How to detect the end of file using fread() then finish the loop when reading binary file? (Do not use feof) - C language

Question

I am writing a code which reads a binary file, read each struct fields seperatly, prints out the results, then stop when the file reading reaches the end.

FYI - The binary file contains multiple records with same struct fields

I could figure out how to do it using feof. but, I would like to run this code using only fread() function(No feof). I was told to find fread return value to detect the end of the file but I am struggling how to do it.

If there are alternate ways to do it Please help me out. below is the code

#define MAX_CT 11
#include<stdio.h>
#include<stdlib.h>

//struct variables
struct test
{
    short int a;
    double b;
    short int c;
    int d;
    float e;
    unsigned short f;
    unsigned char g;
    double h;
    unsigned long i;
    short int j;
    char k;
    int l;
    int m;
    int n;
    char o[MAX_CT];
    char p;
    double q;
};

int main(int argc, char **argv)
{
    struct test a1;
    
    FILE *fp;
    
        //Input Checking Error
        if (argc < 2) {
                fprintf(stderr, "Usage: %s input_file\n", argv[0]);
                exit(1);
        }
    
        //binary file to open for reading
        fp = fopen(argv[1], "rb");
    
        //File Checking Error
        if (fp == NULL){fprintf(stderr, "Cannot open the file %s\n", 
                        argv[1]);
                        exit(1);
        }
    
    //Print the field names
    printf("a, b, c, d, e, f, g, h, i, j, k, l, m, n, o, p, q \n");
    
    //Allocate values into struct using fread
    
    while(1) 
    {
        fread(&a1.a, sizeof(a1.a), 1, fp);
        fread(&a1.b, sizeof(a1.b), 1, fp);
        fread(&a1.c, sizeof(a1.c), 1, fp);
        fread(&a1.d, sizeof(a1.d), 1, fp);
        fread(&a1.e, sizeof(a1.e), 1, fp);
        fread(&a1.f, sizeof(a1.f), 1, fp);
        fread(&a1.g, sizeof(a1.g), 1, fp);
        fread(&a1.h, sizeof(a1.h), 1, fp);
        fread(&a1.i, sizeof(a1.i), 1, fp);
        fread(&a1.j, sizeof(a1.j), 1, fp);
        fread(&a1.k, sizeof(a1.k), 1, fp);
        fread(&a1.l, sizeof(a1.l), 1, fp);
        fread(&a1.m, sizeof(a1.m), 1, fp);
        fread(&a1.n, sizeof(a1.n), 1, fp);
        fread(&a1.o, sizeof(a1.o), 1, fp);
        fread(&a1.p, sizeof(a1.p), 1, fp);
        fread(&a1.q, sizeof(a1.q), 1, fp); 
        
        //if file reading reaches the end, stop the loop 
        //(without feof, is there any other way to finish the loop using fread()????

        if(feof(fp)){
            break;
        }
        
    //Print results
    printf("%d, %f, %i, %d, %f, %d, %d, %f, %lu, %i, %i, %i, %i, %x, %s, %c, %f\n", a1.a, a1.b, 
           a1.c, a1.d, a1.e, a1.f, a1.g, a1.h, a1.i, a1.j, a1.k, a1.l, a1.m, a1.n, a1.o, a1.p, 
           a1.q);
    
    }
    
    //close the file
    fclose(fp);
        
    return 0;
}

From https://en.cppreference.com/w/c/io/fread `fread does not distinguish between end-of-file and error, and callers must use feof and ferror to determine which occurred.` — Retired Ninja, Sep 09 '21 at 04:20
You check the return from `fread()` and when it is `0`, you then check `feof()` and `ferror()` (note: when reading from a network connection, there are further steps involved) — David C. Rankin, Sep 09 '21 at 04:48
staypersistent, `if (fread(&a1.a, sizeof(a1.a), 1, fp) != 1) break; // if (....) break; ...` — chux - Reinstate Monica, Sep 09 '21 at 05:09
@chux-ReinstateMonica I tried that too. Then it does not print out any fread outputs. I even moved the if (....) break; statement after the printf statement, it gives me wrong outputs except the 1st line of the file. — staypersistent, Sep 09 '21 at 05:54
You may want to read [this](https://stackoverflow.com/questions/5431941/why-is-while-feof-file-always-wrong). — n. 1.8e9-where's-my-share m., Sep 09 '21 at 07:27
Unrelated: I don't think `fread`ing a `struct` one field at a time is a good idea. It may be a fun exercise, but don't do that in real code. — n. 1.8e9-where's-my-share m., Sep 09 '21 at 07:42
staypersistent, "it gives me wrong outputs except the 1st line of the file." is telling. Is the file in _lines_ or is it _binary_? Post the file's data content. — chux - Reinstate Monica, Sep 09 '21 at 10:33
`%s` is not the best specifier to print `char o[MAX_CT];` as it is not certianly _null character_ terminated. Use ``%.11s`. — chux - Reinstate Monica, Sep 09 '21 at 10:43

score -5 · Answer 1 · answered Sep 09 '21 at 04:28

-5

i prefer

int c = fgetc(filehandle);

so, I can check:

int c = fgetc(filehandle);
if (c == EOF) {
  fclose(filehandle);
  exit(0);
}

note: for EOF, you can alternatively use '\0'. don't use exit(0); in modern programs. Try to design a program logic, that only need "return ERROR;" or "return SUCCESS;" Because you ran into some trouble if "exit" is nested, and you don't know the position in source. So, good idea is, to "strace", and debug, before you deploy your work.

answered Sep 09 '21 at 04:28

Jens

57
8

2

_"for EOF, you can alternatively use '\0'."_: huh(!?) – Jabberwocky Sep 09 '21 at 05:43
yes, \0 (not ***NULL*** or ***nullptr***) has different things - e.g. mark a char array (char *) in C, and C++ as null-terminated string, mark the (E)nd(O)f(F)ile - EOF. during execution of a binary application, it is used to start counters - computers begin starting to count at 0, not 1. so the numbers have +1 if you calculate with them. \0 can mark a empty database record, and so on, and so forth, ... since '0' is other char as '\0' - caution on the back-slash !!! – Jens Sep 09 '21 at 06:29
3

@Jens `EOF` is definitely different from `'\0'`. `'\0'` does definitely not mark the end of file. The statement _"for EOF, you can alternatively use '\0'"_ is wrong. Read this: https://stackoverflow.com/a/4705984/898348. – Jabberwocky Sep 09 '21 at 06:52
4

The C standard tells us: 7.21.1#3: "The macros are .... `EOF` which expands to an integer constant expression, with **type int and a negative value**, that is returned by several functions to indicate end-of-file, that is, no more input from a stream;" Your suggestion to use `0` is clearly wrong as 0 is not a negative value and cannot be separated from possible `char` values returned by various functions like `fgetc` you use in your answer – Gerhardh Sep 09 '21 at 07:05

How to detect the end of file using fread() then finish the loop when reading binary file? (Do not use feof) - C language

1 Answers1