XWorkflows is a library to add workflows, or state machines, to Python objects.
It has been fully tested with Python 2.7 and all versions from 3.4 to 3.9
- Package on PyPI: http://pypi.python.org/pypi/xworkflows
- Repository and issues on GitHub: http://github.com/rbarrois/xworkflows
- Doc on http://readthedocs.org/docs/xworkflows/
It allows to easilly define a workflow, attach it to a class, and use its transitions:
import xworkflows
class MyWorkflow(xworkflows.Workflow):
# A list of state names
states = (
('foo', "Foo"),
('bar', "Bar"),
('baz', "Baz"),
)
# A list of transition definitions; items are (name, source states, target).
transitions = (
('foobar', 'foo', 'bar'),
('gobaz', ('foo', 'bar'), 'baz'),
('bazbar', 'baz', 'bar'),
)
initial_state = 'foo'
class MyObject(xworkflows.WorkflowEnabled):
state = MyWorkflow()
@xworkflows.transition()
def foobar(self):
return 42
# It is possible to use another method for a given transition.
@xworkflows.transition('gobaz')
def blah(self):
return 13
>>> o = MyObject()
>>> o.state
<StateWrapper: <State: 'foo'>>
>>> o.state.is_foo
True
>>> o.state.name
'foo'
>>> o.state.title
'Foo'
>>> o.foobar()
42
>>> o.state
<StateWrapper: <State: 'bar'>>
>>> o.state.name
'bar'
>>> o.state.title
'Bar'
>>> o.blah()
13
>>> o.state
<StateWrapper: <State: 'baz'>>
>>> o.state.name
'baz'
>>> o.state.title
'Baz'
Custom functions can be hooked to transactions, in order to run before/after a transition, when entering a state, when leaving a state, ...:
class MyObject(xworkflows.WorkflowEnabled):
state = MyWorkflow()
@xworkflows.before_transition('foobar')
def my_hook(self, *args, **kwargs):
# *args and **kwargs are those passed to MyObject.foobar(...)
pass
@xworkflows.on_enter_state('bar')
def my_other_hook(self, result, *args, **kwargs):
# Will be called just after any transition entering 'bar'
# result is the value returned by that transition
# *args, **kwargs are the arguments/keyword arguments passed to the
# transition.
pass