///Essential Javascript — A Javascript Tutorial

Essential Javascript — A Javascript Tutorial

Essential Javascript — A Javascript Tutorial

Javascript is an interpreted language with a C like syntax. While many people
brush the language off as nothing more than a browser scripting language, it
actually supports many advanced concepts such as object-oriented-programing,
recursion, lambda, and closures. It’s a very approachable language for the
beginner that quickly scales to be as powerful a tool as your skills allow.

This reference will cover the basic language constructs. This is not a
beginner’s guide to programming. This article focuses on bringing people who
already know another programming language up to speed on Javascript methodology. Additionally,
this is not an exhaustive language definition, it is a broad overview that will occasionally
focus in on some more advanced concepts. It’s here to get you started, other articles will focus
on making you an expert.

Getting Started

To dive into Javascript all you need is a simple text-editor and a browser. In windows, you can
use notepad under your accessories and Linux and mac users have a similar editor.
Simply create a blank HTML page as such…

<title>Learning Javascript</title>
<p>Hello World!

Save the file then in your browser type in the file name you just created to see
the results. Javascript is interpreted so any changes you make to this file
will show up instantly in the browser the moment you hit the reload button.

In-Line Javascript

To define a Javascript block in your web page, simply use the following block of HTML.

<script type='text/javascript'>
// Your script goes here.

You can place these script blocks anywhere on the page that you wish, there
are some rules and conventions however. If you are generating dynamic content
as the page loads you will want the script blocks to appear where you want
their output to be. For instance, if I wanted to say "Hello World!" I would
want my script block to appear in the <body> area of my web page and not in the <head> section.

Unless your scripts are generating output as the page loads, good
practice says that you should place your scripts at the very bottom of your HTML. The reason for this
is that each time the browser encounters a <script> tag it has to pause,
compile the script, execute the script, then continue on generating the page.
This takes time so if you can get away with it, make sure the browser hits your
scripts at the end of the page instead of the start.

External Javascript

External Javascript is where things get interesting. Any time you have a block
of code which you will want to use on several different web pages you should place
that block in an external Javascript file. The clock on the upper right-hand
corner of this page is a good example. The clock appears on almost every page
on this site and so it is included in my "common.js" file. Every web-page on
the site will load this file and so the clock is available to all of my web-pages.

There’s nothing fancy about an external Javascript file. All it is, is a text file
where you’ve put all your Javascript. Basically everything that would ordinarily go
between the <script> tags can go in your external file. Note
that between was stressed, you can not have the <script> </script> tags
themselves in your external file or you will get errors.

The biggest advantage to having an external Javascript file is that once the
file has been loaded, the script will hang around the browser’s cache which
means if the Javascript is loaded on one page then it’s almost a sure thing that
the next page on the site the user visits will be able to load the file from
the browser’s cache instead of having to reload it over the Internet (This is
an incredibly fast and speedy process).

Including an external file is basically the same as doing an in-line script,
the only difference is that you specify a filename, and there’s no actual
code between <script> and </script>…

<script type='text/javascript' src='common.js'></script>

When the browser encounters this block it will load common.js, evaluate it,
and execute it. Like in-line scripts above you can place this block anywhere
you need the script to be and like in-line scripts you should place these as
close to the bottom of the web-page as you can get away with.

The only difference between in-line Javascript blocks and external
Javascript blocks is that an external Javascript block will pause to load
the external file. If you discount that one thing, there’s no procedural
difference between the two!

Javascript is case sensitive

It should also be noted, before we begin, that Javascript is extremely
case sensitive so if you’re trying to code along with any examples make
sure lowercase is lowercase and uppercase is uppercase. For the most part
Javascript is also a camel-cased language. That is, if you’re trying to express more
than one word you will eliminate the spaces, leave the first letter uncapitalized
and capitalize the first letter of each word. Thus "get element by id" becomes "getElementByID".

By contrast, HTML itself is NOT case sensitive.

Output (writeln)

One of the most important things to do when learning a new language is to
master basic input and output which is why hello world has become almost a
cliche in programming textbooks. For Javascript you need three hello worlds
because there are three ways to communicate with the user, each increasingly
more useful than the last.

