Posix Regex: Using capture groups to replace missing positive lookaheads

regex-example.c

#include <stdio.h>
#include <stdlib.h>
#include <regex.h>
#include <sys/types.h>
#include <assert.h>

static void regex_must_compile(regex_t* regex, const char* pattern, int flags);

int main() {
    const char* input = " \t a b c /* foo bar\nkk";
    printf("input:\n\"%s\"\n\n", input);

    regex_t r = {0};
    regex_must_compile(&r, "^(.*?)(/\\*)", REG_EXTENDED);
    const size_t num_capture_groups = 2;

    const size_t nmatches = num_capture_groups + 1;
    regmatch_t* matches = calloc(nmatches, sizeof(regmatch_t));
    assert(matches);

    if (regexec(&r, input, nmatches, matches, 0) == REG_NOMATCH) {
        puts("no match");
        free(matches);
        regfree(&r);
        return 1;
    }

    puts("match!");
    for (size_t i = 0; i < nmatches; i++) {
        regoff_t start = matches[i].rm_so;
        regoff_t end = matches[i].rm_eo;
        if (end <= start) {
            continue;
        }
        
        printf("match #%zu: (%02d, %02d):\t\"", i, start, end);
        for (regoff_t offset = start; offset < end; offset++) {
            putchar(input[offset]);
        }
        puts("\"");
    }

    free(matches);
    regfree(&r);
    return 0;
}

static void regex_must_compile(regex_t* regex, const char* pattern, int flags)
{
    int rc = regcomp(regex, pattern, flags);
    if (rc != 0) {
        // When we call regerror with 0, 0 as last parameters, it returns the error msg size
        // (already considering the '\0')
        const size_t error_msg_size = regerror(rc, regex, 0, 0);
        char* error_msg = calloc(error_msg_size, sizeof(char));
        if (!error_msg) {
            fprintf(stderr, "error: regex: compilation failed. pattern=\"%s\"\n", pattern);
            exit(EXIT_FAILURE);
        }

        // We call it once again, now with the error_msg buffer to be set
        regerror(rc, regex, error_msg, error_msg_size);

        fprintf(stderr, "error: regex: compilation failed: %s. pattern=\"%s\"\n", error_msg, pattern);
        
        free(error_msg);
        exit(EXIT_FAILURE);
    }
}