Pakete, Repositories und App-Kategorien suchen

Paketdetailsv2.0.0

flutter_data

Die nahtlose Möglichkeit, mit persistierenden Datenmodellen in Flutter zu arbeiten. Inspiriert von Ember Data und ActiveRecord.

AndroidiOSmacOSWindowsLinux
30-Tage-Downloads
856
Likes
373
Pub-Punkte
130

Paket installieren

flutter pub add flutter_data

Paketübersicht

Aus öffentlichen Paketdaten erstellte Orientierung.

Flutter Data bietet einen vereinfachten Ansatz zur Verwaltung persistenter Datenmodelle in Flutter-Anwendungen, der sich von Ember Data und ActiveRecord inspirieren lässt. Es vereinfacht CRUD-Vorgänge, Zustandsverwaltung und Datensynchronisation über Geräte hinweg, indem es eine einheitliche API für die Arbeit mit lokalen und remote Datenquellen bereitstellt. Das Paket integriert sich nahtlos in das Flutter-Ökosystem und unterstützt typensichere Modelle, automatische Serialisierung und eingebautes Caching. Entwickler können Datenmodelle und Beziehungen einfach definieren, wodurch eine effiziente Datenhandhabung ohne Boilerplate-Code möglich ist.

Anwendungsfälle

  • Verwaltung von Datenworkflows mit Fokus auf Offline
  • Synchronisieren von Daten zwischen lokaler Speicherung und APIs
  • Skalierbare CRUD-Anwendungen erstellen
  • Behandlung komplexer Datenbeziehungen
  • Reaktive Datenaktualisierungen implementieren

Hauptmerkmale

  • Automatische Modellserialisierung
  • Eingebaute Persistenz mit SQLite
  • reaktive Zustandsverwaltung
  • Unterstützung für REST- und GraphQL-APIs
  • Typsichere Datenmodelle und Abfragen

Geeignet für

  • Apps, die einen Offline-Zugriff auf Daten erfordern
  • Projekte mit REST- oder GraphQL-Backends
  • Entwickler, die weniger Boilerplate suchen
  • Teams, die eine MVC-ähnliche Datenarchitektur übernehmen
  • Anwendungen mit komplexen Datenbeziehungen

Hinweise

  • Lernkurve für neue Benutzer
  • Eingeschränkte Dokumentation im Vergleich zu etablierten Alternativen
  • Kann zusätzliche Einrichtung für benutzerdefinierte Endpunkte erfordern
  • Nicht ideal für sehr einfache Apps mit geringem Datenbedarf
  • Abhängig von externen Abhängigkeiten wie json_serializable

Original-README

Originaler README-Schnappschuss aus dem verknüpften Quell-Repository. Aktuelle Inhalte finden Sie im Repository.

Flutter Data

tests codecov pub.dev license

Persistent reactive models in Flutter with zero boilerplate

Flutter Data is a local-first data framework with a customizable REST client and powerful model relationships, built on Riverpod.

Inspired by Ember Data and ActiveRecord.

Features

  • Adapters for all models 🚀
    • Default CRUD and custom remote endpoints
    • StateNotifier watcher APIs
  • Built for offline-first 🔌
    • SQLite3-based local storage at its core, with adapters for many other engines: Objectbox, Isar, etc (coming soon)
  • Intuitive APIs, effortless setup 💙
    • Truly configurable and composable via Dart mixins and codegen
    • Built-in Riverpod providers for all models
  • Exceptional relationship support ⚡️
    • Automatically synchronized, fully traversable relationship graph
    • Reactive relationships

👩🏾‍💻 Quick introduction

In Flutter Data, every model gets its default adapter. These adapters can be extended by mixing in custom adapters.

Annotate a model with @DataAdapter and pass a custom adapter:

@JsonSerializable()
@DataAdapter([MyJSONServerAdapter])
class User extends DataModel<User> {
  @override
  final int? id; // ID can be of any type
  final String name;
  User({this.id, required this.name});
  // `User.fromJson` and `toJson` optional
}

mixin MyJSONServerAdapter on RemoteAdapter<User> {
  @override
  String get baseUrl => "https://my-json-server.typicode.com/flutterdata/demo/";
}

After code-gen, Flutter Data will generate the resulting Adapter<User> which is accessible via Riverpod's ref.users or container.users.

@override
Widget build(BuildContext context, WidgetRef ref) {
  final state = ref.users.watchOne(1);
  if (state.isLoading) {
    return Center(child: const CircularProgressIndicator());
  }
  final user = state.model;
  return Text(user.name);
}

To update the user:

TextButton(
  onPressed: () => ref.users.save(User(id: 1, name: 'Updated')),
  child: Text('Update'),
),

ref.users.watchOne(1) will make a background HTTP request (to https://my-json-server.typicode.com/flutterdata/demo/users/1 in this case), deserialize data and listen for any further local changes to the user.

state is of type DataState which has loading, error and data substates.

In addition to the reactivity, models have ActiveRecord-style extension methods so the above becomes:

GestureDetector(
  onTap: () => User(id: 1, name: 'Updated').save(),
  child: Text('Update')
),

Compatibility

Fully compatible with the tools we know and love:

Flutter ✅ Or plain Dart. It does not require Flutter.
json_serializable ✅ Fully supported (but not required)
Riverpod ✅ Supported & automatically wired up
Classic JSON REST API ✅ Built-in support!
JSON:API ✅ Supported via external adapter
Firebase, Supabase, GraphQL ✅ Can be fully supported by writing custom adapters
Freezed ✅ Supported!
Flutter Web - TBD

📲 Apps using Flutter Data in production

logos

Integrations

  • Appwrite Offline. A Flutter Data adapter for Appwrite that provides offline support, real-time updates, and seamless integration with Flutter Data's powerful features.

