ASP.NET

Broadcast from an ASP.NET Web API to a UWP Client using SignalR 2

I recently had to take an existing ASP.NET Web API (not core) that was used by a UWP (Universal Windows Platform) client to pull information and modify it to receive notifications from the Web API under some situations. This post is going to cover the creation of a Web API and UWP projects and the modification of the projects to allow notification from the Web API to the UWP client.

Web API Project Creation

In Visual Studio we need to create a new project for the Web API using File > New > Project.

We will be creating an ASP.NET Web Application (.NET Framework) and I’m just naming the project Web API.

On the next screen select that we want a Web API application.

UPW Project Creation

As above in Visual Studio we need to create a new solution and project for the UWP Client using File > New > Project from a new instance of Visual Studio. Select the Blank App (Universal Windows) template.

Next, you will be prompted for the versions of Window 10 to support with the UWP application. I just took the defaults.

You may also want to set your PC into Developer mode.

Add SignalR to the Web API

Now that the initial project creation is done let’s add SignalR to the WebApi project. Inside of Visual Studio right-click on the project and select Manage NuGet Packages.  Select the Browse tab and in the search box enter SignalR and we are looking for the Microsoft.AspNet.SignalR package. Select it and click the Install button.

Configuration

Now that SignalR is installed it needs to be configured on startup of the application. To do this add a Startup class if it doesn’t exist with the following contents.

public class Startup
{
    public void Configuration(IAppBuilder app)
    {
        var hubConfiguration = new HubConfiguration
        {
            EnableDetailedErrors = true,
            EnableJavaScriptProxies = false
        };
        app.MapSignalR(hubConfiguration);

    }
}

The hubConfiguration bit is only needed if you want to change the default settings. In this example, we don’t have any JavaScript clients so we don’t need those proxies and we are learning so detailed errors will come in handy.

Hub

The first thing we need to do is create a new hub which is what SignalR uses to do all of its magic. Hubs are what enable the server (and/or client depending on how you are using it) to make remote calls. Since we are just doing a broadcast the hub class is empty except for a reference to the class that will be handling when to send the broadcast. This property is very important since it will be the trigger for Broadcaster to get instantiated when a client connects.  The following is the full class.

using Microsoft.AspNet.SignalR;
using Microsoft.AspNet.SignalR.Hubs;

namespace WebApi.Hubs
{
    [HubName("BroadcastHub")]
    public class BroadcastHub : Hub
    {
        private readonly Broadcaster _broadcaster = Broadcaster.Instance;
    }
}

Make note of the hub name used as we will need it on the client side.

Next, we have the class that actually does the broadcasting. The following is the full class.

using System;
using System.Threading;
using Microsoft.AspNet.SignalR;
using Microsoft.AspNet.SignalR.Hubs;

namespace WebApi.Hubs
{
    public class Broadcaster
    {
        private readonly TimeSpan _updateInterval = 
            TimeSpan.FromMilliseconds(250);
        private Timer _timer;

        private static readonly Lazy<Broadcaster> _instance = 
            new Lazy<Broadcaster>(() => new Broadcaster(GlobalHost.ConnectionManager.GetHubContext<BroadcastHub>().Clients));
        public static Broadcaster Instance => _instance.Value;

        public IHubConnectionContext<dynamic> Clients { get; set; }

        public Broadcaster(IHubConnectionContext<dynamic> clients)
        {
            Clients = clients;
            _timer = new Timer(Broadcast, null, _updateInterval, _updateInterval);
        }

        public void Broadcast(object state)
        {
            Clients.All.Broadcast(DateTime.Now);
        }
    }
}

The above is a singleton which gets created with the list of clients for a specific hub which can be seen in the following line. It has been reformatted from above to try and make it a bit more readable.

private static readonly Lazy<Broadcaster> _instance = 
  new Lazy<Broadcaster>(() => 
             new Broadcaster(GlobalHost
                            .ConnectionManager
                            .GetHubContext<BroadcastHub>()
                            .Clients));

The constructor of the class sets up a timer that calls a Broadcast function which sends the current date and time to all the connected clients. This could be any sort of data, but for this sample, I’m keeping it simple.

