Linux codecvt wide string conversion with multibyte chars and locale + concatenation, C++ (clang)

Linux codecvt wide string conversion with multibyte chars and locale + concatenation

//Microsoft (R) C/C++ Optimizing Compiler Version 19.00.23506 for x64

#include <iostream>
#include <string.h>
#include <codecvt>

// =====================================================================================================================
// Returns the length of a wchar_t based string.  This function is necessary when specifying the -fshort-wchar option
size_t  PalWcslen(
    const wchar_t* pWideStr)    ///< [in] wide string to query
{
    size_t      len = 0;

if (pWideStr != nullptr)
    {
        const wchar_t*    pWchar = pWideStr;

while (*pWchar != 0)
        {
            ++pWchar;
            ++len;
        }
    }

return len;
}

// =====================================================================================================================
// Convert UTf-16 string to UTF-8
bool ConvertUtf16StringToUtf8(
    char*          pDst,           ///< [out] dst string
    const wchar_t* pSrc,           ///< [in] src string
    size_t         dstSizeInBytes) ///< size of the destination buffer in bytes
{
    std::codecvt_utf8_utf16<char16_t>   converter;
    mbstate_t                           state = {0};

char*           pCharNext = nullptr;
    const char16_t* pUtf16Next = nullptr;

std::codecvt_base::result retCode = converter.out(state,
                                                     reinterpret_cast<const char16_t*>(pSrc),
                                                     reinterpret_cast<const char16_t*>(pSrc)+PalWcslen(pSrc)+1,
                                                     pUtf16Next,
                                                     pDst,
                                                     pDst + dstSizeInBytes,
                                                     pCharNext);
    return (retCode == std::codecvt_base::ok);
}

// =====================================================================================================================
// Convert UTF-16 or UTF-32 string (depending on wchar_t size) to UTF-8
bool ConvertWcharStringToUtf8(
    char*          pDst,           ///< [out] dst string
    const wchar_t* pSrc,           ///< [in] src string
    size_t         dstSizeInBytes) ///< size of the destination buffer in bytes
{
    bool result = false;
    if (sizeof(wchar_t) == 2)
    {
        result = ConvertUtf16StringToUtf8(pDst, pSrc, dstSizeInBytes);
    }
    else
    {
        std::codecvt_utf8<char32_t> converter;
        mbstate_t                   state = {0};

char*           pCharNext = nullptr;
        const char32_t* pUtf32Next = nullptr;

std::codecvt_base::result retCode = converter.out(state,
                                                         reinterpret_cast<const char32_t*>(pSrc),
                                                         reinterpret_cast<const char32_t*>(pSrc) + PalWcslen(pSrc) + 1,
                                                         pUtf32Next,
                                                         pDst,
                                                         pDst + dstSizeInBytes,
                                                         pCharNext);
        result = (retCode == std::codecvt_base::ok);
    }
    return result;
}

int main()
{
    const wchar_t* ppInput[] = { L"José", L"測試", L"αβ", L"Środa" };
    
    for(size_t i = 0; i < (sizeof(ppInput)/sizeof(ppInput[0])); ++i)
    {
        std::wcout << L"Original string: " << ppInput[i] << std::endl;
        
        char pOutput[256] = { '\0' };
    
        bool ret = ConvertWcharStringToUtf8(pOutput, ppInput[i], sizeof(pOutput) - 1);
                
        std::cout << "\tret = " << ret 
                  << ", output = " << pOutput << std::endl;
        
        strncat(pOutput, "-concatenated", sizeof(pOutput));
        std::cout << "\tconcatenated = " << pOutput << std::endl;
    }
}


run \| edit \| history \| help	0