This article explains the new features in Python 3.10, compared to 3.9.

For full details, see the :ref:`changelog <changelog>`.

• PEP 644, require OpenSSL 1.1.1 or newer

Using enclosing parentheses for continuation across multiple lines in context managers is now supported. This allows formatting a long collection of context managers in multiple lines in a similar way as it was previously possible with import statements. For instance, all these examples are now valid:

with (CtxManager() as example):
    ...

with (
    CtxManager1(),
    CtxManager2()
):
    ...

with (CtxManager1() as example,
      CtxManager2()):
    ...

with (CtxManager1(),
      CtxManager2() as example):
    ...

with (
    CtxManager1() as example1,
    CtxManager2() as example2
):
    ...

it is also possible to use a trailing comma at the end of the enclosed group:

with (
    CtxManager1() as example1,
    CtxManager2() as example2,
    CtxManager3() as example3,
):
    ...

This new syntax uses the non LL(1) capacities of the new parser. Check PEP 617 for more details.

(Contributed by Guido van Rossum, Pablo Galindo and Lysandros Nikolaou in :issue:`12782` and :issue:`40334`.)

When parsing code that contains unclosed parentheses or brackets the interpreter now includes the location of the unclosed bracket of parentheses instead of displaying SyntaxError: unexpected EOF while parsing or pointing to some incorrect location. For instance, consider the following code (notice the unclosed '{'):

expected = {9: 1, 18: 2, 19: 2, 27: 3, 28: 3, 29: 3, 36: 4, 37: 4,
            38: 4, 39: 4, 45: 5, 46: 5, 47: 5, 48: 5, 49: 5, 54: 6,
some_other_code = foo()

Previous versions of the interpreter reported confusing places as the location of the syntax error:

File "example.py", line 3
    some_other_code = foo()
                    ^
SyntaxError: invalid syntax

but in Python 3.10 a more informative error is emitted:

File "example.py", line 1
    expected = {9: 1, 18: 2, 19: 2, 27: 3, 28: 3, 29: 3, 36: 4, 37: 4,
               ^
SyntaxError: '{' was never closed

In a similar way, errors involving unclosed string literals (single and triple quoted) now point to the start of the string instead of reporting EOF/EOL.

These improvements are inspired by previous work in the PyPy interpreter.

(Contributed by Pablo Galindo in :issue:`42864` and Batuhan Taskaya in :issue:`40176`.)

:exc:`SyntaxError` exceptions raised by the interpreter will now highlight the full error range of the expression that constitutes the syntax error itself, instead of just where the problem is detected. In this way, instead of displaying (before Python 3.10):

>>> foo(x, z for z in range(10), t, w)
  File "<stdin>", line 1
    foo(x, z for z in range(10), t, w)
           ^
SyntaxError: Generator expression must be parenthesized

now Python 3.10 will display the exception as:

>>> foo(x, z for z in range(10), t, w)
  File "<stdin>", line 1
    foo(x, z for z in range(10), t, w)
           ^^^^^^^^^^^^^^^^^^^^
SyntaxError: Generator expression must be parenthesized

This improvement was contributed by Pablo Galindo in :issue:`43914`.

A considerable amount of new specialized messages for :exc:`SyntaxError` exceptions have been incorporated. Some of the most notable ones:

• Missing : before blocks:

  >>> if rocket.position > event_horizon
    File "<stdin>", line 1
      if rocket.position > event_horizon
                                        ^
  SyntaxError: expected ':'

  (Contributed by Pablo Galindo in :issue:`42997`)

• Unparenthesised tuples in comprehensions targets:

  >>> {x,y for x,y in zip('abcd', '1234')}
    File "<stdin>", line 1
      {x,y for x,y in zip('abcd', '1234')}
       ^
  SyntaxError: did you forget parentheses around the comprehension target?

  (Contributed by Pablo Galindo in :issue:`43017`)

• Missing commas in collection literals and between expressions:

  >>> items = {
  ... x: 1,
  ... y: 2
  ... z: 3,
    File "<stdin>", line 3
      y: 2
         ^
  SyntaxError: invalid syntax. Perhaps you forgot a comma?

  (Contributed by Pablo Galindo in :issue:`43822`)

• Multiple Exception types without parentheses:

  >>> try:
  ...     build_dyson_sphere()
  ... except NotEnoughScienceError, NotEnoughResourcesError:
    File "<stdin>", line 3
      except NotEnoughScienceError, NotEnoughResourcesError:
             ^
  SyntaxError: multiple exception types must be parenthesized

  (Contributed by Pablo Galindo in :issue:`43149`)

• Missing : and values in dictionary literals:

  >>> values = {
  ... x: 1,
  ... y: 2,
  ... z:
  ... }
    File "<stdin>", line 4
      z:
       ^
  SyntaxError: expression expected after dictionary key and ':'
  
  >>> values = {x:1, y:2, z w:3}
    File "<stdin>", line 1
      values = {x:1, y:2, z w:3}
                          ^
  SyntaxError: ':' expected after dictionary key

  (Contributed by Pablo Galindo in :issue:`43823`)

• Usage of = instead of == in comparisons:

  >>> if rocket.position = event_horizon:
    File "<stdin>", line 1
      if rocket.position = event_horizon:
                         ^
  SyntaxError: cannot assign to attribute here. Maybe you meant '==' instead of '='?

  (Contributed by Pablo Galindo in :issue:`43797`)

• Usage of * in f-strings:

  >>> f"Black holes {*all_black_holes} and revelations"
    File "<stdin>", line 1
      (*all_black_holes)
       ^
  SyntaxError: f-string: cannot use starred expression here

  (Contributed by Pablo Galindo in :issue:`41064`)

Many :exc:`IndentationError` exceptions now have more context regarding what kind of block was expecting an indentation, including the location of the statement:

>>> def foo():
...    if lel:
...    x = 2
  File "<stdin>", line 3
    x = 2
    ^
IndentationError: expected an indented block after 'if' statement in line 2

When printing :exc:`AttributeError`, :c:func:`PyErr_Display` will offer suggestions of similar attribute names in the object that the exception was raised from:

>>> collections.namedtoplo
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
AttributeError: module 'collections' has no attribute 'namedtoplo'. Did you mean: namedtuple?

(Contributed by Pablo Galindo in :issue:`38530`.)

When printing :exc:`NameError` raised by the interpreter, :c:func:`PyErr_Display` will offer suggestions of similar variable names in the function that the exception was raised from:

>>> schwarzschild_black_hole = None
>>> schwarschild_black_hole
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
NameError: name 'schwarschild_black_hole' is not defined. Did you mean: schwarzschild_black_hole?

(Contributed by Pablo Galindo in :issue:`38530`.)

PEP 626 brings more precise and reliable line numbers for debugging, profiling and coverage tools. Tracing events, with the correct line number, are generated for all lines of code executed and only for lines of code that are executed.

The f_lineno attribute of frame objects will always contain the expected line number.

The co_lnotab attribute of code objects is deprecated and will be removed in 3.12. Code that needs to convert from offset to line number should use the new co_lines() method instead.

Structural pattern matching has been added in the form of a match statement and case statements of patterns with associated actions. Patterns consist of sequences, mappings, primitive data types as well as class instances. Pattern matching enables programs to extract information from complex data types, branch on the structure of data, and apply specific actions based on different forms of data.

The generic syntax of pattern matching is:

match subject:
    case <pattern_1>:
        <action_1>
    case <pattern_2>:
        <action_2>
    case <pattern_3>:
        <action_3>
    case _:
        <action_wildcard>

A match statement takes an expression and compares its value to successive patterns given as one or more case blocks. Specifically, pattern matching operates by:

1. using data with type and shape (the subject)
2. evaluating the subject in the match statement
3. comparing the subject with each pattern in a case statement from top to bottom until a match is confirmed.
4. executing the action associated with the pattern of the confirmed match
5. If an exact match is not confirmed, the last case, a wildcard _, if provided, will be used as the matching case. If an exact match is not confirmed and a wildcard case does not exist, the entire match block is a no-op.

Readers may be aware of pattern matching through the simple example of matching a subject (data object) to a literal (pattern) with the switch statement found in C, Java or JavaScript (and many other languages). Often the switch statement is used for comparison of an object/expression with case statements containing literals.

More powerful examples of pattern matching can be found in languages such as Scala and Elixir. With structural pattern matching, the approach is "declarative" and explicitly states the conditions (the patterns) for data to match.

While an "imperative" series of instructions using nested "if" statements could be used to accomplish something similar to structural pattern matching, it is less clear than the "declarative" approach. Instead the "declarative" approach states the conditions to meet for a match and is more readable through its explicit patterns. While structural pattern matching can be used in its simplest form comparing a variable to a literal in a case statement, its true value for Python lies in its handling of the subject's type and shape.

