Adding a note to highlight Damien’s comment below. Damien is on the EF team and does not recommend using the current Feature CTP with the RC. A new version will be available "soon".

In comments of another blog post Ivan reminds me of something else I ran into the the VS2010 RC.

I tweeted on Tuesday that the Self-Tracking Entities template is now “in the box”.

I also blogged on Tuesday about a change to a file, EF.Utility.CS.ttinclude (or EF.Utility.VB.ttinclude) which the EF T4 Templates depends. The RC version is different from the Beta 2 version.. Using the Beta 2 EF POCO Template with VS2010 RC

The EF Feature CTP contains Code Only functionality and an earlier version of the STE template. Like the POCO template, that STE template depends on the Beta 2  version of the ttinclude file. But the STE template in the VS2010 RC depends on the newer version of the ttinclude file.

If you install the EF Feature CTP on top of the VS2010 RC, you will overwrite the Self Tracking Entities template with the old version and it will no longer work.

I tried to outsmart the process by saving the Item Template zip files, installing the CTP and then copy the RC zip files over the ones which  the CTP installed. That didn’t work out very well.

Option 1

However, doing a repair on VS2010 RC after installing the Feature CTP (and in my case totally mucking things up trying to be clever with the zip files) let me have my cake (Code Only support that works with the VS2010 RC -yay) with the icing (the RC version of teh Self-Tracking Entities which got reinstalled during the repair).

Option 2

Another alternative is to run the STE template item before installing the CTP. The STE template item creates two template files, one for the context and the other for the entities. Only the entities template is impacted by the Feature CTP installation. Copy the contents of the entities template file somewhere. Then after installing the CTP, you’ll initially create the Beta 2 version of the entities template file, just copy the code from the new file into the template file that’s created.

Option 3
If you don’t need to use Code Only, don’t install the Feature CTP! 🙂 Originally, the Feature CTP was a source for Code Only support, STEs and the POCO template. The POCO template is now an extension (although see the above link for some current incompatibility issues with the RC) and STEs are now in the box.

Eventually these conflicts will all get sorted out.

I had a problem that stopped me in my tracks for two days.

A RIA Services/Silverlight solution which I had created from the Silverlight Business App solution template, was crashing.

Specifically…

In VS2010 Beta 2, using the web development server, I was able to run the app by right clicking on the host web page and choosing View in Browser.

But any time I tried to run it by debugging (F5), I got the old somethingorother not loaded error. That’s the error the domain context on the client side throws when it tries to load data from the service but no data comes back.

After lots of digging, I concluded that in debug mode the client app was unable to find the service.

I won’t go through the two days of hairpulling, but in the end it turned out that the cause of my problem was a known issue with VS2010’s Intellitrace.

It may have been known, but it was not documented anywhere discoverable.

Wilco Bauwer (another great community person who I seem to have forgotten was sucked up by Microsoft at some point – I’m doing this a lot lately…can’t keep track ) was dragged into the thread by Jeff Handley (another silverlight guru at MS,  who had been valiant in trying to help me). Wilco suggested turning off Intellitrace and voila, two days of misery disappeared.

Update: After realizing that we still want access to it, the team has made the Beta 2 file available again. You can get it from their original blog post, not from the Visual Studio Gallery links but look for a zip file at the bottom of the post.

But there’s more! While I had managed to figure out one big difference (TemplateFileManager name changed) between the old include file and the new include file it wasn’t enough. which is why after a few hours I gave up and went back to the old include file. But Nikola Malovic figured out the second piece which was how to replace the WriteFiles method which is no longer there! Check out his blog post for the super secret trick!

————————————————–

Entity Framework’s T4 templates all have a dependency on a file that includes additional specialized methods.

If you look at the top of the templates (ENtityObject, POCO or Self-Tracking Entiteis) there is a pointer to a ttinclude file.

<#@ include file="EF.Utility.CS.ttinclude"#>

That file, however, changed after Beta 2. The new templates (EntityObject and STE’s both included in VS2010) work wtih the new version of this file. But the POCO template that was released in December doesn’t.

However, I have a trick to have your cake and eat it too until the new POCO template comes along.

