Transact-SQL Syntax ConventionsDetermines the partitioning and ordering of a rowset before the associated window function is applied. That is, the OVER clause defines a window or user-specified set of rows within a query result set. A window function then computes a value for each row in the window. You can use the OVER clause with functions to compute aggregated values such as moving averages, cumulative aggregates, running totals, or a top N per group results.
Transact-SQL Syntax Conventions
-- Syntax for SQL Server, Azure SQL Database, and Azure SQL Data Warehouse
OVER (
[ <PARTITION BY clause> ]
[ <ORDER BY clause> ]
[ <ROW or RANGE clause> ]
)
<PARTITION BY clause> ::=
PARTITION BY value_expression , ... [ n ]
<ORDER BY clause> ::=
ORDER BY order_by_expression
[ COLLATE collation_name ]
[ ASC | DESC ]
[ ,...n ]
<ROW or RANGE clause> ::=
{ ROWS | RANGE } <window frame extent>
<window frame extent> ::=
{ <window frame preceding>
| <window frame between>
}
<window frame between> ::=
BETWEEN <window frame bound> AND <window frame bound>
<window frame bound> ::=
{ <window frame preceding>
| <window frame following>
}
<window frame preceding> ::=
{
UNBOUNDED PRECEDING
| <unsigned_value_specification> PRECEDING
| CURRENT ROW
}
<window frame following> ::=
{
UNBOUNDED FOLLOWING
| <unsigned_value_specification> FOLLOWING
| CURRENT ROW
}
<unsigned value specification> ::=
{ <unsigned integer literal> }
-- Syntax for Parallel Data Warehouse
OVER ( [ PARTITION BY value_expression ] [ order_by_clause ] ) PARTITION BY
Divides the query result set into partitions. The window function is applied to each partition separately and computation restarts for each partition.
value_expression
Specifies the column by which the rowset is partitioned. value_expression can only refer to columns made available by the FROM clause. value_expression cannot refer to expressions or aliases in the select list. value_expression can be a column expression, scalar subquery, scalar function, or user-defined variable.
<ORDER BY clause>
Defines the logical order of the rows within each partition of the result set. That is, it specifies the logical order in which the window functioncalculation is performed.
order_by_expression
Specifies a column or expression on which to sort. order_by_expression can only refer to columns made available by the FROM clause. An integer cannot be specified to represent a column name or alias.
COLLATE collation_name
Specifies that the ORDER BY operation should be performed according to the collation specified in collation_name. collation_name can be either a Windows collation name or a SQL collation name. For more information, see Collation and Unicode Support. COLLATE is applicable only for columns of type char, varchar, nchar, and nvarchar.
ASC | DESC
Specifies that the values in the specified column should be sorted in ascending or descending order. ASC is the default sort order. Null values are treated as the lowest possible values.
ROWS | RANGE
Applies to: SQL Server 2012 (11.x) through SQL Server 2012 (11.x) SQL Server 2017
Further limits the rows within the partition by specifying start and end points within the partition. This is done by specifying a range of rows with respect to the current row either by logical association or physical association. Physical association is achieved by using the ROWS clause.
The ROWS clause limits the rows within a partition by specifying a fixed number of rows preceding or following the current row. Alternatively, the RANGE clause logically limits the rows within a partition by specifying a range of values with respect to the value in the current row. Preceding and following rows are defined based on the ordering in the ORDER BY clause. The window frame “RANGE … CURRENT ROW …” includes all rows that have the same values in the ORDER BY expression as the current row. For example, ROWS BETWEEN 2 PRECEDING AND CURRENT ROW means that the window of rows that the function operates on is three rows in size, starting with 2 rows preceding until and including the current row.
[!NOTE]
ROWS or RANGE requires that the ORDER BY clause be specified. If ORDER BY contains multiple order expressions, CURRENT ROW FOR RANGE considers all columns in the ORDER BY list when determining the current row.
UNBOUNDED PRECEDING
Applies to: SQL Server 2012 (11.x) through SQL Server 2012 (11.x) SQL Server 2017
Specifies that the window starts at the first row of the partition. UNBOUNDED PRECEDING can only be specified as window starting point.
<unsigned value specification> PRECEDING
Specified with <unsigned value specification>to indicate the number of rows or values to precede the current row. This specification is not allowed for RANGE.
CURRENT ROW
Applies to: SQL Server 2012 (11.x) through SQL Server 2012 (11.x) SQL Server 2017
Specifies that the window starts or ends at the current row when used with ROWS or the current value when used with RANGE. CURRENT ROW can be specified as both a starting and ending point.
