Files
yellowjacket/backend/database/search_test.go
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

948 lines
23 KiB
Go

package database
import (
"fmt"
"testing"
"yellowjacket/backend/database/sql/sqlcgen"
)
// seedSearchData inserts ~7 tracks with the full FK chain required for
// FTS5 search tests: artist_credit → recordings → audio_files →
// release_groups → release_group_recordings → search_index.
//
// Track list:
//
// ID 1: "Bohemian Rhapsody" by "Queen" on "A Night at the Opera"
// ID 2: "Halo" by "Beyoncé" on "Lemonade"
// ID 3: "Back in Black" by "AC/DC" on "Back in Black"
// ID 4: "Comfortably Numb" by "Pink Floyd" on "The Dark Side of the Moon"
// ID 5: "Another One Bites the Dust" by "Queen" on "The Game"
// ID 6: "Thunderstruck" by "AC/DC" on "The Razors Edge"
// ID 7: "Queen of the Stone Age" by "Queens of the Stone Age" on "Rated R"
func seedSearchData(t *testing.T, db *DB) {
t.Helper()
type track struct {
id int64
filePath string
title string
artist string // artist_credit text
album string // release_group name
trackNum *int64 // recording track_number (nil = NULL)
discNum *int64 // recording disc_number (nil = NULL)
year int64 // recording year
genre string // genre name (empty = no genre)
composer string // recording composer
lenMs int64 // audio_files length_milliseconds
ftID int64 // file_type_id
sr int64 // sample_rate
bd int64 // bit_depth
ch int64 // channels
br int64 // bitrate
fsize int64 // file_size
}
intPtr := func(v int64) *int64 { return &v }
tracks := []track{
{
1, "/music/queen/bohemian_rhapsody.mp3", "Bohemian Rhapsody", "Queen",
"A Night at the Opera", intPtr(11), intPtr(1), 1975, "Rock",
"Freddie Mercury", 354000, 0, 44100, 16, 2, 320000, 8500000,
},
{
2, "/music/beyonce/halo.flac", "Halo", "Beyoncé", "Lemonade",
intPtr(1), intPtr(1), 2008, "Pop", "Ryan Tedder", 261000, 1,
96000, 24, 2, 1411000, 42000000,
},
{
3, "/music/acdc/back_in_black.mp3", "Back in Black", "AC/DC",
"Back in Black", intPtr(1), intPtr(1), 1980, "Hard Rock",
"Angus Young", 255000, 0, 44100, 16, 2, 320000, 6100000,
},
{
4, "/music/pinkfloyd/comfortably_numb.flac", "Comfortably Numb",
"Pink Floyd", "The Dark Side of the Moon", intPtr(6), intPtr(1),
1979, "Progressive Rock", "David Gilmour", 382000, 1, 96000, 24,
2, 1411000, 54000000,
},
{
5, "/music/queen/another_one_bites_the_dust.mp3",
"Another One Bites the Dust", "Queen", "The Game", intPtr(3),
intPtr(1), 1980, "Funk Rock", "John Deacon", 215000, 0, 44100,
16, 2, 320000, 5200000,
},
{
6, "/music/acdc/thunderstruck.mp3", "Thunderstruck", "AC/DC",
"The Razors Edge", intPtr(1), intPtr(1), 1990, "Hard Rock",
"Angus Young", 292000, 0, 44100, 16, 2, 320000, 7000000,
},
{
7, "/music/qotsa/queen_of_the_stone_age.mp3",
"Queen of the Stone Age", "Queens of the Stone Age", "Rated R",
intPtr(1), intPtr(1), 2000, "Stoner Rock", "Josh Homme", 310000,
0, 44100, 16, 2, 320000, 7400000,
},
}
for _, tr := range tracks {
var genres []string
if tr.genre != "" {
genres = []string{tr.genre}
}
var trackNum, discNum int64
if tr.trackNum != nil {
trackNum = *tr.trackNum
}
if tr.discNum != nil {
discNum = *tr.discNum
}
id := InsertTestTrack(t, db, TestTrack{
FilePath: tr.filePath,
Title: tr.title,
Artist: tr.artist,
Album: tr.album,
Genres: genres,
TrackNumber: trackNum,
DiscNumber: discNum,
Year: tr.year,
LengthMs: tr.lenMs,
})
// The fixtures assert on audio properties and the composer,
// which InsertTestTrack does not carry - they are not part of
// what a seeder should have to know about a track.
if _, err := db.ExecContext(
`UPDATE audio_files
SET file_type_id = ?, sample_rate = ?, bit_depth = ?,
channels = ?, bitrate = ?, file_size = ?, composer = ?
