director

A client Side/Server Side Router

npm install director
3 253 downloads in the last day
17 622 downloads in the last week
83 200 downloads in the last month

Synopsis

Director is a router. Routing is the process of determining what code to run when a URL is requested.

Motivation

A routing library that works in both the browser and node.js environments with as few differences as possible. Simplifies the development of Single Page Apps and Node.js applications. Dependency free (doesn't require jQuery or Express, etc).

Status

Build Status

Features

Usage

Client-side Routing

It simply watches the hash of the URL to determine what to do, for example:

http://foo.com/#/bar

Client-side routing (aka hash-routing) allows you to specify some information about the state of the application using the URL. So that when the user visits a specific URL, the application can be transformed accordingly.

Here is a simple example:

<!DOCTYPE html>
<html>
  <head>
    <meta charset="utf-8">
    <title>A Gentle Introduction</title>
    <script src="https://raw.github.com/flatiron/director/master/build/director.min.js"></script>
    <script>

      var author = function () { console.log("author"); },
          books = function () { console.log("books"); },
          viewBook = function(bookId) { console.log("viewBook: bookId is populated: " + bookId); };

      var routes = {
        '/author': author,
        '/books': [books, function() { console.log("An inline route handler."); }],
        '/books/view/:bookId': viewBook
      };

      var router = Router(routes);
      router.init();

    </script>
  </head>
  <body>
    <ul>
      <li><a href="#/author">#/author</a></li>
      <li><a href="#/books">#/books</a></li>
      <li><a href="#/books/view/1">#/books/view/1</a></li>
    </ul>
  </body>
</html>

Director works great with your favorite DOM library, such as jQuery.

<!DOCTYPE html>
<html>
  <head>
    <meta charset="utf-8">
    <title>A Gentle Introduction 2</title>
    <script src="https://ajax.googleapis.com/ajax/libs/jquery/1.7.1/jquery.min.js"></script>
    <script src="https://raw.github.com/flatiron/director/master/build/director.min.js"></script>
    <script>
    $('document').ready(function(){
      //
      // create some functions to be executed when
      // the correct route is issued by the user.
      //
      var showAuthorInfo = function () { console.log("showAuthorInfo"); },
          listBooks = function () { console.log("listBooks"); },
          allroutes = function() {
            var route = window.location.hash.slice(2),
                sections = $('section'),
                section;
            if ((section = sections.filter('[data-route=' + route + ']')).length) {
              sections.hide(250);
              section.show(250);
            }
          };

      //
      // define the routing table.
      //
      var routes = {
        '/author': showAuthorInfo,
        '/books': listBooks
      };

      //
      // instantiate the router.
      //
      var router = Router(routes);

      //
      // a global configuration setting.
      //
      router.configure({
        on: allroutes
      });
      router.init();
    });
    </script>
  </head>
  <body>
    <section data-route="author">Author Name</section>
    <section data-route="books">Book1, Book2, Book3</section>
    <ul>
      <li><a href="#/author">#/author</a></li>
      <li><a href="#/books">#/books</a></li>
    </ul>
  </body>
</html>

You can find a browser-specific build of director here which has all of the server code stripped away.

Server-Side HTTP Routing

Director handles routing for HTTP requests similar to journey or express:

  //
  // require the native http module, as well as director.
  //
  var http = require('http'),
      director = require('director');

  //
  // create some logic to be routed to.
  //
  function helloWorld() {
    this.res.writeHead(200, { 'Content-Type': 'text/plain' })
    this.res.end('hello world');
  }

  //
  // define a routing table.
  //
  var router = new director.http.Router({
    '/hello': {
      get: helloWorld
    }
  });

  //
  // setup a server and when there is a request, dispatch the
  // route that was requested in the request object.
  //
  var server = http.createServer(function (req, res) {
    router.dispatch(req, res, function (err) {
      if (err) {
        res.writeHead(404);
        res.end();
      }
    });
  });

  //
  // You can also do ad-hoc routing, similar to `journey` or `express`.
  // This can be done with a string or a regexp.
  //
  router.get('/bonjour', helloWorld);
  router.get(/hola/, helloWorld);

  //
  // set the server to listen on port `8080`.
  //
  server.listen(8080);

See Also:

CLI Routing

Director supports Command Line Interface routing. Routes for cli options are based on command line input (i.e. process.argv) instead of a URL.

  var director = require('director');

  var router = new director.cli.Router();

  router.on('create', function () {
    console.log('create something');
  });

  router.on(/destroy/, function () {
    console.log('destroy something');
  });

  // You will need to dispatch the cli arguments yourself
  router.dispatch('on', process.argv.slice(2).join(' '));

Using the cli router, you can dispatch commands by passing them as a string. For example, if this example is in a file called foo.js:

$ node foo.js create
create something
$ node foo.js destroy
destroy something

API Documentation

Constructor

  var router = Router(routes);

Routing Table

An object literal that contains nested route definitions. A potentially nested set of key/value pairs. The keys in the object literal represent each potential part of the URL. The values in the object literal contain references to the functions that should be associated with them. bark and meow are two functions that you have defined in your code.

