I believe I’ve previously covered c# generics covariance and contravariance in the past, now it’s javas turn

As you may or may not know

The term PECS stands for “Producer Extends, Consumer Super,” which is an odd acronym coined by Joshua Block in his Effective Java book, but provides a mnemonic on what to do. It means that if a parameterized type represents a producer, use extends. If it represents a consumer, use super. If the parameter is both, don’t use wildcards at all—the only type that satisfies both requirements is the explicit type itself.

Covariance in java uses the extends keyword (yes even with interfaces), e.g. List<? extends Number> accommodates all types that derive from Number

Contravariance on the other hand uses the super keyword e.g. List<? super Number> accommodates all the types that Number derives from and of course number itself.

So what exactly is PECS recommending we do?

• Use extends when you only get values out of a data structure
• Use super when you only put values into a data structure
• Use the exact type when you plan on doing both

Tip

If you are debugging with VS2017/9 and want to pass environment variables to your container then read this post, if you are looking for picture of cats then sorry but leave a comment how you got here

Step 1

Create a new text file, the name doesn’t matter, but i called mine Dockerfile.env

Step 2

Add this file to your .csproj file.

Step 3

Not really a step but you you can simply query your Environment variable in the usual fashion (Environment.GetEnvironmentVariable())

Note:

Needless to say when you run in production you’ll need to pass the Environment variable according to Docker documentation which I don’t cover here

Hi Everyone,

I thought it worth sharing how to configure Azure Active Directory to work with a .net core 2.2 webapi backend and an angular7 front end that uses ADAL (i.e. v1 of Azure AD)

AD Versions

As you may or may not be aware, Azure AD has two implementations of security protocols, v1 is the common one but v2 is becoming more popular. From an Angular point of view you will pull in either the ADAL library for v1 or the MSAL library for v2, I’m not going to dwell on what the differences are or why to use either, in a recent project I was working on we found that there was no Java springboot support for v2 at the time, so went with the v1 endpoints to get our POC up and running quickly.

Asp Web Api

To configure Asp.net core 2.2 for use with v1 you’ll need a jwt token

Angular 7

For angular7 I used the adal-angular4 library (this is an unfortunate name as it is not limited to v4)

The application settings are configured in the environment

The module then adds adal and interceptors via the providers statement

Now when you make a http request the bearer token will be added by the angular interceptor and recognised by the webapi

The C# compiler defaults to the latest major version of the language that has been released. You may choose to compile any project using a new point release of the language. Choosing a newer version of the language enables your project to make use of the latest language features. In other scenarios, you may need to validate that a project compiles cleanly when using an older version of the language.

This capability decouples the decision to install new versions of the SDK and tools in your development environment from the decision to incorporate new language features in a project. You can install the latest SDK and tools on your build machine. Each project can be configured to use a specific version of the language for its build

Screenshot shows me selecting C# 7.2 for a .net core 2.1 application by changing the advanced options of the project properties build pane

Imagine the scenario you are on a team race, there are a number of stages along the route however only once all you teammates have gotten the the end of a stage can anyone proceed to the next stage.

Now imagine the competitors are threads/tasks and that you had to write this code…. with the .net Barrier class this is quite trivial.

The result looks like this:

In windows we have two types of semaphores, local and named system semaphores.

You can think of a semaphore as a bounder in a nightclub, the responsibility is to only allow a number of people into the club at any one time.

.net has a lightweight semaphore, ‘SemaphoreSlim’ that can be used for local communication (that is to say, system synchronization is not supported)

If you run the code above (e.g. in .net core 2.1 project) you will be presented with the following result

What is happening is that all the tasks try get access to the semaphore, they are all initially blocked until

is called which allows 3 tasks to enter at any one time.

Hi everyone,

Recently I found myself using Azure managed Kubernetes (AKS), however the images I wanted to pull were in AWS ECR. If my k8s cluster was in AWS it would be transparent to me provided the IAM user had permission but, in order to pull such an image from Azure; one can create a secret and to pull the image, sadly (or maybe thankfully) this secret expires after 12 hours so we need to keep refreshing.

Below I present an approach which could be used, it creates a service account (note I apply the permissions to the default account also as some of my deployments dont reference this service account yet) for pulling the image with RBAC, permissions, there is a kubernetes job that will execute immediately and a cronjob that will execute every 8 hours thereafter

Just use this secret in your deployments

Hope this is of benefit to others! remember to update those <TODO> sections!

So I’ve been looking for a serverless framework that can run on-prem and in the cloud, I’ve been leaning towards OpenFaaS as it appears to be gaining more traction, however I love Azure functions and though let’s see if this is a viable solution.

