Why Recursive CTE for series generation in SQL? - Purpose & Use Cases

The Big Idea

What if you could create endless lists of numbers with just one simple query?

The Scenario

Imagine you need to create a list of numbers from 1 to 100 for a report, but you only have a basic table with a few rows. You try to write out each number manually or copy-paste them one by one.

The Problem

Writing out each number by hand is slow and boring. Copy-pasting can cause mistakes like missing numbers or duplicates. If you want a longer list or a different range, you have to redo everything. It's easy to get tired and make errors.

The Solution

Using a Recursive CTE lets you generate a series of numbers automatically inside your SQL query. It starts from a base number and keeps adding the next number until it reaches the limit you set. This saves time, avoids mistakes, and works for any range you want.

Before vs After

✗ Before

SELECT 1 AS number UNION ALL SELECT 2 UNION ALL SELECT 3 UNION ALL ...

✓ After

WITH RECURSIVE numbers AS (SELECT 1 AS number UNION ALL SELECT number + 1 FROM numbers WHERE number < 100) SELECT number FROM numbers;

What It Enables

You can quickly create any sequence of numbers or dates inside your database, making complex reports and calculations much easier.

Real Life Example

A sales manager wants to see daily sales for the past month, even on days with no sales recorded. Using a recursive CTE to generate all dates in the month ensures every day appears in the report.

Key Takeaways

Manual number lists are slow and error-prone.

Recursive CTEs generate sequences automatically and flexibly.

This technique helps create complete reports and data sets easily.

Practice

1. What is the main purpose of using a Recursive CTE in SQL for series generation?

easy

A. To generate a sequence of numbers or dates step-by-step within a query

B. To join two tables based on a common column

C. To delete duplicate rows from a table

D. To create indexes for faster searching

Solution

1. Step 1: Understand Recursive CTE usage

   Recursive CTEs are used to repeatedly execute a query to build results step-by-step.

2. Step 2: Identify series generation purpose

   They are commonly used to generate sequences like numbers or dates without needing extra tables.

3. Final Answer:

   To generate a sequence of numbers or dates step-by-step within a query -> Option A

4. Quick Check:

   Recursive CTE = sequence generation [OK]

Hint: Recursive CTEs build sequences stepwise inside SQL [OK]

Common Mistakes:

• Confusing recursive CTE with JOIN operations
• Thinking recursive CTE deletes duplicates
• Assuming recursive CTE creates indexes

2. Which of the following is the correct syntax to start a recursive CTE named numbers generating numbers from 1?

easy

A. WITH RECURSIVE numbers AS SELECT 1 AS num UNION ALL SELECT num + 1 FROM numbers WHERE num < 5

B. WITH RECURSIVE numbers AS (SELECT 1 AS num UNION ALL SELECT num + 1 FROM numbers WHERE num < 5)

C. WITH numbers RECURSIVE AS (SELECT 1 AS num UNION ALL SELECT num + 1 FROM numbers WHERE num < 5)

D. WITH RECURSIVE numbers (num) AS (SELECT 1 AS n UNION ALL SELECT num + 1 FROM numbers WHERE num < 5)

Solution

1. Step 1: Check correct WITH RECURSIVE syntax

   The correct syntax starts with WITH RECURSIVE cte_name AS ( ... ).

2. Step 2: Validate the recursive query structure

   WITH RECURSIVE numbers AS (SELECT 1 AS num UNION ALL SELECT num + 1 FROM numbers WHERE num < 5) correctly uses SELECT 1 AS num as anchor and recursive part with UNION ALL and stopping condition.

3. Final Answer:

   WITH RECURSIVE numbers AS (SELECT 1 AS num UNION ALL SELECT num + 1 FROM numbers WHERE num < 5) -> Option B

4. Quick Check:

   Correct recursive CTE syntax = WITH RECURSIVE numbers AS (SELECT 1 AS num UNION ALL SELECT num + 1 FROM numbers WHERE num < 5) [OK]

Hint: Start recursive CTE with WITH RECURSIVE cte AS (...) [OK]

Common Mistakes:

• Placing RECURSIVE after CTE name
• Missing parentheses around CTE query
• Omitting AS keyword

3. Given the query:

WITH RECURSIVE seq AS (
  SELECT 1 AS n
  UNION ALL
  SELECT n + 2 FROM seq WHERE n < 7
)
SELECT * FROM seq;

What will be the output rows?

medium

A. [1, 3, 5]

B. [1, 2, 3, 4, 5, 6, 7]

C. [1, 3, 5, 7]

D. [2, 4, 6, 8]

Solution

1. Step 1: Analyze anchor and recursive step

   Anchor row is 1. Recursive step adds 2 each time: 1, 3, 5, 7.

