Closures: Scope Chain Variables - Not sure how variables link up - javascript

Really new to Javascript. This code is taken from MDN.
// global scope
var e = 10;
function sum(a){
return function sum2(b){
return function sum3(c){
// outer functions scope
return function sum4(d){
// local scope
return a + b + c + d + e;
}
}
}
}
var s = sum(1);
var s1 = s(2);
var s2 = s1(3);
var s3 = s2(4);
console.log(s3) //log 20
When I try to input different variable names (EX below) they don't seem to work and I don't understand how everything links up together to spit out the answer 20.
// global scope
var e = 10;
function sum(a){
return function sum2(b){
return function sum3(c){
// outer functions scope
return function sum4(d){
// local scope
return a + b + c + d + e;
}
}
}
}
var w = sum(1);
var x = s(2);
var y = s1(3);
var z = s2(4);
console.log(s3) //log 20
When I change it to this it also does not work. The console tells me that sa is not defined
// global scope
var e = 10;
function sm(a){
return function sa(b){
return function sb(c){
// outer functions scope
return function sc(d){
// local scope
return a + b + c + d + e;
}
}
}
}
var s = sm(1);
var s1 = sa(2);
var s2 = sb(3);
var s3 = sc(4);
console.log(sc) //log 20
I can keep throwing out more examples that don't work. Someone, please help me understand how the first example works.

function sm(a){
return function sa(b){
return function sb(c){
// outer functions scope
return function sc(d){
// local scope
return a + b + c + d + e;
}
}
}
}
The function sm is taking one argument and is returning a function which takes one argument. The function sm returns is not named sb. Try to think of it as the return value of sm .
So, when you do
var s = sm(1);
The returned function is stored in the variable s
Now if you want to run the second function (sa inside sm) you need to invoke s.
var s1 = s(2);
The returned function (sb) is stored inside varibale s1.
Similarly,
var s2 = s1(3);
var s3 = s2(4);
console.log(s3); // 20

Related

variable visibility in javascript

I am struggling to understand how variables are referenced and stay alive in Javascript. In the following I have two types of object, a Note and an IntervalBuilder which takes a Note and creates
a second Note.
function Note() {
this.key = 1 + Math.floor( Math.random() * 13); // from 1 to 13 inclusive
this.setKey = function setKey(i) { key = i; };
this.getKey = function getKey() { return this.key; } ; // {return key} is a ReferenceError!!
}
function IntervalBuilder() {
this.setPerfectFifth = function setPerfectFifth(root) {
this.fifth = new Note();
console.log("this.fifth: " + this.fifth);
this.y = root.key;
console.log("root.key: " + root.key );
console.log("y: " + this.y );
this.fifth.setKey( this.y + 4 );
return this.fifth;
};
}
With the above I can now do this:
var x = new Note();
var ib = new IntervalBuilder();
ib.setPerfectFifth(x);
However, the instance ib now has a member named fifth! What I was hoping for was that I could assign the return value (a Note) from setPerfectFifth to a variable and let fifth vanish. How is that done?
Many thanks for any help, I find lots of this very confusing.
Gerard
Since you titled your quesion variable visibility in javascript what is basically going on is: In this.fifth = new Note(); the keyword this references the instance (the ib of var ib = new ...). So you attach your newly created Note to the instance. In JavaScript, as long as a variable can be reached starting with the global Object (window, when you think of a graph), it won't get garbage-collected away.
What you want is: var fith = new Note(), which will create a local variable which will get freed as soon as the function execution ends. Clearly, every usage of this.fifth then has to be replaced by just fith.
I do not know exactly what you want to achieve, but I think you want the following code structure:
// ==============================
// NOTE "CLASS"
// ==============================
var Note = (function () {
// Constructor
function Note() {
this._key = 1 + Math.floor( Math.random() * 13);
}
// Getter
Note.prototype.getKey = function () {
return this._key;
};
// Setter
Note.prototype.setKey = function (i) {
this._key = i;
};
return Note;
})();
// ==============================
// INTERVAL BUILDER "CLASS"
// ==============================
var IntervalBuilder = (function () {
// Constructor
function IntervalBuilder() {}
// Private members
var fifth = null,
y = 0;
// Setter
IntervalBuilder.prototype.setPerfectFifth = function (root) {
fifth = new Note();
y = root.getKey();
fifth.setKey(y + 4);
return fifth;
};
return IntervalBuilder;
})();
// ==============================
// CLIENT CODE
// ==============================
var x = new Note(),
ib = new IntervalBuilder();
ib.setPerfectFifth(x);

JavaScript Closure - trying to understand following code

Coming from Java background trying to make sense out of the following code.
From:
https://medium.freecodecamp.com/lets-learn-javascript-closures-66feb44f6a44#.cbk6c4e9g
For function bar(c), which line passes argument c into bar(c) as I don't see it here.
Thanks.
var x = 10;
function foo(a) {
var b = 20;
function bar(c) {
var d = 30;
return boop(x + a + b + c + d);
}
function boop(e) {
return e * -1;
}
return bar;
}
var moar = foo(5); // Closure
/*
The function below executes the function bar which was returned
when we executed the function foo in the line above. The function bar
invokes boop, at which point bar gets suspended and boop gets push
onto the top of the call stack (see the screenshot below)
*/
moar(15);
When your first statement of function call var moar = foo(5) is executed
moar variable will be function bar(c){var d=30;return boop(x+a+b+c+d);
check the snippet for understanding
var x = 10;
function foo(a) {
var b = 20;
function bar(c) {
var d = 30;
return boop(x + a + b + c + d);
}
return bar;
}
var moor=foo(10);
console.log(moor);
2.After this statement moar(15),your are actually passing 15 to the bar method
It will execute bar method with c as 15. Now this function would return boop(80) which would just be -80
var x = 10;
function foo(a) {
var b = 20;
function bar(c) {
var d = 30;
return boop(x + a + b + c + d);
}
function boop(e) {
return e * -1;
}
return bar;
}
var moar = foo(5)
console.log(moar(15));
Hope it helps
moar(15) passes 15 into bar which gets copied into parameter c.
The // Closure comment is misleading because it is more useful to think of the closure being configured at the point of declaration of a function.
The reality is that the pointer to the outer lexical environment is configured when a function object is instantiated, and then this pointer is then copied to the execution context associated with any invocations of said function object.

