Files
yellowjacket/backend/database/sql/queries/audio_files.sql
T
yonluandClaude Opus 5 e7748f1fd5
CI / check (push) Successful in 3m7s
CI / e2e (push) Canceled after 1m45s
feat(database): shape the library like files, and shrink the catalog
Plans 013 and 014, the album page that prompted them, and the smaller
fixes they turned up. Changelog, largest first.

## The local library is shaped like files, not like MusicBrainz

`audio_files` carries its own tags and points at `albums` and
`artists`; `file_genres` is the one real many-to-many. `recordings`,
`release_group_recordings`, `artist_credit`, `artist_credit_artist`,
`recording_genres`, `release_groups` and `release_to_rg` are gone from
the local side, and with them a six-way join in every read, a
`MIN(release_group_id)` subquery in eleven queries and a
first-credited-artist subquery in nine. Measured on a real 25,966-file
library, every many-to-many that model expressed was 1:1 in the data.

- Ownership is a file. `GetFilePathsByRecordingMBIDs`,
  `LibraryMBIDIndex.CheckMBIDs`, `collectLibraryEntities` and
  `pruneStaleLocalCrossReferences` all join `audio_files`, so the 812
  orphaned recordings, 216 release groups and 260 artists that library
  carried are now structurally impossible.
- One projection: every track query selects from the `track_metadata`
  view, one row type, one mapper. Nine hand-rolled copies had drifted
  far enough to report different years on different screens.
- `library_id = 0` means every library, so each list query exists once
  instead of scoped and unscoped with a branch at every call site.
- No migration chain. `sql/schemas/` is the one description of the
  shape; `sql/migrations/`, `applyMigrations` and `schema_migrations`
  are squashed away, along with the drift between them that had sqlc
  generating against a stale schema.
- `database.InsertTestTrack` is the one test seeder; twenty test files
  had been assembling the old FK chain each in its own order.

## The catalog stores its ids as bytes

`explore_index`'s three 36-char MBID columns and its entity-type text
are 16 raw bytes and a small integer. The table and its six indexes go
780 MB to 405 MB on a real 2,052,200-row catalog, which is why a fresh
install is ~0.6 GB rather than ~1.0 GB.

- `backend/explore/mbid.go` is the only place the encoding is known;
  everything above it speaks dashed strings.
- `CHECK(length(mbid) = 16)` makes a stringly write fail at the insert
  rather than silently returning no rows, since SQLite does not coerce
  between TEXT and BLOB.
- The importer asks the artifact what encoding it carries and converts
  on the way in, so the artifact already published keeps working and no
  format bump is needed.
- `indexRowColumns`/`scanIndexRow` replace four copies of a 22-column
  list, and `TestStoredEncodingRoundTrips` sweeps every read path.

## An album page that says how much of the album is yours

- One question, asked once: is there a file. `filePaths` is filled by a
  single batched lookup when the tracklist settles, and the badge, the
  Play count, the dimmed rows and every menu item read it — replacing
  four claims of decreasing confidence that could show a green tick on
  an album whose every action did nothing.
- Play, Play 7 of 12, or no play button at all.
- `total_tracks` on `explore_index` (~2 bytes over 400,677 release
  groups) and on `audio_files` from tags that have always carried it:
  a complete MBID-matched album now makes no catalog call at all, where
  it used to spend the most expensive request the app makes.
- A merged cluster shows the running order the most releases agree on,
  and the version list marks the release you own rather than standing a
  synthetic entry in for it.
- `AlbumReleasesFailed`: a slow fetch is no longer reported as a failed
  one by a 12-second timer.
- Rows not in the library are dimmed in place (with `aria-disabled`)
  instead of the owned ones wearing a green tick and a legend.

## Caches and cover art get ceilings

- Only the three tiers of a cover are stored; the full-resolution copy
  nothing rendered was 1,134 MB of a 1.4 GB covers directory.
- One artist portrait is downloaded and the rest are remembered as
  URLs — 4.1 GB of a 5.3 GB cache was candidates no code path reads.
- `browsedArtBudget` and `httpCacheBudget` bound what an age cannot:
  the same install held art for 5,770 artists in a 1,301-artist
  library.
- `OrphanedArtistImagesJob` joined a bare MBID onto a sharded
  directory, so it deleted the rows that were the only record of the
  files it left behind. `explore.ArtistImageDir` is that layout's one
  definition now.

## The autotag queue asks whether there is work

`tagging_items` was a row per album folder, not a queue, and no query
read the `tag_status` column that held the answer. The four queue
queries ask the files, which matters most where it is least visible:
`startPrefetch` was scoring every album in a tagged library against
MusicBrainz.

## Phantom playlist tracks resolve in place

An M3U8 imported before its files leaves phantom rows; they now match
by path and fall back to position, keep their place in the playlist
when resolved, and pair best-first so two phantoms cannot claim the
same file.

