SAMRAI::tbox::MathUtilities< TYPE > Class Template Reference

#include <source/toolbox/base/MathUtilities.h>

List of all members.

Public Member Functions
template<>
bool	isNaN (const float &value)
template<>
bool	isNaN (const double &value)
template<>
bool	Rand (bool low, bool width)
template<>
char	Rand (char low, char width)
template<>
int	Rand (int low, int width)
template<>
float	Rand (float low, float width)
template<>
double	Rand (double low, double width)
template<>
dcomplex	Rand (dcomplex low, dcomplex width)
template<>
dcomplex	Min (dcomplex a, dcomplex b)
template<>
dcomplex	Max (dcomplex a, dcomplex b)
Static Public Member Functions
static TYPE	getZero ()
	Get the value 0.
static TYPE	getOne ()
	Get the value 1.
static TYPE	getSignalingNaN ()
	Get the IEEE signaling NaN on architectures that support it.
static bool	isNaN (const TYPE &value)
	Indicates whether the supplied value is NaN.
static TYPE	getMax ()
	Get max for the templated type.
static TYPE	getMin ()
	Get min for the templated type.
static TYPE	getEpsilon ()
	Get epsilon for the templated type.
static TYPE	getUndefined ()
	Get value to set for undefined data.
static TYPE	Min (TYPE a, TYPE b)
	Compute the minimum of a and b.
static TYPE	Max (TYPE a, TYPE b)
	Compute the maximum of a and b.
static TYPE	Rand (TYPE low, TYPE width)
	Generate a random value from low to low+width.

---

Detailed Description

template<class TYPE>
class SAMRAI::tbox::MathUtilities< TYPE >

Class MathUtilities is a utility for managing basic math functions and the initialization of data to signaling NaNs, and for managing Posix constants like INT_MAX, FLT_MAX, DBL_MAX, etc. Signaling NaNs force a trap if they are used in a numerical operation, so they are a useful way to track uninitialized floating point data. Setting integer values to INT_MAX is a useful way to track uninitialized integer values.

The implementation of this class depends heavily on the particular computer architecture and how it implements floating point arithmetic and hardware traps.

This class has a similiar purpose to

See also:

tbox::IEEE. The primary purpose for replication is due to the use of variable type names in the method names in the IEEE class. This prevents templating based on type. This class is templated on type so it can be used by templated classes.

---

Member Function Documentation

template<class TYPE> TYPE SAMRAI::tbox::MathUtilities< TYPE >::getZero (   )  [inline, static]

Get the value 0.

template<class TYPE> TYPE SAMRAI::tbox::MathUtilities< TYPE >::getOne (   )  [inline, static]

Get the value 1.

template<class TYPE> TYPE SAMRAI::tbox::MathUtilities< TYPE >::getSignalingNaN (   )  [inline, static]

Get the IEEE signaling NaN on architectures that support it. Using this value in a numerical expression will cause a program abort. Valid for float/double only. For non float/double will return false.

template<class TYPE> bool SAMRAI::tbox::MathUtilities< TYPE >::isNaN (  const TYPE &  value  )  [inline, static]

Indicates whether the supplied value is NaN. Valid for float/double only. For non float/double will return false. Parameters: value  Value to test

template<class TYPE> TYPE SAMRAI::tbox::MathUtilities< TYPE >::getMax (   )  [inline, static]

Get max for the templated type.

template<class TYPE> TYPE SAMRAI::tbox::MathUtilities< TYPE >::getMin (   )  [inline, static]

Get min for the templated type.

template<class TYPE> TYPE SAMRAI::tbox::MathUtilities< TYPE >::getEpsilon (   )  [inline, static]

Get epsilon for the templated type.

template<class TYPE> TYPE SAMRAI::tbox::MathUtilities< TYPE >::getUndefined (   )  [inline, static]

Get value to set for undefined data. Typicaly used for initialziation during debugging.

template<class TYPE> TYPE SAMRAI::tbox::MathUtilities< TYPE >::Min (  TYPE  a, TYPE  b )  [inline, static]

Compute the minimum of a and b. Parameters: a  b

template<class TYPE> TYPE SAMRAI::tbox::MathUtilities< TYPE >::Max (  TYPE  a, TYPE  b )  [inline, static]

Compute the maximum of a and b. Parameters: a  b

template<class TYPE> static TYPE SAMRAI::tbox::MathUtilities< TYPE >::Rand (  TYPE  low, TYPE  width )  [static]

Generate a random value from low to low+width. Parameters: low  Starting value for range width  Width of the range.

template<> bool SAMRAI::tbox::MathUtilities< float >::isNaN (  const float &  value  )

template<> bool SAMRAI::tbox::MathUtilities< double >::isNaN (  const double &  value  )

template<> bool SAMRAI::tbox::MathUtilities< bool >::Rand (  bool  low, bool  width )

template<> char SAMRAI::tbox::MathUtilities< char >::Rand (  char  low, char  width )

template<> int SAMRAI::tbox::MathUtilities< int >::Rand (  int  low, int  width )

template<> float SAMRAI::tbox::MathUtilities< float >::Rand (  float  low, float  width )

template<> double SAMRAI::tbox::MathUtilities< double >::Rand (  double  low, double  width )

template<> dcomplex SAMRAI::tbox::MathUtilities< dcomplex >::Rand (  dcomplex  low, dcomplex  width )

template<> dcomplex SAMRAI::tbox::MathUtilities< dcomplex >::Min (  dcomplex  a, dcomplex  b )

template<> dcomplex SAMRAI::tbox::MathUtilities< dcomplex >::Max (  dcomplex  a, dcomplex  b )

---

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

• source/toolbox/base/MathUtilities.h
• source/toolbox/base/MathUtilities.I

---