파이썬 3.13의 새로운 기능에 대해 알아봅시다¶

개선된 대화형 인터프리터¶

파이썬 3.13은 기본적으로 PyPy project의 코드를 기반으로 한 새로운 대화형 (interactive) 셸을 사용합니다. 사용자가 대화형 터미널에서 REPL을 시작하면 다음과 같은 새 기능이 지원됩니다:

• 여러개의 줄을 편집할 때에도 기록을 보존할 수 있습니다.

• help, exit, quit 같은 REPL 전용 명령을 직접 지원합니다. 또한, 함수로 호출할 필요 없이 사용할 수 있습니다.

• 프롬프트와 트레이스백에서 기본적으로 색상이 활성화됩니다.

• F1으로 대화형 도움말을 탐색할 수 있으며, 별도로 명령 기록이 제공됩니다.

• F2로 명령 기록을 탐색할 때, 출력을 건너뛸 뿐 아니라 >>>와 … 프롬프트도 건너뜁니다.

• F3의 “붙여넣기 모드”는 큰 코드 블록을 붙여넣기 쉽게 해 줍니다(F3를 다시 누르면 일반 프롬프트로 다시 돌아갑니다).

새 대화형 셸을 비활성화하려면 PYTHON_BASIC_REPL 환경 변수를 설정하면 됩니다. 대화형 모드에 대한 자세한 내용은 대화형 모드을 참조 부탁드립니다.

(PyPy 프로젝트의 코드를 기반으로, Pablo Galindo Salgado, Łukasz Langa, Lysandros Nikolaou가 gh-111201을 통해 기능을 구현하였습니다. Dino Viehland와 Anthony Shaw가 Windows 지원에 기여하였습니다.)

개선된 에러 메세지¶

• 인터프리터는 터미널에서 트레이스백을 표시할 때 기본적으로 색상을 사용합니다. 이러한 색상 적용을 변경하기 위해 PYTHON_COLORS 환경 변수 및 NO_COLOR 및 FORCE_COLOR 환경 변수를 활용할 수 있으며, 이러한 활용을 통해 색상 적용을 제어할 수 있습니다. (Pablo Galindo Salgado가 gh-112730를 통해 기능을 구현하였습니다.)

• 표준 라이브러리 모듈과 동일한 이름의 스크립트를 작성하는 실수는 (여러 사용자에 의해) 흔히 발생합니다. 표준 라이브러리 모듈과 동일한 이름의 스크립트로 인해 오류가 발생하는 경우에 보다 자세한 에러 메세지 제공함을 통해 보다 도움이 될 수 있도록 하였습니다:

  $ python random.py
  Traceback (most recent call last):
    File "/home/me/random.py", line 1, in <module>
      import random
    File "/home/me/random.py", line 3, in <module>
      print(random.randint(5))
            ^^^^^^^^^^^^^^
  AttributeError: module 'random' has no attribute 'randint' (consider renaming '/home/me/random.py' since it has the same name as the standard library module named 'random' and prevents importing that standard library module)
  

  또한, 스크립트가 임포트(import)하려는 제삼자(third-party) 모듈과 이름이 같아서 오류가 발생하는 경우에도 보다 자세한 에러 메세지 제공함을 통해 보다 도움이 될 수 있도록 하였습니다:

  $ python numpy.py
  Traceback (most recent call last):
    File "/home/me/numpy.py", line 1, in <module>
      import numpy as np
    File "/home/me/numpy.py", line 3, in <module>
      np.array([1, 2, 3])
      ^^^^^^^^
  AttributeError: module 'numpy' has no attribute 'array' (consider renaming '/home/me/numpy.py' if it has the same name as a library you intended to import)
  

  (Shantanu Jain이 gh-95754을 통해 기능을 구현하였습니다.)

• 함수에 잘못된 키워드 인자가 전달되는 경우, 올바른 키워드 인자가 에러 메세지 안에서 제안되도록 하였습니다.

  >>> "Better error messages!".split(max_split=1)
  Traceback (most recent call last):
    File "<python-input-0>", line 1, in <module>
      "Better error messages!".split(max_split=1)
      ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
  TypeError: split() got an unexpected keyword argument 'max_split'. Did you mean 'maxsplit'?
  

  (Pablo Galindo Salgado와 Shantanu Jain이 gh-107944를 통해 기능을 구현하였습니다.)

제한 없는 스레드를 사용하는 파이썬(CPython)¶

CPython은 전역 인터프리터 잠금 (GIL)을 비활성화한 제한 없는 스레드 모드(free-threaded mode)에서 실행을 실험적으로 지원합니다. 이는 실험적 기능이므로 기본 설정으로 활성화되어 있지는 않습니다. 제한 없는 스레드 모드를 실행시키기 위해서는 python3.13t 또는 python3.13t.exe라는 별도의 실행 파일이 필요합니다. free-threaded로 표시된 미리 빌드된 바이너리는 공식 Windows 및 macOS 설치 프로그램의 일부로 설치할 수 있으며, 또는 --disable-gil 옵션을 사용하여 소스 코드로부터 CPython을 빌드할 수 있습니다.

제한 없는 스레드 실행은 사용 가능한 CPU 코어에서 스레드를 병렬로 실행하여 가용 처리 능력을 온전히 활용할 수 있게 합니다. 모든 소프트웨어가 자동적으로 이를 통해 이점을 갖게 되는 것은 아니지만, 스레딩을 염두에 두고 설계된 프로그램은 다중 코어 하드웨어에서 보다 더 빠르게 실행됩니다. 제한 없는 스레드 모드는 실험적이며 개선 작업이 진행 중이므로, 버그가 있을 수 있으며, 단일 스레드를 통한 실행의 성능이 저하될 수 있습니다. 제한 없는 스레드를 사용하는 파이썬(CPython) 빌드는 실행 시 환경 변수 PYTHON_GIL 또는 명령줄 옵션 -X gil=1을 사용해 GIL을 활성화하여 선택적으로 실행할 수 있습니다.

사용 중인 인터프리터가 제한 없는 스레드의 사용을 지원하는지 확인하려면 python -VV 및 sys.version에서 “experimental free-threading build”를 통해 확인할 수 있습니다. 새 sys._is_gil_enabled() 함수를 사용하면 실행 중인 프로세스에서 GIL이 실제로 비활성화되어 있는지 확인할 수 있습니다.

C-API 확장 모듈은 제한 없는 스레드(free-threaded)를 사용하는 빌드용으로 따로 빌드해야 합니다. 전역 인터프리터 잠금(Global Interpreter Lock; GIL) 비활성 실행을 지원하는 확장 모듈은 Py_mod_gil 슬롯을 사용해야 합니다. 개별 단계 초기화(single-phase init)를 사용하는 확장 모듈은 GIL 비활성 실행 지원 여부를 표시하기 위해 PyUnstable_Module_SetGIL()을 사용해야 합니다. 이러한 메커니즘을 사용하지 않는 C 확장을 임포트하면 GIL이 자동으로 활성화됩니다. PYTHON_GIL 환경 변수나 -X gil=0 옵션으로 GIL을 명시적으로 비활성화한 경우는 예외입니다. 제한 없는 스레드를 사용하는 빌드에서 C 확장이 포함된 패키지를 설치하려면 pip 24.1 이상이 필요합니다.

이러한 개선 작업은 많은 개인과 조직, 그리고 제한 없는 스레드(free-threading) 사용 지원을 테스트하고 활성화하는 데 기여한 파이썬 및 다양한 제삼자(third-parth) 프로젝트 커뮤니티들 덕분이었습니다. 주목할 만한 기여자로는 Sam Gross, Ken Jin, Donghee Na, Itamar Oren, Matt Page, Brett Simmers, Dino Viehland, Carl Meyer, Nathan Goldbaum, Ralf Gommers, Lysandros Nikolaou 등이 있으며, 이외에도 많은 분들이 있습니다. 이 기여자들 중 많은 분들이 Meta에 재직하고 있으며, Meta는 이 프로젝트를 지원하기 위해 상당한 엔지니어링 자원을 제공하였습니다.

실험적인 지정 시간 구동 (Just-In-Time; JIT) 컴파일러¶

--enable-experimental-jit 옵션을 사용하여 CPython을 구성하고 빌드하면 일부 파이썬 프로그램의 속도를 높일 수 있는 JIT 컴파일러가 추가됩니다. 윈도우에서 JIT를 활성화하려면 PCbuild/build.bat --experimental-jit을 사용하고, 2단계 (Tier 2) 인터프리터를 활성화하려면 --experimental-jit-interpreter를 사용하세요. 빌드 요구 사항과 추가 지원 정보는 Tools/jit/README.md를 확인 부탁드립니다.

--enable-experimental-jit 옵션은 다음과 같은 특정 값들을 받을 수 있습니다. 특정 값을 지정하지 않으면 기본값은 yes입니다.

• no: 2단계 (Tier 2) 인터프리터 및 JIT 파이프라인을 모두 비활성화합니다.

• yes: JIT (Just-In-Time; 지정 시간 구동) 컴파일러를 활성화합니다. 런타임에 JIT를 비활성화하려면 환경 변수 PYTHON_JIT=0을 설정할 수 있습니다.

• yes-off: JIT를 빌드하되 기본값으로 비활성화합니다. 런타임에 JIT를 활성화하려면 환경 변수 PYTHON_JIT=1을 설정할 수 있습니다.

• interpreter: 2단계 (Tier 2) 인터프리터를 활성화하지만 JIT는 비활성화합니다. 이 인터프리터는 환경 변수 PYTHON_JIT=0을 설정하여 비활성화할 수 있습니다.

내부 아키텍처는 대략적으로 다음과 같습니다:

• 시작 지점은 최적화된 1단계(Tier 1) 바이트 코드입니다. 관련된 보다 구체적인 정보를 위해서는 3.11 버전의 새로운 기능를 열람 부탁드립니다.

• 1단계(Tier 1) 바이트 코드가 빈번하게 실행되는 경우, 완전히 내부용인 새로운 중간 표현(IR)인 2단계 IR(Tier 2 IR)로 변환됩니다. 변환된 이러한 중간 표현은 마이크로 연산(uops)으로 명명되기도 합니다.

• 2단계 IR도 1단계와 같은 스택 기반 가상 머신을 사용하지만, 명령어 형식의 경우 기계어로 변환하기에 보다 더 적합합니다.

• 2단계 IR에는 여러개의 최적화 패스가 있으며, 이는 인터프리터에 의해 처리되어 실행되거나 기계어로 변환되기 전에 적용됩니다.

• 2단계(Tier 2) 인터프리터가 있지만, 주로 앞서 서술한 최적화 파이프라인의 초기 단계를 디버깅하기 위해 만들어진 것입니다. 2단계 인터프리터는 파이썬에서 --enable-experimental-jit=interpreter 옵션을 설정함을 통해 활성화할 수 있습니다.

• 지정 시간 구동 (JIT; Just-In-Time) 컴파일러가 활성화되면, 최적화된 2단계(Tier 2) IR이 기계어로 변환되어 실행됩니다.