The first method is to use the document.writeln(string) command. This can
be used while the page is being constructed. After the page has finished loading
a new document.writeln(string) command will delete the page in most browsers,
so use this only while the page is loading. Here’s how a simple web-page will look…

<script type='text/javascript'>

.writeln('Hello World!');    


As the page is loading, Javascript will encounter this script and it will
output "Hello World!" exactly where the script block appears on the page.

The problem with writeln is that if you use this method after the page has loaded
the browser will destroy the page and start constructing a new one. As you can
see by clicking on this example which
will execute a document.writeln after the page has finished loading.

For the most part, document.writeln is useful only when teaching yourself the
language. Dynamic content during page load is better served by the server-side
scripting languages. That said, document.writeln is very useful in pre-processing
forms before they’re sent to the server — you can basically create a new web-page on the
fly without the need to contact the server.

Output (alert)

The second method is to use a browser alert box. While these are incredibly
useful for debugging (and learning the language), they are a horrible way to
communicate with the user. Alert boxes will stop your scripts from running
until the user clicks the OK button, and it has all the charm and grace
of all those pop-up windows everyone spent so many years trying to get rid of!

<script type='text/javascript'>

('Hello World!');


You can see this alert in action by clicking here!

Output (getElementById)

The last method is the most powerful and the most complex (but don’t worry, it’s really easy!).

Everything on a web page resides in a box. A paragraph (<P>) is a box. When you
mark something as bold you create a little box around that text that will
contain bold text. You can give each and every box in HTML a unique identifier
(an ID), and Javascript can find boxes you have labeled and let you manipulate them.
Well enough verbiage, check out the code!


<div id='feedback'></div>

<script type='text/javascript'>

.getElementById('feedback').innerHTML='Hello World!';


The page is a little bigger now but it’s a lot more powerful and scalable
than the other two. Here we defined a division <div> and named it "feedback".
That HTML has a name now, it is unique and that means we can use Javascript to
find that block, and modify it. We do exactly this in the script below the division!
The left part of the statement says on this web page (document) find a block we’ve
named "feedback" ( getElementById(‘feedback’) ), and change its HTML (innerHTML) to
be ‘Hello World!’.

We can change the contents of ‘feedback‘ at any time, even after the page
has finished loading (which document.writeln can’t do), and without annoying the
user with a bunch of pop-up alert boxes (which alert can’t do!).

It should be mentioned that innerHTML is not a published standard. The
standards provide ways to do exactly what we did in our example above. That
mentioned, innerHTML is supported by every major Browser and in addition
innerHTML works faster, and is easier to use and maintain. It’s, therefore,
not surprising that the vast majority of web pages use innerHTML over
the official standards.

While we used "Hello World!" as our first example, its important to note that, with
the exception of <script> and <style>, you can use full-blown HTML. Which
means instead of just Hello World we could do something like this…


<div id='feedback'></div>

<script type='text/javascript'>

.getElementById('feedback').innerHTML='<P><font color=red>Hello World!</font>';


innerHTML will process your string and basically redraw the web page with the
new content. This is a VERY powerful and easy to use concept. It means you
can basically take an empty HTML element (which our feedback division is) and
suddenly expand it out with as much HTML content as you’d like.

Input (One Click To Rule Them All)

Input, of course, is a little more complicated. For now we’ll just reduce
it to a bare click of the mouse.

If everything in HTML is a box and every box can be given a name,
then every box can be given an event as well and one of those events
we can look for is "onClick". Lets revisit our last example…


<div id='feedback' onClick='goodbye()'>Users without Javascript see this.</div>

<script type='text/javascript'>

.getElementById('feedback').innerHTML='Hello World!';

function goodbye() {
.getElementById('feedback').innerHTML='Goodbye World!';


Here we did two things to the example, first we added an "onClick" event to
our feedback division which tells it to execute a function called goodbye() when
the user clicks on the division. A function is nothing more than a named block
of code. In this example goodbye does the exact same thing as our first hello world
example, it’s just named and inserts ‘Goodbye World!’ instead of ‘Hello World!’. See for yourself!
Click the text below…

Hello World!

