Archive

Posts Tagged ‘IIS’

Gofer – Silverlight

January 4, 2012 9 comments

In our last post, we used Gofer in a Console application and got data from a SQL Server database as well as creating the database from our domain model. Today, we are going to be doing the same thing but we will be doing this with Silverlight.

First, start by create a new Silverlight 5 application.

Make sure that you do NOT enable WCF RIA Services!

Next, we are going to setup of the web project first and then move over to the Silverlight project once we are done. Let’s start with getting Gofer from NuGet. Right-click on your References folder and select Manage NuGet Packages. Next, type in “Gofer” as your search criteria. Select Gofer.Sample as your choice. This package comes with the Gofer library as well as with some helper files to make testing this easier.

Gofer.Sample has a dependency on SwitchBlade and ValueInjecter.

SwitchBlade is another package that I wrote that allows you to host Razor templates outside of ASP.NET and IIS. I will be covering SwitchBlade in a future post.

ValueInjecter is a package like AutoMapper but much more convention-based and easier.

We are also going to need to use Ninject as our DI/IOC container. We will use NuGet to install this package as well:

Finally, we will need the WCF Web API from NuGet as well:

Moving on from adding all of our packages, you will also notice that you have two new template folders for your CRUD and DDL operations. You can modify these templates to shape how you want your SQL code to look when it is used by Gofer.

You will also notice two new files:

Domain.cs – This file represents a sample domain model. It is very similar to what you would see from a Northwind with some slight modifications.

TestDriver.cs – This file is a test driver class that allows us to test Gofer. We will be using a Silverlight version of this file instead. DELETE this file from the project as we do not need it.

Next, let’s add a Global.asax file to the project and modify as shown below:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Web;
using System.Web.Security;
using System.Web.SessionState;

using System.Web.Routing;
using Microsoft.ApplicationServer.Http;
using Ninject;
using Domain;
using Gofer;

namespace GoferSilverlight.Web
{
    public class Global : System.Web.HttpApplication
    {
        public const string NAMESPACE = "Domain";
        public Func<Type,bool> PREDICATE = x => x.Namespace == NAMESPACE;

        protected void Application_Start(object sender, EventArgs e)
        {
            RouteTable.Routes.SetDefaultHttpConfiguration(new WebApiConfiguration()
            {
                CreateInstance = (serviceType, context, request) => CreateInstance(serviceType),
                EnableTestClient = true
            });

            RouteTable.Routes.MapServiceRouteForAssemblyOf<Customers>(PREDICATE);
        }

        private object CreateInstance(Type serviceType)
        {
            object result = null;
            IKernel kernel = new StandardKernel();
            try
            {
                // The following is a sample entry for using the MapServiceRoute method:
                // RouteTable.Routes.MapServiceRoute<GoferService<Customers>>("Customer");
                // Hence the reason we need to pull the generic type from the GoferService.
                var genericType = serviceType.GetGenericArguments().FirstOrDefault();
                SchemaRules rules = GetRules(genericType);
                kernel.Bind(serviceType).ToSelf().WithConstructorArgument("rules", rules);
                result = kernel.Get(serviceType);
            }
            catch { }

            return result;
        }

        #region Rules

        private SchemaRules GetRules(Type type)
        {
            var result = new SchemaRules();
            result.AssemblyOf(type)
                .ShouldMap(PREDICATE)
                .GetSchema();

            result.PerformMigration = true;
            result.ForceNewMigration = false;

            return result;
        }

        #endregion

    };
}

UPDATE: I added a Fun<Type,bool> predicate so you could easily add your own logic for both the “RouteTable.Routes.MapServiceRouteForAssemblyOf<Customers>(PREDICATE)” and the .ShouldMap(PREDICATE) calls.

If you are curious as to what is going on here, please refer to my post Building a Generic Service using WCF Web API – Part II as it walks you through all of these steps.

One thing to point out here is that we are using the same SchemaRules class.

SchemaRules – This tells the Gofer engine what conventions to use for its data access. There is a ton that you can override with this class and we will take a look at that in a later post but this is the bare minimum that you need to get going. Also, you will see two properties that tell the engine whether or not to perform a migration as well as force the migration, meaning that it will drop the database and recreate it if necessary.

