src/multiarray/ucsnarrow.c File Reference

#include <Python.h>
#include <locale.h>
#include <stdio.h>
#include "numpy/arrayobject.h"
#include "numpy/npy_math.h"
#include "npy_config.h"
#include "npy_pycompat.h"
#include "ctors.h"

Defines
#define	NPY_NO_DEPRECATED_API   NPY_API_VERSION
#define	PY_SSIZE_T_CLEAN
#define	_MULTIARRAYMODULE
Functions
NPY_NO_EXPORT int	PyUCS2Buffer_FromUCS4 (Py_UNICODE *ucs2, npy_ucs4 *ucs4, int ucs4length)
NPY_NO_EXPORT int	PyUCS2Buffer_AsUCS4 (Py_UNICODE *ucs2, npy_ucs4 *ucs4, int ucs2len, int ucs4len)
NPY_NO_EXPORT PyUnicodeObject *	PyUnicode_FromUCS4 (char *src, Py_ssize_t size, int swap, int align)

---

Define Documentation

#define _MULTIARRAYMODULE

#define NPY_NO_DEPRECATED_API   NPY_API_VERSION

#define PY_SSIZE_T_CLEAN

---

Function Documentation

NPY_NO_EXPORT int PyUCS2Buffer_AsUCS4	(	Py_UNICODE *	ucs2,
		npy_ucs4 *	ucs4,
		int	ucs2len,
		int	ucs4len
	)

This converts a UCS2 buffer of the given length to UCS4 buffer. It converts up to ucs4len characters of UCS2

It returns the number of characters converted which can be less than ucs2len if there are surrogate pairs in ucs2.

The return value is the actual size of the used part of the ucs4 buffer.

surrogate pair

<

-0xdc00 + 0x10000

NPY_NO_EXPORT int PyUCS2Buffer_FromUCS4	(	Py_UNICODE *	ucs2,
		npy_ucs4 *	ucs4,
		int	ucs4length
	)

Functions only needed on narrow builds of Python for converting back and forth between the NumPy Unicode data-type (always 4-bytes) and the Python Unicode scalar (2-bytes on a narrow build).

The ucs2 buffer must be large enough to hold 2*ucs4length characters due to the use of surrogate pairs.

The return value is the number of ucs2 bytes used-up which is ucs4length + number of surrogate pairs found.

Values above 0xffff are converted to surrogate pairs.

NPY_NO_EXPORT PyUnicodeObject* PyUnicode_FromUCS4	(	char *	src,
		Py_ssize_t	size,
		int	swap,
		int	align
	)

Returns a PyUnicodeObject initialized from a buffer containing UCS4 unicode.

Parameters

<blockquote>

src: char *

Pointer to buffer containing UCS4 unicode.

size: Py_ssize_t

Size of buffer in bytes.

swap: int

If true, the data will be swapped.

align: int

If true, the data will be aligned.

</blockquote>

Returns

new_reference: PyUnicodeObject

swap and align if needed

trim trailing zeros

produce PyUnicode object