When the character string passed to C ++ by JNA is garbled

Case study

When passing Java String type to C ++ with JNA, if it is received as Char * type and assigned to std: string type, it will be garbled.

Character code problem

It seems that ** UTF-16 ** is used in Java, and it seems that ** Shift-JIS ** was required in my C ++ environment this time.

In C ++, it seems that the character code that can be handled by each type changes depending on the environment. Thank you @yumetodo.

I'm guessing that the conversion wasn't working and I got an error.

solution

I have quoted a part of the code here. http://nekko1119.hatenablog.com/entry/2017/01/02/054629

use.cpp

#include <iostream>
#include <string>
#include <codecvt>
#include <cassert>
#include <locale>

void useString(char* str){
  std:string strUtf;
  strUtf = str;
  /* uft-Shift from 8-Convert to JIS*/
  strShiftJis = utf8_to_multi_cppapi(strUtf);
  /*Please like from here on*/
}

std::string utf8_to_multi_cppapi(std::string const& src)
{
  auto const wide = utf8_to_wide_cppapi(src);
  return wide_to_multi_capi(wide);
}

std::string wide_to_multi_capi(std::wstring const& src)
{
  std::size_t converted{};
  std::vector<char> dest(src.size() * sizeof(wchar_t) + 1, '\0');
  if (::_wcstombs_s_l(&converted, dest.data(), dest.size(), src.data(), _TRUNCATE, ::_create_locale(LC_ALL, "jpn")) != 0) {
    throw std::system_error{ errno, std::system_category() };
  }
  return std::string(dest.begin(), dest.end());
}

std::wstring utf8_to_wide_cppapi(std::string const& src)
{
  std::wstring_convert<std::codecvt_utf8_utf16<wchar_t>> converter;
  return converter.from_bytes(src);
}

Conclusion

I'm glad I learned about character codes.

References

http://marupeke296.com/CPP_charUnicodeWideChar.html http://nekko1119.hatenablog.com/entry/2017/01/02/054629 https://cpprefjp.github.io/reference/codecvt/codecvt_utf8_utf16.html https://theolizer.com/cpp-school1/cpp-school1-18/