Trouble using ROW_NUMBER() OVER (PARTITION BY ...)

row_number() over (partition by oracle
row_number() over (partition by multiple columns)
sql row_number() over (partition by performance)
c# row_number() over(partition)
row_number() over(partition by where clause)
row_number() over (partition by snowflake
row_number() over (partition by example in teradata
row_number() over (order by

I'm using SQL Server 2008 R2. I have table called EmployeeHistory with the following structure and sample data:

EmployeeID Date      DepartmentID SupervisorID
10001      20130101  001          10009
10001      20130909  001          10019
10001      20131201  002          10018
10001      20140501  002          10017
10001      20141001  001          10015
10001      20141201  001          10014

Notice that the Employee 10001 has been changing 2 departments and several supervisors over time. What I am trying to do is to list the start and end dates of this employee's employment in each Department ordered by the Date field. So, the output will look like this:

EmployeeID DateStart DateEnd  DepartmentID 
10001      20130101  20131201 001
10001      20131201  20141001 002
10001      20141001  NULL     001

I intended to use partitioning the data using the following query but it failed. The Department changes from 001 to 002 and then back to 001. Obviously I cannot partition by DepartmentID... I'm sure I'm overlooking the obvious. Any help? Thank you, in advance.

SELECT * ,ROW_NUMBER() OVER (PARTITION BY EmployeeID, DepartmentID
ORDER BY [Date]) RN FROM EmployeeHistory

A bit involved. Easiest would be to refer to this SQL Fiddle I created for you that produces the exact result. There are ways you can improve it for performance or other considerations, but this should hopefully at least be clearer than some alternatives.

The gist is, you get a canonical ranking of your data first, then use that to segment the data into groups, then find an end date for each group, then eliminate any intermediate rows. ROW_NUMBER() and CROSS APPLY help a lot in doing it readably.


EDIT 2019:

The SQL Fiddle does in fact seem to be broken, for some reason, but it appears to be a problem on the SQL Fiddle site. Here's a complete version, tested just now on SQL Server 2016:

CREATE TABLE Source
(
  EmployeeID int,
  DateStarted date,
  DepartmentID int
)

INSERT INTO Source
VALUES
(10001,'2013-01-01',001),
(10001,'2013-09-09',001),
(10001,'2013-12-01',002),
(10001,'2014-05-01',002),
(10001,'2014-10-01',001),
(10001,'2014-12-01',001)


SELECT *, 
  ROW_NUMBER() OVER (PARTITION BY EmployeeID ORDER BY DateStarted) AS EntryRank,
  newid() as GroupKey,
  CAST(NULL AS date) AS EndDate
INTO #RankedData
FROM Source
;

UPDATE #RankedData
SET GroupKey = beginDate.GroupKey
FROM #RankedData sup
  CROSS APPLY 
  (
    SELECT TOP 1 GroupKey
    FROM #RankedData sub 
    WHERE sub.EmployeeID = sup.EmployeeID AND
      sub.DepartmentID = sup.DepartmentID AND
      NOT EXISTS 
        (
          SELECT * 
          FROM #RankedData bot 
          WHERE bot.EmployeeID = sup.EmployeeID AND
            bot.EntryRank BETWEEN sub.EntryRank AND sup.EntryRank AND
            bot.DepartmentID <> sup.DepartmentID
        )
      ORDER BY DateStarted ASC
    ) beginDate (GroupKey);

UPDATE #RankedData
SET EndDate = nextGroup.DateStarted
FROM #RankedData sup
  CROSS APPLY 
  (
    SELECT TOP 1 DateStarted
    FROM #RankedData sub
    WHERE sub.EmployeeID = sup.EmployeeID AND
      sub.DepartmentID <> sup.DepartmentID AND
      sub.EntryRank > sup.EntryRank
    ORDER BY EntryRank ASC
  ) nextGroup (DateStarted);

