Backbone-DNode is a server to client integration package for use with, you guessed it, Backbone and DNode. The package brovides both node.js server side code for CRUD and Pubsub routines, as well as the matching client (or server) side routines.
The idea is to make writing a real-time Backbone application as simple as possible, the app is supported on the server side by using the Mongoose ORM for final validation and persistence.
The annotated source can be found here
The project can be installed via NPM, or by cloning this repo into your project.
npm install backbone-dnode
or
git clone git://github.com/sorensen/backbone-dnode.git
cd backbone-dnode
npm link
To run the tests
npm test
or
make test && make clean
Whip up a server and attatch DNode, while using the backbone-dnode methods as middleware.
var express = require('express')
, DNode = require('dnode')
, BackboneDNode = require('backbone-dnode')
, server = express.createServer()
Simply allow the package to be served through your express static if
you have included the package via npm
. Serving up the client side script
can also be done via browserify,
but that is entirely up to you, as this can be done many ways, and I generally
prefer to bundle all client-side javascript into a single minifified file.
server.use(express.static(__dirname + '/node_modules/backbone-dnode/browser'))
Register your Mongoose schemas, and then pass the database instance to the CRUD configuration. At least one mongoose schema must be registered to use the CRUD routines.
var mongoose = require('mongoose')
, Schema = mongoose.Schema
, db = mongoose.connect('mongodb://localhost/db')
Foo = new Schema({
bar: { type: String, index: true }
})
(Optional) Configure the Redis connection if you would like to use Redis as the pubsub mechanics. This will allow you to use other libraries such as Cluster, letting Redis act as the message queue. If you don't use redis, the package will default to a single-threaded mode, which will work fine so long as you don't have multiple instances of node running.
var redis = require('redis')
, pub = redis.createClient()
, sub = redis.createClient()
Start the node server, and attach the backbone-dnode middleware to the DNode instance.
server.listen(8080)
dnode()
.use(BackboneDNode.pubsub({
publish: pub
, subscribe: sub
}))
.use(BackboneDNode.crud({
database: db
}))
.listen(server)
Since the model types are going to be specified on the client side, an array
of whitelisted models can be passed into the crud
and pubsub
midleware.
It is more important to whitelist the crud
middleware.
var mongoose = require('mongoose')
mongoose.model('foo', new mongoose.Schema())
mongoose.model('bar', new mongoose.Schema())
var whitelist = [
'foo'
, 'bar'
]
BackboneDnode.crud({
database: db
, whitelist: whitelist
})
Simply include the client-side part of the package onto the page, which may differ depending on how you serve up your static content.
<script src="/dnode.js"></script>
<script src="/dnode.backbone.js"></script>
The package will need to be configured as well, allowing it to be used
as DNode middleware, if you wish to use the pubsub methods of the package,
enable it, as it is not used by default. This will broadcast all changes
to any models to anyone else connected, otherwise, it will only call back to
the current client, and use the default Backbone success
methods.
DNode()
.use(root.dnodeBackbone({
pubsub: true
}))
.connect()
To connect to node.js and mongoose from the browser (or on the server),
a model type
for mongoose must be specified, as well as overriding the
sync
method on each model, an underscore mixin has been created to
provide optional support based on the model, in case you have different
persistant support in mind.
var foo = Backbone.Model.extend({
type: 'room'
, sync: _.sync
})
Now create the collection, the attributes are set on both the model and collection to ensure that they will both use the same persistance, even if a model is created outside of the collection.
var FooCollection = Backbone.Collection.extend({
url: 'foos'
, type: 'foo'
, sync: _.sync
, model: Foo
})
You can also override the sync method globally, by overriding
the default Backbone.sync
method
Backbone.sync = _.sync
Once the middleware has been established, and a model has been set to use it (or if as been overridden globally), the default Backbone methods will automatically send the changes through the socket (dnode), where they will be mapped to the corresponding Mongoose schema, and then published to the connected clients that have been subscribed to the model or collection's URL.
var options = {}
, foos = new FooCollection()
foos.subscribe(options, function() {
foos.fetch({
finished: function(model, resp, options) {
// The server has responded with the fetched data,
// and has added to the collection
}
, error: function(model, resp, options) {
// Something went wrong, the server has responded with
// an error code for client side handling
}
})
})
When the subscribe
method has returned, you are now able to use all of the default
Backbone model methods and have them interact with the server. When using any of the
Backbone fetch
, save
, create
, or delete
methods, a callback function will be
used when the server responds, and a finished
method will be executed when the middleware
is done with the Backbone integration methods. Can optionally pass in an error
method that
will be triggered if anything goes wrong on the server side. Think of finished
as the
Backbone success
callback when normally using these methods, the name is changed to avoid
conflicts.
foos.create({
bar: 'something'
})
Backbone.fetch() has been overloaded to accept a query
and sorting
argument, which will be
directly used on the server against the Mongoose ORM. The default behavior for passing in silent:true
or add:true
will still be used.
foos.fetch({
query: { bar : 'something' }
, sorting: { sort: [['created', -1]], limit: 20 }
})
(The MIT License)
Copyright (c) 2011-2012 Beau Sorensen mail@beausorensen.com
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.