  //
  // Assign routes to an object literal.
  //
  var routes = {
    //
    // a route which assigns the function `bark`.
    //
    '/dog': bark,
    //
    // a route which assigns the functions `meow` and `scratch`.
    //
    '/cat': [meow, scratch]
  };

  //
  // Instantiate the router.
  //
  var router = Router(routes);

Adhoc Routing

When developing large client-side or server-side applications it is not always possible to define routes in one location. Usually individual decoupled components register their own routes with the application router. We refer to this as Adhoc Routing. Lets take a look at the API director exposes for adhoc routing:

Client-side Routing

  var router = new Router().init();

  router.on('/some/resource', function () {
    //
    // Do something on `/#/some/resource`
    //
  });

HTTP Routing

  var router = new director.http.Router();

  router.get(/\/some\/resource/, function () {
    //
    // Do something on an GET to `/some/resource`
    //
  });

Scoped Routing

In large web appliations, both Client-side and Server-side, routes are often scoped within a few individual resources. Director exposes a simple way to do this for Adhoc Routing scenarios:

  var router = new director.http.Router();

  //
  // Create routes inside the `/users` scope.
  //
  router.path(/\/users\/(\w+)/, function () {
    //
    // The `this` context of the function passed to `.path()`
    // is the Router itself.
    //

    this.post(function (id) {
      //
      // Create the user with the specified `id`.
      //
    });

    this.get(function (id) {
      //
      // Retreive the user with the specified `id`.
      //
    });

    this.get(/\/friends/, function (id) {
      //
      // Get the friends for the user with the specified `id`.
      //
    });
  });

Routing Events

In director, a "routing event" is a named property in the Routing Table which can be assigned to a function or an Array of functions to be called when a route is matched in a call to router.dispatch().

  • on: A function or Array of functions to execute when the route is matched.
  • before: A function or Array of functions to execute before calling the on method(s).

Client-side only

  • after: A function or Array of functions to execute when leaving a particular route.
  • once: A function or Array of functions to execute only once for a particular route.

Configuration

Given the flexible nature of director there are several options available for both the Client-side and Server-side. These options can be set using the .configure() method:

  var router = new director.Router(routes).configure(options);

The options are:

  • recurse: Controls route recursion. Use forward, backward, or false. Default is false Client-side, and backward Server-side.
  • strict: If set to false, then trailing slashes (or other delimiters) are allowed in routes. Default is true.
  • async: Controls async routing. Use true or false. Default is false.
  • delimiter: Character separator between route fragments. Default is /.
  • notfound: A function to call if no route is found on a call to router.dispatch().
  • on: A function (or list of functions) to call on every call to router.dispatch() when a route is found.
  • before: A function (or list of functions) to call before every call to router.dispatch() when a route is found.

Client-side only

  • resource: An object to which string-based routes will be bound. This can be especially useful for late-binding to route functions (such as async client-side requires).
  • after: A function (or list of functions) to call when a given route is no longer the active route.
  • html5history: If set to true and client supports pushState(), then uses HTML5 History API instead of hash fragments. See History API for more information.
  • run_handler_in_init: If html5history is enabled, the route handler by default is executed upon Router.init() since with real URIs the router can not know if it should call a route handler or not. Setting this to false disables the route handler initial execution.

URL Matching

  var router = Router({
    //
    // given the route '/dog/yella'.
    //
    '/dog': {
      '/:color': {
        //
        // this function will return the value 'yella'.
        //
        on: function (color) { console.log(color) }
      }
    }
  });

Routes can sometimes become very complex, simple/:tokens don't always suffice. Director supports regular expressions inside the route names. The values captured from the regular expressions are passed to your listener function.

  var router = Router({
    //
    // given the route '/hello/world'.
    //
    '/hello': {
      '/(\\w+)': {
        //
        // this function will return the value 'world'.
        //
        on: function (who) { console.log(who) }
      }
    }
  });
  var router = Router({
    //
    // given the route '/hello/world/johny/appleseed'.
    //
    '/hello': {
      '/world/?([^\/]*)\/([^\/]*)/?': function (a, b) {
        console.log(a, b);
      }
    }
  });

URL Parameters

When you are using the same route fragments it is more descriptive to define these fragments by name and then use them in your Routing Table or Adhoc Routes. Consider a simple example where a userId is used repeatedly.

