|
|
Open Source .NET eXchange: Mike Hadlow on Implementing the Repository Pattern
Implementing the Repository Pattern
Mike Hadlow starts his talk by summarising the evolution of data access from the early days of using RDO and DAO with Visual Basic, which did not provide a lot of control over the database, to contemporary domain driven approaches. He goes on explaining the different meanings people tend to give to the term “Repository”.
He explains that, the Repository pattern is important for providing dependency-free access to data of any type. Repositories are responsible for persisting entities and value types. The core purpose of the repository is to encapsulate persistence. “The client should appear to be simply using an entity collection and all the details of object relational mapping and specific data access APIs should be hidden behind that collection like interface”.
Mike goes on with his thoughts on some questions like: Should we return IQuaryable objects from our repositories? Should repositories control transactions and who should consume the repositories?
More information, Mike Hadlow's Blog
You can download the slides here
ABOUT MIKE HADLOW
|
Mike Hadlow is a freelance programmer based in Brighton, specialising in enterprise application development with Microsoft tools. Mike is the author of several open source projects including WsdlWorks, a web service test tool and Suteki Shop, an MVC
More about Mike Hadlow
|
|
PODCAST IMPLEMENTING THE REPOSITORY PATTERN
This session took part at the Open Source .NET eXchange. You can view the other 7 podcasts here.
|
|
|
