What is the role and purpose of the replayCache parameter in the SharedFlow implementation?

The replayCache parameter in the SharedFlow implementation determines the number of the most recent emitted values that will be stored and made available to new subscribers upon their connection. This allows new subscribers to immediately receive the latest events, even if they occurred before their subscription. For example, if replayCache is set to 1, a new subscriber will receive the last value emitted in the SharedFlow.

Example usage:

val sharedFlow = MutableSharedFlow<Int>(replay = 1)

sharedFlow.emit(1) // the value will be stored in replayCache

sharedFlow.collect { value ->
    println(value) // the new subscriber will immediately receive 1
}

Thus, replayCache enhances reactivity and data consistency for new subscribers.