Let's look at this example as pattern matching in its simplest form: a value, the subject, being matched to several literals, the patterns. In the example below, status is the subject of the match statement. The patterns are each of the case statements, where literals represent request status codes. The associated action to the case is executed after a match:

def http_error(status):
    match status:
        case 400:
            return "Bad request"
        case 404:
            return "Not found"
        case 418:
            return "I'm a teapot"
        case _:
            return "Something's wrong with the Internet"

If the above function is passed a status of 418, "I'm a teapot" is returned. If the above function is passed a status of 500, the case statement with _ will match as a wildcard, and "Something's wrong with the Internet" is returned. Note the last block: the variable name, _, acts as a wildcard and insures the subject will always match. The use of _ is optional.

You can combine several literals in a single pattern using | ("or"):

case 401 | 403 | 404:
    return "Not allowed"

If we modify the above example by removing the last case block, the example becomes:

def http_error(status):
    match status:
        case 400:
            return "Bad request"
        case 404:
            return "Not found"
        case 418:
            return "I'm a teapot"

Without the use of _ in a case statement, a match may not exist. If no match exists, the behavior is a no-op. For example, if status of 500 is passed, a no-op occurs.

Patterns can look like unpacking assignments, and a pattern may be used to bind variables. In this example, a data point can be unpacked to its x-coordinate and y-coordinate:

# point is an (x, y) tuple
match point:
    case (0, 0):
        print("Origin")
    case (0, y):
        print(f"Y={y}")
    case (x, 0):
        print(f"X={x}")
    case (x, y):
        print(f"X={x}, Y={y}")
    case _:
        raise ValueError("Not a point")

The first pattern has two literals, (0, 0), and may be thought of as an extension of the literal pattern shown above. The next two patterns combine a literal and a variable, and the variable binds a value from the subject (point). The fourth pattern captures two values, which makes it conceptually similar to the unpacking assignment (x, y) = point.

If you are using classes to structure your data, you can use as a pattern the class name followed by an argument list resembling a constructor. This pattern has the ability to capture class attributes into variables:

class Point:
    x: int
    y: int

def location(point):
    match point:
        case Point(x=0, y=0):
            print("Origin is the point's location.")
        case Point(x=0, y=y):
            print(f"Y={y} and the point is on the y-axis.")
        case Point(x=x, y=0):
            print(f"X={x} and the point is on the x-axis.")
        case Point():
            print("The point is located somewhere else on the plane.")
        case _:
            print("Not a point")

You can use positional parameters with some builtin classes that provide an ordering for their attributes (e.g. dataclasses). You can also define a specific position for attributes in patterns by setting the __match_args__ special attribute in your classes. If it's set to ("x", "y"), the following patterns are all equivalent (and all bind the y attribute to the var variable):

Point(1, var)
Point(1, y=var)
Point(x=1, y=var)
Point(y=var, x=1)

Patterns can be arbitrarily nested. For example, if our data is a short list of points, it could be matched like this:

match points:
    case []:
        print("No points in the list.")
    case [Point(0, 0)]:
        print("The origin is the only point in the list.")
    case [Point(x, y)]:
        print(f"A single point {x}, {y} is in the list.")
    case [Point(0, y1), Point(0, y2)]:
        print(f"Two points on the Y axis at {y1}, {y2} are in the list.")
    case _:
        print("Something else is found in the list.")

To this point, the examples have used _ alone in the last case statement. A wildcard can be used in more complex patterns, such as ('error', code, _). For example:

match test_variable:
    case ('warning', code, 40):
        print("A warning has been received.")
    case ('error', code, _):
        print(f"An error {code} occurred.")

In the above case, test_variable will match for ('error', code, 100) and ('error', code, 800).

We can add an if clause to a pattern, known as a "guard". If the guard is false, match goes on to try the next case block. Note that value capture happens before the guard is evaluated:

match point:
    case Point(x, y) if x == y:
        print(f"The point is located on the diagonal Y=X at {x}.")
    case Point(x, y):
        print(f"Point is not on the diagonal.")

Several other key features:

• Like unpacking assignments, tuple and list patterns have exactly the same meaning and actually match arbitrary sequences. Technically, the subject must be a sequence. Therefore, an important exception is that patterns don't match iterators. Also, to prevent a common mistake, sequence patterns don't match strings.

• Sequence patterns support wildcards: [x, y, *rest] and (x, y, *rest) work similar to wildcards in unpacking assignments. The name after * may also be _, so (x, y, *_) matches a sequence of at least two items without binding the remaining items.

• Mapping patterns: {"bandwidth": b, "latency": l} captures the "bandwidth" and "latency" values from a dict. Unlike sequence patterns, extra keys are ignored. A wildcard **rest is also supported. (But **_ would be redundant, so is not allowed.)

• Subpatterns may be captured using the as keyword:

  case (Point(x1, y1), Point(x2, y2) as p2): ...
  

  This binds x1, y1, x2, y2 like you would expect without the as clause, and p2 to the entire second item of the subject.

• Most literals are compared by equality. However, the singletons True, False and None are compared by identity.

• Named constants may be used in patterns. These named constants must be dotted names to prevent the constant from being interpreted as a capture variable:

  from enum import Enum
  class Color(Enum):
      RED = 0
      GREEN = 1
      BLUE = 2
  
  match color:
      case Color.RED:
          print("I see red!")
      case Color.GREEN:
          print("Grass is green")
      case Color.BLUE:
          print("I'm feeling the blues :(")
  

For the full specification see PEP 634. Motivation and rationale are in PEP 635, and a longer tutorial is in PEP 636.

The default encoding of :class:`TextIOWrapper` and :func:`open` is platform and locale dependent. Since UTF-8 is used on most Unix platforms, omitting encoding option when opening UTF-8 files (e.g. JSON, YAML, TOML, Markdown) is a very common bug. For example:

# BUG: "rb" mode or encoding="utf-8" should be used.
with open("data.json") as f:
    data = json.load(f)

To find this type of bug, an optional EncodingWarning is added. It is emitted when :data:`sys.flags.warn_default_encoding <sys.flags>` is true and locale-specific default encoding is used.

-X warn_default_encoding option and :envvar:`PYTHONWARNDEFAULTENCODING` are added to enable the warning.

See :ref:`io-text-encoding` for more information.

This section covers major changes affecting PEP 484 type hints and the :mod:`typing` module.

A new type union operator was introduced which enables the syntax X | Y. This provides a cleaner way of expressing 'either type X or type Y' instead of using :data:`typing.Union`, especially in type hints.

In previous versions of Python, to apply a type hint for functions accepting arguments of multiple types, :data:`typing.Union` was used:

def square(number: Union[int, float]) -> Union[int, float]:
    return number ** 2

Type hints can now be written in a more succinct manner:

def square(number: int | float) -> int | float:
    return number ** 2

This new syntax is also accepted as the second argument to :func:`isinstance` and :func:`issubclass`:

>>> isinstance(1, int | str)
True

See :ref:`types-union` and PEP 604 for more details.

(Contributed by Maggie Moss and Philippe Prados in :issue:`41428`.)

Two new options to improve the information provided to static type checkers for PEP 484's Callable have been added to the :mod:`typing` module.

The first is the parameter specification variable. They are used to forward the parameter types of one callable to another callable -- a pattern commonly found in higher order functions and decorators. Examples of usage can be found in :class:`typing.ParamSpec`. Previously, there was no easy way to type annotate dependency of parameter types in such a precise manner.

The second option is the new Concatenate operator. It's used in conjunction with parameter specification variables to type annotate a higher order callable which adds or removes parameters of another callable. Examples of usage can be found in :class:`typing.Concatenate`.

See :class:`typing.Callable`, :class:`typing.ParamSpec`, :class:`typing.Concatenate`, :class:`typing.ParamSpecArgs`, :class:`typing.ParamSpecKwargs`, and PEP 612 for more details.

(Contributed by Ken Jin in :issue:`41559`, with minor enhancements by Jelle Zijlstra in :issue:`43783`. PEP written by Mark Mendoza.)

PEP 484 introduced the concept of type aliases, only requiring them to be top-level unannotated assignments. This simplicity sometimes made it difficult for type checkers to distinguish between type aliases and ordinary assignments, especially when forward references or invalid types were involved. Compare:

StrCache = 'Cache[str]'  # a type alias
LOG_PREFIX = 'LOG[DEBUG]'  # a module constant

Now the :mod:`typing` module has a special value :data:`TypeAlias` which lets you declare type aliases more explicitly:

StrCache: TypeAlias = 'Cache[str]'  # a type alias
LOG_PREFIX = 'LOG[DEBUG]'  # a module constant

See PEP 613 for more details.

(Contributed by Mikhail Golubev in :issue:`41923`.)

:data:`TypeGuard` has been added to the :mod:`typing` module to annotate type guard functions and improve information provided to static type checkers during type narrowing. For more information, please see :data:`TypeGuard`'s documentation, and PEP 647.

