Compare commits

...
Author SHA1 Message Date
timothy bb310f6bc7 Merge pull request 'chore(dev): husky pre-commit / pre-push / commit-msg guardrails' (#273) from feat/autolint-precommit into main
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 8m40s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 10m47s
Build ErsatzTV Image / Build & push image (amd64) (push) Successful in 9m27s
2026-07-11 19:07:35 +00:00
timothyandClaude Opus 4.8 7ffc12b213 docs(handoff): lore — activating IsConcurrencyToken exposes all root writers (#253 PR3 review)
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 4m47s
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 6m13s
Build ErsatzTV Image / Build & push image (amd64) (push) Has been cancelled
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 20:58:27 +02:00
timothy 2d0935b651 Merge remote-tracking branch 'origin/main' into feat/autolint-precommit
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 8s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 8m14s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 9m53s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
2026-07-11 20:57:05 +02:00
timothy 6c9b998d7d Merge remote-tracking branch 'origin/main' into feat/autolint-precommit
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 10s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 5m13s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 6m26s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
2026-07-11 20:46:01 +02:00
timothy 22440cc1e6 Merge remote-tracking branch 'origin/main' into feat/253-pr3-diff-scalar
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 5s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 8m21s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 10m12s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / Build & push image (amd64) (push) Has been cancelled
Build ErsatzTV Image / Build & test (.NET) (push) Has been cancelled
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Has been cancelled
# Conflicts:
#	docs/decisions.md
2026-07-11 20:45:22 +02:00
timothy 550d7ac984 Merge pull request '#253 PR2 — optimistic-concurrency fan-out (Template, DecoTemplate, Playlist, schedule-items)' (#268) from feat/253-pr2 into main
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 3m49s
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 7m7s
Build ErsatzTV Image / Build & push image (amd64) (push) Successful in 5m38s
2026-07-11 17:54:05 +00:00
timothy f59f2362e2 ci: retrigger — flake in unrelated LibrariesScreen scan-polling test (passes 13/13 locally)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 5s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 7m59s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 10m38s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
2026-07-11 19:45:07 +02:00
timothy c93266b6fb Merge remote-tracking branch 'origin/main' into feat/253-pr2
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 5s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 4m34s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 5m28s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
2026-07-11 19:44:39 +02:00
timothyandClaude Opus 4.8 e61f58cd99 fix(253): force-write non-participating root writers past a concurrent Version bump (PR3 review)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 7s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Failing after 2m33s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 3m30s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Adversarial review caught a HIGH the fan-out introduced: activating Version as an
IsConcurrencyToken on Playout/Collection makes EF append `WHERE Version=@orig` to
EVERY root UPDATE, so a non-If-Match writer that saves via plain SaveChangesAsync
now throws DbUpdateConcurrencyException → 500 when a replace-all editor bumps the
row between its load and save. Realistic two-tab trigger (edit playout settings while
editing its alt-schedules; edit a collection's name while reordering) — a new crash,
previously silent last-write-wins.

Fix: shared ConcurrencyExtensions.SaveChangesForcingVersion — on a concurrency
failure it adopts the stored token as original+current (client-wins merge scoped to
the token, never reverting the concurrent bump) and retries, i.e. Phase-1 force-write
semantics for a missing If-Match. Applied to the exposed UPDATE writers:
UpdatePlayout, Update{Sequential,Scripted,ExternalJson}Playout, UpdateOnDemandCheckpoint,
UpdateCollection. Non-vacuous test proves the write lands and the bump survives.

Deletes + repo-mediated Add* writers (rarer / join-rows-only) re-scoped onto #269.

Refs #253 #269

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 19:36:57 +02:00
timothyandClaude Opus 4.8 f27f458ff5 feat(web): add husky pre-commit/pre-push/commit-msg guardrails
Installs husky git hooks (via web/'s lint-staged + npm, since the JS/TS
project lives in web/ with no root package.json) to catch lint, format,
type, and generated-API-drift errors locally before they reach CI.

Hooks (committed at repo root under .husky/):
- pre-commit: (a) lint-staged runs eslint --fix on staged
  web/src/**/*.{ts,tsx} + a project-wide typecheck; (b) if any *.cs are
  staged, dotnet format --verify-no-changes on just those files (skipped
  when no .cs staged, so web-only commits skip the sln load).
- pre-push: CI-parity gate — cd web && check:api && lint && typecheck &&
  build. Blocks pushing drift or a change that breaks an unstaged file.
- commit-msg: requires a Co-Authored-By trailer (merge commits exempt).

Wiring: web/package.json gains husky + lint-staged devDeps, a lint-staged
config, and a `prepare` script (cd .. && husky) that points git's
core.hooksPath at the repo-root .husky dir on npm install. A fresh
`web/` npm install installs all four hooks automatically.

Monorepo/worktree gotchas handled:
- husky init hard-checks for .git in cwd, so `prepare` cd's to the repo
  root before invoking husky (npm keeps web/node_modules/.bin on PATH).
- git exports GIT_DIR while running hooks; in a worktree/subdir that made
  pre-push's `git diff` (check:api) mislocate the working tree and pass
  silently on drift — pre-push now unsets GIT_DIR/GIT_WORK_TREE/GIT_INDEX_FILE.

docs/ci-cd.md: new "Pre-commit hooks (web/)" section covering all four.

Verified: eslint error blocks commit; clean commit passes; bad-format .cs
blocks (dotnet format ~6-7s scoped), good .cs passes; check:api drift and
a lint error each block `git push --dry-run`, clean state passes; missing
Co-Authored-By blocks commit-msg, present passes; non-web/.cs commits skip
lint/format. npm run lint clean.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 19:31:22 +02:00
timothyandClaude Opus 4.8 bb7b189929 feat(91b): wire deep-scan + external-collections buttons into LibrariesScreen
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 6s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 4m35s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 5m22s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 4m2s
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 6m50s
Build ErsatzTV Image / Build & push image (amd64) (push) Successful in 4m13s
Last SPA pre-work before deleting Blazor Libraries.razor (#91 phase b):
wire the shipped scanLibrary(id, deep) + scanCollections(family, id, deep)
clients (F9 API, #235) into LibrariesScreen so the SPA reaches parity with
Libraries.razor's four scan actions.

- Deep Scan Library button on each remote (Plex/Jellyfin/Emby) library row,
  threading `deep` through the existing optimistic-pending/poll hook (quick +
  deep share the per-library lock).
- External Collections section (quick + deep per remote source). Rows derive
  client-side from getMediaSources(): the media-sources API handler already
  filters each source's `libraries` to sync-enabled entries, so a remote
  source with a non-empty libraries list is exactly GetExternalCollections's
  Libraries.Any(ShouldSyncItems) filter — no new endpoint.
- useCollectionsScan hook: collections scans have no scan-status poll surface
  (the endpoint is library-keyed; Blazor observed collections locks via
  in-process IEntityLocker events), so pending is optimistic + timeout-bounded
  (409 benign, 404/network surfaces the error). Follow-up #271 for a proper
  collections status surface.

Pure SPA change (no backend/OpenAPI). Docs: blazor-route-parity.md §5 (SPA
affordance DONE), decisions.md (derive-vs-endpoint + optimistic-timeout).

Refs #91

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 19:25:56 +02:00
timothyandClaude Opus 4.8 c40ffefa99 feat(253): PR3 optimistic-concurrency fan-out — Diff + Scalar aggregates
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 5s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 8m11s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 9m51s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Fans the frozen Block recipe (api-conventions §7a) across the five Diff/Scalar
replace-all endpoints, completing the #253 PR2→PR4 arc's implementable core:

- #6 Collection custom-order, #7 Playout alternate-schedules, #8 Playout templates
  (shared Playout.Version), #9 MultiCollection, #10 RerunCollection — each: pre-check
  412 as a standalone Either after validation (H2, subtype survives the Join flatten),
  unconditional Version++ (M1), guarded save, controller If-Match/ETag/400/412, SPA
  editor ETag round-trip + 412 conflict dialog.
- H1: the two Playout handlers' catch(Exception)→422 restructured so the guard's
  PreconditionFailedError returns before the catch (412, not 422).
- M2: RerunCollection/Collection refresh runs unconditionally on save; MultiCollection
  keeps its name-only→no-rebuild optimization by bumping on the first (name) save.
- H3: UpdateDefaultDecoHandler bulk-bumps Playout.Version via .SetProperty.
- Shared ConcurrencyHeaders.MalformedIfMatchProblem() for the 400 guard.
- Deferred (→ #269): same-root non-bulk sibling config writers' ETag rotation.

Tests: per-handler pre-check/bump concurrency tests (Playout ×2 incl. non-vacuous
racing-save backstop, Rerun, Multi incl. name-only-no-rebuild M2, Collection);
controller tests get a DefaultHttpContext for the header read/write. Docs:
api-conventions §7a fan-out status, spa-conventions §4a list-editor note, decisions.

Refs #253

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 19:23:48 +02:00
timothy 7bc84fa369 fix(#253 PR2): SPA lint — move conflict-reload gating out of loadItems
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 5s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 5m15s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 6m19s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
The prior fix set setItemsLoading/setItemsLoaded at the top of loadItems, but the
activeId effect calls loadItems synchronously → react-hooks/set-state-in-effect lint
error (Main Lint SPA, the real CI failure). Move the not-loaded/loading gating into
reloadAfterConflict (an event handler, lint-clean), which is exactly the 412 conflict-
reload path Codex's Medium-3 targeted; canEdit stays false through that reload window.
Behavior unchanged; the normal switch/initial-load paths (guarded separately by the
#242 dirty-guard) are left untouched. Local: lint clean, vitest 667, check:api no drift,
full dotnet solution test green.
2026-07-11 19:15:45 +02:00
timothy 1a24298105 fix(#253 PR2): close review findings (ETag/items consistency + SPA load ordering)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 7s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Failing after 40s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 4m2s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Independent Codex review of #268 found two Blockers the fork missed + two Mediums:
- Blocker: replace PUTs returned the handler's item snapshot but re-queried the root
  for the ETag separately, so a racing writer could pair stale items with a newer ETag
  (silent overwrite). All four controllers now reload root-then-items (version-first,
  fail-safe) and 404 when the root is gone between commit and reload — matching the
  Block reference. Fixes the Blocker + the Medium '200 without ETag' case together.
- Blocker: PlaylistsScreen loaded items+root via Promise.all (concurrent), pairing a
  stale name with the current ETag; now sequential (items-with-meta first, then root).
- Medium: SchedulesScreen loadItems now marks not-loaded/loading up front so canEdit is
  false through the 412 conflict reload (no stale-draft edits lost).

Controller unit-test mocks updated to stub the new reload query. Full suite green
(ErsatzTV.Tests 1334, web 667, check:api no drift).
2026-07-11 19:08:25 +02:00
timothyandClaude Opus 4.8 3981abc7f9 docs(handoff): lore — enumerate ALL channel/lock producers before a no-cross-release verdict (#235/#267)
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 8m17s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 10m3s
Build ErsatzTV Image / Build & push image (amd64) (push) Successful in 7m3s
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 19:07:29 +02:00
timothy 4aeabada3b Merge remote-tracking branch 'origin/main' into feat/235-async-contract
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 5s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 8m24s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 10m2s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / Build & push image (amd64) (push) Has been cancelled
Build ErsatzTV Image / Build & test (.NET) (push) Has been cancelled
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Has been cancelled
2026-07-11 18:53:43 +02:00
timothy b063bc45c0 docs(#253 PR2): api-conventions §7a — fan-out landed + schedule body-version nuance
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 6s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 4m28s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 5m29s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
2026-07-11 18:43:21 +02:00
timothyandClaude Opus 4.8 589c35d357 fix(235): make QueueShowScanResult switch total (explicit NotFound + throwing fallback)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 6s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 8m33s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 9m59s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Codex-review Low: NotFound previously reached the catch-all by coincidence; a
future enum value would silently 404. Explicit arm + UnreachableException fallback
so an unmapped outcome fails loudly rather than mis-mapping to 404.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 18:41:33 +02:00
timothy b5c900dfbf chore(#253 PR2): regenerate OpenAPI spec + TS client after fan-out merge 2026-07-11 18:40:43 +02:00
timothy 832d482f2c merge: #253 PR2 schedule-items 2026-07-11 18:38:14 +02:00
timothy 6046bc622e merge: #253 PR2 Playlist 2026-07-11 18:38:05 +02:00
timothy 7b78a393cf merge: #253 PR2 Template+DecoTemplate 2026-07-11 18:37:59 +02:00
timothyandClaude Opus 4.8 787058d18c fix(235): scheduler-safe collections lock ownership (Codex High / Fable reconciliation)
The new POST /api/media-sources/{plex|jellyfin|emby}/{id}/scan-collections
endpoints acquire a per-provider collections lock (409 if held) and hand the
single release to the ScannerService finally. But SchedulerService's periodic
collection scans were enqueued WITHOUT the lock, and ScannerService's finally
released the collections lock whenever held with no ownership check. A
scheduler-queued scan running while an API request held the lock cross-released
the API's lock (#250 bug class), letting a second API request get a spurious
202 instead of 409.

Fix (mirrors the SynchronizePlexLibraryByIdIfNeeded(Unlock: !networksFollow)
library-scan precedent):
- Add `bool Unlock = true` (4th positional param) to the three
  Synchronize{Plex,Jellyfin,Emby}Collections records; default keeps the
  controller + Libraries.razor call sites compiling and releasing on run.
- ScannerService: the three collection finallys now honor `request.Unlock`
  (the concrete typed request is in scope in each method) so a batch member
  with Unlock:false never releases a lock it doesn't own.
- SchedulerService: replace the unlocked per-source enqueue with a lock-once
  per-provider batch — LockX Collections() once, enqueue each source with
  Unlock:isLast (last message owns the release), compensating unlock in catch,
  and SKIP the whole provider loop if the lock is already held. A naive
  "lock-per-source, skip if held" would deterministically starve the 2nd+
  source; lock-once-batch does not.

Tests (ErsatzTV.Tests/Services/): ScannerServiceCollectionLockTests drives the
real ScannerService read loop + real EntityLocker and asserts Unlock:false
leaves a held lock intact while Unlock:true releases (all three providers);
SchedulerServiceCollectionLockTests reflect-invokes ScanPlexMediaSources and
asserts it locks once + skips the enqueue when held, and hands the release to
the last message when acquired. Proven non-vacuous: reverting the Plex fix
fails exactly the three Plex tests.

No OpenAPI/v1.json change (internal channel-message record, not a DTO).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 18:37:35 +02:00
timothyandClaude Opus 4.8 611924c0ee feat(#253 PR2): optimistic-concurrency contract for Template and DecoTemplate
Wire the frozen ETag/If-Match/412 recipe (Block reference implementation)
onto the Template and DecoTemplate aggregates:

- ReplaceTemplateItems / ReplaceDecoTemplateItems commands gain
  Option<int> ExpectedVersion; ToCommand() on the request DTOs threads it
  through from If-Match.
- Handlers introduce the version check as a standalone Either after
  validation (never via Apply), bump Version unconditionally before
  saving, and persist through SaveChangesWithConcurrencyGuard so a losing
  writer maps to 412 instead of 500. DecoTemplate's post-commit playout
  Reset enqueue now only runs after a successful save.
- TemplateViewModel / DecoTemplateViewModel carry Version (header-only,
  not echoed in the response body), populated in Mapper.
- TemplateController / DecoTemplateController: GET items emits a strong
  ETag of the root's version; PUT parses If-Match (400 on malformed),
  threads the expected version into the command, and returns the new
  ETag from the refreshed root on success. Both PUT actions now use the
  handler's returned item list directly instead of re-querying items.
- SPA: templates.ts / decoTemplates.ts gain getXItemsWithMeta and an
  If-Match-aware replaceX; TemplateEditor / DecoTemplateEditor hold the
  ETag in a ref, read items-with-meta first on load, and open a
  "changed elsewhere" ConfirmDialog on a 412 instead of navigating away.

Tests: new ReplaceTemplateItemsHandlerConcurrencyTests /
ReplaceDecoTemplateItemsHandlerConcurrencyTests mirror the Block
concurrency contract tests (stale/matching/absent If-Match, no-op bump,
racing-save 412, non-vacuous backstop). TemplateControllerTests /
DecoTemplateControllerTests gain ETag/If-Match/412 coverage.
TemplatesScreen.test.tsx / DecoTemplatesScreen.test.tsx gain a 412
conflict-dialog test mirroring BlocksScreen's.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 18:36:35 +02:00
timothyandClaude Opus 4.8 5c9f04fdec feat(#253 PR2): optimistic-concurrency on schedule-items aggregate
Wire the frozen #253 ETag/If-Match/412 recipe onto ProgramSchedule /
schedule-items, keeping the PR#258 positional in-place reconcile intact.

Backend:
- ReplaceProgramScheduleItems command gains Option<int> ExpectedVersion;
  ReplaceScheduleItemsRequest.ToCommand threads it.
- Handler: standalone CheckVersion Either AFTER validation (so 412 isn't
  flattened to 422), unconditional Version++ before save, guarded save via
  SaveChangesWithConcurrencyGuard, and 412 propagated without running the
  post-save reload/enqueue.
- ProgramScheduleViewModel + Mapper carry Version.
- ScheduleController: GET /items emits ETag; PUT /items parses If-Match
  (malformed -> 400), threads ExpectedVersion, re-queries for the new ETag,
  and advertises 400/412.
- Sibling config-writers (Add/Delete item, Update schedule) bump Version.

Frontend:
- schedules.ts: getScheduleItemsWithMeta + replaceScheduleItems(ifMatch)
  returning ResponseWithMeta.
- SchedulesScreen: etagRef threaded through the #242 dirty-guard (set from
  load + every successful save); 412 opens a conflict ConfirmDialog whose
  Reload discards the draft and re-runs loadItems.

Tests: handler concurrency suite (stale->412 no mutation + fill-group state
untouched, match/absent success+bump, no-op still bumps, racing save->412);
controller ETag/If-Match/412 cases; SchedulesScreen 412-conflict-dialog test.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 18:36:25 +02:00
timothy 1a8c0f60de feat(playlists): wire optimistic-concurrency contract onto Playlist (#253 PR2)
Fans the frozen ETag/If-Match/412 recipe (Block reference, #253) onto the
Playlist aggregate:

- ReplacePlaylistItems command carries ExpectedVersion; the handler runs
  CheckVersion as a standalone Either after validation (so a stale write
  survives as 412, not flattened to 422 by Apply/Join), bumps Version
  unconditionally before saving, and persists via
  SaveChangesWithConcurrencyGuard (EF concurrency-token backstop).
- PlaylistViewModel carries Version; the items GET sets a strong ETag and
  the PUT parses If-Match, threads it into the command, and returns the
  refreshed ETag on success (400 on a malformed If-Match).
- Sibling item-adding handlers (AddItemsToPlaylist, AddMovie/Episode/
  Season/ShowToPlaylist) bump Version too, since they mutate the same
  editor-visible item list.
- SPA: playlists.ts exposes getPlaylistItemsWithMeta and an
  If-Match-aware updatePlaylist; PlaylistEditor holds the ETag in a ref,
  round-trips it on save, and opens a "changed elsewhere" ConfirmDialog on
  412 (mirrors BlockEditor).

Tests: new ReplacePlaylistItemsHandlerConcurrencyTests (stale/match/
force-write/no-op-bump/racing-save), new PlaylistController tests
(ETag on GET items, 400/412/thread-version/force-write on PUT), and a
vitest 412-conflict-dialog test for PlaylistsScreen. dotnet test:
1304/1304 green. web: npm run typecheck clean, npm run build clean,
vitest 664/664 green.

Ref #253 PR2.
2026-07-11 18:33:40 +02:00
timothy 3efb2ac4e1 Merge remote-tracking branch 'origin/main' into feat/235-async-contract
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 8s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 7m30s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 5m51s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
# Conflicts:
#	docs/decisions.md
2026-07-11 18:11:34 +02:00
timothy 13cd00c8fe Merge remote-tracking branch 'origin/main' into feat/254-mutation-hardening
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 6s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 4m23s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 5m49s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 6m31s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 9m21s
Build ErsatzTV Image / Build & push image (amd64) (push) Successful in 6m10s
# Conflicts:
#	docs/api-conventions.md
#	docs/decisions.md
2026-07-11 18:10:30 +02:00
timothyandClaude Opus 4.8 d32ca976f7 feat(235): SPA clients for deep/collections scan + typed reset-all; docs
- libraries.ts: scanLibrary(id, deep), new scanCollections(source, id, deep),
  corrected stale scanShow status-code comment (400 -> 202/404/409/422)
- playouts.ts: resetAllPlayouts returns typed ResetAllPlayoutsResponseModel body
- libraries.test.ts: deep-scan + scanCollections client tests
- decisions.md: #235 async-op contract + F9 endpoints + accepted-by-design channels note
- blazor-route-parity.md §5: F9 API gate closed; SPA deep/collections buttons = removal-PR work

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 18:10:02 +02:00
timothy 2de091ea4f Merge pull request '#253 PR1 — optimistic-concurrency contract (infra + Block reference)' (#263)
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 8m15s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 10m38s
Build ErsatzTV Image / Build & push image (amd64) (push) Has been cancelled
2026-07-11 16:06:11 +00:00
timothyandClaude Opus 4.8 a27cfc475e chore(235): regenerate OpenAPI artifacts + endpoint index after slice merge
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 18:04:33 +02:00
timothy 61c8556ec7 Merge branch 'feat/235-s2-libraries' into feat/235-async-contract 2026-07-11 18:02:58 +02:00
timothy 321a34c748 Merge branch 'feat/235-s3-playouts' into feat/235-async-contract 2026-07-11 18:02:58 +02:00
timothy b39f398df7 Merge branch 'feat/235-s4-trakt' into feat/235-async-contract 2026-07-11 18:02:58 +02:00
timothyandClaude Opus 4.8 628c9d7228 feat(235): F9 API parity — library deep-scan, external-collections scan, scan-show outcome enum (#235 slice B)
Closes the two F9 Libraries.razor parity gaps and normalizes scan-show error
mapping to ProblemDetails.

TASK 1 — library-wide deep scan:
- QueueLibraryScanByLibraryId gains optional `bool DeepScan = false`; handler
  threads it into ForceSynchronize{Plex,Jellyfin,Emby}LibraryById.
- POST /api/libraries/{id}/scan?deep=false binds it via [FromQuery].

TASK 2 — external-collections scan (new endpoints):
- POST /api/media-sources/{plex|jellyfin|emby}/{id}/scan-collections?deep=false
  acquires the per-source collections lock (§3b: lock IS the running scan → 409),
  enqueues Synchronize{X}Collections(id, ForceScan:true, deep) to the scanner
  channel, returns 202; compensating-unlock on enqueue throw.

