BUG: freq information lost after stack() is called

[MMCMA]

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
idx = pd.date_range('20200101', periods=12, freq='BM')
df = pd.DataFrame(dict(a=range(12)), index=idx)
print(df.index.freq)
>> <BusinessMonthEnd>
df.stack()
print(df.index.freq()
>> None

Issue Description

Some of our CI pipeline unittests failed after upgrading to pandas 1.5.0 (linux). (I can't share this example.)

I tried to replicate the error above locally, and it seems to behave the same way. However, the same issue is only visible in ipython within my pycharm console. When I open a python instance via cmd, I cannot replicate the behaviour, which makes this bug a bit harder to trace I believe.

Expected Behavior

I would expect that the index.freq property would not be lost

Installed Versions

INSTALLED VERSIONS

commit : 87cfe4e
python : 3.9.7.final.0
python-bits : 64
OS : Windows
OS-release : 10
Version : 10.0.19043
machine : AMD64
processor : Intel64 Family 6 Model 142 Stepping 12, GenuineIntel
byteorder : little
LC_ALL : None
LANG : None
LOCALE : English_United States.1252
pandas : 1.5.0
numpy : 1.22.4
pytz : 2021.3
dateutil : 2.8.2
setuptools : 58.0.4
pip : 21.2.4
Cython : 0.29.25
pytest : 6.2.4
hypothesis : None
sphinx : 4.3.1
blosc : None
feather : None
xlsxwriter : None
lxml.etree : 4.6.4
html5lib : None
pymysql : None
psycopg2 : 2.9.3
jinja2 : 3.0.3
IPython : 7.29.0
pandas_datareader: 0.10.0
bs4 : None
bottleneck : None
brotli : None
fastparquet : None
fsspec : None
gcsfs : None
matplotlib : 3.5.0
numba : 0.55.2
numexpr : None
odfpy : None
openpyxl : 3.0.9
pandas_gbq : None
pyarrow : 8.0.0
pyreadstat : None
pyxlsb : None
s3fs : None
scipy : 1.7.3
snappy : None
sqlalchemy : 1.4.31
tables : None
tabulate : 0.8.9
xarray : None
xlrd : None
xlwt : None
zstandard : None
tzdata : None

[vamsi-verma-s]

Hi @MMCMA

I think it is only happening when I do df2=df.stack() and look at df2.index . pls try yourself once.

In [64]: idx = pd.date_range('20200101', periods=12, freq='BM')

In [65]: df = pd.DataFrame(dict(a=range(12)), index=idx)

In [66]: df.index
Out[66]:
DatetimeIndex(['2020-01-31', '2020-02-28', '2020-03-31', '2020-04-30',
               '2020-05-29', '2020-06-30', '2020-07-31', '2020-08-31',
               '2020-09-30', '2020-10-30', '2020-11-30', '2020-12-31'],
              dtype='datetime64[ns]', freq='BM')

In [67]: df2 = df.stack()

In [68]: df.index
Out[68]:
DatetimeIndex(['2020-01-31', '2020-02-28', '2020-03-31', '2020-04-30',
               '2020-05-29', '2020-06-30', '2020-07-31', '2020-08-31',
               '2020-09-30', '2020-10-30', '2020-11-30', '2020-12-31'],
              dtype='datetime64[ns]', freq='BM')

In [69]: df2.index
Out[69]:
MultiIndex([('2020-01-31', 'a'),
            ('2020-02-28', 'a'),
            ('2020-03-31', 'a'),
            ('2020-04-30', 'a'),
            ('2020-05-29', 'a'),
            ('2020-06-30', 'a'),
            ('2020-07-31', 'a'),
            ('2020-08-31', 'a'),
            ('2020-09-30', 'a'),
            ('2020-10-30', 'a'),
            ('2020-11-30', 'a'),
            ('2020-12-31', 'a')],
           )

In [70]: df.index
Out[70]:
DatetimeIndex(['2020-01-31', '2020-02-28', '2020-03-31', '2020-04-30',
               '2020-05-29', '2020-06-30', '2020-07-31', '2020-08-31',
               '2020-09-30', '2020-10-30', '2020-11-30', '2020-12-31'],
              dtype='datetime64[ns]', freq=None)

[MMCMA]

Yes, I can replicate this with pandas 1.5.0, I also checked an older pandas version and the freq property in df.index is not lost

[phofl]

commit fb3e3e60963ae4808ce976fc0f89864122eced15
Author: Luke Manley <[email protected]>
Date:   Tue Mar 15 18:43:02 2022 -0400

    PERF: MultiIndex.values for MI's with DatetimeIndex, TimedeltaIndex, or ExtensionDtype levels (#46288)

cc @lukemanley #46288

Simple reproducer:

idx = pd.date_range('20200101', periods=12, freq='BM')

df = pd.DataFrame(dict(a=range(12)), index=idx)

df2 = df.stack()
df2.index.__repr__()
assert df.index.freq is not None

[MarcoGorelli]

Hey @MMCMA

Thanks for having reported this - there's now a release candidate for pandas 2.0.0, which you can install with

mamba install -c conda-forge/label/pandas_rc pandas==2.0.0rc0

or

python -m pip install --upgrade --pre pandas==2.0.0rc0

If you try it out and report bugs, then we can fix them before the final 2.0.0 release