Another new concept in this example is that we provided some text for people
without Javascript to see. As the page loads it will place "Users without Javascript will
see this." in the division. If the browser has Javascript, and it’s enabled then that
text will be immediately overwritten by the first line in the script which looks up
the division and inserts "Hello World!", overwriting our initial message. This happens so
fast that the process is invisible to the user, they see only the result, not the process. The
goodbye() function is not executed until it’s explicitly called and
that only happens when the user clicks on the division.

While Javascript is nearly universal there are people who surf with it deliberately
turned off and the search bots (googlebot, yahoo’s slurp, etc) also don’t process
your Javascript, so you may want to make allowances for what people and machines are-not

Input (User Input)

Clicks are powerful and easy and you can add an onClick event to pretty much
any HTML element, but sometimes you need to be able to ask for input from the user
and process it. For that you’ll need a basic form element and a button…

<input id='userInput' size=60> <button onClick='userSubmit()'>Submit</button><BR>
<P><div id='result'></div>

Here we create an input field and give it a name of userInput. Then we create a
HTML button with an onClick event that will call the function userSubmit(). These are
all standard HTML form elements but they’re not bound by a <form> tag since we’re
not going to be submitting this information to a server. Instead, when the user clicks the
submit button, the onClick event will call the userSubmit() function…

<script type='text/javascript'>
function userSubmit() {
var UI=document.getElementById('userInput').value;
.getElementById('result').innerHTML='You typed: '+UI;

Here we create a variable called UI which looks up the input field userInput. This
lookup is exactly the same as when we looked up our feedback division in the previous example.
Since the input field has data, we ask for its value and place that value in our UI variable.
The next line looks up the result division and puts our output there. In this case the
output will be "You Typed: " followed by whatever the user had typed into the input field.

Give it a try and see how it works…

We don’t actually need to have a submit button. If you’d like to process the
user input as the user types then simply attach an onKeyup event to the input
field as such…

<input id='userInput' onKeyUp="userSubmit()" size=60><BR>
<P><div id='result'></div>

There’s no need to modify the userSubmit() function. Now whenever a user
presses a key while the userInput box has the focus, for each keypress, userSubmit()
will be called, the value of the input box retrieved, and the result division
updated. And here’s the example…

It’s interesting to note that as you’re typing you’re actually creating HTML and
inserting it into this document which means that html tags you typed will be appropriately
processed. Try typing "The quick<p>brown fox." and you’ll see the line break
correctly inserted in the output.

Javascript is an Event Driven Language

As you can tell from the input examples, Javascript is an event driven language
which means your scripts react to events you set up. Your code isn’t running all
the time, it simply waits until an event starts something up! Going into all the
Javascript events is beyond the scope of this document but here’s a short-list of
common events to get you started.

Event Description
onAbort An image failed to load.
onBeforeUnload The user is navigating away from a page.
onBlur A form field lost the focus (User moved to another field)
onChange The contents of a field has changed.
onClick User clicked on this item.
onDblClick User double-clicked on this item.
onError An error occurred while loading an image.
onFocus User just moved into this form element.
onKeyDown A key was pressed.
onKeyPress A key was pressed OR released.
onKeyUp A key was released.
onLoad This object (iframe, image, script) finished loading.
onMouseDown A mouse button was pressed.
onMouseMove The mouse moved.
onMouseOut A mouse moved off of this element.
onMouseOver The mouse moved over this element.
onMouseUp The mouse button was released.
onReset A form reset button was pressed.
onResize The window or frame was resized.
onSelect Text has been selected.
onSubmit A form’s Submit button has been pressed.
onUnload The user is navigating away from a page.

These events can be attached to most any HTML tag or form element. Of them all onClick will probably
be what you end up using most often.


Javascript supports two types of comments. Double-slashes (//) tell javascript
to ignore everything to the end of the line. You will see them used most often to
describe what is happening on a particular line.

var x=5; // Everything from the // to end of line is ignored(*)
var thingamajig=123.45; // 2 times the price of a whatsit.

