SQL SELECTING ROW [duplicate]

问题:

Possible Duplicate:
SQL SERVER SELECTING ROW

If I have this table below

   id         time
   start      08.00 AM 
   stop       09.00 AM
   stop       09.30 AM
   start      09.30 AM
   start      11.00 AM
   start      11.30 AM
   stop       11.30 AM
   stop       12.00 PM

I want output that looks like output below:

Any solution?

This is how the output I want:

    id         time       end_time    interval 
   start      08.00 AM    09.00 AM      60 minute
   stop       09.00 AM    09.30 AM      30 minute
   start      09.30 AM    11.30 AM      120 minute
   stop       11.30 AM    12.00 PM      30 minute
   stop       12.00 PM     NULL          NULL

回答1:

I want output that only take every first start and every start after stop and the latest stop

That's easily done with the row_number() window function:

; with  numbered as
        (
        select  row_number() over (order by [time]) rn
        ,       count(*) over() as cnt
        ,       *
        from    Table1
        )
select  *
from    numbered cur
left join
        numbered prev
on      cur.rn = prev.rn + 1
left join
        numbered next
on      cur.rn = next.rn - 1
where   cur.rn in (1, cur.cnt) -- first or last row
        or (cur.id = 'start' and prev.id = 'stop') -- start after stop
order by
        cur.rn

Live example on SQL Fiddle.

Not that your example output does not meet your requirements. It includes several rows that match neither of the three criteria (first row, last row, start after stop.)