Discovery's "Suggested albums" surfaced singles and EPs because
_derive_albums filtered with is_core_release (Album/Single/EP). Add a
discovery-only _album_kind_ok gate that, when discover.albumsOnly (new
config, default true), requires primary_type == "Album" with no secondary
types. is_core_release is untouched — the monitor/scan still use it.
Exposed discover.albumsOnly in the discover config route + a Discovery
settings toggle. Worker + web tests added.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds a "Running behind a VPN (Gluetun)" README section: the whole-stack
network_mode: service:gluetun setup (DATABASE_URL @db -> @localhost, ports moved
to gluetun, FIREWALL_INPUT_PORTS + FIREWALL_OUTBOUND_SUBNETS), a compose sketch,
and caveats (Qobuz geo-sensitivity, the IPv6 sysctl not applying but harmless, MB
per-IP rate limit covered by the cache, bind mounts unaffected).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A manually-imported album had no release-group MBID, so it showed the ◉
placeholder until a Scan matched it. Now the import resolves the release group
(searchReleaseGroup, via the shared MB cache) and stores it on the new
LibraryItem.rgMbid column (migration add_library_rg_mbid). The library route
prefers LibraryItem.rgMbid, falling back to the MonitoredRelease join for
scanned items — so a manual album gets cover art immediately. Best-effort: a
no-match / MB outage just leaves rgMbid null.
web 164 tests, tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
- Cover art now loads through GET /api/cover/[rgMbid], which fetches from the
Cover Art Archive once, caches the image on disk (LYRA_COVER_CACHE, default
/tmp/lyra-covers), and serves it with a 30-day Cache-Control — so the browser
stops re-fetching it on every Library / Recently-Pressed render. Missing covers
get a negative marker (re-checked daily). CoverArt points at the endpoint.
- Recently Pressed dedupes by artist+album (re-requesting/re-downloading an album
left multiple completed Requests showing the same album repeatedly); newest of
each is kept.
web 164 tests, tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The real cause of the failed 584MB album upload: Next.js caps the request body
at 10MB for any route that passes through middleware, and the import route went
through the auth gate. Exclude /api/library/import from the middleware matcher so
the body streams uncapped to busboy, and authenticate it inline instead (new
lib/auth.isAuthed reads + verifies the session cookie from the request).
web 161 tests, tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Manual import used request.formData(), which buffers the whole body and caps out
around ~15MB — a real FLAC album is hundreds of MB (Hollywood's Bleeding is
584MB), so it failed with "expected multipart form data". Now the multipart body
is STREAMED with busboy: each file pipes straight to a hidden .import-<uuid>
staging dir on the library filesystem, then the dir is atomically renamed into
place. No full-body buffering, no size cap. Cleaned up on any failure; guards
(400 missing fields/audio, 409 exists) unchanged. web 161 tests.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A real folder drag silently did nothing for a user (the logic is fine — a
simulated drop works — so a real drop was either erroring silently or not
firing). Harden it and give a guaranteed alternative:
- onDrop now captures the entries synchronously, wraps the async walk in
try/catch, and TOASTS on any failure or empty result (no more silent no-op).
- dragEnter preventDefault + dragover dropEffect="copy" for reliable drop-target
registration.
- New "Choose a folder" affordance (a webkitdirectory <input>) that sidesteps
drag entirely — pick the album folder and it collects the files + prefills
artist/album/year from the folder name. Plus "Choose files".
tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Manual import only accepted individual audio files — dragging a folder yielded
nothing (dataTransfer.files doesn't recurse). Now the drop zone walks the entry
tree via webkitGetAsEntry, so you can drag a whole album folder (the natural
shape of a ripped CD) and it collects the audio + cover inside. When a single
folder is dropped, artist/album/year are prefilled by parsing its name
("Artist - Album (Year)"), only filling fields you haven't typed into. Clicking
still opens the multi-file picker. No API change — same files reach the route.
web 161 tests, tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Second Library Management feature. An "Add album" button on the Library page
opens a modal: drag-drop (or pick) the audio files (+ optional cover image),
enter artist/album/year, and Lyra writes the album into the library.
