data_layer_hive 0.0.2

Hive implementation of DataLayer.

Data Layer Hive

An add-on to pkg:data_layer which provides a Hive implementation of the Source interface.

Motivation

On-device caching is complicated. While some reach for SQL in the form of sqlite3, the data that arrives on a device by way of API responses is already denormalized. Instead of re-normalizing it, or forbidding use of JOINs in server-side queries, pkg:data_layer_hive brings persistent, offline-friendly document-based caching to pkg:data_layer.

Index

• Motivation
• Architecture

Architecture

pkg:data_layer_hive provides a HiveSource class which implements the Source interface specifically for caching data using pkg:hive_ce. This allows your Repository objects to persist data durably across application launches.

• Getting Started
• Creating a HiveSource

Getting started

Add the following to your pubspec.yaml:

dependencies:
  data_layer: ^0.0.2-rc.1
  data_layer_hive: ^0.0.2-rc.1

It is recommended to use pkg:hive_ce for its companion package, pkg:hive_ce_generator, which allows you to register your models like so:

@GenerateAdapters([AdapterSpec<MyModel>()])
part 'hive_adapters.g.dart';

See the example directory for a complete working sample of a Repository using both a LocalMemorySource and a HiveSource.

Creating a HiveSource

Instances of a HiveSource meant to be used within the context of a SourceList.

SourceList(
  ...
  sources: [
    HiveSource(
      bindings: userBindings,
      boxName: 'users_box',
      // You probably want to call this code elsewhere and capture its Future
      // for reuse across multiple `HiveSource` instances.
      hiveInit: Hive.initFlutter().then((_) => Hive.registerAdapters()),
    RestSource(...),
  ],
)