Replies: 2 comments 1 reply
-
Not sure if that's possible right now. I wouldn't be opposed to accepting a refactor as long as it still keeps things simple. Maybe we can decouple things, so you can use the 'low-level' stuff directly without relying on the high-level flake-parts layer. Alternatively, maybe there is way for you to use flake-parts just for a subset of your flake (ie., you can use its outputs but without letting it control the flake outputs). cc @roberth |
Beta Was this translation helpful? Give feedback.
-
Also interested as I currently hacked around it in similar way since I’m not even using flakes at work yet (waiting for lazy trees) |
Beta Was this translation helpful? Give feedback.
-
Hi & thank you very much for this awesome stuff!
I'm currently wondering: Is it possible to use process-compose-flake as a stand-alone library, without flake-parts?
I imagine something like this:
And then being able to
my-processes up
ormy-processes down
I know you're probably using the module system provided by flake-parts, and decoupling it would require shipping your own module system. But it would also mean, process-compose-flake could be used within flakes not using flake-parts (flake-utils for example).
Beta Was this translation helpful? Give feedback.
All reactions