API: Disallow sets as index and columns argument in DataFrame constructor

[Dr-Irv]

• closes API: Should DataFrame constructor allow sets for index and columns? #47215
• pandas/tests/frame/test_constructors.py method test_columns_indexes_raise_on_sets
• All code checks passed.
• Added an entry in the latest doc/source/whatsnew/v1.5.0.rst file if fixing a bug or adding a new feature.

Note: This also allowed some typing changes removing some #type: ignore statements

[rhshadrach]

Changes look good, but I think there is still not consensus on if this should go through deprecation first: #47215 (comment)

[rhshadrach]

Why specifically Sequence[str] and not Sequence[Hashable]?

[Dr-Irv]

I've removed Dict and Sequence[str] in the next commit. Sequence[Hashable] will match a string, and that is not a valid value for index or columns

[github-actions]

This pull request is stale because it has been open for thirty days with no activity. Please update and respond to this comment if you're still interested in working on this.

[Dr-Irv]

@jreback have merged with latest master. You were last to review - suggesting using ensure_index(), but that created test failures.

[jreback]

hmm https://github.com/pandas-dev/pandas/runs/7271930917?check_suite_focus=true

i thinksomething changed on master for types......

[Dr-Irv]

hmm https://github.com/pandas-dev/pandas/runs/7271930917?check_suite_focus=true

i thinksomething changed on master for types......

Yes, there was a change made that made the _set_axis() parameter have too wide a type. It included Sequence, and I changed it to list

We have to avoid using Sequence in general, and Sequence[str] specifically because a plain string will match either.

Next commit addresses that.

[mroeschke]

I would be +0 for a deprecation, but I think this is also okay that this is mentioned in the Other API changes section. Off to @jreback

[jreback]

thanks @Dr-Irv lgtm.