Listening to Nodes
Last updated
Last updated
It's been established that we can easily manipulate "nodes" in the DOM, as well as create and remove elements at will. We know that we can, so how exactly do we interact with nodes on the page? Well, we listen for them!
Demonstrate listening to events on a DOM node with addEventListener()
Demonstrate triggering events listeners on DOM nodes
Demonstrate preventing the default behavior for DOM nodes
Explain the difference between bubbling and capturing events
Demonstrate stopping propagated behaviors with stopPropagation()
If you are using the Learn IDE available in your browser, you will automatically clone down the files you need when you click 'Open IDE', but in order to view index.html
, you will need to use httpserver
to serve the HTML page temporarily. In the terminal, type httpserver
and press enter. You will see that Your server is running at ...
followed by a string of numbers and dots. This string is a temporary IP address that is hosting your index.html
file. Copy this string of numbers, open a new tab and paste the string in to the URL bar.
If you are using a standalone text editor such as Sublime or Atom, before we get started, follow to manually fork and clone the lesson repository on GitHub. In your forked and cloned copy, you'll find the index.html
file, which you can then manually open up in the browser. (For instructions on opening HTML files in the browser from the Learn IDE, see .)
addEventListener()
Adding an event listener to a DOM node doesn't require much logic — we just call addEventListener()
on the node. addEventListener()
takes two arguments: the name of the event, and a function to handle the event.
Start by adding a listener for click
events to the main#main
element in index.html
. Once you've opened index.html
in the browser, enter the following in the browser's JS console:
If you click on the main
element (you can click its text, "My ID is 'main'!"), you should see an alert: 'I was clicked!'
. How does this work?
The second argument is a function that accepts the event as its argument. The event has a number of useful properties on it — keypress, keydown, and keyup events, for example, will have a which
property that tells us which key was pressed. In the code snippet example below we add an event listener to the input
element.
If you test out this code, you'll see that pressing "enter" will print 13
in console; pressing "a" prints 65
; etc.
Refresh the page to start with a clean slate. Let's test out preventing the default behavior of the input by keeping it from receiving the "g" character. Enter the following in your console:
Now try to type "g" in the input--not working, right?
Every DOM event
comes with a preventDefault
property. preventDefault
is a function that, when called, will prevent the, well, default event from taking place. It provides us an opportunity to intercept and tweak user interactions (usually in more helpful ways than preventing them from typing "g").
Another, related event property is called stopPropagation
. Like preventDefault
, stopPropagation
is a function that, when called, interrupts the event's normal behavior. In this case, it stops the event from triggering other nodes in the DOM that might be listening for the same event. Yes, one action can trigger multiple events!
DOM events propagate by bubbling (starting at the target node and moving up the DOM tree to the root) and capturing (starting from the target node's parent elements and propagating down the tree until it reaches the target) — by default, events nowadays all bubble. We can verify this behavior by attaching listeners to those nested div
s in index.html
. Enter the following in your console:
Now click on the div
containing "5". You should see
In the console, you can see the event starts at div
5, and then it bubbles up to the topmost node. Along the way, it triggers any other nodes that are listening for the event -- in this case, 'click'
.
Try clicking on a node that's not so deeply nested -- you should still see the event bubble up, starting at the node that you clicked and hitting every node up the tree until it reaches the top.
What about capturing? In order to capture, we need to set the third argument to addEventListener
to true
. Let's try it out.
Now click on div
5. You should see
As you can see, the event propagates from the top of the page towards the target node, triggering event handlers as appropriate along the way.
Notice that the target node is the last node to capture the event, whereas it's the first node to bubble the event up. This is the most important takeaway.
stopPropagation()
Now that we've explained a bit about the dangers and behavior of bubbling and capturing, you can see how events propagate through the DOM. Much of the time, since we're listening for very specific events, this doesn't matter: our events can propagate up or down, and they'll only trigger the event handler(s) that we want them to trigger. Sometimes, as with these div
s, we have a fairly generic event that we want to constrain to its target. That's where stopPropagation
comes in.
Let's rewrite the bubbling example to stop propagation so that only one event is triggered (be sure to reload the page before entering this code!):
Now try clicking on any node — you should only see one log statement!
We covered a lot in this lesson. Feel free to edit index.html
, to write code directly in the document (just put it between tags), and to play around with different events. It's important to practice so you can get the hang of it! You can practice adding an event listener, using different event triggers, and intercepting user interactions with e.preventDefault()
and e.stopPropagation().
The first argument, 'click'
is the name of the event that's being listened for. Click events probably make up a majority of events listened to, but other common events are change
, 'keydown'
, 'keyup'
, 'load'
, 'mouseover'
, 'mouseout'
— the list goes on. You can find a reasonably comprehensive list on .
NOTE: Don't worry if bubbling and capturing seems a bit esoteric. The different event behaviors are the results of the browser wars of the 90s, but most of the time it's safe just to stick to the default (which, for the record, is bubbling). You can read more about bubbling and capture on and
Clone :