Unicode Objects and Codecs

Unicode Objects

Since the implementation of PEP 393 in Python 3.3, Unicode objects internally use a variety of representations, in order to allow handling the complete range of Unicode characters while staying memory efficient. There are special cases for strings where all code points are below 128, 256, or 65536; otherwise, code points must be below 1114112 (which is the full Unicode range).

Py_UNICODE* and UTF-8 representations are created on demand and cached in the Unicode object. The Py_UNICODE* representation is deprecated and inefficient; it should be avoided in performance- or memory-sensitive situations.

Due to the transition between the old APIs and the new APIs, unicode objects can internally be in two states depending on how they were created:

• “canonical” unicode objects are all objects created by a non-deprecated unicode API. They use the most efficient representation allowed by the implementation.

• “legacy” unicode objects have been created through one of the deprecated APIs (typically PyUnicode_FromUnicode()) and only bear the Py_UNICODE* representation; you will have to call PyUnicode_READY() on them before calling any other API.

Unicode Type

These are the basic Unicode object types used for the Unicode implementation in Python:

Py_UCS4

Py_UCS2

Py_UCS1

These types are typedefs for unsigned integer types wide enough to contain characters of 32 bits, 16 bits and 8 bits, respectively. When dealing with single Unicode characters, use Py_UCS4.

New in version 3.3.

Py_UNICODE

This is a typedef of wchar_t, which is a 16-bit type or 32-bit type depending on the platform.

Changed in version 3.3: In previous versions, this was a 16-bit type or a 32-bit type depending on whether you selected a “narrow” or “wide” Unicode version of Python at build time.

PyASCIIObject

PyCompactUnicodeObject

PyUnicodeObject

These subtypes of PyObject represent a Python Unicode object. In almost all cases, they shouldn’t be used directly, since all API functions that deal with Unicode objects take and return PyObject pointers.

New in version 3.3.

PyTypeObject PyUnicode_Type

This instance of PyTypeObject represents the Python Unicode type. It is exposed to Python code as str.

The following APIs are really C macros and can be used to do fast checks and to access internal read-only data of Unicode objects:

int PyUnicode_Check(PyObject *o)

Return true if the object o is a Unicode object or an instance of a Unicode subtype.

int PyUnicode_CheckExact(PyObject *o)

Return true if the object o is a Unicode object, but not an instance of a subtype.

int PyUnicode_READY(PyObject *o)

Ensure the string object o is in the “canonical” representation. This is required before using any of the access macros described below.

Returns 0 on success and -1 with an exception set on failure, which in particular happens if memory allocation fails.

New in version 3.3.

Py_ssize_t PyUnicode_GET_LENGTH(PyObject *o)

Return the length of the Unicode string, in code points. o has to be a Unicode object in the “canonical” representation (not checked).

New in version 3.3.

Py_UCS1* PyUnicode_1BYTE_DATA(PyObject *o)

Py_UCS2* PyUnicode_2BYTE_DATA(PyObject *o)

Py_UCS4* PyUnicode_4BYTE_DATA(PyObject *o)

Return a pointer to the canonical representation cast to UCS1, UCS2 or UCS4 integer types for direct character access. No checks are performed if the canonical representation has the correct character size; use PyUnicode_KIND() to select the right macro. Make sure PyUnicode_READY() has been called before accessing this.

New in version 3.3.

PyUnicode_WCHAR_KIND

PyUnicode_1BYTE_KIND

PyUnicode_2BYTE_KIND

PyUnicode_4BYTE_KIND

Return values of the PyUnicode_KIND() macro.

New in version 3.3.

int PyUnicode_KIND(PyObject *o)

Return one of the PyUnicode kind constants (see above) that indicate how many bytes per character this Unicode object uses to store its data. o has to be a Unicode object in the “canonical” representation (not checked).

New in version 3.3.

void* PyUnicode_DATA(PyObject *o)

Return a void pointer to the raw unicode buffer. o has to be a Unicode object in the “canonical” representation (not checked).

New in version 3.3.

void PyUnicode_WRITE(int kind, void *data, Py_ssize_t index, Py_UCS4 value)

Write into a canonical representation data (as obtained with PyUnicode_DATA()). This macro does not do any sanity checks and is intended for usage in loops. The caller should cache the kind value and data pointer as obtained from other macro calls. index is the index in the string (starts at 0) and value is the new code point value which should be written to that location.