WHERE id = ?`,
tr.ftID, tr.sr, tr.bd, tr.ch, tr.br, tr.fsize, tr.composer, id,
); err != nil {
t.Fatalf("set audio properties for %q: %v", tr.filePath, err)
}
}
}
// ---------------------------------------------------------------------------
// Pure helper tests (no database needed)
// ---------------------------------------------------------------------------
func TestTokeniseForFTS(t *testing.T) {
t.Parallel()
tests := []struct {
name string
input string
want []string
}{
{"simple word", "hello", []string{`"hello"`}},
{"multiple words", "hello world", []string{`"hello"`, `"world"`}},
{"hyphens split", "rock-pop", []string{`"rock"`, `"pop"`}},
{"slashes split", "AC/DC", []string{`"AC"`, `"DC"`}},
{"dots split", "01.track", []string{`"01"`, `"track"`}},
{"underscores split", "my_song", []string{`"my"`, `"song"`}},
{
"double quotes escaped",
`he"llo`,
[]string{`"he""llo"`},
},
{"empty string", "", nil},
{"only separators", "---", nil},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
t.Parallel()
got := tokeniseForFTS(tt.input)
if len(got) != len(tt.want) {
t.Fatalf(
"tokeniseForFTS(%q): got %d tokens %v, want %d tokens %v",
tt.input, len(got), got, len(tt.want), tt.want,
)
}
for i := range got {
if got[i] != tt.want[i] {
t.Errorf(
"tokeniseForFTS(%q)[%d] = %q, want %q",
tt.input, i, got[i], tt.want[i],
)
}
}
})
}
}
func TestBuildFTSQuery(t *testing.T) {
t.Parallel()
tests := []struct {
name string
input string
want string
}{
{"single word", "queen", `"queen"`},
{"multi-word", "bohemian rhapsody", `"bohemian" "rhapsody"`},
{"empty string returns original", "", ""},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
t.Parallel()
got := buildFTSQuery(tt.input)
if got != tt.want {
t.Errorf(
"buildFTSQuery(%q) = %q, want %q",
tt.input, got, tt.want,
)
}
})
}
}
func TestStripExtForSearch(t *testing.T) {
t.Parallel()
tests := []struct {
name string
input string
want string
}{
{"mp3 extension", "song.mp3", "song"},
{"double dot", "my.song.flac", "my.song"},
{"no extension", "noextension", "noextension"},
{"hidden file", ".hidden", ".hidden"},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
t.Parallel()
got := stripExtForSearch(tt.input)
if got != tt.want {
t.Errorf(
"stripExtForSearch(%q) = %q, want %q",
tt.input, got, tt.want,
)
}
})
}
}
// ---------------------------------------------------------------------------
// FTS5 search tests (require database + seeded data)
// ---------------------------------------------------------------------------
func TestSearchFTS_BasicTerm(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
seedSearchData(t, db)
results, err := db.SearchFTS("queen", 10)
if err != nil {
t.Fatalf("SearchFTS(queen): %v", err)
}
// Should find at least "Bohemian Rhapsody" and "Another One Bites the
// Dust" (artist=Queen) plus "Queen of the Stone Age" (title match).
if len(results) < 2 {
t.Fatalf("SearchFTS(queen): got %d results, want >= 2", len(results))
}
// Verify we got the expected Queen tracks by collecting titles.
titles := map[string]bool{}
for _, r := range results {
titles[r.Title] = true
}
for _, want := range []string{"Bohemian Rhapsody", "Another One Bites the Dust"} {
if !titles[want] {
t.Errorf("SearchFTS(queen): missing expected title %q in results %v",
want, titles)
}
}
}
func TestSearchFTS_EmptyQuery(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
seedSearchData(t, db)
// Empty string.
results, err := db.SearchFTS("", 10)
if err != nil {
t.Fatalf("SearchFTS(empty): %v", err)
}
if results != nil {
t.Errorf("SearchFTS(empty): got %v, want nil", results)
}
// Whitespace-only.
