backend: drop unreliable artist *= filter entirely — yt-dlp artist field
is empty for many label-distributed YTM tracks (NoAki, etc.), causing
correct results to be rejected in Steps 1-2 and falling through to Step 3
anyway. Search query already targets the right artist; title keyword +
duration are sufficient. Simplifies cascade from 8→8 steps but with fewer
wasted filter attempts.
frontend: add 3s delay before polling getScanStatus (gives Navidrome time
to actually start the scan), and 5s buffer after scan completes (lets DB
writes settle). Fallback to 30s fixed wait if scan API is unavailable.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
backend: replace title *= (case-sensitive) with title ~= (?i) regex so
"MEMORIES" and "Echoes of Eternity" match regardless of YT title casing.
Refactor buildTitleKeyword → buildTitleKeywords returning array.
frontend: trigger Navidrome startScan after download completes and wait
for scan to finish before running _createNdPlaylist, so newly downloaded
tracks are indexed and searchable when the playlist is built.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- artistFilterWord: use longest \w+ run instead of stripping special chars,
fixes artists like N!GHT where stripping gave NGHT (not a substring)
- buildTitleKeyword: new fn extracts most distinctive title word, added as
title *= filter to all cascade steps to block unrelated fallback downloads
- Mix version specificity: Radio/Extended/Club Mix titles also filter on the
qualifier word so plain versions are rejected
- Switch uploader *= to artist *= so label-uploaded tracks (Monstercat etc.)
match by music metadata artist tag, not channel name
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Code 101 means --max-downloads was reached — the file was downloaded
successfully. Treating it as error caused the ffmpeg/metadata step to
be skipped, leaving files with no tags.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
--playlist-items 1-5 --max-downloads 1 on all filtered steps so yt-dlp
skips non-matching results (wrong artist, sped up, wrong duration) and
lands on the first result that actually passes all active filters.
Unfiltered last-resort fallbacks keep --playlist-items 1.
YouTube fallback uses ytsearch5: to match the 5-result window.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Snapshot dir before each yt-dlp call so hasAudio() only detects
files added by the current attempt, not pre-existing files from
parallel jobs downloading to the same album directory.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Duration: ±15% (min 15s) instead of -40%/+50%
- Variant filter: reject sped/speedup/slowed/reverb/nightcore/karaoke/
parody/remix/cover/acoustic/instrumental unless original title has keyword
- Cascade: YTM+YT tried at each level (dur+artist+variant → no artist → no dur → no filter)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Strips special chars from artist name (☆NoAki☆ → NoAki) and adds
uploader *= check so wrong-artist results are rejected early.
Falls back to duration-only, then no-filter if artist filter yields nothing.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Previous order let wrong YTM results through on the no-filter fallback.
New order: YTM+filter → YT+filter → YTM no filter → YT no filter.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
ytmsearch1: is not a valid yt-dlp scheme — use full music.youtube.com URL with --playlist-items 1.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
official audio consistently breaks remixes and lesser-known artists.
Duration filter is the primary quality check.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Remix titles (remix/edit/flip/vip/etc.) are specific enough without suffix.
Normal tracks get 'official audio' to prefer audio over music videos.
Original official-audio issue was only for remixes — now caught by regex.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adding 'audio' shifts results for lesser-known artists (Kanro, etc.).
Plain 'Artist - Title' matches what users find manually on YouTube.
Duration filter is the reliable quality check against wrong songs.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Remixes already have '(Remix)' etc. in title so no suffix needed.
Normal tracks get 'audio' to filter piano covers and tutorials.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
'official audio' biased towards artist's main channel (wrong for remixes).
No suffix caused piano covers to rank top. 'audio' filters covers/tutorials
without over-specifying to official channels.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
For remixes and edits, 'official audio' biases YouTube search towards the
artist's official channel (wrong song) instead of the specific version.
