Convert const char* to const wchar_t*

Question

I am trying to create a program with Irrlicht that loads certain things from a configuration file written in Lua, one of which is the window title. However, the lua_tostring function returns a const char* while the Irrlicht device's method setWindowCaption expects a const wchar_t*. How can I convert the string returned by lua_tostring?

Rorry

Answer 1

There are multiple questions on SO that address the problem on Windows. Sample posts:

1. char* to const wchar_t * conversion
2. conversion from unsigned char* to const wchar_t*

There is a platform agnostic method posted at http://ubuntuforums.org/showthread.php?t=1579640. The source from this site is (I hope I am not violating any copyright):

#include <locale>
#include <iostream>
#include <string>
#include <sstream>
using namespace std ;

wstring widen( const string& str )
{
    wostringstream wstm ;
    const ctype<wchar_t>& ctfacet = use_facet<ctype<wchar_t>>(wstm.getloc()) ;
    for( size_t i=0 ; i<str.size() ; ++i ) 
              wstm << ctfacet.widen( str[i] ) ;
    return wstm.str() ;
}

string narrow( const wstring& str )
{
    ostringstream stm ;

    // Incorrect code from the link
    // const ctype<char>& ctfacet = use_facet<ctype<char>>(stm.getloc());

    // Correct code.
    const ctype<wchar_t>& ctfacet = use_facet<ctype<wchar_t>>(stm.getloc());

    for( size_t i=0 ; i<str.size() ; ++i ) 
                  stm << ctfacet.narrow( str[i], 0 ) ;
    return stm.str() ;
}

int main()
{
  {
    const char* cstr = "abcdefghijkl" ;
    const wchar_t* wcstr = widen(cstr).c_str() ;
    wcout << wcstr << L'\n' ;
  }
  {  
    const wchar_t* wcstr = L"mnopqrstuvwx" ;
    const char* cstr = narrow(wcstr).c_str() ;
    cout << cstr << '\n' ;
  } 
}

Patronizing

Generally, I've found that dealing with wide text in C++ in Ubuntu requires setting the default locale. It's very ironic how implementations behave. With UTF-8 as in Unix-land the locale hardly matters for narrow-wide conversion, but must be set, while with various single-byte encodings such as in Windows the locale very much matters, but is already set by default. – Saker

Answer 2

You can use mbstowcs:

    wchar_t WBuf[100];
    mbstowcs(WBuf,lua_tostring( /*...*/ ),99);

or more safe:

    const char* sz = lua_tostring(/*...*/);
    std::vector<wchar_t> vec;
    size_t len = strlen(sz);
    vec.resize(len+1);
    mbstowcs(&vec[0],sz,len);
    const wchar_t* wsz = &vec[0];

Ohg

Answer 3

For Unicode:

std::string myString = "Master James";
const char* sz = myString.c_str();
size_t origsizes = strlen(sz) + 1;
const size_t newsizes = 500;
size_t convertedCharP = 0;
wchar_t constTowchar[500];
mbstowcs_s(&convertedCharP, constTowchar, origsizes, sz, _TRUNCATE);
std::wcout << constTowchar << std::endl;

This is working using mbstowcs_s.

Empoverish

Answer 4

Here is a faster version of the code from the first answer, do not forget about std::locale::global(std::locale("")); for some characters

#include <xlocale>
#include <sstream>

[[nodiscard]] inline std::wstring widen(const char* str)
{
    const auto length = strlen(str);

    std::wstring result;
    result.resize(length);

    const auto& facet = use_facet<std::ctype<wchar_t>>(std::wostringstream().getloc());
    std::transform(str, str + length, result.begin(), [&facet](const char ch)
    {
        return facet.widen(ch);
    });
    return result;
}

[[nodiscard]] inline std::string narrow(const wchar_t* str)
{
    const auto length = wcslen(str);

    std::string result;
    result.resize(length);

    const auto& facet = use_facet<std::ctype<wchar_t>>(std::ostringstream().getloc());
    std::transform(str, str + length, result.begin(), [&facet](const wchar_t ch)
    {
        return facet.narrow(ch);
    });
    return result;
}

Patron

While your answer may answer the question, "code only" answers are not accepted. Please add title and some explanation for what's happening. stackoverflow.com/help/how-to-answer for more information. – Calcine