TASK 3 — scan-show normalization:
- New QueueShowScanResult enum; handler returns it instead of bool.
- POST /api/libraries/{id}/scan-show now maps 202/404/409/422 (all errors
  ProblemDetails) instead of 200/404/400-anonymous-object.
- Updated the lone Blazor caller (TelevisionSeasonList.razor).

Tests: LibrariesController (scan deep=true, scan-show enum→status), the three
media-source controllers (scan-collections route/404/409/202/compensating-unlock),
and handler tests for both changed handlers (deep threading + show-scan outcomes).
Docs: api-conventions §3b exemplar + blazor-route-parity §5 F9 gate.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 18:02:17 +02:00
timothyandClaude Opus 4.8 9b73b62527 feat(235): async-op API contract normalization — playouts slice C (#235)
Slice C of the async-op contract normalization:

- channel reset (POST /api/channels/{channelNumber}/playout/reset) now
  returns 202 Accepted (was 200 Ok) — it only queues a background rebuild
- reset-all (POST /api/playouts/reset-all) still 202 but now returns a
  ResetAllPlayoutsResponseModel body reporting QueuedPlayoutIds /
  SkippedLocked / SkippedUnsupported instead of silently swallowing skips;
  handler returns a new ResetAllPlayoutsResult record
- single-playout GET (GET /api/playouts/{id}) now exposes IsLocked on
  PlayoutResponseModel, set from IEntityLocker.IsPlayoutLocked mirroring
  the list projection — gives a polling client the lock flag

Tests: channel reset asserts 202; reset-all asserts 202 + skipped-body
shape; single GET asserts IsLocked; new ResetAllPlayoutsHandlerTests
(in-memory SQLite) asserts locked/ExternalJson/None land in skipped lists
and eligible playouts in queued. docs/api-conventions.md §3a updated.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 17:58:15 +02:00
timothyandClaude Opus 4.8 d02f953922 fix(235-F7): release leaked Trakt lock on worker shutdown
The global Trakt lock is acquired by SchedulerService.RefreshTraktLists /
MatchTraktLists (and TraktController) and released only when the *terminal*
message of a batch — the one carrying Unlock: true (list == traktLists.Last())
— is processed by WorkerService, whose handler (AddTraktListHandler /
MatchTraktListItemsHandler) calls IEntityLocker.UnlockTrakt() in a finally.

WorkerService.ExecuteAsync breaks out of the read loop on
stoppingToken.IsCancellationRequested (and exits on channel completion /
reader cancellation) BEFORE processing the next message. If shutdown lands
after a batch is enqueued but before its terminal Unlock: true message is
handled, UnlockTrakt() never runs and the in-memory Trakt lock leaks for the
rest of the process lifetime (subsequent Trakt operations 409 forever).

Fix (option a): make the batch-release loss-tolerant with a compensating
release in a finally around the read loop — if the Trakt lock is still held
when the worker stops, release it. Chosen over tracking pending ownership
(b) because the lock is a global singleton and WorkerService is its sole
batch-release site, so "held at shutdown" unambiguously means "the terminal
release was lost"; covers all three exit paths (break / channel completion /
cancellation) in one place. Same lock-lifecycle class as #231/#233/#234.

Regression test: WorkerServiceTests gates the first (non-terminal) batch
message on the stopping token, then StopAsync-cancels so the worker breaks
before the terminal Unlock: true message — asserts the lock is released and
the terminal message was never processed. Proven non-vacuous: inverting the
finally condition fails the test.

Backend-only; no controller/DTO/SPA/OpenAPI impact.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 17:57:41 +02:00
timothyandClaude Opus 4.8 5e5f0af684 fix(235-A): normalize async-op error contracts on Maintenance + Troubleshoot controllers (#235)
Slice A of the async-op API contract normalization.

MaintenanceController:
- EmptyTrash error path: was 500 text/plain (error.ToString()); now maps the
  BaseError Left through ApiResults.ToErrorResult() -> 404 (NotFoundError) / 422
  ProblemDetails. Success stays 200 OkResult. Added ProducesResponseType 200 + 422.
- CleanArtwork: fire-and-forget enqueue of DeleteOrphanedArtwork was a silent 200;
  now returns 202 Accepted (AcceptedResult) since it queues background work.
  Added ProducesResponseType 202. (Controller does not derive from ControllerBase,
  so results are built directly as before.)

TroubleshootController.TroubleshootPlayback (GET|HEAD /api/troubleshoot/playback.m3u8):
- Two bare body-less NotFound() call sites conflated "not found" with "prepare/
  playback failure". Both now return a ProblemDetails body:
  * prepare-failure (result.IsLeft): mapped through error.ToErrorResult() -> 404 for
    NotFoundError (unknown media item/channel) else 422 for a validation BaseError.
  * terminal fall-through (prepare ok but no playable output): kept 404 with a
    distinguishing ApiResults.NotFoundProblem(...) detail.
- Added ProducesResponseType 404 + 422 (409 already present).

Consumer check: the SPA (PlaybackTroubleshootingScreen) feeds the playback.m3u8 URL
straight to hls.js via HlsPlayer, which never inspects the HTTP status code — playback
state is surfaced via the separate /api/troubleshoot/playback/status poll. So the
404->422 split for the validation subcase is safe; no player code branches on the
status code.

Tests: MaintenanceControllerTests (200/422/202 + enqueue assertion),
TroubleshootControllerTests (prepare 404 NotFoundError, 422 validation). All green;
Api error-metadata/contract/security scans still pass.

Note: OpenAPI artifacts (v1.json / v1.d.ts) intentionally NOT regenerated here — the
orchestrator regenerates once after all #235 slices merge.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 17:54:21 +02:00
timothyandClaude Opus 4.8 6055bd3687 fix(app): fold Create/DeleteSmartCollection into the post-commit None sweep (#254 review)
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 8m27s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 10m0s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 35s
Independent adversarial review of PR #266 found two sibling handlers with the
identical post-commit `_smartCollectionCache.Refresh(cancellationToken)` pattern
that the sweep missed (only UpdateSmartCollection was caught). Same audit#22 F4
class: a late client-disconnect after the commit lands would throw and leave the
in-memory smart-collection cache stale vs the committed DB. → CancellationToken.None.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 17:49:16 +02:00
timothyandClaude Opus 4.8 a1bd303cce fix(app): post-commit side effects on CancellationToken.None + guide-xml/empty-list hardening (#254)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 6s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Failing after 18m17s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Failing after 18m41s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been cancelled
Backend mutation-hardening cluster from the 2026-07-11 mutation-safety audit
sweep (adversarial-reviewer #22/#23), the parallel-safe backend-isolated slice.

audit#22 F4 — standardize post-commit enqueues on CancellationToken.None:
20 command handlers under MediaCollections/, ProgramSchedules/, Playouts/,
Channels/ threaded the request cancellationToken into work that runs AFTER
SaveChangesAsync commits (WriteAsync rebuild/refresh enqueues, mediator.Publish,
reindex, cache Refresh, and post-commit lookups that gate an enqueue). A late
client-disconnect then turns an already-durable commit into a thrown request AND
drops the side effect. Generalizes the #251 deco-handler fix. Excludes
BuildPlayoutHandler (worker/background token, not a client-disconnect token),
the config/FFmpeg multi-upsert handlers (partial-commit case, separate
follow-up), and response-projection reloads (correctly keep the request token).

audit#22 F2 — DeleteChannelHandler/DeletePlayoutHandler now delete the channel
guide {number}.xml through IFileSystem.File.Delete (observable under
MockFileSystem) and BEFORE the commit (a post-commit delete orphans the xml on a
crash; the xml is regenerable on demand, so pre-commit delete is the safe order).

audit#23 F4 — ReplacePlayoutAlternateScheduleItemsHandler rejects an empty item
list in the handler (not only the controller pre-guard) so a direct caller can't
trip the Max()-on-empty crash.

Docs: api-conventions.md §7a (post-commit token convention + boundaries),
decisions.md entry (rationale, sweep scope, #253 PR2-4 coordination note).
Tests: guide-cache-delete-through-FS for both delete handlers, empty-list guard.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 17:42:45 +02:00
timothy 8e2f995492 Merge pull request 'feat(media-sources): SPA management screens + write REST API (#202)' (#262) from feat/202-media-sources into main
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 4m22s
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 5m31s
Build ErsatzTV Image / Build & push image (amd64) (push) Failing after 18m15s
2026-07-11 15:38:24 +00:00
timothy 7d106bf810 Merge remote-tracking branch 'origin/main' into feat/253-optimistic-concurrency
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 6s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 8m38s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 10m12s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
# Conflicts:
#	docs/decisions.md
2026-07-11 17:37:12 +02:00
timothy 6ed36b4bac Merge remote-tracking branch 'origin/main' into feat/202-media-sources
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 6s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 4m38s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 5m59s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
# Conflicts:
#	docs/blazor-route-parity.md
#	docs/decisions.md
2026-07-11 17:31:04 +02:00
timothyandClaude Opus 4.8 b285e33747 docs(handoff): lore — prose cross-refs go stale after a parallel merge, re-check them (#205/#206)
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 5m19s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 4m19s
Build ErsatzTV Image / Build & push image (amd64) (push) Successful in 4m9s
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 17:22:51 +02:00
timothy 8878bf9e11 docs(review): record deferred If-Match 412-semantics refinement (#265) as an acceptable-defer
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 6s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 5m27s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 9m34s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Codex re-review of the fix commit confirmed both prior findings resolved and raised one
new Medium: RFC 7232 would 412 (not 400) a syntactically-valid but non-matching If-Match
(non-canonical "03", weak W/"3", tag lists, empty, overflow). Deferred to #197 (cold
contract pass) as #265 — fail-safe today (the mutation is rejected, never applied) and no
first-party client is affected. Records the deferral where the #253 fan-out will copy the
parser: a code comment in ConcurrencyHeaders + a note in api-conventions §7a.

Refs #253 #265
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 17:18:35 +02:00
timothy 9b83be57f5 fix(202): clear dirty flag before post-create navigation so the guard doesn't spuriously prompt (live E2E)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 7s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 4m28s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 7m23s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
2026-07-11 17:17:35 +02:00
timothy ee39effe0b fix(review): close client load-TOCTOU + canonicalize If-Match parse (Codex High/Medium)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 7s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 4m39s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 5m30s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Codex independent review of #263 surfaced two defects the fork review missed:

- High — client load TOCTOU: BlockEditor read root metadata (getBlock) and items+ETag
  (getBlockItemsWithMeta) concurrently, so a concurrent write landing between them (with
  the items read resolving last) left a stale root paired with a current ETag → the save
  silently overwrote the concurrent change with no 412. Fix: read items+ETag FIRST, then
  the root metadata, so the captured ETag is never newer than the root version and any
  inconsistency fails safe (save 412s → conflict dialog → reload).
- Medium — `ParseIfMatch` accepted non-canonical strong tags ("03", "+3", " 3 ") as
  version 3. An ETag is opaque; only the exact emitted form is valid. Fix: canonical
  decimal only (`NumberStyles.None` + no leading zeros) → else 400.

Tests: new `ConcurrencyHeadersTests` (canonical parse + padded/signed/whitespace/weak/
unquoted/list/overflow/empty → malformed); `ApiResultsTests` gains the 412 mapping case.
Existing BlocksScreen tests still green (load reordering is behavior-preserving for the
non-concurrent path).

Refs #253
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 17:13:26 +02:00
timothyandClaude Opus 4.8 7608cccebd docs(91b): auth-posture sign-off + rollback-tag procedure for Blazor removal (#205, #206)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 7s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 8m13s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 10m35s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Has been cancelled
Build ErsatzTV Image / Build & push image (amd64) (push) Has been cancelled
Build ErsatzTV Image / Build & test (.NET) (push) Has been cancelled
Resolve the two SHOULD-FIX gate findings from the #91 cold review by making the
removal plan address them explicitly instead of clearing the gate by omission.

Pages (verified in code, not assumed) — OIDC's AuthorizeFolder("/") gates only the
Blazor _Host Razor Page; /app (SPA) and /api/* were already unauthenticated since
phase (a); /iptv JWT + API-key filters are independent of Blazor and survive
removal. Sign-off: no capability lost, no NEW exposure beyond phase (a); real
SPA/API auth deferred to #197. Recorded in docs/decisions.md.

(cut at removal time on the pre-deletion main commit — not a v* tag, no release
build) + the restore path (checkout+build+pin test container, or revert the merge).
Recorded in docs/decisions.md.

Both fold into a new "Section 5 — Removal execution runbook" in blazor-route-parity.md
so the (gated) removal PR has an ordered checklist. Docs-only; no code change.

refs #205 #206 #91

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 17:09:29 +02:00
timothy 9a1ddec71a fix(202): make RemoteLibrariesEditScreen reviewable (NUL delimiter -> \u0000 escape) + Plex library-prefs missing-id 422 guard (review)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 7s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 5m28s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 6m26s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
2026-07-11 16:52:05 +02:00
timothyandClaude Opus 4.8 94ebf34ccd feat(api): optimistic-concurrency contract for replace-all PUTs — PR1 infra + Block reference (#253)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 7s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 5m24s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 4m25s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Adds the shared optimistic-concurrency contract so a stale second tab can no longer
silently overwrite a fresher edit. PR1 lands the infra + the Block reference aggregate;
PRs 2–4 fan the same recipe across the other 8 roots (design: #253#issuecomment-8472).

Contract
- `IVersionedAggregate` (`int Version`) on all 9 replace-all roots (ProgramSchedule,
  Block, Template, DecoTemplate, Playlist, Collection, Playout, MultiCollection,
  RerunCollection), EF-mapped `.IsConcurrencyToken()`; one dual-provider migration
  `AddAggregateVersions` (nullable:false, default 0).
- Strong `ETag` of `Version` on the aggregate GET; `If-Match` on the PUT; mismatch →
  412 (distinct from the §3a 409 build-lock guard). Successful PUT returns the new ETag.
- `PreconditionFailedError : BaseError` → 412 in `ApiResults.ToErrorResult`;
  `ConcurrencyHeaders.ParseIfMatch/SetETag`; malformed If-Match → 400; `*`/absent =
  Phase-1 force-write.

Block reference wiring
- Handler: standalone `Either` via `CheckVersion` AFTER validation (never through
  `Apply`, which Join()-flattens the subtype to 422), unconditional `Version++`,
  `SaveChangesWithConcurrencyGuard` backstop (DbUpdateConcurrencyException → 412).
- `BlockViewModel.Version` (header-only, not echoed in the body); controller sets the
  ETag on GET items and on the successful PUT.
- SPA: `client.requestWithMeta` seam; `blocks.getBlockItemsWithMeta` + `replaceBlock`
  If-Match/ETag round-trip; `BlockEditor` holds the ETag, sends If-Match, and on 412
  opens a blocking "changed elsewhere — reload" dialog.

Tests
- Handler contract tests: stale-If-Match → 412 (no mutation), matching/absent → success
  + bump, no-op save still bumps, and a two-context racing save → 412; proven
  non-vacuous (drop `.IsConcurrencyToken()` → the race test fails).
- Controller tests: malformed If-Match → 400, If-Match threaded to the command, ETag on
  GET/PUT, 412 passthrough. SPA: requestWithMeta ETag, replaceBlock If-Match, 412 dialog.

Docs: api-conventions §7a, spa-conventions §4a, domain-model glossary, decisions log.

Refs #253
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 16:51:59 +02:00
timothyandClaude Opus 4.8 495e450e2c test(ui): extend redirect guard meta-test to Tier-2 pattern templates (#204 review)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 6s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 4m41s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 5m33s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 4m11s
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 6m48s
Build ErsatzTV Image / Build & push image (amd64) (push) Successful in 3m42s
Fork adversarial review nit: Map_Keys_Should_Not_Begin_With_Forbidden_Prefix
covered only Tier-1 Map keys, not the Tier-2 PatternRule templates. That guard
invariant is the load-bearing protection for the un-prefix-guarded /api|/artwork|
/docs|/openapi surface, so make it self-enforcing over ALL rules — a future
prefix-violating template now fails the test instead of slipping through.

Exposes internal LegacyUiRedirects.PatternTemplates (InternalsVisibleTo already
set for ErsatzTV.Tests); stores the raw template on PatternRule.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 16:45:03 +02:00
timothy b9a2fdec50 Merge remote-tracking branch 'origin/main' into feat/202-media-sources
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 7s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 5m27s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 10m4s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
2026-07-11 16:45:03 +02:00
timothy 12cbff01f9 docs(202): parity verdict, domain-model routes, decisions, capability matrix (#202) 2026-07-11 16:43:39 +02:00
timothyandClaude Opus 4.8 8a2238b62e fix(ui): pattern-based legacy→SPA redirect matcher for parameterized routes (#204)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 5s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 7m41s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 10m28s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Extend LegacyUiRedirects from an exact-match dictionary to a two-tier matcher:
Tier 1 keeps the exact Map (now 52 entries incl. the ?kind= browse roots),
Tier 2 adds 36 ordered segment-template PatternRules for id-carrying routes.
{id} is a strict positive integer (non-int/0/neg/overflow falls through), which
also makes the rule set collision-free by construction. New AppendQueryString
helper merges the incoming query into ?kind= targets with '&' (kills the
double-'?' bug); one-line Startup change keeps the redirect GET/HEAD-only 302
before UseRouting.

Completes phase-(a) Step 1 for every PARITY-OK route (#91 phase b); the
catch-all fallback replacing MapFallbackToPage stays with the removal PR.
/media/sources/* (#202) and /system/health remain deliberately un-redirected.

fixes #204

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 16:39:56 +02:00
timothy 7cddbb4bf8 Merge branch 'feat/202-s6b-remote' into feat/202-media-sources
# Conflicts:
#	web/src/App.tsx
2026-07-11 16:38:26 +02:00
timothyandClaude Opus 4.8 edbecc01d7 feat(spa): S6b — Plex/Jellyfin/Emby Remote media-source screens (#202)
Build the Remote media-source SPA screens over the S5 foundation, replacing
the MediaSourceEditorPlaceholder for the plex/jellyfin/emby dispatch branches
only (Local branches left for S6a):

- PlexSourceScreen: pin-flow sign-in / fix-credentials / sign-out with the
  §C1 poll state machine — polls GET /api/media-sources/plex every 2s up to
  150s and keeps polling while authorized-but-locked ("finalizing"); the
  terminal success is the lock releasing. Popup-blocked fallback link. Server
  table (Refresh disabled while locked / Edit Libraries / Edit Path
  Replacements) + sign-out content-removal confirm dialog.
- RemoteSourceScreen (shared Jellyfin/Emby): connect / edit-connection /
  disconnect (warning dialog) + server table.
- RemoteConnectionEditScreen (shared): secure key affordance (§C3/finding 1)
  — address prefilled, "leave blank to keep" when hasApiKey, required on first
  connect; stored key never rendered or requested.
- RemoteLibrariesEditScreen (shared): client-side sortable Name + MediaKind
  columns, per-library sync Switch, one Save; draft keyed by (name,mediaKind)
  not id, refetch after save (ids change on disable, §C4a).
- PathReplacementsEditScreen (shared): row list + selected-row edit form,
  add/remove, one Save; both fields required; family remote-path column label.

All editors use the ChannelEditScreen draft/save model + a shared useDirtyGuard
(registerNavigationGuard + beforeunload), Save gated !valid||!dirty||saving,
draft retained on 422/network, destructive actions gated on saving, 409 →
refetch. Colocated tests cover the poll (waiting→finalizing→success asserting
it does NOT stop at authorized&locked, timeout, budget-exhausted), the secure
key affordance, sortable columns, draft-retained-on-422, dirty-guard veto, and
the disconnect/sign-out dialogs.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 16:33:41 +02:00
timothyandClaude Opus 4.8 ff72eb4d5f feat(spa): build LocalLibraryEditScreen for local media libraries (#202 slice S6a)
Adds the Local library create/edit editor (create at /app/libraries/local/new,
edit at /app/libraries/local/{id}) wired into the S5-built LibrariesRouteScreen
dispatch switch, replacing MediaSourceEditorPlaceholder for the local-new and
local-edit sub-routes only. Remote (Plex/Jellyfin/Emby) branches are untouched
(S6b).

- Name (required) + Media Kind (create-only, disabled+annotated on edit)
- Add Path: path-exists pre-check (L7) + in-draft duplicate detection
  (mediaSources/paths.ts normalizePath)
- Delete path: draft-local removal with a media-item-count confirm dialog
- Move path: dialog filtered to same-MediaKind libraries excluding the source,
  including "(New Library)" which composes createLocalLibrary + moveLocalLibraryPath
  (surfaces the error and leaves the new empty library on a failed move, matching
  Blazor); gated on !dirty to avoid clobbering unsaved edits with the post-move
  refetch
- Draft/saved model with explicit Save (POST L3 / PUT L4), draft retained on
  422/network error, dirty-guard (registerNavigationGuard + beforeunload)
- Delete library (L5) with a media-item-count confirm; 409 refetches detail

Extended the existing App.test.tsx App-owned-popstate regression test (design
§D.2) to exercise the real screen's dirty guard instead of a manually-armed
stand-in, now that S6a has landed the editor it was stubbing out for.

Verification (web/): vitest (632 passed), eslint clean, tsc -b + vite build
clean, check:api reports no drift.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 16:24:55 +02:00
timothyandClaude Opus 4.8 6ce448d265 feat(spa): media-source SPA foundation — shared client, helpers, App-owned popstate wrapper (#202 slice S5)
Owns the shared single-files so the S6a (Local) / S6b (Remote) editor slices touch
disjoint files. Editor screens are stubbed (MediaSourceEditorPlaceholder) for S6.