Duration filter is now the quality check — the plain query is more accurate.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
--max-downloads 1 exits with code 101 which varies by yt-dlp version — caused
ffmpeg to never run, leaving untagged files. Use ytsearch1: + --match-filter instead:
filters top result by duration, exits 0 in all cases, falls back cleanly if no match.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
--playlist-items 1 in baseArgs caused filter to only check result #1 (filter useless).
--max-downloads 1 exits with code 101 which we treated as an error — ffmpeg never ran,
leaving untagged files (→ Unknown Artist/Album in Navidrome). Now code 101 resolves
successfully and filtered search uses commonArgs without --playlist-items.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
When iTunes/Spotify provides a track duration, search 5 YouTube results
and keep only those within 60–150% of the expected length (filters remixes,
prayer videos, etc.). Falls back to ytsearch1: if no match is found.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Replace apk yt-dlp with pip3 install to get current version with ytmsearch support
- Split mainArtist on comma in downloadTrack so multi-artist tracks
always land in the primary artist folder (e.g. "Artist1, Artist2" → Artist1/)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- Switch to YouTube Music search (ytmsearch3) as primary source with
fallback to ytsearch5 if no result found
- Add duration filter (±30s/+45s) and live-stream exclusion via
--match-filter when iTunes trackTimeMillis is available
- Send trackTimeMillis from frontend in download payload
- Pass durationMs through downloadTrack and downloadAlbumTracks
- Use first artist only (split on comma) for folder path on multi-artist tracks
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- itunesProxy: drop 3-arg form (url, options, cb) — use 2-arg (url, cb) like lookupItunesTrackMeta
- _iGet: if proxy returns {results:[]} trust it (iTunes said no results); only fall back to direct Apple if proxy gave no response at all
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Apple now blocks direct browser requests from custom origins. Added backend proxy routes:
GET /api/itunes/search — iTunes search
GET /api/itunes/lookup — iTunes lookup by ID
GET /api/itunes/charts — Apple RSS most-played charts (albums/songs)
All itunesSearch/Lookup/TopCharts/TopSongs functions in api.js now route through backend.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- track-artwork endpoint: split artist on ,/•/ before iTunes search (same as download flow)
- track-artwork endpoint: fall back to existingArt when iTunes returns nothing
- frontend: render any artwork already in Spotify data immediately, don't wait for iTunes
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- nowplaying: add cancellation token to _loadNpLyrics to prevent stale async
calls overwriting newer results (synced→plain race condition)
- nowplaying: replace scrollIntoView with explicit container.scrollTo to prevent
page-level scroll on desktop (topbar shifting / gap at edge)
- backend: remove results[0] fallback in lookupItunesTrackMeta — only use
exact title+artist match to avoid embedding wrong artwork
- components/views: makeSongItem accepts noArtistLink/noAlbumLink opts;
album-detail disables both, playlist-detail disables artist link
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
navidromeArtist: replace ', ' with ' / ' so Navidrome splits
multi-artist strings (e.g. 'Artist A, Artist B' → 'Artist A / Artist B')
with existing Tags.artist.Split = ["/"] config.
Playlist tracks: remove playlist cover URL as defaultArt in _spTrackItem
so that songs where iTunes lookup fails get no embedded artwork instead
of the wrong playlist cover art.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Navidrome splits artist tags on '/' (Tags.artist.Split = ["/"]).
Using ' / ' between multiple artists (both from array format and from
Spotify's comma-separated subtitle) lets Navidrome register them as
separate artists without any config change.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Multi-artist strings like "Kløn, Renzyx" from Spotify's subtitle field
confuse the iTunes search API. Use only the first artist (split on comma)
as the search term, same as the playlist path already does.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Spotify's flat embed format provides no album, year or per-track artwork.
Now calls lookupItunesTrackMeta for every playlist track before downloading:
- Uses first artist of "Artist1, Artist2" subtitle for iTunes search query
(better match), keeps full string as the MP3 tag
- iTunes album/year/trackNum/artwork override the empty Spotify fields
- Falls back to Spotify data if iTunes returns nothing
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Adds POST /api/spotify/track-artwork endpoint that takes an array of
{idx, trackName, artistName} and returns iTunes artwork URLs in batches
of 8 concurrent requests.
