nidas::util::EndianConverter Class Reference

Virtual base class declaring methods for converting numeric values between little-endian and big-endian representations, and for determining the endian represenation of the host system. More...

#include <EndianConverter.h>

Inheritance diagram for nidas::util::EndianConverter:

Public Types
enum	endianness { EC_UNKNOWN_ENDIAN, EC_BIG_ENDIAN, EC_LITTLE_ENDIAN }

Public Member Functions
virtual	~EndianConverter ()
virtual double	doubleValue (const void *) const =0
	Get 8 byte double at address, do endian conversion. More...
virtual double	doubleValue (const double &) const =0
virtual float	floatValue (const void *) const =0
	Get 4 byte float at address, do endian conversion. More...
virtual float	floatValue (const float &) const =0
virtual int32_t	int32Value (const void *) const =0
	Get 4 byte int32 at address, do endian conversion. More...
virtual int32_t	int32Value (const int32_t &) const =0
virtual int64_t	int64Value (const void *) const =0
	Get 8 byte int64_t at address, do endian conversion. More...
virtual int64_t	int64Value (const int64_t &) const =0
virtual uint32_t	uint32Value (const void *) const =0
	Get 4 byte unsigned int32_t at address, do endian conversion. More...
virtual uint32_t	uint32Value (const uint32_t &) const =0
virtual int16_t	int16Value (const void *) const =0
virtual int16_t	int16Value (const int16_t &) const =0
virtual uint16_t	uint16Value (const void *) const =0
virtual uint16_t	uint16Value (const uint16_t &) const =0
virtual void	doubleCopy (const double &, void *) const =0
	Copy 8 byte double to the given address, doing endian conversion. More...
virtual void	floatCopy (const float &, void *) const =0
	Copy 4 byte float to the given address, doing endian conversion. More...
virtual void	int32Copy (const int32_t &, void *) const =0
	Copy 4 byte int to the given address, doing endian conversion. More...
virtual void	int64Copy (const int64_t &, void *) const =0
	Copy 8 byte int64_t to the given address, doing endian conversion. More...
virtual void	uint32Copy (const uint32_t &, void *) const =0
	Copy 4 byte unsigned int to the given address, doing endian conversion. More...
virtual void	int16Copy (const int16_t &, void *) const =0
virtual void	uint16Copy (const uint16_t &, void *) const =0

Static Public Member Functions
static endianness	getHostEndianness ()
	Return endianness value for this host. More...
static const EndianConverter *	getConverter (endianness input, endianness output)
	Return an EndianConverter for converting from one endian to another. More...
static const EndianConverter *	getConverter (endianness input)
	Return an EndianConverter for converting from an endian represenation to the endian representation of th host. More...

Static Public Attributes
static endianness	hostEndianness = EndianConverter::EC_UNKNOWN_ENDIAN

Static Private Member Functions
static endianness	privGetHostEndianness ()

Static Private Attributes
static EndianConverter *	flipConverter = 0
static EndianConverter *	noflipConverter = 0
static Mutex	staticInitMutex = Mutex()

Detailed Description

Virtual base class declaring methods for converting numeric values between little-endian and big-endian representations, and for determining the endian represenation of the host system.

Implementations of this class are meant for serializing/de-serializing binary data. The methods read from an address or write to an address. The addresses do not have to be aligned correctly for the data value type.

Member Enumeration Documentation

enum nidas::util::EndianConverter::endianness

Enumerator
EC_UNKNOWN_ENDIAN
EC_BIG_ENDIAN
EC_LITTLE_ENDIAN

Constructor & Destructor Documentation

virtual nidas::util::EndianConverter::~EndianConverter ( )

inlinevirtual

Member Function Documentation

virtual void nidas::util::EndianConverter::doubleCopy ( const double &  , void *    ) const

pure virtual

Copy 8 byte double to the given address, doing endian conversion.

Pointer to address does not need to be 8-byte aligned.

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

virtual double nidas::util::EndianConverter::doubleValue ( const void *  ) const

pure virtual

Get 8 byte double at address, do endian conversion.

Pointer to address does not need to be 8-byte aligned.

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

virtual double nidas::util::EndianConverter::doubleValue ( const double &  ) const

pure virtual

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

virtual void nidas::util::EndianConverter::floatCopy ( const float &  , void *    ) const

pure virtual

Copy 4 byte float to the given address, doing endian conversion.

Pointer to address does not need to be 4-byte aligned.

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

virtual float nidas::util::EndianConverter::floatValue ( const void *  ) const

pure virtual

Get 4 byte float at address, do endian conversion.

Pointer to address does not need to be 4-byte aligned.

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

virtual float nidas::util::EndianConverter::floatValue ( const float &  ) const

pure virtual

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

const EndianConverter * EndianConverter::getConverter ( EndianConverter::endianness  input, EndianConverter::endianness  output  )

static

Return an EndianConverter for converting from one endian to another.

If both are the same, then the converter that is returned just does memcpy's and does not change the representation. It is a const pointer since all its methods are const, and the pointer is owned by EndianConverter.

