List partitioning by category in PostgreSQL

Introduction

List partitioning helps organize data into smaller parts based on categories. This makes searching and managing data faster and easier.

You have a large table of products and want to separate them by product type.

You want to store customer data divided by country or region.

You manage orders and want to split them by order status like 'pending', 'shipped', or 'cancelled'.

You want to improve query speed by only searching relevant categories.

You want to keep data organized and easier to maintain.

Syntax

PostgreSQL

CREATE TABLE parent_table (
  id SERIAL PRIMARY KEY,
  category TEXT NOT NULL,
  data TEXT
) PARTITION BY LIST (category);

CREATE TABLE partition_name PARTITION OF parent_table
  FOR VALUES IN ('category_value1', 'category_value2');

The parent table holds the structure but no data directly.

Each partition stores rows for specific category values.

Examples

This example creates a products table partitioned by category with two partitions: Electronics and Clothing.

PostgreSQL

CREATE TABLE products (
  product_id SERIAL PRIMARY KEY,
  category TEXT NOT NULL,
  name TEXT
) PARTITION BY LIST (category);

CREATE TABLE products_electronics PARTITION OF products
  FOR VALUES IN ('Electronics');

CREATE TABLE products_clothing PARTITION OF products
  FOR VALUES IN ('Clothing');

Orders table is partitioned by order status with partitions for Pending and Completed orders.

PostgreSQL

CREATE TABLE orders (
  order_id SERIAL PRIMARY KEY,
  status TEXT NOT NULL,
  amount NUMERIC
) PARTITION BY LIST (status);

CREATE TABLE orders_pending PARTITION OF orders
  FOR VALUES IN ('Pending');

CREATE TABLE orders_completed PARTITION OF orders
  FOR VALUES IN ('Completed');

Sample Program

This program creates a customer_data table partitioned by region with USA and Europe partitions. It inserts three customers and selects all rows ordered by ID.

PostgreSQL

CREATE TABLE customer_data (
  customer_id SERIAL PRIMARY KEY,
  region TEXT NOT NULL,
  name TEXT
) PARTITION BY LIST (region);

CREATE TABLE customer_data_usa PARTITION OF customer_data
  FOR VALUES IN ('USA');

CREATE TABLE customer_data_europe PARTITION OF customer_data
  FOR VALUES IN ('Europe');

INSERT INTO customer_data (region, name) VALUES
  ('USA', 'Alice'),
  ('Europe', 'Bob'),
  ('USA', 'Charlie');

SELECT * FROM customer_data ORDER BY customer_id;

Important Notes

List partitioning improves query speed by limiting search to relevant partitions.

Each partition is a separate table storing only rows for its category values.

Be careful to include all possible category values in partitions to avoid errors.

Summary

List partitioning splits a table into parts based on category values.

It helps organize data and speeds up queries by focusing on relevant partitions.

Each partition holds rows for specific category values defined in the parent table.

Practice

1. What is the main purpose of list partitioning by category in PostgreSQL?

easy

A. To split a table into parts based on specific category values

B. To combine multiple tables into one large table

C. To encrypt data in the table for security

D. To create temporary tables for faster queries

Solution

1. Step 1: Understand list partitioning concept

   List partitioning divides a table into smaller parts based on specific category values.

2. Step 2: Identify the main purpose

   This helps organize data and speeds up queries by focusing on relevant partitions.

3. Final Answer:

   To split a table into parts based on specific category values -> Option A

4. Quick Check:

   List partitioning = split by category [OK]

Hint: List partitioning splits tables by category values [OK]

Common Mistakes:

• Confusing list partitioning with table joins
• Thinking it encrypts data
• Assuming it merges tables

2. Which of the following is the correct syntax to create a list partitioned table by category in PostgreSQL?

easy

A. CREATE TABLE sales PARTITION BY GROUP (category);

B. CREATE TABLE sales PARTITION BY RANGE (category);

C. CREATE TABLE sales PARTITION BY HASH (category);

D. CREATE TABLE sales PARTITION BY LIST (category);

Solution

1. Step 1: Identify partition type syntax

   List partitioning uses PARTITION BY LIST in PostgreSQL.

2. Step 2: Match correct syntax

   Only CREATE TABLE sales PARTITION BY LIST (category); uses PARTITION BY LIST with the column name correctly.

