vtkHDFReader::Implementation Class Reference

Implementation for the vtkHDFReader. More...

#include <vtkHDFReaderImplementation.h>

Classes
struct	TypeDescription
	Used to store HDF native types in a map. More...

Public Member Functions
	Implementation (vtkHDFReader *reader)
virtual	~Implementation ()
bool	Open (VTK_FILEPATH const char *fileName)
	Opens this VTK HDF file and checks if it is valid.
void	Close ()
	Closes the VTK HDF file and releases any allocated resources.
int	GetDataSetType ()
	Type of vtkDataSet stored by the HDF file, such as VTK_IMAGE_DATA or VTK_UNSTRUCTURED_GRID, from vtkTypes.h.
const std::array< int, 2 > &	GetVersion ()
	Returns the version of the VTK HDF implementation.
template<typename T >
bool	GetAttribute (const char *attributeName, size_t numberOfElements, T *value)
	Reads an attribute from the /VTKHDF group.
template<typename T >
bool	GetAttribute (hid_t group, const char *attributeName, size_t numberOfElements, T *value)
	Reads an attribute from the group passed to it.
int	GetNumberOfPieces (vtkIdType step=-1)
	Returns the number of partitions for this dataset at the time step step if applicable.
bool	GetPartitionExtent (hsize_t partitionIndex, int *extent)
	For an ImageData, sets the extent for 'partitionIndex'.
std::vector< std::string >	GetArrayNames (int attributeType)
	Returns the names of arrays for 'attributeType' (point or cell).
std::vector< hsize_t >	GetDimensions (const char *dataset)
	Returns the dimensions of a HDF dataset.
bool	FillAMR (vtkOverlappingAMR *data, unsigned int maximumLevelsToReadByDefault, double origin[3], vtkDataArraySelection *dataArraySelection[3])
	Fills the given AMR data with the content of the opened HDF file.
vtkIdType	GetArrayOffset (vtkIdType step, int attributeType, std::string name)
	Methods to query for array offsets when steps are present.
vtkDataArray *	NewArray (int attributeType, const char *name, const std::vector< hsize_t > &fileExtent)
	Reads and returns a new vtkDataArray.
vtkDataArray *	NewArray (int attributeType, const char *name, hsize_t offset, hsize_t size)
	Reads and returns a new vtkDataArray.
vtkAbstractArray *	NewFieldArray (const char *name, vtkIdType offset=-1, vtkIdType size=-1)
	Reads and returns a new vtkDataArray.
vtkDataArray *	NewMetadataArray (const char *name, hsize_t offset, hsize_t size)
	Reads a 1D metadata array in a DataArray or a vector of vtkIdType.
std::vector< vtkIdType >	GetMetadata (const char *name, hsize_t size, hsize_t offset=0)
	Reads a 1D metadata array in a DataArray or a vector of vtkIdType.
std::size_t	GetNumberOfSteps ()
	Read the number of steps from the opened file.
std::size_t	GetNumberOfSteps (hid_t group)
	Read the number of steps from the opened file.
vtkDataArray *	GetStepValues ()
	Read the values of the steps from the open file.
vtkDataArray *	GetStepValues (hid_t group)
	Read the values of the steps from the open file.

