Join the MemSQL Community Today
Get expert advice, develop skills, and connect with others.

REPLACE

Replaces all occurrences of a substring by another string.

Syntax

REPLACE(string, old_string, new_string);

Arguments

  • string: any string
  • old_string: the string to be replaced
  • new_string: the replacement string

Remarks

  • MemSQL uses case-sensitive match to search for the new_string

Return Type

String

Examples

The example below replaces a string in the name column of the company table using the REPLACE function.

SELECT name FROM company WHERE name LIKE "Zumiez Inc.";
****
+------------------------------+
| NAME                         |
+------------------------------+
| Zumiez Inc.                  |
+------------------------------+
1 row in set (242 ms)
SELECT REPLACE(name,'Inc.','Corp.') FROM company WHERE name LIKE "Zumiez Inc.";
****
+------------------------------+
| REPLACE(name,'Inc.','Corp.') |
+------------------------------+
| Zumiez Corp.                 |
+------------------------------+
1 row in set (1.34 sec)

The REPLACE function can also be grouped with other string functions. An example of combining REPLACE with GROUP_CONCAT function is shown below.

SELECT * FROM Emp;
****
+-------------------------------------------------------------+
| Name                         | City                         |
+-------------------------------------------------------------+
| Adam                         | Chicago                      |
+-------------------------------------------------------------+
1 row in set (1.05 sec)
SELECT GROUP_CONCAT(Name, '-', City) FROM Emp;
****
+------------------------------+
| GROUP_CONCAT(Name, '-', Ci...|
+------------------------------+
| Adam-Chicago                 |
+------------------------------+
1 row in set (991 ms)
SELECT GROUP_CONCAT(REPLACE(Name, 'Ad', 'S'),'-', City) FROM Emp;
****
+------------------------------+
| GROUP_CONCAT(REPLACE(Name,...|
+------------------------------+
| Sam-Chicago                  |
+------------------------------+
1 row in set (960 ms)