3. Final Answer:

   CREATE TABLE sales PARTITION BY LIST (category); -> Option D

4. Quick Check:

   List partitioning syntax = PARTITION BY LIST [OK]

Hint: Use PARTITION BY LIST for list partitioning [OK]

Common Mistakes:

• Using PARTITION BY RANGE instead of LIST
• Using PARTITION BY HASH incorrectly
• Using non-existent PARTITION BY GROUP

3. Given the following setup:

CREATE TABLE products (
  id SERIAL PRIMARY KEY,
  name TEXT,
  category TEXT
) PARTITION BY LIST (category);

CREATE TABLE products_electronics PARTITION OF products FOR VALUES IN ('electronics');
CREATE TABLE products_clothing PARTITION OF products FOR VALUES IN ('clothing');

INSERT INTO products (name, category) VALUES ('Phone', 'electronics'), ('Shirt', 'clothing');

SELECT * FROM products WHERE category = 'electronics';

What will the SELECT query return?

medium

A. No rows because category is filtered incorrectly

B. All rows from both partitions

C. Only rows where category is 'electronics', here the 'Phone' row

D. An error because partitions are not queried directly

Solution

1. Step 1: Understand partition filtering

   Query filters category = 'electronics', so only that partition is scanned.

2. Step 2: Check inserted data

   Only 'Phone' has category 'electronics', so only that row is returned.

3. Final Answer:

   Only rows where category is 'electronics', here the 'Phone' row -> Option C

4. Quick Check:

   Partition filter returns matching rows [OK]

Hint: Query filters partition by category value [OK]

Common Mistakes:

• Expecting all rows without filter
• Thinking query causes error
• Assuming partitions must be queried separately

4. Consider this incorrect partition creation:

CREATE TABLE orders (
  id SERIAL PRIMARY KEY,
  product TEXT,
  category TEXT
) PARTITION BY LIST (category);

CREATE TABLE orders_electronics PARTITION OF orders FOR VALUES IN ('electronics', 'gadgets');

What is the error in this partition definition?

medium

A. FOR VALUES IN must list values as separate strings, not combined

B. Partitions cannot have multiple values in FOR VALUES IN clause

C. The partition table name is invalid

D. The parent table cannot be partitioned by LIST

Solution

1. Step 1: Check FOR VALUES IN syntax

   FOR VALUES IN expects a list of values as separate strings, e.g. ('electronics', 'gadgets').

2. Step 2: Identify error in given code

   The code uses multiple values in one partition which is not allowed in PostgreSQL list partitioning.

3. Final Answer:

   Partitions cannot have multiple values in FOR VALUES IN clause -> Option B

4. Quick Check:

   Each partition must have exactly one value [OK]

Hint: Each partition must have a single value in FOR VALUES IN clause [OK]

Common Mistakes:

• Using one string with commas inside
• Misnaming partition tables
• Thinking LIST partitioning is not allowed

5. You want to create a list partitioned table events by event_type with partitions for 'login', 'logout', and 'purchase'. Which of the following is the correct way to create the partitions and insert a new 'purchase' event?

hard

A. CREATE TABLE events PARTITION BY LIST (event_type);
CREATE TABLE events_login PARTITION OF events FOR VALUES IN ('login');
CREATE TABLE events_logout PARTITION OF events FOR VALUES IN ('logout');
CREATE TABLE events_purchase PARTITION OF events FOR VALUES IN ('purchase');
INSERT INTO events (event_type) VALUES ('purchase');

B. CREATE TABLE events PARTITION BY RANGE (event_type);
CREATE TABLE events_login PARTITION OF events FOR VALUES FROM ('login') TO ('logout');
INSERT INTO events (event_type) VALUES ('purchase');

C. CREATE TABLE events PARTITION BY LIST (event_type);
CREATE TABLE events_all PARTITION OF events FOR VALUES IN ('login', 'logout', 'purchase');
INSERT INTO events (event_type) VALUES ('purchase');

D. CREATE TABLE events PARTITION BY HASH (event_type);
CREATE TABLE events_hash PARTITION OF events FOR VALUES IN (1);
INSERT INTO events (event_type) VALUES ('purchase');

Solution

1. Step 1: Choose correct partition type and syntax

   List partitioning by event_type requires PARTITION BY LIST and partitions with FOR VALUES IN for each category.

2. Step 2: Verify partitions and insert

   CREATE TABLE events PARTITION BY LIST (event_type); CREATE TABLE events_login PARTITION OF events FOR VALUES IN ('login'); CREATE TABLE events_logout PARTITION OF events FOR VALUES IN ('logout'); CREATE TABLE events_purchase PARTITION OF events FOR VALUES IN ('purchase'); INSERT INTO events (event_type) VALUES ('purchase'); correctly creates partitions for 'login', 'logout', and 'purchase' and inserts a 'purchase' event into the parent table.

3. Final Answer:

   Option A with correct list partitions and insert -> Option A

4. Quick Check:

   List partitions per category + insert into parent [OK]

Hint: Create partitions per category, insert into parent table [OK]

Common Mistakes:

• Using RANGE or HASH instead of LIST
• Creating one partition for all values
• Inserting into partitions directly instead of parent