Six small UX additions on the Playlist Auto-Sync manager:
- Sidebar gets a "Filter playlists…" search input. Re-renders only
the schedule panel on input so focus is preserved while typing.
- Scheduled cards show a red `!` badge + red border tint when the
last three pipeline runs failed (yellow `⚠` if at least one of the
last few failed). Surfaces chronically broken schedules visually
instead of leaving them indistinguishable from healthy ones.
- Run History tab title shows a red error count badge when there are
failed runs in the loaded window.
- Run History tab body gains All / Errors / Completed filter pills
with per-bucket counts.
- Load-more button at the bottom of the history tab pulls another
50 entries (capped at 500).
- "Run pipeline again" button in the expanded detail of each history
card re-triggers that playlist's pipeline directly.
Also dropped the "Discovered: completed" result pill — `tracks_discovered`
in the result payload is a status string, not a count, and the same
data is already in the before/after stats grid above.
Three problems wrapped into one pass on the Playlist Auto-Sync surface:
1. Visual: the manager modal had its own vibe (radial gradient, pill
tabs, sky-blue chrome) that didn't line up with the rest of the
app. Reworked the modal shell, KPI summary, live pipeline monitor,
tab bar, schedule board sidebar, and column cards to use the
standard SoulSync patterns — gradient `#1a1a1a → #121212`,
accent-tinted 1px border, 20px radius, underline tabs, dense dark
card pattern that Automations + Library pages already use. Modal
now uses near-full screen so there's room for the schedule board
without horizontal scroll pain. Run history cards followed the
same path: slim horizontal row mirroring `.automation-card` plus
an expanded detail that mirrors the Automations run-history modal
(stats-grid + facts row + result pills + log section).
2. Hang: the previous SQL fix for the run-history "in library" count
added `COLLATE NOCASE` on the join columns of `tracks` and
`artists`. SQLite can't use `idx_artists_name` or `idx_tracks_title`
when the comparison collation doesn't match the column collation,
so the join did a full table scan per mirrored playlist track.
~18s per playlist × 30 playlists = `/api/mirrored-playlists` hung
indefinitely and the modal stayed at "Loading schedule…" forever.
Switched the join back to case-sensitive equality (~6ms per
playlist, 3000× faster). Spotify names canonicalize to the same
form as library imports so the recall loss is in the rounding
error of pure case-only mismatches.
3. Slowness: even after the hang fix, each modal open spent ~1.5s
gathering per-playlist status counts. The endpoint looped
`get_mirrored_playlist_status_counts(playlist_id)` per row, which
opened a fresh SQLite connection + PRAGMA setup each time. Added
`get_all_mirrored_playlist_status_counts(profile_id)` which
returns counts for every mirrored playlist owned by the active
profile in 4 batched `GROUP BY` queries over a single connection.
Modal load dropped to ~280ms.
Also fixed: `tracks.artist` reference in `get_mirrored_playlist_status_counts`
that never worked since the schema went relational — the query threw
"no such column", got swallowed by the try/except, and the in-library
count silently defaulted to 0 on every playlist. Rewired to join
through `artists`.
`get_mirrored_playlist_status_counts` (single-playlist) kept for
callers that still want it, but the modal endpoint uses the batched
version.
Render playlist pipeline history rows with DOM construction instead of per-card HTML strings so malformed payload values cannot collapse the row into an empty bordered shell. Add a visible per-row render fallback for bad history records while preserving the expandable detail panel.
Flatten playlist pipeline history cards into a stable header, flow, preview, and detail structure so the run history does not collapse into broken line rows. Keep explicit expand bindings and preserve the richer detail payload rendering.
Bind run history card expand interactions after the modal renders instead of relying on inline handlers, and reshape the run cards into a clearer two-row layout with controlled metadata, preview chips, and roomier expanded detail padding.
Make playlist pipeline run history cards clickable and keyboard-accessible, with expanded detail sections for summary stats, timeline, before/after snapshots, result payload fields, and future run logs. Refresh the card styling so the expanded state remains responsive inside the Auto-Sync modal.
Bring Auto-Sync automation and run-history cards closer to the Automations page pattern with status dots, flow chips, compact metadata, and denser run preview details.
