A C string can contain characters beyond the basic character set. Before printing a multibyte string such as "fär ínóff", call `setlocale(LC_ALL, "")` so character handling uses the installed locale. The locale matters: placing extended characters in a string alone is not enough to make their output work reliably. section 14.3
setlocale(LC_ALL, "");
/* Multibyte character printing only works after the locale
has been switched. */
draw_sep(TOPLEFT " © 2014 j𝜀nz ' g𝜔z,t𝜀t ", TOPRIGHT);
The listing switches the locale before printing text containing extended characters. section 14.3
A multibyte character is a sequence of bytes interpreted as one character of an extended character set. A multibyte string contains such sequences. Its characters contain no null bytes, and the string is null-terminated, so ordinary string operations such as `strcpy` can still work. TAKEAWAY: `strlen` counts bytes, not necessarily the characters that will print; neither it nor the size of a `char` array directly gives the printed length. section 14.3
To parse characters rather than copy bytes, keep conversion state in an `mbstate_t` from `<wchar.h>`. The library provides `mbsrtowcs` for conversion to `wchar_t` characters and `mbrtowc` for conversion of a multibyte character. The example `mbsrlen` uses `mbsrtowcs` to determine a converted length; it is an ad hoc function, not a standard-library function. Parsing must account for conversion status rather than assume that every byte begins a complete character. section 14.3
size_t mbsrlen(char const*s, mbstate_t const*restrict state) {
state = state ? state : MBSTATE;
mbstate_t st = *state;
size_t mblen = mbsrtowcs(nullptr, &s, 0, &st);
if (mblen == mbinvalid) errno = 0;
return mblen;
}
This implementation copies the parsing state before measuring the multibyte string with `mbsrtowcs`. section 14.3
TAKEAWAY: Prefer multibyte strings for extended-character text input and output, but distinguish byte-oriented string operations from character-oriented parsing and measurement. Unicode (ISO 10646) supplies a framework for character codes; it does not remove the need to handle multibyte conversion and its state. section 14.3 section 14.7