Object function scope

factory(n) returns objects with functions.
func1 function definition creates its own scope, and x inside this function references x = n + ''.
But func2 is a reference and the scope is wrong.
Is there a way to return an object from create so its functions were references (not separate definitions)?
Actually, I'm fine with func1 approach while function definition footprint is small. If it is a complex function it would be better not to clone this function into every object comming from factory(n). inner_func may not use this, it is simple function. Also I want to avoid new and this.
var factory = (function(){
var x = '!';
return function create(n){
var x = n + '';
return {
func1: function(y){return inner_func(x, y); },
/* vs */
func2: inner_func_api
}
}
function inner_func_api(y){ return inner_func(x, y); }
function inner_func(a, b){ return a + b; }
}());
var f1 = factory(2);
var f2 = factory(3);
var f1_func1 = f1.func1(4);
var f2_func1 = f2.func1(5);
var f1_func2 = f1.func2(4);
var f2_func2 = f2.func2(5);
console.log(f1_func1, f2_func1); //24 35
console.log(f1_func2, f2_func2); //!4 !5
You could define that function separately from the object initializer on the return statement:
var factory = (function(){
var x = '!';
return function create(n){
var x = n + '';
function func1(y) {
return inner_func(x, y);
}
return {
func1: func1,
/* vs */
func2: inner_func_api
}
}
function inner_func_api(y){ return inner_func(x, y); }
function inner_func(a, b){ return a + b; }
}());
However, it makes no practical difference, and it doesn't matter how big or complicated that function is. Function instances do take up space, but the code for the function is constant (immutable) and doesn't need to be part of every Function object created from the same piece of source code.

Can I pass parameter like reference

I need function change to change variables and return back to Tst1. I expect to get in console:
5
aaa
but have unchanged ones:
6
bbb
My functions:
function change ( aa,bb )
{
aa=5;
bb="aaa";
}
function Tst1()
{
aa=6;
bb="bbb";
change(aa,bb);
console.log (aa);
console.log (bb);
}
One way is to move change() into the function test(). Then it shares the same variables as the calling scope.
'use strict';
function test() {
function change() {
aa = 6;
bb = 76;
}
var aa = 5,
bb = 6;
change();
document.write(aa + " " + bb);
}
test();
JavaScript is like java in that primitives are never passed by reference but objects are always passed by reference. You need to wrap your data in an object and pass that instead:
function change (aa, bb)
{
aa.value = 5;
bb.value = "aaa";
}
function Tst1()
{
aa = { value: 6 };
bb = { value: "bbb" };
change(aa, bb);
console.log (aa.value); // outputs 5
console.log (bb.value); // outputs aaa
}
or you can play with global variable, but it is not a good practice.
var aa,bb;
function change(){
aa=6;
bb=76;
}
function test(){
aa = 5;
bb = 6;
change();
console.log(aa + " " + bb);
}
test();
Short answer: NO, you can't pass primitive parameters by reference in JS.
One alternative solution to the presented here is to return the result values as array of items:
function change ( aa,bb )
{
aa=5;
bb="aaa";
return [aa, bb];
}
function Tst1()
{
aa=6;
bb="bbb";
result = change(aa,bb);
aa = result[0];
bb = result[1];
document.writeln(aa);
document.writeln(bb);
}
Tst1();

can a function reference itself in javascript?

Suppose we define a function that simply increments its input by some stored value dd:
var obj={}
obj.dd=1
obj.f=function(x){
return x+this.dd
}
Alternatively you could create a closure for dd as follows but this would create a static increment as opposed to one that could be altered later:
var dd=1
var f=function(x){
return x+dd
}
We could alternatively store dd in the function itself:
var obj={}
obj.f=function(x){
return x+this.f.dd
}
obj.f.dd=1
I am curious as to whether it is possible for a function to retrieve a variable attached to itself without going through a parent object, something like a self keyword that would refer to the function itself and would allow the following:
var f=function(x){
return x+self.dd
}
f.dd=1
I know it is unnecessary to do such a thing but I think it would be cool if you could.
You can give function literals a name:
var f = function me(x) {
return x + me.dd;
};
f.dd = 1;
This doesn’t work properly in older versions of IE/JScript, though, as me and f don’t reference the same object. The (deprecated and not usable in strict mode) alternative is arguments.callee:
var f = function(x) {
return x + arguments.callee.dd;
};
f.dd = 1;
Also, your note about the closure isn’t quite right; it can be altered later, even through another function:
var dd = 1;
var f = function(x) {
return x + dd;
};
var setdd = function(_dd) {
dd = _dd;
};
A function is an object. If you reference the var holding the function:
var f = function (x) {
return x + f.dd
};
f.dd = 1;
alert(f(1));
result: 2
If the function is named, you can do the same:
function foo(x) {
return x + foo.dd;
}
foo.dd = 1;
alert(foo(1));
result: 2

Categories