Calculate date differences in consecutive rows and generate a sequence

Question

Hi Guys,
I am trying to implement one scenario where in i need to compare the date differences for consecutive rows and generate a sequence against that.
this is the table schema:
create table temp
id int identity(1,1),
emp_id int,
time datetime
insert into temp 
values
(1, '2011-02-20 12:30:00.000'),
(1, '2011-02-20 12:32:34.172'),
(1, '2011-02-20 12:32:34.172'),
(1, '2011-02-20 12:43:21.004'),
(1, '2011-02-20 12:46:39.745'),
(2, '2011-02-20 12:48:06.004'),
(2, '2011-02-20 12:48:06.004'),
(2, '2011-02-20 12:53:07.733'),
(2, '2011-02-20 12:55:30.295');
now, I want to compare the first date-time with the second and so on. now if the date-time difference is same for two consecutive rows then the sequence should  increment by 1 otherwise the sequence again will start from '00' for any unique date-time.
This sequence number should start from '00' from each employee.
I want the output to be like this
ID emp_id
time    
sequence
1 1 2011-02-20 12:30:00.000
00
2  1 2011-02-20 12:32:34.172
00
3  1 2011-02-20 12:32:34.172
01
4  1 2011-02-20 12:32:34.172
02
5  1 2011-02-20 12:46:39.745
00
6  2 
2011-02-20 12:48:06.004 00
7  2 
2011-02-20 12:48:07.003 00
8  2 
2011-02-20 12:48:07.003 01
9  2 
2011-02-20 12:46:39.745 00
Please revert as soon as possible as this is a bit urgent.
Thank You in Advance. :)

Answer 1

create table temp
(id int identity(1,1),
 emp_id int,
 time datetime
insert into temp values
(1, '20110220 12:30:00.000'),
(1, '20110220 12:32:34.172'),
(1, '20110220 12:32:34.172'),
(1, '20110220 12:43:21.004'),
(1, '20110220 12:46:39.745'),
(2, '20110220 12:48:06.004'),
(2, '20110220 12:48:06.004'),
(2, '20110220 12:53:07.733'),
(2, '20110220 12:55:30.295');
go
SELECT id, emp_id, time,
       dateadd(mcs, (row_number() OVER(PARTITION BY emp_id, time ORDER BY
id) - 1) * 10,
               convert(datetime2(5), time)) AS [Sequence]
FROM   temp
ORDER  BY id
go
DROP TABLE temp
Erland Sommarskog, SQL Server MVP, [email protected]