Add MSC3030 experimental client and federation API endpoints to get the closest event to a given timestamp (#9445)

MSC3030: https://github.com/matrix-org/matrix-doc/pull/3030

Client API endpoint. This will also go and fetch from the federation API endpoint if unable to find an event locally or we found an extremity with possibly a closer event we don't know about.
```
GET /_matrix/client/unstable/org.matrix.msc3030/rooms/<roomID>/timestamp_to_event?ts=<timestamp>&dir=<direction>
{
    "event_id": ...
    "origin_server_ts": ...
}
```

Federation API endpoint:
```
GET /_matrix/federation/unstable/org.matrix.msc3030/timestamp_to_event/<roomID>?ts=<timestamp>&dir=<direction>
{
    "event_id": ...
    "origin_server_ts": ...
}
```

Co-authored-by: Erik Johnston <erik@matrix.org>

5 files changed, 208 insertions, 1 deletions

diff --git a/synapse/federation/federation_client.py b/synapse/federation/federation_client.py
index bc3f96c1fc..be1423da24 100644
--- a/synapse/federation/federation_client.py
+++ b/synapse/federation/federation_client.py
@@ -1517,6 +1517,83 @@ class FederationClient(FederationBase):
         self._get_room_hierarchy_cache[(room_id, suggested_only)] = result
         return result
 
+    async def timestamp_to_event(
+        self, destination: str, room_id: str, timestamp: int, direction: str
+    ) -> "TimestampToEventResponse":
+        """
+        Calls a remote federating server at `destination` asking for their
+        closest event to the given timestamp in the given direction. Also
+        validates the response to always return the expected keys or raises an
+        error.
+
+        Args:
+            destination: Domain name of the remote homeserver
+            room_id: Room to fetch the event from
+            timestamp: The point in time (inclusive) we should navigate from in
+                the given direction to find the closest event.
+            direction: ["f"|"b"] to indicate whether we should navigate forward
+                or backward from the given timestamp to find the closest event.
+
+        Returns:
+            A parsed TimestampToEventResponse including the closest event_id
+            and origin_server_ts
+
+        Raises:
+            Various exceptions when the request fails
+            InvalidResponseError when the response does not have the correct
+            keys or wrong types
+        """
+        remote_response = await self.transport_layer.timestamp_to_event(
+            destination, room_id, timestamp, direction
+        )
+
+        if not isinstance(remote_response, dict):
+            raise InvalidResponseError(
+                "Response must be a JSON dictionary but received %r" % remote_response
+            )
+
+        try:
+            return TimestampToEventResponse.from_json_dict(remote_response)
+        except ValueError as e:
+            raise InvalidResponseError(str(e))
+
+
+@attr.s(frozen=True, slots=True, auto_attribs=True)
+class TimestampToEventResponse:
+    """Typed response dictionary for the federation /timestamp_to_event endpoint"""
+
+    event_id: str
+    origin_server_ts: int
+
+    # the raw data, including the above keys
+    data: JsonDict
+
+    @classmethod
+    def from_json_dict(cls, d: JsonDict) -> "TimestampToEventResponse":
+        """Parsed response from the federation /timestamp_to_event endpoint
+
+        Args:
+            d: JSON object response to be parsed
+
+        Raises:
+            ValueError if d does not the correct keys or they are the wrong types
+        """
+
+        event_id = d.get("event_id")
+        if not isinstance(event_id, str):
+            raise ValueError(
+                "Invalid response: 'event_id' must be a str but received %r" % event_id
+            )
+
+        origin_server_ts = d.get("origin_server_ts")
+        if not isinstance(origin_server_ts, int):
+            raise ValueError(
+                "Invalid response: 'origin_server_ts' must be a int but received %r"
+                % origin_server_ts
+            )
+
+        return cls(event_id, origin_server_ts, d)
+
 
 @attr.s(frozen=True, slots=True, auto_attribs=True)
 class FederationSpaceSummaryEventResult:
diff --git a/synapse/federation/federation_server.py b/synapse/federation/federation_server.py
index 8fbc75aa65..cce85526e7 100644
--- a/synapse/federation/federation_server.py
+++ b/synapse/federation/federation_server.py
@@ -110,6 +110,7 @@ class FederationServer(FederationBase):
         super().__init__(hs)
 
         self.handler = hs.get_federation_handler()
+        self.storage = hs.get_storage()
         self._federation_event_handler = hs.get_federation_event_handler()
         self.state = hs.get_state_handler()
         self._event_auth_handler = hs.get_event_auth_handler()
@@ -200,6 +201,48 @@ class FederationServer(FederationBase):
 
         return 200, res
 
+    async def on_timestamp_to_event_request(
+        self, origin: str, room_id: str, timestamp: int, direction: str
+    ) -> Tuple[int, Dict[str, Any]]:
+        """When we receive a federated `/timestamp_to_event` request,
+        handle all of the logic for validating and fetching the event.
+
+        Args:
+            origin: The server we received the event from
+            room_id: Room to fetch the event from
+            timestamp: The point in time (inclusive) we should navigate from in
+                the given direction to find the closest event.
+            direction: ["f"|"b"] to indicate whether we should navigate forward
+                or backward from the given timestamp to find the closest event.
+
+        Returns:
+            Tuple indicating the response status code and dictionary response
+            body including `event_id`.
+        """
+        with (await self._server_linearizer.queue((origin, room_id))):
+            origin_host, _ = parse_server_name(origin)
+            await self.check_server_matches_acl(origin_host, room_id)
+
+            # We only try to fetch data from the local database
+            event_id = await self.store.get_event_id_for_timestamp(
+                room_id, timestamp, direction
+            )
+            if event_id:
+                event = await self.store.get_event(
+                    event_id, allow_none=False, allow_rejected=False
+                )
+
+                return 200, {
+                    "event_id": event_id,
+                    "origin_server_ts": event.origin_server_ts,
+                }
+
+        raise SynapseError(
+            404,
+            "Unable to find event from %s in direction %s" % (timestamp, direction),
+            errcode=Codes.NOT_FOUND,
+        )
+
     async def on_incoming_transaction(
         self,
         origin: str,
diff --git a/synapse/federation/transport/client.py b/synapse/federation/transport/client.py
index fe29bcfd4b..d1f4be641d 100644
--- a/synapse/federation/transport/client.py
+++ b/synapse/federation/transport/client.py
@@ -149,6 +149,42 @@ class TransportLayerClient:
         )
 
