Cloud and Mobile

Practice with Mobile Services of Azure – part 2

By Xiaodong Liang 

In the last post, we draw some geometries on Android and upload to Mobile Service of Azure. As mentioned, Mobile Service of Azure allows you to access the service by HTML/JavaScript. So we could take advantage JavaScript AutoCAD 2014 provides. We will render the geometries in AutoCAD using the new technology.

Here is the sample project. Download AzureMSAutoCAD. The key points are:

1. Create an ASP.NET project. In this demo, it is named AzureMSAutoCAD. Load the necessary Js libs or Js code in Default.aspx.

<%@ Page Title="Home Page" Language="C#" MasterPageFile="~/Site.master" AutoEventWireup="true"

    CodeBehind="Default.aspx.cs" Inherits="AzureMSAutoCAD._Default" %>

 

<asp:Content ID="BodyContent" runat="server" ContentPlaceHolderID="MainContent">  

    <%--load Javascriot lib of AutoCAD--%>

    <script type="text/javascript" src="http://www.autocadws.com/jsapi/v1/Autodesk.AutoCAD.js"></script>   

    <%--load jquery lib--%>

    <script src='https://ajax.aspnetcdn.com/ajax/jQuery/jquery-1.9.1.min.js'></script>

    <%--load the js produced by Azure for your mobile service--%>

    <script src='https://lxdcell.azure-mobile.net/client/MobileServices.Web-1.0.0.min.js'></script>       

    <%--the js for this demo, which gets all records of mobile service and provides the ability the draw the geometries--%>

    <script type="text/javascript" src="Scripts\AzureAutoCAD.js"></script>

    <ul id='todo-items'>

    </ul> 

</asp:Content>

 

2. Similar to Android, you could download the Js demo from Azure. It shows how to access the service, including the access url and key.

     

 

 

 

3. In AzureAutoCAD.js of our demo, I will connect to my mobile service. The code reads each geometry data of the mobile table and produces the rows of a list in the web page.

 

$(function () {

    var client = new WindowsAzure.MobileServiceClient('Access URL', 'Key');

    var mytable = client.getTable('AzureTableItem');

 

    // Read current data and rebuild UI.

    // If you plan to generate complex UIs like this, consider using a JavaScript templating library.

    function refreshTodoItems() {

 

      // read two columns of the table and produce the rows of the list 

        var query = mytable.select("id", "graphicsdata").read().done(function (results) {

            var listItems = $.map(results, function (item) {

                return $('<li>')

    .attr('data-todoitem-data', item.graphicsdata)

    .append($('<button class="item-display">display graphics</button>'))

    .append($('<input class="item-data" readonly = "true">').val(item.id + ":"

                    + item.graphicsdata));

 

            });

 

            $('#todo-items').empty().append(listItems).toggle(listItems.length > 0);

        }, handleError);

}

// …. Other codes

 

 

 

4. Before deploying to Azure, we need to test on local. Firstly, we need to add localhost as the [Allow Requests From Host Names] in the mobile service configuration.

 

This is how it works locally. Beside each row, there is a button, it will draw the geometry in AutoCAD with the corresponding data.

 

 

 

 

5. When the ASP.net project is ready, we could host it to a website. In this demo, I deploy it to Azure (before this, please create a website in Azure). Again, remember to add the website as a host names in mobile service.

 

 

 

6. Now, we will implement the Js code that draws geometry in AutoCAD. I’d recommend you have a look at the tutorial on AutoCAD Js basic my colleague Philippe has delivered. Since the 2014 Js has not much abilities such as creating database entities, we could use extension Js which calls a .NET function. In the tutorial AutoCAD Js basic, Philippe has introduced on extension Js. Here are two more nice blogs which are more helpful to understand the mechanism how to call .NET modules.

 

http://through-the-interface.typepad.com/through_the_interface/2013/05/complementing-autocads-javascript-api-using-net.html

http://through-the-interface.typepad.com/through_the_interface/2013/05/returning-data-from-net-to-javascript-inside-autocad.html

 

7. So, in AzureAutoCAD.js of our demo, it gets the geometry data of the corresponding row, produces a json string, and calls the function of .NET. To make the display more clear, it deletes the older geometry and zoom all after the new geometry is drawn.

 

      // Handle display

    $(document.body).on('click', '.item-display', function () {

 

        //delete he previous entities, just for this demo

        Acad.Editor.executeCommand("delete all ");

 

        //the corresponding geometry data of this row

        var thisata = getTodoItemId(this);        

 

        //execute the .NET  function, passing the geometry data in.

        execAsync(JSON.stringify({

            functionName: 'JsCallDotNet',

            invokeAsCommand: false,

            functionParams: { args: thisata }

        }),

        OnArxSuccess,

        OnArxError);

 

        //zoom all

        Acad.Editor.executeCommand("zoom e");

 

    });

 

8. Next, we prepare a .NET project of AutoCAD, in which one command AdnJsDemo  loads the webpage, the other JsCallDotNet is a JavaScript Callback function.

