Files
yellowjacket/frontend/test/components/album-catalog-scope.test.ts
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

334 lines
10 KiB
TypeScript

/**
* What the album page claims about the catalog while it is waiting.
*
* The scope notice said "No catalog details for this album right now"
* on albums that were matched correctly and whose catalog data arrived
* a few seconds later. The cause was that *not having an answer yet*
* and *having been told there is no answer* were the same state: the
* page inferred a failure from a deadline, and the deadline was 12 s
* against a browse that waits on a 1 req/s limiter shared with
* `PrefetchReleases`, which fires up to eight of them when an artist
* page renders.
*
* So the rule under test is that `unavailable` is only ever reached by
* something *telling* the page the catalog did not answer —
* `AlbumReleasesFailed`, or an empty result after the background fetch
* reported itself done.
*
* A fetch that is merely slow says *nothing at all*. It used to say
* "showing what your library has while the full album details load",
* which is a sentence about the page's own plumbing; the dimmed rows in
* the tracklist carry that information without a banner, so tracks
* arriving dimmed reads as the album filling in.
*/
import { describe, expect, it, beforeEach } from 'vitest';
import type { LitElement } from 'lit';
import '@components/explore-album-details/explore-album-details';
import { stub, emit, flush, resetHarness, calls } from '@test/support/harness';
import { fixture, shadow } from '@test/support/render';
const MBID = 'rg-0001';
/** The scope the notice is currently being rendered with. */
function scope(el: LitElement): string | null {
return shadow(el, 'catalog-scope-notice')?.getAttribute('scope') ?? null;
}
/**
* An album page mid-flight: the release group resolves, but
* `BrowseReleases` returns empty, which is what the local-first backend
* path does on a cold cache while it fetches in the background.
*/
async function coldAlbum(): Promise<LitElement> {
const el = await fixture<LitElement>('explore-album-details', {
releaseGroupMBID: MBID,
albumName: 'Glass Harbour',
});
await flush();
await el.updateComplete;
return el;
}
describe('what the album page says while the catalog is still coming', () => {
beforeEach(() => {
resetHarness();
stub('explore.Service.BrowseReleases', []);
stub('explore.Service.LookupReleaseGroup', {
mbid: MBID,
title: 'Glass Harbour',
artistCredit: 'Tideline',
});
stub('explore.Service.GetThumbnail', '');
stub('library.Library.GetAlbumTracks', []);
stub('library.Library.GetAllLibrariesWithTrackCounts', []);
stub('library.Library.GetAlbumCompleteness', {
owned: 0,
expected: 0,
known: false,
complete: false,
});
stub('library.Library.GetFilePathsByRecordingMBIDs', {});
});
it('does not call a slow fetch a failure', async () => {
const el = await coldAlbum();
// No event either way yet — the background browse is still queued,
// and `catalog` is the silent scope: the notice renders nothing.
expect(scope(el)).toBe('catalog');
});
it('says the catalog is unavailable when the browse reports failing', async () => {
const el = await coldAlbum();
emit('AlbumReleasesFailed', MBID);
await flush();
await el.updateComplete;
expect(scope(el)).toBe('unavailable');
});
it('ignores a failure for a different release group', async () => {
const el = await coldAlbum();
emit('AlbumReleasesFailed', 'rg-9999');
await flush();
await el.updateComplete;
expect(scope(el)).toBe('catalog');
});
it('says unavailable when the catalog answers with nothing', async () => {
const el = await coldAlbum();
// The background fetch reported done, and the re-fetch it prompts
// still comes back empty: the catalog answered, and the answer was
// that it has no releases for this group.
emit('AlbumReleasesReady', MBID);
await flush();
await el.updateComplete;
expect(scope(el)).toBe('unavailable');
});
it('goes quiet once the releases actually arrive', async () => {
const el = await coldAlbum();
stub('explore.Service.BrowseReleases', [
{
mbid: 'rel-1',
title: 'Glass Harbour',
date: '2019-04-01',
tracks: [
{
position: 1,
discNumber: 1,
title: 'Track 1',
length: 200000,
mbid: 'rec-1',
inLibrary: false,
},
],
},
]);
emit('AlbumReleasesReady', MBID);
await flush();
await el.updateComplete;
// `catalog` is the silent scope — the notice renders nothing.
expect(scope(el)).toBe('catalog');
});
});
/**
* The album you already own in full.
*
* Identity comes from the MBID and the tracklist from the files' own
* "5/12" denominators, so between them there is nothing left for a
* browse to answer — and the browse was the expensive part, waiting on
* a 1 req/s limiter behind up to eight queued prefetches.