## Playing a track plays the list it is in

Double-click, and Play on a single row's menu, queue the list as
displayed with `startIndex` on that row — the album page and the track
list used to queue one track and discard the album around it. A
multi-row selection still plays exactly itself.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AfVYUVExXsx1nSWrXN8mAh
2026-08-16 13:58:15 -04:00

210 lines
8.0 KiB
SQL

-- Queries over audio_files and the track_metadata view above it.
--
-- Every query that returns "a track" selects from `track_metadata`,
-- which is the one place the projection is defined. The scoped and
-- unscoped variants that used to be written twice are one query now:
-- library_id 0 means "every library", and `(:id = 0 OR library_id = :id)`
-- costs nothing measurable (23 ms vs 21 ms over 26k rows) because these
-- queries scan either way.
-- ---------------------------------------------------------------------
-- Writes
-- ---------------------------------------------------------------------
-- name: CreateAudioFile :one
INSERT INTO audio_files (
file_path, library_id, file_type_id,
length_milliseconds, sample_rate, bit_depth, channels, bitrate, file_size,
title, artist_credit, artist_id, album_id,
track_number, disc_number, total_tracks, year, composer, comment,
recording_mbid, basename, group_key, modified_at, tag_status
) VALUES (
?, ?, ?,
?, ?, ?, ?, ?, ?,
?, ?, ?, ?,
?, ?, ?, ?, ?, ?,
?, ?, ?, ?, ?
)
RETURNING *;
-- name: UpdateAudioFileTags :exec
-- A rescan of a file whose mtime moved: the tags are re-read and
-- written over the same row. Under the old schema this created a
-- *new* recording and repointed the file at it, abandoning the old one
-- -- which is where 812 orphaned rows and every phantom "you own this"
-- came from. There is nothing to orphan now.
UPDATE audio_files
SET title = ?, artist_credit = ?, artist_id = ?, album_id = ?,
track_number = ?, disc_number = ?, total_tracks = ?, year = ?,
composer = ?, comment = ?, recording_mbid = ?,
sample_rate = ?, bit_depth = ?, channels = ?, bitrate = ?,
file_size = ?, length_milliseconds = ?, modified_at = ?
WHERE id = ?;
-- name: SetAudioFileGroupKey :exec
UPDATE audio_files SET group_key = ? WHERE id = ?;
-- name: PromoteAudioFileTagStatusIfUntagged :exec
-- A rescan re-reads the tags of a file whose mtime moved, so a file
-- another tagger stamped with MBIDs since import arrives here still
-- carrying the 'untagged' status it was created with (only the insert
-- path sets it). Promote it the same way saveAudioFile does.
-- Guarded on 'untagged' so it cannot overwrite a deliberate
-- 'user_skipped_permanent', and so a file losing its MBIDs is left
-- alone -- demotion is the scan's judgement, not this statement's.
UPDATE audio_files
SET tag_status = 'user_confirmed'
WHERE id = ? AND tag_status = 'untagged';
-- name: UpdateAudioFileStat :exec
-- Records the on-disk mtime/size without re-reading tags. Used to
-- backfill the staleness baseline for files the scan skipped, and to
-- re-baseline after YellowJacket's own tag writer rewrites a file.
UPDATE audio_files
SET modified_at = ?, file_size = ?
WHERE id = ?;
-- name: SetAudioFileRecordingMBID :exec
UPDATE audio_files SET recording_mbid = ? WHERE id = ?;
-- name: DeleteAudioFile :exec
DELETE FROM audio_files WHERE id = ?;
-- name: DeleteAllAudioFiles :exec
DELETE FROM audio_files;
-- ---------------------------------------------------------------------
-- Reads: the file row itself
-- ---------------------------------------------------------------------
-- name: GetAudioFile :one
SELECT * FROM audio_files WHERE id = ? LIMIT 1;
-- name: GetAudioFileByPath :one
SELECT * FROM audio_files WHERE file_path = ? LIMIT 1;
-- name: GetAudioFileGroupKey :one
SELECT group_key FROM audio_files WHERE id = ? LIMIT 1;
-- name: GetAllAudioFilePaths :many
SELECT id, file_path FROM audio_files;
-- name: GetAudioFilesByPaths :many
SELECT id, library_id, file_path, group_key FROM audio_files
WHERE file_path IN (sqlc.slice('paths'));
-- name: GetRandomAudioFilePath :one
SELECT file_path FROM audio_files ORDER BY RANDOM() LIMIT 1;
-- name: CountAudioFiles :one
SELECT COUNT(*) AS count FROM audio_files
WHERE library_id = COALESCE(NULLIF(CAST(sqlc.arg(library_id) AS INTEGER), 0), library_id);
