As promised by the author, this book starts with code (“Hello LINQ”), ends with code (a query joining database data with XML data), and is code-heavy in the middle. All the code runs and the output is listed in the book. The code snippets you can steal will quickly recoup the price of the book.

It’s divided into five parts: Introduction, LINQ to Objects, LINQ to XML, LINQ to DataSet, and LINQ to SQL.

Part One has two chapters. In the first chapter, “Hello LINQ,” the author shows how to use LINQ to query XML and SQL databases, and adds that LINQ can also work with DataSets, object, and Entities. He also points out that LINQ (Language Integrated Query) is more of a data-iteration engine, but that Microsoft didn’t want to name its hot new technology “DIE”! Yep, Joseph has a sense of humor, and it pops up from time to time in this book.

The second chapter covers the language enhancements in C# and VB.NET to support LINQ. Note I didn’t specify the version of C# or VB.NET. That’s because the building blocks of LINQ have been being added in one by one since .NET version 2.0. In some cases the author goes into some detail on what problem the language addition was meant to address, and how the addition addresses the problem. The chapter ends with eight pages of details on how the compiler translates LINQ into C# or .NET code. Joseph admits that you’re about as likely to need this information as you need to know what machine code is generated for a foreach loop, but I find it interesting, and I’m impressed that he dug the information up.

The second part covers LINQ to objects, serving also as an overall introduction to LINQ. After a quick intro to LINQ in Chapter Three, Chapter Four covers deferred operations, so-called because they return a type of IEnumerable<T> that will be iterated through at a later time. Chapter Five covers non-deferred types, so-called because they immediately return values of types other than IEnumerable<T>. An example of this would be conversion routines such as ToList.

Part three covers LINQ to XML, with Chapter Six kicking things off with a description of creating a simple XML document using DOM to show the amount of code involved so we can see how big an improvement LINQ to XML is. Chapter Seven covers the LINQ XML API, and discusses the “Halloween” problem, so-called because it was first discussed by experts on Halloween. The problem occurs if part of an XML document is modified while a LINQ query is still iterating over it. Chapter Eight covers LINQ-to-XML operators, such as “DescendantNodes,” which return a sequence containing the descendant nodes of each element or document. Chapter Nine covers transformations and validation. If you plan on using LINQ and XML, the code snippets in this section are worth their weight in gold.

Part four consist of two very short chapters covering LINQ to DataSet. The author chose to cover LINQ DataSet operations before covering LINQ to SQL because LINQ to SQL only works with Microsoft SQL Server, where you can retrieve DataSets from almost any database and then use LINQ to query the results. Chapter 10 covers DataRow and DataTable operators. Chapter 11 quickly covers typed DataSets. As I said, it’s a short section.

The fifth and final part covers LINQ to SQL. Chapter 12 is another short chapter that introduces us to LINQ to SQL. The author points out that most technologies that link relational databases to business classes lose the ability to do queries against the classes. He then discusses a bit about how entity-mapping classes solve this problem by allowing LINQ to run queries and even updates against classes created from SQL databases.

In Chapter 13 the author gives tips and discusses tools. Joseph points out that if we don’t understand his tips now, he’s done his job, because he wants to give us tips before we need them, not after we’ve learned we need them the hard way. The tools he discusses are SQLMetal and Object Relational Designer. Both are included in the .NET 3.5 Framework.

Chapter 14, “LINQ to SQL Database Operations,” is described as “a whirlwind tour of standard database operations.” It covers such things as how LINQ-to-SQL queries are executed in the database, and the issues caused by the fact that LINQ stands for “Language Integrated Query,” where queries are written as close to the programming language (C# in our case) as possible, but have to be translated into true SQL for execution. This, for example,

where c.CustomerID.TrimEnd(‘k’) == “Lazy”

throws an error because TrimEnd is not valid SQL.

where c.CustomerID == “Lazy”. TrimEnd(‘k’)

executes without error because TrimEnd is a valid .NET function. Understand the difference? If not, buy the book and let Joseph explain it. Simply put, in this case the left side is evaluated by SQL and the right side is evaluated by C#.

Chapter 15 covers LINQ-to-SQL entity classes, including generating entity classes, extending entity classes with partial methods, and important System.Data.Linq API classes. Chapter 16 is a long chapter covering the details of the DataContext class, including example code for many of its functions. Chapter 17 covers detecting and resolving concurrency conflicts including pessimistic concurrency and transactions. The final chapter, Chapter 18, covers views, entity class inheritance, and more information on transactions.

This is a well-written book, full of tips and traps, with enough source code snippets to make it worth buying for that reason alone. It’s value-packed with information from someone who clearly knows what he’s talking about. It’s one of the better books I’ve seen on LINQ and will be a welcome addition to my bookshelf.

Title: Pro LINQ: Language Integrated Query in C# 2008
Author: Joseph C. Rattz, Jr.
Publisher: Apress
ISBN: 978-1-59059-789-7
Price: U.S. $44.99