9.6 KiB
title | description | order | type | setup |
---|---|---|---|---|
Set Operator Queries | Learn how to use UNION, INTERSECT, and EXCEPT to combine results from multiple queries | 150 | lesson-challenge | ```sql CREATE TABLE fiction_book ( id INT PRIMARY KEY, title VARCHAR(255), author VARCHAR(255), price DECIMAL(10, 2) ); CREATE TABLE non_fiction_book ( id INT PRIMARY KEY, title VARCHAR(255), author VARCHAR(255), price DECIMAL(10, 2) ); INSERT INTO fiction_book (id, title, author, price) VALUES (1, 'The Great Gatsby', 'F. Scott Fitzgerald', 24.99), (2, '1984', 'George Orwell', 19.99), (3, 'Pride and Prejudice', 'Jane Austen', 15.99), (4, 'The Hobbit', 'J.R.R. Tolkien', 29.99), (5, 'The Doors of Perception', 'Aldous Huxley', 12.99); INSERT INTO non_fiction_book (id, title, author, price) VALUES (1, 'A Brief History of Time', 'Stephen Hawking', 29.99), (2, 'The Art of War', 'Sun Tzu', 19.99), (3, 'Sapiens', 'Yuval Noah Harari', 24.99), (4, 'Pride and Prejudice: A Study Guide', 'John Smith', 12.99), (5, 'The Doors of Perception', 'Aldous Huxley', 12.99); CREATE TABLE summer_reads ( id INT PRIMARY KEY, title VARCHAR(255), author VARCHAR(255), price DECIMAL(10, 2) ); INSERT INTO summer_reads (id, title, author, price) VALUES (1, 'The Great Gatsby', 'F. Scott Fitzgerald', 24.99), (2, 'Sapiens', 'Yuval Noah Harari', 24.99), (3, 'The Art of War', 'Sun Tzu', 19.99); ``` |
Sometimes you need to combine results from multiple queries into a single result set. SQL provides several set operations for this purpose: UNION
, UNION ALL
, INTERSECT
, and EXCEPT
.
Let' take a look at some examples to understand how these set operations work. I have setup the following two tables fiction_book
and non_fiction_book
with the following data:
id | title | author | price |
---|---|---|---|
1 | The Great Gatsby | F. Scott Fitzgerald | 24.99 |
2 | 1984 | George Orwell | 19.99 |
3 | Pride and Prejudice | Jane Austen | 15.99 |
4 | The Hobbit | J.R.R. Tolkien | 29.99 |
5 | The Doors of Perception | Aldous Huxley | 12.99 |
id | title | author | price |
---|---|---|---|
1 | A Brief History of Time | Stephen Hawking | 29.99 |
2 | The Art of War | Sun Tzu | 19.99 |
3 | Sapiens | Yuval Noah Harari | 24.99 |
4 | Pride and Prejudice: A Study Guide | John Smith | 12.99 |
5 | The Doors of Perception | Aldous Huxley | 12.99 |
Let's start with the UNION
operations.
UNION
combines results from two or more queries and removes duplicates. UNION ALL
does the same but keeps duplicates.
UNION Example
Let's find all books across fiction_book
and non_fiction_book
tables using a UNION
operation:
SELECT title, author, price
FROM fiction_book
UNION
SELECT title, author, price
FROM non_fiction_book
ORDER BY title;
The query above will combine the results from both the queries and remove any duplicates. Output will be:
title | author | price |
---|---|---|
1984 | George Orwell | 19.99 |
A Brief History of Time | Stephen Hawking | 29.99 |
Pride and Prejudice | Jane Austen | 15.99 |
Pride and Prejudice: A Study Guide | John Smith | 12.99 |
Sapiens | Yuval Noah Harari | 24.99 |
The Art of War | Sun Tzu | 19.99 |
The Doors of Perception | Aldous Huxley | 12.99 |
The Great Gatsby | F. Scott Fitzgerald | 24.99 |
The Hobbit | J.R.R. Tolkien | 29.99 |
Notice how the The Doors of Perception
book is only listed once even though it appears in both the fiction_book
and non_fiction_book
tables.
