将 zoneinfo 与 pandas.date_range 一起使用

Question

I am trying to use zoneinfo instead of pytz . I am running into a problem using zoneinfo to initiate dates and passing it on to pd.date_range .

Below is an example of doing the exact same thing with pytz and with zoneinfo . But, while passing it to pd.date_range getting an error with the latter.

pytz example:

start_date = datetime(2021, 1, 1, 0, 0, 0)end_date = datetime(2024, 1, 1, 0, 0, 0) # exclusive end range

pt = pytz.timezone('Canada/Pacific')start_date = pt.localize(start_date)end_date = pt.localize(end_date)

pd.date_range(start_date, end_date-timedelta(days=1), freq='d')

zoneinfo example:

start_date1 = '2021-01-01 00:00:00

start_date1 = datetime.strptime(start_date1, '%Y-%m-%d %H:%M:%S').replace(microsecond=0, second=0, minute=0, tzinfo=ZoneInfo("America/Vancouver"))end_date1 = start_date1 + relativedelta(years=3)

pd.date_range(start_date1, end_date1-timedelta(days=1), freq='d')

Yet, when using zoneinfo I get the following error:

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
~/Documents/GitHub/virtual/lib/python3.9/site-packages/pandas/_libs/tslibs/timezones.pyx in pandas._libs.tslibs.timezones.get_dst_info()

AttributeError: 'NoneType' object has no attribute 'total_seconds'

Exception ignored in: 'pandas._libs.tslibs.tzconversion.tz_convert_from_utc_single'
Traceback (most recent call last):
  File "pandas/_libs/tslibs/timezones.pyx", line 266, in pandas._libs.tslibs.timezones.get_dst_info
AttributeError: 'NoneType' object has no attribute 'total_seconds'
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
~/Documents/GitHub/virtual/lib/python3.9/site-packages/pandas/_libs/tslibs/timezones.pyx in pandas._libs.tslibs.timezones.get_dst_info()

AttributeError: 'NoneType' object has no attribute 'total_seconds'

Exception ignored in: 'pandas._libs.tslibs.tzconversion.tz_convert_from_utc_single'
Traceback (most recent call last):
  File "pandas/_libs/tslibs/timezones.pyx", line 266, in pandas._libs.tslibs.timezones.get_dst_info
AttributeError: 'NoneType' object has no attribute 'total_seconds'
---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
/var/folders/vp/7ptlp5l934vdh1lvmpgk4qyc0000gn/T/ipykernel_67190/3566591779.py in <module>
      5 end_date1 = start_date1 + relativedelta(years=3)
      6 
----> 7 pd.date_range(start_date1, end_date1-timedelta(days=1), freq='d')
      8 
      9 # Because certain distributions will be a result of combined distributions,

~/Documents/GitHub/virtual/lib/python3.9/site-packages/pandas/core/indexes/datetimes.py in date_range(start, end, periods, freq, tz, normalize, name, closed, **kwargs)
   1095         freq = "D"
   1096 
-> 1097     dtarr = DatetimeArray._generate_range(
   1098         start=start,
   1099         end=end,

~/Documents/GitHub/virtual/lib/python3.9/site-packages/pandas/core/arrays/datetimes.py in _generate_range(cls, start, end, periods, freq, tz, normalize, ambiguous, nonexistent, closed)
    450 
    451             if tz is not None and index.tz is None:
--> 452                 arr = tzconversion.tz_localize_to_utc(
    453                     index.asi8, tz, ambiguous=ambiguous, nonexistent=nonexistent
    454                 )

~/Documents/GitHub/virtual/lib/python3.9/site-packages/pandas/_libs/tslibs/tzconversion.pyx in pandas._libs.tslibs.tzconversion.tz_localize_to_utc()

~/Documents/GitHub/virtual/lib/python3.9/site-packages/pandas/_libs/tslibs/timezones.pyx in pandas._libs.tslibs.timezones.get_dst_info()

AttributeError: 'NoneType' object has no attribute 'total_seconds'

Testing the parameters:

start_date==start_date1

and

end_date==end_date1 Both tests result in True .

Answer 1

if understanding correctly you want to create a date range (1D freq) using ZoneInfo…if correct I see a few things going on with your code.

#1 When dealing with datetimes be sure the object is in the correct dtype. I believe datetime64 format will work better.

#2 From the provide code I don't think 'strptime' or 'replace' are needed. To access "America/Vancouver" within ZoneInfo you can make it work if you parse start_date1 into years, months, days, hours and minutes.

#3 When start_date1 is parsed, you can add 3 to years (or another number) to create the end date.

The above will create a DatetimeIndex over the specified range.

Datetimes are always tricky. As always you can get to the same destination using different paths…this is just one of them.

start_date_str = '2021-01-01 00:00:00'
start_date_datetime64 = pd.to_datetime(start_date_str) # change dtype  to datetime64

year = start_date_datetime64.year
month = start_date_datetime64.month
day = start_date_datetime64.day
hour = start_date_datetime64.hour
minute = start_date_datetime64.minute

start_date_formatted = dt.datetime(year, month, day, hour, minute, tzinfo=ZoneInfo("America/Vancouver"))
end_date_formatted = dt.datetime(year + 3, month, day, hour, minute, tzinfo=ZoneInfo("America/Vancouver"))

result = pd.date_range(start_date_formatted, end_date_formatted-pd.Timedelta(days=1), freq='d')

OUTPUT- DatetimeIndex, dtype='datetime64[ns, America/Vancouver]', length=1095, freq='D')

Answer 2

This error was a result of compatibility between the pandas version and the nbformat version. Once I updated both to the newest version, the code worked with no error.