peach-reactives 0.0.3

No project description provided

Reactives

Library for easily launching threads that "react" to messages on a queue.

Reactive

At its core, Reactives are independent threads with their own queues that "react" to messages.

Creating a Reactive is as simple as inheriting the class and overriding the on_run(self, msg) function.

from reactive import Reactive

class My_Reactive(Reactive):

    def on_run(self, msg):
        ...

There are 5 functions that the user can override, only the on_run is required. Each function corresponds to a stage in the Lifecycle of a Reactive:

1. on_init(self)
2. on_register(self)
3. on_start(self)
4. on_run(self, msg)
5. on_stop(self)

It is important not to override the __init__(self) magic function.

When a Reactive is instantiated it can be given a unique name, if a name is not provided it will default to the name of the class. No two Reactives can have the same name, this will result in a exception raised during registration with the Messenger. The name is used by the Messenger to know where to send messages.

my_reactive_1 = My_Reactive()           <-- Defaults name to 'my_reactive'
my_reactive_2 = My_Reactive(name="r_2") <-- name is 'r_2'
my_reactive_3 = My_Reactive()           <-- Raises an `AlreadyRegistered' exception

Lifecycle

Each Reactive has a Lifecycle that it follows. Some Lifecycle states are transitioned automatically, others are requested. At each Lifecycle state the user has an opportunity to execute some function.

Init

The Init state is entered when the Reactive is instantiated. This state creates the Reactive, its queue, and any other necesseties. The user defined on_init(self) function is executed at the end of this state. The Init state automatically transitions to the Register state if successfully completed.

Register

The Register state is automatically entered after the Init state is complete. This state registers the Reactive with the Messenger and subscribes to any subjects.
The user defined on_register(self) function is executed at the end of this state. The Register state is held here, waiting further instruction before transitioning to the Start state.

Start

The Start state is entered after user intervention, either manually executing start() or, more preferably, requesting it through the Messenger. This state will execute the user defined on_start(self) function and start a new thread. The Start state will automatically transition to the Run state if successfully completed.

Run

The Run state is automatically entered after the Start state is complete. This state is where the actual "work" happens. The thread will enter a loop listening for messages and running the user defined on_run(self, msg) function, passing along any messages in the queue. The Run state is held in an infinite loop, reacting to messages or an instruction to transition to the Stop state.

Stop

The Stop state is entered after user intervention, either manually executing stop() or, more preferable, requesting it through the Messenger. This state will wait for the thread to complete, clear the queue, and run general cleanup tasks. The user defined on_stop(self) function is executed at the end of this state. The Stop state is the final state in the Lifecycle and will sit here until user intervention requests a transition to the Start state.

Messenger

This is the means of communication with and between Reactives. Other than instantiating the Reactives all interaction is done through the Messenger. The Messenger is a singleton that manages all of the Reactive's queues and any subjects that Reactives can subscribe to.

Direct Message

The Messenger can send messages directly to a Reactive by calling send() with the name of the Reactive along with a message.

from reactive import Messenger
def main():
    messenger = Messenger()
    messenger.send("my_reactive", "test")

Pub-Sub

Reactives can subscribe to subjects for a Pub-Sub style messaging. Reactives register to listen to subjects during instantiation. Messages sent to a subject are then forwarded to all Reactives that subscribe to that subject.

my_reactive = My_Reactive(subjects=['count'])
messenger.send('count', {increment: 2})

Start

Request the Reactive enter the start state of the lifecycle. If a subject is given, all subscribed Reactives will be requested to enter the start state. If no name is given, all registered Reactives will be requested to enter the start state.

messenger.start("r_2")   <-- Will only start the Reactive named 'r_2'
messenger.start("count") <-- Will start all Reactives subscribed to subject 'count'
messenger.start()        <-- Will start all registered Reactives

Stop

Request the Reactive enter the stop state of the lifecycle. If a subject is given, all subscribed Reactives will be requested to enter the stop state. If no name is given, all registered Reactives will be requested to enter the stop state.

messenger.stop("r_2")   <-- Will only stop the Reactive named 'r_2'
messenger.stop("count") <-- Will stop all Reactives subscribed to subject 'count'
messenger.stop()        <-- Will stop all registered Reactives

Status

Request the status of a given Reactive. If a subject is given, the status of all subscribed Reactives will be requested. If no name is given, the status of all registered Reactives will be requested.

messenger.status("r_2")   <-- Will only return the status of the Reactive named 'r_2'
messenger.status("count") <-- Will return the status of all Reactives subscribed to subject 'count'
messenger.status()        <-- Will return the status of all registered Reactives

Getting started

Installation

pip install reactives

Support

I don't get paid for this and it is just a fun experiment so it is provided as-is.

Roadmap

What will we think of next?

Contributing

This is just a fun experiment built for my own exploration, education, and entertainment so no contributions are accepted at this time. I am open to hearing about additional features, improvements, etc..

Authors and acknowledgment

Me, myself, and I

License

Beerware

Project status

Experimental