That document is what I wrote during following that amazing course by Anthony P.Alicea. You can find the course and the first 3.5 hours. This course is kinda old and isn't using ES6 stuff but it is the most interesting piece of JS knowledge I found to that date.
These notes won't replace watching the course, they more of a personal reminder, for me and whoever watch the course.
- javascript is synchronous and single-threaded.
- objects and functions are closely related.
- javascript doesn't care if arguments are not passed to function, they will just be setted to
undefined
- everything is an object or a primitive
The execution context is created in 2 phases:
- creation phase: Global object (variable environment),
this
and outer environment is setup in memory. Also setup the memory space for variables and functions (that step is called hoisting). Anargument
variable is also created for functions. - execution phase: We here have everything setup and it will run the code line by line.
When a variable is called, javascript will does more than check the current execution context. An execution context have a reference to its outer environment. So in a execution stack, the outer environment is not necessarily the execution context below itself. It will instead depends on the lexical environment. So if my function is at the global level, the outer environment will be global
.
function a(){
console.log(example)
}
function b(){
var example = 1
a()
}
var example = 2
b()
// => will return "2" because outer environment of a() is global even thought it's called in b()
function b(){
function a(){
console.log(example)
}
var example = 1
a()
}
var example = 2
b()
// => will return '1' because outer environment of a() is b() as a() sits in b()
The fact of going down through these outer environment is called the scope chain.
Another point of interest when speaking about scope chain is the difference between normal functions and arrow functions.
document.addEventListener('click', function() { console.log(this) });
// => will return the document object as functions are inheriting the execution scope
document.addEventListener('click', () => console.log(this));
// => will return the window object as arrow functions are inheriting the declaration scope
A function is a special type of object. You can assign it a primitive, an object or a function as property. In a function there is also two important property: name and code. A function without a name is called an anonymous function.
When passing a primitive to a variable, javascript will create a copy of that value, stored in its own memory.
a = 3
b = a
b
// => returns 3
a = 2
b
// => still returns 3
When passing an object (or a function as a function is an object), javascript will just reference to the object.
a = {name: "Tom"}
b = a
b
// => returns {name: "Tom"}
a.name = "Margaux"
b
// => returns {name: "Margaux"}
this
is setup during the creation phase of the execution context.
console.log(this)
// => returns the window object
function test(){
console.log(this)
}
test()
// => returns the window object
var test = function(){
console.log(this)
}
test()
// => returns the window object
var object = {
name: "Tom",
test: function(){
console.log(this)
}
}
object.test()
// => returns the object
var object = {
name: "Tom",
test: function(){
function test(){
console.log(this)
}
test()
}
}
object.test()
// => returns the window object !!!
// for this case we usually need a self = this
Since javascript have first class functions, we are open to a lot of functional programming approaches. Let's have an example:
var arr = [1,2,3]
function mapForEach(arr, fn){
var newArr = []
for(var i=0; i < arr.length; i++){
newArr.push(fn(arr[i]))
}
return newArr
}
arr2 = mapForEach(arr, function(item){
return item*10
})
console.log(arr2)
// => return [10,20,30]
undefined
is an internal value of javascript. It will be returned when a variable is initialized(via var
or something). Different from not defined
error that happen when the variable isn't in memory at all.
Never set yourself a value to undefined
yourself. That would be valid but it's make the debugging flawy.
The fact of changing a value to another type. The coercion is used when the equality operator is called. The coercion can act weird sometimes and that's why we will use triple equal ===
for a most "defined" behavior.
In a function javascript creates for us a variable called arguments
. It's an array-like variable that we can use to access parameters like arguments[0]
.
An spread argument can be used for specify an undefined numbers of arguments.
function test(arg1,arg2,...arg3){
console.log(arg3)
}
test("Tom", "Quinonero")
// => will return []
test("Tom", "Quinonero", "Berlin", "JS Expert")
// => will return ["Berlin", "JS Expert"]
function test(){
console.log("I'm a statement")
}
var test = function(){
console.log("I'm an expression")
}
When calling an expression just after it been defined.
var test = function(){
console.log("test")
}()
(function(){
console.log("test")
})()
A feature of javascript. It's a combination of a function and it's lexical environment. The function will "remember" it's outer environment.
function buildFunction(){
var arr = []
for (var i = 0; i < 3; i++){
arr.push(
function() {
console.log(i)
}
)
}
return arr
}
var fs = buildFunction()
fs[0]()
fs[1]()
fs[2]()
// => Return 3 3 3
// Because for doesn't create it's own execution environment
// The outer environment will be the buildFunction and i is 3 in there
All functions have access to a call()
, apply()
and bind()
function.
When creating an expression function, I can change what this
is referring to using bind:
var person = {
firstname: "Tom"
}
var test = function(salutation){
console.log(salutation + ' ' + this.firstname)
}
test()
// => Will return undefined
var testBinded = test.bind(person)
testBinded()
// => Will return "tom"
I can also keep the same function using call:
test.call(person, "hi")
// => Will return "hi tom"
Apply can be used, the difference is that the arguments have to be wrapped into an array:
test.apply(person, ["hi"])
// => Will return "hi tom"
Cool patterns using these functions:
// Borowing a method
var post = {
title: "How to be the best js person in the world",
author: "Tom",
getPostName: function(){
return this.title + " by: " + this.author
}
}
var post2 = {
title: "How to be a bad js person",
author: "Jack"
}
post.getPostName.bind(post2)()
// => Will return "How to be a bad js person by: Jack"
// function currying
function multiply(a,b){
return a*b
}
// we set the function to be a copy of multiply with 2 as the first argument
var multiplyByTwo = multiply.bind(this, 2)
multiplyByTwo(4)
// => will return "8"
In javascript the inheritance is prototypal.
