Oct 08 2009

NerdDinner with Fluent NHibernate Part 3 – The infrastructure

Category: Uncategorizedbengtbe @ 05:00

This is the final post in a series of three where I’m going to see how we can change the NerdDinner project to use Fluent NHibernate instead of LINQ to SQL:

Introduction

In the first post we took a look at the domain model of NerdDinner.com, and recreated and improved the code that was auto generated by the LINQ to SQL designer. In the second post we saw how to map this domain model to the database using Fluent NHibernate. In this final post we are going to take a look at the required changes to the rest of the application.

Session per request

In order for the repository to use NHibernate it needs to get a hold of an ISession. In this application I have chosen to follow the session per request pattern, in which a session is created at the beginning of the HTTP request, and closed at the end of the HTTP request. I also wanted to keep the changes to the NerdDinner project as small as possible; hence the entire infrastructure for NHibernate is isolated in a single class:

public class NHibernateSessionPerRequest : IHttpModule

{

    private static readonly ISessionFactory _sessionFactory;

    static NHibernateSessionPerRequest()

    {

        _sessionFactory = CreateSessionFactory();

    }

    public void Init(HttpApplication context)

    {

        context.BeginRequest += BeginRequest;

        context.EndRequest += EndRequest;

    }

    public static ISession GetCurrentSession()

    {

        return _sessionFactory.GetCurrentSession();

    }

    public void Dispose() { }

    private static void BeginRequest(object sender, EventArgs e)

    {

        ISession session = _sessionFactory.OpenSession();

        session.BeginTransaction();

        CurrentSessionContext.Bind(session);

    }

    private static void EndRequest(object sender, EventArgs e)

    {

        ISession session = CurrentSessionContext.Unbind(_sessionFactory);

        if (session == null) return;

        try

        {

            session.Transaction.Commit();

        }

        catch (Exception)

        {

            session.Transaction.Rollback();

        }

        finally

        {

            session.Close();

            session.Dispose();

        }

    }

    private static ISessionFactory CreateSessionFactory()

    {

        string connString = "NerdDinnerConnectionString";

        FluentConfiguration configuration = Fluently.Configure()

        .Database(MsSqlConfiguration.MsSql2005.ConnectionString(

           x => x.FromConnectionStringWithKey(connString)))

        .ExposeConfiguration(

            c => c.SetProperty("current_session_context_class", "web"))

        .Mappings(m => m.FluentMappings.AddFromAssemblyOf<Dinner>());

        return configuration.BuildSessionFactory();

    }

}


Some comments about the infrastructure:

  • It implements the IHttpModule, in order to plug into the request handling pipeline.
  • Since it is expensive to create an ISessionFactory you only want to create this once. This is why the factory is stored in a static field that is initialized by the static constructor.
  • At the beginning of the HTTP request the ISession is created and “stored” using the CurrentSessionContext.Bind method.
  • Other code can get a hold of this session by calling the GetCurrentSession method.
  • At the end of the HTTP request we “remove” the session by using the CurrentSessionContext.Unbind method. We also commit any changes, and close the session.

There are some drawbacks to this approach. We create a session for every request, even if it is not needed. However, a session is very lightweight, and we can easily tweak this behavior later. Another drawback is that we commit the transaction at the end of the HTTP request. In a more complex business application you probably want more control over your unit of work.

In order to activate this IHttpModule we need to add the following to httpModules/modules section of the Web.config:

<add name=”NHibernateSessionPerRequesttype=”NerdDinner.Models.NHibernateSessionPerRequest” />

Changes to the repository

In the DinnerRepository we are mainly going to use LINQ to NHibernate to replace the LINQ to SQL queries:

public class DinnerRepository : IDinnerRepository

{

    public ISession Session

    {

        get { return NHibernateSessionPerRequest.GetCurrentSession(); }

    }

    public IQueryable<Dinner> FindAllDinners()

    {

        return Session.Linq<Dinner>();

    }

    public IQueryable<Dinner> FindUpcomingDinners()

    {

        return from dinner in FindAllDinners()

               where dinner.EventDate >= DateTime.Now

               orderby dinner.EventDate

               select dinner;