This is just a temporary workaround until a new set of templates hits the airwaves. 🙂

The ttinclude files live here:

C:\Program Files (x86)\Microsoft Visual Studio 10.0\Common7\IDE\Extensions\Microsoft\Entity Framework Tools\Templates\Includes

The new one has a file date of Jan 6, 1010.

You can grab the ttinclude file from Beta 2 and rename it. I’ve renamed mind:

EF.B2.Utility.CS.ttinclude

The file date on that one is Sept 15, 2009.

Then copy it into the Includes folder with the new includes.

Now when you create a new POCO template, you’ll first get error message about TemplateFileManager not existing. In the new include, that’s been renamed to EntityFrameworkTemplateFIleManager. But that’s not hte only change. Fixing that name in your template will not help.

Instead,  just change the reference in the template file to read the Beta 2 include :

<#@ include file="EF.B2.Utility.CS.ttinclude"#>

Then the POCO template will use the Beta 2 version of the include and you can go on your merry way.

Thanks to Dane Morgridge for sending me the VSIX file from the Beta 2 POCO since I had installed directly rather than downloading it to my computer.

The video from my webcast, Top 10 Treats in Entity Framework 4 is available on Oreilly’s website: http://www.oreillynet.com/pub/e/1517.

I just discovered that WebEx saved the chat window from my O’Reilly webcast last month. So I thought I would pick out the quesitons and answer them here although I did answer some of them at the end of the webcast:

Q address NOT Address??

Someone noticed that an entity was named address. The pluralization support in the VS2010 EDM Designer does not fix up bad capitalization from the database which this was meant to highlight. Another viewer replied: it’s address and not Address because the original table was with a lower case ‘a’. VS2010 supports pluralization, but doesn’t change capitalization

Q These classes are not POCO, correct? What do I have to do to have POCO and generate db schema?

In the early part of the demo I was wroking with the default EntityObjects. To use POCO classes to generate db schema you’ll want to skip the model and use the Code-Only support that is part of the EF Feature CTP: Updated Feature CTP Walkthrough- Code Only for Entity Framework

This is awesome. We will all forgot how to sql.
Not a question, but a funny comment I wanted to share.

Q What about update database from model?

Update database from model has been supported sine the first version of EF. It will bring in new tables and new properties in tables but it will not modify property types. Be aware that any changes made manually to the SSDL get overwritten during this process.

Q. How much control does the developer have over the generated SQL DDL?
I answered this one at the end of the webcast. There is a T4 template that drives the DDL generation so that’s customizable.

This thing needs a pause button! LOL
Another funny comment I thought I would share. Yeah it is tough doing these webcasts that have a time limit! 🙂

Q. can you control lazy loading on a per entity level?
I happened to answer this question as I was talking about lazy loading and the person who asked then said: “ah thanks … ok, so per context only.”

Q. How easy is it to mock entities for nUnit or xUnit testing?

The new IObjectSet interface makes it totally possible to mock entities now. Many people have written blog posts about this. Here’s mine: http://thedatafarm.wpengine.com/blog/data-access/agile-entity-framework-4-repository-part-5-iobjectset/

Q: I heard something about "model functions" or "schema functions", implying I can do calculations for fields OR create views over entities in the EDM / .edmx file. Is this true?
Yes. Check out http://thedatafarm.wpengine.com/blog/data-access/ef4-model-defined-functions-level-1-amp-2/

Q: how does Oracle support compare? Is it as functionally complete? Is it as stable/bug free?

Oracle providers are available from DevArt and Data Direct. As far as I know they are both very solid. Functionally complete? Dunno.

LINK to EF POCO Adapter for EF1: code.msdn.microsoft.com/EFPocoAdapter

I hit my head against a wall for a while today when I was trying to create a data service using my EF POCOs. I have my model & context in one assembly and my generated POCO entities in another.

When I ran the service (I had exposed all EntitySets for AllRead), I got the mysterious error:

The server encountered an error processing the request. See server logs for more details.

Even with config.UseVerboseErrors = true; in my service, the message did not change.

I finally saw the real problem first in Intellitrace (I’m using VS2010 and .NET 4.0) and then by setting the service behavior’s IncludeExceptionDetailInFaults to true.

