No matter if the condition is true if always executes in c++

Question

Can someone correct this code please.

#include <iostream>
using namespace std;

int main()
{
    int a;
    int b;
    int c;
    double d;
    double e;
    double f;
    double g;
    f = a / b;
    g = b / a;
    c = 0;
    cin >> a;
    cin >> b;
    f = a / b;
    g = b / a;
    if (a == b)
    {
        cout << a << endl;
        return 0;
    }
    else if (f == int(f))
    {
        cout << a << endl;
        return 0;
    }
start:
    while (a * b > c)
        c = c + 1;
    d = c / a;
    e = c / b;
    if (d == int(d))
        if (e == int(e))
        {
            cout << c << endl;
            return 0;
        }
        else if (d != int(d))
            goto start;
        else if (e != int(e))
            goto start;
    if (a * b <= c)
        cout << a * b << endl;
}

No matter what the

else if(f==int(f))

code is always executed. Eg. I put in 3 and 5 and even though 3/5 gives a decimal the else if is always executed and outputs 3. WHAT AM I MISSING HERE?

can you send the corrected if else part please because i am a beginner — LuckyFire, Feb 23 '20 at 21:39
Please try `d = (1.0*c) / a;e = (1.0*c) / b;`. And similar for f and g. If that helps I will explain. — Yunnosch, Feb 23 '20 at 21:40
Does this answer your question? [Integer division always zero](https://stackoverflow.com/questions/9455271/integer-division-always-zero) — Yksisarvinen, Feb 23 '20 at 21:42
`f = a / b;` at this piont `a` and `b` are uninitialized, leading to undefined behavior. Before doing anything else, fix that. — Timo, Feb 23 '20 at 21:42
first, all the division are integer division, second, what you want to do with `f == int(f)`? create a integer version of f? — Alberto Sinigaglia, Feb 23 '20 at 21:42
`int/int` truncates its result into another `int`, you need to cast to `float` or `double` before dividing — Alan Birtles, Feb 23 '20 at 21:42
Does this answer your question? [Why does division result in zero instead of a decimal?](https://stackoverflow.com/questions/8906722/why-does-division-result-in-zero-instead-of-a-decimal) — Alan Birtles, Feb 23 '20 at 21:43
well on my pc if i insert "1" and "2" that else is fired.. maybe you should read about integer division and narrowing conversion — Alberto Sinigaglia, Feb 23 '20 at 21:44

Yunnosch · Answer 1 · 2020-02-23T21:51:54.230

2

These expression, though they are assigning to double variables, calculate integer divisions.

f = a / b;
g = b / a;
d = c / a;
e = c / b;

I.e. what gets assigned to the doubles are integer values.

Your if-conditions basically check for integer values and hence always evaluate to true.

In order to avoid the integer division and get actual floating point values assigned to the doubles, you need to make sure early that the compiler interprets them accordingly. E.g.:

f = (1.0*a) / b;
g = (1.0*b) / a;
d = (1.0*c) / a;
e = (1.0*c) / b;

And, as a comment points out, better always init all your variables (even if here a,b,c would be enough).

int a=1;
int b=1;
int c01;
double d=1.0;
double e=1.0;
double f=1.0;
double g=1.0;

edited Feb 23 '20 at 21:51

answered Feb 23 '20 at 21:43

Yunnosch

26,130
9
42
54

@Timo I picked up your advice, hope you take it as a compliment. – Yunnosch Feb 23 '20 at 21:52
`(1.0*a) / b` instead of `static_cast(a) / b`? Doing an identitiy operation to force a type conversion feels wrong :/ might aswell do `(0.0 + a) / b` – Timo Feb 23 '20 at 21:54
At least an alternative to potentially risky typecasting or even changing integers to imprecise floating point types. – Yunnosch Feb 23 '20 at 21:56
Well you can't get any safer than `static_cast`. – Timo Feb 23 '20 at 21:57
Or `f=a; f/=b;` – Robert Andrzejuk Feb 23 '20 at 22:09
@RobertAndrzejuk Interesting approach. I think you can make an answer of that, ideally an explained one (and are allowed to do a new one for that, but no guarantees, only my opinion...). – Yunnosch Feb 23 '20 at 22:12

score 2 · Answer 2 · answered Feb 23 '20 at 21:47

The primary error in your code leading to the problem with the if statement is the integer division. You have to cast the operands to doubles to perform floating point division:

cin >> a;
cin >> b;
f = double(a) / double(b);
g = double(b) / double(a);

There are other issues to clean up, but this is the one that leads to your question.

score 1 · Accepted Answer · answered Feb 23 '20 at 21:45

1

You just need to define type a and b as double not int.

answered Feb 23 '20 at 21:45

Mehdi

717
1
7
21

1

This is the simplest solution and it just works. Thank you so much. – LuckyFire Feb 23 '20 at 21:51

No matter if the condition is true if always executes in c++

3 Answers3