Your assignment is to use operator overloading to solve the issues - but you actually don't do so anywhere (apart from the operator+ for your Analyzer class – which is meaningless, though).
Reading your lines, I'd rather assume that you're supposed to write separate classes for each task:
class Minimum
{
std::vector<int> const& values
public:
Minimum(std::vector<int> const& values) : values(values) { }
// calculates minimum / lowest value from member:
int operator()();
};
class Maximum
{
public:
//Maximum(); not needed in this variant
// calculates maximum from parameter
int operator()(std::vector<int> const& values);
};
void test()
{
std::vector<int> values({10, 12, 7});
int min = Minimum(values)();
int max = Maximum()(values);
}
These are two different patterns, for consistency, you should select one and implement all classes alike. In first approach, you can access the vector from any member function without having to pass it around as parameter, in second approach, you can re-use one and the same object to calculate the value on several different vectors (you could still maintain a pointer to the vector to avoid passing it around via parameters...).
Coming back to your original code, unfortunately it is full of errors
vector<int> numbers;
int min = numbers[0]; // vector is yet empty! undefined behaviour!
int max = numbers[0];
Actually, you might want not to use globals at all, see later...
//int sizeofArray; // use numbers.size() instead!
// not an error, but questionable: you have a std::vector already, why do you
// fall back to C-style raw arrays?
void randNumbers(int sizeofArray, int* numbers[])
// ^ array of pointers???
{
for (int index = 0; index < sizeofArray; index++)
{
numbers[index] = (numbers() % 499) + 100;
// you certainly intended to use rand function
}
// return; // just plain obsolete
}
// vector variant:
void randNumbers(unsigned int numberOfValues, std::vector<int>& destination)
// ^ not how many numbers ARE in,
// but how many SHALL be inserted
{
// assuming we want to re-use this function and guarantee that EXACTLY
// 'numberOfValues' values are contained:
destination.clear(); // there might have been some values in already...
// assure sufficently internal memory pre-allocated to prevent
// multiple re-allocations during filling the vector:
destination.reserve(numberOfValues);
while(numberOfValues--)
{
numbers.push_back(rand() * 500 / RAND_MAX + 100);
// modulus is unprecise; this calculation will give you better
// distribution
// however, rather prefer modern C++ random number generators!
// IF you use rand: assure that you call srand, too, but exactly ONCE,
// best right when entering main function
}
}
// C++ random number generator:
void randNumbers(unsigned int numberOfValues, std::vector<int>& destination)
{
static std::uniform_int_distribution<> d(100, 599);
static std::mt19937 g;
destination.clear();
destination.reserve(numberOfValues);
while(numberOfValues--)
{
numbers.push_back(d(g));
}
}
Now you have contradicting function declarations:
int getLowest(const int[], int);
void getLowest(const int numbers[], int sizeofArray) { /* ... */ }
int lowest = getLowest(numbers, sizeofArray);
// again: the vector is yet empty!
// so you certainly won't get the result desired
// however, this won't compile at all: numbers is a std::vector,
// but parameter type is array, so you need:
int lowest = getLowest(numbers.data(), numbers.size());
// ^ replaced the redundant global as well
// move this into your main function AFTER having filled the vector!
// picking int as return value:
int getLowest(const int numbers[], unsigned int sizeofArray)
{
// you'd now have to initialize the global first; better, though:
// return a local variable:
// this assumes that there is at least one element in! check before usage
// and decide what would be the appropriate error handling if the vector
// is empty (return 0? return INT_MIN? throw an execption?)
int min = numbers[0];
for (int i = 1; i < sizeofArray; i++)
{
if (min > numbers[i])
{
min = numbers[i];
// min = lowest; // don't overwrite the minimum again!
}
}
// returning at end of void function is obsolete, don't do that explicitly
// well, with int as return value, as is NOW, you NEED to return:
return min;
}
Maximum analogously, be aware that you did not change the comparison from > to <! Be aware that there are already std::min_element, std::max_element and std::minmax_element which do the same (if not prohibited by the assignment, you should rather use these instead of re-inventing the wheel).
// prefere double! float (on typical machines at least) has same size as int
// and it is quite likely that you will lose precision due to rounding; I
// personally would rather use int64_t instead, so you won't run into rounding
// issues even with double and you'd need quite a large amount of summands
// before overflow can occur...
float getTotal(const int numbers) // just one single number???
{
total = sum(numbers[]);
// index operator cannot be applied on a single int; additionally, you need
// to provide an argument; where is 'sum' function defined at all???
return total;
}
// prefer double again
double getStandardDeviation(int sizeofArray, float numbers[]) // const
// (free standing functions cannot be const)
{
// mean isn't declared/defined anywhere (average instead?)!
// and you need to declare and initialize the sum appropriately:
double sum = 0.0;
float deviation1;
for (int i = 0; i < sizeofArray; i++)
sum += (mean - numbers[i]) * (mean - numbers[i]);
// ^ you need to add, if you want to build sum
// why two variables, even both of same type???
deviation1 = sqrt(sum / sizeofArray - 1);
float deviation = deviation1;
return deviation;
// simplest: drop both deviation and deviation 1 and just do:
return sqrt(sum / sizeofArray - 1);
}
Finally: I don't think that you'd use the resulting string (below) for anything else than printing out to console again, so I'd output to std::cout directly (naming the function 'print'); if at all, I'd provide a std::ostream as parameter to be more flexible:
void print(std::ostream& sout)
{
sout << "STATISTICAL ANALYSIS OF RANDOMLY GENERATED NUMBERS" << endl;
sout << "====================================================" << endl;
sout << left << "Lowest Number:" << left << getLowest() << endl;
sout << left << "Highest Number:" << left << getHighest() << endl;
sout << left << "Numbers Total:" << left << getTotal() << endl;
sout << left << "Numbers Averge:" << left << getAverage() << endl;
sout << left << "Numbers of Standard Deviation:" << left
<< getStandardDeviation() << endl;
}
Now you could pass std::cout to, a std::ostringstream object or even write to file via a std::ofstream...
int main()
{
Analyzer a, b, c; // b, c added by me for illustration only
a + 100;
// the operator accepts another Analyzer object, so you could do
c = a + b;
cout << a; // there's no operator<< overload for Analyzer class
// it is HERE where you'd call all your getXZY functions!
return 0;
}
