- pill: two host dots -> ONE consolidated state (down > slow > up) —
the at-a-glance question is 'can I watch?', which is one state;
per-host detail belongs in the tap panel
- panel: dashed slow-threshold line in the CDN sparkline (drawn at the
server's real threshold, not a client copy) + origin up/down step
sparkline so 'how long has the origin been down' is visible
- server: origin probe keeps a 20-sample hist like the CDN probe;
DEGRADED threshold is now ANIWEB_CDN_KBPS (default 40) and is
returned in /api/cdn so the UI draws the line at the live value
- tests: pill assertions switched to consolidated states; fixtures
carry origin hist + threshold; assert both sparklines + threshold line
- docs: README + SPEC synced; preview on :8322 restarted (pid 44382)
- pill label now shows KB/s when degraded ('cdn 23 KB/s'); tap opens a
panel (replaces the transient toast) with origin/CDN state, KB/s,
a sparkline of the ~1h /api/cdn hist, and last-updated age
- 60s poll while the page is visible (was: load/play/failure only)
- playback failure overlay appends CDN context when degraded/down
- buffer gauge + spinner untouched: client buffer state, not CDN state
- server unchanged: /api/cdn already returned everything
- tests: 3 hermetic browser tests (mocked /api/cdn via page.route —
the throwaway server's probe runs live upstream, so its data is
mocked out of the assertions); 37 passed
- docs: README + SPEC synced
Both docs still described the pre-provider-layer single-source app:
- README: anidb-only framing, missing stats page, 'upstream.py is the
ONLY Cloudflare-facing code', and a self-check command that does not
work (plain script execution breaks the providers' relative imports;
the asserts are anidb-specific under the hianime-first default) —
now documents ANI_SOURCE=anidb + the -m form. Dropped the stale
'episode titles are not available' note (Jikan titles shipped).
- SPEC: 'no database' -> SQLite, architecture -> 3 providers, full
endpoint list (intros/searches/events/stats/cdn, DELETE history,
?all=1, {sub}), auth hardening (per-IP budget, X-Aniweb on /login),
tests/run.py -> pytest + ANIWEB_LIVE_TESTS gating, out-of-scope list
(countdown/intro-skip/titles shipped), risks updated for the
fallback layer, milestone 8 pointer.
README: storage is SQLite (aniweb.db), not JSON files; CDN serves a
single quality (no ABR) — replaced the wrong 1080/720/360 claim;
document the 16-lane parallel fetch, intro-skip, episode names,
buffering UI, /api/cdn + background probe, the ?v= cache-busting rule,
and the client self-report diagnostics.
SPEC: same single-rendition correction (2 places).
PLAN: marked complete (all items shipped; deferred list was stale).
- middleware re-verifies the username against users.json per request:
deleted accounts' signed cookies are rejected immediately
- set_cookie/delete_cookie honor the Secure flag; run.sh sets
ANIWEB_SECURE=1 so sessions only work through the HTTPS proxy
- server/users.py replaces credentials.py: users.json (0600), PBKDF2,
one-time migration of the legacy single password to user admin
(history.json -> history/admin.json)
- accounts are CLI-managed only: python -m server.users add|rm|list
(no self-signup on an internet-exposed app); /signup route removed,
login page back to username + password
- per-user 30-day signed session cookie (payload carries the
username), per-user lockout, GET /api/me, POST /account/delete
- per-user watch history: server/data/history/<user>.json; ANIWEB_DATA
env override for the data dir
- header chip: logged-in user, logout, delete account
- tests/: throwaway-server suite (port 8399, temp data dir) — 9 HTTP
tests + 2 Playwright browser tests (login UI, real video playback
behind auth); runs via python tests/run.py
jozamudi