    }

    public IQueryable<Dinner> FindByLocation(float latitude, float longitude)

    {

        return Session.CreateSQLQuery(

                @”SELECT d.*

                    FROM Dinners d

                    JOIN NearestDinners(:Latitude, :Longitude) n

                      ON d.DinnerId = n.DinnerId

                   WHERE EventDate >= :EventDate

                   ORDER BY EventDate Desc”)

            .AddEntity(typeof(Dinner))

            .SetDouble(“Latitude”, latitude)

            .SetDouble(“Longitude”, longitude)

            .SetDateTime(“EventDate”, DateTime.Now)

            .List<Dinner>().AsQueryable();

    }

    public Dinner GetDinner(int id)

    {

        return Session.Linq<Dinner>()

            .SingleOrDefault(d => d.DinnerID == id);

    }

    public void Add(Dinner dinner)

    {

        Session.SaveOrUpdate(dinner);

    }

    public void Delete(Dinner dinner)

    {

        Session.Delete(dinner);

    }

}

The LINQ to SQL and NHibernate Linq syntax are similar, so the changes to this class were fairly simple. However in the FindByLocation method I had to create a custom SQL query, because I don’t think (not sure) that LINQ to NHibernate supports calling the database function NearestDinners.

Update: It seems that NHibernate Linq supports database functions. However I couldn’t find much information about the functionality except for one simple example in the source code.

The ISession is retrieved by calling the GetCurrentSession() method. In an application with an IoC container I would prefer to inject the session in the constructor.

finally {}

In the final post in this series we have taken a look at the necessary changes to the infrastructure. We used a session per request pattern to create the session, and LINQ to NHibernate in the repository. As you hopefully have seen it was not very difficult to replace LINQ to SQL with Fluent NHibernate. As a benefit the domain part of the application has been greatly improved over the auto generated code from the LINQ to SQL designer.

If you liked this post then please shout and kick me :)

kick it on DotNetKicks.com Shout it

Tags: , , ,


Aug 25 2009

NerdDinner with Fluent NHibernate Part 2 – The mapping

Category: Uncategorizedbengtbe @ 06:00

This is the second post in a series of three where I’m going to see how we can change the NerdDinner project to use Fluent NHibernate instead of LINQ to SQL:

Introduction

In the first post we took a look at the domain model of NerdDinner.com, and recreated and improved the code that was auto generated by the LINQ to SQL designer. In this post we are going to look at how to map this domain model to the database using Fluent NHibernate.

Fluent NHibernate

Fluent NHibernate gives you an alternative to NHibernate’s XML mapping files. It lets you write mapping in strongly typed (and ReSharper friendly) C# code. In order to map an entity, you create a mapping class that derives from ClassMap<T>. All the mapping code is done inside the constructor by calling methods from the base class.

Last week the Release Candidate for Fluent NHibernate 1.0 was released. The code in this post has been updated with the changes in this release.

The mapping for the Dinner entity

The code for DinnerClassMap is shown below:

public class DinnerClassMap : ClassMap<Dinner>

{

    public DinnerClassMap()

    {

        Table(“Dinners”);

        Id(d => d.DinnerID).GeneratedBy.Identity();

        Map(d => d.Title).Not.Nullable().Length(50);

        Map(d => d.EventDate).Not.Nullable();

        Map(d => d.Description).Not.Nullable().Length(256);

        Map(d => d.HostedBy).Not.Nullable().Length(20);

        Map(d => d.ContactPhone).Not.Nullable().Length(20);

        Map(d => d.Address).Not.Nullable().Length(50);

        Map(d => d.Country).Not.Nullable().Length(30);

        Map(d => d.Latitude).Not.Nullable();

        Map(d => d.Longitude).Not.Nullable();

        HasMany(d => d.RSVPs).KeyColumn(“DinnerId”).Cascade.All();

    }

}

Explaination of code:

  • The Table method maps “Dinners” as the name of the table. By default Fluent NHibernate assumes it to be “Dinner”.
  • The Id method maps DinnerId as the primary key, and specifies that an identity column in SQL Server is used to generate ids.
  • The Map method is used to map simple properties to columns.
  • The HasMany method maps the one-to-many relationship to the RSVP table. Cascade.All() is used to make sure that RSVPs are saved/updated/deleded when the dinner is saved/updated/deleded.

