类型提示、链式赋值和多重赋值

Question

我猜这两个问题是相关的，所以我会一起发表：

1. - 是否可以在链式赋值中使用类型提示？

这两个尝试失败了：

>>> def foo(a:int):
...     b: int = c:int = a
  File "<stdin>", line 2
    b: int = c:int = a
              ^
SyntaxError: invalid syntax
>>> def foo(a:int):
...     b = c:int = a
  File "<stdin>", line 2
    b = c:int = a
         ^
SyntaxError: invalid syntax

2. - 是否可以在多个赋值中使用类型提示？

这是我的几次尝试：

>>> from typing import Tuple
>>> def bar(a: Tuple[int]):
...     b: int, c:int = a
  File "<stdin>", line 2
    b: int, c:int = a
          ^
SyntaxError: invalid syntax
>>> def bar(a: Tuple[int]):
...     b, c:Tuple[int] = a
... 
  File "<stdin>", line 2
SyntaxError: only single target (not tuple) can be annotated
>>> def bar(a: Tuple[int]):
...     b, c:int = a
... 
  File "<stdin>", line 2
SyntaxError: only single target (not tuple) can be annotated

我知道在这两种情况下类型都是从a的类型提示中推断出来的，但是我有一个很长的变量列表（在类的__init__中），我想要更加明确。

我正在使用Python 3.6.8。

- steffen

Answer 1

1. As explicitly stated in PEP 526, section "Rejected/postponed proposals", annotations in chained assignments are not supported. Quoting the PEP:

   This has problems of ambiguity and readability similar to tuple unpacking, for example in:
   x: int = y = 1
z = w: int = 1
   it is ambiguous, what should the types of y and z be? Also the second line is difficult to parse.

2. For unpacking, per the same PEP, you should place bare annotations for your variables before the assignment. Example from the PEP:

   # Tuple unpacking with variable annotation syntax
   header: str
   kind: int
   body: Optional[List[str]]
   header, kind, body = message
   

- Mikhail Burshteyn