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

[phofl]

Pandas version checks

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

• [X] 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 : d9cdd2ee5a58015ef6f4d15c7226110c9aab8140 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.