Skip to content

Latest commit

 

History

History
408 lines (290 loc) · 13.8 KB

README.md

File metadata and controls

408 lines (290 loc) · 13.8 KB
Raw

Decimal   Build Status Latest Version Docs Badge

A Decimal number implementation written in pure Rust suitable for financial calculations that require significant integral and fractional digits with no round-off errors.

The binary representation consists of a 96 bit integer number, a scaling factor used to specify the decimal fraction and a 1 bit sign. Because of this representation, trailing zeros are preserved and may be exposed when in string form. These can be truncated using the normalize or round_dp functions.

Installing

$ cargo add rust_decimal

In addition, if you would like to use the optimized macro for convenient creation of decimals:

$ cargo add rust_decimal_macros

Alternatively, you can edit your Cargo.toml directly and run cargo update:

[dependencies]
rust_decimal = "1.36"
rust_decimal_macros = "1.36"

Usage

Decimal numbers can be created in a few distinct ways. The easiest and most efficient method of creating a Decimal is to use the procedural macro that can be enabled using the macros feature:

// Import the `rust_decimal_macros` crate and use the macro directly from there.
use rust_decimal_macros::dec;

let number = dec!(-1.23) + dec!(3.45);
assert_eq!(number, dec!(2.22));
assert_eq!(number.to_string(), "2.22");

Alternatively you can also use one of the Decimal number convenience functions (see the docs for more details):

// Using the prelude can help importing trait based functions (e.g. core::str::FromStr).
use rust_decimal::prelude::*;

// Using an integer followed by the decimal points
let scaled = Decimal::new(202, 2);
assert_eq!("2.02", scaled.to_string());

// From a 128 bit integer
let balance = Decimal::from_i128_with_scale(5_897_932_384_626_433_832, 2);
assert_eq!("58979323846264338.32", balance.to_string());

// From a string representation
let from_string = Decimal::from_str("2.02").unwrap();
assert_eq!("2.02", from_string.to_string());

// From a string representation in a different base
let from_string_base16 = Decimal::from_str_radix("ffff", 16).unwrap();
assert_eq!("65535", from_string_base16.to_string());

// From scientific notation
let sci = Decimal::from_scientific("9.7e-7").unwrap();
assert_eq!("0.00000097", sci.to_string());

// Using the `Into` trait
let my_int: Decimal = 3_i32.into();
assert_eq!("3", my_int.to_string());

// Using the raw decimal representation
let pi = Decimal::from_parts(1_102_470_952, 185_874_565, 1_703_060_790, false, 28);
assert_eq!("3.1415926535897932384626433832", pi.to_string());

Once you have instantiated your Decimal number you can perform calculations with it just like any other number:

use rust_decimal::prelude::*;
use rust_decimal_macros::dec;

let amount = dec!(25.12);
let tax_percentage = dec!(0.085);
let total = amount + (amount * tax_percentage).round_dp(2);
assert_eq!(total, dec!(27.26));

Features

Behavior / Functionality

Database

Serde

borsh

Enables Borsh serialization for Decimal.

c-repr

Forces Decimal to use [repr(C)]. The corresponding target layout is 128 bit aligned.

db-postgres

Enables a PostgreSQL communication module. It allows for reading and writing the Decimal type by transparently serializing/deserializing into the NUMERIC data type within PostgreSQL.

db-tokio-postgres

Enables the tokio postgres module allowing for async communication with PostgreSQL.

db-diesel-postgres

Enable diesel PostgreSQL support.

db-diesel-mysql

Enable diesel MySQL support.

legacy-ops

Warning: This is deprecated and will be removed from a future versions.

As of 1.10 the algorithms used to perform basic operations have changed which has benefits of significant speed improvements. To maintain backwards compatibility this can be opted out of by enabling the legacy-ops feature.

maths

The maths feature enables additional complex mathematical functions such as pow, ln, enf, exp etc. Documentation detailing the additional functions can be found on the MathematicalOps trait.

Please note that ln and log10 will panic on invalid input with checked_ln and checked_log10 the preferred functions to curb against this. When the maths feature was first developed the library would instead return 0 on invalid input. To re-enable this non-panicking behavior, please use the feature: maths-nopanic.

ndarray

Enables arithmetic operations using ndarray on arrays of Decimal.

proptest

Enables a proptest strategy to generate values for Rust Decimal.

rand

Implements rand::distributions::Distribution<Decimal> to allow the creation of random instances.

Note: When using rand::Rng trait to generate a decimal between a range of two other decimals, the scale of the randomly-generated decimal will be the same as the scale of the input decimals (or, if the inputs have different scales, the higher of the two).

rkyv

Enables rkyv serialization for Decimal. In order to avoid breaking changes, this is currently locked at version 0.7.

Supports rkyv's safe API when the rkyv-safe feature is enabled as well.

If rkyv support for versions 0.8 of greater is desired, rkyv's remote derives should be used instead. See examples/rkyv-remote.

rocket-traits

Enable support for Rocket forms by implementing the FromFormField trait.

rust-fuzz

Enable rust-fuzz support by implementing the Arbitrary trait.

serde-float

Note: This feature applies float serialization/deserialization rules as the default method for handling Decimal numbers. See also the serde-with-* features for greater flexibility.