     @log_function
+    async def timestamp_to_event(
+        self, destination: str, room_id: str, timestamp: int, direction: str
+    ) -> Union[JsonDict, List]:
+        """
+        Calls a remote federating server at `destination` asking for their
+        closest event to the given timestamp in the given direction.
+
+        Args:
+            destination: Domain name of the remote homeserver
+            room_id: Room to fetch the event from
+            timestamp: The point in time (inclusive) we should navigate from in
+                the given direction to find the closest event.
+            direction: ["f"|"b"] to indicate whether we should navigate forward
+                or backward from the given timestamp to find the closest event.
+
+        Returns:
+            Response dict received from the remote homeserver.
+
+        Raises:
+            Various exceptions when the request fails
+        """
+        path = _create_path(
+            FEDERATION_UNSTABLE_PREFIX,
+            "/org.matrix.msc3030/timestamp_to_event/%s",
+            room_id,
+        )
+
+        args = {"ts": [str(timestamp)], "dir": [direction]}
+
+        remote_response = await self.client.get_json(
+            destination, path=path, args=args, try_trailing_slash_on_400=True
+        )
+
+        return remote_response
+
+    @log_function
     async def send_transaction(
         self,
         transaction: Transaction,
diff --git a/synapse/federation/transport/server/__init__.py b/synapse/federation/transport/server/__init__.py
index c32539bf5a..abcb8728f5 100644
--- a/synapse/federation/transport/server/__init__.py
+++ b/synapse/federation/transport/server/__init__.py
@@ -22,7 +22,10 @@ from synapse.federation.transport.server._base import (
     Authenticator,
     BaseFederationServlet,
 )
-from synapse.federation.transport.server.federation import FEDERATION_SERVLET_CLASSES
+from synapse.federation.transport.server.federation import (
+    FEDERATION_SERVLET_CLASSES,
+    FederationTimestampLookupServlet,
+)
 from synapse.federation.transport.server.groups_local import GROUP_LOCAL_SERVLET_CLASSES
 from synapse.federation.transport.server.groups_server import (
     GROUP_SERVER_SERVLET_CLASSES,
@@ -324,6 +327,13 @@ def register_servlets(
             )
 
         for servletclass in DEFAULT_SERVLET_GROUPS[servlet_group]:
+            # Only allow the `/timestamp_to_event` servlet if msc3030 is enabled
+            if (
+                servletclass == FederationTimestampLookupServlet
+                and not hs.config.experimental.msc3030_enabled
+            ):
+                continue
+
             servletclass(
                 hs=hs,
                 authenticator=authenticator,
diff --git a/synapse/federation/transport/server/federation.py b/synapse/federation/transport/server/federation.py
index 66e915228c..77bfd88ad0 100644
--- a/synapse/federation/transport/server/federation.py
+++ b/synapse/federation/transport/server/federation.py
@@ -174,6 +174,46 @@ class FederationBackfillServlet(BaseFederationServerServlet):
         return await self.handler.on_backfill_request(origin, room_id, versions, limit)
 
 
+class FederationTimestampLookupServlet(BaseFederationServerServlet):
+    """
+    API endpoint to fetch the `event_id` of the closest event to the given
+    timestamp (`ts` query parameter) in the given direction (`dir` query
+    parameter).
+
+    Useful for other homeservers when they're unable to find an event locally.
+
+    `ts` is a timestamp in milliseconds where we will find the closest event in
+    the given direction.
+
+    `dir` can be `f` or `b` to indicate forwards and backwards in time from the
+    given timestamp.
+
+    GET /_matrix/federation/unstable/org.matrix.msc3030/timestamp_to_event/<roomID>?ts=<timestamp>&dir=<direction>
+    {
+        "event_id": ...
+    }
+    """
+
+    PATH = "/timestamp_to_event/(?P<room_id>[^/]*)/?"
+    PREFIX = FEDERATION_UNSTABLE_PREFIX + "/org.matrix.msc3030"
+
+    async def on_GET(
+        self,
+        origin: str,
+        content: Literal[None],
+        query: Dict[bytes, List[bytes]],
+        room_id: str,
+    ) -> Tuple[int, JsonDict]:
+        timestamp = parse_integer_from_args(query, "ts", required=True)
+        direction = parse_string_from_args(
+            query, "dir", default="f", allowed_values=["f", "b"], required=True
+        )
+
+        return await self.handler.on_timestamp_to_event_request(
+            origin, room_id, timestamp, direction
+        )
+
+
 class FederationQueryServlet(BaseFederationServerServlet):
     PATH = "/query/(?P<query_type>[^/]*)"
 
@@ -683,6 +723,7 @@ FEDERATION_SERVLET_CLASSES: Tuple[Type[BaseFederationServlet], ...] = (
     FederationStateV1Servlet,
     FederationStateIdsServlet,
     FederationBackfillServlet,
+    FederationTimestampLookupServlet,
     FederationQueryServlet,
     FederationMakeJoinServlet,
     FederationMakeLeaveServlet,