Lifetime of an actors note when using mocha specs

Hi @jan-molak,

I did some experiments with the new notes feature using Serenity/JS 3.0.0-rc18 and the mocha template, to keep it simple.

Given an Actor

• using a notepad
• that is not shared
• that is either typed or untyped

then the note is just available in the mocha it block where it was set.

Though I’m in the same spec and in the same describe block, the note is not available when using an untyped note or set back to the initial value when using the typed notepad.

Is this intended? This is basically the scenario which led to my question about the lifecylce of an actor.

As the the Actor gets configured here should the note not be available for an actor called by the same name?

To have the notes available I would have to use a shared notepad, but this would apply to all actors, wouldn’t it?

[viper3400]

Thanks, Jan! This example with

• using different notepads
• for different actors
• speaking different languages

finally helped me to understand the whole (technical) concept behind the actors (and their lifetime) a bit more.

[jan-molak]

initiate the actor in a before hook

means more or less to let an actorCalled(‘somehow’) perform his first Task.

Exactly. Well, technically speaking, even a call to actorCalled('somehow') without specificying the task would instantiate an actor. It’s a bit like initialising a constant before you do something with it.

Would it be possible to build a dependency? My attempt to engage both does not really work.

Right, that’s because the second call to engage resets the Cast set by the first call.

Remember, that you can always do something a bit simpler, which is to have one implementation of the Cast, and differentiate between the actors based on their names.

For example:

class Actors implements Cast {
  prepare(actor: Actor): Actor {
    switch(actor.name) {
      case 'Jean':
        return actor.whoCan(TakeNotes.using(Notepad.with(this.translationsIn('fr-FR'))))
      case 'John'
        return actor.whoCan(TakeNotes.using(Notepad.with(this.translationsIn('en-GB'))))       
      default:
        return actor.whoCan(TakeNotes.using(Notepad.with(this.translationsIn('de-DE'))))
    }
  }
  
  private translationsIn(locale: string) {
      return ...
  }
}

(Maybe, I should also take into account, that one it block should not depend on another one, anyway …)

Yeah, I’d suggest avoiding test scenarios depending on the order of execution, since that prevents people from executing scenarios individually.