I download what is a Preview, so I wasn’t expecting miracles, I’m sharing the reasons why I can’t use it for my own requirements below.

It might save some of you guys the effort, I must reiterate that this is still a preview so some of the stuff I say here will be out of date really quickly!

I have decided against Azure Function On Prem in March 2017 because:

• It needs Sql Server, I can’t rely on having this at least not for some brown field projects I want to use serverless for.
• It needs IIS, I have to run on Linux (might be a solved problem… especially as it’s using the new .net core runtime )
• It only has Javascript and C# language support in preview, I need Java, and Go and Python would be nice to haves
• The packaging was a windows installer, I was hoping for some docker images, I expect this is a solved solution and for now the MSI is a quick win for the developers.

Next it’s down the rabbits burrow with OpenFaas on Kubernetes, cross your fingers for me!

Aside from the above which are mostly external limitations it’s nice to see Azure Functions Running locally

I’ve been quite vocal in the past that the Java language syntax is low on my list of favourite languages. I believe this stems from the fact that I write lots of C# (and even Typescript) these days, and those languages are a lot nicer to work with.

Lately I invested a little time in other languages, Go and Kotlin for the JVM.

While I’m yet to formulate my view on GoLang  I absolutely fell in love with Kotlin from jetbrains; if you work on the JVM I encourage you to check it out, It is simply awesome!

I’ve been thinking a lot about the state of software development recently and moreover taking a step back trying to identify and articulate where I stand on past, present, future trends or concepts and define how to employ these ideas and approaches to become more effective in delivering successful production software.

Now I’m as guilty as the next person of Resume Driven Development, for years I’ve chased that shiny new technology; why? well it was current, it was bleeding edge or at least cutting edge, I gorged on technical articles and videos and spend many a long night trying things out, in fact, I still do and I can’t see that ever changing.

Experience vs. Skill vs. Attitude

However this can lead to having a lot of experience, but experience is cheap (as all that’s needed is time), it doesn’t necessarily lead to skill, this is harder this you have to practise at but a lot of people can do that. But attitude.  You either have it, or you don’t.  The right sort of person is so passionate about coding that can’t be stopped from doing it.  They typically started before high school and never looked back.  They write everything from assembly to JavaScript, on PCs to mobile phones, doing hard core computer graphics to high level social networking. They’ve tried everything!

For these type of people:

• Learning becomes easier – The more one learns the easier new information takes root, it’s simple, you’re training your brain, you’ll find that you will develop an insatiable appetite for yet even more information, for tips, tricks, patterns, practises, anti patterns, frameworks… you name it…
• Value identification – Because Netflix, Amazon, Microsoft [insert big name brand here] says it is shiny is no longer good enough! We’ve all been through this technology hype cycle where, something new comes out, everyone is really excited then one starts using it only to fall into the trough of disillusionment.  

  Once you’ve been through this cycle a few times you know what to expect and you will better qualify where you lie on this trend line when evaluating for example a new technology stack.

My Views

Tooling

I grew up in Microsoft Land, if you take away my tools you’ll take away my productivity. Tooling in my view is critical to being productive. But tooling can be bad, tooling if not understood can abstract you so far away from the underlying technology you may not understand how it works, this could make a really trivial problem very hard to resolve.

The CTO of Expensify once wrote an article something along the lines of “Why I will never hire a .NET developer”. From what I recall he faced some serious backlash on that post, but I understood the point he was trying to get across. I have watched many Asp.NET webforms developers struggle when that community started embracing ASP MVC. For those that don’t know ASP Web Forms to some degree abstracted away web applications so much with the likes of ViewState and Postbacks, that a junior dev would be forgiven for thinking he/she was writing a VB6 application.

The point I’m getting to is that, if you understand how the underlying technology works, then you are in a much better position to leverage any tooling that increases productivity.

DevOps

In my view everyone should be aware of what DevOps means, for a long time developers worked away in their own individual boxes and likewise operations, only to knock heads and point fingers when trying to deploy software. I don’t think the industry can afford to continue with this approach, everyone developing software should be at least aware of what is involved in operations, only by seeing the struggles from another viewpoint can we all work together to streamline this process and improve our products and processes.

Language X versus Language Y

I’ve forgotten more languages than I know at this stage, but what I do know is languages are easy. I think I'm qualified to say that after writing a few applications in objective-C (can I remember much of the syntax? Nope! but I have the github repos to prove it! and give me a few hours and I'll remember it all).

