Node

Purely evented server-side I/O for V8 javascript.

This is an example of a web server written with Node which responds with "Hello World" after waiting two seconds:

new node.http.Server(function (req, res) {
  setTimeout(function () {
    res.sendHeader(200, [["Content-Type", "text/plain"]]);
    res.sendBody("Hello World");
    res.finish();
  }, 2000);
}).listen(8000);
puts("Server running at http://127.0.0.1:8000/");

To run the server, put the code into a file server.js and call it with the node executable

% /usr/local/bin/node server.js
Server running at http://127.0.0.1:8000/

See the API documentation for more examples.

Node is free to download, use, and build upon.

Benchmarks

TODO

Download

TODO

Build

Node aims to support all POSIX operating systems (including Windows with mingw). However at the moment it is only being tested on Macintosh and Linux.

V8, on which Node is built, supports only IA-32 and ARM processors.

The build system requires Python.

./configure
make
make install

API

Conventions: Callbacks are object members which are prefixed with on. All methods and members are camel cased. Constructors always have a capital first letter.

Node supports 3 byte-string encodings: ASCII ("ascii"), UTF-8 ("utf8"), and raw binary ("raw"). It uses strings to represent ASCII and UTF-8 encoded data. For the moment, arrays of integers are used to represent raw binary data—this representation is rather inefficient. This will change in the future, when V8 supports Blob objects.

The following are global functions:

puts(string, callback): Alias for stdout.puts(). Outputs the string and a trailing new-line to stdout.
The callback argument is optional and mostly useless: it will notify the user when the operation has completed. Everything in node is asynchronous; puts() is no exception. This might seem ridiculous but, if for example, one is piping stdout into an NFS file, printf() will block from network latency. There is an internal queue for puts() output, so you can be assured that output will be displayed in the order it was called.
print(string, callback): Like puts() but without the trailing new-line.
node.debug(string): A synchronous output function. Will block the process and output the string immediately to stdout. Use with care.
node.exit(code): Immediately ends the process with the specified code.

Timers

setTimeout(callback, delay): To schedule execution of callback after delay milliseconds. Returns a timeoutId for possible use with clearTimeout().
clearTimeout(timeoutId): Prevents said timeout from triggering.
setInterval(callback, delay): To schedule the repeated execution of callback every delay milliseconds. Returns a intervalId for possible use with clearInterval().
clearInterval(intervalId): Stops a interval from triggering.

node.fs

File I/O is tricky because there are not simple non-blocking ways to do it. Node handles file I/O by employing an internal thread pool to execute file system calls.

This part of the API is split into two parts: simple wrappers around standard POSIX file I/O functions and a user-friendly File object.

POSIX Wrappers

All POSIX wrappers have a similar form. They return undefined and have a callback called on_completion as their last argument. The on_completion callback may be passed many parameters, but the first parameter is always an integer indicating the error status. If the status integer is zero, then the call was successful. Example:

node.fs.unlink("/tmp/hello", function (status) {
  if (status == 0) 
    puts("successfully deleted /tmp/hello");
});

There is no guaranteed ordering to the POSIX wrappers. The following is very much prone to error

node.fs.rename("/tmp/hello", "/tmp/world");
node.fs.stat("/tmp/world", function (status, stats) {
  puts("stats: " + JSON.stringify(stats));
});

because it could be that stat() is executed before the rename(). The correct way to do this, is use the on_completion callback for rename()

node.fs.rename("/tmp/hello", "/tmp/world", function (status) {
  if (status != 0) return;
  node.fs.stat("/tmp/world", function (status, stats) {
    puts("stats: " + JSON.stringify(stats));
  });
});

node.fs.rename(path1, path2, on_completion(status))

rename(2)

node.fs.stat(path, on_completion(status, stats))

stat(2)

node.fs.unlink(path, on_completion(status))

unlink(2)

node.fs.rmdir(path, on_completion(status))

rmdir(2)

node.fs.close(fd, on_completion(status))

close(2)

node.fs.open(path, flags, mode, on_completion(status, fd))

open(2)

The constants like O_CREAT are defined at node.fs.O_CREAT.

node.fs.write(fd, data, position, on_completion(status, written))

Write data to the file specified by fd.

data is either an array of integer (for raw data) or a string for UTF-8 encoded characters.

position refers to the offset from the beginning of the file where this data should be written. If null, the data will be written at the current position.

