- clearSearchResults() advances the search request version, so a search
issued under the previous lock state cannot republish what a relock
just cleared.
- A lock write publishes its store revision only once every touched type
is stamped, so a reload triggered by it cannot read a later type
through its old stamps.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The warm-route hydration reads the cache directly; it now returns nothing
while the lock store withholds everything, like the live reads.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- A hidden-category lookup that lands after a later playback (same
provider id, another playlist) no longer overwrites the newer channel's
category; resolutions are generation- and playlist-checked.
- Withheld Stalker rows are keyed by id, stream_id, movie_id, series_id
or the row's cmd/name, so id-less rows no longer collapse onto one key
and stall paging past locked pages.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- The workspace route resolver awaits ParentalLockService.initialize()
next to the settings load, so no route or catalog activates before the
PIN and lock store are known.
- Every lock write re-reads a failed store before building its edit, so a
recovered store is edited rather than overwritten.
- The deferred hydration reload runs under the publish guard of the
request that deferred it.
- Backup import validates every parental lock entry and rejects a damaged
list instead of erasing the persisted locks on restore.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The Xtream store stays populated after leaving that portal, so its reload
runs on every apply; a locked M3U channel no longer keeps playing behind a
slow database or provider read.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- The Xtream live layout resolves a playing channel's category through
the unfiltered rows when the visible list lacks it (search can play a
hidden category's channel); until that lookup lands the category is
unknown and a relock stops the channel.
- A relock that overtakes the initial hydration now withholds the
category publications too and reloads categories with the content.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- The Xtream category guard no longer runs the item check on category-only
routes (Number(null) is 0), which prompted for the PIN on every unlocked
VOD and series category while the lock was active.
- A lock change during the initial Xtream hydration withholds the rows the
hydration publishes and runs the filtered reload once it has settled,
on every path that marks the content initialized.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- ElectronXtreamDataSource serves no categories, content or search hits
while the lock store withholds everything; its SQLite index may still
carry a stale stamp.
- setupPin decides whether to persist the switch from the settings value
before the PIN is stored, since enabled follows hasPin while the switch
is unknown.
- A lock store recovered by a later read marks its playlists stale so the
index is re-derived, a persisted entry must carry all three lists, and a
stale Stalker search page is dropped before touching the withheld-id
bookkeeping.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- A write that removes a playlist's last lock clears the SQLite index
first and drops the store key afterwards, so an interruption between the
two can only leave a state the startup reconcile repairs toward locked.
- A PIN hash read failure is distinct from an absent PIN: the session stays
locked and every PIN-protected step re-reads it first.
- Backup export awaits parental lock initialization and refuses to run
while the lock store is not readable, since an absent lock field means
"no opinion" on restore.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- A relock now fails closed immediately: the selected detail is stepped
off against the lock store, the catalog lists and stored search results
are emptied, and the filtered reloads publish only while the captured
lock version is still current.
- Backups carry M3U lock titles verbatim (exact dedup), since the locks
match group titles exactly.
- A relock-timeout write that fails reverts the in-memory value and shows
the settings save-failure snackbar.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- The Electron mirror of the feature switch is awaited; a mirror that
cannot be written undoes the settings write, so a reload never starts
from a mirror that disagrees with the persisted switch.
- A failed multi-type re-stamp rolls the store back AND re-stamps every
touched type from it, since earlier types may already carry the new
locks; a failed rollback keeps the playlist stale (fail-closed).
- A persisted lock store whose nested entries are not what writeLocks
produces is a failed read, not an empty store.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- The lock store is readable only once the SQLite index has been re-derived
from it, and a re-stamp that keeps failing keeps the session fail-closed,
so catalog reads can never serve rows stamped unlocked by a stale index.
- While everything is withheld, Stalker rows without a genre are withheld
as well (the store filter and the renderer predicate).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- The window before the initial lock store read settles now withholds
everything, like an unreadable store: settings can report the feature as
on before the locks are known.
- The store commits before the SQLite index re-stamp; a failed re-stamp
now rolls the store back, a failed rollback re-stamps on the next
access, and every launch re-derives the index from the store.
- Xtream category/content reloads fail closed: a rejected reload empties
the affected lists (content types drop back to idle) instead of keeping
rows read under the previous lock state.
- Enabling/disabling the feature persists through one guarded path that
undoes the in-memory switch and skips the Electron mirror on a failed
settings write.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- A Stalker search page issued before a relock was filtered with the
pre-relock withheld set and could still be applied after it; the
staleness check now includes the parental lock version.
- A lock store payload that does not parse or is not an object is a
failed read (everything withheld until it reads again), no longer an
empty store.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- A lock store that cannot be read is no longer treated as empty: while
the lock is active every category is withheld (renderer predicates and
set-based filters alike) until the PIN is entered or the store reads
again, and writes are refused meanwhile so an empty in-memory store can
never wipe the persisted locks. The lock set now lives in its own
ParentalLockLockStore service.
- The M3U group dialog's lock write is awaited and a failed save is
reported in a snackbar instead of being silently dropped.
- The Electron categories.locked re-stamp clears and re-locks inside one
transaction, so a failed restamp keeps the previous index.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- Unreadable settings (IndexedDB load failure) left the feature switch at
its default and announced "unlocked" to the main process. A stored PIN
now stands in for the switch, and without one nothing is announced, so
the worker keeps its mirrored locked default.
