minstrel/internal/library/duplicate_merge_test.go

package library

import (
	"context"
	"errors"
	"io/fs"
	"os"
	"path/filepath"
	"syscall"
	"testing"
	"time"

	"github.com/jackc/pgx/v5/pgtype"
	"github.com/jackc/pgx/v5/pgxpool"

	"git.fabledsword.com/bvandeusen/minstrel/internal/db/dbq"
	"git.fabledsword.com/bvandeusen/minstrel/internal/dbtest"
)

// mergeFixture is a library with one duplicate pair carrying history on both
// copies, and a neighbour track for similarity edges.
type mergeFixture struct {
	pool                  *pgxpool.Pool
	keep, remove, other   dbq.Track
	keepPath, removePath  string
	groupID               pgtype.UUID
	alice, bob            dbq.User
	playlistID            pgtype.UUID
	removePlaylistPos     int32
	aliceEarlierLikeOnRem time.Time
}

func newMergeFixture(t *testing.T) mergeFixture {
	t.Helper()
	pool := newPool(t)
	ctx := context.Background()
	q := dbq.New(pool)
	dir := t.TempDir()
	f := mergeFixture{pool: pool}

	f.keepPath = filepath.Join(dir, "keep.flac")
	f.removePath = filepath.Join(dir, "remove.mp3")
	for _, p := range []string{f.keepPath, f.removePath} {
		if err := os.WriteFile(p, []byte("audio"), 0o644); err != nil {
			t.Fatalf("write %s: %v", p, err)
		}
	}
	var album dbq.Album
	var artist dbq.Artist
	f.keep, album, artist = seedTrack(t, pool, f.keepPath)
	upsert := func(title, path string) dbq.Track {
		t.Helper()
		tr, err := q.UpsertTrack(ctx, dbq.UpsertTrackParams{
			Title: title, AlbumID: album.ID, ArtistID: artist.ID,
			DurationMs: 215000, FilePath: path, FileSize: 100, FileFormat: "mp3",
		})
		if err != nil {
			t.Fatalf("track %s: %v", title, err)
		}
		return tr
	}
	f.remove = upsert("WWW (copy)", f.removePath)
	f.other = upsert("Neighbour", filepath.Join(dir, "other.mp3"))

	mustExec := func(sql string, args ...any) {
		t.Helper()
		if _, err := pool.Exec(ctx, sql, args...); err != nil {
			t.Fatalf("exec %q: %v", sql, err)
		}
	}
	// Only the copy being removed carries a recording MBID.
	mustExec(`UPDATE tracks SET mbid = 'rec-www' WHERE id = $1`, f.remove.ID)

	user := func(name string) dbq.User {
		t.Helper()
		u, err := q.CreateUser(ctx, dbq.CreateUserParams{
			Username: dbtest.TestUserPrefix + name, PasswordHash: "x", ApiTokenHash: name + "-merge-token",
		})
		if err != nil {
			t.Fatalf("user %s: %v", name, err)
		}
		return u
	}
	f.alice, f.bob = user("merge-alice"), user("merge-bob")

	// Alice liked both copies, the removed one first; Bob liked only the removed one.
	f.aliceEarlierLikeOnRem = time.Now().Add(-72 * time.Hour).UTC().Truncate(time.Microsecond)
	mustExec(`INSERT INTO general_likes (user_id, track_id, liked_at) VALUES ($1, $2, $3), ($1, $4, now()), ($5, $2, now())`,
		f.alice.ID, f.remove.ID, f.aliceEarlierLikeOnRem, f.keep.ID, f.bob.ID)

	now := pgtype.Timestamptz{Time: time.Now(), Valid: true}
	session, err := q.InsertPlaySession(ctx, dbq.InsertPlaySessionParams{UserID: f.alice.ID, StartedAt: now})
	if err != nil {
		t.Fatalf("session: %v", err)
	}
	for _, track := range []dbq.Track{f.remove, f.remove, f.keep} {
		if _, err := q.InsertPlayEvent(ctx, dbq.InsertPlayEventParams{
			UserID: f.alice.ID, TrackID: track.ID, SessionID: session.ID, StartedAt: now,
		}); err != nil {
			t.Fatalf("play event: %v", err)
		}
	}
	if _, err := q.InsertSkipEvent(ctx, dbq.InsertSkipEventParams{
		UserID: f.alice.ID, TrackID: f.remove.ID, SessionID: session.ID, SkippedAt: now, PositionMs: 1000,
	}); err != nil {
		t.Fatalf("skip event: %v", err)
	}

	pl, err := q.CreatePlaylist(ctx, dbq.CreatePlaylistParams{UserID: f.alice.ID, Name: "merge-mix"})
	if err != nil {
		t.Fatalf("playlist: %v", err)
	}
	f.playlistID = pl.ID
	entry, err := q.AppendPlaylistTrack(ctx, dbq.AppendPlaylistTrackParams{PlaylistID: pl.ID, TrackID: f.remove.ID})
	if err != nil {
		t.Fatalf("playlist entry: %v", err)
	}
	f.removePlaylistPos = entry.Position