- web/src/api/mediaSources.ts (+test): client module over the new media-source write
  endpoints (local CRUD + move/path-exists; Plex pin-flow/sign-out; shared remote
  state/connection/libraries/path-replacements/refresh; family = only URL variance),
  DTOs re-exported from generated v1, messageFromMediaSourcesError; barrel export.
- web/src/mediaSources/{familyMeta,paths,pinFlowPoll}.ts (+tests): family labels/routes/
  remote-path column naming (owns RemoteFamily); client-side NormalizePath mirror for
  in-draft dup detection; pure §C1 pin-flow poll state machine (waiting/finalizing/
  success/timeout/budget-exhausted), timer-free and fully unit-tested.
- routing.ts parseLibrariesSubRoute + LibrariesSubRoute union split Local vs Remote.
- App.tsx: libraries route allowSubPaths; LibrariesRouteScreen wrapper dispatching a
  flat switch to placeholders; App-owned popstate (finding 4) — App is the single
  popstate owner, consults canLeaveCurrentScreen() and only on approval updates
  librariesSubPath passed DOWN to the wrapper (wrapper never self-listens); state write
  scoped to the libraries route so Playouts/Media pops stay byte-identical (nit 3).
- LibrariesScreen hub wiring: Add-Source menu (Local/Plex/Jellyfin/Emby), remote source
  gear -> family screen, local library row gear -> edit route; removed the disabled
  Scan-All button + the deferred-sources card (§C7/§D.1).
- Tests: App-owned-popstate dirty-guard case (confirm false keeps URL+sub-screen; true
  navigates); mediaSources client URL/verb mapping; pinFlowPoll transitions; familyMeta/
  paths units; hub-wiring navigation.
- docs/spa-conventions.md §8 (resolved sub-path+dirty-guard caveat -> App-owned popstate)
  + §2 exemplar list (guarded-route exception).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 16:11:16 +02:00
timothy c6ca87c06e chore(api): regenerate OpenAPI + TS client for media-source write endpoints (#202 OpenAPI gate) 2026-07-11 15:52:31 +02:00
timothy ee104164db Merge branch 'feat/202-s3-jfemby' into feat/202-media-sources 2026-07-11 15:48:07 +02:00
timothyandClaude Opus 4.8 c617a01e83 feat(api): add Jellyfin/Emby media-source write API (#202 slice S3)
New JellyfinMediaSourcesController (/api/media-sources/jellyfin, J1-J9) and
EmbyMediaSourcesController (/api/media-sources/emby, E1-E9), wrapping the
existing Jellyfin/Emby MediatR commands per the #202 design doc §A.3/§A.4.

Secure connection contract (§C3/§B, finding 1): the connection GET returns
only { address, hasApiKey } — the API key never crosses the wire. The PUT
retains the existing key when the incoming key is blank, sets a new one when
non-blank, and 422s "API key is required" on a blank first connect.

Finding 7 (lock-release discipline): DisconnectJellyfinHandler and
DisconnectEmbyHandler now wrap their work in try/finally so a throw from any
awaited dependency (repo delete, search-index commit, secret store) still
releases the family lock instead of wedging every future disconnect at 409.

Findings 2c/8 (path-replacement cross-source guard): UpdateJellyfinPathReplacementsHandler
and UpdateEmbyPathReplacementsHandler now reject, before any write, an incoming
positive Id that isn't owned by the route's media source, a null item, or a
blank RemotePath/LocalPath — all 422 with no partial mutation. Defense-in-depth
repo fix: the Jellyfin/Emby path-replacement UPDATE SQL in MediaSourceRepository
now scopes by {Jellyfin,Emby}MediaSourceId (was previously unscoped by Id alone,
allowing a PUT to one source to silently overwrite another source's row). The
Plex path-replacement method (~line 397) is untouched — that's slice S2's file.

Library preferences (§C4a): the controller validates the incoming id set
against the source's known libraries (reject foreign ids, require full
coverage, no Id=0) before dispatch, then — for §C7 — LockLibrary + enqueues
the SynchronizeXLibraries/SynchronizeXLibraryByIdIfNeeded pair per enabled
library (compensating unlock if the enqueue throws), and returns the reloaded
list (ids are not stable across a disable).

404s on id-taking endpoints come from a controller pre-check (GetXMediaSourceById
is None), not a handler NotFoundError, since Either.Apply/ToEitherAsync join any
NotFoundError into a flat 422 (finding 9).

Tests: controller route/404/409/422 tests for both families; disconnect
fault-injection tests proving the lock releases even when a dependency throws;
path-replacement handler tests for cross-source-id/blank/null-item rejection
and correct add/update/delete merge; a repository-level test proving the SQL
fix stops a same-family cross-source path-replacement overwrite.

No new commands, no DB migration, no OpenAPI regen (gated until S1-S3 merge
per the design doc's build-slice plan).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 15:47:42 +02:00
timothy 26205b8ccd Merge branch 'feat/202-s1-local' into feat/202-media-sources 2026-07-11 15:43:38 +02:00
timothy fe695166f6 Merge branch 'feat/202-s2-plex' into feat/202-media-sources 2026-07-11 15:43:23 +02:00
timothy f5cf23c952 feat(api): local libraries REST endpoints (#202 slice S1)
Adds LocalLibrariesController (L1-L7: list/get/create/update/delete/move-path/
path-exists) wrapping the existing local-library MediatR commands, mapping to
the shared S0 response DTOs. Per design #202 §A.1/§C5/§C6:

- 404 for L4/L5/L6 comes from a controller pre-check (GetLocalLibraryById is
  None), not the handler -- .Apply/.ToEitherAsync both .Join() a NotFoundError
  into a plain 422, so relying on the handler would be dead code. This is
  check-then-act; a delete racing the pre-check falls through to the handler's
  422, documented in the controller.
- L4/L5 409 via IEntityLocker.IsLibraryLocked(id); L6 resolves the source
  library from the path id (new ILibraryRepository.GetLibraryIdForPath) before
  its own lock check.
- MoveLocalLibraryPathHandler gains same-MediaKind and different-library
  validation (finding 3) -- Blazor only filtered these client-side in the move
  dialog, so an API/MCP client could bypass them.
- CreateLocalLibraryHandler/UpdateLocalLibraryHandler gain a shared
  NewPathsMustExist validation (LocalLibraryHandlerBase) that Directory.Exists-
  checks only new paths (Id < 1); existing rows stay exempt so an unmounted
  share doesn't block a rename. L7 (path-exists) is a controller-local
  IFileSystem check with no command.

Tests: controller route/mediator-arg tests incl. 404-pre-check vs
fall-through-422 and 409-lock cases; handler tests for the move-path
cross-kind/same-library 422s, new-path 422 (missing/mixed), and a lossless
round-trip proving local paths are identified by normalized path string, not
id. Full solution test suite (Scanner/Core/Architecture/Tests/Infrastructure)
green, 0 regressions.

Deviations: none from the S1 slice description. Did not touch
MediaSourceRepository.cs or any Plex/Jellyfin/Emby file (S2/S3 scope). Did not
run the OpenAPI regen scripts (separate gate after S1-S3 merge per design §E).
2026-07-11 15:43:10 +02:00
timothy 4852c36268 feat(api): Plex media-source write API + lock-lifecycle fixes (#202 slice S2)
New PlexMediaSourcesController (/api/media-sources/plex) P1-P8 wrapping
existing MediatR commands: state GET, pin-flow, sign-out, per-server
libraries/path-replacements GET+PUT, and refresh — VMs projected to the
S0 shared DTOs, ApiResults mapping, 404 controller pre-checks, #215-style
409 lock guards, [EndpointGroupName("general")].

Lock-lifecycle hardening (the tricky part):
- TryCompletePlexPinFlowHandler now releases the Plex lock ONLY on its
  non-handoff exits (timeout-throw, poll exception, enqueue exception, the
  dead return-false) via try/catch — NOT an unconditional finally. On
  success the lock is handed off to SynchronizePlexMediaSources (the sole
  releaser after discovery); a finally would double-release and release
  before discovery, re-opening the finding-5 poll race. Fixes the latent
  leak where an abandoned pin flow wedged Plex locked until restart.
- StartPlexPinFlow controller compensates UnlockPlex on the Left branch AND
  any thrown dispatch/enqueue; only the Right/200 path holds the lock.
- SignOutOfPlexHandler wraps its work in try/finally { UnlockPlex() } — a
  terminal handler with no handoff, so unconditional release is correct.
- Post-save library sync enqueues SynchronizePlexLibraryByIdIfNeeded
  (Unlock:false) then SynchronizePlexNetworks (Unlock:true) — one lock, one
  release on the last message, compensating-unlock if the 2nd enqueue throws
  (corrects the Blazor Unlock-ordering bug, finding 6).

Data-integrity hardening:
- UpdatePlexPathReplacementsHandler rejects (422, no mutation) any positive
  Id not owned by the route source, blank RemotePath/LocalPath, and null
  list/items (findings 2c/8).
- MediaSourceRepository Plex path-replacement UPDATE gains
  AND PlexMediaSourceId = @id (Jellyfin/Emby untouched — slice S3).
- ReplaceLibraryPreferences controller validates the id set against the
  source's libraries (rejects unowned + Id=0), returns the reloaded list
  (ids change on disable).

Tests (NUnit/Shouldly/NSubstitute), 34 new, all green: pin-flow lock
released on thrown-cancellation/poll-throw/enqueue-throw AND held on
success (no double-release); sign-out finally-release under a throwing
dependency; cross-source/nonblank/null path-replacement 422s; library-prefs
id-not-owned 422; post-save enqueue exact messages + Unlock flags; full
controller route/404/409/422 coverage.

Refs #202
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 15:42:40 +02:00
timothy 8ee709d815 Merge branch 'feat/202-s4-extract' into feat/202-media-sources 2026-07-11 15:30:11 +02:00
timothy 99fab8e0bb refactor(spa): extract LibrariesScreen from App.tsx (#202 slice S4)
Move the Libraries domain verbatim out of web/src/App.tsx into
web/src/screens/LibrariesScreen.tsx (zero-prop, self-sufficient), mirroring the
ChannelsScreen extraction (#244). Pure structural move: no API, route, CSS, or
visual change. App.tsx retains only the import + the <LibrariesScreen /> dispatch.

- 10 symbols moved (LibrariesLoadingState -> sourceLastScanLabel); App.tsx's
  formatDateTime is inlined into the moved screen so it has no import back into
  App.tsx (behavior-identical), matching the Channels precedent.
- Libraries behavior tests moved to a colocated LibrariesScreen.test.tsx with its
  own scoped fetch mock (renders <LibrariesScreen /> directly); App.test.tsx keeps
  one nav-smoke test for the route.
- Pruned now-dead App.tsx imports (Server, MonitorPlay, Music, FileImage, Folder,
  HardDrive icons; useLibrariesScreenQuery, LibraryScanStatus/MediaSource/
  MediaSourceLibrary types) and the now-dead runPollTick test helper (its doc
  comment named it Libraries-specific).
- Disabled "Add Source"/gear/"Scan All" affordances are unchanged (wired in later
  S5/S6 slices, not here).

Verified: web vitest 584 passed, eslint clean, tsc/vite build clean, check:api no
drift.

refs #202
2026-07-11 15:29:15 +02:00
timothyandClaude Opus 4.8 06c877b5fc docs(handoff): lore — post-commit invalidation on None + SetValues positional reconcile (#251, #252)
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 4m30s
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 5m15s
Build ErsatzTV Image / Build & push image (amd64) (push) Successful in 3m30s
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 15:28:29 +02:00
timothy 87060aa3de Merge pull request 'fix(scheduling): stop silently resetting fill-group state + invalidate deco edits (#251, #252)' (#258) from fix/251-252-scheduling-invalidation into main
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / Build & push image (amd64) (push) Has been cancelled
Build ErsatzTV Image / Build & test (.NET) (push) Has been cancelled
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Has been cancelled
2026-07-11 13:26:12 +00:00
timothy 295004ab12 feat(api): add media-source write API shared DTOs (#202 slice S0)
Response DTOs (ErsatzTV.Core/Api/MediaSources) and request DTOs
(ErsatzTV/Controllers/Api/Requests) per the #202 design doc §B — the
shared shapes that backend slices S1 (local libraries), S2 (Plex), and
S3 (Jellyfin/Emby) will consume. No controllers or handler changes;
DTOs are unused so far.

Notable: RemoteConnectionResponseModel deliberately never carries the
raw API key (secure connection contract); SaveRemoteConnectionRequest's
To{Jellyfin,Emby}Command(existingApiKey) retains the existing key when
the incoming ApiKey is blank/omitted.

No conventions changed; nothing to update in docs/api-conventions.md.
2026-07-11 15:23:14 +02:00
timothyandClaude Opus 4.8 51ac11e5aa fix(review): close post-commit cancellation window + guard default-deco enqueue (Codex Medium/Low)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 4s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 3m55s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 6m58s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Independent Codex review of the fix diff surfaced two real findings the fork pass missed:

- Medium: the #251 affected-playout QUERIES in ReplaceDecoTemplateItemsHandler and
  UpdateDecoHandler still ran on the request `cancellationToken`, so a cancellation
  landing after SaveChanges committed but before those queries executed would throw
  before the CancellationToken.None enqueue — the edit committed but no playout Reset,
  re-opening the stale-content bug in that window. Run the entire post-commit
  invalidation (queries + enqueue) on CancellationToken.None so the side effect can't
  be half-aborted once the data has changed.
- Low: UpdateDefaultDecoHandler enqueued a Reset for request.PlayoutId even when
  ExecuteUpdateAsync matched 0 rows (nonexistent playout), creating a background build
  request for an id that isn't there. Guard the enqueue on rows-updated > 0 so the
  enqueued set equals the affected set. Added a regression test.

Also corrected the ReplaceProgramScheduleItemsHandler comments: the schedule-item
hierarchy is TPT (table-per-type), not TPH — the SetValues reconcile is safe either way
(same-runtime-type guard; no discriminator to corrupt), Codex confirmed.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 15:18:38 +02:00
timothyandClaude Opus 4.8 64473b136d fix(scheduling): stop silently resetting fill-group state + invalidate deco edits (#251, #252)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 5s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 7m56s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 10m22s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
#252: ReplaceProgramScheduleItems deleted and re-inserted every item on every
save (even a no-op PUT-back), and PlayoutScheduleItemFillGroupIndex.ProgramScheduleItemId
is OnDelete(Cascade) — so every schedule save silently wiped persisted
fill-group/shuffle enumerator progression for all playouts using the schedule.
Switch to a positional in-place reconcile: for a same-typed slot, copy scalars via
CurrentValues.SetValues (BuildItem stays the single source of item construction, so
no field is dropped) and rebuild the watermark/graphics join rows, keeping the item
id — and with it the fill-group index. Subtype change / surplus falls back to
delete+insert for that slot only. The request DTO carries no stable item id, so
position is the only key here; true content-aware stable identity is deferred to the
shared concurrency/round-trip contract in #253.

#251: deco / deco-template CONTENT edits (and default-deco assignment) only take
effect on a playout Reset build — deco/break/default-filler content is applied during
Reset, a Continue keeps the frozen filler items, and BlockKey change-detection has no
deco dimension to self-heal. The editors enqueued nothing (a commented-out TODO in
ReplaceDecoTemplateItemsHandler), so filler/break content stayed stale indefinitely
until a manual Reset. Enqueue BuildPlayout(Reset) for exactly the affected playouts:
- ReplaceDecoTemplateItemsHandler: playouts via PlayoutTemplate.DecoTemplateId
- UpdateDecoHandler: playouts via Playout.DecoId and via deco-template items
- UpdateDefaultDecoHandler: the reassigned playout (adjacent same-class fix)
Post-commit enqueues use CancellationToken.None (audit #22 policy).

Tests: DecoInvalidationTests + ReplaceProgramScheduleItemsReconcileTests, each proven
non-vacuous against a negative control (inverted the primitive, verified 0 CS errors so
the --no-build run used a fresh dll). Full ErsatzTV.Tests suite green (1067).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 15:08:58 +02:00
timothyandClaude Opus 4.8 d93bca653a docs(handoff): lore — negative-control a flag fix by inverting, not if(true) (CS0219 stale-dll trap)
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 4m57s
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 6m21s
Build ErsatzTV Image / Build & push image (amd64) (push) Successful in 6m41s
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 14:15:06 +02:00
timothy 77272ed5a7 Merge pull request 'fix(locking): release troubleshooting + playout locks on all terminal paths (#233, #234)' (#250) from fix/233-234-lock-leaks into main
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / Build & push image (amd64) (push) Has been cancelled
Build ErsatzTV Image / Build & test (.NET) (push) Has been cancelled
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Has been cancelled
2026-07-11 12:12:52 +00:00
timothyandClaude Opus 4.8 8cdc963636 fix(review): centralize troubleshooting lock ownership in Handle (Codex High)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 9s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 8m3s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 10m4s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Independent Codex review of PR #250 caught a cross-release the fork missed: the
outer catch in PrepareTroubleshootingPlaybackHandler released the troubleshooting
lock unconditionally, so an exception BEFORE this caller acquired it (e.g. request
cancellation during validation, or a DB error) would release a lock held by another
session.

Fix: track ownership with a Handle-scoped `lockAcquired` flag and gate the catch on
it. Acquisition for the media-item path moves out of GetProcess up into Handle (after
validation succeeds), so one place owns the full lifecycle: acquire -> on Left release
-> on success hand off to StartTroubleshootingPlayback -> on any exception release only
if we own it. GetProcess is now lock-free.

Test: Handle_Should_Not_Release_Lock_It_Never_Acquired (exception before acquisition ->
no Unlock, no cross-release), verified non-vacuous against an inverted-condition
negative control. Strengthened the empty-path test to also assert acquisition happened.

Deferred (noted for close comment): worker-dispatch-failure lock leak and
BuildPlayout silent-skip observability are pre-existing / sanctioned -> #235.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 14:02:03 +02:00
timothyandClaude Opus 4.8 084d4c4ca1 fix(locking): release troubleshooting + playout locks on all terminal paths (#233, #234)
Build ErsatzTV Image / Docs update reminder (pull_request) Successful in 5s
Build ErsatzTV Image / Build & test (.NET) (pull_request) Successful in 5m55s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (pull_request) Successful in 8m24s
Build ErsatzTV Image / Build & push image (amd64) (pull_request) Has been skipped
Consume the EntityLocker ownership contract (#231/#241: Lock* returns true iff
this caller won the slot) at three lock-leak sites surfaced by adversarial-reviewer#20.

#233 (F3) — troubleshooting playback:
- PrepareTroubleshootingPlaybackHandler: both lock sites now acquire via
  `if (!LockTroubleshootingPlayback())` (kills the check-then-set TOCTOU) and the
  empty-media-path Left return releases the lock it acquired — previously it leaked,
  wedging the status endpoint at "running" forever for a file gone from disk.
- TroubleshootController.TroubleshootPlayback: lock conflict is now 409 ProblemDetails
  (was a bare 404, indistinguishable from a bad id); the Prepare-success -> enqueue
  window releases the lock if we never hand off to StartTroubleshootingPlayback.

#234 (F4 + F5.2) — playout builds:
- ExtractEmbeddedSubtitlesHandler: try/finally releases exactly the playouts it
  locked, on every terminal path (cancellation early-return, swallowed cancellation,
  any exception) — no more permanent leaks after cancelled mid-extraction, and no
  cross-release of playouts held by someone else.
- BuildPlayoutHandler: skips (logs, returns Right) when LockPlayout returns false
  instead of building unlocked and cross-releasing the other owner's lock in finally.

Tests: handler-level release-discipline tests (Prepare empty-path, Extract
cancellation + no-cross-release, BuildPlayout skip + finally-release) via the
InMemoryTvContext harness, a TroubleshootController 409 test, and OpenApi contract
cases for the m3u8 endpoint's 409. OpenAPI regenerated. All non-vacuous (F3 verified
against a negative control).