The RSVP mapping

The code for the RSVPClassMap is shown below:

public class RSVPClassMap : ClassMap<RSVP>

{

    public RSVPClassMap()

    {

        Id(r => r.RsvpID).GeneratedBy.Identity(); ;

        Map(r => r.AttendeeName).Not.Nullable().Length(30);

        References(r => r.Dinner, “DinnerId”).Not.Nullable();

    }

}

Explaination of code:

  • No need to call the Table method, since Fluent NHibernate correctly assumes that the table name is “RSVP”.
  • The References method maps the “other side” of the HasMany relationship in the DinnerClassMap.

Is all this mapping code needed?

The mapping code shown above contains a lot of information about the database schema. In this case, where we just wanted to map the domain to an existing database we could remove the following:

  • Not.Nullable()
  • Length(int length)

In Greenfield projects I prefer to start with the domain, and then write the mapping, and finally automatically generate the database schema. You then need to include this information in the mappings. In the post Mapping a Twitter like domain with Fluent NHibernate I showed how you generate the database from the mapping.

If you have an existing database you can also use this to troubleshoot your mapping code. Generate a new database from your schema and check it against the existing database.

Give me a new identity!

NerdDinner uses the identity column of SQL Server as primary key. In most cases this works just fine, however it can cause problems with the Unit of Work functionality of NHibernate. If I designed this application from scratch I would rather choose the Guid.Comb or HiLo algorithm to generate primary keys.

If you want to read more about this, check out the post NHibernate POID Generators revealed.

Do you prefer to live in XML hell?

If you miss the good old days of XML (???), then Fluent NHibernate can help you with this. The code below shows how to export your Fluent Mappings to XML-files:

[TestMethod]

public void ExportMappings()

{

    FluentConfiguration conf =

       NHibernateConfiguration.CreateConfiguration(false);

    conf.Mappings(m =>

    {

        m.FluentMappings.ExportTo(@”PATH_TO_EXPORT_FOLDER”);

    });

    conf.BuildSessionFactory();

}

This technique is actually quite useful for several reasons:

  • First, if you are skeptical about using Fluent NHibernate because it has not reached 1.0, or because it might not support all mapping options, you can easily switch back to XML-files later.
  • Secondly, it can also be useful to troubleshoot mapping problems, because there is currently more information on the web regarding XML-mapping.

finally{}

In this post we have taken a look at how to use Fluent NHibernate to specify the mapping between the domain model and the database. In the next post we will take a look at the rest of the changes that was necessary in order to use the new domain and Fluent NHibernate.

If you liked this post then please shout and kick me :)

kick it on DotNetKicks.com Shout it

Tags: , , ,


Aug 10 2009

NerdDinner with Fluent NHibernate Part 1 – The domain model

Category: Uncategorizedbengtbe @ 06:00

This is the first post in a series of three where I’m going to see how we can change the NerdDinner project to use Fluent NHibernate instead of LINQ to SQL:

Introduction

NerdDinner.com is a site where you can organize dinner meetings for nerds, but even more, it is a great tutorial to the ASP.NET MVC framework. If you want to learn this framework (you should!) download the free tutorial (186 pages) from the site and get going!

NerdDinner uses LINQ to SQL as a data access layer generator. LINQ to SQL is perfectly suitable for this project, since the domain is small, and not very complex. However in these posts I would like to explore how the project would look using a real Object/Relational Mapper (O/RM). More specifically, I’m going to use Fluent NHibernate.

Preperations

First I downloaded the latest source code from CodePlex using the Subversion client TortoiseSVN. Then I added references to the NHibernate, Fluent NHibernate, and LINQ to NHibernate assemblies.

The existing domain model

Most of the domain model in NerdDinner is auto generated from the database as partial classes by the LINQ to SQL designer. You can find the Dinner and RSVP class in the NerdDinner.Designer.cs file; however since they are auto generated they are not very readable.

The Dinner class is also extended (using partial classes) with additional methods in the Dinner.cs file.

