We Keep Coding

JS: Javascript closure issue in prototype instance

As for a JS-newbie, I'm struggling with JS-closure. Googled enough, though can't understand why "this" is not available in "store" function. Any help will be appreciated:
;(function (win) {
var _af = function(storeObjectName, storeObjectValue) {
var self;
if (_af.prototype.instance) {
self = _af.prototype.instance;
} else {
self = Object.create(_af.prototype);
_af.prototype.instance = self;
self._init();
}
if (arguments.length == 1) return self.fire(storeObjectName);
if (arguments.length == 2) self.store(storeObjectName, storeObjectValue);
return self;
};
_af.prototype = {
afVariable: '_af',
afObject: {},
_init : function(){
this.afObject = this.get(self.afVariable);
},
store : (storeObjectName, storeObjectValue)=>{
// This throws the error, that this.get is not defined
this.get('_af');
},
get : storageObject=>{
if (!storageObject)
this.afObject = '_af';
else
this.afObject = '_someother'
}
}
win._af = _af;
}(window));

It's because you use a lambda expression ((...) => ...) instead of function() { ....
In JS this depends a lot on the context. In your case, since the prototype is an object, whenever you want to use this to refer to that object, you must do it in a function defined in the object like:
store : function(storeObjectName, storeObjectValue) {
// This throws the error, that this.get is not defined
this.get('_af');
},
or (also correct):
store(storeObjectName, storeObjectValue) {
// This throws the error, that this.get is not defined
this.get('_af');
},
Note you are doing it correctly for _init, but not for store and get
When you use the lambda syntax, this will refer to the current this, when a function is created, so it will be window.
Also avoid using get as a function / or any other name because it is used for defining getters.

Module Pattern: var module = new function() {...} vs. var module = (function () {...}()); [duplicate]

I'm just trying to understand Javascript a little deeper.
I created a 'class' gameData that I only want ONE of, doesn't need a constructor, or instantiated.
So I created it like so...
var gameData = new function () {
//May need this later
this.init = function () {
};
this.storageAvailable = function () {
if (typeof (Storage) !== "undefined") {
return true;
}
else {
return false;
}
};
}
Realizing that the 'new' keyword doesn't allow it to be instantiated and makes it available LIKE a static class would be in C#.
Am I thinking of this correctly? As static?

No, it is not static because it still has a constructor property pointing to your "anonymous" function. In your example, you could use
var gameData2 = new (gameData.constructor)();
to reinstantiate a second object, so the "class" (instance actually) is not really "static". You are basically leaking the constructor, and possibly the data that is bound to it. Also, a useless prototype object (gameData.constructor.prototype) does get created and is inserted in the prototype chain of gameData, which is not what you want.
Instead, you might use
a single, simple object literal (as in Daff's answer). That means you don't have a constructor, no closure-scoped private variables (you have used none anyway) and no (custom) prototype.
the (revealing) module pattern (as in jAndy's answer). There you'd have an IIFE to create closure-scoped variables, and can return any kind of object.
an actual constructor ("class") that can be instantiated later (when needed), and yields the same singleton object always.
This is what the singleton pattern would look like:
function GameData() {
if (this.constructor.singleton)
return this.constructor.singleton;
else
this.constructor.singleton = this;
// init:
// * private vars
// * public properties
// ...
}
GameData.prototype.storageAvailable = function () {
if (typeof (Storage) !== "undefined") {
return true;
}
else {
return false;
}
};
var gameData = new GameData();
var gameData2 = new GameData();
gameData === gameData2 === GameData.singleton; // true
Yet, the prototype is quite useless because you have only one instance of GameData. It would only get interesting with inheritance.

