/*
 * sample-utf8.c: non-ASCII text in C string literals.
 * Free sample from 99helpers.com/file-types/c
 *
 * Saved as UTF-8 without a byte order mark. GCC and Clang read source files
 * as UTF-8 by default and copy the bytes into the program unchanged.
 * Microsoft's compiler assumes the system code page instead, unless you add
 * /utf-8:
 *     cc -Wall -Wextra -std=c17 sample-utf8.c -o sample-utf8 && ./sample-utf8
 *     cl /W4 /std:c17 /utf-8 sample-utf8.c
 * On Windows, run `chcp 65001` first so the console shows UTF-8 output.
 */
#include <stdio.h>
#include <string.h>

/* strlen counts bytes, not characters. In UTF-8 every character starts with
 * a byte that is not of the form 10xxxxxx, so counting those bytes counts
 * code points (what a reader would mostly call characters). */
static size_t utf8_length(const char *s)
{
    size_t n = 0;
    for (; *s != '\0'; s++) {
        if (((unsigned char)*s & 0xC0) != 0x80) {
            n++;
        }
    }
    return n;
}

int main(void)
{
    static const char *const greetings[] = {
        "Hello",            /* English: 1 byte per letter */
        "Grüß Gott",        /* German: ü and ß take 2 bytes each */
        "Bonjour, ça va ?", /* French */
        "Привет",           /* Russian: Cyrillic, 2 bytes each */
        "Γειά σου",         /* Greek */
        "שלום",             /* Hebrew, written right to left */
        "مرحبا",            /* Arabic, written right to left */
        "こんにちは",       /* Japanese: 3 bytes each */
        "你好",             /* Chinese */
        "€ 12,50 · ½ · ±",  /* symbols */
        "👋🙂",             /* emoji: 4 bytes each */
    };

    for (size_t i = 0; i < sizeof greetings / sizeof greetings[0]; i++) {
        printf("%2zu bytes, %2zu characters: %s\n", strlen(greetings[i]),
               utf8_length(greetings[i]), greetings[i]);
    }
    return 0;
}