- POST /api/library/import (auth-gated, multipart): sanitizes the folder name
like the worker's safe_name, assembles into a sibling ".importing" temp dir
then renames into place (never a half-written album), writes a cover.jpg from
any image, and records a source="manual" LibraryItem with the on-disk
trackNames + a rough quality class from the extension (a later Scan re-probes /
MB-resolves for cover art + hi-res detection). Guards: 400 (missing
artist/album or no audio), 409 (folder exists / already in library).
- Library page: ImportAlbum drop-zone modal wired to refresh the grid.
web 161 tests (write-to-disk + trackNames + guards + filename sanitization).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Also: DELETE now prunes an emptied artist folder (best-effort).
First Library Management feature. On the Library album modal, "Delete album"
(two-step confirm) removes the album:
- deletes the folder on disk (guarded to only ever remove a path strictly inside
the library root, LYRA_LIBRARY_ROOT, default /music),
- drops the LibraryItem,
- sets the matching MonitoredRelease monitored=false + currentQualityClass=null
so the monitor won't immediately re-download it (re-Want later to restore).
Requires the web container to see the library, so docker-compose now bind-mounts
MUSIC_DIR at /music (rw) for web too. New DELETE /api/library/[id] (auth-gated,
path-guarded). web 157 tests (delete happy-path + 404 + outside-root guard).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Completeness only compared the adapter's self-reported track_count to expected,
so a silently-skipped track (0-byte placeholder — see the Qobuz 24K Magic bug)
imported as a "complete" album. Now the verify also counts the NON-EMPTY audio
actually staged and fails "incomplete download" when fewer than expected are
present (guarded with `staged and ...` so fake adapters that stage nothing keep
using the count path). A real broken download now goes needs_attention (or falls
through to another source) instead of importing a dead track.
Full per-track duration matching against MB recording lengths (the rest of #16)
remains a larger follow-up. worker 224 tests / 7-skip.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
docker-compose only health-checked db. Add:
- web: an unauthenticated GET /api/health (excluded from the auth middleware) +
a compose healthcheck that hits it (node global fetch — no curl/wget in image).
- worker: a background daemon thread touches /tmp/worker.alive every ~15s
(independent of the main loop, so it stays fresh through a long download) +
a compose healthcheck on the file's mtime (unhealthy if >90s stale).
Now `docker compose ps` reports real health for all services.
worker 223 tests, web 154, tsc + build + compose config clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Several client mutations either ignored a non-ok response (silent no-op) or
toasted success unconditionally even when the request failed. Sweep them so every
mutation gives feedback:
- artists: toggle auto-monitor, unfollow → error toast on failure.
- discography + wanted: toggle monitor / search-now → success or error toast.
- discover suggestion follow/want/dismiss → error toast (was toasting success
even on failure).
- The Floor: queue request + retry → error toast on failure.
- artist-modal + preview follow/want, Last.fm want → error toast on failure
(closes the deferred "want() shows no toast on non-ok" item).
web 153 tests, tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Bruno Mars "24K Magic" track 1 repeatedly downloaded as a 0-byte file. The
worker log showed the real cause: streamrip's track download hit
HTTPException("got more than 100 headers") — Qobuz's CDN returns >100 response
headers for some tracks and Python's http.client rejects them at the default
_MAXHEADERS=100, so streamrip skips the track and leaves an empty placeholder.
- Raise http.client._MAXHEADERS to 1000 on import of the Qobuz adapter, so those
responses parse and the track downloads.