`node.fs.File`

Easy buffered file object.

Internal request queues exist for each file object so that multiple commands can be issued at once without worry that they will be executed out-of-order. Thus the following is safe:

var file = new node.fs.File();
file.open("/tmp/blah", "w+");
file.write("hello");
file.write("world");
file.close();

Request queues are local to a single file. If one does

fileA.write("hello");
fileB.write("world");

it could be that fileB gets written to before fileA is written to. If a certain operation order is needed involving multiple files, use the completion callbacks:

fileA.write("hello", function () {
  fileB.write("world");
});

new node.fs.File(options={})

Creates a new file object.

The options argument is optional. It can contain the following fields

fd — a file descriptor for the file.
encoding — how file.read() should return data. Either "raw" or "utf8". Defaults to raw.

file.onError = function (method, errno, msg) { }

Callback. This is called internally anytime an error occurs with this file. There are three arguments: the method name, the POSIX errno, and a string describing the error.

Example

var path = "/some/path/that/doesnt/exist";
var file = new node.fs.File();
file.onError = function (method, errno, msg) {
  stderr.puts("An error occurred calling " + method);
  stderr.puts(msg);
  node.exit(1);
}
file.open(path, "w+")

file.open(path, mode, on_completion())

Opens the file at path.

mode is a string: "r" open for reading and writing. "r+" open for only reading. "w" create a new file for reading and writing; if it already exists truncate it. "w+" create a new file for writing only; if it already exists truncate it. "a" create a new file for writing and reading. Writes append to the end of the file. "a+"

The on_completion is a callback that is made without arguments when the operation completes. It is optional. If an error occurred the on_completion callback will not be called, but the file.onError will be called.

file.read(length, position, on_completion(data))

file.write(data, position, on_completion(written))

file.close(on_completion())

`node.tcp`

`node.tcp.Server`

Here is an example of a echo server which listens for connections on port 7000

function Echo (socket) {
  socket.setEncoding("utf8");
  socket.onConnect = function () {
    socket.send("hello\r\n");
  };
  socket.onReceive = function (data) {
    socket.send(data);
  };
  socket.onEOF = function () {
    socket.send("goodbye\r\n");
    socket.close();
  };
}
var server = new node.tcp.Server(Echo, {backlog: 1024});
server.listen(7000, "localhost");

new node.tcp.Server(connection_handler(socket), options={});

Creates a new TCP server.

connection_handler is a callback which is called on each connection. It is given one argument: an instance of node.tcp.Connection.

options for now only supports one option: backlog which should be an integer and describes how large of a connection backlog the operating system should maintain for this server. The backlog defaults to 1024.

server.listen(port, host=null)

Tells the server to listen for TCP connections to port and host. Note, host is optional. If host is not specified the server will accept connections to any IP address on the specified port.

server.close()

Stops the server from accepting new connections.

`node.tcp.Connection`

This object is used as a TCP client and also as a server-side socket for node.tcp.Servers.

new node.tcp.Connection(): Creates a new connection object.
connection.readyState: Either "closed", "open", "readOnly", or "writeOnly".
connection.setEncoding(encoding): Sets the encoding (either "utf8" or "raw") for data that is received.
connection.send(data, encoding="ascii"): Sends data on the connection. The data should be eithre an array of integers (for raw binary) or a string (for utf8 or ascii). The second parameter specifies the encoding in the case of a string—it defaults to ASCII because encoding to UTF8 is rather slow.
connection.close(): Half-closes the connection. I.E. sends a FIN packet. It is possible the server will still send some data. After calling this readyState will be "readOnly".
connection.fullClose(): Close both ends of the connection. Data that is received after this call is responded to with RST packets. If you don't know about this, just use close().
connection.forceClose(): Ensures that no more I/O activity happens on this socket. Only necessary in case of errors (parse error or so).
conneciton.onConnect = function () { };: Call once the connection is established.
conneciton.onReceive = function (data) { };: Called when data is received on the connection. Encoding of data is set by connection.setEncoding(). data will either be a string, in the case of utf8, or an array of integer in the case of raw encoding.
conneciton.onEOF = function () { };: Called when the other end of the connection sends a FIN packet. onReceive will not be called after this. After receiving this readyState will be "writeOnly". You should probably just call connection.close() in this callback.
conneciton.onDisconnect = function () { };: Called once the connection is fully disconnected.
conneciton.onError = function () { };: Called on an error.

