Developmentor Instructor's Blog Posts

Dominick Baier: Web Apps vs Web Services: 302s and 401s are not always good Friends

Fri, 10 Feb 2012 12:28:59 -0600

It is not very uncommon to have web sites that have web UX and services content. The UX part maybe uses WS-Federation (or some other redirect based mechanism). That means whenever an authorization error occurs (401 status code), this is picked by the corresponding redirect module and turned into a redirect (302) to the login page. All is good.

But in services, when you emit a 401, you typically want that status code to travel back to the client agent, so it can do error handling. These two approaches conflict.

If you think (like me) that you should separate UX and services into separate apps, you don’t need to read on. Just do it ;)

If you need to mix both mechanisms in a single app – here’s how I solved it for a project.

I sub classed the redirect module – this was in my case the WIF WS-Federation HTTP module and modified the OnAuthorizationFailed method. In there I check for a special HttpContext item, and if that is present, I suppress the redirect. Otherwise everything works as normal:

class ServiceAwareWSFederationAuthenticationModule : WSFederationAuthenticationModule

{

    protected override void OnAuthorizationFailed(AuthorizationFailedEventArgs e)

    {

        base.OnAuthorizationFailed(e);



        var isService
= HttpContext.Current.Items[AdvertiseWcfInHttpPipelineBehavior.DefaultLabel];



        if (isService
!= null)

        {

            e.RedirectToIdentityProvider
= false;

        }

    }

}

Now the question is, how do you smuggle that value into the HttpContext. If it is a MVC based web service, that’s easy of course. In the case of WCF, one approach that worked for me was to set it in a service behavior (dispatch message inspector to be exact):

public void BeforeSendReply(

ref Message reply, object correlationState)

{

    if (HttpContext.Current
!= null)

    {

        HttpContext.Current.Items[DefaultLabel]
= true;

    }

}

HTH

Dominick Baier: Replacing ASP.NET Forms Authentication with WIF Session Authentication (for the better)

Thu, 09 Feb 2012 15:14:54 -0600

ASP.NET Forms Authentication and WIF Session Authentication (which has *nothing* to do with ASP.NET sessions) are very similar. Both inspect incoming requests for a special cookie that contains identity information, if that cookie is present it gets validated and if that is successful, the identity information is made available to the application via HttpContext.User/Thread.CurrentPrincipal.

The main difference between the two is the identity to cookie serialization engine that sits below. Whereas ForsmAuth can only store the name of the user and an additional UserData string. It is limited to a single cookie and hardcoded to protection via the machine key.

WIF session authentication in turn has these additional features:

Can serialize a complete ClaimsPrincipal (including claims) to the cookie(s).
Has a cookie overflow mechanism when data gets too big. In total it can create up to 8 cookies (á 4 KB) per domain (not that I would recommend round tripping that much data).
Supports server side caching (which is an extensible mechanism).
Has an extensible mechanism for protection (DPAPI by default, RSA as an option for web farms, and machine key based protection is coming in .NET 4.5)

So in other words – session authentication is the superior technology, and if done cleverly enough you can replace FormsAuth without any changes to your application code. The only features missing is the redirect mechanism to a login page and an easy to use API to set authentication cookies. But that’s easy to add ;)

FormsSessionAuthenticationModule
This module is a sub class of the standard WIF session module, adding the following features:

Handling EndRequest to do the redirect on 401s to the login page configured for FormsAuth.
Reads the FormsAuth cookie name, cookie domain, timeout and require SSL settings to configure the module accordingly.
Implements sliding expiration if configured for FormsAuth. It also uses the same algorithm as FormsAuth to calculate when the cookie needs renewal.
Implements caching of the principal on the server side (aka session mode) if configured in an AppSetting.
Supports claims transformation via a ClaimsAuthenticationManager.

As you can see, the whole module is designed to easily replace the FormsAuth mechanism. Simply set the authentication mode to None and register the module. In the spirit of the FormsAuthentication class, there is also now a SessionAuthentication class with the same methods and signatures (e.g. SetAuthCookie and SignOut). The rest of your application code should not be affected.

In addition the session module looks for a HttpContext item called “NoRedirect”. If that exists, the redirect to the login page will *not* happen, instead the 401 is passed back to the client. Very useful if you are implementing services or web APIs where you want the actual status code to be preserved. A corresponding UnauthorizedResult is provided that gives you easy access to the context item.

The download contains a sample app, the module and an inspector for session cookies and tokens. Let’s hope that in .NET 4.5 such a module comes out of the box.

HTH

Maurice de Beijer: Windows Workflow Foundation 3 Types Marked Obsolete in .NET 4.5

Thu, 09 Feb 2012 07:42:18 -0600

People have been wondering for a while what the future of WF3 was since the release of WF4. So far both workflow stacks have coexisted in .NET 4 and there has been no official statement about the future of the older WF3 stack.

That has just changed!

The workflow team at Microsoft has just announced that they are marking the WF3 stack as deprecated with the next release of the .NET framework, .NET 4.5.

Of course the types are still there but you will get compile time warnings for using them and you can expect the classes to be dropped some time in the future.

“Warning BC40000: X is obsolete: ‘WF 3 types are deprecated. Please use WF 4 instead.’”

So if you are sill using WF3 this is a good moment to invest the time and start learning Wf4.

You can read the official announcement here.

Enjoy!

TheProblemSolver
DotNetEvents

Maurice de Beijer: Controlling when the value is updated with Knockout.js

Wed, 08 Feb 2012 04:02:00 -0600

In the previous blog post about Knockout.js I showed why and how to get started with Knockout.js. And I explained that the reason I really like Knockout.js is that it is a very familiar way of working with its MVVM style.

I created a small demo where we could update the first and last name of a person and the ViewModel would combine the two and display the concatenated parts as the complete name. This worked just fine as soon as I started using observables except for one thing. Whenever I start typing the full name isn’t updated until the <input> control losses focus.

Quite often this will be good enough as the resulting value isn’t used right away but there are cases, like here, where it is and we want more frequent updates.

The value binding

As it turns out the value data binding has an additional option, the valueUpdate, that controls when the value, and therefor the computed observable full name is updated. By adding that to the data-bind expression we can get real time updates. There are several options but for these real time scenarios using the value of afterkeydown works best.

The markup now looks like this:

<fieldset>

    <legend>Enter person</legend>

<p>

        First name:

        <input type="text" data-bind="value: firstName, valueUpdate: 'afterkeydown'" />

    </p>

<p>

        last name:

        <input type="text" data-bind="value: lastName, valueUpdate: 'afterkeydown'" />

    </p>

<p>

        Full name: <span data-bind="text: fullName"></span>

    </p>

</fieldset>

The JavaScript is unchanged and looks like this:

$(function () {

    var viewModel = {};

    viewModel.firstName = ko.observable("Maurice");

    viewModel.lastName = ko.observable("de Beijer");

    viewModel.fullName = ko.computed(function () {

        return viewModel.firstName() + " " + viewModel.lastName();

});

    ko.applyBindings(viewModel);

});

With this change the full name is updated with each character entered.

Sweet

Enjoy!

