I have seen the other posts on the forum where the use of transient variables are used to accomplish this task, but does the documentation allude to something new that may be built in to the engine?
where in the docs is this? (I'm lazy ;-) )
Ah... ok, after reading the docs, I understand what was meant (it's kind of hypothetical)
JPDL ,afaik, only allows one leaving transition from a sub-process node. What you can do is have multiple end-states in the sup-process and set a variable to a value to indicate which end-state it is in. Copy that value back to the super process and use a decision node after the sub-process node to take a specific transition
Maybe the docs aren't very clear here, but the quote from the documentation seems to imply two possible ways of handling this. Your explanation takes care of the first way.
This selection can be based on either the sub process execution's variables..
The or clause continues:
or the end state of the sub process (a typical state machine can have multiple end states).
If the only possibility was to use execution variables, it seems they would of left this last part of the sentence off. Of course, I suppose we will have to check out the code if the authors don't weigh in first.