There is no Class in ECMAscript, there is only Object.
When new is used to invoke a function, we call it a constructor function. This function somewhat auto returns a new object once it finished. Any data that is stored within that object using this (which references that newly created object) is returned as property of that object. Beside that, new sets a property called constructor to exactly this function.
In your case, you don't even really require the usage of new, you could easily re-write your code as follows:
var gameData = (function () {
var public = { },
private = { }; // any private data can get stored here
//May need this later
public.init = function () {
};
public.storageAvailable = function () {
if (typeof (Storage) !== "undefined") {
return true;
}
else {
return false;
}
};
return public;
}());
This is called the factory pattern, singleton pattern, module pattern, and there might be some other names.

I think what you are looking for is just a simple JavaScript object:
var gameData = {
//May need this later
init : function () {
},
storageAvailable : function () {
if (typeof (Storage) !== "undefined") {
return true;
}
else {
return false;
}
}
}
If you want to use private variables create a revealing module pattern style wrapper. This is basically what jAndy suggested:
var gameData = (function() {
var private = 'private variable';
return {
//May need this later
init : function () {
},
storageAvailable : function () {
if (typeof (Storage) !== "undefined") {
return true;
} else {
return false;
}
}
}
})();

Which logger architecture in javascript? [duplicate]

Is there a way to make any function output a console.log statement when it's called by registering a global hook somewhere (that is, without modifying the actual function itself) or via some other means?

Here's a way to augment all functions in the global namespace with the function of your choice:
function augment(withFn) {
var name, fn;
for (name in window) {
fn = window[name];
if (typeof fn === 'function') {
window[name] = (function(name, fn) {
var args = arguments;
return function() {
withFn.apply(this, args);
return fn.apply(this, arguments);
}
})(name, fn);
}
}
}
augment(function(name, fn) {
console.log("calling " + name);
});
One down side is that no functions created after calling augment will have the additional behavior.

As to me, this looks like the most elegant solution:
(function() {
var call = Function.prototype.call;
Function.prototype.call = function() {
console.log(this, arguments); // Here you can do whatever actions you want
return call.apply(this, arguments);
};
}());

Proxy Method to log Function calls
There is a new way using Proxy to achieve this functionality in JS.
assume that we want to have a console.log whenever a function of a specific class is called:
class TestClass {
a() {
this.aa = 1;
}
b() {
this.bb = 1;
}
}
const foo = new TestClass()
foo.a() // nothing get logged
we can replace our class instantiation with a Proxy that overrides each property of this class. so:
class TestClass {
a() {
this.aa = 1;
}
b() {
this.bb = 1;
}
}
const logger = className => {
return new Proxy(new className(), {
get: function(target, name, receiver) {
if (!target.hasOwnProperty(name)) {
if (typeof target[name] === "function") {
console.log(
"Calling Method : ",
name,
"|| on : ",
target.constructor.name
);
}
return new Proxy(target[name], this);
}
return Reflect.get(target, name, receiver);
}
});
};
const instance = logger(TestClass)
instance.a() // output: "Calling Method : a || on : TestClass"
check that this actually works in Codepen
Remember that using Proxy gives you a lot more functionality than to just logging console names.
Also this method works in Node.js too.

If you want more targeted logging, the following code will log function calls for a particular object. You can even modify Object prototypes so that all new instances get logging too. I used Object.getOwnPropertyNames instead of for...in, so it works with ECMAScript 6 classes, which don't have enumerable methods.
function inject(obj, beforeFn) {
for (let propName of Object.getOwnPropertyNames(obj)) {
let prop = obj[propName];
if (Object.prototype.toString.call(prop) === '[object Function]') {
obj[propName] = (function(fnName) {
return function() {
beforeFn.call(this, fnName, arguments);
return prop.apply(this, arguments);
}
})(propName);
}
}
}
function logFnCall(name, args) {
let s = name + '(';
for (let i = 0; i < args.length; i++) {
if (i > 0)
s += ', ';
s += String(args[i]);
}
s += ')';
console.log(s);
}
inject(Foo.prototype, logFnCall);

Here's some Javascript which replaces adds console.log to every function in Javascript; Play with it on Regex101:
$re = "/function (.+)\\(.*\\)\\s*\\{/m";
$str = "function example(){}";
$subst = "$& console.log(\"$1()\");";
$result = preg_replace($re, $subst, $str);
It's a 'quick and dirty hack' but I find it useful for debugging. If you have a lot of functions, beware because this will add a lot of code. Also, the RegEx is simple and might not work for more complex function names/declaration.