Normalize sparse playlist pipeline history rows before rendering and add a visible fallback for empty entries so the Run History tab cannot collapse into unreadable divider lines.
Render playlist pipeline history as visible run cards with fallback summaries, preview chips, metadata, Details controls, and an explicit empty result message for sparse payloads.
Give placed playlist cards a dedicated content wrapper, full-width action row, wider board columns, and defensive wrapping so titles, timing badges, and Run now controls stay inside card bounds.
Compact the inactive pipeline monitor, widen schedule columns, increase board gutters, and rework scheduled playlist cards so Run now and remove actions no longer crowd playlist text.
Persist per-playlist pipeline run snapshots from the shared playlist pipeline, expose a history API, and upgrade the Auto-Sync modal with live pipeline monitoring, Run now controls, and a runs-style history tab.
Update the dashboard Quick Actions tile to use the shared accent color variables for lane glow, icon chips, borders, hover states, and keyboard focus while keeping the three-destination launcher responsive.
Replace the old Tools CTA with a unified three-lane dashboard launcher for Tools, Auto-Sync, and Automations, using restrained glass/accent styling and responsive stacked behavior.
Cin review: no JS tests covered the autoSync* helpers, so the
timezone fix shipped without a regression test in the layer where the
bug actually lived. New `tests/static/test_auto_sync.mjs` runs under
`node --test` (built-in runner, no extra deps) and pins:
- `autoSyncTriggerForHours` / `autoSyncHoursFromTrigger` round-trip
for 1h, 4h, 12h, 24h, 48h, 168h. Catches off-by-one in the day vs
hour conversion that backs the schedule board's drag-drop.
- `autoSyncBucketLabel` / `autoSyncIntervalLabel` formatting +
pluralization.
- `autoSyncSourceLabel` known + unknown + falsy.
- Predicates (`autoSyncCanSchedulePlaylist`,
`autoSyncIsPipelineAutomation`, `autoSyncPlaylistIdFromAutomation`,
`autoSyncIsScheduleOwned`) including the `owned_by`-flag /
legacy-name-prefix split from the previous commit.
- `buildAutoSyncScheduleState` partitions board-owned schedules from
custom pipelines correctly.
- `autoSyncNextRunLabel` parses the naive UTC timestamp as UTC, not
local — exactly the regression that took an hour to diagnose this
session. Includes a past-time check ("due now") and a multi-day
case.
- `getMirroredSourceRef` source_ref/description-URL/source_playlist_id
resolution order.
Cross-realm note: vm-sandbox return values fail `deepStrictEqual`
against host-realm objects even when shape matches, so a small
`deepShapeEqual` helper round-trips through JSON for structural
comparison. The `_autoParseUTC` stub mirrors the real implementation
in stats-automations.js so the timezone test exercises both files end
to end.
`tests/test_auto_sync_js.py` is the pytest shim — shells out to
`node --test` and surfaces failures inline. Skips cleanly when node
isn't on PATH or is older than 22, matching the existing
discover-section-controller test pattern.
Also updated SPLIT_MODULES in tests/test_script_split_integrity.py to
include the new auto-sync.js — the onclick-coverage check was failing
because `openAutoSyncScheduleModal` (referenced from index.html via
the Sync page button) now lives in a module the integrity scanner
wasn't searching.
39 new JS test cases, all green via `node --test` and via the pytest
wrapper.
Cin review: stats-automations.js had ~600 lines of new Auto-Sync code
piled into an already-large shared file. Moved into its own module:
- New webui/static/auto-sync.js holds:
- Schedule board state (`AUTO_SYNC_BUCKETS`, `_autoSyncScheduleState`,
`_autoSyncActiveTab`, `mirroredPipelinePollers`)
- All `autoSync*` functions (trigger conversion, render panels,
drag/drop, save/unschedule, schedule modal lifecycle)
- Mirrored-playlist pipeline helpers (`runMirroredPlaylistPipeline`,
`pollMirroredPipelineStatus`, `applyMirroredPipelineState`,
`parseMirroredPipelineResponse`, `editMirroredSourceRef`,
`getMirroredSourceRef`)
- index.html loads auto-sync.js immediately after stats-automations.js
so the older `renderMirroredCard` path can keep reaching these
globals through the window namespace.