New in version 3.3.

Py_UCS4 PyUnicode_READ(int kind, void *data, Py_ssize_t index)

Read a code point from a canonical representation data (as obtained with PyUnicode_DATA()). No checks or ready calls are performed.

New in version 3.3.

Py_UCS4 PyUnicode_READ_CHAR(PyObject *o, Py_ssize_t index)

Read a character from a Unicode object o, which must be in the “canonical” representation. This is less efficient than PyUnicode_READ() if you do multiple consecutive reads.

New in version 3.3.

PyUnicode_MAX_CHAR_VALUE(PyObject *o)

Return the maximum code point that is suitable for creating another string based on o, which must be in the “canonical” representation. This is always an approximation but more efficient than iterating over the string.

New in version 3.3.

int PyUnicode_ClearFreeList()

Clear the free list. Return the total number of freed items.

Py_ssize_t PyUnicode_GET_SIZE(PyObject *o)

Return the size of the deprecated Py_UNICODE representation, in code units (this includes surrogate pairs as 2 units). o has to be a Unicode object (not checked).

Deprecated since version 3.3, will be removed in version 4.0: Part of the old-style Unicode API, please migrate to using PyUnicode_GET_LENGTH().

Py_ssize_t PyUnicode_GET_DATA_SIZE(PyObject *o)

Return the size of the deprecated Py_UNICODE representation in bytes. o has to be a Unicode object (not checked).

Deprecated since version 3.3, will be removed in version 4.0: Part of the old-style Unicode API, please migrate to using PyUnicode_GET_LENGTH().

Py_UNICODE* PyUnicode_AS_UNICODE(PyObject *o)

const char* PyUnicode_AS_DATA(PyObject *o)

Return a pointer to a Py_UNICODE representation of the object. The returned buffer is always terminated with an extra null code point. It may also contain embedded null code points, which would cause the string to be truncated when used in most C functions. The AS_DATA form casts the pointer to const char *. The o argument has to be a Unicode object (not checked).

Changed in version 3.3: This macro is now inefficient – because in many cases the Py_UNICODE representation does not exist and needs to be created – and can fail (return NULL with an exception set). Try to port the code to use the new PyUnicode_nBYTE_DATA() macros or use PyUnicode_WRITE() or PyUnicode_READ().

Deprecated since version 3.3, will be removed in version 4.0: Part of the old-style Unicode API, please migrate to using the PyUnicode_nBYTE_DATA() family of macros.

Unicode Character Properties

Unicode provides many different character properties. The most often needed ones are available through these macros which are mapped to C functions depending on the Python configuration.

int Py_UNICODE_ISSPACE(Py_UNICODE ch)

Return 1 or 0 depending on whether ch is a whitespace character.

int Py_UNICODE_ISLOWER(Py_UNICODE ch)

Return 1 or 0 depending on whether ch is a lowercase character.

int Py_UNICODE_ISUPPER(Py_UNICODE ch)

Return 1 or 0 depending on whether ch is an uppercase character.

int Py_UNICODE_ISTITLE(Py_UNICODE ch)

Return 1 or 0 depending on whether ch is a titlecase character.

int Py_UNICODE_ISLINEBREAK(Py_UNICODE ch)

Return 1 or 0 depending on whether ch is a linebreak character.

int Py_UNICODE_ISDECIMAL(Py_UNICODE ch)

Return 1 or 0 depending on whether ch is a decimal character.

int Py_UNICODE_ISDIGIT(Py_UNICODE ch)

Return 1 or 0 depending on whether ch is a digit character.

int Py_UNICODE_ISNUMERIC(Py_UNICODE ch)

Return 1 or 0 depending on whether ch is a numeric character.

int Py_UNICODE_ISALPHA(Py_UNICODE ch)

Return 1 or 0 depending on whether ch is an alphabetic character.

int Py_UNICODE_ISALNUM(Py_UNICODE ch)

Return 1 or 0 depending on whether ch is an alphanumeric character.

int Py_UNICODE_ISPRINTABLE(Py_UNICODE ch)

Return 1 or 0 depending on whether ch is a printable character. Nonprintable characters are those characters defined in the Unicode character database as “Other” or “Separator”, excepting the ASCII space (0x20) which is considered printable. (Note that printable characters in this context are those which should not be escaped when repr() is invoked on a string. It has no bearing on the handling of strings written to sys.stdout or sys.stderr.)