`node.http`

The HTTP interfaces here are designed to support many features of the protocol which have been traditionally difficult to handle. In particular, large, possibly chunked, messages. The interface is careful to never buffer entire requests or responses—the user is able to stream data.

HTTP message headers are represented by an array of 2-element arrays like this

[ ["Content-Length", "123"]
, ["Content-Type", "text/plain"]
, ["Connection", "keep-alive"]
, ["Accept", "*/*"]
]

Dictionary-like objects are popularly used to represent HTTP headers but they are an incorrect abstraction. It is rare, but possible, to have multiple header lines with the same field. Setting multiple cookies in a single response, for example, can only be done with multiple Cookie lines.

`node.http.Server`

new node.http.Server(request_handler, options);

Creates a new web server.

The options argument is optional. The options argument accepts the same values as the options argument for node.tcp.Server does.

The request_handler is a callback which is made on each request with a ServerRequest and ServerResponse arguments.

server.listen(port, hostname)

Begin accepting connections on the specified port and hostname. If the hostname is omitted, the server will accept connections directed to any address.

server.close()

Stops the server from accepting new connections.

`node.http.ServerRequest`

This object is created internally by a HTTP server—not by the user. It is passed to the user as the first argument to the request_handler callback.

req.method

The request method as a string. Read only. Example: "GET", "DELETE".

req.uri

Request URI. (Object.)

req.uri.anchor

req.uri.query

req.uri.file

req.uri.directory

req.uri.path

req.uri.relative

req.uri.port

req.uri.host

req.uri.password

req.uri.user

req.uri.authority

req.uri.protocol

req.uri.queryKey

req.uri.toString(), req.uri.source

The original URI found in the status line.

req.headers

The request headers expressed as an array of 2-element arrays. Read only.

req.httpVersion

The HTTP protocol version as a string. Read only. Examples: "1.1", "1.0"

req.onBody

Callback. Should be set by the user to be informed of when a piece of the message body is received. Example:

req.onBody = function (chunk) {
  puts("part of the body: " + chunk);
};

A chunk of the body is given as the single argument. The transfer-encoding has been decoded.

The body chunk is either a String in the case of UTF-8 encoding or an array of numbers in the case of raw encoding. The body encoding is set with req.setBodyEncoding().

req.onBodyComplete

Callback. Made exactly once for each message. No arguments. After onBodyComplete is executed onBody will no longer be called.

req.setBodyEncoding(encoding)

Set the encoding for the request body. Either "utf8" or "raw". Defaults to raw.

`node.http.ServerResponse`

This object is created internally by a HTTP server—not by the user. It is passed to the user as the second argument to the request_handler callback.

res.sendHeader(statusCode, headers)

Sends a response header to the request. The status code is a 3-digit HTTP status code, like 404. The second argument, headers, should be an array of 2-element arrays, representing the response headers.

Example:

var body = "hello world";
res.sendHeader(200, [ ["Content-Length", body.length]
                    , ["Content-Type", "text/plain"]
                    ]);

This method must only be called once on a message and it must be called before res.finish() is called.

res.sendBody(chunk, encoding="ascii")

This method must be called after sendHeader was called. It sends a chunk of the response body. This method may be called multiple times to provide successive parts of the body.

If chunk is a string, the second parameter specifies how to encode it into a byte stream. By default the encoding is "ascii".

res.finish()

This method signals that all of the response headers and body has been sent; that server should consider this message complete. The method, res.finish(), MUST be called on each response.

`node.http.Client`

An HTTP client is constructed with a server address as its argument, the returned handle is then used to issue one or more requests. Depending on the server connected to, the client might pipeline the requests or reestablish the connection after each connection. Currently the implementation does not pipeline requests.

