Arbitrary precision decimal numbers

A module with a simple implementation for arbitrary precision decimal numbers. The key difference to most other decimal libraries is that there is no maximum precision for example when doing divisions.

>>> from dezimal import Dezimal
>>> Dezimal("355") / Dezimal("113")
Dezimal('3.14159292035398230088495575221238938053097345132743362831858407079646017699115044247787610619469026548672566371681')

dezimal will calculate as many digits after the decimal point as there are, no ROUNDING_MODE or whatever needed.

If the calculation would result in a period, dezimal detects that and stops the calculation as needed:

>>> Dezimal("1") / Dezimal("7")
Dezimal('0.14285714285714285')

However, it prints a minimum number of digits, as to display at least the same precision as a regular float would (a bit more actually).

>>> Dezimal("1") / Dezimal("3")
Dezimal('0.33333333333333333')
>>> 1.0 / 3.0
0.3333333333333333

These parameters can be controlled by invoking the static div method in Dezimal directly:

>>> Dezimal.div(Dezimal("1"), Dezimal("3"), minscale=5)
Dezimal('0.33333')

Also a maximum precision can be given:

>>> Dezimal.div(Dezimal("355"), Dezimal("113"), maxscale=30)
Dezimal('3.141592920353982300884955752212')

Construction

An instance of Dezimal can be constructed from int, float, bool, or a str. The usual formats like 3, 3.0, 7e1, -7e-9, or +13.03e-3 are understood.

Supported operations

The following operations are supported:

+ - * / // < <= > >= == !=

Why?

For fun.