Check appservice user interest against the local users instead of all users (`get_users_in_room` mis-use) (#13958)

3 files changed, 32 insertions, 4 deletions

diff --git a/synapse/appservice/__init__.py b/synapse/appservice/__init__.py

index 0dfa00df44..500bdde3a9 100644
--- a/synapse/appservice/__init__.py
+++ b/synapse/appservice/__init__.py
@@ -172,12 +172,24 @@ class ApplicationService:
         Returns:
             True if this service would like to know about this room.
         """
-        member_list = await store.get_users_in_room(
+        # We can use `get_local_users_in_room(...)` here because an application service
+        # can only be interested in local users of the server it's on (ignore any remote
+        # users that might match the user namespace regex).
+        #
+        # In the future, we can consider re-using
+        # `store.get_app_service_users_in_room` which is very similar to this
+        # function but has a slightly worse performance than this because we
+        # have an early escape-hatch if we find a single user that the
+        # appservice is interested in. The juice would be worth the squeeze if
+        # `store.get_app_service_users_in_room` was used in more places besides
+        # an experimental MSC. But for now we can avoid doing more work and
+        # barely using it later.
+        local_user_ids = await store.get_local_users_in_room(
             room_id, on_invalidate=cache_context.invalidate
         )
 
         # check joined member events
-        for user_id in member_list:
+        for user_id in local_user_ids:
             if self.is_interested_in_user(user_id):
                 return True
         return False
diff --git a/synapse/storage/databases/main/appservice.py b/synapse/storage/databases/main/appservice.py

index 64b70a7b28..63046c0527 100644
--- a/synapse/storage/databases/main/appservice.py
+++ b/synapse/storage/databases/main/appservice.py
@@ -157,10 +157,23 @@ class ApplicationServiceWorkerStore(RoomMemberWorkerStore):
         app_service: "ApplicationService",
         cache_context: _CacheContext,
     ) -> List[str]:
-        users_in_room = await self.get_users_in_room(
+        """
+        Get all users in a room that the appservice controls.
+
+        Args:
+            room_id: The room to check in.
+            app_service: The application service to check interest/control against
+
+        Returns:
+            List of user IDs that the appservice controls.
+        """
+        # We can use `get_local_users_in_room(...)` here because an application service
+        # can only be interested in local users of the server it's on (ignore any remote
+        # users that might match the user namespace regex).
+        local_users_in_room = await self.get_local_users_in_room(
             room_id, on_invalidate=cache_context.invalidate
         )
-        return list(filter(app_service.is_interested_in_user, users_in_room))
+        return list(filter(app_service.is_interested_in_user, local_users_in_room))
 
 
 class ApplicationServiceStore(ApplicationServiceWorkerStore):
diff --git a/synapse/storage/databases/main/roommember.py b/synapse/storage/databases/main/roommember.py

index ab708b0ba5..e56a13f21e 100644
--- a/synapse/storage/databases/main/roommember.py
+++ b/synapse/storage/databases/main/roommember.py
@@ -152,6 +152,9 @@ class RoomMemberWorkerStore(EventsWorkerStore):
         the forward extremities of those rooms will exclude most members. We may also
         calculate room state incorrectly for such rooms and believe that a member is or
         is not in the room when the opposite is true.
+
+        Note: If you only care about users in the room local to the homeserver, use
+        `get_local_users_in_room(...)` instead which will be more performant.
         """
         return await self.db_pool.simple_select_onecol(
             table="current_state_events",