Namespaces
Variants
Views
Actions

C numeric limits interface

From cppreference.com
< cpp‎ | types
 
 
Utilities library
General utilities
Relational operators (deprecated in C++20)
 
Type support
Basic types
Fixed width integer types (C++11)
Fixed width floating-point types (C++23)
(C++11)    
(C++17)
Numeric limits
C numeric limits interface
Runtime type information
 

See also std::numeric_limits interface.

Contents

[edit] Limits of integer types

Limits of core language integer types
Defined in header <climits>
CHAR_BIT
number of bits in a byte
(macro constant) [edit]
MB_LEN_MAX
maximum number of bytes in a multibyte character
(macro constant) [edit]
CHAR_MIN
minimum value of char
(macro constant) [edit]
CHAR_MAX
maximum value of char
(macro constant) [edit]
SCHAR_MINSHRT_MININT_MINLONG_MINLLONG_MIN
(C++11)
minimum value of signed char, short, int, long and long long respectively
(macro constant) [edit]
SCHAR_MAXSHRT_MAXINT_MAXLONG_MAXLLONG_MAX
(C++11)
maximum value of signed char, short, int, long and long long respectively
(macro constant) [edit]
UCHAR_MAXUSHRT_MAXUINT_MAXULONG_MAXULLONG_MAX
(C++11)
maximum value of unsigned char, unsigned short, unsigned int,
unsigned long and unsigned long long respectively
(macro constant) [edit]
Defined in header <cwchar>
Defined in header <cstdint>
WCHAR_MIN
(C++11)
minimum value of wchar_t
(macro constant) [edit]
WCHAR_MAX
(C++11)
maximum value of wchar_t
(macro constant) [edit]
Limits of library type aliases
Defined in header <cstdint>
PTRDIFF_MIN
(C++11)
minimum value of std::ptrdiff_t
(macro constant) [edit]
PTRDIFF_MAX
(C++11)
maximum value of std::ptrdiff_t
(macro constant) [edit]
SIZE_MAX
(C++11)
maximum value of std::size_t
(macro constant) [edit]
SIG_ATOMIC_MIN
(C++11)
minimum value of std::sig_atomic_t
(macro constant) [edit]
SIG_ATOMIC_MAX
(C++11)
maximum value of std::sig_atomic_t
(macro constant) [edit]
WINT_MIN
(C++11)
minimum value of std::wint_t
(macro constant) [edit]
WINT_MAX
(C++11)
maximum value of std::wint_t
(macro constant) [edit]

[edit] Notes

The types of these constants, other than CHAR_BIT and MB_LEN_MAX, are required to match the results of the integral promotions as applied to objects of the types they describe: CHAR_MAX may have type int or unsigned int, but never char. Similarly USHRT_MAX may not be of an unsigned type: its type may be int.

A freestanding implementation may lack std::sig_atomic_t and/or std::wint_t typedef names, in which case the SIG_ATOMIC_* and/or WINT_* macros are correspondingly absent.

[edit] Example

#include <climits>
#include <cstdint>
#include <iomanip>
#include <iostream>
 
int main()
{
    constexpr int w = 14;
    std::cout << std::left;
#   define COUT(x) std::cout << std::setw(w) << #x << " = " << x << '\n'
 
    COUT( CHAR_BIT       );
    COUT( MB_LEN_MAX     );
    COUT( CHAR_MIN       );
    COUT( CHAR_MAX       );
    COUT( SCHAR_MIN      );
    COUT( SHRT_MIN       );
    COUT( INT_MIN        );
    COUT( LONG_MIN       );
    COUT( LLONG_MIN      );
    COUT( SCHAR_MAX      );
    COUT( SHRT_MAX       );
    COUT( INT_MAX        );
    COUT( LONG_MAX       );
    COUT( LLONG_MAX      );
    COUT( UCHAR_MAX      );
    COUT( USHRT_MAX      );
    COUT( UINT_MAX       );
    COUT( ULONG_MAX      );
    COUT( ULLONG_MAX     );
    COUT( PTRDIFF_MIN    );
    COUT( PTRDIFF_MAX    );
    COUT( SIZE_MAX       );
    COUT( SIG_ATOMIC_MIN );
    COUT( SIG_ATOMIC_MAX );
    COUT( WCHAR_MIN      );
    COUT( WCHAR_MAX      );
    COUT( WINT_MIN       );
    COUT( WINT_MAX       );
}

Possible output:

