how to find the consecutive value in time?

Question

so basically I have a dataframe that looks like this

        df=pd.DataFrame([['PLACE A','4/10/2022 3:19:46PM','TYPE2','ALARM'],    
        ['PLACEB','4/10/2022 12:29:16AM','TYPE1','RESET'],
        ['PLACE A','4/10/2022 10:29:46PM', 'TYPE2','RESET'],
        ['PLACEB','4/10/2022 9:12:31AM','TYPE1','ALARM'],
        ['PLACEB','4/10/2022 11:26:16AM','TYPE1','RESET']])

OUTPUT

    0              1                   2          3
0   PLACE A  4/10/2022 3:19:46PM      TYPE2    ALARM
1   PLACE B  4/10/2022 12:29:16AM     TYPE1    RESET
2   PLACE A  4/10/2022 10:29:46PM     TYPE2    RESET
3   PLACE B  4/10/2022 9:12:31AM      TYPE1    ALARM
4   PLACE B  4/10/2022 11:26:16AM     TYPE1    RESET

I want to get a column with the duration of alarm. I tried splitting the df (one with alarm only and other with reset) and then merge) but there are two problems:

1. if there multiple alarms for the same place and type then each alarm get multiple reset and could show negative duration !
2. there could be reset (from the last day) and alarm only with no reset which would show duration of negative value.

so basically I want to merge if only the time of reset is after the alarm and it should be the first consecutive value for multiple alarms.