 

   [JavaScriptCallback("JsCallDotNet")]

        public string TestDotNetRead(string jsonArgs)

        {

            Document doc = 

              Autodesk.AutoCAD.ApplicationServices.Application.
              DocumentManager.MdiActiveDocument;

            Database db = doc.Database;

            Editor ed = doc.Editor;

 

            //lock the document

            using (doc.LockDocument())

            {

 

                try

                {

                    //parse the geometry data

                    var obj_jsonArgs =
                      JObject.Parse(jsonArgs);

                    var obj_args =
                    (string)obj_jsonArgs["functionParams"]["args"];

 

                    //get the geometry data array

                    JavaScriptSerializer js =
                        new JavaScriptSerializer();

                    JsonPt[] ptArray =
                        js.Deserialize<JsonPt[]>(obj_args);

 

 

                    //draw the lines one by one

                    using (Transaction tx =

                    doc.TransactionManager.StartTransaction())

                    {

                        BlockTableRecord btr =
                tx.GetObject(db.CurrentSpaceId, OpenMode.ForWrite)

                            as BlockTableRecord;

 

                        for (int i = 0; i < ptArray.Length - 1; i++)

                        {

                            JsonPt stPt = ptArray[i];

                            JsonPt endPt = ptArray[i + 1];

 

                            Point3d oDBStPt =
                                new Point3d(stPt.X, stPt.Y, 0);

                            Point3d oDBEndPt =

                                new Point3d(endPt.X, endPt.Y, 0);

 

                            Line oNewLine = 
                                 new Line(oDBStPt, oDBEndPt);

                            oNewLine.Color =  
                                 Color.FromRgb(255, 0, 0);

                            oNewLine.LineWeight =  

                                 LineWeight.LineWeight018;

 

                            btr.AppendEntity(oNewLine);

                     tx.AddNewlyCreatedDBObject(oNewLine, true);

                        }

 

                        tx.Commit();

                    }

 

                }

                catch (System.Exception ex)

                {

                    MessageBox.Show(ex.ToString());

                }

            }

           

 

            return "{\"retCode\":0, \"result\":\"OK\"}";

        }

 

All are completed! Now, netload the .NET project in AutoCAD 2014, run the command AdnJsDemo, it will load our webpage in a panel. After the rows are generated, click the button, the corresponding geometries will drawn in AutoCAD.

 

 

Enjoy it! 

 

 

 

 

Cloud Security: Authenticating your ASP .NET App using OAuth

By Gopinath Taget

OAuth is one of the popular Authentication protocols that is used to provides a layer of security to your web applications using login/password. One of the main uses of OAuth is to allow access to one website using Authentication with another.

For instance, let us say you have a custom website with its own login and password (As a side note, the infrastructure for Authentication is referred to as membership in ASP .NET. You will also find information about it here among many other places on the web. Here is another video talking about it.).

Back on the topic, now let us say you would like to give access to a user of this website using the user’s facebook login. You can do it using OAuth. i.e., you can let users login to your website with their facebook account. Under the hood, your facebook account is associated with the users local membership account.

How difficult is this to implement? With ASP .NET 4.5, it is ridiculously easy. If you want to authenticate your ASP .NET application with users’ Facebook, Twitter, Microsoft or Google accounts have. Here is a video about using it with ASP .NET Forms application and here is a codeproject article about using it with MVC. It is as simple as calling a single method for each of the services:

OAuthWebSecurity.RegisterMicrosoftClient(
    clientId: "123456789123456",
    clientSecret: "1234567892188a3456bq44535pk34rg1");

OAuthWebSecurity.RegisterTwitterClient(
    consumerKey: "123456789123456",
    consumerSecret: "1234567892188a3456bq44535pk34rg1");

OAuthWebSecurity.RegisterFacebookClient(
    appId: "123456789123456",
    appSecret: "1234567892188a3456bq44535pk34rg1");

OAuthWebSecurity.RegisterGoogleClient();

What if you want to authenticate your web application with a provider other than facebook, twitter etc? You can still do it but it needs a little more work. Here is an article that shows how to do this.

An AutoCAD Command Usage Statistics “Magic Ball” on Cloud–Part 1

By Daniel Du

This is my another practice with Windows Azure and WebGL, an AutoCAD Command Usage Statistics “Magic Ball” on Cloud. The idea is to monitor user’s command usage while he is drafting and save it up to cloud, then view the statistics in web page with a rotatable ball, I call it “magic ball”, it does not look like a ball just because I am lazy not to draft much to use many AutoCAD commands.  

It contains three parts – an AutoCAD plugin to monitor command usage statistics, an web service on Windows Azure to receive and save the command statistics, and a Web page client to render the statistics as magic ball with HTML5/WebGL.  In this solution, it contains 5 projects, AcadCommandViewer is the service and web site, which is an ASP.NET MVC website on windows Azure. AcadCommandViewerConsole is the AutoCAD plugin.

In this post, I will introduce how to implement the server side. You probably is thinking to create a Windows Azure project in Visual Studio with Azure SDK. Yes, you are correct, but I found that it is not efficient to launch the project to test when developing, even with Azure Emulator. As it is easy to migrate an ASP.net application to Azure latter, I’d rather start from a simple ASP.net MVC Application first. So I created an ASP.net MVC 4 Web Application in Visual Studio.

First step is to create the Model of ASP.net MVC, by adding a class in Models folder of web application, but considering that the data model will also be used in AutoCAD plugin project as well, so I did some refactoring work to make the model a separate project. It is a very simple one, just one class, code goes as below:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Web;
 
namespace AcadCommandViewer.Models
{
    public class UserCommandsHit
    {
        public int Id { get; set; }
        public string UserName { get; set; }
        public virtual ICollection<CommandHit> CommandHits { get; set; }
 
    }
 
    public class CommandHit
    {
        public int Id { get; set; }
        public string CommandName { get; set; }
        public int HitNumber { get; set; }
    }
}

Model is OK now, next step is to create controller. In Visual Studio Solution Explorer, right click “Controllers” folder, Add –> Controller, I choose “API controller with read/write actions, using Entity Framework” template, which creates a controller automatically with ASP.NET web API and Entity Framework.

Select a “<New data context>” to create a new data context automatically.

If the your model class does show up in “Model class” list, you need add reference to your model project “AcadCommandViewerModels”.