Block quotes begin a comment block with a slash-asterisk (/*) and Javascript will
ignore everything from the start of the comment block until it encounters an asterisk-slash (*/).
Block quotes are useful for temporally disabling large areas of code, or describing the
purpose of a function, or detailing the purpose and providing credits for the script itself.

function whirlymajig(jabberwocky) {
/* Here we take the jabberwocky and insert it in the gire-gimble,
      taking great care to observe the ipsum lorum!   For bor-rath-outgrabe!
      We really should patent this! */

return (jabberwocky*2);

You should note that while comments are useful for maintaining the code, they are a liability
itself in Javascript since they will be transmitted along with the code to each and every page load, which
can create substantial bandwidth penalties and increase the load time of your page for users.

This doesn’t mean you shouldn’t comment your code, just that once your code is "finished" you should
make a backup copy with the comments, then strip out all the comments in the file which is actually
sent to the user. You can automate this process with a minimizing application which you can find
here and
an on-line javascript version here.

The result of minimizing your Javascript is a tiny, compact file which is a fraction of the size
of the original which will save you bandwidth and increase page-load time for your visitors. However
the result is also a very unmaintainable mess which is why you should keep a separate, unminimized (and
heavily commented) version of the original file.

(*) Of special consideration, you should note that the browser itself is ALWAYS
looking for a </script> tag to mark the end of your Javascript and if it
finds that tag, intact, in-one-piece, be it in a string or a comment, it is going
to stop processing Javascript at that point and restart-processing HTML.

 var x=5;
/* The browser will break the Javascript when it sees this </script> tag.
      Everything from tag forward is now being processed as HTML!  This is a bad thing!
      To avoid this you need to avoid using this tag anywhere in your Javascript, and if
      you must have it, you should break the string out like this... */



Javascript is not a
strongly typed language which means you rarely have to concern yourself with the type of
data a variable is storing, only what the variable is storing and in Javascript, variables
can store anything, even functions.

var thisIsAString = 'This is a string';
var alsoAString = '25';
var isANumber = 25;
var isEqual = (alsoAString==isANumber); // This is true, they are both 25.
var isEqual = (alsoAString===isANumber); // False one is a number, the other a string.
var concat=alsoAString + isANumber; // concat is now 2525
var addition=isANumber + isANumber; // addition is now 50

var alsoANumber=3.05; // is equal to 3.05 (usually).
var floatError=0.06+0.01; // is equal to 0.06999999999999999
var anExponent=1.23e+3; // is equal to 1230
var hexadecimal = 0xff; // is equal to 255.
var octal = 0377; // is equal to 255.

var isTrue = true; // This is a boolean, it can be true or false.
var isFalse= false; // This is a boolean, it can be true or false

var isArray = [0, 'one', 2, 3, '4', 5]; // This is an array.
var four = isArray[4]; // assign a single array element to a variable.
// in this case four = '4'

var isObject = { 'color': 'blue', // This is a Javascript object
'dog': 'bark',
'array': [0,1,2,3,4,5],
'myfunc': function () { alert('do something!'); }
var dog = isObject.dog; // dog now stores the string 'bark';
.myfunc(); // creates an alert box with the value "do something!"

var someFunction = function() {
return "I am a function!";
var alsoAFunction = someFunction; //No () so alsoAFunction becomes a function
var result = alsoAFunction(); // alsoAFunction is executed here because ()
// executes the function so result stores the
// return value of the function which is
// "I am a function!"                  

A variable may not be a Javascript reserved word or begin with a number or
any symbol other than $ or _. In Internet explorer you should also avoid variable
names with the same name as html elements you have named. For instance…

 var someDiv = document.getElementByID('someDiv');

…will cause problems in Internet Explorer because the variable name
and the division name are identical.

In recent years a convention has formed around
the use of the $ symbol as various libraries like Prototype and JQuery use it
to look up a named HTML element. For most purposes if you see $(‘something’) in
Javascript you should read that as being document.getElementById(‘something’).
This is not standard Javascript, in order for $(‘something’) to work, you
need to be using a Javascript framework which will define $ as doing something (Like JQuery,
Prototype, etc).

The use of a leading underscore (_) is generally useful to indicate a global
variable or a variable that has been set outside the current scope.

A final consideration on variables is that functions themselves can be defined
like, and act like variables. Once a function has been defined it can be passed
to other functions as an argument (A process knows as lambda), or assigned to other
variables just like a string, array or any other Javascript object. Generally if you
use a function without trailing parenthesis (), the function is treated like a
variable and can be passed and assigned. Trailing parenthesis INVOKE the function,
executing it and passing back the return value (if any).

