Merge (SQL)

A relational database management system uses SQL MERGE (also called upsert) statements to INSERT new records or UPDATE or DELETE existing records depending on whether condition matches. It was officially introduced in the SQL:2003 standard, and expanded in the SQL:2008 standard.

Usage

MERGE INTO tablename USING table_reference ON (condition)
  WHEN MATCHED THEN
    UPDATE SET column1 = value1 [, column2 = value2 ...]
  WHEN NOT MATCHED THEN
    INSERT (column1 [, column2 ...]) VALUES (value1 [, value2 ...]);

A right join is employed over the Target (the INTO table) and the Source (the USING table / view / sub-query)--where Target is the left table and Source is the right one. The four possible combinations yield these rules:

• If the ON field(s) in the Source matches the ON field(s) in the Target, then UPDATE
• If the ON field(s) in the Source does not match the ON field(s) in the Target, then INSERT
• If the ON field(s) does not exist in the Source but does exist in the Target, then no action is performed.
• If the ON field(s) does not exist in either the Source or Target, then no action is performed.

If multiple Source rows match a given Target row, an error is mandated by SQL:2003 standards. You cannot update a Target row multiple times with a MERGE statement

Implementations

Database management systems PostgreSQL, Oracle Database, IBM Db2, Teradata, EXASOL, Firebird, CUBRID, H2, HSQLDB, MS SQL, Vectorwise and Apache Derby support the standard syntax. Some also add non-standard SQL extensions.

Synonymous

Some database implementations adopted the term upsert (a portmanteau of update and insert) to a database statement, or combination of statements, that inserts a record to a table in a database if the record does not exist or, if the record already exists, updates the existing record. This synonym is used in PostgreSQL (v9.5+) and SQLite (v3.24+). It is also used to abbreviate the "MERGE" equivalent pseudo-code.

It is used in Microsoft Azure SQL Database.

Other non-standard implementations

Some other database management systems support this, or very similar behavior, through their own, non-standard SQL extensions.

MySQL, for example, supports the use of INSERT ... ON DUPLICATE KEY UPDATE syntax which can be used to achieve a similar effect with the limitation that the join between target and source has to be made only on PRIMARY KEY or UNIQUE constraints, which is not required in the ANSI/ISO standard. It also supports >REPLACE INTO syntax, which first attempts an insert, and if that fails, deletes the row, if exists, and then inserts the new one. There is also an IGNORE clause for the INSERT statement, which tells the server to ignore "duplicate key" errors and go on (existing rows will not be inserted or updated, but all new rows will be inserted).

SQLite's INSERT OR REPLACE INTO works similarly. It also supports REPLACE INTO as an alias for compatibility with MySQL.

Firebird supports MERGE INTO though fails to throw an error when there are multiple Source data rows. Additionally there is a single-row version, UPDATE OR INSERT INTO tablename (columns) VALUES (values) [MATCHING (columns)], but the latter does not give you the option to take different actions on insert versus update (e.g. setting a new sequence value only for new rows, not for existing ones.)

IBM Db2 extends the syntax with multiple WHEN MATCHED and WHEN NOT MATCHED clauses, distinguishing them with ... AND some-condition guards.

Microsoft SQL Server extends with supporting guards and also with supporting Left Join via WHEN NOT MATCHED BY SOURCE clauses.

PostgreSQL supports merge since version 15 but previously supported merging via INSERT INTO ... ON CONFLICT [ conflict_target ] conflict_action.

CUBRID supports MERGE INTO statement. And supports the use of INSERT ... ON DUPLICATE KEY UPDATE syntax. It also supports REPLACE INTO for compatibility with MySQL.

Apache Phoenix supports UPSERT VALUES and UPSERT SELECT syntax.

Spark SQL supports UPDATE SET * and INSERT * clauses in actions.

Apache Impala supports UPSERT INTO ... SELECT.

Usage in NoSQL

A similar concept is applied in some NoSQL databases.

In MongoDB the fields in a value associated with a key can be updated with an update operation. The update raises an error if the key is not found. In the update operation it is possible to set the upsert flag: in this case a new value is stored associated to the given key if it does not exist, otherwise the whole value is replaced.

In Redis the SET operations sets the value associated with a given key. Redis does not know any detail of the internal structure of the value, so an update would have no meaning. So the SET operation has always a set or replace semantics.

See also

• Join in particular:
  • Join (SQL)
  • join (Unix)

References

• Hsu, Leo; Obe, Regina (May 18, 2008). "Cross Compare of SQL Server, MySQL, and PostgreSQL". Postgres OnLine Journal. Retrieved 8 October 2010.
• Chodorow, Kristina; Mike Dirolf (September 2010). MongoDB: The Definitive Guide. O'Reilly. ISBN 978-1-449-38156-1.

External links

• Oracle 11g Release 2 documentation on MERGE
• Firebird 2.1 documentation on MERGE
• DB2 v9 MERGE statement
• Microsoft SQL Server documentation
• HSQLdb 2.0 Data Change Statements
• H2 (1.2) SQL Syntax page