TheProblemSolver
DotNetEvents

Maurice de Beijer: How to get started with Knockout.js

Mon, 06 Feb 2012 03:31:00 -0600

Once you get into doing more client side JavaScript code with business applications and REST services you are going to run into the question of how to construct the client side HTML required to show the data to the users.

Using jQuery

Assuming most people are going to be using jQuery on the client you might start with some jQuery code to generate HTML.

Your code might look something like this

$(function () {

    $("#btn").click(function () {

        $.getJSON("/services/books").then(function (books) {

            $.each(books, function () {

                $("<li>").text(this.Title + " by " + this.Author).appendTo("#books");

});

});

});

});

The code isn’t very complicated but then we are only showing a simple bit of text. And in all likelihood the actual HTML that needs to be generated will be quite a bit more complex. And the code increases as least as much, but often much more, in complexity.

Using templates

The next step people tend to take is using templates. There are lots of different templating libraries out there with more coming but in this example I am using the JavaScript Micro-Templating function John Resig wrote. If you are looking for others take a look at Underscore.js or JsRender/JsViews (still in preview at the moment). Specially Underscore.js is a nice library to look at because all the additional functions in there.

Using templates we separate the code from the template to generate the required HTML. The code is easier to understand.

$(function () {

    $("#btn").click(function () {

        $.getJSON("/services/books").then(function (books) {

            var html = tmpl("books_tmpl", { books: books });

            $("#books").html(html);

});

});

});

However this comes at a cost as we also need a template and that is a mixture of markup and expressions. The expressions syntax varies depending on the templating engine, in this case it’s <%= JavaScript Expression %>. Not hard to use but it is another bit of executing code that has to be maintained.

<script type="text/html" id="books_tmpl">

  <% for ( var i = 0; i < books.length; i++ ) {

    var book = books[i];%>

    <li><a href="/services/books/<%=book.Id%>"><%=book.Title%> by <%=book.Author%></a></li>

  <% } %>

</script>

As you can see the template is actually embedded as a <script> block. Not a requirement as it is just a string but an easy way to do so. It does however mean that you have to handcraft the HTML with no designer or IntelliSense support.

Using Knockout.js

As soon as we start using Knockout.js this become somewhat more like Silverlight/WPF. Not that it is exactly the same but it uses the same MVVM pattern and data binding that Silverlight developers are used to. Something I really like . Now Knockout.js is my no means the only way to do this, just as with templates there are plenty of other options like Backbone or JavaScriptMVC for example. See this blog post for a longer list of alternatives.

Personally I really like the MVVM and binding style Knockout.js uses.

Instead of splitting the HTML into a static and a dynamically generated part with Knockout.js you embed data binding expressions in your HTML. Notice the data-bind attributes in the following HTML snippet. They indicate data binding syntax. The first two data bind the value property of the <input> control to the firstName and lastName from the ViewModel. The third data binds the text in the <span> against the fullName of the ViewModel.

<fieldset>

    <legend>Enter person</legend>

<p>

        First name:

        <input type="text" data-bind="value: firstName" />

    </p>

<p>

        last name:

        <input type="text" data-bind="value: lastName" />

    </p>

<p>

        Full name: <span data-bind="text: fullName"></span>

    </p>

</fieldset>

So what does the view model look like? It can be a real simple JavaScript object. In fact the following would work. It would be less than perfect thou and we can do better.

var viewModel = {

    firstName: "Maurice",

    lastName: "de Beijer",

    fullName: "Maurice de Beijer"

};

A better way is to use the Knockout functionality and define the data fields using observables. By using Observables we can observer changes being made. And by defining the fullName as a computed observable we can have the <span> with the fullName dynamically update. A much better ViewModel:

var viewModel = {};

viewModel.firstName = ko.observable("Maurice");

viewModel.lastName = ko.observable("de Beijer");

viewModel.fullName = ko.computed(function () {

    return viewModel.firstName() + " " + viewModel.lastName();

});

Of course we still need to tell Knockout to use the ViewModel and that takes all of one line:

ko.applyBindings(viewModel);

So the complete JavaScript in my page is now:

$(function () {

    var viewModel = {};

    viewModel.firstName = ko.observable("Maurice");

    viewModel.lastName = ko.observable("de Beijer");

    viewModel.fullName = ko.computed(function () {

        return viewModel.firstName() + " " + viewModel.lastName();

});

    ko.applyBindings(viewModel);

});

And the nice thing is that updating the first name automatically updates the fullName

This is just the first baby steps using Knockout.js as it is far more capable than just this. In fact we can data bind against arrays of data and create nice list etc. The Knockout.js site has some nice tutorials that will help you get started.

Enjoy!

TheProblemSolver
DotNetEvents

Dominick Baier: Mixing Forms and Token Authentication in a single ASP.NET Application (the Details)

Thu, 02 Feb 2012 01:47:52 -0600

The scenario described in my last post works because of the design around HTTP modules in ASP.NET. Authentication related modules (like Forms authentication and WIF WS-Fed/Sessions) typically subscribe to three events in the pipeline – AuthenticateRequest/PostAuthenticateRequest for pre-processing and EndRequest for post-processing (like making redirects to a login page).

In the pre-processing stage it is the modules’ job to determine the identity of the client based on incoming HTTP details (like a header, cookie, form post) and set HttpContext.User and Thread.CurrentPrincipal. The actual page (in the ExecuteHandler event) “sees” the identity that the last module has set.

So in our case there are three modules in effect:

FormsAuthenticationModule (AuthenticateRequest, EndRequest)
WSFederationAuthenticationModule (AuthenticateRequest, PostAuthenticateRequest, EndRequest)
SessionAuthenticationModule (AuthenticateRequest, PostAuthenticateRequest)

So let’s have a look at the different scenario we have when mixing Forms auth and WS-Federation.

Anoymous request to unprotected resource
This is the easiest case. Since there is no WIF session cookie or a FormsAuth cookie, these modules do nothing. The WSFed module creates an anonymous ClaimsPrincipal and calls the registered ClaimsAuthenticationManager (if any) to transform it. The result (by default an anonymous ClaimsPrincipal) gets set.

Anonymous request to FormsAuth protected resource
This is the scenario where an anonymous user tries to access a FormsAuth protected resource for the first time. The principal is anonymous and before the page gets rendered, the Authorize attribute kicks in. The attribute determines that the user needs authentication and therefor sets a 401 status code and ends the request. Now execution jumps to the EndRequest event, where the FormsAuth module takes over. The module then converts the 401 to a redirect (302) to the forms login page.

If authentication is successful, the login page sets the FormsAuth cookie.

FormsAuth authenticated request to a FormsAuth protected resource
Now a FormsAuth cookie is present, which gets validated by the FormsAuth module. This cookie gets turned into a GenericPrincipal/FormsIdentity combination. The WS-Fed module turns the principal into a ClaimsPrincipal and calls the registered ClaimsAuthenticationManager. The outcome of that gets set on the context.

Anonymous request to STS protected resource
This time the anonymous user tries to access an STS protected resource (a controller decorated with the RequireTokenAuthentication attribute). The attribute determines that the user needs STS authentication by checking the authentication type on the current principal. If this is not Federation, the redirect to the STS will be made.

After successful authentication at the STS, the STS posts the token back to the application (using WS-Federation syntax).

