2.3.2. Database Projects in Visual Studio Code

💡 First Principle: Database projects enable schema management as code, supporting CI/CD for data warehouse schemas separate from the data itself. Instead of making changes directly in production (terrifying), you modify SQL files locally, test them, and deploy through a controlled pipeline.

Scenario: Your data warehouse schema needs to evolve: new columns, modified procedures, updated views. Database projects let you manage these changes in VS Code, test them, and deploy through controlled pipelines.

Creating a Database Project

1. Connect to Fabric Warehouse in VS Code
2. Right-click connection → Create Project From Database
3. Schema imported as SQL files (tables, views, procedures)
4. Modify schema in VS Code
5. Deploy changes back to Fabric

⚠️ Exam Trap: The correct action is Create Project From Database, not "Manage Packages." Managing packages is for library dependencies, not schema extraction. This is a common distractor in exam questions.

Key Trade-Offs:

• Full Schema Import vs. Selective: Importing entire schema captures everything but may include unwanted objects
• Immediate Deploy vs. Pipeline Deploy: Direct deployment is faster for development; pipeline deployment is safer for production