jmdacruz/python-statemachine

Python Finite State Machines made easy.

Python State Machine

Python finite-state machines made easy.

• Free software: MIT license
• Documentation: https://python-statemachine.readthedocs.io.

Getting started

To install Python State Machine, run this command in your terminal:

$ pip install python-statemachine

Define your state machine:

from statemachine import StateMachine, State

class TrafficLightMachine(StateMachine):
    green = State('Green', initial=True)
    yellow = State('Yellow')
    red = State('Red')

    slowdown = green.to(yellow)
    stop = yellow.to(red)
    go = red.to(green)

You can now create an instance:

>>> traffic_light = TrafficLightMachine()

And inspect about the current state:

>>> traffic_light.current_state
State('Green', identifier='green', value='green', initial=True)
>>> traffic_light.current_state == TrafficLightMachine.green == traffic_light.green
True

For each state, there's a dynamically created property in the form is_<state.identifier>, that returns True if the current status matches the query:

>>> traffic_light.is_green
True
>>> traffic_light.is_yellow
False
>>> traffic_light.is_red
False

Query about metadata:

>>> [s.identifier for s in m.states]
['green', 'red', 'yellow']
>>> [t.identifier for t in m.transitions]
['go', 'slowdown', 'stop']

Call a transition:

>>> traffic_light.slowdown()

And check for the current status:

>>> traffic_light.current_state
State('Yellow', identifier='yellow', value='yellow', initial=False)
>>> traffic_light.is_yellow
True

You can't run a transition from an invalid state:

>>> traffic_light.is_yellow
True
>>> traffic_light.slowdown()
Traceback (most recent call last):
...
TransitionNotAllowed: Can't slowdown when in Yellow.

You can also trigger events in an alternative way, calling the run(<transition.identificer>) method:

>>> traffic_light.is_yellow
True
>>> traffic_light.run('stop')
>>> traffic_light.is_red
True

A state machine can be instantiated with an initial value:

>>> machine = TrafficLightMachine(start_value='red')
>>> traffic_light.is_red
True

Models

If you need to persist the current state on another object, or you're using the state machine to control the flow of another object, you can pass this object to the StateMachine constructor:

>>> class MyModel(object):
...     def __init__(self, state):
...         self.state = state
...
>>> obj = MyModel(state='red')
>>> traffic_light = TrafficLightMachine(obj)
>>> traffic_light.is_red
True
>>> obj.state
'red'
>>> obj.state = 'green'
>>> traffic_light.is_green
True
>>> traffic_light.slowdown()
>>> obj.state
'yellow'
>>> traffic_light.is_yellow
True

Callbacks

Callbacks when running events:

from statemachine import StateMachine, State

class TrafficLightMachine(StateMachine):
    "A traffic light machine"
    green = State('Green', initial=True)
    yellow = State('Yellow')
    red = State('Red')

    slowdown = green.to(yellow)
    stop = yellow.to(red)
    go = red.to(green)

    def on_slowdown(self):
        print('Calma, lá!')

    def on_stop(self):
        print('Parou.')

    def on_go(self):
        print('Valendo!')

>>> stm = TrafficLightMachine()
>>> stm.slowdown()
Calma, lá!
>>> stm.stop()
Parou.
>>> stm.go()
Valendo!

Or when entering/exiting states:

from statemachine import StateMachine, State

class TrafficLightMachine(StateMachine):
    "A traffic light machine"
    green = State('Green', initial=True)
    yellow = State('Yellow')
    red = State('Red')

    cycle = green.to(yellow) | yellow.to(red) | red.to(green)

    def on_enter_green(self):
        print('Valendo!')

    def on_enter_yellow(self):
        print('Calma, lá!')

    def on_enter_red(self):
        print('Parou.')

>>> stm = TrafficLightMachine()
>>> stm.cycle()
Calma, lá!
>>> stm.cycle()
Parou.
>>> stm.cycle()
Valendo!

Mixins

Your model can inherited from a custom mixin to auto-instantiate a state machine.

class CampaignMachineWithKeys(StateMachine):
    "A workflow machine"
    draft = State('Draft', initial=True, value=1)
    producing = State('Being produced', value=2)
    closed = State('Closed', value=3)
    cancelled = State('Cancelled', value=4)

    add_job = draft.to.itself() | producing.to.itself()
    produce = draft.to(producing)
    deliver = producing.to(closed)
    cancel = cancelled.from_(draft, producing)


class MyModel(MachineMixin):
    state_machine_name = 'CampaignMachineWithKeys'

    def __init__(self, **kwargs):
        for k, v in kwargs.items():
            setattr(self, k, v)
        super(MyModel, self).__init__()

    def __repr__(self):
        return "{}({!r})".format(type(self).__name__, self.__dict__)


model = MyModel(state='draft')
assert isinstance(model.statemachine, campaign_machine)
assert model.state == 'draft'
assert model.statemachine.current_state == model.statemachine.draft

model.statemachine.cancel()
assert model.state == 'cancelled'