Frontend calls this after rendering the track list, then patches each
.gtrack-art element with the correct album cover as results arrive.
Also updates _spotifyResolved.tracks so downloads use the right artwork.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Spotify's playlist embed uses {title, subtitle, audioPreview} per track
instead of {name, artists, album}. Fixes:
- _spTrackItem: reads title as trackName, subtitle as artistName fallback,
audioPreview.url as previewUrl, isExplicit as boolean
- Route: reads entity.authors[0].name for playlist owner, entity.subtitle
for track artist fallback, entity.title fallback for entity.name
- Playlist tracks get playlist cover art as default artwork (no per-track art)
- Removes temporary debug logging
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- _spTrackItem now unwraps GraphQL-style {data:{__typename,...}} wrappers
- _spAllArtists joins all featured artists (e.g. "Kløn, Renzyx")
- _spArtwork adds albumOfTrack path used by newer Spotify embed
- fetchSpotifyEmbed: added entity.content.items / entity.data fallback paths
- Track download now uses Spotify metadata directly (album, art, year,
trackNum) when provided — avoids iTunes returning wrong album for
multi-artist tracks
- Playlist trackList also tries entity.content.items and entity.tracks.items
- Frontend: redesigned preview card with green top accent, bigger art,
playlist checkbox styled as pill with Spotify color on check, album
button shows actual album name
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Removes the Spotify Web API token approach (blocked on non-Premium
accounts and from datacenter IPs). Now fetches open.spotify.com/embed
pages and parses __NEXT_DATA__ JSON — works without any auth from any IP.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- parseSpotifyUrl now handles intl-de/, intl-en/ and other locale prefixes
- getSpotifyToken uses full browser headers + sp_t cookie for the public endpoint
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Replace Client Credentials flow with Spotify's public web-player token endpoint
(open.spotify.com/get_access_token) which works without any account or Premium.
Remove Spotify credential fields from UI entirely.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- New 'Spotify Importer' page under Entdecken (sidebar + mobile nav)
- Backend: POST /api/spotify/resolve — resolves track/album/playlist URLs via Spotify API
- Backend: new 'playlist' download type accepting pre-resolved track list (no iTunes re-fetch)
- Frontend: paste Spotify link → preview with cover, type badge, track list
- Track: single download or full album via iTunes fallback
- Album: download all tracks with Spotify metadata
- Playlist: download all tracks + optional 'Als Navidrome-Playlist speichern' checkbox
- Navidrome playlist creation: searches library by title/artist after download, creates playlist
- Spotify Client ID + Secret stored in app settings (setup screen + settings modal)
- iTunes metadata tags kept for all downloaded files (existing pipeline unchanged)
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Songs from new releases (Apple RSS feed) have no collectionName, causing
the album tag to be omitted from embedded metadata -> Navidrome shows
[Unknown Album]. Backend now calls iTunes search to fill in album, year,
trackNum, totalTracks, discNum, genre before downloading.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Backend:
- fetchLyrics() queries LRCLIB for plain + synced lyrics
- Synced lyrics saved as .lrc file (Navidrome reads these)
- Plain lyrics embedded via ffmpeg -metadata lyrics=
- ITUNESADVISORY=1 written for explicit tracks
- explicit flag passed through album/track download chain
- fetchItunesTracks + searchItunesAlbumTracks use country=us&explicit=Yes
Frontend:
- explicit:true sent in download payload from iTunes track data
- "E" badge shown on explicit tracks in global search track list
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Backend: set skipped:true on job when track already exists on disk.
Album downloads report "X geladen, Y bereits vorhanden" in the message.
Frontend: pollJob shows "⏭ bereits auf dem Server" instead of "✅ gespeichert"
when the job has skipped:true.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Before starting yt-dlp, check if Artist/Album/Title.{ext} already
exists on disk. If found, log a skip message and return early instead
of re-downloading and overwriting the existing file.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
HoneyandClaude Sonnet 4.6