Detecting Pinch Zoom Via Interactions or Processors

I am using the new Input System to abstract away the specific details behind the devices used, and I really love that aspect of the new system. I am having a big issue trying to lump the mouse wheel and pinch to zoom into a single "zoom" action.

I do know a solution, but I don't like it. I can make an action specifically for touch devices alone that does a zoom specifically with actual knowledge of two touches... but that breaks my design because it makes assumptions about technologies used to cause actions.

So, I find myself trying to somehow compose the Vector2 for the "zoom" action by making a composite control made up of touches. This doesn't seem to work since the composite parts are expected to be up, down, left, and right. I don't see a way I can make this happen and wanted to reach out and see if anyone had ideas for me to try out.

 

I might be making progress here:

Code (CSharp):

1. #if UNITY_EDITOR
2.     [InitializeOnLoad]
3. #endif
4.     public class test : InputProcessor<Vector2>
5.     {
6.        
7. #if UNITY_EDITOR
8.         static test()
9.         {
10.             Initialize();
11.         }
12. #endif
13.  
14.         [RuntimeInitializeOnLoadMethod]
15.         static void Initialize()
16.         {
17.             InputSystem.RegisterProcessor<test>();
18.         }
19.  
20.         public override Vector2 Process(Vector2 value, InputControl control)
21.         {
22.             //$$ snag both positions and then produce your vector2 here
23.             var potentialControl0 = control.device.TryGetChildControl<Vector2Control>("/touch0/position");
24.             var potentialControl1 = control.device.TryGetChildControl<Vector2Control>("/touch1/position");
25.             return value;
26.         }
27.  
28.     }

 

Have you came up with something? Can you share processor final code?
I have a trouble with checking

InputControl

state. It seems like for /touch0/position there is no way to get propert

IsActuated()

(probably because it doesn't support magnitude evaluation). So how you handle case, when input system detect touch0 without touch1?

UPD0: Figured out that for such purposes modifiers suits better

UPD1: finally got it to work (described in this thread)