• 기계어 변환 과정에서는 복사-후-패치 (copy-and-patch)라는 기법을 사용합니다. 런타임 의존성은 없지만, 빌드의 경우 LLVM에 대한 새로운 의존성이 추가되었습니다.

(JIT은 Brandt Bucher에 의해, Haoran Xu와 Fredrik Kjolstad의 논문에 영감을 받아 개발되었습니다. 2단계 IR의 경우 Mark Shannon과 Guido van Rossum에 의해 개발되었습니다. 2단계 IR의 최적화의 경우 Ken Jin에 의해 개발되었습니다.)

locals()의 변형에 대한 의미 구조 정의¶

지금까지 locals()의 반환값을 변경했을 때에 예상되는 결과는 개별적인 파이썬 구현에 맡겨져 있었습니다. Python 3.13부터는 PEP 667이 대부분의 코드 실행 스코프에 대해 CPython의 기존 동작을 표준화하고, 최적화된 스코프(함수, 제너레이터, 코루틴, 컴프리헨션, 제너레이터 표현식)에서는 현재 할당되어 있는 지역 변수들에 대한 서로 독립적인 스냅샷으로 명시적으로 반환하도록 변경합니다. 이는 지역적으로 참조된 비지역 변수들이 종료 시점에서 확인되는 경우을 포함합니다.

앞서 서술한 것과 같이 최적화된 스코프에서 locals()의 의미 구조가 바뀌면서, 명시적인 네임 스페이스를 제공하지 않는 경우, 암시적으로 locals()을 대상으로 실행되는 코드 실행 함수들(예: exec(), eval())의 기본 동작도 영향을 받았습니다. 이전 버전에서는 코드 실행 함수를 호출한 뒤 locals()을 호출하여 변경 사항에 액세스할 수 있는지가 구체적인 개별 구현에 따라 달라졌었습니다. 구체적으로, CPython에서 대부분의 경우 이러한 방식은 무리 없이 잘 동작하였지만, 종종 같은 스코프의 공유 스냅샷이 다른 코드(디버거나 코드 실행 추적 도구 등)에 의해 재설정될 수 있기 때문에 최적화된 스코프에서 실패하는 경우도 있었습니다. Python 3.13에서는 해당 코드를 항상 최적화된 스코프의 지역 변수에 대한 독립적인 스냅샷에 대해서만 실행하므로, 그 변경 사항은 이후의 locals() 호출에서 절대로 보이지 않습니다. 이러한 경우들에서 변경 내용을 확인하기 위해서는 관련 함수에 명시적인 네임 스페이스를 전달해야 합니다. 또는 디스크에서 파이썬 파일을 실행할 때처럼 실행 결과의 네임 스페이스를 반환하는 상위 수준의 코드 실행 API(예를 들어, runpy.run_path())로 코드를 변경하는 것이 보다 바람직할 수 있습니다.

디버거와 유사 도구들이 (앞서 서술한 변화의 영향을 받는 스코프에서) 지역 변수를 안정적으로 업데이트할 수 있도록, FrameType.f_locals는 해당 스코프의 지역 변수와 그 스코프에서 참조되는 비지역 변수에 대한 이어쓰기(write-through) 프록시를 반환합니다. 런타임 의미가 불분명하고 일관성 없이 업데이트되던 (공유된) dict 인스턴스를 반환하지 않습니다.

관련된 C API 변경 사항과 사용 지원 중단 등에 대한 보다 구체적인 정보를 위해서는 PEP 667을 열람 부탁드립니다. 또한 관련된 (코드) 이식에 대한 가이드의 경우 Python API 및 C API을 열람 부탁드립니다.

(Mark Shannon과 Tian Gao가 PEP을 작성하였고 gh-74929를 통해 기능을 구현하였습니다. Guido van Rossum과 Alyssa Coghlan이 설명서를 최신화하였습니다.)

모바일 플랫폼들에 대한 지원¶

PEP 730: iOS가 PEP 11에서 지원하는 플랫폼으로, arm64-apple-ios (2013년 이후 출시된 iPhone 및 iPad 기기)와 arm64-apple-ios-simulator (Apple 실리콘 하드웨어에서 실행되는 Xcode iOS 시뮬레이터)는 3단계 (Tier 3)로 지정되었습니다. x86_64-apple-ios-simulator (구형 x86_64 하드웨어에서 실행되는 Xcode iOS 시뮬레이터)는 3단계 지원 플랫폼은 아니지만, 최선의 노력(best-effort)으로 지원됩니다. (Russell Keith-Magee PEP을 작성하였고 gh-114099를 통해 기능을 구현하였습니다.)

PEP 738: Android가 PEP 11에서 지원하는 플랫폼으로, aarch64-linux-android와 x86_64-linux-android은 3단계 (Tier 3)로 지정되었습니다. 32비트 arm-linux-androideabi와 i686-linux-android는 3단계 지원 플랫폼은 아니지만, 최선의 노력(best-effort)으로 지원됩니다. (Malcolm Smith가 PEP을 작성하였고 gh-116622을 통해 기능을 구현하였습니다.)

argparse¶

• Add the deprecated parameter to the add_argument() and add_parser() methods, to enable deprecating command-line options, positional arguments, and subcommands. (Contributed by Serhiy Storchaka in gh-83648.)

array¶

• Add the 'w' type code (Py_UCS4) for Unicode characters. It should be used instead of the deprecated 'u' type code. (Contributed by Inada Naoki in gh-80480.)

• Register array.array as a MutableSequence by implementing the clear() method. (Contributed by Mike Zimin in gh-114894.)

ast¶

• The constructors of node types in the ast module are now stricter in the arguments they accept, with more intuitive behavior when arguments are omitted.

  If an optional field on an AST node is not included as an argument when constructing an instance, the field will now be set to None. Similarly, if a list field is omitted, that field will now be set to an empty list, and if an expr_context field is omitted, it defaults to Load(). (Previously, in all cases, the attribute would be missing on the newly constructed AST node instance.)

  In all other cases, where a required argument is omitted, the node constructor will emit a DeprecationWarning. This will raise an exception in Python 3.15. Similarly, passing a keyword argument to the constructor that does not map to a field on the AST node is now deprecated, and will raise an exception in Python 3.15.

  These changes do not apply to user-defined subclasses of ast.AST unless the class opts in to the new behavior by defining the AST._field_types mapping.

  (Contributed by Jelle Zijlstra in gh-105858, gh-117486, and gh-118851.)

• ast.parse() now accepts an optional argument optimize which is passed on to compile(). This makes it possible to obtain an optimized AST. (Contributed by Irit Katriel in gh-108113.)

asyncio¶

• asyncio.as_completed() now returns an object that is both an asynchronous iterator and a plain iterator of awaitables. The awaitables yielded by asynchronous iteration include original task or future objects that were passed in, making it easier to associate results with the tasks being completed. (Contributed by Justin Arthur in gh-77714.)

• asyncio.loop.create_unix_server() will now automatically remove the Unix socket when the server is closed. (Contributed by Pierre Ossman in gh-111246.)

• DatagramTransport.sendto() will now send zero-length datagrams if called with an empty bytes object. The transport flow control also now accounts for the datagram header when calculating the buffer size. (Contributed by Jamie Phan in gh-115199.)

• Add Queue.shutdown and QueueShutDown to manage queue termination. (Contributed by Laurie Opperman and Yves Duprat in gh-104228.)

• Add the Server.close_clients() and Server.abort_clients() methods, which more forcefully close an asyncio server. (Contributed by Pierre Ossman in gh-113538.)

• Accept a tuple of separators in StreamReader.readuntil(), stopping when any one of them is encountered. (Contributed by Bruce Merry in gh-81322.)

• Improve the behavior of TaskGroup when an external cancellation collides with an internal cancellation. For example, when two task groups are nested and both experience an exception in a child task simultaneously, it was possible that the outer task group would hang, because its internal cancellation was swallowed by the inner task group.

  In the case where a task group is cancelled externally and also must raise an ExceptionGroup, it will now call the parent task’s cancel() method. This ensures that a CancelledError will be raised at the next await, so the cancellation is not lost.

  An added benefit of these changes is that task groups now preserve the cancellation count (cancelling()).

  In order to handle some corner cases, uncancel() may now reset the undocumented _must_cancel flag when the cancellation count reaches zero.

  (Inspired by an issue reported by Arthur Tacca in gh-116720.)

• When TaskGroup.create_task() is called on an inactive TaskGroup, the given coroutine will be closed (which prevents a RuntimeWarning about the given coroutine being never awaited). (Contributed by Arthur Tacca and Jason Zhang in gh-115957.)

• The function and methods named create_task have received a new **kwargs argument that is passed through to the task constructor. This change was accidentally added in 3.13.3, and broke the API contract for custom task factories. Several third-party task factories implemented workarounds for this. In 3.13.4 and later releases the old factory contract is honored once again (until 3.14). To keep the workarounds working, the extra **kwargs argument still allows passing additional keyword arguments to Task and to custom task factories.

  This affects the following function and methods: asyncio.create_task(), asyncio.loop.create_task(), asyncio.TaskGroup.create_task(). (Contributed by Thomas Grainger in gh-128307.)

base64¶

• Add z85encode() and z85decode() functions for encoding bytes as Z85 data and decoding Z85-encoded data to bytes. (Contributed by Matan Perelman in gh-75299.)

compileall¶

• The default number of worker threads and processes is now selected using os.process_cpu_count() instead of os.cpu_count(). (Contributed by Victor Stinner in gh-109649.)

concurrent.futures¶

• The default number of worker threads and processes is now selected using os.process_cpu_count() instead of os.cpu_count(). (Contributed by Victor Stinner in gh-109649.)

configparser¶

• ConfigParser now has support for unnamed sections, which allows for top-level key-value pairs. This can be enabled with the new allow_unnamed_section parameter. (Contributed by Pedro Sousa Lacerda in gh-66449.)

copy¶

• The new replace() function and the replace protocol make creating modified copies of objects much simpler. This is especially useful when working with immutable objects. The following types support the replace() function and implement the replace protocol:

  • collections.namedtuple()

  • dataclasses.dataclass

  • datetime.datetime, datetime.date, datetime.time

  • inspect.Signature, inspect.Parameter

  • types.SimpleNamespace

  • code objects

  Any user-defined class can also support copy.replace() by defining the __replace__() method. (Contributed by Serhiy Storchaka in gh-108751.)

ctypes¶

• As a consequence of necessary internal refactoring, initialization of internal metaclasses now happens in __init__ rather than in __new__. This affects projects that subclass these internal metaclasses to provide custom initialization. Generally:

  • Custom logic that was done in __new__ after calling super().__new__ should be moved to __init__.

  • To create a class, call the metaclass, not only the metaclass’s __new__ method.

  See gh-124520 for discussion and links to changes in some affected projects.

• ctypes.Structure objects have a new _align_ attribute which allows the alignment of the structure being packed to/from memory to be specified explicitly. (Contributed by Matt Sanderson in gh-112433)

dbm¶

• Add dbm.sqlite3, a new module which implements an SQLite backend, and make it the default dbm backend. (Contributed by Raymond Hettinger and Erlend E. Aasland in gh-100414.)

