Testing Automation: The Big Picture

It’s often useful to take a step back and look at the bigger picture, this is true in different aspects of life such as health or wealth or relationships, and is also true of software development.

When it comes to creating automated tests (as with other aspects of software development) dogmatism and absolutist schools of though can exist.

As with all things, the decision to write tests (and how many tests, what type of tests, test coverage aims, etc.) ultimately should boil down to one question: do they add value to what you are doing?

To be clear, I absolutely believe in the creation of automated tests in many cases, however it is good to not be dogmatic. For example if there is a niche market that is ripe for capitalizing on, and time-to-market is the most important thing to capture this market, then an extensive suite of automated tests may slow down getting to that initial release. This of course assumes that the potential market will have some tolerance for software defects. It also depends on what the product is; medical life-critical software is probably going to have a higher quality requirement than a social media app for example. This can be a trade-off however with shorter term delivery speeds being quicker but at the expense of delivery speed in the long term, if you’re overwhelmed fixing production outages you have very little time to add new features/value.

Another aspect to consider is that of risk. What are the risks associated with defects in the software making their way into production? Different features/application areas may also have different risk profiles; for example  the “share on social media” feature may not be deemed as important as a working shopping cart. It’s also important to remember that risk is not just monetary, in the previous example a broken “share on social media” feature may bring the business into disrepute, aka “reputational risk”.

When it comes to the myriad of different types of tests (unit, integration, subcutaneous, etc.) the testing pyramid is an oft-quoted model of how many of each type of tests to have in the test suite. While the testing pyramid may be of great help for someone new to automated testing to help them learn and navigate their initial steps, as experience grows the model may no longer be optimal to some of the projects that are being worked on. Beyond the testing pyramid the different aspects of test types can be considered such as execution speed, breadth/depth, reliability/brittleness etc.

Automated tests also do not exist in and of themselves, they are part of a bigger set of processes/considerations such as pair programming, code reviews, good management, well-understood requirements, good environment management/DevOps, etc.

If you want to take a step back and look at the big picture, or know a developer or manager who wants to understand the trade-offs/benefits check out my Testing Automation: The Big Picture Pluralsight course.

Installing and Configuring a UPS with Windows 10

The power grid at my home office is sometimes unreliable. In the last year there have been several power outages ranging from several hours to transient outages of a few seconds.

My main PC is a silent desktop which means (unlike a laptop) has no resilience to power outages. This is where an external uninterruptable power supply (UPS ) comes in handy. A UPS is like a big battery that gets charged from main power, but when there is an outage the battery kicks in and provides power to connected devices.

When choosing a UPS, generally speaking, the more expensive they are the longer they can provide power to connected devices (or provide the same or more power for a longer time). Having researched typical power consumptions and armed with the knowledge that my PC uses SSDs, is fan-less, and does not have a high-powered GPU something in the 1000-1500 VA range was required. My main requirement was the ability to save work and safely shut down the PC before loosing all power. If I had more complex energy requirements I would have first purchased a power meter to measure how much power my devices were using before choosing the UPS.

Installing the UPS

Once I understood the approximate power requirements, I selected the Smart-UPS 1000VA LCD 230V from APC.

APC Smart-UPS 1000VA LCD 230V Front

The rear of the unit consists of a number of sockets, for the outputs there are 2 distinct groups that can be configured differently. For example the main group can have the PC and one of the monitors connected. The secondary power group can have non-essential devices attached, in my case the powered studio monitor speakers. The UPS was configured to turn off the secondary group if there is a power outage but after a 15 second delay; this will help with short transient outages. The main group will remain on for as long as the UPC battery has power. There are many different ways to configure this UPS by using the LCD display and front buttons, including the ability to enable/disable audible notifications.

APC Smart-UPS 1000VA LCD 230V rear panel

The rear panel also has a USB connector that can be connected to the PC.

Configuring Windows 10 Power Settings for a UPS

Once everything had been connected (including the UPS-PC USB cable) it can all be switched on. Once the UPS battery is fully charged, turning off the main power at the power outlet causes the UPS to switch to battery power, the PC and single monitor stay on and 15 seconds later the speakers turn off (to save power). Once on battery, the PC and monitor draw about 50W of power and the estimated run time according to the UPS LCD  is about 2 hours.

Because of the USB connection to the PC, Windows 10 also displays the battery status in the system tray:

Windows 10 UPS Battery status

If a power outage occurs when using the PC, once the battery gets low I can manually shutdown the PC to prevent damage/data loss. If however the PC is unattended at the time, I still want it to be safely shutdown.