results, err = db.SearchFTS(" ", 10)
if err != nil {
t.Fatalf("SearchFTS(whitespace): %v", err)
}
if results != nil {
t.Errorf("SearchFTS(whitespace): got %v, want nil", results)
}
}
func TestSearchFTS_SpecialCharacters(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
seedSearchData(t, db)
// "AC/DC" — the tokeniser splits on '/', so "AC" and "DC" both become
// search tokens and match the AC/DC artist in the index.
results, err := db.SearchFTS("AC/DC", 10)
if err != nil {
t.Fatalf("SearchFTS(AC/DC): %v", err)
}
if len(results) < 1 {
t.Fatalf("SearchFTS(AC/DC): got 0 results, want >= 1")
}
// Verify at least one AC/DC track is present.
found := false
for _, r := range results {
if r.Artist == "AC/DC" {
found = true
break
}
}
if !found {
t.Errorf("SearchFTS(AC/DC): no results with Artist='AC/DC'")
}
// Query with embedded double quote — should not error.
results, err = db.SearchFTS(`back"in`, 10)
if err != nil {
t.Fatalf("SearchFTS(quote): %v", err)
}
// We don't assert exact results for the quote test, just no error.
_ = results
}
func TestSearchFTS_MultiWord(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
seedSearchData(t, db)
results, err := db.SearchFTS("bohemian rhapsody", 10)
if err != nil {
t.Fatalf("SearchFTS(multi-word): %v", err)
}
if len(results) == 0 {
t.Fatal("SearchFTS(bohemian rhapsody): got 0 results")
}
// Top result should be the exact title match.
if results[0].Title != "Bohemian Rhapsody" {
t.Errorf(
"SearchFTS(bohemian rhapsody): top result Title = %q, want %q",
results[0].Title, "Bohemian Rhapsody",
)
}
}
func TestSearchFTS_Diacritics(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
seedSearchData(t, db)
// Search without diacritic — should find "Beyoncé" due to
// unicode61 remove_diacritics 2 tokeniser configuration.
results, err := db.SearchFTS("Beyonce", 10)
if err != nil {
t.Fatalf("SearchFTS(Beyonce): %v", err)
}
if len(results) == 0 {
t.Fatal("SearchFTS(Beyonce): got 0 results, want Beyoncé track")
}
found := false
for _, r := range results {
if r.Artist == "Beyoncé" {
found = true
break
}
}
if !found {
t.Error("SearchFTS(Beyonce): no result with Artist='Beyoncé'")
}
}
func TestSearchFTS_Ranking(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
seedSearchData(t, db)
// "Back in Black" appears as both title AND album for track ID 3,
// so it should rank higher than tracks where "black" only appears
// in one column.
results, err := db.SearchFTS("back in black", 10)
if err != nil {
t.Fatalf("SearchFTS(ranking): %v", err)
}
if len(results) == 0 {
t.Fatal("SearchFTS(back in black): got 0 results")
}
// First result should be the "Back in Black" track (title + album match).
if results[0].Title != "Back in Black" {
t.Errorf(
"SearchFTS(ranking): top result = %q by %q, want %q",
results[0].Title, results[0].Artist, "Back in Black",
)
}
}
func TestSearchFTSByFilename(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
seedSearchData(t, db)
// Search by basename — extension is stripped, underscores split.
results, err := db.SearchFTSByFilename("bohemian_rhapsody.mp3", 10)
if err != nil {
t.Fatalf("SearchFTSByFilename: %v", err)
}
if len(results) == 0 {
t.Fatal("SearchFTSByFilename(bohemian_rhapsody.mp3): got 0 results")
}
found := false
for _, r := range results {
if r.Title == "Bohemian Rhapsody" {
found = true
break
}
}
if !found {
t.Error("SearchFTSByFilename: Bohemian Rhapsody not found")
}
// Empty basename.
results, err = db.SearchFTSByFilename("", 10)
if err != nil {
t.Fatalf("SearchFTSByFilename(empty): %v", err)
}
if results != nil {
t.Errorf("SearchFTSByFilename(empty): got %v, want nil", results)
}
}
func TestSearchFTSTracks(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
seedSearchData(t, db)
results, err := db.SearchFTSTracks("queen", 0, 10)
if err != nil {
t.Fatalf("SearchFTSTracks: %v", err)
}
if len(results) == 0 {
t.Fatal("SearchFTSTracks(queen): got 0 results")
}
// Find the Bohemian Rhapsody result and verify all 16 fields.