These APIs can be used for fast direct character conversions:

Py_UNICODE Py_UNICODE_TOLOWER(Py_UNICODE ch)

Return the character ch converted to lower case.

Deprecated since version 3.3: This function uses simple case mappings.

Py_UNICODE Py_UNICODE_TOUPPER(Py_UNICODE ch)

Return the character ch converted to upper case.

Deprecated since version 3.3: This function uses simple case mappings.

Py_UNICODE Py_UNICODE_TOTITLE(Py_UNICODE ch)

Return the character ch converted to title case.

Deprecated since version 3.3: This function uses simple case mappings.

int Py_UNICODE_TODECIMAL(Py_UNICODE ch)

Return the character ch converted to a decimal positive integer. Return -1 if this is not possible. This macro does not raise exceptions.

int Py_UNICODE_TODIGIT(Py_UNICODE ch)

Return the character ch converted to a single digit integer. Return -1 if this is not possible. This macro does not raise exceptions.

double Py_UNICODE_TONUMERIC(Py_UNICODE ch)

Return the character ch converted to a double. Return -1.0 if this is not possible. This macro does not raise exceptions.

These APIs can be used to work with surrogates:

Py_UNICODE_IS_SURROGATE(ch)

Check if ch is a surrogate (0xD800 <= ch <= 0xDFFF).

Py_UNICODE_IS_HIGH_SURROGATE(ch)

Check if ch is a high surrogate (0xD800 <= ch <= 0xDBFF).

Py_UNICODE_IS_LOW_SURROGATE(ch)

Check if ch is a low surrogate (0xDC00 <= ch <= 0xDFFF).

Py_UNICODE_JOIN_SURROGATES(high, low)

Join two surrogate characters and return a single Py_UCS4 value. high and low are respectively the leading and trailing surrogates in a surrogate pair.

Creating and accessing Unicode strings

To create Unicode objects and access their basic sequence properties, use these APIs:

PyObject* PyUnicode_New(Py_ssize_t size, Py_UCS4 maxchar)

Create a new Unicode object. maxchar should be the true maximum code point to be placed in the string. As an approximation, it can be rounded up to the nearest value in the sequence 127, 255, 65535, 1114111.

This is the recommended way to allocate a new Unicode object. Objects created using this function are not resizable.

New in version 3.3.

PyObject* PyUnicode_FromKindAndData(int kind, const void *buffer, Py_ssize_t size)

Create a new Unicode object with the given kind (possible values are PyUnicode_1BYTE_KIND etc., as returned by PyUnicode_KIND()). The buffer must point to an array of size units of 1, 2 or 4 bytes per character, as given by the kind.

New in version 3.3.

PyObject* PyUnicode_FromStringAndSize(const char *u, Py_ssize_t size)

Create a Unicode object from the char buffer u. The bytes will be interpreted as being UTF-8 encoded. The buffer is copied into the new object. If the buffer is not NULL, the return value might be a shared object, i.e. modification of the data is not allowed.

If u is NULL, this function behaves like PyUnicode_FromUnicode() with the buffer set to NULL. This usage is deprecated in favor of PyUnicode_New().

PyObject *PyUnicode_FromString(const char *u)

Create a Unicode object from a UTF-8 encoded null-terminated char buffer u.

PyObject* PyUnicode_FromFormat(const char *format, ...)

Take a C printf()-style format string and a variable number of arguments, calculate the size of the resulting Python unicode string and return a string with the values formatted into it. The variable arguments must be C types and must correspond exactly to the format characters in the format ASCII-encoded string. The following format characters are allowed:

Format Characters	Type	Comment
%%	n/a	The literal % character.
%c	int	A single character, represented as a C int.
%d	int	Exactly equivalent to printf("%d").
%u	unsigned int	Exactly equivalent to printf("%u").
%ld	long	Exactly equivalent to printf("%ld").
%li	long	Exactly equivalent to printf("%li").
%lu	unsigned long	Exactly equivalent to printf("%lu").
%lld	long long	Exactly equivalent to printf("%lld").
%lli	long long	Exactly equivalent to printf("%lli").
%llu	unsigned long long	Exactly equivalent to printf("%llu").
%zd	Py_ssize_t	Exactly equivalent to printf("%zd").
%zi	Py_ssize_t	Exactly equivalent to printf("%zi").
%zu	size_t	Exactly equivalent to printf("%zu").
%i	int	Exactly equivalent to printf("%i").
%x	int	Exactly equivalent to printf("%x").
%s	char*	A null-terminated C character array.
%p	void*	The hex representation of a C pointer. Mostly equivalent to printf("%p") except that it is guaranteed to start with the literal 0x regardless of what the platform’s printf yields.
%A	PyObject*	The result of calling ascii().
%U	PyObject*	A unicode object.
%V	PyObject*, char *	A unicode object (which may be NULL) and a null-terminated C character array as a second parameter (which will be used, if the first parameter is NULL).
%S	PyObject*	The result of calling PyObject_Str().
%R	PyObject*	The result of calling PyObject_Repr().

