ENH: Raise better error message when mapping categorical column with a `dict[str, list]` fails when all categories are mapped

[radusuciu]

Pandas version checks

• I have checked that this issue has not already been reported.

• I have confirmed this bug exists on the latest version of pandas.

• I have confirmed this bug exists on the main branch of pandas.

Reproducible Example

import pandas as pd


s = pd.Series(["foo", "bar"]).astype("category")

map_all = {"foo": [1, 2], "bar": [3, 4]}
map_partial = {"foo": [1, 2]}

# mapping all categories fails
try:
    s.map(map_all)
except Exception as e:
    print("Mapping all categories in categorical:", type(e).__name__, ":", e)

# mapping partial categories works
print("Mapping partial categories in categorical:")
print(s.map(map_partial), sep="\n")

# workaround: convert column to object
print("Mapping all categories in object:")
print(s.astype('object').map(map_all))

# Output:
# Mapping all categories in categorical: TypeError : unhashable type: 'list'
# Mapping partial categories in categorical:
# 0    [1, 2]
# 1       NaN
# dtype: object
# Mapping all categories in object:
# 0    [1, 2]
# 1    [3, 4]
# dtype: object

Issue Description

When Series.map is called on a categorical column and provided with a mapping that maps all categories to a list (or other unhashable object), a TypeError : unhashable type: 'list' is raised. However, if mapping only some of the categories, or if first converting to an object column, the operation succeeds.

Expected Behavior

map on a categorical column should succeed when a full mapping is provided, with the same results as when map is called on an object column:

import pandas as pd

s = pd.Series(["foo", "bar"]).astype("category")
expected_output = s.map({"foo": [1, 2], "bar": [3, 4]})
print(expected_output)

# Expected output:
# 0    [1, 2]
# 1    [3, 4]
# dtype: object

Installed Versions

INSTALLED VERSIONS

commit : 0f43794
python : 3.11.4.final.0
python-bits : 64
OS : Linux
OS-release : 5.15.90.1-microsoft-standard-WSL2
Version : #1 SMP Fri Jan 27 02:56:13 UTC 2023
machine : x86_64
processor :
byteorder : little
LC_ALL : en_US.UTF-8
LANG : en_US.UTF-8
LOCALE : en_US.UTF-8

pandas : 2.0.3
numpy : 1.25.1
pytz : 2023.3
dateutil : 2.8.2
setuptools : 68.0.0
pip : 23.2.1
Cython : None
pytest : 7.4.0
hypothesis : None
sphinx : 7.1.1
blosc : None
feather : None
xlsxwriter : None
lxml.etree : 4.9.3
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : 3.1.2
IPython : 8.14.0
pandas_datareader: None
bs4 : 4.12.2
bottleneck : None
brotli : None
fastparquet : None
fsspec : None
gcsfs : None
matplotlib : 3.7.2
numba : None
numexpr : None
odfpy : None
openpyxl : 3.1.2
pandas_gbq : None
pyarrow : None
pyreadstat : None
pyxlsb : None
s3fs : None
scipy : 1.11.1
snappy : None
sqlalchemy : None
tables : None
tabulate : 0.9.0
xarray : None
xlrd : None
zstandard : None
tzdata : 2023.3
qtpy : None
pyqt5 : None

[JostBrand]

The lists would become your new categories. Pandas tries to hash the categories for faster lookups. But lists are not a hashable data type. You can observe this if you try to construct your desired result:

s = pd.Series([[1, 2],[3, 4]]).astype("category")

TypeError: unhashable type: 'list'

[radusuciu]

I understand, though I think it's potentially confusing that the partial map works, whereas the full map does not. I'm guessing that with the partial map the new column is implicitly cast to an object type column -- could the same behavior be applied to the full map in cases where any of the mapped values are unhashable?

[n-splv]

+1
The error should be more informative