Thread: help on pointers. - return pointer to largest element in array

Morning all.

question from chapter 11, qn 8 modern C programming by king

Write the following function:

Code:

int *find_largest(int a[], int n);

When passed an array a of length n, the function will return a pointer to the array's largest element

my solution

Code:

#include <stdio.h>

int *find_largest(int a[], int n)
{
	int i, x;

	x = 0;
	for (i = 0; i < 6; i++)
	{
		if (a[i] > x)
			x = a[i];
	}

	for (i = 0; i < 6; i++)
	{
		if (a[i] == x)
			return &a[i];
	}
}


int main()
{
	int b[6], i, *p;

	for (i = 0; i < 6; i++)
		scanf("%d", &b[i]);

	p = find_largest(b,6);

	printf("%d", *p);

	getchar();
	return 0;
}

I have traced the code line by line and it prints correctly up to line 31. However, when the program exits (line 35) , it goes to some other screen with very complex code and my output disappears.

Just to clarify: If I just run the code as is, there is no output. However, when I trace the code line by line, the output appears, but then disappears as the program exits
My input was 1 2 3 4 5 6

Except of this

Code:

main.c: In function `find_largest':
main.c:19: warning: control reaches end of non-void function

the rest seems ok.

I also run it and works fine for me.

I do not see the reason of the getchar() at the end, so remove it.

It's the very same thing.

1. Your find_largest function does not make use of the parameter "n".
2. Your find_largest function breaks if your array contains only negative numbers.

Pseudo-code (1 loop pass):

1. Assign a pointer to first element of array
2. Loop through array from index [2,n)
   
   
   • If value at current array index is greater than what the pointer points to then update the pointer to reference the new max value
3. Return the pointer

The second one is better because the first one is wrong. Try entering 6 1 2 3. The first one will print 3.

• You should be thankful to oogabooga for spotting the logical error for you.
• Never leave a program, if you are not sure it's completely correct.
• I think you know why the first code does not work (because you are testing only adjacent elements).
• Why name max as x? Wouldn't it better to give a name to the variable that completely describes its role in your program? Name "x" as "max" is my suggestion.

Originally Posted by std10093

Name "x" as "max" is my suggestion.

I agree with you on being clear in naming variables, but I wouldn't suggest "max" as a name. There are a few too many implementations that define macros named min() and max() in one or more of the standard headers. Admittedly, it's not standard, but it occurs often enough with some well-known compilers/libraries that names min and max are best avoided. The error messages emitted by compilers in such cases are usually rather cryptic as well (as are most errors resulting from an unwanted macro expansion).

I'd probably use a name like "largest" or "maximum" or something like that. Apart from being less ambiguous (is max for maximum, or the name of a soft drink made by pepsi?), it's less likely to run afoul of non-standard but common extensions.