Printable View
Hi everyone. I'm just learning C++. I know that you can write code that will convert a decimal number into an integer, by just dropping the number after the decimal. But how would you write it if you want it to actually round the decimal off to the nearest integer... i.e.- instead of turning 7.9 into 7..... so that 7.9 turns into 8??? Hopefully that makes sense.
Look up the std::nint() [nearest integer] function in <cmath>.
Forgive me, but where do I look that up? Is that something I look up on the compiler?Quote:
Originally Posted by grumpy
<cmath> is described in the C++ standard (which means C++ compilers and libraries support it). The std::nint() function is declared in that header. Alternatively, you can use the <math.h> header from C (which C++ is backward compatible to) and the nint() function in that.
If you have a help file with your compiler, you will usually find the standard library described in that. Otherwise, most basic texts will give you an introduction to functions in the standard library: so look up nint() in the index of such books.
In this case, nint() is a function of the prototype "int nint(double);" or something equivalent for your compiler. The only difference is that the version in <cmath> is placed in namespace std, and the one in <math.h> is not. Those headers provide a number of basic mathematical functions.
Just in case the following doesn't help, here's are three examples of usage (all equivalent, in terms of results).
orCode:// The standard C++ way #1
#include <cmath>
int main()
{
double x = 2.7;
int i = std::nint(x); // i will have a value of 3
return 0;
}
orCode:// The standard C++ way #2
#include <cmath>
using namespace std; // Use this with caution; there are some traps
int main()
{
double x = 2.7;
int i = nint(x); // i will have a value of 3
return 0;
}
Code:/* A standard C way. C++ supports this, but it is often considered bad style in C++ */
#include <math.h>
int main()
{
double x = 2.7;
int i = nint(x); // i will have a value of 3
return 0;
}
simpler
Code:#include <iostream>
#define round_it(d) (static_cast<int>((d)+0.5))
int main()
{
std::cout<<round_it(7.9)<<std::endl;
std::cout<<round_it(2.34)<<std::endl;
std::cout<<round_it(10.001)<<std::endl;
return 0;
}
I disagree that your approach is "simpler". static_cast is a fairly advanced operation, technically. Using it for a simple mathematical operation is using a very big hammer to crack a very small nut.
Using macros in C++ is often viewed as VERY bad style.
Yuck Xerath.
Write a function not a macro.
Code:int round_it(int d)
{
return std::floor(d+0.5);
}
Code:#include <iostream>
#include <cmath>
using namespace std;
int main ()
{
int sum;
int average;
double dec1;
double dec2;
double dec3;
double dec4;
double dec5;
cout << "Enter five decimal numbers: ";
cin >> dec1 >> dec2 >> dec3 >> dec4 >> dec5;
cout << endl;
cout << "The numbers you entered are: \n"
cout << dec1 << dec2 << dec3 << dec4 << dec5;
cout << endl;
double a = dec1
double b = dec2
double c = dec3
double d = dec4
double e = dec5
int h = nint(a);
int i = nint(b);
int j = nint(c);
int k = nint(d);
int l = nint(e);
sum = h + i + j + k + l;
average = (h + i + j + k + l) / 5
cout << endl;
cout << "Sum: " << sum << endl;
cout << "Average: " << average << endl;
return 0;
}
~~~ This is returning a lot of errors for me now. I hope you guys can see what this program is doing and can hopefully give me some advice.... there's probably a way to make it shorter, I just don't know how to do it! You guys have been a great help and I really appreciate it.
since nint apparently doesnt exist in cmath why not just use the code mentioned to create it for yourself...
Code:#include <iostream>
#include <cmath>
using namespace std;
int nint(double d)
{
return std::floor(d+0.5);
}
int main ()
{
int sum;
int average;
double dec1;
double dec2;
double dec3;
double dec4;
double dec5;
cout << "Enter five decimal numbers: ";
cin >> dec1 >> dec2 >> dec3 >> dec4 >> dec5;
cout << endl;
cout << "The numbers you entered are: \n"; //forgot a semicolon
cout << dec1 << " " << dec2 << " " << dec3 << " " << dec4 << " " << dec5; //you probably want spaces between
cout << endl;
double a = dec1; //forgot a semicolon
double b = dec2; //forgot a semicolon
double c = dec3; //forgot a semicolon
double d = dec4; //forgot a semicolon
double e = dec5; //forgot a semicolon
//since there is no nint in cmath we use our own nint function
int h = nint(a);
int i = nint(b);
int j = nint(c);
int k = nint(d);
int l = nint(e);
sum = h + i + j + k + l;
average = sum / 5; //forgot a semicolon and why not just use the sum? and not do it again
cout << endl;
cout << "Sum: " << sum << endl;
cout << "Average: " << average << endl;
cin.ignore(); //catch return
cin.get(); //just to keep output open
return 0;
}
i don't understand why i need more semicolons. They're all there where they need to be, aren't they? I didn't think they went right after "cout" or "double" like that. And what is the "cin.ignore();" and "cin.get();" for?
Well every statement in c++ ends with a semicolon, so basically your cout statements where skipping down untill the first semicolon, so the compiler detected a syntax error. Dont worry about cin.ignore() or cin.get() i just stuck them there so you could see the output even though you arent having trouble with it, it helped me see it so i didnt have to open up the command prompt.
Thank you so much! It works now!
why are macros 'bad style' ? enlighten me with something.
Well for starters they have no respect for scope whatsoever. You should always prefer inline functions to macros.remember macros are expanded before the compiler even sees the code. behaviour is always predictable with inline functions whereas not so with macros.Read meyers effective c++ for more info or in fact just about any book on good coding stylein c++.
I agree, but that isn't a issue to an experienced programmer.Quote:
Originally Posted by Stoned_Coder
But if you want to call it bad style, fine by me.
It is bad style. So ok :).
It's more than bad style. And macros do cause problems, even with experienced programmers. Microsoft added max and min macros to their windows header, and the C++ standards committee added max and min functions to the algorithm header. They cannot co-exist because of the deficiencies of macros.
http://www.parashift.com/c++-faq-lit...s.html#faq-9.5
incredibly the latest VC++ versions are getting standard compliant, so probably those macros will disapear in about 2/3 years :)
Quote:
Originally Posted by prog-bman
this can also be bad stlye.Code:float x;
int i = *(int*)&x;
It all up to the programmers responsability. There's nothing wrong with well used and identified macros.
I doubt it in this case. The macros are in a non-standard header.Quote:
Originally Posted by xErath
It also yields unspecified behaviour. While unspecified behaviour is not evil, it does mean the result is compiler dependent. If a style guide is concerned with portability (e.g. achieving the same results regardless of compiler), I suppose it might describe such tricks might be described as bad behaviour.Quote:
Originally Posted by xErath
Depends on your viewpoint.Quote:
Originally Posted by xErath
Macros, if misused, can be extremely dangerous, as they do text substitution that can make innocent looking code do something it doesn't appear to. Unintended misuse is probably the most dangerous: leaving a trap for some other programmer to encounter and get confused by is probably rarely a good idea. If you think about it, the problem that started this thread is an example can be viewed as an example of unintended misuse.
Technically, there are VERY few things that can be achieved with macros for which C++ does not provide safe alternatives. The only examples where one absolutely can't live without the preprocessor are with #include'ing headers, ability to define include guards and (arguably) macros such as __LINE__ and __FILE__.