summary refs log tree commit diff
path: root/synapse/federation/transport/client.py
diff options
context:
space:
mode:
authorErik Johnston <erik@matrix.org>2021-05-20 16:11:48 +0100
committerGitHub <noreply@github.com>2021-05-20 16:11:48 +0100
commit64887f06fcac63e069364d625d984b4951bf1ffc (patch)
treeeabd30c1ad56d057667271573d3bfc96276d11bd /synapse/federation/transport/client.py
parentAllow a user who could join a restricted room to see it in spaces summary. (#... (diff)
downloadsynapse-64887f06fcac63e069364d625d984b4951bf1ffc.tar.xz
Use ijson to parse the response to `/send_join`, reducing memory usage. (#9958)
Instead of parsing the full response to `/send_join` into Python objects (which can be huge for large rooms) and *then* parsing that into events, we instead use ijson to stream parse the response directly into `EventBase` objects.
Diffstat (limited to 'synapse/federation/transport/client.py')
-rw-r--r--synapse/federation/transport/client.py85
1 files changed, 81 insertions, 4 deletions
diff --git a/synapse/federation/transport/client.py b/synapse/federation/transport/client.py
index 497848a2b7..e93ab83f7f 100644
--- a/synapse/federation/transport/client.py
+++ b/synapse/federation/transport/client.py
@@ -17,13 +17,19 @@ import logging
 import urllib
 from typing import Any, Dict, List, Optional
 
+import attr
+import ijson
+
 from synapse.api.constants import Membership
 from synapse.api.errors import Codes, HttpResponseException, SynapseError
+from synapse.api.room_versions import RoomVersion
 from synapse.api.urls import (
     FEDERATION_UNSTABLE_PREFIX,
     FEDERATION_V1_PREFIX,
     FEDERATION_V2_PREFIX,
 )
+from synapse.events import EventBase, make_event_from_dict
+from synapse.http.matrixfederationclient import ByteParser
 from synapse.logging.utils import log_function
 from synapse.types import JsonDict
 
@@ -240,21 +246,36 @@ class TransportLayerClient:
         return content
 
     @log_function
-    async def send_join_v1(self, destination, room_id, event_id, content):
+    async def send_join_v1(
+        self,
+        room_version,
+        destination,
+        room_id,
+        event_id,
+        content,
+    ) -> "SendJoinResponse":
         path = _create_v1_path("/send_join/%s/%s", room_id, event_id)
 
         response = await self.client.put_json(
-            destination=destination, path=path, data=content
+            destination=destination,
+            path=path,
+            data=content,
+            parser=SendJoinParser(room_version, v1_api=True),
         )
 
         return response
 
     @log_function
-    async def send_join_v2(self, destination, room_id, event_id, content):
+    async def send_join_v2(
+        self, room_version, destination, room_id, event_id, content
+    ) -> "SendJoinResponse":
         path = _create_v2_path("/send_join/%s/%s", room_id, event_id)
 
         response = await self.client.put_json(
-            destination=destination, path=path, data=content
+            destination=destination,
+            path=path,
+            data=content,
+            parser=SendJoinParser(room_version, v1_api=False),
         )
 
         return response
@@ -1053,3 +1074,59 @@ def _create_v2_path(path, *args):
         str
     """
     return _create_path(FEDERATION_V2_PREFIX, path, *args)
+
+
+@attr.s(slots=True, auto_attribs=True)
+class SendJoinResponse:
+    """The parsed response of a `/send_join` request."""
+
+    auth_events: List[EventBase]
+    state: List[EventBase]
+
+
+@ijson.coroutine
+def _event_list_parser(room_version: RoomVersion, events: List[EventBase]):
+    """Helper function for use with `ijson.items_coro` to parse an array of
+    events and add them to the given list.
+    """
+
+    while True:
+        obj = yield
+        event = make_event_from_dict(obj, room_version)
+        events.append(event)
+
+
+class SendJoinParser(ByteParser[SendJoinResponse]):
+    """A parser for the response to `/send_join` requests.
+
+    Args:
+        room_version: The version of the room.
+        v1_api: Whether the response is in the v1 format.
+    """
+
+    CONTENT_TYPE = "application/json"
+
+    def __init__(self, room_version: RoomVersion, v1_api: bool):
+        self._response = SendJoinResponse([], [])
+
+        # The V1 API has the shape of `[200, {...}]`, which we handle by
+        # prefixing with `item.*`.
+        prefix = "item." if v1_api else ""
+
+        self._coro_state = ijson.items_coro(
+            _event_list_parser(room_version, self._response.state),
+            prefix + "state.item",
+        )
+        self._coro_auth = ijson.items_coro(
+            _event_list_parser(room_version, self._response.auth_events),
+            prefix + "auth_chain.item",
+        )
+
+    def write(self, data: bytes) -> int:
+        self._coro_state.send(data)
+        self._coro_auth.send(data)
+
+        return len(data)
+
+    def finish(self) -> SendJoinResponse:
+        return self._response