Vellum/internal/scanner/scan_test.go

package scanner

import (
	"context"
	"database/sql"
	"io"
	"log/slog"
	"os"
	"path/filepath"
	"testing"

	"github.com/mithral/vellum/internal/core"
	"github.com/mithral/vellum/internal/library"
	"github.com/mithral/vellum/migrations"
)

func testDB(t *testing.T) *sql.DB {
	t.Helper()
	db, err := core.OpenDB(filepath.Join(t.TempDir(), "scan.db"))
	if err != nil {
		t.Fatalf("OpenDB: %v", err)
	}
	t.Cleanup(func() { _ = db.Close() })
	log := slog.New(slog.NewTextHandler(io.Discard, nil))
	if err := core.Migrate(context.Background(), db, migrations.FS, log); err != nil {
		t.Fatalf("Migrate: %v", err)
	}
	return db
}

func testLogger() *slog.Logger {
	return slog.New(slog.NewTextHandler(io.Discard, nil))
}

func writeFile(t *testing.T, path, content string) {
	t.Helper()
	if err := os.MkdirAll(filepath.Dir(path), 0o755); err != nil {
		t.Fatal(err)
	}
	if err := os.WriteFile(path, []byte(content), 0o644); err != nil {
		t.Fatal(err)
	}
}

func TestScanLifecycle(t *testing.T) {
	db := testDB(t)
	ctx := context.Background()
	root := t.TempDir()

	writeFile(t, filepath.Join(root, "a.epub"), "aaaaaaaaaa")
	writeFile(t, filepath.Join(root, "b.epub"), "bbbbbbbbbbbbbbbbbbbb")
	writeFile(t, filepath.Join(root, "sub", "c.epub"), "cccccccccccccccccccccccccccccc")
	writeFile(t, filepath.Join(root, "notes.txt"), "not media")
	writeFile(t, filepath.Join(root, ".hidden.epub"), "hidden")
	writeFile(t, filepath.Join(root, ".git", "d.epub"), "in a hidden dir")

	lib, err := library.Create(ctx, db, "Ebooks", "ebooks", root)
	if err != nil {
		t.Fatalf("Create: %v", err)
	}

	// First scan: three supported files, nothing else.
	res, err := Scan(ctx, db, lib, testLogger())
	if err != nil {
		t.Fatalf("Scan: %v", err)
	}
	if res.Added != 3 || res.Scanned != 3 {
		t.Fatalf("first scan added=%d scanned=%d, want 3/3 (%+v)", res.Added, res.Scanned, res)
	}
	if n, _ := library.CountItems(ctx, db, lib.ID); n != 3 {
		t.Errorf("item count = %d, want 3", n)
	}

	// Second scan: nothing changed.
	res, err = Scan(ctx, db, lib, testLogger())
	if err != nil {
		t.Fatal(err)
	}
	if res.Unchanged != 3 || res.Added != 0 || res.Updated != 0 {
		t.Errorf("second scan unchanged=%d added=%d updated=%d, want 3/0/0", res.Unchanged, res.Added, res.Updated)
	}

	// Modify a.epub (different size).
	writeFile(t, filepath.Join(root, "a.epub"), "aaaaaaaaaaaaaaaaaaaaaaaaaaaa")
	res, err = Scan(ctx, db, lib, testLogger())
	if err != nil {
		t.Fatal(err)
	}
	if res.Updated != 1 || res.Unchanged != 2 {
		t.Errorf("after modify updated=%d unchanged=%d, want 1/2", res.Updated, res.Unchanged)
	}

	// Delete b.epub: it goes missing, it is not deleted.
	if err := os.Remove(filepath.Join(root, "b.epub")); err != nil {
		t.Fatal(err)
	}
	res, err = Scan(ctx, db, lib, testLogger())
	if err != nil {
		t.Fatal(err)
	}
	if res.Missing != 1 {
		t.Errorf("after delete missing=%d, want 1", res.Missing)
	}
	if n, _ := library.CountItems(ctx, db, lib.ID); n != 2 {
		t.Errorf("item count (excluding missing) = %d, want 2", n)
	}