(Contributed by Ken Jin and Guido van Rossum in :issue:`43766`. PEP written by Eric Traut.)

• The :class:`int` type has a new method :meth:`int.bit_count`, returning the number of ones in the binary expansion of a given integer, also known as the population count. (Contributed by Niklas Fiekas in :issue:`29882`.)
• The views returned by :meth:`dict.keys`, :meth:`dict.values` and :meth:`dict.items` now all have a mapping attribute that gives a :class:`types.MappingProxyType` object wrapping the original dictionary. (Contributed by Dennis Sweeney in :issue:`40890`.)
• PEP 618: The :func:`zip` function now has an optional strict flag, used to require that all the iterables have an equal length.
• Builtin and extension functions that take integer arguments no longer accept :class:`~decimal.Decimal`s, :class:`~fractions.Fraction`s and other objects that can be converted to integers only with a loss (e.g. that have the :meth:`~object.__int__` method but do not have the :meth:`~object.__index__` method). (Contributed by Serhiy Storchaka in :issue:`37999`.)
• If :func:`object.__ipow__` returns :const:`NotImplemented`, the operator will correctly fall back to :func:`object.__pow__` and :func:`object.__rpow__` as expected. (Contributed by Alex Shkop in :issue:`38302`.)
• Assignment expressions can now be used unparenthesized within set literals and set comprehensions, as well as in sequence indexes (but not slices).
• Functions have a new __builtins__ attribute which is used to look for builtin symbols when a function is executed, instead of looking into __globals__['__builtins__']. The attribute is initialized from __globals__["__builtins__"] if it exists, else from the current builtins. (Contributed by Mark Shannon in :issue:`42990`.)
• Two new builtin functions -- :func:`aiter` and :func:`anext` have been added to provide asynchronous counterparts to :func:`iter` and :func:`next`, respectively. (Contributed by Joshua Bronson, Daniel Pope, and Justin Wang in :issue:`31861`.)
• Static methods (:func:`@staticmethod <staticmethod>`) and class methods (:func:`@classmethod <classmethod>`) now inherit the method attributes (__module__, __name__, __qualname__, __doc__, __annotations__) and have a new __wrapped__ attribute. Moreover, static methods are now callable as regular functions. (Contributed by Victor Stinner in :issue:`43682`.)
• Annotations for complex targets (everything beside simple name targets defined by PEP 526) no longer cause any runtime effects with from __future__ import annotations. (Contributed by Batuhan Taskaya in :issue:`42737`.)
• Class and module objects now lazy-create empty annotations dicts on demand. The annotations dicts are stored in the object’s __dict__ for backwards compatibility. This improves the best practices for working with __annotations__; for more information, please see :ref:`annotations-howto`. (Contributed by Larry Hastings in :issue:`43901`.)
• Annotations consist of yield, yield from, await or named expressions are now forbidden under from __future__ import annotations due to their side effects. (Contributed by Batuhan Taskaya in :issue:`42725`.)
• Usage of unbound variables, super() and other expressions that might alter the processing of symbol table as annotations are now rendered effectless under from __future__ import annotations. (Contributed by Batuhan Taskaya in :issue:`42725`.)

• None yet.

Add missing :meth:`~asyncio.events.AbstractEventLoop.connect_accepted_socket` method. (Contributed by Alex Grönholm in :issue:`41332`.)

Misleading phrase "optional arguments" was replaced with "options" in argparse help. Some tests might require adaptation if they rely on exact output match. (Contributed by Raymond Hettinger in :issue:`9694`.)

The :meth:`~array.array.index` method of :class:`array.array` now has optional start and stop parameters. (Contributed by Anders Lorentsen and Zackery Spytz in :issue:`31956`.)

Add :func:`base64.b32hexencode` and :func:`base64.b32hexdecode` to support the Base32 Encoding with Extended Hex Alphabet.

Add :meth:`~bdb.Breakpoint.clearBreakpoints` to reset all set breakpoints. (Contributed by Irit Katriel in :issue:`24160`.)

Add a :func:`codecs.unregister` function to unregister a codec search function. (Contributed by Hai Shi in :issue:`41842`.)

The __args__ of the :ref:`parameterized generic <types-genericalias>` for :class:`collections.abc.Callable` are now consistent with :data:`typing.Callable`. :class:`collections.abc.Callable` generic now flattens type parameters, similar to what :data:`typing.Callable` currently does. This means that collections.abc.Callable[[int, str], str] will have __args__ of (int, str, str); previously this was ([int, str], str). To allow this change, :class:`types.GenericAlias` can now be subclassed, and a subclass will be returned when subscripting the :class:`collections.abc.Callable` type. Note that a :exc:`TypeError` may be raised for invalid forms of parameterizing :class:`collections.abc.Callable` which may have passed silently in Python 3.9. (Contributed by Ken Jin in :issue:`42195`.)

Add a :func:`contextlib.aclosing` context manager to safely close async generators and objects representing asynchronously released resources. (Contributed by Joongi Kim and John Belmonte in :issue:`41229`.)

Add asynchronous context manager support to :func:`contextlib.nullcontext`. (Contributed by Tom Gringauz in :issue:`41543`.)

Add :class:`AsyncContextDecorator`, for supporting usage of async context managers as decorators.

The extended color functions added in ncurses 6.1 will be used transparently by :func:`curses.color_content`, :func:`curses.init_color`, :func:`curses.init_pair`, and :func:`curses.pair_content`. A new function, :func:`curses.has_extended_color_support`, indicates whether extended color support is provided by the underlying ncurses library. (Contributed by Jeffrey Kintscher and Hans Petter Jansson in :issue:`36982`.)

The BUTTON5_* constants are now exposed in the :mod:`curses` module if they are provided by the underlying curses library. (Contributed by Zackery Spytz in :issue:`39273`.)

Add slots parameter in :func:`dataclasses.dataclass` decorator. (Contributed by Yurii Karabas in :issue:`42269`)

The entire distutils package is deprecated, to be removed in Python 3.12. Its functionality for specifying package builds has already been completely replaced by third-party packages setuptools and packaging, and most other commonly used APIs are available elsewhere in the standard library (such as :mod:`platform`, :mod:`shutil`, :mod:`subprocess` or :mod:`sysconfig`). There are no plans to migrate any other functionality from distutils, and applications that are using other functions should plan to make private copies of the code. Refer to PEP 632 for discussion.

The bdist_wininst command deprecated in Python 3.8 has been removed. The bdist_wheel command is now recommended to distribute binary packages on Windows. (Contributed by Victor Stinner in :issue:`42802`.)

When a module does not define __loader__, fall back to __spec__.loader. (Contributed by Brett Cannon in :issue:`42133`.)

:func:`encodings.normalize_encoding` now ignores non-ASCII characters. (Contributed by Hai Shi in :issue:`39337`.)

:class:`Enum` :func:`__repr__` now returns enum_name.member_name and :func:`__str__` now returns member_name. Stdlib enums available as module constants have a :func:`repr` of module_name.member_name. (Contributed by Ethan Furman in :issue:`40066`.)

Add :class:`enum.StrEnum` for enums where all members are strings. (Contributed by Ethan Furman in :issue:`41816`.)

Add encoding and errors parameters in :func:`fileinput.input` and :class:`fileinput.FileInput`. (Contributed by Inada Naoki in :issue:`43712`.)

:func:`fileinput.hook_compressed` now returns :class:`TextIOWrapper` object when mode is "r" and file is compressed, like uncompressed files. (Contributed by Inada Naoki in :issue:`5758`.)

Add audit hooks for :func:`gc.get_objects`, :func:`gc.get_referrers` and :func:`gc.get_referents`. (Contributed by Pablo Galindo in :issue:`43439`.)

Add the root_dir and dir_fd parameters in :func:`~glob.glob` and :func:`~glob.iglob` which allow to specify the root directory for searching. (Contributed by Serhiy Storchaka in :issue:`38144`.)

The hashlib module requires OpenSSL 1.1.1 or newer. (Contributed by Christian Heimes in PEP 644 and :issue:`43669`.)

The hashlib module has preliminary support for OpenSSL 3.0.0. (Contributed by Christian Heimes in :issue:`38820` and other issues.)

The pure-Python fallback of :func:`~hashlib.pbkdf2_hmac` is deprecated. In the future PBKDF2-HMAC will only be available when Python has been built with OpenSSL support. (Contributed by Christian Heimes in :issue:`43880`.)

The hmac module now uses OpenSSL's HMAC implementation internally. (Contributed by Christian Heimes in :issue:`40645`.)

Make IDLE invoke :func:`sys.excepthook` (when started without '-n'). User hooks were previously ignored. (Patch by Ken Hilton in :issue:`43008`.)

This change was backported to a 3.9 maintenance release.