- Lock applies run one at a time and abandon superseded results; the
Electron data source keys its in-flight share by lock version so a
relock can never reuse an unlock refresh's unfiltered rows.
- The stored in-portal Xtream search is re-run on a lock change.
- Stalker live/radio selections are judged by tv_genre_id, and both live
layouts drop the playback of a channel whose category became withheld.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The Xtream, Stalker and M3U category editors are gated by the PIN only when
they open; an idle relock left them on screen listing locked names with a
lock-rewriting Save. Each now closes itself when the session relocks.
ParentalLockEnforcementService also judges the selected Xtream/Stalker
item by its own category: a detail opened from All, recently added or
search has no selected category to vanish with, so it stayed open after a
relock.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- The idle timer follows the unlocked transition instead of `active`, so the
session that just enabled the lock still locks itself later.
- Stalker search closes an open detail whose genre became withheld on
relock and advances by itself past pages made only of locked rows (only
while they add ids the list has not seen).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- The Xtream detail guard hydrates the PWA session cache before judging an
item on a cold navigation and fails closed when the catalog cannot place
the item.
- The dedicated Stalker search route filters withheld genres, re-fires on
lock changes, judges paging on the raw page and restarts from page 1 on a
lock flip.
- The Xtream category dialog loads its lock candidates through the
capability-selected data source; the PWA source now lists its raw
categories with lock flags, so locks can be configured there too.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- Detail routes check the item's own category: a locked movie or series
paired with an unlocked category id in the URL is still refused.
- `requestUnlock()` awaits the settings load before it can answer "not
active", so a slow startup cannot open a management dialog unguarded.
- `SETTINGS_UPDATE` only persists the `parentalLockEnabled` mirror and
releases the worker on switch-off; it no longer re-locks the worker on
every ordinary settings save under a renderer that shows "unlocked".
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- Xtream and Stalker `:categoryId` routes carry a parental-lock guard: a
locked category reached by URL prompts for the PIN and redirects to the
section root on refusal (Electron row ids are mapped to provider ids).
- PWA search filters withheld categories like the catalog reads.
- Electron warm-cache detection confirms an empty, lock-filtered read with
the unfiltered existence check instead of refetching from the provider.
- A Stalker lock flip past page 1 drops withheld rows at once and restarts
the list from page 1 instead of appending onto stale pages.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- The M3U group dialog opens only after the PIN, like the Xtream and Stalker
dialogs: it lists locked group names and can rewrite the locks.
- Change PIN and Disable always verify the stored hash, even while the
session is unlocked, so an app left unlocked cannot lose its lock.
- Stalker paging judges progress on the raw portal page: withheld ids the
list has not seen count as progress, a page made only of locked rows
requests the next one itself, and the VOD total is reduced by withheld
ids so the grid stops asking once every visible row is in.
- Parental lock contract linked from the agent context map after the
guidance reorganization; bridge helpers split out to stay under the
file-size cap.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Resolves the settings-store defaults split, the electron-conf key list and
the guidance reorganization (CLAUDE.md now imports AGENTS.md; the parental
lock contract is linked from the agent context map instead).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* test(electron): reap Windows process trees and retain smoke diagnostics
* test(electron): require confirmed process exit before relaunch
* test(electron): retain process handle after application exit
* test(electron): bind captured processes to application instances
* docs(tmdb): replace catalog titles in matching examples with stand-ins
The Cyrillic and Arabic examples that document title folding were taken
straight from a user's own portal catalog while the folding bugs were being
diagnosed, and they spread from there into comments, fixtures, assertions and
the architecture doc. A public repository is not the place for someone's
viewing inventory, and the TMDB ids pinned alongside them identify the exact
shows.
Swap in stand-ins that reproduce the property under test rather than the
title: a word whose "й" folds away, one whose "ё" does, a two-word title
carrying a season marker, an Arabic phrase whose initial hamza decomposes into
a separate word. Every replacement was run through the real `normalizeTitle`
and `cleanTitleForSearch` before being written, so the folded keys, the wire
queries and the cache lookup keys are the same shape as before — "Лейка" and
"Леика" still meet on one key while staying two different searches, and
"AR| أمثلة تجريبية" still keys as a hamza split into a space. Real TMDB ids
become synthetic ones. One channel fixture that read as a film title becomes a
plain channel name.
Deliberately left alone: the leading-tag rule's "Akira | 1988" / "Момо | Momo"
examples in `title-normalization.util.ts`. Those are not an inventory — they
are the evidence for a regex decision measured over 1.27M catalog titles, and
inventing replacements would document a measurement that never happened.
No release note: comments, fixtures and docs only, with no behaviour change.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* docs(tmdb): label the folding stand-ins as illustrative, not as history
Review caught that the substitution left synthetic titles inside sentences
that assert observed fact: the architecture doc claimed a particular folded
query "returns zero results while `Лейка` returns the show" and named the
stand-ins as the titles that were searched, missed and negatively cached, and
several comments read the same way. The titles never existed, so the reading
is wrong in the one direction that matters — a future reader debugging a
regression would take them for production evidence.
Keep the claim that is actually established, which is a class-level one:
under the old single-form design every Cyrillic title carrying "й"/"ё" and
every Arabic title carrying a hamza form was searched folded, missed, and
cached as missing for the negative TTL. Present the strings themselves as
illustrative stand-ins chosen to fold the same way. The verified invariant —
what NFD does to those letters, and what the two tiers therefore produce — is
unchanged and still assertable, because it is a property of the text rather
than of any title.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* docs(tmdb): finish the sweep — a missed cache key and a last observed-fact claim
Two spots the first pass left behind:
- `tmdb-search-cache-cleanup.spec.ts` kept a catalog-derived lookup key
verbatim. Only the TMDB id beside it had been swapped, because the sweep
matched the title in its display casing and the key stores it lowercased —
so one item of the inventory stayed in the repository, twice.
