From 097cd3ed726a49ef093ba4d90d9a4bcab6513f21 Mon Sep 17 00:00:00 2001
From: Wesley Aptekar-Cassels <me@wesleyac.com>
Date: Fri, 28 Apr 2023 12:51:44 -0400
Subject: [PATCH] Optimize get_audience by only fetching IDs

Looking at the tracing data from this function in prod, only ~500ms is
spent in the database. My best guess for the rest of the time is
transferring and creating the user objects, which we don't use, since we
simply need the ID.
---
 bookwyrm/activitystreams.py | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/bookwyrm/activitystreams.py b/bookwyrm/activitystreams.py
index 0dcb5262e..5d581d564 100644
--- a/bookwyrm/activitystreams.py
+++ b/bookwyrm/activitystreams.py
@@ -116,7 +116,7 @@ class ActivityStream(RedisStore):
         )
         # direct messages don't appear in feeds, direct comments/reviews/etc do
         if status.privacy == "direct" and status.status_type == "Note":
-            return []
+            return models.User.objects.none()
 
         # everybody who could plausibly see this status
         audience = models.User.objects.filter(
@@ -152,11 +152,11 @@ class ActivityStream(RedisStore):
     def get_audience(self, status):
         """given a status, what users should see it"""
         trace.get_current_span().set_attribute("stream_id", self.key)
-        audience = self._get_audience(status)
+        audience = self._get_audience(status).values_list("id", flat=True)
         status_author = models.User.objects.filter(
             is_active=True, local=True, id=status.user.id
-        )
-        return list({user.id for user in list(audience) + list(status_author)})
+        ).values_list("id", flat=True)
+        return list(set(list(audience) + list(status_author)))
 
     def get_stores_for_users(self, user_ids):
         """convert a list of user ids into redis store ids"""
@@ -186,12 +186,12 @@ class HomeStream(ActivityStream):
         if not audience:
             return []
         # if the user is following the author
-        audience = audience.filter(following=status.user)
+        audience = audience.filter(following=status.user).values_list("id", flat=True)
         # if the user is the post's author
         status_author = models.User.objects.filter(
             is_active=True, local=True, id=status.user.id
-        )
-        return list({user.id for user in list(audience) + list(status_author)})
+        ).values_list("id", flat=True)
+        return list(set(list(audience) + list(status_author)))
 
     def get_statuses_for_user(self, user):
         return models.Status.privacy_filter(
@@ -240,7 +240,7 @@ class BooksStream(ActivityStream):
 
         audience = super()._get_audience(status)
         if not audience:
-            return []
+            return models.User.objects.none()
         return audience.filter(shelfbook__book__parent_work=work).distinct()
 
     def get_audience(self, status):