Read the room version from database when fetching events (#6874)

This is a precursor to giving EventBase objects the knowledge of which room version they belong to.

3 files changed, 79 insertions, 16 deletions

diff --git a/changelog.d/6874.misc b/changelog.d/6874.misc
new file mode 100644
index 0000000000..08aa80bcd9
--- /dev/null
+++ b/changelog.d/6874.misc
@@ -0,0 +1 @@
+Refactoring work in preparation for changing the event redaction algorithm.
diff --git a/synapse/storage/data_stores/main/events_worker.py b/synapse/storage/data_stores/main/events_worker.py
index 47a3a26072..ca237c6f12 100644
--- a/synapse/storage/data_stores/main/events_worker.py
+++ b/synapse/storage/data_stores/main/events_worker.py
@@ -28,9 +28,12 @@ from twisted.internet import defer
 
 from synapse.api.constants import EventTypes
 from synapse.api.errors import NotFoundError
-from synapse.api.room_versions import EventFormatVersions
-from synapse.events import FrozenEvent, event_type_from_format_version  # noqa: F401
-from synapse.events.snapshot import EventContext  # noqa: F401
+from synapse.api.room_versions import (
+    KNOWN_ROOM_VERSIONS,
+    EventFormatVersions,
+    RoomVersions,
+)
+from synapse.events import make_event_from_dict
 from synapse.events.utils import prune_event
 from synapse.logging.context import LoggingContext, PreserveLoggingContext
 from synapse.metrics.background_process_metrics import run_as_background_process
@@ -580,8 +583,49 @@ class EventsWorkerStore(SQLBaseStore):
                 # of a event format version, so it must be a V1 event.
                 format_version = EventFormatVersions.V1
 
-            original_ev = event_type_from_format_version(format_version)(
+            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
+                    )
+                    continue
+
+                # take a wild stab at the room version based on the event format
+                if format_version == EventFormatVersions.V1:
+                    room_version = RoomVersions.V1
+                elif format_version == EventFormatVersions.V2:
+                    room_version = RoomVersions.V3
+                else:
+                    room_version = RoomVersions.V5
+            else:
+                room_version = KNOWN_ROOM_VERSIONS.get(room_version_id)
+                if not room_version:
+                    logger.error(
+                        "Event %s in room %s has unknown room version %s",
+                        event_id,
+                        d["room_id"],
+                        room_version_id,
+                    )
+                    continue
+
+                if room_version.event_format != format_version:
+                    logger.error(
+                        "Event %s in room %s with version %s has wrong format: "
+                        "expected %s, was %s",
+                        event_id,
+                        d["room_id"],
+                        room_version_id,
+                        room_version.event_format,
+                        format_version,
+                    )
+                    continue
+
+            original_ev = make_event_from_dict(
                 event_dict=d,
+                room_version=room_version,
                 internal_metadata_dict=internal_metadata,
                 rejected_reason=rejected_reason,
             )
@@ -661,6 +705,12 @@ class EventsWorkerStore(SQLBaseStore):
            of EventFormatVersions. 'None' means the event predates
            EventFormatVersions (so the event is format V1).
 
+         * room_version_id (str|None): The version of the room which contains the event.
+           Hopefully one of RoomVersions.
+
+           Due to historical reasons, there may be a few events in the database which
+           do not have an associated room; in this case None will be returned here.
+
          * rejected_reason (str|None): if the event was rejected, the reason
            why.
 
@@ -676,17 +726,18 @@ class EventsWorkerStore(SQLBaseStore):
         """
         event_dict = {}
         for evs in batch_iter(event_ids, 200):
-            sql = (
-                "SELECT "
-                " e.event_id, "
-                " e.internal_metadata,"
-                " e.json,"
-                " e.format_version, "
-                " rej.reason "
-                " FROM event_json as e"
-                " LEFT JOIN rejections as rej USING (event_id)"
-                " WHERE "
-            )
+            sql = """\
+                SELECT
+                  e.event_id,
+                  e.internal_metadata,
+                  e.json,
+                  e.format_version,
+                  r.room_version,
+                  rej.reason
+                FROM event_json as e
+                  LEFT JOIN rooms r USING (room_id)
+                  LEFT JOIN rejections as rej USING (event_id)
+                WHERE """
 
             clause, args = make_in_list_sql_clause(
                 txn.database_engine, "e.event_id", evs
@@ -701,7 +752,8 @@ class EventsWorkerStore(SQLBaseStore):
                     "internal_metadata": row[1],
                     "json": row[2],
                     "format_version": row[3],
-                    "rejected_reason": row[4],
+                    "room_version_id": row[4],
+                    "rejected_reason": row[5],
                     "redactions": [],
                 }
 
diff --git a/tests/replication/slave/storage/test_events.py b/tests/replication/slave/storage/test_events.py
index d31210fbe4..f0561b30e3 100644
--- a/tests/replication/slave/storage/test_events.py
+++ b/tests/replication/slave/storage/test_events.py
@@ -15,6 +15,7 @@ import logging
 
 from canonicaljson import encode_canonical_json
 
+from synapse.api.room_versions import RoomVersions
 from synapse.events import FrozenEvent, _EventInternalMetadata, make_event_from_dict
 from synapse.events.snapshot import EventContext
 from synapse.handlers.room import RoomEventSource
@@ -58,6 +59,15 @@ class SlavedEventStoreTestCase(BaseSlavedStoreTestCase):
         self.unpatches = [patch__eq__(_EventInternalMetadata), patch__eq__(FrozenEvent)]
         return super(SlavedEventStoreTestCase, self).setUp()
 
+    def prepare(self, *args, **kwargs):
+        super().prepare(*args, **kwargs)
+
+        self.get_success(
+            self.master_store.store_room(
+                ROOM_ID, USER_ID, is_public=False, room_version=RoomVersions.V1,
+            )
+        )
+
     def tearDown(self):
         [unpatch() for unpatch in self.unpatches]