var br *sqlcgen.TrackMetadatum
for i, r := range results {
if r.Title == "Bohemian Rhapsody" {
br = &results[i]
break
}
}
if br == nil {
t.Fatal("SearchFTSTracks: Bohemian Rhapsody not found")
return
}
// Verify all fields are populated.
checks := []struct {
field string
got any
want any
}{
{"FilePath", br.FilePath, "/music/queen/bohemian_rhapsody.mp3"},
{"LengthMilliseconds", br.LengthMilliseconds, int64(354000)},
{"Title", br.Title, "Bohemian Rhapsody"},
{"ArtistName", br.ArtistName, "Queen"},
{"Album", br.Album, "A Night at the Opera"},
{"Year", br.Year, int64(1975)},
{"Composer", br.Composer, "Freddie Mercury"},
{"SampleRate", br.SampleRate, int64(44100)},
{"BitDepth", br.BitDepth, int64(16)},
{"Channels", br.Channels, int64(2)},
{"Bitrate", br.Bitrate, int64(320000)},
{"FileSize", br.FileSize, int64(8500000)},
}
for _, c := range checks {
if fmt.Sprintf("%v", c.got) != fmt.Sprintf("%v", c.want) {
t.Errorf("SearchFTSTracks: %s = %v, want %v", c.field, c.got, c.want)
}
}
// TrackNumber and DiscNumber are sql.NullInt64.
if !br.TrackNumber.Valid || br.TrackNumber.Int64 != 11 {
t.Errorf("SearchFTSTracks: TrackNumber = %v, want 11", br.TrackNumber)
}
if !br.DiscNumber.Valid || br.DiscNumber.Int64 != 1 {
t.Errorf("SearchFTSTracks: DiscNumber = %v, want 1", br.DiscNumber)
}
// Genre (via recording_genres + genres tables GROUP_CONCAT).
if br.Genre != "Rock" {
t.Errorf("SearchFTSTracks: Genre = %q, want %q", br.Genre, "Rock")
}
// FileType (from file_types table, id=0 → ".mp3").
if br.FileType != ".mp3" {
t.Errorf("SearchFTSTracks: FileType = %q, want %q", br.FileType, ".mp3")
}
}
// ---------------------------------------------------------------------------
// Search index operation tests
// ---------------------------------------------------------------------------
func TestInsertAndDeleteSearchIndex(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
// Set up minimal FK chain for a single track.
InsertTestTrack(t, db, TestTrack{
FilePath: "/test/track.mp3",
Title: "Test Track",
Artist: "Test Artist",
LengthMs: 180000,
})
// Insert into search index.
if err := db.InsertSearchIndex(
1, "/test/track.mp3", "Test Track", "Test Artist", "Test Album",
); err != nil {
t.Fatalf("InsertSearchIndex: %v", err)
}
// Verify it's findable.
results, err := db.SearchFTS("Test Track", 10)
if err != nil {
t.Fatalf("SearchFTS after insert: %v", err)
}
if len(results) == 0 {
t.Fatal("SearchFTS after insert: got 0 results")
}
// DeleteSearchIndex now works with contentless_delete=1.
err = db.DeleteSearchIndex(1)
if err != nil {
t.Fatalf("DeleteSearchIndex: %v", err)
}
// Verify the deleted row is no longer findable.
results, err = db.SearchFTS("Test Track", 10)
if err != nil {
t.Fatalf("SearchFTS after delete: %v", err)
}
if len(results) != 0 {
t.Fatalf(
"SearchFTS after delete: got %d results, want 0",
len(results),
)
}
}
func TestRebuildSearchIndex(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
// Seed the file WITHOUT putting it in search_index.
