Add `getCached` function to `Cache` module about io HOT 7 CLOSED

Closing as completed by #445

from io.

@IMax153 Why does the signature for getOption include Error? Since it will never hit the resolver function shouldn't it be Effect<never, never, Option.Option<Value>> instead?

 /**
     * Retrieves the value associated with the specified key if it exists.
     * Otherwise returns `Option.none`.
     */
    getOption(key: Key): Effect.Effect<never, Error, Option.Option<Value>>;

from io.

cc @tim-smart

from io.

The error is there because a cache entry can still be in a pending state, in the case where the lookup effect is async.

If the cache entry is in a pending state, then it will await the underlying Deferred and pass any errors back to the caller.

from io.

@tim-smart OK makes sense. Perhaps there's a need for another function then. A sort of getImmediateOption that will run sync without any Error only returning the entry if it's actually already cached, ignoring any pending.

I think that may be the more useful behavior. The current getOption is basically just a shorthand for the following.

pipe(
   someCache.contains(key),
   T.ifEffect(
     pipe(cache.get(key), T.map(O.some)), 
     T.succeedNone()
   )

AFAIK there's no way to do a getImmediateOption with the current Cache api. I think it would be useful for situations where you want to take different actions based on whether a value is immediately available because a pending cache request could take an arbitrary amount of time to complete.

from io.

Yes I don't see an issue with another API for getting a completed cache value.

readonly getOptionComplete: (key: Key) => Effect<never, never, Option<Value>>

from io.

Closing as completed by #448

from io.