huddles: Find huddle user ids more efficiently.

We restrict the columns, avoid quadratic looping,
and don't bother with order_by.

We also return the user ids (per recipient) as
sets, since that's how the only caller uses the
info (albeit implicitly via set.union accepting
a list).

zerver/lib/display_recipient.py

@@ -147,7 +147,7 @@ def bulk_fetch_display_recipients(
 
         # Find all user ids whose UserProfiles we will need to fetch:
         user_ids_to_fetch: Set[int] = set()
-        huddle_user_ids: Dict[int, List[int]] = {}
+        huddle_user_ids: Dict[int, Set[int]] = {}
         huddle_user_ids = bulk_get_huddle_user_ids(
             [recipient for recipient in recipients if recipient.type == Recipient.HUDDLE]
         )

zerver/models.py

@@ -3,6 +3,7 @@ import datetime
 import hashlib
 import secrets
 import time
+from collections import defaultdict
 from contextlib import suppress
 from datetime import timedelta
 from email.headerregistry import Address
@@ -2898,7 +2899,7 @@ def get_huddle_user_ids(recipient: Recipient) -> ValuesQuerySet["Subscription",
     )
 
 
-def bulk_get_huddle_user_ids(recipients: List[Recipient]) -> Dict[int, List[int]]:
+def bulk_get_huddle_user_ids(recipients: List[Recipient]) -> Dict[int, Set[int]]:
     """
     Takes a list of huddle-type recipients, returns a dict
     mapping recipient id to list of user ids in the huddle.
@@ -2909,15 +2910,11 @@ def bulk_get_huddle_user_ids(recipients: List[Recipient]) -> Dict[int, List[int]
 
     subscriptions = Subscription.objects.filter(
         recipient__in=recipients,
-    ).order_by("user_profile_id")
+    ).only("user_profile_id", "recipient_id")
 
-    result_dict: Dict[int, List[int]] = {}
-    for recipient in recipients:
-        result_dict[recipient.id] = [
-            subscription.user_profile_id
-            for subscription in subscriptions
-            if subscription.recipient_id == recipient.id
-        ]
+    result_dict: Dict[int, Set[int]] = defaultdict(set)
+    for subscription in subscriptions:
+        result_dict[subscription.recipient_id].add(subscription.user_profile_id)
 
     return result_dict
 

zerver/tests/test_messages.py

@@ -90,9 +90,9 @@ class TestBulkGetHuddleUserIds(ZulipTestCase):
 
         messages = Message.objects.filter(id__in=message_ids).order_by("id")
         first_huddle_recipient = messages[0].recipient
-        first_huddle_user_ids = list(get_huddle_user_ids(first_huddle_recipient))
+        first_huddle_user_ids = set(get_huddle_user_ids(first_huddle_recipient))
         second_huddle_recipient = messages[1].recipient
-        second_huddle_user_ids = list(get_huddle_user_ids(second_huddle_recipient))
+        second_huddle_user_ids = set(get_huddle_user_ids(second_huddle_recipient))
 
         huddle_user_ids = bulk_get_huddle_user_ids(
             [first_huddle_recipient, second_huddle_recipient]