The new domain model

Since most of the existing domain model is auto generated we need to write a new domain model. The new Dinner class looks as follows:

public class Dinner

{

    public Dinner()

    {

        RSVPs = new List<RSVP>();

    }

    public virtual int DinnerID { get; private set; }

    public virtual string Title { get; set; }

    public virtual string Description { get; set; }

    public virtual DateTime EventDate { get; set; }

    public virtual double Latitude { get; set; }

    public virtual double Longitude { get; set; }

    public virtual string Country { get; set; }

    public virtual string Address { get; set; }

    public virtual string HostedBy { get; set; }

    public virtual string ContactPhone { get; set; }

    public virtual IList<RSVP> RSVPs { get; private set;}

    public virtual void AddRSVP(string attendeeName)

    {

        if (dinner.IsUserRegistered(User.Identity.Name)) return;

        RSVPs.Add(new RSVP

                      {

                          AttendeeName = attendeeName,

                          Dinner = this

                      });

    }

    public virtual bool IsHostedBy(string userName) { … }

    public virtual bool IsUserRegistered(string userName) { … }
    public virtual bool IsValid { … }

    public virtual IEnumerable<RuleViolation> GetRuleViolations() { … }

}

The new RSVP class looks as follows:

public class RSVP

{

    public virtual int RsvpID { get; private set; }

    public virtual Dinner Dinner { get; set; }

    public virtual string AttendeeName { get; set; }

}

If you wonder why every method is marked by virtual, this is a requirement of NHibernate in order to support Lazy-Loading. I explained this in more detail in the post Mapping a Twitter like domain with Fluent NHibernate.

The public interface of the domain model is more or less identical to the existing domain model. However I did make some changes.

Protect thy privates!

One of the changes I made was to make some of the property setters private; DinnerID, RSVPs, and RsvpID. Since the consumer code (the controllers) is not supposed to change to these properties we want to communicate this to our fellow programmers. The next change is also about protecting privates.

New method to add a RSVP to a dinner

To the Dinner class I added a new method called AddRSVP. By using terms from Domain Driven Design, the Dinner entity is the aggregate root of the RSVP entity. This basically means that a RSVP must belong to a Dinner; hence I want the Dinner class to control the list of RSVPs.

This change will help reduce code in the controllers and remove duplication. Previously the creating and adding of RSVPs was done both in the RSVPController.Register and the DinnerController.Create action methods. The Register method is shown below:

[Authorize, AcceptVerbs(HttpVerbs.Post)]

public ActionResult Register(int id) {

    Dinner dinner = dinnerRepository.GetDinner(id);

    if (!dinner.IsUserRegistered(User.Identity.Name)) {

        RSVP rsvp = new RSVP();

        rsvp.AttendeeName = User.Identity.Name;

        dinner.RSVPs.Add(rsvp);

        dinnerRepository.Save();

    }

    return Content(“Thanks – we’ll see you there!”);

}

Thanks to our new method, this can now be reduced to:

[Authorize, AcceptVerbs(HttpVerbs.Post)]

public ActionResult Register(int id) {

    Dinner dinner = dinnerRepository.GetDinner(id);

    dinner.AddRSVP(User.Identity.Name);

    return Content(“Thanks – we’ll see you there!”);

}

Similar reduction can be done to the Create method. BTW if you wonder what happened to the call to the Save method then we will come back to this in part 3 (teaser!).

As a rule of thumb always try to keep your action methods as simple as possible!

More object-oriented

The last change is the removal of the DinnerId property from the RSVP class. This property is purly database related, and does not belong in an object-oriented domain model. You can use the Dinner propety to get the DinnerId.

finally{}

In my opinion one of the benefits of using an OR/M like NHibernate as supposed to LINQ to SQL is that the domain model is much clearer. When trying to understand a new project one of the best places to start is by looking at the domain. I also did some changes (improvements?) to the domain model to better communicate intent, and to reduce code duplication.

The next step is to tell NHibernate how to map this domain model to the database. This will be explained in the next post where we look at the mapping.

