Professional ADO.NET 3.5 with LINQ and the Entity Framework

Language Integrated Query (LINQ), as well as the C# 3.0 and VB 9.0 language extensions to support it, is the most import single new feature of Visual Studio 2008 and the .NET Framework 3.x. LINQ is Microsoft’s first attempt to define a universal query language for a diverse set of in-memory collections of generic objects, entities persisted in relational database tables, and element and attributes of XML documents or fragments, as well as a wide variety of other data types, such as RSS and Atom syndication feeds. Microsoft invested millions of dollars in Anders Hejlsberg and his C# design and development groups to add new features to C# 3.0—such as lambda expressions, anonymous types, and extension methods—specifically to support LINQ Standard Query Operators (SQOs) and query expressions as a part of the language itself.

Corresponding additions to VB 9.0 followed the C# team’s lead, but VB’s implementation of LINQ to XML offers a remarkable new addition to the language: XML literals. VB’s LINQ to XML implementation includes XML literals, which treat well-formed XML documents or fragments as part of the VB language, rather than requiring translation of element and attribute names and values from strings to XML DOM nodes and values.

This book concentrates on hands-on development of practical Windows and Web applications that demonstrate C# and VB programming techniques to bring you up to speed on LINQ technologies. The first half of the book covers LINQ Standard Query Operators (SQOs) and the concrete implementations of LINQ for querying collections that implement generic IEnumerable, IQueryable, or both interfaces. The second half is devoted to the ADO.NET Entity Framework, Entity Data Model, Entity SQL (eSQL) and LINQ to Entities. Most code examples emulate real-world data sources, such as the Northwind sample database running on SQL Server 2005 or 2008 Express Edition, and collections derived from its tables. Code examples are C# and VB Windows form or Web site/application projects not, except in the first chapter, simple command-line projects. You can’t gain a feel for the behavior or performance of LINQ queries with “Hello World” projects that process arrays of a few integers or a few first and last names.

This book is intended for experienced .NET developers using C# or VB who want to gain the maximum advantage from the query-processing capabilities of LINQ implementations in Visual Studio 2008—LINQ to Objects, LINQ to SQL, LINQ to DataSets, and LINQ to XML—as well as the object/relational mapping (O/RM) features of VS 2008 SP1’s Entity Framework/Entity Data Model and LINQ to Entities and the increasing number of open-source LINQ implementations by third-party developers.

Basic familiarity with generics and other language features introduced by .NET 2.0, the Visual Studio integrated development environment (IDE), and relational database management systems (RDBMSs), especially Microsoft SQL Server 200x, is assumed. Experience with SQL Server’s Transact-SQL (T-SQL) query language and stored procedures will be helpful but is not required. Proficiency with VS 2005, .NET 2.0, C# 2.0, or VB 8.0 will aid your initial understanding of the book’s C# 3.0 or VB 9.0 code samples but isn’t a prerequisite.

Microsoft’s .NET code samples are primarily written in C#. All code samples in this book’s chapters and sample projects have C# and VB versions unless they’re written in T-SQL or JavaScript.

Professional ADO.NET 3.5: LINQ and the Entity Framework concentrates on programming the System.Linq and System.Linq.Expressions namespaces for LINQ to Objects, System.Data.Linq for LINQ to SQL, System.Data.Linq for LINQ to DataSet, System.Xml.Linq for LINQ to XML, and System.Data.Entity and System.Web.Entity for EF’s Entity SQL.

• “Taking a New Approach to Data Access in ADO.NET 3.5,” uses simple C# and VB code examples to demonstrate LINQ to Objects queries against in-memory objects and databinding with LINQ-populated generic List collections, object/relational mapping (O/RM) with LINQ to SQL, joining DataTables with LINQ to DataSets, creating EntitySets with LINQ to Entities, querying and manipulating XML InfoSets with LINQ to XML, and performing queries against strongly typed XML documents with LINQ to XSD.

• “Understanding LINQ Architecture and Implementation,” begins with the namespaces and C# and VB language extensions to support LINQ, LINQ Standard Query Operators (SQOs), expression trees and compiled queries, and a preview of domain-specific implementations. C# and VB sample projects demonstrate object, array, and collection initializers, extension methods, anonymous types, predicates, lambda expressions, and simple query expressions.

• “Executing LINQ Query Expressions with LINQ to Objects,” classifies the 50 SQOs into operator groups: Restriction, Projection, Partitioning, Join, Concatenation, Ordering, Grouping, Set, Conversion, and Equality, and then lists their keywords in C# and VB. VS 2008 SP1 includes C# and VB versions of the LINQ Project Sample Query Explorer, but the two Explorers don’t use real-world collections as data sources. This describes a LINQ in-memory object generator (LIMOG) utility program that writes C# 3.0 or VB 9.0 class declarations for representative business objects that are more complex than those used by the LINQ Project Sample Query Explorers. Sample C# and VB queries with these business objects as data sources are more expressive than those using a arrays of a few integers or last names.

• “Working with Advanced Query Operators and Expressions,” introduces LINQ queries against object graphs with entities that have related (associated) entities. This begins with examples of aggregate operators, explains use of the Let temporary local variable operator, shows you how to use Group By with aggregate queries, conduct the equivalent of left outer joins, and take advantage of the Contains() SQO to emulate SQL’s IN() function. You learn how to compile queries for improved performance, and create mock object classes for testing without the overhead of queries against relational persistence stores.