Add SignalR to the UWP Client

Now over in the UWP project, we need to add the SignalR client by right-clicking on the project and selecting Manage NuGet Packages. As in the Web API instructions above search for SignalR using the Browse tab. Select the Microsoft.AspNet.SignalR.Client package and click Install.

UI

To keep the sample as simple as possible we are just going to add a TextBlock to the MainPage.xaml to display the results of the broadcast from the Web API. The following is the full code for the page.

<Page
    x:Class="UwpClient.MainPage"
    xmlns="http://schemas.microsoft.com/winfx/2006/xaml/presentation"
    xmlns:x="http://schemas.microsoft.com/winfx/2006/xaml"
    xmlns:local="using:UwpClient"
    xmlns:d="http://schemas.microsoft.com/expression/blend/2008"
    xmlns:mc="http://schemas.openxmlformats.org/markup-compatibility/2006">

    <Grid Background="{ThemeResource ApplicationPageBackgroundThemeBrush}">
        <TextBlock Name="BroadcastResults" />
    </Grid>
</Page>

Now in the code behind we need to connect to the hub on the Web API and define what happens when a broadcast is received. The following is all of the code in the code behind which will be followed up with some comments.

public sealed partial class MainPage : Page
{
    public MainPage()
    {
        this.InitializeComponent();
        InitilizeHub();
    }

    private async void InitilizeHub()
    {
        var hubConnection = new HubConnection("http://localhost:50869");
        var hubProxy = hubConnection.CreateHubProxy("BroadcastHub");

        hubProxy
        .On<DateTime>("Broadcast",
                      async data => 
                            await Dispatcher
                                  .RunAsync(CoreDispatcherPriority.Normal,
                                            () => BroadcastResults.Text = 
                                                          data.ToString()));
        await hubConnection.Start();
    }
}

The hubConnection as you might expect controls the connection with the server and the hubProxy deals with the interaction with a specific hub. In our case, we are saying with the Broadcast function is called on the server we want the client to update the BroadcastResults.Text with the value from the server.

Wrapping Up

Run the Web API and then UWP application and you will see the date time from the server being shown on the client. This is, of course, is the simplest of use cases for SignalR, but it is a great jumping off point to go deeper.

This tutorial from the official docs helped a lot when getting my sample up and running.

The code for this sample can be found here.

Broadcast from an ASP.NET Web API to a UWP Client using SignalR 2 Read More »

More OAuth

Last week I covered using OAuth with Google for authentication. I thought I would try GitHub next only to find out Microsoft does not have a built in provider. After a little searching I came across OWIN OAuth Providers on GitHub. This project has a ton for providers ranging from BattleNet to Yammer.

To install run the following from the package manager console.

Install-Package Owin.Security.Providers

The rest of the process is very similar to last weeks post. I am going to walk through GitHub for this example but the others are very similar. The big difference for each provider is getting API access.

At the top of Statup.Auth.cs in the App_Start folder add the following.

using Owin.Security.Providers.GitHub;

And at the bottom of the ConfigureAuth function.

app.UseGitHubAuthentication(
    clientId: "Your Client ID",
    clientSecret:"Your Client Secret");

The next setup is to set up API access with GitHub. Login to GitHub and go to Developer applications.

oauthgithub

Click the Register new application button in the upper right of the page. On the registration page enter an application name, homepage URL and authorization callback URL. As last week the URL options will be need to be based on your project settings. After entering all your information click Register application.

oauthgithubRegisterApplication

Then next page will show your client ID and client secret which just need to be entered in Statup.Auth.cs. That is all there is to adding an additional OAuth provider and give your users even more authentication options.

oauthgithubLogin

Using OWIN OAuth Providers makes adding access other OAuth providers just as simple as the providers from Microsoft.

More OAuth Read More »

OAuth 2.0 with Google

After getting Basic Authentication and Authorization working I thought it would be neat to add login via Google which involves using OAuth 2.0. To get started I used NuGet to update all the OWIN related packages in my project. I ran across some people who had issues with older versions of OWIN so I recommend getting the latest to avoid any potential problems.

