pd.concat with None

[amgcc]

Describe the bug
concat with None values are valid use but pandas-stubs rejects.

To Reproduce

1. Provide a minimal runnable pandas example that is not properly checked by the stubs.

def x() -> pd.DataFrame:
    df1: pd.DataFrame | None = None
    df2 = pd.DataFrame()
    return pd.concat([df1, df2])

2. Using mypy
3. Error
   List item 0 has incompatible type "DataFrame | None"; expected "DataFrame" [list-item]

Please complete the following information:

• OS: Fedora
• OS Version 7.8
• python version 3.11.7
• version of type checker 1.8.0
• version of installed pandas-stubs 2.1.1

[Dr-Irv]

Thanks for the report. Have to change the Iterable declarations for concat in pandas-stubs/core/reshape/concat.pyi to include None

[amgcc]

I tried to push the fix for this but its rejected. Do you need special permissions to submit PRs to this project? I have never tried submitting changes to a public project before.

remote: Permission to pandas-dev/pandas-stubs.git denied to amgcc.
fatal: unable to access 'https://github.com/pandas-dev/pandas-stubs.git/': The requested URL returned error: 403

[bashtage]

You need to push to your own branch. You should first fork this repo, then clone your fork, make changes, and push your branch to your fork. You can then make a PR from your fork to this repo.

[amgcc]

You need to push to your own branch. You should first fork this repo, then clone your fork, make changes, and push your branch to your fork. You can then make a PR from your fork to this repo.

I tried to clone the repo and create a branch. I'll try this, thanks for the tip.

[amgcc]

Can you tell me if this PR is what you had in mind? If so I can attempt to write a test for it.

[Dr-Irv]

Can you tell me if this PR is what you had in mind? If so I can attempt to write a test for it.

Yes, that's the right change. You just have to add a test.