Merge pull request #2174 from matrix-org/erikj/current_cache_hosts

Add cache for get_current_hosts_in_room

4 files changed, 54 insertions, 5 deletions

diff --git a/synapse/federation/transaction_queue.py b/synapse/federation/transaction_queue.py
index dee387eb7f..695f1a7375 100644
--- a/synapse/federation/transaction_queue.py
+++ b/synapse/federation/transaction_queue.py
@@ -24,7 +24,6 @@ from synapse.util.async import run_on_reactor
 from synapse.util.logcontext import preserve_context_over_fn, preserve_fn
 from synapse.util.retryutils import NotRetryingDestination, get_retry_limiter
 from synapse.util.metrics import measure_func
-from synapse.types import get_domain_from_id
 from synapse.handlers.presence import format_user_presence_state, get_interested_remotes
 import synapse.metrics
 
@@ -183,15 +182,12 @@ class TransactionQueue(object):
                     # Otherwise if the last member on a server in a room is
                     # banned then it won't receive the event because it won't
                     # be in the room after the ban.
-                    users_in_room = yield self.state.get_current_user_in_room(
+                    destinations = yield self.state.get_current_hosts_in_room(
                         event.room_id, latest_event_ids=[
                             prev_id for prev_id, _ in event.prev_events
                         ],
                     )
 
-                    destinations = set(
-                        get_domain_from_id(user_id) for user_id in users_in_room
-                    )
                     if send_on_behalf_of is not None:
                         # If we are sending the event on behalf of another server
                         # then it already has the event and there is no reason to
diff --git a/synapse/replication/slave/storage/events.py b/synapse/replication/slave/storage/events.py
index ab48ff925e..fcaf58b93b 100644
--- a/synapse/replication/slave/storage/events.py
+++ b/synapse/replication/slave/storage/events.py
@@ -144,6 +144,9 @@ class SlavedEventStore(BaseSlavedStore):
         RoomMemberStore.__dict__["_get_joined_users_from_context"]
     )
 
+    get_joined_hosts = DataStore.get_joined_hosts.__func__
+    _get_joined_hosts = RoomMemberStore.__dict__["_get_joined_hosts"]
+
     get_recent_events_for_room = DataStore.get_recent_events_for_room.__func__
     get_room_events_stream_for_rooms = (
         DataStore.get_room_events_stream_for_rooms.__func__
diff --git a/synapse/state.py b/synapse/state.py
index f6b83d888a..02fee47f39 100644
--- a/synapse/state.py
+++ b/synapse/state.py
@@ -176,6 +176,17 @@ class StateHandler(object):
         defer.returnValue(joined_users)
 
     @defer.inlineCallbacks
+    def get_current_hosts_in_room(self, room_id, latest_event_ids=None):
+        if not latest_event_ids:
+            latest_event_ids = yield self.store.get_latest_event_ids_in_room(room_id)
+        logger.debug("calling resolve_state_groups from get_current_hosts_in_room")
+        entry = yield self.resolve_state_groups(room_id, latest_event_ids)
+        joined_hosts = yield self.store.get_joined_hosts(
+            room_id, entry.state_id, entry.state
+        )
+        defer.returnValue(joined_hosts)
+
+    @defer.inlineCallbacks
     def compute_event_context(self, event, old_state=None):
         """Build an EventContext structure for the event.
 
diff --git a/synapse/storage/roommember.py b/synapse/storage/roommember.py
index 7ad2198d96..6ec8a6345d 100644
--- a/synapse/storage/roommember.py
+++ b/synapse/storage/roommember.py
@@ -18,6 +18,7 @@ from twisted.internet import defer
 from collections import namedtuple
 
 from ._base import SQLBaseStore
+from synapse.util.caches import intern_string
 from synapse.util.caches.descriptors import cached, cachedInlineCallbacks
 from synapse.util.stringutils import to_ascii
 
@@ -482,6 +483,44 @@ class RoomMemberStore(SQLBaseStore):
 
         defer.returnValue(False)
 
+    def get_joined_hosts(self, room_id, state_group, state_ids):
+        if not state_group:
+            # If state_group is None it means it has yet to be assigned a
+            # state group, i.e. we need to make sure that calls with a state_group
+            # of None don't hit previous cached calls with a None state_group.
+            # To do this we set the state_group to a new object as object() != object()
+            state_group = object()
+
+        return self._get_joined_hosts(
+            room_id, state_group, state_ids
+        )
+
+    @cachedInlineCallbacks(num_args=2, max_entries=10000, iterable=True)
+    def _get_joined_hosts(self, room_id, state_group, current_state_ids):
+        # We don't use `state_group`, its there so that we can cache based
+        # on it. However, its important that its never None, since two current_state's
+        # with a state_group of None are likely to be different.
+        # See bulk_get_push_rules_for_room for how we work around this.
+        assert state_group is not None
+
+        joined_hosts = set()
+        for (etype, state_key), event_id in current_state_ids.items():
+            if etype == EventTypes.Member:
+                try:
+                    host = get_domain_from_id(state_key)
+                except:
+                    logger.warn("state_key not user_id: %s", state_key)
+                    continue
+
+                if host in joined_hosts:
+                    continue
+
+                event = yield self.get_event(event_id, allow_none=True)
+                if event and event.content["membership"] == Membership.JOIN:
+                    joined_hosts.add(intern_string(host))
+
+        defer.returnValue(joined_hosts)
+
     @defer.inlineCallbacks
     def _background_add_membership_profile(self, progress, batch_size):
         target_min_stream_id = progress.get(