how to use intervalindex with datetimes?

Hello there!

I am struggling with the amazing intervalindex new feature. Consider this simple example:

df = pd.DataFrame({'id' : ['pandas','peanuts'],
                  'inter_left' : [pd.to_datetime('2018-08-01 10:00:00'),
                                  pd.to_datetime('2018-08-01 17:00:00')],
                  'inter_right' : [pd.to_datetime('2018-08-01 12:00:00'),
                                  pd.to_datetime('2018-08-01 18:00:00')]})

Out[81]: 
        id          inter_left         inter_right
0   pandas 2018-08-01 10:00:00 2018-08-01 12:00:00
1  peanuts 2018-08-01 17:00:00 2018-08-01 18:00:00

df_other = pd.DataFrame({'id' : ['jeff','wes'],
                  'inter_left' : [pd.to_datetime('2018-08-01 11:00:00'),
                                  pd.to_datetime('2018-08-01 20:00:00')],
                  'inter_right' : [pd.to_datetime('2018-08-01 13:00:00'),
                                  pd.to_datetime('2018-08-01 22:00:00')]})

Out[85]: 
     id          inter_left         inter_right
0  jeff 2018-08-01 11:00:00 2018-08-01 13:00:00
1   wes 2018-08-01 20:00:00 2018-08-01 22:00:00

Here, each row in df identifies an individual, and inter_left and inter_right are the left and right boundaries of the interval at which id is observed.

I would like to ‘merge’ this dataframe with df_other with the understanding that an id from df_other is merged to an id from df only if its corresponding interval overlaps the one of id.

In the example, only jeff, who is there between 11 and 13, overlaps with pandas, who is there between 10 and 12.

My understanding is that intervalindex can do that. Despite my attempts, I wasnt able to make it work. Is this supposed to be doable?

Thanks!

[jreback]

https://stackoverflow.com/questions/46525786/how-to-join-two-dataframes-for-which-column-values-are-within-a-certain-range/46526249

[toobaz]

I guess this is solved