Recursive CTEs in MySQL

Introduction

Recursive CTEs help you repeat a query step by step until you get all related data. It is like following a chain of connections one by one.

Finding all employees under a manager in a company hierarchy.

Listing all parts needed to build a product from smaller parts.

Exploring family trees to find ancestors or descendants.

Calculating sequences like factorial or Fibonacci numbers in SQL.

Syntax

MySQL

WITH RECURSIVE cte_name (column_list) AS (
  -- Anchor member: base query
  SELECT initial_columns
  FROM table_name
  WHERE condition
  UNION ALL
  -- Recursive member: query referring to cte_name
  SELECT next_columns
  FROM table_name JOIN cte_name ON join_condition
  WHERE recursive_condition
)
SELECT * FROM cte_name;

The anchor member is the starting point of recursion.

The recursive member repeats the query using results from the previous step.

Examples

This example generates numbers from 1 to 5 using recursion.

MySQL

WITH RECURSIVE numbers(n) AS (
  SELECT 1
  UNION ALL
  SELECT n + 1 FROM numbers WHERE n < 5
)
SELECT * FROM numbers;

This shows what happens if the anchor query returns no rows: the whole CTE returns no rows.

MySQL

WITH RECURSIVE empty_cte AS (
  SELECT 1 WHERE FALSE
  UNION ALL
  SELECT 2 FROM empty_cte
)
SELECT * FROM empty_cte;

This example has only one element because the recursive condition stops immediately.

MySQL

WITH RECURSIVE single_node AS (
  SELECT 1 AS id
  UNION ALL
  SELECT id + 1 FROM single_node WHERE id < 1
)
SELECT * FROM single_node;

This example shows recursion stopping at the end of the chain (id = 3).

MySQL

WITH RECURSIVE chain AS (
  SELECT 1 AS id
  UNION ALL
  SELECT id + 1 FROM chain WHERE id < 3
)
SELECT * FROM chain;

Sample Program

This query finds all employees under the top manager, showing their level in the hierarchy.

MySQL

WITH RECURSIVE employee_hierarchy AS (
  -- Anchor member: select the top manager
  SELECT employee_id, manager_id, employee_name, 1 AS level
  FROM employees
  WHERE manager_id IS NULL
  UNION ALL
  -- Recursive member: select employees reporting to previous level
  SELECT e.employee_id, e.manager_id, e.employee_name, eh.level + 1
  FROM employees e
  JOIN employee_hierarchy eh ON e.manager_id = eh.employee_id
)
SELECT employee_id, manager_id, employee_name, level
FROM employee_hierarchy
ORDER BY level, employee_id;

Important Notes

Recursive CTEs run until no new rows are added, so be careful to avoid infinite loops.

Time complexity depends on the depth and branching of recursion; it can grow quickly.

Common mistake: forgetting the anchor member or recursive condition, causing no results or infinite recursion.

Use recursive CTEs when you need to process hierarchical or sequential data inside SQL.

Summary

Recursive CTEs let you repeat queries step by step to explore connected data.

They have an anchor part and a recursive part that refers back to the CTE.

Useful for hierarchies, sequences, and tree-like data in databases.

Practice

1. What is the main purpose of a recursive CTE in MySQL?

easy

A. To create temporary tables for storing large datasets

B. To repeatedly execute a query to explore hierarchical or connected data

C. To speed up simple SELECT queries without joins

D. To permanently store data in the database

Solution

1. Step 1: Understand recursive CTE concept

   A recursive CTE runs a base query and then repeatedly runs a recursive part referring back to itself to explore connected data.

2. Step 2: Identify the purpose

   This process is mainly used to handle hierarchical or tree-like data, such as organizational charts or folder structures.

3. Final Answer:

   To repeatedly execute a query to explore hierarchical or connected data -> Option B

4. Quick Check:

   Recursive CTE = hierarchical data exploration [OK]

Hint: Recursive CTEs repeat queries to explore connected data [OK]

Common Mistakes:

• Confusing recursive CTEs with temporary tables
• Thinking recursive CTEs speed up all queries
• Assuming recursive CTEs permanently store data

2. Which of the following is the correct basic syntax to start a recursive CTE in MySQL?

easy

A. WITH cte_name RECURSIVE (anchor_query UNION recursive_query) SELECT * FROM cte_name;

B. CREATE RECURSIVE TABLE cte_name AS (anchor_query UNION recursive_query);

C. WITH RECURSIVE cte_name AS (anchor_query UNION ALL recursive_query) SELECT * FROM cte_name;

D. SELECT RECURSIVE * FROM cte_name WHERE condition;

Solution

1. Step 1: Recall MySQL recursive CTE syntax

   MySQL uses the keyword WITH RECURSIVE followed by the CTE name and a query combining an anchor and recursive part with UNION ALL.

2. Step 2: Match syntax options

   WITH RECURSIVE cte_name AS (anchor_query UNION ALL recursive_query) SELECT * FROM cte_name; correctly shows this syntax, including UNION ALL and the final SELECT from the CTE.