SELECT * FROM 
(
SELECT *, ROW_NUMBER() OVER (PARTITION BY GroupKey ORDER BY EntryRank ASC) AS GroupRank FROM #RankedData
) FinalRanking
WHERE GroupRank = 1
ORDER BY EntryRank;

DROP TABLE #RankedData
DROP TABLE Source

Row_Number Function With PARTITION BY Clause In SQL Server, It will assign the value 1 for the first row and increase the number of the subsequent rows. Syntax. ROW_NUMBER () OVER ([PARTITION BY� Trouble using ROW_NUMBER() OVER (PARTITION BY. Ask Question Asked 3 years, 2 months ago. Active 3 years, 2 months ago. Viewed 124 times 0. I have this query :

I would do something like this:

;WITH x 
 AS (SELECT *, 
            Row_number() 
              OVER( 
                partition BY employeeid 
                ORDER BY datestart) rn 
     FROM   employeehistory) 
SELECT * 
FROM   x x1 
   LEFT OUTER JOIN x x2 
                ON x1.rn = x2.rn + 1 

Or maybe it would be x2.rn - 1. You'll have to see. In any case, you get the idea. Once you have the table joined on itself, you can filter, group, sort, etc. to get what you need.

Overview of the SQL ROW_NUMBER function, How to use the SQL ROW_NUMBER function with PARTITION. The following example uses PARTITION BY clause on CustomerID and� The ROW_NUMBER function enumerates the rows in the sort order defined in the over clause. ROW_NUMBER() Function without Partition By clause Partition by clause is an optional part of Row_Number function and if you don't use it all the records of the result-set will be considered as a part of single record group or a single partition and then

It looks like a common gaps-and-islands problem. The difference between two sequences of row numbers rn1 and rn2 give the "group" number.

Run this query CTE-by-CTE and examine intermediate results to see how it works.

Sample data

I expanded sample data from the question a little.

DECLARE @Source TABLE
(
    EmployeeID int,
    DateStarted date,
    DepartmentID int
)

INSERT INTO @Source
VALUES
(10001,'2013-01-01',001),
(10001,'2013-09-09',001),
(10001,'2013-12-01',002),
(10001,'2014-05-01',002),
(10001,'2014-10-01',001),
(10001,'2014-12-01',001),

(10005,'2013-05-01',001),
(10005,'2013-11-09',001),
(10005,'2013-12-01',002),
(10005,'2014-10-01',001),
(10005,'2016-12-01',001);

Query for SQL Server 2008

There is no LEAD function in SQL Server 2008, so I had to use self-join via OUTER APPLY to get the value of the "next" row for the DateEnd.

WITH
CTE
AS
(
    SELECT
        EmployeeID
        ,DateStarted
        ,DepartmentID
        ,ROW_NUMBER() OVER (PARTITION BY EmployeeID ORDER BY DateStarted) AS rn1
        ,ROW_NUMBER() OVER (PARTITION BY EmployeeID, DepartmentID ORDER BY DateStarted) AS rn2
    FROM @Source
)
,CTE_Groups
AS
(
    SELECT
        EmployeeID
        ,MIN(DateStarted) AS DateStart
        ,DepartmentID
    FROM CTE
    GROUP BY
        EmployeeID
        ,DepartmentID
        ,rn1 - rn2
)
SELECT
    CTE_Groups.EmployeeID
    ,CTE_Groups.DepartmentID
    ,CTE_Groups.DateStart
    ,A.DateEnd
FROM
    CTE_Groups
    OUTER APPLY
    (
        SELECT TOP(1) G2.DateStart AS DateEnd
        FROM CTE_Groups AS G2
        WHERE
            G2.EmployeeID = CTE_Groups.EmployeeID
            AND G2.DateStart > CTE_Groups.DateStart
        ORDER BY G2.DateStart
    ) AS A
ORDER BY
    EmployeeID
    ,DateStart
;

Query for SQL Server 2012+

Starting with SQL Server 2012 there is a LEAD function that makes this task more efficient.

