* feat(requests): enforce per-profile rating limits in discovery
- Resolve each profile's max content rating and filter discovery, detail, and browse results against it, failing closed on missing ratings
- Reject request submissions for titles above the viewer's ceiling
- Add TMDB GetCertification backed by release_dates/content_ratings with a long-lived cache and singleflight
- Push certification.lte to TMDB for studio/network/genre browse as a cost pre-filter
- Backfill restricted section pages from a fixed window of TMDB pages to keep carousels populated and pagination stable
* fix(requests): address discovery rating review findings
- Preserve backfill overflow: sections use plain TMDB cursor semantics
plus an additive next_page field instead of fixed windows, so an early
stop never drops allowed titles from unconsumed pages (bit hardest at
permissive R/TV-MA ceilings).
- Bound cold-path cost: DiscoverAll backfills at most 2 TMDB pages per
section (vs 5 for a direct section request), capping worst-case cold
certification hydration at 240 lookups instead of 600.
- Keep the TMDB prefilter a superset: rank-3 ceilings now push down
certification.lte=NC-17/TV-MA rather than R, so titles the local
ladder allows can't vanish upstream unrecoverably.
- Fail closed on foreign certifications: enforcement-path lookups use
new US-only pickers (a Canadian PG no longer reads as US PG), while
the display path keeps its any-country fallback. US multi-entry
disagreements prefer the theatrical/real rating over festival NR.
- Detach shared certification fetches from the first caller's context
(WithoutCancel + 30s bound) so one disconnecting client can't fail
the singleflight result for concurrent waiters.
- Advertise enforcement via rating_restrictions_enforced on
/requests/status so clients can feature-detect instead of
version-sniffing.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
* fix(requests): harden rating enforcement per second review pass
- GetDetail gates on the US-only enforcement certification (cached
GetCertification) instead of the display rating, whose any-country
fallback let a foreign "PG" pass the US ladder.
- pickUSMovieCertification takes the strictest recognized US rating when
multiple release entries disagree ([PG, R] -> R); entry order is not
meaningful and enforcement must not admit a title on its most lenient
certificate.
- Certification singleflight uses DoChan so a canceled caller returns
ctx.Err() immediately instead of blocking up to 30s on the detached
shared fetch (which still completes for surviving waiters).
- Viewer rating ceiling resolves once per request and threads through
discover/browse/detail enrichment (enrichPageWithCeiling); DiscoverAll
drops from 12 scope resolutions per load to 1.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
* docs: design spec for multi-instance Sonarr/Radarr request routing
Seerr-style multi-instance arr management inside Silo's request system:
many instances per kind, HD/4K default routing, entitlement-driven
dual-quality fan-out, per-instance anime overrides (keyword 210024),
and a one-to-many media_request_targets model.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* docs: implementation plan for multi-instance arr request routing
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(requests): migration for multi-instance arr routing
Adds migration 169 to convert request_integrations from a one-row-per-kind
table keyed on `kind` to a multi-instance table keyed on `id`, with HD/4K
defaults, anime overrides, and a new one-to-many media_request_targets table
for per-quality fulfillment tracking.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(requests): instance, target, and dual-quality types
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(requests): id-based integration CRUD
Replace upsert-by-kind (UpsertIntegration/UpsertIntegrations) with
GetIntegration, CreateIntegration, UpdateIntegration, DeleteIntegration,
and ClearDefault. Rewrites scanIntegration and integrationColumns to cover
all new multi-instance columns (id, name, is_4k, is_default, is_default_4k,
anime_* fields). Updates the Store interface accordingly.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(requests): target persistence and aggregate status
* feat(tmdb): expose keyword ids and original language on detail
* feat(requests): Seerr-exact anime detection (keyword 210024)
* feat(requests): quality/anime routing engine
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(requests): force_dual_quality setting
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(requests): multi-target fulfillment, reconcile, retry, and instance CRUD
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(api): request integration CRUD endpoints, targets in responses, entitlement wiring
