Geekswithblogs.net

Get Detailed Build Test Results using the TFS API

Tarun Arora — Wed, 16 May 2012 21:37:00 GMT

It’s a Friday evening and you have just checked in that last bit of code, you are waiting for the build to go all green so that you could call it a day. Just then the build summary page comes back red on the test results section. The build summary page tells you that 1 of the test namely HelloWorld.Tests.UnitTest1.Sum_TwoIntNumbers_IncorrectResult_TestMethod has failed…

To know why the test has failed, you have no option but to click on View Test Results link which in turn downloads the trx file from the server so that you can see the error message and possibly identify the root cause of failure from the description error message. If the test list contains more than a 1000 tests you will notice that it takes a lot of time in downloading the test results locally. That’s probably not what you would want to do on a Friday evening…

In this blog post I’ll show you how to use the TFS API to write a simple utility to pull down the test result details programmatically and a bonus power tool for those who read the complete post, let’s get started…

A peek at what we will get to…

1. Connecting to TFS using the API

If you are new to the TFS API, refer to this blog post on getting started with the Visual Studio SDK.

The below snippet will help present a connect to tfs pop up to the user

private TfsTeamProjectCollection _tfs;
private string _selectedTeamProject;

public void ConnectToTfs()
{
       TeamProjectPicker tfsPP = new TeamProjectPicker(TeamProjectPickerMode.SingleProject, false);
       tfsPP.ShowDialog();
       this._tfs = tfsPP.SelectedTeamProjectCollection;
       this._selectedTeamProject = tfsPP.SelectedProjects[0].Name;
}

2. Get All Build Definitions for the selected Team Project

By using the IBuildService you can get access to the QueryBuildDefinition method which takes team project and returns the build definitions associated to the team project.

private IBuildDefinition[] GetAllBuildDefinitionsFromTheTeamProject(_selectTeamProject)
{
      _bs = _tfs.GetService<IBuildServer>();
      return _bs.QueryBuildDefinitions(_selectedTeamProject);
}

3. Get All Builds Associated to the Build Definition

Get All Builds in the build definition for specific build Definition specifications programmatically, this helps you narrow down the search and get the selected few filtered results, like in the snippet below, we specify the maximum number of builds to return, what build quality builds to return, etc.

// Get All Builds for the selected build definition 
private void cbBuildDef_SelectionChanged(object sender, SelectionChangedEventArgs e)
{
       var bdef = (((ComboBox)sender).SelectedItem) as IBuildDefinition;
       var def = _bs.CreateBuildDetailSpec(_selectedTeamProject);
       def.MaxBuildsPerDefinition = 10;
       def.QueryOrder = BuildQueryOrder.FinishTimeDescending;
       def.DefinitionSpec.Name = bdef.Name;
       def.Status = BuildStatus.All;
       var builds = _bs.QueryBuilds(def).Builds;

        cbBuild.ItemsSource = _bs.QueryBuilds(def).Builds;
        cbBuild.DisplayMemberPath = "BuildNumber";
        cbBuild.SelectedValuePath = "Uri";
        cbBuild.SelectedIndex = 0;
}

4. Get all the Test Results associated to a Build

I’ll be using the ITestManagementService to pull all test runs associated to a build. As you can see below by using the QueryResultsByOutcome I can query for tests that passed or failed or tests that errored. You could also use the QueryResultsByOwner to get test results by individual tfs users.

private void GetTestResult(Uri buildUri)
{
       _tms = _tfs.GetService<ITestManagementService>();
       var testRuns = _tms.GetTeamProject(_selectedTeamProject).TestRuns.ByBuild(buildUri);

        foreach (var testRun in testRuns)
        {
             lstTestRunDetails.Items.Add(string.Format("{0}", testRun.Title));
             lstTestRunDetails.Items.Add(string.Format("TestRunId: {0} | TestPlanId: {1}", testRun.Id, testRun.TestPlanId));
             lstTestRunDetails.Items.Add(string.Format("TestSettingsId: {0} | TestEnvironmentId {1} ", testRun.TestSettingsId, testRun.TestEnvironmentId));

             var totalTests = testRun.Statistics.TotalTests;

              foreach (var et in testRun.QueryResultsByOutcome(TestOutcome.Error))
              {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", et.Outcome, et.TestCaseTitle, et.ErrorMessage));
              }

              foreach (var tp in testRun.QueryResultsByOutcome(TestOutcome.Passed))
              {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} ", tp.Outcome, tp.TestCaseTitle));
               }

                foreach (var tf in testRun.QueryResultsByOutcome(TestOutcome.Failed))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", tf.Outcome, tf.TestCaseTitle, tf.ErrorMessage));
                }

                foreach (var tw in testRun.QueryResultsByOutcome(TestOutcome.Warning))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", tw.Outcome, tw.TestCaseTitle, tw.ErrorMessage));
                }

                foreach (var ta in testRun.QueryResultsByOutcome(TestOutcome.Aborted))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", ta.Outcome, ta.TestCaseTitle, ta.ErrorMessage));
                }

                foreach (var tb in testRun.QueryResultsByOutcome(TestOutcome.Blocked))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", tb.Outcome, tb.TestCaseTitle, tb.ErrorMessage));
                }

                foreach (var ti in testRun.QueryResultsByOutcome(TestOutcome.Inconclusive))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", ti.Outcome, ti.TestCaseTitle, ti.ErrorMessage));
                }

                foreach (var to in testRun.QueryResultsByOutcome(TestOutcome.Timeout))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", to.Outcome, to.TestCaseTitle, to.ErrorMessage));
                }

                // Get the test results by user by passing in the Test Foundation Identity
                // testRun.QueryResultsByOwner(TeamFoundationIdentity);
            }

            if(testRuns.Count() == 0)
                lstTestRunDetails.Items.Add("No Test Results have been associated with the selected build");
 }

5. Putting it all together

Download the sample application – wpf, .net 4, you need to have vs sdk installed on your machine to run it.

using System;
using System.Collections.Generic;
using System.Linq;
using System.Text;
using System.Windows;
using System.Windows.Controls;
using System.Windows.Data;
using System.Windows.Documents;
using System.Windows.Input;
using System.Windows.Media;
using System.Windows.Media.Imaging;
using System.Windows.Navigation;
using System.Windows.Shapes;
using Microsoft.TeamFoundation.Build.Client;
using Microsoft.TeamFoundation.Client;
using Microsoft.TeamFoundation.TestManagement.Client;
using Microsoft.TeamFoundation.VersionControl.Client;

namespace TfsApiGetTestResultsAndDetailsByStatus
{
    /// <summary>
    /// Interaction logic for MainWindow.xaml
    /// </summary>
    public partial class MainWindow : Window
    {
        public MainWindow()
        {
            InitializeComponent();
        }

        private TfsTeamProjectCollection _tfs;
        private string _selectedTeamProject;
        private IBuildServer _bs;
        private VersionControlServer _vcs;
        private ITestManagementService _tms;

        public void ConnectToTfs()
        {
            TeamProjectPicker tfsPP = new TeamProjectPicker(TeamProjectPickerMode.SingleProject, false);
            tfsPP.ShowDialog();
            this._tfs = tfsPP.SelectedTeamProjectCollection;
            this._selectedTeamProject = tfsPP.SelectedProjects[0].Name;
        }

        private void btnConnect_Click(object sender, RoutedEventArgs e)
        {
            ConnectToTfs();
            _vcs = _tfs.GetService<VersionControlServer>();
            cbBuildDef.ItemsSource = GetAllBuildDefinitionsFromTheTeamProject();
            cbBuildDef.DisplayMemberPath = "Name";
            cbBuildDef.SelectedValuePath = "Uri";
            cbBuildDef.SelectedIndex = 0;
        }

        private IBuildDefinition[] GetAllBuildDefinitionsFromTheTeamProject()
        {
            _bs = _tfs.GetService<IBuildServer>();
            return _bs.QueryBuildDefinitions(_selectedTeamProject);
        }