To do this in Windows 10, type “power & sleep settings” in the start menu and configure the basic settings. From here, “Additional power settings can be opened” and power plans customized. Once in the plan settings, the “Change advanced power settings” can be selected to get down into the fine details:

Opening additional power settings

Once in the detailed Power Options dialog, in additional to other settings I opened the Battery section and changed the critical battery level to be 20% and the critical battery action to Hibernate. I also modified some o the other battery settings. 20% for critical is very conservative as I wanted to hibernate the PC with plenty of UPS battery capacity remaining.

Hibernating Windows 10 when critical UPS battery level reached

Now when on UPS battery power, Windows will warn me when the battery gets low (I configured this to be 40%) and then auto-hibernate at 20% battery.

It’s a great feeling to know that even when there is a storm blowing in that I will not loose any work and that my main PC hardware is protected from surges and unexpected power loss.

The Joys of Silence

I recently took possession of new desktop PC. As a Pluralsight author one of the unique considerations when choosing a new machine was generated noise. After some research I settled on sourcing the computer from a UK company called QuietPC.com.

NoFan PC case from front

Initially when I started researching what to buy I had assumed that I would pay some performance penalty as I wanted a CPU with fan-less cooling and I didn’t want the additional complexity of something like water cooling.

As the following image shows, I was able to get a quad core Skylake (Core i7 6700K 4.0GHz) without requiring a fan:

Nofan CR-95C Pearl Black IcePipe 95W Fanless CPU Cooler

The humongous thing in the preceding image is a Nofan CR-95C Pearl Black IcePipe 95W Fanless CPU Cooler. This fan-less CPU cooler is based on thermal heatpipes that are able to transfer heat away from the processor, the heat then being dissipated by the huge surface area of the “fins”.

Nofan CR-95C Pearl Black IcePipe 95W Fanless CPU Cooler

The case itself is a Nofan CS-80 Fanless Computer Case coupled with a Nofan P-500A Silent 500W Fanless 80+ GOLD PSU. The case features a vent at the top of the case above the CPU cooler to aid in convection.

In use, the PC is completely silent, no fan noise or electrical hum – the only noise emitted is if you are used the optical drive.

Prior to this machine I was using and was very happy with a Lenovo laptop though under load the fan noise was becoming a little distracting – in all fairness it was an aging machine that had been an 8 hour a day workhorse for a few years.

It’s amazing that once you’ve experienced the joy of completely silent computing, going back to using machines with fans seems archaic. No doubt we’ll eventually have silent, high-performance, fan-less laptops – though the small form factor will present some hard thermal dissipation challenges.

Full hardware specs:

  • Nofan CS-80 Fanless Computer Case
  • Gigabyte GA-Z170XP-SLI LGA1151 ATX Motherboard
  • Intel 6th Gen Core i7 6700K 4.0GHz 91W HD 530 8MB Quad Core CPU
  • Corsair DDR4 Vengeance LPX 32GB (2x16GB) 2400MHz Memory Kit
  • Nofan CR-95C Pearl Black IcePipe 95W Fanless CPU Cooler
  • Nofan P-500A Silent 500W Fanless 80+ GOLD PSU
  • Samsung 950 PRO M.2 512GB NVMe SSD
  • Samsung 850 EVO 250GB 2.5in Solid State Drive
  • Pioneer DVR-221LBK DVD and CD Reader/re-writer
  • Gigabyte Dual Band Wireless-AC GC-WB867D-I Wi-Fi/Bluetooth Card
  • Gigabyte GC-TPM Trusted Platform Module

Winning in 2016

As the year 2015 starts its last slide into 2016, it’s the time of year that I start to think about what my 3 Wins are going to be for next year.

If you’re not familiar with the 3 Wins concept, it’s similar to goal setting but rather than focus on “what will I tick off my todo list” it’s more along the lines of “what will make me feel great, like I’ve accomplished something, like I’ve made progress…”. One way to help come up with three wins is to imagine how your future self will feel when you look back on the year and have accomplished all your 3 Wins.

It’s important to make your 3 Wins achievable, otherwise not achieving any of them could be disheartening and demotivating.

One technique to guide you, while being sounding very “managementy”, is the concept of SMART criteria.

SMART is an acronym that stands for:

  • Specific
  • Measurable
  • Attainable
  • Realistic
  • Time-related (or Time-bound)

Using SMART may help if you are struggling to create your own 3 Wins.