Please note that this is a very broad summary overview of Javascript’s data types.
For more information please see the other articles in this series (listed at the top of the page)
which go into exhaustive detail on each Javascript type.

Variable Scope

Variables in Javascript have FUNCTION scope. That is, all variables are global
unless they are explicitly defined inside a function and even then child-functions
have access to their parent’s variables. If a function defines a new variable WITHOUT
using the var keyword, that variable will be global in scope.

var global = 'this is global';

function scopeFunction() {
= 'This is also global!';
var notGlobal = 'This is private to scopeFunction!';
function subFunction() {
(notGlobal); // We can still access notGlobal in this child function.
= 'No var keyword so this is global!';
var isPrivate = 'This is private to subFunction!';
(stillGlobal); // This is an error since we haven't executed subfunction
(); // execute subfunction
(stillGlobal); // This will output 'No var keyword so this is global!'
(isPrivate); // This generate an error since isPrivate is private to
// subfunction().
(global); // outputs: 'this is global'

(global); // outputs: 'this is global'
(alsoGlobal); // generates an error since we haven't run scopeFunction yet.
(alsoGlobal); // outputs: 'This is also global!';
(notGlobal); // generates an error.

The concept that a variable will continue to exist, and can be referenced after
the function that created it has ceased executing is known as CLOSURE. In the above example,
stillGlobal, and alsoGlobal can be considered closures because they persist after
the function that creates them has ceased to operate. You can do some
pretty fancy stuff
with it later on, but it’s not terribly hard to understand once you
associate it with creating a global scoped variable inside a function.

Special Keywords

Javascript has a few pre-defined variables with special meaning.

NaN — Not a Number (Generated when an arithmetic operation returns an invalid result).
NaN is a weird construct. For one, it is NEVER equal to itself so you can’t simply
check to see if 3/’dog’ == ‘NaN’. You must use the construct isNaN(3/dog) to determine
if the operation failed. In boolean
operations NaN evaluates to false, however 0 also evaluates to false so use isNaN. Since
NaN is never equal to itself you can use this simple trick as well:

if (result != result) { alert('Not a Number!'); }

Infinity is a keyword which is returned when an arithmetic operation overflows
Javascript’s precision which is in the order of 300 digits. You can find the
exact minimum and maximum range for your Javascript implementation using Number.MAX_VALUE
and Number.MIN_VALUE.

null is a reserved word that means "empty". When used in boolean operation it evaluates to

Javascript supports true and false as boolean values.

If a variable hasn’t been declared or assigned yet (an argument to a function which
never received a value, an object property that hasn’t been assigned a value) then
that variable will be given a special undefined value. In boolean operations
undefined evaluates as false. Here’s an example…

function doAlert(sayThis) {
if (sayThis===undefined) { // Check to see if sayThis was passed.
='default value'; // It wasn't so give it a default value
} // End check
(sayThis); // Toss up the alert.

Arithmetic Operators

There’s nothing particularly exotic about the way Javascript does arithmetic.

Operator Description
+ Addition (also string concatention)
Subtration (also unary minus)
* Multiplication
/ Division
% Remainder of division (or modulus)
++ pre or post increment
pre or post decrement

++ and — are C style operators their position around the variable they are
operating on is important. If the operator appears BEFORE the variable they
will be evaluated immediately, if they appear AFTER the variable they will be
evaluated only after the entire line has been resolved. For instance…

 var x = 5;
var y = x++; // y=5, x=6

var x = 5;
var y = ++x; // y=6, x=6

In the first example y is assigned the value of x (5) and then x is incremented
by one because the ++ operator appears AFTER x. In the second example, x is incremented
by one first because ++ appears BEFORE the x, so y is given the value 6.

How this works between C and PhP and Javascript can be somewhat different.
There was a most excellent discussion about this on reddit
if you’d like to delve into those differences.