- `SearchTitleVariant`'s doc comment still asserted that one specific folded
string finds nothing on TMDB. State the mechanism instead: folding rewrites
the letters the search matches on, so the folded form finds nothing there.
`content-search.util.sqlite.spec.ts` gets a channel fixture that no longer
reads as a film title.
`search-text-fold.util.spec.ts` is deliberately left alone. Its "Ёлки" is a
common noun sitting in a Unicode corpus beside "Amélie", "İnşaat" and "Ά",
not an inventory entry — and its rows are precomposed/decomposed PAIRS
(U+0401 against U+0415 U+0308). A textual substitution rewrites only the
composed half and silently turns the pair into two different words; doing it
failed that spec, which is what a fixture encoding an invisible property is
for.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* docs(search): finish the fold fixtures by rewriting both halves of the pair
The last two occurrences lived in the Unicode fold corpus, which an earlier
attempt left alone after a textual substitution failed the spec. The reason it
failed is the point: one row is a precomposed/decomposed PAIR — U+0401 against
U+0415 U+0308 — asserting that both spellings fold to one string. Replacing
the visible text rewrites only the composed half and silently turns the pair
into two different words, which is not a thing a reader or a regex can see.
Rewrite both halves by code point instead, keeping the decomposed half
decomposed: U+0415 U+0308 + the new stem. Verified by decoding every quoted
string in the file afterwards, and the spec passes (534/534).
A repository-wide sweep now finds no occurrence of the replaced titles in
either normalization form.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
Metadata is looked up in the app's own language, so an unrelated older foreign
series can come back under exactly the same localized name as a recent
local-language one. `pickConfidentMatch` admitted the older row through the
series "premiered earlier" tolerance — portals report the running season's
year while TMDB reports the premiere — and then let `pickMostPopular` decide
across every admitted candidate, discarding the year evidence that had just
admitted them. The better-known show won on votes, and the newer series
rendered its poster, cast, genres and rating.
Rank admitted candidates by year evidence first (`yearEvidenceTier`: the
provider's exact year, then a year off by one, then the series tolerance) and
let popularity break ties only inside the strongest tier any candidate
reached. The tolerance stays — three of eight real lookups from one install
depend on it — but it is a last resort, not an equal. Measured over 400
Cyrillic series titles sampled from a real catalog, 20 normalized keys had a
same-titled older series and 16 of those were the more popular row.
The mirror case is accepted knowingly: a long-running show whose stated
season year happens to BE another same-titled show's premiere year now
resolves to the newer show. Only the older show's season air dates could
separate the two and a search response does not carry them, while that shape
needs three coincidences at once against one that needs none.
Search cache keys move to `|v4` with a matching startup cleanup, because a
positive row naming the wrong show stays fresh for 30 days.
Merged with `Build on windows x64` red: the checked-in Windows Embedded MPV
runtime pin points at an upstream release whose retention expired, so that
job fails repository-wide on a cold cache. Unrelated to this change; tracked
separately.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The "current programme" under each Xtream Live TV row was written only on
scroll-into-view, on a new EPG result, or on an EPG offset change. Nothing
re-evaluated it as wall-clock time passed, so once a programme ended the row
stayed on it until the category was left and re-entered. The progress bar
under the row never moved either.
The rows on screen now re-check themselves once a minute. A programme still on
air only has its progress bar advanced, at no request cost; once it ends the
row is re-picked, and only a programme on air or upcoming may replace it, so a
guide that has run out can never walk the row backwards. An exhausted cache is
dropped and refetched at most once per cache lifetime per channel, while an
empty answer from the provider is left alone. What is on screen stays there
until a replacement arrives, so a refreshing row never blanks out.
A programme occupies [start, stop) in every comparison, the visible slice is
read from the viewport rather than remembered, and the two services behind
this are root-provided because a live layout mounts the channel list more than
once over a single EPG queue: EpgRefillLimiter is the floor on refetching an
exhausted guide, keyed by playlist since stream ids are provider-local, and
EpgRefreshCoordinator owns the one timer and merges every mounted list's
request, because the queue is latest-wins and separate timers would cancel
each other on exactly the boundaries that matter.
Contract: docs/architecture/m3u-playlist-module.md ("Xtream channel-row
programme refresh").
Fixes#767. Supersedes #1610.
Co-Authored-By: Justin Willhite <5132924+thejdubb02@users.noreply.github.com>
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Live channels in Favorites and Recently viewed now offer "Open in <playlist>"
for Stalker portals, the counterpart of the VOD "View in portal" action. The
chip in the programme panel and the channel's context menu jump to the channel
inside its portal's Live TV, with its genre selected and the channel playing.
Radio stays hidden: it is a separate legacy-paged section with no
open-on-arrival contract.
The handoff refuses to guess. The store records which genre the rendered ITV
list belongs to, so the deferred play waits for a list that can actually serve
the channel instead of inferring it from array identity. An id claimed by one
channel as its `id` and another as its `stream_id` is ambiguous, so neither is
played and the user still lands in the right genre. The handoff is abandoned
when the user changes genre, search, portal or section first.
