This would presumably be the safest way (Case A):
var myClass = function() { };
myClass.prototype = {
doSomething : function() { alert('Something'); }
};
This is the alternative (Case B):
var myClass = function() {
this.doSomething = function() { alert('Something'); };
};
I'm under the impression that by doing this as shown in Case B, doSomething would be a member, and the function would be defined once for each myClass object I instantiate so that it will exist 100 times in memory for 100 instances, whereas in Case A the function will only exist in one place in memory and different instances will merely reference the prototype.
Am I understanding this correctly?
As a bonus question: When doing it as in Case B, chrome developer gives me intellisense for doSomething, but I must expand __proto__ for an instance to see it. How come it doesn't show up on the object itself? That is, why doesn't prototype members show on the object, but get stuck back down on __proto__? I would have preferred if the __proto__ stack would get flattened and show up on the object directly. Is there another Case that will allow this to happen?
Firstly, in case B, you are merely creating a global function, not attaching it to the instance. You meant:
this.doSomething = function() { }
Secondly, the first will be faster. Though I can't find the link now, jQuery honcho John Resig did a detailed blog post on this showing speed tests on prototypal inheritance of methods vs. methods declared on the instance. Inheritance was notably faster.
In terms of ethos, I've always much favoured inheritance. This is the place for reusable, cross-instance functionality. Adding it to each instance has the sole benefit of allowing you to declare methods inside a single, convenient closure, in your constructor, but that's it.
If this is your reason for liking pattern B, it's possible to do this whilst still a) having the methods inherited; b) not redeclaring them at every instantiation.
function SomeClass() {
if (!SomeClass.prototype.someMethod) {
SomeClass.prototype.someMethod = function() {}
}
}
This will slightly slow down the initial instantiation, though, as it is responsible for setting up the prototype - not really the job of an instantiation process.
There is also a programmatical difference to be aware of between your two cases:
function SomeClass(name) {}
SomeClass.prototype.someMethod = function() {};
var instance = new SomeClass();
console.log(!!instance.someMethod); //true
console.log(instance.hasOwnProperty('someMethod')); //false
The last line is false because the method is inherited, not owned by the instance. With your pattern B, this will resolve to true.
Correct: defining methods in the prototype will create 1 function object, and every instance will reference that 1 function. Defining it in the constructor creates a new function for each instance
Your code needs some work. The way you're defining the constructor, the doSomething function is defined as a global add var to counter that. This still doesn't set doSomething as a property though, it's just a function declared within the scope of the constructor (closure). This is why it doesn't show up in your instance as a method: the function is not attached to this, but even when fix this issue like so, you're still creating new function objects for each instance:
function MyConstructor()//capitalize constructors - conventions are important
{
var someMethod = function(){/*..*/};
this.someMethod = someMethod;
}
Utkanos pointed out what the implications of inheritance and prototype methods are (.hasOwnProperty), and he's absolutely right in that respect (+1). I'd just like to add that the hasOwnProperty method returning false is a trivial matter. Generally speaking, when iterating over an object, and checking which properties and methods are set, and which are not. What you want in most cases are properties, not the methods. So it's in fact better to set them at the prototype level:
for(var name in obj)
{
if (obj.hasOwnProperty(name))
{
//do stuff, here the methods are set # prototype level
}
if (obj.hasOwnPrototype(name) && typeof obj[name] !== 'function')
{
//same stuff, but requires extra check when methods are assigned by constructor
}
}
Related
In the following code I instantiate an object and it inherits properties in two completely different ways.
function C(){
this.k1 = "v1";
}
C.prototype.k2 = "v2";
var o = new C;
The first (k1) seems to attach to the instantiated object (o) as an 'own' property, the second (k2) does not attach to anything in the instantiated object but seems to just access it's value through the _ _ proto _ _ property (or [[prototype]]). In a way it seems like the first way is analogous to (in every computing) making a 'copy' of a file on your desktop whereas the second way is analogous to making an 'alias' of a file on your desktop, and of course, in this analogy I'm makin 'files' are analogous to 'properties'.
