prototyping functions

Ok. Normally, I prototype functions like this:

Code:

---

//prototype
int function(int var1, char var2, char *point, type variable);

//function
int function(int var1, char var2, char *point, type variable)
{
        code;
        etc. etc. etc.;
        [...]
}

---

But, I've also seen in some programs this:

Code:

---

//prototype
int function();

//function
int function(var1,var2,point,variable)
int var1;
char var2;
char *point;
type variable;
{
        code;
        etc. etc. etc.;
        [...]
}

---

Apparently, both of them work. But my question is: What is the difference?

Thanks!

The top one is the "new" ANSI way; the second one is the outdated K&R way.

Cool. Ok so, it is the exact same thing? Nothing special between both?

Also, if I read the standard correctly, the second form doesn't check for type consistency -- so you can't get warnings of the sort "passing double as argument 2 of function function, expecting char".

It should also be mentioned that in this case "new" means "1989" (if not farther back).

Thanks a lot. I'll just keep using the "new" way.

> It should also be mentioned that in this case "new" means "1989" (if not farther back).

K&R2 uses the ANSI-style declarations, and that came out in 1988.

Prototypes were introduced in some compilers and textbooks before the 1989 ANSI Standard (or 1990 ISO standard) were ratified. New standards do not appear by magic: there is a committee process by which they are developed and quite a few compiler vendors and textbook authors attend committees or receive documentation. Hence they are able to anticipate contents of the standards.