Selectively remap with multiple mice?

[pocococo]

I have a 4-button Kensington Expert Mouse trackball as well as a Wacom pen tablet connected to my computer. I like X-Mouse Button Control because it lets me remap my Expert Mouse buttons on Vista, where the Kensington driver isn't supported and doesn't work very well for me. However, I don't want X-Mouse to modify the behavior of the buttons on my Wacom pen.

Is there a way by which X-Mouse could discriminate between events coming from different devices, and avoid remapping button events for devices that the user had excluded? I'm thinking maybe of a list with a checkbox for each device or something.

[phil]

Sorry, no this is not possible.
I have been trying to work out a way of doing this but unfortunately, I have come to the conclusion that it is not feasable from a user level program.

I am still working on it and I may have a breakthrough, but don't count on it - I've tried using direct input and that does tell me where the messages come from but unfortunately, after I receive the mouse messages and not before, which makes it pretty impossible to work with .

Thanks,
Phil

[AlexV]

Have you tried Raw Input? Multiple Mice in Windows XP using Raw Input has a lot of information about this, including example apps and a helper API.

I've just knocked together some quick proof-of-concept code, and it does appear that the Raw Input does occur immediately before the WH_MOUSE mouse hook, as required.

If you do decide to implement multiple mouse detection, can I suggest also allowing per-mouse speed control? For laptop users, the touchpad often requires a very different sensitivity setting to the mouse. Similarly, trackball users may also have a mouse attached, which may need a different sensitivity. There is definitely demand for this, as there is already both a free and a commercial utility to switch between two sensitivities.

They, however, require a hotkey to be pressed to switch sensitivity. If you can detect multiple mice, though, you can do better. Once a mouse is clicked, or moved more than a small tolerance (SM_CX/YDRAG would seem reasonable), it becomes the active mouse and the sensitivity setting set for that mouse is applied. I bet that would bring in a whole new user-base, if they found out about it.

I know you've said that you're now in maintenance mode, and don't intend to add new features to XMBC, but wouldn't this be pretty cool?

Alex

[phil]

Ive tried playing around with raw input a while back and although I managed to pickup the input from different devices, in the hook itself, it is still impossible to know which device sent the message. I could not think of an easy way to tie the two systems together, as there is no guarantee about message ordering etc (if you press a button on each device, say hold one down, and release the other, it gets near impossible).

Phil

[AlexV]

Yes, I see what you're saying. You can't be 100% guaranteed sure where a hooked message came from.

To be honest, though, 99% is almost certainly good enough. How likely is it that a user will hold down a button on both devices at the same time anyway? If you just have it so that moving a device, or clicking any button on it (detected in raw), makes it the 'active' device, then settings assigned to it can be used in the hook.

Thanks for considering it, anyway.

Alex

[AlexV]

Just done a bit more checking. Although the Raw Input comes before the WH_MOUSE hook, it comes after the WH_MOUSE_LL hook, which presumably is the one you are using. I guess that means this won't work after all.

How very irritating.

[phil]

Yes that sounds familiar... my hook was getting called before the raw input. And I have to use the low level hook for various reasons, not least that I had major problems with the normal hook on some OS's.

As I said, I spent some time looking into this and did not come up with any magical ideas... But I havnt totally given up - I just dont have a lot of time to spare for XMBC and things like that certainly take time