Type casting with :: operator in PostgreSQL - Mini Project: Build & Apply

Type Casting with :: Operator in PostgreSQL

📖 Scenario: You are working with a PostgreSQL database that stores product information. Some numeric values are stored as text, but you need to perform calculations on them.

🎯 Goal: Learn how to convert text data to numeric types using the :: type casting operator in PostgreSQL.

📋 What You'll Learn

Create a table with text columns containing numeric values

Add a configuration variable to specify a threshold price

Write a query that casts text to numeric using the :: operator

Complete the query to filter products based on the casted numeric value

💡 Why This Matters

🌍 Real World

In real databases, numeric data is sometimes stored as text. Casting lets you convert it to numbers for calculations.

💼 Career

Database developers and analysts often need to clean and convert data types to run accurate queries and reports.

1

Create the products table with text prices

Create a table called products with two columns: product_name as TEXT and price_text as TEXT. Insert these exact rows: ('Apple', '10'), ('Banana', '5'), ('Cherry', '15').

PostgreSQL

-- Create the products table and insert data
-- Your code here

Hint

Use CREATE TABLE to define the table and INSERT INTO to add rows.

2

Set a price threshold variable

Create a variable called price_threshold and set it to 8 using a WITH clause for use in the query.

PostgreSQL

CREATE TABLE products (
    product_name TEXT,
    price_text TEXT
);

INSERT INTO products (product_name, price_text) VALUES
('Apple', '10'),
('Banana', '5'),
('Cherry', '15');

-- Define price_threshold variable
-- Your code here

Hint

Use a WITH clause to create a named subquery that holds the threshold value.

3

Write a query casting price_text to integer

Write a SELECT query that selects product_name and casts price_text to INTEGER using the :: operator as price from the products table.

PostgreSQL

CREATE TABLE products (
    product_name TEXT,
    price_text TEXT
);

INSERT INTO products (product_name, price_text) VALUES
('Apple', '10'),
('Banana', '5'),
('Cherry', '15');

WITH price_threshold AS (SELECT 8 AS value)
-- Select product_name and cast price_text to integer as price
-- Your code here

Hint

Use price_text::INTEGER to convert the text to a number.

4

Filter products with price above threshold

Complete the query by joining with price_threshold and adding a WHERE clause to select only products where the casted price is greater than price_threshold.value.

PostgreSQL

CREATE TABLE products (
    product_name TEXT,
    price_text TEXT
);

INSERT INTO products (product_name, price_text) VALUES
('Apple', '10'),
('Banana', '5'),
('Cherry', '15');

WITH price_threshold AS (SELECT 8 AS value)
SELECT product_name, price_text::INTEGER AS price
FROM products
-- Join with price_threshold and filter price > price_threshold.value
-- Your code here

Hint

Use a comma join with price_threshold and a WHERE clause to filter.

Practice

1. What does the PostgreSQL expression '123'::integer do?

easy

A. Causes a syntax error because :: is invalid

B. Converts the integer 123 to text '123'

C. Converts the text '123' to an integer number 123

D. Leaves the value as text without any change

Solution

1. Step 1: Understand the :: operator

   The :: operator in PostgreSQL is used to cast or convert a value from one data type to another.

2. Step 2: Apply casting to the given expression

   The expression '123'::integer converts the string '123' into the integer 123.

3. Final Answer:

   Converts the text '123' to an integer number 123 -> Option C

4. Quick Check:

   Type casting '123'::integer = 123 [OK]

Hint: Remember :: changes type, here text to integer [OK]

Common Mistakes:

• Thinking :: converts integer to text
• Assuming :: is invalid syntax
• Believing the value stays as text

2. Which of the following is the correct syntax to cast a column named price to numeric in PostgreSQL?

easy

A. SELECT price AS numeric FROM products;

B. SELECT price to numeric FROM products;

C. SELECT CAST(price numeric) FROM products;

D. SELECT price::numeric FROM products;

Solution

1. Step 1: Recall correct casting syntax

   In PostgreSQL, the correct way to cast a value is using value::type or CAST(value AS type).

