Are wchar_t and char16_t the same thing on Windows?

Question

I have an instance of std::u16string, can I pass its c_str() to a Win32 APIs which expects LPCWSTR, without any kind of conversion? For example, can I safely do this:

auto u16s = std::u16string(u"Hello");
::SetWindowTextW(hWnd, reinterpret_cast<LPCWSTR>(u16s.c_str()));    

Updated, MSDN says here wchar_t is UTF-16LE, while char16_t is just UTF-16 with no endian specified. Is it safe to assume then that char16_t is also always UTF-16LE on Windows? Or would that be MSVC compiler specific, and so it can possibly be UTF-32LE (or maybe UTF-16BE) if I complile with GCC, for example?

Answer 1

I would like to make amends to @jamesdlin's anwser though his answer is correct.

Before C++11, there are char and wchar_t, and hence specialize std::basic_string<> to std::string and std::wstring.

However, the width in bits of wchar_t is platform-specific: on Windows it is 16-bit while on other platforms, its 32-bit.

And with the advent of C++11, the standard adds char16_t to represent 16-bit wide characters; thus on Windows, std::u16string happens to be interchangable with std::wstring in most contexts, because they are both able to represent 16-bit wide characters.

The wchar_t type is an implementation-defined wide character type. In the Microsoft compiler, it represents a 16-bit wide character used to store Unicode encoded as UTF-16LE, the native character type on Windows operating systems.

But the newest MSDN seems to add some aside notes for code using std::wstring yet intend to be portable:

The size of wchar_t is implementation-defined. If your code depends on wchar_t to be a certain size, check your platform's implementation (for example, with sizeof(wchar_t)). If you need a string character type with a width that is guaranteed to remain the same on all platforms, use string, u16string, or u32string.

As for LE(little-endian), it should be architecture-specific, IIRC. and most architectures today use LE.