Creating a CI/CD pipeline for a .NET library: Part 1 - Intro

joaofbantunes

João Antunes

Posted on July 31, 2018

Creating a CI/CD pipeline for a .NET library: Part 1 - Intro

In this series of posts I want to talk about creating a CI/CD (continuous integration/continuous delivery) pipeline for a .NET library that’s distributed as a NuGet package. As reference, I’ll use the GrpcExtensions library that I talked about in a previous post.

Given we’re in this new world of cross platform .NET, I want to build and test stuff in more than just Windows, so I’m using AppVeyor and Travis CI, where the former handles building on Windows and the latter handles the Linux and MacOS work.

Considering the use of more than one provider for the CI/CD pipeline, Cake is useful to define the required tasks in a provider and system agnostic manner. It’s also cool that we can define the build tasks using C#, which is rather easy to understand given the rest of the project is already C#.

Finally, to have visibility on the tests code coverage I’m using Coveralls to host this information.

It would be also good to add some static code analysis to the mix, like SonarQube, but haven’t done it yet, maybe a good topic for another post.

Disclaimer:
I’m assuming the reader understands what I’m talking about from the very small intros I provide, but if I’m assuming wrongly (which is very possible) feel free to ask and I’ll answer and eventually improve the posts to take into account.

Going with a series of posts instead of a giant one to be (hopefully) easier to read.
The following is the list of posts for easy navigation.

The accompanying code for these posts is here (tagged to be sure the code reflects what's written here in the future).

PS: originally posted here.

💖 💪 🙅 🚩
joaofbantunes
João Antunes

Posted on July 31, 2018

Join Our Newsletter. No Spam, Only the good stuff.

Sign up to receive the latest update from our blog.

Related