- stats-automations.js drops 567 lines and gains a one-line breadcrumb
pointing at the new file.
No behavior changes — every function moved verbatim. Globals stay in
the same window namespace, so the still-resident `renderMirroredCard`
keeps calling `runMirroredPlaylistPipeline` / `editMirroredSourceRef`
/ `mirroredPipelinePollers` exactly as before.
Both files pass `node --check`. Full Python suite still green.
The Auto-Sync schedule board was detecting its own automations by
checking `group_name === 'Playlist Auto-Sync' || name.startsWith('Auto-Sync:')`.
That's fragile — renaming the row from the Automations page silently
hands ownership back to the read-only Automation Pipelines tab and the
board stops managing it.
This commit replaces the string convention with an explicit
`automations.owned_by` TEXT column:
- Migration `_add_automation_owned_by_column` adds the column and
backfills `'auto_sync'` for existing rows that match the legacy
`group_name`/`name`-prefix pattern, so users running the migration
don't lose their schedules.
- `database.create_automation` and `database.update_automation` accept
`owned_by` (the latter via its `allowed` kwarg set).
- `core/automation/api.py` forwards `owned_by` on both POST and PUT.
Missing field is left as None, preserving today's behavior for every
caller that doesn't opt in.
- The Auto-Sync schedule board posts `owned_by: 'auto_sync'` and the
detection helper now prefers that signal, falling back to the legacy
name/group convention so any hand-rolled rows still show up.
Tests: three new cases in `tests/automation/test_automation_api.py`
covering create-with-owned-by, create-without (defaults to None), and
update set/clear. The fake DB grew the matching kwarg.
Three changes folded into one perf+cleanup pass:
1. Indexed fast path for the per-artist pool fetch. The previous
`search_tracks(artist=name)` call hit
`unidecode_lower(artists.name) LIKE ?`, a function-in-WHERE that
can't use `idx_artists_name`. New `MusicDatabase.get_artist_tracks_indexed`
does a two-step lookup: exact-name match (indexed) plus a
case-insensitive fallback, then `tracks WHERE artist_id IN (...)`
via `idx_tracks_artist_id`. Drops per-artist fetch from seconds to
milliseconds for the common case. The sync helper falls back to
the old LIKE-based `search_tracks` only when the indexed lookup
finds nothing, preserving diacritic recall and `tracks.track_artist`
feature-artist matches with zero regression.
2. Public text-normalization helper. Lifted the body of
`MusicDatabase._normalize_for_comparison` into
`core/text/normalize.py:normalize_for_comparison` so callers outside
the database layer (matching engine, sync pool, future import-side
comparisons) don't reach across the module boundary into a
leading-underscore "private" method. The DB method now delegates,
so existing internal call sites stay untouched. Sync's lazy pool
now imports the public helper.
3. Artist-name walker extracted. `_artist_name` at module level in
`services/sync_service.py` replaces two near-identical inline
str-or-dict-or-fallback walkers (one in `sync_playlist`, one in
`_find_track_in_media_server`). Returns `''` for None instead of
the literal string `'None'`.
Plus three small tidies from the same review:
- `_POOL_FETCH_LIMIT = 10000` constant in place of the literal at the
pool-fetch call site.
- Trimmed the verbose docstring + comment block on the pool helper.
- Set-intersection predicate for the trigger-shape reset in
`core/automation/api.py` instead of a two-line `or` chain.
Also removed the duplicate `_get_active_media_client()` call at
sync_service.py:212/214 — pre-existing wart that was sitting in the
same block I was editing.
Tests: 21 new tests across `tests/database/`, `tests/sync/`, and
`tests/text/`, plus updates to the existing pool tests to cover the
new fast/fallback split. Full suite stays green (3953 passing).