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* feat(web): multi-instance request integration types and CRUD hooks
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* feat(web): multi-instance arr manager, dual-quality toggle, per-target queue
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(web): UX review fixes for arr manager (delete confirm, switch hints, test feedback, dirty + target status)
* fix(requests): address code-review findings (test-connection by id, HD-only default ceiling, retryable partial failure, idempotent submit, transactional defaults, presence/target reconcile, auto-approve gate)
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix: address CodeRabbit review (anime override fallback, non-null slices, save gate, a11y, DeleteTarget not-found)
- routing: anime fields only override standard root/profile/tags when set,
so enabling anime with blank fields reuses standard values instead of
clearing them into an invalid submission
- api: normalize nil Tags/AnimeTags to [] so they serialize as arrays not null
- web: require an API key before saving a NEW instance; add aria-expanded/
aria-controls to the anime-overrides disclosure toggle
- repo: DeleteTarget returns ErrNotFound when no row was deleted
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
* fix(requests): address PR review findings
---------
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Co-authored-by: Quick <31828688+Quick104@users.noreply.github.com>
- Add TTL caches with singleflight for DiscoverSection, DiscoverPage, and GetExternalIDs to reduce upstream TMDB calls
- Reuse CastCarousel with a fullBleed variant on RequestDetail and drop the bespoke RequestCastRow
- GetExternalIDs now uses the dedicated /movie/{id}/external_ids and
/tv/{id}/external_ids endpoints instead of fetching the full detail
with append_to_response=external_ids. The dedicated payload is
one or two orders of magnitude smaller for the same fields.
- Document PosterPath/BackdropPath on MediaResult as raw TMDB path
fragments that callers must prefix with the image base URL.
- normalizeCast switches from inline insertion sort to sort.SliceStable.
The output is identical; the new form is one line and O(n log n).
- normalizeIntegration no longer reuses integration.Tags' backing
array via Tags[:0]; the slice is callable code, so reusing the
array would silently corrupt the caller's slice if it kept a
reference. Allocate a fresh slice instead.
- HandleGet now requires a profile, matching the rest of the
/requests user-group handlers. Router middleware enforces this
already, but the inline check is defense-in-depth for any future
remount.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
The client previously fell back to a hardcoded project-level API key
when the caller passed an empty string. Once committed the key cannot
be rotated without a rebuild, and the silent fallback meant that
misconfiguration produced unattributed traffic against a shared key.
Remove the default. Clients constructed with an empty key now return
ErrMissingAPIKey from every API call so operators see the
configuration problem immediately.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
- Batch integration upserts in a single transaction
- Treat radarr/sonarr lookup results as arrays and require exact matches
- Prefer queue failures over downloading state when evaluating arr queues
- Allow retrying queued/downloading requests and block declines once fulfillment started
- Fall back to pending when auto-approval integration check fails
- Rename requests query hooks file and fix discover card request affordance
- Add `media_type=all` to request search, backed by TMDB `/search/multi` filtered to movies and series
- Default the Requests page filter to All and refresh search results grid styling
- Refine RequestPosterCard with status accent bar, richer fallback poster, and fluid grid layout
Wire curated TMDB-backed studios/networks/genres discovery into the requests service and UI, replacing on-demand logo fetches with fixed duotone logos and adding browse routes plus tests.
Co-authored-by: Cursor <cursoragent@cursor.com>
- Add GetMediaDetail TMDB client returning normalized detail with cast, crew, recommendations, and certifications
- Add /api/requests/detail/{media_type}/{tmdb_id} endpoint overlaying availability and request state
- Add RequestDetail page and link poster cards to it
- Treat empty/truncated Radarr/Sonarr POST responses as accepted; drop pre-submit existence lookups
- Add request domain, repository, service, and reconcile task
- Add Radarr/Sonarr fulfillment adapters and TMDB discovery
- Expose user and admin request APIs with quota and approval rules
- Add web UI for browsing, requesting, and admin queue management
- Migration 139 introduces media_requests and related tables
