BUG: DataFrame.agg() on empty dataframe returns unexpected result

[kdebrab]

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

index_empty = pd.MultiIndex.from_product([[], []], names=["one", "two"])
df_empty = pd.DataFrame({"a": [], "b": []}, index=index_empty)
df_empty.agg(lambda row: row.name[0], axis=1)

Issue Description

The above returns:

Empty DataFrame
Columns: [a, b]
Index: []

Expected Behavior

The expected result would be an empty series:

Series([], dtype: float64)

Installed Versions

INSTALLED VERSIONS

commit : 0691c5c
python : 3.11.3
python-bits : 64
OS : Windows
OS-release : 10
Version : 10.0.22631
machine : AMD64
processor : Intel64 Family 6 Model 140 Stepping 1, GenuineIntel
byteorder : little
LC_ALL : None
LANG : None
LOCALE : English_Belgium.1252
pandas : 2.2.3
numpy : 2.1.3
pytz : 2024.2
dateutil : 2.9.0.post0
pip : 24.2
Cython : None
sphinx : None
IPython : 8.28.0
adbc-driver-postgresql: None
adbc-driver-sqlite : None
bs4 : None
blosc : None
bottleneck : 1.4.2
dataframe-api-compat : None
fastparquet : None
fsspec : 2024.10.0
html5lib : None
hypothesis : None
gcsfs : None
jinja2 : 3.1.4
lxml.etree : 5.3.0
matplotlib : 3.8.4
numba : None
numexpr : 2.10.2
odfpy : None
openpyxl : 3.1.5
pandas_gbq : None
psycopg2 : None
pymysql : 1.4.6
pyarrow : 18.0.0
pyreadstat : None
pytest : 8.3.3
python-calamine : None
pyxlsb : 1.0.10
s3fs : None
scipy : 1.14.1
sqlalchemy : 2.0.36
tables : 3.10.1
tabulate : 0.9.0
xarray : 2024.10.0
xlrd : None
xlsxwriter : 3.2.0
zstandard : None
tzdata : 2024.2
qtpy : None
pyqt5 : None

[rhshadrach]

Thanks for the report. Closing as a duplicate of #47959.

[kdebrab]

@rhshadrach Thanks for linking with #47959.

I would like to add a bit more context.

Note that e.g. following does work as expected on an empty dataframe:

df_empty.agg(lambda row: row.name, axis=1)
Out[3]: Series([], dtype: float64)

It's only when adding a subscript to row.name (which should be fine as the index is a multiindex), we suddenly get:

df_empty.agg(lambda row: row.name[0], axis=1)
Out[4]: 
Empty DataFrame
Columns: [a, b]
Index: []

[rhshadrach]

@kdebrab - this is because agg will attempt to pass an empty Series, using the result to determine what it should return. By adding [0], this attempt now raises an error, meaning that agg has no information whatsoever on what the UDF you're providing does. As such, it just returns the input as-is.