Function CodePointToUtf8

rtpose_wrapper/src/gtest/gtest-all.cpp:2738–2769

Converts a Unicode code point to a narrow string in UTF-8 encoding. code_point parameter is of type UInt32 because wchar_t may not be wide enough to contain a code point. The output buffer str must containt at least 32 characters. The function returns the address of the output buffer. If the code_point is not a valid Unicode code point (i.e. outside of Unicode range U+0 to U+10FFFF) it will be out

// (i.e. outside of Unicode range U+0 to U+10FFFF) it will be output
// as '(Invalid Unicode 0xXXXXXXXX)'.
char* CodePointToUtf8(UInt32 code_point, char* str) {
  if (code_point <= kMaxCodePoint1) {
    str[1] = '\0';
    str[0] = static_cast<char>(code_point);                          // 0xxxxxxx
  } else if (code_point <= kMaxCodePoint2) {
    str[2] = '\0';
    str[1] = static_cast<char>(0x80 | ChopLowBits(&code_point, 6));  // 10xxxxxx
    str[0] = static_cast<char>(0xC0 | code_point);                   // 110xxxxx
  } else if (code_point <= kMaxCodePoint3) {
    str[3] = '\0';
    str[2] = static_cast<char>(0x80 | ChopLowBits(&code_point, 6));  // 10xxxxxx
    str[1] = static_cast<char>(0x80 | ChopLowBits(&code_point, 6));  // 10xxxxxx
    str[0] = static_cast<char>(0xE0 | code_point);                   // 1110xxxx
  } else if (code_point <= kMaxCodePoint4) {
    str[4] = '\0';
    str[3] = static_cast<char>(0x80 | ChopLowBits(&code_point, 6));  // 10xxxxxx
    str[2] = static_cast<char>(0x80 | ChopLowBits(&code_point, 6));  // 10xxxxxx
    str[1] = static_cast<char>(0x80 | ChopLowBits(&code_point, 6));  // 10xxxxxx
    str[0] = static_cast<char>(0xF0 | code_point);                   // 11110xxx
  } else {
    // The longest string String::Format can produce when invoked
    // with these parameters is 28 character long (not including
    // the terminating nul character). We are asking for 32 character
    // buffer just in case. This is also enough for strncpy to
    // null-terminate the destination string.
    posix::StrNCpy(
        str, String::Format("(Invalid Unicode 0x%X)", code_point).c_str(), 32);
    str[31] = '\0';  // Makes sure no change in the format to strncpy leaves
                     // the result unterminated.
  }
  return str;
}

// The following two functions only make sense if the the system
// uses UTF-16 for wide string encoding. All supported systems