VSTS gives you hope in the Queries interface by providing an option to "Move to team project..."

However this doesn't work for Test Cases as they are classed as hidden and you will get an error message if you try.

Thankfully, if you have Microsoft Test Manager installed, then you have access to Test Case Management CLI (tcm.exe)

To migrate test cases from one project to another, what you have to do is:

• Select Test | Test Plans
• Create a new Test Plan (call it All Test Cases)
• Create a new Query Based Test Suite (call it All Test Cases)

• For the Query, just use "Work Item Type" "In Group" "Microsft.TestCaseCategory"
• Select all your test cases and Create Suite
• You need to note the Test Suite ID as we'll need this in the command line later. I'll refer to this as the <SourceSuiteID>

Now in your destination Team Project (<TargetProjectName>)

• Create a new Test Plan called Imported Test Cases
• Note the Suite ID of this test plan. I'll refer to this as the <TargetSuiteID>

Now fire up the Developer command prompt

C:\Users\ephan>tcm suites /clone 
/collection:https://<youraccount>.visualstudio.com/defaultcollection 
/teamproject:<SourceProjectName> 
/suiteid:<SourceSuiteId> 
/destinationteamproject:"<TargetProjectName>"  
/destinationsuiteid:<TargetSuiteId> 
/overridefield:"Iteration Path"="<TargetProjectName>)"
/overridefield:"Area Path"="<TargetProjectName>" 

Started clone operation with id 1.
Use /status:1 to get progress and completion information of the operation.
Warning: You have cloned the following query based suites as part of this clone operation. You are advised to fix their respective queries in the source and destination suites, after the clone operation completes.
1. All Test Cases -> All Test Cases

It should give you an operation id which you use to query the status of the job.

C:\Users\ephan>tcm suites /status:<Operation ID> /clone /collection:https://<youraccount>.visualstudio.com/defaultcollection /teamproject:<SourceProjectName>

Clone operation 1 is in progress.
Total test cases:    446
Processed:           198
Progress:             44%

C:\Users\ephan>tcm suites /status:<Operation ID> /clone /collection:https://<youraccount>.visualstudio.com/defaultcollection /teamproject:<SourceProjectName>

Clone operation 1 has succeeded. Summary information:
Started on:           3/01/2018 3:26:24 PM
Started by:           Eric Phan
Source suite id:      20105
Destination suite id: 20579
Test cases cloned:    446
Shared steps cloned:  24
Requirements cloned:  0
Completed on:         3/01/2018 3:28:36 PM

Now if we go over to the target project and query the test cases, we can see that they've all been imported.

Note: This only works when you want to migrate within the same VSTS instance/collection

It's a tedious process with many clicks involved, and I think most of the devs here follow the process, sometimes it is still missed. So I began exploring ways to automate the creation of this child testing task.

Thankfully Microsoft has a workflow engine called Flow that plays nicely with VSTS. So here's what you need to do:

Create a new connection to VSTS

1. Select Settings | Connections

1. Select New connection

1. Search for "Visual"
2. Click the "+" next to Visual Studio Team Services

1. Click Create

1. Sign into your VSTS account

Create a new Flow

1. Create a new blank Flow

1. Click on Search hundreds of connectors and triggers
2. Search for "Visual Studio Team Services"
3. Select When a work item is updated

1. Fill in the Account Name, Project Name and Type

1. Click on + New Step
2. Select Add a Condition

1. Setup the condition details by checking when the State is equal to Testing

1. In the If Yes box, select Add an action and look for Create a work item

1. Fill in the details to create a new task

In the above example, I've created a child task to the PBI with title "<PBI's ID> - Testing. It will have the same area and iteration path as the parent PBI and I've setup the relationship using the Link URL and Link Type.

Note: The dynamic content does provide a URL for the PBI, but it links directly to the revision of the PBI, so that URL in the Link URL won't work. That's why in the Link URL field I've had to hand craft the URL and put in the ID of the PBI. The link types are also as follows:

• Dependency-forward - Creates a "Successor" relationship
• Dependency-reverse - Creates a "Predecessor" relationship
• Hierarchy-forward - Creates a "Child" relationship
• Hierarchy-reverse - Creates a "Parent" relationship

Save & Test

You flow should now look something like this.