- _count_staged_audio now skips 0-byte files (a failed-track placeholder is not a
real track) — improves the download-progress estimate and stops counting dead
tracks. (A verify-level "reject 0-byte tracks" safety net is a follow-up under
#16 per-track duration matching — it needs the pipeline fakes to write real
files, deferred to avoid disproportionate churn here.)
worker 223 tests / 7-skip.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Live-testing #9 surfaced a pre-existing gap: one truncated file mutagen chokes
on ("file said 4 bytes, read 0 bytes") raised out of _process_album and aborted
the entire library scan (worklist abandoned). Now scan_chunk catches a per-album
error, rolls back, logs it, counts it as skipped, marks the item done, and
continues — so a single bad file can't block the rest of the library.
worker tests: new case asserts a corrupt album is skipped while a good one still
imports (and captures its on-disk trackNames).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Library modal showed MusicBrainz's canonical tracklist, hiding incomplete
downloads / edition or track-order mismatches, and showed nothing for albums
without an rgMbid. Now it shows the REAL files on disk.
- LibraryItem gains trackNames String[] (migration add_library_track_names):
the album's on-disk "## Title.ext" audio filenames.
- Captured at import (pipeline._import lists the final album folder) and at scan
(scan._record_owned now also refreshes trackNames on re-scan via ON CONFLICT
DO UPDATE + xmax=0 to preserve the newly-imported flag). New shared
library.list_audio_files() helper.
- /api/library exposes trackNames; the AlbumModal prefers on-disk tracks when
present (parsed "## Title" → position/title, zero network, labeled "On disk ·
N tracks"), falling back to the MusicBrainz listing otherwise. Items imported
before this column show the MB fallback until re-scanned.
worker 221 tests / 7-skip, web 153, tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
A worker crash/restart mid-pipeline left its job stuck in matching/matched/
downloading/tagging with no auto-recovery — only the manual Retry (needs_attention
only) could rescue it. Now on startup the worker resets any such orphaned job
back to 'requested' so it gets re-claimed.
Safe at startup: this fresh worker owns none of those jobs, staging is already
cleared (clear_staging_root runs first), and the pipeline redoes intake→import
idempotently with an atomic import. Terminal jobs (imported/needs_attention) are
left untouched; claimedAt/error/downloadProgress are reset. Logged on startup.
worker 218 tests / 7-skip (in-flight→requested, terminal untouched, re-claimable).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds a section at the bottom of the Artists overview listing artists you own
music by (LibraryItem) but don't follow (WatchedArtist), each with a Follow
action — so you can start monitoring artists already in your library.
- /api/artists GET now also returns `ownedNotFollowed`: distinct LibraryItem
artist names not matching the watched roster (case-insensitive; the known
have-via-scan name-match caveat, #18).
- Follow resolves the library name → MBID (/api/mb/artists) then POST /api/artists
with the canonical MB name (mirrors the Last.fm/discover follow flow); a
no-match surfaces a toast. Followed names hide immediately via a session set,
even if the MB canonical name differs from the library name.
web 153 tests (owned-not-followed + case-insensitive match asserted), tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Qobuz/Monitor/Discovery help I just added was always-visible and verbose.
Replace each with a compact "?" button (new reusable HelpButton) that opens the
same content in a popup (reuses the Modal), keeping the settings tabs clean.
web tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Follow affordance in ALL album modals (user request): thread artistMbid through
the remaining callers — Library, Wanted, and artist Discography. /api/library
and /api/wanted now expose artistMbid (from the joined MonitoredRelease); the
artist-detail route already exposed mbid. So every album modal now links the
artist name to their page + shows a Follow/Following button.
Settings polish (#10):
- Qobuz tab: drop the email/password inputs (the engine prefers the more
reliable user-id + auth-token path); a "Clear legacy login" affordance appears
only when old email/password creds are still stored, and the PUT no longer
sends them.
- Clear-button alignment fixed: .field label is a flex row and .field-grid
bottom-aligns inputs, so a set-secret field (with its · set + Clear) lines up
with a plain sibling field.
- Library tab gains a read-only Staging path panel (/staging, STAGING_DIR, why
fast local disk, rebuild-to-change).
- Monitor + Discovery tabs gain grounded helptext (quality-class 1/2/3 cutoff,
poll/retry/upgrade-window; sweep cadence + seeds-per-chunk + ListenBrainz URL).
web 152 tests, tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
AlbumModal showed the artist name as static text — no way to follow the artist
or jump to them. Adds a Follow affordance (todo #13):
- AlbumInfo gains optional `artistMbid`. When present, the modal title links the
artist name to /discover/artist/{mbid} and shows a Follow/Following button that
reflects live follow state.
