decimal128

# Changelog All notable changes to this project will be documented in this file. The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/), and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html). ## [Unreleased] ## [22.0.0] ### Changed - We now emit a single JS file in the `dist` directory rather than several JS module files. No need for ESM or CJS subdirectories anymore. ## [21.2.0] ### Changed - Use [JSBI](https://www.npmjs.com/package/jsbi) in favor of native JS bigints (thanks @amoshydra) - Fix exports in package.json (thanks @amoshydra) ## [21.1.0] ### Removed - Some dead code has been deleted. ## [20.0.0] ### Changed - We now emit CommonJS and ESM modules. The ESM and CJS are now found, respecively, in the `dist/esm` and `dist/cjs` directories. ## [19.0.0] ### Changed - Updated dependencies. No user-visible changes are expected. ## [18.0.0] ### Removed - Support for quantum/precision, including support for trailing zeroes. This is a breaking change if you're using these features. ## [17.0.0] ### Added ## [16.0.0] The purpose of these changes is to align the API with the official spec. ## [15.0.0] - 2024-05-07 - Use the official IEEE 754 rounding names rather than "trunc", "ceil", etc. This is a breaking change if you're using those rounding modes. If not, you shouldn't see any change. ## [14.1.0] - 2024-05-06 ### Added - Support for converting decimals to BigInts (`toBigInt`) and Numbers (`toNumber`). - Relax the constructor to allow all Number arguments (in addition to BigInts and strings). ## [14.0.0] - 2024-04-30 ### Removed - Unsupported/inofficial rounding modes. We support the official IEEE Decimal128 ones: trunc, ceil, floor, halfEven, halfUp. ## [13.0.0] - 2024-04-24 ### Removed - Rounding options for arithmetic operations. The rounding mode is now always, implicitly, `halfEven`. Rounding is still supported in the constructor. ## [12.3.0] - 2024-04-16 ### Added - `compare` method for comparting Decimal128 objects as digit strings, not as mathematical values (models IEEE 754's `compare_total` operation) ## [12.2.0] - 2024-04-15 ### Changed - The remainder operation no longer takes an optional argument for specifying the rounding mode. It is unnecessary. ## [12.1.0] - 2024-04-11 ### Added - New negation operator `neg` ## [12.0.0] - 2024-04-10 ### Removed - The `normalize` option for `lessThan` and `equals` has been removed. The comparison is now always done on the mathematical value. ## [11.2.0] - 2024-03-29 ### Changed - The Decimal128 constructor now accepts BigInt arguments as well as safe integer Numbers ## [11.1.0] - 2024-03-29 ### Changed - For `lessThan` and `equals`: the property of the (optional) second argument `total` has been renamed to `normalize` ## [11.0.0] - 2024-03-26 ### Added - `valueOf` throws unconditioanlly to prevents any kind of mixing of Decimal128 objects with other JS values - `lessThan` to determine whether a Decimal128 is strictly less than another Decimal128 - `equals` to determine whether a Decimal128 value is mathematically equal to another one ### Removed - `cmp` has been removed in favor of `lessThan` and `equals` for a nicer API ## [10.2.1] - 2024-03-22 No code changes. A few more examples of `cmp` in action were added to the test suite. ## [10.2.0] - 2024-03-21 ### Added - `abs` method for computing absolute value ## [10.0.0] - 2024-01-30 ### Changed - Added an option to `cmp` to normalize values values before comparison (default is `true`, i.e., compare mathematical values). ### Removed - `normalize` method (can now be safely defined in user space as `new Decimal128(x.toString())`. - Option to normalize strings in the constructor (we now accept the given digit string as-is) ## [9.1.0] - 2024-01-26 ### Added - Restore possibility to specify rounding mode in arithmetic operations and constructor (#72) ## [9.0.0] - 2024-01-19 ### Changed - Addition, subtraction, multiplication, division, and remainder are now static methods. This is a breaking change. ## [8.0.0] - 2024-01-19 ### Removed - `sqrt` - `pow` - `reciprocal` - `truncate`, `floor`, and `ceil` (use `round` instead) ## [7.1.0] - 2023-12-29 ### Added - Rounding modes for all operations (#54) ## [7.0.1] - 2023-12-21 ### Added - LICENSE file ### Fixed - Ensure that the NPM package is nice and lean (#57) by removing unnecessary files and dependencies ## [7.0.0] - 2023-12-21 ### Added - `toExponentialString`: Output a sign (`+` or `-`) in the exponent part. This is a breaking change. - [Fused multiply-and-add](https://en.wikipedia.org/wiki/Multiply–accumulate_operation#Fused_multiply–add) (`multiplyAdd`) ## [6.2.0] - 2023-12-18 ### Added - Square root operation ## [6.1.0] - 2023-12-11 ### Changed - Internal change in the representation of not-a-number and +/- infinity. No user-visible changes. ## [6.0.0] - 2023-12-08 ### Added - We now support non-normalized decimals, i.e. decimals with a non-zero integer part. This is a breaking change for some use cases. In this approach, trailing zeros are not removed; they are considered significant. For instance, `0.25` plus `0.75` is now `1.00`, not `1`. ## [5.3.0] - 2023-10-25 ### Added - `round` now takes a number argument (default 0) to specify the index of the decimal digit after which rounding should take place (#45) ## [5.2.0] - 2023-10-25 ### Added - `pow` for raising a decimal to a power (#43)