You can actually attach your own function to console.log for everything that loads.
console.log = function(msg) {
// Add whatever you want here
alert(msg);
}

How to dynamically set a function/object name in Javascript as it is displayed in Chrome

This is something which has been bugging me with the Google Chrome debugger and I was wondering if there was a way to solve it.
I'm working on a large Javascript application, using a lot of object oriented JS (using the Joose framework), and when I debug my code, all my classes are given a non-sensical initial display value. To see what I mean, try this in the Chrome console:
var F = function () {};
var myObj = new F();
console.log(myObj);
The output should be a single line which you can expand to see all the properties of myObj, but the first thing you see is just ▶ F.
My issue is that because of my OO framework, every single object instantiated gets the same 'name'. The code which it looks is responsible for this is like so:
getMutableCopy : function (object) {
var f = function () {};
f.prototype = object;
return new f();
}
Which means that in the debugger, the initial view is always ▶ f.
Now, I really don't want to be changing anything about how Joose instantiates objects (getMutableCopy...?), but if there was something I could add to this so that I could provide my own name, that would be great.
Some things that I've looked at, but couldn't get anywhere with:
> function foo {}
> foo.name
"foo"
> foo.name = "bar"
"bar"
> foo.name
"foo" // <-- looks like it is read only

Object.defineProperty(fn, "name", { value: "New Name" });
Will do the trick and is the most performant solution. No eval either.

I've been playing around with this for the last 3 hours and finally got it at least somewhat elegant using new Function as suggested on other threads:
/**
* JavaScript Rename Function
* #author Nate Ferrero
* #license Public Domain
* #date Apr 5th, 2014
*/
var renameFunction = function (name, fn) {
return (new Function("return function (call) { return function " + name +
" () { return call(this, arguments) }; };")())(Function.apply.bind(fn));
};
/**
* Test Code
*/
var cls = renameFunction('Book', function (title) {
this.title = title;
});
new cls('One Flew to Kill a Mockingbird');
If you run the above code, you should see the following output to your console:
Book {title: "One Flew to Kill a Mockingbird"}

Combine usage of computed property name to dynamically name a property, and inferred function naming to give our anonymous function that computed property name:
const name = "aDynamicName"
const tmp = {
[name]: function(){
return 42
}
}
const myFunction= tmp[name]
console.log(myFunction) //=> [Function: aDynamicName]
console.log(myFunction.name) //=> 'aDynamicName'
One could use whatever they want for 'name' here, to create a function with whatever name they want.
If this isn't clear, let's break down the two pieces of this technique separately:
Computed Property Names
const name = "myProperty"
const o = {
[name]: 42
}
console.log(o) //=> { myProperty: 42 }
We can see that the property name assigned on o was myProperty, by way of computed property naming. The []'s here cause JS to lookup the value inside the bracket, and to use that for the property name.
Inferred Function Naming
const o = {
myFunction: function(){ return 42 }
}
console.log(o.myFunction) //=> [Function: myFunction]
console.log(o.myFunction.name) //=> 'myFunction'
Here we use inferred function naming. The language looks at the name of wherever the function is being assigned to, & gives the function that inferred name.
We can combine these two techniques, as shown in the beginning. We create an anonymous function, which gets it's name via inferred function naming, from a computed property name, which is the dynamic name we wanted to create. Then we have to extract the newly created function from the object it is embedded inside of.
Example Using Stack Trace
Naming a supplied anonymous function
// Check the error stack trace to see the given name
function runAnonFnWithName(newName, fn) {
const hack = { [newName]: fn };
hack[newName]();
}
runAnonFnWithName("MyNewFunctionName", () => {
throw new Error("Fire!");
});