Last but not least, give the flow a name and click "Create Flow"

Now you just need to go and update a PBI's state to "Testing" or whatever state you want and see the flow in action. If you click into the flow, you'll see the history of whenever it ran. If you need to tweak things, you can click on one of the previous runs, and run the flow away (after you've made changes)

I would love to have an event on Pull Request completed, so I can even automate the change of state from Committed to Testing.

The way Exclaimer works is that it reads profile info from Azure AD and generates a signature during message transport and applies it to the message. It's actually quite a neat solution.

At SMEx, we have three offices and we include the address of the office in the signature. So instead of creating three different signatures in Exclaimer, we wanted one signature that can pull the address from the user's profile attributes.

In AzureAD we put each user into an AD Group by office so we just need to update the same address for all users in a group. Now whilst Azure AD provides a nice UI for updating profile attributes, it can become tedious if you need to update many users. 

The solution was PowerShell and the AzureAD module

All you need to do is:

1. Import the AzureAD powershell module using 
   1. Import-Module AzureAD
2. Connect to Azure Active Directory using
   1. Connect-AzureAD
   2. This pops open a Microsoft Live login window

1. Get a list of AD Groups and find the ID of the group to update
   1. Get-AzureADGroup

1. Update the profile attribute for all users in the group (in this case we are updating the Department field)
   1. Get-AzureAdGroupMember -ObjectId "<Id of the group>" | ForEach-Object -Process { Set-AzureADUser -ObjectId $_.Mail -Department "New Address/Value here" }

Below is a Gist you can use. All you need to do is replace the GroupID and the value to update. If you want to update an attribute other than Department you can see the list of available attributes here

• It's fast
• It's highly customizable
• And did I mention fast?

So as a .NET Core developer, here's my list of must have extensions:

• C# - you need this to get intelli sense, Go to definition, Find all references
• MS Build Project Tools - Gives you NuGet package intellisense in the *.proj files

• Visual Studio Team Services - Lets to integrate with TFS/Visual Studio.com and access Work Items
• vscode-icons - Turns VS Code from this

to this

I think these 4 extensions are the bare minimum, but the following are highly recommended:

• GitLens - Give you more git stats using CodeLens
• C# Extensions - Give you more refactoring and text expansion snippets for C#

Do you have other awesome extensions that you use? Let me know in the comment.

Note: We made a conscious decision not to automatically deploy to QA as the testers were being interrupted as when a developer commits code, the code got built and pushed to QA giving them a few minutes of down time.

So to save the tester roughly 10 minutes each morning waiting for environments to be deployed to QA, I decided to automate this process.

Unfortunately, Octopus Deploy doesn't have built in capabilities to schedule a deployment at a given time every day, but thankfully Octopus Deploy has a great API that will let you automate pretty much everything.

This is where the new Azure Functions and the Trigger come in handy.

1. In Microsoft Azure create a new Function App

2. Create a New Function

3. Select TimerTrigger - C# and fill in a Cron expression that represents when you want to run the function

4. We want to add the Octopus.Client as a NuGet reference, so select your function | View Files | Add | Add a new project.json file with the following content  

{
  "frameworks": {
    "net46":{
      "dependencies": {
        "Octopus.Client": "4.5.0"
      }
    }
   }
}

5. Now select the run.csx and paste in the code that will trigger the deployments

using System;
using Octopus.Client;
using Octopus.Client.Model;


public static async Task Run(TimerInfo myTimer, TraceWriter log)
{
    log.Info($"DeployToQA function executed at: {DateTime.Now}");
    log.Info($"Octopus URL: {GetSetting("OctopusUrl")}");
    var endpoint = new OctopusServerEndpoint(GetSetting("OctopusUrl"), GetSetting("OctopusApiKey"));
    var environmentId = GetSetting("OctopusQAEnvironmentId");

    using (var client = await OctopusAsyncClient.Create(endpoint))
    {
        var projects = await client.Repository.Projects.FindMany(x => !x.IsDisabled);
        foreach (var project in projects)
        {
            log.Info($"Checking project: {project.Name}");
            // Get the latest release
            var releases = await client.Repository.Projects.GetReleases(project);
            var release = releases.Items.FirstOrDefault();
            if (release != null) 
            {
                log.Info($"Latest release is {release.Version} with Id {release.Id}");
                // See if this release is already deployed to QA
                var deploymentQA = await client.Repository.Deployments.FindOne(x => x.ProjectId == project.Id && x.ReleaseId == release.Id && x.EnvironmentId == environmentId);
                if (deploymentQA == null)
                {                    
                    log.Info($"Deploying release {release.Version} to QA");
                    await client.Repository.Deployments.Create(new Octopus.Client.Model.DeploymentResource
                    {
                        EnvironmentId = environmentId,
                        ProjectId = project.Id,
                        ReleaseId = release.Id
                    });
                }
                else {
                    log.Info($"Release {release.Version} is already on QA");

