Type is not infered when iterating over `df.columns`

[alkatar21]

I don't know if currently only pyright: basic support is targeted? If not it would be nice if this could be infered, since as far as I know it should always be str.

# pyright: strict
import pandas as pd

df = pd.DataFrame()

for name in df.columns:  # Type of "name" is unknown Pylance(reportUnknownVariableType)
    print(name)

[Dr-Irv]

The type of df.columns is Index. An Index can contain any hashable type, not just strings. For example, here is where column names could be integers and strings. Also, column "names" could be a MultiIndex.

>>> import pandas as pd
>>> df=pd.DataFrame({1: [10,20], "abc": [5, 6]})
>>> df
    1  abc
0  10    5
1  20    6
>>> df.columns
Index([1, 'abc'], dtype='object')

From a static typing perspective, there is no way to know the type inside of an Index. It is known at runtime via Index.dtype.

[twoertwein]

Could silence this pyright error by changing this line

pandas-stubs/pandas-stubs/core/indexes/base.pyi

Line 209 in f2b783c

def __iter__(self) -> Iterator: ...

to def __iter__(self) -> Iterator[Hashable]: ... but hashable might not be very helpful.

[Dr-Irv]

Good idea @twoertwein . I did this in #35