Event Emitter
var events = require('events');
var eventEmitter = new events.EventEmitter();
Http Server Request
var http = require('http');
var server = http.createServer(function(req, res){})
Creating instance for new events.EventEmitter() not for http.createServer()
The http library exports the Server constructor so just like with eventEmitters, you can do:
let server = new http.Server(fn);
This, is exactly the same way that you do things with the events library.
http.createServer() is a factory function. It creates a Server object for you. It's not entirely obvious why they decided to also export a factory function, but you can use either the factory function or the constructor. If you look at the code for http.createServer(), you can see that this is all it does:
function createServer(requestListener) {
return new Server(requestListener);
}
where Server is the same thing as http.Server. So, the http library just offers two ways of creating a server, one of which works exactly like the events library.
Related
I'm currently experimenting with NodeJS streams and had a deeper look into the http.ServerResponse object that's being created upon every http request of the http.createServer request handler.
What I'm now trying to do is having the exact same API of the http.ServerResponse object in a different process connected via another arbitrary method (for instance using Streams) and pipe all output of this object (including headers!) to the actual request, like the following:
-[http]-> server1 -[stream]-> server2 -[stream]-> server1 -[http]->
I've tried a couple of variants, like the following (only local) example:
var http = require('http');
var net = require('net');
var through = require('through');
var StreamWrap = require('_stream_wrap').StreamWrap;
http.createServer(function(req, _res) {
var resStream = through();
resStream.pipe(_res.socket);
resStream.__proto__.cork = function() {};
resStream.__proto__.uncork = function() {};
var resSocket = new StreamWrap(resStream);
var res = new http.ServerResponse(req);
res.assignSocket(resSocket);
res.writeHead(201, {'Content-Type': 'text/plain'});
res.write('hello');
res.end();
}).listen(8000, function() {
console.log("Server listening");
});
which should essentially send the raw HTTP message to the underlying socket (or not?), but for some reason I'm getting a Segmentation fault: 11 - I'm also not sure whether I'm using the StreamWrap object correctly. But I think that this is not an optimal solution as the ServerResponse object handles a lot of things with regards to the socket internally.
So what would be the best way to tackle this?
If I'm piping directly to the response object, this results only in writing to the HTTP body (see also this answer), and I'm loosing the header information. So would it be best to separate header and body and then transfer them separately? How do I know when the headers are being set final on the ServerResponse object, and when the data starts (apart from other options such as trailing headers)?
Another option would be to remotely call methods on the ServerResponse Object i.e. via dnode, but isn't there a better way to do this? I would really love to be able to use Express for instance on the remotely connected server, that's why I want to keep the http.ServerResponse API.
I'm glad for any input!
Thanks!
I have a messaging controller setup in sails.js and want to test it with multiple clients to see if the pub/sub code works. I set up a test file
var socketIOClient = require('socket.io-client');
var sailsIOClient = require('sails.io.js');
var socket1 = socketIOClient;
var client1 = sailsIOClient(socket1);
var socket2 = socketIOClient;
var client2 = sailsIOClient(socket2);
var socket3 = socketIOClient('http://localhost:1337', {'force new connection': true});
var client3 = sailsIOClient(socket2);
...
client1.socket.get... works and says it is subscribed.
client1.socket.post... works and posts a message to the DB.
So I want to test that a client can receive the notification when a new message is posted. However, when I post from either client1 or client2, it posts from both. Essentially, they are linked to the same socket object or something like that, but I don't know where. So I want to connect multiple sockets, and I've tried variations like socket3 and client3, but get the following problem:
client3.socket.get... and client3.socket.post... and other variations (forceNew, multiplexing, etc.) each hang up and don't resolve.
Example of hang up:
sails.log('posting...');
client3.socket.post('/v1.0/messaging', data, function(body, JWR){
sails.log('posted');
done();
});
Only posting... is logged in this way, but posted is logged if using client1 or client2.
My question:
How can I connect multiple clients to my sails api to test if my pub/sub controller works?
