In this article, we will be discussing about, the MERGE Statement in SQL Server. The MERGE Statement is a T-SQL enhancement, which was originally shipped with SQL Server 2008.
What is the MERGE Statement in SQL Server?
So what actually is the MERGE Statement? When and how should it be used?
The MERGE Statement actually allows joining a source with a target table and then based on the results of the join, it performs insert, update, or delete operations on the target table.
A simplified syntax example of the MERGE Statement is the following:
MERGE [tableA] as target using [tableB] as source on target.id = source.id
when matched then update set target.col1 = source.col1, target.col2 = source.col2
when not matched then insert values (source.id, source.col1, source.col2)
when not matched by source then delete;
So let’s analyze the above syntax. I will use some steps for that.
Step 1: Declare your source and target tables.
Step 2: Declare a joining condition on the two tables (in our example is the target.id = source.id condition).
Step 3: Set the action for the when matched case. This case means that a record was found that exists in both tables. To this end, you should perform an update in order to synchronize the two records.
Step 4: Set the action for the when not matched case. This case means that a record was found in the source table which does not exist in the target table. To this end, you should perform an insert in order to copy the record from the source to the target.
Step 5: Set the action for the when not matched by source case. This case means that a record was found in the target table which does not exist in the source table. To this end, you can delete this record (if the purpose of this operation is the synchronization of the two tables).
Please note that the MERGE statement should always end with a semicolon (;).
Strengthen you SQL Server Development Skills – Enroll to our Online Course!
Via the course, you will sharpen your SQL Server database programming skills via a large set of tips on T-SQL and database development techniques. The course, among other, features over than 30 live demonstrations!
(Lifetime Access/ Live Demos / Downloadable Resources and more!)
Artemakis Artemiou is a Senior SQL Server Architect, Author, a 9 Times Microsoft Data Platform MVP (2009-2018) and a Udemy Instructor. He has over 15 years of experience in the IT industry in various roles. Artemakis is the founder of SQLNetHub and TechHowTos.com. Artemakis is the creator of the well-known software tools Snippets Generator and DBA Security Advisor. Also, he is the author of many eBooks on SQL Server. Artemakis currently serves as the President of the Cyprus .NET User Group (CDNUG) and the International .NET Association Country Leader for Cyprus (INETA). Moreover, Artemakis teaches on Udemy, you can check his courses here.
Views: 1,659
1 thought on “The MERGE Statement in SQL Server”
Something I would also like to mention regarding the MERGE statement is that by adding in the statement the following code: OUTPUT $action, Inserted.*, Deleted.*; every time the MERGE statement is executed you will get information on which records were updated, inserted and deleted.
So based on this post's example the code can be modified as follows in order to include the OUTPUT clause:
————————— MERGE [tableA] as target using [tableB] as source on target.id = source.id when matched then update set target.col1 = source.col1, target.col2 = source.col2 when not matched then insert values (source.id, source.col1, source.col2) when not matched by source then delete OUTPUT $action, Inserted.*, Deleted.*; —————————
This site uses cookies for personalized content and the best possible experience. By continuing to browse this site, you agree to this use. Find out more.
Something I would also like to mention regarding the MERGE statement is that by adding in the statement the following code: OUTPUT $action, Inserted.*, Deleted.*; every time the MERGE statement is executed you will get information on which records were updated, inserted and deleted.
So based on this post's example the code can be modified as follows in order to include the OUTPUT clause:
—————————
MERGE [tableA] as target using [tableB] as source on target.id = source.id
when matched then update set target.col1 = source.col1, target.col2 = source.col2
when not matched then insert values (source.id, source.col1, source.col2)
when not matched by source then delete
OUTPUT $action, Inserted.*, Deleted.*;
—————————