Migrations

As your app grows, you may want to change the table structure for your drift database: New features need new columns or tables, and outdated columns may have to be altered or removed altogether. When making changes to your database schema, you need to write migrations enabling users with an old version of your app to convert to the database expected by the latest version. Drift provides a set of APIs to make writing migrations easy.

Basics

Drift provides a migration API that can be used to gradually apply schema changes after bumping the schemaVersion getter inside the Database class. To use it, override the migration getter.

Here's an example: Let's say you wanted to add a due date to your todo entries (v2 of the schema). Later, you decide to also add a priority column (v3 of the schema).

class Todos extends Table {
  IntColumn get id => integer().autoIncrement()();
  TextColumn get title => text().withLength(min: 6, max: 10)();
  TextColumn get content => text().named('body')();
  IntColumn get category => integer().nullable()();
  DateTimeColumn get dueDate =>
      dateTime().nullable()(); // new, added column in v2
  IntColumn get priority => integer().nullable()(); // new, added column in v3
}

We can now change the database class like this:

@override
int get schemaVersion => 3; // bump because the tables have changed.

@override
MigrationStrategy get migration {
  return MigrationStrategy(
    onCreate: (Migrator m) async {
      await m.createAll();
    },
    onUpgrade: (Migrator m, int from, int to) async {
      if (from < 2) {
        // we added the dueDate property in the change from version 1 to
        // version 2
        await m.addColumn(todos, todos.dueDate);
      }
      if (from < 3) {
        // we added the priority property in the change from version 1 or 2
        // to version 3
        await m.addColumn(todos, todos.priority);
      }
    },
  );
}
// The rest of the class can stay the same

You can also add individual tables or drop them - see the reference of Migrator for all the available options.

You can also use higher-level query APIs like select, update or delete inside a migration callback. However, be aware that drift expects the latest schema when creating SQL statements or mapping results. For instance, when adding a new column to your database, you shouldn't run a select on that table before you've actually added the column. In general, try to avoid running queries in migration callbacks if possible.

sqlite can feel a bit limiting when it comes to migrations - there only are methods to create tables and columns. Existing columns can't be altered or removed. A workaround is described here, it can be used together with customStatement to run the statements. Alternatively, complex migrations help automating this.

Tips

To ensure your schema stays consistent during a migration, you can wrap it in a transaction block. However, be aware that some pragmas (including foreign_keys) can't be changed inside transactions. Still, it can be useful to:

• always re-enable foreign keys before using the database, by enabling them in beforeOpen.
• disable foreign-keys before migrations
• run migrations inside a transaction
• make sure your migrations didn't introduce any inconsistencies with PRAGMA foreign_key_check.

With all of this combined, a migration callback can look like this:

return MigrationStrategy(
  onUpgrade: (m, from, to) async {
    // disable foreign_keys before migrations
    await customStatement('PRAGMA foreign_keys = OFF');

    await transaction(() async {
      // put your migration logic here
    });

    // Assert that the schema is valid after migrations
    if (kDebugMode) {
      final wrongForeignKeys =
          await customSelect('PRAGMA foreign_key_check').get();
      assert(wrongForeignKeys.isEmpty,
          '${wrongForeignKeys.map((e) => e.data)}');
    }
  },
  beforeOpen: (details) async {
    await customStatement('PRAGMA foreign_keys = ON');
    // ....
  },
);

Complex migrations

Sqlite has builtin statements for simple changes, like adding columns or dropping entire tables. More complex migrations require a 12-step procedure that involves creating a copy of the table and copying over data from the old table. Drift 3.4 introduced the TableMigration api to automate most of this procedure, making it easier and safer to use.