Protected Member Functions
hid_t	OpenDataSet (hid_t group, const char *name, hid_t *nativeType, std::vector< hsize_t > &dims)
	Opens the hdf5 dataset given the 'group' and 'name'.
template<typename T >
hid_t	TemplateTypeToHdfNativeType ()
	Convert C++ template type T to HDF5 native type this can be constexpr in C++17 standard.
template<typename T >
vtkDataArray *	NewVtkDataArray ()
	Create a vtkDataArray based on the C++ template type T.
void	BuildTypeReaderMap ()
	Builds a map between native types and GetArray routines for that type.
TypeDescription	GetTypeDescription (hid_t type)
	Associates a struc of three integers with HDF type.
vtkDataArray *	NewArrayForGroup (hid_t group, const char *name, const std::vector< hsize_t > &fileExtent)
	Reads a vtkDataArray of type T from the attributeType, dataset The array has type 'T' and 'numberOfComponents'.
vtkDataArray *	NewArrayForGroup (hid_t dataset, hid_t nativeType, const std::vector< hsize_t > &dims, const std::vector< hsize_t > &fileExtent)
	Reads a vtkDataArray of type T from the attributeType, dataset The array has type 'T' and 'numberOfComponents'.
template<typename T >
vtkDataArray *	NewArray (hid_t dataset, const std::vector< hsize_t > &fileExtent, hsize_t numberOfComponents)
	Reads a vtkDataArray of type T from the attributeType, dataset The array has type 'T' and 'numberOfComponents'.
template<typename T >
bool	NewArray (hid_t dataset, const std::vector< hsize_t > &fileExtent, hsize_t numberOfComponents, T *data)
	Reads a vtkDataArray of type T from the attributeType, dataset The array has type 'T' and 'numberOfComponents'.
vtkStringArray *	NewStringArray (hid_t dataset, hsize_t size)
	Reads a vtkDataArray of type T from the attributeType, dataset The array has type 'T' and 'numberOfComponents'.

Detailed Description

Implementation for the vtkHDFReader.

Opens, closes and reads information from a VTK HDF file.

Definition at line 28 of file vtkHDFReaderImplementation.h.

Constructor & Destructor Documentation

Implementation()

vtkHDFReader::Implementation::Implementation

(

vtkHDFReader *

reader

)

~Implementation()

virtual vtkHDFReader::Implementation::~Implementation ( )

virtual

Member Function Documentation

Open()

bool vtkHDFReader::Implementation::Open

(

VTK_FILEPATH const char *

fileName

)

Opens this VTK HDF file and checks if it is valid.

Close()

void vtkHDFReader::Implementation::Close

(

)

Closes the VTK HDF file and releases any allocated resources.

GetDataSetType()

int vtkHDFReader::Implementation::GetDataSetType ( )

inline

Type of vtkDataSet stored by the HDF file, such as VTK_IMAGE_DATA or VTK_UNSTRUCTURED_GRID, from vtkTypes.h.

Definition at line 45 of file vtkHDFReaderImplementation.h.

GetVersion()

const std::array< int, 2 > & vtkHDFReader::Implementation::GetVersion ( )

inline

Returns the version of the VTK HDF implementation.

Definition at line 49 of file vtkHDFReaderImplementation.h.

GetAttribute() [1/2]

template<typename T >

bool vtkHDFReader::Implementation::GetAttribute	(	const char *	attributeName,
		size_t	numberOfElements,
		T *	value
	)

Reads an attribute from the /VTKHDF group.

GetAttribute() [2/2]

template<typename T >

bool vtkHDFReader::Implementation::GetAttribute	(	hid_t	group,
		const char *	attributeName,
		size_t	numberOfElements,
		T *	value
	)

Reads an attribute from the group passed to it.

GetNumberOfPieces()

int vtkHDFReader::Implementation::GetNumberOfPieces

(

vtkIdType

step = -1

)

Returns the number of partitions for this dataset at the time step step if applicable.

GetPartitionExtent()

bool vtkHDFReader::Implementation::GetPartitionExtent	(	hsize_t	partitionIndex,
		int *	extent
	)

For an ImageData, sets the extent for 'partitionIndex'.

Returns true for success and false otherwise.

GetArrayNames()

std::vector< std::string > vtkHDFReader::Implementation::GetArrayNames

(

int

attributeType

)

Returns the names of arrays for 'attributeType' (point or cell).

NewArray() [1/4]

vtkDataArray * vtkHDFReader::Implementation::NewArray	(	int	attributeType,
		const char *	name,
		const std::vector< hsize_t > &	fileExtent
	)

Reads and returns a new vtkDataArray.

The actual type of the array depends on the type of the HDF array. The array is read from the PointData or CellData groups depending on the 'attributeType' parameter. There are two versions: a first one that reads from a 3D array using a fileExtent, and a second one that reads from a linear array using an offset and size. The array has to be deleted by the user.