You might have 3 Wins for your work/career and you may also have 3 Wins for personal/health/family/etc. related things. Also you can use 3 Wins daily, weekly, monthly, quarterly, or at any scale you wish; you could even have decade or lifetime 3 Wins.

Reducing MVVM Light Viewmodel Code with Fody Property Dependencies and Custom Property Changed Methods

In other previous articles I’ve written about Fody and this article we’ll see how we can use the PropertyChanged.Fody addin with MVVM Light viewmodels.

Windows Store app screenshot

The preceding screenshot shows the app in action: a name TextBox is twoway data-bound to the viewmodel, when a different name is typed, the TextBlock shows the uppercase version.

The user can also type a color into the color box and a bit of reflection happens to convert this string to a color from the Colors class.

Windows Store app screenshot with different bg color

The following code shows the (non Fody-fied) viewmodel:

More...

Say Goodbye to Boring INotifyPropertyChanged Implementation in Universal Windows Apps with Fody

We need to implement INotifyPropertyChanged to allow bindings to be updated. For example in a simple viewmodel to add 2 numbers:

namespace UniFodyDemo
{    
    internal class CalculatorViewModel : INotifyPropertyChanged
    {
        private ICommand _addCommand;
        private string _firstNumber;
        private string _secondNumber;

        public CalculatorViewModel()
        {
            FirstNumber = "5";
            SecondNumber = "10";
            Result = "";

            AddCommand = new RelayCommand(Add);
        }


        // Use strings to simplify demo code (so no val converter)
        public string FirstNumber
        {
            get { return _firstNumber; }
            set
            {
                _firstNumber = value;
                OnPropertyChanged();
            }
        }

        public string SecondNumber
        {
            get { return _secondNumber; }
            set
            {
                _secondNumber = value;
                OnPropertyChanged();
            }
        }

        public string Result { get; set; }

        public ICommand AddCommand
        {
            get { return _addCommand; }
            private set
            {
                _addCommand = value;
                OnPropertyChanged();
            }
        }

        public event PropertyChangedEventHandler PropertyChanged;

        private void Add(object obj)
        {
            var a = int.Parse(FirstNumber);
            var b = int.Parse(SecondNumber);

            Result = (a + b).ToString();
        }

        protected virtual void OnPropertyChanged([CallerMemberName] string propertyName = null)
        {
            var handler = PropertyChanged;
            if (handler != null) handler(this, new PropertyChangedEventArgs(propertyName));
        }
    }
}

There’s more code here than we’d need if we didn’t have to implement INotifyPropertyChanged. Really the important things are the 3 properties that we bind to and the add command / method.

The XAML in both the Windows Store app and Windows Phone app looks like:

<Grid Background="{ThemeResource ApplicationPageBackgroundThemeBrush}">
    <StackPanel Width="200">
        <TextBlock Text="{Binding FirstNumber, Mode=TwoWay}"></TextBlock>
        <TextBlock Text="{Binding SecondNumber, Mode=TwoWay}"></TextBlock>
        <Button Command="{Binding AddCommand}">Add</Button>
        <TextBlock Text="{Binding Result, Mode=TwoWay}"></TextBlock>
    </StackPanel>
</Grid>

Without implementing INotifyProperty changed, clicking the Add button won’t be of much use, we wont see the result.

The solution looks like the following, the shared project contains the viewmodel.

image

It would be great not have to write this same INotifyProperty stuff for everything we want to bind to/from. Using a NuGet package called PropertyChanged.Fody we can achieve this.

First this package needs to be installed into both the Windows Store and Windows Phone app projects. This is because the shared code in the viewmodel isn’t compiled in that shared project, rather the source code is compiled individually in both the Store/Phone projects.

Now the viewmodel can be simplified:

namespace UniFodyDemo
{
    [ImplementPropertyChanged]
    class CalculatorViewModel
    {
        public CalculatorViewModel()
        {
            FirstNumber = "5";
            SecondNumber = "10";
            Result = "";

            AddCommand = new RelayCommand(Add);
        }

        private void Add(object obj)
        {
            var a = int.Parse(FirstNumber);
            var b = int.Parse(SecondNumber);

            Result = (a + b).ToString();
        }
        

        // Use strings to simplify demo code (so no val converter)
        public string FirstNumber { get; set; }
        public string SecondNumber { get; set; }
        public string Result { get; set; }

        public ICommand AddCommand { get; private set; }
    }
}

Notice the [ImplementPropertyChanged] attribute which means INotifyPropertyChanged will be automatically implemented for us. Notice that it’s much easier to see what’s essential in the viewmodel without all the noise of INotifyPropertyChanged.

