python / python/cpython

`__future__` annotations breaks `TypedDict` `__required/optional_keys__`

未关闭
#97,727 7 条评论 1 个 reaction 已指派 0 人 在 GitHub 查看

还没有人认领这个 Issue。

stdlib topic-typing type-bug
主要语言
Python
星标
77.2k
派生
36k
平均合并
1 天 9 小时
30 天内合并 PR
558

描述

Bug report

from __future__ import annotations appears to break TypedDict required/optional keys ending up in __required_keys__ and __optional_keys__. mypy works as expected though.

Using the example from https://peps.python.org/pep-0655/#usage-in-python-3-11 as the base

$ cat t.py
from __future__ import annotations

from typing_extensions import NotRequired, TypedDict

class Dog(TypedDict):
    name: str
    owner: NotRequired[str]

print("required", Dog.__required_keys__)
print("optional", Dog.__optional_keys__)
$ python3 t.py
required frozenset({'name', 'owner'})
optional frozenset()

With the __future__ import removed, works as expected:

$ cat t.py
from typing_extensions import NotRequired, TypedDict

class Dog(TypedDict):
    name: str
    owner: NotRequired[str]

print("required", Dog.__required_keys__)
print("optional", Dog.__optional_keys__)
$ python3 t.py
required frozenset({'name'})
optional frozenset({'owner'})

Note: breaks across different variations of total and Required/NotRequired and typing_extensions vs typing imports, above is just one example.

https://peps.python.org/pep-0655/#how-to-teach-this contains an example with the __future__ annotations import in place with no mention that it would not cause __required_keys__ and __optional_keys__ becoming populated as expected, so I'm assuming this is a bug.

Your environment

  • CPython versions tested on:

    • 3.9.7 + NotRequired and TypedDict imports from typing_extensions
    • 3.10.7 + above mentioned imports from typing_extensions
    • current 3.11.0rc2+ + above mentioned imports from typing_extensions
    • current 3.11.0rc2+ above mentioned imports from typing
  • Operating system and architecture: Linux x86_64

贡献指南

打开贡献指南

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

调研方向

首先,在列出的 Python 版本中分别使用和不使用 from __future__ import annotations 运行 t.py 复现,并分别使用 typingtyping_extensions。跟踪 TypedDictRequiredNotRequired 的入口点,以确定延迟注解如何影响键的分类。对于报告的各个变体,当 __required_keys____optional_keys__ 与声明的注解一致时,即表示完成。

由索引模型根据 Issue 内容生成。

评估

技术栈
python
领域
compilers
Issue 类型
缺陷
难度
4/5
预计耗时
3-5 天
活跃度
停滞
描述清晰度
描述清楚
新手友好度
42/100

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。