• Allow removing all items from the database through the new clear() methods of the GDBM and NDBM database objects. (Contributed by Donghee Na in gh-107122.)

dis¶

• Change the output of dis module functions to show logical labels for jump targets and exception handlers, rather than offsets. The offsets can be added with the new -O command-line option or the show_offsets argument. (Contributed by Irit Katriel in gh-112137.)

• get_instructions() no longer represents cache entries as separate instructions. Instead, it returns them as part of the Instruction, in the new cache_info field. The show_caches argument to get_instructions() is deprecated and no longer has any effect. (Contributed by Irit Katriel in gh-112962.)

doctest¶

• doctest output is now colored by default. This can be controlled via the new PYTHON_COLORS environment variable as well as the canonical NO_COLOR and FORCE_COLOR environment variables. See also Controlling color. (Contributed by Hugo van Kemenade in gh-117225.)

• The DocTestRunner.run() method now counts the number of skipped tests. Add the DocTestRunner.skips and TestResults.skipped attributes. (Contributed by Victor Stinner in gh-108794.)

email¶

• Headers with embedded newlines are now quoted on output. The generator will now refuse to serialize (write) headers that are improperly folded or delimited, such that they would be parsed as multiple headers or joined with adjacent data. If you need to turn this safety feature off, set verify_generated_headers. (Contributed by Bas Bloemsaat and Petr Viktorin in gh-121650.)

• getaddresses() and parseaddr() now return ('', '') pairs in more situations where invalid email addresses are encountered instead of potentially inaccurate values. The two functions have a new optional strict parameter (default True). To get the old behavior (accepting malformed input), use strict=False. getattr(email.utils, 'supports_strict_parsing', False) can be used to check if the strict parameter is available. (Contributed by Thomas Dwyer and Victor Stinner for gh-102988 to improve the CVE 2023-27043 fix.)

enum¶

• EnumDict has been made public to better support subclassing EnumType.

fractions¶

• Fraction objects now support the standard format specification mini-language rules for fill, alignment, sign handling, minimum width, and grouping. (Contributed by Mark Dickinson in gh-111320.)

glob¶

• Add translate(), a function to convert a path specification with shell-style wildcards to a regular expression. (Contributed by Barney Gale in gh-72904.)

importlib¶

• The following functions in importlib.resources now allow accessing a directory (or tree) of resources, using multiple positional arguments (the encoding and errors arguments in the text-reading functions are now keyword-only):

  • is_resource()

  • open_binary()

  • open_text()

  • path()

  • read_binary()

  • read_text()

  These functions are no longer deprecated and are not scheduled for removal. (Contributed by Petr Viktorin in gh-116608.)

• contents() remains deprecated in favor of the fully-featured Traversable API. However, there is now no plan to remove it. (Contributed by Petr Viktorin in gh-116608.)

io¶

• The IOBase finalizer now logs any errors raised by the close() method with sys.unraisablehook. Previously, errors were ignored silently by default, and only logged in Python Development Mode or when using a Python debug build. (Contributed by Victor Stinner in gh-62948.)

ipaddress¶

• Add the IPv4Address.ipv6_mapped property, which returns the IPv4-mapped IPv6 address. (Contributed by Charles Machalow in gh-109466.)

• Fix is_global and is_private behavior in IPv4Address, IPv6Address, IPv4Network, and IPv6Network. (Contributed by Jakub Stasiak in gh-113171.)

itertools¶

• batched() has a new strict parameter, which raises a ValueError if the final batch is shorter than the specified batch size. (Contributed by Raymond Hettinger in gh-113202.)

marshal¶

• Add the allow_code parameter in module functions. Passing allow_code=False prevents serialization and de-serialization of code objects which are incompatible between Python versions. (Contributed by Serhiy Storchaka in gh-113626.)

math¶

• The new function fma() performs fused multiply-add operations. This computes x * y + z with only a single round, and so avoids any intermediate loss of precision. It wraps the fma() function provided by C99, and follows the specification of the IEEE 754 “fusedMultiplyAdd” operation for special cases. (Contributed by Mark Dickinson and Victor Stinner in gh-73468.)

mimetypes¶

• Add the guess_file_type() function to guess a MIME type from a filesystem path. Using paths with guess_type() is now soft deprecated. (Contributed by Serhiy Storchaka in gh-66543.)

mmap¶

• mmap is now protected from crashing on Windows when the mapped memory is inaccessible due to file system errors or access violations. (Contributed by Jannis Weigend in gh-118209.)

• mmap has a new seekable() method that can be used when a seekable file-like object is required. The seek() method now returns the new absolute position. (Contributed by Donghee Na and Sylvie Liberman in gh-111835.)

• The new UNIX-only trackfd parameter for mmap controls file descriptor duplication; if false, the file descriptor specified by fileno will not be duplicated. (Contributed by Zackery Spytz and Petr Viktorin in gh-78502.)

multiprocessing¶

• The default number of worker threads and processes is now selected using os.process_cpu_count() instead of os.cpu_count(). (Contributed by Victor Stinner in gh-109649.)

os¶

• Add process_cpu_count() function to get the number of logical CPU cores usable by the calling thread of the current process. (Contributed by Victor Stinner in gh-109649.)

• cpu_count() and process_cpu_count() can be overridden through the new environment variable PYTHON_CPU_COUNT or the new command-line option -X cpu_count. This option is useful for users who need to limit CPU resources of a container system without having to modify application code or the container itself. (Contributed by Donghee Na in gh-109595.)

• Add a low level interface to Linux’s timer file descriptors via timerfd_create(), timerfd_settime(), timerfd_settime_ns(), timerfd_gettime(), timerfd_gettime_ns(), TFD_NONBLOCK, TFD_CLOEXEC, TFD_TIMER_ABSTIME, and TFD_TIMER_CANCEL_ON_SET (Contributed by Masaru Tsuchiyama in gh-108277.)

• lchmod() and the follow_symlinks argument of chmod() are both now available on Windows. Note that the default value of follow_symlinks in lchmod() is False on Windows. (Contributed by Serhiy Storchaka in gh-59616.)

• fchmod() and support for file descriptors in chmod() are both now available on Windows. (Contributed by Serhiy Storchaka in gh-113191.)

• On Windows, mkdir() and makedirs() now support passing a mode value of 0o700 to apply access control to the new directory. This implicitly affects tempfile.mkdtemp() and is a mitigation for CVE 2024-4030. Other values for mode continue to be ignored. (Contributed by Steve Dower in gh-118486.)

• posix_spawn() now accepts None for the env argument, which makes the newly spawned process use the current process environment. (Contributed by Jakub Kulik in gh-113119.)

• posix_spawn() can now use the POSIX_SPAWN_CLOSEFROM attribute in the file_actions parameter on platforms that support posix_spawn_file_actions_addclosefrom_np(). (Contributed by Jakub Kulik in gh-113117.)

os.path¶

• Add isreserved() to check if a path is reserved on the current system. This function is only available on Windows. (Contributed by Barney Gale in gh-88569.)

• On Windows, isabs() no longer considers paths starting with exactly one slash (\ or /) to be absolute. (Contributed by Barney Gale and Jon Foster in gh-44626.)

• realpath() now resolves MS-DOS style file names even if the file is not accessible. (Contributed by Moonsik Park in gh-82367.)

pathlib¶

• Add UnsupportedOperation, which is raised instead of NotImplementedError when a path operation isn’t supported. (Contributed by Barney Gale in gh-89812.)

• Add a new constructor for creating Path objects from ‘file’ URIs (file:///), Path.from_uri(). (Contributed by Barney Gale in gh-107465.)

• Add PurePath.full_match() for matching paths with shell-style wildcards, including the recursive wildcard “**”. (Contributed by Barney Gale in gh-73435.)

• Add the PurePath.parser class attribute to store the implementation of os.path used for low-level path parsing and joining. This will be either posixpath or ntpath.

• Add recurse_symlinks keyword-only argument to Path.glob() and rglob(). (Contributed by Barney Gale in gh-77609.)

• Path.glob() and rglob() now return files and directories when given a pattern that ends with “**”. Previously, only directories were returned. (Contributed by Barney Gale in gh-70303.)

• Add the follow_symlinks keyword-only argument to Path.is_file, Path.is_dir, Path.owner(), and Path.group(). (Contributed by Barney Gale in gh-105793 and Kamil Turek in gh-107962.)

pdb¶

• breakpoint() and set_trace() now enter the debugger immediately rather than on the next line of code to be executed. This change prevents the debugger from breaking outside of the context when breakpoint() is positioned at the end of the context. (Contributed by Tian Gao in gh-118579.)

• sys.path[0] is no longer replaced by the directory of the script being debugged when sys.flags.safe_path is set. (Contributed by Tian Gao and Christian Walther in gh-111762.)

• zipapp is now supported as a debugging target. (Contributed by Tian Gao in gh-118501.)

• Add ability to move between chained exceptions during post-mortem debugging in pm() using the new exceptions [exc_number] command for Pdb. (Contributed by Matthias Bussonnier in gh-106676.)

• Expressions and statements whose prefix is a pdb command are now correctly identified and executed. (Contributed by Tian Gao in gh-108464.)

queue¶

• Add Queue.shutdown and ShutDown to manage queue termination. (Contributed by Laurie Opperman and Yves Duprat in gh-104750.)

random¶

• Add a command-line interface. (Contributed by Hugo van Kemenade in gh-118131.)

re¶

• Rename re.error to PatternError for improved clarity. re.error is kept for backward compatibility.

shutil¶

• Support the dir_fd and follow_symlinks keyword arguments in chown(). (Contributed by Berker Peksag and Tahia K in gh-62308)

site¶

• .pth files are now decoded using UTF-8 first, and then with the locale encoding if UTF-8 decoding fails. (Contributed by Inada Naoki in gh-117802.)

sqlite3¶

• A ResourceWarning is now emitted if a Connection object is not closed explicitly. (Contributed by Erlend E. Aasland in gh-105539.)

• Add the filter keyword-only parameter to Connection.iterdump() for filtering database objects to dump. (Contributed by Mariusz Felisiak in gh-91602.)

ssl¶

• The create_default_context() API now includes VERIFY_X509_PARTIAL_CHAIN and VERIFY_X509_STRICT in its default flags.

  참고

  VERIFY_X509_STRICT may reject pre-RFC 5280 or malformed certificates that the underlying OpenSSL implementation might otherwise accept. Whilst disabling this is not recommended, you can do so using:

  import ssl
  
  ctx = ssl.create_default_context()
  ctx.verify_flags &= ~ssl.VERIFY_X509_STRICT
  

  (Contributed by William Woodruff in gh-112389.)

statistics¶

• Add kde() for kernel density estimation. This makes it possible to estimate a continuous probability density function from a fixed number of discrete samples. (Contributed by Raymond Hettinger in gh-115863.)

• Add kde_random() for sampling from an estimated probability density function created by kde(). (Contributed by Raymond Hettinger in gh-115863.)

subprocess¶

• The subprocess module now uses the posix_spawn() function in more situations.

