Be more tolerant of membership events in unknown rooms (#8110)

It turns out that not all out-of-band membership events are labelled as such,
so we need to be more accepting here.

2 files changed, 27 insertions, 6 deletions

diff --git a/synapse/events/__init__.py b/synapse/events/__init__.py
index cc5deca75b..67db763dbf 100644
--- a/synapse/events/__init__.py
+++ b/synapse/events/__init__.py
@@ -133,6 +133,8 @@ class _EventInternalMetadata(object):
         rejection. This is needed as those events are marked as outliers, but
         they still need to be processed as if they're new events (e.g. updating
         invite state in the database, relaying to clients, etc).
+
+        (Added in synapse 0.99.0, so may be unreliable for events received before that)
         """
         return self._dict.get("out_of_band_membership", False)
 
diff --git a/synapse/storage/databases/main/events_worker.py b/synapse/storage/databases/main/events_worker.py
index 4a3333c0db..e1241a724b 100644
--- a/synapse/storage/databases/main/events_worker.py
+++ b/synapse/storage/databases/main/events_worker.py
@@ -620,19 +620,38 @@ class EventsWorkerStore(SQLBaseStore):
             room_version_id = row["room_version_id"]
 
             if not room_version_id:
-                # this should only happen for out-of-band membership events
-                if not internal_metadata.get("out_of_band_membership"):
-                    logger.warning(
-                        "Room %s for event %s is unknown", d["room_id"], event_id
+                # this should only happen for out-of-band membership events which
+                # arrived before #6983 landed. For all other events, we should have
+                # an entry in the 'rooms' table.
+                #
+                # However, the 'out_of_band_membership' flag is unreliable for older
+                # invites, so just accept it for all membership events.
+                #
+                if d["type"] != EventTypes.Member:
+                    raise Exception(
+                        "Room %s for event %s is unknown" % (d["room_id"], event_id)
                     )
-                    continue
 
-                # take a wild stab at the room version based on the event format
+                # so, assuming this is an out-of-band-invite that arrived before #6983
+                # landed, we know that the room version must be v5 or earlier (because
+                # v6 hadn't been invented at that point, so invites from such rooms
+                # would have been rejected.)
+                #
+                # The main reason we need to know the room version here (other than
+                # choosing the right python Event class) is in case the event later has
+                # to be redacted - and all the room versions up to v5 used the same
+                # redaction algorithm.
+                #
+                # So, the following approximations should be adequate.
+
                 if format_version == EventFormatVersions.V1:
+                    # if it's event format v1 then it must be room v1 or v2
                     room_version = RoomVersions.V1
                 elif format_version == EventFormatVersions.V2:
+                    # if it's event format v2 then it must be room v3
                     room_version = RoomVersions.V3
                 else:
+                    # if it's event format v3 then it must be room v4 or v5
                     room_version = RoomVersions.V5
             else:
                 room_version = KNOWN_ROOM_VERSIONS.get(room_version_id)