Question

I would like to create an integer counter for duplicate entries on an intermediate table in an import step. Let’s say I have a table called FinalTable, which looks like this:

FinalTable

ID       Date          SameDateCounter
1003     2012/01/01    NULL
1004     2012/02/01    NULL
1005     2012/03/01    1
1006     2012/03/01    2
1007     2012/03/01    3
1008     2012/04/01    1
1009     2012/04/01    2

And has a unique constraint on (Date, SameDateCounter). In the intermediate step the data is in IntermediateTable, which looks exactly like FinalTable except that it doesn’t have any constraints. So the starting point for this problem is here:.

IntermediateTable:

ID       Date          SameDateCounter
1003     2012/01/01    NULL
1004     2012/02/01    NULL
1005     2012/03/01    NULL
1006     2012/03/01    NULL
1007     2012/03/01    NULL
1008     2012/04/01    NULL
1009     2012/04/01    NULL

And I need to generate the values for SameDateCounter, starting at one, where the date appears more than once.

Answer 1

You can use the ROW_NUMBER() function combined with a PARTITION BY clause:

;WITH Data AS
(  
   SELECT
       ID, Date,
       SameDateCounter = ROW_NUMBER() OVER (PARTITION BY Date ORDER BY ID DESC)
   FROM dbo.YourTableNameHere
)
SELECT *
FROM Data

The PARTITION BY Date clause groups every separate value of Date into its own “partition”, and ROW_NUMBER() hands out numbers starting at 1, for each partition.

So your SameDayCounter will start at 1 for each new day and will show the rows for that particular day.