failure to convert numpy.float64 into System.Double #1936

Environment

• Pythonnet version: 3.0
• Python version: 3.x
• Operating System: Windows/Linux
• .NET Runtime: x

Details

Passing in a numpy array to a C# function that takes a double[] parameter fails.

python code:
test_array = np.array([123, 345], np.float)
CSharpClass.Test(test_array)

C# code
public static void Test(double[] TestData) {...}

I already have a fix for this issue.

In Runtime.cs, change PyFloat_Check(...) to

    internal static bool PyFloat_Check(BorrowedReference ob)
    {
        return PyObject_TypeCheck(ob, PyFloatType);
        //return PyObject_TYPE(ob) == PyFloatType; // existing code
    }

This will convert any type that is derived from PyFloatType, which includes numpy.float64