Additional shorthand operators exist when working on the same variable. For instance…

   x = x + 5; // is the same as...
+= 5;

Logical and Comparison Operators

Javascript supports equality checking (==) and identity checking (===).
Equality checks for equality regardless of type. Therefore 25 and ’25’ will evaluate as
true. Identity checking checks not only for equality but type equality as well so 25 and
’25’ will evaluate as false because, while both are 25, one is a string and the other a number.
Note that a single equal sign is an assignment statement! x=5 will assign 5 to x, while x==5 will
see if x is equal to 5, and x===5 will check to see if x is identical to 5.

In addition to == and ===, you can check for not equal (!=) and not identical (!==).

Operator Description
= Assignment x=5; // assigns 5 to x
== Equality, is x==5?
=== Identity, is x 5 and a number as well?
!= Not equal, is x unequal to 5?
!== Not identical, is x unequal to the Number 5?
! Not, if not(false) is true.
|| OR, is (x==5) OR (y==5)
&& And, is (x==5) AND (y==5)
< Less than. is x less than 5?
<= Less than or equal. is x less than or equal to 5?
> Greater than. is x greater than 5?
>= Greater than or qual. is x greater than or equal to 5?

JavaScript Conditionals: IF

The if statement lets you execute a block of code if some test is passed.

var x=5;
if (x==5) {
('x is equal to 5!');

You can also use an else clause to execute code if the test fails.

var x=5;
if (x==5) {
('x is equal to 5!');
} else {
('x is not equal to 5!');

An elseif statement also exists which allows for better formating of
long conditional tests.

var x=5;
if (x==1) {
('x is equal to 1!');
} else if (x==2) {
('x is equal to 2!');
} else if (x==5) {
('x is equal to 5!');
} else {
('x isn't 1, 2 or 5!');

JavaScript Conditionals: SWITCH

If you’re going to be doing a large number of tests, it makes sense to use
a switch statement instead of nested ifs. Switches in javascript are quite
powerful, allowing evaluations on both the switch and the case.

var x=5;
switch (x) {
case 1: alert('x is equal to 1!'; break;
case 2: alert('x is equal to 2!'; break;
case 5: alert('x is equal to 5!'; break;
default: alert("x isn't 1, 2 or 5!");

Note that if you omit the break statement that ALL of the code
to the end of the switch statement will be executed. So if x is actually equal to
5 and there is no break statement, an alert for "x is equal to 5" will appear as well
as an alert for "x isn’t 1,2, or 5!".

Sometimes it makes more sense to do the evaluation in the case statement itself.
In this case you’d use true, false, or an expression which evaluates to true or false
in the switch statement.

var x=5;
switch (true) {
case (x==1): alert('x is equal to 1!'; break;
case (x==2): alert('x is equal to 2!'; break;
case (x==5): alert('x is equal to 5!'; break;
default: alert("x isn't 1, 2 or 5!");

JavaScript Conditionals: Shorthand Assignment

Javascript supports more advanced constructs. Often you will see code like the

function doAddition(firstVar, secondVar) {
var first = firstVar  || 5;
var second= secondVar || 10;
return first+second;


Here Javascript uses a logical OR (||) to determine if the passed variables actually
have a value. In the example we call doAddition with a value of 12 but we neglect to
pass a second argument. When we create the first variable firstVar is a
non-falsey value (IE it’s actually defined) so Javascript assigns firstVar to
first. secondVar was never passed a value so it is undefined which
evaluates to false so here the variable second will be assigned a
default value of 10.

You should note that zero evaluates as false so if you pass zero as either firstVar
or secondVar the default values will be assigned and NOT zero. In our example above
it is impossible for first or second to be assigned a zero.

In psuedo code…

var someVariable = (assign if this is truthy) || (assign this if first test evaluates false)

JavaScript Conditionals: Ternary Operators

Ternary operators are a shorthand if/else block who’s syntax can be a bit
confusing when you’re dealing with OPC (Other People’s Code). The syntax boils down
to this.

var userName = 'Bob';
var hello = (userName=='Bob') ? 'Hello Bob!' : 'Hello Not Bob!'; 

In this example the statement to be evaluated is (userName==’Bob’).
The question marks ends the statement and begins the conditionals. If UserName
is, indeed, Bob then the first block ‘Hello Bob!’ will be returned and assigned
to our hello variable. If userName isn’t Bob then the second
block (‘Hello Not Bob!’) is returned and assigned to our hello variable.

