Fix CLI semantics: exit codes, --prune, dedup, deep-verify

2026-05-01 07:04:37 +02:00
parent f97a1dc2eb
commit 6821215b0e
12 changed files with 321 additions and 288 deletions
@@ -541,3 +541,142 @@ func TestBackupAndRestore(t *testing.T) {

 	t.Log("Backup and restore test completed successfully")
 }
+
+// TestEndToEndFileStorage exercises the full backup → restore loop against the
+// real `file://` storage backend (FileStorer) on a real OS filesystem. This is
+// the closest local approximation of a production backup: encrypted blobs get
+// written to disk, the metadata SQLite database is exported through the same
+// blobgen pipeline as a real backup, and restoration reads them back through
+// the public Vaultik.Restore entrypoint. It is the canonical end-to-end smoke
+// test for 1.0.
+func TestEndToEndFileStorage(t *testing.T) {
+	log.Initialize(log.Config{})
+
+	// Real OS filesystem (SQLite + FileStorer both need it).
+	fs := afero.NewOsFs()
+	tempDir, err := os.MkdirTemp("", "vaultik-e2e-")
+	require.NoError(t, err)
+	defer func() { _ = os.RemoveAll(tempDir) }()
+
+	dataDir := filepath.Join(tempDir, "source")
+	storeDir := filepath.Join(tempDir, "remote")
+	restoreDir := filepath.Join(tempDir, "restored")
+	dbPath := filepath.Join(tempDir, "index.sqlite")
+
+	// Write a representative mix of file sizes:
+	//   - empty file
+	//   - tiny text file
+	//   - file just under chunk boundary
+	//   - file forcing multiple chunks
+	//   - nested subdirectories
+	chunkSize := int64(64 * 1024)
+	maxBlobSize := int64(512 * 1024)
+
+	testFiles := map[string][]byte{
+		filepath.Join(dataDir, "empty.txt"):                {},
+		filepath.Join(dataDir, "small.txt"):                []byte("hello vaultik"),
+		filepath.Join(dataDir, "subdir", "medium.bin"):     bytesPattern("medium-", int(chunkSize/2)),
+		filepath.Join(dataDir, "subdir", "large.bin"):      bytesPattern("large-", int(chunkSize*4)),
+		filepath.Join(dataDir, "deep", "nest", "leaf.txt"): []byte("leaf"),
+	}
+
+	for path, content := range testFiles {
+		require.NoError(t, fs.MkdirAll(filepath.Dir(path), 0o755))
+		require.NoError(t, afero.WriteFile(fs, path, content, 0o644))
+	}
+
+	// FileStorer is the real-world local-disk backend.
+	storer, err := storage.NewFileStorer(storeDir)
+	require.NoError(t, err)
+
+	agePublicKey := "age1ezrjmfpwsc95svdg0y54mums3zevgzu0x0ecq2f7tp8a05gl0sjq9q9wjg"
+	ageSecretKey := "AGE-SECRET-KEY-19CR5YSFW59HM4TLD6GXVEDMZFTVVF7PPHKUT68TXSFPK7APHXA2QS2NJA5"
+
+	cfg := &config.Config{
+		AgeRecipients:    []string{agePublicKey},
+		AgeSecretKey:     ageSecretKey,
+		CompressionLevel: 3,
+		Hostname:         "test-host",
+	}
+
+	ctx := context.Background()
+
+	db, err := database.New(ctx, dbPath)
+	require.NoError(t, err)
+	defer func() { _ = db.Close() }()
+
+	repos := database.NewRepositories(db)
+
+	sm := snapshot.NewSnapshotManager(snapshot.SnapshotManagerParams{
+		Repos:   repos,
+		Storage: storer,
+		Config:  cfg,
+	})
+	sm.SetFilesystem(fs)
+
+	scanner := snapshot.NewScanner(snapshot.ScannerConfig{
+		FS:               fs,
+		Storage:          storer,
+		ChunkSize:        chunkSize,
+		MaxBlobSize:      maxBlobSize,
+		CompressionLevel: cfg.CompressionLevel,
+		AgeRecipients:    cfg.AgeRecipients,
+		Repositories:     repos,
+	})
+
+	snapshotID, err := sm.CreateSnapshotWithName(ctx, cfg.Hostname, "e2e", "test-version", "test-git")
+	require.NoError(t, err)
+
+	scanResult, err := scanner.Scan(ctx, dataDir, snapshotID)
+	require.NoError(t, err)
+	require.Greater(t, scanResult.FilesScanned, 0)
+	require.Greater(t, scanResult.BlobsCreated, 0)
+
+	require.NoError(t, sm.CompleteSnapshot(ctx, snapshotID))
+	require.NoError(t, sm.ExportSnapshotMetadata(ctx, dbPath, snapshotID))
+
+	// Verify the backup actually landed on disk under blobs/ and metadata/.
+	blobInfo, err := os.Stat(filepath.Join(storeDir, "blobs"))
+	require.NoError(t, err)
+	require.True(t, blobInfo.IsDir())
+	metaInfo, err := os.Stat(filepath.Join(storeDir, "metadata", snapshotID))
+	require.NoError(t, err)
+	require.True(t, metaInfo.IsDir())
+
+	// Tear down the source DB before restore — restore must work using only
+	// the remote bytes plus the secret key, with no help from the local index.
+	require.NoError(t, db.Close())
+
+	restoreVaultik := &vaultik.Vaultik{
+		Config:  cfg,
+		Storage: storer,
+		Fs:      fs,
+		Stdout:  io.Discard,
+		Stderr:  io.Discard,
+	}
+	restoreVaultik.SetContext(ctx)
+
+	require.NoError(t, restoreVaultik.Restore(&vaultik.RestoreOptions{
+		SnapshotID: snapshotID,
+		TargetDir:  restoreDir,
+		Verify:     true,
+	}))
+
+	// Byte-equality compare every original against its restored copy.
+	for origPath, expected := range testFiles {
+		restoredPath := filepath.Join(restoreDir, origPath)
+		got, err := afero.ReadFile(fs, restoredPath)
+		require.NoError(t, err, "restored file missing: %s", restoredPath)
+		require.Equalf(t, expected, got, "byte-equality failed for %s", origPath)
+	}
+}
+
+// bytesPattern returns a deterministic byte slice of length n with a tag prefix,
+// useful for forcing chunker behavior with reproducible content.
+func bytesPattern(tag string, n int) []byte {
+	out := make([]byte, n)
+	for i := range out {
+		out[i] = byte(tag[i%len(tag)] ^ byte(i&0xff))
+	}
+	return out
+}