Displaying a wchar_c and its code point from std::wcin - c++

Question

In C++, I am trying to display the codepoint of a wchar_t retrieved from std::wcin in MessageBoxW().
My source file is encoded in UTF-8.
If I declare my wchar_t in the source of my program, and give it an initial value, I get the display of the Unicode character and its codepoint in MessageBoxW().
However, if I retrieve the wchar_t from std::wcin, the Unicode character entered is not interpreted correctly.
Can you tell me what my error is?
I compile my code with MinGW GCC version 6.3 32-bit.
Do I need to use a particular C++ option, or C++ version?
Here is the code that works:
#include <Windows.h>
#include <stdio.h>
int main()
{
wchar_t c = L'−';
wchar_t *c1 = &c;
wchar_t buff[1024];
swprintf(buff, L"The code point of %c is %d.", c1, c);
MessageBoxW(NULL, buff, L"", MB_OK);
}
Here is the code that interprets an erroneous character, but it does not give any error when compiling:
#include <Windows.h>
#include <iostream>
#include <stdio.h>
int main()
{
wchar_t c;
std::wcout << "Enter a wchar";
std::wcin >> c;
wchar_t *c1 = &c;
wchar_t buff[1024];
swprintf(buff, L"The code point of %c is %d.", c1, c);
MessageBoxW(NULL, buff, L"", MB_OK);
}

Answer 1

Finally, I got it!
Many thanks to #RemyLebeau and #n.m for their help.
I only needed the last part of the code given by #RemyLebeau.
Here is the code that works here very well with any typed character.
P.S. It's missing the verification of the size of the string entered by the user, he must enter only one character.
Any idea of correction or improvement would be much appreciated.
#include <stdio.h>
#include <Windows.h>
#include <iostream>
int main()
{
std::wcout << "Enter a wchar";
std::wstring s;
wchar_t buffer[4] = {};
DWORD numRead = 0;
if (ReadConsoleW(GetStdHandle(STD_INPUT_HANDLE), buffer, 4, &numRead, NULL))
{
s.append(buffer, numRead);
}
wchar_t buff[1024];
const wchar_t* c = s.c_str();
swprintf(buff, L"The codepoint of %s is %u.", c, *c);
MessageBoxW(NULL, buff, L"", MB_OK);
}
Best regards.