Store room_versions in EventBase objects (#6875)

This is a bit fiddly because it all has to be done on one fell swoop:

* Wherever we create a new event, pass in the room version (and check it matches the format version)
* When we prune an event, use the room version of the unpruned event to create the pruned version.
* When we pass an event over the replication protocol, pass the room version over alongside it, and use it when deserialising the event again.

5 files changed, 63 insertions, 32 deletions

diff --git a/changelog.d/6875.misc b/changelog.d/6875.misc
new file mode 100644
index 0000000000..08aa80bcd9
--- /dev/null
+++ b/changelog.d/6875.misc
@@ -0,0 +1 @@
+Refactoring work in preparation for changing the event redaction algorithm.
diff --git a/synapse/events/__init__.py b/synapse/events/__init__.py
index 7307116556..533ba327f5 100644
--- a/synapse/events/__init__.py
+++ b/synapse/events/__init__.py
@@ -15,9 +15,10 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+import abc
 import os
 from distutils.util import strtobool
-from typing import Optional, Type
+from typing import Dict, Optional, Type
 
 import six
 
@@ -199,15 +200,25 @@ class _EventInternalMetadata(object):
         return self._dict.get("redacted", False)
 
 
-class EventBase(object):
+class EventBase(metaclass=abc.ABCMeta):
+    @property
+    @abc.abstractmethod
+    def format_version(self) -> int:
+        """The EventFormatVersion implemented by this event"""
+        ...
+
     def __init__(
         self,
-        event_dict,
-        signatures={},
-        unsigned={},
-        internal_metadata_dict={},
-        rejected_reason=None,
+        event_dict: JsonDict,
+        room_version: RoomVersion,
+        signatures: Dict[str, Dict[str, str]],
+        unsigned: JsonDict,
+        internal_metadata_dict: JsonDict,
+        rejected_reason: Optional[str],
     ):
+        assert room_version.event_format == self.format_version
+
+        self.room_version = room_version
         self.signatures = signatures
         self.unsigned = unsigned
         self.rejected_reason = rejected_reason
@@ -303,7 +314,13 @@ class EventBase(object):
 class FrozenEvent(EventBase):
     format_version = EventFormatVersions.V1  # All events of this type are V1
 
-    def __init__(self, event_dict, internal_metadata_dict={}, rejected_reason=None):
+    def __init__(
+        self,
+        event_dict: JsonDict,
+        room_version: RoomVersion,
+        internal_metadata_dict: JsonDict = {},
+        rejected_reason: Optional[str] = None,
+    ):
         event_dict = dict(event_dict)
 
         # Signatures is a dict of dicts, and this is faster than doing a
@@ -326,8 +343,9 @@ class FrozenEvent(EventBase):
 
         self._event_id = event_dict["event_id"]
 
-        super(FrozenEvent, self).__init__(
+        super().__init__(
             frozen_dict,
+            room_version=room_version,
             signatures=signatures,
             unsigned=unsigned,
             internal_metadata_dict=internal_metadata_dict,
@@ -352,7 +370,13 @@ class FrozenEvent(EventBase):
 class FrozenEventV2(EventBase):
     format_version = EventFormatVersions.V2  # All events of this type are V2
 
-    def __init__(self, event_dict, internal_metadata_dict={}, rejected_reason=None):
+    def __init__(
+        self,
+        event_dict: JsonDict,
+        room_version: RoomVersion,
+        internal_metadata_dict: JsonDict = {},
+        rejected_reason: Optional[str] = None,
+    ):
         event_dict = dict(event_dict)
 
         # Signatures is a dict of dicts, and this is faster than doing a
@@ -377,8 +401,9 @@ class FrozenEventV2(EventBase):
 
         self._event_id = None
 
-        super(FrozenEventV2, self).__init__(
+        super().__init__(
             frozen_dict,
+            room_version=room_version,
             signatures=signatures,
             unsigned=unsigned,
             internal_metadata_dict=internal_metadata_dict,
@@ -445,7 +470,7 @@ class FrozenEventV3(FrozenEventV2):
         return self._event_id
 
 
-def event_type_from_format_version(format_version: int) -> Type[EventBase]:
+def _event_type_from_format_version(format_version: int) -> Type[EventBase]:
     """Returns the python type to use to construct an Event object for the
     given event format version.
 
@@ -474,5 +499,5 @@ def make_event_from_dict(
     rejected_reason: Optional[str] = None,
 ) -> EventBase:
     """Construct an EventBase from the given event dict"""
-    event_type = event_type_from_format_version(room_version.event_format)
-    return event_type(event_dict, internal_metadata_dict, rejected_reason)
+    event_type = _event_type_from_format_version(room_version.event_format)
+    return event_type(event_dict, room_version, internal_metadata_dict, rejected_reason)
diff --git a/synapse/events/utils.py b/synapse/events/utils.py
index f70f5032fb..bc6f98ae3b 100644
--- a/synapse/events/utils.py
+++ b/synapse/events/utils.py
@@ -35,26 +35,20 @@ from . import EventBase
 SPLIT_FIELD_REGEX = re.compile(r"(?<!\\)\.")
 
 
-def prune_event(event):
+def prune_event(event: EventBase) -> EventBase:
     """ Returns a pruned version of the given event, which removes all keys we
     don't know about or think could potentially be dodgy.
 
