Plan: how to implement input processing

[keesvandieren]

How can we implement input?

• can we reuse GWT implementation?
• what are bytecoder specifics we need to take care of?
• how much work is the minimal implementation?
• Focus first on mouse clicks / tabs, keyboard input lower prio for us

[yuripourre]

I think GWT can be used as a basis. It can copy the hookEvents method from DefaultGwtInput

 val w = Window.window()
 val d = w.document()
 d.addEventListener("mousemove", this);
 d.addEventListener("mousedown", this);
 d.addEventListener("mouseup", this);
 d.addEventListener("keyup", this);
//...

The BytecoderInput class should implement EventListener<Event> and inside the run method, check the type of the event and cast them for specific usage (just like gwt implementation is doing).

The steps would be:

• Add type to event (Merged)
• Add button to MouseEvent (Merged)
• Add key and code to KeyboardEvent (Merged)
• Create MouseWheel event (Merged)
• Update BytecoderInput adding hookEvents method
• Think about how to handle events outside the canvas element (calculate relative positions, handle focus, etc).

[yuripourre]

@keesvandieren @CoenRijsdijk I updated my comment to reflect the current state of Bytecoder. I added a task list that can work as a plan.

What do you think?