The automatically created controller is pretty good, but there are some minor problems with update action. Entity Framework does not update the children elements, so we need to do some changes to the code, please refer to the code and comments inline:

using System;
using System.Collections.Generic;
using System.Data;
using System.Data.Entity;
using System.Data.Entity.Infrastructure;
using System.Linq;
using System.Net;
using System.Net.Http;
using System.Web;
using System.Web.Http;
using AcadCommandViewer.Models;
 
namespace AcadCommandViewer.Controllers
{
    public class AcadCommandsController : ApiController
    {
        private AcadCommandViewerContext db = new AcadCommandViewerContext();
 
        // GET api/AcadCommands
        public IEnumerable<UserCommandsHit> GetCommandsDatas()
        {
            return db.UserCommandsHits.OrderBy(p => p.UserName)
                .AsEnumerable();
        }
 
        // GET api/AcadCommands/5
        public UserCommandsHit GetCommandsData(int id)
        {
            UserCommandsHit commandsdata = db.UserCommandsHits.Find(id);
            if (commandsdata == null)
            {
                throw new HttpResponseException(
                    Request.CreateResponse(HttpStatusCode.NotFound));
            }
 
            return commandsdata;
        }
 
        // GET api/AcadCommands?username=Daniel
        public UserCommandsHit GetCommandsData(string userName)
        {
            UserCommandsHit commandsdata = db.UserCommandsHits
                .FirstOrDefault<UserCommandsHit>(p => 
                    p.UserName.ToUpper() == userName.ToUpper());
            if (commandsdata == null)
            {
                throw new HttpResponseException(
                    Request.CreateResponse(HttpStatusCode.NotFound));
            }
 
            return commandsdata;
        }
 
        // PUT api/AcadCommands/5
        public HttpResponseMessage PutCommandsData(int id, 
                            UserCommandsHit commandsdata)
        {
            if (ModelState.IsValid && id == commandsdata.Id)
            {
                var usrCmdHitInDb = db.UserCommandsHits.Find(id);
                //http://stackoverflow.com/questions/7968598/entity-4-1-updating-an-existing-parent-entity-with-new-child-entities
                //SetValues never updates navigation properties.
                //EF doesn't have any magic to update the children 
                // - which means: adding new children, deleting 
                // removed children, updating existing children
                // this procedure forces you to delete the old 
                // children also from the database and insert the new one
                db.Entry(usrCmdHitInDb).CurrentValues.SetValues(commandsdata);
                //workaroud is to remove 
                foreach (var ch in usrCmdHitInDb.CommandHits.ToList())
                {
                    usrCmdHitInDb.CommandHits.Remove(ch);
                }
 
                usrCmdHitInDb.CommandHits.Clear();
 
                foreach (var item in commandsdata.CommandHits)
                {
                    usrCmdHitInDb.CommandHits.Add(item);
                }
 
                try
                {
                    db.SaveChanges();
                }
                catch (DbUpdateConcurrencyException)
                {
                    return Request.CreateResponse(HttpStatusCode.NotFound);
                }
 
                return Request.CreateResponse(HttpStatusCode.OK);
            }
            else
            {
                return Request.CreateResponse(HttpStatusCode.BadRequest);
            }
        }
 
        // POST api/AcadCommands
        public HttpResponseMessage PostCommandsData(UserCommandsHit commandsdata)
        {
            if (ModelState.IsValid)
            {
                db.UserCommandsHits.Add(commandsdata);
                db.SaveChanges();
 
                HttpResponseMessage response = Request
                    .CreateResponse(HttpStatusCode.Created, commandsdata);
                response.Headers.Location = new Uri(
                    Url.Link("DefaultApi", 
                            new { id = commandsdata.Id })
                    );
                return response;
            }
            else
            {
                return Request.CreateResponse(HttpStatusCode.BadRequest);
            }
        }
 
        // DELETE api/AcadCommands/5
        public HttpResponseMessage DeleteCommandsData(int id)
        {
            UserCommandsHit commandsdata = db.UserCommandsHits.Find(id);
            if (commandsdata == null)
            {
                return Request.CreateResponse(HttpStatusCode.NotFound);
            }
 
            db.UserCommandsHits.Remove(commandsdata);
 
            try
            {
                db.SaveChanges();
            }
            catch (DbUpdateConcurrencyException)
            {
                return Request.CreateResponse(HttpStatusCode.NotFound);
            }
 
            return Request.CreateResponse(HttpStatusCode.OK, commandsdata);
        }
 
        protected override void Dispose(bool disposing)
        {
            db.Dispose();
            base.Dispose(disposing);
        }
    }
}

 

Please note that the context class is using SQL Express, it is easy to migrate to SQL Azure latter:

  <connectionStrings>
    <add name="AcadCommandViewerContext" connectionString="Data Source=.\SQLEXPRESS;
 Initial Catalog=AcadCommandViewerContext; Integrated Security=True; 
MultipleActiveResultSets=True"
      providerName="System.Data.SqlClient" />
    
  </connectionStrings>

Now make some sample data for testing, back to AcadCommandViewer project and add a class --SampleData.cs -- in Models folder:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Web;
 
namespace AcadCommandViewer.Models
{
    public static class SampleData
    {
        public static UserCommandsHit danielCmdHits = new UserCommandsHit
        {
            Id = 1,
            UserName = "Daniel",
            CommandHits = new CommandHit[] {
                    new CommandHit{
                        CommandName = "PLINE",
                        HitNumber  = 1
                    },
                    new CommandHit{
                        CommandName = "ZOOM",
                        HitNumber  = 2
                    },
                    new CommandHit{
                        CommandName = "LINE",
                        HitNumber  = 3
                    },
                    new CommandHit{
                        CommandName = "Save",
                        HitNumber  = 4
                    }
                }
        };
 
        public static UserCommandsHit jerryCmdHits = new UserCommandsHit
             {
                 Id = 2,
                 UserName = "Jerry",
                 CommandHits = new CommandHit[]
                {
                    new CommandHit{
                        CommandName = "CIRCLE",
                        HitNumber  = 2
                    },
                    new CommandHit{
                        CommandName = "Quit",
                        HitNumber  = 1
                    }
                }
             };
 
