Package org.apache.ignite.internal

Class GridJobResultImpl

java.lang.Object

org.apache.ignite.internal.GridJobResultImpl

All Implemented Interfaces:

ComputeJobResult

public class GridJobResultImpl
extends Object
implements ComputeJobResult

Class provides implementation for job result.

Constructor Summary

Constructors

Constructor	Description
GridJobResultImpl(ComputeJob job, IgniteUuid jobId, ClusterNode node, GridJobSiblingImpl sib)

Method Summary

Modifier and Type	Method	Description
void	clearData()	Clears stored job data.
Object	getData()	Gets data returned by remote job if it didn't fail.
IgniteException	getException()	Gets exception produced by execution of remote job, or null if remote execution finished normally and did not produce any exceptions.
ComputeJob	getJob()	Gets local instance of remote job returned by ComputeTask.map(List, Object) method.
ComputeJobContext	getJobContext()	Gets job context.
ClusterNode	getNode()	Gets node this job executed on.
GridJobSiblingImpl	getSibling()
boolean	hasResponse()
boolean	isCancelled()	Gets job cancellation status.
boolean	isOccupied()
void	onResponse(@Nullable Object data, @Nullable IgniteException ex, @Nullable Map<Object,Object> jobAttrs, boolean isCancelled)
void	resetResponse()
void	setNode(ClusterNode node)
void	setOccupied(boolean isOccupied)
String	toString()

Methods inherited from class java.lang.Object

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

Constructor Detail

GridJobResultImpl

public GridJobResultImpl(ComputeJob job,
                         IgniteUuid jobId,
                         ClusterNode node,
                         GridJobSiblingImpl sib)

Parameters:

job - Job instance.

jobId - ID of the job.

node - Node from where this result was received.

sib - Sibling associated with this result.

Method Detail

getJob

public ComputeJob getJob()

Gets local instance of remote job returned by ComputeTask.map(List, Object) method.

Specified by:

getJob in interface ComputeJobResult

Returns:

Local instance of remote job returned by ComputeTask.map(List, Object) method.

getJobContext

public ComputeJobContext getJobContext()

Gets job context. Use job context to access job unique ID or to get/set jobs attributes. Context is attached to a job and travels with it wherever it goes. For example, if a job gets failed-over from one node to another, then its context will be failed over with it and all attributes that were set on the job on the first node will be available on the new node.

Specified by:

getJobContext in interface ComputeJobResult

Returns:

Job context.

getSibling

public GridJobSiblingImpl getSibling()

Returns:

Sibling associated with this result.

getNode

public ClusterNode getNode()

Gets node this job executed on.

Specified by:

getNode in interface ComputeJobResult

Returns:

Node this job executed on.

getData

public Object getData()

Gets data returned by remote job if it didn't fail. This data is the object returned from ComputeJob.execute() method.

Note that if task is annotated with ComputeTaskNoResultCache annotation, then job results will not be cached and will be available only in ComputeTask.result(ComputeJobResult, List) method for every individual job, but not in ComputeTask.reduce(List) method. This feature was added to avoid excessive storing of overly large results.

Specified by:

getData in interface ComputeJobResult

Returns:

Data returned by remote job's ComputeJob.execute() method if it didn't fail.

getException

public IgniteException getException()

Gets exception produced by execution of remote job, or null if remote execution finished normally and did not produce any exceptions.

Specified by:

getException in interface ComputeJobResult

Returns:

IgniteException produced by execution of remote job or null if no exception was produced.

Note that if remote job resulted in RuntimeException or Error then they will be wrapped into ComputeUserUndeclaredException returned by this method.

If job on remote node was rejected (cancelled while it was on waiting queue), then ComputeExecutionRejectedException will be returned.

If node on which job was computing failed, then ClusterTopologyException is returned.

isCancelled

public boolean isCancelled()

Gets job cancellation status. Returns true if job received cancellation request on remote node. Note that job, after receiving cancellation request, will still need to finish and return, hence ComputeJobResult.getData() method may contain execution result even if the job was canceled.

Job can receive cancellation request if the task was explicitly cancelled from future (see IgniteFuture.cancel()) or if task completed prior to getting results from all remote jobs.

Specified by:

isCancelled in interface ComputeJobResult

Returns:

true if job received cancellation request and false otherwise.

setNode

public void setNode(ClusterNode node)

Parameters:

node - Node from where this result was received.

onResponse

public void onResponse(@Nullable
                       @Nullable Object data,
                       @Nullable
                       @Nullable IgniteException ex,
                       @Nullable
                       @Nullable Map<Object,Object> jobAttrs,
                       boolean isCancelled)

Parameters:

data - Job data.

ex - Job exception.

jobAttrs - Job attributes.

isCancelled - Whether job was cancelled or not.

setOccupied

public void setOccupied(boolean isOccupied)

Parameters:

isOccupied - True if job for this response is being sent.

isOccupied

public boolean isOccupied()

Returns:

True if job for this response is being sent.

clearData

public void clearData()

Clears stored job data.

resetResponse

public void resetResponse()

hasResponse

public boolean hasResponse()

Returns:

true if remote job responded.

toString

public String toString()

Overrides:

toString in class Object