fix piece type inference for non-inline preparations

model/migrate.go

@@ -262,9 +262,11 @@ func fixPostgresSequences(db *gorm.DB) error {
 }
 
 // inferPieceTypes sets piece_type for cars that predate the column.
-// A piece is "data" if any of its blocks reference files (contain file content).
-// A piece is "dag" if none of its blocks reference files (directory metadata only).
-// This is idempotent - only updates rows where piece_type is NULL or empty.
+// for inline preps, a car is "data" if any of its car_blocks reference files.
+// for non-inline preps, car_blocks don't reference files (data is on disk),
+// so we fall back to num_of_files > 0 which is only set by the packer.
+// everything else is "dag" (directory metadata only).
+// idempotent - only updates rows where piece_type is NULL or empty.
 func inferPieceTypes(db *gorm.DB) error {
 	dialect := db.Dialector.Name()
 
@@ -288,15 +290,14 @@ func inferPieceTypes(db *gorm.DB) error {
 	if dialect == "sqlite" {
 		query = `
 			UPDATE cars SET piece_type = (
-				CASE WHEN EXISTS (
+				CASE WHEN num_of_files > 0 OR EXISTS (
 					SELECT 1 FROM car_blocks WHERE car_blocks.car_id = cars.id AND car_blocks.file_id IS NOT NULL
 				) THEN 'data' ELSE 'dag' END
 			) WHERE piece_type IS NULL OR piece_type = ''`
 	} else {
-		// postgres/mysql support correlated subquery in CASE
 		query = `
 			UPDATE cars c SET piece_type = CASE
-				WHEN EXISTS (
+				WHEN c.num_of_files > 0 OR EXISTS (
 					SELECT 1 FROM car_blocks cb WHERE cb.car_id = c.id AND cb.file_id IS NOT NULL
 				) THEN 'data' ELSE 'dag'
 			END WHERE c.piece_type IS NULL OR c.piece_type = ''`

model/migrate_test.go

@@ -76,3 +76,61 @@ func TestFKSetNullOnDelete(t *testing.T) {
 		require.Nil(t, loadedCarBlock.FileID)
 	})
 }
+
+func TestInferPieceTypes(t *testing.T) {
+	testutil.All(t, func(ctx context.Context, t *testing.T, db *gorm.DB) {
+		prep := model.Preparation{Name: "test", MaxSize: 1024, PieceSize: 1024}
+		require.NoError(t, db.Create(&prep).Error)
+
+		storage := model.Storage{Name: "test", Type: "local", Path: "/tmp"}
+		require.NoError(t, db.Create(&storage).Error)
+
+		attachment := model.SourceAttachment{PreparationID: prep.ID, StorageID: storage.ID}
+		require.NoError(t, db.Create(&attachment).Error)
+
+		file := model.File{Path: "test.txt", Size: 100, AttachmentID: &attachment.ID}
+		require.NoError(t, db.Create(&file).Error)
+
+		// inline data piece: car_blocks reference files
+		inlineCar := model.Car{PieceSize: 1024, PreparationID: &prep.ID, AttachmentID: &attachment.ID}
+		require.NoError(t, db.Create(&inlineCar).Error)
+		cb := model.CarBlock{CarOffset: 0, CarID: &inlineCar.ID, FileID: &file.ID}
+		require.NoError(t, db.Create(&cb).Error)
+
+		// non-inline data piece: no file refs in car_blocks, but num_of_files > 0
+		nonInlineCar := model.Car{PieceSize: 1024, NumOfFiles: 5, PreparationID: &prep.ID, AttachmentID: &attachment.ID}
+		require.NoError(t, db.Create(&nonInlineCar).Error)
+
+		// dag piece: no file refs, num_of_files == 0
+		dagCar := model.Car{PieceSize: 1024, PreparationID: &prep.ID, AttachmentID: &attachment.ID}
+		require.NoError(t, db.Create(&dagCar).Error)
+
+		// all should have empty piece_type
+		for _, id := range []model.CarID{inlineCar.ID, nonInlineCar.ID, dagCar.ID} {
+			var c model.Car
+			require.NoError(t, db.First(&c, id).Error)
+			require.Empty(t, c.PieceType)
+		}
+
+		// run migration
+		require.NoError(t, model.AutoMigrate(db))
+
+		var c1, c2, c3 model.Car
+
+		require.NoError(t, db.First(&c1, inlineCar.ID).Error)
+		require.Equal(t, model.DataPiece, c1.PieceType, "inline car with file refs should be data")
+
+		require.NoError(t, db.First(&c2, nonInlineCar.ID).Error)
+		require.Equal(t, model.DataPiece, c2.PieceType, "non-inline car with num_of_files > 0 should be data")
+
+		require.NoError(t, db.First(&c3, dagCar.ID).Error)
+		require.Equal(t, model.DagPiece, c3.PieceType, "car with no file refs and num_of_files == 0 should be dag")
+
+		// idempotent: running again should not change anything
+		require.NoError(t, model.AutoMigrate(db))
+
+		var c4 model.Car
+		require.NoError(t, db.First(&c4, nonInlineCar.ID).Error)
+		require.Equal(t, model.DataPiece, c4.PieceType)
+	})
+}