std::sqrt(std::valarray)

template< class T >
valarray<T> sqrt( const valarray<T>& va );

For each element in va computes the square root of the value of the element.

Parameters

Return value

Value array containing square roots of the values in va.

Notes

Unqualified function (sqrt) is used to perform the computation. If such function is not available, std::sqrt is used due to argument-dependent lookup.

The function can be implemented with the return type different from std::valarray. In this case, the replacement type has the following properties:

• All const member functions of std::valarray are provided.
• std::valarray, std::slice_array, std::gslice_array, std::mask_array and std::indirect_array can be constructed from the replacement type.
• All functions accepting an argument of type const std::valarray& except begin() and end() (since C++11) should also accept the replacement type.
• All functions accepting two arguments of type const std::valarray& should accept every combination of const std::valarray& and the replacement type.
• The return type does not add more than two levels of template nesting over the most deeply-nested argument type.

Possible implementation

template< class T >
valarray<T> sqrt( const valarray<T>& va )
{
    valarray<T> other = va;
    for (T &i : other) {
        i = sqrt(i);
    }
    return other; // proxy object may be returned
}

Example

#include <cstddef>
#include <valarray>
#include <iostream>
 
int main()
{
    std::valarray<double> a(1, 8);
    std::valarray<double> b{1, 2, 3, 4, 5, 6, 7, 8};
    std::valarray<double> c = -b;
    // literals must also be of type T until LWG3074 (double in this case)
    std::valarray<double> d = std::sqrt(b * b - 4.0 * a * c);
    std::valarray<double> x1 = (-b - d) / (2.0 * a);
    std::valarray<double> x2 = (-b + d) / (2.0 * a);
    std::cout << "quadratic equation:  root 1:    root 2:\n";
    for (std::size_t i = 0; i < a.size(); ++i) {
        std::cout << a[i] << "\u00B7x\u00B2 + " << b[i] << "\u00B7x + "
                  << c[i] << " = 0  " << std::fixed << x1[i] << "  "
                  << x2[i] << std::defaultfloat << '\n';
    }
}

quadratic equation:  root 1:    root 2:
1·x² + 1·x + -1 = 0  -1.618034  0.618034
1·x² + 2·x + -2 = 0  -2.732051  0.732051
1·x² + 3·x + -3 = 0  -3.791288  0.791288
1·x² + 4·x + -4 = 0  -4.828427  0.828427
1·x² + 5·x + -5 = 0  -5.854102  0.854102
1·x² + 6·x + -6 = 0  -6.872983  0.872983
1·x² + 7·x + -7 = 0  -7.887482  0.887482
1·x² + 8·x + -8 = 0  -8.898979  0.898979

See also