const-decimal
is a fixed precision decimal backed by an underlying integer.
This means it can compute Add/Sub/Mul/Div* operations within it's range without
precision loss. This crate was created as there was no pre-existing crate that
offered these features for decimal numbers (fixed
cannot represent decimals
with precision).
This is a fairly simple crate and is mainly intended to serve as a lossless representation of decimal values. While math operations are supported it is not intended to be a general purpose maths library.
* Division can of course truncate like normal division.
The goals in order of priority are:
- Safety: Panic on over/underflow (even in release).
- Precision: Use integers instead of floats.
- Performance: Use primitive underlying types were possible.