Bdtjtk

So after playing a little bit with Craig Estey's answer, I managed to get the following piece of code:

#include <string.h>

#include <stdio.h>

#include <stdlib.h>

#include <errno.h>

#include <unistd.h> 

#include <sys/types.h> 

#include <sys/stat.h> 

#include <fcntl.h> 





int main(int argc,char **argv){

    char *filename = argv[1];

    if(filename == NULL) {

        printf("Please specify a filename.n");

        exit(1);

    }

    FILE *fi = fopen(filename,"r");

    printf("Finding size...n");

    int size = 0;

    while(fscanf(fi,"%*d") != -1) size++;

    rewind(fi);

    int numbers[size];

    while(fscanf(fi,"%d",&numbers[size]) != 1);

    fclose(fi);

    printf("Size of array: %dnn", size);

    int idx=0,idx2=1,idx3=1,idx4=1;

    printf("Elements: n");

    while (idx < size){

        printf("%dn",numbers[idx]);

        idx++;

    }

    int maximum = numbers[0];

    while (idx2 < size){

            if (numbers[idx2] > maximum){

                maximum  = numbers[idx2];

        }

        idx2++;

    }

    printf("nMax: %dn",maximum);

    int minimum = numbers[0];

    while (idx3 < size){

            if (numbers[idx3] < minimum){

                minimum  = numbers[idx3];

        }

        idx3++;

    }

    printf("Min: %dn",minimum);

    int sum = numbers[0];

    while (idx4 < size){

        sum = sum + numbers[idx4];

        idx4++;

    }

    printf("Total: %dn",sum);

}

Problem is, now it executes without halting, but still doesn't provide the right answer. My 'abcd.txt' file contains the following numbers:

5564

4324

863

98743

However, my result after executing reader.c is:

./reader abcd.txt

Finding size...

Size of array: 4



Elements: 

0

0

1384783917

32713



Max: 1384783917

Min: 0

Total: 1384816630

Why is that, now? I cannot find why is it different than in the answer below. If I execute the exact code in the answer, it does return the right answer. Thanks in advance.

This is prefaced by my top comments.

You can't do: printf(numbers[idx2]); (i.e. the first argument has to be a format string). So, do: printf("%dn",numbers[idx2]);

Doing sizeof(numbers) / sizeof(int) does not give an accurate count of the number actually filled in. It gives the maximum which will have garbage values at the end.

Because, there were so many errors, I couldn't list them all. I had to completely refactor your code. Just compare it in detail to yours to see the difference. You'll learn much more and more quickly than trying to patch what you already have.

#include <errno.h>

#include <fcntl.h>

#include <stdio.h>

#include <stdlib.h>

#include <string.h>

#include <sys/stat.h>

#include <sys/types.h>

#include <unistd.h>



long int

findSize(char *filename)

{

    FILE *fp = fopen(filename, "r");



    if (fp == NULL) {

        printf("findSize: unable to open file '%s' -- %sn",

            filename,strerror(errno));

        exit(1);

    }



    fseek(fp, 0L, SEEK_END);

    long int res = ftell(fp);



    fclose(fp);

    return res;

}



int

main(int argc, char **argv)

{



    char *filename = argv[1];

    if (filename == NULL) {

        printf("please specify a filenamen");

        exit(1);

    }



    printf("findSizen");

    int numbers[findSize(filename)];



    printf("scanfn");

    FILE *fi = fopen(filename,"r");

    int size = 0;

    while (1) {

        if (fscanf(fi,"%d",&numbers[size]) != 1)

            break;

        printf("scanf: %dn",numbers[size]);

        ++size;

    }

    fclose(fi);



    printf("size: %dn",size);

    for (int idx2 = 0;  idx2 < size;  ++idx2)

        printf("%dn",numbers[idx2]);



    int maximum = numbers[0];

    for (int idx3 = 1;  idx3 < size;  ++idx3) {

        if (numbers[idx3] > maximum)

            maximum = numbers[idx3];

    }

    printf("Max: %dn", maximum);



    int minimum = numbers[0];

    for (int idx4 = 1;  idx4 < size;  ++idx4) {

        if (numbers[idx4] < minimum)

            minimum = numbers[idx4];

    }

    printf("Min: %dn", minimum);



    int sum = 0;

    for (int idx5 = 0;  idx5 < size;  ++idx5)

        sum = sum + numbers[idx5];

    printf("Total: %dn", sum);



    return 0;

}

UPDATE:

First of all, thanks a lot for your work. Looking at your code (I cannot call it anymore my code as it's completely different)

Yes, it was intended as a guide for you to write/rewrite your own code, which you've done.

I see that the findSize function is mostly irrelevant.

It is technically correct, but would allocate [much] more space than was needed.

I tried to put another fscanf instead of calling a findSize function, but I don't know how this function works exactly.

Good for you in figuring out this alternate sizing algorithm.

Let me see if I get it: If there is a third argument, it stores each read element into the third argument (which should be a variable). But if there is no third element, and you put an * in the second argument, it doesn't store the read elements anywhere. Am I right?

Believe it or not, I don't use fscanf too much as it can be slow, so I'm not familiar with some of the more esoteric options. I've tested your new sizing loop and it appears to work, so I think your analysis may be correct.

Also, we could simply introduce an extra scalar int variable and do:

int size = 0;

while (1) {

    int val;

    if (fscanf(fi,"%d",&val) != 1)

        break;

    ++size;

}

However, you're actual second fscanf loop that is used to read in the numbers will not work:

while (fscanf(fi, "%d", &numbers[size]) != 1);

This puts all numbers in the same index, namely size, which is one beyond the end of the numbers array, so this is undefined behavior.

Also, the loop condition is reversed from what it should be, so only one number would be filled in.

To fix this, we need an additional index variable and need to change the sense of the while loop:

int idx0 = 0;

while (fscanf(fi, "%d", &numbers[idx0]) == 1)

    ++idx0;

After this change, the rest of your updated code [min/max/sum] works fine.

Some notes on style:

As you noticed, my original posted code was a bit different (e.g. replacing while loops with for loops).

But, one thing I didn't do was reuse the index variable. That is, there isn't a need to have separate ones for each loop. They could all be changed to a single idx variable.

Using different symbols isn't wrong, and can add to clarity. But, idx, idx1, idx2 aren't as descriptive as they could be.

If you'd like to keep separate variables, I'd suggest more descriptive names such as: inpidx, minidx, maxidx, and sumidx