variable defined in the header by #define isn't set by the code

Question

I made a very simple code which calculates the values of a function:

#define MIN 0.00001
#define MAX 1
#define MAXP 10000


 double func_log(double x)
{
    double r=RPWN;
printf(" MAXP = %e  \n",MAXP);
printf(" MIN = %e  \n",MIN);
printf(" MAX = %e  \n",MAX);

    double y;
    if (x>r || x==0) return(0.0);
    else {
        y = r+((r*r)/(2.*x)-x/2.)*log((1+r/x)/(r/x-1));  
    }
    return y;       

}

That fonction is in a file.c which is compiled with other source files. The problem is that MAXP cannot be set to its value. Here is what I get when I ask to print out the values of MIN, MAXP, and MAXP:

 MAXP = 4.940656e-324  
 MIN = 1.000000e-05  
 MAX = 1.000000e+00

I really don't understand why I get 4.940656e-324 for MAXP, I never had such issue. Also when I write #define MAXP 10000., I then get MAXP = 1.000000e+04. I compile with gcc, does someone has any clue?

MAX and MAXP are integer constants being printed as doubles. Happiness is not one of the options. Add decimal points to the definitions. — Jonathan Leffler, Nov 23 '17 at 15:14
Enable compiler warnings. Voting to close this as simple typo. — Lundin, Nov 23 '17 at 15:14
@Lundin not necessarily a typo, the question is OK. For a beginner this behaviour may actually look strange — Jabberwocky, Nov 23 '17 at 15:16
@MichaelWalz Fine, then close as duplicate to [What Are Integer Literal Type?](https://stackoverflow.com/questions/41405578/what-are-integer-literal-type-and-how-they-are-strored). — Lundin, Nov 23 '17 at 15:30
@Lundin not close enough for me. And the OP will probably not understand — Jabberwocky, Nov 23 '17 at 15:40

Sergey Kalinichenko · Accepted Answer · 2017-11-23T15:19:19.797

The problem is not in the value of MAXP, but in the format specifier %e that you are using to print it.

MAXP is not a variable, it is a macro. Hence, it has no type associated with it. The compiler figures out the type of 1000 from the context, which in case of printf means int (the default interpretation of 1000). In a different context 1000 could be interpreted as a float or a double, e.g. float f = MAXP; An important thing to remember is that MAXP is textually replaced with 1000 in the body of your code.

If you want MAXP to be interpreted as a double by default, add .0 at the end:

#define MAXP 10000.0

If you want it to remain an integer literal, use %d to print it

printf(" MAXP = %d  \n", MAXP);

or add an explicit cast:

printf(" MAXP = %e  \n", (double)MAXP);

In fact, you don't even need the `0` at the end (in case there is some possibility of it being mistaken by the programmer as a significant figure), you just need the decimal point or an exponent part. — Ian Abbott, Nov 23 '17 at 15:18

score 2 · Answer 2 · answered Nov 23 '17 at 15:14

2

The problem is that %e prints float and double types, but your MAXP and MAX type integer.

You can use %d to print integers.

answered Nov 23 '17 at 15:14

Bastian

10,403
1
31
40

variable defined in the header by #define isn't set by the code

2 Answers2