Postback from STS authentication
After the postback, the WS-Fed module finds the token response and validates the contained token. If successful, the token gets transformed by the ClaimsAuthenticationManager, and the outcome is a) stored in a session cookie, and b) set on the context.

STS authenticated request to an STS protected resource
This time the WIF Session authentication module kicks in because it can find the previously issued session cookie. The module re-hydrates the ClaimsPrincipal from the cookie and sets it.

FormsAuth and STS authenticated request to a protected resource
This is kind of an odd case – e.g. the user first authenticated using Forms and after that using the STS. This time the FormsAuth module does its work, and then afterwards the session module stomps over the context with the session principal. In other words, the STS identity wins.

What about roles?
A common way to set roles in ASP.NET is to use the role manager feature. There is a corresponding HTTP module for that (RoleManagerModule) that handles PostAuthenticateRequest. Does this collide with the above combinations?

No it doesn’t! When the WS-Fed module turns existing principals into a ClaimsPrincipal (like it did with the FormsIdentity), it also checks for RolePrincipal (which is the principal type created by role manager), and turns the roles in role claims. Nice!

But as you can see in the last scenario above, this might result in unnecessary work, so I would rather recommend consolidating all role work (and other claims transformations) into the ClaimsAuthenticationManager. In there you can check for the authentication type of the incoming principal and act accordingly.

HTH

Dominick Baier: Mixing Forms and Token Authentication in a single ASP.NET Application

Wed, 01 Feb 2012 23:08:38 -0600

I recently had the task to find out how to mix ASP.NET Forms Authentication with WIF’s WS-Federation. The FormsAuth app did already exist, and a new sub-directory of this application should use ADFS for authentication. Minimum changes to the existing application code would be a plus ;)

Since the application is using ASP.NET MVC this was quite easy to accomplish – WebForms would be a little harder, but still doable. I will discuss the MVC solution here.

To solve this problem, I made the following changes to the standard MVC internet application template:

Added WIF’s WSFederationAuthenticationModule and SessionAuthenticationModule to the modules section.
Add a WIF configuration section to configure the trust with ADFS.
Added a new authorization attribute. This attribute will go on controller that demand ADFS (or STS in general) authentication.

The attribute logic is quite simple – it checks for authenticated users – and additionally that the authentication type is set to Federation. If that’s the case all is good, if not, the redirect to the STS will be triggered.

public class RequireTokenAuthenticationAttribute : AuthorizeAttribute

{

    protected override bool AuthorizeCore(HttpContextBase httpContext)

    {

        if (httpContext.User.Identity.IsAuthenticated
&&

            httpContext.User.Identity.AuthenticationType.Equals(

WIF.AuthenticationTypes.Federation, StringComparison.OrdinalIgnoreCase))

        {

            return true;

        }

            


        return false;

    }



    protected override void HandleUnauthorizedRequest(AuthorizationContext filterContext)

    {            


        //
do the redirect to the STS

        var message
= FederatedAuthentication.WSFederationAuthenticationModule.CreateSignInRequest(

"passive", 


filterContext.HttpContext.Request.RawUrl, 


false);

        filterContext.Result = new RedirectResult(message.RequestUrl);

    }

}

That’s it ;) If you want to know why this works (and a possible gotcha) – read my next post.

Tony Sneed: 2011 Recap

Tue, 31 Jan 2012 13:15:00 -0600

After a rather long break from blogging, it’s time for me to jump back in! I thought I would start by taking a look back at 2011 and recapping some of my experiences, with a sneak peek at blogging topics … Continue reading →

Pinku Surana: You’re throwing money away!

Mon, 30 Jan 2012 21:40:19 -0600

I am for the first time seriously considering making a bid on a home. Coincidentally, two seemingly intelligent women chattering next to me at lunch today agreed that when renting “you are throwing money away”, whereas buying a house means you will (somehow) get that money back. It’s bizarre that most people don’t understand the [...]

Jason Diamond: WebSocket Webinar Demo Code

Mon, 30 Jan 2012 14:08:01 -0600

Thanks to those who attended today’s webinar. The demo code is located here until it (and the recording) get uploaded to the DM website later this week. Feel free to email with questions!

Pinku Surana: PyPy is fast

Fri, 27 Jan 2012 23:05:15 -0600

I wrote a small Python program that maintains a heap. It took 20 minutes to process a very large data file with Python 2.7. On a whim I tried PyPy. It finished the task in 1 minute. That’s 20X faster without making any changes to my code. Why isn’t everyone using PyPy?

Llewellyn Falco: Creating Namers

Fri, 27 Jan 2012 19:35:00 -0600

If your current testing framework is not supported, you can create your own namer to allow you to use it.
Check out this namer for ideas how to make your own.

Maurice de Beijer: DotNed Podcast: Koen Zwikstra over Silverlight 5 en de toekomst van Silverlight

Fri, 27 Jan 2012 04:39:00 -0600

Er is weer een nieuwe DotNed podcast online. In deze podcast spreekt Maurice de Beijer met Koen Zwikstra over de recente Silverlight 5 release en hoe de toekomst er voor Silverlight ontwikkelaars uitziet. Verder vertelt hij over Silverlight Spy een runtime inspector waarmee je willekeurige Silverlight applicaties kan inspecteren. Hij kondigt ook nog even aan dat hij druk bezig is met een Spy voor Metro applicaties op Windows 8.

Links:

Blog: http://firstfloorsoftware.com/blog/
First Floor Software: http://firstfloorsoftware.com/
Silverlight Spy: http://firstfloorsoftware.com/silverlightspy/
Balder - 3D engine for Silverlight: http://balder.codeplex.com/

enjoy!

TheProblemSolver
DotNetEvents

Dominick Baier: Thinktecture IdentityServer Azure Edition RC

Fri, 27 Jan 2012 03:20:37 -0600

I found some time over the holidays to finalize the Azure edition of IdentityServer.

http://identityserver.codeplex.com/releases/view/81206

The biggest difference to the on-premise version (and earlier Azure betas) is, that by default IdSrv now uses Azure Storage for all data storage (configuration & user data). This means that there is no need anymore for SQL Azure (which is still supported out of the box – just not the default anymore).

The download includes a readme file with setup instructions. In a nutshell:

Create a new hosted service and upload your certificates
Modify the service configuration file in the download to your needs (signing cert, connection strings to storage…)
Deploy the package via the portal or other tools
Use the new Powershell scripts to add users

If you encounter any problem, please give me feedback.

Mark Blomsma: Type-Safe Include extension method for Entity Framework

Tue, 24 Jan 2012 14:47:02 -0600

If you’re interested in the Include method in a LINQ to Entities queries to be type-safe, then Joe Ferner has the answer for you in this post: Type-Safe Entity Framework Include: http://www.nearinfinity.com/blogs/joe_ferner/type-safe_entity_framework_inc.html

Maurice de Beijer: Deploying SqlServerCe with an ASP.NET MVC application

Tue, 24 Jan 2012 04:50:00 -0600

Using Entity Framework Code First together with SqlServerCe is a great way to work with small databases in ASP.NET MVC applications. I have several web applications running on shared hosting sites where SqlServerCe is the database engine under the hood. Deploying is also quite easy. I typically use the build in Publish Web option which works great over FTP.

