BUG: Fix Index.get_level_values() mishandling of boolean, pd.NA, np.nan, and pd.NaT levels #62175
Conversation
pandas/core/indexes/base.py
Outdated
| """ | ||
| if isinstance(level, int): | ||
| if type(level) is int: |
There was a problem hiding this comment.
Just a simple edge case where the Index.name is also an int
| if type(level) is int: | |
| if type(level) is int: | |
| if isinstance(self.name, int) and level == self.name: | |
| return |
| raise KeyError( | ||
| f"Requested level ({level}) does not match index name ({self.name})" | ||
| ) | ||
|
|
There was a problem hiding this comment.
I think that this is a better verification that validates what is expected from the documentation.
| elif isinstance(level, str) and isinstance(self.name, str) and level != self.name: | |
| raise KeyError( | |
| f"Requested level ({level}) does not match index name ({self.name})" | |
| ) |
pandas/core/indexes/base.py
Outdated
| """ | ||
| if isinstance(level, int): | ||
| if type(level) is int: |
…d improve error messaging
…enhance error handling for boolean levels
|
@jbrockmendel Can you explain why these checks are constantly failing? Am I overlooking something? |
…el and index name
pandas/_libs/__init__.py
Outdated
| import pandas._libs.pandas_parser # isort: skip # type: ignore[reportUnusedImport] | ||
| import pandas._libs.pandas_datetime # noqa: F401 # isort: skip # type: ignore[reportUnusedImport] | ||
| from pandas._libs.interval import Interval | ||
| from pandas._libs.missing import NA |
There was a problem hiding this comment.
seems this change is unrelated to the rest of the PR?
There was a problem hiding this comment.
Yes, actually, I was trying something and forgot to undo this change. It's fixed in the latest commit.
pandas/core/indexes/base.py
Outdated
| if isinstance(level, int): | ||
| # Explicitly raise for missing/null values to match pandas convention | ||
| # Also reject all NA-like values (np.nan, pd.NA, pd.NaT, etc.) | ||
| if isna(level): |
There was a problem hiding this comment.
what if index.name is a na-like scalar?
pandas/core/indexes/base.py
Outdated
| ) | ||
|
|
||
| # Reject booleans unless the index name is actually a boolean and matches | ||
| if isinstance(level, bool): |
There was a problem hiding this comment.
isnt this handled in the existing elif level != self.name check?
There was a problem hiding this comment.
Makes sense... I'll keep the same.
It's really common to have level=name=None, which is getting caught by the isna(level) check you added. |
… improve error messaging
…alues and streamline error messaging
|
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. |
3 participants
levelcorrectly #62169