An SQL UPDATE statement changes the data of one or more records in a table. Either all the rows can be updated, or a subset may be chosen using a condition.
The UPDATE
statement has the following form:[1]
UPDATE
table_name SET
column_name = value [, ''column_name'' = ''value ...''] ['''<code>WHERE</code>''' ''condition'']
For the UPDATE
to be successful, the user must have data manipulation privileges (UPDATE
privilege) on the table or column and the updated value must not conflict with all the applicable constraints (such as primary keys, unique indexes, CHECK
constraints, and NOT NULL
constraints).
In some databases, such as PostgreSQL, when a FROM clause is present, what essentially happens is that the target table is joined to the tables mentioned in the fromlist, and each output row of the join represents an update operation for the target table. When using FROM, one should ensure that the join produces at most one output row for each row to be modified. In other words, a target row shouldn't join to more than one row from the other table(s). If it does, then only one of the join rows will be used to update the target row, but which one will be used is not readily predictable.[2]
Because of this indeterminacy, referencing other tables only within sub-selects is safer, though often harder to read and slower than using a join.
MySQL does not conform to ANSI standard.[3]
Set the value of column C1 in table T to 1, only in those rows where the value of column C2 is "a".
In table T, set the value of column C1 to 9 and the value of C3 to 4 for all rows for which the value of column C2 is "a".
Increase value of column C1 by 1 if the value in column C2 is "a".
Prepend the value in column C1 with the string "text" if the value in column C2 is "a".
Set the value of column C1 in table T1 to 2, only if the value of column C2 is found in the sublist of values in column C3 in table T2 having the column C4 equal to 0.
Complex conditions and JOINs are also possible:
Some databases allow the non-standard use of the FROM clause:
Or on Oracle systems (assuming there is an index on classification.articleID):
UPDATE
statements to become an infinite loop when the [[where (SQL)|WHERE]]
clause and one or more SET
clauses may utilize an intertwined index.