It is difficult to coordinate a change like that having only one-way communication).
I really think we need to fix that at some point, such that:
* Downstream connections can send CopyData messages *up* the COPY BOTH protocol, where they're passed to a hook on the output plugin; and
* Output plugins can hook the walsender's event loop (latch set, etc) and send their own messages without being driven by a logical decoding event .
I wanted to do that some time ago but ran into some issues and time constraints. Because of the need to support older versions I'm now committed to an approach using direct libpq connections and function calls instead, but it seems like a real shame to do that when the replication protocol connection is *right there*...