                }
            }
        }
    }

}

public static string GetSetting(string name) =>  System.Environment.GetEnvironmentVariable(name, EnvironmentVariableTarget.Process);

This code loops through all the active projects, gets the latest release and checks to see if it has been deployed to the QA environment already. If it hasn't then it creates the deployment.

6. Next setup the application variables that the script uses. Click on the Function App in the left nav then select Application Settings

7. Add keys for:

• OctopusUrl - URL to your Octopus Deploy server
• OctopusApiKey - API key (Generate one from your profile page in Octopus Deploy)
• OctopusQAEnvironmentId - The ID of the environment (edit the environment, the ID is in the URL, it should be in the form "Environments-<id>"

8. Now you can test your function. Go back to the run.csx and click Run

Now every morning when the tester comes in, there's a fresh deployment for them to start testing on!

If you haven't decided to make the move to SSRS 2016 you should check out What's New in Reporting Services (SSRS). The main features we care about are:

• Better Portal
• Mobile Reports
• PowerBI integration
• Better HTML5 report rendering

As part of the migration process, we needed to assess which reports were actively being used so that we only migrated the useful reports and not the legacy baggage.

Thankfully, this is easily done as SSRS keeps an execution log for all the reports that it renders. The only thing you'll need to consider is how long that execution log tracks data for. The default is 60 days. If you want to update this to a longer period then the rule Do you keep track of which reports are being executed?. Thankfully at SSW, this was already set to retain the logs for 365 days.

So all that's left to do is write a query against the ExecutionLog table in the ReportServer database

WITH RankedReports
AS
(SELECT ReportID,
        TimeStart,
        UserName, 
        RANK() OVER (PARTITION BY ReportID ORDER BY TimeStart DESC) AS iRank
   FROM dbo.ExecutionLog t1
        JOIN 
        dbo.Catalog t2
          ON t1.ReportID = t2.ItemID
)
SELECT t2.Name AS ReportName,
       MAX(t1.TimeStart) LastAccessed,
       --t1.UserName,
       t2.Path,    
       SUBSTRING(t2.Path, 0, CHARINDEX('/', t2.Path, 2)) ParentFolder,
       t1.ReportID
  FROM RankedReports t1
       JOIN 
       dbo.Catalog t2
         ON t1.ReportID = t2.ItemID
 WHERE t1.iRank = 1
GROUP BY t2.Name, Path, ReportID
ORDER BY MAX(t1.TimeStart) DESC;

Credit goes to Russell Fox for this query (I've modified this to give distinct reports)

The error reported is "The job has been abandoned because agent <agent name> did not renew the lock. Ensure agent is running, not sleeping, and has not lost communication with the service."

Investigating the _diag folder of the build agent itself reveals the actual error.

---------------------------------------------------------------------------
The session for this agent already exists. Sleeping for 30 seconds before next retry. Attempts=1/10.
---------------------------------------------------------------------------
Microsoft.TeamFoundation.DistributedTask.WebApi.TaskAgentSessionConflictException: The task agent Agent-SYDTFSBLDP01-1 already has an active session for owner SYDTFSBLDP01.
   at Microsoft.VisualStudio.Services.WebApi.VssHttpClientBase.HandleResponse(HttpResponseMessage response)
   at Microsoft.VisualStudio.Services.WebApi.VssHttpClientBase.<SendAsync>d__79.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at Microsoft.VisualStudio.Services.WebApi.VssHttpClientBase.<SendAsync>d__76`1.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at Microsoft.VisualStudio.Services.WebApi.VssHttpClientBase.<SendAsync>d__65`1.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at Microsoft.VisualStudio.Services.WebApi.VssHttpClientBase.<SendAsync>d__64`1.MoveNext()
  at Microsoft.VisualStudio.Services.WebApi.VssHttpClientBase.HandleResponse(HttpResponseMessage response)
   at Microsoft.VisualStudio.Services.WebApi.VssHttpClientBase.<SendAsync>d__79.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at Microsoft.VisualStudio.Services.WebApi.VssHttpClientBase.<SendAsync>d__76`1.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at Microsoft.VisualStudio.Services.WebApi.VssHttpClientBase.<SendAsync>d__65`1.MoveNext()
--- End of stack trace from previous location where exception was thrown ---
   at System.Runtime.CompilerServices.TaskAwaiter.ThrowForNonSuccess(Task task)
   at System.Runtime.CompilerServices.TaskAwaiter.HandleNonSuccessAndDebuggerNotification(Task task)
   at Microsoft.VisualStudio.Services.WebApi.VssHttpClientBase.<SendAsync>d__64`1.MoveNext()

