This implies that you can not call an action from an action (a.k.a. killing action-** c**omposition), right?
Unfortunately this makes the github actions DSL an extremely limited programming language.
Until now I discovered: minimal if , no loops and no composition.
This basically boils down to a single level sequential list of commands, only suitable for simple use-cases. Duplication is my only resort, which is not particularly state-of-the-art :-(.
- What was the reasoning behind this design?
- Are there any plans to introduce composition and/or loop-constructs at a later stage?
P.S. Don’t get me wrong here, I like github actions a lot for what it can do and how easy it is to use and extend. I love it so much that I would like to use it for more and more… but that requires a more and more mature DSL…
So, keep up the good work and keep evolving!