BUG: Let melt name multiple variable columns for labels from a MultiIndex
#58088
Conversation
|
Thanks for the PR! Sorry this hasn't been reviewed yet. |
|
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. |
|
Hi! Just checking in as this issue has been affecting my code. I wondered if anything needed to be done or if this will be merged in a later release? |
| tm.assert_frame_equal(df.melt(var_name=["first", "second"]), expected) | ||
| tm.assert_frame_equal(df.melt(var_name=["first"]), expected[["first", "value"]]) | ||
|
|
||
| with pytest.raises( |
There was a problem hiding this comment.
Could you make this a separate test?
pandas/core/reshape/melt.py
Outdated
| @@ -217,7 +218,15 @@ def melt( | |||
| frame.columns.name if frame.columns.name is not None else "variable" | |||
| ] | |||
| elif is_list_like(var_name): | |||
| raise ValueError(f"{var_name=} must be a scalar.") | |||
| if isinstance(frame.columns, MultiIndex): | |||
| var_name = list(var_name) | |||
There was a problem hiding this comment.
Is the list cast needed here? it should be Sized to compute len
There was a problem hiding this comment.
var_name will be iterated on twice, so I think there's a pathological case where you could have a single-use object that has __len__ and __iter__.
There was a problem hiding this comment.
Ah true. Could you use is_iterator to check for the single use object and cast to list if so?
doc/source/whatsnew/v2.2.2.rst
Outdated
| @@ -15,6 +15,7 @@ Fixed regressions | |||
| ~~~~~~~~~~~~~~~~~ | |||
| - :meth:`DataFrame.__dataframe__` was producing incorrect data buffers when the a column's type was a pandas nullable on with missing values (:issue:`56702`) | |||
| - :meth:`DataFrame.__dataframe__` was producing incorrect data buffers when the a column's type was a pyarrow nullable on with missing values (:issue:`57664`) | |||
| - :meth:`DataFrame.melt` would not accept multiple names in ``var_name`` when the columns were a :class:`MultiIndex` (:issue:`58033`) | |||
There was a problem hiding this comment.
Could you move this to v3.0.0rst?
mroeschke
added
Reshaping
|
Thanks @rob-sil |
|
Thanks for moving this forward! |
|
my code works in jupyter notebook with 'var_names' parameter but not in py file, it gives the same error when running in py file. Interesting. |
meltname the columns for the labels of aMultiIndex#58033doc/source/whatsnew/v2.2.2.rstfile if fixing a bug or adding a new feature.If there are fewer names provided in
var_namethan levels in the columnsMultiIndex, thenmeltwill just take the first several. I personally find this a bit strange, but it keeps with the previous behavior ofmeltwith aMultiIndexwhere you could pass it a single name and it would only use one level of theMultiIndex.