache.crunch.impl.mr.exec.MRExecutor	plan()
<S> PCollection<S>	read(Source<S> source) Converts the given Source into a PCollection that is available to jobs run using this Pipeline instance.
<K,V> PTable<K,V>	read(TableSource<K,V> source) A version of the read method for TableSource instances that map to PTables.
PCollection<String>	readTextFile(String pathName) A convenience method for reading a text file.
PipelineResult	run() Constructs and executes a series of MapReduce jobs in order to write data to the output targets.
void	setConfiguration(org.apache.hadoop.conf.Configuration conf) Set the Configuration to use with this pipeline.
void	write(PCollection<?> pcollection, Target target) Write the given collection to the given target on the next pipeline run.
void	write(PCollection<?> pcollection, Target target, Target.WriteMode writeMode) Write the contents of the PCollection to the given Target, using the storage format specified by the target and the given WriteMode for cases where the referenced Target already exists.
<T> void	writeTextFile(PCollection<T> pcollection, String pathName) A convenience method for writing a text file.

Methods inherited from class java.lang.Object

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

Constructor Detail

MRPipeline

public MRPipeline(Class<?> jarClass)

Instantiate with a default Configuration and name.

Parameters:

jarClass - Class containing the main driver method for running the pipeline

MRPipeline

public MRPipeline(Class<?> jarClass,
          String name)

Instantiate with a custom pipeline name. The name will be displayed in the Hadoop JobTracker.

Parameters:

jarClass - Class containing the main driver method for running the pipeline

name - Display name of the pipeline

MRPipeline

public MRPipeline(Class<?> jarClass,
          org.apache.hadoop.conf.Configuration conf)

Instantiate with a custom configuration and default naming.

Parameters:

jarClass - Class containing the main driver method for running the pipeline

conf - Configuration to be used within all MapReduce jobs run in the pipeline

MRPipeline

public MRPipeline(Class<?> jarClass,
          String name,
          org.apache.hadoop.conf.Configuration conf)

Instantiate with a custom name and configuration. The name will be displayed in the Hadoop JobTracker.

Parameters:

jarClass - Class containing the main driver method for running the pipeline

name - Display name of the pipeline

conf - Configuration to be used within all MapReduce jobs run in the pipeline

Method Detail

getConfiguration

public org.apache.hadoop.conf.Configuration getConfiguration()

Description copied from interface: Pipeline

Returns the Configuration instance associated with this pipeline.

Specified by:

getConfiguration in interface Pipeline

setConfiguration

public void setConfiguration(org.apache.hadoop.conf.Configuration conf)

Description copied from interface: Pipeline

Set the Configuration to use with this pipeline.

Specified by:

setConfiguration in interface Pipeline

plan

public org.apache.crunch.impl.mr.exec.MRExecutor plan()

run

public PipelineResult run()

Description copied from interface: Pipeline

Constructs and executes a series of MapReduce jobs in order to write data to the output targets.

Specified by:

run in interface Pipeline

done

public PipelineResult done()

Description copied from interface: Pipeline

Run any remaining jobs required to generate outputs and then clean up any intermediate data files that were created in this run or previous calls to run.

Specified by:

done in interface Pipeline

read

public <S> PCollection<S> read(Source<S> source)

Description copied from interface: Pipeline

Converts the given Source into a PCollection that is available to jobs run using this Pipeline instance.

Specified by:

read in interface Pipeline

Parameters:

source - The source of data

Returns:

A PCollection that references the given source

read

public <K,V> PTable<K,V> read(TableSource<K,V> source)

Description copied from interface: Pipeline

A version of the read method for TableSource instances that map to PTables.

Specified by:

read in interface Pipeline

Parameters:

source - The source of the data

Returns:

A PTable that references the given source

readTextFile

public PCollection<String> readTextFile(String pathName)

Description copied from interface: Pipeline

A convenience method for reading a text file.

Specified by:

readTextFile in interface Pipeline

write

public void write(PCollection<?> pcollection,
         Target target)

Description copied from interface: Pipeline

Write the given collection to the given target on the next pipeline run. The system will check to see if the target's location already exists using the WriteMode.DEFAULT rule for the given Target.

Specified by:

write in interface Pipeline

Parameters:

pcollection - The collection

target - The output target

write

public void write(PCollection<?> pcollection,
         Target target,
         Target.WriteMode writeMode)

Description copied from interface: Pipeline

Write the contents of the PCollection to the given Target, using the storage format specified by the target and the given WriteMode for cases where the referenced Target already exists.

Specified by:

write in interface Pipeline

Parameters:

pcollection - The collection

target - The target to write to

writeMode - The strategy to use for handling existing outputs

materialize

public <T> Iterable<T> materialize(PCollection<T> pcollection)

Description copied from interface: Pipeline

Create the given PCollection and read the data it contains into the returned Collection instance for client use.

Specified by:

materialize in interface Pipeline

Parameters:

pcollection - The PCollection to materialize

Returns:

the data from the PCollection as a read-only Collection

getMaterializeSourceTarget

public <T> ReadableSource<T> getMaterializeSourceTarget(PCollection<T> pcollection)

Retrieve a ReadableSourceTarget that provides access to the contents of a PCollection. This is primarily intended as a helper method to materialize(PCollection). The underlying data of the ReadableSourceTarget may not be actually present until the pipeline is run.

Parameters:

pcollection - The collection for which the ReadableSourceTarget is to be retrieved

Returns:

The ReadableSourceTarget

Throws:

IllegalArgumentException - If no ReadableSourceTarget can be retrieved for the given PCollection

createIntermediateOutput

public <T> SourceTarget<T> createIntermediateOutput(PType<T> ptype)

createTempPath

public org.apache.hadoop.fs.Path createTempPath()

writeTextFile

public <T> void writeTextFile(PCollection<T> pcollection,
                     String pathName)

Description copied from interface: Pipeline

A convenience method for writing a text file.

Specified by:

writeTextFile in interface Pipeline

getNextAnonymousStageId

public int getNextAnonymousStageId()

enableDebug

public void enableDebug()

Description copied from interface: Pipeline

Turn on debug logging for jobs that are run from this pipeline.

Specified by:

enableDebug in interface Pipeline

getName

public String getName()

Description copied from interface: Pipeline

Returns the name of this pipeline.

Specified by:

getName in interface Pipeline

Returns:

Name of the pipeline