2. Step 2: Identify the correct option

   SELECT price::numeric FROM products; uses price::numeric, which is valid syntax. Options B, C, and D have incorrect syntax.

3. Final Answer:

   SELECT price::numeric FROM products; -> Option D

4. Quick Check:

   Correct cast syntax uses :: [OK]

Hint: Use ::type after value for casting [OK]

Common Mistakes:

• Using 'to' instead of :: or CAST
• Incorrect CAST syntax missing AS
• Using AS for casting instead of aliasing

3. What is the output of this query?

SELECT '2023-06-01'::date + 5;

medium

A. '2023-06-06' (date 5 days after June 1, 2023)

B. 2023-06-01 05:00:00 (timestamp with 5 hours added)

C. Error: cannot add integer to date

D. '2023-06-01 5' (string concatenation)

Solution

1. Step 1: Cast string to date

   The expression '2023-06-01'::date converts the text to a date value representing June 1, 2023.

2. Step 2: Add integer to date

   Adding 5 to a date adds 5 days, resulting in June 6, 2023.

3. Final Answer:

   '2023-06-06' (date 5 days after June 1, 2023) -> Option A

4. Quick Check:

   Date + integer adds days [OK]

Hint: Adding integer to date adds days [OK]

Common Mistakes:

• Thinking integer adds hours or seconds
• Expecting string concatenation
• Assuming error on adding integer to date

4. Identify the error in this query:

SELECT 'abc'::integer;

medium

A. Runtime error: invalid input syntax for integer

B. Syntax error due to invalid cast operator

C. Returns NULL because 'abc' is not a number

D. No error, returns 0

Solution

1. Step 1: Understand casting invalid text to integer

   Trying to cast a non-numeric string like 'abc' to integer causes a runtime error in PostgreSQL.

2. Step 2: Identify error type

   The error message is 'invalid input syntax for integer', indicating the cast failed at runtime.

3. Final Answer:

   Runtime error: invalid input syntax for integer -> Option A

4. Quick Check:

   Invalid cast causes runtime error [OK]

Hint: Casting non-numeric text to integer causes error [OK]

Common Mistakes:

• Expecting NULL instead of error
• Thinking syntax error instead of runtime error
• Assuming automatic conversion to 0

5. You have a table orders with a text column order_date storing dates as strings like '06/01/2023'. You want to find all orders after June 1, 2023. Which query correctly uses type casting with :: to achieve this?

hard

A. SELECT * FROM orders WHERE order_date > '2023-06-01'::date;

B. SELECT * FROM orders WHERE order_date::date > '2023-06-01';

C. SELECT * FROM orders WHERE order_date::text > '2023-06-01'::text;

D. SELECT * FROM orders WHERE order_date::integer > 20230601;

Solution

1. Step 1: Understand data types involved

   The column order_date is text, so to compare dates properly, it must be cast to date type.

2. Step 2: Choose correct casting and comparison

   SELECT * FROM orders WHERE order_date::date > '2023-06-01'; casts order_date to date and compares it to a date literal, which is correct. SELECT * FROM orders WHERE order_date > '2023-06-01'::date; compares text to date, which is invalid. SELECT * FROM orders WHERE order_date::text > '2023-06-01'::text; compares text to text, which is lexicographical, not date. SELECT * FROM orders WHERE order_date::integer > 20230601; tries to cast text to integer, which will cause error.

3. Final Answer:

   SELECT * FROM orders WHERE order_date::date > '2023-06-01'; -> Option B

4. Quick Check:

   Cast text column to date before date comparison [OK]

Hint: Cast text column to date before comparing dates [OK]

Common Mistakes:

• Comparing text directly to date
• Casting to wrong type like integer
• Comparing text lexicographically instead of date