I'm wondering if this is the correct way of imagining what's taking place and if so what the benefits and drawbacks are to using one method over the other. I'm assuming that using the prototype property saves memory because it doesn't force the newly instantiated object (o) to have keys (k1) as an 'own' property, but maybe I'm wrong.
Why would anyone choose one method over another?
I'm wondering if this is the correct way of imagining what's taking place
Yes.
Why would anyone choose one method over another?
.prototype
values are shared amongst all instances
therefore takes less memory and is faster: Defining methods via prototype vs using this in the constructor - really a performance difference?
This is especially beneficial for methods (with their dynamic this binding)
this
values are specific to each instance
methods defined there are closure which have access to the constructor scope and therefore allows true privacy.
The own properties are necessary for the data that distinguishes the instances.
Think of the prototype as what is inherited. In a classic OOP world, it's almost like a parent class.
The advantage of using a prototype is that for instances of C which do not customize k2, there will be only a single value stored in memory. If you set the property on an instance, however, you create a new property that is within the object directly (hasOwnProperty will return true).
The disadvantage is that it may not be clear to all who read your code how that value gets set.
Also, keep in mind, if a default does not make sense, and you always need to compute it, then putting it in the prototype makes little sense.
FYI, this is why you will often see methods bound to an object via its prototype, since they just define code and are generally stateless by themselves.
function C() {
this.blah = function () {
return 1;
};
// other stuff
}
var a = new C();
var b = new C();
You have 2 functions named blah in memory, not 1.
Versus:
function C() {
// other stuff
}
C.prototype.blah = function () {
return 1;
};
var a = new C();
var b = new C();
Now you only have a single blah function in memory, bound to 2 separate objects.
(Obviously I'm not sure what the title of this should be.)
I'd like to be able to modify the prototypes of native objects with minimal risk (because trust me; I have heard over and over again how it's a mortal javascript sin).
My train of thought goes something like this: I can greatly reduce the possibility of conflicts if I can add my custom functions to a single object on a prototype rather than directly adding them to the prototype itself.
If I have the code,
String.prototype.foo = function() {
//do stuff
};
...obviously the keyword this references the string that called it. However, if I do something like,
String.prototype.foo = {};
String.prototype.foo.bar = function() {
//do stuff
};
...the keyword this references the object foo.
If there some way to overcome this?
The only way to overcome this would be to override the context using call:
String.prototype.foo.bar.call('hello'))
which works, but is pretty ridiculous.
You're better off (as you've already heard) avoiding modifying native prototypes, and instead having an object with your methods accepting the string as a parameter.
Well, there's a couple of ways to do this, depending on how much work you want to put in.
Using bind is the most straightforward way, but you have to define the string as a variable so you can pass a reference of it to bind:
String.prototype.foo = function() { console.log(this); }
String.prototype.foo.bar = function() { console.log(this); }
var g = "hi there";
g.foo() // Yields g
g.foo.bar() // Yields String.prototype.foo()
g.foo.bar.bind(g)() // Yields g again.
There may be another very hackish way to produce the desired result by creating getters and setters on String.prototype.foo so that String.prototype.foo.bar activates a function that returns a function bound to the instance that foo refers to. ??? Confusing.
What might be the best solution to reduce the possibility of conflicts, is to take advantage of prototypal inheritance and create your own sub prototype of the native String.
function MyString(string) {
this.foo = function() { //whatever };
this.foo.bar = (function() {}).bind(this);
this.toString = function() { return string; } // because toString is not generic.
}
MyString.prototype = new String();
Here, you're creating your own "sub-prototype" of the String prototype. It inherits all the properties of the String prototype and adds its own, all without altering the native strings at all. BONUS: this.foo and this.foo.bar will both refer to your instance.
var instance = new MyString("hi");
instance.foo(); // this == instance
instance.foo.bar(); // this == instance
instance.replace("i", "e"); // returns "he"
This maybe wasn't the answer you were looking for, but hopefully it will at least be helpful.