Although it is ugly, you could cheat via eval():
function copy(parent, name){
name = typeof name==='undefined'?'Foobar':name;
var f = eval('function '+name+'(){};'+name);
f.prototype = parent;
return new f();
}
var parent = {a:50};
var child = copy(parent, 'MyName');
console.log(child); // Shows 'MyName' in Chrome console.
Beware: You can only use names which would be valid as function names!
Addendum: To avoid evaling on every object instantiation, use a cache:
function Cache(fallback){
var cache = {};
this.get = function(id){
if (!cache.hasOwnProperty(id)){
cache[id] = fallback.apply(null, Array.prototype.slice.call(arguments, 1));
}
return cache[id];
}
}
var copy = (function(){
var cache = new Cache(createPrototypedFunction);
function createPrototypedFunction(parent, name){
var f = eval('function '+name+'(){};'+name);
f.prototype = parent;
return f;
}
return function(parent, name){
return new (cache.get(name, parent, typeof name==='undefined'?'Foobar':name));
};
})();

This won't totally solve your problem, but I would suggest overriding the toString method on the class's prototype. For instance:
my_class = function () {}
my_class.prototype.toString = function () { return 'Name of Class'; }
You'll still see the original class name if you enter an instance of my_class directly in the console (I don't think it's possible to do anything about this), but you'll get the nice name in error messages, which I find very helpful. For instance:
a = new my_class()
a.does_not_exist()
Will give the error message: "TypeError: Object Name of Class has no method 'does_not_exist'"

If you want to dynamically create a named function. You can use new Function to create your named function.
function getMutableCopy(fnName,proto) {
var f = new Function(`function ${fnName}(){}; return ${fnName}`)()
f.prototype = proto;
return new f();
}
getMutableCopy("bar",{})
// ▶ bar{}

Similar to #Piercey4 answer, but I had to set the name for the instance as well:
function generateConstructor(newName) {
function F() {
// This is important:
this.name = newName;
};
Object.defineProperty(F, 'name', {
value: newName,
writable: false
});
return F;
}
const MyFunc = generateConstructor('MyFunc');
const instance = new MyFunc();
console.log(MyFunc.name); // prints 'MyFunc'
console.log(instance.name); // prints 'MyFunc'

normally you use window[name] like
var name ="bar";
window["foo"+name] = "bam!";
foobar; // "bam!"
which would lead you to a function like:
function getmc (object, name) {
window[name] = function () {};
window[name].prototype = object;
return new window[name]();
}
but then
foo = function(){};
foobar = getmc(foo, "bar");
foobar; // ▶ window
foobar.name; // foo
x = new bar; x.name; // foo .. not even nija'ing the parameter works
and since you can't eval a return statement (eval("return new name()");), I think you're stuck

I think this is the best way to dynamically set the name of a function :
Function.prototype.setName = function (newName) {
Object.defineProperty(this,'name', {
get : function () {
return newName;
}
});
}
Now you just need to call the setName method
function foo () { }
foo.name; // returns 'foo'
foo.setName('bar');
foo.name; // returns 'bar'
foo.name = 'something else';
foo.name; // returns 'bar'
foo.setName({bar : 123});
foo.name; // returns {bar : 123}

Based on the answer of #josh, this prints in a console REPL, shows in console.log and shows in the debugger tooltip:
var fn = function() {
return 1917;
};
fn.oldToString = fn.toString;
fn.toString = function() {
return "That fine function I wrote recently: " + this.oldToString();
};
var that = fn;
console.log(that);
Inclusion of fn.oldToString() is a magic which makes it work. If I exclude it, nothing works any more.