What is hard with a language is; its eco system and frameworks, therein is where the effort lies. I grew up in C++ and loved it I gorged on it however when .NET arrived on the scene I knew its days were numbered (at least for the type of applications I was interested in). To this day C# is my favourite language. That said I also use Java day to day, and while I firmly believe it’s a less coherent language, the java ecosystem is much richer than its .NET counterpart. Which would I choose for a new project? This depends on the task I’m trying to achieve but moreover the team with whom I’d be working with and their technical capabilities. It’s good to expose yourself to different languages which will force you to learn different approaches to achieving an end goal, be a polyglot!

Try pick the Best language for the Job, keeping in mind there are many interpretations of what best means!

Static vs Dynamic Languages

A few years ago, after I’d heavily invested in my Javascript knowledge, I ended up writing a high level design document for a rather sophisticated Web application framework, I distinctly remember deliberating to no end on whether to insist on JavaScript or Typescript, the exact phrase I remember leaving my lips was “My heart says JavaScript, my Head says Typescript”, well with Hindsight, I’m glad to say my head was proven correct, I ended up prototyping and working on the initial version of that product and I firmly believe that Typescript has made large scale web applications more tolerable. I think dynamic languages are great for POC’s and early versions of products, why you can nearly type in the code and see the results in real time. However dynamic languages are harder to support, tooling is more difficult and a compile is your first unit test!

SOAP vs REST

Unfortunately, there is a lot of misinformation and misconceptions around REST, in fact they can’t really be compared directly given one is a protocol and the other an architectural style. But If i was to go out on a limb on the general expectation of what this comparison means, then I would say choose REST.

I listened to a talk from the Uber CTO talking about what he learned after writing 1000 microservices, and one of his points was the REST approach requires tooling that supports building and documenting the API. One such example I have experience with is Swagger. Tooling always wins out in my book, I don’t want if at all possible to have to learn the Swagger syntax, much the same way for these last 20 years I’ve never explicitly learned WSDL, once I understand how it works and what it’s for I’ll use a tool to generate it faster and with less errors that I could ever hope to achieve.

Clouds…

For years I’ve been saying, “bet on the web”, now I say “bet on the cloud”.

I think public clouds are the single biggest game changers in todays software industry, by lowering the barrier to entry, companies can get off the ground quickly, with limited resources and short times to market. My preference for Azure (where possible) is no secret.

A client of mine once told me how back in the .COM boom he was paying 100k $ a month to have a geo-redundant solution that involved a container up some mountain some where in the US. With the cloud the infrastructure already exists, and with a flip of a switch it’s possible, sometimes with no changes to the underlying application itself.

But sadly cloud can be used in all the wrong ways and the costs can quickly ramp up. I would suggest the following guidelines when looking at the cloud

• If you are lifting and shifting, then you are probably doing it wrong
• If you stay in IaaS you are missing out on the advantages of PaaS
• If you use serverless, then use it correctly, don’t turn your application into one giant RPC call just to be sexy.

Understand your Business

You may have very smart people in your business that can’t innovate as they don’t understand the big picture, identify these people and ensure they are in a position to contribute. Developers and business need to speak the same language if innovation is to be expected.

Be aware of the innovators dilemma and don’t stagnate,.

Microservices

How many times have you heard people moving to microservice architecture to escape the problems they have with their current monolithic approach?

Yes microservices can help one avoid some of the pitfalls of monolits however I would argue that if someone can’t write a good monolit then they won’t write a good set of microservices. Some backgrounds like the Microsoft stack naturally fit well into Microservices while other like J2EE are in stark contrast.

The Promise

Microservices are certainly one of the areas where people encounter the hype cycle in early attempts. They break up big application in the hope of:

• having greater flexibility and scalability
• having to ability to deploy services independently with different teams
• have more agility

The Result

• It is more brittle than ever before
• Performance is terrible
• Services needed to be deployed together and in specific order
• It was impossible to follow the flow of messages through the system.
• Distributed systems are HARD
• Eventual consistency is a paradigm shift
• Legacy habits created a distributed “big ball of mud”

I’m not looking to frighten anyone, I’m pro microservice but one needs to understand what they are buying into and what approaches exist to overcome these limitations.

Architecture

Architecture is never just about the technology, it’s about the team, are they agile, do they “do agile” or neither, is there a DevOps culture, what are the underlying business decisions driving the change?

Keep the follow in mind when architecting a solution:

• Keep it simple
• Don't build what you don’t need
• Don’t build what you might need
• TCO and ROI are important
• Research new approaches,
  e.g. I’d rather drink battery acid than do 2 phase commit, there are other approaches…. research them!

Always be learning!