cancel
Showing results for 
Search instead for 
Did you mean: 

Lane outside a Pool - Activity Designer vs JBPM Designer difference

lucabur
Champ in-the-making
Champ in-the-making
Hi,
I have some .bpmn files created with the JBPM 5 Eclipse Designer, and any attempt to import / open them in Activity Designer fails with the following message:

java.lang.IllegalArgumentException: findParentPool was called on a lane that has no pool as a parent
   at org.activiti.designer.eclipse.navigator.diagram.LaneDiagramTreeNode.findParentProcess(LaneDiagramTreeNode.java:71)

As the error clearly states, the problem is that the diagram contains some lanes without an enclosing pool. Trying to build such a diagram with Activity Designer is not possible, because a lane can not be created outside a pool.
The JBPM 5 Designer however allows this construct and this is the cause of the problem.
I have three questions:
1 who is right with respect of the BPMN 2.0 rules ?
2 if Activity Designer is correctly implementing the BPMN 2.0 rules, a more gracefully error message when opening such a wrong diagram is imho required.
3 otherwise, if the right behaviour is thr JBPM 5 Designer behaviour, the Activiti Designer has a bug.

Before filing a bug issue (which one ?) I would like to have the advice of anyone more expert than me on the matter.

Thank you.
2 REPLIES 2

jbarrez
Star Contributor
Star Contributor
Of course, I'm biased 😉

1) My understanding is that a lane cannot exist without a pool (even if there is only one). It's a while I read that section of the spec though, but it sounds right to me.

2) Correct.

3) See 1 🙂

You can definitely file a bug, already to get a better exception stack trace. And reference this forum posts, so we can look it up in the spec when we get to the issue.

ngigiwaithaka
Champ in-the-making
Champ in-the-making
I am thinking, the solution or bug to be filled could be in helping Activiti Designer import jBPM5 designed .bpmn files without regard as to whether jBPM interpreted the spec correctly.

It would be nice for the two leading open source BPMs to inter-operate to avoid some form of vendor lock-in as is happening in this case.

P.S
Whats the critical reason to leaving jBPM 5?

Regards
Waithaka Ngigi
A1 Workpoint