With ECMAScript2015 (ES2015, ES6) language specification, it is possible to dynamically set a function name without the use of slow and unsafe eval function and without Object.defineProperty method which both corrupts function object and does not work in some crucial aspects anyway.
See, for example, this nameAndSelfBind function that is able to both name anonymous functions and renaming named functions, as well as binding their own bodies to themselves as this and storing references to processed functions to be used in an outer scope (JSFiddle):
(function()
{
// an optional constant to store references to all named and bound functions:
const arrayOfFormerlyAnonymousFunctions = [],
removeEventListenerAfterDelay = 3000; // an auxiliary variable for setTimeout
// this function both names argument function and makes it self-aware,
// binding it to itself; useful e.g. for event listeners which then will be able
// self-remove from within an anonymous functions they use as callbacks:
function nameAndSelfBind(functionToNameAndSelfBind,
name = 'namedAndBoundFunction', // optional
outerScopeReference) // optional
{
const functionAsObject = {
[name]()
{
return binder(...arguments);
}
},
namedAndBoundFunction = functionAsObject[name];
// if no arbitrary-naming functionality is required, then the constants above are
// not needed, and the following function should be just "var namedAndBoundFunction = ":
var binder = function()
{
return functionToNameAndSelfBind.bind(namedAndBoundFunction, ...arguments)();
}
// this optional functionality allows to assign the function to a outer scope variable
// if can not be done otherwise; useful for example for the ability to remove event
// listeners from the outer scope:
if (typeof outerScopeReference !== 'undefined')
{
if (outerScopeReference instanceof Array)
{
outerScopeReference.push(namedAndBoundFunction);
}
else
{
outerScopeReference = namedAndBoundFunction;
}
}
return namedAndBoundFunction;
}
// removeEventListener callback can not remove the listener if the callback is an anonymous
// function, but thanks to the nameAndSelfBind function it is now possible; this listener
// removes itself right after the first time being triggered:
document.addEventListener("visibilitychange", nameAndSelfBind(function(e)
{
e.target.removeEventListener('visibilitychange', this, false);
console.log('\nEvent listener 1 triggered:', e, '\nthis: ', this,
'\n\nremoveEventListener 1 was called; if "this" value was correct, "'
+ e.type + '"" event will not listened to any more');
}, undefined, arrayOfFormerlyAnonymousFunctions), false);
// to prove that deanonymized functions -- even when they have the same 'namedAndBoundFunction'
// name -- belong to different scopes and hence removing one does not mean removing another,
// a different event listener is added:
document.addEventListener("visibilitychange", nameAndSelfBind(function(e)
{
console.log('\nEvent listener 2 triggered:', e, '\nthis: ', this);
}, undefined, arrayOfFormerlyAnonymousFunctions), false);
// to check that arrayOfFormerlyAnonymousFunctions constant does keep a valid reference to
// formerly anonymous callback function of one of the event listeners, an attempt to remove
// it is made:
setTimeout(function(delay)
{
document.removeEventListener('visibilitychange',
arrayOfFormerlyAnonymousFunctions[arrayOfFormerlyAnonymousFunctions.length - 1],
false);
console.log('\nAfter ' + delay + 'ms, an event listener 2 was removed; if reference in '
+ 'arrayOfFormerlyAnonymousFunctions value was correct, the event will not '
+ 'be listened to any more', arrayOfFormerlyAnonymousFunctions);
}, removeEventListenerAfterDelay, removeEventListenerAfterDelay);
})();

I have not seen anyone mention the use of ES6 Proxies. Which in my opinion solve this problem beautifully. So here it is.
function shadow(object, secondObject) {
return new Proxy(object, {
get(target, prop, receiver) {
if (secondObject.hasOwnProperty(prop)) return secondObject[prop];
return Reflect.get(...arguments);
}
})
}
let t=function namedFunction(a,b,c){return a+b+c;}
console.log(t.name)//read only property
let f=shadow(t,{name:"addition"})
console.log(f.name)

Method delegation in Javascript/jQuery?

