nanovdb::io Namespace Reference

Classes
struct	FileHeader
	Data encoded at the head of each segment of a file or stream. More...
struct	FileMetaData

Enumerations
enum class	Codec : uint16_t {   NONE = 0 , ZIP = 1 , BLOSC = 2 , End = 3 ,   StrLen = 6 + End }
	Define compression codecs. More...

Functions
__hostdev__ const char *	toStr (char *dst, Codec codec)
__hostdev__ Codec	toCodec (const char *str)
template<typename StreamT>
void	writeUncompressedGrid (StreamT &os, const GridData *gridData, bool raw=false)
	This is a standalone alternative to io::writeGrid(...,Codec::NONE) defined in util/IO.h Unlike the latter this function has no dependencies at all, not even NanoVDB.h, so it also works if client code only includes PNanoVDB.h!
template<typename StreamT, typename ValueT>
void	writeUncompressedGrid (StreamT &os, const GridData *gridData, const ValueT *blindData, GridBlindDataSemantic semantic=GridBlindDataSemantic::Unknown, bool raw=false)
	Write an IndexGrid to a stream and append blind data.
template<typename GridHandleT, template< typename... > class VecT>
void	writeUncompressedGrids (const char *fileName, const VecT< GridHandleT > &handles, bool raw=false)
	write multiple NanoVDB grids to a single file, without compression.
template<typename GridHandleT, typename StreamT, template< typename... > class VecT>
VecT< GridHandleT >	readUncompressedGrids (StreamT &is, const typename GridHandleT::BufferType &pool=typename GridHandleT::BufferType())
	read all uncompressed grids from a stream and return their handles.
template<typename GridHandleT, template< typename... > class VecT>
VecT< GridHandleT >	readUncompressedGrids (const char *fileName, const typename GridHandleT::BufferType &buffer=typename GridHandleT::BufferType())
	Read a multiple un-compressed NanoVDB grids from a file and return them as a vector.

Enumeration Type Documentation

Codec

enum class Codec : uint16_t

strong

Define compression codecs.

Note

NONE is the default, ZIP is slow but compact and BLOSC offers a great balance.

Exceptions

NanoVDB

optionally supports ZIP and BLOSC compression and will throw an exception if its support is required but missing.

Enumerator
NONE
ZIP
BLOSC
End
StrLen

Function Documentation

readUncompressedGrids() [1/2]

template<typename GridHandleT, template< typename... > class VecT>

VecT< GridHandleT > readUncompressedGrids	(	const char *	fileName,
		const typename GridHandleT::BufferType &	buffer = typename GridHandleT::BufferType() )

Read a multiple un-compressed NanoVDB grids from a file and return them as a vector.

readUncompressedGrids() [2/2]

template<typename GridHandleT, typename StreamT, template< typename... > class VecT>

VecT< GridHandleT > readUncompressedGrids	(	StreamT &	is,
		const typename GridHandleT::BufferType &	pool = typename GridHandleT::BufferType() )

read all uncompressed grids from a stream and return their handles.

Exceptions

std::invalid_argument

if stream does not contain a single uncompressed valid NanoVDB grid

StreamT class must support: "bool read(char*, size_t)" and "void skip(uint32_t)"

toCodec()

__hostdev__ Codec toCodec ( const char * str )

inline

toStr()

__hostdev__ const char * toStr ( char * dst, Codec codec )

inline

writeUncompressedGrid() [1/2]

template<typename StreamT>

void writeUncompressedGrid	(	StreamT &	os,
		const GridData *	gridData,
		bool	raw = false )

This is a standalone alternative to io::writeGrid(...,Codec::NONE) defined in util/IO.h Unlike the latter this function has no dependencies at all, not even NanoVDB.h, so it also works if client code only includes PNanoVDB.h!

Writes a raw NanoVDB buffer, possibly with multiple grids, to a stream WITHOUT compression. It follows all the conventions in util/IO.h so the stream can be read by all existing client code of NanoVDB.

Note

This method will always write uncompressed grids to the stream, i.e. Blosc or ZIP compression is never applied! This is a fundamental limitation and feature of this standalone function.

Exceptions

std::invalid_argument

if buffer does not point to a valid NanoVDB grid.

Warning

This is pretty ugly code that involves lots of pointer and bit manipulations - not for the faint of heart :)

writeUncompressedGrid() [2/2]

template<typename StreamT, typename ValueT>

void writeUncompressedGrid	(	StreamT &	os,
		const GridData *	gridData,
		const ValueT *	blindData,
		GridBlindDataSemantic	semantic = GridBlindDataSemantic::Unknown,
		bool	raw = false )

Write an IndexGrid to a stream and append blind data.

Template Parameters

StreamT	Type of stream to write the IndexGrid and blind data to
ValueT	Type of the blind data

Parameters

os	Output stream to write to
gridData	GridData containing an IndexGrid WITHOUT existing blind data
blindData	Raw point to array of blind data
semantic	GridBlindDataSemantic of the blind data
raw	If true the IndexGrid and blind data are streamed raw, i.e. without a file header.

writeUncompressedGrids()

template<typename GridHandleT, template< typename... > class VecT>

void writeUncompressedGrids	(	const char *	fileName,
		const VecT< GridHandleT > &	handles,
		bool	raw = false )

write multiple NanoVDB grids to a single file, without compression.

Note

To write all grids in a single GridHandle simply use handle.write("fieNane")