How to create multibyte characters in C

Question

During my study of character encoding in C and C++ I came across two general ways of encoding: multibyte characters and wide characters. In order to strengthen my understanding of those systems (benefits and drawbacks) I wanted to do some examples. Doing examples with wide characters is not a problem due to the native support with the wchar_t type. But when I wanted to create a string which contains those so called multibyte characters I came to a problem.

How can I actually create a multibyte character string which uses an encoding that works with a char array (using Visual C++)? This kind of encoding sure does exist: http://www.gnu.org/software/libc/manual/html_node/Shift-State.html. But I read only about it and never saw an actual example. Or do you have to create your own encoding for this kind of string?

Answer 1

If you are able to create a wide character string literal, simply omitting the L should give you a multibyte character string literal with an implementation defined encoding (gcc has an option to chose it, I don't know about visual C++).

If you have a wide character string, you can get the equivalent multibyte string according to the C locale using the functions wcstombs (in <stdlib.h>) and wcsrtombs (in <wchar.h>).

C++ locale system also provides a way to do that conversion. (Look for the in and out member of the codecvt facet, I won't provide here a tutorial on their use, the site cppreference has example codes, for instance for out).

I'm not sure you'll be able to find easily support either on Unix or on Windows for an encoding with a shift state. You should search for encoding for China, Japan, Korea, Vietman (for instance ISO 2022-JP, but it seems to me that Unix tend to use EUC-JP instead and Windows Shift JIS).