W3cubDocs

/C++

std::rint, std::rintf, std::rintl, std::lrint, std::lrintf, std::lrintl, std::llrint, std::llrintf

Defined in header `<cmath>`
float rint ( float arg ); float rintf( float arg );	(1)	(since C++11)
double rint ( double arg );	(2)	(since C++11)
long double rint ( long double arg ); long double rintl( long double arg );	(3)	(since C++11)
double rint ( IntegralType arg );	(4)	(since C++11)
long lrint ( float arg ); long lrintf( float arg );	(5)	(since C++11)
long lrint ( double arg );	(6)	(since C++11)
long lrint ( long double arg ); long lrintl( long double arg );	(7)	(since C++11)
long lrint ( IntegralType arg );	(8)	(since C++11)
long long llrint ( float arg ); long long llrintf( float arg );	(9)	(since C++11)
long long llrint ( double arg );	(10)	(since C++11)
long long llrint ( long double arg ); long long llrintl( long double arg );	(11)	(since C++11)
long long llrint ( IntegralType arg );	(12)	(since C++11)

1-3) Rounds the floating-point argument arg to an integer value (in floating-point format), using the current rounding mode.

5-7, 9-11) Rounds the floating-point argument arg to an integer value, using the current rounding mode.

4,8,12) A set of overloads or a function template accepting an argument of any integral type. Equivalent to (2,6,10), respectively (the argument is cast to double).

Parameters

arg

floating point value

Return value

If no errors occur, the nearest integer value to arg, according to the current rounding mode, is returned.

Error handling

Errors are reported as specified in math_errhandling.

If the result of std::lrint or std::llrint is outside the range representable by the return type, a domain error or a range error may occur.

If the implementation supports IEEE floating-point arithmetic (IEC 60559), For the std::rint function:

If arg is ±∞, it is returned, unmodified
If arg is ±0, it is returned, unmodified
If arg is NaN, NaN is returned

For std::lrint and std::llrint functions:

If arg is ±∞, FE_INVALID is raised and an implementation-defined value is returned
If the result of the rounding is outside the range of the return type, FE_INVALID is raised and an implementation-defined value is returned
If arg is NaN, FE_INVALID is raised and an implementation-defined value is returned

Notes

POSIX specifies that all cases where std::lrint or std::llrint raise FE_INEXACT are domain errors.

As specified in math_errhandling, FE_INEXACT may be (but isn't required to be on non-IEEE floating-point platforms) raised by std::rint when rounding a non-integer finite value.

The only difference between std::rint and std::nearbyint is that std::nearbyint never raises FE_INEXACT.

The largest representable floating-point values are exact integers in all standard floating-point formats, so std::rint never overflows on its own; however the result may overflow any integer type (including std::intmax_t), when stored in an integer variable.

If the current rounding mode is...

FE_DOWNWARD, then std::rint is equivalent to std::floor.
FE_UPWARD, then std::rint is equivalent to std::ceil.
FE_TOWARDZERO, then std::rint is equivalent to std::trunc
FE_TONEAREST, then std::rint differs from std::round in that halfway cases are rounded to even rather than away from zero.

Example

#include <iostream>
#include <cmath>
#include <cfenv>
#include <climits>
 
int main()
{
#pragma STDC FENV_ACCESS ON
    std::fesetround(FE_TONEAREST);
    std::cout << "rounding to nearest (halfway cases to even):\n"
              << "rint(+2.3) = " << std::rint(2.3)
              << "  rint(+2.5) = " << std::rint(2.5)
              << "  rint(+3.5) = " << std::rint(3.5) << '\n'
              << "rint(-2.3) = " << std::rint(-2.3)
              << "  rint(-2.5) = " << std::rint(-2.5)
              << "  rint(-3.5) = " << std::rint(-3.5) << '\n';
 
    std::fesetround(FE_DOWNWARD);
    std::cout << "rounding down:\n" 
              << "rint(+2.3) = " << std::rint(2.3)
              << "  rint(+2.5) = " << std::rint(2.5)
              << "  rint(+3.5) = " << std::rint(3.5) << '\n'
              << "rint(-2.3) = " << std::rint(-2.3)
              << "  rint(-2.5) = " << std::rint(-2.5)
              << "  rint(-3.5) = " << std::rint(-3.5) << '\n'
              << "rounding down with lrint\n" 
              << "lrint(+2.3) = " << std::lrint(2.3)
              << "  lrint(+2.5) = " << std::lrint(2.5)
              << "  lrint(+3.5) = " << std::lrint(3.5) << '\n'
              << "lrint(-2.3) = " << std::lrint(-2.3)
              << "  lrint(-2.5) = " << std::lrint(-2.5)
              << "  lrint(-3.5) = " << std::lrint(-3.5) << '\n';
 
    std::cout << "lrint(-0.0) = " << std::lrint(-0.0)  << '\n'
              << "lrint(-Inf) = " << std::lrint(-INFINITY) << '\n';
 
    // error handling
    std::feclearexcept(FE_ALL_EXCEPT);
    std::cout << "std::rint(0.1) = " << std::rint(.1) << '\n';
    if (std::fetestexcept(FE_INEXACT))
              std::cout << "    FE_INEXACT was raised\n";
 
    std::feclearexcept(FE_ALL_EXCEPT);
    std::cout << "std::lrint(LONG_MIN-2048.0) = "
              << std::lrint(LONG_MIN-2048.0) << '\n';
    if (std::fetestexcept(FE_INVALID))
              std::cout << "    FE_INVALID was raised\n";
}

Possible output:

rounding to nearest (halfway cases to even): 
rint(+2.3) = 2  rint(+2.5) = 2  rint(+3.5) = 4
rint(-2.3) = -2  rint(-2.5) = -2  rint(-3.5) = -4
rounding down:
rint(+2.3) = 2  rint(+2.5) = 2  rint(+3.5) = 3
rint(-2.3) = -3  rint(-2.5) = -3  rint(-3.5) = -4
rounding down with lrint
lrint(+2.3) = 2  lrint(+2.5) = 2  lrint(+3.5) = 3
lrint(-2.3) = -3  lrint(-2.5) = -3  lrint(-3.5) = -4
lrint(-0.0) = 0
lrint(-Inf) = -9223372036854775808
std::rint(0.1) = 0
    FE_INEXACT was raised
std::lrint(LONG_MIN-2048.0) = -9223372036854775808
    FE_INVALID was raised