PostgreSQL UNION ALL

Updated May 3, 2023

Introduction to PostgreSQL UNION ALL

The PostgreSQL UNION ALL operator provides the combined result sets of more than one SELECT statement. The result contains all rows from the query without removing duplicate rows between multiple SELECT statements. Generally, the UNION operator is used in the reporting or data warehouse system to combine the rows from similar tables that are not normalized.

Syntax:

SELECT
expression_1,
expression_2
…
expression_n
FROM
table_name_1
[WHERE conditions]
UNION ALL
SELECT
expression_1,
expression_2
…
expression_n
FROM
table_name_2;
[WHERE conditions]

Explanation:

• It defines the list of column names or calculations we want to fetch using expression_1, expression_2 … expression_n, etc.
• Also, it defines the name of the tables from which you want to fetch the record. There should be at least one table name specified in the FROM clause.
• The WHERE clause is optional; you can specify the conditions that should be met for selecting a record.
• Define the UNION ALL keywords between the two SELECT statements.

How UNION ALL clause works in PostgreSQL?

• The result of the SELECT statements defined in UNION ALL combined.
• UNION ALL does not remove duplicate rows from the result set; if the SELECT statement contains duplicate rows in the result set, the UNION ALL operator considers the same row while combining the result set.
• If you want to sort the result set, specify the ORDER BY clause at the end of the query.

Examples of PostgreSQL UNION ALL

Let us consider the following table structure to understand the examples of the UNION ALL operator.

Profit2020Q1: This table stores the profit of quarter 1 for the year 2020 for an organization.

Profit2020Q2: This table stores the profit of quarter 2 for the year 2020 for an organization.

CREATE TABLE Profit2020Q1(
unit_name VARCHAR (50) NOT NULL,
total_amount numeric(10,2)
);
CREATE TABLE Profit2020Q2(
unit_name VARCHAR (50) NOT NULL,
total_amount numeric(10,2)
);

The above statements will create both Profit2020Q1and Profit2020Q2 tables; now insert some data in the above table as follows:

INSERT INTO Profit2020Q1 (unit_name, total_amount)
values
('Production', 10025530.23),
('Marketing', 20025530.35),
('R&D', 30025530.21),
('HR', 24325530.35),
('Accounting', 76025530.35),
('Finance', 89025530.35),
('Purchasing', 33325530.35);

INSERT INTO Profit2020Q2 (unit_name, total_amount)
values
('Production', 20025530.23),
('Marketing', 30025530.35),
('R&D', 40025530.21),
('HR', 34325530.35),
('Accounting', 86025530.35),
('Finance', 99025530.35),
('Purchasing', 33325530.35);

Now verify the inserted data by retrieving the data from both of the tables as follows:

select * from Profit2020Q1 ;

select * from Profit2020Q2 ;

Example #1: UNION ALL

SELECT *
FROM
Profit2020Q1
UNION ALL
SELECT *
FROM
Profit2020Q2;

SELECT *
FROM
Profit2020Q1
UNION ALL
SELECT *
FROM
Profit2020Q2
ORDER BY
unit_name ASC,
total_amount DESC;