I can't test it right now, but you could try this
var socketIOClient = require('socket.io-client');
var sailsIOClient = require('sails.io.js');
// Instantiate the socket client (`io`)
var io = sailsIOClient(socketIOClient);
// prevents socket to connect with it's default origin
io.sails.autoConnect = false
// Ask the client to create two socket connections
var socket1 = io.sails.connect('http://localhost:1337');
var socket2 = io.sails.connect('http://localhost:1337');
// Test it
socket1.get(url, data, cb)
socket1.post(url, data, cb)
socket2.get(url, data, cb)
socket2.post(url, data, cb)
// If you want to configure and use the "eager" instance
io.sails.url = 'http://localhost:1337';
io.socket.get(url, data, cb)
This way, you would create several SailsSocket instance instead of using the "eager" instance.
When you use sails.io.js in a browser, io.socket contains the socket instance (called "eager instance" in the comments) which will automatically try to connect using the host that the js file was served from. io.sails.connect() allows you to create other instances.
The correct syntax for actual version of socket.io should be
//first socket
var socket1 = socketIOClient('http://localhost:1337', {'forceNew: true});
//second socket
var socket2 = socketIOClient('http://localhost:1337', {'forceNew: true});
See socket.io docs http://socket.io/blog/socket-io-1-2-0/#
I've been trying to create (basically) a factory function that configures and builds an expressjs server for a dozen smaller specialized servers I have. For part of this I want to augment the listen function.
I would like to know the best way to go about this. I'm also looking for a reusable design choice here.
Server is created normally:
var httpServer = express();
...
Because of the way express is designed (Not sure if I am correct) I cannot access a {whatever}.prototype.listen. So I have come up with two approaches.
Using an additional variable in the current scope:
var oldListen = httpServer.listen;
httpServer.listen = function(callback){
...
oldListen.call(httpServer, options.port, options.host, function(){
...
if ( typeof callback == 'function' ) callback();
});
};
Which works and is fairly straight forward but then I have a variable hoisting wart. I also have a closure solution, but I think it may be too obtuse to be practical:
httpServer.listen = (function(superListen){
return function(callback){
...
superListen.call(httpServer, options.port, options.host, function(){
...
if ( typeof callback == 'function' ) callback();
});
};
})(httpServer.listen);
Both examples are part of the factory context and I am intentionally reducing the arguments passed to the function.
Any help would be appreciated.
If you insist on "overloading", make sure you implement the original footprint (such is the nature of overloading). Express listen is just an alias to node's internal http listen method:
server.listen(port, [host], [backlog], [callback]);
UPDATE: Express even suggests using node's server API for custom implementations: http://expressjs.com/4x/api.html#app.listen
Otherwise, you should create your own custom listen method which would be defined like:
httpServer.myCustomListen = function (callback) {
httpServer.listen.call(httpServer, options.port, options.host, callback);
}
The second option is your best bet, but in order for it to work, you must extend the express library. Express is open source and hosted on Github. Fork it and modify it as you please. Periodically pull in new updates so you stay up-to-date with the core library. I do this all the time with node modules.
There are two benefits from doing it this way:
You have complete control to customize the code however you see fit while staying up to date with the code written by the original authors.
If you find a bug or build a cool feature, you can submit a pull request to benefit the community at large.
You would first fork the repository, then grab the URL for your fork, clone it, and then add a reference to the original "upstream" repo:
git clone [url_to your_fork]
cd express
git remote add upstream git#github.com:strongloop/express.git
Then you can push changes to your own repo (git push). If you want to get updates from the original repo, you can pull from the upstream repo: git pull upstream master.
If you want to add your custom fork of express as an npm module for a project, you would use the following:
npm install git://github.com/[your_user_name]/express.git --save
As Victor's answer pointed out, express's prototype is in express/lib/application.js. That file is used to build express and is exported via the application namespace in express/lib/express.js. Therefore, the .listen function can be referenced using express.appliction.listen.
One can use this method then: (similar to Victor's method)
var express = require('express');
express.application._listen = express.application.listen;
express.application.listen = function(callback) {
return this._listen(options.port, options.host, callback);
};
One can also use Lo-dash's _.wrap function if you don't want to store the base function in a variable yourself. It would look something like this:
var express = require('express');
var _ = require('lodash');
express.application.listen = _.wrap(express.application.listen, function(listenFn) {
return listenFn(options.port, options.host, callback); // Called with the same this
};
However, using these methods would run into the problems that you mentioned in your question (variable hoisting, creating an extra variable). To solve this, I would usually create my own subclass of express.application and replace the .listen function in that subclass and tell express to use that subclass instead. Due to express's current structure, however, you cannot replace express.application with your own subclass without overriding the express() function itself.