	mustExec(`INSERT INTO track_tags (track_id, tag, weight) VALUES ($1, 'j-pop', 1), ($1, 'house', 0.5), ($2, 'house', 0.9)`,
		f.remove.ID, f.keep.ID)
	mustExec(`INSERT INTO track_similarity (track_a_id, track_b_id, score, source) VALUES
		($1, $3, 0.8, 'listenbrainz'),
		($2, $3, 0.7, 'listenbrainz'),
		($1, $2, 0.99, 'listenbrainz'),
		($3, $1, 0.6, 'musicbrainz_tag')`, f.remove.ID, f.keep.ID, f.other.ID)

	if err := pool.QueryRow(ctx,
		`INSERT INTO duplicate_groups (member_key, tier) VALUES ('merge-fixture', 'exact') RETURNING id`,
	).Scan(&f.groupID); err != nil {
		t.Fatalf("group: %v", err)
	}
	mustExec(`INSERT INTO duplicate_group_members (group_id, track_id) VALUES ($1, $2), ($1, $3)`,
		f.groupID, f.keep.ID, f.remove.ID)
	return f
}

func (f mergeFixture) count(t *testing.T, sql string, args ...any) int {
	t.Helper()
	var n int
	if err := f.pool.QueryRow(context.Background(), sql, args...).Scan(&n); err != nil {
		t.Fatalf("count %q: %v", sql, err)
	}
	return n
}

// The #3911 proof: after a merge, every piece of history the removed copy held
// is on the copy kept, nothing is doubled, and the removed copy — row and file —
// is gone.
func TestMergeDuplicateGroup_Integration(t *testing.T) {
	f := newMergeFixture(t)
	ctx := context.Background()

	res, err := MergeDuplicateGroup(ctx, f.pool, nil, "", f.groupID, f.keep.ID)
	if err != nil {
		t.Fatalf("merge: %v", err)
	}
	if len(res.Removed) != 1 || res.Removed[0].FilePath != f.removePath || res.Survivor.TrackID != f.keep.ID {
		t.Fatalf("result = %+v, want the removed copy reported and the kept one as survivor", res)
	}
	if res.PlayEvents != 2 || res.SkipEvents != 1 || res.Likes != 2 || res.PlaylistEntries != 1 {
		t.Errorf("moved = plays %d skips %d likes %d playlist %d, want 2, 1, 2, 1",
			res.PlayEvents, res.SkipEvents, res.Likes, res.PlaylistEntries)
	}

	if _, err := os.Stat(f.removePath); !errors.Is(err, os.ErrNotExist) {
		t.Errorf("removed copy's file still on disk: %v", err)
	}
	if _, err := os.Stat(f.keepPath); err != nil {
		t.Errorf("kept copy's file is gone: %v", err)
	}
	if n := f.count(t, `SELECT count(*) FROM tracks WHERE id = $1`, f.remove.ID); n != 0 {
		t.Errorf("removed copy's row still exists")
	}

	// Likes: one per user, Alice's dated to her earlier like.
	if n := f.count(t, `SELECT count(*) FROM general_likes WHERE track_id = $1`, f.keep.ID); n != 2 {
		t.Errorf("likes on the kept copy = %d, want 2 (Alice once, Bob)", n)
	}
	var aliceLiked time.Time
	if err := f.pool.QueryRow(ctx, `SELECT liked_at FROM general_likes WHERE user_id = $1 AND track_id = $2`,
		f.alice.ID, f.keep.ID).Scan(&aliceLiked); err != nil {
		t.Fatalf("alice's like: %v", err)
	}
	if !aliceLiked.Equal(f.aliceEarlierLikeOnRem) {
		t.Errorf("alice's like dated %v, want her earlier like %v", aliceLiked, f.aliceEarlierLikeOnRem)
	}

	// Plays and skips move exactly: none lost, none invented.
	if n := f.count(t, `SELECT count(*) FROM play_events WHERE track_id = $1`, f.keep.ID); n != 3 {
		t.Errorf("plays on the kept copy = %d, want 3", n)
	}
	if n := f.count(t, `SELECT count(*) FROM skip_events WHERE track_id = $1`, f.keep.ID); n != 1 {
		t.Errorf("skips on the kept copy = %d, want 1", n)
	}

	// The playlist entry stays where it was and now plays the kept copy.
	if n := f.count(t, `SELECT count(*) FROM playlist_tracks WHERE playlist_id = $1 AND position = $2 AND track_id = $3`,
		f.playlistID, f.removePlaylistPos, f.keep.ID); n != 1 {
		t.Errorf("playlist entry at position %d does not point at the kept copy", f.removePlaylistPos)
	}

