After finishing last project the time has come to prepare myself for next projects. And for that, I had to setup development tools and environments. Shouldn’t take too long right? Unless… There are more than two of them.
In this post there are mostly my reflections on working on .NET/JS project using Windows and Linux operating systems simultaneously and how to select good IDE and VCS for that kind of set-up.
The more the merier?
As for me, I am currently using four different development setups — two on notebooks (one with Win10/Fedora GNU/Linux combination, second with only Fedora) and two PCs (Win10/Fedora & only Win10 this time). As most of my projects are mostly .NET focused that would imply that on one notebook I cannot do any of my projects. Fortunately .NET Core happened.
.NET and Linux/Git cooperation
Now, as long as I use .NET Core as my base for C#-related programming everything should go as expected. Unless you take into consideration some issues with .NET Core support on Fedora, but based on my experience it can change from one dnf update to another. But then another issue emerges with Visual Studio Code (I still cannot bring myself to love that little monster). Probably it is finally time to use a JetBrains Rider 30 days trial for my own purposes and then consider buying the license for myself.
Oh, and another pain point is that Visual Studio doesn’t offer native support for normal Git, and only for GitHub. Of course there are some third party libraries, and for those the most promising one is GitLab Extension, but for that I need to once again make my GitLab alive — as currently is is down after some system update two years ago — I mostly did use it for my CLI/Python/TeX projects then, so I didn’t need integration with any IDE anyway so I didn’t care about it then.
Where should I start?
As during this weekend I was at my parents house where I have only Windows 10, I wanted to start preparing everything for next projects (new version of my website, and two other projects using Angular/Android/ASP.NET WebAPI2). But as those projects were supposed to be a little more private project and I didn’t want to buy GitHub Developer Plan I needed my own private repo once more. So I figured out that running GitLab again would be a good start for me (but it is still in TODO’s).
And then configuring all my environments to make sure I can just switch to another location/device and still code everything as planned. That should include installing NodeJS, Android SDK (perhaps Xamarin?) and .NET Core SDK everywhere, creating GitLab project and cloning repository on every machine. Doesn’t sound so bad, right? More work now can keep you have less work-load in a longer run after all.
Unless OpenSSH on Windows surprises you — but about that there will be another post in the upcoming days.
Originally published at www.blog.smyk.it on February 18, 2018.