Apress - Pro LINQ Object Relational Mapping In CSharp 2008

Introduction

It is nearly impossible today to write enterprise software without the use of one or more relational databases. Granted, in some cases the data is transient and not stored in a database, but for the most part software needs to consume and manipulate the data in a database. Easy enough, right? You put the lime in the coconut and you’ve got yourself a data-aware software application. Wrong! There are hundreds of ways to connect software systems to databases and thousands of people who think they have the skeleton key for data access layers. I can’t say that I have the end-all pattern for data access, but I do have an efficient, repeatable way to apply industry design patterns to build scalable object-oriented persistence layers.

Object-relational mapping (ORM) has been a gray area in Microsoft development for many years. It’s not that Microsoft language developers didn’t understand ORM; in fact, the opposite is true, as is exemplified by the glut of third-party .NET ORM tools on the market. The struggle has come more from the lack of native tools with the object-oriented and object-persistence capacity to effectively work in this arena. With the inception of .NET, Microsoft overcame the first obstacle by developing an object-oriented environment and framework. The second obstacle, the native object persistence layer, is only now being realized with the introduction of the upcoming data access enhancements in Visual Studio 2008. The gray area is no longer gray, and the .NET developers of the world finally have the native tools required to build modular, reusable data access layers.

Working as an architect and consultant, I have noticed a severe dearth in the .NET community when it comes to the finer points of using design patterns to build data access layers. The Java camp has followed the patterns = reuse mantra for a long time, but the .NET side of the house is just starting to move in that direction. After scouring the Internet and bookstores, I have been shocked at how few books address using object-relational mapping patterns with .NET. The idea for this book has been in the back of my mind for a while, but I was always hesitant because of the deficiency in the native Microsoft tools. Now, with the Language-Integrated Query (LINQ) suite and the ADO.NET Entity Framework (EF), the object-relational mapping pattern can finally be realized in the .NET space. Although there are numerous books about LINQ, this book goes further and ties together the use of ORM design patterns with LINQ and Visual Studio 2008.

Before the naysayers start in on me for not writing this entire book about the ADO.NET EF, the “true” ORM tool that Microsoft is developing, let me say that I understand that EF is expected to be a far more sophisticated ORM tool than [LINQ] to SQL, the .NET Language-Integrated Query for Relational Databases, a subset of the LINQ suite. I also understand that some people are cursing my name right now because I’m calling EF an ORM, but a cat is a cat even if you shave off its hair and call it a dog. Bottom line, with VS 2008 there are two ORM tools: LINQ to SQL, which is not getting the recognition it deserves, and EF, which might be getting too much attention. The focus of this book is ORM with LINQ and C# 2008. This includes EF and LINQ to SQL, and therefore this text covers both.

This text can be utilized as a practical guide for ORM with the .NET Framework. Although some of this book is based on theory and design patterns, the focus is not an academic or theoretical discussion. However, it is important for everyone who is using an ORM tool to understand that certain principles and patterns lay the foundation of what you are doing. After reading this text, you will have knowledge of ORM and LINQ, and knowledge of the patterns you need to write robust software applications. Additionally, by walking through some real-world examples, you will have the tools you need to move forward in developing ORM solutions in .NET.