initializer element is not constant...

Hi
I'm trying to compile the following code, but gcc 4.2 returns the messages
"global.c:8: error: initializer element is not constant"
"global.c:9: error: initializer element is not constant"

Code:

---

#include <stdio.h>
#include <math.h>
#include <stdlib.h>
/*file global.c*/

static long double h=1e-2;
static long double t_max=2.0*3.1415926535897932384626433832795;
static long double y_max=100.0;
static unsigned int lin=(int)(y_max/h+2);
static unsigned int col=(int)(t_max/h+1);

int main()
{
  return 0; 
}

---

Can anyone help me? Nothing that I read on the subject is much clear about this... :(

If you want global variables just remove the static bit:

Code:

---

long double h=1e-2;

---

Or if you want constants use a define:

Code:

---

#define t_max 2.0*3.1415926535897932384626433832795

---

With #define it would probably be better to enclose the expression with parentheses. A macro name would also be typically uppercase.

Ok.
As a matter of fact I want them to be global and static as well.
I read about the #define environment to create macros but I'm not sure if

Code:

---

#define  t_max 2.0*3.1415926535897932384626433832795

---

results in a floating point t_max or what.

Thanks.

Also, the lines it's complaining about are lin and col. There's no reason the compiler should think that an expression like (int)(t_max/h+1) is constant, since it isn't. You could wrap it all up in parentheses and put it in a #define macro as per above.

Quote:

---

Originally Posted by John Connor

Ok.
As a matter of fact I want them to be global and static as well.
I read about the #define environment to create macros but I'm not sure if

Code:

---

#define t_max  t_max 2.0*3.1415926535897932384626433832795

---

results in a floating point t_max or what.

Thanks.

---

That's going to get you a lot of errors.

#define macros are just search-n-replace; if you did

Code:

---

#define t_max (2.0*3.1415926535897932384626433832795)

---

would simply make the preprocessor remove all occurrences of the token t_max and replace it with that ... thing on the right. The compiler would never see a variable t_max at all.

Quote:

---

Originally Posted by tabstop

That's going to get you a lot of errors.

#define macros are just search-n-replace; if you did

Code:

---

#define t_max (2.0*3.1415926535897932384626433832795)

---

would simply make the preprocessor remove all occurrences of the token t_max and replace it with that ... thing on the right. The compiler would never see a variable t_max at all.

---

Thanks I realised my mistake in the previos reply and removed the t_max t_max...

Now, if I use the expression

Code:

---

#define t_max 2.0*3.1415926535897932384626433832795

---

Can i call t_max expecting it to be double precision?

Thanks

Quote:

---

Originally Posted by John Connor

Thanks I realised my mistake in the previos reply and removed the t_max t_max...

Now, if I use the expression

Code:

---

#define t_max 2.0*3.1415926535897932384626433832795

---

Can i call t_max expecting it to be double precision?

Thanks

---

Yes, in that any constant with a decimal point in it in a C program is double unless you tag it otherwise: 1.0 is double, 1.0f is float.

Quote:

---

Originally Posted by tabstop

Yes, in that any constant with a decimal point in it in a C program is double unless you tag it otherwise: 1.0 is double, 1.0f is float.

---

Thanks a lot! That really helped me.
:)

Quote:

---

Originally Posted by John Connor

As a matter of fact I want them to be global and static as well.

---

I don't think you know what you mean by your wanting constants to be static. What are you expecting this to mean for them, and why?

Quote:

---

Originally Posted by iMalc

I don't think you know what you mean by your wanting constants to be static. What are you expecting this to mean for them, and why?

---

Hi
At the first time, wen I wrote the code

Code:

---

#include <stdio.h>
#include <math.h>
#include <stdlib.h>
/*file global.c*/

static long double h=1e-2;
static long double t_max=2.0*3.1415926535897932384626433832795;
static long double y_max=100.0;
static unsigned int lin=(int)(y_max/h+2);
static unsigned int col=(int)(t_max/h+1);

int main()
{
  return 0; 
}

---

h,t_max,y_max, lin and col were all global "variables". Weren't they?

Now wen a say global and static I mean that they can all be accessed by whatever function my program can possibly have, but can' be reassigned by any of them.

But I Didn't get it, I think...

Thanks

Right, static doesn't mean what you think it does. const is what you are after.

static means that the variable can't be seen outside the current compilation unit (that's a source-file in most cases, but if you do weird stuff, it may be different).

And I think the error you are getting is caused by the fact that the compiler can't quite figure out your rather complicated math in

Code:

---

static unsigned int lin=(int)(y_max/h+2);
static unsigned int col=(int)(t_max/h+1);

---

--
Mats

Quote:

---

Originally Posted by matsp

Right, static doesn't mean what you think it does. const is what you are after.

static means that the variable can't be seen outside the current compilation unit (that's a source-file in most cases, but if you do weird stuff, it may be different).

And I think the error you are getting is caused by the fact that the compiler can't quite figure out your rather complicated math in

Code:

---

static unsigned int lin=(int)(y_max/h+2);
static unsigned int col=(int)(t_max/h+1);

---

--
Mats

---

Hi, thanks for the tip.
The program is running ok after I changed things based on what was told here...

Sorry for bothering you all with newbie questions, but I think I'm starting to learn it right... I have read a lot but I have never practiced it...

Thanks