Skip to main content

Redux implementation for Python

Project description

🚀 Python Redux

🌟 Overview

Python Redux is a Redux implementation for Python, bringing Redux's state management architecture to Python applications.

⚙️ Features

  • Redux API for Python developers.

  • Reduce boilerplate by dropping type property, payload classes and action creators:

    • Each action is a subclass of BaseAction.
    • Its type is checked by utilizing isinstance (no need for type property).
    • Its payload are its direct properties (no need for payload property).
    • Its creator is its auto-generated constructor.

    ➡️ Sample usage

  • Use type annotations for all its API.

  • Immutable state management for predictable state updates using python-immutable.

  • Offers a streamlined, native API for handling action side-effects asynchronously, eliminating the necessity for more intricate utilities such as redux-thunk or redux-saga.

  • Incorporates the autorun decorator, inspired by the mobx framework, to better integrate with elements of the software following procedural patterns.

  • Supports middlewares.

📦 Installation

The package handle in PyPI is python-redux

Pip

pip install python-redux

Poetry

poetry add python-redux

🛠 Usage

Side Effects - store.subscribe_event

Reducers can return either a simple state (like normal redux) or a combination of a state, bunch of actions (which will be dispatched immediately) and a bunch of events. These events will be consumed by event listeners registered via subscribe_event asynchronously in separate worker threads.

See todo sample below or check the demo to see it in action.

Autorun Decorator - store.autorun

Inspired by MobX's autorun and reaction, python-redux introduces the autorun decorator. This decorator requires a selector function as an argument. The selector is a function that accepts the store instance and returns a derived object from the store's state. The primary function of autorun is to establish a subscription to the store. Whenever the store is changed, autorun executes the selector with the updated store. Importantly, the decorated function is triggered only if there is a change in the selector's return value. This mechanism ensures that the decorated function runs in response to relevant state changes, enhancing efficiency and responsiveness in the application.

See todo sample below or check the demo to see it in action.

Combining reducers - combine_reducers

You can compose high level reducers by combining smaller reducers using combine_reducers utility function. This works mostly the same as the JS redux library version except that it provides a mechanism to dynamically add/remove reducers to/from it. This is done by generating an id and returning it along the generated reducer. This id is used to refer to this reducer in the future. Let's assume you composed a reducer like this:

reducer, reducer_id = combine_reducers(
    state_type=StateType,
    first=straight_reducer,
    second=second_reducer,
)

You can then add a new reducer to it using the reducer_id like this:

store.dispatch(
    CombineReducerRegisterAction(
        _id=reducer_id,
        key='third',
        third=third_reducer,
    ),
)

You can also remove a reducer from it like this:

store.dispatch(
    CombineReducerRegisterAction(
        _id=reducer_id,
        key='second',
    ),
)

Without this id, all the combined reducers in the store tree would register third reducer and unregister second reducer, but thanks to this reducer_id, these actions will only target the desired combined reducer.

🔎 Sample Usage

Minimal todo application store implemented using python-redux:

# state:
class ToDoItem(Immutable):
    id: str
    content: str
    timestamp: float


class ToDoState(Immutable):
    items: Sequence[ToDoItem]


# actions:
class AddTodoItemAction(BaseAction):
    content: str
    timestamp: float


class RemoveTodoItemAction(BaseAction):
    id: str


# events:
# events are non-pure side effects, like fetching information via an api, or anything
class CallApi(BaseEvent):
    parameters: object


# reducer:
def reducer(
    state: ToDoState | None,
    action: BaseAction,
) -> ReducerResult[ToDoState, BaseAction, BaseEvent]:
    if state is None:
        return ToDoState(
            items=[
                ToDoItem(
                    id=uuid.uuid4().hex,
                    content='Initial Item',
                    timestamp=time.time(),
                ),
            ],
        )
    if isinstance(action, AddTodoItemAction):
        return replace(
            state,
            items=[
                *state.items,
                ToDoItem(
                    id=uuid.uuid4().hex,
                    content=action.content,
                    timestamp=action.timestamp,
                ),
            ],
        )
    if isinstance(action, RemoveTodoItemAction):
        return CompleteReducerResult(
            state=replace(
                state,
                actions=[item for item in state.items if item.id != action.id],
            ),
            events=[CallApi(parameters={})],
        )
    return state


store = create_store(reducer)


# subscription:
dummy_render = print
store.subscribe(dummy_render)


# autorun:
@store.autorun(
  lambda state: state.items[0].content if len(state.items) > 0 else None,
)
def reaction(content: str | None) -> None:
    print(content)


# event listener, note that this will run async in a separate thread, so it can include
# io operations like network calls, etc:
dummy_api_call = print
store.subscribe_event(CallApi, lambda event: dummy_api_call(event.parameters))

# dispatch:
store.dispatch(AddTodoItemAction(content='New Item', timestamp=time.time()))

store.dispatch(FinishAction())

🎉 Demo

For a detailed example, see demo.py in the GitHub repository.

🤝 Contributing

Contributions following Python best practices are welcome.

🔒 License

Refer to the repository for license details.

Project details


Release history Release notifications | RSS feed

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

python_redux-0.9.6.tar.gz (13.2 kB view hashes)

Uploaded Source

Built Distribution

python_redux-0.9.6-py3-none-any.whl (12.6 kB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page