        // Get All Builds for the selected build definition 
        private void cbBuildDef_SelectionChanged(object sender, SelectionChangedEventArgs e)
        {
            var bdef = (((ComboBox)sender).SelectedItem) as IBuildDefinition;
            var def = _bs.CreateBuildDetailSpec(_selectedTeamProject);
            def.MaxBuildsPerDefinition = 10;
            def.QueryOrder = BuildQueryOrder.FinishTimeDescending;
            def.DefinitionSpec.Name = bdef.Name;
            def.Status = BuildStatus.All;
            var builds = _bs.QueryBuilds(def).Builds;

            cbBuild.ItemsSource = _bs.QueryBuilds(def).Builds;
            cbBuild.DisplayMemberPath = "BuildNumber";
            cbBuild.SelectedValuePath = "Uri";
            cbBuild.SelectedIndex = 0;
        }

        private void cbBuild_SelectionChanged(object sender, SelectionChangedEventArgs e)
        {
            var build = (((ComboBox)sender).SelectedItem) as IBuildDetail;
            if (build == null && ((ComboBox)sender).Items.Count != 0)
            {
                build = ((ComboBox)sender).Items[0] as IBuildDetail;
                var def = _bs.CreateBuildDetailSpec(_selectedTeamProject);
                GetTestResult(build.Uri);
            }
        }

        private void GetTestResult(Uri buildUri)
        {
            _tms = _tfs.GetService<ITestManagementService>();
            var testRuns = _tms.GetTeamProject(_selectedTeamProject).TestRuns.ByBuild(buildUri);

            foreach (var testRun in testRuns)
            {
                lstTestRunDetails.Items.Add(string.Format("{0}", testRun.Title));
                lstTestRunDetails.Items.Add(string.Format("TestRunId: {0} | TestPlanId: {1}", testRun.Id, testRun.TestPlanId));
                lstTestRunDetails.Items.Add(string.Format("TestSettingsId: {0} | TestEnvironmentId {1} ", testRun.TestSettingsId, testRun.TestEnvironmentId));

                var totalTests = testRun.Statistics.TotalTests;

                foreach (var et in testRun.QueryResultsByOutcome(TestOutcome.Error))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", et.Outcome, et.TestCaseTitle, et.ErrorMessage));
                }

                foreach (var tp in testRun.QueryResultsByOutcome(TestOutcome.Passed))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} ", tp.Outcome, tp.TestCaseTitle));
                }

                foreach (var tf in testRun.QueryResultsByOutcome(TestOutcome.Failed))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", tf.Outcome, tf.TestCaseTitle, tf.ErrorMessage));
                }

                foreach (var tw in testRun.QueryResultsByOutcome(TestOutcome.Warning))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", tw.Outcome, tw.TestCaseTitle, tw.ErrorMessage));
                }

                foreach (var ta in testRun.QueryResultsByOutcome(TestOutcome.Aborted))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", ta.Outcome, ta.TestCaseTitle, ta.ErrorMessage));
                }

                foreach (var tb in testRun.QueryResultsByOutcome(TestOutcome.Blocked))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", tb.Outcome, tb.TestCaseTitle, tb.ErrorMessage));
                }

                foreach (var ti in testRun.QueryResultsByOutcome(TestOutcome.Inconclusive))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", ti.Outcome, ti.TestCaseTitle, ti.ErrorMessage));
                }

                foreach (var to in testRun.QueryResultsByOutcome(TestOutcome.Timeout))
                {
                    lstTestRunDetails.Items.Add(string.Format("{0}: {1} - {2}", to.Outcome, to.TestCaseTitle, to.ErrorMessage));
                }

                // Get the test results by user by passing in the Test Foundation Identity
                // testRun.QueryResultsByOwner(TeamFoundationIdentity);
            }

            if(testRuns.Count() == 0)
                lstTestRunDetails.Items.Add("No Test Results have been associated with the selected build");
        }
    }
}

6. And the Bonus Power tool…

Thank you for reading this far. If you haven’t already download the Community TFS Build Manager extension from the visual studio extension gallery do so now. The Community TFS Build Manager (TBM) is an open source build management solution developed by the ALM rangers, the build manager simplifies the management of builds in medium to large Team Foundation Server environments. Some common limitations you may run into while dealing with builds through the Visual Studio Build Explorer are not being able to perform bulk operations on builds, create a relationship diagram of the build controller/agents, managing build process templates, etc. The community TFS build manager address that and many other such gaps you may encounter while managing builds through the visual studio build explorer. You can grab the source and read more about the project on CodePlex.

Recently a new feature namely Build Notes has been added to the community TFS Build Manager that let’s you generate an ms word report of the build summary. A sample attached below… Look at the test result section, the trx file is parsed for you to get you a test result list filtered by test result outcome, where failed test also carries the error message. Watch this video or read this post here on what you can achieve with the build notes.

Thank you for taking the time out and reading this blog post. If you enjoyed the post, remember to subscribe to http://feeds.feedburner.com/TarunArora.

Check out the other posts on cool tools using TFS API.

SQL Server 2008 R2 setup fails on a Windows Virtual PC (VPC) with error code 2337

Daniel Hester — Wed, 16 May 2012 13:17:04 GMT

When installing SQL Server 2008 R2 (Developer, Standard, or Enterprise) within a Virtual PC you may suffer the following problem:

“The following error has occurred: The installer has encountered an unexpected error installing this package. This may indicate a problem with this package. The error code is 2337.”

The error occurs in the following package: sql_engine_core_shared

It turns out that this problem is with the DVD/ISO Mounting Feature in Windows Virtual PC. To workaround the problem you need to expand the ISO with WinRAR – not WinZip and not by copying the contents of the mounted ISO to the VPC’s hard drive. You need to use WinRAR to extract the ISO; extracting the contents of the ISO with anything else just won’t solve the problem.

Microsoft have acknowledged the problem here:

https://connect.microsoft.com/SQLServer/feedback/details/550502/r2-rc0-install-problem-on-virtual-pc-2007

WinRAR can be found here:

http://www.win-rar.com/download.html

Self Host Web Api's for Integration Testing

Justin Hoffman — Wed, 16 May 2012 09:27:52 GMT

Recently I had a need to write some tests making Web Api calls that would be deployable to the build server without any special configuration.What I did was utilize the Self Hosting feature in the Asp.NetWebApi. First, I created a simple Api controller .

// GET /api/values
    public class ValuesController : ApiController
    {
        public SomeObjects GetByName(string firstName, string lastName)
        {
            var someObjects= new List<SomeObject>()
                               {
                                    new SomeObject
                                    {
                                        FirstName = "Tom",
                                        LastName = "Jones",
                                        Id= 1213
                                    },
                                     new SomeObject
                                    {
                                        FirstName = "Jane",
                                        LastName = "Doe",
                                        Id= 1523
                                    },
                                     new SomeObject
                                    {
                                        FirstName = "John",
                                        LastName = "Doe",
                                        Id= 3123
                                    }
                               };

            return someObjects.Where(x => x.FirstName == firstName && x.LastName == lastName).SingleOrDefault();
        }
    }

Next, I created a simple utility class to execute tests against the hosted service

public static class TestHostingUtil
    {
        public static void ExecuteHostedTest(Action<Task> continuationAction, string routeTemplate = null)
        {
            var routeTempl = "api/{controller}/{id}";
            if (!string.IsNullOrEmpty(routeTemplate))
            {
                routeTempl = routeTemplate;
            }

            var serverConfig = new HttpSelfHostConfiguration("http://localhost:8080/");
            serverConfig.Routes.MapHttpRoute("default", routeTempl, new { id = RouteParameter.Optional, folder = "TestApiControllers" });

            var server = new HttpSelfHostServer(serverConfig);
            
            server.OpenAsync().ContinueWith(continuationAction).Wait();
            server.CloseAsync();
        }
    }

Note the "folder" param in the HttpRoute registration. In the test project this is the folder under which I added the Api controllers, if they were not in a seperate folder you would leave this blank.

Finally, the usuage for your tests:

[TestMethod]
        public void Should_Perform_Get()
        {
            TestHostingUtil.ExecuteHostedTest(task =>
                                  {

                                      // Arrange
                                      var webUtil=
                                          new WebUtil
                                              <SomeObject, SomeObject>(
                                              "http://localhost:8080/api/values?firstName=Tom&lastName=Jones");

                                      // Act
                                      var result = webUtil.Get();

                                      // Assert
                                      result.FirstName.Should().Be("Tom");
                                  });
        }

Microsoft Test Manager–Test Cases Not in a Test Plan

Bob Hardister — Wed, 16 May 2012 06:21:13 GMT

Visual Studio and Team Foundation Server (TFS) 2010 provided support for system testing. Test cases were added as a work item type. But test plans and suites are not work items and this greatly limits the visibility into the management of test artifacts. For example, there is no easy way out-of-the-box to find test cases that don’t belong to a test plan.

Here’s the SQL that creates a list of test cases providing the test case:

ID
Title
State
Assigned To
Area Path
Iteration Path
Plan and Suite Path

You can use the output to identify what test cases are in what plan and suite; and more to the point, which test cases are not in a plan.

Using the Amrein SQL Viewer Web Part you can easily throw this up as a web part on the project portal site where users can globally access, sort and filter the data.

SELECT tco.System_Id as ID
,tco.System_Title as Title
,tco.System_State as State
,ISNULL(p.Name,'null') as Assigned
,tco.AreaPath as Area
,tco.IterationPath as Iteration
,ISNULL(s.SuitePath,'null') as 'Plan'
FROM Tfs_01Warehouse.dbo.vDimTestCaseOverlay tco inner join Tfs_01Warehouse.dbo.DimPerson p
ON tco.System_AssignedTo__PersonSK=p.PersonSK left outer join Tfs_{collection DB}.dbo.tbl_SuiteEntry se ON tco.System_Id=se.TestCaseId left outer join Tfs_01Warehouse.dbo.vDimTestSuiteOverlay s
ON se.SuiteId=s.TestSuiteID
WHERE tco.TeamProjectSK={#} /* team project */
ORDER BY tco.System_Id DESC

Mobile responsive site design in Expression Web

ihaynes — Wed, 16 May 2012 08:35:12 GMT

Although it doesn’t have any specific tools to help responsive or mobile site design this is still possible in Expression Web and I’ve created a number of snippets to help the process.