	// The row still exists, flagged missing.
	var missing int
	if err := db.QueryRowContext(ctx,
		`SELECT missing FROM media_items WHERE library_id = ? AND path = ?`, lib.ID, "b.epub").Scan(&missing); err != nil {
		t.Fatalf("missing item row was deleted: %v", err)
	}
	if missing != 1 {
		t.Errorf("missing flag = %d, want 1", missing)
	}

	// Restore b.epub.
	writeFile(t, filepath.Join(root, "b.epub"), "bbbbbbbbbbbbbbbbbbbb")
	res, err = Scan(ctx, db, lib, testLogger())
	if err != nil {
		t.Fatal(err)
	}
	if res.Restored != 1 {
		t.Errorf("after restore restored=%d, want 1 (%+v)", res.Restored, res)
	}
	if n, _ := library.CountItems(ctx, db, lib.ID); n != 3 {
		t.Errorf("item count after restore = %d, want 3", n)
	}
}

func TestScanGroupsDirectoryKinds(t *testing.T) {
	db := testDB(t)
	ctx := context.Background()
	root := t.TempDir()

	// An audiobook is a folder of chapters.
	writeFile(t, filepath.Join(root, "Author", "Book", "01 - Chapter One.mp3"), "a")
	writeFile(t, filepath.Join(root, "Author", "Book", "02 - Chapter Two.mp3"), "b")
	writeFile(t, filepath.Join(root, "Author", "Book", "10 - Chapter Ten.mp3"), "c")
	// A loose file at the root groups under the library name.
	writeFile(t, filepath.Join(root, "stray.mp3"), "d")

	lib, err := library.Create(ctx, db, "Audiobooks", "audiobooks", root)
	if err != nil {
		t.Fatalf("Create: %v", err)
	}

	res, err := Scan(ctx, db, lib, testLogger())
	if err != nil {
		t.Fatalf("Scan: %v", err)
	}
	if res.Added != 4 {
		t.Errorf("added = %d, want 4 (four tracks)", res.Added)
	}

	// Two items: the directory, and the root group.
	var items int
	if err := db.QueryRowContext(ctx,
		`SELECT count(*) FROM media_items WHERE library_id = ?`, lib.ID).Scan(&items); err != nil {
		t.Fatal(err)
	}
	if items != 2 {
		t.Fatalf("items = %d, want 2", items)
	}

	var title string
	if err := db.QueryRowContext(ctx,
		`SELECT title FROM media_items WHERE library_id = ? AND path = ?`, lib.ID, "Author/Book").Scan(&title); err != nil {
		t.Fatalf("grouped item not found: %v", err)
	}
	if title != "Book" {
		t.Errorf("grouped title = %q, want Book", title)
	}

	// Tracks must come back in natural order, not lexical.
	rows, err := db.QueryContext(ctx, `
		SELECT f.path FROM media_files f JOIN media_items i ON i.id = f.item_id
		WHERE i.path = ? ORDER BY f.id`, "Author/Book")
	if err != nil {
		t.Fatal(err)
	}
	defer rows.Close()
	var paths []string
	for rows.Next() {
		var p string
		if err := rows.Scan(&p); err != nil {
			t.Fatal(err)
		}
		paths = append(paths, p)
	}
	want := []string{
		"Author/Book/01 - Chapter One.mp3",
		"Author/Book/02 - Chapter Two.mp3",
		"Author/Book/10 - Chapter Ten.mp3",
	}
	if len(paths) != len(want) {
		t.Fatalf("track count = %d, want %d: %v", len(paths), len(want), paths)
	}
	for i := range want {
		if paths[i] != want[i] {
			t.Errorf("track %d = %q, want %q", i, paths[i], want[i])
		}
	}
}

func TestScanUnreadableRoot(t *testing.T) {
	db := testDB(t)
	ctx := context.Background()
	root := t.TempDir()
	writeFile(t, filepath.Join(root, "a.epub"), "x")

	lib, err := library.Create(ctx, db, "Ebooks", "ebooks", root)
	if err != nil {
		t.Fatal(err)
	}
	// Index it once so there is a known file to go missing.
	if _, err := Scan(ctx, db, lib, testLogger()); err != nil {
		t.Fatal(err)
	}

