[#BATCH-90] StepExecution and StepExecutionContext are parallel domains, and StepExecution is by comparison anaemic

Spring Batch

StepExecution and StepExecutionContext are parallel domains, and StepExecution is by comparison anaemic

Created: 11/Aug/07 06:05 AM Updated: 07/Aug/08 10:06 AM

Component/s:

Core

Affects Version/s:

1.0-m2

Fix Version/s:

1.0-m3

Security Level:

Public (Public Issues)

Time Tracking:

Not Specified

Issue Links:

Depends

This issue depends on:

~~BATCH-124~~ Potential Bug in JobExecutionContext?

This issue is depended on by:

~~BATCH-124~~ Potential Bug in JobExecutionContext?

~~BATCH-102~~ Normalization of domain object dependencies

Description

« Hide

StepExecution and StepExecutionContext are parallel domains and it is unnecessary - all the features of the *Context could be put in the *Execution. StepExecution and JobExecution are classic anaemic domain objects.

All

Comments

Work Log

Change History

FishEye

Builds

Sort Order:

[ Permlink | « Hide ]

Lucas Ward added a comment - 12/Sep/07 03:24 PM

I agree that StepExecution and JobExecution are fairly anaemic, but aren't JobInstance, StepInstance, and their *Context variants just as anaemic? Also, wouldn't getting rid of this 'anaemicness' require a substantial reworking of how the executors interact with the repository? (not that it would be a bad thing)

[ Permlink | « Hide ]

Dave Syer added a comment - 16/Sep/07 04:34 PM

Substantial re-working might be what it takes. Actually I was also thinking that JobExecutor should return a JobExecution (which itself could contain the ExitStatus). Another interface change, but we should do it now, before people get used to the wrong way of doing it.

[ Permlink | « Hide ]

Dave Syer added a comment - 30/Sep/07 12:39 PM

* Removed setter for jobIdentifier in JobInstance.
* Did similar job for the StepInstance and StepExecution.
* Pushed transient functionality from *ExecutionContext to *Execution

Still more to do...

[ Permlink | « Hide ]

Dave Syer added a comment - 05/Oct/07 05:06 AM

Resolving this one and creating new issue to deal with *Executor* interfaces (~~BATCH-159~~)

[ Permlink | « Hide ]

Dave Syer added a comment - 07/Aug/08 10:06 AM

Assume closed as resolved and released