Develop Reference

Maintaining a map of which keys are down in the presence of Undo, Copy, etc

I'm building a web application with a fullscreen canvas that needs to react to continuous and discrete keyboard input.
My current approach is similar to the one in this question: JavaScript keep track of which keys are down accurately.
I maintain a set of which keys are currently pressed using keyEvent.code to differentiate keys. I query this set in the requestAnimationFrame loop of my application. I update the set in event handlers for the 'keyup' and 'keydown' events attached to window.
The Problem
When pressing and releasing Cmd+Z to undo, the 'keydown' event fires but the 'keyup' event does not (at least in the latest Firefox, Chrome, and Edge on macOS 10.15.6).
Consequently, my set contains an entry for 'KeyZ' even though the Z key on the keyboard is not being held. Similarly, Cmd+C, Cmd+V, and many other system shortcuts seem to hijack the 'keyup' event.
Why am I not getting a 'keyup' event in this case? What can I do to ensure that my set accurately reflects the state of the currently held keys on the keyboard?
What I've Tried
This seems like it could be related to event propagation, so I tried using keyEvent.stopPropagation() and keyEvent.preventDefault() in the keydown event handler.
I've tried pressing Control+Z instead, which does fire a 'keyup' event.
Problem Reproduction
const listenerTarget = window;
const useCapture = false;
const heldKeys = new Set();
listenerTarget.addEventListener('keydown', (keyEvent) => {
const code = keyEvent.code;
if (!keyEvent.repeat) {
console.log(`${code} went down`);
heldKeys.add(code);
} else {
console.log(`${code} went down (repeat)`);
}
keyEvent.stopPropagation();
keyEvent.preventDefault();
}, useCapture);
listenerTarget.addEventListener('keyup', (keyEvent) => {
const code = keyEvent.code;
// Why does this not fire if
// the 'keydown' happened in combination with
// a Meta key?
console.log(`${code} went up`);
heldKeys.delete(code);
}, useCapture);
Workaround Update (11/03/2020)
I found When CMD key is kept pressed, keyup is not triggered for any other key and http://web.archive.org/web/20160304022453/http://bitspushedaround.com/on-a-few-things-you-may-not-know-about-the-hellish-command-key-and-javascript-events/, which indicate that this is known behavior. My current workaround is to not add codes to the set of held keys if keyEvent.metaKey is set. Doing this at least ensures that the set of held keys doesn't contain phantom entries. I'll keep this question open in case someone can think of a better approach.

For me it works just fine on the lates chrome version..
Sorry.

This isn't super helpful, but there's a bunch of good info here: https://unixpapa.com/js/key.html. You've probably already seen that in other answers though.
I say not super helpful because it doesn't really answer your particular question, but it does show that there's definite weirdness between browser versions and both modifier keys and branded keys. I suspect the answer is that the branded keys like the windows key and cmd key just act weird because they tend to trigger system level events. I would just use different keys, but hopefully someone else will know a way around this other than "use different keys".
For windows people, you can produce really similar results by doing Windows Key + Z. You'll only get up event for Z and no down events.

Javascript Keyup isn't called when Command and another is pressed

This is Mac only problem; I've tried this on windows and it works fine.
I have a script that saves which keys are pressed on keydown and deletes them on keyup.
$(function(){
var keys = [];
$(document).keydown(function(event) {
keys[event.which] = true;
});
$(document).keyup(function(event) {
delete keys[event.which];
console.log(keys);
});
});
All I am doing right now is console logging whatever is left after the keyup, which should be nothing. This works as expected when you press any number of keys simultaneously.
HOWEVER, when you press command and any other key, the other key is NOT released! So console.log will show that key to be true. This will remain true until you press that key itself.
This only happens on a Mac, and it only happens when one of the keys pressed is the Command key. Here is a very simple plunker with the above example.
Any ideas?

The Mac is modifying your key whenever you press Command, and thus the normal keyup event never fires.
C = command key and K = non-command key
As you press C and K, they register normally. While they are both simultaneously pressed, the Mac captures K and modifies it. In modifying K, the Mac somehow makes K's keyup event not fire as it is supposed to. C's keyup works as expected, however.
Since K's keyup never actually fires, it won't correctly delete the matching element from keys. Later on when you press K without C, the keydown event overwrites the existing keydown in keys. And when K's keyup correctly fires, it works as expected.
In addition to all the normal keys used to input ASCII characters,
keyboards typically have many special purpose keys that do other
things. These do not necessarily generate the same events as normal
keys, and they show less consistency across browsers.
JavaScript Madness: Keyboard Events. Potentially helpful article for all key-related problems.

I can't fix this Mac issue, but here is my way of getting around it.
This answer will help you if you are trying to have keyboard-shortcut behavior, where the user presses CMD+S to do save, or something like that. This answer does not apply to people who may be building a game or something where their keyboard's keydown states need to be known at each run frame. Sorry!
In the KeyboardEvent returned by keydown, you can do the following
$(document).keydown(function(keyboardEvent) {
if (keyboardEvent.metaKey){
// on Mac, CMD is down ...or Meta key is down on pc
console.log(keyboardEvent.meta + " & command key are down")
switch (keyboardEvent.which) {
...
}
}
});
If your keyboard shortcut overlaps with the browser's, you need to make sure to cancel the propagation of the keyboard event,
keyboardEvent.preventDefault()
I hope this helps people who want keyboard shortcut functionality that is Mac compatible!

evt.which gives different output on keyup and keypress function [duplicate]

Related: JavaScript KeyCode vs CharCode
Here is some code you can try at home or in a jsfiddle:
el.addEventListener( 'keyup', function( e ) {
console.log( 'Keyup event' );
console.log( e.keyCode );
} );
el.addEventListener( 'keypress', function( e ) {
console.log( 'Keypress event' );
console.log( e.keyCode );
} );
Why is the keyCode different?
I can understand why one should use keypress only, but what I don't understand is how two key events, given the same hit key on the keyboard, give different keyCodes.
PS: I'm not worrying about legacy browsers support, I tried this in Chrome and was surprised, and couldn't find an explanation.