In psudeo code…

var someVariable = (condition to test) ? (condition true) : (condition false);

The question mark (?) and colon (:) tend to get lost in complex
expressions as you can see in this example taken from wikipedia (but which will also
work in Javascript if the various variables are assigned…)

for (i = 0; i < MAX_PATTERNS; i++)
[i].ShowWindow(m_data.fOn[i] ? SW_SHOW : SW_HIDE);

So while quick and efficient, they do tend to reduce the maintainability/readability of the code.

JavaScript Loops: FOR

The for loop follows basic C syntax, consisting of an initialization,
an evaluation, and an increment.

for (var i=0; (i<5); i++) {
.writeln('I is equal to '+i+'<br>');
// outputs:
// I is equal to 0
// I is equal to 1
// I is equal to 2
// I is equal to 3
// I is equal to 4

This is actually an extreme simplification of what a for statement can do. On the
other end of the spectrum, consider this shuffle prototype which will randomly shuffle
the contents of an array. Here, everything is defined, and executed within the context
of the for statement itself, needing no additional block to handle the code.

Array.prototype.shuffle = function (){ 
for(var rnd, tmp, i=this.length; i; rnd=parseInt(Math.random()*i), tmp=this[--i], this[i]=this[rnd], this[rnd]=tmp);

JavaScript Loops: FOR/IN

Javascript has a variant of the for loop when dealing with Javascript objects.

Consider the following object…

var myObject = { 'animal' : 'dog',
'growls' : true,
'hasFleas': true,
'loyal' : true } 

We can loop through these values with the following construct.

var myObject = { 'animal' : 'dog',
'growls' : true,
'hasFleas': true,
'loyal' : true }
for (var property in myObject) {
.writeln(property + ' contains ' + myObject[property]+'<br>');
// Outputs:
// animal contains dog
// growls contains true
// hasFleas contains true
// loyal contains true

What this essentially does is assign the property name to the variable property.
We can then access myObject through an associative array style syntax.
For instance the first itteration of the loop assigns animal to property
and myObject["animal"] will return dog.

There is a big caveat here in that properties and methods added by prototyping
will also show up in these types of loops. Therefore it’s best to always check to
make sure you are dealing with data and not a function as such…

for (var property in myObject) {
if (typeof(myObject[property]) != 'function') {
.writeln(property + ' contains ' + myObject[property]+'<br>');

The typeof check to screen out functions will ensure that your for/in loops will
extract only data and not methods that may be added by popular javascript libraries like

JavaScript Loops: WHILE

while loops in Javascript also follow basic C syntax and are easy to understand and use.

The while loop will continue to execute until its test condition evaluates to
false or the loop encounters a break statement.

var x = 1;
while (x<5) {
= x +1;

var x = 1;
while (true) {
= x + 1;
if (x>=5) {

Sometimes it makes more sense to evaluate the test condition at the end of
the loop instead of the beginning. So for this Javascript supports a do/while

var x=1;
do {
= x + 1;
} while (x < 5);

Bringing It All Together

So now you know how to do basic input and output, how to do conditional branching
and how to do loops. That’s pretty much everything you need to know about any programming
language to get started! So lets wrap everything we’ve done so far into one simple web

First we’ll define our web page.

<title>My First Javascript</title>
      Hello World!

Now to this we will add an input line, a drop down box, and a submit button.
Notice we give an ID to the form elements, and have the submit button call a function
when it’s clicked. A function which we will write later. We’ve also added some
descriptive text.

<title>My First Javascript</title>
      Hello World!
<p>Say what? <input id="sayThis" size=40>
<P>How many times? <select id='howMany'>
<option value=1>1</option>
<option value=5 selected>5</option>
<option value=10>10</option>
<option value=20>20</option>
<p><button onClick='doLoop()'>Do It!</button>

Now we need to add a division where the output will occurr. We’ll add this right
below the form elements, specifically the "Do It!" button.

<title>My First Javascript</title>
      Hello World!
<p>Say what? <input id="sayThis" size=40>
<P>How many times? <select id='howMany'>
<option value=1>1</option>
<option value=5 selected>5</option>
<option value=10>10</option>
<option value=20>20</option>
<p><button onClick='doLoop()'>Do It!</button>
<p><div id="results"></div>                        

The script for this will be simple enough. When the button is clicked the
doLoop() function will be called. doLoop() will lookup the value of
the sayThis input field, then repeat that howMany times.

