Trying to emulate hitbox controller rules

[zehthales]

I'm playing Street Fighter V on steam with my keyboard. The steam key mapping for Keyboard do something like the key left pressed together with key right have the output of always following the opponent position.

At hitbox controller is simpler, right cancels left and up cancel down. Basically right+left=down+up=stand.

I tried to something like this with input-remaper, but no success by now.

Note: the remapping works inside the game, I just don't know how to implement it

[jonasBoss]

Did I get this correct: if you press left+right or up+down you want the output to be no key pressed?

This is a bit difficult on the current release, but if you switch to beta you can do this:

• left+right -> disable (literally just write "disable" in the output field)
• up+down -> disable

On the current release you will probably need 6 mappings:

• left -> KEY_LEFT
• right -> KEY_RIGHT
• left+right -> event(EV_KEY, KEY_LEFT, 0).event(EV_KEY, KEY_RIGHT, 0) this will release both the left an right key
• the same for up and down

[zehthales]

The unmask command had no output. After this, trying to enable says that still masked, but the beta version is fully working now. I even tried you tip on remapping keyboard into gamepad analog or axis

Thank you very much!

[jonasBoss]

I think on beta this is as close as it gets:

{
    "1,17,1": {
        "target_uinput": "keyboard",
        "output_symbol": "set(y_state, 1)\n.key_down(KEY_UP)\n.hold(\n\tif_eq(\n\t\t$y_state, 0,\n\t\tset(y_state, 1).key_down(KEY_UP),\n\t)\n)\n.if_eq(\n\t$y_state, 0,\n\tset(y_state, -1).key_down(KEY_DOWN),\n\tkey_up(KEY_UP)\n)"
    },
    "1,31,1": {
        "target_uinput": "keyboard",
        "output_symbol": "set(y_state, -1)\n.key_down(KEY_DOWN)\n.hold(\n\tif_eq(\n\t\t$y_state, 0,\n\t\tset(y_state, -1).key_down(KEY_DOWN),\n\t)\n)\n.if_eq(\n\t$y_state, 0,\n\tset(y_state, 1).key_down(KEY_UP),\n\tkey_up(KEY_DOWN)\n)"
    },
    "1,17,1+1,31,1": {
        "target_uinput": "keyboard",
        "output_symbol": "key_up(KEY_UP).key_up(KEY_DOWN)\n.hold()\n.set(y_state, 0)"
    }
}

This will map the "w" and "s" keys to the up and down arrow keys. There is one issue:
pressing "w" then "s" and then releasing "w" will result in a stuck down arrow until "s" is pressed and released again.

same thing but with the gamepad axis instead:

{
    "1,30,1": {
        "target_uinput": "gamepad",
        "output_symbol": "set(x_state, 1)\n.event(3,0,32768)\n.hold(\n\tif_eq(\n\t\t$x_state, 0,\n\t\tset(x_state, 1).event(3,0,32768),\n\t)\n)\n.if_eq(\n\t$x_state, 0,\n\tset(x_state, -1).event(3,0,-32768),\n\tevent(3,0,0)\n)"
    },
    "1,32,1": {
        "target_uinput": "gamepad",
        "output_symbol": "set(x_state, -1)\n.event(3,0,-32768)\n.hold(\n\tif_eq(\n\t\t$x_state, 0,\n\t\tset(x_state, -1).event(3,0,-32768),\n\t)\n)\n.if_eq(\n\t$x_state, 0,\n\tset(x_state, 1).event(3,0,32768),\n\tevent(3,0,0)\n)\n"
    },
    "1,30,1+1,32,1": {
        "target_uinput": "gamepad",
        "output_symbol": "event(3,0,0)\n.hold()\n.set(x_state, 0)"
    }
}

[zehthales]

This is very close! Solve most of the buffers problem related with steam mapping. If we could add a releasing information to input key to avoid that problem...

[zehthales]

What happens if I write "1,30,1+1,32,1+1,30,0" ?

[jonasBoss]

you will get an error or the mapping will just not work.

I am very certain that you can not write a macro that solves this entirely. Because the macro is missing information about the keys which are pressed and released.
The solution above is trying to infer information by observing when which mapping is triggered and released.
If we press "w" then "s" and then release "w" no more macro is running and we end up with a stuck key.

[zehthales]

too bad! But your solution was very nice! Thank you for the help