Please help me decide whether I should use a function's prototype object and the "new" keyword or completely stay away from constructor functions.
Situation:
Function called widget() that will be called 10-15 times to initialize each widget on the page. widget() contains quite a few internal methods.
Each time widget() is called, the function needs to return an object that acts as an API to operate on the widget.
Question
1) Do I put all the internal methods inside Widget() under its prototype property? It does not make sense but the main reason for this is to not re-instantiate the internal functions every time widget() is called.
But if I do put the internal functions in prototype, each instantiated w object (w = new Widget();) has access to internal private methods.
2) If I stay away from constructor functions and new keyword and structure my code as down below, how do I fix the performance concern of the internal functions getting re-instantiated every time widget() is called.
function widget()
{
var returnObj = {};
/* Add internal functions but this will be re-instantiated every time */
return returnObj;
}
You have a bit of a tradeoff here. As you seem to already understand, methods you put on the .prototype are publicly available, but that is the most efficient places to put methods as they are automatically added to all new copies of that object in a very efficient manner. When using .prototype for methods, there is only one copy of your methods and a reference to that single copy is automatically added to all new instantiations of that object.
But, javascript doesn't have private methods built-in and the only work-around for that involves not using the .prototype for them or for any methods that need to call the private methods.
This article by Doug Crockford is a pretty good description of how you can create privacy for either data or methods in any object.
In either case, I don't see any reason to avoid using the new keyword to create new objects. You can make either .prototype or private methods work with new.
But, if you want to achieve truly private methods, then you can't use .prototype for either the private methods or any methods that need to access them so you have to decide which is more important to you. There is no single correct answer because your need for privacy is situation-specific.
In my coding, I generally don't enforce privacy and I do use .prototype and new. I designate "non-public" methods on the prototype by starting their name with an underscore. This is a notational convention, not an access enforcement scheme.
In answer to your second question about avoiding the new operator and reinstantiating methods, I'd just ask why you're doing this? What are you gaining? I'm not aware of any downsides to using new. As best I understand your decision about whether to use .prototype vs. manually create/assign methods in your constructor should be about the need for private methods.
FYI, 15 objects is hardly going to create a significant difference in performance either way here. I would evaluate your need for true privacy and make your decision based on that. If you HAVE to enforce privacy, then go with the Crockford method for implementing private methods. If you don't HAVE to have true privacy, then use .prototype. I don't see a reason here to avoid using new in either case.
You can use a metaconstructor* pattern to get around this.
function defineCtor(metaCtor) {
var proto = new metaCtor();
var ctor = proto.hasOwnProperty('constructor') ?
proto.constructor : new Function();
return (ctor.prototype = proto).constructor = ctor;
}
Now you have a function that constructs constructors (or more accurately constructs prototypes and returns constructors).
var Widget = defineCtor(function() {
function doInternalStuff() {
// ...cant see me
}
// this function ends up on the prototype
this.getFoo = function() { return doInternalStuff(); };
});
// ...
var myWidget = new Widget();
Explanation
defineCtor takes a single anonymous function as a property. It invokes the function with new, creating an object. It assigns the object as the prototype property of a new constructor function (either an empty function, or the generated prototype object's own constructor property), and returns that function.
This provides a closure for your internal functions, addressing your question 1, and sets up the constructor/prototype pair for you, addressing question 2.
Comparison
Compare the defineCtor technique to the following two examples.
This example uses the prototype, and has problem 1: the internal stuff is not encapsulated.
function Widget(options) {
this.options = options;
}
Widget.prototype = {
getFoo: function() {
return doInternalStuff();
}
};
// How to encapsulate this?
function doInternalStuff() { /* ... */ }
This example sets up everything in a constructor, and has problem 2: each time it constructs an object, it instantiates new function objects for each property.
function Widget(options) {
this.options = options;
function doInternalStuff() { /* ... */ }
this.getFoo = function() {
return doInternalStuff();
};
}
This example uses the technique described above to provide encapsulation while still leveraging the prototype:
var Widget = defineCtor(function() {
// ^
// This function runs once, constructing the prototype.