  Notably, when close_fds is True (the default), posix_spawn() will be used when the C library provides posix_spawn_file_actions_addclosefrom_np(), which includes recent versions of Linux, FreeBSD, and Solaris. On Linux, this should perform similarly to the existing Linux vfork() based code.

  A private control knob subprocess._USE_POSIX_SPAWN can be set to False if you need to force subprocess to never use posix_spawn(). Please report your reason and platform details in the issue tracker if you set this so that we can improve our API selection logic for everyone. (Contributed by Jakub Kulik in gh-113117.)

sys¶

• Add the _is_interned() function to test if a string was interned. This function is not guaranteed to exist in all implementations of Python. (Contributed by Serhiy Storchaka in gh-78573.)

tempfile¶

• On Windows, the default mode 0o700 used by tempfile.mkdtemp() now limits access to the new directory due to changes to os.mkdir(). This is a mitigation for CVE 2024-4030. (Contributed by Steve Dower in gh-118486.)

time¶

• On Windows, monotonic() now uses the QueryPerformanceCounter() clock for a resolution of 1 microsecond, instead of the GetTickCount64() clock which has a resolution of 15.6 milliseconds. (Contributed by Victor Stinner in gh-88494.)

• On Windows, time() now uses the GetSystemTimePreciseAsFileTime() clock for a resolution of 1 microsecond, instead of the GetSystemTimeAsFileTime() clock which has a resolution of 15.6 milliseconds. (Contributed by Victor Stinner in gh-63207.)

tkinter¶

• Add tkinter widget methods: tk_busy_hold(), tk_busy_configure(), tk_busy_cget(), tk_busy_forget(), tk_busy_current(), and tk_busy_status(). (Contributed by Miguel, klappnase and Serhiy Storchaka in gh-72684.)

• The tkinter widget method wm_attributes() now accepts the attribute name without the minus prefix to get window attributes, for example w.wm_attributes('alpha') and allows specifying attributes and values to set as keyword arguments, for example w.wm_attributes(alpha=0.5). (Contributed by Serhiy Storchaka in gh-43457.)

• wm_attributes() can now return attributes as a dict, by using the new optional keyword-only parameter return_python_dict. (Contributed by Serhiy Storchaka in gh-43457.)

• Text.count() can now return a simple int when the new optional keyword-only parameter return_ints is used. Otherwise, the single count is returned as a 1-tuple or None. (Contributed by Serhiy Storchaka in gh-97928.)

• Support the “vsapi” element type in the element_create() method of tkinter.ttk.Style. (Contributed by Serhiy Storchaka in gh-68166.)

• Add the after_info() method for Tkinter widgets. (Contributed by Cheryl Sabella in gh-77020.)

• Add a new copy_replace() method to PhotoImage to copy a region from one image to another, possibly with pixel zooming, subsampling, or both. (Contributed by Serhiy Storchaka in gh-118225.)

• Add from_coords parameter to the PhotoImage methods copy(), zoom() and subsample(). Add zoom and subsample parameters to the PhotoImage method copy(). (Contributed by Serhiy Storchaka in gh-118225.)

• Add the PhotoImage methods read() to read an image from a file and data() to get the image data. Add background and grayscale parameters to the write() method. (Contributed by Serhiy Storchaka in gh-118271.)

traceback¶

• Add the exc_type_str attribute to TracebackException, which holds a string display of the exc_type. Deprecate the exc_type attribute, which holds the type object itself. Add parameter save_exc_type (default True) to indicate whether exc_type should be saved. (Contributed by Irit Katriel in gh-112332.)

• Add a new show_group keyword-only parameter to TracebackException.format_exception_only() to (recursively) format the nested exceptions of a BaseExceptionGroup instance. (Contributed by Irit Katriel in gh-105292.)

types¶

• SimpleNamespace can now take a single positional argument to initialise the namespace’s arguments. This argument must either be a mapping or an iterable of key-value pairs. (Contributed by Serhiy Storchaka in gh-108191.)

typing¶

• PEP 705: Add ReadOnly, a special typing construct to mark a TypedDict item as read-only for type checkers.

• PEP 742: Add TypeIs, a typing construct that can be used to instruct a type checker how to narrow a type.

• Add NoDefault, a sentinel object used to represent the defaults of some parameters in the typing module. (Contributed by Jelle Zijlstra in gh-116126.)

• Add get_protocol_members() to return the set of members defining a typing.Protocol. (Contributed by Jelle Zijlstra in gh-104873.)

• Add is_protocol() to check whether a class is a Protocol. (Contributed by Jelle Zijlstra in gh-104873.)

• ClassVar can now be nested in Final, and vice versa. (Contributed by Mehdi Drissi in gh-89547.)

unicodedata¶

• Update the Unicode database to version 15.1.0. (Contributed by James Gerity in gh-109559.)

venv¶

• Add support for creating source control management (SCM) ignore files in a virtual environment’s directory. By default, Git is supported. This is implemented as opt-in via the API, which can be extended to support other SCMs (EnvBuilder and create()), and opt-out via the CLI, using --without-scm-ignore-files. (Contributed by Brett Cannon in gh-108125.)

warnings¶

• PEP 702: The new warnings.deprecated() decorator provides a way to communicate deprecations to a static type checker and to warn on usage of deprecated classes and functions. A DeprecationWarning may also be emitted when a decorated function or class is used at runtime. (Contributed by Jelle Zijlstra in gh-104003.)

xml¶

• Allow controlling Expat >=2.6.0 reparse deferral (CVE 2023-52425) by adding five new methods:

  • xml.etree.ElementTree.XMLParser.flush()

  • xml.etree.ElementTree.XMLPullParser.flush()

  • xml.parsers.expat.xmlparser.GetReparseDeferralEnabled()

  • xml.parsers.expat.xmlparser.SetReparseDeferralEnabled()

  • xml.sax.expatreader.ExpatParser.flush()

  (Contributed by Sebastian Pipping in gh-115623.)

• Add the close() method for the iterator returned by iterparse() for explicit cleanup. (Contributed by Serhiy Storchaka in gh-69893.)

zipimport¶

• Add support for ZIP64 format files. Everybody loves huge data, right? (Contributed by Tim Hatch in gh-94146.)

PEP 594: Remove “dead batteries” from the standard library¶

PEP 594 proposed removing 19 modules from the standard library, colloquially referred to as ‘dead batteries’ due to their historic, obsolete, or insecure status. All of the following modules were deprecated in Python 3.11, and are now removed:

• aifc

  • standard-aifc: Use the redistribution of aifc library from PyPI.

• audioop

  • audioop-lts: Use audioop-lts library from PyPI.

• chunk

  • standard-chunk: Use the redistribution of chunk library from PyPI.

• cgi and cgitb

  • cgi.FieldStorage can typically be replaced with urllib.parse.parse_qsl() for GET and HEAD requests, and the email.message module or the multipart library for POST and PUT requests.

  • cgi.parse() can be replaced by calling urllib.parse.parse_qs() directly on the desired query string, unless the input is multipart/form-data, which should be replaced as described below for cgi.parse_multipart().

  • cgi.parse_header() can be replaced with the functionality in the email package, which implements the same MIME RFCs. For example, with email.message.EmailMessage:

    from email.message import EmailMessage
    
    msg = EmailMessage()
    msg['content-type'] = 'application/json; charset="utf8"'
    main, params = msg.get_content_type(), msg['content-type'].params
    

  • cgi.parse_multipart() can be replaced with the functionality in the email package, which implements the same MIME RFCs, or with the multipart library. For example, the email.message.EmailMessage and email.message.Message classes.

  • standard-cgi: and standard-cgitb: Use the redistribution of cgi and cgitb library from PyPI.

• crypt and the private _crypt extension. The hashlib module may be an appropriate replacement when simply hashing a value is required. Otherwise, various third-party libraries on PyPI are available:

  • bcrypt: Modern password hashing for your software and your servers.

  • argon2-cffi: The secure Argon2 password hashing algorithm.

  • legacycrypt: ctypes wrapper to the POSIX crypt library call and associated functionality.

  • crypt_r: Fork of the crypt module, wrapper to the crypt_r(3) library call and associated functionality.

  • standard-crypt and deprecated-crypt-alternative: Use the redistribution of crypt and reimplementation of _crypt libraries from PyPI.

• imghdr: The filetype, puremagic, or python-magic libraries should be used as replacements. For example, the puremagic.what() function can be used to replace the imghdr.what() function for all file formats that were supported by imghdr.

  • standard-imghdr: Use the redistribution of imghdr library from PyPI.

• mailcap: Use the mimetypes module instead.

  • standard-mailcap: Use the redistribution of mailcap library from PyPI.

• msilib

• nis

• nntplib: Use the pynntp library from PyPI instead.

  • standard-nntplib: Use the redistribution of nntplib library from PyPI.

• ossaudiodev: For audio playback, use the pygame library from PyPI instead.

• pipes: Use the subprocess module instead. Use shlex.quote() to replace the undocumented pipes.quote function.

  • standard-pipes: Use the redistribution of pipes library from PyPI.

• sndhdr: The filetype, puremagic, or python-magic libraries should be used as replacements.

  • standard-sndhdr: Use the redistribution of sndhdr library from PyPI.

• spwd: Use the python-pam library from PyPI instead.

• sunau

  • standard-sunau: Use the redistribution of sunau library from PyPI.

• telnetlib, Use the telnetlib3 or Exscript libraries from PyPI instead.

  • standard-telnetlib: Use the redistribution of telnetlib library from PyPI.

• uu: Use the base64 module instead, as a modern alternative.

  • standard-uu: Use the redistribution of uu library from PyPI.

• xdrlib

  • standard-xdrlib: Use the redistribution of xdrlib library from PyPI.

(Contributed by Victor Stinner and Zachary Ware in gh-104773 and gh-104780.)

2to3¶

• Remove the 2to3 program and the lib2to3 module, previously deprecated in Python 3.11. (Contributed by Victor Stinner in gh-104780.)

builtins¶

• Remove support for chained classmethod descriptors (introduced in gh-63272). These can no longer be used to wrap other descriptors, such as property. The core design of this feature was flawed and led to several problems. To “pass-through” a classmethod, consider using the __wrapped__ attribute that was added in Python 3.10. (Contributed by Raymond Hettinger in gh-89519.)

• Raise a RuntimeError when calling frame.clear() on a suspended frame (as has always been the case for an executing frame). (Contributed by Irit Katriel in gh-79932.)

configparser¶

• Remove the undocumented LegacyInterpolation class, deprecated in the docstring since Python 3.2, and at runtime since Python 3.11. (Contributed by Hugo van Kemenade in gh-104886.)

importlib.metadata¶

• Remove deprecated subscript (__getitem__()) access for EntryPoint objects. (Contributed by Jason R. Coombs in gh-113175.)

locale¶

• Remove the locale.resetlocale() function, deprecated in Python 3.11. Use locale.setlocale(locale.LC_ALL, "") instead. (Contributed by Victor Stinner in gh-104783.)

opcode¶

• Move opcode.ENABLE_SPECIALIZATION to _opcode.ENABLE_SPECIALIZATION. This field was added in 3.12, it was never documented, and is not intended for external use. (Contributed by Irit Katriel in gh-105481.)