NewArray() [2/4]

vtkDataArray * vtkHDFReader::Implementation::NewArray	(	int	attributeType,
		const char *	name,
		hsize_t	offset,
		hsize_t	size
	)

Reads and returns a new vtkDataArray.

The actual type of the array depends on the type of the HDF array. The array is read from the PointData or CellData groups depending on the 'attributeType' parameter. There are two versions: a first one that reads from a 3D array using a fileExtent, and a second one that reads from a linear array using an offset and size. The array has to be deleted by the user.

NewFieldArray()

vtkAbstractArray * vtkHDFReader::Implementation::NewFieldArray	(	const char *	name,
		vtkIdType	offset = -1,
		vtkIdType	size = -1
	)

Reads and returns a new vtkDataArray.

The actual type of the array depends on the type of the HDF array. The array is read from the PointData or CellData groups depending on the 'attributeType' parameter. There are two versions: a first one that reads from a 3D array using a fileExtent, and a second one that reads from a linear array using an offset and size. The array has to be deleted by the user.

NewMetadataArray()

vtkDataArray * vtkHDFReader::Implementation::NewMetadataArray	(	const char *	name,
		hsize_t	offset,
		hsize_t	size
	)

Reads a 1D metadata array in a DataArray or a vector of vtkIdType.

We read either the whole array for the vector version or a slice specified with (offset, size). For an error we return nullptr or an empty vector.

GetMetadata()

std::vector< vtkIdType > vtkHDFReader::Implementation::GetMetadata	(	const char *	name,
		hsize_t	size,
		hsize_t	offset = 0
	)

Reads a 1D metadata array in a DataArray or a vector of vtkIdType.

We read either the whole array for the vector version or a slice specified with (offset, size). For an error we return nullptr or an empty vector.

GetDimensions()

std::vector< hsize_t > vtkHDFReader::Implementation::GetDimensions

(

const char *

dataset

)

Returns the dimensions of a HDF dataset.

FillAMR()

bool vtkHDFReader::Implementation::FillAMR	(	vtkOverlappingAMR *	data,
		unsigned int	maximumLevelsToReadByDefault,
		double	origin[3],
		vtkDataArraySelection *	dataArraySelection[3]
	)

Fills the given AMR data with the content of the opened HDF file.

The number of level to read is limited by the maximumLevelsToReadByDefault argument. maximumLevelsToReadByDefault == 0 means to read all levels (no limit). Only the selected data array in dataArraySelection are added to the AMR data. Returns true on success.

GetNumberOfSteps() [1/2]

std::size_t vtkHDFReader::Implementation::GetNumberOfSteps

(

)

Read the number of steps from the opened file.

GetNumberOfSteps() [2/2]

std::size_t vtkHDFReader::Implementation::GetNumberOfSteps

(

hid_t

group

)

Read the number of steps from the opened file.

GetStepValues() [1/2]

vtkDataArray * vtkHDFReader::Implementation::GetStepValues

(

)

Read the values of the steps from the open file.

GetStepValues() [2/2]

vtkDataArray * vtkHDFReader::Implementation::GetStepValues

(

hid_t

group

)

Read the values of the steps from the open file.

GetArrayOffset()

vtkIdType vtkHDFReader::Implementation::GetArrayOffset	(	vtkIdType	step,
		int	attributeType,
		std::string	name
	)

Methods to query for array offsets when steps are present.

OpenDataSet()

hid_t vtkHDFReader::Implementation::OpenDataSet ( hid_t  group, const char *  name, hid_t *  nativeType, std::vector< hsize_t > &  dims  )

protected

Opens the hdf5 dataset given the 'group' and 'name'.

Returns the hdf dataset and sets 'nativeType' and 'dims'.

TemplateTypeToHdfNativeType()

template<typename T >

hid_t vtkHDFReader::Implementation::TemplateTypeToHdfNativeType ( )

protected

Convert C++ template type T to HDF5 native type this can be constexpr in C++17 standard.

NewVtkDataArray()

template<typename T >

vtkDataArray * vtkHDFReader::Implementation::NewVtkDataArray ( )

protected

Create a vtkDataArray based on the C++ template type T.