If you liked this post then please shout and kick me :)

kick it on DotNetKicks.com Shout it

Tags: , , ,


May 24 2009

Mapping a Twitter like domain with Fluent NHibernate

Category: Uncategorizedbengtbe @ 10:00

I’m currently learning NHibernate, and one of the best ways to learn is to blog about it :) Since I’m not a big fan of XML files, I also wanted to use Fluent NHibernate to do the mapping. As an example I will use a social messaging domain, similar to Twitter. I will also use a Top-down approach, starting with the domain model, writing the mapping, and finally creating the database schema using the SchemaExport class in NHibernate.

The domain model

Let’s start with a look at the domain model:

Quacker Domain

The domain model mainly consists of the User entity and the Message entity and their associations:

  • The user has a list of Messages that he/she has posted.
  • The user has a list of other users that are interested in his/hers messages, called Followers.
  • The user has a list of other users whose messages he/she is interested, called Following.
  • The message has a PostedBy reference to the user who posted it.

Both entities derive from the Entity<T> base class that contains the Id property and implements Equals and GetHashCode methods.

The domain classes

Let’s take a close look at the domain classes. Below is the class for the Message entity:

public class Message : Entity<Message>

{

    public virtual string Text { get; set; }

    public virtual User PostedBy { get; set; }

    public virtual DateTime CreatedAt { get; set; }

}

It contains the message text, a reference to the user who posted it, and the time it was created.

Below is the class for the User entity:

public class User : Entity<User>

{

    public User()

    {

        Name = Url = Email = “”;

        Followers = new List<User>();

        Following = new List<User>();

        Messages = new List<Message>();

    }

    public virtual string Name { get; set; }

    public virtual string Username { get; set; }

    public virtual string Email { get; set; }

    public virtual string Url { get; set; }

    public virtual IList<User> Followers { get; private set; }

    public virtual IList<User> Following { get; private set; }

    public virtual IList<Message> Messages { get; private set; }

    public virtual void AddFollowing(User user)

    {

        if (Following.Contains(user)) return;

        Following.Add(user);

        user.Followers.Add(this);

    }

    public virtual void AddMessage(string messageText)

    {

        var message = new Message

                          {

                              Text = messageText,

                              PostedBy = this,

                              CreatedAt = DateTime.Now

                          };

        Messages.Add(message);

    }

}

 

The User class contains some properties about the user, a list of followers, a list of following users, and a list of messages. It also contains two methods:

  • AddFollowing(User user) - Adds another user to the list of users that the current user is following. Since this is a bidirectional relationship, it also adds the current user to the other user’s followers.
  • AddMessage(string messageText) – Adds a new message that the user has posted. Sets the PostedBy to the current user.

Why is everything virtual?

When using NHibernate the domain model can be close to Persistence Ignorant (PI), which means that you don’t have to take any special considerations when designing the domain model, like derive from a specific base class. However to support transparent lazy-loading, NHibernate needs to return a proxy that inherit from your entity and override the properties.

If you dislike marking everything with virtual you can turn of lazy-loading, but I don’t recommend doing this.

Mapping with Fluent NHibernate

As mentioned, Fluent NHibernate lets you write mapping with strongly typed C# code, instead of the traditional NHibernate XML mapping files. This allows you to use tools like ReSharper to alter both the domain and mapping files when refactoring.

The mapping files must derive from ClassMap<T> and the mapping code is done inside the constructor using lambda expressions and a fluent syntax.

Let’s take a look at the mapping file for the Message entity:

public class MessageClassMap : ClassMap<Message>

{

    public MessageClassMap()

    {

        Id(m => m.Id).GeneratedBy.GuidComb();

        Map(m => m.Text).Not.Nullable().WithLengthOf(140);

        Map(m => m.CreatedAt).Not.Nullable();

        References(m => m.PostedBy, “PostedBy”).Not.Nullable();

    }

}

 

Explanation of the mapping:

  • The Id method maps the Id property as the identifier of the Message entity. It should be generated using the GuidComb algorithm.
  • The Map method maps the Text property. It also specifies that is should not be null, and that the max length is 120.
  • The Map method also maps the CreatedAt property.
  • The References method maps the PostedBy property as a many-to-one relationship between the message and user. Hence, one message belongs to a single user, but a user can have many messages. It also specifies that the column name in the database should be PostedBy.

