Push-To-Talk won't stop with some bindings

[two-facedjanus]

I got a new mouse today, the Evoluent 4 ergonomic, and after fiddling with the drivers which had some bizarre issue wherein the mouse couldn't be used for push-to-talk I finally got it working, but there's a new problem. Two of the buttons on the mouse are currently mapped to "Back" and "Forward" and a third button is also mapped to "Forward". Whenever I'm using push-to-talk with a "back" or "forward" mapped key, when I release, PTT will not stop and I continually transmit despite my best efforts to stop it using that key.

To fix this I have to go into the settings and change my PTT button to something else like tilde or shift or what have you and push it to toggle PTT.

So for some reason these back and forward bound buttons are unable to stop transmitting when used for PTT. Additionally if I'm using a binding for "ctrl" or "alt" the same problem exists. I'm not entirely sure if this is a driver issue with the Evoluent 4 mouse though. If I change the binding to something else like F12 using the mouse software, it functions as intended and PTT will stop transmitting when I release the button.

EDIT: Not every key can be bound to this mouse. Some functions are unique, like "Secondary Functions" which Mumble is incapable of registering, and most of the key combinations interfere with other programs, namely the F1-F12 buttons which are used for games and such. Rebinding is not a viable solution for this.

[rawnar]

The Push-To-Talk system listens to the events pressed and released. When the signal pressed is emitted mumble will start capturing the audio input stream, until the released signal is emitted. It could be that some of the bindings do not transmit the released signal or send a double pressed signal when pressed. This will cause mumble not to stop transmitting the audio stream. If it is a double pressed signal, then it can probably be fixed from within mumble.

Try this link to see which events are send. It works on Java, so it will not be exactly the same as Qt but it may help locate the problem.

These unique function are probably not sending any signal to the event system of windows, but are only used for the drivers itself or internally by the mouse. My mouse also has this button but the function of it is handle inside the mouse and has nothing to do with the drivers.

[two-facedjanus]

I can hardly imagine that I'm using this page right but activating that link with a normal mousebind sends the signals:

mousedown

mouseup

click

Trying it with the mousewheel gets:

mousedown

mouseup

Trying it with something set to "forward" or "back" doesn't yield anything in the javascript window so I'm not sure what it's transmitting.

[rawnar]

Probably the java script is not recording it because the actions is not performed on the link. Darn.

Okay I found a program that should do the trick. Follow this link and get spy++. I have not tested it myself, so be careful.

[two-facedjanus]

I'm afraid I'm not sure what to make of any of this.

Under the Windows tab I found something named "Evoluent1" under its particular windows handle and the associated processes and threads. That's the name of the mouse manufacturer.

Further than this I really can't determine anything other than the mouse is being managed by a program. I can't seem to detect any mouse presses or anything of the like. Trying to log messages for the process doesn't yield anything. I dunno if I'm just using the program wrong because I've never seen it in my life (likely) or if it's not doing what I thought it would.

[rawnar]

I tried the program as well. You need to start logging the messages from your mouse actions. But on my setup the program was not giving me any output. This could be related to the fact it is a 32-bit program, running on a 64-bit operation system.

This kind of sucks, that there is no program like xev on windows to monitor the mouse events. At this moment I am out of idea's, but maybe the guys in the IRC can help you.