     This is used when we "redact" an event. We want to remove all fields that
     the user has specified, but we do want to keep necessary information like
     type, state_key etc.
-
-    Args:
-        event (FrozenEvent)
-
-    Returns:
-        FrozenEvent
     """
     pruned_event_dict = prune_event_dict(event.get_dict())
 
-    from . import event_type_from_format_version
+    from . import make_event_from_dict
 
-    pruned_event = event_type_from_format_version(event.format_version)(
-        pruned_event_dict, event.internal_metadata.get_dict()
+    pruned_event = make_event_from_dict(
+        pruned_event_dict, event.room_version, event.internal_metadata.get_dict()
     )
 
     # Mark the event as redacted
diff --git a/synapse/replication/http/federation.py b/synapse/replication/http/federation.py
index 8794720101..7e23b565b9 100644
--- a/synapse/replication/http/federation.py
+++ b/synapse/replication/http/federation.py
@@ -18,7 +18,7 @@ import logging
 from twisted.internet import defer
 
 from synapse.api.room_versions import KNOWN_ROOM_VERSIONS
-from synapse.events import event_type_from_format_version
+from synapse.events import make_event_from_dict
 from synapse.events.snapshot import EventContext
 from synapse.http.servlet import parse_json_object_from_request
 from synapse.replication.http._base import ReplicationEndpoint
@@ -38,6 +38,9 @@ class ReplicationFederationSendEventsRestServlet(ReplicationEndpoint):
         {
             "events": [{
                 "event": { .. serialized event .. },
+                "room_version": .., // "1", "2", "3", etc: the version of the room
+                                    // containing the event
+                "event_format_version": .., // 1,2,3 etc: the event format version
                 "internal_metadata": { .. serialized internal_metadata .. },
                 "rejected_reason": ..,   // The event.rejected_reason field
                 "context": { .. serialized event context .. },
@@ -73,6 +76,7 @@ class ReplicationFederationSendEventsRestServlet(ReplicationEndpoint):
             event_payloads.append(
                 {
                     "event": event.get_pdu_json(),
+                    "room_version": event.room_version.identifier,
                     "event_format_version": event.format_version,
                     "internal_metadata": event.internal_metadata.get_dict(),
                     "rejected_reason": event.rejected_reason,
@@ -95,12 +99,13 @@ class ReplicationFederationSendEventsRestServlet(ReplicationEndpoint):
             event_and_contexts = []
             for event_payload in event_payloads:
                 event_dict = event_payload["event"]
-                format_ver = event_payload["event_format_version"]
+                room_ver = KNOWN_ROOM_VERSIONS[event_payload["room_version"]]
                 internal_metadata = event_payload["internal_metadata"]
                 rejected_reason = event_payload["rejected_reason"]
 
-                EventType = event_type_from_format_version(format_ver)
-                event = EventType(event_dict, internal_metadata, rejected_reason)
+                event = make_event_from_dict(
+                    event_dict, room_ver, internal_metadata, rejected_reason
+                )
 
                 context = EventContext.deserialize(
                     self.storage, event_payload["context"]
diff --git a/synapse/replication/http/send_event.py b/synapse/replication/http/send_event.py
index 84b92f16ad..b74b088ff4 100644
--- a/synapse/replication/http/send_event.py
+++ b/synapse/replication/http/send_event.py
@@ -17,7 +17,8 @@ import logging
 
 from twisted.internet import defer
 
-from synapse.events import event_type_from_format_version
+from synapse.api.room_versions import KNOWN_ROOM_VERSIONS
+from synapse.events import make_event_from_dict
 from synapse.events.snapshot import EventContext
 from synapse.http.servlet import parse_json_object_from_request
 from synapse.replication.http._base import ReplicationEndpoint
@@ -37,6 +38,9 @@ class ReplicationSendEventRestServlet(ReplicationEndpoint):
 
         {
             "event": { .. serialized event .. },
+            "room_version": .., // "1", "2", "3", etc: the version of the room
+                                // containing the event
+            "event_format_version": .., // 1,2,3 etc: the event format version
             "internal_metadata": { .. serialized internal_metadata .. },
             "rejected_reason": ..,   // The event.rejected_reason field
             "context": { .. serialized event context .. },
@@ -77,6 +81,7 @@ class ReplicationSendEventRestServlet(ReplicationEndpoint):
 
         payload = {
             "event": event.get_pdu_json(),
+            "room_version": event.room_version.identifier,
             "event_format_version": event.format_version,
             "internal_metadata": event.internal_metadata.get_dict(),
             "rejected_reason": event.rejected_reason,
@@ -93,12 +98,13 @@ class ReplicationSendEventRestServlet(ReplicationEndpoint):
             content = parse_json_object_from_request(request)
 
             event_dict = content["event"]
-            format_ver = content["event_format_version"]
+            room_ver = KNOWN_ROOM_VERSIONS[content["room_version"]]
             internal_metadata = content["internal_metadata"]
             rejected_reason = content["rejected_reason"]
 
-            EventType = event_type_from_format_version(format_ver)
-            event = EventType(event_dict, internal_metadata, rejected_reason)
+            event = make_event_from_dict(
+                event_dict, room_ver, internal_metadata, rejected_reason
+            )
 
             requester = Requester.deserialize(self.store, content["requester"])
             context = EventContext.deserialize(self.storage, content["context"])