TransformRange – scale value from one range to another

[DruggedBunnyParticipant]

Had to look this one up! Main intended use for me is to scale joystick input values (-1.0 to 1.0) into other ranges (eg. 0.0 to 1.0)…

Monkey

#Import "<std>" ' NB. Input range (from_max - from_min) must be non-zero -- mathematically impossible to scale a zero-size range into a non zero-size range! Function TransformRange:Float (input_value:Float, from_min:Float, from_max:Float, to_min:Float, to_max:Float) ' Algorithm via jerryjvl at https://stackoverflow.com/questions/929103/convert-a-number-range-to-another-range-maintaining-ratio Local from_delta:Float = from_max - from_min ' Input range, eg. 0.0 - 1.0 Local to_delta:Float = to_max - to_min ' Output range, eg. 5.0 - 10.0 Assert (from_delta <> 0.0, "TransformRange: Invalid input range!") Return (((input_value - from_min) * to_delta) / from_delta) + to_min End Function Main () ' Transform value of 0.5 in range 0.0 - 1.0 into range 0.5 - 1.0... Local value:Float = 0.5 ' Half-way into input range 0.0 - 1.0 Local from_range_min:Float = 0.0 Local from_range_max:Float = 1.0 Local to_range_min:Float = 5.0 Local to_range_max:Float = 10.0 ' Answer = 7.5 -- halfway between 5 and 10. Print TransformRange (value, from_range_min, from_range_max, to_range_min, to_range_max) End

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

#Import "<std>"   ' NB. Input range (from_max - from_min) must be non-zero -- mathematically impossible to scale a zero-size range into a non zero-size range!   Function TransformRange:Float (input_value:Float, from_min:Float, from_max:Float, to_min:Float, to_max:Float)   ' Algorithm via jerryjvl at https://stackoverflow.com/questions/929103/convert-a-number-range-to-another-range-maintaining-ratio Local from_delta:Float = from_max - from_min ' Input range, eg. 0.0 - 1.0 Local to_delta:Float = to_max - to_min ' Output range, eg. 5.0 - 10.0 Assert (from_delta <> 0.0, "TransformRange: Invalid input range!") Return (((input_value - from_min) * to_delta) / from_delta) + to_min End   Function Main ()   ' Transform value of 0.5 in range 0.0 - 1.0 into range 0.5 - 1.0... Local value:Float = 0.5 ' Half-way into input range 0.0 - 1.0 Local from_range_min:Float = 0.0 Local from_range_max:Float = 1.0   Local to_range_min:Float = 5.0 Local to_range_max:Float = 10.0 ' Answer = 7.5 -- halfway between 5 and 10. Print TransformRange (value, from_range_min, from_range_max, to_range_min, to_range_max)   End

 

Not sure if the correct mathematical term is ‘transform’, but that’s what I’m calling it…

[nerobotParticipant]

Useful function!

I made this func a bit universal – by using generic for any number types.

Monkey

Function TransformRange<T>:T (input_value:T, from_min:T, from_max:T, to_min:T, to_max:T) Where T Implements INumeric ' Algorithm via jerryjvl at https://stackoverflow.com/questions/929103/convert-a-number-range-to-another-range-maintaining-ratio Local from_delta := from_max - from_min ' Input range, eg. 0.0 - 1.0 Local to_delta := to_max - to_min ' Output range, eg. 5.0 - 10.0 Assert (from_delta <> Null, "TransformRange: Invalid input range!") Return (((input_value - from_min) * to_delta) / from_delta) + to_min End

1 2 3 4 5 6 7 8 9 10 11

Function TransformRange<T>:T (input_value:T, from_min:T, from_max:T, to_min:T, to_max:T) Where T Implements INumeric ' Algorithm via jerryjvl at https://stackoverflow.com/questions/929103/convert-a-number-range-to-another-range-maintaining-ratio Local from_delta := from_max - from_min ' Input range, eg. 0.0 - 1.0 Local to_delta := to_max - to_min ' Output range, eg. 5.0 - 10.0 Assert (from_delta <> Null, "TransformRange: Invalid input range!") Return (((input_value - from_min) * to_delta) / from_delta) + to_min End

Note 1: I replaced Assert (from_delta <> 0.0,... with Assert (from_delta <> Null,... to have comparison with _0_ for int-based and 0.0 for float-based types.

Note 2: Seems that comparison floats with any concrete number is a bad idea, because 0.0 is about 0.0000000001 or so. But it works!

Note 3: Bad idea to use it for integers because of division operation… so maybe better where-clause would be Where T=Float Or T=Double.

[DruggedBunnyParticipant]

Nice, I briefly thought about doing <T> but didn’t really need it.

Not really familiar with Where, but I think I see how that works, which is cool.

I’d say it’s a bug, though, if float zero ever evaluates to 0.0000000000001! Although floats can definitely not represent all numbers, and direct value comparisons are generally a bad idea, floating point zero should still always be 0.0[… 0] no matter what — all bits being set to zero.

Already found mine really useful, great for getting finer control from joystick inputs!

[Author]

Posts