• Remove opcode.is_pseudo(), opcode.MIN_PSEUDO_OPCODE, and opcode.MAX_PSEUDO_OPCODE, which were added in Python 3.12, but were neither documented nor exposed through dis, and were not intended to be used externally. (Contributed by Irit Katriel in gh-105481.)

optparse¶

• This module is no longer considered soft deprecated. While argparse remains preferred for new projects that aren’t using a third party command line argument processing library, there are aspects of the way argparse works that mean the lower level optparse module may provide a better foundation for writing argument processing libraries, and for implementing command line applications which adhere more strictly than argparse does to various Unix command line processing conventions that originate in the behaviour of the C getopt() function . (Contributed by Alyssa Coghlan and Serhiy Storchaka in gh-126180.)

pathlib¶

• Remove the ability to use Path objects as context managers. This functionality was deprecated and has had no effect since Python 3.9. (Contributed by Barney Gale in gh-83863.)

re¶

• Remove the undocumented, deprecated, and broken re.template() function and re.TEMPLATE / re.T flag. (Contributed by Serhiy Storchaka and Nikita Sobolev in gh-105687.)

tkinter.tix¶

• Remove the tkinter.tix module, deprecated in Python 3.6. The third-party Tix library which the module wrapped is unmaintained. (Contributed by Zachary Ware in gh-75552.)

turtle¶

• Remove the RawTurtle.settiltangle() method, deprecated in the documentation since Python 3.1 and at runtime since Python 3.11. (Contributed by Hugo van Kemenade in gh-104876.)

typing¶

• Remove the typing.io and typing.re namespaces, deprecated since Python 3.8. The items in those namespaces can be imported directly from the typing module. (Contributed by Sebastian Rittau in gh-92871.)

• Remove the keyword-argument method of creating TypedDict types, deprecated in Python 3.11. (Contributed by Tomas Roun in gh-104786.)

unittest¶

• Remove the following unittest functions, deprecated in Python 3.11:

  • unittest.findTestCases()

  • unittest.makeSuite()

  • unittest.getTestCaseNames()

  Use TestLoader methods instead:

  • loadTestsFromModule()

  • loadTestsFromTestCase()

  • getTestCaseNames()

  (Contributed by Hugo van Kemenade in gh-104835.)

• Remove the untested and undocumented TestProgram.usageExit() method, deprecated in Python 3.11. (Contributed by Hugo van Kemenade in gh-104992.)

urllib¶

• Remove the cafile, capath, and cadefault parameters of the urllib.request.urlopen() function, deprecated in Python 3.6. Use the context parameter instead with an SSLContext instance. The ssl.SSLContext.load_cert_chain() function can be used to load specific certificates, or let ssl.create_default_context() select the operating system’s trusted certificate authority (CA) certificates. (Contributed by Victor Stinner in gh-105382.)

webbrowser¶

• Remove the untested and undocumented MacOSX class, deprecated in Python 3.11. Use the MacOSXOSAScript class (introduced in Python 3.2) instead. (Contributed by Hugo van Kemenade in gh-104804.)

• Remove the deprecated MacOSXOSAScript._name attribute. Use the MacOSXOSAScript.name attribute instead. (Contributed by Nikita Sobolev in gh-105546.)

Pending removal in Python 3.14¶

• argparse: The type, choices, and metavar parameters of argparse.BooleanOptionalAction are deprecated and will be removed in 3.14. (Contributed by Nikita Sobolev in gh-92248.)

• ast: The following features have been deprecated in documentation since Python 3.8, now cause a DeprecationWarning to be emitted at runtime when they are accessed or used, and will be removed in Python 3.14:

  • ast.Num

  • ast.Str

  • ast.Bytes

  • ast.NameConstant

  • ast.Ellipsis

  Use ast.Constant instead. (Contributed by Serhiy Storchaka in gh-90953.)

• asyncio:

  • The child watcher classes asyncio.MultiLoopChildWatcher, asyncio.FastChildWatcher, asyncio.AbstractChildWatcher and asyncio.SafeChildWatcher are deprecated and will be removed in Python 3.14. (Contributed by Kumar Aditya in gh-94597.)

  • asyncio.set_child_watcher(), asyncio.get_child_watcher(), asyncio.AbstractEventLoopPolicy.set_child_watcher() and asyncio.AbstractEventLoopPolicy.get_child_watcher() are deprecated and will be removed in Python 3.14. (Contributed by Kumar Aditya in gh-94597.)

  • The get_event_loop() method of the default event loop policy now emits a DeprecationWarning if there is no current event loop set and it decides to create one. (Contributed by Serhiy Storchaka and Guido van Rossum in gh-100160.)

• email: Deprecated the isdst parameter in email.utils.localtime(). (Contributed by Alan Williams in gh-72346.)

• importlib.abc deprecated classes:

  • importlib.abc.ResourceReader

  • importlib.abc.Traversable

  • importlib.abc.TraversableResources

  Use importlib.resources.abc classes instead:

  • importlib.resources.abc.Traversable

  • importlib.resources.abc.TraversableResources

  (Contributed by Jason R. Coombs and Hugo van Kemenade in gh-93963.)

• itertools had undocumented, inefficient, historically buggy, and inconsistent support for copy, deepcopy, and pickle operations. This will be removed in 3.14 for a significant reduction in code volume and maintenance burden. (Contributed by Raymond Hettinger in gh-101588.)

• multiprocessing: The default start method will change to a safer one on Linux, BSDs, and other non-macOS POSIX platforms where 'fork' is currently the default (gh-84559). Adding a runtime warning about this was deemed too disruptive as the majority of code is not expected to care. Use the get_context() or set_start_method() APIs to explicitly specify when your code requires 'fork'. See 컨텍스트 및 시작 방법.

• pathlib: is_relative_to() and relative_to(): passing additional arguments is deprecated.

• pkgutil: pkgutil.find_loader() and pkgutil.get_loader() now raise DeprecationWarning; use importlib.util.find_spec() instead. (Contributed by Nikita Sobolev in gh-97850.)

• pty:

  • master_open(): use pty.openpty().

  • slave_open(): use pty.openpty().

• sqlite3:

  • version and version_info.

  • execute() and executemany() if named placeholders are used and parameters is a sequence instead of a dict.

• urllib: urllib.parse.Quoter is deprecated: it was not intended to be a public API. (Contributed by Gregory P. Smith in gh-88168.)

Pending removal in Python 3.15¶

• The import system:

  • Setting __cached__ on a module while failing to set __spec__.cached is deprecated. In Python 3.15, __cached__ will cease to be set or take into consideration by the import system or standard library. (gh-97879)

  • Setting __package__ on a module while failing to set __spec__.parent is deprecated. In Python 3.15, __package__ will cease to be set or take into consideration by the import system or standard library. (gh-97879)

• ctypes:

  • The undocumented ctypes.SetPointerType() function has been deprecated since Python 3.13.

• http.server:

  • The obsolete and rarely used CGIHTTPRequestHandler has been deprecated since Python 3.13. No direct replacement exists. Anything is better than CGI to interface a web server with a request handler.

  • The --cgi flag to the python -m http.server command-line interface has been deprecated since Python 3.13.

• importlib:

  • load_module() method: use exec_module() instead.

• locale:

  • The getdefaultlocale() function has been deprecated since Python 3.11. Its removal was originally planned for Python 3.13 (gh-90817), but has been postponed to Python 3.15. Use getlocale(), setlocale(), and getencoding() instead. (Contributed by Hugo van Kemenade in gh-111187.)

• pathlib:

  • PurePath.is_reserved() has been deprecated since Python 3.13. Use os.path.isreserved() to detect reserved paths on Windows.

• platform:

  • java_ver() has been deprecated since Python 3.13. This function is only useful for Jython support, has a confusing API, and is largely untested.

• sysconfig:

  • The check_home argument of sysconfig.is_python_build() has been deprecated since Python 3.12.

• threading:

  • RLock() will take no arguments in Python 3.15. Passing any arguments has been deprecated since Python 3.14, as the Python version does not permit any arguments, but the C version allows any number of positional or keyword arguments, ignoring every argument.

• types:

  • types.CodeType: Accessing co_lnotab was deprecated in PEP 626 since 3.10 and was planned to be removed in 3.12, but it only got a proper DeprecationWarning in 3.12. May be removed in 3.15. (Contributed by Nikita Sobolev in gh-101866.)

• typing:

  • The undocumented keyword argument syntax for creating NamedTuple classes (for example, Point = NamedTuple("Point", x=int, y=int)) has been deprecated since Python 3.13. Use the class-based syntax or the functional syntax instead.

  • When using the functional syntax of TypedDicts, failing to pass a value to the fields parameter (TD = TypedDict("TD")) or passing None (TD = TypedDict("TD", None)) has been deprecated since Python 3.13. Use class TD(TypedDict): pass or TD = TypedDict("TD", {}) to create a TypedDict with zero field.

  • The typing.no_type_check_decorator() decorator function has been deprecated since Python 3.13. After eight years in the typing module, it has yet to be supported by any major type checker.

• wave:

  • The getmark(), setmark(), and getmarkers() methods of the Wave_read and Wave_write classes have been deprecated since Python 3.13.

• zipimport:

  • load_module() has been deprecated since Python 3.10. Use exec_module() instead. (Contributed by Jiahao Li in gh-125746.)

Pending removal in Python 3.16¶

• The import system:

  • Setting __loader__ on a module while failing to set __spec__.loader is deprecated. In Python 3.16, __loader__ will cease to be set or taken into consideration by the import system or the standard library.

• array:

  • The 'u' format code (wchar_t) has been deprecated in documentation since Python 3.3 and at runtime since Python 3.13. Use the 'w' format code (Py_UCS4) for Unicode characters instead.

• asyncio:

  • asyncio.iscoroutinefunction() is deprecated and will be removed in Python 3.16; use inspect.iscoroutinefunction() instead. (Contributed by Jiahao Li and Kumar Aditya in gh-122875.)

  • asyncio policy system is deprecated and will be removed in Python 3.16. In particular, the following classes and functions are deprecated:

    • asyncio.AbstractEventLoopPolicy

    • asyncio.DefaultEventLoopPolicy

    • asyncio.WindowsSelectorEventLoopPolicy

    • asyncio.WindowsProactorEventLoopPolicy

    • asyncio.get_event_loop_policy()

    • asyncio.set_event_loop_policy()

    Users should use asyncio.run() or asyncio.Runner with loop_factory to use the desired event loop implementation.

    For example, to use asyncio.SelectorEventLoop on Windows:

    import asyncio
    
    async def main():
        ...
    
    asyncio.run(main(), loop_factory=asyncio.SelectorEventLoop)
    

    (Contributed by Kumar Aditya in gh-127949.)

• builtins:

  • Bitwise inversion on boolean types, ~True or ~False has been deprecated since Python 3.12, as it produces surprising and unintuitive results (-2 and -1). Use not x instead for the logical negation of a Boolean. In the rare case that you need the bitwise inversion of the underlying integer, convert to int explicitly (~int(x)).

