What is the difference between writing exists=owner outside the scope of an owner and inside the scope?

[sommio]

What is the difference between

potential = {
    uses_district_set = city_world
    exists = owner
    owner = { has_valid_civic = civic_machine_servitor }
}

and

potential = {
    uses_district_set = city_world
    owner = { exists = owner has_valid_civic = civic_machine_servitor }
}

for example?

 

[Mr. Morague]

With the second one, you're scoping to the owner immediately. If there isn't an owner, you'd be scoping to something that doesn't exist and get a nice little error.

 

[Ryika]

Potential for an error message aside, if you swap into the owner scope and do the exists = owner check there, then you're doing that check on the owner, not the planet. So you're essentially looking for the planet owner's owner, which I think will always return false, since empires do not have owners.

 

[Firegolem]

Potential for an error message aside, if you swap into the owner scope and do the exists = owner check there, then you're doing that check on the owner, not the planet. So you're essentially looking for the planet owner's owner, which I think will always return false, since empires do not have owners.

The game does not return false, because the game is very indulgent here (the same with species = { exists = species } ).

There also exists planet_owner (but I guess it is not really intended for planet scope ^^).

PS: The need of exists = owner is mostly a bug (as a PDX dev told, and at some point probably superfluous).