HTML5 and CSS3 are the key to responsive design, so the snippets cover the new HTML5 layout elements, form input types and microdata tags, and the CSS3 snippets cover box-radius, box-shadow, gradients etc. The mobile snippets include the CSS3 media queries, mobile meta tags and the responsd.js shim that enables older non-HTML5/CSS3 browsers to use the media queries.

The snippets are downloadable from the Expression Gallery:

http://gallery.expression.microsoft.com/HTML5Snippets

http://gallery.expression.microsoft.com/CSS3Snippets

http://gallery.expression.microsoft.com/MobileSnippets

I’ve also created a sample mobile-friendly responsive template for EW.

http://gallery.expression.microsoft.com/MobileTemplate

One caveat about using the CSS3 media queries with a mobile-first layout (the preferred way to use them) is that the Design view in Expression Web doesn’t recognise the queries and therefore displays the mobile view rather than the appropriate wide screen view. You may therefore have to reply more on browser preview when laying out pages.

SharePoint Saturday NYC Update from the Trenches

MOSSLover — Tue, 15 May 2012 23:49:04 GMT

So as you know I am yet again involved in SPS NYC with my awesome teammates, Greg Hurlman, Jason Gallicchio, Tasha Scott, and Tom Daly. The speakers have finally been picked and a first draft schedule has been posted at http://www.sharepointsaturday.org/ny. We are also still looking for sponsors if you guys are interested you can also click on the link to grab the form, sign it, and send it to sponsors@spsnyc.com. We are also looking for volunteers and you can fill out the form at the same place listed above. We are offering a t-shirt and a gift card for those who dedicate a couple hours of service to the event. I believe this year will be the best event yet. I am also considering wearing a Viking Helmut with horns if at least one other person will wear a similar item. I would consider dressing up if I could get another person to dress up, but I think that’s asking for too much. Also, I need to run around and help the others put out fires, so I’m not so sure full on dress attire would work out.

Anyway, registration is not yet posted for those of you guys who wish to attend. We might post it more towards the end of June. I am not yet sure what the decision was on the exact date, but normally we post about 6 weeks prior to the event. I hope to see you all this year at the event. Yet again my dad is going to show up, so stop by, attend a few sessions, and say hi to everyone.

Technorati Tags: SPS NYC,SharePoint Saturday,SharePoint

Integrating SharePoint with Fruity Products

MOSSLover — Tue, 15 May 2012 23:42:17 GMT

So as you can see this blog has not been the best in the world lately. Life changes and burn out are killing me lately. Don’t get me wrong I love my job. I love writing things in Visual Studio, in fact this week I got to work with MVC3 for the first time ever. I am completely hooked on MVC3 and will further explore that in the upcoming months. I have entertaining a completely different idea, because I think it would be fun and interesting. The downside is it involves me buying an Apple computer.

I have been fighting buying a Macbook Pro for ages. I love building Windows PCs from scratch. I love knowing that I built the computer that I am using. I guess it stems back from my grandparents teaching me that hard work leads to a more fulfilling accomplishment. Anyway, I am not switching over to the darkside. I just built myself a gaming PC (core i7 process, 24 gb of ram-32 when I RMA the last two chips, 4 hard drives – main being an SSD, and an NVIDIA GTX 570 card). I just want to branch out and learn some new tech, so I figure if I have some spare time (which is like asking for a unicorn), I will start posting more about my new development. Anyway, I am waiting for the WWDC and then I am going to decide on what flavor of Macbook Pro to purchase. It turns out a Macbook Air won’t get me what I want programmatically the processor might not be quite up to par for hardcore development. I wouldn’t want to limit myself to anything softcore right?

So check back here periodically or on twitter. I’m sure I will post something about my endeavors. In the meanwhile I am still working on my personal life and I am happy to say I have friends in the area. For the first time in a very long time some of them are actually female. As you all know my friends for the longest time were members of the SharePoint and .Net Community mainly and of the male breed. So things are working out. I still love every single day in NYC. It is the first city I can 100% say I belong in. If I have my way I will stay in the metro area for a long time. Anyway, I hope you all are having a good time and goodnight.

Technorati Tags: SharePoint and Apple,Microsoft and Apple,Apple

Let’s MEET Windows Azure

Shaun — Wed, 16 May 2012 00:37:18 GMT

Microsoft had just released a new website http://www.meetwindowsazure.com/ which announced an outstanding event named “MEET Windows Azure”. There will be some awesome new features and releases being published around the Windows Azure Platform, the cloud computing platform and cloud-based technologies Microsoft offers. From what I know, this might be the biggest innovation in Windows Azure till now.

The event will be hosted at 1pm PDT, June 7th at the Madrone Studio in San Francisco with limited sites. But it will opened to everyone who is interested in through the live online. If you are willing to join the live stream, like me, go to http://register.meetwindowsazure.com/ to register as soon as possible.

Hope this helps,

Shaun

All documents and related graphics, codes are provided "AS IS" without warranty of any kind.
Copyright © Shaun Ziyan Xu. This work is licensed under the Creative Commons License.

IQueryable Dynamic Expressions

Jon Canning — Tue, 15 May 2012 19:19:42 GMT

public static class QueryableExtensions
    {
        public static IQueryable<T> Filter<T>(this IQueryable<T> queryable, IEnumerable<KeyValuePair<string, string>> filters) where T : class
        {
            foreach (var filter in filters)
            {
                var propertyName = typeof(T).GetProperties().Select(x => x.Name).SingleOrDefault(x => x.ToLower() == filter.Key.ToLower());
                if (propertyName.IsNullOrWhiteSpace()) continue;
                var parameterExpression = Expression.Parameter(typeof(T));
                var propertyExpression = Expression.Property(parameterExpression, propertyName);
                var toStringCallExpression = Expression.Call(propertyExpression, typeof(T).GetMethod("ToString"));
                var toLowerCallExpression = Expression.Call(toStringCallExpression, typeof(string).GetMethod("ToLower", Type.EmptyTypes));
                var constantExpression = Expression.Constant(filter.Value.ToLower());
                var equalExpression = Expression.Equal(toLowerCallExpression, constantExpression);
                var methodCallExpression = Expression.Call(typeof(Queryable), "Where", new[] { typeof(T) }, queryable.Expression, Expression.Lambda<Func<T, bool>>(equalExpression, new[] { parameterExpression }));
                queryable = queryable.Provider.CreateQuery(methodCallExpression) as IQueryable<T>;
            }
            return queryable;
        }
    }

Old SQL, Top and a Variable

Bunch — Tue, 15 May 2012 11:48:32 GMT

I came across this the other day. Using SQL 2000 I could not get a Top statement to work with a variable. I had forgotten that only with SQL 2005 and higher could you do this trick. I found a good way around it, you can read about it here. Anyway this works on SQL 2000:

declare @top int
set @top = 5

BEGIN
set rowcount @top
select Some_ID
from tblStuff
set rowcount 0
END

The code above would return the first five records from tblStuff. For any of the newer SQL Server version you can simply use the @top variable in Top like:

select top (@top) Some_ID
from tblStuff

Tags: SQL

Sass vs Less

Renso — Tue, 15 May 2012 08:42:24 GMT

I know there are afew places where you can check out the differences between Sass or Less, but I thought it may be helpful for others to know why I chose one over the other.

Here are some of my sources:

http://sass-lang.com/
http://www.dotlesscss.org/
http://stackoverflow.com/questions/8411066/less-vs-sass-vs
http://coding.smashingmagazine.com/2011/09/09/an-introduction-to-less-and-comparison-to-sass/

Since my main decision was affected by my IDE I need to mention my development environment:

Visual Studio 2010 with Mindscape's Web Workbench

In short I chose Sass (Syntactically Awesome Style Sheet). My main reason was that when you save the file in VS, it automatically regenerates the CSS file for SCSS on the fly, so you are able to see right away what CSS is being generated and can run your app right away. I am not in the favor of Less in this regard, as although you can specify set of handlers etc to interpret the less file on the fly, I would never use it to interpret on the fly, especially not on production servers.

Also the Ruby team now ships officially with CoffeeScript and Sass by default and will be included by default in the gems when you deploy your Ruby app. Maybe another reason to use it if you're in the Rails framework.

Any comments very welcome as I am am not sure if my assumptions and assertions are true.

Passing in report parameters to TFS 2010 OLAP Web Parts on the Team Portal

Bob Hardister — Tue, 15 May 2012 06:48:00 GMT

There is some helpful information from Ewald Hofman on passing in parameters to the TFS SQL Server Analysis Services OLAP reports.

But, the 2010 schema has changed. You can use the SQL Server Mgmgt Studio to access the metadata and query the tables to seee the new schema and record values.

As an example of the Task Burndown report on the TFS 2010 Project Portal, here's the URL that worked in a browser:

http://{your tfs server}/Reports/Pages/Report.aspx?ItemPath=%2fTfsReports%2fDefaultCollection%2fCcMsd1.0%2fDashboards%2fBurndown&rc%3atoolbar=false&rs%3aCommand=Render&IterationParam=[IterationName].[ParentIterationSK].%26[{the IterationSK value}]

Here's the URL that was accepted in the web part:

/sites/DefaultCollection/CcMsd1.0/_layouts/TfsRedirect.aspx?tf:Type=Report&tf:ReportName=Dashboards%2fBurndown&tf:ShowToolbar=0&Width=381pt&Height= 180pt&tf:Command=Render&tf:Parameters=IterationParam=[IterationName].[ParentIterationSK].%26[17]

To-date though, even though the report processes without error, the passed in parameters don't make any difference/change in the report. So, I gave up on this approach and was somewhat successful with this approach (see my subsequent blog article).

