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.
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:
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.
ALTER TABLE author DROP CONSTRAINT author_pkey;
Build apps and APIs 10x faster
Built-in authorization and caching
8x more performant than hand-rolled APIs