When using SqlServerCe you have to make sure you add the deployable dependencies. Forgetting those results in pretty clear error messages.

However once that is done I always run into another issue. When a page actually tries to use SqlServerCe I will see a security exception:

System.UnauthorizedAccessException: Access is denied. (Exception from HRESULT: 0x80070005 (E_ACCESSDENIED))

The solution is simple enough. Go to whatever security portal your provider is using, mine is using Plesk, and make sure the user that is actually running the web site has the required read/write privileges to work with the Bin folder.

Enjoy!

TheProblemSolver
DotNetEvents

Michael Kennedy: Screencast: Building ASP.NET MVC Forms with Razor (ASP.NET MVC Foundations Series)

Mon, 23 Jan 2012 16:45:53 -0600

[note: This screencast has been adapted from my earlier blog post.]

In this ASP.NET MVC Foundations screencast, we’re going to look at building an ASP.NET MVC page which allows users to create and edit objects in our domain. We’ll cover just the basics of using HTML helpers to map model properties to our HTML form and Model Binding to convert our HTML form back into our rich domain object.

We’ll start with a very basic store website which has read-only data and we'll add the ability to create and edit products in our store. Be sure to watch in HD mode for a crisp screen.

All content copyright Michael C. Kennedy. All information, source code, and especially tools are provided as is and on a "use at your own risk" basis.

Maurice de Beijer: HTML5, Internet Explorer and automatic updates

Mon, 23 Jan 2012 04:24:00 -0600

One of the problems with HTML5, or development at the cutting edge of the web, has always the need to support a large number of different browsers. Depending on who you ask the numbers will vary slightly but is usually boils down to Internet Explorer being the biggest and Chrome and FireFox each taking quite a big chunk of the stats and the remainder of the browsers filling up the gaps.

So that means that any public facing website should at the very least support the three major browsers, not to bad right?

Turns out that live isn’t quite as good as that. With Chrome and FireFox most users are at or close to the last revision of the browser but with Internet Explorer this isn’t the case. In fact as the chart below shows the majority is still using IE8 with IE9 only accounting for a little over a third of all IE users.

So what the big difference?

There are 2 reasons.

First of all IE9 is only available on Windows Vista and Windows 7 and it turns out there is still a substantial number of people using Windows XP. For all those users Internet Explorer 8 is the latest version of the browser they can use. And that is not going to change when Internet Explorer 10 ships. However that number of XP users isn’t that large.

The second and more important reason is that Microsoft doesn’t automatically update Internet Explorer when a newer version is available. You can download it if you want to but if you don’t explicitly do so nothing happens. And by contrast all recent versions of FireFox and Chrome are self updating, so whenever a new version is available it is downloaded without any explicit user action. The benefit is that with both Chrome and FireFox a web developer wanting to do cutting edge HTML5 stuff can be pretty confident that a user has an up to date browser. Well except with Internet explorer that is

Microsoft has seen the update light

Fortunately that won’t be a problem much longer. Recently Microsoft has announced that it will start auto updating Internet Explorer using Windows Update. So no longer are we dependent on users going in and manually doing an update, instead if they don’t take any action they will be automatically upgraded to the last possible version of IE. This means that we should soon see at leas two thirds of all IE users use IE9, and more important for HTML5 developers, a more compliant browser with a much faster JavaScript engine.

Of course IE9 still isn’t perfect because it supports less HTML5 features than both Chrome and FireFox but the fact that the JavaScript engine is much faster makes it much easier when using polyfills to insert missing pieces. And the fast JavaScript engine and more compliant rendering engine is great news for all web developers

And the this update will mean that adoption of IE10 will be a lot faster when it ships later this year.

The auto update doesn’t start worldwide right away. At first it will start in Australia and Brazil only. But when that is done they plan on doing so in more countries around the world.

A big step forwards for Microsoft and a huge step forward for HTML5 developers all over the world

Enjoy!

TheProblemSolver
DotNetEvents

Michael Kennedy: Building ASP.NET MVC Forms with Razor (ASP.NET MVC Foundations Series)

Fri, 20 Jan 2012 14:30:04 -0600

[Update: Want to watch this as a screencast rather than article? Look here...]

In this ASP.NET MVC Foundations article, we’re going to look at building an ASP.NET MVC page which allows users to create and edit objects in our domain. We’ll cover just the basics of using HTML helpers to map model properties to our HTML form and Model Binding to convert our HTML form back into our rich domain object.

We’ll start with a very basic store website (downloads here: BasicMvcForms_starter.zip and BasicMvcForms_final.zip)which has a database and some basic products already listed:

Notice that we have five products. There links to edit and create products. Currently, they don’t do much:

First we’ll need to add action methods. That’s easy enough. Here’s the create code:

And similarly, the edit code:

Now that we have action methods for edit and create, we’ll need the corresponding views. Use the Visual Studio tooling to create two empty views as follows:

In the resulting dialog, choose strongly-typed with Product and an empty view without referencing the scripts (we do this globally already). In practice, you might choose “edit” and “create” to help jump-start the Razor code. In this example, we’ll do that from scratch so empty is what we want.

Now our links to edit and create products work. However they don’t have any content in their views. We’ll use the HTML Helper methods to convert our product into forms ready for the editing.

It all starts with Html.BeginForm(). We’ll define a form using this MVC convention and helper method as follows:

Next, we use the Model property of the view and the HTML helper methods to define the input fields. Note that we’re using Html.TextBoxFor() and Html.LabelFor() to create the fields. Our Featured property is a Boolean, so we’ll use Html.CheckBoxFor() on that one.

Once we flesh out the other properties, we’ll be finished with the create view. And it turns out the edit view is identical. There are tricks to share them across actions (e.g. PartialViews) but for our simple example, we’ll just copy / paste between the two views.

Last thing we need is a submit button to submit the form.

Now we should have a nice usable form to create products (or edit them if you copy & pasted that view). Here’s the edit view:

The final step is to capture the form data on the controller methods and update the database. This is where it gets interesting.

We will define a second method for each action (create & edit) which accept the post. We do not want our “show the form to start editing” code to mix with the “save the data and move on” code. We’ll achieve this separation using two attributes HttpGet & HttpPost.

Notice the original Create method has the GET attribute. This displays the form to begin creating a product. The second one is more interesting. It adds the product to the DB and returns to the product list. It also only accepts POST requests. Notice that it accepts a Product parameter which is populated using model binding by ASP.NET MVC.

Learn this pattern! It’s super common in MVC. I’ll call it the Get+Post+Redirect pattern.

1. HttpGet method shows form, returns View().
2. HttpPost method accepts the model (which is populated using model binding)
3. HttpPost method then updates the data and redirects to a new view.

Edit is similar. Often, websites use AutoMapper to do the manual copy / update you see here.

Now we have a fully functioning store (albeit a simple one). One glaring omision is validation. We’ll cover that in another post.

Conclusion

To wrap up, we took a basic MVC website and went through these steps:

- Added edit and create methods
- The new methods returned the correct model
- Added strongly-typed views for each method
- Used Html.BeginForm() and related Html.XXXXFor() methods to build out the form
- Added a submit button
- Implemented the Get+Post+Redirect pattern in edit and create.
- Took the rest of the day off. :)

Good luck with your websites and happy POSTing.

