Really the parent of the history state will be entered or not dependending
on the LCA and the if the transition is internal or not, so I think the
ancestors should be calculated with the parent of the history only (not the


So in the example, when the transition r-->h occurs:
LCA= root (and yes, I ment scxml)
exit state r
enter state s
enter state h, so
--> enter state s22
--> properAncestors(s22,h.parent)= properAncestors(s22,s)= [s2]
-- -->enter s2

If the transition would be from a children of s to h with internal flag
then s will not be entered.

Do you agree with this?

Received on Wednesday, 14 November 2012 17:09:27 UTC