生成类似于.pyi文件的visual studio代码

Question

在我的示例中，visual studio代码中的.pyi文件(python扩展名)中的_csv.pyi文件提供了更多关于类型提示的详细信息，与mypy生成的.pyi文件相比

例如，在下面的代码中，您可以看到两个.pyi文件

visual代码_csv.pyi：

from typing import Any, Iterable, Iterator, Protocol, Union
from typing_extensions import Literal

__version__: str

QUOTE_ALL: Literal[1]
QUOTE_MINIMAL: Literal[0]
QUOTE_NONE: Literal[3]
QUOTE_NONNUMERIC: Literal[2]

class Error(Exception): ...

class Dialect:
    delimiter: str
    quotechar: str | None
    escapechar: str | None
    doublequote: bool
    skipinitialspace: bool
    lineterminator: str
    quoting: int
    strict: int
    def __init__(self) -> None: ...

_DialectLike = Union[str, Dialect, type[Dialect]]

class _reader(Iterator[list[str]]):
    dialect: Dialect
    line_num: int
    def __next__(self) -> list[str]: ...

class _writer:
    dialect: Dialect
    def writerow(self, row: Iterable[Any]) -> Any: ...
    def writerows(self, rows: Iterable[Iterable[Any]]) -> None: ...

class _Writer(Protocol):
    def write(self, __s: str) -> object: ...

def writer(csvfile: _Writer, dialect: _DialectLike = ..., **fmtparams: Any) -> _writer: ...
def reader(csvfile: Iterable[str], dialect: _DialectLike = ..., **fmtparams: Any) -> _reader: ...
def register_dialect(name: str, dialect: Any = ..., **fmtparams: Any) -> None: ...
def unregister_dialect(name: str) -> None: ...
def get_dialect(name: str) -> Dialect: ...
def list_dialects() -> list[str]: ...
def field_size_limit(new_limit: int = ...) -> int: ...

类型顽固的_csv.pyi：

from typing import Any, ClassVar

QUOTE_ALL: int
QUOTE_MINIMAL: int
QUOTE_NONE: int
QUOTE_NONNUMERIC: int
_dialects: dict

class Dialect:
    delimiter: ClassVar[getset_descriptor] = ...
    doublequote: ClassVar[member_descriptor] = ...
    escapechar: ClassVar[getset_descriptor] = ...
    lineterminator: ClassVar[getset_descriptor] = ...
    quotechar: ClassVar[getset_descriptor] = ...
    quoting: ClassVar[getset_descriptor] = ...
    skipinitialspace: ClassVar[member_descriptor] = ...
    strict: ClassVar[member_descriptor] = ...
    @classmethod
    def __init__(cls, *args, **kwargs) -> None: ...

class Error(Exception): ...

def field_size_limit(*args, **kwargs) -> Any: ...
def get_dialect(name) -> Any: ...
def list_dialects() -> Any: ...
def reader(*args, **kwargs) -> Any: ...
def register_dialect(*args, **kwargs) -> Any: ...
def unregister_dialect(name) -> Any: ...
def writer(*args, **kwargs) -> Any: ...

正如你所看到的，visual代码有一个比mypy强的更精确的类型提示。

，为什么是这种情况？用于生成这些.pyi文件的哪个工具visual studio代码(python扩展精确)？

Answer 1

_cvs.pyi是与mypy一起安装的typeshed的一部分。你可以在这里看到它：https://github.com/python/typeshed/blob/master/stdlib/_csv.pyi