Two pre-existing defects surfaced during review and are fixed here:
- A blank provider id shadowed a valid one. `a ?? b` keeps an empty string, so
a channel with a blank `stream_id` was stored with no identity at all and
could not be selected, played or found again. Six writers had that shape and
three private copies of the skip-the-blank rule; all now go through one
`firstNonBlankStalkerId` helper.
- Route-session readiness could publish before the store held the portal's
row. The constructor starts one sync and the first navigation starts another,
and the second skipped the bootstrap because the playlist id was claimed
before the awaits that install it. Arrivals are now serialized, the id is
claimed only after the store write resolves, and only the newest sync
publishes readiness.
Both fixes carry regression tests that fail on the old behavior.
Xtream and Stalker live cards on the dashboard carried no XMLTV key, so the
rails never asked anything for them and showed only the LIVE chip. Their
programme now comes from the portal, one card at a time and only once the card
is on screen.
- `lib-dashboard-rail` reports the cards inside its viewport through an
IntersectionObserver rooted at the track; `DashboardLiveEpgPresenter` unions
them with the pinned hero row and hands the set to
`DashboardPortalLiveEpgPresenter`.
- `DashboardPortalLiveEpgService` runs the bounded queue — two requests in
flight, 200 ms apart, one card each — through
`StreamResolverService.loadEpgForItems`, publishing every answer the moment
it lands, so the page never waits and a slow portal delays no other card. A
card scrolled past before its turn is never requested.
- A programme is trusted for 60 s, an empty answer for 30 s (the resolver
reports a dead portal and a guide-less channel identically), and a
completion captures both the display offset and the EPG source revision,
requeueing itself when either moved.
- The presenter hands its wanted set back on destroy. Desktop only: the shared
collection resolver is gated on the local XMLTV bridge.
- A shimmer placeholder shows only before a card's first portal answer; M3U
cards keep the batched XMLTV lookup.
Every dashboard title carried a "Xtream · Series" / "Stalker · Movie" subtitle. Provider kind and content kind are the app's own taxonomy, not a property of the title, and they are identical on every card in a rail — so the one line that could tell two cards apart said nothing.
The hero now shows the source name alone, through `playlistDisplayLabel` (a stored playlist name is routinely the pasted URL with credentials, or a MAC). Continue Watching cards show what actually varies: the "S1·E5" chip plus "12 min left". Favorites keep the title alone, Recently Added keeps the source name, and a meta row with nothing in it is no longer rendered.
Stalker shows filed under Movies had no badge, no progress and no resume. An embedded-VOD row announces its episodes through a `series[]` array and carries no `is_series` flag, so `extractStalkerItemType` reports `movie` on purpose — the item must keep routing to the VOD catalog — while its progress is a set of episode positions keyed by the parent id, which the dashboard was looking up as a single `vod` row and never finding.
Split the two questions: `PortalActivityItem.watch_kind` records the progress model when it differs from the routing type, and every reader that has to choose goes through `resolvePortalActivityWatchKind` instead of `type`. That makes the resume handoff reachable for Stalker, so wire it through `STALKER_SERIES_RESUME_TARGET`: consumed once after the series positions are read, hydrating a lazy Ministra season first with a bounded two-attempt retry, and playing nothing at all when the position read failed rather than restarting the episode from zero.
Also fixes the global-recent route template, which bound `[seriesResume]` only on its Xtream branch.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* refactor(collections): split the unified collection page and live tab
Both files sat far above the workspace's 400-line hard maximum and were
only green because `tools/eslint/max-lines-baseline.mjs` exempted them.
No behavior change: every template binding, public signal and handler the
components exposed still resolves the same way, and the two baseline
entries are gone.
`unified-collection-page.component.ts` (966 → 398 lines) keeps the view
surface and delegates:
- `unified-collection-data.service.ts` — component-provided; owns the
rows, the favorite uid set, the loading/reload indicators and every
mutation, plus `loadedRequest` (the PR #1636 invariant, now read through
one `mutationRequest` computed instead of two ad-hoc fallbacks)
- `unified-collection-load.ts` — the reload key and the load effect
- `unified-collection-scope.ts` — the This-playlist/All-playlists toggle
- `unified-collection-content-type.ts` — the Live/Movies/Series tab
- `unified-collection-history.ts` — the `window.history.state` view entry
- `unified-collection-detail-state.ts` / `-detail-navigation.ts` — the
inline detail and where an item this route cannot render goes instead
- `unified-collection-clear-action.ts`, `-labels.ts`, `-favorites-sort.ts`
`unified-live-tab.component.ts` (1051 → 426 lines) continues the pattern
its siblings already established:
- `unified-live-selection.ts` (+ `-selection-generation.ts`) — activating
a row while the mounted player stays alive
- `unified-live-selection-view.ts` — what the selection implies about the
source and the player surface
- `unified-live-epg-view.ts`, `unified-live-epg-map.ts` — the EPG panel
and the rail's now-playing map
- `unified-live-catchup.ts` — the timeshift override
- `unified-live-recording-metadata.ts`, `unified-live-channel-rows.ts`
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
* refactor(collections): move the collection data service into data-access
Codex review on PR #1642: `UnifiedCollectionDataService` loads and persists
favorites/recent rows, which CLAUDE.md places in
`@iptvnator/portal/shared/data-access`, not in the `type:ui` project that
renders them — and the dialog-scoped `SourceCleanupService` already sets that
precedent for a component-provided stateful collection service. It was also
the only non-root `@Injectable()` in `portal-shared-ui`.