- Cheers
@mkennedy

All content copyright Michael C. Kennedy. All information, source code, and especially tools are provided as is and on a "use at your own risk" basis.

Maurice de Beijer: Getting Started with WCF and Rest material

Fri, 20 Jan 2012 02:24:29 -0600

Thanks everyone for joining in with the DevelopMentor webinar I did last night on Getting Started with WCF and Rest.

If you want to take another look at the slides or samples you can download them using the links below:

For more info about the WCF Web API you can check the WCF CodePlex site.

Enjoy!

TheProblemSolver
DotNetEvents

Andy Clymer: Developmentor has moved

Tue, 17 Jan 2012 15:24:00 -0600

In addition to working hard at RSK myself and Richard are also instructors for Developmtor. Both of us author and teach a range of courses for them. This year has seen Developmentor move from Hammersmith to central London, for all of you who are based in and around London they are offering an evening of tech and nibbles to show case the new location. Details below, hopefully see some of you there.

Come and see us! On Monday 6th February 2012, 5-7pm to cement the move to our new London City location in Moorgate, we invite you to our house warming, entertainment provided by Richard Blewett and Andy Clymerwho will be revealing the magic behind the new and exciting C#5 Async and AWait keywords. To register your interest for this free event, please email us salesuk@develop.com or call 01242 525108 .

Mark Blomsma: Generating an index in the database using Entity Framework Code First

Tue, 17 Jan 2012 11:24:40 -0600

If you’re using Entity Framework Code First, then you may want to create an index on some of your table. The way to do this is to call “Database.ExecuteSqlCommand” in the “Seed” method of your database initializer. The sample below shows how it is done (thanks to Rolf for pointing me in the right direction):

using System;

using System.Collections.Generic;

using System.Linq;

using System.Text;

using System.Data.Entity;

using System.ComponentModel.DataAnnotations;

namespace CodeFirstPlayground

  class Program

    static void Main( string[]
args )

      Database.SetInitializer<CodeFirstSampleModel>( new CodeFirstSampleDbInitializer()
);

      using ( var model = new CodeFirstSampleModel()
)

        var query = from c in model.Customers

                    where c.Name != null

                    select c;

        foreach ( var item in query
)

          Console.WriteLine( item.Name );

      Console.ReadLine();

  public class CodeFirstSampleModel
: DbContext

    public CodeFirstSampleModel()

      : base( "CodeFirstSampleDB" )

      Customers = this.Set<Customer>();

    public DbSet<Customer> Customers
{ get; set; }

  public class CodeFirstSampleDbInitializer
: DropCreateDatabaseAlways<CodeFirstSampleModel>

    protected override void Seed(
CodeFirstSampleModel context )

      context.Database.ExecuteSqlCommand( "CREATE
INDEX IX_Customer_Name ON Customers (Name) " );

      Customer c = new Customer()

        Name = "Mark",

        LastOrder = DateTime.Now

};

      context.Customers.Add( c );

      base.Seed( context );

  public class Customer

    [Key]

    public int Id
{ get; set; }

    [Required]

    [MaxLength( 50, ErrorMessage = "Customer
name must be 50 characters or less." )]

    public string Name {
get; set; }

    public DateTime? LastOrder { get; set; }

Richard Blewett: DevelopMentor Moorgate Office Launch

Tue, 17 Jan 2012 02:18:38 -0600

To celebrate DevelopMentor’s move to offices at 120 Moorgate in London, myself and Andy Clymer are presenting an evening of C# 5 Async. The next version of the C# language has built-in support for async processing. We’ll show you how it improves over the previous models and how it works under the covers.

This event is on Monday the 6th February from 5-7 and is free to attend, just register with the DevelopMentor office on 01242 525 108 or email salesuk@develop.com

Dominick Baier: Google Apps and IdentityServer

Mon, 16 Jan 2012 02:15:17 -0600

nice!

Andy Clymer: Code smarter with Design Patterns

Thu, 12 Jan 2012 16:47:00 -0600

Just finished updating the Code Smarter with Design Patterns for Developmentor.

The course traditionally has focused on GoF design patterns, in this release I have added MVC and Repository pattern, to round off the offering.

Both these patterns build upon the GoF patterns discussed earlier in the course, and show how the basic GoF patterns are often compounded to produce elegant solutions.

Other enhancements include addition of Prototype and Builder patterns, and a general upgrade to the latest features offered by the .NET framework.

To learn more click here

Maurice de Beijer: Wat doe ik met HTML5 op de Microsoft TechDays 2012?

Thu, 12 Jan 2012 16:24:24 -0600

Uiteraard ben ik ook dit jaar weer aanwezig op de Microsoft TechDays 2012. Voor degene die het gemist hebben de TechDays zijn dit jaar van 15 t/m 17 februari in Den Haag.

Ik doe dit jaar twee sessies:

Op vrijdag middag doe ik een sessie over het maken van een HTML5 client voor REST services.
Het maken van een REST service is pas deel een van een oplossing. Nadat deze er is moet er meestal ook een grafische interface voor gemaakt worden. En in de huidige tijd, waar het ondersteunen van diverse devices steeds belangrijker wordt, komt het maar al te vaak voor dat deze gebruikers interface in HTML geschreven wordt. Helaas blijken de verschillende browsers maar al te vaak voor extra problemen te zorgen die het leven van de ontwikkelaar een stuk moeilijker maken. In deze sessie laat Maurice de Beijer zien hoe je effectief de verschillende problemen waar je als ontwikkelaar tegenaan loopt kan oplossen.
Op donderdag avond tijdens de Geek Night doe ik mijn populaire HTML5 sessie nog een keer.
Maar uiteraard moet dat tijdens de Geek Night wel een beetje anders dus het wordt een sessie met een twist

Tot daar!

TheProblemSolver
DotNetEvents

Maurice de Beijer: HTML5 Presentatie slides en demo

Fri, 06 Jan 2012 06:31:46 -0600

De PowerPoint slides en de demo code van mijn HTML5 presentatie van gisteravond zijn nu beschikbaar.

Voor de liefhebbers ook online via SlideShare.

HTML5 Overview

View more presentations from Maurice Beijer

Met dank aan 4DotNet die de sponsor was van deze bijeenkomst.

Enjoy!

TheProblemSolver
DotNetEvents

Mark Blomsma: SharePoint UG Maine

Wed, 04 Jan 2012 04:55:41 -0600

Just a quick service announcement:

Winxnet, in Portland, has started hosting the first Maine based SharePoint User Group.

For more information visit: www.winxnet.com/spugme

Maurice de Beijer: DotNed podcast: Freena Eijffinger over Autisme en de Surface

Wed, 04 Jan 2012 04:47:00 -0600

In deze podcast spreekt Maurice de Beijer met Freena Eijffinger over autisme en hoe Freena de Microsoft Surface tafel gebruikt om autisme te diagnotiseren bij kinderen.

Met dank aan onze sponsor: Red-Gate, ingeniously simple tools

Links:

Autitouch: http://www.autitouch.com/
Freena: http://www.freena.nl/
Hart van Nederland: http://www.hartvannederland.nl/nederland/2011/autisme-snel-vastgesteld-door-game/
BizSpark: http://www.microsoft.com/bizspark/

Enjoy!

