Version 0.0.2

CHANGELOG.md

@@ -6,4 +6,8 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to the
 [Haskell Package Versioning Policy](https://pvp.haskell.org/).
 
-## Unreleased
+## [0.0.2] - 2024-06-26
+### Added
+* Basic functionality exposed via `Decimal (m :: Type) (s :: Nat)` type
+* Test suite
+* Benchmark suite

README.md

@@ -4,7 +4,21 @@
 [![Test Coverage](https://coveralls.io/repos/github/EduardSergeev/fixed-decimal/badge.svg)](https://coveralls.io/github/EduardSergeev/fixed-decimal)
 [![Documentation](https://eduardsergeev.github.io/fixed-decimal/haddock.svg)](https://eduardsergeev.github.io/fixed-decimal/haddock/)
 
-[Fixed-precision decimals](https://en.wikipedia.org/wiki/Fixed-point_arithmetic) for Haskell
+Fixed-precision decimals for Haskell
+
+# The purpose of this package
+
+To provide simple high-performance [fixed-precision decimals](https://en.wikipedia.org/wiki/Fixed-point_arithmetic) data types for Haskell.  
+
+High-performance in this implementation is achieved because:
+
+1. The implementation is fixed-precision on type-level and is never recalculated so internally only fast integral-arithmetic is used;
+1. It never checks for arithmetic overflows or precision overflow;
+1. It does not do any rounding of any sort.
+
+It is assumed that the user of this package will choose an appropriate (large enough) [Integral](https://hackage.haskell.org/package/base/docs/Prelude.html#t:Integral) type to store mantissa and will specify an appropriate scale (the length of the fractional part) to accommodate any possible values user's code can produce.
+
+As a result the [performance of this implementation](https://eduardsergeev.github.io/fixed-decimal/ubuntu-latest/results.html), especially for smaller `Integral` mantissas is in-par (or faster) than standard `Double` and much faster than arbitrary precision decimal implementations. 
 
 # How to use it
 
@@ -48,7 +62,7 @@ To use mantissa of arbitrary length use `Integer`:
 0.33333333333333333333333333333333333333333333333333
 ```
 
-# Why
+# Fixed-precision vs floating point numeric types
 
 In comparison to floating-point numeric data types like `Float` or `Double` fixed-precision decimals can store decimal numbers with exact precision.