Add Admin API to Fetch Messages Within a Particular Window (#13672)

This adds two new admin APIs that allow us to fetch messages from a room within a particular time.

3 files changed, 132 insertions, 13 deletions

diff --git a/synapse/handlers/pagination.py b/synapse/handlers/pagination.py
index a0c39778ab..1f83bab836 100644
--- a/synapse/handlers/pagination.py
+++ b/synapse/handlers/pagination.py
@@ -26,6 +26,7 @@ from synapse.events.utils import SerializeEventConfig
 from synapse.handlers.room import ShutdownRoomResponse
 from synapse.logging.opentracing import trace
 from synapse.metrics.background_process_metrics import run_as_background_process
+from synapse.rest.admin._base import assert_user_is_admin
 from synapse.storage.state import StateFilter
 from synapse.streams.config import PaginationConfig
 from synapse.types import JsonDict, Requester, StreamKeyType
@@ -423,6 +424,7 @@ class PaginationHandler:
         pagin_config: PaginationConfig,
         as_client_event: bool = True,
         event_filter: Optional[Filter] = None,
+        use_admin_priviledge: bool = False,
     ) -> JsonDict:
         """Get messages in a room.
 
@@ -432,10 +434,16 @@ class PaginationHandler:
             pagin_config: The pagination config rules to apply, if any.
             as_client_event: True to get events in client-server format.
             event_filter: Filter to apply to results or None
+            use_admin_priviledge: if `True`, return all events, regardless
+                of whether `user` has access to them. To be used **ONLY**
+                from the admin API.
 
         Returns:
             Pagination API results
         """
+        if use_admin_priviledge:
+            await assert_user_is_admin(self.auth, requester)
+
         user_id = requester.user.to_string()
 
         if pagin_config.from_token:
@@ -458,12 +466,14 @@ class PaginationHandler:
         room_token = from_token.room_key
 
         async with self.pagination_lock.read(room_id):
-            (
-                membership,
-                member_event_id,
-            ) = await self.auth.check_user_in_room_or_world_readable(
-                room_id, requester, allow_departed_users=True
-            )
+            (membership, member_event_id) = (None, None)
+            if not use_admin_priviledge:
+                (
+                    membership,
+                    member_event_id,
+                ) = await self.auth.check_user_in_room_or_world_readable(
+                    room_id, requester, allow_departed_users=True
+                )
 
             if pagin_config.direction == "b":
                 # if we're going backwards, we might need to backfill. This
@@ -475,7 +485,7 @@ class PaginationHandler:
                         room_id, room_token.stream
                     )
 
-                if membership == Membership.LEAVE:
+                if not use_admin_priviledge and membership == Membership.LEAVE:
                     # If they have left the room then clamp the token to be before
                     # they left the room, to save the effort of loading from the
                     # database.
@@ -528,12 +538,13 @@ class PaginationHandler:
         if event_filter:
             events = await event_filter.filter(events)
 
-        events = await filter_events_for_client(
-            self._storage_controllers,
-            user_id,
-            events,
-            is_peeking=(member_event_id is None),
-        )
+        if not use_admin_priviledge:
+            events = await filter_events_for_client(
+                self._storage_controllers,
+                user_id,
+                events,
+                is_peeking=(member_event_id is None),
+            )
 
         # if after the filter applied there are no more events
         # return immediately - but there might be more in next_token batch
diff --git a/synapse/rest/admin/__init__.py b/synapse/rest/admin/__init__.py
index fa3266720b..bac754e1b1 100644
--- a/synapse/rest/admin/__init__.py
+++ b/synapse/rest/admin/__init__.py
@@ -61,9 +61,11 @@ from synapse.rest.admin.rooms import (
     MakeRoomAdminRestServlet,
     RoomEventContextServlet,
     RoomMembersRestServlet,
+    RoomMessagesRestServlet,
     RoomRestServlet,
     RoomRestV2Servlet,
     RoomStateRestServlet,
+    RoomTimestampToEventRestServlet,
 )
 from synapse.rest.admin.server_notice_servlet import SendServerNoticeServlet
 from synapse.rest.admin.statistics import UserMediaStatisticsRestServlet
@@ -271,6 +273,8 @@ def register_servlets(hs: "HomeServer", http_server: HttpServer) -> None:
     DestinationResetConnectionRestServlet(hs).register(http_server)
     DestinationRestServlet(hs).register(http_server)
     ListDestinationsRestServlet(hs).register(http_server)
+    RoomMessagesRestServlet(hs).register(http_server)
+    RoomTimestampToEventRestServlet(hs).register(http_server)
 