`tests/automation/test_automation_api.py` gains three update_automation
tests covering the schedule-shape reset:
- trigger_config change blanks next_run
- trigger_type change blanks next_run
- non-trigger field (name) leaves next_run alone
`tests/sync/test_sync_candidate_pool.py` is new — nine tests for the
lazy artist track pool in PlaylistSyncService:
- candidate_pool=None disables pooling and skips the DB call
- first lookup for an artist fetches and caches
- second lookup for the same artist reuses the cache (zero DB calls)
- empty result still cached so the next call short-circuits without SQL
- defensive None return coerced to []
- search_tracks exception returns None and does NOT poison the cache
- pool key is normalized so casing variants share a single fetch
- different artists get separate pool entries
- server_source plumbing survives the trip to search_tracks
All assertions go through fakes / MagicMock — no real DB, no
web_server.py import, no AST-parsing.
Syncing a playlist where most tracks weren't in the library was burning
~30 SQL queries per missed track. `services/sync_service.py` walked
each Spotify track through `check_track_exists` with no
`candidate_tracks`, hitting the legacy title-variation × artist-variation
grid in `database/music_database.py:6041-6069` for every miss. The
`sync_match_cache` only covered matches, so misses re-paid the full
lookup cost every sync. A 30-track playlist with a 30% match rate
(Discover Weekly was 9/30 in the test run) was taking ~4m14s, almost
entirely in the matching phase.
`check_track_exists` already accepts a `candidate_tracks` kwarg that
skips the SQL widening and scores against an in-memory list (the
batched path at `music_database.py:6031`, originally added for artist
discography iteration). The sync service just wasn't using it.
This commit wires that path in via a lazy per-artist pool:
- `sync_playlist` creates an empty `candidate_pool` dict and passes it
to each `_find_track_in_media_server` call.
- `_get_or_fetch_artist_candidates` runs SQL for an artist only on the
first track that needs them — playlists where every track is already
in `sync_match_cache` pay zero pool cost (no upfront delay).
- Subsequent misses for the same artist hit the memoized list and skip
the per-variation SQL grid.
- Artists with no library tracks still get a cached empty list, which
triggers the batched path's instant short-circuit instead of falling
into the SQL widening.
- Any pool fetch failure returns None so the caller falls through to
the original per-track SQL loop, so the worst case is the old
behavior, never a regression.
On a 30-track / 25-unique-artist playlist with a cold cache the SQL
fan-out drops from ~900 queries to ~25; with a warm cache it drops to
zero (no pool fetches at all). Applies to every entry point that goes
through `sync_playlist`: manual sync, auto-sync schedules, the
`playlist_pipeline` automation action, and the Sync All button.
The Playlist Auto-Sync schedule board was showing "next in 8h" on every
card regardless of the configured interval. Root cause: backend stores
next_run as a naive UTC string ("2026-05-25 05:00:00") and the new
auto-sync renderer was parsing it with plain `new Date(...)`, which
treats unmarked timestamps as local time. On Pacific time that offsets
the displayed countdown by ~8 hours. Auto-Sync now routes through the
existing `_autoParseUTC` helper that the rest of the Automations page
already uses, so countdowns line up with the wall clock.
A separate correctness fix in the automation update API: when a PUT
changes `trigger_type` or `trigger_config`, the stored `next_run` is
now blanked before the engine reschedules. Previously the scheduler's
restart-survival path would preserve a stale future timestamp from the
prior interval, so dragging a playlist from the 8h column to the 1h
column kept firing at the old 8h mark. Boot-time restart behavior is
unchanged — only user-driven schedule changes reset the clock.
Modal restyle: the Auto-Sync manager's hardcoded sky-blue palette is
replaced with `var(--accent-rgb)` everywhere so the modal honors the
user's chosen accent color. Tinted glow on the modal border, tabbed
header active state, scheduled-playlist chips, scrollbars, and a new
drag-over highlight on columns all follow the accent theme. The
column drag-over state is wired through new ondragleave handling so
the highlight clears reliably when leaving a column.
Constrain Auto-Sync columns inside the modal with per-column vertical scrolling, add responsive layouts for narrower and shorter viewports, and separate schedule interval labels from next-run timing.
Also prevents unsupported mirrored sources from being scheduled into the playlist pipeline while still showing them as unavailable in the sidebar.
Upgrade the Auto-Sync modal into a tabbed manager with a richer schedule board and a separate read-only Automation Pipelines tab for existing playlist_pipeline automations.