CHAR_BIT       = 8
MB_LEN_MAX     = 16
CHAR_MIN       = -128
CHAR_MAX       = 127
SCHAR_MIN      = -128
SHRT_MIN       = -32768
INT_MIN        = -2147483648
LONG_MIN       = -9223372036854775808
LLONG_MIN      = -9223372036854775808
SCHAR_MAX      = 127
SHRT_MAX       = 32767
INT_MAX        = 2147483647
LONG_MAX       = 9223372036854775807
LLONG_MAX      = 9223372036854775807
UCHAR_MAX      = 255
USHRT_MAX      = 65535
UINT_MAX       = 4294967295
ULONG_MAX      = 18446744073709551615
ULLONG_MAX     = 18446744073709551615
PTRDIFF_MIN    = -9223372036854775808
PTRDIFF_MAX    = 9223372036854775807
SIZE_MAX       = 18446744073709551615
SIG_ATOMIC_MIN = -2147483648
SIG_ATOMIC_MAX = 2147483647
WCHAR_MIN      = -2147483648
WCHAR_MAX      = 2147483647
WINT_MIN       = 0
WINT_MAX       = 4294967295

[edit] Limits of floating-point types

Defined in header <cfloat>
FLT_RADIX
the radix (integer base) used by the representation of all three floating-point types
(macro constant) [edit]
DECIMAL_DIG
(C++11)
conversion from long double to decimal with at least DECIMAL_DIG digits and back to long double is the identity conversion: this is the decimal precision required to serialize/deserialize a long double (see also std::numeric_limits::max_digits10)
(macro constant) [edit]
FLT_DECIMAL_DIGDBL_DECIMAL_DIGLDBL_DECIMAL_DIG
(C++17)
conversion from float/double/long double to decimal with at least FLT_DECIMAL_DIG/DBL_DECIMAL_DIG/LDBL_DECIMAL_DIG digits and back is the identity conversion: this is the decimal precision required to serialize/deserialize a floating-point value (see also std::numeric_limits::max_digits10). Defined to at least 6, 10, and 10 respectively, or 9 for IEEE float and 17 for IEEE double.
(macro constant) [edit]
FLT_MINDBL_MINLDBL_MIN
minimum normalized positive value of float, double and long double respectively
(macro constant) [edit]
FLT_TRUE_MINDBL_TRUE_MINLDBL_TRUE_MIN
(C++17)
minimum positive value of float, double and long double respectively
(macro constant) [edit]
FLT_MAXDBL_MAXLDBL_MAX
maximum finite value of float, double and long double respectively
(macro constant) [edit]
FLT_EPSILONDBL_EPSILONLDBL_EPSILON
difference between 1.0 and the next representable value for float, double and long double respectively
(macro constant) [edit]
FLT_DIGDBL_DIGLDBL_DIG
number of decimal digits that are guaranteed to be preserved in text → float/double/long double → text roundtrip without change due to rounding or overflow (see std::numeric_limits::digits10 for explanation)
(macro constant) [edit]
FLT_MANT_DIGDBL_MANT_DIGLDBL_MANT_DIG
number of base FLT_RADIX digits that can be represented without losing precision for float, double and long double respectively
(macro constant) [edit]
FLT_MIN_EXPDBL_MIN_EXPLDBL_MIN_EXP
minimum negative integer such that FLT_RADIX raised by power one less than that integer is a normalized float, double and long double respectively
(macro constant) [edit]
FLT_MIN_10_EXPDBL_MIN_10_EXPLDBL_MIN_10_EXP
minimum negative integer such that 10 raised to that power is a normalized float, double and long double respectively
(macro constant) [edit]
FLT_MAX_EXPDBL_MAX_EXPLDBL_MAX_EXP
maximum positive integer such that FLT_RADIX raised by power one less than that integer is a representable finite float, double and long double respectively
(macro constant) [edit]
FLT_MAX_10_EXPDBL_MAX_10_EXPLDBL_MAX_10_EXP
maximum positive integer such that 10 raised to that power is a representable finite float, double and long double respectively
(macro constant) [edit]
default rounding mode of floating-point arithmetic
(macro constant) [edit]
specifies in what precision all arithmetic operations are done
(macro constant) [edit]
FLT_HAS_SUBNORMDBL_HAS_SUBNORMLDBL_HAS_SUBNORM
(C++17)
specifies whether the type supports subnormal (denormal) numbers:
-1 – indeterminable, 0 – absent, 1 – present
(macro constant) [edit]

