reactive_orm #

|

⚠️ Note: reactive_orm stands for Object Reactive Model — inspired by ORM patterns, but it is not a database ORM. It is a lightweight, field-level state management solution for Flutter UI.

🎬 Demo #

✨ Core Philosophy #

• Models are plain Dart objects.
• State changes happen via normal field mutation.
• UI reacts automatically, with optional field-specific reactivity.
• No ChangeNotifier, providers, streams, or extra boilerplate.
• Supports object-wise, field-wise, and nested reactivity.
• ORM-inspired design: models behave like domain entities, but in-memory.

✨ Features #

• ✅ Reactive models with automatic UI updates
• ✅ Object-wise reactivity (entire model rebuilds)
• ✅ Field-wise reactivity (only selected fields rebuild)
• ✅ Nested & shared models supported
• ✅ Multiple widgets can listen to the same model
• ✅ Minimal boilerplate
• ✅ ORM-style mental model

🆚 Comparison #

🚀 Getting Started #

Installation #

dependencies:
  reactive_orm: ^0.0.6

🧩 Basic Usage #

1️⃣ Create a Reactive Model

import 'package:reactive_orm/reactive_orm.dart';

class Task extends ReactiveModel {
  String _title;
  bool _completed = false;
  String _status = "Idle";

  Task({required String title}) : _title = title;

  String get title => _title;
  set title(String value) {
    if (_title != value) {
      _title = value;
      notifyListeners('title');
    }
  }

  bool get completed => _completed;
  set completed(bool value) {
    if (_completed != value) {
      _completed = value;
      notifyListeners('completed');
    }
  }

  String get status => _status;
  set status(String value) {
    if (_status != value) {
      _status = value;
      notifyListeners('status');
    }
  }
}

2️⃣ Object-wise Reactivity (Whole Object)

Any field change rebuilds the widget:

final task = Task(title: "Object-wise");

ReactiveBuilder<Task>(
  model: task,
  builder: (t) => ListTile(
    title: Text(t.title),
    subtitle: Text(t.status),
    trailing: Checkbox(
      value: t.completed,
      onChanged: (v) => t.completed = v!,
    ),
  ),
);

3️⃣ Field-wise Reactivity (Optimized)

Widget rebuilds only when specified fields change:

final task = Task(title: "Field-wise");

ReactiveBuilder<Task>(
  model: task,
  fields: ['completed', 'status'],
  builder: (t) => ListTile(
    title: Text(t.title),
    subtitle: Text(t.status),
    trailing: Checkbox(
      value: t.completed,
      onChanged: (v) => t.completed = v!,
    ),
  ),
);

• Rebuilds only when completed or status changes.
• Changes to other fields are ignored.

🔗 Relationship Patterns #

1-> Many → One (Aggregation)

Multiple models feed a single reactive observer:

class Dashboard extends ReactiveModel {
  final List<Task> sources;
  Dashboard(this.sources) {
    for (final task in sources) addNested(task);
  }
}

final dashboard = Dashboard([task1, task2]);

ReactiveBuilder<Dashboard>(
  model: dashboard,
  builder: (_) => Text("Dashboard updated"),
);

• ✔ Any task change updates the dashboard automatically.

2-> Many ↔ Many (Shared Models)

Same model instance used across multiple parents:

class Group extends ReactiveModel {
  final String name;
  final List<Task> tasks;

  Group({required this.name, required this.tasks}) {
    for (final task in tasks) addNested(task);
  }
}

• ✔ One task update reflects everywhere.
• ✔ No duplication or manual syncing required.

🧠 How It Works (High Level) #

• Models extend ReactiveModel.
• Field setters call notifyListeners(fieldName) when the value changes.
• ReactiveBuilder widgets listen to either:
  • Whole model (object-wise)
  • Specific fields (field-wise)
• Nested models propagate changes upward automatically.
• Widgets rebuild safely, respecting Flutter lifecycle.

🛣 Roadmap #

• Batch updates / transactions
• Async persistence hooks
• Database adapters (optional)
• DevTools / debug inspector
• Optional code generation

🧪 Status #

• Version: 0.0.6
• Stability: Stable / suitable for prototyping and early production
• Use case: Learning, prototyping, early production experiments

📌 Summary #

reactive_orm is ideal when you want:

• Clean Dart models with fine-grained reactivity
• ORM-like mental model, but in-memory and UI-focused
• Minimal boilerplate, field-level and object-level reactivity
• Nested & shared models without manual wiring
• A lightweight state management solution that scales with Flutter apps