The fact that we call a parameter to execute "initial-something", feels a bit odd to me.
It looks like it would be better to distinguish the "static" part of the context and the "runtime" part of the context, so that we have better distinction between the two computations.
So I would like to propose two changes:
The config will accept the :initial-ctx key, signalling to the framework that we want the context to be initialized with a value xyz
The parameter passed to execute becomes the "runtime" context, which will be merged with the :initial-ctx and threaded through the steps.
I think this would improve naming and semantics will be more clear.
The fact that we call a parameter to
execute"initial-something", feels a bit odd to me.It looks like it would be better to distinguish the "static" part of the context and the "runtime" part of the context, so that we have better distinction between the two computations.
So I would like to propose two changes:
configwill accept the:initial-ctxkey, signalling to the framework that we want the context to be initialized with a value xyzexecutebecomes the "runtime" context, which will be merged with the:initial-ctxand threaded through the steps.I think this would improve naming and semantics will be more clear.