Reputation: 2541
I am creating a custom object to be used in some internal applications where I work. I researched some ways to go about doing this - and this is what I came out with.
function ISGrader(text)
{
this.text = text;
this.printInfo = function(){
alert("Object working " + text);
}
this.putGrade = function(score)
{
alert(score);
}
}
I believe this shows constructor-type functionality, as well as some simple starter methods that I will build on.
Is the above good practice or is there another way that is more standard?
Upvotes: 2
Views: 219
Reputation: 141859
I prefer a pattern similar to the one below. You can think of it as a 4-step approach:
(function(parent) {
// 1. Declare private variables and functions that will be
// accessible by everybody within the scope of this
// function, but not outside of it.
var doSomethingAwesome = function() { .. }; // private function
var coolInteger = 42; // private variable
// 2. Create the constructor function
function ISGrader() {
..
}
// 3. Create shared public methods on the prototype object.
// These will be created only once, and shared between all objects
// which is more efficient that re-creating each method for each object.
ISGrader.prototype.printInfo = function() { .. };
ISGrader.prototype.putGrade = function(score) { .. };
// 4. Expose the constructor to the outside world.
parent.ISGrader = ISGrader;
})(window);
The reason why everything is enclosed inside a self-executing anonymous function is to ensure the private variables we create inside don't leak outside to the enclosing scope, and to basically keep things clean.
Another benefit of declaring the constructor like this is that you can change the parent object easily from say window
to a further namespaced object by changing a single word.
Upvotes: 2
Reputation: 25135
It is always recommended to do it using `prototype'. This way you can also inherit it's properties and create new one.
var ISGrader = function(text) {
this.text = text;
var _privateVar = text;
this.updatePrivateVar = function(newText) {
_privateVar = newText;
alert("private variable updated");
}
}
ISGrader.prototype.text = "";
ISGrader.prototype.printInfo = function() {
alert("Object working " + this.text);
}
ISGrader.prototype.putGrade = function(score) {
alert(score);
}
var isGrader = new ISGrader("hello");
isGrader.printInfo();
// Inherit and create a new definition
var ISGrader2 = function() {}
ISGrader2.prototype = new ISGrader();
var isGrader2 = new ISGrader("hello2");
isGrader2.printInfo();
isGrader2.updatePrivateVar("hello3");
demo : http://jsfiddle.net/rkspP/3/
Upvotes: 1
Reputation: 15666
If you are planning on creating multiple ISGrader
objects on a single page, it's more memory efficient to stick the functions in a prototype object assigned to ISGrader
like this:
function ISGrader(text) {
this.text = text;
}
ISGrader.prototype = {
printInfo: function() {
alert("Object working " + this.text);
},
putGrade: function(score) {
alert(score);
}
}
Upvotes: 0
Reputation: 7517
While not really an answer, I recommend Douglas Crockford's book JavaScript: The Good Parts as it does a good job of introducing you to the "good parts" of the language and discusses the pros and cons of the different ways to create objects in JavaScript.
You can also review this resource if you're just looking for an explanation of member visibility in JavaScript objects: http://javascript.crockford.com/private.html
Upvotes: 0
Reputation: 46647
I prefer this pattern (IIFE), but it is purely opinion:
var ISGrader = (function (text) {
// anything declared here is "private"
var printInfo = function() {
alert("Object working " + text);
};
var putGrade = function (score) {
alert(score);
};
// put "publicly" accesible things in the returned object
return {
text: text,
printInfo: printInfo,
putGrade: putGrade
};
})(text);
Upvotes: 2