Battlefront3.net

[Scott]

Thanks for the help. I'm just starting on C++ and am getting stuck quite often. I'll probably end up join a dedicated forum or mailbag soon. If you don't mind, I do have one more question...

In the following code:

Code: Select all

#include <iostream>

using namespace std;

int Divide(unsigned short int x, unsigned short int y);

int main()
{
    unsigned short int x, y, divide;
    
    cout << "Please enter a number...\n";
    cin >> x;
    cout << "Please enter another number...\n";
    cin >> y;
    
    divide = Divide(x,y);
    
    if (divide != -1)
    cout << x << " / " << y << " = " << divide << endl;
    else
    cout << "Error, cannot divide by 0";
  
    char response;
    cin >> response;
    return 0;
}

int Divide(unsigned short int x, unsigned short int y)
{
    if (y == 0)
    return -1;
    else
    return (x/y);
}

I get the error: 18 [Warning] comparison is always true due to limited range of data type. I understand what this means but no matter what I do with the if statements it always ends being always false or always true.

[Matthew]

Your function returns a signed integer which is right but the division variable is an unsigned short integer. This means it can't hold a negative value.

Change this:

Code: Select all

unsigned short int x, y, divide;

To this:

Code: Select all

unsigned short int x, y;
int divide;

That will make the divide variable a signed 4 byte (32 bit) integer. You can't use a signed short integer if you will be using unsigned short integers for the arguments because the answer could potentially be above the range of the signed short datatype. A 4 byte integer should be used in this case.

Datatypes can get confusing. It's best to familiarise yourself with the data ranges of different data types.

[Scott]

Didn't notice I had different integer types. Thanks again

[Matthew]

You could have the result as an unsigned short integer and use a flag to represent division by 0, keeping it all in a structure. That may be unnecessary but I think the solution is more elegant and sophisticated that way. I'm going to look into C++ programming again because I want to try out that idea.

I haven't touched C++ for almost a year now. I was halfway through a project in making a menu GUI library for the PSP. I gave up when I replaced the project with a python project.

It took me a while because I didn't know "Y" was different from 'Y'. The later being a char which is what I wanted but here it is.

Code: Select all

#include <iostream>

using namespace std;

struct divide_answer{
	unsigned short int val;
	bool nan;
};
divide_answer divide(unsigned short int x, unsigned short int y){
	divide_answer ans;
	if (y == 0){
		ans.nan = true;
	}else{
		ans.nan = false;
		ans.val = x/y;
	}
	return ans;
}
int main(){
	unsigned short int x, y;
	char leave_ans;
	while(true){
		divide_answer resultant;
		cout << "Please enter number 1: ";
		cin >> x;
		cout << "Please enter number 2: ";
		cin >> y;
		resultant = divide(x,y);
		if(resultant.nan){
			cout << "Error: Cannot divide by 0";
		}else{
			cout << x << " / " << y << " = " << resultant.val;
		}
		cout << "\n";
		while(true){
			cout << "Do you wish to leave? (Y/N): ";
			cin >> leave_ans;
			if(leave_ans == 'Y'){
				exit(0);
			}else if(leave_ans == 'N'){
				break;
			}
		}
	}
}

To show you how much easier python is at the same thing:

Code: Select all

def divide(x,y):
	if y == 0:
		return False
	else:
		return x/y
while 1:
	num1 = raw_input("Please enter number 1: ")
	num2 = raw_input("Please enter number 2: ")
	ans = divide(int(num1),int(num2))
	if ans:
		print num1 + " / " + num2 + " = " + str(ans)
	else:
		print "Error: Division by zero not possible."
	while 1:
		ans = raw_input("Do you wish to leave? (Y/N): ")
		if ans == "Y":
			raise (SystemExit)
		elif ans == "N":
			break

It does pretty much exactly the same thing.

[Scott]

I considered learning Python after I had learned some Ruby but decided to try C++ instead. It is a very difficult language. One specific thing I'm confused about is the practical function of function recursion. I don't see how this could help when there are easier ways to get most of the tasks that recursion would be used for, done. Anyway, thanks again for the help.

[Matthew]

Recursion is very useful. Take this factorial function in python as a good simple example:

Code: Select all

def factorial(n):
	if n == 0:
		return 1
	elif n < 0:
		return n * factorial(abs(n) - 1)
	else:
		return n * factorial(n - 1)

>>> factorial(0)
1
>>> factorial(1)
1
>>> factorial(2)
2
>>> factorial(3)
6
>>> factorial(4)
24
>>> factorial(5)
120
>>> factorial(-1)
-1
>>> factorial(-2)
-2
>>> factorial(-3)
-6
>>> factorial(-4)
-24
>>> factorial(-5)
-120
>>>

[Scott]

Well I know of uses (Fibonacci number, powers, etc) but I can't see when I would use it. Although I said the same about arrays when I first started Ruby.

[Matthew]

In my game I've made a recursive function for several things. One of the things is the function which processes a bullet line. It calls itself again for ricochets. Another one was to recursively update the offset values for my Surface objects when I redraw their parent at another position.

One thing I should do all the time now but didn't used to, is set constants for magic numbers.

For my game I am very likely going to rewrite parts in C++. You can write python modules in C++, so it's very easy to extend python with C++. Well I'm hoping anyway.

[Scott]

Well I'm just in the beginning stages of C++ and am trying to use simpler techniques. Right now I'm trying to make Tic-Tac-Toe which is harder than I expected...

[Matthew]

I tried to make battleships in VB at college. That was command line based but I was trying to make a nice interface and I never finished it. I do get to the stage of places the ships in a grid. The grid was made out of lots of dots and the ships were represented by letters.

Will you use a command based idea or an interactive grid?

[Scott]

I'm using a command line interface. The controls are basically the number pad. 1 is the bottom left, 5 is the middle, etc. I'm storing all of the info in arrays and want to eventually have a computer opponent, but I need to learn more about random numbers in C++ first.

[Matthew]

C++ uses a simple rand() function but it gives you an integer. To convert it to a floating point number with the range (n) 0 < n < 1 you'll need to divide the result by RAND_MAX which is the maximum result. Of-course you'll need to convert to float .

If you want a random integer you don't need to convert to a floating pint number first, you can simply times the result by your range and then divide by RAND_MAX.

[Scott]

Do you know how to make small noises relatively simply? By scanning some source I found \a which makes a beep for every \a. It's neat to have for some programs (like the cola machine program I made when practicing with switches).

[Matthew]

I'm sure a simple library can't be too difficult to find.

[Scott]

That would be an option but I'd like to stay close to the STL for now.