 
 
 
        public static UserCommandsHit[] userCmdsHits = new UserCommandsHit[]
        {
            danielCmdHits,
            jerryCmdHits
        };
 
 
 
    }
}

 

To generate the test data in database automatically, I need to enable database migration. Open Package Manager Console from Tools—> Library Package Manager, and type commands as below:

PM> Enable-Migrations

A class named as Configuration.cs and a folder ”Migrations” will to added into Visual Studio project.  Open the class and implement the Seed method to populate the test data :

 

namespace AcadCommandViewer.Migrations
{
    using System;
    using System.Data.Entity;
    using System.Data.Entity.Migrations;
    using System.Linq;
    using AcadCommandViewer.Models;
 
    internal sealed class Configuration : DbMigrationsConfiguration<AcadCommandViewer.Models.AcadCommandViewerContext>
    {
        public Configuration()
        {
            AutomaticMigrationsEnabled = true;
        }
 
        protected override void Seed(AcadCommandViewer.Models.AcadCommandViewerContext context)
        {
            //  This method will be called after migrating to the latest version.
 
            //  You can use the DbSet<T>.AddOrUpdate() helper extension method 
            //  to avoid creating duplicate seed data. E.g.
            //
            //    context.People.AddOrUpdate(
            //      p => p.FullName,
            //      new Person { FullName = "Andrew Peters" },
            //      new Person { FullName = "Brice Lambson" },
            //      new Person { FullName = "Rowan Miller" }
            //    );
            //
 
            //SampleData s = new SampleData();
            context.UserCommandsHits.AddOrUpdate(
                p => p.UserName,
                SampleData.userCmdsHits
                );
        }
    }
}

Then run Update-Database command in package Manager Console to create the database and test data.

PM> Update-Database

 

Okay, with that we are done the RESTful server side. I can use the tool fiddler to verify whether my Web Service works.  The usage of Fiddler to test a RESTful service has been introduced in this post.

Next post will introduce how to implement the viewer side, to render the magic ball with WebGL.

Cloud Security: Securing with SSL – REST based WCF Application

Gopinath Taget

Continuing from my previous post, I wanted to find a way to actually “see” the traffic data as it is transported so I could confirm that SSL is indeed encrypting (as well as for some cheap thrills). I have heard that there are several tools for monitoring web traffic, so I decided to use one of them called Fiddler. Fiddler is an excellent free tool that I have known about for some time but do not have any experience with. Monitoring SSL traffic seemed like a good excuse to learn how to use Fiddler. More about that in the next post.

The first step though is to have a client and a web application that I can monitor. In my previous post I created a WCF web service but I did not have a client application to use that service. So the normal thing to do would be to create a .NET client application for the WCF service. Right?

Well, I am not normal so I decided to create another WCF service. And this service is special because it is a REST based service. REST is a very popular technology for creating web services because it allows clients and web servers to exchange data in human readable XML or JSON form. This means any web browser can display the data and I don’t have to create a separate client to show the data received from the web service. This is – like we say in India – Clever no?

I used this tutorial to create a simple REST based web service using WCF. I then modified its configuration file to look like this:

<?xml version="1.0"?>

<configuration>

 

  <system.web>

    <compilation debug="true" targetFramework="4.0" />

  </system.web>

  <system.serviceModel>

    <bindings>

      <basicHttpBinding>

        <binding name="TransportSecurity">

          <security mode="Transport">

            <transport clientCredentialType="None" />

          </security>

        </binding>

 

      </basicHttpBinding>

      <webHttpBinding>

        <binding name="webHttpTransportSecurity">

          <security mode="Transport" />

        </binding>

      </webHttpBinding>

    </bindings>

    <services>

      <service name="RestBasedWCFApp.RestServiceImpl"

               behaviorConfiguration="ServiceBehaviour">

        <host>

          <baseAddresses>

            <add baseAddress="https://localhost/RestBasedWCFApp/"/>

          </baseAddresses>

        </host>

        <!--Service Endpoints-->

        <!--Unless fully qualified, address is relative to

        base address supplied above-->

        <endpoint address=""

                  binding="webHttpBinding"

                  bindingConfiguration="webHttpTransportSecurity"

                  contract="RestBasedWCFApp.IRestServiceImpl"

                  behaviorConfiguration="web">

        <!--Upon deployment, the following identity element should

        be removed or replaced to reflect the identity under which

        the deployed service runs. If removed, WCF will infer an

        appropriate identity automatically.-->

        </endpoint>

          <endpoint address="mex"

                  binding="mexHttpsBinding"

                  contract="IMetadataExchange" />

 

      </service>

    </services>

    <behaviors>

      <serviceBehaviors>

        <behavior name="ServiceBehaviour">

          <!-- To avoid disclosing metadata information, set the

          value below to false and remove the metadata endpoint

          above before deployment -->

          <serviceMetadata httpsGetEnabled="true"

                           httpGetEnabled="true"/>

          <!-- To receive exception details in faults for debugging

          purposes, set the value below to true.  Set to false

          before deployment to avoid disclosing exception

          information -->

          <serviceDebug includeExceptionDetailInFaults="false"/>

        </behavior>

        <behavior name="MyServiceBehavior">

          <serviceMetadata httpGetEnabled="false"

                           httpsGetEnabled="true"

                           httpsGetUrl="" />

          <serviceDebug

            includeExceptionDetailInFaults="false"/>

        </behavior>

      </serviceBehaviors>

      <endpointBehaviors>

        <behavior name="web">

          <webHttp/>

        </behavior>

      </endpointBehaviors>

    </behaviors>

    <serviceHostingEnvironment multipleSiteBindingsEnabled="true" />

  </system.serviceModel>

<system.webServer>

    <modules runAllManagedModulesForAllRequests="true"/>

  </system.webServer>

 

</configuration>

Here is the REST based sample itself. In my next post, I will talk about my experiences with Fiddler monitoring the REST based service.

