SyncedDB makes it easy to write offline-first applications with real-time syncing and server-side persistence.
SyncedDB makes web applications work beautifully both online and offline.
You can write your client as if everything was stored offline! SyncedDB takes care of synchronizing the local database to other clients in real time.
Since the widestream adoption of IndexedDB writing web applications with full offline support has been viable. But when storing data offline web applications typically lack the ability to seamlessly make a user's data available across devices. SyncedDB is a library that gives web applications the best of both worlds: a fully functional offline experience with real-time or on-demand synchronization of data when online.
SyncedDB was built with the following design goal: Be as simple as possible while still providing all the features and flexibility necessary to easily create efficient and secure real-time synchronizing web applications that works offline.
SyncedDB is a lightweight layer on top of IndexedDB. It strips away all the boilerplate that the IndexedDB API requires by introducing implicit transactions, convenience methods and promises for all asynchronous operations.
Server side SyncedDB stores a list of changes that clients can request/subscribe and post/publish to. The SyncedDB client communicates with the backend through WebSockets to achieve synchronization in real time. Furthermore, the client provides elegant conflict handling and events for reacting to changes published from the server.
Client
var stores = {
tasks: [ // One store named 'tasks'.
['byCreation', 'createdAt'] // With one index into the 'createdAt' property.
]
};
var db = syncedDB.open({ // Open database.
name: 'todoApp',
version: 1,
stores: stores,
remote: 'localhost:8080',
});
db.tasks.put({ // Add one task to database.
description: 'Task description',
finished: false,
createdAt: Date.now()
});
db.tasks.byCreation.getAll() // Get all task elements sorted after creation and output them on the console.
.then(function(tasks) {
tasks.forEach(console.log);
});
db.tasks.on('add', function(e) { // Add a handler if a new task element is pushed from remote.
console.log("New task from server: ", e.record); // Handle task by printing it on the console.
});
// Start syncing continuously, the server will now.
// push and pull changes in real time.
db.sync('tasks', {continuously: true});
Server
var Server = require('synceddb-server');
// Persistence within memory (you can use other adaptors for PostgreSQL, MySQL & CouchDB.
var sdbPersistence = require('synceddb-persistence-memory');
var server = new Server({
port: 8080,
store: sdbPersistence.create(),
});
Run the example by starting the server with node and run the client in a browser incognito mode. Then restart the client and watch the console output.
See a more sophisticated version of this example here
Some libraries cater to multiple storage backends and thus end up with a limited feature set following the lowest common denominator. Others implement a new database on top of the browser's native storage facilities. This highly increases complexity and reduces performance. By being a small wrapper around IndexedDB, SyncedDB gains some of its key features: simplicity, power and performance.
The SyncedDB backend was designed to be as flexible as possible. Users can easily plug in any database they want, create custom message handlers at relevant points and extend the communication between the client and the server with custom messages.
SyncedDB is still under development. Expect rough edges.
Persistence options are provided based on the following currently supported databases:
SyncedDB makes it easy to use different server side persistence strategies. These are easy to write (take a look at the existing options) and a test suite is provided.