แก้ไข

แชร์ผ่าน

Use the inserted and deleted tables

  • บทความ

Applies to: SQL Server Azure SQL Database Azure SQL Managed Instance

DML trigger statements use two special tables: the deleted and inserted tables. SQL Server automatically creates and manages these tables. You can use these temporary, memory-resident tables to test the effects of certain data modifications and to set conditions for DML trigger actions. You cannot directly modify the data in the tables or perform data definition language (DDL) operations on the tables, such as CREATE INDEX.

Understanding the inserted and deleted tables

In DML triggers, the inserted and deleted tables are primarily used to perform the following:

  • Extend referential integrity between tables.

  • Insert or update data in base tables underlying a view.

  • Test for errors and take action based on the error.

  • Find the difference between the state of a table before and after a data modification and take actions based on that difference.

The deleted table stores copies of the affected rows in the trigger table before they were changed by a DELETE or UPDATE statement (the trigger table is the table on which the DML trigger runs). During the execution of a DELETE or UPDATE statement, the affected rows are first copied from the trigger table and transferred to the deleted table.

The inserted table stores copies of the new or changed rows after an INSERT or UPDATE statement. During the execution of an INSERT or UPDATE statement, the new or changed rows in the trigger table are copied to the inserted table. The rows in the inserted table are copies of the new or updated rows in the trigger table.

An update transaction is similar to a delete operation followed by an insert operation. During the execution of an UPDATE statement, the following sequence of events occurs:

  1. The original row is copied from the trigger table to the deleted table.
  2. The trigger table is updated with the new values from the UPDATE statement.
  3. The updated row in the trigger table is copied to the inserted table.

This allows you to compare the contents of the row before the update (in the deleted table) with the new row values after the update (in the inserted table).

When you set trigger conditions, use the inserted and deleted tables appropriately for the action that fired the trigger. Although referencing the deleted table when testing an INSERT or the inserted table when testing a DELETE does not cause any errors, these trigger test tables do not contain any rows in these cases.

Note

If trigger actions depend on the number of rows a data modification effects, use tests (such as an examination of @@ROWCOUNT) for multirow data modifications (an INSERT, DELETE, or UPDATE based on a SELECT statement), and take appropriate actions. For more information, see Create DML Triggers to Handle Multiple Rows of Data.

SQL Server does not allow for text, ntext, or image column references in the inserted and deleted tables for AFTER triggers. However, these data types are included for backward compatibility purposes only. The preferred storage for large data is to use the varchar(max), nvarchar(max), and varbinary(max) data types. Both AFTER and INSTEAD OF triggers support varchar(max), nvarchar(max), and varbinary(max) data in the inserted and deleted tables. For more information, see CREATE TRIGGER (Transact-SQL).

Example: Use the inserted table in a trigger to enforce business rules

Because CHECK constraints can reference only the columns on which the column-level or table-level constraint is defined, any cross-table constraints (in this case, business rules) must be defined as triggers.

The following example creates a DML trigger. This trigger checks to make sure the credit rating for the vendor is good when an attempt is made to insert a new purchase order into the PurchaseOrderHeader table. To obtain the credit rating of the vendor corresponding to the purchase order that was just inserted, the Vendor table must be referenced and joined with the inserted table. If the credit rating is too low, a message is displayed and the insertion does not execute.

USE AdventureWorks2022;
GO
IF OBJECT_ID ('Purchasing.LowCredit','TR') IS NOT NULL
   DROP TRIGGER Purchasing.LowCredit;
GO
-- This trigger prevents a row from being inserted in the Purchasing.PurchaseOrderHeader table
-- when the credit rating of the specified vendor is set to 5 (below average).  
  
