pytz UTC 转换

发布于 2024-08-03 15:15:24 字数 838 浏览 2 评论 0原文

将简单时间和 tzinfo 转换为 UTC 时间的正确方法是什么? 假设我有:

d = datetime(2009, 8, 31, 22, 30, 30)
tz = timezone('US/Pacific')

第一种方式,受 pytz 启发:

d_tz = tz.normalize(tz.localize(d))
utc = pytz.timezone('UTC')
d_utc = d_tz.astimezone(utc)

第二种方式,来自 UTCDateTimeField

def utc_from_localtime(dt, tz):
    dt = dt.replace(tzinfo=tz)
    _dt = tz.normalize(dt)
    if dt.tzinfo != _dt.tzinfo:
        # Houston, we have a problem...
        # find out which one has a dst offset
        if _dt.tzinfo.dst(_dt):
            _dt -= _dt.tzinfo.dst(_dt)
        else:
            _dt += dt.tzinfo.dst(dt)
    return _dt.astimezone(pytz.utc)

不用说这两种方法对相当多的时区产生不同的结果。

问题是 - 什么是正确的方法?

What is the right way to convert a naive time and a tzinfo into an UTC time?
Say I have:

d = datetime(2009, 8, 31, 22, 30, 30)
tz = timezone('US/Pacific')

First way, pytz inspired:

d_tz = tz.normalize(tz.localize(d))
utc = pytz.timezone('UTC')
d_utc = d_tz.astimezone(utc)

Second way, from UTCDateTimeField

def utc_from_localtime(dt, tz):
    dt = dt.replace(tzinfo=tz)
    _dt = tz.normalize(dt)
    if dt.tzinfo != _dt.tzinfo:
        # Houston, we have a problem...
        # find out which one has a dst offset
        if _dt.tzinfo.dst(_dt):
            _dt -= _dt.tzinfo.dst(_dt)
        else:
            _dt += dt.tzinfo.dst(dt)
    return _dt.astimezone(pytz.utc)

Needless to say those two methods produce different results for quite a few timezones.

Question is - what's the right way?

如果你对这篇内容有疑问,欢迎到本站社区发帖提问 参与讨论,获取更多帮助,或者扫码二维码加入 Web 技术交流群。

扫码二维码加入Web技术交流群

发布评论

需要 登录 才能够评论, 你可以免费 注册 一个本站的账号。

评论(4

稀香 2024-08-10 15:15:24

您的第一个方法似乎是已批准的方法,并且应该了解 DST。

您可以稍微缩短它,因为 pytz.utc = pytz.timezone('UTC'),但您已经知道了:)

tz = timezone('US/Pacific')
def toUTC(d):
    return tz.normalize(tz.localize(d)).astimezone(pytz.utc)

print "Test: ", datetime.datetime.utcnow(), " = ", toUTC(datetime.datetime.now())

Your first method seems to be the approved one, and should be DST-aware.

You could shorten it a tiny bit, since pytz.utc = pytz.timezone('UTC'), but you knew that already :)

tz = timezone('US/Pacific')
def toUTC(d):
    return tz.normalize(tz.localize(d)).astimezone(pytz.utc)

print "Test: ", datetime.datetime.utcnow(), " = ", toUTC(datetime.datetime.now())
耶耶耶 2024-08-10 15:15:24

将 naive 时间和 tzinfo 转换为 utc 时间的正确方法是什么?

此答案列举了将本地时间转换为 UTC 的一些问题

from datetime import datetime
import pytz # $ pip install pytz

d = datetime(2009, 8, 31, 22, 30, 30)
tz = pytz.timezone('US/Pacific')

# a) raise exception for non-existent or ambiguous times
aware_d = tz.localize(d, is_dst=None)
## b) assume standard time, adjust non-existent times
#aware_d = tz.normalize(tz.localize(d, is_dst=False))
## c) assume DST is in effect, adjust non-existent times
#aware_d = tz.normalize(tz.localize(naive_d, is_dst=True))

# convert to UTC
utc_d = aware_d.astimezone(pytz.utc)

What is the right way to convert a naive time and a tzinfo into an utc time?

This answer enumerates some issues with converting a local time to UTC:

from datetime import datetime
import pytz # $ pip install pytz

d = datetime(2009, 8, 31, 22, 30, 30)
tz = pytz.timezone('US/Pacific')

# a) raise exception for non-existent or ambiguous times
aware_d = tz.localize(d, is_dst=None)
## b) assume standard time, adjust non-existent times
#aware_d = tz.normalize(tz.localize(d, is_dst=False))
## c) assume DST is in effect, adjust non-existent times
#aware_d = tz.normalize(tz.localize(naive_d, is_dst=True))

# convert to UTC
utc_d = aware_d.astimezone(pytz.utc)
若能看破又如何 2024-08-10 15:15:24

使用第一种方法。没有理由重新发明时区转换的轮子

Use the first method. There's no reason to reinvent the wheel of timezone conversion

π浅易 2024-08-10 15:15:24
    import pytz
    from django.utils import timezone

    tz = pytz.timezone('America/Los_Angeles')
    time = tz.normalize(timezone.now())
    import pytz
    from django.utils import timezone

    tz = pytz.timezone('America/Los_Angeles')
    time = tz.normalize(timezone.now())
~没有更多了~
我们使用 Cookies 和其他技术来定制您的体验包括您的登录状态等。通过阅读我们的 隐私政策 了解更多相关信息。 单击 接受 或继续使用网站,即表示您同意使用 Cookies 和您的相关数据。
原文