The ALTER TABLE statement modifies a table.
ALTER TABLE tableName { ADD COLUMN columnDefinition | ADD CONSTRAINT clause | DROP [ COLUMN ] columnName [ CASCADE | RESTRICT ] | DROP { PRIMARY KEY | FOREIGN KEY constraintName | UNIQUE constraintName | CHECK constraintName | CONSTRAINT constraintName } | ALTER [ COLUMN ] columnAlteration | LOCKSIZE { ROW | TABLE } }
simpleColumnName [ dataType ] [ columnLevelConstraint ]* [ [ WITH ] DEFAULT defaultConstantExpression | generationClause ]
The syntax for the columnDefinition for a new column is a subset of the syntax for a column in a CREATE TABLE statement.
The syntax of DataType is described in Data types. The DataType can be omitted only if you specify a generationClause. If you omit the DataType, the type of the generated column is the type of the generationClause. If you specify both a DataType and a generationClause, the type of the generationClause must be assignable to DataType.
For details on defaultConstantExpression, see Column default.
columnName SET DATA TYPE BLOB( integer ) | columnName SET DATA TYPE CLOB( integer ) | columnName SET DATA TYPE VARCHAR( integer ) | columnName SET DATA TYPE VARCHAR( integer ) FOR BIT DATA | columnName SET INCREMENT BY integerConstant | columnName RESTART WITH integerConstant | columnName { SET | DROP } NOT NULL | columnName [ NOT ] NULL | columnName [ WITH | SET ] DEFAULT defaultValue | columnName DROP DEFAULT
In the columnAlteration, SET INCREMENT BY integerConstant specifies the interval between consecutive values of the identity column. The next value to be generated for the identity column will be determined from the last assigned value with the increment applied. The column must already be defined with the IDENTITY attribute.
CREATE TABLE tauto(i INT GENERATED BY DEFAULT AS IDENTITY, k INT) CREATE UNIQUE INDEX tautoInd ON tauto(i) INSERT INTO tauto(k) values 1,2
INSERT INTO tauto VALUES (3,3) INSERT INTO tauto VALUES (4,4) INSERT INTO tauto VALUES (5,5)
ALTER TABLE tauto ALTER COLUMN i RESTART WITH 6
ALTER TABLE does not affect any view that references the table being altered. This includes views that have an "*" in their SELECT list. You must drop and re-create those views if you wish them to return the new columns.
Derby raises an error if you try to change the DataType of a generated column to a type which is not assignable from the type of the generationClause. Derby also raises an error if you try to add a DEFAULT clause to a generated column.
The syntax for the columnDefinition for a new column is almost the same as for a column in a CREATE TABLE statement. This syntax allows a column constraint to be placed on the new column within the ALTER TABLE ADD COLUMN statement. However, a column with a NOT NULL constraint can be added to an existing table if you give a default value; otherwise, an exception is thrown when the ALTER TABLE statement is executed.
Just as in CREATE TABLE, if the column definition includes a primary key constraint, the column cannot contain null values, so the NOT NULL attribute must also be specified (SQLSTATE 42831).
If you add a generated column to a table, Derby computes the generated values for all existing rows in the table.
ALTER TABLE ADD COLUMN adds the new column at the end of the table row. If you need to change a column in a way not permitted by ALTER TABLE ALTER COLUMN (for example, if you need to change its data type), the only way to do so is to drop the column and add a new one, and this changes the ordering of the columns.
ALTER TABLE ADD UNIQUE or PRIMARY KEY provide a shorthand method of defining a primary key composed of a single column. If PRIMARY KEY is specified in the definition of column C, the effect is the same as if the PRIMARY KEY(C) clause were specified as a separate clause. The column cannot contain null values, so the NOT NULL attribute must also be specified.
For information on the syntax of constraints, see CONSTRAINT clause. Use the syntax for table-level constraint when adding a constraint with the ADD TABLE ADD CONSTRAINT syntax.
ALTER TABLE DROP COLUMN allows you to drop a column from a table.
The keyword COLUMN is optional.
The keywords CASCADE and RESTRICT are also optional. If you specify neither CASCADE nor RESTRICT, the default is CASCADE.
If you specify RESTRICT, then the column drop will be rejected if it would cause a dependent schema object to become invalid.
If you specify CASCADE, then the column drop should additionally drop other schema objects which have become invalid.
The schema objects which can cause a DROP COLUMN RESTRICT to be rejected include: views, triggers, primary key constraints, foreign key constraints, unique key constraints, check constraints, and column privileges. If one of these types of objects depends on the column being dropped, DROP COLUMN RESTRICT will reject the statement.
Derby also raises an error if you specify RESTRICT when you drop a column referenced by the generationClause of a generated column. However, if you specify CASCADE, the generated column is also dropped with CASCADE semantics.
You may not drop the last (only) column in a table.
CASCADE/RESTRICT doesn't consider whether the column being dropped is used in any indexes. When a column is dropped, it is removed from any indexes which contain it. If that column was the only column in the index, the entire index is dropped.