Some third parties require SSL when using their auth services so I decided to go a head and change my application to use SSL. I hit a couple of issues during this part of the process that I want to point out. The first is that when you change to SSL IIS Express defaults to using port 44300. I missed this the first try and it took a good bit of searching before I spotted the problem. Even after reverting all my changes I could not get my site to load without errors. A reboot got my reverted site working which I am assuming was a configuration issue with IIS Express that got reset on reboot.

To enable SSL select the project in Solution Explorer and press F4 to bring up the properties window (which has different options that the project properties you get if you right-click the project and click properties). Set the SSL Enabled property to true. Copy the SSL URL to use when updating the project.ContactsPropertiesSsl

Back in the Solution Explorer right-click on the project and click properties. On the Web tab under the Servers section paste the SSL URL from above into the Project Url and save.ContactsProjectPropertiesSslUrl

When running the first time after enabling SSL Visual Studio will prompt asking if you would like to trust the self-signed certificate that IIS Express generated. I chose to trust in order to avoid warnings from the browser.

OAuthTrustIISSSL

The last change need in the project is in the Statup.Auth.cs found in the App_Start folder. In the ConfigureAuth function add the following code using your own ClientId and ClientSecret.

app.UseGoogleAuthentication(new GoogleOAuth2AuthenticationOptions()
{
    ClientId = "Your Client ID",
    ClientSecret = "Your Client Secret"
});

If you don’t have a client ID and client secret head to the Google Developers Console. The first step in the process is to Click the Create Project button.

GoogleDevelopersConsoleEnter a project name and project ID. The refresh button on the project ID field will randomly generate project IDs in case the one you want is already in use. When finished click Create.
GoogleDevelopersConsoleNewProjectAfter the project creation process finishes click on the name of your project. Next click on the APIs & auth section and then click the Consent screen option. This set of options determines what the user sees when Google prompts a user for consent to use information from their Google account. The minimum needed is email address and product name. For a live application I would recommend filling out as much as possible to give the user the best experience. When done click Save.GoogleDevelopersConsoleAPIsAndAuthConsentScreenIn the same APIs & auth section click the APIs option. Search for Google+ API. Click the name Google+ API and on the next screen click Enable API.GoogleDevelopersConsoleAPIsAndAuthApisGooglePlusAgain in the APIs & auth section click on Credentials option and then in the OAuth section fo the page click Create new Client ID.
GoogleDevelopersConsoleAPIsAndAuthCredentialsThe following dialog will show. Select the appropriate Application type which is Web application for this example. For Authorized JavaScript origins use the value from Project Url listed above which in my case is https://localhost:44300/. For Authorized redirect URIs the base URI is the same but with an added level. For MVC 5 the redirect should be set to https://localhost:44300/signin-google changing the base URI as needed of course. Click Create Client ID and you will be returned to the previous page that will now list the Client ID and Client Secret needed in Statup.Auth.cs.
GoogleDevelopersConsoleAPIsAndAuthCredentialsCreateClientId

Now the login page will have a button for Google which will allow users to create an account and associate it with their Google account. After the association users will be able to login with their Google account.LoginWithGoogle

 

From this point adding Microsoft, Facebook or Twitter would just be a matter of adding the desired options to ConfigureAuth Startup.Auth.cs and going to each service and requesting API access.

OAuth 2.0 with Google Read More »

Basic Authentication and Authorization

When I created my contact application I did so with authentication set to Individual User Accounts as seen below.
vs2013NewAspProject

 

By choosing an authentication option during project creation Visual Studio takes care of a lot of the work involved with getting authentication and authorization up and running. For individual user accounts the default data store is SQL and Visual Studio sets up the need classes and data using entity framework to create a database and the needed tables to store authentication data. An AccountController and associated views are also created to handle user creation and login.

I am not going to cover everything that Visual Studio created automatically. I will be walking you through the other changes are needed to start using authentication and authorization with all the bits Visual Studio has provided.

The first step is to add AuthorizeAttribute to the FilterConfig which is found in the App_Start folder.

public class FilterConfig
{
    public static void RegisterGlobalFilters(GlobalFilterCollection filters)
    {
        filters.Add(new HandleErrorAttribute());
        filters.Add(new AuthorizeAttribute());
    }
}

Now that the AuthorizeAttribute is added to the applications filters any page that is requested by a user that is not authorized will be redirected to a login page to authenticate.