Liberate Your Apps! Presentation and demo recording

By Philippe Leefsma

Here are links to the cloud content the ADN team was presenting during our DevDays conferences around the world in November-December 2012.

This recording will take you through the basics steps of creating and deploying a concrete WCF web service to Microsoft Azure. This is intended to people with no or little knowledge about cloud technologies who are looking for getting started with web services.

If you are looking for more advanced and detailed material, you can also take a look at my post series “Creating a Cloud based Viewer” I to X, on that same blog.

Power Point Presentation

Code Samples

Recording

Cloud Security: Securing with SSL – Securing a WCF Application

Gopinath Taget

In my previous post, I started talking about SSL and how important certificates are for SSL infrastructure. In this blog I will talk about implementing SSL in a WCF web service.

Before we start anything, please make sure you have a root certificate and a developer certificate as explained in this article. That article also talks about implementing SSL in WCF but I could not get it to work with just the steps in this article. I had to do a couple of other things as well.

Implementing SSL in WCF web service is all about configuration. You will find several tutorials on the web on how to create a simple WCF service, like the one here, here and here so I am not going to go into that. I am using Visual Studio 2010 to create a WCF service and implement SSL in it.

Once you have a WCF service project in Visual Studio 2010, open the web.config file. Almost everything you need to do to get your web services to use SSL goes here. Here is my complete web.config file:

 

<?xml version="1.0"?>

<configuration>

 

  <system.web>

    <compilation debug="true" targetFramework="4.0" />

  </system.web>

  <system.serviceModel>

    <bindings>

      <basicHttpBinding>

        <binding name="TransportSecurity">

          <security mode="Transport">

            <transport clientCredentialType="None" />

          </security>

        </binding>

      </basicHttpBinding>

    </bindings>

    <services>

      <service name="WCFWithSSL_2.WCFSampleService"

               behaviorConfiguration="MyServiceBehavior">

        <host>

          <baseAddresses>

            <add baseAddress="https://localhost/WCFWithSSL_2/"/>

          </baseAddresses>

        </host>

        <endpoint address="Service1.svc"

                  binding="basicHttpBinding"

                  bindingConfiguration="TransportSecurity"

                  contract="WCFWithSSL_2.IWCFSampleService" />

        <endpoint address="mex"

                  binding="mexHttpsBinding"

                  contract="IMetadataExchange" />

      </service>

    </services>

    <behaviors>

      <serviceBehaviors>

        <!--<behavior>

          --><!-- To avoid disclosing metadata information,

          set the value below to false and remove the metadata

          endpoint above before deployment --><!--

          <serviceMetadata httpsGetEnabled="true"/>

          --><!-- To receive exception details in faults for

          debugging purposes, set the value below to true. 

          Set to false before deployment to avoid disclosing

          exception information --><!--

          <serviceDebug includeExceptionDetailInFaults="false"/>

        </behavior>-->

        <behavior name="MyServiceBehavior">

          <serviceMetadata httpsGetEnabled="true" httpsGetUrl="" />

          <serviceDebug includeExceptionDetailInFaults="false"/>

        </behavior>

      </serviceBehaviors>

    </behaviors>

    <serviceHostingEnvironment multipleSiteBindingsEnabled="true" />

  </system.serviceModel>

<system.webServer>

    <modules runAllManagedModulesForAllRequests="true"/>

  </system.webServer>

 

</configuration>

 

I have highlighted the relevant pieces. Arguably, the most important configuration piece is the “binding” tag. We specific the “security mode” property as “Transport”. This specifies that we want to use SSL. The other vital piece of  configuration is the “behavior” where we set the “httpsGetEnabled” property to “true”.

This binding and behavior are then imposed on our service under the “services” tag with the bindingConfiguration and behaviorConfiguration properties.

Finally the endpoint tags specify the actual service on which you would like to impose SSL transport encryption.

You will find that there are two endpoint tags. In the endpoint with the “mex” address, it is important to specify that its binding is HTTPS as well using the “mexHttpsBinding” property value.

In the other endpoint tag, I was not able to get the service working by specifying the service URL completely in the “address” property. I had to create a “baseAddresses” tag and specify a base address and then specify the service address relative to the base address in the endpoint.

These settings *almost* do everything I need but they were not enough. I had to go the the project properties and under the “Web” tab, I had to explicitly specify that Local IIS Web Server as the server to be used for testing:

That is it. I am now able to browse to the service URL:

Just like you would with an unencrypted service:

Also, here is the source of the sample service I created.

Display Civil 3D Surface in Web Page with WebGL–part 3

By Daniel Du

This is continuation of Part 2. In part 2, I create a simplest memory repository, in this part, I would like to create a database repository to make my data persistent. I am using Entity Framework code first programming, which makes it very easy to create the database, actually I do not need to write a single SQL clause to manipulate the database, all stuff can be done in C# with Entity Framework. It is also very flexible if I need to change/reflector my model someday.

Firstly I created a context class, named as SurfaceContext, inherited from DbContext:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Web;
using System.Data.Entity;
using TransferDataModel;
 
namespace SurfaceStoreServiceWebRole.Models
{
    public class SurfaceContext : DbContext
    {
        public SurfaceContext()
            : base("name = SurfaceStoreServiceWebRoleContext")
        {
 
        }
 
        public DbSet<TransferTinSurface> TinSurfaces
        {
            get;
            set;
        }
 
        public DbSet<TransferTriangleFace> TriangleFace
        {
            get;
            set;
        }
  
    }
}

Please note that “SurfaceStoreServiceWebRoleContext”  is my connection string, which is defined in web.config:

    <add name="SurfaceStoreServiceWebRoleContext" connectionString=
"Data Source=.\SQLEXPRESS; 
Initial Catalog=SurfaceStoreServiceWebRoleContext; 
Integrated Security=True; MultipleActiveResultSets=True"
      providerName="System.Data.SqlClient" />

Here I am using SQL Express, but it is easy to refer to a cloud based SQL Azure database latter.