For instance, for a float we create a vtkFloatArray. this can be constexpr in C++17 standard

NewArrayForGroup() [1/2]

vtkDataArray * vtkHDFReader::Implementation::NewArrayForGroup ( hid_t  group, const char *  name, const std::vector< hsize_t > &  fileExtent  )

protected

Reads a vtkDataArray of type T from the attributeType, dataset The array has type 'T' and 'numberOfComponents'.

We are reading fileExtent slab from the array. It returns the array or nullptr in case of an error. There are three cases for fileExtent: fileExtent.size() == 0 - in this case we expect a 1D array and we read the whole array. Used for field arrays. fileExtent.size()>>1 == ndims - in this case we read a scalar fileExtent.size()>>1 + 1 == ndims - in this case we read an array with the number of components > 1.

NewArrayForGroup() [2/2]

vtkDataArray * vtkHDFReader::Implementation::NewArrayForGroup ( hid_t  dataset, hid_t  nativeType, const std::vector< hsize_t > &  dims, const std::vector< hsize_t > &  fileExtent  )

protected

Reads a vtkDataArray of type T from the attributeType, dataset The array has type 'T' and 'numberOfComponents'.

We are reading fileExtent slab from the array. It returns the array or nullptr in case of an error. There are three cases for fileExtent: fileExtent.size() == 0 - in this case we expect a 1D array and we read the whole array. Used for field arrays. fileExtent.size()>>1 == ndims - in this case we read a scalar fileExtent.size()>>1 + 1 == ndims - in this case we read an array with the number of components > 1.

NewArray() [3/4]

template<typename T >

vtkDataArray * vtkHDFReader::Implementation::NewArray ( hid_t  dataset, const std::vector< hsize_t > &  fileExtent, hsize_t  numberOfComponents  )

protected

Reads a vtkDataArray of type T from the attributeType, dataset The array has type 'T' and 'numberOfComponents'.

We are reading fileExtent slab from the array. It returns the array or nullptr in case of an error. There are three cases for fileExtent: fileExtent.size() == 0 - in this case we expect a 1D array and we read the whole array. Used for field arrays. fileExtent.size()>>1 == ndims - in this case we read a scalar fileExtent.size()>>1 + 1 == ndims - in this case we read an array with the number of components > 1.

NewArray() [4/4]

template<typename T >

bool vtkHDFReader::Implementation::NewArray ( hid_t  dataset, const std::vector< hsize_t > &  fileExtent, hsize_t  numberOfComponents, T *  data  )

protected

Reads a vtkDataArray of type T from the attributeType, dataset The array has type 'T' and 'numberOfComponents'.

We are reading fileExtent slab from the array. It returns the array or nullptr in case of an error. There are three cases for fileExtent: fileExtent.size() == 0 - in this case we expect a 1D array and we read the whole array. Used for field arrays. fileExtent.size()>>1 == ndims - in this case we read a scalar fileExtent.size()>>1 + 1 == ndims - in this case we read an array with the number of components > 1.

NewStringArray()

vtkStringArray * vtkHDFReader::Implementation::NewStringArray ( hid_t  dataset, hsize_t  size  )

protected

Reads a vtkDataArray of type T from the attributeType, dataset The array has type 'T' and 'numberOfComponents'.

We are reading fileExtent slab from the array. It returns the array or nullptr in case of an error. There are three cases for fileExtent: fileExtent.size() == 0 - in this case we expect a 1D array and we read the whole array. Used for field arrays. fileExtent.size()>>1 == ndims - in this case we read a scalar fileExtent.size()>>1 + 1 == ndims - in this case we read an array with the number of components > 1.

BuildTypeReaderMap()

void vtkHDFReader::Implementation::BuildTypeReaderMap ( )

protected

Builds a map between native types and GetArray routines for that type.

GetTypeDescription()

TypeDescription vtkHDFReader::Implementation::GetTypeDescription ( hid_t  type )

protected

Associates a struc of three integers with HDF type.

This can be used as key in a map.

---

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

• IO/HDF/vtkHDFReaderImplementation.h