InsertTestTrack(t, db, TestTrack{
FilePath: "/rebuild/track.mp3",
Title: "Rebuild Track",
Artist: "Rebuild Artist",
Album: "Rebuild Album",
LengthMs: 200000,
SkipSearchIndex: true,
})
// Search should return nothing before rebuild.
results, err := db.SearchFTS("Rebuild", 10)
if err != nil {
t.Fatalf("SearchFTS before rebuild: %v", err)
}
if len(results) != 0 {
t.Fatalf("SearchFTS before rebuild: got %d results, want 0", len(results))
}
// Rebuild search index.
if err := db.RebuildSearchIndex(); err != nil {
t.Fatalf("RebuildSearchIndex: %v", err)
}
// Search should now return the track.
results, err = db.SearchFTS("Rebuild", 10)
if err != nil {
t.Fatalf("SearchFTS after rebuild: %v", err)
}
if len(results) == 0 {
t.Fatal("SearchFTS after rebuild: got 0 results, want >= 1")
}
if results[0].Title != "Rebuild Track" {
t.Errorf(
"SearchFTS after rebuild: Title = %q, want %q",
results[0].Title, "Rebuild Track",
)
}
if results[0].Album != "Rebuild Album" {
t.Errorf(
"SearchFTS after rebuild: Album = %q, want %q",
results[0].Album, "Rebuild Album",
)
}
}
func TestClearSearchIndex(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
seedSearchData(t, db)
// Verify data exists.
results, err := db.SearchFTS("queen", 10)
if err != nil {
t.Fatalf("SearchFTS before clear: %v", err)
}
if len(results) == 0 {
t.Fatal("SearchFTS before clear: got 0 results")
}
// ClearSearchIndex drops and recreates the contentless FTS5
// table, which is the only way to clear a content='' table.
err = db.ClearSearchIndex()
if err != nil {
t.Fatalf("ClearSearchIndex: %v", err)
}
// Verify the index is empty after clear.
results, err = db.SearchFTS("queen", 10)
if err != nil {
t.Fatalf("SearchFTS after clear: %v", err)
}
if len(results) != 0 {
t.Fatalf("SearchFTS after clear: got %d results, want 0", len(results))
}
}
// ---------------------------------------------------------------------------
// FTS5 row deletion and update cycle tests
// ---------------------------------------------------------------------------
func TestDeleteSearchIndex(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
seedSearchData(t, db) // 7 tracks
tests := []struct {
name string
rowid int64
searchTerm string
expectEmpty bool // true = search should return 0 results after delete
}{
{
name: "delete existing rowid removes it from search",
rowid: 1, // "Bohemian Rhapsody"
searchTerm: "Bohemian Rhapsody",
expectEmpty: true,
},
{
name: "delete non-existent rowid is a no-op",
rowid: 9999,
searchTerm: "queen",
expectEmpty: false, // other Queen tracks still present
},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
err := db.DeleteSearchIndex(tt.rowid)
if err != nil {
t.Fatalf(
"DeleteSearchIndex(%d): %v",
tt.rowid, err,
)
}
results, err := db.SearchFTS(tt.searchTerm, 10)
if err != nil {
t.Fatalf("SearchFTS(%q): %v", tt.searchTerm, err)
}
if tt.expectEmpty && len(results) != 0 {
t.Errorf(
"SearchFTS(%q) after delete: got %d results, want 0",
tt.searchTerm, len(results),
)
}
if !tt.expectEmpty && len(results) == 0 {
t.Errorf(
"SearchFTS(%q) after delete: got 0 results, want > 0",
tt.searchTerm,
)
}
})
}
// Verify other tracks are unaffected — "Thunderstruck" should
// still be searchable after deleting rowid 1.
results, err := db.SearchFTS("Thunderstruck", 10)
if err != nil {
t.Fatalf("SearchFTS(Thunderstruck): %v", err)
}
if len(results) == 0 {
t.Error("SearchFTS(Thunderstruck): expected results, got 0")
}
}
func TestSearchIndexUpdateCycle(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
// Set up minimal FK chain for a single track at rowid 100.
id := InsertTestTrack(t, db, TestTrack{
FilePath: "/test/update_cycle.mp3",
Title: "Old Title",
Artist: "Old Artist",
LengthMs: 200000,
SkipSearchIndex: true,
})
// 1. Insert with old metadata.
if err := db.InsertSearchIndex(
id, "/test/update_cycle.mp3", "Old Title", "Old Artist", "Old Album",
); err != nil {
t.Fatalf("InsertSearchIndex (old): %v", err)
}
// Verify search for "Old Title" finds it.
results, err := db.SearchFTS("Old Title", 10)
if err != nil {
t.Fatalf("SearchFTS(Old Title): %v", err)
}
if len(results) == 0 {
t.Fatal("SearchFTS(Old Title): got 0 results after insert")
}
// 2. Delete the row.
if err := db.DeleteSearchIndex(id); err != nil {
t.Fatalf("DeleteSearchIndex(%d): %v", id, err)
}
// Verify "Old Title" no longer found.
results, err = db.SearchFTS("Old Title", 10)
if err != nil {
t.Fatalf("SearchFTS(Old Title) after delete: %v", err)
}
if len(results) != 0 {
t.Fatalf(
"SearchFTS(Old Title) after delete: got %d results, want 0",
len(results),
)
}
// 3. Update the file's title in the DB to simulate a tag edit.