`collection-reload-indicator.ts` moves with it: `type:data-access` may not
depend on `type:ui`, and the indicator is the service's own loading state
machine rather than a view. Both are exported from the data-access collection
barrel, so the boundary is now lint-enforced instead of conventional.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
The `pull_request: closed` cleanup is the fast path, not a guarantee: GitHub
does not run that workflow when the head ref is already gone at event time,
which is what Dependabot does when it supersedes one of its own PRs. 15
`test-pr-<n>` drafts had been orphaned that way, 13 of them Dependabot's.
Add a daily scheduled (and manually dispatchable) sweep to the same workflow.
It lists every draft tagged `^test-pr-[0-9]+$`, asks GitHub for that PR's live
state, and deletes only when the PR reports closed. It fails closed: a PR
lookup error leaves the draft untouched, a failed release listing fails the job
rather than sweeping a short list, and only a confirmed HTTP 404 excuses a
failed delete — `gh api` exits 1 for every failure alike, so the re-check reads
the response status instead of the exit code.
Workflow permissions drop to `contents: read`; the event job keeps
`actions: write` + `contents: write`, the sweep takes only `contents: write`.
No new actions. Docs: new "Rolling test drafts" section in
docs/architecture/release-pipeline.md.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Switching the unified Favorites/Recent page between "This playlist" and "All playlists" kept the old items on screen with no feedback until the new query resolved. The skeleton cannot be reused there: it unmounts a playing channel and drops focus from the toggle.
Add a separate reload state that keeps content mounted: after a 180 ms grace period an indeterminate progress bar overlays the header separator, the content region carries aria-busy, and the grid or the live tab's channel rail dims — the player never dims. Fast IndexedDB/SQLite answers show nothing. Only the latest request settles the state.
Also bind Clear and drag reorder to the request that loaded the rows still on screen (`loadedRequest`) instead of `effectiveScope()`, which moves ahead the moment the toggle is clicked.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Bump astro 7.2.4 → 7.2.10 (critical, website build) and retarget the pinned
pnpm overrides for the transitive alerts: js-yaml → 4.3.2 (the one runtime
path, via electron-updater), smol-toml → 1.7.1 (new key for nx's exact 1.6.1
pin), svgo → 4.1.0 (new key for astro's 4.0.2 resolution) and hono → 4.13.5.
Every target stays inside its parent's declared range except nx's exact
smol-toml pin, which is now recorded as the deliberate exception in
docs/architecture/dependency-security-overrides.md.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(xtream): keep a live-channel handoff alive until its playlist catalog is loaded
Arriving at /workspace/xtreams/:id/live from another route with
openXtreamLiveItemId in history state silently did nothing: the Xtream
shell mounts the live layout after its session bootstrap, i.e. after the
arrival's NavigationEnd, so the layout's NavigationEnd subscription never
saw it. When the layout was reused instead (playlist switch), the shared
store still held the previous playlist's catalog at NavigationEnd and the
"not in liveStreams" verdict dropped the pending id.
Read the state once at mount as well, carry openXtreamLivePlaylistId in
the navigation state, and treat a miss as final only once currentPlaylist
is the requested playlist and isContentInitialized is true.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* feat(collections): open a favorite or recent live channel inside its playlist
Live channels watched from Favorites / Recently viewed had no visible
playlist and no way to jump there, unlike movies and series with "View in
portal". Add the live counterpart:
- getLiveCollectionPlaylistNavigation() resolves the channel inside its
playlist (Xtream via the live layout's auto-open state, M3U via
openM3uChannelUrl on the player's all view); Stalker resolves to null
until its ITV layout gets an open-on-arrival contract, so nothing is
shown there instead of landing on the section root.
- app-open-in-playlist-chip, projected into the EPG timeline / list-view
toolbar through a new [epgToolbarAction] slot beside the channel name,
visible in the collapsed state too.
- "Open in <playlist>" entry in the channel row context menu, which also
covers radio rows and rows that are not playing.
Both label with playlistDisplayLabel and reuse PORTALS.VIEW_IN_PORTAL_TOOLTIP.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(xtream): only match an auto-open channel against the requested playlist's catalog
Review finding (Greptile/Codex P1): the playlist check ran only on a miss,
so a colliding provider-local xtream_id in the previous playlist's catalog
was accepted, played the wrong channel and consumed the handoff. Check the
playlist before consulting the catalog at all.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* feat(portal): season thumbnails in the season dropdown + PR #1628 follow-ups
Follow-ups to the season posters shipped in #1628:
- The >6-seasons dropdown (`SeasonTabsComponent`) now carries a 28×42
season thumbnail at the start of each menu row that has a poster and in
the closed trigger for the selected season, fed by a new `seasonPosters`
input from the season container and the fullscreen episode panel. Rows
without a poster get no placeholder, a failed image is dropped, and the
pill row stays text-only as the design review decided.
- The fullscreen season strip's episode count uses its own
`PORTALS.EPISODE_COUNT_ONE/OTHER` keys instead of borrowing the download
manager's; all 18 locales filled through the i18n merger from their
existing `DOWNLOADS.EPISODE_COUNT_*` translations.
- The Stalker mock's serve targets no longer pin `PORT` (an nx:run-commands
`env` entry overrides the shell), and `main.ts` resolves `PORT`, then the
Playwright-side `MOCK_PORT` alias, then 3210 — so `MOCK_PORT=3310` now
relocates the whole E2E run. The Xtream mock honours `XTREAM_MOCK_PORT`
the same way.