Next I created my database repository, implementing the ITinSurfaceRepository interface:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Web;
using System.Data.Entity.Infrastructure;
using TransferDataModel;
 
namespace SurfaceStoreServiceWebRole.Models
{
    public class TinSurfaceDatabaseRepository : ITinSurfaceRepository
    {
        private SurfaceContext db = new SurfaceContext();
 
        public TinSurfaceDatabaseRepository()
        {
            
        }
 
        public IEnumerable<TransferTinSurface> GetAll()
        {
            return db.TinSurfaces.AsEnumerable<TransferTinSurface>();
        }
 
        public TransferTinSurface Get(int id)
        {
            return db.TinSurfaces.Find(id);
 
        }
 
        public TransferTinSurface Add(TransferTinSurface surface)
        {
            if (surface == null)
            {
                throw new ArgumentNullException("surface is null");
            }
            db.TinSurfaces.Add(surface);
            db.SaveChanges();
 
            return surface;
        }
 
        public TransferTinSurface Remove(int id)
        {
            TransferTinSurface surf = Get(id);
            if (surf == null)
            {
                return null;
            }
            db.TinSurfaces.Remove(surf);
 
            try
            {
                db.SaveChanges();
            }
            catch (DbUpdateConcurrencyException)
            {
                return null;
                
            }
 
            return surf;
 
        }
 
        public bool Update(TransferTinSurface updatedSurface)
        {
            var surface = Get(updatedSurface.Id);
            if (surface == null)
            {
                return false;
            }
 
            surface.Name = updatedSurface.Name;
            surface.Description = updatedSurface.Description;
            surface.TriangleFaces = updatedSurface.TriangleFaces;
            try
            {
                db.SaveChanges();
            }
            catch (DbUpdateConcurrencyException)
            {
                return false;
                
            }
 
            return true;
        }
    }
}

For test purpose, I also need some initial data and populate them into database, so here is a sample data class:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Web;
using TransferDataModel;
 
namespace SurfaceStoreServiceWebRole.Models
{
    public static class SampleData
    {
 
        static TransferTriangleFace[] triangleFacesOfSurface1 = new TransferTriangleFace[]
        {
            new TransferTriangleFace{Id = 1,
                P1 = new TransferPoint{X = 10, Y = 0, Z = 0}, 
                P2 = new TransferPoint{X = 0,  Y=10, Z=0}, 
                P3 = new TransferPoint{X=-10, Y=0,Z=0}, 
                Color = "0xff0000"},
            new TransferTriangleFace{Id = 2,
                P1 = new TransferPoint{X = -10, Y = 0,Z = 0}, 
                P2 = new TransferPoint{ X = 0,Y = -10,Z = 0}, 
                P3 = new TransferPoint{X = 10, Y = 0, Z = 0} , 
                Color = "0x00ff00"}
        };
 
        static TransferTriangleFace[] triangleFacesOfSurface2 = new TransferTriangleFace[]
        {
            new TransferTriangleFace{Id = 1,
                P1 = new TransferPoint{X = 10, Y = 10,Z = 0}, 
                P2 = new TransferPoint{X=-10,Y= 0,Z= 0}, 
                P3 = new TransferPoint{X = 15,Y = 0,Z = 0}, 
                Color = "0xff0000"},
            new TransferTriangleFace{Id = 2,
                P1 = new TransferPoint{X = -10,Y = 0,Z = 0}, 
                P2 = new TransferPoint{X = 0,Y = -15, Z = 0}, 
                P3 = new TransferPoint{ X = 15, Y = 0,Z = 0}, 
                Color = "0x00ff00"}
        };
 
        static TransferTinSurface[] sampleTinSurfaces = new TransferTinSurface[]
        {
            new TransferTinSurface { Name = "surface1", 
                Description = "This is the first surface", 
                TriangleFaces = triangleFacesOfSurface1 },
            new TransferTinSurface { Name = "DesignSurface", 
                Description = "This is a design surface", 
                TriangleFaces = triangleFacesOfSurface2 }
        };
 
 
        public static TransferTinSurface[] TinSurfaces
        {
            get
            {
                return sampleTinSurfaces;
            }
        }
 
    }
}

 

Now go to Tools menu—> Library package Manager to open Package Manager Console:

And type command “Enable-Migrations”, please note the intelligence

After enabling migration, a file named as “Configurations.cs” will added into a new folder “Migrations”

Seed method is called after migrating the the latest version by running “Update-Database” command in Package Manager, so I just add my simple code to add some initial data into database context from SampleData class:

namespace SurfaceStoreServiceWebRole.Migrations
{
    using System;
    using System.Data.Entity;
    using System.Data.Entity.Migrations;
    using System.Linq;
    using SurfaceStoreServiceWebRole.Models;
 
    internal sealed class Configuration : DbMigrationsConfiguration<
                SurfaceStoreServiceWebRole.Models.SurfaceContext>
    {
        public Configuration()
        {
            AutomaticMigrationsEnabled = true;
        }
 
        protected override void Seed(SurfaceStoreServiceWebRole.Models.SurfaceContext context)
        {
            //  This method will be called after migrating to the latest version.
 
            //  You can use the DbSet<T>.AddOrUpdate() helper extension method 
            //  to avoid creating duplicate seed data. E.g.
 
            context.TinSurfaces.AddOrUpdate(
                  s => s.Name,
                  SampleData.TinSurfaces);
 
            context.SaveChanges();
 
            
        }
    }
}

 

If you have read part 2 of the serials, you will notice that my first version of data model is named as “TinSurface”, and latter I found it is confusing when I came to Civil 3D plugin programing, so I did some refactoring work. Fortunately, that is easy with Entity Framework Code First programing. I just rename the class name in Visual Studio, the refactor tool of VS2012 ensure to change all relevant code, and run a command in Package Manager:

PM> Add-Migration RefactorClassName
PM> Update-Database

That’s it, all stuffs are done automatically, fantastic! “RefactorClassName” is my migration name, you will noticed that a new class is generated in “Migrations” folder:

After that, I check my database to verify whether my seed data is generated or not. Go Server Explorer and add a Data Connection to .\SQLEXPRESS. Two tables are created and some test data are populated.

