paulmataruso/zulip
1
0
Fork 0
mirror of https://github.com/zulip/zulip.git synced 2025-11-04 14:03:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

realm emoji: Use a single cache for all lookups.

Browse Source 
The active realm emoji are just a subset of all your
realm emoji, so just use a single cache entry per
realm.

Cache misses should be very infrequent per realm.

If a realm has lots of deactivated realm emoji, then
there's a minor expense to deserialize them, but that
is gonna be dwarfed by all the other more expensive
operations in message-send.

I also renamed the two related functions.  I erred on
the side of using somewhat verbose names, as we don't
want folks to confuse the two use cases. Fortunately
there are somewhat natural affordances to use one or
the other, and mypy helps too.

Finally, I use realm_id instead of realm in places
where we don't need the full Realm object.
This commit is contained in:
Steve Howell
2023-07-14 10:37:29 +00:00
committed by Tim Abbott
parent e988cf9b0a
commit b742f1241f
8 changed files with 58 additions and 48 deletions
Download Patch File Download Diff File Expand all files Collapse all files

44
zerver/models.py
View File

@@ -239,14 +239,10 @@ def get_recipient_ids(
return to, recipient_type_str
def get_realm_emoji_cache_key(realm_id: int) -> str:
def get_all_custom_emoji_for_realm_cache_key(realm_id: int) -> str:
return f"realm_emoji:{realm_id}"
def get_active_realm_emoji_cache_key(realm_id: int) -> str:
return f"active_realm_emoji:{realm_id}"
# This simple call-once caching saves ~500us in auth_enabled_helper,
# which is a significant optimization for common_context. Note that
# these values cannot change in a running production system, but do
@@ -837,16 +833,6 @@ class Realm(models.Model): # type: ignore[django-manager-missing] # django-stub
ret[realm_authentication_method.name] = True
return ret
# `realm` instead of `self` here to make sure the parameters of the cache key
# function matches the original method.
@cache_with_key(lambda realm: get_realm_emoji_cache_key(realm.id), timeout=3600 * 24 * 7)
def get_emoji(realm) -> Dict[str, EmojiInfo]: # noqa: N805
return get_realm_emoji_uncached(realm.id)
@cache_with_key(lambda realm: get_active_realm_emoji_cache_key(realm.id), timeout=3600 * 24 * 7)
def get_active_emoji(realm) -> Dict[str, EmojiInfo]: # noqa: N805
return get_active_realm_emoji_uncached(realm.id)
def get_admin_users_and_bots(
self, include_realm_owners: bool = True
) -> QuerySet["UserProfile"]:
@@ -1161,7 +1147,7 @@ class RealmEmoji(models.Model):
return f"{self.realm.string_id}: {self.id} {self.name} {self.deactivated} {self.file_name}"
def get_realm_emoji_dicts(realm_id: int, only_active_emojis: bool = False) -> Dict[str, EmojiInfo]:
def get_all_custom_emoji_for_realm_uncached(realm_id: int) -> Dict[str, EmojiInfo]:
# RealmEmoji objects with file_name=None are still in the process
# of being uploaded, and we expect to be cleaned up by a
# try/finally block if the upload fails, so it's correct to
@@ -1169,8 +1155,6 @@ def get_realm_emoji_dicts(realm_id: int, only_active_emojis: bool = False) -> Di
query = RealmEmoji.objects.filter(realm_id=realm_id).exclude(
file_name=None,
)
if only_active_emojis:
query = query.filter(deactivated=False)
d = {}
from zerver.lib.emoji import get_emoji_url
@@ -1202,16 +1186,15 @@ def get_realm_emoji_dicts(realm_id: int, only_active_emojis: bool = False) -> Di
return d
def get_realm_emoji_uncached(realm_id: int) -> Dict[str, EmojiInfo]:
return get_realm_emoji_dicts(realm_id)
@cache_with_key(get_all_custom_emoji_for_realm_cache_key, timeout=3600 * 24 * 7)
def get_all_custom_emoji_for_realm(realm_id: int) -> Dict[str, EmojiInfo]:
return get_all_custom_emoji_for_realm_uncached(realm_id)
def get_active_realm_emoji_uncached(realm_id: int) -> Dict[str, EmojiInfo]:
realm_emojis = get_realm_emoji_dicts(realm_id, only_active_emojis=True)
d = {}
for emoji_id, emoji_dict in realm_emojis.items():
d[emoji_dict["name"]] = emoji_dict
return d
def get_name_keyed_dict_for_active_realm_emoji(realm_id: int) -> Dict[str, EmojiInfo]:
# It's important to use the cached version here.
realm_emojis = get_all_custom_emoji_for_realm(realm_id)
return {row["name"]: row for row in realm_emojis.values() if not row["deactivated"]}
def flush_realm_emoji(*, instance: RealmEmoji, **kwargs: object) -> None:
@@ -1227,13 +1210,8 @@ def flush_realm_emoji(*, instance: RealmEmoji, **kwargs: object) -> None:
return
realm_id = instance.realm_id
cache_set(
get_realm_emoji_cache_key(realm_id),
get_realm_emoji_uncached(realm_id),
timeout=3600 * 24 * 7,
)
cache_set(
get_active_realm_emoji_cache_key(realm_id),
get_active_realm_emoji_uncached(realm_id),
get_all_custom_emoji_for_realm_cache_key(realm_id),
get_all_custom_emoji_for_realm_uncached(realm_id),
timeout=3600 * 24 * 7,
)