C# For Loop type inference when using var

Question

Consider the following code...

            double total = Int32.MaxValue;
            total++;

            int previousX = 0;

            for (var x = 0; x <= total; x++)
            {
                if (x == Int32.MaxValue)
                {
                    Console.WriteLine("Int32 max reached.");
                }

                if (x < 0)
                {
                    Console.WriteLine("Counter now < 0.");
                }

                previousX = x;
            }

It would appear that, if you use var with a for loop the default type inference is of an int.

Is this correct because if the counter goes past the maximum for an int 32, instead of overflowing the stack it resets itself back to zero and then counts down from zero.

Note: previousX allows you to set breakpoints and see what the previous value of the counter "x" was.

Does anyone know why this happens?

It seems possible to get into a bit of a pickle by using var for the counter of a for loop.

I think the problem is your saying double total = Int32.MaxValue - you shouldnt be declaring total as a double in this case. — Ctrl_Alt_Defeat
– Ctrl_Alt_Defeat, Commented Nov 8, 2011 at 11:44
You have many questions here, can you focus on one at a time? — Lasse V. Karlsen
– Lasse V. Karlsen, Commented Nov 8, 2011 at 11:45
If I understand it right, you wish "total" to be larger than int. If so use Int64 not "double". — Roee Gavirel
– Roee Gavirel, Commented Nov 8, 2011 at 11:46
"[...] it resets itself back to zero and then counts down from zero.". Are you sure about that? That would really surprise me. I'd expect it to jump to Int32.MinValue and count up. And it does so on my machine. — Jens
– Jens, Commented Nov 8, 2011 at 11:47
Yes incrementing (int.MaxValue++) == Int.MinValue (Figuratively). int.MaxValue++ will equal -2147483647 — Lloyd
– Lloyd, Commented Nov 8, 2011 at 11:54

samjudson · Accepted Answer · 2011-11-08 11:51:36Z

6

The type of x is determined by the initial value, which above is the integer literal 0.

If you want x to be a double then use the double literal 0D.

edited Nov 8, 2011 at 11:51

answered Nov 8, 2011 at 11:45

samjudson

57k7 gold badges61 silver badges69 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

ChrisBD · Accepted Answer · 2011-11-08 12:02:40Z

3

It is important to understand that the var keyword does not mean "variant" and does not indicate that the variable is loosely typed, or late-bound. It just means that the compiler determines and assigns the most appropriate type. As you have var x=0 it determines that x is an int in order to match 0. So yes it is correct functionality.

Because of the manner in which negative numbers are represented (twos compliment) as a number wraps over its maximum value it will appear to reset and count down from -1 onwards.

The increment on x++ doesn't occur until that iteration of the loop has completed.

answered Nov 8, 2011 at 12:02

ChrisBD

9,2393 gold badges25 silver badges36 bronze badges

1 Comment

Remotec Over a year ago

+1 for explaining what happens when the number get to its max.

Tudor · Accepted Answer · 2011-11-08 11:45:07Z

2

Of course, if you increment an int past MaxValue, it goes back down to MinValue and starts to increment from there.

answered Nov 8, 2011 at 11:45

Tudor

62.6k13 gold badges105 silver badges147 bronze badges

Comments

HeXanon · Accepted Answer · 2011-11-08 12:17:10Z

2

This is indeed interesting but is unrelated to the use of var. This still happens when using int (or in fact long, short etc).

The actual culprit here is that the MAX value for the loop is larger than the max value for int32. when a numeric value is incremented the number will wrap causing an infinite loop. You can declare var x, int x, short x and ALL of these would loop in a similar manner.

The nature of incrementing numeric values cause 'wrap-around'. You can try it by running the following code and observing the values befor and after the increment. (This also works if you add a decrement).

        int a = Int32.MaxValue;
        Console.WriteLine(a);
        a++;
        Console.WriteLine(a);

        short b = Int16.MaxValue;
        Console.WriteLine(b);
        b++;
        Console.WriteLine(b);

        long c = Int64.MaxValue;
        Console.WriteLine(c);
        c++;
        Console.WriteLine(c);

As an aside, the var in the code snippet posted is being set to an int by the statement var x = 0. The '0' is itself an int and hence the type of x is set to int. If you stated var x = 0F or 0L the type would be different.

edited Nov 8, 2011 at 12:17

answered Nov 8, 2011 at 12:01

HeXanon

5141 gold badge8 silver badges12 bronze badges

1 Comment

Remotec Over a year ago

I wonder why it wraps? Historically these sort of things used to overflow (i'm sure I have seen that). I think that this behavior may give unpredictable results if you are not carefull. Especially, using VAR in this case means you dont really think about the data type where you would if you were forced to specify it.

Piotr Auguscik · Accepted Answer · 2011-11-08 11:45:58Z

1

Var exact type is defined by your usage of = operator. In c# 0 is of type int so compiler is using int.

answered Nov 8, 2011 at 11:45

Piotr Auguscik

3,6911 gold badge25 silver badges31 bronze badges

Comments

Ray · Accepted Answer · 2011-11-08 11:46:27Z

1

There's nothing special going on here. var will be inferred as Int32 because that's what 0 defaults to. If you want a long use var x = 0L. Or just long x = 0

answered Nov 8, 2011 at 11:46

Ray

46.7k30 gold badges128 silver badges170 bronze badges

Comments

Community · Accepted Answer · 2017-05-23 12:03:10Z

1

The var x = 0 is going to infer that x is an integer because you are using an integer literal ("0").

Check out this related post concerning integer overflow:

why this would result in long integer overflow

edited May 23, 2017 at 12:03

CommunityBot

11 silver badge

answered Nov 8, 2011 at 11:47

JohnD

14.9k4 gold badges42 silver badges54 bronze badges

Comments

Ilya Smagin · Accepted Answer · 2011-11-08 11:48:53Z

0

This looks quite simple:

var x = new User()

When compiler sees this line, he replaces var with User, because that's what constructor returns. The same happens with function calls:

var x = y.ToString()

ToString() returns string, so compiler knows what to replace var with.

When you write

var x = 0

you actually call int constructor, so var is replaced with with int

You have three options:

double x = 0 ...
var x = 0.0 ...
var x = 0D ...

Note: D suffix is for double

answered Nov 8, 2011 at 11:48

Ilya Smagin

6,18011 gold badges42 silver badges57 bronze badges

Collectives™ on Stack Overflow

C# For Loop type inference when using var

8 Answers 8

Comments

1 Comment

Comments

1 Comment

Comments

Comments

Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

8 Answers 8

Comments

1 Comment

Comments

1 Comment

Comments

Comments

Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related