Fixes #233, #234

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-07-11 13:44:47 +02:00
timothy 533c6c1e9f Merge pull request 'refactor(spa): extract ChannelsScreen from App.tsx (#244, epic #243 phase 1)' (#249) from feat/244-channels-extract into main
Build ErsatzTV Image / Docs update reminder (push) Has been skipped
Build ErsatzTV Image / Build & test (.NET) (push) Successful in 7m55s
Build ErsatzTV Image / EF migration integrity (SQLite + MySql) (push) Successful in 10m1s
Build ErsatzTV Image / Build & push image (amd64) (push) Successful in 5m30s
2026-07-11 11:38:47 +00:00
302 changed files with 36865 additions and 1483 deletions
+10
View File
@@ -0,0 +1,10 @@
# Enforce the CLAUDE.md protocol: every commit message must carry a Co-Authored-By
# trailer. Merge commits are exempt (their MERGE_MSG has no trailer and shouldn't be
# rewritten).
if git rev-parse -q --verify MERGE_HEAD >/dev/null 2>&1; then
exit 0
fi
grep -q '^Co-Authored-By:' "$1" || {
echo 'husky - commit message missing Co-Authored-By trailer'
exit 1
}
+15
View File
@@ -0,0 +1,15 @@
cd web && npx lint-staged || exit 1
cd ..
# dotnet format on staged .cs files (repo root). Scoped to the staged files so we
# don't pay the full-tree cost; skip entirely when no .cs is staged (avoids the
# ~20-40s sln load for web-only commits).
cs_files=$(git diff --cached --name-only --diff-filter=ACM -- '*.cs')
if [ -n "$cs_files" ]; then
echo "husky - dotnet format (verify) on staged .cs files"
# shellcheck disable=SC2086
dotnet format ErsatzTV.sln --verify-no-changes --include $cs_files || {
echo "husky - dotnet format found issues in staged .cs files; run 'dotnet format ErsatzTV.sln --include <files>' to fix"
exit 1
}
fi
+11
View File
@@ -0,0 +1,11 @@
# Git exports GIT_DIR/GIT_WORK_TREE/GIT_INDEX_FILE while running hooks. In a worktree
# (or any subdir), an explicit GIT_DIR makes nested `git` commands mislocate the working
# tree — notably `check:api`'s `git diff --exit-code` (run from web/) silently reports "no
# diff" and lets drift through. Unset them so nested git rediscovers the repo normally.
unset GIT_DIR GIT_WORK_TREE GIT_INDEX_FILE
# CI-parity checks: catch "green locally, red in CI" before the push leaves the machine.
# check:api guards the generated OpenAPI types (v1.json / v1.d.ts drift); the full
# lint/typecheck/build catch a staged change that breaks an UNstaged file (lint-staged
# only sees staged files).
cd web && npm run check:api && npm run lint && npm run typecheck && npm run build
@@ -55,7 +55,9 @@ public class BulkDeleteChannelsHandler(
}
}
await workerChannel.WriteAsync(new RefreshChannelList(), cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await workerChannel.WriteAsync(new RefreshChannelList(), CancellationToken.None);
return Right<BaseError, Unit>(Unit.Default);
}
@@ -54,7 +54,10 @@ public class BulkMoveChannelsToGroupHandler(
await transaction.CommitAsync(cancellationToken);
searchTargets.SearchTargetsChanged();
await workerChannel.WriteAsync(new RefreshChannelList(), cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await workerChannel.WriteAsync(new RefreshChannelList(), CancellationToken.None);
return Right<BaseError, Unit>(Unit.Default);
}
@@ -68,19 +68,22 @@ public class CreateChannelFromLineupHandler(
}
searchTargets.SearchTargetsChanged();
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await workerChannel.WriteAsync(
new BuildPlayout(prepared.Playout.Id, PlayoutBuildMode.Reset),
cancellationToken);
CancellationToken.None);
// Mirror CreateClassicPlayoutHandler: on-demand playouts must be time-shifted to "now" after build.
if (prepared.Channel.PlayoutMode is ChannelPlayoutMode.OnDemand)
{
await workerChannel.WriteAsync(
new TimeShiftOnDemandPlayout(prepared.Playout.Id, DateTimeOffset.Now, false),
cancellationToken);
CancellationToken.None);
}
await workerChannel.WriteAsync(new RefreshChannelList(), cancellationToken);
await workerChannel.WriteAsync(new RefreshChannelList(), CancellationToken.None);
return new CreateChannelFromLineupResponseModel(
prepared.Channel.Id,
@@ -47,20 +47,24 @@ public class DeleteChannelHandler : IRequestHandler<DeleteChannel, Either<BaseEr
private async Task<Unit> DoDeletion(TvContext dbContext, Channel channel, CancellationToken cancellationToken)
{
// Delete the guide cache file through the filesystem abstraction (so it's observable under a
// MockFileSystem) and BEFORE the commit: deleting after commit orphans {number}.xml if the
// process crashes in between (nothing reaps it, and GetChannelGuideHandler serves everything
// in the cache folder). The guide xml is regenerable on demand, so losing it pre-commit is safe (#254).
string cacheFile = Path.Combine(FileSystemLayout.ChannelGuideCacheFolder, $"{channel.Number}.xml");
if (_fileSystem.File.Exists(cacheFile))
{
_fileSystem.File.Delete(cacheFile);
}
dbContext.Channels.Remove(channel);
await dbContext.SaveChangesAsync(cancellationToken);
_searchTargets.SearchTargetsChanged();
// delete channel data from channel guide cache
string cacheFile = Path.Combine(FileSystemLayout.ChannelGuideCacheFolder, $"{channel.Number}.xml");
if (_fileSystem.File.Exists(cacheFile))
{
File.Delete(cacheFile);
}
// refresh channel list to remove channel that has no playout
await _workerChannel.WriteAsync(new RefreshChannelList(), cancellationToken);
// refresh channel list to remove channel that has no playout — post-commit side effect runs on
// CancellationToken.None so a late request cancellation can't abort it after the delete committed (#254)
await _workerChannel.WriteAsync(new RefreshChannelList(), CancellationToken.None);
return Unit.Default;
}
@@ -157,21 +157,23 @@ public class UpdateChannelHandler(
searchTargets.SearchTargetsChanged();
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
if (c.SubtitleMode != ChannelSubtitleMode.None)
{
Option<Playout> maybePlayout = await dbContext.Playouts
.SelectOneAsync(p => p.ChannelId, p => p.ChannelId == c.Id, cancellationToken);
.SelectOneAsync(p => p.ChannelId, p => p.ChannelId == c.Id, CancellationToken.None);
foreach (Playout playout in maybePlayout)
{
await workerChannel.WriteAsync(new ExtractEmbeddedSubtitles(playout.Id), cancellationToken);
await workerChannel.WriteAsync(new ExtractEmbeddedSubtitles(playout.Id), CancellationToken.None);
}
}
await workerChannel.WriteAsync(new RefreshChannelList(), cancellationToken);
await workerChannel.WriteAsync(new RefreshChannelList(), CancellationToken.None);
if (hasEpgChange)
{
await workerChannel.WriteAsync(new RefreshChannelData(c.Number), cancellationToken);
await workerChannel.WriteAsync(new RefreshChannelData(c.Number), CancellationToken.None);
}
return ProjectToViewModel(c, c.Playouts?.Count ?? 0);
@@ -81,10 +81,12 @@ public class UpdateChannelNumbersHandler(
await transaction.CommitAsync(cancellationToken);
// update channel list and xmltv
await workerChannel.WriteAsync(new RefreshChannelList(), cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await workerChannel.WriteAsync(new RefreshChannelList(), CancellationToken.None);
foreach (var channel in channelsToUpdate)
{
await workerChannel.WriteAsync(new RefreshChannelData(channel.Number), cancellationToken);
await workerChannel.WriteAsync(new RefreshChannelData(channel.Number), CancellationToken.None);
}
return Option<BaseError>.None;
@@ -0,0 +1,89 @@
using ErsatzTV.Core;
using ErsatzTV.Core.Domain;
using ErsatzTV.Core.Errors;
using Microsoft.EntityFrameworkCore;
using Microsoft.EntityFrameworkCore.ChangeTracking;
namespace ErsatzTV.Application;
public static class ConcurrencyExtensions
{
/// <summary>
/// Persist changes that touch a versioned root but do <b>not</b> participate in the If-Match
/// contract (e.g. a playout's settings/schedule-file/on-demand-checkpoint writer, a collection's
/// name edit). Because the root's <c>Version</c> is an <c>IsConcurrencyToken</c>, EF guards every
/// UPDATE of that row with <c>WHERE Version=@orig</c>, so a concurrent bump from a replace-all
/// editor would otherwise surface as an unhandled <see cref="DbUpdateConcurrencyException" /> →
/// 500 (issue #253 / #269). Phase-1 semantics for a missing <c>If-Match</c> is <b>force-write</b>,
/// so on a concurrency failure we adopt the stored token as both original (the retry's WHERE then
/// matches) and current (so we don't revert the concurrent bump) and retry — a client-wins merge
/// scoped to the token; our own modified scalars still win. Bounded to avoid a livelock; if the row
/// was deleted out from under us, that's a genuine conflict and rethrows.
/// </summary>
public static async Task<int> SaveChangesForcingVersion(
this DbContext dbContext,
CancellationToken cancellationToken)
{
for (var attempt = 0; ; attempt++)
{
try
{
return await dbContext.SaveChangesAsync(cancellationToken);
}
catch (DbUpdateConcurrencyException ex) when (attempt < 5)
{
var resolvedAny = false;
foreach (EntityEntry entry in ex.Entries)
{
if (entry.Entity is not IVersionedAggregate)
{
continue;
}
PropertyValues databaseValues = await entry.GetDatabaseValuesAsync(cancellationToken);
if (databaseValues is null)
{
// The row was deleted out from under us — a genuine conflict, not a token race.
throw;
}
PropertyEntry version = entry.Property(nameof(IVersionedAggregate.Version));
object currentVersion = databaseValues[nameof(IVersionedAggregate.Version)]!;
version.OriginalValue = currentVersion;
version.CurrentValue = currentVersion;
resolvedAny = true;
}
if (!resolvedAny)
{
throw;
}
}
}
}
/// <summary>
/// Persist pending changes, mapping the EF optimistic-concurrency failure to
/// <see cref="PreconditionFailedError" /> (→ 412). When a versioned root carries an
/// <c>IsConcurrencyToken</c> column and its <c>Version</c> is bumped before saving, EF emits
/// <c>UPDATE … WHERE Id=@id AND Version=@original</c>; a zero-row result (another writer won
/// the race between our load and save) throws <see cref="DbUpdateConcurrencyException" />.
/// This is the backstop that closes the load→save TOCTOU the handler pre-check cannot.
/// Issue #253.
/// </summary>
public static async Task<Either<BaseError, Unit>> SaveChangesWithConcurrencyGuard(
this DbContext dbContext,
CancellationToken cancellationToken)
{
try
{
await dbContext.SaveChangesAsync(cancellationToken);
return Unit.Default;
}
catch (DbUpdateConcurrencyException)
{
return new PreconditionFailedError(
"The resource was modified by another request. Reload and try again.");
}
}
}
@@ -30,12 +30,21 @@ public class DisconnectEmbyHandler : IRequestHandler<DisconnectEmby, Either<Base
DisconnectEmby request,
CancellationToken cancellationToken)
{
List<int> ids = await _mediaSourceRepository.DeleteAllEmby();
await _searchIndex.RemoveItems(ids);
_searchIndex.Commit();
await _embySecretStore.DeleteAll();
_entityLocker.UnlockRemoteMediaSource<EmbyMediaSource>();
// This is a terminal handler with no lock handoff (unlike the Plex pin-flow handlers) —
// release unconditionally so a throw from any awaited dependency (repo delete, search-index
// commit, secret store) can't wedge the Emby lock until restart (design #202 finding 7).
try
{
List<int> ids = await _mediaSourceRepository.DeleteAllEmby();
await _searchIndex.RemoveItems(ids);
_searchIndex.Commit();
await _embySecretStore.DeleteAll();
return Unit.Default;
return Unit.Default;
}
finally
{
_entityLocker.UnlockRemoteMediaSource<EmbyMediaSource>();
}
}
}
@@ -2,6 +2,6 @@ using ErsatzTV.Core;
namespace ErsatzTV.Application.Emby;
public record SynchronizeEmbyCollections(int EmbyMediaSourceId, bool ForceScan, bool DeepScan)
public record SynchronizeEmbyCollections(int EmbyMediaSourceId, bool ForceScan, bool DeepScan, bool Unlock = true)
: IRequest<Either<BaseError, Unit>>,
IScannerBackgroundServiceRequest;
@@ -1,4 +1,4 @@
using ErsatzTV.Core;
using ErsatzTV.Core;
using ErsatzTV.Core.Domain;
using ErsatzTV.Core.Interfaces.Repositories;
@@ -12,12 +12,29 @@ public class UpdateEmbyPathReplacementsHandler : IRequestHandler<UpdateEmbyPathR
public UpdateEmbyPathReplacementsHandler(IMediaSourceRepository mediaSourceRepository) =>
_mediaSourceRepository = mediaSourceRepository;
public Task<Either<BaseError, Unit>> Handle(
public async Task<Either<BaseError, Unit>> Handle(
UpdateEmbyPathReplacements request,
CancellationToken cancellationToken) =>
Validate(request, cancellationToken)
.MapT(pms => MergePathReplacements(request, pms))
.Bind(v => v.ToEitherAsync());
CancellationToken cancellationToken)
{
Option<EmbyMediaSource> maybeSource =
await _mediaSourceRepository.GetEmby(request.EmbyMediaSourceId, cancellationToken);
return await maybeSource.Match(
Some: async embyMediaSource =>
{
Option<BaseError> maybeError = ValidateItems(request, embyMediaSource);
return await maybeError.Match(
Some: error => Task.FromResult(Left<BaseError, Unit>(error)),
None: async () =>
{
await MergePathReplacements(request, embyMediaSource);
return Right<BaseError, Unit>(Unit.Default);
});
},
None: () => Task.FromResult(
Left<BaseError, Unit>(
BaseError.New($"Emby media source {request.EmbyMediaSourceId} does not exist."))));
}
private Task<Unit> MergePathReplacements(
UpdateEmbyPathReplacements request,
@@ -37,12 +54,38 @@ public class UpdateEmbyPathReplacementsHandler : IRequestHandler<UpdateEmbyPathR
private static EmbyPathReplacement Project(EmbyPathReplacementItem vm) =>
new() { Id = vm.Id, EmbyPath = vm.EmbyPath, LocalPath = vm.LocalPath };
private Task<Validation<BaseError, EmbyMediaSource>> Validate(UpdateEmbyPathReplacements request, CancellationToken cancellationToken) =>
EmbyMediaSourceMustExist(request, cancellationToken);
// Defense-in-depth for design #202 findings 2c/8 — the repo UPDATE is scoped by
// EmbyMediaSourceId, but reject a foreign/blank/null row here too, before any write, so the
// mutation is all-or-nothing.
private static Option<BaseError> ValidateItems(
UpdateEmbyPathReplacements request,
EmbyMediaSource embyMediaSource)
{
List<EmbyPathReplacementItem> items = request.PathReplacements ?? [];
private Task<Validation<BaseError, EmbyMediaSource>> EmbyMediaSourceMustExist(
UpdateEmbyPathReplacements request, CancellationToken cancellationToken) =>
_mediaSourceRepository.GetEmby(request.EmbyMediaSourceId, cancellationToken)
.Map(v => v.ToValidation<BaseError>(
$"Emby media source {request.EmbyMediaSourceId} does not exist."));
if (items.Any(item => item is null))
{
return BaseError.New("Path replacement items must not be null.");
}
if (items.Any(item => string.IsNullOrWhiteSpace(item.EmbyPath) || string.IsNullOrWhiteSpace(item.LocalPath)))
{
return BaseError.New("Each path replacement requires a non-blank Emby path and local path.");
}
var existingIds = (embyMediaSource.PathReplacements ?? new List<EmbyPathReplacement>())
.Map(pr => pr.Id)
.ToList();
var foreignIds = items.Filter(item => item.Id > 0 && !existingIds.Contains(item.Id))
.Map(item => item.Id)
.ToList();
if (foreignIds.Count > 0)
{
return BaseError.New(
$"Path replacement id(s) {string.Join(", ", foreignIds)} do not belong to Emby media source " +
$"{request.EmbyMediaSourceId}.");
}
return Option<BaseError>.None;
}
}
@@ -30,12 +30,21 @@ public class DisconnectJellyfinHandler : IRequestHandler<DisconnectJellyfin, Eit
DisconnectJellyfin request,
CancellationToken cancellationToken)
{
List<int> ids = await _mediaSourceRepository.DeleteAllJellyfin();
await _searchIndex.RemoveItems(ids);
_searchIndex.Commit();
await _jellyfinSecretStore.DeleteAll();
_entityLocker.UnlockRemoteMediaSource<JellyfinMediaSource>();
// This is a terminal handler with no lock handoff (unlike the Plex pin-flow handlers) —
// release unconditionally so a throw from any awaited dependency (repo delete, search-index
// commit, secret store) can't wedge the Jellyfin lock until restart (design #202 finding 7).
try
{
List<int> ids = await _mediaSourceRepository.DeleteAllJellyfin();
await _searchIndex.RemoveItems(ids);
_searchIndex.Commit();
await _jellyfinSecretStore.DeleteAll();
return Unit.Default;
return Unit.Default;
}
finally
{
_entityLocker.UnlockRemoteMediaSource<JellyfinMediaSource>();
}
}
}
@@ -2,6 +2,6 @@ using ErsatzTV.Core;
namespace ErsatzTV.Application.Jellyfin;
public record SynchronizeJellyfinCollections(int JellyfinMediaSourceId, bool ForceScan, bool DeepScan) :
public record SynchronizeJellyfinCollections(int JellyfinMediaSourceId, bool ForceScan, bool DeepScan, bool Unlock = true) :
IRequest<Either<BaseError, Unit>>,
IScannerBackgroundServiceRequest;
@@ -1,4 +1,4 @@
using ErsatzTV.Core;
using ErsatzTV.Core;
using ErsatzTV.Core.Domain;
using ErsatzTV.Core.Interfaces.Repositories;
@@ -12,12 +12,29 @@ public class UpdateJellyfinPathReplacementsHandler : IRequestHandler<UpdateJelly
public UpdateJellyfinPathReplacementsHandler(IMediaSourceRepository mediaSourceRepository) =>
_mediaSourceRepository = mediaSourceRepository;
public Task<Either<BaseError, Unit>> Handle(
public async Task<Either<BaseError, Unit>> Handle(
UpdateJellyfinPathReplacements request,
CancellationToken cancellationToken) =>
Validate(request)
.MapT(pms => MergePathReplacements(request, pms))
.Bind(v => v.ToEitherAsync());
CancellationToken cancellationToken)
{
Option<JellyfinMediaSource> maybeSource =
await _mediaSourceRepository.GetJellyfin(request.JellyfinMediaSourceId);
return await maybeSource.Match(
Some: async jellyfinMediaSource =>
{
Option<BaseError> maybeError = ValidateItems(request, jellyfinMediaSource);
return await maybeError.Match(
Some: error => Task.FromResult(Left<BaseError, Unit>(error)),
None: async () =>
{
await MergePathReplacements(request, jellyfinMediaSource);
return Right<BaseError, Unit>(Unit.Default);
});
},
None: () => Task.FromResult(
Left<BaseError, Unit>(
BaseError.New($"Jellyfin media source {request.JellyfinMediaSourceId} does not exist."))));
}
private Task<Unit> MergePathReplacements(
UpdateJellyfinPathReplacements request,
@@ -37,12 +54,38 @@ public class UpdateJellyfinPathReplacementsHandler : IRequestHandler<UpdateJelly
private static JellyfinPathReplacement Project(JellyfinPathReplacementItem vm) =>
new() { Id = vm.Id, JellyfinPath = vm.JellyfinPath, LocalPath = vm.LocalPath };
private Task<Validation<BaseError, JellyfinMediaSource>> Validate(UpdateJellyfinPathReplacements request) =>
JellyfinMediaSourceMustExist(request);
// Defense-in-depth for design #202 findings 2c/8 — the repo UPDATE is scoped by
// JellyfinMediaSourceId, but reject a foreign/blank/null row here too, before any write, so the
// mutation is all-or-nothing.
private static Option<BaseError> ValidateItems(
UpdateJellyfinPathReplacements request,
JellyfinMediaSource jellyfinMediaSource)
{
List<JellyfinPathReplacementItem> items = request.PathReplacements ?? [];
private Task<Validation<BaseError, JellyfinMediaSource>> JellyfinMediaSourceMustExist(
UpdateJellyfinPathReplacements request) =>
_mediaSourceRepository.GetJellyfin(request.JellyfinMediaSourceId)
.Map(v => v.ToValidation<BaseError>(
$"Jellyfin media source {request.JellyfinMediaSourceId} does not exist."));
if (items.Any(item => item is null))
{
return BaseError.New("Path replacement items must not be null.");
}
if (items.Any(item => string.IsNullOrWhiteSpace(item.JellyfinPath) || string.IsNullOrWhiteSpace(item.LocalPath)))
{
return BaseError.New("Each path replacement requires a non-blank Jellyfin path and local path.");
}
var existingIds = (jellyfinMediaSource.PathReplacements ?? new List<JellyfinPathReplacement>())
.Map(pr => pr.Id)
.ToList();
var foreignIds = items.Filter(item => item.Id > 0 && !existingIds.Contains(item.Id))
.Map(item => item.Id)
.ToList();
if (foreignIds.Count > 0)
{
return BaseError.New(
$"Path replacement id(s) {string.Join(", ", foreignIds)} do not belong to Jellyfin media source " +
$"{request.JellyfinMediaSourceId}.");
}
return Option<BaseError>.None;
}
}
@@ -1,4 +1,5 @@
using System.Threading.Channels;
using System.IO.Abstractions;
using System.Threading.Channels;
using ErsatzTV.Application.MediaSources;
using ErsatzTV.Core;
using ErsatzTV.Core.Domain;
@@ -14,15 +15,18 @@ public class CreateLocalLibraryHandler : LocalLibraryHandlerBase,
{
private readonly IDbContextFactory<TvContext> _dbContextFactory;
private readonly IEntityLocker _entityLocker;
private readonly IFileSystem _fileSystem;
private readonly ChannelWriter<IScannerBackgroundServiceRequest> _scannerWorkerChannel;
public CreateLocalLibraryHandler(
ChannelWriter<IScannerBackgroundServiceRequest> scannerWorkerChannel,
IEntityLocker entityLocker,
IFileSystem fileSystem,
IDbContextFactory<TvContext> dbContextFactory)
{
_scannerWorkerChannel = scannerWorkerChannel;
_entityLocker = entityLocker;
_fileSystem = fileSystem;
_dbContextFactory = dbContextFactory;
}
@@ -31,7 +35,7 @@ public class CreateLocalLibraryHandler : LocalLibraryHandlerBase,
CancellationToken cancellationToken)
{
await using TvContext dbContext = await _dbContextFactory.CreateDbContextAsync(cancellationToken);
Validation<BaseError, LocalLibrary> validation = await Validate(dbContext, request);
Validation<BaseError, LocalLibrary> validation = await Validate(_fileSystem, dbContext, request);
return await validation.Apply(localLibrary => PersistLocalLibrary(dbContext, localLibrary));
}
@@ -61,11 +65,13 @@ public class CreateLocalLibraryHandler : LocalLibraryHandlerBase,
}
private static Task<Validation<BaseError, LocalLibrary>> Validate(
IFileSystem fileSystem,
TvContext dbContext,
CreateLocalLibrary request) =>
MediaSourceMustExist(dbContext, request)
.BindT(localLibrary => NameMustBeValid(request, localLibrary))
.BindT(localLibrary => PathsMustBeValid(dbContext, localLibrary));
.BindT(localLibrary => PathsMustBeValid(dbContext, localLibrary))
.BindT(localLibrary => NewPathsMustExist(fileSystem, localLibrary));
private static Task<Validation<BaseError, LocalLibrary>> MediaSourceMustExist(
TvContext dbContext,
@@ -1,4 +1,5 @@
using ErsatzTV.Core;
using System.IO.Abstractions;
using ErsatzTV.Core;