I have this code:
var myWidget = $('#myWidget');
and calls like this elsewhere:
myWidget.hide();
myWidget.slideToggle();
These work of course because jQuery adds these methods.
Now, let's say I'm doing some refactoring to make myWidget a proper object with its own custom methods and state:
var myWidget = (function() {
// private stuff
var actualJQueryObject = $('#myWidget');
return {
publicMethod: function() {...},
// MAGIC!
}
})()
but I want to have all the calls that expect a jQuery object, which are all around my code, to still work even though myWidget is no longer a jQuery object, because myWidget knows how to delegate these calls to actualJQueryObject.
Is this possible?

You could also extend your jQuery object, with another object that has your custom methods:
var myWidget = function() {
// private stuff
var actualJQueryObject = $('#myWidget');
var extensionMethods = {
publicMethod: function() { alert('public method!'); }
}
return $.extend(actualJQueryObject, extensionMethods);
}();
Just be careful with the name of your extension methods, to not clash with any other jQuery defined function.
You can try the above snippet here.

One option is using the original jquery object as a prototype.
function wrap(jqObject) {
function MyNewType() {
this.changeFontSize = function(a) {
this.css({fontSize : this.size});
};
}
MyNewType.prototype = jqObject;
return new MyNewType;
}
var obj = wrap($('#someitem'));
obj.size = 50; // obj.size
obj.changeFontSize(); // obj.changeFontSize
obj.hide(); // $.hide
obj.fadeIn("slow"); // $.fadeIn

I've written a plugin that might help you. It's basically a plugin for writing plugins. This dev group post explains it and has some code samples:
http://groups.google.com/group/jquery-dev/browse_thread/thread/664cb89b43ccb92c/72cf730045d4333a?hl=en&q=structure+plugin+authoring#72cf730045d4333a
And the source is here:
http://code.google.com/p/jquery-plugin-dev/source/browse/trunk/jquery.plugin.js
EDIT:
I created a function that has similar functionality to that plugin:
jQuerify = function(fn) {
function plugin() {
var instantiate = false;
// check to see if it has any prototyped methods (we only need one iteration to do this)
for (var i in construct.prototype) {
instantiate = true;
break;
}
// if there are prototyped methods, return an instance (since an instance's return value won't vary)
// otherwise just call it using apply so the return value can vary
return instantiate
? new construct(this, arguments)
: construct(this, arguments);
}
function construct(parent, args) {
// 'this' will not mimic jQuery unless given the length property
this.length = 0;
this.selector = parent.selector;
this.context = parent.context;
// mimic 'this' in jQuery, but for the plugin namespace
Array.prototype.push.apply(this, $.makeArray(parent));
// return the constructors return value
// should be 'this' if you want to chain the new namespace
return fn.apply(this, arguments);
}
// copy all static properties and methods
for (var i in fn) {
plugin[i] = fn[i];
}
// allow .fn and copy all instance properties and methods; the last part is for IE
plugin.fn = construct.prototype = plugin.prototype = fn.prototype;
return plugin;
}
This allows you to add custom objects to jQuery as a plugin while using 'this' to refer to the selected objects and also allows you to have an unlimited depth to your namespace:
function test1() {
return this;
}
test1.prototype.getHtml1 = function() {
return $(this).html();
}
function test2() {
return this;
}
test2.prototype.getHtml2 = function() {
return $(this).html();
}
function test3() {
return this;
}
test3.prototype.getHtml3 = function() {
return $(this).html();
}
jQuery.fn.test1 = jQuerify(test1);
jQuery.fn.test1.fn.test2 = jQuerify(test2);
jQuery.fn.test1.fn.test2.fn.test3 = jQuerify(test3);
jQuery(function($) {
alert($('body').test1().getHtml1());
alert($('body').test1().test2().getHtml2());
alert($('body').test1().test2().test3().getHtml3());
});