BETWEEN <window frame bound > AND <window frame bound >
Applies to: SQL Server 2012 (11.x) through SQL Server 2012 (11.x) SQL Server 2017
Used with either ROWS or RANGE to specify the lower (starting) and upper (ending) boundary points of the window. <window frame bound> defines the boundary starting point and <window frame bound> defines the boundary end point. The upper bound cannot be smaller than the lower bound.
UNBOUNDED FOLLOWING
Applies to: SQL Server 2012 (11.x) through SQL Server 2012 (11.x) SQL Server 2017
Specifies that the window ends at the last row of the partition. UNBOUNDED FOLLOWING can only be specified as a window end point. For example RANGE BETWEEN CURRENT ROW AND UNBOUNDED FOLLOWING defines a window that starts with the current row and ends with the last row of the partition.
<unsigned value specification> FOLLOWING
Specified with <unsigned value specification> to indicate the number of rows or values to follow the current row. When <unsigned value specification> FOLLOWING is specified as the window starting point, the ending point must be <unsigned value specification>FOLLOWING. For example, ROWS BETWEEN 2 FOLLOWING AND 10 FOLLOWING defines a window that starts with the second row that follows the current row and ends with the tenth row that follows the current row. This specification is not allowed for RANGE.
unsigned integer literal
Applies to: SQL Server 2012 (11.x) through SQL Server 2012 (11.x) SQL Server 2017
Is a positive integer literal (including 0) that specifies the number of rows or values to precede or follow the current row or value. This specification is valid only for ROWS.
More than one window function can be used in a single query with a single FROM clause. The OVER clause for each function can differ in partitioning and ordering.
If PARTITION BY is not specified, the function treats all rows of the query result set as a single group.
If ROWS/RANGE is specified and <window frame preceding> is used for <window frame extent> (short syntax) then this specification is used for the window frame boundary starting point and CURRENT ROW is used for the boundary ending point. For example “ROWS 5 PRECEDING” is equal to “ROWS BETWEEN 5 PRECEDING AND CURRENT ROW”.
[!NOTE] If ORDER BY is not specified entire partition is used for a window frame. This applies only to functions that do not require ORDER BY clause. If ROWS/RANGE is not specified but ORDER BY is specified, RANGE UNBOUNDED PRECEDING AND CURRENT ROW is used as default for window frame. This applies only to functions that have can accept optional ROWS/RANGE specification. For example, ranking functions cannot accept ROWS/RANGE, therefore this window frame is not applied even though ORDER BY is present and ROWS/RANGE is not.
The OVER clause cannot be used with the CHECKSUM aggregate function.
RANGE cannot be used with <unsigned value specification> PRECEDING or <unsigned value specification> FOLLOWING.
Depending on the ranking, aggregate, or analytic function used with the OVER clause, <ORDER BY clause> and/or the <ROWS and RANGE clause> may not be supported.
The following example shows using the OVER clause with ROW_NUMBER function to display a row number for each row within a partition. The ORDER BY clause specified in the OVER clause orders the rows in each partition by the column SalesYTD. The ORDER BY clause in the SELECT statement determines the order in which the entire query result set is returned.
USE AdventureWorks2012;
GO
SELECT ROW_NUMBER() OVER(PARTITION BY PostalCode ORDER BY SalesYTD DESC) AS "Row Number",
p.LastName, s.SalesYTD, a.PostalCode
FROM Sales.SalesPerson AS s
INNER JOIN Person.Person AS p
ON s.BusinessEntityID = p.BusinessEntityID
INNER JOIN Person.Address AS a
ON a.AddressID = p.BusinessEntityID
WHERE TerritoryID IS NOT NULL
AND SalesYTD <> 0
ORDER BY PostalCode;
GO Here is the result set.
Row Number LastName SalesYTD PostalCode
--------------- ----------------------- --------------------- ----------
1 Mitchell 4251368.5497 98027
2 Blythe 3763178.1787 98027
3 Carson 3189418.3662 98027
4 Reiter 2315185.611 98027
5 Vargas 1453719.4653 98027
6 Ansman-Wolfe 1352577.1325 98027
1 Pak 4116871.2277 98055
2 Varkey Chudukatil 3121616.3202 98055
3 Saraiva 2604540.7172 98055
4 Ito 2458535.6169 98055
5 Valdez 1827066.7118 98055
6 Mensa-Annan 1576562.1966 98055
7 Campbell 1573012.9383 98055
8 Tsoflias 1421810.9242 98055The following example uses the OVER clause with aggregate functions over all rows returned by the query. In this example, using the OVER clause is more efficient than using subqueries to derive the aggregate values.