Hence, what I would do is to take over express.application.listen completely since it is only 2 lines. It is rather simple!
var express = require('express');
var http = require('http');
express.application.listen = function(callback) {
return http.createServer(this).listen(options.port, options.host, callback);
};
You can even make an https option!
var express = require('express');
var http = require('http');
var https = require('https');
express.application.listen = function(callback) {
return (options.https ? http.createServer(this) : https.createServer({ ... }, this))
.listen(options.port, options.host, callback);
};
Note: One of the other answers mentions forking express and modifying it. I would have a tough time justifying that for such a small function.
You should be able to easily overload the express listen function. You can access it in the following Object path: express.application.listen
So, you can implement something like this:
var express = require('express');
express.application.baseListen = express.application.listen;
express.application.listen = function(port) {
console.log('Port is: ' + port);
this.baseListen(port);
};
The implementation of the listen function is in the following path under the express module folder: node_modules\express\lib\application.js
Bind and listen for connections on the given host and port. This method is identical to node's http.Server#listen().
var express = require('express');
var app = express();
app.listen(3000);
The app returned by express() is in fact a JavaScript Function, designed to be passed to node's HTTP servers as a callback to handle requests. This allows you to provide both HTTP and HTTPS versions of your app with the same codebase easily, as the app does not inherit from these (it is simply a callback):
var express = require('express');
var https = require('https');
var http = require('http');
var app = express();
http.createServer(app).listen(80);
https.createServer(options, app).listen(443);
The app.listen() method is a convenience method for the following (if you wish to use HTTPS or provide both, use the technique above):
app.listen = function(){
var server = http.createServer(this);
return server.listen.apply(server, arguments);
};
Reference:http://expressjs.com/api.html
Hope This helps.
I am using session.socket.io for my app
o = require('socket.io').listen(tmpServer),
appCookieParser = express.cookieParser(settings.cookie.secret),
appRedisStore = new RedisStore(),
sessionIO = new SessionSockets(io, appRedisStore, appCookieParser
App.app.use(appCookieParser)
App.app.use(express.session({
store: appRedisStore
}))
Note that App is a global variable which holds some of my app data and some helper functions, as I followed the LearnAllTheNodes application structure.
Then I define the callback on connection
sessionIO.on('connection', socketHandler)
where socketHandler is a function returned by a require call.
module.exports = function(err, socket, session) {
console.log(io.sockets.clients('a'))
socket.join('a')
console.log(io.sockets.clients('a'))
}
According to the documentation I should have access to my io object, but I am always getting the error that io is not defined.
Note that if I am emitting some events or listening for some events in the sockatHandler I am getting/sending them properly. Just I have no access to io.
UPDATE
It turns out it works perfectly if I am implementing the callback function as an anonymous function. The documentation is confusing, it should be: you can use io if you have a reference to it.
You seem to have forgotten to define io which here should be the listener.
According to the code you display, you should probably have something like
var app = express();
var server = http.createServer(app);
var io = socketio.listen(server);
Here's an example of a complete application using express, socket.io and session.socket.io.
To make my app less crowded I've updated the socketHandler so it takes all my socket related objects
socketHandler(sessionIO, io)
and I am handling the socket operations in another module via these objects.
I'd like to augment the prototype of the request object in expressjs, but it isn't clear where this request is defined? I think it is http.ServerRequest, but I can't find that definition either.
What's the right way to do the following...
http.ServerRequest.prototype.redirect = function(path) { }
Express itself adds it's utility methods to http.IncomingMessage.prototype, using this pattern in 2.*:
var http = require('http'),
req = http.IncomingMessage.prototype;
req.foo = function(bar) {
// Do cool stuff
};
And this pattern in 3.*:
var http = require('http');
var req = exports = module.exports = {
__proto__: http.IncomingMessage.prototype
};
It's wise to be careful with monkey patching though, as Vadim Baryshev warns in his answer.
Look at Connect framework and his middleware libs. Every middleware extends request and response objects after their creation. Changing prototype of core objects not the best way because this can lead to unpredictable behavior in other modules.