This sounds very similar to bug 1522089 in that there are two events - one for the selected mute button and one other affecting the media element. I'm moving this to Event Handling to follow suit with bug 1522089.

(In reply to Andreas Pehrson [:pehrsons] from comment #6)

This sounds very similar to bug 1522089 in that there are two events - one for the selected mute button and one other affecting the media element. I'm moving this to Event Handling to follow suit with bug 1522089.

Like bug 1522089, I'm thinking if it's worth having Mike's team taking a look at this. :)

Cristi, can you re-test before we dig in? YouTube just rolled back some key event capture changes that fixed Bug 1522089. Thanks!

(In reply to Mike Taylor [:miketaylr] from comment #8)

Cristi, can you re-test before we dig in? YouTube just rolled back some key event capture changes that fixed Bug 1522089. Thanks!

Sorry I should have mentioned that I tested this again even after bug 1522089 was fixed. It's still good to have Cristi's confirmation. :)

Still reproducible with 66.0a1(20190127213438).

Unfortunately this isn't reproducing for me on any of my systems.

Adding 66.0 and 67.0a1 (2019-03-11) as still affected builds, and Ubuntu 16.04 for the OS section.

Sergiu, does this reproduce for you?

Yes! This is reproducible for me on Firefox nightly 67.0a1 (2019-03-14) (64-bit).
The issue is NOT reproducible if you press the Enter key. Pressing the Enter key just toggles the Mute/Unmute button.

My hunch is that after clicking on the button, the focus is kept on it.
I think that this is a case where one physical command (in this case, pressing the Spacebar) is firing 2 different events.

Another example of this is if you, instead of the Mute/Unmute button, click on the Cog button. The "Autoplay" option is toggled with every press of the Spacebar. The difference here is that the video is not paused.

Ah, now I can see it on Windows 10. Tom, can you re-test on Windows?

I also have bug on youtube, I think it's related to this one.

How to reproduce?

Sorry, for double message.

Reproduce (Windows):

1. Start any video on youtube.
2. Click on pause button on the bottom of player.
3. Press space. Video will play and paused again immediately.

Unassigned P2 and we shipped with this bug since at least 63, so wontfix for 67.

So the symptoms do happen for me with the STR given by alexander (they do not if I focus the play/pause button by tabbing to give the button the focus, but that's neither here nor there). Just for reference, this is with YouTube running Polymer 3.2.0.

After fishing through a ton of red herrings, it turns out that the problem here is that their code has Firefox try to toggle the play/pause button not just for the initial keydown event, but also for the related click event when the button is activated by the spacebar. That is, it does not try to do the toggling in Chrome during click, yet does in Firefox.

That made me wonder whether they are doing something Chrome-specific to avoid the click events, as they're definitely being fired (I can see them in the console when I add a simple listener and use the spacebar). And indeed that must be the case, because when I spoof Firefox's UA String as Chrome or Safari, I can no longer reproduce the problem.

But oddly, it's just spoofing the HTTP headers that is effective. The JS environment variables don't seem to influence the result. And after poring over the responses listed in the network panel, I don't really see anything in the base markup, related scripts, or even CSS that gives me any insight.

I don't really know what else to try at this stage, other than reaching out to YouTube with the STR and my findings, and hoping that helps them find the bug.

I've sent an email to our partner mailing list.

Experienced the same problem on Ubuntu 18.04. As this affects YouTube, one of the most popular websites, it would be nice if it could get higher priority.

it would be nice if it could get higher priority.

In this case, it's YouTube that needs to fix the bug. So the priority doesn't really mean much (since we're not writing code to fix this ourselves).

(In reply to Mike Taylor [:miketaylr] (PTO Dec 21 - Jan 2) from comment #20)

I've sent an email to our partner mailing list.

Perhaps we could follow up with them if there hasn't been any movement so far?

This issue was reproducible on Windows 10 with Firefox Nightly version 74.0a1 (2020-01-14) (64-bit) / Beta version 73.0b4 (64-bit) / Release version 72.0.1 (64-bit). Marked as affected.

(In reply to Botond Ballo [:botond] from comment #23)

Perhaps we could follow up with them if there hasn't been any movement so far?

I can reproduce as well, and pinged our contacts on the thread started in May.

Hi,

I was able to reproduce this issue on Windows for Firefox version Nightly 75.0a1 (2020-02-20). Marking that flag as affected.

I was able to reproduce this issue on latest Nightly version 76.0a1 (2020-03-17) on Ubuntu 18.04, Changed flags accordingly.

This issue is reproducible on Windows 10 with Firefox Nightly version 77.0a1 (2020-04-15) (64-bit)

I'm going to remove the regression keyword, because this issue wasn't caused by us, but by YouTube.

(We probably don't need to update the affected flags for each release as well)

For me, this behaves identically to bug 1481400 FF 76.0.1 and is still broken.

The issue is not reproducible on my side anymore. Following the steps to reproduce, pressing the "Space" key on the keyboard does not trigger the mute/unmute function of the video player:

https://media.giphy.com/media/lvDP1wQf66JJDXB6rk/giphy.gif

Cristian, is the issue reproducible on your side?

Notes:
The button's margins are highlighted on other browsers as well

Tested with:

Browser / Version: Firefox Nightly 101.0a1 (2022-04-21) (64-bit) / Chrome Version Version 100.0.4896.127 (Official Build) (64-bit)
Operating System: Windows 10 PRO x64
Operating System: Ubuntu 20.4 LTS x64

Indeed, the initial issue appears to have been fixed (as only one of the 2x buttons gets triggered now).
Confirming with 102.0a1 (2022-05-03) - Win10 as well.

Updating status to WFM, as we do not know which patch has fixed this.
Thank you for double checking @Raul!