There is one last change that you need to put in place before we can test our service. We will need to modify the Web.config. The following is a sample Web.config that you can pattern against for yourself:

<?xml version="1.0" encoding="utf-8"?>
<!--
  For more information on how to configure your ASP.NET application, please visit
  http://go.microsoft.com/fwlink/?LinkId=169433
  -->
<configuration>
  <appSettings>
    <add key="DB_NAME" value="Example" />

    <add key="DDL_ConnectionString" value="Provider=SQLOLEDB;Server=(local);Database=master;Integrated Security=SSPI;" />
    <add key="DDL_DatabaseType" value="4" />

    <add key="ConnectionString" value="Data Source=(local);Initial Catalog=Example;Integrated Security=SSPI;" />
    <add key="DatabaseType" value="3" />

    <add key="TemplatePath" value="C:\Users\Matt\Documents\visual studio 2010\Projects\GoferSilverlight\GoferSilverlight.Web\CRUD_Templates" />
    <add key="DDL_TemplatePath" value="C:\Users\Matt\Documents\visual studio 2010\Projects\GoferSilverlight\GoferSilverlight.Web\DDL_Templates" />
  </appSettings>

  <system.web>
    <compilation debug="true" targetFramework="4.0" />
  </system.web>
  <system.serviceModel>
    <serviceHostingEnvironment aspNetCompatibilityEnabled="true" />
  </system.serviceModel>
</configuration>

If we run the application, you will see a blank screen but we can still test our service:

I am using the database that I tested with the previous post. Therefore, I knew I had at least one record in the Customers table. Based on the results of the service, I can see that I am getting back a good JSON response.

Ok, our service is ready, now let’s shift gears and see what we must do to test this on the Silverlight side.

We want to share our domain for both the client and server. Right-click your Silverlight project and select “Add | Existing Item…”. Navigate to the web project and select the Domain.cs file and click the down arrow to “Add As Link”. This will give us the domain model definition in our Silverlight project.

Next, let’s add our Gofer.Silverlight package from NuGet. Right-click on the References folder and select Manage NuGet Packages. Next, type in “Gofer” as your search criteria. Select Gofer.Silverlight as your choice.

Gofer.Silverlight has a dependency on Async CTP and HTTP Contrib which are provided as part of the install.

Async CTP is a library that helps make asynchronous programming easier.

Http Contrib is a library that helps make calling the WCF Web API easier from clients such as Silverlight.

You will also notice that a “readme.txt” file added to the project. If you open the file, you will see the you need to add the following code snippet to the constructor of your App.xaml.cs file:

HttpWebRequest.RegisterPrefix("http://", WebRequestCreator.ClientHttp);
HttpWebRequest.RegisterPrefix("https://", WebRequestCreator.ClientHttp);

I wrote a blog post here when I ran into some strange issues trying to test my services for PUT and DELETE behaviors.

Okay, let’s add the following TestDriver.cs class to the Silverlight project:

using System;
using System.Net;
using System.Windows;
using System.Windows.Controls;
using System.Windows.Documents;
using System.Windows.Ink;
using System.Windows.Input;
using System.Windows.Media;
using System.Windows.Media.Animation;
using System.Windows.Shapes;

using Gofer.DataAccess;
using Domain;

namespace GoferSilverlight
{
    public class TestDriver
    {
        public void Run()
        {
            var repo = new SilverlightRepository("Customers");
            var cust = new Customers()
            {
                CompanyName = "Bubba's Repair",
                ContactName = "Billy Bob",
                ContactTitle = "Owner",
                Address = "100 Pecan Street",
                City = "Columbia",
                PostalCode = "29661",
                Country = "USA",
                Phone = "(803) 836-1212",
                Fax = "(803) 836-1213"
            };
            repo.Create<Customers>(cust,
                (item) =>
                {
                    if (item == null)
                    {
                        // Handle data here....
                    }
                },
                (error) =>
                {
                    if (error == null)
                    {
                        // Handle error here....
                    }
                }
            );            
        }
    };
}

As you can see, this is very similar to what we used in the Console application but all calls are asynchronous and I also wanted to have a unique callback for success and errors. If the database did not exist, you could run this just like the Console code and a new database would be created as long as you had the PerformMigration property set to “true” in your Global.asax.cs file.

The final step to get this to work is to add the following code to the constructor of your MainPage.xaml.cs file:

TestDriver td = new TestDriver();
td.Run();

If you add a couple breakpoints as shown below, you should be ready to run the application:

When the debugger hits your breakpoint, you should see something similar to the following:

That’s it! This may seem like a lot of moving pieces to get this working but once you get in the groove, you will see that this is so much easier than dealing with proxies and hidden code generated files from Visual Studio. I personally really like this approach and I look forward to getting your response as well.

In the next couple of posts, I will be digging deeper into to Gofer as a whole to show you everything that you can do.

