std::imag(std::complex) - cppreference.com

From cppreference.com

Defined in header `<complex>`
template< class T > T imag( const std::complex<T>& z );	(1)	(until C++14)
template< class T > constexpr T imag( const std::complex<T>& z );	(1)	(since C++14)
Additional overloads (since C++11)
Defined in header `<complex>`
float imag( float f ); double imag( double f ); long double imag( long double f );	(A)	(until C++14)
constexpr float imag( float f ); constexpr double imag( double f ); constexpr long double imag( long double f );		(since C++14) (until C++23)
template< class FloatingPoint > FloatingPoint imag( FloatingPoint f );		(since C++23)
template< class Integer > double imag( Integer i );	(B)	(until C++14)
template< class Integer > constexpr double imag( Integer i );	(B)	(since C++14)

1) Returns the imaginary part of the complex number z, i.e. z.imag().

A,B) Additional overloads are provided for all integer and floating-point types, which are treated as complex numbers with zero imaginary part.

(since C++11)

1) The imaginary part of z.

A) decltype(f){} (zero).

B) 0.0.

The additional overloads are not required to be provided exactly as (A,B). They only need to be sufficient to ensure that for their argument num:

If num has a standard(until C++23) floating-point type T, then std::imag(num) has the same effect as std::imag(std::complex<T>(num)).
Otherwise, if num has an integer type, then std::imag(num) has the same effect as std::imag(std::complex<double>(num)).