     # Some servlets only get registered for the main process.
     if hs.config.worker.worker_app is None:
diff --git a/synapse/rest/admin/rooms.py b/synapse/rest/admin/rooms.py
index 3d870629c4..747e6fda83 100644
--- a/synapse/rest/admin/rooms.py
+++ b/synapse/rest/admin/rooms.py
@@ -35,6 +35,7 @@ from synapse.rest.admin._base import (
 )
 from synapse.storage.databases.main.room import RoomSortOrder
 from synapse.storage.state import StateFilter
+from synapse.streams.config import PaginationConfig
 from synapse.types import JsonDict, RoomID, UserID, create_requester
 from synapse.util import json_decoder
 
@@ -858,3 +859,106 @@ class BlockRoomRestServlet(RestServlet):
             await self._store.unblock_room(room_id)
 
         return HTTPStatus.OK, {"block": block}
+
+
+class RoomMessagesRestServlet(RestServlet):
+    """
+    Get messages list of a room.
+    """
+
+    PATTERNS = admin_patterns("/rooms/(?P<room_id>[^/]*)/messages$")
+
+    def __init__(self, hs: "HomeServer"):
+        self._hs = hs
+        self._clock = hs.get_clock()
+        self._pagination_handler = hs.get_pagination_handler()
+        self._auth = hs.get_auth()
+        self._store = hs.get_datastores().main
+
+    async def on_GET(
+        self, request: SynapseRequest, room_id: str
+    ) -> Tuple[int, JsonDict]:
+        requester = await self._auth.get_user_by_req(request)
+        await assert_user_is_admin(self._auth, requester)
+
+        pagination_config = await PaginationConfig.from_request(
+            self._store, request, default_limit=10
+        )
+        # Twisted will have processed the args by now.
+        assert request.args is not None
+        as_client_event = b"raw" not in request.args
+        filter_str = parse_string(request, "filter", encoding="utf-8")
+        if filter_str:
+            filter_json = urlparse.unquote(filter_str)
+            event_filter: Optional[Filter] = Filter(
+                self._hs, json_decoder.decode(filter_json)
+            )
+            if (
+                event_filter
+                and event_filter.filter_json.get("event_format", "client")
+                == "federation"
+            ):
+                as_client_event = False
+        else:
+            event_filter = None
+
+        msgs = await self._pagination_handler.get_messages(
+            room_id=room_id,
+            requester=requester,
+            pagin_config=pagination_config,
+            as_client_event=as_client_event,
+            event_filter=event_filter,
+            use_admin_priviledge=True,
+        )
+
+        return HTTPStatus.OK, msgs
+
+
+class RoomTimestampToEventRestServlet(RestServlet):
+    """
+    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 cases like jump to date so you can start paginating messages from
+    a given date in the archive.
+
+    `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 /_synapse/admin/v1/rooms/<roomID>/timestamp_to_event?ts=<timestamp>&dir=<direction>
+    {
+        "event_id": ...
+    }
+    """
+
+    PATTERNS = admin_patterns("/rooms/(?P<room_id>[^/]*)/timestamp_to_event$")
+
+    def __init__(self, hs: "HomeServer"):
+        self._auth = hs.get_auth()
+        self._store = hs.get_datastores().main
+        self._timestamp_lookup_handler = hs.get_timestamp_lookup_handler()
+
+    async def on_GET(
+        self, request: SynapseRequest, room_id: str
+    ) -> Tuple[int, JsonDict]:
+        requester = await self._auth.get_user_by_req(request)
+        await assert_user_is_admin(self._auth, requester)
+
+        timestamp = parse_integer(request, "ts", required=True)
+        direction = parse_string(request, "dir", default="f", allowed_values=["f", "b"])
+
+        (
+            event_id,
+            origin_server_ts,
+        ) = await self._timestamp_lookup_handler.get_event_for_timestamp(
+            requester, room_id, timestamp, direction
+        )
+
+        return HTTPStatus.OK, {
+            "event_id": event_id,
+            "origin_server_ts": origin_server_ts,
+        }