The project goal is to provide an eventstore implementation for node.js:
- load and store events via EventStream object
- event dispatching to your publisher (optional)
- supported Dbs (inmemory, mongodb, redis, tingodb, elasticsearch, azuretable)
- snapshot support
- query your events
npm install eventstore
var eventstore = require('eventstore');
var es = eventstore();
By default the eventstore will use an inmemory Storage.
For logging and debugging you can use debug by TJ Holowaychuk
simply run your process with
DEBUG=eventstore* node app.js
example with mongodb:
var es = require('eventstore')({
type: 'mongodb',
host: 'localhost', // optional
port: 27017, // optional
dbName: 'eventstore', // optional
eventsCollectionName: 'events', // optional
snapshotsCollectionName: 'snapshots', // optional
transactionsCollectionName: 'transactions', // optional
timeout: 10000 // optional
// authSource: 'authedicationDatabase', // optional
// username: 'technicalDbUser', // optional
// password: 'secret' // optional
});
example with redis:
var es = require('eventstore')({
type: 'redis',
host: 'localhost', // optional
port: 6379, // optional
db: 0, // optional
prefix: 'eventstore', // optional
eventsCollectionName: 'events', // optional
snapshotsCollectionName: 'snapshots', // optional
timeout: 10000 // optional
// password: 'secret' // optional
});
example with tingodb:
var es = require('eventstore')({
type: 'tingodb',
dbPath: '/path/to/my/db/file', // optional
eventsCollectionName: 'events', // optional
snapshotsCollectionName: 'snapshots', // optional
transactionsCollectionName: 'transactions', // optional
timeout: 10000 // optional
});
example with elasticsearch:
var es = require('eventstore')({
type: 'elasticsearch',
host: 'localhost:9200', // optional
indexName: 'eventstore', // optional
eventsTypeName: 'events', // optional
snapshotsTypeName: 'snapshots', // optional
log: 'warning', // optional
maxSearchResults: 10000 // optional
});
example with azuretable:
var es = require('eventstore')({
type: 'azuretable',
storageAccount: 'nodeeventstore',
storageAccessKey: 'aXJaod96t980AbNwG9Vh6T3ewPQnvMWAn289Wft9RTv+heXQBxLsY3Z4w66CI7NN12+1HUnHM8S3sUbcI5zctg==',
storageTableHost: 'https://nodeeventstore.table.core.windows.net/',
eventsTableName: 'events', // optional
snapshotsTableName: 'snapshots', // optional
timeout: 10000 // optional
});
if defined the eventstore will try to publish AND set event do dispatched on its own...
es.useEventPublisher(function(evt) {
// bus.emit('event', evt);
});
es.useEventPublisher(function(evt, callback) {
// bus.sendAndWaitForAck('event', evt, callback);
});
es.on('connect', function() {
console.log('storage connected');
});
es.on('disconnect', function() {
console.log('connection to storage is gone');
});
Define which values should be mapped/copied to the payload event.
es.defineEventMappings({
id: 'id',
commitId: 'commitId',
commitSequence: 'commitSequence',
commitStamp: 'commitStamp',
streamRevision: 'streamRevision'
});
es.init(function (err) {
// this callback is called when all is ready...
});
// or
ex.init(); // callback is optional
es.getEventStream('streamId', function(err, stream) {
var history = stream.events; // the original event will be in events[i].payload
// myAggregate.loadFromHistory(history);
});
or
es.getEventStream({
aggregateId: 'myAggregateId',
aggregate: 'person', // optional
context: 'hr' // optional
}, function(err, stream) {
var history = stream.events; // the original event will be in events[i].payload
// myAggregate.loadFromHistory(history);
});
'streamId' and 'aggregateId' are the same... In ddd terms aggregate and context are just to be more precise in language. For example you can have a 'person' aggregate in the context 'human ressources' and a 'person' aggregate in the context of 'business contracts'... So you can have 2 complete different aggregate instances of 2 complete different aggregates (but perhaps with same name) in 2 complete different contexts
you can request an eventstream even by limit the query with a 'minimum revision number' and a 'maximum revision number'
var revMin = 5,
revMax = 8; // if you omit revMax or you define it as -1 it will retrieve until the end
es.getEventStream('streamId' || {/* query */}, revMin, revMax, 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('streamId', function(err, stream) {
stream.addEvent({ my: 'event' });
stream.addEvents([{ my: 'event2' }]);
stream.commit();
// or
stream.commit(function(err, stream) {
console.log(stream.eventsToDispatch); // this is an array containing all added events in this commit.
});
});
if you defined an event publisher function the committed event will be dispatched to the provided publisher
if you just want to load the last event as stream you can call getLastEventAsStream instead of ´getEventStream´.
get snapshot and eventhistory from the snapshot point
es.getFromSnapshot('streamId', function(err, snapshot, stream) {
var snap = snapshot.data;
var history = stream.events; // events history from given snapshot
// myAggregate.loadSnapshot(snap);
// myAggregate.loadFromHistory(history);
});
or
es.getFromSnapshot({
aggregateId: 'myAggregateId',
aggregate: 'person', // optional
context: 'hr' // optional
}, function(err, snapshot, stream) {
var snap = snapshot.data;
var history = stream.events; // events history from given snapshot
// myAggregate.loadSnapshot(snap);
// myAggregate.loadFromHistory(history);
});
you can request a snapshot and an eventstream even by limit the query with a 'maximum revision number'
var revMax = 8; // if you omit revMax or you define it as -1 it will retrieve until the end
es.getFromSnapshot('streamId' || {/* query */}, revMax, function(err, 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('streamId', 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 > myLimit) {
es.createSnapshot({
streamId: 'streamId',
data: myAggregate.getSnap(),
revision: stream.lastRevision,
version: 1 // optional
}, function(err) {
// snapshot saved
});
// or
es.createSnapshot({
aggregateId: 'myAggregateId',
aggregate: 'person', // optional
context: 'hr' // optional
data: myAggregate.getSnap(),
revision: stream.lastRevision,
version: 1 // optional
}, function(err) {
// snapshot saved
});
}
// go on: store new event and commit it
// stream.addEvents...
