C++ Help

[Cruise Control]

I need help in C++. I'm trying to figure out how to detect if a number entered by the user is an integer.

BTW we can't have equations in IF statements.

I don't know how to go about this or what to do from here.

//Chris Taylor
//2/6/06 (0010/0001/0110)
//Terminal 2
//Integer Recognition Program

//Standard header files
#include <iostream.h>
#include <conio.h>
#include <fstream.h>
#include <string.h>
#include <iomanip.h>


int main()
{
 ofstream outfile ("IntegerProgram.txt"); //Creates textfile
 cout << "This is the semi-godlike Integer Detection Program. It will allow you to enter a number, then determine if  the entered number is an intege.\n";


 float integer;
 int remainder;

 cout << "Please enter your number: ";
 cin >> integer;

 remainder = integer/2;

 cout << remainder;

 getch();
 return 0;
 }

[Flying Mullet]

At first glance, I believe that doing a mod by 1 will work. That will return any remainders you may have, then you can compare the remainder against 0 to see if there is one.

cout << "Please enter your number: ";
 cin >> integer;

 remainder = integer%1;
if(remainder == 0)
{
// do integer stuff
}
else
{
// do non-integer stuff
}

[Cruise Control]

Actually, my teacher said I need to typecast a non integer in order to use modulus, what the hell is that? And how do I do this.

[Flying Mullet]

Typcasting is "forcing" data of one type into another. For example, below I am forcing a float of 1.5 to an int of 1.

float number1 = 1.5;
int number2 = (int)number1;

And if you did the opposite, typecast the number 1 (an int) to a float, you would get 1.0.

I'm pretty sure you could do operator overloading to typecast entire objects, but it's been years since I've worked with C++.

And apparently you can't mod a float. I've researched modulus a little more on the web. Here's a thread in another fourm I found on the issue:
http://www.gamedev.net/community/for...opic_id=306374

Like I said, it's been a while since I've used C++ so I'm still a bit rusty. Hope this helps.

[Cruise Control]

It's not working, god I suck. I should reread the entire book.

int main()
{
 ofstream outfile ("IntegerProgram.txt"); //Creates textfile
 cout << "This is the semi-godlike Integer Detection Program. It will allow you to enter a number, then determine if  the entered number is an intege.\n";


float integer;
int remainder = (int)integer;

cout << "Please enter your number: ";
 cin >> integer;

 remainder = integer%1;

 if(remainder == 0)
 {
  cout << "Yes, " << integer << "is an integer.";
 }
 else
 {
  cout << "I'm sorry, " << integer << "is not an integer.";
 }


 getch();
 return 0;
 }

[Flying Mullet]

Well you could always use a brute force approach.

1) Treat the input as a string.
2) Go through the input character by character and verify that each character is only 0-9 or a ".".
3) If there is a ".", check that there is only one ".".
4) If there is no decimal and it's a valid number then you have an integer.
5) If there is a decimal, see if there is any number other than zero following it.
6) If there are only zeros, then you have an integer, otherwise it's a float.

It's not as pretty but it would work.

[Samuraid]

I would personally read the number in as a string and parse the digits to see if they were all numeric and if it had a decimal point (which would mean it was a float not an int). Then one could simply cast it to an int.

EDIT: Ah Mulley you beat me.

[Dr Unne]

How about testing if x - (int)x == 0.0 .

Well you could always use a brute force approach.

1) Treat the input as a string.
2) Go through the input character by character and verify that each character is only 0-9 or a ".".
3) If there is a ".", check that there is only one ".".
4) If there is no decimal and it's a valid number then you have an integer.
5) If there is a decimal, see if there is any number other than zero following it.
6) If there are only zeros, then you have an integer, otherwise it's a float.

It's not as pretty but it would work.

I think it depends on the definition of "is an integer". If you input 1.0, to store it into an int variable it would have to be cast as an int from a float or double at some point. 1.0 is technically a floating point number, mathematically equal to 1 but maybe not computationally equal. Strictly speaking I wouldn't call 1.0 an integer.

Using your definition above though, 1e2 would also be an integer.

My C is more than a bit rusty (and spoiled from Perl) but I think you can enter integers in straight hexidecimal in C, such as 0x100. Or in binary, 0b1001.

You can also have + or - at the beginning of an integer.

[Samuraid]

How about testing if x - (int)x == 0.0 .

Great idea.

I'm not sure if scanf() accepts all number formats inclusively. You would have to specifically scan for either hex or decimal with the format specifier. If I remember correctly, you have to use something like strtol() to be able to input direct hex/octal numbers as well as numbers with leading + and -.

[Flying Mullet]

Great idea.

Yes, very elegant.

[Cruise Control]

How about testing if x - (int)x == 0.0 .

That was brutally efficient, if I wasn't so sure you'd stab me I'd give you a hug.

You helped a friend and I out immensely.

[Mr. Wertz]

Yes, you helped me alot. Lots of props to you.

[Cruise Control]

Can someone give me the coordinates to make a Zanarkand Abes Logo. It's my optional shape for a graphics program.

[edczxcvbnm]

Why don't you print out the logo get some graph paper and plot it out yourself? That is a simple solution.

Note: I have no played FFX thus can't even begin to help with coordinates but I am also very art impared.

[Cruise Control]

Me too. I don't know where to start.