Fixed-size decimal numbers implemented in pure Rust. Suitable for financial, crypto and any other fixed-precision calculations.

API Docs

This crate is inspired by num_bigint and bigdecimal – amazing crates that allow you to store big integers and arbitrary precision fixed-point decimal numbers almost any precision.

BigInt internally uses a Vec of decimal digits the size of which is theoretically limited only by the usize max value or memory capacity.

Under the hood BigDecimal uses a BigInt object, paired with a 64-bit integer which determines the position of the decimal point. Therefore, the precision is not actually arbitrary, but limited to 2 ⁶³ decimal places.

Despite the seemingly undeniable advantages at first glance, this approach also has a number of fundamental disadvantages:

• Non-copyable types for both integers and fixed point numbers.
• Dynamic allocation to store even tiny numbers, for example, 0 or 1.
• Extra dynamic allocation for almost any operation (mathematical operations, parsing, converting, etc.).
• Constant calculations are not available.
• Potentially uncontrolled growth of memory consumption and the need to artificially limit it.

Because most practical problems requiring the use of fixed-point numbers do not require so much limit on the number of digits, such as usize, but as a rule it is limited:

Then most real numbers for financial and other systems requiring accuracy can use 256-bit or even 128-bit integer to store decimal digits.

So In this library, a different approach was chosen.

fastnum provides signed and unsigned exact precision decimal numbers suitable for financial calculations that require significant integral and fractional digits with no round-off errors (such as 0.1 + 0.2 ≠ 0.3).

Any fastnum decimal type consists of an N-bit big unsigned integer, paired with a 64-bit signaling block which contains a 16-bit scaling factor determines the position of the decimal point, sign, special and signaling flags. Trailing zeros are preserved and may be exposed when in string form.

Thus, fixed-point numbers are trivially copyable and don't require any dynamic allocation. This allows you to get additional performance gains by eliminating not only dynamic allocation, like such, but also will get rid of one indirect addressing, which improves cache-friendliness and reduces the CPU load.

• Strictly exact precision: no round-off errors (such as 0.1 + 0.2 ≠ 0.3).
• Special values: fastnum support ±0, ±Infinity and NaN special values with IEEE 754 semantic.
• Blazing fast: fastnum numerics are as fast as native types, well almost :).
• Trivially copyable types: all fastnum numerics are trivially copyable (both integer and decimal, ether signed and unsigned) and can be stored on the stack, as they're fixed size.
• No dynamic allocation: no expensive sys-call's, no indirect addressing, cache-friendly.
• Compile-time integer and decimal parsing: all the from_* methods on fastnum integers and decimals are const, which allows parsing of integers and numerics from string slices and floats at compile time. Additionally, the string to be parsed does not have to be a literal: it could, for example, be obtained via include_str!, or env!.
• Const-evaluated in compile time macro-helpers: any type has its own macro helper which can be used for definitions of constants or variables whose value is known in advance. This allows you to perform all the necessary checks at the compile time.
• Short dependencies list by default: fastnum depends only upon bnum by default. All other dependencies are optional. Support for crates such as rand and serde can be enabled with crate features.
• no-std compatible: fastnum can be used in no_std environments.
• wasm compatible: fastnum is fully compatible with WebAssembly.
• const evaluation: nearly all methods defined on fastnum integers and decimals are const, which allows complex compile-time calculations and checks.
• Full range of advanced mathematical functions: exponential, roots, power, logarithmic, and trigonometric functions for working with exact precision decimals. And yes, they're all const too.

To install and use fastnum, simply add the following line to your Cargo.toml file in the [dependencies] section:

fastnum = "0.7"

Or, to enable various fastnum features as well, add, for example, this line instead:

fastnum = { version = "0.7", features = ["serde"] } # enables the "serde" feature

use fastnum::*;

fn main() {
    const ZERO: UD256 = udec256!(0);
    const ONE: UD256 = udec256!(1.0);

    let a = udec256!(12345);

    println!("a = {a}");
}

The numtraits feature includes implementations of traits from the num_traits crate, e.g. AsPrimitive, Signed, etc.

The rand feature allows creation of random fastnum decimals via the rand crate.

The serde feature enables serialization and deserialization of fastnum decimals via the serde crate. More details about serialization and deserialization you can found in

The zeroize feature enables the Zeroize trait from the zeroize crate.

The diesel feature enables serialization and deserialization of fastnum decimals for diesel crate.

The sqlx feature enables serialization and deserialization of fastnum decimals for sqlx crate.

The tokio-postgres feature enables serialization and deserialization of fastnum decimals for tokio-postgres crate.

The utoipa feature enables support of fastnum decimals for autogenerated OpenAPI documentation via the utoipa crate.

^* Precision is arbitrary but fixed.

fastnum is blazing fast. As much as possible given the overhead of arbitrary precision support.

Some benchmark reports are shown below:

Allocate vec![] with N elements.

Perform a + b.

Perform a - b.

Perform a × b.

Perform a ÷ b.

Convert D64/D128 decimals into f64 floating point.

^* bigdecimal to float conversion is not pretty accurate.

You can run benchmark tests with Criterion.rs tool:

cd benchmark
cargo criterion

This crate is tested as well as with specific edge cases.

cargo test --all-features

The current Minimum Supported Rust Version (MSRV) is 1.87.0.

API Docs

NB: fastnum is currently pre-1.0.0. As per the Semantic Versioning guidelines, the public API may contain breaking changes while it is in this stage. However, as the API is designed to be as similar as possible to the API of Rust's primitive types, it is unlikely that there will be a large number of breaking changes.

You can set a few default parameters at compile-time via environment variables:

There are several areas for further work:

• Micro-optimization of big integer types using vector extensions (SSE2, SSE4.2, AVX2, AVX512F, etc.).
• Const trait implementations once they're stabilized in Rust. (rust-lang/rust#67792)
• Integration with a large number of crates (ORM's, auto-docs crates, etc.).

This code is dual-licensed under the permissive MIT & Apache 2.0 licenses.

Unless you explicitly state otherwise, any contribution intentionally submitted for inclusion in the work by you, as defined in the Apache-2.0 license, shall be dual licensed as above, without any additional terms or conditions.

view changelog