3. Final Answer:

   WITH RECURSIVE cte_name AS (anchor_query UNION ALL recursive_query) SELECT * FROM cte_name; -> Option C

4. Quick Check:

   WITH RECURSIVE + UNION ALL = correct syntax [OK]

Hint: Look for WITH RECURSIVE and UNION ALL keywords [OK]

Common Mistakes:

• Using UNION instead of UNION ALL in recursive part
• Trying to CREATE RECURSIVE TABLE (not valid)
• Misplacing RECURSIVE keyword after CTE name

3. Given the table employees(id, manager_id, name), what will this query return?

WITH RECURSIVE subordinates AS (
  SELECT id, name FROM employees WHERE id = 2
  UNION ALL
  SELECT e.id, e.name FROM employees e
  JOIN subordinates s ON e.manager_id = s.id
)
SELECT * FROM subordinates;

medium

A. All employees who report directly or indirectly to employee with id 2

B. Only the employee with id 2

C. All employees who manage employee with id 2

D. Syntax error due to missing anchor query

Solution

1. Step 1: Analyze anchor query

   The anchor selects the employee with id = 2 as the starting point.

2. Step 2: Analyze recursive part

   The recursive part joins employees whose manager_id matches the current subordinates' id, thus finding direct reports, then their reports, and so on.

3. Step 3: Understand final output

   The query returns employee 2 and all employees reporting to them directly or indirectly.

4. Final Answer:

   All employees who report directly or indirectly to employee with id 2 -> Option A

5. Quick Check:

   Recursive join on manager_id finds all subordinates [OK]

Hint: Recursive join on manager_id finds all subordinates [OK]

Common Mistakes:

• Confusing manager and subordinate roles
• Thinking it returns only direct reports
• Assuming syntax error due to anchor query presence

4. Identify the error in this recursive CTE:

WITH RECURSIVE cte AS (
  SELECT id FROM items WHERE parent_id IS NULL
  UNION
  SELECT i.id FROM items i JOIN cte ON i.parent_id = cte.id
)
SELECT * FROM cte;

medium

A. Anchor query should use WHERE parent_id = 0 instead of IS NULL

B. Recursive part should use LEFT JOIN instead of JOIN

C. CTE name cannot be 'cte', must be more descriptive

D. Missing UNION ALL causes infinite recursion or wrong results

Solution

1. Step 1: Check recursive CTE syntax

   Recursive CTEs require UNION ALL to combine anchor and recursive parts to avoid duplicate elimination.

2. Step 2: Understand effect of UNION vs UNION ALL

   Using UNION removes duplicates each iteration, which can cause infinite recursion or incorrect results in recursive queries.

3. Final Answer:

   Missing UNION ALL causes infinite recursion or wrong results -> Option D

4. Quick Check:

   Recursive CTEs must use UNION ALL [OK]

Hint: Always use UNION ALL in recursive CTEs to avoid infinite loops [OK]

Common Mistakes:

• Using UNION instead of UNION ALL in recursive CTE
• Assuming NULL vs 0 matters for recursion logic
• Thinking JOIN type affects recursion correctness

5. You want to list all ancestors of an employee in a company hierarchy stored in employees(id, manager_id, name). Which recursive CTE query correctly finds all managers up the chain for employee with id 5?

hard

A. WITH RECURSIVE ancestors AS (
SELECT id, manager_id, name FROM employees WHERE id = 5
UNION ALL
SELECT e.id, e.manager_id, e.name FROM employees e
JOIN ancestors a ON e.id = a.manager_id
)
SELECT * FROM ancestors WHERE id != 5;

B. WITH RECURSIVE ancestors AS (
SELECT id, manager_id, name FROM employees WHERE id = 5
UNION ALL
SELECT e.id, e.manager_id, e.name FROM employees e
JOIN ancestors a ON e.manager_id = a.id
)
SELECT * FROM ancestors WHERE id != 5;

C. WITH RECURSIVE ancestors AS (
SELECT id, manager_id, name FROM employees WHERE manager_id = 5
UNION ALL
SELECT e.id, e.manager_id, e.name FROM employees e
JOIN ancestors a ON e.manager_id = a.id
)
SELECT * FROM ancestors;

D. WITH RECURSIVE ancestors AS (
SELECT id, manager_id, name FROM employees WHERE id = 5
UNION
SELECT e.id, e.manager_id, e.name FROM employees e
JOIN ancestors a ON e.id = a.manager_id
)
SELECT * FROM ancestors WHERE id != 5;

Solution

1. Step 1: Understand the goal

   We want to find all managers (ancestors) of employee 5, moving up the chain via manager_id.

2. Step 2: Analyze join condition

   To move up, join employees where employee id matches the manager_id of current ancestor (e.id = a.manager_id).

3. Step 3: Check anchor and recursive parts

   Anchor selects employee 5; recursive part finds their manager, then that manager's manager, etc.

4. Final Answer:

   WITH RECURSIVE ancestors AS (SELECT id, manager_id, name FROM employees WHERE id = 5 UNION ALL SELECT e.id, e.manager_id, e.name FROM employees e JOIN ancestors a ON e.id = a.manager_id) SELECT * FROM ancestors WHERE id != 5; -> Option A

5. Quick Check:

   Join on e.id = a.manager_id moves up hierarchy [OK]

Hint: Join on e.id = a.manager_id to move up the hierarchy [OK]

Common Mistakes:

• Reversing join condition to move down instead of up
• Using UNION instead of UNION ALL causing errors
• Starting anchor with manager_id = 5 instead of id = 5