The first dataframe can be convert to datatime format with pandas. The second dataframe missed some fields in input time string, what’s proper way to covert the time strings to datetime object?

import pandas as pd

data = [
    '+2h03m46s456ms',   
]
df = pd.DataFrame(data, columns=['ts'])
df['ts'] = pd.to_datetime(df['ts'],format="+%Hh%Mm%Ss%fms")

print(df)

data2=[
    '0',        
    '+456ms',    
    '+46s456ms',    
    '+3m46s456ms'
    ]
df2 = pd.DataFrame(data2, columns=['ts'])
#df2['ts'] = pd.to_datetime(df2['ts'],format="+%Hh%Mm%Ss%fms")
print(df2)

Output:

                       ts
0 1900-01-01 02:03:46.456
            ts
0            0
1       +456ms
2    +46s456ms
3  +3m46s456ms