FixedPoint

FixedPoint provides a simple decimal fixed point type for the D language.

Usage

The Fixed struct provides all the functionality needed. It behaves like other numerical types. Binary operations between two Fixed numbers will always result into a new Fixed. This will have the precision and type of the higher or bigger operand respectively.

import std.conv : to;
import fixedpoint.fixed : Fixed;

alias MyFixed = Fixed!4;

void main()
{
    auto f1 = MyFixed("21.5");
    assert(f1.toString == "21.5000");
    assert(f1.to!int == 21);
    assert(f1.to!double == 21.5);
    assert(f1 + 1 == MyFixed("22.5"));
    assert(f1 + 1 == MyFixed("22.5"));
    auto f2 = MyFixed("20.5");
    assert(f1 > f2);
    assert(f1+f2 == MyFixed("42"));
}

Related Work

There are similar packages which provide some kind of decimal or fixed-point types and arithmetic. However they don't seem to be maintained, or are too big for my purposes. There is fixed, which is very similar and inspired this project. stdxdecimal seems to have same goals as well. For arbitrary precision, you might want to try bigfixed. decimal is the IEEE-754-2008 compliant decimal data type.

Licence

FixedPoint is released with the Boost license (like most D projects). See here for more details.