TheProblemSolver
DotNetEvents

Mark Blomsma: Microsoft Most Valuable Professional 2012

Mon, 02 Jan 2012 05:47:23 -0600

I’m honored to have been awarded the Microsoft Most Valuable Professional 2012 Award for contributions in the Visual C# technical communities. I’m honored to be counted amongst the 228 people world wide to receive this award.

“At Microsoft, we believe that technical communities enhance people’s lives and the industry’s success because independent experts, like you, help others extract greater value from products and technologies through the free and objective exchange of knowledge. As a Microsoft MVP, you are part of a highly select group of experts that represent technology’s best and brightest who share a deep commitment to community and a willingness to help others.”
- Heather Kostes

Thank you Microsoft and thank you Heather! I’m looking forward to organizing more events again this year for the Maine Developer Network and hanging out with the Bangor Area .NET Developers and I’m looking forward to Maine Code Camp #3 and also to meeting my fellow MVPs and product groups in Seattle at the MVP Summit 2012.

- Mark

Llewellyn Falco: Using Reporters in Approval Tests

Sat, 31 Dec 2011 17:33:00 -0600

Today I pushed new versions of ApprovalTests for both C# and Java to SourceForge. Updates include new capabilities around the Reporters feature set for ApprovalTests. These enhancements were driven by feature requests from Peik Aschan(@peikas). He and I pair-programmed remotely (he lives in Finland) to extend ApprovalTests per his ‘asks’. If you are using ApprovalTests and have an idea for a feature, please ping me via twitter @LlewellynFalco and I’ll pair with you to code up your idea as well.

Why Use Reporters?

I assume that you have tried out Reporters if you are reading this blog, however, just to get us started, I’ll remind you that a Reporter is called when an ApprovalTest fails. The three reasons for using different types of reporters are as below.

Visualizing Results

The output of a failed ApprovalTest can range from simple (a text file) to complex (*.png, *.html, *.mp3, etc…) If you get a *.html file, then you do not usually want to see the ‘page source’ output, rather you can usually more easily view (and approve) a HTML-rendered view. Because of this, you might want to decorate your class or method with

[UseReporter(typeof(FileLauncherReporter))]

Comparing Results

Of course, not only do ApprovalTests Reporters tell you the results, but also the results can help you determine what has changed since they last passed. While the screenshot above will help you to understand what the web page LOOKS like, it won’t help you to figure out why the test is no longer passing. To get more granular information, you may want to use a different Reporter, such as the DiffReporter. Output is shown below.

[UseReporter(typeof(DiffReporter))]

Approving the Results

The last reason to use a particular Reporter is to make the ability for you to approve the dersired output of the test easier. The DiffReporter is most often used for this, because you can just right click in your particular differencing tool (TortoiseMerge is shown above, BeyondCompare, WinMerge or KDiff can also be used) and approve the output ‘approve whole file’.

In this release of ApprovalTests, I’ve added a new reporter to increase the flexibility of approving. That reporter is called the ClipboardReporter and here is how it works. Rather than launching the output in any tool, this reporter creates the command-line output needed so that you can move the results file to the approved file quickly. It automatically adds this command to your clipboard:

[UseReporter(typeof(ClipboardReporter))]

Simply open a command prompt and past the contents of the clipboard in to the command prompt window to approve the file.

How to use Reporters

In the release, we have added the ability to decorate an assembly (C# only), class or method with multiple reporters. This is easy to do, just add the reporters of interest to the level, separated by commas. The current list of reporters is as follows:

C# Reporters	Java Reporters
BeyondCompareReporter.cs ClipboardReporter.cs DiffReporter.cs FileLauncherReporter.cs ImageReporter.cs MsTestReporter.cs MultiReporter.cs NotepadLauncher.cs NUnitReporter.cs QuietReporter.cs WinMergeReporter.cs	ClipboardReporter.java DiffReporter.java EnvironmentAwareReporter.java ExecutableQueryFailure.java FileLauncherReporter.java FirstWorkingReporterChain.java GenericDiffReporter.java ImageDiffReporter.java ImageWebReporter.java JunitReporter.java MultiReporter.java NotePadLancher.java QuietReporter.java TextWebReporter.java TortoiseDiffReporter.java WinMergeReporter.java

Here is the code example for the scenario described above, i.e. both HTML (browser) and Diff (source comparison) using Tortoise Diff.

1: using ApprovalTests.Reporters;

   2:  using NUnit.Framework;

3:

   4:  namespace ApprovalTests.Tests.Html

   5:  {

   6:      [TestFixture]

   7:      [UseReporter(typeof(DiffReporter), typeof(FileLauncherReporter))]

   8:      public class HtmlTest

   9:      {

  10:          [Test]

  11:          public static void TestHtml()

  12:          {

  13:              Approvals.ApproveHtml("<html><body><div style='font-family:Broadway;font-size:18'> Web Page from ApprovalTests</div></body></html>");

  14:          }

  15:      }

  16:  }

I have found that using the right Reporter or Reporters at the right time in the testing cycle has made me more productive. Let me know how it works for you.
Happy testing!

Dominick Baier: Troopers 2012

Sat, 31 Dec 2011 05:25:48 -0600

My last advice for 2011:

Get a ticket for Troopers 2012 before it is sold out.

If you like to learn about IPv6, Android, iOS, SAP or cloud security (and much more) – that’s the place to be!

Pinku Surana: Weather Prediction Error

Fri, 30 Dec 2011 14:20:35 -0600

I wrote a Python script to download the 1 through 5 day predictions for the weather in the top 50 American cities from the Weather Underground. I’ve got almost 2 years of data now. Here are the RMSE for predictions for all cities. I believe the RMSE for all cities is 2.24, 2.89, 3.68, 4.29, [...]

Mark Blomsma: How to discover what font was used

Mon, 26 Dec 2011 06:34:25 -0600

Sometimes you’re working on a website and you get some images with text in them, but no one remembers what fonts was used in the image. No fear! There is a website call “What The Font” that will take your picture and tell you what font was used: http://new.myfonts.com/WhatTheFont/

Maurice de Beijer: TechDays wildcast sessies

Fri, 23 Dec 2011 04:34:00 -0600

De TechDays, zie #TechDaysNL, komen er alweer snel aan. Maar net zoals vorig jaar zijn er ook dit jaar weer een aantal wildcard sessies die afhankelijk van de stemmen al dan niet op het programma komen.

Ga dus snel naar de lijst en stem op je favoriete sessie!

Enjoy!

www.TheProblemSolver.nl
www.dotnetevents.nl

Dominick Baier: ACS for free ‘til December 2012

Wed, 21 Dec 2011 03:43:16 -0600

see here:

http://blogs.msdn.com/b/windowsazure/archive/2011/12/20/important-announcements-regarding-the-access-control-service.aspx

Dominick Baier: Walkthrough for setting up IdentityServer v1.0

Wed, 21 Dec 2011 03:16:00 -0600

Claudio Sanchez did it again! thanks!

http://claudioasanchez.blogspot.com/2011/12/walk-though-of-provisioning-identity.html

Maurice de Beijer: #DotNed podcast: Bart De Smet over RX (Deel 2 van 2)

Tue, 20 Dec 2011 04:28:00 -0600

In deze podcast spreekt Maurice de Beijer met Bart De Smet over het RX framework. Bart werkt bij Microsoft aan de RX librray en verteld hoe deze ontwikkeling oorspronkelijk is begonnen als een onderdeel van een veel groter cloud programmability project.