using ErsatzTV.Core.Domain;
using ErsatzTV.Infrastructure.Data;
using Microsoft.EntityFrameworkCore;
@@ -14,6 +15,28 @@ public abstract class LocalLibraryHandlerBase
.Bind(_ => request.NotLongerThan(50)(c => c.Name))
.Map(_ => localLibrary).AsTask();
/// <summary>
/// Validates that every NEW path (<c>Id &lt; 1</c> — see design #202 §C2) exists on the
/// filesystem. Existing rows are exempt: an unmounted share must not block saving a rename,
/// matching Blazor's behavior of only checking existence when a path is added.
/// </summary>
protected static Task<Validation<BaseError, LocalLibrary>> NewPathsMustExist(
IFileSystem fileSystem,
LocalLibrary localLibrary)
{
List<string> missing = localLibrary.Paths
.Filter(p => p.Id < 1)
.Filter(p => !fileSystem.Directory.Exists(p.Path))
.Map(p => p.Path)
.ToList();
Validation<BaseError, LocalLibrary> result = missing.Count == 0
? Success<BaseError, LocalLibrary>(localLibrary)
: Fail<BaseError, LocalLibrary>($"Path(s) do not exist on the filesystem: {string.Join(", ", missing)}");
return result.AsTask();
}
protected static async Task<Validation<BaseError, LocalLibrary>> PathsMustBeValid(
TvContext dbContext,
LocalLibrary localLibrary,
@@ -79,10 +79,31 @@ public class MoveLocalLibraryPathHandler : IRequestHandler<MoveLocalLibraryPath,
private static async Task<Validation<BaseError, Parameters>> Validate(
TvContext dbContext,
MoveLocalLibraryPath request,
CancellationToken cancellationToken) =>
(await LibraryPathMustExist(dbContext, request, cancellationToken),
await LocalLibraryMustExist(dbContext, request, cancellationToken))
.Apply((libraryPath, localLibrary) => new Parameters(libraryPath, localLibrary));
CancellationToken cancellationToken)
{
Validation<BaseError, Parameters> parameters =
(await LibraryPathMustExist(dbContext, request, cancellationToken),
await LocalLibraryMustExist(dbContext, request, cancellationToken))
.Apply((libraryPath, localLibrary) => new Parameters(libraryPath, localLibrary));
return parameters
.Bind(TargetLibraryMustDiffer)
.Bind(TargetLibraryMustMatchMediaKind);
}
// Blazor's move dialog filters the target-library picker to same-kind, source-excluded
// libraries only (MoveLocalLibraryPathDialog.razor:82); an API/MCP client bypasses that
// client-side filter today, so #202 moves both invariants into the handler (design #202 §C5,
// finding 3).
private static Validation<BaseError, Parameters> TargetLibraryMustDiffer(Parameters parameters) =>
parameters.LibraryPath.LibraryId == parameters.Library.Id
? Fail<BaseError, Parameters>("Target library must be different from the source path's current library")
: Success<BaseError, Parameters>(parameters);
private static Validation<BaseError, Parameters> TargetLibraryMustMatchMediaKind(Parameters parameters) =>
parameters.LibraryPath.Library.MediaKind != parameters.Library.MediaKind
? Fail<BaseError, Parameters>("Target library must have the same media kind as the source path's library")
: Success<BaseError, Parameters>(parameters);
private static Task<Validation<BaseError, LibraryPath>> LibraryPathMustExist(
TvContext dbContext,
@@ -8,4 +8,4 @@ public enum QueueLibraryScanResult
AlreadyScanning
}
public record QueueLibraryScanByLibraryId(int LibraryId) : IRequest<QueueLibraryScanResult>;
public record QueueLibraryScanByLibraryId(int LibraryId, bool DeepScan = false) : IRequest<QueueLibraryScanResult>;
@@ -66,7 +66,7 @@ public class QueueLibraryScanByLibraryIdHandler(
new SynchronizePlexLibraries(library.MediaSourceId),
cancellationToken);
await scannerWorker.WriteAsync(
new ForceSynchronizePlexLibraryById(library.Id, false),
new ForceSynchronizePlexLibraryById(library.Id, request.DeepScan),
cancellationToken);
break;
case JellyfinLibrary:
@@ -74,7 +74,7 @@ public class QueueLibraryScanByLibraryIdHandler(
new SynchronizeJellyfinLibraries(library.MediaSourceId),
cancellationToken);
await scannerWorker.WriteAsync(
new ForceSynchronizeJellyfinLibraryById(library.Id, false),
new ForceSynchronizeJellyfinLibraryById(library.Id, request.DeepScan),
cancellationToken);
break;
case EmbyLibrary:
@@ -82,7 +82,7 @@ public class QueueLibraryScanByLibraryIdHandler(
new SynchronizeEmbyLibraries(library.MediaSourceId),
cancellationToken);
await scannerWorker.WriteAsync(
new ForceSynchronizeEmbyLibraryById(library.Id, false),
new ForceSynchronizeEmbyLibraryById(library.Id, request.DeepScan),
cancellationToken);
break;
}
@@ -1,3 +1,14 @@
namespace ErsatzTV.Application.Libraries;
public record QueueShowScanByLibraryId(int LibraryId, int ShowId, string ShowTitle, bool DeepScan) : IRequest<bool>;
public enum QueueShowScanResult
{
Queued,
NotFound,
Unsupported,
SyncDisabled,
AlreadyScanning,
ScanFailed
}
public record QueueShowScanByLibraryId(int LibraryId, int ShowId, string ShowTitle, bool DeepScan)
: IRequest<QueueShowScanResult>;
@@ -19,9 +19,9 @@ public class QueueShowScanByLibraryIdHandler(
IMediator mediator,
ChannelWriter<IBackgroundServiceRequest> workerChannel,
ILogger<QueueShowScanByLibraryIdHandler> logger)
: IRequestHandler<QueueShowScanByLibraryId, bool>
: IRequestHandler<QueueShowScanByLibraryId, QueueShowScanResult>
{
public async Task<bool> Handle(QueueShowScanByLibraryId request, CancellationToken cancellationToken)
public async Task<QueueShowScanResult> Handle(QueueShowScanByLibraryId request, CancellationToken cancellationToken)
{
await using TvContext dbContext = await dbContextFactory.CreateDbContextAsync(cancellationToken);
@@ -42,14 +42,14 @@ public class QueueShowScanByLibraryIdHandler(
if (!shouldSyncItems)
{
logger.LogWarning("Library sync is disabled for library id {Id}", library.Id);
return false;
return QueueShowScanResult.SyncDisabled;
}
// Check if library is already being scanned - return false if locked
// A false from LockLibrary means a scan is already in progress; we own no release.
if (!locker.LockLibrary(library.Id))
{
logger.LogWarning("Library {Id} is already being scanned, cannot scan individual show", library.Id);
return false;
return QueueShowScanResult.AlreadyScanning;
}
logger.LogDebug(
@@ -60,41 +60,43 @@ public class QueueShowScanByLibraryIdHandler(
try
{
var success = false;
QueueShowScanResult outcome;
switch (library)
{
case PlexLibrary:
Either<BaseError, string> plexResult = await mediator.Send(
new SynchronizePlexShowById(library.Id, request.ShowId, request.DeepScan),
cancellationToken);
success = plexResult.IsRight;
outcome = plexResult.IsRight ? QueueShowScanResult.Queued : QueueShowScanResult.ScanFailed;
break;
case JellyfinLibrary:
Either<BaseError, string> jellyfinResult = await mediator.Send(
new SynchronizeJellyfinShowById(library.Id, request.ShowId, request.DeepScan),
cancellationToken);
success = jellyfinResult.IsRight;
outcome = jellyfinResult.IsRight ? QueueShowScanResult.Queued : QueueShowScanResult.ScanFailed;
break;
case EmbyLibrary:
Either<BaseError, string> embyResult = await mediator.Send(
new SynchronizeEmbyShowById(library.Id, request.ShowId, request.DeepScan),
cancellationToken);
success = embyResult.IsRight;
outcome = embyResult.IsRight ? QueueShowScanResult.Queued : QueueShowScanResult.ScanFailed;
break;
case LocalLibrary:
logger.LogWarning("Single show scanning is not supported for local libraries");
outcome = QueueShowScanResult.Unsupported;
break;
default:
logger.LogWarning("Unknown library type for library {Id}", library.Id);
outcome = QueueShowScanResult.Unsupported;
break;
}
if (success && request.DeepScan)
if (outcome == QueueShowScanResult.Queued && request.DeepScan)
{
await workerChannel.WriteAsync(new ExtractEmbeddedShowSubtitles(request.ShowId), cancellationToken);
}
return success;
return outcome;
}
finally
{
@@ -103,6 +105,6 @@ public class QueueShowScanByLibraryIdHandler(
}
}
return false;
return QueueShowScanResult.NotFound;
}
}
@@ -1,4 +1,5 @@
using System.Threading.Channels;
using System.IO.Abstractions;
using System.Threading.Channels;
using Dapper;
using ErsatzTV.Application.MediaSources;
using ErsatzTV.Core;
@@ -17,17 +18,20 @@ public class UpdateLocalLibraryHandler : LocalLibraryHandlerBase,
{
private readonly IDbContextFactory<TvContext> _dbContextFactory;
private readonly IEntityLocker _entityLocker;
private readonly IFileSystem _fileSystem;
private readonly ChannelWriter<IScannerBackgroundServiceRequest> _scannerWorkerChannel;
private readonly ISearchIndex _searchIndex;
public UpdateLocalLibraryHandler(
ChannelWriter<IScannerBackgroundServiceRequest> scannerWorkerChannel,
IEntityLocker entityLocker,
IFileSystem fileSystem,
ISearchIndex searchIndex,
IDbContextFactory<TvContext> dbContextFactory)
{
_scannerWorkerChannel = scannerWorkerChannel;
_entityLocker = entityLocker;
_fileSystem = fileSystem;
_searchIndex = searchIndex;
_dbContextFactory = dbContextFactory;
}
@@ -37,7 +41,8 @@ public class UpdateLocalLibraryHandler : LocalLibraryHandlerBase,
CancellationToken cancellationToken)
{
await using TvContext dbContext = await _dbContextFactory.CreateDbContextAsync(cancellationToken);
Validation<BaseError, Parameters> validation = await Validate(dbContext, request, cancellationToken);
Validation<BaseError, Parameters> validation =
await Validate(_fileSystem, dbContext, request, cancellationToken);
return await validation.Apply(parameters => UpdateLocalLibrary(dbContext, parameters));
}
@@ -114,13 +119,15 @@ public class UpdateLocalLibraryHandler : LocalLibraryHandlerBase,
}
private static Task<Validation<BaseError, Parameters>> Validate(
IFileSystem fileSystem,
TvContext dbContext,
UpdateLocalLibrary request,
CancellationToken cancellationToken) =>
LocalLibraryMustExist(dbContext, request, cancellationToken)
.BindT(parameters => NameMustBeValid(request, parameters.Incoming).MapT(_ => parameters))
.BindT(parameters => PathsMustBeValid(dbContext, parameters.Incoming, parameters.Existing.Id)
.MapT(_ => parameters));
.MapT(_ => parameters))
.BindT(parameters => NewPathsMustExist(fileSystem, parameters.Incoming).MapT(_ => parameters));
private static Task<Validation<BaseError, Parameters>> LocalLibraryMustExist(
TvContext dbContext,
@@ -32,6 +32,10 @@ public class AddEpisodeToPlaylistHandler(IDbContextFactory<TvContext> dbContextF
};
parameters.Playlist.Items.Add(playlistItem);
// Mutates the playlist's editor-visible item list, so bump the concurrency token (issue #253).
parameters.Playlist.Version++;
await dbContext.SaveChangesAsync();
return Unit.Default;
}
@@ -81,13 +81,15 @@ public class AddItemsToCollectionHandler :
if (await dbContext.SaveChangesAsync(cancellationToken) > 0)
{
await _searchChannel.WriteAsync(new ReindexMediaItems(toAddIds.ToArray()), cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await _searchChannel.WriteAsync(new ReindexMediaItems(toAddIds.ToArray()), CancellationToken.None);
// refresh all playouts that use this collection
foreach (int playoutId in await _mediaCollectionRepository
.PlayoutIdsUsingCollection(request.CollectionId))
{
await _channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh), cancellationToken);
await _channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh), CancellationToken.None);
}
}
@@ -69,6 +69,9 @@ public class AddItemsToPlaylistHandler : IRequestHandler<AddItemsToPlaylist, Eit
}
}
// Mutates the playlist's editor-visible item list, so bump the concurrency token (issue #253).
playlist.Version++;
await dbContext.SaveChangesAsync(cancellationToken);
return Unit.Default;
@@ -32,6 +32,10 @@ public class AddMovieToPlaylistHandler(IDbContextFactory<TvContext> dbContextFac
};
parameters.Playlist.Items.Add(playlistItem);
// Mutates the playlist's editor-visible item list, so bump the concurrency token (issue #253).
parameters.Playlist.Version++;
await dbContext.SaveChangesAsync();
return Unit.Default;
}
@@ -32,6 +32,10 @@ public class AddSeasonToPlaylistHandler(IDbContextFactory<TvContext> dbContextFa
};
parameters.Playlist.Items.Add(playlistItem);
// Mutates the playlist's editor-visible item list, so bump the concurrency token (issue #253).
parameters.Playlist.Version++;
await dbContext.SaveChangesAsync();
return Unit.Default;
}
@@ -32,6 +32,10 @@ public class AddShowToPlaylistHandler(IDbContextFactory<TvContext> dbContextFact
};
parameters.Playlist.Items.Add(playlistItem);
// Mutates the playlist's editor-visible item list, so bump the concurrency token (issue #253).
parameters.Playlist.Version++;
await dbContext.SaveChangesAsync();
return Unit.Default;
}
@@ -42,7 +42,10 @@ public class CreateSmartCollectionHandler :
await dbContext.SmartCollections.AddAsync(smartCollection, cancellationToken);
await dbContext.SaveChangesAsync(cancellationToken);
_searchTargets.SearchTargetsChanged();
await _smartCollectionCache.Refresh(cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await _smartCollectionCache.Refresh(CancellationToken.None);
return ProjectToViewModel(smartCollection);
}
@@ -48,7 +48,10 @@ public class DeleteSmartCollectionHandler : IRequestHandler<DeleteSmartCollectio
dbContext.SmartCollections.Remove(smartCollection);
await dbContext.SaveChangesAsync(cancellationToken);
_searchTargets.SearchTargetsChanged();
await _smartCollectionCache.Refresh(cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await _smartCollectionCache.Refresh(CancellationToken.None);
return Unit.Default;
}
@@ -63,14 +63,16 @@ public class RemoveItemsFromCollectionHandler : IRequestHandler<RemoveItemsFromC
if (itemsToRemove.Count != 0 && await dbContext.SaveChangesAsync(cancellationToken) > 0)
{
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await _searchChannel.WriteAsync(
new ReindexMediaItems(itemsToRemove.Select(mi => mi.Id).ToArray()),
cancellationToken);
CancellationToken.None);
// refresh all playouts that use this collection
foreach (int playoutId in await _mediaCollectionRepository.PlayoutIdsUsingCollection(collection.Id))
{
await _channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh), cancellationToken);
await _channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh), CancellationToken.None);
}
}
@@ -2,5 +2,9 @@ using ErsatzTV.Core;
namespace ErsatzTV.Application.MediaCollections;
public record ReplacePlaylistItems(int PlaylistId, string Name, List<ReplacePlaylistItem> Items)
public record ReplacePlaylistItems(
int PlaylistId,
string Name,
List<ReplacePlaylistItem> Items,
Option<int> ExpectedVersion = default)
: IRequest<Either<BaseError, List<PlaylistItemViewModel>>>;
@@ -15,10 +15,21 @@ public class ReplacePlaylistItemsHandler(IDbContextFactory<TvContext> dbContextF
{
await using TvContext dbContext = await dbContextFactory.CreateDbContextAsync(cancellationToken);
Validation<BaseError, Playlist> validation = await Validate(dbContext, request, cancellationToken);
return await validation.Apply(ps => Persist(dbContext, request, ps, cancellationToken));
// Introduce the optimistic-concurrency check as a standalone Either AFTER the validation
// pipeline (never via Apply) so a PreconditionFailedError survives to a 412 and is not
// flattened to a generic 422 by Join() (issue #253 / api-conventions §7a).
// LanguageExtensions.ToEither joins the Seq<BaseError> to a single BaseError (the native
// Validation.ToEither() would keep Seq and is called explicitly here to avoid that shadow).
Either<BaseError, Playlist> validated = LanguageExtensions.ToEither(validation)
.Bind(playlist => playlist.CheckVersion(request.ExpectedVersion));
return await validated.Match(
Right: playlist => Persist(dbContext, request, playlist, cancellationToken),
Left: error => Task.FromResult<Either<BaseError, List<PlaylistItemViewModel>>>(error));
}
private static async Task<List<PlaylistItemViewModel>> Persist(
private static async Task<Either<BaseError, List<PlaylistItemViewModel>>> Persist(
TvContext dbContext,
ReplacePlaylistItems request,
Playlist playlist,
@@ -30,9 +41,15 @@ public class ReplacePlaylistItemsHandler(IDbContextFactory<TvContext> dbContextF
dbContext.RemoveRange(playlist.Items);
playlist.Items = request.Items.Map(i => BuildItem(playlist, i.Index, i)).ToList();
await dbContext.SaveChangesAsync(cancellationToken);
// Unconditional bump: EF emits the root UPDATE only when a scalar actually differs, so a
// same-value/no-op save would otherwise write no root row and neither fire the concurrency
// token nor rotate other clients' ETags. Bumping guarantees both on every save (issue #253).
playlist.Version++;
return playlist.Items.Map(Mapper.ProjectToViewModel).ToList();
// Save through the guard so an EF concurrency failure (a racing writer won between our load
// and save) maps to 412 rather than surfacing as a 500.
Either<BaseError, Unit> saved = await dbContext.SaveChangesWithConcurrencyGuard(cancellationToken);
return saved.Map(_ => playlist.Items.Map(Mapper.ProjectToViewModel).ToList());
}
private static PlaylistItem BuildItem(Playlist playlist, int index, ReplacePlaylistItem item) =>
@@ -4,6 +4,7 @@ namespace ErsatzTV.Application.MediaCollections;
public record UpdateCollectionCustomOrder(
int CollectionId,
List<MediaItemCustomOrder> MediaItemCustomOrders) : IRequest<Either<BaseError, Unit>>;
List<MediaItemCustomOrder> MediaItemCustomOrders,
Option<int> ExpectedVersion = default) : IRequest<Either<BaseError, Unit>>;
public record MediaItemCustomOrder(int MediaItemId, int CustomIndex);
@@ -32,13 +32,22 @@ public class UpdateCollectionCustomOrderHandler : IRequestHandler<UpdateCollecti
{
await using TvContext dbContext = await _dbContextFactory.CreateDbContextAsync(cancellationToken);
Validation<BaseError, Collection> validation = await Validate(dbContext, request, cancellationToken);
return await validation.Apply(c => ApplyUpdateRequest(dbContext, c, request));
// Optimistic-concurrency check as a standalone Either AFTER validation, never via Apply (which
// Join()s the error Seq and would flatten PreconditionFailedError to a 422) — issue #253 §7a.
Either<BaseError, Collection> validated = LanguageExtensions.ToEither(validation)
.Bind(c => c.CheckVersion(request.ExpectedVersion));
return await validated.Match(
Right: c => ApplyUpdateRequest(dbContext, c, request, cancellationToken),
Left: error => Task.FromResult<Either<BaseError, Unit>>(error));
}
private async Task<Unit> ApplyUpdateRequest(
private async Task<Either<BaseError, Unit>> ApplyUpdateRequest(
TvContext dbContext,
Collection c,
UpdateCollectionCustomOrder request)
UpdateCollectionCustomOrder request,
CancellationToken cancellationToken)
{
foreach (MediaItemCustomOrder updateItem in request.MediaItemCustomOrders)
{
@@ -51,14 +60,24 @@ public class UpdateCollectionCustomOrderHandler : IRequestHandler<UpdateCollecti
}
}
if (await dbContext.SaveChangesAsync() > 0)
// Unconditional bump (issue #253 §7a / M1) then guarded save (→ 412 on a lost race).
c.Version++;
Either<BaseError, Unit> saveResult = await dbContext.SaveChangesWithConcurrencyGuard(cancellationToken);
if (saveResult.IsLeft)
{
// refresh all playouts that use this collection
foreach (int playoutId in await _mediaCollectionRepository
.PlayoutIdsUsingCollection(request.CollectionId))
{
await _channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh));
}
return saveResult;
}
// Refresh all playouts that use this collection. The old `SaveChangesAsync() > 0` gate is always
// true once the version bumps unconditionally (M2), so run the refresh on any successful save.
// Post-commit enqueue on CancellationToken.None so a late cancellation can't drop the rebuild
// after the commit landed (#254 / §7b).
foreach (int playoutId in await _mediaCollectionRepository
.PlayoutIdsUsingCollection(request.CollectionId))
{
await _channel.WriteAsync(
new BuildPlayout(playoutId, PlayoutBuildMode.Refresh),
CancellationToken.None);
}
return Unit.Default;
@@ -59,13 +59,17 @@ public class UpdateCollectionHandler : IRequestHandler<UpdateCollection, Either<
c.UseCustomPlaybackOrder = useCustomPlaybackOrder;
}
if (await dbContext.SaveChangesAsync(cancellationToken) > 0 && request.UseCustomPlaybackOrder.IsSome)
// Force-write past a concurrent Version bump from the custom-order editor (this name/flag writer
// doesn't participate in If-Match, so the active token must not 500 a benign race) — #253/#269.
if (await dbContext.SaveChangesForcingVersion(cancellationToken) > 0 && request.UseCustomPlaybackOrder.IsSome)
{
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
// refresh all playouts that use this collection
foreach (int playoutId in await _mediaCollectionRepository.PlayoutIdsUsingCollection(
request.CollectionId))
{
await _channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh), cancellationToken);
await _channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh), CancellationToken.None);
}
}
@@ -12,4 +12,5 @@ public record UpdateMultiCollectionItem(
public record UpdateMultiCollection(
int MultiCollectionId,
string Name,
List<UpdateMultiCollectionItem> Items) : IRequest<Either<BaseError, Unit>>;
List<UpdateMultiCollectionItem> Items,
Option<int> ExpectedVersion = default) : IRequest<Either<BaseError, Unit>>;
@@ -36,10 +36,18 @@ public class UpdateMultiCollectionHandler : IRequestHandler<UpdateMultiCollectio
{
await using TvContext dbContext = await _dbContextFactory.CreateDbContextAsync(cancellationToken);
Validation<BaseError, MultiCollection> validation = await Validate(dbContext, request, cancellationToken);
return await validation.Apply(c => ApplyUpdateRequest(dbContext, c, request, cancellationToken));
// Optimistic-concurrency check as a standalone Either AFTER validation, never via Apply (which
// Join()s the error Seq and would flatten PreconditionFailedError to a 422) — issue #253 §7a.
Either<BaseError, MultiCollection> validated = LanguageExtensions.ToEither(validation)
.Bind(c => c.CheckVersion(request.ExpectedVersion));
return await validated.Match(
Right: c => ApplyUpdateRequest(dbContext, c, request, cancellationToken),
Left: error => Task.FromResult<Either<BaseError, Unit>>(error));
}
private async Task<Unit> ApplyUpdateRequest(