});
es.getUndispatchedEvents(function(err, evts) {
// or es.getUndispatchedEvents('streamId', function(err, evts) {
// or es.getUndispatchedEvents({ // free choice (all, only context, only aggregate, only aggregateId...)
// context: 'hr',
// aggregate: 'person',
// aggregateId: 'uuid'
// }, function(err, evts) {
// all undispatched events
console.log(evts);
// dispatch it and set the event as dispatched
for (var e in evts) {
var evt = evts[r];
es.setEventToDispatched(evt, function(err) {});
// or
es.setEventToDispatched(evt.id, function(err) {});
}
});
for replaying your events or for rebuilding a viewmodel or just for fun...
skip, limit always optional
var skip = 0,
limit = 100; // if you omit limit or you define it as -1 it will retrieve until the end
es.getEvents(skip, limit, function(err, evts) {
// if (events.length === amount) {
// events.next(function (err, nextEvts) {}); // just call next to retrieve the next page...
// } else {
// // finished...
// }
});
// or
es.getEvents('streamId', skip, limit, function(err, evts) {
// if (events.length === amount) {
// events.next(function (err, nextEvts) {}); // just call next to retrieve the next page...
// } else {
// // finished...
// }
});
// or
es.getEvents({ // free choice (all, only context, only aggregate, only aggregateId...)
context: 'hr',
aggregate: 'person',
aggregateId: 'uuid'
}, skip, limit, function(err, evts) {
// if (events.length === amount) {
// events.next(function (err, nextEvts) {}); // just call next to retrieve the next page...
// } else {
// // finished...
// }
});
by revision
revMin, revMax always optional
var revMin = 5,
revMax = 8; // if you omit revMax or you define it as -1 it will retrieve until the end
es.getEventsByRevision('streamId', revMin, revMax, function(err, evts) {});
// or
es.getEventsByRevision({
aggregateId: 'myAggregateId',
aggregate: 'person', // optional
context: 'hr' // optional
}, revMin, revMax, function(err, evts) {});
by commitStamp
skip, limit always optional
var skip = 0,
limit = 100; // if you omit limit or you define it as -1 it will retrieve until the end
es.getEventsSince(new Date(2015, 5, 23), skip, limit, function(err, evts) {
// if (events.length === amount) {
// events.next(function (err, nextEvts) {}); // just call next to retrieve the next page...
// } else {
// // finished...
// }
});
// or
es.getEventsSince(new Date(2015, 5, 23), limit, function(err, evts) {
// if (events.length === amount) {
// events.next(function (err, nextEvts) {}); // just call next to retrieve the next page...
// } else {
// // finished...
// }
});
// or
es.getEventsSince(new Date(2015, 5, 23), function(err, evts) {
// if (events.length === amount) {
// events.next(function (err, nextEvts) {}); // just call next to retrieve the next page...
// } else {
// // finished...
// }
});
for example to obtain the last revision nr
es.getLastEvent('streamId', function(err, evt) {
});
// or
es.getLastEvent({ // free choice (all, only context, only aggregate, only aggregateId...)
context: 'hr',
aggregate: 'person',
aggregateId: 'uuid'
} function(err, evt) {
});
es.getNewId(function(err, newId) {
if(err) {
console.log('ohhh :-(');
return;
}
console.log('the new id is: ' + newId);
});
Inserting multiple events (documents) in mongodb, is not atomic.
For the eventstore tries to repair itself when calling getEventsByRevision
.
But if you want you can trigger this from outside:
es.store.getPendingTransactions(function(err, txs) {
if(err) {
console.log('ohhh :-(');
return;
}
// txs is an array of objects like:
// {
// _id: '/* the commitId of the committed event stream */',
// events: [ /* all events of the committed event stream */ ],
// aggregateId: 'aggregateId',
// aggregate: 'aggregate', // optional
// context: 'context' // optional
// }
es.store.getLastEvent({
aggregateId: txs[0].aggregateId,
aggregate: txs[0].aggregate, // optional
context: txs[0].context // optional
}, function (err, lastEvent) {
if(err) {
console.log('ohhh :-(');
return;
}
es.store.repairFailedTransaction(lastEvent, function (err) {
if(err) {
console.log('ohhh :-(');
return;
}
console.log('everything is fine');
});
});
});
- nodeCQRS A CQRS sample integrating eventstore
- Jonathan Oliver's EventStore for .net.
Currently these databases are supported:
- inmemory
- mongodb (node-mongodb-native)
- redis (redis)
- tingodb (tingodb)
- azuretable (azure-storage)
You can use your own db implementation by extending this...
var Store = require('eventstore').Store,
util = require('util'),
_ = require('lodash');
function MyDB(options) {
options = options || {};
Store.call(this, options);
}
util.inherits(MyDB, Store);
_.extend(MyDB.prototype, {
// ...
});
module.exports = MyDB;
and you can use it in this way
var es = require('eventstore')({
type: MyDB
});
// es.init...
Copyright (c) 2016 Adriano Raiano, Jan Muehlemann
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.