• functools:

  • Calling the Python implementation of functools.reduce() with function or sequence as keyword arguments has been deprecated since Python 3.14.

• logging:

  Support for custom logging handlers with the strm argument is deprecated and scheduled for removal in Python 3.16. Define handlers with the stream argument instead. (Contributed by Mariusz Felisiak in gh-115032.)

• mimetypes:

  • Valid extensions start with a ‘.’ or are empty for mimetypes.MimeTypes.add_type(). Undotted extensions are deprecated and will raise a ValueError in Python 3.16. (Contributed by Hugo van Kemenade in gh-75223.)

• shutil:

  • The ExecError exception has been deprecated since Python 3.14. It has not been used by any function in shutil since Python 3.4, and is now an alias of RuntimeError.

• symtable:

  • The Class.get_methods method has been deprecated since Python 3.14.

• sys:

  • The _enablelegacywindowsfsencoding() function has been deprecated since Python 3.13. Use the PYTHONLEGACYWINDOWSFSENCODING environment variable instead.

• sysconfig:

  • The sysconfig.expand_makefile_vars() function has been deprecated since Python 3.14. Use the vars argument of sysconfig.get_paths() instead.

• tarfile:

  • The undocumented and unused TarFile.tarfile attribute has been deprecated since Python 3.13.

Pending removal in Python 3.17¶

• collections.abc:

  • collections.abc.ByteString is scheduled for removal in Python 3.17.

    Use isinstance(obj, collections.abc.Buffer) to test if obj implements the buffer protocol at runtime. For use in type annotations, either use Buffer or a union that explicitly specifies the types your code supports (e.g., bytes | bytearray | memoryview).

    ByteString was originally intended to be an abstract class that would serve as a supertype of both bytes and bytearray. However, since the ABC never had any methods, knowing that an object was an instance of ByteString never actually told you anything useful about the object. Other common buffer types such as memoryview were also never understood as subtypes of ByteString (either at runtime or by static type checkers).

    See PEP 688 for more details. (Contributed by Shantanu Jain in gh-91896.)

• typing:

  • Before Python 3.14, old-style unions were implemented using the private class typing._UnionGenericAlias. This class is no longer needed for the implementation, but it has been retained for backward compatibility, with removal scheduled for Python 3.17. Users should use documented introspection helpers like typing.get_origin() and typing.get_args() instead of relying on private implementation details.

  • typing.ByteString, deprecated since Python 3.9, is scheduled for removal in Python 3.17.

    Use isinstance(obj, collections.abc.Buffer) to test if obj implements the buffer protocol at runtime. For use in type annotations, either use Buffer or a union that explicitly specifies the types your code supports (e.g., bytes | bytearray | memoryview).

    ByteString was originally intended to be an abstract class that would serve as a supertype of both bytes and bytearray. However, since the ABC never had any methods, knowing that an object was an instance of ByteString never actually told you anything useful about the object. Other common buffer types such as memoryview were also never understood as subtypes of ByteString (either at runtime or by static type checkers).

    See PEP 688 for more details. (Contributed by Shantanu Jain in gh-91896.)

Pending removal in future versions¶

The following APIs will be removed in the future, although there is currently no date scheduled for their removal.

• argparse:

  • Nesting argument groups and nesting mutually exclusive groups are deprecated.

  • Passing the undocumented keyword argument prefix_chars to add_argument_group() is now deprecated.

  • The argparse.FileType type converter is deprecated.

• builtins:

  • Generators: throw(type, exc, tb) and athrow(type, exc, tb) signature is deprecated: use throw(exc) and athrow(exc) instead, the single argument signature.

  • Currently Python accepts numeric literals immediately followed by keywords, for example 0in x, 1or x, 0if 1else 2. It allows confusing and ambiguous expressions like [0x1for x in y] (which can be interpreted as [0x1 for x in y] or [0x1f or x in y]). A syntax warning is raised if the numeric literal is immediately followed by one of keywords and, else, for, if, in, is and or. In a future release it will be changed to a syntax error. (gh-87999)

  • Support for __index__() and __int__() method returning non-int type: these methods will be required to return an instance of a strict subclass of int.

  • Support for __float__() method returning a strict subclass of float: these methods will be required to return an instance of float.

  • Support for __complex__() method returning a strict subclass of complex: these methods will be required to return an instance of complex.

  • Delegation of int() to __trunc__() method.

  • Passing a complex number as the real or imag argument in the complex() constructor is now deprecated; it should only be passed as a single positional argument. (Contributed by Serhiy Storchaka in gh-109218.)

• calendar: calendar.January and calendar.February constants are deprecated and replaced by calendar.JANUARY and calendar.FEBRUARY. (Contributed by Prince Roshan in gh-103636.)

• codecs: use open() instead of codecs.open(). (gh-133038)

• codeobject.co_lnotab: use the codeobject.co_lines() method instead.

• datetime:

  • utcnow(): use datetime.datetime.now(tz=datetime.UTC).

  • utcfromtimestamp(): use datetime.datetime.fromtimestamp(timestamp, tz=datetime.UTC).

• gettext: Plural value must be an integer.

• importlib:

  • cache_from_source() debug_override parameter is deprecated: use the optimization parameter instead.

• importlib.metadata:

  • EntryPoints tuple interface.

  • Implicit None on return values.

• logging: the warn() method has been deprecated since Python 3.3, use warning() instead.

• mailbox: Use of StringIO input and text mode is deprecated, use BytesIO and binary mode instead.

• os: Calling os.register_at_fork() in multi-threaded process.

• pydoc.ErrorDuringImport: A tuple value for exc_info parameter is deprecated, use an exception instance.

• re: More strict rules are now applied for numerical group references and group names in regular expressions. Only sequence of ASCII digits is now accepted as a numerical reference. The group name in bytes patterns and replacement strings can now only contain ASCII letters and digits and underscore. (Contributed by Serhiy Storchaka in gh-91760.)

• sre_compile, sre_constants and sre_parse modules.

• shutil: rmtree()’s onerror parameter is deprecated in Python 3.12; use the onexc parameter instead.

• ssl options and protocols:

  • ssl.SSLContext without protocol argument is deprecated.

  • ssl.SSLContext: set_npn_protocols() and selected_npn_protocol() are deprecated: use ALPN instead.

  • ssl.OP_NO_SSL* options

  • ssl.OP_NO_TLS* options

  • ssl.PROTOCOL_SSLv3

  • ssl.PROTOCOL_TLS

  • ssl.PROTOCOL_TLSv1

  • ssl.PROTOCOL_TLSv1_1

  • ssl.PROTOCOL_TLSv1_2

  • ssl.TLSVersion.SSLv3

  • ssl.TLSVersion.TLSv1

  • ssl.TLSVersion.TLSv1_1

• threading methods:

  • threading.Condition.notifyAll(): use notify_all().

  • threading.Event.isSet(): use is_set().

  • threading.Thread.isDaemon(), threading.Thread.setDaemon(): use threading.Thread.daemon attribute.

  • threading.Thread.getName(), threading.Thread.setName(): use threading.Thread.name attribute.

  • threading.currentThread(): use threading.current_thread().

  • threading.activeCount(): use threading.active_count().

• typing.Text (gh-92332).

• The internal class typing._UnionGenericAlias is no longer used to implement typing.Union. To preserve compatibility with users using this private class, a compatibility shim will be provided until at least Python 3.17. (Contributed by Jelle Zijlstra in gh-105499.)

• unittest.IsolatedAsyncioTestCase: it is deprecated to return a value that is not None from a test case.

• urllib.parse deprecated functions: urlparse() instead

  • splitattr()

  • splithost()

  • splitnport()

  • splitpasswd()

  • splitport()

  • splitquery()

  • splittag()

  • splittype()

  • splituser()

  • splitvalue()

  • to_bytes()

• wsgiref: SimpleHandler.stdout.write() should not do partial writes.

• xml.etree.ElementTree: Testing the truth value of an Element is deprecated. In a future release it will always return True. Prefer explicit len(elem) or elem is not None tests instead.

• sys._clear_type_cache() is deprecated: use sys._clear_internal_caches() instead.

새로운 기능들¶

• Add the PyMonitoring C API for generating PEP 669 monitoring events:

  • PyMonitoringState

  • PyMonitoring_FirePyStartEvent()

  • PyMonitoring_FirePyResumeEvent()

  • PyMonitoring_FirePyReturnEvent()

  • PyMonitoring_FirePyYieldEvent()

  • PyMonitoring_FireCallEvent()

  • PyMonitoring_FireLineEvent()

  • PyMonitoring_FireJumpEvent()

  • PyMonitoring_FireBranchEvent

  • PyMonitoring_FireCReturnEvent()

  • PyMonitoring_FirePyThrowEvent()

  • PyMonitoring_FireRaiseEvent()

  • PyMonitoring_FireCRaiseEvent()

  • PyMonitoring_FireReraiseEvent()

  • PyMonitoring_FireExceptionHandledEvent()

  • PyMonitoring_FirePyUnwindEvent()

  • PyMonitoring_FireStopIterationEvent()

  • PyMonitoring_EnterScope()

  • PyMonitoring_ExitScope()

  (Contributed by Irit Katriel in gh-111997).

• Add PyMutex, a lightweight mutex that occupies a single byte, and the new PyMutex_Lock() and PyMutex_Unlock() functions. PyMutex_Lock() will release the GIL (if currently held) if the operation needs to block. (Contributed by Sam Gross in gh-108724.)

• Add the PyTime C API to provide access to system clocks:

  • PyTime_t.

  • PyTime_MIN and PyTime_MAX.

  • PyTime_AsSecondsDouble().

  • PyTime_Monotonic().

  • PyTime_MonotonicRaw().

  • PyTime_PerfCounter().

  • PyTime_PerfCounterRaw().

  • PyTime_Time().

  • PyTime_TimeRaw().

  (Contributed by Victor Stinner and Petr Viktorin in gh-110850.)

• Add the PyDict_ContainsString() function with the same behavior as PyDict_Contains(), but key is specified as a const char* UTF-8 encoded bytes string, rather than a PyObject*. (Contributed by Victor Stinner in gh-108314.)

• Add the PyDict_GetItemRef() and PyDict_GetItemStringRef() functions, which behave similarly to PyDict_GetItemWithError(), but return a strong reference instead of a borrowed reference. Moreover, these functions return -1 on error, removing the need to check PyErr_Occurred(). (Contributed by Victor Stinner in gh-106004.)

• Add the PyDict_SetDefaultRef() function, which behaves similarly to PyDict_SetDefault(), but returns a strong reference instead of a borrowed reference. This function returns -1 on error, 0 on insertion, and 1 if the key was already present in the dictionary. (Contributed by Sam Gross in gh-112066.)

• Add the PyDict_Pop() and PyDict_PopString() functions to remove a key from a dictionary and optionally return the removed value. This is similar to dict.pop(), though there is no default value, and KeyError is not raised for missing keys. (Contributed by Stefan Behnel and Victor Stinner in gh-111262.)