_, err = db.ExecContext(
"UPDATE audio_files SET title = 'New Title' WHERE file_path = '/test/update_cycle.mp3'",
)
if err != nil {
t.Fatalf("update title: %v", err)
}
// 4. Re-insert the row with new metadata.
if err := db.InsertSearchIndex(
id, "/test/update_cycle.mp3", "New Title", "New Artist", "New Album",
); err != nil {
t.Fatalf("InsertSearchIndex (new): %v", err)
}
// 5. Verify "New Title" is now findable.
results, err = db.SearchFTS("New Title", 10)
if err != nil {
t.Fatalf("SearchFTS(New Title): %v", err)
}
if len(results) == 0 {
t.Fatal("SearchFTS(New Title): got 0 results after reinsert")
}
// 6. Verify "Old Title" still returns no results (no ghost entries).
results, err = db.SearchFTS("Old Title", 10)
if err != nil {
t.Fatalf("SearchFTS(Old Title) after reinsert: %v", err)
}
if len(results) != 0 {
t.Fatalf(
"SearchFTS(Old Title) after reinsert: got %d results, want 0 (ghost entry)",
len(results),
)
}
}
func TestClearSearchIndexPreservesSchema(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
seedSearchData(t, db) // 7 tracks
// Verify data exists before clear.
results, err := db.SearchFTS("queen", 10)
if err != nil {
t.Fatalf("SearchFTS before clear: %v", err)
}
if len(results) == 0 {
t.Fatal("SearchFTS before clear: got 0 results")
}
// Clear the search index.
if err := db.ClearSearchIndex(); err != nil {
t.Fatalf("ClearSearchIndex: %v", err)
}
// Verify all data is gone.
results, err = db.SearchFTS("queen", 10)
if err != nil {
t.Fatalf("SearchFTS after clear: %v", err)
}
if len(results) != 0 {
t.Fatalf(
"SearchFTS after clear: got %d results, want 0",
len(results),
)
}
// Re-insert data and verify the recreated table supports
// both insert and delete (contentless_delete=1 preserved).
if err := db.InsertSearchIndex(
1, "/music/queen/bohemian_rhapsody.mp3",
"Bohemian Rhapsody", "Queen", "A Night at the Opera",
); err != nil {
t.Fatalf("InsertSearchIndex after clear: %v", err)
}
results, err = db.SearchFTS("Bohemian", 10)
if err != nil {
t.Fatalf("SearchFTS after reinsert: %v", err)
}
if len(results) == 0 {
t.Fatal("SearchFTS after reinsert: got 0 results")
}
// Verify delete still works on the recreated table.
if err := db.DeleteSearchIndex(1); err != nil {
t.Fatalf("DeleteSearchIndex after clear+reinsert: %v", err)
}
results, err = db.SearchFTS("Bohemian", 10)
if err != nil {
t.Fatalf("SearchFTS after clear+reinsert+delete: %v", err)
}
if len(results) != 0 {
t.Fatalf(
"SearchFTS after clear+reinsert+delete: got %d results, want 0",
len(results),
)
}
}
// ---------------------------------------------------------------------------
// Schema constraint tests
// ---------------------------------------------------------------------------
func TestSearchIndexSchema(t *testing.T) {
t.Parallel()
db := NewTestDB(t)
// Verify the UNIQUE index on artist_credit_artist exists by
// attempting a duplicate insert. First, create the prerequisites.
_, err := db.ExecContext(
"INSERT INTO artists (id, name) VALUES (1, 'Test')",
)
if err != nil {
t.Fatalf("insert artist: %v", err)
}
// The credit tables this used to assert a UNIQUE constraint on are
// gone; a file names its artist directly, and artists are unique by
// name, which is asserted below.
_, err = db.ExecContext(
"INSERT INTO artists (id, name) VALUES (2, 'Test')",
)
if err == nil {
t.Error("duplicate artist name should fail, got nil error")
}
}