How I Customized Report Parameters for Team Foundation Server 2010

Bob Hardister — Tue, 15 May 2012 06:46:27 GMT

My goal was to have the burndown chart on the team project portal main page display the current iteration.

The default report displays the tasks for the entire team project starting 5 days before the current date and ending 25 days after the current date. FRUSTRATING, to say the least! I mean, what a useless report! Who's Agile book did they get that one from?

What I wanted was to see the burndown for the current iteration. That is, tasks for that iteration only and that iteration starting on a fixed start date and stopping on a fixed finish date.

First, I tried to implement this by passing in the parameters in the web part link url. I talked to about 4 different Microsoft people at PDC; specialist in VS ALM, SSRS, SharePoint. No one knew how to do this. See my original article on this here.

I have yet to be successful on the first approach. Which seems the most obvious way to approach it. I hope that this secret makes its way into the TFS community at some point. No wonder so few use TFS SSRS reporting! Can you sense the frustration?

Then, I came across at article by John Socha-Leialoha on Customizing Report Parameters for Team Foundation Server 2010. This article describes how I used his approach to be successful.

First, save a copy of the report definition so you can restore it if things get messed up.
To access the report definition, from the team project main page, select Reports from the left column and then navigate through the SQl Server Reporting Services (SSRS) location hierarchy to find the report. Click on it to open it. Go to the Properties tab and select the General panel. Click Edit to save a copy of the report definition file. You will need to click the Update button to browse, select and then apply the file to restore the definition
Now, to the point: select the Properties panel and set the iteration path on the report parameters page as John's blog specifies. Note: the parameter string for iteration path [Work Item].[Iteration Hierarchy].[All] must be changed to [Work Item].[Iteration Hierarchy].[Iteration1].&[{ParentIterationSK}]&[{IterationSK}]. I had to use the MS SQL Server Mgmgt Studio (SSMS) to find out what these were. I connected to my TFS server using the SSMS, opened the Tfs_Warehouse Database and queried the rows in the dbo.DimIteration table. From here you can see what the SK values are based on the IterationPath you are interested in. Note2: I don't know what "[Iteration1]" means for sure. It may have to do with the depth of the IterationPath Hierarchy. In my case, I'm only one level down from the root of the iteration path. So, in this case, I must use "[Iteration1]" regardless of the real iteration I want to report on. So, if the full parameter for my first iteration is: [Work Item].[Iteration Hierarchy].[Iteration1].&[14]&[17] and the SK for my second iteration is 18, my parameter for my project's second iteration looks like this: [Work Item].[Iteration Hierarchy].[Iteration1].&[14]&[18]
Next, click the "Override Default" for StartDateParam and EndDateParam. Note: I think this is permanent for the active defintion. I think you can restore the original defintion to reset the default. Type in the start and end date you want (mm/dd/yyyy). The time will default to 12 AM, but you can change this after you have applied the first explicit date values.
Now view and display the report. You should see the report you want :)
Next, rename the report to reflect the iteration it's for
Now create the linked reports as described by John in his blog
Finally, you can easily modify the web part link url to containt the correct report name.

UPDATE:
In subsequent work I confirmed the syntax of the IterationParm string. For example, take -

[Work Item].[Iteration Hierarchy].[Iteration1].&[14]&[18]

The [Iteration1] node does prescribe the depth of the hierarchy. [Iteration1] means root to 1st level. [Iteration2] means root to 1st level to 2nd level] and so forth. The node IDs correspond to the nodes at each level.

So [Work Item].[Iteration Hierarchy].[Iteration4].&[14]&[18]&[54]&[72] specifies an iteration path node at the root>1st>2nd>3rd level.

You can obtain the SSRS report properties IterationParm value using SQL Server Management Studio. Connect using SQL Server Analysis Services and create an MDX query of the work item.iteration hiearchy path node you want to report on. See this great article by John Socha-Leiloha for more details.

Good hunting!

Using a TFS 2010 custom bug work item definition with the MTLM Test Runner

Bob Hardister — Tue, 15 May 2012 06:44:29 GMT

We don’t use the term “Bug,” rather we use “Software Action Request” or “SAR.” I renamed the Bug WI definition from Bug to SAR. I added a few custom fields to our SAR WI definition, but all the Microsoft fields are used and unchanged.

What I want is for the MTLM client to be configured to open a SAR rather than a Bug. In other words, out-of-the-box when a user is manually executing a test case with the MTLM Test Runner client and one of the steps fail, the user clicks a button to create a Bug or link to an existing Bug.

The answer was provided by Rubel Singal in this blog article. It works! Thank you Rubel!

Happy trails!

Updating an incorrect reason code after a state transition

Bob Hardister — Tue, 15 May 2012 06:32:11 GMT

When using TFS work items you may find that when you transition from one state to another that you selected the wrong reason code, or you found out shortly afterwards that the reason code you selected is incorrect.

Well, you may be surprised to find that reason codes (System.Reason) cannot be updated within the same state.

After some looking into, I came to understand the intent of reason code is to document the "reason" the state changed. As opposed to indicate within state changes. However, it was also clear that no out-of-the-box provision is made for correcting the reason code when the wrong one was used. I did find two different approaches that used a custom field:

One replaces System.Reason with a custom field. Another uses a custom field that gets populated from System.Reason and can then be updated.

I appreciate the approach of using a custom field, as shown in the article. However, wanting to minimize custom code, I took the following approach:

Using the Process Editor:

I simply created an "Update [state] Reason Code" state
The transition reason to this state is "Current reason code is incorrect."
The transition back provides the list of reason codes to chose from

This was easy and leverages the existing fields and functionality. It also provides a nice, concise history and allows the application of permissions to limit who can do this. I found I didn't need to use this on most of my states because many of them can already go back to the immediately prior state and so it amounts to the same thing. Using it on the initial state (Submitted) and Close will be very helpful.

Azure Presentation at VS Live!

Steve Michelotti — Tue, 15 May 2012 06:20:14 GMT

Tomorrow I will be presenting “Building Scalable Apps in Windows Azure” at VS Live in New York City. I will be co-presenting with my colleague Vishwas Lele of Azure fame. Should be a great event!

"Read It Later Pro" has been ruined into Pocket, but RIL Free is still usable.

Michael Freidgeim — Tue, 15 May 2012 11:13:17 GMT

I loved RIL Pro for iPad and used it every day, but about a month ago I was forced to upgrade to new version, named Pocket.

The new version has so many problems, that it's almost unusable.

The main issue is that iPad app doesn't show many of the saved links.

I've posted my opinion that it would be better to have Pocket as a new separate application and recommend users to install it side by side with RIL Pro and try before replace at Pocket forum APR 20, 2012

http://help.getpocket.com/customer/portal/questions/290569-auto-update-to-pocket-wasn-t-a-god-idea .It's still unanswered.

I found that there a many other threads that report different issues with a new version and I've posted a few comments, supporting suggestions to fix the issues, that were broken by the Pocket.(e.g. here)

Pocket agents answered only a small number of questions, and many of their answers were unreasonable- like "We removed this feature because it is not convenient for our future plans"

Example of this is still stored in Google cache

24Apr I've posted a request to restore RIL Pro in AppStore until they will fix the problems in pocket.

From 27 April they introduced moderation on the public forum and stopped show new posts.

Actually "moderation" is not a correct work, it's converted the forum into Q&A site, where they show only posts with answers that they like.

Furthermore they removed many discussion threads that existed on the forum before the "moderation".

When I pointed to this in email, the answer was un-reasonable-that their forum is actually not a forum, but "a helpful support site where users can get their questions answered"

In fact users post public questions and do not see them posted for a week or forever.

For couple of my questions I received email starting with the words

"An agent has responded to your question (Link to Question<http://readitlater.assistly.com/customer/portal/questions/306173-sync-in-ril-free-stopped-working>)."

But the link is not working, because they deleted the discussion from the forum.

At the beginning of May Pocket support answered that the "known issue with a partial list sync" is addressed and will be in the next upgrade.

Despite the statement in upgrade notes, next upgrade didn't help with the problem(at least for me).

Support replied to me, that "it seems as though a small handful of users are still experiencing this sync discrepancy."

27Apr I've reported that "Sync in RIL Free stopped working" and it was on forum at least until 6 May(according Google cache)

10 May I received a reply via email(but post has been removed from the forum)

"The fix is to log out of Read It Later Free, and then log back in. Although your local cache will be cleared, the app will go through your list and re-download your items."

It was the first useful advice after introducing Pocket. Now I am able to use RIL Free. With all it's limitation RIL Free is in much more workable conditions, than Pocket. It shows all articles that I've saved and it allowed to rename titles. The only feature from RIL Pro, that I really missing, is the ability to filter only untagged articles.

So now I can read all my saved articles in RIL Free and wait until they will fix the Pocket.