USE AdventureWorks2012;
GO
SELECT SalesOrderID, ProductID, OrderQty
,SUM(OrderQty) OVER(PARTITION BY SalesOrderID) AS Total
,AVG(OrderQty) OVER(PARTITION BY SalesOrderID) AS "Avg"
,COUNT(OrderQty) OVER(PARTITION BY SalesOrderID) AS "Count"
,MIN(OrderQty) OVER(PARTITION BY SalesOrderID) AS "Min"
,MAX(OrderQty) OVER(PARTITION BY SalesOrderID) AS "Max"
FROM Sales.SalesOrderDetail
WHERE SalesOrderID IN(43659,43664);
GO Here is the result set.
SalesOrderID ProductID OrderQty Total Avg Count Min Max
------------ ----------- -------- ----------- ----------- ----------- ------ ------
43659 776 1 26 2 12 1 6
43659 777 3 26 2 12 1 6
43659 778 1 26 2 12 1 6
43659 771 1 26 2 12 1 6
43659 772 1 26 2 12 1 6
43659 773 2 26 2 12 1 6
43659 774 1 26 2 12 1 6
43659 714 3 26 2 12 1 6
43659 716 1 26 2 12 1 6
43659 709 6 26 2 12 1 6
43659 712 2 26 2 12 1 6
43659 711 4 26 2 12 1 6
43664 772 1 14 1 8 1 4
43664 775 4 14 1 8 1 4
43664 714 1 14 1 8 1 4
43664 716 1 14 1 8 1 4
43664 777 2 14 1 8 1 4
43664 771 3 14 1 8 1 4
43664 773 1 14 1 8 1 4
43664 778 1 14 1 8 1 4 The following example shows using the OVER clause with an aggregate function in a calculated value.
USE AdventureWorks2012;
GO
SELECT SalesOrderID, ProductID, OrderQty
,SUM(OrderQty) OVER(PARTITION BY SalesOrderID) AS Total
,CAST(1. * OrderQty / SUM(OrderQty) OVER(PARTITION BY SalesOrderID)
*100 AS DECIMAL(5,2))AS "Percent by ProductID"
FROM Sales.SalesOrderDetail
WHERE SalesOrderID IN(43659,43664);
GO Here is the result set. Notice that the aggregates are calculated by SalesOrderID and the Percent by ProductID is calculated for each line of each SalesOrderID.
SalesOrderID ProductID OrderQty Total Percent by ProductID
------------ ----------- -------- ----------- ---------------------------------------
43659 776 1 26 3.85
43659 777 3 26 11.54
43659 778 1 26 3.85
43659 771 1 26 3.85
43659 772 1 26 3.85
43659 773 2 26 7.69
43659 774 1 26 3.85
43659 714 3 26 11.54
43659 716 1 26 3.85
43659 709 6 26 23.08
43659 712 2 26 7.69
43659 711 4 26 15.38
43664 772 1 14 7.14
43664 775 4 14 28.57
43664 714 1 14 7.14
43664 716 1 14 7.14
43664 777 2 14 14.29
43664 771 3 14 21.4
43664 773 1 14 7.14
43664 778 1 14 7.14
(20 row(s) affected) The following example uses the AVG and SUM functions with the OVER clause to provide a moving average and cumulative total of yearly sales for each territory in the Sales.SalesPerson table. The data is partitioned by TerritoryID and logically ordered by SalesYTD. This means that the AVG function is computed for each territory based on the sales year. Notice that for TerritoryID 1, there are two rows for sales year 2005 representing the two sales people with sales that year. The average sales for these two rows is computed and then the third row representing sales for the year 2006 is included in the computation.
USE AdventureWorks2012;
GO
SELECT BusinessEntityID, TerritoryID
,DATEPART(yy,ModifiedDate) AS SalesYear
,CONVERT(varchar(20),SalesYTD,1) AS SalesYTD
,CONVERT(varchar(20),AVG(SalesYTD) OVER (PARTITION BY TerritoryID
ORDER BY DATEPART(yy,ModifiedDate)
),1) AS MovingAvg
,CONVERT(varchar(20),SUM(SalesYTD) OVER (PARTITION BY TerritoryID
ORDER BY DATEPART(yy,ModifiedDate)
),1) AS CumulativeTotal
FROM Sales.SalesPerson
WHERE TerritoryID IS NULL OR TerritoryID < 5
ORDER BY TerritoryID,SalesYear; Here is the result set.
