vtkTableToCoordinatesInternal Namespace Reference

Functions
bool	ComputeUniqueAndSorted (vtkDataArray *input, vtkDataArray *output)
	Fill output array with unique, sorted values from the given input.
void	ConstructSpace (vtkFieldData *space, vtkDataSetAttributes *inputPoints)
	Construct a N-dimension space from the list of points.
void	MapSpaceToInput (vtkFieldData *space, vtkDataSetAttributes *inputPoints, vtkIdTypeArray *inputToSpaceMap, vtkIdType invalidMapping)
	Map every point index from space to the randomly ordered and sparsed source arrays from inputPoints.
vtkIdType	GetPointIdxInSpace (vtkFieldData *space, vtkDataSetAttributes *inputPoints, vtkIdType inputPointId)
	Given a structured N-dimension space, compute the structured coordinates from an inputPoint and return its idx in space.
void	GetSurroundingPoints (vtkFieldData *space, const std::vector< double > &point, vtkIdTypeArray *pointIdx, vtkDoubleArray *weights)
	Given a random point, return its surrounding point in space with associated weight.

Function Documentation

ComputeUniqueAndSorted()

bool vtkTableToCoordinatesInternal::ComputeUniqueAndSorted	(	vtkDataArray *	input,
		vtkDataArray *	output )

Fill output array with unique, sorted values from the given input.

Return true on success, false on any error.

ConstructSpace()

void vtkTableToCoordinatesInternal::ConstructSpace	(	vtkFieldData *	space,
		vtkDataSetAttributes *	inputPoints )

Construct a N-dimension space from the list of points.

• inputPoints: a list of points in a N-dimension space. Each tuple of inputPoints is a point in N-dimension, N being the number of arrays.
• space: the output N-dimension space. Each output array in space contains unique and sorted values, extracted from the inputPoints array. In space, each array can have its own number of tuples (i.e. each axis has its own dimension) This can be seen as a N-dimension extension of a vtkRectilinearGrid.

Non numerical input arrays are skipped.

See also

ComputeUniqueAndSorted

MapSpaceToInput()

void vtkTableToCoordinatesInternal::MapSpaceToInput	(	vtkFieldData *	space,
		vtkDataSetAttributes *	inputPoints,
		vtkIdTypeArray *	inputToSpaceMap,
		vtkIdType	invalidMapping )

Map every point index from space to the randomly ordered and sparsed source arrays from inputPoints.

• space: an extension to N-dimension of a vtkRectilinearGrid. Thus, arrays in space should have unique and ordered values. But they can have different number of tuples.

• inputPoints: a list of point in space, without any requirement on order. So each arrays should have the same number of tuples.

• inputToSpaceMap: the map filled by this method. Some inputPoint may not exists in space. In that case, inputToSpaceMap will point to invalidMapping id.
• invalidMapping: the value to use in the map for invalid input point.

See also

ConstructSpace to get such space from the inputPoint container.

GetPointIdxInSpace()

vtkIdType vtkTableToCoordinatesInternal::GetPointIdxInSpace	(	vtkFieldData *	space,
		vtkDataSetAttributes *	inputPoints,
		vtkIdType	inputPointId )

Given a structured N-dimension space, compute the structured coordinates from an inputPoint and return its idx in space.

GetSurroundingPoints()

void vtkTableToCoordinatesInternal::GetSurroundingPoints	(	vtkFieldData *	space,
		const std::vector< double > &	point,
		vtkIdTypeArray *	pointIdx,
		vtkDoubleArray *	weights )

Given a random point, return its surrounding point in space with associated weight.

• space: as filled by ConstructSpace. Each array has unique, sorted values.
• point: coordinates of a random point in space.
• pointIdx: output variable with the list of surrounding point indices. In some way, they represent the structured cell that includes the given point.
• weights: output variable with the weights of each surrounding point, for linear interpolation purpose.