Question

I have table of items, every one has it’s start time and end time.

For example:

| id      | item | start      | end        |
+---------+------+------------+------------+
|    2703 | 223  | 2003-12-30 | 2004-01-16 |
|   47029 | 223  | 2004-04-21 | 2005-01-06 |
| 1386593 | 223  | 2005-04-19 | 2005-04-19 |
| 1386739 | 223  | 2005-04-19 | 2006-12-07 |
| 1432862 | 223  | 2006-02-08 | 2006-02-10 |

I’d like to find gaps between items (days between one’s end with next start). Gap between first two items in example is 93 days, between last 2 there is overlap 302 days, which is counted as zero. So gaps for example table are:

96
103
0
0

My goal is to find average gap (49.75 in this example) or other-words sum of gap days (199).

Optional goals to have: min gap (0) and max gap (103).

My solution for now is iterate over rows and calculate it outside database, but i am looking for prompt solution (inside MySQL). I looked over the aggregate functions but don’t have no idea is this possible or not. How is the best way to find such average gap?

Answer 1

select avg(if (diff < 0, 0, diff))
from (
    select datediff(min(t2.start), t1.end) as diff
    from Table1 t1
    inner join Table1 t2 on t1.id < t2.id
    group by t1.end
) a

SQL Fiddle Example