	// Tags are a union; the kept copy's own weight wins where both had the tag.
	if n := f.count(t, `SELECT count(*) FROM track_tags WHERE track_id = $1`, f.keep.ID); n != 2 {
		t.Errorf("tags on the kept copy = %d, want 2 (house, j-pop)", n)
	}
	if n := f.count(t, `SELECT count(*) FROM track_tags WHERE track_id = $1 AND tag = 'house' AND weight = 0.9`, f.keep.ID); n != 1 {
		t.Errorf("the kept copy's own house weight was overwritten")
	}

	// Similarity: rewritten onto the kept copy, no duplicate edge, no self-edge,
	// nothing left pointing at the removed copy.
	if n := f.count(t, `SELECT count(*) FROM track_similarity WHERE track_a_id = $1 AND track_b_id = $2 AND source = 'listenbrainz'`,
		f.keep.ID, f.other.ID); n != 1 {
		t.Errorf("listenbrainz edge keep→other = %d rows, want exactly 1", n)
	}
	if n := f.count(t, `SELECT count(*) FROM track_similarity WHERE track_a_id = $1 AND track_b_id = $2 AND source = 'musicbrainz_tag'`,
		f.other.ID, f.keep.ID); n != 1 {
		t.Errorf("musicbrainz_tag edge other→keep was not carried over")
	}
	if n := f.count(t, `SELECT count(*) FROM track_similarity WHERE track_a_id = track_b_id`); n != 0 {
		t.Errorf("a self-edge was written")
	}

	// The removed copy's recording MBID is inherited; the group is closed.
	if n := f.count(t, `SELECT count(*) FROM tracks WHERE id = $1 AND mbid = 'rec-www'`, f.keep.ID); n != 1 {
		t.Errorf("the kept copy did not inherit the recording MBID")
	}
	if n := f.count(t, `SELECT count(*) FROM duplicate_groups WHERE id = $1 AND status = 'merged' AND resolved_at IS NOT NULL`, f.groupID); n != 1 {
		t.Errorf("group was not marked merged")
	}

	// A second merge of the same group is refused rather than repeated.
	if _, err := MergeDuplicateGroup(ctx, f.pool, nil, "", f.groupID, f.keep.ID); !errors.Is(err, ErrDuplicateGroupNotPending) {
		t.Errorf("second merge err = %v, want ErrDuplicateGroupNotPending", err)
	}
}

// When the removed copy's file cannot go, nothing may change: its likes, plays
// and row stay exactly where they were, and the group stays pending.
func TestMergeDuplicateGroup_UnremovableFileChangesNothing(t *testing.T) {
	f := newMergeFixture(t)
	stubRemoveFile(t, func(name string) error {
		return &fs.PathError{Op: "remove", Path: name, Err: syscall.EROFS}
	})

	_, err := MergeDuplicateGroup(context.Background(), f.pool, nil, "", f.groupID, f.keep.ID)
	var fre *FileRemoveError
	if !errors.As(err, &fre) || !fre.NotWritable() {
		t.Fatalf("err = %v, want a not-writable *FileRemoveError", err)
	}
	if n := f.count(t, `SELECT count(*) FROM tracks WHERE id = $1`, f.remove.ID); n != 1 {
		t.Errorf("the copy's row was deleted although its file was not")
	}
	if n := f.count(t, `SELECT count(*) FROM general_likes WHERE track_id = $1`, f.remove.ID); n != 2 {
		t.Errorf("likes on the copy = %d, want both still there", n)
	}
	if n := f.count(t, `SELECT count(*) FROM play_events WHERE track_id = $1`, f.remove.ID); n != 2 {
		t.Errorf("plays on the copy = %d, want both still there", n)
	}
	if n := f.count(t, `SELECT count(*) FROM duplicate_groups WHERE id = $1 AND status = 'pending'`, f.groupID); n != 1 {
		t.Errorf("group left pending = false, want it still pending")
	}
}

func TestMergeDuplicateGroup_SurvivorMustBeAMember(t *testing.T) {
	f := newMergeFixture(t)
	var stranger pgtype.UUID
	stranger.Bytes[15], stranger.Valid = 0xEE, true

	_, err := MergeDuplicateGroup(context.Background(), f.pool, nil, "", f.groupID, stranger)
	if !errors.Is(err, ErrSurvivorNotInGroup) {
		t.Fatalf("err = %v, want ErrSurvivorNotInGroup", err)
	}
	if _, err := os.Stat(f.removePath); err != nil {
		t.Errorf("a refused merge removed a file: %v", err)
	}
	if n := f.count(t, `SELECT count(*) FROM duplicate_groups WHERE id = $1 AND status = 'pending'`, f.groupID); n != 1 {
		t.Errorf("a refused merge changed the group")
	}
}