Thread: Define a macro to sum the elements of an array

Your compiler should be giving you some feedback. What are the error messages and what do you understand of them?

The error message is surprising vague. Anyway, for starters, the macro parameter should be a name only. The [10] does not belong there. Within the macro, you should be using the array name when you want to refer to the array. array[10] is actually an element of the array (that does not exist).

Also, be careful of the difference between braces and parentheses, and you should not have any characters after the \ that ends each physical line.

Originally Posted by laserlight

Your compiler should be giving you some feedback.

You have a talent for understatement, laserlight

In terms of specifics for the OP (which I have picked up only on a quick look)

1) Take the array dimension out of the argument list of the macro definition. That will probably cause the preprocessor to reject the definition, and explain why subsequent attempts to use the macro give a compile error referring to the name of the macro.

2) The body (expansion) of the macro should not be delimited by brackets (). Use curly braces instead.

3) If you want the size of the array, compute sizeof(array) not sizeof(array[10]). sizeof(array[10]) gives the size of an element of the array, not the size of thw whole array. Since array is an array of int, computing sizeof(array[10])/sizeof(int) will always give a value of 1.

4) If you are going to report error messages in a forum, don't only report the last error message you see. The compiler (or, more specifically, the preprocessor) will report an error at the macro definition before it will emit the error message you reported. By being selective, you have omitted information that is directly relevant to your problem. Incidentally, if you had taken time to interpret ALL the error messages, you could have puzzled out a solution to your problem for yourself.

5) Your macro needs to define the variable i and initialise it to zero, otherwise it will trigger other error messages (assuming you fix problems pointed out above) from the compiler.

6) Your macro is attempting to loop over array elements to a maximum index value of arraySize. Array indices run from 0 to size-1, not from some random value to size.

7) Bear in mind that your code will not work for pointers. That means if you pass the array to a function, the macro will not work in the body of that function.

8) Your would be better off writing a function to do what you want, rather than a macro. If for no other reason that avoiding the silliness you have introduced in your macro - do such silliness in a function rather than in a (broken) macro, and the compiler will emit meaningful error messages.

Originally Posted by volte

/*Define and use a macro SUMARRAY to sum the values of a numeric array.*/

Code:

int sum( int array[], size_t nelm )
{
    int s = 0;
    size_t x = 0;
    for( x = 0; x < nelm; x++ )
        s += array[ x ];
    return s;
}
/* and now the magic */
#define SUMARRAY sum

Or you could make it an actual macro, but you will need to pass the value to store the sum in, initialized to zero:

Code:

#define SUMARRAY(a,b,c)do(c)+=(a)[(int)--(b)];while((b))

Hm...that looks about right.


Quzah.