DELETE with RETURNING clause in PostgreSQL - Time & Space Complexity

Time Complexity: DELETE with RETURNING clause

O(n)

Understanding Time Complexity

When we delete rows from a table and ask the database to return the deleted rows, it takes some time to find and remove those rows.

We want to understand how the time needed grows as the number of rows to delete grows.

Scenario Under Consideration

Analyze the time complexity of the following code snippet.

DELETE FROM employees
WHERE department_id = 5
RETURNING employee_id, name;
    

This code deletes all employees in department 5 and returns their IDs and names.

Identify Repeating Operations

Identify the loops, recursion, array traversals that repeat.

• Primary operation: Scanning the table or index to find rows matching the condition.
• How many times: Once for each row in the table or index scanned.

How Execution Grows With Input

As the number of rows to delete grows, the database must check more rows and delete more data.

Input Size (n)	Approx. Operations
10	About 10 row checks and deletes
100	About 100 row checks and deletes
1000	About 1000 row checks and deletes

Pattern observation: The work grows roughly in direct proportion to the number of rows deleted.

Final Time Complexity

Time Complexity: O(n)

This means the time to delete and return rows grows linearly with how many rows match the condition.

Common Mistake

[X] Wrong: "Deleting rows with RETURNING is instant no matter how many rows match."

[OK] Correct: The database must find and remove each matching row and prepare the returned data, so more rows mean more work and more time.

Interview Connect

Understanding how DELETE with RETURNING scales helps you explain database behavior clearly and shows you think about efficiency in real tasks.

Self-Check

"What if we added an index on department_id? How would the time complexity change?"

Practice

1. What does the RETURNING clause do in a DELETE statement in PostgreSQL?

easy

A. It counts the number of rows in the table.

B. It prevents rows from being deleted.

C. It updates rows instead of deleting them.

D. It returns the rows that were deleted.

Solution

1. Step 1: Understand DELETE operation

   The DELETE statement removes rows from a table based on a condition.

2. Step 2: Role of RETURNING clause

   The RETURNING clause outputs the rows that were deleted, showing their data immediately.

3. Final Answer:

   It returns the rows that were deleted. -> Option D

4. Quick Check:

   RETURNING shows deleted rows [OK]

Hint: RETURNING shows deleted rows immediately [OK]

Common Mistakes:

• Thinking RETURNING prevents deletion
• Confusing RETURNING with UPDATE
• Assuming RETURNING counts rows

2. Which of the following is the correct syntax to delete rows from users where age < 18 and return all deleted rows?

easy

A. DELETE FROM users WHERE age < 18 RETURNING *;

B. DELETE users WHERE age < 18 RETURNING *;

C. DELETE FROM users RETURNING * WHERE age < 18;

D. DELETE FROM users WHERE age < 18 RETURNING all;

Solution

1. Step 1: Check DELETE syntax

   The correct DELETE syntax is: DELETE FROM table WHERE condition.

2. Step 2: Use RETURNING clause correctly

   RETURNING * must come after WHERE clause to return all columns of deleted rows.

3. Final Answer:

   DELETE FROM users WHERE age < 18 RETURNING *; -> Option A

4. Quick Check:

   Correct DELETE syntax with RETURNING * [OK]

Hint: DELETE FROM ... WHERE ... RETURNING *; is correct syntax [OK]

Common Mistakes:

• Omitting FROM keyword
• Placing RETURNING before WHERE
• Using RETURNING all instead of *

3. Given the table products with rows:
{id:1, name:'Pen', stock:10}, {id:2, name:'Pencil', stock:0}, {id:3, name:'Eraser', stock:5}
What will this query return?

DELETE FROM products WHERE stock = 0 RETURNING id, name;

medium

A. [{id:1, name:'Pen'}, {id:3, name:'Eraser'}]

B. []

C. [{id:2, name:'Pencil'}]

D. Syntax error

Solution

1. Step 1: Identify rows matching condition

   Only the product with stock = 0 is {id:2, name:'Pencil'}.

2. Step 2: DELETE removes matching rows and RETURNING outputs specified columns

   The query deletes that row and returns its id and name.

3. Final Answer:

   [{id:2, name:'Pencil'}] -> Option C

4. Quick Check:

   Deleted rows with stock=0 returned [OK]

Hint: RETURNING shows deleted rows matching WHERE [OK]

Common Mistakes:

• Returning all rows instead of matching ones
• Expecting deleted rows to remain in table
• Confusing syntax causing errors

4. What is wrong with this query?

DELETE FROM orders RETURNING * WHERE status = 'cancelled';

medium

A. The WHERE clause must come before RETURNING.

B. RETURNING * is not allowed in DELETE.

C. Missing semicolon at the end.

D. DELETE cannot filter rows.

Solution

1. Step 1: Check clause order in DELETE

   In PostgreSQL, WHERE clause must come before RETURNING clause.

2. Step 2: Identify error in query

   The query places RETURNING * before WHERE, which is invalid syntax.

3. Final Answer:

   The WHERE clause must come before RETURNING. -> Option A

4. Quick Check:

   WHERE before RETURNING in DELETE [OK]

Hint: WHERE clause always before RETURNING [OK]

Common Mistakes:

• Placing RETURNING before WHERE
• Thinking RETURNING is disallowed
• Ignoring clause order rules

5. You want to delete all employees from staff who have salary < 30000 and return their id and name. Which query correctly does this and also returns the count of deleted rows?

hard

A. DELETE FROM staff WHERE salary < 30000 RETURNING id, name, COUNT(*);

B. WITH deleted AS (DELETE FROM staff WHERE salary < 30000 RETURNING id, name) SELECT COUNT(*) FROM deleted;

C. DELETE FROM staff RETURNING id, name WHERE salary < 30000; SELECT COUNT(*) FROM staff;

D. DELETE FROM staff WHERE salary < 30000; SELECT id, name FROM staff RETURNING COUNT(*);

Solution

1. Step 1: Delete rows with condition and return columns

   Using DELETE ... WHERE salary < 30000 RETURNING id, name returns deleted rows.

2. Step 2: Use WITH clause to capture deleted rows and count them

   WITH deleted AS (...) lets us select COUNT(*) from deleted rows in one query.

3. Final Answer:

   WITH deleted AS (DELETE FROM staff WHERE salary < 30000 RETURNING id, name) SELECT COUNT(*) FROM deleted; -> Option B

4. Quick Check:

   Use WITH + RETURNING to delete and count [OK]

Hint: Use WITH clause to count deleted rows with RETURNING [OK]

Common Mistakes:

• Trying to use COUNT(*) inside RETURNING
• Placing WHERE after RETURNING
• Separating DELETE and count without capturing deleted rows