From a481fb9f98ad10e5e129bdc7664c59498a7332f6 Mon Sep 17 00:00:00 2001
From: Patrick Cloke <clokep@users.noreply.github.com>
Date: Fri, 10 Feb 2023 08:09:47 -0500
Subject: Refactor get_user_devices_from_cache to avoid mutating cached values.
 (#15040)

The previous version of the code could mutate a cached value,
but only if the input requested all devices of a user *and* a specific
device.

To avoid this nonsensical situation we no longer fetch a specific
device ID if all of a user's devices are returned.
---
 synapse/storage/databases/main/devices.py | 31 +++++++++++++++++--------------
 1 file changed, 17 insertions(+), 14 deletions(-)

(limited to 'synapse/storage/databases')

diff --git a/synapse/storage/databases/main/devices.py b/synapse/storage/databases/main/devices.py
index 766c2052fb..85c1778a81 100644
--- a/synapse/storage/databases/main/devices.py
+++ b/synapse/storage/databases/main/devices.py
@@ -746,42 +746,45 @@ class DeviceWorkerStore(RoomMemberWorkerStore, EndToEndKeyWorkerStore):
     @trace
     @cancellable
     async def get_user_devices_from_cache(
-        self, query_list: List[Tuple[str, Optional[str]]]
+        self, user_ids: Set[str], user_and_device_ids: List[Tuple[str, str]]
     ) -> Tuple[Set[str], Dict[str, Dict[str, JsonDict]]]:
         """Get the devices (and keys if any) for remote users from the cache.
 
         Args:
-            query_list: List of (user_id, device_ids), if device_ids is
-                falsey then return all device ids for that user.
+            user_ids: users which should have all device IDs returned
+            user_and_device_ids: List of (user_id, device_ids)
 
         Returns:
             A tuple of (user_ids_not_in_cache, results_map), where
             user_ids_not_in_cache is a set of user_ids and results_map is a
             mapping of user_id -> device_id -> device_info.
         """
-        user_ids = {user_id for user_id, _ in query_list}
-        user_map = await self.get_device_list_last_stream_id_for_remotes(list(user_ids))
+        unique_user_ids = user_ids | {user_id for user_id, _ in user_and_device_ids}
+        user_map = await self.get_device_list_last_stream_id_for_remotes(
+            list(unique_user_ids)
+        )
 
         # We go and check if any of the users need to have their device lists
         # resynced. If they do then we remove them from the cached list.
         users_needing_resync = await self.get_user_ids_requiring_device_list_resync(
-            user_ids
+            unique_user_ids
         )
         user_ids_in_cache = {
             user_id for user_id, stream_id in user_map.items() if stream_id
         } - users_needing_resync
-        user_ids_not_in_cache = user_ids - user_ids_in_cache
+        user_ids_not_in_cache = unique_user_ids - user_ids_in_cache
 
+        # First fetch all the users which all devices are to be returned.
         results: Dict[str, Dict[str, JsonDict]] = {}
-        for user_id, device_id in query_list:
-            if user_id not in user_ids_in_cache:
-                continue
-
-            if device_id:
+        for user_id in user_ids:
+            if user_id in user_ids_in_cache:
+                results[user_id] = await self.get_cached_devices_for_user(user_id)
+        # Then fetch all device-specific requests, but skip users we've already
+        # fetched all devices for.
+        for user_id, device_id in user_and_device_ids:
+            if user_id in user_ids_in_cache and user_id not in user_ids:
                 device = await self._get_cached_user_device(user_id, device_id)
                 results.setdefault(user_id, {})[device_id] = device
-            else:
-                results[user_id] = await self.get_cached_devices_for_user(user_id)
 
         set_tag("in_cache", str(results))
         set_tag("not_in_cache", str(user_ids_not_in_cache))
-- 
cgit 1.4.1