Function WideStringToUtf8

rtpose_wrapper/src/gtest/gtest-all.cpp:2807–2829

Converts a wide string to a narrow string in UTF-8 encoding. The wide string is assumed to have the following encoding: UTF-16 if sizeof(wchar_t) == 2 (on Windows, Cygwin, Symbian OS) UTF-32 if sizeof(wchar_t) == 4 (on Linux) Parameter str points to a null-terminated wide string. Parameter num_chars may additionally limit the number of wchar_t characters processed. -1 is used when the entire strin

// and contains invalid UTF-16 surrogate pairs, values in those pairs
// will be encoded as individual Unicode characters from Basic Normal Plane.
String WideStringToUtf8(const wchar_t* str, int num_chars) {
  if (num_chars == -1)
    num_chars = static_cast<int>(wcslen(str));

  ::std::stringstream stream;
  for (int i = 0; i < num_chars; ++i) {
    UInt32 unicode_code_point;

    if (str[i] == L'\0') {
      break;
    } else if (i + 1 < num_chars && IsUtf16SurrogatePair(str[i], str[i + 1])) {
      unicode_code_point = CreateCodePointFromUtf16SurrogatePair(str[i],
                                                                 str[i + 1]);
      i++;
    } else {
      unicode_code_point = static_cast<UInt32>(str[i]);
    }

    char buffer[32];  // CodePointToUtf8 requires a buffer this big.
    stream << CodePointToUtf8(unicode_code_point, buffer);
  }
  return StringStreamToString(&stream);
}

// Converts a wide C string to a String using the UTF-8 encoding.
// NULL will be converted to "(null)".