This is what the detailed exception looks like:

The server encountered an error processing the request. The exception message is ‘Value cannot be null. Parameter name: key’. See server logs for more details. The exception stack trace is:

at System.ThrowHelper.ThrowArgumentNullException(ExceptionArgument argument) at System.Collections.Generic.Dictionary`2.FindEntry(TKey key) at System.Collections.Generic.Dictionary`2.get_Item(TKey key) at System.Data.Services.Providers.ObjectContextServiceProvider.PopulateMetadata(IDictionary`2 knownTypes, IDictionary`2 childTypes, IDictionary`2 entitySets) at System.Data.Services.Providers.BaseServiceProvider.PopulateMetadata() at System.Data.Services.DataService`1.CreateProvider() at System.Data.Services.DataService`1.EnsureProviderAndConfigForRequest() at System.Data.Services.DataService`1.HandleRequest() at System.Data.Services.DataService`1.ProcessRequestForMessage(Stream messageBody) at SyncInvokeProcessRequestForMessage(Object , Object[] , Object[] ) at System.ServiceModel.Dispatcher.SyncMethodInvoker.Invoke(Object instance, Object[] inputs, Object[]& outputs) at System.ServiceModel.Dispatcher.DispatchOperationRuntime.InvokeBegin(MessageRpc& rpc) at System.ServiceModel.Dispatcher.ImmutableDispatchRuntime.ProcessMessage5(MessageRpc& rpc) at System.ServiceModel.Dispatcher.ImmutableDispatchRuntime.ProcessMessage41(MessageRpc& rpc) at System.ServiceModel.Dispatcher.ImmutableDispatchRuntime.ProcessMessage4(MessageRpc& rpc) at System.ServiceModel.Dispatcher.ImmutableDispatchRuntime.ProcessMessage31(MessageRpc& rpc) at System.ServiceModel.Dispatcher.ImmutableDispatchRuntime.ProcessMessage3(MessageRpc& rpc) at System.ServiceModel.Dispatcher.ImmutableDispatchRuntime.ProcessMessage2(MessageRpc& rpc) at System.ServiceModel.Dispatcher.ImmutableDispatchRuntime.ProcessMessage11(MessageRpc& rpc) at System.ServiceModel.Dispatcher.ImmutableDispatchRuntime.ProcessMessage1(MessageRpc& rpc) at System.ServiceModel.Dispatcher.MessageRpc.Process(Boolean isOperationContextSet

Key in here for me are “PopulateMetadata” and argument is null.

Looks like we can’t read the metadata, i.e. the CSDL, MSL and/or SSDL files. I’ve got he connection string in web.config so that ‘s not it.

With this info in hand, I found this forum post which clarified the issue: Fail to work with POCO ModelContainer which entities are located in

I neede to force the metadata to load into the application process. I have a handy trick for this which is simpler than the one suggested in the forum.

I blogged about it in 2008: Quick Trick for forcing MetadataWorkspace ItemCollections to load.

So, I added this override to the data service:

     protected override BAPOCOs CreateDataSource()
     {
       var context = new BAPOCOs();
       var tracestring = context.CreateQuery<Contact>("BAPOCOs.Contacts").ToTraceString();
       return context;
     }

Note that calling ToTraceString does not cause a command to be executed on the server.

And my service was back in working condition, however…

There is one thing I’m not happy with. I originally had a test in there to see if the metadata was already loaded. If so, then no need to call ToTraceString. The metadata is supposed to remain in teh app process so after the first hit, I should be skipping the ToTraceString code. However, it was being hit every time and if I forced it to skip, then I was back to the original error.

This seems like an expensive waste of resources and if that’s the case, I’m doing it on EVERY call, I’d rather use the more complex code which explicitly loads the MDW rather than my little trick which is not a big deal if I’m only doing it once. The upside is that this is essentially an Entity SQL query and by default ESQL queries are cached and do not need to be recompiled. So perhaps it’s not so bad. Either way, I’d like to get this solution to a more satisfactory state.

I have asked about this in the above mentioned forum thread.  And will add to this blog post when I get more info.

