data-observer

This library wraps Object.observe and Array.observe to make them more usable. Warning: This library has not yet been tested in production.

Overview

Observing Objects

To observe an object, pass it to the observe function to return a scalar observation.

import observe from 'data-observer';

let object = {a: 1, b: 2, c: 3};

// observe a property
let observation = observe(object, 'a')
observation.getValue(); // => 1
observation.onDidChangeValue((a) => { // called when `object.a` changes
  console.log('a changed:', a)
});

// observe multiple properties
// pass a function to combine them into a single value
let observation = observe(object, 'a', 'b', 'c', (a, b, c) => a + b + c)
observation.getValue(); // => 6
observation.onDidChangeValue((value) => { // called when `object.a`, `.b`, or `.c` change
  console.log('a, b, or c changed:', value)
});

// observe the whole object
let observation = observe(object);
observation.getValue() // object
let disposable = observation.onDidChangeValue((object) => { // called when object changes
  console.log('object changed', object);
})

Observing Arrays

This library is much better than the raw Array.observe API, because it efficiently coalesces the overlapping change events delivered by the raw API into a minimal list of non-overlapping changes. If you apply these changes to a copy of the original array in order, it will equal the current value of the observed array.

let array = ['a', 'b', 'c', 'd', 'e'];
let original = array.slice();

observe(array).onDidChangeValues(changes => { // called when array changes
  for (let {index, removedCount, added} of changes) {
    original.splice(index, removedCount, ...added);
  }
});

Mapping Over Observations

The observations are returned from the observe function support a map operation, which maps a transformation function over the current value(s) of the observation. Here we map value => value + 1 over an array observation.

let array = [1, 2, 3];
let observation = observe(array);
let mappedObservation = observation.map(value => value + 1);
mappedObservation.getValues(); // => [2, 3, 4]

The transform will also be applied to all added values passed to onDidChangeValues listeners.

Mapping over scalar observations is similar:

let object = {a: 'hello'};
let observation = observe(object, 'a');
let mappedObservation = observation.map(value => value.toUpperCase());
mappedObservation.getValue(); // => 'HELLO';

You can chain calls to map if you're so inclined:

let doubleMapObservation =
  observation.map(value => value + 1).map(value => value * 6);

You can also apply a map operation by passing a transform function as the last argument to observe:

let observation = observe(array, value => value + 1);

API

observe(object, [...propertyNames], [transformFunction])

Observe an entire object or specific properties on an object.

• object The object to observe.
• ...propertyNames The names of one or more properties to observe. If more than one property name is provided, you must supply a transformFunction to combine them.
• transformationFunction An optional function taking one or more observed values as arguments that returns a single combined value for the observation. This argument is required if more than one property name is supplied.

Returns a scalar observation.

observe(array, [transformFunction])

Observe an array.

• array The array to observe.
• transformFunction An optional function transforming the values of the observed array.

Returns an array observation.

Scalar Observations

These are returned when observing an object or an object's fields. Scalar observations represent a single value that changes over time.

getValue()

Get the current value of the observation.

onDidChangeValue(fn)

Subscribe to changes to the observation's value.

• fn A function that will be called with the current value of the observation whenever it changes.

Returns a Disposable on which .dispose() can be called to cancel the subscription.

map(transformFn)

Build a new observation based on applying the given transform function to the this observation's value.

• transformFn A function that transforms the observation's value.

Returns a new scalar observation.

Array Observations

getValues()

Get the current values of the observed array.

onDidChangeValues(fn)

Subscribe to changes to the observed array. Applying the changes passed to the listener function to a copy of the array as it existed before being changed in order via splice should update the copy to match the current state of the array.

• fn A function that will be called with an array of changes whenever the observed array changes.
  • changes An Array of change objects with the following keys:
    • index Where the change starts.
    • removedCount The number of elements that were removed.
    • added The elements that were added.

Returns a Disposable on which .dispose() can be called to cancel the subscription.

map(transformFn)

Build a new observation based on applying the given transform function to the this observation's values.

• transformFn A function that transforms the observation's values.

Returns a new array observation.

Rationale

I built this library so I can use it from a view framework that I haven't finished writing yet. It will combine virtual-DOM diffing with data-model observation. Stay tuned.

Ideas

Array observations could be extended with relational methods, such as filter, join, flatMap, etc. This would enable us to use the ideas of functional relational programming described in Out of the Tarpit (PDF) while still using simple JavaScript primitives. Scalar observations could similarly be extended with an API reminiscent of Rx.