Add a Sync-page Auto-Sync manager with source-grouped mirrored playlists, interval columns, and drag/drop scheduling backed by playlist_pipeline automations.
Schedules created by the board are editable there, while existing custom pipeline automations are shown as locked automation-managed entries.
Rename the manual pipeline button to Auto-Sync and make non-JSON endpoint failures show an actionable restart message instead of a raw JSON parse error.
Expose playlist-native run and status endpoints that reuse the shared mirrored playlist pipeline engine while routing progress into playlist UI state.
Add a Run Pipeline action to mirrored playlist cards and modals with live status polling, and make the shared pipeline lock atomic for manual and scheduled callers.
Extract the all-in-one mirrored playlist lifecycle into core/playlists/pipeline.py so automation becomes a thin adapter. Preserve the existing automation action and behavior while making the pipeline reusable by future direct playlist UI controls.
Return normalized source_ref metadata from mirrored playlist APIs so the UI no longer has to infer editable refresh links from description fields. Accept Spotify embed URLs during source-ref repair and add coverage for source-ref health reporting.
Centralize mirrored playlist source reference normalization so edited links and IDs are stored consistently. Preserve URL-backed refresh refs, surface missing-source refresh failures, count background sync failures in pipeline summaries, and retry guarded automation skips after a short delay instead of losing a scheduled run. Add focused coverage for source refs, mirrored playlist source updates, refresh failures, and guarded retry behavior.
Delay torrent and usenet album-bundle dispatch until missing-track analysis confirms there is work to do, matching the Soulseek album flow and avoiding release downloads for already-owned albums.
Clear private album-bundle staging state when a release-level source intentionally falls back to per-track mode so workers can use the normal staging/search path instead of an empty private bundle directory.
Verified by user: focused downloads master tests passed, 2 passed.
Stop appending persistent download history once the unified downloads payload reaches the requested limit. This keeps the Downloads tab history tail bounded even if the history provider returns more rows than requested, while preserving existing live-task total behavior.
Use Zustand's public shallow selector export so the import page resolves correctly under the installed Zustand 5 package. This fixes the Vite boot overlay without changing import workflow behavior.
Keep album-bundle staging from replacing known per-track album numbers with the filename parser's default when staged files do not expose a real track number. Carry staging tag numbers through the cache, fall back to task metadata for private release staging, and cap hybrid album batches to one worker when Soulseek is first in the source order.
- pass release metadata through album search normalization
- surface release format, country, label, and disambiguation in React import cards
- add coverage for search normalization and import route rendering
- describe the implemented nested /import route structure
- document the route-local workflow store and stable draft state
- update testing, risk, and cleanup notes to match the current code
- keep the /import loader from turning transient staging fetch failures into route errors
- keep cached auto-import status and results visible during refetch failures
- show inline notices only when there is no stale data to fall back to
- add regression coverage for staging, status, and results failure paths
- move the shared badge primitive and styling out of the form component module
- keep primary-button badge styling working via a data-slot hook instead of a form-local class
- update the import pages and primitive tests to consume the new home for Badge
- thread primary_source through album and track search payloads while keeping per-result source on the returned rows
- reuse the shared Notice primitive for fallback and error messaging in the import pages
- update the import route tests and shell route smoke coverage for the new behavior
- fold Show and Notice into a single primitives module with one shared stylesheet
- keep the primitives barrel export intact while shrinking the folder footprint
- consolidate the primitive tests into one combined suite
- Keep option buttons transparent by default and subtle on hover
- Use the ghost style for inactive auto-import filters so the active one stands out
- Keep OptionButton aligned with the existing button variant API
- keep only semantic data attributes on the form primitives
- move variant styling into nested CSS module selectors
- preserve the existing visual treatment while simplifying the component layer
- replace direct fetch stubs with shared MSW handlers
- keep fetch spying only for request assertions
- cover the shell prefetch with an issues counts handler
- let the singles action buttons use the default size again
- remove redundant type="button" props from import controls
- switch import page conditional classes to clsx object notation
- drop route-test assertions that pinned compact auto-import sizes
