Question

I’ll try to explain my scenario as best i can;

At each application tick I query the current state of the keyboard and mouse and wrap them in individual classes and data structures. For the keyboard it’s an array of my Keys enum (one item for each of the keys that are currently pressed) and for the mouse it’s a class containing coordinate delta’s and bools for each buttons pressed.

I also have a rudimentary state machine managed via a state manager class which maintains the stack and marshalls the states.

All I want to know is, how best to pass the input (snapshots) to the individual states my app can be in at any one time?

I would like to process as much of the input as possible away from the individual states, so as to reduce repeating logic within the states.

Or would it be better to keep the input snapshots as pure as possible and pass them to the states so they can keep they’re input-specific-logic hidden?

Note
This structure is similiar to how I imagine a game would work, and although this application is not a game it does need to be processed as quickly as possible.

Answer 1

Why are you querying the state of the keyboard and mouse with each tick? A much better and traditional solution would be to capture events fired from the keyboard and mouse. That way you only need to update the state when you HAVE to.