Daml Script allocating parties in Distributed Topologies, and subsequent commands

Assuming that there are a couple of Canton Participant nodes and I want to allocate parties on these participants using a single Daml Script.

The participants.json contains the config for all participants, and the Script is using the AllocatePartyOn command.

What I am interested in is the party_participants config. Given that the parties are only allocated during the runtime of the script, I of course can not statically set the partyId in the static config.

Does that mean that
a) The script runner is smart enough to remember that when you allocate a party on a node, it will send all subsequent commands from that party to the same node.
b) The script runner tries to fall back to match the display name of the party with the party_participants entry.
c) Something completely different?

It’s (a): The script runner automatically adds any newly allocated parties to the party_participants mapping and subsequent commands are sent to the corresponding participant.