[jboss-user] [JBoss jBPM] - Re: Fork + Join

[nlmarco]

Hello Tom,

thanks a lot for your quick reply! I built a test case and hope you can either tell me what I'm doing wrong or fix the bug in jBPM ;-)

This is my test process definition: http://www.nightlabs.de/~marco/jbpm/2006-12-18.00/processimage.jpg

Here is a little test application (incl. the process definition): http://www.nightlabs.de/~marco/jbpm/2006-12-18.00/jbpmfork.tar.gz

And here are screen shots (of the above test application) showing two wrong/strange behaviours:

http://www.nightlabs.de/~marco/jbpm/2006-12-18.00/forktest-a.html

As you see in A3, I was able to move the root token from the Fork node "fork1" to the State "state_left_1", even though the child tokens didn't reach the Join node yet. This is not a serious problem, though, because I could find out that the root token has active child tokens and prevent the user from doing the transition "to_state_left_1". However, is this behaviour really intended? Shouldn't jBPM prevent signalling the root token while there are active children?

http://www.nightlabs.de/~marco/jbpm/2006-12-18.00/forktest-b.html

This is what I've described in my previous posting. As you see in B3 and B4, I move the two child tokens to the Join node "join1". Correctly, the root token jumps from "fork1" to "state_after_join" as soon as the second child token reached "join1". B5 demonstrates, however, that I can continue moving the second child token after it has reached "join1" - it does not end in this Join node.

Btw, I'm using jBPM version 3.1.2 with MySQL 5.0.18.

Best regards, Marco :-)

View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3994796#3994796

Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3994796