How to get the exact time after converting from UTC to any other time zone in Python?

Question

I've a data with time columns, which I have to convert from utc to ist

example

utc_datetime         tag_value   inserted_at
2018-12-03 16:48:32  3.38        2018-12-03 22:18:48 

I've used the following code to convert time in UTC to IST

data = data.rename(columns={"utc_datetime": "ist_datetime"})
data['ist_datetime'] = pd.to_datetime(data['ist_datetime'], errors='coerce') 
data['ist_datetime'] = pd.to_datetime(data['ist_datetime'], errors='coerce')
data['inserted_at'] = pd.to_datetime(data['inserted_at'], errors='coerce')
data['ist_datetime'] = data['ist_datetime'].dt.tz_localize('UTC').dt.tz_convert('Asia/Kolkata')
data['inserted_at'] = data['inserted_at'].dt.tz_localize('UTC').dt.tz_convert('Asia/Kolkata')

For which I'm getting the output like

What should I do to get the exact converted time and not + 5:30 with the original time?

Eg - '16:18:32' to '21:48:32' and not '16:18:32 + 5:30'

Thanks

Answer 1

From your comment, I think you may be mistaken about the source data.

For example:

import datetime
import pandas as pd

dt = pd.to_datetime([datetime.datetime(2018, 1, 1, 0, 0, 0)])
print(dt)
print(dt.tz_localize('UTC'))
print(dt.tz_localize('UTC').tz_convert('Asia/Kolkata'))

Works exactly as you expect:

DatetimeIndex(['2018-01-01'], dtype='datetime64[ns]', freq=None)
DatetimeIndex(['2018-01-01 00:00:00+00:00'], dtype='datetime64[ns, UTC]', freq=None)
DatetimeIndex(['2018-01-01 05:30:00+05:30'], dtype='datetime64[ns, Asia/Kolkata]', freq=None)

You don't provide us with the source data, only the result you think is wrong - you'd have to provide a more complete example, with output of the source data to make clear what the problem is, if you believe there still actually is one.