DOC: Updated set_index doc with a warning #60990
Conversation
SaraInCode
commented
Feb 23, 2025
SaraInCode
commented
- closes DOC: No warning in set_index() that previous index column is removed. #60973
- Tests added and passed if fixing a bug or adding a new feature
- All code checks passed.
- Added type annotations to new arguments/methods/functions.
pandas/core/frame.py
Outdated
| @@ -5868,6 +5868,10 @@ def set_index( | |||
| columns or arrays (of the correct length). The index can replace the | |||
| existing index or expand on it. | |||
|
|
|||
| .. warning:: | |||
There was a problem hiding this comment.
Instead, could you show this as an example in the Examples section instead?
There was a problem hiding this comment.
I'll probably be removing the warning but I could show an example
pandas/core/frame.py
Outdated
| Setting a new index will remove the current index column, | ||
| unless you first call `reset_index`. |
There was a problem hiding this comment.
This is only true when append=False. As mentioned in the issue, this seems to me to be redundant with the line above:
The index can replace the existing index or expand on it.
There was a problem hiding this comment.
Thanks for the review. Though it seems redundant with the line you mentioned. When you look at,
append: bool, default False
Whether to append columns to existing index.
it doesn't directly explain the behavior of replacing the existing index (by default). Does this statement makes it clear?
append: bool, default False
Whether to append new columns to the existing index . If set to True, the current index will remain unchanged and new columns will be added.
There was a problem hiding this comment.
Agreed we should add something here, but saying it is unchanged and columns are added seems inconsistent to me. I also think the True behavior is already clear by Whether to append new columns to the existing index. However the False behavior could perhaps use elaboration. I would suggest:
When False, the current index will be dropped from the DataFrame.
There was a problem hiding this comment.
But not strongly opposed to adding elaboration for the True case either.
pandas/core/frame.py
Outdated
|
|
||
| Append a column to the existing index: | ||
|
|
||
| >>> df.set_index("month", inplace=True) |
There was a problem hiding this comment.
We will be deprecating inplace in the near-ish future. Can you do df = df.set_index(...) instead.
pandas/core/frame.py
Outdated
|
|
||
| Append a column to the existing index: | ||
|
|
||
| >>> a = df.set_index("month") |
There was a problem hiding this comment.
I had suggested df. I would stay away from a as it has no particular meeting. Is there a reason to not use df here?
There was a problem hiding this comment.
I thought not to disturb the original dataframe.
There was a problem hiding this comment.
Nevertheless, I've updated to df.
|
Thanks @SaraInCode |