	// Remove the root out from under the library.
	if err := os.RemoveAll(root); err != nil {
		t.Fatal(err)
	}

	res, err := Scan(ctx, db, lib, testLogger())
	if err != nil {
		t.Fatalf("scan should not fail on a missing root: %v", err)
	}
	if res.Added != 0 {
		t.Errorf("added=%d, want 0", res.Added)
	}
	if res.Missing != 1 {
		t.Errorf("missing=%d, want 1 (existing item now vanished)", res.Missing)
	}
	if len(res.Errors) == 0 {
		t.Error("expected the walk error to be recorded")
	}
}
func TestMissingSinceLifecycle(t *testing.T) {
	db := testDB(t)
	ctx := context.Background()
	root := t.TempDir()
	writeFile(t, filepath.Join(root, "a.epub"), "aaaaaaaaaa")
	writeFile(t, filepath.Join(root, "b.epub"), "bbbbbbbbbbbbbbbbbbbb")

	lib, err := library.Create(ctx, db, "Ebooks", "ebooks", root)
	if err != nil {
		t.Fatalf("Create: %v", err)
	}
	if _, err := Scan(ctx, db, lib, testLogger()); err != nil {
		t.Fatalf("first Scan: %v", err)
	}

	// A present file has no missing timestamp.
	var since string
	if err := db.QueryRowContext(ctx,
		`SELECT missing_since FROM media_files WHERE path = ?`, "b.epub").Scan(&since); err != nil {
		t.Fatal(err)
	}
	if since != "" {
		t.Errorf("present file missing_since = %q, want empty", since)
	}

	// Delete b.epub: it goes missing and the timestamp is recorded.
	if err := os.Remove(filepath.Join(root, "b.epub")); err != nil {
		t.Fatal(err)
	}
	if _, err := Scan(ctx, db, lib, testLogger()); err != nil {
		t.Fatalf("missing Scan: %v", err)
	}
	if err := db.QueryRowContext(ctx,
		`SELECT missing_since FROM media_files WHERE path = ?`, "b.epub").Scan(&since); err != nil {
		t.Fatal(err)
	}
	if since == "" {
		t.Fatal("missing_since was not set on the transition into missing")
	}
	first := since

	// A rescan must not refresh the age.
	if _, err := Scan(ctx, db, lib, testLogger()); err != nil {
		t.Fatalf("rescan: %v", err)
	}
	if err := db.QueryRowContext(ctx,
		`SELECT missing_since FROM media_files WHERE path = ?`, "b.epub").Scan(&since); err != nil {
		t.Fatal(err)
	}
	if since != first {
		t.Errorf("missing_since reset on rescan: %q -> %q", first, since)
	}

	// ListMissing surfaces the file with its age.
	entries, err := library.ListMissing(ctx, db, []int64{lib.ID})
	if err != nil {
		t.Fatalf("ListMissing: %v", err)
	}
	if len(entries) != 1 {
		t.Fatalf("ListMissing returned %d entries, want 1", len(entries))
	}
	e := entries[0]
	if e.Title != "b" || e.Path != "b.epub" || e.Kind != "ebooks" || e.LibraryID != lib.ID {
		t.Errorf("unexpected missing entry: %+v", e)
	}
	if e.MissingSince != first {
		t.Errorf("entry missing_since = %q, want %q", e.MissingSince, first)
	}

	// Restore the file: missing_since must be cleared, not left behind.
	writeFile(t, filepath.Join(root, "b.epub"), "bbbbbbbbbbbbbbbbbbbb")
	if _, err := Scan(ctx, db, lib, testLogger()); err != nil {
		t.Fatalf("restore Scan: %v", err)
	}
	if err := db.QueryRowContext(ctx,
		`SELECT missing_since FROM media_files WHERE path = ?`, "b.epub").Scan(&since); err != nil {
		t.Fatal(err)
	}
	if since != "" {
		t.Errorf("missing_since after restore = %q, want empty", since)
	}
	if got, _ := library.ListMissing(ctx, db, []int64{lib.ID}); len(got) != 0 {
		t.Errorf("ListMissing after restore returned %d entries, want 0", len(got))
	}
}

func TestPurgeMissingOnlyDeletesMissingItems(t *testing.T) {
	db := testDB(t)
	ctx := context.Background()
	root := t.TempDir()

