Welcome back! In the previous lessons, you learned how to set up your project and design database models for recipes, ingredients, and reviews. As you continue building your AI Cooking Helper, you might find yourself needing to clear out all the information in your database. The process of resetting your database is common during development and testing when you want to start fresh or remove test data that is no longer needed.

Resetting your database means deleting all the records from your tables. This helps you:

• Remove test or sample data before finalizing your application.
• Fix mistakes if you accidentally added incorrect or malformed data.
• Start over with a clean slate to test new features.

In this lesson, you will learn how to create a management script that safely deletes all recipes, ingredients, reviews, and their relationships. Since this is a destructive action, we will also implement safety checks to prevent accidental data loss.

Before we write the script, let’s review our database setup. You previously created models to represent the core parts of the application:

• Recipe
• Ingredient
• Review
• RecipeIngredient (The intermediate model connecting recipes and ingredients)

The RecipeIngredient model is particularly important because it handles the many-to-many relationship. It links specific ingredients to specific recipes, often including details such as the amount needed.

Here is a summary of the relationships:

Deleting all data is a permanent step. Once the records are gone, they cannot be recovered without a backup. To prevent accidents, it is standard practice to add a confirmation prompt.

You can use the input() function to pause the script and wait for the user to confirm their intent:

• The script asks the user to type y.
• If the user types anything else (or just presses Enter), the script stops and prints Aborted.

When deleting related data, we want to ensure the operation is "atomic." This means atomic transactions ensure that either all the data is deleted successfully or none of it is. This prevents a situation where your ingredients are deleted but your recipes remain, leaving your database in a confusing, half-empty state.

By using a transaction block, we group our deletion commands together. If something goes wrong during the process, the database rolls back to its original state.

Even within a transaction, it is best practice to delete data in an order that respects relationships:

1. RecipeIngredient: Delete the links between recipes and ingredients first.
2. Review: Delete the reviews associated with recipes.
3. Recipe: Delete the main recipe records.
4. Ingredient: Delete the ingredient records.

Using the model manager, we can target all records at once: RecipeIngredient.objects.all().delete()

Let’s look at how to build the complete script. We will place this in a scripts folder within your project.

Since this script runs independently from your main web server, you must perform an environment setup to tell it where to find your project settings and initialize the application systems.

• This setup ensures that when you import your models, the script knows which database to connect to and how to handle the data.

Once the environment is ready, we import the necessary tools and models to perform the reset logic.

• transaction.atomic(): Wraps the deletions in an atomic block. If the script crashes halfway through, no data is actually deleted.
• objects.all().delete(): This is the standard way to clear a table.

Finally, we ensure the function runs when the script is executed from the command line:

If you run the reset_database.py script and confirm:

If you change your mind:

In this lesson, you learned how to manage your database data by:

• Setting up an external script that initializes your project's environment and settings.
• Implementing a safety prompt to prevent accidental data loss.
• Using atomic transactions to ensure all-or-nothing data deletion.
• Clearing specific tables using the model manager in the correct order.

This script is a powerful utility that you will use frequently during the development phase. In the next part of the course, we will start looking at how to populate our database with fresh, structured data using similar scripting techniques.