You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Sep 14, 2023. It is now read-only.
#226 states the intention to invert the flow of types. We initially planned to encapsulate all types (RPC and FRAME types) in a Config, which would flow through the DX. Now, we're going in the other direction (initial work in #229): type-safe bindings wrap the dynamic usage (potentially injecting pre-computed codecs, keys, misc.). This brings up some important questions: should we even bother encoding static type info in a Config?
Of course, this would result in type errors in the current RPC client. Perhaps this is a good thing though... it does feel like too much typing at too low a level.
#226 states the intention to invert the flow of types. We initially planned to encapsulate all types (RPC and FRAME types) in a
Config
, which would flow through the DX. Now, we're going in the other direction (initial work in #229): type-safe bindings wrap the dynamic usage (potentially injecting pre-computed codecs, keys, misc.). This brings up some important questions: should we even bother encoding static type info in aConfig
?Let's take a look at the current
Config
:What would happen if we got rid of the RPC phantoms?
Of course, this would result in type errors in the current RPC client. Perhaps this is a good thing though... it does feel like too much typing at too low a level.
We can address this after #251.
The text was updated successfully, but these errors were encountered: