Is there any difference between if(a==5) or if(5==a) in C#?

Question

Just I would like to know, is there any difference between

if (a==5) or if (5==a)

in C#, Which one is better?

Answer 1

There's no difference - assuming that "a" is an integer.

I know some people prefer if (5==a) because in c & c++ if you wrote if (5=a) by mistake you'd get a compiler error while if (a=5) would result in a bug.

C# raises a compiler error in the latter case, so it's not an issue.

Answer 2

I'd actually say there is a difference, but it's not a technical one (as everyone has well covered already) - readability. It matters and the first form is much more natural.

Answer 3

The if(5 == a) construct is common in C/C++ because boolean values are represented using ints. Thus if you write a = 5 by mistake this can be evaluated in the context of the if, which is most likely not what you wanted.

In C# there's no implicit conversion from int to bool, so if you type = instead of == you'll get a compile error.

Answer 4

no difference, it is an old habit to avoid if(a=5) in c/c++.

These questions/answers are about the same:

• (0 == variable) or (null == obj): An outdated practice in C#?
• Checking for null, which is better? "null ==" or "==null"
• Why does one often see "null != variable" instead of "variable != null" in C#?

Answer 5

Both are equivalent. I remember when I used to code in C, I preferred 'if (5==a)' because it guarantees that I haven't typed 5=a accidentally as the compiler would throw error. This would not happen if we write 'if (a=5)'. Though it is a typo, it would not generate any compiler error and would go unnoticed.

But, in C# it is not the case. There is no logical reason to write 'if (5==a)'. If we had written 'if(a=5)', the compiler would throw an error. So in C# use 'if(a==5)'!

Answer 6

With correct design, there is no difference between "a == 5" and "5 == a". But there is some special situation, where has "a == 5" and "5 == a" different behaviour. It's very unpropably, but it is posible.

Nevertheless this example is constructed for demonstration of the situation, and I does not recomend do thinks such this.

Example:

public class BadClass {
    public int Value;

    public static implicit operator int( BadClass c ) {
        return c.Value;
    }
    //public static implicit operator BadClass( int n ) {
    //    return new BadClass { Value = n };
    //}

    public static bool operator ==( BadClass c, int n ) {
        return (c.Value + 1 == n);
    }
    public static bool operator !=( BadClass c, int n ) {
        return (c.Value + 1 != n);
    }

    public override bool Equals( object obj ) {
        if ( obj is int ) {
            return (this == (int)obj);
        }
        else {
            return base.Equals( obj );
        }
    }

    public override int GetHashCode() {
        return base.GetHashCode();
    }
}
...
BadClass a = new BadClass { Value = 13 };
var rslt_1 = (13 == a); //there will be true
var rslt_2 = (a == 13); //there will be false

Answer 7

Other than the safety if (5==a) gives you, No.

Answer 8

Only difference is if you forget the second equals, the first version is still a valid statement whereas the second version will raise a compile time error.

HTH

cheers,