dbindjs
dbindjs is data binding for Javascript.
Import with import { dbind } from 'dbindjs'
Run in Browser
Probably the easiest way to make dbindjs run in browser is to upload the file dist/dbindjs.js to the scripts' location of your webserver and use a standard import as follows:
<html>
<head>
...
</head>
<body>
<script type="module">
import { dbind } from './scripts/dbindjs.js'
// ...
</script>
</body>
</html>Examples
A couple of examples are provided here below. For more see examples folder on GitHub and dbindjs Documentation.
Basic dbindjs
import { dbind } from 'dbindjs'
dbind({
// Property, Basic
// when the value of this property changes, the bindings depending on it automatically update
a: 1,
// Property, Basic
b: 1,
// Binding, Basic
// defines a binding relation (which may include internal bindings) within the pool,
// and uses results outside of the pool
c: function () {
// use references to basic properties of the binding pool
const dep = this.d()
const inrc = this.a / this.b / dep
console.log(inrc)
// use result somewhere else outside of the binding pool
// ...
},
// Binding, Internal
// defines a binding relation which is used by other bindings in the binding pool
d: function () {
const inrd = this.a + this.b + this.f
console.log(inrd)
// output for other bindings
return inrd
},
// Binding, Complex
// defines a binding relation (which may include internal bindings) within the pool,
// uses results outside of the pool
// and can be used by other bindings in the binding pool
e: function () {
// use references to basic properties of the binding pool
const dep = this.d()
const inre = this.a + this.b + this.f + dep
console.log(inre)
// use result somewhere else outside of the binding pool
// ...
// output for other bindings
return inre
},
// Binding, Basic
f: 2
// ...
// ...
})
// trigger the bindings by changing the values of some dependencies
dbind({ a: 23, f: 45 })
Neural networks with dbindjs
This example simulates a fully connected neural network of maxNeurons 'neurons'
(100000). The number of 'neurons' can be increased or decreased to match testing
machine capabilities. When using the default value (100000 'neurons'), the example
should run pretty fast on a standard personal computer.
Changes to any components of this network trigger propagation to all 'neurons'.
All sorts of actions can be triggered. This example prints a result if 'neuron' data meets a certain condition.
import { dbind } from 'dbindjs'
const maxNeurons = 100000
const miniBrain = {}
for (let i = 0; i < maxNeurons; i++) {
miniBrain['n' + i] = {
synapseFiringTrigger: 0,
neuronData: {
a: Math.random() * 100,
b: Math.random() * 100000
},
// important! remember fat arrow functions loose 'this',
// so don't use them with dbindjs binding functions
act: function () {
if (this.neuronData.a > this.neuronData.b) {
console.log('n' + i + ': ', this.neuronData)
}
// alter data in order to create randomness for the next calls
// this.neuronData.a = Math.random() * 100
// this.neuronData.b = Math.random() * 100000
}
}
}
const desc = {
action1: function () {
for (const prop in this) {
if (miniBrain.hasOwnProperty(prop) && miniBrain[prop].hasOwnProperty('act')) {
miniBrain[prop].act()
}
}
}
}
for (var prop in miniBrain) {
desc[prop] = miniBrain[prop].synapseFiringTrigger
}
dbind(desc)
// pick a random neuron
const neuronId = 'n' + Math.round(Math.random() * maxNeurons)
const shakeTheNetwork = {}
shakeTheNetwork[neuronId] = dbind.propstore[neuronId].value++
dbind(shakeTheNetwork)
Others
dbindjs incorporates a couple of advantages over existing Javascript data binding libraries:
- data binding only
- merged updates
- separation through namespaces
- advanced fine-tuning
- consistent protection
- pause/resume features
- introduces the concept of 'binding pool'
Further developments arriving soon.
dbindjs is © Copyright 2017-2023 Nicolae Iotu, nicolae.g.iotu@gmail.com