std::logb,std::logbf,std::logbl (3) - Linux Manuals

std::logb,std::logbf,std::logbl: std::logb,std::logbf,std::logbl

NAME

std::logb,std::logbf,std::logbl - std::logb,std::logbf,std::logbl

Synopsis

Defined in header <cmath>
float logb ( float arg ); (1) (since C++11)
float logbf( float arg );
double logb ( double arg ); (2) (since C++11)
long double logb ( long double arg ); (3) (since C++11)
long double logbl( long double arg );
double logb ( IntegralType arg ); (4) (since C++11)

1-3) Extracts the value of the unbiased radix-independent exponent from the floating-point argument arg, and returns it as a floating-point value.
4) A set of overloads or a function template accepting an argument of any integral_type. Equivalent to (2) (the argument is cast to double).
Formally, the unbiased exponent is the signed integral part of log
r|arg| (returned by this function as a floating-point value), for non-zero arg, where r is std::numeric_limits<T>::radix and T is the floating-point type of arg. If arg is subnormal, it is treated as though it was normalized.

Parameters

arg - floating point value

Return value

If no errors occur, the unbiased exponent of arg is returned as a signed floating-point value.
If a domain error occurs, an implementation-defined value is returned
If a pole error occurs, -HUGE_VAL, -HUGE_VALF, or -HUGE_VALL is returned.

Error handling

Errors are reported as specified in math_errhandling.
Domain or range error may occur if arg is zero.
If the implementation supports IEEE floating-point arithmetic (IEC 60559),

* If arg is ±0, -∞ is returned and FE_DIVBYZERO is raised.
* If arg is ±∞, +∞ is returned
* If arg is NaN, NaN is returned.
* In all other cases, the result is exact (FE_INEXACT is never raised) and the_current_rounding_mode is ignored

Notes

POSIX_requires that a pole error occurs if arg is ±0.
The value of the exponent returned by std::logb is always 1 less than the exponent retuned by std::frexp because of the different normalization requirements: for the exponent e returned by std::logb, |arg*r-e
| is between 1 and r (typically between 1 and 2), but for the exponent e returned by std::frexp, |arg*2-e
| is between 0.5 and 1.

Example

Compares different floating-point decomposition functions
// Run this code

  #include <iostream>
  #include <cmath>
  #include <limits>
  #include <cfenv>
  #pragma STDC FENV_ACCESS ON
  int main()
  {
      double f = 123.45;
      std::cout << "Given the number " << f << " or " << std::hexfloat
                << f << std::defaultfloat << " in hex,\n";

      double f3;
      double f2 = std::modf(f, &f3);
      std::cout << "modf() makes " << f3 << " + " << f2 << '\n';

      int i;
      f2 = std::frexp(f, &i);
      std::cout << "frexp() makes " << f2 << " * 2^" << i << '\n';

      i = std::ilogb(f);
      std::cout << "logb()/ilogb() make " << f/std::scalbn(1.0, i) << " * "
                << std::numeric_limits<double>::radix
                << "^" << std::ilogb(f) << '\n';

      // error handling
      std::feclearexcept(FE_ALL_EXCEPT);
      std::cout << "logb(0) = " << std::logb(0) << '\n';
      if (std::fetestexcept(FE_DIVBYZERO))
          std::cout << " FE_DIVBYZERO raised\n";
  }

Possible output:

  Given the number 123.45 or 0x1.edccccccccccdp+6 in hex,
  modf() makes 123 + 0.45
  frexp() makes 0.964453 * 2^7
  logb()/ilogb() make 1.92891 * 2^6
  logb(0) = -Inf
      FE_DIVBYZERO raised