<Refresher>

When you pass parameters in VB, you use the ByVal and ByRef keyworks. The default in VB is ByRef with a nuance (see ByRef Parameters Passed by Value)  In C#, no keyword means by value and alternatively, you use the ref keyword to pass a reference.

The key difference is that when you pass a variable by value , you are just passing a copy. The originating variable cannot be replaced.

When passing by reference, you are working with the actual variable and can replace it.

The part that frequently evades developers is that if the value is a reference type (i.e. does not contain it’s values directly compared to, for example, an int which does) you can modify those related values whether you pass by reference or by value.

</Refresher>

Now let’s look at this with respect to an ObjectContext being passed as a parameter. An ObjectContext has it’s own values but also has references to other values (e.g. the ObjectStateEntry types that it maintains to keep track of entities).

Whether you pass the context by value or reference, the entities that the context is tracking will be impacted when you execute queries or call SaveChanges with the context.

When passing it by value you won’t replace the context, e.g, passedinContext=new MyEntities() won’t affect the originating context.

When passing it by ref, you CAN replace the context e.g., passedinContext=new MyEntities() will reinstantiate the originating context. And if you replace the context, you will lose ALL of the state information for the currently tracked entities. Those entities will be orphaned – just sitting in memory with nobody tracking them and if you do reattach them to the new instance of the context, they will all be unchanged.

That’s a big problem.

So unless you are trying to achieve some specific behavior, you should definitely pass byval. In fact you should define your method signature that takes the parameter to ensure that you are passing by value.

Tomorrow at 10am PST/1pm EST, I’ll be presenting a live web cast for O’Reilly Media called the Top 10 Treats in Entity Framework 4. The web cast is one hour (including intro and questions). Short & sweet.

You can register at http://www.oreillynet.com/pub/e/1517.

The webcast will be recorded and available at the same link approximately one week later.

Tour the Top 10 Treats in Entity Framework 4

Wednesday, January 27, 2010

10am PT, San Francisco
6pm – London | 1pm – New York | Thu, Jan 28th at 5am – Sydney | Thu, Jan 28th at 3am – Tokyo | Thu, Jan 28th at 2am – Beijing | 10:30pm – Mumbai

Presented by: Julia Lerman

Duration: Approximately 60 minutes.

Cost: Free

Visual Studio 2010 and .NET 4.0 bring us a new version of ADO.NET Entity Framework, called Entity Framework 4. From the designer, to the APIs to entire new feature sets to support agile development, even former EF skeptics are getting excited about this version. There is a lot to see. In this one hour webcast, Julie Lerman, who is currently working on the 2nd edition of Programming Entity Framework, will highlight the ten most significant changes to the Entity Framework.

About Julia Lerman

Julia Lerman is the leading independent authority on the Entity Framework and has been using and teaching the technology since its inception two years ago. She is well known in the .NET community as a Microsoft MVP, ASPInsider and INETA Speaker. She is a prolific blogger, a frequent presenter at technical conferences around the world, including DevConnections and TechEd and she writes articles for many well–known technical publications.

Julia lives in Vermont where she runs the Vermont.NET User Group, was a founding board member of the Vermont Software Developers Alliance, and is a member of the Champlain College Software Engineering Advisory Board. You can read her blog at www.thedatafarm.com/blog.

One of the benefits of Visual Studio’s multi-targeting is that if you are still stuck using the .NET 3.5 version of Entity Framework and are unable to move to EF4 (so sad), you can still get the new designer goodies for a .NET 3.5 project.  Remember that the designer is a VS2010 feature, not a .NET 4.0 feature.

Be sure to select .NET 3.5 when you create your project:

When you add the model, you’ll notice that the foreign key checkbox is disabled. FK support is not only an EF4 designer feature but it is dependent on .NET 4. But you can still get the much needed pluralization support.

Once the model is created, you can take advantage of new designer features such as the complex type support. Complex tyeps existed in .NET 3.5 but the designer did not support them making it painful to take advantage of them.

Here is that complex type in an entity that was retrieved from the database.

Model First works too:

Not every new designer feature will work with an older model though.