-- name: GetLibraryMaxModifiedAt :one
-- Newest recorded mtime in a library, for the startup soft scan. 0 when
-- the library is empty or no row has a baseline yet.
SELECT CAST(COALESCE(MAX(modified_at), 0) AS INTEGER) FROM audio_files
WHERE library_id = ?;
-- ---------------------------------------------------------------------
-- Reads: tracks
-- ---------------------------------------------------------------------
-- name: GetTracks :many
SELECT * FROM track_metadata
WHERE library_id = COALESCE(NULLIF(CAST(sqlc.arg(library_id) AS INTEGER), 0), library_id);
-- name: GetTrackByPath :one
SELECT * FROM track_metadata WHERE file_path = ? LIMIT 1;
-- name: GetTracksByAlbum :many
SELECT * FROM track_metadata
WHERE album_id = sqlc.arg(album_id)
AND library_id = COALESCE(NULLIF(CAST(sqlc.arg(library_id) AS INTEGER), 0), library_id)
ORDER BY disc_number, track_number;
-- name: GetTracksByGenre :many
SELECT tm.* FROM track_metadata tm
JOIN file_genres fg ON fg.audio_file_id = tm.id
JOIN genres g ON g.id = fg.genre_id
WHERE g.name = sqlc.arg(genre)
AND tm.library_id = COALESCE(NULLIF(CAST(sqlc.arg(library_id) AS INTEGER), 0), tm.library_id);
-- name: LookupTrackMetaByPaths :many
SELECT id, file_path, title, artist_name, album, cover_art_path,
artist_mbid, release_group_mbid, recording_mbid
FROM track_metadata
WHERE file_path IN (sqlc.slice('paths'));
-- name: SearchTracksByBasename :many
SELECT id, file_path, length_milliseconds, title, artist_name, album
FROM track_metadata
WHERE file_path IN (
SELECT file_path FROM audio_files WHERE basename = sqlc.arg(basename)
)
LIMIT sqlc.arg(lim);
-- ---------------------------------------------------------------------
-- Reads: file paths, grouped by whatever the caller asked about
-- ---------------------------------------------------------------------
-- These answer "what can I play" and they all ask audio_files, because
-- that is the only table whose rows are files. Grouped rather than
-- flattened because the caller owns the order.
-- name: GetFilePathsByAlbums :many
-- The library filter is applied in Go rather than here: sqlc numbers a
-- named parameter (?2) but expands a slice into N placeholders, so the
-- two together bind the wrong values - GetFilePathsByAlbums([1,2], 0)
-- read album id 2 as the library id. Returning library_id and
-- filtering the (small) result is the version that cannot be wrong.
SELECT album_id, library_id, file_path FROM audio_files
WHERE album_id IN (sqlc.slice('album_ids'))
ORDER BY disc_number, track_number;
-- name: GetFilePathsByRecordingMBIDs :many
-- The ownership question in its only honest form: which of these
-- catalog recordings has a *file* behind it. Asked of audio_files, so
-- a metadata row with no file cannot answer yes.
SELECT recording_mbid, library_id, file_path FROM audio_files
WHERE recording_mbid IN (sqlc.slice('mbids'))
ORDER BY file_path;
-- name: GetFilePathsByGenres :many
SELECT g.name AS genre, af.library_id, af.file_path
FROM audio_files af
JOIN file_genres fg ON fg.audio_file_id = af.id
JOIN genres g ON g.id = fg.genre_id
WHERE g.name IN (sqlc.slice('genres'))
ORDER BY af.disc_number, af.track_number;
-- name: GetFilePathsByArtistMBID :many
SELECT DISTINCT af.file_path
FROM audio_files af
JOIN artists a ON a.id = af.artist_id
WHERE a.mbid = ?;
-- ---------------------------------------------------------------------
-- Ownership, asked in bulk
-- ---------------------------------------------------------------------
-- name: OwnedRecordingMBIDs :many
-- Which of these recording MBIDs are actually in the library. This is
-- what marks a catalog tracklist owned; it used to be
-- `SELECT mbid FROM recordings`, which answered yes for 129 tracks in a
-- real library that had no file at all.
SELECT DISTINCT recording_mbid FROM audio_files
WHERE recording_mbid IN (sqlc.slice('mbids'));
-- name: OwnedAlbumMBIDs :many
SELECT DISTINCT al.mbid FROM albums al
JOIN audio_files af ON af.album_id = al.id
WHERE al.mbid IN (sqlc.slice('mbids'));
-- name: OwnedArtistMBIDs :many
SELECT DISTINCT a.mbid FROM artists a
JOIN audio_files af ON af.artist_id = a.id
WHERE a.mbid IN (sqlc.slice('mbids'));
-- name: GetAudioFilesInLibrary :many
SELECT * FROM audio_files WHERE library_id = ?;