Advertisements

Gofer – Console

January 3, 2012 1 comment

By now, I am sure that you are tired of reading and just want to play with whatever I have been talking about. Well, that is exactly what we are doing to do.

First, start by creating a new Console Application. Next we are going to access my libraries using NuGet. Right-click on your References folder and select Manage NuGet Packages. Next, type in “Gofer” as your search criteria. Select Gofer.Sample as your choice. This package comes with the Gofer library as well as with some helper files to make testing this easier.

Gofer.Sample has a dependency on SwitchBlade and ValueInjecter.

SwitchBlade is another package that I wrote that allows you to host Razor templates outside of ASP.NET and IIS. I will be covering SwitchBlade in a future post.

ValueInjecter is a package like AutoMapper but much more convention-based and easier.

You will also notice that you have two new template folders for your CRUD and DDL operations. You can modify these templates to shape how you want your SQL code to look when it is used by Gofer.

You will also notice two new files:

Domain.cs – This file represents a sample domain model. It is very similar to what you would see from a Northwind with some slight modifications.

TestDriver.cs – This file is a test driver class that allows us to test Gofer.

In your Program.cs file, add the following code snippet to your Main method:

TestDriver td = new TestDriver();
td.Run();

Here is what the TestDriver class looks like:

public class TestDriver
{
    public void Run()
    {
        SchemaRules rules = new SchemaRules();
        rules.AssemblyOf<Customers>()
            .ShouldMap(x => x.Namespace == "Domain")
            .GetSchema();

        rules.PerformMigration = true;
        rules.ForceNewMigration = true;

        var repo = new Repository<Customers>(rules);
        var cust = new Customers() {
                CompanyName = "Bubba's Repair",
                ContactName = "Billy Bob",
                ContactTitle = "Owner",
                Address = "100 Pecan Street",
                City = "Columbia",
                PostalCode = "29661",
                Country = "USA",
                Phone = "(803) 836-1212",
                Fax = "(803) 836-1213"
        };
        var id = repo.Insert(cust);
        var ds = repo.Get().ToList();

        Console.WriteLine("Press any key to exit...");
        Console.ReadKey();
    }
};

As you can see, we are using two main classes from Gofer: SchemaRules and Repository.

SchemaRules – This tells the Gofer engine what conventions to use for its data access. There is a ton that you can override with this class and we will take a look at that in a later post but this is the bare minimum that you need to get going. Also, you will see two properties that tell the engine whether or not to perform a migration as well as force the migration, meaning that it will drop the database and recreate it if necessary.

Repository – This is our data access class that facilitates getting data from the Gofer engine.

There is one last change that you need to put in place before you continue. The package will have also provided you with an App.config that you will need to complete. The following is a sample App.config that you can pattern against for yourself:

<?xml version="1.0" encoding="utf-8"?>
<configuration>
  <appSettings>
    <add key="DB_NAME" value="Example" />

    <add key="DDL_ConnectionString" value="Provider=SQLOLEDB;Server=(local);Database=master;Integrated Security=SSPI;" />
    <add key="DDL_DatabaseType" value="4" />

    <add key="ConnectionString" value="Data Source=(local);Initial Catalog=Example;Integrated Security=SSPI;" />
    <add key="DatabaseType" value="3" />

    <add key="TemplatePath" value="C:\Users\Matt\Documents\Visual Studio 2010\Projects\GoferConsole\GoferConsole\CRUD_Templates" />
    <add key="DDL_TemplatePath" value="C:\Users\Matt\Documents\Visual Studio 2010\Projects\GoferConsole\GoferConsole\DDL_Templates" />
  </appSettings>
</configuration>

I used the name “Example” for the name of the database we will be using for data access. There are two connection strings since we will have one for our data access as well as one for our creation statements. You will see two different DatabaseType values you can leave for now. We will go into how you can go against any back-end system in a future post. Finally, there are two paths for the CRUD and DDL templates. Make sure that you update the paths to the directory where these folders are located on your machine.

If you run your application and you left PerformMigration to true, Gofer will create the database for you. It will then try and insert a new record and the pull all records from the Customers table.

NOTE: I did need to change my project type from the Client Profile to the full .NET 4 Framework.

Here is one last tidbit, if you put a break point after your insert statement, you can access a SqlTrace property on the Repository instance. This will show you what was executed and any error messages coming back from SQL Server. This is really helpful especially when you are migrating changes over to the database. Gofer does this automatically when you have the PerformMigration property set to true.