🚨 BREAKING CHANGES IN 2.0

  • All methods are now directly on Adapter, there is no Repository, RemoteAdapter or LocalAdapter. Any method you are looking for is probably on Adapter, for example, findAll from LocalAdapter is now called findAllLocal
  • For initialization we no longer call the configure... method on the Riverpod overrides, we just do localStorageProvider.overrideWithValue and pass a LocalStorage instance; the actual initialization is done via initializeFlutterData which needs an adapter map. An adapterProvidersMap is conveniently code-generated and available on main.data.dart
  • Offline operations are still supported, but retries are not, it is up to clients to implement them

📚 Public API

Initialization

First you need to supply a local storage provider, via Riverpod configuration. A popular option for the base directory is using the path_provider package.

ProviderScope(
  overrides: [
    localStorageProvider.overrideWithValue(
      LocalStorage(
        baseDirFn: () async {
          return (await getApplicationSupportDirectory()).path;
        },
        busyTimeout: 5000,
        clear: LocalStorageClearStrategy.never,
      ),
    )
  ],
  // ...
),

And initialize like so:

return Scaffold(
  body: ref.watch(initializeFlutterData(adapterProvidersMap)).when(
    data: (_) => child,
    error: (e, _) => const Text('Error'),
    loading: () => const Center(child: CircularProgressIndicator()),
  ),
Adapters

Method names should be self explanatory. All of these methods have a reasonable default implementation.

Local storage

All models are identified by keys, that might be associated to an id (either self-assigned or fetched from a remote source).

Keys have the format model#5.

List<T> findAllLocal();

List<T> findManyLocal(Iterable<String> keys);

T? findOneLocal(String? key);

T? findOneLocalById(Object id);

bool exists(String key);

T saveLocal(T model, {bool notify = true});

Future<List<String>?> saveManyLocal(Iterable<DataModelMixin> models,
      {bool notify = true, bool async = true});

void deleteLocal(T model, {bool notify = true});

void deleteLocalById(Object id, {bool notify = true});

void deleteLocalByKeys(Iterable<String> keys, {bool notify = true});

Future<void> clearLocal({bool notify = true});

int get countLocal;

Set<String> get keys;
Remote & watchers
Future<List<T>> findAll({
    bool remote = true,
    bool background = false,
    Map<String, dynamic>? params,
    Map<String, String>? headers,
    bool syncLocal = false,
    OnSuccessAll<T>? onSuccess,
    OnErrorAll<T>? onError,
    DataRequestLabel? label,
  });

Future<T?> findOne(
    Object id, {
    bool remote = true,
    bool background = false,
    Map<String, dynamic>? params,
    Map<String, String>? headers,
    OnSuccessOne<T>? onSuccess,
    OnErrorOne<T>? onError,
    DataRequestLabel? label,
  });

Future<T> save(
    T model, {
    bool remote = true,
    Map<String, dynamic>? params,
    Map<String, String>? headers,
    OnSuccessOne<T>? onSuccess,
    OnErrorOne<T>? onError,
    DataRequestLabel? label,
  });

Future<T?> delete(
    Object model, {
    bool remote = true,
    Map<String, dynamic>? params,
    Map<String, String>? headers,
    OnSuccessOne<T>? onSuccess,
    OnErrorOne<T>? onError,
    DataRequestLabel? label,
  });

Set<OfflineOperation<T>> get offlineOperations;

DataState<List<T>> watchAll({
    bool remote = false,
    Map<String, dynamic>? params,
    Map<String, String>? headers,
    bool syncLocal = false,
    String? finder,
    DataRequestLabel? label,
  });

DataState<T?> watchOne(
    Object model, {
    bool remote = false,
    Map<String, dynamic>? params,
    Map<String, String>? headers,
    AlsoWatch<T>? alsoWatch,
    String? finder,
    DataRequestLabel? label,
  });

DataStateNotifier<List<T>> watchAllNotifier(
      {bool remote = false,
      Map<String, dynamic>? params,
      Map<String, String>? headers,
      bool syncLocal = false,
      String? finder,
      DataRequestLabel? label});

DataStateNotifier<T?> watchOneNotifier(Object model,
      {bool remote = false,
      Map<String, dynamic>? params,
      Map<String, String>? headers,
      AlsoWatch<T>? alsoWatch,
      String? finder,
      DataRequestLabel? label});

final coreNotifierThrottleDurationProvider;
Serialization
Future<Map<String, dynamic>> serialize(T model,
      {bool withRelationships = true});

Future<DeserializedData<T>> deserialize(Object? data,
      {String? key, bool async = true});

Future<DeserializedData<T>> deserializeAndSave(Object? data,
      {String? key, bool notify = true, bool ignoreReturn = false});

➕ Questions and collaborating

Please use Github to ask questions, open issues and send PRs. Thanks!

Tests can be run with: dart test

📝 License

See LICENSE.

Kategorien

Abhängigkeiten

Abhängigkeiten 17 Einträge

Quell-Repository

Projekte, die dieses Paket verwenden

Noch keine Daten verfügbar.