handling datetime.timedelta type variable in a for loop-CodePudding

For the code below, I am getting an unexpected result.

for j in range(len(bc)):
    df.loc[df['dd'] == 0, 'time'] = datetime(2021,7,11,df.iloc[j]['hh'],df.iloc[j]['mm'],df.iloc[j]['ss']) 
    df.loc[df['dd'] == 1, 'time'] = datetime(2021,7,12,df.iloc[j]['hh'],df.iloc[j]['mm'],df.iloc[j]['ss'])
print(df.head(5))

What I get

            index  dd  hh  mm  ss                time
0 0 days 04:52:00   0   4  52   0 2021-07-11 04:56:00   
1 1 days 04:53:00   1   4  53   0 2021-07-11 04:56:00   
2 0 days 04:54:00   0   4  54   0 2021-07-11 04:56:00   
3 1 days 04:55:00   1   4  55   0 2021-07-11 04:56:00   
4 0 days 04:56:00   0   4  56   0 2021-07-11 04:56:00

What I want

            index  dd  hh  mm  ss                time
0 0 days 04:52:00   0   4  52   0 2021-07-11 04:52:00   
1 1 days 04:53:00   1   4  53   0 2021-07-12 04:53:00   
2 0 days 04:54:00   0   4  54   0 2021-07-11 04:54:00   
3 1 days 04:55:00   1   4  55   0 2021-07-12 04:55:00   
4 0 days 04:56:00   0   4  56   0 2021-07-11 04:56:00

CodePudding user response：

You could do:

df['time'] = df.apply(
    lambda row: datetime(
        2021, 7, 11 if row['dd'] == 0 else 12, row['hh'], row['mm'], row['ss']
    ),
    axis=1
)