An Introduction to Enterprise Design Patterns
0 likes86 views
This document provides an overview of several enterprise design patterns, including dependency injection, the onion architecture, the repository pattern, unit of work pattern, and view models. It discusses the benefits of each pattern such as loose coupling, increased testability and code reuse. It also provides examples of implementing the repository and unit of work patterns when accessing data with Entity Framework.
An Introduction to Enterprise Design Patterns
- 1. Enterprise Design Patterns Intro Designing your app the correct way – SSW Module 1725
- 2. Module Overview 1. Intro to Dependency Injection 2. Intro to the Onion Architecture 3. Intro to Enterprise Data Access a) Repository Pattern b) Unit of Work Pattern 4. View Models
- 3. Are you using Dependency Injection? Tried it? Like it? Used it in a project?
- 4. Dependencies
- 7. The answer: Dependency Injection Loosely coupled classes Increased code reuse Maintainable code Testable methods All dependencies are specified in one place Class dependencies are clearly visible in the constructor
- 8. Lets look at the OrderProcessor. Spot the Code Smell ?
- 9. Look Mum. No Dependencies !
- 10. Are you using Dependency Injection? Tried it? Like it? Used for new project? Refactored existing code? Implemented everywhere?
- 12. Enterprise Data Access Patterns
- 13. The Provided Solution Structure Our Solution Structure
- 14. Repository
- 15. Directly Accessing Data Duplicated code / No code re-use Hard to centralize data-related policies (i.e. caching) Hard to write unit tests
- 16. IIS Controller EF DBContext Database No Repository Direct access to the database from the controller IIS Controller EF Repository: IRepository Database With Repository Abstraction layer between controller and database context. Unit tests can use a custom persistence later to facilitate testing Test Project Controller Mock Repository : IRepository Mock Persistence
- 17. Use the Repository Pattern Isolates the data layer better unit testing support Centralised, consistent access rules, logic and caching Maintainability improved code re-use Readability e.g. GetTopSellingItemsByCategory – Simple method name, complex SQL Query http://blog.damianbrady.com.au/2012/03/07/a-generic-crud-repository-for-entity-framework/ http://rules.ssw.com.au/SoftwareDevelopment/RulestobetterArchitectureandCodeReview/Pages/RepositoryPattern.aspx
- 18. IIS Controller LEGACY DATABASE: IRepository Database Use Repositories to abstract legacy data access Test Project Controller Mock Repository : IRepository Mock Persistence IIS Controller EF Repository: IRepository Database New Functionality Repositories – Migrating to modern data access
- 19. Implementing the Repository Pattern with Code First
- 21. Generic Repository - Overview
- 22. Generic Repository – Db Context
- 23. Generic Repository - Implementation
- 24. The Repositories
- 25. Repositories in the store controller
- 26. Unit of Work Coordinates multiple repositories with a single db context UOW.SaveChanges() commits the changes made to all of the repositories
- 27. IIS Controller Unit of Work Database Unit of Work Ensures that multiple repositories share a single database context Test Project Controller Mock Unit Of Work Mock Persistence Repository Mock Repository Repository Mock Repository DbContext
- 28. Implementing the Unit Of Work Pattern
- 30. Where Is My Context?
- 31. Where the Magic Happens….. Our Dependency Injection Container !
- 32. Implementing the Repository Pattern with EDMX
- 33. I Need Some Context
- 34. Code First Context vs Edmx Generated Context
- 35. Generating Entities In My Domain Project From the EDMX
- 36. Update the T4 template to point to the EDMX
- 37. DB First - Repository + Unit of Work Entity Framework 4 instructions: http://blogs.msdn.com/b/adonet/archive/2009/05/21/poco-in-the-entity- framework-part-1-the-experience.aspx
- 38. View Models models that describe a view pass data from a controller to a view. provides a convenient way of representing complex view data
- 39. Benefits of View Models Remove logic from your views – viewmodels are easily unit tested Security – remove properties from models you don’t want in the view Loose coupling –views are not coupled to the domain DataAnnotations describe the model
- 41. Summary 1. Intro to Dependency Injection 2. Intro to the Onion Architecture 3. Intro to Enterprise Data Access a) Repository Pattern b) Unit of Work Pattern 4. View Models