CREATE TABLE with PostgreSQL types

Introduction

We use CREATE TABLE to make a new table in a database. Tables store data in rows and columns.

When you want to save information like users, products, or orders.

When starting a new project that needs organized data storage.

When you need to define what kind of data each column will hold.

When you want to set rules for data, like making sure a number is always positive.

When you want to prepare a database to accept data from an app or website.

Syntax

PostgreSQL

CREATE TABLE table_name (
  column_name1 data_type1 [constraints],
  column_name2 data_type2 [constraints],
  ...
);

Replace table_name with your table's name.

Choose data_type based on the kind of data you want to store, like INTEGER for numbers or TEXT for words.

Examples

This creates an employees table with an auto-incrementing ID, a name that cannot be empty, a salary with two decimals, and a hire date.

PostgreSQL

CREATE TABLE employees (
  id SERIAL PRIMARY KEY,
  name TEXT NOT NULL,
  salary NUMERIC(10,2),
  hire_date DATE
);

This table uses a unique UUID for product IDs, limits product names to 100 characters, stores price as money, and tracks if the product is in stock.

PostgreSQL

CREATE TABLE products (
  product_id UUID PRIMARY KEY,
  product_name VARCHAR(100) NOT NULL,
  price MONEY,
  in_stock BOOLEAN DEFAULT TRUE
);

Sample Program

This creates a books table with an auto-incrementing ID, title and author as text fields that cannot be empty, a year as a number, and a price with two decimals.

PostgreSQL

CREATE TABLE books (
  book_id SERIAL PRIMARY KEY,
  title VARCHAR(255) NOT NULL,
  author TEXT NOT NULL,
  published_year INTEGER,
  price NUMERIC(6,2)
);

Important Notes

SERIAL is a shortcut for auto-incrementing integers.

VARCHAR(n) limits text length to n characters.

Use NOT NULL to make sure a column always has a value.

Summary

CREATE TABLE makes a new table to store data.

Choose the right data_type for each column to match your data.

Use constraints like PRIMARY KEY and NOT NULL to keep data organized and valid.

Practice

1. Which PostgreSQL data type is best to store a person's full name in a CREATE TABLE statement?

easy

A. INTEGER

B. VARCHAR or TEXT

C. BOOLEAN

D. DATE

Solution

1. Step 1: Understand the data type purpose

   Names are text data, so we need a type that stores characters.

2. Step 2: Choose appropriate text type in PostgreSQL

   VARCHAR or TEXT types store strings of characters, perfect for names.

3. Final Answer:

   VARCHAR or TEXT -> Option B

4. Quick Check:

   Text data = VARCHAR or TEXT [OK]

Hint: Use VARCHAR or TEXT for any text data like names [OK]

Common Mistakes:

• Using INTEGER for text data
• Using BOOLEAN for names
• Using DATE for text fields

2. Which of the following is the correct syntax to create a table named employees with an integer primary key id and a text column name?

easy

A. CREATE TABLE employees (id INTEGER PRIMARY KEY, name TEXT);

B. CREATE employees TABLE (id INT PRIMARY, name STRING);

C. CREATE TABLE employees {id INT PRIMARY KEY, name TEXT};

D. CREATE TABLE employees (id INT, name TEXT PRIMARY KEY);

Solution

1. Step 1: Check the basic CREATE TABLE syntax

   PostgreSQL uses: CREATE TABLE table_name (column definitions); with commas separating columns.

2. Step 2: Verify data types and constraints

   INTEGER or INT is valid; PRIMARY KEY follows the column type; TEXT is valid for strings.

3. Final Answer:

   CREATE TABLE employees (id INTEGER PRIMARY KEY, name TEXT); -> Option A

4. Quick Check:

   Correct syntax and types = CREATE TABLE employees (id INTEGER PRIMARY KEY, name TEXT); [OK]

Hint: Use parentheses and commas correctly in CREATE TABLE [OK]

Common Mistakes:

• Wrong keyword order like PRIMARY before type
• Using curly braces instead of parentheses
• Wrong data types like STRING instead of TEXT

3. Given the table creation:
CREATE TABLE products (product_id SERIAL PRIMARY KEY, name VARCHAR(50) NOT NULL, price NUMERIC(10,2));
What will be the data type of product_id and what does SERIAL do?

medium

A. product_id is an auto-incrementing integer used as a unique ID

B. product_id is a string that stores numbers

C. product_id is a floating-point number

D. product_id is a boolean flag

Solution

1. Step 1: Understand SERIAL type in PostgreSQL

   SERIAL creates an integer column that auto-increments with each new row, perfect for unique IDs.