If we decompile the Windows store app we can see the implementation:

namespace UniFodyDemo
{
    internal class CalculatorViewModel : INotifyPropertyChanged
    {
        public ICommand AddCommand
        {
            get
            {
                return this.u003cAddCommandu003ek__BackingField;
            }
            private set
            {
                if (this.u003cAddCommandu003ek__BackingField == value)
                {
                    return;
                }
                this.u003cAddCommandu003ek__BackingField = value;
                this.OnPropertyChanged("AddCommand");
            }
        }

        public string FirstNumber
        {
            get
            {
                return this.u003cFirstNumberu003ek__BackingField;
            }
            set
            {
                if (String.Equals(this.u003cFirstNumberu003ek__BackingField, value, 4))
                {
                    return;
                }
                this.u003cFirstNumberu003ek__BackingField = value;
                this.OnPropertyChanged("FirstNumber");
            }
        }

        public string Result
        {
            get
            {
                return this.u003cResultu003ek__BackingField;
            }
            set
            {
                if (String.Equals(this.u003cResultu003ek__BackingField, value, 4))
                {
                    return;
                }
                this.u003cResultu003ek__BackingField = value;
                this.OnPropertyChanged("Result");
            }
        }

        public string SecondNumber
        {
            get
            {
                return this.u003cSecondNumberu003ek__BackingField;
            }
            set
            {
                if (String.Equals(this.u003cSecondNumberu003ek__BackingField, value, 4))
                {
                    return;
                }
                this.u003cSecondNumberu003ek__BackingField = value;
                this.OnPropertyChanged("SecondNumber");
            }
        }

        public CalculatorViewModel()
        {
            this.FirstNumber = "5";
            this.SecondNumber = "10";
            this.Result = "";
            this.AddCommand = new RelayCommand(new Action<object>(this, CalculatorViewModel.Add));
        }

        private void Add(object obj)
        {
            int num = Int32.Parse(this.FirstNumber);
            int num1 = Int32.Parse(this.SecondNumber);
            this.Result = (num + num1).ToString();
        }

        public virtual void OnPropertyChanged(string propertyName)
        {
            PropertyChangedEventHandler propertyChangedEventHandler = this.PropertyChanged;
            if (propertyChangedEventHandler != null)
            {
                propertyChangedEventHandler(this, new PropertyChangedEventArgs(propertyName));
            }
        }

        public event PropertyChangedEventHandler PropertyChanged;
    }
}

So not only with Universal projects can we share code via the shared code project, adding Fody into the mix is even more awesome.

For a list of available addins or to contribute (or even create your own addin) check out the GitHub project site.

Also to learn more about Fody feel free to check out my Pluralsight course.

Write Less Repetitive Boilerplate Code with Fody

My newest Pluralsight course on Fody was just released.

Fody is a tool that frees us up from having to write repetitive boilerplate code. For example we can automate IDisposable implementation or automatically add Equals(), GetHashCode() and equality operators.

Fody is not one tool, but a collection of individual “addins” around the core Fody, with the core GitHub members: Simon Cropp, Rafał Jasica, and Cameron MacFarland.

Automatically Implementing Equality

As an example of what’s possible with Fody, we can auto-implement equality.

So first off we install the Equals Fody addin via NuGet:

PM> Install-Package Equals.Fody

Now we can add the [Equals] attribute to indicate that we want to auto-implement equality.

We can opt-out specific properties so they won’t be considered during equality by applying the [IgnoreDuringEquals] attribute.

We can also provide custom equality logic by creating a method and decorating it with the [CustomEqualsInternal] attribute. So for example if either altitude is –1 then consider altitudes equal.

The following code shows these attributes in action:

[Equals]
public class Location
{
    [IgnoreDuringEquals] 
    public int Altitude { get; set; }
    
    public int Lat { get; set; }
    public int Long { get; set; }

    [CustomEqualsInternal]
    private bool CustomEqualsLogic(Location other)
    {
        // this method is evaluated if the other auto-equal properties are equal

        if (this.Altitude == other.Altitude)
        {
            return true;
        }

        if (this.Altitude == -1 || other.Altitude == -1)
        {
            return true;
        }

        return false;
    } 
}

For a list of available addins or to contribute (or even create your own addin) check out the GitHub project site.

To learn more about how Fody works, how to use some of the available addins, and how to create your own check out my Pluralsight course.