- Follow state fetched via a new lightweight GET /api/artists?mbid= check
(avoids pulling the full watched list + discography); Follow reuses
POST /api/artists. Non-ok POST surfaces a toast (also chips at todo #14).
- Wired artistMbid in the three discovery contexts where following is valuable:
Discover, Discover artist preview, and Last.fm. Library/discography callers
pass none (already-followed artists) so the button stays hidden — graceful.
web 152 tests (the ?mbid= follow-state check asserted), tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
/discover felt "stale until I hit Discover now": it showed the result string but
never WHEN the sweep last ran, and "Discovering…" ended the moment the worker
claimed the request (requested→false) rather than when the sweep finished.
- /api/discover/run GET now also returns `inProgress` (discover.inProgress) and
`lastRunAt` (the discover.result Config row's updatedAt = last completion).
- The client treats requested||inProgress as "running", so "Discovering…"
persists for the whole sweep, and polls until both clear before refreshing.
- The tool row shows "Last discovered {N ago} · {result}", a sweeping message
while running, and a clear "Not yet run" empty state.
web 151 tests (inProgress + lastRunAt asserted), tsc + build clean. No worker
change (signals already written); NOT a page cache (would worsen staleness).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Floor bar used a file-count poller that jumps 0→~99% for small/fast albums
(streamrip reports only 0/1). But yt-dlp and slskd already emit real byte-level
progress via the on_progress callback the pipeline was discarding
(lambda _pct: None).
- Thread a throttled on_progress → Job.downloadProgress (write on a >=1% move or
every 0.5s; clamped 0..1; write failures swallowed). Called synchronously from
adapter.download on the pipeline thread, so it reuses the main conn safely.
- The file-count poller stays as the FALLBACK: it skips writing while a
`reports` flag is set (an adapter is driving real progress), and the flag is
cleared per attempt so a fall-through to streamrip still gets the estimate.
- Smooth bars for YouTube + Soulseek now; Qobuz/streamrip keeps the file-count
estimate until its per-track callback is hooked later.
worker 216 tests / 7-skip (on_progress write/throttle/clamp covered; existing
pipeline tests now exercise the fake's on_progress end-to-end).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
/lastfm hit the Last.fm API live on every load / period switch / page. Cache the
raw Last.fm payload in the shared ApiCache table (12h TTL) while keeping the
own-state annotation (followed/owned/wanted) computed LIVE on top — so a follow
made after the payload was cached shows immediately.
- cached() gains a { force } option to bypass a fresh row (still serves stale if
the forced fetch fails); CACHE_TTL.HOUR added.
- All three routes wrap their Last.fm call: lfm-top-artists / lfm-top-albums
:{user}:{period}:{page}, lfm-artist-plays:{user}:{artist} (keys normalized).
- A "Refresh" button on /lastfm force-bypasses the cache (?refresh=1).
- Factored the copy-pasted getCreds/PERIODS/LIMIT into a route-only _creds.ts
(closes the Last.fm-routes DRY todo #19) — prisma stays out of any client
bundle since lib/lastfm.ts is imported by a client component.
web 150 tests (cache-hit + refresh + own-state-live asserted), tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The worker's MB browser (discography + artist search, used by monitor sweeps,
library scans, and Last.fm discovery) re-fetched every time. Route it through
the same ApiCache table + logical keys the web app uses, so a discography one
process fetched warms the other, and repeat lookups stop hitting MB.
- apicache.py cached_api(): read-through, serve-stale-on-outage, never caches
null, best-effort (a cache-DB error degrades to a live fetch, never breaks
scan/sweep/discovery). Its own DEDICATED autocommit connection so it can't
commit the worker's in-flight transaction, and autocommit avoids freezing
Postgres now() (which would break TTL math).
- CachingMbBrowser decorates MusicBrainzBrowser, serializing to the SAME
camelCase JSON shape the web writes (round-trip tested). registry.build_browser
wraps it when a DSN is present; the Last.fm source's browser rides it too.
