Re: What's wrong with keyCode and String.fromCharCode
- From: RobG <rgqld@xxxxxxxxxxxx>
- Date: Tue, 28 Feb 2006 02:00:14 GMT
Pugi! wrote:
"VK" <schools_ring@xxxxxxxxx> schreef in bericht news:1141062704.880634.86380@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Pugi! wrote:
It is example of adding an eventlistener to a textarea. The events works
alright, but when I press 'a' the alert gives me 'A', when I press '8' I get
'h', when I press '0' I get ''', when I press 'à' I get '0', ....
Has probably something to do with the fact that I don't live in the UK or
USA. But it shouldn't matter; a key is a key.
No it's not ;-) Do not mix keyboard *scancodes* with *keycode*.
Scancode is the system value assigned to each key on your keyboard.
This value is indead layout-independent. Keycode is Unicode value going
to receiver: it depends on current layout, shift state etc.
In your particular case though it may help to monitor 'okeypress'
instead of 'onkeydown' as you're doing right now.
---
Well, actually I used onkeyup. But onkeypress only made it work for IE. So I looked elswewhere in the code and keyCode is for IE, while Mozilla (and rest) prefers charCode. But according to the author keyCode works fine in Mozilla, guess he didn't debug his code hard enough.
It depends on the event you are using. Keydown fires when the key is pressed down. keypress fires when the key is released (it is almost identical to keyup);
<URL:http://www.w3.org/TR/html4/interact/scripts.html#adef-onkeypress>
Different browsers treat events differently. The Gecko DOM event object has both keyCode and charCode properties. In IE, the event object has a keyCode property but no charCode property.
In IE, the keyCode property for the keydown event is the ASCII (or Unicode) code of the key - for the 'W' key that's '87' which maps to uppercase 'W' on my keyboard. The keypress event takes into consideration any modifier keys that might have been pressed at the same time, and since the 'shift' key wasn't pressed, we get a keyCode of '119' or lowercase 'w'.
In the same situation in a Gecko browser, the keydown event will report the keyCode of '87' and a charCode of '0'. The keypress event will report a keyCode of '0' and a charCode of '119', so you need to look at the right property.
A shortcut for Gecko browsers here is to use the old 'which' property that is effectively the same as IE's keyCode property:
function showKey(e)
{
var e = e || window.event;
var kCode;
if (e.which){
kCode = e.which;
} else if (e.keyCode){
kCode = e.keyCode;
}
alert(kCode + ': ' + String.fromCharCode(kCode))
}
or more concisely:
function showKey(e)
{
var e = e || window.event;
var kCode = e.which || e.keyCode;
alert(kCode + ': ' + String.fromCharCode(kCode))
}
kCode should also have taken into consideration the keyboard layout you are using, but that might be unreliable. It is much more reliable to look at the text that was entered rather than grabbing keyCodes and trying to convert them to characters.
I think the above will be much more reliable cross-browser than the one below.
function aKeyWasPressed(e) {
if (window.event) {
var key = window.event.keyCode;
} else {
var key = e.charCode; // changed keyCode to charCode for Mozilla
}
alert('You pressed the key: ' + String.fromCharCode(key));
}
If you want to know what key was pressed, use keydown. If you want to know what character was entered, use keyup or keypress and use the input's value to determine the character that was entered (or maybe deleted).
Make sure your algorithm can cope with the entry or removal of multiple characters without any key presses.
--
Rob
.
- References:
- Prev by Date: Re: Check null or 0
- Next by Date: Re: undefined field error message
- Previous by thread: Re: What's wrong with keyCode and String.fromCharCode
- Next by thread: Slight problem with script
- Index(es):
Relevant Pages
|
