How to check if element of array is int or float?

Question

I have 2 arrays, one is commented, I would like to make universal printf that will plot the values:

int values [] = { 88, 56, 100, 2, 25 };
//float values[] = {88.5f, 56.5f, 100.0f, 2.234f, 88.12f};

if (value[0] is int) {
    for(i; i < 5; ++i)
    printf("%d ", *(values + i));
} else {
    for(i; i < 5; ++i)
    printf("%f ", *(values + i));
}

Is there any way to check it? For example when I wanted to see if element is char or int then I used sizeof

Yes I know. But I need to do program for classes. Ill comment the int values and uncomment float values then start program, then Ill comment float and uncomment int. Those arrays are used for sorting. — F1sher
– F1sher, Commented Jan 5, 2014 at 21:16
I dont need a way to store 2 value types in one. I need info how to find out if the element of array is int or float. — F1sher
– F1sher, Commented Jan 5, 2014 at 21:17
You are probably trying to do the wrong thing. This said, you can use in the latest C standard, _Generic(values[0], int: 0, float:1, default:2) — Pascal Cuoq
– Pascal Cuoq, Commented Jan 5, 2014 at 21:19
_Generic(values[0], int: 0, float:1, default: 2); thats interesting but won't compile. I'm latest Dev-C++ 5.5.3 currently. [Error] expected expression before 'int' — F1sher
– F1sher, Commented Jan 5, 2014 at 21:22
You forgot to initialize i to 0. And why are you using *(values + i) rather than values[i]? — Keith Thompson
– Keith Thompson, Commented Jan 5, 2014 at 21:51

Keith Thompson · Accepted Answer · 2014-01-05 21:39:48Z

3

With the caveat that this is a lot of sophistication for a “feature” that you are only going to use while debugging your program, you can use C11's _Generic construct:

_Generic(values[0], int: 0, float:1, default:2)

What you should probably do instead that would be more consistent with your intentions is conditional compilation:

#define FLOAT_CASE

#ifdef FLOAT_CASE
float values[] = {88.5f, 56.5f, 100.0f, 2.234f, 88.12f};
#else
int values [] = { 88, 56, 100, 2, 25 };
#endif

... // all the code that is independent of the type of values here

for(i; i < 5; ++i)
#ifdef FLOAT_CASE
  printf("%f ", *(values + i));
#else
  printf("%d ", *(values + i));
#endif

edited Jan 5, 2014 at 21:39

Keith Thompson

266k46 gold badges447 silver badges668 bronze badges

answered Jan 5, 2014 at 21:35

Pascal Cuoq

80.6k8 gold badges168 silver badges293 bronze badges

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

2 Comments

Keith Thompson Over a year ago

Why do you write *(value + i) rather than values[i]? And you forgot to initialize i to 0.

Pascal Cuoq Over a year ago

@KeithThompson You should direct all these remarks to the OP. I took program lines verbatim from the question to make it clear that my answer was to use #ifdef or _Generic, not to fix stylistic issues (that wouldn't help with the problem of writing generic code for int and float)

Marian · Accepted Answer · 2014-01-06 11:32:41Z

2

An ugly solution: if you only need to distinct between int and float you can assign a non-integer value, say 0.1 to the variable and then test the actual value. This gives something like:

#define IS_INT_TYPE(x) (x=0.1, x==0)

However, this will destroy the original value of the macro parameter. If you use gcc, you can use typeof() extension to cast the value:

#define IS_INT_TYPE(x) (((typeof(x)) 0.1) == 0)

And Pascal's notes gave me the idea to the solution:

#define IS_INT_TYPE(x) (((x*0+1)/2) == 0)

edited Jan 6, 2014 at 11:32

answered Jan 5, 2014 at 21:40

Marian

7,4922 gold badges24 silver badges36 bronze badges

3 Comments

Pascal Cuoq Over a year ago

If you want to go in this direction, a solution if one knows that x is neither 0 nor within a factor of two of the limits of its type is x / (x + x) == 0

Pascal Cuoq Over a year ago

Removing the limitation for 0: x==0 ? ((x+1) / 2 == 0) : …

Pascal Cuoq Over a year ago

Only requiring x, if float, to be finite: x == 0 ? ((x+1) / 2 == 0) : (x/x) / (x/x + x/x) == 0. [after seeing edit] Well, okay, your solution is clearly better but still requires x to be finite. This is easy to test: !(x < INT_MIN) && !(x > INT_MAX) && …

Collectives™ on Stack Overflow

How to check if element of array is int or float?

2 Answers 2

2 Comments

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

2 Comments

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related