passnote_generator 1.1.2

Code generator for Passnote document models: one @Document annotation yields the codec, the document-id extractor, and the per-field CRDT schema.

passnote_generator

Code generator for Passnote document models: one annotation yields the codec, the document id extractor, and the per-field CRDT schema — no hand-written maps, and the three can never drift apart.

Use

Add to a package that defines document models:

dev_dependencies:
  build_runner: ^2.15.0
  passnote_generator: ^1.1.2

Annotate the model and add a part directive:

import 'package:passnote/passnote.dart';

part 'task.g.dart';

@Document()
class Task {
  const Task({required this.id, required this.title, this.qty = 0});

  @DocumentId()
  final String id;

  final String title; // Unannotated: merges last-write-wins.

  @Crdt(PnCounter())
  final num qty; // Concurrent increments sum.

  static final Codec<Task> codec = _$TaskCodec;
  static const MergeStrategy merge = _$TaskMergeStrategy;
}

Then generate:

dart run build_runner build --delete-conflicting-outputs

And use the generated artefacts when opening the collection:

final tasks = db.collection<Task>('tasks', Task.codec, merge: Task.merge);

What the generator expects

• An unnamed constructor whose parameters (named or positional) match the instance fields by name. Parameter default values fill fields a stored document lacks.
• Exactly one @DocumentId() field, a non-nullable String.
• Field types: String, int, double, num, bool, their nullable forms, List of a scalar, Map<String, scalar>. Anything else fails the build with a message naming the field — write a manual JsonCodec for exotic shapes.
• A @Crdt(PnCounter()) field is a number; a @Crdt(OrSet()) field is a List.