MySQL, LINQ and the

ADO.NET Entity Framework

Presented by,
MySQL & O’Reilly Media, Inc. Reggie Burnett, MySQL
David Sceppa, Microsoft
Agenda
 From DataSets to data access layers
 Introducing LINQ
 Using LINQ to access your data
 The ADO.NET Entity Framework
 Introducing ADO.NET Data Services
 Summary
Working with DataSets
 Basic example
//Execute a query, retrieve and store the results
connectionString = "Data Source=localhost;User ID=...;";
commandText = "SELECT CompanyName, City FROM Customers " +
"WHERE Country = ?Country";
adapter = new MySqlDataAdapter(commandText, connectionString);
adapter.SelectCommand.Parameters.AddWithValue("?Country", "Brazil");
table = new DataTable();
adapter.Fill(table);

//Display the results

DataGridView resultsGrid = new DataGridView();
resultsGrid.DataSource = table;

//Submit pending changes

commandBuilder = new MySqlCommandBuilder(adapter);
adapter.Update(table);
Working with DataSets
 Benefit
• Rapid development - Designers, wizards,
bound controls make it easy to write code
and generate user interface
 Drawbacks
• Developers must become SQL experts to
build advanced queries
• Code becomes more complex when app
model and database schema diverge
Moving to data access layers
 Focus on your application model
• Queries expressed in terms of app model
• Classes track their own changes
 Data access layer responsible for
• Translating into SQL queries
• Returning results as objects
• Submitting pending changes
Moving to data access layers
 Basic example
//Partial class to separate data access code
public class Customer {

//Add methods to return data

public static List<Customer> GetCustomers(string country) {
List<Customer> results = new List<Customer> ();
MySqlConnection connection = new MySqlConnection(connectString);
connection.Open();
MySqlCommand command = new MySqlCommand(queryString, connection);

command.Parameters.AddWithValue("?Country", country);
MySqlDataReader reader = command.ExecuteReader();
while (reader.Read())
results.Add(Customer.CreateFromReader(reader));
return results;
}

//Support tracking and submitting changes

private string origCustomerId, origCompanyName, ...;
public void SubmitChanges() { ... }
}
Moving to data access layers
 Benefit
• Division of labor - Manage data access code
separately, only subset of developers need to
become SQL gurus
 Drawbacks
• Data access code quickly gets complex
 Classes that span multiple tables
 Handling inheritance hierarchies
 Note: Patterns and frameworks emerge
Introducing LINQ
 What is LINQ?
• Stands for Language INtegrated Query
• Allows developers to query data structures
using SQL-like syntax from within their
application’s code
• Is available to both C# and Visual Basic
developers in Visual Studio 2008
Introducing LINQ
 LINQ Example - Querying an array
//Create an array of integers
int[] myarray = new int[] { 49, 28, 20, 15, 25,
23, 24, 10, 7, 34 };

//Create aa query
//Create query for
for odd
odd numbers,
numbers sorted in descending order
var oddNumbers
var oddNumbers == from
from ii in
in myarray
myarray where i % 2 == 1 select i;
where i % 2 == 1
//Compose the original
select query
orderby i; to create a query for odd numbers
i descending
var sorted = fromselect
i in oddNumbers
i; orderby i descending select i;

//Display the results of the query

foreach (int i in oddNumbers)
Console.WriteLine(i);
LINQ-enabled classes
 Collection classes extended
• Arrays, Hashtable, List<T>, etc.
 LINQ to DataSet
• Extended for untyped and strongly typed
 LINQ to XML
 Data access technologies
• LINQ to SQL
• LINQ to Entities
Agenda
 From DataSets to data access layers
 Introducing LINQ
 Using LINQ to access your data
 The ADO.NET Entity Framework
 Introducing ADO.NET Data Services
 Summary
LINQ to DataSet
 DataSet class enhanced to support LINQ
• Available via typed and untyped DataSet
Dim ordersQuery = From o In dataSet.Orders _
Where o.CustomerID = "ALFKI" _
Select o

For Each o As NorthwindDataSet.OrdersRow In ordersQuery

Console.WriteLine("{0} {1,10:d}", _
o.OrderID, o.OrderDate)
Next o
LINQ to DataSet
 Benefits
• Quickest way to use LINQ w/ .NET 2.0
feature set
• Supports query capabilities previously
unavailable using a DataSet
• Returns results as series of DataRows or as
anonymous types
 Limitations
• All client-side processing
LINQ to SQL
 Benefits
• Rapid application development scenarios
• Query processing performed at the server
• Designer makes mapping classes to tables
simple
 Limitations
• Supports limited mapping scenarios
• No provider model for extensibility
Agenda
 From DataSets to data access layers
 Introducing LINQ
 Using LINQ to access your data
 The ADO.NET Entity Framework
 Introducing ADO.NET Data Services
 Summary
LINQ to Entities
 What is LINQ to Entities?
• LINQ implementation of the Entity Data Model
 What is the Entity Data Model?
• Definition for your application model
• Map between app model, database schema
• Advanced mapping scenarios supported
 One entity mapped across multiple tables
 Multiple inheritance hierarchy mappings
 Many-to-many without "link" table in model
LINQ to DataSet
 DataSet class enhanced to support LINQ
• Available via typed and untyped DataSet
Dim ordersQuery = From o In dataSet.Orders _
Where o.CustomerID = "ALFKI" _
Select o