- Hourly prune drops ApiCache rows > 90d so the table stays bounded.
- worker 213 tests / 7-skip.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Reopening an artist (discography), an album (tracklist), or repeating a lookup
re-hit MusicBrainz every time with no cache and no throttle. Adds a generic
Postgres-backed read-through cache keyed on the LOGICAL operation (not the
request URL) so the worker and web app share entries.
- New ApiCache(key, json, fetchedAt) model + migration (generic name so the
Last.fm browse cache can ride the same table).
- lib/apicache.ts cached(): read-through, TTL from fetchedAt (tunable, no
migration), serve-stale-on-outage, never caches null.
- Wrap the high-level MB fns with logical keys + TTLs: discography &
tracklists & artist-name 30d, searchReleaseGroup 7d, searchArtists 1d.
Search keys normalized (trim+lowercase) to match the worker byte-for-byte.
- album-modal: a session-scoped Map<rgMbid,Track[]> so reopening is instant
with zero network (complements the DB cache).
- Test setup clears ApiCache between tests. web 147 tests, tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
searchReleaseGroup used a strict exact-phrase releasegroup:"<album>", so
Last.fm album names with per-release edition/version qualifiers ("The Stranger
(Legacy Edition)", "Nevermind - Remastered") returned 0 hits → 404 → "No
MusicBrainz match", even though MB knows the artist and the base title.
Add stripEditionQualifiers (removes a trailing edition parenthetical/bracket or
" - " suffix only when it names an edition — deluxe/remaster/legacy/edition/…,
never a plain title parenthetical) and search in decreasing precision: exact
phrase on the given title → exact phrase on the stripped title → a loose
unquoted fallback. Clean titles still resolve in one request; albumPreferenceRank
ranking unchanged. Fixes both callers (/api/mb/release-group + /api/wanted).
Verified live against real MB: "The Stranger (Legacy Edition)"→The Stranger,
"Nevermind - Remastered"→Nevermind, Continuum unchanged. web 141 tests.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Modal focus-trap effect depended on [open, onClose] and called
panelRef.focus() on every run. Callers that pass a fresh onClose each render
(the common case, e.g. artists-client's closeAdd) re-ran the effect on every
parent re-render — so typing one char into the add-artist search (setQuery →
re-render → new onClose) yanked focus off the <input> back onto the panel,
making the field untypable.
Hold the latest onClose in a ref and depend on [open] alone, so focus-on-open
runs once per open. Fixes every Modal caller, not just add-artist.
Verified live via Playwright: typing "radiohead" one key at a time keeps the
value and document.activeElement on the search input.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Pre-merge review of the P0 branch:
- Open redirect: login redirected to an unvalidated ?next param, so a crafted
/login?next=https://evil (or //evil) bounced off-site after login. Now only
same-origin relative paths are honored (else fall back to "/").
- Remove the unused useRouter import/binding in the login form.
- Memoize expectedToken() keyed on (password, secret) so middleware stops
recomputing an HMAC on every gated request.
Deferred (low-value/latent, noted): Soulseek cover.jpg isn't retrieved into
staging (UI still shows Cover Art Archive art); _retrieve's basename-keyed map
could undercount same-named files but source_refs come from a single slskd dir;
SignOut visibility rides a server-env read (cosmetic).
web 138 tests green, tsc + build clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
.env.example shipped a valid, PUBLIC base64 key — if a user didn't replace it,
every credential got encrypted with a publicly-known key and everything still
"worked," so the footgun was silent.
- .env.example now ships an obvious placeholder (CHANGE_ME_generate_with_...)
with a "you MUST replace this" note and the openssl hint.
- web + worker validate LYRA_SECRET_KEY at startup: missing / placeholder /
wrong-length ⇒ loud FATAL banner and refuse to start (web via the Next.js
instrumentation hook, worker via _require_secret_key before touching the DB).
The old public example key ⇒ a loud WARNING but not fatal, since an existing
install may have encrypted its creds under it (rotating needs re-entry).
