Pallas Math

Fixed-precision arithmetic for the parts of the Cardano protocol that need to compute ln, exp, and pow to high precision — most notably the reward and pool-saturation calculations. Backed by dashu for the underlying big-integer representation.

Usage

use pallas_math::math::{FixedDecimal, FixedPrecision};

let x = FixedDecimal::from_str("2", 34)?;   // 2.0 with 34 fractional digits
let y = x.ln();                             // ≈ 0.6931471805599453…
println!("ln(2) ≈ {y}");

Overview

• math — the public surface: FixedDecimal (the fixed-precision number type) and the FixedPrecision trait it implements (new, from_str, precision, exp, ln, pow, exp_cmp, round/floor/ceil/trunc).
• math_dashu — dashu-backed implementation that FixedDecimal aliases.
• DEFAULT_PRECISION (34), and the lazy ZERO / ONE / MINUS_ONE constants for convenience.