Split period into two dates when the date has the same delimiter

Use a regex to split on the dash with surrounding spaces:

out = (df['period'].str.split(r's+-s+',expand=True)
         .set_axis(['period_start', 'period_end'], axis=1)
       )

or to remove the column and create new ones:

df[['period_start', 'period_end']] = df.pop('period').str.split(r's+-s+',expand=True) 

output:

  period_start  period_end
0   31-02-2022  30.09.2022
1   31.01.2022  31.12.2022
2   28.02.2019  30-04-2020

the problem is you were trying to split on dash, and there’s many dashes in the one row, this work :

df[['period_start','period_end']]= df['period'].str.split(' - ',expand=True) 

because we split on space + dash

You can use pandas.Series.str.extract to capture the two dates.

Try this :

out = (
        df['period'].str.replace('s*', '', regex=True)
                    .str.extract('(d{2}.d{2}.d{4}).(d{2}.d{2}.d{4})',expand=True)
                    .apply(pd.to_datetime, dayfirst=True)
                    .apply(lambda x: x.dt.strftime('%d.%m.%Y'))
                    .rename(columns={0: 'period_start', 1: 'period_end'})
      )

# Output :

print(out)

  period_start  period_end
0   28.02.2022  30.09.2022
1   31.01.2022  31.12.2022
2   28.02.2019  30.04.2020
3   20.01.2019  22.02.2020
4   19.03.2020  24.05.2021
5   13.09.2022  12.10.2022