An unrecognized format character causes all the rest of the format string to be copied as-is to the result string, and any extra arguments discarded.

Note

The width formatter unit is number of characters rather than bytes. The precision formatter unit is number of bytes for "%s" and "%V" (if the PyObject* argument is NULL), and a number of characters for "%A", "%U", "%S", "%R" and "%V" (if the PyObject* argument is not NULL).

Changed in version 3.2: Support for "%lld" and "%llu" added.

Changed in version 3.3: Support for "%li", "%lli" and "%zi" added.

Changed in version 3.4: Support width and precision formatter for "%s", "%A", "%U", "%V", "%S", "%R" added.

PyObject* PyUnicode_FromFormatV(const char *format, va_list vargs)

Identical to PyUnicode_FromFormat() except that it takes exactly two arguments.

PyObject* PyUnicode_FromEncodedObject(PyObject *obj, const char *encoding, const char *errors)

Return value: New reference.

Decode an encoded object obj to a Unicode object.

bytes, bytearray and other bytes-like objects are decoded according to the given encoding and using the error handling defined by errors. Both can be NULL to have the interface use the default values (see Built-in Codecs for details).

All other objects, including Unicode objects, cause a TypeError to be set.

The API returns NULL if there was an error. The caller is responsible for decref’ing the returned objects.

Py_ssize_t PyUnicode_GetLength(PyObject *unicode)

Return the length of the Unicode object, in code points.

New in version 3.3.

Py_ssize_t PyUnicode_CopyCharacters(PyObject *to, Py_ssize_t to_start, PyObject *from, Py_ssize_t from_start, Py_ssize_t how_many)

Copy characters from one Unicode object into another. This function performs character conversion when necessary and falls back to memcpy() if possible. Returns -1 and sets an exception on error, otherwise returns the number of copied characters.

New in version 3.3.

Py_ssize_t PyUnicode_Fill(PyObject *unicode, Py_ssize_t start, Py_ssize_t length, Py_UCS4 fill_char)

Fill a string with a character: write fill_char into unicode[start:start+length].

Fail if fill_char is bigger than the string maximum character, or if the string has more than 1 reference.

Return the number of written character, or return -1 and raise an exception on error.

New in version 3.3.

int PyUnicode_WriteChar(PyObject *unicode, Py_ssize_t index, Py_UCS4 character)

Write a character to a string. The string must have been created through PyUnicode_New(). Since Unicode strings are supposed to be immutable, the string must not be shared, or have been hashed yet.

This function checks that unicode is a Unicode object, that the index is not out of bounds, and that the object can be modified safely (i.e. that it its reference count is one).

New in version 3.3.

Py_UCS4 PyUnicode_ReadChar(PyObject *unicode, Py_ssize_t index)

Read a character from a string. This function checks that unicode is a Unicode object and the index is not out of bounds, in contrast to the macro version PyUnicode_READ_CHAR().

New in version 3.3.

PyObject* PyUnicode_Substring(PyObject *str, Py_ssize_t start, Py_ssize_t end)

Return a substring of str, from character index start (included) to character index end (excluded). Negative indices are not supported.

New in version 3.3.

Py_UCS4* PyUnicode_AsUCS4(PyObject *u, Py_UCS4 *buffer, Py_ssize_t buflen, int copy_null)

Copy the string u into a UCS4 buffer, including a null character, if copy_null is set. Returns NULL and sets an exception on error (in particular, a SystemError if buflen is smaller than the length of u). buffer is returned on success.

New in version 3.3.

Py_UCS4* PyUnicode_AsUCS4Copy(PyObject *u)

Copy the string u into a new UCS4 buffer that is allocated using PyMem_Malloc(). If this fails, NULL is returned with a MemoryError set. The returned buffer always has an extra null code point appended.