T4 code generation won’t. You can select “Create Code Generation Item” from the context menu, but no templates will be available.

EF 3.5 uses a different type of code generation.

Function Import will NOT provide the nifty EF4 support for mapping stored procedure results to Complex Types:

Without the Foreign Key support, the model will use what’s called “Independent Associations” in other words, the same type of associations that we were limited to in .NET 3.5 where the association was dependent on the mappings.

Even though the designer has a referential constraint in the properties window, you can’t build it with this model because the foreign key is not available.

(With an EF4 model, you can use either Independent Associations or Foreign Key Associations.)

So if you are stuck building EF v3.5 apps (oh please please move to EF4, life will be SO much happier!), you can still get a lot of benefit from using VS2010’s designer.

Just a friendly reminder that I’m passing on from one of the tech reviewers of my book who (thankfully) noticed that I had left a note in there about Single & SingleOrDefault not being supported.

They were not in .NET 3.5 SP1 Entity Framework, but they are now supported in EF4.

Single is the way to go when you only expect one item to be in your result set.

As noted by many, the benefit of Single is highlighted in testing. Single expects only one item to be in the result set. If there are more, then an exception is thrown and this can help you discover logic errors in your application.

For a list of supported & unsupported LINQ methods used (or not used) in LINQ to ENtities, see the MSDN Topic: Supported and Unsupported LINQ Methods (LINQ to Entities)

1. Agile Entity Framework 4 Repository: Part 1- Model and POCO Classes
2. Agile Entity Framework 4 Repository: Part 2- The Repository
3. Agile EF4 Repository: Part 3 -Fine Tuning the Repository
4. Agile EF4 Repository: Part 4: Compiled LINQ Queries
5. Agile Entity Framework 4 Repository Part 5: IObjectSet and Include
6. Agile Entity Framework 4 Repository: Part 6: Mocks & Unit Tests

In Part 2 of this blog series, I created a context interface, IBAGAContext, that returns IObjectSet types rather than ObjectSet types from the EntitySets.

Example:

     IObjectSet<Contact> Contacts{get;}
     IObjectSet<Customer> Customers {get;}
    

IObjectSet is the new EF4 interface that is used to create the ObjectSets returned by the default context.

For a point of comparison, here’s a method from a default code generated context:

 public ObjectSet<Activity> Activities
        {
            get
            {
                if ((_Activities == null))
                {
                    _Activities = base.CreateObjectSet<Activity>("Activities");
                }
                return _Activities;
            }
        }
        private ObjectSet<Activity> _Activities;

ObjectSet implements IObjectSet which provides it with set operations. It also inherits from QueryObject which gives it lots of EF capabilities. One of the methods that come with QueryObject is Include which provides eager loading.

context.Customers.Include(“Orders”) blah blah

(A few people, including Matthieu Mezil, have created overloads for Include that take lambdas instead of magic strings. Here’s Matthieu’s second of two posts which contains a link to the first: Entity Framework Include with Function Next. Additionally, Jamie Philips had a lambda function for EF1 that allowed further filtering. Last time we talked, he reported that it wasn’t working with EF4 POCOs.)

But Include is not available with IObjectSet.

In a recent comment, Nick asked how I was using Include on an IObjectSet.I realize I neglected to add that into the relevant post.

With nods to Jamie Phillips again, Include came from his POCO sample when he and I co-presented back-to-back EF4 POCO sessions at the New England Code Camp in October.

Jamie created an Include extension method to replace the Include supplied by the ObjectQuery. I placed the method in a separate class file in the same project that contains the model and the BAGAContext file because it simply calls the real ObjectQuery.Include.

namespace ExtensionMethods
{
  public static class MyExtensions
  {
    public static IQueryable<TSource> Include<TSource>
      (this IQueryable<TSource> source, string path)
    {
      var objectQuery = source as ObjectQuery<TSource>;
      if (objectQuery != null)
      {
        return objectQuery.Include(path);
      }
      return source;
    }
  }
}

In the repositories in Part 3 of this series, this is how I was able to use the Include method.

Jamie did write a follow-up post about the problem with the lambdas and POCO classes: Include() Extension Method and POCO