I’m running on a Windows Insider Slow build so that I can leverage WSL 2, the Windows Subsystem for Linux v 2. Its pretty incredible, because there’s now a Linux kernel inside Windows. Ubuntu is fast, its a wonderful development experience all my favorite linux tools. I can’t wait for this to be out of preview this year and in the mainstream windows releases.
I’m also using the latest version of Docker Desktop, with WSL2 support. What this means is that instead of using Hyper-V to run a Moby Linux VM, docker runs directly on WSL2. It also has built-in Kubernetes support.
This blog was posted as part of the Third Annual C# Advent. Make sure to check out everyone else’s work when you’re done here
Dependency Injection, or DI, is a Software Architecture Design Pattern. DI is something that comes up during discussions on SOLID, IoC (Inversion of Control), testing, and refactoring. I want to speak on each of these briefly because DI touches all of these. But before I really dive into things, I want to define what a dependency is. A dependency is any object that another object requires. So all of those classes, services, and libraries that we use to build our applications are dependencies.
I have been starting to speak and present a lot more, and was looking into great tools like Carnac and KeyPosé. But I just found out today about a feature I didn’t know existed inside Visual Studio Code, Screencast mode. This was introduced in January 2019. How did I miss it?
To enable and use Screencast mode, Open the command palette, Ctrl + Shift + P.
When first enabled, Screencast Mode is not what I wanted. It shows EVERY keypress. It also shows a little higher on the screen that I prefer. It also puts a little red circle everywhere I click the mouse, which is nice. Let’s configure it and see if we can clean up some of that. Open the command palette again (Ctrl + Shift + P) and go to the user settings.
I’ve been a developer for a long time, writing APIs and clients to consume them. When an API is around long enough, it needs to change. I’ve versioned APIs in the past using a number of different techniques. Some successful, some painful. Now I realize this discussion is like the VI/Emacs conflict, the Tab/Space wars, and the Spanish Inquisition, but it is a good topic to look at. There are a few main styles when it comes to API versioning:
I’ve got a template straight from Microsoft. I want this wired into a CI/CD pipeline to I can play around and get quick feedback. I’m going to use Azure DevOps to help make all this possible. Let’s get those templates into a repository to get started. New repository, initialize it, add new files.
Next, I’m going to create a new resource group to play around with my web app resources.
Time to Dive in # I’m one of those guys that likes to learn by doing. Reading the documentation is great, and I do that a lot. But for me to really grok something, I need to play with it, run it, and probably blow it up.
If you missed part 1, read along and come back. I need a WebApp setup for my sample project. I realized I can do it a few ways. Some of the ways are very manual, some are repeatable, but one stood out to me.
