Great tool! I've recently started experimenting with PostgreSQL logical replication as part of a project where we’re pruning data from our main database and arc

Great tool! I've recently started experimenting with PostgreSQL logical replication as part of a project where we’re pruning data from our main database and archiving it to S3 for future use. However, some tables need to be available in a secondary Postgres instance specifically for read operations, which is where replication comes in. I’m encountering a challenge due to two types of delete operations in the main database. The first delete operation is for data pruning, and I don’t want these deletions to be reflected in the replicated database. The second delete operation is for rollback purposes, which alters the data state, and these deletions should indeed be replicated to maintain consistency. Is there a way to distinguish between these two delete operations so that only the rollback deletions are replicated? And would your tool support this kind of use case?