The conclusion I've had is that the Pocket company doesn't respect their customers, and I should not rely on their products too much. However the application is quite convenient, and I am familiar with it, so I will continue to use it.

The company's behaviour reminds me how Google changed layout of iGoogle tabs and ignored hundreds of complains. Note that recently (4 years later) Google returned to layout similar to original.

2012 Transit of Venus - First Web Camera located.

TATWORTH — Tue, 15 May 2012 10:18:40 GMT

Since the 2012 Transit of Venus is the last transit of Venus across the face of the Sun as viewed from Earth, for many years, being able to locate a web cam feed that will be in clear daylight for the whole of the transit period. So far I have only located one.

This is at: http://www.transitofvenus.com.au/HOME.html

Historically the transit of Venus was very important for determining the absolute size of the solar system, however high power radar has allowed direct measurement of distance at any time.

Debugging Castle WcfFacility installers

jeremyj — Tue, 15 May 2012 08:00:59 GMT

Stepping through IWindsorInstaller implementations in web services created using the WcfFacility^[1] is not as immediately accessible compared to debugging the start up of a console application. Just pressing F5 with the web service set as a start up project causes the start up code to have already been run before the debugger attaches to the IIS worker process. Fortunately there is a simple way of causing the start up code to be re-run whilst remaining attached to the IIS worker process, so that you can step through your Castle installers and access the debugger views.

Once you've attached to the worker process under the debugger you can just rebuild the web service project to force IIS to re-initialize the newly built assembly when it is next exercised.

Once you've attached to the worker process under the debugger^[2] you can just rebuild the web service project to force IIS to re-initialize the newly built assembly when it is next exercised. The IDE won't normally let you do this^[3] so you have to build the web service project outside the IDE that is running the debugger. Of the various ways of doing this i'd like to suggest running the .csproj file directly in MSBuild.exe on the command line (because it means that just pressing the up arrow and enter in the command window will easily let you re-build). So with the debugger attached to the IIS worker process of your web service, just re-build the web service project on the command line e.g.

C:\Windows\Microsoft.NET\Framework\v4.0.30319\MSBuild.exe My.Web.Service.csproj

As soon as you call any of its operations IIS will re-initialize the service causing the breakpoints to be hit. I hope this saved you some frustration as it took me a while to figure this one out. Happy Debugging.

[1] If you are using the Castle WcfFacility your My.Web.Service.svc will probably look something like this -

<%@ ServiceHost Language="C#"

Debug="true"

Service="My.Web.Service"

Factory="Castle.Facilities.WcfIntegration.DefaultServiceHostFactory, Castle.Facilities.WcfIntegration" %>

[2] I.e. either you've just hit F5 with the web service project as the start up project, or you've attached to the worker process by selecting it in the processes list (If the worker process isn't showing in the list it might be because it isn't yet running, the worker process hosting the application pool can be started just by calling one of the web service operations. Alternatively you might have to tick the box to include processes from all sessions.)

[3] I've definitely seen a bug in VS2010 SP1 where under rare circumstances it will let you, i have been able to do this once but now have no idea how to reproduce this issue now...

Apache 2.4 and the missing SSL certificate chain issue, my notes.

TJ — Mon, 14 May 2012 22:50:10 GMT

I'm running Mahara in a Windows Server 2008 environment using Apache web server and just for fun I wanted to enable SSL :-).

I setup SSL and tested the setup using a self-signed certificate, all good, SSL was working.

I requested a server certificate from our corporate CA, which is an intermediate CA for GlobalSign, this was done using an online MS certificate server service.

When the certificate was ready, I revisit the site, and the certificate is installed automagically into the current user certificate store.

Okay, so now I have a SSL certificate stored in the Windows certificate store, but I want to use it with Apache, so I'm sort of blending the MS way of doing things with the Apache way of doing things.

Here is what I did once I had the proper certificate.

First of I needed to export the certificate from the current users certificate store so it can be used by Apache, to do this.

Fire up an MMC console, and add a certificate snap-in, selecting My users account
Once opened, browse to the new certificate
Right click on it and export
Select, Yes, export the private key
Select, Include all certificates in the certification path if possible and Export all extended properties
Enter a password <importprivatekeypassword> and confirm
An then enter a descriptive filename, I used "newcert-19-05-2012-with-privte-key.pfx"

Next I need to convert the certificate to something Apache understands, and at the same time export the private key, as Apache needs the private key in a separate file when running on Windows.

I copied "newcert-19-05-2012-with-privte-key.pfx" to the same folder as openssl.exe is located, c:\web\apache\bin, just to make it easier.

I start an elevated command prompt

CD to the folder where openssl.exe is located

Run openssl and I get the OpenSSL> prompt.

Now I'm going to run the following commands

To export the private key file from the pfx file
- pkcs12 -in newcert-19-04-2012-with-privte-key.pfx -nocerts -out key.pem
To export the certificate file from the pfx file
- pkcs12 -in newcert-19-04-2012-with-privte-key.pfx -clcerts -nokeys -out cert.pem
To remove the passphrase from the private key
- rsa -in key.pem -out server.key

This is what it looked like at the OpenSSL> prompt

OpenSSL> pkcs12 -in newcert-19-04-2012-with-privte-key.pfx -nocerts -out key.pem

Enter Import Password:

MAC verified OK

Enter PEM pass phrase:

Verifying - Enter PEM pass phrase:

OpenSSL> pkcs12 -in newcert-19-04-2012-with-privte-key.pfx -clcerts -nokeys -out cert.pem

Enter Import Password:

MAC verified OK

OpenSSL> rsa -in key.pem -out server.key

Enter pass phrase for key.pem:

writing RSA key

OpenSSL>

Now I have a server certificate "cert.pem", and a key file, "server.key"

I copied these to where I'm storing my certificate files, c:\web\apache\cert

Next updat httpd-ssl.conf...

SSLCertificateFile "c:/web/apache/cert/cert.pem"

SSLCertificateKeyFile "c:/web/apache/cert/server.key "

Restart Apache

At this point the certificate worked, sort of...

It was fine in Mozilla Firefox, but Internet Explorer couldn't see the certificate chain.

To fix this I had to download the CA chain from the same server the certificate was requested, an then extract each of the 4 certificates in the certificate chain. All the guides I read said I need the certificates as Base-64, and to use the Unix CAT command to copy them to the one file. I tried doing this with a standard text editor, but that didn't work, I ended up using the DOS (Windows?) TYPE command.

So download the CA chain file

Double click on the CA chain file, and then drill down to the CA and intermediate certificates

Right click on each of the certificates starting from the top level certificate and export them as Base-64 encoded X.509 (.CER) files

I exported them as cert1.cer, cert2.cer, cert3.cer, cert4.cer

Now concatenate the certificates into one file using the "type" command

C:>Type cert1.cer cert2.cer cert3.cer cert4.cer >server-ca.crt

Then I copied "server-ca.crt" to my certificate folder, c:\web\apache\cert

Update httpd-ssl.conf

SSLCertificateChainFile "q:/webroot/apache/cert/server-ca.crt"

Restart Apache

Test, all good...

Hope this helps someone, or me if I try to do this again :-)...

The Forefront Identity Manager Service cannot connect to the SQL Database Server.

Leonard Mwangi — Mon, 14 May 2012 16:01:10 GMT

After configuring user profile you get the correlation id and in event logs you get

The Forefront Identity Manager Service cannot connect to the SQL Database Server.

The SQL Server could not be contacted. The connection failure may be due to a network failure, firewall configuration error, or other connection issue. Additionally, the SQL Server connection information could be configured incorrectly.

Verify that the SQL Server is reachable from the Forefront Identity Manager Service computer. Ensure that SQL Server is running, that the network connection is active, and that the firewall is configured properly. Last, verify the connection information has been configured properly. This configuration is stored in the Windows Registry.

Open central administration à security à Configure Service Account (On the drop down pick Farm Account) and ensure that the select and account for the component is the same as the one you are using to configure FIM (User profiles) the OK.

That should fix that issue.

Happy FIMing

Is It Done? Really?

Jonathan Mills — Mon, 14 May 2012 08:29:35 GMT

Everyone has done it, you know you have too. When someone asks you for a status update on a new feature you just finished coding, you say "That's Done!" Think about what you just said for a minute, is it really done? Can someone use that feature in production for its intended purpose? That's really what done means, right? To the end user, done means its up and running, in production, providing value. Code complete is just another stage in the pipeline getting us out to really done.

You may say, that is a semantic argument. Your piece is done, so you get to say done. I would argue, its a mindset discussion. As developers, if we think about our piece, and only our piece, we say done when our piece is done. Agile is about picking your head up and looking at more than just your piece. Collaboration and a sense of unity across the entire product team is critical to a successful agile organization. We as developers need to move away from our tasks as "in dev" and "done". We have to get to the point where we think of a feature in development, or in test, or in requirements. We need to pick our heads up and see the whole picture, not just our piece. Next time your manager or scrum master asks you for a status on a feature you just completed coding, resist the urge. Instead of saying, "that's done", give them the actual status of the feature. Your larger team will thank you for it.