To start the migration, drift will create a new instance of the table with the current schema. Next, it will copy over rows from the old table. In most cases, for instance when changing column types, we can't just copy over each row without changing its content. Here, you can use a columnTransformer to apply a per-row transformation. The columnTransformer is a map from columns to the sql expression that will be used to copy the column from the old table. For instance, if we wanted to cast a column before copying it, we could use:

columnTransformer: {
  todos.category: todos.category.cast<int>(),
}

Internally, drift will use a INSERT INTO SELECT statement to copy old data. In this case, it would look like INSERT INTO temporary_todos_copy SELECT id, title, content, CAST(category AS INT) FROM todos. As you can see, drift will use the expression from the columnTransformer map and fall back to just copying the column otherwise. If you're introducing new columns in a table migration, be sure to include them in the newColumns parameter of TableMigration. Drift will ensure that those columns have a default value or a transformation in columnTransformer. Of course, drift won't attempt to copy newColumns from the old table either.

Regardless of whether you're implementing complex migrations with TableMigration or by running a custom sequence of statements, we strongly recommend to write integration tests covering your migrations. This helps to avoid data loss caused by errors in a migration.

Here are some examples demonstrating common usages of the table migration api:

Changing the type of a column

Let's say the category column in Todos used to be a non-nullable text() column that we're now changing to a nullable int. For simplicity, we assume that category always contained integers, they were just stored in a text column that we now want to adapt.

class Todos extends Table {
  IntColumn get id => integer().autoIncrement()();
  TextColumn get title => text().withLength(min: 6, max: 10)();
  TextColumn get content => text().named('body')();
-  IntColumn get category => text()();
+  IntColumn get category => integer().nullable()();
}

After re-running your build and incrementing the schema version, you can write a migration:

return MigrationStrategy(
  onUpgrade: (m, old, to) async {
    if (old <= yourOldVersion) {
      await m.alterTable(
        TableMigration(todos, columnTransformer: {
          todos.category: todos.category.cast<int>(),
        }),
      );
    }
  },
);

The important part here is the columnTransformer - a map from columns to expressions that will be used to copy the old data. The values in that map refer to the old table, so we can use todos.category.cast<int>() to copy old rows and transform their category. All columns that aren't present in columnTransformer will be copied from the old table without any transformation.

Changing column constraints

When you're changing columns constraints in a way that's compatible to existing data (e.g. changing non-nullable columns to nullable columns), you can just copy over data without applying any transformation:

await m.alterTable(TableMigration(todos));

Deleting columns

Deleting a column that's not referenced by a foreign key constraint is easy too:

await m.alterTable(TableMigration(yourTable));

To delete a column referenced by a foreign key, you'd have to migrate the referencing tables first.

Renaming columns

If you're renaming a column in Dart, note that the easiest way is to just rename the getter and use named: TextColumn newName => text().named('old_name')(). That is fully backwards compatible and doesn't require a migration.

If you know your app runs on sqlite 3.25.0 or later (it does if you're using sqlite3_flutter_libs), you can also use the renameColumn api in Migrator:

m.renameColumn(yourTable, 'old_column_name', yourTable.newColumn);

If you do want to change the actual column name in a table, you can write a columnTransformer to use an old column with a different name:

await m.alterTable(
  TableMigration(
    yourTable,
    columnTransformer: {
      yourTable.newColumn: const CustomExpression('old_column_name')
    },
  )
)

Migrating views, triggers and indices

When changing the definition of a view, a trigger or an index, the easiest way to update the database schema is to drop and re-create the element. With the Migrator API, this is just a matter of calling await drop(element) followed by await create(element), where element is the trigger, view or index to update.

Note that the definition of a Dart-defined view might change without modifications to the view class itself. This is because columns from a table are referenced with a getter. When renaming a column through .named('name') in a table definition without renaming the getter, the view definition in Dart stays the same but the CREATE VIEW statement changes.

A headache-free solution to this problem is to just re-create all views in a migration, for which the Migrator provides the recreateAllViews method.