2. Step 2: Confirm data type and purpose

   It is an integer type, not string or float, used as a primary key to uniquely identify rows.

3. Final Answer:

   product_id is an auto-incrementing integer used as a unique ID -> Option A

4. Quick Check:

   SERIAL = auto-increment integer [OK]

Hint: SERIAL means auto-increment integer primary key [OK]

Common Mistakes:

• Thinking SERIAL is a string type
• Confusing SERIAL with floating-point
• Assuming SERIAL is a boolean

4. Identify the error in this table creation statement:
CREATE TABLE orders (order_id INT PRIMARY KEY, order_date DATE NOT NULL, total_amount FLOAT(10,2));

medium

A. Missing comma between columns

B. PRIMARY KEY cannot be used on INT columns

C. FLOAT(10,2) is invalid syntax in PostgreSQL

D. DATE cannot be NOT NULL

Solution

1. Step 1: Check FLOAT data type usage

   PostgreSQL FLOAT accepts at most one precision argument (1-53 digits); FLOAT(10,2) uses two arguments like NUMERIC, which is invalid syntax for FLOAT. Use NUMERIC(10,2) for precise money values.

2. Step 2: Verify other parts of the statement

   INT can be PRIMARY KEY, DATE can be NOT NULL, commas are correctly placed.

3. Final Answer:

   FLOAT(10,2) is invalid syntax in PostgreSQL -> Option C

4. Quick Check:

   FLOAT(p,s) invalid; use NUMERIC(p,s) [OK]

Hint: FLOAT(p) ok, but not FLOAT(p,s); use NUMERIC(p,s) for money [OK]

Common Mistakes:

• Using FLOAT(10,2) in PostgreSQL
• Thinking PRIMARY KEY can't be INT
• Assuming DATE can't be NOT NULL

5. You want to create a table users with columns: user_id as a UUID primary key, email as unique and not null, and created_at as timestamp with default current time. Which is the correct CREATE TABLE statement?

hard

A. CREATE TABLE users (user_id SERIAL PRIMARY KEY, email VARCHAR UNIQUE, created_at TIMESTAMP);

B. CREATE TABLE users (user_id UUID, email TEXT NOT NULL, created_at DATE DEFAULT NOW());

C. CREATE TABLE users (user_id UUID PRIMARY KEY, email VARCHAR(255) NOT NULL, created_at TIMESTAMP DEFAULT NOW());

D. CREATE TABLE users (user_id UUID PRIMARY KEY, email VARCHAR(255) UNIQUE NOT NULL, created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP);

Solution

1. Step 1: Check UUID primary key syntax

   UUID is a valid type for unique IDs; PRIMARY KEY constraint is correctly placed.

2. Step 2: Validate email constraints and type

   VARCHAR(255) with UNIQUE and NOT NULL ensures emails are unique and always present.

3. Step 3: Confirm timestamp default value

   DEFAULT CURRENT_TIMESTAMP sets creation time automatically; this is correct syntax.

4. Final Answer:

   CREATE TABLE users (user_id UUID PRIMARY KEY, email VARCHAR(255) UNIQUE NOT NULL, created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP); -> Option D

5. Quick Check:

   UUID PK, UNIQUE email, timestamp default = CREATE TABLE users (user_id UUID PRIMARY KEY, email VARCHAR(255) UNIQUE NOT NULL, created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP); [OK]

Hint: Use UUID for IDs, UNIQUE + NOT NULL for emails, CURRENT_TIMESTAMP for defaults [OK]

Common Mistakes:

• Using SERIAL instead of UUID for user_id
• Missing UNIQUE on email
• Using NOW() instead of CURRENT_TIMESTAMP in DEFAULT