  //
  // Create a router. This could also be director.cli.Router() or
  // director.http.Router().
  //
  var router = new director.Router();

  //
  // A route could be defined using the `userId` explicitly.
  //
  router.on(/([\w-_]+)/, function (userId) { });

  //
  // Define a shorthand for this fragment called `userId`.
  //
  router.param('userId', /([\\w\\-]+)/);

  //
  // Now multiple routes can be defined with the same
  // regular expression.
  //
  router.on('/anything/:userId', function (userId) { });
  router.on('/something-else/:userId', function (userId) { });

Route Recursion

Can be assigned the value of forward or backward. The recurse option will determine the order in which to fire the listeners that are associated with your routes. If this option is NOT specified or set to null, then only the listeners associated with an exact match will be fired.

No recursion, with the URL /dog/angry

  var routes = {
    '/dog': {
      '/angry': {
        //
        // Only this method will be fired.
        //
        on: growl
      },
      on: bark
    }
  };

  var router = Router(routes);

Recursion set to backward, with the URL /dog/angry

  var routes = {
    '/dog': {
      '/angry': {
        //
        // This method will be fired first.
        //
        on: growl
      },
      //
      // This method will be fired second.
      //
      on: bark
    }
  };

  var router = Router(routes).configure({ recurse: 'backward' });

Recursion set to forward, with the URL /dog/angry

  var routes = {
    '/dog': {
      '/angry': {
        //
        // This method will be fired second.
        //
        on: growl
      },
      //
      // This method will be fired first.
      //
      on: bark
    }
  };

  var router = Router(routes).configure({ recurse: 'forward' });

Breaking out of recursion, with the URL /dog/angry

  var routes = {
    '/dog': {
      '/angry': {
        //
        // This method will be fired first.
        //
        on: function() { return false; }
      },
      //
      // This method will not be fired.
      //
      on: bark
    }
  };

  //
  // This feature works in reverse with recursion set to true.
  //
  var router = Router(routes).configure({ recurse: 'backward' });

Async Routing

Before diving into how Director exposes async routing, you should understand Route Recursion. At it's core route recursion is about evaluating a series of functions gathered when traversing the Routing Table.

Normally this series of functions is evaluated synchronously. In async routing, these functions are evaluated asynchronously. Async routing can be extremely useful both on the client-side and the server-side:

  • Client-side: To ensure an animation or other async operations (such as HTTP requests for authentication) have completed before continuing evaluation of a route.
  • Server-side: To ensure arbitrary async operations (such as performing authentication) have completed before continuing the evaluation of a route.

The method signatures for route functions in synchronous and asynchronous evaluation are different: async route functions take an additional next() callback.

Synchronous route functions

  var router = new director.Router();

  router.on('/:foo/:bar/:bazz', function (foo, bar, bazz) {
    //
    // Do something asynchronous with `foo`, `bar`, and `bazz`.
    //
  });

Asynchronous route functions

  var router = new director.http.Router().configure({ async: true });

  router.on('/:foo/:bar/:bazz', function (foo, bar, bazz, next) {
    //
    // Go do something async, and determine that routing should stop
    //
    next(false);
  });

Resources

Available on the Client-side only. An object literal containing functions. If a host object is specified, your route definitions can provide string literals that represent the function names inside the host object. A host object can provide the means for better encapsulation and design.


  var router = Router({

    '/hello': {
      '/usa': 'americas',
      '/china': 'asia'
    }

  }).configure({ resource: container }).init();

  var container = {
    americas: function() { return true; },
    china: function() { return true; }
  };

History API

Available on the Client-side only. Director supports using HTML5 History API instead of hash fragments for navigation. To use the API, pass {html5history: true} to configure(). Use of the API is enabled only if the client supports pushState().

Using the API gives you cleaner URIs but they come with a cost. Unlike with hash fragments your route URIs must exist. When the client enters a page, say http://foo.com/bar/baz, the web server must respond with something meaningful. Usually this means that your web server checks the URI points to something that, in a sense, exists, and then serves the client the JavaScript application.

If you're after a single-page application you can not use plain old <a href="/bar/baz"> tags for navigation anymore. When such link is clicked, web browsers try to ask for the resource from server which is not of course desired for a single-page application. Instead you need to use e.g. click handlers and call the setRoute() method yourself.

Attach Properties To this

Generally, the this object bound to route handlers, will contain the request in this.req and the response in this.res. One may attach additional properties to this with the router.attach method:

  var director = require('director');

  var router = new director.http.Router().configure(options);

  //
  // Attach properties to `this`
  //
  router.attach(function () {
    this.data = [1,2,3];
  });

  //
  // Access properties attached to `this` in your routes!
  //
  router.get('/hello', function () {
    this.res.writeHead(200, { 'content-type': 'text/plain' });

    //
    // Response will be `[1,2,3]`!
    //
    this.res.end(this.data);
  });

This API may be used to attach convenience methods to the this context of route handlers.