To fix this issue, you need to re-run the ConfigureAgent.cmd and on the first option "Do you want to also replace the server registration (default is N)" Choose "Y"

PS C:\Agents\2> .\ConfigureAgent.cmd
An existing configuration file was detected.  This will update the local agent settings.  Do you want to also replace th
e server registration (default is N)? Y

This will remove the existing agent registration and clean up the locked "session".

For the last month or so I've been living with my Visual Studio 2015 hanging for up to two minutes when simply right clicking on the web project and selecting Publish (before the publishing UI comes up)

Well today I got fed up with waiting and went to search for solutions. I stumbled across this Stack Overflow answer  from Mike.

Following the instructions:

1. Open the Properties of the web application you are trying to publish
2. Select Package/Publish SQL (click Enable This Page if available) 

1. Click Import from web.config
2. Uncheck all the options

Now right clicking and Publishing only takes a few seconds before the Publishing UI comes up

Last week I discovered that the solution wide NuGet package manage in Visual Studio 2015 has a "Consolidate" tab for this exact purpose. 

Simply go to Tools | NuGet Package Manager | Manage NuGet Packages for Solution...

Then click on the Consolidate tab

Select the out of date projects, then click Install

 For me the biggest positives were:

1. Quietness and smoothness of the ride (Yes after having kids, quietness is important! No hooning while the kids are asleep!)
2. The instant feedback when you push the accelerator and how easily it goes uphills
3. The 17” center console (I pretty much have maps + rear view camera on all the time)
4. The turn by turn navigation right next to the speedo
5. Space, lots of it for passengers and groceries and whatever else you need
6. Child seat anchors are nicely hidden so if you don’t use them you can’t even see that they’re there
7. The awesome reversing camera and markings that change based on how you rotate the wheel make it really easy to reverse.
8. The mobile app that allows you to turn the car on/off, lock, turn on climate control (air conditioning), see the charging status and where the car is parked
9. Cruise control is pretty smart too, slows and accelerates when needed
10. Michelle’s favorite – heated seats and the ability to turn on climate control before you enter the car, so it can be as warm or cool as you want

 There are a lot of little nice touches too like:

1. Mirrors turn downwards when you reverse
2. The lights on the handles when they extend out
3. Bluetooth phone connectivity just resumes playing when you get in the car
4. The charge port opening when the charger gets close

The negatives were:

1. It’s wider than you expect, so you have to be a bit cautious on the narrow Sydney streets or when passing between trucks :)
2. I had to reboot the car/consoles once during the two weeks when the 3G connection stopped working (still had many bars of reception, but it was not getting any internet connectivity, so google maps were limited)
3. The slowness of the home charger (~13km per hour) – so from flat it would take 24 hrs to fully charge. This would obviously not be a problem once more super charging infrastructure is in place.

 Improvements:

1. Auto speed detection by sign reading is a bit flakey. When driving in highways and freeways it incorrectly detects the exit lane speeds when you’re not exiting, causing the speeding warning to incorrectly chime.
2. If you forget to close the charge port and are driving, it should just close it for you. 
3. Having the heavy regenerative braking on sort of detracts away from defensive driving where you want to rest your foot on the brake, so that if there’s any surprises, your foot can slam on the brakes quicker. So I’m a bit up in the air about this feature. I’m wondering if there could be some sort of hybrid, where you can take your foot off the accelerator, the light regenerative braking kicks in, and when you apply some pressure to the brakes, the heavy regenerative braking kicks in.