Add a Shell sidebar. Move the primary prompt ('>>>') to the sidebar. Add secondary prompts ('...') to the sidebar. Left click and optional drag selects one or more lines of text, as with the editor line number sidebar. Right click after selecting text lines displays a context menu with 'copy with prompts'. This zips together prompts from the sidebar with lines from the selected text. This option also appears on the context menu for the text. (Contributed by Tal Einat in :issue:`37903`.)

Use spaces instead of tabs to indent interactive code. This makes interactive code entries 'look right'. Making this feasible was a major motivation for adding the shell sidebar. Contributed by Terry Jan Reedy in :issue:`37892`.)

We expect to backport these shell changes to a future 3.9 maintenance release.

Highlight the new :ref:`soft keywords <soft-keywords>` :keyword:`match`, :keyword:`case <match>`, and :keyword:`_ <wildcard-patterns>` in pattern-matching statements. However, this highlighting is not perfect and will be incorrect in some rare cases, including some _-s in case patterns. (Contributed by Tal Einat in bpo-44010.)

Feature parity with importlib_metadata 4.4 (history).

:ref:`importlib.metadata entry points <entry-points>` now provides a nicer experience for selecting entry points by group and name through a new :class:`importlib.metadata.EntryPoints` class. See the Compatibility Note in the docs for more info on the deprecation and usage.

Added :func:`importlib.metadata.packages_distributions` for resolving top-level Python modules and packages to their :class:`importlib.metadata.Distribution`.

When a module does not define __loader__, fall back to __spec__.loader. (Contributed by Brett Cannon in :issue:`42133`.)

Add :func:`inspect.get_annotations`, which safely computes the annotations defined on an object. It works around the quirks of accessing the annotations on various types of objects, and makes very few assumptions about the object it examines. :func:`inspect.get_annotations` can also correctly un-stringize stringized annotations. :func:`inspect.get_annotations` is now considered best practice for accessing the annotations dict defined on any Python object; for more information on best practices for working with annotations, please see :ref:`annotations-howto`. Relatedly, :func:`inspect.signature`, :func:`inspect.Signature.from_callable`, and :func:`inspect.Signature.from_function` now call :func:`inspect.get_annotations` to retrieve annotations. This means :func:`inspect.signature` and :func:`inspect.Signature.from_callable` can also now un-stringize stringized annotations. (Contributed by Larry Hastings in :issue:`43817`.)

When a module does not define __loader__, fall back to __spec__.loader. (Contributed by Brett Cannon in :issue:`42133`.)

Add :func:`os.cpu_count()` support for VxWorks RTOS. (Contributed by Peixing Xin in :issue:`41440`.)

Add a new function :func:`os.eventfd` and related helpers to wrap the eventfd2 syscall on Linux. (Contributed by Christian Heimes in :issue:`41001`.)

Add :func:`os.splice()` that allows to move data between two file descriptors without copying between kernel address space and user address space, where one of the file descriptors must refer to a pipe. (Contributed by Pablo Galindo in :issue:`41625`.)

Add :data:`~os.O_EVTONLY`, :data:`~os.O_FSYNC`, :data:`~os.O_SYMLINK` and :data:`~os.O_NOFOLLOW_ANY` for macOS. (Contributed by Dong-hee Na in :issue:`43106`.)

:func:`os.path.realpath` now accepts a strict keyword-only argument. When set to True, :exc:`OSError` is raised if a path doesn't exist or a symlink loop is encountered. (Contributed by Barney Gale in :issue:`43757`.)

Add slice support to :attr:`PurePath.parents <pathlib.PurePath.parents>`. (Contributed by Joshua Cannon in :issue:`35498`)

Add negative indexing support to :attr:`PurePath.parents <pathlib.PurePath.parents>`. (Contributed by Yaroslav Pankovych in :issue:`21041`)

Add :meth:`Path.hardlink_to <pathlib.Path.hardlink_to>` method that supersedes :meth:`~pathlib.Path.link_to`. The new method has the same argument order as :meth:`~pathlib.Path.symlink_to`. (Contributed by Barney Gale in :issue:`39950`.)

:meth:`pathlib.Path.stat` and :meth:`~pathlib.Path.chmod` now accept a follow_symlinks keyword-only argument for consistency with corresponding functions in the :mod:`os` module. (Contributed by Barney Gale in :issue:`39906`.)

Add :func:`platform.freedesktop_os_release()` to retrieve operation system identification from freedesktop.org os-release standard file. (Contributed by Christian Heimes in :issue:`28468`)

:mod:`pprint` can now pretty-print :class:`dataclasses.dataclass` instances. (Contributed by Lewis Gaul in :issue:`43080`.)

Add --quiet option to command-line interface of :mod:`py_compile`. (Contributed by Gregory Schevchenko in :issue:`38731`.)

Add an end_lineno attribute to the Function and Class objects in the tree returned by :func:`pyclbr.readline` and :func:`pyclbr.readline_ex`. It matches the existing (start) lineno. (Contributed by Aviral Srivastava in :issue:`38307`.)

The :mod:`shelve` module now uses :data:`pickle.DEFAULT_PROTOCOL` by default instead of :mod:`pickle` protocol 3 when creating shelves. (Contributed by Zackery Spytz in :issue:`34204`.)

Add :func:`~statistics.covariance`, Pearson's :func:`~statistics.correlation`, and simple :func:`~statistics.linear_regression` functions. (Contributed by Tymoteusz Wołodźko in :issue:`38490`.)

When a module does not define __loader__, fall back to __spec__.loader. (Contributed by Brett Cannon in :issue:`42133`.)

The exception :exc:`socket.timeout` is now an alias of :exc:`TimeoutError`. (Contributed by Christian Heimes in :issue:`42413`.)

Add option to create MPTCP sockets with IPPROTO_MPTCP (Contributed by Rui Cunha in :issue:`43571`.)

The ssl module requires OpenSSL 1.1.1 or newer. (Contributed by Christian Heimes in PEP 644 and :issue:`43669`.)

The ssl module has preliminary support for OpenSSL 3.0.0 and new option :data:`~ssl.OP_IGNORE_UNEXPECTED_EOF`. (Contributed by Christian Heimes in :issue:`38820`, :issue:`43794`, :issue:`43788`, :issue:`43791`, :issue:`43799`, :issue:`43920`, :issue:`43789`, and :issue:`43811`.)

Deprecated function and use of deprecated constants now result in a :exc:`DeprecationWarning`. The following features have been deprecated since Python 3.6, Python 3.7, or OpenSSL 1.1.0: :data:`~ssl.OP_NO_SSLv2`, :data:`~ssl.OP_NO_SSLv3`, :data:`~ssl.OP_NO_TLSv1`, :data:`~ssl.OP_NO_TLSv1_1`, :data:`~ssl.OP_NO_TLSv1_2`, :data:`~ssl.OP_NO_TLSv1_3`, :data:`~ssl.PROTOCOL_SSLv2`, :data:`~ssl.PROTOCOL_SSLv3`, :data:`~ssl.PROTOCOL_SSLv23`, :data:`~ssl.PROTOCOL_TLSv1`, :data:`~ssl.PROTOCOL_TLSv1_1`, :data:`~ssl.PROTOCOL_TLSv1_2`, :data:`~ssl.PROTOCOL_TLS`, :func:`~ssl.wrap_socket`, :func:`~ssl.match_hostname`, :func:`~ssl.RAND_pseudo_bytes`, :func:`~ssl.RAND_egd`, :meth:`ssl.SSLSocket.selected_npn_protocol`, :meth:`ssl.SSLContext.set_npn_protocols`. (Contributed by Christian Heimes in :issue:`43880`.)

The ssl module now has more secure default settings. Ciphers without forward secrecy or SHA-1 MAC are disabled by default. Security level 2 prohibits weak RSA, DH, and ECC keys with less than 112 bits of security. :class:`~ssl.SSLContext` defaults to minimum protocol version TLS 1.2. Settings are based on Hynek Schlawack's research. (Contributed by Christian Heimes in :issue:`43998`.)

The deprecated protocols SSL 3.0, TLS 1.0, and TLS 1.1 are no longer officially supported. Python does not block them actively. However OpenSSL build options, distro configurations, vendor patches, and cipher suites may prevent a successful handshake.

Add a timeout parameter to the :func:`ssl.get_server_certificate` function. (Contributed by Zackery Spytz in :issue:`31870`.)

The ssl module uses heap-types and multi-phase initialization. (Contributed by Christian Heimes in :issue:`42333`.)

A new verify flag :data:`~ssl.VERIFY_X509_PARTIAL_CHAIN` has been added. (Contributed by l0x in :issue:`40849`.)

Add audit events for :func:`~sqlite3.connect/handle`, :meth:`~sqlite3.Connection.enable_load_extension`, and :meth:`~sqlite3.Connection.load_extension`. (Contributed by Erlend E. Aasland in :issue:`43762`.)