In my next post, I will be basically doing the same thing but using Gofer over the web for Silverlight without any need for a proxy! My main goal for Gofer is simplicity and allowing us to get back to focusing on our business rules and domain models. Gofer has a lot of extensibility and we will be going into this in future posts as well.

If you start playing with this, remember that it is the tip of the iceberg and I will be going into more depth on multiple levels. Hope you like…

HTML 5 Video tag and IIS

December 5, 2011 1 comment

HTML 5 is pretty awesome and it is only going to get better as far as removing dependendies on third-party plugins for simple things like watching movies or playing music.

Visual Studio 2010 is a great development environment. I really don’t believe that there is a better one on the the market for what it does. One thing that it does too well is setup your virtual environments for debugging your web sites either through Cassini or IIS Express. I don’t mind that it does all these neat tricks to make my development experience enjoyable but I do get concerned when I try and deploy my working application to IIS and it stops working completely. I don’t mind running into issues when I deploy an application to IIS but it is always nice to know what the problem is.

The most common reason for this error is the fact that IIS does not have any MIME types setup for the video that you are trying to stream. We can fix this in a couple ways:

  • Go into IIS and manually add in the .mp4 MIME type. If you are going to be hosting a lot of video and don’t ever want to worry about this again on your server, then this is probably the best solution.

  • If you don’t have full control over your IIS box but you still want to have your video working you can make a simple entry in your Web.config file as follows:
    <?xml version="1.0" encoding="utf-8"?>
    <!--
      For more information on how to configure your ASP.NET application, please visit
      http://go.microsoft.com/fwlink/?LinkId=152368
      -->
    <configuration>
      ...
      <system.webServer>
        <staticContent>
          <mimeMap fileExtension=".mp3" mimeType="audio/mpeg" />
          <mimeMap fileExtension=".mp4" mimeType="video/mp4" />
        </staticContent>
      </system.webServer>
    </configuration>
    

As you can see both examples are fairly easy to do but it can be somewhat frustrating when it is working perfectly from Visual Studio 2010 but does not work once you publish your solution.

Be careful not to try and implement both of these or you will get the following error:

Hope this helps…

NOTE: I also wanted to point out that since this is still new. I had to use several browsers to make sure that it was working and not just the browser itself. I found that if I couldn’t get it to work in IE 9, it would work on Chrome. So don’t give up if you are finding issues with this. Also, I have been able to play my videos on the HP TouchPad, Apple iPad, Asus EeePad, and my Windows Phone 7. All pretty cool!

Having problems with IIS caching your Silverlight application?

August 26, 2011 10 comments

Microsoft has made developing Silverlight applications really easy when using Visual Studio. Once you are finished with your initial version, you want to deploy it and make it available to your clients. Within Visual Studio you can simply perform a Publish operation on your web site that is hosting your Silverlight application.

One of the strongest frustrations that I hear from clients is that when they deploy their application and then make a minor change and redeploy, the end user never sees a changed until the clear their cache on their browser. This makes it look like your software is buggy and that Silverlight is not a stable platform. It is even harder to control this when you are dealing with a multi-tenant style enterprise application where your modules are separate XAP files.

If you have designed your application to have a quick startup and are only bringing down the required XAP files, then the following solution might work for you. In most of the applications that I design, I have a shell XAP and a separate security XAP that are downloaded immediately and once a user has successfully authenticated then I bring down any other required XAPs. I typically use Prism from the Patterns and Practices Team at Microsoft to handle all of this for me but you can do this through MEF as well.

Okay, here are the steps to ensure that your end users never have to clear their cache when using your Silverlight application regardless of how many times you push any new updates.

This solution is based on IIS 7.0 but you can also do this in 6.0 as well.

Go to your Silverlight application in IIS 7.0 and click on the application.

Next double-click on the HTTP Response Headers icon

Next click on the Set Common Headers link on the top right

Finally, check on the Expire Web content checkbox and use the default setting for immediate.

That is all you should have to do to get your application running like a champ. There is the side effect of the application being downloaded every time but if you have mitigated this with good Silverlight design, then you should be okay from a user’s experience perspective. I have seen other solutions that mention versioning your assembly but it doesn’t work so well when all that has changed is a XAP file that is not the controlling shell application.

If you are only dealing with a single XAP and you want some way to get around this, you should read this blog post by Lars Holm Jensen. This only works if you have made modifications to the primary XAP file but is a good solution for your smaller Silverlight applications.

If there is a better solution to the multi-XAP scenario then I would love to hear from you.

Hope this helps!

Categories: English Tags: , , ,