FP_NORMAL, FP_SUBNORMAL, FP_ZERO, FP_INFINITE, FP_NAN

 

C++

语言

标准库头文件

概念

工具库

字符串库

容器库

算法库

迭代器库

数值库

输入/输出库

本地化库

正则表达式库 (C++11)

原子操作库 (C++11)

线程支持库 (C++11)

文件系统库 (C++17)

技术规范

 

数值库

常用数学函数

特殊数学函数 (C++17)

浮点环境 (C++11)

复数

数值数组

伪随机数生成

编译时有理数算术 (C++11)

数值算法

gcd (C++17)

lcm (C++17)

通用数值运算

iota (C++11)

accumulate

inner_product

adjacent_difference

partial_sum

 

常用数学函数

函数
基本运算
abs(int)labsllabsimaxabs (C++11) abs(float)fabs divldivlldivimaxdiv (C++11) fmod	remainder (C++11) remquo (C++11) fma (C++11) fmax (C++11) fmin (C++11) fdim (C++11) nannanfnanl (C++11) (C++11) (C++11)
指数函数
exp exp2 (C++11) expm1 (C++11)	log log10 log1p (C++11) log2 (C++11)
幂函数
sqrt cbrt (C++11)	hypot (C++11) pow
三角与双曲函数
sin cos tan asin acos atan atan2	sinh cosh tanh asinh (C++11) acosh (C++11) atanh (C++11)

误差与伽马函数
erf (C++11) erfc (C++11)	lgamma (C++11) tgamma (C++11)
临近整数的浮点运算
ceil floor roundlroundllround (C++11) (C++11) (C++11)	trunc (C++11) nearbyint (C++11) rintlrintllrint (C++11) (C++11) (C++11)
浮点操作函数
ldexp scalbnscalbln (C++11) (C++11) ilogb (C++11) logb (C++11)	frexp modf nextafternexttoward (C++11) (C++11) copysign (C++11)
分类/比较
fpclassify (C++11) isfinite (C++11) isinf (C++11) isnan (C++11) isnormal (C++11) signbit (C++11)	isgreater (C++11) isgreaterequal (C++11) isless (C++11) islessequal (C++11) islessgreater (C++11) isunordered (C++11)
宏常量
HUGE_VALFHUGE_VALHUGE_VALL (C++11) (C++11) INFINITY (C++11) NAN (C++11) math_errhandlingMATH_ERRNOMATH_ERREXCEPT (C++11)	FP_NORMALFP_SUBNORMALFP_ZEROFP_INFINITEFP_NAN (C++11) (C++11) (C++11) (C++11) (C++11)

 

定义于头文件 <cmath>
#define FP_NORMAL    /*implementation defined*/		(C++11 起)
#define FP_SUBNORMAL /*implementation defined*/		(C++11 起)
#define FP_ZERO      /*implementation defined*/		(C++11 起)
#define FP_INFINITE  /*implementation defined*/		(C++11 起)
#define FP_NAN       /*implementation defined*/		(C++11 起)

FP_NORMAL 、 FP_SUBNORMAL 、 FP_ZERO 、 FP_INFINITE 、 FP_NAN 宏各代表一个独自的浮点数类别。它们都展开成整数常量表达式。

常量	解释
FP_NORMAL	指示值为正规，即不是无穷大、非正规、非数或零
FP_SUBNORMAL	指示值为非正规
FP_ZERO	指示值为正或负零
FP_INFINITE	指示值无法以底层类型表示（正或负无穷大）
FP_NAN	指示值是非数（ NaN ）

示例

运行此代码

#include <iostream>
#include <cmath>
#include <cfloat>
 
const char* show_classification(double x) {
    switch(std::fpclassify(x)) {
        case FP_INFINITE:  return "Inf";
        case FP_NAN:       return "NaN";
        case FP_NORMAL:    return "normal";
        case FP_SUBNORMAL: return "subnormal";
        case FP_ZERO:      return "zero";
        default:           return "unknown";
    }
}
int main()
{
    std::cout << "1.0/0.0 is " << show_classification(1/0.0) << '\n'
              << "0.0/0.0 is " << show_classification(0.0/0.0) << '\n'
              << "DBL_MIN/2 is " << show_classification(DBL_MIN/2) << '\n'
              << "-0.0 is " << show_classification(-0.0) << '\n'
              << "1.0 is " << show_classification(1.0) << '\n';
}

输出：

1.0/0.0 is Inf
0.0/0.0 is NaN
DBL_MIN/2 is subnormal
-0.0 is zero
1.0 is normal

参阅

fpclassify (C++11)	归类给定的浮点值 (函数)
FP_categories的 C 文档