- Validation lives in an edge-safe web/src/lib/secret-key.ts (no node:crypto)
so the instrumentation hook bundles for both runtimes; crypto.ts re-exports.
- README emphasizes generating a fresh key + keeping it stable.
Verified live: worker exits 1 on placeholder/unset before any DB call; web
instrumentation throws and refuses to serve on the placeholder. web 138 tests,
worker 204/7-skip, tsc clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The web app had no auth — anyone reaching the host could browse the library,
trigger downloads, and read/overwrite the stored credentials in Settings. Adds
an opt-in single-shared-password gate:
- Next.js middleware redirects unauthenticated page requests to /login and
returns 401 for /api/*; static assets are excluded via matcher.
- Auth is enabled ONLY when LYRA_PASSWORD is set (unset ⇒ app stays open, so
existing deployments are unchanged). Documented in README "Security".
- The session cookie is httpOnly and carries HMAC-SHA256(password) keyed by
LYRA_SECRET_KEY — unforgeable, and the plaintext password never touches the
client. Edge-safe Web Crypto so one lib/auth.ts serves middleware + route.
- Cookie is NOT Secure: Lyra serves plain HTTP on the LAN; a Secure cookie would
loop. Works behind an HTTPS proxy too.
- /login page + form, a Sign out button (shown only when auth is enabled), and
chrome (nav + worker-status poll) suppressed on /login.
- docker-compose passes LYRA_PASSWORD to web; .env.example documents it.
Verified live end-to-end (built + next start): no-cookie page→307 /login,
API→401, wrong password→401, correct→200+cookie, authed page/API→200, forged
cookie→401, and the auth-disabled (no LYRA_PASSWORD) path stays fully open.
web 133 tests green, tsc clean.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
All durable state (library, monitor/wanted lists, discovery, config, and the
encrypted credentials) lived only in the single postgres-data volume with no
backup — one volume loss = total loss. Adds a db-backup compose sidecar that
pg_dumps immediately on start then every BACKUP_INTERVAL_SECONDS (default daily),
keeping the last BACKUP_KEEP (default 7) custom-format dumps in ${BACKUP_DIR}.
Travels with the stack (no host cron), starts with `up -d`. .env.example
documents the knobs; README adds a "Backup & restore" section with pg_restore
steps and an off-box note. Verified live: sidecar wrote a valid 128K dump of the
real lyra DB (pg_restore -l lists all tables + _prisma_migrations).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
SlskdClient enqueued the transfer and polled to Completed but never moved any
files into `dest` — slskd saves them to its OWN downloads dir, so staging stayed
empty and every Soulseek job failed "incomplete download". Qobuz/YouTube write
straight into staging and were unaffected.
- SlskdClient._retrieve() moves the finished files out of slskd's downloads dir
into staging after completion, matching tolerantly by basename and preferring
a parent-folder + size match (slskd's on-disk layout varies by version).
- download() returns the moved count as track_count so the pipeline's
completeness check reflects what actually landed; raises loudly if nothing was
found (misconfigured/unmounted downloads dir).
- Downloads dir resolves slskd.downloadsDir Config > SLSKD_DOWNLOADS_ROOT env >
/slskd-downloads default; docker-compose mounts ${SLSKD_DOWNLOADS_DIR} there.
- .env.example documents SLSKD_DOWNLOADS_DIR; retrieval unit-tested offline.
