Skip to content

Latest commit

 

History

History
138 lines (84 loc) · 3.54 KB

content.md

File metadata and controls

138 lines (84 loc) · 3.54 KB
Raw

Introduction

The project goal is to provide an eventstore implementation for node.js:

  • load and store events via EventStream object
  • event dispatching to your publisher
  • supported Dbs (MongoDb, Redis)
  • snapshot support

Installation

npm install eventstore

choose one of the existing storage implementation or provide your own:

// for mongoDb
npm install eventstore.mongoDb

// for redis
npm install eventstore.redis

Usage

Require the module and init the eventstore:

var eventstore = require('eventstore');

var es = eventstore.createStore(); // optional pass in your options

By default the eventstore will use an inMemory Storage, a fakePublisher and no logger.

To use the provided console.logger you could create the eventstore with option eventstore.createStore({logger: 'console'});.

Provide implementation for storage and publishing events

Example will use redis storage, but same will work for mongoDb.

var storage = require('eventstore.redis');

es.configure(function() {
    es.use(storage.createStorage());
    es.use(publisher); // your publisher must provide function 'publisher.publish(event)'
    // es.use(logger);
}).start();

Work with the eventstore

get the eventhistory of an aggregate

es.getEventStream(aggregateId, function(err, stream) {                    
    var history = stream.events; // the original event will be in events[i].payload

    // myAggregate.loadFromHistory(history);
});

store a new event and commit it to store

es.getEventStream(aggregateId, function(err, stream) {                    
    
    stream.addEvent(new event);
    stream.commit();

});

the committed event will be dispatched to the provided publisher.

you can even get an eventstream from a starting revision to an ending revision number:

es.getEventStream(aggregateId, 5 /*minRev*/, 10 /*maxRev*/, function(err, stream) {                    
});

Work with snapshotting

get snapshot and eventhistory from the snapshot point

es.getFromSnapshot(aggregateId, function(err, snapshot, stream) {

    var snap = snapshot.data;
    var history = stream.events; // events history from given snapshot

    myAggregate.loadSnapshot(snap);
    myAggregate.loadFromHistory(history);

});

create a snapshot point

es.getFromSnapshot(aggregateId, function(err, snapshot, stream) {

    var snap = snapshot.data;
    var history = stream.events; // events history from given snapshot

    myAggregate.loadSnapshot(snap);
    myAggregate.loadFromHistory(history);

    // create a new snapshot depending on your rules
    if (history.length > myRange) {
      es.createSnapshot(aggregateId, stream.currentRevision(), myAggregate.getSnap());
    }

    // go on: store new event and commit it

});

Sample Integration

  • nodeCQRS A CQRS sample integrating eventstore

Annotated Code

You can find the code documentation here.

Inspiration

Release Notes

v0.6.0

  • removed couchDb implementation
  • rewritten tests in mocha and expect.js
  • updated to node.js 0.6.15

v0.5.0

  • simplified API for storage usage
  • if possible fork dispatching to own childprocess
  • optimized lastRevision handling

v0.3.0

  • eventstreams
  • snapshoting
  • get all events with paging for replay
  • console.logger
  • db implementations for mongoDb, couchDb, redis