For Each o As NorthwindDataSet.OrdersRow In ordersQuery

Console.WriteLine("{0} {1,10:d}", _
o.OrderID, o.OrderDate)
Next o
LINQ to Entities or LINQ to SQL?
 LINQ to SQL
• Shipped with Visual Studio 2008 and .NET 3.5
• Emphasis on rapid application development
• Supports Microsoft SQL Server family of databases
 LINQ to Entities
• Will ship as an update to .NET 3.5
• Offers a provider model for third-party databases
• Designed for enterprise-grade data scenarios
• Higher level of abstraction for programming databases
• Just one layer of the overall ADO.NET Entity Framework
LINQ to Entities: Generating Queries
 Converting LINQ queries to SQL
• Compiler converts code into a LINQ expression tree
• LINQ to Entities converts LINQ expression tree into
a DbCommandTree based on mapping information
 DbCommandTree expressed in terms of the
database schema
• ADO.NET provider generates a DbCommand
• LINQ to Entities executes the DbCommand,
assembles results into the structure(s) specified in
the LINQ query
Agenda
 From DataSets to data access layers
 Introducing LINQ
 Using LINQ to access your data
 The ADO.NET Entity Framework
 Introducing ADO.NET Data Services
 Summary
Entity Framework: LINQ and more
 ADO.NET Entity Framework Layers
• First implementation of
Entity Data Model (EDM) LINQ to Entities
• Includes a text-based query
query language (Entity SQL) ObjectServices
• New Provider – EntityClient
• EntityClient
New ORM stack (Object Services)
leverages Entity Client
• ADO.NET Provider
LINQ support over Entities
• An evolution of ADO.NET
preserving customer investment
Entity Framework Query Options
 LINQ to Entities
var ordersQuery = from o in context.Orders
where o.Customers.CustomerID == "ALFKI"
select o;

foreach (var o in ordersQuery)

Console.WriteLine("{0} {1:d}", o.OrderID, o.OrderDate);
 Sweet spot:
• Strongly typed queries and results
 Compile-time type checking and Intellisense
• Results returned as objects: entities / projections
• Queries must be well defined at compile time
Entity Framework Query Options
 ObjectQuery<T> / ObjectQuery(Of T)
Dim eql As String = "SELECT VALUE o FROM Orders AS o " & _
" WHERE o.Customer.CustomerID = 'ALFKI'"
Dim ordersQuery = context.CreateQuery(Of Order)(esql)

For Each o As Order In ordersQuery

Console.WriteLine("{0} {1,10:d} ", o.OrderID, o.OrderDate)
Next o

 Sweet spot:
• Strongly typed results, text-based queries
• Results returned as objects: entities / projections
• Queries more loosely defined at compile-time
Entity Framework Query Options
 EntityClient Provider
string eSql = "SELECT VALUE o FROM NorthwindEntities.Orders AS o " +
"WHERE o.Customers.CustomerID = 'ALFKI'";
EntityCommand cmd = new EntityCommand(eSql, connectionString);
EntityDataReader rdr = cmd.ExecuteReader();
while (rdr.Read())
Console.WriteLine("{0} {1:d}", rdr["OrderID"], rdr["OrderDate"]);

 Sweet spot:
• Untyped queries and results
• Results returned as DataReaders
• Queries can be generated purely at run time
 No application model classes needed at compile time
Agenda
 From DataSets to data access layers
 Introducing LINQ
 Using LINQ to access your data
 The ADO.NET Entity Framework
 Introducing ADO.NET Data Services
 Summary
ADO.NET Data Services
 Formerly known as "Project Astoria"
• Part of ASP.NET Extensions Preview
• Integrated into .NET Framework going forward
 Targeting Web development technologies
• Silverlight and AJAX
• Data returned via Web-friendly formats
 ATOM (XML-based) and JSON
ADO.NET Data Services
 Service exposed via lightweight data access API
• Supply both location and query as a URI:
 http://myserver/data.svc/Customers[ALFKI]/Orders
• Query converted to LINQ inside of service
• Supports submitting changes
 Specialized Entity Framework logic for submitting changes
 Online data hosting service available
 For more information:
• http://astoria.mslivelabs.com
Agenda
 From DataSets to data access layers
 Introducing LINQ
 Using LINQ to access your data
 The ADO.NET Entity Framework
 Introducing ADO.NET Data Services
 Summary
Summary - LINQ
 Represents a revolution for developers
• Query your objects using SQL-like syntax
• LINQ to DataSet
 Rich query scenarios for data residing in a DataSet
• LINQ to Entities
 Provider model for working with other data stores
 Supports enterprise mapping scenarios
Summary - ADO.NET Entity Framework
 More than just LINQ to Entities
• Also supports text-based query language – ESQL
• ObjectQuery<T> for strongly typed results
• EntityClient for reader-based results
 Enterprise-grade data scenarios
 Provider model for working with other data stores
Summary - ADO.NET Data Services
 Designed for Silverlight, AJAX environments
 Lightweight API, connect and query via URI
 Retrieve results via familiar formats
• ATOM
• JSON
Questions?