Categories:

### Determining how much the user has scrolled the page using JavaScript or jQuery

Created: Jan 22nd, 2016

One of the projects I recently worked on required me to not just detect how much the user has scrolled the page in terms of an absolute value (ie: pixels), but also, as a percentage of the total scrollable length of the document. With the later, we can perform tasks based on a more robust calculation of how much the document has been scrolled, such as at 50% or 100% (very bottom). In this quick tutorial, we'll see how to do all of the above, using plain JavaScript, and for the jQuery addict, using that library as well. Lets get started!

### Detecting the number of pixels the user has scrolled

To detect how much the user has scrolled the page vertically in terms of pixels from the very top, in JavaScript, we would probe either window.pageYOffset, or in older versions of IE, one of several variants of document.body.scrollTop, whichever property is supported:

var scrollTop = window.pageYOffset || (document.documentElement || document.body.parentNode || document.body).scrollTop

Using jQuery instead, the equivalent would be:

var scrollTop = \$(window).scrollTop()
Try it

### Detecting how far the user has scrolled the page percentage wise

To determine how far the user has scrolled as a percentage of the total document, we need to first find out what the total scrollable length of the document is. On a page with no scrollbars for example, that length is 0, while for longer pages, that length is basically represented by the white part inside the browser scrollbar UI:

So first thing's first before. To get the total scrollable area of a document, we need to retrieve the following two measurements of the page:

1. The height of the browser window
2. The height of the entire document

By subtracting 1 from 2, we get the total scrollable area of the document.

#### - Getting the height of the browser window

To get the height of the browser window in JavaScript, use window.innerHeight, and for IE8 or below, a variant of document.body.clientHeight:

var winheight = window.innerHeight || (document.documentElement || document.body).clientHeight

In jQuery, the code is more succinct as usual:

var winheight = \$(window).height()

#### - Getting the height of the entire document

For retrieving the height of the entire document in JavaScript, parts of which may be hidden behind the browser frame, we need to examine multiple properties and select the largest value out of the bunch. This is due to inconsistencies in some browsers when it comes to interpreting the height of a document with no vertical scrollbar- what we want in this case is simply the window's height, though some browsers (ie: Mobile Chrome in Android) will return the height of the content contained inside the document instead. To get the desired document height across the board, we can use James Padolsey's function for that:

function getDocHeight() {
var D = document;
return Math.max(
D.body.scrollHeight, D.documentElement.scrollHeight,
D.body.offsetHeight, D.documentElement.offsetHeight,
D.body.clientHeight, D.documentElement.clientHeight
)
}

var docheight = getDocHeight()

And in jQuery, the task is disposed of much easily:

var docheight = \$(document).height()

#### - Putting it all together

We have all the pieces now for determining how much the user has scrolled in terms of percentage. First up is the JavaScript version- we'll attach all our code to window's onscroll event to see the output change as the user scrolls:

JavaScript version:

function amountscrolled(){
var winheight= window.innerHeight || (document.documentElement || document.body).clientHeight
var docheight = getDocHeight()
var scrollTop = window.pageYOffset || (document.documentElement || document.body.parentNode || document.body).scrollTop
var trackLength = docheight - winheight
var pctScrolled = Math.floor(scrollTop/trackLength * 100) // gets percentage scrolled (ie: 80 or NaN if tracklength == 0)
console.log(pctScrolled + '% scrolled')
}

amountscrolled()
}, false)

Move your eyes down to the trackLength variable, which gets the total available scroll length of the document. The variable will contain 0 if the page is NOT scrollable. The pctScrolled variable then divides the scrollTop variable (amount the use has scrolled) with trackLength to derive how much the user has scrolled percentage wise.

Here is the jQuery equivalent:

jQuery version:

function amountscrolled(){
var winheight = \$(window).height()
var docheight = \$(document).height()
var scrollTop = \$(window).scrollTop()
var trackLength = docheight - winheight
var pctScrolled = Math.floor(scrollTop/trackLength * 100) // gets percentage scrolled (ie: 80 NaN if tracklength == 0)
console.log(pctScrolled + '% scrolled')
}

\$(window).on("scroll", function(){
amountscrolled()
})
Try it

### Applying some optimizations inside onscroll

Running code inside window's onscroll event can be an expensive affair, as the code is often invoked many times per second as the user obliviously scrolls. Looking at our code for getting how many percent of the page the user has scrolled above, here's a couple of optimizations we can perform:

• Cache the winheight, docheight, and trackLength variables' values, and only refresh them when the page has resized, instead of each time the page is scrolled.

• Throttle the code inside window onscroll so they are executed only once per a certain period (ie: 300 milliseconds), using JavaScript's setTimeout() method.

With that said, here is the JavaScript version of our optimized code to get how much the user has scrolled percent wise as he/she scrolls:

var winheight, docheight, trackLength, throttlescroll

function getmeasurements(){
winheight= window.innerHeight || (document.documentElement || document.body).clientHeight
docheight = getDocHeight()
trackLength = docheight - winheight
}

function amountscrolled(){
var scrollTop = window.pageYOffset || (document.documentElement || document.body.parentNode || document.body).scrollTop
var pctScrolled = Math.floor(scrollTop/trackLength * 100) // gets percentage scrolled (ie: 80 or NaN if tracklength == 0)
console.log(pctScrolled + '% scrolled')
}

getmeasurements()

getmeasurements()
}, false)

clearTimeout(throttlescroll)
throttlescroll = setTimeout(function(){ // throttle code inside scroll to once every 50 milliseconds
amountscrolled()
}, 50)
}, false)

### Conclusion

Detecting how much the user has scrolled forms the basis for many interesting effects, such as parallax or opt-in boxes that only pop up when you've reached the bottom of the page. We now have the essential parts to implement these effects ourselves.

End of Tutorial

Partners