Post-migration callbacks

The beforeOpen parameter in MigrationStrategy can be used to populate data after the database has been created. It runs after migrations, but before any other query. Note that it will be called whenever the database is opened, regardless of whether a migration actually ran or not. You can use details.hadUpgrade or details.wasCreated to check whether migrations were necessary:

beforeOpen: (details) async {
    if (details.wasCreated) {
      final workId = await into(categories).insert(Category(description: 'Work'));

      await into(todos).insert(TodoEntry(
            content: 'A first todo entry',
            category: null,
            targetDate: DateTime.now(),
      ));

      await into(todos).insert(
            TodoEntry(
              content: 'Rework persistence code',
              category: workId,
              targetDate: DateTime.now().add(const Duration(days: 4)),
      ));
    }
},

You could also activate pragma statements that you need:

beforeOpen: (details) async {
  if (details.wasCreated) {
    // ...
  }
  await customStatement('PRAGMA foreign_keys = ON');
}

During development

During development, you might be changing your schema very often and don't want to write migrations for that yet. You can just delete your apps' data and reinstall the app - the database will be deleted and all tables will be created again. Please note that uninstalling is not enough sometimes - Android might have backed up the database file and will re-create it when installing the app again.

You can also delete and re-create all tables every time your app is opened, see this comment on how that can be achieved.

Verifying migrations

Drift contains experimental support to verify the integrity of your migrations in unit tests.

To support this feature, drift can help you generate

• a json representation of your database schema
• test databases operating on an older schema version

By using those test databases, drift can help you test migrations from and to any schema version.

Setup

To use this feature, drift needs to know all schemas of your database. A schema is the set of all tables, triggers and indices that you use in your database.

You can use the CLI tools to export a json representation of your schema. In this guide, we'll assume a file layout like the following, where my_app is the root folder of your project:

my_app
  .../
  lib/
    database/
      database.dart
      database.g.dart
  test/
    generated_migrations/
      schema.dart
      schema_v1.dart
      schema_v2.dart
  drift_schemas/
    drift_schema_v1.json
    drift_schema_v2.json
  pubspec.yaml

The generated migrations implementation and the schema jsons will be generated by drift. To start writing schemas, create an empty folder named drift_schemas in your project. Of course, you can also choose a different name or use a nested subfolder if you want to.

Exporting the schema

To begin, let's create the first schema representation:

$ mkdir drift_schemas
$ dart run drift_dev schema dump lib/database/database.dart drift_schemas/

This instructs the generator to look at the database defined in lib/database/database.dart and extract its schema into the new folder.

After making a change to your database schema, you can run the command again. For instance, let's say we made a change to our tables and increased the schemaVersion to 2. To dump the new schema, just run the command again:

$ dart run drift_dev schema dump lib/database/database.dart drift_schemas/

You'll need to run this command every time you change the schema of your database and increment the schemaVersion.

Drift will name the files in the folder drift_schema_vX.json, where X is the current schemaVersion of your database. If drift is unable to extract the version from your schemaVersion getter, provide the full path explicitly:

$ dart run drift_dev schema dump lib/database/database.dart drift_schemas/drift_schema_v3.json

Generating test code

After you exported the database schema into a folder, you can generate old versions of your database class based on those schema files. For verifications, drift will generate a much smaller database implementation that can only be used to test migrations.

You can put this test code whereever you want, but it makes sense to put it in a subfolder of test/. If we wanted to write them to test/generated_migrations/, we could use

$ dart run drift_dev schema generate drift_schemas/ test/generated_migrations/

Writing tests

After that setup, it's finally time to write some tests! For instance, a test could look like this:

import 'package:my_app/database/database.dart';

import 'package:test/test.dart';
import 'package:drift_dev/api/migrations.dart';

// The generated directory from before.
import 'generated_migrations/schema.dart';

void main() {
  late SchemaVerifier verifier;

  setUpAll(() {
    // GeneratedHelper() was generated by drift, the verifier is an api
    // provided by drift_dev.
    verifier = SchemaVerifier(GeneratedHelper());
  });

  test('upgrade from v1 to v2', () async {
    // Use startAt(1) to obtain a database connection with all tables
    // from the v1 schema.
    final connection = await verifier.startAt(1);
    final db = MyDatabase.connect(connection);

