Type conversion c++, quadmath

Question

The quadmath library does not have much documentation online and I would like to perform two simple actions using __complex128 type.

Is there a way to easily print the value inside a variable of this type?
How can I convert back a variable from __complex128 to complex<double>, or even the single __float128 real and imaginary parts to double?

The second question is more important because if I can convert it I can simply use cout!

EDIT: Everything comes to the following question. How do I convert __float128 to double?

These links might help: [\[1\]](https://gcc.gnu.org/ml/gcc/2016-07/msg00120.html) [\[2\]](https://stackoverflow.com/a/41776202/3282436) — 0x5453, Jun 27 '17 at 13:28
Thank you for the links but it gives me an error: undefined reference to quadmath_snprintf. However I need mostly the conversion — NSZ, Jun 27 '17 at 13:37
@NSZ: That's s linker error. A wild guess would be that you have to link with libquadmath (try `-lquadmath`). — You, Jun 27 '17 at 15:20
@You: Thanks for the answer. Sorry for my newbeness, but where do I put this command ( I am on windows). — NSZ, Jun 27 '17 at 15:48
@NSZ: Wherever linker flags go, which depends on what compiler/build system you use. — You, Jun 27 '17 at 16:32

score 2 · Accepted Answer · answered Jun 27 '17 at 15:18

Converting a __float128 to a double does not require anything special:

const __float128 x{};
const double y = x;

You might want a static_cast for clarity (and less warnings).

Constructing a std::complex<T> from __complex128 is also not difficult:

auto bar(__complex128 y)
{
    return std::complex<double>{
        static_cast<double>(__real__ y),
        static_cast<double>(__imag__ y)
    };
}

gsamaras · Answer 2 · 2017-06-27T16:08:46.623

0

Try this:

std::complex<double> x;
__complex128 y;
x.real(__real__ y)
x.imag(__imag__ y)

edited Jun 27 '17 at 16:08

answered Jun 27 '17 at 13:54

gsamaras

It doesn't work. The error message is: "error: invalid use of member (did you forget the '&' ?)" – NSZ Jun 27 '17 at 13:59
[`real`](http://en.cppreference.com/w/cpp/numeric/complex/real) and [`imag`](http://en.cppreference.com/w/cpp/numeric/complex/imag) are member _functions_ of `std::complex`. – You Jun 27 '17 at 15:09
Probably need `x.real(__real__ y)` etc. – n. m. could be an AI Jun 27 '17 at 15:22
Updated the answer per @n.m.'s comment. Thank you guys. – gsamaras Jun 27 '17 at 16:09

2 Answers2