[edit] Example

#include <cfloat>
#include <iomanip>
#include <iostream>
 
int main()
{
    int w = 16;
    std::cout << std::left; // std::cout << std::setprecision(53);
#   define COUT(x) std::cout << std::setw(w) << #x << " = " << x << '\n'
 
    COUT( FLT_RADIX        );
    COUT( DECIMAL_DIG      );
    COUT( FLT_DECIMAL_DIG  );
    COUT( DBL_DECIMAL_DIG  );
    COUT( LDBL_DECIMAL_DIG );
    COUT( FLT_MIN          );
    COUT( DBL_MIN          );
    COUT( LDBL_MIN         );
    COUT( FLT_TRUE_MIN     );
    COUT( DBL_TRUE_MIN     );
    COUT( LDBL_TRUE_MIN    );
    COUT( FLT_MAX          );
    COUT( DBL_MAX          );
    COUT( LDBL_MAX         );
    COUT( FLT_EPSILON      );
    COUT( DBL_EPSILON      );
    COUT( LDBL_EPSILON     );
    COUT( FLT_DIG          );
    COUT( DBL_DIG          );
    COUT( LDBL_DIG         );
    COUT( FLT_MANT_DIG     );
    COUT( DBL_MANT_DIG     );
    COUT( LDBL_MANT_DIG    );
    COUT( FLT_MIN_EXP      );
    COUT( DBL_MIN_EXP      );
    COUT( LDBL_MIN_EXP     );
    COUT( FLT_MIN_10_EXP   );
    COUT( DBL_MIN_10_EXP   );
    COUT( LDBL_MIN_10_EXP  );
    COUT( FLT_MAX_EXP      );
    COUT( DBL_MAX_EXP      );
    COUT( LDBL_MAX_EXP     );
    COUT( FLT_MAX_10_EXP   );
    COUT( DBL_MAX_10_EXP   );
    COUT( LDBL_MAX_10_EXP  );
    COUT( FLT_ROUNDS       );
    COUT( FLT_EVAL_METHOD  );
    COUT( FLT_HAS_SUBNORM  );
    COUT( DBL_HAS_SUBNORM  );
    COUT( LDBL_HAS_SUBNORM );
}

Possible output:

FLT_RADIX        = 2
DECIMAL_DIG      = 21
FLT_DECIMAL_DIG  = 9
DBL_DECIMAL_DIG  = 17
LDBL_DECIMAL_DIG = 21
FLT_MIN          = 1.17549e-38
DBL_MIN          = 2.22507e-308
LDBL_MIN         = 3.3621e-4932
FLT_TRUE_MIN     = 1.4013e-45
DBL_TRUE_MIN     = 4.94066e-324
LDBL_TRUE_MIN    = 3.6452e-4951
FLT_MAX          = 3.40282e+38
DBL_MAX          = 1.79769e+308
LDBL_MAX         = 1.18973e+4932
FLT_EPSILON      = 1.19209e-07
DBL_EPSILON      = 2.22045e-16
LDBL_EPSILON     = 1.0842e-19
FLT_DIG          = 6
DBL_DIG          = 15
LDBL_DIG         = 18
FLT_MANT_DIG     = 24
DBL_MANT_DIG     = 53
LDBL_MANT_DIG    = 64
FLT_MIN_EXP      = -125
DBL_MIN_EXP      = -1021
LDBL_MIN_EXP     = -16381
FLT_MIN_10_EXP   = -37
DBL_MIN_10_EXP   = -307
LDBL_MIN_10_EXP  = -4931
FLT_MAX_EXP      = 128
DBL_MAX_EXP      = 1024
LDBL_MAX_EXP     = 16384
FLT_MAX_10_EXP   = 38
DBL_MAX_10_EXP   = 308
LDBL_MAX_10_EXP  = 4932
FLT_ROUNDS       = 1
FLT_EVAL_METHOD  = 0
FLT_HAS_SUBNORM  = 1
DBL_HAS_SUBNORM  = 1
LDBL_HAS_SUBNORM = 1

[edit] Defect reports

The following behavior-changing defect reports were applied retroactively to previously published C++ standards.

DR Applied to Behavior as published Correct behavior
LWG 416 C++98 it was unclear whether the types of the macros in <climits> are guaranteed
to match the type to which they refer (C++ refers to C, and C says no)
clarified as not
guaranteed

[edit] See also

C documentation for Numeric limits