Overall it’s an excellent car. It a lot wider than you expect, it drives quietly and smoothly. And the best part is, that it's brains are upgradable over the air thanks to it's 3G connection. I can't wait to take it for a spin again once Auto Pilot is rolled out.

Looking through the KendoUI docs and forums yielded plenty of results, but this was all for client side binding and formatting.

.ClientFooterTemplate("Sum: #= kendo.toString(sum, 'n0') #")

This didn't work for my scenario as I wasn't using AJAX or client side binding, I was using a local binding to an MVC model.

The trick was to fiddle with the .FooterTemplate fluent method to format the value. So instead of just doing:

.FooterTemplate(f => f.Sum)

You can do:

.FooterTemplate(f => { return (object)((decimal)f.Sum.Value).ToString("N0"); })

The trick here is that the FooterTemplate method takes an Func<GridAggregateResult, object>, so have to cast the GridAggregateResult to our number type, format it and cast it back to an object.

So these are static elements and not very informational. What I find when I'm using the Team Explorer window is that I'm usually in the My Work or Pending Changes hubs and that I rarely use the Home hub.

I don't see the home screen as being useful at all when I can also get to the hubs using the dropdown menu

What would make me use the Home hub more is if there was useful information presented on it:

What I like about this mockup is that it makes the Home hub more useful for a developer with live information without needing to go into each hub.

So at a glance I can know:

• what I'm currently working on,
• what changes I have to commit,
• how the last build went and if it's safe to get latest
• who the last person to checkin code was

I understand that having this live information will impact on performance of this window (which is what the VS team spent a lot of time trying to optimize in VS2012 and VS2013), but I feel that the information is useful enough for this to be an opt in option that can be turned on in Settings 

Agree? Disagree? I would love to know your thoughts...

EDIT: Added a user voice suggestion. If you agree please vote :)

I have been using Font Awesome over the last year for all my web projects. It's goal is to do away with all the images that you need for a modern web application and replace them with a single font.

Why replace images with a font you may ask?

There are several advantages:

• Fonts are scalable, so they look crisp at whatever size you want them
• You can use all your font styles on these icons to style them (like size and colour)
• All modern browsers support the download of custom fonts (there is a fallback to image sprites for old browsers)

Using it on your project is easy, just add the following stylesheet to your app:

<link href="//netdna.bootstrapcdn.com/font-awesome/3.2.1/css/font-awesome.css" rel="stylesheet">

Then you have access to over 350 icons. Here are a few that I use all the time: 

The HTML is easy, just <i class='icon-edit'></i>, <i class='icon-trash'></i> 

You can then add modifiers to control the size, rotation and even animation of the icon

• <i class='icon-cloud-download'></i>
• <i class='icon-cloud-download icon-2x'></i>
• <i class='icon-cloud-download icon-3x'></i>
• <i class='icon-cloud-download icon-4x'></i>
• <i class='icon-cloud-download icon-spin'></i>
• <i class='icon-cloud-download icon-rotate-90'></i>

Since it's a font, you can change the colour, simply by applying the "color" css style

You can even combine icons together: 

<span class="icon-stack icon-4x">
  <i class="icon-check-empty icon-stack-base"></i>
  <i class="icon-twitter"></i>
</span>

These icons also play nicely with Twitter Bootstrap. Here is a practical example where you change the save icon to a spinner to indicate the AJAX call is being made. Click the "Submit Changes" button to see it in action

<button id='submitButton' class='btn btn-default'><i class='icon-save'></i> Submit Changes</button>
 
<script type='text/javascript'>
  $(function() {
    $("#submitButton").click(function() {
     $("i", this).attr("class", "icon-refresh icon-spin");
      setTimeout(function() { $("#submitButton i").attr("class", "icon-save"); }, 2000);
    });
  });
</script>

You can use these icons to spice up your forms: 

<form>
  <div class="input-prepend">
    <span class="add-on"><i class="icon-envelope"></i></span>
    <input class="span2" type="text" placeholder="Email address">
  </div>
  <div class="input-prepend">
    <span class="add-on"><i class="icon-key"></i></span>
    <input class="span2" type="password" placeholder="Password">
  </div>
</form>

This library is a must for all web projects :)