Why set operations matter in PostgreSQL

Introduction

Set operations help combine or compare data from different tables easily. They let you find common, different, or all unique data quickly.

You want to find all customers who bought from either store A or store B.

You need to see which products are sold in both online and physical stores.

You want to list all unique email addresses from two mailing lists.

You want to find customers who bought from store A but not from store B.

Syntax

PostgreSQL

SELECT column_list FROM table1
UNION | UNION ALL | INTERSECT | EXCEPT
SELECT column_list FROM table2;

All SELECT statements must have the same number of columns and compatible data types.

UNION removes duplicates, UNION ALL keeps duplicates.

Examples

Finds all unique customer names from both stores.

PostgreSQL

SELECT name FROM customers_storeA
UNION
SELECT name FROM customers_storeB;

Finds products sold both online and in stores.

PostgreSQL

SELECT product_id FROM online_sales
INTERSECT
SELECT product_id FROM store_sales;

Finds emails in list1 but not in list2.

PostgreSQL

SELECT email FROM mailing_list1
EXCEPT
SELECT email FROM mailing_list2;

Sample Program

This query lists all unique customers from both stores, sorted by name.

PostgreSQL

CREATE TABLE storeA_customers (name TEXT);
CREATE TABLE storeB_customers (name TEXT);

INSERT INTO storeA_customers VALUES ('Alice'), ('Bob'), ('Charlie');
INSERT INTO storeB_customers VALUES ('Bob'), ('Diana');

SELECT name FROM storeA_customers
UNION
SELECT name FROM storeB_customers
ORDER BY name;

Important Notes

Set operations work best when columns match in type and order.

UNION removes duplicates, but UNION ALL is faster if duplicates are okay.

Use INTERSECT to find common data, EXCEPT to find differences.

Summary

Set operations combine or compare data from multiple queries.

They help find unique, common, or different data easily.

Remember to match columns and data types in all SELECTs.

Practice

1. What is the main purpose of set operations like UNION and INTERSECT in PostgreSQL?

easy

A. To delete rows from multiple tables

B. To create new tables from existing ones

C. To update data in a table

D. To combine or compare results from multiple SELECT queries

Solution

1. Step 1: Understand set operations

   Set operations like UNION and INTERSECT combine or compare results from multiple SELECT queries.

2. Step 2: Identify their purpose

   They help find unique, common, or different data across queries without modifying tables.

3. Final Answer:

   To combine or compare results from multiple SELECT queries -> Option D

4. Quick Check:

   Set operations = combine/compare SELECT results [OK]

Hint: Set operations combine query results, not modify tables [OK]

Common Mistakes:

• Thinking set operations modify tables
• Confusing set operations with DML commands
• Assuming set operations create new tables

2. Which of the following is the correct syntax to combine two SELECT queries using UNION in PostgreSQL?

easy

A. SELECT col1 FROM table1 JOIN SELECT col1 FROM table2;

B. SELECT col1 FROM table1 UNION SELECT col1 FROM table2;

C. SELECT col1 FROM table1 + SELECT col1 FROM table2;

D. SELECT col1 FROM table1 INTERSECT SELECT col1 FROM table2;

Solution

1. Step 1: Recall UNION syntax

   The correct syntax uses UNION keyword between two SELECT statements with matching columns.

2. Step 2: Check options

   Only SELECT col1 FROM table1 UNION SELECT col1 FROM table2; is correct. JOIN is for table joins; + is invalid syntax; INTERSECT is a different operation.

3. Final Answer:

   SELECT col1 FROM table1 UNION SELECT col1 FROM table2; -> Option B

4. Quick Check:

   UNION syntax = SELECT ... UNION SELECT ... [OK]

Hint: UNION joins SELECTs with UNION keyword, no + or JOIN [OK]

Common Mistakes:

• Using + instead of UNION
• Confusing JOIN with UNION
• Mixing INTERSECT with UNION syntax

3. Given these tables:
table1: (id) values (1), (2), (3);
table2: (id) values (2), (3), (4);
What is the result of this query?

SELECT id FROM table1 INTERSECT SELECT id FROM table2;

medium

A. [2, 3]

B. [1, 2, 3, 4]

C. [1]

D. [4]

Solution

1. Step 1: Understand INTERSECT operation

   INTERSECT returns only values present in both queries.

2. Step 2: Compare table1 and table2 ids

   table1 has 1,2,3; table2 has 2,3,4; common values are 2 and 3.

3. Final Answer:

   [2, 3] -> Option A

4. Quick Check:

   INTERSECT = common values [OK]

Hint: INTERSECT returns only common rows between queries [OK]

Common Mistakes:

• Including all values from both tables
• Confusing INTERSECT with UNION
• Returning values only from one table

4. Consider this query:

SELECT id FROM table1 UNION ALL SELECT id FROM table2;

It returns duplicate rows. How can you fix it to return only unique rows?

medium

A. Replace UNION ALL with UNION

B. Add DISTINCT inside each SELECT

C. Use INTERSECT instead of UNION ALL

D. Add GROUP BY id at the end

Solution

1. Step 1: Understand UNION ALL vs UNION

   UNION ALL returns all rows including duplicates; UNION removes duplicates.

2. Step 2: Fix query to remove duplicates

   Replacing UNION ALL with UNION removes duplicates automatically.

3. Final Answer:

   Replace UNION ALL with UNION -> Option A

4. Quick Check:

   UNION removes duplicates, UNION ALL does not [OK]

Hint: Use UNION (not UNION ALL) to remove duplicates [OK]

Common Mistakes:

• Using UNION ALL when duplicates are unwanted
• Adding DISTINCT inside each SELECT unnecessarily
• Using INTERSECT which returns only common rows

5. You want to find all unique customer IDs who made purchases in either 2022 or 2023 from two tables:
purchases_2022(customer_id) and purchases_2023(customer_id).
Which query correctly returns this list without duplicates?

hard

A. SELECT customer_id FROM purchases_2022 UNION ALL SELECT customer_id FROM purchases_2023;

B. SELECT customer_id FROM purchases_2022 INTERSECT SELECT customer_id FROM purchases_2023;

C. SELECT customer_id FROM purchases_2022 UNION SELECT customer_id FROM purchases_2023;

D. SELECT DISTINCT customer_id FROM purchases_2022, purchases_2023;

Solution

1. Step 1: Understand the goal

   We want all unique customer IDs from either year, combined without duplicates.

2. Step 2: Evaluate options

   SELECT customer_id FROM purchases_2022 INTERSECT SELECT customer_id FROM purchases_2023; returns only customers in both years (common). SELECT customer_id FROM purchases_2022 UNION SELECT customer_id FROM purchases_2023; returns all unique customers from both years. SELECT customer_id FROM purchases_2022 UNION ALL SELECT customer_id FROM purchases_2023; returns all including duplicates. SELECT DISTINCT customer_id FROM purchases_2022, purchases_2023; creates a cross join, not union.

3. Final Answer:

   SELECT customer_id FROM purchases_2022 UNION SELECT customer_id FROM purchases_2023; -> Option C

4. Quick Check:

   UNION = unique combined rows [OK]

Hint: Use UNION to combine unique rows from multiple tables [OK]

Common Mistakes:

• Using INTERSECT to get all customers
• Using UNION ALL causing duplicates
• Using cross join instead of union