 

Next is to switch to the database repository in my TinSurfacesController. Note that calling new TinSurfaceDatabaseRepository() in the controller is not the best design, because it ties the controller to a particular implementation of ITinSurfaceRepository. For a better approach, see Using the Web API Dependency Resolver. I would let it be by now.

        static readonly ITinSurfaceRepository repository =
            new TinSurfaceDatabaseRepository();
        //static readonly ITinSurfaceRepository repository = 
        //    new TinSurfaceMemoryRepository();

 

Okay, with that I am working with a database now. With test of fiddler like part 2, I can retrieve data from database, and save data into database as well.

Display Civil 3D Surface in Web Page with WebGL - part2

By Daniel Du

In part 1 of this serials, I learned how to draw a basic triangular face in a web page with WebGL, and the coordinates of the triangular face are hardcoded.  I would like to get them from a cloud service. I am planning to use Windows Azure to host this service and use SQL Azure to store my surface data. In this post, I will describe how I implement the web service.

With Visual Studio 2010 + Windows Azure SDK, I can create a Windows Azure Cloud Service easily. I selected a ASP.NET MVC 4 as web role. Windows Azure SDK helps me to setup the local emulation environment,  so that I can develop and test my cloud service locally, even without knowing it actually is a cloud service. But debugging such a cloud project on a local cloud emulation environment is pretty slow, and it is not necessary for me at this stage, so I just removed the cloud project, and just work on the ASP.net MVC project. After all, I can add the cloud project and publish it to Windows Azure latter. So currently, I am just working on a ASP.NET MVC 4 web application.

If you have read the [Creating a cloud based viewer] serials of my colleague Philippe Leefsma, you will know that it is possible to create a RESTful web service with WCF. but I would like to try the new ASP.NET Web API in my project. I seems that Web API is much easier for me than WCF in this project, but Web API is not a replacement of WCF for everywhere, this article describe the difference and which one you should pick.

Firstly I will create my models. It is pretty straight forward, code snippet goes as below:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Web;
 
namespace SurfaceStoreServiceWebRole.Models
{
    public class TinSurface 
    {
        
        public int Id { get; set; }
        public string Name { get; set; }
        public string Description { get; set; }
        public virtual ICollection<TriangleFace> TriangleFaces 
        { get; set; }
 
    }
 
 
    public class TriangleFace
    {
        public int Id { get; set; }
        public virtual Point P1 { get; set; }
        public virtual Point P2 { get; set; }
        public virtual Point P3 { get; set; }
        public string Color { get; set; }
 
    }
 
    public class Point
    {
        public double X { get; set; }
        public double Y { get; set; }
        public double Z { get; set; }
    }
}

 

Then add a repository interface and a class to implement this interface. The repository is used to handle the data from various data sources, it can be a database or a simple array in memory for testing, and the service only ask data from the repository, in this way, we can change the backing store without rewriting the service class.

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
 
namespace SurfaceStoreServiceWebRole.Models
{
    interface ITinSurfaceRepository
    {
        IEnumerable<TinSurface> GetAll();
        TinSurface Get(int id);
        TinSurface Add(TinSurface surface);
        TinSurface Remove(int id);
        bool Update(TinSurface surface);
    }
}

Firstly add a simple repository in memory, just for quick test.

using System;
using System.Collections.Generic;
using System.Linq;
using System.Web;
 
namespace SurfaceStoreServiceWebRole.Models
{
    //very simple implementation, no error checking, just for quick test
    public class TinSurfaceMemoryRepository : ITinSurfaceRepository
    {
        List<TinSurface> surfacesStorage = new List<TinSurface>();
        public TinSurfaceMemoryRepository()
        {
            //add some test data
            surfacesStorage.Add(
                new TinSurface
                {
                    Id = 1,
                    Name = "surface1",
                    Description = "this is test surface 1",
                    TriangleFaces = new List<TriangleFace>(){ 
                        new TriangleFace{ 
                            Id = 1, 
                            P1 = new Point{X = 10, Y = 10, Z= 0}, 
                            P2 = new Point{X = -10, Y = 0, Z= 0},
                            P3 = new Point{X = 15, Y = 10, Z= 0},
                            Color = "0xffffff"}
                    }
                });
 
        }
 
        public IEnumerable<TinSurface> GetAll()
        {
            return surfacesStorage.AsEnumerable<TinSurface>();
        }
 
        public TinSurface Get(int id)
        {
            return surfacesStorage.Find(p => p.Id == id);
        }
 
        public TinSurface Add(TinSurface surface)
        {
            surfacesStorage.Add(surface);
            return surface;
        }
 
        public TinSurface Remove(int id)
        {
            TinSurface surf = Get(id);
            surfacesStorage.Remove(surf);
            return surf;
        }
 
        public bool Update(TinSurface surface)
        {
            TinSurface surf = Get(surface.Id);
            if (surf != null)
            {
                surfacesStorage.Remove(surf);
                surfacesStorage.Add(surface);
                return true;
            }
            else
            {
                return false;
            }
        }
    }
}

 

