peter/siftlode
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

fix(quota): don't let the demo account persist deep_requested (S1, finding 2)

Browse Source 
The demo account is shared, so writing deep_requested onto its subscription polluted
shared state (harmless today — the scheduler's deep paths are human-only — but a
future scheduler path trusting the flag would leak quota). Guard the flag WRITE at the
source (not user.is_demo), which also makes deep_turned_on always False for demo, so
the redundant is_demo check on the immediate backfill is removed.
This commit is contained in:
peter
2026-07-22 00:00:44 +02:00
parent cb4d06bb3d
commit f76cd6ef36
1 changed files with 6 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files
backend/app/routes/channels.py
+6 -9
View File
@@ -391,7 +391,10 @@ def update_channel(
sub.priority = int(payload["priority"]) sub.priority = int(payload["priority"])
if "hidden" in payload: if "hidden" in payload:
sub.hidden = bool(payload["hidden"]) sub.hidden = bool(payload["hidden"])
if "deep_requested" in payload: # The demo account is SHARED, so it must not persist deep_requested (a quota-relevant flag) onto
# the shared subscription — skip the write entirely for demo, which also keeps deep_turned_on
# False so the immediate backfill below can never fire for it.
if "deep_requested" in payload and not user.is_demo:
# Opt this channel into (or out of) full-history backfill. The deep scheduler # Opt this channel into (or out of) full-history backfill. The deep scheduler
# picks the flag up on its next run; turning it off won't undo already-fetched # picks the flag up on its next run; turning it off won't undo already-fetched
# videos, it just stops further deep paging if nobody else still wants it. # videos, it just stops further deep paging if nobody else still wants it.
@@ -403,14 +406,8 @@ def update_channel(
# Opting a channel into full history should give an immediate feed: if we haven't even # Opting a channel into full history should give an immediate feed: if we haven't even
# fetched its recent uploads yet, do that one channel now (cheap) instead of waiting for # fetched its recent uploads yet, do that one channel now (cheap) instead of waiting for
# the scheduler. Deep paging still follows on the scheduler's next run (recent-then-deep). # the scheduler. Deep paging still follows on the scheduler's next run (recent-then-deep).
# Not for the demo account: it must not be able to trigger an immediate quota-burning backfill # (deep_turned_on is only ever True for a non-demo user — see the guard above.)
# (the scheduler's deep paths are already human-only). Same guard as channel_detail's enrichment. if deep_turned_on and channel is not None and channel.recent_synced_at is None:
if (
deep_turned_on
and channel is not None
and channel.recent_synced_at is None
and not user.is_demo
):
with quota.attribute(user.id, quota.QuotaAction.VIDEOS_BACKFILL_RECENT): with quota.attribute(user.id, quota.QuotaAction.VIDEOS_BACKFILL_RECENT):
run_recent_backfill(db, [channel], max_channels=1) run_recent_backfill(db, [channel], max_channels=1)