aws/aws-sdk-pandas

Timezone issue

Kayden-lolasery opened this issue · 1 comments

Describe the bug

UTC to UTC + 8 is not working as intended (it is an hour behind)
image

How to Reproduce

import pytz

# argus_path is a s3 path
zzzzz = wr.s3.describe_objects(argus_path, last_modified_begin=dt(2024, 7, 20, 19, 0, tzinfo=pytz.timezone('Asia/Singapore')))
print([zzzzz[i]['LastModified'] +timedelta(hours=8) for i in zzzzz])

Expected behavior

i would expect last_modified_begin=dt(2024, 7, 20, 19, 0, tzinfo=pytz.timezone('Asia/Singapore')) to give me everything including and after 7pm but it doesnt.

Your project

No response

Screenshots

image

OS

Win

Python version

3.9

AWS SDK for pandas version

3.2.1

Additional context

No response

seems like pytz library changed in behaviour ... perhaps it does some DST during its conversion.
specifying a datetime pytz's timezone object results in:
<DstTzInfo 'Asia/Singapore' LMT+6:55:00 STD>
while the dt.now() timezone object results in:
<DstTzInfo 'Asia/Singapore' +08+8:00:00 STD>

I'll use zoneinfo.ZoneInfo instead.