private async Task<Either<BaseError, Unit>> ApplyUpdateRequest(
TvContext dbContext,
MultiCollection c,
UpdateMultiCollection request,
@@ -47,8 +55,16 @@ public class UpdateMultiCollectionHandler : IRequestHandler<UpdateMultiCollectio
{
c.Name = request.Name;
// save name first so playouts don't get rebuilt for a name change
await dbContext.SaveChangesAsync(cancellationToken);
// Bump the version on this first save (issue #253 §7a): it rotates other clients' ETags and,
// via IsConcurrencyToken, closes the load→save race — a stale writer's WHERE Version=@orig hits
// 0 rows → PreconditionFailedError (412). Saving the name first also keeps a name-only change
// from triggering a playout rebuild (the item save below stays gated on real item changes).
c.Version++;
Either<BaseError, Unit> nameSave = await dbContext.SaveChangesWithConcurrencyGuard(cancellationToken);
if (nameSave.IsLeft)
{
return nameSave;
}
var toAdd = request.Items
.Filter(i => i.CollectionId.HasValue)
@@ -124,10 +140,12 @@ public class UpdateMultiCollectionHandler : IRequestHandler<UpdateMultiCollectio
_searchTargets.SearchTargetsChanged();
// refresh all playouts that use this collection
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
foreach (int playoutId in await _mediaCollectionRepository.PlayoutIdsUsingMultiCollection(
request.MultiCollectionId))
{
await _channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh), cancellationToken);
await _channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh), CancellationToken.None);
}
}
@@ -13,5 +13,6 @@ public record UpdateRerunCollection(
SmartCollectionViewModel SmartCollection,
NamedMediaItemViewModel MediaItem,
PlaybackOrder FirstRunPlaybackOrder,
PlaybackOrder RerunPlaybackOrder)
PlaybackOrder RerunPlaybackOrder,
Option<int> ExpectedVersion = default)
: IRequest<Either<BaseError, Unit>>;
@@ -22,10 +22,18 @@ public class UpdateRerunCollectionHandler(
{
await using TvContext dbContext = await dbContextFactory.CreateDbContextAsync(cancellationToken);
Validation<BaseError, RerunCollection> validation = await Validate(dbContext, request, cancellationToken);
return await validation.Apply(c => ApplyUpdateRequest(dbContext, c, request, cancellationToken));
// Optimistic-concurrency check as a standalone Either AFTER validation, never via Apply (which
// Join()s the error Seq and would flatten PreconditionFailedError to a 422) — issue #253 §7a.
Either<BaseError, RerunCollection> validated = LanguageExtensions.ToEither(validation)
.Bind(c => c.CheckVersion(request.ExpectedVersion));
return await validated.Match(
Right: c => ApplyUpdateRequest(dbContext, c, request, cancellationToken),
Left: error => Task.FromResult<Either<BaseError, Unit>>(error));
}
private async Task<Unit> ApplyUpdateRequest(
private async Task<Either<BaseError, Unit>> ApplyUpdateRequest(
TvContext dbContext,
RerunCollection c,
UpdateRerunCollection request,
@@ -40,15 +48,23 @@ public class UpdateRerunCollectionHandler(
c.FirstRunPlaybackOrder = request.FirstRunPlaybackOrder;
c.RerunPlaybackOrder = request.RerunPlaybackOrder;
// rebuild playouts
if (await dbContext.SaveChangesAsync(cancellationToken) > 0)
// Unconditional bump (issue #253 §7a / M1) then guarded save, which maps a racing
// DbUpdateConcurrencyException to PreconditionFailedError (→ 412).
c.Version++;
Either<BaseError, Unit> saveResult = await dbContext.SaveChangesWithConcurrencyGuard(cancellationToken);
if (saveResult.IsLeft)
{
// refresh all playouts that use this rerun collection
foreach (int playoutId in await mediaCollectionRepository.PlayoutIdsUsingRerunCollection(
request.RerunCollectionId))
{
await channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh), cancellationToken);
}
return saveResult;
}
// Refresh all playouts that use this rerun collection. The old `SaveChangesAsync() > 0` gate is
// always true once the version bumps unconditionally (M2), so run the refresh on any successful
// save. Post-commit enqueue on CancellationToken.None so a late cancellation can't drop the
// rebuild after the commit landed (#254).
foreach (int playoutId in await mediaCollectionRepository.PlayoutIdsUsingRerunCollection(
request.RerunCollectionId))
{
await channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh), CancellationToken.None);
}
return Unit.Default;
@@ -69,12 +69,15 @@ public class
if (await dbContext.SaveChangesAsync(cancellationToken) > 0)
{
_searchTargets.SearchTargetsChanged();
await _smartCollectionCache.Refresh(cancellationToken);
// post-commit side effects run on CancellationToken.None so a late request cancellation
// can't abort them after the commit landed (#254)
await _smartCollectionCache.Refresh(CancellationToken.None);
// refresh all playouts that use this smart collection
foreach (int playoutId in await _mediaCollectionRepository.PlayoutIdsUsingSmartCollection(request.Id))
{
await _channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh), cancellationToken);
await _channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Refresh), CancellationToken.None);
}
}
@@ -57,7 +57,9 @@ public class UpdateTraktListHandler(
{
try
{
await workerChannel.WriteAsync(new MatchTraktListItems(traktList.Id), cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request
// cancellation can't abort it after the commit landed (#254)
await workerChannel.WriteAsync(new MatchTraktListItems(traktList.Id), CancellationToken.None);
}
catch
{
@@ -12,14 +12,16 @@ internal static class Mapper
collection.Id,
collection.Name,
collection.UseCustomPlaybackOrder,
MediaItemState.Normal);
MediaItemState.Normal,
collection.Version);
internal static MultiCollectionViewModel ProjectToViewModel(MultiCollection multiCollection) =>
new(
multiCollection.Id,
multiCollection.Name,
Optional(multiCollection.MultiCollectionItems).Flatten().Map(ProjectToViewModel).ToList(),
Optional(multiCollection.MultiCollectionSmartItems).Flatten().Map(ProjectToViewModel).ToList());
Optional(multiCollection.MultiCollectionSmartItems).Flatten().Map(ProjectToViewModel).ToList(),
multiCollection.Version);
internal static SmartCollectionViewModel ProjectToViewModel(SmartCollection collection) =>
new(collection.Id, collection.Name, collection.Query);
@@ -49,7 +51,8 @@ internal static class Mapper
_ => null
},
collection.FirstRunPlaybackOrder,
collection.RerunPlaybackOrder);
collection.RerunPlaybackOrder,
collection.Version);
internal static TraktListViewModel ProjectToViewModel(TraktList traktList) =>
new(
@@ -89,7 +92,7 @@ internal static class Mapper
new(playlistGroup.Id, playlistGroup.Name, playlistGroup.Playlists.Count, playlistGroup.IsSystem);
internal static PlaylistViewModel ProjectToViewModel(Playlist playlist) =>
new(playlist.Id, playlist.PlaylistGroupId, playlist.Name, playlist.IsSystem);
new(playlist.Id, playlist.PlaylistGroupId, playlist.Name, playlist.IsSystem, playlist.Version);
internal static PlaylistItemViewModel ProjectToViewModel(PlaylistItem playlistItem) =>
new(
@@ -8,7 +8,10 @@ public record MediaCollectionViewModel(
int Id,
string Name,
bool UseCustomPlaybackOrder,
MediaItemState State) : MediaCardViewModel(
MediaItemState State,
// Optimistic-concurrency token (issue #253), header-only via ETag; 0 for the selection/placeholder
// constructions that don't read a real collection. Set from the entity by the Mapper projection.
int Version = 0) : MediaCardViewModel(
Id,
Name,
string.Empty,
@@ -4,4 +4,6 @@ public record MultiCollectionViewModel(
int Id,
string Name,
List<MultiCollectionItemViewModel> Items,
List<MultiCollectionSmartItemViewModel> SmartItems);
List<MultiCollectionSmartItemViewModel> SmartItems,
// Optimistic-concurrency token (issue #253), header-only via ETag; 0 for selection placeholders.
int Version = 0);
@@ -1,3 +1,3 @@
namespace ErsatzTV.Application.MediaCollections;
public record PlaylistViewModel(int Id, int PlaylistGroupId, string Name, bool IsSystem);
public record PlaylistViewModel(int Id, int PlaylistGroupId, string Name, bool IsSystem, int Version);
@@ -12,4 +12,6 @@ public record RerunCollectionViewModel(
SmartCollectionViewModel SmartCollection,
NamedMediaItemViewModel MediaItem,
PlaybackOrder FirstRunPlaybackOrder,
PlaybackOrder RerunPlaybackOrder);
PlaybackOrder RerunPlaybackOrder,
// Optimistic-concurrency token (issue #253), header-only via ETag.
int Version = 0);
@@ -62,9 +62,20 @@ public class BuildPlayoutHandler : IRequestHandler<BuildPlayout, Either<BaseErro
public async Task<Either<BaseError, Unit>> Handle(BuildPlayout request, CancellationToken cancellationToken)
{
// respect the EntityLocker ownership contract: LockPlayout returns true only for the caller
// that performed the 0->1 transition. If another operation already holds this playout's lock
// (a concurrent build, or a subtitle extraction), skip rather than build unlocked and then
// cross-release the other owner's lock in the finally.
if (!await _entityLocker.LockPlayout(request.PlayoutId))
{
_logger.LogDebug(
"Playout {PlayoutId} is already locked; skipping this build request",
request.PlayoutId);
return Unit.Default;
}
try
{
await _entityLocker.LockPlayout(request.PlayoutId);
if (request.Mode is not PlayoutBuildMode.Reset)
{
// this needs to happen before we load the playout in this handler because it modifies items, etc
@@ -35,15 +35,18 @@ public class CreateScriptedPlayoutHandler(
{
await dbContext.Playouts.AddAsync(playout, cancellationToken);
await dbContext.SaveChangesAsync(cancellationToken);
await channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Reset), cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Reset), CancellationToken.None);
if (playout.Channel.PlayoutMode is ChannelPlayoutMode.OnDemand)
{
await channel.WriteAsync(
new TimeShiftOnDemandPlayout(playout.Id, DateTimeOffset.Now, false),
cancellationToken);
CancellationToken.None);
}
await channel.WriteAsync(new RefreshChannelList(), cancellationToken);
await channel.WriteAsync(new RefreshChannelList(), CancellationToken.None);
return new CreatePlayoutResponse(playout.Id);
}
@@ -33,15 +33,18 @@ public class CreateSequentialPlayoutHandler(
{
await dbContext.Playouts.AddAsync(playout, cancellationToken);
await dbContext.SaveChangesAsync(cancellationToken);
await channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Reset), cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Reset), CancellationToken.None);
if (playout.Channel.PlayoutMode is ChannelPlayoutMode.OnDemand)
{
await channel.WriteAsync(
new TimeShiftOnDemandPlayout(playout.Id, DateTimeOffset.Now, false),
cancellationToken);
CancellationToken.None);
}
await channel.WriteAsync(new RefreshChannelList(), cancellationToken);
await channel.WriteAsync(new RefreshChannelList(), CancellationToken.None);
return new CreatePlayoutResponse(playout.Id);
}
@@ -28,20 +28,25 @@ public class DeletePlayoutHandler(
foreach (Playout playout in maybePlayout)
{
dbContext.Playouts.Remove(playout);
await dbContext.SaveChangesAsync(cancellationToken);
// delete channel data from channel guide cache
// Delete the guide cache file through the filesystem abstraction (observable under a
// MockFileSystem) and BEFORE the commit: deleting after commit orphans {number}.xml if the
// process crashes in between. The guide xml is regenerable on demand, so losing it pre-commit
// is safe (#254).
string cacheFile = Path.Combine(FileSystemLayout.ChannelGuideCacheFolder, $"{playout.Channel.Number}.xml");
if (fileSystem.File.Exists(cacheFile))
{
File.Delete(cacheFile);
fileSystem.File.Delete(cacheFile);
}
// refresh channel list to remove channel that has no playout
await workerChannel.WriteAsync(new RefreshChannelList(), cancellationToken);
dbContext.Playouts.Remove(playout);
await dbContext.SaveChangesAsync(cancellationToken);
await mediator.Publish(new PlayoutUpdatedNotification(playout.Id, false), cancellationToken);
// post-commit side effects run on CancellationToken.None so a late request cancellation can't
// abort them after the delete committed (#254)
// refresh channel list to remove channel that has no playout
await workerChannel.WriteAsync(new RefreshChannelList(), CancellationToken.None);
await mediator.Publish(new PlayoutUpdatedNotification(playout.Id, false), CancellationToken.None);
}
return maybePlayout
@@ -2,5 +2,8 @@ using ErsatzTV.Core;
namespace ErsatzTV.Application.Playouts;
public record ReplacePlayoutAlternateScheduleItems(int PlayoutId, List<ReplacePlayoutAlternateSchedule> Items)
public record ReplacePlayoutAlternateScheduleItems(
int PlayoutId,
List<ReplacePlayoutAlternateSchedule> Items,
Option<int> ExpectedVersion = default)
: IRequest<Either<BaseError, Unit>>;
@@ -20,7 +20,13 @@ public class ReplacePlayoutAlternateScheduleItemsHandler(
ReplacePlayoutAlternateScheduleItems request,
CancellationToken cancellationToken)
{
// TODO: validate that items is not empty
// The handler reads the highest-index item as the default schedule (Max() below), so an empty
// list is invalid — reject it here rather than letting Max() throw. The controller pre-guards
// too, but a direct handler caller (MCP, test, reuse) must get a clean 422, not a raw crash (#254).
if (request.Items.Count == 0)
{
return BaseError.New("Playout alternate schedule items must not be empty");
}
try
{
@@ -34,6 +40,18 @@ public class ReplacePlayoutAlternateScheduleItemsHandler(
foreach (Playout playout in maybePlayout)
{
// Optimistic-concurrency pre-check (issue #253 §7a): reject a stale If-Match with a
// PreconditionFailedError (→ 412) before any mutation. Introduced as a standalone Either,
// and returned directly (a value, not a throw) so it escapes the catch(Exception) below
// rather than being reshaped into a bare 422 (§9/H1).
Either<BaseError, Playout> versionCheck = playout.CheckVersion(request.ExpectedVersion);
if (versionCheck.IsLeft)
{
return versionCheck.Match<Either<BaseError, Unit>>(
Left: error => error,
Right: _ => Unit.Default);
}
var existingScheduleMap = new Dictionary<DateTimeOffset, ProgramSchedule>();
var daysToCheck = new List<DateTimeOffset>();
@@ -121,7 +139,20 @@ public class ReplacePlayoutAlternateScheduleItemsHandler(
playout.ProgramScheduleId = highest.ProgramScheduleId;
}
await dbContext.SaveChangesAsync(cancellationToken);
// Unconditional bump (issue #253 §7a / M1): EF emits the root UPDATE only when a scalar
// actually differs, so a no-op PUT-back would otherwise neither fire the concurrency token
// nor rotate other clients' ETags.
playout.Version++;
// Guarded save maps a racing DbUpdateConcurrencyException to PreconditionFailedError (→ 412)
// as a return value, so a lost race short-circuits here before the post-commit refresh block
// and escapes the catch(Exception) below as a 412, not a 422 (§9/H1).
Either<BaseError, Unit> saveResult =
await dbContext.SaveChangesWithConcurrencyGuard(cancellationToken);
if (saveResult.IsLeft)
{
return saveResult;
}
if (hasDefaultScheduleChange)
{
@@ -152,9 +183,11 @@ public class ReplacePlayoutAlternateScheduleItemsHandler(
existingValue.Name,
schedule.Name);
// post-commit enqueue runs on CancellationToken.None: the schedule change is
// already committed, so a late cancellation must not drop the rebuild (#254)
await channel.WriteAsync(
new BuildPlayout(request.PlayoutId, PlayoutBuildMode.Refresh),
cancellationToken);
CancellationToken.None);
break;
}
@@ -1,3 +1,3 @@
namespace ErsatzTV.Application.Playouts;
public record ResetAllPlayouts : IRequest;
public record ResetAllPlayouts : IRequest<ResetAllPlayoutsResult>;
@@ -11,33 +11,49 @@ public class ResetAllPlayoutsHandler(
IEntityLocker locker,
ChannelWriter<IBackgroundServiceRequest> channel,
IDbContextFactory<TvContext> dbContextFactory)
: IRequestHandler<ResetAllPlayouts>
: IRequestHandler<ResetAllPlayouts, ResetAllPlayoutsResult>
{
public async Task Handle(ResetAllPlayouts request, CancellationToken cancellationToken)
public async Task<ResetAllPlayoutsResult> Handle(
ResetAllPlayouts request,
CancellationToken cancellationToken)
{
await using TvContext dbContext = await dbContextFactory.CreateDbContextAsync(cancellationToken);
var queued = new List<int>();
var skippedLocked = new List<int>();
var skippedUnsupported = new List<int>();
foreach (Playout playout in await dbContext.Playouts.ToListAsync(cancellationToken))
{
switch (playout.ScheduleKind)
{
case PlayoutScheduleKind.Classic:
if (!locker.IsPlayoutLocked(playout.Id))
if (locker.IsPlayoutLocked(playout.Id))
{
skippedLocked.Add(playout.Id);
}
else
{
await channel.WriteAsync(
new BuildPlayout(playout.Id, PlayoutBuildMode.Refresh),
cancellationToken);
queued.Add(playout.Id);
}
break;
case PlayoutScheduleKind.Block:
case PlayoutScheduleKind.Sequential:
case PlayoutScheduleKind.Scripted:
if (!locker.IsPlayoutLocked(playout.Id))
if (locker.IsPlayoutLocked(playout.Id))
{
skippedLocked.Add(playout.Id);
}
else
{
await channel.WriteAsync(
new BuildPlayout(playout.Id, PlayoutBuildMode.Reset),
cancellationToken);
queued.Add(playout.Id);
}
break;
@@ -45,8 +61,11 @@ public class ResetAllPlayoutsHandler(
case PlayoutScheduleKind.None:
default:
// external json cannot be reset
skippedUnsupported.Add(playout.Id);
continue;
}
}
return new ResetAllPlayoutsResult(queued, skippedLocked, skippedUnsupported);
}
}
@@ -40,9 +40,13 @@ public class
{
playout.ScheduleFile = request.ScheduleFile;
if (await dbContext.SaveChangesAsync(cancellationToken) > 0)
// Force-write past a concurrent Version bump from a replace-all editor (schedule-file writer
// doesn't participate in If-Match, so an active token must not 500 a benign race) — #253/#269.
if (await dbContext.SaveChangesForcingVersion(cancellationToken) > 0)
{
await _workerChannel.WriteAsync(new RefreshChannelData(playout.Channel.Number), cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await _workerChannel.WriteAsync(new RefreshChannelData(playout.Channel.Number), CancellationToken.None);
}
return new PlayoutNameViewModel(
@@ -56,7 +60,8 @@ public class
playout.DailyRebuildTime,
playout.BuildStatus,
playout.DecoId,
playout.Deco?.Name);
playout.Deco?.Name,
playout.Version);
}
private static Task<Validation<BaseError, Playout>> Validate(
@@ -61,7 +61,10 @@ public class UpdateOnDemandCheckpointHandler(
playout.Channel.Name,
playout.OnDemandCheckpoint);
await dbContext.SaveChangesAsync(cancellationToken);
// Force-write past a concurrent Version bump from a replace-all editor: the on-demand
// checkpoint writer touches the Playout root but doesn't participate in If-Match, so the
// active concurrency token must not turn a benign race into a 500 (#253/#269).
await dbContext.SaveChangesForcingVersion(cancellationToken);
}
}
}
@@ -35,7 +35,9 @@ public class UpdatePlayoutHandler : IRequestHandler<UpdatePlayout, Either<BaseEr
playout.DailyRebuildTime = dailyRebuildTime;
}
await dbContext.SaveChangesAsync(cancellationToken);
// Force-write past a concurrent Version bump from a replace-all editor (this settings writer
// doesn't participate in If-Match, so a missing token = force-write, not a 500) — #253/#269.
await dbContext.SaveChangesForcingVersion(cancellationToken);
return new PlayoutNameViewModel(
playout.Id,
@@ -48,7 +50,8 @@ public class UpdatePlayoutHandler : IRequestHandler<UpdatePlayout, Either<BaseEr
playout.DailyRebuildTime,
playout.BuildStatus,
playout.DecoId,
playout.Deco?.Name);
playout.Deco?.Name,
playout.Version);
}
private static Task<Validation<BaseError, Playout>> Validate(
@@ -35,9 +35,13 @@ public class
{
playout.ScheduleFile = request.ScheduleFile;
if (await dbContext.SaveChangesAsync(cancellationToken) > 0)
// Force-write past a concurrent Version bump from a replace-all editor (schedule-file writer
// doesn't participate in If-Match, so an active token must not 500 a benign race) — #253/#269.
if (await dbContext.SaveChangesForcingVersion(cancellationToken) > 0)
{
await workerChannel.WriteAsync(new RefreshChannelData(playout.Channel.Number), cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await workerChannel.WriteAsync(new RefreshChannelData(playout.Channel.Number), CancellationToken.None);
}
return new PlayoutNameViewModel(
@@ -51,7 +55,8 @@ public class
playout.DailyRebuildTime,
playout.BuildStatus,
playout.DecoId,
playout.Deco?.Name);
playout.Deco?.Name,
playout.Version);
}
private async Task<Validation<BaseError, Playout>> Validate(
@@ -40,9 +40,13 @@ public class
{
playout.ScheduleFile = request.ScheduleFile;
if (await dbContext.SaveChangesAsync(cancellationToken) > 0)
// Force-write past a concurrent Version bump from a replace-all editor (schedule-file writer
// doesn't participate in If-Match, so an active token must not 500 a benign race) — #253/#269.
if (await dbContext.SaveChangesForcingVersion(cancellationToken) > 0)
{
await _workerChannel.WriteAsync(new RefreshChannelData(playout.Channel.Number), cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
await _workerChannel.WriteAsync(new RefreshChannelData(playout.Channel.Number), CancellationToken.None);
}
return new PlayoutNameViewModel(
@@ -56,7 +60,8 @@ public class
playout.DailyRebuildTime,
playout.BuildStatus,
playout.DecoId,
playout.Deco?.Name);
playout.Deco?.Name,
playout.Version);
}
private static Task<Validation<BaseError, Playout>> Validate(
+2 -1
View File
@@ -19,7 +19,8 @@ internal static class Mapper
playout.DecoId,
// the paged-playouts query does not eager-load Deco (the list response does not surface
// the default deco); GetPlayoutById includes it for the detail response
playout.Deco?.Name);
playout.Deco?.Name,
playout.Version);
internal static PlayoutItemViewModel ProjectToViewModel(PlayoutItem playoutItem) =>
new(
@@ -13,7 +13,8 @@ public record PlayoutNameViewModel(
TimeSpan? DbDailyRebuildTime,
PlayoutBuildStatus BuildStatus,
int? DecoId,
string DecoName)
string DecoName,
int Version)
{
public Option<TimeSpan> DailyRebuildTime => Optional(DbDailyRebuildTime);
@@ -30,6 +30,7 @@ public class GetPlayoutByIdHandler(IDbContextFactory<TvContext> dbContextFactory