With authorization and authentication now in play any controller action that should allow anonymous usage will need to have the AllowAnonymous attribute added. The following example is using AllowAnonymous on the index action of the home controller.

public class HomeController : Controller
{
    [AllowAnonymous]
    public ActionResult Index()
    {
        return View();
    }
}

That is all there is to get the very basics going. This is a big subject, but Visual Studio and .Net do a lot of work to make getting basics up and going simple.

Basic Authentication and Authorization Read More »

Adding Paging

As the number of contacts grows it is overwhelming to see in single page. To address this I am going to add paging to the contact list and only show 10 contacts per page.

To generate a large set of realistic date I am using Mockaroo. They offers data in CSV, Tab-Delimited, SQL, Excel, JSON and DBUnit XML with up to 1,000 rows of data per generation for free. The data Mockaroo is way better than the crazy stuff I was coming up with. If you need more than 1,000 rows at a time they do have a pay option.

For paging I am going to take advantage of the PagedList.Mvc NuGet package. Using the package manager console it can be installed with the Install-Package PagedList.Mvc command as seen in the following screenshotPackageManager-PagedListMvc.

To start using PagedList.Mvc I need the following changes to ContactsController. Add new using for PagedList and the index action needs a page parameter. As part of this change I also refactored the contact query to the GetContacts function so the Index action is much less cluttered than before. The function is now shown below but I moved the distinct city query to the GetDistinctCities function.

        
public ActionResult Index(string filterCity, string filterSearch, int? page)
{
    ViewBag.filterCity = filterCity;
    ViewBag.filterSearch = filterSearch;
    ViewBag.city = new SelectList(GetDistinctCities());

    var contacts = GetContacts(filterCity, 
                               filterSearch)
                   .ToPagedList(page ?? 1, 10);

    if (Request.IsAjaxRequest())
    {
        return PartialView("_ContactList", contacts);
    }

    return View(contacts);
}

private IQueryable<Contact> GetContacts(string city, string search)
{
    var contacts = from c in db.Contacts
                   select c;

    if (!string.IsNullOrWhiteSpace(search))
    {
        contacts = contacts.Where(c => c.Name.Contains(search));
    }

    if (!string.IsNullOrWhiteSpace(city))
    {
        contacts = contacts.Where(c => c.City == city);
    }
     return contacts.OrderBy(c => c.Name);
}

Notice that I am now saving the parameter values of the city and search filters to the ViewBag in the index action which will be used when changing pages when a filter is in play.

In GetContacts I had to add an order by statement to the contacts being returned. This is to make sure the list is always in the same order. It will also be a good spot to add different ordering options in the future.

The last change in the Index action was to take the return value from GetContacts and call ToPagedList on it. This function takes a page number and number of items per page and returns a paged list as the name implies. In my case if page is null I am using page 1 and each page will contain 10 items.

In both Index.cshtml and the partial view _ContactList.cshtml the models need to be changed to PagedList.IPagedList. In addition @Html.DisplayNameFor also needs to be changed from model.Name to model.FirstOrDefault().Name. The last step is to add the paged list control to the bottom of the view. Here is the code from the contact list partial view.

@using PagedList.Mvc
@model PagedList.IPagedList<Contacts.Models.Contact>

<p>
    @Html.ActionLink("Create New", "Create")
</p>
<table class="table">
    <tr>
        <th>
            @Html.DisplayNameFor(model => model.FirstOrDefault().Name)
        </th>
        <th>
            @Html.DisplayNameFor(model => model.FirstOrDefault().Address)
        </th>
        <th>
            @Html.DisplayNameFor(model => model.FirstOrDefault().City)
        </th>
        <th>
            @Html.DisplayNameFor(model => model.FirstOrDefault().State)
        </th>
        <th>
            @Html.DisplayNameFor(model => model.FirstOrDefault().ZipCode)
        </th>
        <th>
            @Html.DisplayNameFor(model => model.FirstOrDefault().Country)
        </th>
        <th></th>
    </tr>

