dtype Class Reference

#include <numpy.h>

Inheritance diagram for dtype:

Public Member Functions
ssize_t	alignment () const
	Alignment of the data type. More...
char	byteorder () const
	Single character for byteorder. More...
char	char_ () const
	dtype (const buffer_info &info)
	dtype (const char *format)
	dtype (const pybind11::str &format)
	dtype (const std::string &format)
	dtype (int typenum)
	dtype (list names, list formats, list offsets, ssize_t itemsize)
std::uint64_t	flags () const
	Flags for the array descriptor. More...
bool	has_fields () const
	Returns true for structured data types. More...
ssize_t	itemsize () const
	Size of the data type in bytes. More...
char	kind () const
int	num () const
	type number of dtype. More...
Public Member Functions inherited from object
template<typename T >
T	cast () &&
template<typename T >
T	cast () const &
template<>
void	cast () const &
	object ()=default
	object (const object &o)
	Copy constructor; always increases the reference count. More...
	object (handle h, borrowed_t)
	object (handle h, stolen_t)
	object (object &&other) noexcept
	Move constructor; steals the object from other and preserves its reference count. More...
object &	operator= (const object &other)
object &	operator= (object &&other) noexcept
	PYBIND11_DEPRECATED ("Use reinterpret_borrow<object>() or reinterpret_steal<object>()") object(handle h
handle	release ()
	~object ()
	Destructor; automatically calls handle::dec_ref() More...
Public Member Functions inherited from handle
template<typename T >
T	cast () const
template<>
void	cast () const
bool	check () const
const handle &	dec_ref () const &
	handle ()=default
	The default constructor creates a handle with a nullptr-valued pointer. More...
template<typename T , detail::enable_if_t< detail::all_of< detail::none_of< std::is_base_of< handle, T >, detail::is_pyobj_ptr_or_nullptr_t< T >>, std::is_convertible< T, PyObject * >>::value, int > = 0>
	handle (T &obj)
	Enable implicit conversion through T::operator PyObject *(). More...
template<typename T , detail::enable_if_t< detail::is_pyobj_ptr_or_nullptr_t< T >::value, int > = 0>
	handle (T ptr)
const handle &	inc_ref () const &
	operator bool () const
	Return true when the handle wraps a valid Python object. More...
bool	operator!= (const handle &h) const
bool	operator== (const handle &h) const
PyObject *&	ptr ()
PyObject *	ptr () const
	Return the underlying PyObject * pointer. More...

Static Public Member Functions
static dtype	from_args (const object &args)
	This is essentially the same as calling numpy.dtype(args) in Python. More...
template<typename T >
static dtype	of ()
	Return dtype associated with a C++ type. More...
Static Public Member Functions inherited from handle
static std::size_t	inc_ref_counter ()

Private Member Functions
dtype	strip_padding (ssize_t itemsize)

Static Private Member Functions
static object &	_dtype_from_pep3118 ()

Additional Inherited Members
Public Attributes inherited from object
bool	is_borrowed: handle(h) { if (is_borrowed) { inc_ref()
Protected Attributes inherited from handle
PyObject *	m_ptr = nullptr

Detailed Description

Definition at line 636 of file numpy.h.

Constructor & Destructor Documentation

dtype() [1/6]

dtype::dtype ( const buffer_info &  info )

inlineexplicit

Definition at line 640 of file numpy.h.

dtype() [2/6]

dtype::dtype ( const pybind11::str &  format )

inlineexplicit

Definition at line 648 of file numpy.h.

dtype() [3/6]

dtype::dtype ( const std::string &  format )

inlineexplicit

Definition at line 650 of file numpy.h.

dtype() [4/6]

dtype::dtype ( const char *  format )

inlineexplicit

Definition at line 652 of file numpy.h.

dtype() [5/6]

dtype::dtype ( list  names, list  formats, list  offsets, ssize_t  itemsize  )

inline

Definition at line 654 of file numpy.h.

dtype() [6/6]

dtype::dtype ( int  typenum )

inlineexplicit

Return dtype for the given typenum (one of the NPY_TYPES). https://numpy.org/devdocs/reference/c-api/array.html#c.PyArray_DescrFromType

Definition at line 665 of file numpy.h.

Member Function Documentation

_dtype_from_pep3118()

static object& dtype::_dtype_from_pep3118 ( )

inlinestaticprivate

Definition at line 764 of file numpy.h.

alignment()

ssize_t dtype::alignment ( ) const

inline

Alignment of the data type.

Definition at line 743 of file numpy.h.

byteorder()

char dtype::byteorder ( ) const

inline

Single character for byteorder.

Definition at line 737 of file numpy.h.

char_()

char dtype::char_ ( ) const

inline

Single-character for dtype's type. For example, float is 'f', double 'd', int 'i', and long 'l'.

Definition at line 721 of file numpy.h.

flags()

std::uint64_t dtype::flags ( ) const

inline

Flags for the array descriptor.

Definition at line 755 of file numpy.h.

from_args()

static dtype dtype::from_args ( const object &  args )

inlinestatic

This is essentially the same as calling numpy.dtype(args) in Python.

Definition at line 673 of file numpy.h.

has_fields()

bool dtype::has_fields ( ) const

inline

Returns true for structured data types.

Definition at line 703 of file numpy.h.

itemsize()

ssize_t dtype::itemsize ( ) const

inline

Size of the data type in bytes.

Definition at line 691 of file numpy.h.

kind()

char dtype::kind ( ) const

inline

Single-character code for dtype's kind. For example, floating point types are 'f' and integral types are 'i'.

Definition at line 717 of file numpy.h.

num()

int dtype::num ( ) const

inline

type number of dtype.

Definition at line 729 of file numpy.h.

of()

template<typename T >

static dtype dtype::of ( )

inlinestatic

Return dtype associated with a C++ type.

Definition at line 683 of file numpy.h.

strip_padding()

dtype dtype::strip_padding ( ssize_t  itemsize )

inlineprivate

Definition at line 774 of file numpy.h.

---

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

• numpy.h