p.DailyRebuildTime,
p.BuildStatus,
p.DecoId,
p.DecoId == null ? null : p.Deco.Name));
p.DecoId == null ? null : p.Deco.Name,
p.Version));
}
}
@@ -0,0 +1,6 @@
namespace ErsatzTV.Application.Playouts;
public record ResetAllPlayoutsResult(
List<int> QueuedPlayoutIds,
List<int> SkippedLocked,
List<int> SkippedUnsupported);
@@ -27,12 +27,22 @@ public class SignOutOfPlexHandler : IRequestHandler<SignOutOfPlex, Either<BaseEr
public async Task<Either<BaseError, Unit>> Handle(SignOutOfPlex request, CancellationToken cancellationToken)
{
List<int> ids = await _mediaSourceRepository.DeleteAllPlex();
await _searchIndex.RemoveItems(ids);
_searchIndex.Commit();
await _plexSecretStore.DeleteAll();
_entityLocker.UnlockPlex();
// Terminal handler (no lock handoff): release the Plex lock on EVERY exit via finally so a throw
// from any awaited dependency (repo delete, search-index commit, secret store) cannot wedge Plex
// locked at 409 until restart (#202 §C6 / finding 7). This is the UNCONDITIONAL-finally case —
// contrast the pin-flow handlers, which hand the lock off and must NOT use a blanket finally.
try
{
List<int> ids = await _mediaSourceRepository.DeleteAllPlex();
await _searchIndex.RemoveItems(ids);
_searchIndex.Commit();
await _plexSecretStore.DeleteAll();
return Unit.Default;
return Unit.Default;
}
finally
{
_entityLocker.UnlockPlex();
}
}
}
@@ -2,5 +2,5 @@ using ErsatzTV.Core;
namespace ErsatzTV.Application.Plex;
public record SynchronizePlexCollections(int PlexMediaSourceId, bool ForceScan, bool DeepScan)
public record SynchronizePlexCollections(int PlexMediaSourceId, bool ForceScan, bool DeepScan, bool Unlock = true)
: IRequest<Either<BaseError, Unit>>, IScannerBackgroundServiceRequest;
@@ -1,5 +1,6 @@
using System.Threading.Channels;
using System.Threading.Channels;
using ErsatzTV.Core;
using ErsatzTV.Core.Interfaces.Locking;
using ErsatzTV.Core.Interfaces.Plex;
namespace ErsatzTV.Application.Plex;
@@ -7,34 +8,59 @@ namespace ErsatzTV.Application.Plex;
public class TryCompletePlexPinFlowHandler : IRequestHandler<TryCompletePlexPinFlow, Either<BaseError, bool>>
{
private readonly ChannelWriter<IPlexBackgroundServiceRequest> _channel;
private readonly IEntityLocker _entityLocker;
private readonly IPlexTvApiClient _plexTvApiClient;
public TryCompletePlexPinFlowHandler(
IPlexTvApiClient plexTvApiClient,
ChannelWriter<IPlexBackgroundServiceRequest> channel)
ChannelWriter<IPlexBackgroundServiceRequest> channel,
IEntityLocker entityLocker)
{
_plexTvApiClient = plexTvApiClient;
_channel = channel;
_entityLocker = entityLocker;
}
public async Task<Either<BaseError, bool>>
Handle(TryCompletePlexPinFlow request, CancellationToken cancellationToken)
{
// Lock-release discipline (#202 §C1.6 / §C6): the Plex lock this pin flow holds is released
// ONLY on non-handoff exits — the 2-minute timeout (Task.Delay throws
// OperationCanceledException), a poll exception, a failed enqueue, or the (effectively dead)
// return-false at loop entry. On SUCCESS the lock is HANDED OFF to SynchronizePlexMediaSources,
// whose handler is the sole releaser after server discovery (SynchronizePlexMediaSourcesHandler).
// This is deliberately NOT an unconditional finally: a blanket release here would double-release
// AND release before discovery, re-opening the finding-5 race (an empty server list reading as
// success). Contrast the terminal SignOutOfPlexHandler, which DOES use finally (it has no handoff).
using var cts = new CancellationTokenSource(TimeSpan.FromMinutes(2));
using var linkedTokenSource = CancellationTokenSource.CreateLinkedTokenSource(cts.Token, cancellationToken);
CancellationToken token = linkedTokenSource.Token;
while (!token.IsCancellationRequested)
try
{
bool result = await _plexTvApiClient.TryCompletePinFlow(request.AuthPin);
if (result)
while (!token.IsCancellationRequested)
{
await _channel.WriteAsync(new SynchronizePlexMediaSources(), token);
return true;
bool result = await _plexTvApiClient.TryCompletePinFlow(request.AuthPin);
if (result)
{
// hand the lock off to the sync handler — do NOT release on this success path
await _channel.WriteAsync(new SynchronizePlexMediaSources(), token);
return true;
}
await Task.Delay(TimeSpan.FromSeconds(1), token);
}
await Task.Delay(TimeSpan.FromSeconds(1), token);
// effectively unreachable (Task.Delay throws on cancellation before the loop condition is
// re-evaluated) but if the flow ever ends here it abandoned without auth → release
_entityLocker.UnlockPlex();
return false;
}
catch (Exception)
{
// non-handoff exit: timeout-throw, poll exception, or failed enqueue — release the lock so an
// abandoned flow does not wedge Plex locked, then rethrow (PlexService logs it as before)
_entityLocker.UnlockPlex();
throw;
}
return false;
}
}
@@ -35,14 +35,56 @@ public class
private static PlexPathReplacement Project(PlexPathReplacementItem vm) =>
new() { Id = vm.Id, PlexPath = vm.PlexPath, LocalPath = vm.LocalPath };
private Task<Validation<BaseError, PlexMediaSource>> Validate(
private async Task<Validation<BaseError, PlexMediaSource>> Validate(
UpdatePlexPathReplacements request,
CancellationToken cancellationToken) =>
PlexMediaSourceMustExist(request, cancellationToken);
CancellationToken cancellationToken)
{
Option<PlexMediaSource> maybeSource =
await _mediaSourceRepository.GetPlex(request.PlexMediaSourceId, cancellationToken);
private Task<Validation<BaseError, PlexMediaSource>> PlexMediaSourceMustExist(
foreach (PlexMediaSource plexMediaSource in maybeSource)
{
return ValidatePathReplacements(request, plexMediaSource);
}
return Fail<BaseError, PlexMediaSource>(
BaseError.New($"Plex media source {request.PlexMediaSourceId} does not exist."));
}
// Programmatic clients now reach this handler directly (#202), so validate what the Blazor form
// enforced plus the cross-source ownership hole (finding 2c/8): reject a null list, null items,
// blank RemotePath/LocalPath, and any positive Id NOT owned by this source — all before any write,
// so there is no partial mutation.
private static Validation<BaseError, PlexMediaSource> ValidatePathReplacements(
UpdatePlexPathReplacements request,
CancellationToken cancellationToken) =>
_mediaSourceRepository.GetPlex(request.PlexMediaSourceId, cancellationToken)
.Map(v => v.ToValidation<BaseError>($"Plex media source {request.PlexMediaSourceId} does not exist."));
PlexMediaSource plexMediaSource)
{
List<PlexPathReplacementItem> items = request.PathReplacements;
if (items is null)
{
return Fail<BaseError, PlexMediaSource>(BaseError.New("[PathReplacements] is required"));
}
if (items.Any(i => i is null))
{
return Fail<BaseError, PlexMediaSource>(BaseError.New("[PathReplacements] must not contain null items"));
}
if (items.Any(i => string.IsNullOrWhiteSpace(i.PlexPath) || string.IsNullOrWhiteSpace(i.LocalPath)))
{
return Fail<BaseError, PlexMediaSource>(
BaseError.New("Each path replacement requires a non-empty RemotePath and LocalPath"));
}
var ownedIds = Optional(plexMediaSource.PathReplacements).Flatten().Map(pr => pr.Id).ToHashSet();
List<int> foreignIds = items.Filter(i => i.Id > 0 && !ownedIds.Contains(i.Id)).Map(i => i.Id).ToList();
if (foreignIds.Count > 0)
{
return Fail<BaseError, PlexMediaSource>(
BaseError.New(
$"Path replacement {foreignIds[0]} does not belong to Plex media source {request.PlexMediaSourceId}"));
}
return Success<BaseError, PlexMediaSource>(plexMediaSource);
}
}
@@ -52,12 +52,17 @@ public class AddProgramScheduleItemHandler : ProgramScheduleItemCommandBase,
ProgramScheduleItem item = BuildItem(programSchedule, nextIndex, request);
programSchedule.Items.Add(item);
// bump the optimistic-concurrency token so this config edit rotates other clients' ETags (#253)
programSchedule.Version++;
await dbContext.SaveChangesAsync(cancellationToken);
// refresh any playouts that use this schedule
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
foreach (Playout playout in programSchedule.Playouts)
{
await _channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Refresh), cancellationToken);
await _channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Refresh), CancellationToken.None);
}
// reload with the full navigation graph before projecting: BuildItem creates the watermark/graphics
@@ -34,11 +34,17 @@ public class DeleteProgramScheduleItemHandler(
List<Playout> playouts = item.ProgramSchedule.Playouts;
dbContext.ProgramScheduleItems.Remove(item);
// bump the optimistic-concurrency token so this config edit rotates other clients' ETags (#253)
item.ProgramSchedule.Version++;
await dbContext.SaveChangesAsync(cancellationToken);
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
foreach (Playout playout in playouts)
{
await channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Refresh), cancellationToken);
await channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Refresh), CancellationToken.None);
}
return Unit.Default;
@@ -44,5 +44,8 @@ public record ReplaceProgramScheduleItem(
string PreferredSubtitleLanguageCode,
ChannelSubtitleMode? SubtitleMode) : IProgramScheduleItemRequest;
public record ReplaceProgramScheduleItems(int ProgramScheduleId, List<ReplaceProgramScheduleItem> Items) : IRequest<
public record ReplaceProgramScheduleItems(
int ProgramScheduleId,
List<ReplaceProgramScheduleItem> Items,
Option<int> ExpectedVersion = default) : IRequest<
Either<BaseError, IEnumerable<ProgramScheduleItemViewModel>>>;
@@ -26,34 +26,103 @@ public class ReplaceProgramScheduleItemsHandler(
Some: async programSchedule =>
{
Validation<BaseError, ProgramSchedule> validation = await Validate(dbContext, request, programSchedule);
return await validation.Apply(ps => PersistItems(dbContext, request, ps, cancellationToken));
// Introduce the optimistic-concurrency check as a standalone Either AFTER the validation
// pipeline (never via Apply) so a PreconditionFailedError survives to a 412 and is not
// flattened to a generic 422 by Join() (issue #253 / api-conventions §7a).
Either<BaseError, ProgramSchedule> validated = LanguageExtensions.ToEither(validation)
.Bind(ps => ps.CheckVersion(request.ExpectedVersion));
return await validated.Match(
Right: ps => PersistItems(dbContext, request, ps, cancellationToken),
Left: error =>
Task.FromResult<Either<BaseError, IEnumerable<ProgramScheduleItemViewModel>>>(error));
},
None: () => Task.FromResult<Either<BaseError, IEnumerable<ProgramScheduleItemViewModel>>>(
new NotFoundError("[ProgramScheduleId] does not exist.")));
}
private async Task<IEnumerable<ProgramScheduleItemViewModel>> PersistItems(
private async Task<Either<BaseError, IEnumerable<ProgramScheduleItemViewModel>>> PersistItems(
TvContext dbContext,
ReplaceProgramScheduleItems request,
ProgramSchedule programSchedule,
CancellationToken cancellationToken)
{
dbContext.RemoveRange(programSchedule.Items);
// reset index starting with zero
programSchedule.Items = [];
// Positional in-place reconcile (rather than delete-and-reinsert): a schedule item owns the
// persisted fill-group/shuffle enumerator state via PlayoutScheduleItemFillGroupIndex, whose
// ProgramScheduleItemId FK is OnDelete(Cascade). Removing and re-inserting every item — as the
// original handler did on every save, including a no-op PUT-back — cascade-deleted that state for
// all playouts using the schedule (#252). Reusing the existing item row for the same-typed slot
// keeps its id, so the cascade never fires and progression survives. The request DTO carries no
// stable item id, so position is the only key available here; true content-aware stable identity
// is deferred to the shared concurrency/round-trip contract in #253.
var orderedItems = request.Items.OrderBy(i => i.Index).ToList();
for (var i = 0; i < orderedItems.Count; i++)
List<ProgramScheduleItem> existingItems = programSchedule.Items.OrderBy(i => i.Index).ToList();
// load the watermark/graphics join rows for the existing items so they can be rebuilt in place
await dbContext.Entry(programSchedule)
.Collection(ps => ps.Items)
.Query()
.Include(i => i.ProgramScheduleItemWatermarks)
.Include(i => i.ProgramScheduleItemGraphicsElements)
.LoadAsync(cancellationToken);
int commonCount = Math.Min(existingItems.Count, orderedItems.Count);
for (var i = 0; i < commonCount; i++)
{
ProgramScheduleItem existing = existingItems[i];
ProgramScheduleItem built = BuildItem(programSchedule, i, orderedItems[i]);
if (existing.GetType() == built.GetType())
{
// same TPT subtype: copy all scalar values in place (BuildItem is the single source of
// item construction, so no field is silently dropped) and rebuild the join rows, keeping
// the item's id — and with it the fill-group index that would otherwise cascade away.
built.Id = existing.Id;
dbContext.Entry(existing).CurrentValues.SetValues(built);
RebuildChildren(existing, orderedItems[i]);
}
else
{
// EF can't change a TPT row's type in place; this slot must be replaced (its fill-group
// index resets, which is acceptable — the item fundamentally changed).
dbContext.Remove(existing);
programSchedule.Items.Add(built);
}
}
// remove surplus existing items
for (int i = commonCount; i < existingItems.Count; i++)
{
dbContext.Remove(existingItems[i]);
}
// add surplus incoming items
for (int i = commonCount; i < orderedItems.Count; i++)
{
programSchedule.Items.Add(BuildItem(programSchedule, i, orderedItems[i]));
}
await dbContext.SaveChangesAsync(cancellationToken);
// Unconditional bump: this handler frequently saves with only CHILD changes and no root-scalar
// change, so without an explicit bump EF would emit no root UPDATE and the concurrency token
// would never fire (nor rotate other clients' ETags). Bumping guarantees both on every save,
// including a no-op same-items PUT-back (issue #253 / api-conventions §7a).
programSchedule.Version++;
// Save through the guard so an EF concurrency failure (a racing writer won between our load and
// save) maps to 412 rather than surfacing as a 500. On failure, propagate the error WITHOUT
// running the post-save reload/enqueue below.
Either<BaseError, Unit> saved = await dbContext.SaveChangesWithConcurrencyGuard(cancellationToken);
if (saved.IsLeft)
{
return saved.Map(_ => (IEnumerable<ProgramScheduleItemViewModel>)[]);
}
// refresh any playouts that use this schedule
// post-commit side effect runs on CancellationToken.None so a late request cancellation
// can't abort it after the commit landed (#254)
foreach (Playout playout in programSchedule.Playouts)
{
await channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Refresh), cancellationToken);
await channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Refresh), CancellationToken.None);
}
// reload with the full navigation graph before projecting: BuildItem creates the watermark/graphics
@@ -68,6 +137,36 @@ public class ReplaceProgramScheduleItemsHandler(
return persisted.Map(ProjectToViewModel).ToList();
}
// Rebuild the watermark/graphics join rows for an item updated in place. Clearing severs the required
// relationship so EF deletes the orphaned join rows (same net effect the delete-and-reinsert path had),
// then re-add from the request. Only the parent item row is preserved — the join rows carry no state.
private static void RebuildChildren(ProgramScheduleItem item, ReplaceProgramScheduleItem request)
{
item.ProgramScheduleItemWatermarks ??= [];
item.ProgramScheduleItemWatermarks.Clear();
foreach (int watermarkId in request.WatermarkIds)
{
item.ProgramScheduleItemWatermarks.Add(
new ProgramScheduleItemWatermark
{
ProgramScheduleItem = item,
WatermarkId = watermarkId
});
}
item.ProgramScheduleItemGraphicsElements ??= [];
item.ProgramScheduleItemGraphicsElements.Clear();
foreach (int graphicsElementId in request.GraphicsElementIds)
{
item.ProgramScheduleItemGraphicsElements.Add(
new ProgramScheduleItemGraphicsElement
{
ProgramScheduleItem = item,
GraphicsElementId = graphicsElementId
});
}
}
private static async Task<Validation<BaseError, ProgramSchedule>> Validate(
TvContext dbContext,
ReplaceProgramScheduleItems request,
@@ -54,6 +54,9 @@ public class UpdateProgramScheduleHandler(
programSchedule.RandomStartPoint = request.RandomStartPoint;
programSchedule.FixedStartTimeBehavior = request.FixedStartTimeBehavior;
// bump the optimistic-concurrency token so this config edit rotates other clients' ETags (#253)
programSchedule.Version++;
await dbContext.SaveChangesAsync();
if (needToRefreshPlayout)
@@ -12,7 +12,8 @@ internal static class Mapper
programSchedule.TreatCollectionsAsShows,
programSchedule.ShuffleScheduleItems,
programSchedule.RandomStartPoint,
programSchedule.FixedStartTimeBehavior);
programSchedule.FixedStartTimeBehavior,
programSchedule.Version);
internal static ProgramScheduleItemViewModel ProjectToViewModel(ProgramScheduleItem programScheduleItem) =>
programScheduleItem switch
@@ -9,4 +9,5 @@ public record ProgramScheduleViewModel(
bool TreatCollectionsAsShows,
bool ShuffleScheduleItems,
bool RandomStartPoint,
FixedStartTimeBehavior FixedStartTimeBehavior);
FixedStartTimeBehavior FixedStartTimeBehavior,
int Version);
@@ -19,7 +19,8 @@ public class GetAllProgramSchedulesHandler(IDbContextFactory<TvContext> dbContex
ps.TreatCollectionsAsShows,
ps.ShuffleScheduleItems,
ps.RandomStartPoint,
ps.FixedStartTimeBehavior))
ps.FixedStartTimeBehavior,
ps.Version))
.ToListAsync(cancellationToken);
}
}
@@ -8,4 +8,5 @@ public record BlockViewModel(
string GroupName,
string Name,
int Minutes,
BlockStopScheduling StopScheduling);
BlockStopScheduling StopScheduling,
int Version);
@@ -9,5 +9,6 @@ public record ReplaceBlockItems(
string Name,
int Minutes,
BlockStopScheduling StopScheduling,
List<ReplaceBlockItem> Items)
List<ReplaceBlockItem> Items,
Option<int> ExpectedVersion = default)
: IRequest<Either<BaseError, Unit>>;
@@ -16,10 +16,21 @@ public class ReplaceBlockItemsHandler(IDbContextFactory<TvContext> dbContextFact
{
await using TvContext dbContext = await dbContextFactory.CreateDbContextAsync(cancellationToken);
Validation<BaseError, Block> validation = await Validate(dbContext, request, cancellationToken);
return await validation.Apply(ps => Persist(dbContext, request, ps, cancellationToken));
// Introduce the optimistic-concurrency check as a standalone Either AFTER the validation
// pipeline (never via Apply) so a PreconditionFailedError survives to a 412 and is not
// flattened to a generic 422 by Join() (issue #253 / api-conventions §7a).
// LanguageExtensions.ToEither joins the Seq<BaseError> to a single BaseError (the native
// Validation.ToEither() would keep Seq and is called explicitly here to avoid that shadow).
Either<BaseError, Block> validated = LanguageExtensions.ToEither(validation)
.Bind(block => block.CheckVersion(request.ExpectedVersion));
return await validated.Match(
Right: block => Persist(dbContext, request, block, cancellationToken),
Left: error => Task.FromResult<Either<BaseError, Unit>>(error));
}
private static async Task<Unit> Persist(
private static async Task<Either<BaseError, Unit>> Persist(
TvContext dbContext,
ReplaceBlockItems request,
Block block,
@@ -33,7 +44,10 @@ public class ReplaceBlockItemsHandler(IDbContextFactory<TvContext> dbContextFact
dbContext.RemoveRange(block.Items);
block.Items = request.Items.Map(i => BuildItem(block, i.Index, i)).ToList();
await dbContext.SaveChangesAsync(cancellationToken);
// Unconditional bump: EF emits the root UPDATE only when a scalar actually differs, so a
// same-value/no-op save would otherwise write no root row and neither fire the concurrency
// token nor rotate other clients' ETags. Bumping guarantees both on every save (issue #253).
block.Version++;
// TODO: refresh any playouts that use this schedule
// foreach (Playout playout in programSchedule.Playouts)
@@ -41,7 +55,9 @@ public class ReplaceBlockItemsHandler(IDbContextFactory<TvContext> dbContextFact
// await _channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Refresh));
// }
return Unit.Default;
// Save through the guard so an EF concurrency failure (a racing writer won between our load
// and save) maps to 412 rather than surfacing as a 500.
return await dbContext.SaveChangesWithConcurrencyGuard(cancellationToken);
}
private static BlockItem BuildItem(Block block, int index, ReplaceBlockItem item)
@@ -6,5 +6,6 @@ public record ReplaceDecoTemplateItems(
int DecoTemplateId,
int DecoTemplateGroupId,
string Name,
List<ReplaceDecoTemplateItem> Items)
List<ReplaceDecoTemplateItem> Items,
Option<int> ExpectedVersion = default)
: IRequest<Either<BaseError, List<DecoTemplateItemViewModel>>>;
@@ -1,12 +1,17 @@
using System.Threading.Channels;
using ErsatzTV.Application.Playouts;
using ErsatzTV.Core;
using ErsatzTV.Core.Domain.Scheduling;
using ErsatzTV.Core.Scheduling;
using ErsatzTV.Infrastructure.Data;
using ErsatzTV.Infrastructure.Extensions;
using Microsoft.EntityFrameworkCore;
namespace ErsatzTV.Application.Scheduling;
public class ReplaceDecoTemplateItemsHandler(IDbContextFactory<TvContext> dbContextFactory)
public class ReplaceDecoTemplateItemsHandler(
IDbContextFactory<TvContext> dbContextFactory,
ChannelWriter<IBackgroundServiceRequest> channel)
: IRequestHandler<ReplaceDecoTemplateItems, Either<BaseError, List<DecoTemplateItemViewModel>>>
{
public async Task<Either<BaseError, List<DecoTemplateItemViewModel>>> Handle(
@@ -15,10 +20,19 @@ public class ReplaceDecoTemplateItemsHandler(IDbContextFactory<TvContext> dbCont
{
await using TvContext dbContext = await dbContextFactory.CreateDbContextAsync(cancellationToken);
Validation<BaseError, DecoTemplate> validation = await Validate(dbContext, request, cancellationToken);
return await validation.Apply(ps => Persist(dbContext, request, ps, cancellationToken));
// Introduce the optimistic-concurrency check as a standalone Either AFTER the validation
// pipeline (never via Apply) so a PreconditionFailedError survives to a 412 and is not
