Docs Menu
Docs Home
/ /
Atlas Device SDKs
/ / /

Define a Realm Object Schema - Flutter SDK

On this page

  • Create Model
  • Import Realm
  • Create Generated File Part Directive
  • Create RealmModel
  • Generate RealmObject
  • Use RealmObject in Application
  • Using Schemas with Device Sync
  • Supported Data Types
  • Property Annotations
  • Define Relationship Properties
  • Map a Model or Class to a Different Name
  • Model Unstructured Data
  • Generate the RealmObject
  • Define an Asymmetric Object

Important

Flutter SDK v2.0.0 Breaking Change to Generated Files

Flutter SDK version 2.0.0 introduces an update to the builder, which impacts how files generate. In v2.0.0 and later, all generated files use the .realm.dart naming convention instead of .g.dart.

This is a breaking change for existing apps. For information on how to upgrade an existing app from an earlier SDK version to v2.0.0 or later, refer to Upgrade to Flutter SDK v2.0.0.

An object schema is a configuration object that defines the properties and relationships of a Realm object. Realm client applications define object schemas with the native class implementation in their respective language using the Object Schema.

Object schemas specify constraints on object properties such as the data type of each property and whether or not a property is required. Schemas can also define relationships between object types in a realm.

1

Import the Realm SDK package at the top of your file.

schemas.dart
import 'package:realm/realm.dart';
schemas.dart
import 'package:realm_dart/realm.dart';
2

Changed in version v2.0.0: Generated files are named .realm.dart instead of .g.dart

Add a part directive to include the RealmObject file that you generate in step 4 in the same package as the file you're currently working on.

schemas.dart
part 'schemas.realm.dart';
3

Create the model for your Realm schema. You must include the annotation RealmModel at the top of the class definition.

You'll use the RealmModel to generate the public RealmObject used throughout the application in step 4.

You can make the model private or public. We recommend making the all models private and defining them in a single file. Prepend the class name with an underscore (_) to make it private.

If you need to define your schema across multiple files, you can make the RealmModel public. Prepend the name with a dollar sign ($) to make the model public. You must do this to generate the RealmObject from the RealmModel, as described in step 4.

Add fields to the RealmModel. You can add all supported data types. Include additional behavior using property annotations.

schemas.dart
@RealmModel()
class _Car {
@PrimaryKey()
late ObjectId id;
late String make;
late String? model;
late int? miles;
}

Note

Class names are limited to a maximum of 57 UTF-8 characters.

4

Changed in version v2.0.0: Generated files are named .realm.dart instead of .g.dart

Generate the RealmObject, which you'll use in your application:

dart run realm generate
dart run realm_dart generate

This command generates the file in the same directory as your model file. It has the name you specified in the part directive of step 2.

Tip

Track the generated file

Track the generated file in your version control system, such as git.

Example

File structure after generating model

.
├── schemas.dart
├── schemas.realm.dart // newly generated file
├── myapp.dart
└── ...rest of application
5

Use the RealmObject that you generated in the previous step in your application. Since you included the generated file as part of the same package where you defined the RealmModel in step 2, access the RealmObject by importing the file with the RealmModel.

myapp.dart
import './schemas.dart';
final hondaCivic = Car(ObjectId(), 'Honda', model: 'Civic', miles: 99);

An App Services Schema is a list of valid object schemas that each define an object type that an App may persist. All synced objects in a realm must conform to the App Services Schema.

Client applications provide a Object Schema when they open a realm. If a realm already contains data, then it already has a schema, and when it is opened, Realm validates the schema on the client against the existing schema.

You can define App Services Schemas in the following ways:

  • Automatically with the Object Schema if development mode is enabled.

  • Explicitly define the App Services Schema with App Services.

In your schema you must use the MapTo("_id") annotation with your primary key in the RealmModel to successfully sync your Object Schema with App Services.

@RealmModel()
class _SyncSchema {
@PrimaryKey()
@MapTo("_id")
late ObjectId id;
// ... other properties
}

For further information on defining your schema and which of these approaches you should consider for your application, refer to the Create a Realm Schema documentation.

Realm schemas support many Dart-language data types, in addition to some Realm-specific types. For a comprehensive reference of all supported data types, refer to Data Types.

Use annotations to add functionality to properties in your Realm object models. You can use annotations for things like marking a property as nullable, setting a primary key, ignoring a property, and more. To learn more about the available property annotations, refer to Property Annotations.

You can define relationships between Realm objects in your schema. The Realm Flutter SDK supports to-one relationships, to-many relationships, inverse relationships, and embedding objects within other objects. To learn more about how to define relationships in your Realm object schema, refer to Relationships.

You can use the MapTo annotation to map a Realm object model or property to a different stored name in Realm. This can be useful in the following scenarios. For example:

  • To make it easier to work across platforms where naming conventions differ. For example, if your Device Sync schema property names use snake case, while your project uses camel case.

  • To change a class or field name without forcing a migration.

  • To support multiple model classes with the same name in different packages.

  • To use a class name that is longer than the 57-character limit enforced by Realm.

If you're using Atlas Device Sync, the name that you specify in the MapTo annotation is used as the persisted App Services Schema name.

