Override Pattern for Fakes

Override Pattern for Fakes

Give each method of a fake a nullable override function field placed right above it, so a test can change one behavior without a new fake.

The override pattern is a testing technique where each method in a fake implementation has a corresponding nullable function field that can be overridden to customize its default behavior in tests. The override field should be placed directly above the method it affects.

Structure

class FakeSomeClass implements SomeInterface {
  // Override field directly above the method it affects
  Future<T> Function(/* same parameters as method */)? overrideMethodName;

  @override
  Future<T> methodName(/* parameters */) async {
    if (overrideMethodName != null) {
      return overrideMethodName!(/* same parameters */);
    }
    // Default implementation
  }
}

// Usage in test
test('handles upload failure', () {
  final fake = FakeApi();
  fake.overrideAddPhoto = (uuid, lat, lon) async {
    throw Exception('Network error');
  };
  // Test upload failure
});

Advantages

• Flexibility: Tests can easily customize behavior without modifying the fake implementation
• Clarity: The relationship between override and method is clear through proximity
• Type Safety: Override functions match the method signature exactly
• Default Behavior: When not overridden, the fake provides sensible defaults
• Test Isolation: Each test can have its own override without affecting others

Best Practices

• Place the override field directly above the method it affects
• Use positional parameters in override functions
• Name parameters in override function types for better readability
• Keep the default implementation simple and predictable (success case)