Abnormal Behaviour of Array-CodePudding

This Code convert a decimal number i to its binary ... Along with storing its value in a array ans.

In This below code i used ans size 100 . Now when i enter input as 0.5 it gives a weird output . But if i change the size of Array to 1000 the it's giving correct output can anyone explain this behaviour.

Code with ans Size 100

#include <stdio.h>

int main() {
    float n;
    printf ("Enter Decimal Number:");
    scanf("%f", &n);

    // Integer Part of Number
    int ip = n;
    // To store the Binary Number
    char ans[100];
    // Index of array
    int i = 0;
    // For integer part
    while(ip != 1 && ip != 0) {
        if ((ip % 2) == 0) {
            ans[i] = '0';
        }
        else {
            ans[i] = '1';
        }
        i  ;
        ip /= 2;
    }
    if (ip) {
        ans[i  ] = '1';
    }
    else{
        ans[i  ] = '0';
    }

    // Reverse the Integer Part
    for (int j = 0; j < (i/2); j  ) {
        char temp = ans[j];
        ans[j] = ans[i - j - 1];
        ans[i - j - 1] = temp;
    }
    ans[i  ] = '.';
    ans[i] = '0';

    // Flaoting Part of Number
    float fp = n - (int) n;

    // For floating Part
    if (fp) {

        for (int j = 0; j < 5; j  ) {
            fp *= 2;
            // Integer part of fp
            int x = fp;
            // floating part  of new fp
            fp = fp - x;
            if (x) {
                ans[i  ] = '1';
            }
            else {
                ans[i  ] = '0';
            }
            if(!fp){
                break;
            }
        }
    }
    printf("%f in Binary is %s", n, ans);
    return 0;
}

OUTPUT

Enter Decimal Number:0.5

0.500000 in Binary is 0.1�

[Program finished]

Code with ans size 1000

#include <stdio.h>

int main() {
    float n;
    printf ("Enter Decimal Number:");
    scanf("%f", &n);

    // Integer Part of Number
    int ip = n;
    // To store the Binary Number
    char ans[1000];
    // Index of array
    int i = 0;
    // For integer part
    while(ip != 1 && ip != 0) {
        if ((ip % 2) == 0) {
            ans[i] = '0';
        }
        else {
            ans[i] = '1';
        }
        i  ;
        ip /= 2;
    }
    if (ip) {
        ans[i  ] = '1';
    }
    else{
        ans[i  ] = '0';
    }

    // Reverse the Integer Part
    for (int j = 0; j < (i/2); j  ) {
        char temp = ans[j];
        ans[j] = ans[i - j - 1];
        ans[i - j - 1] = temp;
    }
    ans[i  ] = '.';
    ans[i] = '0';

    // Flaoting Part of Number
    float fp = n - (int) n;

    // For floating Part
    if (fp) {

        for (int j = 0; j < 5; j  ) {
            fp *= 2;
            // Integer part of fp
            int x = fp;
            // floating part  of new fp
            fp = fp - x;
            if (x) {
                ans[i  ] = '1';
            }
            else {
                ans[i  ] = '0';
            }
            if(!fp){
                break;
            }
        }
    }
    printf("%f in Binary is %s", n, ans);
    return 0;
}

OUTPUT

Enter Decimal Number:0.5

0.500000 in Binary is 0.1

[Program finished]

CodePudding user response：

Your ans string in not null-terminated, so printing a string that is not null-terminated causes UB (undefined behavior). As to why it works when the size is 1000 vs. 100 I cannot say, but the bottom line is that undefined bahavior is undefined... anything could happen. So fix the bug as follows:

  :
  :

// Flaoting Part of Number
float fp = n - ip;

// For floating Part
if (fp) {
    for (int j = 0; j < 5; j  ) {
        fp *= 2;
        // Integer part of fp
        int x = (int)fp;
        // floating part  of new fp
        fp -= x;
        ans[i  ] = x ? '1' : '0';
        if(!fp)
            break;
    }
} else i  ;
ans[i] = '\0';
printf("%f in Binary is %s", n, ans);
return 0;

}

CodePudding user response：

I am turning my comment into an answer.

Your ans array is not null terminated, hence printf("%s", arr) doesn't know where to stop reading and keeps printing whatever garbage follows. If you are going to print it with %s append a null terminator at the end of your data in the array before printing it, or print it character by character in a loop.

You can also initialize your array as char ans[100] = { 0 }, which assigns all elements 0. Which also is the ASCII code for the null terminator, hence when you are done writing your characters, the next character will be a null terminator. This however, doesn't work if you re-write a smaller data to your array, previous characters that are not overwritten will still be printed. You can consider using memset to clean you array with 0 before each time you re-write it.

When it comes to the difference between 100 and 1000 element array, I guess your compiler happens to allocate 100 element array in a part of the memory containing garbage, and 1000 element array in a part that happens to contain 0. This is unpredictable and as you can see, can mislead you. Never, ever use your variables uninitialized.