• Add the PyMapping_GetOptionalItem() and PyMapping_GetOptionalItemString() functions as alternatives to PyObject_GetItem() and PyMapping_GetItemString() respectively. The new functions do not raise KeyError if the requested key is missing from the mapping. These variants are more convenient and faster if a missing key should not be treated as a failure. (Contributed by Serhiy Storchaka in gh-106307.)

• Add the PyObject_GetOptionalAttr() and PyObject_GetOptionalAttrString() functions as alternatives to PyObject_GetAttr() and PyObject_GetAttrString() respectively. The new functions do not raise AttributeError if the requested attribute is not found on the object. These variants are more convenient and faster if the missing attribute should not be treated as a failure. (Contributed by Serhiy Storchaka in gh-106521.)

• Add the PyErr_FormatUnraisable() function as an extension to PyErr_WriteUnraisable() that allows customizing the warning message. (Contributed by Serhiy Storchaka in gh-108082.)

• Add new functions that return a strong reference instead of a borrowed reference for frame locals, globals, and builtins, as part of PEP 667:

  • PyEval_GetFrameBuiltins() replaces PyEval_GetBuiltins()

  • PyEval_GetFrameGlobals() replaces PyEval_GetGlobals()

  • PyEval_GetFrameLocals() replaces PyEval_GetLocals()

  (Contributed by Mark Shannon and Tian Gao in gh-74929.)

• Add the Py_GetConstant() and Py_GetConstantBorrowed() functions to get strong or borrowed references to constants. For example, Py_GetConstant(Py_CONSTANT_ZERO) returns a strong reference to the constant zero. (Contributed by Victor Stinner in gh-115754.)

• Add the PyImport_AddModuleRef() function as a replacement for PyImport_AddModule() that returns a strong reference instead of a borrowed reference. (Contributed by Victor Stinner in gh-105922.)

• Add the Py_IsFinalizing() function to check whether the main Python interpreter is shutting down. (Contributed by Victor Stinner in gh-108014.)

• Add the PyList_GetItemRef() function as a replacement for PyList_GetItem() that returns a strong reference instead of a borrowed reference. (Contributed by Sam Gross in gh-114329.)

• Add the PyList_Extend() and PyList_Clear() functions, mirroring the Python list.extend() and list.clear() methods. (Contributed by Victor Stinner in gh-111138.)

• Add the PyLong_AsInt() function. It behaves similarly to PyLong_AsLong(), but stores the result in a C int instead of a C long. (Contributed by Victor Stinner in gh-108014.)

• Add the PyLong_AsNativeBytes(), PyLong_FromNativeBytes(), and PyLong_FromUnsignedNativeBytes() functions to simplify converting between native integer types and Python int objects. (Contributed by Steve Dower in gh-111140.)

• Add PyModule_Add() function, which is similar to PyModule_AddObjectRef() and PyModule_AddObject(), but always steals a reference to the value. (Contributed by Serhiy Storchaka in gh-86493.)

• Add the PyObject_GenericHash() function that implements the default hashing function of a Python object. (Contributed by Serhiy Storchaka in gh-113024.)

• Add the Py_HashPointer() function to hash a raw pointer. (Contributed by Victor Stinner in gh-111545.)

• Add the PyObject_VisitManagedDict() and PyObject_ClearManagedDict() functions. which must be called by the traverse and clear functions of a type using the Py_TPFLAGS_MANAGED_DICT flag. The pythoncapi-compat project can be used to use these functions with Python 3.11 and 3.12. (Contributed by Victor Stinner in gh-107073.)

• Add the PyRefTracer_SetTracer() and PyRefTracer_GetTracer() functions, which enable tracking object creation and destruction in the same way that the tracemalloc module does. (Contributed by Pablo Galindo in gh-93502.)

• Add the PySys_AuditTuple() function as an alternative to PySys_Audit() that takes event arguments as a Python tuple object. (Contributed by Victor Stinner in gh-85283.)

• Add the PyThreadState_GetUnchecked() function as an alternative to PyThreadState_Get() that doesn’t kill the process with a fatal error if it is NULL. The caller is responsible for checking if the result is NULL. (Contributed by Victor Stinner in gh-108867.)

• Add the PyType_GetFullyQualifiedName() function to get the type’s fully qualified name. The module name is prepended if type.__module__ is a string and is not equal to either 'builtins' or '__main__'. (Contributed by Victor Stinner in gh-111696.)

• Add the PyType_GetModuleName() function to get the type’s module name. This is equivalent to getting the type.__module__ attribute. (Contributed by Eric Snow and Victor Stinner in gh-111696.)

• Add the PyUnicode_EqualToUTF8AndSize() and PyUnicode_EqualToUTF8() functions to compare a Unicode object with a const char* UTF-8 encoded string and 1 if they are equal or 0 otherwise. These functions do not raise exceptions. (Contributed by Serhiy Storchaka in gh-110289.)

• Add the PyWeakref_GetRef() function as an alternative to PyWeakref_GetObject() that returns a strong reference or NULL if the referent is no longer live. (Contributed by Victor Stinner in gh-105927.)

• Add fixed variants of functions which silently ignore errors:

  • PyObject_HasAttrWithError() replaces PyObject_HasAttr().

  • PyObject_HasAttrStringWithError() replaces PyObject_HasAttrString().

  • PyMapping_HasKeyWithError() replaces PyMapping_HasKey().

  • PyMapping_HasKeyStringWithError() replaces PyMapping_HasKeyString().

  The new functions return -1 for errors and the standard 1 for true and 0 for false.

  (Contributed by Serhiy Storchaka in gh-108511.)

Changed C APIs¶

• The keywords parameter of PyArg_ParseTupleAndKeywords() and PyArg_VaParseTupleAndKeywords() now has type char *const* in C and const char *const* in C++, instead of char**. In C++, this makes these functions compatible with arguments of type const char *const*, const char**, or char *const* without an explicit type cast. In C, the functions only support arguments of type char *const*. This can be overridden with the PY_CXX_CONST macro. (Contributed by Serhiy Storchaka in gh-65210.)

• PyArg_ParseTupleAndKeywords() now supports non-ASCII keyword parameter names. (Contributed by Serhiy Storchaka in gh-110815.)

• The PyCode_GetFirstFree() function is now unstable API and is now named PyUnstable_Code_GetFirstFree(). (Contributed by Bogdan Romanyuk in gh-115781.)

• The PyDict_GetItem(), PyDict_GetItemString(), PyMapping_HasKey(), PyMapping_HasKeyString(), PyObject_HasAttr(), PyObject_HasAttrString(), and PySys_GetObject() functions, each of which clears all errors which occurred when calling them now reports these errors using sys.unraisablehook(). You may replace them with other functions as recommended in the documentation. (Contributed by Serhiy Storchaka in gh-106672.)

• Add support for the %T, %#T, %N and %#N formats to PyUnicode_FromFormat():

  • %T: Get the fully qualified name of an object type

  • %#T: As above, but use a colon as the separator

  • %N: Get the fully qualified name of a type

  • %#N: As above, but use a colon as the separator

  See PEP 737 for more information. (Contributed by Victor Stinner in gh-111696.)

• You no longer have to define the PY_SSIZE_T_CLEAN macro before including Python.h when using # formats in format codes. APIs accepting the format codes always use Py_ssize_t for # formats. (Contributed by Inada Naoki in gh-104922.)

• If Python is built in debug mode or with assertions, PyTuple_SET_ITEM() and PyList_SET_ITEM() now check the index argument with an assertion. (Contributed by Victor Stinner in gh-106168.)

Limited C API Changes¶

• The following functions are now included in the Limited C API:

  • PyMem_RawMalloc()

  • PyMem_RawCalloc()

  • PyMem_RawRealloc()

  • PyMem_RawFree()

  • PySys_Audit()

  • PySys_AuditTuple()

  • PyType_GetModuleByDef()

  (Contributed by Victor Stinner in gh-85283, gh-85283, and gh-116936.)

• Python built with --with-trace-refs (tracing references) now supports the Limited API. (Contributed by Victor Stinner in gh-108634.)

Removed C APIs¶

• Remove several functions, macros, variables, etc with names prefixed by _Py or _PY (which are considered private). If your project is affected by one of these removals and you believe that the removed API should remain available, please open a new issue to request a public C API and add cc: @vstinner to the issue to notify Victor Stinner. (Contributed by Victor Stinner in gh-106320.)

• Remove old buffer protocols deprecated in Python 3.0. Use 버퍼 프로토콜 instead.

  • PyObject_CheckReadBuffer(): Use PyObject_CheckBuffer() to test whether the object supports the buffer protocol. Note that PyObject_CheckBuffer() doesn’t guarantee that PyObject_GetBuffer() will succeed. To test if the object is actually readable, see the next example of PyObject_GetBuffer().

  • PyObject_AsCharBuffer(), PyObject_AsReadBuffer(): Use PyObject_GetBuffer() and PyBuffer_Release() instead:

    Py_buffer view;
    if (PyObject_GetBuffer(obj, &view, PyBUF_SIMPLE) < 0) {
        return NULL;
    }
    // Use `view.buf` and `view.len` to read from the buffer.
    // You may need to cast buf as `(const char*)view.buf`.
    PyBuffer_Release(&view);
    

  • PyObject_AsWriteBuffer(): Use PyObject_GetBuffer() and PyBuffer_Release() instead:

    Py_buffer view;
    if (PyObject_GetBuffer(obj, &view, PyBUF_WRITABLE) < 0) {
        return NULL;
    }
    // Use `view.buf` and `view.len` to write to the buffer.
    PyBuffer_Release(&view);
    

  (Contributed by Inada Naoki in gh-85275.)

• Remove various functions deprecated in Python 3.9:

  • PyEval_CallObject(), PyEval_CallObjectWithKeywords(): Use PyObject_CallNoArgs() or PyObject_Call() instead.

    경고

    In PyObject_Call(), positional arguments must be a tuple and must not be NULL, and keyword arguments must be a dict or NULL, whereas the removed functions checked argument types and accepted NULL positional and keyword arguments. To replace PyEval_CallObjectWithKeywords(func, NULL, kwargs) with PyObject_Call(), pass an empty tuple as positional arguments using PyTuple_New(0).

  • PyEval_CallFunction(): Use PyObject_CallFunction() instead.

  • PyEval_CallMethod(): Use PyObject_CallMethod() instead.

  • PyCFunction_Call(): Use PyObject_Call() instead.

  (Contributed by Victor Stinner in gh-105107.)

• Remove the following old functions to configure the Python initialization, deprecated in Python 3.11:

  • PySys_AddWarnOptionUnicode(): Use PyConfig.warnoptions instead.

  • PySys_AddWarnOption(): Use PyConfig.warnoptions instead.

  • PySys_AddXOption(): Use PyConfig.xoptions instead.

  • PySys_HasWarnOptions(): Use PyConfig.xoptions instead.

  • PySys_SetPath(): Set PyConfig.module_search_paths instead.

  • Py_SetPath(): Set PyConfig.module_search_paths instead.

  • Py_SetStandardStreamEncoding(): Set PyConfig.stdio_encoding instead, and set also maybe PyConfig.legacy_windows_stdio (on Windows).

  • _Py_SetProgramFullPath(): Set PyConfig.executable instead.

