Insert a row after a date in a dataframe-CodePudding

I have a Dataframe like :

    date                   col1    col2
0   2022-10-07 04:00:00    x       x1
1   2022-10-08 04:00:00    y       x2

I need to update a row (as dictionary) in a specific date if exist, and if it does not exist, insert the row next to the closest date.

For this new given date 2022-10-07 05:00:00 (one hour later) and dic {col1:z} I would like to get :

    date                   col1   col2
0   2022-10-07 04:00:00    x      x1
1   2022-10-07 05:00:00    z      x1
2   2022-10-08 04:00:00    y      x2

Currently I am doing this:

def write(date,dic):
    m = df['date'] == date
    if m.any():
        df.loc[df['date'] == date, list(dic.keys())] =  list(dic.values()) 
    else:
        df.loc[len(df)] = {**dic, **{'date':date}}

Which means that if I can't find the date, I just add the row to the end of the df, but I want to insert it right after the previous date.

(Please also see that since when I insert/update I only have col1, so col2 value will be copied from previous row somehow with ffill)

CodePudding user response：

You can set the date as index and update like:

df1 = df.set_index('date')
df1.loc[new_date, dic.keys()] = dic.values()
df = df1.sort_index().reset_index().ffill()

It will insert new date if it doesn't exist. If it exists it will update the record at that index.