HTTP Streaming and Body Parsing

When you are performing HTTP routing there are two common scenarios:

  • Buffer the request body and parse it according to the Content-Type header (usually application/json or application/x-www-form-urlencoded).
  • Stream the request body by manually calling .pipe or listening to the data and end events.

By default director.http.Router() will attempt to parse either the .chunks or .body properties set on the request parameter passed to router.dispatch(request, response, callback). The router instance will also wait for the end event before firing any routes.

Default Behavior

  var director = require('director');

  var router = new director.http.Router();

  router.get('/', function () {
    //
    // This will not work, because all of the data
    // events and the end event have already fired.
    //
    this.req.on('data', function (chunk) {
      console.log(chunk)
    });
  });

In flatiron, director is used in conjunction with union which uses a BufferedStream proxy to the raw http.Request instance. union will set the req.chunks property for you and director will automatically parse the body. If you wish to perform this buffering yourself directly with director you can use a simple request handler in your http server:

  var http = require('http'),
      director = require('director');

  var router = new director.http.Router();

  var server = http.createServer(function (req, res) {
    req.chunks = [];
    req.on('data', function (chunk) {
      req.chunks.push(chunk.toString());
    });

    router.dispatch(req, res, function (err) {
      if (err) {
        res.writeHead(404);
        res.end();
      }

      console.log('Served ' + req.url);
    });
  });

  router.post('/', function () {
    this.res.writeHead(200, { 'Content-Type': 'application/json' })
    this.res.end(JSON.stringify(this.req.body));
  });

Streaming Support

If you wish to get access to the request stream before the end event is fired, you can pass the { stream: true } options to the route.

  var director = require('director');

  var router = new director.http.Router();

  router.get('/', { stream: true }, function () {
    //
    // This will work because the route handler is invoked
    // immediately without waiting for the `end` event.
    //
    this.req.on('data', function (chunk) {
      console.log(chunk);
    });
  });

Instance methods

configure(options)

  • options {Object}: Options to configure this instance with.

Configures the Router instance with the specified options. See Configuration for more documentation.

param(token, matcher)

  • token {string}: Named parameter token to set to the specified matcher
  • matcher {string|Regexp}: Matcher for the specified token.

Adds a route fragment for the given string token to the specified regex matcher to this Router instance. See URL Parameters for more documentation.

on(method, path, route)

  • method {string}: Method to insert within the Routing Table (e.g. on, get, etc.).
  • path {string}: Path within the Routing Table to set the route to.
  • route {function|Array}: Route handler to invoke for the method and path.

Adds the route handler for the specified method and path within the Routing Table.

path(path, routesFn)

  • path {string|Regexp}: Scope within the Routing Table to invoke the routesFn within.
  • routesFn {function}: Adhoc Routing function with calls to this.on(), this.get() etc.

Invokes the routesFn within the scope of the specified path for this Router instance.

dispatch(method, path[, callback])

  • method {string}: Method to invoke handlers for within the Routing Table
  • path {string}: Path within the Routing Table to match
  • callback {function}: Invoked once all route handlers have been called.

Dispatches the route handlers matched within the Routing Table for this instance for the specified method and path.

mount(routes, path)

  • routes {object}: Partial routing table to insert into this instance.
  • path {string|Regexp}: Path within the Routing Table to insert the routes into.

Inserts the partial Routing Table, routes, into the Routing Table for this Router instance at the specified path.

Instance methods (Client-side only)

init([redirect])

  • redirect {String}: This value will be used if '/#/' is not found in the URL. (e.g., init('/') will resolve to '/#/', init('foo') will resolve to '/#foo').

Initialize the router, start listening for changes to the URL.

getRoute([index])

  • index {Number}: The hash value is divided by forward slashes, each section then has an index, if this is provided, only that section of the route will be returned.

Returns the entire route or just a section of it.

setRoute(route)

  • route {String}: Supply a route value, such as home/stats.

Set the current route.

setRoute(start, length)

  • start {Number} - The position at which to start removing items.
  • length {Number} - The number of items to remove from the route.

Remove a segment from the current route.

setRoute(index, value)

  • index {Number} - The hash value is divided by forward slashes, each section then has an index.
  • value {String} - The new value to assign the the position indicated by the first parameter.

Set a segment of the current route.

Frequently Asked Questions

What About SEO?

Is using a Client-side router a problem for SEO? Yes. If advertising is a requirement, you are probably building a "Web Page" and not a "Web Application". Director on the client is meant for script-heavy Web Applications.

Licence

(The MIT License)

Copyright (c) 2010 Nodejitsu Inc. http://www.twitter.com/nodejitsu

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.

npm loves you