Now it is time to add a controller, controller is an object that handles HTTP request. In Solution Explorer, right-click the the Controllers folder. Select Add and then select Controller. In the Add Controller wizard, name the controller as "TinSurfacesController". In the Template drop-down list, there are many templates to make your work easy. I select Empty API Controller here.  Then click Add. Then add code as below:

using System;
using System.Collections.Generic;
using System.Linq;
using System.Net;
using System.Net.Http;
using System.Web.Http;
using SurfaceStoreServiceWebRole.Models;
 
 
namespace SurfaceStoreServiceWebRole.Controllers
{
    public class TinSurfacesController : ApiController
    {
        //Calling new ProductRepository() in the controller is not the best design, because it ties the controller to a particular implementation of IProductRepository. For a better approach, see Using the Web API Dependency Resolver.
        //http://www.asp.net/web-api/overview/extensibility/using-the-web-api-dependency-resolver
        //static readonly ITinSurfaceRepository repository = 
        //    new TinSurfaceRepository();
        static readonly ITinSurfaceRepository repository = 
            new TinSurfaceMemoryRepository();
 
        //api/TinSurfaces
        public IEnumerable<TinSurface> GetAllTinSurfaces()
        {
            return repository.GetAll();
        }
 
        //api/TinSurfaces/id
        public TinSurface GetTinSurface(int id)
        {
            TinSurface surface = repository.Get(id);
            if (surface == null)
            {
                throw new HttpResponseException(HttpStatusCode.NotFound);
            }
            return surface;
        }
 
        //GET api/TinSurfaces?name=name
        public IEnumerable<TinSurface> GetTinSurfacesByName(string name)
        {
            return repository.GetAll().Where(
                p => string.Equals(p.Name, name, 
                    StringComparison.OrdinalIgnoreCase));
        
        }
 
        //create,POST  api/TinSurfaces
        public HttpResponseMessage PostTinSurface(TinSurface surface)
        {
            surface = repository.Add(surface);
            var response = Request.CreateResponse<TinSurface>(
                HttpStatusCode.Created, surface);
            string uri = Url.Link("DefaultApi",
                                new TinSurface{Id = surface.Id});
            response.Headers.Location = new Uri(uri);
            return response;
        }
 
        //update, PUT api/TinSurfaces/id
        public void PutTinSurface(int id,TinSurface surface)
        {
            surface.Id = id;
            if (!repository.Update(surface))
            {
                throw new HttpResponseException(HttpStatusCode.NotFound); 
            }
 
        }
 
        //delete, DELETE api/TinSurfaces/id
        public void DeleteTinSurface(int id)
        {
            TinSurface surface = repository.Get(id);
            if (surface == null)
            {
                throw new HttpResponseException(HttpStatusCode.NotFound);
            }
            repository.Remove(id);
        }
 
 
    }
}

Here is the table to of example URLs:

Action	HTTP Method	URL
Get all surfaces	GET	api/TinSurfaces
Get one surface by Id	GET	api/TinSurfaces/1
Get one surface by name	GET	api/TinSurfaces?name=surface1
Create a new surface	POST	api/TinSurfaces
Update a surface with new object	PUT	api/TinSurfaces/1
Delete a surface by Id	DELETE	api/TinSurfaces/1

 

Now it is time to run and test this service. I use Fiddler to test the REST service. Switch Composer tab ,input the url http://localhost:3297/api/tinsurfaces and select “GET” method, then click “Execute” button:

The result HTTP code is 200, that means all work fine, double click the item to check the output in Inspector tab, I saw the result in JSON format.

 

It works pretty good! Okay, I would like to stop here and will continue work on this project in next post.

Cloud Security: Securing with SSL – Certificates

Gopinath Taget

As you move your traditional desktop based applications to the cloud, you naturally begin worrying about security.

Typically, you want to make sure that no unauthorized person or entity gains access to your cloud based applications. There are several things you can do to build robust security around your cloud based applications. One important step is to make your web services secure by using SSL (Secure Socket Layer).

SSL is a method of encrypting the communication between a client application (like a desktop application, mobile application or a web browser) and the web server application running somewhere on the cloud. You might recognize this as the HTTPS URLs that you see from time to time. HTTPS URLs are secure web sites and services that use SSL.

Given that there are many server technologies like ASP .NET, Apache and Java based services, there are several ways to implement SSL security into your web-service.

Over the next several posts, I will talk about various technology components that go into implementing an SSL based web service. In this post I will talk about certificates.

For SSL to work, a verifiable trusted certificate is the key. A certificate provides information on who this certificate belongs to among other things. But this is not enough. You need to be able to trust that certificate. If I tell you that a digital certificate I send to you is trust worthy, you might choose to believe me or not (depending on how well you know me). Given that there are millions of websites, you will not know everyone and it would be impossible to verify each individual certificate before using their websites or services.

To solve this problem, we have Certificate Authorities (CA). If I am a web service provider and you are a client who intends to use my service, we do not need to explicitly know each other to trust each other. If both of us trust a Certificate Authority, than we implicitly trust each other (at least, that is how the system works on the internet now). So when a certificate Authority issues a certificate to a web service vendor, the client which uses that vendor service requests for the vendor certificate and once it is obtained, the client verifies if it is issued by a certificate authority.

This verification is performed by validating the vendor certificate against a certificate belonging to the Certificate Authority itself. This certificate (of the Certificate Authority) is called the Root Certificate. When you install a web browser on your computer, the root certificates of the most popular certificates are automatically installed in a secure location on your machine. You will find information on popular Certificate Authorities by just googling for them.

Now, does this mean you need a certificate from a Certificate Authority (Most of the CAs charge money for issuing certificates) to start building SSL based web services? Not necessarily. You can create certificates for yourself (both root and regular) using the free microsoft windows command-line tool makecert. Since you know yourself ( you do know yourself don’t you?) you can trust your root certificate and any certificate issued with that root certificate (unless you do not trust yourself, which could happen some times). Here are a couple of web articles (first, second) that show how to use makecert to create root and regular certificates.

That is it for now. In my next blog post, I will talk about using these certificates and configuring a WCF Web service to implement SSL.