isalpha('é') returns 0. The narrow character classification functions in <ctype.h> only handle ASCII (values 0-127); everything else is either wrong or undefined behaviour. wctype.h (C95) is the wide-character counterpart that works on the full Unicode character set via wchar_t values.

#include <wctype.h>
#include <locale.h>
 
setlocale(LC_ALL, "");   // needed for correct Unicode classification
 
iswalpha(L'é')   // non-zero: é is a letter
iswdigit(L'7')   // non-zero: decimal digit
iswspace(L' ')   // non-zero: whitespace
iswupper(L'Ä')   // non-zero: uppercase letter
iswlower(L'ö')   // non-zero: lowercase letter
iswpunct(L'!')   // non-zero: punctuation
 
towupper(L'é')   // L'É'
towlower(L'Ä')   // L'ä'

The functions mirror the <ctype.h> set with an isw prefix instead of is, and tow instead of to. Results are locale-dependent, as in <ctype.h>.

<wctype.h> also provides extensible character class testing via wctype_t:

wctype_t alpha_class = wctype("alpha");
if (iswctype(L'ñ', alpha_class))
    wprintf(L"'ñ' is alphabetic\n");

wctrans_t does the same for transformations:

wctrans_t to_upper = wctrans("toupper");
wchar_t result = towctrans(L'é', to_upper);  // L'É'

The named character classes recognised by wctype include: alnum, alpha, blank, cntrl, digit, graph, lower, print, punct, space, upper, xdigit. Locale-specific classes may extend this list.

// compile: gcc -o wcttest wcttest.c
// run: ./wcttest
// description: classify and case-convert a string of wide characters
 
#include <wctype.h>
#include <wchar.h>
#include <locale.h>
 
int main(void) {
    setlocale(LC_ALL, "");
 
    const wchar_t *s = L"héLLo 42!";
    for (const wchar_t *p = s; *p; p++) {
        wprintf(L"'%lc'  alpha=%d  digit=%d  upper=%d  upper→%lc\n",
                *p,
                iswalpha(*p) != 0,
                iswdigit(*p) != 0,
                iswupper(*p) != 0,
                towupper(*p));
    }
    return 0;
}

towupper(L'h') returns L'H'; towupper(L'é') returns L'É' — the accented uppercase form that toupper from <ctype.h> cannot produce. The digit L'4' and punctuation L'!' return themselves unchanged from towupper, as expected.