On Mar 21, 10:03 am, Tim Chase <python.l...@tim.thechases.com> wrote:
> >>> transitions = {
> >>> # values are tuples of (newstate, transition_function)
> >>> STATE_A: [
> >>> (STATE_B, lambda x: x > 5),
> >>> (STATE_C, lambda x: x > 10),
> >>> (STATE_D, lambda x: x > 100),
> >>> ],
> >>> STATE_B: [
> >>> (STATE_A, lambda x: x < 5),
> >>> (STATE_C, lambda x: x > 10),
> >>> ],
> >>> STATE_C: [
> >>> (STATE_B, lambda x: x < 10),
> >>> (STATE_D, lambda x: x > 100),
> >>> ],
> >>> STATE_D: [],
> >>> }
>
> >> And if you don't mind me asking. How do you invoke lambda from
> >> transitions?
>
> > Disregard that. I think I figured it out.
>
> > If you had something like...
>
> >>>> transitions = {1: [2, lambda x: 2*x]}
>
> > You would probably call it like...
>
> >>>> transitions[1][1](4)
> > 8
>
> I tend to use them with tuple assignment which I find reads more
> cleanly than directly indexing:
>
> for input in source():
> available_states = []
> for new_state, function in transitions[state]:
> if function(input):
> available_states.append(new_state)
> do_something(available_states)
>
To my defense, I would like to say that I started with python like 2
weeks ago.
--
http://mail.python.org/mailman/listinfo/python-list