ALTER TABLE DROP CONSTRAINT drops a constraint on an existing table. To drop an unnamed constraint, you must specify the generated constraint name stored in SYS.SYSCONSTRAINTS as a delimited identifier.
Dropping a primary key, unique, or foreign key constraint drops the physical index that enforces the constraint (also known as a backing index).
To increase the width of a column of these types, specify the data type and new size after the column name.
You are not allowed to decrease the width or to change the data type. You are not allowed to increase the width of a column that is part of a primary or unique key referenced by a foreign key constraint or that is part of a foreign key constraint.
To set an interval between consecutive values of the identity column, specify the integerConstant. You must previously define the column with the IDENTITY attribute (SQLSTATE 42837). If there are existing rows in the table, the values in the column for which the SET INCREMENT default was added do not change.
You can add the NOT NULL constraint to an existing column. To do so there must not be existing NULL values for the column in the table.
You can remove the NOT NULL constraint from an existing column. To do so the column must not be used in a PRIMARY KEY constraint.
You can use DEFAULT default-value to change a column default. To disable a previously set default, use DROP DEFAULT (alternatively, you can specify NULL as the default-value).
You can specify a default value for a new column. A default value is the value that is inserted into a column if no other value is specified. If not explicitly specified, the default value of a column is NULL. If you add a default to a new column, existing rows in the table gain the default value in the new column.
For more information about defaults, see CREATE TABLE statement.
The LOCKSIZE clause allows you to override row-level locking for the specific table, if your system uses the default setting of row-level locking. (If your system is set for table-level locking, you cannot change the locking granularity to row-level locking, although Derby allows you to use the LOCKSIZE clause in such a situation without throwing an exception.) To override row-level locking for the specific table, set locking for the table to TABLE. If you created the table with table-level locking granularity, you can change locking back to ROW with the LOCKSIZE clause in the ALTER TABLE STATEMENT. For information about why this is sometimes useful, see Tuning Derby.
-- Add a new column with a column-level constraint -- to an existing table -- An exception will be thrown if the table -- contains any rows -- since the newcol will be initialized to NULL -- in all existing rows in the table ALTER TABLE CITIES ADD COLUMN REGION VARCHAR(26) CONSTRAINT NEW_CONSTRAINT CHECK (REGION IS NOT NULL); -- Add a new unique constraint to an existing table -- An exception will be thrown if duplicate keys are found ALTER TABLE SAMP.DEPARTMENT ADD CONSTRAINT NEW_UNIQUE UNIQUE (DEPTNO); -- add a new foreign key constraint to the -- Cities table. Each row in Cities is checked -- to make sure it satisfied the constraints. -- if any rows don't satisfy the constraint, the -- constraint is not added ALTER TABLE CITIES ADD CONSTRAINT COUNTRY_FK Foreign Key (COUNTRY) REFERENCES COUNTRIES (COUNTRY); -- Add a primary key constraint to a table -- First, create a new table CREATE TABLE ACTIVITIES (CITY_ID INT NOT NULL, SEASON CHAR(2), ACTIVITY VARCHAR(32) NOT NULL); -- You will not be able to add this constraint if the -- columns you are including in the primary key have -- null data or duplicate values. ALTER TABLE Activities ADD PRIMARY KEY (city_id, activity); -- Drop the city_id column if there are no dependent objects: ALTER TABLE Cities DROP COLUMN city_id RESTRICT; -- Drop the city_id column, also dropping all dependent objects: ALTER TABLE Cities DROP COLUMN city_id CASCADE; -- Drop a primary key constraint from the CITIES table ALTER TABLE Cities DROP CONSTRAINT Cities_PK; -- Drop a foreign key constraint from the CITIES table ALTER TABLE Cities DROP CONSTRAINT COUNTRIES_FK; -- add a DEPTNO column with a default value of 1 ALTER TABLE SAMP.EMP_ACT ADD COLUMN DEPTNO INT DEFAULT 1; -- increase the width of a VARCHAR column ALTER TABLE SAMP.EMP_PHOTO ALTER PHOTO_FORMAT SET DATA TYPE VARCHAR(30); -- change the lock granularity of a table ALTER TABLE SAMP.SALES LOCKSIZE TABLE; -- Remove the NOT NULL constraint from the MANAGER column ALTER TABLE Employees ALTER COLUMN Manager NULL; -- Add the NOT NULL constraint to the SSN column ALTER TABLE Employees ALTER COLUMN ssn NOT NULL; -- Change the default value for the SALARY column ALTER TABLE Employees ALTER COLUMN Salary DEFAULT 1000.0 ALTER TABLE Employees ALTER COLUMN Salary DROP DEFAULT
An ALTER TABLE statement causes all statements that are dependent on the table being altered to be recompiled before their next execution. ALTER TABLE is not allowed if there are any open cursors that reference the table being altered.