#include <IEEEFloat32.h>
Public Types | |
| typedef float | FloatType |
| This typedef specifies the native type that will be used to do computations internally. | |
Public Member Functions | |
| IEEEFloat32 () | |
| Default constructor initializes to 0.0;. | |
| IEEEFloat32 (FloatType value) | |
| This constructor initializes the IEEEFloat32 instance to the value specified by its argument. | |
| IEEEFloat32 (unsigned char byte0, unsigned char byte1, unsigned char byte2, unsigned char byte3) | |
| This constructor initializes the IEEEFloat32 instance using its 32 bit binary representation. | |
| IEEEFloat32 (const IEEEFloat32 &source) | |
| This is the copy constructor. | |
| ~IEEEFloat32 () | |
| The destructor destroys the class instance and cleans up any storage. | |
| operator FloatType () const | |
| This conversion operator returns the float as a built-in type. | |
| unsigned char | getByte (size_t index0) |
| This member function returns the requested 8 bits byte from the IEEE floating point representation. | |
| FloatType | getFloat () |
| This member function returns an instance of FloatType having the same value as *this. | |
| void | setValue (FloatType value) |
| This member function sets the IEEEFloat32 instance to the value specified by its argument. | |
| void | setValue (unsigned char byte0, unsigned char byte1, unsigned char byte2, unsigned char byte3) |
| This member function sets the IEEEFloat32 instance using the 32 bit binary representation. | |
This class is useful if, for example, you need to figure out * what specific pattern of bits represents the number 1.347 in 32 bit * IEEE floating point format, and you're not sure that your machine * uses IEEE floating point representation.
The 32-bit IEEE floating point format is one sign bit (s), followed by an 8-bit biased exponent, followed by a fraction from the normalized mantissa (f). This is broken into bytes as follows:
| seeeeeee | efffffff | ffffffff | ffffffff |
This arrangements of bits represents a real number, F, according to the following formula:
F = (-1)^s * 2E(e - 127) * (1.f)_2,
where the notation (1.f)_2 means "the 24 bit binary number consisting of 1, followed by the 23 bits of f, where the 23 bits of f represent the fractional part. That is the first bit of f has a weight of 2E-1, the second has a weight of 2E-2, and so on.
WARNING: This class currently does not handle NaN or Inf values. Furthermore, the value -0.0 is silently converted to 0.0.
Definition at line 53 of file IEEEFloat32.h.
| typedef float dlr::numeric::IEEEFloat32::FloatType |
This typedef specifies the native type that will be used to do computations internally.
It must have at least as many exponent and mantissa bits as IEEE 32-bit floating point representation.
Definition at line 62 of file IEEEFloat32.h.
| dlr::numeric::IEEEFloat32::IEEEFloat32 | ( | ) |
Default constructor initializes to 0.0;.
Definition at line 27 of file IEEEFloat32.cpp.
References setValue().
| dlr::numeric::IEEEFloat32::IEEEFloat32 | ( | FloatType | value | ) |
This constructor initializes the IEEEFloat32 instance to the value specified by its argument.
| value | This argument specifies the value of the float in question. |
Definition at line 39 of file IEEEFloat32.cpp.
References setValue().
| dlr::numeric::IEEEFloat32::IEEEFloat32 | ( | unsigned char | byte0, | |
| unsigned char | byte1, | |||
| unsigned char | byte2, | |||
| unsigned char | byte3 | |||
| ) |
This constructor initializes the IEEEFloat32 instance using its 32 bit binary representation.
| byte0 | This argument represents the first 8 bits of the binary representation (the sign bit and the first 7 exponent bits). | |
| byte1 | This argument represents the second 8 bits of the binary representation (the final exponent bit and the first 7 mantissa bits). | |
| byte2 | This argument represents the third 8 bits of the binary representation (the 8th - 15th mantissa bits). | |
| byte3 | This argument represents the final 8 bits of the binary representation (the 16th - 23rd mantissa bits). |
Definition at line 51 of file IEEEFloat32.cpp.
References setValue().
| dlr::numeric::IEEEFloat32::IEEEFloat32 | ( | const IEEEFloat32 & | source | ) |
This is the copy constructor.
It deep copies its argument.
| source | This argument is the IEEEFloat32 instance to be copied. |
Definition at line 65 of file IEEEFloat32.cpp.
References m_bytes.
| dlr::numeric::IEEEFloat32::~IEEEFloat32 | ( | ) | [inline] |
The destructor destroys the class instance and cleans up any storage.
Definition at line 118 of file IEEEFloat32.h.
| unsigned char dlr::numeric::IEEEFloat32::getByte | ( | size_t | index0 | ) |
This member function returns the requested 8 bits byte from the IEEE floating point representation.
| index0 | This argument specifies which byte to return. If its value is zero, the sign bit and 1st 7 exponent bits will be returned. If its value is one, the final exponent bit and the first 7 mantissa bits will be returned. If its value is two, the subsequent 8 mantissa bits will be returned. If its value is two, the final 8 mantissa bits will be returned. |
Definition at line 77 of file IEEEFloat32.cpp.
| FloatType dlr::numeric::IEEEFloat32::getFloat | ( | ) | [inline] |
This member function returns an instance of FloatType having the same value as *this.
It is provided for those times when an implicit type conversion isn't possible, and a static_cast is too clunky.
Definition at line 159 of file IEEEFloat32.h.
| dlr::numeric::IEEEFloat32::operator FloatType | ( | ) | const [inline] |
This conversion operator returns the float as a built-in type.
Definition at line 128 of file IEEEFloat32.h.
| void dlr::numeric::IEEEFloat32::setValue | ( | unsigned char | byte0, | |
| unsigned char | byte1, | |||
| unsigned char | byte2, | |||
| unsigned char | byte3 | |||
| ) |
This member function sets the IEEEFloat32 instance using the 32 bit binary representation.
| byte0 | This argument represents the first 8 bits of the binary representation (the sign bit and the first 7 exponent bits). | |
| byte1 | This argument represents the second 8 bits of the binary representation (the final exponent bit and the first 7 mantissa bits). | |
| byte2 | This argument represents the third 8 bits of the binary representation (the 8th - 15th mantissa bits). | |
| byte3 | This argument represents the final 8 bits of the binary representation (the 16th - 23rd mantissa bits). |
Definition at line 118 of file IEEEFloat32.cpp.
| void dlr::numeric::IEEEFloat32::setValue | ( | FloatType | value | ) |
This member function sets the IEEEFloat32 instance to the value specified by its argument.
| value | This argument specifies the value of the float in question. |
Definition at line 94 of file IEEEFloat32.cpp.
Referenced by IEEEFloat32().
1.5.8