// flattened to a generic 422 by Join() (issue #253 / api-conventions §7a).
Either<BaseError, DecoTemplate> validated = LanguageExtensions.ToEither(validation)
.Bind(decoTemplate => decoTemplate.CheckVersion(request.ExpectedVersion));
return await validated.Match(
Right: decoTemplate => Persist(dbContext, request, decoTemplate, cancellationToken),
Left: error => Task.FromResult<Either<BaseError, List<DecoTemplateItemViewModel>>>(error));
}
private static async Task<List<DecoTemplateItemViewModel>> Persist(
private async Task<Either<BaseError, List<DecoTemplateItemViewModel>>> Persist(
TvContext dbContext,
ReplaceDecoTemplateItems request,
DecoTemplate decoTemplate,
@@ -31,21 +45,49 @@ public class ReplaceDecoTemplateItemsHandler(IDbContextFactory<TvContext> dbCont
decoTemplate.Items = request.Items.Map(i => BuildItem(decoTemplate, i)).ToList();
await dbContext.SaveChangesAsync(cancellationToken);
// Unconditional bump: EF emits the root UPDATE only when a scalar actually differs, so a
// same-value/no-op save would otherwise write no root row and neither fire the concurrency
// token nor rotate other clients' ETags. Bumping guarantees both on every save (issue #253).
decoTemplate.Version++;
// TODO: refresh any playouts that use this schedule
// foreach (Playout playout in programSchedule.Playouts)
// {
// await _channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Refresh));
// }
// Save through the guard so an EF concurrency failure (a racing writer won between our load
// and save) maps to 412 rather than surfacing as a 500.
Either<BaseError, Unit> saved = await dbContext.SaveChangesWithConcurrencyGuard(cancellationToken);
return await saved.Match(
Right: async _ =>
{
// Deco/break/default-filler content is only (re)applied during a Reset build (a Continue keeps
// the frozen DecoDefault filler items), and BlockKey change-detection has no deco dimension, so
// nothing self-heals a deco-template edit — the editor returned 200 but built filler stayed
// stale until a manual Reset (#251). Enqueue a Reset for every playout that references this
// deco template. This whole post-commit invalidation runs with CancellationToken.None (audit
// #22 policy): once the edit is committed, a late request cancellation must not be able to
// abort the affected-playout query OR the enqueue and leave content stale. Only runs after a
// successful save (issue #253) — a 412/422 must not enqueue a Reset for content that was never
// persisted.
List<int> playoutIds = await dbContext.PlayoutTemplates
.Where(pt => pt.DecoTemplateId == decoTemplate.Id)
.Select(pt => pt.PlayoutId)
.Distinct()
.ToListAsync(CancellationToken.None);
await dbContext.Entry(decoTemplate)
.Collection(t => t.Items)
.Query()
.Include(i => i.Deco)
.LoadAsync(cancellationToken);
foreach (int playoutId in playoutIds)
{
await channel.WriteAsync(
new BuildPlayout(playoutId, PlayoutBuildMode.Reset),
CancellationToken.None);
}
return decoTemplate.Items.Map(Mapper.ProjectToViewModel).ToList();
await dbContext.Entry(decoTemplate)
.Collection(t => t.Items)
.Query()
.Include(i => i.Deco)
.LoadAsync(cancellationToken);
return Right<BaseError, List<DecoTemplateItemViewModel>>(
decoTemplate.Items.Map(Mapper.ProjectToViewModel).ToList());
},
Left: error => Task.FromResult(Left<BaseError, List<DecoTemplateItemViewModel>>(error)));
}
private static DecoTemplateItem BuildItem(DecoTemplate decoTemplate, ReplaceDecoTemplateItem item) =>
@@ -2,5 +2,8 @@ using ErsatzTV.Core;
namespace ErsatzTV.Application.Scheduling;
public record ReplacePlayoutTemplateItems(int PlayoutId, List<ReplacePlayoutTemplate> Items)
public record ReplacePlayoutTemplateItems(
int PlayoutId,
List<ReplacePlayoutTemplate> Items,
Option<int> ExpectedVersion = default)
: IRequest<Option<BaseError>>;
@@ -29,6 +29,17 @@ public class ReplacePlayoutTemplateItemsHandler(
foreach (Playout playout in maybePlayout)
{
// Optimistic-concurrency pre-check (issue #253 §7a): reject a stale If-Match with a
// PreconditionFailedError (→ 412) before any mutation. Returned directly so it escapes the
// catch(Exception) below as a 412 rather than being reshaped into a bare 422 (§9/H1).
Either<BaseError, Playout> versionCheck = playout.CheckVersion(request.ExpectedVersion);
if (versionCheck.IsLeft)
{
return versionCheck.Match<Option<BaseError>>(
Left: error => error,
Right: _ => Option<BaseError>.None);
}
PlayoutTemplate[] existing = playout.Templates.ToArray();
List<ReplacePlayoutTemplate> incoming = request.Items;
@@ -88,7 +99,20 @@ public class ReplacePlayoutTemplateItemsHandler(
}
}
await dbContext.SaveChangesAsync(cancellationToken);
// Unconditional bump (issue #253 §7a / M1): #8 writes no root scalar at all (only the
// Templates child collection), so without this the root UPDATE never fires and the token
// never rotates. Bump then save through the guard, which maps a racing
// DbUpdateConcurrencyException to PreconditionFailedError (→ 412) as a return value.
playout.Version++;
Either<BaseError, Unit> saveResult =
await dbContext.SaveChangesWithConcurrencyGuard(cancellationToken);
if (saveResult.IsLeft)
{
return saveResult.Match<Option<BaseError>>(
Left: error => error,
Right: _ => Option<BaseError>.None);
}
}
return Option<BaseError>.None;
@@ -2,5 +2,10 @@ using ErsatzTV.Core;
namespace ErsatzTV.Application.Scheduling;
public record ReplaceTemplateItems(int TemplateGroupId, int TemplateId, string Name, List<ReplaceTemplateItem> Items)
public record ReplaceTemplateItems(
int TemplateGroupId,
int TemplateId,
string Name,
List<ReplaceTemplateItem> Items,
Option<int> ExpectedVersion = default)
: IRequest<Either<BaseError, List<TemplateItemViewModel>>>;
@@ -15,10 +15,19 @@ public class ReplaceTemplateItemsHandler(IDbContextFactory<TvContext> dbContextF
{
await using TvContext dbContext = await dbContextFactory.CreateDbContextAsync(cancellationToken);
Validation<BaseError, Template> validation = await Validate(dbContext, request, cancellationToken);
return await validation.Apply(ps => Persist(dbContext, request, ps, cancellationToken));
// Introduce the optimistic-concurrency check as a standalone Either AFTER the validation
// pipeline (never via Apply) so a PreconditionFailedError survives to a 412 and is not
// flattened to a generic 422 by Join() (issue #253 / api-conventions §7a).
Either<BaseError, Template> validated = LanguageExtensions.ToEither(validation)
.Bind(template => template.CheckVersion(request.ExpectedVersion));
return await validated.Match(
Right: template => Persist(dbContext, request, template, cancellationToken),
Left: error => Task.FromResult<Either<BaseError, List<TemplateItemViewModel>>>(error));
}
private static async Task<List<TemplateItemViewModel>> Persist(
private static async Task<Either<BaseError, List<TemplateItemViewModel>>> Persist(
TvContext dbContext,
ReplaceTemplateItems request,
Template template,
@@ -30,7 +39,10 @@ public class ReplaceTemplateItemsHandler(IDbContextFactory<TvContext> dbContextF
dbContext.RemoveRange(template.Items);
template.Items = request.Items.Map(i => BuildItem(template, i)).ToList();
await dbContext.SaveChangesAsync(cancellationToken);
// Unconditional bump: EF emits the root UPDATE only when a scalar actually differs, so a
// same-value/no-op save would otherwise write no root row and neither fire the concurrency
// token nor rotate other clients' ETags. Bumping guarantees both on every save (issue #253).
template.Version++;
// TODO: refresh any playouts that use this schedule
// foreach (Playout playout in programSchedule.Playouts)
@@ -38,13 +50,22 @@ public class ReplaceTemplateItemsHandler(IDbContextFactory<TvContext> dbContextF
// await _channel.WriteAsync(new BuildPlayout(playout.Id, PlayoutBuildMode.Refresh));
// }
await dbContext.Entry(template)
.Collection(t => t.Items)
.Query()
.Include(i => i.Block)
.LoadAsync(cancellationToken);
// Save through the guard so an EF concurrency failure (a racing writer won between our load
// and save) maps to 412 rather than surfacing as a 500.
Either<BaseError, Unit> saved = await dbContext.SaveChangesWithConcurrencyGuard(cancellationToken);
return await saved.Match(
Right: async _ =>
{
await dbContext.Entry(template)
.Collection(t => t.Items)
.Query()
.Include(i => i.Block)
.LoadAsync(cancellationToken);
return template.Items.Map(Mapper.ProjectToViewModel).ToList();
return Right<BaseError, List<TemplateItemViewModel>>(
template.Items.Map(Mapper.ProjectToViewModel).ToList());
},
Left: error => Task.FromResult(Left<BaseError, List<TemplateItemViewModel>>(error)));
}
private static TemplateItem BuildItem(Template template, ReplaceTemplateItem item) =>
@@ -1,13 +1,18 @@
using System.Threading.Channels;
using ErsatzTV.Application.Playouts;
using ErsatzTV.Core;
using ErsatzTV.Core.Domain;
using ErsatzTV.Core.Domain.Scheduling;
using ErsatzTV.Core.Scheduling;
using ErsatzTV.Infrastructure.Data;
using ErsatzTV.Infrastructure.Extensions;
using Microsoft.EntityFrameworkCore;
namespace ErsatzTV.Application.Scheduling;
public class UpdateDecoHandler(IDbContextFactory<TvContext> dbContextFactory)
public class UpdateDecoHandler(
IDbContextFactory<TvContext> dbContextFactory,
ChannelWriter<IBackgroundServiceRequest> channel)
: IRequestHandler<UpdateDeco, Either<BaseError, Unit>>
{
public async Task<Either<BaseError, Unit>> Handle(UpdateDeco request, CancellationToken cancellationToken)
@@ -17,7 +22,7 @@ public class UpdateDecoHandler(IDbContextFactory<TvContext> dbContextFactory)
return await validation.Apply(ps => ApplyUpdateRequest(dbContext, ps, request, cancellationToken));
}
private static async Task<Unit> ApplyUpdateRequest(
private async Task<Unit> ApplyUpdateRequest(
TvContext dbContext,
Deco existing,
UpdateDeco request,
@@ -177,6 +182,34 @@ public class UpdateDecoHandler(IDbContextFactory<TvContext> dbContextFactory)
await dbContext.SaveChangesAsync(cancellationToken);
// Deco content (break/default filler/etc.) is only (re)applied during a Reset build, and BlockKey
// change-detection has no deco dimension, so editing a deco left already-built filler/break content
// stale until a manual Reset (#251). Enqueue a Reset for every playout that references this deco —
// directly (Playout.DecoId) or through a deco template that includes it. This whole post-commit
// invalidation runs with CancellationToken.None (audit #22 policy): once the edit is committed, a
// late request cancellation must not be able to abort either the affected-playout queries OR the
// enqueue and leave content stale — the entire side effect has to complete.
List<int> directPlayoutIds = await dbContext.Playouts
.Where(p => p.DecoId == request.DecoId)
.Select(p => p.Id)
.ToListAsync(CancellationToken.None);
List<int> decoTemplateIds = await dbContext.DecoTemplateItems
.Where(i => i.DecoId == request.DecoId)
.Select(i => i.DecoTemplateId)
.Distinct()
.ToListAsync(CancellationToken.None);
List<int> templatePlayoutIds = await dbContext.PlayoutTemplates
.Where(pt => pt.DecoTemplateId.HasValue && decoTemplateIds.Contains(pt.DecoTemplateId.Value))
.Select(pt => pt.PlayoutId)
.ToListAsync(CancellationToken.None);
foreach (int playoutId in directPlayoutIds.Concat(templatePlayoutIds).Distinct())
{
await channel.WriteAsync(new BuildPlayout(playoutId, PlayoutBuildMode.Reset), CancellationToken.None);
}
return Unit.Default;
}
@@ -1,10 +1,15 @@
using System.Threading.Channels;
using ErsatzTV.Application.Playouts;
using ErsatzTV.Core;
using ErsatzTV.Core.Scheduling;
using ErsatzTV.Infrastructure.Data;
using Microsoft.EntityFrameworkCore;
namespace ErsatzTV.Application.Scheduling;
public class UpdateDefaultDecoHandler(IDbContextFactory<TvContext> dbContextFactory)
public class UpdateDefaultDecoHandler(
IDbContextFactory<TvContext> dbContextFactory,
ChannelWriter<IBackgroundServiceRequest> channel)
: IRequestHandler<UpdateDefaultDeco, Option<BaseError>>
{
public async Task<Option<BaseError>> Handle(UpdateDefaultDeco request, CancellationToken cancellationToken)
@@ -13,9 +18,29 @@ public class UpdateDefaultDecoHandler(IDbContextFactory<TvContext> dbContextFact
{
await using TvContext dbContext = await dbContextFactory.CreateDbContextAsync(cancellationToken);
await dbContext.Playouts
int updated = await dbContext.Playouts
.Where(p => p.Id == request.PlayoutId)
.ExecuteUpdateAsync(u => u.SetProperty(p => p.DecoId, p => request.DecoId), cancellationToken);
.ExecuteUpdateAsync(
u => u
.SetProperty(p => p.DecoId, p => request.DecoId)
// Bump the shared Playout.Version so any open alternate-schedule/template editor tab
// (which round-trips this same token) detects the deco change and reloads (#253 §7a).
// Bulk ExecuteUpdate bypasses change-tracking, so bump via SetProperty, not Version++.
.SetProperty(p => p.Version, p => p.Version + 1),
cancellationToken);
// Changing a playout's default deco only takes effect on a Reset build (deco content is applied
// during Reset, and BlockKey has no deco dimension to self-heal) — same stale-until-Reset class
// as #251. Enqueue a Reset for the affected playout, but only if a row was actually updated: a
// request for a nonexistent playout must not queue a build for an id that isn't there (keeps the
// enqueued set equal to the affected set). CancellationToken.None so the rebuild isn't dropped
// after the assignment has already been persisted.
if (updated > 0)
{
await channel.WriteAsync(
new BuildPlayout(request.PlayoutId, PlayoutBuildMode.Reset),
CancellationToken.None);
}
return Option<BaseError>.None;
}
@@ -1,3 +1,3 @@
namespace ErsatzTV.Application.Scheduling;
public record DecoTemplateViewModel(int Id, int DecoTemplateGroupId, string GroupName, string Name);
public record DecoTemplateViewModel(int Id, int DecoTemplateGroupId, string GroupName, string Name, int Version);
+4 -3
View File
@@ -32,7 +32,7 @@ internal static class Mapper
new(blockGroup.Id, blockGroup.Name);
internal static BlockViewModel ProjectToViewModel(Block block) =>
new(block.Id, block.BlockGroupId, block.BlockGroup.Name, block.Name, block.Minutes, block.StopScheduling);
new(block.Id, block.BlockGroupId, block.BlockGroup.Name, block.Name, block.Minutes, block.StopScheduling, block.Version);
internal static BlockItemViewModel ProjectToViewModel(BlockItem blockItem) =>
new(
@@ -65,7 +65,7 @@ internal static class Mapper
new(templateGroup.Id, templateGroup.Name, templateGroup.Templates.Count);
internal static TemplateViewModel ProjectToViewModel(Template template) =>
new(template.Id, template.TemplateGroupId, template.TemplateGroup.Name, template.Name);
new(template.Id, template.TemplateGroupId, template.TemplateGroup.Name, template.Name, template.Version);
internal static TemplateItemViewModel ProjectToViewModel(TemplateItem templateItem)
{
@@ -168,7 +168,8 @@ internal static class Mapper
decoTemplate.Id,
decoTemplate.DecoTemplateGroupId,
decoTemplate.DecoTemplateGroup.Name,
decoTemplate.Name);
decoTemplate.Name,
decoTemplate.Version);
}
internal static DecoTemplateItemViewModel ProjectToViewModel(DecoTemplateItem decoTemplateItem)
@@ -1,3 +1,3 @@
namespace ErsatzTV.Application.Scheduling;
public record TemplateViewModel(int Id, int TemplateGroupId, string GroupName, string Name);
public record TemplateViewModel(int Id, int TemplateGroupId, string GroupName, string Name, int Version);
@@ -59,6 +59,9 @@ public class ExtractEmbeddedSubtitlesHandler : ExtractEmbeddedSubtitlesHandlerBa
string ffmpegPath,
CancellationToken cancellationToken)
{
// track exactly the playouts this handler locked so the finally releases only those,
// on every terminal path (cancellation early-return, swallowed cancellation, any exception).
var lockedPlayoutIds = new List<int>();
try
{
bool useEmbeddedSubtitles = await _configElementRepository
@@ -124,7 +127,10 @@ public class ExtractEmbeddedSubtitlesHandler : ExtractEmbeddedSubtitlesHandlerBa
foreach (int playoutId in playoutIdsToCheck)
{
await _entityLocker.LockPlayout(playoutId);
if (await _entityLocker.LockPlayout(playoutId))
{
lockedPlayoutIds.Add(playoutId);
}
}
_logger.LogDebug("Checking playouts {PlayoutIds} for text subtitles to extract", playoutIdsToCheck);
@@ -179,16 +185,18 @@ public class ExtractEmbeddedSubtitlesHandler : ExtractEmbeddedSubtitlesHandlerBa
}
_logger.LogDebug("Done checking playouts {PlayoutIds} for text subtitles to extract", playoutIdsToCheck);
foreach (int playoutId in playoutIdsToCheck)
{
await _entityLocker.UnlockPlayout(playoutId);
}
}
catch (Exception ex) when (ex is TaskCanceledException or OperationCanceledException)
{
// do nothing
}
finally
{
foreach (int playoutId in lockedPlayoutIds)
{
await _entityLocker.UnlockPlayout(playoutId);
}
}
return Option<BaseError>.None;
}
@@ -52,6 +52,10 @@ public class PrepareTroubleshootingPlaybackHandler(
var currentStreamingLevel = loggingLevelSwitches.StreamingLevelSwitch.MinimumLevel;
loggingLevelSwitches.StreamingLevelSwitch.MinimumLevel = LogEventLevel.Debug;
// tracks whether THIS invocation owns the troubleshooting lock, so the catch/failure paths
// never release a lock held by another session (e.g. an exception before we acquire)
var lockAcquired = false;
try
{
using var logContext = LogContext.PushProperty(InMemoryLogService.CorrelationIdKey, request.SessionId);
@@ -64,12 +68,14 @@ public class PrepareTroubleshootingPlaybackHandler(
return BaseError.New("Channel start is required");
}
if (entityLocker.IsTroubleshootingPlaybackLocked())
// acquire atomically: LockTroubleshootingPlayback returns false if another session
// already holds it (no check-then-set race, no double-owner)
if (!entityLocker.LockTroubleshootingPlayback())
{
return BaseError.New("Troubleshooting playback is locked");
}
entityLocker.LockTroubleshootingPlayback();
lockAcquired = true;
statusStore.Reset();
localFileSystem.EnsureFolderExists(FileSystemLayout.TranscodeTroubleshootingFolder);
@@ -108,6 +114,7 @@ public class PrepareTroubleshootingPlaybackHandler(
Option<double>.None),
cancellationToken);
entityLocker.UnlockTroubleshootingPlayback();
lockAcquired = false;
}
return result.Map(model => new PlayoutItemResult(
@@ -119,6 +126,7 @@ public class PrepareTroubleshootingPlaybackHandler(
if (maybeChannel.IsNone)
{
entityLocker.UnlockTroubleshootingPlayback();
lockAcquired = false;
return BaseError.New($"Channel {request.ChannelId} does not exist");
}
}
@@ -129,19 +137,46 @@ public class PrepareTroubleshootingPlaybackHandler(
request,
cancellationToken);
return await validation.Match(
tuple => GetProcess(
dbContext,
request,
tuple.Item1,
tuple.Item2,
tuple.Item3,
tuple.Item4,
cancellationToken),
async tuple =>
{
// acquire only after validation succeeds; if another session already holds the lock,
// report the conflict. Ownership is tracked so the catch never cross-releases.
if (!entityLocker.LockTroubleshootingPlayback())
{
return BaseError.New("Troubleshooting playback is locked");
}
lockAcquired = true;
Either<BaseError, PlayoutItemResult> processResult = await GetProcess(
dbContext,
request,
tuple.Item1,
tuple.Item2,
tuple.Item3,
tuple.Item4,
cancellationToken);
// success hands the lock off to StartTroubleshootingPlayback; any Left releases it now
if (processResult.IsLeft)
{
entityLocker.UnlockTroubleshootingPlayback();
lockAcquired = false;
}
return processResult;
},
error => Task.FromResult<Either<BaseError, PlayoutItemResult>>(error.Join()));
}
catch (Exception ex)
{
entityLocker.UnlockTroubleshootingPlayback();
// only release if THIS invocation still owns the lock; never release a lock held by another
// session when the failure happened before we acquired (e.g. cancellation during validation)
if (lockAcquired)
{
entityLocker.UnlockTroubleshootingPlayback();
}
await mediator.Publish(
new PlaybackTroubleshootingCompletedNotification(-1, ex, Option<double>.None),
cancellationToken);
@@ -163,12 +198,7 @@ public class PrepareTroubleshootingPlaybackHandler(
FFmpegProfile ffmpegProfile,
CancellationToken cancellationToken)
{
if (entityLocker.IsTroubleshootingPlaybackLocked())
{
return BaseError.New("Troubleshooting playback is locked");
}
entityLocker.LockTroubleshootingPlayback();
// the troubleshooting lock is acquired and released by Handle; GetProcess only builds the process
statusStore.Reset();
localFileSystem.EnsureFolderExists(FileSystemLayout.TranscodeTroubleshootingFolder);
@@ -181,6 +211,8 @@ public class PrepareTroubleshootingPlaybackHandler(
string mediaPath = await GetMediaItemPath(dbContext, mediaItem, cancellationToken);
if (string.IsNullOrEmpty(mediaPath))
{
// Handle releases the lock on this Left return, so troubleshooting isn't wedged
// "running" forever for a file that's gone from disk
logger.LogWarning("Media item {MediaItemId} does not exist on disk; cannot troubleshoot.", mediaItem.Id);
return BaseError.New("Media item does not exist on disk");
}
@@ -0,0 +1,11 @@
using ErsatzTV.Core.Domain;
namespace ErsatzTV.Core.Api.MediaSources;
public record LocalLibraryDetailResponseModel(
int Id,
string Name,
LibraryMediaKind MediaKind,
bool IsLocked,
int MediaItemCount,
List<LocalLibraryPathResponseModel> Paths);
@@ -0,0 +1,3 @@
namespace ErsatzTV.Core.Api.MediaSources;
public record LocalLibraryPathResponseModel(int Id, string Path, int MediaItemCount);
@@ -0,0 +1,5 @@
using ErsatzTV.Core.Domain;
namespace ErsatzTV.Core.Api.MediaSources;
public record LocalLibraryResponseModel(int Id, string Name, LibraryMediaKind MediaKind, bool IsLocked);
@@ -0,0 +1,3 @@
namespace ErsatzTV.Core.Api.MediaSources;
public record LocalPathCheckResponseModel(bool Exists);
@@ -0,0 +1,3 @@
namespace ErsatzTV.Core.Api.MediaSources;
public record PathReplacementResponseModel(int Id, string RemotePath, string LocalPath);
@@ -0,0 +1,3 @@
namespace ErsatzTV.Core.Api.MediaSources;
public record PlexPinFlowResponseModel(string AuthUrl);

Some files were not shown because too many files have changed in this diff Show More