Each object have a proto
property. If a property is not found on the object itself, it will look for it in the prototype of it. And that goes on. That is called the prototype chain.
All object inherits by default from the base object. That object is the base of every object.
All function inherits from the function base object. This object define property as arguments, bind, call, apply ... That's why every function can access these.
Same with arrays.
We can build a object with a function constructor using new
operator:
function Post(){
this.title = "How to construct js objects"
this.author = "Tom"
}
var post = new Post()
console.log(post)
// => Will return Post {title: "How to construct js objects", author: "Tom"}author: "Tom"title: "How to construct js objects"__proto__: Object
If the function doesn't specify a return
, it's acting like we returned this
.
A downside is that if we forgot to use the new
keyword, it will not set the properties at the object level.
A convention for helping avoiding this issue is to start functions constructors name with a capital letter.
var person = {
firstname: "default",
lastname: "default",
greet: function(){
return "Hi " + this.firstname + " " + this.lastname
}
}
var tom = Object.create(person)
tom.greet()
// => will return "Hi default default"
tom.firstname = "Tom"
tom.greet()
// => will return "Hi Tom default"
In javascript, since ES6, you can use class
for for creating an object.
That works like a syntaxic sugar for constructing objects.
It has an extends
possibility for using a prototype.
- asynchronous: Means more than one at a time. Since javascript is synchronous, the event queue is used to handle asynchronous stuff. The asynchronous part will not happen in the javascript engine, it will mostly happens on the render engine of the browser or the HTTP engine.
- block: A portion of code generally delimited by curly brackets
{}
. - callback function: A function you give to another function to be executed when the other function is finished.
- coercion: Converting a value to another type.
- dynamic typing: Type is not defined explicitly, it's figured out when the code is running. It's the opposite of static typing.
- Event queue: When an event happens, it's placed on the event queue. When the execution is finished (read the execution stack is empty), javascript will check periodically the event queue and check if something should be run in that case (eg: if an event is listened). If an event happens before the execution is done, javascript will only process this event when the execution is done.
- execution contexts: A wrapper to help manage the code that is running. The default execution context is the global execution context. In a browser, the global object is the
window
object.this
at the global level is thewindow
object. execution context != scope - execution stack: A stack of execution contexts. When the script on the page is done, the stack is empty.
- expression: A unit of code that results in a value.
- first class function: Anything you can do with other types (assign to var, create them on the fly) can be done with functions. Functions in javascript are first-class functions.
- function currying: Creating a copy of a function but with some preset parameters
- function constructor: A function that is used to construct an object.
- hoisting: Phenomenon that happens during the creation phase of the execution context. It's setting up space for variables and functions into the memory. During this process, all variables are initially set to
undefined
and functions are entirely wrote to the memory. So you can call a function that is defined later in the code event thought it's should not be done. - inheritance: One object gets access to the properties and methods of another object. Javascript uses prototypal inheritance.
- invocation: Running a function. Invocation will create a new execution context and put it at the top of the execution stack. When the invocation is done, the execution context is removed form the execution stack.
- lexical environment: Where something sits physically in the code you write
- literal syntax: A short syntax eg:
[]
fornew Array()
. - mutation: Changing something
- name/value pair: A name which maps to a unique value. A name can only have one value in a given execution context.
- namespace: A container for variables and functions.
- object: A collection of name/value pair. An object can contains a primitive (called a property), another object (called also a property) or a function (called a method).
- operator: An operator is a special function in javascript that use a different syntax. Its usually have two parameters and one result. The double equal comparator
==
is an operator. The plus+
is an operator. The.
is also an operator ! - operator associativity: What order operator functions get called first when the operators have the same precedence. Left-to-right associativity or Right-to-left associativity.
- operator precedence: Define which operator function is called first. The higher precedence win. For example
*
has a higher precedence than+
. - outer environment: A reference to the environment where a function or a variable sits, it's lexical environment. By default (read "when declared at the root"), the outer environment is
global
. - primitives types: The "default" types of javascript. Primitive type mean a single value, not an object. The six primitives types are:
undefined
,null
,boolean
,number
,string
andsymbol
(that is a ES6 type) - reflection: An object can look at itself, listing and changing its properties and methods.
- scope: Where a variable is available in your code and if it's the truly same variable.
- scope chain: The fact that a function or variable call not defined in the current execution context goes down to the outer environment for finding it.
- single threaded: Mean that one command happen at a time.
- synchronous: Mean that the code is happening one line at the time. Similar to single threaded.
- syntax parsers: Program that reads the code and determinate what it does and if its syntax is valid.
- variable environment: Where the variable is. Since every function has it's own execution context a variable with the same name will have no interactions between them.
- whitespace: Invisible character that create space in the code (spaces, tabs, returns).