UNION ALL Example
If you want to keep duplicates, use UNION ALL:
SELECT title, author, price FROM fiction_book
UNION ALL
SELECT title, author, price FROM non_fiction_book
ORDER BY title;
The result will be:
title | author | price |
---|---|---|
1984 | George Orwell | 19.99 |
A Brief History of Time | Stephen Hawking | 29.99 |
Pride and Prejudice | Jane Austen | 15.99 |
Pride and Prejudice: A Study Guide | John Smith | 12.99 |
Sapiens | Yuval Noah Harari | 24.99 |
The Art of War | Sun Tzu | 19.99 |
The Doors of Perception | Aldous Huxley | 12.99 |
The Doors of Perception | Aldous Huxley | 12.99 |
The Great Gatsby | F. Scott Fitzgerald | 24.99 |
The Hobbit | J.R.R. Tolkien | 29.99 |
Notice how the The Doors of Perception
book is listed twice.
INTERSECT
INTERSECT
returns only the rows that appear in both result sets. Let's find books that are in both the fiction_book
and non_fiction_book
tables:
SELECT title, author FROM fiction_book
INTERSECT
SELECT title, author FROM non_fiction_book;
This will return:
title | author |
---|---|
The Doors of Perception | Aldous Huxley |
EXCEPT
EXCEPT
returns rows from the first query that don't appear in the second query. Let's find the books in the fiction_book
table that aren't in the non_fiction_book
table:
SELECT title, author FROM fiction_book
EXCEPT
SELECT title, author FROM non_fiction_book
ORDER BY title;
This will return:
title | author |
---|---|
1984 | George Orwell |
Pride and Prejudice | Jane Austen |
The Great Gatsby | F. Scott Fitzgerald |
The Hobbit | J.R.R. Tolkien |
Requirements
The key requirement for all of these set operation queries listed above is that the columns in both queries must have the same number, order, and data types.
For example, below is an invalid query:
-- ERROR: each UNION query must have the same number of columns
SELECT title, author FROM fiction_book
UNION
SELECT title, author, price FROM non_fiction_book;
The above query will return an error because the fiction_book
table has 2 columns and the non_fiction_book
table has 3 columns.
Similary if we try to combine the following two queries:
-- ERROR: UNION types numeric and character varying cannot be matched
SELECT title, price FROM fiction_book
UNION
SELECT title, author FROM non_fiction_book;
This will also return an error because the price
column from the first query doesn't have the same data type as the author
column from the second query.
Also, another important thing to note for the set operations is that the column names in the result set will be the column names from the first query. For example, if we run the following query:
SELECT title, author as 'writer_name' FROM fiction_book
UNION
SELECT title, author FROM non_fiction_book;
The output will be:
title | writer_name |
---|---|
A Brief History of Time | Stephen Hawking |
The Hobbit | J.R.R. Tolkien |
The Great Gatsby | F. Scott Fitzgerald |
Sapiens | Yuval Noah Harari |
The Art of War | Sun Tzu |
Pride and Prejudice: A Study Guide | John Smith |
1984 | George Orwell |
The Doors of Perception | Aldous Huxley |
Pride and Prejudice | Jane Austen |
Notice how it took the writer_name
column alias from the first query for the name of the column in the result set.
Challenge Time! 🎯
Using the provided tables, write queries to:
- Find all books that cost exactly $19.99 across all categories
- List authors who have written both fiction and non-fiction books
- Find summer reads that aren't in our fiction collection
Solution
-- 1. Books that cost $19.99
SELECT title, author, 'Fiction' as category
FROM fiction_book
WHERE price = 19.99
UNION
SELECT title, author, 'Non-Fiction' as category
FROM non_fiction_book
WHERE price = 19.99;
-- 2. Authors in both categories
SELECT author FROM fiction_book
INTERSECT
SELECT author FROM non_fiction_book;
-- 3. Summer reads not in fiction
SELECT title, author FROM summer_reads
EXCEPT
SELECT title, author FROM fiction_book;
Key Takeaways
- Use UNION to combine results and remove duplicates
- Use UNION ALL when you want to keep duplicates (it's faster than UNION)
- Use INTERSECT to find common rows between results
- Use EXCEPT to find rows in one result that aren't in another
- Column count, order, and data types must match across queries
- These operations are particularly useful when working with similar but separate data sets
Remember: Choose the appropriate set operation based on your needs:
- UNION/UNION ALL for combining results
- INTERSECT for finding common elements
- EXCEPT for finding differences