Who is Expression Web aimed at and at what market?

ihaynes — Mon, 14 May 2012 09:05:30 GMT

In answering this question it helps to look at the overall creation of a website and the processes involved.

The term 'web design' still in common use, is something of a misnomer. 'Web creation' seems to be a better term and it involves three distinct processes.

The design of the ‘look and feel’ of the site or the ‘UI’. This is often (and best) done by someone with artistic or graphic design skills, but with an appreciation of what is possible on the web. This is the real ‘design’ activity.
The translation of this design into a functioning page template, using HTML, CSS and JavaScript. A process known as ‘front end’ development. An increasingly complex task given the rise of the mobile web.
If required, coding in a server-side language of dynamic functionality, possibly with a connection to a data source of some sort. The process of ‘back end’ development.

Now obviously there are many scenarios where two or even all three of these are handled by the same individual but these are still the processes involved and all require skill, experience and appropriate tools for the technologies involved.

It’s clear from these processes where Expression Web sits, as a ‘front end’ development tool and therefore aimed at ‘front end’ developers, not ‘designers’ as such.

It also follows that as a tool for this EW needs to keep up to date with the technologies involved, something that sadly hasn’t happened. Nor does Microsoft offer any other solution for front end development.

Stylecop 4.7.20.0 has been released

TATWORTH — Mon, 14 May 2012 05:46:35 GMT

Stylecop 4.7.20.0 has been released at http://stylecop.codeplex.com/releases/view/79972
Release Notes

Compatible with the Visual Studio 11 Preview.

Install order should be :

VS10
VS11
R#6.1.1 msi (for VS10)
R#6.1.1 vsix (for VS11)
StyleCop 4.7

This version is now compatible with R# 5.1 (5.1.3000.12), R# 6.0 (6.0.2202.688), R# 6.1 (6.1.37.86) and R# 6.1.1 (6.1.1000.82).

Here are the bug details for fixed in 4.7 and closed in 4.7 issues (over 100 issues fixed since 4.6)

Here are the bug details for all issues since 4.3.3.0 that have been fixed and closed (over 450 fixes).

Release Notes
Added support for cultures into the Settings.StyleCop file. Currently supports en-US,en-GB, fr-FR, pl-PL, pt-BR, de-DE and ru-RU.

NYC Metro Area Community Dinner

MOSSLover — Sun, 13 May 2012 22:27:23 GMT

So I have decided to hold a community dinner the first week of June. I invited a bunch of MS community members who run user groups or contribute a lot in the NYC Metro area. If anyone is interested drop me comment on this blog with your email address. I won’t publish the comment with your email in case you don’t want everyone to see your address. I just figured that some people might see this blog that I missed via email. I am also going to spam the main community list for njny metro area. Hopefully, we can get a good turn out from the community at large and have some good dialog discussions on events we are running. I am sure a lot of don’t see each other often so I figured why not. Anyway hope to see you all at the dinner.

Technorati Tags: Microsoft Community,New York Metro Microsoft Community

Project SET Day–Insights and Observations

Lori Lalonde — Sun, 13 May 2012 18:00:35 GMT

A few weeks ago, I had the privilege of being able to participate in Conestoga College’s Software Engineering Technology project demo day as a guest judge, along with 4 other members of the IT community.

This event is held every year at the end of April, and provides graduating students the opportunity to present their group projects to those in attendance, discussing the business problem it solved, the challenges they faced, how they addressed those challenges, and the future direction they have planned for their product.

My first impression when walking into the room was a positive one. The students were dressed to the nines and were very enthusiastic about discussing their projects, as well as answering any questions. They stood at their project booths patiently, willing to discuss their projects one-on-one with passers-by, and to provide a working demo of their product.

Projects ranged from a mobile Android app that would serve as a mobile recording studio for guitarists to an inventory tracking application to an educational application with a web site that tracked the child’s progress which parents could log in to view details on.

During the formal presentation rounds, the students not only demonstrated that they had the technical skills necessary to hit the ground running in the workforce, but they also demonstrated collaborative team work, business acumen, and real passion for their work. They handled questions with ease, provided insights into the obstacles they faced, how they learned from them, and how they overcame them.

Along with the other members of the judging panel, I participated in deliberations and we had to make the difficult decision to select a winning team. After much discussion, we chose a team that had not only developed their intended project, a mobile app to serve data from an existing Health & Safety Watch web site, but had gone above and beyond to redesign the website to conform to the look and feel of the mobile app design, making it simple, clean, concise and easy to use. In fact, I have the site bookmarked and check it daily: http://www.healthandsafetywatch.com

All in all, each student that participated in this event proved that they are all winners who are more than well-equipped to join the workforce and make an impact. If you have the opportunity to mentor a new graduate, I hope you do so wholeheartedly. You may find that, in turn, you will learn from them as well.

C# Tutorial for log4net

huwyss — Sun, 13 May 2012 16:54:59 GMT

Here's a short tutorial on how to use log4net in C#

1. Get log4net from the apache website.

2. Open your project with visual studio.

3. Add the reference to your project: You find the reference in the zip that you just downloaded: \bin\net\xxx\release\log4net.dll. xxx is your .net version.

4. Add the Appender section to your app.config. The following code uses a file for the logging:

<configuration>
<configSections>
    <section name="log4net" type="log4net.Config.Log4NetConfigurationSectionHandler,Log4net"/>
</configSections>
<log4net>
    <root>
      <level value="DEBUG" />
      <appender-ref ref="LogFileAppender" />
    </root>
    <appender name="LogFileAppender" type="log4net.Appender.RollingFileAppender" >
      <param name="File" value="log-file.txt" />
      <param name="AppendToFile" value="true" />
      <rollingStyle value="Size" />
      <maxSizeRollBackups value="10" />
      <maximumFileSize value="10MB" />
      <staticLogFileName value="true" />
      <layout type="log4net.Layout.PatternLayout">
        <param name="ConversionPattern" value="%date [%thread] %-5level %logger [%property{NDC}] - %message%newline" />
      </layout>
    </appender>
</log4net>
</configuration>

5. Use the following code to use the configuration you just added to app.config:

static void Main()
{
      log4net.Config.XmlConfigurator.Configure();
...

6. To log use the following code:

using log4net;
...
private static readonly ILog log = LogManager.GetLogger(typeof(Bar));
log.Debug("this is the first log message");
...

7. If Visual studio doesn't recognize log4net. Configure your project like this:

Visual Studio -> project -> project name properties -> target framework -> .net Framework (not client)

Mother’s Day - To The Forgotten Moms…

D'Arcy Lussier — Sun, 13 May 2012 14:53:34 GMT

Today is a day to celebrate mother’s everywhere. There will be breakfasts cooked, cards opened, gifts given, brunches and dinners shared. There will be church services where mothers are given flowers from their children and sermons will talk of the blessings mothers are. People who aren’t even related will wish others who are mothers a “Happy Mother’s Day”. And all of this will go on without any consideration or thought to the forgotten mothers in our midst.

For many, mother’s day (and father’s day) isn’t a celebration. It’s a reminder. A reminder of what could have been. I reminder of hope lost, of dreams shattered, of “what if’s”.

***

It’s their first ultrasound. They wait with anticipation to see the little heartbeat on the monitor. Something is wrong though…the doctor furrows his brow. He points to where the fetus is attached, and then delivers words like a sledgehammer to the heart: There’s no heartbeat.

***

She cradles her baby in her arms, singing her a lullaby. There’s a knock on the door. It’s time. She hands her daughter to the social worker along with a bag filled with her toys and clothes. She kisses her daughters forehead and says goodbye, watching the social worker get into her car and drive off into the night. The adoption process had gone well and there was no warning signs. But she knew this was a possibility. 16 days ago her daughter was born, 14 days ago she was brought home, and 2 days ago the birth mother changed her mind and wanted her baby back.

***

The local high school is getting ready for graduation. Fundraisers are being organized, dresses are being ordered, and plans for parties are being made. She hears over coffee with friends how their kids are getting ready to leave high school and embark on the next phase of life at college. She hears them reminisce on various events of the last 4 years and how their kids have grown into fine young adults. They finish coffee, say goodbye, and she drives to the florist. Thirty minutes later she drives to the cemetery, placing the flowers on her son’s grave. 7 years later, it still feels like that first night when she got that horrible phone call.

***

So to those moms out there that don’t have a section at Hallmark devoted to them, to those moms who keep their pain hidden and tolerate Mother’s Day, to those moms that never got to meet their children or did for only a short amount of time, and especially to those moms that don’t feel that today applies to them:

This is your day to recognize the sacrifices you made, emotionally/physically/spiritually, in giving your heart to another human being. This is your day to acknowledge the care, nurturing, and love that you showered on your child, regardless of how long you had with them. That doesn’t make the pain and memories hurt any less, but please know that in the frenzy of Mother’s Day you are not forgotten.

You are a mother. This is your day. Happy Mother’s Day.

Quick Analysis of Popular Canon and Nikon DSLR Cameras

Kit Ong — Sun, 13 May 2012 11:31:37 GMT

Photo quality aside, it seems that Canon DSLRs are more popular than Nikon DSLRs right now based on Flickr stats.

