Returning modified rows with RETURNING in PostgreSQL - Time & Space Complexity

Time Complexity: Returning modified rows with RETURNING

O(n)

Understanding Time Complexity

When we update rows in a database and ask it to return the changed rows, we want to know how the time it takes grows as the data grows.

How does the database handle returning modified rows as the number of rows changes?

Scenario Under Consideration

Analyze the time complexity of the following code snippet.

UPDATE employees
SET salary = salary * 1.1
WHERE department = 'Sales'
RETURNING id, salary;
    

This code increases the salary by 10% for all employees in the Sales department and returns their id and new salary.

Identify Repeating Operations

Identify the loops, recursion, array traversals that repeat.

• Primary operation: Scanning all rows in the employees table to find those in the Sales department.
• How many times: Once per row in the table, to check the department and update if needed.

How Execution Grows With Input

As the number of employees grows, the database must check each one to see if they belong to Sales and update their salary if so.

Input Size (n)	Approx. Operations
10	About 10 checks and some updates
100	About 100 checks and more updates
1000	About 1000 checks and many updates

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

Final Time Complexity

Time Complexity: O(n)

This means the time to update and return rows grows linearly with the number of rows in the table.

Common Mistake

[X] Wrong: "The RETURNING clause makes the update run faster because it returns data immediately."

[OK] Correct: Returning rows adds work because the database must collect and send the updated rows, so it does not reduce the time needed to find and update rows.

Interview Connect

Understanding how returning modified rows affects performance helps you explain database behavior clearly and shows you think about real data sizes and costs.

Self-Check

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

Practice

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

easy

A. It delays the update until a commit is issued.

B. It prevents the update from modifying any rows.

C. It returns the rows that were updated immediately after the update.

D. It returns the number of rows in the table before the update.

Solution

1. Step 1: Understand the purpose of RETURNING

   The RETURNING clause is used to get the data of rows affected by an INSERT, UPDATE, or DELETE immediately.

2. Step 2: Apply to UPDATE statement

   When used with UPDATE, RETURNING shows the updated rows without needing a separate SELECT query.

3. Final Answer:

   It returns the rows that were updated immediately after the update. -> Option C

4. Quick Check:

   RETURNING shows changed rows = B [OK]

Hint: RETURNING shows changed rows right after modification [OK]

Common Mistakes:

• Thinking RETURNING delays changes
• Assuming RETURNING prevents updates
• Confusing RETURNING with row count only

2. Which of the following is the correct syntax to return all columns of updated rows in PostgreSQL?

easy

A. UPDATE table_name SET col = val RETURNING ALL COLUMNS;

B. UPDATE table_name SET col = val RETURNING *;

C. UPDATE table_name SET col = val RETURNING ALL;

D. UPDATE table_name SET col = val RETURN *;

Solution

1. Step 1: Recall correct RETURNING syntax

   PostgreSQL uses RETURNING followed by column names or * to return all columns.

2. Step 2: Identify correct option

   Only UPDATE table_name SET col = val RETURNING *; uses RETURNING * which is valid syntax to return all columns.

3. Final Answer:

   UPDATE table_name SET col = val RETURNING *; -> Option B

4. Quick Check:

   RETURNING * returns all columns [OK]

Hint: Use RETURNING * to get all columns after update [OK]

Common Mistakes:

• Using RETURN instead of RETURNING
• Writing RETURNING ALL instead of RETURNING *
• Adding extra keywords like ALL COLUMNS

3. Given the table users(id INT, name TEXT, age INT) with rows:
(1, 'Alice', 30), (2, 'Bob', 25)
What will this query return?

UPDATE users SET age = age + 1 WHERE id = 2 RETURNING id, age;

medium

A. (2, 26)

B. (2, 25)

C. (1, 31)

D. (1, 30)

Solution

1. Step 1: Identify updated row

   The WHERE clause updates only the row with id = 2 (Bob).

2. Step 2: Calculate new age and returned columns

   Bob's age increases from 25 to 26. RETURNING id, age returns (2, 26).

3. Final Answer:

   (2, 26) -> Option A

4. Quick Check:

   Updated row with id=2 returns new age 26 [OK]

Hint: RETURNING shows updated values immediately [OK]

Common Mistakes:

• Returning old values instead of updated
• Confusing which row is updated
• Ignoring WHERE clause effect

4. What is wrong with this query?

UPDATE products SET price = price * 1.1 RETURNING;

medium

A. RETURNING cannot be used with UPDATE statements.

B. The multiplication operator * is invalid in UPDATE.

C. Missing WHERE clause causes syntax error.

D. RETURNING clause requires columns or * to specify what to return.

Solution

1. Step 1: Check RETURNING syntax

   RETURNING must be followed by column names or *; empty RETURNING is invalid syntax.

2. Step 2: Validate other parts

   Multiplication and UPDATE without WHERE are valid; RETURNING with no columns is the error.

3. Final Answer:

   RETURNING clause requires columns or * to specify what to return. -> Option D

4. Quick Check:

   RETURNING needs columns or * [OK]

Hint: Always specify columns or * after RETURNING [OK]

Common Mistakes:

• Leaving RETURNING empty
• Thinking WHERE is mandatory for syntax
• Believing RETURNING disallowed in UPDATE

5. You want to increase the salary by 10% for employees in department 5 and get their id and new salary. Which query correctly does this and returns the updated rows?

hard

A. UPDATE employees SET salary = salary * 1.1 WHERE department_id = 5 RETURNING id, salary;

B. UPDATE employees SET salary = salary + 10 WHERE department_id = 5 RETURNING *;

C. UPDATE employees SET salary = salary * 1.1 RETURNING id, salary WHERE department_id = 5;

D. UPDATE employees SET salary = salary * 1.1 WHERE department_id = 5 RETURN id, salary;

Solution

1. Step 1: Identify correct UPDATE with WHERE and RETURNING

   The query must update only department 5 employees and return id and salary.

2. Step 2: Check syntax correctness

   UPDATE employees SET salary = salary * 1.1 WHERE department_id = 5 RETURNING id, salary; uses correct syntax: UPDATE ... SET ... WHERE ... RETURNING id, salary;

3. Step 3: Eliminate incorrect options

   Adding a fixed 10 instead of multiplying by 1.1 doesn't implement a 10% increase. Placing RETURNING before WHERE is invalid syntax. Using RETURN instead of RETURNING is wrong.

4. Final Answer:

   UPDATE employees SET salary = salary * 1.1 WHERE department_id = 5 RETURNING id, salary; -> Option A

5. Quick Check:

   Correct syntax with WHERE and RETURNING columns [OK]

Hint: RETURNING follows WHERE clause with columns listed [OK]

Common Mistakes:

• Placing RETURNING before WHERE
• Using RETURN instead of RETURNING
• Adding fixed amount instead of percentage