Add :data:`sys.orig_argv` attribute: the list of the original command line arguments passed to the Python executable. (Contributed by Victor Stinner in :issue:`23427`.)

Add :data:`sys.stdlib_module_names`, containing the list of the standard library module names. (Contributed by Victor Stinner in :issue:`42955`.)

:func:`_thread.interrupt_main` now takes an optional signal number to simulate (the default is still :data:`signal.SIGINT`). (Contributed by Antoine Pitrou in :issue:`43356`.)

Add :func:`threading.gettrace` and :func:`threading.getprofile` to retrieve the functions set by :func:`threading.settrace` and :func:`threading.setprofile` respectively. (Contributed by Mario Corchero in :issue:`42251`.)

Add :data:`threading.__excepthook__` to allow retrieving the original value of :func:`threading.excepthook` in case it is set to a broken or a different value. (Contributed by Mario Corchero in :issue:`42308`.)

The :func:`~traceback.format_exception`, :func:`~traceback.format_exception_only`, and :func:`~traceback.print_exception` functions can now take an exception object as a positional-only argument. (Contributed by Zackery Spytz and Matthias Bussonnier in :issue:`26389`.)

Reintroduce the :data:`types.EllipsisType`, :data:`types.NoneType` and :data:`types.NotImplementedType` classes, providing a new set of types readily interpretable by type checkers. (Contributed by Bas van Beek in :issue:`41810`.)

For major changes, see New Features Related to Type Hints.

The behavior of :class:`typing.Literal` was changed to conform with PEP 586 and to match the behavior of static type checkers specified in the PEP.

1. Literal now de-duplicates parameters.

2. Equality comparisons between Literal objects are now order independent.

3. Literal comparisons now respects types. For example, Literal[0] == Literal[False] previously evaluated to True. It is now False. To support this change, the internally used type cache now supports differentiating types.

4. Literal objects will now raise a :exc:`TypeError` exception during equality comparisons if any of their parameters are not :term:`hashable`. Note that declaring Literal with unhashable parameters will not throw an error:

   >>> from typing import Literal
   >>> Literal[{0}]
   >>> Literal[{0}] == Literal[{False}]
   Traceback (most recent call last):
     File "<stdin>", line 1, in <module>
   TypeError: unhashable type: 'set'
   

(Contributed by Yurii Karabas in :issue:`42345`.)

Add new function :func:`typing.is_typeddict` to introspect if an annotation is a :class:`typing.TypedDict`. (Contributed by Patrick Reader in :issue:`41792`)

Subclasses of typing.Protocol which only have data variables declared will now raise a TypeError when checked with isinstance unless they are decorated with :func:`runtime_checkable`. Previously, these checks passed silently. Users should decorate their subclasses with the :func:`runtime_checkable` decorator if they want runtime protocols. (Contributed by Yurii Karabas in :issue:`38908`)

Add new method :meth:`~unittest.TestCase.assertNoLogs` to complement the existing :meth:`~unittest.TestCase.assertLogs`. (Contributed by Kit Yan Choi in :issue:`39385`.)

Python versions earlier than Python 3.10 allowed using both ; and & as query parameter separators in :func:`urllib.parse.parse_qs` and :func:`urllib.parse.parse_qsl`. Due to security concerns, and to conform with newer W3C recommendations, this has been changed to allow only a single separator key, with & as the default. This change also affects :func:`cgi.parse` and :func:`cgi.parse_multipart` as they use the affected functions internally. For more details, please see their respective documentation. (Contributed by Adam Goldschmidt, Senthil Kumaran and Ken Jin in :issue:`42967`.)

Add a :class:`~xml.sax.handler.LexicalHandler` class to the :mod:`xml.sax.handler` module. (Contributed by Jonathan Gossage and Zackery Spytz in :issue:`35018`.)

Add methods related to PEP 451: :meth:`~zipimport.zipimporter.find_spec`, :meth:`zipimport.zipimporter.create_module`, and :meth:`zipimport.zipimporter.exec_module`. (Contributed by Brett Cannon in :issue:`42131`.)

Add :meth:`~zipimport.zipimporter.invalidate_caches` method. (Contributed by Desmond Cheong in :issue:`14678`.)

• Constructors :func:`str`, :func:`bytes` and :func:`bytearray` are now faster (around 30--40% for small objects). (Contributed by Serhiy Storchaka in :issue:`41334`.)
• The :mod:`runpy` module now imports fewer modules. The python3 -m module-name command startup time is 1.4x faster in average. On Linux, python3 -I -m module-name imports 69 modules on Python 3.9, whereas it only imports 51 modules (-18) on Python 3.10. (Contributed by Victor Stinner in :issue:`41006` and :issue:`41718`.)
• The LOAD_ATTR instruction now uses new "per opcode cache" mechanism. It is about 36% faster now for regular attributes and 44% faster for slots. (Contributed by Pablo Galindo and Yury Selivanov in :issue:`42093` and Guido van Rossum in :issue:`42927`, based on ideas implemented originally in PyPy and MicroPython.)
• When building Python with :option:`--enable-optimizations` now -fno-semantic-interposition is added to both the compile and link line. This speeds builds of the Python interpreter created with :option:`--enable-shared` with gcc by up to 30%. See this article for more details. (Contributed by Victor Stinner and Pablo Galindo in :issue:`38980`.)
• Use a new output buffer management code for :mod:`bz2` / :mod:`lzma` / :mod:`zlib` modules, and add .readall() function to _compression.DecompressReader class. bz2 decompression is now 1.09x ~ 1.17x faster, lzma decompression 1.20x ~ 1.32x faster, GzipFile.read(-1) 1.11x ~ 1.18x faster. (Contributed by Ma Lin, reviewed by Gregory P. Smith, in :issue:`41486`)
• When using stringized annotations, annotations dicts for functions are no longer created when the function is created. Instead, they are stored as a tuple of strings, and the function object lazily converts this into the annotations dict on demand. This optimization cuts the CPU time needed to define an annotated function by half. (Contributed by Yurii Karabas and Inada Naoki in :issue:`42202`)
• Substring search functions such as str1 in str2 and str2.find(str1) now sometimes use Crochemore & Perrin's "Two-Way" string searching algorithm to avoid quadratic behavior on long strings. (Contributed by Dennis Sweeney in :issue:`41972`)
• Add micro-optimizations to _PyType_Lookup() to improve type attribute cache lookup performance in the common case of cache hits. This makes the interpreter 1.04 times faster on average. (Contributed by Dino Viehland in :issue:`43452`)
• The following built-in functions now support the faster PEP 590 vectorcall calling convention: :func:`map`, :func:`filter`, :func:`reversed`, :func:`bool` and :func:`float`. (Contributed by Dong-hee Na and Jeroen Demeyer in :issue:`43575`, :issue:`43287`, :issue:`41922`, :issue:`41873` and :issue:`41870`)
• :class:`BZ2File` performance is improved by removing internal RLock. This makes :class:`BZ2File` thread unsafe in the face of multiple simultaneous readers or writers, just like its equivalent classes in :mod:`gzip` and :mod:`lzma` have always been. (Contributed by Inada Naoki in :issue:`43785`).

• Starting in this release, there will be a concerted effort to begin cleaning up old import semantics that were kept for Python 2.7 compatibility. Specifically, :meth:`~importlib.abc.PathEntryFinder.find_loader`/:meth:`~importlib.abc.Finder.find_module` (superseded by :meth:`~importlib.abc.Finder.find_spec`), :meth:`~importlib.abc.Loader.load_module` (superseded by :meth:`~importlib.abc.Loader.exec_module`), :meth:`~importlib.abc.Loader.module_repr` (which the import system takes care of for you), the __package__ attribute (superseded by __spec__.parent), the __loader__ attribute (superseded by __spec__.loader), and the __cached__ attribute (superseded by __spec__.cached) will slowly be removed (as well as other classes and methods in :mod:`importlib`). :exc:`ImportWarning` and/or :exc:`DeprecationWarning` will be raised as appropriate to help identify code which needs updating during this transition.

• The entire distutils namespace is deprecated, to be removed in Python 3.12. Refer to the :ref:`module changes <distutils-deprecated>` section for more information.

• Non-integer arguments to :func:`random.randrange` are deprecated. The :exc:`ValueError` is deprecated in favor of a :exc:`TypeError`. (Contributed by Serhiy Storchaka and Raymond Hettinger in :issue:`37319`.)

• The various load_module() methods of :mod:`importlib` have been documented as deprecated since Python 3.6, but will now also trigger a :exc:`DeprecationWarning`. Use :meth:`~importlib.abc.Loader.exec_module` instead. (Contributed by Brett Cannon in :issue:`26131`.)

