Reading a password into a finite buffer allows shorter prefixes to match as well — Live demo at https://cplayground.com/?p=koala-phil-cod

password_match_prefix.c

// Reading a password into a finite buffer allows shorter prefixes to match as well.
// Live demo at https://cplayground.com/?p=koala-phil-cod

#include <stdio.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>


int main() {
    // The correct password is:
    constexpr char const * const EXPECTED_PASSWORD = "1234";
    // ..but also try to enter '123456'.
    
    // Since the read buffer only has length...
    constexpr size_t MAX_PASSWORD_LEN = 5;
    char password[MAX_PASSWORD_LEN];
    
    // ...we only read so many chars:
    printf("Enter password:\n");
    const ssize_t bytes_read = read(STDIN_FILENO, password, MAX_PASSWORD_LEN-1); // Leave one char for null-terminator
    if (bytes_read <= 0) {
        // errno==0 if bytes_read==0 because of EOF, but too lazy to handle it separately
        perror("read");
        return EXIT_FAILURE;
    }
    
    // Manual null-termination, because read() doesn't do it itself
    password[bytes_read] = '\0';
    
    // Remove potential \n
    password[strcspn(password, "\n")] = '\0';
    
    // The password in the buffer may be shorter than what you entered...
    printf("Password stored in buffer: %s\n", password);
    
    // ...so the comparision is true in more cases than expected
    const bool password_correct = strcmp(password, EXPECTED_PASSWORD) == 0;
    printf("correct? %s\n", password_correct? "true" : "false");
    
    
    // What did we learn? Use getline()
    // Or avoid writing C and use Rust instead :)
    return 0;
}