aivi.core.float

Pure helpers and compiler intrinsics for finite IEEE 754 double-precision values. The MVP keeps general numeric operations here; presentation rounding, percentages, angle wrapping, and curve evaluation belong in application libraries.

0.0 and -0.0 compare equal and address the same map key, including inside compound keys. Their signs are preserved when rendering or serializing values. Runtime deserialization rejects NaN and infinities, just as construction and parseText do.

use aivi.core.float (
    pi
    e
    tau
    negate
    max
    min
    clamp
    lerp
    sign
    between
    isZero
    isPositive
    isNegative
    square
    toRadians
    toDegrees
    floor
    ceil
    round
    sqrt
    abs
    toInt
    fromInt
    toText
    parseText
    approxEq
)

Constants

Value	Meaning
pi	π
e	Euler's number
tau	2π

Pure helpers

Function	Type	Behavior
negate	Float -> Float	Flip the sign
max	Float -> Float -> Float	Return the greater value
min	Float -> Float -> Float	Return the lesser value
clamp	Float -> Float -> Float -> Float	Restrict a value to inclusive bounds
lerp	Float -> Float -> Float -> Float	Linear interpolation
sign	Float -> Float	Return -1.0, 0.0, or 1.0
between	Float -> Float -> Float -> Bool	Inclusive range test
isZero	Float -> Bool	Test against 0.0
isPositive	Float -> Bool	Test above zero
isNegative	Float -> Bool	Test below zero
square	Float -> Float	Multiply a value by itself
toRadians	Float -> Float	Convert degrees to radians
toDegrees	Float -> Float	Convert radians to degrees
approxEq	Float -> Float -> Float -> Bool	approxEq epsilon a b tests abs (a - b) <= epsilon

Compiler intrinsics

Function	Type	Behavior
floor	Float -> Float	Round down
ceil	Float -> Float	Round up
round	Float -> Float	Round to the nearest integral value
sqrt	Float -> Float	Square root
abs	Float -> Float	Absolute value
toInt	Float -> Int	Truncate toward zero
fromInt	Int -> Float	Convert an integer
toText	Float -> Text	Render decimal text
parseText	Text -> Option Float	Parse finite decimal text

use aivi.core.float (
    approxEq
    lerp
    pi
    toRadians
)

value midpoint : Float = lerp 10.0 20.0 0.5
value halfTurn : Bool = approxEq 0.000001 (toRadians 180.0) pi