Hi Christophe,
my mistake, I forgot to add this limitation in the doc. The favor_compile_time_policy loses not only event hierarchies but also kleene events. I don't think there is a way to achieve this with the current implementation.
That's too bad. So I probably have to compile less frequently, to cope with the longer compile-times. :-)
Thanks for clarification, Deniz
Hi Deniz, it turns out my answer was incorrect, or at least incomplete. If you have no submachine, then there might be a way around. If you write: p.process_event(pause); where pause is an eUML event, true, it cannot work. However, this works: any_number_event evpause(pause); p.process_event(evpause); I don't know if it helps you, it might not fit your use case. It doesn't work for submachines at the moment, but as I'm at it, I'll have a look ;-) OTOH, if compile-time is a problem, mybe the full eUML is not the right solution for you, you might want to try writing only the table with eUML, it compiles already faster, or use the functor front-end, which compiles much faster. HTH, Christophe