• :meth:`zimport.zipimporter.load_module` has been deprecated in preference for :meth:`~zipimport.zipimporter.exec_module`. (Contributed by Brett Cannon in :issue:`26131`.)

• The use of :meth:`~importlib.abc.Loader.load_module` by the import system now triggers an :exc:`ImportWarning` as :meth:`~importlib.abc.Loader.exec_module` is preferred. (Contributed by Brett Cannon in :issue:`26131`.)

• The use of :meth:`importlib.abc.MetaPathFinder.find_module` and :meth:`importlib.abc.PathEntryFinder.find_module` by the import system now trigger an :exc:`ImportWarning` as :meth:`importlib.abc.MetaPathFinder.find_spec` and :meth:`importlib.abc.PathEntryFinder.find_spec` are preferred, respectively. You can use :func:`importlib.util.spec_from_loader` to help in porting. (Contributed by Brett Cannon in :issue:`42134`.)

• The use of :meth:`importlib.abc.PathEntryFinder.find_loader` by the import system now triggers an :exc:`ImportWarning` as :meth:`importlib.abc.PathEntryFinder.find_spec` is preferred. You can use :func:`importlib.util.spec_from_loader` to help in porting. (Contributed by Brett Cannon in :issue:`43672`.)

• The various implementations of :meth:`importlib.abc.MetaPathFinder.find_module` ( :meth:`importlib.machinery.BuiltinImporter.find_module`, :meth:`importlib.machinery.FrozenImporter.find_module`, :meth:`importlib.machinery.WindowsRegistryFinder.find_module`, :meth:`importlib.machinery.PathFinder.find_module`, :meth:`importlib.abc.MetaPathFinder.find_module`), :meth:`importlib.abc.PathEntryFinder.find_module` ( :meth:`importlib.machinery.FileFinder.find_module`, ), and :meth:`importlib.abc.PathEntryFinder.find_loader` ( :meth:`importlib.machinery.FileFinder.find_loader` ) now raise :exc:`DeprecationWarning` and are slated for removal in Python 3.12 (previously they were documented as deprecated in Python 3.4). (Contributed by Brett Cannon in :issue:`42135`.)

• :class:`importlib.abc.Finder` is deprecated (including its sole method, :meth:`~importlib.abc.Finder.find_module`). Both :class:`importlib.abc.MetaPathFinder` and :class:`importlib.abc.PathEntryFinder` no longer inherit from the class. Users should inherit from one of these two classes as appropriate instead. (Contributed by Brett Cannon in :issue:`42135`.)

• The deprecations of :mod:`imp`, :func:`importlib.find_loader`, :func:`importlib.util.set_package_wrapper`, :func:`importlib.util.set_loader_wrapper`, :func:`importlib.util.module_for_loader`, :class:`pkgutil.ImpImporter`, and :class:`pkgutil.ImpLoader` have all been updated to list Python 3.12 as the slated version of removal (they began raising :exc:`DeprecationWarning` in previous versions of Python). (Contributed by Brett Cannon in :issue:`43720`.)

• The import system now uses the __spec__ attribute on modules before falling back on :meth:`~importlib.abc.Loader.module_repr` for a module's __repr__() method. Removal of the use of module_repr() is scheduled for Python 3.12. (Contributed by Brett Cannon in :issue:`42137`.)

• :meth:`importlib.abc.Loader.module_repr`, :meth:`importlib.machinery.FrozenLoader.module_repr`, and :meth:`importlib.machinery.BuiltinLoader.module_repr` are deprecated and slated for removal in Python 3.12. (Contributed by Brett Cannon in :issue:`42136`.)

• sqlite3.OptimizedUnicode has been undocumented and obsolete since Python 3.3, when it was made an alias to :class:`str`. It is now deprecated, scheduled for removal in Python 3.12. (Contributed by Erlend E. Aasland in :issue:`42264`.)

• :func:`asyncio.get_event_loop` now emits a deprecation warning if there is no running event loop. In the future it will be an alias of :func:`~asyncio.get_running_loop`. :mod:`asyncio` functions which implicitly create a :class:`~asyncio.Future` or :class:`~asyncio.Task` objects now emit a deprecation warning if there is no running event loop and no explicit loop argument is passed: :func:`~asyncio.ensure_future`, :func:`~asyncio.wrap_future`, :func:`~asyncio.gather`, :func:`~asyncio.shield`, :func:`~asyncio.as_completed` and constructors of :class:`~asyncio.Future`, :class:`~asyncio.Task`, :class:`~asyncio.StreamReader`, :class:`~asyncio.StreamReaderProtocol`. (Contributed by Serhiy Storchaka in :issue:`39529`.)

• The undocumented built-in function sqlite3.enable_shared_cache is now deprecated, scheduled for removal in Python 3.12. Its use is strongly discouraged by the SQLite3 documentation. See the SQLite3 docs for more details. If a shared cache must be used, open the database in URI mode using the cache=shared query parameter. (Contributed by Erlend E. Aasland in :issue:`24464`.)

• The following threading methods are now deprecated:

  • threading.currentThread => :func:`threading.current_thread`
  • threading.activeCount => :func:`threading.active_count`
  • threading.Condition.notifyAll => :meth:`threading.Condition.notify_all`
  • threading.Event.isSet => :meth:`threading.Event.is_set`
  • threading.Thread.setName => :attr:`threading.Thread.name`
  • threading.thread.getName => :attr:`threading.Thread.name`
  • threading.Thread.isDaemon => :attr:`threading.Thread.daemon`
  • threading.Thread.setDaemon => :attr:`threading.Thread.daemon`

  (Contributed by Jelle Zijlstra in :issue:`21574`.)

• :meth:`pathlib.Path.link_to` is deprecated and slated for removal in Python 3.12. Use :meth:`pathlib.Path.hardlink_to` instead. (Contributed by Barney Gale in :issue:`39950`.)

• cgi.log() is deprecated and slated for removal in Python 3.12. (Contributed by Inada Naoki in :issue:`41139`.)

• Removed special methods __int__, __float__, __floordiv__, __mod__, __divmod__, __rfloordiv__, __rmod__ and __rdivmod__ of the :class:`complex` class. They always raised a :exc:`TypeError`. (Contributed by Serhiy Storchaka in :issue:`41974`.)

• The ParserBase.error() method from the private and undocumented _markupbase module has been removed. :class:`html.parser.HTMLParser` is the only subclass of ParserBase and its error() implementation was already removed in Python 3.5. (Contributed by Berker Peksag in :issue:`31844`.)

• Removed the unicodedata.ucnhash_CAPI attribute which was an internal PyCapsule object. The related private _PyUnicode_Name_CAPI structure was moved to the internal C API. (Contributed by Victor Stinner in :issue:`42157`.)

• Removed the parser module, which was deprecated in 3.9 due to the switch to the new PEG parser, as well as all the C source and header files that were only being used by the old parser, including node.h, parser.h, graminit.h and grammar.h.

• Removed the Public C API functions :c:func:`PyParser_SimpleParseStringFlags`, :c:func:`PyParser_SimpleParseStringFlagsFilename`, :c:func:`PyParser_SimpleParseFileFlags` and :c:func:`PyNode_Compile` that were deprecated in 3.9 due to the switch to the new PEG parser.

• Removed the formatter module, which was deprecated in Python 3.4. It is somewhat obsolete, little used, and not tested. It was originally scheduled to be removed in Python 3.6, but such removals were delayed until after Python 2.7 EOL. Existing users should copy whatever classes they use into their code. (Contributed by Dong-hee Na and Terry J. Reedy in :issue:`42299`.)

• Removed the :c:func:`PyModule_GetWarningsModule` function that was useless now due to the _warnings module was converted to a builtin module in 2.6. (Contributed by Hai Shi in :issue:`42599`.)

• Remove deprecated aliases to :ref:`collections-abstract-base-classes` from the :mod:`collections` module. (Contributed by Victor Stinner in :issue:`37324`.)

• The loop parameter has been removed from most of :mod:`asyncio`'s :doc:`high-level API <../library/asyncio-api-index>` following deprecation in Python 3.8. The motivation behind this change is multifold:

  1. This simplifies the high-level API.
  2. The functions in the high-level API have been implicitly getting the current thread's running event loop since Python 3.7. There isn't a need to pass the event loop to the API in most normal use cases.
  3. Event loop passing is error-prone especially when dealing with loops running in different threads.

  Note that the low-level API will still accept loop. See Changes in the Python API for examples of how to replace existing code.

  (Contributed by Yurii Karabas, Andrew Svetlov, Yury Selivanov and Kyle Stanley in :issue:`42392`.)

This section lists previously described changes and other bugfixes that may require changes to your code.

• The etype parameters of the :func:`~traceback.format_exception`, :func:`~traceback.format_exception_only`, and :func:`~traceback.print_exception` functions in the :mod:`traceback` module have been renamed to exc. (Contributed by Zackery Spytz and Matthias Bussonnier in :issue:`26389`.)