New in version 3.3.

Deprecated Py_UNICODE APIs

Deprecated since version 3.3, will be removed in version 4.0.

These API functions are deprecated with the implementation of PEP 393. Extension modules can continue using them, as they will not be removed in Python 3.x, but need to be aware that their use can now cause performance and memory hits.

PyObject* PyUnicode_FromUnicode(const Py_UNICODE *u, Py_ssize_t size)

Return value: New reference.

Create a Unicode object from the Py_UNICODE buffer u of the given size. u may be NULL which causes the contents to be undefined. It is the user’s responsibility to fill in the needed data. The buffer is copied into the new object.

If the buffer is not NULL, the return value might be a shared object. Therefore, modification of the resulting Unicode object is only allowed when u is NULL.

If the buffer is NULL, PyUnicode_READY() must be called once the string content has been filled before using any of the access macros such as PyUnicode_KIND().

Please migrate to using PyUnicode_FromKindAndData(), PyUnicode_FromWideChar() or PyUnicode_New().

Py_UNICODE* PyUnicode_AsUnicode(PyObject *unicode)

Return a read-only pointer to the Unicode object’s internal Py_UNICODE buffer, or NULL on error. This will create the Py_UNICODE* representation of the object if it is not yet available. The buffer is always terminated with an extra null code point. Note that the resulting Py_UNICODE string may also contain embedded null code points, which would cause the string to be truncated when used in most C functions.

Please migrate to using PyUnicode_AsUCS4(), PyUnicode_AsWideChar(), PyUnicode_ReadChar() or similar new APIs.

PyObject* PyUnicode_TransformDecimalToASCII(Py_UNICODE *s, Py_ssize_t size)

Create a Unicode object by replacing all decimal digits in Py_UNICODE buffer of the given size by ASCII digits 0–9 according to their decimal value. Return NULL if an exception occurs.

Py_UNICODE* PyUnicode_AsUnicodeAndSize(PyObject *unicode, Py_ssize_t *size)

Like PyUnicode_AsUnicode(), but also saves the Py_UNICODE() array length (excluding the extra null terminator) in size. Note that the resulting Py_UNICODE* string may contain embedded null code points, which would cause the string to be truncated when used in most C functions.

New in version 3.3.

Py_UNICODE* PyUnicode_AsUnicodeCopy(PyObject *unicode)

Create a copy of a Unicode string ending with a null code point. Return NULL and raise a MemoryError exception on memory allocation failure, otherwise return a new allocated buffer (use PyMem_Free() to free the buffer). Note that the resulting Py_UNICODE* string may contain embedded null code points, which would cause the string to be truncated when used in most C functions.

New in version 3.2.

Please migrate to using PyUnicode_AsUCS4Copy() or similar new APIs.

Py_ssize_t PyUnicode_GetSize(PyObject *unicode)

Return the size of the deprecated Py_UNICODE representation, in code units (this includes surrogate pairs as 2 units).

Please migrate to using PyUnicode_GetLength().

PyObject* PyUnicode_FromObject(PyObject *obj)

Return value: New reference.

Copy an instance of a Unicode subtype to a new true Unicode object if necessary. If obj is already a true Unicode object (not a subtype), return the reference with incremented refcount.

Objects other than Unicode or its subtypes will cause a TypeError.

Locale Encoding

The current locale encoding can be used to decode text from the operating system.

PyObject* PyUnicode_DecodeLocaleAndSize(const char *str, Py_ssize_t len, const char *errors)

Decode a string from the current locale encoding. The supported error handlers are "strict" and "surrogateescape" (PEP 383). The decoder uses "strict" error handler if errors is NULL. str must end with a null character but cannot contain embedded null characters.

Use PyUnicode_DecodeFSDefaultAndSize() to decode a string from Py_FileSystemDefaultEncoding (the locale encoding read at Python startup).

See also

The Py_DecodeLocale() function.

New in version 3.3.

Changed in version 3.6.5: The function now also uses the current locale encoding for the surrogateescape error handler. Previously, Py_DecodeLocale() was used for the surrogateescape, and the current locale encoding was used for strict.

PyObject* PyUnicode_DecodeLocale(const char *str, const char *errors)

Similar to PyUnicode_DecodeLocaleAndSize(), but compute the string length using strlen().