Re: problem with std.variant rounding

[kinke via Digitalmars-d-learn]

On Friday, 28 April 2017 at 16:24:55 UTC, Suliman wrote:
import std.stdio;
import std.variant;
void main()
{
        Variant b = 56.051151;
        float x = b.coerce!float;
        writeln(x);
}

56.0512

void main()
{
    import core.stdc.stdio;
    import std.stdio;

    double d = 56.051151;
    writeln(d);
    printf("%g %f %a\n\n", d, d, d);

    real r = 56.051151L;
    writeln(r);
    printf("%Lg %Lf %La\n", r, r, r);
}

=>

56.0512
56.0512 56.051151 0x1.c068c1db0142fp+5

56.0512
56.0512 56.051151 0x1.c068c1db0142f61ep+5

So using write[ln]() to check floating-point values isn't a good 
idea as you may lose precision; hex formatting (or a proper 
debugger) is a much better choice. Additionally, your value isn't 
*exactly* representable; you may want to read up on 
floating-point representations if that's unclear.