Редагувати

Поділитися через


LAG (Transact-SQL)

Applies to: SQL Server Azure SQL Database Azure SQL Managed Instance Azure Synapse Analytics Analytics Platform System (PDW) SQL analytics endpoint in Microsoft Fabric Warehouse in Microsoft Fabric

Accesses data from a previous row in the same result set without the use of a self-join starting with SQL Server 2012 (11.x). LAG provides access to a row at a given physical offset that comes before the current row. Use this analytic function in a SELECT statement to compare values in the current row with values in a previous row.

Transact-SQL syntax conventions

Syntax

LAG (scalar_expression [ , offset ] [ , default ] ) [ IGNORE NULLS | RESPECT NULLS ]
    OVER ( [ partition_by_clause ] order_by_clause )  

Arguments

scalar_expression

The value to be returned based on the specified offset. It is an expression of any type that returns a single (scalar) value. scalar_expression cannot be an analytic function.

offset
The number of rows back from the current row from which to obtain a value. If not specified, the default is 1. offset can be a column, subquery, or other expression that evaluates to a positive integer or can be implicitly converted to bigint. offset cannot be a negative value or an analytic function.

default
The value to return when offset is beyond the scope of the partition. If a default value is not specified, NULL is returned. default can be a column, subquery, or other expression, but it cannot be an analytic function. default must be type-compatible with scalar_expression.

[ IGNORE NULLS | RESPECT NULLS ]

Applies to: SQL Server (starting with SQL Server 2022 (16.x)), Azure SQL Database, Azure SQL Managed Instance, Azure SQL Edge

IGNORE NULLS - Ignore null values in the dataset when computing the first value over a partition.

RESPECT NULLS - Respect null values in the dataset when computing first value over a partition. RESPECT NULLS is the default behavior if a NULLS option is not specified.

There was a bug fix in SQL Server 2022 CU4 related to IGNORE NULLS in LAG and LEAD.

For more information on this argument in Azure SQL Edge, see Imputing missing values.

OVER ( [ partition_by_clause ] order_by_clause )

partition_by_clause divides the result set produced by the FROM clause into partitions to which the function is applied. If not specified, the function treats all rows of the query result set as a single group. order_by_clause determines the order of the data before the function is applied. If partition_by_clause is specified, it determines the order of the data in the partition. The order_by_clause is required. For more information, see OVER Clause (Transact-SQL).

Return Types

The data type of the specified scalar_expression. NULL is returned if scalar_expression is nullable or default is set to NULL.

General Remarks

LAG is nondeterministic. For more information, see Deterministic and Nondeterministic Functions.

Examples

A. Compare values between years

The following example uses the LAG function to return the difference in sales quotas for a specific employee over previous years. Notice that because there is no lag value available for the first row, the default of zero (0) is returned.

USE AdventureWorks2022;  
GO  
SELECT BusinessEntityID, YEAR(QuotaDate) AS SalesYear, SalesQuota AS CurrentQuota,   
       LAG(SalesQuota, 1,0) OVER (ORDER BY YEAR(QuotaDate)) AS PreviousQuota  
FROM Sales.SalesPersonQuotaHistory  
WHERE BusinessEntityID = 275 AND YEAR(QuotaDate) IN ('2005','2006');  

Here's the result set.

BusinessEntityID SalesYear   CurrentQuota          PreviousQuota  
---------------- ----------- --------------------- ---------------------  
275              2005        367000.00             0.00  
275              2005        556000.00             367000.00  
275              2006        502000.00             556000.00  
275              2006        550000.00             502000.00  
275              2006        1429000.00            550000.00  
275              2006        1324000.00            1429000.00  
  

B. Compare values within partitions

The following example uses the LAG function to compare year-to-date sales between employees. The PARTITION BY clause is specified to divide the rows in the result set by sales territory. The LAG function is applied to each partition separately and computation restarts for each partition. The ORDER BY clause in the OVER clause orders the rows in each partition. The ORDER BY clause in the SELECT statement sorts the rows in the whole result set. Notice that because there is no lag value available for the first row of each partition, the default of zero (0) is returned.

USE AdventureWorks2022;  
GO  
SELECT TerritoryName, BusinessEntityID, SalesYTD,   
       LAG (SalesYTD, 1, 0) OVER (PARTITION BY TerritoryName ORDER BY SalesYTD DESC) AS PrevRepSales  
FROM Sales.vSalesPerson  
WHERE TerritoryName IN (N'Northwest', N'Canada')   
ORDER BY TerritoryName;  

Here's the result set.

TerritoryName            BusinessEntityID SalesYTD              PrevRepSales  
-----------------------  ---------------- --------------------- ---------------------  
Canada                   282              2604540.7172          0.00  
Canada                   278              1453719.4653          2604540.7172  
Northwest                284              1576562.1966          0.00  
Northwest                283              1573012.9383          1576562.1966  
Northwest                280              1352577.1325          1573012.9383  
  

C. Specifying arbitrary expressions

The following example demonstrates specifying various arbitrary expressions and ignoring NULL values in the LAG function syntax.