    @foreach (var item in Model)
    {
        <tr>
            <td>
                @Html.DisplayFor(modelItem => item.Name)
            </td>
            <td>
                @Html.DisplayFor(modelItem => item.Address)
            </td>
            <td>
                @Html.DisplayFor(modelItem => item.City)
            </td>
            <td>
                @Html.DisplayFor(modelItem => item.State)
            </td>
            <td>
                @Html.DisplayFor(modelItem => item.ZipCode)
            </td>
            <td>
                @Html.DisplayFor(modelItem => item.Country)
            </td>
            <td>
                @Html.ActionLink("Edit", "Edit", new { id = item.Id }) |
                @Html.ActionLink("Details", "Details", new { id = item.Id }) |
                @Html.ActionLink("Delete", "Delete", new { id = item.Id })
            </td>
        </tr>
    }

</table>

@Html.PagedListPager(Model,
                     page => Url.Action("Index",
                                        "Contacts",
                                        new { ViewBag.filterCity,
                                              ViewBag.filterSearch,
                                              page }),
                     PagedListRenderOptions.EnableUnobtrusiveAjaxReplacing(
                          new AjaxOptions
                          {
                              HttpMethod = "GET",
                              UpdateTargetId = "contactList"
                          }))

The Url.Action on the PagedListPager is the function that will be called when the user want to changes pages. Index is the action that will be called, Contacts is the controller that the action will be called on. Next is an anonymous type that will be processed to determine what parameters the index action will be called with. It is important that the existing filters are passed to the index action or else when the pager requests a different page it will return all contacts instead of next page of the filtered set of contacts.

The other nice option that the PagedListPager offers is the option to use ajax when requesting a different page. Since the contact page is using ajax when filtering it would be silly not to do the same when paging. To enable ajax just use the PagedListRenderOptions.EnableUnobtrusiveAjaxReplacing with the same AjaxOptions as the filter form.

As a side note I have changed the AjaxOptions on the index view that are used for filtering to match what is shown above for paging. I also removed the accepted verbs from the index action so it only responds to get requests. The request for a new set of data based on a filter change is really more of a get operation than a post operation.

Adding Paging Read More »

Partial View with AJAX

This is step two of applying filters to my contact list without refreshing the whole page. I will be using the partial view created in last week’s post.

I ran into a bit of trouble getting ajax working. It seems that starting in MVC 5 the jQuery plugin needed to make this form of ajax work is no longer included by default. In order to fix use NuGet and install jQuery.Unobtrusive.Ajax.nuget-jqueryajax

After the NuGet package is installed add a new bundle (or add to an existing bundle) for the jQuery unobtrusive files to the BundleConfig found in the App_Start folder.

bundles.Add(new ScriptBundle("~/bundles/jqueryunob").Include(
            "~/Scripts/jquery.unobtrusive*"));

If a new bundle was added then make sure to render the new bundle in the _Layout.cshtml file in the Views/Shared folder.

@Scripts.Render("~/bundles/jqueryunob")

The next step was to change the index action of contacts controller. The index action now needs to accept both gets and posts which can be accomplished via the AcceptVerbs attribute. The second change needed in the index action is to return a partial view instead of a view if the request is an ajax request.

[AcceptVerbs(HttpVerbs.Get | HttpVerbs.Post)]
public ActionResult Index(string city, string search)
{
    var cityList = new List<string>();
    var cityDistinct = from c in db.Contacts
                       orderby c.City
                       select c.City;

    cityList.AddRange(cityDistinct.Distinct());
    ViewBag.city = new SelectList(cityList);

    var contacts = from c in db.Contacts
                   select c;

    if (!string.IsNullOrWhiteSpace(search))
    {
        contacts = contacts.Where(c => c.Name.Contains(search));
    }

    if (!string.IsNullOrWhiteSpace(city))
    {
        contacts = contacts.Where(c => c.City == city);
    }

    if (Request.IsAjaxRequest())
    {
        return PartialView("_ContactList", contacts);
    }

    return View(contacts);
}

Reusing the index action is only one option to implement the needed changes. Another option would be to add another action to the controller that would return the partial view. In this case the  common query code would be moved to a function.