Link

Enjoy!

TheProblemSolver
www.dotnetevents.nl

Pinku Surana: Upgrading Ubuntu to 11.10

Fri, 16 Dec 2011 11:56:57 -0600

I run various versions of Ubuntu within VMware. Every single time I let it do a full upgrade to a new version it fails. This time an upgrade from 11.04 to 11.10 failed to reboot. Here’s the problem and solution. This is my first look at Ubuntu’s Unity. Generally I don’t care about the desktop [...]

Maurice de Beijer: HTML5 sessie bij 4DotNet in Utrecht

Thu, 15 Dec 2011 05:03:00 -0600

Op donderdag 5 januari 2012 beginnen we het nieuwe jaar met een herhaling van de sessie van Maurice de Beijer over HTML5. Deze avond wordt gehost door 4dotnet en zal plaats vinden in Utrecht.

Er wordt al tijden veel over de nieuwe HTML5 standaard gesproken. Maar wanneer kan je als ontwikkelaar nu eigenlijk met HTML5 aan de slag? En als je het kan gaan gebruiken wat heeft het eigenlijk allemaal te bieden? Hoe zit het met al die oudere browsers die mensen misschien nog gebruiken? En dan hebben we natuurlijk nog de vraag wat ik moet doen als ik tussen Silverlight en HTML5 moet kiezen voor een nieuwe applicatie?

In deze sessie geeft Maurice de Beijer antwoord op deze vragen. Hij laat onder meer diverse nieuwe HTML5 onderdelen zien als local storage, web sockets, achtergrondtaken en nieuwe elementen zoals de video tag en het canvas. Daarnaast zal hij laten zien hoe je kan omgaan met browsers die de verschillende nieuwe HTML5 onderdelen nog niet ondersteunen.

HTML5 is niet alleen belangrijk als web-taal, maar ook in de nieuwste versie van Windows zal het maken van applicaties op de Windows Runtime met HTML5 een belangrijke optie worden. Dus als je webontwikkelaar bent, als je XAML ontwikkelaar bent en je wilt de ‘concurrentie’ in de gaten houden, of je wilt gewoon op de hoogte blijven van de laatste technieken: hier moet je bij zijn!

Je kan je hier inschrijven.

www.TheProblemSolver.nl
www.dotnetevents.nl

Maurice de Beijer: HTML5 Video

Tue, 13 Dec 2011 05:05:00 -0600

Video has always been a difficult thing on the web. With HTML5 supporting the <video> element that should become a lot easier right? After all you just add a video element, set the source and you are good to go right?

If you create a page like this:

<!DOCTYPE html>

<html>

    <head>

        <title>HTML5 Video demo</title>

        <style>

            video

                height: 200px;

                width: 400px;

        </style>

    </head>

    <body>

        <video poster="http://images.apple.com/html5/showcase/video/images/tron_legacy.jpg"

               src="http://www.apple.com/105/media/us/html5/showcase/2011/demos/apple-html5-demo-tron-us_848x352.webm"

               controls

               autoplay>

                   HTML5 Video element not supported

        </video>

    </body>

</html>

And run it in Google Chrome its start playing just like this:

Simple enough right?

However when I open the page in Internet Exporer 9 or FireFox we get this:

The video element loads the movie poster but it doesn’t start playing.

And on Safari or an older version of IE we get:

Turns out that on Windows Safari doesn’t support the HTML5 <video> element and wants users to install and use Quicktime instead. And that even though according to CanIUse.com the video element is supported on the current version of Safari I am running

So what gives?

The problem with IE9 and FireFox not playing the movie has to do with video formats. While there is an HTML5 definition of the <video> there is no definition of the video source stream. It is left up to the bowser vendors to decide on the supported standard and implement these. And as the vendors can’t agree on a standard we as developers are basically stuck with having to provide multiple video streams with different encodings Not nice but that is politics for you. There is a good table on Wikipedia describing the formats supported in each browser. But the simple story is we need to supply at least 2 formats, for example H264 and WebM.

The solution

In order to fix this we can supply multiple <source> elements and let the browser pick the appropriate format.

<!DOCTYPE html>

<html>

    <head>

        <title>HTML5 Video demo</title>

        <style>

            video

                height: 200px;

                width: 400px;

        </style>

    </head>

    <body>

        <video poster="http://images.apple.com/html5/showcase/video/images/tron_legacy.jpg"

               controls

               autoplay>

               <source src="http://www.apple.com/105/media/us/html5/showcase/2011/demos/apple-html5-demo-tron-us_848x352.webm"

                       type='video/webm; codecs="vp8, vorbis"'>

               <source src="http://movies.apple.com/media/us/html5/showcase/2011/demos/apple-html5-demo-tron-us_848x352.m4v"

                       type='video/mp4; codecs="avc1.42E01E, mp4a.40.2"'>

                       HTML5 Video element not supported

        </video>

    </body>

</html>

You can try the effect for yourself here.

One more hurdle

So in IE9 and Chrome this works just fine but in FireFox we still don’t get to see the video. This is strange as FireFox is supposed to be able to display WebM videos, what gives?

The problem is in the way the video stream is served from the web. The video stream is returned with an HTTP header:

Content-Type: text/plain; charset=UTF-8

and it turns out FireFox is a bit picky about the Content-Type. It really wants it to be

Content-Type: video/webm

for it to work. Not a problem if you control the server but in this case I don’t so I can’t fix the problem.

Conclusion

HTML5 video is nice but it certainly isn’t as nice as I hoped when I first started looking at it. But having to encode everything twice isn’t that bad so I guess I can live with the it

Enjoy!

www.TheProblemSolver.nl
www.dotnetevents.nl

Mark Blomsma: Silverlight 5 release

Mon, 12 Dec 2011 08:36:09 -0600

Silverlight 5 got released this weekend and can be downloaded here: http://www.silverlight.net/downloads.

Summary of the features

(from the Silverlight 5 download package)

Improved media support

Low Latency Audio Playback
Variable Speed Playback
H/W Decode of H.264 media
DRM Key Rotation/LiveTV Playback
Application-Restricted Media

Improved Text support

Text Tracking & Leading
Linked Text Containers
OpenType and Pixel Snapped Text
Postscript vector printing
Performance improvements for Block Layout Engine.

Building next-generation business applications

PivotViewer
ClickCount
Listbox/ComboBox type-ahead text searching
Ancestor RelativeSource Binding
Implicit DataTemplates
DataContextChanged event
Added PropertyChanged to the UpdateSourceTrigger enum
Save File and Open File Dialog
Databinding Debugging
Custom Markup Extensions
Binding on Style Setters

Silverlight 5 performance improvements

Parser Performance Improvements
Network Latency Improvements
H/W accelerated rendering in IE9 windowless mode
Multicore JIT
64-bit browser support

Graphics improvements

Improved Graphics stack
3D

"Trusted Application" model

Multiple window support
Full-Trust in-browser
In-browser HTML support
Unrestricted File System Access
P/Invoke support

Tools improvements

Visual Studio Team Test support

Dominick Baier: Claims Guide–2nd Edition PDF

Sun, 11 Dec 2011 22:48:49 -0600

http://tinyurl.com/claimsguide2

Maurice de Beijer: HTML5, Google Chrome Frame and older browsers

Thu, 08 Dec 2011 05:04:00 -0600

Consider the following problem:

You might be tempted to start working with HTML5 features in your web applications but you have to support older bowsers.

