将时区指定为字符串的日期被解析为 naive

Question

I'm curious why the timezone in this example, GMT , is not parsed as a valid one:

>>> from datetime import datetime
>>> import pytz
>>> b = 'Mon, 3 Oct 2016 21:24:17 GMT'
>>> fmt = '%a, %d %b %Y %H:%M:%S %Z'
>>> datetime.strptime(b, fmt).astimezone(pytz.utc)
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ValueError: astimezone() cannot be applied to a naive datetime

Doing the same with a -0700 instead of GMT and %z instead of %Z in the format works just fine.

What's the proper way to parse dates ending in string time zones if not this?

Answer 1

Use .replace() method with datetime object to update the time zone info.

>>> datetime.strptime(b, fmt).replace(tzinfo=pytz.utc)
datetime.datetime(2016, 10, 3, 21, 24, 17, tzinfo=<UTC>)

Since you mentioned, .astimezone() is working with %Z instead of %s in the format string. Even though there is z in both the formatting (difference in just case), but they are totally different in terms of what they represent.

As per the strftime 's directive document:

%z : UTC offset in the form +HHMM or -HHMM (empty string if the the object is naive).

%Z : Time zone name (empty string if the object is naive).