// In here, `this` refers to the prototype.
// The real constructor.
this.constructor = function(options) {
// In function properties, `this` is an object instance
// with the outer `this` in its prototype chain.
this.options = options;
};
function doInternalStuff() { /* ... */ }
this.getFoo = function() { return doInternalStuff(); };
});
// ...
var myWidget = new Widget();
This approach has a few benefits, some more immediately obvious than others.
It provides encapsulation. You could do this by wrapping the first "comparison" example in an immediately invoked function, but this approach may be cleaner and more easily "enforced" in a team setting.
It's extensible. You can give your "metaconstructor" functions their own prototypes, with function properties like "extends", "mixin", etc. Then, inside the body of metaCtor, you can write things like this.extends(BaseWidget). The defineCtor API never needs to change for any of this to happen.
It "tricks" Google Closure Compiler, Eclipse, jsdoc, etc. into thinking you are defining the actual constructor function rather than a "meta function." This can be useful in certain situations (the code is "self-documented" in a way these tools understand).
* As far as I know, the word "metaconstructor" is completely made up.
As far as I can tell, there are two main ways of creating functions for an object in javascript. They are:
Method A, make it in the constructor:
function MyObject() {
this.myFunc1 = function() {
...
}
this.myFunc2 = function() {
...
}
...
}
Method B, add it to the prototype:
function MyObject() {
...
}
MyObject.prototype.myFunc1 = function() {
...
}
MyObject.prototype.myFunc2 = function() {
....
}
Obviously if you did:
MyObject.myFunc3 = function() {
....
}
then myFunc3 would become associated with MyObject itself, and not any new objects created with the new keyword. For clarity, we'll call it method C, even though it doesn't work for creating new objects with the new keyword.
So, I would like to know what the differences between the two are. As far as I can tell they have the same effect logically, even if what's happening on the machine is different.
If I were to guess I would say that the only real difference is when you're defining them in the constructor like in method A, it creates a whole new function object for each object that's created, and Method B only keeps one copy of it (in MyObject), that it refers to any time it's called. if this is the case, why would you do it one way over the other. Otherwise, what is the difference between method A and method B.
The advantage of giving a separate function to each object is that you can close over variables in the constructor, essentially allowing for "private data".
function MyObject(a,b) {
var n = a + b; //private variable
this.myFunc1 = function() {
console.log(n);
}
};
vs
function MyObject(a,b) {
this.n = a + b; //public variable
}
MyObject.prototype.myFunc1 = function() {
console.log(this.n);
}
Whether this is a good idea or not depends on who you ask. My personal stance is reserving constructor functions for when I actually use the prototype, as in option #2 and using plain functions (say, make_my_object(a,b)) when using closures, as in option #1.
The idea is that you can modify the prototype at any time and all objects of the type (even those created before the modification) will inherit the changes. This is because, as you mentioned, the prototype is not copied with every new instance.
The MyObject in method A is an instance for inner functions.
You cannot call its functions explicitly outside of it unless object (you can call it a class) was instantiated.
Assume this:
MyObject.MyFunc1(); // will not work
var obj = new MyObject();
obj.MyFunc1(); // will work
so this is the same as any class in other languages. Describing usefulness of classes and their usages goes beyond that question though.
Also to notice:
function MyObject() {
var privateVar = "foo";
this.publicProperty = "bar";
// public function
this.publicFunc = function() {
...
}
// private function
function privateFunc () {
...
}
}
For method B it's same as with method A, the only difference is prototyping is a style of creating object. Some use prototypes for readability or out of preference.
The main advantage in prototypes is that you can extend existing object without touching the original source. You need to be careful with that though.