WITH
CTE
AS
(
    SELECT
        EmployeeID
        ,DateStarted
        ,DepartmentID
        ,ROW_NUMBER() OVER (PARTITION BY EmployeeID ORDER BY DateStarted) AS rn1
        ,ROW_NUMBER() OVER (PARTITION BY EmployeeID, DepartmentID ORDER BY DateStarted) AS rn2
    FROM @Source
)
,CTE_Groups
AS
(
    SELECT
        EmployeeID
        ,MIN(DateStarted) AS DateStart
        ,DepartmentID
    FROM CTE
    GROUP BY
        EmployeeID
        ,DepartmentID
        ,rn1 - rn2
)
SELECT
    CTE_Groups.EmployeeID
    ,CTE_Groups.DepartmentID
    ,CTE_Groups.DateStart
    ,LEAD(CTE_Groups.DateStart) OVER (PARTITION BY CTE_Groups.EmployeeID ORDER BY CTE_Groups.DateStart) AS DateEnd
FROM
    CTE_Groups
ORDER BY
    EmployeeID
    ,DateStart
;

Result

+------------+--------------+------------+------------+
| EmployeeID | DepartmentID | DateStart  |  DateEnd   |
+------------+--------------+------------+------------+
|      10001 |            1 | 2013-01-01 | 2013-12-01 |
|      10001 |            2 | 2013-12-01 | 2014-10-01 |
|      10001 |            1 | 2014-10-01 | NULL       |
|      10005 |            1 | 2013-05-01 | 2013-12-01 |
|      10005 |            2 | 2013-12-01 | 2014-10-01 |
|      10005 |            1 | 2014-10-01 | NULL       |
+------------+--------------+------------+------------+

Why You Can't Use ROW_NUMBER() In Your WHERE Clause, This could be across the entire query result, or within partitions. f) ROW_NUMBER() OVER WNS is equivalent to the <window function>: COUNT The problem with this approach is that if you order by some column from the� In this example, we skipped the PARTITION BY clause, therefore, the ROW_NUMBER() treated the whole result set as a single partition. Using SQL Server ROW_NUMBER() over partitions example. The following example uses the ROW_NUMBER() function to assign a sequential integer to each customer. It resets the number when the city changes:

Row numbers with nondeterministic order, It is inside these parentheses of OVER() that you can specify both PARTITION and ORDER BY clauses,� Now lets remove the duplicates/triplicates in one query in an efficient way using Row_Number() Over() with the Partition By clause. Since we have identified the duplicates/triplicates as the rows

Learn more about How to Use ROW_NUMBER() in DB2, The set of rows on which the ROW_NUMBER() function operates is called a window. Because we did not use the PARTITION BY clause, the ROW_NUMBER() To solve this problem, we can get a list of distinct prices in a CTE, the apply� ROW_NUMBER() OVER (PARTITION BY colx, coly… ORDER BY colz) as aliasname. Let’s take a closer look: ROW_NUMBER() – you’re instructing the query engine to give you back a column with row numbers. These come back as a bigint. OVER – you’re telling it that you’re about to give it some more information.

PostgreSQL ROW_NUMBER() Explained with Practical Examples, In this tutorial, you will learn how to use the Oracle ROW_NUMBER() function to WITH cte_products AS ( SELECT row_number() OVER( PARTITION BY� To add a row number column in front of each row, add a column with the ROW_NUMBER function, in this case named Row#. You must move the ORDER BY clause up to the OVER clause. SELECT ROW_NUMBER() OVER(ORDER BY name ASC) AS Row#, name, recovery_model_desc FROM sys.databases WHERE database_id < 5; Here is the result set.

Comments
  • More than two years later, @coding_idiot, but I've rectified it via a complete script to run on your own server.
  • @DominicP, I think it can be done in a much simpler and efficient way. Have a look at my answer.