When it comes to a host calling a closure, they get 3 parameters:
flags, closure data, and output.
Is my understanding accurate that:
flags is inputs,
closure data is captured variables,
and of course output is output.
the flags bit is arbitrary
in elm, your main function, in the most general case, takes some flags to get started
but if on your platform you want to have two initial arguments, you can (or should be able to, I guess this is untested)
So flags can be used as a input from the platform if wanted, but also can be used for other things?
I guess? the idea here is that the platform can inject some data
that data can come form the platform itself, or might be produced by other roc functions
ok. makes sense
Last updated: Jul 06 2025 at 12:14 UTC