  Use the new PyConfig API of the Python Initialization Configuration instead (PEP 587), added to Python 3.8. (Contributed by Victor Stinner in gh-105145.)

• Remove PyEval_AcquireLock() and PyEval_ReleaseLock() functions, deprecated in Python 3.2. They didn’t update the current thread state. They can be replaced with:

  • PyEval_SaveThread() and PyEval_RestoreThread();

  • low-level PyEval_AcquireThread() and PyEval_RestoreThread();

  • or PyGILState_Ensure() and PyGILState_Release().

  (Contributed by Victor Stinner in gh-105182.)

• Remove the PyEval_ThreadsInitialized() function, deprecated in Python 3.9. Since Python 3.7, Py_Initialize() always creates the GIL: calling PyEval_InitThreads() does nothing and PyEval_ThreadsInitialized() always returns non-zero. (Contributed by Victor Stinner in gh-105182.)

• Remove the _PyInterpreterState_Get() alias to PyInterpreterState_Get() which was kept for backward compatibility with Python 3.8. The pythoncapi-compat project can be used to get PyInterpreterState_Get() on Python 3.8 and older. (Contributed by Victor Stinner in gh-106320.)

• Remove the private _PyObject_FastCall() function: use PyObject_Vectorcall() which is available since Python 3.8 (PEP 590). (Contributed by Victor Stinner in gh-106023.)

• Remove the cpython/pytime.h header file, which only contained private functions. (Contributed by Victor Stinner in gh-106316.)

• Remove the undocumented PY_TIMEOUT_MAX constant from the limited C API. (Contributed by Victor Stinner in gh-110014.)

• Remove the old trashcan macros Py_TRASHCAN_SAFE_BEGIN and Py_TRASHCAN_SAFE_END. Replace both with the new macros Py_TRASHCAN_BEGIN and Py_TRASHCAN_END. (Contributed by Irit Katriel in gh-105111.)

Deprecated C APIs¶

• Deprecate old Python initialization functions:

  • PySys_ResetWarnOptions(): Clear sys.warnoptions and warnings.filters instead.

  • Py_GetExecPrefix(): Get sys.exec_prefix instead.

  • Py_GetPath(): Get sys.path instead.

  • Py_GetPrefix(): Get sys.prefix instead.

  • Py_GetProgramFullPath(): Get sys.executable instead.

  • Py_GetProgramName(): Get sys.executable instead.

  • Py_GetPythonHome(): Get PyConfig.home or the PYTHONHOME environment variable instead.

  (Contributed by Victor Stinner in gh-105145.)

• Soft deprecate the PyEval_GetBuiltins(), PyEval_GetGlobals(), and PyEval_GetLocals() functions, which return a borrowed reference. (Soft deprecated as part of PEP 667.)

• Deprecate the PyImport_ImportModuleNoBlock() function, which is just an alias to PyImport_ImportModule() since Python 3.3. (Contributed by Victor Stinner in gh-105396.)

• Soft deprecate the PyModule_AddObject() function. It should be replaced with PyModule_Add() or PyModule_AddObjectRef(). (Contributed by Serhiy Storchaka in gh-86493.)

• Deprecate the old Py_UNICODE and PY_UNICODE_TYPE types and the Py_UNICODE_WIDE define. Use the wchar_t type directly instead. Since Python 3.3, Py_UNICODE and PY_UNICODE_TYPE are just aliases to wchar_t. (Contributed by Victor Stinner in gh-105156.)

• Deprecate the PyWeakref_GetObject() and PyWeakref_GET_OBJECT() functions, which return a borrowed reference. Replace them with the new PyWeakref_GetRef() function, which returns a strong reference. The pythoncapi-compat project can be used to get PyWeakref_GetRef() on Python 3.12 and older. (Contributed by Victor Stinner in gh-105927.)

Changes in the Python API¶

• PEP 667 introduces several changes to the semantics of locals() and f_locals:

  • Calling locals() in an optimized scope now produces an independent snapshot on each call, and hence no longer implicitly updates previously returned references. Obtaining the legacy CPython behavior now requires explicit calls to update the initially returned dictionary with the results of subsequent calls to locals(). Code execution functions that implicitly target locals() (such as exec and eval) must be passed an explicit namespace to access their results in an optimized scope. (Changed as part of PEP 667.)

  • Calling locals() from a comprehension at module or class scope (including via exec or eval) once more behaves as if the comprehension were running as an independent nested function (i.e. the local variables from the containing scope are not included). In Python 3.12, this had changed to include the local variables from the containing scope when implementing PEP 709. (Changed as part of PEP 667.)

  • Accessing FrameType.f_locals in an optimized scope now returns a write-through proxy rather than a snapshot that gets updated at ill-specified times. If a snapshot is desired, it must be created explicitly with dict or the proxy’s .copy() method. (Changed as part of PEP 667.)

• functools.partial now emits a FutureWarning when used as a method. The behavior will change in future Python versions. Wrap it in staticmethod() if you want to preserve the old behavior. (Contributed by Serhiy Storchaka in gh-121027.)

• An OSError is now raised by getpass.getuser() for any failure to retrieve a username, instead of ImportError on non-Unix platforms or KeyError on Unix platforms where the password database is empty.

• The value of the mode attribute of gzip.GzipFile is now a string ('rb' or 'wb') instead of an integer (1 or 2). The value of the mode attribute of the readable file-like object returned by zipfile.ZipFile.open() is now 'rb' instead of 'r'. (Contributed by Serhiy Storchaka in gh-115961.)

• mailbox.Maildir now ignores files with a leading dot (.). (Contributed by Zackery Spytz in gh-65559.)

• pathlib.Path.glob() and rglob() now return both files and directories if a pattern that ends with “**” is given, rather than directories only. Add a trailing slash to keep the previous behavior and only match directories.

• The threading module now expects the _thread module to have an _is_main_interpreter() function. This function takes no arguments and returns True if the current interpreter is the main interpreter.

  Any library or application that provides a custom _thread module must provide _is_main_interpreter(), just like the module’s other “private” attributes. (gh-112826.)

Changes in the C API¶

• Python.h no longer includes the <ieeefp.h> standard header. It was included for the finite() function which is now provided by the <math.h> header. It should now be included explicitly if needed. Remove also the HAVE_IEEEFP_H macro. (Contributed by Victor Stinner in gh-108765.)

• Python.h no longer includes these standard header files: <time.h>, <sys/select.h> and <sys/time.h>. If needed, they should now be included explicitly. For example, <time.h> provides the clock() and gmtime() functions, <sys/select.h> provides the select() function, and <sys/time.h> provides the futimes(), gettimeofday() and setitimer() functions. (Contributed by Victor Stinner in gh-108765.)

• On Windows, Python.h no longer includes the <stddef.h> standard header file. If needed, it should now be included explicitly. For example, it provides offsetof() function, and size_t and ptrdiff_t types. Including <stddef.h> explicitly was already needed by all other platforms, the HAVE_STDDEF_H macro is only defined on Windows. (Contributed by Victor Stinner in gh-108765.)

• If the Py_LIMITED_API macro is defined, Py_BUILD_CORE, Py_BUILD_CORE_BUILTIN and Py_BUILD_CORE_MODULE macros are now undefined by <Python.h>. (Contributed by Victor Stinner in gh-85283.)

• The old trashcan macros Py_TRASHCAN_SAFE_BEGIN and Py_TRASHCAN_SAFE_END were removed. They should be replaced by the new macros Py_TRASHCAN_BEGIN and Py_TRASHCAN_END.

  A tp_dealloc function that has the old macros, such as:

  static void
  mytype_dealloc(mytype *p)
  {
      PyObject_GC_UnTrack(p);
      Py_TRASHCAN_SAFE_BEGIN(p);
      ...
      Py_TRASHCAN_SAFE_END
  }
  

  should migrate to the new macros as follows:

  static void
  mytype_dealloc(mytype *p)
  {
      PyObject_GC_UnTrack(p);
      Py_TRASHCAN_BEGIN(p, mytype_dealloc)
      ...
      Py_TRASHCAN_END
  }
  

  Note that Py_TRASHCAN_BEGIN has a second argument which should be the deallocation function it is in. The new macros were added in Python 3.8 and the old macros were deprecated in Python 3.11. (Contributed by Irit Katriel in gh-105111.)

• PEP 667 introduces several changes to frame-related functions:

  • The effects of mutating the dictionary returned from PyEval_GetLocals() in an optimized scope have changed. New dict entries added this way will now only be visible to subsequent PyEval_GetLocals() calls in that frame, as PyFrame_GetLocals(), locals(), and FrameType.f_locals no longer access the same underlying cached dictionary. Changes made to entries for actual variable names and names added via the write-through proxy interfaces will be overwritten on subsequent calls to PyEval_GetLocals() in that frame. The recommended code update depends on how the function was being used, so refer to the deprecation notice on the function for details.

  • Calling PyFrame_GetLocals() in an optimized scope now returns a write-through proxy rather than a snapshot that gets updated at ill-specified times. If a snapshot is desired, it must be created explicitly (e.g. with PyDict_Copy()), or by calling the new PyEval_GetFrameLocals() API.

  • PyFrame_FastToLocals() and PyFrame_FastToLocalsWithError() no longer have any effect. Calling these functions has been redundant since Python 3.11, when PyFrame_GetLocals() was first introduced.

  • PyFrame_LocalsToFast() no longer has any effect. Calling this function is redundant now that PyFrame_GetLocals() returns a write-through proxy for optimized scopes.

• Python 3.13 removed many private functions. Some of them can be replaced using these alternatives:

  • _PyDict_Pop(): PyDict_Pop() or PyDict_PopString();

  • _PyDict_GetItemWithError(): PyDict_GetItemRef();

  • _PyErr_WriteUnraisableMsg(): PyErr_FormatUnraisable();

  • _PyEval_SetTrace(): PyEval_SetTrace() or PyEval_SetTraceAllThreads();

  • _PyList_Extend(): PyList_Extend();

  • _PyLong_AsInt(): PyLong_AsInt();

  • _PyMem_RawStrdup(): strdup();

  • _PyMem_Strdup(): strdup();

  • _PyObject_ClearManagedDict(): PyObject_ClearManagedDict();

  • _PyObject_VisitManagedDict(): PyObject_VisitManagedDict();

  • _PyThreadState_UncheckedGet(): PyThreadState_GetUnchecked();

  • _PyTime_AsSecondsDouble(): PyTime_AsSecondsDouble();

  • _PyTime_GetMonotonicClock(): PyTime_Monotonic() or PyTime_MonotonicRaw();

  • _PyTime_GetPerfCounter(): PyTime_PerfCounter() or PyTime_PerfCounterRaw();

  • _PyTime_GetSystemClock(): PyTime_Time() or PyTime_TimeRaw();

  • _PyTime_MAX: PyTime_MAX;

  • _PyTime_MIN: PyTime_MIN;

  • _PyTime_t: PyTime_t;

  • _Py_HashPointer(): Py_HashPointer();

  • _Py_IsFinalizing(): Py_IsFinalizing().

  The pythoncapi-compat project can be used to get most of these new functions on Python 3.12 and older.