The last set of changes needed are in the index view. Instead of using Html.BeginForm Ajax.BeginForm should be used. The section of the page that will be replaced via the ajax request needs to be moved to a div with an id. In the example below I added a div with the id of contactList. It is important that the div id match the UpdateTargetId in AjaxOptions.

@model IEnumerable<Contacts.Models.Contact>

@{
    ViewBag.Title = "Index";
}

<h2>Index</h2>

<p>
    @using (Ajax.BeginForm("Index",
                           "Contacts",
                           new AjaxOptions
                           {
                               UpdateTargetId = "contactList"
                           }))
    {
        <p>
            City: @Html.DropDownList("city",
                                     ViewBag.city as SelectList,
                                     "All",
                                     new {@class = "city",
                                          onchange = "$(this.form).submit();"})
            Name: @Html.TextBox("Search")
            <input type="submit" value="Filter"/>
        </p>
    }
</p>
<div id="contactList">
    @{
        Html.RenderPartial("_ContactList");
    }
</div>

One thing to make special note of is the that the onchange for the city drop down list has change from “this.form.submit();” to “$(this.form).submit();”. Without this change the request will not come through as an ajax request and the full page will refresh instead of just the contact list section. I wasted a lot of time trying to track down why a full page request was happening and the issue ended up being the way that the drop down list was submitting the form.

Partial View with AJAX Read More »

Partial Views

On my contacts list page I want to apply filters without refreshing the whole page. This post is about step one of my first try at accomplishing this goal.

To add a partial view right clicked on the destination folder for the new view and from the Add menu select the View option.AddViewMenu

This brings up the Add View dialog. I am adding a view call _ContactList using the list template since this is going to be a list of contact. Contact is the model the list will be based on and any data access will happen via the ContactDbContext. The last thing to do is check the create as a partial view check box. Clicking add creates a _ContactList.cshtml file.AddViewDialog

The resulting code in _ConactList.cshtml looks like this:

@model IEnumerable<Contacts.Models.Contact>

<p>
    @Html.ActionLink("Create New", "Create")
</p>
<table class="table">
    <tr>
        <th>
            @Html.DisplayNameFor(model => model.Name)
        </th>
        <th>
            @Html.DisplayNameFor(model => model.Address)
        </th>
        <th>
            @Html.DisplayNameFor(model => model.City)
        </th>
        <th>
            @Html.DisplayNameFor(model => model.State)
        </th>
        <th>
            @Html.DisplayNameFor(model => model.ZipCode)
        </th>
        <th>
            @Html.DisplayNameFor(model => model.Country)
        </th>
        <th></th>
    </tr>

@foreach (var item in Model) {
    <tr>
        <td>
            @Html.DisplayFor(modelItem => item.Name)
        </td>
        <td>
            @Html.DisplayFor(modelItem => item.Address)
        </td>
        <td>
            @Html.DisplayFor(modelItem => item.City)
        </td>
        <td>
            @Html.DisplayFor(modelItem => item.State)
        </td>
        <td>
            @Html.DisplayFor(modelItem => item.ZipCode)
        </td>
        <td>
            @Html.DisplayFor(modelItem => item.Country)
        </td>
        <td>
            @Html.ActionLink("Edit", "Edit", new { id=item.Id }) |
            @Html.ActionLink("Details", "Details", new { id=item.Id }) |
            @Html.ActionLink("Delete", "Delete", new { id=item.Id })
        </td>
    </tr>
}

</table>

As you can see by choosing the contact model Visual Studio was able to build a view that displays all the properties of the contact. Since I chose the list template the view is expecting a list of contacts that it will iterate over that list and create a table from the list of contacts.

In my index view I was able to remove all the code associated with the contact list and replace it with the following code which renders my new partial view.

@{
    Html.RenderPartial("_ContactList");
}

This partial view will allow me to have a consistent view of my contact list any where I may need it. It is also my hope that this partial view will allow me to refresh the list portion of the contact list page when a filter is changed.

Partial Views Read More »

Add a Dropdown Filter

On my contacts application I wanted the option to only show contacts from a specific city. To accomplish this I am going to use a dropdown with a distinct list of cities that exist in my contact list.

