Documentation Home
MySQL 5.7 Reference Manual
Related Documentation Download this Manual
PDF (US Ltr) - 37.2Mb
PDF (A4) - 37.2Mb
PDF (RPM) - 36.9Mb
EPUB - 10.5Mb
HTML Download (TGZ) - 10.3Mb
HTML Download (Zip) - 10.3Mb
HTML Download (RPM) - 8.9Mb
Eclipse Doc Plugin (TGZ) - 11.1Mb
Eclipse Doc Plugin (Zip) - 13.3Mb
Man Pages (TGZ) - 203.8Kb
Man Pages (Zip) - 309.1Kb
Info (Gzip) - 3.4Mb
Info (Zip) - 3.4Mb
Excerpts from this Manual

14.2.8 REPLACE Syntax

    [INTO] tbl_name
    [PARTITION (partition_name,...)]
    {VALUES | VALUE} ({expr | DEFAULT},...),(...),...


    [INTO] tbl_name
    [PARTITION (partition_name,...)]
    SET col_name={expr | DEFAULT}, ...


    [INTO] tbl_name
    [PARTITION (partition_name,...)]
    SELECT ...

REPLACE works exactly like INSERT, except that if an old row in the table has the same value as a new row for a PRIMARY KEY or a UNIQUE index, the old row is deleted before the new row is inserted. See Section 14.2.5, “INSERT Syntax”.

REPLACE is a MySQL extension to the SQL standard. It either inserts, or deletes and inserts. For another MySQL extension to standard SQL—that either inserts or updates—see Section, “INSERT ... ON DUPLICATE KEY UPDATE Syntax”.

DELAYED inserts and replaces were deprecated in MySQL 5.6.6. In MySQL 5.7, DELAYED is not supported. The server recognizes but ignores the DELAYED keyword, handles the replace as a nondelayed replace, and generates an ER_WARN_LEGACY_SYNTAX_CONVERTED warning. (REPLACE DELAYED is no longer supported. The statement was converted to REPLACE.) The DELAYED keyword will be removed in a future release.


REPLACE makes sense only if a table has a PRIMARY KEY or UNIQUE index. Otherwise, it becomes equivalent to INSERT, because there is no index to be used to determine whether a new row duplicates another.

Values for all columns are taken from the values specified in the REPLACE statement. Any missing columns are set to their default values, just as happens for INSERT. You cannot refer to values from the current row and use them in the new row. If you use an assignment such as SET col_name = col_name + 1, the reference to the column name on the right hand side is treated as DEFAULT(col_name), so the assignment is equivalent to SET col_name = DEFAULT(col_name) + 1.

If a generated column is replaced explicitly, the only permitted value is DEFAULT. For information about generated columns, see Section, “CREATE TABLE and Generated Columns”.

To use REPLACE, you must have both the INSERT and DELETE privileges for the table.

REPLACE supports explicit partition selection using the PARTITION keyword with a comma-separated list of names of partitions, subpartitions, or both. As with INSERT, if it is not possible to insert the new row into any of these partitions or subpartitions, the REPLACE statement fails with the error Found a row not matching the given partition set. See Section 21.5, “Partition Selection”, for more information.

The REPLACE statement returns a count to indicate the number of rows affected. This is the sum of the rows deleted and inserted. If the count is 1 for a single-row REPLACE, a row was inserted and no rows were deleted. If the count is greater than 1, one or more old rows were deleted before the new row was inserted. It is possible for a single row to replace more than one old row if the table contains multiple unique indexes and the new row duplicates values for different old rows in different unique indexes.

The affected-rows count makes it easy to determine whether REPLACE only added a row or whether it also replaced any rows: Check whether the count is 1 (added) or greater (replaced).

If you are using the C API, the affected-rows count can be obtained using the mysql_affected_rows() function.

You cannot replace into a table and select from the same table in a subquery.

MySQL uses the following algorithm for REPLACE (and LOAD DATA ... REPLACE):

  1. Try to insert the new row into the table

  2. While the insertion fails because a duplicate-key error occurs for a primary key or unique index:

    1. Delete from the table the conflicting row that has the duplicate key value

    2. Try again to insert the new row into the table

It is possible that in the case of a duplicate-key error, a storage engine may perform the REPLACE as an update rather than a delete plus insert, but the semantics are the same. There are no user-visible effects other than a possible difference in how the storage engine increments Handler_xxx status variables.

Because the results of REPLACE ... SELECT statements depend on the ordering of rows from the SELECT and this order cannot always be guaranteed, it is possible when logging these statements for the master and the slave to diverge. For this reason, REPLACE ... SELECT statements are flagged as unsafe for statement-based replication. With this change, such statements produce a warning in the log when using the STATEMENT binary logging mode, and are logged using the row-based format when using MIXED mode. See also Section, “Advantages and Disadvantages of Statement-Based and Row-Based Replication”.

