Unable to change normalize property of objects from to_offset()

[villebro]

Code Sample, a copy-pastable example if possible

from datetime import date

from pandas.tseries.frequencies import to_offset
from pandas.tseries.offsets import DateOffset


dt = date.today()
offset = to_offset("52W")
print(f"type: {type(offset)}\ndict: {offset.__dict__}")
print(f"{dt} -> {dt + offset}")

try:
    offset.normalize = True
except AttributeError as ex:
    print(f"\n{ex}")

offset = DateOffset(normalize=True, **offset.kwds)
print(f"\ntype: {type(offset)}\ndict: {offset.__dict__}")
print(f"{dt} -> {dt + offset}")

Yields

type: <class 'pandas.tseries.offsets.Week'>
dict: {'n': 52, 'normalize': False, '_cache': {}, 'weekday': 6}
2019-08-03 -> 2020-07-26 00:00:00

DateOffset objects are immutable.

type: <class 'pandas.tseries.offsets.DateOffset'>
dict: {'n': 1, 'normalize': True, '_cache': {}, '_offset': relativedelta(weekday=SU), '_use_relativedelta': True, 'weekday': 6}
2019-08-03 -> 2019-08-04 00:00:00

Problem description

Since Pandas 0.24 the DateOffset objects are immutable. Previously it was possible to change the normalize property after receiving an object from to_offset(), but now there doesn't seem to be any mechanism for doing this. While it seems possible to create a new instance of DateOffset with normalize = True, the type of the object changes and doesn't work correctly.

What is the best way of getting an object from to_offset() with normalize set to True? From an end-user perspective, the best solution would probably be to add a normalize attribute to the function, like so:

def to_offset(freq, normalize=False):

Output of pd.show_versions()

INSTALLED VERSIONS

commit : None
python : 3.7.3.final.0
python-bits : 64
OS : Darwin
OS-release : 18.6.0
machine : x86_64
processor : i386
byteorder : little
LC_ALL : en_US.UTF-8
LANG : en_US.UTF-8
LOCALE : en_US.UTF-8

pandas : 0.25.0
numpy : 1.17.0
pytz : 2019.2
dateutil : 2.8.0
pip : 19.2.1
setuptools : 41.0.1
Cython : None
pytest : None
hypothesis : None
sphinx : None
blosc : None
feather : None
xlsxwriter : None
lxml.etree : None
html5lib : None
pymysql : None
psycopg2 : None
jinja2 : None
IPython : None
pandas_datareader: None
bs4 : None
bottleneck : None
fastparquet : None
gcsfs : None
lxml.etree : None
matplotlib : None
numexpr : None
odfpy : None
openpyxl : None
pandas_gbq : None
pyarrow : None
pytables : None
s3fs : None
scipy : None
sqlalchemy : None
tables : None
xarray : None
xlrd : None
xlwt : None
xlsxwriter : None

[jbrockmendel]

Try type(obj)(obj.n, normalize=True, **obj.kwds)

This can/should be better documented.

[villebro]

Thanks, worked like a charm! 👍 Can try to put through a doc PR if needed.