Below is part of the existing contacts controller with the Index action that gets all contacts and returns a view.

public class ContactsController : Controller
{
    private ContactDbContext db = new ContactDbContext();

    // GET: Contacts
    public ActionResult Index()
    {
        var contacts = from c in db.Contacts
                       select c;
        return View(contacts);
    }
}

Using ContactDbContext, which is a entity framework DbContext for contacts, the database will be queried for a list of distinct cities. The list of distinct cities is then stored as a select list in the view bag.

var cityList = new List<string>();
var cityDistinct = from c in db.Contacts
                   orderby c.City
                   select c.City;

cityList.AddRange(cityDistinct.Distinct());
ViewBag.city = new SelectList(cityList);

ViewBag is a dynamic object and is one way to pass data from a controller to a view. A dynamic object basically allows you to add properties just by setting them. In the above example the with ViewBag.city is adding a city property to the ViewBag and setting it equal to a new SelectList.

The resulting index action now takes a city parameter which is used when querying for contact if it has a value.

public ActionResult Index(string city)
{
    var cityList = new List<string>();
    var cityDistinct = from c in db.Contacts
                       orderby c.City
                       select c.City;

    cityList.AddRange(cityDistinct.Distinct());
    ViewBag.city = new SelectList(cityList);

    var contacts = from c in db.Contacts
                   select c;

    if (!string.IsNullOrWhiteSpace(city))
    {
        contacts = contacts.Where(c => c.City == city);
    }

    return View(contacts);
}

In my razor view the following adds the new city filter to the UI.

@using (Html.BeginForm("Index", "Contacts", FormMethod.Get))
{
<p>
    City: @Html.DropDownList("city", 
                             ViewBag.city as SelectList, 
                             "All", 
                             new {@class = "city", 
                                  onchange = "this.form.submit();"})
</p>
}

“All” is the label used for the default empty item. The last bit is for HTML attributes and set the class for the drop down to “city” and makes it so the form submits on change of the drop down. This means that when a new city is selected the view will refresh and only show contacts for the newly selected city.

This is the resulting UI:CityFilterUi

Add a Dropdown Filter Read More »

Data Annotations

Validation via data annotations in .NET is a great feature that is simple to implement. Here is an example model class that is using data annotations.

using System.ComponentModel.DataAnnotations;

namespace Contacts.Models
{
    public class Contact
    {
        [Required]
        [StringLength(50, MinimumLength = 4)]
        public string Name { get; set; }
        [StringLength(2)]
        public string State { get; set; }
        [Display(Name = "Zip Code")]
        [StringLength(10)]
        public string ZipCode { get; set; }
    }
}

First step in using data annotations is to add a using statement for the System.ComponentModel.DataAnnotations namespace. This namespace allows you to add validation attributes to the properties that require validation and contains helper classes to run validate on attributed models.

In the example above Required, StringLength and Display are all validation attributes. For the most part the attributes are self-explanatory, but be aware most attributes also have options to give you more control of the validation. For example required attribute has an AllowEmptyStrings property that if set to true will allow an empty string to pass validation. String length is another example which has options for minimum and maximum lengths for the property being validated.

This is just a small sample of the built-in attributes that exist. Other built-in attributes range  from credit card and email address validation to file extension and range validation. If a built-in validation does not meet your needs then a custom attribute a can be created by deriving from ValidationAttribue.

MVC has a lot of great functionality built around data annotations. For instance as long as a client has JavaScript turned a form will not allow submission until the model passes all the validation conditions. On the server-side inside of a controller to run validation all it takes is to run ModelState.IsValid.

With winforms the story is not quite as simple, but still pretty straight forward. The following is a simple example of calling validation on a model.

private void ValidationExample()
{
    var contact = new Contact { Name = "Bob", State = "TN" };

    var context = new ValidationContext(contact);
    var errors = new List<ValidationResult>();

    if (!Validator.TryValidateObject(contact, context, errors))
    {
        foreach (ValidationResult result in errors)
        {
            //act on error
        }
    }
}

Validator.TryValidateObject takes the model to be validated, a validation context (created with the model to validate) and an empty list of type ValidationResult. The function call returns false if any properties fail validation. In the example above the validation would fail since name only contains three characters the minimum length is set to four characters.

Data Annotations Read More »