BusinessEntityID TerritoryID SalesYear SalesYTD MovingAvg CumulativeTotal
---------------- ----------- ----------- -------------------- -------------------- --------------------
274 NULL 2005 559,697.56 559,697.56 559,697.56
287 NULL 2006 519,905.93 539,801.75 1,079,603.50
285 NULL 2007 172,524.45 417,375.98 1,252,127.95
283 1 2005 1,573,012.94 1,462,795.04 2,925,590.07
280 1 2005 1,352,577.13 1,462,795.04 2,925,590.07
284 1 2006 1,576,562.20 1,500,717.42 4,502,152.27
275 2 2005 3,763,178.18 3,763,178.18 3,763,178.18
277 3 2005 3,189,418.37 3,189,418.37 3,189,418.37
276 4 2005 4,251,368.55 3,354,952.08 6,709,904.17
281 4 2005 2,458,535.62 3,354,952.08 6,709,904.17
(10 row(s) affected)
In this example, the OVER clause does not include PARTITION BY. This means that the function will be applied to all rows returned by the query. The ORDER BY clause specified in the OVER clause determines the logical order to which the AVG function is applied. The query returns a moving average of sales by year for all sales territories specified in the WHERE clause. The ORDER BY clause specified in the SELECT statement determines the order in which the rows of the query are displayed.
SELECT BusinessEntityID, TerritoryID
,DATEPART(yy,ModifiedDate) AS SalesYear
,CONVERT(varchar(20),SalesYTD,1) AS SalesYTD
,CONVERT(varchar(20),AVG(SalesYTD) OVER (ORDER BY DATEPART(yy,ModifiedDate)
),1) AS MovingAvg
,CONVERT(varchar(20),SUM(SalesYTD) OVER (ORDER BY DATEPART(yy,ModifiedDate)
),1) AS CumulativeTotal
FROM Sales.SalesPerson
WHERE TerritoryID IS NULL OR TerritoryID < 5
ORDER BY SalesYear; Here is the result set.
BusinessEntityID TerritoryID SalesYear SalesYTD MovingAvg CumulativeTotal
---------------- ----------- ----------- -------------------- -------------------- --------------------
274 NULL 2005 559,697.56 2,449,684.05 17,147,788.35
275 2 2005 3,763,178.18 2,449,684.05 17,147,788.35
276 4 2005 4,251,368.55 2,449,684.05 17,147,788.35
277 3 2005 3,189,418.37 2,449,684.05 17,147,788.35
280 1 2005 1,352,577.13 2,449,684.05 17,147,788.35
281 4 2005 2,458,535.62 2,449,684.05 17,147,788.35
283 1 2005 1,573,012.94 2,449,684.05 17,147,788.35
284 1 2006 1,576,562.20 2,138,250.72 19,244,256.47
287 NULL 2006 519,905.93 2,138,250.72 19,244,256.47
285 NULL 2007 172,524.45 1,941,678.09 19,416,780.93
(10 row(s) affected) Applies to: SQL Server 2012 (11.x) through SQL Server 2012 (11.x) SQL Server 2017
The following example uses the ROWS clause to define a window over which the rows are computed as the current row and the N number of rows that follow (1 row in this example).
SELECT BusinessEntityID, TerritoryID
,CONVERT(varchar(20),SalesYTD,1) AS SalesYTD
,DATEPART(yy,ModifiedDate) AS SalesYear
,CONVERT(varchar(20),SUM(SalesYTD) OVER (PARTITION BY TerritoryID
ORDER BY DATEPART(yy,ModifiedDate)
ROWS BETWEEN CURRENT ROW AND 1 FOLLOWING ),1) AS CumulativeTotal
FROM Sales.SalesPerson
WHERE TerritoryID IS NULL OR TerritoryID < 5; Here is the result set.
BusinessEntityID TerritoryID SalesYTD SalesYear CumulativeTotal
---------------- ----------- -------------------- ----------- --------------------
274 NULL 559,697.56 2005 1,079,603.50
287 NULL 519,905.93 2006 692,430.38
285 NULL 172,524.45 2007 172,524.45
283 1 1,573,012.94 2005 2,925,590.07
280 1 1,352,577.13 2005 2,929,139.33
284 1 1,576,562.20 2006 1,576,562.20
275 2 3,763,178.18 2005 3,763,178.18
277 3 3,189,418.37 2005 3,189,418.37
276 4 4,251,368.55 2005 6,709,904.17
281 4 2,458,535.62 2005 2,458,535.62 In the following example, the ROWS clause is specified with UNBOUNDED PRECEDING. The result is that the window starts at the first row of the partition.