CREATE TABLE T (a INT, b INT, c INT);   
GO  
INSERT INTO T VALUES (1, 1, -3), (2, 2, 4), (3, 1, NULL), (4, 3, 1), (5, 2, NULL), (6, 1, 5);   
  
SELECT b, c,   
    LAG(2*c, b*(SELECT MIN(b) FROM T), -c/2.0) IGNORE NULLS OVER (ORDER BY a) AS i  
FROM T;  

Here's the result set.

b           c           i  
----------- ----------- -----------  
1           -3          1  
2           4           -2  
1           NULL        8  
3           1           -6  
2           NULL        8  
1           5           2  

D. Use IGNORE NULLS to find non-NULL values

The following sample query demonstrates using the IGNORE NULLS argument.

The IGNORE NULLS argument is used with both LAG and LEAD to demonstrate substitution of NULL values for preceding or next non-NULL values.

  • If the preceding row contained NULL with LAG, then the current row uses the most recent non-NULL value.
  • If the next row contains a NULL with LEAD, then the current row uses the next available non-NULL value.
DROP TABLE IF EXISTS #test_ignore_nulls;
CREATE TABLE #test_ignore_nulls (column_a int, column_b int);
GO

INSERT INTO #test_ignore_nulls VALUES
    (1, 8),
    (2, 9),
    (3, NULL),
    (4, 10),
    (5, NULL),
    (6, NULL),
    (7, 11);

SELECT column_a, column_b,
      [Previous value for column_b] = LAG(column_b) IGNORE NULLS OVER (ORDER BY column_a),
      [Next value for column_b] = LEAD(column_b) IGNORE NULLS OVER (ORDER BY column_a)
FROM #test_ignore_nulls
ORDER BY column_a;

--cleanup
DROP TABLE #test_ignore_nulls;
column_a     column_b    Previous value for column_b    Next value for column_b
1            8           NULL                           9
2            9           8                              10
3            NULL        9                              10
4            10          9                              11
5            NULL        10                             11
6            NULL        10                             11
7            11          10                             NULL

E. Use RESPECT NULLS to keep NULL values

The following sample query demonstrates using the RESPECT NULLS argument, which is the default behavior if not specified, as opposed to the IGNORE NULLS argument in the previous example.

  • If the preceding row contained NULL with LAG, then the current row uses the most recent value.
  • If the next row contains a NULL with LEAD, then the current row uses the next value.
DROP TABLE IF EXISTS #test_ignore_nulls;
CREATE TABLE #test_ignore_nulls (column_a int, column_b int);
GO

INSERT INTO #test_ignore_nulls VALUES
    (1, 8),
    (2, 9),
    (3, NULL),
    (4, 10),
    (5, NULL),
    (6, NULL),
    (7, 11);

SELECT column_a, column_b,
      [Previous value for column_b] = LAG(column_b) RESPECT NULLS OVER (ORDER BY column_a),
      [Next value for column_b] = LEAD(column_b) RESPECT NULLS OVER (ORDER BY column_a)
FROM #test_ignore_nulls
ORDER BY column_a;

--Identical output
SELECT column_a, column_b,
      [Previous value for column_b] = LAG(column_b)  OVER (ORDER BY column_a),
      [Next value for column_b] = LEAD(column_b)  OVER (ORDER BY column_a)
FROM #test_ignore_nulls
ORDER BY column_a;

--cleanup
DROP TABLE #test_ignore_nulls;
column_a     column_b    Previous value for column_b    Next value for column_b
1            8           NULL                           9
2            9           8                              NULL
3            NULL        9                              10
4            10          NULL                           NULL
5            NULL        10                             NULL
6            NULL        NULL                           11
7            11          NULL                           NULL

Examples: Azure Synapse Analytics and Analytics Platform System (PDW)

A. Compare values between quarters

The following example demonstrates the LAG function. The query uses the LAG function to return the difference in sales quotas for a specific employee over previous calendar quarters. Notice that because there is no lag value available for the first row, the default of zero (0) is returned.

-- Uses AdventureWorks  
  
SELECT CalendarYear, CalendarQuarter, SalesAmountQuota AS SalesQuota,  
       LAG(SalesAmountQuota,1,0) OVER (ORDER BY CalendarYear, CalendarQuarter) AS PrevQuota,  
       SalesAmountQuota - LAG(SalesAmountQuota,1,0) OVER (ORDER BY CalendarYear, CalendarQuarter) AS Diff  
FROM dbo.FactSalesQuota  
WHERE EmployeeKey = 272 AND CalendarYear IN (2001, 2002)  
ORDER BY CalendarYear, CalendarQuarter;   

Here's the result set.

Year Quarter  SalesQuota  PrevQuota  Diff  
---- -------  ----------  ---------  -------------  
2001 3        28000.0000      0.0000   28000.0000  
2001 4         7000.0000  28000.0000  -21000.0000  
2001 1        91000.0000   7000.0000   84000.0000  
2002 2       140000.0000  91000.0000   49000.0000  
2002 3         7000.0000 140000.0000  -70000.0000  
2002 4       154000.0000   7000.0000   84000.0000

Next steps