comparing timestamp of the same dataframe in pandas-CodePudding

I am working with a dataframe similar to the one below. I have to compare the timestamp of rows. if the timestamp of the rows is within 1 hour then get the name from the row which has a value and populate it in other rows which don't have.

current data

      timestamp           name      Maths   Science   History
0   2021-08-09 10:18:48   Anni
1   2021-08-09 10:18:51             89                  34
2   2021-08-09 10:19:26                        76
3   2021-08-11 12:39:24             43
4   2021-08-11 12:39:45   Jeff                           65
5   2021-08-11 12:45:05   Jerry                65

expected data

       timestamp          name      Maths   Science   History
0   2021-08-09 10:18:48   Anni
1   2021-08-09 10:18:51   Anni       89                  34
2   2021-08-09 10:19:26   Anni                76
3   2021-08-11 12:39:24   Jeff       43
4   2021-08-11 12:39:45   Jeff                           65
5   2021-08-11 12:45:05   Jerry                65

But I can not find logic to this problem. Any idea?

CodePudding user response：

Can you try this ?

n=0
first = df.timestamp[0]
empty_list = []
for time in df.timestamp:
    diff = time - first
    if diff > pd.Timedelta("1h"):
        n =1
        first = time
    empty_list.append(n)

df["helper"] = empty_list

df["name"] = df.groupby("helper")["name"].ffill().bfill()

del df["helper"]