	// Two audiobooks, each a folder of tracks.
	writeFile(t, filepath.Join(root, "Book A", "01.mp3"), "a1")
	writeFile(t, filepath.Join(root, "Book A", "02.mp3"), "a2")
	writeFile(t, filepath.Join(root, "Book B", "01.mp3"), "b1")
	writeFile(t, filepath.Join(root, "Book B", "02.mp3"), "b2")

	lib, err := library.Create(ctx, db, "Audiobooks", "audiobooks", root)
	if err != nil {
		t.Fatalf("Create: %v", err)
	}
	if _, err := Scan(ctx, db, lib, testLogger()); err != nil {
		t.Fatalf("Scan: %v", err)
	}

	// A second library with its own missing item, to prove purge is scoped.
	otherRoot := t.TempDir()
	writeFile(t, filepath.Join(otherRoot, "gone.epub"), "x")
	other, err := library.Create(ctx, db, "Other", "ebooks", otherRoot)
	if err != nil {
		t.Fatalf("Create other: %v", err)
	}
	if _, err := Scan(ctx, db, other, testLogger()); err != nil {
		t.Fatalf("Scan other: %v", err)
	}
	if err := os.Remove(filepath.Join(otherRoot, "gone.epub")); err != nil {
		t.Fatal(err)
	}
	if _, err := Scan(ctx, db, other, testLogger()); err != nil {
		t.Fatalf("Scan other (missing): %v", err)
	}

	// Remove one track from Book A: the item is not missing, only that file.
	// Remove Book B entirely: the item and both its files are missing.
	if err := os.Remove(filepath.Join(root, "Book A", "02.mp3")); err != nil {
		t.Fatal(err)
	}
	if err := os.RemoveAll(filepath.Join(root, "Book B")); err != nil {
		t.Fatal(err)
	}
	if _, err := Scan(ctx, db, lib, testLogger()); err != nil {
		t.Fatalf("second Scan: %v", err)
	}

	entries, err := library.ListMissing(ctx, db, []int64{lib.ID})
	if err != nil {
		t.Fatalf("ListMissing: %v", err)
	}
	if len(entries) != 3 {
		t.Fatalf("ListMissing returned %d entries, want 3 (Book A/02 + Book B/01 + Book B/02)", len(entries))
	}

	// Purge must remove only Book B.
	n, err := library.PurgeMissing(ctx, db, lib.ID)
	if err != nil {
		t.Fatalf("PurgeMissing: %v", err)
	}
	if n != 1 {
		t.Errorf("purged = %d, want 1 (only the fully missing item)", n)
	}

	// Book A survives, with its present and missing files intact.
	var items int
	if err := db.QueryRowContext(ctx,
		`SELECT count(*) FROM media_items WHERE library_id = ?`, lib.ID).Scan(&items); err != nil {
		t.Fatal(err)
	}
	if items != 1 {
		t.Errorf("media_items after purge = %d, want 1", items)
	}
	var bookATitle string
	if err := db.QueryRowContext(ctx,
		`SELECT title FROM media_items WHERE library_id = ? AND path = ?`, lib.ID, "Book A").Scan(&bookATitle); err != nil {
		t.Fatalf("a non-missing item was purged: %v", err)
	}
	var files int
	if err := db.QueryRowContext(ctx, `
		SELECT count(*) FROM media_files f JOIN media_items i ON i.id = f.item_id
		WHERE i.library_id = ? AND i.path = ?`, lib.ID, "Book A").Scan(&files); err != nil {
		t.Fatal(err)
	}
	if files != 2 {
		t.Errorf("Book A files after purge = %d, want 2", files)
	}

	// The other library's missing item is untouched.
	var otherItems int
	if err := db.QueryRowContext(ctx,
		`SELECT count(*) FROM media_items WHERE library_id = ?`, other.ID).Scan(&otherItems); err != nil {
		t.Fatal(err)
	}
	if otherItems != 1 {
		t.Errorf("other library items after purge = %d, want 1 (purge must be scoped)", otherItems)
	}

	// A second purge has nothing left to remove.
	if n, err := library.PurgeMissing(ctx, db, lib.ID); err != nil || n != 0 {
		t.Errorf("second purge = %d, %v; want 0, nil", n, err)
	}
}