BUG: groupby.first() with None values fills None value with next row

[sehHeiden]

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

python
import geopandas as gpd
from shapely.geometry import Polygon, MultiPolygon

gdf = pgd.Geopandas([{"ID1": "3991b7ab", "ID2": nan,"geometry": POLYGON ((...)), "area_m": 10720.28326}, 
{"ID1": "3991b7ab","ID2": "dc4772ed", "geometry": MULTIPOLYGON(((...)), "area": 0.24245}]

gdf.assign(area=lambda x: x.geometry.area).sort_values(by="area", ascending=False).query("FLUR_OBJECT_UUI == '3991b7ab'").groupby(by="ID1").first()

Issue Description

I have a (Geo)DataFrame with Multiple UUID columns.
When sorting the rows, the first columns, can have a None value. After groupby if ID1, the None value of ID2 is replaced withe next value in that column.

Expected Behavior

The value of ID2 should be None.

Installed Versions

NSTALLED VERSIONS

commit : d9cdd2e
python : 3.11.8.final.0
python-bits : 64
OS : Windows
OS-release : 10
Version : 10.0.22631
machine : AMD64
processor : Intel64 Family 6 Model 186 Stepping 2, GenuineIntel
byteorder : little
LC_ALL : None
LANG : None
LOCALE : de_DE.cp1252
pandas : 2.2.2
numpy : 1.26.3
pytz : 2024.1
dateutil : 2.9.0.post0
setuptools : 69.1.0
pip : 24.0
Cython : None
pytest : 8.2.0
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : 3.1.3
IPython : 8.23.0
pandas_datareader : None
adbc-driver-postgresql: None
adbc-driver-sqlite : None
bs4 : 4.12.3
bottleneck : None
dataframe-api-compat : None
fastparquet : None
fsspec : 2024.3.1
gcsfs : None
matplotlib : None
numba : 0.59.0
numexpr : None
odfpy : None
openpyxl : 3.1.2
pandas_gbq : None
pyarrow : None
pyreadstat : None
python-calamine : None
pyxlsb : None
s3fs : None
scipy : 1.13.0
sqlalchemy : 2.0.30
tables : None
tabulate : None
xarray : 2024.1.1
xlrd : None
zstandard : None
tzdata : 2024.1
qtpy : None
pyqt5 : None

[rhshadrach]

Thanks for the report. Can you simplify this? Produce the DataFrame that goes into the groupby explicitly (i.e. remove the use of assign, sort_values, and query), and show the result you get.

[asishm]

df.groupby.first has a skipna argument that defaults to True https://pandas.pydata.org/docs/reference/api/pandas.core.groupby.DataFrameGroupBy.first.html

Try setting that to False.

In [190]: pd.DataFrame({'a': [1, 1, 2, 2], 'b': [np.nan, '2', '3', '4']}).groupby('a').first(skipna=False)
Out[190]:
     b
a
1  NaN
2    3

In [191]: pd.DataFrame({'a': [1, 1, 2, 2], 'b': [np.nan, '2', '3', '4']}).groupby('a').first()
Out[191]:
   b
a
1  2
2  3

[sehHeiden]

Okay, my bad.

Made sure, that I have > 2.2.1 on all machines.
skipna, did help. Problem was, when I search for the documentation for first. I find the DataFrame.first and not the DataFrameGroupBy.first.
I think, this is a problem, but something different.

[rhshadrach]

@sehHeiden - it's not clear to me if your issue is resolved or not. If it isn't resolved, can you post a reproducible example (see my previous comment).

[sehHeiden]

Solved.
@asishm s solution works. Tested with 2.2.2, but does not work with 2.2.0 as it does not have the skipna parameter.

Optional wish: Do link GroupBy.first in the DataFrame.first documentation.