使用熊猫解析时区变化（由于夏令时）的csv

Question

I'm trying to parse a csv that looks like this

time                                val
28.10.2007 00:00:00.000 GMT+0100    1
28.10.2007 00:01:00.000 GMT+0100    2
28.10.2007 01:00:00.000 GMT-0000    3
28.10.2007 01:01:00.000 GMT-0000    4

To do so, I use

pd.read_csv(f,
            parse_dates=[0],
            dayfirst=True,
            index_col=0)

However, the result looks like this

                           val
time                          
2007-10-28 00:00:00-01:00    1
2007-10-28 00:01:00-01:00    2
2007-10-28 00:00:00-01:00    3
2007-10-28 00:01:00-01:00    4

This cause the 3rd and 4th row to be duplicated values. Is there a way to ask pandas to convert this time to UTC and understand the change in TZ?

Answer 1

I tried this and somehow it works but I don't know if this is something that you want.

df = pd.read_csv('data.csv')

df['time'] = pd.to_datetime(df['time'], format='%d.%m.%Y %H:%M:%S.%f GMT%z')
df['time_'] = pd.to_datetime(df['time'], utc=True)