Floats breaking my code

Question

I've written a program that returns the median value of a user-defined array. While I've put a few checks in my code (array size can not be negative) I keep running into one issue I simply can not fix (for clarity sake, assume strings and alphabetical characters will not be used).

All of my input values are int however the user could just as easily enter in a float. When they do this (either for size of array or entering in the element) it breaks my code. I've tried multiple things to try and catch this, but it seems like the way my program is getting the value doesn't allow for the catch in time.

#include <iostream>
using namespace std;

void sort(int * a,int n)
{
    for(int i=0;i<n;++i)
        for(int j=i+1;j<n;++j)
        {
            if(a[i]>a[j])
            {
                int tmp = a[i];
                a[i] = a[j];
                a[j] = tmp;
            }\
        }
    return;
}

int main()
{
    int n;
    int check;
    int x;
    cout<<"Enter length of array:";
    cin>>n;
    if (n < 0){
        while (n < 0){
        cout << "Please enter a length greater than 0" << endl;
        cin >> n;
        }
    } else if (n % 1 != 0){
        while (n % 1 != 0){
            cout << "Whole numbers only! Try again" << endl;
            cin >> n;
        }
    }
    if (n == 0){
        cout <<"You try to enter numbers, but there's no place to put them." << endl;
        cout << ":(";
        return 0;
    }
    int a[n];
    cout<<"Enter values one by one:\n";
    for(int i=0;i<n;++i){
        cin >> x;
        a[i] = int(x);
    }
    sort(a,n);
    if (n % 2 == 1){
    cout<<"Median is:"<<a[n/2]<<endl;
    }
    else{
    float z = (float(a[n/2]) + float(a[(n/2)-1])) / 2;
    cout << "Median is:" << z << endl;
    }




    return 0;
}

First thing I tried was catching the float like so

`if (n % 1 !=0){
    while(n % 1 !=0){
        cout << "Enter a whole number"
        cin >> n
    }
}`

This still broke my program. The odd thing was that I entered a float and then printed the value of n and it only showed the int value.

I tried using typeid.n() with #include <typeinfo>and comparing that to an int type to check it was the correct value, but that slipped through as well.

I tried doing an int cast, something like int(n) immediately after number was stored in n but before it went into a[n] and yet again, it still broke my code.

How can I check against float user-input and loop them until they give me an int?

Answer 1

You're reading into an int:

int x;
...
cin >> x;

So it will read what it can, then stop at e.g. a . and leave the rest on the stream (like if the user enters "123.4" you'll get 123 and then ".4" won't be consumed from the input stream).

Instead, you could read into a float:

float x;
...
cin >> x;

And do the appropriate math.

Alternatively you could read into a string and parse it into a float. That way you won't get stuck at letters and such either.

And the final option is to read into an int but handle any errors and skip the bad input, which is detailed at How to handle wrong data type input so I won't reproduce it here.

Which option you choose really just depends on what you want the behavior of your program to be and how strictly you want to validate input (e.g. round vs. fail if "2.5" is entered but an integer is expected, how do you want to handle "xyz" as input, etc.).