- `resolveAutoSelectedSeason` gets a direct spec covering every branch.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(xtream-mock): mint marketing asset URLs on the port the server bound
Greptile P1 on #1633: the listener honoured `XTREAM_MOCK_PORT`, but
`marketingAssetOrigin()` still read `PORT` alone, so a run relocated only
through the alias sent every poster/backdrop/logo/episode URL to 3211.
One resolver (`resolveXtreamMockPortString` in `mock-port.ts`: `PORT`,
then `XTREAM_MOCK_PORT`, then 3211) now feeds the environment parser, the
marketing asset origin and the demo-guide origin fallback. A spec pins the
precedence and that `marketingAssetUrl` follows the bound port.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* fix(portals): apply the IPv6->IPv4 fallback budget in the Electron process
The 2500 ms happy-eyeballs attempt timeout from #1404 only ever ran in the
web backend. The Electron main process and its playlist-refresh and EPG
workers kept Node's 250 ms default, so a dual-stack panel hostname behind a
VPN or a slow link failed every connection attempt in a row and tripped the
host connectivity guard. The module now lives in `@iptvnator/shared/host-health`
and every Node isolate that opens connections applies it.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(portals): stop treating a slow panel as a dead one in the host guard
axios raises the same ECONNABORTED whether the SYN went unanswered or the
panel accepted the connection and then thought for longer than the request
budget. Two such timeouts opened the breaker and every request to the panel
was refused for 30 s with "portal is not responding" — the shape behind the
"connection keeps dropping" reports on 0.23 and nightly.
Both transports now report whether the TCP connection was established
(`onConnect`: Electron through a per-request observed agent instead of the
shared keep-alive globalAgent, the web backend through the transport that owns
the ClientRequest), and `classifyHostRequestFailure(error, { connected })`
downgrades a host-level code observed after the handshake to inconclusive.
Redirect attribution keeps precedence. A host that never accepts the
connection trips the guard exactly as before.
The Xtream mock gains a `silent:silent` scenario whose detail actions accept
and never answer, plus a real-socket regression spec for the guard.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(portals): let an accepted connection clear the host-failure streak
Review finding: an unanswered SYN, then an accepted-but-slow timeout, then
another unanswered SYN still reached the two-failure threshold, because the
middle request was merely not counted. An accepted TCP connection is the
reachability the guard measures, so it now reads as `responded` and clears
the streak like an HTTP response would. Regression coverage for the mixed
sequence on one flapping loopback origin (Electron) and through the proxy
route (web backend).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(portals): credit an accepted connection when it happens, not when the request settles
Review findings. A request that connected and then hung for 30 s cleared,
on its eventual timeout, the failures later requests had recorded while it
waited — reopening a host that had just died on evidence older than theirs.
The connect hook now reports the connection the moment it fires through a
new `HostConnectivityGuard.reportConnected`, which clears the failure streak
but closes no open or half-open breaker (the trial keeps its slot until it
settles), and the settled timeout is inconclusive.
Electron also skips the socket observer while an environment proxy
(`http_proxy` / `https_proxy` / `all_proxy`) applies to the request: through
a proxy the socket connects to the proxy, whose handshake proves nothing
about the portal, so those requests keep the pre-observer behaviour.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(portals): decide the proxy exemption with axios' own resolution
Review findings. The hand-rolled environment check ignored `no_proxy`, so a
LAN portal exempted from the proxy lost its connect observer and slow
requests to it still tripped the breaker; it also read the variables with
`??`, letting an empty lowercase one mask a populated uppercase one that
axios would honour. The decision now calls `proxy-from-env`'s
`getProxyForUrl`, the same pinned package axios' http adapter uses,
declared as a direct dependency so the packaged app carries it.
The validated-axios spec clears and restores every proxy variable around each
case, so a runner that exports a proxy cannot change what the cases prove.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* fix(settings): make the update-channel check honest and keep release notes fresh
Settings → About mixed two commit models: the channel select applied on
Save while "Check again" ran immediately against the still-saved channel,
so picking Nightly and checking reported "latest version" for Stable under
a select reading Nightly. The status now carries a badge naming the channel
the verdict describes; while the select shows an unsaved other channel the
verdict is dimmed, a hint names both channels, and the check button becomes
"Save and check for <channel> updates", which submits the form — the main
process already re-checks when the saved channel changes. A download in
flight or finished belongs to the previous channel and keeps the plain
check.