Kill CapsLock with AutoHotkey

capslock key on a keyboard

How often do you actually use CapsLock – deliberately that is?

Most of the time I hit it accidentally then END UP SHOUTING in emails or Tweets. It’s especially a pain sometimes while coding…

We can stop CapsLock from doing it’s capsy thing using AutoHotkey.

The following remaps the CapsLock key to nothing:

Capslock::

Now the CapsLock key doesn’t do anything, the caps LED on the keyboard doesn’t even light up.

We could also remap CapsLock to F6, so we can hit CapsLock in Visual Studio to do a build:

Capslock::F6

This will trigger F6 in all applications, not just in Visual Studio.

To limit the remapping to only when we are in Visual Studio:

SetTitleMatchMode, 2
#IfWinActive Microsoft Visual Studio
Capslock::F6
#IfWinActive

Here the remapping only occurs if the window title contains “Microsoft Visual Studio”. (The SetTitleMatchMode command tells AutoHotKey to look in any part of the title for the match.

Now, outside of Visual Studio, CapsLock behaves normally, but inside VS it will perform a build.

 

To learn more about AutoHotkey, check out my Pluralsight course: Personal Productivity & Performance Tools for Windows Developers.

Doing an Internet Search for the Current Word in Visual Studio with AutoHotkey

I got bored in Visual Studio double clicking a class, copying to clipboard, then heading to browser then pasting into search box.

I wanted to be able to quickly search for something (e.g. in Google) for whatever word the cursor was currently in. Also in other applications.

I have a Microsoft keyboard that has a calculator key that normally opens the Calculator application. I’ve never ever used this and hadn’t even noticed it until now.

image of the calculator key on my keyboard

To map this key in AutoHotkey, “Launch_App2” can be used, so to trigger a script whenever it’s pressed: Launch_App2::

I also wanted the default behaviour to automatically go to the first result. Pressing CTRL-calckey just does a normal search and lists all the results for me to peruse.

Here’s the script – note that it’s a bit rough and ready, for example it uses the clipboard (CTRL-C) so this will overwrite anything you have in it. It also won’t work if the cursor is at start of line, etc.

Launch_App2:: ; Do an I'm Feeling Lucky Google search when calc key pressed
  Clipboard := 

  SendInput, ^{LEFT}^+{RIGHT}
  SendInput, ^c
  ClipWait, 1

  if !(ErrorLevel)  { 
      Run, % "https://www.google.com/search?hl=en&btnI=I%27m+Feeling+Lucky&&q=" Clipboard
  }
return



^Launch_App2:: ; Do a normal Google search when calc key pressed
  Clipboard := 

  SendInput, ^{LEFT}^+{RIGHT}
  SendInput, ^c
  ClipWait, 1

  if !(ErrorLevel)  { 
      Run, % "https://www.google.com/search?hl=en&q=" Clipboard
  }
return

Now if i have my cursor in the word “ActionResult” in Visual Studio, hitting the calculator key takes me to MSDN http://msdn.microsoft.com/en-us/library/system.web.mvc.actionresult%28v=vs.118%29.aspx as this is the first Google search result for “ActionResult”.

 

If you’ve never used AutoHotkey before, to get started check out my Pluralsight course.

Using Glide Paths to Increase Productivity

I first became aware of the concept of “glide paths” in Getting Results the Agile Way. A glide path is an aeronautical term meaning: “the course followed by an aircraft or spacecraft when descending for a landing.” – Dictionary.com

Usually this refers to a low/no power descent which requires little effort.

Applying this concept to the work we do essentially means doing the harder/boring/tougher/difficult/unpleasant things earlier in the week so that towards the end of the week we are “gliding”, with less effort required as we’ve already done the worse tasks.

In addition to applying glide paths to the week, we can apply this to individual days: doing the tougher things at the start so as we get closer to finishing for the day, our tasks get easier.

The same thinking can also be employed by agile teams, for example doing the harder things earlier in the sprint.

The following diagram shows as we get more tired during the week, how the duration of a given task may increase:

image

This is not always the case obviously, everyone has different energy patterns, but becoming more self-aware we can apply ourselves to different kinds of work at the times where we get get the most output. Also it’s interesting that it may be better to do tasks requiring creativity when we’re tired.

Glide paths assume that we have some autonomy over the prioritisation of our tasks, even when we do have autonomy we my have little choice sometimes about when things get done. Even so, glide paths are a useful concept to have in our productivity toolboxes.