slskd runs as an external daemon, so the shared dir must be a host path both
slskd and the worker mount — set once Lyra runs on the slskd host. Not yet
verified live end-to-end (no slskd access from this VM).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
The Last.fm browse page has the same live-external-call-per-load problem;
name the cache table/helper generically (ApiCache) so Last.fm rides the
same read-through, with a manual Refresh button and live own-state
annotation.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
A read-through cache shared by the web app and the worker so reopening an
artist/album (and repeat lookups during scans/discovery) stop re-hitting
MusicBrainz. Postgres-only (Redis disregarded — PG is already shared across
web + worker); keys on the logical operation so both processes share entries;
serve-stale-on-outage; TTL-only invalidation. Four tasks: MbCache table →
web read-through → worker CachingMbBrowser → client Map + pruning.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Bring the README current with everything shipped since it was written:
Last.fm as a second discovery source + the /lastfm browse page, the
/library grid, The Floor's three-phase progress + retry, and the
Pressing Plant design system. Add a "Running it" quick-start. Fix a
pre-existing inaccuracy: slskd is an external daemon reached via
config, not a compose service (stack is db + web + worker).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Replace the misleading 6-micro-step progress bar with three honest
phases (Searching -> Downloading -> Finishing). Downloading now shows
a live determinate percentage sourced from job.downloadProgress (Slice
A), with a "NN% · done/total tracks" caption when the track count is
known. Searching and Finishing stay indeterminate.
- status.ts: describeJob collapses to {label, kind}; drop step/totalSteps
and the STAGES array.
- progress-bar.tsx: ProgressBar takes {kind, indeterminate?, percent?,
caption?} instead of step/total/stageLabel.
- job-row.tsx: JobRow takes a `bar?: ReactNode` slot instead of
step/total/indeterminate/stageLabel props.
- queue.tsx: builds the per-state bar + meta; downloading meta drops
the track count (now shown in the bar). Retry button + error note
from the retry feature preserved exactly.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add Job.downloadProgress (0.0-1.0), derived from a worker background
poller that counts audio files landing in staging / expected track
count, since streamrip only reports 0/1. No UI yet (slice B).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Add POST /api/requests/[id]/retry: 404 if the request/job is missing,
400 if the job isn't in needs_attention, otherwise clears the job's
stale candidates and resets it to state=requested/currentStage=intake
(error/claimedAt cleared) plus request status=pending, in a
transaction, so the worker's requested-only claim re-picks it up.
Wire a Retry button into the attention-row note on the Floor.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Each search/download called asyncio.run, creating and closing a fresh event loop.
streamrip's module-global download-concurrency Semaphore binds to the loop on first
use, so the 2nd+ download reused a Semaphore bound to a closed loop -> "Semaphore is
bound to a different event loop", tracks failed, album arrived short -> "incomplete
download" / needs_attention. Run all streamrip coroutines on one persistent loop.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
_mark_chosen ran only after a download succeeded, so throughout the 'downloading'
state no candidate was chosen — the Floor row's "source · format · tracks" line was
blank (only elapsed showed). Now mark the candidate exclusively at each download
attempt so `chosen` reflects the source in flight (and updates on fall-through).
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Last.fm returns error 27 "Deprecated" for user.getArtistTracks, so the per-artist
plays modal was 502ing for every artist. Pivot to filtering the user's supported
all-time top tracks + top albums (which carry real per-item play counts) to the
artist — exact personal counts, cheaper, and the clicked artists are the user's
top artists so their items cluster near the front. Adds topTracks(); route + modal
return shape unchanged.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds an optional lastfmName prop to ArtistModal. When set (only on
/lastfm) it fetches /api/lastfm/artist-plays independently of the
existing preview fetch and renders a most-played-songs list plus a
most-played-albums list above the release grid. Album Want resolves
via /api/mb/release-group then POSTs /api/discover/want, mirroring
the existing want(r) flow. Own-state chips name-match Last.fm albums
against the already-loaded release grid. Other ArtistModal callers
(discover-client.tsx) pass no lastfmName, so their behavior is
unchanged.
searchReleaseGroup took the first MusicBrainz hit, so a title MB ranks with the
single first (e.g. "Thriller" ties single+album at score 100) resolved to the
single — the Last.fm album modal and the Want flow then showed a 1-track single.
Prefer a clean Album > EP > Single > album-with-secondary, keeping MB order as the
in-tier tiebreak; single-only titles still resolve.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Adds LastfmSource, resolving Last.fm's often-empty artist mbid via the
injected MB browser (name->mbid cached) so every emitted SimilarArtist.mbid
is a real MusicBrainz artist MBID. Registered in build_similarity_sources,
gated on lastfm.api_key so the ListenBrainz-only path is unchanged when
unset.
Jonathan