The events are for completely different purposes. Use keyup and keydown for identifying physical keys and keypress for identifying typed characters. The two are fundamentally different tasks with different events; don't try to mix the two. In particular, keyCode on keypress events is usually redundant and shouldn't be used (except in older IE, but see the linked document below for more on that); for printable keypresses it's usually the same as which and charCode, although there is some variation between browsers.
Jan Wolter's article on key events, already linked to in another answer, is the definitive word on this subject for me and has tables describing what each of the different properties returns for each type of key event and each browser.

There is a good article on quirksmode.org answering exactly that question. You might also want to look at Unixpapa's results.

Well, I stumbled upon one difference when i was trying to copy user's entry from one input of the form to some other part of the form , which I had locked for my for users to edit.
What i found was, that whenever a user moved to the next label using key upon completing the input, one last keyboard entry was missed in the copied entry when I used eventListener to keypress and this got resolved on using keyup.
So, in conclusion Keypress listens to the state at the instant when the key was pressed, leaving aside the result of keypress, whereas keyup listens to the system status after the key has been pressed and includes the result of the keypress.

Trying to figure out if there is a bug in jQuery or if it's something I'm doing

$(document).keydown(function (event)
{
alert(event.which);
});
For the semicolon key, ;, this gives 59 in Firefox and 186 in Chrome. However, from the jQuery reference page for the keydown event, it says
"While browsers use differing properties to store this information, jQuery normalizes the .which property so you can reliably use it to retrieve the key code. This code corresponds to a key on the keyboard, including codes for special keys such as arrows."
Am I missing something?

The which property is a "one stop shop" for which key was pressed, allowing you to ignore the differences between the keyCode and charCode properties. That is the "normalization" that jQuery provides.
The difference in the value of which comes down to a difference between the way the various browsers supply the information - so you'll have to write code to handle the different values that come back. There are a few references to this behavior online.

A quick Google search says you will simply have to test for both. This is a consistent inconsistency with Firefox.

I don't know about jQuery but I'd suggest sticking to keypress events for typing keys and only using keydown events for special keys such as arrows.

Here is the entirety of the "normalization" that jQuery does:
if ( event.which == null ) {
event.which = original.charCode != null ? original.charCode : original.keyCode;
}
Looks like it just gets keyCode if charCode doesn't exist. And charCode is only used if event.which doesn't already exist. It doesn't change the numbers around to make them consistent.

$(window).keypress(function()) does not work in IE7?

This keypress event works fine for me in Chrome and Firefox, but simply isn't being picked up at all in IE7:
$(window).keypress(function(e) {
alert('hello world');
});
Does anyone know alternatives for IE7?
Or is it an error higher up in my JavaScript that means it isn't being picked up in IE7 - in which case, how can I debug it? I have script errors turned on in IE, but nothing is popping up.

IE doesn't support key events on the window.
Place it on the document instead.
$(document).keypress(function(e) {
alert('hello world');
});

Currently I have no IE7 to test it, but I believe you have to assign the event handler to $(document) instead of $(window). Something like:
$(document).keypress(function(e) {
alert("hello world");
});

A couple things to try:
1: Wrap your event code in a document.onready event, like so:
jQuery(function()
{
$(window).keypress(function(e)....
});
This way your event is not attached until the DOM has finished loading.
2: Attach your event to the body or the document instead of the window:
jQuery(function()
{
$("body").keypress(function(e)....
$(document).keypress(function(e)....
});
(I have no reason to think this will work, it's just something I've done when events don't bubble all the way back to document or window for some reason.)

I encountered this problem too. What I discovered is that IE7 has no keyChar member in its implementation of the event object. Thus, it tries to evaluate event.keyChar, gets a null and does nothing. Instead the keyCode is ambiguous returning either the ascii keyChar or the keycode of the key that was pressed... ugh. Since I was having to trap both special keys and keys tied to characters this was my solution.
onkeydown
negate the keyCode and save in a persistent object
myObject.kSave=-event.keyCode;
WHY? because at this point you don't know if the key is a character or a special key !!!
onkeypress
The onkeypress event may not fire if the key is a special key.
For that reason my strategy uses onkeyup to handle special keys and control characters and uses onkeypress to handle characters other than control characters.
When the key is a character, event.keyChar contains the character and its value is greater than 0. However, since IE7 doesn't have an event.keyChar member, use event.keyCode as the character. You can force all control characters (e.g. enter, esc, tab, etc.) to be handled as a special key by preventing the default behavior of control characters and deferring handling to onkeyup. To signal that a character has been handled we must set the saved object to the character we found.
var ch = (event.keyChar == null) ? event.keyCode : event.keyChar;
if (ch >=32) {//ch is a character and not a control character
myObject.kSave = ch; //tells onkeyup that the key was handled
event.preventDefault(); //don't let default processing mess with your handler
switch(ch) {
//your character handler goes here
}
} else if (ch > 0) {//ch is a control character
event.preventDefault();
}
onkeyup
Add the saved value to event.keyCode. If the sum is zero, the onkeypress event did not
handle the keypress because the pressed key was a special key or a control character.
When the sum is positive, there is nothing further to do because the key was a character
and the onkeypress event handler already took care of it.
if (myObject.kSave+event.keyCode == 0) {//this is a special key or control character
event.preventDefault();
switch(event.keyCode) {
//handle special keys here
}
}
One last thing, in your event handler invoke event.preventDefault() if you don't want the default handler to mess with your event handler.