novation-mobile

novation-mobile

novation-mobile is a framework using a NodeJS/MongoDB + Mongoose/Socket.IO/Redis stack.

Get the source from GitHub or install via NPM

npm install novation-mobile --save

Note: this will take a while. We include all the dependencies to run this.

Version

0.5.0

How to use

In a web.js file at your project root, use the following to set up a novation-mobile server:

var nm = require('novation-mobile');

var config = {
  appName: 'ExampleApp',
  server: 'Main',
  port: process.argv[2] || 4050,
  useStaticServer: true,
  favicon: 'favicon.ico',
  envLocation: '_env.js',
  preContent: 'routes.js',
  postContent: 'routes2.js',
  mongooseSchemaLocation: '_schema.js',
  viewEngine: 'jade',
  viewDirectory: 'views',
  publicDirectory: 'public',
  servers: ['Main:' + os.hostname()],
  logger: {
    userName: '',
    password: ''
  },
  api: {
    location: 'api'
  },
  onlineUsersConfig: {
    timer:900
  }
};

nm.extra(__dirname).server(config);

Each option should be customized for your app.

Config Options:

1. appName: Name of your app.
2. server: Name of the server that the current code is running on.
3. port: What port to run server on. Defaults to process.env.PORT and then to 4050.
4. useStaticServer: Wether to allow the server to act as a static server for a specified folder. Used with viewEngine, viewDirectory, and publicDirectory. Defaults to true.
5. favicon: Location of your favicon. Defaults to 'public'.
6. envLocation: Location of your environmental variables.
7. preContent: Location of your routes that run before api routes.
8. postContent: Location of your routes that run after api routes.
9. mongooseSchemaLocation: Location of your mongoose schema. Defaults to '_schema.js'.
10. viewEngine: Which view engine to use. Example: jade, html, handlebars, etc.
11. viewEngine: Which directory to be used to serve views, if using dynamic views.
12. publicDirectory: Which directory to be used as your 'static folder.'
13. servers: An array of servers that is used by redis-logger and socket.io-online-users.
14. logger.username: username to access the redis-logger
15. logger.password: password to access the redis-logger
16. api.location: Location of your api folder.
17. api.version: The version number the server should use for internal calls.
18. api.addSocketsToRoom: A function that is called every API call that allows you to add a socket/user to a room for socket.io. The function has two arguments: (session, socket);
19. onlineUsersConfig: An object with configuration options to use socket.io-online-users.
20. onlineUsersConfig.timer: The buffer time until the server updates the server with who is online.
21. ssl: An object of options to use ssl on your node server.
22. ssl.key: Location of key file to use.
23. ssl.cert: Location of the cert file to use.
24. ssl.port: Port to have your node.js https server run on.
25. sslRedirect: Redirect http to https.
26. dontUseRedisTTL: do not use a ttl for redis.
27. ttl: Time in seconds until redis expires documents. Defaults to 3600.

Components

Environmental Variables

Location: _env.js

Allows you to set environment variables used throughout the app:

exports.configureEnvironment = function(app, process) {
  // required variables
  process.env['SESSION_KEY'] = 'my_express.sid';
  process.env['SESSION_SECRET'] = 'exampleSecret';
  process.env['COOKIE_KEY'] = 'ExampleCookie';
  process.env.MONGO_URI = '';
  process.env.REDIS_URI = 'redis://redis:redis@ip:port/dbindex';

  // add your own
  process.env['SOME_API_KEY'] = 'aaa111nnn123';
};

Routes

Location: routes.js

Allows you to create custom routes for your app.

exports.content = function(app, io) {
  // you can use this page for additional, custom routes
  app.get('/', function(req, res, next) {
    res.send('This is an example server');
  });
};

Standard APIs

Location: api/

Allows you to create APIs that can be accessed by both socket.io and by RESTful requests.

Say I want to call the function 'run' under 'SomeAPI'. I can request the API either using http://localhost:4050/api/SomeAPI/run or by using sockets on the client:

socket.emit('api', 'SomeAPI', 'run', {
  testData: 'I Am Groot'
}, function(err, data) {
  if (err) {
    console.log(err);
  } else {
    console.log(data);
  }
});

The contents of api/SomeAPI.js then look like:

exports.run = function() {
  console.log(data.testData); // prints "I Am Groot"

  var number = Math.random();
  if (number < .5) {
    return fn('This is a standard error message.');
  } else {
    return fn(null, {
      data: 'This the standard way to send data back to the client.'
    });
  }
};

Extras has the following properties:

• mongoose - access to the mongoose variable.
• io
• socket - the particular socket connection, if available
• connectionType - either socket or http.
• fileName - the file that the API is being hit by.
• req - if available
• res - if available
• method - the method that is being called.
• ipAddress
• hostname

API Middleware Example

function testSession(data,fn,session,extras,next){
  if(!session){
    return fn("You have to have a session for this.");
  } else {
    return next();
  }
}

exports.testSession=API2(testSession,testSession,function(data,fn,session,extras){
  fn(null, 'You have a session!');
});

exports.fn=function(){
  fn(null, 'yay!!');
};

exports.staticVariacle=1;

next()

Next allows you to run the next functon in the iteration. If you want to skip all middleware except the last function, run next({ finish: true }).

Also, if you use the middleware and do not provide a connectionType in extras, API2 will add 'internal' to the connectionType.

after()

If you want to run an API after another API is complete, you may add an after() call to the middleware.

var middleware = API('middleware');
var afterware = API('afterware');

exports.run = API2(middleware.checkCredentials, function(data, fn, session, extras) {

  if (!data) {
    return fn('You did not send any data.');
  }

  var number = Math.random();
  console.log('We are sending back this number::', number);
  return fn(null, number);

});

exports.run.after(afterware.testLog);

In the above example, the run() API will use middleware to check access credentials. If the credentials middleware finishes successfully, our API does its work. As soon as fn(null, number) is called, the afterware API called eventLog is triggered. What happens inside the afterware API has no impact on what the run() API does. An afterware API gets the parameters err, res, data, session, extras, and might look something like:

exports.testLog = function(err, res, data, session, extras) {
  if (err) {
    return console.log('The API experienced an error. Log the error to the DB.')
  } else {
    return console.log('We can log the number ' + res + ' to the DB.');
  }
};

API Promises

With 0.5.0 we are introducing promises for our APIs. To turn any of our APIs as a promise, run API.Q. Although it should be compatible with several promise libraries, I recommend using the module, q.

var User = API.Q('User');

User.getData({},'session','extras').then(function(){
  console.log('success',arguments);
},function(){
  console.log('fail',arguments);
});

Mongoose Schema

Location: schema.js

Allows you to create a mongoose schema that can be used throughout your app. Configure your file to look like this:

var mongoose = require('mongoose');
var Schema = mongoose.Schema;

exports.User = mongoose.model('User', new Schema({
  firstName: String,
  lastName: String,
  fullName: String
}));

Note: everything you export in here will be attached to the global scope. It will be accessible throughout your whole server.