Question

Here is the method to convert from managed string to native char pointer, it’ll return error code 42 when the input string is “€” (ASCII 128d):

void StringUtility::ManagedToNative(String^ source, char* pTarget, Int32 targetLength)
{
    if(pTarget == NULL)
        throw gcnew System::ArgumentNullException("The source pointer cannot be empty.");
    if(targetLength <= 0)
        throw gcnew System::ArgumentOutOfRangeException("The target length has to be larger than 0.");

    memset(pTarget, 0, targetLength);
    if(String::IsNullOrEmpty(source))
    {
        pTarget[0] = '\0';
        return;
    }

    // Conversion to char* :
    size_t convertedChars = 0;
    size_t sizeInBytes = targetLength;
    size_t count = (source->Length > targetLength) ? _TRUNCATE : targetLength;
    errno_t err = 0;
    {
        /* minimize the scope of pined object */
        pin_ptr<const wchar_t> wch = PtrToStringChars(source);
        err = wcstombs_s(&convertedChars, pTarget, sizeInBytes, wch, count);
    }

    // if truncate did happen and it's intended, return as successful.
    if( count == _TRUNCATE && err == STRUNCATE )
        return;
    if (err != 0)
        throw gcnew System::InvalidOperationException("convert from String^ to char* failed");
}  

Is it just because I can’t convert any string (with bytes >= 128) using wcstombs_s? (see this)

Answer 1

From MSDN doc about wcstombs_s:

If wcstombs_s encounters a wide character it cannot convert to a multibyte character, it puts 0 in *pReturnValue, sets the destination buffer to an empty string, sets errno to EILSEQ, and returns EILSEQ.

EILSEQ is error code 42.

To convert Unicode string to char* array, I’d suggest get the binary array of the Unicode string and encode it using Base64 and store the encoded result in a char* array.