Enable this so that JSON serialization of Decimal types are sent as a float instead of a string (default).

e.g. with this turned on, JSON serialization would output:

{
  "value": 1.234
}

serde-str

Note: This feature applies string serialization/deserialization rules as the default method for handling Decimal numbers. See also the serde-with-* features for greater flexibility.

This is typically useful for bincode or csv like implementations.

Since bincode does not specify type information, we need to ensure that a type hint is provided in order to correctly be able to deserialize. Enabling this feature on its own will force deserialization to use deserialize_str instead of deserialize_any.

If, for some reason, you also have serde-float enabled then this will use deserialize_f64 as a type hint. Because converting to f64 loses precision, it's highly recommended that you do NOT enable this feature when working with bincode. That being said, this will only use 8 bytes so is slightly more efficient in terms of storage size.

serde-arbitrary-precision

Note: This feature applies arbitrary serialization/deserialization rules as the default method for handling Decimal numbers. See also the serde-with-* features for greater flexibility.

This is used primarily with serde_json and consequently adds it as a "weak dependency". This supports the arbitrary_precision feature inside serde_json when parsing decimals.

This is recommended when parsing "float" looking data as it will prevent data loss.

Please note, this currently serializes numbers in a float like format by default, which can be an unexpected consequence. For greater control over the serialization format, please use the serde-with-arbitrary-precision feature.

serde-with-float

Enable this to access the module for serializing Decimal types to a float. This can be used in struct definitions like so:

#[derive(Serialize, Deserialize)]
pub struct FloatExample {
    #[serde(with = "rust_decimal::serde::float")]
    value: Decimal,
}
#[derive(Serialize, Deserialize)]
pub struct OptionFloatExample {
    #[serde(with = "rust_decimal::serde::float_option")]
    value: Option<Decimal>,
}

Alternatively, if only the serialization feature is desired (e.g. to keep flexibility while deserialization):

#[derive(Serialize, Deserialize)]
pub struct FloatExample {
    #[serde(serialize_with = "rust_decimal::serde::float::serialize")]
    value: Decimal,
}

serde-with-str

Enable this to access the module for serializing Decimal types to a String. This can be used in struct definitions like so:

#[derive(Serialize, Deserialize)]
pub struct StrExample {
    #[serde(with = "rust_decimal::serde::str")]
    value: Decimal,
}
#[derive(Serialize, Deserialize)]
pub struct OptionStrExample {
    #[serde(with = "rust_decimal::serde::str_option")]
    value: Option<Decimal>,
}

This feature isn't typically required for serialization however can be useful for deserialization purposes since it does not require a type hint. Consequently, you can force this for just deserialization by:

#[derive(Serialize, Deserialize)]
pub struct StrExample {
    #[serde(deserialize_with = "rust_decimal::serde::str::deserialize")]
    value: Decimal,
}

serde-with-arbitrary-precision

Enable this to access the module for deserializing Decimal types using the serde_json/arbitrary_precision feature. This can be used in struct definitions like so:

#[derive(Serialize, Deserialize)]
pub struct ArbitraryExample {
    #[serde(with = "rust_decimal::serde::arbitrary_precision")]
    value: Decimal,
}
#[derive(Serialize, Deserialize)]
pub struct OptionArbitraryExample {
    #[serde(with = "rust_decimal::serde::arbitrary_precision_option")]
    value: Option<Decimal>,
}

An unexpected consequence of this feature is that it will serialize as a float like number. To prevent this, you can target the struct to only deserialize with the arbitrary_precision feature:

#[derive(Serialize, Deserialize)]
pub struct ArbitraryExample {
    #[serde(deserialize_with = "rust_decimal::serde::arbitrary_precision::deserialize")]
    value: Decimal,
}

This will ensure that serialization still occurs as a string.

Please see the examples directory for more information regarding serde_json scenarios.

std

Enable std library support. This is enabled by default, however in the future will be opt in. For now, to support no_std libraries, this crate can be compiled with --no-default-features.

Building

Please refer to the Build document for more information on building and testing Rust Decimal.

Minimum Rust Compiler Version

The current minimum compiler version is 1.60.0 which was released on 2022-04-07.

This library maintains support for rust compiler versions that are 4 minor versions away from the current stable rust compiler version. For example, if the current stable compiler version is 1.50.0 then we will guarantee support up to and including 1.46.0. Of note, we will only update the minimum supported version if and when required.

Comparison to other Decimal implementations

During the development of this library, there were various design decisions made to ensure that decimal calculations would be quick, accurate and efficient. Some decisions, however, put limitations on what this library can do and ultimately what it is suitable for. One such decision was the structure of the internal decimal representation.

This library uses a mantissa of 96 bits made up of three 32-bit unsigned integers with a fourth 32-bit unsigned integer to represent the scale/sign (similar to the C and .NET Decimal implementations). This structure allows us to make use of algorithmic optimizations to implement basic arithmetic; ultimately this gives us the ability to squeeze out performance and make it one of the fastest implementations available. The downside of this approach however is that the maximum number of significant digits that can be represented is roughly 28 base-10 digits (29 in some cases).

While this constraint is not an issue for many applications (e.g. when dealing with money), some applications may require a higher number of significant digits to be represented. Fortunately, there are alternative implementations that may be worth investigating, such as:

If you have further questions about the suitability of this library for your project, then feel free to either start a discussion or open an issue and we'll do our best to help.