So for many HTML5 features there are polyfills available that will let you make older browsers like Internet Explorer 7 or 8 work with the newer API. See a possible list of HTML5 polyfills here. However a number of these is JavaScript based and while that works in older browsers their JavaScript engine is often quite slow as well making it hard to work with.

But how about just upgrading the complete browser?

Seems to make far more sense right? Except upgrading from IE8 to IE9 isn’t possible in all cases. One big showstopper can be people still running Windows XP because IE9 requires Windows VISTA as a minimum.

So we can upgrade to another browsers like FireFox or Chrome which has far better HTML5 support. While that technically works it causes headaches as it means people must switch browsers. And worse you might also have to support some site that only works well in IE forcing the users to switch between multiple browsers. Not a problem for techies, I do it all the time, but not much fun for the average end user.

So it turns out there is a great solution out there called Google Chrome Frame.

Users can install Google Chrome Frame in their machine and by default it does nothing noticeable. Only when a specific HTML meta tag X-UA-Compatible is added to the head section does the plugin start working. And then it intercepts the response and starts rendering it using Google Chrome Frame instead. However this is done inside the Internet Explorer window so the user has no clue he or she has just switched to Google Chrome Frame. And just like magic all the cool new HTML5 features supported in Chrome will work in what appears to be IE.

<meta http-equiv="X-UA-Compatible" content="chrome=1">

A simple example

When I try to new up a WebSocket object in IE9 I get the following error.

And with a simple addition of the HTML header the sample starts working. Notice that as far as the end user is concerned he is still using IE9, no changes there whatsoever.

Pretty cool right

And the good thing is that by default Google Chrome Frame is self updating just like Google Chrome so you know that once installed the plugin will always be up to date.

Detecting Google Chrome Frame

If you want to detect if Chrome Frame is installed ahead of time you can look at the User-Agent header the browser sends to the server. If installed and active Chrome Frame will add chromeframe/X.X.X.X where the X.X.X.X is the version number to the User-Agent string.

So why just add it to a single page?

Enabling Chrome Frame for al pages in a website is easy, just add the following to your web.config

<configuration>

  <system.webServer>

     <httpProtocol>

        <customHeaders>

           <add name="X-UA-Compatible" value="chrome=1" />

        </customHeaders>

     </httpProtocol>

  </system.webServer>

</configuration>

You can read the Google Chrome Frame FAQ with lots of extra bits of information here.

Enjoy!

www.TheProblemSolver.nl
www.dotnetevents.nl

Maurice de Beijer: #DotNed podcast: Bart De Smet over RX (Deel 1 van 2)

Wed, 07 Dec 2011 06:12:53 -0600

Links:

Blog: http://community.bartdesmet.net/blogs/bart/Default.aspx
The Reactive Extensions (Rx): http://msdn.microsoft.com/en-us/data/gg577609
Channel9: http://channel9.msdn.com/tags/Rx/
101 Rx Samples: http://rxwiki.wikidot.com/101samples
Forum: http://social.msdn.microsoft.com/Forums/en-US/rx/threads

Enjoy!

www.TheProblemSolver.nl
www.dotnetevents.nl

Dominick Baier: Small (and fixed) Bug in IdentityServer v1.0

Wed, 07 Dec 2011 02:10:24 -0600

It was brought to my attention that fedutil does not work anymore with IdSrv v1 metadata. And I can confirm that.

The reason for this bug is my recent change to the XmlWriter factory methods which have a different default behavior when it comes to encoding.

Since there were only 20 downloads so far – I fixed the bug in-place (shame on me). So when you are one of the early adopters and run into this problem – just re-download IdSrv ;)

HTH

Dominick Baier: Thinktecture IdentityServer v1.0

Tue, 06 Dec 2011 12:14:20 -0600

Yeah – it is finally done. I just uploaded the v1 bits to Codeplex and the documentation to our server. Here’s the official blurb…

Thinktecture IdentityServer is an open source security token service based on Microsoft .NET, ASP.NET MVC, WCF and WIF.

High level features

Multiple protocols support (WS-Trust, WS-Federation, OAuth2, WRAP, JSNotify, HTTP GET)
Multiple token support (SAML 1.1/2.0, SWT)
Out of the box integration with ASP.NET membership, roles and profile
Support for username/password and client certificates authentication
Support for WS-Federation metadata
Support for WS-Trust identity delegation
Extensibility points to customize configuration and user management handling

Disclaimer
I did thorough testing of all features of IdentityServer - but keep in mind that this is an open source project and I am the only architect, developer and tester on the team.
IdentityServer also lacks many of the enterprise-level features like configuration services, proxy support, operations integration etc.
I only recommend using IdentityServer if you also understand how it works (to be able to support it). I am offering consulting to help you with customization and lock down - contact me.

Download. Documentation.

Up next is v1 of the Azure version. Have fun!

Maurice de Beijer: HTML5 in een vogelvlucht

Thu, 01 Dec 2011 05:07:00 -0600

De PowerPoint presentaties van mijn HTML 5 in een vogelvlucht sessie bij Achmea is hier te downloaden en de demo code hier. De presentatie is ook op SlideShare te vinden.

HTML 5 in een vogelvlucht (Dutch)

View more presentations from Maurice Beijer

Enjoy!

www.TheProblemSolver.nl
www.dotnetevents.nl

Andy Clymer: Devweek 2012

Thu, 24 Nov 2011 04:01:00 -0600

RSK guys will be speaking at DevWeek this year, and as usual we will be talking about a large range of cutting edge topics, including

Our fluffy cuddly colouring in guy Dave Wheeler will be doing a huge number of sessions on Windows 8 Metro including the Keynote to get you all fired up, but don’t think they will be all soft, one session includes a COM primer for all those youngsters who didn’t take their COM apprentiship.

Rich will be covering topics from Whats New in WCF 4.5, and the ever cool Reactive Framework, and a deep dive into the GC, something all developers should have intimate knowledge of.

Andy will be covering the new async functionality of C#5, highlighting how you now write both ui and server side code to consume and enable the use of these new wonderful keywords, and doing a whole day of threading as a pre con on Monday

Kevin will be presenting a day of building MVC applications, along with a session on unit testing.

During the breaks Dr Rock Man will also be there to try and solve your specific technical problems or just happy to hear you rant about the code base you have just inherited.

In total we will be presenting a total 19 sessions, plus pre and post conference sessions. Hopefully see you there with what should be a fun pact week..

Pinku Surana: High Performance Counters

Wed, 23 Nov 2011 10:58:29 -0600

There doesn’t appear to be a high performance concurrent counter available in the .NET framework. I timed various simple implementations on my Core 2 Duo laptop: A single thread can increment a static field 1.5 billion times in 4 seconds. Two threads that update a static field runs slightly slower than 4 seconds probably due [...]