2. Step 2: Check stopping condition

   Recursion stops when n < 7 is false. Since 7 is included, sequence ends at 7.

3. Final Answer:

   [1, 3, 5, 7] -> Option C

4. Quick Check:

   Start=1, step=+2, stop at 7 = [1,3,5,7] [OK]

Hint: Add step value repeatedly until condition fails [OK]

Common Mistakes:

• Assuming step is 1 instead of 2
• Stopping before reaching 7
• Including numbers beyond 7

4. Identify the error in this recursive CTE:

WITH RECURSIVE count_up AS (
  SELECT 1 AS num
  UNION ALL
  SELECT num + 1 FROM count_up
)
SELECT * FROM count_up WHERE num <= 3;

medium

A. Using UNION instead of UNION ALL

B. Incorrect anchor query syntax

C. Wrong column alias in SELECT statements

D. Missing stopping condition inside the recursive part causing infinite recursion

Solution

1. Step 1: Check recursive part for stopping condition

   The recursive SELECT lacks a WHERE clause to stop recursion, causing infinite loop.

2. Step 2: Understand WHERE clause placement

   The WHERE clause outside the CTE filters final results but does not stop recursion inside CTE.

3. Final Answer:

   Missing stopping condition inside the recursive part causing infinite recursion -> Option D

4. Quick Check:

   Stopping condition must be inside recursive CTE [OK]

Hint: Put stopping condition inside recursive SELECT, not outside [OK]

Common Mistakes:

• Placing stopping condition only in final SELECT
• Confusing UNION and UNION ALL effects
• Incorrect alias usage

5. You want to generate a list of dates from '2024-01-01' to '2024-01-05' using a recursive CTE. Which query correctly produces this date series?

hard

A. WITH RECURSIVE dates AS (SELECT DATE('2024-01-01') AS dt UNION ALL SELECT dt + INTERVAL '1 day' FROM dates WHERE dt < '2024-01-05') SELECT * FROM dates;

B. WITH RECURSIVE dates AS (SELECT '2024-01-01' AS dt UNION ALL SELECT dt + INTERVAL 1 DAY FROM dates WHERE dt <= '2024-01-05') SELECT * FROM dates;

C. WITH RECURSIVE dates AS (SELECT '2024-01-01' AS dt UNION ALL SELECT dt + 1 FROM dates WHERE dt < '2024-01-05') SELECT * FROM dates;

D. WITH RECURSIVE dates AS (SELECT DATE('2024-01-01') AS dt UNION ALL SELECT DATE(dt, '+1 day') FROM dates WHERE dt < '2024-01-05') SELECT * FROM dates;

Solution

1. Step 1: Understand date addition syntax

   WITH RECURSIVE dates AS (SELECT DATE('2024-01-01') AS dt UNION ALL SELECT dt + INTERVAL '1 day' FROM dates WHERE dt < '2024-01-05') SELECT * FROM dates; uses dt + INTERVAL '1 day' which is standard SQL for adding one day to a date.

2. Step 2: Check stopping condition and anchor

   Anchor selects the start date as DATE type. Recursive part adds one day while date is less than '2024-01-05'.

3. Step 3: Compare other options

   WITH RECURSIVE dates AS (SELECT DATE('2024-01-01') AS dt UNION ALL SELECT DATE(dt, '+1 day') FROM dates WHERE dt < '2024-01-05') SELECT * FROM dates; uses SQLite style DATE(dt, '+1 day'), which is not standard SQL. WITH RECURSIVE dates AS (SELECT '2024-01-01' AS dt UNION ALL SELECT dt + INTERVAL 1 DAY FROM dates WHERE dt <= '2024-01-05') SELECT * FROM dates; uses incorrect WHERE condition allowing extra date. WITH RECURSIVE dates AS (SELECT '2024-01-01' AS dt UNION ALL SELECT dt + 1 FROM dates WHERE dt < '2024-01-05') SELECT * FROM dates; adds integer to date incorrectly.

4. Final Answer:

   WITH RECURSIVE dates AS (SELECT DATE('2024-01-01') AS dt UNION ALL SELECT dt + INTERVAL '1 day' FROM dates WHERE dt < '2024-01-05') SELECT * FROM dates; -> Option A

5. Quick Check:

   Date + INTERVAL '1 day' is standard SQL date addition [OK]

Hint: Use INTERVAL '1 day' to add days in standard SQL [OK]

Common Mistakes:

• Using database-specific date functions in standard SQL
• Incorrect stopping condition allowing extra dates
• Adding integer directly to date without INTERVAL