*/
describe('an album the library already holds in full', () => {
beforeEach(() => {
resetHarness();
stub('explore.Service.BrowseReleases', []);
stub('explore.Service.LookupReleaseGroup', {
mbid: MBID,
title: 'Glass Harbour',
artistCredit: 'Tideline',
});
stub('explore.Service.GetThumbnail', '');
stub('library.Library.GetAllLibrariesWithTrackCounts', []);
stub('library.Library.GetFilePathsByRecordingMBIDs', {});
// A local album's rows carry their file paths, and that is what
// "the library holds this" means now — the page records them as it
// maps the tracks, so nothing has to be asked again later.
stub('library.Library.GetAlbumTracks', [
{
TrackName: 'Track 1',
TrackNumber: 1,
DiscNumber: 1,
TrackLength: '3:20',
FilePath: '/music/glass-harbour/01.mp3',
},
]);
});
it('never asks the catalog', async () => {
stub('library.Library.GetAlbumCompleteness', {
owned: 12,
expected: 12,
known: true,
complete: true,
});
const el = await fixture<LitElement>('explore-album-details', {
releaseGroupMBID: MBID,
localAlbumId: 7,
albumName: 'Glass Harbour',
});
await flush();
await el.updateComplete;
expect(calls('explore.Service.BrowseReleases')).toHaveLength(0);
// And says nothing about it, because nothing is missing.
expect(scope(el)).toBe('catalog');
});
it('still asks when tracks are missing', async () => {
stub('library.Library.GetAlbumCompleteness', {
owned: 9,
expected: 12,
known: true,
complete: false,
});
const el = await fixture<LitElement>('explore-album-details', {
releaseGroupMBID: MBID,
localAlbumId: 7,
albumName: 'Glass Harbour',
});
await flush();
await el.updateComplete;
expect(calls('explore.Service.BrowseReleases').length).toBeGreaterThan(0);
});
it('still asks when the tags never declared a total', async () => {
// Unknown is not incomplete. The catalog is the only way to learn
// the total here, so this is exactly when it is worth asking.
stub('library.Library.GetAlbumCompleteness', {
owned: 9,
expected: 0,
known: false,
complete: false,
});
const el = await fixture<LitElement>('explore-album-details', {
releaseGroupMBID: MBID,
localAlbumId: 7,
albumName: 'Glass Harbour',
});
await flush();
await el.updateComplete;
expect(calls('explore.Service.BrowseReleases').length).toBeGreaterThan(0);
});
it('marks a partly-held album with a ring, and a full one with a tick', async () => {
stub('library.Library.GetAlbumCompleteness', {
owned: 9,
expected: 12,
known: true,
complete: false,
});
const el = await fixture<LitElement>('explore-album-details', {
releaseGroupMBID: MBID,
localAlbumId: 7,
albumName: 'Glass Harbour',
});
await flush();
await el.updateComplete;
const badge = shadow(el, 'library-status-indicator');
expect(badge?.getAttribute('status')).toBe('partial');
});
/**
* The denominator the files could not supply.
*
* A great deal of any library declares no track total at all, and
* "unknown" is a third state that must render as neither complete nor
* incomplete — so an album like this used to wear a plain tick no
* matter how much of it was missing. The catalog carries a per-
* release-group total in the artifact for about two bytes a row, and
* that is what fills the gap: the numerator stays local (how many
* distinct track numbers are on disk), only the denominator is
* borrowed.
*/
it('borrows the catalog total when the tags declared none', async () => {
stub('explore.Service.LookupReleaseGroup', {
mbid: MBID,
title: 'Glass Harbour',
artistCredit: 'Tideline',
totalTracks: 12,
});
stub('library.Library.GetAlbumCompleteness', {
owned: 9,
expected: 0,
known: false,
complete: false,
});
const el = await fixture<LitElement>('explore-album-details', {
releaseGroupMBID: MBID,
localAlbumId: 7,
albumName: 'Glass Harbour',
});
await flush();
await el.updateComplete;
const badge = shadow(el, 'library-status-indicator');
expect(badge?.getAttribute('status')).toBe('partial');
expect((badge as unknown as { expected: number }).expected).toBe(12);
expect((badge as unknown as { owned: number }).owned).toBe(9);
});
/**
* And when neither side can total it, nothing is invented: zero means
* "the catalog does not say", which is the same third state the local
* answer has, so the badge stays a plain tick.
*/
it('draws no ring when neither the tags nor the catalog say', async () => {
stub('library.Library.GetAlbumCompleteness', {
owned: 9,
expected: 0,
known: false,
complete: false,
});
const el = await fixture<LitElement>('explore-album-details', {
releaseGroupMBID: MBID,
localAlbumId: 7,
albumName: 'Glass Harbour',
});
await flush();
await el.updateComplete;
expect(
shadow(el, 'library-status-indicator')?.getAttribute('status'),
).toBe('in-library');
});
});