Below is a ranking of popularity based on daily users (taken from Flickr.com)

Canon Camera (Average daily users)
EOS 5D Mark II (4350) $2500
EOS REBEL T2i / 550D (3908) $500-$600
EOS 7D (3577) $1500-$1800
EOS REBEL T1i / 500D (2507) $400-$500
EOS 60D (2219) $800-$1300

Nikon Camera (Average daily users)
D90 (3985) $1000-$1200
D7000 (2664) $1200-$1500
D3100 (2173) $500-$700
D700 (1912) $2200-$3000
D5000 (1521) $600-$1000 (superseded by D5100)

Key to below stats:

# of items - This is the total number of photos uploaded per model.
Avg. daily users - This is the number of Flickr members who uploaded 1 or more photos yesterday.
Activity Factor The "Activity Factor" is a number derived from photos, members and a model's rank to indicate cameras that are used "a lot".
Type Point & Shoot / DSLR / Cameraphone / Camcorder
Rank A model's rank (within its brand) is based on how many people used it yesterday.

Canon Popularity

Nikon Popularity

Canon DSLR

Nikon DSLR

Links for more info.
http://www.flickr.com/cameras/canon/
http://www.flickr.com/cameras/nikon/
http://en.wikipedia.org/wiki/Template:Canon_DSLR_cameras
http://en.wikipedia.org/wiki/Template:Nikon_DSLR_cameras

How to do debugging with visual studio 2008/2010 ?

Chandradev — Sat, 12 May 2012 12:51:04 GMT

Hi

If you have good debugging skill then you can save lots of time. In visual studio lots of options are available for debugging.

Here are some good URL for debugging. It will be really useful in our daily working project.

http://www.codeproject.com/Articles/309781/Advanced-Debugging-in-Visual-Studio
http://weblogs.asp.net/scottgu/archive/2010/08/18/debugging-tips-with-visual-studio-2010.aspx
http://www.codeproject.com/Articles/79508/Mastering-Debugging-in-Visual-Studio-2010-A-Beginn
http://www.dotnetperls.com/debugging
http://geekswithblogs.net/sdorman/archive/2009/02/14/visual-studio-2008-debugging-ndash-the-watch-window.aspx

Debugging in loop

http://www.c-sharpcorner.com/uploadfile/suchit_84/debugging-tips-using-visual-studio/
http://blog.dreamlabsolutions.com/post/2008/12/11/Debugging-tips-and-tricks-for-Visual-Studio.aspx

Debugging in Store Procedure

http://codeasp.net/blogs/chandradev/microsoft-net/1901/how-to-debug-store-procedure-in-sqlserver

I hope it will help to someone

Links about SQL PIVOT

Michael Freidgeim — Sat, 12 May 2012 08:34:26 GMT

One of my colleagues asked how to create SQL query to show results as columns of a row, that actually stored in multiple rows in the table. Other co-worker suggested to use cursor to manually generate required rows.
I’ve found a few links about SQL PIVOT.
It is easy to start with
http://www.kodyaz.com/articles/t-sql-pivot-tables-in-sql-server-tutorial-with-examples.aspx
or http://programmingsolution.net/post/SQL-Server-Pivot-Converting-Rows-to-Columns-with-Dynamic-Query.aspx

More detailed are articles written by Itzik Ben-Ganhttp://www.sqlmag.com/article/tsql3/pivot-or-unpivot-your-data
and
http://www.sqlmag.com/content1/topic/pivot-or-unpivot-your-data/catpath/tsql3/page/2compaslightly

How to use Pivoting Without Aggregation is described inhttp://www.sqlmag.com/article/tsql3/pivoting-without-aggregation

Promoting BizTalk in your organisation

Michael Stephenson — Fri, 11 May 2012 20:26:26 GMT

This week I was working with a customer and one of the interesting things in the organisation is that they have a fairly large investment in BizTalk but in general few people in the organisation are knowledgeable about how important a role it plays. I decided it might be interesting to do a small PR exercise; I would write a little post with 10 bits of information about their BizTalk investment that they might not be aware of. It was also interesting that I found out a few things that I didn't know myself in terms of the stats.

The idea of the post was as follows:

It should be written in a not overly technical language so it could be given to business people
It should talk about some of the business processes
We would like to talk about some technical things but it must be explained with examples which are not technical
It might reference things that the organisation does particularly well

Below is an example of the article for a fictitious company called Acme.

10 things you might not know about BizTalk at Acme

BizTalk is a black box which plays a very important part in the Acme application architecture. A lot of people have probably heard of it but don't know much about what it does. This article will help to tell you a little more about what it is and its role.

1. What is BizTalk

BizTalk is a product from Microsoft which is a centralised Integration Server platform. It is used to develop solutions to integrate your applications.

As a platform it allows you to develop solutions from simple application connectivity to complex ESB, hybrid Cloud or B2B scenarios. The platform allows you to host these integration solutions side by side so they can benefit from the investment made in each other.

BizTalk provides many "quality of service" features such as a scalable hosting environment and administration and management tools which let you support solutions in production.

2. Where does it fit at Acme

At Acme over the years we have developed many integration solutions with BizTalk. Some of the bigger better known solutions include "process a" instances between application A and application B most of the interaction between these applications happens via BizTalk messaging and workflow. BizTalk deals with complex message transformation between the applications, provides durable messaging to handle any server downtime and fault handling in addition to workflow to manage the system to system integration processing.

In Acme almost all documents which go to customers will go via an application called "application c". BizTalk enables the integration from "application a" into "application c" and BizTalk handles both real time document requests and heavy duty batch processing. BizTalk makes it easier for these two applications to work together to deliver the end customer value.

In Acme approximately 5000+ "process a" instances are processed per day from B2B partners and loaded into "application a".

In Acme all payments from customers go through BizTalk in both real time and for monthly payments.

All orders Acme receives are processed through BizTalk

BizTalk is also involved in many other processes.

3. Number of messages processed

It is estimated that BizTalk has processed in the region of n messages since it was introduced to Acme.

[Note: this was an estimate based on the average number of messages per day from HAT queries]

4. More Stats

Since BizTalk went live it has processed approximately:

£n worth of customer payments
n documents have been sent to customers through BizTalk
n orders from B2B partners have been processed through BizTalk

5. Long running transactions

BizTalk excels in the area of long running transactions. At Acme we have some processes which can take hours or even days to complete. Because BizTalk is able to maintain the state of a workflow the long running process can do some work and then stay in a dormant state until the next part is due to execute or until an application sends the response message back to trigger the next part of the process.

This state persistence is also able to withstand failure conditions or even the whole data centre going down and a disaster recover scenario happening.

6. Load Levelling

One of the key features of BizTalk which plays a very important role in the application performance at Acme is a pattern called load levelling. If you imagine that in some areas of the systems you have different amounts of activity at different times. For example overnight from 10pm onwards we get a large spike in the number of "process d" and "process e" instances to support the overnight batch processing. During the day these processes also execute but it is a much smaller volume. Overnight the input into BizTalk is at too high a rate to be passed through to "application e" at the same rate without significant scaling of these applications. In BizTalk we flatten this load and deliver it to the end system at a controlled rate which the application can manage. Imagine pouring water through a funnel.

7. Fault Handling

Typically applications are prone to errors and downtime. One of the most important features of BizTalk is its ability to recover from errors in an application it is trying to integrate with. If for example "application a" threw an error when BizTalk tried to send a message to it then BizTalk can work out what kind of error it is and if an automatic retry can be attempted. The error may persist and indicate a bigger problem. If the number of automatic retries is exhausted then BizTalk will suspend the message but provide a management interface so that the support team can see the problem, workout what went wrong and can start the retry process again when the issue is resolved.

8..9..10….

Ive removed the last few as they were a little too specific to the customer to display on my blog, but hopefully you get the point.

This communication actually went down really well and I think served as an excellent way to tell people a little about what this expensive black box they occasionally hear about is and why it's important to the organisation.

Has anyone else done anything similar?

Extension method for outputting formatted JSON in LINQPad

Elton Stoneman — Fri, 11 May 2012 17:29:57 GMT

Very simple, but a handy way of viewing nicely-formatted JSON in LINQPad:

I use Json.NET as it does indented formatting, and represents JSON dates properly. To add as an extension, download Json.NET, add a reference to Newtonsoft.Json.dll to your “My Extensions” query and add the following code to your MyExtensions class:

public static object DumpJson(this object value, string description = null)

{

return GetJsonDumpTarget(value).Dump(description);

}

public static object DumpJson(this object value, string description, int depth)

{

return GetJsonDumpTarget(value).Dump(description, depth);

}

public static object DumpJson(this object value, string description, bool toDataGrid)

{

return GetJsonDumpTarget(value).Dump(description, toDataGrid);

}

private static object GetJsonDumpTarget(object value)

{

object dumpTarget = value;

//if this is a string that contains a JSON object, do a round-trip serialization to format it:

var stringValue = value as string;

if (stringValue != null)

{

if (stringValue.Trim().StartsWith("{"))

{

var obj = JsonConvert.DeserializeObject(stringValue);

dumpTarget = JsonConvert.SerializeObject(obj, Newtonsoft.Json.Formatting.Indented);

}

else

{

dumpTarget = stringValue;

}

else

{

dumpTarget = JsonConvert.SerializeObject(value, Newtonsoft.Json.Formatting.Indented);

}

return dumpTarget;

}

Call with .DumpJson() on any object, and it will serialize to formatted JSON and output it. If you pass a string which is already JSON, it round-trip serializes it to get the formatted version (including converting WCF-formatted JSON dates to ISO 8601).

