Scalar Math

Overview

Scalar math functions for floating-point and integer types: rounding, clamping, interpolation, sign, absolute value, power, and square root.

Why Use This?

These provide a consistent, type-safe API for common math operations across all supported numeric types, matching the same naming conventions as the vector operations.

Functions

All functions shown with _f32. Replace with _f64, _i64, _i32, _i16, _i8, _u64, _u32, _u16, _u8 as needed. Integer types do not have rounding, interpolation, fract, pow, or sqrt variants.

Rounding (float types only)

Function	Description
`lm2_floor_f32(a)`	Floor
`lm2_ceil_f32(a)`	Ceil
`lm2_round_f32(a)`	Round to nearest
`lm2_trunc_f32(a)`	Truncate toward zero
`lm2_floor_multiple_f32(a, multiple)`	Floor to nearest multiple
`lm2_ceil_multiple_f32(a, multiple)`	Ceil to nearest multiple
`lm2_round_multiple_f32(a, multiple)`	Round to nearest multiple
`lm2_trunc_multiple_f32(a, multiple)`	Truncate to nearest multiple

Sign and Absolute Value

Function	Types	Description
`lm2_abs_f32(a)`	f64, f32, i64-i8	Absolute value
`lm2_sign_f32(a)`	f64, f32, i64-i8	Returns 1 or -1
`lm2_sign0_f32(a)`	f64, f32, i64-i8	Returns 1, -1, or 0

Comparison

Function	Types	Description
`lm2_min_f32(a, b)`	all 10	Minimum of two values
`lm2_max_f32(a, b)`	all 10	Maximum of two values
`lm2_min_abs_f32(a, b)`	f64, f32, i64-i8	Minimum by absolute value
`lm2_max_abs_f32(a, b)`	f64, f32, i64-i8	Maximum by absolute value
`lm2_clamp_f32(min, value, max)`	all 10	Clamp value to [min, max]

Interpolation (float types only)

Function	Description
`lm2_saturate_f32(value)`	Clamp to [0, 1]
`lm2_lerp_f32(a, t, b)`	Linear interpolation: a + t*(b-a)
`lm2_smoothstep_f32(edge0, x, edge1)`	Smooth Hermite interpolation
`lm2_alpha_f32(a, value, b)`	Inverse lerp: compute t from value

Math (float types only)

Function	Description
`lm2_fract_f32(a)`	Fractional part
`lm2_pow_f32(base, exponent)`	Power
`lm2_sqrt_f32(a)`	Square root

Example

float val = 7.3f;

float clamped = lm2_clamp_f32(0.0f, val, 10.0f);     // 7.3
float rounded = lm2_round_multiple_f32(val, 2.5f);     // 7.5
float t = lm2_alpha_f32(0.0f, val, 10.0f);             // 0.73
float interp = lm2_lerp_f32(0.0f, 0.5f, 100.0f);      // 50.0