CREATE TRIGGER Purchasing.LowCredit ON Purchasing.PurchaseOrderHeader  
AFTER INSERT  
AS  
IF (ROWCOUNT_BIG() = 0)
RETURN;
IF EXISTS (SELECT 1  
           FROM inserted AS i   
           JOIN Purchasing.Vendor AS v   
           ON v.BusinessEntityID = i.VendorID  
           WHERE v.CreditRating = 5  
          )  
BEGIN  
RAISERROR ('A vendor''s credit rating is too low to accept new  
purchase orders.', 16, 1);  
ROLLBACK TRANSACTION;  
RETURN   
END;  
GO  
  
-- This statement attempts to insert a row into the PurchaseOrderHeader table  
-- for a vendor that has a below average credit rating.  
-- The AFTER INSERT trigger is fired and the INSERT transaction is rolled back.  
  
INSERT INTO Purchasing.PurchaseOrderHeader (RevisionNumber, Status, EmployeeID,  
VendorID, ShipMethodID, OrderDate, ShipDate, SubTotal, TaxAmt, Freight)  
VALUES (  
2  
,3  
,261  
,1652  
,4  
,GETDATE()  
,GETDATE()  
,44594.55  
,3567.564  
,1114.8638 );  
GO

Use the inserted and deleted tables in INSTEAD OF triggers

The inserted and deleted tables passed to INSTEAD OF triggers defined on tables follow the same rules as the inserted and deleted tables passed to AFTER triggers. The format of the inserted and deleted tables is the same as the format of the table on which the INSTEAD OF trigger is defined. Each column in the inserted and deleted tables maps directly to a column in the base table.

The following rules regarding when an INSERT or UPDATE statement referencing a table with an INSTEAD OF trigger must supply values for columns are the same as if the table did not have an INSTEAD OF trigger:

  • Values cannot be specified for computed columns or columns with a timestamp data type.

  • Values cannot be specified for columns with an IDENTITY property, unless IDENTITY_INSERT is ON for that table. When IDENTITY_INSERT is ON, INSERT statements must supply a value.

  • INSERT statements must supply values for all NOT NULL columns that do not have DEFAULT constraints.

  • For any columns except computed, identity, or timestamp columns, values are optional for any column that allows nulls, or any NOT NULL column that has a DEFAULT definition.

When an INSERT, UPDATE, or DELETE statement references a view that has an INSTEAD OF trigger, the Database Engine calls the trigger instead of taking any direct action against any table. The trigger must use the information presented in the inserted and deleted tables to build any statements required to implement the requested action in the base tables, even when the format of the information in the inserted and deleted tables built for the view is different from the format of the data in the base tables.

The format of the inserted and deleted tables passed to an INSTEAD OF trigger defined on a view matches the select list of the SELECT statement defined for the view. For example:

USE AdventureWorks2022;  
GO  
CREATE VIEW dbo.EmployeeNames (BusinessEntityID, LName, FName)  
AS  
SELECT e.BusinessEntityID, p.LastName, p.FirstName  
FROM HumanResources.Employee AS e   
JOIN Person.Person AS p  
ON e.BusinessEntityID = p.BusinessEntityID;

The result set for this view has three columns: an int column and two nvarchar columns. The inserted and deleted tables passed to an INSTEAD OF trigger defined on the view also have an int column named BusinessEntityID, an nvarchar column named LName, and an nvarchar column named FName.

The select list of a view can also contain expressions that do not directly map to a single base-table column. Some view expressions, such as a constant or function invocation, may not reference any columns and can be ignored. Complex expressions can reference multiple columns, yet the inserted and deleted tables have only one value for each inserted row. The same issues apply to simple expressions in a view if they reference a computed column that has a complex expression. An INSTEAD OF trigger on the view must handle these types of expressions.

Performance considerations

Because the inserted and deleted tables are virtual, memory-resident tables, properties such as statistics or indexes are not available. Though some cardinality information is exposed from these tables, you should exercise care when considering the number of rows to be temporarily stored there. Inserting a large number of rows in these tables and querying or joining them with other tables may result in sub-optimial query plans and slow query executions. Be sure to carefully design and test your application to meet your query performance needs.

Next steps

For more information, see the overview of DML Triggers.