Here’s the script itself, separate from the HTML…

 function doLoop() {
var sayWhat = document.getElementById('sayThis').value;
var maxLoop = document.getElementById('howMany').value;
var str = ''; // where we'll store our output temporarily.
for (var i=1; (i<=maxLoop); i++) {

And the entire application, all together.

<title>My First Javascript</title>
      Hello World!
<p>Say what? <input id="sayThis" size=40>
<P>How many times? <select id='howMany'>
<option value=1>1</option>
<option value=5 selected>5</option>
<option value=10>10</option>
<option value=20>20</option>
<p><button onClick='doLoop()'>Do It!</button>
<p><div id="results"></div>
<script type='text/html'>
function doLoop() {
var sayWhat = document.getElementById('sayThis').value;
var maxLoop = document.getElementById('howMany').value;
var str = ''; // where we'll store our output temporarily.
for (var i=1; (i<=maxLoop); i++) {

And finally a working example.

Hello World!

Say what?

How many times?


The above example is what’s technically considered Dynamic HTML (or DHTML) because
the contents of the page dynamically change based on user interaction after the page
has finished loading. Going into great detail on DHTML is beyond the scope of this
article but there is one small, trivial jump that can really change how you see HTML
and Javascript.

This page has pulled an HTML element with the use of document.getElementById. Once
we’ve had this assigned to a variable we’ve been able to manipulate the contents of that
element with innerHTML, and in the case of forms find the value of the element with the
value property.

There is also a style property which lets you access, and change, the CSS styles for
that element. There is a nice CSS reference at ILoveJackDaniels.
Most all of the properties you can access are listed on the left and right margin of the
cheat sheet (the interior/brown areas aren’t applicable). The only real trick is that
where you see a dash in the CSS name, for Javascript you need to remove the dash and
capitalize the next letter. So the CSS style background-color becomes
backgroundColor in Javascript.

Here’s an example which will change the background color of a division when
you click on it’s contents.

<div id="example" onClick="colorize()">Click on this text to change the background color</div>
<script type='text/javascript'>
function colorize() {
var element = document.getElementById("example");

And the example.

Click on this text to change the background color

Here we create a division with a name of example which will call a function
called colorize() when that division is clicked. colorize() will
lookup the example division and assign it to the Javascript variable element.
Next, we assign a color of #800 (burgundy) to the element’s style.backgroundColor property.

This actually made the text hard to read so in the next example we’ll add another line
to change the color of the text to white.

<div id="example" onClick="colorize()">Click on this text to change the background color</div>
<script type='text/javascript'>
function colorize() {
var element = document.getElementById("example");

And the new example…

Click on this text to change the background color

And lets go ahead and center the text as well.

<div id="example" onClick="colorize()">Click on this text to change the background color</div>
<script type='text/javascript'>
function colorize() {
var element = document.getElementById("example");
Click on this text to change the background color

The manipulation of an HTML element’s CSS styles is incredibly powerful! You can
hide, or display elements with the style.display property (style.display=none=invisible, style.display=block=visible)
You can let the elements float over the page with style.position (absolute/relative/fixed), you
can change their position on the page by setting style.left and style.top. The possibilities
are literally endless.


As stated at the beginning of this article, this document is intended to bring
a programmer up to speed in Javascript, and is not an exhaustive review of the language.
This article is a part of a larger series of reference articles (listed at the top of
the page). As you master the fundamentals introduced on this page, it’s recommended
that you visit the other reference articles to master the details of the language.

Javascript is a surprisingly deep and well considered language (with a few notable and
notorious exceptions). It is simple enough for beginners to programming to understand
and scales to become as advanced as you need it to be. While it may be frustrating to
work around the various idiosyncrasies of all the browser models and versions, javascript
itself is a joy to work in, and when developing web-applications, the least and most
unobtrusive of all the problems you will have to surmount.

2010-05-25T22:49:11+00:00 December 2nd, 2007|JavaScript|0 Comments

About the Author:


Leave A Comment