"What's new" for a nightly published after the app started failed with a
raw IPC error: each release catalog is a process-lifetime snapshot and a
fully paged list never re-read GitHub. findIndex now reloads the catalog
once when a version is missing, and a newly found update drops every
catalog. The dialog recognises the not-found rejection through the shared
marker text, explains it with the version, and links to the channel's
release list; other failures keep their reason under a localized headline.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(updater): serialize readers of one release catalog
findIndex may rebuild the shared release array while another reader of
the same catalog still holds an index into the old one and dereferences
it after paging further. Every reader now runs through the catalog's
runExclusive queue (getReleaseNotes and the manual-update check), so a
reload can no longer pull the list out from under a navigation.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(settings): attribute a kept download to the channel it was found on
setChannel keeps a download that is running or finished when the saved
channel changes, but status.channel already names the new channel, so the
About badge attributed a Stable download to Nightly and the pending hint
vanished. Every check now stamps status.verdictChannel with the channel it
ran on and setChannel leaves it alone; the badge names that channel, and
while it differs from the saved one a hint says the shown update came from
the other channel and the saved one has not been checked yet.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* refactor(updater): move release-notes reads out of AppUpdateService
AppUpdateReleaseCatalogs (app-update-release-notes.ts) now owns the
per-channel catalogs and both reads the updater performs on them: release
notes with previous/next paging and the newest release for the
manual-install fallback. The service only delegates, shrinking from 610
to 508 lines instead of growing.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* docs(updater): name verdictChannel as the badge's source
The About paragraph still said the badge reads status.channel while the
paragraph below it and the code use status.verdictChannel.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(settings): show the release list, not the earlier release, after a paging failure
A failed Previous/Next keeps the earlier notes for navigation while the
body shows the error, so the dialog's action offered the earlier release
instead of the channel release list. The error is now checked first.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(updater): do not reload the catalog before falling back to latest
A read that falls back to the newest release on a miss (the installed
version's notes, e.g. an unpublished local build) paged the whole list
twice: findIndex reloaded on the miss before index 0 was selected. The
reload is now opt-in per call and off on that path.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Series detail pages now render the selected season's poster as a season
cover next to the season tabs and description, and the fullscreen episode
panel shows the same poster as a season strip above its tabs.
Resolution is TMDB-first, like the show artwork merge: the lazy season
enrichment stores `/tv/{id}/season/{n}` `poster_path` as a w342 URL in
`tmdb_season_posters` (Xtream) or `StalkerSeriesTmdbSeasonsService.posters()`
(Stalker), under the same write-only-if-changed convergence guard as the
season overview. Xtream falls back to the provider's `seasons[].cover_big`/
`cover` when it is an http(s) URL other than the show poster, because panels
repeat the show poster on every season. Stalker is TMDB-only.
The cover column is not rendered for one-season items, seasons without a
poster, or a failed image, so every fallback is today's markup. It is sized
by a new `--season-cover-width` token (96/120/144px per Settings.coverSize).
The hero poster never follows the season.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The two "channel scrolling keeps focus after selection" cases (and the
Xtream "channel focus and separate scrollbar" cases, which press Tab the
same way) failed deterministically on Playwright WebKit while passing on
Chromium and Firefox. Playwright's WebKit emulates Safari's default
keyboard preference, under which plain Tab visits only text fields and
links: from the focused channel pane the key landed on the sidebar search
field instead of the first row button. Option+Tab reaches the button and
then the favorite action in the same DOM order Chromium's Tab follows, so
the app's focus contract (ChannelScrollFocusDirective) is intact and this
is Safari's Tab semantics, not an app bug.
Add a `pressTab` E2E helper that presses Alt+Tab only on webkit and keeps
the literal Tab / Shift+Tab on chromium and firefox, use it at the four
Tab presses toward buttons, and note Safari's behaviour in the keyboard
scrolling contract.
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
`apps/electron-backend/build-embedded-mpv.js` resolves the compiler with
`require.resolve('node-gyp/bin/node-gyp.js')` and its comment claimed the
package was "a declared devDependency" — it never was. node-gyp reached the
tree only as a transitive of `@electron/rebuild`, in pnpm's hidden hoist
(`node_modules/.pnpm/node_modules`). pnpm's `.bin` shims export that
directory on NODE_PATH, which is why `pnpm nx …`, `pnpm run build:backend`
and CI kept building the addon, while a plain
`node apps/electron-backend/build-embedded-mpv.js` on a clean install failed
with "Unable to resolve node-gyp".
Declare node-gyp 12.4.0 (the version already in the lockfile store) as a root
devDependency so the resolution no longer depends on a shim implementation
detail, correct the stale comment, and record the contract in the
embedded-MPV architecture doc plus the Agent Bootstrap notes.
No packaged-build change: the no-runtime skip and its
`embedded-mpv-unavailable.txt` marker are untouched.
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
* feat(shell): zoom shortcuts on Windows and Linux (#1109)
Cmd/Ctrl and +/−/0 (numpad included) now zoom the app on every platform.
Windows/Linux run without a menu (`setMenu(null)`), so the shortcuts are a
renderer key binding in `WorkspaceKeyboardShortcutsService` calling a new
synchronous, preload-local bridge method `adjustZoomLevel`, which steps the
frame-bound temporary level through `webFrame.setZoomLevel` — never a
main-process `webContents.setZoomLevel`, whose per-URL entry the app's
`file://` path routing resets. Step and limits live in
`libs/shared/interfaces` (`stepZoomLevel`: 0.5 per press like Electron's
zoomIn/zoomOut roles, clamped to levels −4…6). On macOS the renderer sees the
key before the application menu, and `preventDefault()` keeps the menu role
from stepping a second time (Electron only performs the menu key equivalent
in its unhandled-keyboard-event hook).
