Find the total time taken by each `id` within a 1 hours - sql

Question

How to find the total time taken by each id.
There must be one activity within an hour (After the start of the one activity),
and if there is any activity before the 1hrs time is elapsed, consider that time in the previous slot, calculate the time differences.
if 1 hrs has already passed now check if there is any other activity within a 1hrs period
Sample data is:
create table test (
id integer not null,
ts datetime not null
);
insert into test values (1, '2012-01-01 08:00');
insert into test values (1, '2012-01-01 08:01');
insert into test values (1, '2012-01-01 08:06');
insert into test values (1, '2012-01-01 08:30');
insert into test values (1, '2012-01-01 09:35');
insert into test values (1, '2012-01-02 16:10');
insert into test values (1, '2012-01-02 16:20');
insert into test values (1, '2012-01-03 06:40');
insert into test values (1, '2012-01-03 06:41');
insert into test values (2, '2012-01-01 08:30');
insert into test values (2, '2012-01-01 09:26');
insert into test values (2, '2012-01-01 10:25');
The output column will be:
insert into test values (1, '2012-01-01 08:00');
insert into test values (1, '2012-01-01 08:01');
insert into test values (1, '2012-01-01 08:06');
insert into test values (1, '2012-01-01 08:30');
Here the differences are exactly: 00:30:00
and there is the record which but it is after 1 hr, so we didn't add this activity on the previous slot and there is no activity up to 1 hour from this time so we do not consider this activity
insert into test values (1, '2012-01-01 09:35');`
Now, this runs for: 00:10:00
insert into test values (1, '2012-01-02 16:10');
insert into test values (1, '2012-01-02 16:20');
Similarly, this runs for 00:01:00
insert into test values (1, '2012-01-03 06:40');
insert into test values (1, '2012-01-03 06:41');
So the total time for id=1 will be 00:41:00.
id | Time duration (hh:mm:ss)
1 | 00:41:00
2 | 01:55:00
I couldn't think of how to start with this problem, any possible head-start would be appreciated. Thank you!

Answer 1

Your rules are a bit hard to follow, but I think I understand them. Count the gap between two rows for the same id if the second is less than an hour after the first. Then aggregate by the id.
So:
select id, sum(datediff(minute, ts, next_ts)) as duration_minutes
from (select t.*,
lead(ts) over (partition by id order by ts) as next_ts
from test t
) t
where datediff(minute, ts, next_ts) < 60
group by id;
Here is a db<>fiddle.
I am reluctant to convert the duration to a time value, because that is limited to just 24 hours.