float

dashu-float

Arbitrary precision floating point number implementation, as a part of the dashu library for arbitrary-precision mathematics. See Docs.rs for the full documentation.

Features

• Supports no_std and written in pure Rust.
• Support arbitrary base and arbitrary rounding mode.
• Support efficient base conversion.
• Small float numbers are inlined on stack.
• Efficient float number parsing and printing with base 2~36.
• Supports the hexadecimal float format used by C++.
• Developer friendly debug printing for float numbers.

Quick example

Binary and decimal float literals, correctly-rounded transcendentals, and the Context layer for explicit precision control:

use dashu::{fbig, dbig};
use dashu_float::{Context, round::mode::HalfEven};

// Binary and decimal float literals — hex floats, precision from digit count
let x = fbig!(0x1.fffp-1); // ≈ 0.9998
let pi = dbig!(3.1415926535897932384626);

// Correctly-rounded transcendentals (Ziv engine)
let e = x.exp(); // exp(0.9998…), binary
let _log_pi = pi.ln();

// Context layer: IEEE 754 binary64 precision, HalfEven rounding
let ctx = Context::<HalfEven>::new(53);
let e53 = ctx.exp(&x.repr(), None).unwrap().value();
assert!(e53 > e); // higher precision, larger value

// set the precision explicitly
let x50 = x.with_precision(50).value();
assert_eq!(x50.precision(), 50);

Optional dependencies

• std (default): enable std support for dependencies.

Performance

Relevant benchmark will be implemented in the built-in benchmark.

License

See the top-level readme.