Makes for a very handy testing rig for REST services.

Work Life Balance

rickpetersen — Fri, 11 May 2012 08:31:34 GMT

Where’s Daddy?

My wife, one year old daughter and I play a game that most other parents have played at some point; my wife will hold her and ask her "Where’s daddy?” and the response is invariably a wide-eyed scan of the room by my daughter, culminating in a big snotty slobbery smile when she locates me. This is a game and is fun for her but it is an important exercise for linking a name and role with a face for her and for enforcing the idea that when she looks for me, she will find me.

So what about when we cannot be there? What about when my six year old daughter asks this question and the answer is “We will see him Friday”? I read the blogs of techies that travel the country speaking at conferences and user groups. Other road warriors spend their time attending or leading developer boot camps and still others are meeting with clients and doing marketing and sales work. Many of these techies have families and my assumption is that they place a high level of importance on their family life.

So how do you do it? How do you manage a passion for technology, and near chemical addiction to learning and sharing information with others, alongside the responsibilities you have within your family?

Techniques for Balance

1. Schedule Time for Life – How much work would you get done if you had absolutely no plan and no schedule? This is not just a reference to scheduling time for your spouse and/or children if you have them, but it is also important to schedule some time alone for reflection or some time with the girls/guys. LifeHacker made reference to this. We have smart phones, free calendar services and a habit of scheduling time at work just take advantage of these facts and treat your non-work time like it is just as important as your work time.

2. Learn to say “no” at work. Your bosses love your dedication and would love you to continue to push yourself and outperform all expectations, but they will not be pleased when you burn yourself out. TalentCulture.com has a good set of suggestions for saying no.

Technologies to help Balance

First, a definition. I am generically defining technology as the application of tools rather than processes to solve a problem. It may not be the right definition but it is my definition. Technology should help you manage your time in such a way that you get things done at work and can feel good about having time away for yourself.

1. Organize your work-life. Once upon a time I tried to use the 43 Folders system but at that time I really did not have much responsibility. I think now I may benefit from taking another look at it.

2. Calendaring. I mentioned this in point one of the techniques section, but set up a calendar account, or use your work one, and schedule time for yourself and your family; as an extension, do not dismiss those appointments when they come up. You need time for you and ignoring that fact will diminish what you can do when you are at work. If you are married, show your spouse how to schedule things on your calendar so that they know they can get your attention. I use my Google/gmail calendar and find it works well.

3. The off button. Use it. I am not going to say you need to unplug for a week at a time or commit infosuicide but just unplug long enough to enjoy something else. Take your wife on a date. Take your son or daughter out to show them they are special… and turn off the phone.

And then… (AND NO AND THEN!)

Technology and our careers are important; some of us are really using our love of technology to make a difference in the world but in the end you are not your job. If you live to work, you will get to the end of the work and have nothing else to do. If you work to live, you will get to the end of the work and be excited to more live more fully. Not only will it be better for your home life, it will be better for your employer; the research supporting this statement is not exactly new.

Work less hours, work harder and more productively during those hours and live a fulfilling life both inside and outside the office.

Pre-Processing Directives for Multi-Platform Support

Joe Mayo — Fri, 11 May 2012 06:33:53 GMT

Over the last several months, I’ve been modifying LINQ to Twitter to support multiple platforms. Originally, I had targeted full profile .NET, ASP.NET Web Forms, and ASP.NET MVC apps. An easy stop along the way was Mono compatibility, but that was simple because Mono is good about running any CLR app. The code-base was good until I started porting to Silverlight.

Porting to Silverlight

Silverlight requires a different type of class library, so right-away I couldn’t use the same libraries as the full profile version. To fix this, I moved the physical files over to the Silverlight library and tried to compile. This didn’t work because certain .NET libraries were either not present, missing features, or required additional code in Silverlight. I didn’t want to maintain two separate code bases, so I left the physical files in the Silverlight project and created file links from the full profile library to the Silverlight files. To create file links, I right-clicked the folder, selected Add Existing Files, and navigated to and selected the files. Instead of clicking the Add button, there’s a drop-down arrow on the right side of the Add button, which has an option named Add As Link, which I selected. This still doesn’t fix the problem with differences in libraries.

To account for the differences in libraries, I used pre-processing directives. Silverlight has a defined SILVERLIGHT symbol, that ships with the default project template, which is what I used. Here’s an example of how I used it:

#if SILVERLIGHT
            WebRequest.RegisterPrefix("http://", WebRequestCreator.ClientHttp);
            WebRequest.RegisterPrefix("https://", WebRequestCreator.ClientHttp);
#endif

The example above demonstrates a Silverlight-only scenario requiring registration of prefixes for HTTP communication. It uses #if and #endif to block off code that will only compile in the Silverlight project, which defines the SILVERLIGHT symbol.

A natural consequence of this strategy is that you only want to use a symbol that is unique to a specific project type if the logic only pertains to that project type. However, sometimes the lines blur. After porting to Silverlight, I went on to create a Windows Phone version of LINQ to Twitter.

Porting to Windows Phone

Since Windows Phone uses Silverlight, I was tempted to use the same define symbol, but that wouldn’t be good because there are situations, like the example above that are unique to Silverlight Web Applications, but not Silverlight for Windows Phone. Fortunately, Windows Phone has a default WINDOWS_PHONE symbol that ships with its templates. Here’s an example for a Silverlight scenario that doesn’t work for Windows Phone:

#if SILVERLIGHT && !WINDOWS_PHONE
            if (!Application.Current.IsRunningOutOfBrowser)
            {
                ProxyUrl =
                        Application.Current.Host.Source.Scheme + "://" +
                        Application.Current.Host.Source.Host + ":" +
                        Application.Current.Host.Source.Port + "/LinqToTwitterProxy.ashx?url="; 
            }
#else
            ProxyUrl = string.Empty;
#endif

In the example above, Silverlight Web Apps require a proxy, but Windows Phone doesn’t. This results in a more sophisticated condition that excludes Windows Phone’s from being configured with a proxy. Notice the AND and NOT operators and the #else clause for refining this condition.

Porting to Client Profile

After supporting Silverlight and Windows Phone, I turned my attention to a recurring problem that I saw happening. It seems that people would try out LINQ to Twitter by starting a new Console app and running a simple query. The problem is that it wasn’t so simple because the default profile of a Console application is Client Profile. The conflict occurs because LINQ to Twitter supports ASP.NET Web Forms and MVC, whose libraries are not included in Client Profile. This prevented the application from building. Most people figured out that they could move forward by opening Properties and and switching the Target framework to one of the non-Client Profile options, but the situation did highlight the usefulness of supporting Client Profile.

Client Profile doesn’t have any default defined symbols, so I created one for the new library named CLIENT_PROFILE. At first, this was very useful because I learned that Client Profile doesn’t include any of the JSON serialization libraries that I was using. Fortunately, I had recently integrated LitJson into LINQ to Twitter, which I could use with the Client Profile library. Here’s an example:

#if SILVERLIGHT || CLIENT_PROFILE
            var array = dictionary.GetValue<list<object>>(key, null);
#else
            var array = dictionary.GetValue<ArrayList>(key, null);
#endif

The example above uses the OR operator in the pre-processing directive to indicate that both Silverlight and Client Profile need the same code – operating on a generic List<object> because ArrayList isn’t part of their profiles.

Porting to Windows 8

Most recently, I ported LINQ to Twitter to support Windows 8 Metro Style Class Libraries. The template for Windows 8 libraries ships in Visual Studio 11 and defines the NETFX_CORE symbol. There are a few significant library changes that I had to account for, but making the code work on multiple platforms, including Windows 8 is doable. Here’s one example:

#if NETFX_CORE
                                            Task.Delay(errorWait);
#else
                                            Thread.Sleep(errorWait);
#endif

In Windows 8, you use the Task Parallel library (TPL), instead of the Thread class and other .NET libraries. Since LINQ to Twitter supports .NET 3.5, but Microsoft doesn’t officially support TPL for .NET 3.5, the Thread class is still necessary. I encountered more scenarios with various types, reflection, and collections that needed special handling with pre-processing directives, all of which were handled similar to the examples you’ve seen so far. Moving the code between Visual Studio 11 and Visual Studio 2010 was seamless because I didn’t change the Target frameworks on any of the other class library types.

Note: You can view the LINQ to Twitter Source Code on CodePlex to see what my check-ins looked like today (5/10/12) to see how I made the LinqToTwitterRT library changes and merged with the existing code base.

Summary

Porting class libraries to different .NET platforms isn’t always easy. Each platform adds and removes features that are appropriate in their own context. My approach was to have a common set of physical files, link to those files from other projects, and then use pre-processing directives to manage the differences. You saw examples of how I did this by porting to Silverlight, Windows Phone, Client Profile, and finally Windows 8. As you can see the code gets a bit fragmented at times, but it isn’t too bad, and most importantly, it works.

@JoeMayo