Example of connecting to google.com

var google = new node.http.Client(80, "google.com");
var req = google.get("/");
req.finish(function (res) {
  puts("STATUS: " + res.statusCode);
  puts("HEADERS: " + JSON.stringify(res.headers));
  res.setBodyEncoding("utf8");
  res.onBody = function (chunk) {
    puts("BODY: " + chunk);
  };
});

new node.http.Client(port, host);

Constructs a new HTTP client. port and host refer to the server to be connected to. A connection is not established until a request is issued.

client.get(path, request_headers);

client.head(path, request_headers);

client.post(path, request_headers);

client.del(path, request_headers);

client.put(path, request_headers);

Issues a request; if necessary establishes connection.

request_headers is optional. request_headers should be an array of 2-element arrays. Additional request headers might be added internally by Node. Returns a ClientRequest object.

Important: the request is not complete. This method only sends the header of the request. One needs to call req.finish() to finalize the request and retrieve the response. (This sounds convoluted but it provides a chance for the user to stream a body to the server with req.sendBody().)

GET and HEAD requests normally are without bodies but HTTP does not forbid it, so neither do we.

`node.http.ClientRequest`

This object is created internally and returned from the request methods of a node.http.Client. It represents an in-progress request whose header has already been sent.

req.sendBody(chunk, encoding="ascii")

Sends a sucessive peice of the body. By calling this method many times, the user can stream a request body to a server—in that case it is suggested to use the

["Transfer-Encoding",
"chunked"]

header line when creating the request.

The chunk argument should be an array of integers or a string.

The encoding argument is optional and only applies when chunk is a string. The encoding argument should be either "utf8" or "ascii". By default the body uses ASCII encoding, as it is faster.

req.finish(response_handler)

Finishes sending the request. If any parts of the body are unsent, it will flush them to the socket. If the request is chunked, this will send the terminating "0\r\n\r\n".

The parameter response_handler is a user-supplied callback which will be executed exactly once when the server response headers have been received. The response_handler callback is executed with one argument: a ClientResponse object.

`node.http.ClientResponse`

This object is created internally and passed to the response_handler callback (is given to the client in req.finish function). The response object appears exactly as the header is completely received but before any part of the response body has been read.

res.statusCode: The 3-digit HTTP response status code. E.G. 404.
res.httpVersion: The HTTP version of the connected-to server. Probably either "1.1" or "1.0".
res.headers: The response headers. An Array of 2-element arrays.
res.onBody: Callback. Should be set by the user to be informed of when a piece of the response body is received. A chunk of the body is given as the single argument. The transfer-encoding has been removed.
The body chunk is either a String in the case of UTF-8 encoding or an array of numbers in the case of raw encoding. The body encoding is set with res.setBodyEncoding().
res.onBodyComplete: Callback. Made exactly once for each message. No arguments. After onBodyComplete is executed onBody will no longer be called.
res.setBodyEncoding(encoding): Set the encoding for the response body. Either "utf8" or "raw". Defaults to raw.

Modules

Node has a simple module loading system. In Node, files and modules are in one-to-one correspondence.

As an example, foo.js loads the module mjsunit.js.

The contents of foo.js:

include("mjsunit");
function onLoad () {
  assertEquals(1, 2);
}

The contents of mjsunit.js:

function fail (expected, found, name_opt) {
  // ...
}
function deepEquals (a, b) {
  // ...
}
exports.assertEquals = function (expected, found, name_opt) {
  if (!deepEquals(found, expected)) {
    fail(expected, found, name_opt);
  }
};

Here the module mjsunit.js has exported the function assertEquals(). mjsunit.js must be in the same directory as foo.js for include() to find it. The module path is relative to the file calling include(). The module path does not include filename extensions like .js.

include() inserts the exported objects from the specified module into the global namespace.

Because file loading does not happen instantaneously, and because Node has a policy of never blocking, the callback onLoad can be set and will notify the user when all the included modules are loaded. Each file/module can have an onLoad callback.

To export an object, add to the special exports object.

The functions fail and deepEquals are not exported and remain private to the module.

require() is like include() except does not polute the global namespace. It returns a namespace object. The exported objects can only be guaranteed to exist after the onLoad() callback is made. For example:

var mjsunit = require("mjsunit");
function onLoad () {
  mjsunit.assertEquals(1, 2);
}

include() and require() cannot be used after onLoad() is called. So put them at the beginning of your file.