Now it looks like it's correct for all rows in my data set. I've been working on a new solution on my own but didn't get correct result when the range...

Thank you Ken for you input. Your solution doesn't expand the intervals and that makes it really fast. But it doesn't produce correct result. In this scenario the final row...

• This reply was modified 1 year ago by  Alfred.
• This reply was modified 1 year ago by  Alfred.

Thank you for your input Steve. Your...

This gives the same results as Ken's query.  (NOTE: I changed your permanent table to a temp table.)

WITH InOutStatuses AS
(
    SELECT ts.GroupingId, i.InOutDate,...

Thanks for your reply.

I've tested the query on a larger set of testdata and it's very fast. But will experiment with distinct as well.

Wow that is so amazing, thank you so much!

One qustion: whats the purpose of

THEN DATEADD(minute, DATEDIFF(minute, '2020', T.StartDateTime), '2020')
ELSE DATEADD(minute, DATEDIFF(minute, '2020', T.EndDateTime), '2020')

instead of

            
October 17, 2023 at 1:11 pm
            
#4309767

            
            
                                                
        

I've done some testing and when I added partition by PersonId and Group7 it gives me exactly what I want.

The trick with datediff  (Boundary5 in CTE Boundaries) was what I...

Thank you all for the tips and insights.

When I discovered this I started by unistall everything that had to do with SQL Server, and installed it again. All previous instances...

When I run that query and display the results as text I goes on and on for over a minute, then I stopped. Displaying over 40K rows.

If I run it...

Thank you for your reply!

Unfortunally I can't see wierd with memory och I/O.

But I just saw that if I display the results to text instead of grid there is no...