simpnumer.h File Reference

Include dependency graph for simpnumer.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Namespaces
	Interpolate
	Taper
	Taper an indexable array from 1 to taperfactor. If lowpos is less than highpos, the samples array[0] to array[lowpos] will be set to zero. If lowpos is more than highpos, the samples array[lowpos] to array[sz-1] will be set to zero. The taper can be either cosine or linear.

Enumerations
enum	Taper::Type { Taper::Cosine, Taper::Linear }

Functions
int	Interpolate::getArrIdxPosition (const float farridx, const int arrsz, float &relpos, const float eps=1e-4)
template<class T >
T	greatestCommonDivisor (T val0, T val1)
template<class T >
StepInterval< T >	getCommonStepInterval (const StepInterval< T > &a, const StepInterval< T > &b)
template<class iT >
iT	exactPower (iT val, iT base)
template<class iT >
iT	nextPower (iT val, iT base)
template<class iT >
iT	getPow2Sz (iT actsz, const bool above=true, iT minsz=1, iT maxsz=0)
template<class iT >
iT	nextPower2 (iT nr, iT minnr, iT maxnr)
template<class iT >
iT	nrBlocks (iT totalsamples, iT basesize, iT overlapsize)
template<class T >
bool	taperArray (T *array, int sz, int lowpos, int highpos, Taper::Type type)
template<class T >
T	sampledGradient (T ym2, T ym1, T y0_, T y1_, T y2)
template<class X , class Y , class RT >
void	getGradient (const X &x, const Y &y, int sz, int firstx, int firsty, RT *gradptr=0, RT *interceptptr=0)
template<class X >
float	variance (const X &x, int sz)
int	solve3DPoly (double a, double b, double c, double &root0, double &root1, double &root2)
template<class T >
bool	holdsClassValue (const T val, const unsigned int maxclss=50)
template<class T >
bool	holdsClassValues (const T *vals, od_int64 sz, const unsigned int maxclss=50, const unsigned int samplesz=100)
template<class T >
bool	isSigned8BitesValue (const T val)
template<class T >
bool	is8BitesData (const T *vals, od_int64 sz, const unsigned int samplesz=100)

Function Documentation

template<class iT >

iT exactPower ( iT  val, iT  base  )

inline

exactPower determines whether a value is a power of a base, i.e. if

val=base^pow. If that is the case, exactPower returns pow, if not, it returns zero.

template<class T >

StepInterval<T> getCommonStepInterval ( const StepInterval< T > &  a, const StepInterval< T > &  b  )

inline

Computes a StepInterval where all samples of both input StepInterval's

are present.

template<class X , class Y , class RT >

void getGradient ( const X &  x, const Y &  y, int  sz, int  firstx, int  firsty, RT *  gradptr = 0, RT *  interceptptr = 0  )

inline

template<class iT >

iT getPow2Sz ( iT  actsz, const bool  above = true, iT  minsz = 1, iT  maxsz = 0  )

inline

template<class T >

T greatestCommonDivisor ( T  val0, T  val1  )

inline

Computes the greatest common divisor from two intigers. Uses the algorithm published by Josef Stein.

template<class T >

bool holdsClassValue ( const T  val, const unsigned int  maxclss = 50  )

inline

template<class T >

bool holdsClassValues ( const T *  vals, od_int64  sz, const unsigned int  maxclss = 50, const unsigned int  samplesz = 100  )

inline

template<class T >

bool is8BitesData ( const T *  vals, od_int64  sz, const unsigned int  samplesz = 100  )

inline

template<class T >

bool isSigned8BitesValue ( const T  val )

inline

template<class iT >

iT nextPower ( iT  val, iT  base  )

inline

template<class iT >

iT nextPower2 ( iT  nr, iT  minnr, iT  maxnr  )

inline

template<class iT >

iT nrBlocks ( iT  totalsamples, iT  basesize, iT  overlapsize  )

inline

Find number of blocks when given total number of samples, the base size for each block and the number of samples overlaped between two blocks.

template<class T >

T sampledGradient ( T  ym2, T  ym1, T  y0_, T  y1_, T  y2  )

inline

Gradient from a series of 4 points, sampled like: x: -2 -1 0 1 2 y: ym2 ym1 y0 y1 y2 The gradient estimation is done at x=0. y0 is generally not needed, but it will be used if there are one or more undefineds. The function will return mUdf(T) if there are too many missing values.

int solve3DPoly ( double  a, double  b, double  c, double &  root0, double &  root1, double &  root2  )

inline

solve3DPoly - finds the roots of the equation

z^3+a*z^2+b*z+c = 0

solve3DPoly returns the number of real roots found.

Algorithms taken from NR, page 185.

The complex roots can be calculated as follows: const double sqrt3_through_2 = Math::Sqrt(3)/2;

root1 = complex_double( 0.5*(A+B)+minus_a_through_3, sqrt3_through_2*(A-B));

root2 = complex_double( 0.5*(A+B)+minus_a_through_3, -sqrt3_through_2*(A-B));

template<class T >

bool taperArray	(	T *	array,
		int	sz,
		int	lowpos,
		int	highpos,
		Taper::Type	type
	)

template<class X >

float variance ( const X &  x, int  sz  )

inline