• “Using LINQ to SQL and the LinqDataSource,” introduces LINQ to SQL as Microsoft’s first O/RM tool to reach released products status and shows you how to autogenerate class files for entity types with the graphical O/R Designer or command-line SqlMetal.exe. This also explains how to edit *.dbml mapping files in the Designer or XML Editor, instantiate DataContext objects, and use LINQ to SQL as a Data Access Layer (DAL) with T-SQL queries or stored procedures. Closes with a tutorial for using the ASP.NET LinqDataSource control with Web sites or applications.

• “Querying DataTables with LINQ to DataSets,” begins with a comparison of DataSet and DataContext objects and features, followed by a description of the DataSetExtensions. Next comes querying untyped and typed DataSets, creating lookup lists, and generating LinqDataViews for databinding with the AsDataView() method. This ends with a tutorial that shows you how to copy LINQ query results to DataTables.

• “Manipulating Documents with LINQ to XML,” describes one of LINQ most powerful capabilities: managing XML Infosets. This demonstrates that LINQ to XML has query and navigation capabilities that equal or surpasses XQuery 1.0 and XPath 2.0. It also shows LINQ to XML document transformation can replace XQuery and XSLT 1.0+ in the majority of common use cases. You learn how to use VB 9.0’s XML literals to constructs XML documents, use GroupJoin() to produce hierarchical documents, and work with XML namespaces in C# and VB.

• “Exploring Third-Party and Emerging LINQ Implementations,” describes Microsoft’s Parallel LINQ (also called PLINQ) for taking advantage of multiple CPU cores in LINQ to Objects queries, LINQ to REST for translating LINQ queries into Representational State Transfer URLs that define requests to a Web service with the HTML GET, POST, PUT, and DELETE methods, and Bart De Smet’s LINQ to Active Directory and LINQ to SharePoint third-party implementations.

• “Raising the Level of Data Abstraction with the Entity Data Model,” starts with a guided tour of the development of EDM and EF as an O/RM tool and heir apparent to ADO.NET DataSets, provides a brief description of the entity-relationship (E-R) data model and diagrams, and then delivers a detailed analysis of EF architecture. Next comes an introduction to the Entity SQL (eSQL) language, eSQL queries, client views, and Object Services, including the ObjectContext, MetadataWorkspace, and ObjectStateManager. Later chapters describe eSQL and these objects in greater detail. Two C# and VB sample projects expand on the eSQL query and Object Services sample code.

• “Defining Conceptual, Mapping, and Storage Schema Layers,” provides detailed insight into the structure of the *.edmx file that generates the *.ssdl (storage schema data language), *.msl (mapping schema language), and *.csdl files at runtime. You learn how to edit the *.edmx file manually to accommodate modifications that the graphic EDM Designer can’t handle. You learn how to implement the Table-per-Hierarchy (TPH) inheritance model and traverse the MetadataWorkspace to obtain property values. Four C# and VB sample projects demonstrate mapping, substituting stored procedures for queries, and TPH inheritance.

• “Introducing Entity SQL,” examines EF’s new eSQL dialect that adds keywords to address the differences between querying entities and relational tables. You learn to use Zlatko Michaelov’s eBlast utility to write and analyze eSQL queries, then dig into differences between eSQL and T-SQL SELECT queries. (eSQL v1 doesn’t support INSERT, UPDATE, DELETE and other SQL Data Manipulation Language constructs). You execute eSQL queries against the EntityClient, measure the performance hit of eSQL compared to T-SQL, execute parameterize eSQL queries, and use SQL Server Compact 3.5 as a data store. C# and VB Sample projects demonstrate the programming techniques.

• “Taking Advantage of Object Services and LINQ to Entities,” concentrates manipulating the Object Services API’s ObjectContext. It continues with demonstrating use of partial classes for the ModelNameEntities and EntityName objects, executing eSQL ObjectQuerys, and deferred or eager loading of associated entities, including ordering and filtering the associated entities. Also covers instructions for composing QueryBuilder methods for ObjectQuerys, LINQ to Entities queries, and parameterizing ObjectQuerys.

• “Updating Entities and Complex Types,” shows you how to perform create, update, and delete (CUD) operations on EntitySets and manage optimistic concurrency conflicts. It starts with a detailed description of the ObjectContext.ObjectStateManager and its child objects, which perform object identification and change tracking operations with EntityKeys. This also covers validation of create and update operations, optimizing the DataContext lifetime, performing updates with stored procedures, and working with complex types.

• "Binding Data Controls to the ObjectContext", describes creating design-time data sources from ObjectContext.EntitySet instances, drag-and-drop addition of BindingNavigator, BindingSource, bound TextBox, and DataGridView controls to Windows forms. You also learn how to update EntityReference and EntitySet values with ComboBox columns in DataGridView controls. (You can’t update EntitySet values directly; you must delete and add a new member having the required value). This concludes with a demonstration of the use of the ASP.NET EntityDataSource control bound to GridView and DropDownList controls.

• “Using the Entity Framework As a Data Source,” concentrates on using EF as a data source for the ADO.NET Data Services Framework (the former codename “Project Astoria” remains in common use), which is the preferred method for deploying EF v1 as a Web service provider. (EF v2 is expected to be able to support n-tier data access with Windows Communication Foundation [WCF] directly). A Windows form example uses Astoria’s .NET 3.5 Client Library to display and update entity instances with the Atom Publication (AtomPub or APP) wire format. The Web form project uses the AJAX Client Library and JavaScript Object Notation (JSON) as the wire format.

Nyckelord: LINQ to SQL, LINQ to Objects, LINQ to DataSet, LINQ to XML, LINQ to ADO.NET Data Services, LINQ to REST, LINQ to Entities, Entity Framework, Entity Data Model, Entity Data Platform, ADO.NET Data Services, Project Astoria, IEnumerable, IQueryable, object/relational mapping, O/RM, Plain Old CLR Objects, POCO, persistence ignorance