SELECT BusinessEntityID, TerritoryID
,CONVERT(varchar(20),SalesYTD,1) AS SalesYTD
,DATEPART(yy,ModifiedDate) AS SalesYear
,CONVERT(varchar(20),SUM(SalesYTD) OVER (PARTITION BY TerritoryID
ORDER BY DATEPART(yy,ModifiedDate)
ROWS UNBOUNDED PRECEDING),1) AS CumulativeTotal
FROM Sales.SalesPerson
WHERE TerritoryID IS NULL OR TerritoryID < 5; Here is the result set.
BusinessEntityID TerritoryID SalesYTD SalesYear CumulativeTotal
---------------- ----------- -------------------- ----------- --------------------
274 NULL 559,697.56 2005 559,697.56
287 NULL 519,905.93 2006 1,079,603.50
285 NULL 172,524.45 2007 1,252,127.95
283 1 1,573,012.94 2005 1,573,012.94
280 1 1,352,577.13 2005 2,925,590.07
284 1 1,576,562.20 2006 4,502,152.27
275 2 3,763,178.18 2005 3,763,178.18
277 3 3,189,418.37 2005 3,189,418.37
276 4 4,251,368.55 2005 4,251,368.55
281 4 2,458,535.62 2005 6,709,904.17
The following example returns the ROW_NUMBER for sales representatives based on their assigned sales quota.
-- Uses AdventureWorks
SELECT ROW_NUMBER() OVER(ORDER BY SUM(SalesAmountQuota) DESC) AS RowNumber,
FirstName, LastName,
CONVERT(varchar(13), SUM(SalesAmountQuota),1) AS SalesQuota
FROM dbo.DimEmployee AS e
INNER JOIN dbo.FactSalesQuota AS sq
ON e.EmployeeKey = sq.EmployeeKey
WHERE e.SalesPersonFlag = 1
GROUP BY LastName, FirstName; Here is a partial result set.
RowNumber FirstName LastName SalesQuota
--------- --------- ------------------ -------------
1 Jillian Carson 12,198,000.00
2 Linda Mitchell 11,786,000.00
3 Michael Blythe 11,162,000.00
4 Jae Pak 10,514,000.00 The following examples show using the OVER clause with aggregate functions. In this example, using the OVER clause is more efficient than using subqueries.
-- Uses AdventureWorks
SELECT SalesOrderNumber AS OrderNumber, ProductKey,
OrderQuantity AS Qty,
SUM(OrderQuantity) OVER(PARTITION BY SalesOrderNumber) AS Total,
AVG(OrderQuantity) OVER(PARTITION BY SalesOrderNumber) AS Avg,
COUNT(OrderQuantity) OVER(PARTITION BY SalesOrderNumber) AS Count,
MIN(OrderQuantity) OVER(PARTITION BY SalesOrderNumber) AS Min,
MAX(OrderQuantity) OVER(PARTITION BY SalesOrderNumber) AS Max
FROM dbo.FactResellerSales
WHERE SalesOrderNumber IN(N'SO43659',N'SO43664') AND
ProductKey LIKE '2%'
ORDER BY SalesOrderNumber,ProductKey; Here is the result set.
OrderNumber Product Qty Total Avg Count Min Max
----------- ------- --- ----- --- ----- --- ---
SO43659 218 6 16 3 5 1 6
SO43659 220 4 16 3 5 1 6
SO43659 223 2 16 3 5 1 6
SO43659 229 3 16 3 5 1 6
SO43659 235 1 16 3 5 1 6
SO43664 229 1 2 1 2 1 1
SO43664 235 1 2 1 2 1 1 The following example shows using the OVER clause with an aggregate function in a calculated value. Notice that the aggregates are calculated by SalesOrderNumber and the percentage of the total sales order is calculated for each line of each SalesOrderNumber.
-- Uses AdventureWorks
SELECT SalesOrderNumber AS OrderNumber, ProductKey AS Product,
OrderQuantity AS Qty,
SUM(OrderQuantity) OVER(PARTITION BY SalesOrderNumber) AS Total,
CAST(1. * OrderQuantity / SUM(OrderQuantity)
OVER(PARTITION BY SalesOrderNumber)
*100 AS DECIMAL(5,2)) AS PctByProduct
FROM dbo.FactResellerSales
WHERE SalesOrderNumber IN(N'SO43659',N'SO43664') AND
ProductKey LIKE '2%'
ORDER BY SalesOrderNumber,ProductKey; The first start of this result set is:
OrderNumber Product Qty Total PctByProduct
----------- ------- --- ----- ------------
SO43659 218 6 16 37.50
SO43659 220 4 16 25.00
SO43659 223 2 16 12.50
SO43659 229 2 16 18.75 Aggregate Functions (Transact-SQL)
Analytic Functions (Transact-SQL)
Excellent blog post about window functions and OVER, on sqlmag.com, by Itzik Ben-Gan