• :mod:`atexit`: At Python exit, if a callback registered with :func:`atexit.register` fails, its exception is now logged. Previously, only some exceptions were logged, and the last exception was always silently ignored. (Contributed by Victor Stinner in :issue:`42639`.)

• :class:`collections.abc.Callable` generic now flattens type parameters, similar to what :data:`typing.Callable` currently does. This means that collections.abc.Callable[[int, str], str] will have __args__ of (int, str, str); previously this was ([int, str], str). Code which accesses the arguments via :func:`typing.get_args` or __args__ need to account for this change. Furthermore, :exc:`TypeError` may be raised for invalid forms of parameterizing :class:`collections.abc.Callable` which may have passed silently in Python 3.9. (Contributed by Ken Jin in :issue:`42195`.)

• :meth:`socket.htons` and :meth:`socket.ntohs` now raise :exc:`OverflowError` instead of :exc:`DeprecationWarning` if the given parameter will not fit in a 16-bit unsigned integer. (Contributed by Erlend E. Aasland in :issue:`42393`.)

• The loop parameter has been removed from most of :mod:`asyncio`'s :doc:`high-level API <../library/asyncio-api-index>` following deprecation in Python 3.8.

  A coroutine that currently looks like this:

  async def foo(loop):
      await asyncio.sleep(1, loop=loop)
  

  Should be replaced with this:

  async def foo():
      await asyncio.sleep(1)
  

  If foo() was specifically designed not to run in the current thread's running event loop (e.g. running in another thread's event loop), consider using :func:`asyncio.run_coroutine_threadsafe` instead.

  (Contributed by Yurii Karabas, Andrew Svetlov, Yury Selivanov and Kyle Stanley in :issue:`42392`.)

• The :data:`types.FunctionType` constructor now inherits the current builtins if the globals dictionary has no "__builtins__" key, rather than using {"None": None} as builtins: same behavior as :func:`eval` and :func:`exec` functions. Defining a function with def function(...): ... in Python is not affected, globals cannot be overriden with this syntax: it also inherits the current builtins. (Contributed by Victor Stinner in :issue:`42990`.)

• The MAKE_FUNCTION instruction now accepts either a dict or a tuple of strings as the function's annotations. (Contributed by Yurii Karabas and Inada Naoki in :issue:`42202`)

• PEP 644: Python now requires OpenSSL 1.1.1 or newer. OpenSSL 1.0.2 is no longer supported. (Contributed by Christian Heimes in :issue:`43669`.)

• The C99 functions :c:func:`snprintf` and :c:func:`vsnprintf` are now required to build Python. (Contributed by Victor Stinner in :issue:`36020`.)

• :mod:`sqlite3` requires SQLite 3.7.15 or higher. (Contributed by Sergey Fedoseev and Erlend E. Aasland :issue:`40744` and :issue:`40810`.)

• The :mod:`atexit` module must now always be built as a built-in module. (Contributed by Victor Stinner in :issue:`42639`.)

• Add :option:`--disable-test-modules` option to the configure script: don't build nor install test modules. (Contributed by Xavier de Gaye, Thomas Petazzoni and Peixing Xin in :issue:`27640`.)

• Add :option:`--with-wheel-pkg-dir=PATH option <--with-wheel-pkg-dir>` to the ./configure script. If specified, the :mod:`ensurepip` module looks for setuptools and pip wheel packages in this directory: if both are present, these wheel packages are used instead of ensurepip bundled wheel packages.

  Some Linux distribution packaging policies recommend against bundling dependencies. For example, Fedora installs wheel packages in the /usr/share/python-wheels/ directory and don't install the ensurepip._bundled package.

  (Contributed by Victor Stinner in :issue:`42856`.)

• Add a new :option:`configure --without-static-libpython option <--without-static-libpython>` to not build the libpythonMAJOR.MINOR.a static library and not install the python.o object file.

  (Contributed by Victor Stinner in :issue:`43103`.)

• The configure script now uses the pkg-config utility, if available, to detect the location of Tcl/Tk headers and libraries. As before, those locations can be explicitly specified with the :option:`--with-tcltk-includes` and :option:`--with-tcltk-libs` configuration options. (Contributed by Manolis Stamatogiannakis in :issue:`42603`.)

• Add :option:`--with-openssl-rpath` option to configure script. The option simplifies building Python with a custom OpenSSL installation, e.g. ./configure --with-openssl=/path/to/openssl --with-openssl-rpath=auto. (Contributed by Christian Heimes in :issue:`43466`.)

The Stable ABI (Application Binary Interface) for extension modules or embedding Python is now explicitly defined. :ref:`stable` describes C API and ABI stability guarantees along with best practices for using the Stable ABI.

(Contributed by Petr Viktorin in PEP 652 and :issue:`43795`.)

• The result of :c:func:`PyNumber_Index` now always has exact type :class:`int`. Previously, the result could have been an instance of a subclass of int. (Contributed by Serhiy Storchaka in :issue:`40792`.)

• Add a new :c:member:`~PyConfig.orig_argv` member to the :c:type:`PyConfig` structure: the list of the original command line arguments passed to the Python executable. (Contributed by Victor Stinner in :issue:`23427`.)

• The :c:func:`PyDateTime_DATE_GET_TZINFO` and :c:func:`PyDateTime_TIME_GET_TZINFO` macros have been added for accessing the tzinfo attributes of :class:`datetime.datetime` and :class:`datetime.time` objects. (Contributed by Zackery Spytz in :issue:`30155`.)

• Add a :c:func:`PyCodec_Unregister` function to unregister a codec search function. (Contributed by Hai Shi in :issue:`41842`.)

• The :c:func:`PyIter_Send` function was added to allow sending value into iterator without raising StopIteration exception. (Contributed by Vladimir Matveev in :issue:`41756`.)

• Add :c:func:`PyUnicode_AsUTF8AndSize` to the limited C API. (Contributed by Alex Gaynor in :issue:`41784`.)

• Add :c:func:`PyModule_AddObjectRef` function: similar to :c:func:`PyModule_AddObject` but don't steal a reference to the value on success. (Contributed by Victor Stinner in :issue:`1635741`.)

• Add :c:func:`Py_NewRef` and :c:func:`Py_XNewRef` functions to increment the reference count of an object and return the object. (Contributed by Victor Stinner in :issue:`42262`.)

• The :c:func:`PyType_FromSpecWithBases` and :c:func:`PyType_FromModuleAndSpec` functions now accept a single class as the bases argument. (Contributed by Serhiy Storchaka in :issue:`42423`.)

• The :c:func:`PyType_FromModuleAndSpec` function now accepts NULL tp_doc slot. (Contributed by Hai Shi in :issue:`41832`.)

• The :c:func:`PyType_GetSlot` function can accept :ref:`static types <static-types>`. (Contributed by Hai Shi and Petr Viktorin in :issue:`41073`.)

• Add a new :c:func:`PySet_CheckExact` function to the C-API to check if an object is an instance of :class:`set` but not an instance of a subtype. (Contributed by Pablo Galindo in :issue:`43277`.)

• Add :c:func:`PyErr_SetInterruptEx` which allows passing a signal number to simulate. (Contributed by Antoine Pitrou in :issue:`43356`.)

• The limited C API is now supported if :ref:`Python is built in debug mode <debug-build>` (if the Py_DEBUG macro is defined). In the limited C API, the :c:func:`Py_INCREF` and :c:func:`Py_DECREF` functions are now implemented as opaque function calls, rather than accessing directly the :c:member:`PyObject.ob_refcnt` member, if Python is built in debug mode and the Py_LIMITED_API macro targets Python 3.10 or newer. It became possible to support the limited C API in debug mode because the :c:type:`PyObject` structure is the same in release and debug mode since Python 3.8 (see :issue:`36465`).

  The limited C API is still not supported in the :option:`--with-trace-refs` special build (Py_TRACE_REFS macro). (Contributed by Victor Stinner in :issue:`43688`.)

• Add the :c:func:`Py_Is(x, y) <Py_Is>` function to test if the x object is the y object, the same as x is y in Python. Add also the :c:func:`Py_IsNone`, :c:func:`Py_IsTrue`, :c:func:`Py_IsFalse` functions to test if an object is, respectively, the None singleton, the True singleton or the False singleton. (Contributed by Victor Stinner in :issue:`43753`.)

• Add new functions to control the garbage collector from C code: :c:func:`PyGC_Enable()`, :c:func:`PyGC_Disable()`, :c:func:`PyGC_IsEnabled()`. These functions allow to activate, deactivate and query the state of the garbage collector from C code without having to import the :mod:`gc` module.

• Add a new :c:data:`Py_TPFLAGS_DISALLOW_INSTANTIATION` type flag to disallow creating type instances. (Contributed by Victor Stinner in :issue:`43916`.)