@RealmModel()
@MapTo('naval_ship')
class _Boat {
@PrimaryKey()
late ObjectId id;
late String name;
late int? maxKnots;
late int? nauticalMiles;
}
class _Vehicle {
@PrimaryKey()
late ObjectId id;
late String? maybeDescription; // optional value
late double milesTravelled = 0; // 0 is default value
@Ignored()
late String notInRealmModel;
@Indexed()
late String make;
@MapTo('wheels') // 'wheels' is property name in the RealmObject
late int numberOfWheels;
}

New in version 2.0.0.

Starting in Flutter SDK version 2.0.0, you can store collections of mixed data within a RealmValue property. You can use this feature to model complex data structures, such as JSON or MongoDB documents, without having to define a strict data model.

Unstructured data is data that doesn't easily conform to an expected schema, making it difficult or impractical to model to individual data classes. For example, your app might have highly variable data or dynamic data whose structure is unknown at runtime.

Storing collections in a mixed property offers flexibility without sacrificing functionality, including performant synchronization when using Device Sync. And you can work with them the same way you would a non-mixed collection:

  • You can nest mixed collections up to 100 levels.

  • You can query on and react to changes on mixed collections.

  • You can find and update individual mixed collection elements.

However, storing data in mixed collections is less performant than using a structured schema or serializing JSON blobs into a single string property.

To model unstructured data in your app, define the appropriate properties in your schema as RealmValue types. You can then set these RealmValue properties as a RealmList or a RealmMap collection of RealmValue elements. Note that RealmValue cannot represent a RealmSet or an embedded object.

For example, you might use a RealmValue that contains a map of mixed data when modeling a variable event log object:

Data model
// Define class with a `RealmValue` property
@RealmModel()
class _EventLog {
@PrimaryKey()
late ObjectId id;
late String eventType;
late DateTime timestamp;
late String userId;
late RealmValue details;
}
Create unstructured data
realm.write(() {
// Add `eventLog` property data as a map of mixed data, which
// also includes nested lists of mixed data
realm.add(EventLog(ObjectId(), 'purchase', DateTime.now(), 'user123',
details: RealmValue.from({
'ipAddress': '192.168.1.1',
'items': [
{'id': 1, 'name': 'Laptop', 'price': 1200.00},
{'id': 2, 'name': 'Mouse', 'price': 49.99}
],
'total': 1249.99
})));
final eventLog = realm.all<EventLog>().first;
final items = eventLog.details.asMap();
print('''
Event Type: ${eventLog.eventType}
Timestamp: ${eventLog.timestamp}
User ID: ${eventLog.userId}
Details:
Item:
''');
for (var item in items.entries) {
print('${item.key}: ${item.value}');
}
Event Type: purchase
Timestamp: 2024-03-18 13:50:58.402979Z
User ID: user123
Details:
Item:
ipAddress: RealmValue(192.168.1.1)
items: RealmValue([RealmValue({id: RealmValue(1), name: RealmValue(Laptop), price: RealmValue(1200.0)}), RealmValue({id: RealmValue(2), name: RealmValue(Mouse), price: RealmValue(49.99)})])
total: RealmValue(1249.99)

Tip

  • Use a map of mixed data types when the type is unknown but each value will have a unique identifier.

  • Use a list of mixed data types when the type is unknown but the order of objects is meaningful.

Changed in version v2.0.0: Generated files are named .realm.dart instead of .g.dart

Once you've completed your Realm model, you must generate the RealmObject class to use it in your application.

Run the following command to generate RealmObjects:

dart run realm generate
dart run realm_dart generate

Running this creates a public class in a new file in the directory where you defined the RealmModel class per the Create Model section.

The generated file has the same base name as the file with your RealmModel, ending with .realm.dart. For example if the file with your RealmModel is named schemas.dart, the generated file will be schemas.realm.dart.

Note

Remember to include the generated file in a part directive in your RealmModel definition file.

schemas.dart
// ...import packages
part 'schemas.realm.dart';
@RealmModel()
// ...model definition

If you'd like to watch your data models to generate RealmObjects whenever there's a change, include the --watch flag in your command.

dart run realm generate --watch
dart run realm_dart generate --watch

To clean the generator caches, include the --clean flag in your command. Cleaning the generator cache can be useful when debugging.

dart run realm generate --clean
dart run realm_dart generate --clean

New in version 1.5.0.

Asymmetric objects require Flexible Sync. To define an asymmetric object, pass ObjectType.asymmetricObject to @RealmModel().

@RealmModel(ObjectType.asymmetricObject)
class _WeatherSensor {
@PrimaryKey()
@MapTo("_id")
late ObjectId id;
late String deviceId;
late double modtemperatureInFahrenheitel;
late double barometricPressureInHg;
late double windSpeedInMph;
}

In Flutter SDK versions 1.5.0 and earlier, you cannot link from asymmetricObject types to RealmObjects. In SDK versions 1.6.0 and later, asymmetricObject types can link to RealmObjects in addition to embedded object types.

Note

Attempting to Read Asymmetric Objects

Asymmetric objects can't be read. If you attempt to query an asymmetric object, you will get the following error: "Error: You can't query an asymmetric class.".

To learn more about Data Ingest, refer to Stream Data to Atlas.

Back

Model Data

Next

Data Types