(as example Prototype framework)
For method C you can call them a static functions. As you said they can be called explicitly by referring through object like:
MyObject.MyFunc1();
So which one to use depends on situation you're handling.
I am trying to create a prototype/class that inherits from the object that is returned by a specific function. But I dont know what that object's name is?
For the instance var xhr = XMLHttpRequest(); what object is returned by xhr.responseXML;? Is it XMLDocument? Or maybe XMLDOM?
Also if I create the object var xmlDoc = new ActiveXObject("Microsoft.XMLDOM"); and call xmlDoc.load("xmlFile.xml" ); will it return the same type of object as xhr.responseXML;?
I am trying to do the following:
function XMLHandler()
{
this.xmlFile = "defaultXML.xml";
}
// Make XMLHandler inherit from Javascript object
XMLHandler.prototype = new XMLDocument();
XMLHandler.prototype.constructor = XMLDocument;
You shouldn't include unrelated questions in the same post. To address your second question about declaring a function within a "class":
Your first syntax, declaring it within the constructor, will pointlessesly overwrite myClass.prototype.publicFunct to point to a newly created function every time the constructor is called. That is, you will be creating an extra copy of the function with each call to the constructor, with myClass.prototype.publicFunct always pointing to the most recently created copy - or undefined until the constructor has been called at least once. Don't do this.
Your second option is simply invalid syntax in JavaScript.
Your third syntax is prefered. This will add the function to the prototype once.
Remember: JavaScript doesn't have classes as such, though you can sort of simulate them if you think it is worth the bother.
I can tell by your recent questions that you are thinking like Java, which this is not.
First question:
responseXML is different for each browser. Firefox gives a nsIDOMDocument, IE gives an IXMLDOMDocument and Webkit browsers depend on the responseType setting but will probably be Document. Since you cannot predict what it will be stop trying to extend it. In most cases the type isn't made available by the browser's API so javascript cannot extend it anyway.
Moreover, since JavaScript's inheritance is not class based you are forced into doing this:
XMLHandler.prototype = new XMLDocument();
...which simply does not work for your purpose. Any instance of XMLHandler will be built on an unrelated, empty document and not the document returned by responseXML. You have to use a wrapper here.
Second Question:
Of your 3 methods the first is equivalent to the last but more wasteful because it repeatedly sets the same function to the same prototype. The second is nonsensical, the syntax is broken. These are your real options:
// Instance method, every instance is given a copy of the function upon "new"
function MyClass()
{
this.publicFunct = function()
{
alert("public function");
};
}
// Prototypal method, only one copy of the function shared by all instances
function MyClass()
{
}
MyClass.prototype.publicFunct = function()
{
alert("public function");
};
// Shorthand method, same as prototypal but handy for several members at once
// It prevents MyClass from being a descendent of another type
function MyClass()
{
}
MyClass.prototype = {
// A colon is only acceptable in object notation
publicFunct: function()
{
alert("public function");
}
};
I would go with a prototypal method for efficiency unless you need to selectively add functions to a class. Your use of "public function" (also "class") seems like another symptom of an OOP background, there aren't any private functions in JavaScript so "public" has no place, all member functions are public. If at some point you do need a private function you can fake the effect with a closure.
(function() {
// Assignments are mostly global
MyClass = function() {};
MyClass.prototype.publicFunct = function()
{
privateFunct();
};
// These statements affect local scope
var foo = 'bar';
function privateFunct()
{
alert("public function");
}
})(); // These extra brackets cause the contents to be executed immediately
Having said that it is rare to need private functions and all JavaScript is visible anyway so it's not secret really. The above could be thwarted like this:
thief = {};
MyClass.prototype.publicFunct.call(thief);
// privateFunct is called by publicFunct in the context of the thief
You might as well accept that functions are public. You can go a step further and give up on classes altogether. Objects are just objects that happen to have some functions and those functions can even be shared with completely different objects.