Referenced by nidas::dynld::isff::CSI_IRGA_Sonic::checkSampleTags(), nidas::dynld::isff::DAUSensor::init(), nidas::dynld::IEEE_Float::init(), nidas::dynld::DSC_Event::init(), nidas::dynld::DSC_PulseCounter::init(), and nidas::dynld::DSC_FreqCounter::init().

const EndianConverter * EndianConverter::getConverter ( EndianConverter::endianness  input )

static

Return an EndianConverter for converting from an endian represenation to the endian representation of th host.

If both are the same, then the converter that is returned just does memcpy's and does not change the representation. It is a const pointer since all its methods are const, and the pointer is owned by EndianConverter.

EndianConverter::endianness EndianConverter::getHostEndianness ( )

static

Return endianness value for this host.

Referenced by nidas::dynld::isff::CSI_IRGA_Sonic::checkSampleTags(), and nidas::dynld::IEEE_Float::init().

virtual void nidas::util::EndianConverter::int16Copy ( const int16_t &  , void *    ) const

pure virtual

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

virtual int16_t nidas::util::EndianConverter::int16Value ( const void *  ) const

pure virtual

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

Referenced by nidas::dynld::isff::CU_Coldwire::process(), and nidas::dynld::isff::CSI_CRX_Binary::process().

virtual int16_t nidas::util::EndianConverter::int16Value ( const int16_t &  ) const

pure virtual

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

virtual void nidas::util::EndianConverter::int32Copy ( const int32_t &  , void *    ) const

pure virtual

Copy 4 byte int to the given address, doing endian conversion.

Pointer to address does not need to be 4-byte aligned.

Implemented in nidas::util::FlipConverter.

virtual int32_t nidas::util::EndianConverter::int32Value ( const void *  ) const

pure virtual

Get 4 byte int32 at address, do endian conversion.

Pointer to address does not need to be 4-byte aligned.

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

Referenced by nidas::dynld::raf::TwoD64_USB::process(), nidas::dynld::raf::TwoD32_USB::process(), nidas::dynld::isff::CSI_CRX_Binary::process(), nidas::dynld::raf::TwoD32_USB::processImage(), nidas::dynld::raf::TwoD64_USB::processSOR(), and ExtractFast2D::run().

virtual int32_t nidas::util::EndianConverter::int32Value ( const int32_t &  ) const

pure virtual

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

virtual void nidas::util::EndianConverter::int64Copy ( const int64_t &  , void *    ) const

pure virtual

Copy 8 byte int64_t to the given address, doing endian conversion.

Pointer to address does not need to be 8-byte aligned.

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

virtual int64_t nidas::util::EndianConverter::int64Value ( const void *  ) const

pure virtual

Get 8 byte int64_t at address, do endian conversion.

Pointer to address does not need to be 8-byte aligned.

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

Referenced by nidas::dynld::raf::TwoD64_USB::processImageRecord().

virtual int64_t nidas::util::EndianConverter::int64Value ( const int64_t &  ) const

pure virtual

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

EndianConverter::endianness EndianConverter::privGetHostEndianness ( )

staticprivate

References VLOG.

virtual void nidas::util::EndianConverter::uint16Copy ( const uint16_t &  , void *    ) const

pure virtual

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

virtual uint16_t nidas::util::EndianConverter::uint16Value ( const void *  ) const

pure virtual

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

Referenced by nidas::dynld::isff::CU_Coldwire::process(), and nidas::dynld::isff::CSI_CRX_Binary::process().

virtual uint16_t nidas::util::EndianConverter::uint16Value ( const uint16_t &  ) const

pure virtual

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

virtual void nidas::util::EndianConverter::uint32Copy ( const uint32_t &  , void *    ) const

pure virtual

Copy 4 byte unsigned int to the given address, doing endian conversion.

Pointer to address does not need to be 4-byte aligned.

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

virtual uint32_t nidas::util::EndianConverter::uint32Value ( const void *  ) const

pure virtual

Get 4 byte unsigned int32_t at address, do endian conversion.

Pointer to address does not need to be 4-byte aligned.

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

Referenced by nidas::dynld::raf::UDPArincSensor::decodeTIMER(), nidas::dynld::isff::CU_Coldwire::process(), and nidas::dynld::raf::UDPArincSensor::process().

virtual uint32_t nidas::util::EndianConverter::uint32Value ( const uint32_t &  ) const

pure virtual

Implemented in nidas::util::NoFlipConverter, and nidas::util::FlipConverter.

Member Data Documentation

EndianConverter * EndianConverter::flipConverter = 0

staticprivate

EndianConverter::endianness EndianConverter::hostEndianness = EndianConverter::EC_UNKNOWN_ENDIAN

static

EndianConverter * EndianConverter::noflipConverter = 0

staticprivate

Mutex EndianConverter::staticInitMutex = Mutex()

staticprivate

---

The documentation for this class was generated from the following files:

• nidas/util/EndianConverter.h
• nidas/util/EndianConverter.cc