    // Use this to run a migration to v2 and then validate that the
    // database has the expected schema.
    await verifier.migrateAndValidate(db, 2);
  });
}

In general, a test looks like this:

1. Use verifier.startAt() to obtain a connection to a database with an initial schema. This database contains all your tables, indices and triggers from that version, created by using Migrator.createAll.
2. Create your application database with that connection - you can forward the DatabaseConnection to the GeneratedDatabase.connect() constructor on the parent class for this.
3. Call verifier.migrateAndValidate(db, version). This will initiate a migration towards the target version (here, 2). Unlike the database created by startAt, this uses the migration logic you wrote for your database.

migrateAndValidate will extract all CREATE statement from the sqlite_schema table and semantically compare them. If it sees anything unexpected, it will throw a SchemaMismatch exception to fail your test.

Verifying data integrity

In addition to the changes made in your table structure, its useful to ensure that data that was present before a migration is still there after it ran. You can use schemaAt to obtain a raw Database from the sqlite3 package in addition to a connection. This can be used to insert data before a migration. After the migration ran, you can then check that the data is still there.

Note that you can't use the regular database class from you app for this, since its data classes always expect the latest schema. However, you can instruct drift to generate older snapshots of your data classes and companions for this purpose. To enable this feature, pass the --data-classes and --companions command-line arguments to the drift_dev schema generate command:

$ dart run drift_dev schema generate --data-classes --companions drift_schemas/ test/generated_migrations/

Then, you can import the generated classes with an alias:

import 'generated_migrations/schema_v1.dart' as v1;
import 'generated_migrations/schema_v2.dart' as v2;

This can then be used to manually create and verify data at a specific version:

void main() {
  // ...
  test('upgrade from v1 to v2', () async {
    final schema = await verifier.schemaAt(1);

    // Add some data to the users table, which only has an id column at v1
    final oldDb = v1.DatabaseAtV1.connect(schema.newConnection());
    await oldDb.into(oldDb.users).insert(const v1.UsersCompanion(id: Value(1)));
    await oldDb.close();

    // Run the migration and verify that it adds the name column.
    final db = Database(schema.newConnection());
    await verifier.migrateAndValidate(db, 2);
    await db.close();

    // Make sure the user is still here
    final migratedDb = v2.DatabaseAtV2.connect(schema.newConnection());
    final user = await migratedDb.select(migratedDb.users).getSingle();
    expect(user.id, 1);
    expect(user.name, 'no name'); // default from the migration
    await migratedDb.close();
  });
}

Verifying a database schema at runtime

Instead (or in addition to) writing tests to ensure your migrations work as they should, you can use a new API from drift_dev 1.5.0 to verify the current schema without any additional setup.

// import the migrations tooling
import 'package:drift_dev/api/migrations.dart';

class MyDatabase extends _$MyDatabase {
  @override
  MigrationStrategy get migration => MigrationStrategy(
        onCreate: (m) async {/* ... */},
        onUpgrade: (m, from, to) async {/* your existing migration logic */},
        beforeOpen: (details) async {
          // your existing beforeOpen callback, enable foreign keys, etc.

          if (kDebugMode) {
            // This check pulls in a fair amount of code that's not needed
            // anywhere else, so we recommend only doing it in debug builds.
            await validateDatabaseSchema();
          }
        },
      );
}

When you use validateDatabaseSchema, drift will transparently:

• collect information about your database by reading from sqlite3_schema.
• create a fresh in-memory instance of your database and create a reference schema with Migrator.createAll().
• compare the two. Ideally, your actual schema at runtime should be identical to the fresh one even though it grew through different versions of your app.

When a mismatch is found, an exception with a message explaining exactly where another value was expected will be thrown. This allows you to find issues with your schema migrations quickly.