neural_stream 0.1.0

A library for neurologically oriented programming

Neural Stream

Neural stream is a library designed to simplify reactive programming.

It provides an interface exactly like a normal stream to make it easy to pickup for anyone already familiar with streams however it has several key differences.

1. Not Type Bound Unlike a regular stream a neural stream is not bound to a specific type. This allows the user to manage all their events from a single stream.

2. Automatic Event Chaining A key difference to a regular stream is that the return value of a listener function is automatically be fed back into the stream.

This provides a mechanism for chaining reactions together without pulling the stream into the scope of the reaction.

The same is occurs with any exceptions thrown during a listener's computation.

Hello World

import 'package:event_stream/event_stream.dart';

void main() {

    NeuralStream stream = NeuralStream();
    
    stream.listen((String text) async {
        print('sub: $text');
    });
    
    stream.add('hello world'); // output: 'sub: hello world'
}

Trigger Custom Events

You can subscribe to any type. Note: a trigger should be immutable.

class HelloWorld {
  final String message;
  HelloWorld(this.message);
}

stream.listen((HelloWorld helloWorld) async {
    print(helloWorld.message);
});

stream.add(HelloWorld('this is an example'));

Multiple subscriptions of same type

You are not bound to listening to a trigger by just one listener Any number of listeners of the same type can be created.

NeuralStream stream = NeuralStream();
    
stream.listen((String text) async {
    print('sub 1: $text');
});
stream.listen((String text) async {
    print('sub 2: $text');
});

stream.add('hello'); 
// output: sub 1: hello
// output: sub 2: hello

Subscriptions are configurable

NeuralStream stream = NeuralStream();

stream.listen((String text) async {
    print('sub 1: $text');
}, max: 2); // limit this subscription to listen to a maximum of 2 events

stream.add('hello 1'); 
stream.add('hello 2'); 
stream.add('hello 3'); 

// output
// sub 1: hello 1
// sub 1: hello 2

Cancel Subscriptions

NeuralStream stream = NeuralStream();

Subscription<String> subscription = stream.listen((String text) async {
    print('sub 1: $text');
});

stream.add('hello before subscription cancelled'); 
stream.cancel(subscription);
stream.add('hello after subscription cancelled'); 
// output: sub 1: hello before subscription cancelled

Why Neural Stream?

A normal stream is bound to a specific type

Another key difference of the super stream architecture is that it automatically passes the result of the reaction back into the stream which binds the reactions together without the user manually having to connect them.

Neurologically Oriented Programming

The inspiration for this architecture came from the brain.

Our brain consists of a network of neurons. Signals travel through our brain from neuron to neuron either activating it or not.

If a neuron is activated it performs some kind of computation and can then produce a new signal which will be once again passed to all the other neurons in the brain.

This structure is useful because the neuron doesn't have to know anything about the rest of the brain, it simply produces a signal and its job is finished.