Swimlanes problem in Process Execution

[egycator]

Hello All,

I'm creating a process definition with one Swimlane

When the process has just one main execution ,the Swimlane is working perfectly

But when adding any block that will create any child executions like "Timer or Fork-Join" the Swimlane for the child  executions is assigned to null.

How can I make these Swimlanes assigned to the same user of the Swimlane of  the Root Execution.

[rebody]

Hi Khaled,

  Could you provide a testcase to reproduce this scenario?  Thank you very much.

[egycator]

Hello HuiSheng,

My test case is attached.

Kindly notice that this problem appears when I assign the task in the runtime by using

taskService.AssignTask(taskId,userId);

But if I used the attribute of assignee in the swimlane tag , this problem doesn't occur

<swimlane name="user" assignee="alex" />

[rebody]

Hi Khaled,

This caused by the execution only find swimlaneImpl in its own scope.  If it can't find the related swimlane, it won't propragate to look for swimlane defined in parent scope.

I don't very clear for the swimlane behaviour between parent and children executions.  So I think I could create a patch and make some discussion about this.

By the way, please use zip to compress the attachment, not rar.

[mwohlf]

it is funny that the child execution/scope is only created when there are timer or variables used in the task,

the simpliest solution would be to lookup the parent scope for the swimlane (see attachment).

Another maybe more consistent approach would be to copy the swimlanes into the child scope like it

is done with variables and timers in ExecutionImpl.createScope(ScopeElementImpl scope)

[rebody]

Hi Michael,

Thank you for the patch and testcase.  It runs perfectly.  I created a issue in JIRA for this issue.

https://jira.jboss.org/jira/browse/JBPM-2860

[egycator]

Hello all,

Thanks for the quick fix , it worked perfectly.

[sebastian.s]

If this can be tested sufficient and this seems to be the case since Khaled is using this feature I'd love to see this scheduled for 4.4 as well. Any opinions on this? Maciej, maybe you would be so kind to have a look at Huisheng Xu's patch and commit it in case it's okay?

[rebody]

Hi Sebastian,

Thank you for noticing this.  I will move this issue to jBPM-4.4.