When modifying an existing table that is not partitioned to accommodate partitioning, or, when modifying the partitioning of an already partitioned table, you may consider altering the table's primary key (see Section 21.6.1, “Partitioning Keys, Primary Keys, and Unique Keys”). You should be aware that, if you do this, the results of REPLACE statements may be affected, just as they would be if you modified the primary key of a nonpartitioned table. Consider the table created by the following CREATE TABLE statement:


When we create this table and run the statements shown in the mysql client, the result is as follows:

mysql> REPLACE INTO test VALUES (1, 'Old', '2014-08-20 18:47:00');
Query OK, 1 row affected (0.04 sec)

mysql> REPLACE INTO test VALUES (1, 'New', '2014-08-20 18:47:42');
Query OK, 2 rows affected (0.04 sec)

mysql> SELECT * FROM test;
| id | data | ts                  |
|  1 | New  | 2014-08-20 18:47:42 |
1 row in set (0.00 sec)

Now we create a second table almost identical to the first, except that the primary key now covers 2 columns, as shown here (emphasized text):

  PRIMARY KEY (id, ts)

When we run on test2 the same two REPLACE statements as we did on the original test table, we obtain a different result:

mysql> REPLACE INTO test2 VALUES (1, 'Old', '2014-08-20 18:47:00');
Query OK, 1 row affected (0.05 sec)

mysql> REPLACE INTO test2 VALUES (1, 'New', '2014-08-20 18:47:42');
Query OK, 1 row affected (0.06 sec)

mysql> SELECT * FROM test2;
| id | data | ts                  |
|  1 | Old  | 2014-08-20 18:47:00 |
|  1 | New  | 2014-08-20 18:47:42 |
2 rows in set (0.00 sec)

This is due to the fact that, when run on test2, both the id and ts column values must match those of an existing row for the row to be replaced; otherwise, a row is inserted.

In MySQL 5.7, a REPLACE statement affecting a partitioned table using a storage engine such as MyISAM that employs table-level locks locks only those partitions containing rows that match the REPLACE statement's WHERE clause, as long as none of the table's partitioning columns are updated; otherwise the entire table is locked. (For storage engines such as InnoDB that employ row-level locking, no locking of partitions takes place.) For more information, see Section 21.6.4, “Partitioning and Locking”.

User Comments
  Posted by no forms on October 28, 2004
Be careful with InnoDB tables and REPLACE:
If you run a replace on existing keys on table T, and table F references T with a forgein key constraint ON DELETE CASCADE, then table T will be updated - but table F will be emptied due to the DELETE before INSERT.

`id` int(10) unsigned NOT NULL auto_increment,

`foreign_id` int(10) unsigned NOT NULL,

<insert numbers 1..1000 into T>
<insert numbers 1..1000 into F>

=> T is updated
=> F is truncated - not the desired effect.

It is best to avoid REPLACE when working with constraints.
  Posted by R Lenard on January 12, 2005
Be careful with REPLACE INTO with server side prepared statements and the 3.1.6 driver - it doesn't support them. The 3.0.x driver did :-(
  Posted by Eric Stevens on February 3, 2005
Performance considerations:

Please note that REPLACE INTO is a much slower performer than an UPDATE statement. Keep in mind that a REPLACE INTO requires a test on the keys, and if a matching unique key is found on any or all columns, a DELETE FROM is executed, then an INSERT is executed. There's a lot of management of rows involved in this, and if you're doing it frequently, you'll hurt your performance unless you simply cannot do with any other syntax.

The only time when I can see where you'd actually need a REPLACE INTO is when you have multiple unique constraints on a table, and need to drop any rows that would match any of the constraints. Then REPLACE INTO becomes more efficient from DELETE FROM... INSERT INTO...

If you're looking at a single unique column table (Primary Key), please use UPDATE, or INSERT. Also, check out INSERT ... ON DUPLIATE KEY UPDATE... as an alternative if you're willing to stick to MySQL 4.1+
  Posted by Rolf Martin-Hoster on May 8, 2006
INNODB mysql 5.0 does not support "DELAYED" but does support LOW_PRIORITY :

mysql> REPLACE DELAYED INTO `online_users` SET `session_id`='3580cc4e61117c0785372c426eddd11c', `user_id` = 'XXX', `page` = '/', `lastview` = NOW();
ERROR 1031 (HY000): Table storage engine for 'online_users' doesn't have this option

  Posted by Atif Ghaffar on September 23, 2007
PLEASE Note that the REPLACE does a DELETE operation.

We did not realize this and had the triggers that should be triggered on DELETE triggered.

After checking all the code, we just found a script that does a replace to refresh the values of some fields.

We should have had used "insert into ... on duplicate update" syntax instead.

  Posted by J Mike on May 4, 2009
If you are using REPLACE INTO... triggers are fired in this order (if delete of duplcate key is used):
- before insert
- before delete
- after delete
- after insert

  Posted by Pablo Fernandez on September 7, 2009
This can also be used to merge databases

  Posted by Abidir Rokhman on February 13, 2011
i already move my blog to another domain and realize that all images on my post still pointing to old domain, so i need to find and replace all url on my post with my new domain. and i found this function useful to update just old domain to all my post.

i am using this query :

UPDATE wp_posts SET post_content = REPLACE(post_content, '', '');

full explanation here
  Posted by Meir Guttman on July 6, 2014
"no forms" tip also applies to a much simpler situation:

Simply having an Auto-Increment as a primary key will insert a new record with the same VALUES (...),(...); whenever the same "REPLACE INTO..." query is executed.

Records then are ADDED, not REPLACE(d)!
  Posted by Nathan Neulinger on April 18, 2015
Should note that this warning about inserts and nulls

"For multiple-row INSERT statements or INSERT INTO ... SELECT statements, the column is set to the implicit default value for the column data type. This is 0 for numeric types, the empty string ('') for string types, and the “zero” value for date and time types."

also appears to apply to a single row "replace into" query, which can be very confusing to debug when it appears to not obey the table constraints and just turns nulls/missing columns into empty strings. This can particularly be a problem if you have a unique constraint on one of those columns.
Sign Up Login You must be logged in to post a comment.