Strange usage of ExecutorEffectsInterpreter in ExecutorEffectHandler

[adolgiy]

Current implementation in 'master' looks like:

typealias ExecutorEffectsInterpreter<Eff, Msg> = ExecutorService.(eff: Eff, listener: (Msg) -> Unit) -> Unit

class ExecutorEffectHandler<Msg : Any, Eff : Any>(
    // ...
) : EffectHandler<Eff, Msg> {

    override fun handleEffect(eff: Eff) {
        effectsExecutorService.run { effectsInterpreter(eff, listener ?: {}) } // !!! THIS
    }
}

Line with comment do nothing except proxying call to extension function via run {} call. Shouldn't it be fixed like this?

effectsExecutorService.execute { effectsInterpreter(eff, listener ?: {}) }

[Mishkun]

EffectsInterpreter is an extension function to ExecutorService, so yeah it is better to rewrite this as

effectsInterpreter(effectsExecutorService, eff, listener ?: {})

[adolgiy]

But in this way it will not be scheduled on given ExecutorService. Shouldn't ExecutorEffectHandler handle all effects on this executor? If should, why would you pass ExecutorService as param?