The target/srcElement
properties are sometimes buggy in Explorer 5 on Mac and unreliable in Netscape 4.
Detecting which key has been pressed is not (yet) possible in Konqueror/Safari, iCab and Opera 5 on Mac.
I prepared a little example (popup) of the various event registration models, event properties and event orders. Thus you can get a quick overview of the possibilities and restrictions of event handling.
On this page I give some example scripts for reading out event properties. There are very serious browser incompatibilities in this area.
As soon as we want to read out information about the event, we’re inundated by an immense amount of properties, most of which do not work in most browsers. See the Event compatibility tables for a quick overview or the W3C DOM Compatibility - Events page for a complete overview.
I am not going to give an alphabetical list of properties, since it wouldn’t help a bit to make you understand all this — the situation is too confusing. Instead I’ve written five scripts to ask five questions of the browser.
I answered the last question in great detail in an Evolt article.
Please note that in these scripts I’ve been very strict in my object detection. I first create cross-browser event access, then I check if the browser supports each individual property before using it.
This is the only question with a true cross-browser answer: use the type
property to read out the type:
function doSomething(e)
{
if (!e) var e = window.event;
alert(e.type);
}
W3C/Netscape says: the target
. No, says Microsoft, the srcElement
. Both properties return the HTML element the event took place on.
function doSomething(e)
{
var targ;
if (!e) var e = window.event;
if (e.target) targ = e.target;
else if (e.srcElement) targ = e.srcElement;
if (targ.nodeType == 3) // defeat Safari bug
targ = targ.parentNode;
}
The last two lines of code are especially for Safari. If an event takes place on an element that contains text, this text node, and not the element, becomes the target of the event. Therefore we check if the target's nodeType
is 3, a text node. If it is we move to its parent node, the HTML element.
Even if an event is captured or bubbles up, the target/srcElement always remains the element the event took place on.
Explorer 5 on Mac has a bug: when the event bubbles up the target/srcElement incorrectly refers to the BODY element.
Netscape 4 is very backwards in this respect. It really recognizes target
only for mouseovers and mouseouts, for most other events it’s impossible to get a good reading.
There are some more targeting properties. I discuss currentTarget on the Event order page and relatedTarget, fromElement and toElement on the Mouse events page.
This one is relatively easy. First get the code of the key (a = 65 in all browsers except Netscape 4, where a = 97). The old Netscape model uses the which
property for key detection, while more modern browsers use keyCode
.
When you’ve read out the key code, send it through the method String.fromCharCode()
to obtain the actual key value, if necessary.
function doSomething(e)
{
var code;
if (!e) var e = window.event;
if (e.keyCode) code = e.keyCode;
else if (e.which) code = e.which;
var character = String.fromCharCode(code);
alert('Character was ' + character);
}
There are some subtleties that may make the key events hard to use. For instance, the keypress
event fires as long as the user keeps the key pressed. However, in most browsers (except for Netscape 4) the keydown
event also keeps firing as long as the key is pressed. I’m not sure this is a good idea, but it’s what happens.
Detecting the key is not yet possible in Konqueror, iCab and Opera 5 on Mac. For more information about key events and which keys they can detect, see this article.
There are two properties for finding out which mouse button has been clicked: which
and button
. Please note that these properties don’t always work on a click
event. To safely detect a mouse button you have to use the mousedown
or mouseup
events.
which
is an old Netscape property. Left button gives a value of 1, middle button (mouse wheel) gives 2, right button gives 3. No problems, except its meager support (and the fact that it’s also used for key detection).
Now button
has been fouled up beyond all recognition. According to W3C its values should be:
According to Microsoft its values should be:
No doubt the Microsoft model is better than W3C’s. 0 should mean “no button pressed”, anything else is illogical.
Besides, only in the Microsoft model button values can be combined, so that 5 would mean “left and middle button”. Not even Explorer 6 actually supports this yet, but in the W3C model such a combination is theoretically impossible: you can never know whether the left button was also clicked.
In my opinion W3C has made some serious mistakes in defining button
. Fortunately only Mozilla follows the spec, all other browsers have opted for the saner solution.
Fortunately you most often want to know if the right button has been clicked. Since W3C and Microsoft happen to agree on this one and give button
a value of 2, you can still detect a right click.
function doSomething(e)
{
var rightclick;
if (!e) var e = window.event;
if (e.which) rightclick = (e.which == 3);
else if (e.button) rightclick = (e.button == 2);
alert('Rightclick: ' + rightclick); // true or false
}
Please note that, although Macs have only one mouse button, Netscape 6 gives a Ctrl–Click a button
value of 2, since Ctrl–Click also brings up the context menu. iCab doesn’t yet support mouse button properties at all and you cannot yet detect a right–click in Opera.
See the Image protection script for a practical example of right–click detection.
As to the mouse position, the situation is horrible. Although there are no less than six mouse coordinates property pairs, there is no reliable cross–browser way to find the mouse coordinates relative to the document we need.
These are the six property pairs — see also the Event compatibility tables or the W3C DOM Compatibility - Events page.
I explained the problem, W3C’s vagueness and the use of pageX/Y
and clientX/Y
in my slightly outdated Evolt article.
The screenX
and screenY
properties are the only ones that are completely cross–browser compatible. They give the mouse position relative to the entire computer screen of the user. Unfortunately this information is completely useless: you never need to know the mouse position relative to the screen — well, maybe only if you want to place another window at the mouse position.
The other three property pairs are unimportant. See the W3C DOM Compatibility - Events page for a description.
This is the correct script for detecting the mouse coordinates:
function doSomething(e)
{
var posx = 0;
var posy = 0;
if (!e) var e = window.event;
if (e.pageX || e.pageY)
{
posx = e.pageX;
posy = e.pageY;
}
else if (e.clientX || e.clientY)
{
posx = e.clientX + document.body.scrollLeft;
posy = e.clientY + document.body.scrollTop;
}
// posx and posy contain the mouse position relative to the document
// Do something with this information
}
If you wish to go through all event pages in order, you should now continue with the Event order page.