BUG: freq after sort values differs between arm Mac and ubuntu

[phofl]

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

pdf = pd.DataFrame({"z": 1})
pdf.z.sort_values()


2020-12-31    1
2021-01-01    1
2021-01-02    1
2021-01-03    1
2021-01-04    1
             ..
2021-04-05    1
2021-04-06    1
2021-04-07    1
2021-04-08    1
2021-04-09    1
Freq: D, Name: z, Length: 100, dtype: int64

Issue Description

This preserves the freq on ubuntu but loses it on Mac

Expected Behavior

don't know, consistency above everything else

I can't test the nightlies on ubuntu right now, that's why it's only 2.2.2

Installed Versions

INSTALLED VERSIONS

commit : d9cdd2e
python : 3.12.4.final.0
python-bits : 64
OS : Darwin
OS-release : 23.3.0
Version : Darwin Kernel Version 23.3.0: Wed Dec 20 21:33:31 PST 2023; root:xnu-10002.81.5~7/RELEASE_ARM64_T8112
machine : arm64
processor : arm
byteorder : little
LC_ALL : None
LANG : None
LOCALE : None.UTF-8

pandas : 2.2.2
numpy : 2.0.0
pytz : 2024.1
dateutil : 2.9.0
setuptools : 70.1.0
pip : 24.0
Cython : None
pytest : 8.2.2
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : 3.1.4
IPython : 8.25.0
pandas_datareader : None
adbc-driver-postgresql: None
adbc-driver-sqlite : None
bs4 : 4.12.3
bottleneck : None
dataframe-api-compat : None
fastparquet : None
fsspec : 2024.6.0
gcsfs : None
matplotlib : None
numba : None
numexpr : None
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : 16.1.0
pyreadstat : None
python-calamine : None
pyxlsb : None
s3fs : 2024.6.0
scipy : None
sqlalchemy : None
tables : None
tabulate : None
xarray : 2024.6.0
xlrd : None
zstandard : None
tzdata : 2024.1
qtpy : None
pyqt5 : None
None

[phofl]

cc @jbrockmendel thoughts here?

[jbrockmendel]

I don't have an ubuntu env ATM, so speculating: sort_values go through nargsort which I suspect may have undefined behavior in non-unique cases.

BTW the example in the OP is not reproducible, needs an index passed to DataFrame.

[Siddharth-Latthe-07]

@phofl pandas has historically had issues with preserving the freq attribute of DatetimeIndex after certain operations. The sort_values method should ideally not affect the freq attribute, but inconsistencies can arise due to differences in underlying implementations or versions.
u can try this code:-

import pandas as pd

# Create a DataFrame with a single column
pdf = pd.DataFrame({"z": 1}, index=pd.date_range(start='2020-12-31', periods=100, freq='D'))

# Sort the values in the column
sorted_values = pdf.z.sort_values()

# Restore the frequency attribute
sorted_values.index.freq = pdf.index.freq

# Print the sorted values
print(sorted_values)

concl:-
the behavior of pandas should ideally be consistent across platforms, there are occasional discrepancies. By explicitly managing the freq attribute, you can ensure your code behaves as expected regardless of the operating system.