• Add a new :c:data:`Py_TPFLAGS_IMMUTABLETYPE` type flag for creating immutable type objects: type attributes cannot be set nor deleted. (Contributed by Victor Stinner and Erlend E. Aasland in :issue:`43908`.)

• The PY_SSIZE_T_CLEAN macro must now be defined to use :c:func:`PyArg_ParseTuple` and :c:func:`Py_BuildValue` formats which use #: es#, et#, s#, u#, y#, z#, U# and Z#. See :ref:`Parsing arguments and building values <arg-parsing>` and the PEP 353. (Contributed by Victor Stinner in :issue:`40943`.)

• Since :c:func:`Py_REFCNT()` is changed to the inline static function, Py_REFCNT(obj) = new_refcnt must be replaced with Py_SET_REFCNT(obj, new_refcnt): see :c:func:`Py_SET_REFCNT()` (available since Python 3.9). For backward compatibility, this macro can be used:

  #if PY_VERSION_HEX < 0x030900A4
  #  define Py_SET_REFCNT(obj, refcnt) ((Py_REFCNT(obj) = (refcnt)), (void)0)
  #endif
  

  (Contributed by Victor Stinner in :issue:`39573`.)

• Calling :c:func:`PyDict_GetItem` without :term:`GIL` held had been allowed for historical reason. It is no longer allowed. (Contributed by Victor Stinner in :issue:`40839`.)

• PyUnicode_FromUnicode(NULL, size) and PyUnicode_FromStringAndSize(NULL, size) raise DeprecationWarning now. Use :c:func:`PyUnicode_New` to allocate Unicode object without initial data. (Contributed by Inada Naoki in :issue:`36346`.)

• The private _PyUnicode_Name_CAPI structure of the PyCapsule API unicodedata.ucnhash_CAPI has been moved to the internal C API. (Contributed by Victor Stinner in :issue:`42157`.)

• :c:func:`Py_GetPath`, :c:func:`Py_GetPrefix`, :c:func:`Py_GetExecPrefix`, :c:func:`Py_GetProgramFullPath`, :c:func:`Py_GetPythonHome` and :c:func:`Py_GetProgramName` functions now return NULL if called before :c:func:`Py_Initialize` (before Python is initialized). Use the new :ref:`Python Initialization Configuration API <init-config>` to get the :ref:`Python Path Configuration. <init-path-config>`. (Contributed by Victor Stinner in :issue:`42260`.)

• :c:func:`PyList_SET_ITEM`, :c:func:`PyTuple_SET_ITEM` and :c:func:`PyCell_SET` macros can no longer be used as l-value or r-value. For example, x = PyList_SET_ITEM(a, b, c) and PyList_SET_ITEM(a, b, c) = x now fail with a compiler error. It prevents bugs like if (PyList_SET_ITEM (a, b, c) < 0) ... test. (Contributed by Zackery Spytz and Victor Stinner in :issue:`30459`.)

• The non-limited API files odictobject.h, parser_interface.h, picklebufobject.h, pyarena.h, pyctype.h, pydebug.h, pyfpe.h, and pytime.h have been moved to the Include/cpython directory. These files must not be included directly, as they are already included in Python.h: :ref:`Include Files <api-includes>`. If they have been included directly, consider including Python.h instead. (Contributed by Nicholas Sim in :issue:`35134`)

• Use the :c:data:`Py_TPFLAGS_IMMUTABLETYPE` type flag to create immutable type objects. Do not rely on :c:data:`Py_TPFLAGS_HEAPTYPE` to decide if a type object is mutable or not; check if :c:data:`Py_TPFLAGS_IMMUTABLETYPE` is set instead. (Contributed by Victor Stinner and Erlend E. Aasland in :issue:`43908`.)

• The undocumented function Py_FrozenMain has been removed from the limited API. The function is mainly useful for custom builds of Python. (Contributed by Petr Viktorin in :issue:`26241`)

• The PyUnicode_InternImmortal() function is now deprecated and will be removed in Python 3.12: use :c:func:`PyUnicode_InternInPlace` instead. (Contributed by Victor Stinner in :issue:`41692`.)

• PyObject_AsCharBuffer(), PyObject_AsReadBuffer(), PyObject_CheckReadBuffer(), and PyObject_AsWriteBuffer() are removed. Please migrate to new buffer protocol; :c:func:`PyObject_GetBuffer` and :c:func:`PyBuffer_Release`. (Contributed by Inada Naoki in :issue:`41103`.)

• Removed Py_UNICODE_str* functions manipulating Py_UNICODE* strings. (Contributed by Inada Naoki in :issue:`41123`.)

  • Py_UNICODE_strlen: use :c:func:`PyUnicode_GetLength` or :c:macro:`PyUnicode_GET_LENGTH`
  • Py_UNICODE_strcat: use :c:func:`PyUnicode_CopyCharacters` or :c:func:`PyUnicode_FromFormat`
  • Py_UNICODE_strcpy, Py_UNICODE_strncpy: use :c:func:`PyUnicode_CopyCharacters` or :c:func:`PyUnicode_Substring`
  • Py_UNICODE_strcmp: use :c:func:`PyUnicode_Compare`
  • Py_UNICODE_strncmp: use :c:func:`PyUnicode_Tailmatch`
  • Py_UNICODE_strchr, Py_UNICODE_strrchr: use :c:func:`PyUnicode_FindChar`

• Removed PyUnicode_GetMax(). Please migrate to new (PEP 393) APIs. (Contributed by Inada Naoki in :issue:`41103`.)

• Removed PyLong_FromUnicode(). Please migrate to :c:func:`PyLong_FromUnicodeObject`. (Contributed by Inada Naoki in :issue:`41103`.)

• Removed PyUnicode_AsUnicodeCopy(). Please use :c:func:`PyUnicode_AsUCS4Copy` or :c:func:`PyUnicode_AsWideCharString` (Contributed by Inada Naoki in :issue:`41103`.)

• Removed _Py_CheckRecursionLimit variable: it has been replaced by ceval.recursion_limit of the :c:type:`PyInterpreterState` structure. (Contributed by Victor Stinner in :issue:`41834`.)

• Removed undocumented macros Py_ALLOW_RECURSION and Py_END_ALLOW_RECURSION and the recursion_critical field of the :c:type:`PyInterpreterState` structure. (Contributed by Serhiy Storchaka in :issue:`41936`.)

• Removed the undocumented PyOS_InitInterrupts() function. Initializing Python already implicitly installs signal handlers: see :c:member:`PyConfig.install_signal_handlers`. (Contributed by Victor Stinner in :issue:`41713`.)

• Remove the PyAST_Validate() function. It is no longer possible to build a AST object (mod_ty type) with the public C API. The function was already excluded from the limited C API (PEP 384). (Contributed by Victor Stinner in :issue:`43244`.)

• Remove the symtable.h header file and the undocumented functions:

  • PyST_GetScope()
  • PySymtable_Build()
  • PySymtable_BuildObject()
  • PySymtable_Free()
  • Py_SymtableString()
  • Py_SymtableStringObject()

  The Py_SymtableString() function was part the stable ABI by mistake but it could not be used, because the symtable.h header file was excluded from the limited C API.

  Use Python :mod:`symtable` module instead. (Contributed by Victor Stinner in :issue:`43244`.)

• Remove :c:func:`PyOS_ReadlineFunctionPointer` from the limited C API headers and from python3.dll, the library that provides the stable ABI on Windows. Since the function takes a FILE* argument, its ABI stability cannot be guaranteed. (Contributed by Petr Viktorin in :issue:`43868`.)

• Remove ast.h, asdl.h, and Python-ast.h header files. These functions were undocumented and excluded from the limited C API. Most names defined by these header files were not prefixed by Py and so could create names conflicts. For example, Python-ast.h defined a Yield macro which was conflict with the Yield name used by the Windows <winbase.h> header. Use the Python :mod:`ast` module instead. (Contributed by Victor Stinner in :issue:`43244`.)

• Remove the compiler and parser functions using struct _mod type, because the public AST C API was removed:

  • PyAST_Compile()
  • PyAST_CompileEx()
  • PyAST_CompileObject()
  • PyFuture_FromAST()
  • PyFuture_FromASTObject()
  • PyParser_ASTFromFile()
  • PyParser_ASTFromFileObject()
  • PyParser_ASTFromFilename()
  • PyParser_ASTFromString()
  • PyParser_ASTFromStringObject()

  These functions were undocumented and excluded from the limited C API. (Contributed by Victor Stinner in :issue:`43244`.)

• Remove the pyarena.h header file with functions:

  • PyArena_New()
  • PyArena_Free()
  • PyArena_Malloc()
  • PyArena_AddPyObject()

  These functions were undocumented, excluded from the limited C API, and were only used internally by the compiler. (Contributed by Victor Stinner in :issue:`43244`.)