Please, observe:
C:\dev\poc\SDR>python
Python 2.7.1 (r271:86832, Nov 27 2010, 17:19:03) [MSC v.1500 64 bit (AMD64)] on win32
Type "help", "copyright", "credits" or "license" for more information.
>>> from pytz import timezone
>>> from datetime import datetime
>>> tz1=timezone('America/Los_Angeles')
>>> tz2=timezone('US/Pacific')
>>> ts1=datetime(2011,8,1,tzinfo=tz1)
>>> ts2=datetime(2011,8,1,tzinfo=tz2)
>>> ts1
datetime.datetime(2011, 8, 1, 0, 0, tzinfo=<DstTzInfo 'America/Los_Angeles' PST-1 day, 16:00:00 STD>)
>>> ts2
datetime.datetime(2011, 8, 1, 0, 0, tzinfo=<DstTzInfo 'US/Pacific' PST-1 day, 16:00:00 STD>)
>>> ts1.astimezone(tz1)
datetime.datetime(2011, 8, 1, 0, 0, tzinfo=<DstTzInfo 'America/Los_Angeles' PST-1 day, 16:00:00 STD>)
>>> ts2.astimezone(tz2)
datetime.datetime(2011, 8, 1, 0, 0, tzinfo=<DstTzInfo 'US/Pacific' PST-1 day, 16:00:00 STD>)
>>> ts1.astimezone(tz2)
datetime.datetime(2011, 8, 1, 1, 0, tzinfo=<DstTzInfo 'US/Pacific' PDT-1 day, 17:00:00 DST>)
>>> ts2.astimezone(tz1)
datetime.datetime(2011, 8, 1, 1, 0, tzinfo=<DstTzInfo 'America/Los_Angeles' PDT-1 day, 17:00:00 DST>)
>>>
Here is what I do not understand. US/Pacific (tz1) and America/Los_Angeles (tz2) are supposed to denote the same time zone, aren’t they? Then how come that datetime.astimezone called to move from one zone to another changes the hour?
Thanks.
Daylight Savings Time. Notice that the last two entries are listing
PDT-1.astimezonetakes into account DST, but only if it actually goes through its full logic.The first two
astimezonecalls in your example don’t go through the full logic, because they short-circuit (since the timezone they’re converting “to” already matches the one they’re converting “from”).(You might wonder why the initial datetimes aren’t using PDT already. This is because the
datetimeconstructor doesn’t take into account daylight savings time, even if you pass it a timezone – it just sets the timezone blindly.)