RE: [misterhouse-users] opinions on change of state_now

[Bruce W. <br...@mi...>]

> >Do you have an example of where it would be nice to do state_now the ECS
> >way?
>
> for example, i have several events that kick off when i wake up
> in the morning (testing 'scott is awake' with a state_now). i
> also have MH set 'scott is awake' to true at 10am - just in
> case i hadn't let MH know... i don't want the set of events to
> occur again if i've already told MH i'm awake. i realize that i
> could have this event at 10am check to make sure i'm not
> already awake, but i'm just suggesting that as an example. it
> becomes more useful when i've got tie_events associated with a
> serial object that represents an X10 keypress. rather
> than doing the test to see what state the object is at every time
> i set a state, i can just set it and the events kicked off by
> monitoring state_now won't get duplicated.
>
> i'm also open to the possibility that i've got ECS myopia...

As David pointed out, it seems a modified state_now would dis-allow
detecting sequential 'Volume up' type commands.

Maybe we could add a state_changed method, which would be like state_now,
but would return data only if the state really changed?

Bruce