Let’s take a look at the mapping file for the User entity:

public class UserClassMap : ClassMap<User>

{

    public UserClassMap()

    {

        Id(u => u.Id);

        Map(u => u.Name).Not.Nullable().WithLengthOf(50);

        Map(u => u.Username).Not.Nullable().WithLengthOf(50).Unique();

        Map(u => u.Email).Not.Nullable().WithLengthOf(100);

        Map(u => u.Url).Not.Nullable().WithLengthOf(100);

        HasMany(u => u.Messages).LazyLoad().Cascade.All()

            .KeyColumnNames.Add(“PostedBy”);

        HasManyToMany(u => u.Following).LazyLoad()

            .WithParentKeyColumn(“FollowerId”)

            .WithChildKeyColumn(“FollowingId”);

        HasManyToMany(u => u.Followers).LazyLoad()

            .WithParentKeyColumn(“FollowingId”)

            .WithChildKeyColumn(“FollowerId”).Inverse();

    }

}

 

Explanation of the mapping:

  • The mapping of the Id, Name, Email, Url does not contain any new syntax.
  • The mapping of the UserName property uses the Unique method to specify uniqueness. This will create a unique constrain in the database schema.
  • The HasMany method maps the Messages property as a one-to-many between the user entity and message entity. The Cascade.All means the message will become persistent when you add it to a persistent user. The messages will also be deleted when you delete the user. It also specifies that the name of the column should be PostedBy in order to make sure that it uses the same column as specified in the message mapping.
  • The HasManyToMany method maps the Following as a many-to-many mapping between users with specified column names.
  • The HasManyToMany method maps the Followers as a many-to-many mapping between users. This is the inverse of the Following association, hence the call to Inverse().

We also call LazyLoad() to specify that we want support for lazy loading of the lists.

Is something missing in the mapping?

You might have noticed that we often don’t specify the table or column name in the mapping. In these cases the table will get the same name as the class, and the column will get the same name as the property. Below is an example of how to specify table and column names:

WithTable(“TBL_MESSAGES”);

Map(c => c.CreatedAt, “COL_CREATEDAT”).Not.Nullable();

You also don’t need to specify the type of the properties (e.g. that PostedBy is of type User). NHibernate uses reflection to determine the type.

Creating the database

In a top-down approach you change the database after you make changes to the domain or the mapping. During development you can recreate the database using the SchemaExport class in a test or console application. Below is an example using an explicit test in NUnit:

    [TestFixture]

    public class DatabaseSetup

    {

        [Test, Explicit]

        public void SetupDatabase()

        {

            FluentConfiguration conf = NHConfiguration.CreateConfiguration(false);

            conf.ExposeConfiguration(BuildSchema)

                .BuildSessionFactory();

        }

        private static void BuildSchema(Configuration conf)

        {

            new SchemaExport(conf).Drop(false, true);

            new SchemaExport(conf).Create(false, true);

        }

    }

}

The test is marked with the Explicit attribute because you want to control when this test is run, e.g. you do not want to run it every time you run your tests. The SetupDatabase() test receives a FluentConfiguration from the NHibernate configuration class in the project. The database is created in the BuildSchema method that is passed to the ExposeConfiguration method. The call to BuildSessionFactory() is needed to trigger the FluentConfiguration to actually execute the call to the BuildSchema method.

Finally the database schema

Below is the database schema that has been created from the domain model and the mappings:

Quacker Database Schema

Some notes about the database schema:

  • The length of the nvarchar fields are according to the WithLengthOf call in the mappings.
  • All fields are not nullable because of the Not.Nullable() call in the mappings.
  • The many-to-many association between users are represented using the UserToUser junction table.
  • The one-to-many association between the message and users are done by the PostedBy column in the Message table.
  • It is not shown in the diagram, but Username has a unique constrain due to the Unique() call in the mappings.

Conclusion

In this post I have shown how to use Fluent NHibernate in a top-down approach in a Twitter like domain. Since I’m fairly new to NHibernate I welcome any comments and suggestions!

kick it on DotNetKicks.com Shout it

Tags: ,