Persistence is unchanged: the main process still reads the live level back
on close, quit and reload. The zoom E2E now drives the real shortcuts (in,
out, numpad, reset). Help dialog entries added and translated for all
locales; contract updated in docs/architecture/workspace-shell.md.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(shell): never step a stored out-of-range zoom level against the request
A level persisted before the shortcuts existed (the macOS menu roles never
clamped, and the store restores any finite level) was clamped BEFORE the
step, so the first zoom-in from level 7 rendered smaller. Step from the raw
level instead: a press further out leaves an out-of-range level where it is,
a press back in lands on the limit.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* docs(shell): state the zoom bridge's return contract precisely
`adjustZoomLevel` steps by `stepZoomLevel`'s rules; a stored out-of-range
level is never moved against the request, so the returned level is not
itself guaranteed to be within `ZOOM_LEVEL_MIN..ZOOM_LEVEL_MAX`.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* docs(release): add a release note for the zoom shortcuts
The Release note gate requires an added `.changes/*.md` for runtime changes;
the shortcuts are a user-visible feature of their own, so they get their own
note and the persistence note stays about persistence.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
The title search sent the folded comparison key (NFD + strip marks + lowercase) as the TMDB query, so every Russian title with й/ё ("Фейк (10 серий)" → "феик") and every Arabic title with hamza missed and was cached as missing for 7 days. Search candidates now carry a provider-spelled wire query beside the folded comparison key; variants are deduplicated and cached per attempted variant by the lowercased query; the search lookup key moves to |v3 and startup deletes the retired |v2 rows under their own app_state marker. Verified on 1.99M live catalog titles (folded key byte-identical). Real-SQLite cleanup coverage runs inside Electron across skipped, previous, pre-person, fresh and repeated startups.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Movie and series detail pages showed two arrow_back controls while the
inline player was open: the shell's sticky arrow (added in #1576 so Back
survives scrolling) meant "Close player" in watch state, while the
now-playing bar carried a second arrow that meant route-level Back — the
same icon with two meanings, next to a "Close player" button that
duplicated the first.
The shell's sticky arrow is now route-level Back in browse and watch
alike, and the bar carries no arrow of its own. Closing the player is
the bar's "Close player" button and Escape, which still unwinds one
level (close, then back). Hosts without a browse Back target (M3U,
downloads) render no arrow in either state.
Unit specs for the shell and the inline player cover the new contract;
the Electron and web E2E helpers that pressed Back from watch are
updated, and the M3U flow closes the player through the bar's button.
Docs, the mirrored CLAUDE.md/AGENTS.md paragraph and a release note
follow the change.
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
The packaged renderer is index.html over file:// with path routing, so
after in-app navigation the document URL names a path with no file behind
it. A main-process reload (macOS View > Reload, DevTools) failed with
ERR_FILE_NOT_FOUND and stranded the window on Chromium's error page; a
renderer-initiated reload (the settings unsaved-changes guard's confirmed
location.reload()) was cancelled by the will-navigate trust check and
silently did nothing.
Both legs now re-load the packaged index with the route in a restoreRoute
query parameter, which main.ts restores with history.replaceState before
Angular bootstraps. The did-fail-load recovery is deferred to the error
page's dom-ready: a load issued from inside the failure event yields a
document that never receives animation frames and never paints.
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Series playing in fullscreen get the same slide-in side panel the live channel list has, with season tabs and the episode list: rest the mouse on the left edge, click it, or press C; pick an episode and it plays inline without leaving fullscreen.
- Panel contract: `FullscreenChannelPanelHost` gains optional `panelSearchEnabled` and `panelKind`; the template context gains `open`. Pointer/keyboard rules and the four live providers are unchanged.
- Series host: `PortalInlinePlayerComponent` provides the token through `createEpisodePanelHost()` and stamps `app-fullscreen-episode-panel` (SeasonTabsComponent over rows with TMDB still or numeral tile, label, runtime, clamped overview, progress, watched check, now-playing marker; playing row centred on open). Episode clicks reuse the Up Next rail's inline path; season tab clicks reach the hosts' `onSeasonSelected` (Xtream TMDB season enrichment, Stalker lazy VOD load with a Retry row after a failed request).
- Gates: `Settings.fullscreenChannelPanel` (label now covers both lists in all locales), episode content only, native-view Embedded MPV withheld by the view, external players excluded.
- Inline-series e2e moved to `xtream-series-playback.e2e.ts` with shared Xtream helpers in a fixture; adds a fullscreen episode switch through the panel.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(playback): make the fullscreen channel panel discoverable and calmer to close
The left-edge hot zone was an invisible 28px strip with nothing telling the
user where the list lived, so "hovering the left side" rarely reached it while
`C` always worked. Mouse movement over the stage now reveals a slim edge hint
tab (CSS chevron, fades after 2.5 s idle, lit while the pointer rests in the
zone), the zone grows to 40px (48px coarse), and a click or tap on the edge
opens at once without the dwell.
Closing follows the pointer more honestly: the mouse-leave grace grows from
420 ms to 1 s, and it applies only once the pointer has engaged with the
panel, so a `C`-opened list no longer closes while the mouse merely roams over
the video under the user's typing. Clicks inside the panel never close it.
Found on the way: closing with Escape while the mouse still rests on the edge
reopened the panel 160 ms later, because Chromium synthesizes a `pointerenter`
on the hot zone the aside slid away from. An explicit close now re-arms the
zone only on the next real pointer move.
Verified live over CDP on HTML5, Video.js, ArtPlayer and Embedded MPV
frame-copy; the hover mechanism itself was sound on every engine.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
* fix(playback): open the panel only on a primary click begun on the edge
Review follow-up. The hot zone opened on every pointerup, so a drag released
over the edge, a right or middle click and a pen barrel button all opened the
panel; now pointerdown records the primary pointer and pointerup must match
it (a leave or cancel forgets the press). The synthetic pointerenter that
follows an explicit close no longer arms the hint either: the edge stays
clear until the next real pointer move, which arms it and starts the dwell.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
---------
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
