Hello everyone !
I am still a beginner, and am posting the code in which I have problems -
I compiled it in VC++ 6.0 standard edition, and it gave two warnings and no errors. The warnings disappear if i use double everywhere instead of float.
the warning (same twice)-
warning C4305: '=' : truncation from 'const double ' to 'float '


float fun1(float,float);
float *fun2(float,float);
int *fun3(int,int);

void main()
	float i,j,k,prod,*l;
	int o=3,m=2,*n;

	printf("the product =\n\n%f",prod);

	printf("\n\nthe product(from function -call by value-return value) = %f",k);

	printf("\n\nThe product (from function -call by value-return pointer)= %f",*l);

	printf("\n\nThe product of integers -return pointer= %d",*n);

float fun1(float i,float j)
	float prod;
	return prod;

float *fun2(float i, float j)
	float *p,prod;
	return p;

int *fun3(int i,int j)
	int prod,*p;
	return p;


I have three questions -
1. If i use float instead of double, i get the answer 25.628401 (which is wrong(-ish), the answer is 25.628400)
2. the first two printf's give the right result, but the last two printf's give the wrong result. the second printf gives 0.000000 and the last printf gives a garbage six digit value.
3. if i replace "return p" by "return &prod" in fun2, VC++ gives a warning
warning C4172: returning address of local variable or temporary
I understand that after control in fun2 or fun3 ends, the local variables "die" and if i use static, i get the right answer, but why is this happening ?
When a variable is declared, a memory "cell" is allocated to it and the name (say i) is given to it. Here, after the control exits from the functions (fun2 or fun3), the memory cell has lost its name(which was "i " till the control was in fun2 or fun3), but the address of the cell remains the same, so i shud be able to access it correctly using pointers. It is happening as if the compiler resets the value of the memory cell (to zero if earlier it was float and garbage if it was int) as soon as the memory cell loses its name. And if this is happening (by any chance), why is the compiler doing this ??

I appreciate any help you can give......

7 Years
Discussion Span
Last Post by s_sridhar

>If i use float instead of double, i get the answer 25.628401 (which is wrong...
1. It's the same question as Why Cessna 182 can't be so fast as F-22?: float data type has precision ~6-7 decimal digits only...

2. and 3. "dye" means dye: memory cells allocated (as usually in so called stack memory) for automatic (local) variables are reusable storage pool. In your case they were overwrited by the next printf call. That's why you got so strange results. Moral: never return pointers to local variables...


you cant pass the address of the local variable. They are destroyed after the function ends. Even the pointer which you declare is local. So, if you want to really use pointers pass all the pointers from the main function into the corresponding function.


>So, if you want to really use pointers pass all the pointers from the main function into the corresponding function.
It's too restrictive and unrealistic requirement.

>Even the pointer which you declare is local.
It's of no importance after return pointer; However fun2 and fan3 OP functions are classic examples of never do that codes (like void main() instead of standard int main() . ;)


>It's too restrictive and unrealistic requirement

Yeah of course, I agree with you, but if he still wants to learn the working of pointers, he can do that.

This article has been dead for over six months. Start a new discussion instead.
Have something to contribute to this discussion? Please be thoughtful, detailed and courteous, and be sure to adhere to our posting rules.