Primary Keys

A primary key is used to identify a row uniquely in a table. It could be a single column or a group of columns (composite). A primary key is one of the constraints that we define to restrict data input and validate.

Primary key is a combination of

• NOT NULL Constraint
• UNIQUE Constraint

Defining a primary key

As we looked at the constraints in the previous section, In the CREATE TABLE statement, you can define a primary key for a column by using the PRIMARY KEY keyword against the required unique column or a combination of columns.

Copy
CREATE TABLE author (
  id integer PRIMARY KEY,
  name text,
);

Here the id column is marked as the primary key.

Adding a primary key later

In case you had forgotten to mark any column as a primary key during table creation, we can also add a primary key later by altering the table using the following command:

Copy
ALTER TABLE author ADD PRIMARY KEY (id);

Removing a primary key

We can remove an existing primary key by using the DROP CONSTRAINT keyword in the alter table syntax.

Copy
ALTER TABLE author DROP CONSTRAINT author_pkey;