/* utf8_scripts - demonstrates UTF8 mode: the subject is decoded into
 * Unicode code points before matching (not just treated as opaque bytes,
 * the way ASCII/BINARY mode do), so \w and IGNORECASE work correctly
 * across scripts, not only ASCII letters, and match positions come in two
 * different units depending on which accessor is used.
 *
 * The subject deliberately mixes Latin (with a combining-free accented
 * letter), Greek, Cyrillic, and a CJK ideograph, each of which encodes to
 * a different number of UTF-8 bytes per code point (1, 2, 2, and 3
 * respectively), to make the code-point/byte distinction concrete rather
 * than something that happens to not matter for the specific example
 * chosen.
 */
#include "regexx.h"
#include <stdio.h>
#include <string.h>

typedef struct { int n; } Ctx;

static void print_word(void *ctx, const Match *m_const) {
    Ctx *c = ctx;
    Match *m = (Match *)m_const;
    const char *g; size_t glen;
    Match_group(m, NULL, 0, &g, &glen);
    int64_t cs, ce, bs, be;
    Match_span(m, 0, &cs, &ce);
    Match_span_byte(m, 0, &bs, &be);
    printf("word %d: '%.*s'  codepoints=[%lld,%lld) (len %lld)  bytes=[%lld,%lld) (len %lld)\n",
           c->n, (int)glen, g,
           (long long)cs, (long long)ce, (long long)(ce - cs),
           (long long)bs, (long long)be, (long long)(be - bs));
    c->n++;
}

int main(void) {
    /* "café" (Latin, e-acute is 2 bytes) + "  " + "Ελλάδα" (Greek) + "  "
     * + "Россия" (Cyrillic) + "  " + "日本" (CJK, 3 bytes per character). */
    const char *subject =
        "caf\xc3\xa9  "
        "\xce\x95\xce\xbb\xce\xbb\xce\xac\xce\xb4\xce\xb1  "
        "\xd0\xa0\xd0\xbe\xd1\x81\xd1\x81\xd0\xb8\xd1\x8f  "
        "\xe6\x97\xa5\xe6\x9c\xac";

    const char *pattern = "\\w+";
    PatternError err; memset(&err, 0, sizeof err);
    Pattern *pat = re_compile(pattern, strlen(pattern), UTF8, &err);
    if (!pat) {
        fprintf(stderr, "compile error: %s\n", err.msg);
        PatternError_free(&err);
        return 1;
    }

    Input *in = Input_from_buffer((const uint8_t *)subject, strlen(subject));
    printf("subject is %zu bytes\n\n", strlen(subject));
    Ctx ctx = { 0 };
    int n = Pattern_finditer(pat, in, 0, -1, print_word, &ctx);
    printf("\ntotal words: %d\n", n);
    printf("\\w under UTF8 mode consults the platform's Unicode tables (glibc\n"
           "wctype.h under the C.utf8 locale, docs/API.md Section 2), so it\n"
           "recognizes Greek, Cyrillic, and CJK letters, not only ASCII ones;\n"
           "the byte length of each word differs from its codepoint length\n"
           "exactly where the script needs more than one byte per character.\n");

    Input_free(in);
    Pattern_free(pat);
    return 0;
}